modelId
stringlengths 5
139
| author
stringlengths 2
42
| last_modified
timestamp[us, tz=UTC]date 2020-02-15 11:33:14
2025-08-27 18:28:06
| downloads
int64 0
223M
| likes
int64 0
11.7k
| library_name
stringclasses 523
values | tags
listlengths 1
4.05k
| pipeline_tag
stringclasses 55
values | createdAt
timestamp[us, tz=UTC]date 2022-03-02 23:29:04
2025-08-27 18:27:40
| card
stringlengths 11
1.01M
|
|---|---|---|---|---|---|---|---|---|---|
ibm-research/ColD-Fusion-itr15-seed4
|
ibm-research
| 2022-12-06T09:50:09Z | 5 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:49:36Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr15-seed2
|
ibm-research
| 2022-12-06T09:49:05Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:48:30Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr15-seed1
|
ibm-research
| 2022-12-06T09:48:27Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:47:50Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr14-seed4
|
ibm-research
| 2022-12-06T09:47:22Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:46:54Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr14-seed3
|
ibm-research
| 2022-12-06T09:46:52Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:46:33Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr14-seed2
|
ibm-research
| 2022-12-06T09:46:30Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:46:18Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr13-seed4
|
ibm-research
| 2022-12-06T09:45:57Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:45:46Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr13-seed3
|
ibm-research
| 2022-12-06T09:45:30Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:45:19Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr12-seed0
|
ibm-research
| 2022-12-06T09:44:37Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:44:27Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr12-seed1
|
ibm-research
| 2022-12-06T09:43:59Z | 5 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:43:49Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr11-seed4
|
ibm-research
| 2022-12-06T09:43:46Z | 6 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:43:36Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr11-seed3
|
ibm-research
| 2022-12-06T09:43:19Z | 5 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:43:08Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr11-seed1
|
ibm-research
| 2022-12-06T09:42:54Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:42:42Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr10-seed4
|
ibm-research
| 2022-12-06T09:42:40Z | 5 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:42:31Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr10-seed3
|
ibm-research
| 2022-12-06T09:42:28Z | 5 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:42:19Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr10-seed0
|
ibm-research
| 2022-12-06T09:42:17Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:42:07Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr10-seed1
|
ibm-research
| 2022-12-06T09:41:51Z | 5 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:41:41Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr9-seed0
|
ibm-research
| 2022-12-06T09:41:08Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:40:57Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
ibm-research/ColD-Fusion-itr9-seed2
|
ibm-research
| 2022-12-06T09:40:52Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"exbert",
"en",
"arxiv:2212.01378",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T09:40:41Z |
---
language: en
tags:
- exbert
license: mit
---
# ColD Fusion model
Finetuned model that aims to be a great base model. It improves over RoBERTa base, trained on 35 datasets.
Full details at [this paper](https://arxiv.org/abs/2212.01378).
## Paper Abstract:
Pretraining has been shown to scale well with compute, data size and data diversity. Multitask learning trains on a
mixture of supervised datasets and produces improved performance compared to self-supervised pretraining. Until now,
massively multitask learning required simultaneous access to all datasets in the mixture and heavy compute resources
that are only available to well-resourced teams.
In this paper, we propose ColD Fusion, a method that provides the benefits of multitask learning but leverages distributed
computation and requires limited communication and no sharing of data. Consequentially, ColD Fusion can create a synergistic
loop, where finetuned models can be recycled to continually improve the pretrained model they are based on. We show that
ColD Fusion yields comparable benefits to multitask pretraining by producing a model that (a) attains strong performance on
all of the datasets it was multitask trained on and (b) is a better starting point for finetuning on unseen datasets. We find
ColD Fusion outperforms RoBERTa and even previous multitask models. Specifically, when training and testing on 35 diverse datasets,
ColD Fusion-based model outperforms RoBERTa by 2.45 points in average without any changes to the architecture.
### How to use
Best way to use is to finetune on your own task, but you can also extract features directly.
To get the features of a given text in PyTorch:
```python
from transformers import RobertaTokenizer, RobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = RobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
```
and in TensorFlow:
```python
from transformers import RobertaTokenizer, TFRobertaModel
tokenizer = RobertaTokenizer.from_pretrained('ibm/ColD-Fusion')
model = TFRobertaModel.from_pretrained('ibm/ColD-Fusion')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='tf')
output = model(encoded_input)
```
## Evaluation results
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
When fine-tuned on downstream tasks, this model achieves the following results:
### BibTeX entry and citation info
```bibtex
@article{ColDFusion,
author = {Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, Leshem ChoshenYinhan Liu and},
title = {ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning},
journal = {CoRR},
volume = {abs/2212.01378},
year = {2022},
url = {https://arxiv.org/abs/2212.01378},
archivePrefix = {arXiv},
eprint = {2212.01378},
}
```
<a href="https://huggingface.co/exbert/?model=ibm/ColD-Fusion">
<img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png">
</a>
|
tomekkorbak/musing_hoover
|
tomekkorbak
| 2022-12-06T09:35:33Z | 1 | 0 |
transformers
|
[
"transformers",
"pytorch",
"gpt2",
"generated_from_trainer",
"en",
"dataset:tomekkorbak/detoxify-pile-chunk3-0-50000",
"dataset:tomekkorbak/detoxify-pile-chunk3-50000-100000",
"dataset:tomekkorbak/detoxify-pile-chunk3-100000-150000",
"dataset:tomekkorbak/detoxify-pile-chunk3-150000-200000",
"dataset:tomekkorbak/detoxify-pile-chunk3-200000-250000",
"dataset:tomekkorbak/detoxify-pile-chunk3-250000-300000",
"dataset:tomekkorbak/detoxify-pile-chunk3-300000-350000",
"dataset:tomekkorbak/detoxify-pile-chunk3-350000-400000",
"dataset:tomekkorbak/detoxify-pile-chunk3-400000-450000",
"dataset:tomekkorbak/detoxify-pile-chunk3-450000-500000",
"dataset:tomekkorbak/detoxify-pile-chunk3-500000-550000",
"dataset:tomekkorbak/detoxify-pile-chunk3-550000-600000",
"dataset:tomekkorbak/detoxify-pile-chunk3-600000-650000",
"dataset:tomekkorbak/detoxify-pile-chunk3-650000-700000",
"dataset:tomekkorbak/detoxify-pile-chunk3-700000-750000",
"dataset:tomekkorbak/detoxify-pile-chunk3-750000-800000",
"dataset:tomekkorbak/detoxify-pile-chunk3-800000-850000",
"dataset:tomekkorbak/detoxify-pile-chunk3-850000-900000",
"dataset:tomekkorbak/detoxify-pile-chunk3-900000-950000",
"dataset:tomekkorbak/detoxify-pile-chunk3-950000-1000000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1000000-1050000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1050000-1100000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1100000-1150000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1150000-1200000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1200000-1250000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1250000-1300000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1300000-1350000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1350000-1400000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1400000-1450000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1450000-1500000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1500000-1550000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1550000-1600000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1600000-1650000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1650000-1700000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1700000-1750000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1750000-1800000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1800000-1850000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1850000-1900000",
"dataset:tomekkorbak/detoxify-pile-chunk3-1900000-1950000",
"license:mit",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] | null | 2022-12-05T16:55:01Z |
---
language:
- en
license: mit
tags:
- generated_from_trainer
datasets:
- tomekkorbak/detoxify-pile-chunk3-0-50000
- tomekkorbak/detoxify-pile-chunk3-50000-100000
- tomekkorbak/detoxify-pile-chunk3-100000-150000
- tomekkorbak/detoxify-pile-chunk3-150000-200000
- tomekkorbak/detoxify-pile-chunk3-200000-250000
- tomekkorbak/detoxify-pile-chunk3-250000-300000
- tomekkorbak/detoxify-pile-chunk3-300000-350000
- tomekkorbak/detoxify-pile-chunk3-350000-400000
- tomekkorbak/detoxify-pile-chunk3-400000-450000
- tomekkorbak/detoxify-pile-chunk3-450000-500000
- tomekkorbak/detoxify-pile-chunk3-500000-550000
- tomekkorbak/detoxify-pile-chunk3-550000-600000
- tomekkorbak/detoxify-pile-chunk3-600000-650000
- tomekkorbak/detoxify-pile-chunk3-650000-700000
- tomekkorbak/detoxify-pile-chunk3-700000-750000
- tomekkorbak/detoxify-pile-chunk3-750000-800000
- tomekkorbak/detoxify-pile-chunk3-800000-850000
- tomekkorbak/detoxify-pile-chunk3-850000-900000
- tomekkorbak/detoxify-pile-chunk3-900000-950000
- tomekkorbak/detoxify-pile-chunk3-950000-1000000
- tomekkorbak/detoxify-pile-chunk3-1000000-1050000
- tomekkorbak/detoxify-pile-chunk3-1050000-1100000
- tomekkorbak/detoxify-pile-chunk3-1100000-1150000
- tomekkorbak/detoxify-pile-chunk3-1150000-1200000
- tomekkorbak/detoxify-pile-chunk3-1200000-1250000
- tomekkorbak/detoxify-pile-chunk3-1250000-1300000
- tomekkorbak/detoxify-pile-chunk3-1300000-1350000
- tomekkorbak/detoxify-pile-chunk3-1350000-1400000
- tomekkorbak/detoxify-pile-chunk3-1400000-1450000
- tomekkorbak/detoxify-pile-chunk3-1450000-1500000
- tomekkorbak/detoxify-pile-chunk3-1500000-1550000
- tomekkorbak/detoxify-pile-chunk3-1550000-1600000
- tomekkorbak/detoxify-pile-chunk3-1600000-1650000
- tomekkorbak/detoxify-pile-chunk3-1650000-1700000
- tomekkorbak/detoxify-pile-chunk3-1700000-1750000
- tomekkorbak/detoxify-pile-chunk3-1750000-1800000
- tomekkorbak/detoxify-pile-chunk3-1800000-1850000
- tomekkorbak/detoxify-pile-chunk3-1850000-1900000
- tomekkorbak/detoxify-pile-chunk3-1900000-1950000
model-index:
- name: musing_hoover
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# musing_hoover
This model was trained from scratch on the tomekkorbak/detoxify-pile-chunk3-0-50000, the tomekkorbak/detoxify-pile-chunk3-50000-100000, the tomekkorbak/detoxify-pile-chunk3-100000-150000, the tomekkorbak/detoxify-pile-chunk3-150000-200000, the tomekkorbak/detoxify-pile-chunk3-200000-250000, the tomekkorbak/detoxify-pile-chunk3-250000-300000, the tomekkorbak/detoxify-pile-chunk3-300000-350000, the tomekkorbak/detoxify-pile-chunk3-350000-400000, the tomekkorbak/detoxify-pile-chunk3-400000-450000, the tomekkorbak/detoxify-pile-chunk3-450000-500000, the tomekkorbak/detoxify-pile-chunk3-500000-550000, the tomekkorbak/detoxify-pile-chunk3-550000-600000, the tomekkorbak/detoxify-pile-chunk3-600000-650000, the tomekkorbak/detoxify-pile-chunk3-650000-700000, the tomekkorbak/detoxify-pile-chunk3-700000-750000, the tomekkorbak/detoxify-pile-chunk3-750000-800000, the tomekkorbak/detoxify-pile-chunk3-800000-850000, the tomekkorbak/detoxify-pile-chunk3-850000-900000, the tomekkorbak/detoxify-pile-chunk3-900000-950000, the tomekkorbak/detoxify-pile-chunk3-950000-1000000, the tomekkorbak/detoxify-pile-chunk3-1000000-1050000, the tomekkorbak/detoxify-pile-chunk3-1050000-1100000, the tomekkorbak/detoxify-pile-chunk3-1100000-1150000, the tomekkorbak/detoxify-pile-chunk3-1150000-1200000, the tomekkorbak/detoxify-pile-chunk3-1200000-1250000, the tomekkorbak/detoxify-pile-chunk3-1250000-1300000, the tomekkorbak/detoxify-pile-chunk3-1300000-1350000, the tomekkorbak/detoxify-pile-chunk3-1350000-1400000, the tomekkorbak/detoxify-pile-chunk3-1400000-1450000, the tomekkorbak/detoxify-pile-chunk3-1450000-1500000, the tomekkorbak/detoxify-pile-chunk3-1500000-1550000, the tomekkorbak/detoxify-pile-chunk3-1550000-1600000, the tomekkorbak/detoxify-pile-chunk3-1600000-1650000, the tomekkorbak/detoxify-pile-chunk3-1650000-1700000, the tomekkorbak/detoxify-pile-chunk3-1700000-1750000, the tomekkorbak/detoxify-pile-chunk3-1750000-1800000, the tomekkorbak/detoxify-pile-chunk3-1800000-1850000, the tomekkorbak/detoxify-pile-chunk3-1850000-1900000 and the tomekkorbak/detoxify-pile-chunk3-1900000-1950000 datasets.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0005
