在神经网络和插入符号 (R) 中设置隐藏层和神经元

Question

我想使用包neuralnet和caret交叉验证神经网络。

可以从这篇文章中复制数据df。

运行neuralnet()函数时，有一个名为hidden的参数，您可以在其中设置每个隐藏层和神经元。假设我想要 2 个分别具有 3 个和 2 个神经元的隐藏层。它会写成hidden = c(3, 2).

但是，由于我想交叉验证它，我决定使用出色的caret包。但是在使用函数train()时，我不知道如何设置层数和神经元数。

有谁知道我在哪里可以添加这些数字？

这是我运行的代码：

nn <- caret::train(DC1 ~ ., data=df, 
method = "neuralnet", 
#tuneGrid = tune.grid.neuralnet,
metric = "RMSE",
trControl = trainControl (
method = "cv", number = 10,
verboseIter = TRUE
))

顺便说一下，我收到一些警告，以前的代码：

predictions failed for Fold01: layer1=3, layer2=0, layer3=0 Error in cbind(1, pred) %*% weights[[num_hidden_layers + 1]] : 
requires numeric/complex matrix/vector arguments

关于如何解决它的想法？

Answer 1

在插入符号中使用神经网络模型以指定三个支持的层中每个层中的隐藏单元的数量时，您可以使用参数layer1、layer2和layer3。我通过检查来源发现了。

library(caret)
grid <-  expand.grid(layer1 = c(32, 16),
layer2 = c(32, 16),
layer3 = 8)

波士顿住房数据的用例：

library(mlbench)
data(BostonHousing)

让我们只为示例选择数字列以使其简单：

BostonHousing[,sapply(BostonHousing, is.numeric)] -> df
nn <- train(medv ~ ., 
data = df, 
method = "neuralnet", 
tuneGrid = grid,
metric = "RMSE",
preProc = c("center", "scale", "nzv"), #good idea to do this with neural nets - your error is due to non scaled data
trControl = trainControl(
method = "cv",
number = 5,
verboseIter = TRUE)
)

零件

preProc = c("center", "scale", "nzv")

对于算法收敛至关重要，神经网络不喜欢未缩放的特征

虽然它超级慢。

nn
#output
Neural Network 
506 samples
12 predictor
Pre-processing: centered (12), scaled (12) 
Resampling: Cross-Validated (5 fold) 
Summary of sample sizes: 405, 404, 404, 405, 406 
Resampling results across tuning parameters:
layer1  layer2  RMSE      Rsquared   MAE     
16      16           NaN        NaN       NaN
16      32      4.177368  0.8113711  2.978918
32      16      3.978955  0.8275479  2.822114
32      32      3.923646  0.8266605  2.783526
Tuning parameter 'layer3' was held constant at a value of 8
RMSE was used to select the optimal model using the smallest value.
The final values used for the model were layer1 = 32, layer2 = 32 and layer3 = 8.