sergiooramas
commented
6 years ago I basically used Theano, so I'm not sure what happens with tensorflow and my code. You can also run the code with Keras 2, you can try that.
Open chenjie04 opened 6 years ago
sergiooramas
commented
6 years ago I basically used Theano, so I'm not sure what happens with tensorflow and my code. You can also run the code with Keras 2, you can try that.
chenjie04
commented
6 years ago thank you very much! I will try that now.
chenjie04
commented
6 years ago hello! I update the Keras(2.1.5) and tensorflow(1.7.0), but it throw out another error :ValueError: Error when checking target: expected activation_3 to have shape (3,) but got array with shape (1,).
The detail show as follow:
Traceback (most recent call last):
File "run_experiments.py", line 343, in
conv2d_1 (Conv2D) (None, 64, 319, 1) 24640
activation_1 (Activation) (None, 64, 319, 1) 0
max_pooling2d_1 (MaxPooling2 (None, 64, 79, 1) 0
dropout_1 (Dropout) (None, 64, 79, 1) 0
conv2d_2 (Conv2D) (None, 128, 76, 1) 32896
activation_2 (Activation) (None, 128, 76, 1) 0
max_pooling2d_2 (MaxPooling2 (None, 128, 38, 1) 0
dropout_2 (Dropout) (None, 128, 38, 1) 0
flatten_1 (Flatten) (None, 4864) 0
dense_1 (Dense) (None, 3) 14595
activation_3 (Activation) (None, 3) 0
Total params: 72,131 Trainable params: 72,131 Non-trainable params: 0
I can't find any solution on the Internet. please help me, thank you!
and I add K.set_image_dim_ordering('th') in models.py, is it cause the problem?
when I run you code on my machine with keras==1.1.0 and tensorflow==1.0.0, I catch the error as below: Traceback (most recent call last): File "run_experiments.py", line 343, in
process(params)
File "/home/jie_chen/recommendation/tartarus/src/train.py", line 322, in process
model = build_model(config)
File "/home/jie_chen/recommendation/tartarus/src/train.py", line 68, in build_model
model = get_model(params)
File "/home/jie_chen/recommendation/tartarus/src/models.py", line 160, in get_model_1
model.add(Flatten())
File "/home/jie_chen/tensorflow11/local/lib/python2.7/site-packages/keras/models.py", line 308, in add
output_tensor = layer(self.outputs[0])
File "/home/jie_chen/tensorflow11/local/lib/python2.7/site-packages/keras/engine/topology.py", line 514, in call
self.add_inbound_node(inbound_layers, node_indices, tensor_indices)
File "/home/jie_chen/tensorflow11/local/lib/python2.7/site-packages/keras/engine/topology.py", line 572, in add_inbound_node
Node.create_node(self, inbound_layers, node_indices, tensor_indices)
File "/home/jie_chen/tensorflow11/local/lib/python2.7/site-packages/keras/engine/topology.py", line 149, in create_node
output_tensors = to_list(outbound_layer.call(input_tensors[0], mask=input_masks[0]))
File "/home/jie_chen/tensorflow11/local/lib/python2.7/site-packages/keras/layers/core.py", line 409, in call
return K.batch_flatten(x)
File "/home/jie_chen/tensorflow11/local/lib/python2.7/site-packages/keras/backend/tensorflow_backend.py", line 823, in batch_flatten
x = tf.reshape(x, tf.pack([-1, prod(shape(x)[1:])]))
AttributeError: 'module' object has no attribute 'pack'
As far as I know, the tf.pack() has renamed as tf.stack(). So ,what can I do to fix the problem? to update the keras? please help! thanks!