Is upsampling loss considered before the concatenation in `ss_fusion_seg.py`?

WHU-Sigma / HyperSIGMA

The official repo for the paper "HyperSIGMA: Hyperspectral Intelligence Comprehension Foundation Model"

155 stars 16 forks source link

Open szubing opened 3 months ago

szubing commented 3 months ago

Sorry, I mean 'lose' no 'loss'

jinyaoWHU commented 3 months ago

@szubing Sorry, I don't understand the meaning of upsampling lose you mentioned. Can you explain what upsampling lose means?