Search is not available for this dataset
pipeline_tag
stringclasses 48
values | library_name
stringclasses 205
values | text
stringlengths 0
18.3M
| metadata
stringlengths 2
1.07B
| id
stringlengths 5
122
| last_modified
null | tags
listlengths 1
1.84k
| sha
null | created_at
stringlengths 25
25
|
|---|---|---|---|---|---|---|---|---|
image-classification
|
transformers
|
# ConvNeXT (base-sized model)
ConvNeXT model trained on ImageNet-1k at resolution 384x384. It was introduced in the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Liu et al. and first released in [this repository](https://github.com/facebookresearch/ConvNeXt).
Disclaimer: The team releasing ConvNeXT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
ConvNeXT is a pure convolutional model (ConvNet), inspired by the design of Vision Transformers, that claims to outperform them. The authors started from a ResNet and "modernized" its design by taking the Swin Transformer as inspiration.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=convnext) to look for
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import ConvNextImageProcessor, ConvNextForImageClassification
import torch
from datasets import load_dataset
dataset = load_dataset("huggingface/cats-image")
image = dataset["test"]["image"][0]
processor = ConvNextImageProcessor.from_pretrained("facebook/convnext-base-384")
model = ConvNextForImageClassification.from_pretrained("facebook/convnext-base-384")
inputs = processor(image, return_tensors="pt")
with torch.no_grad():
logits = model(**inputs).logits
# model predicts one of the 1000 ImageNet classes
predicted_label = logits.argmax(-1).item()
print(model.config.id2label[predicted_label]),
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/convnext).
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2201-03545,
author = {Zhuang Liu and
Hanzi Mao and
Chao{-}Yuan Wu and
Christoph Feichtenhofer and
Trevor Darrell and
Saining Xie},
title = {A ConvNet for the 2020s},
journal = {CoRR},
volume = {abs/2201.03545},
year = {2022},
url = {https://arxiv.org/abs/2201.03545},
eprinttype = {arXiv},
eprint = {2201.03545},
timestamp = {Thu, 20 Jan 2022 14:21:35 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2201-03545.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["vision", "image-classification"], "datasets": ["imagenet-1k"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/tiger.jpg", "example_title": "Tiger"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/teapot.jpg", "example_title": "Teapot"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/palace.jpg", "example_title": "Palace"}]}
|
facebook/convnext-base-384
| null |
[
"transformers",
"pytorch",
"tf",
"convnext",
"image-classification",
"vision",
"dataset:imagenet-1k",
"arxiv:2201.03545",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# ConvNeXT (large-sized model)
ConvNeXT model pre-trained on ImageNet-22k and fine-tuned on ImageNet-1k at resolution 224x224. It was introduced in the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Liu et al. and first released in [this repository](https://github.com/facebookresearch/ConvNeXt).
Disclaimer: The team releasing ConvNeXT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
ConvNeXT is a pure convolutional model (ConvNet), inspired by the design of Vision Transformers, that claims to outperform them. The authors started from a ResNet and "modernized" its design by taking the Swin Transformer as inspiration.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=convnext) to look for
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import ConvNextImageProcessor, ConvNextForImageClassification
import torch
from datasets import load_dataset
dataset = load_dataset("huggingface/cats-image")
image = dataset["test"]["image"][0]
processor = ConvNextImageProcessor.from_pretrained("facebook/convnext-large-224-22k-1k")
model = ConvNextForImageClassification.from_pretrained("facebook/convnext-large-224-22k-1k")
inputs = processor(image, return_tensors="pt")
with torch.no_grad():
logits = model(**inputs).logits
# model predicts one of the 1k ImageNet classes
predicted_label = logits.argmax(-1).item()
print(model.config.id2label[predicted_label]),
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/convnext).
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2201-03545,
author = {Zhuang Liu and
Hanzi Mao and
Chao{-}Yuan Wu and
Christoph Feichtenhofer and
Trevor Darrell and
Saining Xie},
title = {A ConvNet for the 2020s},
journal = {CoRR},
volume = {abs/2201.03545},
year = {2022},
url = {https://arxiv.org/abs/2201.03545},
eprinttype = {arXiv},
eprint = {2201.03545},
timestamp = {Thu, 20 Jan 2022 14:21:35 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2201-03545.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["vision", "image-classification"], "datasets": ["imagenet-21k"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/tiger.jpg", "example_title": "Tiger"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/teapot.jpg", "example_title": "Teapot"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/palace.jpg", "example_title": "Palace"}]}
|
facebook/convnext-large-224-22k-1k
| null |
[
"transformers",
"pytorch",
"tf",
"convnext",
"image-classification",
"vision",
"dataset:imagenet-21k",
"arxiv:2201.03545",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# ConvNeXT (large-sized model)
ConvNeXT model trained on ImageNet-22k at resolution 224x224. It was introduced in the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Liu et al. and first released in [this repository](https://github.com/facebookresearch/ConvNeXt).
Disclaimer: The team releasing ConvNeXT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
ConvNeXT is a pure convolutional model (ConvNet), inspired by the design of Vision Transformers, that claims to outperform them. The authors started from a ResNet and "modernized" its design by taking the Swin Transformer as inspiration.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=convnext) to look for
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import ConvNextFeatureExtractor, ConvNextForImageClassification
import torch
from datasets import load_dataset
dataset = load_dataset("huggingface/cats-image")
image = dataset["test"]["image"][0]
feature_extractor = ConvNextFeatureExtractor.from_pretrained("facebook/convnext-large-224-22k")
model = ConvNextForImageClassification.from_pretrained("facebook/convnext-large-224-22k")
inputs = feature_extractor(image, return_tensors="pt")
with torch.no_grad():
logits = model(**inputs).logits
# model predicts one of the 22k ImageNet classes
predicted_label = logits.argmax(-1).item()
print(model.config.id2label[predicted_label]),
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/convnext).
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2201-03545,
author = {Zhuang Liu and
Hanzi Mao and
Chao{-}Yuan Wu and
Christoph Feichtenhofer and
Trevor Darrell and
Saining Xie},
title = {A ConvNet for the 2020s},
journal = {CoRR},
volume = {abs/2201.03545},
year = {2022},
url = {https://arxiv.org/abs/2201.03545},
eprinttype = {arXiv},
eprint = {2201.03545},
timestamp = {Thu, 20 Jan 2022 14:21:35 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2201-03545.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["vision", "image-classification"], "datasets": ["imagenet-21k"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/tiger.jpg", "example_title": "Tiger"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/teapot.jpg", "example_title": "Teapot"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/palace.jpg", "example_title": "Palace"}]}
|
facebook/convnext-large-224-22k
| null |
[
"transformers",
"pytorch",
"tf",
"safetensors",
"convnext",
"image-classification",
"vision",
"dataset:imagenet-21k",
"arxiv:2201.03545",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# ConvNeXT (large-sized model)
ConvNeXT model trained on ImageNet-1k at resolution 224x224. It was introduced in the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Liu et al. and first released in [this repository](https://github.com/facebookresearch/ConvNeXt).
Disclaimer: The team releasing ConvNeXT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
ConvNeXT is a pure convolutional model (ConvNet), inspired by the design of Vision Transformers, that claims to outperform them. The authors started from a ResNet and "modernized" its design by taking the Swin Transformer as inspiration.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=convnext) to look for
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import ConvNextImageProcessor, ConvNextForImageClassification
import torch
from datasets import load_dataset
dataset = load_dataset("huggingface/cats-image")
image = dataset["test"]["image"][0]
processor = ConvNextImageProcessor.from_pretrained("facebook/convnext-large-224")
model = ConvNextForImageClassification.from_pretrained("facebook/convnext-large-224")
inputs = processor(image, return_tensors="pt")
with torch.no_grad():
logits = model(**inputs).logits
# model predicts one of the 1000 ImageNet classes
predicted_label = logits.argmax(-1).item()
print(model.config.id2label[predicted_label]),
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/convnext).
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2201-03545,
author = {Zhuang Liu and
Hanzi Mao and
Chao{-}Yuan Wu and
Christoph Feichtenhofer and
Trevor Darrell and
Saining Xie},
title = {A ConvNet for the 2020s},
journal = {CoRR},
volume = {abs/2201.03545},
year = {2022},
url = {https://arxiv.org/abs/2201.03545},
eprinttype = {arXiv},
eprint = {2201.03545},
timestamp = {Thu, 20 Jan 2022 14:21:35 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2201-03545.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["vision", "image-classification"], "datasets": ["imagenet-1k"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/tiger.jpg", "example_title": "Tiger"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/teapot.jpg", "example_title": "Teapot"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/palace.jpg", "example_title": "Palace"}]}
|
facebook/convnext-large-224
| null |
[
"transformers",
"pytorch",
"tf",
"convnext",
"image-classification",
"vision",
"dataset:imagenet-1k",
"arxiv:2201.03545",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# ConvNeXT (large-sized model)
ConvNeXT model pre-trained on ImageNet-22k and fine-tuned on ImageNet-1k at resolution 384x384. It was introduced in the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Liu et al. and first released in [this repository](https://github.com/facebookresearch/ConvNeXt).
Disclaimer: The team releasing ConvNeXT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
ConvNeXT is a pure convolutional model (ConvNet), inspired by the design of Vision Transformers, that claims to outperform them. The authors started from a ResNet and "modernized" its design by taking the Swin Transformer as inspiration.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=convnext) to look for
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import ConvNextFeatureExtractor, ConvNextForImageClassification
import torch
from datasets import load_dataset
dataset = load_dataset("huggingface/cats-image")
image = dataset["test"]["image"][0]
feature_extractor = ConvNextFeatureExtractor.from_pretrained("facebook/convnext-large-384-22k-1k")
model = ConvNextForImageClassification.from_pretrained("facebook/convnext-large-384-22k-1k")
inputs = feature_extractor(image, return_tensors="pt")
with torch.no_grad():
logits = model(**inputs).logits
# model predicts one of the 1000 ImageNet classes
predicted_label = logits.argmax(-1).item()
print(model.config.id2label[predicted_label]),
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/convnext).
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2201-03545,
author = {Zhuang Liu and
Hanzi Mao and
Chao{-}Yuan Wu and
Christoph Feichtenhofer and
Trevor Darrell and
Saining Xie},
title = {A ConvNet for the 2020s},
journal = {CoRR},
volume = {abs/2201.03545},
year = {2022},
url = {https://arxiv.org/abs/2201.03545},
eprinttype = {arXiv},
eprint = {2201.03545},
timestamp = {Thu, 20 Jan 2022 14:21:35 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2201-03545.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["vision", "image-classification"], "datasets": ["imagenet-21k", "imagenet-1k"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/tiger.jpg", "example_title": "Tiger"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/teapot.jpg", "example_title": "Teapot"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/palace.jpg", "example_title": "Palace"}]}
|
facebook/convnext-large-384-22k-1k
| null |
[
"transformers",
"pytorch",
"tf",
"convnext",
"image-classification",
"vision",
"dataset:imagenet-21k",
"dataset:imagenet-1k",
"arxiv:2201.03545",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# ConvNeXT (large-sized model)
ConvNeXT model trained on ImageNet-1k at resolution 384x384. It was introduced in the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Liu et al. and first released in [this repository](https://github.com/facebookresearch/ConvNeXt).
Disclaimer: The team releasing ConvNeXT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
ConvNeXT is a pure convolutional model (ConvNet), inspired by the design of Vision Transformers, that claims to outperform them. The authors started from a ResNet and "modernized" its design by taking the Swin Transformer as inspiration.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=convnext) to look for
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import ConvNextFeatureExtractor, ConvNextForImageClassification
import torch
from datasets import load_dataset
dataset = load_dataset("huggingface/cats-image")
image = dataset["test"]["image"][0]
feature_extractor = ConvNextFeatureExtractor.from_pretrained("facebook/convnext-large-384")
model = ConvNextForImageClassification.from_pretrained("facebook/convnext-large-384")
inputs = feature_extractor(image, return_tensors="pt")
with torch.no_grad():
logits = model(**inputs).logits
# model predicts one of the 1000 ImageNet classes
predicted_label = logits.argmax(-1).item()
print(model.config.id2label[predicted_label]),
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/convnext).
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2201-03545,
author = {Zhuang Liu and
Hanzi Mao and
Chao{-}Yuan Wu and
Christoph Feichtenhofer and
Trevor Darrell and
Saining Xie},
title = {A ConvNet for the 2020s},
journal = {CoRR},
volume = {abs/2201.03545},
year = {2022},
url = {https://arxiv.org/abs/2201.03545},
eprinttype = {arXiv},
eprint = {2201.03545},
timestamp = {Thu, 20 Jan 2022 14:21:35 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2201-03545.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["vision", "image-classification"], "datasets": ["imagenet-1k"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/tiger.jpg", "example_title": "Tiger"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/teapot.jpg", "example_title": "Teapot"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/palace.jpg", "example_title": "Palace"}]}
|
facebook/convnext-large-384
| null |
[
"transformers",
"pytorch",
"tf",
"safetensors",
"convnext",
"image-classification",
"vision",
"dataset:imagenet-1k",
"arxiv:2201.03545",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# ConvNeXT (large-sized model)
ConvNeXT model trained on ImageNet-1k at resolution 224x224. It was introduced in the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Liu et al. and first released in [this repository](https://github.com/facebookresearch/ConvNeXt).
Disclaimer: The team releasing ConvNeXT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
ConvNeXT is a pure convolutional model (ConvNet), inspired by the design of Vision Transformers, that claims to outperform them. The authors started from a ResNet and "modernized" its design by taking the Swin Transformer as inspiration.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=convnext) to look for
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import ConvNextImageProcessor, ConvNextForImageClassification
import torch
from datasets import load_dataset
dataset = load_dataset("huggingface/cats-image")
image = dataset["test"]["image"][0]
processor = ConvNextImageProcessor.from_pretrained("facebook/convnext-small-224")
model = ConvNextForImageClassification.from_pretrained("facebook/convnext-small-224")
inputs = processor(image, return_tensors="pt")
with torch.no_grad():
logits = model(**inputs).logits
# model predicts one of the 1000 ImageNet classes
predicted_label = logits.argmax(-1).item()
print(model.config.id2label[predicted_label]),
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/convnext).
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2201-03545,
author = {Zhuang Liu and
Hanzi Mao and
Chao{-}Yuan Wu and
Christoph Feichtenhofer and
Trevor Darrell and
Saining Xie},
title = {A ConvNet for the 2020s},
journal = {CoRR},
volume = {abs/2201.03545},
year = {2022},
url = {https://arxiv.org/abs/2201.03545},
eprinttype = {arXiv},
eprint = {2201.03545},
timestamp = {Thu, 20 Jan 2022 14:21:35 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2201-03545.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["vision", "image-classification"], "datasets": ["imagenet-1k"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/tiger.jpg", "example_title": "Tiger"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/teapot.jpg", "example_title": "Teapot"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/palace.jpg", "example_title": "Palace"}]}
|
facebook/convnext-small-224
| null |
[
"transformers",
"pytorch",
"tf",
"safetensors",
"convnext",
"image-classification",
"vision",
"dataset:imagenet-1k",
"arxiv:2201.03545",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# ConvNeXT (tiny-sized model)
ConvNeXT model trained on ImageNet-1k at resolution 224x224. It was introduced in the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Liu et al. and first released in [this repository](https://github.com/facebookresearch/ConvNeXt).
Disclaimer: The team releasing ConvNeXT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
ConvNeXT is a pure convolutional model (ConvNet), inspired by the design of Vision Transformers, that claims to outperform them. The authors started from a ResNet and "modernized" its design by taking the Swin Transformer as inspiration.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=convnext) to look for
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import ConvNextImageProcessor, ConvNextForImageClassification
import torch
from datasets import load_dataset
dataset = load_dataset("huggingface/cats-image")
image = dataset["test"]["image"][0]
processor = ConvNextImageProcessor.from_pretrained("facebook/convnext-tiny-224")
model = ConvNextForImageClassification.from_pretrained("facebook/convnext-tiny-224")
inputs = processor(image, return_tensors="pt")
with torch.no_grad():
logits = model(**inputs).logits
# model predicts one of the 1000 ImageNet classes
predicted_label = logits.argmax(-1).item()
print(model.config.id2label[predicted_label]),
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/convnext).
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2201-03545,
author = {Zhuang Liu and
Hanzi Mao and
Chao{-}Yuan Wu and
Christoph Feichtenhofer and
Trevor Darrell and
Saining Xie},
title = {A ConvNet for the 2020s},
journal = {CoRR},
volume = {abs/2201.03545},
year = {2022},
url = {https://arxiv.org/abs/2201.03545},
eprinttype = {arXiv},
eprint = {2201.03545},
timestamp = {Thu, 20 Jan 2022 14:21:35 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2201-03545.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["vision", "image-classification"], "datasets": ["imagenet-1k"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/tiger.jpg", "example_title": "Tiger"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/teapot.jpg", "example_title": "Teapot"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/palace.jpg", "example_title": "Palace"}]}
|
facebook/convnext-tiny-224
| null |
[
"transformers",
"pytorch",
"tf",
"convnext",
"image-classification",
"vision",
"dataset:imagenet-1k",
"arxiv:2201.03545",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# ConvNeXT (xlarge-sized model)
ConvNeXT model trained on ImageNet-1k at resolution 224x224. It was introduced in the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Liu et al. and first released in [this repository](https://github.com/facebookresearch/ConvNeXt).
Disclaimer: The team releasing ConvNeXT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
ConvNeXT is a pure convolutional model (ConvNet), inspired by the design of Vision Transformers, that claims to outperform them. The authors started from a ResNet and "modernized" its design by taking the Swin Transformer as inspiration.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=convnext) to look for
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import ConvNextFeatureExtractor, ConvNextForImageClassification
import torch
from datasets import load_dataset
dataset = load_dataset("huggingface/cats-image")
image = dataset["test"]["image"][0]
feature_extractor = ConvNextFeatureExtractor.from_pretrained("facebook/convnext-xlarge-224-22k-1k")
model = ConvNextForImageClassification.from_pretrained("facebook/convnext-xlarge-224-22k-1k")
inputs = feature_extractor(image, return_tensors="pt")
with torch.no_grad():
logits = model(**inputs).logits
# model predicts one of the 1000 ImageNet classes
predicted_label = logits.argmax(-1).item()
print(model.config.id2label[predicted_label]),
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/convnext).
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2201-03545,
author = {Zhuang Liu and
Hanzi Mao and
Chao{-}Yuan Wu and
Christoph Feichtenhofer and
Trevor Darrell and
Saining Xie},
title = {A ConvNet for the 2020s},
journal = {CoRR},
volume = {abs/2201.03545},
year = {2022},
url = {https://arxiv.org/abs/2201.03545},
eprinttype = {arXiv},
eprint = {2201.03545},
timestamp = {Thu, 20 Jan 2022 14:21:35 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2201-03545.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["vision", "image-classification"], "datasets": ["imagenet-21k", "imagenet-1k"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/tiger.jpg", "example_title": "Tiger"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/teapot.jpg", "example_title": "Teapot"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/palace.jpg", "example_title": "Palace"}]}
|
facebook/convnext-xlarge-224-22k-1k
| null |
[
"transformers",
"pytorch",
"tf",
"convnext",
"image-classification",
"vision",
"dataset:imagenet-21k",
"dataset:imagenet-1k",
"arxiv:2201.03545",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# ConvNeXT (xlarge-sized model)
ConvNeXT model trained on ImageNet-22k at resolution 224x224. It was introduced in the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Liu et al. and first released in [this repository](https://github.com/facebookresearch/ConvNeXt).
Disclaimer: The team releasing ConvNeXT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
ConvNeXT is a pure convolutional model (ConvNet), inspired by the design of Vision Transformers, that claims to outperform them. The authors started from a ResNet and "modernized" its design by taking the Swin Transformer as inspiration.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=convnext) to look for
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import ConvNextFeatureExtractor, ConvNextForImageClassification
import torch
from datasets import load_dataset
dataset = load_dataset("huggingface/cats-image")
image = dataset["test"]["image"][0]
feature_extractor = ConvNextFeatureExtractor.from_pretrained("facebook/convnext-xlarge-224-22k")
model = ConvNextForImageClassification.from_pretrained("facebook/convnext-xlarge-224-22k")
inputs = feature_extractor(image, return_tensors="pt")
with torch.no_grad():
logits = model(**inputs).logits
# model predicts one of the 22k ImageNet classes
predicted_label = logits.argmax(-1).item()
print(model.config.id2label[predicted_label]),
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/convnext).
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2201-03545,
author = {Zhuang Liu and
Hanzi Mao and
Chao{-}Yuan Wu and
Christoph Feichtenhofer and
Trevor Darrell and
Saining Xie},
title = {A ConvNet for the 2020s},
journal = {CoRR},
volume = {abs/2201.03545},
year = {2022},
url = {https://arxiv.org/abs/2201.03545},
eprinttype = {arXiv},
eprint = {2201.03545},
timestamp = {Thu, 20 Jan 2022 14:21:35 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2201-03545.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["vision", "image-classification"], "datasets": ["imagenet-21k"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/tiger.jpg", "example_title": "Tiger"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/teapot.jpg", "example_title": "Teapot"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/palace.jpg", "example_title": "Palace"}]}
|
facebook/convnext-xlarge-224-22k
| null |
[
"transformers",
"pytorch",
"tf",
"convnext",
"image-classification",
"vision",
"dataset:imagenet-21k",
"arxiv:2201.03545",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# ConvNeXT (xlarge-sized model)
ConvNeXT model pre-trained on ImageNet-22k and fine-tuned on ImageNet-1k at resolution 384x384. It was introduced in the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Liu et al. and first released in [this repository](https://github.com/facebookresearch/ConvNeXt).
Disclaimer: The team releasing ConvNeXT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
ConvNeXT is a pure convolutional model (ConvNet), inspired by the design of Vision Transformers, that claims to outperform them. The authors started from a ResNet and "modernized" its design by taking the Swin Transformer as inspiration.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=convnext) to look for
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import ConvNextImageProcessor, ConvNextForImageClassification
import torch
from datasets import load_dataset
dataset = load_dataset("huggingface/cats-image")
image = dataset["test"]["image"][0]
processor = ConvNextImageProcessor.from_pretrained("facebook/convnext-xlarge-384-22k-1k")
model = ConvNextForImageClassification.from_pretrained("facebook/convnext-xlarge-384-22k-1k")
inputs = processor(image, return_tensors="pt")
with torch.no_grad():
logits = model(**inputs).logits
# model predicts one of the 1000 ImageNet classes
predicted_label = logits.argmax(-1).item()
print(model.config.id2label[predicted_label]),
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/convnext).
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2201-03545,
author = {Zhuang Liu and
Hanzi Mao and
Chao{-}Yuan Wu and
Christoph Feichtenhofer and
Trevor Darrell and
Saining Xie},
title = {A ConvNet for the 2020s},
journal = {CoRR},
volume = {abs/2201.03545},
year = {2022},
url = {https://arxiv.org/abs/2201.03545},
eprinttype = {arXiv},
eprint = {2201.03545},
timestamp = {Thu, 20 Jan 2022 14:21:35 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2201-03545.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["vision", "image-classification"], "datasets": ["imagenet-21k", "imagenet-1k"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/tiger.jpg", "example_title": "Tiger"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/teapot.jpg", "example_title": "Teapot"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/palace.jpg", "example_title": "Palace"}]}
|
facebook/convnext-xlarge-384-22k-1k
| null |
[
"transformers",
"pytorch",
"tf",
"convnext",
"image-classification",
"vision",
"dataset:imagenet-21k",
"dataset:imagenet-1k",
"arxiv:2201.03545",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# Data2Vec-Audio-Base-100h
[Facebook's Data2Vec](https://ai.facebook.com/research/data2vec-a-general-framework-for-self-supervised-learning-in-speech-vision-and-language/)
The base model pretrained and fine-tuned on 100 hours of Librispeech on 16kHz sampled speech audio. When using the model
make sure that your speech input is also sampled at 16Khz.
[Paper](https://arxiv.org/abs/2202.03555)
Authors: Alexei Baevski, Wei-Ning Hsu, Qiantong Xu, Arun Babu, Jiatao Gu, Michael Auli
**Abstract**
While the general idea of self-supervised learning is identical across modalities, the actual algorithms and objectives differ widely because they were developed with a single modality in mind. To get us closer to general self-supervised learning, we present data2vec, a framework that uses the same learning method for either speech, NLP or computer vision. The core idea is to predict latent representations of the full input data based on a masked view of the input in a self-distillation setup using a standard Transformer architecture. Instead of predicting modality-specific targets such as words, visual tokens or units of human speech which are local in nature, data2vec predicts contextualized latent representations that contain information from the entire input. Experiments on the major benchmarks of speech recognition, image classification, and natural language understanding demonstrate a new state of the art or competitive performance to predominant approaches.
The original model can be found under https://github.com/pytorch/fairseq/tree/main/examples/data2vec .
# Pre-Training method
For more information, please take a look at the [official paper](https://arxiv.org/abs/2202.03555).
# Usage
To transcribe audio files the model can be used as a standalone acoustic model as follows:
```python
from transformers import Wav2Vec2Processor, Data2VecForCTC
from datasets import load_dataset
import torch
# load model and processor
processor = Wav2Vec2Processor.from_pretrained("facebook/data2vec-audio-base-100h")
model = Data2VecForCTC.from_pretrained("facebook/data2vec-audio-base-100h")
# load dummy dataset and read soundfiles
ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
# tokenize
input_values = processor(ds[0]["audio"]["array"],, return_tensors="pt", padding="longest").input_values # Batch size 1
# retrieve logits
logits = model(input_values).logits
# take argmax and decode
predicted_ids = torch.argmax(logits, dim=-1)
transcription = processor.batch_decode(predicted_ids)
```
|
{"language": "en", "license": "apache-2.0", "tags": ["speech"], "datasets": ["librispeech_asr"]}
|
facebook/data2vec-audio-base-100h
| null |
[
"transformers",
"pytorch",
"data2vec-audio",
"automatic-speech-recognition",
"speech",
"en",
"dataset:librispeech_asr",
"arxiv:2202.03555",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# Data2Vec-Audio-Base-10m
[Facebook's Data2Vec](https://ai.facebook.com/research/data2vec-a-general-framework-for-self-supervised-learning-in-speech-vision-and-language/)
The base model pretrained and fine-tuned on 10 minutes of Librispeech on 16kHz sampled speech audio. When using the model
make sure that your speech input is also sampled at 16Khz.
[Paper](https://arxiv.org/abs/2202.03555)
Authors: Alexei Baevski, Wei-Ning Hsu, Qiantong Xu, Arun Babu, Jiatao Gu, Michael Auli
**Abstract**
While the general idea of self-supervised learning is identical across modalities, the actual algorithms and objectives differ widely because they were developed with a single modality in mind. To get us closer to general self-supervised learning, we present data2vec, a framework that uses the same learning method for either speech, NLP or computer vision. The core idea is to predict latent representations of the full input data based on a masked view of the input in a self-distillation setup using a standard Transformer architecture. Instead of predicting modality-specific targets such as words, visual tokens or units of human speech which are local in nature, data2vec predicts contextualized latent representations that contain information from the entire input. Experiments on the major benchmarks of speech recognition, image classification, and natural language understanding demonstrate a new state of the art or competitive performance to predominant approaches.
The original model can be found under https://github.com/pytorch/fairseq/tree/main/examples/data2vec .
# Pre-Training method
For more information, please take a look at the [official paper](https://arxiv.org/abs/2202.03555).
# Usage
To transcribe audio files the model can be used as a standalone acoustic model as follows:
```python
from transformers import Wav2Vec2Processor, Data2VecForCTC
from datasets import load_dataset
import torch
# load model and processor
processor = Wav2Vec2Processor.from_pretrained("facebook/data2vec-audio-base-10m")
model = Data2VecForCTC.from_pretrained("facebook/data2vec-audio-base-10m")
# load dummy dataset and read soundfiles
ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
# tokenize
input_values = processor(ds[0]["audio"]["array"],, return_tensors="pt", padding="longest").input_values # Batch size 1
# retrieve logits
logits = model(input_values).logits
# take argmax and decode
predicted_ids = torch.argmax(logits, dim=-1)
transcription = processor.batch_decode(predicted_ids)
```
|
{"language": "en", "license": "apache-2.0", "tags": ["speech"], "datasets": ["librispeech_asr"]}
|
facebook/data2vec-audio-base-10m
| null |
[
"transformers",
"pytorch",
"data2vec-audio",
"automatic-speech-recognition",
"speech",
"en",
"dataset:librispeech_asr",
"arxiv:2202.03555",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# Data2Vec-Audio-Base-960h
[Facebook's Data2Vec](https://ai.facebook.com/research/data2vec-a-general-framework-for-self-supervised-learning-in-speech-vision-and-language/)
The base model pretrained and fine-tuned on 960 hours of Librispeech on 16kHz sampled speech audio. When using the model
make sure that your speech input is also sampled at 16Khz.
[Paper](https://arxiv.org/abs/2202.03555)
Authors: Alexei Baevski, Wei-Ning Hsu, Qiantong Xu, Arun Babu, Jiatao Gu, Michael Auli
**Abstract**
While the general idea of self-supervised learning is identical across modalities, the actual algorithms and objectives differ widely because they were developed with a single modality in mind. To get us closer to general self-supervised learning, we present data2vec, a framework that uses the same learning method for either speech, NLP or computer vision. The core idea is to predict latent representations of the full input data based on a masked view of the input in a self-distillation setup using a standard Transformer architecture. Instead of predicting modality-specific targets such as words, visual tokens or units of human speech which are local in nature, data2vec predicts contextualized latent representations that contain information from the entire input. Experiments on the major benchmarks of speech recognition, image classification, and natural language understanding demonstrate a new state of the art or competitive performance to predominant approaches.
The original model can be found under https://github.com/pytorch/fairseq/tree/main/examples/data2vec .
# Pre-Training method
For more information, please take a look at the [official paper](https://arxiv.org/abs/2202.03555).
# Usage
To transcribe audio files the model can be used as a standalone acoustic model as follows:
```python
from transformers import Wav2Vec2Processor, Data2VecForCTC
from datasets import load_dataset
import torch
# load model and processor
processor = Wav2Vec2Processor.from_pretrained("facebook/data2vec-audio-base-960h")
model = Data2VecForCTC.from_pretrained("facebook/data2vec-audio-base-960h")
# load dummy dataset and read soundfiles
ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
# tokenize
input_values = processor(ds[0]["audio"]["array"],, return_tensors="pt", padding="longest").input_values # Batch size 1
# retrieve logits
logits = model(input_values).logits
# take argmax and decode
predicted_ids = torch.argmax(logits, dim=-1)
transcription = processor.batch_decode(predicted_ids)
```
## Evaluation
This code snippet shows how to evaluate **facebook/data2vec-audio-base-960h** on LibriSpeech's "clean" and "other" test data.
```python
from transformers import Wav2Vec2Processor, Data2VecForCTC
from datasets import load_dataset
import torch
from jiwer import wer
# load model and processor
processor = Wav2Vec2Processor.from_pretrained("facebook/data2vec-audio-base-960h").to("cuda")
model = Data2VecForCTC.from_pretrained("facebook/data2vec-audio-base-960h")
librispeech_eval = load_dataset("librispeech_asr", "clean", split="test")
def map_to_pred(batch):
input_values = processor(batch["audio"]["array"], return_tensors="pt", padding="longest").input_values
with torch.no_grad():
logits = model(input_values.to("cuda")).logits
predicted_ids = torch.argmax(logits, dim=-1)
transcription = processor.batch_decode(predicted_ids)
batch["transcription"] = transcription
return batch
result = librispeech_eval.map(map_to_pred, batched=True, batch_size=1, remove_columns=["audio"])
print("WER:", wer(result["text"], result["transcription"]))
```
*Result (WER)*:
| "clean" | "other" |
|---|---|
| 2.77 | 7.08 |
|
{"language": "en", "license": "apache-2.0", "tags": ["speech", "hf-asr-leaderboard"], "datasets": ["librispeech_asr"], "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}], "model-index": [{"name": "data2vec-audio-base-960h", "results": [{"task": {"type": "automatic-speech-recognition", "name": "Automatic Speech Recognition"}, "dataset": {"name": "LibriSpeech (clean)", "type": "librispeech_asr", "config": "clean", "split": "test", "args": {"language": "en"}}, "metrics": [{"type": "wer", "value": 2.77, "name": "Test WER"}]}, {"task": {"type": "automatic-speech-recognition", "name": "Automatic Speech Recognition"}, "dataset": {"name": "LibriSpeech (other)", "type": "librispeech_asr", "config": "other", "split": "test", "args": {"language": "en"}}, "metrics": [{"type": "wer", "value": 7.08, "name": "Test WER"}]}]}]}
|
facebook/data2vec-audio-base-960h
| null |
[
"transformers",
"pytorch",
"data2vec-audio",
"automatic-speech-recognition",
"speech",
"hf-asr-leaderboard",
"en",
"dataset:librispeech_asr",
"arxiv:2202.03555",
"license:apache-2.0",
"model-index",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
feature-extraction
|
transformers
|
# Data2Vec-Audio-Base
[Facebook's Data2Vec](https://ai.facebook.com/research/data2vec-a-general-framework-for-self-supervised-learning-in-speech-vision-and-language/)
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/2202.03555)
Authors: Alexei Baevski, Wei-Ning Hsu, Qiantong Xu, Arun Babu, Jiatao Gu, Michael Auli
**Abstract**
While the general idea of self-supervised learning is identical across modalities, the actual algorithms and objectives differ widely because they were developed with a single modality in mind. To get us closer to general self-supervised learning, we present data2vec, a framework that uses the same learning method for either speech, NLP or computer vision. The core idea is to predict latent representations of the full input data based on a masked view of the input in a self-distillation setup using a standard Transformer architecture. Instead of predicting modality-specific targets such as words, visual tokens or units of human speech which are local in nature, data2vec predicts contextualized latent representations that contain information from the entire input. Experiments on the major benchmarks of speech recognition, image classification, and natural language understanding demonstrate a new state of the art or competitive performance to predominant approaches.
The original model can be found under https://github.com/pytorch/fairseq/tree/main/examples/data2vec .
# Pre-Training method
For more information, please take a look at the [official paper](https://arxiv.org/abs/2202.03555).
# 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.
|
{"language": "en", "license": "apache-2.0", "tags": ["speech"], "datasets": ["librispeech_asr"]}
|
facebook/data2vec-audio-base
| null |
[
"transformers",
"pytorch",
"data2vec-audio",
"feature-extraction",
"speech",
"en",
"dataset:librispeech_asr",
"arxiv:2202.03555",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
feature-extraction
|
transformers
|
# Data2Vec-Text base model
Pretrained model on English language using the *data2vec* objective. It was introduced in
[this paper](https://arxiv.org/abs/2202.03555) and first released in
[this repository](https://github.com/pytorch/fairseq/tree/main/examples/data2vec). This model is case-sensitive: it
makes a difference between english and English.
Disclaimer: The team releasing Data2Vec-Text did not write a model card for this model so this model card has been written by
the Hugging Face team.
## Pre-Training method
For more information, please take a look at the [official paper](https://arxiv.org/abs/2202.03555).
## Abstract
*While the general idea of self-supervised learning is identical across modalities, the actual algorithms and objectives differ widely because
they were developed with a single modality in
mind. To get us closer to general self-supervised
learning, we present data2vec, a framework that
uses the same learning method for either speech,
NLP or computer vision. The core idea is to predict latent representations of the full input data
based on a masked view of the input in a selfdistillation setup using a standard Transformer architecture. Instead of predicting modality-specific
targets such as words, visual tokens or units of
human speech which are local in nature, data2vec
predicts contextualized latent representations that
contain information from the entire input. Experiments on the major benchmarks of speech
recognition, image classification, and natural language understanding demonstrate a new state of
the art or competitive performance to predominant approaches.*
## Intended uses & limitations
The model is intended to be fine-tuned on a downstream task.
See the [model hub](https://huggingface.co/models?filter=data2vec-text) to look for fine-tuned versions on a task that
interests you.
Note that this model is primarily aimed at being fine-tuned on tasks that use the whole sentence (potentially masked)
to make decisions, such as sequence classification, token classification or question answering. For tasks such as text
generation you should look at model like GPT2.
## Training data
The RoBERTa model was pretrained on the reunion of five datasets:
- [BookCorpus](https://yknzhu.wixsite.com/mbweb), a dataset consisting of 11,038 unpublished books;
- [English Wikipedia](https://en.wikipedia.org/wiki/English_Wikipedia) (excluding lists, tables and headers) ;
- [CC-News](https://commoncrawl.org/2016/10/news-dataset-available/), a dataset containing 63 millions English news
articles crawled between September 2016 and February 2019.
- [OpenWebText](https://github.com/jcpeterson/openwebtext), an opensource recreation of the WebText dataset used to
train GPT-2,
- [Stories](https://arxiv.org/abs/1806.02847) a dataset containing a subset of CommonCrawl data filtered to match the
story-like style of Winograd schemas.
Together theses datasets weight 160GB of text.
### BibTeX entry and citation info
```bibtex
@misc{https://doi.org/10.48550/arxiv.2202.03555,
doi = {10.48550/ARXIV.2202.03555},
url = {https://arxiv.org/abs/2202.03555},
author = {Baevski, Alexei and Hsu, Wei-Ning and Xu, Qiantong and Babu, Arun and Gu, Jiatao and Auli, Michael},
keywords = {Machine Learning (cs.LG), FOS: Computer and information sciences, FOS: Computer and information sciences},
title = {data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language},
publisher = {arXiv},
year = {2022},
copyright = {arXiv.org perpetual, non-exclusive license}
}
```
|
{"language": "en", "license": "mit", "tags": ["exbert"], "datasets": ["bookcorpus", "wikipedia"]}
|
facebook/data2vec-text-base
| null |
[
"transformers",
"pytorch",
"data2vec-text",
"feature-extraction",
"exbert",
"en",
"dataset:bookcorpus",
"dataset:wikipedia",
"arxiv:2202.03555",
"arxiv:1806.02847",
"license:mit",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# Distilled Data-efficient Image Transformer (base-sized model)
Distilled data-efficient Image Transformer (DeiT) model pre-trained and fine-tuned on ImageNet-1k (1 million images, 1,000 classes) at resolution 224x224. It was first introduced in the paper [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877) by Touvron et al. and first released in [this repository](https://github.com/facebookresearch/deit). However, the weights were converted from the [timm repository](https://github.com/rwightman/pytorch-image-models) by Ross Wightman.
Disclaimer: The team releasing DeiT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
This model is a distilled Vision Transformer (ViT). It uses a distillation token, besides the class token, to effectively learn from a teacher (CNN) during both pre-training and fine-tuning. The distillation token is learned through backpropagation, by interacting with the class ([CLS]) and patch tokens through the self-attention layers.
Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=facebook/deit) to look for
fine-tuned versions on a task that interests you.
### How to use
Since this model is a distilled ViT model, you can plug it into DeiTModel, DeiTForImageClassification or DeiTForImageClassificationWithTeacher. Note that the model expects the data to be prepared using DeiTFeatureExtractor. Here we use AutoFeatureExtractor, which will automatically use the appropriate feature extractor given the model name.
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import AutoFeatureExtractor, DeiTForImageClassificationWithTeacher
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = AutoFeatureExtractor.from_pretrained('facebook/deit-base-distilled-patch16-224')
model = DeiTForImageClassificationWithTeacher.from_pretrained('facebook/deit-base-distilled-patch16-224')
inputs = feature_extractor(images=image, return_tensors="pt")
# forward pass
outputs = model(**inputs)
logits = outputs.logits
# model predicts one of the 1000 ImageNet classes
predicted_class_idx = logits.argmax(-1).item()
print("Predicted class:", model.config.id2label[predicted_class_idx])
```
Currently, both the feature extractor and model support PyTorch. Tensorflow and JAX/FLAX are coming soon.
## Training data
This model was pretrained and fine-tuned with distillation on [ImageNet-1k](http://www.image-net.org/challenges/LSVRC/2012/), a dataset consisting of 1 million images and 1k classes.
## Training procedure
### Preprocessing
The exact details of preprocessing of images during training/validation can be found [here](https://github.com/facebookresearch/deit/blob/ab5715372db8c6cad5740714b2216d55aeae052e/datasets.py#L78).
At inference time, images are resized/rescaled to the same resolution (256x256), center-cropped at 224x224 and normalized across the RGB channels with the ImageNet mean and standard deviation.
### Pretraining
The model was trained on a single 8-GPU node for 3 days. Training resolution is 224. For all hyperparameters (such as batch size and learning rate) we refer to table 9 of the original paper.
## Evaluation results
| Model | ImageNet top-1 accuracy | ImageNet top-5 accuracy | # params | URL |
|---------------------------------------|-------------------------|-------------------------|----------|------------------------------------------------------------------|
| DeiT-tiny | 72.2 | 91.1 | 5M | https://huggingface.co/facebook/deit-tiny-patch16-224 |
| DeiT-small | 79.9 | 95.0 | 22M | https://huggingface.co/facebook/deit-small-patch16-224 |
| DeiT-base | 81.8 | 95.6 | 86M | https://huggingface.co/facebook/deit-base-patch16-224 |
| DeiT-tiny distilled | 74.5 | 91.9 | 6M | https://huggingface.co/facebook/deit-tiny-distilled-patch16-224 |
| DeiT-small distilled | 81.2 | 95.4 | 22M | https://huggingface.co/facebook/deit-small-distilled-patch16-224 |
| **DeiT-base distilled** | **83.4** | **96.5** | **87M** | **https://huggingface.co/facebook/deit-base-distilled-patch16-224** |
| DeiT-base 384 | 82.9 | 96.2 | 87M | https://huggingface.co/facebook/deit-base-patch16-384 |
| DeiT-base distilled 384 (1000 epochs) | 85.2 | 97.2 | 88M | https://huggingface.co/facebook/deit-base-distilled-patch16-384 |
Note that for fine-tuning, the best results are obtained with a higher resolution (384x384). Of course, increasing the model size will result in better performance.
### BibTeX entry and citation info
```bibtex
@misc{touvron2021training,
title={Training data-efficient image transformers & distillation through attention},
author={Hugo Touvron and Matthieu Cord and Matthijs Douze and Francisco Massa and Alexandre Sablayrolles and Hervé Jégou},
year={2021},
eprint={2012.12877},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@misc{wu2020visual,
title={Visual Transformers: Token-based Image Representation and Processing for Computer Vision},
author={Bichen Wu and Chenfeng Xu and Xiaoliang Dai and Alvin Wan and Peizhao Zhang and Zhicheng Yan and Masayoshi Tomizuka and Joseph Gonzalez and Kurt Keutzer and Peter Vajda},
year={2020},
eprint={2006.03677},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@inproceedings{deng2009imagenet,
title={Imagenet: A large-scale hierarchical image database},
author={Deng, Jia and Dong, Wei and Socher, Richard and Li, Li-Jia and Li, Kai and Fei-Fei, Li},
booktitle={2009 IEEE conference on computer vision and pattern recognition},
pages={248--255},
year={2009},
organization={Ieee}
}
```
|
{"license": "apache-2.0", "tags": ["image-classification", "vision"], "datasets": ["imagenet"]}
|
facebook/deit-base-distilled-patch16-224
| null |
[
"transformers",
"pytorch",
"tf",
"deit",
"image-classification",
"vision",
"dataset:imagenet",
"arxiv:2012.12877",
"arxiv:2006.03677",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# Distilled Data-efficient Image Transformer (base-sized model)
Distilled data-efficient Image Transformer (DeiT) model pre-trained at resolution 224x224 and fine-tuned at resolution 384x384 on ImageNet-1k (1 million images, 1,000 classes). It was first introduced in the paper [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877) by Touvron et al. and first released in [this repository](https://github.com/facebookresearch/deit). However, the weights were converted from the [timm repository](https://github.com/rwightman/pytorch-image-models) by Ross Wightman.
