@@ -29,9 +29,9 @@ Here is how to use these techniques in our scripts:
* **Gradient Accumulation**\ : Gradient accumulation can be used by supplying a integer greater than 1 to the ``--gradient_accumulation_steps`` argument. The batch at each step will be divided by this integer and gradient will be accumulated over ``gradient_accumulation_steps`` steps.
* **Multi-GPU**\ : Multi-GPU is automatically activated when several GPUs are detected and the batches are splitted over the GPUs.
* **Distributed training**\ : Distributed training can be activated by supplying an integer greater or equal to 0 to the ``--local_rank`` argument (see below).
* **16-bits training**\ : 16-bits training, also called mixed-precision training, can reduce the memory requirement of your model on the GPU by using half-precision training, basically allowing to double the batch size. If you have a recent GPU (starting from NVIDIA Volta architecture) you should see no decrease in speed. A good introduction to Mixed precision training can be found `here <https://devblogs.nvidia.com/mixed-precision-training-deep-neural-networks/>`_ and a full documentation is `here <https://docs.nvidia.com/deeplearning/sdk/mixed-precision-training/index.html>`_. In our scripts, this option can be activated by setting the ``--fp16`` flag and you can play with loss scaling using the ``--loss_scale`` flag (see the previously linked documentation for details on loss scaling). The loss scale can be zero in which case the scale is dynamically adjusted or a positive power of two in which case the scaling is static.
* **16-bits training**\ : 16-bits training, also called mixed-precision training, can reduce the memory requirement of your model on the GPU by using half-precision training, basically allowing to double the batch size. If you have a recent GPU (starting from NVIDIA Volta architecture) you should see no decrease in speed. A good introduction to Mixed precision training can be found `here <https://devblogs.nvidia.com/mixed-precision-training-deep-neural-networks/>`__ and a full documentation is `here <https://docs.nvidia.com/deeplearning/sdk/mixed-precision-training/index.html>`__. In our scripts, this option can be activated by setting the ``--fp16`` flag and you can play with loss scaling using the ``--loss_scale`` flag (see the previously linked documentation for details on loss scaling). The loss scale can be zero in which case the scale is dynamically adjusted or a positive power of two in which case the scaling is static.
To use 16-bits training and distributed training, you need to install NVIDIA's apex extension `as detailed here <https://github.com/nvidia/apex>`_. You will find more information regarding the internals of ``apex`` and how to use ``apex`` in `the doc and the associated repository <https://github.com/nvidia/apex>`_. The results of the tests performed on pytorch-BERT by the NVIDIA team (and my trials at reproducing them) can be consulted in `the relevant PR of the present repository <https://github.com/huggingface/pytorch-pretrained-BERT/pull/116>`_.
To use 16-bits training and distributed training, you need to install NVIDIA's apex extension `as detailed here <https://github.com/nvidia/apex>`__. You will find more information regarding the internals of ``apex`` and how to use ``apex`` in `the doc and the associated repository <https://github.com/nvidia/apex>`_. The results of the tests performed on pytorch-BERT by the NVIDIA team (and my trials at reproducing them) can be consulted in `the relevant PR of the present repository <https://github.com/huggingface/pytorch-pretrained-BERT/pull/116>`_.
Note: To use *Distributed Training*\ , you will need to run one training script on each of your machines. This can be done for example by running the following command on each server (see `the above mentioned blog post <(https://medium.com/huggingface/training-larger-batches-practical-tips-on-1-gpu-multi-gpu-distributed-setups-ec88c3e51255>`_\ ) for more details):
...
...
@@ -153,10 +153,10 @@ and unpack it to some directory ``$GLUE_DIR``.
--num_train_epochs 3.0 \
--output_dir /tmp/mrpc_output/
Our test ran on a few seeds with `the original implementation hyper-parameters <https://github.com/google-research/bert#sentence-and-sentence-pair-classification-tasks>`_ gave evaluation results between 84% and 88%.
Our test ran on a few seeds with `the original implementation hyper-parameters <https://github.com/google-research/bert#sentence-and-sentence-pair-classification-tasks>`__ gave evaluation results between 84% and 88%.
