neurobase 0.1.0

NeuroBase: a small educational machine learning library with classic models and metrics.

NeuroBase

NeuroBase is a lightweight educational machine-learning library written in pure Python and NumPy. It provides clear implementations of classic algorithms plus a small metrics and optimizer module.

Features

• SimpleLinearRegression, MultipleLinearRegression
• LogisticRegression
• RidgeRegression, LassoRegression
• DecisionTreeRegressorCustom
• KNNClassifier
• KMeansNew
• Gradient boosting (experimental)
• Metrics: MSE, MAE, RMSE, R², adjusted R²
• Optimizers: batch and stochastic gradient descent
• Fully tested with pytest

Installation (local dev)

git clone https://github.com/your-username/neurobase.git
cd neurobase

py -m venv .venv
.venv\Scripts\activate

pip install --upgrade pip
pip install -e .           # editable install of neurobase
pip install pytest