Disclaimer: The team releasing DeiT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
This model is a distilled Vision Transformer (ViT). It uses a distillation token, besides the class token, to effectively learn from a teacher (CNN) during both pre-training and fine-tuning. The distillation token is learned through backpropagation, by interacting with the class ([CLS]) and patch tokens through the self-attention layers.
Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=facebook/deit) to look for
fine-tuned versions on a task that interests you.
### How to use
Since this model is a distilled ViT model, you can plug it into DeiTModel, DeiTForImageClassification or DeiTForImageClassificationWithTeacher. Note that the model expects the data to be prepared using DeiTFeatureExtractor. Here we use AutoFeatureExtractor, which will automatically use the appropriate feature extractor given the model name.
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import AutoFeatureExtractor, DeiTForImageClassificationWithTeacher
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = AutoFeatureExtractor.from_pretrained('facebook/deit-base-distilled-patch16-384')
model = DeiTForImageClassificationWithTeacher.from_pretrained('facebook/deit-base-distilled-patch16-384')
inputs = feature_extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
logits = outputs.logits
# model predicts one of the 1000 ImageNet classes
predicted_class_idx = logits.argmax(-1).item()
print("Predicted class:", model.config.id2label[predicted_class_idx])
```
Currently, both the feature extractor and model support PyTorch. Tensorflow and JAX/FLAX are coming soon.
## Training data
This model was pretrained and fine-tuned with distillation on [ImageNet-1k](http://www.image-net.org/challenges/LSVRC/2012/), a dataset consisting of 1 million images and 1k classes.
## Training procedure
### Preprocessing
The exact details of preprocessing of images during training/validation can be found [here](https://github.com/facebookresearch/deit/blob/ab5715372db8c6cad5740714b2216d55aeae052e/datasets.py#L78).
At inference time, images are resized/rescaled to the same resolution (438x438), center-cropped at 384x384 and normalized across the RGB channels with the ImageNet mean and standard deviation.
### Pretraining
The model was trained on a single 8-GPU node for 3 days. Pre-training resolution is 224. For all hyperparameters (such as batch size and learning rate) we refer to table 9 of the original paper.
## Evaluation results
| Model | ImageNet top-1 accuracy | ImageNet top-5 accuracy | # params | URL |
|-------------------------------------------|-------------------------|-------------------------|----------|------------------------------------------------------------------|
| DeiT-tiny | 72.2 | 91.1 | 5M | https://huggingface.co/facebook/deit-tiny-patch16-224 |
| DeiT-small | 79.9 | 95.0 | 22M | https://huggingface.co/facebook/deit-small-patch16-224 |
| DeiT-base | 81.8 | 95.6 | 86M | https://huggingface.co/facebook/deit-base-patch16-224 |
| DeiT-tiny distilled | 74.5 | 91.9 | 6M | https://huggingface.co/facebook/deit-tiny-distilled-patch16-224 |
| DeiT-small distilled | 81.2 | 95.4 | 22M | https://huggingface.co/facebook/deit-small-distilled-patch16-224 |
| DeiT-base distilled | 83.4 | 96.5 | 87M | https://huggingface.co/facebook/deit-base-distilled-patch16-224 |
| DeiT-base 384 | 82.9 | 96.2 | 87M | https://huggingface.co/facebook/deit-base-patch16-384 |
| **DeiT-base distilled 384 (1000 epochs)** | **85.2** | **97.2** | **88M** | **https://huggingface.co/facebook/deit-base-distilled-patch16-384** |
Note that for fine-tuning, the best results are obtained with a higher resolution (384x384). Of course, increasing the model size will result in better performance.
### BibTeX entry and citation info
```bibtex
@misc{touvron2021training,
title={Training data-efficient image transformers & distillation through attention},
author={Hugo Touvron and Matthieu Cord and Matthijs Douze and Francisco Massa and Alexandre Sablayrolles and Hervé Jégou},
year={2021},
eprint={2012.12877},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@misc{wu2020visual,
title={Visual Transformers: Token-based Image Representation and Processing for Computer Vision},
author={Bichen Wu and Chenfeng Xu and Xiaoliang Dai and Alvin Wan and Peizhao Zhang and Zhicheng Yan and Masayoshi Tomizuka and Joseph Gonzalez and Kurt Keutzer and Peter Vajda},
year={2020},
eprint={2006.03677},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@inproceedings{deng2009imagenet,
title={Imagenet: A large-scale hierarchical image database},
author={Deng, Jia and Dong, Wei and Socher, Richard and Li, Li-Jia and Li, Kai and Fei-Fei, Li},
booktitle={2009 IEEE conference on computer vision and pattern recognition},
pages={248--255},
year={2009},
organization={Ieee}
}
```
|
{"license": "apache-2.0", "tags": ["image-classification", "vision"], "datasets": ["imagenet"]}
|
facebook/deit-base-distilled-patch16-384
| null |
[
"transformers",
"pytorch",
"tf",
"safetensors",
"deit",
"image-classification",
"vision",
"dataset:imagenet",
"arxiv:2012.12877",
"arxiv:2006.03677",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# Data-efficient Image Transformer (base-sized model)
Data-efficient Image Transformer (DeiT) model pre-trained and fine-tuned on ImageNet-1k (1 million images, 1,000 classes) at resolution 224x224. It was first introduced in the paper [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877) by Touvron et al. and first released in [this repository](https://github.com/facebookresearch/deit). However, the weights were converted from the [timm repository](https://github.com/rwightman/pytorch-image-models) by Ross Wightman.
Disclaimer: The team releasing DeiT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
This model is actually a more efficiently trained Vision Transformer (ViT).
The Vision Transformer (ViT) is a transformer encoder model (BERT-like) pre-trained and fine-tuned on a large collection of images in a supervised fashion, namely ImageNet-1k, at a resolution of 224x224 pixels.
Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded. One also adds a [CLS] token to the beginning of a sequence to use it for classification tasks. One also adds absolute position embeddings before feeding the sequence to the layers of the Transformer encoder.
By pre-training the model, it learns an inner representation of images that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled images for instance, you can train a standard classifier by placing a linear layer on top of the pre-trained encoder. One typically places a linear layer on top of the [CLS] token, as the last hidden state of this token can be seen as a representation of an entire image.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=facebook/deit) to look for
fine-tuned versions on a task that interests you.
### How to use
Since this model is a more efficiently trained ViT model, you can plug it into ViTModel or ViTForImageClassification. Note that the model expects the data to be prepared using DeiTFeatureExtractor. Here we use AutoFeatureExtractor, which will automatically use the appropriate feature extractor given the model name.
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import AutoFeatureExtractor, ViTForImageClassification
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = AutoFeatureExtractor.from_pretrained('facebook/deit-base-patch16-224')
model = ViTForImageClassification.from_pretrained('facebook/deit-base-patch16-224')
inputs = feature_extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
logits = outputs.logits
# model predicts one of the 1000 ImageNet classes
predicted_class_idx = logits.argmax(-1).item()
print("Predicted class:", model.config.id2label[predicted_class_idx])
```
Currently, both the feature extractor and model support PyTorch. Tensorflow and JAX/FLAX are coming soon.
## Training data
The ViT model was pretrained on [ImageNet-1k](http://www.image-net.org/challenges/LSVRC/2012/), a dataset consisting of 1 million images and 1k classes.
## Training procedure
### Preprocessing
The exact details of preprocessing of images during training/validation can be found [here](https://github.com/facebookresearch/deit/blob/ab5715372db8c6cad5740714b2216d55aeae052e/datasets.py#L78).
At inference time, images are resized/rescaled to the same resolution (256x256), center-cropped at 224x224 and normalized across the RGB channels with the ImageNet mean and standard deviation.
### Pretraining
The model was trained on a single 8-GPU node for 3 days. Training resolution is 224. For all hyperparameters (such as batch size and learning rate) we refer to table 9 of the original paper.
## Evaluation results
| Model | ImageNet top-1 accuracy | ImageNet top-5 accuracy | # params | URL |
|---------------------------------------|-------------------------|-------------------------|----------|------------------------------------------------------------------|
| DeiT-tiny | 72.2 | 91.1 | 5M | https://huggingface.co/facebook/deit-tiny-patch16-224 |
| DeiT-small | 79.9 | 95.0 | 22M | https://huggingface.co/facebook/deit-small-patch16-224 |
| **DeiT-base** | **81.8** | **95.6** | **86M** | **https://huggingface.co/facebook/deit-base-patch16-224** |
| DeiT-tiny distilled | 74.5 | 91.9 | 6M | https://huggingface.co/facebook/deit-tiny-distilled-patch16-224 |
| DeiT-small distilled | 81.2 | 95.4 | 22M | https://huggingface.co/facebook/deit-small-distilled-patch16-224 |
| DeiT-base distilled | 83.4 | 96.5 | 87M | https://huggingface.co/facebook/deit-base-distilled-patch16-224 |
| DeiT-base 384 | 82.9 | 96.2 | 87M | https://huggingface.co/facebook/deit-base-patch16-384 |
| DeiT-base distilled 384 (1000 epochs) | 85.2 | 97.2 | 88M | https://huggingface.co/facebook/deit-base-distilled-patch16-384 |
Note that for fine-tuning, the best results are obtained with a higher resolution (384x384). Of course, increasing the model size will result in better performance.
### BibTeX entry and citation info
```bibtex
@misc{touvron2021training,
title={Training data-efficient image transformers & distillation through attention},
author={Hugo Touvron and Matthieu Cord and Matthijs Douze and Francisco Massa and Alexandre Sablayrolles and Hervé Jégou},
year={2021},
eprint={2012.12877},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@misc{wu2020visual,
title={Visual Transformers: Token-based Image Representation and Processing for Computer Vision},
author={Bichen Wu and Chenfeng Xu and Xiaoliang Dai and Alvin Wan and Peizhao Zhang and Zhicheng Yan and Masayoshi Tomizuka and Joseph Gonzalez and Kurt Keutzer and Peter Vajda},
year={2020},
eprint={2006.03677},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@inproceedings{deng2009imagenet,
title={Imagenet: A large-scale hierarchical image database},
author={Deng, Jia and Dong, Wei and Socher, Richard and Li, Li-Jia and Li, Kai and Fei-Fei, Li},
booktitle={2009 IEEE conference on computer vision and pattern recognition},
pages={248--255},
year={2009},
organization={Ieee}
}
```
|
{"license": "apache-2.0", "tags": ["image-classification"], "datasets": ["imagenet-1k"]}
|
facebook/deit-base-patch16-224
| null |
[
"transformers",
"pytorch",
"tf",
"vit",
"image-classification",
"dataset:imagenet-1k",
"arxiv:2012.12877",
"arxiv:2006.03677",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# Data-efficient Image Transformer (base-sized model)
Data-efficient Image Transformer (DeiT) model pre-trained and fine-tuned on ImageNet-1k (1 million images, 1,000 classes) at resolution 384x384. It was first introduced in the paper [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877) by Touvron et al. and first released in [this repository](https://github.com/facebookresearch/deit). However, the weights were converted from the [timm repository](https://github.com/rwightman/pytorch-image-models) by Ross Wightman.
Disclaimer: The team releasing DeiT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
This model is actually a more efficiently trained Vision Transformer (ViT).
The Vision Transformer (ViT) is a transformer encoder model (BERT-like) pre-trained at resolution 224 and fine-tuned at resolution 384 on a large collection of images in a supervised fashion, namely ImageNet-1k.
Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded. One also adds a [CLS] token to the beginning of a sequence to use it for classification tasks. One also adds absolute position embeddings before feeding the sequence to the layers of the Transformer encoder.
By pre-training the model, it learns an inner representation of images that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled images for instance, you can train a standard classifier by placing a linear layer on top of the pre-trained encoder. One typically places a linear layer on top of the [CLS] token, as the last hidden state of this token can be seen as a representation of an entire image.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=facebook/deit) to look for
fine-tuned versions on a task that interests you.
### How to use
Since this model is a more efficiently trained ViT model, you can plug it into ViTModel or ViTForImageClassification. Note that the model expects the data to be prepared using DeiTFeatureExtractor. Here we use AutoFeatureExtractor, which will automatically use the appropriate feature extractor given the model name.
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import AutoFeatureExtractor, ViTForImageClassification
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = AutoFeatureExtractor.from_pretrained('facebook/deit-base-patch16-384')
model = ViTForImageClassification.from_pretrained('facebook/deit-base-patch16-384')
inputs = feature_extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
logits = outputs.logits
# model predicts one of the 1000 ImageNet classes
predicted_class_idx = logits.argmax(-1).item()
print("Predicted class:", model.config.id2label[predicted_class_idx])
```
Currently, both the feature extractor and model support PyTorch. Tensorflow and JAX/FLAX are coming soon.
## Training data
The ViT model was pretrained on [ImageNet-1k](http://www.image-net.org/challenges/LSVRC/2012/), a dataset consisting of 1 million images and 1k classes.
## Training procedure
### Preprocessing
The exact details of preprocessing of images during training/validation can be found [here](https://github.com/facebookresearch/deit/blob/ab5715372db8c6cad5740714b2216d55aeae052e/datasets.py#L78).
At inference time, images are resized/rescaled to the same resolution (438x438), center-cropped at 384x384 and normalized across the RGB channels with the ImageNet mean and standard deviation.
### Pretraining
The model was trained on a single 8-GPU node for 3 days. Pre-training resolution is 224. For all hyperparameters (such as batch size and learning rate) we refer to table 9 of the original paper.
## Evaluation results
| Model | ImageNet top-1 accuracy | ImageNet top-5 accuracy | # params | URL |
|---------------------------------------|-------------------------|-------------------------|----------|------------------------------------------------------------------|
| DeiT-tiny | 72.2 | 91.1 | 5M | https://huggingface.co/facebook/deit-tiny-patch16-224 |
| DeiT-small | 79.9 | 95.0 | 22M | https://huggingface.co/facebook/deit-small-patch16-224 |
| DeiT-base | 81.8 | 95.6 | 86M | https://huggingface.co/facebook/deit-base-patch16-224 |
| DeiT-tiny distilled | 74.5 | 91.9 | 6M | https://huggingface.co/facebook/deit-tiny-distilled-patch16-224 |
| DeiT-small distilled | 81.2 | 95.4 | 22M | https://huggingface.co/facebook/deit-small-distilled-patch16-224 |
| DeiT-base distilled | 83.4 | 96.5 | 87M | https://huggingface.co/facebook/deit-base-distilled-patch16-224 |
| **DeiT-base 384** | **82.9** | **96.2** | **87M** | **https://huggingface.co/facebook/deit-base-patch16-384** |
| DeiT-base distilled 384 (1000 epochs) | 85.2 | 97.2 | 88M | https://huggingface.co/facebook/deit-base-distilled-patch16-384 |
Note that for fine-tuning, the best results are obtained with a higher resolution (384x384). Of course, increasing the model size will result in better performance.
### BibTeX entry and citation info
```bibtex
@misc{touvron2021training,
title={Training data-efficient image transformers & distillation through attention},
author={Hugo Touvron and Matthieu Cord and Matthijs Douze and Francisco Massa and Alexandre Sablayrolles and Hervé Jégou},
year={2021},
eprint={2012.12877},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@misc{wu2020visual,
title={Visual Transformers: Token-based Image Representation and Processing for Computer Vision},
author={Bichen Wu and Chenfeng Xu and Xiaoliang Dai and Alvin Wan and Peizhao Zhang and Zhicheng Yan and Masayoshi Tomizuka and Joseph Gonzalez and Kurt Keutzer and Peter Vajda},
year={2020},
eprint={2006.03677},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@inproceedings{deng2009imagenet,
title={Imagenet: A large-scale hierarchical image database},
author={Deng, Jia and Dong, Wei and Socher, Richard and Li, Li-Jia and Li, Kai and Fei-Fei, Li},
booktitle={2009 IEEE conference on computer vision and pattern recognition},
pages={248--255},
year={2009},
organization={Ieee}
}
```
|
{"license": "apache-2.0", "tags": ["image-classification"], "datasets": ["imagenet-1k"]}
|
facebook/deit-base-patch16-384
| null |
[
"transformers",
"pytorch",
"tf",
"vit",
"image-classification",
"dataset:imagenet-1k",
"arxiv:2012.12877",
"arxiv:2006.03677",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# Distilled Data-efficient Image Transformer (small-sized model)
Distilled data-efficient Image Transformer (DeiT) model pre-trained and fine-tuned on ImageNet-1k (1 million images, 1,000 classes) at resolution 224x224. It was first introduced in the paper [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877) by Touvron et al. and first released in [this repository](https://github.com/facebookresearch/deit). However, the weights were converted from the [timm repository](https://github.com/rwightman/pytorch-image-models) by Ross Wightman.
Disclaimer: The team releasing DeiT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
This model is a distilled Vision Transformer (ViT). It uses a distillation token, besides the class token, to effectively learn from a teacher (CNN) during both pre-training and fine-tuning. The distillation token is learned through backpropagation, by interacting with the class ([CLS]) and patch tokens through the self-attention layers.
Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=facebook/deit) to look for
fine-tuned versions on a task that interests you.
### How to use
Since this model is a distilled ViT model, you can plug it into DeiTModel, DeiTForImageClassification or DeiTForImageClassificationWithTeacher. Note that the model expects the data to be prepared using DeiTFeatureExtractor. Here we use AutoFeatureExtractor, which will automatically use the appropriate feature extractor given the model name.
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import AutoFeatureExtractor, DeiTForImageClassificationWithTeacher
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = AutoFeatureExtractor.from_pretrained('facebook/deit-small-distilled-patch16-224')
model = DeiTForImageClassificationWithTeacher.from_pretrained('facebook/deit-small-distilled-patch16-224')
inputs = feature_extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
logits = outputs.logits
# model predicts one of the 1000 ImageNet classes
predicted_class_idx = logits.argmax(-1).item()
print("Predicted class:", model.config.id2label[predicted_class_idx])
```
Currently, both the feature extractor and model support PyTorch. Tensorflow and JAX/FLAX are coming soon.
## Training data
This model was pretrained and fine-tuned with distillation on [ImageNet-1k](http://www.image-net.org/challenges/LSVRC/2012/), a dataset consisting of 1 million images and 1k classes.
## Training procedure
### Preprocessing
The exact details of preprocessing of images during training/validation can be found [here](https://github.com/facebookresearch/deit/blob/ab5715372db8c6cad5740714b2216d55aeae052e/datasets.py#L78).
At inference time, images are resized/rescaled to the same resolution (256x256), center-cropped at 224x224 and normalized across the RGB channels with the ImageNet mean and standard deviation.
### Pretraining
The model was trained on a single 8-GPU node for 3 days. Training resolution is 224. For all hyperparameters (such as batch size and learning rate) we refer to table 9 of the original paper.
## Evaluation results
| Model | ImageNet top-1 accuracy | ImageNet top-5 accuracy | # params | URL |
|---------------------------------------|-------------------------|-------------------------|----------|------------------------------------------------------------------|
| DeiT-tiny | 72.2 | 91.1 | 5M | https://huggingface.co/facebook/deit-tiny-patch16-224 |
| DeiT-small | 79.9 | 95.0 | 22M | https://huggingface.co/facebook/deit-small-patch16-224 |
| DeiT-base | 81.8 | 95.6 | 86M | https://huggingface.co/facebook/deit-base-patch16-224 |
| DeiT-tiny distilled | 74.5 | 91.9 | 6M | https://huggingface.co/facebook/deit-tiny-distilled-patch16-224 |
| **DeiT-small distilled** | **81.2** | **95.4** | **22M** | **https://huggingface.co/facebook/deit-small-distilled-patch16-224** |
| DeiT-base distilled | 83.4 | 96.5 | 87M | https://huggingface.co/facebook/deit-base-distilled-patch16-224 |
| DeiT-base 384 | 82.9 | 96.2 | 87M | https://huggingface.co/facebook/deit-base-patch16-384 |
| DeiT-base distilled 384 (1000 epochs) | 85.2 | 97.2 | 88M | https://huggingface.co/facebook/deit-base-distilled-patch16-384 |
Note that for fine-tuning, the best results are obtained with a higher resolution (384x384). Of course, increasing the model size will result in better performance.
### BibTeX entry and citation info
```bibtex
@misc{touvron2021training,
title={Training data-efficient image transformers & distillation through attention},
author={Hugo Touvron and Matthieu Cord and Matthijs Douze and Francisco Massa and Alexandre Sablayrolles and Hervé Jégou},
year={2021},
eprint={2012.12877},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@misc{wu2020visual,
title={Visual Transformers: Token-based Image Representation and Processing for Computer Vision},
author={Bichen Wu and Chenfeng Xu and Xiaoliang Dai and Alvin Wan and Peizhao Zhang and Zhicheng Yan and Masayoshi Tomizuka and Joseph Gonzalez and Kurt Keutzer and Peter Vajda},
year={2020},
eprint={2006.03677},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@inproceedings{deng2009imagenet,
title={Imagenet: A large-scale hierarchical image database},
author={Deng, Jia and Dong, Wei and Socher, Richard and Li, Li-Jia and Li, Kai and Fei-Fei, Li},
booktitle={2009 IEEE conference on computer vision and pattern recognition},
pages={248--255},
year={2009},
organization={Ieee}
}
```
|
{"license": "apache-2.0", "tags": ["image-classification", "vision"], "datasets": ["imagenet"]}
|
facebook/deit-small-distilled-patch16-224
| null |
[
"transformers",
"pytorch",
"tf",
"deit",
"image-classification",
"vision",
"dataset:imagenet",
"arxiv:2012.12877",
"arxiv:2006.03677",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# Data-efficient Image Transformer (small-sized model)
Data-efficient Image Transformer (DeiT) model pre-trained and fine-tuned on ImageNet-1k (1 million images, 1,000 classes) at resolution 224x224. It was first introduced in the paper [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877) by Touvron et al. and first released in [this repository](https://github.com/facebookresearch/deit). However, the weights were converted from the [timm repository](https://github.com/rwightman/pytorch-image-models) by Ross Wightman.
Disclaimer: The team releasing DeiT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
This model is actually a more efficiently trained Vision Transformer (ViT).
The Vision Transformer (ViT) is a transformer encoder model (BERT-like) pre-trained and fine-tuned on a large collection of images in a supervised fashion, namely ImageNet-1k, at a resolution of 224x224 pixels.
Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded. One also adds a [CLS] token to the beginning of a sequence to use it for classification tasks. One also adds absolute position embeddings before feeding the sequence to the layers of the Transformer encoder.
By pre-training the model, it learns an inner representation of images that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled images for instance, you can train a standard classifier by placing a linear layer on top of the pre-trained encoder. One typically places a linear layer on top of the [CLS] token, as the last hidden state of this token can be seen as a representation of an entire image.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=facebook/deit) to look for
fine-tuned versions on a task that interests you.
### How to use
Since this model is a more efficiently trained ViT model, you can plug it into ViTModel or ViTForImageClassification. Note that the model expects the data to be prepared using DeiTFeatureExtractor. Here we use AutoFeatureExtractor, which will automatically use the appropriate feature extractor given the model name.
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import AutoFeatureExtractor, ViTForImageClassification
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = AutoFeatureExtractor.from_pretrained('facebook/deit-small-patch16-224')
model = ViTForImageClassification.from_pretrained('facebook/deit-small-patch16-224')
inputs = feature_extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
logits = outputs.logits
# model predicts one of the 1000 ImageNet classes
predicted_class_idx = logits.argmax(-1).item()
print("Predicted class:", model.config.id2label[predicted_class_idx])
```
Currently, both the feature extractor and model support PyTorch. Tensorflow and JAX/FLAX are coming soon.
## Training data
The ViT model was pretrained on [ImageNet-1k](http://www.image-net.org/challenges/LSVRC/2012/), a dataset consisting of 1 million images and 1k classes.
## Training procedure
### Preprocessing
The exact details of preprocessing of images during training/validation can be found [here](https://github.com/facebookresearch/deit/blob/ab5715372db8c6cad5740714b2216d55aeae052e/datasets.py#L78).
At inference time, images are resized/rescaled to the same resolution (256x256), center-cropped at 224x224 and normalized across the RGB channels with the ImageNet mean and standard deviation.
### Pretraining
The model was trained on a single 8-GPU node for 3 days. Training resolution is 224. For all hyperparameters (such as batch size and learning rate) we refer to table 9 of the original paper.
## Evaluation results
| Model | ImageNet top-1 accuracy | ImageNet top-5 accuracy | # params | URL |
|---------------------------------------|-------------------------|-------------------------|----------|------------------------------------------------------------------|
| DeiT-tiny | 72.2 | 91.1 | 5M | https://huggingface.co/facebook/deit-tiny-patch16-224 |
| **DeiT-small** | **79.9** | **95.0** | **22M** | **https://huggingface.co/facebook/deit-small-patch16-224** |
| DeiT-base | 81.8 | 95.6 | 86M | https://huggingface.co/facebook/deit-base-patch16-224 |
| DeiT-tiny distilled | 74.5 | 91.9 | 6M | https://huggingface.co/facebook/deit-tiny-distilled-patch16-224 |
| DeiT-small distilled | 81.2 | 95.4 | 22M | https://huggingface.co/facebook/deit-small-distilled-patch16-224 |
| DeiT-base distilled | 83.4 | 96.5 | 87M | https://huggingface.co/facebook/deit-base-distilled-patch16-224 |
| DeiT-base 384 | 82.9 | 96.2 | 87M | https://huggingface.co/facebook/deit-base-patch16-384 |
| DeiT-base distilled 384 (1000 epochs) | 85.2 | 97.2 | 88M | https://huggingface.co/facebook/deit-base-distilled-patch16-384 |
Note that for fine-tuning, the best results are obtained with a higher resolution (384x384). Of course, increasing the model size will result in better performance.
### BibTeX entry and citation info
```bibtex
@misc{touvron2021training,
title={Training data-efficient image transformers & distillation through attention},
author={Hugo Touvron and Matthieu Cord and Matthijs Douze and Francisco Massa and Alexandre Sablayrolles and Hervé Jégou},
year={2021},
eprint={2012.12877},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@misc{wu2020visual,
title={Visual Transformers: Token-based Image Representation and Processing for Computer Vision},
author={Bichen Wu and Chenfeng Xu and Xiaoliang Dai and Alvin Wan and Peizhao Zhang and Zhicheng Yan and Masayoshi Tomizuka and Joseph Gonzalez and Kurt Keutzer and Peter Vajda},
year={2020},
eprint={2006.03677},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@inproceedings{deng2009imagenet,
title={Imagenet: A large-scale hierarchical image database},
author={Deng, Jia and Dong, Wei and Socher, Richard and Li, Li-Jia and Li, Kai and Fei-Fei, Li},
booktitle={2009 IEEE conference on computer vision and pattern recognition},
pages={248--255},
year={2009},
organization={Ieee}
}
```
|
{"license": "apache-2.0", "tags": ["image-classification"], "datasets": ["imagenet-1k"]}
|
facebook/deit-small-patch16-224
| null |
[
"transformers",
"pytorch",
"tf",
"vit",
"image-classification",
"dataset:imagenet-1k",
"arxiv:2012.12877",
"arxiv:2006.03677",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# Distilled Data-efficient Image Transformer (tiny-sized model)
Distilled data-efficient Image Transformer (DeiT) model pre-trained and fine-tuned on ImageNet-1k (1 million images, 1,000 classes) at resolution 224x224. It was first introduced in the paper [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877) by Touvron et al. and first released in [this repository](https://github.com/facebookresearch/deit). However, the weights were converted from the [timm repository](https://github.com/rwightman/pytorch-image-models) by Ross Wightman.
Disclaimer: The team releasing DeiT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
This model is a distilled Vision Transformer (ViT). It uses a distillation token, besides the class token, to effectively learn from a teacher (CNN) during both pre-training and fine-tuning. The distillation token is learned through backpropagation, by interacting with the class ([CLS]) and patch tokens through the self-attention layers.
Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=facebook/deit) to look for
fine-tuned versions on a task that interests you.
### How to use
Since this model is a distilled ViT model, you can plug it into DeiTModel, DeiTForImageClassification or DeiTForImageClassificationWithTeacher. Note that the model expects the data to be prepared using DeiTFeatureExtractor. Here we use AutoFeatureExtractor, which will automatically use the appropriate feature extractor given the model name.
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import AutoFeatureExtractor, DeiTForImageClassificationWithTeacher
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = AutoFeatureExtractor.from_pretrained('facebook/deit-tiny-distilled-patch16-224')
model = DeiTForImageClassificationWithTeacher.from_pretrained('facebook/deit-tiny-distilled-patch16-224')
inputs = feature_extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
logits = outputs.logits
# model predicts one of the 1000 ImageNet classes
predicted_class_idx = logits.argmax(-1).item()
print("Predicted class:", model.config.id2label[predicted_class_idx])
```
Currently, both the feature extractor and model support PyTorch. Tensorflow and JAX/FLAX are coming soon.
## Training data
This model was pretrained and fine-tuned with distillation on [ImageNet-1k](http://www.image-net.org/challenges/LSVRC/2012/), a dataset consisting of 1 million images and 1k classes.
## Training procedure
### Preprocessing
The exact details of preprocessing of images during training/validation can be found [here](https://github.com/facebookresearch/deit/blob/ab5715372db8c6cad5740714b2216d55aeae052e/datasets.py#L78).
At inference time, images are resized/rescaled to the same resolution (256x256), center-cropped at 224x224 and normalized across the RGB channels with the ImageNet mean and standard deviation.
### Pretraining
The model was trained on a single 8-GPU node for 3 days. Training resolution is 224. For all hyperparameters (such as batch size and learning rate) we refer to table 9 of the original paper.
## Evaluation results
| Model | ImageNet top-1 accuracy | ImageNet top-5 accuracy | # params | URL |
|---------------------------------------|-------------------------|-------------------------|----------|------------------------------------------------------------------|
| DeiT-tiny | 72.2 | 91.1 | 5M | https://huggingface.co/facebook/deit-tiny-patch16-224 |
| DeiT-small | 79.9 | 95.0 | 22M | https://huggingface.co/facebook/deit-small-patch16-224 |
| DeiT-base | 81.8 | 95.6 | 86M | https://huggingface.co/facebook/deit-base-patch16-224 |
| **DeiT-tiny distilled** | **74.5** | **91.9** | **6M** | **https://huggingface.co/facebook/deit-tiny-distilled-patch16-224** |
| DeiT-small distilled | 81.2 | 95.4 | 22M | https://huggingface.co/facebook/deit-small-distilled-patch16-224 |
| DeiT-base distilled | 83.4 | 96.5 | 87M | https://huggingface.co/facebook/deit-base-distilled-patch16-224 |
| DeiT-base 384 | 82.9 | 96.2 | 87M | https://huggingface.co/facebook/deit-base-patch16-384 |
| DeiT-base distilled 384 (1000 epochs) | 85.2 | 97.2 | 88M | https://huggingface.co/facebook/deit-base-distilled-patch16-384 |
Note that for fine-tuning, the best results are obtained with a higher resolution (384x384). Of course, increasing the model size will result in better performance.
### BibTeX entry and citation info
```bibtex
@misc{touvron2021training,
title={Training data-efficient image transformers & distillation through attention},
author={Hugo Touvron and Matthieu Cord and Matthijs Douze and Francisco Massa and Alexandre Sablayrolles and Hervé Jégou},
year={2021},
eprint={2012.12877},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@misc{wu2020visual,
title={Visual Transformers: Token-based Image Representation and Processing for Computer Vision},
author={Bichen Wu and Chenfeng Xu and Xiaoliang Dai and Alvin Wan and Peizhao Zhang and Zhicheng Yan and Masayoshi Tomizuka and Joseph Gonzalez and Kurt Keutzer and Peter Vajda},
year={2020},
eprint={2006.03677},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@inproceedings{deng2009imagenet,
title={Imagenet: A large-scale hierarchical image database},
author={Deng, Jia and Dong, Wei and Socher, Richard and Li, Li-Jia and Li, Kai and Fei-Fei, Li},
booktitle={2009 IEEE conference on computer vision and pattern recognition},
pages={248--255},
year={2009},
organization={Ieee}
}
```
|
{"license": "apache-2.0", "tags": ["image-classification", "vision"], "datasets": ["imagenet"]}
|
facebook/deit-tiny-distilled-patch16-224
| null |
[
"transformers",
"pytorch",
"tf",
"deit",
"image-classification",
"vision",
"dataset:imagenet",
"arxiv:2012.12877",
"arxiv:2006.03677",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-classification
|
transformers
|
# Data-efficient Image Transformer (tiny-sized model)
Data-efficient Image Transformer (DeiT) model pre-trained and fine-tuned on ImageNet-1k (1 million images, 1,000 classes) at resolution 224x224. It was first introduced in the paper [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877) by Touvron et al. and first released in [this repository](https://github.com/facebookresearch/deit). However, the weights were converted from the [timm repository](https://github.com/rwightman/pytorch-image-models) by Ross Wightman.
Disclaimer: The team releasing DeiT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
This model is actually a more efficiently trained Vision Transformer (ViT).
The Vision Transformer (ViT) is a transformer encoder model (BERT-like) pre-trained and fine-tuned on a large collection of images in a supervised fashion, namely ImageNet-1k, at a resolution of 224x224 pixels.
Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded. One also adds a [CLS] token to the beginning of a sequence to use it for classification tasks. One also adds absolute position embeddings before feeding the sequence to the layers of the Transformer encoder.
By pre-training the model, it learns an inner representation of images that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled images for instance, you can train a standard classifier by placing a linear layer on top of the pre-trained encoder. One typically places a linear layer on top of the [CLS] token, as the last hidden state of this token can be seen as a representation of an entire image.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=facebook/deit) to look for
fine-tuned versions on a task that interests you.
### How to use
Since this model is a more efficiently trained ViT model, you can plug it into ViTModel or ViTForImageClassification. Note that the model expects the data to be prepared using DeiTFeatureExtractor. Here we use AutoFeatureExtractor, which will automatically use the appropriate feature extractor given the model name.
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import AutoFeatureExtractor, ViTForImageClassification
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = AutoFeatureExtractor.from_pretrained('facebook/deit-tiny-patch16-224')
model = ViTForImageClassification.from_pretrained('facebook/deit-tiny-patch16-224')
inputs = feature_extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
logits = outputs.logits
# model predicts one of the 1000 ImageNet classes
predicted_class_idx = logits.argmax(-1).item()
print("Predicted class:", model.config.id2label[predicted_class_idx])
```
Currently, both the feature extractor and model support PyTorch. Tensorflow and JAX/FLAX are coming soon.
## Training data
The ViT model was pretrained on [ImageNet-1k](http://www.image-net.org/challenges/LSVRC/2012/), a dataset consisting of 1 million images and 1k classes.
## Training procedure
### Preprocessing
The exact details of preprocessing of images during training/validation can be found [here](https://github.com/facebookresearch/deit/blob/ab5715372db8c6cad5740714b2216d55aeae052e/datasets.py#L78).
At inference time, images are resized/rescaled to the same resolution (256x256), center-cropped at 224x224 and normalized across the RGB channels with the ImageNet mean and standard deviation.
### Pretraining
The model was trained on a single 8-GPU node for 3 days. Training resolution is 224. For all hyperparameters (such as batch size and learning rate) we refer to table 9 of the original paper.
## Evaluation results
| Model | ImageNet top-1 accuracy | ImageNet top-5 accuracy | # params | URL |
|---------------------------------------|-------------------------|-------------------------|----------|------------------------------------------------------------------|
| **DeiT-tiny** | **72.2** | **91.1** | **5M** | **https://huggingface.co/facebook/deit-tiny-patch16-224** |
| DeiT-small | 79.9 | 95.0 | 22M | https://huggingface.co/facebook/deit-small-patch16-224 |
| DeiT-base | 81.8 | 95.6 | 86M | https://huggingface.co/facebook/deit-base-patch16-224 |
| DeiT-tiny distilled | 74.5 | 91.9 | 6M | https://huggingface.co/facebook/deit-tiny-distilled-patch16-224 |
| DeiT-small distilled | 81.2 | 95.4 | 22M | https://huggingface.co/facebook/deit-small-distilled-patch16-224 |
| DeiT-base distilled | 83.4 | 96.5 | 87M | https://huggingface.co/facebook/deit-base-distilled-patch16-224 |
| DeiT-base 384 | 82.9 | 96.2 | 87M | https://huggingface.co/facebook/deit-base-patch16-384 |
| DeiT-base distilled 384 (1000 epochs) | 85.2 | 97.2 | 88M | https://huggingface.co/facebook/deit-base-distilled-patch16-384 |
Note that for fine-tuning, the best results are obtained with a higher resolution (384x384). Of course, increasing the model size will result in better performance.
### BibTeX entry and citation info
```bibtex
@misc{touvron2021training,
title={Training data-efficient image transformers & distillation through attention},
author={Hugo Touvron and Matthieu Cord and Matthijs Douze and Francisco Massa and Alexandre Sablayrolles and Hervé Jégou},
year={2021},
eprint={2012.12877},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@misc{wu2020visual,
title={Visual Transformers: Token-based Image Representation and Processing for Computer Vision},
author={Bichen Wu and Chenfeng Xu and Xiaoliang Dai and Alvin Wan and Peizhao Zhang and Zhicheng Yan and Masayoshi Tomizuka and Joseph Gonzalez and Kurt Keutzer and Peter Vajda},
year={2020},
eprint={2006.03677},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@inproceedings{deng2009imagenet,
title={Imagenet: A large-scale hierarchical image database},
author={Deng, Jia and Dong, Wei and Socher, Richard and Li, Li-Jia and Li, Kai and Fei-Fei, Li},
booktitle={2009 IEEE conference on computer vision and pattern recognition},
pages={248--255},
year={2009},
organization={Ieee}
}
```
|
{"license": "apache-2.0", "tags": ["image-classification"], "datasets": ["imagenet"]}
|
facebook/deit-tiny-patch16-224
| null |
[
"transformers",
"pytorch",
"tf",
"vit",
"image-classification",
"dataset:imagenet",
"arxiv:2012.12877",
"arxiv:2006.03677",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
object-detection
|
transformers
|
# DETR (End-to-End Object Detection) model with ResNet-101 backbone (dilated C5 stage)
DEtection TRansformer (DETR) model trained end-to-end on COCO 2017 object detection (118k annotated images). It was introduced in the paper [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872) by Carion et al. and first released in [this repository](https://github.com/facebookresearch/detr).
Disclaimer: The team releasing DETR did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
The DETR model is an encoder-decoder transformer with a convolutional backbone. Two heads are added on top of the decoder outputs in order to perform object detection: a linear layer for the class labels and a MLP (multi-layer perceptron) for the bounding boxes. The model uses so-called object queries to detect objects in an image. Each object query looks for a particular object in the image. For COCO, the number of object queries is set to 100.
The model is trained using a "bipartite matching loss": one compares the predicted classes + bounding boxes of each of the N = 100 object queries to the ground truth annotations, padded up to the same length N (so if an image only contains 4 objects, 96 annotations will just have a "no object" as class and "no bounding box" as bounding box). The Hungarian matching algorithm is used to create an optimal one-to-one mapping between each of the N queries and each of the N annotations. Next, standard cross-entropy (for the classes) and a linear combination of the L1 and generalized IoU loss (for the bounding boxes) are used to optimize the parameters of the model.
## Intended uses & limitations
You can use the raw model for object detection. See the [model hub](https://huggingface.co/models?search=facebook/detr) to look for all available DETR models.
### How to use
Here is how to use this model:
```python
from transformers import DetrFeatureExtractor, DetrForObjectDetection
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = DetrFeatureExtractor.from_pretrained('facebook/detr-resnet-101-dc5')
model = DetrForObjectDetection.from_pretrained('facebook/detr-resnet-101-dc5')
inputs = feature_extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
# model predicts bounding boxes and corresponding COCO classes
logits = outputs.logits
bboxes = outputs.pred_boxes
```
Currently, both the feature extractor and model support PyTorch.
## Training data
The DETR model was trained on [COCO 2017 object detection](https://cocodataset.org/#download), a dataset consisting of 118k/5k annotated images for training/validation respectively.
## Training procedure
### Preprocessing
The exact details of preprocessing of images during training/validation can be found [here](https://github.com/google-research/vision_transformer/blob/master/vit_jax/input_pipeline.py).
Images are resized/rescaled such that the shortest side is at least 800 pixels and the largest side at most 1333 pixels, and normalized across the RGB channels with the ImageNet mean (0.485, 0.456, 0.406) and standard deviation (0.229, 0.224, 0.225).
### Training
The model was trained for 300 epochs on 16 V100 GPUs. This takes 3 days, with 4 images per GPU (hence a total batch size of 64).
## Evaluation results
This model achieves an AP (average precision) of **44.9** on COCO 2017 validation. For more details regarding evaluation results, we refer to table 1 of the original paper.
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2005-12872,
author = {Nicolas Carion and
Francisco Massa and
Gabriel Synnaeve and
Nicolas Usunier and
Alexander Kirillov and
Sergey Zagoruyko},
title = {End-to-End Object Detection with Transformers},
journal = {CoRR},
volume = {abs/2005.12872},
year = {2020},
url = {https://arxiv.org/abs/2005.12872},
archivePrefix = {arXiv},
eprint = {2005.12872},
timestamp = {Thu, 28 May 2020 17:38:09 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2005-12872.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["object-detection"], "datasets": ["coco"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/savanna.jpg", "example_title": "Savanna"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/football-match.jpg", "example_title": "Football Match"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/airport.jpg", "example_title": "Airport"}]}
|
facebook/detr-resnet-101-dc5
| null |
[
"transformers",
"pytorch",
"safetensors",
"detr",
"object-detection",
"dataset:coco",
"arxiv:2005.12872",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-segmentation
|
transformers
|
# DETR (End-to-End Object Detection) model with ResNet-101 backbone
DEtection TRansformer (DETR) model trained end-to-end on COCO 2017 panoptic (118k annotated images). It was introduced in the paper [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872) by Carion et al. and first released in [this repository](https://github.com/facebookresearch/detr).
Disclaimer: The team releasing DETR did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
The DETR model is an encoder-decoder transformer with a convolutional backbone. Two heads are added on top of the decoder outputs in order to perform object detection: a linear layer for the class labels and a MLP (multi-layer perceptron) for the bounding boxes. The model uses so-called object queries to detect objects in an image. Each object query looks for a particular object in the image. For COCO, the number of object queries is set to 100.
The model is trained using a "bipartite matching loss": one compares the predicted classes + bounding boxes of each of the N = 100 object queries to the ground truth annotations, padded up to the same length N (so if an image only contains 4 objects, 96 annotations will just have a "no object" as class and "no bounding box" as bounding box). The Hungarian matching algorithm is used to create an optimal one-to-one mapping between each of the N queries and each of the N annotations. Next, standard cross-entropy (for the classes) and a linear combination of the L1 and generalized IoU loss (for the bounding boxes) are used to optimize the parameters of the model.