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 64
- total_train_batch_size: 1024
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.01
- training_steps: 3147
- mixed_precision_training: Native AMP
### Framework versions
- Transformers 4.24.0
- Pytorch 1.11.0+cu113
- Datasets 2.5.1
- Tokenizers 0.11.6
# Full config
{'dataset': {'datasets': ['tomekkorbak/detoxify-pile-chunk3-0-50000',
'tomekkorbak/detoxify-pile-chunk3-50000-100000',
'tomekkorbak/detoxify-pile-chunk3-100000-150000',
'tomekkorbak/detoxify-pile-chunk3-150000-200000',
'tomekkorbak/detoxify-pile-chunk3-200000-250000',
'tomekkorbak/detoxify-pile-chunk3-250000-300000',
'tomekkorbak/detoxify-pile-chunk3-300000-350000',
'tomekkorbak/detoxify-pile-chunk3-350000-400000',
'tomekkorbak/detoxify-pile-chunk3-400000-450000',
'tomekkorbak/detoxify-pile-chunk3-450000-500000',
'tomekkorbak/detoxify-pile-chunk3-500000-550000',
'tomekkorbak/detoxify-pile-chunk3-550000-600000',
'tomekkorbak/detoxify-pile-chunk3-600000-650000',
'tomekkorbak/detoxify-pile-chunk3-650000-700000',
'tomekkorbak/detoxify-pile-chunk3-700000-750000',
'tomekkorbak/detoxify-pile-chunk3-750000-800000',
'tomekkorbak/detoxify-pile-chunk3-800000-850000',
'tomekkorbak/detoxify-pile-chunk3-850000-900000',
'tomekkorbak/detoxify-pile-chunk3-900000-950000',
'tomekkorbak/detoxify-pile-chunk3-950000-1000000',
'tomekkorbak/detoxify-pile-chunk3-1000000-1050000',
'tomekkorbak/detoxify-pile-chunk3-1050000-1100000',
'tomekkorbak/detoxify-pile-chunk3-1100000-1150000',
'tomekkorbak/detoxify-pile-chunk3-1150000-1200000',
'tomekkorbak/detoxify-pile-chunk3-1200000-1250000',
'tomekkorbak/detoxify-pile-chunk3-1250000-1300000',
'tomekkorbak/detoxify-pile-chunk3-1300000-1350000',
'tomekkorbak/detoxify-pile-chunk3-1350000-1400000',
'tomekkorbak/detoxify-pile-chunk3-1400000-1450000',
'tomekkorbak/detoxify-pile-chunk3-1450000-1500000',
'tomekkorbak/detoxify-pile-chunk3-1500000-1550000',
'tomekkorbak/detoxify-pile-chunk3-1550000-1600000',
'tomekkorbak/detoxify-pile-chunk3-1600000-1650000',
'tomekkorbak/detoxify-pile-chunk3-1650000-1700000',
'tomekkorbak/detoxify-pile-chunk3-1700000-1750000',
'tomekkorbak/detoxify-pile-chunk3-1750000-1800000',
'tomekkorbak/detoxify-pile-chunk3-1800000-1850000',
'tomekkorbak/detoxify-pile-chunk3-1850000-1900000',
'tomekkorbak/detoxify-pile-chunk3-1900000-1950000'],
'is_split_by_sentences': True},
'generation': {'every_n_steps': 16,
'force_call_on': [25354],
'metrics_configs': [{}, {'n': 1}, {'n': 2}, {'n': 5}],
'scenario_configs': [{'generate_kwargs': {'do_sample': True,
'max_length': 128,
'min_length': 10,
'temperature': 0.7,
'top_k': 0,
'top_p': 0.9},
'name': 'unconditional',
'num_samples': 2048},
{'generate_kwargs': {'do_sample': True,
'max_length': 128,
'min_length': 10,
'temperature': 0.7,
'top_k': 0,
'top_p': 0.9},
'name': 'challenging_rtp',
'num_samples': 2048,
'prompts_path': 'resources/challenging_rtp.jsonl'}],
'scorer_config': {'device': 'cuda:0'}},
'kl_gpt3_callback': {'every_n_steps': 16,
'force_call_on': [25354],
'max_tokens': 64,
'num_samples': 4096},
'model': {'from_scratch': True,
'gpt2_config_kwargs': {'reorder_and_upcast_attn': True,
'scale_attn_by': True},
'model_kwargs': {'value_head_config': {'is_detached': False}},
'path_or_name': 'gpt2'},
'objective': {'alpha': 1, 'beta': 10, 'name': 'AWR'},
'tokenizer': {'path_or_name': 'gpt2'},
'training': {'dataloader_num_workers': 0,
'effective_batch_size': 1024,
'evaluation_strategy': 'no',
'fp16': True,
'hub_model_id': 'musing_hoover',
'hub_strategy': 'all_checkpoints',
'learning_rate': 0.0005,
'logging_first_step': True,
'logging_steps': 1,
'num_tokens': 3300000000,
'output_dir': 'training_output104340',
'per_device_train_batch_size': 16,
'push_to_hub': True,
'remove_unused_columns': False,
'save_steps': 1673,
'save_strategy': 'steps',
'seed': 42,
'warmup_ratio': 0.01,
'weight_decay': 0.1}}
# Wandb URL:
https://wandb.ai/tomekkorbak/apo/runs/1nm8napp
|
fanzru/t5-small-finetuned-xsum-xlsum
|
fanzru
| 2022-12-06T09:28:20Z | 11 | 0 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"t5",
"text2text-generation",
"generated_from_trainer",
"dataset:scientific_papers",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-05T11:55:31Z |
---
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- scientific_papers
metrics:
- rouge
model-index:
- name: t5-small-finetuned-xsum-xlsum
results:
- task:
name: Sequence-to-sequence Language Modeling
type: text2text-generation
dataset:
name: scientific_papers
type: scientific_papers
config: pubmed
split: train
args: pubmed
metrics:
- name: Rouge1
type: rouge
value: 14.3541
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# t5-small-finetuned-xsum-xlsum
This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the scientific_papers dataset.
It achieves the following results on the evaluation set:
- Loss: 1.9963
- Rouge1: 14.3541
- Rouge2: 6.1674
- Rougel: 12.2975
- Rougelsum: 13.2515
- Gen Len: 19.0
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:------:|:-------:|:---------:|:-------:|
| 2.3055 | 1.0 | 7496 | 2.0773 | 14.3312 | 6.153 | 12.2551 | 13.2033 | 19.0 |
| 2.2512 | 2.0 | 14992 | 2.0330 | 14.3048 | 6.1346 | 12.2343 | 13.1992 | 19.0 |
| 2.2034 | 3.0 | 22488 | 2.0106 | 14.3866 | 6.1752 | 12.3205 | 13.2743 | 19.0 |
| 2.2054 | 4.0 | 29984 | 2.0004 | 14.3629 | 6.167 | 12.2928 | 13.2506 | 19.0 |
| 2.1944 | 5.0 | 37480 | 1.9963 | 14.3541 | 6.1674 | 12.2975 | 13.2515 | 19.0 |
### Framework versions
- Transformers 4.24.0
- Pytorch 1.13.0+cu117
- Datasets 2.7.1
- Tokenizers 0.13.2
|
ShadoWxShinigamI/SD2-Vray-Style
|
ShadoWxShinigamI
| 2022-12-06T09:24:19Z | 0 | 4 | null |
[
"license:creativeml-openrail-m",
"region:us"
] | null | 2022-12-06T09:18:10Z |
---
license: creativeml-openrail-m
---
##Textual Inversion Embed For SD 2.0 By ShadoWxShinigamI
This embed attempts to emulate the style and lighting of vray renderer. It has been trained for a total of 1000 steps based on 44 of my personal renders.
Model used for training:- SD 2.0 (512 Base). [Works well with the 768 Model]
This embed mixes well with other 2.0 embeds. Mix and have fun!
Examples:-
|
fathyshalab/all-roberta-large-v1-meta-16-16-5-oos
|
fathyshalab
| 2022-12-06T08:47:06Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T08:22:21Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: all-roberta-large-v1-meta-16-16-5-oos
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# all-roberta-large-v1-meta-16-16-5-oos
This model is a fine-tuned version of [sentence-transformers/all-roberta-large-v1](https://huggingface.co/sentence-transformers/all-roberta-large-v1) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.4797
- Accuracy: 0.28
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 2.7721 | 1.0 | 1 | 2.6529 | 0.1889 |
| 2.2569 | 2.0 | 2 | 2.5866 | 0.2333 |
| 1.9837 | 3.0 | 3 | 2.5340 | 0.2644 |
| 1.6425 | 4.0 | 4 | 2.4980 | 0.2756 |
| 1.4612 | 5.0 | 5 | 2.4797 | 0.28 |
### Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
|
AlanB/clip_guided_stable_diffusion_mod
|
AlanB
| 2022-12-06T08:35:13Z | 0 | 2 | null |
[
"license:openrail",
"region:us"
] | null | 2022-11-27T20:37:48Z |
---
license: openrail
---
Modified version of diffusers CLIP-Guided Community Pipeline.
Fixed incompatibility with Stable Diffusion v2, eliminated Safety warning
Made to go with my [Stable Diffusion Deluxe](https://colab.research.google.com/github/Skquark/AI-Friends/blob/main/Stable_Diffusion_Deluxe.ipynb) Notebook.
|
lmqg/t5-base-squad-ae
|
lmqg
| 2022-12-06T08:33:54Z | 40 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"answer extraction",
"en",
"dataset:lmqg/qg_squad",
"arxiv:2210.03992",
"license:cc-by-4.0",
"model-index",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-06T08:32:32Z |
---
license: cc-by-4.0
metrics:
- bleu4
- meteor
- rouge-l
- bertscore
- moverscore
language: en
datasets:
- lmqg/qg_squad
pipeline_tag: text2text-generation
tags:
- answer extraction
widget:
- text: "extract answers: <hl> Beyonce further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, Cadillac Records. <hl> Her performance in the film received praise from critics, and she garnered several nominations for her portrayal of James, including a Satellite Award nomination for Best Supporting Actress, and a NAACP Image Award nomination for Outstanding Supporting Actress."
example_title: "Answering Extraction Example 1"
- text: "extract answers: Beyonce further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, Cadillac Records. <hl> Her performance in the film received praise from critics, and she garnered several nominations for her portrayal of James, including a Satellite Award nomination for Best Supporting Actress, and a NAACP Image Award nomination for Outstanding Supporting Actress. <hl>"
example_title: "Answering Extraction Example 2"
model-index:
- name: lmqg/t5-base-squad-ae
results:
- task:
name: Text2text Generation
type: text2text-generation
dataset:
name: lmqg/qg_squad
type: default
args: default
metrics:
- name: BLEU4 (Answer Extraction)
type: bleu4_answer_extraction
value: 54.28
- name: ROUGE-L (Answer Extraction)
type: rouge_l_answer_extraction
value: 69.72
- name: METEOR (Answer Extraction)
type: meteor_answer_extraction
value: 43.62
- name: BERTScore (Answer Extraction)
type: bertscore_answer_extraction
value: 91.87
- name: MoverScore (Answer Extraction)
type: moverscore_answer_extraction
value: 82.69
- name: AnswerF1Score (Answer Extraction)
type: answer_f1_score__answer_extraction
value: 70.32
- name: AnswerExactMatch (Answer Extraction)
type: answer_exact_match_answer_extraction
value: 59.48
---
# Model Card of `lmqg/t5-base-squad-ae`
This model is fine-tuned version of [t5-base](https://huggingface.co/t5-base) for answer extraction on the [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) (dataset_name: default) via [`lmqg`](https://github.com/asahi417/lm-question-generation).
### Overview
- **Language model:** [t5-base](https://huggingface.co/t5-base)
- **Language:** en
- **Training data:** [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) (default)
- **Online Demo:** [https://autoqg.net/](https://autoqg.net/)
- **Repository:** [https://github.com/asahi417/lm-question-generation](https://github.com/asahi417/lm-question-generation)
- **Paper:** [https://arxiv.org/abs/2210.03992](https://arxiv.org/abs/2210.03992)
### Usage
- With [`lmqg`](https://github.com/asahi417/lm-question-generation#lmqg-language-model-for-question-generation-)
```python
from lmqg import TransformersQG
# initialize model
model = TransformersQG(language="en", model="lmqg/t5-base-squad-ae")
# model prediction
answers = model.generate_a("William Turner was an English painter who specialised in watercolour landscapes")
```
- With `transformers`
```python
from transformers import pipeline
pipe = pipeline("text2text-generation", "lmqg/t5-base-squad-ae")
output = pipe("extract answers: <hl> Beyonce further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, Cadillac Records. <hl> Her performance in the film received praise from critics, and she garnered several nominations for her portrayal of James, including a Satellite Award nomination for Best Supporting Actress, and a NAACP Image Award nomination for Outstanding Supporting Actress.")
```
## Evaluation
- ***Metric (Answer Extraction)***: [raw metric file](https://huggingface.co/lmqg/t5-base-squad-ae/raw/main/eval/metric.first.answer.paragraph_sentence.answer.lmqg_qg_squad.default.json)
| | Score | Type | Dataset |
|:-----------------|--------:|:--------|:---------------------------------------------------------------|
| AnswerExactMatch | 59.48 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) |
| AnswerF1Score | 70.32 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) |
| BERTScore | 91.87 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) |
| Bleu_1 | 64.27 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) |
| Bleu_2 | 60.78 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) |
| Bleu_3 | 57.35 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) |
| Bleu_4 | 54.28 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) |
| METEOR | 43.62 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) |
| MoverScore | 82.69 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) |
| ROUGE_L | 69.72 | default | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) |
## Training hyperparameters
The following hyperparameters were used during fine-tuning:
- dataset_path: lmqg/qg_squad
- dataset_name: default
- input_types: ['paragraph_sentence']
- output_types: ['answer']
- prefix_types: ['ae']
- model: t5-base
- max_length: 512
- max_length_output: 32
- epoch: 8
- batch: 16
- lr: 0.0001
- fp16: False
- random_seed: 1
- gradient_accumulation_steps: 4
- label_smoothing: 0.0
The full configuration can be found at [fine-tuning config file](https://huggingface.co/lmqg/t5-base-squad-ae/raw/main/trainer_config.json).
