ykamikawa / tf-keras-SegNet

SegNet including indices pooling for Semantic Segmentation with tensorflow and keras
178 stars 80 forks source link

How do I alter layers.py code if input_shape is (channels, height, width) #15

Closed nehSgnaiL closed 4 years ago

nehSgnaiL commented 4 years ago

There's a problem with me.

For some reason my "image_dim_ordering": is "th" and I don't know how the layers.py code works. Can anybody show the MaxPoolingWithArgmax2D & MaxUnpooling2D code when input shape is ( channels, height, width )?

Can anyone help? Thanks a lot.