search

Tramac / awesome-semantic-segmentation-pytorch

Semantic Segmentation on PyTorch (include FCN, PSPNet, Deeplabv3, Deeplabv3+, DANet, DenseASPP, BiSeNet, EncNet, DUNet, ICNet, ENet, OCNet, CCNet, PSANet, CGNet, ESPNet, LEDNet, DFANet)
Apache License 2.0
2.8k stars 581 forks source link

dfa_net #130

Open wshj2020 opened 4 years ago

wshj2020 commented 4 years ago

I have met a question, I used dfa to train my own dataset,and it proformed well training, but worse in evaluating. why? Then my input are negative, is this normal? thank you ~

wxk2008 commented 4 years ago

RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [4, 8, 256, 256]], which is output 0 of ReluBackward1, is at version 2; expected version 1 instead. Hint: enable anomaly detection to find the operation that failed to compute its gradient, with torch.autograd.set_detect_anomaly(True)

bbbts commented 2 weeks ago

I have met a question, I used dfa to train my own dataset,and it proformed well training, but worse in evaluating. why? Then my input are negative, is this normal? thank you ~

did you use xception backbone for dfnanet model? could you kindly provide the link to download it? (link to download the .pth file?)