bert README

The project's README from github.com/google-research/bert. As one markdown file: README.md.

BERT

***** New March 11th, 2020: Smaller BERT Models *****

This is a release of 24 smaller BERT models (English only, uncased, trained with WordPiece masking) referenced in Well-Read Students Learn Better: On the Importance of Pre-training Compact Models.

We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher.

Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.

You can download all 24 from here, or individually from the table below:

H=128 H=256 H=512 H=768
L=2 2/128 (BERT-Tiny) 2/256 2/512 2/768
L=4 4/128 4/256 (BERT-Mini) 4/512 (BERT-Small) 4/768
L=6 6/128 6/256 6/512 6/768
L=8 8/128 8/256 8/512 (BERT-Medium) 8/768
L=10 10/128 10/256 10/512 10/768
L=12 12/128 12/256 12/512 12/768 (BERT-Base)

Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model.

Here are the corresponding GLUE scores on the test set:

Model Score CoLA SST-2 MRPC STS-B QQP MNLI-m MNLI-mm QNLI(v2) RTE WNLI AX
BERT-Tiny 64.2 0.0 83.2 81.1/71.1 74.3/73.6 62.2/83.4 70.2 70.3 81.5 57.2 62.3 21.0
BERT-Mini 65.8 0.0 85.9 81.1/71.8 75.4/73.3 66.4/86.2 74.8 74.3 84.1 57.9 62.3 26.1
BERT-Small 71.2 27.8 89.7 83.4/76.2 78.8/77.0 68.1/87.0 77.6 77.0 86.4 61.8 62.3 28.6
BERT-Medium 73.5 38.0 89.6 86.6/81.6 80.4/78.4 69.6/87.9 80.0 79.1 87.7 62.2 62.3 30.5

For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs:

If you use these models, please cite the following paper:

@article{turc2019,
  title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models},
  author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina},
  journal={arXiv preprint arXiv:1908.08962v2 },
  year={2019}
}

***** New May 31st, 2019: Whole Word Masking Models *****

This is a release of several new models which were the result of an improvement the pre-processing code.

In the original pre-processing code, we randomly select WordPiece tokens to mask. For example:

Input Text: the man jumped up , put his basket on phil ##am ##mon ' s head Original Masked Input: [MASK] man [MASK] up , put his [MASK] on phil [MASK] ##mon ' s head

The new technique is called Whole Word Masking. In this case, we always mask all of the the tokens corresponding to a word at once. The overall masking rate remains the same.

Whole Word Masked Input: the man [MASK] up , put his basket on [MASK] [MASK] [MASK] ' s head

The training is identical -- we still predict each masked WordPiece token independently. The improvement comes from the fact that the original prediction task was too 'easy' for words that had been split into multiple WordPieces.

This can be enabled during data generation by passing the flag --do_whole_word_mask=True to create_pretraining_data.py.

Pre-trained models with Whole Word Masking are linked below. The data and training were otherwise identical, and the models have identical structure and vocab to the original models. We only include BERT-Large models. When using these models, please make it clear in the paper that you are using the Whole Word Masking variant of BERT-Large.

Model SQUAD 1.1 F1/EM Multi NLI Accuracy
BERT-Large, Uncased (Original) 91.0/84.3 86.05
BERT-Large, Uncased (Whole Word Masking) 92.8/86.7 87.07
BERT-Large, Cased (Original) 91.5/84.8 86.09
BERT-Large, Cased (Whole Word Masking) 92.9/86.7 86.46

***** New February 7th, 2019: TfHub Module *****

BERT has been uploaded to TensorFlow Hub. See run_classifier_with_tfhub.py for an example of how to use the TF Hub module, or run an example in the browser on Colab.

***** New November 23rd, 2018: Un-normalized multilingual model + Thai + Mongolian *****

We uploaded a new multilingual model which does not perform any normalization on the input (no lower casing, accent stripping, or Unicode normalization), and additionally inclues Thai and Mongolian.

It is recommended to use this version for developing multilingual models, especially on languages with non-Latin alphabets.

This does not require any code changes, and can be downloaded here:

***** New November 15th, 2018: SOTA SQuAD 2.0 System *****

We released code changes to reproduce our 83% F1 SQuAD 2.0 system, which is currently 1st place on the leaderboard by 3%. See the SQuAD 2.0 section of the README for details.