DETR can be naturally extended to perform panoptic segmentation, by adding a mask head on top of the decoder outputs.
## Intended uses & limitations
You can use the raw model for panoptic segmentation. See the [model hub](https://huggingface.co/models?search=facebook/detr) to look for all available DETR models.
### How to use
Here is how to use this model:
```python
from transformers import DetrFeatureExtractor, DetrForSegmentation
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = DetrFeatureExtractor.from_pretrained('facebook/detr-resnet-101-panoptic')
model = DetrForSegmentation.from_pretrained('facebook/detr-resnet-101-panoptic')
# prepare inputs for the model
inputs = feature_extractor(images=image, return_tensors="pt")
# forward pass
outputs = model(**inputs)
# use the `post_process_panoptic` method of `DetrFeatureExtractor` to convert to COCO format
processed_sizes = torch.as_tensor(inputs["pixel_values"].shape[-2:]).unsqueeze(0)
result = feature_extractor.post_process_panoptic(outputs, processed_sizes)[0]
# the segmentation is stored in a special-format png
panoptic_seg = Image.open(io.BytesIO(result["png_string"]))
panoptic_seg = numpy.array(panoptic_seg, dtype=numpy.uint8)
# retrieve the ids corresponding to each mask
panoptic_seg_id = rgb_to_id(panoptic_seg)
```
Currently, both the feature extractor and model support PyTorch.
## Training data
The DETR model was trained on [COCO 2017 panoptic](https://cocodataset.org/#download), a dataset consisting of 118k/5k annotated images for training/validation respectively.
## Training procedure
### Preprocessing
The exact details of preprocessing of images during training/validation can be found [here](https://github.com/facebookresearch/detr/blob/master/datasets/coco_panoptic.py).
Images are resized/rescaled such that the shortest side is at least 800 pixels and the largest side at most 1333 pixels, and normalized across the RGB channels with the ImageNet mean (0.485, 0.456, 0.406) and standard deviation (0.229, 0.224, 0.225).
### Training
The model was trained for 300 epochs on 16 V100 GPUs. This takes 3 days, with 4 images per GPU (hence a total batch size of 64).
## Evaluation results
This model achieves the following results on COCO 2017 validation: a box AP (average precision) of **40.1**, a segmentation AP (average precision) of **33** and a PQ (panoptic quality) of **45.1**.
For more details regarding evaluation results, we refer to table 5 of the original paper.
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2005-12872,
author = {Nicolas Carion and
Francisco Massa and
Gabriel Synnaeve and
Nicolas Usunier and
Alexander Kirillov and
Sergey Zagoruyko},
title = {End-to-End Object Detection with Transformers},
journal = {CoRR},
volume = {abs/2005.12872},
year = {2020},
url = {https://arxiv.org/abs/2005.12872},
archivePrefix = {arXiv},
eprint = {2005.12872},
timestamp = {Thu, 28 May 2020 17:38:09 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2005-12872.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["image-segmentation", "vision"], "datasets": ["coco"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/dog-cat.jpg", "example_title": "Dog & Cat"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/construction-site.jpg", "example_title": "Construction Site"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/apple-orange.jpg", "example_title": "Apple & Orange"}]}
|
facebook/detr-resnet-101-panoptic
| null |
[
"transformers",
"pytorch",
"safetensors",
"detr",
"image-segmentation",
"vision",
"dataset:coco",
"arxiv:2005.12872",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
object-detection
|
transformers
|
# DETR (End-to-End Object Detection) model with ResNet-101 backbone
DEtection TRansformer (DETR) model trained end-to-end on COCO 2017 object detection (118k annotated images). It was introduced in the paper [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872) by Carion et al. and first released in [this repository](https://github.com/facebookresearch/detr).
Disclaimer: The team releasing DETR did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
The DETR model is an encoder-decoder transformer with a convolutional backbone. Two heads are added on top of the decoder outputs in order to perform object detection: a linear layer for the class labels and a MLP (multi-layer perceptron) for the bounding boxes. The model uses so-called object queries to detect objects in an image. Each object query looks for a particular object in the image. For COCO, the number of object queries is set to 100.
The model is trained using a "bipartite matching loss": one compares the predicted classes + bounding boxes of each of the N = 100 object queries to the ground truth annotations, padded up to the same length N (so if an image only contains 4 objects, 96 annotations will just have a "no object" as class and "no bounding box" as bounding box). The Hungarian matching algorithm is used to create an optimal one-to-one mapping between each of the N queries and each of the N annotations. Next, standard cross-entropy (for the classes) and a linear combination of the L1 and generalized IoU loss (for the bounding boxes) are used to optimize the parameters of the model.
## Intended uses & limitations
You can use the raw model for object detection. See the [model hub](https://huggingface.co/models?search=facebook/detr) to look for all available DETR models.
### How to use
Here is how to use this model:
```python
from transformers import DetrImageProcessor, DetrForObjectDetection
import torch
from PIL import Image
import requests
url = "http://images.cocodataset.org/val2017/000000039769.jpg"
image = Image.open(requests.get(url, stream=True).raw)
# you can specify the revision tag if you don't want the timm dependency
processor = DetrImageProcessor.from_pretrained("facebook/detr-resnet-101", revision="no_timm")
model = DetrForObjectDetection.from_pretrained("facebook/detr-resnet-101", revision="no_timm")
inputs = processor(images=image, return_tensors="pt")
outputs = model(**inputs)
# convert outputs (bounding boxes and class logits) to COCO API
# let's only keep detections with score > 0.9
target_sizes = torch.tensor([image.size[::-1]])
results = processor.post_process_object_detection(outputs, target_sizes=target_sizes, threshold=0.9)[0]
for score, label, box in zip(results["scores"], results["labels"], results["boxes"]):
box = [round(i, 2) for i in box.tolist()]
print(
f"Detected {model.config.id2label[label.item()]} with confidence "
f"{round(score.item(), 3)} at location {box}"
)
```
This should output (something along the lines of):
```
Detected cat with confidence 0.998 at location [344.06, 24.85, 640.34, 373.74]
Detected remote with confidence 0.997 at location [328.13, 75.93, 372.81, 187.66]
Detected remote with confidence 0.997 at location [39.34, 70.13, 175.56, 118.78]
Detected cat with confidence 0.998 at location [15.36, 51.75, 316.89, 471.16]
Detected couch with confidence 0.995 at location [-0.19, 0.71, 639.73, 474.17]
```
Currently, both the feature extractor and model support PyTorch.
## Training data
The DETR model was trained on [COCO 2017 object detection](https://cocodataset.org/#download), a dataset consisting of 118k/5k annotated images for training/validation respectively.
## Training procedure
### Preprocessing
The exact details of preprocessing of images during training/validation can be found [here](https://github.com/google-research/vision_transformer/blob/master/vit_jax/input_pipeline.py).
Images are resized/rescaled such that the shortest side is at least 800 pixels and the largest side at most 1333 pixels, and normalized across the RGB channels with the ImageNet mean (0.485, 0.456, 0.406) and standard deviation (0.229, 0.224, 0.225).
### Training
The model was trained for 300 epochs on 16 V100 GPUs. This takes 3 days, with 4 images per GPU (hence a total batch size of 64).
## Evaluation results
This model achieves an AP (average precision) of **43.5** on COCO 2017 validation. For more details regarding evaluation results, we refer to table 1 of the original paper.
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2005-12872,
author = {Nicolas Carion and
Francisco Massa and
Gabriel Synnaeve and
Nicolas Usunier and
Alexander Kirillov and
Sergey Zagoruyko},
title = {End-to-End Object Detection with Transformers},
journal = {CoRR},
volume = {abs/2005.12872},
year = {2020},
url = {https://arxiv.org/abs/2005.12872},
archivePrefix = {arXiv},
eprint = {2005.12872},
timestamp = {Thu, 28 May 2020 17:38:09 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2005-12872.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["object-detection", "vision"], "datasets": ["coco"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/savanna.jpg", "example_title": "Savanna"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/football-match.jpg", "example_title": "Football Match"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/airport.jpg", "example_title": "Airport"}]}
|
facebook/detr-resnet-101
| null |
[
"transformers",
"pytorch",
"safetensors",
"detr",
"object-detection",
"vision",
"dataset:coco",
"arxiv:2005.12872",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-segmentation
|
transformers
|
# DETR (End-to-End Object Detection) model with ResNet-50 backbone (dilated C5 stage)
DEtection TRansformer (DETR) model trained end-to-end on COCO 2017 panoptic (118k annotated images). It was introduced in the paper [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872) by Carion et al. and first released in [this repository](https://github.com/facebookresearch/detr).
Disclaimer: The team releasing DETR did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
The DETR model is an encoder-decoder transformer with a convolutional backbone. Two heads are added on top of the decoder outputs in order to perform object detection: a linear layer for the class labels and a MLP (multi-layer perceptron) for the bounding boxes. The model uses so-called object queries to detect objects in an image. Each object query looks for a particular object in the image. For COCO, the number of object queries is set to 100.
The model is trained using a "bipartite matching loss": one compares the predicted classes + bounding boxes of each of the N = 100 object queries to the ground truth annotations, padded up to the same length N (so if an image only contains 4 objects, 96 annotations will just have a "no object" as class and "no bounding box" as bounding box). The Hungarian matching algorithm is used to create an optimal one-to-one mapping between each of the N queries and each of the N annotations. Next, standard cross-entropy (for the classes) and a linear combination of the L1 and generalized IoU loss (for the bounding boxes) are used to optimize the parameters of the model.
DETR can be naturally extended to perform panoptic segmentation, by adding a mask head on top of the decoder outputs.
## Intended uses & limitations
You can use the raw model for panoptic segmentation. See the [model hub](https://huggingface.co/models?search=facebook/detr) to look for all available DETR models.
### How to use
Here is how to use this model:
```python
from transformers import DetrFeatureExtractor, DetrForSegmentation
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = DetrFeatureExtractor.from_pretrained('facebook/detr-resnet-50-dc5-panoptic')
model = DetrForSegmentation.from_pretrained('facebook/detr-resnet-50-dc5-panoptic')
inputs = feature_extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
# model predicts COCO classes, bounding boxes, and masks
logits = outputs.logits
bboxes = outputs.pred_boxes
masks = outputs.pred_masks
```
Currently, both the feature extractor and model support PyTorch.
## Training data
The DETR model was trained on [COCO 2017 panoptic](https://cocodataset.org/#download), a dataset consisting of 118k/5k annotated images for training/validation respectively.
## Training procedure
### Preprocessing
The exact details of preprocessing of images during training/validation can be found [here](https://github.com/facebookresearch/detr/blob/master/datasets/coco_panoptic.py).
Images are resized/rescaled such that the shortest side is at least 800 pixels and the largest side at most 1333 pixels, and normalized across the RGB channels with the ImageNet mean (0.485, 0.456, 0.406) and standard deviation (0.229, 0.224, 0.225).
### Training
The model was trained for 300 epochs on 16 V100 GPUs. This takes 3 days, with 4 images per GPU (hence a total batch size of 64).
## Evaluation results
This model achieves the following results on COCO 2017 validation: a box AP (average precision) of **40.2**, a segmentation AP (average precision) of **31.9** and a PQ (panoptic quality) of **44.6**.
For more details regarding evaluation results, we refer to table 5 of the original paper.
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2005-12872,
author = {Nicolas Carion and
Francisco Massa and
Gabriel Synnaeve and
Nicolas Usunier and
Alexander Kirillov and
Sergey Zagoruyko},
title = {End-to-End Object Detection with Transformers},
journal = {CoRR},
volume = {abs/2005.12872},
year = {2020},
url = {https://arxiv.org/abs/2005.12872},
archivePrefix = {arXiv},
eprint = {2005.12872},
timestamp = {Thu, 28 May 2020 17:38:09 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2005-12872.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["image-segmentation"], "datasets": ["coco"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/dog-cat.jpg", "example_title": "Dog & Cat"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/construction-site.jpg", "example_title": "Construction Site"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/apple-orange.jpg", "example_title": "Apple & Orange"}]}
|
facebook/detr-resnet-50-dc5-panoptic
| null |
[
"transformers",
"pytorch",
"safetensors",
"detr",
"image-segmentation",
"dataset:coco",
"arxiv:2005.12872",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
object-detection
|
transformers
|
# DETR (End-to-End Object Detection) model with ResNet-50 backbone (dilated C5 stage)
DEtection TRansformer (DETR) model trained end-to-end on COCO 2017 object detection (118k annotated images). It was introduced in the paper [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872) by Carion et al. and first released in [this repository](https://github.com/facebookresearch/detr).
Disclaimer: The team releasing DETR did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
The DETR model is an encoder-decoder transformer with a convolutional backbone. Two heads are added on top of the decoder outputs in order to perform object detection: a linear layer for the class labels and a MLP (multi-layer perceptron) for the bounding boxes. The model uses so-called object queries to detect objects in an image. Each object query looks for a particular object in the image. For COCO, the number of object queries is set to 100.
The model is trained using a "bipartite matching loss": one compares the predicted classes + bounding boxes of each of the N = 100 object queries to the ground truth annotations, padded up to the same length N (so if an image only contains 4 objects, 96 annotations will just have a "no object" as class and "no bounding box" as bounding box). The Hungarian matching algorithm is used to create an optimal one-to-one mapping between each of the N queries and each of the N annotations. Next, standard cross-entropy (for the classes) and a linear combination of the L1 and generalized IoU loss (for the bounding boxes) are used to optimize the parameters of the model.
## Intended uses & limitations
You can use the raw model for object detection. See the [model hub](https://huggingface.co/models?search=facebook/detr) to look for all available DETR models.
### How to use
Here is how to use this model:
```python
from transformers import DetrFeatureExtractor, DetrForObjectDetection
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = DetrFeatureExtractor.from_pretrained('facebook/detr-resnet-50-dc5')
model = DetrForObjectDetection.from_pretrained('facebook/detr-resnet-50-dc5')
inputs = feature_extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
# model predicts bounding boxes and corresponding COCO classes
logits = outputs.logits
bboxes = outputs.pred_boxes
```
Currently, both the feature extractor and model support PyTorch.
## Training data
The DETR model was trained on [COCO 2017 object detection](https://cocodataset.org/#download), a dataset consisting of 118k/5k annotated images for training/validation respectively.
## Training procedure
### Preprocessing
The exact details of preprocessing of images during training/validation can be found [here](https://github.com/google-research/vision_transformer/blob/master/vit_jax/input_pipeline.py).
Images are resized/rescaled such that the shortest side is at least 800 pixels and the largest side at most 1333 pixels, and normalized across the RGB channels with the ImageNet mean (0.485, 0.456, 0.406) and standard deviation (0.229, 0.224, 0.225).
### Training
The model was trained for 300 epochs on 16 V100 GPUs. This takes 3 days, with 4 images per GPU (hence a total batch size of 64).
## Evaluation results
This model achieves an AP (average precision) of **43.3** on COCO 2017 validation. For more details regarding evaluation results, we refer to table 1 of the original paper.
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2005-12872,
author = {Nicolas Carion and
Francisco Massa and
Gabriel Synnaeve and
Nicolas Usunier and
Alexander Kirillov and
Sergey Zagoruyko},
title = {End-to-End Object Detection with Transformers},
journal = {CoRR},
volume = {abs/2005.12872},
year = {2020},
url = {https://arxiv.org/abs/2005.12872},
archivePrefix = {arXiv},
eprint = {2005.12872},
timestamp = {Thu, 28 May 2020 17:38:09 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2005-12872.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["object-detection", "vision"], "datasets": ["coco"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/savanna.jpg", "example_title": "Savanna"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/football-match.jpg", "example_title": "Football Match"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/airport.jpg", "example_title": "Airport"}]}
|
facebook/detr-resnet-50-dc5
| null |
[
"transformers",
"pytorch",
"safetensors",
"detr",
"object-detection",
"vision",
"dataset:coco",
"arxiv:2005.12872",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-segmentation
|
transformers
|
# DETR (End-to-End Object Detection) model with ResNet-50 backbone
DEtection TRansformer (DETR) model trained end-to-end on COCO 2017 panoptic (118k annotated images). It was introduced in the paper [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872) by Carion et al. and first released in [this repository](https://github.com/facebookresearch/detr).
Disclaimer: The team releasing DETR did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
The DETR model is an encoder-decoder transformer with a convolutional backbone. Two heads are added on top of the decoder outputs in order to perform object detection: a linear layer for the class labels and a MLP (multi-layer perceptron) for the bounding boxes. The model uses so-called object queries to detect objects in an image. Each object query looks for a particular object in the image. For COCO, the number of object queries is set to 100.
The model is trained using a "bipartite matching loss": one compares the predicted classes + bounding boxes of each of the N = 100 object queries to the ground truth annotations, padded up to the same length N (so if an image only contains 4 objects, 96 annotations will just have a "no object" as class and "no bounding box" as bounding box). The Hungarian matching algorithm is used to create an optimal one-to-one mapping between each of the N queries and each of the N annotations. Next, standard cross-entropy (for the classes) and a linear combination of the L1 and generalized IoU loss (for the bounding boxes) are used to optimize the parameters of the model.
DETR can be naturally extended to perform panoptic segmentation, by adding a mask head on top of the decoder outputs.
## Intended uses & limitations
You can use the raw model for panoptic segmentation. See the [model hub](https://huggingface.co/models?search=facebook/detr) to look for all available DETR models.
### How to use
Here is how to use this model:
```python
import io
import requests
from PIL import Image
import torch
import numpy
from transformers import DetrFeatureExtractor, DetrForSegmentation
from transformers.models.detr.feature_extraction_detr import rgb_to_id
url = "http://images.cocodataset.org/val2017/000000039769.jpg"
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = DetrFeatureExtractor.from_pretrained("facebook/detr-resnet-50-panoptic")
model = DetrForSegmentation.from_pretrained("facebook/detr-resnet-50-panoptic")
# prepare image for the model
inputs = feature_extractor(images=image, return_tensors="pt")
# forward pass
outputs = model(**inputs)
# use the `post_process_panoptic` method of `DetrFeatureExtractor` to convert to COCO format
processed_sizes = torch.as_tensor(inputs["pixel_values"].shape[-2:]).unsqueeze(0)
result = feature_extractor.post_process_panoptic(outputs, processed_sizes)[0]
# the segmentation is stored in a special-format png
panoptic_seg = Image.open(io.BytesIO(result["png_string"]))
panoptic_seg = numpy.array(panoptic_seg, dtype=numpy.uint8)
# retrieve the ids corresponding to each mask
panoptic_seg_id = rgb_to_id(panoptic_seg)
```
Currently, both the feature extractor and model support PyTorch.
## Training data
The DETR model was trained on [COCO 2017 panoptic](https://cocodataset.org/#download), a dataset consisting of 118k/5k annotated images for training/validation respectively.
## Training procedure
### Preprocessing
The exact details of preprocessing of images during training/validation can be found [here](https://github.com/facebookresearch/detr/blob/master/datasets/coco_panoptic.py).
Images are resized/rescaled such that the shortest side is at least 800 pixels and the largest side at most 1333 pixels, and normalized across the RGB channels with the ImageNet mean (0.485, 0.456, 0.406) and standard deviation (0.229, 0.224, 0.225).
### Training
The model was trained for 300 epochs on 16 V100 GPUs. This takes 3 days, with 4 images per GPU (hence a total batch size of 64).
## Evaluation results
This model achieves the following results on COCO 2017 validation: a box AP (average precision) of **38.8**, a segmentation AP (average precision) of **31.1** and a PQ (panoptic quality) of **43.4**.
For more details regarding evaluation results, we refer to table 5 of the original paper.
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2005-12872,
author = {Nicolas Carion and
Francisco Massa and
Gabriel Synnaeve and
Nicolas Usunier and
Alexander Kirillov and
Sergey Zagoruyko},
title = {End-to-End Object Detection with Transformers},
journal = {CoRR},
volume = {abs/2005.12872},
year = {2020},
url = {https://arxiv.org/abs/2005.12872},
archivePrefix = {arXiv},
eprint = {2005.12872},
timestamp = {Thu, 28 May 2020 17:38:09 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2005-12872.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["image-segmentation", "vision"], "datasets": ["coco"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/football-match.jpg", "example_title": "Football Match"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/dog-cat.jpg", "example_title": "Dog & Cat"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/construction-site.jpg", "example_title": "Construction Site"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/apple-orange.jpg", "example_title": "Apple & Orange"}]}
|
facebook/detr-resnet-50-panoptic
| null |
[
"transformers",
"pytorch",
"detr",
"image-segmentation",
"vision",
"dataset:coco",
"arxiv:2005.12872",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
object-detection
|
transformers
|
# DETR (End-to-End Object Detection) model with ResNet-50 backbone
DEtection TRansformer (DETR) model trained end-to-end on COCO 2017 object detection (118k annotated images). It was introduced in the paper [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872) by Carion et al. and first released in [this repository](https://github.com/facebookresearch/detr).
Disclaimer: The team releasing DETR did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
The DETR model is an encoder-decoder transformer with a convolutional backbone. Two heads are added on top of the decoder outputs in order to perform object detection: a linear layer for the class labels and a MLP (multi-layer perceptron) for the bounding boxes. The model uses so-called object queries to detect objects in an image. Each object query looks for a particular object in the image. For COCO, the number of object queries is set to 100.
The model is trained using a "bipartite matching loss": one compares the predicted classes + bounding boxes of each of the N = 100 object queries to the ground truth annotations, padded up to the same length N (so if an image only contains 4 objects, 96 annotations will just have a "no object" as class and "no bounding box" as bounding box). The Hungarian matching algorithm is used to create an optimal one-to-one mapping between each of the N queries and each of the N annotations. Next, standard cross-entropy (for the classes) and a linear combination of the L1 and generalized IoU loss (for the bounding boxes) are used to optimize the parameters of the model.
## Intended uses & limitations
You can use the raw model for object detection. See the [model hub](https://huggingface.co/models?search=facebook/detr) to look for all available DETR models.
### How to use
Here is how to use this model:
```python
from transformers import DetrImageProcessor, DetrForObjectDetection
import torch
from PIL import Image
import requests
url = "http://images.cocodataset.org/val2017/000000039769.jpg"
image = Image.open(requests.get(url, stream=True).raw)
# you can specify the revision tag if you don't want the timm dependency
processor = DetrImageProcessor.from_pretrained("facebook/detr-resnet-50", revision="no_timm")
model = DetrForObjectDetection.from_pretrained("facebook/detr-resnet-50", revision="no_timm")
inputs = processor(images=image, return_tensors="pt")
outputs = model(**inputs)
# convert outputs (bounding boxes and class logits) to COCO API
# let's only keep detections with score > 0.9
target_sizes = torch.tensor([image.size[::-1]])
results = processor.post_process_object_detection(outputs, target_sizes=target_sizes, threshold=0.9)[0]
for score, label, box in zip(results["scores"], results["labels"], results["boxes"]):
box = [round(i, 2) for i in box.tolist()]
print(
f"Detected {model.config.id2label[label.item()]} with confidence "
f"{round(score.item(), 3)} at location {box}"
)
```
This should output:
```
Detected remote with confidence 0.998 at location [40.16, 70.81, 175.55, 117.98]
Detected remote with confidence 0.996 at location [333.24, 72.55, 368.33, 187.66]
Detected couch with confidence 0.995 at location [-0.02, 1.15, 639.73, 473.76]
Detected cat with confidence 0.999 at location [13.24, 52.05, 314.02, 470.93]
Detected cat with confidence 0.999 at location [345.4, 23.85, 640.37, 368.72]
```
Currently, both the feature extractor and model support PyTorch.
## Training data
The DETR model was trained on [COCO 2017 object detection](https://cocodataset.org/#download), a dataset consisting of 118k/5k annotated images for training/validation respectively.
## Training procedure
### Preprocessing
The exact details of preprocessing of images during training/validation can be found [here](https://github.com/google-research/vision_transformer/blob/master/vit_jax/input_pipeline.py).
Images are resized/rescaled such that the shortest side is at least 800 pixels and the largest side at most 1333 pixels, and normalized across the RGB channels with the ImageNet mean (0.485, 0.456, 0.406) and standard deviation (0.229, 0.224, 0.225).
### Training
The model was trained for 300 epochs on 16 V100 GPUs. This takes 3 days, with 4 images per GPU (hence a total batch size of 64).
## Evaluation results
This model achieves an AP (average precision) of **42.0** on COCO 2017 validation. For more details regarding evaluation results, we refer to table 1 of the original paper.
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2005-12872,
author = {Nicolas Carion and
Francisco Massa and
Gabriel Synnaeve and
Nicolas Usunier and
Alexander Kirillov and
Sergey Zagoruyko},
title = {End-to-End Object Detection with Transformers},
journal = {CoRR},
volume = {abs/2005.12872},
year = {2020},
url = {https://arxiv.org/abs/2005.12872},
archivePrefix = {arXiv},
eprint = {2005.12872},
timestamp = {Thu, 28 May 2020 17:38:09 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2005-12872.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["object-detection", "vision"], "datasets": ["coco"], "widget": [{"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/savanna.jpg", "example_title": "Savanna"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/football-match.jpg", "example_title": "Football Match"}, {"src": "https://huggingface.co/datasets/mishig/sample_images/resolve/main/airport.jpg", "example_title": "Airport"}]}
|
facebook/detr-resnet-50
| null |
[
"transformers",
"pytorch",
"detr",
"object-detection",
"vision",
"dataset:coco",
"arxiv:2005.12872",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
feature-extraction
|
transformers
|
# Vision Transformer (base-sized model, patch size 16) trained using DINO
Vision Transformer (ViT) model trained using the DINO method. It was introduced in the paper [Emerging Properties in Self-Supervised Vision Transformers](https://arxiv.org/abs/2104.14294) by Mathilde Caron, Hugo Touvron, Ishan Misra, Hervé Jégou, Julien Mairal, Piotr Bojanowski, Armand Joulin and first released in [this repository](https://github.com/facebookresearch/dino).
Disclaimer: The team releasing DINO did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
The Vision Transformer (ViT) is a transformer encoder model (BERT-like) pretrained on a large collection of images in a self-supervised fashion, namely ImageNet-1k, at a resolution of 224x224 pixels.
Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded. One also adds a [CLS] token to the beginning of a sequence to use it for classification tasks. One also adds absolute position embeddings before feeding the sequence to the layers of the Transformer encoder.
Note that this model does not include any fine-tuned heads.
By pre-training the model, it learns an inner representation of images that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled images for instance, you can train a standard classifier by placing a linear layer on top of the pre-trained encoder. One typically places a linear layer on top of the [CLS] token, as the last hidden state of this token can be seen as a representation of an entire image.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=google/vit) to look for
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model:
```python
from transformers import ViTImageProcessor, ViTModel
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
processor = ViTImageProcessor.from_pretrained('facebook/dino-vitb16')
model = ViTModel.from_pretrained('facebook/dino-vitb16')
inputs = processor(images=image, return_tensors="pt")
outputs = model(**inputs)
last_hidden_states = outputs.last_hidden_state
```
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2104-14294,
author = {Mathilde Caron and
Hugo Touvron and
Ishan Misra and
Herv{\'{e}} J{\'{e}}gou and
Julien Mairal and
Piotr Bojanowski and
Armand Joulin},
title = {Emerging Properties in Self-Supervised Vision Transformers},
journal = {CoRR},
volume = {abs/2104.14294},
year = {2021},
url = {https://arxiv.org/abs/2104.14294},
archivePrefix = {arXiv},
eprint = {2104.14294},
timestamp = {Tue, 04 May 2021 15:12:43 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2104-14294.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["dino", "vision"], "datasets": ["imagenet-1k"]}
|
facebook/dino-vitb16
| null |
[
"transformers",
"pytorch",
"tf",
"vit",
"feature-extraction",
"dino",
"vision",
"dataset:imagenet-1k",
"arxiv:2104.14294",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
feature-extraction
|
transformers
|
# Vision Transformer (base-sized model, patch size 8) trained using DINO
Vision Transformer (ViT) model trained using the DINO method. It was introduced in the paper [Emerging Properties in Self-Supervised Vision Transformers](https://arxiv.org/abs/2104.14294) by Mathilde Caron, Hugo Touvron, Ishan Misra, Hervé Jégou, Julien Mairal, Piotr Bojanowski, Armand Joulin and first released in [this repository](https://github.com/facebookresearch/dino).
Disclaimer: The team releasing DINO did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
The Vision Transformer (ViT) is a transformer encoder model (BERT-like) pretrained on a large collection of images in a self-supervised fashion, namely ImageNet-1k, at a resolution of 224x224 pixels.
Images are presented to the model as a sequence of fixed-size patches (resolution 8x8), which are linearly embedded. One also adds a [CLS] token to the beginning of a sequence to use it for classification tasks. One also adds absolute position embeddings before feeding the sequence to the layers of the Transformer encoder.
Note that this model does not include any fine-tuned heads.
By pre-training the model, it learns an inner representation of images that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled images for instance, you can train a standard classifier by placing a linear layer on top of the pre-trained encoder. One typically places a linear layer on top of the [CLS] token, as the last hidden state of this token can be seen as a representation of an entire image.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=google/vit) to look for
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model:
```python
from transformers import ViTImageProcessor, ViTModel
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
processor = ViTImageProcessor.from_pretrained('facebook/dino-vitb8')
model = ViTModel.from_pretrained('facebook/dino-vitb8')
inputs = processor(images=image, return_tensors="pt")
outputs = model(**inputs)
last_hidden_states = outputs.last_hidden_state
```
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2104-14294,
author = {Mathilde Caron and
Hugo Touvron and
Ishan Misra and
Herv{\'{e}} J{\'{e}}gou and
Julien Mairal and
Piotr Bojanowski and
Armand Joulin},
title = {Emerging Properties in Self-Supervised Vision Transformers},
journal = {CoRR},
volume = {abs/2104.14294},
year = {2021},
url = {https://arxiv.org/abs/2104.14294},
archivePrefix = {arXiv},
eprint = {2104.14294},
timestamp = {Tue, 04 May 2021 15:12:43 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2104-14294.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["dino", "vision"], "datasets": ["imagenet-1k"]}
|
facebook/dino-vitb8
| null |
[
"transformers",
"pytorch",
"vit",
"feature-extraction",
"dino",
"vision",
"dataset:imagenet-1k",
"arxiv:2104.14294",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
feature-extraction
|
transformers
|
# Vision Transformer (small-sized model, patch size 16) trained using DINO
Vision Transformer (ViT) model trained using the DINO method. It was introduced in the paper [Emerging Properties in Self-Supervised Vision Transformers](https://arxiv.org/abs/2104.14294) by Mathilde Caron, Hugo Touvron, Ishan Misra, Hervé Jégou, Julien Mairal, Piotr Bojanowski, Armand Joulin and first released in [this repository](https://github.com/facebookresearch/dino).
Disclaimer: The team releasing DINO did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
The Vision Transformer (ViT) is a transformer encoder model (BERT-like) pretrained on a large collection of images in a self-supervised fashion, namely ImageNet-1k, at a resolution of 224x224 pixels.
Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded. One also adds a [CLS] token to the beginning of a sequence to use it for classification tasks. One also adds absolute position embeddings before feeding the sequence to the layers of the Transformer encoder.
Note that this model does not include any fine-tuned heads.
By pre-training the model, it learns an inner representation of images that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled images for instance, you can train a standard classifier by placing a linear layer on top of the pre-trained encoder. One typically places a linear layer on top of the [CLS] token, as the last hidden state of this token can be seen as a representation of an entire image.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=google/vit) to look for
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model:
```python
from transformers import ViTImageProcessor, ViTModel
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
processor = ViTImageProcessor.from_pretrained('facebook/dino-vits16')
model = ViTModel.from_pretrained('facebook/dino-vits16')
inputs = processor(images=image, return_tensors="pt")
outputs = model(**inputs)
last_hidden_states = outputs.last_hidden_state
```
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2104-14294,
author = {Mathilde Caron and
Hugo Touvron and
Ishan Misra and
Herv{\'{e}} J{\'{e}}gou and
Julien Mairal and
Piotr Bojanowski and
Armand Joulin},
title = {Emerging Properties in Self-Supervised Vision Transformers},
journal = {CoRR},
volume = {abs/2104.14294},
year = {2021},
url = {https://arxiv.org/abs/2104.14294},
archivePrefix = {arXiv},
eprint = {2104.14294},
timestamp = {Tue, 04 May 2021 15:12:43 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2104-14294.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["dino", "vision"], "datasets": ["imagenet-1k"]}
|
facebook/dino-vits16
| null |
[
"transformers",
"pytorch",
"vit",
"feature-extraction",
"dino",
"vision",
"dataset:imagenet-1k",
"arxiv:2104.14294",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
feature-extraction
|
transformers
|
# Vision Transformer (small-sized model, patch size 8) trained using DINO
Vision Transformer (ViT) model trained using the DINO method. It was introduced in the paper [Emerging Properties in Self-Supervised Vision Transformers](https://arxiv.org/abs/2104.14294) by Mathilde Caron, Hugo Touvron, Ishan Misra, Hervé Jégou, Julien Mairal, Piotr Bojanowski, Armand Joulin and first released in [this repository](https://github.com/facebookresearch/dino).
Disclaimer: The team releasing DINO did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
The Vision Transformer (ViT) is a transformer encoder model (BERT-like) pretrained on a large collection of images in a self-supervised fashion, namely ImageNet-1k, at a resolution of 224x224 pixels.
Images are presented to the model as a sequence of fixed-size patches (resolution 8x8), which are linearly embedded. One also adds a [CLS] token to the beginning of a sequence to use it for classification tasks. One also adds absolute position embeddings before feeding the sequence to the layers of the Transformer encoder.
Note that this model does not include any fine-tuned heads.
By pre-training the model, it learns an inner representation of images that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled images for instance, you can train a standard classifier by placing a linear layer on top of the pre-trained encoder. One typically places a linear layer on top of the [CLS] token, as the last hidden state of this token can be seen as a representation of an entire image.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=google/vit) to look for
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model:
```python
from transformers import ViTImageProcessor, ViTModel
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
processor = ViTImageProcessor.from_pretrained('facebook/dino-vits8')
model = ViTModel.from_pretrained('facebook/dino-vits8')
inputs = processor(images=image, return_tensors="pt")
outputs = model(**inputs)
last_hidden_states = outputs.last_hidden_state
```
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2104-14294,
author = {Mathilde Caron and
Hugo Touvron and
Ishan Misra and
Herv{\'{e}} J{\'{e}}gou and
Julien Mairal and
Piotr Bojanowski and
Armand Joulin},
title = {Emerging Properties in Self-Supervised Vision Transformers},
journal = {CoRR},
volume = {abs/2104.14294},
year = {2021},
url = {https://arxiv.org/abs/2104.14294},
archivePrefix = {arXiv},
eprint = {2104.14294},
timestamp = {Tue, 04 May 2021 15:12:43 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2104-14294.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"license": "apache-2.0", "tags": ["dino", "vision"], "datasets": ["imagenet-1k"]}
|
facebook/dino-vits8
| null |
[
"transformers",
"pytorch",
"vit",
"feature-extraction",
"dino",
"vision",
"dataset:imagenet-1k",
"arxiv:2104.14294",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
# `dpr-ctx_encoder-multiset-base`
## Table of Contents
- [Model Details](#model-details)
- [How To Get Started With the Model](#how-to-get-started-with-the-model)
- [Uses](#uses)
- [Risks, Limitations and Biases](#risks-limitations-and-biases)
- [Training](#training)
- [Evaluation](#evaluation-results)
- [Environmental Impact](#environmental-impact)
- [Technical Specifications](#technical-specifications)
- [Citation Information](#citation-information)
- [Model Card Authors](#model-card-authors)
## Model Details
**Model Description:** [Dense Passage Retrieval (DPR)](https://github.com/facebookresearch/DPR) is a set of tools and models for state-of-the-art open-domain Q&A research. `dpr-ctx_encoder-multiset-base` is the context encoder trained using the [Natural Questions (NQ) dataset](https://huggingface.co/datasets/nq_open), [TriviaQA](https://huggingface.co/datasets/trivia_qa), [WebQuestions (WQ)](https://huggingface.co/datasets/web_questions), and [CuratedTREC (TREC)](https://huggingface.co/datasets/trec).
- **Developed by:** See [GitHub repo](https://github.com/facebookresearch/DPR) for model developers
- **Model Type:** BERT-based encoder
- **Language(s):** [CC-BY-NC-4.0](https://github.com/facebookresearch/DPR/blob/main/LICENSE), also see [Code of Conduct](https://github.com/facebookresearch/DPR/blob/main/CODE_OF_CONDUCT.md)
- **License:** English
- **Related Models:**
- [`dpr-question_encoder-multiset-base`](https://huggingface.co/facebook/dpr-question_encoder-multiset-base)
- [`dpr-reader-multiset-base`](https://huggingface.co/facebook/dpr-reader-multiset-base)
- [`dpr-question-encoder-single-nq-base`](https://huggingface.co/facebook/dpr-question_encoder-single-nq-base)
- [`dpr-reader-single-nq-base`](https://huggingface.co/facebook/dpr-reader-single-nq-base)
- [`dpr-ctx_encoder-single-nq-base`](https://huggingface.co/facebook/dpr-ctx_encoder-single-nq-base)
- **Resources for more information:**
- [Research Paper](https://arxiv.org/abs/2004.04906)
- [GitHub Repo](https://github.com/facebookresearch/DPR)
- [Hugging Face DPR docs](https://huggingface.co/docs/transformers/main/en/model_doc/dpr)
- [BERT Base Uncased Model Card](https://huggingface.co/bert-base-uncased)
## How to Get Started with the Model
Use the code below to get started with the model.
```python
from transformers import DPRContextEncoder, DPRContextEncoderTokenizer
tokenizer = DPRContextEncoderTokenizer.from_pretrained("facebook/dpr-ctx_encoder-multiset-base")
model = DPRContextEncoder.from_pretrained("facebook/dpr-ctx_encoder-multiset-base")
input_ids = tokenizer("Hello, is my dog cute ?", return_tensors="pt")["input_ids"]
embeddings = model(input_ids).pooler_output
```
## Uses
#### Direct Use
`dpr-ctx_encoder-multiset-base`, [`dpr-question_encoder-multiset-base`](https://huggingface.co/facebook/dpr-question_encoder-multiset-base), and [`dpr-reader-multiset-base`](https://huggingface.co/facebook/dpr-reader-multiset-base) can be used for the task of open-domain question answering.
#### Misuse and Out-of-scope Use
The model should not be used to intentionally create hostile or alienating environments for people. In addition, the set of DPR models was not trained to be factual or true representations of people or events, and therefore using the models to generate such content is out-of-scope for the abilities of this model.
## Risks, Limitations and Biases
**CONTENT WARNING: Readers should be aware this section may contain content that is disturbing, offensive, and can propogate historical and current stereotypes.**
Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al. (2021)](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)). Predictions generated by the model can include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.
## Training
#### Training Data
This model was trained using the following datasets:
- **[Natural Questions (NQ) dataset](https://huggingface.co/datasets/nq_open)** ([Lee et al., 2019](https://aclanthology.org/P19-1612/); [Kwiatkowski et al., 2019](https://aclanthology.org/Q19-1026/))
- **[TriviaQA](https://huggingface.co/datasets/trivia_qa)** ([Joshi et al., 2017](https://aclanthology.org/P17-1147/))
- **[WebQuestions (WQ)](https://huggingface.co/datasets/web_questions)** ([Berant et al., 2013](https://aclanthology.org/D13-1160/))
- **[CuratedTREC (TREC)](https://huggingface.co/datasets/trec)** ([Baudiš & Šedivý, 2015](https://www.aminer.cn/pub/599c7953601a182cd263079b/reading-wikipedia-to-answer-open-domain-questions))
#### Training Procedure
The training procedure is described in the [associated paper](https://arxiv.org/pdf/2004.04906.pdf):
> Given a collection of M text passages, the goal of our dense passage retriever (DPR) is to index all the passages in a low-dimensional and continuous space, such that it can retrieve efficiently the top k passages relevant to the input question for the reader at run-time.
> Our dense passage retriever (DPR) uses a dense encoder EP(·) which maps any text passage to a d- dimensional real-valued vectors and builds an index for all the M passages that we will use for retrieval. At run-time, DPR applies a different encoder EQ(·) that maps the input question to a d-dimensional vector, and retrieves k passages of which vectors are the closest to the question vector.
The authors report that for encoders, they used two independent BERT ([Devlin et al., 2019](https://aclanthology.org/N19-1423/)) networks (base, un-cased) and use FAISS ([Johnson et al., 2017](https://arxiv.org/abs/1702.08734)) during inference time to encode and index passages. See the paper for further details on training, including encoders, inference, positive and negative passages, and in-batch negatives.
## Evaluation
The following evaluation information is extracted from the [associated paper](https://arxiv.org/pdf/2004.04906.pdf).
#### Testing Data, Factors and Metrics
The model developers report the performance of the model on five QA datasets, using the top-k accuracy (k ∈ {20, 100}). The datasets were [NQ](https://huggingface.co/datasets/nq_open), [TriviaQA](https://huggingface.co/datasets/trivia_qa), [WebQuestions (WQ)](https://huggingface.co/datasets/web_questions), [CuratedTREC (TREC)](https://huggingface.co/datasets/trec), and [SQuAD v1.1](https://huggingface.co/datasets/squad).
#### Results
| | Top 20 | | | | | Top 100| | | | |
|:----:|:------:|:---------:|:--:|:----:|:-----:|:------:|:---------:|:--:|:----:|:-----:|
| | NQ | TriviaQA | WQ | TREC | SQuAD | NQ | TriviaQA | WQ | TREC | SQuAD |
| | 79.4 | 78.8 |75.0| 89.1 | 51.6 | 86.0 | 84.7 |82.9| 93.9 | 67.6 |
## Environmental Impact
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). We present the hardware type and based on the [associated paper](https://arxiv.org/abs/2004.04906).
- **Hardware Type:** 8 32GB GPUs
- **Hours used:** Unknown
- **Cloud Provider:** Unknown
- **Compute Region:** Unknown
- **Carbon Emitted:** Unknown
## Technical Specifications
See the [associated paper](https://arxiv.org/abs/2004.04906) for details on the modeling architecture, objective, compute infrastructure, and training details.
## Citation Information
```bibtex
@inproceedings{karpukhin-etal-2020-dense,
title = "Dense Passage Retrieval for Open-Domain Question Answering",
author = "Karpukhin, Vladimir and Oguz, Barlas and Min, Sewon and Lewis, Patrick and Wu, Ledell and Edunov, Sergey and Chen, Danqi and Yih, Wen-tau",
booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
month = nov,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/2020.emnlp-main.550",
doi = "10.18653/v1/2020.emnlp-main.550",
pages = "6769--6781",
}
```
## Model Card Authors
This model card was written by the team at Hugging Face.
|
{"language": "en", "license": "cc-by-nc-4.0", "tags": ["dpr"], "datasets": ["nq_open"], "inference": false}
|
facebook/dpr-ctx_encoder-multiset-base
| null |
[
"transformers",
"pytorch",
"tf",
"dpr",
"en",
"dataset:nq_open",
"arxiv:2004.04906",
"arxiv:1702.08734",
"arxiv:1910.09700",
"license:cc-by-nc-4.0",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
# `dpr-ctx_encoder-single-nq-base`
## Table of Contents
- [Model Details](#model-details)
- [How To Get Started With the Model](#how-to-get-started-with-the-model)
- [Uses](#uses)
- [Risks, Limitations and Biases](#risks-limitations-and-biases)
- [Training](#training)
- [Evaluation](#evaluation-results)
- [Environmental Impact](#environmental-impact)
- [Technical Specifications](#technical-specifications)
- [Citation Information](#citation-information)
- [Model Card Authors](#model-card-authors)
## Model Details
**Model Description:** [Dense Passage Retrieval (DPR)](https://github.com/facebookresearch/DPR) is a set of tools and models for state-of-the-art open-domain Q&A research. `dpr-ctx_encoder-single-nq-base` is the Context Encoder trained using the [Natural Questions (NQ) dataset](https://huggingface.co/datasets/nq_open) ([Lee et al., 2019](https://aclanthology.org/P19-1612/); [Kwiatkowski et al., 2019](https://aclanthology.org/Q19-1026/)).
