ValueError due to dimension mismatch

[AnkitaBeniwal]

Hi,

Could you please help me out with this one, I am using the same dataset as yours and just ran the training script "ValueError: Negative dimension size caused by subtracting 3 from 1 for 'conv2d_1/convolution' (op: 'Conv2D') with input shapes: [?,1,32,32], [3,3,32,32]."

Second what is the configuration of ~/.keras/keras.json like about backend and image_ordering

[CattleOfRa]

Also having this same issue. Has it been resolved?

[emadeldeen24]

In line 228 in train.py update the code to become:

x = Convolution2D(n_filters, (kH, kW), subsample=(sH, sW),
                          border_mode=border_mode, init=init_conv, data_format = 'channels_first')(x)