CLIP (#11445)

* begin second draft

* fix import, style

* add loss

* fix embeds, logits_scale, and projection

* fix imports

* add conversion script

* add feature_extractor and processor

* style

* add tests for tokenizer, extractor and processor

* add vision model tests

* add weight init

* add more tests

* fix save_load  test

* model output, dosstrings, causal mask

* config doc

* add clip model tests

* return dict

* bigin integration test

* add integration tests

* fix-copies

* fix init

* Clip => CLIP

* fix module name

* docs

* fix doc

* output_dim => projection_dim

* fix checkpoint names

* remoe fast tokenizer file

* fix conversion script

* fix tests, quality

* put causal mask on device

* Apply suggestions from code review
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* fix attribute test

* style

* address sylvains comments

* style

* fix docstrings

* add qucik_gelu in activations, docstrings

* clean-up attention test

* fix act fun

* fix config

* fix torchscript tests

* even batch_size

* remove comment

* fix ouput tu_tuple

* fix save load tests

* fix add tokens test

* add fast tokenizer

* update copyright

* new processor API

* fix docs

* docstrings

* docs

* fix doc

* fix doc

* fix tokenizer

* fix import in doc example

* Apply suggestions from code review
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* check types of config

* valhalla => openai

* load image using url

* fix test

* typo
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

README.md

@@ -200,6 +200,7 @@ Current number of checkpoints: ![](https://img.shields.io/endpoint?url=https://h
 1. **[BlenderbotSmall](https://huggingface.co/transformers/model_doc/blenderbot_small.html)** (from Facebook) released with the paper [Recipes for building an open-domain chatbot](https://arxiv.org/abs/2004.13637) by Stephen Roller, Emily Dinan, Naman Goyal, Da Ju, Mary Williamson, Yinhan Liu, Jing Xu, Myle Ott, Kurt Shuster, Eric M. Smith, Y-Lan Boureau, Jason Weston.
 1. **[BORT](https://huggingface.co/transformers/model_doc/bort.html)** (from Alexa) released with the paper [Optimal Subarchitecture Extraction For BERT](https://arxiv.org/abs/2010.10499) by Adrian de Wynter and Daniel J. Perry.
 1. **[CamemBERT](https://huggingface.co/transformers/model_doc/camembert.html)** (from Inria/Facebook/Sorbonne) released with the paper [CamemBERT: a Tasty French Language Model](https://arxiv.org/abs/1911.03894) by Louis Martin*, Benjamin Muller*, Pedro Javier Ortiz Suárez*, Yoann Dupont, Laurent Romary, Éric Villemonte de la Clergerie, Djamé Seddah and Benoît Sagot.
+1. **[CLIP](https://huggingface.co/transformers/model_doc/camembert.html)** 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. **[ConvBERT](https://huggingface.co/transformers/model_doc/convbert.html)** (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. **[CPM](https://huggingface.co/transformers/model_doc/cpm.html)** (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.
 1. **[CTRL](https://huggingface.co/transformers/model_doc/ctrl.html)** (from Salesforce) released with the paper [CTRL: A Conditional Transformer Language Model for Controllable Generation](https://arxiv.org/abs/1909.05858) by Nitish Shirish Keskar*, Bryan McCann*, Lav R. Varshney, Caiming Xiong and Richard Socher.

docs/source/index.rst

@@ -114,142 +114,146 @@ conversion utilities for the following models:
 11. :doc:`CamemBERT <model_doc/camembert>` (from Inria/Facebook/Sorbonne) released with the paper `CamemBERT: a Tasty
     French Language Model <https://arxiv.org/abs/1911.03894>`__ by Louis Martin*, Benjamin Muller*, Pedro Javier Ortiz
     Suárez*, Yoann Dupont, Laurent Romary, Éric Villemonte de la Clergerie, Djamé Seddah and Benoît Sagot.
-12. :doc:`ConvBERT <model_doc/convbert>` (from YituTech) released with the paper `ConvBERT: Improving BERT with
+12. :doc:`CLIP <model_doc/camembert>` 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.
+13. :doc:`ConvBERT <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.
-13. :doc:`CPM <model_doc/cpm>` (from Tsinghua University) released with the paper `CPM: A Large-scale Generative
+14. :doc:`CPM <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.
-14. :doc:`CTRL <model_doc/ctrl>` (from Salesforce) released with the paper `CTRL: A Conditional Transformer Language
+15. :doc:`CTRL <model_doc/ctrl>` (from Salesforce) released with the paper `CTRL: A Conditional Transformer Language
     Model for Controllable Generation <https://arxiv.org/abs/1909.05858>`__ by Nitish Shirish Keskar*, Bryan McCann*,
     Lav R. Varshney, Caiming Xiong and Richard Socher.
-15. :doc:`DeBERTa <model_doc/deberta>` (from Microsoft) released with the paper `DeBERTa: Decoding-enhanced BERT with
+16. :doc:`DeBERTa <model_doc/deberta>` (from Microsoft) released with the paper `DeBERTa: Decoding-enhanced BERT with
     Disentangled Attention <https://arxiv.org/abs/2006.03654>`__ by Pengcheng He, Xiaodong Liu, Jianfeng Gao, Weizhu
     Chen.
-16. :doc:`DeBERTa-v2 <model_doc/deberta_v2>` (from Microsoft) released with the paper `DeBERTa: Decoding-enhanced BERT
+17. :doc:`DeBERTa-v2 <model_doc/deberta_v2>` (from Microsoft) released with the paper `DeBERTa: Decoding-enhanced BERT
     with Disentangled Attention <https://arxiv.org/abs/2006.03654>`__ by Pengcheng He, Xiaodong Liu, Jianfeng Gao,
     Weizhu Chen.
-17. :doc:`DeiT <model_doc/deit>` (from Facebook) released with the paper `Training data-efficient image transformers &
+18. :doc:`DeiT <model_doc/deit>` (from Facebook) released with the paper `Training data-efficient image transformers &
     distillation through attention <https://arxiv.org/abs/2012.12877>`__ by Hugo Touvron, Matthieu Cord, Matthijs
     Douze, Francisco Massa, Alexandre Sablayrolles, Hervé Jégou.
-18. :doc:`DialoGPT <model_doc/dialogpt>` (from Microsoft Research) released with the paper `DialoGPT: Large-Scale
+19. :doc:`DialoGPT <model_doc/dialogpt>` (from Microsoft Research) released with the paper `DialoGPT: Large-Scale
     Generative Pre-training for Conversational Response Generation <https://arxiv.org/abs/1911.00536>`__ by Yizhe
     Zhang, Siqi Sun, Michel Galley, Yen-Chun Chen, Chris Brockett, Xiang Gao, Jianfeng Gao, Jingjing Liu, Bill Dolan.
-19. :doc:`DistilBERT <model_doc/distilbert>` (from HuggingFace), released together with the paper `DistilBERT, a
+20. :doc:`DistilBERT <model_doc/distilbert>` (from HuggingFace), released together with the paper `DistilBERT, a
     distilled version of BERT: smaller, faster, cheaper and lighter <https://arxiv.org/abs/1910.01108>`__ by Victor
     Sanh, Lysandre Debut and Thomas Wolf. The same method has been applied to compress GPT2 into `DistilGPT2
     <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__, RoBERTa into `DistilRoBERTa
     <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__, Multilingual BERT into
     `DistilmBERT <https://github.com/huggingface/transformers/tree/master/examples/distillation>`__ and a German
     version of DistilBERT.
-20. :doc:`DPR <model_doc/dpr>` (from Facebook) released with the paper `Dense Passage Retrieval for Open-Domain
+21. :doc:`DPR <model_doc/dpr>` (from Facebook) released with the paper `Dense Passage Retrieval for Open-Domain
     Question Answering <https://arxiv.org/abs/2004.04906>`__ by Vladimir Karpukhin, Barlas Oğuz, Sewon Min, Patrick
     Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, and Wen-tau Yih.
-21. :doc:`ELECTRA <model_doc/electra>` (from Google Research/Stanford University) released with the paper `ELECTRA:
+22. :doc:`ELECTRA <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.
-22. :doc:`FlauBERT <model_doc/flaubert>` (from CNRS) released with the paper `FlauBERT: Unsupervised Language Model
+23. :doc:`FlauBERT <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.
-23. :doc:`Funnel Transformer <model_doc/funnel>` (from CMU/Google Brain) released with the paper `Funnel-Transformer:
+24. :doc:`Funnel Transformer <model_doc/funnel>` (from CMU/Google Brain) released with the paper `Funnel-Transformer:
     Filtering out Sequential Redundancy for Efficient Language Processing <https://arxiv.org/abs/2006.03236>`__ by
     Zihang Dai, Guokun Lai, Yiming Yang, Quoc V. Le.
-24. :doc:`GPT <model_doc/gpt>` (from OpenAI) released with the paper `Improving Language Understanding by Generative
+25. :doc:`GPT <model_doc/gpt>` (from OpenAI) released with the paper `Improving Language Understanding by Generative
     Pre-Training <https://blog.openai.com/language-unsupervised/>`__ by Alec Radford, Karthik Narasimhan, Tim Salimans
     and Ilya Sutskever.
-25. :doc:`GPT-2 <model_doc/gpt2>` (from OpenAI) released with the paper `Language Models are Unsupervised Multitask
+26. :doc:`GPT-2 <model_doc/gpt2>` (from OpenAI) released with the paper `Language Models are Unsupervised Multitask
     Learners <https://blog.openai.com/better-language-models/>`__ by Alec Radford*, Jeffrey Wu*, Rewon Child, David
     Luan, Dario Amodei** and Ilya Sutskever**.
-26. :doc:`GPT Neo <model_doc/gpt_neo>` (from EleutherAI) released in the repository `EleutherAI/gpt-neo
+27. :doc:`GPT Neo <model_doc/gpt_neo>` (from EleutherAI) released in the repository `EleutherAI/gpt-neo
     <https://github.com/EleutherAI/gpt-neo>`__ by Sid Black, Stella Biderman, Leo Gao, Phil Wang and Connor Leahy.
-27. :doc:`I-BERT <model_doc/ibert>` (from Berkeley) released with the paper `I-BERT: Integer-only BERT Quantization
+28. :doc:`I-BERT <model_doc/ibert>` (from Berkeley) released with the paper `I-BERT: Integer-only BERT Quantization
     <https://arxiv.org/abs/2101.01321>`__ by Sehoon Kim, Amir Gholami, Zhewei Yao, Michael W. Mahoney, Kurt Keutzer
-28. :doc:`LayoutLM <model_doc/layoutlm>` (from Microsoft Research Asia) released with the paper `LayoutLM: Pre-training
+29. :doc:`LayoutLM <model_doc/layoutlm>` (from Microsoft Research Asia) released with the paper `LayoutLM: Pre-training
     of Text and Layout for Document Image Understanding <https://arxiv.org/abs/1912.13318>`__ by Yiheng Xu, Minghao Li,
     Lei Cui, Shaohan Huang, Furu Wei, Ming Zhou.
-29. :doc:`LED <model_doc/led>` (from AllenAI) released with the paper `Longformer: The Long-Document Transformer
+30. :doc:`LED <model_doc/led>` (from AllenAI) released with the paper `Longformer: The Long-Document Transformer
     <https://arxiv.org/abs/2004.05150>`__ by Iz Beltagy, Matthew E. Peters, Arman Cohan.
-30. :doc:`Longformer <model_doc/longformer>` (from AllenAI) released with the paper `Longformer: The Long-Document
+31. :doc:`Longformer <model_doc/longformer>` (from AllenAI) released with the paper `Longformer: The Long-Document
     Transformer <https://arxiv.org/abs/2004.05150>`__ by Iz Beltagy, Matthew E. Peters, Arman Cohan.
-31. :doc:`LUKE <model_doc/luke>` (from Studio Ousia) released with the paper `LUKE: Deep Contextualized Entity
+32. :doc:`LUKE <model_doc/luke>` (from Studio Ousia) released with the paper `LUKE: Deep Contextualized Entity
     Representations with Entity-aware Self-attention <https://arxiv.org/abs/2010.01057>`__ by Ikuya Yamada, Akari Asai,
     Hiroyuki Shindo, Hideaki Takeda, Yuji Matsumoto.
-32. :doc:`LXMERT <model_doc/lxmert>` (from UNC Chapel Hill) released with the paper `LXMERT: Learning Cross-Modality
+33. :doc:`LXMERT <model_doc/lxmert>` (from UNC Chapel Hill) released with the paper `LXMERT: Learning Cross-Modality
     Encoder Representations from Transformers for Open-Domain Question Answering <https://arxiv.org/abs/1908.07490>`__
     by Hao Tan and Mohit Bansal.
-33. :doc:`M2M100 <model_doc/m2m_100>` (from Facebook) released with the paper `Beyond English-Centric Multilingual
+34. :doc:`M2M100 <model_doc/m2m_100>` (from Facebook) released with the paper `Beyond English-Centric Multilingual
     Machine Translation <https://arxiv.org/abs/2010.11125>`__ by 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.
-34. :doc:`MarianMT <model_doc/marian>` Machine translation models trained using `OPUS <http://opus.nlpl.eu/>`__ data by
+35. :doc:`MarianMT <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.
-35. :doc:`MBart <model_doc/mbart>` (from Facebook) released with the paper `Multilingual Denoising Pre-training for
+36. :doc:`MBart <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.
-36. :doc:`MBart-50 <model_doc/mbart>` (from Facebook) released with the paper `Multilingual Translation with Extensible
+37. :doc:`MBart-50 <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.
-37. :doc:`Megatron-BERT <model_doc/megatron_bert>` (from NVIDIA) released with the paper `Megatron-LM: Training
+38. :doc:`Megatron-BERT <model_doc/megatron_bert>` (from NVIDIA) released with the paper `Megatron-LM: Training
     Multi-Billion Parameter Language Models Using Model Parallelism <https://arxiv.org/abs/1909.08053>`__ by Mohammad
     Shoeybi, Mostofa Patwary, Raul Puri, Patrick LeGresley, Jared Casper and Bryan Catanzaro.
-38. :doc:`Megatron-GPT2 <model_doc/megatron_gpt2>` (from NVIDIA) released with the paper `Megatron-LM: Training
+39. :doc:`Megatron-GPT2 <model_doc/megatron_gpt2>` (from NVIDIA) released with the paper `Megatron-LM: Training
     Multi-Billion Parameter Language Models Using Model Parallelism <https://arxiv.org/abs/1909.08053>`__ by Mohammad
     Shoeybi, Mostofa Patwary, Raul Puri, Patrick LeGresley, Jared Casper and Bryan Catanzaro.
-39. :doc:`MPNet <model_doc/mpnet>` (from Microsoft Research) released with the paper `MPNet: Masked and Permuted
+40. :doc:`MPNet <model_doc/mpnet>` (from Microsoft Research) released with the paper `MPNet: Masked and Permuted
     Pre-training for Language Understanding <https://arxiv.org/abs/2004.09297>`__ by Kaitao Song, Xu Tan, Tao Qin,
     Jianfeng Lu, Tie-Yan Liu.
-40. :doc:`MT5 <model_doc/mt5>` (from Google AI) released with the paper `mT5: A massively multilingual pre-trained
+41. :doc:`MT5 <model_doc/mt5>` (from Google AI) released with the paper `mT5: A massively multilingual pre-trained
     text-to-text transformer <https://arxiv.org/abs/2010.11934>`__ by Linting Xue, Noah Constant, Adam Roberts, Mihir
     Kale, Rami Al-Rfou, Aditya Siddhant, Aditya Barua, Colin Raffel.
-41. :doc:`Pegasus <model_doc/pegasus>` (from Google) released with the paper `PEGASUS: Pre-training with Extracted
+42. :doc:`Pegasus <model_doc/pegasus>` (from Google) released with the paper `PEGASUS: Pre-training with Extracted
     Gap-sentences for Abstractive Summarization <https://arxiv.org/abs/1912.08777>`__> by Jingqing Zhang, Yao Zhao,
     Mohammad Saleh and Peter J. Liu.
-42. :doc:`ProphetNet <model_doc/prophetnet>` (from Microsoft Research) released with the paper `ProphetNet: Predicting
+43. :doc:`ProphetNet <model_doc/prophetnet>` (from Microsoft Research) released with the paper `ProphetNet: Predicting
     Future N-gram for Sequence-to-Sequence Pre-training <https://arxiv.org/abs/2001.04063>`__ by Yu Yan, Weizhen Qi,
     Yeyun Gong, Dayiheng Liu, Nan Duan, Jiusheng Chen, Ruofei Zhang and Ming Zhou.
-43. :doc:`Reformer <model_doc/reformer>` (from Google Research) released with the paper `Reformer: The Efficient
+44. :doc:`Reformer <model_doc/reformer>` (from Google Research) released with the paper `Reformer: The Efficient
     Transformer <https://arxiv.org/abs/2001.04451>`__ by Nikita Kitaev, Łukasz Kaiser, Anselm Levskaya.
-44. :doc:`RoBERTa <model_doc/roberta>` (from Facebook), released together with the paper a `Robustly Optimized BERT
+45. :doc:`RoBERTa <model_doc/roberta>` (from Facebook), released together with the paper a `Robustly Optimized BERT
     Pretraining Approach <https://arxiv.org/abs/1907.11692>`__ by Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar
     Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, Veselin Stoyanov.
-45. :doc:`SpeechToTextTransformer <model_doc/speech_to_text>` (from Facebook), released together with the paper
+46. :doc:`SpeechToTextTransformer <model_doc/speech_to_text>` (from Facebook), released together 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, Dmytro Okhonko, Juan Pino.
-46. :doc:`SqueezeBert <model_doc/squeezebert>` released with the paper `SqueezeBERT: What can computer vision teach NLP
+47. :doc:`SqueezeBert <model_doc/squeezebert>` released with the paper `SqueezeBERT: What can computer vision teach NLP
     about efficient neural networks? <https://arxiv.org/abs/2006.11316>`__ by Forrest N. Iandola, Albert E. Shaw, Ravi
     Krishna, and Kurt W. Keutzer.
-47. :doc:`T5 <model_doc/t5>` (from Google AI) released with the paper `Exploring the Limits of Transfer Learning with a
+48. :doc:`T5 <model_doc/t5>` (from Google AI) released with the paper `Exploring the Limits of Transfer Learning with a
     Unified Text-to-Text Transformer <https://arxiv.org/abs/1910.10683>`__ 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.
-48. :doc:`TAPAS <model_doc/tapas>` (from Google AI) released with the paper `TAPAS: Weakly Supervised Table Parsing via
+49. :doc:`TAPAS <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.
-49. :doc:`Transformer-XL <model_doc/transformerxl>` (from Google/CMU) released with the paper `Transformer-XL:
+50. :doc:`Transformer-XL <model_doc/transformerxl>` (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.
-50. :doc:`Vision Transformer (ViT) <model_doc/vit>` (from Google AI) released with the paper `An Image is Worth 16x16
+51. :doc:`Vision Transformer (ViT) <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.
-51. :doc:`Wav2Vec2 <model_doc/wav2vec2>` (from Facebook AI) released with the paper `wav2vec 2.0: A Framework for
+52. :doc:`Wav2Vec2 <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.
-52. :doc:`XLM <model_doc/xlm>` (from Facebook) released together with the paper `Cross-lingual Language Model
+53. :doc:`XLM <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.
-53. :doc:`XLM-ProphetNet <model_doc/xlmprophetnet>` (from Microsoft Research) released with the paper `ProphetNet:
+54. :doc:`XLM-ProphetNet <model_doc/xlmprophetnet>` (from Microsoft Research) released with the paper `ProphetNet:
     Predicting Future N-gram for Sequence-to-Sequence Pre-training <https://arxiv.org/abs/2001.04063>`__ by Yu Yan,
     Weizhen Qi, Yeyun Gong, Dayiheng Liu, Nan Duan, Jiusheng Chen, Ruofei Zhang and Ming Zhou.
-54. :doc:`XLM-RoBERTa <model_doc/xlmroberta>` (from Facebook AI), released together with the paper `Unsupervised
+55. :doc:`XLM-RoBERTa <model_doc/xlmroberta>` (from Facebook AI), released together with the paper `Unsupervised
     Cross-lingual Representation Learning at Scale <https://arxiv.org/abs/1911.02116>`__ by Alexis Conneau*, Kartikay
     Khandelwal*, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke
     Zettlemoyer and Veselin Stoyanov.
-55. :doc:`XLNet <model_doc/xlnet>` (from Google/CMU) released with the paper `​XLNet: Generalized Autoregressive
+56. :doc:`XLNet <model_doc/xlnet>` (from Google/CMU) released with the paper `​XLNet: Generalized Autoregressive
     Pretraining for Language Understanding <https://arxiv.org/abs/1906.08237>`__ by Zhilin Yang*, Zihang Dai*, Yiming
     Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le.
-56. :doc:`XLSR-Wav2Vec2 <model_doc/xlsr_wav2vec2>` (from Facebook AI) released with the paper `Unsupervised
+57. :doc:`XLSR-Wav2Vec2 <model_doc/xlsr_wav2vec2>` (from Facebook AI) released with the paper `Unsupervised
     Cross-Lingual Representation Learning For Speech Recognition <https://arxiv.org/abs/2006.13979>`__ by Alexis
     Conneau, Alexei Baevski, Ronan Collobert, Abdelrahman Mohamed, Michael Auli.
 