- **Developed by:** See [GitHub repo](https://github.com/facebookresearch/DPR) for model developers
- **Model Type:** BERT-based encoder
- **Language(s):** [CC-BY-NC-4.0](https://github.com/facebookresearch/DPR/blob/main/LICENSE), also see [Code of Conduct](https://github.com/facebookresearch/DPR/blob/main/CODE_OF_CONDUCT.md)
- **License:** English
- **Related Models:**
- [`dpr-question-encoder-single-nq-base`](https://huggingface.co/facebook/dpr-question_encoder-single-nq-base)
- [`dpr-reader-single-nq-base`](https://huggingface.co/facebook/dpr-reader-single-nq-base)
- [`dpr-ctx_encoder-multiset-base`](https://huggingface.co/facebook/dpr-ctx_encoder-multiset-base)
- [`dpr-question_encoder-multiset-base`](https://huggingface.co/facebook/dpr-question_encoder-multiset-base)
- [`dpr-reader-multiset-base`](https://huggingface.co/facebook/dpr-reader-multiset-base)
- **Resources for more information:**
- [Research Paper](https://arxiv.org/abs/2004.04906)
- [GitHub Repo](https://github.com/facebookresearch/DPR)
- [Hugging Face DPR docs](https://huggingface.co/docs/transformers/main/en/model_doc/dpr)
- [BERT Base Uncased Model Card](https://huggingface.co/bert-base-uncased)
## How to Get Started with the Model
Use the code below to get started with the model.
```python
>>> from transformers import DPRContextEncoder, DPRContextEncoderTokenizer
>>> tokenizer = DPRContextEncoderTokenizer.from_pretrained("facebook/dpr-ctx_encoder-single-nq-base")
>>> model = DPRContextEncoder.from_pretrained("facebook/dpr-ctx_encoder-single-nq-base")
>>> input_ids = tokenizer("Hello, is my dog cute ?", return_tensors="pt")["input_ids"]
>>> embeddings = model(input_ids).pooler_output
```
## Uses
#### Direct Use
`dpr-ctx_encoder-single-nq-base`, [`dpr-question-encoder-single-nq-base`](https://huggingface.co/facebook/dpr-question_encoder-single-nq-base), and [`dpr-reader-single-nq-base`](https://huggingface.co/facebook/dpr-reader-single-nq-base) can be used for the task of open-domain question answering.
#### Misuse and Out-of-scope Use
The model should not be used to intentionally create hostile or alienating environments for people. In addition, the set of DPR models was not trained to be factual or true representations of people or events, and therefore using the models to generate such content is out-of-scope for the abilities of this model.
## Risks, Limitations and Biases
**CONTENT WARNING: Readers should be aware this section may contain content that is disturbing, offensive, and can propogate historical and current stereotypes.**
Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al. (2021)](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)). Predictions generated by the model can include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.
## Training
#### Training Data
This model was trained using the [Natural Questions (NQ) dataset](https://huggingface.co/datasets/nq_open) ([Lee et al., 2019](https://aclanthology.org/P19-1612/); [Kwiatkowski et al., 2019](https://aclanthology.org/Q19-1026/)). The model authors write that:
> [The dataset] was designed for end-to-end question answering. The questions were mined from real Google search queries and the answers were spans in Wikipedia articles identified by annotators.
#### Training Procedure
The training procedure is described in the [associated paper](https://arxiv.org/pdf/2004.04906.pdf):
> Given a collection of M text passages, the goal of our dense passage retriever (DPR) is to index all the passages in a low-dimensional and continuous space, such that it can retrieve efficiently the top k passages relevant to the input question for the reader at run-time.
> Our dense passage retriever (DPR) uses a dense encoder EP(·) which maps any text passage to a d- dimensional real-valued vectors and builds an index for all the M passages that we will use for retrieval. At run-time, DPR applies a different encoder EQ(·) that maps the input question to a d-dimensional vector, and retrieves k passages of which vectors are the closest to the question vector.
The authors report that for encoders, they used two independent BERT ([Devlin et al., 2019](https://aclanthology.org/N19-1423/)) networks (base, un-cased) and use FAISS ([Johnson et al., 2017](https://arxiv.org/abs/1702.08734)) during inference time to encode and index passages. See the paper for further details on training, including encoders, inference, positive and negative passages, and in-batch negatives.
## Evaluation
The following evaluation information is extracted from the [associated paper](https://arxiv.org/pdf/2004.04906.pdf).
#### Testing Data, Factors and Metrics
The model developers report the performance of the model on five QA datasets, using the top-k accuracy (k ∈ {20, 100}). The datasets were [NQ](https://huggingface.co/datasets/nq_open), [TriviaQA](https://huggingface.co/datasets/trivia_qa), [WebQuestions (WQ)](https://huggingface.co/datasets/web_questions), [CuratedTREC (TREC)](https://huggingface.co/datasets/trec), and [SQuAD v1.1](https://huggingface.co/datasets/squad).
#### Results
| | Top 20 | | | | | Top 100| | | | |
|:----:|:------:|:---------:|:--:|:----:|:-----:|:------:|:---------:|:--:|:----:|:-----:|
| | NQ | TriviaQA | WQ | TREC | SQuAD | NQ | TriviaQA | WQ | TREC | SQuAD |
| | 78.4 | 79.4 |73.2| 79.8 | 63.2 | 85.4 | 85.0 |81.4| 89.1 | 77.2 |
## Environmental Impact
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). We present the hardware type and based on the [associated paper](https://arxiv.org/abs/2004.04906).
- **Hardware Type:** 8 32GB GPUs
- **Hours used:** Unknown
- **Cloud Provider:** Unknown
- **Compute Region:** Unknown
- **Carbon Emitted:** Unknown
## Technical Specifications
See the [associated paper](https://arxiv.org/abs/2004.04906) for details on the modeling architecture, objective, compute infrastructure, and training details.
## Citation Information
```bibtex
@inproceedings{karpukhin-etal-2020-dense,
title = "Dense Passage Retrieval for Open-Domain Question Answering",
author = "Karpukhin, Vladimir and Oguz, Barlas and Min, Sewon and Lewis, Patrick and Wu, Ledell and Edunov, Sergey and Chen, Danqi and Yih, Wen-tau",
booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
month = nov,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/2020.emnlp-main.550",
doi = "10.18653/v1/2020.emnlp-main.550",
pages = "6769--6781",
}
```
## Model Card Authors
This model card was written by the team at Hugging Face.
|
{"language": "en", "license": "cc-by-nc-4.0", "tags": ["dpr"], "datasets": ["nq_open"], "inference": false}
|
facebook/dpr-ctx_encoder-single-nq-base
| null |
[
"transformers",
"pytorch",
"tf",
"dpr",
"en",
"dataset:nq_open",
"arxiv:2004.04906",
"arxiv:1702.08734",
"arxiv:1910.09700",
"license:cc-by-nc-4.0",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
feature-extraction
|
transformers
|
# `dpr-question_encoder-multiset-base`
## Table of Contents
- [Model Details](#model-details)
- [How To Get Started With the Model](#how-to-get-started-with-the-model)
- [Uses](#uses)
- [Risks, Limitations and Biases](#risks-limitations-and-biases)
- [Training](#training)
- [Evaluation](#evaluation-results)
- [Environmental Impact](#environmental-impact)
- [Technical Specifications](#technical-specifications)
- [Citation Information](#citation-information)
- [Model Card Authors](#model-card-authors)
## Model Details
**Model Description:** [Dense Passage Retrieval (DPR)](https://github.com/facebookresearch/DPR) is a set of tools and models for state-of-the-art open-domain Q&A research. `dpr-question_encoder-multiset-base` is the question encoder trained using the [Natural Questions (NQ) dataset](https://huggingface.co/datasets/nq_open), [TriviaQA](https://huggingface.co/datasets/trivia_qa), [WebQuestions (WQ)](https://huggingface.co/datasets/web_questions), and [CuratedTREC (TREC)](https://huggingface.co/datasets/trec).
- **Developed by:** See [GitHub repo](https://github.com/facebookresearch/DPR) for model developers
- **Model Type:** BERT-based encoder
- **Language(s):** [CC-BY-NC-4.0](https://github.com/facebookresearch/DPR/blob/main/LICENSE), also see [Code of Conduct](https://github.com/facebookresearch/DPR/blob/main/CODE_OF_CONDUCT.md)
- **License:** English
- **Related Models:**
- [`dpr-ctx_encoder-multiset-base`](https://huggingface.co/facebook/dpr-ctx_encoder-multiset-base)
- [`dpr-reader-multiset-base`](https://huggingface.co/facebook/dpr-reader-multiset-base)
- [`dpr-ctx_encoder-single-nq-base`](https://huggingface.co/facebook/dpr-ctx_encoder-single-nq-base)
- [`dpr-question_encoder-single-nq-base`](https://huggingface.co/facebook/dpr-question_encoder-single-nq-base)
- [`dpr-reader-single-nq-base`](https://huggingface.co/facebook/dpr-reader-single-nq-base)
- **Resources for more information:**
- [Research Paper](https://arxiv.org/abs/2004.04906)
- [GitHub Repo](https://github.com/facebookresearch/DPR)
- [Hugging Face DPR docs](https://huggingface.co/docs/transformers/main/en/model_doc/dpr)
- [BERT Base Uncased Model Card](https://huggingface.co/bert-base-uncased)
## How to Get Started with the Model
Use the code below to get started with the model.
```python
from transformers import DPRQuestionEncoder, DPRQuestionEncoderTokenizer
tokenizer = DPRQuestionEncoderTokenizer.from_pretrained("facebook/dpr-question_encoder-multiset-base")
model = DPRQuestionEncoder.from_pretrained("facebook/dpr-question_encoder-multiset-base")
input_ids = tokenizer("Hello, is my dog cute ?", return_tensors="pt")["input_ids"]
embeddings = model(input_ids).pooler_output
```
## Uses
#### Direct Use
`dpr-question_encoder-multiset-base`, [`dpr-ctx_encoder-multiset-base`](https://huggingface.co/facebook/dpr-ctx_encoder-multiset-base), and [`dpr-reader-multiset-base`](https://huggingface.co/facebook/dpr-reader-multiset-base) can be used for the task of open-domain question answering.
#### Misuse and Out-of-scope Use
The model should not be used to intentionally create hostile or alienating environments for people. In addition, the set of DPR models was not trained to be factual or true representations of people or events, and therefore using the models to generate such content is out-of-scope for the abilities of this model.
## Risks, Limitations and Biases
**CONTENT WARNING: Readers should be aware this section may contain content that is disturbing, offensive, and can propogate historical and current stereotypes.**
Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al., 2021](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al., 2021](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)). Predictions generated by the model can include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.
## Training
#### Training Data
This model was trained using the following datasets:
- **[Natural Questions (NQ) dataset](https://huggingface.co/datasets/nq_open)** ([Lee et al., 2019](https://aclanthology.org/P19-1612/); [Kwiatkowski et al., 2019](https://aclanthology.org/Q19-1026/))
- **[TriviaQA](https://huggingface.co/datasets/trivia_qa)** ([Joshi et al., 2017](https://aclanthology.org/P17-1147/))
- **[WebQuestions (WQ)](https://huggingface.co/datasets/web_questions)** ([Berant et al., 2013](https://aclanthology.org/D13-1160/))
- **[CuratedTREC (TREC)](https://huggingface.co/datasets/trec)** ([Baudiš & Šedivý, 2015](https://www.aminer.cn/pub/599c7953601a182cd263079b/reading-wikipedia-to-answer-open-domain-questions))
#### Training Procedure
The training procedure is described in the [associated paper](https://arxiv.org/pdf/2004.04906.pdf):
> Given a collection of M text passages, the goal of our dense passage retriever (DPR) is to index all the passages in a low-dimensional and continuous space, such that it can retrieve efficiently the top k passages relevant to the input question for the reader at run-time.
> Our dense passage retriever (DPR) uses a dense encoder EP(·) which maps any text passage to a d- dimensional real-valued vectors and builds an index for all the M passages that we will use for retrieval. At run-time, DPR applies a different encoder EQ(·) that maps the input question to a d-dimensional vector, and retrieves k passages of which vectors are the closest to the question vector.
The authors report that for encoders, they used two independent BERT ([Devlin et al., 2019](https://aclanthology.org/N19-1423/)) networks (base, un-cased) and use FAISS ([Johnson et al., 2017](https://arxiv.org/abs/1702.08734)) during inference time to encode and index passages. See the paper for further details on training, including encoders, inference, positive and negative passages, and in-batch negatives.
## Evaluation
The following evaluation information is extracted from the [associated paper](https://arxiv.org/pdf/2004.04906.pdf).
#### Testing Data, Factors and Metrics
The model developers report the performance of the model on five QA datasets, using the top-k accuracy (k ∈ {20, 100}). The datasets were [NQ](https://huggingface.co/datasets/nq_open), [TriviaQA](https://huggingface.co/datasets/trivia_qa), [WebQuestions (WQ)](https://huggingface.co/datasets/web_questions), [CuratedTREC (TREC)](https://huggingface.co/datasets/trec), and [SQuAD v1.1](https://huggingface.co/datasets/squad).
#### Results
| | Top 20 | | | | | Top 100| | | | |
|:----:|:------:|:---------:|:--:|:----:|:-----:|:------:|:---------:|:--:|:----:|:-----:|
| | NQ | TriviaQA | WQ | TREC | SQuAD | NQ | TriviaQA | WQ | TREC | SQuAD |
| | 79.4 | 78.8 |75.0| 89.1 | 51.6 | 86.0 | 84.7 |82.9| 93.9 | 67.6 |
## Environmental Impact
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). We present the hardware type and based on the [associated paper](https://arxiv.org/abs/2004.04906).
- **Hardware Type:** 8 32GB GPUs
- **Hours used:** Unknown
- **Cloud Provider:** Unknown
- **Compute Region:** Unknown
- **Carbon Emitted:** Unknown
## Technical Specifications
See the [associated paper](https://arxiv.org/abs/2004.04906) for details on the modeling architecture, objective, compute infrastructure, and training details.
## Citation Information
```bibtex
@inproceedings{karpukhin-etal-2020-dense,
title = "Dense Passage Retrieval for Open-Domain Question Answering",
author = "Karpukhin, Vladimir and Oguz, Barlas and Min, Sewon and Lewis, Patrick and Wu, Ledell and Edunov, Sergey and Chen, Danqi and Yih, Wen-tau",
booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
month = nov,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/2020.emnlp-main.550",
doi = "10.18653/v1/2020.emnlp-main.550",
pages = "6769--6781",
}
```
## Model Card Authors
This model card was written by the team at Hugging Face.
|
{"language": "en", "license": "cc-by-nc-4.0", "tags": ["dpr"], "datasets": ["nq_open", "trivia_qa", "web_questions", "trec"], "inference": false}
|
facebook/dpr-question_encoder-multiset-base
| null |
[
"transformers",
"pytorch",
"tf",
"dpr",
"feature-extraction",
"en",
"dataset:nq_open",
"dataset:trivia_qa",
"dataset:web_questions",
"dataset:trec",
"arxiv:2004.04906",
"arxiv:1702.08734",
"arxiv:1910.09700",
"license:cc-by-nc-4.0",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
feature-extraction
|
transformers
|
# `dpr-question_encoder-single-nq-base`
## Table of Contents
- [Model Details](#model-details)
- [How To Get Started With the Model](#how-to-get-started-with-the-model)
- [Uses](#uses)
- [Risks, Limitations and Biases](#risks-limitations-and-biases)
- [Training](#training)
- [Evaluation](#evaluation-results)
- [Environmental Impact](#environmental-impact)
- [Technical Specifications](#technical-specifications)
- [Citation Information](#citation-information)
- [Model Card Authors](#model-card-authors)
## Model Details
**Model Description:** [Dense Passage Retrieval (DPR)](https://github.com/facebookresearch/DPR) is a set of tools and models for state-of-the-art open-domain Q&A research. `dpr-question_encoder-single-nq-base` is the question encoder trained using the [Natural Questions (NQ) dataset](https://huggingface.co/datasets/nq_open) ([Lee et al., 2019](https://aclanthology.org/P19-1612/); [Kwiatkowski et al., 2019](https://aclanthology.org/Q19-1026/)).
- **Developed by:** See [GitHub repo](https://github.com/facebookresearch/DPR) for model developers
- **Model Type:** BERT-based encoder
- **Language(s):** [CC-BY-NC-4.0](https://github.com/facebookresearch/DPR/blob/main/LICENSE), also see [Code of Conduct](https://github.com/facebookresearch/DPR/blob/main/CODE_OF_CONDUCT.md)
- **License:** English
- **Related Models:**
- [`dpr-ctx_encoder-single-nq-base`](https://huggingface.co/facebook/dpr-ctx_encoder-single-nq-base)
- [`dpr-reader-single-nq-base`](https://huggingface.co/facebook/dpr-reader-single-nq-base)
- [`dpr-ctx_encoder-multiset-base`](https://huggingface.co/facebook/dpr-ctx_encoder-multiset-base)
- [`dpr-question_encoder-multiset-base`](https://huggingface.co/facebook/dpr-question_encoder-multiset-base)
- [`dpr-reader-multiset-base`](https://huggingface.co/facebook/dpr-reader-multiset-base)
- **Resources for more information:**
- [Research Paper](https://arxiv.org/abs/2004.04906)
- [GitHub Repo](https://github.com/facebookresearch/DPR)
- [Hugging Face DPR docs](https://huggingface.co/docs/transformers/main/en/model_doc/dpr)
- [BERT Base Uncased Model Card](https://huggingface.co/bert-base-uncased)
## How to Get Started with the Model
Use the code below to get started with the model.
```python
from transformers import DPRQuestionEncoder, DPRQuestionEncoderTokenizer
tokenizer = DPRQuestionEncoderTokenizer.from_pretrained("facebook/dpr-question_encoder-single-nq-base")
model = DPRQuestionEncoder.from_pretrained("facebook/dpr-question_encoder-single-nq-base")
input_ids = tokenizer("Hello, is my dog cute ?", return_tensors="pt")["input_ids"]
embeddings = model(input_ids).pooler_output
```
## Uses
#### Direct Use
`dpr-question_encoder-single-nq-base`, [`dpr-ctx_encoder-single-nq-base`](https://huggingface.co/facebook/dpr-ctx_encoder-single-nq-base), and [`dpr-reader-single-nq-base`](https://huggingface.co/facebook/dpr-reader-single-nq-base) can be used for the task of open-domain question answering.
#### Misuse and Out-of-scope Use
The model should not be used to intentionally create hostile or alienating environments for people. In addition, the set of DPR models was not trained to be factual or true representations of people or events, and therefore using the models to generate such content is out-of-scope for the abilities of this model.
## Risks, Limitations and Biases
**CONTENT WARNING: Readers should be aware this section may contain content that is disturbing, offensive, and can propogate historical and current stereotypes.**
Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al., 2021](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al., 2021](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)). Predictions generated by the model can include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.
## Training
#### Training Data
This model was trained using the [Natural Questions (NQ) dataset](https://huggingface.co/datasets/nq_open) ([Lee et al., 2019](https://aclanthology.org/P19-1612/); [Kwiatkowski et al., 2019](https://aclanthology.org/Q19-1026/)). The model authors write that:
> [The dataset] was designed for end-to-end question answering. The questions were mined from real Google search queries and the answers were spans in Wikipedia articles identified by annotators.
#### Training Procedure
The training procedure is described in the [associated paper](https://arxiv.org/pdf/2004.04906.pdf):
> Given a collection of M text passages, the goal of our dense passage retriever (DPR) is to index all the passages in a low-dimensional and continuous space, such that it can retrieve efficiently the top k passages relevant to the input question for the reader at run-time.
> Our dense passage retriever (DPR) uses a dense encoder EP(·) which maps any text passage to a d- dimensional real-valued vectors and builds an index for all the M passages that we will use for retrieval. At run-time, DPR applies a different encoder EQ(·) that maps the input question to a d-dimensional vector, and retrieves k passages of which vectors are the closest to the question vector.
The authors report that for encoders, they used two independent BERT ([Devlin et al., 2019](https://aclanthology.org/N19-1423/)) networks (base, un-cased) and use FAISS ([Johnson et al., 2017](https://arxiv.org/abs/1702.08734)) during inference time to encode and index passages. See the paper for further details on training, including encoders, inference, positive and negative passages, and in-batch negatives.
## Evaluation
The following evaluation information is extracted from the [associated paper](https://arxiv.org/pdf/2004.04906.pdf).
#### Testing Data, Factors and Metrics
The model developers report the performance of the model on five QA datasets, using the top-k accuracy (k ∈ {20, 100}). The datasets were [NQ](https://huggingface.co/datasets/nq_open), [TriviaQA](https://huggingface.co/datasets/trivia_qa), [WebQuestions (WQ)](https://huggingface.co/datasets/web_questions), [CuratedTREC (TREC)](https://huggingface.co/datasets/trec), and [SQuAD v1.1](https://huggingface.co/datasets/squad).
#### Results
| | Top 20 | | | | | Top 100| | | | |
|:----:|:------:|:---------:|:--:|:----:|:-----:|:------:|:---------:|:--:|:----:|:-----:|
| | NQ | TriviaQA | WQ | TREC | SQuAD | NQ | TriviaQA | WQ | TREC | SQuAD |
| | 78.4 | 79.4 |73.2| 79.8 | 63.2 | 85.4 | 85.0 |81.4| 89.1 | 77.2 |
## Environmental Impact
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). We present the hardware type and based on the [associated paper](https://arxiv.org/abs/2004.04906).
- **Hardware Type:** 8 32GB GPUs
- **Hours used:** Unknown
- **Cloud Provider:** Unknown
- **Compute Region:** Unknown
- **Carbon Emitted:** Unknown
## Technical Specifications
See the [associated paper](https://arxiv.org/abs/2004.04906) for details on the modeling architecture, objective, compute infrastructure, and training details.
## Citation Information
```bibtex
@inproceedings{karpukhin-etal-2020-dense,
title = "Dense Passage Retrieval for Open-Domain Question Answering",
author = "Karpukhin, Vladimir and Oguz, Barlas and Min, Sewon and Lewis, Patrick and Wu, Ledell and Edunov, Sergey and Chen, Danqi and Yih, Wen-tau",
booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
month = nov,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/2020.emnlp-main.550",
doi = "10.18653/v1/2020.emnlp-main.550",
pages = "6769--6781",
}
```
## Model Card Authors
This model card was written by the team at Hugging Face.
|
{"language": "en", "license": "cc-by-nc-4.0", "tags": ["dpr"], "datasets": ["nq_open"], "inference": false}
|
facebook/dpr-question_encoder-single-nq-base
| null |
[
"transformers",
"pytorch",
"tf",
"dpr",
"feature-extraction",
"en",
"dataset:nq_open",
"arxiv:2004.04906",
"arxiv:1702.08734",
"arxiv:1910.09700",
"license:cc-by-nc-4.0",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
# `dpr-reader-multiset-base`
## Table of Contents
- [Model Details](#model-details)
- [How To Get Started With the Model](#how-to-get-started-with-the-model)
- [Uses](#uses)
- [Risks, Limitations and Biases](#risks-limitations-and-biases)
- [Training](#training)
- [Evaluation](#evaluation-results)
- [Environmental Impact](#environmental-impact)
- [Technical Specifications](#technical-specifications)
- [Citation Information](#citation-information)
- [Model Card Authors](#model-card-authors)
## Model Details
**Model Description:** [Dense Passage Retrieval (DPR)](https://github.com/facebookresearch/DPR) is a set of tools and models for state-of-the-art open-domain Q&A research. `dpr-reader-multiset-base` is the reader model trained using the [Natural Questions (NQ) dataset](https://huggingface.co/datasets/nq_open), [TriviaQA](https://huggingface.co/datasets/trivia_qa), [WebQuestions (WQ)](https://huggingface.co/datasets/web_questions), and [CuratedTREC (TREC)](https://huggingface.co/datasets/trec).
- **Developed by:** See [GitHub repo](https://github.com/facebookresearch/DPR) for model developers
- **Model Type:** BERT-based encoder
- **Language(s):** [CC-BY-NC-4.0](https://github.com/facebookresearch/DPR/blob/main/LICENSE), also see [Code of Conduct](https://github.com/facebookresearch/DPR/blob/main/CODE_OF_CONDUCT.md)
- **License:** English
- **Related Models:**
- [`dpr-question_encoder-multiset-base`](https://huggingface.co/facebook/dpr-question_encoder-multiset-base)
- [`dpr-ctx_encoder-multiset-base`](https://huggingface.co/facebook/dpr-ctx_encoder-multiset-base)
- [`dpr-question-encoder-single-nq-base`](https://huggingface.co/facebook/dpr-question_encoder-single-nq-base)
- [`dpr-reader-single-nq-base`](https://huggingface.co/facebook/dpr-reader-single-nq-base)
- [`dpr-ctx_encoder-single-nq-base`](https://huggingface.co/facebook/dpr-ctx_encoder-single-nq-base)
- **Resources for more information:**
- [Research Paper](https://arxiv.org/abs/2004.04906)
- [GitHub Repo](https://github.com/facebookresearch/DPR)
- [Hugging Face DPR docs](https://huggingface.co/docs/transformers/main/en/model_doc/dpr)
- [BERT Base Uncased Model Card](https://huggingface.co/bert-base-uncased)
## How to Get Started with the Model
Use the code below to get started with the model.
```python
from transformers import DPRReader, DPRReaderTokenizer
tokenizer = DPRReaderTokenizer.from_pretrained("facebook/dpr-reader-multiset-base")
model = DPRReader.from_pretrained("facebook/dpr-reader-multiset-base")
encoded_inputs = tokenizer(
questions=["What is love ?"],
titles=["Haddaway"],
texts=["'What Is Love' is a song recorded by the artist Haddaway"],
return_tensors="pt",
)
outputs = model(**encoded_inputs)
start_logits = outputs.start_logits
end_logits = outputs.end_logits
relevance_logits = outputs.relevance_logits
```
## Uses
#### Direct Use
`dpr-reader-multiset-base`, [`dpr-question_encoder-multiset-base`](https://huggingface.co/facebook/dpr-question_encoder-multiset-base), and [`dpr-ctx_encoder-multiset-base`](https://huggingface.co/facebook/dpr-ctx_encoder-multiset-base) can be used for the task of open-domain question answering.
#### Misuse and Out-of-scope Use
The model should not be used to intentionally create hostile or alienating environments for people. In addition, the set of DPR models was not trained to be factual or true representations of people or events, and therefore using the models to generate such content is out-of-scope for the abilities of this model.
## Risks, Limitations and Biases
**CONTENT WARNING: Readers should be aware this section may contain content that is disturbing, offensive, and can propogate historical and current stereotypes.**
Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al. (2021)](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)). Predictions generated by the model can include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.
## Training
#### Training Data
This model was trained using the following datasets:
- **[Natural Questions (NQ) dataset](https://huggingface.co/datasets/nq_open)** ([Lee et al., 2019](https://aclanthology.org/P19-1612/); [Kwiatkowski et al., 2019](https://aclanthology.org/Q19-1026/))
- **[TriviaQA](https://huggingface.co/datasets/trivia_qa)** ([Joshi et al., 2017](https://aclanthology.org/P17-1147/))
- **[WebQuestions (WQ)](https://huggingface.co/datasets/web_questions)** ([Berant et al., 2013](https://aclanthology.org/D13-1160/))
- **[CuratedTREC (TREC)](https://huggingface.co/datasets/trec)** ([Baudiš & Šedivý, 2015](https://www.aminer.cn/pub/599c7953601a182cd263079b/reading-wikipedia-to-answer-open-domain-questions))
#### Training Procedure
The training procedure is described in the [associated paper](https://arxiv.org/pdf/2004.04906.pdf):
> Given a collection of M text passages, the goal of our dense passage retriever (DPR) is to index all the passages in a low-dimensional and continuous space, such that it can retrieve efficiently the top k passages relevant to the input question for the reader at run-time.
> Our dense passage retriever (DPR) uses a dense encoder EP(·) which maps any text passage to a d- dimensional real-valued vectors and builds an index for all the M passages that we will use for retrieval. At run-time, DPR applies a different encoder EQ(·) that maps the input question to a d-dimensional vector, and retrieves k passages of which vectors are the closest to the question vector.
The authors report that for encoders, they used two independent BERT ([Devlin et al., 2019](https://aclanthology.org/N19-1423/)) networks (base, un-cased) and use FAISS ([Johnson et al., 2017](https://arxiv.org/abs/1702.08734)) during inference time to encode and index passages. See the paper for further details on training, including encoders, inference, positive and negative passages, and in-batch negatives.
## Evaluation
The following evaluation information is extracted from the [associated paper](https://arxiv.org/pdf/2004.04906.pdf).
#### Testing Data, Factors and Metrics
The model developers report the performance of the model on five QA datasets, using the top-k accuracy (k ∈ {20, 100}). The datasets were [NQ](https://huggingface.co/datasets/nq_open), [TriviaQA](https://huggingface.co/datasets/trivia_qa), [WebQuestions (WQ)](https://huggingface.co/datasets/web_questions), [CuratedTREC (TREC)](https://huggingface.co/datasets/trec), and [SQuAD v1.1](https://huggingface.co/datasets/squad).
#### Results
| | Top 20 | | | | | Top 100| | | | |
|:----:|:------:|:---------:|:--:|:----:|:-----:|:------:|:---------:|:--:|:----:|:-----:|
| | NQ | TriviaQA | WQ | TREC | SQuAD | NQ | TriviaQA | WQ | TREC | SQuAD |
| | 79.4 | 78.8 |75.0| 89.1 | 51.6 | 86.0 | 84.7 |82.9| 93.9 | 67.6 |
## Environmental Impact
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). We present the hardware type and based on the [associated paper](https://arxiv.org/abs/2004.04906).
- **Hardware Type:** 8 32GB GPUs
- **Hours used:** Unknown
- **Cloud Provider:** Unknown
- **Compute Region:** Unknown
- **Carbon Emitted:** Unknown
## Technical Specifications
See the [associated paper](https://arxiv.org/abs/2004.04906) for details on the modeling architecture, objective, compute infrastructure, and training details.
## Citation Information
```bibtex
@inproceedings{karpukhin-etal-2020-dense,
title = "Dense Passage Retrieval for Open-Domain Question Answering",
author = "Karpukhin, Vladimir and Oguz, Barlas and Min, Sewon and Lewis, Patrick and Wu, Ledell and Edunov, Sergey and Chen, Danqi and Yih, Wen-tau",
booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
month = nov,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/2020.emnlp-main.550",
doi = "10.18653/v1/2020.emnlp-main.550",
pages = "6769--6781",
}
```
## Model Card Authors
This model card was written by the team at Hugging Face.
|
{"language": "en", "license": "cc-by-nc-4.0", "tags": ["dpr"], "datasets": ["nq_open", "trivia_qa", "web_questions", "trec"], "inference": false}
|
facebook/dpr-reader-multiset-base
| null |
[
"transformers",
"pytorch",
"tf",
"dpr",
"en",
"dataset:nq_open",
"dataset:trivia_qa",
"dataset:web_questions",
"dataset:trec",
"arxiv:2004.04906",
"arxiv:1702.08734",
"arxiv:1910.09700",
"license:cc-by-nc-4.0",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
`dpr-reader-single-nq-base`
# Table of Contents
- [Model Details](#model-details)
- [How To Get Started With the Model](#how-to-get-started-with-the-model)
- [Uses](#uses)
- [Risks, Limitations and Biases](#risks-limitations-and-biases)
- [Training](#training)
- [Evaluation](#evaluation-results)
- [Environmental Impact](#environmental-impact)
- [Technical Specifications](#technical-specifications)
- [Citation Information](#citation-information)
- [Model Card Authors](#model-card-authors)
## Model Details
**Model Description:** [Dense Passage Retrieval (DPR)](https://github.com/facebookresearch/DPR) is a set of tools and models for state-of-the-art open-domain Q&A research. `dpr-reader-single-nq-base` is the reader model trained using the [Natural Questions (NQ) dataset](https://huggingface.co/datasets/nq_open) ([Lee et al., 2019](https://aclanthology.org/P19-1612/); [Kwiatkowski et al., 2019](https://aclanthology.org/Q19-1026/)).
- **Developed by:** See [GitHub repo](https://github.com/facebookresearch/DPR) for model developers
- **Model Type:** QA Reader Model
- **Language(s):** [CC-BY-NC-4.0](https://github.com/facebookresearch/DPR/blob/main/LICENSE), also see [Code of Conduct](https://github.com/facebookresearch/DPR/blob/main/CODE_OF_CONDUCT.md)
- **License:** English
- **Related Models:**
- [`dpr-ctx_encoder-single-nq-base`](https://huggingface.co/facebook/dpr-ctx_encoder-single-nq-base)
- [`dpr-question_encoder-single-nq-base`](https://huggingface.co/facebook/dpr-question_encoder-single-nq-base)
- [`dpr-ctx_encoder-multiset-base`](https://huggingface.co/facebook/dpr-ctx_encoder-multiset-base)
- [`dpr-question_encoder-multiset-base`](https://huggingface.co/facebook/dpr-question_encoder-multiset-base)
- [`dpr-reader-multiset-base`](https://huggingface.co/facebook/dpr-reader-multiset-base)
- **Resources for more information:**
- [Research Paper](https://arxiv.org/abs/2004.04906)
- [GitHub Repo](https://github.com/facebookresearch/DPR)
- [Hugging Face DPR docs](https://huggingface.co/docs/transformers/main/en/model_doc/dpr)
- [BERT Base Uncased Model Card](https://huggingface.co/bert-base-uncased)
## How to Get Started with the Model
Use the code below to get started with the model.
```python
from transformers import DPRReader, DPRReaderTokenizer
tokenizer = DPRReaderTokenizer.from_pretrained("facebook/dpr-reader-single-nq-base")
model = DPRReader.from_pretrained("facebook/dpr-reader-single-nq-base")
encoded_inputs = tokenizer(
questions=["What is love ?"],
titles=["Haddaway"],
texts=["'What Is Love' is a song recorded by the artist Haddaway"],
return_tensors="pt",
)
outputs = model(**encoded_inputs)
start_logits = outputs.start_logits
end_logits = outputs.end_logits
relevance_logits = outputs.relevance_logits
```
## Uses
#### Direct Use
`dpr-reader-single-nq-base`, [`dpr-ctx_encoder-single-nq-base`](https://huggingface.co/facebook/dpr-ctx_encoder-single-nq-base), and [`dpr-question_encoder-single-nq-base`](https://huggingface.co/facebook/dpr-question_encoder-single-nq-base) can be used for the task of open-domain question answering.
#### Misuse and Out-of-scope Use
The model should not be used to intentionally create hostile or alienating environments for people. In addition, the set of DPR models was not trained to be factual or true representations of people or events, and therefore using the models to generate such content is out-of-scope for the abilities of this model.
## Risks, Limitations and Biases
**CONTENT WARNING: Readers should be aware this section may contain content that is disturbing, offensive, and can propogate historical and current stereotypes.**
Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al., 2021](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al., 2021](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)). Predictions generated by the model can include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.
## Training
#### Training Data
This model was trained using the [Natural Questions (NQ) dataset](https://huggingface.co/datasets/nq_open) ([Lee et al., 2019](https://aclanthology.org/P19-1612/); [Kwiatkowski et al., 2019](https://aclanthology.org/Q19-1026/)). The model authors write that:
> [The dataset] was designed for end-to-end question answering. The questions were mined from real Google search queries and the answers were spans in Wikipedia articles identified by annotators.
#### Training Procedure
The training procedure is described in the [associated paper](https://arxiv.org/pdf/2004.04906.pdf):
> Given a collection of M text passages, the goal of our dense passage retriever (DPR) is to index all the passages in a low-dimensional and continuous space, such that it can retrieve efficiently the top k passages relevant to the input question for the reader at run-time.
> Our dense passage retriever (DPR) uses a dense encoder EP(·) which maps any text passage to a d- dimensional real-valued vectors and builds an index for all the M passages that we will use for retrieval. At run-time, DPR applies a different encoder EQ(·) that maps the input question to a d-dimensional vector, and retrieves k passages of which vectors are the closest to the question vector.
The authors report that for encoders, they used two independent BERT ([Devlin et al., 2019](https://aclanthology.org/N19-1423/)) networks (base, un-cased) and use FAISS ([Johnson et al., 2017](https://arxiv.org/abs/1702.08734)) during inference time to encode and index passages. See the paper for further details on training, including encoders, inference, positive and negative passages, and in-batch negatives.
## Evaluation
The following evaluation information is extracted from the [associated paper](https://arxiv.org/pdf/2004.04906.pdf).
#### Testing Data, Factors and Metrics
The model developers report the performance of the model on five QA datasets, using the top-k accuracy (k ∈ {20, 100}). The datasets were [NQ](https://huggingface.co/datasets/nq_open), [TriviaQA](https://huggingface.co/datasets/trivia_qa), [WebQuestions (WQ)](https://huggingface.co/datasets/web_questions), [CuratedTREC (TREC)](https://huggingface.co/datasets/trec), and [SQuAD v1.1](https://huggingface.co/datasets/squad).
#### Results
| | Top 20 | | | | | Top 100| | | | |
|:----:|:------:|:---------:|:--:|:----:|:-----:|:------:|:---------:|:--:|:----:|:-----:|
| | NQ | TriviaQA | WQ | TREC | SQuAD | NQ | TriviaQA | WQ | TREC | SQuAD |
| | 78.4 | 79.4 |73.2| 79.8 | 63.2 | 85.4 | 85.0 |81.4| 89.1 | 77.2 |
## Environmental Impact
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). We present the hardware type and based on the [associated paper](https://arxiv.org/abs/2004.04906).
- **Hardware Type:** 8 32GB GPUs
- **Hours used:** Unknown
- **Cloud Provider:** Unknown
- **Compute Region:** Unknown
- **Carbon Emitted:** Unknown
## Technical Specifications
See the [associated paper](https://arxiv.org/abs/2004.04906) for details on the modeling architecture, objective, compute infrastructure, and training details.
## Citation Information
```bibtex
@inproceedings{karpukhin-etal-2020-dense,
title = "Dense Passage Retrieval for Open-Domain Question Answering",
author = "Karpukhin, Vladimir and Oguz, Barlas and Min, Sewon and Lewis, Patrick and Wu, Ledell and Edunov, Sergey and Chen, Danqi and Yih, Wen-tau",
booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
month = nov,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/2020.emnlp-main.550",
doi = "10.18653/v1/2020.emnlp-main.550",
pages = "6769--6781",
}
```
## Model Card Authors
This model card was written by the team at Hugging Face.
|
{"language": "en", "license": "cc-by-nc-4.0", "tags": ["dpr"], "datasets": ["nq_open"], "inference": false}
|
facebook/dpr-reader-single-nq-base
| null |
[
"transformers",
"pytorch",
"tf",
"dpr",
"en",
"dataset:nq_open",
"arxiv:2004.04906",
"arxiv:1702.08734",
"arxiv:1910.09700",
"license:cc-by-nc-4.0",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
fill-mask
|
transformers
|
This repository has been deprecated and will be deleted shortly. All ESM models have been moved to their official names to match their naming at the original FAIR repo. You can now find the ESM-1b model at [facebook/esm1b_t33_650M_UR50S](https://huggingface.co/facebook/esm1b_t33_650M_UR50S).
|
{}
|
facebook/esm-1b
| null |
[
"transformers",
"pytorch",
"safetensors",
"esm",
"fill-mask",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text-to-speech
|
fairseq
|
# fastspeech2-en-200_speaker-cv4
[FastSpeech 2](https://arxiv.org/abs/2006.04558) text-to-speech model from fairseq S^2 ([paper](https://arxiv.org/abs/2109.06912)/[code](https://github.com/pytorch/fairseq/tree/main/examples/speech_synthesis)):
- English
- 200 male/female voices (random speaker when using the widget)
- Trained on [Common Voice v4](https://commonvoice.mozilla.org/en/datasets)
## Usage
```python
from fairseq.checkpoint_utils import load_model_ensemble_and_task_from_hf_hub
from fairseq.models.text_to_speech.hub_interface import TTSHubInterface
import IPython.display as ipd
models, cfg, task = load_model_ensemble_and_task_from_hf_hub(
"facebook/fastspeech2-en-200_speaker-cv4",
arg_overrides={"vocoder": "hifigan", "fp16": False}
)
model = models[0]
TTSHubInterface.update_cfg_with_data_cfg(cfg, task.data_cfg)
generator = task.build_generator(model, cfg)
text = "Hello, this is a test run."
sample = TTSHubInterface.get_model_input(task, text)
wav, rate = TTSHubInterface.get_prediction(task, model, generator, sample)
ipd.Audio(wav, rate=rate)
```
See also [fairseq S^2 example](https://github.com/pytorch/fairseq/blob/main/examples/speech_synthesis/docs/common_voice_example.md).
## Citation
```bibtex
@inproceedings{wang-etal-2021-fairseq,
title = "fairseq S{\^{}}2: A Scalable and Integrable Speech Synthesis Toolkit",
author = "Wang, Changhan and
Hsu, Wei-Ning and
Adi, Yossi and
Polyak, Adam and
Lee, Ann and
Chen, Peng-Jen and
Gu, Jiatao and
Pino, Juan",
booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing: System Demonstrations",
month = nov,
year = "2021",
address = "Online and Punta Cana, Dominican Republic",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.emnlp-demo.17",
doi = "10.18653/v1/2021.emnlp-demo.17",
pages = "143--152",
}
```
|
{"language": "en", "library_name": "fairseq", "tags": ["fairseq", "audio", "text-to-speech", "multi-speaker"], "datasets": ["common_voice"], "task": "text-to-speech", "widget": [{"text": "Hello, this is a test run.", "example_title": "Hello, this is a test run."}]}
|
facebook/fastspeech2-en-200_speaker-cv4
| null |
[
"fairseq",
"audio",
"text-to-speech",
"multi-speaker",
"en",
"dataset:common_voice",
"arxiv:2006.04558",
"arxiv:2109.06912",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text-to-speech
|
fairseq
|
# fastspeech2-en-ljspeech
[FastSpeech 2](https://arxiv.org/abs/2006.04558) text-to-speech model from fairseq S^2 ([paper](https://arxiv.org/abs/2109.06912)/[code](https://github.com/pytorch/fairseq/tree/main/examples/speech_synthesis)):
- English
- Single-speaker female voice
- Trained on [LJSpeech](https://keithito.com/LJ-Speech-Dataset/)
## Usage
```python
from fairseq.checkpoint_utils import load_model_ensemble_and_task_from_hf_hub
from fairseq.models.text_to_speech.hub_interface import TTSHubInterface
import IPython.display as ipd
models, cfg, task = load_model_ensemble_and_task_from_hf_hub(
"facebook/fastspeech2-en-ljspeech",
arg_overrides={"vocoder": "hifigan", "fp16": False}
)
model = models[0]
TTSHubInterface.update_cfg_with_data_cfg(cfg, task.data_cfg)
generator = task.build_generator(model, cfg)
text = "Hello, this is a test run."
sample = TTSHubInterface.get_model_input(task, text)
wav, rate = TTSHubInterface.get_prediction(task, model, generator, sample)
ipd.Audio(wav, rate=rate)
```
See also [fairseq S^2 example](https://github.com/pytorch/fairseq/blob/main/examples/speech_synthesis/docs/ljspeech_example.md).
