This repo contains examples/code for the article on KDNuggets. To view post, click here
We will use the same red wine quality data to compare SHAP and InterpretML. The target value of this dataset is the quality rating from low to high (0–10). The input variables are the content of each wine sample including fixed acidity, volatile acidity, citric acid, residual sugar, chlorides, free sulfur dioxide, total sulfur dioxide, density, pH, sulphates and alcohol. There are 1,599 wine samples.
Data Source: Kaggle
NOTE: If you are unable to view graphs, please paste the notebook link here: https://nbviewer.jupyter.org/