Add LayoutLMv2 + LayoutXLM

.circleci/config.yml

@@ -798,6 +798,44 @@ jobs:
             - run: pip install requests
             - run: python ./utils/link_tester.py
 
+    run_tests_layoutlmv2:
+        working_directory: ~/transformers
+        docker:
+            - image: circleci/python:3.7
+        environment:
+            OMP_NUM_THREADS: 1
+            TRANSFORMERS_IS_CI: yes
+        resource_class: xlarge
+        parallelism: 1
+        steps:
+            - checkout
+            - restore_cache:
+                  keys:
+                      - v0.4-torch-{{ checksum "setup.py" }}
+                      - v0.4-{{ checksum "setup.py" }}
+            - run: sudo apt-get -y update && sudo apt-get install -y libsndfile1-dev
+            - run: pip install --upgrade pip
+            - run: pip install .[torch,testing,vision]
+            - run: pip install torchvision
+            - run: python -m pip install 'git+https://github.com/facebookresearch/detectron2.git'
+            - run: sudo apt install tesseract-ocr
+            - run: pip install pytesseract
+            - save_cache:
+                  key: v0.4-torch-{{ checksum "setup.py" }}
+                  paths:
+                      - '~/.cache/pip'
+            - run: python utils/tests_fetcher.py | tee test_preparation.txt
+            - store_artifacts:
+                  path: ~/transformers/test_preparation.txt
+            - run: |
+                  if [ -f test_list.txt ]; then
+                    python -m pytest -n 1 tests/*layoutlmv2* --dist=loadfile -s --make-reports=tests_layoutlmv2 --durations=100
+                  fi
+            - store_artifacts:
+                  path: ~/transformers/tests_output.txt
+            - store_artifacts:
+                  path: ~/transformers/reports
+
 # TPU JOBS
     run_examples_tpu:
         docker:
@@ -852,6 +890,7 @@ workflows:
             - run_tests_onnxruntime
             - run_tests_hub
             - build_doc
+            - run_tests_layoutlmv2
             - deploy_doc: *workflow_filters
     nightly:
         triggers:

README.md

@@ -244,6 +244,8 @@ Min, Patrick Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, and Wen-tau Yih.
 1. **[Hubert](https://huggingface.co/transformers/model_doc/hubert.html)** (from Facebook) released with the paper [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://arxiv.org/abs/2106.07447) by Wei-Ning Hsu, Benjamin Bolte, Yao-Hung Hubert Tsai, Kushal Lakhotia, Ruslan Salakhutdinov, Abdelrahman Mohamed.
 1. **[I-BERT](https://huggingface.co/transformers/model_doc/ibert.html)** (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
 1. **[LayoutLM](https://huggingface.co/transformers/model_doc/layoutlm.html)** (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.
+1. **[LayoutLMv2](https://huggingface.co/transformers/model_doc/layoutlmv2.html)** (from Microsoft Research Asia) released with the paper [LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding](https://arxiv.org/abs/2012.14740) by Yang Xu, Yiheng Xu, Tengchao Lv, Lei Cui, Furu Wei, Guoxin Wang, Yijuan Lu, Dinei Florencio, Cha Zhang, Wanxiang Che, Min Zhang, Lidong Zhou.
+1. **[LayoutXLM](https://huggingface.co/transformers/model_doc/layoutlmv2.html)** (from Microsoft Research Asia) released with the paper [LayoutXLM: Multimodal Pre-training for Multilingual Visually-rich Document Understanding](https://arxiv.org/abs/2104.08836) by Yiheng Xu, Tengchao Lv, Lei Cui, Guoxin Wang, Yijuan Lu, Dinei Florencio, Cha Zhang, Furu Wei.
 1. **[LED](https://huggingface.co/transformers/model_doc/led.html)** (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.
 1. **[Longformer](https://huggingface.co/transformers/model_doc/longformer.html)** (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.
 1. **[LUKE](https://huggingface.co/transformers/model_doc/luke.html)** (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.

