Possible bug in number of tensors for layer fusion

[TheOnlyError]

The figure below shows that the value of a decoder neuron is based on 5 other values:

When trying to reproduce this in the code, I get the following:

The number of tensors is equal to 4 instead of 5. This suggest there is a bug in the code around the following lines:

keras-unet-collection/keras_unet_collection/_model_unet_3plus_2d.py

Lines 142 to 150 in d30f14a

	for i in range(depth_decode):
	f = filter_num_skip[i]
	# collecting tensors for layer fusion
	X_fscale = []
	# for each upsampling level, loop over all available downsampling levels (similar to the unet++)
	for lev in range(depth_decode):

I think the second for loop should have depth_decode+1 as the upper boundary in the range. Is this indeed a bug?

[ziyangwang007]

I also noticed that the X1EN can not be copied and crop to the decoder part through multi-scale skip connections each time. I agree with you.