CodeTrans model for git commit message generation

Pretrained model on git commit using the t5 small model architecture. It was first released in this repository. This model is trained on tokenized git commit: it works best with tokenized git commit.

Model description

This CodeTrans model is based on the t5-small model. It has its own SentencePiece vocabulary model. It used single-task training on Git Commit Message Generation dataset.

Intended uses & limitations

The model could be used to generate the git commit message for the git commit changes or be fine-tuned on other relevant tasks. It can be used on unparsed and untokenized commit changes. However, if the change is tokenized, the performance should be better.

How to use

Here is how to use this model to generate git commit message using Transformers SummarizationPipeline:

from transformers import AutoTokenizer, AutoModelWithLMHead, SummarizationPipeline

pipeline = SummarizationPipeline(
    tokenizer=AutoTokenizer.from_pretrained("SEBIS/code_trans_t5_small_commit_generation", skip_special_tokens=True),

tokenized_code = "new file mode 100644 index 000000000 . . 892fda21b Binary files / dev / null and b / src / plugins / gateway / lib / joscar . jar differ"

Run this example in colab notebook.

Training data

The supervised training tasks datasets can be downloaded on Link

Evaluation results

For the git commit message generation task, different models achieves the following results on different programming languages (in BLEU score):

Test results :

Language / Model Java
CodeTrans-ST-Small 39.61
CodeTrans-ST-Base 38.67
CodeTrans-TF-Small 44.22
CodeTrans-TF-Base 44.17
CodeTrans-TF-Large 44.41
CodeTrans-MT-Small 36.17
CodeTrans-MT-Base 39.25
CodeTrans-MT-Large 41.18
CodeTrans-MT-TF-Small 43.96
CodeTrans-MT-TF-Base 44.19
CodeTrans-MT-TF-Large 44.34
State of the art 32.81

Created by Ahmed Elnaggar | LinkedIn and Wei Ding | LinkedIn

Downloads last month
Hosted inference API