@@ -284,6 +288,8 @@ Flax), PyTorch, and/or TensorFlow.
 +-----------------------------+----------------+----------------+-----------------+--------------------+--------------+
 |       BlenderbotSmall       |       ✅       |       ❌       |       ✅        |         ✅         |      ❌      |
 +-----------------------------+----------------+----------------+-----------------+--------------------+--------------+
+|            CLIP             |       ✅       |       ✅       |       ✅        |         ❌         |      ❌      |
++-----------------------------+----------------+----------------+-----------------+--------------------+--------------+
 |            CTRL             |       ✅       |       ❌       |       ✅        |         ✅         |      ❌      |
 +-----------------------------+----------------+----------------+-----------------+--------------------+--------------+
 |          CamemBERT          |       ✅       |       ✅       |       ✅        |         ✅         |      ❌      |
@@ -461,6 +467,7 @@ Flax), PyTorch, and/or TensorFlow.
     model_doc/blenderbot_small
     model_doc/bort
     model_doc/camembert
+    model_doc/clip
     model_doc/convbert
     model_doc/cpm
     model_doc/ctrl

docs/source/model_doc/clip.rst

+.. 
+    Copyright 2021 The HuggingFace Team. All rights reserved.
+
+    Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
+    the License. You may obtain a copy of the License at
+
+        http://www.apache.org/licenses/LICENSE-2.0
+
+    Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
+    an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
+    specific language governing permissions and limitations under the License.
+
+CLIP
+-----------------------------------------------------------------------------------------------------------------------
+
+Overview
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The CLIP model was proposed in `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. CLIP
+(Contrastive Language-Image Pre-Training) is a neural network trained on a variety of (image, text) pairs. It can be
+instructed in natural language to predict the most relevant text snippet, given an image, without directly optimizing
+for the task, similarly to the zero-shot capabilities of GPT-2 and 3.
+
+The abstract from the paper is the following:
+
+*State-of-the-art computer vision systems are trained to predict a fixed set of predetermined object categories. This
+restricted form of supervision limits their generality and usability since additional labeled data is needed to specify
+any other visual concept. Learning directly from raw text about images is a promising alternative which leverages a
+much broader source of supervision. We demonstrate that the simple pre-training task of predicting which caption goes
+with which image is an efficient and scalable way to learn SOTA image representations from scratch on a dataset of 400
+million (image, text) pairs collected from the internet. After pre-training, natural language is used to reference
+learned visual concepts (or describe new ones) enabling zero-shot transfer of the model to downstream tasks. We study
+the performance of this approach by benchmarking on over 30 different existing computer vision datasets, spanning tasks
+such as OCR, action recognition in videos, geo-localization, and many types of fine-grained object classification. The
+model transfers non-trivially to most tasks and is often competitive with a fully supervised baseline without the need
+for any dataset specific training. For instance, we match the accuracy of the original ResNet-50 on ImageNet zero-shot
+without needing to use any of the 1.28 million training examples it was trained on. We release our code and pre-trained
+model weights at this https URL.*
+
+Usage
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+CLIP is a multi-modal vision and language model. It can be used for image-text similarity and for zero-shot image
+classification. CLIP uses a ViT like transformer to get visual features and a causal language model to get the text
+features. Both the text and visual features are then projected to a latent space with identical dimension. The dot
+product between the projected image and text features is then used as a similar score.
+
+To feed images to the Transformer encoder, each image is split into a sequence of fixed-size non-overlapping patches,
+which are then linearly embedded. A [CLS] token is added to serve as representation of an entire image. The authors
+also add absolute position embeddings, and feed the resulting sequence of vectors to a standard Transformer encoder.
+The :class:`~transformers.CLIPFeatureExtractor` can be used to resize (or rescale) and normalize images for the model.
+
+The :class:`~transformers.CLIPTokenizer` is used to encode the text. The :class:`~transformers.CLIPProcessor` wraps
+:class:`~transformers.CLIPFeatureExtractor` and :class:`~transformers.CLIPTokenizer` into a single instance to both
+encode the text and prepare the images. The following example shows how to get the image-text similarity scores using
+:class:`~transformers.CLIPProcessor` and :class:`~transformers.CLIPModel`.
+
+
+.. code-block::
+
+        >>> import torch
+        >>> from PIL import Image
+        >>> import requests
+
+        >>> from transformers import CLIPProcessor, CLIPModel
+
+        >>> model = CLIPModel.from_pretrained("openai/clip-vit-base-patch32")
+        >>> processor = CLIPProcessor.from_pretrained("openai/clip-vit-base-patch32")
+
+        >>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
+        >>> image = Image.open(requests.get(url, stream=True).raw)
+
+        >>> inputs = processor(text=["a photo of a cat", "a photo of a dog"], images=image, return_tensors="pt", padding=True)
+
+        >>> outputs = model(**inputs)
+        >>> logits_per_image = outputs.logits_per_image # this is the image-text similarity score
+        >>> probs = logits_per_image.softmax(dim=1) # we can take the softmax to get the label probabilities
+
+
+This model was contributed by `valhalla <https://huggingface.co/valhalla>`__. The original code can be found `here
+<https://github.com/openai/CLIP>`__.
+
+CLIPConfig
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CLIPConfig
+    :members: from_text_vision_configs
+
+
+CLIPTextConfig
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CLIPTextConfig
+    :members:
+
+
+CLIPVisionConfig
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CLIPVisionConfig
+    :members:
+
+
+
+CLIPTokenizer
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CLIPTokenizer
+    :members: build_inputs_with_special_tokens, get_special_tokens_mask,
+        create_token_type_ids_from_sequences, save_vocabulary
+
+CLIPTokenizerFast
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CLIPTokenizerFast
+    :members:
+
+
+CLIPFeatureExtractor
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CLIPFeatureExtractor
+    :members:
+
+
+CLIPProcessor
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CLIPProcessor
+    :members:
+
+
+
+CLIPModel
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CLIPModel
+    :members: forward, get_text_features, get_image_features
+
+
+CLIPTextModel
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CLIPTextModel
+    :members: forward
+
+
+CLIPVisionModel
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.CLIPVisionModel
+    :members: forward

src/transformers/__init__.py

@@ -166,6 +166,13 @@ _import_structure = {
         "BlenderbotSmallTokenizer",
     ],
     "models.camembert": ["CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP", "CamembertConfig"],
+    "models.clip": [
+        "CLIP_PRETRAINED_CONFIG_ARCHIVE_MAP",
+        "CLIPConfig",
+        "CLIPTextConfig",
+        "CLIPTokenizer",
+        "CLIPVisionConfig",
+    ],
     "models.convbert": ["CONVBERT_PRETRAINED_CONFIG_ARCHIVE_MAP", "ConvBertConfig", "ConvBertTokenizer"],
     "models.cpm": ["CpmTokenizer"],
     "models.ctrl": ["CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP", "CTRLConfig", "CTRLTokenizer"],
@@ -315,6 +322,7 @@ else:
 # tokenizers-backed objects
 if is_tokenizers_available():
     # Fast tokenizers
+    _import_structure["models.clip"].append("CLIPTokenizerFast")
     _import_structure["models.convbert"].append("ConvBertTokenizerFast")
     _import_structure["models.albert"].append("AlbertTokenizerFast")
     _import_structure["models.bart"].append("BartTokenizerFast")
@@ -390,6 +398,8 @@ else:
 # Vision-specific objects
 if is_vision_available():
     _import_structure["image_utils"] = ["ImageFeatureExtractionMixin"]
+    _import_structure["models.clip"].append("CLIPFeatureExtractor")
+    _import_structure["models.clip"].append("CLIPProcessor")
     _import_structure["models.deit"].append("DeiTFeatureExtractor")
     _import_structure["models.vit"].append("ViTFeatureExtractor")
 else:
@@ -498,6 +508,7 @@ if is_torch_available():
             "AutoModelWithLMHead",
         ]
     )
+
     _import_structure["models.bart"].extend(
         [
             "BART_PRETRAINED_MODEL_ARCHIVE_LIST",
@@ -588,6 +599,15 @@ if is_torch_available():
             "CamembertModel",
         ]
     )
+    _import_structure["models.clip"].extend(
+        [
+            "CLIP_PRETRAINED_MODEL_ARCHIVE_LIST",
+            "CLIPModel",
+            "CLIPPreTrainedModel",
+            "CLIPTextModel",
+            "CLIPVisionModel",
+        ]
+    )
     _import_structure["models.convbert"].extend(
         [
             "CONVBERT_PRETRAINED_MODEL_ARCHIVE_LIST",
@@ -1566,6 +1586,13 @@ if TYPE_CHECKING:
         BlenderbotSmallTokenizer,
     )
     from .models.camembert import CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP, CamembertConfig
+    from .models.clip import (
+        CLIP_PRETRAINED_CONFIG_ARCHIVE_MAP,
+        CLIPConfig,
+        CLIPTextConfig,
+        CLIPTokenizer,
+        CLIPVisionConfig,
+    )
     from .models.convbert import CONVBERT_PRETRAINED_CONFIG_ARCHIVE_MAP, ConvBertConfig, ConvBertTokenizer
     from .models.cpm import CpmTokenizer
     from .models.ctrl import CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP, CTRLConfig, CTRLTokenizer
@@ -1715,6 +1742,7 @@ if TYPE_CHECKING:
         from .models.bert import BertTokenizerFast
         from .models.big_bird import BigBirdTokenizerFast
         from .models.camembert import CamembertTokenizerFast
+        from .models.clip import CLIPTokenizerFast
         from .models.convbert import ConvBertTokenizerFast
         from .models.deberta import DebertaTokenizerFast
         from .models.distilbert import DistilBertTokenizerFast
@@ -1763,6 +1791,7 @@ if TYPE_CHECKING:
 
     if is_vision_available():
         from .image_utils import ImageFeatureExtractionMixin
+        from .models.clip import CLIPFeatureExtractor, CLIPProcessor
         from .models.deit import DeiTFeatureExtractor
         from .models.vit import ViTFeatureExtractor
     else:
@@ -1936,6 +1965,13 @@ if TYPE_CHECKING:
             CamembertForTokenClassification,
             CamembertModel,
         )
+        from .models.clip import (
+            CLIP_PRETRAINED_MODEL_ARCHIVE_LIST,
+            CLIPModel,
+            CLIPPreTrainedModel,
+            CLIPTextModel,
+            CLIPVisionModel,
+        )
         from .models.convbert import (
             CONVBERT_PRETRAINED_MODEL_ARCHIVE_LIST,
             ConvBertForMaskedLM,

src/transformers/activations.py

@@ -52,6 +52,10 @@ def gelu_fast(x):
     return 0.5 * x * (1.0 + torch.tanh(x * 0.7978845608 * (1.0 + 0.044715 * x * x)))
 
 
+def quick_gelu(x):
+    return x * torch.sigmoid(1.702 * x)
+
+
 def _silu_python(x):
     """
     See Gaussian Error Linear Units (Hendrycks et al., https://arxiv.org/abs/1606.08415) where the SiLU (Sigmoid Linear
@@ -85,6 +89,7 @@ ACT2FN = {
     "tanh": torch.tanh,
     "gelu_new": gelu_new,
     "gelu_fast": gelu_fast,
+    "quick_gelu": quick_gelu,
     "mish": mish,
     "linear": linear_act,
     "sigmoid": torch.sigmoid,

src/transformers/convert_slow_tokenizer.py

@@ -701,6 +701,29 @@ class BigBirdConverter(SpmConverter):
         )
 
