pip install datsets transformers tokenizers
and install pytorch here
pip install pandas tqdm github
- Preprocessing
data/preprocessing.ipynb
- Training and masking testing
train_issue_bert.ipynb
| id | name | language | total | issues | forked |
|---|---|---|---|---|---|
| 1334 | rails | Ruby | 37188 | 12833 | 24355 |
| 1018 | node | NaN | 11155 | 7211 | 3944 |
| 4095 | elasticsearch | Java | 10157 | 6587 | 3570 |
| 340 | netty | Java | 9313 | 4647 | 4666 |
| 6815 | vagrant | Ruby | 9018 | 6618 | 2400 |
It was used to filter prject names
| body | id | title |
|---|---|---|
| Bumps [rfc3986](https://github.com/python-hype... | 596220186 | Bump rfc3986 from 1.3.2 to 1.4.0 |
| Bumps [boto3](https://github.com/boto/boto3... | 596136922 | Bump boto3 from 1.12.36 to 1.12.38 |
| Bumps [botocore](https://github.com/boto/botoc... | 596133846 | Bump botocore from 1.15.36 to 1.15.38 |
| Translations update from [Weblate](https://hos... | 596128315 | Translations update from Weblate |
| this will help to ensure that new projects are... | 596064726 | add rel="nofollow" to trending/latest on index... |
We extracted this data from our issue database with the name we chose from the project data.