From 9ae22fe3c1b81f99a764d382054b6ebe2b025bd4 Mon Sep 17 00:00:00 2001 From: Lysandre Date: Mon, 10 Oct 2022 16:45:44 -0400 Subject: [PATCH 1/6] Release: v4.23.0 --- README.md | 12 ++++++------ README_ko.md | 12 ++++++------ README_zh-hans.md | 12 ++++++------ README_zh-hant.md | 12 ++++++------ examples/flax/question-answering/run_qa.py | 2 +- examples/flax/text-classification/run_flax_glue.py | 2 +- examples/flax/token-classification/run_flax_ner.py | 2 +- .../audio-classification/run_audio_classification.py | 2 +- examples/pytorch/contrastive-image-text/run_clip.py | 2 +- .../image-classification/run_image_classification.py | 2 +- .../run_image_classification_no_trainer.py | 2 +- examples/pytorch/image-pretraining/run_mae.py | 2 +- examples/pytorch/image-pretraining/run_mim.py | 2 +- examples/pytorch/language-modeling/run_clm.py | 2 +- .../pytorch/language-modeling/run_clm_no_trainer.py | 2 +- examples/pytorch/language-modeling/run_mlm.py | 2 +- .../pytorch/language-modeling/run_mlm_no_trainer.py | 2 +- examples/pytorch/language-modeling/run_plm.py | 2 +- examples/pytorch/multiple-choice/run_swag.py | 2 +- .../pytorch/multiple-choice/run_swag_no_trainer.py | 2 +- examples/pytorch/question-answering/run_qa.py | 2 +- .../pytorch/question-answering/run_qa_beam_search.py | 2 +- .../run_qa_beam_search_no_trainer.py | 2 +- .../pytorch/question-answering/run_qa_no_trainer.py | 2 +- .../pytorch/question-answering/run_seq2seq_qa.py | 2 +- .../run_semantic_segmentation.py | 2 +- .../run_semantic_segmentation_no_trainer.py | 2 +- .../speech-recognition/run_speech_recognition_ctc.py | 2 +- .../run_speech_recognition_seq2seq.py | 2 +- examples/pytorch/summarization/run_summarization.py | 2 +- .../summarization/run_summarization_no_trainer.py | 2 +- examples/pytorch/text-classification/run_glue.py | 2 +- .../text-classification/run_glue_no_trainer.py | 2 +- examples/pytorch/text-classification/run_xnli.py | 2 +- examples/pytorch/token-classification/run_ner.py | 2 +- .../token-classification/run_ner_no_trainer.py | 2 +- examples/pytorch/translation/run_translation.py | 2 +- .../translation/run_translation_no_trainer.py | 2 +- examples/tensorflow/multiple-choice/run_swag.py | 2 +- examples/tensorflow/question-answering/run_qa.py | 2 +- .../tensorflow/summarization/run_summarization.py | 2 +- examples/tensorflow/text-classification/run_glue.py | 2 +- examples/tensorflow/translation/run_translation.py | 2 +- setup.py | 2 +- src/transformers/__init__.py | 2 +- 45 files changed, 65 insertions(+), 65 deletions(-) diff --git a/README.md b/README.md index 4cdf062c188ea1..a1c6c4a77041a6 100644 --- a/README.md +++ b/README.md @@ -278,7 +278,7 @@ Current number of checkpoints: ![](https://img.shields.io/endpoint?url=https://h 1. **[CANINE](https://huggingface.co/docs/transformers/model_doc/canine)** (from Google Research) released with the paper [CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representation](https://arxiv.org/abs/2103.06874) by Jonathan H. Clark, Dan Garrette, Iulia Turc, John Wieting. 1. **[CLIP](https://huggingface.co/docs/transformers/model_doc/clip)** (from OpenAI) released with the paper [Learning Transferable Visual Models From Natural Language Supervision](https://arxiv.org/abs/2103.00020) by Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, Ilya Sutskever. 1. **[CodeGen](https://huggingface.co/docs/transformers/model_doc/codegen)** (from Salesforce) released with the paper [A Conversational Paradigm for Program Synthesis](https://arxiv.org/abs/2203.13474) by Erik Nijkamp, Bo Pang, Hiroaki Hayashi, Lifu Tu, Huan Wang, Yingbo Zhou, Silvio Savarese, Caiming Xiong. -1. **[Conditional DETR](https://huggingface.co/docs/transformers/main/model_doc/conditional_detr)** (from Microsoft Research Asia) released with the paper [Conditional DETR for Fast Training Convergence](https://arxiv.org/abs/2108.06152) by Depu Meng, Xiaokang Chen, Zejia Fan, Gang Zeng, Houqiang Li, Yuhui Yuan, Lei Sun, Jingdong Wang. +1. **[Conditional DETR](https://huggingface.co/docs/transformers/model_doc/conditional_detr)** (from Microsoft Research Asia) released with the paper [Conditional DETR for Fast Training Convergence](https://arxiv.org/abs/2108.06152) by Depu Meng, Xiaokang Chen, Zejia Fan, Gang Zeng, Houqiang Li, Yuhui Yuan, Lei Sun, Jingdong Wang. 1. **[ConvBERT](https://huggingface.co/docs/transformers/model_doc/convbert)** (from YituTech) released with the paper [ConvBERT: Improving BERT with Span-based Dynamic Convolution](https://arxiv.org/abs/2008.02496) by Zihang Jiang, Weihao Yu, Daquan Zhou, Yunpeng Chen, Jiashi Feng, Shuicheng Yan. 1. **[ConvNeXT](https://huggingface.co/docs/transformers/model_doc/convnext)** (from Facebook AI) released with the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Zhuang Liu, Hanzi Mao, Chao-Yuan Wu, Christoph Feichtenhofer, Trevor Darrell, Saining Xie. 1. **[CPM](https://huggingface.co/docs/transformers/model_doc/cpm)** (from Tsinghua University) released with the paper [CPM: A Large-scale Generative Chinese Pre-trained Language Model](https://arxiv.org/abs/2012.00413) by Zhengyan Zhang, Xu Han, Hao Zhou, Pei Ke, Yuxian Gu, Deming Ye, Yujia Qin, Yusheng Su, Haozhe Ji, Jian Guan, Fanchao Qi, Xiaozhi Wang, Yanan Zheng, Guoyang Zeng, Huanqi Cao, Shengqi Chen, Daixuan Li, Zhenbo Sun, Zhiyuan Liu, Minlie Huang, Wentao Han, Jie Tang, Juanzi Li, Xiaoyan Zhu, Maosong Sun. @@ -300,7 +300,7 @@ Current number of checkpoints: ![](https://img.shields.io/endpoint?url=https://h 1. **[ELECTRA](https://huggingface.co/docs/transformers/model_doc/electra)** (from Google Research/Stanford University) released with the paper [ELECTRA: Pre-training text encoders as discriminators rather than generators](https://arxiv.org/abs/2003.10555) by Kevin Clark, Minh-Thang Luong, Quoc V. Le, Christopher D. Manning. 1. **[EncoderDecoder](https://huggingface.co/docs/transformers/model_doc/encoder-decoder)** (from Google Research) released with the paper [Leveraging Pre-trained Checkpoints for Sequence Generation Tasks](https://arxiv.org/abs/1907.12461) by Sascha Rothe, Shashi Narayan, Aliaksei Severyn. 1. **[ERNIE](https://huggingface.co/docs/transformers/model_doc/ernie)** (from Baidu) released with the paper [ERNIE: Enhanced Representation through Knowledge Integration](https://arxiv.org/abs/1904.09223) by Yu Sun, Shuohuan Wang, Yukun Li, Shikun Feng, Xuyi Chen, Han Zhang, Xin Tian, Danxiang Zhu, Hao Tian, Hua Wu. -1. **[ESM](https://huggingface.co/docs/transformers/main/model_doc/esm)** (from Meta AI) are transformer protein language models. **ESM-1b** was released with the paper [Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences](https://www.pnas.org/content/118/15/e2016239118) by Alexander Rives, Joshua Meier, Tom Sercu, Siddharth Goyal, Zeming Lin, Jason Liu, Demi Guo, Myle Ott, C. Lawrence Zitnick, Jerry Ma, and Rob Fergus. **ESM-1v** was released with the paper [Language models enable zero-shot prediction of the effects of mutations on protein function](https://doi.org/10.1101/2021.07.09.450648) by Joshua Meier, Roshan Rao, Robert Verkuil, Jason Liu, Tom Sercu and Alexander Rives. **ESM-2** was released with the paper [Language models of protein sequences at the scale of evolution enable accurate structure prediction](https://doi.org/10.1101/2022.07.20.500902) by Zeming Lin, Halil Akin, Roshan Rao, Brian Hie, Zhongkai Zhu, Wenting Lu, Allan dos Santos Costa, Maryam Fazel-Zarandi, Tom Sercu, Sal Candido, Alexander Rives. +1. **[ESM](https://huggingface.co/docs/transformers/model_doc/esm)** (from Meta AI) are transformer protein language models. **ESM-1b** was released with the paper [Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences](https://www.pnas.org/content/118/15/e2016239118) by Alexander Rives, Joshua Meier, Tom Sercu, Siddharth Goyal, Zeming Lin, Jason Liu, Demi Guo, Myle Ott, C. Lawrence Zitnick, Jerry Ma, and Rob Fergus. **ESM-1v** was released with the paper [Language models enable zero-shot prediction of the effects of mutations on protein function](https://doi.org/10.1101/2021.07.09.450648) by Joshua Meier, Roshan Rao, Robert Verkuil, Jason Liu, Tom Sercu and Alexander Rives. **ESM-2** was released with the paper [Language models of protein sequences at the scale of evolution enable accurate structure prediction](https://doi.org/10.1101/2022.07.20.500902) by Zeming Lin, Halil Akin, Roshan Rao, Brian Hie, Zhongkai Zhu, Wenting Lu, Allan dos Santos Costa, Maryam Fazel-Zarandi, Tom Sercu, Sal Candido, Alexander Rives. 1. **[FlauBERT](https://huggingface.co/docs/transformers/model_doc/flaubert)** (from CNRS) released with the paper [FlauBERT: Unsupervised Language Model Pre-training for French](https://arxiv.org/abs/1912.05372) by Hang Le, Loïc Vial, Jibril Frej, Vincent Segonne, Maximin Coavoux, Benjamin Lecouteux, Alexandre Allauzen, Benoît Crabbé, Laurent Besacier, Didier Schwab. 1. **[FLAVA](https://huggingface.co/docs/transformers/model_doc/flava)** (from Facebook AI) released with the paper [FLAVA: A Foundational Language And Vision Alignment Model](https://arxiv.org/abs/2112.04482) by Amanpreet Singh, Ronghang Hu, Vedanuj Goswami, Guillaume Couairon, Wojciech Galuba, Marcus Rohrbach, and Douwe Kiela. 1. **[FNet](https://huggingface.co/docs/transformers/model_doc/fnet)** (from Google Research) released with the paper [FNet: Mixing Tokens with Fourier Transforms](https://arxiv.org/abs/2105.03824) by James Lee-Thorp, Joshua Ainslie, Ilya Eckstein, Santiago Ontanon. @@ -329,7 +329,7 @@ Current number of checkpoints: ![](https://img.shields.io/endpoint?url=https://h 1. **[M-CTC-T](https://huggingface.co/docs/transformers/model_doc/mctct)** (from Facebook) released with the paper [Pseudo-Labeling For Massively Multilingual Speech Recognition](https://arxiv.org/abs/2111.00161) by Loren Lugosch, Tatiana Likhomanenko, Gabriel Synnaeve, and Ronan Collobert. 1. **[M2M100](https://huggingface.co/docs/transformers/model_doc/m2m_100)** (from Facebook) released with the paper [Beyond English-Centric Multilingual Machine Translation](https://arxiv.org/abs/2010.11125) by Angela Fan, Shruti Bhosale, Holger Schwenk, Zhiyi Ma, Ahmed El-Kishky, Siddharth Goyal, Mandeep Baines, Onur Celebi, Guillaume Wenzek, Vishrav Chaudhary, Naman Goyal, Tom Birch, Vitaliy Liptchinsky, Sergey Edunov, Edouard Grave, Michael Auli, Armand Joulin. 1. **[MarianMT](https://huggingface.co/docs/transformers/model_doc/marian)** Machine translation models trained using [OPUS](http://opus.nlpl.eu/) data by Jörg Tiedemann. The [Marian Framework](https://marian-nmt.github.io/) is being developed by the Microsoft Translator Team. -1. **[MarkupLM](https://huggingface.co/docs/transformers/main/model_doc/markuplm)** (from Microsoft Research Asia) released with the paper [MarkupLM: Pre-training of Text and Markup Language for Visually-rich Document Understanding](https://arxiv.org/abs/2110.08518) by Junlong Li, Yiheng Xu, Lei Cui, Furu Wei. +1. **[MarkupLM](https://huggingface.co/docs/transformers/model_doc/markuplm)** (from Microsoft Research Asia) released with the paper [MarkupLM: Pre-training of Text and Markup Language for Visually-rich Document Understanding](https://arxiv.org/abs/2110.08518) by Junlong Li, Yiheng Xu, Lei Cui, Furu Wei. 1. **[MaskFormer](https://huggingface.co/docs/transformers/model_doc/maskformer)** (from Meta and UIUC) released with the paper [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://arxiv.org/abs/2107.06278) by Bowen Cheng, Alexander G. Schwing, Alexander Kirillov. 1. **[mBART](https://huggingface.co/docs/transformers/model_doc/mbart)** (from Facebook) released with the paper [Multilingual Denoising Pre-training for Neural Machine Translation](https://arxiv.org/abs/2001.08210) by Yinhan Liu, Jiatao Gu, Naman Goyal, Xian Li, Sergey Edunov, Marjan Ghazvininejad, Mike Lewis, Luke Zettlemoyer. 1. **[mBART-50](https://huggingface.co/docs/transformers/model_doc/mbart)** (from Facebook) released with the paper [Multilingual Translation with Extensible Multilingual Pretraining and Finetuning](https://arxiv.org/abs/2008.00401) by Yuqing Tang, Chau Tran, Xian Li, Peng-Jen Chen, Naman Goyal, Vishrav Chaudhary, Jiatao Gu, Angela Fan. @@ -375,7 +375,7 @@ Current number of checkpoints: ![](https://img.shields.io/endpoint?url=https://h 1. **[T5v1.1](https://huggingface.co/docs/transformers/model_doc/t5v1.1)** (from Google AI) released in the repository [google-research/text-to-text-transfer-transformer](https://github.com/google-research/text-to-text-transfer-transformer/blob/main/released_checkpoints.md#t511) by Colin Raffel and Noam Shazeer and Adam Roberts and Katherine Lee and Sharan Narang and Michael Matena and Yanqi Zhou and Wei Li and Peter J. Liu. 1. **[TAPAS](https://huggingface.co/docs/transformers/model_doc/tapas)** (from Google AI) released with the paper [TAPAS: Weakly Supervised Table Parsing via Pre-training](https://arxiv.org/abs/2004.02349) by Jonathan Herzig, Paweł Krzysztof Nowak, Thomas Müller, Francesco Piccinno and Julian Martin Eisenschlos. 1. **[TAPEX](https://huggingface.co/docs/transformers/model_doc/tapex)** (from Microsoft Research) released with the paper [TAPEX: Table Pre-training via Learning a Neural SQL Executor](https://arxiv.org/abs/2107.07653) by Qian Liu, Bei Chen, Jiaqi Guo, Morteza Ziyadi, Zeqi Lin, Weizhu Chen, Jian-Guang Lou. -1. **[Time Series Transformer](https://huggingface.co/docs/transformers/main/model_doc/time_series_transformer)** (from HuggingFace). +1. **[Time Series Transformer](https://huggingface.co/docs/transformers/model_doc/time_series_transformer)** (from HuggingFace). 1. **[Trajectory Transformer](https://huggingface.co/docs/transformers/model_doc/trajectory_transformers)** (from the University of California at Berkeley) released with the paper [Offline Reinforcement Learning as One Big Sequence Modeling Problem](https://arxiv.org/abs/2106.02039) by Michael Janner, Qiyang Li, Sergey Levine 1. **[Transformer-XL](https://huggingface.co/docs/transformers/model_doc/transfo-xl)** (from Google/CMU) released with the paper [Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context](https://arxiv.org/abs/1901.02860) by Zihang Dai*, Zhilin Yang*, Yiming Yang, Jaime Carbonell, Quoc V. Le, Ruslan Salakhutdinov. 