Keras: ValueError: ('The specified size contains a dimension with value <= 0', (-11520,))

[Simrankgp]

I am using keras 2.0.2 and trying to run a Conv net

patch_size = 15

print(train_set.shape, 'train samples')

batch_size = 1
nb_classes = 6
nb_epoch = 100

Y_train = np_utils.to_categorical(y_train, nb_classes)

nb_filters = [30, 30]

nb_pool = [2, 2]

nb_conv = [5, 5]

train_set = train_set.astype('float32')

train_set -= np.mean(train_set)

train_set /= np.max(train_set)


model = Sequential()
model.add(Conv3D(nb_filters[0],(nb_conv[0],nb_conv[0],nb_conv[0]),input_shape=(1, img_rows, img_cols, patch_size), activation='relu'))
model.add(MaxPooling3D(pool_size=(nb_pool[0], nb_pool[0], nb_pool[0])))
model.add(Conv3D(nb_filters[0],(nb_conv[0] - 2,nb_conv[0] - 2,nb_conv[0] - 2),input_shape=(1, img_rows, img_cols, patch_size), activation='relu'))

model.add(MaxPooling3D(pool_size=(nb_pool[0], nb_pool[0], nb_pool[0])))
model.add(Conv3D(nb_filters[0],(nb_conv[0]-3, nb_conv[0]-3, nb_conv[0]-3), input_shape=(1, img_rows, img_cols, patch_size), activation='tanh'))
model.add(MaxPooling3D(pool_size=(nb_pool[0], nb_pool[0], nb_pool[0])))
model.add(Conv3D(nb_filters[0],(nb_conv[0]-3, nb_conv[0]-3, nb_conv[0]-3), input_shape=(1, img_rows, img_cols, patch_size), activation='tanh'))


model.add(Dropout(0.5))

model.add(Flatten())

print (model.output_shape)
model.add(Dense(128, kernel_initializer='normal', activation='relu'))

It gives me the train_sample size as = (97, 1, 28, 28, 15)
But then it gives the following error in the dense command

(None,-90)
Traceback (most recent call last):
  File "F:/Project/codes/foreg.py", line 144, in <module>
    print (model.output_shape)
  File "C:\Users\lenov\Anaconda3\envs\3dcnn\lib\site-packages\keras\models.py", line 455, in add
    output_tensor = layer(self.outputs[0])
  File "C:\Users\lenov\Anaconda3\envs\3dcnn\lib\site-packages\keras\engine\topology.py", line 528, in __call__
    self.build(input_shapes[0])
  File "C:\Users\lenov\Anaconda3\envs\3dcnn\lib\site-packages\keras\layers\core.py", line 827, in build
    constraint=self.kernel_constraint)
  File "C:\Users\lenov\Anaconda3\envs\3dcnn\lib\site-packages\keras\engine\topology.py", line 364, in add_weight
    weight = K.variable(initializer(shape), dtype=K.floatx(), name=name)
  File "C:\Users\lenov\Anaconda3\envs\3dcnn\lib\site-packages\keras\initializers.py", line 73, in __call__
    dtype=dtype, seed=self.seed)
  File "C:\Users\lenov\Anaconda3\envs\3dcnn\lib\site-packages\keras\backend\theano_backend.py", line 1960, in random_normal
    return rng.normal(size=shape, avg=mean, std=stddev, dtype=dtype)
  File "C:\Users\lenov\Anaconda3\envs\3dcnn\lib\site-packages\theano\sandbox\rng_mrg.py", line 1574, in normal
    nstreams=nstreams)
  File "C:\Users\lenov\Anaconda3\envs\3dcnn\lib\site-packages\theano\sandbox\rng_mrg.py", line 1344, in uniform
    size)
ValueError: ('The specified size contains a dimension with value <= 0', (-11520,))

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed after 30 days if no further activity occurs, but feel free to re-open a closed issue if needed.

[mclark4386]

Did you ever find a solution to this I'm running into something very like it.

[erralves]

@Simrankgp, @mclark4386,

This looks like a problem with the inputs size :
Using Theano as backend you state the inputs this way : inputs = Input(shape=(3,227,227))
Using Tensorflow as backend : inputs = Input(shape=(227,227,3))

I suggest you to use Theano as backend. Then, in your keras.json, you can adjust "image_data_format": "channels_last" if you want declare in Tensorflow manner.