zijundeng/pytorch-semantic-segmentation

ResNetDUC model can not converge in the dataset CamVid?

guanfuchen opened this issue · 2 comments

guanfuchen commented

image

First thanks for your implementing for ResNetDUC, I try to train on the dataset CamVid, but it is trained slow and seem not converge. What should I do to get a better model?

zijundeng commented

Do you check whether there is any bug, such as forgetting to call loss.backward() ?

guanfuchen commented

Sorry because it trains too slow, I thought it seems not converge.