Live Arena

Functions used to create and run live version of Arena

create_arena()

Creates arena object

push_model()

Adds model to arena

push_observations()

Adds new observations to arena

print(<arena_live>)

Prints live arena summary

run_server()

Run server providing data for live Arena

Static Arena

Functions used to create and upload JSON file for Arena

create_arena()

Creates arena object

push_model()

Adds model to arena

push_observations()

Adds new observations to arena

print(<arena_static>)

Prints static arena summary

upload_arena()

Upload generated json file from static arena

Plots

Internal functions for generating plot’s data.

get_global_plots()

Internal function for calculating global plots

get_local_plots()

Internal function for calculating local plots for all observations

get_feature_importance()

Internal function for calculating feature importance

get_accumulated_dependence()

Internal function for calculating Accumulated Dependence

get_break_down()

Internal function for calculating Break Down

get_shap_values()

Internal function for calculating Shapley Values

get_ceteris_paribus()

Internal function for calculating Ceteris Paribus

get_funnel_measure()

Internal function for calculating funnel measure

get_metrics()

Internal function for calculating model performance metrics

get_partial_dependence()

Internal function for calculating Partial Dependence

get_rec()

Internal function for calculating regression error characteristic

get_roc()

Internal function for calculating receiver operating curve

Misc

Supplementary functions

get_json_structure()

Prepare object ready to change into json

get_observations_list()

Generates list of rownames of each observation from each batch

get_variables_list()

Generates list of unique variables(without target) from each explainer and dataset

validate_new_model()

Checks if it is safe do add a new model to the arena object

validate_new_observations()

Checks if it is safe do add new observations to the arena object

split_multiclass_explainer()

Splits multiclass explainer into multiple classification explainers