I have multiple classification machine learning models with all different accuracy. When I run my xgBOOST (using library(caret)) in the console, I get an accuracy of 0.7586. But when I knit my Rmarkdown, the accuracy of the same model is 0.8621. I have no idea why this is different.
I followed the suggestions of this link, but nothing worked: https://community.rstudio.com/t/console-and-rmd-output-differ-same-program-used-but-the-calculation-gives-a-different-result/67873/3
I also followed the suggestions of problem, but nothing worked: Statistics Result in R Markdown is different from the Knit Output (All Format: Word, HTML, PDF)
At last I tried this, but also nothing worked: sample function gives different result in console and in knitted document when seed is set
Here is my code which I run the same in console and Rmarkdown but with different accuracy:
# Data
data <- data[!is.na(data$var1),]
# Change levels of var1
levels(data$var1)=c("No","Yes")
#Data Preparation and Preprocessing
# Create the training and test datasets
set.seed(100)
# Step 1: Get row numbers for the training data
trainRowNumbers <- createDataPartition(data$var1, p=0.8, list=FALSE)
# Step 2: Create the training dataset
trainset <- data[trainRowNumbers,]
# Step 3: Create the test dataset
testset <- data[-trainRowNumbers,]
# Store Y for later use.
y = trainset$var1
# Create the knn imputation model on the training data
preProcess_missingdata_model <- preProcess(as.data.frame(trainset), method= c("knnImpute"))
preProcess_missingdata_model
# Create the knn imputation model on the testset data
preProcess_missingdata_model_test <- preProcess(as.data.frame(testset), method = c("knnImpute"))
preProcess_missingdata_model_test
# Use the imputation model to predict the values of missing data points
library(RANN) # required for knnInpute
trainset <- predict(preProcess_missingdata_model, newdata = trainset)
anyNA(trainset)
# Use the imputation model to predict the values of missing data points
library(RANN) # required for knnInpute
testset <- predict(preProcess_missingdata_model_test, newdata = testset)
anyNA(testset)
# Append the Y variable
trainset$var1 <- y
# Run algorithms using 5-fold cross validation
control <- trainControl(method="cv",
number=5,
repeats = 5,
savePredictions = "final",
search = "grid",
classProbs = TRUE)
metric <- "Accuracy"
# Make Valid Column Names
colnames(trainset) <- make.names(colnames(trainset))
colnames(testset) <- make.names(colnames(testset))
# xgBOOST
set.seed(7)
fit.xgbDART <- train(var1~., data = trainset, method = "xgbTree", metric = metric, trControl = control, verbose = FALSE, tuneLength = 7, nthread = 1)
# estimate skill of xgBOOST on the testset dataset
predictions <- predict(fit.xgbDART, testset)
cm <- caret::confusionMatrix(predictions, testset$var1, mode='everything')
cm
My RNGKind is:
RNGkind()
[1] "L'Ecuyer-CMRG" "Inversion" "Rejection"
always add the function :
set.seed(544)
This function sets the starting number used to generate a sequence of random numbers – it ensures that you get the same result if you start with that same seed each time you run the same process. For example, if I use the sample() function immediately after setting a seed, I will always get the same sample.
This is my suggestion on where to use set.seed()
# Data
data <- data[!is.na(data$var1),]
# Change levels of var1
levels(data$var1)=c("No","Yes")
#Data Preparation and Preprocessing
# Create the training and test datasets
# Step 1: Get row numbers for the training data
set.seed(100)
trainRowNumbers <- createDataPartition(data$var1, p=0.8, list=FALSE)
# Step 2: Create the training dataset
trainset <- data[trainRowNumbers,]
# Step 3: Create the test dataset
testset <- data[-trainRowNumbers,]
# Store Y for later use.
y = trainset$var1
# Create the knn imputation model on the training data
set.seed(100)
preProcess_missingdata_model <- preProcess(as.data.frame(trainset), method= c("knnImpute"))
preProcess_missingdata_model
# Create the knn imputation model on the testset data
set.seed(100)
preProcess_missingdata_model_test <- preProcess(as.data.frame(testset), method = c("knnImpute"))
preProcess_missingdata_model_test
# Use the imputation model to predict the values of missing data points
library(RANN) # required for knnInpute
trainset <- predict(preProcess_missingdata_model, newdata = trainset)
anyNA(trainset)
# Use the imputation model to predict the values of missing data points
library(RANN) # required for knnInpute
testset <- predict(preProcess_missingdata_model_test, newdata = testset)
anyNA(testset)
# Append the Y variable
trainset$var1 <- y
# Run algorithms using 5-fold cross validation
set.seed(100)
control <- trainControl(method="cv",
number=5,
repeats = 5,
savePredictions = "final",
search = "grid",
classProbs = TRUE)
metric <- "Accuracy"
# Make Valid Column Names
colnames(trainset) <- make.names(colnames(trainset))
colnames(testset) <- make.names(colnames(testset))
# xgBOOST
set.seed(7)
fit.xgbDART <-
train(
var1 ~ .,
data = trainset,
method = "xgbTree",
metric = metric,
trControl = control,
verbose = FALSE,
tuneLength = 7,
nthread = 1
)
# estimate skill of xgBOOST on the testset dataset
predictions <- predict(fit.xgbDART, testset)
cm <- caret::confusionMatrix(predictions, testset$var1, mode='everything')
I keep running into an error while attempting to plot variable importance from ensemble of models.
