graphe_3Sets {DEMOVA} | R Documentation |
Calulate the predicted values for the external validation set and trace the graph experimental values vs predicted values for training, test and external validation sets.
graphe_3Sets(fit, mydata, mynewdata, mynewdata2, n)
fit |
Multi linear regression between property and selected descriptors (lm object) |
mydata |
Dataframe containing names and values of response and descriptors |
mynewdata |
Dataframe containing property and selected descriptors values for the test set |
mynewdata2 |
Dataframe containing property and selected descriptors values for the external validation set |
n |
Numbers of selected descriptors of the regression (determined using select_MLR) |
Rext , Rext2 |
return a list containing the value of the determination coefficient of the test set and of the external validation set |
Graphe_3sets.tiff |
Image representing experimental values vs predicted values for the all three sets |
# This function have to be run last!
## "Test_set.csv" should be with the following form
## ID property SelectedDesc1 SelectedDesc2 ...
# new_nom<-'Test_set.csv'
# newdata<-read.csv(new_nom,header=TRUE , sep=" ")
# mynewdata=newdata[,2:dim[2]]
## "External_set.csv" should be with the following form
## ID property SelectedDesc1 SelectedDesc2 ...
# new_nom2<-'External_set.csv'
# newdata2<-read.csv(new_nom2,header=TRUE , sep=" ")
# mynewdata2=newdata2[,2:dim[2]]
#graphe_3Sets(fit,mynewdata,mynewdata2,dim(MLR)[2])