很抱歉今天出现了所有与purrr相关的问题,仍在努力找出如何有效利用它。
因此,在So的一些帮助下,我设法使随机护林员模型基于来自data.frame的输入值运行。这是使用purrr::pmap
完成的。然而,我不明白返回值是如何从被调用的函数中生成的。考虑这个例子:
library(ranger)
data(iris)
Input_list <- list(iris1 = iris, iris2 = iris) # let's assume these are different input tables
# the data.frame with the values for the function
hyper_grid <- expand.grid(
Input_table = names(Input_list),
mtry = c(1,2),
Classification = TRUE,
Target = "Species")
> hyper_grid
Input_table mtry Classification Target
1 iris1 1 TRUE Species
2 iris2 1 TRUE Species
3 iris1 2 TRUE Species
4 iris2 2 TRUE Species
# the function to be called for each row of the `hyper_grid`df
fit_and_extract_metrics <- function(Target, Input_table, Classification, mtry,...) {
RF_train <- ranger(
dependent.variable.name = Target,
mtry = mtry,
data = Input_list[[Input_table]], # referring to the named object in the list
classification = Classification) # otherwise regression is performed
RF_train$confusion.matrix
}
# the pmap call using a row of hyper_grid and the function in parallel
purrr::pmap(hyper_grid, fit_and_extract_metrics)
它应该返回4倍的3*3混淆矩阵,因为iris$Species
中有3个级别,但它返回的是巨大的混淆矩阵。有人能向我解释一下发生了什么事吗?
第一行:
> purrr::pmap(hyper_grid, fit_and_extract_metrics)
[[1]]
predicted
true 4.4 4.7 4.8 4.9 5 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 5.9 6 6.1 6.2 6.3 6.4
4.3 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
4.4 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
4.5 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
4.6 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0
4.7 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
4.8 0 0 1 3 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0
4.9 0 0 1 2 2 0 0 0 0 0 0 0 0 0 1 0 0 0 0
5 0 0 0 1 9 0 0 0 0 0 0 0 0 0 0 0 0 0 0
5.1 0 0 0 0 0 8 0 0 0 1 0 0 0 0 0 0 0 0 0
这里的问题是因为传递给函数的参数是级别,而不是字符。这使护林员功能失灵。要解决这个问题,您只需要在expand.grid
:中设置stringsAsFactors = FALSE
hyper_grid <- expand.grid(
Input_table = names(Input_list),
mtry = c(1,2),
Classification = TRUE,
Target = "Species", stringsAsFactors = FALSE)
你会得到:
[[1]]
predicted
true setosa versicolor virginica
setosa 50 0 0
versicolor 0 46 4
virginica 0 4 46
[[2]]
predicted
true setosa versicolor virginica
setosa 50 0 0
versicolor 0 46 4
virginica 0 5 45
[[3]]
predicted
true setosa versicolor virginica
setosa 50 0 0
versicolor 0 47 3
virginica 0 3 47
[[4]]
predicted
true setosa versicolor virginica
setosa 50 0 0
versicolor 0 47 3
virginica 0 3 47