Pytorch Implementation the paper Auto-DeepLab Hierarchical Neural Architecture Search for Semantic Image Segmentation
410
stars
97
forks
source link
Given groups=1, weight of size 48 160 1 1, expected input[2, 1280, 25, 25] to have 160 channels, but got 1280 channels instead #62
Open
mx1mx2 opened 4 years ago