 
+class CLIPConverter(Converter):
+    def converted(self) -> Tokenizer:
+        vocab = self.original_tokenizer.encoder
+        merges = list(self.original_tokenizer.bpe_ranks.keys())
+
+        tokenizer = Tokenizer(
+            BPE(
+                vocab=vocab,
+                merges=merges,
+                dropout=None,
+                continuing_subword_prefix="",
+                end_of_word_suffix="</w>",
+                fuse_unk=False,
+            )
+        )
+
+        tokenizer.pre_tokenizer = pre_tokenizers.ByteLevel(add_prefix_space=self.original_tokenizer.add_prefix_space)
+        tokenizer.decoder = decoders.ByteLevel()
+        tokenizer.post_processor = processors.ByteLevel(trim_offsets=False)
+
+        return tokenizer
+
+
 SLOW_TO_FAST_CONVERTERS = {
     "AlbertTokenizer": AlbertConverter,
     "BartTokenizer": RobertaConverter,
@@ -708,6 +731,7 @@ SLOW_TO_FAST_CONVERTERS = {
     "BertTokenizer": BertConverter,
     "BigBirdTokenizer": BigBirdConverter,
     "CamembertTokenizer": CamembertConverter,
+    "CLIPTokenizer": CLIPConverter,
     "ConvBertTokenizer": BertConverter,
     "DebertaTokenizer": DebertaConverter,
     "DistilBertTokenizer": BertConverter,

src/transformers/models/__init__.py

@@ -30,6 +30,7 @@ from . import (
     blenderbot,
     blenderbot_small,
     camembert,
+    clip,
     convbert,
     cpm,
     ctrl,

src/transformers/models/auto/configuration_auto.py

@@ -33,6 +33,7 @@ from ..blenderbot_small.configuration_blenderbot_small import (
     BlenderbotSmallConfig,
 )
 from ..camembert.configuration_camembert import CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP, CamembertConfig
+from ..clip.configuration_clip import CLIP_PRETRAINED_CONFIG_ARCHIVE_MAP, CLIPConfig
 from ..convbert.configuration_convbert import CONVBERT_PRETRAINED_CONFIG_ARCHIVE_MAP, ConvBertConfig
 from ..ctrl.configuration_ctrl import CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP, CTRLConfig
 from ..deberta.configuration_deberta import DEBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP, DebertaConfig
@@ -90,6 +91,7 @@ ALL_PRETRAINED_CONFIG_ARCHIVE_MAP = dict(
     (key, value)
     for pretrained_map in [
         # Add archive maps here
+        CLIP_PRETRAINED_CONFIG_ARCHIVE_MAP,
         BIGBIRD_PEGASUS_PRETRAINED_CONFIG_ARCHIVE_MAP,
         DEIT_PRETRAINED_CONFIG_ARCHIVE_MAP,
         LUKE_PRETRAINED_CONFIG_ARCHIVE_MAP,
@@ -144,6 +146,7 @@ ALL_PRETRAINED_CONFIG_ARCHIVE_MAP = dict(
 CONFIG_MAPPING = OrderedDict(
     [
         # Add configs here
+        ("clip", CLIPConfig),
         ("bigbird_pegasus", BigBirdPegasusConfig),
         ("deit", DeiTConfig),
         ("luke", LukeConfig),
@@ -204,6 +207,7 @@ CONFIG_MAPPING = OrderedDict(
 MODEL_NAMES_MAPPING = OrderedDict(
     [
         # Add full (and cased) model names here
+        ("clip", "CLIP"),
         ("bigbird_pegasus", "BigBirdPegasus"),
         ("deit", "DeiT"),
         ("luke", "LUKE"),

src/transformers/models/auto/modeling_auto.py

@@ -81,6 +81,7 @@ from ..camembert.modeling_camembert import (
     CamembertForTokenClassification,
     CamembertModel,
 )
+from ..clip.modeling_clip import CLIPModel
 from ..convbert.modeling_convbert import (
     ConvBertForMaskedLM,
     ConvBertForMultipleChoice,
@@ -299,6 +300,7 @@ from .configuration_auto import (
     BlenderbotConfig,
     BlenderbotSmallConfig,
     CamembertConfig,
+    CLIPConfig,
     ConvBertConfig,
     CTRLConfig,
     DebertaConfig,
@@ -352,6 +354,7 @@ logger = logging.get_logger(__name__)
 MODEL_MAPPING = OrderedDict(
     [
         # Base model mapping
+        (CLIPConfig, CLIPModel),
         (BigBirdPegasusConfig, BigBirdPegasusModel),
         (DeiTConfig, DeiTModel),
         (LukeConfig, LukeModel),

src/transformers/models/clip/__init__.py

+# flake8: noqa
+# There's no way to ignore "F401 '...' imported but unused" warnings in this
+# module, but to preserve other warnings. So, don't check this module at all.
+
+# Copyright 2021 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from typing import TYPE_CHECKING
+
+from ...file_utils import _BaseLazyModule, is_tokenizers_available, is_torch_available, is_vision_available
+
+
+_import_structure = {
+    "configuration_clip": ["CLIP_PRETRAINED_CONFIG_ARCHIVE_MAP", "CLIPConfig", "CLIPTextConfig", "CLIPVisionConfig"],
+    "tokenization_clip": ["CLIPTokenizer"],
+}
+
+if is_tokenizers_available():
+    _import_structure["tokenization_clip_fast"] = ["CLIPTokenizerFast"]
+
+if is_vision_available():
+    _import_structure["feature_extraction_clip"] = ["CLIPFeatureExtractor"]
+    _import_structure["processing_clip"] = ["CLIPProcessor"]
+
+if is_torch_available():
+    _import_structure["modeling_clip"] = [
+        "CLIP_PRETRAINED_MODEL_ARCHIVE_LIST",
+        "CLIPModel",
+        "CLIPPreTrainedModel",
+        "CLIPTextModel",
+        "CLIPVisionModel",
+    ]
+
+
+if TYPE_CHECKING:
+    from .configuration_clip import CLIP_PRETRAINED_CONFIG_ARCHIVE_MAP, CLIPConfig, CLIPTextConfig, CLIPVisionConfig
+    from .tokenization_clip import CLIPTokenizer
+
+    if is_tokenizers_available():
+        from .tokenization_clip_fast import CLIPTokenizerFast
+
+    if is_vision_available():
+        from .feature_extraction_clip import CLIPFeatureExtractor
+        from .processing_clip import CLIPProcessor
+
+    if is_torch_available():
+        from .modeling_clip import (
+            CLIP_PRETRAINED_MODEL_ARCHIVE_LIST,
+            CLIPModel,
+            CLIPPreTrainedModel,
+            CLIPTextModel,
+            CLIPVisionModel,
+        )
+
+
+else:
+    import importlib
+    import os
+    import sys
+
+    class _LazyModule(_BaseLazyModule):
+        """
+        Module class that surfaces all objects but only performs associated imports when the objects are requested.
+        """
+
+        __file__ = globals()["__file__"]
+        __path__ = [os.path.dirname(__file__)]
+
+        def _get_module(self, module_name: str):
+            return importlib.import_module("." + module_name, self.__name__)
+
+    sys.modules[__name__] = _LazyModule(__name__, _import_structure)

src/transformers/models/clip/configuration_clip.py

+# coding=utf-8
+# Copyright 2021 The HuggingFace Inc. team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" CLIP model configuration """
+
+import copy
+
+from ...configuration_utils import PretrainedConfig
+from ...utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+CLIP_PRETRAINED_CONFIG_ARCHIVE_MAP = {
+    "openai/clip-vit-base-patch32": "https://huggingface.co/openai/clip-vit-base-patch32/resolve/main/config.json",
+    # See all CLIP models at https://huggingface.co/models?filter=clip
+}
+
+
+class CLIPTextConfig(PretrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a :class:`~transformers.CLIPModel`. It is used to
+    instantiate an CLIP model according to the specified arguments, defining the model architecture. Instantiating a
+    configuration with the defaults will yield a similar configuration to that of the CLIP
+    `openai/clip-vit-base-patch32 <https://huggingface.co/openai/clip-vit-base-patch32>`__ architecture.
+
+    Configuration objects inherit from :class:`~transformers.PretrainedConfig` and can be used to control the model
+    outputs. Read the documentation from :class:`~transformers.PretrainedConfig` for more information.
+
+
+    Args:
+        vocab_size (:obj:`int`, `optional`, defaults to 49408):
+            Vocabulary size of the CLIP text model. Defines the number of different tokens that can be represented by
+            the :obj:`inputs_ids` passed when calling :class:`~transformers.CLIPModel`.
+        hidden_size (:obj:`int`, `optional`, defaults to 512):
+            Dimensionality of the encoder layers and the pooler layer.
+        intermediate_size (:obj:`int`, `optional`, defaults to 2048):
+            Dimensionality of the "intermediate" (i.e., feed-forward) layer in the Transformer encoder.
+        num_hidden_layers (:obj:`int`, `optional`, defaults to 12):
+            Number of hidden layers in the Transformer encoder.
+        num_attention_heads (:obj:`int`, `optional`, defaults to 8):
+            Number of attention heads for each attention layer in the Transformer encoder.
+        max_position_embeddings (:obj:`int`, `optional`, defaults to 77):
+            The maximum sequence length that this model might ever be used with. Typically set this to something large
+            just in case (e.g., 512 or 1024 or 2048).
+        hidden_act (:obj:`str` or :obj:`function`, `optional`, defaults to :obj:`"quick_gelu"`):
+            The non-linear activation function (function or string) in the encoder and pooler. If string,
+            :obj:`"gelu"`, :obj:`"relu"`, :obj:`"selu"` and :obj:`"gelu_new"` :obj:`"quick_gelu"` are supported.
+        layer_norm_eps (:obj:`float`, `optional`, defaults to 1e-5):
+            The epsilon used by the layer normalization layers.
+        attention_dropout (:obj:`float`, `optional`, defaults to 0.0):
+            The dropout ratio for the attention probabilities.
+        dropout (:obj:`float`, `optional`, defaults to 0.0):
+            The dropout probabilitiy for all fully connected layers in the embeddings, encoder, and pooler.
+        initializer_range (:obj:`float`, `optional`, defaults to 0.02):
+            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+        initializer_factor (:obj:`float`, `optional`, defaults to 1):
+            A factor for initializing all weight matrices (should be kept to 1, used internally for initialization
+            testing).
+        gradient_checkpointing (:obj:`bool`, `optional`, defaults to :obj:`False`):
+            If True, use gradient checkpointing to save memory at the expense of slower backward pass.
+
+        Example::
+
+        >>> from transformers import CLIPTextModel, CLIPTextConfig
+
+        >>> # Initializing a CLIPTextModel with openai/clip-vit-base-patch32 style configuration
+        >>> configuration = CLIPTextConfig()
+
+        >>> # Initializing a CLIPTextConfig from the openai/clip-vit-base-patch32 style configuration
+        >>> model = CLIPTextModel(configuration)
+
+        >>> # Accessing the model configuration
+        >>> configuration = model.config
+    """
+    model_type = "clip_text_model"
+
+    def __init__(
+        self,
+        vocab_size=49408,
+        hidden_size=512,
+        intermediate_size=2048,
+        num_hidden_layers=12,
+        num_attention_heads=8,
+        max_position_embeddings=77,
+        hidden_act="quick_gelu",
+        layer_norm_eps=1e-5,
+        dropout=0.0,
+        attention_dropout=0.0,
+        initializer_range=0.02,
+        initializer_factor=1.0,
+        pad_token_id=1,
+        bos_token_id=0,
+        eos_token_id=2,
+        gradient_checkpointing=False,
+        **kwargs
+    ):
+        super().__init__(pad_token_id=pad_token_id, bos_token_id=bos_token_id, eos_token_id=eos_token_id, **kwargs)
+
+        self.vocab_size = vocab_size
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.dropout = dropout
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.max_position_embeddings = max_position_embeddings
+        self.layer_norm_eps = layer_norm_eps
+        self.hidden_act = hidden_act
+        self.initializer_range = initializer_range
+        self.initializer_factor = initializer_factor
+        self.attention_dropout = attention_dropout
+        self.gradient_checkpointing = gradient_checkpointing
+
+
+class CLIPVisionConfig(PretrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a :class:`~transformers.CLIPModel`. It is used to
+    instantiate an CLIP model according to the specified arguments, defining the model architecture. Instantiating a
+    configuration with the defaults will yield a similar configuration to that of the CLIP
+    `openai/clip-vit-base-patch32 <https://huggingface.co/openai/clip-vit-base-patch32>`__ architecture.
+
+    Configuration objects inherit from :class:`~transformers.PretrainedConfig` and can be used to control the model
+    outputs. Read the documentation from :class:`~transformers.PretrainedConfig` for more information.
+
+
+    Args:
+        hidden_size (:obj:`int`, `optional`, defaults to 768):
+            Dimensionality of the encoder layers and the pooler layer.
+        intermediate_size (:obj:`int`, `optional`, defaults to 3072):
+            Dimensionality of the "intermediate" (i.e., feed-forward) layer in the Transformer encoder.
+        num_hidden_layers (:obj:`int`, `optional`, defaults to 12):
+            Number of hidden layers in the Transformer encoder.
+        num_attention_heads (:obj:`int`, `optional`, defaults to 12):
+            Number of attention heads for each attention layer in the Transformer encoder.
+        image_size (:obj:`int`, `optional`, defaults to 224):
+            The size (resolution) of each image.
+        patch_size (:obj:`int`, `optional`, defaults to 32):
+            The size (resolution) of each patch.
+        hidden_act (:obj:`str` or :obj:`function`, `optional`, defaults to :obj:`"quick_gelu"`):
+            The non-linear activation function (function or string) in the encoder and pooler. If string,
+            :obj:`"gelu"`, :obj:`"relu"`, :obj:`"selu"` and :obj:`"gelu_new"` :obj:`"quick_gelu"` are supported.
+        layer_norm_eps (:obj:`float`, `optional`, defaults to 1e-5):
+            The epsilon used by the layer normalization layers.
+        dropout (:obj:`float`, `optional`, defaults to 0.0):
+            The dropout probabilitiy for all fully connected layers in the embeddings, encoder, and pooler.
+        attention_dropout (:obj:`float`, `optional`, defaults to 0.0):
+            The dropout ratio for the attention probabilities.
+        initializer_range (:obj:`float`, `optional`, defaults to 0.02):
+            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+        initializer_factor (:obj:`float`, `optional`, defaults to 1):
+            A factor for initializing all weight matrices (should be kept to 1, used internally for initialization
+            testing).
+        gradient_checkpointing (:obj:`bool`, `optional`, defaults to :obj:`False`):
+            If True, use gradient checkpointing to save memory at the expense of slower backward pass.
+
+        Example::
+
+        >>> from transformers import CLIPVisionModel, CLIPVisionConfig
+
+        >>> # Initializing a CLIPVisionModel with openai/clip-vit-base-patch32 style configuration
+        >>> configuration = CLIPVisionConfig()
+
+        >>> # Initializing a CLIPVisionModel model from the openai/clip-vit-base-patch32 style configuration
+        >>> model = CLIPVisionModel(configuration)
+
+        >>> # Accessing the model configuration
+        >>> configuration = model.config
+    """
+
+    model_type = "clip_vision_model"
+
+    def __init__(
+        self,
+        hidden_size=768,
+        intermediate_size=3072,
+        num_hidden_layers=12,
+        num_attention_heads=12,
+        image_size=224,
+        patch_size=32,
+        hidden_act="quick_gelu",
+        layer_norm_eps=1e-5,
+        dropout=0.0,
+        attention_dropout=0.0,
+        initializer_range=0.02,
+        initializer_factor=1.0,
+        gradient_checkpointing=False,
+        **kwargs
+    ):
+        super().__init__(**kwargs)
+
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.dropout = dropout
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.patch_size = patch_size
+        self.image_size = image_size
+        self.initializer_range = initializer_range
+        self.initializer_factor = initializer_factor
+        self.attention_dropout = attention_dropout
+        self.layer_norm_eps = layer_norm_eps
+        self.hidden_act = hidden_act
+        self.gradient_checkpointing = gradient_checkpointing
+
+
+class CLIPConfig(PretrainedConfig):
+    r"""
+    :class:`~transformers.CLIPConfig` is the configuration class to store the configuration of a
+    :class:`~transformers.CLIPModel`. It is used to instantiate CLIP model according to the specified arguments,
+    defining the text model and vision model configs.
+
+    Configuration objects inherit from :class:`~transformers.PretrainedConfig` and can be used to control the model
+    outputs. Read the documentation from :class:`~transformers.PretrainedConfig` for more information.
+
+    Args:
+        projection_dim: (:obj:`int`, `optional`, defaults to 512):
+            Dimentionality of text and vision projection layers.
+        kwargs (`optional`):
+            Dictionary of keyword arguments. Notably:
+
+                - **text_config** (:class:`~transformers.CLIPTextConfig`, `optional`) -- An instance of a configuration
+                  object that defines the text model config.
+                - **vision_config** (:class:`~transformers.CLIPVisionConfig`, `optional`) -- An instance of a
+                  configuration object that defines the vision model config.
+    """
+
+    model_type = "clip"
+    is_composition = True
+
+    def __init__(self, text_config_dict=None, vision_config_dict=None, projection_dim=512, **kwargs):
+        super().__init__(text_config_dict=text_config_dict, vision_config_dict=vision_config_dict, **kwargs)
+
+        if text_config_dict is None:
+            text_config_dict = {}
+            logger.info("text_config_dict is None. Initializing the CLIPTextConfig with default values.")
+
+        if vision_config_dict is None:
+            vision_config_dict = {}
+            logger.info("vision_config_dict is None. initializing the CLIPVisionConfig with default values.")
+
+        self.text_config = CLIPTextConfig(**text_config_dict)
+        self.vision_config = CLIPVisionConfig(**vision_config_dict)
+
+        self.projection_dim = projection_dim
+        self.initializer_factor = 1.0
+
+    @classmethod
+    def from_text_vision_configs(cls, text_config: CLIPTextConfig, vision_config: CLIPVisionConfig, **kwargs):
+        r"""
+        Instantiate a :class:`~transformers.CLIPConfig` (or a derived class) from clip text model configuration and
+        clip vision model configuration.
+
+        Returns:
+            :class:`CLIPConfig`: An instance of a configuration object
+        """
+
+        return cls(text_config_dict=text_config.to_dict(), vision_config_dict=vision_config.to_dict(), **kwargs)
+
+    def to_dict(self):
+        """
+        Serializes this instance to a Python dictionary. Override the default
+        :meth:`~transformers.PretrainedConfig.to_dict`.
+
+        Returns:
+            :obj:`Dict[str, any]`: Dictionary of all the attributes that make up this configuration instance,
+        """
+        output = copy.deepcopy(self.__dict__)
+        output["text_config"] = self.text_config.to_dict()
+        output["vision_config"] = self.vision_config.to_dict()
+        output["model_type"] = self.__class__.model_type
+        return output

