- Implemented decision trees from scratch using numpy and pandas
- Two types of decision trees implemented :-
- Vanilla - Simple division of nodes based on Information Gain
- Prune - Nodes are pruned to prevent overfitting based on validation set
- python3 ID3.py train.csv test.csv prune
<train-set-percent><validation-set-percent> - python3 ID3.py train.csv test.csv vanilla
<train-set-percent> - python3 plots_prune.py
- python3 plots_without_prune.py