srirams04 / decision-tree-from-scratch Goto Github PK

The decision tree implementation does not use existing machine learning libraries like scikit-learn. It is implemented from scratch using Python (NumPy and Pandas) and works for four cases:

• Discrete features and discrete output
• Discrete features and real output
• Real features and discrete output
• Real features and real output

The decision tree can switch between GiniIndex or Information Gain (Entropy) as the criteria for splitting for discrete output, and uses Information Gain (MSE) as the criteria for splitting for real output.

• metrics.py: Contains the performance metrics functions.

• usage.py: This file can be run to check the implementation.

• tree (Directory): Module for decision tree.

  • base.py : Contains Decision Tree Class.
  • utils.py: Contains all utility functions.
  • __init__.py: Do not edit this

This task was completed as part of the Machine Learning course at IIT Gandhinagar, under Prof. Nipun Batra.