## Citation
```
@inproceedings{ushio-etal-2022-generative,
title = "{G}enerative {L}anguage {M}odels for {P}aragraph-{L}evel {Q}uestion {G}eneration",
author = "Ushio, Asahi and
Alva-Manchego, Fernando and
Camacho-Collados, Jose",
booktitle = "Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing",
month = dec,
year = "2022",
address = "Abu Dhabi, U.A.E.",
publisher = "Association for Computational Linguistics",
}
```
|
fathyshalab/all-roberta-large-v1-meta-8-16-5-oos
|
fathyshalab
| 2022-12-06T08:22:01Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T07:55:47Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: all-roberta-large-v1-meta-8-16-5-oos
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# all-roberta-large-v1-meta-8-16-5-oos
This model is a fine-tuned version of [sentence-transformers/all-roberta-large-v1](https://huggingface.co/sentence-transformers/all-roberta-large-v1) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.4797
- Accuracy: 0.28
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 2.7721 | 1.0 | 1 | 2.6529 | 0.1889 |
| 2.2569 | 2.0 | 2 | 2.5866 | 0.2333 |
| 1.9837 | 3.0 | 3 | 2.5340 | 0.2644 |
| 1.6425 | 4.0 | 4 | 2.4980 | 0.2756 |
| 1.4612 | 5.0 | 5 | 2.4797 | 0.28 |
### Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
|
betbhai9/Betbhai9
|
betbhai9
| 2022-12-06T08:11:04Z | 0 | 0 | null |
[
"region:us"
] | null | 2022-12-06T08:09:43Z |
The Betbhai9id se ap online earning karsaktehai, but iskeliyeapko online play karnapadega aur play karnekeliyeapko pay karnapadega. Uskeliyeapko ek id ki need padegi to apko wo ki zrurathogi. [Betbhai9](https://betbhai9.app) id ham user ko provide kartehai. Ap hamari website par visit kareurwhatsappke throw Betbhai9id le saktehai.
|
fathyshalab/all-roberta-large-v1-meta-4-16-5-oos
|
fathyshalab
| 2022-12-06T07:55:27Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T07:31:16Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: all-roberta-large-v1-meta-4-16-5-oos
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# all-roberta-large-v1-meta-4-16-5-oos
This model is a fine-tuned version of [sentence-transformers/all-roberta-large-v1](https://huggingface.co/sentence-transformers/all-roberta-large-v1) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.4797
- Accuracy: 0.28
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 2.7721 | 1.0 | 1 | 2.6529 | 0.1889 |
| 2.2569 | 2.0 | 2 | 2.5866 | 0.2333 |
| 1.9837 | 3.0 | 3 | 2.5340 | 0.2644 |
| 1.6425 | 4.0 | 4 | 2.4980 | 0.2756 |
| 1.4612 | 5.0 | 5 | 2.4797 | 0.28 |
### Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
|
HaojiePan/wav2vec2-base-ft-keyword-spotting
|
HaojiePan
| 2022-12-06T07:33:58Z | 10 | 0 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"wav2vec2",
"audio-classification",
"generated_from_trainer",
"dataset:superb",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] |
audio-classification
| 2022-12-06T07:13:37Z |
---
license: apache-2.0
tags:
- audio-classification
- generated_from_trainer
datasets:
- superb
metrics:
- accuracy
model-index:
- name: wav2vec2-base-ft-keyword-spotting
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# wav2vec2-base-ft-keyword-spotting
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the superb dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0795
- Accuracy: 0.9829
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 0
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5.0
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 0.5546 | 1.0 | 399 | 0.4250 | 0.9618 |
| 0.2128 | 2.0 | 798 | 0.1331 | 0.9781 |
| 0.1763 | 3.0 | 1197 | 0.0935 | 0.9807 |
| 0.1485 | 4.0 | 1596 | 0.0852 | 0.9828 |
| 0.1335 | 5.0 | 1995 | 0.0795 | 0.9829 |
### Framework versions
- Transformers 4.25.1
- Pytorch 1.10.0+cu111
- Datasets 2.7.1
- Tokenizers 0.13.2
|
fathyshalab/all-roberta-large-v1-meta-2-16-5-oos
|
fathyshalab
| 2022-12-06T07:30:57Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T07:07:07Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: all-roberta-large-v1-meta-2-16-5-oos
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# all-roberta-large-v1-meta-2-16-5-oos
This model is a fine-tuned version of [sentence-transformers/all-roberta-large-v1](https://huggingface.co/sentence-transformers/all-roberta-large-v1) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.4797
- Accuracy: 0.28
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 2.7721 | 1.0 | 1 | 2.6529 | 0.1889 |
| 2.2569 | 2.0 | 2 | 2.5866 | 0.2333 |
| 1.9837 | 3.0 | 3 | 2.5340 | 0.2644 |
| 1.6425 | 4.0 | 4 | 2.4980 | 0.2756 |
| 1.4612 | 5.0 | 5 | 2.4797 | 0.28 |
### Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
|
AlekseyKorshuk/6.7b-ri-reproduce-combined-4-gpu-20-val-v3
|
AlekseyKorshuk
| 2022-12-06T07:07:47Z | 5 | 0 |
transformers
|
[
"transformers",
"pytorch",
"opt",
"text-generation",
"generated_from_trainer",
"license:other",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2022-12-05T11:54:11Z |
---
license: other
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: 6.7b-ri-reproduce-combined-4-gpu-20-val-v3
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# 6.7b-ri-reproduce-combined-4-gpu-20-val-v3
This model is a fine-tuned version of [facebook/opt-6.7b](https://huggingface.co/facebook/opt-6.7b) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 3.9434
- Accuracy: 0.0329
- Perplexity: 51.5916
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 9e-07
- train_batch_size: 1
- eval_batch_size: 8
- seed: 100
- distributed_type: multi-GPU
- num_devices: 4
- total_train_batch_size: 4
- total_eval_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- num_epochs: 15.0
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Perplexity |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:----------:|
| 2.5731 | 1.0 | 79 | 2.6113 | 0.0317 | 13.6171 |
| 2.206 | 2.0 | 158 | 2.4805 | 0.0328 | 11.9469 |
| 1.9105 | 3.0 | 237 | 2.4512 | 0.0333 | 11.6019 |
| 1.6301 | 4.0 | 316 | 2.5078 | 0.0345 | 12.2780 |
| 1.3733 | 5.0 | 395 | 2.6816 | 0.0342 | 14.6090 |
| 1.1337 | 6.0 | 474 | 3.0078 | 0.0330 | 20.2431 |
| 0.9619 | 7.0 | 553 | 3.1777 | 0.0330 | 23.9923 |
| 0.798 | 8.0 | 632 | 3.2559 | 0.0330 | 25.9419 |
| 0.6653 | 9.0 | 711 | 3.4277 | 0.0331 | 30.8068 |
| 0.552 | 10.0 | 790 | 3.5566 | 0.0333 | 35.0453 |
| 0.4568 | 11.0 | 869 | 3.7324 | 0.0324 | 41.7802 |
| 0.3756 | 12.0 | 948 | 3.8184 | 0.0328 | 45.5295 |
| 0.3119 | 13.0 | 1027 | 3.8477 | 0.0331 | 46.8831 |
| 0.2448 | 14.0 | 1106 | 3.9062 | 0.0329 | 49.7122 |
| 0.1986 | 15.0 | 1185 | 3.9434 | 0.0329 | 51.5916 |
### Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.12.1+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1
|
huggingtweets/emilyhxrrera-floguo-lucy_guo-saraduit-shrawberryy
|
huggingtweets
| 2022-12-06T06:51:32Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"gpt2",
"text-generation",
"huggingtweets",
"en",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2022-12-06T06:51:23Z |
---
language: en
thumbnail: https://github.com/borisdayma/huggingtweets/blob/master/img/logo.png?raw=true
tags:
- huggingtweets
widget:
- text: "My dream is"
---
<div class="inline-flex flex-col" style="line-height: 1.5;">
<div class="flex">
<div
style="display:inherit; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('https://pbs.twimg.com/profile_images/1596198683179159557/-l7jFkeQ_400x400.jpg')">
</div>
<div
style="display:inherit; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('https://pbs.twimg.com/profile_images/1472319181097824256/hY5RmhQs_400x400.jpg')">
</div>
<div
style="display:inherit; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('https://pbs.twimg.com/profile_images/1571251248342650882/6YDG9PGc_400x400.jpg')">
</div>
</div>
<div style="text-align: center; margin-top: 3px; font-size: 16px; font-weight: 800">🤖 AI CYBORG 🤖</div>
<div style="text-align: center; font-size: 16px; font-weight: 800">floguo & em herrera is in NY 🌃 & Shravani🍓 & Sara Du & Lucy Guo (Hiring Engineers & Designers)</div>
<div style="text-align: center; font-size: 14px;">@emilyhxrrera-floguo-lucy_guo-saraduit-shrawberryy</div>
</div>
I was made with [huggingtweets](https://github.com/borisdayma/huggingtweets).
Create your own bot based on your favorite user with [the demo](https://colab.research.google.com/github/borisdayma/huggingtweets/blob/master/huggingtweets-demo.ipynb)!
## How does it work?
The model uses the following pipeline.
To understand how the model was developed, check the [W&B report](https://wandb.ai/wandb/huggingtweets/reports/HuggingTweets-Train-a-Model-to-Generate-Tweets--VmlldzoxMTY5MjI).
## Training data
The model was trained on tweets from floguo & em herrera is in NY 🌃 & Shravani🍓 & Sara Du & Lucy Guo (Hiring Engineers & Designers).
| Data | floguo | em herrera is in NY 🌃 | Shravani🍓 | Sara Du | Lucy Guo (Hiring Engineers & Designers) |
| --- | --- | --- | --- | --- | --- |
| Tweets downloaded | 3193 | 3234 | 1049 | 1635 | 3239 |
| Retweets | 662 | 488 | 92 | 17 | 68 |
| Short tweets | 423 | 829 | 328 | 287 | 275 |
| Tweets kept | 2108 | 1917 | 629 | 1331 | 2896 |
[Explore the data](https://wandb.ai/wandb/huggingtweets/runs/3kqf9fmj/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
## Training procedure
The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on @emilyhxrrera-floguo-lucy_guo-saraduit-shrawberryy's tweets.
Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/wandb/huggingtweets/runs/1h2quh2b) for full transparency and reproducibility.
At the end of training, [the final model](https://wandb.ai/wandb/huggingtweets/runs/1h2quh2b/artifacts) is logged and versioned.
## How to use
You can use this model directly with a pipeline for text generation:
```python
from transformers import pipeline
generator = pipeline('text-generation',
model='huggingtweets/emilyhxrrera-floguo-lucy_guo-saraduit-shrawberryy')
generator("My dream is", num_return_sequences=5)
```
## Limitations and bias
The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias).
In addition, the data present in the user's tweets further affects the text generated by the model.
## About
*Built by Boris Dayma*
[](https://twitter.com/intent/follow?screen_name=borisdayma)
For more details, visit the project repository.
[](https://github.com/borisdayma/huggingtweets)
|
m-aliabbas/idrak_wav2vec_tr
|
m-aliabbas
| 2022-12-06T06:38:53Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"wav2vec2",
"automatic-speech-recognition",
"generated_from_trainer",
"dataset:common_voice",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] |
automatic-speech-recognition
| 2022-12-06T05:57:48Z |
---
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- common_voice
model-index:
- name: idrak_wav2vec_tr
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# idrak_wav2vec_tr
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the common_voice dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0003
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 30
- mixed_precision_training: Native AMP
### Framework versions
- Transformers 4.11.3
- Pytorch 1.10.0+cu113
- Datasets 1.18.3
- Tokenizers 0.10.3
|
fathyshalab/all-roberta-large-v1-small_talk-8-16-5-oos
|
fathyshalab
| 2022-12-06T06:18:14Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T05:54:47Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: all-roberta-large-v1-small_talk-8-16-5-oos
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# all-roberta-large-v1-small_talk-8-16-5-oos
This model is a fine-tuned version of [sentence-transformers/all-roberta-large-v1](https://huggingface.co/sentence-transformers/all-roberta-large-v1) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.3566
- Accuracy: 0.3855
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 2.7259 | 1.0 | 1 | 2.5917 | 0.2551 |
| 2.217 | 2.0 | 2 | 2.5059 | 0.3275 |
| 1.7237 | 3.0 | 3 | 2.4355 | 0.3768 |
| 1.4001 | 4.0 | 4 | 2.3837 | 0.3739 |
| 1.1937 | 5.0 | 5 | 2.3566 | 0.3855 |
### Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
|
Nhat1904/32-shot-twitter
|
Nhat1904
| 2022-12-06T06:17:21Z | 1 | 0 |
sentence-transformers
|
[
"sentence-transformers",
"pytorch",
"mpnet",
"feature-extraction",
"sentence-similarity",
"transformers",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
sentence-similarity
| 2022-12-06T06:17:07Z |
---
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- transformers
---
# {MODEL_NAME}
This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search.
<!--- Describe your model here -->
## Usage (Sentence-Transformers)
Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed:
```
pip install -U sentence-transformers
```
Then you can use the model like this:
```python
from sentence_transformers import SentenceTransformer
sentences = ["This is an example sentence", "Each sentence is converted"]
model = SentenceTransformer('{MODEL_NAME}')
embeddings = model.encode(sentences)
print(embeddings)
```
## Usage (HuggingFace Transformers)
Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings.