1. **[TrOCR](https://huggingface.co/docs/transformers/model_doc/trocr)** (from Microsoft), released together with the paper [TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models](https://arxiv.org/abs/2109.10282) by Minghao Li, Tengchao Lv, Lei Cui, Yijuan Lu, Dinei Florencio, Cha Zhang, Zhoujun Li, Furu Wei. @@ -388,12 +388,12 @@ Current number of checkpoints: ![](https://img.shields.io/endpoint?url=https://h 1. **[Vision Transformer (ViT)](https://huggingface.co/docs/transformers/model_doc/vit)** (from Google AI) released with the paper [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://arxiv.org/abs/2010.11929) by Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, Neil Houlsby. 1. **[VisualBERT](https://huggingface.co/docs/transformers/model_doc/visual_bert)** (from UCLA NLP) released with the paper [VisualBERT: A Simple and Performant Baseline for Vision and Language](https://arxiv.org/pdf/1908.03557) by Liunian Harold Li, Mark Yatskar, Da Yin, Cho-Jui Hsieh, Kai-Wei Chang. 1. **[ViTMAE](https://huggingface.co/docs/transformers/model_doc/vit_mae)** (from Meta AI) released with the paper [Masked Autoencoders Are Scalable Vision Learners](https://arxiv.org/abs/2111.06377) by Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Dollár, Ross Girshick. -1. **[ViTMSN](https://huggingface.co/docs/transformers/main/model_doc/vit_msn)** (from Meta AI) released with the paper [Masked Siamese Networks for Label-Efficient Learning](https://arxiv.org/abs/2204.07141) by Mahmoud Assran, Mathilde Caron, Ishan Misra, Piotr Bojanowski, Florian Bordes, Pascal Vincent, Armand Joulin, Michael Rabbat, Nicolas Ballas. +1. **[ViTMSN](https://huggingface.co/docs/transformers/model_doc/vit_msn)** (from Meta AI) released with the paper [Masked Siamese Networks for Label-Efficient Learning](https://arxiv.org/abs/2204.07141) by Mahmoud Assran, Mathilde Caron, Ishan Misra, Piotr Bojanowski, Florian Bordes, Pascal Vincent, Armand Joulin, Michael Rabbat, Nicolas Ballas. 1. **[Wav2Vec2](https://huggingface.co/docs/transformers/model_doc/wav2vec2)** (from Facebook AI) released with the paper [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://arxiv.org/abs/2006.11477) by Alexei Baevski, Henry Zhou, Abdelrahman Mohamed, Michael Auli. 1. **[Wav2Vec2-Conformer](https://huggingface.co/docs/transformers/model_doc/wav2vec2-conformer)** (from Facebook AI) released with the paper [FAIRSEQ S2T: Fast Speech-to-Text Modeling with FAIRSEQ](https://arxiv.org/abs/2010.05171) by Changhan Wang, Yun Tang, Xutai Ma, Anne Wu, Sravya Popuri, Dmytro Okhonko, Juan Pino. 1. **[Wav2Vec2Phoneme](https://huggingface.co/docs/transformers/model_doc/wav2vec2_phoneme)** (from Facebook AI) released with the paper [Simple and Effective Zero-shot Cross-lingual Phoneme Recognition](https://arxiv.org/abs/2109.11680) by Qiantong Xu, Alexei Baevski, Michael Auli. 1. **[WavLM](https://huggingface.co/docs/transformers/model_doc/wavlm)** (from Microsoft Research) released with the paper [WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://arxiv.org/abs/2110.13900) by Sanyuan Chen, Chengyi Wang, Zhengyang Chen, Yu Wu, Shujie Liu, Zhuo Chen, Jinyu Li, Naoyuki Kanda, Takuya Yoshioka, Xiong Xiao, Jian Wu, Long Zhou, Shuo Ren, Yanmin Qian, Yao Qian, Jian Wu, Michael Zeng, Furu Wei. -1. **[Whisper](https://huggingface.co/docs/transformers/main/model_doc/whisper)** (from OpenAI) released with the paper [Robust Speech Recognition via Large-Scale Weak Supervision](https://cdn.openai.com/papers/whisper.pdf) by Alec Radford, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey, Ilya Sutskever. +1. **[Whisper](https://huggingface.co/docs/transformers/model_doc/whisper)** (from OpenAI) released with the paper [Robust Speech Recognition via Large-Scale Weak Supervision](https://cdn.openai.com/papers/whisper.pdf) by Alec Radford, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey, Ilya Sutskever. 1. **[X-CLIP](https://huggingface.co/docs/transformers/model_doc/xclip)** (from Microsoft Research) released with the paper [Expanding Language-Image Pretrained Models for General Video Recognition](https://arxiv.org/abs/2208.02816) by Bolin Ni, Houwen Peng, Minghao Chen, Songyang Zhang, Gaofeng Meng, Jianlong Fu, Shiming Xiang, Haibin Ling. 1. **[XGLM](https://huggingface.co/docs/transformers/model_doc/xglm)** (From Facebook AI) released with the paper [Few-shot Learning with Multilingual Language Models](https://arxiv.org/abs/2112.10668) by Xi Victoria Lin, Todor Mihaylov, Mikel Artetxe, Tianlu Wang, Shuohui Chen, Daniel Simig, Myle Ott, Naman Goyal, Shruti Bhosale, Jingfei Du, Ramakanth Pasunuru, Sam Shleifer, Punit Singh Koura, Vishrav Chaudhary, Brian O'Horo, Jeff Wang, Luke Zettlemoyer, Zornitsa Kozareva, Mona Diab, Veselin Stoyanov, Xian Li. 1. **[XLM](https://huggingface.co/docs/transformers/model_doc/xlm)** (from Facebook) released together with the paper [Cross-lingual Language Model Pretraining](https://arxiv.org/abs/1901.07291) by Guillaume Lample and Alexis Conneau. diff --git a/README_ko.md b/README_ko.md index 7d72b9f6a3ac92..3453ce57b1bf32 100644 --- a/README_ko.md +++ b/README_ko.md @@ -228,7 +228,7 @@ Flax, PyTorch, TensorFlow 설치 페이지에서 이들을 conda로 설치하는 1. **[CANINE](https://huggingface.co/docs/transformers/model_doc/canine)** (from Google Research) released with the paper [CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representation](https://arxiv.org/abs/2103.06874) by Jonathan H. Clark, Dan Garrette, Iulia Turc, John Wieting. 1. **[CLIP](https://huggingface.co/docs/transformers/model_doc/clip)** (from OpenAI) released with the paper [Learning Transferable Visual Models From Natural Language Supervision](https://arxiv.org/abs/2103.00020) by Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, Ilya Sutskever. 1. **[CodeGen](https://huggingface.co/docs/transformers/model_doc/codegen)** (from Salesforce) released with the paper [A Conversational Paradigm for Program Synthesis](https://arxiv.org/abs/2203.13474) by Erik Nijkamp, Bo Pang, Hiroaki Hayashi, Lifu Tu, Huan Wang, Yingbo Zhou, Silvio Savarese, Caiming Xiong. -1. **[Conditional DETR](https://huggingface.co/docs/transformers/main/model_doc/conditional_detr)** (from Microsoft Research Asia) released with the paper [Conditional DETR for Fast Training Convergence](https://arxiv.org/abs/2108.06152) by Depu Meng, Xiaokang Chen, Zejia Fan, Gang Zeng, Houqiang Li, Yuhui Yuan, Lei Sun, Jingdong Wang. +1. **[Conditional DETR](https://huggingface.co/docs/transformers/model_doc/conditional_detr)** (from Microsoft Research Asia) released with the paper [Conditional DETR for Fast Training Convergence](https://arxiv.org/abs/2108.06152) by Depu Meng, Xiaokang Chen, Zejia Fan, Gang Zeng, Houqiang Li, Yuhui Yuan, Lei Sun, Jingdong Wang. 1. **[ConvBERT](https://huggingface.co/docs/transformers/model_doc/convbert)** (from YituTech) released with the paper [ConvBERT: Improving BERT with Span-based Dynamic Convolution](https://arxiv.org/abs/2008.02496) by Zihang Jiang, Weihao Yu, Daquan Zhou, Yunpeng Chen, Jiashi Feng, Shuicheng Yan. 1. **[ConvNeXT](https://huggingface.co/docs/transformers/model_doc/convnext)** (from Facebook AI) released with the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Zhuang Liu, Hanzi Mao, Chao-Yuan Wu, Christoph Feichtenhofer, Trevor Darrell, Saining Xie. 1. **[CPM](https://huggingface.co/docs/transformers/model_doc/cpm)** (from Tsinghua University) released with the paper [CPM: A Large-scale Generative Chinese Pre-trained Language Model](https://arxiv.org/abs/2012.00413) by Zhengyan Zhang, Xu Han, Hao Zhou, Pei Ke, Yuxian Gu, Deming Ye, Yujia Qin, Yusheng Su, Haozhe Ji, Jian Guan, Fanchao Qi, Xiaozhi Wang, Yanan Zheng, Guoyang Zeng, Huanqi Cao, Shengqi Chen, Daixuan Li, Zhenbo Sun, Zhiyuan Liu, Minlie Huang, Wentao Han, Jie Tang, Juanzi Li, Xiaoyan Zhu, Maosong Sun. @@ -250,7 +250,7 @@ Flax, PyTorch, TensorFlow 설치 페이지에서 이들을 conda로 설치하는 1. **[ELECTRA](https://huggingface.co/docs/transformers/model_doc/electra)** (from Google Research/Stanford University) released with the paper [ELECTRA: Pre-training text encoders as discriminators rather than generators](https://arxiv.org/abs/2003.10555) by Kevin Clark, Minh-Thang Luong, Quoc V. Le, Christopher D. Manning. 1. **[EncoderDecoder](https://huggingface.co/docs/transformers/model_doc/encoder-decoder)** (from Google Research) released with the paper [Leveraging Pre-trained Checkpoints for Sequence Generation Tasks](https://arxiv.org/abs/1907.12461) by Sascha Rothe, Shashi Narayan, Aliaksei Severyn. 1. **[ERNIE](https://huggingface.co/docs/transformers/model_doc/ernie)** (from Baidu) released with the paper [ERNIE: Enhanced Representation through Knowledge Integration](https://arxiv.org/abs/1904.09223) by Yu Sun, Shuohuan Wang, Yukun Li, Shikun Feng, Xuyi Chen, Han Zhang, Xin Tian, Danxiang Zhu, Hao Tian, Hua Wu. -1. **[ESM](https://huggingface.co/docs/transformers/main/model_doc/esm)** (from Meta AI) are transformer protein language models. **ESM-1b** was released with the paper [Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences](https://www.pnas.org/content/118/15/e2016239118) by Alexander Rives, Joshua Meier, Tom Sercu, Siddharth Goyal, Zeming Lin, Jason Liu, Demi Guo, Myle Ott, C. Lawrence Zitnick, Jerry Ma, and Rob Fergus. **ESM-1v** was released with the paper [Language models enable zero-shot prediction of the effects of mutations on protein function](https://doi.org/10.1101/2021.07.09.450648) by Joshua Meier, Roshan Rao, Robert Verkuil, Jason Liu, Tom Sercu and Alexander Rives. **ESM-2** was released with the paper [Language models of protein sequences at the scale of evolution enable accurate structure prediction](https://doi.org/10.1101/2022.07.20.500902) by Zeming Lin, Halil Akin, Roshan Rao, Brian Hie, Zhongkai Zhu, Wenting Lu, Allan dos Santos Costa, Maryam Fazel-Zarandi, Tom Sercu, Sal Candido, Alexander Rives. +1. **[ESM](https://huggingface.co/docs/transformers/model_doc/esm)** (from Meta AI) are transformer protein language models. **ESM-1b** was released with the paper [Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences](https://www.pnas.org/content/118/15/e2016239118) by Alexander Rives, Joshua Meier, Tom Sercu, Siddharth Goyal, Zeming Lin, Jason Liu, Demi Guo, Myle Ott, C. Lawrence Zitnick, Jerry Ma, and Rob Fergus. **ESM-1v** was released with the paper [Language models enable zero-shot prediction of the effects of mutations on protein function](https://doi.org/10.1101/2021.07.09.450648) by Joshua Meier, Roshan Rao, Robert Verkuil, Jason Liu, Tom Sercu and Alexander Rives. **ESM-2** was released with the paper [Language models of protein sequences at the scale of evolution enable accurate structure prediction](https://doi.org/10.1101/2022.07.20.500902) by Zeming Lin, Halil Akin, Roshan Rao, Brian Hie, Zhongkai Zhu, Wenting Lu, Allan dos Santos Costa, Maryam Fazel-Zarandi, Tom Sercu, Sal Candido, Alexander Rives. 1. **[FlauBERT](https://huggingface.co/docs/transformers/model_doc/flaubert)** (from CNRS) released with the paper [FlauBERT: Unsupervised Language Model Pre-training for French](https://arxiv.org/abs/1912.05372) by Hang Le, Loïc Vial, Jibril Frej, Vincent Segonne, Maximin Coavoux, Benjamin Lecouteux, Alexandre Allauzen, Benoît Crabbé, Laurent Besacier, Didier Schwab. 1. **[FLAVA](https://huggingface.co/docs/transformers/model_doc/flava)** (from Facebook AI) released with the paper [FLAVA: A Foundational Language And Vision Alignment Model](https://arxiv.org/abs/2112.04482) by Amanpreet Singh, Ronghang Hu, Vedanuj Goswami, Guillaume Couairon, Wojciech Galuba, Marcus Rohrbach, and Douwe Kiela. 1. **[FNet](https://huggingface.co/docs/transformers/model_doc/fnet)** (from Google Research) released with the paper [FNet: Mixing Tokens with Fourier Transforms](https://arxiv.org/abs/2105.03824) by James Lee-Thorp, Joshua Ainslie, Ilya Eckstein, Santiago Ontanon. @@ -279,7 +279,7 @@ Flax, PyTorch, TensorFlow 설치 페이지에서 이들을 conda로 설치하는 1. **[M-CTC-T](https://huggingface.co/docs/transformers/model_doc/mctct)** (from Facebook) released with the paper [Pseudo-Labeling For Massively Multilingual Speech Recognition](https://arxiv.org/abs/2111.00161) by Loren Lugosch, Tatiana Likhomanenko, Gabriel Synnaeve, and Ronan Collobert. 1. **[M2M100](https://huggingface.co/docs/transformers/model_doc/m2m_100)** (from Facebook) released with the paper [Beyond English-Centric Multilingual Machine Translation](https://arxiv.org/abs/2010.11125) by Angela Fan, Shruti Bhosale, Holger Schwenk, Zhiyi Ma, Ahmed El-Kishky, Siddharth Goyal, Mandeep Baines, Onur Celebi, Guillaume Wenzek, Vishrav Chaudhary, Naman Goyal, Tom Birch, Vitaliy Liptchinsky, Sergey Edunov, Edouard Grave, Michael Auli, Armand Joulin. 1. **[MarianMT](https://huggingface.co/docs/transformers/model_doc/marian)** Machine translation models trained using [OPUS](http://opus.nlpl.eu/) data by Jörg Tiedemann. The [Marian Framework](https://marian-nmt.github.io/) is being developed by the Microsoft Translator Team. -1. **[MarkupLM](https://huggingface.co/docs/transformers/main/model_doc/markuplm)** (from Microsoft Research Asia) released with the paper [MarkupLM: Pre-training of Text and Markup Language for Visually-rich Document Understanding](https://arxiv.org/abs/2110.08518) by Junlong Li, Yiheng Xu, Lei Cui, Furu Wei. +1. **[MarkupLM](https://huggingface.co/docs/transformers/model_doc/markuplm)** (from Microsoft Research Asia) released with the paper [MarkupLM: Pre-training of Text and Markup Language for Visually-rich Document Understanding](https://arxiv.org/abs/2110.08518) by Junlong Li, Yiheng Xu, Lei Cui, Furu Wei. 1. **[MaskFormer](https://huggingface.co/docs/transformers/model_doc/maskformer)** (from Meta and UIUC) released with the paper [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://arxiv.org/abs/2107.06278) by Bowen Cheng, Alexander G. Schwing, Alexander Kirillov. 1. **[mBART](https://huggingface.co/docs/transformers/model_doc/mbart)** (from Facebook) released with the paper [Multilingual Denoising Pre-training for Neural Machine Translation](https://arxiv.org/abs/2001.08210) by Yinhan Liu, Jiatao Gu, Naman Goyal, Xian Li, Sergey Edunov, Marjan Ghazvininejad, Mike Lewis, Luke Zettlemoyer. 