SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Add m2m100 (#10236)

Browse Source 
* m2m_100

* no layernorm_embedding

* sinusoidal positional embeddings

* update pos embeddings

* add default config values

* tokenizer

* add conversion script

* fix config

* fix pos embed

* remove _float_tensor

* update tokenizer

* update lang codes

* handle lang codes

* fix pos embeds

* fix spm key

* put embedding weights on device

* remove qa and seq classification heads

* fix convert script

* lang codes pn one line

* fix embeds

* fix tokenizer

* fix tokenizer

* add fast tokenizer

* style

* M2M100MT => M2M100

* fix copyright, style

* tokenizer converter

* vocab file

* remove fast tokenizer

* fix embeds

* fix tokenizer

* fix tests

* add tokenizer tests

* add integration test

* quality

* fix model name

* fix test

* doc

* doc

* fix doc

* add copied from statements

* fix tokenizer tests

* apply review suggestions

* fix urls

* fix shift_tokens_right

* apply review suggestions

* fix

* fix doc

* add lang code to id

* remove unused function

* update checkpoint names

* fix copy

* fix tokenizer

* fix checkpoint names

* fix merge issue

* style
This commit is contained in:
Suraj Patil
2021-03-06 22:14:16 +05:30
committed by GitHub
parent fd01104435
commit f6e74a63ca
17 changed files with 2699 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files

43
docs/source/index.rst
View File

@@ -161,57 +161,61 @@ and conversion utilities for the following models:
26. :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.
27. :doc:`MarianMT <model_doc/marian>` Machine translation models trained using `OPUS <http://opus.nlpl.eu/>`__ data by
27. :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.
28. :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.
28. :doc:`MBart <model_doc/mbart>` (from Facebook) released with the paper `Multilingual Denoising Pre-training for
29. :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.
29. :doc:`MBart-50 <model_doc/mbart>` (from Facebook) released with the paper `Multilingual Translation with Extensible
30. :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.
30. :doc:`MPNet <model_doc/mpnet>` (from Microsoft Research) released with the paper `MPNet: Masked and Permuted
31. :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.
31. :doc:`MT5 <model_doc/mt5>` (from Google AI) released with the paper `mT5: A massively multilingual pre-trained
32. :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.
32. :doc:`Pegasus <model_doc/pegasus>` (from Google) released with the paper `PEGASUS: Pre-training with Extracted
33. :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.
33. :doc:`ProphetNet <model_doc/prophetnet>` (from Microsoft Research) released with the paper `ProphetNet: Predicting
34. :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.
34. :doc:`Reformer <model_doc/reformer>` (from Google Research) released with the paper `Reformer: The Efficient
35. :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.
35. :doc:`RoBERTa <model_doc/roberta>` (from Facebook), released together with the paper a `Robustly Optimized BERT
36. :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.
36. :doc:`SqueezeBert <model_doc/squeezebert>` released with the paper `SqueezeBERT: What can computer vision teach NLP
37. :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.
37. :doc:`T5 <model_doc/t5>` (from Google AI) released with the paper `Exploring the Limits of Transfer Learning with a
38. :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.
38. :doc:`TAPAS <model_doc/tapas>` (from Google AI) released with the paper `TAPAS: Weakly Supervised Table Parsing via
39. :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.
39. :doc:`Transformer-XL <model_doc/transformerxl>` (from Google/CMU) released with the paper `Transformer-XL:
40. :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.
40. :doc:`Wav2Vec2 <model_doc/wav2vec2>` (from Facebook AI) released with the paper `wav2vec 2.0: A Framework for
41. :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.
41. :doc:`XLM <model_doc/xlm>` (from Facebook) released together with the paper `Cross-lingual Language Model
42. :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.
42. :doc:`XLM-ProphetNet <model_doc/xlmprophetnet>` (from Microsoft Research) released with the paper `ProphetNet:
43. :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.
43. :doc:`XLM-RoBERTa <model_doc/xlmroberta>` (from Facebook AI), released together with the paper `Unsupervised
44. :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.
44. :doc:`XLNet <model_doc/xlnet>` (from Google/CMU) released with the paper `​XLNet: Generalized Autoregressive
45. :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.
@@ -276,6 +280,8 @@ TensorFlow and/or Flax.
+-----------------------------+----------------+----------------+-----------------+--------------------+--------------+
| Longformer | ✅ | ✅ | ✅ | ✅ | ❌ |
+-----------------------------+----------------+----------------+-----------------+--------------------+--------------+
| M2M100 | ✅ | ❌ | ✅ | ❌ | ❌ |
+-----------------------------+----------------+----------------+-----------------+--------------------+--------------+
| MPNet | ✅ | ✅ | ✅ | ✅ | ❌ |
+-----------------------------+----------------+----------------+-----------------+--------------------+--------------+
| Marian | ✅ | ❌ | ✅ | ✅ | ❌ |
@@ -416,6 +422,7 @@ TensorFlow and/or Flax.
model_doc/longformer
model_doc/lxmert
model_doc/marian
model_doc/m2m_100
model_doc/mbart
model_doc/mobilebert
model_doc/mpnet