**Fast run with apex and 16 bit precision: fine-tuning on MRPC in 27 seconds!**
First install apex as indicated `here <https://github.com/NVIDIA/apex>`_.
First install apex as indicated `here <https://github.com/NVIDIA/apex>`__.
Then run
.. code-block:: shell
...
...
@@ -520,7 +520,7 @@ and unpack it to some directory ``$GLUE_DIR``.
--num_train_epochs 3.0 \
--output_dir /tmp/mrpc_output/
Our test ran on a few seeds with `the original implementation hyper-parameters <https://github.com/zihangdai/xlnet#1-sts-b-sentence-pair-relevance-regression-with-gpus>`_ gave evaluation results between 84% and 88%.
Our test ran on a few seeds with `the original implementation hyper-parameters <https://github.com/zihangdai/xlnet#1-sts-b-sentence-pair-relevance-regression-with-gpus>`__ gave evaluation results between 84% and 88%.
**Distributed training**
Here is an example using distributed training on 8 V100 GPUs to reach XXXX:
@@ -52,13 +52,13 @@ Here are some information on these models:
This PyTorch implementation of BERT is provided with `Google's pre-trained models <https://github.com/google-research/bert>`_\ , examples, notebooks and a command-line interface to load any pre-trained TensorFlow checkpoint for BERT is also provided.
**OpenAI GPT** was released together 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.
This PyTorch implementation of OpenAI GPT is an adaptation of the `PyTorch implementation by HuggingFace <https://github.com/huggingface/pytorch-openai-transformer-lm>`_ and is provided with `OpenAI's pre-trained model <https://github.com/openai/finetune-transformer-lm>`_ and a command-line interface that was used to convert the pre-trained NumPy checkpoint in PyTorch.
This PyTorch implementation of OpenAI GPT is an adaptation of the `PyTorch implementation by HuggingFace <https://github.com/huggingface/pytorch-openai-transformer-lm>`_ and is provided with `OpenAI's pre-trained model <https://github.com/openai/finetune-transformer-lm>`__ and a command-line interface that was used to convert the pre-trained NumPy checkpoint in PyTorch.
**Google/CMU's Transformer-XL** was released together with the paper `Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context <http://arxiv.org/abs/1901.02860>`_ by Zihang Dai\ *, Zhilin Yang*\ , Yiming Yang, Jaime Carbonell, Quoc V. Le, Ruslan Salakhutdinov.
This PyTorch implementation of Transformer-XL is an adaptation of the original `PyTorch implementation <https://github.com/kimiyoung/transformer-xl>`_ which has been slightly modified to match the performances of the TensorFlow implementation and allow to re-use the pretrained weights. A command-line interface is provided to convert TensorFlow checkpoints in PyTorch models.
**OpenAI GPT-2** was released together 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**.
This PyTorch implementation of OpenAI GPT-2 is an adaptation of the `OpenAI's implementation <https://github.com/openai/gpt-2>`_ and is provided with `OpenAI's pre-trained model <https://github.com/openai/gpt-2>`_ and a command-line interface that was used to convert the TensorFlow checkpoint in PyTorch.
This PyTorch implementation of OpenAI GPT-2 is an adaptation of the `OpenAI's implementation <https://github.com/openai/gpt-2>`_ and is provided with `OpenAI's pre-trained model <https://github.com/openai/gpt-2>`__ and a command-line interface that was used to convert the TensorFlow checkpoint in PyTorch.
* ``bert-base-chinese``: Chinese Simplified and Traditional, 12-layer, 768-hidden, 12-heads, 110M parameters
* ``bert-base-german-cased``: Trained on German data only, 12-layer, 768-hidden, 12-heads, 110M parameters `Performance Evaluation <https://deepset.ai/german-bert>`_
* ``bert-base-german-cased``: Trained on German data only, 12-layer, 768-hidden, 12-heads, 110M parameters `Performance Evaluation <https://deepset.ai/german-bert>`__
* ``bert-large-uncased-whole-word-masking``: 24-layer, 1024-hidden, 16-heads, 340M parameters - Trained with Whole Word Masking (mask all of the the tokens corresponding to a word at once)
* ``bert-large-cased-whole-word-masking``: 24-layer, 1024-hidden, 16-heads, 340M parameters - Trained with Whole Word Masking (mask all of the the tokens corresponding to a word at once)
* ``bert-large-uncased-whole-word-masking-finetuned-squad``: The ``bert-large-uncased-whole-word-masking`` model finetuned on SQuAD (using the ``run_bert_squad.py`` examples). Results: *exact_match: 86.91579943235573, f1: 93.1532499015869*
...
