1

很抱歉今天所有与 purrr 相关的问题，仍在试图弄清楚如何有效地利用它。

因此，在 SO 的帮助下，我设法根据来自 data.frame 的输入值运行随机森林护林员模型。这是使用purrr::pmap. 但是，我不明白返回值是如何从被调用函数生成的。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 3*3 混淆矩阵的 4 倍，因为 中有 3 个级别iris$Species，而是返回巨大的混淆矩阵。有人可以向我解释发生了什么吗？

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
4

1 回答 1

1

这里的问题是因为传递给函数的参数是级别，而不是字符。这触发了游侠功能。要解决这个问题，您需要做的就是stringsAsFactors = FALSE在expand.grid：

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
于 2020-03-31T19:12:14.440 回答