dfs99 / ID3_DecisionTree

The repository itself contains an implementation for the well-know ID3 algorithm. The implementation carried out uses a LIFO to apply backtracking instead of recursive calls.

ID3 Decision Tree Algorithm

The repository itself contains an implementation for the well-known ID3 algorithm. The implementation carried out uses a LIFO to apply backtracking instead of recursive calls.

In order to use the implementation you must consider the following advices:

1-.) The ID3 implementation takes as parameter a pd.Dataframe that only works with discrete attributes.

2-.) The class to be classified will be always the pd.Dataframe's last column.

3-.) The method 'generate_tree()' has verbose=True as default. It will print out traces of the tree.

An example is given below:

import pandas as pd
from ID3DecisionTree import ID3DecisionTree

if __name__ == '__main__':
    tree = ID3DecisionTree(pd.read_csv("dataset/any_csv.csv"))
    tree.generate_tree()