## Citation
```bibtex
@inproceedings{wang-etal-2021-fairseq,
title = "fairseq S{\^{}}2: A Scalable and Integrable Speech Synthesis Toolkit",
author = "Wang, Changhan and
Hsu, Wei-Ning and
Adi, Yossi and
Polyak, Adam and
Lee, Ann and
Chen, Peng-Jen and
Gu, Jiatao and
Pino, Juan",
booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing: System Demonstrations",
month = nov,
year = "2021",
address = "Online and Punta Cana, Dominican Republic",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.emnlp-demo.17",
doi = "10.18653/v1/2021.emnlp-demo.17",
pages = "143--152",
}
```
|
{"language": "en", "library_name": "fairseq", "tags": ["fairseq", "audio", "text-to-speech"], "datasets": ["ljspeech"], "task": "text-to-speech", "widget": [{"text": "Hello, this is a test run.", "example_title": "Hello, this is a test run."}]}
|
facebook/fastspeech2-en-ljspeech
| null |
[
"fairseq",
"audio",
"text-to-speech",
"en",
"dataset:ljspeech",
"arxiv:2006.04558",
"arxiv:2109.06912",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
feature-extraction
|
transformers
|
# Hubert-Base
[Facebook's Hubert](https://ai.facebook.com/blog/hubert-self-supervised-representation-learning-for-speech-recognition-generation-and-compression)
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/2106.07447)
Authors: Wei-Ning Hsu, Benjamin Bolte, Yao-Hung Hubert Tsai, Kushal Lakhotia, Ruslan Salakhutdinov, Abdelrahman Mohamed
**Abstract**
Self-supervised approaches for speech representation learning are challenged by three unique problems: (1) there are multiple sound units in each input utterance, (2) there is no lexicon of input sound units during the pre-training phase, and (3) sound units have variable lengths with no explicit segmentation. To deal with these three problems, we propose the Hidden-Unit BERT (HuBERT) approach for self-supervised speech representation learning, which utilizes an offline clustering step to provide aligned target labels for a BERT-like prediction loss. A key ingredient of our approach is applying the prediction loss over the masked regions only, which forces the model to learn a combined acoustic and language model over the continuous inputs. HuBERT relies primarily on the consistency of the unsupervised clustering step rather than the intrinsic quality of the assigned cluster labels. Starting with a simple k-means teacher of 100 clusters, and using two iterations of clustering, the HuBERT model either matches or improves upon the state-of-the-art wav2vec 2.0 performance on the Librispeech (960h) and Libri-light (60,000h) benchmarks with 10min, 1h, 10h, 100h, and 960h fine-tuning subsets. Using a 1B parameter model, HuBERT shows up to 19% and 13% relative WER reduction on the more challenging dev-other and test-other evaluation subsets.
The original model can be found under https://github.com/pytorch/fairseq/tree/master/examples/hubert .
# Usage
See [this blog](https://huggingface.co/blog/fine-tune-wav2vec2-english) for more information on how to fine-tune the model. Note that the class `Wav2Vec2ForCTC` has to be replaced by `HubertForCTC`.
|
{"language": "en", "license": "apache-2.0", "tags": ["speech"], "datasets": ["librispeech_asr"]}
|
facebook/hubert-base-ls960
| null |
[
"transformers",
"pytorch",
"tf",
"hubert",
"feature-extraction",
"speech",
"en",
"dataset:librispeech_asr",
"arxiv:2106.07447",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
feature-extraction
|
transformers
|
# Hubert-Large
[Facebook's Hubert](https://ai.facebook.com/blog/hubert-self-supervised-representation-learning-for-speech-recognition-generation-and-compression)
The large 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.
The model was pretrained on [Libri-Light](https://github.com/facebookresearch/libri-light).
[Paper](https://arxiv.org/abs/2106.07447)
Authors: Wei-Ning Hsu, Benjamin Bolte, Yao-Hung Hubert Tsai, Kushal Lakhotia, Ruslan Salakhutdinov, Abdelrahman Mohamed
**Abstract**
Self-supervised approaches for speech representation learning are challenged by three unique problems: (1) there are multiple sound units in each input utterance, (2) there is no lexicon of input sound units during the pre-training phase, and (3) sound units have variable lengths with no explicit segmentation. To deal with these three problems, we propose the Hidden-Unit BERT (HuBERT) approach for self-supervised speech representation learning, which utilizes an offline clustering step to provide aligned target labels for a BERT-like prediction loss. A key ingredient of our approach is applying the prediction loss over the masked regions only, which forces the model to learn a combined acoustic and language model over the continuous inputs. HuBERT relies primarily on the consistency of the unsupervised clustering step rather than the intrinsic quality of the assigned cluster labels. Starting with a simple k-means teacher of 100 clusters, and using two iterations of clustering, the HuBERT model either matches or improves upon the state-of-the-art wav2vec 2.0 performance on the Librispeech (960h) and Libri-light (60,000h) benchmarks with 10min, 1h, 10h, 100h, and 960h fine-tuning subsets. Using a 1B parameter model, HuBERT shows up to 19% and 13% relative WER reduction on the more challenging dev-other and test-other evaluation subsets.
The original model can be found under https://github.com/pytorch/fairseq/tree/master/examples/hubert .
# Usage
See [this blog](https://huggingface.co/blog/fine-tune-wav2vec2-english) for more information on how to fine-tune the model. Note that the class `Wav2Vec2ForCTC` has to be replaced by `HubertForCTC`.
|
{"language": "en", "license": "apache-2.0", "tags": ["speech"], "datasets": ["libri-light"]}
|
facebook/hubert-large-ll60k
| null |
[
"transformers",
"pytorch",
"tf",
"hubert",
"feature-extraction",
"speech",
"en",
"dataset:libri-light",
"arxiv:2106.07447",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# Hubert-Large-Finetuned
[Facebook's Hubert](https://ai.facebook.com/blog/hubert-self-supervised-representation-learning-for-speech-recognition-generation-and-compression)
The large model fine-tuned on 960h of Librispeech on 16kHz sampled speech audio. When using the model make sure that your speech input is also sampled at 16Khz.
The model is a fine-tuned version of [hubert-large-ll60k](https://huggingface.co/facebook/hubert-large-ll60k).
[Paper](https://arxiv.org/abs/2106.07447)
Authors: Wei-Ning Hsu, Benjamin Bolte, Yao-Hung Hubert Tsai, Kushal Lakhotia, Ruslan Salakhutdinov, Abdelrahman Mohamed
**Abstract**
Self-supervised approaches for speech representation learning are challenged by three unique problems: (1) there are multiple sound units in each input utterance, (2) there is no lexicon of input sound units during the pre-training phase, and (3) sound units have variable lengths with no explicit segmentation. To deal with these three problems, we propose the Hidden-Unit BERT (HuBERT) approach for self-supervised speech representation learning, which utilizes an offline clustering step to provide aligned target labels for a BERT-like prediction loss. A key ingredient of our approach is applying the prediction loss over the masked regions only, which forces the model to learn a combined acoustic and language model over the continuous inputs. HuBERT relies primarily on the consistency of the unsupervised clustering step rather than the intrinsic quality of the assigned cluster labels. Starting with a simple k-means teacher of 100 clusters, and using two iterations of clustering, the HuBERT model either matches or improves upon the state-of-the-art wav2vec 2.0 performance on the Librispeech (960h) and Libri-light (60,000h) benchmarks with 10min, 1h, 10h, 100h, and 960h fine-tuning subsets. Using a 1B parameter model, HuBERT shows up to 19% and 13% relative WER reduction on the more challenging dev-other and test-other evaluation subsets.
The original model can be found under https://github.com/pytorch/fairseq/tree/master/examples/hubert .
# Usage
The model can be used for automatic-speech-recognition as follows:
```python
import torch
from transformers import Wav2Vec2Processor, HubertForCTC
from datasets import load_dataset
processor = Wav2Vec2Processor.from_pretrained("facebook/hubert-large-ls960-ft")
model = HubertForCTC.from_pretrained("facebook/hubert-large-ls960-ft")
ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
input_values = processor(ds[0]["audio"]["array"], return_tensors="pt").input_values # Batch size 1
logits = model(input_values).logits
predicted_ids = torch.argmax(logits, dim=-1)
transcription = processor.decode(predicted_ids[0])
# ->"A MAN SAID TO THE UNIVERSE SIR I EXIST"
```
|
{"language": "en", "license": "apache-2.0", "tags": ["speech", "audio", "automatic-speech-recognition", "hf-asr-leaderboard"], "datasets": ["libri-light", "librispeech_asr"], "model-index": [{"name": "hubert-large-ls960-ft", "results": [{"task": {"type": "automatic-speech-recognition", "name": "Automatic Speech Recognition"}, "dataset": {"name": "LibriSpeech (clean)", "type": "librispeech_asr", "config": "clean", "split": "test", "args": {"language": "en"}}, "metrics": [{"type": "wer", "value": 1.9, "name": "Test WER"}]}]}]}
|
facebook/hubert-large-ls960-ft
| null |
[
"transformers",
"pytorch",
"tf",
"hubert",
"automatic-speech-recognition",
"speech",
"audio",
"hf-asr-leaderboard",
"en",
"dataset:libri-light",
"dataset:librispeech_asr",
"arxiv:2106.07447",
"license:apache-2.0",
"model-index",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
feature-extraction
|
transformers
|
# Hubert-Extra-Large
[Facebook's Hubert](https://ai.facebook.com/blog/hubert-self-supervised-representation-learning-for-speech-recognition-generation-and-compression)
The extra large model pretrained on 16kHz sampled speech audio. When using the model make sure that your speech input is also sampled at 16Khz. Note that this model should be fine-tuned on a downstream task, like Automatic Speech Recognition, Speaker Identification, Intent Classification, Emotion Recognition, etc...
The model was pretrained on [Libri-Light](https://github.com/facebookresearch/libri-light).
[Paper](https://arxiv.org/abs/2106.07447)
Authors: Wei-Ning Hsu, Benjamin Bolte, Yao-Hung Hubert Tsai, Kushal Lakhotia, Ruslan Salakhutdinov, Abdelrahman Mohamed
**Abstract**
Self-supervised approaches for speech representation learning are challenged by three unique problems: (1) there are multiple sound units in each input utterance, (2) there is no lexicon of input sound units during the pre-training phase, and (3) sound units have variable lengths with no explicit segmentation. To deal with these three problems, we propose the Hidden-Unit BERT (HuBERT) approach for self-supervised speech representation learning, which utilizes an offline clustering step to provide aligned target labels for a BERT-like prediction loss. A key ingredient of our approach is applying the prediction loss over the masked regions only, which forces the model to learn a combined acoustic and language model over the continuous inputs. HuBERT relies primarily on the consistency of the unsupervised clustering step rather than the intrinsic quality of the assigned cluster labels. Starting with a simple k-means teacher of 100 clusters, and using two iterations of clustering, the HuBERT model either matches or improves upon the state-of-the-art wav2vec 2.0 performance on the Librispeech (960h) and Libri-light (60,000h) benchmarks with 10min, 1h, 10h, 100h, and 960h fine-tuning subsets. Using a 1B parameter model, HuBERT shows up to 19% and 13% relative WER reduction on the more challenging dev-other and test-other evaluation subsets.
The original model can be found under https://github.com/pytorch/fairseq/tree/master/examples/hubert .
# Usage
See [this blog](https://huggingface.co/blog/fine-tune-wav2vec2-english) for more information on how to fine-tune the model. Note that the class `Wav2Vec2ForCTC` has to be replaced by `HubertForCTC`.
|
{"language": "en", "license": "apache-2.0", "tags": ["speech"], "datasets": ["libri-light"]}
|
facebook/hubert-xlarge-ll60k
| null |
[
"transformers",
"pytorch",
"tf",
"hubert",
"feature-extraction",
"speech",
"en",
"dataset:libri-light",
"arxiv:2106.07447",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# Hubert-Extra-Large-Finetuned
[Facebook's Hubert](https://ai.facebook.com/blog/hubert-self-supervised-representation-learning-for-speech-recognition-generation-and-compression)
The extra large model fine-tuned on 960h of Librispeech on 16kHz sampled speech audio. When using the model make sure that your speech input is also sampled at 16Khz.
The model is a fine-tuned version of [hubert-xlarge-ll60k](https://huggingface.co/facebook/hubert-xlarge-ll60k).
[Paper](https://arxiv.org/abs/2106.07447)
Authors: Wei-Ning Hsu, Benjamin Bolte, Yao-Hung Hubert Tsai, Kushal Lakhotia, Ruslan Salakhutdinov, Abdelrahman Mohamed
**Abstract**
Self-supervised approaches for speech representation learning are challenged by three unique problems: (1) there are multiple sound units in each input utterance, (2) there is no lexicon of input sound units during the pre-training phase, and (3) sound units have variable lengths with no explicit segmentation. To deal with these three problems, we propose the Hidden-Unit BERT (HuBERT) approach for self-supervised speech representation learning, which utilizes an offline clustering step to provide aligned target labels for a BERT-like prediction loss. A key ingredient of our approach is applying the prediction loss over the masked regions only, which forces the model to learn a combined acoustic and language model over the continuous inputs. HuBERT relies primarily on the consistency of the unsupervised clustering step rather than the intrinsic quality of the assigned cluster labels. Starting with a simple k-means teacher of 100 clusters, and using two iterations of clustering, the HuBERT model either matches or improves upon the state-of-the-art wav2vec 2.0 performance on the Librispeech (960h) and Libri-light (60,000h) benchmarks with 10min, 1h, 10h, 100h, and 960h fine-tuning subsets. Using a 1B parameter model, HuBERT shows up to 19% and 13% relative WER reduction on the more challenging dev-other and test-other evaluation subsets.
The original model can be found under https://github.com/pytorch/fairseq/tree/master/examples/hubert .
# Usage
The model can be used for automatic-speech-recognition as follows:
```python
import torch
from transformers import Wav2Vec2Processor, HubertForCTC
from datasets import load_dataset
processor = Wav2Vec2Processor.from_pretrained("facebook/hubert-xlarge-ls960-ft")
model = HubertForCTC.from_pretrained("facebook/hubert-xlarge-ls960-ft")
ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
input_values = processor(ds[0]["audio"]["array"], return_tensors="pt").input_values # Batch size 1
logits = model(input_values).logits
predicted_ids = torch.argmax(logits, dim=-1)
transcription = processor.decode(predicted_ids[0])
# ->"A MAN SAID TO THE UNIVERSE SIR I EXIST"
```
|
{"language": "en", "license": "apache-2.0", "tags": ["speech", "audio", "automatic-speech-recognition", "hf-asr-leaderboard"], "datasets": ["libri-light", "librispeech_asr"], "model-index": [{"name": "hubert-large-ls960-ft", "results": [{"task": {"type": "automatic-speech-recognition", "name": "Automatic Speech Recognition"}, "dataset": {"name": "LibriSpeech (clean)", "type": "librispeech_asr", "config": "clean", "split": "test", "args": {"language": "en"}}, "metrics": [{"type": "wer", "value": 1.8, "name": "Test WER"}]}]}]}
|
facebook/hubert-xlarge-ls960-ft
| null |
[
"transformers",
"pytorch",
"tf",
"safetensors",
"hubert",
"automatic-speech-recognition",
"speech",
"audio",
"hf-asr-leaderboard",
"en",
"dataset:libri-light",
"dataset:librispeech_asr",
"arxiv:2106.07447",
"license:apache-2.0",
"model-index",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null | null |
# <p align="center"> IC-GAN: Instance-Conditioned GAN </p>
Official Pytorch code of [Instance-Conditioned GAN](https://arxiv.org/abs/2109.05070) by Arantxa Casanova, Marlène Careil, Jakob Verbeek, Michał Drożdżal, Adriana Romero-Soriano.
## Generate images with IC-GAN in a Colab Notebook
We provide a [Google Colab notebook](https://colab.research.google.com/github/facebookresearch/ic_gan/blob/main/inference/icgan_colab.ipynb) to generate images with IC-GAN and its class-conditional counter part. We also invite users to check out the [demo on Replicate](https://replicate.ai/arantxacasanova/ic_gan), courtesy of [Replicate](https://replicate.ai/home).
The figure below depicts two instances, unseen during training and downloaded from [Creative Commons search](https://search.creativecommons.org), and the generated images with IC-GAN and class-conditional IC-GAN when conditioning on the class "castle":
Additionally, and inspired by [this Colab](https://colab.research.google.com/github/eyaler/clip_biggan/blob/main/ClipBigGAN.ipynb), we provide the funcionality in the same Colab notebook to guide generations with text captions, using the [CLIP model](https://github.com/openai/CLIP).
As an example, the following Figure shows three instance conditionings and a text caption (top), followed by the resulting generated images with IC-GAN (bottom), when optimizing the noise vector following CLIP's gradient for 100 iterations.
*Credit for the three instance conditionings, from left to right, that were modified with a resize and central crop:* [1: "Landscape in Bavaria" by shining.darkness, licensed under CC BY 2.0](https://search.creativecommons.org/photos/92ef279c-4469-49a5-aa4b-48ad746f2dc4), [2: "Fantasy Landscape - slolsss" by Douglas Tofoli is marked with CC PDM 1.0](https://search.creativecommons.org/photos/13646adc-f1df-437a-a0dd-8223452ee46c), [3: "How to Draw Landscapes Simply" by Kuwagata Keisai is marked with CC0 1.0](https://search.creativecommons.org/photos/2ab9c3b7-de99-4536-81ed-604ee988bd5f)
## Requirements
* Python 3.8
* Cuda v10.2 / Cudnn v7.6.5
* gcc v7.3.0
* Pytorch 1.8.0
* A conda environment can be created from `environment.yaml` by entering the command: `conda env create -f environment.yml`, that contains the aforemention version of Pytorch and other required packages.
* Faiss: follow the instructions in the [original repository](https://github.com/facebookresearch/faiss).
## Overview
This repository consists of four main folders:
* `data_utils`: A common folder to obtain and format the data needed to train and test IC-GAN, agnostic of the specific backbone.
* `inference`: Scripts to test the models both qualitatively and quantitatively.
* `BigGAN_PyTorch`: It provides the training, evaluation and sampling scripts for IC-GAN with a BigGAN backbone. The code base comes from [Pytorch BigGAN repository](https://github.com/ajbrock/BigGAN-PyTorch), made available under the MIT License. It has been modified to [add additional utilities](#biggan-changelog) and it enables IC-GAN training on top of it.
* `stylegan2_ada_pytorch`: It provides the training, evaluation and sampling scripts for IC-GAN with a StyleGAN2 backbone. The code base comes from [StyleGAN2 Pytorch](https://github.com/NVlabs/stylegan2-ada-pytorch), made available under the [Nvidia Source Code License](https://nvlabs.github.io/stylegan2-ada-pytorch/license.html). It has been modified to [add additional utilities](#stylegan-changelog) and it enables IC-GAN training on top of it.
## (Python script) Generate images with IC-GAN
Alternatively, we can <b> generate images with IC-GAN models </b> directly from a python script, by following the next steps:
1) Download the desired pretrained models (links below) and the [pre-computed 1000 instance features from ImageNet](https://dl.fbaipublicfiles.com/ic_gan/stored_instances.tar.gz) and extract them into a folder `pretrained_models_path`.
| model | backbone | class-conditional? | training dataset | resolution | url |
|-------------------|-------------------|-------------------|---------------------|--------------------|--------------------|
| IC-GAN | BigGAN | No | ImageNet | 256x256 | [model](https://dl.fbaipublicfiles.com/ic_gan/icgan_biggan_imagenet_res256.tar.gz) |
| IC-GAN (half capacity) | BigGAN | No | ImageNet | 256x256 | [model](https://dl.fbaipublicfiles.com/ic_gan/icgan_biggan_imagenet_res256_halfcap.tar.gz) |
| IC-GAN | BigGAN | No | ImageNet | 128x128 | [model](https://dl.fbaipublicfiles.com/ic_gan/icgan_biggan_imagenet_res128.tar.gz) |
| IC-GAN | BigGAN | No | ImageNet | 64x64 | [model](https://dl.fbaipublicfiles.com/ic_gan/icgan_biggan_imagenet_res64.tar.gz) |
| IC-GAN | BigGAN | Yes | ImageNet | 256x256 | [model](https://dl.fbaipublicfiles.com/ic_gan/cc_icgan_biggan_imagenet_res256.tar.gz) |
| IC-GAN (half capacity) | BigGAN | Yes | ImageNet | 256x256 | [model](https://dl.fbaipublicfiles.com/ic_gan/cc_icgan_biggan_imagenet_res256_halfcap.tar.gz) |
| IC-GAN | BigGAN | Yes | ImageNet | 128x128 | [model](https://dl.fbaipublicfiles.com/ic_gan/cc_icgan_biggan_imagenet_res128.tar.gz) |
| IC-GAN | BigGAN | Yes | ImageNet | 64x64 | [model](https://dl.fbaipublicfiles.com/ic_gan/cc_icgan_biggan_imagenet_res64.tar.gz) |
| IC-GAN | BigGAN | Yes | ImageNet-LT | 256x256 | [model](https://dl.fbaipublicfiles.com/ic_gan/cc_icgan_biggan_imagenetlt_res256.tar.gz) |
| IC-GAN | BigGAN | Yes | ImageNet-LT | 128x128 | [model](https://dl.fbaipublicfiles.com/ic_gan/cc_icgan_biggan_imagenetlt_res128.tar.gz) |
| IC-GAN | BigGAN | Yes | ImageNet-LT | 64x64 | [model](https://dl.fbaipublicfiles.com/ic_gan/cc_icgan_biggan_imagenetlt_res64.tar.gz) |
| IC-GAN | BigGAN | No | COCO-Stuff | 256x256 | [model](https://dl.fbaipublicfiles.com/ic_gan/icgan_biggan_coco_res256.tar.gz) |
| IC-GAN | BigGAN | No | COCO-Stuff | 128x128 | [model](https://dl.fbaipublicfiles.com/ic_gan/icgan_biggan_coco_res128.tar.gz) |
| IC-GAN | StyleGAN2 | No | COCO-Stuff | 256x256 | [model](https://dl.fbaipublicfiles.com/ic_gan/icgan_stylegan2_coco_res256.tar.gz) |
| IC-GAN | StyleGAN2 | No | COCO-Stuff | 128x128 | [model](https://dl.fbaipublicfiles.com/ic_gan/icgan_stylegan2_coco_res128.tar.gz) |
2) Execute:
```
python inference/generate_images.py --root_path [pretrained_models_path] --model [model] --model_backbone [backbone] --resolution [res]
```
* `model` can be chosen from `["icgan", "cc_icgan"]` to use the IC-GAN or the class-conditional IC-GAN model respectively.
* `backbone` can be chosen from `["biggan", "stylegan2"]`.
* `res` indicates the resolution at which the model has been trained. For ImageNet, choose one in `[64, 128, 256]`, and for COCO-Stuff, one in `[128, 256]`.
This script results in a .PNG file where several generated images are shown, given an instance feature (each row), and a sampled noise vector (each grid position).
<b>Additional and optional parameters</b>:
* `index`: (None by default), is an integer from 0 to 999 that choses a specific instance feature vector out of the 1000 instances that have been selected with k-means on the ImageNet dataset and stored in `pretrained_models_path/stored_instances`.
* `swap_target`: (None by default) is an integer from 0 to 999 indicating an ImageNet class label. This label will be used to condition the class-conditional IC-GAN, regardless of which instance features are being used.
* `which_dataset`: (ImageNet by default) can be chosen from `["imagenet", "coco"]` to indicate which dataset (training split) to sample the instances from.
* `trained_dataset`: (ImageNet by default) can be chosen from `["imagenet", "coco"]` to indicate the dataset in which the IC-GAN model has been trained on.
* `num_imgs_gen`: (5 by default), it changes the number of noise vectors to sample per conditioning. Increasing this number results in a bigger .PNG file to save and load.
* `num_conditionings_gen`: (5 by default), it changes the number of conditionings to sample. Increasing this number results in a bigger .PNG file to save and load.
* `z_var`: (1.0 by default) controls the truncation factor for the generation.
* Optionally, the script can be run with the following additional options `--visualize_instance_images --dataset_path [dataset_path]` to visualize the ground-truth images corresponding to the conditioning instance features, given a path to the dataset's ground-truth images `dataset_path`. Ground-truth instances will be plotted as the leftmost image for each row.
## Data preparation
<div id="data-preparation">
<details>
<summary>ImageNet</summary>
<br>
<ol>
<li>Download dataset from <a href="https://image-net.org/download.php"> here </a>.
</li>
<li>Download <a href="https://github.com/facebookresearch/swav"> SwAV </a> feature extractor weights from <a href="https://dl.fbaipublicfiles.com/deepcluster/swav_800ep_pretrain.pth.tar"> here </a>. </li>
<li> Replace the paths in data_utils/prepare_data.sh: <code>out_path</code> by the path where hdf5 files will be stored, <code>path_imnet</code> by the path where ImageNet dataset is downloaded, and <code>path_swav</code> by the path where SwAV weights are stored. </li>
<li> Execute <code>./data_utils/prepare_data.sh imagenet [resolution]</code>, where <code>[resolution]</code> can be an integer in {64,128,256}. This script will create several hdf5 files:
<ul> <li> <code>ILSVRC[resolution]_xy.hdf5</code> and <code>ILSVRC[resolution]_val_xy.hdf5</code>, where images and labels are stored for the training and validation set respectively. </li>
<li> <code>ILSVRC[resolution]_feats_[feature_extractor]_resnet50.hdf5</code> that contains the instance features for each image. </li>
<li> <code>ILSVRC[resolution]_feats_[feature_extractor]_resnet50_nn_k[k_nn].hdf5</code> that contains the list of [k_nn] neighbors for each of the instance features. </li> </ul> </li>
</ol> </br>
</details>
<details>
<summary>ImageNet-LT</summary>
<br>
<ol>
<li>Download ImageNet dataset from <a href="https://image-net.org/download.php"> here </a>. Following <a href="https://github.com/zhmiao/OpenLongTailRecognition-OLTR"> ImageNet-LT </a>, the file <code>ImageNet_LT_train.txt</code> can be downloaded from <a href="https://drive.google.com/drive/u/1/folders/1j7Nkfe6ZhzKFXePHdsseeeGI877Xu1yf" > this link </a> and later stored in the folder <code>./BigGAN_PyTorch/imagenet_lt</code>.
</li>
<li>Download the pre-trained weights of the ResNet on ImageNet-LT from <a href="https://dl.fbaipublicfiles.com/classifier-balancing/ImageNet_LT/models/resnet50_uniform_e90.pth"> this link</a>, provided by the <a href="https://github.com/facebookresearch/classifier-balancing"> classifier-balancing repository </a>. </li>
<li> Replace the paths in data_utils/prepare_data.sh: <code>out_path</code> by the path where hdf5 files will be stored, <code>path_imnet</code> by the path where ImageNet dataset is downloaded, and <code>path_classifier_lt</code> by the path where the pre-trained ResNet50 weights are stored. </li>
<li> Execute <code>./data_utils/prepare_data.sh imagenet_lt [resolution]</code>, where <code>[resolution]</code> can be an integer in {64,128,256}. This script will create several hdf5 files:
<ul> <li> <code>ILSVRC[resolution]longtail_xy.hdf5</code>, where images and labels are stored for the training and validation set respectively. </li>
<li> <code>ILSVRC[resolution]longtail_feats_[feature_extractor]_resnet50.hdf5</code> that contains the instance features for each image. </li>
<li> <code>ILSVRC[resolution]longtail_feats_[feature_extractor]_resnet50_nn_k[k_nn].hdf5</code> that contains the list of [k_nn] neighbors for each of the instance features. </li> </ul> </li>
</ol> </br>
</details>
<details>
<summary>COCO-Stuff</summary>
<br>
<ol>
<li>Download the dataset following the <a href="https://github.com/WillSuen/LostGANs/blob/master/INSTALL.md"> LostGANs' repository instructions </a>.
</li>
<li>Download <a href="https://github.com/facebookresearch/swav"> SwAV </a> feature extractor weights from <a href="https://dl.fbaipublicfiles.com/deepcluster/swav_800ep_pretrain.pth.tar"> here </a>. </li>
<li> Replace the paths in data_utils/prepare_data.sh: <code>out_path</code> by the path where hdf5 files will be stored, <code>path_imnet</code> by the path where ImageNet dataset is downloaded, and <code>path_swav</code> by the path where SwAV weights are stored. </li>
<li> Execute <code>./data_utils/prepare_data.sh coco [resolution]</code>, where <code>[resolution]</code> can be an integer in {128,256}. This script will create several hdf5 files:
<ul> <li> <code>COCO[resolution]_xy.hdf5</code> and <code>COCO[resolution]_val_test_xy.hdf5</code>, where images and labels are stored for the training and evaluation set respectively. </li>
<li> <code>COCO[resolution]_feats_[feature_extractor]_resnet50.hdf5</code> that contains the instance features for each image. </li>
<li> <code>COCO[resolution]_feats_[feature_extractor]_resnet50_nn_k[k_nn].hdf5</code> that contains the list of [k_nn] neighbors for each of the instance features. </li> </ul> </li>
</ol> </br>
</details>
<details>
<summary>Other datasets</summary>
<br>
<ol>
<li>Download the corresponding dataset and store in a folder <code>dataset_path</code>.
</li>
<li>Download <a href="https://github.com/facebookresearch/swav"> SwAV </a> feature extractor weights from <a href="https://dl.fbaipublicfiles.com/deepcluster/swav_800ep_pretrain.pth.tar"> here </a>. </li>
<li> Replace the paths in data_utils/prepare_data.sh: <code>out_path</code> by the path where hdf5 files will be stored and <code>path_swav</code> by the path where SwAV weights are stored. </li>
<li> Execute <code>./data_utils/prepare_data.sh [dataset_name] [resolution] [dataset_path]</code>, where <code>[dataset_name]</code> will be the dataset name, <code>[resolution]</code> can be an integer, for example 128 or 256, and <code>dataset_path</code> contains the dataset images. This script will create several hdf5 files:
<ul> <li> <code>[dataset_name][resolution]_xy.hdf5</code>, where images and labels are stored for the training set. </li>
<li> <code>[dataset_name][resolution]_feats_[feature_extractor]_resnet50.hdf5</code> that contains the instance features for each image. </li>
<li> <code>[dataset_name][resolution]_feats_[feature_extractor]_resnet50_nn_k[k_nn].hdf5</code> that contains the list of <code>k_nn</code> neighbors for each of the instance features. </li> </ul> </li>
</ol> </br>
</details>
<details>
<summary>How to subsample an instance feature dataset with k-means</summary>
<br>
To downsample the instance feature vector dataset, after we have prepared the data, we can use the k-means algorithm:
<code>
python data_utils/store_kmeans_indexes.py --resolution [resolution] --which_dataset [dataset_name] --data_root [data_path]
</code>
<ul> <li> Adding <code>--gpu</code> allows the faiss library to compute k-means leveraging GPUs, resulting in faster execution. </li>
<li> Adding the parameter <code>--feature_extractor [feature_extractor]</code> chooses which feature extractor to use, with <code>feature_extractor</code> in <code>['selfsupervised', 'classification'] </code>, if we are using swAV as feature extactor or the ResNet pretrained on the classification task on ImageNet, respectively. </li>
<li> The number of k-means clusters can be set with <code>--kmeans_subsampled [centers]</code>, where <code>centers</code> is an integer. </li> </ul>
</br>
</details>
</div>
## How to train the models
#### BigGAN or StyleGAN2 backbone
Training parameters are stored in JSON files in `[backbone_folder]/config_files/[dataset]/*.json`, where `[backbone_folder]` is either BigGAN_Pytorch or stylegan2_ada_pytorch and `[dataset]` can either be ImageNet, ImageNet-LT or COCO_Stuff.
```
cd BigGAN_PyTorch
python run.py --json_config config_files/<dataset>/<selected_config>.json --data_root [data_root] --base_root [base_root]
```
or
```
cd stylegan_ada_pytorch
python run.py --json_config config_files/<dataset>/<selected_config>.json --data_root [data_root] --base_root [base_root]
```
where:
* `data_root` path where the data has been prepared and stored, following the previous section (<a href="./README.md#data-preparation">Data preparation</a>).
* `base_root` path where to store the model weights and logs.
Note that one can create other JSON files to modify the training parameters.
#### Other backbones
To be able to run IC-GAN with other backbones, we provide some orientative steps:
* Place the new backbone code in a new folder under `ic_gan` (`ic_gan/new_backbone`).
* Modify the relevant piece of code in the GAN architecture to allow instance features as conditionings (for both generator and discriminator).
* Create a `trainer.py` file with the training loop to train an IC-GAN with the new backbone. The `data_utils` folder provides the tools to prepare the dataset, load the data and conditioning sampling to train an IC-GAN. The IC-GAN with BigGAN backbone [`trainer.py`](BigGAN_PyTorch/trainer.py) file can be used as an inspiration.
## How to test the models
<b>To obtain the FID and IS metrics on ImageNet and ImageNet-LT</b>:
1) Execute:
```
python inference/test.py --json_config [BigGAN-PyTorch or stylegan-ada-pytorch]/config_files/<dataset>/<selected_config>.json --num_inception_images [num_imgs] --sample_num_npz [num_imgs] --eval_reference_set [ref_set] --sample_npz --base_root [base_root] --data_root [data_root] --kmeans_subsampled [kmeans_centers] --model_backbone [backbone]
```
To obtain the tensorflow IS and FID metrics, use an environment with the Python <3.7 and Tensorflow 1.15. Then:
2) Obtain Inception Scores and pre-computed FID moments:
```
python ../data_utils/inception_tf13.py --experiment_name [exp_name] --experiment_root [base_root] --kmeans_subsampled [kmeans_centers]
```
For stratified FIDs in the ImageNet-LT dataset, the following parameters can be added `--which_dataset 'imagenet_lt' --split 'val' --strat_name [stratified_split]`, where `stratified_split` can be in `[few,low, many]`.
3) (Only needed once) Pre-compute reference moments with tensorflow code:
```
python ../data_utils/inception_tf13.py --use_ground_truth_data --data_root [data_root] --split [ref_set] --resolution [res] --which_dataset [dataset]
```
4) (Using this [repository](https://github.com/bioinf-jku/TTUR)) FID can be computed using the pre-computed statistics obtained in 2) and the pre-computed ground-truth statistics obtain in 3). For example, to compute the FID with reference ImageNet validation set:
```python TTUR/fid.py [base_root]/[exp_name]/TF_pool_.npz [data_root]/imagenet_val_res[res]_tf_inception_moments_ground_truth.npz ```
<b>To obtain the FID metric on COCO-Stuff</b>:
1) Obtain ground-truth jpeg images: ```python data_utils/store_coco_jpeg_images.py --resolution [res] --split [ref_set] --data_root [data_root] --out_path [gt_coco_images] --filter_hd [filter_hd] ```
2) Store generated images as jpeg images: ```python sample.py --json_config ../[BigGAN-PyTorch or stylegan-ada-pytorch]/config_files/<dataset>/<selected_config>.json --data_root [data_root] --base_root [base_root] --sample_num_npz [num_imgs] --which_dataset 'coco' --eval_instance_set [ref_set] --eval_reference_set [ref_set] --filter_hd [filter_hd] --model_backbone [backbone] ```
3) Using this [repository](https://github.com/bioinf-jku/TTUR), compute FID on the two folders of ground-truth and generated images.
where:
* `dataset`: option to select the dataset in `['imagenet', 'imagenet_lt', 'coco']
* `exp_name`: name of the experiment folder.
* `data_root`: path where the data has been prepared and stored, following the previous section ["Data preparation"](#data-preparation).
* `base_root`: path where to find the model (for example, where the pretrained models have been downloaded).
* `num_imgs`: needs to be set to 50000 for ImageNet and ImageNet-LT (with validation set as reference) and set to 11500 for ImageNet-LT (with training set as reference). For COCO-Stuff, set to 75777, 2050, 675, 1375 if using the training, evaluation, evaluation seen or evaluation unseen set as reference.
* `ref_set`: set to `'val'` for ImageNet, ImageNet-LT (and COCO) to obtain metrics with the validation (evaluation) set as reference, or set to `'train'` for ImageNet-LT or COCO to obtain metrics with the training set as reference.
* `kmeans_centers`: set to 1000 for ImageNet and to -1 for ImageNet-LT.
* `backbone`: model backbone architecture in `['biggan','stylegan2']`.
* `res`: integer indicating the resolution of the images (64,128,256).
* `gt_coco_images`: folder to store the ground-truth JPEG images of that specific split.
* `filter_hd`: only valid for `ref_set=val`. If -1, use the entire evaluation set; if 0, use only conditionings and their ground-truth images with seen class combinations during training (eval seen); if 1, use only conditionings and their ground-truth images with unseen class combinations during training (eval unseen).
## Utilities for GAN backbones
We change and provide extra utilities to facilitate the training, for both BigGAN and StyleGAN2 base repositories.
### BigGAN change log
The following changes were made:
* BigGAN architecture:
* In `train_fns.py`: option to either have the optimizers inside the generator and discriminator class, or directly in the `G_D` wrapper module. Additionally, added an option to augment both generated and real images with augmentations from [DiffAugment](https://github.com/mit-han-lab/data-efficient-gans).
* In `BigGAN.py`: added a function `get_condition_embeddings` to handle the conditioning separately.
* Small modifications to `layers.py` to adapt the batchnorm function calls to the pytorch 1.8 version.
* Training utilities:
* Added `trainer.py` file (replacing train.py):
* Training now allows the usage of DDP for faster single-node and multi-node training.
* Training is performed by epochs instead of by iterations.
* Option to stop the training by using early stopping or when experiments diverge.
* In `utils.py`:
* Replaced `MultiEpochSampler` for `CheckpointedSampler` to allow experiments to be resumable when using epochs and fixing a bug where `MultiEpochSampler` would require a long time to fetch data permutations when the number of epochs increased.
* ImageNet-LT: Added option to use different class distributions when sampling a class label for the generator.
* ImageNet-LT: Added class balancing (uniform and temperature annealed).
* Added data augmentations from [DiffAugment](https://github.com/mit-han-lab/data-efficient-gans).
* Testing utilities:
* In `calculate_inception_moments.py`: added option to obtain moments for ImageNet-LT dataset, as well as stratified moments for many, medium and few-shot classes (stratified FID computation).
* In `inception_utils.py`: added option to compute [Precision, Recall, Density, Coverage](https://github.com/clovaai/generative-evaluation-prdc) and stratified FID.
* Data utilities:
* In `datasets.py`, added option to load ImageNet-LT dataset.
* Added ImageNet-LT.txt files with image indexes for training and validation split.
* In `utils.py`:
* Separate functions to obtain the data from hdf5 files (`get_dataset_hdf5`) or from directory (`get_dataset_images`), as well as a function to obtain only the data loader (`get_dataloader`).
* Added the function `sample_conditionings` to handle possible different conditionings to train G with.
* Experiment utilities:
* Added JSON files to launch experiments with the proposed hyper-parameter configuration.
* Script to launch experiments with either the [submitit tool](https://github.com/facebookincubator/submitit) or locally in the same machine (run.py).
### StyleGAN2 change log
<div id="stylegan-changelog">
<ul>
<li> Multi-node DistributedDataParallel training. </li>
<li> Added early stopping based on the training FID metric. </li>
<li> Automatic checkpointing when jobs are automatically rescheduled on a cluster. </li>
<li> Option to load dataset from hdf5 file. </li>
<li> Replaced the usage of Click python package by an `ArgumentParser`. </li>
<li> Only saving best and last model weights. </li>
</ul>
</div>
## Acknowledgements
We would like to thanks the authors of the [Pytorch BigGAN repository](https://github.com/ajbrock/BigGAN-PyTorch) and [StyleGAN2 Pytorch](https://github.com/NVlabs/stylegan2-ada-pytorch), as our model requires their repositories to train IC-GAN with BigGAN or StyleGAN2 bakcbone respectively.
Moreover, we would like to further thank the authors of [generative-evaluation-prdc](https://github.com/clovaai/generative-evaluation-prdc), [data-efficient-gans](https://github.com/mit-han-lab/data-efficient-gans), [faiss](https://github.com/facebookresearch/faiss) and [sg2im](https://github.com/google/sg2im) as some components were borrowed and modified from their code bases. Finally, we thank the author of [WanderCLIP](https://colab.research.google.com/github/eyaler/clip_biggan/blob/main/WanderCLIP.ipynb) as well as the following repositories, that we use in our Colab notebook: [pytorch-pretrained-BigGAN](https://github.com/huggingface/pytorch-pretrained-BigGAN) and [CLIP](https://github.com/openai/CLIP).
## License
The majority of IC-GAN is licensed under CC-BY-NC, however portions of the project are available under separate license terms: BigGAN and [PRDC](https://github.com/facebookresearch/ic_gan/blob/main/data_utils/compute_pdrc.py) are licensed under the MIT license; [COCO-Stuff loader](https://github.com/facebookresearch/ic_gan/blob/main/data_utils/cocostuff_dataset.py) is licensed under Apache License 2.0; [DiffAugment](https://github.com/facebookresearch/ic_gan/blob/main/BigGAN_PyTorch/diffaugment_utils.py) is licensed under BSD 2-Clause Simplified license; StyleGAN2 is licensed under a NVIDIA license, available here: https://github.com/NVlabs/stylegan2-ada-pytorch/blob/main/LICENSE.txt. In the Colab notebook, [CLIP](https://github.com/openai/CLIP) and [pytorch-pretrained-BigGAN](https://github.com/huggingface/pytorch-pretrained-BigGAN) code is used, both licensed under the MIT license.
## Disclaimers
THE DIFFAUGMENT SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
THE CLIP SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
THE PYTORCH-PRETRAINED-BIGGAN SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
## Cite the paper
If this repository, the paper or any of its content is useful for your research, please cite:
```
@inproceedings{casanova2021instanceconditioned,
title={Instance-Conditioned GAN},
author={Arantxa Casanova and Marlène Careil and Jakob Verbeek and Michal Drozdzal and Adriana Romero-Soriano},
booktitle={Advances in Neural Information Processing Systems (NeurIPS)},
year={2021}
}
```
|
{"license": "cc-by-nc-4.0", "tags": ["image-generation", "conditional-image-generation", "generative-model"], "library": "pytorch"}
|
facebook/ic_gan
| null |
[
"image-generation",
"conditional-image-generation",
"generative-model",
"arxiv:2109.05070",
"license:cc-by-nc-4.0",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
# M2M100 1.2B
M2M100 is a multilingual encoder-decoder (seq-to-seq) model trained for Many-to-Many multilingual translation.