I have ensemble of models I've fitted and now I am trying to create multiple variable importance plots for each algorithm I've fitted. I am using varImp() function from caret to extract variable importance, then plot() it. To fit ensemble of models, I am using caretEnsemble package.
Thank you for any help, please see the example of code below.
# Caret ensemble is needed to produce list of models
library(caret)
library(caretEnsemble)
# Set algorithms I wish to fit
my_algorithms <- c("glmnet", "svmRadial", "rf", "nnet", "knn", "rpart")
# Define controls
my_controls <- trainControl(
method = "cv",
savePredictions = "final",
number = 3
)
# Run the models all at once with caretEnsemble
my_list_of_models <- caretEnsemble::caretList(Species ~ .,
data = iris,
trControl = my_controls,
methodList = my_algorithms)
# Subset models
list_of_algorithms <- my_list_of_models[my_algorithms]
# Create first for loop to extract variable importance via caret::varImp()
importance <- list()
for (algo in seq_along(list_of_algorithms)) {
importance[[algo]] <- varImp(list_of_algorithms[[algo]])
}
# Create second loop to go over extracted importance and plot it using plot()
importance_plots <- list()
for (imp in seq_along(importance)) {
importance_plots[[imp]] <- plot(importance[[imp]])
}
# Error occurs during the second for loop:
Error in data.frame(values = unlist(unname(x)), ind, stringsAsFactors = FALSE):arguments imply differing number of rows: 16,
I've come up with the solution to the problem above and decided to post it as my own answer. I've written a small function to plot variable importance without relying on caret helper functions to create plots. I used dotplot and levelplot because caret returns data.frame that differs based on provided algorithm. It may not work on different algorithms and models that didn't fit.
# Libraries ---------------------------------------------------------------
library(caret) # To train ML algorithms
library(dplyr) # Required for %>% operators in custom function below
library(caretEnsemble) # To train multiple caret models
library(lattice) # Required for plotting, should be loaded alongside caret
library(gridExtra) # Required for plotting multiple plots
# Custom function ---------------------------------------------------------
# The function requires list of models as input and is used in for loop
plot_importance <- function(importance_list, imp, algo_names) {
importance <- importance_list[[imp]]$importance
model_title <- algo_names[[imp]]
if (ncol(importance) < 2) { # Plot dotplot if dim is ncol < 2
importance %>%
as.matrix() %>%
dotplot(main = model_title)
} else { # Plot heatmap if ncol > 2
importance %>%
as.matrix() %>%
levelplot(xlab = NULL, ylab = NULL, main = model_title, scales = list(x = list(rot = 45)))
}
}
# Tuning parameters -------------------------------------------------------
# Set algorithms I wish to fit
# Rather than using methodList as provided above, I've switched to tuneList because I need to control tuning parameters of random forest algorithm.
my_algorithms <- list(
glmnet = caretModelSpec(method = "glmnet"),
rpart = caretModelSpec(method = "rpart"),
svmRadial = caretModelSpec(method = "svmRadial"),
rf = caretModelSpec(method = "rf", importance = TRUE), # Importance is not computed for "rf" by default
nnet = caretModelSpec(method = "nnet"),
knn = caretModelSpec(method = "knn")
)
# Define controls
my_controls <- trainControl(
method = "cv",
savePredictions = "final",
number = 3
)
# Run the models all at once with caretEnsemble
my_list_of_models <- caretList(Species ~ .,
data = iris,
tuneList = my_algorithms,
trControl = my_controls
)
# Extract variable importance ---------------------------------------------
importance <- lapply(my_list_of_models, varImp)
# Plotting variable immportance -------------------------------------------
# Create second loop to go over extracted importance and plot it using plot()
importance_plots <- list()
for (imp in seq_along(importance)) {
# importance_plots[[imp]] <- plot(importance[[imp]])
importance_plots[[imp]] <- plot_importance(importance_list = importance, imp = imp, algo_names = names(my_list_of_models))
}
# Multiple plots at once
do.call("grid.arrange", c(importance_plots))
This question is a continuation of the same thread here. Below is a minimal working example taken from this book:
Wehrens R. Chemometrics with R multivariate data analysis in the
natural sciences and life sciences. 1st edition. Heidelberg; New York:
Springer. 2011. (page 250).
