texts
list | meta
dict | scores
list | avg_score
float64 0
1
| num_sents
int64 1
32.7k
|
|---|---|---|---|---|
[
"Development of hybrid polymer scaffolds for potential applications in ligament and tendon tissue engineering.",
"\nFibre-based scaffolds have been widely used for tendon and ligament tissue engineering. ",
"Knitted scaffolds have been proved to favour collagenous matrix deposition which is crucial for tendon/ligament reconstruction. ",
"However, such scaffolds have the limitation of being dependent on a gel system for cell seeding, which is unstable in a dynamic environment such as the knee joint. ",
"This study developed three types of hybrid scaffolds, based on knitted biodegradable polyester scaffolds, aiming to improve mechanical properties and cell attachment and proliferation on the scaffolds. ",
"The hybrid scaffolds were created by coating the knitted scaffolds with a thin film of poly (epsilon-caprolactone) (group I), poly (D, L-lactide-co-glycolide) nanofibres (group II) and type 1 collagen (group III). ",
"Woven scaffolds were also fabricated and compared with the various hybrid scaffolds in terms of their mechanical properties during in vitro degradation and cell attachment and growth. ",
"This study demonstrated that the coating techniques could modulate the mechanical properties and facilitate cell attachment and proliferation in the hybrid scaffold, which could be applied with promise in tissue engineering of tendons/ligaments."
] |
{
"pile_set_name": "PubMed Abstracts"
}
|
[
0.0006109485984779894,
0.0006726640276610851,
0.0007438769098371267,
0.0009938295697793365,
0.000715737696737051,
0.0007465517264790833,
0.0006263580289669335,
0.0005857687792740762
] | 0.000712 | 8 |
[
"---\nabstract: 'In randomized trials, appropriately adjusting for baseline variables and short-term outcomes can lead to increased precision and reduced sample size. ",
"We examine the impact of such adjustment in group sequential designs, i.e., designs with preplanned interim analyses where enrollment may be stopped early for efficacy or futility. ",
"We address the following questions: how much precision gain can be obtained by appropriately adjusting for baseline variables and a short-term outcome? ",
"How is this precision gain impacted by factors such as the proportion of pipeline participants (those who enrolled but haven’t yet had their primary outcomes measured) and treatment effect heterogeneity? ",
"What is the resulting impact on power and average sample size in a group sequential design? ",
"We derive an asymptotic formula that decomposes the overall precision gain from adjusting for baseline variables and a short-term outcome into contributions from factors mentioned above, for efficient estimators in the model that only assumes randomization and independent censoring. ",
"We use our formula to approximate the precision gain from a targeted minimum loss-based estimator applied to data from a completed trial of a new surgical intervention for stroke. ",
"Our formula implies that (for an efficient estimator) adjusting for a prognostic baseline variable leads to at least as much asymptotic precision gain as adjusting for an equally prognostic short-term outcome. ",
"In many cases, such as our stroke trial application, the former leads to substantially greater precision gains than the latter. ",
"In our simulation study, we show how precision gains from adjustment can be converted into sample size reductions (even when there is no treatment effect).'",
"\nauthor:\n- 'Tianchen Qian[^1], Michael Rosenblum[^2], and Huitong Qiu[^3]'\nbibliography:\n- 'MISTIE\\_TMLE\\_short-term.bib'\ntitle: 'Improving Precision through Adjustment for Prognostic Variables in Group Sequential Trial Designs: Impact of Baseline Variables, Short-Term Outcomes, and Treatment Effect Heterogeneity'\n---\n\n[*Keywords:*]{} Short-term outcome; Semiparametric efficiency; Targeted minimum loss-based estimator\n\nIntroduction {#sec:introduction}\n============\n\nGroup sequential designs for randomized clinical trials involve interim analyses that may result in early stopping for efficacy or futility. ",
"We consider trial designs where the primary outcome is measured at a fixed time (called the delay) after enrollment. ",
"Prior to observing the primary outcome, participants may have baseline variables and a short-term outcome measured. ",
"For example, in the MISTIE-II trial [@MISTIE_trial2008] for evaluating a surgical treatment for intracerebral hemorrhage the primary outcome is the modified Rankin Scale (mRS), which measures degree of disability, 180 days after enrollment. ",
"A short-term outcome is mRS 30 days after enrollment. ",
"The baseline (pre-randomization) variables include age and measures of stroke severity.",
"\n\nEstimators that adjust for baseline variables are used in randomized trials because of the potential for increased precision and reduced sample size. ",
"@pocock2002review, who surveyed 50 clinical trial reports from major medical journals, found that 36 adjusted for baseline variables. ",
"Adjusting for prognostic baseline variables is recommended by regulators, e.g., @ICH9 [@EMAguideline2015; @FDA-guideline2019]. ",
"Also, in order to address missing outcomes, it may be useful to appropriately adjust for baseline and post-randomization variables (e.g., short-term outcomes) [@committee2010national; @EMAmissingdata]. ",
"Various methods for adjustment in randomized trials have been developed; see, e.g., @leon2003semiparametric [@davidian2005semiparametric; @tsiatis2008covariate; @rubin2008covariate; @moore2009binary; @moore2011relativeefficiency]. ",
"There has also been discussion on the benefits and limitations of adjustment, compared to the standard, unadjusted estimator [@yang2001efficiency; @freedman2008regression; @lin2013agnostic]. ",
"@moore2011relativeefficiency present a formula for the precision gain from adjusting for baseline variables in single stage trials.",
"\n\nWe focus on estimating the average treatment effect for continuous or binary outcomes. ",
"We derive a formula for the asymptotic precision gain (measured by the relative efficiency compared to the unadjusted estimator) due to adjustment for baseline variables and a short-term outcome using efficient estimators, that is, estimators that extract the maximum prognostic information from these variables. ",
"The formula can be used in trial planning to approximate the precision gain from adjustment, which can translate to reduced sample size.",
"\n\nWe show, using data of the MISTIE-II trial, how a modified version of these formulas can be used to approximate the precision gain from using the targeted minimum loss-based estimator (TMLE) of @Gruber2012TMLE while accounting for model misspecification. ",
"We highlight a few implications of our formula. ",
"Holding other factors fixed, larger treatment effect heterogeneity decreases the value added from adjusting for baseline variables of all participants; however, for the pipeline participants, adjusting for their baseline variables leads to increased precision only when there is treatment effect heterogeneity. ",
"Adjusting for prognostic baseline variables typically leads to precision gains even if all participants have their primary outcomes observed. ",
"In contrast, adjusting for a prognostic short-term outcome can only improve precision when there are participants with the short-term outcome observed but the primary outcome unobserved. ",
"In Section \\[sec:notation\\], we introduce notation and assumptions. ",
"In Section \\[sec:theory-ate\\], we present our formula for precision gain. ",
"We show how to approximate the precision gain from using a TMLE in Section \\[sec:tmle\\]. ",
"In Section \\[sec:application\\], we apply our formula to data from the MISTIE-II trial. ",
"In Section \\[sec:simulation\\], we conduct simulation studies. ",
"Limitations and open problems are discussed in Section \\[sec:discussion\\].",
"\n\nNotation and Assumptions {#sec:notation}\n========================\n\nFor participant $i$, let $A_i$ denote the indicator of study arm assignment. ",
"We assume $A_i$ is binary-valued with $A_i=1$ for treatment and $A_i=0$ for control. ",
"Denote by $W_i$ a vector of baseline variables measured before randomization. ",
"All variables must be preplanned in the trial protocol. ",
"Denote by $L_i$ a short-term outcome, which is observed at time $d_L$ after enrollment. ",
"Though all of our results hold if $L_i$ is any prespecified variable measured after randomization, not necessarily the outcome measured at an earlier time, we focus on the case of short-term outcomes. ",
"Also, for simplicity, we focus on cases with binary $L_i$; the results can be generalized to multidimensional, continuous $L_i$. Denote by $Y_i$ the primary outcome (continuous or binary-valued), which is observed at time $d_Y$ after enrollment with $d_Y \\geq d_L$. The delays $d_L$ and $d_Y$ are prespecified and are common to all participants. ",
"When followed up completely, participant $i$ contributes full data $(W_i, A_i, L_i, Y_i)$.\n\nLet $n$ denote the sample size. ",
"We assume that the set of vectors $\\{(W_i, A_i, L_i, Y_i)\\}_{1\\leq i \\leq n}$ are independent and identically distributed draws from the unknown joint distribution $P_0$ on generic data vector $(W, A, L, Y)$. The following assumption follows from randomization:\n\n\\[assump:randomization\\] The study arm assignment $A$ is independent of the baseline variables $W$ and $P(A=1)=P(A=0)=1/2$.\n\nResults for randomization probabilities other than $1/2$ are in Supplementary Material \\[proof:thm:avar-ate\\].",
"\n\nDefine the average treatment effect as $\\Delta = E(Y|A=1)-E(Y|A=0)$. The goal is to test the null hypothesis of no average treatment benefit: $H_0: \\Delta \\leq 0$ versus $H_1: \\Delta > 0$.\n\nWe assume that participants are enrolled at a constant rate. ",
"Since not all participants have full data observed at an interim analysis, we introduce indicators $C^L$ and $C^Y$ to denote that $L$ and $Y$, respectively, have been observed at or before a given analysis time. ",
"For a participant, $C^L = 1$ if and only if $L$ is observed, and $C^Y = 1$ if and only if $Y$ is observed. ",
"These variables get updated at each analysis. ",
"We make the following assumptions:\n\n\\[assump:independent-censoring\\] (Independent Censoring) $(C^L, C^Y)$ is independent of $(W,A,L,Y)$.\n\n\\[assump:monotone-censoring\\] (Monotone Censoring) $C^L \\geq C^Y$ with probability $1$.\n\nAn example where Assumptions \\[assump:independent-censoring\\] and \\[assump:monotone-censoring\\] hold is when administrative censoring is the only source of missingness and there are no changes over time in the population who are enrolled; this is what we simulate in Section \\[sec:simulation\\]. ",
"At any analysis time, each enrolled participant has one of the following missingness patterns:\n\n- $(C^L, C^Y) = (0,0)$: a pipeline participant with only $(W,A)$ observed;\n\n- $(C^L, C^Y) = (1,0)$: a pipeline participant with $(W,A,L)$ observed;\n\n- $(C^L, C^Y) = (1,1)$: a participant with $(W,A,L,Y)$ observed.",
"\n\nWe assume a nonparametric model for the joint distribution of the variables $(W, A, L, Y,C_L,C_Y)$, except that we make Assumptions 1-3. ",
"The semiparametric efficiency results in the following sections are with respect to this model.",
"\n\nFor a group sequential design with $K$ stages, we consider the asymptotic setting where the maximum sample size goes to infinity such that at each interim analysis the proportions of enrolled participants with final and short-term outcomes observed converge to constants $p_y$ and $p_l$, respectively (where these limit proportions may differ by analysis time, but for conciseness we suppress their dependence on the analysis time). ",
"This asymptotic regime corresponds to fixing the delay times $d_L, d_Y$ and analysis times, while increasing the enrollment rate. ",
"We only consider analysis times with $p_y > 0$. The unadjusted estimator of $\\Delta$ is the difference between the two arms of the sample mean of the primary outcome $Y$, using data from all participants with $Y$ observed, i.e., $\\sum_i A_i C^Y_i Y_i/\\sum_i A_i C^Y_i - \\sum_i (1-A_i) C^Y_i Y_i/\\sum_i (1-A_i) C^Y_i.$ All estimators of the average treatment effect $\\Delta$ that we consider are regular, asymptotically linear (RAL), defined, e.g., by [@BKRW1993]. ",
"Any such estimator $\\widehat{\\Delta}$ of $\\Delta$ is asymptotically normal, i.e., $\\sqrt{n}(\\widehat{\\Delta} - \\Delta) \\stackrel{d}{\\to} N(0, \\sigma^2)$ as sample size $n$ goes to infinity, where $\\stackrel{d}{\\to}$ denotes convergence in distribution; the variance $\\sigma^2$ is called the asymptotic variance of $\\widehat{\\Delta}$ and is denoted by ${\\text{AVar}}(\\widehat{\\Delta})$. For example, the asymptotic variance of the unadjusted estimator is $2\\{{\\text{Var}}(Y|A=1)+{\\text{Var}}(Y|A=0)\\}/p_y$ at an analysis time where $p_y$ fraction of the enrolled participants have their primary outcome observed. ",
"The asymptotic relative efficiency (ARE) between two RAL estimators $\\widehat{\\Delta}_1$ and $\\widehat{\\Delta}_2$ of $\\Delta$ is the inverse of the ratio of their asymptotic variances: ${\\text{ARE}}(\\widehat{\\Delta}_1, \\widehat{\\Delta}_2) = {\\text{AVar}}(\\widehat{\\Delta}_2)/{\\text{AVar}}(\\widehat{\\Delta}_1).$ The asymptotic sample size reduction for a Wald test of single stage trial by using estimator $\\widehat{\\Delta}_1$ instead of $\\widehat{\\Delta}_2$ is $1 - {\\text{ARE}}(\\widehat{\\Delta}_1, \\widehat{\\Delta}_2)^{-1}$. See Section \\[appen:are-sample-size-reduction\\] of the Supplementary Material for a more detailed discussion.",
"\n\nFor any random vector $X$ and $a\\in\\{0,1\\}$, let $E_a(Y \\mid X)$ and ${\\text{Var}}_a(Y \\mid X)$ denote $E(Y\\mid X, A=a)$ and ${\\text{Var}}(Y\\mid X, A=a)$, respectively. ",
"For example, $E_1(Y \\mid W) = E(Y \\mid W, A=1)$, and $E_1(Y) = E(Y \\mid A=1)$. When no subscript is used on $E$ or ${\\text{Var}}$, these refer to expectation and variance, respectively, under the population distribution of the corresponding variables.",
"\n\nBest Possible Precision Gain {#sec:theory-ate}\n============================\n\nFormula for Precision Gain\n--------------------------\n\nThe following result gives the best possible asymptotic variance that can be achieved by a RAL estimator of the average treatment effect $\\Delta$, in terms of the proportions $p_l, p_y$ of enrolled participants with $L$ and $Y$ observed, respectively, at a given analysis time:\n\n\\[thm:avar-ate\\] Suppose Assumptions \\[assump:randomization\\], \\[assump:independent-censoring\\], and \\[assump:monotone-censoring\\] hold. ",
"The asymptotic variance of any RAL estimator of $\\Delta$ is at least $$\\begin{aligned}\n & {\\text{Var}}\\{E_1(Y \\mid W) - E_0(Y \\mid W) \\}\n+ \\sum_{a\\in\\{0,1\\}} \\frac{2}{p_l} {\\text{Var}}_a\\{ E_a(Y \\mid L,W) - E_a(Y \\mid W) \\} \\nonumber \\\\\n+ & \\sum_{a\\in\\{0,1\\}} \\frac{2}{p_y} {\\text{Var}}_a\\{ Y - E_a(Y \\mid L,W) \\}. ",
"\\label{eq:thm:var-bound-ate}\\end{aligned}$$\n\nLemma \\[thm:avar-ate\\] is a consequence of the efficient influence function of $\\Delta$ in our semiparametric model that only makes the assumptions in Section \\[sec:notation\\]. ",
"This efficient influence function is given in Section \\[proof:thm:avar-ate\\] of the Supplementary Material, and follows from @robins1992semipara [@Scharfstein1999; @van2003semipara]. ",
"The TMLE estimator of @Gruber2012TMLE, which will be discussed in Section \\[sec:tmle\\], achieves this variance lower bound when all working models are correct. ",
"When certain working models for this estimator are misspecified, one can approximate the corresponding ${\\text{ARE}}$ by substituting regression model fits for conditional expectations in (\\[eq:thm:var-bound-ate\\]); see Section \\[subsec:are-fitted\\].",
"\n\nThe first term in (\\[eq:thm:var-bound-ate\\]) characterizes the variance in the conditional treatment effect across different levels of $W$. We define the *treatment effect heterogeneity*, denoted by $\\gamma$, by dividing the variance of the conditional treatment effect by the sum of variances of $Y$ in each arm: $$\\begin{aligned}\n\\gamma = \\frac{{\\text{Var}}\\{E_1(Y \\mid W) - E_0(Y \\mid W) \\}}{\\sum_{a\\in\\{0,1\\}} {\\text{Var}}_a(Y)}. ",
"\\label{def:gamma}\\end{aligned}$$ The treatment effect heterogeneity $\\gamma$ is invariant to linear transformations of $Y$, and is nonnegative. ",
"When $\\gamma = 0$ there is no treatment effect heterogeneity, i.e., the conditional treatment effect $E_1(Y \\mid W)-E_0(Y \\mid W)=\\Delta$ with probability 1.",
"\n\nFor each arm $a \\in \\{0,1\\}$, the variance of $Y$ given $A=a$ can be decomposed as follows: $${\\text{Var}}_a(Y) = {\\text{Var}}_a\\{Y - E_a(Y\\mid L,W)\\} + {\\text{Var}}_a\\{ E_a(Y\\mid L,W) - E_a(Y \\mid W) \\} + {\\text{Var}}\\{E_a(Y\\mid W)\\}, \\label{var_decomp_simple}$$ as proved in Section \\[proof:lem:vardecomp\\] in the Supplementary Material. ",
"The last term in the display above motivates the following definition of the proportion of the variance in $Y$ explained by $W$ (summed across arms): $$R^2_W = \\frac{\\sum_{a\\in\\{0,1\\}} {\\text{Var}}\\{E_a(Y \\mid W)\\} }{\\sum_{a\\in\\{0,1\\}} {\\text{Var}}_a(Y)} = 1 - \\frac{\\sum_{a\\in\\{0,1\\}} {\\text{Var}}_a\\{Y - E_a(Y \\mid W)\\} }{\\sum_{a\\in\\{0,1\\}} {\\text{Var}}_a(Y)} \n.$$ Similarly, the middle term in (\\[var\\_decomp\\_simple\\]) motivates the following definition of the proportion of additional variance in $Y$ explained by $L$ after accounting for $W$ (summed across arms): $$R^2_{L \\mid W} = \\frac{\\sum_{a\\in\\{0,1\\}} {\\text{Var}}_a\\{E_a(Y \\mid L, W) - E_a(Y \\mid W)\\} }{\\sum_{a\\in\\{0,1\\}} {\\text{Var}}_a(Y) }. ",
"\\label{def:RsquaredLgivenW}$$ A RAL estimator of $\\Delta$ is efficient (at a given data generating distribution) if it achieves the asymptotic variance lower bound (\\[eq:thm:var-bound-ate\\]) in the semiparametric model.",
"\n\n\\[cor:re-ate\\]\n\nSuppose Assumptions \\[assump:randomization\\], \\[assump:independent-censoring\\], and \\[assump:monotone-censoring\\] hold. ",
"Consider any analysis time. ",
"The ARE between any efficient RAL estimator of $\\Delta$ and the unadjusted estimator is $${\\text{ARE}}= \\frac{1}{ 1 + (p_y/2) \\gamma - R^2_W - (1 - p_y / p_l) R^2_{L\\mid W} }. ",
"\\label{eq:re-ate}$$\n\nThe denominator of the right side of the above display never exceeds 1, which follows since $\\gamma \\leq 2 R^2_W$ by the Cauchy-Schwarz inequality.",
"\n\nImpact of Baseline Variables on Precision Gain {#sec:impactbaseline}\n----------------------------------------------\n\nIn order to isolate the impact of baseline variables $W$ on the ARE , we consider the case where there is no impact of $L$, i.e., when $R^2_{L \\mid W} = 0$. This is the case, for example, when $L$ is independent of $Y$ given $A$ and $W$. Then the ${\\text{ARE}}$ from adjusting for baseline variables $W$ depends on three factors: the proportion of variance in $Y$ explained by $W$ ($R^2_W$), the proportion of participants with $Y$ observed among those enrolled ($p_y$), and the treatment effect heterogeneity ($\\gamma$). ",
"We plot in Figure \\[fig:are-w-ate\\] the ${\\text{ARE}}$ versus $p_y$, considering different combinations of $R^2_W$ and $\\gamma$.\n\n![",
"Asymptotic relative efficiency between an efficient estimator and the unadjusted estimator for estimating $\\Delta$, when only the baseline variable $W$ is prognostic ($R^2_{L \\mid W} = 0$ in all curves).[]{data-label=\"fig:are-w-ate\"}](REW_ATE_onlylinetype.png){width=\"80.00000%\"}\n\nA summary of what is happening in Figure \\[fig:are-w-ate\\] is that with $p_y=1$ and $R^2_W$ fixed, greater treatment effect heterogeneity ($\\gamma$) lessens the precision gain from prognostic baseline variables. ",
"However, with fixed $R^2_W>0$ and $\\gamma>0$, some precision gain is restored as we decrease $p_y$ from $1$ to $0$ (i.e., as we move from right to left along any line with $\\gamma>0$ in Figure \\[fig:are-w-ate\\]). ",
"We describe the intuition for these phenomena below. ",
"The key idea is that precision gains result from adjusting for chance imbalances in $W$ both (i) between study arms among the participants with $Y$ observed, and (ii) between the full sample and the subset with $Y$ observed. ",
"Whether the adjustments in (i) and (ii) lead to precision gains or not is determined by the treatment effect heterogeneity $\\gamma$, which influences whether the estimated means in each arm are adjusted in opposite directions (leading to variance reduction) or in the same direction (leading to cancellation and no impact). ",
"This is analogous to constructive versus destructive interference between waves, which can lead to cancellation or amplification as the waves come together.",
"\n\nTo give intuition for the phenomena in Figure \\[fig:are-w-ate\\], for the remainder of this subsection we fix $R^2_W$ and consider the simple case of a single, binary-valued $W$ representing being less than 65 years old at baseline, and primary outcome $Y$ being the indicator of having mRS at most 3 (a good outcome) at 180 days. ",
"We focus on the stratum $W=1$ and assume that the under 65 subset of the study population would have better outcomes on average than the overall study population if all were assigned to the control arm, i.e., $E(Y|A=0,W=1)-E(Y|A=0)>0$. We next consider $p_y=1$ and the opposite extremes of no treatment effect heterogeneity (where efficiency gains from adjusting for $W$ are maximal) and maximum treatment effect heterogeneity (where there are no efficiency gains from adjusting for $W$).",
"\n\nFirst, consider the case of $p_y=1$ and no treatment effect heterogeneity ($\\gamma=0$), i.e., $E(Y|A=1,W)-E(Y|A=0,W)$ equals the constant $E(Y|A=1)-E(Y|A=0)=\\Delta$. Combining this with the assumption above that those under 65 have better outcomes on average than the overall study population under assignment to the control arm ($E(Y|A=0,W=1)-E(Y|A=0)>0$), it follows that those under 65 also have better outcomes on average than the overall study population under assignment to the [*treatment*]{} arm, i.e., $E(Y|A=1,W=1)-E(Y|A=1)=E(Y|A=0,W=1)-E(Y|A=0)$. If by chance there are proportionally more under 65 participants assigned to the treatment arm than the control arm, then the unadjusted estimator of the mean outcome in the treatment (control) arm is biased upward (downward) conditional on the chance imbalance [@jiang2019robust]. ",
"An efficient estimator adjusts for this by decreasing the unadjusted estimate of $E(Y|A=1)$ and by increasing the unadjusted estimate of $E(Y|A=0)$; the net impact is to decrease the estimate of $\\Delta$. Alternatively, if by chance there are proportionally fewer under 65 participants assigned to the treatment arm, then by a symmetric argument an efficient estimator adjusts for this by increasing the unadjusted estimate of $\\Delta$. The overall impact of such adjustments across many hypothetical trials, is to remove the variance caused by chance imbalances across arms in the stratum $W=1$. This is, intuitively, why there is an efficiency gain when $\\gamma=0$.\n\nNow consider $p_y=1$ and the opposite extreme of maximum treatment effect heterogeneity ($\\gamma=2R^2_W$), which occurs when $E(Y|A=1,W)+E(Y|A=0,W)$ equals the constant $E(Y|A=1)+E(Y|A=0)$. An analogous argument (given in Section \\[appen:adjustmentintuition\\] of the Supplementary Material) to the previous paragraph shows that adjustment for chance imbalances in $W$ leads to equal changes (in the same direction) in the estimated means in each arm; the net impact is that the difference between estimated means across arms is not changed (due to cancellation). ",
"Adjustment has no impact in this case, and leads to no variance reduction.",
"\n\nDefine the chance imbalance between arms in the proportion of participants with $W=1$ as $\\widehat{I}=\\sum_{i=1}^n A_iW_i / \\sum_{i=1}^n A_i - \\sum_{i=1}^n (1-A_i)W_i / \\sum_{i=1}^n (1-A_i)$. E.g., $\\widehat{I}>0$ means that there are proportionally more over 65 participants in the treatment arm than the control arm. ",
"An efficient estimator will adjust for the resulting conditional bias (see e.g., [@jiang2019robust]) by decreasing the unadjusted estimate of the treatment arm mean outcome $E(Y|A=1)$ by $E(Y|A=1,W=1)*\\widehat{I}$. Analogously, an efficient estimator will increase the unadjusted estimator of the control arm mean outcome $E(Y|A=0)$ by $E(Y|A=0,W=1)*\\widehat{I}$. The net impact on the efficient estimator of the average treatment effect $\\Delta=E(Y|A=1)-E(Y|A=0)$ is to subtract $\\{E(Y|A=1,W=1)+E(Y|A=0,W=1)\\}\\widehat{I}$ from the unadjusted estimator. ",
"In this way, the efficient estimator adjusts for chance imbalances in $W$ across arms by subtracting $\\{E(Y|A=1,W=1)+E(Y|A=0,W=1)\\}|\\widehat{I}|$ for trials with $\\widehat{I}>0$ and adding $\\{E(Y|A=1,W=1)+E(Y|A=0,W=1)\\}|\\widehat{I}|$ for trials with $\\widehat{I}<0$. Removing this variation, which is due to imbalances in $W$ across arms and not due to the treatment effect, from the unadjusted estimator leads to a variance reduction of $Var\\{E(Y|A=1,W)+E(Y|A=0,W)\\}$, asymptotically, in the efficient estimator [@jiang2019robust]. ",
"We next relate this characterization of the asymptotic variance reduction to the impact of $R^2_W$, $p_y$, and $\\gamma$ on the ${\\text{ARE}}$.\n\nGiven fixed $R^2_W>0$, $p_y=1$, and $Var_0(Y)=Var_1(Y)=1/2$, the variance reduction due to adjusting for $W$ is maximal when there is no treatment effect heterogeneity, i.e., when $E_1(Y|W)-E_0(Y|W)$ is the same (equal to $\\Delta$) across all strata of $W$. This is because of the relationship $Var\\{E(Y|A=1,W)+E(Y|A=1,W)\\}= 2R^2_W - \\gamma$, i.e., the variance reduction from adjusting for baseline variables $W$ as described above equals twice the variance in $Y$ explained by $W$ minus the treatment effect heterogeneity. ",
"This relationship implies that greater treatment effect heterogeneity leads to lower precision gains. ",
"In the case where $\\gamma$ is the maximum possible value, i.e., $2R^2_W$, there is no variance reduction due to adjustment. ",
"This case corresponds to $E(Y|A=1,W)=-E(Y|A=0,W)$, i.e., the adjustment $\\{E(Y|A=1,W)+E(Y|A=0,W)\\} \\widehat{I}$ in the previous paragraph is always $0$ (and so has no impact). ",
"Above, we considered $p_y=1$, i.e., no pipeline participants. ",
"We now fix $R^2_W>0$ and $\\gamma>0$, and consider the impact of decreasing $p_y$ from $1$ to $0$. This leads to increased ${\\text{ARE}}$ because an efficient estimator can extract information from pipeline participants if $\\gamma>0$; in contrast, the unadjusted estimator does not use any information from pipeline participants. ",
"Intuition for why $\\gamma>0$ is necessary in order for pipeline participant information to be useful is given in Section \\[appen:adjustmentintuition\\] of the Supplementary Material.",
"\n\nImpact of a Short-term Outcome on Precision Gain {#impactshortterm}\n------------------------------------------------\n\nTo isolate the impact of a short-term outcome $L$ on precision gain, we set $R^2_W = 0$ in so that only $L$ is prognostic. ",
"The precision gain from adjusting for short-term outcome $L$ depends on two factors: the proportion of variance in $Y$ explained by $L$ after accounting for $W$ ($R^2_{L \\mid W}$), and the proportion of participants with $Y$ observed among those with $L$ observed ($p_y / p_l$). ",
"In Figure \\[fig:are-l-ate\\], we plot the ${\\text{ARE}}$ against $p_y / p_l$, and consider two values of $R^2_{L \\mid W}$.\n\n![",
"Asymptotic relative efficiency between an efficient estimator and the unadjusted estimator for estimating $\\Delta$, when only the short-term outcome $L$ is prognostic ($R^2_W = 0$ in both curves). ",
"The solid line corresponds to a higher prognostic value $R^2_{L \\mid W}=0.25$ and the dashed line corresponds to a lower prognostic value $R^2_{L \\mid W}=0.1$. []{data-label=\"fig:are-l-ate\"}](REL_ATE.png){width=\"48.00000%\"}\n\nSmaller $p_y/p_l$ and larger $R^2_{L \\mid W}$ generally increase the precision gain from adjusting for $L$; this is because $L$ adds value just for participants with $L$ but not $Y$ observed, and $R^2_{L \\mid W}$ quantifies the prognostic value of $L$ beyond the variance in $Y$ explained by $W$. When there are no pipeline participants, which would occur at the end of a trial with no early stopping or dropout, we have $p_y/p_l = 1$ and the ${\\text{ARE}}$ is 1, i.e., adjusting for $L$ is useless. ",
"This is because adjusting for $L$ helps by, in each arm separately, accounting for chance imbalances between the participants with $Y$ observed and the participants with $L$ but not $Y$ observed; when $p_y/p_l = 1$, these groups are identical and no adjustment can be made.",
"\n\nRecall that we are operating in the semiparametric model defined by Assumptions 1-3; in particular, we are not making any assumptions about the relationships among the variables $(Y,A,L)$. Therefore, we cannot adjust for observed imbalances in $L$ between arms $A \\in \\{0,1\\}$, since these imbalances may be due to the impact of arm assignment. ",
"Given fixed $R^2_{L \\mid W}>0$, larger $p_y / p_l$ decreases the precision gain from adjusting for $L$, as seen in Figure \\[fig:are-l-ate\\]. ",
"In a trial with constant enrollment rate and $d_L < d_Y$, the precision gain from adjusting for $L$ typically attenuates at later stages of a group sequential design due to $p_y / p_l$ being nondecreasing over time. ",
"In the case where missingness is only due to administrative censoring, $p_y / p_l$ starts at $0$ at time $d_L$ and eventually becomes 1 after enrollment stops and enough time has elapsed for all enrolled participants to have $Y$ observed.",
"\n\nComparison of Equally Prognostic Baseline Variable and Short-term Outcome {#impactcomparison}\n-------------------------------------------------------------------------\n\nFor any $q: 0 < q \\leq 1$, we compare the ${\\text{ARE}}$ between two cases: $R^2_W = q, R^2_{L \\mid W} = 0$ (only baseline variable prognostic) and $R^2_W = 0, R^2_{L \\mid W} = q$ (only short-term outcome prognostic). ",
"The ${\\text{ARE}}$ in the former case is larger or equal to that in the latter case. ",
"Equality occurs if and only if $p_l=1$ and treatment effect heterogeneity in the first case is the maximum possible ($\\gamma=2R^2_W=2q)$, as proved in Section \\[appen:proof-section3.4\\] of the Supplementary Material. ",
"Intuition for this is given in Section \\[appen:adjustmentintuition\\] of the Supplementary Material. ",
"The equality can be seen graphically in that the two lines in Figure \\[fig:are-l-ate\\] (corresponding to prognostic value $0.25$ and $0.1$) are identical to the two corresponding lines in Figure \\[fig:are-w-ate\\] with $\\gamma=2R^2_W$; the horizontal axes of the two figures are identical in the special case of $p_l=1$.\n\nEstimator that Adjusts for Baseline Variables and Short-term Outcomes {#sec:tmle}\n=====================================================================\n\nTargeted Minimum Loss-Based Estimator\n-------------------------------------\n\nThe targeted minimum loss-based estimator (TMLE) of [@Gruber2012TMLE], which builds on the ideas of , is implemented in the package [@ltmlePackage]. ",
"We use it to estimate $\\Delta$ and call it the *adjusted estimator*.",
"\n\nOur implementation of this TMLE for a binary outcome $Y$ involves fitting logistic regression working models. ",
"Such models are fit first for the conditional probabilities of censoring given the observed history before censoring: $P(C^L=1 \\mid A, W)$ and $P(C^Y=1 \\mid L, A, W)$. We also fit such a model for study arm assignment $P(A=1 \\mid W)$. Lastly, we fit logistic regression models for $E(Y| L,A,W)$ and for $E(Y| A,W)$, given the corresponding observed histories; we refer to these as outcome regression models. ",
"The last model is fit using sequential regression, an idea from [@Robins2000TMLE; @BangRobins2005TMLE]; see our R code (Github link given at end of paper) for the implementation of the software of [@ltmlePackage] that we used. ",
"All models involve an intercept and main terms for each regressor variable. ",
"Each regression is performed using all participants for whom the relevant variables are uncensored. ",
"For a continuous-valued outcome $Y$, linear outcome regression models could be used or, if the outcome is bounded then it can be rescaled to the interval $[0,1]$ and logistic regression can be used as described by [@grubervanderLaan2010]. ",
"The adjusted estimator initially uses the aforementioned regression model fits. ",
"It then updates each outcome regression model by adding a new covariate built from the other model fits, as described by [@Gruber2012TMLE]. ",
"The final estimator of the average treatment effect $\\Delta$ is based on the updated outcome regression model fit of $E(Y|A,W)$. It is computed by first generating a prediction $Y_{a,i}$ of the outcome $Y$ under each hypothetical assignment to arm $a \\in \\{0,1\\}$ for each participant $i$ by substituting $W=W_i$ and $A=a$ into this model fit. ",
"The estimator of $\\Delta$ is $(1/n)\\sum_{i=1}^n (Y_{1,i}-Y_{0,i})$.\n\nAssumptions 1-3 from Section \\[sec:notation\\] imply that the censoring models and the model for $P(A=1|W)$ are correctly specified. ",
"The results of @Gruber2012TMLE [Section 4] imply that the adjusted estimator is a consistent estimator of the average treatment effect $\\Delta$. This holds regardless of the correctness of the outcome regression models. ",
"That is, these models could be arbitrarily misspecified and still the adjusted estimator converges in probability to $\\Delta$ as sample size goes to infinity. ",
"If the outcome regression models are correct, then the adjusted estimator achieves the asymptotic variance lower bound in (\\[eq:thm:var-bound-ate\\]), in which case it is semiparametric, locally efficient [@Gruber2012TMLE Section 4]. ",
"Other estimators with the above properties include the augmented, inverse probability of treatment (and censoring) estimators of @Robins2000TMLE and @BangRobins2005TMLE.",
"\n\nApproximating the Relative Efficiency between the Adjusted Estimator and the Unadjusted Estimator {#subsec:are-fitted}\n-------------------------------------------------------------------------------------------------\n\nWe focused on the asymptotic relative efficiency between an efficient estimator and the unadjusted estimator in Section \\[sec:theory-ate\\]. ",
"To connect our results regarding efficient estimators to the adjusted estimator, we use results from @Gruber2012TMLE [Section 5.3]. ",
"These results imply that if all regression models used in the TMLE are correctly specified, then plugging the corresponding model fits into the formulas (\\[def:gamma\\])-(\\[def:RsquaredLgivenW\\]) and then using these estimates of $\\gamma,R^2_W,R^2_{L \\mid W}$ in (\\[eq:re-ate\\]) results in a consistent estimator of the ${\\text{ARE}}$ (called the plug-in estimator). ",
"This estimator of the ${\\text{ARE}}$ is asymptotically conservative, meaning that the ${\\text{ARE}}$s may be underestimated but not overestimated, if the models for censoring and for $P(A=1|W)$ are correctly specified; since these models are correct under Assumptions \\[assump:randomization\\]-\\[assump:monotone-censoring\\], it follows that the plug-in estimator can serve as a (possibly conservative) asymptotic approximation to the ${\\text{ARE}}$ between the adjusted estimator and the unadjusted estimator.",
"\n\nDenote by $\\widehat{E}(Y \\mid X=x)$ the predicted value of $E(Y \\mid X=x)$ from a model fit. ",
"Let $\\widehat{{\\text{Var}}}(\\cdot)$ denote the sample variance over participants from both arms, and $\\widehat{{\\text{Var}}}_a(\\cdot)$ denote the sample variance over participants from arm $a$. We estimate $R^2_W$, $R^2_{L\\mid W}$, and $\\gamma$, respectively, by $$\\widehat{R}^2_W = \\frac{ \\sum_{a\\in\\{0,1\\}} \\widehat{{\\text{Var}}}\\{ \\widehat{E}(Y \\mid W, A=a) \\}}{ \\sum_{a\\in\\{0,1\\}} \\widehat{{\\text{Var}}}_a(Y) },$$ $$\\widehat{R}^2_{L \\mid W} = \\frac{ \\sum_{a\\in\\{0,1\\}} \\widehat{{\\text{Var}}}_a\\{\\widehat{E}(Y \\mid W, L, A=a) - \\widehat{E}(Y \\mid W, A=a)\\} }{ \\sum_{a\\in\\{0,1\\}} \\widehat{{\\text{Var}}}_a(Y) },$$ $$\\widehat\\gamma = \\frac{\\widehat{\\text{Var}}\\{ \\widehat{E}(Y \\mid W, A=1) - \\widehat{E}(Y \\mid W, A=0) \\}}{\\sum_{a\\in\\{0,1\\}} \\widehat{{\\text{Var}}}_a(Y)}.$$ In Sections \\[sec:application\\]-\\[sec:simulation\\] when calculating the conditional expectations $\\widehat{E}$ in the formulas for $\\widehat{R}^2_W$ and $\\widehat{R}^2_{L \\mid W}$, we used logistic regression model fits with main terms. ",
"When calculating $\\widehat\\gamma$, we used similar model fits except including interactions, since we wanted to give an opportunity to detect treatment effect heterogeneity. ",
"The estimated $R$-squared quantities above are relevant for continuous and binary outcomes in measuring the precision gain from adjustment when estimating $\\Delta$; this generalizes an idea from @moore2009binary, who give an R-squared formula related to $\\widehat{R}^2_W$ except that it marginalizes over $A$.",
"\n\nApplying Precision Gain Formula to Trial Data {#sec:application}\n=============================================\n\nTrial Example: MISTIE-II {#subsec:mistie}\n------------------------\n\nMISTIE-II is a Phase II randomized clinical trial evaluating a new surgical treatment for intracerebral hemorrhage. ",
"The treatment is called Minimally-Invasive Surgery Plus rt-PA for Intracerebral Hemorrhage Evacuation, abbreviated as MISTIE [@MISTIE_trial2008]. ",
"In the MISTIE-II dataset, the primary and short-term outcomes of each participant correspond to the Modified Rankin Scale (mRS), measured at different times after enrollment. ",
"We use the variables in Table \\[tab:MISTIEvars\\]. ",
"The primary outcome $Y$ is a binary indicator of a successful outcome ($\\mbox{mRS}\\leq3$) at 180 days after enrollment. ",
"The short-term outcomes $L^{(1)}$ and $L^{(2)}$ are the indicators of mRS no larger than 4 at 30 and 90 days after enrollment, respectively. ",
"The cutoff 4 for $L^{(1)}$ and $L^{(2)}$ was chosen because there were relatively few participants with mRS 3 or less at 30 or 90 days after enrollment. ",
"The treatment assignment indicator $A$ denotes whether a participant is assigned to the surgical treatment ($A=1$) or to standard of care ($A=0$). ",
"Baseline variables $W^{(1)}$-$W^{(4)}$ are age (dichotomized at 65 years), NIHSS (NIH Stroke Scale total score, quantifying stroke-caused impairment), ICH (intracerebral hemorrhage volume), and GCS (Glasgow Coma Scale), all measured before randomization. ",
"The baseline variables (except for age) are treated as continuous variables in the regression models. ",
"The dataset has 100 participants.",
"\n\n Name Description \n ----------- ----------------------------------------------------------- --\n $W^{(1)}$ baseline age (years), dichotomized at $\\leq 65$ \n $W^{(2)}$ baseline NIHSS, ordinal \n $W^{(3)}$ baseline ICH, ordinal \n $W^{(4)}$ baseline GCS, ordinal \n $A$ treatment indicator, 1 being MISTIE \n $L^{(1)}$ mRS at 30 days, dichotomized by thresholding at $\\leq 4$ \n $L^{(2)}$ mRS at 90 days, dichotomized by thresholding at $\\leq 4$ \n $Y$ mRS at 180 days, dichotomized by thresholding at $\\leq 3$ \n\n : Variables that we use from the MISTIE-II dataset. ",
"Ordinal-valued basline variables below are treated as continuous in our regression models. []{",
"data-label=\"tab:MISTIEvars\"}\n\nWe let $W^\\text{full} = (W^{(1)}, W^{(2)}, W^{(3)}, W^{(4)})$ and $L^\\text{full} = (L^{(1)}, L^{(2)})$. These variables are used in constructing data generating distributions for our simulations in Section \\[sec:simulation\\]. ",
"However, only the smaller subsets of variables $W = (W^{(1)}, W^{(4)})$ and $L = L^{(1)}$ are made available to the adjusted estimator (both in Section \\[subsec:application-precision-gain\\] and \\[sec:simulation\\]). ",
"We included the “extra” variables $W^{(2)}, W^{(3)}, L^{(2)}$ (which the adjusted estimator does not have access to) in our data generating distributions for simulations so that the regression models used by the adjusted estimator will be misspecified; this was done since we expect at least some misspecification to occur in practice.",
"\n\nApproximate Precision Gain by Using the Adjusted Estimator on MISTIE-II Data {#subsec:application-precision-gain}\n----------------------------------------------------------------------------\n\nApplying the method in Section \\[subsec:are-fitted\\] to the MISTIE-II data, the estimated R-squared quantities are $\\widehat{R}^2_W = 0.36$, $\\widehat{R}^2_{L \\mid W} = 0.08$, and $\\widehat\\gamma = 0.02$. This indicates that the baseline variables are strongly prognostic, that (after accounting for the baseline variables) the short-term outcome $L$ is only mildly prognostic, and that there is little (if any) treatment effect heterogeneity. ",
"At the end of a trial, if every participant has $Y$ observed, we have $p_l = p_y = 1$. Substituting these and the estimated values $\\widehat{R}^2_W, \\widehat{R}^2_{L \\mid W}, \\widehat{\\gamma}$ into (\\[eq:re-ate\\]) gives an approximate ${\\text{ARE}}$ of $1.53$ comparing the TMLE versus the unadjusted estimator. ",
"If instead there had been no treatment effect heterogeneity ($\\widehat \\gamma = 0$), then the approximate ${\\text{ARE}}$ would be slightly larger ($1.55$), assuming $\\widehat R^2_W$ and $\\widehat R^2_{L \\mid W}$ remain fixed. ",
"Now consider a hypothetical interim analysis where 80% of the enrolled participants have $Y$ observed and 95% of the enrolled participants have $L$ observed (e.g., at the 3rd interim analysis of the group sequential design with ${n_{\\rm max}}= 480$ considered in Section \\[subsec:simulationresults-2\\]; see Table \\[tab:sample-size\\] in the Supplementary Material). ",
"Then we have $p_y = 0.8$, $p_l = 0.95$, and the approximate ${\\text{ARE}}$ is $1.56$.\n\nEquivalently, in each of the three aforementioned situations, the adjusted estimator approximately reduces the required sample size to achieve a desired power by $35\\%$, $36\\%$, and $36\\%$, respectively, compared to the unadjusted estimator. (",
"The asymptotic sample size reduction is computed as $1-1/{\\text{ARE}}$ as described in Section \\[sec:theory-singlearm\\] of the Supplementary Material.) ",
"These are all quite similar, which is primarily due to the relatively small values of $\\widehat \\gamma$ and $\\widehat R^2_{L \\mid W}$. The magnitudes of the AREs above are not unusual in the MISTIE-II population–see @optimising2009should–and result from some baseline variables (especially NIHSS) being strongly prognostic for the outcome. ",
"Approximations of the ${\\text{ARE}}$ as above can be applied to trial planning, where one may use previous data to estimate the variance explained by baseline variables and short-term outcomes and the treatment effect heterogeneity, and use to get a rough projection of the precision gain from adjusting for baseline variables and short-term outcomes. ",
"If treatment effect heterogeneity is difficult to estimate *a priori*, one could use $\\gamma = 0$ as the best case scenario (no treatment effect heterogeneity, corresponding to maximal precision gain), with the expectation that $\\gamma > 0$ will result in attenuated precision gain. ",
"A conservative approach is to assume no precision gain when selecting the initial sample size per stage, but then use preplanned sample-size re-estimation (using the estimated variance) to potentially shrink the per-stage sample size to reflect the estimated precision gains based on accrued data. ",
"Such adaptations based on estimating a nuisance parameter are generally acceptable to regulators such as the U.S. Food and Drug Adminsitration [@fda:2016; @FDA]. ",
"This approach has important limitations given in Section \\[sec:discussion\\].",
"\n\nSimulations of Group Sequential Design {#sec:simulation}\n======================================\n\nData Generating Distributions for Simulated Trials Based on MISTIE-II {#subsec:dgm}\n---------------------------------------------------------------------\n\nWe conducted simulation studies to assess how good of an approximation is provided by the precision gain formula (\\[eq:re-ate\\]), and to examine the sample size reduction by using the adjusted versus unadjusted estimator in a group sequential design. ",
"In order to mimic key features of the MISTIE-II trial, we use a resampling-based algorithm to generate participants for our simulated trials. ",
"This algorithm generates simulated trials with the following properties:\n\n1. ",
" The treatment assignment is independent of baseline variables.",
"\n\n2. ",
" The relative efficiency between the adjusted estimator and the unadjusted estimator is similar to that calculated directly from the MISTIE-II dataset.",
"\n\nWe define eight data generating distributions, called “settings”. ",
"The goal is to consider four situations involving $W$ and/or $L$ being prognostic or not for $Y$; these are denoted by ${progn_{W,L}}$ (both prognostic), ${progn_{W}}$ (only $W$ prognostic), ${progn_{L}}$ (only $L$ prognostic), and ${progn_{\\emptyset}}$ (neither prognostic). ",
"For each of these four situations (called “prognostic settings”), we construct distributions with the following two average treatment effects: $\\Delta=0$ (no effect) or $\\Delta=0.122$ (benefit equal to the unadjusted estimate from the MISTIE-II dataset). ",
"Details of the data generating algorithm for the eight settings are in Section \\[appen:dgm\\] of the Supplementary Material. ",
"Table \\[tab:rsq-mistie\\] gives the approximate $R$-squared and $\\gamma$ values for the eight settings. ",
"The only source of missing data in our simulations is administrative censoring, i.e., participants who enrolled but have not been in the trial long enough to have outcomes measured.",
"\n\n ------------------ ------------ ---------------- -- ------------ ---------------- -- ------------ ---------------- -- ------------ ----------------\n \n $\\Delta=0$ $\\Delta=0.122$ $\\Delta=0$ $\\Delta=0.122$ $\\Delta=0$ $\\Delta=0.122$ $\\Delta=0$ $\\Delta=0.122$\n $R^2_{W}$ 0.35 0.36 0.35 0.36 0 0 0 0\n $R^2_{L \\mid W}$ 0.08 0.07 0 0 0.30 0.30 0 0\n $\\gamma$ 0 0.01 0 0.01 0 0 0 0\n ------------------ ------------ ---------------- -- ------------ ---------------- -- ------------ ---------------- -- ------------ ----------------\n\n : Approximate $R$-squared and $\\gamma$ for each of the eight data generating distributions (settings) in our simulation study, computed by simulating $10^6$ participants under each distribution and computing $\\widehat{R}^2_W, \\widehat{R}^2_{L \\mid W}, \\widehat{\\gamma}$, respectively.[]{data-label=\"tab:rsq-mistie\"}\n\nGroup Sequential Trial Design Analysis Timing and Early Stopping Rule {#subsec:design}\n---------------------------------------------------------------------\n\nWe assume that $d_L = 30$ days and $d_Y = 180$ days. ",
"The goal is to control Type I error at level 0.025 and to have at least 80% power to reject the null hypothesis $H_0: \\Delta \\leq 0$ when $\\Delta = 0.122$.\n\nWe used the group sequential design framework in @hampson2013group, which involves interim analyses where a choice is made to stop or continue enrollment; if enrollment is stopped, then one waits until all pipeline participants complete the trial and then a hypothesis test (called a decision analysis) is conducted. ",
"Full details of the design are in Supplementary Material \\[appen:gst-multiple-testing\\]. ",
"Denote by $K$ the total number of stages. ",
"We set $K=5$ stages in our simulated trials. ",
"Participants were enrolled at the constant rate of 140 participants per year based on the projection for the enrollment rate in the MISTIE-III trial [@MistieIIITrial]. ",
"Given the maximum sample size ${n_{\\rm max}}$, the timing of interim analyses is chosen such that for $1\\leq k \\leq 4$, at the $k$-th interim analysis there are $(k/K){n_{\\rm max}}$ participants with $Y$ observed.",
"\n\nFor either estimator the corresponding Wald statistics at different analysis times have (asymptotically) a multivariate normal distribution, where the information at a given analysis time is the reciprocal of the estimator variance. (",
"This relies on Assumptions 1-3 and holds regardless of the correctness of the outcome regression models used in the adjusted estimator.) ",
"The covariance matrix can be estimated, e.g., using the nonparametric bootstrap. ",
"To reduce the computational burden in our simulation studies, for any data generating distribution and ${n_{\\rm max}}$ we precompute the covariance matrix of our statistics at different analysis times, which is then used to determine stopping boundaries. ",
"The efficacy/futility stopping boundaries at each stage are calculated using the error spending approach presented in Supplementary Material \\[appen:gst-boundary\\]. ",
"We used error spending functions $f(t) = 0.025\\min(t^2,1)$ for Type I error and $g(t) = 0.2\\min(t^2,1)$ for Type II error, where $t$ denotes the information time (observed information divided by maximum information).",
"\n\nFor each combination of prognostic setting (${{\\rm progn}_{W,L}}, {{\\rm progn}_{W}}, {{\\rm progn}_{L}}, {{\\rm progn}_{\\emptyset}}$) and estimator (unadjusted or adjusted), we use binary search to find the minimum total sample size ${n_{\\rm max}}$ required to achieve approximately 80% power under $\\Delta=0.122$ at $0.025$ Type I error using the group sequential design. ",
"In practice, this could be approximated from accruing data by estimating the variance of the adjusted estimator and then using sample size re-estimation to set ${n_{\\rm max}}$. However, we assumed that this was known in order to reduce the computational burden. ",
"For the unadjusted estimator, in every setting we let ${n_{\\rm max}}$ be 480. ",
"For the adjusted estimator, ${n_{\\rm max}}$ is set to be 300 under all settings with ${{\\rm progn}_{W,L}}$ and ${{\\rm progn}_{W}}$, and 480 under all settings with ${{\\rm progn}_{L}}$ and ${{\\rm progn}_{\\emptyset}}$. We simulated 50,000 trials under each of the eight settings. ",
"The number of accrued participants, error spending functions, and the stopping boundaries for each analysis are listed in Table \\[tab:error-boundary\\] in the Supplementary Material.",
"\n\nSimulation Results: Sample Size Reduction from Adjustment {#subsec:simulationresults-2}\n---------------------------------------------------------\n\nTable \\[tab:poweress\\] lists the simulation-based ${n_{\\rm max}}$, Type I error (obtained under $\\Delta=0$), power (obtained under $\\Delta=0.122$), and the expected sample size (ESS) for each of the eight settings for each estimator. ",
"For each setting, the empirical Type I error rate is controlled at 0.025 and the power is approximately 80%. ",
"ESS is calculated as the number of enrolled participants when the trial stops, averaging over 50,000 simulations. ",
"The performance of the unadjusted estimator is the same under all prognostic settings and is summarized in the first row.",
"\n\nCompared to the unadjusted estimator, the adjusted estimator substantially reduces the sample size when the baseline variables are prognostic. ",
"In particular, comparing the first row (unadjusted) versus the third row (adjusted ${{\\rm progn}_{W}}$), the adjusted estimator reduces the maximum sample size from 480 to 300, the expected sample size under $\\Delta=0$ from 318 to 227, and the expected sample size under $\\Delta=0.122$ from 382 to 260. ",
"The sample size reduction due to the prognostic short-term outcome is very small. ",
"When neither $W$ nor $L$ is prognostic (${{\\rm progn}_{\\emptyset}}$), performance of the two estimators is similar.",
"\n\n Estimator Progn. ",
"set. ",
" ${n_{\\rm max}}$ Type I error Power ESS($\\Delta=0$) ESS($\\Delta=0.122$)\n ------------ ----------------------------- ----------------- -------------- ------- ----------------- ---------------------\n unadjusted - 480 0.0250 0.811 318 382\n \n adjusted ${{\\rm progn}_{W,L}}$ 300 0.0254 0.791 225 259\n ${{\\rm progn}_{W}}$ 300 0.0256 0.805 227 260\n ${{\\rm progn}_{L}}$ 480 0.0253 0.805 309 375\n ${{\\rm progn}_{\\emptyset}}$ 480 0.0248 0.811 321 384\n\n : The maximum sample size (${n_{\\rm max}}$) and empirical Type I error, power, and expected sample size (ESS) under $\\Delta=0$ and $\\Delta=0.122$ for each estimator under each prognostic setting. ",
"The ${n_{\\rm max}}$ was chosen for each prognostic setting in order to achieve approximately 80% power under $\\Delta=0.122$.[]{data-label=\"tab:poweress\"}\n\nSimulation Results: Relative Efficiency {#subsec:simulationresults}\n---------------------------------------\n\nIn this subsection, we simulated trials with no early stopping, i.e., each simulated trial always enrolls ${n_{\\rm max}}$ participants and continues follow-up until all participants have $Y$ observed. ",
"We then look back at each simulated trial and compute what each estimator’s value would be at each of the interim and decision analysis times. ",
"At any analysis time, the number of participants with $Y$ observed, $L$ observed, and only $(A,W)$ observed, respectively, is fixed; therefore, we can make a direct comparison between the ratio of estimator variances in our simulations and the predicted ${\\text{ARE}}$ from our formula (which assumes constant $p_l,p_y$).",
"\n\nIn order to make direct comparisons between estimators at the same sample size, we change (just for this subsection) the setting of ${n_{\\rm max}}$. The maximum sample size ${n_{\\rm max}}$ is set to be 300 under ${{\\rm progn}_{W,L}}$ and ${{\\rm progn}_{W}}$, and is set to 480 under ${{\\rm progn}_{L}}$ and ${{\\rm progn}_{\\emptyset}}$. Under each setting the same ${n_{\\rm max}}$ is used for both the unadjusted estimator and the adjusted estimator. ",
"Table \\[tab:re\\] lists the approximate, asymptotic relative efficiency (ARE) computed by evaluating (\\[eq:re-ate\\]) but with conditional expectations replaced by empirical estimates as discussed in Section \\[subsec:are-fitted\\], and the relative efficiency (RE) from simulation. ",
"Under most settings, the ARE predicted by the theory is similar to the RE computed from the simulation. ",
"There are some discrepancies between these at the earlier analysis times. ",
"We think that this is due to the relatively small sample sizes at these analysis times compared to the number of variables adjusted for, which can lead to model overfit and so reduced performance of the adjusted estimator. ",
"For example, at interim analysis 1 under ${{\\rm progn}_{W,L}}$ there are only 60 participants with the primary outcome observed but 4 variables get adjusted for in the regression model for $E(Y|W,A,L)$; this violates the rule of thumb that one should have at least 20 observations per term in the regression model. ",
"We discuss possible remedies for this in Section \\[sec:discussion\\].",
"\n\nA comparison between ${{\\rm progn}_{W}}$ (when only $W$ is prognostic) and ${{\\rm progn}_{L}}$ (when only $L$ is prognostic) shows that there is a much larger precision gain in the former case. ",
"This is despite the fact that $W$ and $L$ have roughly similar prognostic values for $Y$ marginally, as shown in Table \\[tab:rsq-mistie\\]. ",
"The reason, as discussed in Sections \\[sec:impactbaseline\\]-\\[impactcomparison\\], is that there is relatively little treatment effect heterogeneity, which means that the baseline variables contribute to precision gains in two ways: allowing adjustment for chance imbalances (i) between study arms among the participants with $Y$ observed, and (ii) between the full sample and the subset with $Y$ observed. ",
"The short-term outcome can only contribute in the second way, and so may lead to smaller gains. ",
"Under ${{\\rm progn}_{L}}$, comparing interim analyses to decision analyses shows that $L$ improves estimation precision only at interim analyses (i.e., when there are pipeline participants), which is in line with our theoretical results.",
"\n\nDiscussion {#sec:discussion}\n==========\n\nWe considered independent censoring, and the only source of missing data in our simulations was administrative censoring. ",
"If outcomes are missing at random [@rubin1976MCAR], then the adjusted estimator may still be consistent under correct specification of certain working models, while the unadjusted estimator may be biased. ",
"If dropout is missing not at random [@rubin1976MCAR], then both estimators may be inconsistent and our formulas will not work.",
"\n\nThe theoretical ARE can be larger than the simulation-based ARE, especially at the early interim analyses. ",
"One reason is that in practice there is a finite sample penalty for adjustment (for each degree of freedom in the models fit) that is not reflected in the theoretical calculations, and that dissipates as sample size grows. ",
"This issue may be mitigated by only scheduling interim analyses for times when at least 50% of the participants have primary outcomes observed, and by constraining the number of variables adjusted for such that one has at least 20 participants with primary outcomes observed per variable.",
"\n\nThe sample size reductions due to improved precision from adjustment came primarily from selecting smaller ${n_{\\rm max}}$. Since the prognostic value of variables is typically not precisely known before the trial starts, one may use a preplanned rule for sample size re-estimation to set ${n_{\\rm max}}$ based on accruing data. ",
"First, the original, maximum sample size is set conservatively assuming no gains from adjustment. ",
"During the trial the asymptotic variance of the adjusted estimator is estimated and ${n_{\\rm max}}$ is set to be the sample size for which the desired power is achieved when $\\Delta$ equals the a priori specified, clinically meaningful, minimum treatment effect. ",
"The sample size re-estimation should only be conducted after a sufficient number of participants have had their primary outcomes measured, e.g., when roughly 50% of the originally planned sample size have this measured.",
"\n\nThis approach is only feasible if the delay time $d_Y$ is not too long compared to the enrollment rate (otherwise the originally planned sample size will all be enrolled before a sufficient number of primary outcomes are observed). ",
"This was feasible in the MISTIE trial context with $d_Y=180$ days and enrollment rate 140 per year, since sample size re-estimation could be done at 2.1 years. ",
"At this time, 230 (48%) of the originally planned sample size (480) have primary outcomes observed. ",
"So restricting to approximately 300 participants (the required ${n_{\\rm max}}$ when $W$ is as prognostic as in the MISTIE-II data) is possible.",
"\n\nA limitation of our simulation study is that we precomputed ${n_{\\rm max}}$ for each setting, in order to save computation time. ",
"In practice, this would be calculated as a function of the estimated variance using accruing data in the trial. ",
"A future area of research is to run simulation studies to evaluate the resulting impact on power and sample size.",
"\n\ncode [@R] for our simulations can be downloaded at <https://github.com/tqian/gst_tmle>.",
"\n\nAcknowledgments {#acknowledgments .unnumbered}\n===============\n\nThis work was supported by the Patient-Centered Outcomes Research Institute (ME-1306-03198), the US Food and Drug Administration (HHSF223201400113C), and NIH grant UL1TR001079. ",
"This work is solely the responsibility of the authors and does not represent the views of the above agencies. ",
"We thank Mary Joy Argo for helpful comments.",
"\n\nSupplementary Material {#supplementary-material .unnumbered}\n======================\n\nIn Section \\[appen:are-sample-size-reduction\\], we discuss the relationship between asymptotic relative efficiency and sample size reduction for a Wald test of a single stage trial. ",
"In Section \\[appen:gst\\], we present the group sequential design used in the simulation studies in Section \\[sec:simulation\\]. ",
"In Section \\[appen:dgm\\], we specify the data generating distribution used in the simulation studies in Section \\[sec:simulation\\]. ",
"Intuition for the impact of treatment effect heterogeneity on precision gains is given in Section \\[appen:adjustmentintuition\\]. ",
"In Section \\[appen:theory-predictive-prognostic\\], we provide additional theoretical results regarding the precision gain from adjusting for prognostic baseline variables, when the baseline variable is either purely predictive or purely prognostic. ",
"In Section \\[sec:theory-singlearm\\], we provide theoretical results regarding the precision gain from adjusting for prognostic baseline variables and the short-term outcome, when the parameter is $E(Y \\mid A=a)$ for each $a\\in\\{0,1\\}$ (rather than the average treatment effect). ",
"Simulations to evaluate the theory in Section \\[sec:theory-singlearm\\] are provided in Section \\[appen:re-h1\\]. ",
"Section \\[appen:sec:proof\\] includes proofs of the results in the main paper as well as results in Section \\[appen:theory-predictive-prognostic\\] and Section \\[sec:theory-singlearm\\]. ",
"Auxiliary lemmas that are used in the proofs in Section \\[appen:sec:proof\\] are themselves proven in Section \\[appen:proof-aux\\].",
"\n\nRelationship between asymptotic relative efficiency and sample size reduction for a Wald test of a single stage trial {#appen:are-sample-size-reduction}\n=====================================================================================================================\n\nConsider one-sided Wald tests of the null hypothesis $H_0:\\Delta \\leq 0$ versus the alternative $H_1:\\Delta > 0$, using RAL estimators $\\widehat{\\Delta}_1$ and $\\widehat{\\Delta}_2$, respectively (where the test statistics are the estimators divided by their standard errors). ",
"The asymptotic relative efficiency can be interpreted as the limit as sample size goes to infinity of the inverse of the ratio of the required sample sizes for the two estimators to achieve a given power at local alternatives [@vdv Section 8.2]. ",
"For example, ${\\text{ARE}}(\\widehat{\\Delta}_1, \\widehat{\\Delta}_2) = 1.2$ means that by using $\\widehat\\Delta_1$ instead of $\\widehat\\Delta_2$, the required sample size is reduced by $1 - 1 / 1.2 \\approx 17\\%$ asymptotically.",
"\n\nFull Description of Group Sequential Design Used in Simulation {#appen:gst}\n==============================================================\n\nComponents of the group sequential design used in simulation are presented in the following order: the multiple testing procedure (Section \\[appen:gst-multiple-testing\\]), the computation of the test boundaries (Section \\[appen:gst-boundary\\]), the sample size at each analysis for the simulated trials (Section \\[appen:sample-size-each-analysis\\]), and the value of Type I error and Type II error spent and the testing boundary at each analysis (Section \\[appen:testing-boundary\\]).",
"\n\nMultiple Testing Procedure {#appen:gst-multiple-testing}\n--------------------------\n\nIn the simulation studies in Section \\[sec:simulation\\] we use the group sequential test from @hampson2013group with $K$ stages. ",
"Given $\\alpha, \\beta \\in (0,1)$, the design goal is to control the Type I error rate at level $\\alpha$ and have power $1-\\beta$ at alternative $\\Delta=\\delta>0$. Such a group sequential test can terminate enrollment at an interim analysis, and if such early stopping happens follow-up continues until all pipeline participants have $Y$ observed before conducting a decision analysis to reject or accept $H_{0}$. For each stage $k$, denote by $S_{k}$ and $\\tilde{S}_{k}$ the test statistics at the $k$th interim analysis (where the decision to stop or continue enrollment occurs) and the $k$th decision analysis (where the hypothesis test is conducted), respectively; let $u_{k}$ and $l_{k}$ denote the efficacy and futility boundaries for terminating enrollment at interim analysis $k$, and let $c_{k}$ denote the critical value for the corresponding decision analysis. ",
"These are used in the group sequential testing procedure below, reproduced from @hampson2013group [Figure 1]:\n\nComputation of Test Boundaries {#appen:gst-boundary}\n------------------------------\n\nFollowing @hampson2013group, consider a Type I error spending function $f(\\cdot)$ and Type II error spending function $g(\\cdot)$, which are non-decreasing with $f\\left(0\\right)=g\\left(0\\right)=0$ and $f\\left(t\\right)=\\alpha$ and $g\\left(t\\right)=\\beta$ for $t\\geq1$. The maximum information level ${\\cal I}_{\\max}$ is chosen depending on the power goal and the error spending functions. ",
"Denote by ${\\cal I}_{k}$ and $\\tilde{{\\cal I}}_{k}$ the information levels at the $k$-th interim analysis and decision analysis, respectively. ",
"Denote by ${\\cal C}_{k}=\\left(l_{k},u_{k}\\right)$ the critical region at interim analysis $k$, $1\\leq k \\leq K$. The test boundaries $u_k$, $l_k$, and $c_k$ are calculated by (12)-(15) in @hampson2013group [Section 4.1.1], and we paraphrase as follows. ",
"Let $u_1$ and $l_1$ be the solutions to $$P(S_1\\geq u_1;\\Delta=0)=f({\\cal I}_1/{\\cal I}_{\\max})\\qquad\\mbox{and}\\qquad P(S_1\\leq l_1;\\Delta=\\delta)=g({\\cal I}_1/{\\cal I}_{\\max}). ",
"\\nonumber$$ For $2\\leq k\\leq K-1$, $u_k$ is the solution to $$P(S_1\\in{\\cal C}_1,\\ldots,S_{k-1}\\in{\\cal C}_{k-1},S_k\\geq u_k;\\Delta=0)=f({\\cal I}_k/{\\cal I}_{\\max})-f({\\cal I}_{k-1}/{\\cal I}_{\\max}), \\label{eq:err_bdry_effic}$$ and $l_k$ is the solution to $$P(S_1\\in{\\cal C}_1,\\ldots,S_{k-1}\\in{\\cal C}_{k-1},S_k\\leq l_k;\\Delta=\\delta)=g({\\cal I}_k/{\\cal I}_{\\max})-g({\\cal I}_{k-1}/{\\cal I}_{\\max}). ",
"\\label{eq:err_bdry_fut}$$ For $1\\leq k\\leq K-1$, the critical value $c_k$ is the solution to $$\\begin{aligned}\nP(S_1\\in{\\cal C}_1,\\ldots,S_{k-1}\\in{\\cal C}_{k-1}, & S_k\\geq u_k,\\tilde{S}_k<c_k;\\Delta=0) \\nonumber \\\\\n& = P(S_1\\in{\\cal C}_1,\\ldots,S_{k-1}\\in{\\cal C}_{k-1},S_k\\leq l_k,\\tilde{S}_k\\geq c_k;\\Delta=0). ",
" \\nonumber $$ And the critical value $c_K$ for the last stage is the solution to $$P(S_1\\in{\\cal C}_1,\\ldots,S_{K-1}\\in{\\cal C}_{K-1},\\tilde{S}_K\\geq c_K;\\Delta=0)=\\alpha-f({\\cal I}_{K-1}/{\\cal I}_{\\max}). ",
" \\nonumber $$\n\nSample Size at Each Analysis {#appen:sample-size-each-analysis}\n----------------------------\n\nTable \\[tab:sample-size\\] lists the sample size and analysis timing of the group sequential designs with ${n_{\\rm max}}=480$ and ${n_{\\rm max}}=300$ used in Section \\[subsec:simulationresults\\]. ",
"For $1\\leq k \\leq 4$, Analysis $k$ indicates interim analysis at stage $k$ and $k^*$ indicates the corresponding decision analysis if enrollment is early stopped at that stage. ",
"Analysis $5^*$ indicates the final decision analysis. ",
"There is not any interim analysis for the final stage. ",
"Fully observed participants are those with $W,L,Y$ observed; partially observed participants are those with $W,L$ but not $Y$ observed. ",
"The three groups of participants listed in Table \\[tab:sample-size\\] are inclusive of all enrollees and mutually exclusive.",
"\n\n[llrrrrrrrrr]{} Analysis & & 1 & $1^*$ & 2 & $2^*$ & 3 & $3^*$ & 4 & $4^*$ & $5^*$\\\n& &\\\nTime (year) & & 1.2 & 1.7 & 1.9 & 2.4 & 2.6 & 3.0 & 3.2 & 3.7 & 3.9\\\n\\# Fully observed ($W,L,Y$) & & 96 & 165 & 192 & 261 & 288 & 357 & 384 & 453 & 480\\\n\\# Partially observed ($W,L$) only & & 57 & 0 & 57 & 0 & 57 & 0 & 57 & 0 & 0\\\n\\# Pipeline with only $W$ observed & & 12 & 0 & 12 & 0 & 12 & 0 & 12 & 0 & 0\\\n\\\n& &\\\nTime (year) & & 0.9 & 1.4 & 1.4 & 1.8 & 1.8 & 2.3 & 2.2 & 2.6 & 2.6\\\n\\# Fully observed ($W,L,Y$) & & 60 & 129 & 120 & 189 & 180 & 249 & 240 & 300 & 300\\\n\\# Partially observed ($W,L$) only & & 57 & 0 & 57 & 0 & 57 & 0 & 57 & 0 & 0\\\n\\# Pipeline with only $W$ observed & & 12 & 0 & 12 & 0 & 12 & 0 & 3 & 0 & 0\\\n\\\n\nType I Error and Type II Error Spent and Testing Boundary {#appen:testing-boundary}\n---------------------------------------------------------\n\nTable \\[tab:error-boundary\\] lists the Type I error per stage $f({\\cal I}_k/{\\cal I}_{\\max})-f({\\cal I}_{k-1}/{\\cal I}_{\\max})$, Type II error per stage $g({\\cal I}_k/{\\cal I}_{\\max})-g({\\cal I}_{k-1}/{\\cal I}_{\\max})$, and the testing boundaries at each stage under different settings used in the simulation studies. ",
"Given any prognostic setting ${{\\rm progn}_{W,L}}, {{\\rm progn}_{W}}, {{\\rm progn}_{L}}, {{\\rm progn}_{\\emptyset}}$ and estimator pair, the designs are the same for $\\Delta=0$ and $\\Delta=0.122$.\n\n[lrrrrr]{} Analysis $(k)$ & 1 & 2 & 3 & 4 & 5\\\n&\\\nType I error per stage $\\times 10^{-3}$ & 0.9 & 3.2 & 5.1 & 7.2 & 8.6\\\nType II error per stage $\\times 10^{-3}$ & 7.5 & 25.4 & 41.1 & 57.3 & 68.6\\\nEfficacy boundary at interim analysis ($u_k$) & 3.11 & 2.71 & 2.47 & 2.27 & -\\\nFutility boundary at interim analysis ($l_k$) & -1.20 & -0.08 & 0.70 & 1.36 & -\\\nCritical value at decision analysis ($c_k$) & 1.30 & 1.54 & 1.74 & 1.91 & 2.07\\\n\\\n&\\\nType I error per stage $\\times 10^{-3}$ & 0.8 & 2.9 & 5.0 & 7.0 & 9.3\\\nType II error per stage $\\times 10^{-3}$ & 6.5 & 23.4 & 40.1 & 55.8 & 74.1\\\nEfficacy boundary at interim analysis ($u_k$) & 3.16 & 2.73 & 2.48 & 2.28 & -\\\nFutility boundary at interim analysis ($l_k$) & -1.26 & -0.14 & 0.66 & 1.33 & -\\\nCritical value at decision analysis ($c_k$) & 1.32 & 1.55 & 1.75 & 1.91 & 2.06\\\n\\\n&\\\nType I error per stage $\\times 10^{-3}$ & 1.2 & 3.2 & 5.3 & 7.5 & 7.8\\\nType II error per stage $\\times 10^{-3}$ & 9.3 & 25.8 & 42.6 & 60.2 & 62.1\\\nEfficacy boundary at interim analysis ($u_k$) & 3.05 & 2.68 & 2.44 & 2.24 & -\\\nFutility boundary at interim analysis ($l_k$) & -0.99 & 0.04 & 0.80 & 1.47 & -\\\nCritical value at decision analysis ($c_k$) & 1.24 & 1.50 & 1.72 & 1.91 & 2.06\\\n\\\n&\\\nType I error per stage $\\times 10^{-3}$ & 0.9 & 2.9 & 5.1 & 6.9 & 9.3\\\nType II error per stage $\\times 10^{-3}$ & 7.3 & 22.9 & 40.6 & 54.9 & 74.3\\\nEfficacy boundary at interim analysis ($u_k$) & 3.12 & 2.73 & 2.47 & 2.28 & -\\\nFutility boundary at interim analysis ($l_k$) & -1.16 & -0.08 & 0.72 & 1.37 & -\\\nCritical value at decision analysis ($c_k$) & 1.24 & 1.50 & 1.72 & 1.91 & 2.04\\\n\\\n&\\\nType I error per stage $\\times 10^{-3}$ & 1.0 & 3.0 & 4.9 & 6.9 & 9.2\\\nType II error per stage $\\times 10^{-3}$ & 7.8 & 24.1 & 39.6 & 54.9 & 73.6\\\nEfficacy boundary at interim analysis ($u_k$) & 3.10 & 2.72 & 2.48 & 2.29 & -\\\nFutility boundary at interim analysis ($l_k$) & -1.37 & -0.43 & 0.24 & 0.82 & -\\\nCritical value at decision analysis ($c_k$) & 1.18 & 1.36 & 1.53 & 1.67 & 2.11\\\n\\\n\nResampling-based Algorithm to Simulate Trial Data {#appen:dgm}\n=================================================\n\nFor the data generating distribution to have the two properties in Section \\[subsec:dgm\\], we design the algorithm to generate a simulated trial of $n$ independent and identically distributed samples from the 100 participants in MISTIE-II dataset. ",
"Recall that the three properties are:\n\n1. ",
" The treatment assignment is independent of baseline variables.",
"\n\n2. ",
" The relative efficiency between the adjusted estimator and the unadjusted estimator in the simulated data is similar to that calculated directly from the MISTIE-II dataset.",
"\n\nBriefly, the algorithm ensures property (i) by adding a “twin” with identical baseline variables and opposite treatment assignment to each participant in the MISTIE-II dataset. ",
"The variables $L^\\text{full}$ and $Y$ for each “twin” are then generated using regression model fits. ",
"The original MISTIE-II data and the set of “twins” results in a 200 participant data set. ",
"This is done once, before any of our simulations are conducted. ",
"To generate each simulated data set, we resample participant vectors from this 200 participant data set with replacement and then make modifications to the replicate data set depending on the desired simulation setting. ",
"Details of the algorithm are given below.",
"\n\n*Step 1: Construct a set of 100 pairs of “twins”.* ",
"We start with the 100-participant MISTIE-II dataset, and we augment the data with a hypothetical “twin” for each participant. ",
"A “twin” has identical baseline variables as the original participant, but opposite treatment assignment. ",
"We fit logistic regression models for $L^{(1)}$ on $(W^\\text{full},A)$, for $L^{(2)}$ on $(W^\\text{full},A,L^{(1)})$, and for $Y$ on $(W^\\text{full},A,L^{(1)},L^{(2)})$, using the original 100 participants in MISTIE-II dataset. ",
"The preliminary $L_i$ and $Y_i$ of each newly added twin are then predicted based on these logistic regression models by rounding the predicted success probability to be $0$ or $1$. The indicator of whether a participant is an original participant in the MISTIE-II data set or a hypothetical twin is included as a variable in this augmented data set of 200 participants, which will be used in Step 2. ",
"Step 1 is only done once at the beginning of the simulation.",
"\n\n*Step 2: Sample $n$ participants from the augmented data set with 200 participants and calibrate the treatment effect on the sampled participants.* ",
"We sample $n$ participants uniformly with replacement from the augmented dataset constructed in Step 1. ",
"Then, for each participant in the $n$ participants whose “hypothetical twin indicator” is true, with probability $0.03$ we reset its $Y_i$ to equal $A_i$. This resetting step increases the treatment effect of the augmented data to $0.122$, matching that of the original data.",
"\n\n*Step 3: Calibrate relative efficiency between the adjusted estimator and the unadjusted estimator.* ",
"For $a\\in\\{ 0,1\\}$, we empirically estimate the marginal distribution $p_{Y;a}=P(Y=1\\mid A=a)$ using the corresponding sample proportions in the original data. ",
"Then, for each participant in the $n$ participants whose “hypothetical twin indicator” is true, with probability $0.164$ we reset $Y_i$ by a realization of an independent Bernoulli random draw with success probability $p_{Y;A_i}$. This resetting step adds random noise to reduce the prognostic value in $W$ and $L$, so that the relative efficiency between the adjusted estimator and the unadjusted estimator for the simulated data is comparable to the estimated ${\\text{ARE}}$ from the original MISTIE-II dataset.",
"\n\nWe then make modifications that are described next to assess validity of the theory under various settings. ",
"We define eight data generating distributions, called “settings”, each based on modifying the aforementioned data generating process. ",
"The goal is to consider four situations involving $W$ and/or $L$ being prognostic or not for $Y$; these are denoted by ${progn_{W,L}}$ (both prognostic), ${progn_{W}}$ (only $W$ prognostic), ${progn_{L}}$ (only $L$ prognostic), and ${progn_{\\emptyset}}$ (neither prognostic). ",
"For each of these four situations, we consider the following two average treatment effects: $\\Delta=0$ (no effect) or $\\Delta=0.122$ (benefit).",
"\n\nWe first describe how we generated simulated trials with $\\Delta=0.122$. The aforementioned data generating algorithm corresponds to ${progn_{W,L}}$. To create data generating distributions with ${progn_{W}}$, we first generate each participant’s data as in ${progn_{W,L}}$ and then replace $L$ by an independent draw from the marginal distribution in the MISTIE-II data. ",
"Similarly, in order to make only $L$ prognostic for $Y$ (${progn_{L}}$), we first generate each participant’s data as in ${progn_{W,L}}$ and then replace $W$ by an independent draw from the marginal distribution in the MISTIE-II data. ",
"Lastly, to make neither $W$ nor $L$ prognostic, we first generate each participant’s data as in ${progn_{W,L}}$ and then replace each of $W$ and $L$ by independent draws from the corresponding marginal distributions in the MISTIE-II data, respectively.",
"\n\nTo generate participants under $\\Delta=0$, the data generating process above is followed, and then each study arm assignment $A$ is replaced by an independent Bernoulli draw with probability $1/2$. Under the data generating process with $\\Delta=0.122$, there is slight treatment effect heterogeneity $\\gamma \\approx 0.01$. By construction, under $\\Delta=0$ there is no treatment effect heterogeneity ($\\gamma = 0$).",
"\n\nIntuition for Impact of Treatment Effect Heterogeneity on Precision Gains {#appen:adjustmentintuition}\n=========================================================================\n\nTo give intuition for the phenomena in Figure \\[fig:are-w-ate\\] in Section \\[sec:impactbaseline\\], for the remainder of this appendix we fix $R^2_W$ and consider the simple case of a single, binary-valued $W$ representing being less than 65 years old at baseline, and primary outcome $Y$ being the indicator of having mRS at most 3 (a good outcome) at 180 days. ",
"We focus on the stratum $W=1$ and assume that the under 65 subset of the study population would have better outcomes on average than the overall study population if all were assigned to the control arm, i.e., $E(Y|A=0,W=1)-E(Y|A=0)>0$. We next consider $p_y=1$ and the extreme of maximum treatment effect heterogeneity (where there are no efficiency gains from adjusting for $W$). (",
"The opposite extreme of $p_y=1$ and no treatment effect heterogeneity is discussed in detail in Section \\[sec:impactbaseline\\].)",
"\n\nConsider $p_y=1$ and with maximum treatment effect heterogeneity ($\\gamma = 2R^2_W$), which occurs when $E(Y|A=1,W)+E(Y|A=0,W)$ equals the constant $E(Y|A=1)+E(Y|A=0)$. It follows from the assumption above that those under 65 have better outcomes on average than the overall study population in the control arm ($E(Y|A=0,W=1)-E(Y|A=0)>0$), that those under 65 have worse outcomes on average than the overall study population under assignment to the treatment arm (and by precisely the same magnitude), which follows since $E(Y|A=1,W=1)-E(Y|A=1)= -\\{E(Y|A=0,W=1)-E(Y|A=0)\\}$. If by chance there are proportionally more under 65 participants assigned to the treatment arm than the control arm, then the unadjusted estimator of the mean outcome in each arm is biased downward, and by the same amount, conditional on the chance imbalance. ",
"Since this bias cancels out when estimating $\\Delta$, an efficient estimator of $\\Delta$ makes no adjustment. ",
"Alternatively, if by chance there are proportionally fewer under 65 participants assigned to the treatment arm, then by a symmetric argument an efficient estimator of $\\Delta$ makes no adjustment. ",
"The overall impact is no adjustment due to chance imbalances in the stratum $W=1$, which means that no improvement in variance is made compared to the unadjusted estimator. ",
"This is, intuitively, why there is no efficiency gain under maximum treatment effect heterogeneity ($\\gamma=2R^2_W$) when there are no pipeline participants ($p_y=1$).",
"\n\nWe next give intuition for why one needs $\\gamma>0$ in order for pipeline participants to contribute useful information toward estimating $\\Delta$ (when $p_y < 1$). ",
"Consider the case where those in stratum $W=1$ benefit more than the overall population, i.e., $E(Y|A=1,W=1)-E(Y|A=0,W=1)>\\Delta$; this is only possible if $\\gamma>0$. Assume that adjustment for chance imbalance in the stratum $W=1$ between arms has already been done to the unadjusted estimator (using only those with $Y$ observed), as described above. ",