src/transformers/models/clip/convert_clip_original_pytorch_to_hf.py

+# coding=utf-8
+# Copyright 2021 The HuggingFace Inc. team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import argparse
+
+import torch
+
+from clip import load
+from transformers import CLIPConfig, CLIPModel
+
+
+def copy_attn_layer(hf_attn_layer, pt_attn_layer):
+    q_proj, k_proj, v_proj = pt_attn_layer.in_proj_weight.chunk(3, dim=0)
+    q_proj_bias, k_proj_bias, v_proj_bias = pt_attn_layer.in_proj_bias.chunk(3, dim=0)
+
+    out_proj_weights = pt_attn_layer.out_proj.weight
+    out_proj_bias = pt_attn_layer.out_proj.bias
+
+    hf_attn_layer.q_proj.weight.data = q_proj
+    hf_attn_layer.q_proj.bias.data = q_proj_bias
+
+    hf_attn_layer.k_proj.weight.data = k_proj
+    hf_attn_layer.k_proj.bias.data = k_proj_bias
+
+    hf_attn_layer.v_proj.weight.data = v_proj
+    hf_attn_layer.v_proj.bias.data = v_proj_bias
+
+    hf_attn_layer.out_proj.weight = out_proj_weights
+    hf_attn_layer.out_proj.bias = out_proj_bias
+
+
+def copy_mlp(hf_mlp, pt_mlp):
+    copy_linear(hf_mlp.fc1, pt_mlp.c_fc)
+    copy_linear(hf_mlp.fc2, pt_mlp.c_proj)
+
+
+def copy_linear(hf_linear, pt_linear):
+    hf_linear.weight = pt_linear.weight
+    hf_linear.bias = pt_linear.bias
+
+
+def copy_layer(hf_layer, pt_layer):
+    # copy layer norms
+    copy_linear(hf_layer.layer_norm1, pt_layer.ln_1)
+    copy_linear(hf_layer.layer_norm2, pt_layer.ln_2)
+
+    # copy MLP
+    copy_mlp(hf_layer.mlp, pt_layer.mlp)
+
+    # copy attn
+    copy_attn_layer(hf_layer.self_attn, pt_layer.attn)
+
+
+def copy_layers(hf_layers, pt_layers):
+    for hf_layer, pt_layer in zip(hf_layers, pt_layers):
+        copy_layer(hf_layer, pt_layer)
+
+
+def copy_encoder(hf_encoder, pt_model):
+    # copy  embeds
+    hf_encoder.embeddings.token_embedding.weight = pt_model.token_embedding.weight
+    hf_encoder.embeddings.position_embedding.weight.data = pt_model.positional_embedding
+
+    # copy layer norm
+    copy_linear(hf_encoder.final_layer_norm, pt_model.ln_final)
+
+    # copy hidden layers
+    copy_layers(hf_encoder.encoder.layers, pt_model.transformer.resblocks)
+
+
+def copy_text_model_and_projection(hf_model, pt_model):
+    # copy projection
+    hf_model.text_projection.weight.data = pt_model.text_projection.data.T
+
+    # copy text encoder
+    copy_encoder(hf_model.text_model, pt_model)
+
+
+def copy_vison_model_and_projection(hf_model, pt_model):
+    # copy projection
+    hf_model.visual_projection.weight.data = pt_model.visual.proj.data.T
+
+    # copy layer norms
+    copy_linear(hf_model.vision_model.pre_layrnorm, pt_model.visual.ln_pre)
+    copy_linear(hf_model.vision_model.post_layernorm, pt_model.visual.ln_post)
+
+    # copy embeds
+    hf_model.vision_model.embeddings.patch_embedding.weight.data = pt_model.visual.conv1.weight.data
+    hf_model.vision_model.embeddings.class_embedding = pt_model.visual.class_embedding
+    hf_model.vision_model.embeddings.position_embedding.weight.data = pt_model.visual.positional_embedding.data
+
+    # copy encoder
+    copy_layers(hf_model.vision_model.encoder.layers, pt_model.visual.transformer.resblocks)
+
+
+@torch.no_grad()
+def convert_clip_checkpoint(checkpoint_path, pytorch_dump_folder_path, config_path=None):
+    """
+    Copy/paste/tweak model's weights to transformers design.
+    """
+    if config_path is not None:
+        config = CLIPConfig.from_pretrained(config_path)
+    else:
+        config = CLIPConfig(projection_dim=512, text_config={}, vision_config={})
+
+    hf_model = CLIPModel(config).eval()
+
+    pt_model, _ = load(checkpoint_path, jit=False)
+    pt_model = pt_model.eval()
+
+    copy_text_model_and_projection(hf_model, pt_model)
+    copy_vison_model_and_projection(hf_model, pt_model)
+    hf_model.logit_scale = pt_model.logit_scale
+
+    input_ids = torch.arange(0, 77).unsqueeze(0)
+    pixel_values = torch.randn(1, 3, 224, 224)
+
+    hf_logits_per_image, hf_logits_per_text = hf_model(
+        input_ids=input_ids, pixel_values=pixel_values, return_dict=True
+    )[1:3]
+    pt_logits_per_image, pt_logits_per_text = pt_model(pixel_values, input_ids)
+
+    assert torch.allclose(hf_logits_per_image, pt_logits_per_image, atol=1e-3)
+    assert torch.allclose(hf_logits_per_text, pt_logits_per_text, atol=1e-3)
+
+    hf_model.save_pretrained(pytorch_dump_folder_path)
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--pytorch_dump_folder_path", default=None, type=str, help="Path to the output PyTorch model.")
+    parser.add_argument("--checkpoint_path", default=None, type=str, help="Path to fairseq checkpoint")
+    parser.add_argument("--config_path", default=None, type=str, help="Path to hf config.json of model to convert")
+    args = parser.parse_args()
+
+    convert_clip_checkpoint(args.checkpoint_path, args.pytorch_dump_folder_path, args.config_path)

src/transformers/models/clip/feature_extraction_clip.py

+# coding=utf-8
+# Copyright 2021 The HuggingFace Inc. team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Feature extractor class for CLIP."""
+
+from typing import List, Optional, Union
+
+import numpy as np
+from PIL import Image
+
+from ...feature_extraction_utils import BatchFeature, FeatureExtractionMixin
+from ...file_utils import TensorType
+from ...image_utils import ImageFeatureExtractionMixin, is_torch_tensor
+from ...utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+
+class CLIPFeatureExtractor(FeatureExtractionMixin, ImageFeatureExtractionMixin):
+    r"""
+    Constructs a CLIP feature extractor.
+
+    This feature extractor inherits from :class:`~transformers.FeatureExtractionMixin` which contains most of the main
+    methods. Users should refer to this superclass for more information regarding those methods.
+
+    Args:
+        do_resize (:obj:`bool`, `optional`, defaults to :obj:`True`):
+            Whether to resize the input to a certain :obj:`size`.
+        size (:obj:`int`, `optional`, defaults to 224):
+            Resize the input to the given size. Only has an effect if :obj:`do_resize` is set to :obj:`True`.
+        resample (:obj:`int`, `optional`, defaults to :obj:`PIL.Image.BICUBIC`):
+            An optional resampling filter. This can be one of :obj:`PIL.Image.NEAREST`, :obj:`PIL.Image.BOX`,
+            :obj:`PIL.Image.BILINEAR`, :obj:`PIL.Image.HAMMING`, :obj:`PIL.Image.BICUBIC` or :obj:`PIL.Image.LANCZOS`.
+            Only has an effect if :obj:`do_resize` is set to :obj:`True`.
+        do_center_crop (:obj:`bool`, `optional`, defaults to :obj:`True`):
+            Whether to crop the input at the center. If the input size is smaller than :obj:`crop_size` along any edge,
+            the image is padded with 0's and then center cropped.
+        crop_size (:obj:`int`, `optional`, defaults to 224):
+            Desired output size when applying center-cropping. Only has an effect if :obj:`do_center_crop` is set to
+            :obj:`True`.
+        do_normalize (:obj:`bool`, `optional`, defaults to :obj:`True`):
+            Whether or not to normalize the input with :obj:`image_mean` and :obj:`image_std`.
+        image_mean (:obj:`List[int]`, defaults to :obj:`[0.485, 0.456, 0.406]`):
+            The sequence of means for each channel, to be used when normalizing images.
+        image_std (:obj:`List[int]`, defaults to :obj:`[0.229, 0.224, 0.225]`):
+            The sequence of standard deviations for each channel, to be used when normalizing images.
+    """
+
+    model_input_names = ["pixel_values"]
+
+    def __init__(
+        self,
+        do_resize=True,
+        size=224,
+        resample=Image.BICUBIC,
+        do_center_crop=True,
+        crop_size=224,
+        do_normalize=True,
+        image_mean=None,
+        image_std=None,
+        **kwargs
+    ):
+        super().__init__(**kwargs)
+        self.do_resize = do_resize
+        self.size = size
+        self.resample = resample
+        self.do_center_crop = do_center_crop
+        self.crop_size = crop_size
+        self.do_normalize = do_normalize
+        self.image_mean = image_mean if image_mean is not None else [0.48145466, 0.4578275, 0.40821073]
+        self.image_std = image_std if image_std is not None else [0.26862954, 0.26130258, 0.27577711]
+
+    def __call__(
+        self,
+        images: Union[
+            Image.Image, np.ndarray, "torch.Tensor", List[Image.Image], List[np.ndarray], List["torch.Tensor"]  # noqa
+        ],
+        return_tensors: Optional[Union[str, TensorType]] = None,
+        **kwargs
+    ) -> BatchFeature:
+        """
+        Main method to prepare for the model one or several image(s).
+
+        .. warning::
+
+           NumPy arrays and PyTorch tensors are converted to PIL images when resizing, so the most efficient is to pass
+           PIL images.
+
+        Args:
+            images (:obj:`PIL.Image.Image`, :obj:`np.ndarray`, :obj:`torch.Tensor`, :obj:`List[PIL.Image.Image]`, :obj:`List[np.ndarray]`, :obj:`List[torch.Tensor]`):
+                The image or batch of images to be prepared. Each image can be a PIL image, NumPy array or PyTorch
+                tensor. In case of a NumPy array/PyTorch tensor, each image should be of shape (C, H, W), where C is a
+                number of channels, H and W are image height and width.
+
+            return_tensors (:obj:`str` or :class:`~transformers.file_utils.TensorType`, `optional`, defaults to :obj:`'np'`):
+                If set, will return tensors of a particular framework. Acceptable values are:
+
+                * :obj:`'tf'`: Return TensorFlow :obj:`tf.constant` objects.
+                * :obj:`'pt'`: Return PyTorch :obj:`torch.Tensor` objects.
+                * :obj:`'np'`: Return NumPy :obj:`np.ndarray` objects.
+                * :obj:`'jax'`: Return JAX :obj:`jnp.ndarray` objects.
+
+        Returns:
+            :class:`~transformers.BatchFeature`: A :class:`~transformers.BatchFeature` with the following fields:
+
+            - **pixel_values** -- Pixel values to be fed to a model.
+        """
+        # Input type checking for clearer error
+        valid_images = False
+
+        # Check that images has a valid type
+        if isinstance(images, (Image.Image, np.ndarray)) or is_torch_tensor(images):
+            valid_images = True
+        elif isinstance(images, (list, tuple)):
+            if len(images) == 0 or isinstance(images[0], (Image.Image, np.ndarray)) or is_torch_tensor(images[0]):
+                valid_images = True
+
+        if not valid_images:
+            raise ValueError(
+                "Images must of type `PIL.Image.Image`, `np.ndarray` or `torch.Tensor` (single example),"
+                "`List[PIL.Image.Image]`, `List[np.ndarray]` or `List[torch.Tensor]` (batch of examples)."
+            )
+
+        is_batched = bool(
+            isinstance(images, (list, tuple))
+            and (isinstance(images[0], (Image.Image, np.ndarray)) or is_torch_tensor(images[0]))
+        )
+
+        if not is_batched:
+            images = [images]
+
+        # transformations (resizing + center cropping + normalization)
+        if self.do_resize and self.size is not None and self.resample is not None:
+            images = [self.resize(image=image, size=self.size, resample=self.resample) for image in images]
+        if self.do_center_crop and self.crop_size is not None:
+            images = [self.center_crop(image, self.crop_size) for image in images]
+        if self.do_normalize:
+            images = [self.normalize(image=image, mean=self.image_mean, std=self.image_std) for image in images]
+
+        # return as BatchFeature
+        data = {"pixel_values": images}
+        encoded_inputs = BatchFeature(data=data, tensor_type=return_tensors)
+
+        return encoded_inputs