The example was taken from this book and its package ChemometricsWithR. It highlighted some pitfalls when modeling using cross-validation techniques.
The Aim:
A cross-validated methodology using the same set of repeated CV to perform a known strategy of PLS followed typically by LDA or cousins like logistic regression, SVM, C5.0, CART, with the spirit of caret package. So PLS would be needed every time before calling the waiting classifier in order to classify PLS score space instead of the observations themselves. The nearest approach in the caret package is doing PCA as a pre-processing step before modeling with any classifier. Below is a PLS-LDA procedure with only one cross-validation to test performance of the classifier, there was no 10-fold CV or any repetition. The code below was taken from the mentioned book but with some corrections otherwise throws error:
library(ChemometricsWithR)
data(prostate)
prostate.clmat <- classvec2classmat(prostate.type) # convert Y to a dummy var
odd <- seq(1, length(prostate.type), by = 2) # training
even <- seq(2, length(prostate.type), by = 2) # holdout test
prostate.pls <- plsr(prostate.clmat ~ prostate, ncomp = 16, validation = "CV", subset=odd)
Xtst <- scale(prostate[even,], center = colMeans(prostate[odd,]), scale = apply(prostate[odd,],2,sd))
tst.scores <- Xtst %*% prostate.pls$projection # scores for the waiting trained LDA to test
prostate.ldapls <- lda(scores(prostate.pls)[,1:16],prostate.type[odd]) # LDA for scores
table(predict(prostate.ldapls, new = tst.scores[,1:16])$class, prostate.type[even])
predictionTest <- predict(prostate.ldapls, new = tst.scores[,1:16])$class)
library(caret)
confusionMatrix(data = predictionTest, reference= prostate.type[even]) # from caret
Output:
Confusion Matrix and Statistics
Reference
Prediction bph control pca
bph 4 1 9
control 1 35 7
pca 34 4 68
Overall Statistics
Accuracy : 0.6564
95% CI : (0.5781, 0.7289)
No Information Rate : 0.5153
P-Value [Acc > NIR] : 0.0001874
Kappa : 0.4072
Mcnemar's Test P-Value : 0.0015385
Statistics by Class:
Class: bph Class: control Class: pca
Sensitivity 0.10256 0.8750 0.8095
Specificity 0.91935 0.9350 0.5190
Pos Pred Value 0.28571 0.8140 0.6415
Neg Pred Value 0.76510 0.9583 0.7193
Prevalence 0.23926 0.2454 0.5153
Detection Rate 0.02454 0.2147 0.4172
Detection Prevalence 0.08589 0.2638 0.6503
Balanced Accuracy 0.51096 0.9050 0.6643
However, the confusion matrix didn't match that in the book, anyway the code in the book did break, but this one here worked with me!
Notes:
Although this was only one CV, but the intention is to agree on this methodology first, sd and mean of the train set were applied on the test set, PLUS transformed into PLS scores based a specific number of PC ncomp. I want this to occur every round of the CV in the caret. If the methodology as code is correct here, then it can serve, may be, as a good start for a minimal work example while modifying the code of the caret package.
Side Notes:
It can be very messy with scaling and centering, I think some of the PLS functions in R do scaling internally, with or without centering, I am not sure, so building a custom model in caret should be handled with care to avoid both lack or multiple scalings or centerings (I am on my guards with these things).
Perils of multiple centering/scaling
The code below is just to show how multliple centering/scaling can change the data, only centering is shown here but the same problem with scaling applies too.
set.seed(1)
x <- rnorm(200, 2, 1)
xCentered1 <- scale(x, center=TRUE, scale=FALSE)
xCentered2 <- scale(xCentered1, center=TRUE, scale=FALSE)
xCentered3 <- scale(xCentered2, center=TRUE, scale=FALSE)
sapply (list(xNotCentered= x, xCentered1 = xCentered1, xCentered2 = xCentered2, xCentered3 = xCentered3), mean)
Output:
xNotCentered xCentered1 xCentered2 xCentered3
2.035540e+00 1.897798e-16 -5.603699e-18 -5.332377e-18
Please drop a comment if I am missing something somewhere in this course. Thanks.