It was introduced in this [paper](https://arxiv.org/abs/2010.11125) and first released in [this](https://github.com/pytorch/fairseq/tree/master/examples/m2m_100) repository.
The model that can directly translate between the 9,900 directions of 100 languages.
To translate into a target language, the target language id is forced as the first generated token.
To force the target language id as the first generated token, pass the `forced_bos_token_id` parameter to the `generate` method.
*Note: `M2M100Tokenizer` depends on `sentencepiece`, so make sure to install it before running the example.*
To install `sentencepiece` run `pip install sentencepiece`
```python
from transformers import M2M100ForConditionalGeneration, M2M100Tokenizer
hi_text = "जीवन एक चॉकलेट बॉक्स की तरह है।"
chinese_text = "生活就像一盒巧克力。"
model = M2M100ForConditionalGeneration.from_pretrained("facebook/m2m100_1.2B")
tokenizer = M2M100Tokenizer.from_pretrained("facebook/m2m100_1.2B")
# translate Hindi to French
tokenizer.src_lang = "hi"
encoded_hi = tokenizer(hi_text, return_tensors="pt")
generated_tokens = model.generate(**encoded_hi, forced_bos_token_id=tokenizer.get_lang_id("fr"))
tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
# => "La vie est comme une boîte de chocolat."
# translate Chinese to English
tokenizer.src_lang = "zh"
encoded_zh = tokenizer(chinese_text, return_tensors="pt")
generated_tokens = model.generate(**encoded_zh, forced_bos_token_id=tokenizer.get_lang_id("en"))
tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
# => "Life is like a box of chocolate."
```
See the [model hub](https://huggingface.co/models?filter=m2m_100) to look for more fine-tuned versions.
## Languages covered
Afrikaans (af), Amharic (am), Arabic (ar), Asturian (ast), Azerbaijani (az), Bashkir (ba), Belarusian (be), Bulgarian (bg), Bengali (bn), Breton (br), Bosnian (bs), Catalan; Valencian (ca), Cebuano (ceb), Czech (cs), Welsh (cy), Danish (da), German (de), Greeek (el), English (en), Spanish (es), Estonian (et), Persian (fa), Fulah (ff), Finnish (fi), French (fr), Western Frisian (fy), Irish (ga), Gaelic; Scottish Gaelic (gd), Galician (gl), Gujarati (gu), Hausa (ha), Hebrew (he), Hindi (hi), Croatian (hr), Haitian; Haitian Creole (ht), Hungarian (hu), Armenian (hy), Indonesian (id), Igbo (ig), Iloko (ilo), Icelandic (is), Italian (it), Japanese (ja), Javanese (jv), Georgian (ka), Kazakh (kk), Central Khmer (km), Kannada (kn), Korean (ko), Luxembourgish; Letzeburgesch (lb), Ganda (lg), Lingala (ln), Lao (lo), Lithuanian (lt), Latvian (lv), Malagasy (mg), Macedonian (mk), Malayalam (ml), Mongolian (mn), Marathi (mr), Malay (ms), Burmese (my), Nepali (ne), Dutch; Flemish (nl), Norwegian (no), Northern Sotho (ns), Occitan (post 1500) (oc), Oriya (or), Panjabi; Punjabi (pa), Polish (pl), Pushto; Pashto (ps), Portuguese (pt), Romanian; Moldavian; Moldovan (ro), Russian (ru), Sindhi (sd), Sinhala; Sinhalese (si), Slovak (sk), Slovenian (sl), Somali (so), Albanian (sq), Serbian (sr), Swati (ss), Sundanese (su), Swedish (sv), Swahili (sw), Tamil (ta), Thai (th), Tagalog (tl), Tswana (tn), Turkish (tr), Ukrainian (uk), Urdu (ur), Uzbek (uz), Vietnamese (vi), Wolof (wo), Xhosa (xh), Yiddish (yi), Yoruba (yo), Chinese (zh), Zulu (zu)
## BibTeX entry and citation info
```
@misc{fan2020englishcentric,
title={Beyond English-Centric Multilingual Machine Translation},
author={Angela Fan and Shruti Bhosale and Holger Schwenk and Zhiyi Ma and Ahmed El-Kishky and Siddharth Goyal and Mandeep Baines and Onur Celebi and Guillaume Wenzek and Vishrav Chaudhary and Naman Goyal and Tom Birch and Vitaliy Liptchinsky and Sergey Edunov and Edouard Grave and Michael Auli and Armand Joulin},
year={2020},
eprint={2010.11125},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
|
{"language": ["multilingual", "af", "am", "ar", "ast", "az", "ba", "be", "bg", "bn", "br", "bs", "ca", "ceb", "cs", "cy", "da", "de", "el", "en", "es", "et", "fa", "ff", "fi", "fr", "fy", "ga", "gd", "gl", "gu", "ha", "he", "hi", "hr", "ht", "hu", "hy", "id", "ig", "ilo", "is", "it", "ja", "jv", "ka", "kk", "km", "kn", "ko", "lb", "lg", "ln", "lo", "lt", "lv", "mg", "mk", "ml", "mn", "mr", "ms", "my", "ne", "nl", false, "ns", "oc", "or", "pa", "pl", "ps", "pt", "ro", "ru", "sd", "si", "sk", "sl", "so", "sq", "sr", "ss", "su", "sv", "sw", "ta", "th", "tl", "tn", "tr", "uk", "ur", "uz", "vi", "wo", "xh", "yi", "yo", "zh", "zu"], "license": "mit"}
|
facebook/m2m100_1.2B
| null |
[
"transformers",
"pytorch",
"rust",
"m2m_100",
"text2text-generation",
"multilingual",
"af",
"am",
"ar",
"ast",
"az",
"ba",
"be",
"bg",
"bn",
"br",
"bs",
"ca",
"ceb",
"cs",
"cy",
"da",
"de",
"el",
"en",
"es",
"et",
"fa",
"ff",
"fi",
"fr",
"fy",
"ga",
"gd",
"gl",
"gu",
"ha",
"he",
"hi",
"hr",
"ht",
"hu",
"hy",
"id",
"ig",
"ilo",
"is",
"it",
"ja",
"jv",
"ka",
"kk",
"km",
"kn",
"ko",
"lb",
"lg",
"ln",
"lo",
"lt",
"lv",
"mg",
"mk",
"ml",
"mn",
"mr",
"ms",
"my",
"ne",
"nl",
"no",
"ns",
"oc",
"or",
"pa",
"pl",
"ps",
"pt",
"ro",
"ru",
"sd",
"si",
"sk",
"sl",
"so",
"sq",
"sr",
"ss",
"su",
"sv",
"sw",
"ta",
"th",
"tl",
"tn",
"tr",
"uk",
"ur",
"uz",
"vi",
"wo",
"xh",
"yi",
"yo",
"zh",
"zu",
"arxiv:2010.11125",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
# M2M100 418M
M2M100 is a multilingual encoder-decoder (seq-to-seq) model trained for Many-to-Many multilingual translation.
It was introduced in this [paper](https://arxiv.org/abs/2010.11125) and first released in [this](https://github.com/pytorch/fairseq/tree/master/examples/m2m_100) repository.
The model that can directly translate between the 9,900 directions of 100 languages.
To translate into a target language, the target language id is forced as the first generated token.
To force the target language id as the first generated token, pass the `forced_bos_token_id` parameter to the `generate` method.
*Note: `M2M100Tokenizer` depends on `sentencepiece`, so make sure to install it before running the example.*
To install `sentencepiece` run `pip install sentencepiece`
```python
from transformers import M2M100ForConditionalGeneration, M2M100Tokenizer
hi_text = "जीवन एक चॉकलेट बॉक्स की तरह है।"
chinese_text = "生活就像一盒巧克力。"
model = M2M100ForConditionalGeneration.from_pretrained("facebook/m2m100_418M")
tokenizer = M2M100Tokenizer.from_pretrained("facebook/m2m100_418M")
# translate Hindi to French
tokenizer.src_lang = "hi"
encoded_hi = tokenizer(hi_text, return_tensors="pt")
generated_tokens = model.generate(**encoded_hi, forced_bos_token_id=tokenizer.get_lang_id("fr"))
tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
# => "La vie est comme une boîte de chocolat."
# translate Chinese to English
tokenizer.src_lang = "zh"
encoded_zh = tokenizer(chinese_text, return_tensors="pt")
generated_tokens = model.generate(**encoded_zh, forced_bos_token_id=tokenizer.get_lang_id("en"))
tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
# => "Life is like a box of chocolate."
```
See the [model hub](https://huggingface.co/models?filter=m2m_100) to look for more fine-tuned versions.
## Languages covered
Afrikaans (af), Amharic (am), Arabic (ar), Asturian (ast), Azerbaijani (az), Bashkir (ba), Belarusian (be), Bulgarian (bg), Bengali (bn), Breton (br), Bosnian (bs), Catalan; Valencian (ca), Cebuano (ceb), Czech (cs), Welsh (cy), Danish (da), German (de), Greeek (el), English (en), Spanish (es), Estonian (et), Persian (fa), Fulah (ff), Finnish (fi), French (fr), Western Frisian (fy), Irish (ga), Gaelic; Scottish Gaelic (gd), Galician (gl), Gujarati (gu), Hausa (ha), Hebrew (he), Hindi (hi), Croatian (hr), Haitian; Haitian Creole (ht), Hungarian (hu), Armenian (hy), Indonesian (id), Igbo (ig), Iloko (ilo), Icelandic (is), Italian (it), Japanese (ja), Javanese (jv), Georgian (ka), Kazakh (kk), Central Khmer (km), Kannada (kn), Korean (ko), Luxembourgish; Letzeburgesch (lb), Ganda (lg), Lingala (ln), Lao (lo), Lithuanian (lt), Latvian (lv), Malagasy (mg), Macedonian (mk), Malayalam (ml), Mongolian (mn), Marathi (mr), Malay (ms), Burmese (my), Nepali (ne), Dutch; Flemish (nl), Norwegian (no), Northern Sotho (ns), Occitan (post 1500) (oc), Oriya (or), Panjabi; Punjabi (pa), Polish (pl), Pushto; Pashto (ps), Portuguese (pt), Romanian; Moldavian; Moldovan (ro), Russian (ru), Sindhi (sd), Sinhala; Sinhalese (si), Slovak (sk), Slovenian (sl), Somali (so), Albanian (sq), Serbian (sr), Swati (ss), Sundanese (su), Swedish (sv), Swahili (sw), Tamil (ta), Thai (th), Tagalog (tl), Tswana (tn), Turkish (tr), Ukrainian (uk), Urdu (ur), Uzbek (uz), Vietnamese (vi), Wolof (wo), Xhosa (xh), Yiddish (yi), Yoruba (yo), Chinese (zh), Zulu (zu)
## BibTeX entry and citation info
```
@misc{fan2020englishcentric,
title={Beyond English-Centric Multilingual Machine Translation},
author={Angela Fan and Shruti Bhosale and Holger Schwenk and Zhiyi Ma and Ahmed El-Kishky and Siddharth Goyal and Mandeep Baines and Onur Celebi and Guillaume Wenzek and Vishrav Chaudhary and Naman Goyal and Tom Birch and Vitaliy Liptchinsky and Sergey Edunov and Edouard Grave and Michael Auli and Armand Joulin},
year={2020},
eprint={2010.11125},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
|
{"language": ["multilingual", "af", "am", "ar", "ast", "az", "ba", "be", "bg", "bn", "br", "bs", "ca", "ceb", "cs", "cy", "da", "de", "el", "en", "es", "et", "fa", "ff", "fi", "fr", "fy", "ga", "gd", "gl", "gu", "ha", "he", "hi", "hr", "ht", "hu", "hy", "id", "ig", "ilo", "is", "it", "ja", "jv", "ka", "kk", "km", "kn", "ko", "lb", "lg", "ln", "lo", "lt", "lv", "mg", "mk", "ml", "mn", "mr", "ms", "my", "ne", "nl", false, "ns", "oc", "or", "pa", "pl", "ps", "pt", "ro", "ru", "sd", "si", "sk", "sl", "so", "sq", "sr", "ss", "su", "sv", "sw", "ta", "th", "tl", "tn", "tr", "uk", "ur", "uz", "vi", "wo", "xh", "yi", "yo", "zh", "zu"], "license": "mit"}
|
facebook/m2m100_418M
| null |
[
"transformers",
"pytorch",
"rust",
"m2m_100",
"text2text-generation",
"multilingual",
"af",
"am",
"ar",
"ast",
"az",
"ba",
"be",
"bg",
"bn",
"br",
"bs",
"ca",
"ceb",
"cs",
"cy",
"da",
"de",
"el",
"en",
"es",
"et",
"fa",
"ff",
"fi",
"fr",
"fy",
"ga",
"gd",
"gl",
"gu",
"ha",
"he",
"hi",
"hr",
"ht",
"hu",
"hy",
"id",
"ig",
"ilo",
"is",
"it",
"ja",
"jv",
"ka",
"kk",
"km",
"kn",
"ko",
"lb",
"lg",
"ln",
"lo",
"lt",
"lv",
"mg",
"mk",
"ml",
"mn",
"mr",
"ms",
"my",
"ne",
"nl",
"no",
"ns",
"oc",
"or",
"pa",
"pl",
"ps",
"pt",
"ro",
"ru",
"sd",
"si",
"sk",
"sl",
"so",
"sq",
"sr",
"ss",
"su",
"sv",
"sw",
"ta",
"th",
"tl",
"tn",
"tr",
"uk",
"ur",
"uz",
"vi",
"wo",
"xh",
"yi",
"yo",
"zh",
"zu",
"arxiv:2010.11125",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-segmentation
|
transformers
|
# MaskFormer
MaskFormer model trained on ADE20k semantic segmentation (base-sized version, Swin backbone). It was introduced in the paper [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://arxiv.org/abs/2107.06278) and first released in [this repository](https://github.com/facebookresearch/MaskFormer/blob/da3e60d85fdeedcb31476b5edd7d328826ce56cc/mask_former/modeling/criterion.py#L169).
Disclaimer: The team releasing MaskFormer did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
MaskFormer addresses instance, semantic and panoptic segmentation with the same paradigm: by predicting a set of masks and corresponding labels. Hence, all 3 tasks are treated as if they were instance segmentation.
## Intended uses & limitations
You can use this particular checkpoint for semantic segmentation. See the [model hub](https://huggingface.co/models?search=maskformer) to look for other
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model:
```python
from transformers import MaskFormerFeatureExtractor, MaskFormerForInstanceSegmentation
from PIL import Image
import requests
url = "https://huggingface.co/datasets/hf-internal-testing/fixtures_ade20k/resolve/main/ADE_val_00000001.jpg"
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = MaskFormerFeatureExtractor.from_pretrained("facebook/maskformer-swin-base-ade")
inputs = feature_extractor(images=image, return_tensors="pt")
model = MaskFormerForInstanceSegmentation.from_pretrained("facebook/maskformer-swin-base-ade")
outputs = model(**inputs)
# model predicts class_queries_logits of shape `(batch_size, num_queries)`
# and masks_queries_logits of shape `(batch_size, num_queries, height, width)`
class_queries_logits = outputs.class_queries_logits
masks_queries_logits = outputs.masks_queries_logits
# you can pass them to feature_extractor for postprocessing
# we refer to the demo notebooks for visualization (see "Resources" section in the MaskFormer docs)
predicted_semantic_map = feature_extractor.post_process_semantic_segmentation(outputs, target_sizes=[image.size[::-1]])[0]
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/maskformer).
|
{"license": "other", "tags": ["vision", "image-segmentation"], "datasets": ["scene_parse_150"], "widget": [{"src": "https://huggingface.co/datasets/hf-internal-testing/fixtures_ade20k/resolve/main/ADE_val_00000001.jpg", "example_title": "House"}, {"src": "https://huggingface.co/datasets/hf-internal-testing/fixtures_ade20k/resolve/main/ADE_val_00000002.jpg", "example_title": "Castle"}]}
|
facebook/maskformer-swin-base-ade
| null |
[
"transformers",
"pytorch",
"maskformer",
"vision",
"image-segmentation",
"dataset:scene_parse_150",
"arxiv:2107.06278",
"license:other",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-segmentation
|
transformers
|
# MaskFormer
MaskFormer model trained on COCO panoptic segmentation (base-sized version, Swin backbone). It was introduced in the paper [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://arxiv.org/abs/2107.06278) and first released in [this repository](https://github.com/facebookresearch/MaskFormer/blob/da3e60d85fdeedcb31476b5edd7d328826ce56cc/mask_former/modeling/criterion.py#L169).
Disclaimer: The team releasing MaskFormer did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
MaskFormer addresses instance, semantic and panoptic segmentation with the same paradigm: by predicting a set of masks and corresponding labels. Hence, all 3 tasks are treated as if they were instance segmentation.
## Intended uses & limitations
You can use this particular checkpoint for semantic segmentation. See the [model hub](https://huggingface.co/models?search=maskformer) to look for other
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model:
```python
from transformers import MaskFormerFeatureExtractor, MaskFormerForInstanceSegmentation
from PIL import Image
import requests
# load MaskFormer fine-tuned on COCO panoptic segmentation
feature_extractor = MaskFormerFeatureExtractor.from_pretrained("facebook/maskformer-swin-base-coco")
model = MaskFormerForInstanceSegmentation.from_pretrained("facebook/maskformer-swin-base-coco")
url = "http://images.cocodataset.org/val2017/000000039769.jpg"
image = Image.open(requests.get(url, stream=True).raw)
inputs = feature_extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
# model predicts class_queries_logits of shape `(batch_size, num_queries)`
# and masks_queries_logits of shape `(batch_size, num_queries, height, width)`
class_queries_logits = outputs.class_queries_logits
masks_queries_logits = outputs.masks_queries_logits
# you can pass them to feature_extractor for postprocessing
result = feature_extractor.post_process_panoptic_segmentation(outputs, target_sizes=[image.size[::-1]])[0]
# we refer to the demo notebooks for visualization (see "Resources" section in the MaskFormer docs)
predicted_panoptic_map = result["segmentation"]
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/maskformer).
|
{"license": "other", "tags": ["vision", "image-segmentation"], "datasets": ["coco"], "widget": [{"src": "http://images.cocodataset.org/val2017/000000039769.jpg", "example_title": "Cats"}, {"src": "http://images.cocodataset.org/val2017/000000039770.jpg", "example_title": "Castle"}]}
|
facebook/maskformer-swin-base-coco
| null |
[
"transformers",
"pytorch",
"maskformer",
"vision",
"image-segmentation",
"dataset:coco",
"arxiv:2107.06278",
"license:other",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-segmentation
|
transformers
|
# MaskFormer
MaskFormer model trained on ADE20k semantic segmentation (large-sized version, Swin backbone). It was introduced in the paper [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://arxiv.org/abs/2107.06278) and first released in [this repository](https://github.com/facebookresearch/MaskFormer/blob/da3e60d85fdeedcb31476b5edd7d328826ce56cc/mask_former/modeling/criterion.py#L169).
Disclaimer: The team releasing MaskFormer did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
MaskFormer addresses instance, semantic and panoptic segmentation with the same paradigm: by predicting a set of masks and corresponding labels. Hence, all 3 tasks are treated as if they were instance segmentation.
## Intended uses & limitations
You can use this particular checkpoint for semantic segmentation. See the [model hub](https://huggingface.co/models?search=maskformer) to look for other
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model:
```python
from transformers import MaskFormerImageProcessor, MaskFormerForInstanceSegmentation
from PIL import Image
import requests
url = "https://huggingface.co/datasets/hf-internal-testing/fixtures_ade20k/resolve/main/ADE_val_00000001.jpg"
image = Image.open(requests.get(url, stream=True).raw)
processor = MaskFormerImageProcessor.from_pretrained("facebook/maskformer-swin-large-ade")
inputs = processor(images=image, return_tensors="pt")
model = MaskFormerForInstanceSegmentation.from_pretrained("facebook/maskformer-swin-large-ade")
outputs = model(**inputs)
# model predicts class_queries_logits of shape `(batch_size, num_queries)`
# and masks_queries_logits of shape `(batch_size, num_queries, height, width)`
class_queries_logits = outputs.class_queries_logits
masks_queries_logits = outputs.masks_queries_logits
# you can pass them to processor for postprocessing
# we refer to the demo notebooks for visualization (see "Resources" section in the MaskFormer docs)
predicted_semantic_map = processor.post_process_semantic_segmentation(outputs, target_sizes=[image.size[::-1]])[0]
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/maskformer).
|
{"license": "other", "tags": ["vision", "image-segmentation"], "datasets": ["scene_parse_150"], "widget": [{"src": "https://huggingface.co/datasets/hf-internal-testing/fixtures_ade20k/resolve/main/ADE_val_00000001.jpg", "example_title": "House"}, {"src": "https://huggingface.co/datasets/hf-internal-testing/fixtures_ade20k/resolve/main/ADE_val_00000002.jpg", "example_title": "Castle"}]}
|
facebook/maskformer-swin-large-ade
| null |
[
"transformers",
"pytorch",
"maskformer",
"vision",
"image-segmentation",
"dataset:scene_parse_150",
"arxiv:2107.06278",
"license:other",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-segmentation
|
transformers
|
# MaskFormer
MaskFormer model trained on COCO panoptic segmentation (large-sized version, Swin backbone). It was introduced in the paper [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://arxiv.org/abs/2107.06278) and first released in [this repository](https://github.com/facebookresearch/MaskFormer/blob/da3e60d85fdeedcb31476b5edd7d328826ce56cc/mask_former/modeling/criterion.py#L169).
Disclaimer: The team releasing MaskFormer did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
MaskFormer addresses instance, semantic and panoptic segmentation with the same paradigm: by predicting a set of masks and corresponding labels. Hence, all 3 tasks are treated as if they were instance segmentation.
## Intended uses & limitations
You can use this particular checkpoint for semantic segmentation. See the [model hub](https://huggingface.co/models?search=maskformer) to look for other
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model:
```python
from transformers import MaskFormerImageProcessor, MaskFormerForInstanceSegmentation
from PIL import Image
import requests
# load MaskFormer fine-tuned on COCO panoptic segmentation
processor = MaskFormerImageProcessor.from_pretrained("facebook/maskformer-swin-large-coco")
model = MaskFormerForInstanceSegmentation.from_pretrained("facebook/maskformer-swin-large-coco")
url = "http://images.cocodataset.org/val2017/000000039769.jpg"
image = Image.open(requests.get(url, stream=True).raw)
inputs = processor(images=image, return_tensors="pt")
outputs = model(**inputs)
# model predicts class_queries_logits of shape `(batch_size, num_queries)`
# and masks_queries_logits of shape `(batch_size, num_queries, height, width)`
class_queries_logits = outputs.class_queries_logits
masks_queries_logits = outputs.masks_queries_logits
# you can pass them to processor for postprocessing
result = processor.post_process_panoptic_segmentation(outputs, target_sizes=[image.size[::-1]])[0]
# we refer to the demo notebooks for visualization (see "Resources" section in the MaskFormer docs)
predicted_panoptic_map = result["segmentation"]
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/maskformer).
|
{"license": "other", "tags": ["vision", "image-segmentation"], "datasets": ["coco"], "widget": [{"src": "http://images.cocodataset.org/val2017/000000039769.jpg", "example_title": "Cats"}, {"src": "http://images.cocodataset.org/val2017/000000039770.jpg", "example_title": "Castle"}]}
|
facebook/maskformer-swin-large-coco
| null |
[
"transformers",
"pytorch",
"safetensors",
"maskformer",
"vision",
"image-segmentation",
"dataset:coco",
"arxiv:2107.06278",
"license:other",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-segmentation
|
transformers
|
# MaskFormer
MaskFormer model trained on ADE20k semantic segmentation (small-sized version, Swin backbone). It was introduced in the paper [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://arxiv.org/abs/2107.06278) and first released in [this repository](https://github.com/facebookresearch/MaskFormer/blob/da3e60d85fdeedcb31476b5edd7d328826ce56cc/mask_former/modeling/criterion.py#L169).
Disclaimer: The team releasing MaskFormer did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
MaskFormer addresses instance, semantic and panoptic segmentation with the same paradigm: by predicting a set of masks and corresponding labels. Hence, all 3 tasks are treated as if they were instance segmentation.
## Intended uses & limitations
You can use this particular checkpoint for semantic segmentation. See the [model hub](https://huggingface.co/models?search=maskformer) to look for other
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model:
```python
from transformers import MaskFormerFeatureExtractor, MaskFormerForInstanceSegmentation
from PIL import Image
import requests
url = "https://huggingface.co/datasets/hf-internal-testing/fixtures_ade20k/resolve/main/ADE_val_00000001.jpg"
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = MaskFormerFeatureExtractor.from_pretrained("facebook/maskformer-swin-small-ade")
inputs = feature_extractor(images=image, return_tensors="pt")
model = MaskFormerForInstanceSegmentation.from_pretrained("facebook/maskformer-swin-small-ade")
outputs = model(**inputs)
# model predicts class_queries_logits of shape `(batch_size, num_queries)`
# and masks_queries_logits of shape `(batch_size, num_queries, height, width)`
class_queries_logits = outputs.class_queries_logits
masks_queries_logits = outputs.masks_queries_logits
# you can pass them to feature_extractor for postprocessing
# we refer to the demo notebooks for visualization (see "Resources" section in the MaskFormer docs)
predicted_semantic_map = feature_extractor.post_process_semantic_segmentation(outputs, target_sizes=[image.size[::-1]])[0]
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/maskformer).
|
{"license": "other", "tags": ["vision", "image-segmentation"], "datasets": ["scene_parse_150"], "widget": [{"src": "https://huggingface.co/datasets/hf-internal-testing/fixtures_ade20k/resolve/main/ADE_val_00000001.jpg", "example_title": "House"}, {"src": "https://huggingface.co/datasets/hf-internal-testing/fixtures_ade20k/resolve/main/ADE_val_00000002.jpg", "example_title": "Castle"}]}
|
facebook/maskformer-swin-small-ade
| null |
[
"transformers",
"pytorch",
"safetensors",
"maskformer",
"vision",
"image-segmentation",
"dataset:scene_parse_150",
"arxiv:2107.06278",
"license:other",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-segmentation
|
transformers
|
# MaskFormer
MaskFormer model trained on COCO panoptic segmentation (small-sized version, Swin backbone). It was introduced in the paper [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://arxiv.org/abs/2107.06278) and first released in [this repository](https://github.com/facebookresearch/MaskFormer/blob/da3e60d85fdeedcb31476b5edd7d328826ce56cc/mask_former/modeling/criterion.py#L169).
Disclaimer: The team releasing MaskFormer did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
MaskFormer addresses instance, semantic and panoptic segmentation with the same paradigm: by predicting a set of masks and corresponding labels. Hence, all 3 tasks are treated as if they were instance segmentation.
## Intended uses & limitations
You can use this particular checkpoint for semantic segmentation. See the [model hub](https://huggingface.co/models?search=maskformer) to look for other
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model:
```python
from transformers import MaskFormerFeatureExtractor, MaskFormerForInstanceSegmentation
from PIL import Image
import requests
# load MaskFormer fine-tuned on COCO panoptic segmentation
feature_extractor = MaskFormerFeatureExtractor.from_pretrained("facebook/maskformer-swin-small-coco")
model = MaskFormerForInstanceSegmentation.from_pretrained("facebook/maskformer-swin-small-coco")
url = "http://images.cocodataset.org/val2017/000000039769.jpg"
image = Image.open(requests.get(url, stream=True).raw)
inputs = feature_extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
# model predicts class_queries_logits of shape `(batch_size, num_queries)`
# and masks_queries_logits of shape `(batch_size, num_queries, height, width)`
class_queries_logits = outputs.class_queries_logits
masks_queries_logits = outputs.masks_queries_logits
# you can pass them to feature_extractor for postprocessing
result = feature_extractor.post_process_panoptic_segmentation(outputs, target_sizes=[image.size[::-1]])[0]
# we refer to the demo notebooks for visualization (see "Resources" section in the MaskFormer docs)
predicted_panoptic_map = result["segmentation"]
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/maskformer).
|
{"license": "other", "tags": ["vision", "image-segmentation"], "datasets": ["coco"], "widget": [{"src": "http://images.cocodataset.org/val2017/000000039769.jpg", "example_title": "Cats"}, {"src": "http://images.cocodataset.org/val2017/000000039770.jpg", "example_title": "Castle"}]}
|
facebook/maskformer-swin-small-coco
| null |
[
"transformers",
"pytorch",
"safetensors",
"maskformer",
"vision",
"image-segmentation",
"dataset:coco",
"arxiv:2107.06278",
"license:other",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-segmentation
|
transformers
|
# MaskFormer
MaskFormer model trained on ADE20k semantic segmentation (tiny-sized version, Swin backbone). It was introduced in the paper [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://arxiv.org/abs/2107.06278) and first released in [this repository](https://github.com/facebookresearch/MaskFormer/blob/da3e60d85fdeedcb31476b5edd7d328826ce56cc/mask_former/modeling/criterion.py#L169).
Disclaimer: The team releasing MaskFormer did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
MaskFormer addresses instance, semantic and panoptic segmentation with the same paradigm: by predicting a set of masks and corresponding labels. Hence, all 3 tasks are treated as if they were instance segmentation.
## Intended uses & limitations
You can use this particular checkpoint for semantic segmentation. See the [model hub](https://huggingface.co/models?search=maskformer) to look for other
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model:
```python
from transformers import MaskFormerFeatureExtractor, MaskFormerForInstanceSegmentation
from PIL import Image
import requests
url = "https://huggingface.co/datasets/hf-internal-testing/fixtures_ade20k/resolve/main/ADE_val_00000001.jpg"
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = MaskFormerFeatureExtractor.from_pretrained("facebook/maskformer-swin-tiny-ade")
inputs = feature_extractor(images=image, return_tensors="pt")
model = MaskFormerForInstanceSegmentation.from_pretrained("facebook/maskformer-swin-tiny-ade")
outputs = model(**inputs)
# model predicts class_queries_logits of shape `(batch_size, num_queries)`
# and masks_queries_logits of shape `(batch_size, num_queries, height, width)`
class_queries_logits = outputs.class_queries_logits
masks_queries_logits = outputs.masks_queries_logits
# you can pass them to feature_extractor for postprocessing
# we refer to the demo notebooks for visualization (see "Resources" section in the MaskFormer docs)
predicted_semantic_map = feature_extractor.post_process_semantic_segmentation(outputs, target_sizes=[image.size[::-1]])[0]
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/maskformer).
|
{"license": "other", "tags": ["vision", "image-segmentation"], "datasets": ["scene_parse_150"], "widget": [{"src": "https://huggingface.co/datasets/hf-internal-testing/fixtures_ade20k/resolve/main/ADE_val_00000001.jpg", "example_title": "House"}, {"src": "https://huggingface.co/datasets/hf-internal-testing/fixtures_ade20k/resolve/main/ADE_val_00000002.jpg", "example_title": "Castle"}]}
|
facebook/maskformer-swin-tiny-ade
| null |
[
"transformers",
"pytorch",
"safetensors",
"maskformer",
"vision",
"image-segmentation",
"dataset:scene_parse_150",
"arxiv:2107.06278",
"license:other",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
image-segmentation
|
transformers
|
# MaskFormer
MaskFormer model trained on COCO panoptic segmentation (tiny-sized version, Swin backbone). It was introduced in the paper [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://arxiv.org/abs/2107.06278) and first released in [this repository](https://github.com/facebookresearch/MaskFormer/blob/da3e60d85fdeedcb31476b5edd7d328826ce56cc/mask_former/modeling/criterion.py#L169).
Disclaimer: The team releasing MaskFormer did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
MaskFormer addresses instance, semantic and panoptic segmentation with the same paradigm: by predicting a set of masks and corresponding labels. Hence, all 3 tasks are treated as if they were instance segmentation.
## Intended uses & limitations
You can use this particular checkpoint for semantic segmentation. See the [model hub](https://huggingface.co/models?search=maskformer) to look for other
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model:
```python
from transformers import MaskFormerFeatureExtractor, MaskFormerForInstanceSegmentation
from PIL import Image
import requests
# load MaskFormer fine-tuned on COCO panoptic segmentation
feature_extractor = MaskFormerFeatureExtractor.from_pretrained("facebook/maskformer-swin-tiny-coco")
model = MaskFormerForInstanceSegmentation.from_pretrained("facebook/maskformer-swin-tiny-coco")
url = "http://images.cocodataset.org/val2017/000000039769.jpg"
image = Image.open(requests.get(url, stream=True).raw)
inputs = feature_extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
# model predicts class_queries_logits of shape `(batch_size, num_queries)`
# and masks_queries_logits of shape `(batch_size, num_queries, height, width)`
class_queries_logits = outputs.class_queries_logits
masks_queries_logits = outputs.masks_queries_logits
# you can pass them to feature_extractor for postprocessing
result = feature_extractor.post_process_panoptic_segmentation(outputs, target_sizes=[image.size[::-1]])[0]
# we refer to the demo notebooks for visualization (see "Resources" section in the MaskFormer docs)
predicted_panoptic_map = result["segmentation"]
```
For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/maskformer).
|
{"license": "other", "tags": ["vision", "image-segmentation"], "datasets": ["coco"], "widget": [{"src": "http://images.cocodataset.org/val2017/000000039769.jpg", "example_title": "Cats"}, {"src": "http://images.cocodataset.org/val2017/000000039770.jpg", "example_title": "Castle"}]}
|
facebook/maskformer-swin-tiny-coco
| null |
[
"transformers",
"pytorch",
"safetensors",
"maskformer",
"vision",
"image-segmentation",
"dataset:coco",
"arxiv:2107.06278",
"license:other",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
translation
|
transformers
|
# mBART-50 many to many multilingual machine translation
This model is a fine-tuned checkpoint of [mBART-large-50](https://huggingface.co/facebook/mbart-large-50). `mbart-large-50-many-to-many-mmt` is fine-tuned for multilingual machine translation. It was introduced in [Multilingual Translation with Extensible Multilingual Pretraining and Finetuning](https://arxiv.org/abs/2008.00401) paper.
The model can translate directly between any pair of 50 languages. To translate into a target language, the target language id is forced as the first generated token. To force the target language id as the first generated token, pass the `forced_bos_token_id` parameter to the `generate` method.
```python
from transformers import MBartForConditionalGeneration, MBart50TokenizerFast
article_hi = "संयुक्त राष्ट्र के प्रमुख का कहना है कि सीरिया में कोई सैन्य समाधान नहीं है"
article_ar = "الأمين العام للأمم المتحدة يقول إنه لا يوجد حل عسكري في سوريا."
model = MBartForConditionalGeneration.from_pretrained("facebook/mbart-large-50-many-to-many-mmt")
tokenizer = MBart50TokenizerFast.from_pretrained("facebook/mbart-large-50-many-to-many-mmt")
# translate Hindi to French
tokenizer.src_lang = "hi_IN"
encoded_hi = tokenizer(article_hi, return_tensors="pt")
generated_tokens = model.generate(
**encoded_hi,
forced_bos_token_id=tokenizer.lang_code_to_id["fr_XX"]
)
tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
# => "Le chef de l 'ONU affirme qu 'il n 'y a pas de solution militaire dans la Syrie."
# translate Arabic to English
tokenizer.src_lang = "ar_AR"
encoded_ar = tokenizer(article_ar, return_tensors="pt")
generated_tokens = model.generate(
**encoded_ar,
forced_bos_token_id=tokenizer.lang_code_to_id["en_XX"]
)
tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
# => "The Secretary-General of the United Nations says there is no military solution in Syria."
```
See the [model hub](https://huggingface.co/models?filter=mbart-50) to look for more fine-tuned versions.
## Languages covered
Arabic (ar_AR), Czech (cs_CZ), German (de_DE), English (en_XX), Spanish (es_XX), Estonian (et_EE), Finnish (fi_FI), French (fr_XX), Gujarati (gu_IN), Hindi (hi_IN), Italian (it_IT), Japanese (ja_XX), Kazakh (kk_KZ), Korean (ko_KR), Lithuanian (lt_LT), Latvian (lv_LV), Burmese (my_MM), Nepali (ne_NP), Dutch (nl_XX), Romanian (ro_RO), Russian (ru_RU), Sinhala (si_LK), Turkish (tr_TR), Vietnamese (vi_VN), Chinese (zh_CN), Afrikaans (af_ZA), Azerbaijani (az_AZ), Bengali (bn_IN), Persian (fa_IR), Hebrew (he_IL), Croatian (hr_HR), Indonesian (id_ID), Georgian (ka_GE), Khmer (km_KH), Macedonian (mk_MK), Malayalam (ml_IN), Mongolian (mn_MN), Marathi (mr_IN), Polish (pl_PL), Pashto (ps_AF), Portuguese (pt_XX), Swedish (sv_SE), Swahili (sw_KE), Tamil (ta_IN), Telugu (te_IN), Thai (th_TH), Tagalog (tl_XX), Ukrainian (uk_UA), Urdu (ur_PK), Xhosa (xh_ZA), Galician (gl_ES), Slovene (sl_SI)
## BibTeX entry and citation info
```
@article{tang2020multilingual,
title={Multilingual Translation with Extensible Multilingual Pretraining and Finetuning},
author={Yuqing Tang and Chau Tran and Xian Li and Peng-Jen Chen and Naman Goyal and Vishrav Chaudhary and Jiatao Gu and Angela Fan},
year={2020},
eprint={2008.00401},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
|
{"language": ["multilingual", "ar", "cs", "de", "en", "es", "et", "fi", "fr", "gu", "hi", "it", "ja", "kk", "ko", "lt", "lv", "my", "ne", "nl", "ro", "ru", "si", "tr", "vi", "zh", "af", "az", "bn", "fa", "he", "hr", "id", "ka", "km", "mk", "ml", "mn", "mr", "pl", "ps", "pt", "sv", "sw", "ta", "te", "th", "tl", "uk", "ur", "xh", "gl", "sl"], "tags": ["mbart-50"], "pipeline_tag": "translation"}
|
facebook/mbart-large-50-many-to-many-mmt
| null |
[
"transformers",
"pytorch",
"tf",
"jax",
"rust",
"safetensors",
"mbart",
"text2text-generation",
"mbart-50",
"translation",
"multilingual",
"ar",
"cs",
"de",
"en",
"es",
"et",
"fi",
"fr",
"gu",
"hi",
"it",
"ja",
"kk",
"ko",
"lt",
"lv",
"my",
"ne",
"nl",
"ro",
"ru",
"si",
"tr",
"vi",
"zh",
"af",
"az",
"bn",
"fa",
"he",
"hr",
"id",
"ka",
"km",
"mk",
"ml",
"mn",
"mr",
"pl",
"ps",
"pt",
"sv",
"sw",
"ta",
"te",
"th",
"tl",
"uk",
"ur",
"xh",
"gl",
"sl",
"arxiv:2008.00401",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
# mBART-50 many to one multilingual machine translation
This model is a fine-tuned checkpoint of [mBART-large-50](https://huggingface.co/facebook/mbart-large-50). `mbart-large-50-many-to-many-mmt` is fine-tuned for multilingual machine translation. It was introduced in [Multilingual Translation with Extensible Multilingual Pretraining and Finetuning](https://arxiv.org/abs/2008.00401) paper.
The model can translate directly between any pair of 50 languages.
```python
from transformers import MBartForConditionalGeneration, MBart50TokenizerFast
article_hi = "संयुक्त राष्ट्र के प्रमुख का कहना है कि सीरिया में कोई सैन्य समाधान नहीं है"
article_ar = "الأمين العام للأمم المتحدة يقول إنه لا يوجد حل عسكري في سوريا."
model = MBartForConditionalGeneration.from_pretrained("facebook/mbart-large-50-many-to-one-mmt")
tokenizer = MBart50TokenizerFast.from_pretrained("facebook/mbart-large-50-many-to-one-mmt")
# translate Hindi to English
tokenizer.src_lang = "hi_IN"
encoded_hi = tokenizer(article_hi, return_tensors="pt")
generated_tokens = model.generate(**encoded_hi)
tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
# => "The head of the UN says there is no military solution in Syria."
# translate Arabic to English
tokenizer.src_lang = "ar_AR"
encoded_ar = tokenizer(article_ar, return_tensors="pt")
generated_tokens = model.generate(**encoded_ar)
tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
# => "The Secretary-General of the United Nations says there is no military solution in Syria."
```
See the [model hub](https://huggingface.co/models?filter=mbart-50) to look for more fine-tuned versions.
## Languages covered
Arabic (ar_AR), Czech (cs_CZ), German (de_DE), English (en_XX), Spanish (es_XX), Estonian (et_EE), Finnish (fi_FI), French (fr_XX), Gujarati (gu_IN), Hindi (hi_IN), Italian (it_IT), Japanese (ja_XX), Kazakh (kk_KZ), Korean (ko_KR), Lithuanian (lt_LT), Latvian (lv_LV), Burmese (my_MM), Nepali (ne_NP), Dutch (nl_XX), Romanian (ro_RO), Russian (ru_RU), Sinhala (si_LK), Turkish (tr_TR), Vietnamese (vi_VN), Chinese (zh_CN), Afrikaans (af_ZA), Azerbaijani (az_AZ), Bengali (bn_IN), Persian (fa_IR), Hebrew (he_IL), Croatian (hr_HR), Indonesian (id_ID), Georgian (ka_GE), Khmer (km_KH), Macedonian (mk_MK), Malayalam (ml_IN), Mongolian (mn_MN), Marathi (mr_IN), Polish (pl_PL), Pashto (ps_AF), Portuguese (pt_XX), Swedish (sv_SE), Swahili (sw_KE), Tamil (ta_IN), Telugu (te_IN), Thai (th_TH), Tagalog (tl_XX), Ukrainian (uk_UA), Urdu (ur_PK), Xhosa (xh_ZA), Galician (gl_ES), Slovene (sl_SI)
## BibTeX entry and citation info
```
@article{tang2020multilingual,
title={Multilingual Translation with Extensible Multilingual Pretraining and Finetuning},
author={Yuqing Tang and Chau Tran and Xian Li and Peng-Jen Chen and Naman Goyal and Vishrav Chaudhary and Jiatao Gu and Angela Fan},
year={2020},
eprint={2008.00401},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
|
{"language": ["multilingual", "ar", "cs", "de", "en", "es", "et", "fi", "fr", "gu", "hi", "it", "ja", "kk", "ko", "lt", "lv", "my", "ne", "nl", "ro", "ru", "si", "tr", "vi", "zh", "af", "az", "bn", "fa", "he", "hr", "id", "ka", "km", "mk", "ml", "mn", "mr", "pl", "ps", "pt", "sv", "sw", "ta", "te", "th", "tl", "uk", "ur", "xh", "gl", "sl"], "tags": ["mbart-50"]}
|
facebook/mbart-large-50-many-to-one-mmt
| null |
[
"transformers",
"pytorch",
"tf",
"jax",
"mbart",
"text2text-generation",
"mbart-50",
"multilingual",
"ar",
"cs",
"de",
"en",
"es",
"et",
"fi",
"fr",
"gu",
"hi",
"it",
"ja",
"kk",
"ko",
"lt",
"lv",
"my",
"ne",
"nl",
"ro",
"ru",
"si",
"tr",
"vi",
"zh",
"af",
"az",
"bn",
"fa",
"he",
"hr",
"id",
"ka",
"km",
"mk",
"ml",
"mn",
"mr",
"pl",
"ps",
"pt",
"sv",
"sw",
"ta",
"te",
"th",
"tl",
"uk",
"ur",
"xh",
"gl",
"sl",
"arxiv:2008.00401",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
# mBART-50 one to many multilingual machine translation
This model is a fine-tuned checkpoint of [mBART-large-50](https://huggingface.co/facebook/mbart-large-50). `mbart-large-50-one-to-many-mmt` is fine-tuned for multilingual machine translation. It was introduced in [Multilingual Translation with Extensible Multilingual Pretraining and Finetuning](https://arxiv.org/abs/2008.00401) paper.
