bert-score 0.3.3

PyTorch implementation of BERT score

BERTScore

Automatic Evaluation Metric described in the paper BERTScore: Evaluating Text Generation with BERT (ICLR 2020).

News:

• Updated to version 0.3.3
  • Fixing the bug with empty strings issue #47.
  • Supporting 6 ELECTRA models and 24 smaller BERT models.
  • A new Google sheet for keeping the performance (i.e., pearson correlation with human judgment) of different models on WMT16 to-English.
  • Including the script for tuning the best number of layers of an English pre-trained model on WMT16 to-English data (See the details).
• Updated to version 0.3.2
  • Bug fixed: fixing the bug in v0.3.1 when having multiple reference sentences.
  • Supporting multiple reference sentences with our command line tool.
• Updated to version 0.3.1
  • A new BERTScorer object that caches the model to avoid re-loading it multiple times. Please see our jupyter notebook example for the usage.
  • Supporting multiple reference sentences for each example. The score function now can take a list of lists of strings as the references and return the score between the candidate sentence and its closest reference sentence.
• Updated to version 0.3.0
  • Supporting Baseline Rescaling: we apply a simple linear transformation to enhance the readability of BERTscore using pre-computed "baselines". It has been pointed out (e.g. by #20, #23) that the numerical range of BERTScore is exceedingly small when computed with RoBERTa models. In other words, although BERTScore correctly distinguishes examples through ranking, the numerical scores of good and bad examples are very similar. We detail our approach in a separate post.

Please see release logs for older updates.

Authors:

• Tianyi Zhang*
• Varsha Kishore*
• Felix Wu*
• Kilian Q. Weinberger
• Yoav Artzi

*: Equal Contribution

Overview

BERTScore leverages the pre-trained contextual embeddings from BERT and matches words in candidate and reference sentences by cosine similarity. It has been shown to correlate with human judgment on sentence-level and system-level evaluation. Moreover, BERTScore computes precision, recall, and F1 measure, which can be useful for evaluating different language generation tasks.

For an illustration, BERTScore precision can be computed as

If you find this repo useful, please cite:

@inproceedings{bert-score,
  title={BERTScore: Evaluating Text Generation with BERT},
  author={Tianyi Zhang* and Varsha Kishore* and Felix Wu* and Kilian Q. Weinberger and Yoav Artzi},
  booktitle={International Conference on Learning Representations},
  year={2020},
  url={https://openreview.net/forum?id=SkeHuCVFDr}
}

Installation

• Python version >= 3.6
• PyTorch version >= 1.0.0

Install from pypi with pip by

pip install bert-score

Install latest unstable version from the master branch on Github by:

pip install git+https://github.com/Tiiiger/bert_score

Install it from the source by:

git clone https://github.com/Tiiiger/bert_score
cd bert_score
pip install .

and you may test your installation by:

python -m unittest discover

Usage

Command Line Interface (CLI)

We provide a command line interface (CLI) of BERTScore as well as a python module. For the CLI, you can use it as follows:

1. To evaluate English text files:

We provide example inputs under ./example.

bert-score -r example/refs.txt -c example/hyps.txt --lang en

You will get the following output at the end:

roberta-large_L17_no-idf_version=0.3.0(hug_trans=2.3.0) P: 0.957378 R: 0.961325 F1: 0.959333

where "roberta-large_L17_no-idf_version=0.3.0(hug_trans=2.3.0)" is the hash code.

Starting from version 0.3.0, we support rescaling the scores with baseline scores

bert-score -r example/refs.txt -c example/hyps.txt --lang en --rescale-with-baseline

You will get:

roberta-large_L17_no-idf_version=0.3.0(hug_trans=2.3.0)-rescaled P: 0.747044 R: 0.770484 F1: 0.759045

This makes the range of the scores larger and more human-readable. Please see this post for details.

When having multiple reference sentences, please use

bert-score -r example/refs.txt example/refs2.txt -c example/hyps.txt --lang en

where the -r argument supports an arbitrary number of reference files. Each reference file should have the same number of lines as your candidate/hypothesis file. The i-th line in each reference file corresponds to the i-th line in the candidate file.

2. To evaluate text files in other languages:

We currently support the 104 languages in multilingual BERT (full list).

Please specify the two-letter abbreviation of the language. For instance, using --lang zh for Chinese text.

See more options by bert-score -h.

3. To load your own custom model: Please specify the path to the model and the number of layers to use by --model and --num_layers.

bert-score -r example/refs.txt -c example/hyps.txt --model path_to_my_bert --num_layers 9

4. To visualize matching scores:

bert-score-show --lang en -r "There are two bananas on the table." -c "On the table are two apples." -f out.png

The figure will be saved to out.png.

Python Function

For the python module, we provide a demo. Please refer to bert_score/score.py for more details.

Running BERTScore can be computationally intensive (because it uses BERT :p). Therefore, a GPU is usually necessary. If you don't have access to a GPU, you can try our demo on Google Colab

Practical Tips

• Report the hash code (e.g., roberta-large_L17_no-idf_version=0.3.0(hug_trans=2.3.0)-rescaled) in your paper so that people know what setting you use. This is inspired by sacreBLEU. Changes in huggingface's transformers version may also affect the score (See issue #46).
• Unlike BERT, RoBERTa uses GPT2-style tokenizer which creates addition " " tokens when there are multiple spaces appearing together. It is recommended to remove addition spaces by sent = re.sub(r' +', ' ', sent) or sent = re.sub(r'\s+', ' ', sent).
• Using inverse document frequency (idf) on the reference sentences to weigh word importance may correlate better with human judgment. However, when the set of reference sentences become too small, the idf score would become inaccurate/invalid. We now make it optional. To use idf, please set --idf when using the CLI tool or idf=True when calling bert_score.score function.
• When you are low on GPU memory, consider setting batch_size when calling bert_score.score function.
• To use a particular model please set -m MODEL_TYPE when using the CLI tool or model_type=MODEL_TYPE when calling bert_score.score function.
• We tune layer to use based on WMT16 metric evaluation dataset. You may use a different layer by setting -l LAYER or num_layers=LAYER. To tune the best layer for your custom model, please follow the instructions in tune_layers folder.
• Limitation: Because BERT, RoBERTa, and XLM with learned positional embeddings are pre-trained on sentences with max length 512, BERTScore is undefined between sentences longer than 510 (512 after adding [CLS] and [SEP] tokens). The sentences longer than this will be truncated. Please consider using XLNet which can support much longer inputs.

Default Behavior

Default Model

Language	Model
en	roberta-large
en-sci	scibert-scivocab-uncased
zh	bert-base-chinese
others	bert-base-multilingual-cased

Default Layers

Please see this Google sheet for the supported models and their performance.

Acknowledgement

This repo wouldn't be possible without the awesome bert, fairseq, and transformers.