If you want to fit these types of models with caret, you would need to use the latest version on CRAN. The last update was created so that people can use non-standard models as they see fit.
My approach below is to jointly fit the PLS and other model (I used random forest in the example below) and tune them at the same time. So for each fold, a 2D grid of ncomp and mtry is used.
The "trick" is to attached the PLS loadings to the random forest object so that they can be used during prediction time. Here is the code that defines the model (classification only):
modelInfo <- list(label = "PLS-RF",
library = c("pls", "randomForest"),
type = "Classification",
parameters = data.frame(parameter = c('ncomp', 'mtry'),
class = c("numeric", 'numeric'),
label = c('#Components',
'#Randomly Selected Predictors')),
grid = function(x, y, len = NULL) {
grid <- expand.grid(ncomp = seq(1, min(ncol(x) - 1, len), by = 1),
mtry = 1:len)
grid <- subset(grid, mtry <= ncomp)
},
loop = NULL,
fit = function(x, y, wts, param, lev, last, classProbs, ...) {
## First fit the pls model, generate the training set scores,
## then attach what is needed to the random forest object to
## be used later
pre <- plsda(x, y, ncomp = param$ncomp)
scores <- pls:::predict.mvr(pre, x, type = "scores")
mod <- randomForest(scores, y, mtry = param$mtry, ...)
mod$projection <- pre$projection
mod
},
predict = function(modelFit, newdata, submodels = NULL) {
scores <- as.matrix(newdata) %*% modelFit$projection
predict(modelFit, scores)
},
prob = NULL,
varImp = NULL,
predictors = function(x, ...) rownames(x$projection),
levels = function(x) x$obsLevels,
sort = function(x) x[order(x[,1]),])
and here is the call to train:
library(ChemometricsWithR)
data(prostate)
set.seed(1)
inTrain <- createDataPartition(prostate.type, p = .90)
trainX <-prostate[inTrain[[1]], ]
trainY <- prostate.type[inTrain[[1]]]
testX <-prostate[-inTrain[[1]], ]
testY <- prostate.type[-inTrain[[1]]]
## These will take a while for these data
set.seed(2)
plsrf <- train(trainX, trainY, method = modelInfo,
preProc = c("center", "scale"),
tuneLength = 10,
trControl = trainControl(method = "repeatedcv",
repeats = 5))
## How does random forest do on its own?
set.seed(2)
rfOnly <- train(trainX, trainY, method = "rf",
tuneLength = 10,
trControl = trainControl(method = "repeatedcv",
repeats = 5))
Just for kicks, I got:
> getTrainPerf(plsrf)
TrainAccuracy TrainKappa method
1 0.7940423 0.65879 custom
> getTrainPerf(rfOnly)
TrainAccuracy TrainKappa method
1 0.7794082 0.6205322 rf
and
> postResample(predict(plsrf, testX), testY)
Accuracy Kappa
0.7741935 0.6226087
> postResample(predict(rfOnly, testX), testY)
Accuracy Kappa
0.9032258 0.8353982
Max
Based on Max's valuable comments, I felt the need to have IRIS referee, which is famous for classification, and more importantly the Species outcome has more than two classes, which would be a good data set to test the PLS-LDA custom model in caret:
data(iris)
names(iris)
head(iris)
dim(iris) # 150x5
set.seed(1)
inTrain <- createDataPartition(y = iris$Species,
## the outcome data are needed
p = .75,
## The percentage of data in the
## training set
list = FALSE)
## The format of the results
## The output is a set of integers for the rows of Iris
## that belong in the training set.
training <- iris[ inTrain,] # 114
testing <- iris[-inTrain,] # 36
ctrl <- trainControl(method = "repeatedcv",
repeats = 5,
classProbs = TRUE)
set.seed(2)
plsFitIris <- train(Species ~ .,
data = training,
method = "pls",
tuneLength = 4,
trControl = ctrl,
preProc = c("center", "scale"))
plsFitIris
plot(plsFitIris)
set.seed(2)
plsldaFitIris <- train(Species ~ .,
data = training,
method = modelInfo,
tuneLength = 4,
trControl = ctrl,
preProc = c("center", "scale"))
plsldaFitIris
plot(plsldaFitIris)
Now comparing the two models:
getTrainPerf(plsFitIris)
TrainAccuracy TrainKappa method
1 0.8574242 0.7852462 pls
getTrainPerf(plsldaFitIris)
TrainAccuracy TrainKappa method
1 0.975303 0.9628179 custom
postResample(predict(plsFitIris, testing), testing$Species)
Accuracy Kappa
0.750 0.625
postResample(predict(plsldaFitIris, testing), testing$Species)
Accuracy Kappa
0.9444444 0.9166667
So, finally there was the EXPECTED difference, and improvement in the metrics. So this would support Max's notion, that two-class problems because of Bayes' probabilistic approach of plsda function both lead to the same results.
