tartarus icon indicating copy to clipboard operation
tartarus copied to clipboard

Published 20 hours ago • verified publisher icon sergiooramas

which version of tensorflow do you use in the project?

Open chenjie04 opened this issue 6 years ago • 3 comments

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!

chenjie04 avatar Apr 03 '18 13:04 chenjie04

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.

sergiooramas avatar Apr 03 '18 14:04 sergiooramas

thank you very much! I will try that now.

chenjie04 avatar Apr 04 '18 00:04 chenjie04

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 process(params) File "/home/jie_chen/recommendation/tartarus/src/train.py", line 397, in process callbacks=[early_stopping]) File "/home/jie_chen/tensorflow11/local/lib/python2.7/site-packages/keras/legacy/interfaces.py", line 91, in wrapper return func(*args, **kwargs) File "/home/jie_chen/tensorflow11/local/lib/python2.7/site-packages/keras/models.py", line 1276, in fit_generator initial_epoch=initial_epoch) File "/home/jie_chen/tensorflow11/local/lib/python2.7/site-packages/keras/legacy/interfaces.py", line 91, in wrapper return func(*args, **kwargs) File "/home/jie_chen/tensorflow11/local/lib/python2.7/site-packages/keras/engine/training.py", line 2160, in fit_generator val_x, val_y, val_sample_weight) File "/home/jie_chen/tensorflow11/local/lib/python2.7/site-packages/keras/engine/training.py", line 1480, in _standardize_user_data exception_prefix='target') File "/home/jie_chen/tensorflow11/local/lib/python2.7/site-packages/keras/engine/training.py", line 123, in _standardize_input_data str(data_shape)) ValueError: Error when checking target: expected activation_3 to have shape (3,) but got array with shape (1,) And the summary as this:

Layer (type) Output Shape Param #

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?

chenjie04 avatar Apr 04 '18 01:04 chenjie04