The model can translate English to other 49 languages mentioned below.
To translate into a target language, the target language id is forced as the first generated token. To force the
target language id as the first generated token, pass the `forced_bos_token_id` parameter to the `generate` method.
```python
from transformers import MBartForConditionalGeneration, MBart50TokenizerFast
article_en = "The head of the United Nations says there is no military solution in Syria"
model = MBartForConditionalGeneration.from_pretrained("facebook/mbart-large-50-one-to-many-mmt")
tokenizer = MBart50TokenizerFast.from_pretrained("facebook/mbart-large-50-one-to-many-mmt", src_lang="en_XX")
model_inputs = tokenizer(article_en, return_tensors="pt")
# translate from English to Hindi
generated_tokens = model.generate(
**model_inputs,
forced_bos_token_id=tokenizer.lang_code_to_id["hi_IN"]
)
tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
# => 'संयुक्त राष्ट्र के नेता कहते हैं कि सीरिया में कोई सैन्य समाधान नहीं है'
# translate from English to Chinese
generated_tokens = model.generate(
**model_inputs,
forced_bos_token_id=tokenizer.lang_code_to_id["zh_CN"]
)
tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
# => '联合国首脑说,叙利亚没有军事解决办法'
```
See the [model hub](https://huggingface.co/models?filter=mbart-50) to look for more fine-tuned versions.
## Languages covered
Arabic (ar_AR), Czech (cs_CZ), German (de_DE), English (en_XX), Spanish (es_XX), Estonian (et_EE), Finnish (fi_FI), French (fr_XX), Gujarati (gu_IN), Hindi (hi_IN), Italian (it_IT), Japanese (ja_XX), Kazakh (kk_KZ), Korean (ko_KR), Lithuanian (lt_LT), Latvian (lv_LV), Burmese (my_MM), Nepali (ne_NP), Dutch (nl_XX), Romanian (ro_RO), Russian (ru_RU), Sinhala (si_LK), Turkish (tr_TR), Vietnamese (vi_VN), Chinese (zh_CN), Afrikaans (af_ZA), Azerbaijani (az_AZ), Bengali (bn_IN), Persian (fa_IR), Hebrew (he_IL), Croatian (hr_HR), Indonesian (id_ID), Georgian (ka_GE), Khmer (km_KH), Macedonian (mk_MK), Malayalam (ml_IN), Mongolian (mn_MN), Marathi (mr_IN), Polish (pl_PL), Pashto (ps_AF), Portuguese (pt_XX), Swedish (sv_SE), Swahili (sw_KE), Tamil (ta_IN), Telugu (te_IN), Thai (th_TH), Tagalog (tl_XX), Ukrainian (uk_UA), Urdu (ur_PK), Xhosa (xh_ZA), Galician (gl_ES), Slovene (sl_SI)
## BibTeX entry and citation info
```
@article{tang2020multilingual,
title={Multilingual Translation with Extensible Multilingual Pretraining and Finetuning},
author={Yuqing Tang and Chau Tran and Xian Li and Peng-Jen Chen and Naman Goyal and Vishrav Chaudhary and Jiatao Gu and Angela Fan},
year={2020},
eprint={2008.00401},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
|
{"language": ["multilingual", "ar", "cs", "de", "en", "es", "et", "fi", "fr", "gu", "hi", "it", "ja", "kk", "ko", "lt", "lv", "my", "ne", "nl", "ro", "ru", "si", "tr", "vi", "zh", "af", "az", "bn", "fa", "he", "hr", "id", "ka", "km", "mk", "ml", "mn", "mr", "pl", "ps", "pt", "sv", "sw", "ta", "te", "th", "tl", "uk", "ur", "xh", "gl", "sl"], "tags": ["mbart-50"]}
|
facebook/mbart-large-50-one-to-many-mmt
| null |
[
"transformers",
"pytorch",
"tf",
"jax",
"mbart",
"text2text-generation",
"mbart-50",
"multilingual",
"ar",
"cs",
"de",
"en",
"es",
"et",
"fi",
"fr",
"gu",
"hi",
"it",
"ja",
"kk",
"ko",
"lt",
"lv",
"my",
"ne",
"nl",
"ro",
"ru",
"si",
"tr",
"vi",
"zh",
"af",
"az",
"bn",
"fa",
"he",
"hr",
"id",
"ka",
"km",
"mk",
"ml",
"mn",
"mr",
"pl",
"ps",
"pt",
"sv",
"sw",
"ta",
"te",
"th",
"tl",
"uk",
"ur",
"xh",
"gl",
"sl",
"arxiv:2008.00401",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text2text-generation
|
transformers
|
# mBART-50
mBART-50 is a multilingual Sequence-to-Sequence model pre-trained using the "Multilingual Denoising Pretraining" objective. It was introduced in [Multilingual Translation with Extensible Multilingual Pretraining and Finetuning](https://arxiv.org/abs/2008.00401) paper.
## Model description
mBART-50 is a multilingual Sequence-to-Sequence model. It was introduced to show that multilingual translation models can be created through multilingual fine-tuning.
Instead of fine-tuning on one direction, a pre-trained model is fine-tuned on many directions simultaneously. mBART-50 is created using the original mBART model and extended to add extra 25 languages to support multilingual machine translation models of 50 languages. The pre-training objective is explained below.
**Multilingual Denoising Pretraining**: The model incorporates N languages by concatenating data:
`D = {D1, ..., DN }` where each Di is a collection of monolingual documents in language `i`. The source documents are noised using two schemes,
first randomly shuffling the original sentences' order, and second a novel in-filling scheme,
where spans of text are replaced with a single mask token. The model is then tasked to reconstruct the original text.
35% of each instance's words are masked by random sampling a span length according to a Poisson distribution `(λ = 3.5)`.
The decoder input is the original text with one position offset. A language id symbol `LID` is used as the initial token to predict the sentence.
## Intended uses & limitations
`mbart-large-50` is pre-trained model and primarily aimed at being fine-tuned on translation tasks. It can also be fine-tuned on other multilingual sequence-to-sequence tasks.
See the [model hub](https://huggingface.co/models?filter=mbart-50) to look for fine-tuned versions.
## Training
As the model is multilingual, it expects the sequences in a different format. A special language id token is used as a prefix in both the source and target text. The text format is `[lang_code] X [eos]` with `X` being the source or target text respectively and `lang_code` is `source_lang_code` for source text and `tgt_lang_code` for target text. `bos` is never used. Once the examples are prepared in this format, it can be trained as any other sequence-to-sequence model.
```python
from transformers import MBartForConditionalGeneration, MBart50TokenizerFast
model = MBartForConditionalGeneration.from_pretrained("facebook/mbart-large-50")
tokenizer = MBart50TokenizerFast.from_pretrained("facebook/mbart-large-50", src_lang="en_XX", tgt_lang="ro_RO")
src_text = " UN Chief Says There Is No Military Solution in Syria"
tgt_text = "Şeful ONU declară că nu există o soluţie militară în Siria"
model_inputs = tokenizer(src_text, return_tensors="pt")
with tokenizer.as_target_tokenizer():
labels = tokenizer(tgt_text, return_tensors="pt").input_ids
model(**model_inputs, labels=labels) # forward pass
```
## Languages covered
Arabic (ar_AR), Czech (cs_CZ), German (de_DE), English (en_XX), Spanish (es_XX), Estonian (et_EE), Finnish (fi_FI), French (fr_XX), Gujarati (gu_IN), Hindi (hi_IN), Italian (it_IT), Japanese (ja_XX), Kazakh (kk_KZ), Korean (ko_KR), Lithuanian (lt_LT), Latvian (lv_LV), Burmese (my_MM), Nepali (ne_NP), Dutch (nl_XX), Romanian (ro_RO), Russian (ru_RU), Sinhala (si_LK), Turkish (tr_TR), Vietnamese (vi_VN), Chinese (zh_CN), Afrikaans (af_ZA), Azerbaijani (az_AZ), Bengali (bn_IN), Persian (fa_IR), Hebrew (he_IL), Croatian (hr_HR), Indonesian (id_ID), Georgian (ka_GE), Khmer (km_KH), Macedonian (mk_MK), Malayalam (ml_IN), Mongolian (mn_MN), Marathi (mr_IN), Polish (pl_PL), Pashto (ps_AF), Portuguese (pt_XX), Swedish (sv_SE), Swahili (sw_KE), Tamil (ta_IN), Telugu (te_IN), Thai (th_TH), Tagalog (tl_XX), Ukrainian (uk_UA), Urdu (ur_PK), Xhosa (xh_ZA), Galician (gl_ES), Slovene (sl_SI)
## BibTeX entry and citation info
```
@article{tang2020multilingual,
title={Multilingual Translation with Extensible Multilingual Pretraining and Finetuning},
author={Yuqing Tang and Chau Tran and Xian Li and Peng-Jen Chen and Naman Goyal and Vishrav Chaudhary and Jiatao Gu and Angela Fan},
year={2020},
eprint={2008.00401},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
|
{"language": ["multilingual", "ar", "cs", "de", "en", "es", "et", "fi", "fr", "gu", "hi", "it", "ja", "kk", "ko", "lt", "lv", "my", "ne", "nl", "ro", "ru", "si", "tr", "vi", "zh", "af", "az", "bn", "fa", "he", "hr", "id", "ka", "km", "mk", "ml", "mn", "mr", "pl", "ps", "pt", "sv", "sw", "ta", "te", "th", "tl", "uk", "ur", "xh", "gl", "sl"], "license": "mit", "tags": ["mbart-50"]}
|
facebook/mbart-large-50
| null |
[
"transformers",
"pytorch",
"tf",
"mbart",
"text2text-generation",
"mbart-50",
"multilingual",
"ar",
"cs",
"de",
"en",
"es",
"et",
"fi",
"fr",
"gu",
"hi",
"it",
"ja",
"kk",
"ko",
"lt",
"lv",
"my",
"ne",
"nl",
"ro",
"ru",
"si",
"tr",
"vi",
"zh",
"af",
"az",
"bn",
"fa",
"he",
"hr",
"id",
"ka",
"km",
"mk",
"ml",
"mn",
"mr",
"pl",
"ps",
"pt",
"sv",
"sw",
"ta",
"te",
"th",
"tl",
"uk",
"ur",
"xh",
"gl",
"sl",
"arxiv:2008.00401",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
translation
|
transformers
|
#### mbart-large-cc25
Pretrained (not finetuned) multilingual mbart model.
Original Languages
```
export langs=ar_AR,cs_CZ,de_DE,en_XX,es_XX,et_EE,fi_FI,fr_XX,gu_IN,hi_IN,it_IT,ja_XX,kk_KZ,ko_KR,lt_LT,lv_LV,my_MM,ne_NP,nl_XX,ro_RO,ru_RU,si_LK,tr_TR,vi_VN,zh_CN
```
Original Code: https://github.com/pytorch/fairseq/tree/master/examples/mbart
Docs: https://huggingface.co/transformers/master/model_doc/mbart.html
Finetuning Code: examples/seq2seq/finetune.py (as of Aug 20, 2020)
Can also be finetuned for summarization.
|
{"language": ["en", "ar", "cs", "de", "et", "fi", "fr", "gu", "hi", "it", "ja", "kk", "ko", "lt", "lv", "my", "ne", "nl", "ro", "ru", "si", "tr", "vi", "zh", "multilingual"], "tags": ["translation"]}
|
facebook/mbart-large-cc25
| null |
[
"transformers",
"pytorch",
"tf",
"mbart",
"text2text-generation",
"translation",
"en",
"ar",
"cs",
"de",
"et",
"fi",
"fr",
"gu",
"hi",
"it",
"ja",
"kk",
"ko",
"lt",
"lv",
"my",
"ne",
"nl",
"ro",
"ru",
"si",
"tr",
"vi",
"zh",
"multilingual",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
translation
|
transformers
|
### mbart-large-en-ro
This is mbart-large-cc25, finetuned on wmt_en_ro.
It scores BLEU 28.1 without post processing and BLEU 38 with postprocessing. Instructions in `romanian_postprocessing.md`
Original Code: https://github.com/pytorch/fairseq/tree/master/examples/mbart
Docs: https://huggingface.co/transformers/master/model_doc/mbart.html
Finetuning Code: examples/seq2seq/finetune.py (as of Aug 20, 2020)
|
{"language": ["en", "ro"], "license": "mit", "tags": ["translation"]}
|
facebook/mbart-large-en-ro
| null |
[
"transformers",
"pytorch",
"tf",
"safetensors",
"mbart",
"translation",
"en",
"ro",
"license:mit",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
fill-mask
|
transformers
|
# Muppet: Massive Multi-task Representations with Pre-Finetuning
# RoBERTa base model
This is a Massive Multi-task Pre-finetuned version of Roberta base. It was introduced in
[this paper](https://arxiv.org/abs/2101.11038). The model improves over roberta-base in a wide range of GLUE, QA tasks (details can be found in the paper). The gains in
smaller datasets are significant.
Note: This checkpoint does not contain the classificaiton/MRC heads used during pre-finetuning due to compatibility issues and hence you might get slightly lower performance than that reported in the paper on some datasets
## Model description
RoBERTa is a transformers model pretrained on a large corpus of English data in a self-supervised fashion. This means
it was pretrained on the raw texts only, with no humans labelling them in any way (which is why it can use lots of
publicly available data) with an automatic process to generate inputs and labels from those texts.
More precisely, it was pretrained with the Masked language modeling (MLM) objective. Taking a sentence, the model
randomly masks 15% of the words in the input then run the entire masked sentence through the model and has to predict
the masked words. This is different from traditional recurrent neural networks (RNNs) that usually see the words one
after the other, or from autoregressive models like GPT which internally mask the future tokens. It allows the model to
learn a bidirectional representation of the sentence.
This way, the model learns an inner representation of the English language that can then be used to extract features
useful for downstream tasks: if you have a dataset of labeled sentences for instance, you can train a standard
classifier using the features produced by the BERT model as inputs.
## Intended uses & limitations
You can use the raw model for masked language modeling, but it's mostly intended to be fine-tuned on a downstream task.
See the [model hub](https://huggingface.co/models?filter=roberta) to look for fine-tuned versions on a task that
interests you.
Note that this model is primarily aimed at being fine-tuned on tasks that use the whole sentence (potentially masked)
to make decisions, such as sequence classification, token classification or question answering. For tasks such as text
generation you should look at model like GPT2.
## Evaluation results
When fine-tuned on downstream tasks, this model achieves the following results:
Glue test results:
| Model | MNLI | QQP | QNLI | SST-2 | CoLA | STS-B | MRPC | RTE | SQuAD|
|:----:|:----:|:----:|:----:|:-----:|:----:|:-----:|:----:|:----:|:----:|
| Roberta-base | 87.6 | 91.9 | 92.8 | 94.8 | 63.6 | 91.2 | 90.2 | 78.7 | 82.6|
| MUPPET Roberta-base | 88.1 | 91.9 | 93.3 | 96.7 | - | - | 91.7 | 87.8 | 86.6|
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2101-11038,
author = {Armen Aghajanyan and
Anchit Gupta and
Akshat Shrivastava and
Xilun Chen and
Luke Zettlemoyer and
Sonal Gupta},
title = {Muppet: Massive Multi-task Representations with Pre-Finetuning},
journal = {CoRR},
volume = {abs/2101.11038},
year = {2021},
url = {https://arxiv.org/abs/2101.11038},
archivePrefix = {arXiv},
eprint = {2101.11038},
timestamp = {Sun, 31 Jan 2021 17:23:50 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2101-11038.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"language": "en", "license": "mit", "tags": ["exbert"], "datasets": ["bookcorpus", "wikipedia"]}
|
facebook/muppet-roberta-base
| null |
[
"transformers",
"pytorch",
"roberta",
"fill-mask",
"exbert",
"en",
"dataset:bookcorpus",
"dataset:wikipedia",
"arxiv:2101.11038",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
fill-mask
|
transformers
|
# Muppet: Massive Multi-task Representations with Pre-Finetuning
# RoBERTa large model
This is a Massive Multi-task Pre-finetuned version of Roberta large. It was introduced in
[this paper](https://arxiv.org/abs/2101.11038). The model improves over roberta-base in a wide range of GLUE, QA tasks (details can be found in the paper). The gains in
smaller datasets are significant.
Note: This checkpoint does not contain the classificaiton/MRC heads used during pre-finetuning due to compatibility issues and hence you might get slightly lower performance than that reported in the paper on some datasets
## Model description
RoBERTa is a transformers model pretrained on a large corpus of English data in a self-supervised fashion. This means
it was pretrained on the raw texts only, with no humans labelling them in any way (which is why it can use lots of
publicly available data) with an automatic process to generate inputs and labels from those texts.
More precisely, it was pretrained with the Masked language modeling (MLM) objective. Taking a sentence, the model
randomly masks 15% of the words in the input then run the entire masked sentence through the model and has to predict
the masked words. This is different from traditional recurrent neural networks (RNNs) that usually see the words one
after the other, or from autoregressive models like GPT which internally mask the future tokens. It allows the model to
learn a bidirectional representation of the sentence.
This way, the model learns an inner representation of the English language that can then be used to extract features
useful for downstream tasks: if you have a dataset of labeled sentences for instance, you can train a standard
classifier using the features produced by the BERT model as inputs.
## Intended uses & limitations
You can use the raw model for masked language modeling, but it's mostly intended to be fine-tuned on a downstream task.
See the [model hub](https://huggingface.co/models?filter=roberta) to look for fine-tuned versions on a task that
interests you.
Note that this model is primarily aimed at being fine-tuned on tasks that use the whole sentence (potentially masked)
to make decisions, such as sequence classification, token classification or question answering. For tasks such as text
generation you should look at model like GPT2.
## Evaluation results
When fine-tuned on downstream tasks, this model achieves the following results:
Glue test results:
| Model | MNLI | QQP | QNLI | SST-2 | CoLA | STS-B | MRPC | RTE | SQuAD|
|:----:|:----:|:----:|:----:|:-----:|:----:|:-----:|:----:|:----:|:----:|
| Roberta-large | 90.2 | 92.2 | 94.7 | 96.4 | 63.6 | 91.2 | 90.9 | 88.1 | 88.7|
| MUPPET Roberta-large | 90.8 | 92.2 | 94.9 | 97.4 | - | - | 91.4 | 92.8 | 89.4|
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2101-11038,
author = {Armen Aghajanyan and
Anchit Gupta and
Akshat Shrivastava and
Xilun Chen and
Luke Zettlemoyer and
Sonal Gupta},
title = {Muppet: Massive Multi-task Representations with Pre-Finetuning},
journal = {CoRR},
volume = {abs/2101.11038},
year = {2021},
url = {https://arxiv.org/abs/2101.11038},
archivePrefix = {arXiv},
eprint = {2101.11038},
timestamp = {Sun, 31 Jan 2021 17:23:50 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2101-11038.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"language": "en", "license": "mit", "tags": ["exbert"], "datasets": ["bookcorpus", "wikipedia"]}
|
facebook/muppet-roberta-large
| null |
[
"transformers",
"pytorch",
"roberta",
"fill-mask",
"exbert",
"en",
"dataset:bookcorpus",
"dataset:wikipedia",
"arxiv:2101.11038",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
## RAG
This is a non-finetuned version of the RAG-Sequence model of the the paper [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://arxiv.org/pdf/2005.11401.pdf)
by Patrick Lewis, Ethan Perez, Aleksandara Piktus et al.
Rag consits of a *question encoder*, *retriever* and a *generator*. The retriever should be a `RagRetriever` instance. The *question encoder* can be any model that can be loaded with `AutoModel` and the *generator* can be any model that can be loaded with `AutoModelForSeq2SeqLM`.
This model is a non-finetuned RAG-Sequence model and was created as follows:
```python
from transformers import RagTokenizer, RagRetriever, RagSequenceForGeneration, AutoTokenizer
model = RagSequenceForGeneration.from_pretrained_question_encoder_generator("facebook/dpr-question_encoder-single-nq-base", "facebook/bart-large")
question_encoder_tokenizer = AutoTokenizer.from_pretrained("facebook/dpr-question_encoder-single-nq-base")
generator_tokenizer = AutoTokenizer.from_pretrained("facebook/bart-large")
tokenizer = RagTokenizer(question_encoder_tokenizer, generator_tokenizer)
model.config.use_dummy_dataset = True
model.config.index_name = "exact"
retriever = RagRetriever(model.config, question_encoder_tokenizer, generator_tokenizer)
model.save_pretrained("./")
tokenizer.save_pretrained("./")
retriever.save_pretrained("./")
```
Note that the model is *uncased* so that all capital input letters are converted to lower-case.
## Usage:
*Note*: the model uses the *dummy* retriever as a default. Better results are obtained by using the full retriever,
by setting `config.index_name="legacy"` and `config.use_dummy_dataset=False`.
The model can be fine-tuned as follows:
```python
from transformers import RagTokenizer, RagRetriever, RagTokenForGeneration
tokenizer = RagTokenizer.from_pretrained("facebook/rag-sequence-base")
retriever = RagRetriever.from_pretrained("facebook/rag-sequence-base")
model = RagTokenForGeneration.from_pretrained("facebook/rag-sequence-base", retriever=retriever)
input_dict = tokenizer.prepare_seq2seq_batch("who holds the record in 100m freestyle", "michael phelps", return_tensors="pt")
outputs = model(input_dict["input_ids"], labels=input_dict["labels"])
loss = outputs.loss
# train on loss
```
|
{"license": "apache-2.0", "thumbnail": "https://huggingface.co/front/thumbnails/facebook.png"}
|
facebook/rag-sequence-base
| null |
[
"transformers",
"pytorch",
"rag",
"arxiv:2005.11401",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
## RAG
This is the RAG-Sequence Model of the the paper [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://arxiv.org/pdf/2005.11401.pdf)
by Patrick Lewis, Ethan Perez, Aleksandara Piktus et al.
The model is a *uncased* model, which means that capital letters are simply converted to lower-case letters.
The model consits of a *question_encoder*, *retriever* and a *generator*. The retriever extracts relevant passages from the *wiki_dpr* `train` datasets, which is linked above.
The question_encoder and retriever are based on `facebook/dpr-question_encoder-single-nq-base` and `facebook/bart-large`, which were jointly finetuned on
on the *wiki_dpr* QA dataset in an end-to-end fashion.
## Usage:
**Note**: In the usage example below only the *dummy* retriever of *wiki_dpr* is used because the complete *lecagy* index requires over 75 GB of RAM.
The model can generate answers to any factoid question as follows:
```python
from transformers import RagTokenizer, RagRetriever, RagSequenceForGeneration
tokenizer = RagTokenizer.from_pretrained("facebook/rag-sequence-nq")
retriever = RagRetriever.from_pretrained("facebook/rag-sequence-nq", index_name="exact", use_dummy_dataset=True)
model = RagSequenceForGeneration.from_pretrained("facebook/rag-sequence-nq", retriever=retriever)
input_dict = tokenizer.prepare_seq2seq_batch("how many countries are in europe", return_tensors="pt")
generated = model.generate(input_ids=input_dict["input_ids"])
print(tokenizer.batch_decode(generated, skip_special_tokens=True)[0])
# should give 54 => google says either 44 or 51
```
|
{"language": "en", "license": "apache-2.0", "datasets": ["wiki_dpr"], "thumbnail": "https://huggingface.co/front/thumbnails/facebook.png"}
|
facebook/rag-sequence-nq
| null |
[
"transformers",
"pytorch",
"tf",
"rag",
"en",
"dataset:wiki_dpr",
"arxiv:2005.11401",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
## RAG
This is a non-finetuned version of the RAG-Token model of the the paper [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://arxiv.org/pdf/2005.11401.pdf)
by Patrick Lewis, Ethan Perez, Aleksandara Piktus et al.
Rag consits of a *question encoder*, *retriever* and a *generator*. The retriever should be a `RagRetriever` instance. The *question encoder* can be any model that can be loaded with `AutoModel` and the *generator* can be any model that can be loaded with `AutoModelForSeq2SeqLM`.
This model is a non-finetuned RAG-Token model and was created as follows:
```python
from transformers import RagTokenizer, RagRetriever, RagTokenForGeneration, AutoTokenizer
model = RagTokenForGeneration.from_pretrained_question_encoder_generator("facebook/dpr-question_encoder-single-nq-base", "facebook/bart-large")
question_encoder_tokenizer = AutoTokenizer.from_pretrained("facebook/dpr-question_encoder-single-nq-base")
generator_tokenizer = AutoTokenizer.from_pretrained("facebook/bart-large")
tokenizer = RagTokenizer(question_encoder_tokenizer, generator_tokenizer)
model.config.use_dummy_dataset = True
model.config.index_name = "exact"
retriever = RagRetriever(model.config, question_encoder_tokenizer, generator_tokenizer)
model.save_pretrained("./")
tokenizer.save_pretrained("./")
retriever.save_pretrained("./")
```
Note that the model is *uncased* so that all capital input letters are converted to lower-case.
## Usage:
*Note*: the model uses the *dummy* retriever as a default. Better results are obtained by using the full retriever,
by setting `config.index_name="legacy"` and `config.use_dummy_dataset=False`.
The model can be fine-tuned as follows:
```python
from transformers import RagTokenizer, RagRetriever, RagTokenForGeneration
tokenizer = RagTokenizer.from_pretrained("facebook/rag-token-base")
retriever = RagRetriever.from_pretrained("facebook/rag-token-base")
model = RagTokenForGeneration.from_pretrained("facebook/rag-token-base", retriever=retriever)
input_dict = tokenizer.prepare_seq2seq_batch("who holds the record in 100m freestyle", "michael phelps", return_tensors="pt")
outputs = model(input_dict["input_ids"], labels=input_dict["labels"])
loss = outputs.loss
# train on loss
```
|
{"language": "en", "license": "apache-2.0", "datasets": ["wiki_dpr"], "thumbnail": "https://huggingface.co/front/thumbnails/facebook.png"}
|
facebook/rag-token-base
| null |
[
"transformers",
"pytorch",
"rag",
"en",
"dataset:wiki_dpr",
"arxiv:2005.11401",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
null |
transformers
|
## RAG
This is the RAG-Token Model of the the paper [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://arxiv.org/pdf/2005.11401.pdf)
by Patrick Lewis, Ethan Perez, Aleksandara Piktus et al.
The model is a *uncased* model, which means that capital letters are simply converted to lower-case letters.
The model consists of a *question_encoder*, *retriever* and a *generator*. The retriever extracts relevant passages from the *wiki_dpr* `train` datasets, which is linked above.
The question_encoder and retriever are based on `facebook/dpr-question_encoder-single-nq-base` and `facebook/bart-large`, which were jointly finetuned on
on the *wiki_dpr* QA dataset in an end-to-end fashion.
## Usage:
**Note**: In the usage example below only the *dummy* retriever of *wiki_dpr* is used because the complete *lecagy* index requires over 75 GB of RAM.
The model can generate answers to any factoid question as follows:
```python
from transformers import RagTokenizer, RagRetriever, RagTokenForGeneration
tokenizer = RagTokenizer.from_pretrained("facebook/rag-token-nq")
retriever = RagRetriever.from_pretrained("facebook/rag-token-nq", index_name="exact", use_dummy_dataset=True)
model = RagTokenForGeneration.from_pretrained("facebook/rag-token-nq", retriever=retriever)
input_dict = tokenizer.prepare_seq2seq_batch("who holds the record in 100m freestyle", return_tensors="pt")
generated = model.generate(input_ids=input_dict["input_ids"])
print(tokenizer.batch_decode(generated, skip_special_tokens=True)[0])
# should give michael phelps => sounds reasonable
```
|
{"language": "en", "license": "apache-2.0", "datasets": ["wiki_dpr"], "thumbnail": "https://huggingface.co/front/thumbnails/facebook.png"}
|
facebook/rag-token-nq
| null |
[
"transformers",
"pytorch",
"tf",
"rag",
"en",
"dataset:wiki_dpr",
"arxiv:2005.11401",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-LARGE-LIBRISPEECH-ASR
`s2t-large-librispeech-asr` is a Speech to Text Transformer (S2T) model trained for automatic speech recognition (ASR).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is an end-to-end sequence-to-sequence transformer model. It is trained with standard
autoregressive cross-entropy loss and generates the transcripts autoregressively.
## Intended uses & limitations
This model can be used for end-to-end speech recognition (ASR).
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-large-librispeech-asr")
processor = Speech2Textprocessor.from_pretrained("facebook/s2t-large-librispeech-asr")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
input_features = processor(
ds["speech"][0],
sampling_rate=16_000,
return_tensors="pt"
).input_features # Batch size 1
generated_ids = model.generate(input_ids=input_features)
transcription = processor.batch_decode(generated_ids)
```
#### Evaluation on LibriSpeech Test
The following script shows how to evaluate this model on the [LibriSpeech](https://huggingface.co/datasets/librispeech_asr)
*"clean"* and *"other"* test dataset.
```python
from datasets import load_dataset, load_metric
from transformers import Speech2TextForConditionalGeneration, Speech2TextProcessor
import soundfile as sf
librispeech_eval = load_dataset("librispeech_asr", "clean", split="test") # change to "other" for other test dataset
wer = load_metric("wer")
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-large-librispeech-asr").to("cuda")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-large-librispeech-asr", do_upper_case=True)
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
librispeech_eval = librispeech_eval.map(map_to_array)
def map_to_pred(batch):
features = processor(batch["speech"], sampling_rate=16000, padding=True, return_tensors="pt")
input_features = features.input_features.to("cuda")
attention_mask = features.attention_mask.to("cuda")
gen_tokens = model.generate(input_ids=input_features, attention_mask=attention_mask)
batch["transcription"] = processor.batch_decode(gen_tokens, skip_special_tokens=True)
return batch
result = librispeech_eval.map(map_to_pred, batched=True, batch_size=8, remove_columns=["speech"])
print("WER:", wer(predictions=result["transcription"], references=result["text"]))
```
*Result (WER)*:
| "clean" | "other" |
|:-------:|:-------:|
| 3.3 | 7.5 |
## Training data
The S2T-LARGE-LIBRISPEECH-ASR is trained on [LibriSpeech ASR Corpus](https://www.openslr.org/12), a dataset consisting of
approximately 1000 hours of 16kHz read English speech.
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 10,000.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively.
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": "en", "license": "mit", "tags": ["audio", "automatic-speech-recognition", "hf-asr-leaderboard"], "datasets": ["librispeech_asr"], "model-index": [{"name": "hubert-large-ls960-ft", "results": [{"task": {"type": "automatic-speech-recognition", "name": "Automatic Speech Recognition"}, "dataset": {"name": "LibriSpeech (clean)", "type": "librispeech_asr", "config": "clean", "split": "test", "args": {"language": "en"}}, "metrics": [{"type": "wer", "value": 3.3, "name": "Test WER"}]}, {"task": {"type": "automatic-speech-recognition", "name": "Automatic Speech Recognition"}, "dataset": {"name": "LibriSpeech (other)", "type": "librispeech_asr", "config": "other", "split": "test", "args": {"language": "en"}}, "metrics": [{"type": "wer", "value": 7.5, "name": "Test WER"}]}]}]}
|
facebook/s2t-large-librispeech-asr
| null |
[
"transformers",
"pytorch",
"tf",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"hf-asr-leaderboard",
"en",
"dataset:librispeech_asr",
"arxiv:2010.05171",
"arxiv:1904.08779",
"license:mit",
"model-index",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-MEDIUM-LIBRISPEECH-ASR
`s2t-medium-librispeech-asr` is a Speech to Text Transformer (S2T) model trained for automatic speech recognition (ASR).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is an end-to-end sequence-to-sequence transformer model. It is trained with standard
autoregressive cross-entropy loss and generates the transcripts autoregressively.
## Intended uses & limitations
This model can be used for end-to-end speech recognition (ASR).
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-medium-librispeech-asr")
processor = Speech2Textprocessor.from_pretrained("facebook/s2t-medium-librispeech-asr")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
input_features = processor(
ds["speech"][0],
sampling_rate=16_000,
return_tensors="pt"
).input_features # Batch size 1
generated_ids = model.generate(input_features=input_features)
transcription = processor.batch_decode(generated_ids)
```
#### Evaluation on LibriSpeech Test
The following script shows how to evaluate this model on the [LibriSpeech](https://huggingface.co/datasets/librispeech_asr)
*"clean"* and *"other"* test dataset.
```python
from datasets import load_dataset
from evaluate import load
from transformers import Speech2TextForConditionalGeneration, Speech2TextProcessor
librispeech_eval = load_dataset("librispeech_asr", "clean", split="test") # change to "other" for other test dataset
wer = load("wer")
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-medium-librispeech-asr").to("cuda")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-medium-librispeech-asr", do_upper_case=True)
def map_to_pred(batch):
features = processor(batch["audio"]["array"], sampling_rate=16000, padding=True, return_tensors="pt")
input_features = features.input_features.to("cuda")
attention_mask = features.attention_mask.to("cuda")
gen_tokens = model.generate(input_features=input_features, attention_mask=attention_mask)
batch["transcription"] = processor.batch_decode(gen_tokens, skip_special_tokens=True)[0]
return batch
result = librispeech_eval.map(map_to_pred, remove_columns=["audio"])
print("WER:", wer.compute(predictions=result["transcription"], references=result["text"]))
```
*Result (WER)*:
| "clean" | "other" |
|:-------:|:-------:|
| 3.5 | 7.8 |
## Training data
The S2T-MEDIUM-LIBRISPEECH-ASR is trained on [LibriSpeech ASR Corpus](https://www.openslr.org/12), a dataset consisting of
approximately 1000 hours of 16kHz read English speech.
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 10,000.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively.
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": "en", "license": "mit", "tags": ["audio", "automatic-speech-recognition"], "datasets": ["librispeech_asr"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}]}
|
facebook/s2t-medium-librispeech-asr
| null |
[
"transformers",
"pytorch",
"tf",
"safetensors",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"en",
"dataset:librispeech_asr",
"arxiv:2010.05171",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-MEDIUM-MUSTC-MULTILINGUAL-ST
`s2t-medium-mustc-multilingual-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Multilingual Speech Translation (ST).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
transcripts/translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end English speech to French text translation.
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
For multilingual speech translation models, `eos_token_id` is used as the `decoder_start_token_id` and
the target language id is forced as the first generated token. To force the target language id as the first
generated token, pass the `forced_bos_token_id` parameter to the `generate()` method. The following
example shows how to transate English speech to French and German text using the `facebook/s2t-medium-mustc-multilingual-st`
checkpoint.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-medium-mustc-multilingual-st")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-medium-mustc-multilingual-st")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
ds = ds.map(map_to_array)
inputs = processor(ds["speech"][0], sampling_rate=16_000, return_tensors="pt")
# translate English Speech To French Text
generated_ids = model.generate(
input_ids=inputs["input_features"],
attention_mask=inputs["attention_mask"],
forced_bos_token_id=processor.tokenizer.lang_code_to_id["fr"]
)
translation_fr = processor.batch_decode(generated_ids)
# translate English Speech To German Text
generated_ids = model.generate(
input_ids=inputs["input_features"],
attention_mask=inputs["attention_mask"],
forced_bos_token_id=processor.tokenizer.lang_code_to_id["de"]
)
translation_de = processor.batch_decode(generated_ids, skip_special_tokens=True)
```
## Training data
The s2t-medium-mustc-multilingual-st is trained on [MuST-C](https://ict.fbk.eu/must-c/).
MuST-C is a multilingual speech translation corpus whose size and quality facilitates the training of end-to-end systems
for speech translation from English into several languages. For each target language, MuST-C comprises several hundred
hours of audio recordings from English TED Talks, which are automatically aligned at the sentence level with their manual
transcriptions and translations.
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 10,000.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
model training and for better performance the encoder is pre-trained for multilingual ASR. For multilingual models, target language ID token
is used as target BOS.
## Evaluation results
MuST-C test results (BLEU score):
| En-De | En-Nl | En-Es | En-Fr | En-It | En-Pt | En-Ro | En-Ru |
|:-----:|:-----:|:-----:|:-----:|:-----:|:-----:|:-----:|:-----:|
| 24.5 | 28.6 | 28.2 | 34.9 | 24.6 | 31.1 | 23.8 | 16.0 |
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": ["en", "de", "nl", "es", "fr", "it", "pt", "ro", "ru"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition"], "datasets": ["mustc"], "pipeline_tag": "automatic-speech-recognition"}
|
facebook/s2t-medium-mustc-multilingual-st
| null |
[
"transformers",
"pytorch",
"tf",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"speech-translation",
"en",
"de",
"nl",
"es",
"fr",
"it",
"pt",
"ro",
"ru",
"dataset:mustc",
"arxiv:2010.05171",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-SMALL-COVOST2-CA-EN-ST
`s2t-small-covost2-ca-en-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Speech Translation (ST).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
transcripts/translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end Catalan speech to English text translation.
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-covost2-ca-en-st")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-covost2-ca-en-st")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
inputs = processor(
ds["speech"][0],
sampling_rate=48_000,
return_tensors="pt"
)
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
translation = processor.batch_decode(generated_ids, skip_special_tokens=True)
```
## Training data
The s2t-small-covost2-ca-en-st is trained on Catalan-English subset of [CoVoST2](https://github.com/facebookresearch/covost).
CoVoST is a large-scale multilingual ST corpus based on [Common Voice](https://arxiv.org/abs/1912.06670), created to to foster
ST research with the largest ever open dataset
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using character based SentencePiece vocab.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
model training and for better performance the encoder is pre-trained for English ASR.
## Evaluation results
CoVOST2 test results for ca-en (BLEU score): 17.85
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": ["ca", "en"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition"], "datasets": ["covost2"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}]}
|
facebook/s2t-small-covost2-ca-en-st
| null |
[
"transformers",
"pytorch",
"tf",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"speech-translation",
"ca",
"en",
"dataset:covost2",
"arxiv:2010.05171",
"arxiv:1912.06670",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-SMALL-COVOST2-DE-EN-ST
`s2t-small-covost2-de-en-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Speech Translation (ST).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
transcripts/translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end German speech to English text translation.
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-covost2-de-en-st")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-covost2-de-en-st")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
inputs = processor(
ds["speech"][0],
sampling_rate=48_000,
return_tensors="pt"
)
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
translation = processor.batch_decode(generated_ids, skip_special_tokens=True)
```
## Training data
The s2t-small-covost2-de-en-st is trained on German-English subset of [CoVoST2](https://github.com/facebookresearch/covost).
CoVoST is a large-scale multilingual ST corpus based on [Common Voice](https://arxiv.org/abs/1912.06670), created to to foster
ST research with the largest ever open dataset
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using character based SentencePiece vocab.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
model training and for better performance the encoder is pre-trained for English ASR.
## Evaluation results
CoVOST2 test results for de-en (BLEU score): 17.58
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": ["de", "en"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition"], "datasets": ["covost2"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}]}
|
facebook/s2t-small-covost2-de-en-st
| null |
[
"transformers",
"pytorch",
"tf",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"speech-translation",
"de",
"en",
"dataset:covost2",
"arxiv:2010.05171",
"arxiv:1912.06670",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-SMALL-COVOST2-EN-CA-ST
`s2t-small-covost2-en-ca-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Speech Translation (ST).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
transcripts/translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end English speech to Catlan text translation.
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-covost2-en-ca-st")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-covost2-en-ca-st")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
inputs = processor(
ds["speech"][0],
sampling_rate=48_000,
return_tensors="pt"
)
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
translation = processor.batch_decode(generated_ids, skip_special_tokens=True)
```
## Training data
The s2t-small-covost2-en-ca-st is trained on English-Catlan subset of [CoVoST2](https://github.com/facebookresearch/covost).
CoVoST is a large-scale multilingual ST corpus based on [Common Voice](https://arxiv.org/abs/1912.06670), created to to foster
ST research with the largest ever open dataset
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using character based SentencePiece vocab.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
model training and for better performance the encoder is pre-trained for English ASR.
## Evaluation results
CoVOST2 test results for en-ca (BLEU score): 21.68
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": ["en", "ca"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition"], "datasets": ["covost2"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}]}
|
facebook/s2t-small-covost2-en-ca-st
| null |
[
"transformers",
"pytorch",
"tf",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"speech-translation",
"en",
"ca",
"dataset:covost2",
"arxiv:2010.05171",
"arxiv:1912.06670",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-SMALL-COVOST2-EN-DE-ST
`s2t-small-covost2-en-de-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Speech Translation (ST).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
transcripts/translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end English speech to German text translation.
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-covost2-en-de-st")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-covost2-en-de-st")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
inputs = processor(
ds["speech"][0],
sampling_rate=48_000,
return_tensors="pt"
)
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
translation = processor.batch_decode(generated_ids, skip_special_tokens=True)
```
## Training data
The s2t-small-covost2-en-de-st is trained on English-German subset of [CoVoST2](https://github.com/facebookresearch/covost).
CoVoST is a large-scale multilingual ST corpus based on [Common Voice](https://arxiv.org/abs/1912.06670), created to to foster
ST research with the largest ever open dataset
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using character based SentencePiece vocab.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
model training and for better performance the encoder is pre-trained for English ASR.
## Evaluation results
CoVOST2 test results for en-de (BLEU score): 16.29
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": ["en", "de"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition"], "datasets": ["covost2"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}]}
|
facebook/s2t-small-covost2-en-de-st
| null |
[
"transformers",
"pytorch",
"tf",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"speech-translation",
"en",
"de",
"dataset:covost2",
"arxiv:2010.05171",
"arxiv:1912.06670",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-SMALL-COVOST2-EN-ET-ST
`s2t-small-covost2-en-et-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Speech Translation (ST).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
transcripts/translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end English speech to Estonian text translation.
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-covost2-en-et-st")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-covost2-en-et-st")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
inputs = processor(
ds["speech"][0],
sampling_rate=48_000,
return_tensors="pt"
)
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
translation = processor.batch_decode(generated_ids, skip_special_tokens=True)
```
## Training data
The s2t-small-covost2-en-et-st is trained on English-Estonian subset of [CoVoST2](https://github.com/facebookresearch/covost).
CoVoST is a large-scale multilingual ST corpus based on [Common Voice](https://arxiv.org/abs/1912.06670), created to to foster
ST research with the largest ever open dataset
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using character based SentencePiece vocab.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
model training and for better performance the encoder is pre-trained for English ASR.