1. **[mBART-50](https://huggingface.co/docs/transformers/model_doc/mbart)** (from Facebook) released with the paper [Multilingual Translation with Extensible Multilingual Pretraining and Finetuning](https://arxiv.org/abs/2008.00401) by Yuqing Tang, Chau Tran, Xian Li, Peng-Jen Chen, Naman Goyal, Vishrav Chaudhary, Jiatao Gu, Angela Fan. @@ -325,7 +325,7 @@ Flax, PyTorch, TensorFlow 설치 페이지에서 이들을 conda로 설치하는 1. **[T5v1.1](https://huggingface.co/docs/transformers/model_doc/t5v1.1)** (from Google AI) released in the repository [google-research/text-to-text-transfer-transformer](https://github.com/google-research/text-to-text-transfer-transformer/blob/main/released_checkpoints.md#t511) by Colin Raffel and Noam Shazeer and Adam Roberts and Katherine Lee and Sharan Narang and Michael Matena and Yanqi Zhou and Wei Li and Peter J. Liu. 1. **[TAPAS](https://huggingface.co/docs/transformers/model_doc/tapas)** (from Google AI) released with the paper [TAPAS: Weakly Supervised Table Parsing via Pre-training](https://arxiv.org/abs/2004.02349) by Jonathan Herzig, Paweł Krzysztof Nowak, Thomas Müller, Francesco Piccinno and Julian Martin Eisenschlos. 1. **[TAPEX](https://huggingface.co/docs/transformers/model_doc/tapex)** (from Microsoft Research) released with the paper [TAPEX: Table Pre-training via Learning a Neural SQL Executor](https://arxiv.org/abs/2107.07653) by Qian Liu, Bei Chen, Jiaqi Guo, Morteza Ziyadi, Zeqi Lin, Weizhu Chen, Jian-Guang Lou. -1. **[Time Series Transformer](https://huggingface.co/docs/transformers/main/model_doc/time_series_transformer)** (from HuggingFace). +1. **[Time Series Transformer](https://huggingface.co/docs/transformers/model_doc/time_series_transformer)** (from HuggingFace). 1. **[Trajectory Transformer](https://huggingface.co/docs/transformers/model_doc/trajectory_transformers)** (from the University of California at Berkeley) released with the paper [Offline Reinforcement Learning as One Big Sequence Modeling Problem](https://arxiv.org/abs/2106.02039) by Michael Janner, Qiyang Li, Sergey Levine 1. **[Transformer-XL](https://huggingface.co/docs/transformers/model_doc/transfo-xl)** (from Google/CMU) released with the paper [Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context](https://arxiv.org/abs/1901.02860) by Zihang Dai*, Zhilin Yang*, Yiming Yang, Jaime Carbonell, Quoc V. Le, Ruslan Salakhutdinov. 1. **[TrOCR](https://huggingface.co/docs/transformers/model_doc/trocr)** (from Microsoft), released together with the paper [TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models](https://arxiv.org/abs/2109.10282) by Minghao Li, Tengchao Lv, Lei Cui, Yijuan Lu, Dinei Florencio, Cha Zhang, Zhoujun Li, Furu Wei. @@ -338,12 +338,12 @@ Flax, PyTorch, TensorFlow 설치 페이지에서 이들을 conda로 설치하는 1. **[Vision Transformer (ViT)](https://huggingface.co/docs/transformers/model_doc/vit)** (from Google AI) released with the paper [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://arxiv.org/abs/2010.11929) by Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, Neil Houlsby. 1. **[VisualBERT](https://huggingface.co/docs/transformers/model_doc/visual_bert)** (from UCLA NLP) released with the paper [VisualBERT: A Simple and Performant Baseline for Vision and Language](https://arxiv.org/pdf/1908.03557) by Liunian Harold Li, Mark Yatskar, Da Yin, Cho-Jui Hsieh, Kai-Wei Chang. 1. **[ViTMAE](https://huggingface.co/docs/transformers/model_doc/vit_mae)** (from Meta AI) released with the paper [Masked Autoencoders Are Scalable Vision Learners](https://arxiv.org/abs/2111.06377) by Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Dollár, Ross Girshick. -1. **[ViTMSN](https://huggingface.co/docs/transformers/main/model_doc/vit_msn)** (from Meta AI) released with the paper [Masked Siamese Networks for Label-Efficient Learning](https://arxiv.org/abs/2204.07141) by Mahmoud Assran, Mathilde Caron, Ishan Misra, Piotr Bojanowski, Florian Bordes, Pascal Vincent, Armand Joulin, Michael Rabbat, Nicolas Ballas. +1. **[ViTMSN](https://huggingface.co/docs/transformers/model_doc/vit_msn)** (from Meta AI) released with the paper [Masked Siamese Networks for Label-Efficient Learning](https://arxiv.org/abs/2204.07141) by Mahmoud Assran, Mathilde Caron, Ishan Misra, Piotr Bojanowski, Florian Bordes, Pascal Vincent, Armand Joulin, Michael Rabbat, Nicolas Ballas. 1. **[Wav2Vec2](https://huggingface.co/docs/transformers/model_doc/wav2vec2)** (from Facebook AI) released with the paper [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://arxiv.org/abs/2006.11477) by Alexei Baevski, Henry Zhou, Abdelrahman Mohamed, Michael Auli. 1. **[Wav2Vec2-Conformer](https://huggingface.co/docs/transformers/model_doc/wav2vec2-conformer)** (from Facebook AI) released with the paper [FAIRSEQ S2T: Fast Speech-to-Text Modeling with FAIRSEQ](https://arxiv.org/abs/2010.05171) by Changhan Wang, Yun Tang, Xutai Ma, Anne Wu, Sravya Popuri, Dmytro Okhonko, Juan Pino. 1. **[Wav2Vec2Phoneme](https://huggingface.co/docs/transformers/model_doc/wav2vec2_phoneme)** (from Facebook AI) released with the paper [Simple and Effective Zero-shot Cross-lingual Phoneme Recognition](https://arxiv.org/abs/2109.11680) by Qiantong Xu, Alexei Baevski, Michael Auli. 1. **[WavLM](https://huggingface.co/docs/transformers/model_doc/wavlm)** (from Microsoft Research) released with the paper [WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://arxiv.org/abs/2110.13900) by Sanyuan Chen, Chengyi Wang, Zhengyang Chen, Yu Wu, Shujie Liu, Zhuo Chen, Jinyu Li, Naoyuki Kanda, Takuya Yoshioka, Xiong Xiao, Jian Wu, Long Zhou, Shuo Ren, Yanmin Qian, Yao Qian, Jian Wu, Michael Zeng, Furu Wei. -1. **[Whisper](https://huggingface.co/docs/transformers/main/model_doc/whisper)** (from OpenAI) released with the paper [Robust Speech Recognition via Large-Scale Weak Supervision](https://cdn.openai.com/papers/whisper.pdf) by Alec Radford, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey, Ilya Sutskever. +1. **[Whisper](https://huggingface.co/docs/transformers/model_doc/whisper)** (from OpenAI) released with the paper [Robust Speech Recognition via Large-Scale Weak Supervision](https://cdn.openai.com/papers/whisper.pdf) by Alec Radford, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey, Ilya Sutskever. 1. **[X-CLIP](https://huggingface.co/docs/transformers/model_doc/xclip)** (from Microsoft Research) released with the paper [Expanding Language-Image Pretrained Models for General Video Recognition](https://arxiv.org/abs/2208.02816) by Bolin Ni, Houwen Peng, Minghao Chen, Songyang Zhang, Gaofeng Meng, Jianlong Fu, Shiming Xiang, Haibin Ling. 1. **[XGLM](https://huggingface.co/docs/transformers/model_doc/xglm)** (From Facebook AI) released with the paper [Few-shot Learning with Multilingual Language Models](https://arxiv.org/abs/2112.10668) by Xi Victoria Lin, Todor Mihaylov, Mikel Artetxe, Tianlu Wang, Shuohui Chen, Daniel Simig, Myle Ott, Naman Goyal, Shruti Bhosale, Jingfei Du, Ramakanth Pasunuru, Sam Shleifer, Punit Singh Koura, Vishrav Chaudhary, Brian O'Horo, Jeff Wang, Luke Zettlemoyer, Zornitsa Kozareva, Mona Diab, Veselin Stoyanov, Xian Li. 1. **[XLM](https://huggingface.co/docs/transformers/model_doc/xlm)** (from Facebook) released together with the paper [Cross-lingual Language Model Pretraining](https://arxiv.org/abs/1901.07291) by Guillaume Lample and Alexis Conneau. diff --git a/README_zh-hans.md b/README_zh-hans.md index 474ba18d58319d..6b9d5168feace9 100644 --- a/README_zh-hans.md +++ b/README_zh-hans.md @@ -252,7 +252,7 @@ conda install -c huggingface transformers 1. **[CANINE](https://huggingface.co/docs/transformers/model_doc/canine)** (来自 Google Research) 伴随论文 [CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representation](https://arxiv.org/abs/2103.06874) 由 Jonathan H. Clark, Dan Garrette, Iulia Turc, John Wieting 发布。 1. **[CLIP](https://huggingface.co/docs/transformers/model_doc/clip)** (来自 OpenAI) 伴随论文 [Learning Transferable Visual Models From Natural Language Supervision](https://arxiv.org/abs/2103.00020) 由 Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, Ilya Sutskever 发布。 1. **[CodeGen](https://huggingface.co/docs/transformers/model_doc/codegen)** (来自 Salesforce) 伴随论文 [A Conversational Paradigm for Program Synthesis](https://arxiv.org/abs/2203.13474) 由 Erik Nijkamp, Bo Pang, Hiroaki Hayashi, Lifu Tu, Huan Wang, Yingbo Zhou, Silvio Savarese, Caiming Xiong 发布。 -1. **[Conditional DETR](https://huggingface.co/docs/transformers/main/model_doc/conditional_detr)** (来自 Microsoft Research Asia) 伴随论文 [Conditional DETR for Fast Training Convergence](https://arxiv.org/abs/2108.06152) 由 Depu Meng, Xiaokang Chen, Zejia Fan, Gang Zeng, Houqiang Li, Yuhui Yuan, Lei Sun, Jingdong Wang 发布。 +1. **[Conditional DETR](https://huggingface.co/docs/transformers/model_doc/conditional_detr)** (来自 Microsoft Research Asia) 伴随论文 [Conditional DETR for Fast Training Convergence](https://arxiv.org/abs/2108.06152) 由 Depu Meng, Xiaokang Chen, Zejia Fan, Gang Zeng, Houqiang Li, Yuhui Yuan, Lei Sun, Jingdong Wang 发布。 1. **[ConvBERT](https://huggingface.co/docs/transformers/model_doc/convbert)** (来自 YituTech) 伴随论文 [ConvBERT: Improving BERT with Span-based Dynamic Convolution](https://arxiv.org/abs/2008.02496) 由 Zihang Jiang, Weihao Yu, Daquan Zhou, Yunpeng Chen, Jiashi Feng, Shuicheng Yan 发布。 1. **[ConvNeXT](https://huggingface.co/docs/transformers/model_doc/convnext)** (来自 Facebook AI) 伴随论文 [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) 由 Zhuang Liu, Hanzi Mao, Chao-Yuan Wu, Christoph Feichtenhofer, Trevor Darrell, Saining Xie 发布。 1. **[CPM](https://huggingface.co/docs/transformers/model_doc/cpm)** (来自 Tsinghua University) 伴随论文 [CPM: A Large-scale Generative Chinese Pre-trained Language Model](https://arxiv.org/abs/2012.00413) 由 Zhengyan Zhang, Xu Han, Hao Zhou, Pei Ke, Yuxian Gu, Deming Ye, Yujia Qin, Yusheng Su, Haozhe Ji, Jian Guan, Fanchao Qi, Xiaozhi Wang, Yanan Zheng, Guoyang Zeng, Huanqi Cao, Shengqi Chen, Daixuan Li, Zhenbo Sun, Zhiyuan Liu, Minlie Huang, Wentao Han, Jie Tang, Juanzi Li, Xiaoyan Zhu, Maosong Sun 发布。 @@ -274,7 +274,7 @@ conda install -c huggingface transformers 1. **[ELECTRA](https://huggingface.co/docs/transformers/model_doc/electra)** (来自 Google Research/Stanford University) 伴随论文 [ELECTRA: Pre-training text encoders as discriminators rather than generators](https://arxiv.org/abs/2003.10555) 由 Kevin Clark, Minh-Thang Luong, Quoc V. Le, Christopher D. Manning 发布。 1. **[EncoderDecoder](https://huggingface.co/docs/transformers/model_doc/encoder-decoder)** (来自 Google Research) 伴随论文 [Leveraging Pre-trained Checkpoints for Sequence Generation Tasks](https://arxiv.org/abs/1907.12461) 由 Sascha Rothe, Shashi Narayan, Aliaksei Severyn 发布。 1. **[ERNIE](https://huggingface.co/docs/transformers/model_doc/ernie)** (来自 Baidu) 伴随论文 [ERNIE: Enhanced Representation through Knowledge Integration](https://arxiv.org/abs/1904.09223) by Yu Sun, Shuohuan Wang, Yukun Li, Shikun Feng, Xuyi Chen, Han Zhang, Xin Tian, Danxiang Zhu, Hao Tian, Hua Wu 发布。 -1. **[ESM](https://huggingface.co/docs/transformers/main/model_doc/esm)** (from Meta AI) are transformer protein language models. **ESM-1b** was released with the paper [Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences](https://www.pnas.org/content/118/15/e2016239118) by Alexander Rives, Joshua Meier, Tom Sercu, Siddharth Goyal, Zeming Lin, Jason Liu, Demi Guo, Myle Ott, C. Lawrence Zitnick, Jerry Ma, and Rob Fergus. **ESM-1v** was released with the paper [Language models enable zero-shot prediction of the effects of mutations on protein function](https://doi.org/10.1101/2021.07.09.450648) by Joshua Meier, Roshan Rao, Robert Verkuil, Jason Liu, Tom Sercu and Alexander Rives. **ESM-2** was released with the paper [Language models of protein sequences at the scale of evolution enable accurate structure prediction](https://doi.org/10.1101/2022.07.20.500902) by Zeming Lin, Halil Akin, Roshan Rao, Brian Hie, Zhongkai Zhu, Wenting Lu, Allan dos Santos Costa, Maryam Fazel-Zarandi, Tom Sercu, Sal Candido, Alexander Rives. +1. **[ESM](https://huggingface.co/docs/transformers/model_doc/esm)** (from Meta AI) are transformer protein language models. **ESM-1b** was released with the paper [Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences](https://www.pnas.org/content/118/15/e2016239118) by Alexander Rives, Joshua Meier, Tom Sercu, Siddharth Goyal, Zeming Lin, Jason Liu, Demi Guo, Myle Ott, C. Lawrence Zitnick, Jerry Ma, and Rob Fergus. **ESM-1v** was released with the paper [Language models enable zero-shot prediction of the effects of mutations on protein function](https://doi.org/10.1101/2021.07.09.450648) by Joshua Meier, Roshan Rao, Robert Verkuil, Jason Liu, Tom Sercu and Alexander Rives. **ESM-2** was released with the paper [Language models of protein sequences at the scale of evolution enable accurate structure prediction](https://doi.org/10.1101/2022.07.20.500902) by Zeming Lin, Halil Akin, Roshan Rao, Brian Hie, Zhongkai Zhu, Wenting Lu, Allan dos Santos Costa, Maryam Fazel-Zarandi, Tom Sercu, Sal Candido, Alexander Rives. 1. **[FlauBERT](https://huggingface.co/docs/transformers/model_doc/flaubert)** (来自 CNRS) 伴随论文 [FlauBERT: Unsupervised Language Model Pre-training for French](https://arxiv.org/abs/1912.05372) 由 Hang Le, Loïc Vial, Jibril Frej, Vincent Segonne, Maximin Coavoux, Benjamin Lecouteux, Alexandre Allauzen, Benoît Crabbé, Laurent Besacier, Didier Schwab 发布。 1. **[FLAVA](https://huggingface.co/docs/transformers/model_doc/flava)** (来自 Facebook AI) 伴随论文 [FLAVA: A Foundational Language And Vision Alignment Model](https://arxiv.org/abs/2112.04482) 由 Amanpreet Singh, Ronghang Hu, Vedanuj Goswami, Guillaume Couairon, Wojciech Galuba, Marcus Rohrbach, and Douwe Kiela 发布。 1. **[FNet](https://huggingface.co/docs/transformers/model_doc/fnet)** (来自 Google Research) 伴随论文 [FNet: Mixing Tokens with Fourier Transforms](https://arxiv.org/abs/2105.03824) 由 James Lee-Thorp, Joshua Ainslie, Ilya Eckstein, Santiago Ontanon 发布。 @@ -303,7 +303,7 @@ conda install -c huggingface transformers 1. **[M-CTC-T](https://huggingface.co/docs/transformers/model_doc/mctct)** (来自 Facebook) 伴随论文 [Pseudo-Labeling For Massively Multilingual Speech Recognition](https://arxiv.org/abs/2111.00161) 由 Loren Lugosch, Tatiana Likhomanenko, Gabriel Synnaeve, and Ronan Collobert 发布。 