title
stringlengths
12
112
published
stringlengths
19
23
url
stringlengths
28
28
video_id
stringlengths
11
11
channel_id
stringclasses
5 values
id
stringlengths
16
31
text
stringlengths
0
596
start
float64
0
37.8k
end
float64
2.18
37.8k
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t664.0
the maximum value
664
668
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t666.0
we want to select is 3 in this case.
666
670
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t668.0
So it's going to be the number
668
672
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t670.0
of sentences
670
674
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t672.0
minus 2.
672
676
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t674.0
Ok.
674
678
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t676.0
Now
676
680
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t678.0
what we do is we do our 50-50
678
682
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t680.0
randomize or not randomize
680
684
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t682.0
sentence B.
682
686
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t684.0
So if random.
684
688
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t686.0
random so this will just
686
690
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t688.0
select random float between values
688
692
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t690.0
0 and 1.
690
694
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t692.0
If that
692
696
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t694.0
is greater than 0.5
694
698
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t696.0
then let's say we'll
696
700
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t698.0
make this our random selection.
698
702
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t700.0
Ok. So for the random
700
704
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t702.0
selection
702
706
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t704.0
what we do is sentence B
704
708
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t706.0
.append
706
710
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t708.0
and then here we would
708
712
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t710.0
append a random sentence
710
714
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t712.0
from our bag up here.
712
716
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t714.0
So to
714
718
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t716.0
do that we would just write bag
716
720
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t718.0
and then in here we need to
718
722
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t720.0
select a random integer
720
724
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t722.0
like we did up here. Ok. So
722
726
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t724.0
we're going to use that same
724
728
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t726.0
function. So random.
726
730
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t728.0
randint
728
732
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t730.0
and that needs to be between 0
730
734
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t732.0
and the length of
732
736
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t734.0
our bag minus 1.
734
738
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t736.0
So we use bag
736
740
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t738.0
size that's why we
738
742
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t740.0
have it.
740
744
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t742.0
So bag size minus 1.
742
746
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t744.0
Ok. Now we'll select a random
744
748
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t746.0
sentence B from that bag for
746
750
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t748.0
us. And
748
752
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t750.0
as well as that we
750
754
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t752.0
also want to set a label.
752
756
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t754.0
So our label in this case
754
758
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t756.0
would be
756
760
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t758.0
a 1. So
758
762
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t760.0
we have the 0 which means it is the next sentence.
760
764
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t762.0
We have a 1 which means it is
762
766
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t764.0
not the next sentence. So we
764
768
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t766.0
set 1.
766
770
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t768.0
Now our sentence A
768
772
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t770.0
it our sentence A gets selected.
770
774
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t772.0
It's the same thing no
772
776
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t774.0
matter whether we have
774
778
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t776.0
the random sentence
776
780
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t778.0
B or the not random sentence B.
778
782
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t780.0
So we can actually write
780
784
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t782.0
our sentence A append
782
786
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t784.0
up here.
784
788
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t786.0
And this is just
786
790
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t788.0
going to be sentences.
788
792
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t790.0
And in the
790
794
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t792.0
index we have start which is
792
796
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t794.0
our value from here.
794
798
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t796.0
Ok. So we
796
800
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t798.0
have the random option.
798
802
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t800.0
Now let's
800
804
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t802.0
do our not random option.
802
806
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t804.0
So in here we
804
808
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t806.0
write sentence B append.
806
810
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t808.0
And
808
812
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t810.0
this needs to append sentences
810
816
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t814.0
start plus
814
818
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t816.0
1. So the following sentence
816
820
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t818.0
after our sentence A.
818
824
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t822.0
And our label here
822
826
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t824.0
would be 0
824
828
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t826.0
which means it is the next sentence.
826
830
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t828.0
So let's
828
832
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t830.0
there's quite a lot of code. Let's run
830
834
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t832.0
that and see what we get.
832
836
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t834.0
Ok.
834
838
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t836.0
Now
836
840
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t838.0
what I want to do
838
842
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t840.0
is let's have a look
840
844
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t842.0
at the first
842
846
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t844.0
few labels. See if we have a
844
848
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t846.0
mix of different ones in there.
846
850
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t848.0
Ok. We just have 1, 1, 1. So I'm going to
848
852
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t850.0
rerun this because I want to show you the
850
854
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t852.0
difference between 0's and 1's here.
852
856
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t854.0
Ok. So we
854
858
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t856.0
have these. So let me
856
860
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t858.0
print out
858
862
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t860.0
what we have.
860
864
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t862.0
So
862
866
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t864.0
for i in
864
868
Training BERT #4 - Train With Next Sentence Prediction (NSP)
2021-05-27 16:15:39 UTC
https://youtu.be/x1lAcT3xl5M
x1lAcT3xl5M
UCv83tO5cePwHMt1952IVVHw
x1lAcT3xl5M-t866.0
range 3. So I'm just
866
870