我正在尝试将package caret和nnet一起使用，但是得到了以下错误。我使用警告（）来查看它是什么，但在互联网上找不到任何东西。请告诉我。感谢。
> str(Std_data)
'data.frame': 1628 obs. of 18 variables:
$ Seed : num 1 1 1 1 1 1 1 1 1 1 ...
$ Pyth : num 0.882 0.904 0.295 0.882 0.671 ...
$ EFG : num 0.709 -0.212 -0.992 0.709 -0.291 ...
$ PPP : num 0.506 0.2393 0.3952 0.506 0.0309 ...
...
$ PD : num 2.17 2.03 1.47 1.4 1.19 ...
$ OppSeed: num 16 16 16 9 16 5 3 3 12 3 ...
$ Win : Factor w/ 2 levels "0","1": 2 2 2 2 2 2 1 1 2 1 ...
> model <- train(Win ~ . - PD, data=Std_data, method='nnet', linout=TRUE, trace = FALSE,
+ tuneGrid=expand.grid(.size=c(2:8),.decay=c(0,0.0001, 0.001,0.01,0.1)), trControl = trainControl(method = "cv"))
Something is wrong; all the Accuracy metric values are missing:
...
Error in train.default(x, y, weights = w, ...) : Stopping
In addition: There were 50 or more warnings (use warnings() to see the first 50)
> warnings()
Warning messages:
1: In eval(expr, envir, enclos) :
model fit failed for Fold01: size=2, decay=0e+00 Error in nnet.default(x, y, w, entropy = TRUE, ...) :
entropy fit only for logistic units
答案 0 :(得分：1)
如果将linout设置为False，您的代码可能会有效。 Linout设置为线性输出单元的True开关。您的Win变量是二进制变量而不是连续变量。至少它只会消除物流单位信息的熵拟合。