1. **[M2M100](https://huggingface.co/docs/transformers/model_doc/m2m_100)** (来自 Facebook) 伴随论文 [Beyond English-Centric Multilingual Machine Translation](https://arxiv.org/abs/2010.11125) 由 Angela Fan, Shruti Bhosale, Holger Schwenk, Zhiyi Ma, Ahmed El-Kishky, Siddharth Goyal, Mandeep Baines, Onur Celebi, Guillaume Wenzek, Vishrav Chaudhary, Naman Goyal, Tom Birch, Vitaliy Liptchinsky, Sergey Edunov, Edouard Grave, Michael Auli, Armand Joulin 发布。 1. **[MarianMT](https://huggingface.co/docs/transformers/model_doc/marian)** 用 [OPUS](http://opus.nlpl.eu/) 数据训练的机器翻译模型由 Jörg Tiedemann 发布。[Marian Framework](https://marian-nmt.github.io/) 由微软翻译团队开发。 -1. **[MarkupLM](https://huggingface.co/docs/transformers/main/model_doc/markuplm)** (来自 Microsoft Research Asia) 伴随论文 [MarkupLM: Pre-training of Text and Markup Language for Visually-rich Document Understanding](https://arxiv.org/abs/2110.08518) 由 Junlong Li, Yiheng Xu, Lei Cui, Furu Wei 发布。 +1. **[MarkupLM](https://huggingface.co/docs/transformers/model_doc/markuplm)** (来自 Microsoft Research Asia) 伴随论文 [MarkupLM: Pre-training of Text and Markup Language for Visually-rich Document Understanding](https://arxiv.org/abs/2110.08518) 由 Junlong Li, Yiheng Xu, Lei Cui, Furu Wei 发布。 1. **[MaskFormer](https://huggingface.co/docs/transformers/model_doc/maskformer)** (from Meta and UIUC) released with the paper [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://arxiv.org/abs/2107.06278) by Bowen Cheng, Alexander G. Schwing, Alexander Kirillov >>>>>>> Fix rebase 1. **[mBART](https://huggingface.co/docs/transformers/model_doc/mbart)** (来自 Facebook) 伴随论文 [Multilingual Denoising Pre-training for Neural Machine Translation](https://arxiv.org/abs/2001.08210) 由 Yinhan Liu, Jiatao Gu, Naman Goyal, Xian Li, Sergey Edunov, Marjan Ghazvininejad, Mike Lewis, Luke Zettlemoyer 发布。 1. **[mBART-50](https://huggingface.co/docs/transformers/model_doc/mbart)** (来自 Facebook) 伴随论文 [Multilingual Translation with Extensible Multilingual Pretraining and Finetuning](https://arxiv.org/abs/2008.00401) 由 Yuqing Tang, Chau Tran, Xian Li, Peng-Jen Chen, Naman Goyal, Vishrav Chaudhary, Jiatao Gu, Angela Fan 发布。 @@ -349,7 +349,7 @@ conda install -c huggingface transformers 1. **[T5v1.1](https://huggingface.co/docs/transformers/model_doc/t5v1.1)** (来自 Google AI) 伴随论文 [google-research/text-to-text-transfer-transformer](https://github.com/google-research/text-to-text-transfer-transformer/blob/main/released_checkpoints.md#t511) 由 Colin Raffel and Noam Shazeer and Adam Roberts and Katherine Lee and Sharan Narang and Michael Matena and Yanqi Zhou and Wei Li and Peter J. Liu 发布。 1. **[TAPAS](https://huggingface.co/docs/transformers/model_doc/tapas)** (来自 Google AI) 伴随论文 [TAPAS: Weakly Supervised Table Parsing via Pre-training](https://arxiv.org/abs/2004.02349) 由 Jonathan Herzig, Paweł Krzysztof Nowak, Thomas Müller, Francesco Piccinno and Julian Martin Eisenschlos 发布。 1. **[TAPEX](https://huggingface.co/docs/transformers/model_doc/tapex)** (来自 Microsoft Research) 伴随论文 [TAPEX: Table Pre-training via Learning a Neural SQL Executor](https://arxiv.org/abs/2107.07653) 由 Qian Liu, Bei Chen, Jiaqi Guo, Morteza Ziyadi, Zeqi Lin, Weizhu Chen, Jian-Guang Lou 发布。 -1. **[Time Series Transformer](https://huggingface.co/docs/transformers/main/model_doc/time_series_transformer)** (from HuggingFace). +1. **[Time Series Transformer](https://huggingface.co/docs/transformers/model_doc/time_series_transformer)** (from HuggingFace). 1. **[Trajectory Transformer](https://huggingface.co/docs/transformers/model_doc/trajectory_transformers)** (from the University of California at Berkeley) released with the paper [Offline Reinforcement Learning as One Big Sequence Modeling Problem](https://arxiv.org/abs/2106.02039) by Michael Janner, Qiyang Li, Sergey Levine 1. **[Transformer-XL](https://huggingface.co/docs/transformers/model_doc/transfo-xl)** (来自 Google/CMU) 伴随论文 [Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context](https://arxiv.org/abs/1901.02860) 由 Zihang Dai*, Zhilin Yang*, Yiming Yang, Jaime Carbonell, Quoc V. Le, Ruslan Salakhutdinov 发布。 1. **[TrOCR](https://huggingface.co/docs/transformers/model_doc/trocr)** (来自 Microsoft) 伴随论文 [TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models](https://arxiv.org/abs/2109.10282) 由 Minghao Li, Tengchao Lv, Lei Cui, Yijuan Lu, Dinei Florencio, Cha Zhang, Zhoujun Li, Furu Wei 发布。 @@ -362,12 +362,12 @@ conda install -c huggingface transformers 1. **[Vision Transformer (ViT)](https://huggingface.co/docs/transformers/model_doc/vit)** (来自 Google AI) 伴随论文 [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://arxiv.org/abs/2010.11929) 由 Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, Neil Houlsby 发布。 1. **[VisualBERT](https://huggingface.co/docs/transformers/model_doc/visual_bert)** (来自 UCLA NLP) 伴随论文 [VisualBERT: A Simple and Performant Baseline for Vision and Language](https://arxiv.org/pdf/1908.03557) 由 Liunian Harold Li, Mark Yatskar, Da Yin, Cho-Jui Hsieh, Kai-Wei Chang 发布。 1. **[ViTMAE](https://huggingface.co/docs/transformers/model_doc/vit_mae)** (来自 Meta AI) 伴随论文 [Masked Autoencoders Are Scalable Vision Learners](https://arxiv.org/abs/2111.06377) 由 Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Dollár, Ross Girshick 发布。 -1. **[ViTMSN](https://huggingface.co/docs/transformers/main/model_doc/vit_msn)** (来自 Meta AI) 伴随论文 [Masked Siamese Networks for Label-Efficient Learning](https://arxiv.org/abs/2204.07141) by Mahmoud Assran, Mathilde Caron, Ishan Misra, Piotr Bojanowski, Florian Bordes, Pascal Vincent, Armand Joulin, Michael Rabbat, Nicolas Ballas 发布. +1. **[ViTMSN](https://huggingface.co/docs/transformers/model_doc/vit_msn)** (来自 Meta AI) 伴随论文 [Masked Siamese Networks for Label-Efficient Learning](https://arxiv.org/abs/2204.07141) by Mahmoud Assran, Mathilde Caron, Ishan Misra, Piotr Bojanowski, Florian Bordes, Pascal Vincent, Armand Joulin, Michael Rabbat, Nicolas Ballas 发布. 1. **[Wav2Vec2](https://huggingface.co/docs/transformers/model_doc/wav2vec2)** (来自 Facebook AI) 伴随论文 [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://arxiv.org/abs/2006.11477) 由 Alexei Baevski, Henry Zhou, Abdelrahman Mohamed, Michael Auli 发布。 1. **[Wav2Vec2-Conformer](https://huggingface.co/docs/transformers/model_doc/wav2vec2-conformer)** (来自 Facebook AI) 伴随论文 [FAIRSEQ S2T: Fast Speech-to-Text Modeling with FAIRSEQ](https://arxiv.org/abs/2010.05171) 由 Changhan Wang, Yun Tang, Xutai Ma, Anne Wu, Sravya Popuri, Dmytro Okhonko, Juan Pino 发布。 1. **[Wav2Vec2Phoneme](https://huggingface.co/docs/transformers/model_doc/wav2vec2_phoneme)** (来自 Facebook AI) 伴随论文 [Simple and Effective Zero-shot Cross-lingual Phoneme Recognition](https://arxiv.org/abs/2109.11680) 由 Qiantong Xu, Alexei Baevski, Michael Auli 发布。 1. **[WavLM](https://huggingface.co/docs/transformers/model_doc/wavlm)** (from Microsoft Research) released with the paper [WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://arxiv.org/abs/2110.13900) by Sanyuan Chen, Chengyi Wang, Zhengyang Chen, Yu Wu, Shujie Liu, Zhuo Chen, Jinyu Li, Naoyuki Kanda, Takuya Yoshioka, Xiong Xiao, Jian Wu, Long Zhou, Shuo Ren, Yanmin Qian, Yao Qian, Jian Wu, Michael Zeng, Furu Wei. -1. **[Whisper](https://huggingface.co/docs/transformers/main/model_doc/whisper)** (来自 OpenAI) 伴随论文 [Robust Speech Recognition via Large-Scale Weak Supervision](https://cdn.openai.com/papers/whisper.pdf) 由 Alec Radford, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey, Ilya Sutskever 发布。 +1. **[Whisper](https://huggingface.co/docs/transformers/model_doc/whisper)** (来自 OpenAI) 伴随论文 [Robust Speech Recognition via Large-Scale Weak Supervision](https://cdn.openai.com/papers/whisper.pdf) 由 Alec Radford, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey, Ilya Sutskever 发布。 1. **[X-CLIP](https://huggingface.co/docs/transformers/model_doc/xclip)** (来自 Microsoft Research) 伴随论文 [Expanding Language-Image Pretrained Models for General Video Recognition](https://arxiv.org/abs/2208.02816) 由 Bolin Ni, Houwen Peng, Minghao Chen, Songyang Zhang, Gaofeng Meng, Jianlong Fu, Shiming Xiang, Haibin Ling 发布。 1. **[XGLM](https://huggingface.co/docs/transformers/model_doc/xglm)** (From Facebook AI) released with the paper [Few-shot Learning with Multilingual Language Models](https://arxiv.org/abs/2112.10668) by Xi Victoria Lin, Todor Mihaylov, Mikel Artetxe, Tianlu Wang, Shuohui Chen, Daniel Simig, Myle Ott, Naman Goyal, Shruti Bhosale, Jingfei Du, Ramakanth Pasunuru, Sam Shleifer, Punit Singh Koura, Vishrav Chaudhary, Brian O'Horo, Jeff Wang, Luke Zettlemoyer, Zornitsa Kozareva, Mona Diab, Veselin Stoyanov, Xian Li. 1. **[XLM](https://huggingface.co/docs/transformers/model_doc/xlm)** (来自 Facebook) 伴随论文 [Cross-lingual Language Model Pretraining](https://arxiv.org/abs/1901.07291) 由 Guillaume Lample and Alexis Conneau 发布。 diff --git a/README_zh-hant.md b/README_zh-hant.md index ebc75e23adb6fb..bcf13d8ea47722 100644 --- a/README_zh-hant.md +++ b/README_zh-hant.md @@ -264,7 +264,7 @@ conda install -c huggingface transformers 1. **[CANINE](https://huggingface.co/docs/transformers/model_doc/canine)** (from Google Research) released with the paper [CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representation](https://arxiv.org/abs/2103.06874) by Jonathan H. Clark, Dan Garrette, Iulia Turc, John Wieting. 1. **[CLIP](https://huggingface.co/docs/transformers/model_doc/clip)** (from OpenAI) released with the paper [Learning Transferable Visual Models From Natural Language Supervision](https://arxiv.org/abs/2103.00020) by Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, Ilya Sutskever. 1. **[CodeGen](https://huggingface.co/docs/transformers/model_doc/codegen)** (from Salesforce) released with the paper [A Conversational Paradigm for Program Synthesis](https://arxiv.org/abs/2203.13474) by Erik Nijkamp, Bo Pang, Hiroaki Hayashi, Lifu Tu, Huan Wang, Yingbo Zhou, Silvio Savarese, Caiming Xiong. -1. **[Conditional DETR](https://huggingface.co/docs/transformers/main/model_doc/conditional_detr)** (from Microsoft Research Asia) released with the paper [Conditional DETR for Fast Training Convergence](https://arxiv.org/abs/2108.06152) by Depu Meng, Xiaokang Chen, Zejia Fan, Gang Zeng, Houqiang Li, Yuhui Yuan, Lei Sun, Jingdong Wang. +1. **[Conditional DETR](https://huggingface.co/docs/transformers/model_doc/conditional_detr)** (from Microsoft Research Asia) released with the paper [Conditional DETR for Fast Training Convergence](https://arxiv.org/abs/2108.06152) by Depu Meng, Xiaokang Chen, Zejia Fan, Gang Zeng, Houqiang Li, Yuhui Yuan, Lei Sun, Jingdong Wang. 1. **[ConvBERT](https://huggingface.co/docs/transformers/model_doc/convbert)** (from YituTech) released with the paper [ConvBERT: Improving BERT with Span-based Dynamic Convolution](https://arxiv.org/abs/2008.02496) by Zihang Jiang, Weihao Yu, Daquan Zhou, Yunpeng Chen, Jiashi Feng, Shuicheng Yan. 1. **[ConvNeXT](https://huggingface.co/docs/transformers/model_doc/convnext)** (from Facebook AI) released with the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Zhuang Liu, Hanzi Mao, Chao-Yuan Wu, Christoph Feichtenhofer, Trevor Darrell, Saining Xie. 1. **[CPM](https://huggingface.co/docs/transformers/model_doc/cpm)** (from Tsinghua University) released with the paper [CPM: A Large-scale Generative Chinese Pre-trained Language Model](https://arxiv.org/abs/2012.00413) by Zhengyan Zhang, Xu Han, Hao Zhou, Pei Ke, Yuxian Gu, Deming Ye, Yujia Qin, Yusheng Su, Haozhe Ji, Jian Guan, Fanchao Qi, Xiaozhi Wang, Yanan Zheng, Guoyang Zeng, Huanqi Cao, Shengqi Chen, Daixuan Li, Zhenbo Sun, Zhiyuan Liu, Minlie Huang, Wentao Han, Jie Tang, Juanzi Li, Xiaoyan Zhu, Maosong Sun. @@ -286,7 +286,7 @@ conda install -c huggingface transformers 1. **[ELECTRA](https://huggingface.co/docs/transformers/model_doc/electra)** (from Google Research/Stanford University) released with the paper [ELECTRA: Pre-training text encoders as discriminators rather than generators](https://arxiv.org/abs/2003.10555) by Kevin Clark, Minh-Thang Luong, Quoc V. Le, Christopher D. Manning. 1. **[EncoderDecoder](https://huggingface.co/docs/transformers/model_doc/encoder-decoder)** (from Google Research) released with the paper [Leveraging Pre-trained Checkpoints for Sequence Generation Tasks](https://arxiv.org/abs/1907.12461) by Sascha Rothe, Shashi Narayan, Aliaksei Severyn. 1. **[ERNIE](https://huggingface.co/docs/transformers/model_doc/ernie)** (from Baidu) released with the paper [ERNIE: Enhanced Representation through Knowledge Integration](https://arxiv.org/abs/1904.09223) by Yu Sun, Shuohuan Wang, Yukun Li, Shikun Feng, Xuyi Chen, Han Zhang, Xin Tian, Danxiang Zhu, Hao Tian, Hua Wu. -1. **[ESM](https://huggingface.co/docs/transformers/main/model_doc/esm)** (from Meta AI) are transformer protein language models. **ESM-1b** was released with the paper [Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences](https://www.pnas.org/content/118/15/e2016239118) by Alexander Rives, Joshua Meier, Tom Sercu, Siddharth Goyal, Zeming Lin, Jason Liu, Demi Guo, Myle Ott, C. Lawrence Zitnick, Jerry Ma, and Rob Fergus. **ESM-1v** was released with the paper [Language models enable zero-shot prediction of the effects of mutations on protein function](https://doi.org/10.1101/2021.07.09.450648) by Joshua Meier, Roshan Rao, Robert Verkuil, Jason Liu, Tom Sercu and Alexander Rives. **ESM-2** was released with the paper [Language models of protein sequences at the scale of evolution enable accurate structure prediction](https://doi.org/10.1101/2022.07.20.500902) by Zeming Lin, Halil Akin, Roshan Rao, Brian Hie, Zhongkai Zhu, Wenting Lu, Allan dos Santos Costa, Maryam Fazel-Zarandi, Tom Sercu, Sal Candido, Alexander Rives. +1. **[ESM](https://huggingface.co/docs/transformers/model_doc/esm)** (from Meta AI) are transformer protein language models. **ESM-1b** was released with the paper [Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences](https://www.pnas.org/content/118/15/e2016239118) by Alexander Rives, Joshua Meier, Tom Sercu, Siddharth Goyal, Zeming Lin, Jason Liu, Demi Guo, Myle Ott, C. Lawrence Zitnick, Jerry Ma, and Rob Fergus. **ESM-1v** was released with the paper [Language models enable zero-shot prediction of the effects of mutations on protein function](https://doi.org/10.1101/2021.07.09.450648) by Joshua Meier, Roshan Rao, Robert Verkuil, Jason Liu, Tom Sercu and Alexander Rives. **ESM-2** was released with the paper [Language models of protein sequences at the scale of evolution enable accurate structure prediction](https://doi.org/10.1101/2022.07.20.500902) by Zeming Lin, Halil Akin, Roshan Rao, Brian Hie, Zhongkai Zhu, Wenting Lu, Allan dos Santos Costa, Maryam Fazel-Zarandi, Tom Sercu, Sal Candido, Alexander Rives. 