"Consider the proportion of participants having $W=1$ among all participants and also among the subset of participants with $Y$ observed. ",
"If the former proportion is larger than the latter, then the stratum $W=1$ (of participants who benefit more than average) is underrepresented among those with outcomes observed. ",
"An efficient estimator adjusts the estimate of $\\Delta$ upward to compensate. ",
"Alternatively, if the former proportion is smaller than the latter, by a symmetric argument an efficient estimator adjusts the estimate of $\\Delta$ downward. ",
"In this way, variance due to imbalance in the proportion with $W=1$ between the overall sample (including pipeline participants) and those with $Y$ observed is removed. ",
"The only cases where adjusting for the aforementioned imbalance at each level of $W$ has no impact are when $E(Y|A=1,W)-E(Y|A=0,W) = \\Delta$, i.e., zero treatment effect heterogeneity, or $p_y=1$ (no pipeline participants).",
"\n\nAdditional Results: When Baseline Variable is Purely Predictive or Purely Prognostic. {#",
"appen:theory-predictive-prognostic}\n=====================================================================================\n\nThe following theorem examines the impact of treatment effect heterogeneity on precision gain from adjusting for prognostic baseline variables under two extreme cases. ",
"Define ${\\text{ARE}}({\\text{efficient}}, {\\text{unadjusted}})$ to be the asymptotic relative efficiency in (\\[eq:re-ate\\]). ",
"More generally, ${\\text{ARE}}(\\widehat{\\Delta}_1, \\widehat{\\Delta}_2)$ denotes the asymptotic relative efficiency of the estimators $\\widehat{\\Delta}_1, \\widehat{\\Delta}_2$.\n\n\\[cor:prognostic-predictive\\] Suppose Assumptions \\[assump:randomization\\], \\[assump:independent-censoring\\], and \\[assump:monotone-censoring\\] hold. ",
"In addition, assume $R^2_{L \\mid W} = 0$. For the estimand $E(Y\\mid A=1) - E(Y\\mid A=0)$, we have the following result regarding the asymptotic relative efficiency between an efficient RAL estimator and the unadjusted estimator.",
"\n\n1. ",
" (Maximal treatment effect heterogeneity.)",
"\n\n If ${\\text{Var}}\\{E_1(Y \\mid W) - E_0(Y \\mid W)\\} > 0$ and ${\\text{Var}}\\{E(Y \\mid W)\\} = 0$, then $\\gamma = 2 R_W^2$ and $${\\text{ARE}}({\\text{efficient}}, {\\text{unadjusted}}) = \\frac{1}{1 - (1-p_y) R^2_W}. ",
"\\label{eq:rsq-onlyW-ate-predictive}$$\n\n2. ",
" (No treatment effect heterogeneity.)",
"\n\n If ${\\text{Var}}\\{E_1(Y \\mid W) - E_0(Y \\mid W)\\} = 0$ and ${\\text{Var}}\\{E(Y \\mid W)\\} > 0$, then $\\gamma = 0$ and $${\\text{ARE}}({\\text{efficient}}, {\\text{unadjusted}}) = \\frac{1}{1 - R^2_W}. ",
"\\label{eq:rsq-onlyW-ate-prognostic}$$\n\nWe call a baseline variable $W$ *purely predictive* if it satisfies the conditions in Corollary \\[cor:prognostic-predictive\\](i), as it predicts the conditional average treatment effect (i.e., it identifies which strata of $W$ benefit from the treatment on average), but it does not explain the variance in $Y$ marginally (i.e., margining out $A$). ",
"For such a baseline variable, ${\\text{ARE}}= 1$ in (\\[eq:rsq-onlyW-ate-predictive\\]) when $p_y=1$. This means that adjusting for a purely predictive baseline variable results in no precision gain when there are no missing primary outcomes. ",
"An extreme case of this, for illustration only, is the following data generating distribution where the baseline variable is perfectly correlated with the outcome within each arm but still contributes nothing to the precision gain: $W,A$ are independently distributed as $\\text{Bernoulli}(0.5)$ and $Y = {{\\mathds 1}}(A=W)$, where ${{\\mathds 1}}(\\cdot)$ is the indicator function. ",
"In this case it is straightforward to verify that $R^2_W = 1$, but adjusting for $W$ would not impact asymptotic variance compared to the unadjusted estimator if all participants have $Y$ observed $(p_y=1)$ since $\\gamma=2$. Though we don’t expect to encounter a purely predictive variable in practice, the above thought experiment shows how $W$ could explain treatment effect heterogeneity (which is useful on its own) while being useless for improving precision in estimating the average treatment effect.",
"\n\nWe call a baseline variable $W$ *purely prognostic* if it satisfies the conditions in Corollary \\[cor:prognostic-predictive\\](ii), as the treatment effect is constant across strata of $W$, and $W$ explains some of the marginal variance in $Y$. For such a baseline variable, the ${\\text{ARE}}$ in (\\[eq:rsq-onlyW-ate-prognostic\\]) does not depend on $p_y$. An intuition for why this holds, as discussed in Section \\[sec:impactbaseline\\], is that adjusting for such a baseline variable reduces variance by correcting for chance imbalances between arms among those who have $Y$ observed, while the baseline variables for the pipeline participants contain no information about $\\Delta$ since $\\gamma=0$ (and we are ignoring $L$ here).",
"\n\nPrecision Gain When Estimating the Treatment Specific Mean $E(Y|A=a)$ {#sec:theory-singlearm}\n=====================================================================\n\nIn this section we focus on estimating the population mean outcome under assignment to a single arm (called the treatment specific mean): $E(Y\\mid A=a)$ for each $a\\in\\{0,1\\}$ separately. ",
"The following theorem gives the lower bound on the asymptotic variance for all regular asymptotically linear (RAL) estimators of $E(Y\\mid A=a)$.\n\n\\[thm:avar\\]\n\nDenote by $p_a = P(A=a)$ for $a\\in\\{0, 1\\}$. Assume $p_y > 0$ and $0<p_a<1$. Under Assumptions \\[assump:randomization\\], \\[assump:independent-censoring\\], and \\[assump:monotone-censoring\\] (except allowing $p_a \\neq 1/2$) the semiparametric lower bound on the asymptotic variance of all RAL estimators for $E(Y\\mid A=a)$ is $${\\text{Var}}_a\\{E_a(Y \\mid W)\\} + \\frac{1}{p_a p_l} {\\text{Var}}_a\\{ E_a(Y \\mid L,W) - E_a(Y \\mid W) \\} + \\frac{1}{p_a p_y} {\\text{Var}}_a\\{ Y - E_a(Y \\mid L,W) \\}. ",
"\\label{eq:thm:var-bound}$$\n\nAnalogous to the $R$-squared quantities defined in Section \\[sec:theory-ate\\], we define new $R$-squared quantities relevant to estimating the population mean of the primary outcome $Y$ in each arm separately. ",
"For each $a \\in \\{0,1\\}$, define the proportion of variance in $Y$ explained by $W$ in arm $a$ as $\nR^2_{W;a} = {\\text{Var}}\\{E_a(Y \\mid W)\\} / {\\text{Var}}_a(Y)\n$; the proportion of additional variance in $Y$ explained by $L$ after accounting for $W$ in arm $a$ as $\nR^2_{L \\mid W;a} = {\\text{Var}}_a\\{E_a(Y \\mid L, W) - E_a(Y \\mid W)\\} / {\\text{Var}}_a(Y)\n$. ",
"We also will refer to the proportion of variance in $Y$ explained by $L$ alone in arm $a$, defined as $R^2_{L;a} = {\\text{Var}}_a\\{E_a(Y \\mid L)\\} / {\\text{Var}}_a(Y)$.\n\nThe following result gives the ${\\text{ARE}}$ between an efficient RAL estimator and the unadjusted estimator for $E(Y\\mid A=a)$.\n\n\\[cor:re\\]\n\nAssume all conditions in Theorem \\[thm:avar\\] hold. ",
"For each arm $a\\in\\{0,1\\}$, the asymptotic relative efficiency between an efficient RAL estimator and the unadjusted estimator of the treatment specific mean $E(Y\\mid A=a)$ is $$\\frac{1}{1 - (1 - p_a p_y) R^2_{W;a} - (1 - p_y / p_l) R^2_{L\\mid W;a}}. ",
"\\label{eq:re}$$\n\n[.5]{} ![[]{",
"data-label=\"fig:are-l\"}](REW.png \"fig:\"){width=\"\\linewidth\"}\n\n[.5]{} ![[]{",
"data-label=\"fig:are-l\"}](REL.png \"fig:\"){width=\"\\linewidth\"}\n\nWhen only $W$ is prognostic ($R^2_{L \\mid W;a} = 0$), the asymptotic relative efficiency between an efficient RAL estimator and the unadjusted estimator is equal to when the $L$ variable is not available to the efficient RAL estimator (i.e., when the efficient RAL estimator adjusts for $W$ alone). ",
"Similarly, when only $L$ is prognostic ($R^2_{W;a} = 0$), the asymptotic relative efficiency is equal to when the $W$ variable is not available to the efficient RAL estimator (i.e., when only adjusts for $L$ alone). ",
"In Figure \\[fig:are-w\\] we illustrate the implication of Theorem \\[cor:re\\] under those special cases. ",
"We plot ${\\text{ARE}}$ against $p_y$ when only $W$ is prognostic ($R^2_{L \\mid W;a} = 0$), and in Figure \\[fig:are-l\\] we plot ${\\text{ARE}}$ against $p_y / p_l$ when only $L$ is prognostic ($R^2_{W;a} = 0$). ",
"In each plot, we separately consider prognostic value being $0.1$ and $0.25$. We consider three implications of the curves in Figure \\[fig:are\\]. ",
"First, the precision gain (${\\text{ARE}}$) from adjusting for a prognostic baseline variable increases with a greater proportion of pipeline participants (i.e., smaller $p_y$). ",
"The reason is that each additional pipeline participant (who has observed baseline variables but missing outcome) contributes some information to the adjusted estimator (through the baseline variables) but no information to the unadjusted estimator. ",
"For a similar reason, the precision gain from adjusting for a prognostic short-term outcome increases with a greater proportion of participants with $W,L$ but not $Y$ observed (i.e., smaller $p_y/p_l$). ",
"Second, when every participant has their primary outcome observed (e.g., at the final analysis time of a group sequential design with no dropouts), adjusting for the prognostic baseline variables still improves estimation precision as long as $R^2_{W;a} > 0$, but adjusting for prognostic short-term outcome no longer does. ",
"Third, for any given $(p_y,p_l)$, adjusting for a prognostic baseline variable alone always leads to larger precision gain than adjusting for an equally prognostic short-term outcome alone.",
"\n\nFollowing the discussion in Section \\[appen:are-sample-size-reduction\\], we define the asymptotic equivalent reduction in sample size (${\\text{AERSS}}$) of $\\widehat{\\Delta}_1$ compared to $\\widehat{\\Delta}_2$ as $${\\text{AERSS}}(\\widehat{\\Delta}_1, \\widehat{\\Delta}_2) = 1 - {\\text{ARE}}(\\widehat{\\Delta}_1, \\widehat{\\Delta}_2)^{-1}. ",
"\\label{def:redss}$$ Adjusting for prognostic $W$ alone with $R^2_{W;a}=q$ (for $0<q<1$) yields ${\\text{AERSS}}= q(1 - p_y/2)$, and adjusting for prognostic $L$ alone with $R^2_{L;a} = q$ yields ${\\text{AERSS}}= q(1 - p_y/p_l)$. The ratio of the two ${\\text{AERSS}}$s equals (assuming $q>0$) $$r = \\frac{{\\text{AERSS}}\\text{ from adjusting only for prognostic } W \\text{ with } R^2_{W;a}=q}{{\\text{AERSS}}\\text{ from adjusting only for prognostic } L \\text{ with } R^2_{L;a}=q} = \\frac{1 - p_y/2}{1 - p_y/p_l}. ",
"\\label{def:ratio-redss}$$ $r>1$ means that the sample size reduction from adjusting for a prognostic $W$ is larger than that from adjusting for an equally prognostic $L$; $r<1$ means the opposite. ",
"Figure \\[fig:contour\\] plots $r$ against $(p_l, p_y)$ under the constraint $0 < p_y \\leq p_l \\leq 1$. For all such $(p_l, p_y)$, $r > 1$. In addition, we have $r \\geq 2$ if $p_y \\geq 2/3$. This means that if at most $1/3$ of the enrollees are in the pipeline, the sample size reduction from adjusting for a prognostic short-term outcome alone is at most half of that from adjusting for an equally prognostic baseline variable alone. ",
"Roughly speaking, small $p_y$ makes $r$ close to 1, meaning that adjusting for $W$ or adjusting for $L$ results in a similar sample size reduction when there are relatively few with $Y$ observed. ",
"This may occur at early stages of a group sequential design, if the delay of the primary outcome is long relative to the enrollment rate.",
"\n\nSimulation Results for Relative Efficiency When Estimating the Treatment Specific Mean {#appen:re-h1}\n======================================================================================\n\nThe simulation setup, including the data generating distributions (settings), are the same as Section \\[sec:simulation\\]. ",
"Table \\[tab:rsq-mistie-a0a1\\] gives the values of $R^2_{W;a}$ and $R^2_{L \\mid W;a}$ for $a\\in\\{0,1\\}$, which are defined in the paragraph after Theorem \\[thm:avar\\]. ",
"Tables \\[tab:re-H0\\] and \\[tab:re-H1\\] list the asymptotic relative efficiency (ARE) approximated by substituting model-based estimates of the conditional expectations in the quantities in (\\[eq:re\\]).",
"\n\n -------------------- ------------ ---------------- -- ------------ ---------------- -- ------------ ---------------- -- ------------ ----------------\n \n $\\Delta=0$ $\\Delta=0.122$ $\\Delta=0$ $\\Delta=0.122$ $\\Delta=0$ $\\Delta=0.122$ $\\Delta=0$ $\\Delta=0.122$\n $R^2_{W;0}$ 0.35 0.33 0.35 0.33 0 0 0 0\n $R^2_{L \\mid W;0}$ 0.08 0.06 0 0 0.30 0.28 0 0\n $R^2_{W;1}$ 0.35 0.38 0.35 0.38 0 0 0 0\n $R^2_{L \\mid W;1}$ 0.08 0.08 0 0 0.30 0.31 0 0\n -------------------- ------------ ---------------- -- ------------ ---------------- -- ------------ ---------------- -- ------------ ----------------\n\n : $R^2_{W;a}$ and $R^2_{L \\mid W;a}$ for $a\\in\\{0,1\\}$ for the data generating distributions from the settings in Section \\[sec:simulation\\], approximated based on a simulated dataset with 1,000,000 participants under each setting.[]{data-label=\"tab:rsq-mistie-a0a1\"}\n\nProofs {#appen:sec:proof}\n======\n\nWe prove the results in the main paper as well as results in Section \\[appen:theory-predictive-prognostic\\] and Section \\[sec:theory-singlearm\\]. ",
"Auxiliary lemmas (used in these proofs) are proved Section \\[appen:proof-aux\\].",
"\n\nIdentification of the average treatment effect\n----------------------------------------------\n\nUnder Assumptions \\[assump:randomization\\]-\\[assump:monotone-censoring\\], the average treatment effect can be expressed as follows: $$\\begin{aligned}\nE(Y \\mid A=1) - E(Y \\mid A=0) & =& E [ E \\{ E (Y \\mid L,W,A=1,C^Y=1) \\mid W,C^L=1,A=1\\} ] \\nonumber \\\\\n&& - E [ E \\{ E (Y \\mid L,W,A=0,C^Y=1) \\mid W,C^L=1,A=0\\} ] \\label{eq:gcomputation}\\end{aligned}$$ The proof of (\\[eq:gcomputation\\]) is given as follows.",
"\n\nIt suffices to show that for each $a\\in\\{0,1\\}$ the following holds: $$E(Y \\mid A=a) = E [ E \\{ E (Y \\mid L,W,A=a,C^Y=1) \\mid W,A=a,C^L=1\\} ]. ",
"\\label{eq:gcomputation-proofuse-1}$$ Using the law of iterated expectation twice, we deduce $$E(Y \\mid A=a) = E [ E \\{ E (Y \\mid L,W,A=a) \\mid W,A=a\\} ]. ",
"\\label{eq:gcomputation-proofuse-3}$$ Let $f(L,W) = E (Y \\mid L,W,A=a)$. By Assumption \\[assump:independent-censoring\\] we have $$E(f(L,W) \\mid A=a) = E(f(L,W) \\mid A=a, C^L=1), \\label{eq:gcomputation-proofuse-4}$$ and $$E(Y \\mid L,W,A=a) = E(Y \\mid L,W,A=a, C^L=1, C^Y=1). ",
"\\label{eq:gcomputation-proofuse-5}$$ Equations (\\[eq:gcomputation-proofuse-3\\])-(\\[eq:gcomputation-proofuse-5\\]) together yield $$E(Y \\mid A=a) = E [ E \\{ E (Y \\mid L,W,A=a,C^L=1,C^Y=1) \\mid W,A=a,C^L=1\\} ]. ",
"\\label{eq:gcomputation-proofuse-6}$$ Equation (\\[eq:gcomputation-proofuse-6\\]) and Assumption \\[assump:monotone-censoring\\] yield (\\[eq:gcomputation-proofuse-1\\]). ",
"This completes the proof.",
"\n\nLemma on Variance Decomposition {#proof:lem:vardecomp}\n-------------------------------\n\n\\[lem:vardecomp\\]\n\nFor study arm $a\\in\\{0,1\\}$, we have the following decomposition of the variance of $Y$ in that arm: $${\\text{Var}}_a(Y) = {\\text{Var}}_a\\{E_a(Y\\mid W)\\} + {\\text{Var}}_a\\{ E_a(Y\\mid L,W) - E_a(Y \\mid W) \\} + {\\text{Var}}_a\\{Y - E_a(Y\\mid L,W)\\}. ",
"\\label{eq:vardecomp}$$ In addition, we have $$R^2_{W;a} + R^2_{L\\mid W;a} + R^2_{r;a} = 1, \\quad \\mbox{ for each } a \\in \\{0,1\\}. ",
"\\label{eq:rsqsum1}$$\n\nAdding and subtracting terms, we have $${\\text{Var}}_a(Y) = {\\text{Var}}_a \\big[ \\{Y - E_a(Y\\mid L,W)\\} + \\{E_a(Y\\mid L,W) - E_a(Y \\mid W)\\} + E_a(Y\\mid W) \\big].$$ So for proving (\\[eq:vardecomp\\]), it suffices to establish the following: $$\\begin{aligned}\n{\\text{Cov}}_a \\big\\{ Y - E_a(Y\\mid L,W), E_a(Y\\mid L,W) - E_a(Y \\mid W) \\big\\} &= 0, \\label{eq:vardecomp-proofuse-1} \\\\\n{\\text{Cov}}_a \\big\\{ Y - E_a(Y\\mid L,W), E_a(Y\\mid W) \\big\\} &= 0, \\label{eq:vardecomp-proofuse-2} \\\\\n{\\text{Cov}}_a \\big\\{ E_a(Y\\mid L,W) - E_a(Y \\mid W), E_a(Y\\mid W) \\big\\} &= 0, \\label{eq:vardecomp-proofuse-3}\\end{aligned}$$ where ${\\text{Cov}}_a$ denotes the conditional covariance given $A=a$.\n\nFirst, we have $$\\begin{aligned}\n& {\\text{Cov}}_a \\big\\{ Y - E_a(Y\\mid L,W), E_a(Y\\mid L,W) - E_a(Y \\mid W) \\big\\} \\nonumber\\\\\n= & E_a \\big\\{ Y E_a(Y\\mid L,W) - E_a(Y\\mid L,W)^2\n - Y E_a(Y \\mid W) + E_a(Y\\mid L,W) E_a(Y \\mid W) \\big\\}. ",
"\\label{eq:vardecomp-proofuse-4}\\end{aligned}$$ By Lemma \\[lem:cond-exp\\] with $X = Z = (L,W)$, we have $$E_a \\big\\{ Y E_a(Y\\mid L,W) \\big\\} = E_a \\big\\{ E_a(Y \\mid L,W)^2 \\big\\}. ",
"\\label{eq:vardecomp-proofuse-5}$$ By Lemma \\[lem:cond-exp\\] with $X = W$ and $Z = (L,W)$, we have $$E_a \\big\\{ Y E_a(Y \\mid W) \\big\\} = E_a \\big\\{ E_a(Y \\mid L,W) E_a(Y\\mid W) \\big\\}. ",
"\\label{eq:vardecomp-proofuse-6}$$ Equations (\\[eq:vardecomp-proofuse-4\\]), (\\[eq:vardecomp-proofuse-5\\]), and (\\[eq:vardecomp-proofuse-6\\]) imply (\\[eq:vardecomp-proofuse-1\\]).",
"\n\nSecond, we have $${\\text{Cov}}_a \\big\\{ Y - E_a(Y\\mid L,W), E_a(Y\\mid W) \\big\\} = E_a \\big\\{ Y E_a(Y\\mid W) - E_a(Y\\mid L,W) E_a(Y\\mid W) \\big\\}. ",
"\\label{eq:vardecomp-proofuse-7}$$ Equations (\\[eq:vardecomp-proofuse-7\\]) and (\\[eq:vardecomp-proofuse-6\\]) imply (\\[eq:vardecomp-proofuse-2\\]).",
"\n\nThird, since $E_a(Y\\mid L,W) - E_a(Y \\mid W)$ has expectation zero, we have $${\\text{Cov}}_a \\big\\{ E_a(Y\\mid L,W) - E_a(Y \\mid W), E_a(Y\\mid W) \\big\\}\n= E_a \\big\\{ E_a(Y\\mid L,W) E_a(Y\\mid W) - E_a(Y \\mid W)^2 \\big\\}. ",
"\\label{eq:vardecomp-proofuse-8}$$ In Lemma \\[lem:cond-exp\\], letting $X=Z=W$ and replacing $Y$ in the lemma by $E(Y \\mid L,W)$ implies $$E_a \\big\\{ E_a(Y \\mid L,W) E_a(Y\\mid W) \\big\\} = E_a \\big\\{ E_a(Y \\mid W)^2 \\big\\}. ",
"\\label{eq:vardecomp-proofuse-9}$$ Equations (\\[eq:vardecomp-proofuse-8\\]) and (\\[eq:vardecomp-proofuse-9\\]) imply (\\[eq:vardecomp-proofuse-3\\]).",
"\n\nThis proves (\\[eq:vardecomp\\]). ",
"Equation (\\[eq:rsqsum1\\]) follows immediately from (\\[eq:vardecomp\\]) and the definition of $R^2_{W;a}$, $R^2_{L \\mid W;a}$ and $R^2_{r;a}$. This completes the proof for Lemma \\[lem:vardecomp\\].",
"\n\nProof of Theorem \\[thm:avar\\] {#proof:thm:avar}\n-----------------------------\n\nIn the proof, we will use equation (\\[eq:gcomputation-proofuse-6\\]) derived earlier; we rewrite it below: $$E(Y \\mid A=a) = E [ E \\{ E (Y \\mid L,W,A=a,C^L=1,C^Y=1) \\mid W,A=a,C^L=1\\} ]. ",
"\\label{eq:thm:avar-proofuse-1}$$ Treating the missingness indicators $C^L$ and $C^Y$ as binary treatments, the right-hand side of (\\[eq:thm:avar-proofuse-1\\]) becomes the average of outcome $Y$ under time dependent treatment assignment: $A=a, C^L=1, C^Y=1$. Because there is no measurement made between $A$ and $C^L$, we can combine the two as a single treatment $\\tilde{A}$, with $\\tilde{A}=1$ if and only if $A=a$ and $C^L=1$. Equation (\\[eq:thm:avar-proofuse-1\\]) becomes $$E(Y \\mid A=a) = E [ E \\{ E (Y \\mid L,W,\\tilde{A}=1,C^Y=1) \\mid W,\\tilde{A}=1\\} ]. ",
"\\label{eq:thm:avar-proofuse-1.5}$$ Using the fact that $L$ is binary-valued, by equations (24) and (28) in @rosenblum2011EIF or Theorem 1 in @van2010EIF, the efficient influence function for (\\[eq:thm:avar-proofuse-1.5\\]) is $$D(W,\\tilde{A},L,C^Y,Y) = D_0(W) + D_1(W,\\tilde{A},L) + D_2(W,\\tilde{A},L,C^Y,Y), \\label{eq:thm:avar-proofuse-d}$$ where $$D_0(W) = E(Y \\mid W,\\tilde{A}=1,C^Y=1) - E(Y \\mid \\tilde{A}=1), \\label{eq:thm:avar-proofuse-d0}$$ $$D_1(W,\\tilde{A},L) = \\frac{\n{{\\mathds 1}}(\\tilde{A}=1) \\big\\{ E(Y \\mid L,W, \\tilde{A}=1, C^Y=1) - E(Y \\mid W,\\tilde{A}=1,C^Y=1) \\big\\}\n}{P(\\tilde{A}=1\\mid W)}, \\label{eq:thm:avar-proofuse-d1}$$ and $$D_2(W,\\tilde{A},L,C^Y,Y) = \\frac{\n{{\\mathds 1}}(C^Y=1) {{\\mathds 1}}(\\tilde{A}=1) \\big\\{ Y - E(Y \\mid L,W, \\tilde{A}=1, C^Y=1) \\big\\}\n}{P(C^Y=1 \\mid L,W,\\tilde{A}=1)P(\\tilde{A}=1\\mid W)}. ",
"\\label{eq:thm:avar-proofuse-d2}$$\n\nBy randomization and independent censoring assumptions, (\\[eq:thm:avar-proofuse-d0\\]) simplifies to $$D_0(W) = E(Y \\mid W, A=a) - E(Y \\mid A=a); \\label{eq:thm:avar-proofuse-d0-simp}\n$$ equation (\\[eq:thm:avar-proofuse-d1\\]) simplifies to $$D_1(W,\\tilde{A},L) = \\frac{\n{{\\mathds 1}}(\\tilde{A}=1) \\big\\{ E(Y \\mid L,W, A=a) - E(Y \\mid W,A=a) \\big\\}\n}{P(\\tilde{A}=1)}; \\label{eq:thm:avar-proofuse-d1-simp}\n$$ equation (\\[eq:thm:avar-proofuse-d2\\]) simplifies to $$D_2(W,\\tilde{A},L,C^Y,Y) = \\frac{\n{{\\mathds 1}}(C^Y=1) {{\\mathds 1}}(\\tilde{A}=1) \\big\\{ Y - E(Y \\mid L,W,A=a) \\big\\}\n}{P(C^Y=1 \\mid C^L=1)P(\\tilde{A}=1)}. ",
"\\label{eq:thm:avar-proofuse-d2-simp}\n$$\n\nThe following lemma states that $D_0$, $D_1$, and $D_2$ are pairwise uncorrelated.",
"\n\n\\[lem:thm-proofuse\\]\n\nWe have $$\\begin{aligned}\n{\\text{Cov}}\\{ D_0(W), D_1(W,\\tilde{A},L)\\} & = 0, \\label{eq:lem:thm-proofuse-toshow-d0d1}\\\\\n{\\text{Cov}}\\{ D_0(W), D_2(W,\\tilde{A},L,C^Y,Y)\\} & = 0, \\label{eq:lem:thm-proofuse-toshow-d0d2}\\\\\n{\\text{Cov}}\\{ D_1(W,\\tilde{A},L), D_2(W,\\tilde{A},L,C^Y,Y)\\} & = 0. ",
"\\label{eq:lem:thm-proofuse-toshow-d1d2}\\end{aligned}$$\n\nLemma \\[lem:thm-proofuse\\] implies $${\\text{Var}}\\{D(W,\\tilde{A},L,C^Y,Y)\\} = {\\text{Var}}\\{D_0(W)\\} + {\\text{Var}}\\{D_1(W,\\tilde{A},L)\\} + {\\text{Var}}\\{D_2(W,\\tilde{A},L,C^Y,Y)\\}. ",
"\\label{eq:thm:avar-proofuse-2}$$ By (\\[eq:thm:avar-proofuse-d0-simp\\]) we have $${\\text{Var}}\\{D_0(W)\\} = {\\text{Var}}\\{E_a(Y \\mid W)\\} = {\\text{Var}}_a\\{E_a(Y \\mid W)\\}, \\label{eq:thm:avar-proofuse-vard0}$$ where the last equality follows from randomization assumption. ",
"By (\\[eq:thm:avar-proofuse-d1-simp\\]) we have $E\\{D_1(W,\\tilde{A},L)\\}=0$, so it follows from randomization and independent censoring that $$\\begin{aligned}\n{\\text{Var}}\\{D_1(W,\\tilde{A},L)\\} & = E \\Big[ \\frac{ {{\\mathds 1}}(\\tilde{A}=1)^2 \\big\\{ E(Y \\mid L,W, A=a) - E(Y \\mid W,A=a) \\big\\}^2 }{P(\\tilde{A}=1)^2} \\Big] \\nonumber \\\\\n& = \\frac{E\\{{{\\mathds 1}}(\\tilde{A}=1)\\}}{P(\\tilde{A}=1)^2} E\\big[ \\big\\{ E(Y \\mid L,W, A=a) - E(Y \\mid W,A=a) \\big\\}^2 \\mid A=a \\big]. ",
"\\label{eq:thm:avar-proofuse-vard1-1}\\end{aligned}$$ By independent censoring we have $P(\\tilde{A}=1) = p_a p_l$. It then follows from (\\[eq:thm:avar-proofuse-vard1-1\\]) and randomization assumption that $${\\text{Var}}\\{D_1(W,\\tilde{A},L)\\} = \\frac{1}{p_a p_l} {\\text{Var}}_a\\big\\{ E_a(Y \\mid L,W) - E_a(Y \\mid W) \\big\\}. ",
"\\label{eq:thm:avar-proofuse-vard1}$$ Similarly, (\\[eq:thm:avar-proofuse-d2-simp\\]) together with randomization and monotone censoring imply $${\\text{Var}}\\{D_2(W,\\tilde{A},L,C^Y,Y)\\} = \\frac{1}{p_a p_y} {\\text{Var}}_a\\big\\{ Y - E_a(Y \\mid L,W) \\big\\}. ",
"\\label{eq:thm:avar-proofuse-vard2}$$\n\nBecause the semiparametric lower bound on the asymptotic variance for an estimand equals the variance of the efficient influence function, by (\\[eq:thm:avar-proofuse-2\\]), (\\[eq:thm:avar-proofuse-vard0\\]), (\\[eq:thm:avar-proofuse-vard1\\]), and (\\[eq:thm:avar-proofuse-vard2\\]) we proved (\\[eq:thm:var-bound\\]).",
"\n\nProof of Theorem \\[cor:re\\] {#proof:cor:re}\n---------------------------\n\nThe unadjusted estimator $\\widehat{\\tau}$ for $E(Y \\mid A=a)$ is defined as $$\\widehat{\\tau} = \\frac{\\sum_{i=1}^n Y_i {{\\mathds 1}}(A_i = a, C^Y_i = 1)}{\\sum_{i=1}^n {{\\mathds 1}}(A_i = a, C^Y_i = 1)}. ",
"\\label{def:unadj-noncausal}$$ Under Assumptions \\[assump:randomization\\] and \\[assump:independent-censoring\\], $\\widehat{\\tau}$ is unbiased: $$\\begin{aligned}\nE(\\widehat{\\tau}) = E \\Big[ E\\Big\\{ \\frac{\\sum_{i=1}^n Y_i {{\\mathds 1}}(A_i = a, C^Y_i = 1)}{\\sum_{i=1}^n {{\\mathds 1}}(A_i = a, C^Y_i = 1)} \\Big| A_1,\\ldots,A_n,C_{Y_1},\\ldots,C_{Y_n} \\Big\\} \\Big] = E(Y \\mid A = a).\\end{aligned}$$\n\nIn the following we calculate the asymptotic variance of $\\widehat{\\tau}$. $$\\begin{aligned}\n\\sqrt{n}\\{\\widehat{\\tau} - E(Y \\mid A=a)\\} & = \\frac{ \\frac{1}{\\sqrt{n}}\\sum_{i=1}^n Y_i {{\\mathds 1}}(A_i = a, C^Y_i = 1)}{ \\frac{1}{n}\\sum_{i=1}^n {{\\mathds 1}}(A_i = a, C^Y_i = 1)} - \\sqrt{n}E(Y \\mid A=a) \\nonumber \\\\\n&= \\frac{ \\frac{1}{\\sqrt{n}}\\sum_{i=1}^n \\{Y_i - E(Y \\mid A=a)\\} {{\\mathds 1}}(A_i = a, C^Y_i = 1)}{ \\frac{1}{n}\\sum_{i=1}^n {{\\mathds 1}}(A_i = a, C^Y_i = 1)}. ",
"\\label{eq:unadj-asymp-1}\\end{aligned}$$ By Weak Law of Large Numbers and the independent censoring assumption, $$\\frac{1}{n}\\sum_{i=1}^n {{\\mathds 1}}(A_i = a, C^Y_i = 1) \\stackrel{P}{\\to} p_a p_y, \\label{eq:unadj-asymp-2}$$ where $\\stackrel{P}{\\to}$ denotes convergence in probability. ",
"By Central Limit Theorem, $$\\frac{1}{\\sqrt{n}}\\sum_{i=1}^n \\{Y_i - E(Y \\mid A=a)\\} {{\\mathds 1}}(A_i = a, C^Y_i = 1) \\stackrel{d}{\\to} N(0, \\sigma^2), \\label{eq:unadj-asymp-3}$$ where by randomization and independent censoring we have $$\\begin{aligned}\n\\sigma^2 & = {\\text{Var}}\\big[ \\{Y - E(Y \\mid A=a)\\} {{\\mathds 1}}(A = a, C^Y = 1) \\big] = E \\big[ \\{Y - E(Y \\mid A=a)\\}^2 {{\\mathds 1}}(A = a, C^Y = 1)^2 \\big] \\nonumber \\\\\n& = p_a p_y {\\text{Var}}(Y \\mid A=a). ",
"\\label{eq:unadj-asymp-4}\\end{aligned}$$ Combining (\\[eq:unadj-asymp-1\\])-(\\[eq:unadj-asymp-4\\]), it follows from Slutsky’s theorem that $$\\sqrt{n}\\{\\widehat{\\tau} - E(Y \\mid A=a)\\} \\stackrel{d}{\\to} N\\big(0, (p_a p_y)^{-1} {\\text{Var}}(Y \\mid A=a)\\big). ",
"\\nonumber$$ So the asymptotic variance of $\\widehat{\\tau}$ is $(p_a p_y)^{-1} {\\text{Var}}(Y \\mid A=a)$, which by randomization yields (recall that by definition ${\\text{Var}}_a(Y) = {\\text{Var}}(Y \\mid A=a)$) $${\\text{AVar}}({\\text{unadjusted}}) = \\frac{1}{p_a p_y} {\\text{Var}}_a(Y). ",
"\\label{eq:cor-proofuse-1}$$ Equations (\\[eq:cor-proofuse-1\\]) and (\\[eq:rsqsum1\\]) imply $${\\text{AVar}}({\\text{unadjusted}}) = \\frac{1}{p_a p_y} {\\text{Var}}_a(Y) (R^2_{W;a} + R^2_{L\\mid W;a} + R^2_{r;a}). ",
"\\label{eq:cor-proofuse-2}$$\n\nOn the other hand, Theorem \\[thm:avar\\] and the definition of $R^2_W$ and $R^2_r$ imply $${\\text{AVar}}({\\text{efficient}}) = {\\text{Var}}_a(Y) ( R^2_{W;a} + \\frac{1}{p_a p_l} R^2_{L\\mid W;a} + \\frac{1}{p_a p_y} R^2_{r;a} ). ",
"\\label{eq:cor-proofuse-3}$$ Equations (\\[eq:cor-proofuse-2\\]), (\\[eq:cor-proofuse-3\\]), and (\\[eq:rsqsum1\\]) yield (\\[eq:re\\]). ",
"The proof is thus finished.",
"\n\nGeneralization and proof of Lemma \\[thm:avar-ate\\] {#proof:thm:avar-ate}\n--------------------------------------------------\n\nWe provide proof for a generalization of Lemma \\[thm:avar-ate\\], which allows for constant randomization probability other than $1/2$.\n\n\\[thm:avar-ate-2\\] Suppose Assumptions \\[assump:randomization\\], \\[assump:independent-censoring\\], and \\[assump:monotone-censoring\\] hold. ",
"Define $P(A=a) = p_a$ for $a \\in \\{0,1\\}$. The lower bound on the asymptotic variance of all RAL estimators of $E(Y\\mid A=1) - E(Y\\mid A=0)$ is $$\\begin{aligned}\n & {\\text{Var}}\\{E_1(Y \\mid W) - E_0(Y \\mid W) \\}\n+ \\sum_{a\\in\\{0,1\\}} \\frac{1}{p_a p_l} {\\text{Var}}_a\\{ E_a(Y \\mid L,W) - E_a(Y \\mid W) \\} \\nonumber \\\\\n+ & \\sum_{a\\in\\{0,1\\}} \\frac{1}{p_a p_y} {\\text{Var}}_a\\{ Y - E_a(Y \\mid L,W) \\}. ",
"\\label{eq:thm:var-bound-ate-2}\\end{aligned}$$\n\nFor notation simplicity, denote by $Q(\\cdot)$ the conditional expectation $E(Y\\mid \\cdot)$. Using the derivation in (\\[eq:thm:avar-proofuse-1\\])-(\\[eq:thm:avar-proofuse-d2-simp\\]) twice for $A=1$ and $A=0$, we get the efficient influence function $D(W,A,C^L,L,C^Y,Y)$ for $E(Y \\mid A=1) - E(Y \\mid A=0)$: $$D(W,A,C^L,L,C^Y,Y) = D_0(W) + D_1(W,A,C^L,L) + D_2(W,A,C^L,L,C^Y,Y),$$ where $$D_0(W) = \\big\\{ Q(W,A=1) - Q(A=1) \\big\\} - \\big\\{ Q(W,A=0) - Q(A=0) \\big\\}, \\label{eq:thm:avar-ate-proofuse-d0}$$ $$\\begin{aligned}\nD_1(W,A,C^L,L) = & \\frac{A C^L}{p_1 p_l} \\big\\{ Q(W,L,A=1) - Q(W,A=1) \\big\\} \\nonumber \\\\\n& - \\frac{(1-A)C^L}{p_0 p_l} \\big\\{ Q(W,L,A=0) - Q(W,A=0) \\big\\} , \\label{eq:thm:avar-ate-proofuse-d1}\\end{aligned}$$ and $$D_2(W,A,C^L,L,C^Y,Y) = \\frac{A C^Y}{p_1 p_y} \\big\\{ Y - Q(W,L,A=1) \\big\\} - \\frac{(1-A) C^Y}{p_0 p_y} \\big\\{ Y - Q(W,L,A=0) \\big\\}. ",
"\\label{eq:thm:avar-ate-proofuse-d2}$$\n\nThe following lemma states that $D_0$, $D_1$, and $D_2$ are pairwise uncorrelated.",
"\n\n\\[lem:thm-ate-proofuse\\]\n\nWe have $$\\begin{aligned}\n{\\text{Cov}}\\{ D_0(W), D_1(W,A,C^L,L)\\} & = 0, \\label{eq:lem:thm-ate-proofuse-toshow-d0d1}\\\\\n{\\text{Cov}}\\{ D_0(W), D_2(W,A,C^L,L,C^Y,Y)\\} & = 0, \\label{eq:lem:thm-ate-proofuse-toshow-d0d2}\\\\\n{\\text{Cov}}\\{ D_1(W,A,C^L,L), D_2(W,A,C^L,L,C^Y,Y)\\} & = 0. ",
"\\label{eq:lem:thm-ate-proofuse-toshow-d1d2}\\end{aligned}$$\n\nLemma \\[lem:thm-ate-proofuse\\] implies $$\\begin{aligned}\n{\\text{Var}}\\{D(W,A,C^L,L,C^Y,Y)\\} = & {\\text{Var}}\\{D_0(W)\\} + {\\text{Var}}\\{D_1(W,A,C^L,L)\\} \\nonumber \\\\\n& + {\\text{Var}}\\{D_2(W,A,C^L,L,C^Y,Y)\\}. ",
"\\label{eq:thm:avar-ate-proofuse-2}\\end{aligned}$$ By \\[eq:thm:avar-ate-proofuse-d0\\] we have $${\\text{Var}}\\{D_0(W)\\} = {\\text{Var}}\\{Q(W,A=1) - Q(W,A=0)\\}. ",
"\\label{eq:thm:avar-ate-proofuse-vard0}$$ By (\\[eq:thm:avar-ate-proofuse-d1\\]) we have $$\\begin{aligned}\n{\\text{Var}}\\{D_1(W,A,C^L,L)\\} = & E \\Big[ \\frac{A C^L}{p_1^2 p_l^2} \\big\\{ Q(W,L,A=1) - Q(W,A=1) \\big\\}^2 \\Big] \\nonumber \\\\\n& + E \\Big[ \\frac{(1-A) C^L}{p_0^2 p_l^2} \\big\\{ Q(W,L,A=0) - Q(W,A=0) \\big\\}^2 \\Big] \\label{eq:thm:avar-ate-proofuse-vard1-1} \\\\\n& = \\sum_{a\\in\\{0,1\\}} \\frac{1}{p_a p_l} E \\Big[ \\big\\{ Q(W,L,A=a) - Q(W,A=a) \\big\\}^2 \\mid A=a \\Big] \\label{eq:thm:avar-ate-proofuse-vard1-2} \\\\\n& = \\sum_{a\\in\\{0,1\\}} \\frac{1}{p_a p_l} {\\text{Var}}\\big\\{ Q(W,L,A=a) - Q(W,A=a) \\mid A=a \\big\\}. ",
"\\label{eq:thm:avar-ate-proofuse-vard1}\\end{aligned}$$ The step from (\\[eq:thm:avar-ate-proofuse-vard1-1\\]) to (\\[eq:thm:avar-ate-proofuse-vard1-2\\]) utilizes the independent censoring and randomization assumptions. ",
"Similarly, (\\[eq:thm:avar-ate-proofuse-d2\\]) together with randomization and independent censoring imply $${\\text{Var}}\\{D_2(W,A,C^L,L,C^Y,Y)\\} = \\sum_{a\\in\\{0,1\\}} \\frac{1}{p_a p_l} {\\text{Var}}\\big\\{ Y - Q(W,L,A=a) \\mid A=a \\big\\}. ",
"\\label{eq:thm:avar-ate-proofuse-vard2}$$\n\nBecause the semiparametric lower bound on the asymptotic variance for an estimand equals the variance of the efficient influence function, by (\\[eq:thm:avar-ate-proofuse-2\\]), (\\[eq:thm:avar-ate-proofuse-vard0\\]), (\\[eq:thm:avar-ate-proofuse-vard1\\]), and (\\[eq:thm:avar-ate-proofuse-vard2\\]) we proved Lemma \\[thm:avar-ate\\].",
"\n\nProof of Theorem \\[cor:re-ate\\] {#proof:cor:re-ate}\n-------------------------------\n\nThe unadjusted estimator for the average treatment effect is $$\\widehat{\\tau} = \\frac{\\sum_{i=1}^n Y_i {{\\mathds 1}}(A_i = 1, C^Y_i = 1)}{\\sum_{i=1}^n {{\\mathds 1}}(A_i = 1, C^Y_i = 1)} - \\frac{\\sum_{i=1}^n Y_i {{\\mathds 1}}(A_i = 0, C^Y_i = 1)}{\\sum_{i=1}^n {{\\mathds 1}}(A_i = 0, C^Y_i = 1)}. ",
"\\nonumber$$ Similar to the derivation from (\\[def:unadj-noncausal\\]) to (\\[eq:cor-proofuse-1\\]), when estimating the average treatment effect with $P(A=1) = P(A=0) = 1/2$, we have $${\\text{AVar}}({\\text{unadjusted}}) = \\frac{2}{p_y} \\sum_{a \\in \\{0,1\\}} {\\text{Var}}_a(Y).$$ The result in Theorem \\[cor:re-ate\\] then follows immediately from Lemma \\[thm:avar-ate\\].",
"\n\nProof of Corollary \\[cor:prognostic-predictive\\] {#proof:cor:prognostic-predictive}\n------------------------------------------------\n\n1. ",
" Because $$\\begin{aligned}\n {\\text{Var}}\\{E(Y \\mid W)\\} &= {\\text{Var}}\\{0.5 E(Y \\mid W, A=1) + 0.5 E(Y \\mid W, A=1)\\}, \\nonumber\\end{aligned}$$ ${\\text{Var}}\\{E(Y \\mid W)\\} = 0$ implies $$\\begin{aligned}\n {\\text{Var}}\\{E(Y \\mid W, A=1)\\} + {\\text{Var}}\\{E(Y \\mid W, A=0)\\} = -2{\\text{Cov}}\\{E(Y \\mid W, A=1),E(Y \\mid W, A=0). ",
"\\label{eq:proofuse:cor:prognostic-predictive-1}\\end{aligned}$$ By (\\[eq:proofuse:cor:prognostic-predictive-1\\]) and the definition of $\\gamma$ and $R^2_W$, we have $\\gamma = 2 R^2_W$. This combined with (\\[eq:re-ate\\]) with $R^2_{L \\mid W} = 0$ imply (\\[eq:rsq-onlyW-ate-predictive\\]).",
"\n\n2. ",
" By the definition of $\\gamma$ and ${\\text{Var}}\\{E(Y \\mid W, A=1) - E(Y \\mid W, A=0)\\} = 0$, (\\[eq:rsq-onlyW-ate-prognostic\\]) follows immediately from (\\[eq:re-ate\\]) with $R^2_{L \\mid W} = 0$.\n\nProof of statements in Section \\[impactcomparison\\] {#appen:proof-section3.4}\n---------------------------------------------------\n\nFor ease of reading we restate the claims to be proven in Section \\[impactcomparison\\]: We compare the ${\\text{ARE}}$ between two cases: $R^2_W = q >0, R^2_{L \\mid W} = 0$ (only baseline variable prognostic) and $R^2_W = 0, R^2_{L \\mid W} = q >0$ (only short-term outcome prognostic). ",
"Regardless of the value of $q>0$, the ${\\text{ARE}}$ in the former case is larger or equal to that in the latter case. ",
"Equality occurs if and only if $p_l=1$ and $W$ is uncorrelated with $Y$ (marginally) in the former case. ",
"The latter condition is equivalent to the treatment effect heterogeneity being the maximum possible $\\gamma=2R^2_W=2q$.\n\nWe first show that $\\gamma \\leq 2R^2_W$ with equality holds only when $E_1(Y\\mid W) = - E_0(Y \\mid W)$ almost surely. ",
"By Cauchy-Schwarz inequality we have $$\\begin{aligned}\n & {\\text{Var}}\\{ E_1(Y \\mid W) - E_0(Y \\mid W) \\} \\\\\n & = {\\text{Var}}\\{ E_1(Y \\mid W)\\} + {\\text{Var}}\\{ E_0(Y \\mid W) \\} - 2 {\\text{Cov}}\\{E_1(Y \\mid W), E_0(Y \\mid W) \\} \\\\\n & \\leq {\\text{Var}}\\{ E_1(Y \\mid W)\\} + {\\text{Var}}\\{ E_0(Y \\mid W) \\} + 2 \\left[{\\text{Var}}\\{ E_1(Y \\mid W)\\}{\\text{Var}}\\{ E_0(Y \\mid W) \\}\\right]^{1/2} \\\\\n & \\leq 2 {\\text{Var}}\\{ E_1(Y \\mid W)\\} + 2 {\\text{Var}}\\{ E_0(Y \\mid W) \\},\\end{aligned}$$ where both inequalities becomes equality if and only if $E_1(Y \\mid W) = - E_0(Y \\mid W)$ almost surely. ",
"This proves the claim.",
"\n\nTheorem \\[cor:re-ate\\] states that the ${\\text{ARE}}$ equals (where we indicate dependence on the arguments by explicitly writing them out) $${\\text{ARE}}(R^2_W, R^2_{L\\mid W}, \\gamma, p_y, p_l) = \\frac{1}{ 1 + (p_y/2) \\gamma - R^2_W - (1 - p_y / p_l) R^2_{L\\mid W} }. ",
"\\nonumber$$ When $R^2_W = q >0, R^2_{L \\mid W} = 0$, $$\\begin{aligned}\n {\\text{ARE}}(q, 0, \\gamma, p_y, p_l) = \\frac{1}{ 1 + (p_y/2) \\gamma - q } \\geq \\frac{1}{1 - (1 - p_y)q},\\end{aligned}$$ with equality holds if and only if $\\gamma = 2q$. When $R^2_W = 0, R^2_{L \\mid W} = q >0$ (which implies $\\gamma = 0$ because $\\gamma \\leq 2R^2_W$), $${\\text{ARE}}(0, R^2_{L\\mid W}, 0, p_y, p_l) = \\frac{1}{ 1 - (1 - p_y / p_l) q }. ",
"\\nonumber$$ Because $p_l \\leq 1$, ${\\text{ARE}}(q, 0, \\gamma, p_y, p_l) \\geq {\\text{ARE}}(0, R^2_{L\\mid W}, 0, p_y, p_l)$ with equality holds if and only if $p_l = 1$ and $\\gamma = 2q$, i.e., $E_1(Y\\mid W) = - E_0(Y \\mid W)$ almost surely.",
"\n\nLastly, we show that $E_1(Y\\mid W) = - E_0(Y \\mid W)$ almost surely implies that $W$ is uncorrelated with $Y$ (marginally): $$\\begin{aligned}\n & {\\text{Cov}}(Y, W) = E\\{Y (W - E(W))\\} = E\\{E(Y \\mid W) (W - E(W))\\} \\\\\n & = E\\{E(Y \\mid W) (W - E(W))\\} = E\\left[\\{0.5 E_1(Y\\mid W) + 0.5 E_0(Y\\mid W)\\} (W - E(W))\\right] = 0.\\end{aligned}$$ This completes the proof.",
"\n\nProof of Auxiliary Lemmas {#appen:proof-aux}\n=========================\n\nAdditional Supporting Lemmas\n----------------------------\n\n\\[lem:cond-exp\\]\n\nConsider three random variables $X$, $Y$, and $Z$. Denote by $\\sigma(Z)$ the $\\sigma$-field generated by $Z$. If $X \\in \\sigma(Z)$, then $$E\\big\\{Y E(Y \\mid X)\\big\\} = E\\big\\{ E(Y\\mid Z) E(Y\\mid X) \\big\\}. ",
"\\label{eq:cond-exp}$$\n\nBy the law of iterated expectation, we have $$E\\big\\{Y E(Y \\mid X)\\big\\} = E\\big[ E \\big\\{Y E(Y \\mid X) \\mid Z\\big\\} \\big]. ",
"\\label{eq:cond-exp-proofuse-1}$$ Because $X \\in \\sigma(Z)$, we have $E(Y \\mid X) \\in \\sigma(X) \\subset \\sigma(Z)$. This implies $$E \\big\\{Y E(Y \\mid X) \\mid Z\\big\\} = E(Y \\mid X) E (Y \\mid Z). ",
"\\label{eq:cond-exp-proofuse-2}$$ Equations (\\[eq:cond-exp-proofuse-1\\]) and (\\[eq:cond-exp-proofuse-2\\]) imply (\\[eq:cond-exp\\]). ",
"This completes the proof.",
"\n\n\\[lem:cond-exp-ate\\]\n\nConsider three random variables $W$, $L$, and $Y$. For any measurable functions $f(W)$ and $g(W,L)$, we have $$\\begin{aligned}\nE\\big[ f(W) \\big\\{ E(Y \\mid W,L) - E(Y \\mid W) \\big\\} \\big] & = 0, \\label{eq:cond-exp-ate-1} \\\\\nE\\big[ g(W,L) \\big\\{ Y - E(Y \\mid W,L) \\big\\} \\big] & = 0. ",
"\\label{eq:cond-exp-ate-2}\\end{aligned}$$\n\nBy the law of iterated expectation, we have $$\\begin{aligned}\nE \\big\\{ g(W) E(Y \\mid W,L) \\big\\} &= E \\big[ E \\big\\{ g(W) E(Y \\mid W,L) \\mid W \\big\\} \\big ] \\nonumber \\\\\n&= E \\big[ g(W) E \\big\\{ E(Y \\mid W,L) \\mid W \\big\\} \\big ] \\nonumber \\\\\n&= E \\big\\{ g(W) E(Y \\mid W) \\big\\}, \\nonumber\\end{aligned}$$ which proves (\\[eq:cond-exp-ate-1\\]).",
"\n\nSimilarly, we have $$E \\big\\{ g(W,L) Y \\big\\} = E \\big[ E \\big\\{ g(W,L) Y \\mid W,L \\big\\} \\big] = E \\big\\{ g(W,L) E(Y \\mid W,L) \\big\\}, \\nonumber$$ which proves (\\[eq:cond-exp-ate-2\\]).",
"\n\nProof of Lemma \\[lem:thm-proofuse\\] {#proof:lem:thm-proofuse}\n-----------------------------------\n\nBy (\\[eq:thm:avar-proofuse-d0-simp\\]) and (\\[eq:thm:avar-proofuse-d1-simp\\]) we have $${\\text{Cov}}(D_0, D_1) \\propto E \\big[ {{\\mathds 1}}(\\tilde{A}=1)\n\\{ E_a(Y \\mid W) - E_a(Y) \\} \\{ E_a(Y \\mid L,W) - E_a(Y \\mid W) \\} \\big]. ",
"\\label{eq:lem:thm-proofuse-d0d1}$$ Combining (\\[eq:vardecomp-proofuse-3\\]), (\\[eq:lem:thm-proofuse-d0d1\\]), and Assumptions \\[assump:randomization\\] and \\[assump:independent-censoring\\], we derive (\\[eq:lem:thm-proofuse-toshow-d0d1\\]).",
"\n\nBy (\\[eq:thm:avar-proofuse-d0-simp\\]) and (\\[eq:thm:avar-proofuse-d2-simp\\]) we have $${\\text{Cov}}(D_0, D_2) \\propto E \\big[ {{\\mathds 1}}(\\tilde{A}=1,C^Y=1)\n\\{ E_a(Y \\mid W) - E_a(Y) \\} \\{ Y - E_a(Y \\mid L,W) \\} \\big]. ",
"\\label{eq:lem:thm-proofuse-d0d2}$$ Combining (\\[eq:vardecomp-proofuse-2\\]), (\\[eq:lem:thm-proofuse-d0d2\\]), and Assumptions \\[assump:randomization\\] and \\[assump:independent-censoring\\], we derive (\\[eq:lem:thm-proofuse-toshow-d0d2\\]).",
"\n\nBy (\\[eq:thm:avar-proofuse-d1-simp\\]) and (\\[eq:thm:avar-proofuse-d2-simp\\]) we have $${\\text{Cov}}(D_1, D_2) \\propto E \\big[ {{\\mathds 1}}(\\tilde{A}=1,C^Y=1)\n\\{ E_a(Y \\mid L,W) - E_a(Y \\mid W) \\} \\{ Y - E_a(Y \\mid L,W) \\} \\big]. ",
"\\label{eq:lem:thm-proofuse-d1d2}$$ Combining (\\[eq:vardecomp-proofuse-1\\]), (\\[eq:lem:thm-proofuse-d1d2\\]), and Assumptions \\[assump:randomization\\] and \\[assump:independent-censoring\\], we derive (\\[eq:lem:thm-proofuse-toshow-d1d2\\]).",
"\n\nThis completes the proof.",
"\n\nProof of Lemma \\[lem:thm-ate-proofuse\\] {#proof:lem:thm-ate-proofuse}\n---------------------------------------\n\nFor notation simplicity, we use $E_1(\\cdot)$ and $E_0(\\cdot)$ to denote $E(\\cdot \\mid A=1)$ and $E(\\cdot \\mid A=0)$, respectively. ",
"By (\\[eq:thm:avar-ate-proofuse-d0\\]) and (\\[eq:thm:avar-ate-proofuse-d1\\]) we have $$\\begin{aligned}\n{\\text{Cov}}(D_0, D_1) \\propto & E \\big[ A \\big\\{ Q(W,A=1) - Q(W,A=0) \\big\\} \\big\\{ Q(W,L,A=1) - Q(W,A=1) \\big\\} \\big] \\nonumber \\\\\n& - E \\big[ (1-A) \\big\\{ Q(W,A=1) - Q(W,A=0) \\big\\} \\big\\{ Q(W,L,A=0) - Q(W,A=0) \\big\\} \\big] \\nonumber \\\\\n&= \\sum_{a\\in\\{0,1\\}} E_a \\big[ \\big\\{ E_1(Y \\mid W) - E_0(Y \\mid W) \\big\\} \\big\\{ E_a(Y \\mid W,L) - E_a(Y \\mid W) \\big\\} \\times P(A=a). ",
"\\label{eq:lem:thm-ate-proofuse-d0d1}\\end{aligned}$$ Both terms in (\\[eq:lem:thm-ate-proofuse-d0d1\\]) equals 0 by (\\[eq:cond-exp-ate-1\\]) in Lemma \\[lem:cond-exp-ate\\] with $f(W) = E_1(Y \\mid W) - E_0(Y \\mid W)$. This yields (\\[eq:lem:thm-ate-proofuse-toshow-d0d1\\]).",
"\n\nBy (\\[eq:thm:avar-ate-proofuse-d0\\]) and (\\[eq:thm:avar-ate-proofuse-d2\\]) we have $$\\begin{aligned}\n{\\text{Cov}}(D_0, D_2) \\propto & E \\big[ A \\big\\{ Q(W,A=1) - Q(W,A=0) \\big\\} \\big\\{ Y - Q(W,L,A=1) \\big\\} \\big] \\nonumber \\\\\n& - E \\big[ (1-A) \\big\\{ Q(W,A=1) - Q(W,A=0) \\big\\} \\big\\{ Y - Q(W,L,A=0) \\big\\} \\big] \\nonumber \\\\\n&= \\sum_{a\\in\\{0,1\\}} E_a \\big[ \\big\\{ E_1(Y \\mid W) - E_0(Y \\mid W) \\big\\} \\big\\{ Y - E_a(Y \\mid W,L) \\big\\} \\times P(A=a). ",
"\\label{eq:lem:thm-ate-proofuse-d0d2}\\end{aligned}$$ Both terms in (\\[eq:lem:thm-ate-proofuse-d0d2\\]) equals 0 by (\\[eq:cond-exp-ate-2\\]) in Lemma \\[lem:cond-exp-ate\\] with $g(W,L) = E_1(Y \\mid W) - E_0(Y \\mid W)$. This yields (\\[eq:lem:thm-ate-proofuse-toshow-d0d2\\]).",
"\n\nBy (\\[eq:thm:avar-ate-proofuse-d1\\]) and (\\[eq:thm:avar-ate-proofuse-d2\\]) we have $$\\begin{aligned}\n{\\text{Cov}}(D_1, D_2) \\propto & E \\big[ A \\big\\{ Q(W,L,A=1) - Q(W,A=1) \\big\\} \\big\\{ Y - Q(W,L,A=1) \\big\\} \\big] \\nonumber \\\\\n& - E \\big[ (1-A) \\big\\{ Q(W,L,A=0) - Q(W,A=0) \\big\\} \\big\\{ Y - Q(W,L,A=0) \\big\\} \\big] \\nonumber \\\\\n&= \\sum_{a\\in\\{0,1\\}} E_a \\big[ \\big\\{ E_a(Y \\mid W,L) - E_a(Y \\mid W) \\big\\} \\big\\{ Y - E_a(Y \\mid W,L) \\big\\} \\times P(A=a). ",
"\\label{eq:lem:thm-ate-proofuse-d1d2}\\end{aligned}$$ Both terms in (\\[eq:lem:thm-ate-proofuse-d1d2\\]) equals 0 by (\\[eq:cond-exp-ate-2\\]) in Lemma \\[lem:cond-exp-ate\\] with $g(W,L) = E_a(Y \\mid W,L) - E_a(Y \\mid W)$. This yields (\\[eq:lem:thm-ate-proofuse-toshow-d1d2\\]).",
"\n\nThis completes the proof.",
"\n\n[^1]: Department of Statistics, Harvard University. ",
"qiantianchen@fas.harvard.edu\n\n[^2]: Department of Biostatistics, Johns Hopkins University.",
"\n\n[^3]: Vatic Labs, New York.",
"\n"
] |
{
"pile_set_name": "ArXiv"
}
|
[
0.000551402335986495,
0.0005811179871670902,
0.0005382592789828777,
0.0006088991649448872,
0.0006535504944622517,
0.0005746125243604183,
0.0005994821549393237,
0.0005939181428402662,
0.0005535827367566526,
0.0005390347214415669,
0.0008055515354499221,
0.0005407682037912309,
0.0005187744973227382,
0.0006588387186639011,
0.0006052831304259598,
0.0005822200910188258,
0.0005876081995666027,
0.0005878065130673349,
0.0005771228461526334,
0.0005256467266008258,
0.0006164649385027587,
0.0005859589437022805,
0.0005914000212214887,
0.0005779981147497892,
0.0005799957434646785,
0.0005358292837627232,
0.0006362742860801518,
0.0005361775401979685,
0.000586594280321151,
0.0005508105386979878,
0.0006234230822883546,
0.0005791062139905989,
0.0005975489038974047,
0.0006200451753102243,
0.0005597994895651937,
0.0005559600540436804,
0.000576195539906621,
0.0013671473134309053,
0.0007456312887370586,
0.0009004038292914629,
0.0006127195665612817,
0.000806130119599402,
0.0005570989451371133,
0.0007180136744864285,
0.003125799121335149,
0.0008549421327188611,
0.0009182561188936234,
0.000575014972127974,
0.0008917191298678517,
0.0005458834930323064,
0.00993399415165186,
0.0006862103473395109,
0.000615466560702771,
0.0005454039783217013,
0.0005830486188642681,
0.0006097517325542867,
0.0038013977464288473,
0.00316808489151299,
0.0015284634428098798,
0.0017979186959564686,
0.0009823959553614259,
0.007797676604241133,
0.028353752568364143,
0.0010329816723242402,
0.0006265780539251864,
0.0006571683916263282,
0.0007382100448012352,
0.002471188548952341,
0.0021995699498802423,
0.0009484839392825961,
0.007592506241053343,
0.01566900871694088,
0.0013107511913403869,
0.047963947057724,
0.0006192157161422074,
0.006381366401910782,
0.0013303020969033241,
0.0007684368174523115,
0.0023074436467140913,
0.0007982406532391906,
0.0011135849636048079,
0.0005235827411524951,
0.0006061357562430203,
0.0005850217421539128,
0.0007360552554018795,
0.000693732697982341,
0.0008124447776935995,
0.0010254429653286934,
0.0008674113196320832,
0.0005881769466213882,
0.003844393650069833,
0.0011349688284099102,
0.0012155105359852314,
0.0009894538670778275,
0.0006506310892291367,
0.0006862231530249119,
0.0032360292971134186,
0.0006330882897600532,
0.000899669190403074,
0.0006072668475098908,
0.0006786760059185326,
0.0006555791478604078,
0.004030948039144278,
0.0006748587475158274,
0.001322112395428121,
0.0006588943069800735,
0.0006309475284069777,
0.0010550995357334614,
0.0006473732646554708,
0.0009557505836710334,
0.0007194606005214155,
0.0010189828462898731,
0.0007169985910877585,
0.0005722053465433419,
0.0011867852881550789,
0.0007646591402590275,
0.0007225408917292953,
0.0007710132631473243,
0.0006854080129414797,
0.0005952297360636294,
0.0005640835152007639,
0.0005942349089309573,
0.0006390208727680147,
0.0006215394241735339,
0.0009680602815933526,
0.0018040764844045043,
0.0006544517818838358,
0.0007511273724958301,
0.0006393674993887544,
0.0007118202629499137,
0.0007511195144616067,
0.0005630025407299399,
0.0009771399199962616,
0.0010419372702017426,
0.0016901401104405522,
0.04125811159610748,
0.0007252084906212986,
0.0006324418354779482,
0.0028635659255087376,
0.0009568530949763954,
0.0005626045167446136,
0.0006194612942636013,
0.0006812431383877993,
0.0006991575937718153,
0.0006371700437739491,
0.0006308624288067222,
0.0007393364794552326,
0.0005949166952632368,
0.0006139485631138086,
0.0009470603545196354,
0.0005836669588461518,
0.0007188626332208514,
0.0006740045500919223,
0.0006265004049055278,
0.0009271990275010467,
0.0016329881036654115,
0.0016167615540325642,
0.0005764832021668553,
0.0008423950057476759,
0.0006857605185359716,
0.0007072300068102777,
0.0005559998680837452,
0.0006695648189634085,
0.0005929470062255859,
0.0006008644122630358,
0.0006206544348970056,
0.0007551501039415598,
0.0005614191177301109,
0.0005811454029753804,
0.0005771704018115997,
0.001185585861094296,
0.0005966645549051464,
0.000557677005417645,
0.001269704895094037,
0.0006063521723262966,
0.0005629287916235626,
0.0006478021969087422,
0.0006313794292509556,
0.0031432348769158125,
0.0007849651738069952,
0.000554640602786094,
0.0010614409111440182,
0.0006533897831104696,
0.0005635923589579761,
0.0007002364727668464,
0.0006366209127008915,
0.000662033271510154,
0.000600865576416254,
0.0006108804955147207,
0.0006083945045247674,
0.0016466445522382855,
0.0008110076887533069,
0.0006462113233283162,
0.0011364336824044585,
0.000848292896989733,
0.0005597106646746397,
0.000662222970277071,
0.0005868792068213224,
0.0005713298451155424,
0.0006278340006247163,
0.0006482524331659079,
0.0006403700681403279,
0.0006125770742073655,
0.0010998479556292295,
0.0009651772561483085,
0.0012539272429421544,
0.0020204102620482445,
0.0007998968940228224,
0.0005306961829774082,
0.0006680481601506472,
0.0008993354858830571,
0.0006464961916208267,
0.0006159389740787446,
0.0006187054677866399,
0.00062932213768363,
0.0006567025557160378,
0.0005607425700873137,
0.0008351734722964466,
0.0008483813726343215,
0.0005952212959527969,
0.0005403843824751675,
0.0006079464219510555,
0.0008612046367488801,
0.0006320805987343192,
0.0008826101548038423,
0.0005537692923098803,
0.0005905826692469418,
0.0005734419682994485,
0.0006123074563220143,
0.000571709533687681,
0.0006209769635461271,
0.0005451989127323031,
0.0005700659821741283,
0.0005620947922579944,
0.0005896480870433152,
0.0010838541202247143,
0.0007046060636639595,
0.0005629408988170326,
0.0005481478874571621,
0.0006086502107791603,
0.0006543601048178971,
0.0006369350012391806,
0.0005534313386306167,
0.0006228547426871955,
0.0005613730172626674,
0.0005759194027632475,
0.0005804196698591113,
0.000564908841624856,
0.0005877958610653877,
0.0006139342440292239,
0.000715371803380549,
0.000734084052965045,
0.0024785439018160105,
0.00059665204025805,
0.0017422348028048873,
0.0006624283851124346,
0.0007308403728529811,
0.0014141477877274156,
0.0009017689153552055,
0.0008500183466821909,
0.0018846581224352121,
0.038854874670505524,
0.19536344707012177,
0.25781840085983276,
0.012963971123099327,
0.0008172106463462114,
0.0009456163388676941,
0.0005975036765448749,
0.0005606712657026947,
0.0008509040926583111,
0.0005857079522684216,
0.006073240656405687,
0.0038556531071662903,
0.0006679119542241096,
0.0005771704018115997,
0.001185585861094296,
0.0005954629741609097,
0.0005940500413998961,
0.0007588728913106024,
0.0005908723105676472,
0.000600861560087651,
0.0005503715947270393,
0.0005431838217191398,
0.0007935075554996729,
0.0005913918139412999,
0.0006213075830601156,
0.0007339874864555895,
0.0006292941980063915,
0.0005867799045518041,
0.0007132374448701739,
0.0006935806013643742,
0.0008132802322506905,
0.0006442180019803345,
0.0008190477383323014,
0.000842807290609926,
0.000555583683308214,
0.0005557553377002478,
0.001269704895094037,
0.0006079261656850576,
0.0005937092937529087,
0.0008389150025323033,
0.0007654509390704334,
0.0006464860634878278,
0.000811561883892864,
0.0007984045660123229,
0.000808390905149281,
0.0016406594077125192,
0.0007924383389763534,
0.0006452607922255993,
0.0006525145727209747,
0.0007135171908885241,
0.0006065521738491952,
0.0019117676420137286,
0.0008205382036976516,
0.0007255693199113011,
0.0006632597651332617,
0.000754646840505302,
0.0006210855790413916,
0.0007477097678929567,
0.0006270924350246787,
0.0009806629968807101,
0.001078184461221099,
0.004553331062197685,
0.0010430014226585627,
0.0009391900966875255,
0.0006091484683565795,
0.019464686512947083,
0.0011318717151880264,
0.0006419276469387114,
0.013829757459461689,
0.0008223769254982471,
0.0007649500039406121,
0.0007153333281166852,
0.0006379018304869533,
0.0008769270498305559,
0.002131375018507242,
0.009155341424047947,
0.0007746661431156099,
0.009590511210262775,
0.002255477709695697,
0.00800847727805376,
0.0042640757746994495,
0.00084189313929528,
0.0009868912165984511,
0.000856242433656007,
0.0007421725313179195,
0.012349638156592846,
0.0006350326584652066,
0.0006938658771105111,
0.0006075261626392603,
0.0006227143458090723,
0.0005853070179000497,
0.0006893155514262617,
0.0009046168997883797,
0.005645815748721361,
0.0009111089166253805,
0.0007650082698091865,
0.0008554208325222135,
0.0005523167783394456,
0.0007736484403721988,
0.001046955119818449,
0.0006846755859442055,
0.014430413953959942,
0.0007935793837532401,
0.004400831647217274,
0.007346542552113533,
0.005388077348470688,
0.019850218668580055,
0.0055444613099098206,
0.0009758769301697612,
0.0015256991609930992,
0.005661977920681238,
0.007841982878744602,
0.037142880260944366,
0.029865091666579247,
0.06606198847293854,
0.001278056064620614,
0.037579528987407684,
0.0011178868589922786,
0.050703346729278564,
0.03895123675465584,
0.001137048122473061,
0.0009523913613520563,
0.001999996602535248,
0.0018743440741673112,
0.0017531777266412973,
0.04683160036802292,
0.020583685487508774,
0.0009916570270434022,
0.010702408850193024,
0.013757659122347832,
0.0017646771157160401,
0.0703904777765274,
0.022755607962608337,
0.028437796980142593,
0.001200117520056665,
0.012210801243782043,
0.20879371464252472,
0.004021580331027508,
0.014896159060299397,
0.24525432288646698,
0.03287171944975853,
0.012266023084521294,
0.01410844549536705,
0.0017170502105727792,
0.0006777239614166319,
0.004915405996143818,
0.11571520566940308,
0.04098803177475929,
0.0009808378526940942,
0.007987537421286106,
0.014553177170455456,
0.003568087238818407,
0.13712641596794128,
0.0009182015201076865,
0.019983965903520584,
0.0011422389652580023,
0.020738903433084488,
0.003092752769589424,
0.0013617753284052014,
0.02806553803384304,
0.0025862690526992083,
0.001185585861094296,
0.0017920969985425472,
0.0007813534466549754,
0.0007478600600734353,
0.0014417849015444517,
0.014352833852171898,
0.0007304302416741848,
0.007440466433763504,
0.019975734874606133,
0.010449306108057499,
0.00944708101451397,
0.014686482958495617,
0.02675272710621357,
0.026109764352440834,
0.001518465462140739,
0.0015256991609930992,
0.03392180800437927,
0.05565278232097626,
0.06432954967021942,
0.02224118635058403,
0.0022660873364657164,
0.01212739385664463,
0.0021944455802440643,
0.015418889001011848,
0.0019024541834369302,
0.0015256991609930992,
0.004353970754891634,
0.09244323521852493,
0.005516575183719397,
0.12484148889780045,
0.007358120754361153,
0.11550917476415634,
0.0066011818125844,
0.0015256991609930992,
0.0006426788168027997,
0.0006997278542257845,
0.0007052976870909333,
0.001995444530621171
] | 0.007125 | 460 |
[
"Q:\n\nhtml table export to excel borders are not showing in excelsheet\n\nI am exporting a html table to excel-sheet.",
"\nEverything is working fine but borders in excel-sheet is not visible.",
"\n\n$(\"#btnExport\").click(function (e) {\r\n window.open('data:application/vnd.ms-excel,' + $('#dvData').html());\r\n e.preventDefault();\r\n});\nbody {\r\n font-size: 12pt;\r\n font-family: Calibri;\r\n padding : 10px;\r\n}\r\ntable {\r\n border: 1px solid black;\r\n}\r\nth {\r\n border: 1px solid black;\r\n padding: 5px;\r\n background-color:grey;\r\n color: white;\r\n}\r\ntd {\r\n border: 1px solid black;\r\n padding: 5px;\r\n}\r\ninput {\r\n font-size: 12pt;\r\n font-family: Calibri;\r\n}\n<script src=\"https://cdnjs.cloudflare.com/ajax/libs/jquery/3.2.1/jquery.min.js\"></script>\r\n<input type=\"button\" id=\"btnExport\" value=\" Export Table data into Excel \" />\r\n<br/>\r\n<br/>\r\n<div id=\"dvData\">\r\n <table>\r\n <tr>\r\n <th>Column One</th>\r\n <th>Column Two</th>\r\n <th>Column Three</th>\r\n </tr>\r\n <tr>\r\n <td>row1 Col1</td>\r\n <td>row1 Col2</td>\r\n <td>row1 Col3</td>\r\n </tr>\r\n <tr>\r\n <td>row2 Col1</td>\r\n <td>row2 Col2</td>\r\n <td>row2 Col3</td>\r\n </tr>\r\n <tr>\r\n <td>row3 Col1</td>\r\n <td>row3 Col2</td>\r\n <td>row3 Col3</td>\r\n </tr>\r\n </table>\r\n</div>\n\nJsfiddle\nI've tried to change px into em and increasing their size, but they didn't worked.",
"\nNote: I am using MS Office 2013.",
"\n\nA:\n\n<meta http-equiv=\"content-type\" content=\"text/plain; charset=UTF-8\"/>\n\nMay be useful below script : \n<script type=\"text/javascript\">\n var tableToExcel = (function() {\n var uri = 'data:application/vnd.ms-excel;base64,'\n , template = '<html xmlns:o=\"urn:schemas-microsoft-com:office:office\" xmlns:x=\"urn:schemas-microsoft-com:office:excel\" xmlns=\"http://www.w3.org/TR/REC-html40\"><head><!--[if gte mso 9]><xml><x:ExcelWorkbook><x:ExcelWorksheets><x:ExcelWorksheet><x:Name>{worksheet}</x:Name><x:WorksheetOptions><x:DisplayGridlines/></x:WorksheetOptions></x:ExcelWorksheet></x:ExcelWorksheets></x:ExcelWorkbook></xml><![endif]--><meta http-equiv=\"content-type\" content=\"text/plain; charset=UTF-8\"/></head><body><table>{table}</table></body></html>'\n , base64 = function(s) { return window.btoa(unescape(encodeURIComponent(s))) }\n , format = function(s, c) { return s.replace(/{(\\w+)}/g, function(m, p) { return c[p]; }) }\n return function(table, name) {\n if (!",
"table.nodeType) table = document.getElementById(table)\n var ctx = {worksheet: name || 'Worksheet', table: table.innerHTML}\n window.location.href = uri + base64(format(template, ctx))\n }\n })()\n </script>\n\nLive Example : http://jsfiddle.net/cmewv/537/\n\n"
] |
{
"pile_set_name": "StackExchange"
}
|
[
0.0006219136412255466,
0.0006035135011188686,
0.002369845751672983,
0.0005680656176991761,
0.000837914296425879,
0.0007138436194509268
] | 0.000953 | 6 |
[
"Tuesday, August 24, 2010\n\nMy mother lives in Brunswick, Maine, roughly two-and-a-half hours from my home in Londonderry, New Hampshire. ",
"With a full-time job, writing, three kids and a granddaughter who lives with us, it's hard to get up to see her as often as I should, but this past Monday I was cruising up the Maine Turnpike, mid-morning, coffee in hand, when it occurred to me I was hungry. ",
"I had driven an hour-and-a-half, after all, with no breakfast.",
"Luckily for me, there's a Burger King at the traveler's plaza at Exit Three, so I pulled in, grabbed an order of French Toast Sticks, and hit the road again. ",
"They were pretty good, too.",
"Does everyone else know they serve maple syrup with those things? ",
"Because I sure didn't. ",
"But it wasn't a problem until I had finished eating, at which point I packed up my napkin in the little box the French Toast Sticks came in and stuck it in the Burger King paper bag and rolled the bag into a ball to throw it on the floor of the car......and ruptured the little packet the syrup came in. ",
"It oozed out over my hands, flowing thickly onto my lap and under my butt, soaking into the seat. ",
"At seventy miles an hour. ",
"In the passing lane.",
"By the time I had pulled back across traffic into the breakdown lane and got the car stopped - getting sticky syrup all over the steering wheel and turn signal, by the way - my jeans looked like, well, you can use your imagination if you really feel the need.",
"I found a sweater one of my daughters had left in the back seat, sopped up as much of the syrup as I could, all with the traffic zooming by a couple of feet away, and then sat on the sweater and drove the rest of the way to Brunswick, where I walked into Sears and bought a new pair of jeans. ",
"Then I drove to my mother's house.",
"I think I would rather have gone hungry.----------There's only one week left to enter my website contest to win a free copy of every book I ever publish...check out the details at www.allanleverone.com...\n\nAbout Me\n\nI'm a 53 year old author living in New Hampshire with my wife Sue, three children, one adorable granddaughter and a cat who has used up eight lives.",
"\nI'm the author of the Amazon Top 25 overall paid bestseller, THE LONELY MILE, as well as three other novels, three novellas and two story collections.",
"\nA 2012 Derringer Award winner for excellence in short mystery fiction as well as a 2011 Pushcart Prize nominee, my work has been featured in Needle: A Magazine of Noir, Shroud Magazine, Shotgun Honey, Morpheus Tales, Twisted Dreams, Mysterical-E and many others, as well as the print anthologies INTRIGUE, TEN FOR TEN and NORTHERN HAUNTS.",
"\nI am a fan of Lawrence Block, Donald Westlake, Lee Child, Tom Piccirilli, Dave Zeltserman and Vincent Zandri, among many others."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0006616573664359748,
0.0006495910929515958,
0.001320359413512051,
0.0008759023039601743,
0.0005854956107214093,
0.0006139290053397417,
0.0008490533800795674,
0.0012412481009960175,
0.7251771092414856,
0.0007159344968385994,
0.0007664699223823845,
0.0012013145023956895,
0.0012400327250361443,
0.009102017618715763,
0.0008744334918446839,
0.0006115314899943769,
0.0007635520305484533,
0.0007366615463979542
] | 0.041555 | 18 |
[
"Rainforest Plants – Coconut Palm\n\nGeneral Description: The Coconut Palm Tree's scientific name is Cocos nucifera. ",
"It is a slender, tall tree, growing to 100 feet and topped with palm fronds at the top of the trunk. ",
"Botanists believe the Coconut Palm evolved in the Indo-Pacific Ocean region because its genetic diversity is most obvious there.",
"\n\nNative to Southeast Asia and the southwest Pacific, Cocos nucifera now grows in the tropics around the world. ",
"This palm is dispensed by water, which is why most Coconut Palms grace beaches where its fruit washed to shore.",
"\n\nCoconut palms are warm weather trees, having little tolerance for the cold. ",
"Botanists believe they cross-pollinate, with some dwarf species probably self-pollinating.",
"\n\nUses: Coconut Palms are known for their wood, milk and medicinal value. ",
"Wood from the trunks provide material for building bridges (small), because they are fairly straight, strong and resistant to salt. ",
"The fruit most of us know as 'coconut' grows from the center of the fronds where the fronds attach to the trunk. ",
"Coconut milk is made from grated coconut, and adding hot water or milk; coconut oil and other compounds are extracted in this process. ",
"Coconut pulp and milk are highly nutritious, with an abundance of fiber, vitamins and minerals.",
"\n\nCoconut is also valued as an herbal remedy in Pakistan to treat rat bites. ",
"These fruits are used broadly in Hindu religious ceremonies, as an offering to the gods. ",
"They are also smashed on the ground or on a structure as an initiation or inauguration of a new project, much like Western civilization uses champagne on the bow of a ship. ",
"Coconut oil is used broadly in cooking, soaps and skin products; it is known for its healing properties. ",
"The oil is extracted and processed for worldwide distribution.",
"\n\nCoconuts have an extensive history, going back thousands of years, and are considered by many civilizations as a primary and valuable source of food.",
"\n\nDisclaimer: The statements contained herein have not been evaluated by the Food and Drug Administration.",
"\n\nAny reference to medicinal use is not intended to treat, cure, mitigate or prevent any disease."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0006300494424067438,
0.0007555981865152717,
0.0005817500059492886,
0.0008530725026503205,
0.0007018412579782307,
0.0007358427974395454,
0.001743373810313642,
0.0005584200844168663,
0.0006486269412562251,
0.0006067432695999742,
0.0006148146349005401,
0.0005823450046591461,
0.005017520859837532,
0.0006098537123762071,
0.0015399130061268806,
0.0005899755633436143,
0.0006474328110925853,
0.0005790297291241586,
0.0006601187051273882,
0.0007338039577007294
] | 0.00097 | 20 |
[
"Introduction\n============\n\nImplantation is one of the most interesting biological events ([@B1]) and its failure remains a major problem in infertility treatment. ",
"One of the reasons for this failure is suspected to be due to impaired uterine receptivity because of high serum estradiol concentrations induced by ovulation induction treatments ([@B2]).",
"\n\nOne specific morphologic marker that has been proposed to be associated with the window of implantation is the appearance of pinopodes ([@B3]). ",
"Pinopodes are surface projections of the endometrial cells involved in uterine pinocytosis in mice and rats, but not in cows, humans or rabbits ([@B4]). ",
"Presence and development of pinopodes mainly depend on the ovarian hormones, especially progesterone ([@B5]). ",
"Pinopodes on the endometrial surface have been suggested to be ultrastructural markers of the implantation window ([@B6]). ",
"The surfaces of pinopodes may have some receptors for adhesion molecules, which are essential for embryo implantation. ",
"In mice, ovarian hyperstimulation is known to have a detrimental effect on the formation and disappearance of pinopodes ([@B4]). ",
"Administration of exogenous gonadotropic hormones such as human menopausal gonadotropin) )HMG ( and human chorionic gonadotropic (HCG) leads to increased secretion of estrogen and progesterone. ",
"Steroid hormones and their receptors have been suggested to be involved in the regulation of pinopode formation ([@B7]). ",
"Implantation failure remains an unsolved problem in reproductive medicine and is considered as a major cause of infertility in otherwise healthy women ([@B8]). ",
"The vasodilatory effect of sildenafil may also improve perfusion of uterus and ovaries after application to women with poor endometrial response and reduced uterine blood flow ([@B9]).",
"\n\nPhosphodiesterase (PDE) is a family of isoenzymes that hydrolyzes cAMP and cGMP. ",
"Specific inhibitors of PDE subtypes have been identified that can augment the effects of cyclic nucleotides on target tissues, such as human spermatozoa ([@B10]). ",
"Sildenafil citrate (Viagra) is a newly developed, type 5-specific PDE inhibitor that prevents the breakdown of cGMP and potentiates the effects of NO (Nitric Oxide) on vascular smooth muscle ([@B10]). ",
"Since its introduction in 1997, sildenafil has been used with great success in the treatment of male erectile dysfunction ([@B11]), but fewer researches have evaluated its effects in woman. ",
"The availability of sildenafil has enabled us to reap the benefits of NO on the uterus, while minimizing its side effects ([@B10]). ",
"Effects of sildenafil on perfusion of uterus and ovaries are controversial. ",
"Early reports on the benefit of sildenafil in assisted reproduction should be evaluated by placebo-controlled studies ([@B9]). ",
"In addition; there may be a role for sildenafil in the treatment of female sexual dysfunction ([@B12]). ",
"The aim of the present study was to find out whether mice uterine pinopodes are affected ultrastructurally by treatment with sildenafil citrate during immediately before implantation ([@B13]).",
"\n\nMaterials and Methods\n=====================\n\n***Animals***\n\nThirty adult (3.5 months) Syrian female and 20 adult Syrian male mice (mean weight, 25±5 g) were kept under standard laboratory conditions. ",
"The mice were acclimatized for 1 week under a 12 hr: 12 hr light: dark cycle at room temperature of 22±2 °C. ",
"Female mice were randomly divided into 3 groups.",
"\n\n***Group A***\n\n*Non-stimulated control group*\n\nA total number of 10 mice were placed in the control group and, every 2 female mice were placed with 1 male mouse in one cage, for mating.",
"\n\n***Group B***\n\n*Hyperstimulated group*\n\nThe mice of this group (n=10) were hyperstimulated by intraperitoneal (IP) injection of 7.5 IU HMG (Menogon, Ferring, Pharmaceuticals, Germany) and HCG (PREGNYL, Organon, Netherlands) at an interval of 48 hr.",
"\n\n***Group C***\n\n*Hyperstimulated and sildenafil citrate administrated group*\n\nIn this group the mice (n=10) were hyperstimulated and mated in the same way as group B, then they received IP injections of sildenafil citrate (3 mg/mouse) (Rouz Darou, Iran) at 24, 48 and 72 hr interval after HMG injection.",
"\n\n***Tissue preparation***\n\nNinety six hr after HMG injection, the mice in experimental groups together with control group mice were sacrificed and their uterine were flashed for blastocyst. ",
"Uterine specimen, only from those whose uterine contained blastocyst (at least 10 in each group) was prepared (the middle one-third) transmission electron microscope (TEM) for as follows:\n\nSamples were washed by normal saline and afterwards were washed with phosphate buffer for 10 min (three times) and were cut into 1×1×1mm by scalpel (all samples were taken from the middle third of the uterus mice from both sides). ",
"Tissues were fixed with 2% glutaraldehyde in buffered paraformaldeide (Thuringowa, Australia) at room temperature for 30 min, and at 4 °C overnight, and then post-fixed in 1% osmium tetroxide for 2 hr at dark room. ",
"Specimens were washed with 0.1 M phosphate buffer (pH=7.4) for 10 min (three times). ",
"The tissues were dehydrated by increasing concentrations (30 to 100%) of ethanol followed by propylene oxide then were embedded in pure resin. ",
"Semithin sections (500 nm thick) were stained with 2% toluidine blue. ",
"Ultrathin sections (90-150 nm thick) (ultramicrotome, Zeiss, Germany) were stained with uranyl acetate and aqueous lead citrate. ",
"The sections were examined using Zeiss transmission electron microscope (LEO 906, Germany).",
"\n\nResults\n=======\n\n***Non-stimulated control group***\n\nIn the non- hyperstimulated mice of group A, epithelial cells (luminal) of uterine were oval with euchromatin nuclei surrounded by thin marginal heterochromatin. ",
"Their cytoplasm was rich in organelles such as rough endoplasmic reticulum, polyribosome and mitochondria. ",
"The microvilli of the uterine epithelium cells were abundant and large, 96 hr after mating ([Figure 1](#F1){ref-type=\"fig\"} and [Figure 2. ",
"A](#F2){ref-type=\"fig\"}). ",
"No developed pinopodes in luminal cells microvillis were observed in this group ([Figure 1](#F1){ref-type=\"fig\"}).",
"\n\n***Hyperstimulated*** ***group***\n\nIn the hyperstimulated mice of group B, 96 hr after HMG injection, the microvilli decreased in number and length. ",
"Smooth and slender membrane projections form, arising from the entire cell apex (developing pinopodes). ",
"Some cells had no microvilli and were transformed to pinopodes. ",
"In this group the pinopodes were well developed and some of them had short microvilli on their surface. ",
"The nuclei of this group like group A, were oval and mainly euchromatic with a prominent nucleolus and heterochromatin dispersed peripherally; and the cytoplasmic organelles were found in the cells ([Figure 2. ",
"C](#F2){ref-type=\"fig\"}).",
"\n\n***Hyperstimulated*** ***+*** ***sildenafil citrate*** ***injected group***\n\nIn the hyperstimulated + sildenafil citrate injected groups (group C) 96 hr after HMG injection the short microvilli covered with glycocalyx were seen. ",
"In this group the apical surface of cells were similar to stimulated group and different from non-stimulated one. ",
"The euchromatic and oval nuclei with distinctive nucleoli and double membranes were situated at the mid -part of the cells. ",
"The organelles of the samples from hyperstimulated + sildenafil citrate injected animals were the same as control group ([Figure 2. ",
"D](#F2){ref-type=\"fig\"}).",
"\n\n{#F1}\n\n{#F2}\n\nDiscussion\n==========\n\nThe surface of endometrium undergoes cyclic changes which influences the embryo attachment and implantation. ",
"High levels of hormone produced during ovarian hyperstimulation, may disturb the structure of the endometrium ([@B5]). ",
"The relative contributions of the endometrium for successful implantation are not known, and there are no accepted criteria for the evaluation of endometrial receptivity ([@B14]). ",
"The presence and development of pinopodes is dependent on the ovarian hormones, especially progesterone ([@B15]). ",
"The premature appearance of pinopodes after hyperstimulation has been reported in some studies, but Nikas *et al* showed that the ovarian stimulation did not affect the quantity and life span of the endometrial pinopodes in human ([@B16]).",
"\n\nOur ultrastructural studies showed that 96 hr after mating or HMG injection, pinopodes were not visible in the control group and all the epithelial cells had microvilli on the apical surface. ",
"But hyperstimulated group had pinopodes at the same time. ",
"The pinopodes were observed for a short time, 24 to 48 hr, during implantation in mammals, ([@B17]) depending on the ovarian hormones, especially progesterone ([@B18]).",
"\n\nKolb *et al* reported that although the ultrastructural features of the endometrium in the luteal phase of the ovarian hyperstimolation are better than natural phases, but may shift the window of implantation. ([",
"@B6]). ",
"In contrast, there are some reports on a high incidence of dysfunction of endometrium under high physiological level of estrogen and progesterone. ",
"The high level of these hormones could affect the endometrial receptivity ([@B5]). ",
"Ertzeid reported that ovarian stimulation impairs implantation and fetal development in mice ([@B14]). ",
"Previous researches showed a delay in maturation of endometrium epithelium and stroma after ovarian stimulation in human and animals ([@B19]).",
"\n\nMice have commonly been used as animal models in reproductive development research ([@B20]). ",
"Within the last few years, sildenafil citrate (Viagra) has been used successfully for the treatment of penile erectile dysfunction ([@B11]). ",
"Sildenafil citrate promotes smooth muscle relaxation by preventing the degradation of the second messenger cGMP by phosphodiesterase, PDE5 ([@B12]). ",
"The results showed that developed pinopodes are visible in group receiving sildenafil citrate 96 hr after injection HMG. ",
"This may either be a direct effect of the drug on the endometriom, such as inhibition of type 5-specific phosphodiesterase or potential effects of NO on vascular smooth muscle.",
"\n\nUsing a cross-over study design, Sher and Fisch demonstrated the ability of sildenafil to modulate uterine artery blood flow and improve endometrial pattern and thickness. ",
"While improving uterine blood flow in the proliferative phase, NO may have detrimental effects on the level of the endometrium during the implantation window ([@B10]). ",
"But Barroso *et al* showed that the NO mediated release of cytokines such as tumour necrosis factor-α from activated natural killer cells have been implicated as a cause of implantation failure ([@B21]). ",
"It may be beneficial to minimize endometrial NO exposure at the time of embryo transfer and we suggest that discontinuing sildenafil citrate administrated 24 and 72 hr before the day of HCG administration, has improvs effects on pinopodes formation.",
"\n\nIt was Barroso *et al* who, for the first time, demonstrated that higher concentrations of NO inhibit both embryo development *in vitro* and implantation *in vivo* in mice. ",
"Embryos fail to implant if either the uterine receptivity or the development of embryos is impaired, while at higher concentrations, NO is cytotoxic ([@B21]). ",
"Since [Sher and](http://www.ncbi.nlm.nih.gov/pubmed?term=Sher%20G%5BAuthor%5D&cauthor=true&cauthor_uid=10739824) Fish investigation, we have used sildenafil to improve the uterine artery blood flow and endometrial pinopodes and thus, uterine receptivity ([@B10]).",
"\n\nConclusions\n===========\n\nTo conclude, our study on mice endometrial samples shows that after administration of HMG & HCG and sildenafil citrate, the well organized pinopodes were expressed over the surface of mouse endometrium. ",
"It seems that ovarian hyperstimulation by sildenafil citrate injection in mice could cause premature expression of pinopodes on the pre implantation time. ",
"It seems that sildenafil citrate injection may be more helpful and may increase the pregnancy rate; however, further studies are needed.",
"\n\nThis study was financially supported by Drugs Applied Research Centre, Tabriz University of Medical Science, Tabriz, Iran. ",
"Authors have no conflict of interests.",
"\n"
] |
{
"pile_set_name": "PubMed Central"
}
|
[
0.0031095952726900578,
0.0010213070781901479,
0.001093630911782384,
0.002250195713713765,
0.00853568222373724,
0.00075922766700387,
0.0007282610167749226,
0.003252509981393814,
0.0020372371654957533,
0.0007203978020697832,
0.008487923070788383,
0.008092627860605717,
0.0024270364083349705,
0.0012917885323986411,
0.0011295414296910167,
0.002063463209196925,
0.0006251694285310805,
0.01880313828587532,
0.0005394675536081195,
0.00929381512105465,
0.0007774711702950299,
0.0007924530655145645,
0.0005871945177204907,
0.002583377528935671,
0.0009375959634780884,
0.0006787099409848452,
0.0008297809399664402,
0.0009719103691168129,
0.0006911652744747698,
0.000700335600413382,
0.0007021308992989361,
0.000681900477502495,
0.0008801817893981934,
0.0012707519344985485,
0.0005654770066030324,
0.0007539248908869922,
0.0011625959305092692,
0.0008825016557238996,
0.0006724726408720016,
0.0007195909274742007,
0.0009706444106996059,
0.0006866974290460348,
0.003004827070981264,
0.0008601135341450572,
0.0007570937741547823,
0.0006912795943208039,
0.0033656861633062363,
0.0006270556477829814,
0.0006363252177834511,
0.0006800158880650997,
0.0006880800356157124,
0.0019055769080296159,
0.0008168201893568039,
0.0009431688813492656,
0.0007169463206082582,
0.0008168201893568039,
0.0007121154339984059,
0.000738457718398422,
0.003838286502286792,
0.0006608167313970625,
0.0062459842301905155,
0.001385524170473218,
0.0006521319155581295,
0.00175144849345088,
0.0012336352374404669,
0.0009560005273669958,
0.0008230271632783115,
0.001123203313909471,
0.000793415296357125,
0.0018591727130115032,
0.0011845005210489035,
0.0006538594607263803,
0.002083598170429468,
0.0012595850275829434,
0.0007699871202930808,
0.0008648780058138072,
0.0006423600134439766,
0.000820237270090729,
0.0008151546935550869,
0.0006525400094687939,
0.0006916954298503697,
0.0009034597314894199,
0.0007341826567426324,
0.0008719716570340097,
0.003241156693547964,
0.0005657001747749746,
0.0005689616082236171,
0.0006800272385589778,
0.001995444530621171
] | 0.001742 | 89 |
[
"/* Copyright 2015 The TensorFlow Authors. ",
"All Rights Reserved.",
"\n\nLicensed under the Apache License, Version 2.0 (the \"License\");\nyou may not use this file except in compliance with the License.",
"\nYou may obtain a copy of the License at\n\n http://www.apache.org/licenses/LICENSE-2.0\n\nUnless required by applicable law or agreed to in writing, software\ndistributed under the License is distributed on an \"AS IS\" BASIS,\nWITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.",
"\nSee the License for the specific language governing permissions and\nlimitations under the License.",
"\n==============================================================================*/\n\n#ifndef TENSORFLOW_STREAM_EXECUTOR_LIB_PTR_UTIL_H_\n#define TENSORFLOW_STREAM_EXECUTOR_LIB_PTR_UTIL_H_\n\nnamespace perftools {\nnamespace gputools {\nnamespace port {\n\n// Trait to select overloads and return types for MakeUnique.",
"\ntemplate <typename T>\nstruct MakeUniqueResult {\n using scalar = std::unique_ptr<T>;\n};\ntemplate <typename T>\nstruct MakeUniqueResult<T[]> {\n using array = std::unique_ptr<T[]>;\n};\ntemplate <typename T, size_t N>\nstruct MakeUniqueResult<T[N]> {\n using invalid = void;\n};\n\n// MakeUnique<T>(...) is an early implementation of C++14 std::make_unique.",
"\n// It is designed to be 100% compatible with std::make_unique so that the\n// eventual switchover will be a simple renaming operation.",
"\ntemplate <typename T, typename... Args>\ntypename MakeUniqueResult<T>::scalar MakeUnique(Args&&... args) { // NOLINT\n return std::unique_ptr<T>(\n new T(std::forward<Args>(args)...)); // NOLINT(build/c++11)\n}\n\n// Overload for array of unknown bound.",
"\n// The allocation of arrays needs to use the array form of new,\n// and cannot take element constructor arguments.",
"\ntemplate <typename T>\ntypename MakeUniqueResult<T>::array MakeUnique(size_t n) {\n return std::unique_ptr<T>(new typename std::remove_extent<T>::type[n]());\n}\n\n// Reject arrays of known bound.",
"\ntemplate <typename T, typename... Args>\ntypename MakeUniqueResult<T>::invalid MakeUnique(Args&&... /* args */) =\n delete; // NOLINT\n\n} // namespace port\n} // namespace gputools\n} // namespace perftools\n\n\n#endif // TENSORFLOW_STREAM_EXECUTOR_LIB_PTR_UTIL_H_\n"
] |
{
"pile_set_name": "Github"
}
|
[
0.0006782020791433752,
0.0006133938441053033,
0.0005655200802721083,
0.0005397882778197527,
0.0005530848284251988,
0.002414906630292535,
0.0012754208873957396,
0.0006453970563597977,
0.0007360609015449882,
0.0006637975457124412,
0.0009444493916817009,
0.001244266633875668
] | 0.000906 | 12 |
[
"Women's erotic rape fantasies: an evaluation of theory and research.",
"\nThis article is the first systematic review of the research literature on women's rape fantasies. ",
"Current research indicates that between 31% and 57% of women have fantasies in which they are forced into sex against their will, and for 9% to 17% of women these are a frequent or favorite fantasy experience. ",
"Erotic rape fantasies are paradoxical: they do not appear to make sense. ",
"Why would a person have an erotic and pleasurable fantasy about an event that, in real life, would be abhorrent and traumatic? ",
"In this article, the major theories of women's rape fantasies are evaluated both rationally and empirically. ",
"These theories explain rape fantasies in terms of masochism, sexual blame avoidance, openness to sexuality, sexual desirability, male rape culture, biological predisposition to surrender, sympathetic physiological activation, and adversary transformation. ",
"This article evaluates theory and research, makes provisional judgments as to which theories appear to be most viable, and begins the task of theoretical integration to arrive at a more complete and internally consistent explanation for why many women engage in erotic rape fantasies. ",
"Methodological critiques and programs for future research are presented throughout."