```python
from transformers import AutoTokenizer, AutoModel
import torch
#Mean Pooling - Take attention mask into account for correct averaging
def mean_pooling(model_output, attention_mask):
token_embeddings = model_output[0] #First element of model_output contains all token embeddings
input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)
# Sentences we want sentence embeddings for
sentences = ['This is an example sentence', 'Each sentence is converted']
# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}')
model = AutoModel.from_pretrained('{MODEL_NAME}')
# Tokenize sentences
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
# Compute token embeddings
with torch.no_grad():
model_output = model(**encoded_input)
# Perform pooling. In this case, mean pooling.
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
print("Sentence embeddings:")
print(sentence_embeddings)
```
## Evaluation Results
<!--- Describe how your model was evaluated -->
For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME})
## Training
The model was trained with the parameters:
**DataLoader**:
`torch.utils.data.dataloader.DataLoader` of length 384 with parameters:
```
{'batch_size': 10, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}
```
**Loss**:
`sentence_transformers.losses.CosineSimilarityLoss.CosineSimilarityLoss`
Parameters of the fit()-Method:
```
{
"epochs": 1,
"evaluation_steps": 0,
"evaluator": "NoneType",
"max_grad_norm": 1,
"optimizer_class": "<class 'torch.optim.adamw.AdamW'>",
"optimizer_params": {
"lr": 2e-05
},
"scheduler": "WarmupLinear",
"steps_per_epoch": 384,
"warmup_steps": 39,
"weight_decay": 0.01
}
```
## Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: MPNetModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
)
```
## Citing & Authors
<!--- Describe where people can find more information -->
|
StonyBrookNLP/teabreac-preasm-large-iirc-retrieved
|
StonyBrookNLP
| 2022-12-06T06:05:43Z | 5 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T23:11:18Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/teabreac-preasm-large-iirc-retrieved"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"Who scored the first touchdown of the game?\n" +
"... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/teabreac-preasm-large-iirc-gold
|
StonyBrookNLP
| 2022-12-06T06:03:50Z | 5 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T23:09:27Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/teabreac-preasm-large-iirc-gold"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"Who scored the first touchdown of the game?\n" +
"... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/teabreac-preasm-large-drop
|
StonyBrookNLP
| 2022-12-06T06:02:04Z | 5 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T23:07:27Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/teabreac-preasm-large-drop"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"Who scored the first touchdown of the game?\n" +
"... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/teabreac-preasm-large
|
StonyBrookNLP
| 2022-12-06T05:59:46Z | 5 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T23:05:35Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
# NOTE: This model is only pretrained on TeaBReaC, and not on any real QA dataset.
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/teabreac-preasm-large"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"Who scored the first touchdown of the game?\n" +
"... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/teabreac-nt5-small-tatqa
|
StonyBrookNLP
| 2022-12-06T05:58:33Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T23:05:16Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/teabreac-nt5-small-tatqa"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"answer_me: Who scored the first touchdown of the game?" +
"context: ... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/teabreac-nt5-small-iirc-gold
|
StonyBrookNLP
| 2022-12-06T05:57:25Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T23:04:20Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/teabreac-nt5-small-iirc-gold"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"answer_me: Who scored the first touchdown of the game?" +
"context: ... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/teabreac-t5-3b-tatqa
|
StonyBrookNLP
| 2022-12-06T05:53:08Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T22:56:36Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/teabreac-t5-3b-tatqa"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"answer_me: Who scored the first touchdown of the game?" +
"context: ... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/teabreac-t5-3b-iirc-gold
|
StonyBrookNLP
| 2022-12-06T05:33:31Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T22:36:44Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/teabreac-t5-3b-iirc-gold"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"answer_me: Who scored the first touchdown of the game?" +
"context: ... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
dung1308/RM_system_not_mixed__NLP_model_80_20_CPU
|
dung1308
| 2022-12-06T05:31:31Z | 3 | 0 |
transformers
|
[
"transformers",
"tf",
"roberta",
"fill-mask",
"generated_from_keras_callback",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
fill-mask
| 2022-12-05T16:45:43Z |
---
tags:
- generated_from_keras_callback
model-index:
- name: dung1308/RM_system_not_mixed__NLP_model_80_20_CPU
results: []
---
<!-- This model card has been generated automatically according to the information Keras had access to. You should
probably proofread and complete it, then remove this comment. -->
# dung1308/RM_system_not_mixed__NLP_model_80_20_CPU
This model is a fine-tuned version of [vinai/phobert-base](https://huggingface.co/vinai/phobert-base) on an unknown dataset.
It achieves the following results on the evaluation set:
- Train Loss: 4.2681
- Validation Loss: 4.2461
- Epoch: 3
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'WarmUp', 'config': {'initial_learning_rate': 2e-05, 'decay_schedule_fn': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': -356, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, '__passive_serialization__': True}, 'warmup_steps': 1000, 'power': 1.0, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
- training_precision: mixed_float16
### Training results
| Train Loss | Validation Loss | Epoch |
|:----------:|:---------------:|:-----:|
| 5.1497 | 4.3969 | 0 |
| 4.3110 | 4.2424 | 1 |
| 4.2373 | 4.2722 | 2 |
| 4.2681 | 4.2461 | 3 |
### Framework versions
- Transformers 4.18.0
- TensorFlow 2.8.0
- Datasets 2.7.0
- Tokenizers 0.11.0
|
fathyshalab/all-roberta-large-v1-small_talk-2-16-5-oos
|
fathyshalab
| 2022-12-06T05:30:14Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T05:06:55Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: all-roberta-large-v1-small_talk-2-16-5-oos
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# all-roberta-large-v1-small_talk-2-16-5-oos
This model is a fine-tuned version of [sentence-transformers/all-roberta-large-v1](https://huggingface.co/sentence-transformers/all-roberta-large-v1) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.3566
- Accuracy: 0.3855
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 2.7259 | 1.0 | 1 | 2.5917 | 0.2551 |
| 2.217 | 2.0 | 2 | 2.5059 | 0.3275 |
| 1.7237 | 3.0 | 3 | 2.4355 | 0.3768 |
| 1.4001 | 4.0 | 4 | 2.3837 | 0.3739 |
| 1.1937 | 5.0 | 5 | 2.3566 | 0.3855 |
### Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
|
StonyBrookNLP/teabreac-t5-3b-drop
|
StonyBrookNLP
| 2022-12-06T05:27:19Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T07:38:36Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/teabreac-t5-3b-drop"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"answer_me: Who scored the first touchdown of the game?" +
"context: ... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/teabreac-t5-3b
|
StonyBrookNLP
| 2022-12-06T05:21:13Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T22:23:02Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
# NOTE: This model is only pretrained on TeaBReaC, and not on any real QA dataset.
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/teabreac-t5-3b"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"answer_me: Who scored the first touchdown of the game?" +
"context: ... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/teabreac-t5-large-tatqa
|
StonyBrookNLP
| 2022-12-06T05:16:13Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T07:37:43Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/teabreac-t5-large-tatqa"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"answer_me: Who scored the first touchdown of the game?" +
"context: ... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/teabreac-t5-large-numglue
|
StonyBrookNLP
| 2022-12-06T05:14:10Z | 5 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T07:35:56Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/teabreac-t5-large-numglue"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"answer_me: Who scored the first touchdown of the game?" +
"context: ... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/teabreac-t5-large-drop
|
StonyBrookNLP
| 2022-12-06T05:08:03Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T07:30:47Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/teabreac-t5-large-drop"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"answer_me: Who scored the first touchdown of the game?" +
"context: ... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/preasm-large-tatqa
|
StonyBrookNLP
| 2022-12-06T04:55:56Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T07:17:29Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/preasm-large-tatqa"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"Who scored the first touchdown of the game?\n" +
"... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/preasm-large-numglue
|
StonyBrookNLP
| 2022-12-06T04:54:11Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T07:15:46Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/preasm-large-numglue"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"Who scored the first touchdown of the game?\n" +
"... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
alanila/autotrain-tc_ac-2349273884
|
alanila
| 2022-12-06T04:51:06Z | 1 | 0 |
transformers
|
[
"transformers",
"pytorch",
"autotrain",
"text-classification",
"en",
"dataset:alanila/autotrain-data-tc_ac",
"co2_eq_emissions",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T04:49:49Z |
---
tags:
- autotrain
- text-classification
language:
- en
widget:
- text: "I love AutoTrain 🤗"
datasets:
- alanila/autotrain-data-tc_ac
co2_eq_emissions:
emissions: 1.196433244085964
---
# Model Trained Using AutoTrain
- Problem type: Multi-class Classification
- Model ID: 2349273884
- CO2 Emissions (in grams): 1.1964
## Validation Metrics
- Loss: 1.271
- Accuracy: 0.517
- Macro F1: 0.465
- Micro F1: 0.517
- Weighted F1: 0.437
- Macro Precision: 0.495
- Micro Precision: 0.517
- Weighted Precision: 0.488
- Macro Recall: 0.501
- Micro Recall: 0.517
- Weighted Recall: 0.517
## Usage
You can use cURL to access this model:
```
$ curl -X POST -H "Authorization: Bearer YOUR_API_KEY" -H "Content-Type: application/json" -d '{"inputs": "I love AutoTrain"}' https://api-inference.huggingface.co/models/alanila/autotrain-tc_ac-2349273884
```
Or Python API:
```
from transformers import AutoModelForSequenceClassification, AutoTokenizer
model = AutoModelForSequenceClassification.from_pretrained("alanila/autotrain-tc_ac-2349273884", use_auth_token=True)
tokenizer = AutoTokenizer.from_pretrained("alanila/autotrain-tc_ac-2349273884", use_auth_token=True)
inputs = tokenizer("I love AutoTrain", return_tensors="pt")
outputs = model(**inputs)
```
|
StonyBrookNLP/preasm-large-iirc-gold
|
StonyBrookNLP
| 2022-12-06T04:50:11Z | 6 | 1 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T07:12:15Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/preasm-large-iirc-gold"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"Who scored the first touchdown of the game?\n" +
"... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/preasm-large-drop
|
StonyBrookNLP
| 2022-12-06T04:48:33Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T07:10:16Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/preasm-large-drop"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"Who scored the first touchdown of the game?\n" +
"... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/nt5-small-tatqa
|
StonyBrookNLP
| 2022-12-06T04:47:10Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T07:10:00Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/nt5-small-tatqa"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"answer_me: Who scored the first touchdown of the game?" +
"context: ... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/nt5-small-drop
|
StonyBrookNLP
| 2022-12-06T04:45:52Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T07:08:58Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/nt5-small-drop"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"answer_me: Who scored the first touchdown of the game?" +
"context: ... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
fathyshalab/all-roberta-large-v1-work-8-16-5-oos
|
fathyshalab
| 2022-12-06T04:18:45Z | 5 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T03:55:31Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: all-roberta-large-v1-work-8-16-5-oos
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# all-roberta-large-v1-work-8-16-5-oos
This model is a fine-tuned version of [sentence-transformers/all-roberta-large-v1](https://huggingface.co/sentence-transformers/all-roberta-large-v1) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.3586
- Accuracy: 0.3689
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 2.8058 | 1.0 | 1 | 2.6169 | 0.2356 |
| 2.3524 | 2.0 | 2 | 2.5215 | 0.2978 |
| 1.9543 | 3.0 | 3 | 2.4427 | 0.3422 |
| 1.5539 | 4.0 | 4 | 2.3874 | 0.36 |
| 1.4133 | 5.0 | 5 | 2.3586 | 0.3689 |
### Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
|
StonyBrookNLP/t5-large-numglue
|
StonyBrookNLP
| 2022-12-06T04:10:26Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T06:44:29Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/t5-large-numglue"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"answer_me: Who scored the first touchdown of the game?" +
"context: ... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/t5-large-iirc-retrieved
|
StonyBrookNLP
| 2022-12-06T04:08:49Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T06:42:48Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/t5-large-iirc-retrieved"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"answer_me: Who scored the first touchdown of the game?" +
"context: ... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
StonyBrookNLP/t5-large-iirc-gold
|
StonyBrookNLP
| 2022-12-06T04:07:07Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering, multi-step-reasoning, multi-hop-reasoning",
"arxiv:2205.12496",
"license:cc-by-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-12-03T06:41:16Z |
---
tags:
- question-answering, multi-step-reasoning, multi-hop-reasoning
thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
license: cc-by-4.0
---
# What's this?
This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
We release the following models:
- **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
- **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
- **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
# How to use it?
Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
```python
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
model_name = "StonyBrookNLP/t5-large-iirc-gold"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
enable_digit_tokenization(tokenizer)
input_texts = [
"answer_me: Who scored the first touchdown of the game?" +
"context: ... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
# Note: some models have slightly different qn/ctxt format. See the github repo.
]
input_ids = tokenizer(
input_texts, return_tensors="pt",
truncation=True, max_length=800,
add_special_tokens=True, padding=True,
)["input_ids"]
generated_ids = model.generate(input_ids, min_length=1, max_length=50)
generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
generated_predictions = [
tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
]
# => ["Chaz Schilens"]
```
|
fathyshalab/all-roberta-large-v1-work-4-16-5-oos
|
fathyshalab
| 2022-12-06T03:55:07Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T03:31:50Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: all-roberta-large-v1-work-4-16-5-oos
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# all-roberta-large-v1-work-4-16-5-oos
This model is a fine-tuned version of [sentence-transformers/all-roberta-large-v1](https://huggingface.co/sentence-transformers/all-roberta-large-v1) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.3586
- Accuracy: 0.3689
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 2.8058 | 1.0 | 1 | 2.6169 | 0.2356 |
| 2.3524 | 2.0 | 2 | 2.5215 | 0.2978 |
| 1.9543 | 3.0 | 3 | 2.4427 | 0.3422 |
| 1.5539 | 4.0 | 4 | 2.3874 | 0.36 |
| 1.4133 | 5.0 | 5 | 2.3586 | 0.3689 |
### Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
|
lilanxi0221/distilbert-base-uncased-finetuned-cola
|
lilanxi0221
| 2022-12-06T03:54:02Z | 5 | 0 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"distilbert",
"text-classification",
"generated_from_trainer",
"dataset:glue",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-02T16:36:24Z |
---
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- matthews_correlation
model-index:
- name: distilbert-base-uncased-finetuned-cola
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: glue
type: glue
config: cola
split: train
args: cola
metrics:
- name: Matthews Correlation
type: matthews_correlation
value: 0.5552849676135797
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# distilbert-base-uncased-finetuned-cola
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the glue dataset.
It achieves the following results on the evaluation set:
- Loss: 0.7539
- Matthews Correlation: 0.5553
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Matthews Correlation |
|:-------------:|:-----:|:----:|:---------------:|:--------------------:|
| 0.5245 | 1.0 | 535 | 0.5223 | 0.4063 |
| 0.3574 | 2.0 | 1070 | 0.4856 | 0.5079 |
| 0.2461 | 3.0 | 1605 | 0.5503 | 0.5279 |
| 0.1909 | 4.0 | 2140 | 0.6974 | 0.5288 |
| 0.1451 | 5.0 | 2675 | 0.7539 | 0.5553 |
### Framework versions
- Transformers 4.25.1
- Pytorch 1.12.1+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2
|
arb9p4/ppo-LunarLander-v2
|
arb9p4
| 2022-12-06T03:40:46Z | 0 | 0 |
stable-baselines3
|
[
"stable-baselines3",
"LunarLander-v2",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2022-12-06T03:40:26Z |
---
library_name: stable-baselines3
tags:
- LunarLander-v2
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: PPO
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: LunarLander-v2
type: LunarLander-v2
metrics:
- type: mean_reward
value: 257.80 +/- 15.76
name: mean_reward
verified: false
---
# **PPO** Agent playing **LunarLander-v2**
This is a trained model of a **PPO** agent playing **LunarLander-v2**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3).
