How to get the Gini coefficient using random forests in the caret R package? -

i'm trying understand difference between random forest implementation in randomforest package , in caret package.

for example, specifies 2000 trees mtry = 2 in randomforest , show gini coefficient each predictor:

library(randomforest) library(tidyr)  rf1 <- randomforest(species ~ ., data = iris,                        ntree = 2000, mtry = 2,                       importance = true) data.frame(rf = sort(importance(rf1)[, "meandecreasegini"], decreasing = true)) %>% add_rownames() %>% rename(predictor = rowname) #      predictor       rf # 1  petal.width 45.57974 # 2 petal.length 41.61171 # 3 sepal.length  9.59369 # 4  sepal.width  2.47010 

i'm trying same info in caret, don't know how specify number of trees, or how gini coefficient:

rf2 <- train(species ~ ., data = iris, method = "rf",               metric = "kappa",                tunegrid = data.frame(mtry = 2)) varimp(rf2) # not gini coefficient #              overall # petal.length 100.000 # petal.width   99.307 # sepal.width    0.431 # qsepal.length  0.000 

also, confusion matrix of rf1 has errors , of rf2 doesn't. parameter causing difference?:

# rf1 confusion matrix: #            setosa versicolor virginica class.error # setosa         50          0         0        0.00 # versicolor      0         47         3        0.06 # virginica       0          4        46        0.08  table(predict(rf2, iris), iris$species) #             setosa versicolor virginica #  setosa         50          0         0 #  versicolor      0         50         0 #  virginica       0          0        50 

this quick , dirty. know isn't right way test performance of classifier, dont' understand difference in results.

this might answer question - see 2nd post:

caret: using random forest , include cross-validation

randomforest sampling replacement. if use "rf" in caret, need specify trcontrol in train::caret(); want same resampling method used in caret i.e. bootstrap, need set trcontrol="oob". trcontrol list of values defines how function acts; can set "cv" cross validation, "repeatedcv" repeated cross validation etc. see caret package documentation more info.

you should same result using randomforest, remember set seeds properly.