] |
{
"pile_set_name": "PubMed Abstracts"
}
|
[
0.2173812836408615,
0.07577446848154068,
0.0231462549418211,
0.2602206766605377,
0.0008867920842021704,
0.03200133517384529,
0.029390163719654083,
0.03229762241244316,
0.0005937452660873532
] | 0.074632 | 9 |
[
"Christopher Wilson\n\nContact Information\n\nWebsite\n\nBiography Summary\n\nChristopher Wilson is an Electrical and Computer Engineering Associate Professor cross-appointed to the Department of Physics and Astronomy at the University of Waterloo.",
"\n\nIn 2004, Professor Wilson moved to Sweden to work on a quantum computing project at Chalmers University of Technology. ",
"Along with his team, he started working on superconducting circuits for quantum computing. ",
"Around 2007, they realized that the work could allow them to measure the virtual photons inside a vacuum. ",
"These virtual photons are generated and annihilated in pairs. ",
"About 40 years ago, it was suggested that a mirror moving near the speed of light could capture some of these photons. ",
"The effect had never been observed, due to the difficulty of moving a massive object at that speed. ",
"They made an electronic 'mirror' that they could effectively move at one-quarter of the speed of light using magnetic fields. ",
"This allowed the team to separate the pairs, stopping them from annihilating and turning them into real photons that they could observe.",
"\n\nProfessor Wilson is continuing his work on quantum information, microwave quantum optics, and nonlinear dynamics at the University of Waterloo. ",
"He has received numerous awards for his research, including the Wallmark Prize for 2012 awarded by the Royal Swedish Academy for his work on the Dynamical Casimir effect (DCE). ",
"Professor Wilson’s work on DCE was named in the Top 5 Breakthrough of 2011 by Physics World and #1 Reader's Choice on Nature News."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0006085025961510837,
0.0006115497671999037,
0.0006495898705907166,
0.0006447678315453231,
0.0009095217101275921,
0.000570553878787905,
0.0006701528327539563,
0.0007175830542109907,
0.000687992840539664,
0.0005664204363711178,
0.0005729210679419339,
0.00062713393708691
] | 0.000653 | 12 |
[
"Q:\n\nPosting to \"me/myapp:myaction\" results in my open graph story only displaying in my 'recent activity' section\n\nI am trying to get my open graph story to also appear on my timeline and be seen on the news feed. ",
"I authorize the user by calling\n NSArray *permissions = [[NSArray alloc] initWithObjects:\n @\"email\", @\"publish_actions\",\n nil];\n [FBSession openActiveSessionWithPublishPermissions:permissions\n defaultAudience:FBSessionDefaultAudienceFriends\n allowLoginUI:YES\n completionHandler:^(FBSession *session, FBSessionState state, NSError *error) {//some handler code}\n\nAnd post my open graph story by calling \n NSMutableDictionary<FBGraphObject> *action = [FBGraphObject graphObject];\n action[@\"event\"] = myEventUrlString;\n\n[FBRequestConnection startForPostWithGraphPath:@\"me/myapp:myaction\" graphObject:action completionHandler:^(FBRequestConnection *connection, id result, NSError *error) {//some handler code}\n\nThe call is successful and I get an id back but what I am consistently seeing is that my open graph story is not appearing on my timeline, not appearing in the news feed and only seems to show in the 'recent activity' section of my page.",
"\n\nA:\n\nYou can \"explicitly share\" your action for it to appear on the user's timeline: https://developers.facebook.com/docs/opengraph/guides/explicit-sharing/\nBe sure to read the \"Inappropriate use\" section of that page to ensure you're meeting the guidelines.",
"\nOnce you enable it for your action in the app settings, you should do something like the following to enable it per action:\nNSMutableDictionary<FBGraphObject> *action = [FBGraphObject graphObject];\naction[@\"event\"] = myEventUrlString;\naction[@\"explicitly_shared\"] = @\"true\";\n\n"
] |
{
"pile_set_name": "StackExchange"
}
|
[
0.0006097348523326218,
0.004299357533454895,
0.0005316216847859323,
0.0007682153955101967
] | 0.001552 | 4 |
[
"First allow me to be very candid. ",
"I have been in the medical billing and coding industry for many years now and I have yet to see or hear of a legitimate company hiring home based medical billers to process their excess claims workload in exchange for either a salary or commission. ",
"If getting a job from home doing medical billing was a reality I (and several of my medical billing business owners associates) would have done that with at least 2-3 different companies to increase our profit potential instead of going through drama of starting a home based medical billing business from scratch.",
"\n\nAs of August 2007 I professionally recommend the following 2 books to anyone that’s looking to start a home based medical billing business. ",
"These books put all the aspects of this industry on table, they are very easy to read and the use of current examples make it very simple for the reader to put the information into action:\n\nLorelys in Miami, Florida said: I am single mom that wants to spend more time with my kids. ",
"I work to pay daycare and well want to do something from work but don't know how or where to start. ",
"Can someone help???",
"\n\nFound this while seeking my own education. ",
"You might look them up. ",
"This is a partial list from the Texas Unemployment Commission approved schools\n\nI have one more question-is it imperative that I am certified in coding to start this business? ",
"I am already a Ceritified Medical Billing and Coding Specialist. (",
"Received from a local Technical School last year.",
"\n\nIt is very unlikely that you will receive the type of response you are looking for. ",
"The topic of marketing a home based medical billingbusiness is not one that can just be expressed in a paragraph or two and this is why the marketing phase of the business is the hardest part for most to deal with.",
"\n\nIf you have never marketed anything besides yourself at an interview (and most do not do that correctly) then selling yourself and your company is very, very scary at first.",
"\n\nThe skill of marketing and selling is learned over time. ",
"As you become more confident in your abilities to make physician see what is wrong with their businesses and what you can do to help them you will become more knowledgeable and that will translate into profits and more business.",
"\n\nSo for the benefit of everyone that reads this page I am going to give you all the expert advice on home based medical billing that I have come across while being in this business:\n\nMedical Billing Beginners Book= This the best book I have ever had the pleasure of reading concerning home based medical billing. ",
"medicalbillingbooks.tripod.com\n\nHow To Jump Start A Successful Medical Billing Service= Another book that I just finished that does a decent job of covering the aspects of doing business as a home based business by Vanessa Best. ",
"www.urlfreeze.com/medbill/L-312b/\n\nMedical Billing Training Courses= This website instructs beginners as to what they must consider before they choose an online medical billing training course to enroll in. ",
"medicalbillingbooks.tripod.com/id2.html\n\nSince my post above was cut short (I know I can ramble on at times) here is the remaining links to the other home based medical billing resources that everyone interested in this field should have in their library:\n\nHow To Market A Medical Billing Business Book= This medical billing book exposes you to the 12 strategies that you should be using every day to gain and increase your business. ",
"You get the \"strategies\", how to implement them, along with phone scripts, letters, and other marketing resources. ",
"These methods work, because they increase our business every time we use them. ",
"www.urlfreeze.com/medbill/How_To_Market_A_Medical_Billing_Business_Book/\n\nThe Basics of Medical Billing Book= This medical billing book is a must read before you get to the marketing phase of your set-up. ",
"Once you absorb the information contained within you will have no fear educating your potential clients and their staff about how and why they should improve the efficiency and cash flow of their office. ",
"www.urlfreeze.com/medbill/Basics_of_Medical_Billing_Book/\n\nHow to Correctly Complete a CMS 1500 Form Book - Line by Line, Box by Box= This medical billing book walks you step by step, line-by-line, box-by-box in easy to understand language through properly preparing the CMS 1500 form so you can prevent getting \"Denied Claims\" which will get you and your clients get paid faster.",
"\nwww.urlfreeze.com/medbill/How_to_Complete_a_CMS_1500_Form/\n\nWould you suggest that if you want to start a at home medical billingbusiness, its best to start at a doctor's office to get more experience? ",
"I am about to start working on my medical claim and billing course. ",
"I have been going back and forth about this decision for about 2 months. ",
"I didn't want to just jump in and do the class unless I know I want to do this. ",
"I would not mind working at home but I have thought it would be best to work somewhere to get more experience and go from there. ",
"How can anyone just start without having any type of experience?",
"\n\nIs there really anything in this world that you can work from home with out having to pay in money to get a job?? ",
"I have 2 kids to care for and i don't want to have to pay for daycare and all the other things that cost money...I would rather stay home and do it myself\n\nIf you can make money doing this---How do you get started? ",
"I have taken the course and I have front office, back office, and other medical experience. ",
"Getting started is the hardest. ",
"Can anyone lend a helping hand???",
"\n\nI am really interested in doing medical claims and billing from home, but I have a few questions. ",
"First of all, I am a certified Medical Assistant, so I've already taken a medical billing course. ",
"Is that enough training to be hired by a company online? ",
"Secondly, I made some stupid choices when I was younger and I am now a convicted felon...how hard will it be for me to find a job? ",
"Does that matter considering that I will be working from home and not in an office?",
"\n\nUnfortunately, I don't know of any doctor that would turn over their billing to you.",
"\n\nAs a part of a biller's compliance plan that is required to detect and prevent fraud and abuse, employers and providers are supposed to check anyone that works for them to see if they have been excluded from submitting claims to the Medicare and Medicaid program and to pre-screen employees as a reasonable step to implement in your compliance plan. ",
"Best practices would exclude you as a convicted felon.",
"\n\nI do think that you might get away without disclosing this fact to some extent, but this is something that would likely come back to haunt you and the providers you were to work with. ",
"You definitely would not be eligible for errors & omissions insurance, which you would want to have.",
"\n\nI don't think it is realistic to believe you could own and operate a professional and successful billing business or be hired to work in a provider or billing office with your background. ",
"I know this is probably not a popular answer, but it's an honest one.",
"\n\nis your heart set on the home-based business or are you willing to work as a medical biller as well? ",
"I say this because--about 11 years ago, I worked for a medical billing company that did have a convicted felon on the staff--in fact, the employee was convicted of insurancefraud working at an insurance company--I am not sure what was involved during the hiring process, but the employee was there for 5 years; so there are those out there that may give you a chance.",
"\n\nSince 1998, the Office of InspectorGeneral has issued 3rd party billing guidelines that basically address this. ",
"Page 12 of 15 of these guidelines say:\n\n>>2. ",
"New Employee Policy For all new employees who have discretionary authority to make decisions that may involve compliance with the law or compliance oversight,\nbilling companies should conduct a reasonable and prudent background investigation, including a reference check, as part of every such employment application. ",
"The application should specifically require the applicant to disclose any criminal conviction, as defined by 42 U.S.C. 1320a-7(i), or exclusion action. ",
"Pursuant\nto the compliance program, billing company policies should prohibit the employment of individuals who have been recently convicted of a criminal offense related to health care or who are listed as debarred, excluded or otherwise ineligible for participation in Federalhealth care programs. ",
"In addition, pending the resolution of any criminal charges or proposed debarment or exclusion, the OIG recommends that\nsuch individuals should be removed from direct responsibility for, or involvement, in any Federal health care program. ",
"Similarly, with regard to current employees or independent contractors, if resolution of the matter results in conviction, debarment or exclusion, then the billing company should remove the individual from direct responsibility for or involvement with all Federal health care programs.",
"\n\nNot knowing what you were convicted of makes it hard to say firmly that you couldn't or wouldn't be hired or that you can't do billing from home but I don't think you can reasonably expect to be welcomed into an office easily. ",
"Again, good luck to you.",
"\n\nI have one more question-is it imperative that I am certified in coding to start this business? ",
"I am already a Ceritified Medical Billing and Coding Specialist. (",
"Received from a local Technical School last year.",
"\n\nI am a certified billing and coding specialist also as of 06/08. ",
"I am currently working in a private PCP office doing a variety of field related tasks including billing and coding. ",
"I would love to work from home in billing/coding only. ",
"Please keep me informed of your business opportunity status. ",
"I will apply!",
"\n\nSomeone earlier said that the US Career Institute was a joke. ",
"Does anyone have an opinion on this or on other correspondence type courses. ",
"Who is the most reputable and who does the best job of preparing you for a medical coding and/or billing career?",
"\n\nnone there are non that really prepare you all the courses are accelerated now which makes it impossible for u to have evough experirnce to survive unless you push your self.. my opionion. ",
"I went to anthem insitute not at all what I was expecting and if i knew then what i know now i would have took colladge courses instead!",
"\n\nAlicia in Jacksonville, Florida said: I am really interested in doing medical claims and billing from home, but I have a few questions. ",
"First of all, I am a certified Medical Assistant, so I've already taken a medical billing course. ",
"Is that enough training to be hired by a company online? ",
"Secondly, I made some stupid choices when I was younger and I am now a convicted felon...how hard will it be for me to find a job? ",
"Does that matter considering that I will be working from home and not in an office?",
"\n\nweiler in Las Vegas, Nevada said: starting a successul at home business like this is not a good idea if you are NEW, you need contacts, experience and clients, try PureMedicalJobsDOTCOM and other similar sites good luck\n\nYes, experience is preferred, but we have people that only have education and no work experience that are able to obtain clients and become successful all the time. ",
"There is nothing saying you can't do it, but obviously, if you have experience, it will help you become successful sooner. ",
"We have sample contracts, billing forms, marketing assistance and much more as an association for our members to download. ",
"We have more than 200 useful forms, articles and samples for people that are just starting or that have experience...\n\nPeople are driving is swarms to start a medical billingbusiness from home but the facts are that their is a lot that goes into this. ",
"First you are responsible for the doctors income. ",
"If you don't collect the money he can lose his practice. ",
"You don't need to go to college to be a medical biller but you should know the basics and be willing to do anything and talk to anyone to get a claim paid on.",
"\n\nI just found this cool website that offers over 12 hours of free medical billing training videos. ",
"You can watch from the setup of a practice management software, to the actual steps of posting charges and payments including ERA into a system. ",
"Also follow along as they send statements and work collections.",
"\n\nHere is the info check out videotrainingpro.com/ and scroll down and click on the Free Medical Billing option. ",
"It takes you to a webpage that lists 113 free training videos - click on the one you want, click play and then maximize the screen for full viewing. ",
"~ pretty cool\n\nThose lists that CMS (Medicare) used to maintain on providers that didn't file claims electronically are not kept any longer because it's mandatory for doctors to file claims electronically to Medicare now.",
"\n\nIf some company told you they would provide you with a list, it was probably a scam because those lists haven't been updated in years.",
"\n\nThe best list will be the one you create. ",
"You can, however, use the information from the NPPES database file for updated addresses of providers.",
"\n\nThank you for responding I have already signed up with a franchise medical billing company I did all my research before I sent in my money and they had wonderful record but I would like to know if you know of any one else that franchise with this company before and how is business going for them and do doctors really use indidviduals that do outsource billing from there house someone told me that they would rather use a commericial billing companies it is going to be hard for me to get a doctor signed up I have not passed out my brochures yet because I still training on the marketing to fit my needs and what is comfortable for me because the way the company is training me is like telemarketing and that is being to pushy and I do not like pushy sale people and also I am making me up some postcards and I would like to use the postcards to market instead of the brochure they gave me I like to make up my own do you think by me starting off I should use they business name since they have been in business for 17 years or should I use my business name or just wait to use mines until I gain doctors and gain more reputation with the business first and the name of the business I franchise with is called USA For Healthclaims\n\npower for a purpose in Franklin, Tennessee said: I have a very lucrative home based business that is very exciting. ",
"It is in the medical field. ",
"If I can help, please contact me for further information.",
"\n\nI have been doing medical billing for 15 yrs. ",
"I started out working in physicians offices, attending Medicare seminars and learning all I can. ",
"I have worked mainly for ophthamologist and cardiologist. ",
"I am currently working from home as an A/R specialist with a well established company of 30 yrs. ",
"I got tired of the unprofessionalism of some of the co-workers and decided to ask if I could work from home. ",
"Thankfully, they agreed and I have been happily working from home for over 2 yrs now.",
"\nIt's pretty easy work but you have to know what you are doing to be able to do this from home. ",
"There is alot to it and you have to do alot of investigational work to resolve the problem. ",
"And there are so many different insurances out there and they all have guidelines of their own so it is not always just set in stone across the board. ",
"I would suggest you try to find a job as a medical biller/A/R specialist with a physician office and learn all you can and show your employer that you are trustworthy to work from home.",
"It takes dedication and discipline to work successfully from home. ",
"I wish you luck....\n\nI am looking to do Medical Billing from. ",
"I have been doing this for seven year now. ",
"I want to send time with my daughter a little bit more. ",
"If anyone can help me please email me at monicaneal8814@yahoo.com\n\nI talked with a person who does medical billing and to make life easier and affordable she is using the InstaClaim.net program and says it is way affordable and can do electronicclaims as well. ",
"I just wanted to put this information out there to help anyone looking for an affordable solution to medical billing. ",
"Good luck in your work I sympathize with your efforts to be moms too.",
"\n\nWhen I had a billing service I didn't have the Erorr & omission insurance however; I knew what I was doing and I had a lot of \"protection\" from my contract with the doctor where I placed most all scenarios back on the doctors shoulders including not being responsible if the doctor coded incorrectly since I wasn't getting his medical records on every patient etc.",
"\n\nPS - if you are employed with the billing service you should be covered under their umbrella of insurances. ",
"If you are a independent contractor then you would be responsible for any claims brought against you which is where having the insurance would be helpful.",
"\n\nI am a bit confused with your frustration. ",
"On your site, you offer the \"Certified Medical Billing Professional (CMBP) Designation \". ",
"So, you award this credential to your students without them taking an exam? ",
"You award the credential for what--them completing the program? ",
"If so, that is fine. ",
"I have no problem with that....if you wish to award a credential without an exam, then, of course-that is your choice. ",
"No-where on my site do you see \"national\". ",
"But if a prospective student or medical biller asks me if the credential is national--I explain to them, because the credential is NOT state specific, they can 'take' their credential with them if they move from state to state.",
"\n\nWhat is your problem if myself and others offer a credential WITH the student successfully passing an exam and you offering a credential WITHOUT the student taking an exam? ",
"I don't understand?",
"\n\nI'm looking into taking a medical billing & claims specialist course. ",
"I'm comparing two at the moment, US Career Institute and Allied. ",
"Allied is more money, however, it includes a big section on Medisoft. ",
"Is Medisoft something I need to be trained on to get a job?",
"\n\nMediSoft is NOT something you need to be concerned about. ",
"It is a program that many billers used to utilize for claims processing, but that was before there were a lot of choices. ",
"In the past several years, there have become a lot more effective programs available and many MediSoft users are no longer using that program as they have switched to something that better meets their needs either financially or practically.",
"\n\nHave you actually review the Better Business Bureau reports for these companies? ",
"That's something you really should do as a matter of research.",
"\n\npower for a purpose in Franklin, Tennessee said: I have a very lucrative home based business that is very exciting. ",
"It is in the medical field. ",
"If I can help, please contact me for further information.",
"\n\nCan you send me your contact info? ",
"I am in the process of trying to work from home.",
"\n\nI just started my medical billingbusiness from home and I am having the hardest time trying to get a doctor to sign up did you work in the field before you got started please give me some tips on how to get a doctor I am trying to find networking groups to join but i do not know which ones to join\n\nThe good news is that you have not diluded yourself into a work from home \"job\" but decided to start your own business! ",
"The key is NOT to give up. ",
"What have you done in terms of marketing so far? ",
"Direct mail? ",
"Internet? ",
"Get your name out there as much as possible! ",
"Hand your cards out to your attorney, your accountant, offer a referal incentive! ",
"Networking with other billers is good to a point..but you need to look at where the connections to the providers are and also think outside the box. ",
"Benefits sell! ",
"Tell & Show potential clients that you can offer MORE than just billing.",
"\n\nI have done direct mailing and have not gotten any responses I do not know any attorneys and I have no healthinsurance so I really do not know any private doctors and most of my friends and most go to the free health clinic like me and medical billing is all the services that I have to offer right now I am looking into taking some accounting classess. ",
"Do your services help business find doctors thank you for responding\n\nShay in Vandalia, Ohio said: US Career Institute is a joke, Went there and NO JOB in the Medical Billing field. ",
"I work\nin A Nursing Home. ",
"Please that is a waste of money and time.",
"\n\nYou say it is a joke its not. ",
"Obviously you don't know what your doing for that career. ",
"I took the same class and it worked for me they lesson's taught you everything. ",
"I am stariting to work from home and I finished that class in January after having a baby, so it isn't a joke.",
"\n\nWhen I had a billing service I didn't have the Erorr & omission insurance however; I knew what I was doing and I had a lot of \"protection\" from my contract with the doctor where I placed most all scenarios back on the doctors shoulders including not being responsible if the doctor coded incorrectly since I wasn't getting his medical records on every patient etc.",
"\n\nYour contract and just knowing what you are doing is NOT even remotely good reason for not having Errors & omissions insurance. ",
"We live in a time where anyone can sue anyone for anything. ",
"What that means is that EVEN if you do EVERYTHING right, in order to defend yourself you will need to be financially ABLE to defend yourself. ",
"I don't know about you but most medical billing companies don't have thousands of dollars sitting around just in case you need to hire a lawyer to defend yourself. ",
"Again, this doesn't matter who's right or wrong, it costs you money to even have to defend yourself, that's what E/O insurance is for."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0005812417948618531,
0.0005750242853537202,
0.000698560499586165,
0.0005670072860084474,
0.0007598201045766473,
0.0009621492936275899,
0.0009421553113497794,
0.001005392405204475,
0.0006476392154581845,
0.0006904884940013289,
0.0005650093080475926,
0.0006161579512991011,
0.0006194299785420299,
0.000591713993344456,
0.0010215440997853875,
0.000600950326770544,
0.0005893393536098301,
0.0006148609099909663,
0.0006489225197583437,
0.0005850325687788427,
0.0006223589298315346,
0.0005882400437258184,
0.000637653807643801,
0.0006669730646535754,
0.0005813661846332252,
0.000833675847388804,
0.0006769842002540827,
0.0006432995432987809,
0.0005441746325232089,
0.0028984409291297197,
0.0005847516003996134,
0.0007094211759977043,
0.0025373452808707952,
0.002407459309324622,
0.0005781256477348506,
0.0008248738013207912,
0.00113677850458771,
0.0005361120565794408,
0.0006414017407223582,
0.0007296990370377898,
0.24443869292736053,
0.0009653386659920216,
0.0008023286354728043,
0.0005843797116540372,
0.07377449423074722,
0.00357833132147789,
0.000719634466804564,
0.0005749491974711418,
0.000667072890792042,
0.0010484907543286681,
0.0006383427535183728,
0.000600348983425647,
0.0007367478683590889,
0.0006042346358299255,
0.0006657128105871379,
0.0006570106488652527,
0.0006842368748039007,
0.0005765208043158054,
0.0007152705802582204,
0.0005884152487851679,
0.0006325638387352228,
0.0005650093080475926,
0.0006161579512991011,
0.0005587810301221907,
0.0005578128038905561,
0.0005729385884478688,
0.0005448376177810133,
0.0014567149337381124,
0.006008030381053686,
0.0005753016448579729,
0.0009853298543021083,
0.01200326718389988,
0.0009571247501298785,
0.0005370073486119509,
0.0006414017407223582,
0.0007296990370377898,
0.24443869292736053,
0.0009653386659920216,
0.0006979499012231827,
0.000600490893702954,
0.0005267317756079137,
0.0006556829321198165,
0.0006543740746565163,
0.015080573037266731,
0.001652171486057341,
0.0005610850057564676,
0.0005691522965207696,
0.0005254866555333138,
0.0006177164614200592,
0.0007094465545378625,
0.0005968326004222035,
0.0015892470255494118,
0.0006244861287996173,
0.0005384008400142193,
0.0006462362362071872,
0.0006881937733851373,
0.0005405592382885516,
0.0007225425797514617,
0.0005404948606155813,
0.000794567633420229,
0.0007444800576195121,
0.0006854008533991873,
0.0006362357526086271,
0.00062272121431306,
0.0018751543248072267,
0.0007019676850177348,
0.000883866217918694,
0.0005440457607619464,
0.0006168290274217725,
0.0006087738438509405,
0.0008848533616401255,
0.0005611887318082154,
0.0005296278977766633,
0.14777779579162598,
0.0006225962424650788,
0.0005805101245641708,
0.0007339350413531065,
0.015156131237745285,
0.000593523436691612,
0.0007324251928366721,
0.0006815688102506101,
0.0007184767164289951,
0.0006427874905057251,
0.0007554840412922204,
0.0005872281035408378,
0.0033493663650006056,
0.001119876280426979,
0.0006208644481375813,
0.0005498797981999815,
0.000650664558634162,
0.0014629372162744403,
0.0011996865505352616,
0.0006135118892416358,
0.0005891380715183914,
0.0006311670877039433,
0.0005977439577691257,
0.0005565499304793775,
0.0006881937733851373,
0.0005405592382885516,
0.0009985649958252907,
0.0007604393758811057,
0.009575322270393372,
0.0008885606657713652,
0.0006654920289292932,
0.0007891276036389172,
0.0011160933645442128,
0.09061089158058167,
0.00774376280605793,
0.0005628761718980968,
0.010231416672468185,
0.000574957113713026,
0.0007271250942721963,
0.0010767828207463026,
0.0021084966138005257,
0.03729664906859398,
0.0013587174471467733,
0.0010099433129653335,
0.0020544042345136404,
0.0015762796392664313,
0.0006225962424650788,
0.0038844950031489134,
0.030075088143348694,
0.012500172480940819,
0.0011569021735340357,
0.026534147560596466
] | 0.006724 | 165 |
[
"---\nabstract: 'Based on $5.8 \\times 10^7 {J/\\psi}$ events detected in BESII, the branching fractions of ${J/\\psi}\\to\\phi\\eta$ and $\\phi{\\eta^{\\prime}}$ are measured for different $\\eta$ and ${\\eta^{\\prime}}$ decay modes. ",
"The results are significantly higher than previous measurements. ",
"An upper limit on $B({J/\\psi}{\\rightarrow}\\phi{\\pi^0})$ is also obtained.'",
"\nauthor:\n- |\n M. Ablikim$^{1}$, J. Z. Bai$^{1}$, Y. Ban$^{11}$, J. G. Bian$^{1}$, X. Cai$^{1}$, J. F. Chang$^{1}$, H. F. Chen$^{17}$, H. S. Chen$^{1}$, H. X. Chen$^{1}$, J. C. Chen$^{1}$, Jin Chen$^{1}$, Jun Chen$^{7}$, M. L. Chen$^{1}$, Y. B. Chen$^{1}$, S. P. Chi$^{2}$, Y. P. Chu$^{1}$, X. Z. Cui$^{1}$, H. L. Dai$^{1}$, Y. S. Dai$^{19}$, Z. Y. Deng$^{1}$, L. Y. Dong$^{1}$$^a$, Q. F. Dong$^{15}$, S. X. Du$^{1}$, Z. Z. Du$^{1}$, J. Fang$^{1}$, S. S. Fang$^{2}$, C. D. Fu$^{1}$, H. Y. Fu$^{1}$, C. S. Gao$^{1}$, Y. N. Gao$^{15}$, M. Y. Gong$^{1}$, W. X. Gong$^{1}$, S. D. Gu$^{1}$, Y. N. Guo$^{1}$, Y. Q. Guo$^{1}$, Z. J. Guo$^{16}$, F. A. Harris$^{16}$, K. L. He$^{1}$, M. He$^{12}$, X. He$^{1}$, Y. K. Heng$^{1}$, H. M. Hu$^{1}$, T. Hu$^{1}$, G. S. Huang$^{1}$$^b$, X. P. Huang$^{1}$, X. T. Huang$^{12}$, X. B. Ji$^{1}$, C. H. Jiang$^{1}$, X. S. Jiang$^{1}$, D. P. Jin$^{1}$, S. Jin$^{1}$, Y. Jin$^{1}$, Yi Jin$^{1}$, Y. F. Lai$^{1}$, F. Li$^{1}$, G. Li$^{2}$, H. H. Li$^{1}$, J. Li$^{1}$, J. C. Li$^{1}$, Q. J. Li$^{1}$, R. Y. Li$^{1}$, S. M. Li$^{1}$, W. D. Li$^{1}$, W. G. Li$^{1}$, X. L. Li$^{8}$, X. Q. Li$^{10}$, Y. L. Li$^{4}$, Y. F. Liang$^{14}$, H. B. Liao$^{6}$, C. X. Liu$^{1}$, F. Liu$^{6}$, Fang Liu$^{17}$, H. H. Liu$^{1}$, H. M. Liu$^{1}$, J. Liu$^{11}$, J. B. Liu$^{1}$, J. P. Liu$^{18}$, R. G. Liu$^{1}$, Z. A. Liu$^{1}$, Z. X. Liu$^{1}$, F. Lu$^{1}$, G. R. Lu$^{5}$, H. J. Lu$^{17}$, J. G. Lu$^{1}$, C. L. Luo$^{9}$, L. X. Luo$^{4}$, X. L. Luo$^{1}$, F. C. Ma$^{8}$, H. L. Ma$^{1}$, J. M. Ma$^{1}$, L. L. Ma$^{1}$, Q. M. Ma$^{1}$, X. B. Ma$^{5}$, X. Y. Ma$^{1}$, Z. P. Mao$^{1}$, X. H. Mo$^{1}$, J. Nie$^{1}$, Z. D. Nie$^{1}$, S. L. Olsen$^{16}$, H. P. Peng$^{17}$, N. D. Qi$^{1}$, C. D. Qian$^{13}$, H. Qin$^{9}$, J. F. Qiu$^{1}$, Z. Y. Ren$^{1}$, G. Rong$^{1}$, L. Y. Shan$^{1}$, L. Shang$^{1}$, D. L. Shen$^{1}$, X. Y. Shen$^{1}$, H. Y. Sheng$^{1}$, F. Shi$^{1}$, X. Shi$^{11}$$^c$, H. S. Sun$^{1}$, J. F. Sun$^{1}$, S. S. Sun$^{1}$, Y. Z. Sun$^{1}$, Z. J. Sun$^{1}$, X. Tang$^{1}$, N. Tao$^{17}$, Y. R. Tian$^{15}$, G. L. Tong$^{1}$, G. S. Varner$^{16}$, D. Y. Wang$^{1}$, J. Z. Wang$^{1}$, K. Wang$^{17}$, L. Wang$^{1}$, L. S. Wang$^{1}$, M. Wang$^{1}$, P. Wang$^{1}$, P. L. Wang$^{1}$, S. Z. Wang$^{1}$, W. F. Wang$^{1}$$^d$ Y. F. Wang$^{1}$, Z. Wang$^{1}$, Z. Y. Wang$^{1}$, Zhe Wang$^{1}$, Zheng Wang$^{2}$, C. L. Wei$^{1}$, D. H. Wei$^{1}$, N. Wu$^{1}$, Y. M. Wu$^{1}$, X. M. Xia$^{1}$, X. X. Xie$^{1}$, B. Xin$^{8}$$^b$, G. F. Xu$^{1}$, H. Xu$^{1}$, S. T. Xue$^{1}$, M. L. Yan$^{17}$, F. Yang$^{10}$, H. X. Yang$^{1}$, J. Yang$^{17}$, Y. X. Yang$^{3}$, M. Ye$^{1}$, M. H. Ye$^{2}$, Y. X. Ye$^{17}$, L. H. Yi$^{7}$, Z. Y. Yi$^{1}$, C. S. Yu$^{1}$, G. W. Yu$^{1}$, C. Z. Yuan$^{1}$, J. M. Yuan$^{1}$, Y. Yuan$^{1}$, S. L. Zang$^{1}$, Y. Zeng$^{7}$, Yu Zeng$^{1}$, B. X. Zhang$^{1}$, B. Y. Zhang$^{1}$, C. C. Zhang$^{1}$, D. H. Zhang$^{1}$, H. Y. Zhang$^{1}$, J. Zhang$^{1}$, J. W. Zhang$^{1}$, J. Y. Zhang$^{1}$, Q. J. Zhang$^{1}$, S. Q. Zhang$^{1}$, X. M. Zhang$^{1}$, X. Y. Zhang$^{12}$, Y. Y. Zhang$^{1}$, Yiyun Zhang$^{14}$, Z. P. Zhang$^{17}$, Z. Q. Zhang$^{5}$, D. X. Zhao$^{1}$, J. B. Zhao$^{1}$, J. W. Zhao$^{1}$, M. G. Zhao$^{10}$, P. P. Zhao$^{1}$, W. R. Zhao$^{1}$, X. J. Zhao$^{1}$, Y. B. Zhao$^{1}$, Z. G. Zhao$^{1}$$^e$, H. Q. Zheng$^{11}$, J. P. Zheng$^{1}$, L. S. Zheng$^{1}$, Z. P. Zheng$^{1}$, X. C. Zhong$^{1}$, B. Q. Zhou$^{1}$, G. M. Zhou$^{1}$, L. Zhou$^{1}$, N. F. Zhou$^{1}$, K. J. Zhu$^{1}$, Q. M. Zhu$^{1}$, Y. C. Zhu$^{1}$, Y. S. Zhu$^{1}$, Yingchun Zhu$^{1}$$^f$, Z. A. Zhu$^{1}$, B. A. Zhuang$^{1}$, X. A. Zhuang$^{1}$, B. S. Zou$^{1}$\\\n (BES Collaboration)\\\n $^{1}$[*Institute of High Energy Physics, Beijing 100049, People’s Republic of China* ]{}\\\n $^{2}$[*China Center for Advanced Science and Technology, Beijing 100080, People’s Republic of China*]{}\\\n $^{3}$[*Guangxi Normal University, Guilin 541004, People’s Republic of China* ]{}\\\n $^{4}$ [*Guangxi University, Nanning 530004, People’s Republic of China*]{}\\\n $^{5}$ [*Henan Normal University, Xinxiang 453002, People’s Republic of China*]{}\\\n $^{6}$[*Huazhong Normal University, Wuhan 430079, People’s Republic of China*]{}\\\n $^{7}$ [*Hunan University, Changsha 410082, People’s Republic of China*]{}\\\n $^{8}$[*Liaoning University, Shenyang 110036, People’s Republic of China*]{}\\\n $^{9}$[*Nanjing Normal University, Nanjing 210097, People’s Republic of China*]{}\\\n $^{10}$ [*Nankai University, Tianjin 300071, People’s Republic of China*]{}\\\n $^{11}$ [*Peking University, Beijing 100871, People’s Republic of China*]{}\\\n $^{12}$ [*Shandong University, Jinan 250100, People’s Republic of China*]{}\\\n $^{13}$ [*Shanghai Jiaotong University, Shanghai 200030, People’s Republic of China*]{}\\\n $^{14}$ [*Sichuan University, Chengdu 610064, People’s Republic of China*]{}\\\n $^{15}$ [*Tsinghua University, Beijing 100084, People’s Republic of China*]{}\\\n $^{16}$ [*University of Hawaii, Honolulu, Hawaii 96822, USA*]{}\\\n $^{17}$ [*University of Science and Technology of China, Hefei 230026, People’s Republic of China*]{}\\\n $^{18}$ [*Wuhan University, Wuhan 430072, People’s Republic of China*]{}\\\n $^{19}$ [*Zhejiang University, Hangzhou 310028, People’s Republic of China*]{}\\\n $^{a}$ Current address: Iowa State University, Ames, Iowa 50011-3160, USA.\\\n $^{b}$ Current address: Purdue University, West Lafayette, Indiana 47907, USA.\\\n $^{c}$ Current address: Cornell University, Ithaca, New York 14853, USA.\\\n $^{d}$ Current address: Laboratoire de l’Acc[é]{}l[é]{}ratear Lin[é]{}aire, F-91898 Orsay, France.\\\n $^{e}$ Current address: University of Michigan, Ann Arbor, Michigan 48109, USA.\\\n $^{f}$ Current address: DESY, D-22607, Hamburg, Germany.\\\ntitle: ' **Measurements of ${J/\\psi}$ decays into $\\phi{\\pi^0}$, $\\phi\\eta$, and $\\phi{\\eta^{\\prime}}$** '\n---\n\n0.2cm\n\nIntroduction {#introd}\n============\n\nThe decay of the $J/\\psi$ into a vector and pseudoscalar meson pair, $J/\\psi \\to V P$ with $V$ and $P$ representing vector and pseudoscalar mesons, can proceed via strong and electromagnetic reactions. ",
"A well measured set of all possible decays of $J/\\psi \\to V P$ allows one to systematically study the quark gluon contents of pseudoscalar mesons, SU(3) breaking, as well as determine the electromagnetic and doubly suppressed OZI amplitudes in two-body $J/\\psi$ decays [@theory]. ",
"MARKIII [@mark2; @mark3] and DM2 [@dm2] measured many $J/\\psi \\to V P$ decays and obtained the $\\eta-\\eta'$ mixing angle, the quark content of the $\\eta$ and $\\eta'$, and much more. ",
"Recently, a sample of $5.8 \\times 10^7 {J/\\psi}$ events was accumulated with the upgraded Beijing Spectrometer (BESII) [@besii], which offers a unique opportunity to measure precisely the full set of $J/\\psi \\to V P$ decays. ",
"In an earlier analysis based on this data set, the branching fraction of ${J/\\psi}{\\rightarrow}{\\pi^+}{\\pi^-}{\\pi^0}$ was measured to be $(2.10\\pm0.12)\\%$ [@besrhopi], which is higher than the PDG [@pdg2004] value by about 30%. ",
"This indicates a higher branching fraction for ${J/\\psi}{\\rightarrow}\\rho\\pi$ than those from older experiments [@frhopi], since the dominant dynamics in ${J/\\psi}\\to {\\pi^+}{\\pi^-}{\\pi^0}$ is ${J/\\psi}{\\rightarrow}\\rho\\pi$. Therefore, remeasuring the branching fractions of all ${J/\\psi}{\\rightarrow}VP$ decay modes becomes very important. ",
"In this paper, $J/\\psi \\to \\phi \\pi^0$, $\\phi \\eta$, and $\\phi {\\eta^{\\prime}}$ are studied, based on the BESII $5.8 \\times 10^7 {J/\\psi}$ events.",
"\n\nThe BES Detector {#BESD}\n================\n\nThe upgraded Beijing Spectrometer detector (BESII) is located at the Beijing Electron-Positron Collider (BEPC). ",
"BESII is a large solid-angle magnetic spectrometer which is described in detail in Ref.",
" [@besii]. ",
"The momentum of charged particles is determined by a 40-layer cylindrical main drift chamber (MDC) which has a momentum resolution of $\\sigma_{p}$/p=$1.78\\%\\sqrt{1+p^2}$ ($p$ in GeV/c). ",
"Particle identification is accomplished using specific ionization ($dE/dx$) measurements in the drift chamber and time-of-flight (TOF) information in a barrel-like array of 48 scintillation counters. ",
"The $dE/dx$ resolution is $\\sigma_{dE/dx}\\simeq8.0\\%$; the TOF resolution for Bhabha events is $\\sigma_{TOF}= 180$ ps. ",
"Radially outside of the time-of-flight counters is a 12-radiation-length barrel shower counter (BSC) comprised of gas proportional tubes interleaved with lead sheets. ",
"The BSC measures the energy and direction of photons with resolutions of $\\sigma_{E}/E\\simeq21\\%\\sqrt{E}$ ($E$ in GeV), $\\sigma_{\\phi}=7.9$ mrad, and $\\sigma_{z}=2.3$ cm. ",
"The iron flux return of the magnet is instrumented with three double layers of proportional counters (MUC) that are used to identify muons.",
"\n\nA GEANT3 based Monte Carlo package (SIMBES) with detailed consideration of the detector performance is used. ",
"The consistency between data and Monte Carlo has been carefully checked in many high purity physics channels, and the agreement is reasonable. ",
"The detection efficiency and mass resolution for each decay mode are obtained from a Monte Carlo simulation which takes into account the angular distributions appropriate for the different final states [@rhopi].",
"\n\nanalysis\n========\n\nIn this analysis, the $\\phi$ meson is observed in its ${K^+}{K^-}$ decay mode, and the pseudoscalar mesons are detected in the modes: ${\\pi^0}{\\rightarrow}{\\gamma}{\\gamma}$; $\\eta{\\rightarrow}{\\gamma}{\\gamma}$, ${\\gamma}{\\pi^+}{\\pi^-}$, and ${\\pi^+}{\\pi^-}{\\pi^0}$; and ${\\eta^{\\prime}}{\\rightarrow}{\\gamma}{\\gamma}$, ${\\gamma}{\\pi^+}{\\pi^-}({\\gamma}\\rho)$, and ${\\pi^+}{\\pi^-}\\eta~(\\eta{\\rightarrow}{\\gamma}{\\gamma})$. Using multiple $\\eta$ and ${\\eta^{\\prime}}$ decay modes allows us to crosscheck our measurements, as well as obtain higher precision. ",
"Possible final states of ${J/\\psi}{\\rightarrow}\\phi{\\pi^0},~\\phi\\eta$, and $\\phi{\\eta^{\\prime}}$ are then ${K^+}{K^-}{\\gamma}{\\gamma}$, ${K^+}{K^-}{\\pi^+}{\\pi^-}{\\gamma}$, and ${K^+}{K^-}{\\pi^+}{\\pi^-}{\\gamma}{\\gamma}$. Candidate events are required to satisfy the following common selection criteria:\n\n1. ",
" The events must have the correct number of charged tracks with net charge zero. ",
"Each track must be well fitted to a helix, originating from the interaction region of R$_{xy}<$0.02 m and $|z| <$ 0.2 m, and have a polar angle, $\\theta$, in the range $|\\cos \\theta| <$ 0.8.",
"\n\n2. ",
" Events should have at least the minimum number of isolated photons associated with the different final states. ",
"Isolated photons are those that have energy deposited in the BSC greater than 60 MeV, the angle between the direction at the first hit layer of the BSC and the developing direction of the cluster less than 30$^\\circ$, and the angle between photons and any charged tracks larger than 10$^\\circ$.\n\n3. ",
" For each charged track in an event, $\\chi^{2}_{PID}(i)$ is determined using both $dE/dx$ and TOF information:\n\n $\\chi^{2}_{PID}(i)$=$\\chi^{2}_{dE/dx}(i)$+$\\chi^{2}_{TOF}(i)$\n\n A charged track is identified as a $\\pi$ or $K$ if its $\\chi^{2}_{PID}$ is less than those for any other assignment. ",
"To reject background events, two charged tracks are required to be identified as kaons in ${J/\\psi}{\\rightarrow}\\phi{\\pi^0}$. For the other channels, at least one charged track must be identified as a kaon in the event selection.",
"\n\n4. ",
" The selected events are fitted kinematically. ",
"The kinematic fit adjusts the track energy and momentum within the measured errors so as to satisfy energy and momentum conservation for the given event hypothesis. ",
"This improves resolution, selects the correct charged-particle assignment for the tracks, and reduces background. ",
"When the number of photons in an event exceeds the minimum, all combinations are tried, and the combination with the smallest $\\chi^{2}$ is retained.",
"\n\nThe branching fraction is calculated using\n\n$$\\begin{aligned}\n\\lefteqn{B({J/\\psi}{\\rightarrow}\\phi P) = } \\\\ \n& & \\frac{N_{obs}}{N_{{J/\\psi}}\\cdot\n{\\varepsilon}\\cdot B(\\phi{\\rightarrow}{K^+}{K^-})\\cdot B(P{\\rightarrow}X)},\n\\label{forbr}\\end{aligned}$$\n\nwhere $N_{obs}$ is the number of events observed (or the upper limit), $N_{{J/\\psi}}$ is the number of ${J/\\psi}$ events, $(5.77\\pm\n0.27)\\times 10^7$, determined from the number of inclusive 4-prong hadronic decays [@fangss], ${\\varepsilon}$ is the detection efficiency obtained from Monte Carlo simulation, and $B(\\phi{\\rightarrow}{K^+}{K^-})$ and $B(P{\\rightarrow}X)$ are the branching fractions of $\\phi{\\rightarrow}{K^+}{K^-}$ and pseudoscalar decays from the PDG [@pdg2004], respectively.",
"\n\n${J/\\psi}{\\rightarrow}\\phi{\\gamma}{\\gamma}$\n-------------------------------------------\n\nEvents with two oppositely charged tracks and at least two or three isolated photons are selected. ",
"A 4C-fit is performed to the $K^+ K^- \\gamma \\gamma$ hypothesis, and $\\chi^2 < 15$ is required. ",
"To reject possible background events from ${J/\\psi}{\\rightarrow}{\\gamma}{K^+}{K^-}{\\pi^0}$, the 4C-fit probability for the assignment ${J/\\psi}{\\rightarrow}{K^+}{K^-}{\\gamma}{\\gamma}$ must be larger than that of ${K^+}{K^-}{\\gamma}{\\gamma}{\\gamma}$.\n\nAfter this selection, the scatter plot (Figure \\[dphieta2gam\\]) of $m_{{K^+}{K^-}}$ versus $m_{{\\gamma}{\\gamma}}$ shows two clusters corresponding to $\\phi {\\eta^{\\prime}}$ and $\\phi\\eta$, but there is no clear accumulation of events for $\\phi{\\pi^0}$. To obtain the $m_{{\\gamma}{\\gamma}}$ distribution recoiling against $\\phi$, the ${K^+}{K^-}$ invariant mass is required to be in the $\\phi$ mass region, $|m_{{K^+}{K^-}}-1.02|<0.02$ GeV/c$^2$.\n\n### ${J/\\psi}{\\rightarrow}\\phi{\\pi^0}$\n\nFigure \\[piofit\\](a) shows the $m_{{\\gamma}{\\gamma}}$ invariant mass distribution after the above selection; no clear ${\\pi^0}$ signal is observed. ",
"The Bayesian method is used to determine the upper limit on the ${J/\\psi}{\\rightarrow}\\phi{\\pi^0}$ branching fraction. ",
"A Breit-Wigner convoluted with a Gaussian plus a polynomial background function are used to fit the $m_{{\\gamma}{\\gamma}}$ spectrum. ",
"The ${\\pi^0}$ mass and width are fixed to PDG values. ",
"The mass resolution, obtained from Monte Carlo simulation, is 17.7 MeV/c$^2$. At the 90% confidence level, the number of $\\phi{\\pi^0}$ events is 24. ",
"Taking into account the detection efficiency, $(16.63\\pm 0.20)\\%$, the upper limit on the branching fraction is\n\n$B({J/\\psi}{\\rightarrow}\\phi{\\pi^0})<5.10\\times 10^{-6}$\n\n### ${J/\\psi}{\\rightarrow}\\phi\\eta$\n\nFigure \\[piofit\\](b) shows the $m_{{\\gamma}{\\gamma}}$ distribution; an $\\eta$ signal is clearly seen. ",
"The fit of this distribution with a Breit-Wigner convoluted with a Gaussian plus a second order polynomial background function gives $2086\\pm 42 $ $\\phi\\eta$ events with a $\\eta$ mass of $549.0\\pm 0.5$ MeV/c$^2$. The background events, $152\\pm 17$, are estimated from the $\\phi$ sidebands, defined by 0.98 GeV/c$^2 <m_{{K^+}{K^-}}<1.00$ GeV/c$^2$ and 1.04 GeV/c$^2<m_{{K^+}{K^-}}<1.06$ GeV/c$^2$. After subtracting background and correcting for detection efficiency, $(19.98 \\pm0.22)$%, the ${J/\\psi}{\\rightarrow}\\phi\\eta$ branching fraction is obtained\n\n$B({J/\\psi}{\\rightarrow}\\phi\\eta)=(8.67\\pm 0.19)\\times 10^{-4}$,\n\nwhere the error is statistical only.",
"\n\n### ${J/\\psi}{\\rightarrow}\\phi{\\eta^{\\prime}}$\n\nThe distribution of $m_{{\\gamma}{\\gamma}}$ in ${\\eta^{\\prime}}$ mass region recoiling against the $\\phi$ is shown in Figure \\[piofit\\](c). ",
"A fit of the ${\\eta^{\\prime}}$ peak with a Breit-Wigner and a second order backgound polynomial yields $68\\pm 15$ $\\phi{\\eta^{\\prime}}$ events with the peak at $958.1\\pm2.6$ MeV/c$^2$. No obvious signal is observed for the distribution of $m_{{\\gamma}{\\gamma}}$ recoiling against $\\phi$ sidebands (0.98 GeV/c$^2<m_{{K^+}{K^-}}<1.0$ GeV/c$^2$ and 1.04 GeV/c$^2<m_{{K^+}{K^-}}<1.06$ GeV/$c^2$). ",
"The detection efficiency is $(18.57\\pm 0.22)\\%$, and the corresponding branching fraction is determined to be\n\n$B({J/\\psi}{\\rightarrow}\\phi{\\eta^{\\prime}})=(6.10\\pm 1.34)\\times 10^{-4}$,\n\nwhere the error is only the statistical error.",
"\n\n${J/\\psi}{\\rightarrow}\\phi{\\gamma}{\\pi^+}{\\pi^-}$ {#phietag2pi}\n-------------------------------------------------\n\nFor $J/\\psi \\to \\phi \\eta$, $\\eta \\to \\gamma \\pi^+ \\pi^-$, events with four well-reconstructed charged tracks and at least one isolated photon are required. ",
"To select the pions and kaons from amongst the tracks, 4C fits are applied for one of the following three cases: (1) if only one charged track is identified as a kaon using particle identification, then the other charged tracks are assumed, one at a time, to be a kaon, while the other two are assumed to be pions; (2) if two charged tracks are identified as kaons, then the other two tracks are assumed to be pions; (3) if three or four charged tracks are identified as kaons, then the particle identification information is ignored and all combinations of two kaon and two pion tracks are kinematicaly fitted. ",
"For each case, the hypothesis with the smallest $\\chi^{2}$ is selected. ",
"We further require that the probability of the 4C fit for the ${J/\\psi}{\\rightarrow}{K^+}{K^-}{\\pi^+}{\\pi^-}{\\gamma}$ assignment is larger than those of ${K^+}{K^-}{\\pi^+}{\\pi^-}$ and ${K^+}{K^-}{\\pi^+}{\\pi^-}{\\gamma}{\\gamma}$.\n\nThe scatter plot of $m_{{K^+}{K^-}}$ versus $m_{{\\gamma}{\\pi^+}{\\pi^-}}$ is shown in Figure \\[dphietag2pi\\], where ${J/\\psi}{\\rightarrow}\\phi\\eta$ and ${J/\\psi}{\\rightarrow}\\phi{\\eta^{\\prime}}$ decays are apparent. ",
"For the scatter plot of $m_{{\\pi^+}{\\pi^-}}$ versus $m_{{\\gamma}{\\pi^+}{\\pi^-}}$, shown in Figure \\[dphietapgrho\\], the ${\\eta^{\\prime}}$ - $\\rho$ signal corresponds to the decay ${\\eta^{\\prime}}{\\rightarrow}{\\gamma}\\rho$. The other cluster is from $\\eta {\\rightarrow}{\\gamma}{\\pi^+}{\\pi^-}$ and $\\eta{\\rightarrow}{\\pi^0}{\\pi^+}{\\pi^-}$ background events.",
"\n\n### ${J/\\psi}{\\rightarrow}\\phi\\eta$\n\nFigure \\[mgpipi\\] shows the ${\\gamma}{\\pi^+}{\\pi^-}$ invariant mass recoiling against the $\\phi$, defined by $|m_{K^+K^-}-1.02|<0.02$ GeV/c$^2$. A clear $\\eta$ signal is observed. ",
"The peak on the left side of the $\\eta$ in Figure \\[mgpipi\\] comes from ${J/\\psi}{\\rightarrow}\\phi\\eta~(\\eta{\\rightarrow}{\\pi^+}{\\pi^-}{\\pi^0})$ with one photon missing; this is confirmed by Monte-Carlo simulation. ",
"This peak cannot be described by a simple Breit-Wigner due to its asymmetric shape. ",
"To obtain the shape of the peak, a Monte-carlo sample of ${J/\\psi}{\\rightarrow}\\phi\\eta~(\\eta{\\rightarrow}{\\pi^+}{\\pi^-}{\\pi^0})$ is generated and a fit is made to the peak. ",
"The ${\\gamma}{\\pi^+}{\\pi^-}$ mass distribution is then fitted with this shape, a Breit-Wigner to describe the $\\eta$ signal, and a polynomial background. ",
"The fit, shown in Figure \\[mgpipi\\], yields $134 \\pm 14$ $\\eta$ events with a mass at $548.9\\pm0.9$ MeV/c$^2$. The detection efficiency obtained from Monte Carlo simulation is $(10.32\\pm 0.16)\\%$, and the corresponding branching fraction is\n\n$B({J/\\psi}{\\rightarrow}\\phi\\eta)=(9.79\\pm 1.02)\\times 10^{-4},$\n\nwhere the error is only the statistical error.",
"\n\n### ${J/\\psi}{\\rightarrow}\\phi{\\eta^{\\prime}}$\n\nAfter requiring $|m_{{K^+}{K^-}}-1.02|<0.02$ GeV/c$^2$ and 0.3 GeV/c$^2<m_{{\\pi^+}{\\pi^-}}<0.95$ GeV/c$^2$, the distribution of ${\\gamma}{\\pi^+}{\\pi^-}$ invariant mass recoiling against the $\\phi$ is shown in Figure \\[fitphietapg2pi\\]; a fit with a Breit-Wigner convoluted with a Gaussian and a second order polynomial gives $462\\pm 29\n$ events with a peak at $957.4\\pm0.7$ MeV/c$^2$. The detection efficiency obtained from Monte Carlo simulation is $(9.80\\pm0.16)$%, and the branching fraction obtained is\n\n$B({J/\\psi}{\\rightarrow}\\phi{\\eta^{\\prime}})=(5.64\\pm0.35)\\times 10^{-4}$.\n\n${J/\\psi}{\\rightarrow}\\phi{\\pi^+}{\\pi^-}{\\gamma}{\\gamma}$\n---------------------------------------------------------\n\nFor the $\\eta{\\rightarrow}{\\pi^+}{\\pi^-}{\\pi^0}$ case, events with four well reconstructed charged tracks and at least two isolated photons are selected. ",
"A 4C kinematic fit to the ${K^+}{K^-}{\\pi^+}{\\pi^-}{\\gamma}{\\gamma}$ hypothsis is applied, as described in [ Section \\[phietag2pi\\] ]{} for ${J/\\psi}{\\rightarrow}\\phi{\\gamma}{\\pi^+}{\\pi^-}$, and the case with the smallest $\\chi^2$ is selected.",
"\n\nAfter the above selection and with the requirement that $m_{{\\gamma}{\\gamma}}$ be consistent with a ${\\pi^0}$, (0.095 GeV/c$^2$ $< m_{\\gamma \\gamma} <\n0.175$ GeV/c$^2$), the ${J/\\psi}{\\rightarrow}\\phi\\eta$ decay is clearly observed in the scatter plot of $m_{{K^+}{K^-}}$ versus $m_{{\\pi^+}{\\pi^-}{\\gamma}{\\gamma}}$, shown in Figure \\[dphieta3pi1\\](a). ",
"Requiring 0.5 GeV/c$^2$ $< m_{{\\gamma}{\\gamma}} < $ 0.6 GeV/c$^2$, the scatter plot in Figure \\[dphieta3pi1\\](b) shows clean $\\phi{\\eta^{\\prime}}$ signals. ",
"The decays of $\\eta{\\rightarrow}{\\pi^+}{\\pi^-}{\\pi^0}$ and ${\\eta^{\\prime}}{\\rightarrow}{\\pi^+}{\\pi^-}\\eta$ are also observed in the scatter plot of $m_{{\\gamma}{\\gamma}}$ versus $m_{{\\pi^+}{\\pi^-}\\eta}$, shown in Figure \\[dphietap2pieta\\].",
"\n\n### ${J/\\psi}{\\rightarrow}\\phi\\eta$\n\nThe $m_{K^+K^-}$ invariant mass spectrum recoiling against the $\\eta$, shown in Figure \\[m3pi\\], is used to get the $\\phi\\eta$ signals. ",
"A Breit-Wigner convoluted with a Gaussian to account for the $\\phi$ mass resolution plus a second order polynomial are used to fit the $m_{K^+K^-}$ mass distribution. ",
"A total of $350\\pm$11 events with a $\\phi$ mass at $1020.4\\pm0.3$ MeV/c$^2$ from $\\phi$ decay are obtained in the fit, which using the detection efficiency of $(5.81\\pm0.12)$% corresponds to a branching fraction of\n\n$B({J/\\psi}{\\rightarrow}\\phi\\eta)=(9.41\\pm0.30)\\times 10^{-4}$.\n\nHere, the error is only the statistical error.",
"\n\n### ${J/\\psi}{\\rightarrow}\\phi{\\eta^{\\prime}}$\n\nAfter requiring $0.5 <m_{\\gamma\\gamma} <0.6$ GeV/c$^2$ and $m_{{\\pi^+}{\\pi^-}}<0.45$ GeV/c$^2$, the ${\\pi^+}{\\pi^-}{\\gamma}{\\gamma}$ mass recoiling against the $\\phi$($|m_{{K^+}{K^-}}-1.02|<0.02$ GeV/c$^2$), shows a clean ${\\eta^{\\prime}}$ peak, as seen in Figure \\[fitphietap2pieta\\]. ",
"No clear signal is observed for $\\phi$ sidebands (0.98 GeV/c$^2<m_{{K^+}{K^-}}<1.0$ GeV/c$^2$ and 1.04 GeV/c$^2<m_{{K^+}{K^-}}<1.06$ GeV/c$^2$). ",
"The fit of $m_{{\\pi^+}{\\pi^-}{\\gamma}{\\gamma}}$ yields $198 \\pm 12$ events with a peak at $959.2\\pm1.4$ MeV/c$^2$, and the detection efficiency for this channel is $(7.83\\pm0.14)$%, which gives\n\n$B({J/\\psi}{\\rightarrow}\\phi{\\eta^{\\prime}})=(5.11\\pm0.31)\\times 10^{-4}.$\n\nHere, the error is statistical only.",
"\n\nSystematic Errors {#J-sys}\n-----------------\n\nIn this analysis, the systematic errors on the branching fractions mainly come from the following sources:\n\n### MDC tracking efficiency\n\nThe MDC tracking efficiency is measured in clean channels like ${J/\\psi}{\\rightarrow}\\Lambda\\bar{\\Lambda}$ and $\\psi(2S){\\rightarrow}{\\pi^+}{\\pi^-}{J/\\psi},~{J/\\psi}{\\rightarrow}\\mu^+\\mu^-$. It is found that the Monte Carlo simulation agrees with data within 1-2% for each charged track. ",
"Therefore 4% is taken as the systematic error on the tracking efficiency for the channels with two charged tracks and 8% for the channels with four charged tracks in the final states.",
"\n\n### Particle ID\n\nThe particle identification (PID) efficiency of the kaon is studied from ${J/\\psi}{\\rightarrow}{K^+}{K^-}{\\pi^0}$ and ${J/\\psi}{\\rightarrow}\\phi\\eta$. The results indicate that the kaon PID efficiency for data agrees well with that of the Monte Carlo simulation in the kaon momentum region less than 1.0 GeV/c. In the analysis of ${J/\\psi}{\\rightarrow}\\phi{\\pi^0}$, where two charged tracks are required to be kaons, the PID efficiency difference between data and Monte Carlo simulation is about 3.4%. ",
"In other decay modes, at least one charged track is required to be identified as a kaon, so the difference from PID is less than 1%. ",
"Here, the difference of the PID efficiencies between data and Monte Carlo simulation is taken as one of the systematic errors.",
"\n\n### Photon detection efficiency\n\nFor the decay modes analyzed in this paper, one or two photons are involved in the final states. ",
"The photon detection efficiency is studied from ${J/\\psi}{\\rightarrow}\\rho^0{\\pi^0}$ in Ref.",
" [@besrhopi]. ",
"The results indicate that the difference between the detection efficiency of data and MC simulation is less than 2% for each photon.",
"\n\n### Kinematic fit \n\nThe kinematic fit is a useful tool to improve resolution and reduce background. ",
"The systematic error from the kinematic fit is studied with the clean channel ${J/\\psi}{\\rightarrow}{\\pi^+}{\\pi^-}{\\pi^0}$, as described in Ref.",
" [@besrhopi]. ",
"The conclusion is that the kinematic fit efficiency difference between data and Monte Carlo simulation is about 4.1%. ",
"Using the same method, the decay mode ${J/\\psi}{\\rightarrow}{\\pi^+}{\\pi^-}{\\pi^+}{\\pi^-}{\\pi^0}$ is also analyzed, and the kinematic fit efficiency difference between data and Monte-Carlo is about 4.3%. ",
"In this paper, 5% is conservatively taken to be the systematic error from the kinematic fit for all analyzed decay modes.",
"\n\n### Selection criteria\n\nThe systematic errors for additional selection criteria in specific decay modes are estimated by comparing the efficiency difference with and without the criterion or replacing it with a very loose requirement. ",
"The study indicates that they are not large compared with other systematic errors. ",
"The results are listed in Table \\[toterr\\]\n\n### Uncertainty from hadronic interaction model\n\nDifferent simulations of the hadronic interaction lead to different efficiencies. ",
"In this analysis, two models, FLUKA [@FLUKA] and GCALOR [@GCALOR], are used in simulating hadronic interactions in the Monte-Carlo. ",
"The difference of the detection efficiencies from these two Monte Carlo models is about 3%, which is taken as the systematic error.",
"\n\n### Uncertainty of background\n\nThe uncertainties of the background in each channel are estimated by changing the background shape in the fit. ",
"The results are listed in Table \\[toterr\\].",
"\n\n### Intermediate decay branching fractions\n\nThe branching fractions of $\\phi{\\rightarrow}{K^+}{K^-}$ and the pseudoscalar decays are taken from the PDG. ",
"The errors of these branching fractions are systematic errors in our measurements and are listed in Table \\[toterr\\].",
"\n\nThe systematic error contributions studied above, the error due to the uncertainty of the number of ${J/\\psi}$ events, and the statistical error of the Monte-Carlo samples are all listed in Table \\[toterr\\]. ",
"The total systematic error is the sum of them added in quadrature.",
"\n\n ${J/\\psi}{\\rightarrow}$ $\\phi{\\pi^0}$ \n ---------------------------- ------------------------------ ------------------------------ ------------------------------------ -------------------------------------------- ------------------------------ ------------------------------------ --------------------------------------------\n Final states ${K^+}{K^-}{\\gamma}{\\gamma}$ ${K^+}{K^-}{\\gamma}{\\gamma}$ ${K^+}{K^-}{\\pi^+}{\\pi^-}{\\gamma}$ ${K^+}{K^-}{\\pi^+}{\\pi^-}{\\gamma}{\\gamma}$ ${K^+}{K^-}{\\gamma}{\\gamma}$ ${K^+}{K^-}{\\pi^+}{\\pi^-}{\\gamma}$ ${K^+}{K^-}{\\pi^+}{\\pi^-}{\\gamma}{\\gamma}$\n Error Sources \n MDC tracking 4 4 8 8 4 8 8\n Particle ID 3.4 $<$1 $<$1 $<$1 $<$1 $<$1 $<$1\n Kinematic fit 5 5 5 5 5 5 5\n Photon efficiency 4 4 2 4 4 2 4\n Selection criteria 2.4 2.4 2.8 1 2.4 2.9 2.2\n MC sample 1.2 1.2 1.5 2.1 1.2 1.6 1.8\n Hadronic interaction model 3 3 3 3 3 3 3\n Background uncertainty 16.7 3.9 1.5 3.4 1.5 2.0 1.5\n Intermediate decays 1.2 1.4 2.7 2.2 6.7 3.6 3.6\n Total ${J/\\psi}$ events 4.7 4.7 4.7 4.7 4.7 4.7 4.7\n Total systematic error 19.7 10.7 12.0 12.6 12.0 12.4 12.7\n\n\\[toterr\\]\n\nResults and Discussion\n======================\n\nThe branching fractions of ${J/\\psi}$ decaying into $\\phi{\\pi^0}$, $\\phi\\eta$, and $\\phi{\\eta^{\\prime}}$, measured into different final states, are listed in Table \\[brpv\\]. ",
"The average value is the weighted mean of the results from the different decay modes, and the PDG value is the world average taken from Ref.",
" [@pdg2004]. ",
"The world averages mainly come from MarkIII and DM2. ",
"The results obtained here are not in good agreement with previous measurements. ",
"Just as for the branching fraction of ${J/\\psi}{\\rightarrow}{\\pi^+}{\\pi^-}{\\pi^0}$, the branching fraction of ${J/\\psi}{\\rightarrow}\\phi\\eta$ and $\\phi{\\eta^{\\prime}}$ are higher than those in the PDG.",
"\n\nIn this paper, we measured the branching fractions of ${J/\\psi}$ decays into $\\phi$ plus a pseudoscalar. ",
"The three branching fractions are not sufficient for a detailed study of pseudoscalar mixing, SU(3) breaking, and the contribution from doubly suppressed OZI processes using the phenomenological model in Ref.",
" [@theory]. ",
"However the inconsistency between the results from BESII and those from former measurements emphasize the importance for such a study. ",
"After measuring the other decay modes of ${J/\\psi}{\\rightarrow}VP$, such as ${J/\\psi}{\\rightarrow}\\omega{\\pi^0}$, $\\omega\\eta$, $\\omega{\\eta^{\\prime}}$, $\\rho\\eta$, $\\rho{\\eta^{\\prime}}$, and $K^*K$, it will be important to extract physics with all the relevant measurements again.",
"\n\n ${J/\\psi}{\\rightarrow}$ Final states Branching Fraction ($\\times 10^{-4}$)\n ------------------------- -------------------------------------------- ---------------------------------------\n $\\phi{\\pi^0}$ ${K^+}{K^-}{\\gamma}{\\gamma}$ $<0.064$ (C.L. 90%) [@foot]\n ${K^+}{K^-}{\\gamma}{\\gamma}$ 8.67$\\pm$0.19$\\pm$0.93\n ${K^+}{K^-}{\\pi^+}{\\pi^-}{\\gamma}$ 9.79$\\pm$1.02$\\pm$1.17\n $\\phi\\eta$ ${K^+}{K^-}{\\pi^+}{\\pi^-}{\\gamma}{\\gamma}$ 9.41$\\pm$0.30$\\pm$1.19\n Average 8.99$\\pm$0.18$\\pm$0.89\n PDG $6.5\\pm0.7$\n ${K^+}{K^-}{\\gamma}{\\gamma}$ 6.10$\\pm$1.34$\\pm$0.73\n $\\phi{\\eta^{\\prime}}$ ${K^+}{K^-}{\\pi^+}{\\pi^-}{\\gamma}$ 5.64$\\pm$0.35$\\pm$0.70\n ${K^+}{K^-}{\\pi^+}{\\pi^-}{\\gamma}{\\gamma}$ 5.11$\\pm$0.31$\\pm$0.65\n Average 5.40$\\pm$0.25$\\pm$0.56\n PDG $3.3\\pm 0.4$\n\n : Branching fractions of ${J/\\psi}{\\rightarrow}\\phi{\\pi^0}$, $\\phi\\eta$, and $\\phi{\\eta^{\\prime}}$.\n\n\\[brpv\\]\n\nThe BES collaboration thanks the staff of BEPC and computing center for their hard efforts. ",
"This work is supported in part by the National Natural Science Foundation of China under contracts Nos. ",
"19991480, 10225524, 10225525, the Chinese Academy of Sciences under contract No. ",
"KJ 95T-03, the 100 Talents Program of CAS under Contract Nos. ",
"U-11, U-24, U-25, and the Knowledge Innovation Project of CAS under Contract Nos. ",
"U-602, U-34 (IHEP); by the National Natural Science Foundation of China under Contract No. ",
"10175060 (USTC), and No. ",
"10225522 (Tsinghua University) and by the Department of Energy under Contract No. ",
"DE-FG03-94ER40833 (U Hawaii).",
"\n\n[120]{} H. E. Haber, J. Perrier, Phys. ",
"Rev. D [**32**]{}, 2961 (1985). ",
"R. M. Baltrusaitis [*et al.*]{}, ",
"Phys. ",
"Rev. D [**32**]{}, 2883 (1985). ",
"D. Coffman [*et al.*]{}, ",
"Phys. ",
"Rev. D [**38**]{}, 22695 (1988). ",
"J. Jousset [*et al.*]{}, ",
"Phys. ",
"Rev. D [**41**]{}, 1389 (1990). ",
"J. Z. Bai [*et al.*]{}, ",
"Nucl. ",
"Instrum. ",
"Methods A [**458**]{}, 627 (2001).",
"\n\nJ. Z. Bai [*et al.*]{}, ",
"Phys. ",
"Rev. D [**70**]{}, 012005 (2004). ",
"S. Eidelman [*et al.*]{}",
" (Particle Data Group), Phys. ",
"Lett. ",
"B [**592**]{}, 1 (2004), and references therein.",
"\n\nJ. J. Aubert [*et al.*]{}, ",
"Phys. ",
"Rev. Lett. [**",
"33**]{}, 1404 (1974); J. E. Augustin [*et al.*]{}, ",
"Phys. ",
"Rev. Lett. [**",
"33**]{}, 1406 (1974); B. Jean-Marie [*et al.*]{}, ",
"Phys. ",
"Rev. Lett. [**",
"36**]{}, 291 (1976); W. Braunschweig [*et al.*]{}, ",
"Phys. ",
"Lett. [**",
"63B**]{}, 487 (1976); W. Bartel et al., ",
"Phys. ",
"Lett. ",
" [**64B**]{}, 483 (1976); PLUTO collaboration, Phys. ",
"Lett. [**",
"72B**]{}, 493 (1978); DASP collaboration, Phys. ",
"Lett. [**",
"74B**]{}, 292 (1978); D. Coffman et al., ",
"Phys. ",
"Rev. D [**38**]{}, 2695 (1988); J. Z. Bai et al., ",
"Phys. ",
"Rev. D [**54**]{}, 1221 (1996).",
"\n\nThe angular distribution is described by\n\n${\\frac{d^3\\sigma}{d\\cos\\theta_V d\\cos\\theta_1 d\\phi_1}}\n={\\sin^2\\theta_1[1+\\cos^2\\theta_V+\\sin^2\\theta_V\\cos(2\\phi_1)]}$\n\nwhere $\\theta_V$ is the angle between the vector meson and the positron direction. ",
"$\\theta_1$ and $\\phi_1$ describe the decay products of the vector meson in its helicity frame. ",
"For $\\phi{\\rightarrow}{K^+}{K^-}$, $\\theta_1$ and $\\phi_1$ are the polar and azimuthal angles of the momentum of $K$ with respect to the helicity direction of the $\\phi$.\n\nS. S. Fang [*et al.*]{}, ",
"High Energy Phys. ",
"Nucl. ",
"Phys. [**",
"27**]{}, 277 (2003) (in Chinese).",
"\n\nK. Hanssgen, H.-J.Mohring and J. Ranft, Nucl. ",
"Sci. ",
"Eng. [**",
"88**]{}, 551 (1984); J. Ranft and S. Ritter, Z. Phys. ",
"C [**20**]{}, 347 (1983); A. Fasso et al., ",
"FLUKA 92, Proceedings of the Workshop on Simulating Accelerator Radiation Environments, Santa Fe, 1993. ",
"C. Zeitnitz and T. A. Gabriel, Nucl. ",
"Instrum. ",
"Methods A [**349**]{}, 106 (1994).",
"\n"
] |
{
"pile_set_name": "ArXiv"
}
|
[
0.0009777327068150043,
0.0006395344971679151,
0.0007961041992530227,
0.044114403426647186,
0.0008754635928198695,
0.0016783070750534534,
0.0006234614411368966,
0.0008212164393626153,
0.0011702714255079627,
0.001339172595180571,
0.0006193588487803936,
0.0006810207851231098,
0.0009024475584737957,
0.0006143481004983187,
0.000630836293566972,
0.0015607503009960055,
0.0007881190394982696,
0.0007796669378876686,
0.0006698875222355127,
0.0005683036288246512,
0.0005345529643818736,
0.0005523444269783795,
0.0032533511985093355,
0.0021294166799634695,
0.0006711348542012274,
0.0006726014544256032,
0.001185585861094296,
0.0005449815071187913,
0.0006694120238535106,
0.0009762731497175992,
0.0006988798268139362,
0.0012900278670713305,
0.000560414744541049,
0.0006145167862996459,
0.0005457448423840106,
0.0007354449480772018,
0.0034858169965445995,
0.0009821646381169558,
0.0009797053644433618,
0.0029176122043281794,
0.0008704105275683105,
0.0011772800935432315,
0.0008937644306570292,
0.000625329150352627,
0.0009574485593475401,
0.0020436339545994997,
0.003760221181437373,
0.00496775284409523,
0.0007302997400984168,
0.0028731017373502254,
0.0007201044354587793,
0.0007099927752278745,
0.006971090100705624,
0.00791178923100233,
0.003303329460322857,
0.0009269680012948811,
0.0007491401047445834,
0.0007062469958327711,
0.0008862708345986903,
0.0007969224243424833,
0.002764855744317174,
0.0030491796787828207,
0.0025563521776348352,
0.001432066666893661,
0.0069923666305840015,
0.0035730376839637756,
0.0011089768959209323,
0.001037410693243146,
0.0070175714790821075,
0.0011336570605635643,
0.0012886800104752183,
0.0008829381549730897,
0.0006417392287403345,
0.0007634286885149777,
0.0007761971792206168,
0.0010279121343046427,
0.0005680304602719843,
0.000667957472614944,
0.0009268001886084676,
0.0006009511998854578,
0.0007076190668158233,
0.0009168428368866444,
0.0009268001886084676,
0.0006361496634781361,
0.0009839385747909546,
0.0006709855515509844,
0.0006452776724472642,
0.0006933601689524949,
0.0006126987864263356,
0.000625535671133548,
0.0006456252303905785,
0.0005844134138897061,
0.0006254394538700581,
0.0009392232168465853,
0.0010585840791463852,
0.0005997304688207805,
0.0009177317260764539,
0.05088989436626434,
0.0006362654967233539,
0.0007722838781774044,
0.0006974067073315382,
0.0006138702156022191,
0.0024264014791697264,
0.0007295731920748949,
0.0007048830739222467,
0.0007208414608612657,
0.0005700542824342847,
0.0027211999986320734,
0.022511687129735947,
0.0005868040025234222,
0.000660644203890115,
0.0006272594910115004,
0.00062683189753443,
0.0006051675300113857,
0.0006733874906785786,
0.0005936069064773619,
0.002852067584171891,
0.0007926131365820765,
0.0007113771280273795,
0.0007922789664007723,
0.0012429299531504512,
0.0006993392016738653,
0.0007063732482492924,
0.0012429299531504512,
0.0007316273986361921,
0.0007408285164274275,
0.0012429299531504512,
0.0007136226049624383,
0.0007172921323217452,
0.001070261001586914,
0.0013108622515574098,
0.0006654416210949421,
0.0007172921323217452,
0.0012429299531504512,
0.0007382758194580674,
0.0007317254203371704,
0.0006266467389650643,
0.0019435402937233448,
0.0006491724634543061,
0.000724396260920912,
0.0012429299531504512,
0.0009571398259140551,
0.0007113835890777409,
0.0012429299531504512,
0.0009571398259140551,
0.0006860612193122506,
0.0012429299531504512,
0.0009571398259140551,
0.0007271891226992011,
0.0012429299531504512,
0.0018622804200276732,
0.0007194363279268146,
0.0012429299531504512,
0.0019435402937233448,
0.0007376629509963095,
0.0018622804200276732,
0.0007098228670656681,
0.0018622804200276732,
0.0006689560250379145,
0.0012429299531504512,
0.0006867438787594438,
0.0012429299531504512,
0.000714822846930474,
0.002225861418992281,
0.0007482076180167496,
0.0016889135586097836,
0.0008881429093889892,
0.001070261001586914,
0.001444116816855967,
0.0006507777725346386,
0.0008355921600013971,
0.0008689719252288342,
0.0008401460945606232,
0.0008089133771136403,
0.000728427607100457,
0.0006216929177753627,
0.0008570729405619204,
0.0013108622515574098,
0.0006694503244943917,
0.001995444530621171
] | 0.00186 | 180 |
[
"Nicola Sturgeon to warn David Cameron: More powers or another referendum\n\nScottish First Minister Nicola Sturgeon has warned David Cameron that she could demand another independence referendum if he does not agree to more powers for Holyrood.",
"\n\nThe SNP leader is due to meet Cameron on Friday 15 May and will demand control over the minimum wage, national insurance contributions, welfare, business taxes and equality policy.",
"\n\nSpeaking in the Scottish parliament yesterday, Sturgeon warned Cameron that, if he did not listen, he could face another Scottish referendum.",
"\n\nShe said: \"I can't impose a referendum against the will of the Scottish people, but nor can David Cameron rule out a referendum against the will of the people. ",
"It will be the people who decide.",
"\n\n\"And what happens to public opinion on this question in the years ahead will depend not just on what the SNP and the Scottish government do, but also on the respect shown to the decisions the people of Scotland have made.",
"\n\n\"How David Cameron, his government and the Westminster system choose to respond to the message Scotland has sent will be crucial to how we move forward.\"",
"\n\nCameron has ruled out granting the Scots another referendum on independence, but the nationalists are hoping that their sweeping victory on 7 May might force his hand.",
"\n\nSturgeon also urged Scottish Labour to back her plans, by saying that Scottish unions supported her.",
"\n\nMeanwhile, the SNP appointed Alex Salmond as its foreign affairs spokesman in a move that Willie Rennie, the Scottish Liberal Democrat leader, said was like \"putting Mr Bean in charge of the World Bank\".",
"\n\nSalmond said last night that his party would \"provide a strong, consistent voice which is pro-European, pro-developing world and against military adventurism\"."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0008532866486348212,
0.0006288780714385211,
0.003482495667412877,
0.0007056149770505726,
0.000645578489638865,
0.0005524115404114127,
0.0005415592459030449,
0.0016302065923810005,
0.0008984485757537186,
0.000859397929161787,
0.0007968973368406296
] | 0.001054 | 11 |
[
"Thanks to development studios Lizardcube, Guard Crush Games, and Dotemu Streets of Rage 4 is coming after 25 years of the series laying untouched. ",
"The game will feature hand-drawn graphics from the team behind the Wonder Boy: The Dragon’s Trap remake. ",
"No release date is announced as of yet, but we did get a pretty sweet trailer. ",
"Check it out here –\n\nLike this: Like Loading..."
] |
{
"pile_set_name": "OpenWebText2"
}
|
[
0.000677076168358326,
0.0006744664860889316,
0.000571158598177135,
0.0007009498658590019
] | 0.000656 | 4 |
[
"Cholesterol metabolism in relation to aging and dietary fat in rats and humans.",
"\nA review of research in the authors' laboratories regarding effects of dietary fat polyunsaturation upon longevity in rats and some aspects of the regulation of cholesterol metabolism with regard to age of rats and humans is presented. ",
"The longevity of the rat was found to be enhanced by consumption of dietary fat providing a polyunsaturated to saturated fatty acid (P/S) ratio of 0.3 to 1, corresponding to about 5-12% of energy (en%) as linoleate, compared with less or more polyunsaturated fat. ",
"Mechanisms of the effects of the fats upon cholesterol metabolism were studied. ",
"With advancing age, there seems to be a decline in the rate of catabolism of cholesterol, resulting in longer retention in the body of the rat. ",
"In the human, there seems to be a decline in regulation of uptake of cholesterol by leukocytes and, therefore, perhaps other tissues, resulting in increased synthesis of cholesterol by the peripheral tissues. ",
"Moderate rather than high dietary consumption of polyunsaturated fat seems to be favorable to metabolic processes contributing to longevity."
] |
{
"pile_set_name": "PubMed Abstracts"
}
|
[
0.02049202285706997,
0.0005927961319684982,
0.003356806468218565,
0.0008525963057763875,
0.0009221590589731932,
0.0006796284578740597,
0.001470560790039599
] | 0.004052 | 7 |
[
"Structurally engineered anodic alumina nanotubes as nano-carriers for delivery of anticancer therapeutics.",
"\nHere, we report a study on the biocompatibility, cell uptake and in vitro delivery of tumor necrosis factor-related apoptosis-inducing ligand (Apo2L/TRAIL) by new nano-carriers called anodic alumina nanotubes (AANTs) for potential cancer therapy. ",
"AANTs were electrochemically engineered by a unique pulse anodization process, which enables precise control of the nanotube geometry, and used here as nano-carriers for drug delivery. ",
"In vitro cytotoxicity and cell uptake of AANTs was assessed using MDA-MB231-TXSA human breast cancer cells and mouse RAW 264.7 macrophage cells. ",
"AANTs exhibited excellent biocompatibility in both cell lines over a time course of five days even at a maximum concentration of AANTs of 100 μgmL(-1). ",
"Transmission electron microscopy and fluorescence microscopy confirmed a significant uptake of AANTs by RAW 264.7 cells and breast cancer cells. ",
"AANTs loaded with the pro-apoptotic protein Apo2L/TRAIL showed exceptional loading capacity (104 ± 14.4 μgmg(-1) of AANTs) and demonstrated significant decrease in viability of MDA-MB231-TXSA cancer cells due to apoptosis induction. ",
"These results demonstrate that AANTs are promising nano-carriers for drug delivery applications."
] |
{
"pile_set_name": "PubMed Abstracts"
}
|
[
0.000771278515458107,
0.0006189929554238915,
0.0006141663179732859,
0.0007786842179484665,
0.0006229600985534489,
0.0006874805549159646,
0.0007056835456751287,
0.0006247817073017359
] | 0.000678 | 8 |
[
"1. ",
"Field of the Invention\nThe present invention generally relates to a thin film transistor liquid crystal display, and more particularly to a thin film transistor liquid crystal display capable of fast driving and having enhanced display quality.",
"\n2. ",
"Description of the Related Art\nThe thin film transistor liquid crystal display (hereinafter \"TFT-LCD\") has advantages of excellent response characteristics and is appropriate for high number of pixels, so that the TFT-LCD is able to realize high display quality and large size of display devices comparable with the cathode ray tube (CRT).",
"\nThe TFT-LCD comprises a TFT array substrate in which a thin film transistor and a pixel electrode are formed, a color filter substrate in which a color filter and a counter electrode are formed, and a liquid crystal layer sandwiched between the TFT array substrate and the color filter substrate.",
"\nIn the TFT-LCD, it is required to obtain excellent display quality that a first signal applied from data line should be uniformly maintained until a second signal is applied. ",
"So as to uniformly maintain the applied signal, a storage electrode for obtaining a storage capacitance C.sub.st is provided at each pixel. ",
"The storage electrode is provided in the form of a line separated from a gate line, or in another form protruding from the gate line. ",
"The former type is called as \"storage on common\" and the later \"storage on gate\".",
"\nFIG. ",
"1 is a plane view for showing a TFT array substrate provided with a conventional storage electrode of the storage on gate type. ",
"As shown in the drawing, a plurality of gate lines 1 are arranged in rows and a plurality of data bus lines 4 are arranged perpendicular to the gate lines 1. ",
"A storage electrode 2 is provided in the form protruding from the gate line 1 within the pixel area defined by a pair of gate lines 1 and a pair of data lines 4. ",
"A pixel electrode 6 made of a transparent metal layer, for example an ITO metal layer, is disposed to be overlapped with the storage electrode 2 within the pixel area.",
"\nA TFT 10 is formed at a portion of intersection of the gate line 1 and the data line 4. ",
"The TFT 10 includes a gate electrode, i.e. a part of the gate line 1, a source electrode 7 and a drain electrode 8 disposed to be overlapped with the gate electrode. ",
"The drain electrode 8 has a form withdrawn from the data line 4, and the source electrode 7 is disposed to be opposed with the drain electrode 8 and in contact with the pixel electrode 6.",
"\nHowever, it is very difficult to utilize the fast operation in the TFT-LCD having the above-described TFT array substrate since there is occurred RC-Delay due to a storage capacitance C.sub.st, a parasitic capacitance C.sub.gs between a part of the gate line i.e. the gate electrode and the source electrode, and a parasitic capacitance C.sub.gd between the gate electrode and the drain electrode defined as following equation 1. ",
"EQU .tau.=R.multidot.",
"C.sub.total =R.multidot.(.SIGMA.C.sub.gs +.SIGMA.C.sub.gd.",
"SIGMA.C.sub.st) equation 1\nHerein, .tau. ",
"means a degree of signal delay and R means resistance value of the gate line.",
"\nFurther, the TFT array substrate as constituted above is formed conventionally by a divisional exposure process. ",
"However, as shown in FIGS. ",
"2a and 2b, the overlapping rate of a part of the gate line 1, i.e. the gate electrode and the source electrode 7, and the overlapping rate of the gate electrode and the drain electrode 8 are changed due to a misalign of an exposing mask, thereby occurring changes in the value of parasitic capacitance. ",
"Therefore, as shown in following equation 2, display quality is degraded due to the difference of brightness between the respective divisional exposure regions.",
"\n{character pullout}V.sub.p =C.sub.gs /C.sub.gs +C.sub.lc +C.sub.st X {character pullout}V.sub.g equation 2\nHerein, {character pullout}V.sub.p means the variation of pixel voltage, i.e. the kick-back voltage, C.sub.gs means a parasitic capacitance between the gate electrode and the source electrode, C.sub.lc means a parasitic capacitance of liquid crystal, and {character pullout}V.sub.g means the variation of gate voltage."