1. **[FlauBERT](https://huggingface.co/docs/transformers/model_doc/flaubert)** (from CNRS) released with the paper [FlauBERT: Unsupervised Language Model Pre-training for French](https://arxiv.org/abs/1912.05372) by Hang Le, Loïc Vial, Jibril Frej, Vincent Segonne, Maximin Coavoux, Benjamin Lecouteux, Alexandre Allauzen, Benoît Crabbé, Laurent Besacier, Didier Schwab. 1. **[FLAVA](https://huggingface.co/docs/transformers/model_doc/flava)** (from Facebook AI) released with the paper [FLAVA: A Foundational Language And Vision Alignment Model](https://arxiv.org/abs/2112.04482) by Amanpreet Singh, Ronghang Hu, Vedanuj Goswami, Guillaume Couairon, Wojciech Galuba, Marcus Rohrbach, and Douwe Kiela. 1. **[FNet](https://huggingface.co/docs/transformers/model_doc/fnet)** (from Google Research) released with the paper [FNet: Mixing Tokens with Fourier Transforms](https://arxiv.org/abs/2105.03824) by James Lee-Thorp, Joshua Ainslie, Ilya Eckstein, Santiago Ontanon. @@ -315,7 +315,7 @@ conda install -c huggingface transformers 1. **[M-CTC-T](https://huggingface.co/docs/transformers/model_doc/mctct)** (from Facebook) released with the paper [Pseudo-Labeling For Massively Multilingual Speech Recognition](https://arxiv.org/abs/2111.00161) by Loren Lugosch, Tatiana Likhomanenko, Gabriel Synnaeve, and Ronan Collobert. 1. **[M2M100](https://huggingface.co/docs/transformers/model_doc/m2m_100)** (from Facebook) released with the paper [Beyond English-Centric Multilingual Machine Translation](https://arxiv.org/abs/2010.11125) by Angela Fan, Shruti Bhosale, Holger Schwenk, Zhiyi Ma, Ahmed El-Kishky, Siddharth Goyal, Mandeep Baines, Onur Celebi, Guillaume Wenzek, Vishrav Chaudhary, Naman Goyal, Tom Birch, Vitaliy Liptchinsky, Sergey Edunov, Edouard Grave, Michael Auli, Armand Joulin. 1. **[MarianMT](https://huggingface.co/docs/transformers/model_doc/marian)** Machine translation models trained using [OPUS](http://opus.nlpl.eu/) data by Jörg Tiedemann. The [Marian Framework](https://marian-nmt.github.io/) is being developed by the Microsoft Translator Team. -1. **[MarkupLM](https://huggingface.co/docs/transformers/main/model_doc/markuplm)** (from Microsoft Research Asia) released with the paper [MarkupLM: Pre-training of Text and Markup Language for Visually-rich Document Understanding](https://arxiv.org/abs/2110.08518) by Junlong Li, Yiheng Xu, Lei Cui, Furu Wei. +1. **[MarkupLM](https://huggingface.co/docs/transformers/model_doc/markuplm)** (from Microsoft Research Asia) released with the paper [MarkupLM: Pre-training of Text and Markup Language for Visually-rich Document Understanding](https://arxiv.org/abs/2110.08518) by Junlong Li, Yiheng Xu, Lei Cui, Furu Wei. 1. **[MaskFormer](https://huggingface.co/docs/transformers/model_doc/maskformer)** (from Meta and UIUC) released with the paper [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://arxiv.org/abs/2107.06278) by Bowen Cheng, Alexander G. Schwing, Alexander Kirillov 1. **[mBART](https://huggingface.co/docs/transformers/model_doc/mbart)** (from Facebook) released with the paper [Multilingual Denoising Pre-training for Neural Machine Translation](https://arxiv.org/abs/2001.08210) by Yinhan Liu, Jiatao Gu, Naman Goyal, Xian Li, Sergey Edunov, Marjan Ghazvininejad, Mike Lewis, Luke Zettlemoyer. 1. **[mBART-50](https://huggingface.co/docs/transformers/model_doc/mbart)** (from Facebook) released with the paper [Multilingual Translation with Extensible Multilingual Pretraining and Finetuning](https://arxiv.org/abs/2008.00401) by Yuqing Tang, Chau Tran, Xian Li, Peng-Jen Chen, Naman Goyal, Vishrav Chaudhary, Jiatao Gu, Angela Fan. @@ -361,7 +361,7 @@ conda install -c huggingface transformers 1. **[T5v1.1](https://huggingface.co/docs/transformers/model_doc/t5v1.1)** (from Google AI) released with the paper [google-research/text-to-text-transfer-transformer](https://github.com/google-research/text-to-text-transfer-transformer/blob/main/released_checkpoints.md#t511) by Colin Raffel and Noam Shazeer and Adam Roberts and Katherine Lee and Sharan Narang and Michael Matena and Yanqi Zhou and Wei Li and Peter J. Liu. 1. **[TAPAS](https://huggingface.co/docs/transformers/model_doc/tapas)** (from Google AI) released with the paper [TAPAS: Weakly Supervised Table Parsing via Pre-training](https://arxiv.org/abs/2004.02349) by Jonathan Herzig, Paweł Krzysztof Nowak, Thomas Müller, Francesco Piccinno and Julian Martin Eisenschlos. 1. **[TAPEX](https://huggingface.co/docs/transformers/model_doc/tapex)** (from Microsoft Research) released with the paper [TAPEX: Table Pre-training via Learning a Neural SQL Executor](https://arxiv.org/abs/2107.07653) by Qian Liu, Bei Chen, Jiaqi Guo, Morteza Ziyadi, Zeqi Lin, Weizhu Chen, Jian-Guang Lou. -1. **[Time Series Transformer](https://huggingface.co/docs/transformers/main/model_doc/time_series_transformer)** (from HuggingFace). +1. **[Time Series Transformer](https://huggingface.co/docs/transformers/model_doc/time_series_transformer)** (from HuggingFace). 1. **[Trajectory Transformer](https://huggingface.co/docs/transformers/model_doc/trajectory_transformers)** (from the University of California at Berkeley) released with the paper [Offline Reinforcement Learning as One Big Sequence Modeling Problem](https://arxiv.org/abs/2106.02039) by Michael Janner, Qiyang Li, Sergey Levine 1. **[Transformer-XL](https://huggingface.co/docs/transformers/model_doc/transfo-xl)** (from Google/CMU) released with the paper [Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context](https://arxiv.org/abs/1901.02860) by Zihang Dai*, Zhilin Yang*, Yiming Yang, Jaime Carbonell, Quoc V. Le, Ruslan Salakhutdinov. 1. **[TrOCR](https://huggingface.co/docs/transformers/model_doc/trocr)** (from Microsoft) released with the paper [TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models](https://arxiv.org/abs/2109.10282) by Minghao Li, Tengchao Lv, Lei Cui, Yijuan Lu, Dinei Florencio, Cha Zhang, Zhoujun Li, Furu Wei. @@ -374,12 +374,12 @@ conda install -c huggingface transformers 1. **[Vision Transformer (ViT)](https://huggingface.co/docs/transformers/model_doc/vit)** (from Google AI) released with the paper [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://arxiv.org/abs/2010.11929) by Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, Neil Houlsby. 1. **[VisualBERT](https://huggingface.co/docs/transformers/model_doc/visual_bert)** (from UCLA NLP) released with the paper [VisualBERT: A Simple and Performant Baseline for Vision and Language](https://arxiv.org/pdf/1908.03557) by Liunian Harold Li, Mark Yatskar, Da Yin, Cho-Jui Hsieh, Kai-Wei Chang. 1. **[ViTMAE](https://huggingface.co/docs/transformers/model_doc/vit_mae)** (from Meta AI) released with the paper [Masked Autoencoders Are Scalable Vision Learners](https://arxiv.org/abs/2111.06377) by Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Dollár, Ross Girshick. -1. **[ViTMSN](https://huggingface.co/docs/transformers/main/model_doc/vit_msn)** (from Meta AI) released with the paper [Masked Siamese Networks for Label-Efficient Learning](https://arxiv.org/abs/2204.07141) by Mahmoud Assran, Mathilde Caron, Ishan Misra, Piotr Bojanowski, Florian Bordes, Pascal Vincent, Armand Joulin, Michael Rabbat, Nicolas Ballas. +1. **[ViTMSN](https://huggingface.co/docs/transformers/model_doc/vit_msn)** (from Meta AI) released with the paper [Masked Siamese Networks for Label-Efficient Learning](https://arxiv.org/abs/2204.07141) by Mahmoud Assran, Mathilde Caron, Ishan Misra, Piotr Bojanowski, Florian Bordes, Pascal Vincent, Armand Joulin, Michael Rabbat, Nicolas Ballas. 1. **[Wav2Vec2](https://huggingface.co/docs/transformers/model_doc/wav2vec2)** (from Facebook AI) released with the paper [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://arxiv.org/abs/2006.11477) by Alexei Baevski, Henry Zhou, Abdelrahman Mohamed, Michael Auli. 1. **[Wav2Vec2-Conformer](https://huggingface.co/docs/transformers/model_doc/wav2vec2-conformer)** (from Facebook AI) released with the paper [FAIRSEQ S2T: Fast Speech-to-Text Modeling with FAIRSEQ](https://arxiv.org/abs/2010.05171) by Changhan Wang, Yun Tang, Xutai Ma, Anne Wu, Sravya Popuri, Dmytro Okhonko, Juan Pino. 1. **[Wav2Vec2Phoneme](https://huggingface.co/docs/transformers/model_doc/wav2vec2_phoneme)** (from Facebook AI) released with the paper [Simple and Effective Zero-shot Cross-lingual Phoneme Recognition](https://arxiv.org/abs/2109.11680) by Qiantong Xu, Alexei Baevski, Michael Auli. 1. **[WavLM](https://huggingface.co/docs/transformers/model_doc/wavlm)** (from Microsoft Research) released with the paper [WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://arxiv.org/abs/2110.13900) by Sanyuan Chen, Chengyi Wang, Zhengyang Chen, Yu Wu, Shujie Liu, Zhuo Chen, Jinyu Li, Naoyuki Kanda, Takuya Yoshioka, Xiong Xiao, Jian Wu, Long Zhou, Shuo Ren, Yanmin Qian, Yao Qian, Jian Wu, Michael Zeng, Furu Wei. -1. **[Whisper](https://huggingface.co/docs/transformers/main/model_doc/whisper)** (from OpenAI) released with the paper [Robust Speech Recognition via Large-Scale Weak Supervision](https://cdn.openai.com/papers/whisper.pdf) by Alec Radford, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey, Ilya Sutskever. +1. **[Whisper](https://huggingface.co/docs/transformers/model_doc/whisper)** (from OpenAI) released with the paper [Robust Speech Recognition via Large-Scale Weak Supervision](https://cdn.openai.com/papers/whisper.pdf) by Alec Radford, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey, Ilya Sutskever. 1. **[X-CLIP](https://huggingface.co/docs/transformers/model_doc/xclip)** (from Microsoft Research) released with the paper [Expanding Language-Image Pretrained Models for General Video Recognition](https://arxiv.org/abs/2208.02816) by Bolin Ni, Houwen Peng, Minghao Chen, Songyang Zhang, Gaofeng Meng, Jianlong Fu, Shiming Xiang, Haibin Ling. 1. **[XGLM](https://huggingface.co/docs/transformers/model_doc/xglm)** (From Facebook AI) released with the paper [Few-shot Learning with Multilingual Language Models](https://arxiv.org/abs/2112.10668) by Xi Victoria Lin, Todor Mihaylov, Mikel Artetxe, Tianlu Wang, Shuohui Chen, Daniel Simig, Myle Ott, Naman Goyal, Shruti Bhosale, Jingfei Du, Ramakanth Pasunuru, Sam Shleifer, Punit Singh Koura, Vishrav Chaudhary, Brian O'Horo, Jeff Wang, Luke Zettlemoyer, Zornitsa Kozareva, Mona Diab, Veselin Stoyanov, Xian Li. 1. **[XLM](https://huggingface.co/docs/transformers/model_doc/xlm)** (from Facebook) released together with the paper [Cross-lingual Language Model Pretraining](https://arxiv.org/abs/1901.07291) by Guillaume Lample and Alexis Conneau. diff --git a/examples/flax/question-answering/run_qa.py b/examples/flax/question-answering/run_qa.py index c044ff628c8abb..1dffacaccceb2c 100644 --- a/examples/flax/question-answering/run_qa.py +++ b/examples/flax/question-answering/run_qa.py @@ -61,7 +61,7 @@ logger = logging.getLogger(__name__) # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") Array = Any Dataset = datasets.arrow_dataset.Dataset diff --git a/examples/flax/text-classification/run_flax_glue.py b/examples/flax/text-classification/run_flax_glue.py index 29c2b511fcaa4f..7bd03160cf859c 100755 --- a/examples/flax/text-classification/run_flax_glue.py +++ b/examples/flax/text-classification/run_flax_glue.py @@ -54,7 +54,7 @@ logger = logging.getLogger(__name__) # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") Array = Any Dataset = datasets.arrow_dataset.Dataset diff --git a/examples/flax/token-classification/run_flax_ner.py b/examples/flax/token-classification/run_flax_ner.py index 929e249f446a99..0a762185a6cb88 100644 --- a/examples/flax/token-classification/run_flax_ner.py +++ b/examples/flax/token-classification/run_flax_ner.py @@ -55,7 +55,7 @@ logger = logging.getLogger(__name__) # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/token-classification/requirements.txt") diff --git a/examples/pytorch/audio-classification/run_audio_classification.py b/examples/pytorch/audio-classification/run_audio_classification.py index e7d7f30a67dd04..eb2f4e8a7bb041 100644 --- a/examples/pytorch/audio-classification/run_audio_classification.py +++ b/examples/pytorch/audio-classification/run_audio_classification.py @@ -45,7 +45,7 @@ logger = logging.getLogger(__name__) # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.14.0", "To fix: pip install -r examples/pytorch/audio-classification/requirements.txt") diff --git a/examples/pytorch/contrastive-image-text/run_clip.py b/examples/pytorch/contrastive-image-text/run_clip.py index 1296d0675fa9c6..b4a31fb49d4e13 100644 --- a/examples/pytorch/contrastive-image-text/run_clip.py +++ b/examples/pytorch/contrastive-image-text/run_clip.py @@ -54,7 +54,7 @@ logger = logging.getLogger(__name__) # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/contrastive-image-text/requirements.txt") diff --git a/examples/pytorch/image-classification/run_image_classification.py b/examples/pytorch/image-classification/run_image_classification.py index eb20c9efedf77c..8e6871a0982616 100644 --- a/examples/pytorch/image-classification/run_image_classification.py +++ b/examples/pytorch/image-classification/run_image_classification.py @@ -55,7 +55,7 @@ logger = logging.getLogger(__name__) # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/image-classification/requirements.txt") diff --git a/examples/pytorch/image-classification/run_image_classification_no_trainer.py b/examples/pytorch/image-classification/run_image_classification_no_trainer.py index 902ba962986e88..42ec7dc970eaf4 100644 --- a/examples/pytorch/image-classification/run_image_classification_no_trainer.py +++ b/examples/pytorch/image-classification/run_image_classification_no_trainer.py @@ -53,7 +53,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") logger = get_logger(__name__) diff --git a/examples/pytorch/image-pretraining/run_mae.py b/examples/pytorch/image-pretraining/run_mae.py index 9e8c366d3d2728..e210464ece94b4 100644 --- a/examples/pytorch/image-pretraining/run_mae.py +++ b/examples/pytorch/image-pretraining/run_mae.py @@ -43,7 +43,7 @@ logger = logging.getLogger(__name__) # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/image-pretraining/requirements.txt") diff --git a/examples/pytorch/image-pretraining/run_mim.py b/examples/pytorch/image-pretraining/run_mim.py index c05c1b1333f606..49ffe999524f55 100644 --- a/examples/pytorch/image-pretraining/run_mim.py +++ b/examples/pytorch/image-pretraining/run_mim.py @@ -48,7 +48,7 @@ logger = logging.getLogger(__name__) # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/image-pretraining/requirements.txt") diff --git a/examples/pytorch/language-modeling/run_clm.py b/examples/pytorch/language-modeling/run_clm.py index f0a63ef732d14d..fe03cde7c8c71d 100755 --- a/examples/pytorch/language-modeling/run_clm.py +++ b/examples/pytorch/language-modeling/run_clm.py @@ -54,7 +54,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/language-modeling/requirements.txt") diff --git a/examples/pytorch/language-modeling/run_clm_no_trainer.py b/examples/pytorch/language-modeling/run_clm_no_trainer.py index 3e1c9048334455..165eded1e71431 100755 --- a/examples/pytorch/language-modeling/run_clm_no_trainer.py +++ b/examples/pytorch/language-modeling/run_clm_no_trainer.py @@ -57,7 +57,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") logger = get_logger(__name__) diff --git a/examples/pytorch/language-modeling/run_mlm.py b/examples/pytorch/language-modeling/run_mlm.py index 576fc3528febae..d18b08eac1ddb0 100755 --- a/examples/pytorch/language-modeling/run_mlm.py +++ b/examples/pytorch/language-modeling/run_mlm.py @@ -53,7 +53,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/language-modeling/requirements.txt") diff --git a/examples/pytorch/language-modeling/run_mlm_no_trainer.py b/examples/pytorch/language-modeling/run_mlm_no_trainer.py index 9ce3cdd09f83bd..1134d1a0c12e71 100755 --- a/examples/pytorch/language-modeling/run_mlm_no_trainer.py +++ b/examples/pytorch/language-modeling/run_mlm_no_trainer.py @@ -57,7 +57,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") logger = get_logger(__name__) require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/language-modeling/requirements.txt") diff --git a/examples/pytorch/language-modeling/run_plm.py b/examples/pytorch/language-modeling/run_plm.py index 12430b2a8985fc..e445feb878714b 100755 --- a/examples/pytorch/language-modeling/run_plm.py +++ b/examples/pytorch/language-modeling/run_plm.py @@ -47,7 +47,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/language-modeling/requirements.txt") diff --git a/examples/pytorch/multiple-choice/run_swag.py b/examples/pytorch/multiple-choice/run_swag.py index c8084e25a22138..536fe648974f33 100755 --- a/examples/pytorch/multiple-choice/run_swag.py +++ b/examples/pytorch/multiple-choice/run_swag.py @@ -47,7 +47,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") logger = logging.getLogger(__name__) diff --git a/examples/pytorch/multiple-choice/run_swag_no_trainer.py b/examples/pytorch/multiple-choice/run_swag_no_trainer.py index 7235d9494e30a0..eebb2e55d1eb8c 100755 --- a/examples/pytorch/multiple-choice/run_swag_no_trainer.py +++ b/examples/pytorch/multiple-choice/run_swag_no_trainer.py @@ -56,7 +56,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") logger = get_logger(__name__) # You should update this to your particular problem to have better documentation of `model_type` diff --git a/examples/pytorch/question-answering/run_qa.py b/examples/pytorch/question-answering/run_qa.py index 65940cd889fb9c..1240623b525531 100755 --- a/examples/pytorch/question-answering/run_qa.py +++ b/examples/pytorch/question-answering/run_qa.py @@ -49,7 +49,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/question-answering/requirements.txt") diff --git a/examples/pytorch/question-answering/run_qa_beam_search.py b/examples/pytorch/question-answering/run_qa_beam_search.py index a0a1a6c51c4e52..ba7be8e60eddb0 100755 --- a/examples/pytorch/question-answering/run_qa_beam_search.py +++ b/examples/pytorch/question-answering/run_qa_beam_search.py @@ -48,7 +48,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/question-answering/requirements.txt") diff --git a/examples/pytorch/question-answering/run_qa_beam_search_no_trainer.py b/examples/pytorch/question-answering/run_qa_beam_search_no_trainer.py index e1a70a980d3cb9..a5318f00c55acc 100644 --- a/examples/pytorch/question-answering/run_qa_beam_search_no_trainer.py +++ b/examples/pytorch/question-answering/run_qa_beam_search_no_trainer.py @@ -56,7 +56,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/question-answering/requirements.txt") diff --git a/examples/pytorch/question-answering/run_qa_no_trainer.py b/examples/pytorch/question-answering/run_qa_no_trainer.py index 49d5b752bfb564..0d5889705b1c41 100755 --- a/examples/pytorch/question-answering/run_qa_no_trainer.py +++ b/examples/pytorch/question-answering/run_qa_no_trainer.py @@ -57,7 +57,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/question-answering/requirements.txt") diff --git a/examples/pytorch/question-answering/run_seq2seq_qa.py b/examples/pytorch/question-answering/run_seq2seq_qa.py index abcc11fbb3edb7..4965fbc521da24 100644 --- a/examples/pytorch/question-answering/run_seq2seq_qa.py +++ b/examples/pytorch/question-answering/run_seq2seq_qa.py @@ -45,7 +45,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/question-answering/requirements.txt") diff --git a/examples/pytorch/semantic-segmentation/run_semantic_segmentation.py b/examples/pytorch/semantic-segmentation/run_semantic_segmentation.py index bf8099135702bb..7f8a5774acc5b8 100644 --- a/examples/pytorch/semantic-segmentation/run_semantic_segmentation.py +++ b/examples/pytorch/semantic-segmentation/run_semantic_segmentation.py @@ -51,7 +51,7 @@ logger = logging.getLogger(__name__) # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=2.0.0", "To fix: pip install -r examples/pytorch/semantic-segmentation/requirements.txt") diff --git a/examples/pytorch/semantic-segmentation/run_semantic_segmentation_no_trainer.py b/examples/pytorch/semantic-segmentation/run_semantic_segmentation_no_trainer.py index 8eb18434879acb..77217354d13f49 100644 --- a/examples/pytorch/semantic-segmentation/run_semantic_segmentation_no_trainer.py +++ b/examples/pytorch/semantic-segmentation/run_semantic_segmentation_no_trainer.py @@ -50,7 +50,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") logger = get_logger(__name__) diff --git a/examples/pytorch/speech-recognition/run_speech_recognition_ctc.py b/examples/pytorch/speech-recognition/run_speech_recognition_ctc.py index 904a297c5a86d4..7d35467712022b 100755 --- a/examples/pytorch/speech-recognition/run_speech_recognition_ctc.py +++ b/examples/pytorch/speech-recognition/run_speech_recognition_ctc.py @@ -50,7 +50,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.18.0", "To fix: pip install -r examples/pytorch/speech-recognition/requirements.txt") diff --git a/examples/pytorch/speech-recognition/run_speech_recognition_seq2seq.py b/examples/pytorch/speech-recognition/run_speech_recognition_seq2seq.py index e372bd833aac8d..75fcbd7ac03c52 100755 --- a/examples/pytorch/speech-recognition/run_speech_recognition_seq2seq.py +++ b/examples/pytorch/speech-recognition/run_speech_recognition_seq2seq.py @@ -48,7 +48,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.18.0", "To fix: pip install -r examples/pytorch/speech-recognition/requirements.txt") diff --git a/examples/pytorch/summarization/run_summarization.py b/examples/pytorch/summarization/run_summarization.py index e8969d216f945a..60db3e80a64bd1 100755 --- a/examples/pytorch/summarization/run_summarization.py +++ b/examples/pytorch/summarization/run_summarization.py @@ -52,7 +52,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/summarization/requirements.txt") diff --git a/examples/pytorch/summarization/run_summarization_no_trainer.py b/examples/pytorch/summarization/run_summarization_no_trainer.py index 9b8ee1a1dc876d..be11d6bd4ba335 100644 --- a/examples/pytorch/summarization/run_summarization_no_trainer.py +++ b/examples/pytorch/summarization/run_summarization_no_trainer.py @@ -56,7 +56,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") logger = get_logger(__name__) require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/summarization/requirements.txt") diff --git a/examples/pytorch/text-classification/run_glue.py b/examples/pytorch/text-classification/run_glue.py index 3f97fc3f5e1b44..3eb423f0861db4 100755 --- a/examples/pytorch/text-classification/run_glue.py +++ b/examples/pytorch/text-classification/run_glue.py @@ -48,7 +48,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/text-classification/requirements.txt") diff --git a/examples/pytorch/text-classification/run_glue_no_trainer.py b/examples/pytorch/text-classification/run_glue_no_trainer.py index ff972a73bdf0d6..f7ea1cb225a0af 100644 --- a/examples/pytorch/text-classification/run_glue_no_trainer.py +++ b/examples/pytorch/text-classification/run_glue_no_trainer.py @@ -48,7 +48,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") logger = get_logger(__name__) diff --git a/examples/pytorch/text-classification/run_xnli.py b/examples/pytorch/text-classification/run_xnli.py index 2b512a2bb1b302..55523edfc71312 100755 --- a/examples/pytorch/text-classification/run_xnli.py +++ b/examples/pytorch/text-classification/run_xnli.py @@ -48,7 +48,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/text-classification/requirements.txt") diff --git a/examples/pytorch/token-classification/run_ner.py b/examples/pytorch/token-classification/run_ner.py index 6541593c5d0e09..52cbbb87bdc686 100755 --- a/examples/pytorch/token-classification/run_ner.py +++ b/examples/pytorch/token-classification/run_ner.py @@ -49,7 +49,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/token-classification/requirements.txt") diff --git a/examples/pytorch/token-classification/run_ner_no_trainer.py b/examples/pytorch/token-classification/run_ner_no_trainer.py index b8c3a9eae8a755..71120b31432892 100755 --- a/examples/pytorch/token-classification/run_ner_no_trainer.py +++ b/examples/pytorch/token-classification/run_ner_no_trainer.py @@ -55,7 +55,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") logger = get_logger(__name__) require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/token-classification/requirements.txt") diff --git a/examples/pytorch/translation/run_translation.py b/examples/pytorch/translation/run_translation.py index 9df352f63dd997..ce8a04f9aee04b 100755 --- a/examples/pytorch/translation/run_translation.py +++ b/examples/pytorch/translation/run_translation.py @@ -52,7 +52,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/translation/requirements.txt") diff --git a/examples/pytorch/translation/run_translation_no_trainer.py b/examples/pytorch/translation/run_translation_no_trainer.py index 8018652eae2c18..ca439b0e1708d3 100644 --- a/examples/pytorch/translation/run_translation_no_trainer.py +++ b/examples/pytorch/translation/run_translation_no_trainer.py @@ -57,7 +57,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") logger = get_logger(__name__) require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/translation/requirements.txt") diff --git a/examples/tensorflow/multiple-choice/run_swag.py b/examples/tensorflow/multiple-choice/run_swag.py index 34d8281ac5a03e..87cfa807832a90 100644 --- a/examples/tensorflow/multiple-choice/run_swag.py +++ b/examples/tensorflow/multiple-choice/run_swag.py @@ -50,7 +50,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") logger = logging.getLogger(__name__) diff --git a/examples/tensorflow/question-answering/run_qa.py b/examples/tensorflow/question-answering/run_qa.py index 9bba2e54e915ba..71e478bc9f1dc9 100755 --- a/examples/tensorflow/question-answering/run_qa.py +++ b/examples/tensorflow/question-answering/run_qa.py @@ -48,7 +48,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") logger = logging.getLogger(__name__) diff --git a/examples/tensorflow/summarization/run_summarization.py b/examples/tensorflow/summarization/run_summarization.py index 21ad8e4fa02af9..75ef60dc61d9a7 100644 --- a/examples/tensorflow/summarization/run_summarization.py +++ b/examples/tensorflow/summarization/run_summarization.py @@ -53,7 +53,7 @@ # region Checking dependencies # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/summarization/requirements.txt") diff --git a/examples/tensorflow/text-classification/run_glue.py b/examples/tensorflow/text-classification/run_glue.py index ab91e84a59653a..dcc9478e4d83ec 100644 --- a/examples/tensorflow/text-classification/run_glue.py +++ b/examples/tensorflow/text-classification/run_glue.py @@ -47,7 +47,7 @@ # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") task_to_keys = { "cola": ("sentence", None), diff --git a/examples/tensorflow/translation/run_translation.py b/examples/tensorflow/translation/run_translation.py index 6632dbe9105867..52b7832705edef 100644 --- a/examples/tensorflow/translation/run_translation.py +++ b/examples/tensorflow/translation/run_translation.py @@ -56,7 +56,7 @@ # region Dependencies and constants # Will error if the minimal version of Transformers is not installed. Remove at your own risks. -check_min_version("4.23.0.dev0") +check_min_version("4.23.0") require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/summarization/requirements.txt") diff --git a/setup.py b/setup.py index 84c5df4793ef02..842e3e075723b6 100644 --- a/setup.py +++ b/setup.py @@ -408,7 +408,7 @@ def run(self): setup( name="transformers", - version="4.23.0.dev0", # expected format is one of x.y.z.dev0, or x.y.z.rc1 or x.y.z (no to dashes, yes to dots) + version="4.23.0", # expected format is one of x.y.z.dev0, or x.y.z.rc1 or x.y.z (no to dashes, yes to dots) author="The Hugging Face team (past and future) with the help of all our contributors (https://github.com/huggingface/transformers/graphs/contributors)", author_email="transformers@huggingface.co", description="State-of-the-art Machine Learning for JAX, PyTorch and TensorFlow", diff --git a/src/transformers/__init__.py b/src/transformers/__init__.py index b634ffea72b119..df633eaa7882b7 100755 --- a/src/transformers/__init__.py +++ b/src/transformers/__init__.py @@ -22,7 +22,7 @@ # to defer the actual importing for when the objects are requested. This way `import transformers` provides the names # in the