Predict aspects

Functions for local model explanations. Also known as aspect importance.

aspect_importance() lime() predict_aspects()

Calculates importance of variable groups (called aspects) for a selected observation

aspect_importance_single()

Aspects importance for single aspects

plot(<aspect_importance>)

Function for plotting aspect_importance results

print(<aspect_importance>)

Function for printing aspect_importance results

get_sample()

Function for getting binary matrix

Triplot

Plots for hierarchical feature grouping.

calculate_triplot() print(<triplot>) model_triplot() predict_triplot()

Calculate triplot that sums up automatic aspect/feature importance grouping

plot(<triplot>)

Plots triplot

hierarchical_importance() plot(<hierarchical_importance>)

Calculates importance of hierarchically grouped aspects

Variables grouping

Function for variables grouping.

cluster_variables()

Creates a cluster tree from numeric features

plot(<cluster_variables>)

Plots tree with correlation values

list_variables()

Cuts tree at custom height and returns a list

group_variables()

Helper function that combines clustering variables and creating aspect list