Implementation of CNN+ bidirectional LSTM for videos about keras-extra HOT 6 CLOSED

@anayebi Why have you closed the issue?

from keras-extra.

Already answered here (putting the link for reference): keras-team/keras#2646 (comment)

from keras-extra.

@anayebi Can you give me the link to Keras 0.3.0?

from keras-extra.

If you already have Keras 1.0.2 installed, isn't it easier for you to just use the TimeDistributed wrapper on Convolution2D? It ought to do the same thing as TimeDistributedConvolution2D.

from keras-extra.

@anayebi ok i used the exact same example code that you have provided in the README -
model.add(TimeDistributed(Convolution2D(8, 4, 4, border_mode='valid', input_shape=(10,1,32,32)))) model.add(TimeDistributed(MaxPooling2D(pool_size=(2, 2),border_mode='valid'))) model.add(Activation('relu')) model.add(TimeDistributed(Convolution2D(8, 3, 3, border_mode='valid'))) model.add(Activation('relu')) model.add(TimeDistributedFlatten()) model.add(Activation('relu')) model.add(GRU(output_dim=100,return_sequences=True)) model.add(GRU(output_dim=50,return_sequences=False)) model.add(Dropout(.2)) model.add(Dense(1))

I still get an error -

File "/home/toothless/daksh/ml/project/src/model/model2.py", line 133, in buildModel
model.add(TimeDistributed(Convolution2D(8, 4, 4, border_mode='valid', input_shape=(10,1,32,32))))
File "/usr/local/lib/python2.7/dist-packages/keras/models.py", line 102, in add
raise Exception('The first layer in a Sequential model must '
Exception: The first layer in a Sequential model must get an input_shape or batch_input_shape argument.

I have added the input_shape argument to the first layer, but still it throws an error. Please help me out.

from keras-extra.

@anayebi Just to add.. I am getting the following error while building the model itself. I have not used model.fit() still.

from keras-extra.