v0.1.0
[0.1.0] - Initial Release
The first official release of PyLWL includes the following features:
馃搧 Project Structure
- Added essential project files:
CODE_OF_CONDUCT.md,MANIFEST.in,LICENSE,CITATION.cff, andrequirements.txt - Added structured layout for examples, tests, and documentation site
馃 Core Modules
- Implemented shared utility modules:
verifier: for validating inputs and parametersscaler: for feature normalizationscorer: for evaluation metricsdata_preparer: for dataset scaling and splittingkernel: for defining and managing kernel functions
馃П Base Framework
- Introduced
BaseModelclass inbase_model.pyfor consistent model design and shared logic
馃攳 Core Models
- Implemented classic locally weighted models in
classic_lw.py:LwClassifier: Locally Weighted ClassifierLwRegressor: Locally Weighted Regressor
- Implemented gradient descent-enhanced variants in
gd_lw.py:GdLwClassifier: Gradient-Descent Locally Weighted ClassifierGdLwRegressor: Gradient-Descent Locally Weighted Regressor
馃殌 Tooling & Deployment
- Added GitHub Actions workflow for PyPI publishing
- Added working examples, test cases, and documentation starter site