## Usage (with Stable-baselines3)
TODO: Add your code
```python
from stable_baselines3 import ...
from huggingface_sb3 import load_from_hub
...
```
|
nlpconnect/deberta-v3-xsmall-squad2
|
nlpconnect
| 2022-12-06T03:37:01Z | 15 | 0 |
transformers
|
[
"transformers",
"pytorch",
"deberta-v2",
"question-answering",
"dataset:squad_v2",
"license:apache-2.0",
"model-index",
"endpoints_compatible",
"region:us"
] |
question-answering
| 2022-08-28T10:24:54Z |
---
license: apache-2.0
datasets:
- squad_v2
model-index:
- name: nlpconnect/deberta-v3-xsmall-squad2
results:
- task:
type: question-answering
name: Question Answering
dataset:
name: squad_v2
type: squad_v2
config: squad_v2
split: validation
metrics:
- type: exact_match
value: 79.3917
name: Exact Match
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZTFiMWI5YzFlMDZhMzc2NDIwYjNiZmIyMThmOWQxYjFjZmM2ZDQ0OGM2NmNlNmI3Y2U2N2JjMmVkZTgyZjNiOCIsInZlcnNpb24iOjF9.MCw9UJ3MI3Lf5hvOgk7Lw2xZfN4678p7ebG3vnGXX_Avw6fELTPwxZ9qGA-9tL00p4NxaSb3Cx6XAFvWetAIBA
- type: f1
value: 82.6738
name: F1
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiMjdiYWY2MzU4YjZhMWQzZGJhZTk3NzU3Y2UwYmQ4MzliZmQxOGUxZDllN2Y0ZmZhYjVlNTE0MzY1MjU5OWMwMCIsInZlcnNpb24iOjF9.zeWLwXy77n0YKxGA5gjySe8p-_nPQxbiPnvQU2tF45IyMmlYKUuLeq4hJnNe-5NgriTf8xkBJBE7Cr5lWHy_Cw
- task:
type: question-answering
name: Question Answering
dataset:
name: squad
type: squad
config: plain_text
split: validation
metrics:
- type: exact_match
value: 84.9246
name: Exact Match
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZGJhYmU0Y2I4Y2UyOGVlOTlkMmQ2OTcyMTZkNTkwNTMzNzhmNzZiYjU4ZDkxMGM5NzAyMjk1M2ExNGIzOWU4NCIsInZlcnNpb24iOjF9.ql1rCId6lQ7Uwq2spG3q2fFppkFGHA1IWQjvyPRhvKdRNzApBO0mu9JjMAv4uNKZX-kmGEkI018_9tAzN7kwDw
- type: f1
value: 91.6201
name: F1
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZjBjMmI0OTFmODVjMzllZDM0NTdmNjU4NGI4NzA4NTJhOWVkMDQ5OTY0MDcyMWEwZTFkODNlY2VhZjU2NWJmZSIsInZlcnNpb24iOjF9.rGvF60bfWIXzB66C7fkdxCtZvRZ_m3onbLaNbs7M4M0Fk27xnMat6IAy1DeTztkOKLoiD2s2NQH6wXid83cgCw
---
# Deberta-v3-xsmall-squad2
## What is SQuAD?
Stanford Question Answering Dataset (SQuAD) is a reading comprehension dataset, consisting of questions posed by crowdworkers on a set of Wikipedia articles, where the answer to every question is a segment of text, or span, from the corresponding reading passage, or the question might be unanswerable.
SQuAD2.0 combines the 100,000 questions in SQuAD1.1 with over 50,000 unanswerable questions written adversarially by crowdworkers to look similar to answerable ones. To do well on SQuAD2.0, systems must not only answer questions when possible, but also determine when no answer is supported by the paragraph and abstain from answering.
## Inference
```python
from transformers import pipeline
qa = pipeline("question-answering", model="nlpconnect/deberta-v3-xsmall-squad2")
result = qa(context="My name is Sarah and I live in London", question="Where do I live?")
```
## Accuracy
```json
squad_v2 = {'exact': 79.392,
'f1': 82.674}
squad = {'exact': 84.925,
'f1': 91.620}
```
|
neulab/reatt-large-nq-fiqa
|
neulab
| 2022-12-06T03:13:27Z | 58 | 2 |
transformers
|
[
"transformers",
"pytorch",
"t5",
"text2text-generation",
"question-answering",
"en",
"arxiv:2212.02027",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
question-answering
| 2022-12-04T22:40:39Z |
---
language: en
tags:
- question-answering
---
# ReAtt
ReAtt is a retrieval-augmented model for knowledge-intensive tasks proposed in [Retrieval as Attention: End-to-end Learning of Retrieval and Reading within a Single Transformer](https://arxiv.org/pdf/2212.02027.pdf). The original Github repository is [https://github.com/jzbjyb/ReAtt](https://github.com/jzbjyb/ReAtt).
## Description
`neulab/reatt-large-nq-fid` (based on T5 architecture) is initialized with `neulab/reatt-large-nq` and adapted on FiQA dataset with end-to-end retrieval-augmented training.
## Usage
Please refer to [https://github.com/jzbjyb/ReAtt](https://github.com/jzbjyb/ReAtt) for instructions to use this model.
## Reference
```bibtex
@inproceedings{jiang-etal-2022-reatt,
title = {Retrieval as Attention: End-to-end Learning of Retrieval and Reading within a Single Transformer},
author = {Zhengbao Jiang and Luyu Gao and Jun Araki and Haibo Ding and Zhiruo Wang and Jamie Callan and Graham Neubig},
booktitle = {Conference on Empirical Methods in Natural Language Processing (EMNLP)},
address = {Abu Dhabi, UAE},
month = {December},
year = {2022}
}
```
|
Nhat1904/test_trainer_XLNET_3ep_5e-5
|
Nhat1904
| 2022-12-06T03:10:16Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"xlnet",
"text-classification",
"generated_from_trainer",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T01:30:37Z |
---
license: mit
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: test_trainer_XLNET_3ep_5e-5
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# test_trainer_XLNET_3ep_5e-5
This model is a fine-tuned version of [xlnet-base-cased](https://huggingface.co/xlnet-base-cased) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5405
- Accuracy: 0.8773
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 0.7984 | 1.0 | 1125 | 0.6647 | 0.7923 |
| 0.5126 | 2.0 | 2250 | 0.4625 | 0.862 |
| 0.409 | 3.0 | 3375 | 0.5405 | 0.8773 |
### Framework versions
- Transformers 4.25.1
- Pytorch 1.12.1+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2
|
Nadav/bert-base-historic-english-cased-squad-en
|
Nadav
| 2022-12-06T02:57:42Z | 5 | 0 |
transformers
|
[
"transformers",
"pytorch",
"bert",
"question-answering",
"generated_from_trainer",
"license:mit",
"endpoints_compatible",
"region:us"
] |
question-answering
| 2022-12-06T00:58:39Z |
---
license: mit
tags:
- generated_from_trainer
model-index:
- name: bert-base-historic-english-cased-squad-en
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# bert-base-historic-english-cased-squad-en
This model is a fine-tuned version of [dbmdz/bert-base-historic-english-cased](https://huggingface.co/dbmdz/bert-base-historic-english-cased) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.7739
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.2
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 2.2943 | 1.0 | 4686 | 1.9503 |
| 2.0811 | 2.0 | 9372 | 1.7739 |
### Framework versions
- Transformers 4.24.0
- Pytorch 1.13.0+cu117
- Datasets 2.7.1
- Tokenizers 0.13.2
|
timaos/distilbert-base-uncased-finetuned-cola
|
timaos
| 2022-12-06T02:41:15Z | 3 | 0 |
transformers
|
[
"transformers",
"tf",
"tensorboard",
"distilbert",
"text-classification",
"generated_from_keras_callback",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-06T02:20:46Z |
---
license: apache-2.0
tags:
- generated_from_keras_callback
model-index:
- name: timaos/distilbert-base-uncased-finetuned-cola
results: []
---
<!-- This model card has been generated automatically according to the information Keras had access to. You should
probably proofread and complete it, then remove this comment. -->
# timaos/distilbert-base-uncased-finetuned-cola
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on an unknown dataset.
It achieves the following results on the evaluation set:
- Train Loss: 0.1915
- Validation Loss: 0.5237
- Train Matthews Correlation: 0.5123
- Epoch: 2
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- optimizer: {'name': 'Adam', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 1602, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False}
- training_precision: float32
### Training results
| Train Loss | Validation Loss | Train Matthews Correlation | Epoch |
|:----------:|:---------------:|:--------------------------:|:-----:|
| 0.5210 | 0.4500 | 0.5041 | 0 |
| 0.3169 | 0.4527 | 0.5280 | 1 |
| 0.1915 | 0.5237 | 0.5123 | 2 |
### Framework versions
- Transformers 4.24.0
- TensorFlow 2.10.0
- Datasets 2.5.2
- Tokenizers 0.13.2
|
daripaez/ppo-LunarLander-v2
|
daripaez
| 2022-12-06T02:22:02Z | 1 | 0 |
stable-baselines3
|
[
"stable-baselines3",
"LunarLander-v2",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2022-12-06T02:21:39Z |
---
library_name: stable-baselines3
tags:
- LunarLander-v2
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: PPO
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: LunarLander-v2
type: LunarLander-v2
metrics:
- type: mean_reward
value: 254.94 +/- 20.69
name: mean_reward
verified: false
---
# **PPO** Agent playing **LunarLander-v2**
This is a trained model of a **PPO** agent playing **LunarLander-v2**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3).
## Usage (with Stable-baselines3)
TODO: Add your code
```python
from stable_baselines3 import ...
from huggingface_sb3 import load_from_hub
...
```
|
supermy/jinyong-gpt2
|
supermy
| 2022-12-06T02:13:48Z | 419 | 1 |
transformers
|
[
"transformers",
"pytorch",
"gpt2",
"text-generation",
"zh",
"dataset:jinyong",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2022-12-02T23:50:36Z |
---
language: zh
datasets: jinyong
inference:
parameters:
max_length: 108
num_return_sequences: 1
do_sample: True
widget:
- text: "杨过朗声说道：今番良晤，豪兴不浅，他日江湖相逢，再当杯酒言欢。咱们就此别过。 -"
example_title: "神雕侠侣"
- text: "乱世之际，人不如狗。 -"
example_title: "射雕英雄传"
---
# 飞雪连天射白鹿，笑书神侠倚碧鸳
## Model description
AI生成金庸小说，给出开头续写。
## How to use
使用 pipeline 调用模型:
```python
>>> # 调用微调后的模型
>>> senc="这些雪花落下来,多么白,多么好看.过几天太阳出来,每一片 雪花都变得无影无踪.到得明年冬天,又有许很多多雪花,只不过已不是 今年这些雪花罢了。"
>>> model_id="jinyong-gpt2-finetuning"
>>> from transformers import AutoTokenizer, GPT2LMHeadModel, TextGenerationPipeline
>>> tokenizer = AutoTokenizer.from_pretrained(model_id)
>>> model = GPT2LMHeadModel.from_pretrained(model_id)
>>> text_generator = TextGenerationPipeline(model, tokenizer)
>>> text_generator.model.config.pad_token_id = text_generator.model.config.eos_token_id
>>> text_generator( senc,max_length=108, do_sample=True)
[{'generated_text': '这些雪花落下来,多么白,多么好看.过几天太阳出来,每一片 雪花都变得无影无踪.到得明年冬天,又有许很多多雪花,只不过已不是 今年这些雪花罢了。 反正 老天爷 有眼 ， 不知 哪里 是甚么 风 险 ？” 正 说到此处 ， 突然 听得 谢逊 啸声 渐近 ， 忍不住 张口 惊呼 ， 一齐 向他 扑去 ， 只听 谢逊 一声 怒吼 ， 跟着 左手 用力 拍 出一掌 ， 以 掌力 化开 。 众人 吃了一惊 ， 同时 从 海 道 中 跃出 ， 双双 倒退 。 张翠山和殷素素 对望一眼 ， 均想 以 这两 大高手 之力 如何 抵挡 ， 以 今日 之力 如何 攻敌 之'}]
>>>
```
Here is how to use this model to get the features of a given text in PyTorch:
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("supermy/jinyong-gpt2")
model = AutoModelForCausalLM.from_pretrained("supermy/jinyong-gpt2")
```
## Training data
此数据集基于金庸的【飞雪连天射白鹿，笑书神侠倚碧鸳】小说集训练。
## 统计信息
```
```
## Training procedure
基于模型：[GPT2](https://huggingface.co/gpt2)
训练环境：英伟达16G显卡
bpe分词："vocab_size"=30000
```
[INFO|trainer.py:1608] 2022-12-02 19:52:59,024 >> ***** Running training *****
[INFO|trainer.py:1609] 2022-12-02 19:52:59,024 >> Num examples = 9443
[INFO|trainer.py:1610] 2022-12-02 19:52:59,024 >> Num Epochs = 108
[INFO|trainer.py:1611] 2022-12-02 19:52:59,024 >> Instantaneous batch size per device = 12
[INFO|trainer.py:1612] 2022-12-02 19:52:59,024 >> Total train batch size (w. parallel, distributed & accumulation) = 12
[INFO|trainer.py:1613] 2022-12-02 19:52:59,024 >> Gradient Accumulation steps = 1
[INFO|trainer.py:1614] 2022-12-02 19:52:59,024 >> Total optimization steps = 84996
[INFO|trainer.py:1616] 2022-12-02 19:52:59,025 >> Number of trainable parameters = 124439808
[INFO|trainer.py:1608] 2022-12-03 21:44:00,182 >> ***** Running training *****
[INFO|trainer.py:1609] 2022-12-03 21:44:00,182 >> Num examples = 9443
[INFO|trainer.py:1610] 2022-12-03 21:44:00,182 >> Num Epochs = 216
[INFO|trainer.py:1611] 2022-12-03 21:44:00,182 >> Instantaneous batch size per device = 12
[INFO|trainer.py:1612] 2022-12-03 21:44:00,182 >> Total train batch size (w. parallel, distributed & accumulation) = 12
[INFO|trainer.py:1613] 2022-12-03 21:44:00,182 >> Gradient Accumulation steps = 1
[INFO|trainer.py:1614] 2022-12-03 21:44:00,182 >> Total optimization steps = 169992
[INFO|trainer.py:1616] 2022-12-03 21:44:00,183 >> Number of trainable parameters = 124439808
[INFO|trainer.py:1637] 2022-12-03 21:44:00,184 >> Continuing training from checkpoint, will skip to saved global_step
[INFO|trainer.py:1638] 2022-12-03 21:44:00,184 >> Continuing training from epoch 107
[INFO|trainer.py:1639] 2022-12-03 21:44:00,184 >> Continuing training from global step 84500
[INFO|trainer.py:1608] 2022-12-05 07:36:13,626 >> ***** Running training *****
[INFO|trainer.py:1609] 2022-12-05 07:36:13,626 >> Num examples = 9443
[INFO|trainer.py:1610] 2022-12-05 07:36:13,626 >> Num Epochs = 368
[INFO|trainer.py:1611] 2022-12-05 07:36:13,626 >> Instantaneous batch size per device = 12
[INFO|trainer.py:1612] 2022-12-05 07:36:13,626 >> Total train batch size (w. parallel, distributed & accumulation) = 12
[INFO|trainer.py:1613] 2022-12-05 07:36:13,626 >> Gradient Accumulation steps = 1
[INFO|trainer.py:1614] 2022-12-05 07:36:13,626 >> Total optimization steps = 289616
[INFO|trainer.py:1616] 2022-12-05 07:36:13,627 >> Number of trainable parameters = 124439808
[INFO|trainer.py:1637] 2022-12-05 07:36:13,628 >> Continuing training from checkpoint, will skip to saved global_step
[INFO|trainer.py:1638] 2022-12-05 07:36:13,628 >> Continuing training from epoch 255
[INFO|trainer.py:1639] 2022-12-05 07:36:13,628 >> Continuing training from global step 201000
{'loss': 8.0431, 'learning_rate': 4.970998635229893e-05, 'epoch': 0.64}
{'loss': 7.4867, 'learning_rate': 4.94158548637583e-05, 'epoch': 1.27}
{'loss': 7.322, 'learning_rate': 4.912172337521766e-05, 'epoch': 1.91}
......