src/transformers/models/clip/modeling_clip.py

+# coding=utf-8
+# Copyright 2021 The OpenAI Team Authors and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" PyTorch CLIP model. """
+
+
+from typing import Any, Optional, Tuple
+
+import torch
+import torch.nn.functional as F
+import torch.utils.checkpoint
+from torch import nn
+
+from ...activations import ACT2FN
+from ...file_utils import (
+    ModelOutput,
+    add_start_docstrings,
+    add_start_docstrings_to_model_forward,
+    replace_return_docstrings,
+)
+from ...modeling_outputs import BaseModelOutput, BaseModelOutputWithPooling
+from ...modeling_utils import PreTrainedModel
+from ...utils import logging
+from .configuration_clip import CLIPConfig, CLIPTextConfig, CLIPVisionConfig
+
+
+logger = logging.get_logger(__name__)
+
+
+CLIP_PRETRAINED_MODEL_ARCHIVE_LIST = [
+    "openai/clip-vit-base-patch32",
+    # See all CLIP models at https://huggingface.co/models?filter=clip
+]
+
+
+# Copied from transformers.models.bart.modeling_bart._expand_mask
+def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Optional[int] = None):
+    """
+    Expands attention_mask from `[bsz, seq_len]` to `[bsz, 1, tgt_seq_len, src_seq_len]`.
+    """
+    bsz, src_len = mask.size()
+    tgt_len = tgt_len if tgt_len is not None else src_len
+
+    expanded_mask = mask[:, None, None, :].expand(bsz, 1, tgt_len, src_len).to(dtype)
+
+    inverted_mask = 1.0 - expanded_mask
+
+    return inverted_mask.masked_fill(inverted_mask.bool(), torch.finfo(dtype).min)
+
+
+# contrastive loss function, adapted from
+# https://sachinruk.github.io/blog/pytorch/pytorch%20lightning/loss%20function/gpu/2021/03/07/CLIP.html
+def contrastive_loss(logits: torch.Tensor, dim: int) -> torch.Tensor:
+    neg_ce = torch.diag(F.log_softmax(logits, dim=dim))
+    return -neg_ce.mean()
+
+
+def clip_loss(similarity: torch.Tensor) -> torch.Tensor:
+    caption_loss = contrastive_loss(similarity, dim=0)
+    image_loss = contrastive_loss(similarity, dim=1)
+    return (caption_loss + image_loss) / 2.0
+
+
+class CLIPOutput(ModelOutput):
+    """
+    Args:
+        loss (:obj:`torch.FloatTensor` of shape :obj:`(1,)`, `optional`, returned when :obj:`return_loss` is :obj:`True`):
+            Contrastive loss for image-text similarity.
+        logits_per_image:(:obj:`torch.FloatTensor` of shape :obj:`(image_batch_size, text_batch_size)`):
+            The scaled dot product scores between :obj:`image_embeds` and :obj:`text_embeds`. This represents the
+            image-text similarity scores.
+        logits_per_text:(:obj:`torch.FloatTensor` of shape :obj:`(text_batch_size, image_batch_size)`):
+            The scaled dot product scores between :obj:`text_embeds` and :obj:`image_embeds`. This represents the
+            text-image similarity scores.
+        text_embeds(:obj:`torch.FloatTensor` of shape :obj:`(batch_size, output_dim`):
+            The text embeddings obtained by applying the projection layer to the pooled output of
+            :class:`~transformers.CLIPTextModel`.
+        image_embeds(:obj:`torch.FloatTensor` of shape :obj:`(batch_size, output_dim`):
+            The image embeddings obtained by applying the projection layer to the pooled output of
+            :class:`~transformers.CLIPVisionModel`.
+        text_model_output(:obj:`BaseModelOutputWithPooling`):
+            The output of the :class:`~transformers.CLIPTextModel`.
+        vision_model_output(:obj:`BaseModelOutputWithPooling`):
+            The output of the :class:`~transformers.CLIPVisionModel`.
+    """
+
+    loss: Optional[torch.FloatTensor] = None
+    logits_per_image: torch.FloatTensor = None
+    logits_per_text: torch.FloatTensor = None
+    text_embeds: torch.FloatTensor = None
+    image_embeds: torch.FloatTensor = None
+    text_model_output: BaseModelOutputWithPooling = None
+    vision_model_output: BaseModelOutputWithPooling = None
+
+    def to_tuple(self) -> Tuple[Any]:
+        return tuple(
+            self[k] if k not in ["text_model_output", "vision_model_output"] else getattr(self, k).to_tuple()
+            for k in self.keys()
+        )
+
+
+class CLIPVisionEmbeddings(nn.Module):
+    def __init__(self, config: CLIPVisionConfig):
+        super().__init__()
+        self.config = config
+        self.embed_dim = config.hidden_size
+        self.image_size = config.image_size
+        self.patch_size = config.patch_size
+
+        self.class_embedding = nn.Parameter(torch.randn(self.embed_dim))
+
+        self.patch_embedding = nn.Conv2d(
+            in_channels=3, out_channels=self.embed_dim, kernel_size=self.patch_size, stride=self.patch_size, bias=False
+        )
+
+        self.num_patches = (self.image_size // self.patch_size) ** 2
+        self.num_positions = self.num_patches + 1
+        self.position_embedding = nn.Embedding(self.num_positions, self.embed_dim)
+        self.register_buffer("position_ids", torch.arange(self.num_positions).expand((1, -1)))
+
+    def forward(self, pixel_values):
+        batch_size = pixel_values.shape[0]
+        patch_embeds = self.patch_embedding(pixel_values)  # shape = [*, width, grid, grid]
+        patch_embeds = patch_embeds.flatten(2).transpose(1, 2)
+
+        class_embeds = self.class_embedding.expand(batch_size, 1, -1)
+        embeddings = torch.cat([class_embeds, patch_embeds], dim=1)
+        embeddings = embeddings + self.position_embedding(self.position_ids)
+        return embeddings
+
+
+class CLIPTextEmbeddings(nn.Module):
+    def __init__(self, config: CLIPTextConfig):
+        super().__init__()
+        embed_dim = config.hidden_size
+
+        self.token_embedding = nn.Embedding(config.vocab_size, embed_dim)
+        self.position_embedding = nn.Embedding(config.max_position_embeddings, embed_dim)
+
+        # position_ids (1, len position emb) is contiguous in memory and exported when serialized
+        self.register_buffer("position_ids", torch.arange(config.max_position_embeddings).expand((1, -1)))
+
+    def forward(self, input_ids=None, position_ids=None, inputs_embeds=None):
+        seq_length = input_ids.shape[-1] if input_ids is not None else inputs_embeds.shape[-2]
+
+        if position_ids is None:
+            position_ids = self.position_ids[:, :seq_length]
+
+        if inputs_embeds is None:
+            inputs_embeds = self.token_embedding(input_ids)
+
+        position_embeddings = self.position_embedding(position_ids)
+        embeddings = inputs_embeds + position_embeddings
+
+        return embeddings
+
+
+class CLIPAttention(nn.Module):
+    """Multi-headed attention from 'Attention Is All You Need' paper"""
+
+    def __init__(self, config):
+        super().__init__()
+        self.config = config
+        self.embed_dim = config.hidden_size
+        self.num_heads = config.num_attention_heads
+        self.head_dim = self.embed_dim // self.num_heads
+        assert (
+            self.head_dim * self.num_heads == self.embed_dim
+        ), f"embed_dim must be divisible by num_heads (got `embed_dim`: {self.embed_dim} and `num_heads`: {self.num_heads})."
+        self.scale = self.head_dim ** -0.5
+        self.dropout = config.attention_dropout
+
+        self.k_proj = nn.Linear(self.embed_dim, self.embed_dim)
+        self.v_proj = nn.Linear(self.embed_dim, self.embed_dim)
+        self.q_proj = nn.Linear(self.embed_dim, self.embed_dim)
+        self.out_proj = nn.Linear(self.embed_dim, self.embed_dim)
+
+    def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
+        return tensor.view(bsz, seq_len, self.num_heads, self.head_dim).transpose(1, 2).contiguous()
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: Optional[torch.Tensor] = None,
+        causal_attention_mask: Optional[torch.Tensor] = None,
+        output_attentions: bool = False,
+    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
+        """Input shape: Batch x Time x Channel"""
+
+        bsz, tgt_len, embed_dim = hidden_states.size()
+
+        # get query proj
+        query_states = self.q_proj(hidden_states) * self.scale
+        key_states = self._shape(self.k_proj(hidden_states), -1, bsz)
+        value_states = self._shape(self.v_proj(hidden_states), -1, bsz)
+
+        proj_shape = (bsz * self.num_heads, -1, self.head_dim)
+        query_states = self._shape(query_states, tgt_len, bsz).view(*proj_shape)
+        key_states = key_states.view(*proj_shape)
+        value_states = value_states.view(*proj_shape)
+
+        src_len = key_states.size(1)
+        attn_weights = torch.bmm(query_states, key_states.transpose(1, 2))
+
+        if attn_weights.size() != (bsz * self.num_heads, tgt_len, src_len):
+            raise ValueError(
+                f"Attention weights should be of size {(bsz * self.num_heads, tgt_len, src_len)}, but is {attn_weights.size()}"
+            )
+
+        # apply the causal_attention_mask first
+        if causal_attention_mask is not None:
+            if causal_attention_mask.size() != (bsz, 1, tgt_len, src_len):
+                raise ValueError(
+                    f"Attention mask should be of size {(bsz, 1, tgt_len, src_len)}, but is {causal_attention_mask.size()}"
+                )
+            attn_weights = attn_weights.view(bsz, self.num_heads, tgt_len, src_len) + causal_attention_mask
+            attn_weights = attn_weights.view(bsz * self.num_heads, tgt_len, src_len)
+
+        if attention_mask is not None:
+            if attention_mask.size() != (bsz, 1, tgt_len, src_len):
+                raise ValueError(
+                    f"Attention mask should be of size {(bsz, 1, tgt_len, src_len)}, but is {causal_attention_mask.size()}"
+                )
+            attn_weights = attn_weights.view(bsz, self.num_heads, tgt_len, src_len) + attention_mask
+            attn_weights = attn_weights.view(bsz * self.num_heads, tgt_len, src_len)
+
+        attn_weights = F.softmax(attn_weights, dim=-1)
+
+        if output_attentions:
+            # this operation is a bit akward, but it's required to
+            # make sure that attn_weights keeps its gradient.
+            # In order to do so, attn_weights have to reshaped
+            # twice and have to be reused in the following
+            attn_weights_reshaped = attn_weights.view(bsz, self.num_heads, tgt_len, src_len)
+            attn_weights = attn_weights_reshaped.view(bsz * self.num_heads, tgt_len, src_len)
+        else:
+            attn_weights_reshaped = None
+
+        attn_probs = F.dropout(attn_weights, p=self.dropout, training=self.training)
+
+        attn_output = torch.bmm(attn_probs, value_states)
+
+        if attn_output.size() != (bsz * self.num_heads, tgt_len, self.head_dim):
+            raise ValueError(
+                f"`attn_output` should be of size {(bsz, self.num_heads, tgt_len, self.head_dim)}, but is {attn_output.size()}"
+            )
+
+        attn_output = attn_output.view(bsz, self.num_heads, tgt_len, self.head_dim)
+        attn_output = attn_output.transpose(1, 2)
+        attn_output = attn_output.reshape(bsz, tgt_len, embed_dim)
+
+        attn_output = self.out_proj(attn_output)
+
+        return attn_output, attn_weights_reshaped
+
+
+class CLIPMLP(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.config = config
+        self.activation_fn = ACT2FN[config.hidden_act]
+        self.fc1 = nn.Linear(config.hidden_size, config.intermediate_size)
+        self.fc2 = nn.Linear(config.intermediate_size, config.hidden_size)
+
+    def forward(self, hidden_states):
+        hidden_states = self.fc1(hidden_states)
+        hidden_states = self.activation_fn(hidden_states)
+        hidden_states = self.fc2(hidden_states)
+        return hidden_states
+
+
+class CLIPEncoderLayer(nn.Module):
+    def __init__(self, config: CLIPConfig):
+        super().__init__()
+        self.embed_dim = config.hidden_size
+        self.self_attn = CLIPAttention(config)
+        self.layer_norm1 = nn.LayerNorm(self.embed_dim)
+        self.mlp = CLIPMLP(config)
+        self.layer_norm2 = nn.LayerNorm(self.embed_dim)
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: torch.Tensor,
+        causal_attention_mask: torch.Tensor,
+        output_attentions: bool = False,
+    ):
+        """
+        Args:
+            hidden_states (:obj:`torch.FloatTensor`): input to the layer of shape :obj:`(seq_len, batch, embed_dim)`
+            attention_mask (:obj:`torch.FloatTensor`): attention mask of size
+                :obj:`(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
+            layer_head_mask (:obj:`torch.FloatTensor`): mask for attention heads in a given layer of size
+                :obj:`(config.encoder_attention_heads,)`.
+            output_attentions (:obj:`bool`, `optional`):
+                Whether or not to return the attentions tensors of all attention layers. See ``attentions`` under
+                returned tensors for more detail.
+        """
+        residual = hidden_states
+
+        hidden_states = self.layer_norm1(hidden_states)
+        hidden_states, attn_weights = self.self_attn(
+            hidden_states=hidden_states,
+            attention_mask=attention_mask,
+            causal_attention_mask=causal_attention_mask,
+            output_attentions=output_attentions,
+        )
+        hidden_states = residual + hidden_states
+
+        residual = hidden_states
+        hidden_states = self.layer_norm2(hidden_states)
+        hidden_states = self.mlp(hidden_states)
+        hidden_states = residual + hidden_states
+
+        outputs = (hidden_states,)
+
+        if output_attentions:
+            outputs += (attn_weights,)
+
+        return outputs
+
+
+class CLIPPreTrainedModel(PreTrainedModel):
+    """
+    An abstract class to handle weights initialization and a simple interface for downloading and loading pretrained
+    models.
+    """
+
+    config_class = CLIPConfig
+    base_model_prefix = "clip"
+    _keys_to_ignore_on_load_missing = [r"position_ids"]
+
+    def _init_weights(self, module):
+        """Initialize the weights"""
+        factor = self.config.initializer_factor
+        if isinstance(module, CLIPTextEmbeddings):
+            module.token_embedding.weight.data.normal_(mean=0.0, std=factor * 0.02)
+            module.position_embedding.weight.data.normal_(mean=0.0, std=factor * 0.02)
+        elif isinstance(module, CLIPVisionEmbeddings):
+            factor = self.config.initializer_factor
+            nn.init.normal_(module.class_embedding, mean=0.0, std=module.embed_dim ** -0.5 * factor)
+            nn.init.normal_(module.patch_embedding.weight, std=module.config.initializer_range * factor)
+            nn.init.normal_(module.position_embedding.weight, std=module.config.initializer_range * factor)
+        elif isinstance(module, CLIPAttention):
+            factor = self.config.initializer_factor
+            in_proj_std = (module.embed_dim ** -0.5) * ((2 * module.config.num_hidden_layers) ** -0.5) * factor
+            out_proj_std = (module.embed_dim ** -0.5) * factor
+            nn.init.normal_(module.q_proj.weight, std=in_proj_std)
+            nn.init.normal_(module.k_proj.weight, std=in_proj_std)
+            nn.init.normal_(module.v_proj.weight, std=in_proj_std)
+            nn.init.normal_(module.out_proj.weight, std=out_proj_std)
+        elif isinstance(module, CLIPMLP):
+            factor = self.config.initializer_factor
+            in_proj_std = (
+                (module.config.hidden_size ** -0.5) * ((2 * module.config.num_hidden_layers) ** -0.5) * factor
+            )
+            fc_std = (2 * module.config.hidden_size) ** -0.5 * factor
+            nn.init.normal_(module.fc1.weight, std=fc_std)
+            nn.init.normal_(module.fc2.weight, std=in_proj_std)
+        elif isinstance(module, CLIPModel):
+            nn.init.normal_(
+                module.text_projection.weight,
+                std=module.text_embed_dim ** -0.5 * self.config.initializer_factor,
+            )
+            nn.init.normal_(
+                module.visual_projection.weight,
+                std=module.vision_embed_dim ** -0.5 * self.config.initializer_factor,
+            )
+
+        if isinstance(module, nn.LayerNorm):
+            module.bias.data.zero_()
+            module.weight.data.fill_(1.0)
+        if isinstance(module, nn.Linear) and module.bias is not None:
+            module.bias.data.zero_()
+
+
+CLIP_START_DOCSTRING = r"""
+    This model is a PyTorch `torch.nn.Module <https://pytorch.org/docs/stable/nn.html#torch.nn.Module>`_ subclass. Use
+    it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and
+    behavior.
+
+    Parameters:
+        config (:class:`~transformers.CLIPConfig`): Model configuration class with all the parameters of the model.
+            Initializing with a config file does not load the weights associated with the model, only the
+            configuration. Check out the :meth:`~transformers.PreTrainedModel.from_pretrained` method to load the model
+            weights.
+"""
+
+CLIP_TEXT_INPUTS_DOCSTRING = r"""
+    Args:
+        input_ids (:obj:`torch.LongTensor` of shape :obj:`(batch_size, sequence_length)`):
+            Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide
+            it.
+
+            Indices can be obtained using :class:`~transformers.CLIPTokenizer`. See
+            :meth:`transformers.PreTrainedTokenizer.encode` and :meth:`transformers.PreTrainedTokenizer.__call__` for
+            details.
+
+            `What are input IDs? <../glossary.html#input-ids>`__
+        attention_mask (:obj:`torch.Tensor` of shape :obj:`(batch_size, sequence_length)`, `optional`):
+            Mask to avoid performing attention on padding token indices. Mask values selected in ``[0, 1]``:
+
+            - 1 for tokens that are **not masked**,
+            - 0 for tokens that are **masked**.
+
+            `What are attention masks? <../glossary.html#attention-mask>`__
+        position_ids (:obj:`torch.LongTensor` of shape :obj:`(batch_size, sequence_length)`, `optional`):
+            Indices of positions of each input sequence tokens in the position embeddings. Selected in the range ``[0,
+            config.max_position_embeddings - 1]``.
+
+            `What are position IDs? <../glossary.html#position-ids>`_
+        output_attentions (:obj:`bool`, `optional`):
+            Whether or not to return the attentions tensors of all attention layers. See ``attentions`` under returned
+            tensors for more detail.
+        output_hidden_states (:obj:`bool`, `optional`):
+            Whether or not to return the hidden states of all layers. See ``hidden_states`` under returned tensors for
+            more detail.
+        return_dict (:obj:`bool`, `optional`):
+            Whether or not to return a :class:`~transformers.file_utils.ModelOutput` instead of a plain tuple.
+"""
+
+CLIP_VISION_INPUTS_DOCSTRING = r"""
+    Args:
+        pixel_values (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, num_channels, height, width)`):
+            Pixel values. Padding will be ignored by default should you provide it. Pixel values can be obtained using
