Closed yl3639 closed 6 years ago
Hi, thanks for the code sharing. I am just going through the code and I am wondering why the raw_output raw_output = net.layers['fc1_voc12']
has shape (2,41,41,21)?
Best
batch size, height, width, number of classes. the input image is being downsampled throughout the network until approximately (1/8) of the original resolution.
cool, I see! thanks!
Hi, thanks for the code sharing. I am just going through the code and I am wondering why the raw_output raw_output = net.layers['fc1_voc12']
has shape (2,41,41,21)?
Best