{'loss': 3.901, 'learning_rate': 2.5010882865076008e-05, 'epoch': 108.01}
{'loss': 3.8959, 'learning_rate': 2.4863817120805686e-05, 'epoch': 108.64}
......
{'loss': 3.1625, 'learning_rate': 4.6090404254317857e-07, 'epoch': 214.1}
{'loss': 3.1592, 'learning_rate': 3.1413242976140055e-07, 'epoch': 214.74}
{'loss': 3.1625, 'learning_rate': 1.6706668549108195e-07, 'epoch': 215.37}
{'train_runtime': 72271.9602, 'train_samples_per_second': 28.222, 'train_steps_per_second': 2.352, 'train_loss': 1.7180436183842016, 'epoch': 216.0}
{'loss': 2.7087, 'learning_rate': 4.2642671675598036e-08, 'epoch': 367.85}
{'train_runtime': 74859.0808, 'train_samples_per_second': 46.421, 'train_steps_per_second': 3.869, 'train_loss': 0.8725239146935282, 'epoch': 368.0}
***** train metrics *****
epoch = 368.0
train_loss = 0.8725
train_runtime = 20:47:39.08
train_samples = 9443
train_samples_per_second = 46.421
train_steps_per_second = 3.869
12/06/2022 04:23:55 - INFO - __main__ - *** Evaluate ***
[INFO|trainer.py:2929] 2022-12-06 04:23:55,953 >> ***** Running Evaluation *****
[INFO|trainer.py:2931] 2022-12-06 04:23:55,953 >> Num examples = 283
[INFO|trainer.py:2934] 2022-12-06 04:23:55,954 >> Batch size = 12
100%|██████████| 24/24 [00:07<00:00, 3.20it/s]
[INFO|modelcard.py:449] 2022-12-06 04:24:04,760 >> Dropping the following result as it does not have all the necessary fields:
{'task': {'name': 'Causal Language Modeling', 'type': 'text-generation'}, 'metrics': [{'name': 'Accuracy', 'type': 'accuracy', 'value': 0.19599206157122803}]}
***** eval metrics *****
epoch = 368.0
eval_accuracy = 0.196
eval_loss = 7.9524
eval_runtime = 0:00:07.87
eval_samples = 283
eval_samples_per_second = 35.94
eval_steps_per_second = 3.048
perplexity = 2842.2766
```
|
Murple/wav2vec2-base-4k
|
Murple
| 2022-12-06T01:52:44Z | 1 | 0 |
transformers
|
[
"transformers",
"pytorch",
"wav2vec2",
"pretraining",
"speech",
"multilingual",
"dataset:librispeech_asr",
"dataset:Murple/ksponspeech",
"dataset:Murple/csj",
"dataset:Murple/mmcrsc",
"arxiv:2006.11477",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null | 2022-12-06T01:36:44Z |
---
language: multilingual
datasets:
- librispeech_asr
- Murple/ksponspeech
- Murple/csj
- Murple/mmcrsc
tags:
- speech
license: apache-2.0
---
# Wav2Vec2-Base
[Facebook's Wav2Vec2](https://ai.facebook.com/blog/wav2vec-20-learning-the-structure-of-speech-from-raw-audio/)
The base model pretrained on 16kHz sampled speech audio. When using the model make sure that your speech input is also sampled at 16Khz.
**Note**: This model does not have a tokenizer as it was pretrained on audio alone. In order to use this model **speech recognition**, a tokenizer should be created and the model should be fine-tuned on labeled text data. Check out [this blog](https://huggingface.co/blog/fine-tune-wav2vec2-english) for more in-detail explanation of how to fine-tune the model.
[Paper](https://arxiv.org/abs/2006.11477)
Authors: Alexei Baevski, Henry Zhou, Abdelrahman Mohamed, Michael Auli
**Abstract**
We show for the first time that learning powerful representations from speech audio alone followed by fine-tuning on transcribed speech can outperform the best semi-supervised methods while being conceptually simpler. wav2vec 2.0 masks the speech input in the latent space and solves a contrastive task defined over a quantization of the latent representations which are jointly learned. Experiments using all labeled data of Librispeech achieve 1.8/3.3 WER on the clean/other test sets. When lowering the amount of labeled data to one hour, wav2vec 2.0 outperforms the previous state of the art on the 100 hour subset while using 100 times less labeled data. Using just ten minutes of labeled data and pre-training on 53k hours of unlabeled data still achieves 4.8/8.2 WER. This demonstrates the feasibility of speech recognition with limited amounts of labeled data.
The original model can be found under https://github.com/pytorch/fairseq/tree/master/examples/wav2vec#wav2vec-20.
# Usage
See [this notebook](https://colab.research.google.com/drive/1FjTsqbYKphl9kL-eILgUc-bl4zVThL8F?usp=sharing) for more information on how to fine-tune the model.
The results can be seen [here](https://wandb.ai/toraruka623/wav2vec2-pretraining/reports/Wav2Vec2-base-4k--VmlldzozMDkxMDk3?accessToken=lfn2kwe9pzmvdonhx7hihd9nf13wzby7odu0iakdubwep3le4ywirxc3gx9w66fi)
|
saphvis/px-mixedbag
|
saphvis
| 2022-12-06T01:52:25Z | 0 | 0 | null |
[
"license:creativeml-openrail-m",
"region:us"
] | null | 2022-12-06T01:47:43Z |
---
license: creativeml-openrail-m
---
|
jnick/ppo-LunarLander-v2
|
jnick
| 2022-12-06T01:39:23Z | 1 | 0 |
stable-baselines3
|
[
"stable-baselines3",
"LunarLander-v2",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2022-12-06T01:38:49Z |
---
library_name: stable-baselines3
tags:
- LunarLander-v2
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: PPO
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: LunarLander-v2
type: LunarLander-v2
metrics:
- type: mean_reward
value: 267.30 +/- 18.25
name: mean_reward
verified: false
---
# **PPO** Agent playing **LunarLander-v2**
This is a trained model of a **PPO** agent playing **LunarLander-v2**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3).
## Usage (with Stable-baselines3)
TODO: Add your code
```python
from stable_baselines3 import ...
from huggingface_sb3 import load_from_hub
...
```
|
Nadav/bert-base-historic-multilingual-64k-td-cased-squad-en
|
Nadav
| 2022-12-06T01:04:49Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"bert",
"question-answering",
"generated_from_trainer",
"license:mit",
"endpoints_compatible",
"region:us"
] |
question-answering
| 2022-12-05T23:08:03Z |
---
license: mit
tags:
- generated_from_trainer
model-index:
- name: bert-base-historic-multilingual-64k-td-cased-squad-en
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# bert-base-historic-multilingual-64k-td-cased-squad-en
This model is a fine-tuned version of [dbmdz/bert-base-historic-multilingual-64k-td-cased](https://huggingface.co/dbmdz/bert-base-historic-multilingual-64k-td-cased) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.5474
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.2
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 1.9315 | 1.0 | 4659 | 1.7399 |
| 1.5775 | 2.0 | 9318 | 1.5474 |
### Framework versions
- Transformers 4.24.0
- Pytorch 1.13.0+cu117
- Datasets 2.7.1
- Tokenizers 0.13.2
|
sree2910/tonality
|
sree2910
| 2022-12-06T00:57:13Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"distilbert",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-05T15:40:07Z |
---
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: tonality
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# tonality
This model is a fine-tuned version of [distilbert-base-cased](https://huggingface.co/distilbert-base-cased) on an unknown dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
### Framework versions
- Transformers 4.24.0
- Pytorch 1.13.0+cpu
- Datasets 2.7.1
- Tokenizers 0.13.2
|
Nadav/bert-base-historic-multilingual-cased-squad-en
|
Nadav
| 2022-12-06T00:54:11Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"bert",
"question-answering",
"generated_from_trainer",
"license:mit",
"endpoints_compatible",
"region:us"
] |
question-answering
| 2022-12-05T22:51:39Z |
---
license: mit
tags:
- generated_from_trainer
model-index:
- name: bert-base-historic-multilingual-cased-squad-en
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# bert-base-historic-multilingual-cased-squad-en
This model is a fine-tuned version of [dbmdz/bert-base-historic-multilingual-cased](https://huggingface.co/dbmdz/bert-base-historic-multilingual-cased) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.5307
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.2
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 1.881 | 1.0 | 4820 | 1.5507 |
| 1.5883 | 2.0 | 9640 | 1.5307 |
### Framework versions
- Transformers 4.24.0
- Pytorch 1.13.0+cu117
- Datasets 2.7.1
- Tokenizers 0.13.2
|
flamesbob/dpin-model
|
flamesbob
| 2022-12-06T00:15:36Z | 0 | 0 | null |
[
"license:creativeml-openrail-m",
"region:us"
] | null | 2022-12-04T21:39:01Z |
---
license: creativeml-openrail-m
---
|
fathyshalab/all-roberta-large-v1-travel-4-16-5-oos
|
fathyshalab
| 2022-12-06T00:14:15Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-05T23:50:51Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: all-roberta-large-v1-travel-4-16-5-oos
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# all-roberta-large-v1-travel-4-16-5-oos
This model is a fine-tuned version of [sentence-transformers/all-roberta-large-v1](https://huggingface.co/sentence-transformers/all-roberta-large-v1) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.1384
- Accuracy: 0.4289
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 2.7625 | 1.0 | 1 | 2.5258 | 0.2933 |
| 2.0955 | 2.0 | 2 | 2.3775 | 0.3333 |
| 1.7076 | 3.0 | 3 | 2.2590 | 0.38 |
| 1.3257 | 4.0 | 4 | 2.1788 | 0.4089 |
| 1.1109 | 5.0 | 5 | 2.1384 | 0.4289 |
### Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
|
delmaksym/ppo-Huggy
|
delmaksym
| 2022-12-05T23:51:10Z | 4 | 0 |
ml-agents
|
[
"ml-agents",
"tensorboard",
"onnx",
"unity-ml-agents",
"deep-reinforcement-learning",
"reinforcement-learning",
"ML-Agents-Huggy",
"region:us"
] |
reinforcement-learning
| 2022-12-05T23:51:04Z |
---
tags:
- unity-ml-agents
- ml-agents
- deep-reinforcement-learning
- reinforcement-learning
- ML-Agents-Huggy
library_name: ml-agents
---
# **ppo** Agent playing **Huggy**
This is a trained model of a **ppo** agent playing **Huggy** using the [Unity ML-Agents Library](https://github.com/Unity-Technologies/ml-agents).
## Usage (with ML-Agents)
The Documentation: https://github.com/huggingface/ml-agents#get-started
We wrote a complete tutorial to learn to train your first agent using ML-Agents and publish it to the Hub:
### Resume the training
```
mlagents-learn <your_configuration_file_path.yaml> --run-id=<run_id> --resume
```
### Watch your Agent play
You can watch your agent **playing directly in your browser:**.