] |
{
"pile_set_name": "USPTO Backgrounds"
}
|
[
0.0009391900966875255,
0.0006004453753121197,
0.001185585861094296,
0.0005736000020988286,
0.000690500601194799,
0.0005787605186924338,
0.0005887992447242141,
0.0005778835038654506,
0.0005977853434160352,
0.0008292015991173685,
0.0006208558916114271,
0.0006018952699378133,
0.0006051097880117595,
0.0006174601730890572,
0.0006766599835827947,
0.0006772009655833244,
0.0006136663141660392,
0.000642020080704242,
0.0010337411658838391,
0.003473594319075346,
0.0007197448867373168,
0.000664907565806061,
0.0005777268670499325,
0.0006229138234630227,
0.0006116107106208801,
0.000550556811504066,
0.0009057855932042003
] | 0.000792 | 27 |
[
"Judy Hoffman\n\nJudy Hoffman may refer to:\n\n Judy Hoffman (artist)\n Judy Hoffman (filmmaker)"
] |
{
"pile_set_name": "Wikipedia (en)"
}
|
[
0.0007007221574895084
] | 0.000701 | 1 |
[
"Q:\n\nDoes Kubernetes support connection draining?",
"\n\nDoes Kubernetes support connection draining?",
"\nFor example, my deployment rolls out a new version of my web app container. ",
"\nIn connection draining mode Kubernetes should spin up a new container from the new image and route all new traffic coming to my service to this new instance. ",
"The old instance should remain alive long enough to send a response for existing connections.",
"\n\nA:\n\nKubernetes does support connection draining, but how it happens is controlled by the Pods, and is called graceful termination.",
"\nGraceful Termination\nLet's take an example of a set of Pods serving traffic through a Service. ",
"This is a simplified example, the full details can be found in the documentation.",
"\n\nThe system (or a user) notifies the API that the Pod needs to stop.",
"\nThe Pod is set to the Terminating state. ",
"This removes it from a Service serving traffic. ",
"Existing connections are maintained, but new connections should stop as soon as the load balancers recognize the change.",
"\nThe system sends SIGTERM to all containers in the Pod.",
"\nThe system waits terminationGracePeriodSeconds (default 30s), or until the Pod completes on it's own.",
"\nIf containers in the Pod are still running, they are sent SIGKILL and terminated immediately. ",
"At this point the Pod is forcefully terminated if it is still running.",
"\n\nThis not only covers the simple termination case, but the exact same process is used in rolling update deployments, each Pod is terminated in the exact same way and is given the opportunity to clean up.",
"\nUsing Graceful Termination For Connection Draining\nIf you do not handle SIGTERM in your app, your Pods will immediately terminate, since the default action of SIGTERM is to terminate the process immediately, and the grace period is not used since the Pod exits on its own.",
"\nIf you need \"connection draining\", this is the basic way you would implement it in Kubernetes:\n\nHandle the SIGTERM signal, and clean up your connections in a way that your application decides. ",
"This may simply be \"do nothing\" to allow in-flight connections to clear out. ",
"Long running connections may be terminated in a way that is (more) friendly to client applications.",
"\nSet the terminationGracePeriodSeconds long enough for your Pod to clean up after itself.",
"\n\n"
] |
{
"pile_set_name": "StackExchange"
}
|
[
0.0007918888004496694,
0.0007777053979225457,
0.0007508397102355957,
0.0006999388569965959,
0.000598931044805795,
0.0006139960023574531,
0.000554097699932754,
0.0005564038292504847,
0.0007242639549076557,
0.0005915501969866455,
0.0007548701833002269,
0.0005953131476417184,
0.0010509949643164873,
0.0008384051034227014,
0.0012198048643767834,
0.0006052548997104168,
0.0005904441932216287,
0.0006136873853392899,
0.0006065224297344685,
0.0007248540641739964,
0.0005636299028992653,
0.013433998450636864,
0.001995444530621171
] | 0.001315 | 23 |
[
"Rural-Urban Differences in Access to Specialist Providers of Colorectal Cancer Care in the United States: A Physician Workforce Issue.",
"\nAlthough early detection and treatment of colorectal cancer has been shown to improve outcomes, geographic proximity may influence access to these services. ",
"To examine the disparities that may exist in colorectal cancer screening and treatment by comparing the distribution of providers of these services in rural and urban counties in the United States. ",
"A retrospective population-based study using data obtained from the 2009 Area Resource File for the entire US population within each county. ",
"Counties in the United States were categorized as rural or urban using rural-urban continuum codes as our primary exposure. ",
"The proportion of gastroenterologists, general surgeons, and radiation oncologists per 100,000 people in each county was estimated as primary outcomes. ",
"Multivariate linear regression analysis adjusted for county-level socioeconomic variables, such as percentages of females, blacks, population without insurance, those with a high school diploma, and median household income, to estimate the relative density of each category of these providers between urban and rural counties. ",
"In total, 3220 counties were identified, comprising 1807 rural and 1413 urban counties. ",
"An unadjusted analysis showed an increased density of gastroenterologists, general surgeons, and radiation oncologists per 100,000 people in urban vs rural counties. ",
"A multivariable analysis revealed a significantly higher density of gastroenterologists (1.63; 95% CI, 1.40-1.85; P < .001), general surgeons (2.01; 95% CI, 1.28-2.73; P < .001), and radiation oncologists (0.68; 95% CI, 0.59-0.77; P < .001) per 100,000 people living in urban vs rural counties. ",
"A rural-urban disparity exists in the density of gastroenterologists, general surgeons, and radiation oncologists who traditionally provide colorectal cancer screening services and treatment. ",
"This might affect access to these services and may negatively influence outcomes for colorectal cancer in rural areas."
] |
{
"pile_set_name": "PubMed Abstracts"
}
|
[
0.0006645274697802961,
0.0007590139284729958,
0.0006121443002484739,
0.0005573339876718819,
0.0005654136184602976,
0.0005834897747263312,
0.000763112329877913,
0.0006620989297516644,
0.0005761423381045461,
0.0006309541640803218,
0.0007514986209571362,
0.0013376392889767885
] | 0.000705 | 12 |
[
"syntax = \"proto3\";\n\npackage org.platanios.tensorflow.proto;\n\nimport \"any.proto\";\nimport \"graph.proto\";\nimport \"op_def.proto\";\nimport \"tensor_shape.proto\";\nimport \"types.proto\";\nimport \"saved_object_graph.proto\";\nimport \"saver.proto\";\nimport \"struct.proto\";\n\noption cc_enable_arenas = true;\noption java_outer_classname = \"MetaGraphProtos\";\noption java_multiple_files = true;\noption java_package = \"org.platanios.tensorflow.proto\";\noption go_package = \"github.com/tensorflow/tensorflow/tensorflow/go/core/core_protos_go_proto\";\n\n// NOTE: This protocol buffer is evolving, and will go through revisions in the\n// coming months.",
"\n//\n// Protocol buffer containing the following which are necessary to restart\n// training, run inference. ",
"It can be used to serialize/de-serialize memory\n// objects necessary for running computation in a graph when crossing the\n// process boundary. ",
"It can be used for long term storage of graphs,\n// cross-language execution of graphs, etc.",
"\n// MetaInfoDef\n// GraphDef\n// SaverDef\n// CollectionDef\n// TensorInfo\n// SignatureDef\nmessage MetaGraphDef {\n // Meta information regarding the graph to be exported. ",
" To be used by users\n // of this protocol buffer to encode information regarding their meta graph.",
"\n message MetaInfoDef {\n // User specified Version string. ",
"Can be the name of the model and revision,\n // steps this model has been trained to, etc.",
"\n string meta_graph_version = 1;\n\n // A copy of the OpDefs used by the producer of this graph_def.",
"\n // Descriptions and Ops not used in graph_def are stripped out.",
"\n OpList stripped_op_list = 2;\n\n // A serialized protobuf. ",
"Can be the time this meta graph is created, or\n // modified, or name of the model.",
"\n google.",
"Any any_info = 3;\n\n // User supplied tag(s) on the meta_graph and included graph_def.",
"\n //\n // MetaGraphDefs should be tagged with their capabilities or use-cases.",
"\n // Examples: \"train\", \"serve\", \"gpu\", \"tpu\", etc.",
"\n // These tags enable loaders to access the MetaGraph(s) appropriate for a\n // specific use-case or runtime environment.",
"\n repeated string tags = 4;\n\n // The __version__ string of the tensorflow build used to write this graph.",
"\n // This will be populated by the framework, which will overwrite any user\n // supplied value.",
"\n string tensorflow_version = 5;\n\n // The __git_version__ string of the tensorflow build used to write this\n // graph. ",
"This will be populated by the framework, which will overwrite any\n // user supplied value.",
"\n string tensorflow_git_version = 6;\n\n // A flag to denote whether default-valued attrs have been stripped from\n // the nodes in this graph_def.",
"\n bool stripped_default_attrs = 7;\n\n // FunctionDef name to aliases mapping.",
"\n map<string, string> function_aliases = 8;\n }\n MetaInfoDef meta_info_def = 1;\n\n // GraphDef.",
"\n GraphDef graph_def = 2;\n\n // SaverDef.",
"\n SaverDef saver_def = 3;\n\n // collection_def: Map from collection name to collections.",
"\n // See CollectionDef section for details.",
"\n map<string, CollectionDef> collection_def = 4;\n\n // signature_def: Map from user supplied key for a signature to a single\n // SignatureDef.",
"\n map<string, SignatureDef> signature_def = 5;\n\n // Asset file def to be used with the defined graph.",
"\n repeated AssetFileDef asset_file_def = 6;\n\n // Extra information about the structure of functions and stateful objects.",
"\n SavedObjectGraph object_graph_def = 7;\n}\n\n// CollectionDef should cover most collections.",
"\n// To add a user-defined collection, do one of the following:\n// 1. ",
"For simple data types, such as string, int, float:\n// tf.add_to_collection(\"your_collection_name\", your_simple_value)\n// strings will be stored as bytes_list.",
"\n//\n// 2. ",
"For Protobuf types, there are three ways to add them:\n// 1) tf.add_to_collection(\"your_collection_name\",\n// your_proto.",
"SerializeToString())\n//\n// collection_def {\n// key: \"user_defined_bytes_collection\"\n// value {\n// bytes_list {\n// value: \"queue_name: \\\"test_queue\\\"\\n\"\n// }\n// }\n// }\n//\n// or\n//\n// 2) tf.add_to_collection(\"your_collection_name\", str(your_proto))\n//\n// collection_def {\n// key: \"user_defined_string_collection\"\n// value {\n// bytes_list {\n// value: \"\\n\\ntest_queue\"\n// }\n// }\n// }\n//\n// or\n//\n// 3) any_buf = any_pb2.Any()\n// tf.add_to_collection(\"your_collection_name\",\n// any_buf.",
"Pack(your_proto))\n//\n// collection_def {\n// key: \"user_defined_any_collection\"\n// value {\n// any_list {\n// value {\n// type_url: \"type.googleapis.com/tensorflow.QueueRunnerDef\"\n// value: \"\\n\\ntest_queue\"\n// }\n// }\n// }\n// }\n//\n// 3. ",
"For Python objects, implement to_proto() and from_proto(), and register\n// them in the following manner:\n// ops.register_proto_function(\"your_collection_name\",\n// proto_type,\n// to_proto=YourPythonObject.to_proto,\n// from_proto=YourPythonObject.from_proto)\n// These functions will be invoked to serialize and de-serialize the\n// collection. ",
"For example,\n// ops.register_proto_function(ops.",
"GraphKeys.",
"GLOBAL_VARIABLES,\n// proto_type=variable_pb2.VariableDef,\n// to_proto=Variable.to_proto,\n// from_proto=Variable.from_proto)\nmessage CollectionDef {\n // NodeList is used for collecting nodes in graph. ",
"For example\n // collection_def {\n // key: \"summaries\"\n // value {\n // node_list {\n // value: \"input_producer/ScalarSummary:0\"\n // value: \"shuffle_batch/ScalarSummary:0\"\n // value: \"ImageSummary:0\"\n // }\n // }\n message NodeList {\n repeated string value = 1;\n }\n\n // BytesList is used for collecting strings and serialized protobufs. ",
"For\n // example:\n // collection_def {\n // key: \"trainable_variables\"\n // value {\n // bytes_list {\n // value: \"\\n\\017conv1/weights:0\\022\\024conv1/weights/Assign\n // \\032\\024conv1/weights/read:0\"\n // value: \"\\n\\016conv1/biases:0\\022\\023conv1/biases/Assign\\032\n // \\023conv1/biases/read:0\"\n // }\n // }\n // }\n message BytesList {\n repeated bytes value = 1;\n }\n\n // Int64List is used for collecting int, int64 and long values.",
"\n message Int64List {\n repeated int64 value = 1 [packed = true];\n }\n\n // FloatList is used for collecting float values.",
"\n message FloatList {\n repeated float value = 1 [packed = true];\n }\n\n // AnyList is used for collecting Any protos.",
"\n message AnyList {\n repeated google.",
"Any value = 1;\n }\n\n oneof kind {\n NodeList node_list = 1;\n BytesList bytes_list = 2;\n Int64List int64_list = 3;\n FloatList float_list = 4;\n AnyList any_list = 5;\n }\n}\n\n// Information about a Tensor necessary for feeding or retrieval.",
"\nmessage TensorInfo {\n // For sparse tensors, The COO encoding stores a triple of values, indices,\n // and shape.",
"\n message CooSparse {\n // The shape of the values Tensor is [?]. ",
" Its dtype must be the dtype of\n // the SparseTensor as a whole, given in the enclosing TensorInfo.",
"\n string values_tensor_name = 1;\n\n // The indices Tensor must have dtype int64 and shape [?, ?].",
"\n string indices_tensor_name = 2;\n\n // The dynamic logical shape represented by the SparseTensor is recorded in\n // the Tensor referenced here. ",
" It must have dtype int64 and shape [?].",
"\n string dense_shape_tensor_name = 3;\n }\n\n // Generic encoding for composite tensors.",
"\n message CompositeTensor {\n // The serialized TypeSpec for the composite tensor.",
"\n TypeSpecProto type_spec = 1;\n\n // A TensorInfo for each flattened component tensor.",
"\n repeated TensorInfo components = 2;\n }\n\n oneof encoding {\n // For dense `Tensor`s, the name of the tensor in the graph.",
"\n string name = 1;\n // There are many possible encodings of sparse matrices\n // (https://en.wikipedia.org/wiki/Sparse_matrix). ",
" Currently, TensorFlow\n // uses only the COO encoding. ",
" This is supported and documented in the\n // SparseTensor Python class.",
"\n CooSparse coo_sparse = 4;\n // Generic encoding for CompositeTensors.",
"\n CompositeTensor composite_tensor = 5;\n }\n DataType dtype = 2;\n // The static shape should be recorded here, to the extent that it can\n // be known in advance. ",
" In the case of a SparseTensor, this field describes\n // the logical shape of the represented tensor (aka dense_shape).",
"\n TensorShapeProto tensor_shape = 3;\n}\n\n// SignatureDef defines the signature of a computation supported by a TensorFlow\n// graph.",
"\n//\n// For example, a model with two loss computations, sharing a single input,\n// might have the following signature_def map.",
"\n//\n// Note that across the two SignatureDefs \"loss_A\" and \"loss_B\", the input key,\n// output key, and method_name are identical, and will be used by system(s) that\n// implement or rely upon this particular loss method. ",
"The output tensor names\n// differ, demonstrating how different outputs can exist for the same method.",
"\n//\n// signature_def {\n// key: \"loss_A\"\n// value {\n// inputs {\n// key: \"input\"\n// value {\n// name: \"input:0\"\n// dtype: DT_STRING\n// tensor_shape: ...\n// }\n// }\n// outputs {\n// key: \"loss_output\"\n// value {\n// name: \"loss_output_A:0\"\n// dtype: DT_FLOAT\n// tensor_shape: ...\n// }\n// }\n// }\n// ...\n// method_name: \"some/package/compute_loss\"\n// }\n// signature_def {\n// key: \"loss_B\"\n// value {\n// inputs {\n// key: \"input\"\n// value {\n// name: \"input:0\"\n// dtype: DT_STRING\n// tensor_shape: ...\n// }\n// }\n// outputs {\n// key: \"loss_output\"\n// value {\n// name: \"loss_output_B:0\"\n// dtype: DT_FLOAT\n// tensor_shape: ...\n// }\n// }\n// }\n// ...\n// method_name: \"some/package/compute_loss\"\n// }\nmessage SignatureDef {\n // Named input parameters.",
"\n map<string, TensorInfo> inputs = 1;\n // Named output parameters.",
"\n map<string, TensorInfo> outputs = 2;\n // Extensible method_name information enabling third-party users to mark a\n // SignatureDef as supporting a particular method. ",
"This enables producers and\n // consumers of SignatureDefs, e.g. a model definition library and a serving\n // library to have a clear hand-off regarding the semantics of a computation.",
"\n //\n // Note that multiple SignatureDefs in a single MetaGraphDef may have the same\n // method_name. ",
"This is commonly used to support multi-headed computation,\n // where a single graph computation may return multiple results.",
"\n string method_name = 3;\n}\n\n// An asset file def for a single file or a set of sharded files with the same\n// name.",
"\nmessage AssetFileDef {\n // The tensor to bind the asset filename to.",
"\n TensorInfo tensor_info = 1;\n // The filename within an assets directory. ",
"Note: does not include the path\n // prefix, i.e. directories. ",
"For an asset at /tmp/path/vocab.txt, the filename\n // would be \"vocab.txt\".",
"\n string filename = 2;\n}\n"
] |
{
"pile_set_name": "Github"
}
|
[
0.0007503536180593073,
0.000690981512889266,
0.0006629424751736224,
0.0005483961431309581,
0.0007702656439505517,
0.0006310786120593548,
0.0006878947024233639,
0.0005677083972841501,
0.0007505160174332559,
0.0007913634181022644,
0.0009509383235126734,
0.0005899564712308347,
0.0009281968814320862,
0.000657101976685226,
0.0005752539727836847,
0.0006002874579280615,
0.000641722115688026,
0.0007479912601411343,
0.0006022108718752861,
0.0007414998835884035,
0.0005974521045573056,
0.0007988075376488268,
0.0011970453197136521,
0.0011803415836766362,
0.0014481202233582735,
0.000939826131798327,
0.0005976707907393575,
0.0007706723990850151,
0.0007040113559924066,
0.0006347754970192909,
0.0007556626223959029,
0.000967489555478096,
0.000754284905269742,
0.0014022699324414134,
0.0006900678272359073,
0.006178948562592268,
0.0010440292535349727,
0.0007201661937870085,
0.0006967699155211449,
0.0011394666507840157,
0.0006990323308855295,
0.0007357032736763358,
0.0011838942300528288,
0.0007948834099806845,
0.000707095256075263,
0.0008104228763841093,
0.0011024042032659054,
0.0005851851310580969,
0.0006877000560052693,
0.0006754136993549764,
0.0006223648670129478,
0.0006014712271280587,
0.000586514885071665,
0.0006990456604398787,
0.0006270825979299843,
0.0006717482465319335,
0.0007218209793791175,
0.0005874540074728429,
0.0006772532942704856,
0.0005896453512832522,
0.0007279595010913908,
0.000574302626773715,
0.0006151138222776353,
0.0006998421740718186,
0.0006197521579451859,
0.0006881714798510075,
0.0006005752948112786,
0.0009055617265403271,
0.0006641775835305452,
0.0006561126792803407,
0.0005960566340945661,
0.0006709826411679387,
0.0006117424927651882,
0.0007694428786635399,
0.0007011466659605503,
0.0006296001374721527,
0.0006385189481079578,
0.0005707882810384035,
0.0008553588413633406
] | 0.000813 | 79 |
[
"Q:\n\nCreate datatable from unbound datagridview\n\nI want a button that creates a datatable to then bulk copy to a SQL database. ",
"I have been able to merge datatables starting with a datatable created from the SQL table, but I have not been able to just make a datatable from scratch.",
"\nPrivate Sub btnSave_Click(ByVal sender As System.",
"Object, ByVal e As System.",
"EventArgs) Handles btnSave.",
"Click\n\n Dim table As New DataTable\n Dim row As DataRow\n\n Dim TableName As String = \"SQLLocation_\"\n\n table.",
"Columns.",
"Add(TableName & \"Number\", GetType(Int64))\n table.",
"Columns.",
"Add(TableName & \"AnotherNumber\", GetType(Int16))\n table.",
"Columns.",
"Add(TableName & \"Name\", GetType(String))\n table.",
"Columns.",
"Add(TableName & \"Port\", GetType(Int32))\n table.",
"Columns.",
"Add(TableName & \"OnOff\", GetType(Boolean))\n\n For Each drthree As DataGridViewRow In DataGridView3.Rows\n row = table.",
"NewRow 'Create new row\n table.",
"Rows.",
"Add(row)\n Next\n\nEnd Sub\n\nThis code just creates a bunch of blank rows. ",
"Eventually I will add on this code to create the new table.",
"\nUsing destinationConnection As SqlConnection = _\n New SqlConnection(sConnectionString)\n destinationConnection.",
"Open()\n\n ' Set up the bulk copy object. ",
" \n ' The column positions in the source data reader \n ' match the column positions in the destination table, \n ' so there is no need to map columns. ",
"\n\n Using bulkCopy As SqlBulkCopy = _\n New SqlBulkCopy(destinationConnection)\n bulkCopy.",
"DestinationTableName = _\n \"dbo.",
"SQLLocation_Tbl\"\n\n Try\n ' Write from the source to the destination.",
"\n bulkCopy.",
"WriteToServer(table)\n\n Catch ex As Exception\n MessageBox.",
"Show(ex.",
"ToString, _\n \"ERROR\", MessageBoxButtons.",
"OK, MessageBoxIcon.",
"Exclamation)\n System.",
"Threading.",
"Thread.",
"CurrentThread.",
"Abort()\n Finally\n ' Close the SqlDataReader. ",
"The SqlBulkCopy \n ' object is automatically closed at the end \n ' of the Using block.",
"\n End Try\n End Using\n\n End Using\nEnd Sub\n\nA:\n\nThis will allow you to iterate through a datagridview and create a table based on its columns and rows. ",
" It would create columns in the datatable with the same names, rather than it looks like you were trying to do. ",
"\n Dim table As New DataTable()\n For Each col As DataGridViewColumn In dgv.",
"Columns\n table.",
"Columns.",
"Add(col.",
"Name, col.",
"ValueType)\n table.",
"Columns(col.",
"Name).Caption = col.",
"HeaderText\n Next\n\n For Each row As DataGridViewRow In dgv.",
"Rows\n Dim drNewRow As DataRow = table.",
"NewRow()\n For Each col As DataColumn In table.",
"Columns\n drNewRow(col.",
"ColumnName) = row.",
"Cells(col.",
"ColumnName).Value\n Next\n table.",
"Rows.",
"Add(drNewRow)\n Next\n\nIf a different name is a requirement you can retain your current method of creating the columns, but then when you loop through creating each row you will have to know which columns values match the ones you want and map them instead of using that inner For Each.",
"\nSomething like:\n drNewRow(\"SQLLocation_Number\") = row.cell(dgv.",
"Columns(0).ColumnName).Value\n\n"
] |
{
"pile_set_name": "StackExchange"
}
|
[
0.000695524737238884,
0.0005727616371586919,
0.0009635143214836717,
0.0010692771757021546,
0.0007700414280407131,
0.0009402919095009565,
0.0007132727769203484,
0.0006434234092012048,
0.0007132727769203484,
0.0006576142041012645,
0.0007132727769203484,
0.0006468373467214406,
0.0007132727769203484,
0.0006392292561940849,
0.0007132727769203484,
0.0008756704046390951,
0.0008442476391792297,
0.0007687400793656707,
0.0010874060681089759,
0.0005918409442529082,
0.0008291300036944449,
0.0007940029609017074,
0.0006016480620019138,
0.0008806582191027701,
0.0007523470558226109,
0.0007326129707507789,
0.0014548704493790865,
0.001148577081039548,
0.0009276372729800642,
0.0009456266998313367,
0.0006407772889360785,
0.0007380524766631424,
0.0009957521688193083,
0.0010222275741398335,
0.001150845899246633,
0.004101594444364309,
0.0010101761436089873,
0.0013873666757717729,
0.0006357899983413517,
0.0006832206854596734,
0.0006977244629524648,
0.0007132727769203484,
0.0007647189195267856,
0.0007040807395242155,
0.0006554689025506377,
0.0006777778617106378,
0.0007231271010823548,
0.000623483385425061,
0.0024404143914580345,
0.0006363352295011282,
0.0007386470679193735,
0.000809012446552515,
0.0007044831872917712,
0.0007068159757182002,
0.0007687400793656707,
0.0006802434218116105,
0.0006180706550367177,
0.0006882039597257972
] | 0.000881 | 58 |
[
"Swimming photochromic azobenzene single crystals in triacrylate solution.",
"\nSelf-motion of a growing single crystal of azobenzene chromophore in triacrylate solution (TA) is investigated in relation to the solid-liquid phase diagram bound by the solidus and liquidus lines. ",
"Upon thermal quenching from the isotropic melt to the crystal + liquid gap, various single crystals develop in a manner dependent on concentration and supercooling depth. ",
"During the crystal growth, TA solvent is rejected from the growing faceted fronts, enriching with TA in close proximity to the crystal-solution interface. ",
"The concentration gradient that formed as the result of TA expulsion induces convective flows in the solution and generates spatial variability of surface tension usually responsible for Marangoni effect. ",
"Either or both of these phenomena may have contributed to the observed self-motion including swimming, sinking, and floating of the azobenzene rhomboidal crystal in TA solution. ",
"A stationary rhomboidal crystal is also shown to swim upon irradiation with the UV light because of a mechanical torque generated by the trans-cis isomerization. ",
"Judging from the sinking or floating behavior of the azobenzene crystal, it may be inferred that the nucleation occurs at the solution-air interface."
] |
{
"pile_set_name": "PubMed Abstracts"
}
|
[
0.0008077672100625932,
0.0006302116089500487,
0.0006717514479532838,
0.00055285933194682,
0.0006273522158153355,
0.0006154926959425211,
0.0007110565202310681,
0.0005398881039582193
] | 0.000645 | 8 |
[
"6. ",
"Διαφάνεια στην περιφερειακή πολιτική και στη χρηματοδότησή της (\n- Πριν από την ψηφοφορία:\nΜιχάλης Τρεμόπουλος\nεισηγητής. - ",
"Κύριε Πρόεδρε, αγαπητοί συνάδελφοι, κίνητρο για αυτήν την έκθεση σχετικά με τη διαφάνεια στην περιφερειακή πολιτική και τη χρηματοδότησή της ήταν το γεγονός ότι η πλήρης παρουσίαση των δικαιούχων που εισπράττουν κονδύλια της Ευρωπαϊκής Ένωσης επιτρέπει τη συμμετοχή πολιτών σε μια ουσιαστική συζήτηση για το πώς δαπανάται το ευρωπαϊκό δημόσιο χρήμα. ",
"Αυτό είναι απαραίτητο για τη λειτουργία της δημοκρατίας σε ευρωπαϊκό επίπεδο.",
"\n\" έκθεση αυτή, που μου ανατέθηκε από την Επιτροπή Περιφερειακής Ανάπτυξης, περιέχει συστάσεις που θα πρέπει να περιληφθούν σε μελλοντικούς κανονισμούς για τα διαρθρωτικά ταμεία, όπως η παροχή επιπλέον απαραίτητων πληροφοριών κατά τη δημοσίευση των καταλόγων των δικαιούχων και οι επαρκώς δεσμευτικοί κανόνες για την εταιρική σχέση.",
"\nΆλλες προτάσεις που μπορούν να υλοποιηθούν στο πλαίσιο των σημερινών προγραμμάτων συνοχής είναι, για παράδειγμα:\nΟ ορισμός, εκ μέρους της Επιτροπής, μιας πιο λεπτομερούς και περιγραφικής παρουσίασης που θα προσδιορίζει τη δομή, τη μορφή και το περιεχόμενο των πληροφοριών που πρέπει να παρέχονται.",
"\nΑκόμη, η σύνδεση της ευρωπαϊκής πρωτοβουλίας για τη διαφάνεια με τους νέους δημοσιονομικούς ελέγχους και το λογιστικό έλεγχο.",
"\nΜια σκληρότερη γραμμή από τους ελεγκτές για τις απαιτήσεις επικοινωνίας και πληροφόρησης, συμπεριλαμβανομένης και της \"κατονομασίας και διαπόμπευσης\" και της χρήσης των δημοσιονομικών διορθώσεων σε επιβεβαιωμένα κρούσματα απάτης.",
"\nΜια στενότερη εμπλοκή περιφερειακών και τοπικών αρχών και άλλων συναφών εταίρων σε όλα τα στάδια του προγραμματισμού και της εφαρμογής της συνοχής καθώς και πλήρης πρόσβασή τους σε όλα τα έγγραφα του έργου.",
"\nΠερισσότερη καθοδήγηση από την Επιτροπή σχετικά με τον τρόπο εφαρμογής της ρήτρας εταιρικής σχέσης στην πράξη στο πλαίσιο των τρεχόντων προγραμμάτων και βελτίωση της διαφάνειας όσον αφορά την κοινοτική χρηματοδότηση των μεγάλων έργων.",
"\nΘα ήθελα για ακόμη μια φορά να ευχαριστήσω τους σκιώδεις εισηγητές των άλλων πολιτικών ομάδων για αυτό το τελικό κείμενο που μπορέσαμε να επιτύχουμε.",
"\n"
] |
{
"pile_set_name": "EuroParl"
}
|
[
0.001202064217068255,
0.0009588348329998553,
0.0013696521054953337,
0.0010581217939034104,
0.0010918352054432034,
0.0012625163653865457,
0.0010994032490998507,
0.0010133233154192567,
0.0009855644311755896,
0.0011232455726712942,
0.0009703476680442691,
0.001995444530621171
] | 0.001178 | 12 |
[
"Amazing Fun Facts About Valentines Day\n\nThere are many theories on how Valentine’s Day began. ",
"The holiday may go back to ancient Roman times, when a festival called Lupercalia was held on February 15. ",
"Lupercalia was a celebration honouring Juno, the Roman deity of women and marriage. ",
"During the festival, women wrote love notes called billets and put them in a large jar or urn. ",
"Each man would select a note and pursue the woman whose name was on it. ",
"Another theory is that Valentine’s Day began with one or more early Christian saints. ",
"According to one legend, the Roman Emperor Claudius II (A.D. 200s) forbade all marriages, believing that single men made better soldiers.",
"\n\nBut a priest named Valentine broke the law and married many couples in secret. ",
"Another story tells of an early Christian named Valentine who befriended children. ",
"When he was imprisoned by the Romans for his faith, Valentine’s little friends gave him notes and cards through his jail window. ",
"This may explain the tradition of exchanging cards on Valentine’s Day. ",
"Most of the Christian stories agree that Valentine was killed by the Romans on February 14 around A.D. 269. ",
"People began to make and sell valentines on a commercial level in the mid-1800s. ",
"During this time, it also became affordable to mail valentines. (",
"Previously, postage was so expensive that cards were almost always hand-delivered.)",
"\n\nIn the 1830s, Esther A. Howland of Worcester, Massachusetts, became the first American to make mass-produced Valentines. ",
"She hired a staff of women and set up an assembly line: One group glued on flowers, others added lace, and some painted leaves. ",
"Her sales amounted to about $130,000 annually. ",
"Today, there are almost 2,000 greeting card publishers in the United States. ",
"Flowers are symbolic of love blooming and waning. ",
"Red is symbolic of strong emotions; yellow, of jealousy. ",
"Other flowers also have symbolic meanings. ",
"The forget-me-not represents true love. ",
"The periwinkle is used to express early friendship (blue) and happy memories (white).",
"\n\nDid you know?",
"\n\n1. ",
"Hundreds of years ago, children in England dressed up as adults on Valentine’s Day and sang songs door to door.",
"\n\n2. ",
"There is also an old English myth claiming that February 14 is the day the birds of spring choose their mates.",
"\n\n3. ",
"The British artist Kate Greenaway was famous for her Valentines decorated with lovely children and gardens.",
"\n\n4. ",
"The language of flowers was developed in the 1600s in Constantinople and Persia.",
"\n\n5. ",
"The red rose is very popular and was said to be the favourite flower of Venus, the deity of love exchanged during this day.",
"\n\n6. ",
"How you combined a bouquet for your Valentine could say quite a lot ."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.000603809836320579,
0.000672784517519176,
0.0013294649543240666,
0.011405046097934246,
0.000690518761985004,
0.0007618283852934837,
0.0007430550176650286,
0.0010633395286276937,
0.0006737864459864795,
0.003734011435881257,
0.0006390604539774358,
0.0008780269417911768,
0.0006753142806701362,
0.0005913833738304675,
0.0006137739401310682,
0.0006683000246994197,
0.0007525373948737979,
0.0006583883077837527,
0.0006852646474726498,
0.0017182538285851479,
0.0025529733393341303,
0.000590665964409709,
0.000997339142486453,
0.0006036216509528458,
0.0009163962095044553,
0.0009391900966875255,
0.001216667820699513,
0.001185585861094296,
0.000796004431322217,
0.00117425003554672,
0.0007218815153464675,
0.0012900278670713305,
0.000559487147256732,
0.0013671480119228363,
0.0006242941599339247,
0.001202064217068255,
0.0007061922224238515
] | 0.00127 | 37 |
[
"Consolata Missionaries\n\nThe Consolata Missionaries are an Italian Roman Catholic missionary congregation for men and women.",
"\n\nHistory \nThe Instituto Missioni Della Consolata (hence the abbreviation I.M.C. and M.C. for its female congregation) was established on 29 January 1901 by the beatified Italian priest Giuseppe Allamano. ",
"For men, it is a Clerical Religious Congregation of Pontifical Right . ",
"Its headquarters are at Viale della Mura Aurelie 11-13, 00165 in Rome, Italy.",
"\n\nStatistics \nIn 2012, the male branch of the congregation had 218 houses and 1,000 members, including 758 priests. ",
"Sister Leonella Sgorbati, M.C., was beatified as a martyr in 2017.",
"\n\nSuperiors general \n Bishop Filippo Perlo (1926.02.16 – 1929.01.11)\n Domenico Fiorina (1949.09.01 – 1969)\n Mario Bianchi (1970 - 1981)\n Giuseppe Inverardi (1982 - 1993)\n Pietro Trabucco (1993 – 2005.05.10)\n Aquileo Fiorentini (2005.05.10 – 2011.06.08)\n Stefano Camerlengo (2011.06.08 – present)\nThe current Superior General of the Sisters of Consolata Missionaries is Simona Brambilla since 7 June 2011.",
"\n\nPrelates from their ranks \nDeceased (by year of death)\n 1930: Fr. ",
"Giuseppe Balbo, Apostolic Prefect of Meru (Kenya)\n 1933: Fr. ",
"Gaudenzio Barlassina, Apostolic Prefect of Kaffa (Ethiopia)\n 1935: Fr. ",
"Francesco Cagliero, Prefect Apostolic of Iringa (Tanzania)\n 1944: Bishop Giuseppe Perrachon, Apostolic Vicar emeritus of Nyeri (Kenya)\n 1948: Bishop Gabriele Perlo, Apostolic Vicar emeritus of Mogadishu (Somalia)\n 1948: Bishop Filippo Perlo, Apostolic Vicar emeritus of Kenya (Kenya) and Superior General emeritus of Consolata Missionaries\n 1953: Bishop Luigi Santa, Bishop of Rimini (Italy)\n 1960: Bishop Antonio Torasso, Apostolic Vicar of Florencia (Colombia)\n 1965: Bishop Attilio Beltramino, Bishop of Iringa (Tanzania)\n 1966: Bishop José Nepote-Fus, Bishop-prelate emeritus of Roraima (Brazil)\n 1976: Bishop Lawrence Victor Bessone, Bishop of Meru (Kenya)\n 1978: Bishop Carlo Re, Bishop emeritus of Ampurias e Tempio (Italy)\n 1990: Bishop Carlo Maria Cavallera, Bishop emeritus of Marsabit (Kenya)\n 2012: Bishop Angelo Cuniberti, Apostolic Vicar emeritus of Florencia (Colombia)\n 2014: Bishop José Luis Serna Alzate, Bishop emeritus of Líbano–Honda (Colombia)\n 2014: Bishop Servílio Conti, Bishop-Prelate emeritus of Roraima (Brazil)Alive\n Archbishop Luis Augusto Castro Quiroga, Metropolitan Archbishop of Tunja (Colombia) and President of the Episcopal Conference of Colombia\n Bishop Evaristo Marc Chengula, Bishop of Mbeya (Tanzania)\n Bishop Giovanni Crippa, Bishop of Estância (Brazil)\n Bishop Carillo Gritti, Bishop-Prelate of Itacoatiara (Brazil)\n Bishop Anthony Ireri Mukobo, Apostolic Vicar of Isiolo (Kenya) and Titular Bishop of Rusguniæ\n Bishop Peter Kihara Kariuki, Bishop of Marsabit (Kenya)\n Bishop Francisco Lerma Martínez, Bishop of Gurué (Mozambique)\n Bishop Aldo Mongiano, Bishop emeritus of Roraima (Brazil)\n Bishop Francisco Javier Múnera Correa, Apostolic Vicar of San Vicente del Caguán (Colombia) and Titular Bishop of Aquæ novæ in Numidia\n Bishop Virgilio Pante, Bishop of Maralal (Kenya)\n Bishop Joaquím Humberto Pinzón Güiza, Apostolic Vicar of Puerto Leguízamo–Solano (Colombia) and Titular Bishop of Otočac\n Bishop José Luís Gerardo Ponce de León, Bishop of Manzini (Swaziland).",
"\n Bishop Elio Rama, Bishop of Pinheiro (Brazil)\n Bishop Ambrogio Ravasi, Bishop emeritus of Marsabit (Kenya)\n Bishop Inácio Saure, Bishop of Tete (Mozambique)\n Bishop Walmir Alberto Valle, Bishop emeritus of Joaçaba (Brazil)\n Fr. ",
"Alberto Trevisiol, Rector Magnificus of the Pontifical Urbaniana University\n\nSource and external links \n I.M.C. website, mainly in Italian\n GigaCatholic, with many biographical links\n\nReferences\n\nCategory:Catholic orders and societies"
] |
{
"pile_set_name": "Wikipedia (en)"
}
|
[
0.002231588354334235,
0.0006976194563321769,
0.0007458853651769459,
0.00062905641971156,
0.0007069810526445508,
0.000963327125646174,
0.000664174382109195,
0.0007234176155179739,
0.0006059619481675327,
0.0006657886551693082,
0.0008386787958443165,
0.0007486808462999761,
0.0005913300556130707
] | 0.000832 | 13 |
[
"The French Parliament is considering a legislative provision that would ban strong encryption by requiring tech companies to configure their systems so that police and intelligence agencies could always access their data.",
"\n\nThe amendment to the vast “Digital Republic” bill was introduced in the French National Assembly, parliament’s lower house, by eighteen politicians from the conservative Republican Party."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.000582201755605638,
0.0006819053669460118
] | 0.000632 | 2 |
[
"Le parquet d'Evry (Essonne) a requis 4 mois de prison avec sursis contre Richard Trinquier, le maire (Debout la France) de Wissous jugé pour avoir menacé des gens du voyage avec un katana et un revolver à la ceinture. ",
"La décision sera rendue le 21 novembre.",
"\n\nLe 8 avril dernier, le maire de Wissous (Essonne) a menacé des gens des voyages (qui s’étaient installés sur le parking d’une crèche de sa commune) avec un katana qu’il a brandi et avec lequel il a fait des moulinets. ",
"Richard Trinquier explique avoir agi \"par urgence\" estimant qu’il était \"de son devoir d’intervenir\". \"",
"Je représente la nation, j’ai pris cet engagement et je suis dans mes fonctions\". ",
"Le procureur de la République lui rappelle qu’un maire \"n’est pas habilité à intervenir en flagrance\". ",
"Un maire, précise le Procureur \"a en effet la qualité d’OPJ mais pas l’habilitation\". ",
"Il a donc requis 4 mois de prisons avec sursis à son encontre.",
"\n\nUne plaque de shérif\n\nSelon les termes de M. Cabut, Procureur d'Evry, \"M. Trinquier a donné à la fonction de maire, un côté pitrerie et guignolesque, qui donne un côté un peu Kill Bill, en plus, avec une plaque de shérif\". ",
"Sur ce point, les policiers qui sont intervenus ce jour là ont en effet constaté une plaque en forme de shérif sur la veste du maire où il était écrit : \"Maire, OPJ\". ",
"Il dira à ces deux policiers qu’\"avec ces gens là, c’est comme ça qu’il faut faire !\". ",
"Le tribunal de son côté, lui signifie qu’on \"imagine mal la maire de Paris se balader avec un sabre dans la capitale\".",
"\n\nLe maire explique avoir agit à cause de la défaillance de l'Etat\n\n\"Ça peut vous paraître inhabituel pour un maire de sortir un katana car il faut une formation mais moi je l’ai eu la formation après mon passage chez les parachutistes\". ",
"A ce sujet, le Procureur lui rappelle qu’\"un maire n’est pas un ancien para, même si vous avez peut être la nostalgie de cette période passée chez les para\".",
"Richard Trinquier dit avoir agit à la place de la police nationale qui n’a pas voulu se déplacer. ",
"Face aux installations illégales de gens du voyage explique M. Trinquier, \"l’action du Préfet,c’est le néant et on laisse les maires agir seul\". ",
"Malgré ces déclaration, il a été établi que la BAC s’est tout de même rendu sur place ce dimanche avant l’intervention du maire.",
"\n\nAlcool et Glock 45\n\nCe jour là, M Trinquier explique avoir bu un apéritif et deux verres de vin. ",
"Appelé par son adjoint, il a donc décider d’exclure cette communauté de gens du voyage. ",
"Plusieurs membres de cette communauté ont décrit des cris, des pleurs, nombre d’entre eux ont été effrayés par le comportement du maire. ",
"Les policiers qui ont interpellé le maire de cette commune de l’Essonne ont également constaté la présence d’un pistolet, Glock 45 qu’il avait à la ceinture. ",
"Le maire explique pour sa part que son arme était sous le siège de sa voiture.",
"\n\nMenacé par 6 hommes dont un avec un fusil à pompe\n\nRichard Trinquier explique à la barre avoir été menacé par un homme qui le visait avec un fusil à pompe. ",
"Il évoque également la présence de six personnes armées. ",
"Selon l'enquête, seul un homme a sorti une arme. ",
"Le procureur a également requis 4 mois de prison avec sursis pour un des membres de la communauté du voyage qui a sorti un fusil de chasse, cassé (qui n’était pas enclenché)."
] |
{
"pile_set_name": "OpenWebText2"
}
|
[
0.0011837657075375319,
0.0010058345505967736,
0.004485089797526598,
0.0013756519183516502,
0.0011407530400902033,
0.0019001701148226857,
0.004518238827586174,
0.0316057913005352,
0.008195747621357441,
0.0022000125609338284,
0.44237950444221497,
0.000718838709872216,
0.08723848313093185,
0.03723869472742081,
0.019301049411296844,
0.0019423465710133314,
0.019387202337384224,
0.001492135925218463,
0.02768177166581154,
0.00877627544105053,
0.0795878991484642,
0.001573594519868493,
0.009798471815884113,
0.00534382089972496,
0.007104680873453617,
0.010387690737843513
] | 0.031445 | 26 |
[
"One of the best features of No Man’s Sky’s Beyond update is the ability to tame and ride animals. ",
"It might not be the most useful, but it’s still just the best.",
"\n\nTurning a space crab into a tame and rideable mount takes a bit of work, though. ",
"Each animal requires a specific advanced bait that can only be made in a Nutrient Processor. ",
"This guide will walk you through all of the steps to get you mounted up.",
"\n\nTable of contents\n\nFind 10 Salvaged Data\n\nBefore you can make your Nutrient Processor, you’ll need to buy the blueprints. ",
"And to buy the blueprint, you’ll need 10 Salvaged Data.",
"\n\nThe easiest and fastest way to collect Salvaged Data is in Buried Technology Modules. ",
"You’ll find these on just about every planet (they’ve been on every planet we've visited, but it’s a big universe). ",
"They’ll show up in your Scanner and in your Analysis Visor as a two-bar Wi-Fi signal-like icon.",
"\n\nMake your way to them, and dig them up with your Terrain Manipulator. ",
"You’ll collect at least one Salvaged Data (usually more) for each.",
"\n\nNutrient Processor blueprint\n\nOnce you have 10 Salvaged Data, you can go buy the blueprint for the Nutrient Processor. ",
"You cannot get the blueprint from your base’s Construction Research Unit, though. ",
"This blueprint is only found on the Anomaly. ",
"Head into space and call in the new social hub from your menu.",
"\n\nWhen you land, head to the left side. ",
"All the way in the back, past the Teleporter room, you’ll find a room with several vendors. ",
"On the left side, between the Ship and Exosuit upgrades, you’ll find the Construction Research Station. ",
"Tab over to the Technology Modules tree. ",
"The Nutrient Processor is on the right side.",
"\n\nConstruct a Nutrient Processor\n\nYou’ll need two Metal Plating (50 Ferrite Dust each), one Hermetic Seal (30 Condensed Carbon), and 25 Sodium for your Nutrient Processor. ",
"Once you’ve collected everything, the Nutrient Processor is in your Construction menu under Tech > Portable > Nutrient Processor.",
"\n\nFigure out which Advanced Bait you need\n\nEach animal has a preferred bait that will tame it in No Man’s Sky. ",
"You can find this in your Analysis Visor, right at the top when you’re looking at a creature. ",
"You also have a list of all Creature Bait in your Utilities menu.",
"\n\nMake Advanced Bait\n\nTo make Advanced Bait, you’ll need some Mordite or some Faecium. ",
"You’ll get Mordite from certain plants or when you kill wildlife (you monster). ",
"Faecium is, well, poop. ",
"Find an animal and feed it Creature Pellets (10 Carbon) to make it drop Faecium.",
"\n\nYou’ll also need some local vegetables. ",
"Each planet will have planetary flora that shows up in your scanner with a wheat icon. ",
"They’ll have names like Sweet Root or Frozen Tubers. ",
"These are the things you’ll use on each planet to make Advanced Bait.",
"\n\nTake your Mordite or Faecium and your local flora to your Nutrient Processor. ",
"It might take some experimentation to get the Advanced Bait you’re after.",
"\n\nRide ‘em, Space Cowboy\n\nNow that you’ve got your new friend’s favorite food, head out and find them again. ",
"You’ll probably need some regular Creature Bait to get them to hang around, and then you can drop the right Advanced Bait. ",
"This will tame your animal friends. ",
"Once you have their attention, approach them. ",
"You’ll get the option to collect an animal byproduct like milk or to mount up.",
"\n\nHappy trails!"
] |
{
"pile_set_name": "OpenWebText2"
}
|
[
0.0011433657491579652,
0.0006189201376400888,
0.0006928565562702715,
0.0007246077875606716,
0.0006842633010819554,
0.0006255905609577894,
0.0006158280302770436,
0.0005509660113602877,
0.0006307422299869359,
0.002055829158052802,
0.02346973866224289,
0.0005682824994437397,
0.0005644672783091664,
0.0006139558972790837,
0.000575900252442807,
0.0013100940268486738,
0.0325668565928936,
0.0014916316140443087,
0.0006517373258247972,
0.0006138160824775696,
0.0006624457892030478,
0.0005893329507671297,
0.0005689478130079806,
0.0021948006469756365,
0.0010545839322730899,
0.003927044104784727,
0.2546609342098236,
0.8154216408729553,
0.40465787053108215,
0.3274715840816498,
0.0014739324105903506,
0.0006944162887521088,
0.001407626667059958,
0.001601419411599636,
0.06771350651979446,
0.001654545427300036,
0.008976642973721027,
0.019368181005120277,
0.027334537357091904,
0.0011156495893374085,
0.02877647615969181,
0.0008781892829574645
] | 0.048642 | 42 |
[
"MWP is owned and operated by FRS Global Marketing . ",
"We are dedicated to manufacturing quality handmade military planes (model aeroplane), civilian model airplanes and commercial airline model aeroplanes and etc. ",
"These model planes are crafted out of premium Mahogany wood with precise detail. ",
"We specialize in model airplanes for award ceremonies, permanent change of station (PCS) or retirements. ",
"MWP can provide the model planes of your choice. ",
"We can customize Air Force model planes, Army model plane, Navy model planes, Marines model planes and commerical Airliner model planes. ",
"This includes model airplanes, handmade helicopter models, model boats and much more. ",
"If you can not find what you are looking for, just send a picture, we can make a wooden model airplane display to suit your needs. ",
"See how they are made."
] |
{
"pile_set_name": "OpenWebText2"
}
|
[
0.0007784115732647479,
0.0006088028312660754,
0.0005907236482016742,
0.0005194657715037465,
0.0005631863605231047,
0.0007105091353878379,
0.0006065998459234834,
0.0007301501464098692,
0.0006069098599255085
] | 0.000635 | 9 |
[
"HP's last Woodstock model. ",
"A very cute little unit! ",
"Like all\nthe Woodstocks\nvery small and handy. ",
"Compared to the HP-25C it is much more powerful\nand\novercomes all of the former's programming limitations: Twice as much\nprogram\nmemory, subroutine calls, numeric labels, instruction insertion and\ndeletion\nand more. ",
"Most importantly it fixes the HP-25C's number termination bug\n-\nor was that a feature?",
"\nThe only thing that didn't change compared to the HP-25C is execution\nspeed.",
"\nWhen compared with Classic models it falls in the category of the HP-67\nbut\nlacks the card reader. ",
"Compared to Voyager models it's feature\nset comes close to the HP-11C.\nFeature wise the unit is identical to the HP-19C."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0007305674371309578,
0.0007955164182931185,
0.0009386909659951925,
0.000717294926289469,
0.0007784990593791008,
0.0006693992763757706,
0.0006635578465647995,
0.0006086818175390363
] | 0.000738 | 8 |
[
"Surgery backlog crisis looming\n\nDoctors and healthcare leaders are worried about Canada’s growing backlog of urgent surgeries as non-emergency procedures have been dramatically scaled back or delayed due to COVID-19 for almost two months now.",
"\n\nSince early March, Sunnybrook Hospital in Toronto has reduced the number of operating rooms available for scheduled surgeries from about 25 to four to free up human resources, medications, equipment, and beds for patients with COVID-19. ",
"In the last week, the hospital made available two more operating rooms to keep up with the rising number of patients who need surgery urgently.",
"\n\n“We’ve really been banging the drum about how much danger these patients are under if the slowdown continues at this rate,” says Dr. Shady Ashamalla, a surgical oncologist at Sunnybrook. “",
"The administration has been listening,” he adds.",
"\n\nWith less operating space available for scheduled procedures, Sunnybrook has prioritized patients at risk of worse outcomes or death if they don’t get surgery in the next two weeks, Ashamalla says. ",
"The hospital will soon expand that window to four weeks.",
"\n\nPatients who can wait longer are being monitored more closely than usual to ensure their conditions don’t worsen. “",
"Every morning, we review our list of patients we believe to be in that two-week window,” Ashamalla says. ",
"Patients flagged for urgent surgery typically receive it within 48 hours.",
"\n\nAshamalla believes his team has “kept everyone safe” so far, but some patients will suffer worse outcomes if provinces move too slowly to resume elective procedures.",
"\n\nDr. Andrew Krahn, a cardiologist at St. Paul’s Hospital in Vancouver, says the slowdown of elective procedures in British Columbia has led to an uptick in “more complex procedures with more risk” as patients’ conditions have worsened. ",
"And delays will continue even after surgeries resume as usual. “",
"There are probably about 15,000 to 20,000 people in Canada who are on a waitlist for cardiac surgeries normally and adding two months [of delayed cases] is going to make that number a fair bit larger,” he explains.",
"\n\nWhile a patient’s prognosis is the main consideration in prioritizing surgeries, there are many other factors to weigh, Krahn says. ",
"For example, it’s easier to fit in patients who won’t require lengthy hospital stays compared to those undergoing cardiac surgery who may be in the hospital for a week and need intensive care beds, some of which hospitals are still reserving in case of a surge in COVID-19 cases. ",
"Decisions are also being coordinated across jurisdictions. “",
"In one day, I have the same conversation at all three levels: locally, regionally and provincially,” Krahn says.",
"\n\nAccording to Dr. Danielle Martin, executive vice president and chief medical executive at Women’s College Hospital in Toronto, one “unintended but welcome consequence” of COVID-19 is the way hospital leaders are collaborating regionally to make the most of limited resources. ",
"While Women’s College typically only does outpatient surgeries, the hospital has taken on procedures for complex patients requiring overnight stays. ",
"It has also made operating space available for teams from other hospitals.",
"\n\nMartin says prioritization decisions are becoming more fraught as the slowdown continues. ",
"In the early days, surgery to remove a recently discovered melanoma cancer could wait, but now those cases are getting to the point where delaying them further is risky.",
"\n\nIn a recent CMAJ commentary, Martin and Women’s College Hospital Surgeon-in-Chief Dr. David Urbach warned that returning to “business as usual” will not be acceptable. ",
"They called for a switch to a team-based approach, where patients see the first available surgeon in a team responsible for their care, rather than waiting to see specific providers.",
"\n\nMartin adds that many research bodies and organizations, including the Wait Time Alliance and Cancer Care Ontario, are advising hospitals on how to “work down the backlog as quickly and as equitably as possible.”",
"\n\nHow quickly will depend on whether Canada sees a drop in hospitalization rates for COVID-19. ",
"So far, that hasn’t happened, says Dr. Joshua Tepper, president and chief executive officer of North York General Hospital in Toronto. ",
"Patients undergoing surgery and those hospitalized with COVID-19 draw on the same resources, including ventilators, intensive care beds, and drugs like propofol, which is now in short supply globally. ",
"Many health professionals needed in operating rooms, such as anesthesiologists, have been called to COVID-19 wards. ",
"And shortages of protective equipment persist across health systems.",
"\n\n“It’s easy to forget how integrated our healthcare system is,” says Tepper. “",
"Nobody operates without gloves and nobody operates without gowns, and that supply chain is under stress.”"
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0005734635633416474,
0.0005856046918779612,
0.0005650788079947233,
0.0007137803477235138,
0.0006817322573624551,
0.0008321265922859311,
0.0006207440746948123,
0.0005823707906529307,
0.0005756504833698273,
0.0006111565162427723,
0.0006654184544458985,
0.0006319634267129004,
0.0006479834555648267,
0.0005682971677742898,
0.000746501493267715,
0.0006298117223195732,
0.0005700648762285709,
0.0006094745476730168,
0.0005895685753785074,
0.0008185819606296718,
0.000621513114310801,
0.0007081978837959468,
0.001352368970401585,
0.0006202334188856184,
0.0006045022746548057,
0.0006026650662533939,
0.0007117860368452966,
0.0005799073842354119,
0.0008196781855076551,
0.0005989447236061096,
0.0005818767240270972,
0.0010781694436445832,
0.0008152451482601464
] | 0.000682 | 33 |
[
"Where did the idea for Elara come from?",
"\n\nDan Grinberg: I had previously built and sold a foodservice glove company and saw an opportunity to start a new one. ",
"This time around I wanted to blend my industry experience with my longstanding involvement with hunger relief and the Island Harvest Food Bank. ",
"This was around 2010, and I was noticing that a growing number of companies were embedding a social mission in their business models. ",
"TOMS Shoes is a popular example. ",
"TOMS gives a pair of shoes to a child in need for every pair someone buys. ",
"Along those lines, we launched Elara in 2011 with hunger relief in our DNA. ",
"Our signature program is called One Case • One Meal. ",
"For every case purchased by a customer we help provide a meal for a person struggling with hunger. ",
"I believe that we are going to see a huge increase in Social Enterprise business models and wanted to be on the forefront of it.",
"\n\nHow have the needs of that distributor and end user evolved?",
"\n\nI started in this industry in 1990 when the use of gloves in foodservice barely existed. ",
"Today gloves use is required by food codes and corporate policies around the country. ",
"What we see evolving is that many end-users are looking for new glove options to save time and money and to increase productivity. ",
"In response we have developed some neat new products. ",
"For example, new materials that last longer so there is less breakage and waste. ",
"This saves money. ",
"At the same time, distributors are looking for ways to help their customers improve their sales and profits. ",
"The more progressive distributors are partnering with like-minded vendors who can offer new products and ideas to their end user accounts.",
"\n\nFrom branding standpoint, operators and distributors have to stand out from their ever-increasing competition while also keeping up with changing consumer preferences and expectations. ",
"Great food and products are essential, but differentiation has to extend to the overall experiences a customer has with your company. ",
"How you greet a guest, how a DSR interacts with a chef, how you connect with the community, how you use social media- all these things impact a customer’s decision to support your establishment or the one next door.",
"\n\nDo those customer agendas include being good neighbors and citizens? ",
"Green and sustainable etc.?",
"\n\nNot only is that important, it’s where the consumer is going. ",
"Studies show that a large and growing number of consumers are more likely to support a company that is involved in corporate social responsibility. ",
"These consumers also expect companies to be involved in solving critical social and environmental issues. ",
"This, in turn, creates more trust and loyalty to such companies. ",
"There is an ROI to social responsibility if done right.",
"\n\nHow did Elara get involved with fighting hunger?",
"\n\nElara was launched with hunger relief embedded in the business model from day one. ",
"Our gloves are used while meals are being prepared; yet some 49 million Americans are struggling to get their next meal. ",
"So hunger relief is a natural fit for us. ",
"I also have experience on the food bank side as a volunteer and board member of Island Harvest, so we want to support the great work food banks do in their communities.",
"\n\nHow does that embrace Elara’s commitment to the industry and community it serves?",
"\n\nThe better we are at serving our customers with great products and service, the more we will grow. ",
"Growth, in turn, will enable Elara to donate an increasing number of meals. ",
"This is a great example of how social enterprise can work. ",
"Companies can do well while doing good.",
"\n\nWhy Island Harvest?",
"\n\nWe are based on Long Island and Island Harvest is the Island’s largest food bank, providing food and services for 300,000 people struggling with hunger. ",
"The organization has a proactive culture with strong leadership and a committed staff. ",
"Island Harvest consistently ranks high on Charity Navigator as well. ",
"Even though I am on their board, we would still designate Island Harvest as our local beneficiary due to its reach and reputation. ",
"I should mention that City Harvest and about twenty other food banks around the country are beneficiaries of the One Case One • Meal program as well. ",
"We donate to food banks designated by our customers in their local markets.",
"\n\nWhat led to the partnership with DiCarlo? ",
"What are the goals for the partnership?",
"\n\nIt started with an alignment of values. ",
"Community involvement and hunger relief are important to the DiCarlo family and the entire DiCarlo organization. ",
"DiCarlo donates food, helps Island Harvest with food purchasing and provides Island Harvest with use of its facility, among other things. ",
"When we presented the One Case • One Meal concept to Vinnie DiCarlo, he saw right away the positive impact it could have and was the first to champion the program in the Metro New York area. ",
"DiCarlo has since helped contribute tens of thousands of meals. ",
"We are also working together to raise awareness of the hunger issue out in the market so more folks can get involved.",
"\n\nHow does the case donation program work?",
"\n\nIt’s very simple. ",
"We add up the number of cases we sell. ",
"We then contribute to designated food banks to help provide them with the same number of meals. ",
"The food banks, in turn, distribute meals to individuals in need through food pantries, meal programs and other agencies.",
"\n\nWhat are the goals for the program?",
"\n\nWe are growing our business locally and expanding nationally. ",
"This means we can feed more people who are food insecure. ",
"What’s incredible is that more than 49 million Americans are in this situation. ",
"Children, senior, single parent households, veterans, the underemployed. ",
"You would be surprised by how pervasive hunger is. ",
"So our goal is to assist as many as possible by partnering with our customers, supporting local food banks and raising awareness so that others can take up the cause. ",
"There is certainly power in numbers!",
"\n\nHow can restaurateurs and Foodservice operators get involved?",
"\n\nThe good news is that many in the restaurant industry are already involved in the fight against hunger- through food donations, special events, conducting food drives and volunteering. ",
"As an example, Shake Shack has an annual “Great American Shake Sale” to benefit No Kid Hungry. ",
"There are many creative things you can do. ",
"The nice thing about One Case • One Meal is that just by purchasing gloves you already need you are getting involved in the cause! ",
"Elara can also provide graphics and POS materials so that your customers see that you are involved. ",
"According to the studies, your customers are looking for this.",
"\n\nTotal Food Service is a monthly B2B foodservice publication and website covering foodservice and hospitality news, industry trends and exclusive interviews.",
"\n\nABOUT US\n\nSince 1990, Total Food Service has been serving the Metro New York foodservice industry, including restaurant operators, chefs, dealers, consultants, suppliers and distributors. ",
"Each month, we deliver the latest news, insights, and interviews focused exclusively for the world’s largest food service and hospitality capital of the word, Metro New York.",
"\n\nFrom exclusive interviews to the latest news on products, trends, associations, and events – Total Food Service covers all aspects of the Metro New York food service industry. ",
"Whether it’s our monthly print or our digital issue, to our website and social media channels like Twitter and Facebook, we deliver a package of must read news and information to build your business."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0007303251186385751,
0.0006981012993492186,
0.0005623723845928907,
0.0005368244019336998,
0.0007205801084637642,
0.05360502377152443,
0.0005473720375448465,
0.00056527677224949,
0.0011342338984832168,
0.0005155573016963899,
0.0007203785935416818,
0.0005939574912190437,
0.0005786769324913621,
0.0005745470989495516,
0.0005975363310426474,
0.012535962276160717,
0.0006725091952830553,
0.0005482191918417811,
0.0005430725286714733,
0.0005747683462686837,
0.0005171940429136157,
0.0006609370466321707,
0.0006164361257106066,
0.000603964610490948,
0.0006315356586128473,
0.0005282182828523219,
0.0005782080115750432,
0.0005471704644151032,
0.0006412956863641739,
0.0013217830564826727,
0.0005558946868404746,
0.0015981262549757957,
0.0008356179459951818,
0.000517999695148319,
0.0005633688997477293,
0.0006476306007243693,
0.0005950229824520648,
0.0006088814698159695,
0.0005566570907831192,
0.0006980457110330462,
0.0009447702905163169,
0.0005367079866118729,
0.000564184389077127,
0.0006028561037965119,
0.000515662191901356,
0.0008403422543779016,
0.000798730761744082,
0.0008100591949187219,
0.0005583210149779916,
0.0006152267451398075,
0.0005850217421539128,
0.0005302577628754079,
0.0005961207789368927,
0.0005598629941232502,
0.0007043734076432884,
0.0006670255097560585,
0.0007253155927173793,
0.0005215398850850761,
0.0006251872982829809,
0.0006756680086255074,
0.0005980714340694249,
0.17061421275138855,
0.0017536141676828265,
0.001184533815830946,
0.03940103203058243,
0.000558445870410651,
0.0008486401056870818,
0.0009330696775577962,
0.0005939569091424346,
0.002446308033540845,
0.0006342898705042899,
0.0006291775498539209,
0.0005746599053964019,
0.0005715166917070746,
0.0007529401336796582,
0.0005709604592993855,
0.0005186083726584911,
0.0005821554805152118,
0.0006374086369760334
] | 0.004162 | 79 |
[
"270 F.2d 891\nSECURITIES AND EXCHANGE COMMISSION, Plaintiff-Appellee,v.FARM AND HOME AGENCY, INC., ",
"an Indiana corporation, Philip H. Meade, Chester R. Koza, E. Bernie Shelton and W. Harold Hilbert, Defendants-Appellants.",
"\nNo. ",
"12602.",
"\nUnited States Court of Appeals Seventh Circuit.",
"\nOctober 7, 1959.",
"\nRehearing Denied November 10, 1959.",
"\n\nLuis Kutner, Chicago, Ill., Joseph M. Shannon, Indianapolis, Ind. (John M. Bowlus, Chicago, Ill., of counsel), for appellants.",
"\nThomas G. Meeker, Gen. Counsel, David Ferber, Asst. ",
"Gen. Counsel, Theodore Zimmerman, Atty., ",
"S.E.C., Alexandria, Va., Thomas B. Hart, Regional Adm'r, John I. Mayer, Asst. ",
"Regional Adm'r, John J. Enright, Atty., ",
"S.E.C., Chicago, Ill., for appellee.",
"\nBefore MAJOR, SCHNACKENBERG and CASTLE, Circuit Judges.",
"\nSCHNACKENBERG, Circuit Judge.",
"\n\n\n1\nOn April 16, 1958, a final judgment enjoining Farm and Home Agency, Inc., an Indiana corporation, Philip H. Meade, Chester R. Koza, E. Bernie Shelton and W. Harold Hilbert, herein referred to as defendants, from violating sections 5(a) and 5(c) of the Securities Act of 1933, 15 U.S.C.A. § 77e(a) and (c), was entered by consent in the district court. ",
"The judgment was based upon a complaint filed by the Securities and Exchange Commission, plaintiff, alleging that defendants had engaged in acts and practices which constituted violations of said sections, which proscribe using the mails and means and instruments of transportation and communication in interstate commerce of unregistered securities, and which complaint sought to enjoin those alleged violations. ",
"The securities involved were the common capital stock of Farm and Home Agency, Inc.\n\n\n2\nOn November 24, 1958, defendants asked leave to file a motion to vacate the consent judgment. ",
"That motion, with an accompanying brief, appears in the record.",
"\n\n\n3\nOn December 1, 1958, defendants filed a motion for a restraining order seeking to enjoin the Securities and Exchange Commission or persons in active concert with it, from investigating, prosecuting, or presenting any evidence to a grand jury pertaining to \"alleged violations\" of section 5 of the Securities Act \"or any other matters or things set forth in the consent decree of April 16, 1958.\"",
"\n\n\n4\nAfter consideration of briefs and oral argument, the district court on January 12, 1959, entered the following order, from which this appeal was taken:\n\n\n5\n\"The defendants' motion for restraining order and motion for leave to file instanter the motion to vacate and set aside the decree of permanent injunction entered April 16, 1958 and in support thereof, file their brief herewith, being submitted for the consideration of the Court, and the Court being duly advised, now denies each of said motions.\"",
"\n\n\n6\nRelying upon rule 60(b) of the Federal Rules of Civil Procedure, 28 U.S.C.A., Rule 60(b), defendants contend that the judgment of April 16, 1958 was void in that the district court had no jurisdiction over either the parties to the judgment or the subject matter. ",
"They also charge that the decree was entered by virtue of a mistake of law and fact by the parties to the action \"because the defendants adopted the legal advice of their counsel which coincided with the jurisdictional assertions of the plaintiff\", and because \"the defendants and each of them were incorrectly advised as to their legal rights in the premises\"; and \"that all of the defendants were and are engaged in the business of insurance and were, therefore, under the supervision of the Department of Insurance of Indiana\".",
"\n\n\n7\nIt is apparent that, as the time to appeal has expired, defendants are asking us to review a decree enjoining them from violating the Securities Act, to the entry of which they fully consented while represented by counsel. § ",
"22(a) of the Securities Act1 grants to district courts of the United States jurisdiction of \"offenses and violations\" under the Securities Act, 15 U.S.C.A. § 77v(a). ",
"There is, therefore, no question that the court below had jurisdiction to enter the consent judgment. ",
"We cannot review alleged error in this proceeding. ",
"No appeal was taken. ",
"Numerous cases support the view that a motion to vacate a judgment under Rule 60(b) is addressed to the sound legal discretion of the court and that its determination will not be disturbed upon appeal except for an abuse of discretion. ",
"Elgin Nat. ",
"Watch Co. v. Barrett, 5 Cir., ",
"213 F.2d 776, 780; Independence Lead Mines Co. v. Kingsbury, 9 Cir., ",
"175 F.2d 983, 988; Western Union Telegraph Co. v. Dismang, 10 Cir., ",
"106 F.2d 362, 364; Bush v. Bush, 61 App.",
"D.C. 357, 63 F.2d 134, 135; and Jones v. Jones, 7 Cir., ",
"217 F.2d 239, 241.",
"\n\n\n8\nThe relief sought by the restraining order is ancillary to defendants' motion to vacate the consent decree, and, therefore, the right to such restraining order falls when we affirm the district court's order denying defendants' motion to file a motion to vacate the consent decree.",
"\n\n\n9\nFor the reasons herein set forth, the order of the district court is affirmed.",
"\n\n\n10\nAffirmed.",
"\n\n\n\nNotes:\n\n\n1\n 15 U.S.C.A. § 77v(a)\n\n\n"
] |
{
"pile_set_name": "FreeLaw"
}
|
[
0.0007084831013344228,
0.000610369723290205,
0.0013785817427560687,
0.0009461178560741246,
0.000894161406904459,
0.0007552821771241724,
0.0006981781334616244,
0.0006484363111667335,
0.009743615053594112,
0.0006265706033445895,
0.004542707931250334,
0.0008786525577306747,
0.0007349313818849623,
0.0007956018089316785,
0.0009109118836931884,
0.0006542874616570771,
0.0006645676330663264,
0.0006154633592814207,
0.0006072118412703276,
0.0006948396912775934,
0.00058214325690642,
0.0006770289619453251,
0.0006406863103620708,
0.0005733291036449373,
0.0007026242674328387,
0.0006447745836339891,
0.0006505731143988669,
0.000792524020653218,
0.0006619598716497421,
0.000866593501996249,
0.0008137174299918115,
0.0008117375546135008,
0.0009217923507094383,
0.0010067976545542479,
0.0008844785625115037,
0.0010353658581152558,
0.0007099013892002404,
0.0006295329658314586,
0.0008257554145529866,
0.000738701201044023
] | 0.001082 | 40 |
[
"\n74 F.Supp.2d 1345 (1999)\nThe UNITED STATES, Plaintiff\nv.\nJoseph ALMANY, d/b/a. J.A. Imports, David Jordan, Inc., and Far West Insurance Company, Defendants.",
"\nSlip Op. ",
"99-120.",
"\nCourt No. ",
"96-02-00384.",
"\nUnited States Court of International Trade.",
"\nNovember 3, 1999.",
"\n*1346 David W. Ogden, Acting Assistant Attorney General, David M. Cohen, Director, Commercial Litigation Branch, Civil Division, U.S. Department of Justice, Washington, DC (A. David Lafer, Franklin E. White, Jr., Washington, DC), for plaintiff.",
"\nJoseph Almany, defendant, pro se.",
"\nLaw Offices of David K. Geren (David K. Geren, Encino, CA), for defendant David Jordan, Inc.\n\nMEMORANDUM\nMUSGRAVE, Judge.",
"\nFamiliarity with the prior proceedings of this case is presumed. ",
"Defendants Joseph Almany, d/b/a J.A. Imports and David Jordan, Inc. have judgment by default entered against them for lost duties in the amount of $5,016.87, plus interest, payable to the United States Customs Service, resulting from a violation of 19 U.S.C. § 1592. ",
"See Slip Op. ",
"98-72 (June 3, 1998). ",
"The United States has moved for clarification or modification of the Judgment to encompass fraudulent culpability against defendants, arguing that certain requests for admissions were deemed \"conclusively established\" by reason of defendants' failure(s) to respond thereto and this Court's Order of January 13, 1997. ",
"Relevant portions of those requests read as follows:\nRequest for Admission No. ",
"2:\nDuring the years 1989 through 1991, Joseph Almany was the owner, president, and sole proprietor of J.A. Imports, and directed the activities of that business.",
"\nRequest for Admission No. ",
"3:\nDuring the year 1991, J.A. Imports imported and distributed merchandise provided to it by Harbour City Trading Company.",
"\nRequest for Admission No. ",
"4:\nDuring the years 1989 and 1991, J.A. Imports imported and distributed merchandise provided to it by First Union Enterprises Company.",
"\nRequest for Admission No. ",
"5:\nDuring the years 1989 and 1991, J.A. Imports imported and distributed merchandise provided to it by the Hongsinga Watch Co. Ltd.\nRequest for Admission No. ",
"6:\nWith respect to ... [twelve] Customs entries ... (identified in Exhibit A to plaintiff's complaint) Hongsinga Watch Co. Ltd. provided J.A. Imports with two different invoices for the merchandise entered by J.A. Imports.",
"\nRequest for Admission No. ",
"7:\nWith respect to each of the entries identified in Request for Admission No. ",
"6 above, one of the two invoices Hongsinga Watch Co. Ltd. provided J.A. Imports stated a purchase price for the merchandise entered by J.A. Imports which was lower than the actual purchase price paid by J.A. Imports for that merchandise.",
"\nRequest for Admission No. ",
"8:\nWith respect to each of the entries identified in Request for Admission No. ",
"6 above, one of the two invoices Hongsinga Watch Co. Ltd. provided J.A. Imports stated a purchase price for the merchandise entered by J.A. Imports which reflected the actual purchase price of the merchandise and which was higher than the purchase price stated on the other invoice.",
"\nRequest for Admission No. ",
"9:\nWith respect to each of the entries identified in Request for Admission No. ",
"6 above, J.A. Imports used the invoice provided by Hongsinga Watch Co. Ltd. which reflected the lower purchase price rather than invoice provided by Hongsinga Watch Co. Ltd. which reflected the higher purchase price as the basis *1347 for the declaration of the value of the merchandise to the U.S. Customs Service.",
"\nRequest for Admission No. ",
"10:\nWith respect to each of the entries identified in Request for Admission No. ",
"6 above, Joseph Almany and J.A. Imports knew that, by reporting to Customs the lower price listed in one of the two invoices provided by Hongsinga Watch Co. Ltd., J.A. Imports would pay a lower amount in duties than it would have to pay if Joseph Almany and J.A. Imports had declared to Customs the price listed in the higher priced invoice provided by Hongsinga Watch Co. Ltd. with respect to the same entry.",
"\nRequest for Admission No. ",
"11:\nWith respect to each of the entries identified in Request for Admission No. ",
"6 above, by relying upon the lower priced invoice provided by Hongsinga Watch Co. Ltd. as the basis of the declaration of the value of the merchandise to Customs, J.A. Imports paid lower duties than the amount required by law.",
"\nRequest for Admission No. ",
"12:\nWith respect to each of the entries identified in Request for Admission No. ",
"6 above, J.A. Imports reported to Customs a value for the merchandise covered by the entry that was lower than the actual purchase price and did so with an intent to deceive.",
"\n...\nRequest for Admission No. ",
"27:\nDavid Jordan, Inc. is the same entity formerly known as J.A. Imports.",
"\nRequest for Admission No. ",
"28:\nAdmit the genuineness of Exhibit 1 ... as a true and correct copy of the RIDER [to Customs' Form 301] NAME CHANGE OF PRINCIPAL executed on April 17, 1993, on behalf of J.A. Imports [and naming David Jordan, Inc. as principal \"responsible for any act secured by this bond under principal's former name.\"] ...",
"\nPlaintiff's Motion for a Court Order Stating that Plaintiff's First Set of Requests for Admissions to Defendants Joseph Almany d/b/a J.A. Imports and David Jordan Inc. Are Admitted Pursuant to Rule 36(a), Exhibit 1 (and \"Exhibit 1\" attached thereto) (emphasis in original). ",
"Requests for Admission 13-19 and 20-26 followed the pattern established in 6-12 with respect to five entries supported by invoices purportedly prepared by First Union Enterprises Company and six entries supported by invoices purportedly prepared by Harbour City Trading Company, respectively.",
"\nJoseph Almany's \"response\" to the government's motion states merely:\nActing on my own behalf, I hereby respond to [chamber's] letter dated April 22, 1999 and I am willing to pay the sum of $5,016.87 in order to satisfy the judgment entered on June 6, 1998 in regards to case or court # XX-XX-XXXXX United States of America vs. Joseph Almany, J.A. Imports and Far West Insurance Company.",
"\nBe it my understanding and condition, however, that no other amount such as interest will be added to said $5,016.87. ",
"As you are probably aware, the case has been dragging on for almost 10 years. ",
"This payment would also close this matter once and for all and will therefore release me and/or any other defendant of any further liability relying [sic] to this case.",
"\nAn individual may choose representation pro se, however a corporation must be represented by an attorney admitted to the bar of the court. ",
"CIT Rules 75(a),(b)(1). ",
"There has been no response to date from counsel for David Jordan, Inc., although a response was ordered and ample time allowed. ",
"See Slip Op. ",
"99-77 (Aug. 9, 1999); cf. ",
"Slip Op. ",
"98-66 (May 19, 1998).",
"\nThe Court anticipated the government's reply: that the amount of duties Mr. Almany states he is now \"willing\" to pay, without interest, is less than the amount he has been ordered to pay, that the customs duties owed remain unpaid, and that the civil penalty of $413,138.00 imposed by the U.S. Customs Service for fraudulent *1348 violations of 19 U.S.C. § 1592 has not been paid.",
"\nThis Court ought not to countenance Mr. Almany's audacity in suggesting that the Government settle the entire case against all defendants (including the lawful penalty of $413,138.00 for fraudulent violations of 19 U.S.C. § 1592 sought in the complaint) merely to receive a payment that this Court ordered Mr. Almany to pay more than a year ago. ",
"Rather, in accordance with its June 3, 1998 Order, this Court should again order Mr. Almany to pay the customs duties owed, plus interest, by a check payable to the \"United States Treasury\" and mailed to counsel for plaintiff. ",
"Further, this Court should hold that the defendants Joseph Almany d/b/a J.A. Imports and David Jordan, Inc. are jointly and severally liable for payment of a civil penalty, resulting from their fraudulent violations of 19 U.S.C. § 1592, in an amount to be determined in further proceedings.",
"\nPlaintiff's Reply at 3 (emphasis in original).",
"\nSlip Op. ",
"99-77 countenanced only due administration of justice. ",
"Mr. Almany's response provides no basis for denying the government's motion. ",
"Judgment will enter accordingly.",
"\n\nJUDGMENT\nUpon consideration of the motion of The United States of America, plaintiff, for clarification or modification of the Court's opinion and order dated June 3, 1998, and upon all other papers and proceedings herein, it is\nORDERED that the motion of The United States of America, plaintiff, is granted, and it is further\nORDERED, ADJUDGED AND DECREED that defendants Joseph Almany, d/b/a J.A. Imports and David Jordan, Inc. are jointly and severally liable to The United States of America, plaintiff, for the sum of US$5,016.87, representing the amount of duties lost by reason of defendants violations of 19 U.S.C. § 1592, plus interest, and it is further\nORDERED that, in accordance with 19 U.S.C. § 1505(c) (1994), 19 C.F.R. §§ 24.3(e) and 24.3a, and Travenol Labs., ",
"Inc. v. U.S., 118 F.3d 749 (Fed.",
"Cir.1997), interest on the duties owing and unpaid shall accrue and be assessed monthly as of each monthiversary date on the basis of a 360-day year of 12 monthly periods from, and including, March 22, 1995, the date on which the U.S. Customs Service demanded payment of duties owing from defendant Joseph Almany, to, but excluding, the monthly period within which payment for the full amount owing is made, in accordance with interest rates published by the U.S. Customs Service for overdue accounts, to wit:\n\n\nBeginning Ending Under-payment\n Date Date Percentage (per annum)\n03/22/95 03/31/95 9%\n04/01/95 06/30/95 10%\n07/01/95 03/31/96 9%\n04/01/96 06/30/96 8%\n07/01/96 03/31/98 9%\n04/01/98 12/31/98 8%\n01/01/99 03/31/99 7%\n04/01/99 09/30/99 8%\n\nand accruing at such rates for overdue accounts as the Secretary of the Treasury shall continue to publish from time to time, and it is further\nORDERED that payment of amounts owed by defendants shall be by bank or cashier's or certified check in United States Dollars acceptable for deposit by a Federal Reserve bank or other designated depositary (or by other arrangement which counsel for plaintiff may, in its sole discretion, find acceptable) made payable to \"United States Treasury\" and received by certified, first class mail addressed to either A. David Lafer, Senior Trial Counsel or Franklin E. White, Jr., Trial Attorney, in either case at Commercial Litigation Branch, Civil Division, Department of Justice, 1100 L Street, N.W., Room 11026, Washington, D.C., 20530, U.S.A., telephone (202) 307-6289; and it is further\nORDERED, ADJUDGED AND DECREED that the conduct of defendants *1349 Joseph Almany, d/b/a J.A. Imports and David Jordan, Inc., as described in the complaint of The United States of America, plaintiff, and in Plaintiff's First Request for Admissions and as deemed conclusively established by Order dated January 13, 1999, constitutes fraudulent violations of 19 U.S.C. § 1592, and it is further\nORDERED that defendants Joseph Almany, d/b/a J.A. Imports and David Jordan, Inc. are jointly and severally liable to The United States of America, plaintiff, for a civil penalty by reason of fraudulent violations of 19 U.S.C. § 1592, and it is further\nORDERED that the parties propose scheduling, jointly or separately, for determining the amount of the civil penalty.",
"\n"
] |
{
"pile_set_name": "FreeLaw"
}
|
[
0.0007833812851458788,
0.002210701582953334,
0.00089447753271088,
0.0007400541217066348,
0.0012918297434225678,
0.0007957401685416698,
0.0007268143235705793,
0.0006479568546637893,
0.0007743598544038832,
0.0006444145110435784,
0.0005541719147004187,
0.0006727803265675902,
0.0009612592402845621,
0.0006603889632970095,
0.0006294061313383281,
0.0005592392408289015,
0.0005941063282079995,
0.0005987804033793509,
0.0005660037859342992,
0.0005987804033793509,
0.0005793954478576779,
0.0005987804033793509,
0.0005802355008199811,
0.0006093151168897748,
0.0005987804033793509,
0.00059019576292485,
0.000676951021887362,
0.0005987804033793509,
0.0005924685392528772,
0.0006442696321755648,
0.0005987804033793509,
0.0005911996122449636,
0.0006384446751326323,
0.0005987804033793509,
0.0005906443111598492,
0.0006253732717595994,
0.0005987804033793509,
0.0005922407144680619,
0.0006916476413607597,
0.0005987804033793509,
0.0005895703216083348,
0.0007740131113678217,
0.0006491053500212729,
0.000671404181048274,
0.0005987804033793509,
0.0005912035121582448,
0.0006111151305958629,
0.0005643056356348097,
0.000606596062425524,
0.0006961413309909403,
0.0005788087728433311,
0.0009485440095886588,
0.0006271047168411314,
0.0006943910848349333,
0.0005580034339800477,
0.0009612592402845621,
0.0006747336592525244,
0.002210701582953334,
0.0006659345817752182,
0.000685632461681962,
0.0018132448894903064,
0.0006319688400253654,
0.0007141552632674575,
0.0006125482032075524,
0.002210701582953334,
0.0008667833753861487,
0.0007079464849084616,
0.0006022194866091013,
0.0006824011215940118,
0.00111823296174407,
0.0007264396990649402,
0.001995444530621171
] | 0.000774 | 72 |
[
"Mapping the History of Our School: District\n83\nPrimary and Secondary Sources\n\nNote - These links are to resources\nrelated to various schools, but may not be specific to your school.",
"\nMaterials such as these are commonly kept for Nebraska communities.",
"\nThese are the types of materials to look for pertaining to your\nschool district.",
"\n\nBooks\nThe NSHS Library/Archives has several books on the history of\nschools. ",
"Although this is not a circulating collection, these\nbooks can be viewed at the State Historical Society and may be\navailable through your local historical society or library.",
"\n\n50 Years of Sheridan: A History of Sheridan School, 1926-1976\n\nAlder Grove: Community, School, Church, 1867-1992\n\nBloomington Public School: The History of the Institution\nand Its People, 1871-1991\n\nHistory of Pilger High School, Pilger Nebraska\n\nPages of History: Nebraska High Schools, Present and Past,\nPublic and Private, 1854-1994\n\nPublic School District Reorganization and Consolidation\nin Adams County, Nebraska, 1949-1989"
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0005395695916377008,
0.0005396928172558546,
0.0005968459299765527,
0.0006059368024580181,
0.0005782797234132886,
0.0006191785214468837
] | 0.00058 | 6 |
[
"/*\n * Copyright 2015 Red Hat, Inc. and/or its affiliates.",
"\n *\n * Licensed under the Apache License, Version 2.0 (the \"License\");\n * you may not use this file except in compliance with the License.",
"\n * You may obtain a copy of the License at\n *\n * http://www.apache.org/licenses/LICENSE-2.0\n *\n * Unless required by applicable law or agreed to in writing, software\n * distributed under the License is distributed on an \"AS IS\" BASIS,\n * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.",
"\n * See the License for the specific language governing permissions and\n * limitations under the License.",
"\n */\n\npackage org.optaplanner.examples.investment.persistence;\n\nimport org.optaplanner.examples.common.persistence.",
"AbstractSolutionImporter;\nimport org.optaplanner.examples.common.persistence.",
"ImportDataFilesTest;\nimport org.optaplanner.examples.investment.app.",
"InvestmentApp;\nimport org.optaplanner.examples.investment.domain.",
"InvestmentSolution;\n\npublic class InvestmentImporterTest extends ImportDataFilesTest<InvestmentSolution> {\n\n @Override\n protected AbstractSolutionImporter<InvestmentSolution> createSolutionImporter() {\n return new InvestmentImporter();\n }\n\n @Override\n protected String getDataDirName() {\n return InvestmentApp.",
"DATA_DIR_NAME;\n }\n}\n"
] |
{
"pile_set_name": "Github"
}
|
[
0.0007464065565727651,
0.0005749124684371054,
0.0005534342490136623,
0.0005547429318539798,
0.0007904841331765056,
0.000839308719150722,
0.000806970929261297,
0.0007678135298192501,
0.001032635336741805,
0.0007584914565086365
] | 0.000743 | 10 |
[
"Advisory ID: DRUPAL-SA-CORE-2012-003\n\nProject: Drupal core\n\nVersion: 7.x\n\nDate: 2012-October-17\n\nSecurity risk: Highly critical\n\nExploitable from: Remote\n\nVulnerability: Information Disclosure, Arbitrary PHP code execution\n\nDescription\n\nMultiple vulnerabilities were discovered in Drupal core.",
"\n\nArbitrary PHP code execution\n\nA bug in the installer code was identified that allows an attacker to re-install Drupal using an external database server under certain transient conditions. ",
"This could allow the attacker to execute arbitrary PHP code on the original server.",
"\n\nThis vulnerability is mitigated by the fact that the re-installation can only be successful if the site's settings.php file or sites directories are writeable by or owned by the webserver user. ",
"Configuring the Drupal installation to be owned by a different user than the webserver user (and not to be writeable by the webserver user) is a recommended security best practice. ",
"However, in all cases the transient conditions expose information to an attacker who accesses install.php, and therefore this security update should be applied to all Drupal 7 sites.",
"\n\nCVE: CVE-2012-4553\n\nInformation disclosure - OpenID module\n\nFor sites using the core OpenID module, an information disclosure vulnerability was identified that allows an attacker to read files on the local filesystem by attempting to log in to the site using a malicious OpenID server.",
"\n\nCVE: CVE-2012-4554\n\nVersions affected\n\nDrupal core 7.x versions prior to 7.16.",
"\n\nDrupal 6 is not affected.",
"\n\nSolution\n\nInstall the latest version:\n\nIf you use Drupal 7.x, upgrade to Drupal core 7.16.",
"\n\nIf you are unable to deploy the security release immediately, removing or blocking access to install.php is a sufficient mitigation step for the arbitrary PHP code execution vulnerability.",
"\n\nAlso see the Drupal core project page.",
"\n\nReported by\n\nThe arbitrary PHP code execution vulnerability was reported by Heine Deelstra and Noam Rathaus working with Beyond Security's SecuriTeam Secure Disclosure Program. ",
"Heine Deelstra is also a member of the Drupal Security Team.",
"\n\nThe information disclosure vulnerability in the OpenID module was reported by Reginaldo Silva.",
"\n\nFixed by\n\nContact and More Information\n\nThe Drupal security team can be reached at security at drupal.org or via the contact form at http://drupal.org/contact.",
"\n\nLearn more about the Drupal Security team and their policies, writing secure code for Drupal, and securing your site."
