PPtreeregViz (version 2.0.5)

subpick: projection pursuit submodular pick algorithm PP SP-LIME

Description

Pick several data containing various information for each final node for PPTreereg submodular Pick (SP-LIME) was developed (Ribeiro et al., 2016) to selects representative data with important information to determine the reliability of model based on the LIME algorithm. In order to extract data for each final node in the PPTreereg model, PP SP-LIME was proposed based on SP-LIME.

Usage

subpick(data_long, final.leaf, obsnum = 5)

Value

Observation names and their original values as data

Arguments

data_long

ppshapr_prep class object.

final.leaf

location of final leaf

obsnum

The number of budgets (instance to be selected). Default value is 1.

References

Ribeiro, Marco Tulio, Sameer Singh, and Carlos Guestrin. "" Why should i trust you?" Explaining the predictions of any classifier." Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. 2016. tools:::Rd_expr_doi("10.1145/2939672.2939778") https://github.com/marcotcr/lime/blob/master/lime/submodular_pick.py

Examples

data("dataXY")
Model <- PPTreereg(Y~., data = dataXY, DEPTH = 2)
shap_long=ppshapr_prep(Model,final.rule =3,method="simple")
subpick(shap_long,final.leaf = 1, obsnum = 5)