1. Go to https://huggingface.co/spaces/unity/ML-Agents-Huggy
2. Step 1: Write your model_id: delmaksym/ppo-Huggy
3. Step 2: Select your *.nn /*.onnx file
4. Click on Watch the agent play 👀
|
fathyshalab/all-roberta-large-v1-travel-2-16-5-oos
|
fathyshalab
| 2022-12-05T23:50:30Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-05T23:33:19Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: all-roberta-large-v1-travel-2-16-5-oos
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# all-roberta-large-v1-travel-2-16-5-oos
This model is a fine-tuned version of [sentence-transformers/all-roberta-large-v1](https://huggingface.co/sentence-transformers/all-roberta-large-v1) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.1384
- Accuracy: 0.4289
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 2.7625 | 1.0 | 1 | 2.5258 | 0.2933 |
| 2.0955 | 2.0 | 2 | 2.3775 | 0.3333 |
| 1.7076 | 3.0 | 3 | 2.2590 | 0.38 |
| 1.3257 | 4.0 | 4 | 2.1788 | 0.4089 |
| 1.1109 | 5.0 | 5 | 2.1384 | 0.4289 |
### Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
|
famube/autotrain-ciap2-2347173866
|
famube
| 2022-12-05T23:32:01Z | 2 | 0 |
transformers
|
[
"transformers",
"pytorch",
"autotrain",
"text-classification",
"pt",
"dataset:famube/autotrain-data-ciap2",
"co2_eq_emissions",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-05T22:49:37Z |
---
tags:
- autotrain
- text-classification
language:
- pt
widget:
- text: "febre"
- text: "dor de cabeça"
- text: "corpo inteiro doendo"
datasets:
- famube/autotrain-data-ciap2
co2_eq_emissions:
emissions: 4.825567476024859
---
# Model Trained Using AutoTrain
- Problem type: Multi-class Classification
- Model ID: 2347173866
- CO2 Emissions (in grams): 4.8256
## Validation Metrics
- Loss: 1.932
- Accuracy: 0.681
- Macro F1: 0.609
- Micro F1: 0.681
- Weighted F1: 0.622
- Macro Precision: 0.592
- Micro Precision: 0.681
- Weighted Precision: 0.610
- Macro Recall: 0.669
- Micro Recall: 0.681
- Weighted Recall: 0.681
## Usage
You can use cURL to access this model:
```
$ curl -X POST -H "Authorization: Bearer YOUR_API_KEY" -H "Content-Type: application/json" -d '{"inputs": "I love AutoTrain"}' https://api-inference.huggingface.co/models/famube/autotrain-ciap2-2347173866
```
Or Python API:
```
from transformers import AutoModelForSequenceClassification, AutoTokenizer
model = AutoModelForSequenceClassification.from_pretrained("famube/autotrain-ciap2-2347173866", use_auth_token=True)
tokenizer = AutoTokenizer.from_pretrained("famube/autotrain-ciap2-2347173866", use_auth_token=True)
inputs = tokenizer("I love AutoTrain", return_tensors="pt")
outputs = model(**inputs)
```
|
fathyshalab/all-roberta-large-v1-auto_and_commute-16-16-5-oos
|
fathyshalab
| 2022-12-05T23:01:50Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-05T22:36:13Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: all-roberta-large-v1-auto_and_commute-16-16-5-oos
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# all-roberta-large-v1-auto_and_commute-16-16-5-oos
This model is a fine-tuned version of [sentence-transformers/all-roberta-large-v1](https://huggingface.co/sentence-transformers/all-roberta-large-v1) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.2614
- Accuracy: 0.4289
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 2.7929 | 1.0 | 1 | 2.5690 | 0.2667 |
| 2.267 | 2.0 | 2 | 2.4558 | 0.3533 |
| 1.8495 | 3.0 | 3 | 2.3630 | 0.3911 |
| 1.4397 | 4.0 | 4 | 2.2956 | 0.4133 |
| 1.2985 | 5.0 | 5 | 2.2614 | 0.4289 |
### Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
|
Nadav/MacBERTh-squad-en
|
Nadav
| 2022-12-05T22:47:54Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"bert",
"question-answering",
"generated_from_trainer",
"endpoints_compatible",
"region:us"
] |
question-answering
| 2022-12-05T20:37:59Z |
---
tags:
- generated_from_trainer
model-index:
- name: MacBERTh-squad-en
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# MacBERTh-squad-en
This model is a fine-tuned version of [emanjavacas/MacBERTh](https://huggingface.co/emanjavacas/MacBERTh) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 2.1805
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.2
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:-----:|:---------------:|
| 2.5789 | 1.0 | 5110 | 2.3494 |
| 2.2681 | 2.0 | 10220 | 2.1805 |
### Framework versions
- Transformers 4.24.0
- Pytorch 1.13.0+cu117
- Datasets 2.7.1
- Tokenizers 0.13.2
|
sd-concepts-library/vie-proceres
|
sd-concepts-library
| 2022-12-05T22:41:59Z | 0 | 0 | null |
[
"license:mit",
"region:us"
] | null | 2022-12-05T22:41:55Z |
---
license: mit
---
### vie-proceres on Stable Diffusion
This is the `<vie-proceres>` concept taught to Stable Diffusion via Textual Inversion. You can load this concept into the [Stable Conceptualizer](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/stable_conceptualizer_inference.ipynb) notebook. You can also train your own concepts and load them into the concept libraries using [this notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/sd_textual_inversion_training.ipynb).
Here is the new concept you will be able to use as a `style`:
|
alanila/autotrain-acc_keys-2347073860
|
alanila
| 2022-12-05T22:34:34Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"bert",
"fill-mask",
"autotrain",
"text-classification",
"unk",
"dataset:alanila/autotrain-data-acc_keys",
"co2_eq_emissions",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-05T22:27:11Z |
---
tags:
- autotrain
- text-classification
language:
- unk
widget:
- text: "I love AutoTrain 🤗"
datasets:
- alanila/autotrain-data-acc_keys
co2_eq_emissions:
emissions: 1.3599341780747405
---
# Model Trained Using AutoTrain
- Problem type: Multi-class Classification
- Model ID: 2347073860
- CO2 Emissions (in grams): 1.3599
## Validation Metrics
- Loss: 1.255
- Accuracy: 0.500
- Macro F1: 0.445
- Micro F1: 0.500
- Weighted F1: 0.421
- Macro Precision: 0.498
- Micro Precision: 0.500
- Weighted Precision: 0.508
- Macro Recall: 0.481
- Micro Recall: 0.500
- Weighted Recall: 0.500
## Usage
You can use cURL to access this model:
```
$ curl -X POST -H "Authorization: Bearer YOUR_API_KEY" -H "Content-Type: application/json" -d '{"inputs": "I love AutoTrain"}' https://api-inference.huggingface.co/models/alanila/autotrain-acc_keys-2347073860
```
Or Python API:
```
from transformers import AutoModelForSequenceClassification, AutoTokenizer
model = AutoModelForSequenceClassification.from_pretrained("alanila/autotrain-acc_keys-2347073860", use_auth_token=True)
tokenizer = AutoTokenizer.from_pretrained("alanila/autotrain-acc_keys-2347073860", use_auth_token=True)
inputs = tokenizer("I love AutoTrain", return_tensors="pt")
outputs = model(**inputs)
```
|
zzmez/ppo-LunarLander-v2
|
zzmez
| 2022-12-05T22:29:25Z | 0 | 0 |
stable-baselines3
|
[
"stable-baselines3",
"LunarLander-v2",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2022-12-05T21:43:32Z |
---
library_name: stable-baselines3
tags:
- LunarLander-v2
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: PPO
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: LunarLander-v2
type: LunarLander-v2
metrics:
- type: mean_reward
value: 260.80 +/- 24.69
name: mean_reward
verified: false
---
# **PPO** Agent playing **LunarLander-v2**
This is a trained model of a **PPO** agent playing **LunarLander-v2**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3).
## Usage (with Stable-baselines3)
TODO: Add your code
```python
from stable_baselines3 import ...
from huggingface_sb3 import load_from_hub
...
```
|
Nadav/bert-base-historic-multilingual-64k-td-cased-squad-nl
|
Nadav
| 2022-12-05T22:28:57Z | 10 | 0 |
transformers
|
[
"transformers",
"pytorch",
"bert",
"question-answering",
"generated_from_trainer",
"license:mit",
"endpoints_compatible",
"region:us"
] |
question-answering
| 2022-12-05T20:30:03Z |
---
license: mit
tags:
- generated_from_trainer
model-index:
- name: bert-base-historic-multilingual-64k-td-cased-squad-nl
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# bert-base-historic-multilingual-64k-td-cased-squad-nl
This model is a fine-tuned version of [dbmdz/bert-base-historic-multilingual-64k-td-cased](https://huggingface.co/dbmdz/bert-base-historic-multilingual-64k-td-cased) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.6382
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.2
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 2.0101 | 1.0 | 4659 | 1.8679 |
| 1.6528 | 2.0 | 9318 | 1.6382 |
### Framework versions
- Transformers 4.24.0
- Pytorch 1.13.0+cu117
- Datasets 2.7.1
- Tokenizers 0.13.2
|
fathyshalab/all-roberta-large-v1-auto_and_commute-4-16-5-oos
|
fathyshalab
| 2022-12-05T22:11:24Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-05T21:47:30Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: all-roberta-large-v1-auto_and_commute-4-16-5-oos
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# all-roberta-large-v1-auto_and_commute-4-16-5-oos
This model is a fine-tuned version of [sentence-transformers/all-roberta-large-v1](https://huggingface.co/sentence-transformers/all-roberta-large-v1) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.2614
- Accuracy: 0.4289
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 2.7929 | 1.0 | 1 | 2.5690 | 0.2667 |
| 2.267 | 2.0 | 2 | 2.4558 | 0.3533 |
| 1.8495 | 3.0 | 3 | 2.3630 | 0.3911 |
| 1.4397 | 4.0 | 4 | 2.2956 | 0.4133 |
| 1.2985 | 5.0 | 5 | 2.2614 | 0.4289 |
### Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
|
hannoh/05_model_sales_external_imbalanced
|
hannoh
| 2022-12-05T22:00:24Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"distilbert",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-05T21:34:00Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: 05_model_sales_external_imbalanced
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# 05_model_sales_external_imbalanced
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2421
- Accuracy: 0.9294
- F1: 0.3654
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
### Framework versions
- Transformers 4.25.1
- Pytorch 1.12.1+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2
|
huggingtweets/jellynatelli-raspberryl0ver
|
huggingtweets
| 2022-12-05T21:59:30Z | 4 | 0 |
transformers
|
[
"transformers",
"pytorch",
"gpt2",
"text-generation",
"huggingtweets",
"en",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2022-12-05T21:59:22Z |
---
language: en
thumbnail: https://github.com/borisdayma/huggingtweets/blob/master/img/logo.png?raw=true
tags:
- huggingtweets
widget:
- text: "My dream is"
---
<div class="inline-flex flex-col" style="line-height: 1.5;">
<div class="flex">
<div
style="display:inherit; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('https://pbs.twimg.com/profile_images/1552729971956727808/zVaFH3ex_400x400.jpg')">
</div>
<div
style="display:inherit; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('https://pbs.twimg.com/profile_images/1583521884590772232/DGBIkzGk_400x400.jpg')">
</div>
<div
style="display:none; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('')">
</div>
</div>
<div style="text-align: center; margin-top: 3px; font-size: 16px; font-weight: 800">🤖 AI CYBORG 🤖</div>
<div style="text-align: center; font-size: 16px; font-weight: 800">🌞 & 9</div>
<div style="text-align: center; font-size: 14px;">@jellynatelli-raspberryl0ver</div>
</div>
I was made with [huggingtweets](https://github.com/borisdayma/huggingtweets).
Create your own bot based on your favorite user with [the demo](https://colab.research.google.com/github/borisdayma/huggingtweets/blob/master/huggingtweets-demo.ipynb)!
## How does it work?
The model uses the following pipeline.
To understand how the model was developed, check the [W&B report](https://wandb.ai/wandb/huggingtweets/reports/HuggingTweets-Train-a-Model-to-Generate-Tweets--VmlldzoxMTY5MjI).
## Training data
The model was trained on tweets from 🌞 & 9.
| Data | 🌞 | 9 |
| --- | --- | --- |
| Tweets downloaded | 1797 | 3205 |
| Retweets | 413 | 202 |
| Short tweets | 206 | 633 |
| Tweets kept | 1178 | 2370 |
[Explore the data](https://wandb.ai/wandb/huggingtweets/runs/2nlgvuz7/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
## Training procedure
The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on @jellynatelli-raspberryl0ver's tweets.
Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/wandb/huggingtweets/runs/1pu0nfyz) for full transparency and reproducibility.
At the end of training, [the final model](https://wandb.ai/wandb/huggingtweets/runs/1pu0nfyz/artifacts) is logged and versioned.
## How to use
You can use this model directly with a pipeline for text generation:
```python
from transformers import pipeline
generator = pipeline('text-generation',
model='huggingtweets/jellynatelli-raspberryl0ver')
generator("My dream is", num_return_sequences=5)
```
## Limitations and bias
The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias).
In addition, the data present in the user's tweets further affects the text generated by the model.
## About
*Built by Boris Dayma*
[](https://twitter.com/intent/follow?screen_name=borisdayma)
For more details, visit the project repository.
[](https://github.com/borisdayma/huggingtweets)
|
fathyshalab/all-roberta-large-v1-auto_and_commute-2-16-5-oos
|
fathyshalab
| 2022-12-05T21:47:10Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-05T21:20:02Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: all-roberta-large-v1-auto_and_commute-2-16-5-oos
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# all-roberta-large-v1-auto_and_commute-2-16-5-oos
This model is a fine-tuned version of [sentence-transformers/all-roberta-large-v1](https://huggingface.co/sentence-transformers/all-roberta-large-v1) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.2614
- Accuracy: 0.4289
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 2.7929 | 1.0 | 1 | 2.5690 | 0.2667 |
| 2.267 | 2.0 | 2 | 2.4558 | 0.3533 |
| 1.8495 | 3.0 | 3 | 2.3630 | 0.3911 |
| 1.4397 | 4.0 | 4 | 2.2956 | 0.4133 |
| 1.2985 | 5.0 | 5 | 2.2614 | 0.4289 |
### Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
|
gemasphi/setfit-ss-paraphrase-multilingual-mpnet-base-v2
|
gemasphi
| 2022-12-05T21:45:41Z | 5 | 0 |
sentence-transformers
|
[
"sentence-transformers",
"pytorch",
"xlm-roberta",
"feature-extraction",
"sentence-similarity",
"transformers",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] |
sentence-similarity
| 2022-12-05T21:45:17Z |
---
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- transformers
---
# {MODEL_NAME}
This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search.
