search

lukovkin / ufcnn-keras

Implementation of UFCNN in Keras
MIT License
93 stars 26 forks source link

TypeError: unsupported operand type(s) for +: 'NoneType' and 'int' #4

Closed developeralgo8888 closed 8 years ago

developeralgo8888 commented 8 years ago

Getting type error on the current Commit.

Using Theano backend. Using gpu device 6: GeForce GTX TITAN X (CNMeM is enabled with initial size: 90.0% of memory, CuDNN 4004) /home/anaconda3/lib/python3.5/site-packages/theano/tensor/signal/downsample.py:5: UserWarning: downsample module has been moved to the pool module. warnings.warn("downsample module has been moved to the pool module.") Running model: tradcom Traceback (most recent call last): File "UFCNN1.py", line 623, in loss="categorical_crossentropy", sequence_length=sequence_length, optimizer=sgd ) File "UFCNN1.py", line 215, in ufcnn_model init) File "UFCNN1.py", line 131, in ufcnn_model_concat model.add_node(Convolution1D(nb_filter=nb_filter, filter_length=filter_length, border_mode='valid', init=init, input_shape=(sequence_length, features)), name='conv1', input='input_padding') File "/home/anaconda3/lib/python3.5/site-packages/keras/layers/containers.py", line 359, in add_node layer.set_previous(self.nodes[input]) File "/home/anaconda3/lib/python3.5/site-packages/keras/layers/core.py", line 98, in set_previous assert self.input_ndim == len(layer.output_shape), ('Incompatible shapes: layer expected input with ndim=' + File "/home/anaconda3/lib/python3.5/site-packages/keras/layers/convolutional.py", line 707, in output_shape input_shape[1] + self.padding * 2, TypeError: unsupported operand type(s) for +: 'NoneType' and 'int'

lukovkin commented 8 years ago

Works on Python 2.7, Python 3.5 on CPU with latest master branch of Keras. Please check.

developeralgo8888 commented 8 years ago

Upgraded Keras 0.3.2 master branch from the source

developeralgo8888 commented 8 years ago

Issue resolved