从R中的列表中提取结果

Question

我有一个名为 bag 的列表。这是来自mlr包（伪）代码的许多回归方法的结果如下所示。我想从每个任务 中提取汇总的效果结果。例如，我如何访问＆＃34; Aggr。 PERF＆＃34;来自$ visc.1 $ regr.rpart和$ visc.2 $ regr.rpart的值。我可以单独访问它们，但我认为必须有一个更简单的方法。

> class(bag)

[1] "list"

代码生成包（不包括数据集）：

library('mlr')
dataset = read.csv("dataset.csv")
regr.task = makeRegrTask(id = "dataset", data = dataset, target = "target")
# feature reduction
fv = generateFilterValuesData(regr.task)

# resampling description    
rdesc = makeResampleDesc("Holdout")

####################
bag = NULL
####################

for (i in 1:2)
{

  cols = c(order(fv$data$randomForestSRC.rfsrc, decreasing = TRUE)[1:i], ncol(dataset))
  dataset.ig = dataset[, cols]
  iid = paste(c("dataset", i), collapse = ".")
  regr.task = makeRegrTask(id = iid,
                           data = dataset.ig,
                           target = "dataset")

  # learners
  lrns = list(
    "regr.bcart"
    ,"regr.fnn"
    ,"regr.rpart"
  )

      set.seed(0, "L'Ecuyer")
      bmr = benchmark(lrns, regr.task, rdesc, show.info = FALSE)

      #########################
      bag = c(bag, bmr)
      #########################
}

MWC：

for (i in seq(1,42,3))
{
    print (bag[i]$results)
}

输出：

$visc.1
$visc.1$regr.bcart
Resample Result
Task: visc.1
Learner: regr.bcart
Aggr perf: mse.test.mean=5.14e+03
Runtime: 0.181672

$visc.1$regr.cforest
Resample Result
Task: visc.1
Learner: regr.cforest
Aggr perf: mse.test.mean=4.92e+03
Runtime: 0.103091

$visc.1$regr.fnn
Resample Result
Task: visc.1
Learner: regr.fnn
Aggr perf: mse.test.mean=2.44e+03
Runtime: 0.0151947

$visc.1$regr.mars
Resample Result
Task: visc.1
Learner: regr.mars
Aggr perf: mse.test.mean=3.77e+03
Runtime: 0.0163326

$visc.1$regr.rpart
Resample Result
Task: visc.1
Learner: regr.rpart
Aggr perf: mse.test.mean=4.77e+03
Runtime: 0.0265007


$visc.2
$visc.2$regr.bcart
Resample Result
Task: visc.2
Learner: regr.bcart
Aggr perf: mse.test.mean=5.14e+03
Runtime: 0.177014

$visc.2$regr.cforest
Resample Result
Task: visc.2
Learner: regr.cforest
Aggr perf: mse.test.mean=4.95e+03
Runtime: 0.115235

$visc.2$regr.fnn
Resample Result
Task: visc.2
Learner: regr.fnn
Aggr perf: mse.test.mean=3.25e+03
Runtime: 0.0116491

$visc.2$regr.mars
Resample Result
Task: visc.2
Learner: regr.mars
Aggr perf: mse.test.mean=2.67e+03
Runtime: 0.0153017

$visc.2$regr.rpart
Resample Result
Task: visc.2
Learner: regr.rpart
Aggr perf: mse.test.mean=4.77e+03
Runtime: 0.0252295

修改 我还保存了dput（bag [1：2]）here。

Answer 1

根据上述mtcars示例，从bag列创建列表。

io <-list(mtcars$mpg, mtcars$cyl, mtcars$disp)
names(io) <- c("mpg","cyl","disp")
class(io)
[1] "list"

io
$mpg
 [1] 21.0 21.0 22.8 21.4 18.7 18.1 14.3 24.4 22.8 19.2 17.8 16.4 17.3 15.2 10.4 10.4 14.7 32.4 30.4 33.9 21.5
[22] 15.5 15.2 13.3 19.2 27.3 26.0 30.4 15.8 19.7 15.0 21.4

$cyl
 [1] 6 6 4 6 8 6 8 4 4 6 6 8 8 8 8 8 8 4 4 4 4 8 8 8 8 4 4 4 8 6 8 4

$disp
 [1] 160.0 160.0 108.0 258.0 360.0 225.0 360.0 146.7 140.8 167.6 167.6 275.8 275.8 275.8 472.0 460.0 440.0
[18]  78.7  75.7  71.1 120.1 318.0 304.0 350.0 400.0  79.0 120.3  95.1 351.0 145.0 301.0 121.0

实施 MWC ：

req_rows <- seq(1,length(io$mpg),3)
req_rows
[1]  1  4  7 10 13 16 19 22 25 28 31


# rather than using loop, use this
io$mpg[io_seq]
[1] 21.0 21.4 14.3 19.2 17.3 10.4 30.4 15.5 19.2 30.4 15.0