SEBIS/code_trans_t5_large_commit_generation_multitask_finetune

1---2tags:3- summarization4widget:5- text: "new file mode 100644 index 000000000 . . 892fda21b Binary files / dev / null and b / src / plugins / gateway / lib / joscar . jar differ"6 7---8 9 10# CodeTrans model for git commit message generation11Pretrained model on git commit using the t5 large model architecture. It was first released in12[this repository](https://github.com/agemagician/CodeTrans). This model is trained on tokenized git commit: it works best with tokenized git commit.13 14 15## Model description16 17This CodeTrans model is based on the `t5-large` model. It has its own SentencePiece vocabulary model.  It used multi-task training on 13 supervised tasks in the software development domain and 7 unsupervised datasets. It is then fine-tuned on the git commit message generation task for the java commit changes.18 19## Intended uses & limitations20 21The 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.22 23### How to use24 25Here is how to use this model to generate git commit message using Transformers SummarizationPipeline:26 27```python28from transformers import AutoTokenizer, AutoModelWithLMHead, SummarizationPipeline29 30pipeline = SummarizationPipeline(31    model=AutoModelWithLMHead.from_pretrained("SEBIS/code_trans_t5_large_commit_generation_multitask_finetune"),32    tokenizer=AutoTokenizer.from_pretrained("SEBIS/code_trans_t5_large_commit_generation_multitask_finetune", skip_special_tokens=True),33    device=034)35 36tokenized_code = "new file mode 100644 index 000000000 . . 892fda21b Binary files / dev / null and b / src / plugins / gateway / lib / joscar . jar differ"37pipeline([tokenized_code])38```39Run this example in [colab notebook](https://github.com/agemagician/CodeTrans/blob/main/prediction/multitask/fine-tuning/commit%20generation/large_model.ipynb).40## Training data41 42The supervised training tasks datasets can be downloaded on [Link](https://www.dropbox.com/sh/488bq2of10r4wvw/AACs5CGIQuwtsD7j_Ls_JAORa/finetuning_dataset?dl=0&subfolder_nav_tracking=1)43 44 45## Training procedure46 47### Multi-task Pretraining48 49The model was trained on a single TPU Pod V3-8 for 500,000 steps in total, using sequence length 512 (batch size 4096).50It has a total of approximately 220M parameters and was trained using the encoder-decoder architecture.51The optimizer used is AdaFactor with inverse square root learning rate schedule for pre-training.52 53### Fine-tuning54 55This model was then fine-tuned on a single TPU Pod V2-8 for 3,000 steps in total, using sequence length 512 (batch size 256), using only the dataset only containing commit changes.56 57 58## Evaluation results59 60For the git commit message generation task, different models achieves the following results on different programming languages (in BLEU score):61 62Test results :63 64|   Language / Model   |      Java      |65| -------------------- | :------------: |66|   CodeTrans-ST-Small    |     39.61      |67|   CodeTrans-ST-Base     |     38.67      |68|   CodeTrans-TF-Small    |     44.22      |69|   CodeTrans-TF-Base     |     44.17      |70|   CodeTrans-TF-Large    |   **44.41**    |71|   CodeTrans-MT-Small    |     36.17      |72|   CodeTrans-MT-Base     |     39.25      |73|   CodeTrans-MT-Large    |     41.18      |74|   CodeTrans-MT-TF-Small |     43.96      |75|   CodeTrans-MT-TF-Base  |     44.19      |76|   CodeTrans-MT-TF-Large |     44.34      |77|   State of the art   |     32.81      |78 79 80 81> Created by [Ahmed Elnaggar](https://twitter.com/Elnaggar_AI) | [LinkedIn](https://www.linkedin.com/in/prof-ahmed-elnaggar/) and Wei Ding | [LinkedIn](https://www.linkedin.com/in/wei-ding-92561270/)82 83