Giters

JDAI-CV / FADA

(ECCV 2020) Classes Matter: A Fine-grained Adversarial Approach to Cross-domain Semantic Segmentation

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

why batch_size // 2 ?

desti-nation opened this issue · comments

commented

cfg.SOLVER.BATCH_SIZE//2

commented

Hi, it is because each batch contains the same number of images from source domain and target domain. Thus, each dataloader will load half of the total batch size images when training is performed.