+            :class:`~transformers.CLIPFeatureExtractor`. See :meth:`transformers.CLIPFeatureExtractor.__call__` for
+            details.
+        output_attentions (:obj:`bool`, `optional`):
+            Whether or not to return the attentions tensors of all attention layers. See ``attentions`` under returned
+            tensors for more detail.
+        output_hidden_states (:obj:`bool`, `optional`):
+            Whether or not to return the hidden states of all layers. See ``hidden_states`` under returned tensors for
+            more detail.
+        return_dict (:obj:`bool`, `optional`):
+            Whether or not to return a :class:`~transformers.file_utils.ModelOutput` instead of a plain tuple.
+"""
+
+CLIP_INPUTS_DOCSTRING = r"""
+    Args:
+        input_ids (:obj:`torch.LongTensor` of shape :obj:`(batch_size, sequence_length)`):
+            Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide
+            it.
+
+            Indices can be obtained using :class:`~transformers.CLIPTokenizer`. See
+            :meth:`transformers.PreTrainedTokenizer.encode` and :meth:`transformers.PreTrainedTokenizer.__call__` for
+            details.
+
+            `What are input IDs? <../glossary.html#input-ids>`__
+        attention_mask (:obj:`torch.Tensor` of shape :obj:`(batch_size, sequence_length)`, `optional`):
+            Mask to avoid performing attention on padding token indices. Mask values selected in ``[0, 1]``:
+
+            - 1 for tokens that are **not masked**,
+            - 0 for tokens that are **masked**.
+
+            `What are attention masks? <../glossary.html#attention-mask>`__
+        position_ids (:obj:`torch.LongTensor` of shape :obj:`(batch_size, sequence_length)`, `optional`):
+            Indices of positions of each input sequence tokens in the position embeddings. Selected in the range ``[0,
+            config.max_position_embeddings - 1]``.
+
+            `What are position IDs? <../glossary.html#position-ids>`_
+        pixel_values (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, num_channels, height, width)`):
+            Pixel values. Padding will be ignored by default should you provide it. Pixel values can be obtained using
+            :class:`~transformers.CLIPFeatureExtractor`. See :meth:`transformers.CLIPFeatureExtractor.__call__` for
+            details.
+        return_loss (:obj:`bool`, `optional`):
+            Whether or not to return the contrastive loss.
+        output_attentions (:obj:`bool`, `optional`):
+            Whether or not to return the attentions tensors of all attention layers. See ``attentions`` under returned
+            tensors for more detail.
+        output_hidden_states (:obj:`bool`, `optional`):
+            Whether or not to return the hidden states of all layers. See ``hidden_states`` under returned tensors for
+            more detail.
+        return_dict (:obj:`bool`, `optional`):
+            Whether or not to return a :class:`~transformers.file_utils.ModelOutput` instead of a plain tuple.
+"""
+
+
+class CLIPEncoder(nn.Module):
+    """
+    Transformer encoder consisting of :obj:`config.num_hidden_layers` self attention layers. Each layer is a
+    :class:`~transformers.CLIPEncoderLayer`.
+
+    Args:
+        config: CLIPConfig
+        embed_tokens (torch.nn.Embedding): output embedding
+    """
+
+    def __init__(self, config: CLIPConfig):
+        super().__init__()
+        self.config = config
+        self.layers = nn.ModuleList([CLIPEncoderLayer(config) for _ in range(config.num_hidden_layers)])
+
+    def forward(
+        self,
+        inputs_embeds,
+        attention_mask=None,
+        causal_attention_mask=None,
+        output_attentions=None,
+        output_hidden_states=None,
+        return_dict=None,
+    ):
+        r"""
+        Args:
+            inputs_embeds (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`, `optional`):
+                Optionally, instead of passing :obj:`input_ids` you can choose to directly pass an embedded
+                representation. This is useful if you want more control over how to convert :obj:`input_ids` indices
+                into associated vectors than the model's internal embedding lookup matrix.
+            attention_mask (:obj:`torch.Tensor` of shape :obj:`(batch_size, sequence_length)`, `optional`):
+                Mask to avoid performing attention on padding token indices. Mask values selected in ``[0, 1]``:
+
+                - 1 for tokens that are **not masked**,
+                - 0 for tokens that are **masked**.
+
+                `What are attention masks? <../glossary.html#attention-mask>`__
+            causal_attention_mask (:obj:`torch.Tensor` of shape :obj:`(batch_size, sequence_length)`, `optional`):
+                Causal mask for the text model. Mask values selected in ``[0, 1]``:
+
+                - 1 for tokens that are **not masked**,
+                - 0 for tokens that are **masked**.
+
+                `What are attention masks? <../glossary.html#attention-mask>`__
+            output_attentions (:obj:`bool`, `optional`):
+                Whether or not to return the attentions tensors of all attention layers. See ``attentions`` under
+                returned tensors for more detail.
+            output_hidden_states (:obj:`bool`, `optional`):
+                Whether or not to return the hidden states of all layers. See ``hidden_states`` under returned tensors
+                for more detail.
+            return_dict (:obj:`bool`, `optional`):
+                Whether or not to return a :class:`~transformers.file_utils.ModelOutput` instead of a plain tuple.
+        """
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        encoder_states = () if output_hidden_states else None
+        all_attentions = () if output_attentions else None
+
+        hidden_states = inputs_embeds
+        for idx, encoder_layer in enumerate(self.layers):
+            if output_hidden_states:
+                encoder_states = encoder_states + (hidden_states,)
+            if getattr(self.config, "gradient_checkpointing", False) and self.training:
+
+                def create_custom_forward(module):
+                    def custom_forward(*inputs):
+                        return module(*inputs, output_attentions)
+
+                    return custom_forward
+
+                layer_outputs = torch.utils.checkpoint.checkpoint(
+                    create_custom_forward(encoder_layer),
+                    hidden_states,
+                    attention_mask,
+                    causal_attention_mask,
+                )
+            else:
+                layer_outputs = encoder_layer(
+                    hidden_states,
+                    attention_mask,
+                    causal_attention_mask,
+                    output_attentions=output_attentions,
+                )
+
+                hidden_states = layer_outputs[0]
+
+            if output_attentions:
+                all_attentions = all_attentions + (layer_outputs[1],)
+
+        if output_hidden_states:
+            encoder_states = encoder_states + (hidden_states,)
+
+        if not return_dict:
+            return tuple(v for v in [hidden_states, encoder_states, all_attentions] if v is not None)
+        return BaseModelOutput(
+            last_hidden_state=hidden_states, hidden_states=encoder_states, attentions=all_attentions
+        )
+
+
+class CLIPTextTransformer(nn.Module):
+    def __init__(self, config: CLIPTextConfig):
+        super().__init__()
+        self.config = config
+        embed_dim = config.hidden_size
+        self.embeddings = CLIPTextEmbeddings(config)
+        self.encoder = CLIPEncoder(config)
+        self.final_layer_norm = nn.LayerNorm(embed_dim)
+
+    @add_start_docstrings_to_model_forward(CLIP_TEXT_INPUTS_DOCSTRING)
+    @replace_return_docstrings(output_type=BaseModelOutputWithPooling, config_class=CLIPTextConfig)
+    def forward(
+        self,
+        input_ids=None,
+        attention_mask=None,
+        position_ids=None,
+        output_attentions=None,
+        output_hidden_states=None,
+        return_dict=None,
+    ):
+        r"""
+        Returns:
+
+        """
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        if input_ids is None:
+            raise ValueError("You have to specify either input_ids")
+
+        input_shape = input_ids.size()
+        input_ids = input_ids.view(-1, input_shape[-1])
+
+        hidden_states = self.embeddings(input_ids=input_ids, position_ids=position_ids)
+
+        bsz, seq_len = input_shape
+        # CLIP's text model uses causal mask, prepare it here.
+        # https://github.com/openai/CLIP/blob/cfcffb90e69f37bf2ff1e988237a0fbe41f33c04/clip/model.py#L324
+        causal_attention_mask = self._build_causal_attention_mask(bsz, seq_len).to(hidden_states.device)
+        # expand attention_mask
+        if attention_mask is not None:
+            # [bsz, seq_len] -> [bsz, 1, tgt_seq_len, src_seq_len]
+            attention_mask = _expand_mask(attention_mask, hidden_states.dtype)
+
+        encoder_outputs = self.encoder(
+            inputs_embeds=hidden_states,
+            attention_mask=attention_mask,
+            causal_attention_mask=causal_attention_mask,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        last_hidden_state = encoder_outputs[0]
+        last_hidden_state = self.final_layer_norm(last_hidden_state)
+
+        # text_embeds.shape = [batch_size, n_ctx, transformer.width]
+        # take features from the eot embedding (eot_token is the highest number in each sequence)
+        pooled_output = last_hidden_state[torch.arange(last_hidden_state.shape[0]), input_ids.argmax(dim=-1)]
+
+        if not return_dict:
+            return (last_hidden_state, pooled_output) + encoder_outputs[1:]
+
+        return BaseModelOutputWithPooling(
+            last_hidden_state=last_hidden_state,
+            pooler_output=pooled_output,
+            hidden_states=encoder_outputs.hidden_states,
+            attentions=encoder_outputs.attentions,
+        )
+
+    def _build_causal_attention_mask(self, bsz, seq_len):
+        # lazily create causal attention mask, with full attention between the vision tokens
+        # pytorch uses additive attention mask; fill with -inf
+        mask = torch.empty(bsz, seq_len, seq_len)
+        mask.fill_(float("-inf"))
+        mask.triu_(1)  # zero out the lower diagonal
+        mask = mask.unsqueeze(1)  # expand mask
+        return mask
+
+
+class CLIPTextModel(CLIPPreTrainedModel):
+    config_class = CLIPTextConfig
+
+    def __init__(self, config: CLIPTextConfig):
+        super().__init__(config)
+        self.text_model = CLIPTextTransformer(config)
+        self.init_weights()
+
+    def get_input_embeddings(self) -> nn.Module:
+        return self.text_model.embeddings.token_embedding
+
+    def set_input_embeddings(self, value):
+        self.text_model.embeddings.token_embedding = value
+
+    @add_start_docstrings_to_model_forward(CLIP_TEXT_INPUTS_DOCSTRING)
+    @replace_return_docstrings(output_type=BaseModelOutputWithPooling, config_class=CLIPTextConfig)
+    def forward(
+        self,
+        input_ids=None,
+        attention_mask=None,
+        position_ids=None,
+        output_attentions=None,
+        output_hidden_states=None,
+        return_dict=None,
+    ):
+        r"""
+        Returns:
+
+        """
+        return self.text_model(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+
+class CLIPVisionTransformer(nn.Module):
+    def __init__(self, config: CLIPVisionConfig):
+        super().__init__()
+        self.config = config
+        embed_dim = config.hidden_size
+
+        self.embeddings = CLIPVisionEmbeddings(config)
+        self.pre_layrnorm = nn.LayerNorm(embed_dim)
+        self.encoder = CLIPEncoder(config)
+        self.post_layernorm = nn.LayerNorm(embed_dim)
+
+    @add_start_docstrings_to_model_forward(CLIP_VISION_INPUTS_DOCSTRING)
+    @replace_return_docstrings(output_type=BaseModelOutputWithPooling, config_class=CLIPVisionConfig)
+    def forward(
+        self,
+        pixel_values=None,
+        output_attentions=None,
+        output_hidden_states=None,
+        return_dict=None,
+    ):
+        r"""
+        Returns:
+
+        """
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        if pixel_values is None:
+            raise ValueError("You have to specify pixel_values")
+
+        hidden_states = self.embeddings(pixel_values)
+        hidden_states = self.pre_layrnorm(hidden_states)
+
+        encoder_outputs = self.encoder(
+            inputs_embeds=hidden_states,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        last_hidden_state = encoder_outputs[0]
+        pooled_output = last_hidden_state[:, 0, :]
+        pooled_output = self.post_layernorm(pooled_output)
+
+        if not return_dict:
+            return (last_hidden_state, pooled_output) + encoder_outputs[1:]
+
+        return BaseModelOutputWithPooling(
+            last_hidden_state=last_hidden_state,
+            pooler_output=pooled_output,
+            hidden_states=encoder_outputs.hidden_states,
+            attentions=encoder_outputs.attentions,
+        )
+
+
+class CLIPVisionModel(CLIPPreTrainedModel):
+    config_class = CLIPVisionConfig
+
+    def __init__(self, config: CLIPVisionConfig):
+        super().__init__(config)
+        self.vision_model = CLIPVisionTransformer(config)
+        self.init_weights()
+
+    def get_input_embeddings(self) -> nn.Module:
+        return self.vision_model.embeddings.patch_embedding
+
+    @add_start_docstrings_to_model_forward(CLIP_VISION_INPUTS_DOCSTRING)
+    @replace_return_docstrings(output_type=BaseModelOutputWithPooling, config_class=CLIPVisionConfig)
+    def forward(
+        self,
+        pixel_values=None,
+        output_attentions=None,
+        output_hidden_states=None,
+        return_dict=None,
+    ):
+        r"""
+        Returns:
+
+        """
+        return self.vision_model(
+            pixel_values=pixel_values,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+
+@add_start_docstrings(CLIP_START_DOCSTRING)
+class CLIPModel(CLIPPreTrainedModel):
+    config_class = CLIPConfig
+
+    def __init__(self, config: CLIPConfig):
+        super().__init__(config)
+
+        if not isinstance(config.text_config, CLIPTextConfig):
+            raise ValueError(
+                f"config.text_config is expected to be of type CLIPTextConfig but is of type {type(config.text_config)}."
+            )
+
+        if not isinstance(config.vision_config, CLIPVisionConfig):
+            raise ValueError(
+                f"config.vision_config is expected to be of type CLIPVisionConfig but is of type {type(config.vision_config)}."
+            )
+
+        text_config = config.text_config
+        vision_config = config.vision_config
+
+        self.projection_dim = config.projection_dim
+        self.text_embed_dim = text_config.hidden_size
+        self.vision_embed_dim = vision_config.hidden_size
+
+        self.text_model = CLIPTextTransformer(text_config)
+        self.vision_model = CLIPVisionTransformer(vision_config)
+
+        self.visual_projection = nn.Linear(self.vision_embed_dim, self.projection_dim, bias=False)
+        self.text_projection = nn.Linear(self.text_embed_dim, self.projection_dim, bias=False)
+        self.logit_scale = nn.Parameter(torch.ones([]))
+
+        self.init_weights()
+
+    @add_start_docstrings_to_model_forward(CLIP_TEXT_INPUTS_DOCSTRING)
+    def get_text_features(
+        self,
+        input_ids=None,
+        attention_mask=None,
+        position_ids=None,
+        output_attentions=None,
+        output_hidden_states=None,
+        return_dict=None,
+    ):
+        r"""
+        Returns:
+            text_features (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, output_dim`): The text embeddings
+            obtained by applying the projection layer to the pooled output of :class:`~transformers.CLIPTextModel`.
+        """
+        text_outputs = self.text_model(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        pooled_output = text_outputs[1]
+        text_features = self.text_projection(pooled_output)
+
+        return text_features
+
+    @add_start_docstrings_to_model_forward(CLIP_VISION_INPUTS_DOCSTRING)
+    def get_image_features(
+        self,
+        pixel_values=None,
+        output_attentions=None,
+        output_hidden_states=None,
+        return_dict=None,
+    ):
+        r"""
+        Returns:
+            image_features (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, output_dim`): The image embeddings
+            obtained by applying the projection layer to the pooled output of :class:`~transformers.CLIPVisionModel`.
+        """
+        vision_outputs = self.vision_model(
+            pixel_values=pixel_values,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        pooled_output = vision_outputs[1]  # pooled_output
+        image_features = self.visual_projection(pooled_output)
+
+        return image_features
+
+    @add_start_docstrings_to_model_forward(CLIP_INPUTS_DOCSTRING)
+    @replace_return_docstrings(output_type=CLIPOutput, config_class=CLIPConfig)
+    def forward(
+        self,
+        input_ids=None,
+        pixel_values=None,
+        attention_mask=None,
+        position_ids=None,
+        return_loss=None,
+        output_attentions=None,
+        output_hidden_states=None,
+        return_dict=None,
+    ):
+        r"""
+        Returns:
+
+        """
+        return_dict = return_dict if return_dict is not None else self.config.return_dict
+        vision_outputs = self.vision_model(
+            pixel_values=pixel_values,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        text_outputs = self.text_model(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        image_embeds = vision_outputs[1]
+        image_embeds = self.visual_projection(image_embeds)
+
+        text_embeds = text_outputs[1]
+        text_embeds = self.text_projection(text_embeds)
+
+        # normalized features
+        image_embeds = image_embeds / image_embeds.norm(dim=-1, keepdim=True)
+        text_embeds = text_embeds / text_embeds.norm(dim=-1, keepdim=True)
+
+        # cosine similarity as logits
+        logit_scale = self.logit_scale.exp()
+        logits_per_text = torch.matmul(text_embeds, image_embeds.t()) * logit_scale
+        logits_per_image = logits_per_text.T
+
+        loss = None
+        if return_loss:
+            loss = clip_loss(logits_per_text)
+
+        if not return_dict:
+            output = (logits_per_image, logits_per_text, text_embeds, image_embeds, text_outputs, vision_outputs)
+            return ((loss,) + output) if loss is not None else output
+
+        return CLIPOutput(
+            loss=loss,
+            logits_per_image=logits_per_image,
+            logits_per_text=logits_per_text,
+            text_embeds=text_embeds,
+            image_embeds=image_embeds,
+            text_model_output=text_outputs,
+            vision_model_output=vision_outputs,
+        )

