CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generation

This is the official PyTorch implementation for the following paper at EMNLP 2021 from Salesforce Research:
Title: CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generation [blog]
Authors: Yue Wang, Weishi Wang, Shafiq Joty, Steven C.H. Hoi
Abstract: Pre-trained models for Natural Languages (NL) like BERT and GPT have been recently shown to transfer well to Programming Languages (PL) and largely benefit a broad set of code-related tasks. Despite their success, most current methods either rely on an encoder-only (or decoder-only) pre-training that is suboptimal for generation (resp. understanding) tasks or process the code snippet in the same way as NL, neglecting the special characteristics of PL such as token types. We present CodeT5, a unified pre-trained encoder-decoder Transformer model that better leverages the code semantics conveyed from the developer-assigned identifiers. Our model employs a unified framework to seamlessly support both code understanding and generation tasks and allows for multi-task learning. Besides, we propose a novel identifier-aware pre-training task that enables the model to distinguish which code tokens are identifiers and to recover them when they are masked. Furthermore, we propose to exploit the user-written code comments with a bimodal dual generation task for better NL-PL alignment. Comprehensive experiments show that CodeT5 significantly outperforms prior methods on understanding tasks such as code defect detection and clone detection, and generation tasks across various directions including PL-NL, NL-PL, and PL-PL. Further analysis reveals that our model can better capture semantic information from code.

Requirements

• Pytorch==1.7.1
• tensorboard==2.4.1
• transformers==4.6.1
• tree_sitter==0.2.2

Download

• Pre-trained checkpoint
• Fine-tuning data

Fine-tuning

Go to sh folder, you can use run_exp.py to run a broad set of experiments by simply passing the model_tag, task, sub_task, and other hyper-parameters. For example, if you want to run CodeT5-base on code summarization task for Ruby, you can type the following command:

python run_exp.py --model_tag codet5_base --task summarize --sub_task ruby

Then it will save the fine-tuning checkpoints into sh/saved_models, the evaluation results to sh/results, the training curves to sh/tensorboard, which can be visualized using tensorboard.

File Structure

├── CODE_OF_CONDUCT.md
├── README.md
├── SECURITY.md
├── CodeT5.png
├── _utils.py
├── configs.py
├── models.py
├── run_clone.py
├── run_gen.py
├── utils.py
├── LICENSE.txt
├── data
│   ├── clone
│   ├── concode
│   ├── defect
│   ├── refine
│   │   ├── medium
│   │   └── small
│   ├── summarize
│   │   ├── go
│   │   ├── java
│   │   ├── javascript
│   │   ├── php
│   │   ├── python
│   │   └── ruby
│   └── translate
├── evaluator
│   ├── CodeBLEU
│   │   ├── keywords
│   │   └── parser
├── pretrained_models
│   └── codet5_base
├── sh
│   ├── exp_with_args.sh
│   ├── run_exp.py
│   ├── results
│   ├── saved_models
│   └── tensorboard
└── tokenizer
    └── salesforce
        ├── codet5-merges.txt
        └── codet5-vocab.json    

Citation

If you find this code to be useful for your research, please consider citing.

@article{CodeT5,
      title={CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generation}, 
      author={Yue Wang, Weishi Wang, Shafiq Joty, Steven C.H. Hoi},
      year={2021},
      journal={arXiv preprint arXiv:2109.00859},
}