ciu (version 0.8)

ciu.explain.long.data.frame: Produce CIU explanations as "long" data.frame

Description

This function takes a CIU object and calculates CIU values for the whole data set given as parameter to ciu.new or given as the data parameter.

Usage

ciu.explain.long.data.frame(CIU, data = NULL, out.ind = 1, neutral.CU = 0.5)

Value

data.frame of class "ciu.result.long.data.frame".

Arguments

CIU

CIU object, as created by ciu.new.

data

Data to use as data.frame (default: NULL). If NULL, then use the data of the CIU object, if that exists. This data.frame must contain only feature (input) values, not output value(s).

out.ind

Index of output to explain. Default: 1.

neutral.CU

Neutral CU value(s). Default is 0.5.

Examples

if (FALSE) {
# Boston data set with GBM model.
library(MASS)
library(caret)
kfoldcv <- trainControl(method="cv", number=10)
gbm <- caret::train(medv ~ ., Boston, method="gbm", trControl=kfoldcv)
ciu <- ciu.new(gbm, medv~., Boston)
df <- ciu.explain.long.data.frame(ciu)
head(df)
# Only get results for a part of the data set.
ciu.explain.long.data.frame(ciu, data=subset(Boston[1:10,], select=-medv))
}