***** New November 5th, 2018: Third-party PyTorch and Chainer versions of BERT available *****

NLP researchers from HuggingFace made a PyTorch version of BERT available which is compatible with our pre-trained checkpoints and is able to reproduce our results. Sosuke Kobayashi also made a Chainer version of BERT available (Thanks!) We were not involved in the creation or maintenance of the PyTorch implementation so please direct any questions towards the authors of that repository.

***** New November 3rd, 2018: Multilingual and Chinese models available *****

We have made two new BERT models available:

We use character-based tokenization for Chinese, and WordPiece tokenization for all other languages. Both models should work out-of-the-box without any code changes. We did update the implementation of BasicTokenizer in tokenization.py to support Chinese character tokenization, so please update if you forked it. However, we did not change the tokenization API.

For more, see the Multilingual README.

***** End new information *****

Introduction

BERT, or Bidirectional Encoder Representations from Transformers, is a new method of pre-training language representations which obtains state-of-the-art results on a wide array of Natural Language Processing (NLP) tasks.

Our academic paper which describes BERT in detail and provides full results on a number of tasks can be found here: https://arxiv.org/abs/1810.04805.

To give a few numbers, here are the results on the SQuAD v1.1 question answering task:

SQuAD v1.1 Leaderboard (Oct 8th 2018) Test EM Test F1
1st Place Ensemble - BERT 87.4 93.2
2nd Place Ensemble - nlnet 86.0 91.7
1st Place Single Model - BERT 85.1 91.8
2nd Place Single Model - nlnet 83.5 90.1

And several natural language inference tasks:

System MultiNLI Question NLI SWAG
BERT 86.7 91.1 86.3
OpenAI GPT (Prev. SOTA) 82.2 88.1 75.0

Plus many other tasks.

Moreover, these results were all obtained with almost no task-specific neural network architecture design.

If you already know what BERT is and you just want to get started, you can download the pre-trained models and run a state-of-the-art fine-tuning in only a few minutes.

What is BERT?

BERT is a method of pre-training language representations, meaning that we train a general-purpose "language understanding" model on a large text corpus (like Wikipedia), and then use that model for downstream NLP tasks that we care about (like question answering). BERT outperforms previous methods because it is the first unsupervised, deeply bidirectional system for pre-training NLP.

Unsupervised means that BERT was trained using only a plain text corpus, which is important because an enormous amount of plain text data is publicly available on the web in many languages.

Pre-trained representations can also either be context-free or contextual, and contextual representations can further be unidirectional or bidirectional. Context-free models such as word2vec or GloVe generate a single "word embedding" representation for each word in the vocabulary, so bank would have the same representation in bank deposit and river bank. Contextual models instead generate a representation of each word that is based on the other words in the sentence.

BERT was built upon recent work in pre-training contextual representations — including Semi-supervised Sequence Learning, Generative Pre-Training, ELMo, and ULMFit — but crucially these models are all unidirectional or shallowly bidirectional. This means that each word is only contextualized using the words to its left (or right). For example, in the sentence I made a bank deposit the unidirectional representation of bank is only based on I made a but not deposit. Some previous work does combine the representations from separate left-context and right-context models, but only in a "shallow" manner. BERT represents "bank" using both its left and right context — I made a ... deposit — starting from the very bottom of a deep neural network, so it is deeply bidirectional.

BERT uses a simple approach for this: We mask out 15% of the words in the input, run the entire sequence through a deep bidirectional Transformer encoder, and then predict only the masked words. For example:

Input: the man went to the [MASK1] . he bought a [MASK2] of milk.
Labels: [MASK1] = store; [MASK2] = gallon

In order to learn relationships between sentences, we also train on a simple task which can be generated from any monolingual corpus: Given two sentences A and B, is B the actual next sentence that comes after A, or just a random sentence from the corpus?

Sentence A: the man went to the store .
Sentence B: he bought a gallon of milk .
Label: IsNextSentence

Sentence A: the man went to the store .
Sentence B: penguins are flightless .
Label: NotNextSentence

We then train a large model (12-layer to 24-layer Transformer) on a large corpus (Wikipedia + BookCorpus) for a long time (1M update steps), and that's BERT.

Using BERT has two stages: Pre-training and fine-tuning.

Pre-training is fairly expensive (four days on 4 to 16 Cloud TPUs), but is a one-time procedure for each language (current models are English-only, but multilingual models will be released in the near future). We are releasing a number of pre-trained models from the paper which were pre-trained at Google. Most NLP researchers will never need to pre-train their own model from scratch.

Fine-tuning is inexpensive. All of the results in the paper can be replicated in at most 1 hour on a single Cloud TPU, or a few hours on a GPU, starting from the exact same pre-trained model. SQuAD, for example, can be trained in around 30 minutes on a single Cloud TPU to achieve a Dev F1 score of 91.0%, which is the single system state-of-the-art.

