Davy592's picture
First commit
79c84e2
raw
history blame contribute delete
623 Bytes
"""Nygaard Code Comment Classification package.
This package provides utilities to download, clean, featurize, train, and
evaluate models for code comment classification across multiple languages
(`java`, `python`, and `pharo`).
Available submodules
--------------------
- `config`: Project paths, constants, and logging setup.
- `dataset`: Dataset download and cleaning routines.
- `features`: Feature extraction (embeddings and labels) utilities.
- `plots`: Basic plotting CLI scaffold.
- `modeling.train`: Training entry points (CatBoost and SetFit).
- `modeling.predict`: Evaluation and MLflow logging utilities.
"""