RuntimeError: Given groups=1, weight of size 48 160 1 1, expected input[2, 640, 3, 3] to have 160 channels, but got 640 channels instead

[mx1mx2]

when do i run train.py?

[adityalokesh]

mx1mx2 need help running this could you pls help me !
adityapesu@gmail.com

[ZhengPeng7]

Have you fixed this error?

[ZhengPeng7]

print(model) and search 160, you will find the only 160 is the input_channel of the decoder.

It's the low_level_inplanes controls the input_channel, which defined by filter_multiplier.
Then see the ln25 in build_autodeeplab.py, change the param filter_multiplier from args.filter_multiplier * args.block_multiplier to args.filter_multiplier * args.block_multiplier * filter_param_dict[network_path[2]], which is then adaptive.

Thanks to this issue #93.