The other important aspect of BERT is that it can be adapted to many types of NLP tasks very easily. In the paper, we demonstrate state-of-the-art results on sentence-level (e.g., SST-2), sentence-pair-level (e.g., MultiNLI), word-level (e.g., NER), and span-level (e.g., SQuAD) tasks with almost no task-specific modifications.

What has been released in this repository?

We are releasing the following:

All of the code in this repository works out-of-the-box with CPU, GPU, and Cloud TPU.

Pre-trained models

We are releasing the BERT-Base and BERT-Large models from the paper. 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).

These models are all released under the same license as the source code (Apache 2.0).

For information about the Multilingual and Chinese model, see the Multilingual README.

When using a cased model, make sure to pass --do_lower=False to the training scripts. (Or pass do_lower_case=False directly to FullTokenizer if you're using your own script.)

The links to the models are here (right-click, 'Save link as...' on the name):

Each .zip file contains three items:

bert (google-research/bert)

Searching Database…

40K ·GitHub
Bert
archived
What is this?Google's official TensorFlow implementation of BERT together with pre-trained checkpoints — a method for pre-training language representations that set state-of-the-art results on a wide range of NLP tasks, from SQuAD question answering to natural language inference.
Is it a library I install into my app, or how does it ship?Neither a pip package nor an app: it's research code — the model architecture plus data-generation and fine-tuning scripts — released alongside downloadable pre-trained checkpoints (cased and uncased) that you fine-tune on your own task. BERT is also published as a TensorFlow Hub module.
How do I get it running?Download one of the pre-trained checkpoints linked in the README and fine-tune it with the included scripts on your task; Google says you can go from download to a state-of-the-art fine-tuning in only a few minutes.
Is this something that runs on my own machine, or does it need Google's cloud?It runs locally: plain Python and TensorFlow code plus checkpoint files on your own hardware, with no accounts or API keys. The cloud side — Cloud TPUs or the Colab example in the browser — is optional, not a requirement.
So what's the catch before I invest time in it?The repository is archived, so it's read-only and no longer gets updates or fixes. The releases are also partial: Whole Word Masking comes only in BERT-Large, and the 24 compact models are English-only and uncased.
Will my own GPU cut it, or do I need a Cloud TPU?Fine-tuning is light: the paper's results replicate in at most 1 hour on a single Cloud TPU or a few hours on an ordinary GPU. Only pre-training from scratch is TPU-scale — four days on 4 to 16 Cloud TPUs — and the released checkpoints make it unnecessary for most people.
Why was BERT such a breakthrough compared to word2vec or GPT?Word2vec-style embeddings give every word one fixed vector, and earlier contextual models like ELMo and OpenAI GPT were unidirectional or only shallowly bidirectional. BERT was the first unsupervised, deeply bidirectional pre-training system and beat the previous state of the art — for example 86.7 vs 82.2 accuracy on MultiNLI against OpenAI GPT.
How was BERT trained in the first place?On Wikipedia plus BookCorpus for 1M update steps, learning two tasks: predict 15% of the words masked out of the input through a deep bidirectional Transformer encoder, and decide whether sentence B genuinely follows sentence A. A later Whole Word Masking variant masks whole words at once and lifts uncased BERT-Large SQuAD 1.1 F1 from 91.0 to 92.8.
Which languages can I use it for besides English?There is a Multilingual Cased model covering 104 languages and a separate Chinese model handling Simplified and Traditional characters with character-based tokenization. Google recommends the Multilingual Cased checkpoint, especially for languages with non-Latin alphabets.
How much accuracy do the smaller models give up?On the GLUE test set they score 64.2 (BERT-Tiny), 65.8 (BERT-Mini), 71.2 (BERT-Small) and 73.5 (BERT-Medium) — well below full-size BERT. They target restricted-compute settings and work best in knowledge distillation, fine-tuned with labels produced by a larger, more accurate teacher model.
What are similar projects?Chinese-BERT-wwm, spaCy, HanLP, MSA, funNLP, gpt-2, vision_transformer, awesome-nlp
Which topics is it in?named-entity-recognition, question-answering, language-model, neural-networks, pretrained-models, tensorflow, paper, natural-language-processing, embeddings, google, fine-tuning, academic-research
Creator
google-research
Size
0.3 MB
Stars
40K
Updated
CommitMar 11
Age
7y
Commits · 90 Days