...
@@ -93,7 +93,7 @@ where
* ``bert_config.json`` or ``openai_gpt_config.json`` a configuration file for the model, and
* ``pytorch_model.bin`` a PyTorch dump of a pre-trained instance of ``BertForPreTraining``\ , ``OpenAIGPTModel``\ , ``TransfoXLModel``\ , ``GPT2LMHeadModel`` (saved with the usual ``torch.save()``\ )
If ``PRE_TRAINED_MODEL_NAME_OR_PATH`` is a shortcut name, the pre-trained weights will be downloaded from AWS S3 (see the links `here <https://github.com/huggingface/pytorch-pretrained-BERT/tree/master/pytorch_pretrained_bert/modeling.py>`_\ ) and stored in a cache folder to avoid future download (the cache folder can be found at ``~/.pytorch_pretrained_bert/``\ ).
If ``PRE_TRAINED_MODEL_NAME_OR_PATH`` is a shortcut name, the pre-trained weights will be downloaded from AWS S3 (see the links `here <https://github.com/huggingface/pytorch-pretrained-BERT/tree/master/pytorch_pretrained_bert/modeling.py>`__\ ) and stored in a cache folder to avoid future download (the cache folder can be found at ``~/.pytorch_pretrained_bert/``\ ).
*
``cache_dir`` can be an optional path to a specific directory to download and cache the pre-trained model weights. This option is useful in particular when you are using distributed training: to avoid concurrent access to the same weights you can set for example ``cache_dir='./pretrained_model_{}'.format(args.local_rank)`` (see the section on distributed training for more information).
...
...
@@ -102,7 +102,7 @@ where
* ``state_dict``\ : an optional state dictionnary (collections.OrderedDict object) to use instead of Google pre-trained models
* ``*inputs``\ , `**kwargs`: additional input for the specific Bert class (ex: num_labels for BertForSequenceClassification)
``Uncased`` means that the text has been lowercased before WordPiece tokenization, e.g., ``John Smith`` becomes ``john smith``. The Uncased model also strips out any accent markers. ``Cased`` means that the true case and accent markers are preserved. Typically, the Uncased model is better unless you know that case information is important for your task (e.g., Named Entity Recognition or Part-of-Speech tagging). For information about the Multilingual and Chinese model, see the `Multilingual README <https://github.com/google-research/bert/blob/master/multilingual.md>`_ or the original TensorFlow repository.
``Uncased`` means that the text has been lowercased before WordPiece tokenization, e.g., ``John Smith`` becomes ``john smith``. The Uncased model also strips out any accent markers. ``Cased`` means that the true case and accent markers are preserved. Typically, the Uncased model is better unless you know that case information is important for your task (e.g., Named Entity Recognition or Part-of-Speech tagging). For information about the Multilingual and Chinese model, see the `Multilingual README <https://github.com/google-research/bert/blob/master/multilingual.md>`__ or the original TensorFlow repository.
When using an ``uncased model``\ , make sure to pass ``--do_lower_case`` to the example training scripts (or pass ``do_lower_case=True`` to FullTokenizer if you'reusingyourownscriptandloadingthetokenizeryour-self.).
...
...
@@ -152,7 +152,7 @@ This section explain how you can save and re-load a fine-tuned model (BERT, GPT,
There are three types of files you need to save to be able to reload a fine-tuned model:
* the model it-self which should be saved following PyTorch serialization `best practices <https://pytorch.org/docs/stable/notes/serialization.html#best-practices>`_\ ,
* the model it-self which should be saved following PyTorch serialization `best practices <https://pytorch.org/docs/stable/notes/serialization.html#best-practices>`__\ ,
* the configuration file of the model which is saved as a JSON file, and
* the vocabulary (and the merges for the BPE-based models GPT and GPT-2).