] |
{
"pile_set_name": "OpenWebText2"
}
|
[
0.0006695147603750229,
0.0009336344664916396,
0.0011094427900388837,
0.0009535844437777996,
0.0007353271939791739,
0.0006966182263568044,
0.0007797642610967159,
0.0006900628795847297,
0.0007399501046165824,
0.0007642348646186292,
0.00085692189168185,
0.00062494573649019,
0.0007230237242765725,
0.000811051286291331,
0.0005676440196111798,
0.000603223976213485,
0.0006697180215269327
] | 0.000761 | 17 |
[
"Drug abuse among state hospital psychiatric inpatients with particular reference to PCP.",
"\nThe pattern of alcohol and drug abuse among 78 consecutive newly admitted patients to a middle Tennessee state hospital psychiatric ward was examined. ",
"There was a high prevalence of alcohol, marijuana, hypnotic sedative, and stimulant abuse among this patient population within 1 month prior to their hospitalization. ",
"However, only 1 patient gave a history of PCP abuse and none of the 76 plasma samples analyzed by the GCN and GC2-N methods was positive for PCP or its common analogues. ",
"This study does not support a high prevalence of PCP abuse among this group of psychiatric patients."
] |
{
"pile_set_name": "PubMed Abstracts"
}
|
[
0.0010308767668902874,
0.0007123604882508516,
0.001464066794142127,
0.000674278533551842,
0.001165973604656756
] | 0.00101 | 5 |
[
"To Celebrate 50 years of Star Trek I designed an illustration for Constellation's Star Trek 50th anniversary Art Exhibition & Auction. ",
"My illustration will many sold at the exhibition to go towards the British Heart Foundation."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0005732201389037073,
0.000583849148824811
] | 0.000579 | 2 |
[
"NEW CHICKS!!!!",
"\n\nHere's some pics of my new chicks...\nThese are supposed to be buff Orpingtons and Aracaunas???There were about 60 chicks together and the had dry feces and pee on them. ",
"Probably shouldn't have bought them but I felt bad.."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.01720779947936535,
0.4519317150115967,
0.0008654789999127388
] | 0.156668 | 3 |
[
"Two investors have fil ed cases in the high court asking for IMA Group’s assets to be seized and a CBI probe\n\nBENGALURU: Sleuths who entered the Jayanagar showroom of IMA Jewels — one of the entities of IMA Group being probed for its Ponzi operations — found 90% of the ornaments in the display units had disappeared.",
"\n\nThe Special Investigation Team suspects the IMG Group founder-owner Mohammed Mansoor Khan may have taken the jewellery with him when he fled the country on June 8. ",
"The remaining jewellery is being evaluated and could be worth Rs 20 crore, SIT sources said. ",
"The inspection was conducted on Monday.",
"\n\nSIT sleuths have searched the houses of two women, including a former wife of Khan, and seized gold ornaments, Rs 2 lakh in cash and a laptop used by him.",
"\n\nWith 1.25 lakh people having lost their money to Ponzi scams in Bengaluru in the past decade, new Bengaluru City police commissioner Alok Kumar said preventing such fraudulent schemes will be one of his priorities.",
"\n\nThe Karnataka government said it was considering setting up fast-track courts to try Ponzi cases, and revenue minister RV Deshpande told TOI the government was contemplating a new law in this regard.",
"\n\n"
] |
{
"pile_set_name": "OpenWebText2"
}
|
[
0.0010758506832644343,
0.0006696389173157513,
0.0005662970943376422,
0.0006124530336819589,
0.0018937465501949191,
0.004150527995079756,
0.0007109710713848472,
0.001995444530621171
] | 0.001459 | 8 |
[
"Physicians' perceptions of cancer care for elderly patients: a qualitative sociological study based on a pilot geriatric oncology program.",
"\nThe aim of this study was to document physicians' perceptions of cancer care for elderly patients within an oncogeriatric coordination pilot unit (UPCOG) created in Paris, France. ",
"We focused on how physicians apply new cancer care practices, how they establish new teamwork, and their experience of oncogeriatrics in everyday practice. ",
"Qualitative methods were used, including a literature review, observation of working sessions in the oncogeriatric pilot unit, and semi-structured interviews with 28 physicians. ",
"The results show how physicians' differing perceptions of geriatric oncology can hinder routine collaboration."
] |
{
"pile_set_name": "PubMed Abstracts"
}
|
[
0.0007853950373828411,
0.000608367845416069,
0.0005970181082375348,
0.0005789177375845611,
0.0009496232378296554
] | 0.000704 | 5 |
[
"Festival for Peace\n\nThe Festival for Peace was an all day concert event produced at Shea Stadium in Queens, NY on August 6, 1970.",
"\nIt was the second event of a series planned to raise funds for anti-war political candidates in the early 1970s. ",
"The first, the Winter Festival for Peace, took place in Madison Square Garden earlier in the year. ",
"The date selected for the Summer event was of particular interest as it was also the 25th anniversary of the U.S. first use of an atomic weapon in the bombing of Hiroshima, Japan on August 6, 1945 in World War II. ",
" The concert, advertised as the Summer Festival for Peace, was scheduled for 10:00 AM – 10:00 PM although several last-minute performers and extended sets added about two hours. ",
"Seating was General Admission by tier in the stadium.",
"\n\nVery little media has survived and no film of this concert has surfaced publicly despite the fact that it featured such historic performers as Janis Joplin, Paul Simon, Creedence Clearwater Revival, Steppenwolf, The James Gang, Miles Davis, Johnny Winter, Herbie Hancock, Dionne Warwick, John Sebastian, The Rascals, the cast of \"Hair,\" Pacific Gas & Electric, Ten Wheel Drive and a dozen other important acts of the period. ",
"The Summer Festival for Peace was the first major concert at Shea Stadium after the last performance of The Beatles in 1966. ",
"A wide selection of still photos shot by photographer Ken Davidoff are the only readily available visual documentation of the concert itself.",
"\n\nIt proved to be one of the last performances for Janis Joplin who died only two months later, as well as a reunion and last performance with her former band, Big Brother & the Holding Company. ",
"When the concert was first announced, Joplin was not scheduled to perform, but Big Brother was on the bill. ",
"She was in NYC to do two appearances on Dick Cavett's television show with her new band Full Tilt Boogie and decided to perform with her former band while at Shea. ",
"During the August 3rd appearance with Cavett, Joplin announced her intention to play at the Festival, spoke of the show and described the concert as being produced by Peter Yarrow (of the singing group Peter, Paul & Mary). ",
" During the concert, Joplin sang a duet of \"What the World Needs Now\" with Dionne Warwick.",
"\n\nOther sources confirm that Mr. Yarrow, by then a well-known peace advocate, together with Phil Friedmann (an Amherst graduate who worked in the campaign for the Democratic nomination of Senator Eugene McCarthy for President) produced the Summer Festival after their huge success of the Winter Festival for Peace at Madison Square Garden in February, 1970.",
"\n\nThe importance of these concerts were manifold. ",
"First, unlike the for-profit Woodstock Music & Arts Fair that became increasingly political as it unfolded, the Festivals for Peace were the first large venue U.S. events which were produced with the sole intention of fund raising for political, and specifically anti-war, purposes: not unusual later but not seen prior to 1970. ",
"Secondly, again in contrast to Woodstock where performers insisted on being paid, Peter Yarrow and Friedmann were able to convince the top acts of the day (including many that were paid at Woodstock like CCR, Hendrix and Joplin) to donate their time and performances to the Festival for Peace shows just months after Woodstock.",
"\n\nThis was the first time that the world's biggest rock, jazz, blues and folk performers came together and donated their performances to aid a specific social/political agenda. ",
"The Summer Festival for Peace was the first of many, more publicized benefit concerts in the future. ",
"As such it paved the way for The Concert for Bangladesh (August 1, 1971), Farm Aid (September 22, 1985), Live Aid, etc. ",
"by demonstrating the fundraising potential for such large scale musical events.",
"\n\nReferences\n\nCategory:Benefit concerts in the United States\nCategory:Rock festivals in the United States\nCategory:Shea Stadium"
] |
{
"pile_set_name": "Wikipedia (en)"
}
|
[
0.0006305138231255114,
0.0007305109756998718,
0.0005761008942499757,
0.000549852498807013,
0.0005656025605276227,
0.0006232273299247026,
0.0008233563276007771,
0.0006332841003313661,
0.0005859075463376939,
0.001858190749771893,
0.0007308661588467658,
0.0011781388893723488,
0.0006461010780185461,
0.0007529598660767078,
0.0005841461243107915,
0.0005838588695041835,
0.0006461780867539346,
0.000689188193064183,
0.0005870025488547981,
0.0006397917168214917,
0.0005568001070059836,
0.000586730195209384,
0.0005798774654977024
] | 0.00071 | 23 |
[
"About 805 results for \"wharton school\"\n\nPunit Goenka, MD & CEO, ZEEL delivered the keynote on Saturday at the 19th Wharton India Economic Forum at The Wharton School, University of Pennsylvania. ",
"Speaking on the theme of the conference 'India: Delivering the Dream', he addressed the changed business ...DNA, 1 week ago\n\nGUWAHATI, Dec 26 - The City Police last night arrested Terrence Wharton, principal of the Mount Hermon School, Darjeeling, in connection with the suicide of a class VIII student of the school from Guwahati, Debahuti Talukdar. ",
"Wharton was produced at the CJM ...\n\nMarch 03--Patrick T. Harker, the former Wharton School dean who is president of the University of Delaware, will leave that job July 1 to serve as the next president of the Federal Reserve Bank of Philadelphia. ",
"The move will give Harker a voice, on ...\n\nThe Wharton School is tapping into a combination of business school trends in its choice of speakers for its MBA graduation ceremony this year. ",
"Not only can Neil Blumenthal and Dave Gilboa showcase the promise of a new generation of leaders, ...\n\nA first of its kind study conducted by researchers at the University of Minnesota and the Wharton School, University of Pennsylvania has found that Republicans and Independents are more likely to disapprove of neuroscience-based legal reforms if the ...\n\nAdam Stein (left) and Tom Arnold (right) became friends at the Wharton School of the University of Pennsylvania and together co-founded TerraPass, a pioneering greenhouse gas management company. ",
"This business partnership continues today at Gridium, ...\n\nClaire Spinks, headteacher of Wharton Church of England Primary School, jetted off to Cyprus during the summer term Claire Spinks, headteacher of Wharton Church of England Primary School, went on holiday to Cyprus during term time Photo: AP\n\n( Poets&Quants ) — These days, you can't walk the hallways of a business school campus and not overhear excited conversations among MBA students about their business plans and startup ideas. ",
"Talk of using the MBA experience as a place to fine-tune ...\n\n(Bloomberg) -- BMW AG will test its self-proclaimed image as the maker of the ultimate driving machine with a wagon designed more for parents picking up kids at school than for zipping along the autobahn. ",
"The van-like 2-Series Gran Tourer offers ..."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0005734649603255093,
0.0006015635444782674,
0.000621447863522917,
0.0005794612225145102,
0.0006690395530313253,
0.0005710677360184491,
0.0005781832151114941,
0.0008641650783829391
] | 0.000632 | 8 |
[
"If you were not redirected by your browser, click"
] |
{
"pile_set_name": "OpenWebText2"
}
|
[
0.0015619306359440088
] | 0.001562 | 1 |
[
"Q:\n\nWhy is it allowed to reinterpret_cast integral, enumeration and pointer-to-member types to themselves?",
"\n\nIn this recent question, we saw that it's not allowed to reinterpret_cast some custom class type instance to itself; struct A{}; reinterpret_cast<A>(A{}); is invalid (it works only through references or pointers). ",
"Which seems to make sense, because of the lack of real-world scenarios where such identity conversion is necessary.",
"\nChecking the corresponding standard clause, we have in [expr.reinterpret.cast] (emphasis mine):\n\n1 [...] Conversions that can be performed explicitly using reinterpret_cast are listed below.",
"\n No other conversion can be performed explicitly using reinterpret_cast.",
"\n2 [...] An expression of integral, enumeration, pointer, or pointer-to-member type can be explicitly converted to its own type; such a cast yields the value of its operand.",
"\n\nSo reinterpret_cast<int>(42) is allowed, while the same cast with a struct A{} is not. ",
"Why?",
"\n\nA:\n\nIt was part of resolving DR 799. ",
"The issue was as follows:\n\nThe note in 8.2.10 [expr.reinterpret.cast] paragraph 2 says,\n\nSubject to the restrictions in this section, an expression may be cast\n to its own type using a reinterpret_cast operator. ",
"\n\nHowever, there is nothing in the normative text that permits this conversion, and\n paragraph 1 forbids any conversion not explicitly permitted.",
"\n\nThe idea in the note was deemed worthwhile, that reinterpret_cast should be allowed to do the identity conversion. ",
"So the normative text you ask about was added. ",
"I can assume the restriction to some fundamental types is a cautious first (and maybe even only) step. ",
"Since it doesn't open the can of worms associated with class types and the need to call their constructors. ",
"reinterpret_cast is all about not creating new objects, and one can do that with fundamental types. ",
"Not sure the same applies to class types.",
"\n\n"
] |
{
"pile_set_name": "StackExchange"
}
|
[
0.0007224727305583656,
0.00076880375854671,
0.0005909567698836327,
0.0006145812221802771,
0.000716913491487503,
0.0007033763104118407,
0.000862527231220156,
0.0009686383418738842,
0.000672588066663593,
0.0006402358412742615,
0.0007328790379688144,
0.000592501019127667,
0.000651493261102587,
0.0006147754029370844,
0.001318453811109066,
0.0008447103318758309,
0.0006810402846895158,
0.001995444530621171
] | 0.000816 | 18 |
[
"Aleksandr Sergeyevich Kovalyov\n\nAleksandr Sergeyevich Kovalyov (; born 21 February 1982) is a former Russian professional footballer.",
"\n\nClub career\nHe played 3 seasons in the Russian Football National League for FC Dynamo Bryansk, FC Spartak Nizhny Novgorod and FC SKA-Energiya Khabarovsk.",
"\n\nReferences\n\nExternal links\n \n\nCategory:1982 births\nCategory:Living people\nCategory:Russian footballers\nCategory:Association football defenders\nCategory:PFC CSKA Moscow players\nCategory:FC Oryol players\nCategory:FC Zhenis Astana players\nCategory:FC Dynamo Bryansk players\nCategory:FC SKA-Khabarovsk players\nCategory:FC Dynamo Barnaul players\nCategory:Kazakhstan Premier League players\nCategory:Russian expatriate footballers\nCategory:Expatriate footballers in Kazakhstan\nCategory:Russian expatriate sportspeople in Kazakhstan"
] |
{
"pile_set_name": "Wikipedia (en)"
}
|
[
0.000844488269649446,
0.001221001730300486,
0.0006551494589075446
] | 0.000907 | 3 |
[
"Atlantic City is open as follows: Monday to Thursday:.00am.00am Friday and Saturday:.00am.00am Sunday:.00am.00am Bad Angel Schillerstraße 11A, 80336 Munich Close to the Altstadt and the main railway station, Bad Angel is centrally situated in Munich. ",
"Gay and lesbian travelers should neither experience any trouble: Munich has a large and vivid gay and lesbian community and the Rosa Liste, a gay rights party, has been part of the city government since 1996. ",
"Beidseitig, 69, KB, Schmusen/Kuscheln, extra langes Vorspiel, Masturbation, Intimmassage, erotische Massage, FS, Fuß und Schuherotik, Duschspaß, Zungenküsse.uvm. ",
"Prices vary as much as the selection of ladies but expect to pay a minimum of 150 per hour with the average rate of an escort in Munich being closer to 250 per hour. ",
"You can watch transsexual live sex shows in Munich as long as you are connected to internet. ",
"Lass dich fallen in Meine Hände oder brauchst leichte Erziehung Erotische Art. ",
"The area directly around Munich Central Station ( Hauptbahnhof ) has numerous youth hostels, and upscale hotels like Le Meridien and Sofitel. ",
"There are currently 20 ladies working on their books within the Munich area, but they do offer the services of any of their escorts from across Europe; however, you must be prepared to pay for their travel and a minimum charge of 12 hours. ",
"Instead of the action centring around a sauna or spa, the main focus of the facilities is around the erotic movie cinema."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0006618472398258746,
0.0049417894333601,
0.46382299065589905,
0.0006973007693886757,
0.813312828540802,
0.10667750984430313,
0.0006152545684017241,
0.0005599238211289048,
0.0006358847604133189
] | 0.154658 | 9 |
[
"Q:\n\nAdd weight to randomisation\n\nI'm struggling to find any resources that help with this question:\nSay I want to generate a random number between 1 and 5, but wish the randomisation to be weighted toward the bottom end (like if generating a random number of children a couple might have, which is more likely to be 1 than 5). ",
"Where should I start? ",
"Would this involve creating a complex function to map that curve or is there a method available to help with this?",
"\nFor very simple sets I suppose I could create an array, so for the above example I could do, for example:\nvar children = [1,1,1,1,2,2,2,3,4,5];\nreturn children[Math.floor(Math.random() * children.length)];\n\nBut this lacks flexibility and assumes the data set is being created manually within the script.",
"\n\nA:\n\nThe \"children birth probability\" has Gaussian distribution, which is a \"continuous limit\" of binomial distribution. ",
"What about using binomial distribution?",
"\nvar val = Math.random()*5 + Math.random()*5;\n// \"val\" will be between 0 and 10, but values around 5 are more probable\nreturn 1 + Math.floor(Math.abs(val-5));\n\nJust put it into function and make \"5\" a parameter. ",
"\nSmall test (http://jsfiddle.net/yt6zvs8n/):\n100 output values: 1, 1, 2, 3, 2, 4, 2, 2, 2, 3, 3, 1, 2, 1, 4, 1, 2, 3, 2, 1, 4, 4, 2, 1, 4, 3, 2, 2, 1, 3, 2, 4, 1, 3, 2, 4, 2, 2, 1, 2, 1, 1, 1, 4, 1, 4, 1, 3, 1, 2, 1, 2, 1, 2, 2, 3, 2, 1, 3, 1, 2, 1, 3, 2, 1, 1, 1, 1, 2, 1, 1, 2, 1, 3, 4, 1, 3, 1, 2, 5, 1, 3, 1, 3, 1, 3, 1, 4, 1, 3, 1, 3, 2, 2, 3, 1, 3, 4, 1, 3\nOccurences (histogram): 38, 28, 21, 12, 1\n\nA:\n\nWell as long as you work with integers, where N is the max child value and N is not so big you will need O(n) space and time\nI'll answer assuming that you only use integers going from 0 to N, see in USAGE section of my answer why you don't need to specify them by hand when their weight is 0.",
"\nvar weights = [0,4,3,1,1,1];\n\nvar number = weights.reduce(function (previous, current, index) {\n var value = Math.random() * (current|0);\n if (value > previous[0]) {\n return [value, index];\n }\n return previous;\n}, [Math.random() * (weights[0]|0), 0])[0];\n\nExplanation\nUsing indices as integers, we have:\n\n0 with a weight of 0 (it will never appear)\n1 with a weight of 4\n2 with a weight of 3\n3 with a weight of 1\n4 with a weight of 1\n5 with a weight of 1\n\nUsage\nAs it is not very easy to think in term of range or if you don't want to specify all values until N manually, you can use an object and convert it to an array by specifying a key length equal to N + 1, like this\nvar N = 130, \n weights = Array.apply(null, {\n 12: 1,\n 53: 3,\n 130: 10,\n length: N + 1\n });\n\n"
] |
{
"pile_set_name": "StackExchange"
}
|
[
0.0006404726300388575,
0.0007921524811536074,
0.0006996595184318721,
0.0007028308464214206,
0.0008721394115127623,
0.000776617438532412,
0.001038281829096377,
0.0016164237167686224,
0.0008773858426138759
] | 0.000891 | 9 |
[
"Key workers in Cairngorms abused for staying in holiday homes Published duration 29 April Related Topics Coronavirus pandemic\n\nimage copyright Getty Images\n\nA national park has called for abuse aimed at people staying in holiday accommodation to stop.",
"\n\nCairngorms National Park Authority said some essential workers, including NHS staff and delivery drivers, were using the accommodation during the lockdown.",
"\n\nIt said this had resulted in a \"small number of incidents\" of abuse on social media.",
"\n\nAlmost all park accommodation is currently closed due to Covid-19 restrictions.",
"\n\nThe park authority said the abuse stemmed from people mistakenly believing the \"guests\" were flouting lockdown rules.",
"\n\nHowever, the accommodation is being used by key workers who would otherwise be homeless.",
"\n\n'Nip this in the bud'\n\nAn open letter has been posted, calling for \"understanding\" about why some holiday accommodation remained in use.",
"\n\nGrant Moir, chief executive of the park authority, told BBC Scotland: \"We want to make sure we nip this in the bud and focus on the great work by communities across the park.",
"\n\n\"In Badenoch and Strathspey alone about 900 people have volunteered to help with deliveries, and we are seeing volunteering in Tomintoul and Deeside.\"",
"\n\nThe letter, jointly signed by Cairngorms Business Partnership chief executive, Mark Tate said: \"A face we don't recognise is not necessarily a stranger and our plea is not to publicly criticise and jump to any judgement.\"",
"\n\nPark and business leaders said concerns about people flouting restrictions should still reported by calling police on the telephone number 101."
] |
{
"pile_set_name": "OpenWebText2"
}
|
[
0.0007368658552877605,
0.0007974404725246131,
0.0007130950107239187,
0.0006627103430218995,
0.0010622810805216432,
0.001000325195491314,
0.0005738014588132501,
0.0005687953671440482,
0.0006020068540237844,
0.0006446515908464789,
0.0006298312218859792
] | 0.000727 | 11 |
[
"Sandeep,\n\nThis is Dr. Ueneohara. ",
"The spellchecker has done it for me. ",
"Sorry.",
"\n\nPlease, make sure that I am involved in the meeting. ",
"Also,\ninclude Bob Lee, Rakesh, Vasant, Zimin, Krishna, Paulo, Alex,\nTanya, Rabi.",
"\n\nVince\n\n\n\n -----Original Message-----\nFrom: \tKohli, Sandeep \nSent:\tThursday, July 19, 2001 8:37 AM\nTo:\tKaminski, Vince J; Brady, Mark\nCc:\t'vkaminski@aol.com'\nSubject:\tRE: Boltzmann Model Demonstration from Toshiba in Tokyo\n\nMark,\n\nI will be in Houston over the period mentioned (July 31st to August 4th), and will be glad to meet with Dr. O'Hara. ",
" My extension at the office is: 713-853-5188. ",
" My mobile number is 713-857-6826.",
"\n\nRegards,\nSandeep.",
"\n\n -----Original Message-----\nFrom: \tKaminski, Vince J \nSent:\tThursday, July 19, 2001 8:23 AM\nTo:\tBrady, Mark\nCc:\tKohli, Sandeep; 'vkaminski@aol.com'\nSubject:\tRE: Boltzmann Model Demonstration from Toshiba in Tokyo\n\nMark,\n\nThe members of my group attend INFORMS conferences\nand I shall check today if we are covering this specific\nmeeting.",
"\n\nWe shall be glad to speak with Dr. O'Hara when he comes to Houston.",
"\nI shall be gone on Friday, but you can contact Sandeep Kohli in Houston to discuss the details.",
"\nAny date in the time interval given by you works for us.",
"\n\nVince\n\n -----Original Message-----\nFrom: \tBrady, Mark \nSent:\tThursday, July 19, 2001 3:47 AM\nTo:\tvkamins@enron.com\nCc:\tSangster, Barry; Sasaki, Tsuyoshi; Crenshaw, Shirley\nSubject:\tBoltzmann Model Demonstration from Toshiba in Tokyo\nImportance:\tHigh\n\nVince,\n\nEnron Japan received a presentation today from a nuclear physicist working for Toshiba who has developed an application of the Boltzmann model to finance. ",
" We were introduced to this group inside Toshiba through the Carlyle Group's Japan office. ",
" Dr. Uenohara, from Toshiba, is travelling to New York City to speak at the 11th INFORMS Applied Probability Society Conference. ",
" The URL is: http://www.cap.columbia.edu/informs-aps/ap-2001.html\n\nDr. Uenohara is interested in applying this model to power markets and would very much like to speak with someone from Enron during his trip to the US. ",
" Is anyone from Enron going to be attending the conference? ",
" If there is someone going, it would be worthwhile to facilitate a meeting. ",
" If not, Dr. Uehara has stated he would travel to Houston between the dates of July 31 and August 4 to explain his model and the possibility of working with Enron. ",
" Toshiba is currently working with Daiwa here in Japan, however they told us in the meeting that Daiwa was way too slow and they were looking for other companies to work with in the application of the model to different types of markets. ",
" Please give me the name of a suitable contact in the event that we are able to arrange a visit for Dr. Uenohara to Houston.",
"\n\nToshiba will be sending me the presentations we received today and I will forward them to you as soon as I receive them. ",
"\n\nRegards,\nMark Brady\n\nMark Brady\nEnron Japan Corp.\nOtemachi 1st Square Bldg., ",
"West 11th Floor\n1-5-1 Otemachi, Chiyoda-ku\nTokyo 100-0004\nTel: 03.5219.4579 Fax: 03.5219.4510\nwww.enron.co.jp"
] |
{
"pile_set_name": "Enron Emails"
}
|
[
0.0008183811441995203,
0.006584938149899244,
0.000904906599316746,
0.0005165701149962842,
0.000697038252837956,
0.0005711052799597383,
0.0006847839104011655,
0.000722577387932688,
0.0009486023336648941,
0.0005509914481081069,
0.0006760518881492317,
0.0005958035471849144,
0.0005698378081433475,
0.0006520983879454434,
0.0005366562982089818,
0.0005927853635512292,
0.0005239417660050094,
0.0006819138070568442,
0.0005211415700614452,
0.0005268792156130075,
0.0005816934281028807,
0.0005341898067854345,
0.0005295226583257318,
0.000792428501881659,
0.0010717628756538033
] | 0.000895 | 25 |
[
"Salad Cutter Bowl,Vegetable Cutter Bowl\n\nCLEAN & SAFE: Cutting board free, and it is no need for you to clean to much containers anymore. ",
"Just cutting all kind of fruit or vegetable directly in the cutter bowl. ",
"And no need to worried about cutting your finger anymore.",
"\n\nPRACTICAL & MULTIFUNCTION: The bowl can be used as a strainer, with which can wash your lettuce together, the pedestal can be used as a cut board, just sets them up for a perfect Cross cutting. ",
"Spin the strainer and slice your fruit or vegetable smaller.",
"\n\nFAST & EASY: Cutting all Vegetable in a bowl, finish your salad quickly. ",
"It is easier and more convenient for you to make healthy salad everyday.",
"\n\nHIGH QUALITY KITCHENWARE: Each product is made using high quality ABS plastic. ",
"Safe, durable and FDA and LFGB approved.",
"\n\nGREAT HELPER FOR HEALTHY LIFE: The salad container matches with a variety of foods, keeping nutrition balanced. ",
"It is also the great helper for a family to live a healthy life.",
"\n\nSubscribe\n\nSign up to get the latest on sales, new releases and more …\n\nYou Can Trust Us!",
"\n\nWe've been in business for over 30 years. ",
"All of our products come with a 100% Satisfaction Guarantee. ",
"You have a full 60 days to give all our As Seen On TV products a try. ",
"If you're not completely happy, please return the product and you'll be fully refunded (less shipping & handling costs), no questions asked!"
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.000668513064738363,
0.0007215812802314758,
0.8584503531455994,
0.03360212594270706,
0.00221400149166584,
0.0008310239645652473,
0.000554275989998132,
0.0006481125601567328,
0.0005474414210766554,
0.0005670216050930321,
0.0007323609315790236,
0.0006691563175991178,
0.0005885498831048608,
0.0006356142112053931,
0.0015520923770964146,
0.000645406253170222
] | 0.056477 | 16 |
[
"Peter Davies, PhD\n\nIN THE NEWS\n\nSizzling Summer Science at New York Genome Center’s Five Points Lectures\n\nWednesday mornings this summer, the New York Genome Center (NYGC) will host a superb line-up of Five Points Lectures, bringing outstanding scientists from near and far to discuss fresh and diverse genomic research, in technical...\n\nGreg Hannon Joins the New York Genome Center as Associate Core Member and Director of Cancer Genomics\n\nThe New York Genome Center (NYGC) welcomes Dr. Greg Hannon as Associate Core Member and Director of Cancer Genomics.",
"\nAt NYGC Dr. Hannon will facilitate studies of cancer genomes, and run a research group focused on...\n\nExciting June Lineup of Speakers at New York Genome Center Five Points Lecture series\n\nIn June, the New York Genome Center (NYGC) welcomes four superb speakers for their Five Points Lectures, technically detailed weekly talks that bring scientists from New York and beyond to discuss their work, in depth,...\n\nSee The New York Genome Center in our news and media archives VIEW MORE"
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0006006179610267282,
0.0005519289406947792
] | 0.000576 | 2 |
[
"Mamas Expo September 22-23: Sponsors Highlights\n\nGet ready for the Mamas Network’s Mamas Expo on September 22-23 at Museum of the Moving Image in Queens for parents of pre-K to pre-teens. ",
"Parents will receive an exclusive look at the newest resources and products. ",
"They’ll also gain insider access to tons of vendors, hear parenting experts speak, view special screenings of family-friendly films and explore exhibitions. ",
"We’ve highlighted some of the many Mama’s Expo sponsors lined up for the big event.",
"\n\nClickandImprove.com - Click and Improve delivers a new hassle-free, worry-free way to purchase home improvement and repair services online. ",
"www.clickandimprove.com\n\nDolphin Organics – Dolphin Organics is a line of natural and organic personal care products for kids including skincare for preteens. ",
"There’s nothing artificial, and even its labels are very clear and honest with its ingredients. ",
"www.dolphinorganics.com\n\nEmerald Brand – Emerald brand is the industry leader in eco-friendly and green consumables. ",
"It offers alternative to non-green products in five product categories including break-room, facilities, washroom, jan-san, home and kitchen. ",
"www.emeraldecoproducts.com\n\nPolkadot What- Polkadot What lets girls customize their very own leggings by picking and choosing designs and styles in a simple two-step process. ",
"www.polkadotwhat.com\n\nQueens Zoo / Central Park Zoo / Prospect Park Zoo - Queens Zoo and its sister zoos are proud sponsors of the Mamas Expo. ",
"View animals from North and South America including eagles, bison, sea lions, pumas, Andean bears and more all year long. ",
"www.queenszoo.com\n\nStay tuned for highlights of more vendors and sponsors over the next few weeks!",
"\n\nMamas Expo\nSaturday and Sunday, September 22-23, 2012 from 10 am – 4 pm\nMuseum of the Moving Image\n36-01 35th Avenue\nAstoria, NY 11106Note: The Mamas Expo is free with the paid museum admission at the Museum of the Moving Image."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0010626961011439562,
0.0005587174091488123,
0.00053377979202196,
0.0005459634121507406,
0.0011001424863934517,
0.003873692126944661,
0.0005790325230918825,
0.0006627255352213979,
0.0006359084509313107,
0.0010019381297752261,
0.0008103467989712954,
0.0006985216168686748,
0.0005469819298014045,
0.0006164921796880662
] | 0.000945 | 14 |
[
"Vegan Gear by Hill Killer All orders of our new Vegan Flag gear qualify for a free Team Vegan water bottle (while supplies last). ",
"Read our Vegan Athlete blog series\n\nSort by Featured Best Selling Alphabetically, A-Z Alphabetically, Z-A Price, low to high Price, high to low Date, new to old Date, old to new"
] |
{
"pile_set_name": "OpenWebText2"
}
|
[
0.0009180313209071755,
0.001145711401477456
] | 0.001032 | 2 |
[
"[Sclerothalamectomy (STE): stable postoperative intraocular pressure regulation in primary open-angle and pseudoexfoliative glaucoma].",
"\nSince the 70th years we operated the hemorrhagic glaucoma by a ciliary body exposure after Benedikt. ",
"In a few patients we noted a stable intraocular pressure regulation for years, often in these cases where there is no filtering bleb. ",
"This signifies, that the ciliary body is heavily involved in the resorption of the anterior chamber fluid. ",
"During the following next years we further developed this operating technique and we use it today successfully in the operative treatment of primary open angle- and pseudoexfoliative glaucoma. ",
"Since march 1996 we performed a sclerothalamectomy in 46 eyes. ",
"In approximate the half of cases the operation was done combined with a cataract surgery. ",
"27 eyes had a primary open angle and 19 eyes a pseudoexfoliative glaucoma with a mean preoperative intra-ocular pressure of 29.79 +/- 7.96 mm Hg respectively of 33.58 +/- 9.32 mm Hg. ",
"The mean follow-up was in the case of primary open angle glaucoma 18.3 +/- 8.9 (median 20) and in the case of pseudoexfoliative glaucoma 15.8 +/- 11 (median 16) months. ",
"The mean postoperative intra-ocular pressure, which range in the group of primary open angle glaucoma 13.9 +/- 1.6 mm Hg and in the group of pseudoexfoliative glaucoma 12.9 +/- 2.5 mm Hg, was significant lower with significant less postoperative medication compared with the preoperative values during the complete follow-up (p < 0.01). ",
"There was no evidence in a significant difference between the both groups regarding the postoperative complication rate and intra-ocular pressure. ",
"We observed altogether in 7 eyes a flat filtering bleb, 5/27 in the primary open angle glaucoma and 2/19 in the pseudoexfoliative glaucoma group. ",
"In all of other cases we didn't state a filtering bleb. ",
"The sclerothalamectomy leeds to a long-term stable intra-ocular pressure with a contemporary low complication rate in the case of primary open angle- and pseudoexfoliative glaucoma."
] |
{
"pile_set_name": "PubMed Abstracts"
}
|
[
0.0013410367537289858,
0.008823031559586525,
0.0005569285131059587,
0.0007628817693330348,
0.0016029411926865578,
0.0019361177692189813,
0.0013434879947453737,
0.006036497186869383,
0.0009853917872533202,
0.0008985333261080086,
0.000763194402679801,
0.0025187309365719557,
0.0006155393784865737,
0.0017440932570025325
] | 0.002138 | 14 |
[
"Q:\n\nCommutations relations of C,P,T transformations with Lorentz group\n\nAlmost any QFT textbook discusses the C,P,T symmetry operators which are charge conjugation, parity transformation, time reversal respectively. ",
"\n\nI failed so far to find any discussion of the commutation relations of these operators with the Lorentz group (or its generators). ",
"I would be happy to have a reference. ",
"\n\nA:\n\nWeinberg vol 1 Eqs (2.6.7) to (2.6.12) for $P$ and $T$.\n\n"
] |
{
"pile_set_name": "StackExchange"
}
|
[
0.0007242750725708902,
0.0007147003198042512,
0.0005232255789451301,
0.013517770916223526
] | 0.00387 | 4 |
[
"Gérard Barray\n\nGérard Barray (born 2 November 1931 in Toulouse) is a French comedian and film and television actor.",
"\n\nEarly life and education\nBarray's parents split up quickly and his mother, who came from Montauban decided to return to her hometown with her little boy. ",
"Around the age of 15, he discovered a passion for jazz; he participated in a few shows in nightclubs while pursuing his studies and obtained a bachelor's degree at the Faculty of Toulouse.",
"\nCamille Ricard, an actress and teacher at the Conservatory of Toulouse, who advised him to go to Paris with a letter of recommendation for a friend, Noel Roquevert. ",
"Barray enrolled at the Cours Simon, a drama school in Paris. ",
"Four years later, Gérard Barray won the Jury Prize.",
"\n\nCareer\nIt will then excel in the roles of knights with a big heart. ",
"He starred as D'Artagnan, Pardaillan, Surcouf and Scaramouche. ",
"In total there practice gender in a dozen feature films, most of which are box-office success, widely known abroad. ",
"Besides films swashbuckling as Pardaillan and Scaramouche and adventure films like Surcouf, Barray turned police commissioner in two San Antonio movies. ",
"In 1969, he starred beside young actress Claude Jade in \"The Witness\". ",
"He played Van Britten, a mysterious museum curator who seduces a young English teacher. ",
"It was his last major role.",
"\n\nFor Claude Berri he played in (1970) as Richard, a super star and rather temperamental actor. ",
"His comeback in 1997 was in Alejandro Amenabar's \"Abre los ojos\" as Devernois, a TV man.",
"\n\nGérard Barray was appointed an Officer in the Order of Arts and Letters in January 2010.",
"\n\nSelected filmography\n\nReferences\n\nExternal links\n\nCategory:1931 births\nCategory:Living people\nCategory:French male film actors\nCategory:French male television actors"
] |
{
"pile_set_name": "Wikipedia (en)"
}
|
[
0.0013615833595395088,
0.0020541823469102383,
0.0006668586283922195,
0.0006526376819238067,
0.0007120902882888913,
0.0006897308048792183,
0.0005792927695438266,
0.0008914332720451057,
0.0007796423742547631,
0.0007851222180761397,
0.0006222468218766153,
0.001249757595360279,
0.001441239146515727,
0.002004475798457861,
0.000783243274781853,
0.0006232252926565707,
0.0007617336814291775
] | 0.00098 | 17 |
[
"//\n// Generated by class-dump 3.5 (64 bit).",
"\n//\n// class-dump is Copyright (C) 1997-1998, 2000-2001, 2004-2013 by Steve Nygard.",
"\n//\n\n#import \"ABMainListOutlineView.h\"\n\n@interface CNContactListView : ABMainListOutlineView\n{\n}\n\n- (void)awakeFromNib;\n\n@end\n\n"
] |
{
"pile_set_name": "Github"
}
|
[
0.0053914207965135574,
0.00948100071400404,
0.0007386716897599399
] | 0.005204 | 3 |
[
"\n202 Or. ",
"456 (1954)\n274 P.2d 798\nSTATE OF OREGON\nv.\nCAPUTO\nSupreme Court of Oregon.",
"\nArgued September 8, 1954.",
"\nAffirmed October 6, 1954.",
"\nPetition for rehearing denied November 3, 1954.",
"\n*458 David W. Dardano and Leo Levenson, of Portland, argued the cause for appellant. ",
"With them on the brief was Larry Landgraver, of Portland.",
"\nJames J. Kennedy, Deputy District Attorney for Multnomah County argued the cause for respondent. ",
"With him on the brief were John B. McCourt, District Attorney for Multnomah County, and Robert R. Carney and Horace B. Fenton, Deputy District Attorneys for Multnomah County.",
"\nBefore LATOURETTE, Chief Justice, and WARNER, ROSSMAN, LUSK, BRAND and PERRY, Justices.",
"\nAFFIRMED.",
"\nBRAND, J.\nOn 27 February 1953 an indictment was filed which reads as follows:\n\"LARRY CAPUTO is accused by the Grand Jury of the County of Multnomah and State of Oregon, by this indictment of the crime of CONTRIBUTING TO THE DELINQUENCY OF A MINOR committed as follows:\n\"The said LARRY CAPUTO on the 17th day of December, A.D. 1952, in the County of Multnomah and State of Oregon, then and there being, and one Lila Victor then and there being an unmarried female child under the age of eighteen (18) years, the said Larry Caputo did then and there unlawfully and feloniously do an act, to-wit: did then and there by threats, commands and persuasion induce and persuade the said Lila Victor to engage in prostitution, which course of conduct did manifestly then and there cause the said Lila Victor to become a delinquent child, * * *.\"",
"\n*459 On arraignment the defendant stated his true name to be Adolph A. Caputo. ",
"He entered a plea of not guilty and upon trial by jury was convicted. ",
"Defendant moved for judgment notwithstanding the verdict and in the alternative for a new trial. ",
"Both motions were denied and defendant was sentenced to imprisonment in the penitentiary for the maximum period of two years. ",
"He appeals. ",
"The statute, for violation of which, the defendant was convicted, reads as follows:\n\"When a child is a delinquent child as defined by any statute of this state, any person responsible for, or by any act encouraging, causing or contributing to the delinquency of such child, or any person who by threats, command or persuasion, endeavors to induce any child to perform any act or follow any course of conduct which would cause it to become a delinquent child, or any person who does any act which manifestly tends to cause any child to become a delinquent child, shall be punished upon conviction by a fine of not more than $1,000, or by imprisonment in the county jail for a period not exceeding one year, or both, or by imprisonment in the penitentiary for a period not exceeding five years.\" ",
"ORS 167.210.",
"\nIn the statute regulating juvenile court proceedings we find the following provision:\n\"(1) `Delinquent child' includes any child under the age of 18 years who violates any law of this state or any city or village ordinance, or who is incorrigible, or who is a persistent truant from school, or who associates with criminals or reputed criminals, or vicious or immoral persons, or who is growing up in idleness or crime, or who frequents, visits, or is found in any disorderly house, bawdy house, or house of ill fame, or any house or place where fornication is enacted, or in any saloon, bar-room or drinking shop or place, or any place where spirituous liquors, or wine, or intoxicating or malt *460 liquors are sold at retail, exchanged or given away, or who patronizes, frequents, visits or is found in any gaming house, or in any place where any gaming device is or shall be operated.",
"\n\"(2) `Child dependency,' `dependent children' and `neglected children,' unless otherwise required by context, have the meaning given those terms by ORS 419.102.\" ",
"ORS 419.502.",
"\nAssignments of error 1, 2, 3, 8, 10 and 12 raise substantially the same issues. ",
"It is argued that the court erred in refusing to direct a verdict of not guilty and in denying motions for judgment notwithstanding the verdict and for a new trial. ",
"It is contended that there was no convincing evidence that the defendant had committed the acts charged and that the evidence established nothing more than a mere suspicion of guilt. ",
"It is repeatedly argued that the prosecutrix had admitted the indiscriminate practice of intercourse with men and that because of this fact she had already become a delinquent child and a prostitute before she met the defendant. ",
"It is therefore contended that it was impossible for the defendant to have caused her to \"become\" what she already was. ",
"The prosecuting witness testified with extraordinary frankness concerning her conduct which proved her to have been a wild and irresponsible child before she met the defendant. ",
"She admitted sexual intercourse with several young men shortly before meeting the defendant. ",
"She had voluntarily gone to the judge of the domestic relations court and had been committed for a time to the Hillcrest school. ",
"We conclude that she was a delinquent child, 16 years of age, when she met the defendant. ",
"The evidence shows that she was driven out to the Shadows Club where the defendant was employed, by three young men, with at least two of whom she had enjoyed *461 sexual relations. ",
"The three boys went into the club and she stayed outside in the car. ",
"She testified that the defendant came out to the car and that they had a conversation. ",
"Concerning its substance, she testified:\n\"A * * * to the effect that, would I be willing to have him send gentlemen to see me for money for the purpose of prostitution, and from the conversation I am definitely sure that he knew that I had never done prostitution work for any money or for any ____\n* * * * *\n\"A Well, he asked me if I would be willing to have him send gentlemen to me, or men to me, from ____ I gather it was The Shadows club, because ____ well, here I go trying to tell again.",
"\n\"Q Go ahead.",
"\n\"A (Continuing) And he told me that if I was interested that I could give him a call, and from the conversation I gathered that the fellows that I had been with that evening had suggested it to him, but the conversation just ranged around the subject that would I be willing to have him send men to me, and I hadn't quite made up my mind `Yes' or `No.' ",
"I mean I just hadn't made any decision, but I guess in my mind I had kind of made up my mind that I would, and he told me to call him if I did decide, and I did call him * * *\n* * * * *\n\"Q That night when you were out in the car talking to the defendant, was anything said about money, about money again?",
"\n\"A Yes; it was with the understanding that the money would pass through him, from the fellow to him, and I presume that was under the ____ oh, I don't know ____ under the idea that I didn't know who he was sending, and probably he wasn't too sure of the fellows he was sending, and it wasn't *462 too definite whether I would ever be able to keep care of the money or not.",
"\n\"Q Was anything said to you as to who would pay you?",
"\n\"A Yes, it was under the agreement that he would pay me.",
"\n\"Q Did he say that to you?",
"\n\"A Yes, he did.",
"\n\"Q That night?",
"\n\"A Yes.",
"\n\"Q Did he say anything that night about the amount of money you would receive in connection with this type of work?",
"\n\"A No. ",
"I can't say as he said any definite figures, no.",
"\n* * * * *\n\"A Well, he did make a reference in the conversation about the way I was dressed, and that I was never secure in having a roof over my head, and that by earning money that I would be able to buy decent clothes, and that I would know where my next meal was coming from, and that I would have a roof over my head all the time.\"",
"\nThe defendant Caputo as a witness admitted leaving the club and talking with the prosecuting witness but stated that he only went out to tell her that she could not come in. ",
"The prosecuting witness testified further that pursuant to the arrangement she telephoned to the defendant and he sent her a man who went with her to his apartment. ",
"She said that they did not have intercourse but that she contacted Caputo at the club and he asked her if the man had been satisfied and gave her $20. ",
"Her testimony was that before this time she was substantially without funds. ",
"She testified that she gave $10 to one Voorhees the same night at a bootleg establishment, and her testimony was confirmed in that respect by Voorhees as a witness. *",
"463 A few days later she again telephoned Caputo at the club and received instructions to come out, which she did. ",
"At the club she was joined by a merchant seaman. ",
"They had intercourse and the next day she returned to the Shadows Club and received $20 from Caputo. ",
"She also testified to another similar transaction and another $20 payment from the defendant. ",
"After that she started working as a paid prostitute in three places in Oregon and in Tacoma, Washington. ",
"This continued until her arrest on 17 February 1952.",
"\nAlthough the defendant denied that he had entered into the alleged arrangements with her, he admitted that she was in the Shadows Club four or five times.",
"\n1-3. ",
"In view of defendant's contention that the testimony raised a mere suspicion of guilt, we deem it proper to say that we think her memory of the details of her varied and frequent activities is little short of remarkable. ",
"In our opinion, the record portrays a highly irresponsible child, but an intelligent and credible witness. ",
"In any event, we are not authorized to grant a new trial for insufficiency of the evidence unless we can affirmatively say that there was no substantial evidence to support the verdict. ",
"Oregon Constitution, Article VII, section 3; State of Oregon v. Moore, 194 Or 232, 241 P2d 455. ",
"The defendant cites five cases concerning the necessity of cogent and convincing evidence in cases based solely on circumstantial evidence. ",
"In such cases this court recognizes its duty to consider the sufficiency of the evidence under the rules set forth in State v. Dennis, 177 Or 73, 159 P2d 838, 161 P2d 670; State v. Long, 195 Or 81, 244 P2d 1033. ",
"This, however, was a case of direct and substantial testimony which, if believed by the jury, warranted conviction. ",
"Of course it was necessary that proof of guilt *464 be established beyond reasonable doubt, but the question in this case was one for the jury.",
"\nThe argument of the defendant is that it was impossible to cause the prosecuting witness to \"become\" a delinquent child when she already was one. ",
"The argument raises questions both as to the proper construction of the statute and as to the material allegations of the indictment. ",
"The construction suggested is of the type which moved the Supreme Court of Pennsylvania to say:\n\"Considering the beneficent purpose of the legislation, no court should be astute in finding reasons to relieve those who violate its provisions.\" ",
"Commonwealth v. Jordan, 136 Pa Super 242, 7 A2d 523, 528.",
"\nA similar argument was considered by the Supreme Court of Washington. ",
"An indictment charged that the defendant did \"willfully * * * encourage, cause, and contribute to the delinquency\" of a female child by attempting to indulge in sexual intercourse with her, the said child being \"then and there a * * * delinquent person.\" ",
"The court said:\n\"It is urged that the words `encourage, cause, and contribute' cannot be used together in this form, since the word `cause' must refer to those acts which bring about or are the initial cause of the delinquency, and the information charged herein that the minor was already a delinquent child.",
"\n* * * * *\n\"We think the word `cause' may well be used with reference to acts which bring about or assist in the continuance of a state of delinquency, as well as to refer to the acts which bring about the initial step of delinquency.\" ",
"State v. Strom, 144 Wash 334, 258 P 15.",
"\n*465 4. ",
"To become a delinquent child is not analogous to becoming a member of a lodge or of the I.W.W. See State v. Laundy, 103 Or 443, 204 P 958, 206 P 290. ",
"Having joined a lodge, we suppose one could not become more a member, but one may become a more delinquent child. ",
"It is our duty to construe the statute in accordance with the provisions of ORS 161.050. ",
"Even in states unlike our own, where the rule of strict construction of all criminal statutes still applies, it is held that in cases of this kind \"in arriving at a proper construction we are not required to close our eyes to the broad underlying policy and the dominant purpose of the whole law of which the section under consideration is a part. * * *\" ",
"State v. Adams, 95 Wash 189, 163 P 403, 404. ",
"See also State v. Dunn, 53 Or 304, 99 P 278, 100 P 258; Commonwealth v. Jordan, supra.",
"\n5, 6. ",
"The provision of ORS 167.210, supra, do not stop with the portion which makes it a crime to do an act which \"would cause\" or \"manifestly tend\" to cause a child to become delinquent. ",
"It also provides that \"When a child is a delinquent child * * * any person * * * by any act * * * contributing to the delinquency * * * shall be punished. * * *\" ",
"The defendant was informed by the indictment that he was charged with the crime of contributing to the delinquency of a minor and it alleges that he did induce and persuade the child to engage in prostitution. ",
"In view of the nature of the act charged we think it was unnecessary for the indictment to add that the act \"caused\" the child to \"become\" delinquent. ",
"The crime was sufficiently described when the defendant was notified that he had contributed to the child's delinquency by inducing her to engage in prostitution. ",
"We doubt if any one will have the hardihood to argue that such an act does not contribute to delinquency, *466 whether the child was already delinquent or for the first time became so.",
"\nAlthough the child had indulged in somewhat indiscriminate sexual intercourse before meeting the defendant, there is substantial evidence that she had not therefore become a prostitute for pay. ",
"In truth the jury were entitled to find that she became more delinquent by reason of the defendant's acts.",
"\nThe defendant relies upon certain language found in State v. Moore, supra. ",
"Our discussion in that case related, not to the indictment, but to the evidence. ",
"We said:\n\"* * * Of course, it is further necessary to prove that the act or acts so established manifestly tended to cause the child to become a delinquent child. ",
"In other words, it is necessary to establish by the evidence: (1) one or more of the acts of misconduct specifically alleged, and (2) that such act or acts manifestly tended to cause the child to become a delinquent child.\"",
"\nThe case was decided upon the ground that the specific act charged against the defendant was not proven. ",
"We did not intend to intimate that there is, as it were, an open season as to any child who has once become delinquent. ",
"It is unthinkable that the legislature intended that the waywardness of the child should be a defense to a man who attempts to lead her into a life of commercialized degeneracy.",
"\nThe foregoing disposes of assignments of error 1, 2, 3, 8, 10 and 12.",
"\n7. ",
"The trial court read to the jury the provisions of ORS 167.210, supra, which defines the crime of contributing to the delinquency. ",
"No error was committed thereby. ",
"Counsel has confused the act charged against the defendant with the alleged effect of the act. ",
"Upon *467 proof of the act here charged the jury could infer that the defendant by the act contributed to delinquency.",
"\n8. ",
"The refusal of the court to give an instruction on the burden of proof as requested by the defendant is assigned as error. ",
"The instruction given on that subject was one commonly employed by courts. ",
"The jury was informed that the plea of not guilty put in issue every material allegation of the indictment that all of the presumptions were in favor of innocence and that \"defendant is presumed to be innocent until such time, if that ever arrives, when he is convicted to your satisfaction and beyond a reasonable doubt. ",
"The state must prove beyond a reasonable doubt each and every material allegation of the indictment before there can be a conviction.\" ",
"The assignment is without merit.",
"\nThe court refused to give the following requested instruction:\n\"The Court instructs the jury that if from all the evidence in the case you believe beyond a reasonable doubt and to a moral certainty that the testimony of Lila Victor bears on its face indications of irresponsibility or improbability and is contradicted by other evidence then in that event I instruct you that her testimony unless corroborated is insufficient to support a conviction and in such event you must return a verdict of not guilty.\"",
"\nIn several respects the instruction was not a correct statement of the law. ",
"No law requires the corroboration of the testimony of an irresponsible witness. ",
"The word \"irresponsible\" may mean \"incapable of incurring responsibility. ",
"Not able to answer for consequences; unaccountable, undependable, carefree.\" ",
"Webster's New International Dictionary, 2d Ed. ",
"The testimony of such a witness may or may not be highly credible.",
"\n*468 9. ",
"We find no error in the refusal of the court to give instructions referred to in assignments of error 7, 8 and 9. ",
"So far as necessary they were covered by the instructions of the court.",
"\n10. ",
"Assignment of error 11 is without merit. ",
"The prosecuting witness testified briefly in answer to a question on cross examination by the defendant concerning a conversation:\n\"* * * Bob Voorhees and Jim Ciotti had been kind of talking to me about it, and I guess kind of getting me prepared for what Mr. Caputo was going to say.\"",
"\nDefendant moved that the answer be stricken. ",
"The court said:\n\"* * * I will instruct the jury to disregard the conversation with Voorhees and Ciotti, not in the presence of the defendant. ",
"You will try to erase that completely from your minds and not consider it in deciding the case.\"",
"\nThe last assignment of error covers many issues already discussed and urges upon us that the court erred in failing to grant a new trial for the reasons stated in the motion therefor. ",
"The motion states:\n\"* * * the defendant will show that one of the jurors sitting in the above entitled case namely Arthur Raymond Oberg was committed to the mental hospital for veterans at Roseburg, Oregon, by committment issued on February 19, 1946, being signed by the Honorable Ashby C. Dickson and being case No. ",
"16801, Department of Probate, and was not discharged as cured by the veterans' hospital but merely `Discharged from the Records'. ",
"That said committment was signed after a hearing of the petition of a sister of the juror and examination by two competent physicians and psychiatrists upon a finding that he was suffering from paranoic hallucinations. *",
"469 Section 2-407 Sub-section 33 O.C.L.A. holds that a condition once shown to exist is presumed to continue until evidence to the contrary is offered.\"",
"\n11. ",
"The alternative motion for a new trial was filed on 20 April 1953. ",
"It was denied on 22 April 1953. ",
"The motion for a new trial does not even contain an unsworn allegation that the juror had been committed seven years before. ",
"It merely states that defendant \"will show\" such event to have happened. ",
"We have searched the record in vain for any affidavit supporting the motion or purported motion. ",
"If defendant desired to raise this issue it was his duty at the least to present an affidavit setting forth the relevant facts. ",
"This he failed to do. ",
"The motion is not before us. ",
"Karberg v. Leahy, 144 Or 687, 26 P2d 56. ",
"Since the defendant has attempted to raise this issue we deem it proper in the interests of justice to add that the able trial judge considered the motion as if the matter were properly before him, and in denying it, expressed the opinion that the trial \"was absolutely fair\" and that the juror was sane at the time of trial.",
"\nThe judgment of conviction is affirmed.",
"\n"
] |
{
"pile_set_name": "FreeLaw"
}
|
[
0.0008862337563186884,
0.0007657759706489742,
0.0006515845889225602,
0.0006679625366814435,
0.0006822739378549159,
0.0007057040929794312,
0.0006087610963732004,
0.0007601753459312022,
0.0007699299021624029,
0.000668852124363184,
0.0007905580569058657,
0.0012663962552323937,
0.000825725554022938,
0.0015227774856612086,
0.0006801886484026909,
0.0006719124503433704,
0.012800758704543114,
0.002320016035810113,
0.0010160693200305104,
0.0012948372168466449,
0.0009017448173835874,
0.001099191838875413,
0.0006294630584307015,
0.0010019883047789335,
0.0007743908208794892,
0.16233976185321808,
0.000658158038277179,
0.002851233584806323,
0.42134058475494385,
0.0006098002195358276,
0.0014969861367717385,
0.013056694529950619,
0.0015861630672588944,
0.0008007578435353935,
0.0008425459382124245,
0.0021681024227291346,
0.0005807729903608561,
0.0010767622152343392,
0.000648906163405627,
0.011776067316532135,
0.0006642338703386486,
0.0015408920589834452,
0.0006765185389667749,
0.001225260435603559,
0.000660285702906549,
0.0006421822472475469,
0.000977435614913702,
0.0006085537024773657,
0.0007496735779568553,
0.0008197231218218803,
0.0006168698891997337,
0.000990046071819961,
0.0006657664198428392,
0.0008632585522718728,
0.0007128029246814549,
0.0006069775554351509,
0.008151370100677013,
0.0005959137342870235,
0.2542627453804016,
0.0006919511943124235,
0.0006371490308083594,
0.0008184516918845475,
0.0006242938688956201,
0.0007790137315168977,
0.0007940676296129823,
0.0007289053173735738,
0.0006522685871459544,
0.000697310664691031,
0.0007170915487222373,
0.000682055193465203,
0.008503054268658161,
0.000609645911026746,
0.0006462642922997475,
0.000781613402068615,
0.0006222874508239329,
0.27198871970176697,
0.00153326743748039,
0.0006413464434444904,
0.0008050880278460681,
0.001243838807567954,
0.008772583678364754,
0.02162577398121357,
0.0006712167523801327,
0.000617179786786437,
0.0007972269668243825,
0.0007212671334855258,
0.001079862006008625,
0.0019568621646612883,
0.1005871519446373,
0.0015656062169000506,
0.0007035073358565569,
0.0033449663314968348,
0.0023120343685150146,
0.06564316898584366,
0.0006636018515564501,
0.0006707742577418685,
0.000558946980163455,
0.02744469605386257,
0.009261076338589191,
0.0006354391807690263,
0.0011109060142189264,
0.005891794804483652,
0.0007123163086362183,
0.0013101339573040605,
0.0007184890564531088,
0.000657329335808754,
0.0006295467610470951,
0.0008929481264203787,
0.0014581186696887016,
0.0008006804273463786,
0.0005829811561852694,
0.0007748769130557775,
0.0007084541139192879,
0.0006653125747106969,
0.0007449229015037417,
0.0006155384471639991,
0.0008115700911730528,
0.0011692463885992765,
0.0008696287986822426,
0.0007530528237111866,
0.0005976189859211445,
0.00147302879486233,
0.0006472545210272074,
0.0006167928222566843,
0.0011672148248180747,
0.0007136136991903186,
0.0005990783683955669,
0.0009698614594526589,
0.0008199543808586895,
0.000786679913289845,
0.0006320914253592491,
0.0006845297175459564,
0.0006338248495012522,
0.014291354455053806,
0.0006783261778764427,
0.0011401979718357325,
0.0007431162521243095,
0.0006192133296281099,
0.0010332365054637194,
0.0006092360708862543,
0.0006134333088994026,
0.0005904357531107962,
0.0016263198340311646,
0.0006513597909361124,
0.0008830118458718061,
0.0006122960476204753,
0.0005949971964582801,
0.001995444530621171
] | 0.010381 | 148 |
[
"The Emerging Role of Stereotactic Ablative Radiotherapy for Primary Renal Cell Carcinoma: A Systematic Review and Meta-Analysis.",
"\nStereotactic ablative radiotherapy (SABR) is an emerging treatment option for primary renal cell carcinoma (RCC). ",
"To systematically review the literature on SABR for primary RCC and perform a meta-analysis evaluating local control (LC), toxicity, and renal function. ",
"A PROSPERO-registered (#115573), Preferred Reporting Items for Systematic Review and Meta-analyses (PRISMA)-based systematic review of the literature was conducted (1995-2019). ",
"Studies of SABR targeting primary RCC tumors were included, while those targeting only metastases were excluded. ",
"The primary outcome was LC defined as tumor size reduction and/or absence of local progression. ",
"Secondary outcomes included toxicity (Common Terminology Criteria for Adverse Events) and renal function (change in estimated glomerular filtration rate [eGFR]). ",
"Weighted random-effect meta-analyses using the DerSimonian and Laird method were conducted for primary and secondary outcomes. ",
"The I2 statistic and Cochran's Q test were used to assess heterogeneity. ",
"From 2386 PubMed entries and 924 meeting abstracts, 26 studies were identified (11 prospective trials), including 383 tumors in 372 patients, most of whom were deemed inoperable. ",
"Weighted averages (ranges) of median follow-up, median age, and mean tumor size were 28.0 (5.8-79.2)mo, 70.4 (62-83)yr, and 4.6 (2.3-9.5)cm, respectively. ",
"RCC histology was confirmed in 78.9% of patients who underwent pretreatment biopsy. ",
"Dose fractionation varied, but 26Gy in one fraction and 40Gy in five fractions were most common. ",
"The random-effect estimates for LC, grade 3-4 toxicity, and post-SABR eGFR change were 97.2% (95% confidence interval [CI]: 93.9-99.5%, I2=20%), 1.5% (95% CI: 0-4.3%, I2=0%), and -7.7ml/min (95% CI: -12.5 to -2.8, I2=2%), respectively, and heterogeneity was minimal. ",
"Six patients with pre-existing renal dysfunction (2.9%) required dialysis. ",
"Renal SABR is locally effective and associated with low toxicity rates for primary RCC, despite treatment of larger tumors in older, mostly medically inoperable patients. ",
"Stereotactic ablative radiotherapy is a high-precision, noninvasive radiation treatment requiring few outpatient visits, and represents a safe and effective management option for primary renal cell carcinoma."
] |
{
"pile_set_name": "PubMed Abstracts"
}
|
[
0.0005810919683426619,
0.0007181505206972361,
0.00054478261154145,
0.0005632612155750394,
0.0005462900153361261,
0.000573956232983619,
0.0005965636228211224,
0.0005748294643126428,
0.0006763521814718843,
0.0005788829876109958,
0.0006112211849540472,
0.0006127954693511128,
0.0007400555186904967,
0.0006235375185497105,
0.0007409277022816241,
0.0006091403192840517,
0.0007445233641192317
] | 0.000626 | 17 |
[
"INTRODUCTION\n============\n\nUrinary tract infections (UTIs) are serious health affecting problems worldwide.[@B1] *E. coli, E. faecalis, K. pneumoniae, S. marcescens, P. aeruginosa, S. saprophyticus, S. aureus* and *Proteus mirabilis* are most common bacteria causing UTIs in human beings.- The *E. coli* accounts for approximately 85% of community acquired UTIs and 50% of hospital acquired UTIs.[@B5] Different factors like age, gender, immuno-suppression and urological instruments may affect prevalence of UTIs.[@B6] Catheter-associated UTIs are one of the most dangerous health risks contributing 34% of all health care associated infections.[@B7]\n\nThe emergence of extended-spectrum beta-lactamases has threatened the empirical use of cephalosporins and ciprofloxacin.[@B8]^,^[@B9] Microorganisms use various mechanisms to develop drug resistance, such as recombination of foreign DNA in bacterial chromosome, horizontal gene transfer and alteration in genetic material.[@B10] Resistance pattern of microorganisms vary from country to country, state to state, large hospital to small hospital and hospital to community. ",
"In Pakistan, the problem of antibiotic resistance is compounding because of overuse and misuse of antibiotics.[@B6]^,^[@B11] There is no systematic national surveillance of antibiotic resistance and insufficient data is available to quantify the problem.[@B12] Detection of UTI causing pathogens and resistance of these pathogens to commonly prescribed antibiotics in clinical set ups is essential and helpful in improving the efficacy of empirical treatment.[@B13] Objective of the present study was to highlight the bacterial etiology of UTIs and determination of resistance pattern of *E. coli* isolates.",
"\n\nMETHODS\n=======\n\nThe observational and prospective study was conducted at Mayo Hospital Lahore, which is one of the oldest and biggest hospitals in Punjab.",
"\n\n***Sample collection and Isolation of Bacteria:***Urine samples (n=500) were collected from patients in different wards (n=400) and outpatient department (n=100) from Mayo Hospital, Lahore. ",
"Samples were centrifuged and sediments were cultured primarily on blood agar and macConkey's agar by spread plate technique. ",
"Bacterial colonies having different morphology were selected, purified and identified by their biochemical profiles.",
"\n\n***Multiple drug resistance:*** Antibiotic sensitivity pattern of *E. coli* isolates was determined on Muller Hinton agar plates by Kirby-Bauer disc diffusion.[@B14] Isolates were declared as sensitive or resistant on the basis of zone of inhibition following the criteria of Clinical Laboratory standards Institute.",
"\n\nRESULTS\n=======\n\nBacterial etiology of Urinary tract infections (UTIs) in patients admitted in or visiting Mayo hospital, Lahore as out patient was determined. ",
"Resistance pattern of *Escherichia coli* against a number of antibiotics was also checked. ",
"Bacteria were successfully isolated from 402/500 samples. ",
"Rate of isolation of bacterial etiological agent from female samples (87.5%) was not-significantly higher as compared to male (71.3%) patients. ",
"Out of 402 bacterial isolates from patients, rate of *E. coli* (321, 80%) isolation was highest followed by *Staphylococcus aureus* (38 9.4%), *Proteus species (*22 5.4%) and *Pseudomonas spp* (21 5.2%). *",
"E. coli* exhibited highest resistance to penicillin/amoxicillin (100%) followed by cefotaxime (89.7%), ceftazidime/cephradin (73.8%), tetracycline (69.4%), doxycycline (66.6%), augmentin (62.6%), gentamycin (59.8%), cefuroxime (58.2%), ciprofloxacin (54.2%), cefaclor (50%), aztreonam (44.8%), ceftriaxone/imipenem (43.3%), streptomycin (30%), kanamycin (19.9%), tazocin (14%), amikacin (12.7%) and norfloxacin (11.2%) ([Table1](#T1){ref-type=\"table\"}). ",
"Out of 321 *E. coli*, 261 (81%) were multiple drug resistant and 5 isolates were extensively drug resistant. ",
"Multiple drug resistance was defined as resistance to three or more than three different antibiotic classes tested.",
"\n\nDISCUSSION\n==========\n\nUTIs are caused by microbial invasion and subsequent multiplication in urinary tract.[@B15] Eighty percent of the patients with UTI had bacterial etiology in this study. ",
"Although the infection rate was higher in female (87.5%) patients as compared to male (71.3%), it was not-significant, which is in accordance with finding of Shah et al.[@B16] Rate of bacterial isolation was highest in elderly patients (\\>50 years), which is in accordance with Iqbal et al.6 *E. coli* was observed as the most common etiologic agent of UTI, which is also in accordance with previous studies.[@B1]^,^[@B17]^,^[@B18]\n\nAntibiotics are amongst the most important achievements of the twentieth century, used to kill or inhibit the growth of microorganisms. ",
"Antibiotic resistance in *E. coli* isolated from UTIs is increasing day by day, making it a major public health problem. ",
"So it is very important to determine the antibiotic resistance patterns in *E. coli* isolates for proper and accurate prescriptions.",
"\n\n###### \n\nAntibiotic sensitivity pattern of *E. coli* isolates\n\n *Antibiotics* *Codes* *Disks* *Resistant* *Intermediate* *Sensitive* \n ------------------------ --------- ----------- ------------- ---------------- ------------- ------ ----- ------\n Cephradine (CR) CR 30 µg 237 73.8 28 8.7 56 17.4\n Amikacin AK 30 µg 41 12.7 56 17.4 224 71.7\n Streptomycin S 10 ug 96 30 76 23.6 225 70\n Norfloxacin NOR 10 ug 36 11.2 101 31.4 184 58.9\n Ciprofloxacin CIP 5 ug 174 54.2 53 16.5 94 29.2\n Imipenem IPM 10 ug 139 43.3 55 17.1 127 39.5\n Cefuroxime CXM 30 ug 187 58.2 60 18.6 74 23.0\n Augmentin AMC 30 ug 201 62.6 40 12.4 80 24.9\n Ceftriaxone CRO 30 ug 139 43.3 60 18.6 122 38.0\n Gentamicin CN 10 ug 192 59.8 44 13.7 85 26.4\n Aztreonam ATM 30 ug 144 44.8 60 18.6 177 55.1\n Doxycycline DO 30 ug 214 66.6 33 10.2 74 23\n Pipracillin-Tazobactam TZP 100/10 ug 0 0 30 9.6 291 90.6\n Ceftazidime CAZ 30 ug 237 73.8 32 9.9 52 16.1\n Tetracycline TE 30 ug 223 69.4 29 9 69 21.4\n Cefaclor CEC 30 ug 160 50 60 18.6 101 31.4\n Tazocin TZP 110/10 ug 48 14.9 0 0 273 85\n Levofloxacin LEV 5 ug 0 0 21 6.5 300 93.4\n Kanamycin K 30 ug 64 19.9 100 31.1 157 48.9\n Meropenem MEM 10 ug 0 0 0 0 321 100\n Amoxicillin AMC 20 ug 321 100 0 0 0 0\n Pencillin P 10 u 321 100 0 0 0 0\n Tobramycin TOB 10 ug 0 0 0 0 321 100\n Cefotaxime CTX 30 ug 288 89.7 0 0 33 10.2\n\nUTIs caused by antibiotic resistant and multiple drug resistant bacteria have been increased in recent times. ",
"Complications in UTIs have increased because of the prevalence of extended spectrum beta-lactamases (ESBL) producing bacterial pathogens which are also causing many management and epidemiological issues. ",
"There were times almost a decade ago, when most of the ESBLs producing organisms were *Klebsiella* *spp.* ",
"and mostly were nosocomial. ",
"But in recent times the problem has been compounded by the prevalence of ESBL and MDR *E. coli* as well. ",
"Most of the ESBL *E. coli* are resistant to a wide range of beta lactams including cephalosporins, penicillins and piperacillin/tazobactam, and non beta lactams including fluoroquinolones, trimethoprim and gentamycin. ",
"One of the major reasons for this high resistance can be co-expressed resistance mechanisms in the species of different pathogens isolated from patients of urinary tract infections admitted to different wards of Mayo hospital Lahore, Pakistan. ",
"In the present study we analyzed their antibiotic sensitivity pattern was determined by Kirby Bauer technique.",
"\n\nIn present study all *E. coli* species (n=321) were resistant to penicillin and amoxicillin indicating a cautious use of these antibiotics for the treatment of urinary tract infections. ",
"In different parts of the world, resistance of *E. coli* to penicillins group of antibiotics have been on higher side and is increasing day by day, but there are only few reports which indicates 100% resistance to penicillins[@B19] Resistance to the combination of amoxicillin and a beta lactam inhibitor (augmentin 62.6%) was also on the higher side. ",
"Similar kinds of results, where beta lactam inhibitors increase the efficiency of penicillin group of antibiotic against *E. coli*, have been reported in previous studies.[@B20] Resistance to other beta lactam antibiotics including cefotaxime (89.7%), ceftazidime (73.8%), cephradin (73.8%), cefuroxime (58.26), cefaclor (50%), Ceftrioxone (43.3%) was also very high rendering many of these inefficient for empirical prescription of these antibiotics to treat UTIs. ",
"Previous studies in Pakistan have also shown very high antibiotic resistance in *E. coli* against cephalosporins and penicillins.[@B21]\n\nGenerally, in developing countries like Pakistan, penicillines and cephalosporins are not active against the UTI infections and our results suggest that these antibiotics should not be used in the treatment of UTIs. ",
"Inefficiency of penicillins and cephalosporins in this study does not indicate that these antibiotics are not in use in any part of world to treat UTIs caused by *E. coli*. ",
"In some of the recent reports a higher number of *E. coli* was found sensitive to penicillins or cephalosporins from European countries.[@B22] A decade before, these antibiotics were active against *E. coli* even in Pakistan.[@B20]\n\nIn this study, the resistance of *E. coli* against aztreonam and imipenem was 44.8% and 43.3%, respectively, which is higher than previous studies.[@B23]^,^[@B24] Higher resistance in *E. coli* against carbapenams indicates that these may have been misused and overused in health care set ups. ",
"Tazocin, a combination of piperacillin and beta lactamases inhibitor tazobactam, showed best results, for which resistance in *E. coli* was only 14% suggesting that this antibiotic can still be used for the treatment of UTIs.[@B25] Although, tetracycline group of antibiotics are not used now a days for human infectious agents, *E. coli* were highly resistant to tetracycline (69.4%) and doxycycline (67.6). ",
"In the present study variable resistance patterns were found for the aminoglycosides. *",
"E. coli* were highly resistant to gentamycin, while low level of resistance was for kanamycin (19.9%), and amikacin (12.7%).",
"\n\nQuinolones, especially ciprofloxacin have been used for *E. coli* infections in recent past. ",
"In the present study however *E. coli* were highly resistant to ciprofloaxacin (54.2%), which is consistent with the previous reports.[@B26] Other fluoroquinolones such as norfloxacin (11.2% resistance) and levofloaxacin (all sensitive) were found efficient for the *E. coli*. ",
"Other studies from the different parts of the world also show that quinolones are still active against UTI infections.[@B26] Multiple drug resistance (MDR) and extensive drug resistance (XDR) was also determined in this study. ",
"MDR is described as resistant to at least one member from three different antibiotic classes being used for the treatment of *E. coli*, while extensive drug resistance (XDR) is described as resistance to at least one member of all but two antibiotic classes. ",
"MDR and XDR *E. coli* in this study were 81% and 8.7% respectively. ",
"The antibiotics active against the XDR were amikacin and norfloxacin generally. ",
"It is concluded that higher level of antibiotic resistance, MDR and XDR is present in *E. coli*. ",
"To treat the UTIs caused by *E. coli* combination therapy especially amikacin and ciprofloaxacin may provide better results. ",
"Antibiotic resistance in *E. coli* isolated from UTIs insinuates for its close monitoring and prescription of antibiotics after the culture sensitivity tests.",
"\n\nAuthors Contributions:\n======================\n\nAll the authors have contributed significantly in study design, experimentation, data analysis and manuscript drafting.",
"\n\n***Conflict of interest:*** We declare that the authors have no conflict of interest.",
"\n"
] |
{
"pile_set_name": "PubMed Central"
}
|
[
0.0009534926502965391,
0.000641222984995693,
0.0005968431360088289,
0.0006110641406849027,
0.0007101815426722169,
0.000568951538298279,
0.0005732518038712442,
0.0006931411917321384,
0.0006298339576460421,
0.0005542843719013035,
0.0006306126597337425,
0.0006943957414478064,
0.0012456178665161133,
0.0010656563099473715,
0.0006607398972846568,
0.0008044633432291448,
0.0008644957561045885,
0.0008715396397747099,
0.0005745008238591254,
0.015341862104833126,
0.0005693275015801191,
0.0011888903100043535,
0.0011511254124343395,
0.0006287064170464873,
0.0018799406243488193,
0.0006127952365204692,
0.0005392562597990036,
0.0012009694473817945,
0.0006681366940028965,
0.0007478426559828222,
0.0007129982113838196,
0.0010624206624925137,
0.000704597623553127,
0.0008328540716320276,
0.0005351474392227829,
0.0007935086614452302,
0.0006784690776839852,
0.0009628261323086917,
0.0005560451536439359,
0.0006536064902320504,
0.0009705421980470419,
0.0008103869622573256,
0.0006499449955299497,
0.0010789416264742613,
0.000578198058065027,
0.0006195383612066507,
0.0006919363513588905,
0.001995444530621171
] | 0.001112 | 48 |
[
"\"We see this on TV too often,” Rizzo said. “",
"I feel like it’s all the time. ",
"There’s a cycle to it. ",
"We get horrified that this violence is inflicted on our kids. ",
"We get angry that there’s nothing we can do and nothing is done about it. ",
"And then we ultimately get immune and move on to something else. ",
"But then it happens in our own town — in your own school, or movie theater, or nightclub, or church. ",
"And we realize that it can happen to us, in our safe and tightknit community, Parkland.\""
] |
{
"pile_set_name": "OpenWebText2"
}
|
[
0.0005949498154222965,
0.0008042495464906096,
0.000762653537094593,
0.03890275955200195,
0.05935798957943916,
0.0010802859906107187,
0.0023858549538999796,
0.0012971707619726658
] | 0.013148 | 8 |
[
"Maya Moore is a true soldier of social justice, and she proved in a way no one..."
] |
{
"pile_set_name": "OpenWebText2"
}
|
[
0.0011779144406318665
] | 0.001178 | 1 |
[
"BCLocalRoot.ca is an online grocery store offering 100% BC only food and beverage products with delivery and curbside pick-up for Metro Vancouver residents. ",
"The company is a social enterprise on mission to engage and empower our communities to choose an economic and food secure future by purchasing BC only food and beverage products.",
"\n\nBCLocalRoot.ca currently has 400+ BC grown and made products from 60+ brands representing 9 regions in the province and new products are added every week. ",
"With more than 4,000 BC food and beverage brands, we believe that everything we truly need is right here at home in BC!",
"\n\nVancity members get 10% off their first order using the code VANCITYBCLR10 at checkout."
] |
{
"pile_set_name": "OpenWebText2"
}
|
[
0.0007268905756063759,
0.0005900326650589705,
0.000662325881421566,
0.0005734947044402361,
0.0011041667312383652
] | 0.000731 | 5 |
[
"Content search results\n\n... in the jail???????",
"\nTanvir: ?????",
"\nAd : catch him, but in some places most of em' are killers\nTanvir: That i feel is a talk ... muslims did that.... then .... who is gaining..... US.... but look who is gaining.... we are just loosing....\nTanvir: ....Even in a ...\n\n... appeal. ",
"He has got a nice, muscular body and looks stunning. ",
"But he always appears in front of everyone as a stupid person. ",
"His face looks so good but his expression says that he’s the stupidest men on the earth. ",
"This is ...\n\n... manufacturers. ",
"Nike generally constitutes a regular size but slightly wider baseball shoe, whilst Adidas has a regular width yet ... down is still in the hands of the passage of time is gone, but the two men do not know, this is their first time so hearttalk,\n...\n\n... nike free run nz the two are nestling together, but just can not warm. ",
"Ye Zhong Ngok clutch her gently, remember a few days ago ... I only serve you one thing, that you married a good wife, but you're not her.",
"\nthere was nothing surprising, but the problem is the ...\n\n... product, not just comes from hands connected with designers, but additionally comes from requirements connected with consumers. ",
"Please ... will be regret it!It's not your first time that as a host, but she is still tense and fork out more attation to get ready for it.",
"\n...\n\n... 1TLYP9F5 moon it's iconic beauty are still a mystery, but the leading theory is that nike show where Europa's crust cracked open ... and currently holds the money in the bank briefcase. ",
"But Ambrose hasn't won a Pay-Per-View match since the split in June and ... him as a standout superstar on Steve Austin's podcast. ",
"But despite all this, Ambrose is still in danger of being forced into a ...\n\n... that he faces competition for the beginning function but is centered on strengthening his very own game and becoming much better ... for the Massive 10 title this season Andre Tippett Jersey, but Gardner is defiant.&ldquoI appear forward to getting the ... is unaccustomed to playing that function, but for the team to beat its rivals on the street this season, Gardner demands ...\n\n... nike air max 180 shoes. ",
"They truly not only smaller but in addition dignity. ",
"It's perpetually the colorways with 'original' ... is the weighty hitter overseas (particularly in Europe) - but let's pass over our digits and expect these come on our side. ",
"With the ..."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.01585960015654564,
0.0016455395380035043,
0.02398422174155712,
0.0012161730555817485,
0.91755610704422,
0.9174887537956238,
0.000808173674158752,
0.0006897446583025157,
0.0018137997249141335,
0.0006723915576003492,
0.00787216518074274,
0.000687692838255316,
0.0007957060588523746,
0.0011750326957553625,
0.0006338143139146268,
0.0006166191305965185,
0.0009074610425159335
] | 0.111437 | 17 |
[
"Q:\n\nRead text file using FileReader\n\nI've written the following code in an attempt to read the contents of a .txt file\n<!",
"DOCTYPE html>\n<html>\n\n<input type=\"file\" id=\"files\" name=\"file\" />\n<div id=\"container\" style=\"height: 500px; min-width: 500px\"></div>\n\n<script>\n function handleFileSelect(evt) \n {\n var files = evt.target.files; // FileList object\n\n // Loop through the FileList and render image files as thumbnails.",
"\n for (var i = 0, f; f = files[i]; i++) \n {\n\n var reader = new FileReader();\n reader.onload = (function(theFile) \n {\n var contents = theFile.target.result;\n var lines = contents.split('\\n');\n\n })(f);\n\n reader.readAsText(f);\n }\n }\n\n document.getElementById('files').addEventListener('change', handleFileSelect, false);\n</script>\n</html>\n\nUsing firebug I set a break on var contents = theFile.target.result; but nothing is being returned. ",
"Can anyone spot what's wrong?",
"\nThank you\n\nA:\n\nTry (onload with closure):\nfunction handleFileSelect(evt)\n{\n var files = evt.target.files; // FileList object\n\n // Loop through the FileList and render image files as thumbnails.",
"\n for (var i = 0, f; f = files[i]; i++)\n {\n\n var reader = new FileReader();\n reader.onload = (function(reader)\n {\n return function()\n {\n var contents = reader.result;\n var lines = contents.split('\\n');\n //////\n document.getElementById('container').innerHTML=contents;\n }\n })(reader);\n\n reader.readAsText(f);\n }\n}\n\nOr (without closure):\nfunction handleFileSelect(evt)\n{\n var files = evt.target.files; // FileList object\n\n // Loop through the FileList and render image files as thumbnails.",
"\n for (var i = 0, f; f = files[i]; i++)\n {\n\n var reader = new FileReader();\n reader.onload = function(event)\n {\n // NOTE: event.target point to FileReader\n var contents = event.target.result;\n var lines = contents.split('\\n');\n //////\n document.getElementById('container').innerHTML=contents;\n };\n\n reader.readAsText(f);\n }\n}\n\n"
] |
{
"pile_set_name": "StackExchange"
}
|
[
0.000885104644112289,
0.0007286602049134672,
0.0021390507463365793,
0.0007622208795510232,
0.000700331584084779,
0.0010530625004321337,
0.0013145686825737357
] | 0.001083 | 7 |
[
"Q:\n\nselect/sum of a column with different conditions in 1 query\n\n$pos = select * from score_history where content_id = 6 && val = 1 \n$neg = select * from score_history where content_id = 6 && val = -1 \n\ni want to get the pos and neg scores in one query \nbut i dont want to use join \nso perhaps some sort of IF/case statement ?",
"\ni've this but as you can guess it fails\nSELECT count(*) as total , \nCASE \n WHEN `val` = 1 THEN count(*) as `pos` \n WHEN `val` = -1 THEN count(*) as `neg` \nEND\nFROM score_history WHERE `content_id` = '46083' \";\n\nis there any way to do this without using join or sub query ?",
"\n\nA:\n\nYou can make use of the flexibility of MySQL to handle booleans and integers:\nSELECT count(*) total, sum(val = 1) pos, sum(val = -1) neg\nFROM score_history \nWHERE content_id = '46083';\n\nWhenever the condition is true it is a 1. ",
"Otherwise a 0. ",
"No CASE needed nor GROUP BY.",
"\n\n"
] |
{
"pile_set_name": "StackExchange"
}
|
[
0.001987806521356106,
0.006183694116771221,
0.007147057913243771,
0.0026371150743216276,
0.0006135812727734447,
0.001995444530621171
] | 0.003427 | 6 |
[
"WASHINGTON (MarketWatch) — U.S. F-22 stealth fighter jets arrived in South Korea on Sunday for exercises as part of an effort to demonstrate advanced military capabilities, seeking to deter provocations from North Korea, according to reports.",
"\n\nThe move come after North Korea warned South Korea earlier in the weekend that the Korean peninsula had entered a “state of war,” according to an Associated Press report. ",
"Pyongyang threatened to shut down a border factory that is a symbol of North and South Korean cooperation, according to the report."
