DunnBC22/codet5-small-Generate_Docstrings_for_Python-Condensed

1---2license: apache-2.03---4 5# codet5-small-Generate_Docstrings_for_Python-Condensed6 7This model is a fine-tuned version of [Salesforce/codet5-small](https://huggingface.co/Salesforce/codet5-small) on the None dataset.8It achieves the following results on the evaluation set:9- Loss: 2.144410- Rouge1: 0.382811- Rouge2: 0.221412- Rougel: 0.358313- Rougelsum: 0.366114- Gen Len: 12.665615 16## Model description17 18This model is trained to predict the docstring (the output) for a function (the input).19 20For more information on how it was created, check out the following link: https://github.com/DunnBC22/NLP_Projects/blob/main/Generate%20Docstrings/Smol%20Dataset/Code_T5_Project-Small%20Checkpoint.ipynb21 22For this model, I trimmed some of the longer samples to quicken the pace of training on consumer hardware.23 24## Intended uses & limitations25 26This model is intended to demonstrate my ability to solve a complex problem using technology.27 28## Training and evaluation data29 30Dataset Source: calum/the-stack-smol-python-docstrings (from HuggingFace Datasets; https://huggingface.co/datasets/calum/the-stack-smol-python-docstrings)31 32## Training procedure33 34### Training hyperparameters35 36The following hyperparameters were used during training:37- learning_rate: 2e-0538- train_batch_size: 1639- eval_batch_size: 1640- seed: 4241- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-0842- lr_scheduler_type: linear43- num_epochs: 444 45### Training results46 47| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |48|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:|49| 2.9064        | 1.0   | 965  | 2.3096          | 0.3695 | 0.2098 | 0.3464 | 0.3529    | 11.7285 |50| 2.4836        | 2.0   | 1930 | 2.2051          | 0.38   | 0.2176 | 0.3554 | 0.3635    | 12.9401 |51| 2.3669        | 3.0   | 2895 | 2.1548          | 0.3842 | 0.2219 | 0.3595 | 0.3674    | 13.0029 |52| 2.3254        | 4.0   | 3860 | 2.1444          | 0.3828 | 0.2214 | 0.3583 | 0.3661    | 12.6656 |53 54 55### Framework versions56 57- Transformers 4.26.158- Pytorch 1.12.159- Datasets 2.9.060- Tokenizers 0.12.161 62 63## License Notice64This model is a fine-tuned derivative of a pretrained model.65Users must comply with the original model license.66 67 68## Dataset Notice69This model was fine-tuned on third-party datasets which may have separate licenses or usage restrictions.