You need to wrap the CV around both PLS and LDA.
Yes, both plsr and lda center the data their own way
I had a closer look at caret::preProcess (): as it is defined now, you will not be able to use PLS as preprocessing method because it is supervised but caret::preProcess () uses unsupervised methods only (there is no way to hand over the dependent variable). This would probably make patching rather difficult.
So inside the caret framework, you'll need to go for a custom model.
If the scenario were to custom a model of PLS-LDA type, according to the code kindly provided by Max (maintainer of CARET), something is not corect in this code, but I didn't figure it out, because I used the Sonar data set the same in caret vignette and tried to reproduce the result one time using method="pls" and another time using the below custom model for PLS-LDA, the results were exactly identical even to the last digit, which was nonsensical. For benchmarking, one need a known data set (I think a cross-validated PLS-LDA for iris data set would fit here as it is famous for this type of analysis and there should be somewhere a cross-validated treatment of it), everything should be the same (the set.seed(xxx) and the no of K-CV repitition) except the code in question so as to rightly compare and to judge the code below:
modelInfo <- list(label = "PLS-LDA",
library = c("pls", "MASS"),
type = "Classification",
parameters = data.frame(parameter = c("ncomp"),
class = c("numeric"),
label = c("#Components")),
grid = function(x, y, len = NULL) {
grid <- expand.grid(ncomp = seq(1, min(ncol(x) - 1, len), by = 1))
},
loop = NULL,
fit = function(x, y, wts, param, lev, last, classProbs, ...) {
## First fit the pls model, generate the training set scores,
## then attach what is needed to the lda object to
## be used later
pre <- plsda(x, y, ncomp = param$ncomp)
scores <- pls:::predict.mvr(pre, x, type = "scores")
mod <- lda(scores, y, ...)
mod$projection <- pre$projection
mod
},
predict = function(modelFit, newdata, submodels = NULL) {
scores <- as.matrix(newdata) %*% modelFit$projection
predict(modelFit, scores)$class
},
prob = function(modelFit, newdata, submodels = NULL) {
scores <- as.matrix(newdata) %*% modelFit$projection
predict(modelFit, scores)$posterior
},
varImp = NULL,
predictors = function(x, ...) rownames(x$projection),
levels = function(x) x$obsLevels,
sort = function(x) x[order(x[,1]),])
Based on Zach's request, the code below is for method="pls" in caret, exactly the same concrete example in caret vigenette on CRAN:
library(mlbench) # data set from here
data(Sonar)
dim(Sonar) # 208x60
set.seed(107)
inTrain <- createDataPartition(y = Sonar$Class,
## the outcome data are needed
p = .75,
## The percentage of data in the
## training set
list = FALSE)
## The format of the results
## The output is a set of integers for the rows of Sonar
## that belong in the training set.
training <- Sonar[ inTrain,] #157
testing <- Sonar[-inTrain,] # 51
ctrl <- trainControl(method = "repeatedcv",
repeats = 3,
classProbs = TRUE,
summaryFunction = twoClassSummary)
set.seed(108)
plsFitSon <- train(Class ~ .,
data = training,
method = "pls",
tuneLength = 15,
trControl = ctrl,
metric = "ROC",
preProc = c("center", "scale"))
plsFitSon
plot(plsFitSon) # might be slightly difference than what in the vignette due to radnomness
Now, the code below is a pilot run to classify Sonar data using the custom model PLS-LDA which is under question, it is expected to come up with any numbers apart from identical with those using PLS only:
set.seed(108)
plsldaFitSon <- train(Class ~ .,
data = training,
method = modelInfo,
tuneLength = 15,
trControl = ctrl,
metric = "ROC",
preProc = c("center", "scale"))
Now comparing the results between the two models:
getTrainPerf(plsFitSon)
TrainROC TrainSens TrainSpec method
1 0.8741154 0.7638889 0.8452381 pls
getTrainPerf(plsldaFitSon)
TrainROC TrainSens TrainSpec method
1 0.8741154 0.7638889 0.8452381 custom
postResample(predict(plsFitSon, testing), testing$Class)
Accuracy Kappa
0.745098 0.491954
postResample(predict(plsldaFitSon, testing), testing$Class)
Accuracy Kappa
0.745098 0.491954
So, the results are exactly the same which cannot be. As if the lda model were not added?