timingSamps error using formula interface

[LluisRamon]

Hi Max,

I think I have found a bug. I followed your instructions:

• Start a new R session
• Install the latest version of caret: update.packages(oldPkgs="caret", ask=FALSE)
• Write a minimal reproducible example
• run sessionInfo()

Minimal dataset:

library("caret")

data(kyphosis, package = "rpart")

fitControl <- trainControl(method = "cv",
                           timingSamps = 5)

Minimal, runnable code:

# Using formula interface doesn't work
train(Kyphosis ~ ., data = kyphosis, 
      method = "rpart", 
      trControl = fitControl)
# Error in eval(expr, envir, enclos) : object 'Age' not found 

# Using non-formula interface works
train(x = kyphosis[, -1], y = kyphosis$Kyphosis,
      method = "rpart",
      trControl = fitControl)

Session Info:

R version 3.2.3 (2015-12-10)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 14.04.3 LTS

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               LC_TIME=en_US.UTF-8       
 [4] LC_COLLATE=en_US.UTF-8     LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                  LC_ADDRESS=C              
[10] LC_TELEPHONE=C             LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] rpart_4.1-10    caret_6.0-68    ggplot2_2.0.0   lattice_0.20-33

loaded via a namespace (and not attached):
 [1] Rcpp_0.12.3        magrittr_1.5       splines_3.2.3      MASS_7.3-44        munsell_0.4.2     
 [6] colorspace_1.2-6   foreach_1.4.3      minqa_1.2.4        stringr_1.0.0      car_2.1-2         
[11] plyr_1.8.3         tools_3.2.3        nnet_7.3-11        pbkrtest_0.4-6     parallel_3.2.3    
[16] grid_3.2.3         gtable_0.1.2       nlme_3.1-126       mgcv_1.8-7         quantreg_5.21     
[21] e1071_1.6-7        class_7.3-14       MatrixModels_0.4-1 iterators_1.0.8    lme4_1.1-11       
[26] Matrix_1.2-3       nloptr_1.0.4       reshape2_1.4.1     codetools_0.2-14   stringi_1.0-1     
[31] compiler_3.2.3     scales_0.3.0       stats4_3.2.3       SparseM_1.7  

Thank you very much for your great package!!

[topepo]

It was related to the assumption that the predictors would be in a data frame. It works now...

Thanks,

Max

> library("caret")
Loading required package: lattice
Loading required package: ggplot2
Need help? Try the ggplot2 mailing list:
http://groups.google.com/group/ggplot2.
> 
> data(kyphosis, package = "rpart")
> 
> fitControl <- trainControl(method = "cv",
+                            timingSamps = 5)
> train(Kyphosis ~ ., data = kyphosis, 
+       method = "rpart", 
+       trControl = fitControl)
Loading required package: rpart
CART 

81 samples
 3 predictor
 2 classes: 'absent', 'present' 

No pre-processing
Resampling: Cross-Validated (10 fold) 
Summary of sample sizes: 73, 74, 72, 72, 73, 74, ... 
Resampling results across tuning parameters:

  cp          Accuracy   Kappa     
  0.00000000  0.8244048  0.41896846
  0.01960784  0.8244048  0.41896846
  0.17647059  0.7535714  0.04381318

Accuracy was used to select the optimal model using  the largest value.
The final value used for the model was cp = 0.01960784. 
> # Error in eval(expr, envir, enclos) : object 'Age' not found 
> 
> # Using non-formula interface works
> train(x = kyphosis[, -1], y = kyphosis$Kyphosis,
+       method = "rpart",
+       trControl = fitControl)
CART 

81 samples
 3 predictor
 2 classes: 'absent', 'present' 

No pre-processing
Resampling: Cross-Validated (10 fold) 
Summary of sample sizes: 73, 73, 73, 73, 73, 72, ... 
Resampling results across tuning parameters:

  cp          Accuracy   Kappa    
  0.00000000  0.7940476  0.3621554
  0.01960784  0.7940476  0.3621554
  0.17647059  0.7690476  0.1507268

Accuracy was used to select the optimal model using  the largest value.
The final value used for the model was cp = 0.01960784.