Typo in Custom Model
miqbal23 opened this issue · comments
I found a typo at Custom Model section, 2nd cell
# Layer 2
if use_layers == 2:
model.add(Convolution2D(conv_filters, 3, 3, activation=cnn_activation, border_mode='valid', input_shape=input_shape))
#model.add(BatchNormalization())
model.add(Activation(cnn_act)) <---
model.add(MaxPooling2D(pool_size=(2, 2)))
shouldn't cnn_act
be cnn_activation
? Because in the previous cell there's a variable called cnn_activation
that contains the information needed for activation function of the layer
That's correct.
However it is already present in Convolution2D(conv_filters, 3, 3, activation=cnn_activation, ...)
so this line can be removed / commented out.
(thats what i did in the latest version available in the repo).