] |
{
"pile_set_name": "OpenWebText2"
}
|
[
0.0006333025521598756,
0.0010734796524047852,
0.01663561351597309
] | 0.006114 | 3 |
[
"Tarts and Liars\n\nBackstory: I have just come out of a long term relationship and was making the most of being single. ",
"This was sent around 8 a.m.\n\nMum: Are you at home?Me: NoMum: I know your not, I drove past your house and your car wasn’t there.",
"Me: Then why ask?Mum: I was checking to see if you were going to try and lie to me.",
"Me: Why would I do that?Mum: TART!",
"\n\nWhen liver or kidney function is strained it will show up as\nnitrogen in the blood. ",
"Next, check any cables, both power and network, to ensure they are securely attached to the appropriate\ndevices. ",
"Tbh in text speak\n\nThey cited a period in their history, during their wandering in the wilderness, when their “. ",
"The 480i means that the signal carries 480 horizontal lines of information, the i is for interlaced, meaning each line is refreshed alternately.",
"\n\nThe reasons you want to get the ability to\nbecome able to make use of both telephones at once inside the same conversation may go\non, but you get the point. ",
"These energy grid lines inside the earth are electrically charged lines\nthat are similar to your body’s acupuncture meridian system lines.",
"\nMilwaukee electric drill They appear in timeless and\nbeautifully soft fabrics to improve the beauty of the home.",
"\n\nIf you decided to acquire the 3600 m – Ah battery for XBOX 360 controller, this will\nlikely be the perfect item to accompany it. ‘",
"Their brain wave patterns are abnormal and stay\nlike that for the long period.",
"\n\nMake bound to massage in a very circular motion for three to minutes and still\nhave the least number of pressure within the eye area.",
"\nTo cope using the hot summer you must eat more fruits\nand supply one’s body with the vitamins A,C and E. Anti aging hand cream Shahnaz Hussain is considered to become the empress on the world of natural Beauty Products.",
"\n\nNot getting professional consultation before treatment might cause irreparable problems for skin. ",
"Incase you are trying to find\na darker look, and consider the employing liquid eyeliner rather than brush and pencil eyeliners.",
"\n\nmarket and rumors persist that DISH Network and DIRECTV may yet merge in order that\nthey. ",
"Some experts recommend to shut all the\nlights inside room and to install a light behind the TV.",
"\nInsignia tv not working The Vauxhall Insignia is constantly on the draw the interest of buyers at the seventh\nplace with 3,912 units sold.",
"\n\nGenerally, the higher the rate, a little more smoothly the tv screen will be qualified to generate moving pictures.",
"\nIf you wish to connect the Wii on the LCD TV applying this\ncable, the steps are very straightforward.",
"\n\nConnectivity is fixed to Verizon’s CDMA network frequencies of 800 and 1900, 3G is handled via EVDO\nrev. ",
"On the other hand, it revealed a treasure trove of information towards the science computer and\ntemporarily revealed the invisible Malecorman ship to everyone’s monitors.",
"\n5″ led light bar This company takes it on the next level by ensuring everything they are doing is eco-friendly, while using LED lighting to encouraging upcycling by employees.",
"\n\nCCTVs are predominantly employed in places where external monitoring like people movement and activities in public areas forums are required.",
"\nThis can be a detailed review, based on research rather than personal experience.",
"\n\nBest dvr security system CCTV methods are 1 with\nthe old and popular safety systems that people today are employing nowadays.",
"\n\nCCTV cameras are highly effective in cutting crime rate and are useful for monitoring purposes.",
"\nAll the dealings will probably be reported on the recording\ndevice within the DVR security system so that any bad things can be prosecuted using the right evidence.",
"\n\nAnd it’ll only take a little shorter than an hour to push the\nbutton and test it. ",
"That means you can have paid for your unlimited Skype subscription and Skype number\nfor nothing in any way if you planned on with your smart phone to\nutilize Skype over Wi-Fi.",
"\n\nIf you would like to switch off the wireless adapter around the desktop PC you will\nhave to adopt another route. ",
"Your mobiles equipment including a\nlaptop or telephone (via Wi – Fi capabilities) can join for the network\nso lengthy since it in number of its signal. ",
"Internet speed cnet In sum\n- an extraordinary figure, that allows to keep in contact with a lots of friends, partners, customers, contractors and acquaintances.",
"\n\nThe most prominent part within the World of Warcraft activity is questing\nand when your character completes the mission the player\ngets experience points, items or in-game money.",
"\nAlthough Intel own a top end CPU market generally Gaming PC builds is depending\non Intel i5 and i7 CPU using the latter being the fastest.",
"\nIbuypower review 2015 While you desire a hard disk that\n\nWell, most don’t realize this, but Epic Games has been around for (drumroll, please) 20 freaking\nyears. ",
"The game board is split up into two boards with Breakout on the left, and Pong\nfor the right. ",
"Gaming mouse for large hands 2015 Now\nit’s tough to say that there’s anything generally wrong with the game,\nexcept for the blatant lack of being able to invert the mouse controls.",
"\n\nWhy this duo stumbled on pass is tough to say, however it\ndid, and so we’ve got the Habu, an inexpensive and competitive gaming mouse.",
"\nIf you happen to be really into games and another of the most significant\naccessories you may need is really a mouse.",
"\n\nThe former High School Musical star has been captured leaving the BBC One Radio in London, England on Wednesday, November 30,\n2011 using a short spikey hair. ",
"On your first stop down the path\nof internet radio apps, have a look at iheartradio. ",
"Uv 82 vs uv5r This is\nfast, clear and effective nevertheless for that, you need the proper device that could serve the need.",
"\n\nPut the decision-in number on your Outlook Calendar using the appointment in order that it pops up a couple of minutes before.",
"\n\nHt baofeng bf888s Increasing demands on frequency spectrum, a finite resource,\nforce users to look for further spectrum-efficient systems,\ntogether with increased features. ",
"However, in the event you function a company such\nas being a hotel, you will want a number of channels.",
"\n\nAlthough she agreed to use the toy bear and liked him, she was reluctant to tap on the\nstuffed toy itself, so Al suggested that she place Tappy on the chair beside her.",
"\nThese waiting room toys are devised to occupy\nminimum space on the floor and also have immovable parts that reduce instances of untidy office environments.",
"\nAdult nerf guns These stuffed toy animals supply\na bit of reassurance to a puppy.",
"\n\nThere are 216 Rare Earth Magnets using the toy set ,whenever you order because of this\ntoy,you will obtain a complete list of 216 magnetic ball ,each\nhaving 5mm in diameter. ",
"Many people believe by using any toy that a young\nchild uses, they are able to learn and turn into educated on the\nuses of numerous aspects of their surroundings.",
"\n\nThis is completed while it truly is transmitting car stereo of top quality concurrently.",
"\nIn fact, 85% of those facilities are usually sold with connectors, including those\npurchased in Europe. ",
"Rca to cable converter box Even much better than that though, the remote\ndevice has glow-in-the-dark buttons, so despite how low you turn the lights down,\nyou’ll always be in a position to find the best control.",
"\n\nThey’re even cuter on a large screen where you may share them using the whole family.",
"\nYou better put money into some draperies (blinds) or you are able\nto install the TV within a room with less windows.",
"\n\nOne in the things you will must determine is if you want your security\ncamera systems to have night vision. ",
"Out of the innovations\ndone the creation from the ultra sophisticated wireless security camera systems surely takes the cake.",
"\nSecurity cameras ipad These smoke detectors have a low profile camera inside\nthat can record everything that is certainly happening around the\nhouse.",
"\n\nBeing in a position to monitor the perimeter\nof your dwelling with the help of a security alarm camera may be the only method to\ntruly keep the house safe from an invasion. ",
"Thus, it’s obvious that wireless security surveillance cameras enjoy considerably more advantages instead of wired ones.",
"\n\nAlienware products possess a unique turn to them, and this also is further enhanced because of the Alien – FX lighting system\nthat may be adjusted by anyone. ",
"In addition,\nwhen you’re using a laptop, it truly is easy to forget the correct body\nposture and this may affect the muscles negatively. ",
"Best\ngaming laptops 2016 Anyone planning to buy a gaming\nlaptop probably will end up taking a look\nat these two companies.",
"\n\nIf the Num Lock is put off can exist only dedicated buttons, but laptop keyboards\nfunctionality is lost. ",
"monitor, CPU, and keyboard) are already built-into one device.",
"\n\nThe most prominent part in the World of Warcraft online game is questing and once your\ncharacter completes the mission you gets experience points,\nitems or in-game money. ",
"And you can find several ways you’ll be able to entertain yourself\nwith one of these gaming laptops. ",
"Ibuypower desktop reviews Recommendation: Anti-Ghosting and a\n1000Hz polling rate are fairly standard of all gaming\nkeyboards, so there’s no real reason to accept anything less.",
"\n\nIf an out-of-the-box gaming experience will be your preference a pre-built gaming system will be a a great deal more accessible\nas well as simple option. ",
"This game was better on single player, with entertaining missions,\nbut on multi-player side wasn.",
"\n\nThough its maximum resolution is 720×480 for NTSC\n(no support for HDTV) the functionality just for this price range is impressive.",
"\nHowever, we can easily prevent crimes from happening in the very\nfirst place by installing a CCTV system. ",
"Cctv systems Similarly\nif it truly is too low, your cameras might not work effectively and they are more\nsusceptible to being damaged or vandalised.",
"\n\nYou can merely set up just wherever you wish and start recording.",
"\n\nThe modern CCTV systems can alert you via your cellular phone no matter in which you are\nwithin the world.",
"\n\nThe mini bar might be mounted about the dash, deck or about the roof from the vehicle.",
"\nOn some cell and smartphones, there could be an icon from the notification bar that tells you for those who have your SIM card installed, however, some phones may not have this.",
"\nBuy led light bar Red and blue lights are exclusive to cop cars\nhowever, and the light bars happen to be mounted around the roof from the vehicles.",
"\n\nFire trucks and ambulances are also most important\nemergency response vehicles which use lightbars and strobe lights in red\namber and white colors. ",
"LED lights conserve energy to your great extent nonetheless it is important\nto look at note of the requirements prior to deciding to invest\nin vehicle lighting.",
"\n\nPerformance is extremely good in both daytime and night time\nwhile using infrared lighting selection for low light conditions.",
"\nAs far being a WM6, smartphone is worried, the Asus M530W may seem to\nhave it all. ",
"Real – Pad is pre-loaded with numerous fantastic and all\nto easy to use apps like the Real – Quick Fix, games, entertainment\nand AARP tools.",
"\n\nMany security parameters are followed to reduce the possible\nthreats. ",
"The best protective solution to safeguard out there online threats like phishing, spyware, and hackers is\nid theft monitoring.",
"\n\nIt makes the means of identifying and recognizing the suspects should of theft and other criminal\nactivities very easy. ",
"For under $170, this may be the best high-tech beauty gift for the people looking to avoid, correct and repair acne.",
"\nThe last technology These gadgets resemble covered cameraswhich\nare widely-used to spy on folks.",
"\n\nLooking with the very best quality cable, telephone, and net provider.",
"\nWith the help of mobile phones some victims who are stuck under this debris\nhave even been capable to call for help and direct individuals to their\nlocation.",
"\n\nAll a baby has to grow up well is being attended to regularly.",
"\nIt is scientifically proven that couples that hug and touch like they used to do in the beginning stay together longer and with a more satisfying relationship.",
"\nBest pocket pussy 2015 Although phthalates have\nbeen utilized in common household items and medical instruments considering that the 1920.",
"\n\nIn an expression, you might be the controller of one’s individual\nlife and hope all of you use a awesome spring break.",
"\nDai-Do is often a organization which offers an aluminum alloy vibrator.",
"\n\nHi! ",
"I really could have sworn I’ve visited this web site before but after going through a number of the articles I realized it’s new\nto me. ",
"Anyways, I’m certainly delighted I came across it and I’ll\nbe bookmarking it and checking back frequently!"
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.047095783054828644,
0.043073296546936035,
0.01837742142379284,
0.22177059948444366,
0.00308116152882576,
0.0006215107277967036,
0.000588459603022784,
0.0005729385884478688,
0.0005954076186753809,
0.0008498661336489022,
0.0005500752595253289,
0.0006657032645307481,
0.0023406879045069218,
0.0007500684005208313,
0.0007975134649313986,
0.0009331497130915523,
0.000970915425568819,
0.000656307500321418,
0.011703163385391235,
0.0006888913339935243,
0.0005544913001358509,
0.0005364143871702254,
0.0007946653640829027,
0.0008984652813524008,
0.000581786094699055,
0.0005791580188088119,
0.0005658646114170551,
0.0006564491777680814,
0.0006507612997666001,
0.0006418661796487868,
0.0007479434716515243,
0.0008174875401891768,
0.0006261685048229992,
0.000604042608756572,
0.0005228297086432576,
0.0007303191814571619,
0.0007171588949859142,
0.054922547191381454,
0.0015564562054350972,
0.0010483060032129288,
0.0006995401345193386,
0.000672403781209141,
0.0010013965656980872,
0.0009367204620502889,
0.0005770180141553283,
0.0006859563291072845,
0.0006017828127369285,
0.0005552576039917767,
0.0007869543042033911,
0.0008117603720165789,
0.13428360223770142,
0.0006567955715581775,
0.0007453904836438596,
0.0005792072042822838,
0.0005856152856722474,
0.0006381837301887572,
0.0006200753268785775,
0.0034729475155472755,
0.0006929216324351728,
0.0007590562454424798,
0.001147327246144414,
0.0007035143789835274,
0.0006069422233849764,
0.0006080698221921921,
0.0006079670274630189,
0.0005394303007051349,
0.0007585923303849995,
0.0006681853556074202,
0.0007832410046830773,
0.0006991772097535431,
0.0006396978278644383,
0.0005987381446175277,
0.0007549709407612681,
0.0005748853436671197,
0.0006440618308261037,
0.0009871061192825437,
0.0006127598462626338,
0.0007647725287824869,
0.0007074814639054239,
0.0005660774186253548,
0.0006963192136026919,
0.0007787725189700723,
0.0005185962654650211,
0.0005650017410516739,
0.0008499977411702275,
0.0006913105025887489,
0.0006655891193076968,
0.0018989704549312592,
0.0007569562876597047,
0.036705199629068375,
0.004040715284645557,
0.0005339732160791755,
0.0005647126235999167,
0.006295826751738787,
0.0005673635751008987,
0.514226496219635,
0.0011853250907734036,
0.0011281842598691583,
0.001105394447222352,
0.0006003143498674035,
0.000605391338467598
] | 0.011532 | 101 |
[
"<!",
"DOCTYPE HTML>\n<html>\n <head>\n <script src=\"https://ajax.googleapis.com/ajax/libs/jquery/2.2.4/jquery.min.js\"></script>\n <script type=\"text/javascript\" src=\"../../../tests/qunit/qunit-bootstrap.js\"></script>\n <script type=\"text/javascript\" src=\"../../../mwEmbedLoader.php\"></script>\n <script type=\"text/javascript\" src=\"../../../docs/js/doc-bootstrap.js\"></script>\n <script type=\"text/javascript\" src=\"../../RaptMedia/tests/RaptMedia.qunit.js\"></script>\n </head>\n<body>\n\n <div id=\"kaltura_player\" style=\"width: 560px; height: 395px\"></div>\n <div>\n <input class=\"changeMedia\" type=\"button\" data-entryId=\"404\" value=\"Missing Entry\" />\n <input class=\"changeMedia\" type=\"button\" data-entryId=\"0_kh14ze3t\" value=\"RAPTMedia Demo Playlist\" />\n <input class=\"changeMedia\" type=\"button\" data-entryId=\"1_4bzj26am\" value=\"Kaltura University Intro Video\" />\n </div>\n <script>\n \n var d = new Date(\"December 10, 2016 12:00:00\");\n var cacheSt = (parseInt(d.getTime()/1000)).toString();\n\n kWidget.featureConfig({\n \"targetId\": \"kaltura_player\",\n \"wid\": \"_811441\",\n \"uiconf_id\": 33991361,\n \"entry_id\": \"0_kh14ze3t\",\n \"flashvars\": {\n \"EmbedPlayer.",
"WebKitPlaysInline\": true,\n \"dash.plugin\": true,\n \"raptMedia.plugin\": true,\n \"raptMediaDurationLabel.plugin\": true,\n \"raptMediaScrubber.plugin\": true,\n \"durationLabel.plugin\": false,\n \"scrubber.plugin\": false,\n \"largePlayBtn.plugin\": false\n },\n \"cache_st\": cacheSt\n });\n\n kWidget.addReadyCallback(function(playerId){\n this.kdp = document.getElementById( playerId );\n var _this = this;\n $( 'input.changeMedia' ).click(function(){\n _this.kdp.sendNotification( \"changeMedia\", { 'entryId' : $(this).attr('data-entryId') });\n });\n });\n </script>\n</body>\n</html>\n"
] |
{
"pile_set_name": "Github"
}
|
[
0.006595523562282324,
0.0008812391315586865,
0.0011052039917558432
] | 0.002861 | 3 |
[
"t t be (75/125)/((-56)/(-5)). ",
"Is t at most as big as 1?",
"\nTrue\nSuppose 4*q = 3*k + 30, -6*q + 2*q = 0. ",
"Suppose -30*u + 32*u = -3*t - 32, -3*u - 4*t = 45. ",
"Which is smaller: k or u?",
"\nk\nLet k(l) = 11*l + 471. ",
"Let i be k(-43). ",
"Let m = 164001 + -1809374/11. ",
"Let q = 487 + m. Which is smaller: q or i?",
"\ni\nLet x(s) = 40*s - 3. ",
"Let q be x(-2). ",
"Let y = q + 71. ",
"Which is bigger: y or -11?",
"\n-11\nLet y = -16.26 - -0.26. ",
"Let s = y + 16.1. ",
"Which is smaller: s or 3/13?",
"\ns\nLet w(s) = s + 2. ",
"Let o be w(-11). ",
"Let v be -4 + (-8 - 6) + 7 + -1. ",
"Which is smaller: o or v?",
"\nv\nLet r(h) = -h**3 - 9*h**2 + 24*h + 2. ",
"Suppose 0 = -2*n - 26 + 4. ",
"Let a be r(n). ",
"Which is smaller: -21 or a?",
"\n-21\nSuppose 52 = 2*w - 5*a, 4*a = 8*w - 9*w + 26. ",
"Which is greater: 24 or w?",
"\nw\nLet c be -3 + (-3)/(-3) + 3. ",
"Suppose -c = -2*n + 17. ",
"Is n not equal to 7?",
"\nTrue\nLet b(g) = -8*g - 79. ",
"Let a be b(-10). ",
"Let y = 2/15 + 1/5. ",
"Does a = y?",
"\nFalse\nLet p = -13611/41 - -332. ",
"Suppose -2*v - v + 12 = 0. ",
"Let l be (v - 4)*3/(-9). ",
"Is p > l?",
"\nTrue\nLet x = 1027/6 + -122165/714. ",
"Is x greater than or equal to 0?",
"\nTrue\nLet w = -7825 + 923343/118. ",
"Are w and 0 non-equal?",
"\nTrue\nLet i = 50 + -34. ",
"Let n = -16.2 + i. Which is bigger: n or 5?",
"\n5\nLet b = 1545/11 - 141. ",
"Is -1 greater than or equal to b?",
"\nFalse\nLet f = -5/21 + 1/6. ",
"Suppose -c = 5*h + 19, -5*c + 4*h = 7 - 28. ",
"Does f = c?",
"\nFalse\nLet i be 48/(-192) - 126/8. ",
"Do -16 and i have the same value?",
"\nTrue\nLet b(q) = 2*q**2 + q**3 + 4*q**3 + 2 - q - 3*q**3. ",
"Let r be b(-2). ",
"Let u = -63/22 + 4/11. ",
"Which is bigger: r or u?",
"\nu\nLet r = -2801 + 2837. ",
"Is -5 at most as big as r?",
"\nTrue\nLet p = 55 + -38. ",
"Suppose 3*b - 5*t - p = -5, 3*b + 5*t = -18. ",
"Is b <= -3/2?",
"\nFalse\nLet n(u) = u**3 + 8*u**2 - 3*u - 18. ",
"Let k be n(-8). ",
"Let s be -5*(0 - k/(-15)). ",
"Which is smaller: s or -3/2?",
"\ns\nSuppose 5*q + 58 + 22 = 5*u, -u + 52 = -4*q. ",
"Which is smaller: q or -53/4?",
"\n-53/4\nLet s = 27805/3 + -9402. ",
"Let c = -135 - s. Suppose -2*f - 19 = -4*q + q, 2*f = -3*q + 11. ",
"Is f smaller than c?",
"\nTrue\nLet o = 525 - 533.84. ",
"Let j = o - 0.16. ",
"Let u = 8.6 + j. Which is smaller: u or 2?",
"\nu\nLet a = -0.1 - -0.1. ",
"Let b = 10.96 + -37.96. ",
"Does b = a?",
"\nFalse\nLet a(u) = -u**2 - 7*u - 1. ",
"Suppose -m + 2*f = -f + 20, -m = -2*f + 15. ",
"Let l be a(m). ",
"Which is smaller: 10 or l?",
"\nl\nLet b(g) = 5*g + 9. ",
"Let n = 21 + -27. ",
"Let i be b(n). ",
"Let a = i + 21. ",
"Which is smaller: a or 2/41?",
"\na\nLet k = 3/188 - 597/2068. ",
"Suppose 0 = d + s - 4, 0*s + 4 = -d + s. Which is bigger: d or k?",
"\nd\nLet y = 13 + -34. ",
"Let s = 75 + -53.95. ",
"Let m = s + y. Which is bigger: m or 0?",
"\nm\nSuppose 4*j - 10 = 5*l - j, 22 = 4*l + 2*j. ",
"Let q be (1 - 4/10)/(l/(-2)). ",
"Is 23 greater than or equal to q?",
"\nTrue\nLet p = 261/280 + -19/20. ",
"Is 0 greater than p?",
"\nTrue\nSuppose s = -43 + 150. ",
"Let o = s + -23. ",
"Which is smaller: 83 or o?",
"\n83\nLet t = 0.2 + 2.4. ",
"Let x = -2.76 + t. Let h = 0 + x. Is -1 greater than or equal to h?",
"\nFalse\nLet b = -9 - -13. ",
"Suppose r = b*z - 4, 0 = z + 3*z - 5*r + 12. ",
"Let j = z - 5. ",
"Does j = 2?",
"\nFalse\nLet o be (-2)/16 + 212/(-352). ",
"Suppose -j - 5 = 2*s + 4, -2*j - 5*s - 22 = 0. ",
"Is j smaller than o?",
"\nTrue\nLet h be (1 - 2) + (-50)/(-51). ",
"Let i be (-1)/((1 + 2)/(18/6)). ",
"Is h < i?",
"\nFalse\nLet o = -3348 - -3340. ",
"Suppose -5*g = 2*t + 2*t - 15, 2*g = -2. ",
"Suppose -2*n = -4*u - 8, 2*n - n = -t*u - 3. ",
"Is u at least as big as o?",
"\nTrue\nLet r = 6 + -7. ",
"Let k be 0/(-5 + 2 - r). ",
"Is k > -2/7?",
"\nTrue\nLet r = -102.4 - 8.6. ",
"Which is smaller: -2 or r?",
"\nr\nLet v = -505.7 + 529. ",
"Let a = -22 + v. Let u = a + -1. ",
"Is 2/3 at least as big as u?",
"\nTrue\nSuppose 5*h = 3*t + 12, -5*t + 5*h = 4*h + 20. ",
"Let p be 590/385 + t/(-22). ",
"Let u(k) = k**3 + 21*k**2 - 24*k - 43. ",
"Let v be u(-22). ",
"Which is smaller: v or p?",
"\nv\nLet a = -42 + 83. ",
"Let o = a + -41.04. ",
"Which is smaller: o or -0.1?",
"\n-0.1\nSuppose 4*m = -4*p - 504, -4*p + 0*p + 2*m - 486 = 0. ",
"Let j = p + 121. ",
"Suppose 2*h + 9 = -h. ",
"Is h bigger than j?",
"\nFalse\nLet g = -0.863 + 0.737. ",
"Is g > 4?",
"\nFalse\nLet b = -27.86 + 29.46. ",
"Is b less than -4?",
"\nFalse\nLet x(b) = b - 1. ",
"Let v be x(4). ",
"Let j be 3 + v/(-24)*65. ",
"Let i = j + 631/120. ",
"Is -1 <= i?",
"\nTrue\nLet q be 20596/57 + (-4)/(-6). ",
"Is -2/3 at most q?",
"\nTrue\nLet a = 243 - 326. ",
"Is 1 at least a?",
"\nTrue\nLet p = 0.3 + -2. ",
"Let j = p - -1.77. ",
"Which is greater: -2/9 or j?",
"\nj\nLet y = -9.12 - -9.02. ",
"Are 104 and y unequal?",
"\nTrue\nLet r = 20 - 18. ",
"Let m = r - 5. ",
"Which is greater: m or 1?",
"\n1\nLet u = -534/11 - -548/11. ",
"Is u smaller than 1/4?",
"\nFalse\nSuppose 0 = 2*k - 3*y - 9, -3 - 62 = -5*k - y. Let u be (-4)/3*k/8. ",
"Which is greater: u or 0?",
"\n0\nLet z be 11/(198/(-28))*(-2)/4. ",
"Let d(l) = -l**3 - 2*l**2 + 1. ",
"Let q be d(-1). ",
"Is z less than q?",
"\nFalse\nLet k(h) = -2*h**3 - 5*h**2 + 1. ",
"Let w be k(-2). ",
"Let r be w + 76/(0 - -4). ",
"Is 15 at least r?",
"\nFalse\nLet n = -10 + 9. ",
"Let q = n - -2. ",
"Let h be 16/40 + (-6)/(-10). ",
"Do h and q have different values?",
"\nFalse\nLet j = 375659/21105 + 2/4221. ",
"Let c = j - 1671/95. ",
"Which is greater: c or 0?",
"\nc\nLet o(r) = -r**3 + 10*r**2 + 34*r - 212. ",
"Let h be o(10). ",
"Is 133 equal to h?",
"\nFalse\nSuppose -5*i = -5*o - 115, -5*o + 60 = 7*i - 5*i. ",
"Let a = i + -11. ",
"Is 14 <= a?",
"\nTrue\nLet o = 463 + -806. ",
"Which is greater: o or -345?",
"\no\nLet y be (21/8)/(63/462). ",
"Let j = -37/2 + y. Is 0 greater than or equal to j?",
"\nFalse\nLet n be 1 + 10/(-8) - 1995/(-140). ",
"Let s = n + 12. ",
"Is 26 smaller than s?",
"\nFalse\nLet t be ((-2)/5)/((-110)/(-125)). ",
"Let f(v) = -v**2 - 4*v - 3. ",
"Let i be f(-3). ",
"Which is smaller: i or t?",
"\nt\nSuppose 0 = 3*c - 4*u - 28, 2*c + 0*u - 2*u = 16. ",
"Let i = 18 - 21. ",
"Let p = i + c. Is p greater than or equal to 2?",
"\nFalse\nLet f(y) = -y**2 - 3*y + 2. ",
"Let z be f(3). ",
"Suppose 37*x - 66 = 4*u + 40*x, -u + x - 13 = 0. ",
"Which is greater: u or z?",
"\nu\nLet p = -3.6 + 3.11. ",
"Let k = p - 7.51. ",
"Do k and 5 have the same value?",
"\nFalse\nLet k(i) = -133*i + 77. ",
"Let y(l) = -7*l + 4. ",
"Let n(q) = -3*k(q) + 56*y(q). ",
"Let f be n(2). ",
"Suppose f*w = 4*w - 3. ",
"Which is greater: w or -5/7?",
"\n-5/7\nLet l be (27/6)/(1 - -2). ",
"Let h = -23 + 23.1. ",
"Let k = -0.1 + h. Which is bigger: k or l?",
"\nl\nLet k be ((-2)/8 - (-3)/6)*(-804)/(-18). ",
"Suppose -4*t + 16 = 2*c, 5*c + 3*t - 46 = -t. ",
"Is k > c?",
"\nTrue\nSuppose -y - 8 = -3*y. ",
"Suppose y*k + 3*t = -26, 5*k - 4*t = -20 + 3. ",
"Which is smaller: -4 or k?",
"\nk\nSuppose -g + 3 + 1 = 0. ",
"Suppose 6 = -z + g*z. ",
"Suppose -z*x - l + 0*l = -5, 3*x = -4*l - 5. ",
"Is 4 greater than x?",
"\nFalse\nLet b = 1 - 7. ",
"Let v(f) = -2*f**2 + 19*f - 4. ",
"Let h(z) = 6*z**2 - 55*z + 12. ",
"Let q(g) = -6*h(g) - 17*v(g). ",
"Let y be q(4). ",
"Does y = b?",
"\nFalse\nSuppose 2*r = -3*q - 62, -r - 13*q + 10*q - 34 = 0. ",
"Which is smaller: -2 or r?",
"\nr\nLet u = -1519321/42 + 36178. ",
"Let r = 23/6 - u. Which is greater: r or 0?",
"\nr\nLet z be (6 - 4)/(-68)*13. ",
"Let g = -1/34 - z. Let f = g - 29/34. ",
"Is -7 at least as big as f?",
"\nFalse\nSuppose 0 = -18*d + 117*d - 33858. ",
"Is d !",
"= 342?",
"\nFalse\nLet g be -1*(-14)/(-91) + 38/(-260). ",
"Which is smaller: g or -17?",
"\n-17\nLet p be (-2)/(-2)*-4 + 1. ",
"Let c be p + -46*(-1)/14. ",
"Let d = -25/76 - -3/38. ",
"Are d and c equal?",
"\nFalse\nSuppose 4*b + 13 = 5*t, -5*t - 2*b - 4 + 5 = 0. ",
"Let s be (6/(-8))/(t/4). ",
"Suppose 5*v + 12 = -8, 0 = -4*o - v - 8. ",
"Which is smaller: o or s?",
"\ns\nLet o = -26 - -26. ",
"Let u be ((0 + -1)*(o - 1))/3. ",
"Let m be (-6)/2*8/(-6). ",
"Which is smaller: u or m?",
"\nu\nLet w be (-50963)/164 + 3/4. ",
"Are -2 and w equal?",
"\nFalse\nLet z = -12 - -16. ",
"Suppose 0 = -4*t + z. Let o = -38 - -37. ",
"Which is smaller: o or t?",
"\no\nLet i = 79 + 6. ",
"Let y be (80/i)/(1*2). ",
"Which is smaller: 0 or y?",
"\n0\nLet n = -103 + 105. ",
"Suppose 0 = 2*v - n, -7*v + 2*v - 63 = 4*x. ",
"Is x greater than -19?",
"\nTrue\nLet g = -9 + 1. ",
"Let c = -1 - g. Suppose -c*f - 10 = -2*f. ",
"Which is greater: f or 0?",
"\n0\nLet a = -1.5 - -2. ",
"Let p = -0.5 - a. Is p at least as big as -11?",
"\nTrue\nLet j = 2 + 1. ",
"Suppose 2*s - 9 = s + 4*k, 3*k = 5*s - 28. ",
"Suppose s*c + 19 = g, -11 = j*c + 1. ",
"Is g > -1/2?",
"\nFalse\nLet d(h) = -2*h**2 - h. Let z(v) = -v**2 - v - 1. ",
"Let k(w) = -2*d(w) + 3*z(w). ",
"Let n be k(3). ",
"Let l = n - 17/5. ",
"Which is smaller: -8 or l?",
"\n-8\nLet u be 132*(5/(-10))/(-1). ",
"Let i be 3/(-9) + u/9. ",
"Is i greater than or equal to 8?",
"\nFalse\nLet g be (308/(-5) - -2) + 14/(-35). ",
"Let x be (0 + (-8)/6)*(-345)/g. ",
"Which is smaller: -2 or x?",
"\nx\nLet g be -2 + 3 + -31*1. ",
"Let p be ((-4)/5)/(g/(-225)). ",
"Let c be 27/p - 8/(-16). ",
"Which is smaller: -5 or c?",
"\n-5\nLet w be (4/14)/((-24)/70). ",
"Let v(d) = d**2 - 7*d - 3. ",
"Let s be v(8). ",
"Suppose -2*t - 16 = 4*x + 2*t, -s*t = 15. ",
"Which is bigger: x or w?",
"\nw\nSuppose 0 = 55*l - 0 - 55. ",
"Which i"
] |
{
"pile_set_name": "DM Mathematics"
}
|
[
0.00208419025875628,
0.0015861268620938063,
0.31054818630218506,
0.04806134104728699,
0.0013946486869826913,
0.11570105701684952,
0.0010262419236823916,
0.0007813813863322139,
0.0016011944971978664,
0.027683917433023453,
0.0011540739797055721,
0.0011825236724689603,
0.0009185549570247531,
0.006054895929992199,
0.002011829288676381,
0.0007404809002764523,
0.005282217636704445,
0.004008484072983265,
0.0024843448773026466,
0.0008965565357357264,
0.054024044424295425,
0.007110179867595434,
0.009218597784638405,
0.0007151761674322188,
0.005647863261401653,
0.000851765216793865,
0.00963140930980444,
0.004291870165616274,
0.0012773021589964628,
0.009943349286913872,
0.0012532228138297796,
0.0019830302335321903,
0.0017787342658266425,
0.001258858828805387,
0.001170995645225048,
0.0013840219471603632,
0.0016775819240137935,
0.0009261169470846653,
0.0015696656191721559,
0.0009740889072418213,
0.0015888360794633627,
0.0010275915265083313,
0.0029322514310479164,
0.0008764383383095264,
0.001091211917810142,
0.004039521794766188,
0.0029532613698393106,
0.0028755369130522013,
0.0014110287884250283,
0.0007063453667797148,
0.010670059360563755,
0.0014881457900628448,
0.0017641339218243957,
0.0019044997170567513,
0.0016982194501906633,
0.003109632059931755,
0.001329222577624023,
0.01214654091745615,
0.0009950360981747508,
0.19225472211837769,
0.004857474938035011,
0.009146123193204403,
0.0007581337704323232,
0.020299842581152916,
0.000710050982888788,
0.0012779635144397616,
0.05644092708826065,
0.0012047957861796021,
0.0009189129923470318,
0.009332319721579552,
0.0017856683116406202,
0.02798619121313095,
0.0011708379024639726,
0.001681627589277923,
0.07152363657951355,
0.03804934024810791,
0.00231058313511312,
0.0007340004667639732,
0.04294697567820549,
0.0015921670710667968,
0.020049717277288437,
0.003044125158339739,
0.0006991312839090824,
0.0016212589107453823,
0.04846028611063957,
0.002086582826450467,
0.0010229406179860234,
0.012514248490333557,
0.006242483854293823,
0.0008953263750299811,
0.001111143035814166,
0.0008194552501663566,
0.0012901733862236142,
0.0009603447397239506,
0.026978621259331703,
0.0007696500979363918,
0.004280872642993927,
0.005407294724136591,
0.002020995132625103,
0.04512857273221016,
0.0018103437032550573,
0.0011724935611709952,
0.0021676465403288603,
0.002851892728358507,
0.002085315063595772,
0.0011570029892027378,
0.001047565252520144,
0.009613423608243465,
0.0026939320378005505,
0.012670997530221939,
0.19496788084506989,
0.19100268185138702,
0.0011946327285841107,
0.004900471307337284,
0.0009978518355637789,
0.0010986322304233909,
0.0007363783661276102,
0.0014126745518296957,
0.025870686396956444,
0.01590539887547493,
0.00633419631049037,
0.0006927128997631371,
0.9913731217384338,
0.0028410328086465597,
0.0007682156865485013,
0.0049132974818348885,
0.00592104671522975,
0.0008167758351191878,
0.010921178385615349,
0.001246386207640171,
0.00193941465113312,
0.01761918142437935,
0.004195243585854769,
0.0009001815924420953,
0.0028471481055021286,
0.0012130949180573225,
0.0018535155104473233,
0.0014895981876179576,
0.0022615077905356884,
0.0008890116005204618,
0.001453686272725463,
0.000778554764110595,
0.0008410636219196022,
0.0009232676820829511,
0.0009603753569535911,
0.001955724786967039,
0.001639772322960198,
0.0008476743241772056,
0.003989838529378176,
0.0009016486583277583,
0.0009403098374605179,
0.0024367175064980984,
0.00088578334543854,
0.001587547012604773,
0.0013237830717116594,
0.8755831718444824,
0.0016241921111941338,
0.002578542334958911,
0.007481714244931936,
0.0011911573819816113,
0.006922740489244461,
0.8142926096916199,
0.002001199871301651,
0.0009553113486617804,
0.001054977416060865,
0.004821301903575659,
0.002226137788966298,
0.0012832003412768245,
0.0008165217004716396,
0.0014867690624669194,
0.0009060540469363332,
0.0010336274281144142,
0.17245565354824066,
0.006072382442653179,
0.001443778513930738,
0.013028671965003014,
0.004874697886407375,
0.001250878325663507,
0.0015774740604683757,
0.000834205886349082,
0.00811596680432558,
0.0023589949123561382,
0.002082978608086705,
0.001809420995414257,
0.0009112608386203647,
0.0017376189352944493,
0.1263306736946106,
0.0021320628002285957,
0.0008700278121978045,
0.17862309515476227,
0.0009613137226551771,
0.005634131841361523,
0.06906608492136002,
0.0024999231100082397,
0.0027803550474345684,
0.0012678636703640223,
0.012873276136815548,
0.002169250976294279,
0.000839287182316184,
0.01306008081883192,
0.005277219694107771,
0.5775719881057739,
0.006558471824973822,
0.9120721817016602,
0.0008663396001793444,
0.0011885336134582758,
0.0019690701737999916,
0.006873240228742361,
0.005311358720064163,
0.01169730257242918,
0.0012976910220459104,
0.0011616416741162539,
0.9427455067634583,
0.0007676796521991491,
0.0017683315090835094,
0.02237272635102272,
0.003703697118908167,
0.0010588474106043577,
0.0018125423230230808,
0.7154614329338074,
0.004724419210106134,
0.012065669521689415,
0.0012168079847469926,
0.0015945867635309696,
0.003732785815373063,
0.0007363783661276102,
0.0017643033061176538,
0.0032406144309788942,
0.007368855644017458,
0.011641702614724636,
0.011564040556550026,
0.001096179592423141,
0.002671710914000869,
0.0009829150512814522,
0.002738428534939885,
0.0007531971205025911,
0.00288203451782465,
0.001990998862311244,
0.0010405464563518763,
0.005473054945468903,
0.008617433719336987,
0.0009512134129181504,
0.021596180275082588,
0.0008817346533760428,
0.018694499507546425,
0.02803047187626362,
0.0016309501370415092,
0.0008318040054291487,
0.00839325413107872,
0.0010519165080040693,
0.00205856841057539,
0.018428850919008255,
0.00106815155595541,
0.004362667910754681,
0.0020094048231840134,
0.0008630678639747202,
0.0020390083082020283,
0.02481301873922348,
0.0009727389551699162,
0.0015099066076800227,
0.31028053164482117,
0.001934452448040247,
0.006603189278393984,
0.0041514369659125805,
0.0012894067913293839,
0.886013388633728,
0.019814569503068924,
0.0009278962970711291,
0.01982012763619423,
0.3376535475254059,
0.003706692485138774,
0.0025202264077961445,
0.0007214039796963334,
0.002135351300239563,
0.0035463448148220778,
0.001454438199289143,
0.0011961203999817371,
0.0015572654083371162,
0.0007442221394740045,
0.005784122738987207,
0.0010521213989704847,
0.0009389574406668544,
0.0007037748582661152,
0.0023189655039459467,
0.012474282644689083,
0.0010119996732100844,
0.02899617701768875,
0.0010034077567979693,
0.001665486954152584,
0.0025489151012152433
] | 0.035913 | 290 |
[
"This invention relates to an implantable infusion system. ",
"In particular, it relates to an improvement in an infusate accumulator used with a valve system in such a system operating at positive pressure to dispense medication in accordance with different specified flow rates.",
"\nImplantable infusion pumps utilizing an accumulator positioned between a pair of valves which alternately open and close are disclosed in U.S. Pat. ",
"Nos. ",
"4,838,887 and 5,049,141, commonly assigned. ",
"The disclosure of the '887 patent is expressly incorporated herein by reference. ",
"In such a system, as illustrated in FIG. ",
"1, a drug reservoir 10 is refillable by means of a septum 12. ",
"The reservoir comprises a sealed housing 14 containing a bellows element 16 having a chamber 18 which comprises the drug reservoir. ",
"The bellows 16 separates the housing 14 into a second zone 20 which is normally filled with a two-phase fluid. ",
"The fluid, normally Freon, vaporizes and compresses the bellows 16 thus providing a release pressure to the reservoir 18 through the outlet leading to the infusion site. ",
"During the refill process chamber 18 is loaded with medication via the septum 12. ",
"The two-phase fluid is then pressurized condensing a portion of the vapor and returning it to the liquid phase.",
"\nSuch systems also employ an outlet filter 24 and a side port 27 for direct bolus injections. ",
"The reservoir and fluid delivery techniques are well established in constant flow systems. ",
"To provide for increased accuracy in delivery dosage, the use of an accumulator 30 has been proposed for use in a number of applications.",
"\nThese systems employ a metering assembly comprising two normally closed valves 26 and 28. ",
"Interposed between the two valves is an accumulator 30. ",
"The valves 26 and 28 are controlled electronically by the module 32 which may be programmed utilizing an external programmer 34.",
"\nIn operation of such a system, a constant positive pressure flow from the reservoir 10 to the inlet valve 26 occurs. ",
"Valve 26 is opened while the outlet valve 28 is closed. ",
"This loads the accumulator 30 with a predetermined amount of medication. ",
"The valve 26 is then closed and outlet valve 28 opened so that the contents of the accumulator 30 are delivered to the catheter 36 for delivery to the infusate site. ",
"The rate of switching of the valves 26 and 28 thus determines the frequency of pumping through the system and therefore determines the delivery rate of medication through the catheter 36. ",
"For each cycle a constant amount is delivered, the accumulator volume.",
"\nIn the system illustrated in FIG. ",
"1, the accumulator 30 has two alternative forms as illustrated in FIGS. ",
"2 and 3. ",
"For purposes of convenience, FIGS. ",
"2 and 3 utilize the same numbering as in the '887 patent. ",
"As illustrated in FIG. ",
"2, the inlet is represented by numeral 58 wherein fluid enters the accumulator from the inlet valve 26. ",
"The accumulator comprises a diaphragm 90, a backing plate 92, an end cap 94, and fill tube 96 and a spacer plate 98. ",
"The diaphragm 90 deflects in response to fluid entry of the chamber 102. ",
"The backing plate 92 acts as a mechanical stop to limit motion of the diaphragm. ",
"Similarly, the spacer plate 98 is used to limit diaphragm motion during discharge, that is, the passage of fluid through the outlet 59.",
"\nThe use of the spacer plate 98 as a mechanical stop creates a problem when fragile materials such as insulin are used. ",
"The material is locally compressed and crushed by the mechanical action of the diaphragm contacting the stop. ",
"This breaks down the fluid. ",
"Since the internal volumes are quite small, this in turn can result in residue formation inside the accumulator.",
"\nThe fill tube 96 is used to supply an inert gas to the chamber 104. ",
"The purpose of chamber 104 is to provide a region which is charged at a pressure lower than that of the infusate pressure in drug reservoir 18 so that accumulator chamber 102 fills when inlet valve 26 is opened, but higher in pressure than that of the catheter 36 to allow the diaphragm 90 to deflect back to the spacer plate 98 position as the chamber 102 empties when the outlet valve 28 is opened. ",
"Additional details of this accumulator may be found in U.S. Pat. ",
"No. ",
"4,838,887.",
"\nFIG. ",
"3 illustrates an alternative configuration. ",
"In this alternative arrangement, the backing plate 92 comprises three elements which electrically isolate the center of the plate from the diaphragm 90. ",
"A supply of inert gas is still supplied via the feed fill tube 96. ",
"A lead 110 is attached to a flange forming a portion of the end cap assembly. ",
"A ceramic cup 113, which is lined with metal 111, provides a conductive path between a stop 114 and the lead 110. ",
"The diaphragm 90 is used as a moving switch contact. ",
"This provides a signal indicating that the accumulator is full, that is, the diaphragm in an upward position contacting the stop 114. ",
"This electrical signal is used for diagnostic determinations of the system such as leaks in the valve. ",
"Again, more complete details as to the use of the modified accumulator of FIG. ",
"3 are found in the '887 patent.",
"\nA problem with these types of accumulators is the formation of precipitate on the spacer plate. ",
"Over time this build-up deteriorates the flow accuracy in the system and in an extreme case causes blockage. ",
"U.S. Pat. ",
"No. ",
"5,049,141 addresses this problem by minimizing the contact points with the diaphragm. ",
"This solution reduces stress points on the fluid within the accumulator. ",
"There are however still points of contact and therefore potential damage to the material."
] |
{
"pile_set_name": "USPTO Backgrounds"
}
|
[
0.0005573942326009274,
0.0006002031150273979,
0.000705494312569499,
0.0008322071516886353,
0.000668034132104367,
0.0005657395231537521,
0.0006089586531743407,
0.0007794302655383945,
0.0006036034901626408,
0.0007696382817812264,
0.0007378925220109522,
0.0006464028265327215,
0.0006858412525616586,
0.0008590577053837478,
0.0005738476756960154,
0.0005576801486313343,
0.0007020271732471883,
0.0008532822248525918,
0.0006854361854493618,
0.0005864989943802357,
0.0006977025768719614,
0.0008477793890051544,
0.0009013011585921049,
0.000617732061073184,
0.0006318333325907588,
0.0006367951282300055,
0.0007096219924278557,
0.0007354761473834515,
0.0005884096608497202,
0.0006117870798334479,
0.0006542113842442632,
0.0005930893821641803,
0.000902269093785435,
0.0007562666432932019,
0.001439877087250352,
0.0008915658108890057,
0.000653607479762286,
0.0008555148961022496,
0.0014072251506149769,
0.0006410200730897486,
0.0007237689569592476,
0.0008028207812458277,
0.0009994925931096077,
0.0013785817427560687,
0.0009357182425446808,
0.0008292015991173685,
0.0006367802270688117,
0.0006526532815769315,
0.000813114398624748,
0.0007001474150456488,
0.000661083497107029,
0.0009431486832909286,
0.0013673542998731136,
0.0005781479412689805,
0.000694083166308701,
0.0006333880010060966,
0.0010408275993540883,
0.0007496242760680616,
0.0010884626535698771,
0.0013785817427560687,
0.00455812131986022,
0.0006990080000832677,
0.000998621224425733
] | 0.000845 | 63 |
[
"Subjective social status and shaming experiences in relation to adolescent depression.",
"\nTo investigate the associations among social status, shaming experiences, and adolescent depression using a status-shaming model. ",
"Population-based, self-report cohort study. ",
"Västmanland, Sweden. ",
"A cohort of 5396 students in grade 9 (age 15-16 years) and the second year of high school (age 17-18 years). ",
"Intervention Participants completed the anonymous questionnaire Survey of Adolescent Life in Västmanland-2006 during class hours. ",
"We investigated the prevalence of depression according to the Depression Self-Rating Scale of the Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition, Text Revision), A-criterion, regarding subjective social status, shaming experiences, and social background. ",
"Social status was measured as (1) attributed status of a family's socioeconomic and social standing and (2) acquired status of peer group and school. ",
"Binary logistic regressions were used for the analyses. ",
"Shaming experiences and low social status interacted with depression. ",
"If shaming experiences were present, participants with both high and low attributed status were at increased risk for depression (odds ratio [low and high groups, respectively], 5.4-6.9), whereas medium status seemed to have a protective function. ",
"For acquired status, the highest elevated risk was found in participants with low status (odds ratio [girls and boys, respectively], 6.7-8.6). ",
"Social status may influence the risk for depression when an individual is subjected to shaming experiences. ",
"The present study contributes to the mapping of the influence of social status on health and may have essential implications for understanding, preventing, and treating adolescent depression."
] |
{
"pile_set_name": "PubMed Abstracts"
}
|
[
0.002978577744215727,
0.0013788201613351703,
0.0006057759746909142,
0.000761610921472311,
0.0006430065259337425,
0.0006284883711487055,
0.0006474331021308899,
0.000613111536949873,
0.0006195487221702933,
0.012437084689736366,
0.0006898452411405742,
0.0006474300753325224,
0.0033531873486936092,
0.0008442255202680826
] | 0.001918 | 14 |
[
"TICKETS HERE 5 BUCKS EACH: http:// www.brownpapertickets.com/ event/2526661\n\nAn hour long comedy showcase at Destiny City Comics featuring three of your soon to be favorite stand-up comedians: Chad Denick, Lisa Wallen, Jim Stewart Allen, and also the host, Taylor Bonzer!",
"\n\nThe venue is small so grab your tickets ASAP! ",
"If we sell enough we’ll do more of these!",
"\n\nRSVP on Facebook"
] |
{
"pile_set_name": "OpenWebText2"
}
|
[
0.0009826677851378918,
0.0014802676159888506,
0.005562423728406429,
0.000791488157119602
] | 0.002204 | 4 |
[
"Q:\n\nAndroid Wear:CardScrollView cannot scroll?",
"\n\nI use CardScrollView and CardFrame in my project, but the CardScrollView can not scroll, here is my code, any suggestion? ",
"\nCardScrollViewActivity.java\npublic class CardScrollViewActivity extends Activity {\n\n private CardFrame cardFrame;\n @Override\n protected void onCreate(Bundle savedInstanceState) {\n super.onCreate(savedInstanceState);\n setContentView(R.layout.card_scroll_view);\n cardFrame=(CardFrame)findViewById(R.id.card_frame);\n cardFrame.setExpansionDirection(CardFrame.",
"EXPAND_UP);\n cardFrame.setExpansionEnabled(true);\n cardFrame.setExpansionFactor(CardFrame.",
"NO_EXPANSION);\n }\n}\n\ncard_scroll_view.xml\n<android.support.wearable.view.",
"CardScrollView\n xmlns:android=\"http://schemas.android.com/apk/res/android\"\n android:orientation=\"vertical\"\n android:layout_width=\"match_parent\"\n android:layout_height=\"wrap_content\">\n\n <android.support.wearable.view.",
"CardFrame\n android:id=\"@+id/card_frame\"\n android:layout_width=\"match_parent\"\n android:layout_height=\"wrap_content\">\n <TextView\n android:layout_width=\"match_parent\"\n android:layout_height=\"wrap_content\"\n android:padding=\"5dp\"\n android:text=\"TestTextView\\nTestTextView\\nTestTextView\\nTestTextView\\nTestTextView\\nTestTextView\\nTestTextView\\nTestTextView\"/>\n </android.support.wearable.view.",
"CardFrame>\n</android.support.wearable.view.",
"CardScrollView>\n\nA:\n\nCardScrollView does not seem to be made to working like ScrollView. ",
"\nYou can just wrap CardFrame with ScrollView.",
"\n<ScrollView xmlns:android=\"http://schemas.android.com/apk/res/android\"\n android:layout_width=\"match_parent\"\n android:layout_height=\"match_parent\">\n\n <FrameLayout\n android:layout_width=\"match_parent\"\n android:layout_height=\"wrap_content\"\n android:orientation=\"vertical\">\n\n <android.support.wearable.view.",
"CardFrame\n android:id=\"@+id/cardFrame\"\n android:layout_width=\"match_parent\"\n android:layout_height=\"wrap_content\">\n\n <TextView\n android:id=\"@+id/textView\"\n android:layout_width=\"wrap_content\"\n android:layout_height=\"wrap_content\" />\n\n </android.support.wearable.view.",
"CardFrame>\n\n </FrameLayout>\n\n</ScrollView>\n\n"
] |
{
"pile_set_name": "StackExchange"
}
|
[
0.000916278746444732,
0.0006460970616899431,
0.0008789112907834351,
0.0007102597155608237,
0.0009387929458171129,
0.0006747465231455863,
0.0008288270910270512,
0.0007524617831222713,
0.0007824096246622503,
0.00058320676907897,
0.0006584454094991088,
0.0006866224575787783,
0.0007479548803530633
] | 0.000754 | 13 |
[
"Facebook Blogging\n\nEdward Hugh has a lively and enjoyable Facebook community where he publishes frequent breaking news economics links and short updates. ",
"If you would like to receive these updates on a regular basis and join the debate please invite Edward as a friend by clicking the Facebook link at the top of the right sidebar.",
"\n\nWednesday, November 08, 2006\n\nFirst Impressions\n\nWell since I need to be pretty up front in saying that at this point in time I have next to no specific idea about or feeling for the Hungarian economy, perhaps you will ask, and you would have every right to do this, what the hell am I doing here. ",
"This would be a good question.",
"\n\nAll I can say is that I have some very general understanding of macroeconomic processes, and how things work, and that I am struck by the situation in Hungary, by how complex it is, by how many people's aspirations are tied with in the hope that all will be well as it ends well, and how difficult it is to foresee that sort of outcome.",
"\n\nMaybe at this point I should put on my CD of Bartok's quartets (actually at this point they have a piano concierto from Shostakovitch on the radio, I certainly hope that isn't a portend).",
"\n\nHungary's government has decided on Wednesday that corporates will be allowed to write off their research and development expenses from the tax base of the “solidarity tax\" that was implemented on 1 September, government spokeswoman Emese Danks told a press conference on Wednesday.",
"\n\nThe decision to go soft on the new tax was triggered by Audi's recent decision to suspend investments in Hungary until it comes to terms on the solidarity tax and other tax-related issues with the cabinet.",
"\n\nThe 4% solidarity tax is payable on top of a 16% corporate tax and its base is the pre-tax profit.",
"\n\nHowever, Audi, which spends about HUF 250 bn on investments in Hungary a year, is exempt from corporate tax until 2011.",
"\n\nNow the base of both the 16% corporate tax and the 4% solidarity tax will be a tax base that excludes R&D spending.",
"\n\nThe cabinet had earlier said the solidarity tax would bring HUF 150 billion to state coffers, which figure has today been revised upward to HUF 170 billion. ",
"Taking last year's R&D spending into consideration (cc. ",
"HUF 100 bn), the allowance will cost the budget some HUF 5 billion next year.",
"\n\nNow this decision is certainly a stand-down for PM Gyurcsány who had been insisting on the full application of the tax, and it isn't clear to me, as a newcomer to all this, just what relations are like between Gyurcsány and his economics minister János Kóka, doubtless i will get hold of this as I advance. ",
"OK, that will do to get me started.",
"\n\nEdward's Catalan Blog\n\nDaniel Antal, Budapest\n\nHungarian Spectrum\n\nEU8 Related Weblogs\n\nAbout\n\nEdward 'the bonobo' is a Catalan economist of British extraction based in Barcelona. ",
"By inclination he is a macro economist, but his obsession with trying to understand the economic impact of demographic changes has often taken him far from home, off and away from the more tranquil and placid pastures of the dismal science, into the bracken and thicket of demography, anthropology, biology, sociology and systems theory. ",
"All of which has lead him to ask himself whether Thomas Wolfe was not in fact right when he asserted that the fact of the matter is \"you can never go home again\".",
"\n\nHe is currently working on a book with the provisional working title \"Population, the Ultimate Non-renewable Resource\".",
"\n\nApart from his participation in A Fistful of Euros, Edward also writes regularly for the demography blog Demography Matters. ",
"He also contributes to the Indian Economy blog . ",
"His personal weblog is Bonobo Land . ",
"Edward's website can be found at EdwardHugh.net."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0005819116486236453,
0.0005411796737462282,
0.030806221067905426,
0.0005660486058332026,
0.000545561546459794,
0.0006701892707496881,
0.0005725116352550685,
0.0005572808440774679,
0.0009032698580995202,
0.0006667744601145387,
0.0009092162363231182,
0.0008073558565229177,
0.0006554079591296613,
0.0010737087577581406,
0.0006786236190237105,
0.0007420328911393881,
0.0006103299674578011,
0.0009317455696873367,
0.002425598446279764,
0.0005974455852992833,
0.000702976540196687,
0.0006080938619561493,
0.0007621202967129648,
0.0006383063155226409
] | 0.002023 | 24 |
[
"package kingpin\n\nimport (\n\t\"encoding/hex\"\n\t\"fmt\"\n\t\"net\"\n\t\"regexp\"\n\t\"strconv\"\n\t\"time\"\n)\n\n// This file is autogenerated by \"go generate .\". ",
"Do not modify.",
"\n\n// -- bool Value\ntype boolValue struct{ v *bool }\n\nfunc newBoolValue(p *bool) *boolValue {\n\treturn &boolValue{p}\n}\n\nfunc (f *boolValue) Set(s string) error {\n\tv, err := strconv.",
"ParseBool(s)\n\tif err == nil {\n\t\t*f.v = (bool)(v)\n\t}\n\treturn err\n}\n\nfunc (f *boolValue) Get() interface{} { return (bool)(*f.v) }\n\nfunc (f *boolValue) String() string { return fmt.",
"Sprintf(\"%v\", *f.v) }\n\n// Bool parses the next command-line value as bool.",
"\nfunc (p *parserMixin) Bool() (target *bool) {\n\ttarget = new(bool)\n\tp.BoolVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) BoolVar(target *bool) {\n\tp.SetValue(newBoolValue(target))\n}\n\n// BoolList accumulates bool values into a slice.",
"\nfunc (p *parserMixin) BoolList() (target *[]bool) {\n\ttarget = new([]bool)\n\tp.BoolListVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) BoolListVar(target *[]bool) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newBoolValue(v.(*bool))\n\t}))\n}\n\n// -- string Value\ntype stringValue struct{ v *string }\n\nfunc newStringValue(p *string) *stringValue {\n\treturn &stringValue{p}\n}\n\nfunc (f *stringValue) Set(s string) error {\n\tv, err := s, error(nil)\n\tif err == nil {\n\t\t*f.v = (string)(v)\n\t}\n\treturn err\n}\n\nfunc (f *stringValue) Get() interface{} { return (string)(*f.v) }\n\nfunc (f *stringValue) String() string { return string(*f.v) }\n\n// String parses the next command-line value as string.",
"\nfunc (p *parserMixin) String() (target *string) {\n\ttarget = new(string)\n\tp.StringVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) StringVar(target *string) {\n\tp.SetValue(newStringValue(target))\n}\n\n// Strings accumulates string values into a slice.",
"\nfunc (p *parserMixin) Strings() (target *[]string) {\n\ttarget = new([]string)\n\tp.StringsVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) StringsVar(target *[]string) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newStringValue(v.(*string))\n\t}))\n}\n\n// -- uint Value\ntype uintValue struct{ v *uint }\n\nfunc newUintValue(p *uint) *uintValue {\n\treturn &uintValue{p}\n}\n\nfunc (f *uintValue) Set(s string) error {\n\tv, err := strconv.",
"ParseUint(s, 0, 64)\n\tif err == nil {\n\t\t*f.v = (uint)(v)\n\t}\n\treturn err\n}\n\nfunc (f *uintValue) Get() interface{} { return (uint)(*f.v) }\n\nfunc (f *uintValue) String() string { return fmt.",
"Sprintf(\"%v\", *f.v) }\n\n// Uint parses the next command-line value as uint.",
"\nfunc (p *parserMixin) Uint() (target *uint) {\n\ttarget = new(uint)\n\tp.UintVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) UintVar(target *uint) {\n\tp.SetValue(newUintValue(target))\n}\n\n// Uints accumulates uint values into a slice.",
"\nfunc (p *parserMixin) Uints() (target *[]uint) {\n\ttarget = new([]uint)\n\tp.UintsVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) UintsVar(target *[]uint) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newUintValue(v.(*uint))\n\t}))\n}\n\n// -- uint8 Value\ntype uint8Value struct{ v *uint8 }\n\nfunc newUint8Value(p *uint8) *uint8Value {\n\treturn &uint8Value{p}\n}\n\nfunc (f *uint8Value) Set(s string) error {\n\tv, err := strconv.",
"ParseUint(s, 0, 8)\n\tif err == nil {\n\t\t*f.v = (uint8)(v)\n\t}\n\treturn err\n}\n\nfunc (f *uint8Value) Get() interface{} { return (uint8)(*f.v) }\n\nfunc (f *uint8Value) String() string { return fmt.",
"Sprintf(\"%v\", *f.v) }\n\n// Uint8 parses the next command-line value as uint8.",
"\nfunc (p *parserMixin) Uint8() (target *uint8) {\n\ttarget = new(uint8)\n\tp.Uint8Var(target)\n\treturn\n}\n\nfunc (p *parserMixin) Uint8Var(target *uint8) {\n\tp.SetValue(newUint8Value(target))\n}\n\n// Uint8List accumulates uint8 values into a slice.",
"\nfunc (p *parserMixin) Uint8List() (target *[]uint8) {\n\ttarget = new([]uint8)\n\tp.Uint8ListVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) Uint8ListVar(target *[]uint8) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newUint8Value(v.(*uint8))\n\t}))\n}\n\n// -- uint16 Value\ntype uint16Value struct{ v *uint16 }\n\nfunc newUint16Value(p *uint16) *uint16Value {\n\treturn &uint16Value{p}\n}\n\nfunc (f *uint16Value) Set(s string) error {\n\tv, err := strconv.",
"ParseUint(s, 0, 16)\n\tif err == nil {\n\t\t*f.v = (uint16)(v)\n\t}\n\treturn err\n}\n\nfunc (f *uint16Value) Get() interface{} { return (uint16)(*f.v) }\n\nfunc (f *uint16Value) String() string { return fmt.",
"Sprintf(\"%v\", *f.v) }\n\n// Uint16 parses the next command-line value as uint16.",
"\nfunc (p *parserMixin) Uint16() (target *uint16) {\n\ttarget = new(uint16)\n\tp.Uint16Var(target)\n\treturn\n}\n\nfunc (p *parserMixin) Uint16Var(target *uint16) {\n\tp.SetValue(newUint16Value(target))\n}\n\n// Uint16List accumulates uint16 values into a slice.",
"\nfunc (p *parserMixin) Uint16List() (target *[]uint16) {\n\ttarget = new([]uint16)\n\tp.Uint16ListVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) Uint16ListVar(target *[]uint16) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newUint16Value(v.(*uint16))\n\t}))\n}\n\n// -- uint32 Value\ntype uint32Value struct{ v *uint32 }\n\nfunc newUint32Value(p *uint32) *uint32Value {\n\treturn &uint32Value{p}\n}\n\nfunc (f *uint32Value) Set(s string) error {\n\tv, err := strconv.",
"ParseUint(s, 0, 32)\n\tif err == nil {\n\t\t*f.v = (uint32)(v)\n\t}\n\treturn err\n}\n\nfunc (f *uint32Value) Get() interface{} { return (uint32)(*f.v) }\n\nfunc (f *uint32Value) String() string { return fmt.",
"Sprintf(\"%v\", *f.v) }\n\n// Uint32 parses the next command-line value as uint32.",
"\nfunc (p *parserMixin) Uint32() (target *uint32) {\n\ttarget = new(uint32)\n\tp.Uint32Var(target)\n\treturn\n}\n\nfunc (p *parserMixin) Uint32Var(target *uint32) {\n\tp.SetValue(newUint32Value(target))\n}\n\n// Uint32List accumulates uint32 values into a slice.",
"\nfunc (p *parserMixin) Uint32List() (target *[]uint32) {\n\ttarget = new([]uint32)\n\tp.Uint32ListVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) Uint32ListVar(target *[]uint32) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newUint32Value(v.(*uint32))\n\t}))\n}\n\n// -- uint64 Value\ntype uint64Value struct{ v *uint64 }\n\nfunc newUint64Value(p *uint64) *uint64Value {\n\treturn &uint64Value{p}\n}\n\nfunc (f *uint64Value) Set(s string) error {\n\tv, err := strconv.",
"ParseUint(s, 0, 64)\n\tif err == nil {\n\t\t*f.v = (uint64)(v)\n\t}\n\treturn err\n}\n\nfunc (f *uint64Value) Get() interface{} { return (uint64)(*f.v) }\n\nfunc (f *uint64Value) String() string { return fmt.",
"Sprintf(\"%v\", *f.v) }\n\n// Uint64 parses the next command-line value as uint64.",
"\nfunc (p *parserMixin) Uint64() (target *uint64) {\n\ttarget = new(uint64)\n\tp.Uint64Var(target)\n\treturn\n}\n\nfunc (p *parserMixin) Uint64Var(target *uint64) {\n\tp.SetValue(newUint64Value(target))\n}\n\n// Uint64List accumulates uint64 values into a slice.",
"\nfunc (p *parserMixin) Uint64List() (target *[]uint64) {\n\ttarget = new([]uint64)\n\tp.Uint64ListVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) Uint64ListVar(target *[]uint64) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newUint64Value(v.(*uint64))\n\t}))\n}\n\n// -- int Value\ntype intValue struct{ v *int }\n\nfunc newIntValue(p *int) *intValue {\n\treturn &intValue{p}\n}\n\nfunc (f *intValue) Set(s string) error {\n\tv, err := strconv.",
"ParseFloat(s, 64)\n\tif err == nil {\n\t\t*f.v = (int)(v)\n\t}\n\treturn err\n}\n\nfunc (f *intValue) Get() interface{} { return (int)(*f.v) }\n\nfunc (f *intValue) String() string { return fmt.",
"Sprintf(\"%v\", *f.v) }\n\n// Int parses the next command-line value as int.",
"\nfunc (p *parserMixin) Int() (target *int) {\n\ttarget = new(int)\n\tp.IntVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) IntVar(target *int) {\n\tp.SetValue(newIntValue(target))\n}\n\n// Ints accumulates int values into a slice.",
"\nfunc (p *parserMixin) Ints() (target *[]int) {\n\ttarget = new([]int)\n\tp.IntsVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) IntsVar(target *[]int) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newIntValue(v.(*int))\n\t}))\n}\n\n// -- int8 Value\ntype int8Value struct{ v *int8 }\n\nfunc newInt8Value(p *int8) *int8Value {\n\treturn &int8Value{p}\n}\n\nfunc (f *int8Value) Set(s string) error {\n\tv, err := strconv.",
"ParseInt(s, 0, 8)\n\tif err == nil {\n\t\t*f.v = (int8)(v)\n\t}\n\treturn err\n}\n\nfunc (f *int8Value) Get() interface{} { return (int8)(*f.v) }\n\nfunc (f *int8Value) String() string { return fmt.",
"Sprintf(\"%v\", *f.v) }\n\n// Int8 parses the next command-line value as int8.",
"\nfunc (p *parserMixin) Int8() (target *int8) {\n\ttarget = new(int8)\n\tp.Int8Var(target)\n\treturn\n}\n\nfunc (p *parserMixin) Int8Var(target *int8) {\n\tp.SetValue(newInt8Value(target))\n}\n\n// Int8List accumulates int8 values into a slice.",
"\nfunc (p *parserMixin) Int8List() (target *[]int8) {\n\ttarget = new([]int8)\n\tp.Int8ListVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) Int8ListVar(target *[]int8) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newInt8Value(v.(*int8))\n\t}))\n}\n\n// -- int16 Value\ntype int16Value struct{ v *int16 }\n\nfunc newInt16Value(p *int16) *int16Value {\n\treturn &int16Value{p}\n}\n\nfunc (f *int16Value) Set(s string) error {\n\tv, err := strconv.",
"ParseInt(s, 0, 16)\n\tif err == nil {\n\t\t*f.v = (int16)(v)\n\t}\n\treturn err\n}\n\nfunc (f *int16Value) Get() interface{} { return (int16)(*f.v) }\n\nfunc (f *int16Value) String() string { return fmt.",
"Sprintf(\"%v\", *f.v) }\n\n// Int16 parses the next command-line value as int16.",
"\nfunc (p *parserMixin) Int16() (target *int16) {\n\ttarget = new(int16)\n\tp.Int16Var(target)\n\treturn\n}\n\nfunc (p *parserMixin) Int16Var(target *int16) {\n\tp.SetValue(newInt16Value(target))\n}\n\n// Int16List accumulates int16 values into a slice.",
"\nfunc (p *parserMixin) Int16List() (target *[]int16) {\n\ttarget = new([]int16)\n\tp.Int16ListVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) Int16ListVar(target *[]int16) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newInt16Value(v.(*int16))\n\t}))\n}\n\n// -- int32 Value\ntype int32Value struct{ v *int32 }\n\nfunc newInt32Value(p *int32) *int32Value {\n\treturn &int32Value{p}\n}\n\nfunc (f *int32Value) Set(s string) error {\n\tv, err := strconv.",
"ParseInt(s, 0, 32)\n\tif err == nil {\n\t\t*f.v = (int32)(v)\n\t}\n\treturn err\n}\n\nfunc (f *int32Value) Get() interface{} { return (int32)(*f.v) }\n\nfunc (f *int32Value) String() string { return fmt.",
"Sprintf(\"%v\", *f.v) }\n\n// Int32 parses the next command-line value as int32.",
"\nfunc (p *parserMixin) Int32() (target *int32) {\n\ttarget = new(int32)\n\tp.Int32Var(target)\n\treturn\n}\n\nfunc (p *parserMixin) Int32Var(target *int32) {\n\tp.SetValue(newInt32Value(target))\n}\n\n// Int32List accumulates int32 values into a slice.",
"\nfunc (p *parserMixin) Int32List() (target *[]int32) {\n\ttarget = new([]int32)\n\tp.Int32ListVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) Int32ListVar(target *[]int32) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newInt32Value(v.(*int32))\n\t}))\n}\n\n// -- int64 Value\ntype int64Value struct{ v *int64 }\n\nfunc newInt64Value(p *int64) *int64Value {\n\treturn &int64Value{p}\n}\n\nfunc (f *int64Value) Set(s string) error {\n\tv, err := strconv.",
"ParseInt(s, 0, 64)\n\tif err == nil {\n\t\t*f.v = (int64)(v)\n\t}\n\treturn err\n}\n\nfunc (f *int64Value) Get() interface{} { return (int64)(*f.v) }\n\nfunc (f *int64Value) String() string { return fmt.",
"Sprintf(\"%v\", *f.v) }\n\n// Int64 parses the next command-line value as int64.",
"\nfunc (p *parserMixin) Int64() (target *int64) {\n\ttarget = new(int64)\n\tp.Int64Var(target)\n\treturn\n}\n\nfunc (p *parserMixin) Int64Var(target *int64) {\n\tp.SetValue(newInt64Value(target))\n}\n\n// Int64List accumulates int64 values into a slice.",
"\nfunc (p *parserMixin) Int64List() (target *[]int64) {\n\ttarget = new([]int64)\n\tp.Int64ListVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) Int64ListVar(target *[]int64) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newInt64Value(v.(*int64))\n\t}))\n}\n\n// -- float64 Value\ntype float64Value struct{ v *float64 }\n\nfunc newFloat64Value(p *float64) *float64Value {\n\treturn &float64Value{p}\n}\n\nfunc (f *float64Value) Set(s string) error {\n\tv, err := strconv.",
"ParseFloat(s, 64)\n\tif err == nil {\n\t\t*f.v = (float64)(v)\n\t}\n\treturn err\n}\n\nfunc (f *float64Value) Get() interface{} { return (float64)(*f.v) }\n\nfunc (f *float64Value) String() string { return fmt.",
"Sprintf(\"%v\", *f.v) }\n\n// Float64 parses the next command-line value as float64.",
"\nfunc (p *parserMixin) Float64() (target *float64) {\n\ttarget = new(float64)\n\tp.Float64Var(target)\n\treturn\n}\n\nfunc (p *parserMixin) Float64Var(target *float64) {\n\tp.SetValue(newFloat64Value(target))\n}\n\n// Float64List accumulates float64 values into a slice.",
"\nfunc (p *parserMixin) Float64List() (target *[]float64) {\n\ttarget = new([]float64)\n\tp.Float64ListVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) Float64ListVar(target *[]float64) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newFloat64Value(v.(*float64))\n\t}))\n}\n\n// -- float32 Value\ntype float32Value struct{ v *float32 }\n\nfunc newFloat32Value(p *float32) *float32Value {\n\treturn &float32Value{p}\n}\n\nfunc (f *float32Value) Set(s string) error {\n\tv, err := strconv.",
"ParseFloat(s, 32)\n\tif err == nil {\n\t\t*f.v = (float32)(v)\n\t}\n\treturn err\n}\n\nfunc (f *float32Value) Get() interface{} { return (float32)(*f.v) }\n\nfunc (f *float32Value) String() string { return fmt.",
"Sprintf(\"%v\", *f.v) }\n\n// Float32 parses the next command-line value as float32.",
"\nfunc (p *parserMixin) Float32() (target *float32) {\n\ttarget = new(float32)\n\tp.Float32Var(target)\n\treturn\n}\n\nfunc (p *parserMixin) Float32Var(target *float32) {\n\tp.SetValue(newFloat32Value(target))\n}\n\n// Float32List accumulates float32 values into a slice.",
"\nfunc (p *parserMixin) Float32List() (target *[]float32) {\n\ttarget = new([]float32)\n\tp.Float32ListVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) Float32ListVar(target *[]float32) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newFloat32Value(v.(*float32))\n\t}))\n}\n\n// DurationList accumulates time.",
"Duration values into a slice.",
"\nfunc (p *parserMixin) DurationList() (target *[]time.",
"Duration) {\n\ttarget = new([]time.",
"Duration)\n\tp.DurationListVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) DurationListVar(target *[]time.",
"Duration) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newDurationValue(v.(*time.",
"Duration))\n\t}))\n}\n\n// IPList accumulates net.",
"IP values into a slice.",
"\nfunc (p *parserMixin) IPList() (target *[]net.",
"IP) {\n\ttarget = new([]net.",
"IP)\n\tp.IPListVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) IPListVar(target *[]net.",
"IP) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newIPValue(v.(*net.",
"IP))\n\t}))\n}\n\n// TCPList accumulates *net.",
"TCPAddr values into a slice.",
"\nfunc (p *parserMixin) TCPList() (target *[]*net.",
"TCPAddr) {\n\ttarget = new([]*net.",
"TCPAddr)\n\tp.TCPListVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) TCPListVar(target *[]*net.",
"TCPAddr) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newTCPAddrValue(v.(**net.",
"TCPAddr))\n\t}))\n}\n\n// ExistingFiles accumulates string values into a slice.",
"\nfunc (p *parserMixin) ExistingFiles() (target *[]string) {\n\ttarget = new([]string)\n\tp.ExistingFilesVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) ExistingFilesVar(target *[]string) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newExistingFileValue(v.(*string))\n\t}))\n}\n\n// ExistingDirs accumulates string values into a slice.",
"\nfunc (p *parserMixin) ExistingDirs() (target *[]string) {\n\ttarget = new([]string)\n\tp.ExistingDirsVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) ExistingDirsVar(target *[]string) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newExistingDirValue(v.(*string))\n\t}))\n}\n\n// ExistingFilesOrDirs accumulates string values into a slice.",
"\nfunc (p *parserMixin) ExistingFilesOrDirs() (target *[]string) {\n\ttarget = new([]string)\n\tp.ExistingFilesOrDirsVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) ExistingFilesOrDirsVar(target *[]string) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newExistingFileOrDirValue(v.(*string))\n\t}))\n}\n\n// -- *regexp.",
"Regexp Value\ntype regexpValue struct{ v **regexp.",
"Regexp }\n\nfunc newRegexpValue(p **regexp.",
"Regexp) *regexpValue {\n\treturn ®expValue{p}\n}\n\nfunc (f *regexpValue) Set(s string) error {\n\tv, err := regexp.",
"Compile(s)\n\tif err == nil {\n\t\t*f.v = (*regexp.",
"Regexp)(v)\n\t}\n\treturn err\n}\n\nfunc (f *regexpValue) Get() interface{} { return (*regexp.",
"Regexp)(*f.v) }\n\nfunc (f *regexpValue) String() string { return fmt.",
"Sprintf(\"%v\", *f.v) }\n\n// Regexp parses the next command-line value as *regexp.",
"Regexp.",
"\nfunc (p *parserMixin) Regexp() (target **regexp.",
"Regexp) {\n\ttarget = new(*regexp.",
"Regexp)\n\tp.RegexpVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) RegexpVar(target **regexp.",
"Regexp) {\n\tp.SetValue(newRegexpValue(target))\n}\n\n// RegexpList accumulates *regexp.",
"Regexp values into a slice.",
"\nfunc (p *parserMixin) RegexpList() (target *[]*regexp.",
"Regexp) {\n\ttarget = new([]*regexp.",
"Regexp)\n\tp.RegexpListVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) RegexpListVar(target *[]*regexp.",
"Regexp) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newRegexpValue(v.(**regexp.",
"Regexp))\n\t}))\n}\n\n// -- net.",
"IP Value\ntype resolvedIPValue struct{ v *net.",
"IP }\n\nfunc newResolvedIPValue(p *net.",
"IP) *resolvedIPValue {\n\treturn &resolvedIPValue{p}\n}\n\nfunc (f *resolvedIPValue) Set(s string) error {\n\tv, err := resolveHost(s)\n\tif err == nil {\n\t\t*f.v = (net.",
"IP)(v)\n\t}\n\treturn err\n}\n\nfunc (f *resolvedIPValue) Get() interface{} { return (net.",
"IP)(*f.v) }\n\nfunc (f *resolvedIPValue) String() string { return fmt.",
"Sprintf(\"%v\", *f.v) }\n\n// Resolve a hostname or IP to an IP.",
"\nfunc (p *parserMixin) ResolvedIP() (target *net.",
"IP) {\n\ttarget = new(net.",
"IP)\n\tp.ResolvedIPVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) ResolvedIPVar(target *net.",
"IP) {\n\tp.SetValue(newResolvedIPValue(target))\n}\n\n// ResolvedIPList accumulates net.",
"IP values into a slice.",
"\nfunc (p *parserMixin) ResolvedIPList() (target *[]net.",
"IP) {\n\ttarget = new([]net.",
"IP)\n\tp.ResolvedIPListVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) ResolvedIPListVar(target *[]net.",
"IP) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newResolvedIPValue(v.(*net.",
"IP))\n\t}))\n}\n\n// -- []byte Value\ntype hexBytesValue struct{ v *[]byte }\n\nfunc newHexBytesValue(p *[]byte) *hexBytesValue {\n\treturn &hexBytesValue{p}\n}\n\nfunc (f *hexBytesValue) Set(s string) error {\n\tv, err := hex.",
"DecodeString(s)\n\tif err == nil {\n\t\t*f.v = ([]byte)(v)\n\t}\n\treturn err\n}\n\nfunc (f *hexBytesValue) Get() interface{} { return ([]byte)(*f.v) }\n\nfunc (f *hexBytesValue) String() string { return fmt.",
"Sprintf(\"%v\", *f.v) }\n\n// Bytes as a hex string.",
"\nfunc (p *parserMixin) HexBytes() (target *[]byte) {\n\ttarget = new([]byte)\n\tp.HexBytesVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) HexBytesVar(target *[]byte) {\n\tp.SetValue(newHexBytesValue(target))\n}\n\n// HexBytesList accumulates []byte values into a slice.",
"\nfunc (p *parserMixin) HexBytesList() (target *[][]byte) {\n\ttarget = new([][]byte)\n\tp.HexBytesListVar(target)\n\treturn\n}\n\nfunc (p *parserMixin) HexBytesListVar(target *[][]byte) {\n\tp.SetValue(newAccumulator(target, func(v interface{}) Value {\n\t\treturn newHexBytesValue(v.(*[]byte))\n\t}))\n}\n"
] |
{
"pile_set_name": "Github"
}
|
[
0.0006647541304118931,
0.001070800586603582,
0.04633273556828499,
0.12071548402309418,
0.0036532310768961906,
0.0047792778350412846,
0.018875718116760254,
0.0009708197321742773,
0.0014000090304762125,
0.02048404887318611,
0.001295194262638688,
0.0008520034025423229,
0.001216556061990559,
0.025463733822107315,
0.0011259934399276972,
0.0008239441085606813,
0.0016306077595800161,
0.02704741433262825,
0.0010376193094998598,
0.0007665841840207577,
0.002023225650191307,
0.027041524648666382,
0.001161863561719656,
0.0009129085228778422,
0.002168856794014573,
0.023557424545288086,
0.0011198587017133832,
0.0009186333045363426,
0.001804375322535634,
0.01071926299482584,
0.0012729560257866979,
0.001119959051720798,
0.0032822166103869677,
0.02297470159828663,
0.0012612417340278625,
0.0009786600712686777,
0.0029648973140865564,
0.02410605177283287,
0.0011793930316343904,
0.0009850782807916403,
0.003634912194684148,
0.0244474895298481,
0.0013180834939703345,
0.001108091906644404,
0.0035599099937826395,
0.021136000752449036,
0.0012609619880095124,
0.0010517766932025552,
0.002259747125208378,
0.00913997646421194,
0.0010956085752695799,
0.0008304492803290486,
0.0020200922153890133,
0.010951568372547626,
0.0011497741797938943,
0.0008425266714766622,
0.0008788701961748302,
0.0006232083542272449,
0.0006888082716614008,
0.0006431996007449925,
0.0007182874251157045,
0.0006583252688869834,
0.0008439878583885729,
0.0007046235259622335,
0.0009085615747608244,
0.0008121816208586097,
0.0011505510192364454,
0.0008220385061576962,
0.0012331100879237056,
0.0007754131802357733,
0.0010772162349894643,
0.0008998853154480457,
0.0009637209586799145,
0.0008784641977399588,
0.0008355976897291839,
0.0009391994681209326,
0.0009890679502859712,
0.0011162678711116314,
0.0032280811574310064,
0.004102805629372597,
0.008664648048579693,
0.015032368712127209,
0.01995346136391163,
0.009245933033525944,
0.0012121022446081042,
0.0007795810815878212,
0.0009092006948776543,
0.0009637745679356158,
0.0011782858055084944,
0.0007992640603333712,
0.000737772963475436,
0.000965220620855689,
0.0009928372455760837,
0.001037467154674232,
0.0008137546246871352,
0.0010201595723628998,
0.0011924845166504383,
0.0010897833853960037,
0.002392841735854745,
0.006875739898532629,
0.004358063917607069,
0.0009393650689162314,
0.0007660616538487375,
0.0009135350119322538,
0.0009074424160644412,
0.0007337798015214503,
0.0007046235259622335,
0.0007736432598903775,
0.0008121816208586097,
0.0008509405888617039,
0.0007927416008897126,
0.0038875907193869352,
0.022619158029556274,
0.004614129662513733,
0.0008425631676800549,
0.0009844430023804307
] | 0.005564 | 116 |
[
"'Agents of SHIELD': How many people have watched that pilot so far?",
"\n\nMarvel's Agents of S.H.I.E.L.D. type TV Show network ABC genre Superhero Where to watch Close Streaming Options\n\nMore than you think.",
"\n\nABC released detailed breakdown of how many people have seen the first episode of Marvel’s Agents of S.H.I.E.L.D. (seriously, I just need to create a Mac keyboard shortcut for that title with all those periods in it).",
"\n\nYou’ll recall — because you read every ratings story we post, right? — ",
"12.1 million people watched the premiere last Tuesday, based on the adjusted afternoon nationals from Nielsen.",
"\n\nThen another 3.7 million watched on DVR in the next couple days after the premiere.",
"\n\nThen another 4.7 million saw the show’s repeat on Thursday.",
"\n\nAnd another 1.6 million streamed the pilot online.",
"\n\nSo the grand total is now 22.1 million viewers (and that’s still climbing, since there’s still more data coming in). ",
"Stacks up, huh?",
"\n\nGranted, that’s not as impressive as, say, CBS’ Big Bang Theory bagging 20 million viewers within six hours of airing on Thursday night. ",
"But it’s still pretty great and will make things interesting on Wednesday morning with the show’s second episode numbers come in.",
"\n\nHere’s a chart:"
] |
{
"pile_set_name": "OpenWebText2"
}
|
[
0.0006147589883767068,
0.0006017998093739152,
0.0005900771357119083,
0.0006220541545189917,
0.0005828383727930486,
0.0006272588507272303,
0.0007458381005562842,
0.0006169956759549677,
0.0006982203922234476,
0.001006804290227592,
0.0007324499310925603,
0.0005748765543103218,
0.0008372291340492666
] | 0.000681 | 13 |
[
"Elastic instability\n\nElastic instability is a form of instability occurring in elastic systems, such as buckling of beams and plates subject to large compressive loads.",
"\n\nThere are a lot of ways to study this kind of instability. ",
"One of them is to use the method of incremetal deformations based on superposing a small perturbation on an equilibrium solution.",
"\n\nSingle degree of freedom-systems\n\nConsider as a simple example a rigid beam of length L, hinged in one end and free in the other, and having an angular spring attached to the hinged end. ",
"The beam is loaded in the free end by a force F acting in the compressive axial direction of the beam, see the figure to the right.",
"\n\nMoment equilibrium condition\n\nAssuming a clockwise angular deflection , the clockwise moment exerted by the force becomes . ",
"The moment equilibrium equation is given by \n\nwhere is the spring constant of the angular spring (Nm/radian). ",
"Assuming is small enough, implementing the Taylor expansion of the sine function and keeping the two first terms yields \n\nwhich has three solutions, the trivial , and \n\nwhich is imaginary (i.e. not physical) for and real otherwise. ",
"This implies that for small compressive forces, the only equilibrium state is given by , while if the force exceeds the value there is suddenly another mode of deformation possible.",
"\n\nEnergy method\nThe same result can be obtained by considering energy relations. ",
"The energy stored in the angular spring is \n\nand the work done by the force is simply the force multiplied by the vertical displacement of the beam end, which is . ",
"Thus, \n\nThe energy equilibrium condition now yields as before (besides from the trivial ).",
"\n\nStability of the solutions\n\nAny solution is stable iff a small change in the deformation angle results in a reaction moment trying to restore the original angle of deformation. ",
"The net clockwise moment acting on the beam is \n\nAn infinitesimal clockwise change of the deformation angle results in a moment \n\nwhich can be rewritten as \n\nsince due to the moment equilibrium condition. ",
"Now, a solution is stable iff a clockwise change results in a negative change of moment and vice versa. ",
"Thus, the condition for stability becomes\n\nThe solution is stable only for , which is expected. ",
"By expanding the cosine term in the equation, the approximate stability condition is obtained:\n\nfor , which the two other solutions satisfy. ",
"Hence, these solutions are stable.",
"\n\nMultiple degrees of freedom-systems \n\nBy attaching another rigid beam to the original system by means of an angular spring a two degrees of freedom-system is obtained. ",
"Assume for simplicity that the beam lengths and angular springs are equal. ",
"The equilibrium conditions become\n\nwhere and are the angles of the two beams. ",
"Linearizing by assuming these angles are small yields\n\nThe non-trivial solutions to the system is obtained by finding the roots of the determinant of the system matrix, i.e. for \n\nThus, for the two degrees of freedom-system there are two critical values for the applied force F. These correspond to two different modes of deformation which can be computed from the nullspace of the system matrix. ",
"Dividing the equations by yields\n\nFor the lower critical force the ratio is positive and the two beams deflect in the same direction while for the higher force they form a \"banana\" shape. ",
"These two states of deformation represent the buckling mode shapes of the system.",
"\n\nSee also\n\n Buckling\n Cavitation (elastomers)\n Drucker stability\n\nFurther reading\n\nTheory of elastic stability, S. Timoshenko and J. Gere\n\nCategory:Continuum mechanics\nCategory:Structural analysis\nCategory:Mechanics"
] |
{
"pile_set_name": "Wikipedia (en)"
}
|
[
0.0007675164961256087,
0.0007612078916281462,
0.0008063740679062903,
0.000723799632396549,
0.0006936884019523859,
0.00065616425126791,
0.0005918744718655944,
0.0006913592806085944,
0.0006328859599307179,
0.0005774028832092881,
0.0006484118639491498,
0.000588107155635953,
0.0005580558790825307,
0.0006436686962842941,
0.0006167596438899636,
0.0005865333369001746,
0.0007966012926772237,
0.0005775520112365484,
0.0006064920453354716,
0.0006381478160619736,
0.0005851677851751447,
0.0006423478480428457,
0.0006946538924239576,
0.0005886115832254291,
0.001576215960085392
] | 0.00069 | 25 |
[
"High-performance asymmetric supercapacitors based on monodisperse MnO nanocrystals with high energy densities.",
"\nMonodisperse spherical MnO nanocrystals (NCs) with a size of 22.5 nm were synthesized by the thermal decomposition of manganese oleate in the presence of oleic acid and 1-octadecene. ",
"The as-synthesized MnO NCs show superior electrochemical performances with a specific capacitance of 736.4 F g-1 at a current density of 1 A g-1 and retain 93.3% of initial specific capacitance after 5000 cycles. ",
"The MnO NC electrode was successfully assembled in an asymmetric supercapacitor as the cathode with an activated carbon (AC) electrode as the anode. ",
"The as-fabricated device can demonstrate remarkable performance with an energy density of 44.2 W h kg-1, a power density of 900 W kg-1, and excellent cycling stability. ",
"This work provides a new direction for MnO nanomaterials towards high-performance energy storage devices."