docs/source/index.rst

@@ -202,99 +202,106 @@ Supported models
 34. :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.
-35. :doc:`LED <model_doc/led>` (from AllenAI) released with the paper `Longformer: The Long-Document Transformer
+35. :doc:`LayoutLMv2 <model_doc/layoutlmv2>` (from Microsoft Research Asia) released with the paper `LayoutLMv2:
+    Multi-modal Pre-training for Visually-Rich Document Understanding <https://arxiv.org/abs/2012.14740>`__ by Yang Xu,
+    Yiheng Xu, Tengchao Lv, Lei Cui, Furu Wei, Guoxin Wang, Yijuan Lu, Dinei Florencio, Cha Zhang, Wanxiang Che, Min
+    Zhang, Lidong Zhou.
+36. :doc:`LayoutXLM <model_doc/layoutlmv2>` (from Microsoft Research Asia) released with the paper `LayoutXLM:
+    Multimodal Pre-training for Multilingual Visually-rich Document Understanding <https://arxiv.org/abs/2104.08836>`__
+    by Yiheng Xu, Tengchao Lv, Lei Cui, Guoxin Wang, Yijuan Lu, Dinei Florencio, Cha Zhang, Furu Wei.
+37. :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.
-36. :doc:`Longformer <model_doc/longformer>` (from AllenAI) released with the paper `Longformer: The Long-Document
+38. :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.
-37. :doc:`LUKE <model_doc/luke>` (from Studio Ousia) released with the paper `LUKE: Deep Contextualized Entity
+39. :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.
-38. :doc:`LXMERT <model_doc/lxmert>` (from UNC Chapel Hill) released with the paper `LXMERT: Learning Cross-Modality
+40. :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.
-39. :doc:`M2M100 <model_doc/m2m_100>` (from Facebook) released with the paper `Beyond English-Centric Multilingual
+41. :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.
-40. :doc:`MarianMT <model_doc/marian>` Machine translation models trained using `OPUS <http://opus.nlpl.eu/>`__ data by
+42. :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.
-41. :doc:`MBart <model_doc/mbart>` (from Facebook) released with the paper `Multilingual Denoising Pre-training for
+43. :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.
-42. :doc:`MBart-50 <model_doc/mbart>` (from Facebook) released with the paper `Multilingual Translation with Extensible
+44. :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.
-43. :doc:`Megatron-BERT <model_doc/megatron_bert>` (from NVIDIA) released with the paper `Megatron-LM: Training
+45. :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.
-44. :doc:`Megatron-GPT2 <model_doc/megatron_gpt2>` (from NVIDIA) released with the paper `Megatron-LM: Training
+46. :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.
-45. :doc:`MPNet <model_doc/mpnet>` (from Microsoft Research) released with the paper `MPNet: Masked and Permuted
+47. :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.
-46. :doc:`MT5 <model_doc/mt5>` (from Google AI) released with the paper `mT5: A massively multilingual pre-trained
+48. :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.
-47. :doc:`Pegasus <model_doc/pegasus>` (from Google) released with the paper `PEGASUS: Pre-training with Extracted
+49. :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.
-48. :doc:`ProphetNet <model_doc/prophetnet>` (from Microsoft Research) released with the paper `ProphetNet: Predicting
+50. :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.
-49. :doc:`Reformer <model_doc/reformer>` (from Google Research) released with the paper `Reformer: The Efficient
+51. :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.
-50. :doc:`RemBERT <model_doc/rembert>` (from Google Research) released with the paper `Rethinking embedding coupling in
+52. :doc:`RemBERT <model_doc/rembert>` (from Google Research) released with the paper `Rethinking embedding coupling in
     pre-trained language models <https://arxiv.org/pdf/2010.12821.pdf>`__ by Hyung Won Chung, Thibault Févry, Henry
     Tsai, M. Johnson, Sebastian Ruder.
-51. :doc:`RoBERTa <model_doc/roberta>` (from Facebook), released together with the paper a `Robustly Optimized BERT
+53. :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.
-52. :doc:`RoFormer <model_doc/roformer>` (from ZhuiyiTechnology), released together with the paper a `RoFormer:
+54. :doc:`RoFormer <model_doc/roformer>` (from ZhuiyiTechnology), released together with the paper a `RoFormer:
