Nearest Instance Counterfactual explanations
Nearest Instance Counterfactual Explanations (NICE)
NICE is an algorithm to generate Counterfactual Explanations for heterogeneous tabular data. Our approach exploits information from a nearest instance to speed up the search process and guarantee that an explanation will be found.
Install NICE through Pypi
pip install NICEx
pip install git git+https://github.com/ADMantwerp/nice.git
NICE requires acces to the prediction score and trainingdata to generate counterfactual explanations.
from nice import NICE # Initialize NICE by specifing the optimization strategy NICE_explainer = NICE(optimization='sparsity') # Fit our NICE explainer on the training data and classifier NICE_explainer.fit(predict_fn,X_train,cat_feat,num_feat,y_train,optimization='sparsity') # explain an instance NICE_explainer.explain(x)
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.