CFNOW

CF - Nearest Optimal Wololo

The simplest way to generate counterfactuals for any tabular dataset and model.

This package finds an optimal point (closer to the input dataset point), which the classification is different from the original classification (i.e. "flips" the classification of the original input by minimally changin it).

Minimal example:

import cfnow

# Generating a sample model

# Selecting a random point

# Here we can see the original class

# Then, we use CFNOW to generate the minimum modification to change the classification

Improving your results

The minimal example above considers all features as numerical continuous, however, some datasets can have categorical (binary or one-hot encoded) features. CFNOW can handle these data types in a simple way as demonstrated below:

I have binary categorical features!

import cfnow

# Generating a sample model

I have one-hot encoded features!

import cfnow

# Generating a sample model

I have one-hot and binary categorical features!

import cfnow

# Generating a sample model