] |
{
"pile_set_name": "PubMed Abstracts"
}
|
[
0.0009825049201026559,
0.0008095295052044094,
0.0006507207290269434,
0.0006542316405102611,
0.0005762397777289152,
0.0005719002801924944
] | 0.000708 | 6 |
[
"Erst 1977 erschien der erste Spielautomat der Gauselmann Gruppe mit dem auch gleich das allseits bekannte Symbol die „Merkursonne“ eingeführt wurde. ",
"1984 gelang dann der Durchbruch und Gauselmann übernahm die Marktführung unter den Automatenherstellern. ",
"Es dauerte noch lange, bis das die Spielautomaten ihren Weg ins Internet fanden.",
"\n\nDer erste Spielautomat wurde Ende des 19. ",
"Jahrhunderts in den USA entwickelt und in Betrieb genommen. ",
"Die ersten Spielautomaten, die mechanisch betrieben wurden und über einen Hebelarm an der rechten Seite verfügten, über die die Spiele ausgelöst wurden, trugen im Volksmund den Namen „Einarmiger Bandit“. ",
"Bis das die ersten Spielautomaten in Deutschland erschienen, dauerte es noch eine Weile.",
"\n\nSpielautomaten online spielen\n\nSpielautomaten sind die meist gespielten Spiele\n\nMit Abstand sind die Spielautomaten heute die Casino Spiele die am meisten gespielt werden. ",
"Wenn du genug hast von Einsatzlimits, Öffnungszeiten und ständig besetzen Spielautomaten, dann solltest du online spielen. ",
"In einem online Casino kannst du zu jeder Zeit um echtes Geld spielen und von vielen weiteren Vorteilen profitieren.",
"\n\nEs ist nicht übertrieben, wenn behauptet wird, dass es zahlreiche Vorteile gibt, die dafür sprechen Spielautomaten online zu spielen.",
"\n\ndu kannst rund um die Uhr spielen\n\ndie profitierst von höheren Auszahlungsquoten\n\ndu kannst höhere Gewinne erzielen – teilweise Jackpots in fünf- oder sechsstelliger Höhe\n\ndu kannst mehr einsetzen\n\ndu musst nicht darauf warten, dass dein Lieblingsslot frei wird\n\nZudem übertrifft die Spielauswahl in einem online Casino, die der Spielotheken und Spielhallen bei weitem und zugleich kannst du die Hits aus diesen auch online spielen.",
"\n\nWarum online um Echtgeld an Spielautomaten spielen?",
"\n\nEs ist nicht abzustreiten, dass bei keinem anderen Spiel wie den Spielautomaten, die Auswahl so riesig ist. ",
"Du triffst die Spielautomaten in den unterschiedlichsten Einrichtungen an. ",
"Dazu zählen zum einen die großen und kleinen Casinos und Spielbanken, sowie die Automatencasinos und Spielhallen in den Städten. ",
"Zudem sind die Spiele von Merkur und Novoline in den Gastwirtschaftsbetrieben zu finden und darunter befinden sich Top-Titel wie Book of Ra, Magic Mirror, Super 7 Reels, Jokers Cap oder Lord of the Ocean. ",
"Auch an den stark frequentierten Verkehrspunkten, wie Bahnhöfen, Flughäfen und dergleichen sind die Spielautomaten zu finden. ",
"Kurzum: Offtline gibt es ein riesiges Angebot an Spielautomaten, mit sehr unterschiedlichen Titeln. ",
"Warum solltest du also nicht dieses Angebot nutzen, und anstelle dessen deine Einsätze im Internet tätigen und per Flash Game die Rollen in Schwung bringen?",
"\n\nEiner der Hauptgründe dafür, der vielen weiteren kleineren Argumenten vorausgeht, ist die Auszahlungsquote. ",
"Bei den vorab genannten Einrichtungen kannst du davon ausgehen, dass diese nicht mehr als 60 % beträgt und das bedeutet es werden nur 60 Cent von jedem eingesetzten Euro als Gewinn ausbezahlt. ",
"Schaust du dir die Auszahlungsquoten der online Spielautomaten an, dann erkennst du, dass die Auszahlungsquote in der Regel 95 % und mehr beträgt. ",
"Der Slot Book of Ra, der ein Klassiker ist und zu den beliebtesten Novoline-Slots gehört, bringt sogar eine Auszahlungsquote von 96,99 %. ",
"Andere Automaten reichen sogar teilweise an die 99 % Marke heran.",
"\n\nBeliebte Spiele mit hohen Auszahlungsquoten\n\nLucky Lady’s Charm (Novoline): 97.10 %\n\nBook of Ra Deluxe (Novoline): 96,99 %\n\nTriple Chance (Merkur): 96,11 %\n\nGold of Persia (Merkur): 95,97 %\n\nSouth Park Reel Chaos (NetEnt): 96,80 %\n\nThe Soprans (Playtech): 96,75 %\n\nHolmes and the Stolen Stones (Yggdrasil): 96,20 %\n\nWelche Vorteile bietet die Nutzung der online Slots noch?",
"\n\nMeldest du dich in einem online Casino an, um an den virtuellen Spielautomaten deine Echtgeld-Einsätze vorzunehmen, dann profitierst du nicht nur von einer höheren Auszahlungsquote. ",
"Sondern auch von der Tatsache, dass du jeden Slot gratis im Spielgeld-Modus spielen kannst. ",
"Auch das spricht für die online Casinos. ",
"Den so kannst du aus den hunderten von Spielen deinen Lieblingsslot finden, ohne ein finanzielles Risiko einzugehen oder einen neuen Slot antesten, ob dieser deinen Ansprüchen gerecht wird. ",
"Im Anschluss kannst du dann echtes Geld setzen und Gewinne einfahren.",
"\n\nHinzu kommt ein weiterer Vorteil: Fast jedes online Casino bietet einen Bonus auf die erste Einzahlung an. ",
"Neben diesen gibt es später auch die Möglichkeit von weiteren Bonus-Angeboten und Promotion zu profitieren. ",
"Hinzu kommt noch ein weiterer Fakt: du kannst dich auch an den klassischen Casino-Spielen versuchen, denn die online Casinos bieten auch Roulette, Baccarat, Blackjack und viele weitere Tisch- und Kartenspiele an. ",
"Ziehst du die echte Casino-Atmosphäre vor, dann musst du auch auf diese nicht verzichten, denn viele Anbieter offerieren auch ein Live-Casino. ",
"Bei diesen werden Live-HD-Streams genutzt, um die Spieltische in Echtzeit auf deinen Bildschirm zu bringen.",
"\n\nGewinne erzielen: Automaten mit Jackpot oder ohne nutzen?",
"\n\nDiese Frage kann nicht pauschal beantwortet werden. ",
"Denn es kommt ganz darauf an, was du einsetzen möchtest und wie dein Wunsch nach Gewinnen aussieht. ",
"Zudem spielt auch der Spaßfaktor eine große Rolle bei der Auswahl des online Spielautomaten. ",
"Spielst du beispielsweise deinen Lieblingsslot der über keinen Jackpot verfügt, aber du dennoch gute Gewinne einstreichst, dann ist das doch klasse, oder? ",
"Doch suchst du den Kick, der unwiderruflich von den Jackpot-Slots ausgeht, und das entsprechende Spiel sagt dir zu, dann ist es nicht verkehrt, diesen zu spielen. ",
"Du musst dir nur über eines im klaren sein: Du kannst deinen kleinen Echtgeld-Einsatz zwar in einen Millionen-Gewinn umwandeln, aber das ist nicht die Regel.",
"\n\nSoll es der Spielautomat mit Jackpot sein, dann stellt sich die Frage, ob diese einen hohen oder niedrigen Jackpot aufweisen soll. ",
"Auch hier ist wieder nur eine pauschale Antwort möglich. ",
"Der Grund ist, dass nicht alle Jackpots gleich wachsen und bei der gleichen Summe ausgezahlt werden. ",
"Es gibt beispielsweise Jackpot-Slots, die bis zu einem vier- oder fünfstelligen Betrag anwachsen und wenn du zu denen zählst, die gern ein- bis zwei Dutzend Tausender gewinnen möchte, dann solltest du in das Spiel einsteigen. ",
"Aber dagegen gibt es noch die Automaten, die dir sogar Millionengewinnen bescheren können, wie beispielsweise der Rekord-Slot Mega Moolah von Microgaming.",
"\n\nWenn du in einen Slot mit Jackpot einsteigst, achte auf die letzten Auszahlungen und entscheide, wann du mit dem Spiel beginnst. ",
"Ein weiterer Slot der das Interesse der Millionärsanwärter auf sich zieht ist Millionaire Genie.",
"\n\nDie Vor- und Nachteile vom Echtgeld Einsatz bei online Slots\n\nEin online Slot ist einfach, zu bedienen, und ist daher für jedermann geeignet. ",
"Je nach Entwickler, Thema und Alter des Spiels bringt ein online Spielautomat eine Menge Spaß und zudem sehr oft einen kleineren oder größeren Gewinn. ",
"Dazu kommen noch die Spiele mit Jackpots oder hohen Hauptgewinnen. ",
"Doch neben den positiven Aspekten gibt es auch wie so oft einige wenige negative Punkte, die es zu bedenken gibt.",
"\n\nDie Vorteile\n\nhöhere Auszahlungsquoten als in den Spielhallen\n\nschnelle Spielrunden für eilige Spieler\n\nJackpot-Spielautomaten die teilweise Millionengewinne in Aussicht stellen\n\nUnterhaltsame Inhalte im Gegensatz zu Karten- und Tischspielen\n\nes sind kaum Kenntnisse notwendig – Strategien sind nicht notwendig\n\nRiesige Auswahl – angesagte Titel bekannter und namhafter Hersteller\n\nsind oftmals mit Spielgeld kostenlos spielbar\n\nder Casino Bonus kann für die Slots eingesetzt werden\n\nUmsätze an den Spielautomaten werden auf die Bonusbedingungen zu 100 % angerechnet.",
"\n\nDie Nachteile\n\nbei längeren Spielen gibt es oftmals nur kleine oder gar keine Gewinne\n\nEs besteht aufgrund der Spielgeschwindigkeit und der nicht notwendigen Kenntnisse Suchtgefahr\n\nEin Slot kann schnell langweilig werden aufgrund des Themas\n\nviele online Casinos arbeiten nur mit einem Anbieter und damit fehlt eine entsprechende Auswahl an aktuell angesagten Titeln\n\nTipps, um online Automatenspiele zu zocken.",
"\n\nAuch wenn es einige negative Punkte gibt, so können diese allerdings bei einem gewissenhaften Spiel in Schach gehalten werden. ",
"Wenn du online an den Spielautomaten um Echtgeld zocken möchtest, dann solltest du die folgenden Tipps beherzigen:\n\nBeachte die Auszahlungsquote: Wie bereits erwähnt, ist die Auswahl riesig, aber viele Spiele ähneln sich auch. ",
"Bei der Auswahl solltest du auf die Auszahlungsrate der Spiele (ATP) achten. ",
"Diese kannst du im Paytable bzw. ",
"den Spielregeln finden. ",
"Dabei sollte der Wert so hoch wie möglich ausfallen. ",
"Im besten Fall über 95 %.",
"\n\nWähle den richtigen Bonus: Geht es um das Freispielen von einem Bonus, dann musst du dir bei den Spielautomaten keine Sorgen machen, denn diese werden stets zu 100 % auf die Umsatzbedingungen angerechnet. ",
"Einzig und allein die Höhe der Umsatzbedingungen ist interessant und die sollten niemals höher als das 50-fache des Bonusbetrages sein.",
"\n\nBudget und Limits festlegen: Da das Spiel an den Spielautomaten sehr schnell ist, kannst du sehr schnell den Überblick verlieren und das nicht nur im Bezug auf die Zeit, sondern auch auf die Gewinne bzw. ",
"Verluste. ",
"Lege vor dem Beginn einer Spielsession genau fest, wie lange du spielen möchtest und wie viel du einsetzen willst. ",
"Beende dein Spiel, wenn das Limit erreicht wurde, und nehme die Gewinne aus der Session mit."
] |
{
"pile_set_name": "OpenWebText2"
}
|
[
0.0694836676120758,
0.011554614640772343,
0.24977602064609528,
0.031170852482318878,
0.002944269450381398,
0.369714617729187,
0.015386869199573994,
0.3465043902397156,
0.06347093731164932,
0.007772011216729879,
0.09825163334608078,
0.45641350746154785,
0.04163748398423195,
0.22001142799854279,
0.03353122994303703,
0.007415593136101961,
0.008051732555031776,
0.06763388216495514,
0.1955646574497223,
0.43763241171836853,
0.009719527326524258,
0.08717969805002213,
0.02880888432264328,
0.0023766898084431887,
0.42208054661750793,
0.014020715840160847,
0.029173584654927254,
0.06082439050078392,
0.013346335850656033,
0.0927155539393425,
0.06445670127868652,
0.019141631200909615,
0.18314413726329803,
0.5478388667106628,
0.044814202934503555,
0.04486272484064102,
0.06308092176914215,
0.6999180912971497,
0.4772505462169647,
0.06454965472221375,
0.16982164978981018,
0.20843924582004547,
0.06992799043655396,
0.6446555852890015,
0.22922977805137634,
0.3685641884803772,
0.30020418763160706,
0.10049047321081161,
0.47866517305374146,
0.05341504514217377,
0.015594976022839546,
0.09308002889156342,
0.7208813428878784,
0.7700130343437195,
0.27324068546295166,
0.09409314393997192,
0.6872910261154175,
0.2409587800502777,
0.03673122450709343,
0.6838706731796265,
0.02056232839822769,
0.05936155095696449,
0.001163580222055316,
0.14630217850208282,
0.03249617666006088,
0.03833246976137161,
0.0018730475567281246,
0.04198508337140083,
0.08334305882453918
] | 0.179244 | 69 |
[
"Influence of baseline airway calibre and pulmonary emphysema on bronchial responsiveness in patients with chronic obstructive pulmonary disease.",
"\nBronchial hyper-responsiveness (BHR) is widely observed in patients with chronic obstructive pulmonary disease (COPD). ",
"However, its clinical significance in COPD has not yet been established. ",
"To determine the factors that influence BHR in COPD, multiple linear regression analysis was used to analyse the relationship between BHR to methacholine and baseline forced expiratory volume in 1 s (FEV1), vital capacity (VC), residual volume (RV)/total lung capacity (TLC), static compliance (Cst), transfer coefficient of the lung (Kco), and the percentage of low attenuation area (%LAA) determined by computed tomographic (CT) scan. ",
"Bronchial responsiveness to methacholine was determined in 63 patients with COPD by the dosimeter method and expressed as PD20FEV1. ",
"Residual volume and TLC were determined by body plethysmography. ",
"The percentage of low attenuation area was defined as the percentage of area less than -960 Hounsfield unit on a CT scan of the thorax. ",
"Forced expiratory volume in 1s (%predicted) and %LAA correlated with BHR (P = 0.023 and 0.020, respectively), while VC (%pred), RV/TLC, Cst and Kco did not. ",
"The coefficient of determination was 0.43. ",
"The regression analysis showed that a 10% increase in FEV1 (%pred) and a 10% decrease in %LAA would increase log(PD20FEV1) by 0.145 and 0.117, respectively. ",
"A considerable proportion of the BHR in COPD is related to functional abnormalities and morphologic changes of emphysema, such as decreased baseline airway calibre and destruction of the lung parenchyma."
] |
{
"pile_set_name": "PubMed Abstracts"
}
|
[
0.0010154475457966328,
0.001424393500201404,
0.000575416546780616,
0.0006158905453048646,
0.0007959745125845075,
0.0006217879708856344,
0.0006988145178183913,
0.0008257412700913846,
0.0006231890292838216,
0.0006988918175920844,
0.0010240195551887155
] | 0.000811 | 11 |
[
"---\nabstract: 'Fighting against computer malware require a mandatory step of reverse engineering. ",
"As soon as the code has been disassemblied/decompiled (including a dynamic analysis step), there is a hope to understand what the malware actually does and to implement a detection mean. ",
"This also applies to protection of software whenever one wishes to analyze them. ",
"In this paper, we show how to amour code in such a way that reserse engineering techniques (static and dymanic) are absolutely impossible by combining malicious cryptography techniques developped in our laboratory and new types of programming (k-ary codes). ",
"Suitable encryption algorithms combined with new cryptanalytic approaches to ease the protection of (malicious or not) binaries, enable to provide both total code armouring and large scale polymorphic features at the same time. ",
"A simple 400 Kb of executable code enables to produce a binary code and around $2^{140}$ mutated forms natively while going far beyond the old concept of decryptor.'",
"\nauthor:\n- Eric Filiol\ntitle: 'Malicious cryptography techniques for unreversable (malicious or not) binaries'\n---\n\nIntroduction\n============\n\nMalicious cryptography and malicious mathematics malicious are an emerging field \\[2, 3,4\\] that finds its origin in the initial work of Young and Yung on the use of asymmetric cryptography in the design of dedicated offensive functions for money extorsion (cryptovirology) \\[5\\]. ",
"But this initial approach is very limited and gives only a very small insight of the way mathematics and cryptography can be actually perverted by malware.",
"\n\nMalicious cryptology and malicious mathematics make in fact explode Young and Yung’s narrow vision. ",
"This results in an unlimited, fascinating yet disturbing field of research and experimentation. ",
"This new domain covers several fields and topics (non-exhaustive list):\n\n- Use of cryptography and mathematics to develop “[*super malware*]{}” ([*über-malware*]{}) which evade any kind of detection by implementing:\n\n - Optimized propagation and attack techniques (e.g. by using biased or specific random number generator) [@roblot].",
"\n\n - Sophisticated self-protection techniques. ",
"The malware code protect itself and its own functional activity by using strong cryptography-based tools [@filiol_eicar2005].",
"\n\n - Code invisibility techniques using testing simulability [@filiol_josse2007].",
"\n\n - Code mutation techniques (by combining the different methods of the previous three items).",
"\n\n- Use of complexity theory or computability theory to design undetectable malware [@filiol_hacklu2008].",
"\n\n- Use of malware to perform applied cryptanalysis operations (theft of secrets elements such as passwords or secret keys, static or on-the-fly modification of systems to reduce their cryptographic strength [@filiol_dct10] ...) by a direct action on the cryptographic algorithm or its environment.",
"\n\n- Design and implementation of encryption systems with hidden mathematical trapdoors. ",
"The knowledge of the trap (by the system designer only) enables to break the system very efficiently. ",
"Despite the fact that system is open and public, the trapdoor must remain undetectable. ",
"This can also apply to the keys themselves in the case of asymmetric cryptography [@Erra_Grenier2009].",
"\n\nTo summarize, we can define malicious cryptography and malicious mathematics as the interconnection and interdependence of computer virology with cryptology and mathematics for their mutual benefit. ",
"The possibilities are virtually infinite. ",
"It is worth mentioning that this vision and the underlying techniques can be translated for “non-malicious” use as the protection of legitimate codes (e.g. for copyright protection purposes) against static (reverse engineering) and dynamic (sandboxing, virtualization) analyses. ",
"If simple and classical obfuscation techniques are bound to fail as proved by Barak and al. [",
"@barak] theoretically (for a rather restricted model of computation), new models of computation [@kary; @gold] and new malware techniques [@bcps] have proved that efficient code protection can be achieved practically.",
"\n\nIf the techniques arising from this new domain are conceptually attractive and potentially powerful, their operational implementation is much more difficult and tricky. ",
"This requires a very good mastery of algorithms and low-level programming. ",
"Above all the prior operational thinking of the attack is unavoidable. ",
"The same code implementing the same techniques in two different contexts will have quite different outcomes. ",
"In particular, we must never forget that the code analysis is static (disassembly/decompilation) but also dynamic (debugging, functional analysis ...). ",
"Code encryption usually protects – possibly – against static code analysis only.",
"\n\nIn this paper we will show how the techniques of malicious cryptography enable to implement total amoring of programs, thus prohibiting any reverse engineering operation. ",
"The main interest of that approach lies in the fact that TRANSEC properties are achieved at the same time. ",
"In other words, the protected binary has the same entropy as any legitimate, unprotected code. ",
"This same technique can also achieve a certain level of polymorphism/metamorphism at the same time: a suitable 59-bit key stream cipher is sufficient to generate up to $2^{140}$ variants very simply. ",
"More interestingly, the old fashioned concept of decryptor which usually constitutes a potential signature and hence a weakness, is totally revisited.",
"\n\nThese techniques have been implemented by the author in the LibThor which has been written and directed by Anthony Desnos [@libthor1; @libthor2].",
"\n\nThe paper is organized as follows. ",
"Section \\[cacm\\] recalls some important aspect about code protection and discusses key points about code armouring and code mutation. ",
"Then Section \\[cs\\] presents the context we consider to apply our technique: we aim at protecting critical portions of code that are first transformed into an intermediate representation (IR), then into the bytecode. ",
"This code itself is that of virtual machine used to provide efficent protection against dynamic analysis. ",
"Section \\[mprng\\] presents the different models and techniques of malicious pseudo-random number generator (PRNG). ",
"Finally Section \\[imp\\_k\\] presents implementation issue which must be considered to achieve protection against both static and dynamic analyses.",
"\n\nCode Armouring and Code Mutation {#cacm}\n================================\n\nWe will not recall in details what these two techniques are. ",
"The reader may consult [@filiol_bk] for a complete presentation with respect to them. ",
"As far as malicious cryptography is concerned, we are just going to discuss some critical points which are bottlenecks in their effective implementation. ",
"Most of the times the attempts to play with these techniques fail due to a clumsy use of cryptography.",
"\n\nCode armouring [@filiol_eicar2005] consists in writing a code so as to delay, complicate or even prevent its analysis. ",
"As for code mutation techniques they strive to limit (polymorphism) or to remove (metamorphism) any fixed component (invariant) from one mutated version to another. ",
"The purpose is to circumvent the notion of antiviral signature, may it be a a simple sequence of contiguous or nor contiguous bytes or meta-structures such as Control Flow Graphs (CFG) and other traces describing the code functional (behavioral) structure.",
"\n\nIn all these cases, the general approach is to encrypt the code to protect or to use special techniques for generating random data. ",
"But encryption and generation of randomness relates to two major practical problems: their characteristic entropy profile and the secret elements (keys) management.",
"\n\nEntropy profile\n---------------\n\nThe whole problem lies in the fact that code armouring and code mutation involve random data. ",
"These must be generated on-the-fly and in the context of metamorphism, the generator itself must be too. ",
"For sake of simplicity, we shall speak of [*Pseudo-Random Number Generator*]{} (PRNG) to describe both a random number generator and an encryption system. ",
"The difference lies in the fact that in the latter case either random data produced from the expansion of the key are combined with the plaintext (stream ciphers) or they are the result of the combination of the key with the plaintext (block ciphers).",
"\n\nThe whole issue lies in the generation of a so-called “good” randomness. ",
"Except that in the context of malicious cryptography [@filiol_bk2], the term “good” does not necessarily correspond to what cryptographers usually mean. ",
"In fact, it is better – yet a simplified but sufficient reduction as a first approximation – to use the concept of entropy [@csw2008]. ",
"In the same way, the term of random data will indifferently describe the random data themselves or the result of encryption.",
"\n\nConsider a (malicious) code as an information source $X$. When parsed, the source outputs characters taking the possible values $x_i \\quad (i = 0,\\ldots , 255)$, each with a probability $p_i = P[X = x_i]$. Then the entropy $H(X)$ of the source is the following sum[^1]: $$H(X) = \\sum_{i = 0}^{255} -p_i\\log_2(p_i)$$ Random data, by nature will exhibit a high entropy value[^2]. ",
"On the contrary, non random data exhibit a low entropy profile (they are easier or less difficult to predict).",
"\n\nFrom the attacker’s point of view[^3] the presence of random data means that something is hidden but he has to make the difference between legitimate data (e.g. use of packers to protect code against piracy) and illegitimate data (e.g. malware code). ",
"In the NATO terminology – at the present time it is the most precise and accurate one as far as InfoSec is concerned– random data relate to a COMSEC ([*COMmunication SECurity*]{}) aspect only.",
"\n\nFor the attacker (automated software or human expert), the problem is twofold: first detect random data parts inside a code and then decrypt them. ",
"In this respect, any code area exhibiting a high entropy profile must be considered as suspicious. ",
"To prevent attention to be focused on those random parts, is it possible to add some TRANSEC ([ *TRANSmission SECurity*]{}) aspect. ",
"The most famous one is steganography but for malware or program protection purposes it is not directly usable (data cannot be directly executed) and we have to find different ways. ",
"The other solution is to use malicious statistics as defined and exposed in [@csw2008].",
"\n\nIt is also possible to break randomness by using noisy encoding techniques like Perseus [@perseus]. ",
"In this case the code remains executable while being protected AND exhibiting low entropy profile at the same time (or the entropy profile of any arbitrary kind of data). ",
"As for an exemple, on a 175-bytes script, the unprotected code has an entropy of 3.90. ",
"An encrypted version (by combining simple transpositions with simple substitutions) of that code has an entropy equal to 5.5. ",
"When protected by Perseus the entropy is about 2.57. ",
"So it is very close to a normal program and thus evade entropy-based detection.",
"\n\nThis applies well on any data used for code mutation (e.g. junk code insertion), including specific subsets of code as CFGs: randomly mutated CFG must exhibit the same profile as any normal CFG would. ",
"Otherwise, considering the COMSEC aspect only is bound to make the code detection very easy.",
"\n\nKey management\n--------------\n\nEncrypting a code or a piece of code implies its preliminary deciphering whenever it is executed. ",
"But in all of the cases – except those involving money extortion introduced Young and Yung [@yy] – the key must be accessible to the code itself to decipher. ",
"Consequently in a way or another it is contained in a more or less obfuscated form inside the code. ",
"Therefore is it accessible to the analyst who will always succeed in finding and accessing it. ",
"Instead of performing cryptanalysis, a simple decoding/deciphering operation is sufficient.",
"\n\nIt is therefore necessary to consider keys that are external to the encrypted code. ",
"Two cases are possible [@filiol_bk2]:\n\n- Environmental key management introduced in [@riordan] and developped in [@filiol_bk2]. ",
"The code gathers information in its execution environment and calculates the key repeatedly. ",
"The correct key will be computed when and only when the suitable conditions will be realized in the code environment – which is usually under the control of the code designer. ",
"The security model should prohibit dictionary attacks or environment reduction attacks (enabling reduced exhaustive search) by the code analyst. ",
"Consequently the analyst must examine the code in an controlled dynamic area (sandbox or virtual machine) and wait until suitable conditions are met without knowing when they will be. ",
"However it is possible to build more operational scenarii for this case and to detect that the code is being analyzed and controlled [@filiol_bk2].",
"\n\n- Use of $k$-ary codes [@karyd; @kary] in which a program is no longer a single monolithic binary entity but a set of binaries and non executable files (working in a serial mode or in a parallel mode) to produce a desired final (malicious or not) action. ",
"Then the analyst has a reduced view on the whole code only since generally he can access a limited subset of this $k$-set. ",
"In the context of (legitimate) code protection, one of the files will be a kernel-land module communicating with a userland code to protect. ",
"The code without the appropriate operating environment – with a user-specific configuration by the administrator – will never work. ",
"This solution has the great advantage of hiding (by outsourcing it), the encryption system itself. ",
"It is one particular instance with respect to this last solution that we present in this paper.",
"\n\nCase Studies {#cs}\n============\n\nIn this section we are going to illustrate our approach with simple but powerful cases. ",
"Without loss of generality and for sake of clarity, we consider a reduced case in which only a very few instructions are protected against any disassembly attempt. ",
"Of course any real case – as we did in LibThor – will consider far more instructions, especially all critical ones (those defining the CFG for instance).",
"\n\nThe working context\n-------------------\n\nThe LibThor library initiated and mainly developped by Anthony Desnos [@libthor1; @libthor2] uses virtual machines (VM) as the core tool. ",
"Virtual Machines offer a powerful protection for an algorithm or anything else that we would like to protect against reverse engineering. ",
"We can have a lot of different VMs piled up (like Russian dolls) into a software. ",
"This technique must be combined with classical techniques since it is just a one more step towards code protection.",
"\n\nIn LibThor, the main goal of VMs is to build a simple code which interprets another one. ",
"The idea is to take a piece of x86 assembly instructions and to have a simple, dynamic, metamorphic VMs very quickly to interpret it. ",
"It is worth stressing on the fact that you can embed different VM into the same program to protect differents parts.",
"\n\nIn Desnos’ LibThor VMs are codes which interpret an intermediate representation (IR) derived from the REIL language [@reil]. ",
"So a translation between x86 assembly instructions and the chosen IR is performed. ",
"The VM is totaly independent from the LIBC or anything like that. ",
"The main goal of the VM is to run an algorithm into an encapsulated object (which can be loaded anywhere; in fact it is a simple shellcode), but the main feature of the VM is to have a different version of it at each generation for a same code so if we want to protect a simple instruction $X$, we can build a new VM every time to protect the same $X$. This new VM will be different from the previous one.",
"\n\nMoreover everything must be dynamic. ",
"Therefore we must have simple working rules:\n\n- the bytecode must be dynamic and hidden by using the malicious cryptography tools (malicious PNRG) we are exposing hereafter;\n\n- information must be dynamic (area for each context, the offset for each register in the context...);\n\n- the code must be transformed with the internal LibThor metamorphic package (in which malicious PNRG can used used too);\n\n- integer constants can be hidden with LibThor internal junk package. ",
"This is very useful to hide opcode values, register offsets or anything which represent any invariant in a program. ",
"Here again malicious PRNGs play a critical role.",
"\n\nDuring the code generation (mutation), we use the LibThor metamorphic package a lot of times on a function. ",
"We have several steps during a generation:\n\n1. ",
" Obfuscation of C source code.",
"\n\n2. ",
" Compiling the C source code of the VM into a library.",
"\n\n3. ",
" Extraction of interesting functions from the library.",
"\n\n4. ",
" Transformation of x86 assembly instructions into IR.",
"\n\n5. ",
" Obfuscation of IR by using metamorphic package.",
"\n\n6. ",
" Transformation of IR into bytecode.",
"\n\n7. ",
" Creation of dynamic functions to handle the bytecode.",
"\n\n8. ",
" Obfuscation of functions by using the metamorphic package. ",
"The mutated code is produced here and the malicious PRNG is mainly involved here.",
"\n\n9. ",
" Assembly of all parts to have a simple shellcode.",
"\n\nOur PRNG is essentially dedicated to the protection of the bytecode which is the final result of the transformation: [X86 ASM]{} $\\rightarrow$ [REIL IR]{} $\\rightarrow$ [bytecode]{}. ",
"Here is an illustrating example:\n\n [X86 ASM] MOV EAX, 0x3 [B803000000]\n [REIL IR] STR (0x3, B4, 1, 0), (EAX, B4, 0, 0)\n [BYTECODES] 0xF1010000 0x40004 0x3 0x0 0x6A\n\nWe have five fields in the bytecode corresponding respectively to :\n\n- [0xF1010000]{}\n\n - [0xF1]{}: the opcode of the instruction (STR),\n\n - [0x01]{}: specifies that it is an integer value,\n\n - [0x00]{}: useless with respect to this instruction,\n\n - [0x00]{}: specifies that it is a register.",
"\n\n- [0x40004]{}\n\n - [0x04]{}: the size of the first operand,\n\n - [0x00]{}: useless with respect this instruction,\n\n - [0x04]{}: the size of the third operand,\n\n- [0x3]{}: direct value of the integer,\n\n- [0x0]{}: useless with respect to this instruction,\n\n- [0x6A]{}: value of the register.",
"\n\nIn this setting the [0x00]{} values (useless with respect this instruction) contribute directly to the TRANSEC aspect. ",
"Now that the working environment is defined, let us explain how a malicious PRNG can efficiently protect the opcode values while generating them dynamically.",
"\n\nMalicious PRNG {#mprng}\n==============\n\nSophisticated polymorphic/metamorphic or obfuscation techniques must rely on PRNG (Pseudo-Random Number Generator). ",
"In the context of this paper, the aim is to generate sequences of random numbers (here bytecode values) on-the-fly while hiding the code behavior.",
"\n\nSequences are precomputed and we have to design a generator which will afterwards output those data. ",
"The idea is that any data produced by the resulting generator will be first used by the code as a valid address, and then will itself seed the PNRG to produce the next random data.",
"\n\nThree cases are to be considered:\n\n1. ",
" the code is built from any arbitrary random sequence;\n\n2. ",
" the sequence is given by a (non yet protected) instance of bytecode and we have to design an instance of PNRG accordingly;\n\n3. ",
" a more interesting problem lies in producing random data that can be somehow interpreted by a PRNG as meaningful instructions like [jump 0x89]{} directly.",
"\n\nThis relates to interesting problems of PRNG cryptanalysis. ",
"We are going to address these three cases.",
"\n\nFrom a general point of view it is necessary to recall that for both three cases the malware author needs reproducible random sequences. ",
"By reproducible (hence the term of pseudo-random), we mean that the malware will replay this sequence to operate its course of actions. ",
"The reproducibility condition implies to consider a [*deterministic Finite-State Machine*]{} (dFSM). ",
"The general scheme of how this dFSM is working is illustrated as follows. ",
"Without the dFSM, any instruction data whenever executed produced a data used the next instruction (e.g. an address, an operand...). ",
"$$I_0 \\rightarrow D_0 \\rightarrow I_1 \\rightarrow D_1 \\ldots \\rightarrow D_i \\rightarrow I_(i+1) \\rightarrow \\ldots$$\n\nAny analysis of the code will easily reveal to the malware analyst all the malware internals since all instructions are hardcoded. ",
"But if a few data/instructions are kept under an encrypted form, and deciphered at execution only, the analysis is likely to be far more difficult (up to decryptor and the secret key protection issue). ",
"It is denied of a priori analysis capabilities. ",
"So we intend to have $$I_0 \\rightarrow D'_0 \\rightarrow I_1 \\rightarrow D_1 \\ldots \\rightarrow D'_i \\rightarrow I_(i+1) \\rightarrow \\ldots$$ where $dFSM(D'_i) = D_i$ for all i. Upon execution, we just have to input data $D'_i$ into the dFSM which will then output the data $D_i$. A few critical points are worth stressing on\n\n1. ",
" no key is neither required nor used;\n\n2. ",
" instructions can similarly protected as well.",
"\n\nOf course to be useful as a prevention tool against analysis, the dFSM must itself be obfuscated and protected against analysis. ",
"But this last point is supposed to be fulfilled. ",
"We will show a more powerful implementation by using $K$-ary malware in Section \\[imp\\_k\\].",
"\n\n(Malware) Code Built From an Arbitrary Sequence\n-----------------------------------------------\n\nIn this case, the sequence is arbitrary chosen before the design of the code and hence the code is written directly from this arbitrary sequence. ",
"This case is the most simple to manage. ",
"We just have to choose carefully the dFSM we need. ",
"One of the best choice is to take a congruential generator since it implies a very reduced algorithm with simple instructions.",
"\n\nLet us consider $X_0$ an initial value and the corresponding equation $$x_(i+1) = a*X_i + b \\qquad mod(N)$$ where $a$ is the multiplier, $b$ is the increment and $N$ is the modulus. ",
"Since the length of the sequence involved in the malware design is rather very short (up to a few tens of bytes), the choice of those parameters is not as critical as it would be for practical cryptographic applications. ",
"In this respect, one can refer to Knuth’s seminal book to get the best sets of parameters [@knuth].",
"\n\nHere are a few such examples among many others:\n\nStandard minimal generator\n\n: $a = 16,807 - b = 0 - N = 2^{31} - 1$.\n\nVAX-Marsaglia generator\n\n: $a = 16,645 - b = 0 - N = 2^{32}$.\n\nLavaux & Jenssens generator\n\n: $a = 31,167,285 - b = 0 - N = 2^{48}$.\n\nHaynes generator\n\n: $a = 6,364,136,223,846,793,005 - b = 0 - N = 2^{64}$.\n\nKuth’s generator\n\n: $a = 22~695~477 - b = 1 - N = 2^{32}$ and $X_{n + 1} >>= 16$.\n\nOf course the choice of the modulus is directly depending on the data type used in the malware.",
"\n\nAnother interesting approach is to consider hash functions and S/key. ",
"The principle is almost the same. ",
"We take a $(m,n)$ hash function $H$ which produces a $n$-bit output from a $m$-bit input with $m > n$. In our case we can build $m$ in the following way\n\n m = <data to protect><padding of random data><size of data>\n\nor equivalently\n\n m = D_i <random data> |D_i|\n\nThen we choose a $m$-bit initialization vector (IV) and we compute the random sequence as follows $$IV \\rightarrow D_i = H(IV) \\rightarrow x = H^{|D_i|}(D_i) \\rightarrow y = H^{|x|}(x) \\rightarrow H^{|y|}(y) \\rightarrow$$ The iteration value $|D_i|$ can be used to get one or more required arbitrary value thus anticipating the next case. ",
"Of course the nature of the hash function is also a key parameter: you can either use existing hash function (e.g MD5, SHA-1, RIPEMD 160, SHA-2...) and keep only a subset of the output bit; or you can design your own hash function as explained in [@knuth].",
"\n\nRandom Sequence Coming From a Arbitrary (Malware) Code\n------------------------------------------------------\n\nIn this slightly different case, the sequence is determined by a (non yet protected) instance of a code. ",
"This issue is then to design or use an instance of PRNG accordingly. ",
"This is of course a far more difficult issue which implies cryptanalytic techniques. ",
"To formalize the problem we have a sequence $$X_0, X_1, X_2 \\ldots x_i \\ldots X_n$$ which represents critical data (addresses, ASM instructions, operands...) of a particular instance of a (malware) code. ",
"As for example let us consider three series of 32-bit integers describing bytecode values as defined in Section \\[cs\\]:\n\n 0x2F010000 0x040004 0x3 0x0 0x89 (1)\n 0x3D010000 0x040004 0x3 0x0 0x50 (2) \n 0x5010000 0x040004 0x3 0x0 0x8D (3) \n\nThey are just different instances of the same instruction [@libthor2]. ",
"The aim is to have these data in the code under a non hardcoded form. ",
"Then we intend to code them under an obfuscated form, e.g. $$K_0, K_1, K_2, \\ldots K_i, \\ldots K_n \\ldots$$ We then have to find a dFSM such that $$X_0 = dFSM(K_0), X_1 = dFSM(K_1) \\ldots X_i = dFSM(K_i) \\ldots$$ The notation $K_i$ directly suggests that the quantity input to the dFSM is a key in a cryptographic context but these keys have to exhibit local low entropy profile at the same time. ",
"So the malicious PRNG must take this into account as well. ",
"In this case, we have to face a two-fold cryptanalytic issue:\n\n- either fix the output value $X_i$ and find out the key $K_i$ which outputs $X_i$ for an arbitrary dFSM,\n\n- or for an arbitrary set of pairs $(X_i, K_i)$ design a unique suitable dFSM for those pairs.",
"\n\nThe first case directly relates to a cryptanalytic problem while the second refers more to the problem of designing cryptographic dFSMs with trapdoors. ",
"In our context of malicious cryptography, the trapdoors here are precisely the arbitrary pairs of values $(X_i, K_i)$ while the dFSM behaves for any other pair as a strong cryptosystem [@eciw2010]. ",
"This second issue is far more complex to address and will not be exposed in this paper (research under way; to be published later).",
"\n\nWe will focus on the first case which has been partially addressed for real-life cryptosystem like [*Bluetooth*]{} E0 [@e0] in the context of zero knowledge-like proof of cryptanalysis. ",
"But in our case we do not need to consider such systems and much simpler dFSM can be built conveniently for our purposes: sequences of data we use are rather short.",
"\n\nWe have designed several such dFSMs and we have proceeded to their cryptanalysis to find out the keys $K_i$ which output the values $X_i$.\n\nAs we will see, those dFSM have to exhibit additional features in order to\n\n- be used for code mutation purposes,\n\n- exhibit TRANSEC properties. ",
"In other words, if we have $Y = dFSM(X)$, then $X$ and $Y$ must have the same entropy profile. ",
"Replacing $X$ with a $Y$ having a higher entropy profile would focus the analyst’s attention.",
"\n\nWithout loss of generality, let us consider the mathematical description of a 59-key bit stream cipher (we currently work on block cipher which should offer more interesting features; to be continued …) we have designed for that purpose. ",
"Other dFSMs with larger key size (up to 256) have been also designed (available upon request).",
"\n\nThe principle is very simple yet powerful: three registers $R_1, R_2$ and $R_3$ are filtered by a Boolean function $F$, thus outputing bits $s_t$ (or bytes in a vectorized version) that are combined to the plaintext (Figure \\[combsys\\]).",
"\n\n![",
"Malicious 59-bit key deterministic finite state-machine (stream cipher)[]{data-label=\"combsys\"}](./combsys2_crop.pdf){width=\"\\textwidth\"}\n\nThe value $K_i$ initializes the content of registers $R_1, R_2$ and $R_3$ at time instant $t = 0$, and outputs bits $s^t$ which represent the binary version of values $X_i$.\n\nThe linear feedback polynomials driving the registers are the following: $$\\begin{aligned}\nP_1(x) & = & x^{17} \\oplus x^{15} \\oplus x^{14} \\oplus x^{13} \\oplus x^{11} \\oplus x^{10} \\oplus x^{9} \\oplus x^{8} \n \\oplus x^6 \\oplus x^5 \\oplus x^4 \\oplus x^2 \\oplus 1 \\\\\nP_2(x) & = & x^{19} \\oplus x^{18} \\oplus x^{16} \\oplus x^{15} \\oplus x^{11} \\oplus x^{10} \\oplus x^5 \\oplus x^3 \\oplus x^2 \\oplus x \\oplus 1 \\\\ \nP_3(x) & = & x^{23} \\oplus x^{22} \\oplus x^{21} \\oplus x^{20} \\oplus x^{17} \\oplus x^{16} \\oplus x^{15} \\oplus x^{12} \\oplus x^{10} \\oplus x^{8} \\oplus x^7 \\oplus x \\oplus 1\\end{aligned}$$ where $\\oplus$ denotes the exclusive OR.",
"\n\nThe combination function is the [*Majority function*]{} applied on three input bits and is given by $$f(x_1, x_2, x_3) = x_1x_2 \\oplus x_1x_3 \\oplus x_2x_3.$$ An efficient implementation in C language is freely available by contacting the author. ",
"For our purposes, we will use it in a procedure whose prototype is given by\n\n void sco(unsigned long long int * X, unsigned long long int K)\n {\n /* K obfuscated value (input), X unobfuscated value (output) */\n /* (array of 8 unsigned char) by SCO */\n ...\n }\n\nNow according to the level of obfuscation you need, different ways exist to protect your critical data inside the code (series of integers (1), (2) and (3) above). ",
"We are going to detail two of them.",
"\n\n### Concatenated bytecodes\n\nThe dFSM outputs critical data under a concatenated form to produce chunks of code corresponding to the exact entropy of the input value $(K_i)$, thus preventing any local increase of the code entropy. ",
"For the dFSM considered, it means that we output series (1), (2) and (3) under the following form\n\n 1)--> 0x2F01000000040004000000030000000000000089 \n 2)--> 0x3D01000000040004000000030000000000000050\n 3)--> 0x050100000004000400000003000000000000008D \n\nLet us detail the first output sequence (1). ",
"It will be encoded as three 59-bit outputs $M_1, M_2$ and $M_3$\n\n M_1 = 0x0BC04000000LL;\n M_2 = 0x080008000000060LL;\n M_3 = 0x000000000000089LL; \n\nTo transform $M_1, M_2$ and $M_3$ back into five 32-bit values $X_1, X_2, \nX_3, X_4$ and $X_5$, we use the following piece of code:\n\n /* Generate the M_i values */\n sco(&M_1, K_1);\n sco(&M_2, K_2);\n sco(&M_3, K_3);\n\n X_1 = M_1 >> 10; /* X_1 = 0x2F010000L */ \n X_2 = ((M_2 >> 37) | (M_1 << 22)) & 0xFFFFFFFFL \n /* X_2 = 0x00040004L */\n X_3 = (M_2 >> 5) & 0xFFFFFFFFL; /* X_3 = 0x3 */\n X_4 = ((M_3 >> 32) | (M_2 << 27)) & 0xFFFFFFFFL;\n /* X_4 = 0x0 */\n X_5 = M_3 & 0xFFFFFFFFL; /* X_5 = 0x89 */\n\nValues $M_1, M_2$ and $M_3$ will be stored in the code as the values $K_1, K_2$ and $K_3$ with $dFSM(K_i) = M_i$:\n\n K_1 = 0x6AA006000000099LL;\n K_2 = 0x500403000015DC8LL;\n K_3 = 0x0E045100001EB8ALL; \n\nSimilarly we have for sequence (2)\n\n M_1 = 0x0F404000000LL; K_1 = 0x7514360000053C0LL;\n M_2 = 0x080008000000060LL; K_2 = 0x4C07A200000A414LL;\n M_3 = 0x000000000000050LL; K_3 = 0x60409500001884ALL;\n\nand for sequence (3)\n\n M_1 = 0x01404000000LL; K_1 = 0x76050E00001F0B1LL;\n M_2 = 0x080008000000060LL; K_2 = 0x00000010C80C460LL;\n M_3 = 0x00000000000008DLL; K_3 = 0x000000075098031LL;\n\nThe main interest of that method is that the interpretation of code is not straightforward. ",
"Code/data alignment does not follow any logic (that is precisely why we have chosen a 59-bit FSM which is far better that a seemingly more obvious 64-bit FSM ; any prime value is optimal).",
"\n\nMoreover, as we can notice, the $K_i$ values are themselves sparse as unobfuscated opcodes are (structural aspect). ",
"Additionally, their entropy profile (quantitative aspects) is very similar to the $M_i$ values (and hence the $X_i$ ones). ",
"This implies that any detection techniques based on local entropy picks is bound to fail.",
"\n\nDue to the careful design of our 59-bit dFSM, we managed to obtain a unicity distance which is greater than 59 bits (the unicity distance is the minimal size for a dFSM output to be produced by a single secret key). ",
"In our case a large number of different 59-bit keys can output an arbitrary output sequence. ",
"Here are the results for the three series (1), (2) and (3) (Table \\[res\\]):\n\n Serie $M_i$ values Number of secret keys $K_i$\n ------- -------------- -----------------------------\n (1) $M_1$ 314 (file res11)\n (1) $M_2$ 2,755 (file res12)\n (1) $M_3$ 8,177 (file res13)\n (2) $M_1$ 319 (file res21)\n (2) $M_2$ 2,755 (file res22)\n (2) $M_3$ 26,511 (file res23)\n (3) $M_1$ 9,863 (file res31)\n (3) $M_2$ 2,755 (file res32)\n (3) $M_3$ 3,009 (file res33)\n\n : Number of possible keys for a given output value[]{data-label=\"res\"}\n\nThis implies that we can randomly select our 9 $M_i$ values and thus we have $$\\begin{aligned}\n & & 314 \\times (2,755)^3 \\times 8,177 \\times 319 \\times 26,511 \\times 9,863 \\times 3,009 \\\\\n & = & 13,475,238,762,538,894,122,655,502,879,250 \\end{aligned}$$ different possible code variants. ",
"It is approximatively equal to $2^{103}$ variants. ",
"The different files ($resij$ with $i, j \\in \\{1, 2, 3\\}$) whose name is given in right column of Table \\[res\\] are freely available upon request.",
"\n\nTo build a variant, you just have to select data in the nine files randomly according to the following piece of code (code extract to generate the $M_1$ values for the first serie (1) only; refer to Section \\[imp\\_k\\] for a secure implementation):\n\n f1 = fopen(\"res11\",\"r\"); \n f2 = fopen(\"res12\",\"r\");\n f3 = fopen(\"res13\",\"r\");\n \n randval = (314.0*(rand()/(1 + RAND_MAX)); \n for(i = 0; i < randval; i++) \n fscanf(f1, “%lX %lx %lx\\n”, &y1,&y2, &y3); \n K_1 = y1 | (y2 << 17) | (y3 << 36) ; \n /* do the same for values M_2 and M_3 of serie (1) */\n .... \n /* repeat the same for series (2) and (3) */\n .... \n /* Generate M_1 value for series(1) */ \n sco(&M_1, K_1);\n\nThe different files ($resij$ with $i, j \\in \\{1, 2, 3\\}$) can be stored in a program whose size is less than 400 Kb.",
"\n\n### Non concatenated bytecodes\n\nIn this second case, the dFSM outputs 59-bit chunks of data whose only the 32 least significant bits are useful. ",
"In this case we output five 59-bit chunks of data $M_1, M_2, M_3, M_4$ and $M_5$. For sequence (1) we have\n\n M_1 = 0x???????2F010000LL; \n M_2 = 0x???????00040004LL; \n M_3 = 0x???????00000003LL; \n M_4 = 0x???????00000000LL; \n M_5 = 0x???????00000089LL;\n\nwhere ` ?",
"` describes any random nibble. ",
"We get the $X_i$ values from the $M_i$ values with $X_i = M_i \\quad \\& \\quad 0xFFFFFFFFL;$\n\nThe main interest of that method is that it naturally and very simply provides increased polymorphism compared to the previous approach. ",
"Indeed about $2^{140}$ 5-tuples $(K_1, K_2, K_3, K_4, K_5)$ whenever input in our dFSM produces 5-tuples $(X_1, X_2, X_3, X_4, X_5)$. Then we can produce a huge number of different instances of the same code by randomly choosing any possible 5-tuples. ",
"By increasing size of the memory of the FSM we even can arbitrarily increase the number of possible polymorphic instances.",
"\n\nOperational Implementation {#imp_k}\n==========================\n\nThe obfuscation techniques we have presented before, which are based on malicous cryptography (malicious, biased PRNG) and cryptanalysis techniques (to precompute inputs to the dFSM) require obviously to protect the dFSM itself very strongly. ",
"Indeed obtaining the $X_i$ values from the $K_i$ is straightforward whenever we have the dFSM. ",
"We would then come back to the weak existing implementations and reduce the problem to a simple decoding scheme.",
"\n\nBut this obfuscation is impossible to reverse in the context of $k$-ary malware [@karyd; @kary]. ",
"In this case, the viral information is not contained in a single code as usual malware do, but it is split into $k$ different innocent-looking files whose combined action -– serially or in parallel -– results in the actual malware behavior and in the metamorphic code instance generation.",
"\n\nJust imagine a 2-ary code (we can take of course $k > 2$) made of parts $V_1$ and $V_2$. Part $V_1$ just embeds the dFSM and wait in memory for values $K_i$ coming from part $V_2$. Its role is to compute $dFSM(K_i)$ values and send them back to part $V_2$ according to the protocol summarized by Figure \\[fig2\\].",
"\n\n![",
"2-ary implementation of malicious PRNG[]{data-label=\"fig2\"}](./kary_imp_crop.pdf){width=\"\\textwidth\"}\n\nThe interpretation of data by the reverser (and hence the reversing of the code) is no longer possible since the dFSM is deported in the part $V_1$ and the analysis of part $V_2$ would require to reconstruct the dFSM and to break it. ",
"This is of course impossible since the code does not contain a sufficient quantity of encrypted information.",
"\n\nIn the case of metamorphism, the part $V_1$ will output a random $K_i$ value instead during the code instance generation.",
"\n\nFrom a practical point of view, we have considered different implementations.",
"\n\nCommunication pipes\n\n: Due to the relationship between father and children processes, only parallel class A or C [@kary] k-ary codes can be implemented. ",
"It is not the most optimal solution.",
"\n\nNamed communication pipes\n\n: In this case, k-ary parallel class B codes can be efficiently implemented (the most powerful class since there is no reference in any part \\[hence information\\] to other any part).",
"\n\nSystem V IPC\n\n: This is the most powerful method since everything is located into shared memory.",
"\n\nThe source codes of our implementation will be soon publicly available.",
"\n\nIn the context of code protection for legitimate purpose, the part $V_2$ will be a kernel-land module can be user/host specific. ",
"It can also be a program located on a server outside the code analyst’s scope.",
"\n\nConclusion\n==========\n\nWe have shown in this paper that it is possible to prevent code analysis from reversing while ensuring a high level of metamorphism. ",
"Of course, we have to apply those techniques to all critical parts of the code (especially those which determine the execution flow). ",
"In this case it is obvious that static analysis is no longer possible. ",
"As far as dynamic analysis is concerned, the analyst has to perform debugging and sandboxing. ",
"But using delay detection technique [@filiol_bk2] can trigger a different code behaviour and code mutation.",
"\n\nOur current research and experimentation work is related to non deterministic FSMs. ",
"In this case any internal state of the FSM results in many possible outcome (next state at time instant $t+1$). ",
"Even if is likely to be far beyond any AV software detection capability, human analysis becomes impossible. ",
"Non deterministic contexts result in untractable complexity.",
"\n\nAcknowledgement {#acknowledgement .unnumbered}\n===============\n\nThanks to Anthony Desnos for his support with respect to virtual machines and IR and for many other reasons: Geoffroy Gueguen (and his awful vanWinjgaarden formal grammars), Eloi Vanderbéken (aka [*Baboon*]{}) for his funny way of thinking security.",
"\n\n[50]{} B. Barak, O. Goldreich, R. Impagliazzo, S. Rudich, A. Sahai, S. Vadhan and K. Yang (2001). ",
"On the (im)possibility of obfuscation Programs. ",
"In: [*Advances in Cryptology - CRYPTO 2001*]{}, LNCS 2139, pp. ",
"1-8, Springer Verlag. ",
"Philippe Beaucamps and Eric Filiol (2006). ",
"On the possibility of practically obfuscating programs - Towards a unified perspective of code protection. [*",
"Journal in Computer Virology*]{}, 2(4), WTCV’06 Special Issue, G. Bonfante editor, Springer Verlag. ",
"Anthony Desnos (2009). ",
"Implementation of $K$-ary Viruses in Python. ",
"Hack.lu 2009 Conference. ",
"Anthony Desnos (2010). ",
"Dynamic, Metamorphic (and Open Source) Virtual Machines. ",
"To appear in the [*Proceedings of the Hack.lu 2010 Conference*]{}, Luxembourg. ",
"Thomas Dullien and Sebastian Porst (2009). ",
"REIL: a Platform Independent Intermediate Representation of Disassembled Code for Static Code Analysis. ",
"CanSecWest 2009, Vancouver, Canada. ",
"Available on <http://www.zynamics.com/downloads/csw09.pdf> Robert Erra and Christophe Grenier (2009). ",
"How to choose RSA keys (the art of RSA: past, present and future)? ",
"iAWACS 2009 conference. ",
"Available on <http://www.esiea-recherche.eu/iawacs_2009_papers.html> Eric Filiol. ",
"Strong Cryptography Armoured Computer Viruses Forbidding Code Analysis: the [bradley]{} virus. ",
"In: [*Proceedings of the 14th EICAR Conference*]{}, pp.- 201–217, may 2005, Malta. ",
"Eric Filiol (2005). [*",
"Computer viruses: from theory to applications*]{}, IRIS International Series, Springer Verlag France, ISBN 2-287-23939-1. ",
"Eric Filiol (2007). [*",
"Techniques virales avancées*]{}, collection IRIS, Springer Verlag France, ISBN 978-2-287-33887-8. ",
"Eric Filiol and Sébastien Josse. ",
"A Statistical Model for Viral Detection Undecidability. ",
"EICAR 2007 Special Issue, V. Broucek ed., [*",
"Journal in Computer Virology*]{}, 3 (2). ",
"Springer Verlag. ",
"Eric Filiol, Edouard Franc, Alessandro Gubbioli, Benoit Moquet and Guillaume Roblot (2007). ",
"Combinatorial Optimisation of Worm Propagation on an Unknown Network. [*",
"International Journal in Computer Science*]{}, 2 (2), pp. ",
"124–130, 2007. ",
"Eric Filiol (2007). ",
"Formalisation and implementation aspects of $K$-ary (malicious) codes. [*",
"Journal in Computer Virology*]{}, vol. ",
"3, issue 3. ",
"Springer Verlag. ",
"Eric Filiol (2007). ",
"Zero knowledge-like Proof of Cryptanalysis of Bluetooth Encryption. [*",
"International Journal in Information Theory*]{}, 3 (4), pp. ",
"40ff, <http://www.waset.org/journals/ijit/v3/v3-4-40.pdf> Eric Filiol and Fred Raynal (2008).[*Malicious cryptography ...reloaded - and also malicious statistics*]{}. ",
"CanSecWest Conference, Vancouver, Canada, March 2008. ",
"Available on <http://cansecwest.com/csw08/csw08-raynal.pdf> Eric Filiol. ",
"The malware of the future: when mathematics are on the bad side. [*",
"Opening Keynote Talk*]{}, Hack.lu 2008, Luxembourg, October 2008. ",
"Available on the Hack.lu website. ",
"Eric Filiol (2010) Anti-forensics Techniques Based on Malicious Cryptography. ",
"In: [*Proceedings of the 9th European Conference in Information Warfare ECIW 2010*]{}, Thessaloniki, Greece. ",
"Eric Filiol (2010) Dynamic cryptographic trapdoors. ",
"Submitted. ",
"Shafi Goldwasser and Guy N. Rothblum (2007). ",
"On Best-possible Obfuscation. ",
"In: [*Proceedings of the 4th Theory of Cryptography Conference - TCC’07*]{}. ",
"LNCS 4392, pp. ",
"194–213, Springer Verlag. ",
"Donald E. Knuth (1998). [*",
"The Art of Computer Programming: Seminumerical Algorithms*]{}, Volume 2, Addison-Wesley. ",
"LibThor Website (2010) [*Playing with virology for fun and profit*]{} <http://libthor.avcave.org>. ",
"Available on october 2010. ",
"Perseus homepage. ",
"<http://www.esiea-recherche.eu/perseus.html> James Riordan and Bruce Schneier (1998). ",
"Environmental Key Generation Towards Clueless Agents. ",
"in: [*Mobile Agents and Security*]{}, G. Vigna ed., ",
"Lecture Notes in Computer Science, pp. ",
"15–24, Springer Verlag. ",
"Adam Young and Moti Yung (1999). [*",
"Malicious Cryptography - Exposing Cryptovirology*]{}. ",
"Wiley. ",
"ISBN 0-7645-4975-8.",
"\n\n[^1]: Let us note that here the entropy considers single characters or 1-grams only. ",
"A more accurate value would consider all the possible $n$-grams and would compute entropy when $n \\rightarrow \\infty$.\n\n[^2]: This means that the uncertainty is maximal whenever trying to predict the next value output by the source $X$.\n\n[^3]: Do not forget to reverse the view dually: here the attacker is any code analyst who wants to have a deep insight into the binary code. ",
"However we must keep in mind that what can do a human analyst, will be most of the time very difficult (from a computing point of view at least) for automated program as AV software are.",
"\n"
] |
{
"pile_set_name": "ArXiv"
}
|
[
0.0009405154851265252,
0.0006212362204678357,
0.0006508444785140455,
0.0006923855980858207,
0.0006518836598843336,
0.0007127947174012661,
0.001885971869342029,
0.0010448198299854994,
0.007520771119743586,
0.0005487537709996104,
0.0006894298712722957,
0.0006436862167902291,
0.0016708935145288706,
0.0007062911172397435,
0.0005633060936816037,
0.0009145527146756649,
0.0009097196743823588,
0.0006018680287525058,
0.000621322076767683,
0.0006746079307049513,
0.0006215921603143215,
0.0010807155631482601,
0.0006579611799679697,
0.000588762282859534,
0.0011696417350322008,
0.0006352986674755812,
0.0005990337813273072,
0.0005898521631024778,
0.0006337756058201194,
0.0005741171189583838,
0.0006454403628595173,
0.0006421969155780971,
0.0007604053826071322,
0.0005495426012203097,
0.0009883793536573648,
0.0006209278362803161,
0.0006433510570786893,
0.0007128599099814892,
0.0005558660486713052,
0.0005905249272473156,
0.0005653085536323488,
0.0008722695056349039,
0.000977825839072466,
0.0006169747794046998,
0.0008702923660166562,
0.000529505021404475,
0.0007053507724776864,
0.0014111028285697103,
0.0007329554646275938,
0.0006290183519013226,
0.0006987269734963775,
0.0006858693668618798,
0.0005827097920700908,
0.0009861176367849112,
0.000572064658626914,
0.0006422970327548683,
0.0006823276053182781,
0.0010750290239229798,
0.0007782112807035446,
0.0006495685083791614,
0.000632976705674082,
0.0037756627425551414,
0.000828948337584734,
0.0008264094940386713,
0.0005587725318036973,
0.0008813818567432463,
0.000539252650924027,
0.0005697754095308483,
0.0007092724554240704,
0.0007336682756431401,
0.0011228241492062807,
0.0007721668807789683,
0.0012388852192088962,
0.0007397417211905122,
0.031889498233795166,
0.0008693027775734663,
0.0006880176952108741,
0.0005723558715544641,
0.0007361355819739401,
0.0007017154712229967,
0.0006606404203921556,
0.0005452529876492918,
0.0008447739528492093,
0.0005899190437048674,
0.0005670718965120614,
0.0005840144585818052,
0.0005568802589550614,
0.0007555510383099318,
0.0005530993803404272,
0.0006589128170162439,
0.0016845215577632189,
0.0009847023757174611,
0.0005877457442693412,
0.0010470079723745584,
0.0006766552105545998,
0.0005420346278697252,
0.0006138412281870842,
0.0005861220415681601,
0.0005528210895135999,
0.00066038453951478,
0.0006478797295130789,
0.0006935558631084859,
0.0006073611439205706,
0.000696733477525413,
0.0006344920839183033,
0.0006033018580637872,
0.0007778310100547969,
0.0006699535879306495,
0.0008021703106351197,
0.0007410760154016316,
0.0006080924649722874,
0.0008373028831556439,
0.0006485055200755596,
0.21602298319339752,
0.0005563073209486902,
0.00059639842947945,
0.0008644001791253686,
0.001185585861094296,
0.0006839959532953799,
0.00117425003554672,
0.0005778171471320093,
0.0012900278670713305,
0.0007435926236212254,
0.0013671480119228363,
0.0009302939288318157,
0.001202064217068255,
0.0006448485655710101,
0.0013101339573040605,
0.0005912041524425149,
0.0014581186696887016,
0.0007629097672179341,
0.008680756203830242,
0.00135024543851614,
0.0006227110861800611,
0.0008012097096070647,
0.0014607085613533854,
0.013779226690530777,
0.0006159328040666878,
0.0007875049486756325,
0.03395908698439598,
0.0005795133765786886,
0.0006196104222908616,
0.0006746214930899441,
0.0006231008446775377,
0.000800071400590241,
0.0005931126652285457,
0.015298730693757534,
0.0008107665926218033,
0.0005750494892708957,
0.0007766929338686168,
0.0011808578856289387,
0.00075879180803895,
0.0005822351085953414,
0.0006632774602621794,
0.014864773489534855,
0.0005521028069779277,
0.0006486736820079386,
0.005083650816231966,
0.0007627949235029519,
0.000613835989497602,
0.0006712771719321609,
0.0005895054200664163,
0.011610886082053185,
0.0034606154076755047,
0.0005911418702453375,
0.0005986710893921554,
0.0005956778186373413,
0.00446232408285141,
0.0006769906030967832,
0.00052559410687536,
0.0014805187238380313,
0.0006324804271571338,
0.0006123462226241827,
0.002750464016571641,
0.0006500193849205971,
0.001761899096891284,
0.001398092252202332,
0.0006616245955228806,
0.002815967658534646,
0.0009082408505491912,
0.0005940539995208383,
0.0027774216141551733,
0.03919487074017525,
0.0010740094585344195,
0.0006780237308703363,
0.000910410366486758,
0.0005610219086520374,
0.0007093218737281859,
0.000585028959903866,
0.0006160220364108682,
0.0015528729418292642,
0.0010563667165115476,
0.0005776502657681704,
0.0005846426938660443,
0.0008795232279226184,
0.0019055769080296159,
0.028893737122416496,
0.0006898182909935713,
0.0017277761362493038,
0.0005733952857553959,
0.0008410343434661627,
0.0007962350500747561,
0.07880240678787231,
0.0006619440391659737,
0.0012455401010811329,
0.0009414119413122535,
0.0010629963362589478,
0.0006514504784718156,
0.0006375800003297627,
0.0009774192003533244,
0.0007433064747601748,
0.0007116107153706253,
0.0010818296577781439,
0.0006888732314109802,
0.007274607662111521,
0.0015183411305770278,
0.003098580287769437,
0.0009900290751829743,
0.0006540458416566253,
0.008377180434763432,
0.0020230128429830074,
0.000766384822782129,
0.00405431492254138,
0.0012836605310440063,
0.0010681536514312029,
0.0019055769080296159,
0.002026585629209876,
0.000691685825586319,
0.0006644403329119086,
0.0005343570956028998,
0.00071693368954584,
0.0007760215667076409,
0.0006209745770320296,
0.0006511006504297256,
0.0005484666908159852,
0.0005772471195086837,
0.0006480615702457726,
0.0006211870349943638,
0.0005666828365065157,
0.0006033590761944652,
0.0005762999644502997,
0.0006529241800308228,
0.0006738824886269867,
0.0006389176123775542,
0.0005987815093249083,
0.0006829690537415445,
0.0007380616152659059,
0.0006821351707912982,
0.0006949380040168762,
0.0007031990098766983,
0.0006504286429844797,
0.0006274936022236943,
0.0006099769379943609,
0.0006740932003594935,
0.0006822437280789018,
0.00486097764223814,
0.0008166030165739357,
0.0006728204316459596,
0.0006127119995653629,
0.0006497446447610855,
0.0006539245950989425,
0.0006178432959131896,
0.0006230693543329835,
0.0006191342836245894,
0.0005921970005147159,
0.0006479877629317343,
0.000638603582046926,
0.001556670875288546,
0.0005903206183575094,
0.0007043210207484663,
0.0007475334568880498,
0.0007045197417028248,
0.0006761649274267256,
0.0007395838620141149,
0.0006106499931775033,
0.0006832096260040998,
0.0007160771056078374,
0.0006898639840073884,
0.0006030293297953904,
0.0007764890906400979,
0.0006898212595842779,
0.0007643259014002979,
0.0006878567510284483,
0.0012972343247383833,
0.0007494617602787912,
0.0007373400148935616,
0.0006898639840073884,
0.0006878567510284483,
0.0013844964560121298,
0.0006446341285482049,
0.0011233771219849586,
0.00059486337704584,
0.0007155805360525846,
0.004570068791508675,
0.0006333939963951707,
0.0008642165921628475,
0.0011720212642103434,
0.0005928062601014972,
0.0007801366737112403,
0.0006649179849773645,
0.0007557498174719512,
0.000979678356088698,
0.0007477589533664286,
0.0007786834612488747,
0.0006842518341727555,
0.0007255084929056466,
0.0007801098399795592,
0.0010577549692243338,
0.0005804567481391132,
0.0016373113030567765,
0.0007029169355519116,
0.0007081185467541218,
0.0006670054281130433,
0.0006508661317639053,
0.0006518898881040514,
0.0006647237460128963,
0.008302364498376846,
0.0008483115234412253,
0.0011227159993723035,
0.0006542718037962914,
0.0008650746312923729,
0.0006107768276706338,
0.001995444530621171
] | 0.002482 | 323 |
[
"My Blog.",
"\n\nAn Israeli Diplomat's wife's car was bombed in Delhi recently, allegedly by a limpet bomb attached by a motorcycle rider on the streets of Delhi. ",
"Hawks on all sides are pointing fingers at Iran, but the Indian police claim they haven't found conclusive evidence supporting that view yet. ",
"I spent some time thinking beyond the standard narratives about the attack. ",
"Here are some ideas to chew on:\n\nIndia has been a reasonably staunch ally of Iran on the international stage, but is increasingly under pressure by the West to toe their line instead, and abandon Iran. ",
"India's growing proximity to the West and to Israel, especially in matters related to defense equipment purchases, is a compelling reason for India to throw Iran under the bus. ",
"But the dilemma is, how can they change tune without being viewed by the other developing countries (and their BRIC buddies) as spineless turncoats? ",
"Solution: a fake attack that can be pinned on Iran. ",
"Find a convenient thug to plant the bomb (which, for all the deadly images of the Toyota Innova in flames, actually resulted in no deaths), do a quick \"investigation\", without actually blaming Iran, allow people to speculate to that effect, and use that as an excuse for the chilling of relations with a decades-old ally. ",
"Convenient!!",
"\n\nIsrael is looking for reasons to strike against Iran, and the Hawks in the US (especially in the US Media - look at this article on Slate) are quietly standing by allowing the rhetoric to heat up. ",
"What stops the Mossad from picking a soft target like India, Georgia and Thailand to \"attack\" one of their own diplomats / embassies and blame Iran? ",
"They've been known to do stranger things!",
"\n\nPoint is, there's always more to these things than meets the eye. ",
"Will we ever find out? ",
"Perhaps not. ",
"But that shouldn't stop us from keeping a sharp eye out for a rat.",
"\n\nAbout me\n\nExecutive, entrepreneur, investor and mentor to social entrepreneurs, golf and squash addict, author of thrillers... In short, an amateur dabbler in new experiences, and provoker of thoughts."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0008494722424075007,
0.008286913856863976,
0.0006078545702621341,
0.0005558925331570208,
0.0006277654320001602,
0.001042168471030891,
0.0012301403330639005,
0.01667069084942341,
0.007079394534230232,
0.002280624583363533,
0.0007560382364317775,
0.0023759775795042515,
0.0006349448813125491,
0.0006405638414435089,
0.001262931851670146,
0.0008334283484145999,
0.07659250497817993,
0.0007426515221595764
] | 0.006837 | 18 |
[
"Seating: Formabilio: Facile\n\nFormabilio’s Facile is an outdoor sofa; its cushions have pegs on the bottoms that fit into holes in a wood base, allowing them to be rearranged to create different seating configurations. ",
"The natural plywood base measures 70 inches long, 26 inches deep, and 26 inches high. ",
"Facile is available with two or three seat cushions in charcoal, cream, or sky blue upholstery."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.02070145308971405,
0.0006723832339048386,
0.011914764530956745
] | 0.011096 | 3 |
[
"/* This Source Code Form is subject to the terms of the Mozilla Public\n * License, v. 2.0. ",
"If a copy of the MPL was not distributed with this\n * file, You can obtain one at http://mozilla.org/MPL/2.0/. */\n\n/* db_json is a json database driver. ",
" It is designed for use in\n * local development, is intended to be extremely easy to maintain,\n * have minimal dependencies on 3rd party libraries, and we could\n * care less if it performs well with more than 10 or so users.",
"\n */\nconst\ndbutils = require('./dbutils.js'),\npath = require('path'),\nfs = require('fs'),\nsecrets = require('../secrets.js'),\njsel = require('JSONSelect'),\nlogger = require('../logging/logging.js').logger,\nconfiguration = require('../configuration.js'),\nprimary = require('../primary.js'),\ntemp = require('temp');\n\n// existsSync moved from path in 0.6.x to fs in 0.8.x\nif (typeof fs.existsSync === 'function') {\n var existsSync = fs.existsSync;\n} else {\n var existsSync = path.existsSync;\n}\n\n// a little alias for stringify\nconst ESC = JSON.stringify;\n\nvar dbPath = path.join(configuration.get('var_path'), \"authdb.json\");\n\n/* The JSON database. ",
"The structure is thus:\n * {\n * \"users\":[\n * {\n * \"id\": <numerical user id>,\n * \"password\": <string password and salt info>,\n * \"lastPasswordReset\": <seconds-since-epoch, integer>,\n * \"failedAuthTries\": <number-of-consecutive-failed-authentication-attempts, integer>,\n * \"emails\":{\n * \"syncer@somehost.com\":{\n * \"type\": <string secondary|primary>,\n * \"verified\":<boolean>\n * }\n * }\n * }\n * ],\n * \"stagedEmails\":{},\n * \"staged\":{},\n * \"idp\":{}\n * }\n */\n\nfunction now() { return Math.floor(new Date().getTime() / 1000); }\n\nfunction getNextUserID() {\n var max = 1;\n jsel.forEach(\".id\", db.users, function(id) {\n if (id >= max) max = id + 1;\n });\n return max;\n}\n\nvar db = {\n users: [ ],\n stagedEmails: { },\n staged: { },\n idp: { }\n};\n\nfunction flush() {\n try {\n fs.writeFileSync(dbPath, JSON.stringify(db));\n } catch (e) {\n logger.error(\"Cannot save database to \" + dbPath);\n }\n}\n\nfunction sync() {\n // the database not existing yet just means its empty, don't log an error\n if (existsSync(dbPath)) {\n try {\n db = JSON.parse(fs.readFileSync(dbPath));\n\n // FIXME:\n // at this point db might be missing some important fields\n // we may want to fix this. ",
"In the meantime, delete your old json db\n } catch(e) {\n logger.error(\"Cannot read database from \" + dbPath);\n }\n } else {\n logger.debug(\"Database doesn't exist (yet): \" + dbPath);\n }\n}\n\nexports.open = function(cfg, cb) {\n if (cfg && cfg.path) {\n dbPath = cfg.path;\n }\n logger.debug(\"opening JSON database: \" + dbPath);\n\n sync();\n process.nextTick(function() { cb(null); });\n};\n\nexports.closeAndRemove = function(cb) {\n // if the file cannot be removed, it's not an error, just means it was never\n // written or deleted by a different process\n try { fs.unlinkSync(dbPath); } catch(e) { }\n process.nextTick(function() { cb(null); });\n};\n\nexports.close = function(cb) {\n // don't flush database here to disk, the database is flushed synchronously when\n // written - If we were to flush here we could overwrite changes made by\n // another process - see issue #557\n process.nextTick(function() { cb(null); });\n};\n\nexports.emailKnown = function(email, cb) {\n sync();\n var m = jsel.match(\".emails .\" ",
"+ ESC(email), db.users);\n process.nextTick(function() { cb(null, m.length > 0); });\n};\n\nexports.emailIsVerified = function(email, cb) {\n sync();\n var m = jsel.match(\".emails .\" ",
"+ ESC(email), db.users);\n process.nextTick(function() {\n if (!",
"m.length) cb(\"no such email\");\n else cb(null, m[0].verified);\n });\n};\n\nexports.emailType = function(email, cb) {\n sync();\n var m = jsel.match(\".emails .\" ",
"+ ESC(email), db.users);\n process.nextTick(function() { cb(null, m.length ? ",
"m[0].type : undefined); });\n};\n\nexports.emailInfo = function(email, cb) {\n sync();\n // emailInfo is case insensitive, so search the database and find the\n // \"canonical\" or normalized form of the email address\n db.users.forEach(function(user) {\n Object.keys(user.emails).forEach(function(foundEmail) {\n if (foundEmail.toLowerCase() === email.toLowerCase()) email = foundEmail;\n });\n });\n var m = jsel.match(\".emails .?\", [ ",
"email ], db.users);\n if (!",
"m || !",
"m.length) return process.nextTick(function() {\n cb(null, null);\n });\n var response = {\n lastUsedAs: m[0].type || \"secondary\",\n verified: m[0].verified,\n normalizedEmail: email\n };\n // if the email is known, let's determine if the account has a password\n m = jsel.match(\":root > object:has(.emails > .?) ",
"> .password\", [email], db.users);\n response.hasPassword = m.length ? (",
"typeof m[0] === 'string') : false;\n process.nextTick(function() { cb(null, response); });\n};\n\nexports.isStaged = function(email, cb) {\n if (cb) {\n setTimeout(function() {\n sync();\n cb(null, db.stagedEmails.hasOwnProperty(email));\n }, 0);\n }\n};\n\nexports.lastStaged = function(email, cb) {\n if (cb) {\n sync();\n var d;\n if (db.stagedEmails.hasOwnProperty(email)) {\n d = new Date(db.staged[db.stagedEmails[email]].when);\n }\n setTimeout(function() { cb(null, d); }, 0);\n }\n};\n\nexports.emailsBelongToSameAccount = function(lhs, rhs, cb) {\n sync();\n var m = jsel.match(\".emails:has(.\"+ESC(lhs)+\"):has(.\"+ESC(rhs)+\")\", db.users);\n process.nextTick(function() {\n cb(null, m && m.length === 1);\n });\n};\n\nexports.emailToUID = function(email, cb) {\n sync();\n var m = jsel.match(\":root > object:has(.emails > .\" ",
"+ ESC(email) + \") > .id\", db.users);\n if (m.length === 0) m = undefined;\n else m = m[0];\n process.nextTick(function() {\n cb(null, m);\n });\n};\n\nexports.userOwnsEmail = function(uid, email, cb) {\n sync();\n var m = jsel.match(\":root > object:has(:root > .id:expr(x=\" + ESC(uid) + \")):has(.emails > .\" ",
"+ ESC(email) + \")\", db.users);\n var owned = m && m.length === 1;\n process.nextTick(function() {\n cb(null, owned);\n });\n};\n\nfunction addEmailToAccount(userID, email, type, cb) {\n // validate 'type' isn't bogus\n if ([ 'secondary', 'primary' ].indexOf(type) === -1) {\n return process.nextTick(function() {\n cb(\"invalid type\");\n });\n }\n\n process.nextTick(function() {\n sync();\n var emails = jsel.match(\":has(.id:expr(x=\"+ ESC(userID) +\")) > .emails\", db.users);\n if (emails && emails.length > 0) {\n emails[0][email] = {\n type: type,\n verified: true\n };\n flush();\n }\n cb(null);\n });\n}\n\nfunction deleteOldSecrets(email) {\n var secret = db.stagedEmails[email];\n if (secret) {\n delete db.stagedEmails[email];\n delete db.staged[secret];\n }\n}\n\nexports.stageUser = function(email, hash, cb) {\n secrets.generate(48, function(secret) {\n // overwrite previously staged users\n sync();\n deleteOldSecrets(email);\n db.staged[secret] = {\n type: \"add_account\",\n email: email,\n passwd: hash,\n when: (new Date()).getTime()\n };\n db.stagedEmails[email] = secret;\n flush();\n process.nextTick(function() { cb(null, secret); });\n });\n};\n\nexports.stageEmail = function(existing_user, new_email, hash, cb) {\n secrets.generate(48, function(secret) {\n // overwrite previously staged users\n sync();\n deleteOldSecrets(new_email);\n db.staged[secret] = {\n type: \"add_email\",\n existing_user: existing_user,\n email: new_email,\n passwd: hash,\n when: (new Date()).getTime()\n };\n db.stagedEmails[new_email] = secret;\n flush();\n\n process.nextTick(function() { cb(null, secret); });\n });\n};\n\nexports.createUserWithPrimaryEmail = function(email, cb) {\n var emailVal = { };\n emailVal[email] = {\n type: 'primary',\n verified: true\n };\n var uid = getNextUserID();\n var lastPasswordReset = now();\n db.users.push({\n id: uid,\n password: null,\n lastPasswordReset: lastPasswordReset,\n failedAuthTries: 0,\n emails: emailVal\n });\n flush();\n process.nextTick(function() {\n cb(null, uid, lastPasswordReset);\n });\n};\n\nexports.haveVerificationSecret = function(secret, cb) {\n process.nextTick(function() {\n sync();\n cb(null, !!(",
"db.staged[secret]));\n });\n};\n\n\nexports.emailForVerificationSecret = function(secret, cb) {\n process.nextTick(function() {\n sync();\n if (!",
"db.staged[secret]) return cb(\"no such secret\");\n cb(null, db.staged[secret].email, db.staged[secret].existing_user, db.staged[secret].passwd);\n });\n};\n\nexports.authForVerificationSecret = function(secret, cb) {\n process.nextTick(function() {\n sync();\n if (!",
"db.staged[secret]) return cb(\"no such secret\");\n\n if (db.staged[secret].passwd) {\n return cb(null, db.staged[secret].passwd, db.staged[secret].existing_user);\n }\n\n exports.checkAuth(db.staged[secret].existing_user, function (err, hash) {\n logger.debug(\"check auth \" + err + \" \" + hash);\n cb(err, hash, db.staged[secret].existing_user);\n });\n });\n};\n\n\nexports.verificationSecretForEmail = function(email, cb) {\n setTimeout(function() {\n sync();\n cb(null, db.stagedEmails[email]);\n }, 0);\n};\n\nfunction getAndDeleteRowForSecret(secret, cb) {\n sync();\n if (!",
"db.staged.hasOwnProperty(secret)) return cb(\"unknown secret\");\n\n // simply move from staged over to the emails \"database\"\n var o = db.staged[secret];\n delete db.staged[secret];\n delete db.stagedEmails[o.email];\n flush();\n\n process.nextTick(function() { cb(null, o); });\n}\n\n// either a email re-verification, or an email addition - we treat these things\n// the same\nexports.completeConfirmEmail = function(secret, cb) {\n getAndDeleteRowForSecret(secret, function(err, o) {\n dbutils.withType(o.email, function (type) {\n exports.emailKnown(o.email, function(err, known) {\n function addIt() {\n addEmailToAccount(o.existing_user, o.email, type, function(e) {\n var hash = o.passwd;\n if(e || typeof hash !",
"== 'string') return cb(e, o.email, o.existing_user);\n\n // a hash was specified, update the password for the user\n exports.emailToUID(o.email, function(err, uid) {\n if(err) return cb(err, o.email, o.existing_user);\n\n exports.updatePassword(uid, hash, false, function(err) {\n cb(err || null, o.email, o.existing_user);\n });\n });\n });\n }\n if (known) {\n removeEmailNoCheck(o.email, function (err) {\n if (err) cb(err);\n else addIt();\n });\n } else {\n addIt();\n }\n });\n });\n });\n};\n\nexports.completeCreateUser = function(secret, cb) {\n getAndDeleteRowForSecret(secret, function(err, o) {\n dbutils.withType(o.email, function (type) {\n exports.emailKnown(o.email, function(err, known) {\n function createAccount() {\n var emailVal = {};\n emailVal[o.email] = { type: type, verified: true };\n var uid = getNextUserID();\n var hash = o.passwd;\n db.users.push({\n id: uid,\n password: hash,\n lastPasswordReset: now(),\n failedAuthTries: 0,\n emails: emailVal\n });\n flush();\n cb(null, o.email, uid);\n }\n\n // if this email address is known and a user has completed a re-verification of this email\n // address, remove the email from the old account that it was associated with, and then\n // create a brand new account with only this email.",
"\n // NOTE: this might be sub-optimal, but it's a dead simple approach that mitigates many attacks\n // and gives us reasonable behavior (without explicitly supporting) in the face of shared email\n // addresses.",
"\n if (known) {\n removeEmailNoCheck(o.email, function (err) {\n if (err) cb(err);\n else createAccount();\n });\n } else {\n createAccount();\n }\n });\n });\n });\n};\n\nexports.completePasswordReset = function(secret, password, cb) {\n getAndDeleteRowForSecret(secret, function(err, o) {\n exports.emailKnown(o.email, function(err) {\n if (err) return cb(err);\n\n exports.emailToUID(o.email, function(err, uid) {\n if (err) return cb(err);\n\n // if for some reason the email is associated with a different user now than when\n // the action was initiated, error out.",
"\n if (uid !",
"== o.existing_user) {\n return cb(\"cannot update password, data inconsistency\");\n }\n\n sync();\n // flip the verification bit on all emails for the user other than the one just verified\n var emails = jsel.match(\":has(.id:expr(x=?)) ",
"> .emails\", [ uid ], db.users)[0];\n\n Object.keys(emails).forEach(function(email) {\n if (email !",
"== o.email && emails[email].type === 'secondary') {\n emails[email].verified = false;\n } else if (email === o.email) {\n emails[email].verified = true;\n }\n });\n flush();\n\n // update the password!",
"\n exports.updatePassword(uid, password || o.passwd, true, function(err) {\n cb(err, o.email, uid);\n });\n });\n });\n });\n};\n\nexports.addPrimaryEmailToAccount = function(userID, emailToAdd, cb) {\n sync();\n exports.emailKnown(emailToAdd, function(err, known) {\n function addIt() {\n addEmailToAccount(userID, emailToAdd, 'primary', cb);\n }\n if (known) {\n removeEmailNoCheck(emailToAdd, function (err) {\n if (err) cb(err);\n else addIt();\n });\n } else {\n addIt();\n }\n });\n};\n\nexports.checkAuth = function(userID, cb) {\n sync();\n var hash, failedAuthTries;\n if (userID) {\n var m = jsel.match(\":root > object:has(:root > .id:expr(x=?))\", [ ",
"userID ], db.users);\n if (m.length === 1) {\n hash = m[0].password;\n failedAuthTries = m[0].failedAuthTries;\n }\n }\n process.nextTick(function() { cb(null, hash, failedAuthTries); });\n};\n\nexports.incAuthFailures = function(userID, cb) {\n if (typeof userID === 'string') userID = parseInt(userID, 10);\n sync();\n var err;\n if (userID) {\n var m = jsel.match(\":root > object:has(:root > .id:expr(x=?))\", [ ",
"userID ], db.users);\n if (m.length === 1) {\n m[0].failedAuthTries++;\n flush();\n } else {\n err = \"no such user\";\n }\n }\n process.nextTick(function() { cb(err); });\n};\n\nexports.clearAuthFailures = function(userID, cb) {\n if (typeof userID === 'string') userID = parseInt(userID, 10);\n sync();\n var err;\n if (userID) {\n var m = jsel.match(\":root > object:has(:root > .id:expr(x=?))\", [ ",
"userID ], db.users);\n if (m.length === 1) {\n m[0].failedAuthTries = 0;\n flush();\n } else {\n err = \"no such user\";\n }\n }\n process.nextTick(function() { cb(err); });\n};\n\nexports.lastPasswordReset = function(userID, cb) {\n sync();\n var m;\n if (userID) {\n m = jsel.match(\":root > object:has(:root > .id:expr(x=\" + ESC(userID) + \")) > .lastPasswordReset\", db.users);\n if (m.length === 0) m = undefined;\n else m = m[0];\n }\n process.nextTick(function() { cb(null, m); });\n};\n\nexports.userKnown = function(userID, cb) {\n sync();\n var m = jsel.match(\":root > object:has(:root > .id:expr(x=\" + ESC(userID) + \"))\", db.users);\n var known = (m.length !",
"== 0);\n var hasPass = (known && typeof m[0].password === 'string' && m[0].password.length > 0);\n process.nextTick(function() { cb(null, known, hasPass); });\n};\n\nexports.updatePassword = function(userID, hash, invalidateSessions, cb) {\n sync();\n var m = jsel.match(\":root > object:has(.id:expr(x=\" + ESC(userID) + \"))\", db.users);\n var err;\n if (m.length === 0) err = \"no such email address\";\n else {\n m[0].password = hash;\n if (invalidateSessions) {\n m[0].lastPasswordReset = now();\n }\n m[0].failedAuthTries = 0;\n }\n flush();\n process.nextTick(function() { cb(err); });\n};\n\nexports.listEmails = function(uid, cb) {\n sync();\n var emails = jsel.match(\":has(.id:expr(x=\"+ ESC(uid) +\")) > .emails\", db.users);\n process.nextTick(function() {\n if (!",
"emails || emails.length !",
"== 1) {\n cb(\"no such user: \" + uid);\n return;\n }\n // simply return the emails associated with this user\n cb(null, Object.keys(emails[0]));\n });\n};\n\nexports.emailLastUsedAs = function(email, cb) {\n sync();\n var emails = jsel.match(\".?\", [ ",
"email ], db.users);\n process.nextTick(function () {\n if (!",
"emails || emails.length !",
"== 1) {\n cb('emailLastUsedAs Expected 1 row, got ' + emails.length + ' for ' + email);\n return;\n }\n cb(null, emails[0].type);\n });\n};\n\nconst typeEnum = ['primary', 'secondary'];\n\nexports.updateEmailLastUsedAs = function(email, type, cb) {\n if (typeEnum.indexOf(type) === -1) return cb && cb('Invalid type for updating email.type');\n sync();\n var emails = jsel.match(\".\" ",
"+ ESC(email), db.users);\n emails[0].type = type;\n flush();\n process.nextTick(function () {\n cb(null);\n });\n};\n\n\nexports.removeEmail = function(authenticated_user, email, cb) {\n sync();\n var m = jsel.match(\":has(.id:expr(x=\" + ESC(authenticated_user) + \")) .emails:has(.\"+ESC(email)+\")\", db.users);\n\n if (m.length) {\n var emails = m[0];\n delete emails[email];\n flush();\n }\n setTimeout(function() { cb(null); }, 0);\n};\n\nfunction removeEmailNoCheck(email, cb) {\n sync();\n var m = jsel.match(\".emails:has(.\"+ESC(email)+\")\", db.users);\n if (m.length) {\n var emails = m[0];\n delete emails[email];\n flush();\n }\n process.nextTick(function() { cb(null); });\n}\n\nexports.cancelAccount = function(authenticated_uid, cb) {\n sync();\n\n for (var i = 0; i < db.users.length; i++) {\n if (db.users[i].id === authenticated_uid) break;\n }\n\n if (i < db.users.length) {\n var m = jsel.match(\":root > object:has(.existing_user:expr(x=\" + ESC(authenticated_uid) + \"))\", db.staged);\n if (m.length) {\n m.forEach(function(staged) {\n var token = db.stagedEmails[staged.email];\n delete db.stagedEmails[staged.email];\n delete db.staged[token];\n });\n }\n\n db.users.splice(i, 1);\n flush();\n }\n\n process.nextTick(function() { cb(null); });\n};\n\nexports.createUnverifiedUser = function(email, hash, cb) {\n sync();\n var emailVal = {};\n emailVal[email] = {\n type: 'secondary',\n verified: false\n };\n var row = {\n id: getNextUserID(),\n lastPasswordReset: now(),\n password: hash,\n emails: emailVal\n };\n db.users.push(row);\n flush();\n exports.stageEmail(row.id, email, hash, function(err, secret) {\n cb(err, row.id, secret);\n });\n};\n\nexports.addTestUser = function(email, hash, cb) {\n sync();\n removeEmailNoCheck(email, function() {\n var emailVal = {};\n emailVal[email] = {\n type: 'secondary',\n verified: true\n };\n db.users.push({\n id: getNextUserID(),\n password: hash,\n lastPasswordReset: now(),\n emails: emailVal\n });\n flush();\n cb(null);\n });\n};\n\nexports.ping = function(cb) {\n process.nextTick(function() { cb(null); });\n};\n\nexports.updateIDPLastSeen = function(domain, cb) {\n sync();\n db.idp[domain] = new Date().getTime();\n flush();\n process.nextTick(function() { cb(null); });\n};\n\nexports.getIDPLastSeen = function(domain, cb) {\n sync();\n var lastSeen = db.idp[domain] ? ",
"new Date(db.idp[domain]) : null;\n process.nextTick(function() { cb(null, lastSeen); });\n};\n\nexports.forgetIDP = function(domain, cb) {\n sync();\n delete db.idp[domain];\n flush();\n process.nextTick(function() { cb(null); });\n};\n"
] |
{
"pile_set_name": "Github"
}
|
[
0.0006754195201210678,
0.0007588891894556582,
0.000781429756898433,
0.0013185485731810331,
0.0011885721469298005,
0.008866635151207447,
0.0008245963836088777,
0.0008847449789755046,
0.0007883345824666321,
0.0007349642692133784,
0.0012998724123463035,
0.0008490048930980265,
0.008812237530946732,
0.0008947960450313985,
0.0006804304430261254,
0.0016780338482931256,
0.0011978179682046175,
0.0016838848823681474,
0.0009605881059542298,
0.0011944090947508812,
0.00136384682264179,
0.0018624186050146818,
0.0019229038152843714,
0.0007117324275895953,
0.0014390971045941114,
0.0013425906654447317,
0.0007495785248465836,
0.0007605727878399193,
0.0009930250234901905,
0.0011987489415332675,
0.0018860320560634136,
0.0019529955461621284,
0.002662050537765026,
0.0037335518281906843,
0.000763014773838222,
0.0008340870263054967,
0.0008151760557666421,
0.000763014773838222,
0.000846724200528115,
0.0033957939594984055,
0.0031003099866211414
] | 0.001687 | 41 |
[
"Well, I've been tinkering with my new Aurora system all day and so far it hasnt been good. ",
"I've installed FarCry and the graphics look horrible....................which makes no sense at all. ",
"Everything is to bright and has a 'washed out' look. ",
"Ahh!! ",
"I've updated my video drivers and still looks screwed up. ",
"Any ideas?",
"\n\nAlrighty!! ",
"So far the 1.3 FarCry patch worked. ",
"Everything seems to be good...........for now. ",
"Before I installed the new patch the FarCry Auto Configurator kept saying that my system needed to be put on all 'low' settings......................NO WAY. ",
"Now, after the patch...............it says 'Very High' settings.............SWEET!!"
