decisive-tree 0.1.1

A collection of very decisive decision trees

Decisive Tree

A from-scratch Python implementation of the ID3, C4.5, and CART decision tree algorithms, created for educational purposes.

This project provides a clear and understandable look into the core mechanics of these foundational machine learning models, from the splitting criteria to the recursive tree-building process. It also includes a suite of from-scratch utilities for data preprocessing and model evaluation.

Key Features

• ID3 Algorithm: Implements the classic algorithm using Information Gain, perfect for datasets with nominal features.
• C4.5 Algorithm: An extension of ID3 using Gain Ratio to handle both nominal (multi-way split) and continuous features (binary split).
• CART Algorithm: Implements Classification and Regression Trees using Gini Impurity (for classification) and Variance Reduction (for regression) with strictly binary splits for all feature types.
• Utilities: Includes custom, understandable implementations of:
  • Data splitting and preprocessing (split_data, one_hot_encode, ordinal_encode).
  • Model evaluation (get_confusion_matrix, calculate_metrics, plot_confusion_matrix).
• Tree Visualization: All models include a .plot() method for a simple console-based visualization of the resulting tree structure.

---

Installation

You can install decisive-tree directly from PyPI:

pip install decisive-tree