namespace without actually importing anything (and especially none of the backends). -__version__ = "4.23.0.dev0" +__version__ = "4.23.0" from typing import TYPE_CHECKING From c8bc0a0b021d3f3815f96e2a716b8744f1ed6275 Mon Sep 17 00:00:00 2001 From: Arthur <48595927+ArthurZucker@users.noreply.github.com> Date: Tue, 11 Oct 2022 13:17:53 +0200 Subject: [PATCH 2/6] Fix whisper for `pipeline` (#19482) * update feature extractor params * update attention mask handling * fix doc and pipeline test * add warning when skipping test * add whisper translation and transcription test * fix build doc test --- .../whisper/feature_extraction_whisper.py | 14 ++++++ .../models/whisper/modeling_whisper.py | 40 +++++++--------- ..._pipelines_automatic_speech_recognition.py | 48 +++++++++++++++++++ tests/pipelines/test_pipelines_common.py | 12 ++++- 4 files changed, 90 insertions(+), 24 deletions(-) diff --git a/src/transformers/models/whisper/feature_extraction_whisper.py b/src/transformers/models/whisper/feature_extraction_whisper.py index dda53dffaafd0a..0d6bbd9ed18bbf 100644 --- a/src/transformers/models/whisper/feature_extraction_whisper.py +++ b/src/transformers/models/whisper/feature_extraction_whisper.py @@ -218,6 +218,7 @@ def __call__( return_attention_mask: Optional[bool] = None, padding: Optional[str] = "max_length", max_length: Optional[int] = None, + sampling_rate: Optional[int] = None, **kwargs ) -> BatchFeature: """ @@ -261,6 +262,19 @@ def __call__( The value that is used to fill the padding values / vectors. """ + if sampling_rate is not None: + if sampling_rate != self.sampling_rate: + raise ValueError( + f"The model corresponding to this feature extractor: {self} was trained using a sampling rate of" + f" {self.sampling_rate}. Please make sure that the provided `raw_speech` input was sampled with" + f" {self.sampling_rate} and not {sampling_rate}." + ) + else: + logger.warning( + "It is strongly recommended to pass the `sampling_rate` argument to this function. " + "Failing to do so can result in silent errors that might be hard to debug." + ) + is_batched = bool( isinstance(raw_speech, (list, tuple)) and (isinstance(raw_speech[0], np.ndarray) or isinstance(raw_speech[0], (tuple, list))) diff --git a/src/transformers/models/whisper/modeling_whisper.py b/src/transformers/models/whisper/modeling_whisper.py index ef23914b8ce77d..a6e50153375c25 100644 --- a/src/transformers/models/whisper/modeling_whisper.py +++ b/src/transformers/models/whisper/modeling_whisper.py @@ -31,13 +31,22 @@ Seq2SeqModelOutput, ) from ...modeling_utils import PreTrainedModel -from ...utils import add_start_docstrings, add_start_docstrings_to_model_forward, logging, replace_return_docstrings +from ...utils import ( + add_code_sample_docstrings, + add_start_docstrings, + add_start_docstrings_to_model_forward, + logging, + replace_return_docstrings, +) from .configuration_whisper import WhisperConfig logger = logging.get_logger(__name__) _CONFIG_FOR_DOC = "WhisperConfig" +_CHECKPOINT_FOR_DOC = "openai/whisper-tiny" +_PROCESSOR_FOR_DOC = "openai/whisper-tiny" +_EXPECTED_OUTPUT_SHAPE = [1, 2, 512] WHISPER_PRETRAINED_MODEL_ARCHIVE_LIST = [ @@ -982,7 +991,14 @@ def get_decoder(self): return self.decoder @add_start_docstrings_to_model_forward(WHISPER_INPUTS_DOCSTRING) - @replace_return_docstrings(output_type=Seq2SeqLMOutput, config_class=_CONFIG_FOR_DOC) + @add_code_sample_docstrings( + processor_class=_PROCESSOR_FOR_DOC, + checkpoint=_CHECKPOINT_FOR_DOC, + output_type=Seq2SeqModelOutput, + config_class=_CONFIG_FOR_DOC, + expected_output=_EXPECTED_OUTPUT_SHAPE, + modality="audio", + ) def forward( self, input_features=None, @@ -999,26 +1015,6 @@ def forward( output_hidden_states=None, return_dict=None, ): - r""" - Returns: - - Example: - - ```python - >>> import torch - >>> from transformers import WhisperModel, WhisperFeatureExtractor - >>> from datasets import load_dataset - - >>> model = WhisperModel.from_pretrained("openai/whisper-base") - >>> feature_extractor = WhisperFeatureExtractor.from_pretrained("openai/whisper-base") - >>> ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation") - >>> inputs = feature_extractor(ds[0]["audio"]["array"], return_tensors="pt") - >>> input_features = inputs.input_features - >>> decoder_input_ids = torch.tensor([[1, 1]]) * model.config.decoder_start_token_id - >>> last_hidden_state = model(input_features, decoder_input_ids=decoder_input_ids).last_hidden_state - >>> list(last_hidden_state.shape) - [1, 2, 512] - ```""" output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions output_hidden_states = ( diff --git a/tests/pipelines/test_pipelines_automatic_speech_recognition.py b/tests/pipelines/test_pipelines_automatic_speech_recognition.py index d4fcbf5f78146c..f73fda39e97b65 100644 --- a/tests/pipelines/test_pipelines_automatic_speech_recognition.py +++ b/tests/pipelines/test_pipelines_automatic_speech_recognition.py @@ -26,6 +26,8 @@ AutoTokenizer, Speech2TextForConditionalGeneration, Wav2Vec2ForCTC, + WhisperForConditionalGeneration, + WhisperProcessor, ) from transformers.pipelines import AutomaticSpeechRecognitionPipeline, pipeline from transformers.pipelines.audio_utils import chunk_bytes_iter @@ -308,6 +310,52 @@ def test_simple_s2t(self): output = asr(data) self.assertEqual(output, {"text": "Un uomo disse all'universo: \"Signore, io esisto."}) + @slow + @require_torch + @require_torchaudio + def test_simple_whisper_asr(self): + speech_recognizer = pipeline( + task="automatic-speech-recognition", + model="openai/whisper-tiny.en", + framework="pt", + ) + ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation") + filename = ds[0]["file"] + output = speech_recognizer(filename) + self.assertEqual(output, {"text": " Mr. Quilter is the apostle of the middle classes, and we are glad to"}) + + @slow + @require_torch + @require_torchaudio + def test_simple_whisper_translation(self): + speech_recognizer = pipeline( + task="automatic-speech-recognition", + model="openai/whisper-large", + framework="pt", + ) + ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation").sort("id") + filename = ds[40]["file"] + output = speech_recognizer(filename) + self.assertEqual(output, {"text": " A man said to the universe, Sir, I exist."}) + + model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-large") + tokenizer = AutoTokenizer.from_pretrained("openai/whisper-large") + feature_extractor = AutoFeatureExtractor.from_pretrained("openai/whisper-large") + + speech_recognizer_2 = AutomaticSpeechRecognitionPipeline( + model=model, tokenizer=tokenizer, feature_extractor=feature_extractor + ) + output_2 = speech_recognizer_2(filename) + self.assertEqual(output, output_2) + + processor = WhisperProcessor(feature_extractor, tokenizer) + model.config.forced_decoder_ids = processor.get_decoder_prompt_ids(task="transcribe", language="it") + speech_translator = AutomaticSpeechRecognitionPipeline( + model=model, tokenizer=tokenizer, feature_extractor=feature_extractor + ) + output_3 = speech_translator(filename) + self.assertEqual(output_3, {"text": " Un uomo ha detto allo universo, Sir, esiste."}) + @slow @require_torch @require_torchaudio diff --git a/tests/pipelines/test_pipelines_common.py b/tests/pipelines/test_pipelines_common.py index 0f03a42440d79c..9449b380949880 100644 --- a/tests/pipelines/test_pipelines_common.py +++ b/tests/pipelines/test_pipelines_common.py @@ -178,8 +178,16 @@ def __repr__(self): class PipelineTestCaseMeta(type): def __new__(mcs, name, bases, dct): def gen_test(ModelClass, checkpoint, tiny_config, tokenizer_class, feature_extractor_class): - @skipIf(tiny_config is None, "TinyConfig does not exist") - @skipIf(checkpoint is None, "checkpoint does not exist") + @skipIf( + tiny_config is None, + "TinyConfig does not exist, make sure that you defined a `_CONFIG_FOR_DOC` variable in the modeling" + " file", + ) + @skipIf( + checkpoint is None, + "checkpoint does not exist, make sure that you defined a `_CHECKPOINT_FOR_DOC` variable in the" + " modeling file", + ) def test(self): if ModelClass.__name__.endswith("ForCausalLM"): tiny_config.is_encoder_decoder = False From bd469c40659ce76c81f69c7726759d249b4aef49 Mon Sep 17 00:00:00 2001 From: Sylvain Gugger Date: Tue, 11 Oct 2022 07:21:08 -0400 Subject: [PATCH 3/6] Release: v4.23.1 --- setup.py | 2 +- src/transformers/__init__.py | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/setup.py b/setup.py index 842e3e075723b6..c28380ac0e9755 100644 --- a/setup.py +++ b/setup.py @@ -408,7 +408,7 @@ def run(self): setup( name="transformers", - version="4.23.0", # expected format is one of x.y.z.dev0, or x.y.z.rc1 or x.y.z (no to dashes, yes to dots) + version="4.23.1", # expected format is one of x.y.z.dev0, or x.y.z.rc1 or x.y.z (no to dashes, yes to dots) author="The Hugging Face team (past and future) with the help of all our contributors (https://github.com/huggingface/transformers/graphs/contributors)", author_email="transformers@huggingface.co", description="State-of-the-art Machine Learning for JAX, PyTorch and TensorFlow", diff --git a/src/transformers/__init__.py b/src/transformers/__init__.py index df633eaa7882b7..ba24b9886d7902 100755 --- a/src/transformers/__init__.py +++ b/src/transformers/__init__.py @@ -22,7 +22,7 @@ # to defer the actual importing for when the objects are requested. This way `import transformers` provides the names # in the namespace without actually importing anything (and especially none of the backends). -__version__ = "4.23.0" +__version__ = "4.23.1" from typing import TYPE_CHECKING From 53c407db24ca7ee5ee53eab413d3185915bdee5a Mon Sep 17 00:00:00 2001 From: Konstantin Date: Thu, 13 Oct 2022 21:41:34 +0000 Subject: [PATCH 4/6] Add recipe_name to default file names --- src/transformers/models/bert/modeling_bert.py | 25 +++++++++++++++++-- src/transformers/utils/__init__.py | 1 + 2 files changed, 24 insertions(+), 2 deletions(-) diff --git a/src/transformers/models/bert/modeling_bert.py b/src/transformers/models/bert/modeling_bert.py index 11664f66cba8e7..fe79c728ee6cfe 100755 --- a/src/transformers/models/bert/modeling_bert.py +++ b/src/transformers/models/bert/modeling_bert.py @@ -240,6 +240,22 @@ def forward( return embeddings +class QATMatMul(nn.Module): + def __init__(self): + super().__init__() + + # behaves like normal torch.matmul unless a SparseML QuantizationModifier + # is initialized + self.wrap_qat = True + self.qat_wrapper_kwargs = { + "num_inputs": 2, + "input_qconfigs": ["asymmetric", "symmetric"], + } + + def forward(self, a: torch.Tensor, b: torch.Tensor): + return torch.matmul(a, b) + + class BertSelfAttention(nn.Module): def __init__(self, config, position_embedding_type=None): super().__init__() @@ -257,6 +273,11 @@ def __init__(self, config, position_embedding_type=None): self.key = nn.Linear(config.hidden_size, self.all_head_size) self.value = nn.Linear(config.hidden_size, self.all_head_size) + # non-parameterized matmuls will behave as normal torch.matmul ops unless + # Quantization-Aware-Training is invoked + self.attention_scores_matmul = QATMatMul() + self.context_layer_matmul = QATMatMul() + self.dropout = nn.Dropout(config.attention_probs_dropout_prob) self.position_embedding_type = position_embedding_type or getattr( config, "position_embedding_type", "absolute" @@ -320,7 +341,7 @@ def forward( past_key_value = (key_layer, value_layer) # Take the dot product between "query" and "key" to get the raw attention scores. - attention_scores = torch.matmul(query_layer, key_layer.transpose(-1, -2)) + attention_scores = self.attention_scores_matmul(query_layer, key_layer.transpose(-1, -2)) if self.position_embedding_type == "relative_key" or self.position_embedding_type == "relative_key_query": seq_length = hidden_states.size()[1] @@ -354,7 +375,7 @@ def forward( if head_mask is not None: attention_probs = attention_probs * head_mask - context_layer = torch.matmul(attention_probs, value_layer) + context_layer = self.context_layer_matmul(attention_probs, value_layer) context_layer = context_layer.permute(0, 2, 1, 3).contiguous() new_context_layer_shape = context_layer.size()[:-2] + (self.all_head_size,) diff --git a/src/transformers/utils/__init__.py b/src/transformers/utils/__init__.py index 2269f225485820..caa323cc5e40ed 100644 --- a/src/transformers/utils/__init__.py +++ b/src/transformers/utils/__init__.py @@ -165,6 +165,7 @@ CONFIG_NAME = "config.json" FEATURE_EXTRACTOR_NAME = "preprocessor_config.json" MODEL_CARD_NAME = "modelcard.json" +RECIPE_NAME = "recipe.yaml" SENTENCEPIECE_UNDERLINE = "▁" SPIECE_UNDERLINE = SENTENCEPIECE_UNDERLINE # Kept for backward compatibility From 46f78c11496c5814758d48794322d3b4288a5cd4 Mon Sep 17 00:00:00 2001 From: Konstantin Gulin <66528950+KSGulin@users.noreply.github.com> Date: Tue, 18 Oct 2022 16:22:42 +0200 Subject: [PATCH 5/6] Upgrade to transformers release V4.23.1 (#62) * Update trainer and model flows to accommodate sparseml Disable FP16 on QAT start (#12) * Override LRScheduler when using LRModifiers * Disable FP16 on QAT start * keep wrapped scaler object for training after disabling Using QATMatMul in DistilBERT model class (#41) Removed double quantization of output of context layer. (#45) Fix DataParallel validation forward signatures (#47) * Fix: DataParallel validation forward signatures * Update: generalize forward_fn selection Best model after epoch (#46) fix sclaer check for non fp16 mode in trainer (#38) Mobilebert QAT (#55) * Remove duplicate quantization of vocabulary. enable a QATWrapper for non-parameterized matmuls in BERT self attention (#9) * Utils and auxillary changes update Zoo stub loading for SparseZoo 1.1 refactor (#54) add flag to signal NM integration is active (#32) Add recipe_name to file names * Fix errors introduced in manual cherry-pick upgrade Co-authored-by: Benjamin Fineran --- src/transformers/hf_argparser.py | 47 +++++++++++++++++-- .../models/distilbert/modeling_distilbert.py | 42 +++++++++++++++-- .../models/mobilebert/modeling_mobilebert.py | 19 +++++++- src/transformers/trainer.py | 38 +++++++++++---- src/transformers/trainer_seq2seq.py | 2 +- src/transformers/utils/import_utils.py | 3 ++ 6 files changed, 133 insertions(+), 18 deletions(-) diff --git a/src/transformers/hf_argparser.py b/src/transformers/hf_argparser.py index 06a10ff5a0554b..fdae8a690f25d0 100644 --- a/src/transformers/hf_argparser.py +++ b/src/transformers/hf_argparser.py @@ -21,9 +21,16 @@ from inspect import isclass from pathlib import Path from typing import Any, Dict, Iterable, NewType, Optional, Tuple, Union, get_type_hints - +import os import yaml +from sparsezoo import Model + +from .utils.logging import get_logger + + +logger = get_logger(__name__) + DataClass = NewType("DataClass", Any) DataClassType = NewType("DataClassType", Any) @@ -229,12 +236,17 @@ def parse_args_into_dataclasses( # additional namespace. outputs.append(namespace) if return_remaining_strings: - return (*outputs, remaining_args) + return tuple( + *[_download_dataclass_zoo_stub_files(output) for output in outputs], + remaining_args, + ) else: if remaining_args: raise ValueError(f"Some specified arguments are not used by the HfArgumentParser: {remaining_args}") - return (*outputs,) + return tuple( + [_download_dataclass_zoo_stub_files(output) for output in outputs] + ) def parse_dict(self, args: Dict[str, Any], allow_extra_keys: bool = False) -> Tuple[DataClass, ...]: """ @@ -262,7 +274,9 @@ def parse_dict(self, args: Dict[str, Any], allow_extra_keys: bool = False) -> Tu outputs.append(obj) if not allow_extra_keys and unused_keys: raise ValueError(f"Some keys are not used by the HfArgumentParser: {sorted(unused_keys)}") - return tuple(outputs) + return tuple( + [_download_dataclass_zoo_stub_files(output) for output in outputs] + ) def parse_json_file(self, json_file: str, allow_extra_keys: bool = False) -> Tuple[DataClass, ...]: """ @@ -305,3 +319,28 @@ def parse_yaml_file(self, yaml_file: str, allow_extra_keys: bool = False) -> Tup """ outputs = self.parse_dict(yaml.safe_load(Path(yaml_file).read_text()), allow_extra_keys=allow_extra_keys) return tuple(outputs) + +def _download_dataclass_zoo_stub_files(data_class: DataClass): + for name, val in data_class.