evaluating_model: Evaluating QSAR regression model

In angy89/hyQSAR: Hybrid QSAR model

Description

This function compute the internal and external evaluation metrics for a QSAR model

Usage

evaluating_model(bm, X_train, X_test, y_train, y_test, md_list, gene_list,
  inner.train.prop, nPermValm = 25, nPermScr = 25)

Arguments

bm	is a model obtained from the RCVLassoPar function
X_train	is the matrix of the training dataset with samples on the rows and features on the column. The X_train matrix has to be already transformed.
X_test	is the matrix of the test dataset with samples on the rows and features on the column. The X_train matrix has to be already transformed.
y_train	is the numeric vector of response variable for training samples
y_test	is the numeric vector of response variable for test samples
inner.train.prop	is the percentage of samples from the train test to be used as training set in the random-split method. Default value is 0.9
nPermValm	is the number of random split iteration to compute the validation metrics
nPermScr	is the number of random split iteration to compute the y-scrambling test
p_train	is the percentage of samples to be used in the training set

Value

a list with the following components:

Metrics	a dataframe with the internal and external metrics computed for the identified models
WP	a list with percentage of applicability domain of training and test plot and an object to be plot with the williams plot function

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