Local Interpretable (Model-Agnostic) Visual Explanations
Add black box predictions to generated dataset
LIME kernel equal to the inverse of euclidean distance.
Fit white box model to the simulated data.
LIME kernel from the original article with sigma = 1.
LIME kernel that treats all observations as equally similar to observa...
live: visualizing interpretable models to explain black box models.
Function that starts a Shiny app which helps use LIVE.
Fit local model around the observation: shortcut for DALEX explainer o...
Local permutation variable importance
Plotting white box models.
Plot local permutation importance
Generic print function for live explainer
Generic print function for class live_explorer
Print method for local_permutation_importance class
Generate dataset for local exploration.
Interpretability of complex machine learning models is a growing concern. This package helps to understand key factors that drive the decision made by complicated predictive model (so called black box model). This is achieved through local approximations that are either based on additive regression like model or CART like model that allows for higher interactions. The methodology is based on Tulio Ribeiro, Singh, Guestrin (2016) <doi:10.1145/2939672.2939778>. More details can be found in Staniak, Biecek (2018) <doi:10.32614/RJ-2018-072>.