Displaying 5 results from an estimated 5 matches for "rfmodel".
Did you mean:
fmodel
2013 Feb 07
1
Saving model and other objects from caret
Say I train a model in caret, e.g.:
RFmodel <- train(X,Y,method='rf',trControl=myCtrl,tuneLength=1)
How can I save this to disk and load it later in R?
How about an object of the class "resamples"?
resamps <- resamples(
list( RF = RFmodel,
SVM = SVMmodel,
KNN = KNNmodel,...
2017 Oct 22
0
Test set and Train set in Caret package train function
...know how we can get train set and test set for each fold of 5 fold cross validation in Caret package? Imagine if I want to do cross validation by random forest method, I do the following in Caret:
set.seed(12)
train_control <- trainControl(method="cv", number=5,savePredictions = TRUE)
rfmodel <- train(Species~., data=iris, trControl=train_control, method="rf")
first_holdout <- subset(rfmodel$pred, Resample == "Fold1")
str(first_holdout)
'data.frame': 90 obs. of 5 variables:
$ pred : Factor w/ 3 levels "setosa","versicolor",..: 1...
2013 Feb 07
4
Sourcing my file does not print command outputs
I looked at the documentation of source() and summary(), and I could not
find the reason why calling something like:
> summary(resamps)
from the command line, works (it prints the summary)
whereas calling
summary(resampls)
from a file that I source with source("my_file.r") does not print anything.
How can I get summary(resamps) to print when I source a file with this
command?
2013 Feb 07
0
FW: Sourcing my file does not print command outputs
...y
Y <- as.factor(ifelse(sign(Y)>0,'X1','X0'))
#Create bootstrap samples for fitting models
library(caret)
print("Creating bootstrap samples")
tmp <- createResample(Y,times = 25)
myCtrl <- trainControl(method = "boot", index = tmp, timingSamps = 10)
RFmodel <- train(X,Y,method='rf',trControl=myCtrl,tuneLength=1)
NNmodel <- train(X,Y,method='nnet',trControl=myCtrl,tuneLength=3, trace = FALSE)
## GLMnet = GLMmodel,
...
#Assess re-sampled (out of sample) accuracy
print("Assessing re-sampled (OOB) accuracy&quo...
2013 Feb 10
1
Training with very few positives
...e the following setup:
========================================
library(caret)
tmp <- createDataPartition(Y, p = 9/10, times = 3, list = TRUE)
myCtrl <- trainControl(method = "boot", index = tmp, timingSamps = 2,
classProbs = TRUE, summaryFunction = twoClassSummary)
RFmodel <- train(X,Y,method='rf',trControl=myCtrl,tuneLength=1,
metric="ROC")
SVMmodel <- train(X,Y,method='svmRadial',trControl=myCtrl,tuneLength=3,
metric="ROC")
KNNmodel <- train(X,Y,method='knn',trControl=myCtrl,tuneLength=10,
metric="RO...