src/transformers/models/clip/processing_clip.py

+# coding=utf-8
+# Copyright 2021 The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""
+Image/Text processor class for CLIP
+"""
+from ...tokenization_utils_base import BatchEncoding
+from .feature_extraction_clip import CLIPFeatureExtractor
+from .tokenization_clip import CLIPTokenizer
+
+
+class CLIPProcessor:
+    r"""
+    Constructs a CLIP processor which wraps a CLIP feature extractor and a CLIP tokenizer into a single processor.
+
+    :class:`~transformers.CLIPProcessor` offers all the functionalities of :class:`~transformers.CLIPFeatureExtractor`
+    and :class:`~transformers.CLIPTokenizer`. See the :meth:`~transformers.CLIPProcessor.__call__` and
+    :meth:`~transformers.CLIPProcessor.decode` for more information.
+
+    Args:
+        feature_extractor (:class:`~transformers.CLIPFeatureExtractor`):
+            The feature extractor is a required input.
+        tokenizer (:class:`~transformers.CLIPTokenizer`):
+            The tokenizer is a required input.
+    """
+
+    def __init__(self, feature_extractor, tokenizer):
+        if not isinstance(feature_extractor, CLIPFeatureExtractor):
+            raise ValueError(
+                f"`feature_extractor` has to be of type CLIPFeatureExtractor, but is {type(feature_extractor)}"
+            )
+        if not isinstance(tokenizer, CLIPTokenizer):
+            raise ValueError(f"`tokenizer` has to be of type CLIPTokenizer, but is {type(tokenizer)}")
+
+        self.feature_extractor = feature_extractor
+        self.tokenizer = tokenizer
+        self.current_processor = self.feature_extractor
+
+    def save_pretrained(self, save_directory):
+        """
+        Save a CLIP feature extractor object and CLIP tokenizer object to the directory ``save_directory``, so that it
+        can be re-loaded using the :func:`~transformers.CLIPProcessor.from_pretrained` class method.
+
+        .. note::
+
+            This class method is simply calling :meth:`~transformers.PreTrainedFeatureExtractor.save_pretrained` and
+            :meth:`~transformers.tokenization_utils_base.PreTrainedTokenizer.save_pretrained`. Please refer to the
+            docstrings of the methods above for more information.
+
+        Args:
+            save_directory (:obj:`str` or :obj:`os.PathLike`):
+                Directory where the feature extractor JSON file and the tokenizer files will be saved (directory will
+                be created if it does not exist).
+        """
+
+        self.feature_extractor.save_pretrained(save_directory)
+        self.tokenizer.save_pretrained(save_directory)
+
+    @classmethod
+    def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
+        r"""
+        Instantiate a :class:`~transformers.CLIPProcessor` from a pretrained CLIP processor.
+
+        .. note::
+
+            This class method is simply calling CLIPFeatureExtractor's
+            :meth:`~transformers.PreTrainedFeatureExtractor.from_pretrained` and CLIPTokenizer's
+            :meth:`~transformers.tokenization_utils_base.PreTrainedTokenizer.from_pretrained`. Please refer to the
+            docstrings of the methods above for more information.
+
+        Args:
+            pretrained_model_name_or_path (:obj:`str` or :obj:`os.PathLike`):
+                This can be either:
+
+                - a string, the `model id` of a pretrained feature_extractor hosted inside a model repo on
+                  huggingface.co. Valid model ids can be located at the root-level, like ``clip-vit-base-patch32``, or
+                  namespaced under a user or organization name, like ``openai/clip-vit-base-patch32``.
+                - a path to a `directory` containing a feature extractor file saved using the
+                  :meth:`~transformers.PreTrainedFeatureExtractor.save_pretrained` method, e.g.,
+                  ``./my_model_directory/``.
+                - a path or url to a saved feature extractor JSON `file`, e.g.,
+                  ``./my_model_directory/preprocessor_config.json``.
+
+            **kwargs
+                Additional keyword arguments passed along to both :class:`~transformers.PreTrainedFeatureExtractor` and
+                :class:`~transformers.PreTrainedTokenizer`
+        """
+        feature_extractor = CLIPFeatureExtractor.from_pretrained(pretrained_model_name_or_path, **kwargs)
+        tokenizer = CLIPTokenizer.from_pretrained(pretrained_model_name_or_path, **kwargs)
+
+        return cls(feature_extractor=feature_extractor, tokenizer=tokenizer)
+
+    def __call__(self, text=None, images=None, return_tensors=None, **kwargs):
+        """
+        Main method to prepare for the model one or several sequences(s) and image(s). This method forwards the
+        :obj:`text` and :obj:`kwargs` arguments to CLIPTokenizer's :meth:`~transformers.CLIPTokenizer.__call__` if
+        :obj:`text` is not :obj:`None` to encode the text. To prepare the image(s), this method forwards the
+        :obj:`images` and :obj:`kwrags` arguments to CLIPFeatureExtractor's
+        :meth:`~transformers.CLIPFeatureExtractor.__call__` if :obj:`images` is not :obj:`None`. Please refer to the
+        doctsring of the above two methods for more information.
+
+        Args:
+            text (:obj:`str`, :obj:`List[str]`, :obj:`List[List[str]]`):
+                The sequence or batch of sequences to be encoded. Each sequence can be a string or a list of strings
+                (pretokenized string). If the sequences are provided as list of strings (pretokenized), you must set
+                :obj:`is_split_into_words=True` (to lift the ambiguity with a batch of sequences).
+            images (:obj:`PIL.Image.Image`, :obj:`np.ndarray`, :obj:`torch.Tensor`, :obj:`List[PIL.Image.Image]`, :obj:`List[np.ndarray]`, :obj:`List[torch.Tensor]`):
+                The image or batch of images to be prepared. Each image can be a PIL image, NumPy array or PyTorch
+                tensor. In case of a NumPy array/PyTorch tensor, each image should be of shape (C, H, W), where C is a
+                number of channels, H and W are image height and width.
+
+            return_tensors (:obj:`str` or :class:`~transformers.file_utils.TensorType`, `optional`):
+                If set, will return tensors of a particular framework. Acceptable values are:
+
+                * :obj:`'tf'`: Return TensorFlow :obj:`tf.constant` objects.
+                * :obj:`'pt'`: Return PyTorch :obj:`torch.Tensor` objects.
+                * :obj:`'np'`: Return NumPy :obj:`np.ndarray` objects.
+                * :obj:`'jax'`: Return JAX :obj:`jnp.ndarray` objects.
+
+        Returns:
+            :class:`~transformers.BatchEncoding`: A :class:`~transformers.BatchEncoding` with the following fields:
+
+            - **input_ids** -- List of token ids to be fed to a model. Returned when :obj:`text` is not :obj:`None`.
+            - **attention_mask** -- List of indices specifying which tokens should be attended to by the model (when
+              :obj:`return_attention_mask=True` or if `"attention_mask"` is in :obj:`self.model_input_names` and if
+              :obj:`text` is not :obj:`None`).
+            - **pixel_values** -- Pixel values to be fed to a model. Returned when :obj:`images` is not :obj:`None`.
+        """
+
+        if text is None and images is None:
+            raise ValueError("You have to specify either text or images. Both cannot be none.")
+
+        if text is not None:
+            encoding = self.tokenizer(text, return_tensors=return_tensors, **kwargs)
+
+        if images is not None:
+            image_features = self.feature_extractor(images, return_tensors=return_tensors, **kwargs)
+
+        if text is not None and images is not None:
+            encoding["pixel_values"] = image_features.pixel_values
+            return encoding
+        elif text is not None:
+            return encoding
+        else:
+            return BatchEncoding(data=dict(**image_features), tensor_type=return_tensors)
+
+    def batch_decode(self, *args, **kwargs):
+        """
+        This method forwards all its arguments to CLIPTokenizer's
+        :meth:`~transformers.PreTrainedTokenizer.batch_decode`. Please refer to the docstring of this method for more
+        information.
+        """
+        return self.tokenizer.batch_decode(*args, **kwargs)
+
+    def decode(self, *args, **kwargs):
+        """
+        This method forwards all its arguments to CLIPTokenizer's :meth:`~transformers.PreTrainedTokenizer.decode`.
+        Please refer to the docstring of this method for more information.
+        """
+        return self.tokenizer.decode(*args, **kwargs)

