1

为了在 R 中拟合分类模型，一直在使用library(KerasR). 控制学习率和KerasR说

compile(optimizer=Adam(lr = 0.001, beta_1 = 0.9, beta_2 = 0.999, epsilon = 1e-08, decay = 0, clipnorm = -1, clipvalue = -1), loss      = 'binary_crossentropy', metrics   =  c('categorical_accuracy') )

但它给了我这样的错误

模块错误$keras.optimizers$Adam(lr = lr, beta_1 = beta_2, beta_2 = beta_2, : 尝试应用非函数

我也用过keras_compile仍然得到同样的错误。我可以在编译时更改优化器，但最大学习率为 0.01，我想尝试 0.2。

model <- keras_model_sequential()

model %>% layer_dense(units = 512, activation = 'relu',  input_shape =  ncol(X_train)) %>% 
  layer_dropout(rate = 0.2) %>% 
  layer_dense(units = 128, activation = 'relu')%>%
  layer_dropout(rate = 0.1) %>% 
  layer_dense(units = 2, activation = 'sigmoid')%>%
compile( 
  optimizer = 'Adam', 
  loss      = 'binary_crossentropy',
  metrics   =  c('categorical_accuracy') 
)
4

1 回答 1

2

我认为问题在于您同时使用了两个不同的kerasR库keras。您应该只使用其中之一。首先，您使用keras_model_sequential的是来自的函数，keras然后您尝试使用Adam来自kerasR库的函数。您可以在这里找到这两个库之间的区别：https ://www.datacamp.com/community/tutorials/keras-r-deep-learning#differences

以下代码对我有用，它只使用keras库。

library(keras)
model <- keras_model_sequential()

model %>% 
  layer_dense(units = 512, activation = 'relu',  input_shape =  ncol(X_train)) %>% 
  layer_dropout(rate = 0.2) %>% 
  layer_dense(units = 128, activation = 'relu')%>%
  layer_dropout(rate = 0.1) %>% 
  layer_dense(units = 2, activation = 'sigmoid')%>%
  compile(optimizer=optimizer_adam(lr = 0.2), loss= 'binary_crossentropy', metrics   =  c('accuracy') )
于 2020-08-25T03:38:35.443 回答