r-caret

问题 I'm using R programming. I divided the data as train & test for predicting accuracy. This is my code: library("tree") credit<-read.csv("C:/Users/Administrator/Desktop/german_credit (2).csv") library("caret") set.seed(1000) intrain<-createDataPartition(y=credit$Creditability,p=0.7,list=FALSE) train<-credit[intrain, ] test<-credit[-intrain, ] treemod<-tree(Creditability~. , data=train) plot(treemod) text(treemod) cv.trees<-cv.tree(treemod,FUN=prune.tree) plot(cv.trees) prune.trees<-prune.tree

问题 Is there a way to get caret to use RMSE with a binary classification problem? If you try to use metric = "RMSE" with a classification problem you will receive the message: Error in train.default(x, y, weights = w, ...) : Metric RMSE not applicable for classification models Which makes sense. But is there a way to define a custom metric? For example, if your outcome is 0 or 1 , you can define the error as outcome - p where p is the probability predicted by the model. EDIT ====================

问题 Here is my code: library(MASS) library(caret) df <- Boston set.seed(3721) cv.10.folds <- createFolds(df$medv, k = 10) lasso_grid <- expand.grid(fraction=c(1,0.1,0.01,0.001)) lasso <- train(medv ~ ., data = df, preProcess = c("center", "scale"), method ='lasso', tuneGrid = lasso_grid, trControl= trainControl(method = "cv", number = 10, index = cv.10.folds)) lasso Unlike linear model, I cannot find the coefficients of Lasso regression model from summary(lasso). How should I do that? Or maybe I

问题 I am working with cross-validation data (10-fold repeated 5 times) from a SVM-RFE model generated with the caret package. I know that caret package works with pROC package when computing metrics but I need to use ROCR package in order to obtain the average ROC. However, I noticed that the average AUC values were not the same when using each package, so I am not sure if I should use both packages indistinctively. The code I used to prove that is: predictions_NG3<-list() labels_NG3<-list()

问题 I am using createFolds() in R (version: 3.3.0) to create train/test partitions. To make results reproducible, I used set.seed() with a seed value of 10. As expected, the results (generated folds) were reproducible. But once I loaded caret package just after setting the seed. And then used the createFolds function, I found that the created folds were different (although still reproducible). Specifically, the created folds differ in the following two cases: Case 1: library(caret) set.seed(10)

问题 I used the caret package to train a random forest, including repeated cross-validation. I’d like to know whether the OOB, as in the original RF by Breiman, is used or whether this is replaced by the cross-validation. If it is replaced, do I have the same advantages as described in Breiman 2001, like increased accuracy by reducing the correlation between input data? As OOB is drawn with replacement and CV is drawn without replacement, are both procedures comparable? What is the OOB estimate of

问题 I used the caret package to train a random forest, including repeated cross-validation. I’d like to know whether the OOB, as in the original RF by Breiman, is used or whether this is replaced by the cross-validation. If it is replaced, do I have the same advantages as described in Breiman 2001, like increased accuracy by reducing the correlation between input data? As OOB is drawn with replacement and CV is drawn without replacement, are both procedures comparable? What is the OOB estimate of

问题 I'm running a straightforward linear regression model fit on the following dataframe: > str(model_data_rev) 'data.frame': 128857 obs. of 12 variables: $ ENTRY_4 : num 186 218 208 235 256 447 471 191 207 250 ... $ ENTRY_8 : num 724 769 791 777 707 237 236 726 773 773 ... $ ENTRY_12: num 2853 2989 3174 3027 3028 ... $ ENTRY_16: num 2858 3028 3075 2992 3419 ... $ ENTRY_20: num 7260 7188 7587 7560 7165 ... $ EXIT_4 : num 70 82 105 114 118 204 202 99 73 95 ... $ EXIT_8 : num 1501 1631 1594 1576

问题 Is there anyway we could print iteration when using the caret::train function in parallel? I know there is option called verbose but it seems it doesn't print anything if I use multicore. 回答1: I've found a solution. All we need is to register cores via makeCluster function. library(doSNOW) cl <- makeCluster(30, outfile="") registerDoSNOW(cl) In this way, the log will be printed in console. I've tested on regular R/ Rstudio/ Rserver in mac/window/ubuntu (even AWS) For example, iris <- iris[1

问题 I am facing the following errors while trying to install the caret package in R. g++: error: /tmp/Rtmp2Tos7n/R.INSTALL2e6e30153a74/nloptr/nlopt-2.4.2/lib/libnlopt_cxx.a: No such file or directory make: *** [nloptr.so] Error 1 ERROR: compilation failed for package ‘nloptr’ * removing ‘/rmt/csfiles/pgrads/mava290/R/x86_64-suse-linux-gnu-library/3.1/nloptr’ Warning in install.packages : installation of package ‘nloptr’ had non-zero exit status ERROR: dependency ‘nloptr’ is not available for