williams_plot_linear: A function to create the williams plot and compute the...

Description Usage Arguments Value

View source: R/williams_plot.R

Description

This function create the williams plot for a specific model and compute the percentage of training and test point falling in the applicability domain

Usage

1
williams_plot_linear(X_train, X_test, y_train, y_test, model)

Arguments

X_train

is the training dataset matrix with samples on rows and features on columns.

X_test

is the test dataset matrix with samples on rows and features on columns.

y_train

is the vector of the training response.

y_test

is the vector of the test response.

model

is a fitted linear model for which we want to evaluate the applicability domain

beta

is a vector of the relevant beta values for the regression model

Value

ADVal is a vector with the percentage of training and test sample falling in the applicability domain

DTP is a list containing the data to plot the williams plot. In particular it contain the leverages (lev) the residual (res), the colors of training/test points (col) and the h* value (h_star)


angy89/hyQSAR documentation built on Sept. 24, 2019, 7:31 a.m.