__dict__.items(): + if not isinstance(val, str) or "recipe" in name or not val.startswith("zoo:"): + continue + + logger.info(f"Downloading framework files for SparseZoo stub: {val}") + + zoo_model = Model(val) + framework_file_paths = [file.path for file in zoo_model.training.default.files] + assert framework_file_paths, "Unable to download any framework files for SparseZoo stub {val}" + framework_file_names = [os.path.basename(path) for path in framework_file_paths] + if "pytorch_model.bin" not in framework_file_names or ("config.json" not in framework_file_names): + raise RuntimeError( + "Unable to find 'pytorch_model.bin' and 'config.json' in framework " + f"files downloaded from {val}. Found {framework_file_names}. Check " + "if the given stub is for a transformers repo model" + ) + framework_dir_path = Path(framework_file_paths[0]).parent.absolute() + + logger.info(f"Overwriting argument {name} to downloaded {framework_dir_path}") + + data_class.__dict__[name] = str(framework_dir_path) + + return data_class diff --git a/src/transformers/models/distilbert/modeling_distilbert.py b/src/transformers/models/distilbert/modeling_distilbert.py index a2713128e901a3..dd31960cefdc97 100755 --- a/src/transformers/models/distilbert/modeling_distilbert.py +++ b/src/transformers/models/distilbert/modeling_distilbert.py @@ -89,6 +89,38 @@ def _create_sinusoidal_embeddings(n_pos: int, dim: int, out: torch.Tensor): out.detach_() +class QATAttentionScores(nn.Module): + def __init__(self): + super().__init__() + + # behaves like normal torch.matmul unless a SparseML QuantizationModifier + # is initialized + self.wrap_qat = True + self.qat_wrapper_kwargs = { + "num_inputs": 2, + "input_qconfigs": ["asymmetric", "symmetric"], + } + + def forward(self, a: torch.Tensor, b: torch.Tensor): + return torch.matmul(a, b) + +class QATContextLayer(nn.Module): + def __init__(self): + super().__init__() + + # behaves like normal torch.matmul unless a SparseML QuantizationModifier + # is initialized + self.wrap_qat = True + self.qat_wrapper_kwargs = { + "num_inputs": 2, + "num_outputs": 0, + "input_qconfigs": ["asymmetric", "symmetric"], + } + + def forward(self, a: torch.Tensor, b: torch.Tensor): + return torch.matmul(a, b) + + class Embeddings(nn.Module): def __init__(self, config: PretrainedConfig): super().__init__() @@ -150,6 +182,11 @@ def __init__(self, config: PretrainedConfig): self.pruned_heads: Set[int] = set() + # non-parameterized matmuls will behave as normal torch.matmul ops unless + # Quantization-Aware-Training is invoked + self.attention_scores_matmul = QATAttentionScores() + self.context_layer_matmul = QATContextLayer() + def prune_heads(self, heads: List[int]): attention_head_size = self.dim // self.n_heads if len(heads) == 0: @@ -207,7 +244,7 @@ def unshape(x: torch.Tensor) -> torch.Tensor: v = shape(self.v_lin(value)) # (bs, n_heads, k_length, dim_per_head) q = q / math.sqrt(dim_per_head) # (bs, n_heads, q_length, dim_per_head) - scores = torch.matmul(q, k.transpose(2, 3)) # (bs, n_heads, q_length, k_length) + scores = self.attention_scores_matmul(q, k.transpose(2, 3)) # (bs, n_heads, q_length, k_length) mask = (mask == 0).view(mask_reshp).expand_as(scores) # (bs, n_heads, q_length, k_length) scores = scores.masked_fill( mask, torch.tensor(torch.finfo(scores.dtype).min) @@ -220,7 +257,7 @@ def unshape(x: torch.Tensor) -> torch.Tensor: if head_mask is not None: weights = weights * head_mask - context = torch.matmul(weights, v) # (bs, n_heads, q_length, dim_per_head) + context = self.context_layer_matmul(weights, v) # (bs, n_heads, q_length, dim_per_head) context = unshape(context) # (bs, q_length, dim) context = self.out_lin(context) # (bs, q_length, dim) @@ -645,7 +682,6 @@ def forward( loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`. """ return_dict = return_dict if return_dict is not None else self.config.use_return_dict - dlbrt_output = self.distilbert( input_ids=input_ids, attention_mask=attention_mask, diff --git a/src/transformers/models/mobilebert/modeling_mobilebert.py b/src/transformers/models/mobilebert/modeling_mobilebert.py index 6bc306a6e05eb3..af7ceeee02a903 100644 --- a/src/transformers/models/mobilebert/modeling_mobilebert.py +++ b/src/transformers/models/mobilebert/modeling_mobilebert.py @@ -170,6 +170,23 @@ def forward(self, input_tensor: torch.Tensor) -> torch.Tensor: NORM2FN = {"layer_norm": nn.LayerNorm, "no_norm": NoNorm} +class QATEmbeddingTransformation(nn.Module): + def __init__(self, embedded_input_size, hidden_size): + super().__init__() + + # Behaves like normal Linear module unless a SparseML QuantizationModifier + # is initialized. + # When initialized, does not quantize inputs. + # Only weights are quantized (inputs come quantized from embeddings) + self.linear = nn.Linear(embedded_input_size, hidden_size) + self.wrap_qat = True + self.qat_wrapper_kwargs = { + "num_inputs": 0, + "num_outputs": 1, + } + + def forward(self, x: torch.Tensor): + return self.linear(x) class MobileBertEmbeddings(nn.Module): """Construct the embeddings from word, position and token_type embeddings.""" @@ -186,7 +203,7 @@ def __init__(self, config): embed_dim_multiplier = 3 if self.trigram_input else 1 embedded_input_size = self.embedding_size * embed_dim_multiplier - self.embedding_transformation = nn.Linear(embedded_input_size, config.hidden_size) + self.embedding_transformation = QATEmbeddingTransformation(embedded_input_size, config.hidden_size) self.LayerNorm = NORM2FN[config.normalization_type](config.hidden_size) self.dropout = nn.Dropout(config.hidden_dropout_prob) diff --git a/src/transformers/trainer.py b/src/transformers/trainer.py index 214e7a9789d2c7..abb967a2e216dd 100755 --- a/src/transformers/trainer.py +++ b/src/transformers/trainer.py @@ -1687,6 +1687,10 @@ def _inner_training_loop( _ = list(train_dataloader.sampler) for epoch in range(epochs_trained, num_train_epochs): + if self.use_cuda_amp and hasattr(self, "qat_active") and callable(self.qat_active) and self.qat_active(epoch): + logger.info("entering QAT phase, disabling FP16 training") + self.scaler._enabled = False + if isinstance(train_dataloader, DataLoader) and isinstance(train_dataloader.sampler, DistributedSampler): train_dataloader.sampler.set_epoch(epoch) elif hasattr(train_dataloader, "dataset") and isinstance(train_dataloader.dataset, IterableDatasetShard): @@ -2167,7 +2171,12 @@ def _save_checkpoint(self, model, trial, metrics=None): torch.save(self.scaler.state_dict(), os.path.join(output_dir, SCALER_NAME)) # Determine the new best metric / best model checkpoint - if metrics is not None and self.args.metric_for_best_model is not None: + if ( + metrics is not None + and self.args.metric_for_best_model is not None + and self.args.best_model_after_epoch is not None + and self.state.epoch > self.args.best_model_after_epoch + ): metric_to_check = self.args.metric_for_best_model if not metric_to_check.startswith("eval_"): metric_to_check = f"eval_{metric_to_check}" @@ -2421,14 +2430,14 @@ def _prepare_inputs(self, inputs: Dict[str, Union[torch.Tensor, Any]]) -> Dict[s return inputs - def compute_loss_context_manager(self): + def compute_loss_context_manager(self, enabled): """ A helper wrapper to group together context managers. """ return ContextManagers( [ self.torchdynamo_smart_context_manager(), - self.autocast_smart_context_manager(), + self.autocast_smart_context_manager(enabled=enabled), ] ) @@ -2438,7 +2447,7 @@ def torchdynamo_smart_context_manager(self): """ return self.ctx_manager_torchdynamo - def autocast_smart_context_manager(self): + def autocast_smart_context_manager(self, enabled): """ A helper wrapper that creates an appropriate context manager for `autocast` while feeding it the desired arguments, depending on the situation. @@ -2448,10 +2457,10 @@ def autocast_smart_context_manager(self): ctx_manager = ( torch.cpu.amp.autocast(dtype=self.amp_dtype) if self.use_cpu_amp - else torch.cuda.amp.autocast(dtype=self.amp_dtype) + else torch.cuda.amp.autocast(dtype=self.amp_dtype, enabled=enabled) ) else: - ctx_manager = torch.cuda.amp.autocast() + ctx_manager = torch.cuda.amp.autocast(enabled=enabled) else: ctx_manager = contextlib.nullcontext() if sys.version_info >= (3, 7) else contextlib.suppress() @@ -2482,7 +2491,7 @@ def training_step(self, model: nn.Module, inputs: Dict[str, Union[torch.Tensor, loss_mb = smp_forward_backward(model, inputs, self.args.gradient_accumulation_steps) return loss_mb.reduce_mean().detach().to(self.args.device) - with self.compute_loss_context_manager(): + with self.compute_loss_context_manager(enabled=hasattr(self, "scaler") and self.scaler.is_enabled()): loss = self.compute_loss(model, inputs) if self.args.n_gpu > 1: @@ -2939,7 +2948,14 @@ def evaluation_loop( observed_num_examples = 0 # Main evaluation loop + module_forward_fn = model.module.forward if isinstance(model, nn.DataParallel) else model.forward for step, inputs in enumerate(dataloader): + inputs = { + k: inputs[k] + for k in inputs + if k in list(inspect.signature(module_forward_fn).parameters.keys()) + } + # Update the observed num examples observed_batch_size = find_batch_size(inputs) if observed_batch_size is not None: @@ -3191,7 +3207,9 @@ def prediction_step( logits = smp_nested_concat(logits_mb) else: if has_labels: - with self.compute_loss_context_manager(): + with self.compute_loss_context_manager( + enabled=hasattr(self, "scaler") and self.scaler.is_enabled() + ): loss, outputs = self.compute_loss(model, inputs, return_outputs=True) loss = loss.mean().detach() @@ -3201,7 +3219,9 @@ def prediction_step( logits = outputs[1:] else: loss = None - with self.compute_loss_context_manager(): + with self.compute_loss_context_manager( + enabled=hasattr(self, "scaler") and self.scaler.is_enabled() + ): outputs = model(**inputs) if isinstance(outputs, dict): logits = tuple(v for k, v in outputs.items() if k not in ignore_keys) diff --git a/src/transformers/trainer_seq2seq.py b/src/transformers/trainer_seq2seq.py index 7689998c051b8f..39bbc10d4faad7 100644 --- a/src/transformers/trainer_seq2seq.py +++ b/src/transformers/trainer_seq2seq.py @@ -208,7 +208,7 @@ def prediction_step( generated_tokens = self._pad_tensors_to_max_len(generated_tokens, gen_kwargs["max_new_tokens"] + 1) with torch.no_grad(): - with self.compute_loss_context_manager(): + with self.compute_loss_context_manager(enabled=hasattr(self, "scaler") and self.scaler.is_enabled()): outputs = model(**inputs) if has_labels: if self.label_smoother is not None: diff --git a/src/transformers/utils/import_utils.py b/src/transformers/utils/import_utils.py index 140cfc78f4edd1..c73d58fb5db767 100644 --- a/src/transformers/utils/import_utils.py +++ b/src/transformers/utils/import_utils.py @@ -1016,6 +1016,9 @@ class _LazyModule(ModuleType): Module class that surfaces all objects but only performs associated imports when the objects are requested. """ + # flag to signal NM integration is active + NM_INTEGRATED = True + # Very heavily inspired by optuna.integration._IntegrationModule # https://github.com/optuna/optuna/blob/master/optuna/integration/__init__.py def __init__(self, name, module_file, import_structure, module_spec=None, extra_objects=None): From b9429eb1e0b04b08b046251fb16a71c99735d157 Mon Sep 17 00:00:00 2001 From: "dependabot[bot]" <49699333+dependabot[bot]@users.noreply.github.com> Date: Tue, 21 Feb 2023 15:36:48 +0000 Subject: [PATCH 6/6] Bump onnx in /examples/research_projects/decision_transformer Bumps [onnx](https://github.com/onnx/onnx) from 1.11.0 to 1.13.0. - [Release notes](https://github.com/onnx/onnx/releases) - [Changelog](https://github.com/onnx/onnx/blob/main/docs/Changelog.md) - [Commits](https://github.com/onnx/onnx/compare/v1.11.0...v1.13.0) --- updated-dependencies: - dependency-name: onnx dependency-type: direct:production ... Signed-off-by: dependabot[bot] --- .../research_projects/decision_transformer/requirements.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/examples/research_projects/decision_transformer/requirements.txt b/examples/research_projects/decision_transformer/requirements.txt index 6d393d15df2557..f479f379a2f0b9 100644 --- a/examples/research_projects/decision_transformer/requirements.txt +++ b/examples/research_projects/decision_transformer/requirements.txt @@ -119,7 +119,7 @@ nltk==3.7 numba==0.55.1 numpy==1.22.3 oauthlib==3.2.1 -onnx==1.11.0 +onnx==1.13.0 onnxconverter-common==1.9.0 opt-einsum==3.3.0 optax==0.1.1