Segmentation Size

[mavaylon1]

I was looking at "image_segmentation_generator" in the data_loader.py and printed out the shape of the individual output image and mask shapes. I got: img shape (416, 608, 3) seg_shape (252928, 51)

I was wondering if you could explain the seg shape. I know the seg_shape is [(wh),51], where 51 is the number of classes for the data you gave. I'm confused by why is it like that? for example say we had a 3 class scenario and the image also has 3 channels. Would it still be [(wh),3], where 3 is the classes/channels?