src/transformers/models/clip/tokenization_clip.py

+# coding=utf-8
+# Copyright 2021 The Open AI Team Authors and The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Tokenization classes for CLIP."""
+
+import json
+import os
+from functools import lru_cache
+from typing import List, Optional, Tuple
+
+import regex as re
+from transformers.models.bert.tokenization_bert import BasicTokenizer
+
+from ...tokenization_utils import AddedToken, PreTrainedTokenizer
+from ...utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {
+    "vocab_file": "vocab.json",
+    "merges_file": "merges.txt",
+}
+
+PRETRAINED_VOCAB_FILES_MAP = {
+    "vocab_file": {
+        "openai/clip-vit-base-patch32": "https://huggingface.co/openai/clip-vit-base-patch32/resolve/main/vocab.json",
+    },
+    "merges_file": {
+        "openai/clip-vit-base-patch32": "https://huggingface.co/openai/clip-vit-base-patch32/resolve/main/merges.txt",
+    },
+}
+
+PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
+    "openai/clip-vit-base-patch32": 77,
+}
+
+
+PRETRAINED_INIT_CONFIGURATION = {
+    "openai/clip-vit-base-patch32": {"do_lower_case": True},
+}
+
+
+@lru_cache()
+def bytes_to_unicode():
+    """
+    Returns list of utf-8 byte and a mapping to unicode strings. We specifically avoids mapping to whitespace/control
+    characters the bpe code barfs on.
+
+    The reversible bpe codes work on unicode strings. This means you need a large # of unicode characters in your vocab
+    if you want to avoid UNKs. When you're at something like a 10B token dataset you end up needing around 5K for
+    decent coverage. This is a signficant percentage of your normal, say, 32K bpe vocab. To avoid that, we want lookup
+    tables between utf-8 bytes and unicode strings.
+    """
+    bs = (
+        list(range(ord("!"), ord("~") + 1)) + list(range(ord("¡"), ord("¬") + 1)) + list(range(ord("®"), ord("ÿ") + 1))
+    )
+    cs = bs[:]
+    n = 0
+    for b in range(2 ** 8):
+        if b not in bs:
+            bs.append(b)
+            cs.append(2 ** 8 + n)
+            n += 1
+    cs = [chr(n) for n in cs]
+    return dict(zip(bs, cs))
+
+
+def get_pairs(word):
+    """
+    Return set of symbol pairs in a word.
+
+    Word is represented as tuple of symbols (symbols being variable-length strings).
+    """
+    pairs = set()
+    prev_char = word[0]
+    for char in word[1:]:
+        pairs.add((prev_char, char))
+        prev_char = char
+    return pairs
+
+
+def whitespace_clean(text):
+    text = re.sub(r"\s+", " ", text)
+    text = text.strip()
+    return text
+
+
+class CLIPTokenizer(PreTrainedTokenizer):
+    """
+    Construct a CLIP tokenizer. Based on byte-level Byte-Pair-Encoding.
+
+    This tokenizer has been trained to treat spaces like parts of the tokens (a bit like sentencepiece) so a word will
+    be encoded differently whether it is at the beginning of the sentence (without space) or not:
+
+
+    You can get around that behavior by passing ``add_prefix_space=True`` when instantiating this tokenizer or when you
+    call it on some text, but since the model was not pretrained this way, it might yield a decrease in performance.
+
+    .. note::
+
+        When used with ``is_split_into_words=True``, this tokenizer will add a space before each word (even the first
+        one).
+
+    This tokenizer inherits from :class:`~transformers.PreTrainedTokenizer` which contains most of the main methods.
+    Users should refer to this superclass for more information regarding those methods.
+
+    Args:
+        vocab_file (:obj:`str`):
+            Path to the vocabulary file.
+        merges_file (:obj:`str`):
+            Path to the merges file.
+        errors (:obj:`str`, `optional`, defaults to :obj:`"replace"`):
+            Paradigm to follow when decoding bytes to UTF-8. See `bytes.decode
+            <https://docs.python.org/3/library/stdtypes.html#bytes.decode>`__ for more information.
+        unk_token (:obj:`str`, `optional`, defaults to :obj:`<|endoftext|>`):
+            The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this
+            token instead.
+        bos_token (:obj:`str`, `optional`, defaults to :obj:`<|endoftext|>`):
+            The beginning of sequence token.
+        eos_token (:obj:`str`, `optional`, defaults to :obj:`<|endoftext|>`):
+            The end of sequence token.
+        add_prefix_space (:obj:`bool`, `optional`, defaults to :obj:`False`):
+            Whether or not to add an initial space to the input. This allows to treat the leading word just as any
+            other word. (CLIP tokenizer detect beginning of words by the preceding space).
+    """
+
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+    model_input_names = ["input_ids", "attention_mask"]
+
+    def __init__(
+        self,
+        vocab_file,
+        merges_file,
+        errors="replace",
+        unk_token="<|endoftext|>",
+        bos_token="<|startoftext|>",
+        eos_token="<|endoftext|>",
+        pad_token="<|endoftext|>",  # hack to enable padding
+        add_prefix_space=False,
+        do_lower_case=True,
+        **kwargs
+    ):
+        bos_token = AddedToken(bos_token, lstrip=False, rstrip=False) if isinstance(bos_token, str) else bos_token
+        eos_token = AddedToken(eos_token, lstrip=False, rstrip=False) if isinstance(eos_token, str) else eos_token
+        unk_token = AddedToken(unk_token, lstrip=False, rstrip=False) if isinstance(unk_token, str) else unk_token
+
+        super().__init__(
+            errors=errors,
+            unk_token=unk_token,
+            bos_token=bos_token,
+            eos_token=eos_token,
+            pad_token=pad_token,
+            add_prefix_space=add_prefix_space,
+            do_lower_case=do_lower_case,
+            **kwargs,
+        )
+
+        try:
+            import ftfy
+
+            self.fix_text = ftfy.fix_text
+        except ImportError:
+            logger.warning("ftfy or spacy is not installed using BERT BasicTokenizer instead of ftfy.")
+            self.nlp = BasicTokenizer(do_lower_case=True)
+            self.fix_text = None
+
+        with open(vocab_file, encoding="utf-8") as vocab_handle:
+            self.encoder = json.load(vocab_handle)
+        self.decoder = {v: k for k, v in self.encoder.items()}
+        self.errors = errors  # how to handle errors in decoding
+        self.byte_encoder = bytes_to_unicode()
+        self.byte_decoder = {v: k for k, v in self.byte_encoder.items()}
+        with open(merges_file, encoding="utf-8") as merges_handle:
+            bpe_merges = merges_handle.read().split("\n")[1 : 49152 - 256 - 2 + 1]
+        bpe_merges = [tuple(merge.split()) for merge in bpe_merges]
+        self.bpe_ranks = dict(zip(bpe_merges, range(len(bpe_merges))))
+        self.cache = {"<|startoftext|>": "<|startoftext|>", "<|endoftext|>": "<|endoftext|>"}
+        self.add_prefix_space = add_prefix_space
+
+        self.pat = re.compile(
+            r"""<\|startoftext\|>|<\|endoftext\|>|'s|'t|'re|'ve|'m|'ll|'d|[\p{L}]+|[\p{N}]|[^\s\p{L}\p{N}]+""",
+            re.IGNORECASE,
+        )
+
+    # Very ugly hack to enable padding
+    @property
+    def pad_token_id(self) -> Optional[int]:
+        """
+        :obj:`Optional[int]`: Id of the padding token in the vocabulary. Returns :obj:`None` if the token has not been
+        set.
+        """
+        return 0
+
+    @property
+    def vocab_size(self):
+        return len(self.encoder)
+
+    def get_vocab(self):
+        return dict(self.encoder, **self.added_tokens_encoder)
+
+    def build_inputs_with_special_tokens(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and
+        adding special tokens. A CLIP sequence has the following format:
+
+        - single sequence: ``<|startoftext|> X <|endoftext|>``
+
+        Pairs of sequences are not the expected use case, but they will be handled without a separator.
+
+        Args:
+            token_ids_0 (:obj:`List[int]`):
+                List of IDs to which the special tokens will be added.
+            token_ids_1 (:obj:`List[int]`, `optional`):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            :obj:`List[int]`: List of `input IDs <../glossary.html#input-ids>`__ with the appropriate special tokens.
+        """
+        if token_ids_1 is None:
+            return [self.bos_token_id] + token_ids_0 + [self.eos_token_id]
+        return [self.bos_token_id] + token_ids_0 + token_ids_1 + [self.eos_token_id]
+
+    def get_special_tokens_mask(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
+    ) -> List[int]:
+        """
+        Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
+        special tokens using the tokenizer ``prepare_for_model`` method.
+
+        Args:
+            token_ids_0 (:obj:`List[int]`):
+                List of IDs.
+            token_ids_1 (:obj:`List[int]`, `optional`):
+                Optional second list of IDs for sequence pairs.
+            already_has_special_tokens (:obj:`bool`, `optional`, defaults to :obj:`False`):
+                Whether or not the token list is already formatted with special tokens for the model.
+
+        Returns:
+            :obj:`List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
+        """
+
+        if already_has_special_tokens:
+            return super().get_special_tokens_mask(
+                token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
+            )
+
+        if token_ids_1 is None:
+            return [1] + ([0] * len(token_ids_0)) + [1]
+        return [1] + ([0] * len(token_ids_0)) + ([0] * len(token_ids_1)) + [1]
+
+    def bpe(self, token):
+        if token in self.cache:
+            return self.cache[token]
+        word = tuple(token[:-1]) + (token[-1] + "</w>",)
+        pairs = get_pairs(word)
+
+        if not pairs:
+            return token + "</w>"
+
+        while True:
+            bigram = min(pairs, key=lambda pair: self.bpe_ranks.get(pair, float("inf")))
+            if bigram not in self.bpe_ranks:
+                break
+            first, second = bigram
+            new_word = []
+            i = 0
+            while i < len(word):
+                try:
+                    j = word.index(first, i)
+                except ValueError:
+                    new_word.extend(word[i:])
+                    break
+                else:
+                    new_word.extend(word[i:j])
+                    i = j
+
+                if word[i] == first and i < len(word) - 1 and word[i + 1] == second:
+                    new_word.append(first + second)
+                    i += 2
+                else:
+                    new_word.append(word[i])
+                    i += 1
+            new_word = tuple(new_word)
+            word = new_word
+            if len(word) == 1:
+                break
+            else:
+                pairs = get_pairs(word)
+        word = " ".join(word)
+        self.cache[token] = word
+        return word
+
+    def _tokenize(self, text):
+        """Tokenize a string."""
+        bpe_tokens = []
+        if self.fix_text is None:
+            text = " ".join(self.nlp.tokenize(text))
+        else:
+            text = whitespace_clean(self.fix_text(text)).lower()
+
+        for token in re.findall(self.pat, text):
+            token = "".join(
+                self.byte_encoder[b] for b in token.encode("utf-8")
+            )  # Maps all our bytes to unicode strings, avoiding controle tokens of the BPE (spaces in our case)
+            bpe_tokens.extend(bpe_token for bpe_token in self.bpe(token).split(" "))
+        return bpe_tokens
+
+    def _convert_token_to_id(self, token):
+        """Converts a token (str) in an id using the vocab."""
+        return self.encoder.get(token, self.encoder.get(self.unk_token))
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (str) using the vocab."""
+        return self.decoder.get(index)
+
+    def convert_tokens_to_string(self, tokens):
+        """Converts a sequence of tokens (string) in a single string."""
+        text = "".join(tokens)
+        text = bytearray([self.byte_decoder[c] for c in text]).decode("utf-8", errors=self.errors).replace("</w>", " ")
+        return text
+
+    def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        if not os.path.isdir(save_directory):
+            logger.error("Vocabulary path ({}) should be a directory".format(save_directory))
+            return
+        vocab_file = os.path.join(
+            save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
+        )
+        merge_file = os.path.join(
+            save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["merges_file"]
+        )
+
+        with open(vocab_file, "w", encoding="utf-8") as f:
+            f.write(json.dumps(self.encoder, ensure_ascii=False))
+
+        index = 0
+        with open(merge_file, "w", encoding="utf-8") as writer:
+            writer.write("#version: 0.2\n")
+            for bpe_tokens, token_index in sorted(self.bpe_ranks.items(), key=lambda kv: kv[1]):
+                if index != token_index:
+                    logger.warning(
+                        "Saving vocabulary to {}: BPE merge indices are not consecutive."
+                        " Please check that the tokenizer is not corrupted!".format(merge_file)
+                    )
+                    index = token_index
+                writer.write(" ".join(bpe_tokens) + "\n")
+                index += 1
+
+        return vocab_file, merge_file
+
+    def prepare_for_tokenization(self, text, is_split_into_words=False, **kwargs):
+        add_prefix_space = kwargs.pop("add_prefix_space", self.add_prefix_space)
+        if is_split_into_words or add_prefix_space:
+            text = " " + text
+        return (text, kwargs)

src/transformers/models/clip/tokenization_clip_fast.py

+# coding=utf-8
+# Copyright 2021 The Open AI Team Authors and The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Tokenization classes for OpenAI GPT."""
+
+
+import json
+from typing import Optional, Tuple
+
+from tokenizers import pre_tokenizers
+
+from ...tokenization_utils_base import BatchEncoding
+from ...tokenization_utils_fast import PreTrainedTokenizerFast
+from ...utils import logging
+from .tokenization_clip import CLIPTokenizer
+
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "vocab.json", "merges_file": "merges.txt", "tokenizer_file": "tokenizer.json"}
+
+PRETRAINED_VOCAB_FILES_MAP = {
+    "vocab_file": {
+        "openai/clip-vit-base-patch32": "https://huggingface.co/openai/clip-vit-base-patch32/resolve/main/vocab.json",
+    },
+    "merges_file": {
+        "openai/clip-vit-base-patch32": "https://huggingface.co/openai/clip-vit-base-patch32/resolve/main/merges.txt",
+    },
+    "tokenizer_file": {
+        "openai/clip-vit-base-patch32": "https://huggingface.co/openai/clip-vit-base-patch32/resolve/main/tokenizer.json",
+    },
+}
+
+PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
+    "openai/clip-vit-base-patch32": 77,
+}
+
+
+class CLIPTokenizerFast(PreTrainedTokenizerFast):
+    """
+    Construct a "fast" CLIP tokenizer (backed by HuggingFace's `tokenizers` library). Based on byte-level
+    Byte-Pair-Encoding.
+
+    This tokenizer has been trained to treat spaces like parts of the tokens (a bit like sentencepiece) so a word will
+    be encoded differently whether it is at the beginning of the sentence (without space) or not:
+
+    ::
+
+        >>> from transformers import CLIPTokenizerFast
+        >>> tokenizer = CLIPTokenizerFast.from_pretrained("openai/clip-vit-base-patch32")
+        >>> tokenizer("Hello world")['input_ids']
+        [15496, 995]
+        >>> tokenizer(" Hello world")['input_ids']
+        [18435, 995]
+
+    You can get around that behavior by passing ``add_prefix_space=True`` when instantiating this tokenizer or when you
+    call it on some text, but since the model was not pretrained this way, it might yield a decrease in performance.
+
+    .. note::
+
+        When used with ``is_split_into_words=True``, this tokenizer needs to be instantiated with
+        ``add_prefix_space=True``.
+
+    This tokenizer inherits from :class:`~transformers.PreTrainedTokenizerFast` which contains most of the main
+    methods. Users should refer to this superclass for more information regarding those methods.
+
+    Args:
+        vocab_file (:obj:`str`):
+            Path to the vocabulary file.
+        merges_file (:obj:`str`):
+            Path to the merges file.
+        errors (:obj:`str`, `optional`, defaults to :obj:`"replace"`):
+            Paradigm to follow when decoding bytes to UTF-8. See `bytes.decode
+            <https://docs.python.org/3/library/stdtypes.html#bytes.decode>`__ for more information.
+        unk_token (:obj:`str`, `optional`, defaults to :obj:`<|endoftext|>`):
+            The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this
+            token instead.
+        bos_token (:obj:`str`, `optional`, defaults to :obj:`<|endoftext|>`):
+            The beginning of sequence token.
+        eos_token (:obj:`str`, `optional`, defaults to :obj:`<|endoftext|>`):
+            The end of sequence token.
+        add_prefix_space (:obj:`bool`, `optional`, defaults to :obj:`False`):
+            Whether or not to add an initial space to the input. This allows to treat the leading word just as any
+            other word. (CLIP tokenizer detect beginning of words by the preceding space).
+        trim_offsets (:obj:`bool`, `optional`, defaults to :obj:`True`):
+            Whether or not the post-processing step should trim offsets to avoid including whitespaces.
+    """
+
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+    model_input_names = ["input_ids", "attention_mask"]
+    slow_tokenizer_class = CLIPTokenizer
+
+    def __init__(
+        self,
+        vocab_file,
+        merges_file,
+        tokenizer_file=None,
+        unk_token="<|endoftext|>",
+        bos_token="<|startoftext|>",
+        eos_token="<|endoftext|>",
+        pad_token="<|endoftext|>",  # hack to enable padding
+        add_prefix_space=False,
+        **kwargs
+    ):
+        super().__init__(
+            vocab_file,
+            merges_file,
+            tokenizer_file=tokenizer_file,
+            unk_token=unk_token,
+            bos_token=bos_token,
+            eos_token=eos_token,
+            pad_token=pad_token,
+            add_prefix_space=add_prefix_space,
+            **kwargs,
+        )
+
+        pre_tok_state = json.loads(self.backend_tokenizer.pre_tokenizer.__getstate__())
+        if pre_tok_state.get("add_prefix_space", add_prefix_space) != add_prefix_space:
+            pre_tok_class = getattr(pre_tokenizers, pre_tok_state.pop("type"))
+            pre_tok_state["add_prefix_space"] = add_prefix_space
+            self.backend_tokenizer.pre_tokenizer = pre_tok_class(**pre_tok_state)
+
+        self.add_prefix_space = add_prefix_space
+
+    # Very ugly hack to enable padding
+    @property
+    def pad_token_id(self) -> Optional[int]:
+        """
+        :obj:`Optional[int]`: Id of the padding token in the vocabulary. Returns :obj:`None` if the token has not been
+        set.
+        """
+        return 0
+
+    def _batch_encode_plus(self, *args, **kwargs) -> BatchEncoding:
+        is_split_into_words = kwargs.get("is_split_into_words", False)
+        assert self.add_prefix_space or not is_split_into_words, (
+            f"You need to instantiate {self.__class__.__name__} with add_prefix_space=True "
+            "to use it with pretokenized inputs."
+        )
+
+        return super()._batch_encode_plus(*args, **kwargs)
+
+    def _encode_plus(self, *args, **kwargs) -> BatchEncoding:
+        is_split_into_words = kwargs.get("is_split_into_words", False)
+
+        assert self.add_prefix_space or not is_split_into_words, (
+            f"You need to instantiate {self.__class__.__name__} with add_prefix_space=True "
+            "to use it with pretokenized inputs."
+        )
+
+        return super()._encode_plus(*args, **kwargs)
+
+    def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        files = self._tokenizer.model.save(save_directory, name=filename_prefix)
+        return tuple(files)

src/transformers/utils/dummy_pt_objects.py

@@ -888,6 +888,45 @@ class CamembertModel:
         requires_backends(self, ["torch"])
 
 
+CLIP_PRETRAINED_MODEL_ARCHIVE_LIST = None
+
+
+class CLIPModel:
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+    @classmethod
+    def from_pretrained(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+
+class CLIPPreTrainedModel:
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+    @classmethod
+    def from_pretrained(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+
+class CLIPTextModel:
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+    @classmethod
+    def from_pretrained(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+
+class CLIPVisionModel:
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+    @classmethod
+    def from_pretrained(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+
 CONVBERT_PRETRAINED_MODEL_ARCHIVE_LIST = None
 
 

src/transformers/utils/dummy_tokenizers_objects.py

@@ -56,6 +56,15 @@ class CamembertTokenizerFast:
         requires_backends(self, ["tokenizers"])
 
 
+class CLIPTokenizerFast:
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["tokenizers"])
+
+    @classmethod
+    def from_pretrained(self, *args, **kwargs):
+        requires_backends(self, ["tokenizers"])
+
+
 class ConvBertTokenizerFast:
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["tokenizers"])

src/transformers/utils/dummy_vision_objects.py

@@ -7,6 +7,16 @@ class ImageFeatureExtractionMixin:
         requires_backends(self, ["vision"])
 
 
+class CLIPFeatureExtractor:
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["vision"])
+
+
+class CLIPProcessor:
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["vision"])
+
+
 class DeiTFeatureExtractor:
     def __init__(self, *args, **kwargs):
         requires_backends(self, ["vision"])