"rf" Error when using only one PC from PCA

[baderstine]

If you are filing a bug, make sure these boxes are checked before submitting your issue— thank you!

• Start a new R session
• Install the latest version of caret: update.packages(oldPkgs="caret", ask=FALSE)
• Write a minimal reproducible example
• Do not use parallel processing in the code (unless you are certain that the issue is about parallel processing).
• run sessionInfo()

Minimal, reproducible example:

library(caret)
dat = iris
dat$Species2Class = factor(dat$Species == "setosa", levels = c("TRUE", "FALSE"), labels = c("SETOSA", "OTHER"))

# constants used for random seeds
n = 2
r = 5

# set seeds to use for model reproducibility with parallel processing
set.seed(123)
seeds <- vector(mode = "list", length = (n*r)+1)

# for the cross-validation models
for(i in 1:((n*r))) seeds[[i]] <- sample.int(10000,5000)

# for the final model
seeds[[((n*r)+1)]] <- sample.int(1000,1)

fitControl <- trainControl(
  seeds = seeds,
  method = "repeatedcv",
  number = n,
  repeats = r,
  savePredictions = 'final',
  classProbs = T,
  preProcOptions = list(pcaComp = 1),
  summaryFunction = prSummary)

set.seed(1)
response = "Species2Class"
preds = names(dat)[1:4]
model.fmla = as.formula(paste0(response, " ~ ", paste0(preds, collapse="+")))

# run the model 
fit_model <- train(form = model.fmla, 
                   data = dat, 
                   method = "rf",
                   trControl = fitControl,
                   preProcess = c("pca"),
                   metric = "AUC")

# OUTPUT: 
# Error in { : task 1 failed - "subscript out of bounds"
# In addition: There were 50 or more warnings (use warnings() to see the first 50)

This happens when attempting to use only the first principal component. With 2 or more it works fine.

Session Info:

>sessionInfo()

R version 3.6.3 (2020-02-29)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS Catalina 10.15.7

Matrix products: default
BLAS:   /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.6/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

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

other attached packages:
[1] caret_6.0-86    ggplot2_3.3.1   lattice_0.20-41

loaded via a namespace (and not attached):
 [1] Rcpp_1.0.4.6         pillar_1.4.4         compiler_3.6.3       gower_0.2.1          plyr_1.8.6          
 [6] iterators_1.0.12     class_7.3-17         tools_3.6.3          rpart_4.1-15         ipred_0.9-9         
[11] lubridate_1.7.9      lifecycle_0.2.0      tibble_3.0.1         nlme_3.1-148         gtable_0.3.0        
[16] pkgconfig_2.0.3      rlang_0.4.6          Matrix_1.2-18        foreach_1.5.0        rstudioapi_0.11     
[21] prodlim_2019.11.13   stringr_1.4.0        withr_2.2.0          dplyr_1.0.0          MLmetrics_1.1.1     
[26] pROC_1.16.2          generics_0.0.2       vctrs_0.3.1          recipes_0.1.12       stats4_3.6.3        
[31] grid_3.6.3           nnet_7.3-14          tidyselect_1.1.0     data.table_1.12.8    glue_1.4.1          
[36] R6_2.4.1             survival_3.2-3       lava_1.6.7           ROCR_1.0-11          reshape2_1.4.4      
[41] purrr_0.3.4          magrittr_1.5         ModelMetrics_1.2.2.2 scales_1.1.1         codetools_0.2-16    
[46] ellipsis_0.3.1       MASS_7.3-51.6        splines_3.6.3        randomForest_4.6-14  timeDate_3043.102   
[51] colorspace_1.4-1     stringi_1.4.6        munsell_0.5.0        crayon_1.3.4        

[topepo]

Should work now