125
docs/source/model_doc/m2m_100.rst Normal file
View File

@@ -0,0 +1,125 @@
..
Copyright 2020 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.
M2M100
-----------------------------------------------------------------------------------------------------------------------
Overview
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The M2M100 model was proposed in `Beyond English-Centric Multilingual Machine Translation
<https://arxiv.org/abs/2010.11125>`__ by Angela Fan, Shruti Bhosale, Holger Schwenk, Zhiyi Ma, Ahmed El-Kishky,
Siddharth Goyal, Mandeep Baines, Onur Celebi, Guillaume Wenzek, Vishrav Chaudhary, Naman Goyal, Tom Birch, Vitaliy
Liptchinsky, Sergey Edunov, Edouard Grave, Michael Auli, Armand Joulin.
The abstract from the paper is the following:
*Existing work in translation demonstrated the potential of massively multilingual machine translation by training a
single model able to translate between any pair of languages. However, much of this work is English-Centric by training
only on data which was translated from or to English. While this is supported by large sources of training data, it
does not reflect translation needs worldwide. In this work, we create a true Many-to-Many multilingual translation
model that can translate directly between any pair of 100 languages. We build and open source a training dataset that
covers thousands of language directions with supervised data, created through large-scale mining. Then, we explore how
to effectively increase model capacity through a combination of dense scaling and language-specific sparse parameters
to create high quality models. Our focus on non-English-Centric models brings gains of more than 10 BLEU when directly
translating between non-English directions while performing competitively to the best single systems of WMT. We
open-source our scripts so that others may reproduce the data, evaluation, and final M2M-100 model.*
Training and Generation
_______________________________________________________________________________________________________________________
M2M100 is a multilingual encoder-decoder (seq-to-seq) model primarily intended for translation tasks. As the model is
multilingual it expects the sequences in a certain format: A special language id token is used as prefix in both the
source and target text. The source text format is :obj:`[lang_code] X [eos]`, where :obj:`lang_code` is source language
id for source text and target language id for target text, with :obj:`X` being the source or target text.
- Supervised Training
.. code-block::
from transformers import M2M100Config, M2M100ForConditionalGeneration, M2M100Tokenizer
model = M2M100ForConditionalGeneration.from_pretrained('facebook/m2m100_418M')
tokenizer = M2M100Tokenizer.from_pretrained('facebook/m2m100_418M', src_lang="en", tgt_lang="fr")
src_text = "Life is like a box of chocolates."
tgt_lang = "La vie est comme une boîte de chocolat."
model_inputs = tokenizer(src_text, return_tensors="pt")
with tokenizer.as_target_tokenizer():
labels = tokenizer(tgt_text, return_tensors="pt").input_ids
loss = model(**model_inputs, labels=labels) # forward pass
- Generation
M2M100 uses the :obj:`eos_token_id` as the :obj:`decoder_start_token_id` for generation with the target language id
being forced as the first generated token. To force the target language id as the first generated token, pass the
`forced_bos_token_id` parameter to the `generate` method. The following example shows how to translate between
Hindi to French and Chinese to English using the `facebook/m2m100_418M` checkpoint.
.. code-block::
>>> from transformers import M2M100ForConditionalGeneration, M2M100Tokenizer
>>> hi_text = "जीवन एक चॉकलेट बॉक्स की तरह है।"
>>> chinese_text = "生活就像一盒巧克力。"
>>> model = M2M100ForConditionalGeneration.from_pretrained("facebook/m2m100_418M")
>>> tokenizer = M2M100Tokenizer.from_pretrained("facebook/m2m100_418M")
>>> # translate Hindi to French
>>> tokenizer.src_lang = "hi"
>>> encoded_hi = tokenizer(hi_text, return_tensors="pt")
>>> generated_tokens = model.generate(**encoded_hi, forced_bos_token_id=tokenizer.get_lang_id("fr"))
>>> tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
"La vie est comme une boîte de chocolat."
>>> # translate Chinese to English
>>> tokenizer.src_lang = "ar_AR"
>>> encoded_zh = tokenizer(chinese_text, return_tensors="pt")
>>> generated_tokens = model.generate(**encoded_zh, forced_bos_token_id=tokenizer.get_lang_id("en"))
>>> tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
"Life is like a box of chocolate."
M2M100Config
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autoclass:: transformers.M2M100Config
:members:
M2M100Tokenizer
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autoclass:: transformers.M2M100Tokenizer
:members: build_inputs_with_special_tokens, get_special_tokens_mask,
create_token_type_ids_from_sequences, save_vocabulary
M2M100Model
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autoclass:: transformers.M2M100Model
:members: forward
M2M100ForConditionalGeneration
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autoclass:: transformers.M2M100ForConditionalGeneration
:members: forward

6
docs/source/pretrained_models.rst
View File

@@ -365,6 +365,12 @@ For the full list, refer to `https://huggingface.co/models <https://huggingface.
| | ``reformer-crime-and-punishment`` | | 6-layer, 256-hidden, 2-heads, 3M parameters |
| | | | Trained on English text: Crime and Punishment novel by Fyodor Dostoyevsky. |
+--------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
| M2M100 | ``facebook/m2m100_418M`` | | 24-layer, 1024-hidden, 16-heads, 418M parameters |
| | | | multilingual machine translation model for 100 languages |
| +------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
| | ``facebook/m2m100_1.2B`` | | 48-layer, 1024-hidden, 16-heads, 1.2B parameters |
| | | | multilingual machine translation model for 100 languages |
+--------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+
| MarianMT | ``Helsinki-NLP/opus-mt-{src}-{tgt}`` | | 12-layer, 512-hidden, 8-heads, ~74M parameter Machine translation models. Parameter counts vary depending on vocab size. |
| | | | (see `model list <https://huggingface.co/Helsinki-NLP>`_) |
+--------------------+------------------------------------------------------------+---------------------------------------------------------------------------------------------------------------------------------------+