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0007614560308866203,
0.04212772473692894,
0.0014786446699872613,
0.003615802153944969,
0.006816387642174959,
0.0007957352208904922,
0.0011098741088062525,
0.0007818792364560068,
0.0006516212015412748,
0.0011416124179959297,
0.0013274267548695207
] | 0.00551 | 11 |
[
"Q:\n\nCannot instantiate an ObjectInputStream\n\ni'm trying to instantiate an ObjectInputStream in a client on line 24 but it won't, I can instantiate an ObjectOutputStream but not Input. ",
"\n private class ClientThread implements Runnable{\n Socket s;\n ObjectInputStream inObject;\n ObjectOutputStream outObject;\n ServerSocket ss;\n Integer portNo;\n ClientThread(int portNo){\n try{\n this.portNo = portNo;\n ss = new ServerSocket(portNo); \n } \n catch(IOException e){\n System.out.println(\"CT : \"+e.getMessage());\n }\n }\n\n public void run(){\n\n boolean hasOpponent = false;\n try{\n while(!hasOpponent){\n s = ss.accept(); \n if(s !",
"= null){\n inObject = new ObjectInputStream(s.getInputStream()); // line 24 \n System.out.println(\"1\"); \n hasOpponent = true; \n game = new Game(name1, name2); \n }\n }\n }\n catch(Exception e){\n System.out.println(\"D \"+ e.getMessage());\n }\n\nA:\n\nYou haven't said what does happen but I suspect it is blocking.",
"\nObjectInputStream blocks until it get a header from an ObjectOutputStream which may require a flush() from the other end.",
"\n\n"
] |
{
"pile_set_name": "StackExchange"
}
|
[
0.000739714247174561,
0.012232186272740364,
0.006088922265917063,
0.001499259378761053,
0.001995444530621171
] | 0.004511 | 5 |
[
"Synuchus breviusculus\n\nSynuchus breviusculus is a species of ground beetle in the subfamily Harpalinae. ",
"It was described by Mannerheim in 1849.",
"\n\nReferences\n\nCategory:Synuchus\nCategory:Beetles described in 1849"
] |
{
"pile_set_name": "Wikipedia (en)"
}
|
[
0.004841086454689503,
0.0005897678784094751,
0.0006232576561160386
] | 0.002018 | 3 |
[
"Former Hawick High School pupil named as one of USA’s leading movers and shakers\n\nHilary Mactaggart\n\nA former Hawick High School pupil is named in a list of 50 “thinkers, doers and visionaries” transforming politics in the US.",
"\n\nAngus Deaton and his wife Anne Case, both professors at Princeton University in New Jersey, have been afforded places on the elite list assembled by American political magazine Politico.",
"\n\nSir Angus, a professor of economics and international affairs, was awarded the $900,000 dollar Nobel Prize for Economics last year, presented to him by King Carl XVI Gustaf of Sweden at a ceremony in Stockholm.",
"\n\nThe 70-year-old, not to be confused with former TV presenter Angus Deayton, was born in Edinburgh and attended the city’s James Gillespie’s Boys School before he moved with his family to Bowden, near Melrose, as a child.",
"\n\nTogether with his sister Mairi, he was a pupil at Newtown St Boswells Primary School before passing the old 11-plus examination, leading to a daily journey to Hawick High School for the next two years.",
"\n\nIn his Nobel autobiography, Sir Angus, knighted earlier this year, wrote of his father’s long-held wish that he should attend Fettes College in Edinburgh, one of Scotland’s premier educational establishments.",
"\n\nAt that time, the family could not afford it, but the autobiography reveals: “It turned out that Fettes admitted two foundation scholars a year, and several teachers at Hawick donated their time to train me for the competitive examination.",
"\n\n“They must have done this out of dedication to teaching. ",
"Certainly, my father had no money to pay them.",
"\n\n“I worked very hard over many months, becoming quite sick at the time of the exams, but won one of the scholarships.”",
"\n\nThe Politico listing – the latest in a long line of accolades bestowed on the academic – followed publication of a paper in which Sir Angus and his wife researched data from national surveys and discovered that from 1999 to 2013, death rates increased for white non-Hispanic middle-aged Americans, particularly those without college degrees.",
"\n\nThe paper described a quiet epidemic plaguing middle-aged white Americans.",
"\n\nThe Deatons argued that it could not just be a matter of wage stagnation hurting this class of US society.",
"\n\nThe data also pointed to an upturn in what they called “deaths of despair” – increases in mortality due to suicide, alcohol and drug use, particularly heroin and prescription opioids, and related diseases, including liver problems and cirrhosis."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0007922934019006789,
0.0006107046501711011,
0.0005826108972541988,
0.0011994432425126433,
0.0009435914107598364,
0.0005919263930991292,
0.0006378726684488356,
0.000580484454985708,
0.003480571787804365,
0.0006116305594332516,
0.002539760200306773,
0.01721956953406334,
0.0015244053211063147,
0.0006962054176256061
] | 0.002287 | 14 |
[
"Imaging techniques for measuring adipose-tissue distribution in the abdomen: a comparison between computed tomography and 1.5-tesla magnetic resonance spin-echo imaging.",
"\nEight subjects were examined both by abdominal X-ray computed transverse axial tomography (CT) and magnetic resonance imaging (MRI) (SE) (TR/TE, 200 ms/15 ms); another eight volunteers were subjected to three MRI scans to test the reliability of repeated measures. ",
"Correlations between fat area measures obtained by CT and by MRI for subcutaneous fat, total fat, and visceral vs. subcutaneous-fat ratio were highly significant (r = 0.93, 0.91, and 0.94, respectively; p < 0.01), and the standard errors of estimation were 9.99, 23.87, and 0.0047. ",
"The average errors of the method for different fat areas were 2.20 cm2 (intra-examination variance) and 3.75 cm2 (inter-examination variance) for visceral and 0.82 cm2 (intra-examination variance) and 1.29 cm2 (inter-examination variance) for subcutaneous fat areas, respectively. ",
"These results suggest that SE MRI is a practical approach to evaluate body fat distribution without the exposure to radiation. ",
"The reproducibility of SE MRI for the determination of fat areas is high; variation is small and acceptable. ",
"However, it is difficult to determine which estimate of fat area should be accepted when there is a discrepancy between MRI and CT measures."
] |
{
"pile_set_name": "PubMed Abstracts"
}
|
[
0.0005869502201676369,
0.0005683521158061922,
0.001889911014586687,
0.0016041718190535903,
0.0010524783283472061,
0.0009136529988609254,
0.000971709843724966
] | 0.001084 | 7 |
[
"Q:\n\nDrupal Webform hook_webform_submission_insert not firing\n\nI am trying to use the hook_webform_submission_insert in my theme template.php file. ",
"I have 2 other webform hooks currently running in here and they work just fine. ",
"I am trying to get the submission data after it has been submitted. ",
"Below is my code. ",
"\nfunction acquarius_hook_webform_submission_insert($node, $submission){\n var_dump($node);\n var_dump($submission);\n}\n\nI am sure I am missing something small here but everything I try seems to fail.",
"\n\nA:\n\nYou need to replace hook keyword with your theme name. ",
"Also add hook functions in module.",
"\n YOURTHEMENAME_webform_submission_insert($node, $submission){\n// give your code here\n}\n\n"
] |
{
"pile_set_name": "StackExchange"
}
|
[
0.0007652759086340666,
0.0006213522283360362,
0.0005722866044379771,
0.0007543298415839672,
0.0015151910483837128,
0.0007325334008783102,
0.0006428153137676418,
0.0008490898180752993
] | 0.000807 | 8 |
[
"angular.module(\"ngLocale\", [], [\"$provide\", function($provide) {\nvar PLURAL_CATEGORY = {ZERO: \"zero\", ONE: \"one\", TWO: \"two\", FEW: \"few\", MANY: \"many\", OTHER: \"other\"};\n$provide.value(\"$locale\", {\n \"DATETIME_FORMATS\": {\n \"AMPMS\": {\n \"0\": \"AM\",\n \"1\": \"PM\"\n },\n \"DAY\": {\n \"0\": \"Pazar\",\n \"1\": \"Pazartesi\",\n \"2\": \"Salı\",\n \"3\": \"Çarşamba\",\n \"4\": \"Perşembe\",\n \"5\": \"Cuma\",\n \"6\": \"Cumartesi\"\n },\n \"MONTH\": {\n \"0\": \"Ocak\",\n \"1\": \"Şubat\",\n \"2\": \"Mart\",\n \"3\": \"Nisan\",\n \"4\": \"Mayıs\",\n \"5\": \"Haziran\",\n \"6\": \"Temmuz\",\n \"7\": \"Ağustos\",\n \"8\": \"Eylül\",\n \"9\": \"Ekim\",\n \"10\": \"Kasım\",\n \"11\": \"Aralık\"\n },\n \"SHORTDAY\": {\n \"0\": \"Paz\",\n \"1\": \"Pzt\",\n \"2\": \"Sal\",\n \"3\": \"Çar\",\n \"4\": \"Per\",\n \"5\": \"Cum\",\n \"6\": \"Cmt\"\n },\n \"SHORTMONTH\": {\n \"0\": \"Oca\",\n \"1\": \"Şub\",\n \"2\": \"Mar\",\n \"3\": \"Nis\",\n \"4\": \"May\",\n \"5\": \"Haz\",\n \"6\": \"Tem\",\n \"7\": \"Ağu\",\n \"8\": \"Eyl\",\n \"9\": \"Eki\",\n \"10\": \"Kas\",\n \"11\": \"Ara\"\n },\n \"fullDate\": \"d MMMM y EEEE\",\n \"longDate\": \"d MMMM y\",\n \"medium\": \"d MMM y HH:mm:ss\",\n \"mediumDate\": \"d MMM y\",\n \"mediumTime\": \"HH:mm:ss\",\n \"short\": \"dd MM yyyy HH:mm\",\n \"shortDate\": \"dd MM yyyy\",\n \"shortTime\": \"HH:mm\"\n },\n \"NUMBER_FORMATS\": {\n \"CURRENCY_SYM\": \"TL\",\n \"DECIMAL_SEP\": \",\",\n \"GROUP_SEP\": \".\",",
"\n \"PATTERNS\": {\n \"0\": {\n \"gSize\": 3,\n \"lgSize\": 3,\n \"macFrac\": 0,\n \"maxFrac\": 3,\n \"minFrac\": 0,\n \"minInt\": 1,\n \"negPre\": \"-\",\n \"negSuf\": \"\",\n \"posPre\": \"\",\n \"posSuf\": \"\"\n },\n \"1\": {\n \"gSize\": 3,\n \"lgSize\": 3,\n \"macFrac\": 0,\n \"maxFrac\": 2,\n \"minFrac\": 2,\n \"minInt\": 1,\n \"negPre\": \"(\",\n \"negSuf\": \" \\u00A4)\",\n \"posPre\": \"\",\n \"posSuf\": \" \\u00A4\"\n }\n }\n },\n \"id\": \"tr\",\n \"pluralCat\": function (n) { return PLURAL_CATEGORY.OTHER;}\n});\n}]);"
] |
{
"pile_set_name": "Github"
}
|
[
0.0016024099895730615,
0.0010369418887421489
] | 0.00132 | 2 |
[
"Corn Hole Tournament\n\nJune 10, 2017 @ 11:00 am - 8:00 pm\n\nJoin us at Put-in-Bay, Ohio for the best setting for a Corn Hole Tournament! ",
"Listen to FREE Music, grab a drink & enjoy the Lake Erie views at Perry’s Victory and International Peace Memorial. ",
"Two American Cornhole Association sanctioned tournaments (one for singles, the other for two-player teams) will be held in conjunction with the Put-in-Bay Music Festival. ",
"Pre-registration is required at putinbayohiocornhole.com and you must be at least 18 years old to enter. ",
"Please feel free to bring lawn chairs and blankets. ",
"Food, such as Italian sausage sandwiches, gyros, fries, and kid friendly fare like ice cream will be available on site for purchase as well as soft drinks, water, craft beer and local wine.",
"\n\nMore on the Put-in-Bay Music Festival Sponsored by Miller Boat Line: The musical performances will be eclectic and range from zydeco, bluegrass, grassroots, jazz, to blues. ",
"Eight groups are scheduled to perform, including Strung Like a Horse, Mo’Mojo, The Floorwalkers, Emily Keener, JP and the Chatfield Boys, The Flyin Jays, and more. ",
"Come to Put-in-Bay and enjoy the all day FREE Music Festival."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0006634484161622822,
0.0006676827906630933,
0.0006750449538230896,
0.06620801240205765,
0.0005295007140375674,
0.0007676529348827899,
0.0005793005111627281,
0.0013641255209222436,
0.0006205092067830265
] | 0.008008 | 9 |
[
"Ever since I was a kid, I’ve been afraid of the sea. ",
"I’ve never had trouble swimming, and I’ve never dreaded finding myself in a Blake-Lively-clutching-a-rock-type situation. ",
"My particular fear is rooted in taking that last step beyond where the ocean floor is visible. ",
"I don’t get scared because I know what’s out there. ",
"I get scared because I don’t.",
"\n\nAs their contract stalemate with Ryan Fitzpatrick barrels toward training camp, the Jets are wading into their own terrifying version of the unknown. ",
"By now, we know where both parties stand. ",
"The New York Daily News reported earlier this month that Fitzpatrick is willing to take a one-year, $12 million deal, about $4 million less than his original asking price. ",
"With their bloated salary cap, the Jets have balked at slapping such a gaudy price tag on Fitzpatrick for 2016.",
"\n\nThe lasting effects of a 2015 spending spree, a combined $11.5 million paid this offseason to Matt Forte, Ryan Clady, and Steve McClendon, and a $15.7 million franchise-tag offer extended to Muhammad Wilkerson (he hasn’t signed) have left the Jets with just over $3 million in projected cap space heading into this fall. ",
"Second-year general manager Mike Maccagnan’s plan since taking over was to use every dollar saved by predecessor John Idzik and throw it at the Jets’ myriad problems. ",
"So here we are, on the brink of a future in which Geno Smith and Christian Hackenberg battle it out for New York’s starting job in the most depressing Thunderdome ever conceived. ",
"Two men enter. ",
"One fan base leaves disappointed. ",
"The whole fiasco makes it worth considering what the Jets are putting at risk by swimming farther into murky waters.",
"\n\nLooking at Fitzpatrick’s overall production from last season, it makes sense that he believes he should be paid. ",
"He threw for 3,905 yards with 31 touchdowns, and the Jets finished 10th in the league in passing DVOA, up from 27th the year before. ",
"The current quarterback landscape is broken up into the haves (really, every quarterback who has earned a second contract), and the have-nots (players still on rookie deals). ",
"Fitzpatrick’s $12 million asking price would tie him with Houston’s Brock Osweiler for 21st in the NFL among quarterback cap hits, and both Osweiler and the Dolphins’ Ryan Tannehill — ranked 22nd — are slated for huge raises in upcoming years. ",
"With that in mind, if Paxton Lynch wins the job in Denver, only Cleveland’s Robert Griffin III and Buffalo’s Tyrod Taylor would be cheaper, nonrookie starting quarterbacks than Fitzpatrick in 2016.",
"\n\nThe Jets don’t want to give Fitzpatrick that money because they believe their offensive success last season was based on the heroics of receivers Brandon Marshall and Eric Decker, and that Fitzpatrick — who only ascended to the starting job because Geno Smith and Tywin Lannister don’t share similar philosophies about paying debts — was not as good as his numbers indicate; the front office feels that he was league average at best last year and one of the NFL’s lesser quarterbacks at worst. ",
"As such, the team believes that its 10–6 turnaround campaign was the product of trading for Marshall and buying $149 million worth of defensive backs last offseason.",
"\n\nStill, drawing a line in the sand with Fitzpatrick is a decidedly process-over-results move from Maccagnan, and considering history, one that requires plenty of self-restraint. ",
"Even if offensive coordinator Chan Gailey’s scheme and a great supporting cast boosted Fitzpatrick’s production, teams simply don’t move on from quarterbacks who throw 30 touchdown passes in a season. ",
"As in, no team ever has. ",
"Throughout NFL history, zero QBs have thrown 30 touchdowns in a season and not entered the following fall as his team’s starting quarterback.",
"\n\nAnd usually, this is logical. ",
"Most 30-touchdown seasons belong to some of the best QBs in the league, and only 10 players (Brett Favre, Drew Brees, Tom Brady, Aaron Rodgers, Dan Marino, Philip Rivers, Peyton and Eli Manning, Tony Romo and Kurt Warner) have done it more than twice. ",
"Quarterback movement is also rare. ",
"Only four projected starters for 2016 were signed away as free agents by their current teams: the Bills’ Taylor, the Browns’ Griffin, the Saints’ Brees, and the Texans’ Osweiler.",
"\n\nBut this offseason has provided a couple of different instances of teams playing chicken with productive quarterbacks they weren’t ready to commit to in the long term. ",
"Kirk Cousins is a 27-year-old who threw for 4,166 yards with 29 touchdowns last year; Washington elected to give him the franchise tag and burn $19.95 million (the second-highest base salary in the NFL this year, behind only Ndamukong Suh) instead of offering a multiyear deal. ",
"Denver took it further. ",
"Instead of forking over a massive contract to Osweiler, the Broncos let the Texans — searching for their own way out of QB purgatory — hand the 25-year-old a four-year, $72 million deal.",
"\n\nThe situations in Denver and New York mirror each other on several levels. ",
"The Broncos had the league’s best defense in 2015, while the Jets, who finished fifth in defensive DVOA under first-year coach Todd Bowles, weren’t far behind. ",
"A pair of star receivers define both offenses. ",
"Both teams have used the franchise tag on pass rushers (Wilkerson for the Jets, Von Miller for the Broncos) seeking long-term deals. ",
"And in both places, it seems management believes that winning can happen in spite of quarterback play; by electing not to pay Osweiler, Denver has shown it’s possible for a team — one that just won the freaking Super Bowl, no less — to stick to its plan and not panic to secure a quarterback with whom the front office isn’t thrilled.",
"\n\nAnd all of that is fine … until it’s November, the Jets are 3–5, and Marshall is yelling at Smith on the sideline after uncorking his third interception of the day. ",
"Again, whether New York’s season pans out won’t be what determines if moving on from Fitzpatrick was the right choice. ",
"That decision should be based on all of the information available now, and the bulk of that — and essentially all logic and reason — points to the prudency of not giving $12 million to a 33-year-old, below-average quarterback just because Marshall is good at stealing cornerbacks’ lunch money.",
"\n\nThe problem, though — and the reason I understand the trepidation some Jets’ fans have about entering this fall with some combination of Smith and Hackenberg under center — is that fear is inherently irrational. ",
"I know, if I were to take a quick dip in the Pacific, the odds of my needing to call Quint would be exceedingly low. ",
"Sometimes, though, that rationale just doesn’t matter.",
"\n\nEven if Fitzpatrick might as well have been punting the ball downfield at times last year, and even if his shortcomings led to the Jets’ Week 17 loss to the Bills and a place on the couch during the playoffs, Fitzpatrick — in this offense, with these receivers — remains a known quantity. ",
"NFL decision-makers have been falling back on familiar-if-uninspiring options forever. ",
"It’s among their favorite pastimes. ",
"Hell, Mike Mularkey is set to start his third stint as an NFL head coach.",
"\n\nAll of the pieces may be in place for another mediocre quarterback to come into an offense with Marshall, Decker, and now Forte and throw for 30 touchdowns. ",
"But to this point, no other QB has. ",
"No other quarterback in Jets’ history has. ",
"Replacing one player with another is about more than swapping skill sets and rerunning the projections, especially at quarterback. ",
"Command of both the huddle and the offense matter, and even if Smith isn’t a markedly inferior quarterback than Fitzpatrick, there is still no way of knowing whether he can get a combined 2,400 yards and 16 touchdowns out of Decker and Marshall, who has a history of souring on his quarterbacks. ",
"We know what a Fitzpatrick-led Jets offense looks like, and even if that offense sputtered at times because of him, walking away from something that worked is rarely easy.",
"\n\nAs Fitzpatrick’s contract stalemate continues to dominate the headlines, it’s hard to know if the Jets are doing the right thing here. ",
"But whereas questions now abound, in nine months we’ll have plenty of answers — about the Jets and about the costs and benefits of diving into the unknown quarterback depths. ",
"Denver is hoping that the Texans were suckers for paying for Osweiler. ",
"Houston is hoping that the Broncos are stuck defending their Super Bowl title with Lynch. ",
"And as of now, it appears the job in New York is Geno Smith’s to lose. ",
"For so many reasons, it looks to be the frugal, smart move. ",
"But sometimes the smart move is scary.",
"\n\nAn earlier version of this piece omitted Buffalo’s starting quarterback, Tyrod Taylor, who also would be less expensive than Ryan Fitzpatrick in 2016 and also was signed as a free agent by his current team."
] |
{
"pile_set_name": "OpenWebText2"
}
|
[
0.00213999068364501,
0.000803000177256763,
0.0009666216792538762,
0.0016406758222728968,
0.0016091649886220694,
0.001220705104060471,
0.0006059927982278168,
0.0005936322850175202,
0.011216433718800545,
0.0010300109861418605,
0.0010416385484859347,
0.003734057303518057,
0.0011993113439530134,
0.0006786035955883563,
0.0010348664363846183,
0.000609379552770406,
0.0019750408828258514,
0.0009311088360846043,
0.000872313161380589,
0.0008315041195601225,
0.003839533543214202,
0.0006883800379000604,
0.0005819122307002544,
0.0010698391124606133,
0.0008638271247036755,
0.0030333863105624914,
0.0006935499259270728,
0.000753291416913271,
0.0007723685703240335,
0.0007809104863554239,
0.004247254226356745,
0.003268213476985693,
0.00069506122963503,
0.010344253852963448,
0.0005886387662030756,
0.0008795055910013616,
0.0007571919122710824,
0.0007224486907944083,
0.009398719295859337,
0.004407945089042187,
0.0006767064332962036,
0.005915828514844179,
0.004267749842256308,
0.008170347660779953,
0.0008820518269203603,
0.0010672117350623012,
0.0005926684243604541,
0.0010291468352079391,
0.15654443204402924,
0.060498472303152084,
0.0007603788981214166,
0.0028733902145177126,
0.001307674334384501,
0.0050114658661186695,
0.000785741547588259,
0.0006601277855224907,
0.000614900782238692,
0.4880143702030182,
0.0009474827093072236,
0.008121354505419731,
0.0008008688455447555,
0.0006891448283568025,
0.0008315939921885729
] | 0.013241 | 63 |
[
"/* Teensy 3.x, LC ADC library\n * https://github.com/pedvide/ADC\n * Copyright (c) 2015 Pedro Villanueva\n *\n * Permission is hereby granted, free of charge, to any person obtaining\n * a copy of this software and associated documentation files (the\n * \"Software\"), to deal in the Software without restriction, including\n * without limitation the rights to use, copy, modify, merge, publish,\n * distribute, sublicense, and/or sell copies of the Software, and to\n * permit persons to whom the Software is furnished to do so, subject to\n * the following conditions:\n *\n * The above copyright notice and this permission notice shall be\n * included in all copies or substantial portions of the Software.",
"\n *\n * THE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND,\n * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF\n * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND\n * NONINFRINGEMENT. ",
"IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS\n * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN\n * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN\n * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE\n * SOFTWARE.",
"\n */\n\n/* ADC_Module.cpp: Implements the fuctions of a Teensy 3.x, LC ADC module\n *\n */\n\n\n\n#include \"ADC_Module.h\"\n\n\n/* Constructor\n* Point the registers to the correct ADC module\n* Copy the correct channel2sc1a\n* Call init\n* The very long initializer list could be shorter using some kind of struct?",
"\n*/\nADC_Module::ADC_Module(uint8_t ADC_number, const uint8_t* const a_channel2sc1a, const ADC_NLIST* const a_diff_table) :\n ADC_num(ADC_number)\n , channel2sc1a(a_channel2sc1a)\n , diff_table(a_diff_table)\n , adc_offset((uint32_t)0x20000)\n , ADC_SC1A(&ADC0_SC1A + adc_offset*ADC_num)\n , ADC_SC1B(&ADC0_SC1B + adc_offset*ADC_num)\n , ADC_CFG1(&ADC0_CFG1 + adc_offset*ADC_num)\n , ADC_CFG2(&ADC0_CFG2 + adc_offset*ADC_num)\n , ADC_RA(&ADC0_RA + adc_offset*ADC_num)\n , ADC_RB(&ADC0_RB + adc_offset*ADC_num)\n , ADC_CV1(&ADC0_CV1 + adc_offset*ADC_num)\n , ADC_CV2(&ADC0_CV2 + adc_offset*ADC_num)\n , ADC_SC2(&ADC0_SC2 + adc_offset*ADC_num)\n , ADC_SC3(&ADC0_SC3 + adc_offset*ADC_num)\n , ADC_PGA(&ADC0_PGA + adc_offset*ADC_num)\n , ADC_OFS(&ADC0_OFS + adc_offset*ADC_num)\n , ADC_PG(&ADC0_PG + adc_offset*ADC_num)\n , ADC_MG(&ADC0_MG + adc_offset*ADC_num)\n , ADC_CLPD(&ADC0_CLPD + adc_offset*ADC_num)\n , ADC_CLPS(&ADC0_CLPS + adc_offset*ADC_num)\n , ADC_CLP4(&ADC0_CLP4 + adc_offset*ADC_num)\n , ADC_CLP3(&ADC0_CLP3 + adc_offset*ADC_num)\n , ADC_CLP2(&ADC0_CLP2 + adc_offset*ADC_num)\n , ADC_CLP1(&ADC0_CLP1 + adc_offset*ADC_num)\n , ADC_CLP0(&ADC0_CLP0 + adc_offset*ADC_num)\n , ADC_CLMD(&ADC0_CLMD + adc_offset*ADC_num)\n , ADC_CLMS(&ADC0_CLMS + adc_offset*ADC_num)\n , ADC_CLM4(&ADC0_CLM4 + adc_offset*ADC_num)\n , ADC_CLM3(&ADC0_CLM3 + adc_offset*ADC_num)\n , ADC_CLM2(&ADC0_CLM2 + adc_offset*ADC_num)\n , ADC_CLM1(&ADC0_CLM1 + adc_offset*ADC_num)\n , ADC_CLM0(&ADC0_CLM0 + adc_offset*ADC_num)\n , PDB0_CHnC1(&PDB0_CH0C1 + ADC_num*0xA)\n , IRQ_ADC(IRQ_ADC0 + ADC_num*1)\n {\n\n\n\n\n\n // call our init\n analog_init();\n\n\n\n}\n\n/* Initialize stuff:\n* - Start Vref module\n* - Clear all fail flags\n* - Internal reference (default: external vcc)\n* - Mux between a and b channels (b channels)\n* - Calibrate with 32 averages and low speed\n* - When first calibration is done it sets:\n* - Resolution (default: 10 bits)\n* - Conversion speed and sampling time (both set to medium speed)\n* - Averaging (set to 4)\n*/\nvoid ADC_Module::analog_init() {\n\n // default settings:\n /*\n - 10 bits resolution\n - 4 averages\n - vcc reference\n - no interrupts\n - pga gain=1\n - conversion speed = medium\n - sampling speed = medium\n initiate to 0 (or 1) so the corresponding functions change it to the correct value\n */\n analog_res_bits = 0;\n analog_max_val = 0;\n analog_num_average = 0;\n analog_reference_internal = 2;\n var_enableInterrupts = 0;\n pga_value = 1;\n\n conversion_speed = 0;\n sampling_speed = 0;\n\n calibrating = 0;\n\n fail_flag = ADC_ERROR_CLEAR; // clear all errors\n\n num_measurements = 0;\n\n // select b channels\n // *ADC_CFG2_muxsel = 1;\n setBit(ADC_CFG2, ADC_CFG2_MUXSEL_BIT);\n\n // set reference to vcc\n setReference(ADC_REF_3V3);\n\n // set resolution to 10\n setResolution(10);\n\n // the first calibration will use 32 averages and lowest speed,\n // when this calibration is over the averages and speed will be set to default by wait_for_cal and init_calib will be cleared.",
"\n init_calib = 1;\n setAveraging(32);\n setConversionSpeed(ADC_LOW_SPEED);\n setSamplingSpeed(ADC_LOW_SPEED);\n\n // begin init calibration\n calibrate();\n}\n\n// starts calibration\nvoid ADC_Module::calibrate() {\n\n __disable_irq();\n\n calibrating = 1;\n // *ADC_SC3_cal = 0; // stop possible previous calibration\n clearBit(ADC_SC3, ADC_SC3_CAL_BIT);\n // *ADC_SC3_calf = 1; // clear possible previous error\n setBit(ADC_SC3, ADC_SC3_CALF_BIT);\n // *ADC_SC3_cal = 1; // start calibration\n setBit(ADC_SC3, ADC_SC3_CAL_BIT);\n\n __enable_irq();\n}\n\n\n/* Waits until calibration is finished and writes the corresponding registers\n*\n*/\nvoid ADC_Module::wait_for_cal(void) {\n uint16_t sum;\n\n while(getBit(ADC_SC3, ADC_SC3_CAL_BIT)) { // Bit ADC_SC3_CAL in register ADC0_SC3 cleared when calib. ",
"finishes.",
"\n yield();\n }\n\n if(getBit(ADC_SC3, ADC_SC3_CALF_BIT)) { // calibration failed\n fail_flag |= ADC_ERROR_CALIB; // the user should know and recalibrate manually\n }\n\n __disable_irq();\n if (calibrating) {\n sum = *ADC_CLPS + *ADC_CLP4 + *ADC_CLP3 + *ADC_CLP2 + *ADC_CLP1 + *ADC_CLP0;\n sum = (sum / 2) | 0x8000;\n *ADC_PG = sum;\n\n sum = *ADC_CLMS + *ADC_CLM4 + *ADC_CLM3 + *ADC_CLM2 + *ADC_CLM1 + *ADC_CLM0;\n sum = (sum / 2) | 0x8000;\n *ADC_MG = sum;\n\n calibrating = 0;\n }\n __enable_irq();\n\n // the first calibration uses 32 averages and lowest speed,\n // when this calibration is over, set the averages and speed to default.",
"\n if(init_calib) {\n\n // set conversion speed to medium\n setConversionSpeed(ADC_MED_SPEED);\n\n // set sampling speed to medium\n setSamplingSpeed(ADC_MED_SPEED);\n\n // number of averages to 4\n setAveraging(4);\n\n init_calib = 0; // clear\n }\n\n}\n\n//! ",
"Starts the calibration sequence, waits until it's done and writes the results\n/** Usually it's not necessary to call this function directly, but do it if the \"enviroment\" changed\n* significantly since the program was started.",
"\n*/\nvoid ADC_Module::recalibrate() {\n\n calibrate();\n\n wait_for_cal();\n}\n\n\n\n/////////////// METHODS TO SET/GET SETTINGS OF THE ADC ////////////////////\n\n\n/* Set the voltage reference you prefer, default is 3.3V\n* It needs to recalibrate\n* Use ADC_REF_3V3, ADC_REF_1V2 (not for Teensy LC) or ADC_REF_EXT\n*/\nvoid ADC_Module::setReference(uint8_t type) {\n if (analog_reference_internal==type) { // don't need to change anything\n return;\n }\n\n\n\n if (type == ADC_REF_ALT) { // 1.2V ref for Teensy 3.x, 3.3 VDD for Teensy LC\n // internal reference requested\n\n startInternalReference(); // enable VREF if Teensy 3.x\n\n analog_reference_internal = ADC_REF_ALT;\n\n // *ADC_SC2_ref = 1; // uses bitband: atomic\n setBit(ADC_SC2, ADC_SC2_REFSEL0_BIT);\n\n } else if(type == ADC_REF_DEFAULT) { // ext ref for all Teensys, vcc also for Teensy 3.x\n // vcc or external reference requested\n\n stopInternalReference(); // disable 1.2V reference source when using the external ref (p. 102, 3.7.1.7)\n\n analog_reference_internal = ADC_REF_DEFAULT;\n\n // *ADC_SC2_ref = 0; // uses bitband: atomic\n clearBit(ADC_SC2, ADC_SC2_REFSEL0_BIT);\n }\n\n calibrate();\n}\n\n//! ",
"Start the 1.2V internal reference (if present)\nvoid ADC_Module::startInternalReference() {\n#if ADC_USE_INTERNAL\n VREF_TRM = VREF_TRM_CHOPEN | 0x20; // enable module and set the trimmer to medium (max=0x3F=63)\n VREF_SC = VREF_SC_VREFEN | VREF_SC_REGEN | VREF_SC_ICOMPEN | VREF_SC_MODE_LV(1); // (=0xE1) enable 1.2 volt ref with all compensations\n#endif\n}\n\n//! ",
"Stops the internal reference\nvoid ADC_Module::stopInternalReference() {\n#if ADC_USE_INTERNAL\n VREF_SC = 0;\n#endif\n}\n\n\n/* Change the resolution of the measurement\n* For single-ended measurements: 8, 10, 12 or 16 bits.",
"\n* For differential measurements: 9, 11, 13 or 16 bits.",
"\n* If you want something in between (11 bits single-ended for example) select the inmediate higher\n* and shift the result one to the right.",
"\n*\n* It doesn't recalibrate\n*/\nvoid ADC_Module::setResolution(uint8_t bits) {\n\n if(analog_res_bits==bits) {\n return;\n }\n\n uint8_t config;\n\n if (calibrating) wait_for_cal();\n\n if (bits <8) {\n config = 8;\n } else if (bits >= 14) {\n config = 16;\n } else {\n config = bits;\n }\n\n // conversion resolution\n // single-ended 8 bits is the same as differential 9 bits, etc.",
"\n if ( (config == 8) || (config == 9) ) {\n // *ADC_CFG1_mode1 = 0;\n // *ADC_CFG1_mode0 = 0;\n clearBit(ADC_CFG1, ADC_CFG1_MODE1_BIT);\n clearBit(ADC_CFG1, ADC_CFG1_MODE0_BIT);\n analog_max_val = 255; // diff mode 9 bits has 1 bit for sign, so max value is the same as single 8 bits\n } else if ( (config == 10 )|| (config == 11) ) {\n // *ADC_CFG1_mode1 = 1;\n // *ADC_CFG1_mode0 = 0;\n setBit(ADC_CFG1, ADC_CFG1_MODE1_BIT);\n clearBit(ADC_CFG1, ADC_CFG1_MODE0_BIT);\n analog_max_val = 1023;\n } else if ( (config == 12 )|| (config == 13) ) {\n // *ADC_CFG1_mode1 = 0;\n // *ADC_CFG1_mode0 = 1;\n clearBit(ADC_CFG1, ADC_CFG1_MODE1_BIT);\n setBit(ADC_CFG1, ADC_CFG1_MODE0_BIT);\n analog_max_val = 4095;\n } else {\n // *ADC_CFG1_mode1 = 1;\n // *ADC_CFG1_mode0 = 1;\n setBit(ADC_CFG1, ADC_CFG1_MODE1_BIT);\n setBit(ADC_CFG1, ADC_CFG1_MODE0_BIT);\n analog_max_val = 65535;\n }\n\n analog_res_bits = config;\n\n // no recalibration is needed when changing the resolution, p. 619\n\n}\n\n/* Returns the resolution of the ADC\n*\n*/\nuint8_t ADC_Module::getResolution() {\n return analog_res_bits;\n}\n\n/* Returns the maximum value for a measurement, that is: 2^resolution-1\n*\n*/\nuint32_t ADC_Module::getMaxValue() {\n return analog_max_val;\n}\n\n\n// Sets the conversion speed\n/*\n* \\param speed can be ADC_VERY_LOW_SPEED, ADC_LOW_SPEED, ADC_MED_SPEED, ADC_HIGH_SPEED_16BITS, ADC_HIGH_SPEED or ADC_VERY_HIGH_SPEED\n ADC_VERY_LOW_SPEED is guaranteed to be the lowest possible speed within specs for resolutions less than 16 bits (higher than 1 MHz),\n it's different from ADC_LOW_SPEED only for 24, 4 or 2 MHz.",
"\n ADC_LOW_SPEED is guaranteed to be the lowest possible speed within specs for all resolutions (higher than 2 MHz).",
"\n ADC_MED_SPEED is always >= ADC_LOW_SPEED and <= ADC_HIGH_SPEED.",
"\n ADC_HIGH_SPEED_16BITS is guaranteed to be the highest possible speed within specs for all resolutions (lower or eq than 12 MHz).",
"\n ADC_HIGH_SPEED is guaranteed to be the highest possible speed within specs for resolutions less than 16 bits (lower or eq than 18 MHz).",
"\n ADC_VERY_HIGH_SPEED may be out of specs, it's different from ADC_HIGH_SPEED only for 48, 40 or 24 MHz.",
"\n* It doesn't recalibrate at the end.",
"\n*/\nvoid ADC_Module::setConversionSpeed(uint8_t speed) {\n\n if(speed==conversion_speed) { // no change\n return;\n }\n\n if (calibrating) wait_for_cal();\n\n // internal asynchronous clock settings: fADK = 2.4, 4.0, 5.2 or 6.2 MHz\n if(speed >= ADC_ADACK_2_4) {\n setBit(ADC_CFG2, ADC_CFG2_ADACKEN_BIT); // enable ADACK (takes max 5us to be ready)\n setBit(ADC_CFG1, ADC_CFG1_ADICLK1_BIT); // select ADACK as clock source\n setBit(ADC_CFG1, ADC_CFG1_ADICLK0_BIT);\n\n clearBit(ADC_CFG1, ADC_CFG1_ADIV0_BIT); // select divider 1\n clearBit(ADC_CFG1, ADC_CFG1_ADIV1_BIT); // we could divide this clk, but it would be too small for ADC use.",
"\n\n if(speed == ADC_ADACK_2_4) {\n clearBit(ADC_CFG2, ADC_CFG2_ADHSC_BIT);\n setBit(ADC_CFG1, ADC_CFG1_ADLPC_BIT);\n } else if(speed == ADC_ADACK_4_0) {\n setBit(ADC_CFG2, ADC_CFG2_ADHSC_BIT);\n setBit(ADC_CFG1, ADC_CFG1_ADLPC_BIT);\n } else if(speed == ADC_ADACK_5_2) {\n clearBit(ADC_CFG2, ADC_CFG2_ADHSC_BIT);\n clearBit(ADC_CFG1, ADC_CFG1_ADLPC_BIT);\n } else if(speed == ADC_ADACK_6_2) {\n setBit(ADC_CFG2, ADC_CFG2_ADHSC_BIT);\n clearBit(ADC_CFG1, ADC_CFG1_ADLPC_BIT);\n }\n conversion_speed = speed;\n return;\n }\n\n\n // normal bus clock used\n\n // *ADC_CFG2_adacken = 0; // disable the internal asynchronous clock\n clearBit(ADC_CFG2, ADC_CFG2_ADACKEN_BIT);\n\n uint32_t ADC_CFG1_speed; // store the clock and divisor\n\n if(speed == ADC_VERY_LOW_SPEED) {\n // *ADC_CFG2_adhsc = 0; // no high-speed config\n // *ADC_CFG1_adlpc = 1; // use low power conf.",
"\n clearBit(ADC_CFG2, ADC_CFG2_ADHSC_BIT);\n setBit(ADC_CFG1, ADC_CFG1_ADLPC_BIT);\n\n ADC_CFG1_speed = ADC_CFG1_VERY_LOW_SPEED;\n\n } else if(speed == ADC_LOW_SPEED) {\n // *ADC_CFG2_adhsc = 0; // no high-speed config\n // *ADC_CFG1_adlpc = 1; // use low power conf.",
"\n clearBit(ADC_CFG2, ADC_CFG2_ADHSC_BIT);\n setBit(ADC_CFG1, ADC_CFG1_ADLPC_BIT);\n\n ADC_CFG1_speed = ADC_CFG1_LOW_SPEED;\n\n } else if(speed == ADC_MED_SPEED) {\n // *ADC_CFG2_adhsc = 0; // no high-speed config\n // *ADC_CFG1_adlpc = 0; // no low power conf.",
"\n clearBit(ADC_CFG2, ADC_CFG2_ADHSC_BIT);\n clearBit(ADC_CFG1, ADC_CFG1_ADLPC_BIT);\n\n ADC_CFG1_speed = ADC_CFG1_MED_SPEED;\n\n } else if(speed == ADC_HIGH_SPEED_16BITS) {\n // *ADC_CFG2_adhsc = 1; // high-speed config: add 2 ADCK\n // *ADC_CFG1_adlpc = 0; // no low power conf.",
"\n setBit(ADC_CFG2, ADC_CFG2_ADHSC_BIT);\n clearBit(ADC_CFG1, ADC_CFG1_ADLPC_BIT);\n\n ADC_CFG1_speed = ADC_CFG1_HI_SPEED_16_BITS;\n\n } else if(speed == ADC_HIGH_SPEED) {\n // *ADC_CFG2_adhsc = 1; // high-speed config: add 2 ADCK\n // *ADC_CFG1_adlpc = 0; // no low power conf.",
"\n setBit(ADC_CFG2, ADC_CFG2_ADHSC_BIT);\n clearBit(ADC_CFG1, ADC_CFG1_ADLPC_BIT);\n\n ADC_CFG1_speed = ADC_CFG1_HI_SPEED;\n\n } else if(speed == ADC_VERY_HIGH_SPEED) { // this speed is most likely out of specs, so accurancy can be bad\n // *ADC_CFG2_adhsc = 1; // high-speed config: add 2 ADCK\n // *ADC_CFG1_adlpc = 0; // no low power conf.",
"\n setBit(ADC_CFG2, ADC_CFG2_ADHSC_BIT);\n clearBit(ADC_CFG1, ADC_CFG1_ADLPC_BIT);\n\n ADC_CFG1_speed = ADC_CFG1_VERY_HIGH_SPEED;\n\n } else {\n fail_flag |= ADC_ERROR_OTHER;\n return;\n }\n\n // clock source is bus or bus/2\n // *ADC_CFG1_adiclk1 = !!(",
"ADC_CFG1_speed & ADC_CFG1_ADICLK_MASK_1); // !!",
"x converts the number x to either 0 or 1.",
"\n // *ADC_CFG1_adiclk0 = !!(",
"ADC_CFG1_speed & ADC_CFG1_ADICLK_MASK_0);\n changeBit(ADC_CFG1, ADC_CFG1_ADICLK1_BIT, !!(",
"ADC_CFG1_speed & ADC_CFG1_ADICLK_MASK_1));\n changeBit(ADC_CFG1, ADC_CFG1_ADICLK0_BIT, !!(",
"ADC_CFG1_speed & ADC_CFG1_ADICLK_MASK_0));\n\n // divisor for the clock source: 1, 2, 4 or 8.",
"\n // so total speed can be: bus, bus/2, bus/4, bus/8 or bus/16.",
"\n // *ADC_CFG1_adiv1 = !!(",
"ADC_CFG1_speed & ADC_CFG1_ADIV_MASK_1);\n // *ADC_CFG1_adiv0 = !!(",
"ADC_CFG1_speed & ADC_CFG1_ADIV_MASK_0);\n changeBit(ADC_CFG1, ADC_CFG1_ADIV1_BIT, !!(",
"ADC_CFG1_speed & ADC_CFG1_ADIV_MASK_1));\n changeBit(ADC_CFG1, ADC_CFG1_ADIV0_BIT, !!(",
"ADC_CFG1_speed & ADC_CFG1_ADIV_MASK_0));\n\n conversion_speed = speed;\n\n}\n\n\n// Sets the sampling speed\n/* Increase the sampling speed for low impedance sources, decrease it for higher impedance ones.",
"\n* \\param speed can be ADC_VERY_LOW_SPEED, ADC_LOW_SPEED, ADC_MED_SPEED, ADC_HIGH_SPEED or ADC_VERY_HIGH_SPEED\n ADC_VERY_LOW_SPEED is the lowest possible sampling speed (+24 ADCK).",
"\n ADC_LOW_SPEED adds +16 ADCK.",
"\n ADC_MED_SPEED adds +10 ADCK.",
"\n ADC_HIGH_SPEED (or ADC_HIGH_SPEED_16BITS) adds +6 ADCK.",
"\n ADC_VERY_HIGH_SPEED is the highest possible sampling speed (0 ADCK added).",
"\n* It doesn't recalibrate at the end.",
"\n*/\nvoid ADC_Module::setSamplingSpeed(uint8_t speed) {\n\n if(speed==sampling_speed) { // no change\n return;\n }\n\n if (calibrating) wait_for_cal();\n\n // Select between the settings\n if(speed == ADC_VERY_LOW_SPEED) {\n // *ADC_CFG1_adlsmp = 1; // long sampling time enable\n // *ADC_CFG2_adlsts1 = 0; // maximum sampling time (+24 ADCK)\n // *ADC_CFG2_adlsts0 = 0;\n setBit(ADC_CFG1, ADC_CFG1_ADLSMP_BIT);\n clearBit(ADC_CFG2, ADC_CFG2_ADLSTS1_BIT);\n clearBit(ADC_CFG2, ADC_CFG2_ADLSTS0_BIT);\n\n } else if(speed == ADC_LOW_SPEED) {\n // *ADC_CFG1_adlsmp = 1; // long sampling time enable\n // *ADC_CFG2_adlsts1 = 0;// high sampling time (+16 ADCK)\n // *ADC_CFG2_adlsts0 = 1;\n setBit(ADC_CFG1, ADC_CFG1_ADLSMP_BIT);\n clearBit(ADC_CFG2, ADC_CFG2_ADLSTS1_BIT);\n setBit(ADC_CFG2, ADC_CFG2_ADLSTS0_BIT);\n\n } else if(speed == ADC_MED_SPEED) {\n // *ADC_CFG1_adlsmp = 1; // long sampling time enable\n // *ADC_CFG2_adlsts1 = 1;// medium sampling time (+10 ADCK)\n // *ADC_CFG2_adlsts0 = 0;\n setBit(ADC_CFG1, ADC_CFG1_ADLSMP_BIT);\n setBit(ADC_CFG2, ADC_CFG2_ADLSTS1_BIT);\n clearBit(ADC_CFG2, ADC_CFG2_ADLSTS0_BIT);\n\n } else if( (speed == ADC_HIGH_SPEED) || (speed == ADC_HIGH_SPEED_16BITS) ) {\n // *ADC_CFG1_adlsmp = 1; // long sampling time enable\n // *ADC_CFG2_adlsts1 = 1;// low sampling time (+6 ADCK)\n // *ADC_CFG2_adlsts0 = 1;\n setBit(ADC_CFG1, ADC_CFG1_ADLSMP_BIT);\n setBit(ADC_CFG2, ADC_CFG2_ADLSTS1_BIT);\n setBit(ADC_CFG2, ADC_CFG2_ADLSTS0_BIT);\n\n } else if(speed == ADC_VERY_HIGH_SPEED) {\n // *ADC_CFG1_adlsmp = 0; // shortest sampling time\n clearBit(ADC_CFG1, ADC_CFG1_ADLSMP_BIT);\n\n } else { // incorrect speeds have no effect.",
"\n return;\n }\n\n sampling_speed = speed;\n\n}\n\n\n/* Set the number of averages: 0, 4, 8, 16 or 32.",
"\n*\n*/\nvoid ADC_Module::setAveraging(uint8_t num) {\n\n if (calibrating) wait_for_cal();\n\n if (num <= 1) {\n num = 0;\n // *ADC_SC3_avge = 0;\n clearBit(ADC_SC3, ADC_SC3_AVGE_BIT);\n } else {\n // *ADC_SC3_avge = 1;\n setBit(ADC_SC3, ADC_SC3_AVGE_BIT);\n if (num <= 4) {\n num = 4;\n // *ADC_SC3_avgs0 = 0;\n // *ADC_SC3_avgs1 = 0;\n clearBit(ADC_SC3, ADC_SC3_AVGS0_BIT);\n clearBit(ADC_SC3, ADC_SC3_AVGS1_BIT);\n } else if (num <= 8) {\n num = 8;\n // *ADC_SC3_avgs0 = 1;\n // *ADC_SC3_avgs1 = 0;\n setBit(ADC_SC3, ADC_SC3_AVGS0_BIT);\n clearBit(ADC_SC3, ADC_SC3_AVGS1_BIT);\n } else if (num <= 16) {\n num = 16;\n // *ADC_SC3_avgs0 = 0;\n // *ADC_SC3_avgs1 = 1;\n clearBit(ADC_SC3, ADC_SC3_AVGS0_BIT);\n setBit(ADC_SC3, ADC_SC3_AVGS1_BIT);\n } else {\n num = 32;\n // *ADC_SC3_avgs0 = 1;\n // *ADC_SC3_avgs1 = 1;\n setBit(ADC_SC3, ADC_SC3_AVGS0_BIT);\n setBit(ADC_SC3, ADC_SC3_AVGS1_BIT);\n }\n }\n analog_num_average = num;\n}\n\n\n/* Enable interrupts: An ADC Interrupt will be raised when the conversion is completed\n* (including hardware averages and if the comparison (if any) is true).",
"\n*/\nvoid ADC_Module::enableInterrupts() {\n\n if (calibrating) wait_for_cal();\n\n var_enableInterrupts = 1;\n // *ADC_SC1A_aien = 1;\n setBit(ADC_SC1A, ADC_SC1A_AIEN_BIT);\n\n NVIC_ENABLE_IRQ(IRQ_ADC);\n}\n\n/* Disable interrupts\n*\n*/\nvoid ADC_Module::disableInterrupts() {\n\n var_enableInterrupts = 0;\n // *ADC_SC1A_aien = 0;\n clearBit(ADC_SC1A, ADC_SC1A_AIEN_BIT);\n\n NVIC_DISABLE_IRQ(IRQ_ADC);\n\n}\n\n\n/* Enable DMA request: An ADC DMA request will be raised when the conversion is completed\n* (including hardware averages and if the comparison (if any) is true).",
"\n*/\nvoid ADC_Module::enableDMA() {\n\n if (calibrating) wait_for_cal();\n\n // *ADC_SC2_dma = 1;\n setBit(ADC_SC2, ADC_SC2_DMAEN_BIT);\n}\n\n/* Disable ADC DMA request\n*\n*/\nvoid ADC_Module::disableDMA() {\n\n // *ADC_SC2_dma = 0;\n clearBit(ADC_SC2, ADC_SC2_DMAEN_BIT);\n}\n\n\n/* Enable the compare function: A conversion will be completed only when the ADC value\n* is >= compValue (greaterThan=1) or < compValue (greaterThan=0)\n* Call it after changing the resolution\n* Use with interrupts or poll conversion completion with isADC_Complete()\n*/\nvoid ADC_Module::enableCompare(int16_t compValue, bool greaterThan) {\n\n if (calibrating) wait_for_cal(); // if we modify the adc's registers when calibrating, it will fail\n\n // *ADC_SC2_cfe = 1; // enable compare\n // *ADC_SC2_cfgt = (int32_t)greaterThan; // greater or less than?",
"\n setBit(ADC_SC2, ADC_SC2_ACFE_BIT);\n changeBit(ADC_SC2, ADC_SC2_ACFGT_BIT, greaterThan);\n\n *ADC_CV1 = (int16_t)compValue; // comp value\n}\n\n/* Enable the compare function: A conversion will be completed only when the ADC value\n* is inside (insideRange=1) or outside (=0) the range given by (lowerLimit, upperLimit),\n* including (inclusive=1) the limits or not (inclusive=0).",
"\n* See Table 31-78, p. 617 of the freescale manual.",
"\n* Call it after changing the resolution\n*/\nvoid ADC_Module::enableCompareRange(int16_t lowerLimit, int16_t upperLimit, bool insideRange, bool inclusive) {\n\n if (calibrating) wait_for_cal(); // if we modify the adc's registers when calibrating, it will fail\n\n // *ADC_SC2_cfe = 1; // enable compare\n // *ADC_SC2_cren = 1; // enable compare range\n setBit(ADC_SC2, ADC_SC2_ACFE_BIT);\n setBit(ADC_SC2, ADC_SC2_ACREN_BIT);\n\n if(insideRange && inclusive) { // True if value is inside the range, including the limits. ",
"CV1 <= CV2 and ACFGT=1\n // *ADC_SC2_cfgt = 1;\n setBit(ADC_SC2, ADC_SC2_ACFGT_BIT);\n\n *ADC_CV1 = (int16_t)lowerLimit;\n *ADC_CV2 = (int16_t)upperLimit;\n } else if(insideRange && !",
"inclusive) {// True if value is inside the range, excluding the limits. ",
"CV1 > CV2 and ACFGT=0\n // *ADC_SC2_cfgt = 0;\n clearBit(ADC_SC2, ADC_SC2_ACFGT_BIT);\n\n *ADC_CV2 = (int16_t)lowerLimit;\n *ADC_CV1 = (int16_t)upperLimit;\n } else if(!insideRange && inclusive) { // True if value is outside of range or is equal to either limit. ",
"CV1 > CV2 and ACFGT=1\n // *ADC_SC2_cfgt = 1;\n setBit(ADC_SC2, ADC_SC2_ACFGT_BIT);\n\n *ADC_CV2 = (int16_t)lowerLimit;\n *ADC_CV1 = (int16_t)upperLimit;\n } else if(!insideRange && !",
"inclusive) { // True if value is outside of range and not equal to either limit. ",
"CV1 > CV2 and ACFGT=0\n // *ADC_SC2_cfgt = 0;\n clearBit(ADC_SC2, ADC_SC2_ACFGT_BIT);\n\n *ADC_CV1 = (int16_t)lowerLimit;\n *ADC_CV2 = (int16_t)upperLimit;\n }\n}\n\n/* Disable the compare function\n*\n*/\nvoid ADC_Module::disableCompare() {\n\n // *ADC_SC2_cfe = 0;\n clearBit(ADC_SC2, ADC_SC2_ACFE_BIT);\n}\n\n/* Enables the PGA and sets the gain\n* Use only for signals lower than 1.2 V\n* \\param gain can be 1, 2, 4, 8, 16 32 or 64\n*\n*/\nvoid ADC_Module::enablePGA(uint8_t gain) {\n#if ADC_USE_PGA\n\n if (calibrating) wait_for_cal();\n\n uint8_t setting;\n if(gain <= 1) {\n setting = 0;\n } else if(gain<=2) {\n setting = 1;\n } else if(gain<=4) {\n setting = 2;\n } else if(gain<=8) {\n setting = 3;\n } else if(gain<=16) {\n setting = 4;\n } else if(gain<=32) {\n setting = 5;\n } else { // 64\n setting = 6;\n }\n\n *ADC_PGA = ADC_PGA_PGAEN | ADC_PGA_PGAG(setting);\n pga_value=1<<setting;\n#endif\n}\n\n/* Returns the PGA level\n* PGA level = from 0 to 64\n*/\nuint8_t ADC_Module::getPGA() {\n return pga_value;\n}\n\n//! ",
"Disable PGA\nvoid ADC_Module::disablePGA() {\n#if ADC_USE_PGA\n // *ADC_PGA_pgaen = 0;\n clearBit(ADC_PGA, ADC_PGA_PGAEN_BIT);\n#endif\n pga_value = 1;\n}\n\n\n//////////////// INFORMATION ABOUT VALID PINS //////////////////\n\n// check whether the pin is a valid analog pin\nbool ADC_Module::checkPin(uint8_t pin) {\n\n if(pin>ADC_MAX_PIN) {\n return false; // all others are invalid\n }\n\n // translate pin number to SC1A number, that also contains MUX a or b info.",
"\n const uint8_t sc1a_pin = channel2sc1a[pin];\n\n // check for valid pin\n if( (sc1a_pin&ADC_SC1A_CHANNELS) == ADC_SC1A_PIN_INVALID ) {\n return false; // all others are invalid\n }\n\n return true;\n}\n\n// check whether the pins are a valid analog differential pins (including PGA if enabled)\nbool ADC_Module::checkDifferentialPins(uint8_t pinP, uint8_t pinN) {\n if(pinP>ADC_MAX_PIN) {\n return false; // all others are invalid\n }\n\n // translate pinP number to SC1A number, to make sure it's differential\n uint8_t sc1a_pin = channel2sc1a[pinP];\n\n if( !(",
"sc1a_pin&ADC_SC1A_PIN_DIFF) ) {\n return false; // all others are invalid\n }\n\n // get SC1A number, also whether it can do PGA\n sc1a_pin = getDifferentialPair(pinP);\n\n // the pair can't be measured with this ADC\n if( (sc1a_pin&ADC_SC1A_CHANNELS) == ADC_SC1A_PIN_INVALID ) {\n fail_flag |= ADC_ERROR_WRONG_PIN;\n return false; // all others are invalid\n }\n\n #if ADC_USE_PGA\n // check if PGA is enabled, and whether the pin has access to it in this ADC module\n if( isPGAEnabled() && !(",
"sc1a_pin&ADC_SC1A_PIN_PGA) ) {\n return false;\n }\n #endif // ADC_USE_PGA\n\n return true;\n}\n\n\n//////////////// HELPER METHODS FOR CONVERSION /////////////////\n\n// Starts a single-ended conversion on the pin (sets the mux correctly)\n// Doesn't do any of the checks on the pin\n// It doesn't change the continuous conversion bit\nvoid ADC_Module::startReadFast(uint8_t pin) {\n\n // translate pin number to SC1A number, that also contains MUX a or b info.",
"\n const uint8_t sc1a_pin = channel2sc1a[pin];\n\n if(sc1a_pin&ADC_SC1A_PIN_MUX) { // mux a\n clearBit(ADC_CFG2, ADC_CFG2_MUXSEL_BIT);\n } else { // mux b\n setBit(ADC_CFG2, ADC_CFG2_MUXSEL_BIT);\n }\n\n // select pin for single-ended mode and start conversion, enable interrupts if requested\n __disable_irq();\n *ADC_SC1A = (sc1a_pin&ADC_SC1A_CHANNELS) + var_enableInterrupts*ADC_SC1_AIEN;\n __enable_irq();\n\n}\n\n// Starts a differential conversion on the pair of pins\n// Doesn't do any of the checks on the pins\n// It doesn't change the continuous conversion bit\nvoid ADC_Module::startDifferentialFast(uint8_t pinP, uint8_t pinN) {\n\n // get SC1A number\n uint8_t sc1a_pin = getDifferentialPair(pinP);\n\n #if ADC_USE_PGA\n // check if PGA is enabled\n if(isPGAEnabled()) {\n sc1a_pin = 0x2; // PGA always uses DAD2\n }\n #endif // ADC_USE_PGA\n\n __disable_irq();\n *ADC_SC1A = ADC_SC1_DIFF + (sc1a_pin&ADC_SC1A_CHANNELS) + var_enableInterrupts*ADC_SC1_AIEN;\n __enable_irq();\n\n}\n\n\n\n//////////////// BLOCKING CONVERSION METHODS //////////////////\n/*\n This methods are implemented like this:\n\n 1. ",
"Check that the pin is correct\n 2. ",
"if calibrating, wait for it to finish before modifiying any ADC register\n 3. ",
"Check if we're interrupting a measurement, if so store the settings.",
"\n 4. ",
"Disable continuous conversion mode and start the current measurement\n 5. ",
"Wait until it's done, and check whether the comparison (if any) was succesful.",
"\n 6. ",
"Get the result.",
"\n 7. ",
"If step 3. ",
"is true, restore the previous ADC settings\n\n*/\n\n\n/* Reads the analog value of the pin.",
"\n* It waits until the value is read and then returns the result.",
"\n* If a comparison has been set up and fails, it will return ADC_ERROR_VALUE.",
"\n* Set the resolution, number of averages and voltage reference using the appropriate functions.",
"\n*/\nint ADC_Module::analogRead(uint8_t pin) {\n\n //digitalWriteFast(LED_BUILTIN, HIGH);\n\n // check whether the pin is correct\n if(!checkPin(pin)) {\n fail_flag |= ADC_ERROR_WRONG_PIN;\n return ADC_ERROR_VALUE;\n }\n\n // increase the counter of measurements\n num_measurements++;\n\n //digitalWriteFast(LED_BUILTIN, !",
"digitalReadFast(LED_BUILTIN));\n\n if (calibrating) wait_for_cal();\n\n //digitalWriteFast(LED_BUILTIN, !",
"digitalReadFast(LED_BUILTIN));\n\n // check if we are interrupting a measurement, store setting if so.",
"\n // vars to save the current state of the ADC in case it's in use\n ADC_Config old_config = {0};\n uint8_t wasADCInUse = isConverting(); // is the ADC running now?",
"\n\n if(wasADCInUse) { // this means we're interrupting a conversion\n // save the current conversion config, we don't want any other interrupts messing up the configs\n __disable_irq();\n //digitalWriteFast(LED_BUILTIN, !",
"digitalReadFast(LED_BUILTIN) );\n saveConfig(&old_config);\n __enable_irq();\n }\n\n\n // no continuous mode\n singleMode();\n\n startReadFast(pin); // start single read\n\n // wait for the ADC to finish\n while(isConverting()) {\n yield();\n }\n\n // it's done, check if the comparison (if any) was true\n int32_t result;\n __disable_irq(); // make sure nothing interrupts this part\n if (isComplete()) { // conversion succeded\n result = (uint16_t)*ADC_RA;\n } else { // comparison was false\n fail_flag |= ADC_ERROR_COMPARISON;\n result = ADC_ERROR_VALUE;\n }\n __enable_irq();\n\n // if we interrupted a conversion, set it again\n if (wasADCInUse) {\n //digitalWriteFast(LED_BUILTIN, !",
"digitalReadFast(LED_BUILTIN) );\n __disable_irq();\n loadConfig(&old_config);\n __enable_irq();\n }\n\n num_measurements--;\n return result;\n\n} // analogRead\n\n\n\n/* Reads the differential analog value of two pins (pinP - pinN)\n* It waits until the value is read and then returns the result\n* If a comparison has been set up and fails, it will return ADC_ERROR_DIFF_VALUE\n* Set the resolution, number of averages and voltage reference using the appropriate functions\n*/\nint ADC_Module::analogReadDifferential(uint8_t pinP, uint8_t pinN) {\n\n if(!checkDifferentialPins(pinP, pinN)) {\n fail_flag |= ADC_ERROR_WRONG_PIN;\n return ADC_ERROR_VALUE; // all others are invalid\n }\n\n // increase the counter of measurements\n num_measurements++;\n\n // check for calibration before setting channels,\n // because conversion will start as soon as we write to *ADC_SC1A\n if (calibrating) wait_for_cal();\n\n uint8_t res = getResolution();\n\n // vars to saved the current state of the ADC in case it's in use\n ADC_Config old_config = {0};\n uint8_t wasADCInUse = isConverting(); // is the ADC running now?",
"\n\n if(wasADCInUse) { // this means we're interrupting a conversion\n // save the current conversion config, we don't want any other interrupts messing up the configs\n __disable_irq();\n saveConfig(&old_config);\n __enable_irq();\n }\n\n // no continuous mode\n singleMode();\n\n startDifferentialFast(pinP, pinN); // start conversion\n\n // wait for the ADC to finish\n while( isConverting() ) {\n yield();\n //digitalWriteFast(LED_BUILTIN, !",
"digitalReadFast(LED_BUILTIN) );\n }\n\n // it's done, check if the comparison (if any) was true\n int32_t result;\n __disable_irq(); // make sure nothing interrupts this part\n if (isComplete()) { // conversion succeded\n result = (int16_t)(int32_t)(*ADC_RA); // cast to 32 bits\n if(res==16) { // 16 bit differential is actually 15 bit + 1 bit sign\n result *= 2; // multiply by 2 as if it were really 16 bits, so that getMaxValue gives a correct value.",
"\n }\n } else { // comparison was false\n result = ADC_ERROR_VALUE;\n fail_flag |= ADC_ERROR_COMPARISON;\n }\n __enable_irq();\n\n // if we interrupted a conversion, set it again\n if (wasADCInUse) {\n __disable_irq();\n loadConfig(&old_config);\n __enable_irq();\n }\n\n num_measurements--;\n return result;\n\n} // analogReadDifferential\n\n\n\n/////////////// NON-BLOCKING CONVERSION METHODS //////////////\n/*\n This methods are implemented like this:\n\n 1. ",
"Check that the pin is correct\n 2. ",
"if calibrating, wait for it to finish before modifiying any ADC register\n 3. ",
"Check if we're interrupting a measurement, if so store the settings (in a member of the class, so it can be accessed).",
"\n 4. ",
"Disable continuous conversion mode and start the current measurement\n\n The fast methods only do step 4.",
"\n\n*/\n\n\n/* Starts an analog measurement on the pin.",
"\n* It returns inmediately, read value with readSingle().",
"\n* If the pin is incorrect it returns ADC_ERROR_VALUE.",
"\n*/\nbool ADC_Module::startSingleRead(uint8_t pin) {\n\n // check whether the pin is correct\n if(!checkPin(pin)) {\n fail_flag |= ADC_ERROR_WRONG_PIN;\n return false;\n }\n\n if (calibrating) wait_for_cal();\n\n // save the current state of the ADC in case it's in use\n adcWasInUse = isConverting(); // is the ADC running now?",
"\n\n if(adcWasInUse) { // this means we're interrupting a conversion\n // save the current conversion config, the adc isr will restore the adc\n __disable_irq();\n saveConfig(&adc_config);\n __enable_irq();\n }\n\n // no continuous mode\n singleMode();\n\n // start measurement\n startReadFast(pin);\n\n return true;\n}\n\n\n/* Start a differential conversion between two pins (pinP - pinN).",
"\n* It returns inmediately, get value with readSingle().",
"\n* Incorrect pins will return ADC_ERROR_DIFF_VALUE.",
"\n* Set the resolution, number of averages and voltage reference using the appropriate functions\n*/\nbool ADC_Module::startSingleDifferential(uint8_t pinP, uint8_t pinN) {\n\n if(!checkDifferentialPins(pinP, pinN)) {\n fail_flag |= ADC_ERROR_WRONG_PIN;\n return false; // all others are invalid\n }\n\n // check for calibration before setting channels,\n // because conversion will start as soon as we write to *ADC_SC1A\n if (calibrating) wait_for_cal();\n\n // vars to saved the current state of the ADC in case it's in use\n adcWasInUse = isConverting(); // is the ADC running now?",
"\n\n if(adcWasInUse) { // this means we're interrupting a conversion\n // save the current conversion config, we don't want any other interrupts messing up the configs\n __disable_irq();\n saveConfig(&adc_config);\n __enable_irq();\n }\n\n // no continuous mode\n singleMode();\n\n // start the conversion\n startDifferentialFast(pinP, pinN);\n\n return true;\n}\n\n\n\n///////////// CONTINUOUS CONVERSION METHODS ////////////\n/*\n This methods are implemented like this:\n\n 1. ",
"Check that the pin is correct\n 2. ",
"If calibrating, wait for it to finish before modifiying any ADC register\n 4. ",
"Enable continuous conversion mode and start the current measurement\n\n*/\n\n/* Starts continuous conversion on the pin\n * It returns as soon as the ADC is set, use analogReadContinuous() to read the values\n * Set the resolution, number of averages and voltage reference using the appropriate functions BEFORE calling this function\n*/\nbool ADC_Module::startContinuous(uint8_t pin) {\n\n // check whether the pin is correct\n if(!checkPin(pin)) {\n fail_flag |= ADC_ERROR_WRONG_PIN;\n return false;\n }\n\n // check for calibration before setting channels,\n if (calibrating) wait_for_cal();\n\n // increase the counter of measurements\n num_measurements++;\n\n // set continuous conversion flag\n continuousMode();\n\n startReadFast(pin);\n\n return true;\n}\n\n\n/* Starts continuous and differential conversion between the pins (pinP-pinN)\n * It returns as soon as the ADC is set, use analogReadContinuous() to read the value\n * Set the resolution, number of averages and voltage reference using the appropriate functions BEFORE calling this function\n*/\nbool ADC_Module::startContinuousDifferential(uint8_t pinP, uint8_t pinN) {\n\n if(!checkDifferentialPins(pinP, pinN)) {\n fail_flag |= ADC_ERROR_WRONG_PIN;\n return false; // all others are invalid\n }\n\n // increase the counter of measurements\n num_measurements++;\n\n // check for calibration before setting channels,\n // because conversion will start as soon as we write to *ADC_SC1A\n if (calibrating) wait_for_cal();\n\n // save the current state of the ADC in case it's in use\n uint8_t wasADCInUse = isConverting(); // is the ADC running now?",
"\n\n if(wasADCInUse) { // this means we're interrupting a conversion\n // save the current conversion config, we don't want any other interrupts messing up the configs\n __disable_irq();\n saveConfig(&adc_config);\n __enable_irq();\n }\n\n // set continuous mode\n continuousMode();\n\n // start conversions\n startDifferentialFast(pinP, pinN);\n\n return true;\n}\n\n\n/* Stops continuous conversion\n*/\nvoid ADC_Module::stopContinuous() {\n\n // set channel select to all 1's (31) to stop it.",
"\n *ADC_SC1A = 0x1F + var_enableInterrupts*ADC_SC1_AIEN;;\n\n // decrease the counter of measurements (unless it's 0)\n if(!num_measurements) {\n num_measurements--;\n }\n\n\n return;\n}\n\n//////////// PDB ////////////////\n//// Only works for Teensy 3.0 and 3.1, not LC (it doesn't have PDB)\n\n#if ADC_USE_PDB\n\n// frequency in Hz\nvoid ADC_Module::startPDB(uint32_t freq) {\n\n if (!(",
"SIM_SCGC6 & SIM_SCGC6_PDB)) { // setup PDB\n SIM_SCGC6 |= SIM_SCGC6_PDB; // enable pdb clock\n }\n\n if(freq>F_BUS) return; // too high\n if(freq<1) return; // too low\n\n // mod will have to be a 16 bit value\n // we detect if it's higher than 0xFFFF and scale it back acordingly.",
"\n uint32_t mod = (F_BUS / freq);\n\n uint8_t prescaler = 0; // from 0 to 7: factor of 1, 2, 4, 8, 16, 32, 64 or 128\n uint8_t mult = 0; // from 0 to 3, factor of 1, 10, 20 or 40\n\n // if mod is too high we need to use prescaler and mult to bring it down to a 16 bit number\n const uint32_t min_level = 0xFFFF;\n if(mod>min_level) {\n if( mod < 2*min_level ) {\n prescaler = 1;\n }\n else if( mod < 4*min_level ) {\n prescaler = 2;\n }\n else if( mod < 8*min_level ) {\n prescaler = 3;\n }\n else if( mod < 10*min_level ) {\n mult = 1;\n }\n else if( mod < 16*min_level ) {\n prescaler = 4;\n }\n else if( mod < 20*min_level ) {\n mult = 2;\n }\n else if( mod < 32*min_level ) {\n prescaler = 5;\n }\n else if( mod < 40*min_level ) {\n mult = 3;\n }\n else if( mod < 64*min_level ) {\n prescaler = 6;\n }\n else if( mod < 128*min_level ) {\n prescaler = 7;\n }\n else if( mod < 160*min_level ) { // 16*10\n prescaler = 4;\n mult = 1;\n }\n else if( mod < 320*min_level ) { // 16*20\n prescaler = 4;\n mult = 2;\n }\n else if( mod < 640*min_level ) { // 16*40\n prescaler = 4;\n mult = 3;\n }\n else if( mod < 1280*min_level ) { // 32*40\n prescaler = 5;\n mult = 3;\n }\n else if( mod < 2560*min_level ) { // 64*40\n prescaler = 6;\n mult = 3;\n }\n else if( mod < 5120*min_level ) { // 128*40\n prescaler = 7;\n mult = 3;\n }\n else { // frequency too low\n return;\n }\n\n mod >>= prescaler;\n if(mult>0) {\n mod /= 10;\n mod >>= (mult-1);\n }\n }\n\n setHardwareTrigger(); // trigger ADC with hardware\n\n PDB0_IDLY = 1; // the pdb interrupt happens when IDLY is equal to CNT+1\n\n PDB0_MOD = (uint16_t)(mod-1);\n\n PDB0_SC = PDB_CONFIG | PDB_SC_PRESCALER(prescaler) | PDB_SC_MULT(mult) | PDB_SC_LDOK; // load all new values\n\n PDB0_SC = PDB_CONFIG | PDB_SC_PRESCALER(prescaler) | PDB_SC_MULT(mult) | PDB_SC_SWTRIG; // start the counter!",
"\n\n *PDB0_CHnC1 = PDB_CHnC1_TOS_1 | PDB_CHnC1_EN_1; // enable pretrigger 0 (SC1A)\n\n NVIC_ENABLE_IRQ(IRQ_PDB);\n\n}\n\nvoid ADC_Module::stopPDB() {\n if (!(",
"SIM_SCGC6 & SIM_SCGC6_PDB)) { // if PDB clock wasn't on, return\n setSoftwareTrigger();\n return;\n }\n PDB0_SC = 0;\n setSoftwareTrigger();\n\n NVIC_DISABLE_IRQ(IRQ_PDB);\n}\n\n#endif\n"
] |
{
"pile_set_name": "Github"
}
|
[
0.0006722227553837001,
0.0007172609912231565,
0.0007966035627759993,
0.02446136809885502,
0.0032598611433058977,
0.002619669772684574,
0.0007301521836780012,
0.00924395490437746,
0.0011216234415769577,
0.0006013843812979758,
0.0019478988833725452,
0.0013359587173908949,
0.0007452451973222196,
0.0006030348595231771,
0.0007362885517068207,
0.001345984754152596,
0.0025484166108071804,
0.0007733830716460943,
0.0008469251333735883,
0.0006956399884074926,
0.0006949459202587605,
0.00070658401818946,
0.0007907196413725615,
0.0020360779017210007,
0.004719589836895466,
0.0017778617329895496,
0.0016989429714158177,
0.002783098490908742,
0.0031169310677796602,
0.005970133934170008,
0.003505395259708166,
0.001209772890433669,
0.0008969029877334833,
0.002588775707408786,
0.0012229379499331117,
0.0013380927266553044,
0.0006933450349606574,
0.0007634745561517775,
0.0016120272921398282,
0.0013412071857601404,
0.001311760162934661,
0.0012109476374462247,
0.0008385832770727575,
0.0015007620677351952,
0.14740034937858582,
0.11904938519001007,
0.04718223959207535,
0.0007695191306993365,
0.0007907196413725615,
0.0046007982455194,
0.0007803779444657266,
0.012002957053482533,
0.0013717231340706348,
0.0013119002105668187,
0.0009425644529983401,
0.0005967524484731257,
0.0010617478983476758,
0.005445239599794149,
0.0006147988606244326,
0.0014014235930517316,
0.004541073460131884,
0.0006526242941617966,
0.003935173619538546,
0.003126673400402069,
0.002421889454126358,
0.004204032011330128,
0.0016459375619888306,
0.004461156670004129,
0.001379584544338286,
0.0009186670067720115,
0.0006580605986528099,
0.0012900278670713305,
0.0009138084133155644,
0.000684433791320771,
0.001202064217068255,
0.014012708328664303,
0.0013101339573040605,
0.0007336483104154468,
0.000702662393450737,
0.0007117530913092196,
0.0007272973307408392,
0.0006133862771093845,
0.0011123190633952618,
0.0007553923060186207,
0.0005903876735828817,
0.0019606275018304586,
0.0011479186359792948,
0.001217796467244625,
0.0011357669718563557,
0.0010200338438153267,
0.0008892064215615392,
0.0011379116913303733,
0.001379584544338286,
0.0009186670067720115,
0.000555577571503818,
0.0012900278670713305,
0.0008030693861655891,
0.0011275105644017458,
0.000584679190069437,
0.0012752260081470013,
0.0030473480001091957,
0.0015534821432083845,
0.0006056089187040925,
0.001006223144941032,
0.0009456915431655943,
0.0011804085224866867,
0.001379584544338286,
0.0009258735226467252,
0.0016056301537901163,
0.001228103879839182,
0.0008675704593770206,
0.0014718440361320972,
0.010609414428472519,
0.002410082146525383,
0.0016218548407778144
] | 0.004745 | 115 |
[
"I’m not sure how I feel about the SKA Mystery Sock. ",
"I think it’s the yarn that’s making me re-consider it. ",
"The yarn is a bit on the bright, eye searing side. ",
"I’m using Yarntini merino/nylon sock in Strawberry Frenzy, which was part of the Semi-Solid Sock Club from last year. ",
"It’s not very “semi” to me, but a very solid bright red. ",
"It’s even just a bit brighter then the picture shows. ",
"Definitely going to finish them. ",
"They may end up as a gift, though.",
"\n\nSKA November Mystery Sock KAL\n\nThe Barking Dog Yarns Opposites Attract Mystery KAL is going well. ",
"The clues are coming out faster, so I started on clue 2 on Saturday. ",
"The yarn is a fun concept. ",
"Two skeins that go together, but are different. ",
"You’ll see what I mean when I start the second sock. ",
"I took pictures just after finishing clue 1. ",
"Definitely want to get more of this yarn!",
"\n\nBDY Opposites Attract Mystery KAL\n\nBDY Opposites Attract Mystery KAL\n\nI was a little hesitant to join the Woolly Wormhead Mystery Hat KAL since I don’t wear a lot of hats. ",
"I looked through my stash and found that I had yarn for both hats on hand, so I figured I’d give it a go. ",
"If I don’t care for the hats, I’m sure someone I know would love and wear them. ",
"For Pattern A, I’m using Wollmeise 80/20 Twin in Lowenzahn. ",
"For Pattern B, I’m using Elsebeth Lavold Silky Wool in a lovely orange color (color 81). ",
"I finished clue 1 for Pattern B the same day the clue came out. ",
"Not the case for Pattern A, though! ",
"I was working Pattern A on US 1.5, 40″ circular and something about it was hurting my hands. ",
"Probably the twisted stitches, which I’ve found is a different motion then knits or purls. ",
"That combined with fiddling with the big cable likely wasn’t helping much. ",
"I did do the brim shorter then the clue said to for a few reasons. ",
"It was difficult to work on without pain in my hands and wrists after a short time, I’m not a fan of 3″ brims on hats (2″ is about my maximum), and I am just a tad worried I might not have enough yarn (I’m using about half of a leftover skein of Wollmeise). ",
"The picture below doesn’t show it, but I am finished with clue 1 on Pattern A and the brim is 2″.\n\nPattern A\n\nPattern B\n\nDo you do mystery knit-a-longs? ",
"Why/why not? ",
"I’m always curious what other people think about knit-a-longs where you don’t know what the pattern will be like other then what the finished item itself will be (ie a sock or a hat).",
"\n\nShare this:\n\nLike this:\n\nI’m a thirty-something who lives in Cincinnati, OH.",
"\nCrafting, for me, started young - I can’t even recall a time I wasn’t crafty in some way. ",
"My first foray into yarn was doing plastic canvas embroidery and making Barbie doll furniture. ",
"Not long after that I learned to crochet from a book my grandmother gave me after asking to learn what she was doing (she crochets, afghans mostly these days). ",
"After that, I took up counted cross stitch and then came sewing when I took two fashion design classes in high school. ",
"My mom had a sewing machine from my great grandmother and I taught myself to sew from patterns on it. ",
"I still use this same sewing machine today!",
"\nI started knitting in 2004 when a co-worker was pregnant with her first child and I wanted to make something for her. ",
"I bought a “learn to knit kit” from Lion Brand for a baby hat and booties set. ",
"I did make the baby hat and booties (with very few problems) and ended up gifting them to the co-worker. ",
"I wanted to make a blanket, but I wasn’t quite able to get that done in time since it took a lot longer then I’d imagined! ",
"It wasn’t until the end of 2004/early 2005 that knitting really took a hold of me. ",
"Now I love knitting and almost always have a project with me, even if it’s just something small when I’m out and about. ",
"I like a wide range of projects depending on my mood. ",
"Sometimes I like a good challenge knit and other times I enjoy a lot of stockinette or garter stitch. ",
"I love Ravelry for keeping track of my projects and for finding new things to knit - I feel like I’m always spending time on there! ",
"It’s also a great place for getting to know other knitters, crocheters, and various fiber enthusiasts.",
"\nIn the fall of 2008, I purchased a spinning wheel from someone on Craigslist. ",
"It’s a Babe Fiber Starter, single treadle wheel and after a bit of trouble getting used to it, I’ve managed to spin up some fiber. ",
"In January of 2012, I found someone local that was selling their Lendrum DT and it came to live with me! ",
"The Babe now lives with Jen (aka piddleloop) and she’s learning to spin. ",
"I’m still trying to find time to spin along with knitting, crocheting (occasionally), and whatever else is going on, but I work it in here and there.",
"\nI’m sure there’s something I might have left out here, so if there’s anything else you want to know about me, just ask :o)\nView All Posts\n\nOne thought on “Mystery KAL’s Aplenty This Month”\n\nI always debate on whether or not to do a myster KAL…seems like so much knitting for something I may not like…now I can get into the hats because they are small and don’t generally take much time…I love the look of hats but then I rarely wear them."
] |
{
"pile_set_name": "Pile-CC"
}
|
[
0.0012503814650699496,
0.0007311778608709574,
0.0006833411753177643,
0.000945393810980022,
0.00093259965069592,
0.0006318800151348114,
0.004856709856539965,
0.0006190732237882912,
0.0026506998110562563,
0.0006045203772373497,
0.0006264902767725289,
0.0015209545381367207,
0.0032983580604195595,
0.0006360699771903455,
0.001267433282919228,
0.009352832101285458,
0.0007863077917136252,
0.004420380108058453,
0.0006279991357587278,
0.0006204054807312787,
0.0005887392326258123,
0.0008625711780041456,
0.0010027417447417974,
0.0013493287842720747,
0.0019093977753072977,
0.0007895379094406962,
0.001384912640787661,
0.0006775350775569677,
0.0007860747282393277,
0.0006869716453365982,
0.0006857533589936793,
0.0006581948255188763,
0.0007261953433044255,
0.0016696277307346463,
0.0005863432888872921,
0.0011375920148566365,
0.0007124123512767255,
0.0006022651214152575,
0.0019182085525244474,
0.0013130439911037683,
0.0010077342158183455,
0.0007342099561356008,
0.0006288337172009051,
0.0006167387473396957,
0.0007181388791650534,
0.0006207827827893198,
0.0005869955639354885,
0.0007246979512274265,
0.0006743501871824265,
0.0007407875382341444,
0.012232191860675812,
0.0007402882911264896,
0.0007177413790486753
] | 0.001469 | 53 |
[
"// Copyright John Maddock 2008.",
"\n// Use, modification and distribution are subject to the\n// Boost Software License, Version 1.0. (",
"See accompanying file\n// LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt)\n\n#ifndef BOOST_MATH_DISTIBUTIONS_DETAIL_GENERIC_QUANTILE_HPP\n#define BOOST_MATH_DISTIBUTIONS_DETAIL_GENERIC_QUANTILE_HPP\n\nnamespace boost{ namespace math{ namespace detail{\n\ntemplate <class Dist>\nstruct generic_quantile_finder\n{\n typedef typename Dist::value_type value_type;\n typedef typename Dist::policy_type policy_type;\n\n generic_quantile_finder(const Dist& d, value_type t, bool c)\n : dist(d), target(t), comp(c) {}\n\n value_type operator()(const value_type& x)\n {\n return comp ?",
"\n value_type(target - cdf(complement(dist, x)))\n : value_type(cdf(dist, x) - target);\n }\n\nprivate:\n Dist dist;\n value_type target;\n bool comp;\n};\n\ntemplate <class T, class Policy>\ninline T check_range_result(const T& x, const Policy& pol, const char* function)\n{\n if((x >= 0) && (x < tools::min_value<T>()))\n return policies::raise_underflow_error<T>(function, 0, pol);\n if(x <= -tools::max_value<T>())\n return -policies::raise_overflow_error<T>(function, 0, pol);\n if(x >= tools::max_value<T>())\n return policies::raise_overflow_error<T>(function, 0, pol);\n return x;\n}\n\ntemplate <class Dist>\ntypename Dist::value_type generic_quantile(const Dist& dist, const typename Dist::value_type& p, const typename Dist::value_type& guess, bool comp, const char* function)\n{\n typedef typename Dist::value_type value_type;\n typedef typename Dist::policy_type policy_type;\n typedef typename policies::normalise<\n policy_type, \n policies::promote_float<false>, \n policies::promote_double<false>, \n policies::discrete_quantile<>,\n policies::assert_undefined<> >::type forwarding_policy;\n\n //\n // Special cases first:\n //\n if(p == 0)\n {\n return comp\n ? ",
"check_range_result(range(dist).second, forwarding_policy(), function)\n : check_range_result(range(dist).first, forwarding_policy(), function);\n }\n if(p == 1)\n {\n return !",
"comp\n ? ",
"check_range_result(range(dist).second, forwarding_policy(), function)\n : check_range_result(range(dist).first, forwarding_policy(), function);\n }\n\n generic_quantile_finder<Dist> f(dist, p, comp);\n tools::eps_tolerance<value_type> tol(policies::digits<value_type, forwarding_policy>() - 3);\n boost::uintmax_t max_iter = policies::get_max_root_iterations<forwarding_policy>();\n std::pair<value_type, value_type> ir = tools::bracket_and_solve_root(\n f, guess, value_type(2), true, tol, max_iter, forwarding_policy());\n value_type result = ir.first + (ir.second - ir.first) / 2;\n if(max_iter >= policies::get_max_root_iterations<forwarding_policy>())\n {\n return policies::raise_evaluation_error<value_type>(function, \"Unable to locate solution in a reasonable time:\"\n \" either there is no answer to quantile\"\n \" or the answer is infinite. ",
" Current best guess is %1%\", result, forwarding_policy());\n }\n return result;\n}\n\n}}} // namespaces\n\n#endif // BOOST_MATH_DISTIBUTIONS_DETAIL_GENERIC_QUANTILE_HPP\n\n"
] |
{
"pile_set_name": "Github"
}
|
[
0.000821832800284028,
0.0005921193514950573,
0.000751947402022779,
0.0011086188023909926,
0.0008813193417154253,
0.000833220430649817,
0.0008797751506790519,
0.0008408985449932516
] | 0.000839 | 8 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.