Artificial neural network - compilation error
I am learning Deep Learning on my own and am having trouble doing ANN. That's what I'm doing:
ANN initialization (I pre-split the dataset):
classifier = Sequential()
Adding the input layer and the first hidden layer:
classifier.add(Dense(input_dim = 11, kernel_initializer = 'uniform', activation = 'relu', units = 6))
Adding a second hidden layer:
classifier.add(Dense(units = 6, kernel_initializer = 'uniform', activation = 'relu'))
Adding an output layer:
classifier.add(Dense(units = 1, kernel_initializer = 'uniform', activation = 'sigmoid'))
Compiling ANN using stochastic gradient descent:
classifier.compile(optimizer = 'adam', loss = 'binary_crossentropy', metrics = ['accuracy'])
After that, when I select and run the last command, I get the error:
TypeError: sigmoid_cross_entropy_with_logits() got an unexpected keyword argument 'labels'
I noticed that when I use it loss = mean_squared_error
, it compiles fine. Can you tell me what's going on?
Sypder
and Python
the latter, in the day when I publish it.
Windows 10
... Thanos, TensorFlow and Keras recent
Thanks in advance.
source to share
Update your tensorflow version with nightly build:
https://github.com/tensorflow/tensorflow#installation
see this issue: https://github.com/carpedm20/DCGAN-tensorflow/issues/84
source to share