Does not work with tensorflow-gpu

[Vovak1919]

After installing tensorflow-gpu, errors appeared in the model.py module:
Line 138

	def conv_layer(self, x, filters, kernel_size):
		x = Conv2D(
		filters = filters
		, kernel_size = kernel_size
		, data_format="channels_first"
		, padding = 'same'
		, use_bias=False
		, activation='linear'
		, kernel_regularizer = regularizers.l2(self.reg_const)
		)(x)
		x = BatchNormalization(axis=1)(x)

150 x = BatchNormalization(axis=1)(x)
^
ValueError
Shape must be rank 1 but is rank 0 for 'batch_normalization_1/cond/Reshape_4' (op: 'Reshape') with input shapes: [1,75,1,1], [].
File "C:\DRL\model.py", line 150, in conv_layer
x = BatchNormalization(axis=1)(x)
File "C:\DRL\model.py", line 225, in _build_model
x = self.conv_layer(main_input, self.hidden_layers[0]['filters'], self.hidden_layers[0]['kernel_size'])
File "C:\DRL\model.py", line 114, in init
self.model = self._build_model()
File "C:\DRL\Untitled-1.py", line 67, in
current_NN = Residual_CNN(config.REG_CONST, config.LEARNING_RATE, (2,) + env.grid_shape, env.action_size, config.HIDDEN_CNN_LAYERS)

[Vovak1919]

Shape:TensorShape([Dimension(None), Dimension(75), Dimension(6), Dimension(7)])
dims:[Dimension(None), Dimension(75), Dimension(6), Dimension(7)]
0:Dimension(None)
1:Dimension(75)
2:Dimension(6)
3:Dimension(7)
len:4
ndims:4
_dims:[Dimension(None), Dimension(75), Dimension(6), Dimension(7)]
_ndims:4
_tf_api_names:('TensorShape',)
_tf_api_names_v1:('TensorShape',)
value_index:0
_consumers:[]
_dtype:tf.float32
_id:21

[Vovak1919]

The module is called from here.

# create an untrained neural network objects from the config file
current_NN = Residual_CNN(config.REG_CONST, config.LEARNING_RATE, (2,) + env.grid_shape,   env.action_size, config.HIDDEN_CNN_LAYERS)

Call Stack:
conv_layer (c:\DRL\model.py:150)
_build_model (c:\DRL\model.py:225)
init (c:\DRL\model.py:114)
(c:\DRL\Untitled-1.py:67)

[acautin]

@Vovak1919 at least for

Shape must be rank 1 but is rank 0 for 'batch_normalization_1/cond/Reshape_4' (op: 'Reshape') with input shapes: [1,75,1,1], [].

It seems to be a problem with keras, the version 2.1.5 worked for me. (It should also be solved on master but not yet on the pip version)

Ref: keras-team/keras#10648

[Vovak1919]

Thank you, earned after correcting the file tensorflow_backend.py
keras-team/keras@e3a2f7d

[ahmedosama9777]

What did you do as a correction?