     Enhanced Transformer with Rotary Position Embedding <https://arxiv.org/pdf/2104.09864v1.pdf>`__ by Jianlin Su and
     Yu Lu and Shengfeng Pan and Bo Wen and Yunfeng Liu.
-53. :doc:`SpeechToTextTransformer <model_doc/speech_to_text>` (from Facebook), released together with the paper
+55. :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.
-54. `Splinter <https://huggingface.co/transformers/master/model_doc/splinter.html>`__ (from Tel Aviv University),
+56. `Splinter <https://huggingface.co/transformers/master/model_doc/splinter.html>`__ (from Tel Aviv University),
     released together with the paper `Few-Shot Question Answering by Pretraining Span Selection
     <https://arxiv.org/abs/2101.00438>`__ by Ori Ram, Yuval Kirstain, Jonathan Berant, Amir Globerson, Omer Levy.
-55. :doc:`SqueezeBert <model_doc/squeezebert>` released with the paper `SqueezeBERT: What can computer vision teach NLP
+57. :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.
-56. :doc:`T5 <model_doc/t5>` (from Google AI) released with the paper `Exploring the Limits of Transfer Learning with a
+58. :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.
-57. :doc:`TAPAS <model_doc/tapas>` (from Google AI) released with the paper `TAPAS: Weakly Supervised Table Parsing via
+59. :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.
-58. :doc:`Transformer-XL <model_doc/transformerxl>` (from Google/CMU) released with the paper `Transformer-XL:
+60. :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.
-59. :doc:`Vision Transformer (ViT) <model_doc/vit>` (from Google AI) released with the paper `An Image is Worth 16x16
+61. :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.
-60. :doc:`VisualBERT <model_doc/visual_bert>` (from UCLA NLP) released with the paper `VisualBERT: A Simple and
+62. :doc:`VisualBERT <model_doc/visual_bert>` (from UCLA NLP) released with the paper `VisualBERT: A Simple and
     Performant Baseline for Vision and Language <https://arxiv.org/pdf/1908.03557>`__ by Liunian Harold Li, Mark
     Yatskar, Da Yin, Cho-Jui Hsieh, Kai-Wei Chang.
-61. :doc:`Wav2Vec2 <model_doc/wav2vec2>` (from Facebook AI) released with the paper `wav2vec 2.0: A Framework for
+63. :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.
-62. :doc:`XLM <model_doc/xlm>` (from Facebook) released together with the paper `Cross-lingual Language Model
+64. :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.
-63. :doc:`XLM-ProphetNet <model_doc/xlmprophetnet>` (from Microsoft Research) released with the paper `ProphetNet:
+65. :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.
-64. :doc:`XLM-RoBERTa <model_doc/xlmroberta>` (from Facebook AI), released together with the paper `Unsupervised
+66. :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.
-65. :doc:`XLNet <model_doc/xlnet>` (from Google/CMU) released with the paper `​XLNet: Generalized Autoregressive
+67. :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.
-66. :doc:`XLSR-Wav2Vec2 <model_doc/xlsr_wav2vec2>` (from Facebook AI) released with the paper `Unsupervised
+68. :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.
 
@@ -372,6 +379,8 @@ Flax), PyTorch, and/or TensorFlow.
 +-----------------------------+----------------+----------------+-----------------+--------------------+--------------+
 |          LayoutLM           |       ✅       |       ✅       |       ✅        |         ✅         |      ❌      |
 +-----------------------------+----------------+----------------+-----------------+--------------------+--------------+
+|         LayoutLMv2          |       ✅       |       ✅       |       ✅        |         ❌         |      ❌      |
++-----------------------------+----------------+----------------+-----------------+--------------------+--------------+
 |             LED             |       ✅       |       ✅       |       ✅        |         ✅         |      ❌      |
 +-----------------------------+----------------+----------------+-----------------+--------------------+--------------+
 |         Longformer          |       ✅       |       ✅       |       ✅        |         ✅         |      ❌      |
@@ -550,6 +559,8 @@ Flax), PyTorch, and/or TensorFlow.
     model_doc/herbert
     model_doc/ibert
     model_doc/layoutlm
+    model_doc/layoutlmv2
+    model_doc/layoutxlm
     model_doc/led
     model_doc/longformer
     model_doc/luke