randomforest variable importance: results of mlr3 is different from randomForest packages

[slecee]

i want to know why the results are different? Thanks for the help!

mlr3

library(mlr3)library(mlr3verse)
library(mlr3learners)
library(randomForest)

tasks = as_task_classif(iris, target = 'Species')
learners = lrn("classif.randomForest", predict_type = 
"prob",importance= c('gini'))
set.seed(123, kind = "Mersenne-Twister")
split = partition(tasks)
split

a = learners$train(tasks,row_ids = split$train)
a$model$importance

setosa  versicolor  virginica MeanDecreaseAccuracy MeanDecreaseGini
Petal.Length 0.318035851 0.266299298 0.30394373          0.291450470        29.166804
Petal.Width  0.343312702 0.285824160 0.24060252          0.287471876        29.279836
Sepal.Length 0.044516441 0.017425256 0.03575046          0.032833987         7.045327
Sepal.Width  0.007014524 0.009913653 0.00355260          0.006423498         1.783527

randomForest

set.seed(123, kind = "Mersenne-Twister")
tmp <- randomForest(iris[split$train,1:4], 
                    iris$Species[split$train], 
                    importance = TRUE)
tmp[["importance"]]

 setosa versicolor   virginica MeanDecreaseAccuracy MeanDecreaseGini
Sepal.Length 0.028980891 0.01123579 0.041315479          0.027508393         6.729108
Sepal.Width  0.007498441 0.01019763 0.006658933          0.008441336         1.853704
Petal.Length 0.300187608 0.25654441 0.310950138          0.285067350        28.924225
Petal.Width  0.361163535 0.29720214 0.250040471          0.299030008        29.750160

actually,setting the seed does not yield the same results!

[be-marc]

I already answered your question #974.

[slecee]

The result is still different...