<!--- Describe your model here -->
## Usage (Sentence-Transformers)
Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed:
```
pip install -U sentence-transformers
```
Then you can use the model like this:
```python
from sentence_transformers import SentenceTransformer
sentences = ["This is an example sentence", "Each sentence is converted"]
model = SentenceTransformer('{MODEL_NAME}')
embeddings = model.encode(sentences)
print(embeddings)
```
## Usage (HuggingFace Transformers)
Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings.
```python
from transformers import AutoTokenizer, AutoModel
import torch
#Mean Pooling - Take attention mask into account for correct averaging
def mean_pooling(model_output, attention_mask):
token_embeddings = model_output[0] #First element of model_output contains all token embeddings
input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)
# Sentences we want sentence embeddings for
sentences = ['This is an example sentence', 'Each sentence is converted']
# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}')
model = AutoModel.from_pretrained('{MODEL_NAME}')
# Tokenize sentences
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
# Compute token embeddings
with torch.no_grad():
model_output = model(**encoded_input)
# Perform pooling. In this case, mean pooling.
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
print("Sentence embeddings:")
print(sentence_embeddings)
```
## Evaluation Results
<!--- Describe how your model was evaluated -->
For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME})
## Training
The model was trained with the parameters:
**DataLoader**:
`torch.utils.data.dataloader.DataLoader` of length 1320 with parameters:
```
{'batch_size': 32, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}
```
**Loss**:
`sentence_transformers.losses.CosineSimilarityLoss.CosineSimilarityLoss`
Parameters of the fit()-Method:
```
{
"epochs": 1,
"evaluation_steps": 0,
"evaluator": "NoneType",
"max_grad_norm": 1,
"optimizer_class": "<class 'torch.optim.adamw.AdamW'>",
"optimizer_params": {
"lr": 2e-05
},
"scheduler": "WarmupLinear",
"steps_per_epoch": 1320,
"warmup_steps": 132,
"weight_decay": 0.01
}
```
## Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: XLMRobertaModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
)
```
## Citing & Authors
<!--- Describe where people can find more information -->
|
gemasphi/mcontriever-msmarco
|
gemasphi
| 2022-12-05T21:01:55Z | 3 | 0 |
sentence-transformers
|
[
"sentence-transformers",
"pytorch",
"bert",
"feature-extraction",
"sentence-similarity",
"transformers",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] |
sentence-similarity
| 2022-12-05T21:01:38Z |
---
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- transformers
---
# {MODEL_NAME}
This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search.
<!--- Describe your model here -->
## Usage (Sentence-Transformers)
Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed:
```
pip install -U sentence-transformers
```
Then you can use the model like this:
```python
from sentence_transformers import SentenceTransformer
sentences = ["This is an example sentence", "Each sentence is converted"]
model = SentenceTransformer('{MODEL_NAME}')
embeddings = model.encode(sentences)
print(embeddings)
```
## Usage (HuggingFace Transformers)
Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings.
```python
from transformers import AutoTokenizer, AutoModel
import torch
#Mean Pooling - Take attention mask into account for correct averaging
def mean_pooling(model_output, attention_mask):
token_embeddings = model_output[0] #First element of model_output contains all token embeddings
input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)
# Sentences we want sentence embeddings for
sentences = ['This is an example sentence', 'Each sentence is converted']
# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}')
model = AutoModel.from_pretrained('{MODEL_NAME}')
# Tokenize sentences
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
# Compute token embeddings
with torch.no_grad():
model_output = model(**encoded_input)
# Perform pooling. In this case, mean pooling.
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
print("Sentence embeddings:")
print(sentence_embeddings)
```
## Evaluation Results
<!--- Describe how your model was evaluated -->
For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME})
## Training
The model was trained with the parameters:
**DataLoader**:
`torch.utils.data.dataloader.DataLoader` of length 1320 with parameters:
```
{'batch_size': 32, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}
```
**Loss**:
`sentence_transformers.losses.CosineSimilarityLoss.CosineSimilarityLoss`
Parameters of the fit()-Method:
```
{
"epochs": 1,
"evaluation_steps": 0,
"evaluator": "NoneType",
"max_grad_norm": 1,
"optimizer_class": "<class 'torch.optim.adamw.AdamW'>",
"optimizer_params": {
"lr": 2e-05
},
"scheduler": "WarmupLinear",
"steps_per_epoch": 1320,
"warmup_steps": 132,
"weight_decay": 0.01
}
```
## Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
)
```
## Citing & Authors
<!--- Describe where people can find more information -->
|
fathyshalab/all-roberta-large-v1-home-16-16-5-oos
|
fathyshalab
| 2022-12-05T20:53:01Z | 3 | 0 |
transformers
|
[
"transformers",
"pytorch",
"roberta",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2022-12-05T17:43:17Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: all-roberta-large-v1-home-16-16-5-oos
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# all-roberta-large-v1-home-16-16-5-oos
This model is a fine-tuned version of [sentence-transformers/all-roberta-large-v1](https://huggingface.co/sentence-transformers/all-roberta-large-v1) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.3789
- Accuracy: 0.3356
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 2.7614 | 1.0 | 1 | 2.6146 | 0.1889 |
| 2.2082 | 2.0 | 2 | 2.5232 | 0.2667 |
| 1.8344 | 3.0 | 3 | 2.4516 | 0.2933 |
| 1.4601 | 4.0 | 4 | 2.4033 | 0.3267 |
| 1.2748 | 5.0 | 5 | 2.3789 | 0.3356 |
### Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
|
harryrudolph/first_model
|
harryrudolph
| 2022-12-05T20:32:51Z | 0 | 0 |
stable-baselines3
|
[
"stable-baselines3",
"LunarLander-v2",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2022-12-05T20:32:16Z |
---
library_name: stable-baselines3
tags:
- LunarLander-v2
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: PPO
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: LunarLander-v2
type: LunarLander-v2
metrics:
- type: mean_reward
value: -137.96 +/- 24.14
name: mean_reward
verified: false
---
# **PPO** Agent playing **LunarLander-v2**
This is a trained model of a **PPO** agent playing **LunarLander-v2**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3).
## Usage (with Stable-baselines3)
TODO: Add your code
```python
from stable_baselines3 import ...
from huggingface_sb3 import load_from_hub
...
```
|
gemasphi/setfit-ss-distiluse-base-multilingual-cased-v2
|
gemasphi
| 2022-12-05T20:14:47Z | 10 | 0 |
sentence-transformers
|
[
"sentence-transformers",
"pytorch",
"distilbert",
"feature-extraction",
"sentence-similarity",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] |
sentence-similarity
| 2022-12-05T20:14:30Z |
---
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
---
# {MODEL_NAME}
This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 512 dimensional dense vector space and can be used for tasks like clustering or semantic search.
<!--- Describe your model here -->
## Usage (Sentence-Transformers)
Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed:
```
pip install -U sentence-transformers
```
Then you can use the model like this:
```python
from sentence_transformers import SentenceTransformer
sentences = ["This is an example sentence", "Each sentence is converted"]
model = SentenceTransformer('{MODEL_NAME}')
embeddings = model.encode(sentences)
print(embeddings)
```
## Evaluation Results
<!--- Describe how your model was evaluated -->
For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME})
## Training
The model was trained with the parameters:
**DataLoader**:
`torch.utils.data.dataloader.DataLoader` of length 1320 with parameters:
```
{'batch_size': 32, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}
```
**Loss**:
`sentence_transformers.losses.CosineSimilarityLoss.CosineSimilarityLoss`
Parameters of the fit()-Method:
```
{
"epochs": 1,
"evaluation_steps": 0,
"evaluator": "NoneType",
"max_grad_norm": 1,
"optimizer_class": "<class 'torch.optim.adamw.AdamW'>",
"optimizer_params": {
"lr": 2e-05
},
"scheduler": "WarmupLinear",
"steps_per_epoch": 1320,
"warmup_steps": 132,
"weight_decay": 0.01
}
```
## Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: DistilBertModel
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
(2): Dense({'in_features': 768, 'out_features': 512, 'bias': True, 'activation_function': 'torch.nn.modules.activation.Tanh'})
)
```
## Citing & Authors
<!--- Describe where people can find more information -->
|
edgertej/poebert-clean-checkpoint-finetuned-poetry-foundation-clean
|
edgertej
| 2022-12-05T20:09:48Z | 3 | 0 |
transformers
|
[
"transformers",
"tf",
"bert",
"fill-mask",
"generated_from_keras_callback",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
fill-mask
| 2022-12-05T19:11:18Z |
---
license: apache-2.0
tags:
- generated_from_keras_callback
model-index:
- name: edgertej/poebert-clean-checkpoint-finetuned-poetry-foundation-clean
results: []
---
<!-- This model card has been generated automatically according to the information Keras had access to. You should
probably proofread and complete it, then remove this comment. -->
# edgertej/poebert-clean-checkpoint-finetuned-poetry-foundation-clean
This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-cased) on an unknown dataset.
It achieves the following results on the evaluation set:
- Train Loss: 3.8658
- Validation Loss: 3.6186
- Epoch: 2
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- optimizer: {'name': 'Adam', 'learning_rate': 3e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
- training_precision: float32
### Training results
| Train Loss | Validation Loss | Epoch |
|:----------:|:---------------:|:-----:|
| 4.0379 | 3.6686 | 0 |
| 3.9346 | 3.6478 | 1 |
| 3.8658 | 3.6186 | 2 |
### Framework versions
- Transformers 4.19.2
- TensorFlow 2.9.1
- Datasets 2.4.0
- Tokenizers 0.12.1
|
mdcox/distilbert-base-uncased-finetuned-ner
|
mdcox
| 2022-12-05T19:45:32Z | 12 | 0 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"distilbert",
"token-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
token-classification
| 2022-12-05T19:10:37Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: distilbert-base-uncased-finetuned-ner
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# distilbert-base-uncased-finetuned-ner
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0048
- Precision: 0.9203
- Recall: 0.9777
- F1: 0.9482
- Accuracy: 0.9984
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
### Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
| No log | 1.0 | 358 | 0.0067 | 0.9229 | 0.9332 | 0.9280 | 0.9978 |
| 0.0545 | 2.0 | 716 | 0.0052 | 0.9167 | 0.9800 | 0.9473 | 0.9984 |
| 0.0052 | 3.0 | 1074 | 0.0048 | 0.9203 | 0.9777 | 0.9482 | 0.9984 |
### Framework versions
- Transformers 4.25.1
- Pytorch 1.12.1+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2
|
nudro/sd-class-butterflies-32
|
nudro
| 2022-12-05T19:33:43Z | 3 | 0 |
diffusers
|
[
"diffusers",
"pytorch",
"unconditional-image-generation",
"diffusion-models-class",
"license:mit",
"diffusers:DDPMPipeline",
"region:us"
] |
unconditional-image-generation
| 2022-12-05T19:33:35Z |
---
license: mit
tags:
- pytorch
- diffusers
- unconditional-image-generation
- diffusion-models-class
---
# Model Card for Unit 1 of the [Diffusion Models Class 🧨](https://github.com/huggingface/diffusion-models-class)
This model is a diffusion model for unconditional image generation of cute 🦋.
## Usage
```python
from diffusers import DDPMPipeline
pipeline = DDPMPipeline.from_pretrained('nudro/sd-class-butterflies-32')
image = pipeline().images[0]
image
```
|
Subsets and Splits
Filtered Qwen2.5 Distill Models
Identifies specific configurations of models by filtering cards that contain 'distill', 'qwen2.5', '7b' while excluding certain base models and incorrect model ID patterns, uncovering unique model variants.
Filtered Model Cards Count
Finds the count of entries with specific card details that include 'distill', 'qwen2.5', '7b' but exclude certain base models, revealing valuable insights about the dataset's content distribution.
Filtered Distill Qwen 7B Models
Filters for specific card entries containing 'distill', 'qwen', and '7b', excluding certain strings and patterns, to identify relevant model configurations.
Filtered Qwen-7b Model Cards
The query performs a detailed filtering based on specific keywords and excludes certain entries, which could be useful for identifying a specific subset of cards but does not provide deeper insights or trends.
Filtered Qwen 7B Model Cards
The query filters for specific terms related to "distilled" or "distill", "qwen", and "7b" in the 'card' column but excludes certain base models, providing a limited set of entries for further inspection.
Qwen 7B Distilled Models
The query provides a basic filtering of records to find specific card names that include keywords related to distilled Qwen 7b models, excluding a particular base model, which gives limited insight but helps in focusing on relevant entries.
Qwen 7B Distilled Model Cards
The query filters data based on specific keywords in the modelId and card fields, providing limited insight primarily useful for locating specific entries rather than revealing broad patterns or trends.
Qwen 7B Distilled Models
Finds all entries containing the terms 'distilled', 'qwen', and '7b' in a case-insensitive manner, providing a filtered set of records but without deeper analysis.
Distilled Qwen 7B Models
The query filters for specific model IDs containing 'distilled', 'qwen', and '7b', providing a basic retrieval of relevant entries but without deeper analysis or insight.
Filtered Model Cards with Distill Qwen2.
Filters and retrieves records containing specific keywords in the card description while excluding certain phrases, providing a basic count of relevant entries.
Filtered Model Cards with Distill Qwen 7
The query filters specific variations of card descriptions containing 'distill', 'qwen', and '7b' while excluding a particular base model, providing limited but specific data retrieval.
Distill Qwen 7B Model Cards
The query filters and retrieves rows where the 'card' column contains specific keywords ('distill', 'qwen', and '7b'), providing a basic filter result that can help in identifying specific entries.