## Evaluation results
CoVOST2 test results for en-et (BLEU score): 13.01
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": ["en", "et"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition"], "datasets": ["covost2"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}]}
|
facebook/s2t-small-covost2-en-et-st
| null |
[
"transformers",
"pytorch",
"tf",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"speech-translation",
"en",
"et",
"dataset:covost2",
"arxiv:2010.05171",
"arxiv:1912.06670",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-SMALL-COVOST2-EN-FA-ST
`s2t-small-covost2-en-fa-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Speech Translation (ST).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
transcripts/translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end English speech to Farsi text translation.
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-covost2-en-fa-st")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-covost2-en-fa-st")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
inputs = processor(
ds["speech"][0],
sampling_rate=48_000,
return_tensors="pt"
)
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
translation = processor.batch_decode(generated_ids, skip_special_tokens=True)
```
## Training data
The s2t-small-covost2-en-fa-st is trained on English-Farsi subset of [CoVoST2](https://github.com/facebookresearch/covost).
CoVoST is a large-scale multilingual ST corpus based on [Common Voice](https://arxiv.org/abs/1912.06670), created to to foster
ST research with the largest ever open dataset
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using character based SentencePiece vocab.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
model training and for better performance the encoder is pre-trained for English ASR.
## Evaluation results
CoVOST2 test results for en-fa (BLEU score): 11.43
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": ["en", "fa"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition"], "datasets": ["covost2"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}]}
|
facebook/s2t-small-covost2-en-fa-st
| null |
[
"transformers",
"pytorch",
"tf",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"speech-translation",
"en",
"fa",
"dataset:covost2",
"arxiv:2010.05171",
"arxiv:1912.06670",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-SMALL-COVOST2-ES-EN-ST
`s2t-small-covost2-es-en-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Speech Translation (ST).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
transcripts/translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end Spanish speech to English text translation.
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-covost2-es-en-st")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-covost2-es-en-st")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
inputs = processor(
ds["speech"][0],
sampling_rate=48_000,
return_tensors="pt"
)
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
translation = processor.batch_decode(generated_ids, skip_special_tokens=True)
```
## Training data
The s2t-small-covost2-es-en-st is trained on Spanish-English subset of [CoVoST2](https://github.com/facebookresearch/covost).
CoVoST is a large-scale multilingual ST corpus based on [Common Voice](https://arxiv.org/abs/1912.06670), created to to foster
ST research with the largest ever open dataset
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using character based SentencePiece vocab.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
model training and for better performance the encoder is pre-trained for English ASR.
## Evaluation results
CoVOST2 test results for es-en (BLEU score): 22.31
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": ["es", "en"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition"], "datasets": ["covost2"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}]}
|
facebook/s2t-small-covost2-es-en-st
| null |
[
"transformers",
"pytorch",
"tf",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"speech-translation",
"es",
"en",
"dataset:covost2",
"arxiv:2010.05171",
"arxiv:1912.06670",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-SMALL-COVOST2-FR-EN-ST
`s2t-small-covost2-fr-en-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Speech Translation (ST).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
transcripts/translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end French speech to English text translation.
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-covost2-fr-en-st")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-covost2-fr-en-st")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
inputs = processor(
ds["speech"][0],
sampling_rate=48_000,
return_tensors="pt"
)
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
translation = processor.batch_decode(generated_ids, skip_special_tokens=True)
```
## Training data
The s2t-small-covost2-fr-en-st is trained on French-English subset of [CoVoST2](https://github.com/facebookresearch/covost).
CoVoST is a large-scale multilingual ST corpus based on [Common Voice](https://arxiv.org/abs/1912.06670), created to to foster
ST research with the largest ever open dataset
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using character based SentencePiece vocab.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
model training and for better performance the encoder is pre-trained for English ASR.
## Evaluation results
CoVOST2 test results for fr-en (BLEU score): 26.25
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": ["fr", "en"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition"], "datasets": ["covost2"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}]}
|
facebook/s2t-small-covost2-fr-en-st
| null |
[
"transformers",
"pytorch",
"tf",
"safetensors",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"speech-translation",
"fr",
"en",
"dataset:covost2",
"arxiv:2010.05171",
"arxiv:1912.06670",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-SMALL-LIBRISPEECH-ASR
`s2t-small-librispeech-asr` is a Speech to Text Transformer (S2T) model trained for automatic speech recognition (ASR).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is an end-to-end sequence-to-sequence transformer model. It is trained with standard
autoregressive cross-entropy loss and generates the transcripts autoregressively.
## Intended uses & limitations
This model can be used for end-to-end speech recognition (ASR).
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
*Note: The feature extractor depends on [torchaudio](https://github.com/pytorch/audio) and the tokenizer depends on [sentencepiece](https://github.com/google/sentencepiece)
so be sure to install those packages before running the examples.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-librispeech-asr")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-librispeech-asr")
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
input_features = processor(
ds[0]["audio"]["array"],
sampling_rate=16_000,
return_tensors="pt"
).input_features # Batch size 1
generated_ids = model.generate(input_features=input_features)
transcription = processor.batch_decode(generated_ids)
```
#### Evaluation on LibriSpeech Test
The following script shows how to evaluate this model on the [LibriSpeech](https://huggingface.co/datasets/librispeech_asr)
*"clean"* and *"other"* test dataset.
```python
from datasets import load_dataset
from evaluate import load
from transformers import Speech2TextForConditionalGeneration, Speech2TextProcessor
librispeech_eval = load_dataset("librispeech_asr", "clean", split="test") # change to "other" for other test dataset
wer = load("wer")
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-librispeech-asr").to("cuda")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-librispeech-asr", do_upper_case=True)
def map_to_pred(batch):
features = processor(batch["audio"]["array"], sampling_rate=16000, padding=True, return_tensors="pt")
input_features = features.input_features.to("cuda")
attention_mask = features.attention_mask.to("cuda")
gen_tokens = model.generate(input_features=input_features, attention_mask=attention_mask)
batch["transcription"] = processor.batch_decode(gen_tokens, skip_special_tokens=True)[0]
return batch
result = librispeech_eval.map(map_to_pred, remove_columns=["audio"])
print("WER:", wer.compute(predictions=result["transcription"], references=result["text"]))
```
*Result (WER)*:
| "clean" | "other" |
|:-------:|:-------:|
| 4.3 | 9.0 |
## Training data
The S2T-SMALL-LIBRISPEECH-ASR is trained on [LibriSpeech ASR Corpus](https://www.openslr.org/12), a dataset consisting of
approximately 1000 hours of 16kHz read English speech.
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 10,000.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively.
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": "en", "license": "mit", "tags": ["speech", "audio", "automatic-speech-recognition", "hf-asr-leaderboard"], "datasets": ["librispeech_asr"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}], "model-index": [{"name": "s2t-small-librispeech-asr", "results": [{"task": {"type": "automatic-speech-recognition", "name": "Automatic Speech Recognition"}, "dataset": {"name": "LibriSpeech (clean)", "type": "librispeech_asr", "config": "clean", "split": "test", "args": {"language": "en"}}, "metrics": [{"type": "wer", "value": 4.3, "name": "Test WER"}]}, {"task": {"type": "automatic-speech-recognition", "name": "Automatic Speech Recognition"}, "dataset": {"name": "LibriSpeech (other)", "type": "librispeech_asr", "config": "other", "split": "test", "args": {"language": "en"}}, "metrics": [{"type": "wer", "value": 9.0, "name": "Test WER"}]}]}]}
|
facebook/s2t-small-librispeech-asr
| null |
[
"transformers",
"pytorch",
"tf",
"safetensors",
"speech_to_text",
"automatic-speech-recognition",
"speech",
"audio",
"hf-asr-leaderboard",
"en",
"dataset:librispeech_asr",
"arxiv:2010.05171",
"arxiv:1904.08779",
"license:mit",
"model-index",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-SMALL-MUSTC-EN-DE-ST
`s2t-small-mustc-en-de-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Speech Translation (ST).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
transcripts/translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end English speech to German text translation.
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-mustc-en-de-st")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-mustc-en-de-st")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
inputs = processor(
ds["speech"][0],
sampling_rate=16_000,
return_tensors="pt"
)
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
translation = processor.batch_decode(generated_ids, skip_special_tokens=True)
```
## Training data
The s2t-small-mustc-en-de-st is trained on English-German subset of [MuST-C](https://ict.fbk.eu/must-c/).
MuST-C is a multilingual speech translation corpus whose size and quality facilitates the training of end-to-end systems
for speech translation from English into several languages. For each target language, MuST-C comprises several hundred
hours of audio recordings from English TED Talks, which are automatically aligned at the sentence level with their manual
transcriptions and translations.
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 8,000.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
model training and for better performance the encoder is pre-trained for English ASR.
## Evaluation results
MuST-C test results for en-de (BLEU score): 22.7
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": ["en", "de"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition"], "datasets": ["mustc"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}]}
|
facebook/s2t-small-mustc-en-de-st
| null |
[
"transformers",
"pytorch",
"tf",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"speech-translation",
"en",
"de",
"dataset:mustc",
"arxiv:2010.05171",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-SMALL-MUSTC-EN-ES-ST
`s2t-small-mustc-en-es-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Speech Translation (ST).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
transcripts/translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end English speech to Spanish text translation.
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-mustc-en-es-st")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-mustc-en-es-st")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
inputs = processor(
ds["speech"][0],
sampling_rate=16_000,
return_tensors="pt"
)
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
translation = processor.batch_decode(generated_ids, skip_special_tokens=True)
```
## Training data
The s2t-small-mustc-en-es-st is trained on English-Spanish subset of [MuST-C](https://ict.fbk.eu/must-c/).
MuST-C is a multilingual speech translation corpus whose size and quality facilitates the training of end-to-end systems
for speech translation from English into several languages. For each target language, MuST-C comprises several hundred
hours of audio recordings from English TED Talks, which are automatically aligned at the sentence level with their manual
transcriptions and translations.
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 8,000.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
model training and for better performance the encoder is pre-trained for English ASR.
## Evaluation results
MuST-C test results for en-es (BLEU score): 27.2
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": ["en", "es"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition"], "datasets": ["mustc"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}]}
|
facebook/s2t-small-mustc-en-es-st
| null |
[
"transformers",
"pytorch",
"tf",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"speech-translation",
"en",
"es",
"dataset:mustc",
"arxiv:2010.05171",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-SMALL-MUSTC-EN-FR-ST
`s2t-small-mustc-en-fr-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Speech Translation (ST).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
transcripts/translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end English speech to French text translation.
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-mustc-en-fr-st")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-mustc-en-fr-st")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
inputs = processor(
ds["speech"][0],
sampling_rate=16_000,
return_tensors="pt"
)
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
translation = processor.batch_decode(generated_ids, skip_special_tokens=True)
```
## Training data
The s2t-small-mustc-en-fr-st is trained on English-French subset of [MuST-C](https://ict.fbk.eu/must-c/).
MuST-C is a multilingual speech translation corpus whose size and quality facilitates the training of end-to-end systems
for speech translation from English into several languages. For each target language, MuST-C comprises several hundred
hours of audio recordings from English TED Talks, which are automatically aligned at the sentence level with their manual
transcriptions and translations.
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 8,000.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
model training and for better performance the encoder is pre-trained for English ASR.
## Evaluation results
MuST-C test results for en-fr (BLEU score): 32.9
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": ["en", "fr"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition"], "datasets": ["mustc"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}]}
|
facebook/s2t-small-mustc-en-fr-st
| null |
[
"transformers",
"pytorch",
"tf",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"speech-translation",
"en",
"fr",
"dataset:mustc",
"arxiv:2010.05171",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-SMALL-MUSTC-EN-IT-ST
`s2t-small-mustc-en-it-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Speech Translation (ST).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
transcripts/translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end English speech to Italian text translation.
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-mustc-en-it-st")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-mustc-en-it-st")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
inputs = processor(
ds["speech"][0],
sampling_rate=16_000,
return_tensors="pt"
)
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
translation = processor.batch_decode(generated_ids, skip_special_tokens=True)
```
## Training data
The s2t-small-mustc-en-it-st is trained on English-Italian subset of [MuST-C](https://ict.fbk.eu/must-c/).
MuST-C is a multilingual speech translation corpus whose size and quality facilitates the training of end-to-end systems
for speech translation from English into several languages. For each target language, MuST-C comprises several hundred
hours of audio recordings from English TED Talks, which are automatically aligned at the sentence level with their manual
transcriptions and translations.
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 8,000.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
model training and for better performance the encoder is pre-trained for English ASR.
## Evaluation results
MuST-C test results for en-it (BLEU score): 22.7
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": ["en", "it"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition"], "datasets": ["mustc"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}]}
|
facebook/s2t-small-mustc-en-it-st
| null |
[
"transformers",
"pytorch",
"tf",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"speech-translation",
"en",
"it",
"dataset:mustc",
"arxiv:2010.05171",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-SMALL-MUSTC-EN-NL-ST
`s2t-small-mustc-en-nl-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Speech Translation (ST).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
transcripts/translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end English speech to Dutch text translation.
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-mustc-en-nl-st")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-mustc-en-nl-st")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
inputs = processor(
ds["speech"][0],
sampling_rate=16_000,
return_tensors="pt"
)
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
translation = processor.batch_decode(generated_ids, skip_special_tokens=True)
```
## Training data
The s2t-small-mustc-en-nl-st is trained on English-Dutch subset of [MuST-C](https://ict.fbk.eu/must-c/).
MuST-C is a multilingual speech translation corpus whose size and quality facilitates the training of end-to-end systems
for speech translation from English into several languages. For each target language, MuST-C comprises several hundred
hours of audio recordings from English TED Talks, which are automatically aligned at the sentence level with their manual
transcriptions and translations.
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 8,000.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
model training and for better performance the encoder is pre-trained for English ASR.
## Evaluation results
MuST-C test results for en-nl (BLEU score): 27.3
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": ["en", "nl"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition"], "datasets": ["mustc"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}]}
|
facebook/s2t-small-mustc-en-nl-st
| null |
[
"transformers",
"pytorch",
"tf",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"speech-translation",
"en",
"nl",
"dataset:mustc",
"arxiv:2010.05171",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-SMALL-MUSTC-EN-PT-ST
`s2t-small-mustc-en-pt-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Speech Translation (ST).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
transcripts/translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end English speech to Portuguese text translation.
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-mustc-en-pt-st")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-mustc-en-pt-st")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
inputs = processor(
ds["speech"][0],
sampling_rate=16_000,
return_tensors="pt"
)
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
translation = processor.batch_decode(generated_ids, skip_special_tokens=True)
```
## Training data
The s2t-small-mustc-en-pt-st is trained on English-Portuguese subset of [MuST-C](https://ict.fbk.eu/must-c/).
MuST-C is a multilingual speech translation corpus whose size and quality facilitates the training of end-to-end systems
for speech translation from English into several languages. For each target language, MuST-C comprises several hundred
hours of audio recordings from English TED Talks, which are automatically aligned at the sentence level with their manual
transcriptions and translations.
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 8,000.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
model training and for better performance the encoder is pre-trained for English ASR.
## Evaluation results
MuST-C test results for en-pt (BLEU score): 28.1
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": ["en", "pt"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition"], "datasets": ["mustc"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}]}
|
facebook/s2t-small-mustc-en-pt-st
| null |
[
"transformers",
"pytorch",
"tf",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"speech-translation",
"en",
"pt",
"dataset:mustc",
"arxiv:2010.05171",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-SMALL-MUSTC-EN-RO-ST
`s2t-small-mustc-en-ro-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Speech Translation (ST).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
transcripts/translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end English speech to Romanian text translation.
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-mustc-en-ro-st")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-mustc-en-ro-st")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
inputs = processor(
ds["speech"][0],
sampling_rate=16_000,
return_tensors="pt"
)
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
translation = processor.batch_decode(generated_ids, skip_special_tokens=True)
```
## Training data
The s2t-small-mustc-en-ro-st is trained on English-Romanian subset of [MuST-C](https://ict.fbk.eu/must-c/).
MuST-C is a multilingual speech translation corpus whose size and quality facilitates the training of end-to-end systems
for speech translation from English into several languages. For each target language, MuST-C comprises several hundred
hours of audio recordings from English TED Talks, which are automatically aligned at the sentence level with their manual
transcriptions and translations.
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 8,000.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
model training and for better performance the encoder is pre-trained for English ASR.
## Evaluation results
MuST-C test results for en-ro (BLEU score): 21.9
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": ["en", "ro"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition"], "datasets": ["mustc"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}]}
|
facebook/s2t-small-mustc-en-ro-st
| null |
[
"transformers",
"pytorch",
"tf",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"speech-translation",
"en",
"ro",
"dataset:mustc",
"arxiv:2010.05171",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T-SMALL-MUSTC-EN-RU-ST
`s2t-small-mustc-en-ru-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Speech Translation (ST).
The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
## Model description
S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
transcripts/translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end English speech to Russian text translation.
See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio) to extract the
filter bank features. Make sure to install the `torchaudio` package before running this example.*
You could either install those as extra speech dependancies with
`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly
with `pip install torchaudio sentencepiece`.
```python
import torch
from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
from datasets import load_dataset
import soundfile as sf
model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-mustc-en-ru-st")
processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-mustc-en-ru-st")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset(
"patrickvonplaten/librispeech_asr_dummy",
"clean",
split="validation"
)
ds = ds.map(map_to_array)
inputs = processor(
ds["speech"][0],
sampling_rate=16_000,
return_tensors="pt"
)
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
translation = processor.batch_decode(generated_ids, skip_special_tokens=True)
```
## Training data
The s2t-small-mustc-en-ru-st is trained on English-Russian subset of [MuST-C](https://ict.fbk.eu/must-c/).
MuST-C is a multilingual speech translation corpus whose size and quality facilitates the training of end-to-end systems
for speech translation from English into several languages. For each target language, MuST-C comprises several hundred
hours of audio recordings from English TED Talks, which are automatically aligned at the sentence level with their manual
transcriptions and translations.
## Training procedure
### Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
is applied to each example.
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 8,000.
### Training
The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
model training and for better performance the encoder is pre-trained for English ASR.
## Evaluation results
MuST-C test results for en-ru (BLEU score): 15.3
### BibTeX entry and citation info
```bibtex
@inproceedings{wang2020fairseqs2t,
title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
year = {2020},
}
```
|
{"language": ["en", "ru"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition"], "datasets": ["mustc"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Librispeech sample 1", "src": "https://cdn-media.huggingface.co/speech_samples/sample1.flac"}, {"example_title": "Librispeech sample 2", "src": "https://cdn-media.huggingface.co/speech_samples/sample2.flac"}]}
|
facebook/s2t-small-mustc-en-ru-st
| null |
[
"transformers",
"pytorch",
"tf",
"speech_to_text",
"automatic-speech-recognition",
"audio",
"speech-translation",
"en",
"ru",
"dataset:mustc",
"arxiv:2010.05171",
"arxiv:1904.08779",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T2-Wav2Vec2-CoVoST2-EN-AR-ST
`s2t-wav2vec2-large-en-ar` is a Speech to Text Transformer model trained for end-to-end Speech Translation (ST).
The S2T2 model was proposed in [Large-Scale Self- and Semi-Supervised Learning for Speech Translation](https://arxiv.org/pdf/2104.06678.pdf) and officially released in
[Fairseq](https://github.com/pytorch/fairseq/blob/6f847c8654d56b4d1b1fbacec027f47419426ddb/fairseq/models/wav2vec/wav2vec2_asr.py#L266).
## Model description
S2T2 is a transformer-based seq2seq (speech encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a pretrained [Wav2Vec2](https://huggingface.co/transformers/model_doc/wav2vec2.html) as the encoder and a transformer-based decoder. The model is trained with standard autoregressive cross-entropy loss and generates the translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end English speech to Arabic text translation.
See the [model hub](https://huggingface.co/models?filter=speech2text2) to look for other S2T2 checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
You can use the model directly via the ASR pipeline
```python
from datasets import load_dataset
from transformers import pipeline
librispeech_en = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
asr = pipeline("automatic-speech-recognition", model="facebook/s2t-wav2vec2-large-en-ar", feature_extractor="facebook/s2t-wav2vec2-large-en-ar")
translation = asr(librispeech_en[0]["file"])
```
or step-by-step as follows:
```python
import torch
from transformers import Speech2Text2Processor, SpeechEncoderDecoder
from datasets import load_dataset
import soundfile as sf
model = SpeechEncoderDecoder.from_pretrained("facebook/s2t-wav2vec2-large-en-ar")
processor = Speech2Text2Processor.from_pretrained("facebook/s2t-wav2vec2-large-en-ar")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
ds = ds.map(map_to_array)
inputs = processor(ds["speech"][0], sampling_rate=16_000, return_tensors="pt")
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
transcription = processor.batch_decode(generated_ids)
```
## Evaluation results
CoVoST-V2 test results for en-ar (BLEU score): **20.2**
For more information, please have a look at the [official paper](https://arxiv.org/pdf/2104.06678.pdf) - especially row 10 of Table 2.
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2104-06678,
author = {Changhan Wang and
Anne Wu and
Juan Miguel Pino and
Alexei Baevski and
Michael Auli and
Alexis Conneau},
title = {Large-Scale Self- and Semi-Supervised Learning for Speech Translation},
journal = {CoRR},
volume = {abs/2104.06678},
year = {2021},
url = {https://arxiv.org/abs/2104.06678},
archivePrefix = {arXiv},
eprint = {2104.06678},
timestamp = {Thu, 12 Aug 2021 15:37:06 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2104-06678.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"language": ["en", "ar"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition", "speech2text2"], "datasets": ["covost2", "librispeech_asr"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Common Voice 1", "src": "https://cdn-media.huggingface.co/speech_samples/common_voice_en_18301577.mp3"}, {"example_title": "Common Voice 2", "src": "https://cdn-media.huggingface.co/speech_samples/common_voice_en_99987.mp3"}, {"example_title": "Common Voice 3", "src": "https://cdn-media.huggingface.co/speech_samples/common_voice_en_99988.mp3"}]}
|
facebook/s2t-wav2vec2-large-en-ar
| null |
[
"transformers",
"pytorch",
"speech-encoder-decoder",
"automatic-speech-recognition",
"audio",
"speech-translation",
"speech2text2",
"en",
"ar",
"dataset:covost2",
"dataset:librispeech_asr",
"arxiv:2104.06678",
"license:mit",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T2-Wav2Vec2-CoVoST2-EN-CA-ST
`s2t-wav2vec2-large-en-ca` is a Speech to Text Transformer model trained for end-to-end Speech Translation (ST).
The S2T2 model was proposed in [Large-Scale Self- and Semi-Supervised Learning for Speech Translation](https://arxiv.org/pdf/2104.06678.pdf) and officially released in
[Fairseq](https://github.com/pytorch/fairseq/blob/6f847c8654d56b4d1b1fbacec027f47419426ddb/fairseq/models/wav2vec/wav2vec2_asr.py#L266).
## Model description
S2T2 is a transformer-based seq2seq (speech encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a pretrained [Wav2Vec2](https://huggingface.co/transformers/model_doc/wav2vec2.html) as the encoder and a transformer-based decoder. The model is trained with standard autoregressive cross-entropy loss and generates the translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end English speech to Catalan text translation.
See the [model hub](https://huggingface.co/models?filter=speech2text2) to look for other S2T2 checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
You can use the model directly via the ASR pipeline
```python
from datasets import load_dataset
from transformers import pipeline
librispeech_en = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
asr = pipeline("automatic-speech-recognition", model="facebook/s2t-wav2vec2-large-en-ca", feature_extractor="facebook/s2t-wav2vec2-large-en-ca")
translation = asr(librispeech_en[0]["file"])
```
or step-by-step as follows:
```python
import torch
from transformers import Speech2Text2Processor, SpeechEncoderDecoder
from datasets import load_dataset
import soundfile as sf
model = SpeechEncoderDecoder.from_pretrained("facebook/s2t-wav2vec2-large-en-ca")
processor = Speech2Text2Processor.from_pretrained("facebook/s2t-wav2vec2-large-en-ca")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
ds = ds.map(map_to_array)
inputs = processor(ds["speech"][0], sampling_rate=16_000, return_tensors="pt")
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
transcription = processor.batch_decode(generated_ids)
```
## Evaluation results
CoVoST-V2 test results for en-ca (BLEU score): **34.1**
For more information, please have a look at the [official paper](https://arxiv.org/pdf/2104.06678.pdf) - especially row 10 of Table 2.
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2104-06678,
author = {Changhan Wang and
Anne Wu and
Juan Miguel Pino and
Alexei Baevski and
Michael Auli and
Alexis Conneau},
title = {Large-Scale Self- and Semi-Supervised Learning for Speech Translation},
journal = {CoRR},
volume = {abs/2104.06678},
year = {2021},
url = {https://arxiv.org/abs/2104.06678},
archivePrefix = {arXiv},
eprint = {2104.06678},
timestamp = {Thu, 12 Aug 2021 15:37:06 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2104-06678.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"language": ["en", "ca"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition", "speech2text2"], "datasets": ["covost2", "librispeech_asr"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Common Voice 1", "src": "https://cdn-media.huggingface.co/speech_samples/common_voice_en_18301577.mp3"}, {"example_title": "Common Voice 2", "src": "https://cdn-media.huggingface.co/speech_samples/common_voice_en_99989.mp3"}, {"example_title": "Common Voice 3", "src": "https://cdn-media.huggingface.co/speech_samples/common_voice_en_9999.mp3"}]}
|
facebook/s2t-wav2vec2-large-en-ca
| null |
[
"transformers",
"pytorch",
"speech-encoder-decoder",
"automatic-speech-recognition",
"audio",
"speech-translation",
"speech2text2",
"en",
"ca",
"dataset:covost2",
"dataset:librispeech_asr",
"arxiv:2104.06678",
"license:mit",
"endpoints_compatible",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T2-Wav2Vec2-CoVoST2-EN-DE-ST
`s2t-wav2vec2-large-en-de` is a Speech to Text Transformer model trained for end-to-end Speech Translation (ST).
The S2T2 model was proposed in [Large-Scale Self- and Semi-Supervised Learning for Speech Translation](https://arxiv.org/pdf/2104.06678.pdf) and officially released in
[Fairseq](https://github.com/pytorch/fairseq/blob/6f847c8654d56b4d1b1fbacec027f47419426ddb/fairseq/models/wav2vec/wav2vec2_asr.py#L266).
## Model description
S2T2 is a transformer-based seq2seq (speech encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a pretrained [Wav2Vec2](https://huggingface.co/transformers/model_doc/wav2vec2.html) as the encoder and a transformer-based decoder. The model is trained with standard autoregressive cross-entropy loss and generates the translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end English speech to German text translation.
See the [model hub](https://huggingface.co/models?filter=speech2text2) to look for other S2T2 checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
You can use the model directly via the ASR pipeline
```python
from datasets import load_dataset
from transformers import pipeline
librispeech_en = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
asr = pipeline("automatic-speech-recognition", model="facebook/s2t-wav2vec2-large-en-de", feature_extractor="facebook/s2t-wav2vec2-large-en-de")
translation_de = asr(librispeech_en[0]["file"])
```
or step-by-step as follows:
```python
import torch
from transformers import Speech2Text2Processor, SpeechEncoderDecoder
from datasets import load_dataset
import soundfile as sf
model = SpeechEncoderDecoder.from_pretrained("facebook/s2t-wav2vec2-large-en-de")
processor = Speech2Text2Processor.from_pretrained("facebook/s2t-wav2vec2-large-en-de")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
ds = ds.map(map_to_array)
inputs = processor(ds["speech"][0], sampling_rate=16_000, return_tensors="pt")
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
transcription = processor.batch_decode(generated_ids)
```
## Evaluation results
CoVoST-V2 test results for en-de (BLEU score): **26.5**
For more information, please have a look at the [official paper](https://arxiv.org/pdf/2104.06678.pdf) - especially row 10 of Table 2.
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2104-06678,
author = {Changhan Wang and
Anne Wu and
Juan Miguel Pino and
Alexei Baevski and
Michael Auli and
Alexis Conneau},
title = {Large-Scale Self- and Semi-Supervised Learning for Speech Translation},
journal = {CoRR},
volume = {abs/2104.06678},
year = {2021},
url = {https://arxiv.org/abs/2104.06678},
archivePrefix = {arXiv},
eprint = {2104.06678},
timestamp = {Thu, 12 Aug 2021 15:37:06 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2104-06678.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"language": ["en", "de"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition", "speech2text2"], "datasets": ["covost2", "librispeech_asr"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Common Voice 1", "src": "https://cdn-media.huggingface.co/speech_samples/common_voice_en_18301577.mp3"}, {"example_title": "Common Voice 2", "src": "https://cdn-media.huggingface.co/speech_samples/common_voice_en_99985.mp3"}, {"example_title": "Common Voice 3", "src": "https://cdn-media.huggingface.co/speech_samples/common_voice_en_99986.mp3"}]}
|
facebook/s2t-wav2vec2-large-en-de
| null |
[
"transformers",
"pytorch",
"speech-encoder-decoder",
"automatic-speech-recognition",
"audio",
"speech-translation",
"speech2text2",
"en",
"de",
"dataset:covost2",
"dataset:librispeech_asr",
"arxiv:2104.06678",
"license:mit",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
automatic-speech-recognition
|
transformers
|
# S2T2-Wav2Vec2-CoVoST2-EN-TR-ST
`s2t-wav2vec2-large-en-tr` is a Speech to Text Transformer model trained for end-to-end Speech Translation (ST).
The S2T2 model was proposed in [Large-Scale Self- and Semi-Supervised Learning for Speech Translation](https://arxiv.org/pdf/2104.06678.pdf) and officially released in
[Fairseq](https://github.com/pytorch/fairseq/blob/6f847c8654d56b4d1b1fbacec027f47419426ddb/fairseq/models/wav2vec/wav2vec2_asr.py#L266).
## Model description
S2T2 is a transformer-based seq2seq (speech encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
Translation (ST). It uses a pretrained [Wav2Vec2](https://huggingface.co/transformers/model_doc/wav2vec2.html) as the encoder and a transformer-based decoder. The model is trained with standard autoregressive cross-entropy loss and generates the translations autoregressively.
## Intended uses & limitations
This model can be used for end-to-end English speech to Turkish text translation.
See the [model hub](https://huggingface.co/models?filter=speech2text2) to look for other S2T2 checkpoints.
### How to use
As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
transcripts by passing the speech features to the model.
You can use the model directly via the ASR pipeline
```python
from datasets import load_dataset
from transformers import pipeline
librispeech_en = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
asr = pipeline("automatic-speech-recognition", model="facebook/s2t-wav2vec2-large-en-tr", feature_extractor="facebook/s2t-wav2vec2-large-en-tr")
translation = asr(librispeech_en[0]["file"])
```
or step-by-step as follows:
```python
import torch
from transformers import Speech2Text2Processor, SpeechEncoderDecoder
from datasets import load_dataset
import soundfile as sf
model = SpeechEncoderDecoder.from_pretrained("facebook/s2t-wav2vec2-large-en-tr")
processor = Speech2Text2Processor.from_pretrained("facebook/s2t-wav2vec2-large-en-tr")
def map_to_array(batch):
speech, _ = sf.read(batch["file"])
batch["speech"] = speech
return batch
ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
ds = ds.map(map_to_array)
inputs = processor(ds["speech"][0], sampling_rate=16_000, return_tensors="pt")
generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
transcription = processor.batch_decode(generated_ids)
```
## Evaluation results
CoVoST-V2 test results for en-tr (BLEU score): **17.5**
For more information, please have a look at the [official paper](https://arxiv.org/pdf/2104.06678.pdf) - especially row 10 of Table 2.
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2104-06678,
author = {Changhan Wang and
Anne Wu and
Juan Miguel Pino and
Alexei Baevski and
Michael Auli and
Alexis Conneau},
title = {Large-Scale Self- and Semi-Supervised Learning for Speech Translation},
journal = {CoRR},
volume = {abs/2104.06678},
year = {2021},
url = {https://arxiv.org/abs/2104.06678},
archivePrefix = {arXiv},
eprint = {2104.06678},
timestamp = {Thu, 12 Aug 2021 15:37:06 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2104-06678.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
|
{"language": ["en", "tr"], "license": "mit", "tags": ["audio", "speech-translation", "automatic-speech-recognition", "speech2text2"], "datasets": ["covost2", "librispeech_asr"], "pipeline_tag": "automatic-speech-recognition", "widget": [{"example_title": "Common Voice 1", "src": "https://cdn-media.huggingface.co/speech_samples/common_voice_en_99989.mp3"}, {"example_title": "Common Voice 2", "src": "https://cdn-media.huggingface.co/speech_samples/common_voice_en_99986.mp3"}, {"example_title": "Common Voice 3", "src": "https://cdn-media.huggingface.co/speech_samples/common_voice_en_99987.mp3"}]}
|
facebook/s2t-wav2vec2-large-en-tr
| null |
[
"transformers",
"pytorch",
"speech-encoder-decoder",
"automatic-speech-recognition",
"audio",
"speech-translation",
"speech2text2",
"en",
"tr",
"dataset:covost2",
"dataset:librispeech_asr",
"arxiv:2104.06678",
"license:mit",
"endpoints_compatible",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text-to-speech
|
fairseq
|
# tts_transformer-ar-cv7
[Transformer](https://arxiv.org/abs/1809.08895) text-to-speech model from fairseq S^2 ([paper](https://arxiv.org/abs/2109.06912)/[code](https://github.com/pytorch/fairseq/tree/main/examples/speech_synthesis)):
- Arabic
- Single-speaker male voice
- Trained on [Common Voice v7](https://commonvoice.mozilla.org/en/datasets)
## Usage
```python
from fairseq.checkpoint_utils import load_model_ensemble_and_task_from_hf_hub
from fairseq.models.text_to_speech.hub_interface import TTSHubInterface
import IPython.display as ipd
models, cfg, task = load_model_ensemble_and_task_from_hf_hub(
"facebook/tts_transformer-ar-cv7",
arg_overrides={"vocoder": "hifigan", "fp16": False}
)
model = models[0]
TTSHubInterface.update_cfg_with_data_cfg(cfg, task.data_cfg)
generator = task.build_generator(model, cfg)
text = "مرحبًا ، هذا اختبار تشغيل."
sample = TTSHubInterface.get_model_input(task, text)
wav, rate = TTSHubInterface.get_prediction(task, model, generator, sample)
ipd.Audio(wav, rate=rate)
```
See also [fairseq S^2 example](https://github.com/pytorch/fairseq/blob/main/examples/speech_synthesis/docs/common_voice_example.md).
## Citation
```bibtex
@inproceedings{wang-etal-2021-fairseq,
title = "fairseq S{\^{}}2: A Scalable and Integrable Speech Synthesis Toolkit",
author = "Wang, Changhan and
Hsu, Wei-Ning and
Adi, Yossi and
Polyak, Adam and
Lee, Ann and
Chen, Peng-Jen and
Gu, Jiatao and
Pino, Juan",
booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing: System Demonstrations",
month = nov,
year = "2021",
address = "Online and Punta Cana, Dominican Republic",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.emnlp-demo.17",
doi = "10.18653/v1/2021.emnlp-demo.17",
pages = "143--152",
}
```
|
{"language": "ar", "library_name": "fairseq", "tags": ["fairseq", "audio", "text-to-speech"], "datasets": ["common_voice"], "task": "text-to-speech", "widget": [{"text": "\u0645\u0631\u062d\u0628\u064b\u0627 \u060c \u0647\u0630\u0627 \u0627\u062e\u062a\u0628\u0627\u0631 \u062a\u0634\u063a\u064a\u0644.", "example_title": "Hello, this is a test run."}]}
|
facebook/tts_transformer-ar-cv7
| null |
[
"fairseq",
"audio",
"text-to-speech",
"ar",
"dataset:common_voice",
"arxiv:1809.08895",
"arxiv:2109.06912",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text-to-speech
|
fairseq
|
# tts_transformer-en-200_speaker-cv4
[Transformer](https://arxiv.org/abs/1809.08895) text-to-speech model from fairseq S^2 ([paper](https://arxiv.org/abs/2109.06912)/[code](https://github.com/pytorch/fairseq/tree/main/examples/speech_synthesis)):
- English
- 200 male/female voices (random speaker when using the widget)
- Trained on [Common Voice v4](https://commonvoice.mozilla.org/en/datasets)
## Usage
```python
from fairseq.checkpoint_utils import load_model_ensemble_and_task_from_hf_hub
from fairseq.models.text_to_speech.hub_interface import TTSHubInterface
import IPython.display as ipd
models, cfg, task = load_model_ensemble_and_task_from_hf_hub(
"facebook/tts_transformer-en-200_speaker-cv4",
arg_overrides={"vocoder": "hifigan", "fp16": False}
)
model = models[0]
TTSHubInterface.update_cfg_with_data_cfg(cfg, task.data_cfg)
generator = task.build_generator(model, cfg)
text = "Hello, this is a test run."
sample = TTSHubInterface.get_model_input(task, text)
wav, rate = TTSHubInterface.get_prediction(task, model, generator, sample)
ipd.Audio(wav, rate=rate)
```
See also [fairseq S^2 example](https://github.com/pytorch/fairseq/blob/main/examples/speech_synthesis/docs/common_voice_example.md).
## Citation
```bibtex
@inproceedings{wang-etal-2021-fairseq,
title = "fairseq S{\^{}}2: A Scalable and Integrable Speech Synthesis Toolkit",
author = "Wang, Changhan and
Hsu, Wei-Ning and
Adi, Yossi and
Polyak, Adam and
Lee, Ann and
Chen, Peng-Jen and
Gu, Jiatao and
Pino, Juan",
booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing: System Demonstrations",
month = nov,
year = "2021",
address = "Online and Punta Cana, Dominican Republic",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.emnlp-demo.17",
doi = "10.18653/v1/2021.emnlp-demo.17",
pages = "143--152",
}
```
|
{"language": "en", "library_name": "fairseq", "tags": ["fairseq", "audio", "text-to-speech", "multi-speaker"], "datasets": ["common_voice"], "task": "text-to-speech", "widget": [{"text": "Hello, this is a test run.", "example_title": "Hello, this is a test run."}]}
|
facebook/tts_transformer-en-200_speaker-cv4
| null |
[
"fairseq",
"audio",
"text-to-speech",
"multi-speaker",
"en",
"dataset:common_voice",
"arxiv:1809.08895",
"arxiv:2109.06912",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text-to-speech
|
fairseq
|
# tts_transformer-en-ljspeech
[Transformer](https://arxiv.org/abs/1809.08895) text-to-speech model from fairseq S^2 ([paper](https://arxiv.org/abs/2109.06912)/[code](https://github.com/pytorch/fairseq/tree/main/examples/speech_synthesis)):
- English
- Single-speaker female voice
- Trained on [LJSpeech](https://keithito.com/LJ-Speech-Dataset/)
## Usage
```python
from fairseq.checkpoint_utils import load_model_ensemble_and_task_from_hf_hub
from fairseq.models.text_to_speech.hub_interface import TTSHubInterface
import IPython.display as ipd
models, cfg, task = load_model_ensemble_and_task_from_hf_hub(
"facebook/tts_transformer-en-ljspeech",
arg_overrides={"vocoder": "hifigan", "fp16": False}
)
model = models[0]
TTSHubInterface.update_cfg_with_data_cfg(cfg, task.data_cfg)
generator = task.build_generator(model, cfg)
text = "Hello, this is a test run."
sample = TTSHubInterface.get_model_input(task, text)
wav, rate = TTSHubInterface.get_prediction(task, model, generator, sample)
ipd.Audio(wav, rate=rate)
```
See also [fairseq S^2 example](https://github.com/pytorch/fairseq/blob/main/examples/speech_synthesis/docs/ljspeech_example.md).
## Citation
```bibtex
@inproceedings{wang-etal-2021-fairseq,
title = "fairseq S{\^{}}2: A Scalable and Integrable Speech Synthesis Toolkit",
author = "Wang, Changhan and
Hsu, Wei-Ning and
Adi, Yossi and
Polyak, Adam and
Lee, Ann and
Chen, Peng-Jen and
Gu, Jiatao and
Pino, Juan",
booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing: System Demonstrations",
month = nov,
year = "2021",
address = "Online and Punta Cana, Dominican Republic",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.emnlp-demo.17",
doi = "10.18653/v1/2021.emnlp-demo.17",
pages = "143--152",
}
```
|
{"language": "en", "library_name": "fairseq", "tags": ["fairseq", "audio", "text-to-speech"], "datasets": ["ljspeech"], "task": "text-to-speech", "widget": [{"text": "Hello, this is a test run.", "example_title": "Hello, this is a test run."}]}
|
facebook/tts_transformer-en-ljspeech
| null |
[
"fairseq",
"audio",
"text-to-speech",
"en",
"dataset:ljspeech",
"arxiv:1809.08895",
"arxiv:2109.06912",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
text-to-speech
|
fairseq
|
# tts_transformer-es-css10
[Transformer](https://arxiv.org/abs/1809.08895) text-to-speech model from fairseq S^2 ([paper](https://arxiv.org/abs/2109.06912)/[code](https://github.com/pytorch/fairseq/tree/main/examples/speech_synthesis)):
- Spanish
- Single-speaker male voice
- Trained on [CSS10](https://github.com/Kyubyong/css10)
## Usage
Dependencies
```sh
pip install fairseq sentencepiece
```
```python
from fairseq.checkpoint_utils import load_model_ensemble_and_task_from_hf_hub
from fairseq.models.text_to_speech.hub_interface import TTSHubInterface
import IPython.display as ipd
models, cfg, task = load_model_ensemble_and_task_from_hf_hub(
"facebook/tts_transformer-es-css10",
arg_overrides={"vocoder": "hifigan", "fp16": False}
)
model = models[0]
TTSHubInterface.update_cfg_with_data_cfg(cfg, task.data_cfg)
generator = task.build_generator([model], cfg)
text = "Hola, esta es una prueba."
sample = TTSHubInterface.get_model_input(task, text)
wav, rate = TTSHubInterface.get_prediction(task, model, generator, sample)
ipd.Audio(wav, rate=rate)
```
See also [fairseq S^2 example](https://github.com/pytorch/fairseq/blob/main/examples/speech_synthesis/docs/common_voice_example.md).
## Citation
```bibtex
@inproceedings{wang-etal-2021-fairseq,
title = "fairseq S{\^{}}2: A Scalable and Integrable Speech Synthesis Toolkit",
author = "Wang, Changhan and
Hsu, Wei-Ning and
Adi, Yossi and
Polyak, Adam and
Lee, Ann and
Chen, Peng-Jen and
Gu, Jiatao and
Pino, Juan",
booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing: System Demonstrations",
month = nov,
year = "2021",
address = "Online and Punta Cana, Dominican Republic",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.emnlp-demo.17",
doi = "10.18653/v1/2021.emnlp-demo.17",
pages = "143--152",
}
```
|
{"language": "es", "library_name": "fairseq", "tags": ["fairseq", "audio", "text-to-speech"], "datasets": ["css10"], "task": "text-to-speech", "widget": [{"text": "Hola, esta es una prueba.", "example_title": "Hello, this is a test run."}]}
|
facebook/tts_transformer-es-css10
| null |
[
"fairseq",
"audio",
"text-to-speech",
"es",
"dataset:css10",
"arxiv:1809.08895",
"arxiv:2109.06912",
"has_space",
"region:us"
] | null |
2022-03-02T23:29:05+00:00
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.