error happened while trainning with multi-label(self-data)：

Alibaba-MIIL / ML_Decoder

Official PyTorch implementation of "ML-Decoder: Scalable and Versatile Classification Head" (2021)

MIT License

317 stars 53 forks source link

error happened while trainning with multi-label(self-data)： #9

Closed fourmi1995 closed 2 years ago

fourmi1995 commented 2 years ago

cmd: CUDA_VISIBLE_DEVICES=6 python train.py --data=/DATA2/disk1/fourmi/images/data/images/data/ --model-name=tresnet_l --image-size=448

File "/home/fourmi/.conda/envs/ASL/lib/python3.6/site-packages/inplaceabn/functions.py", line 85, in forward count = count.to(dtype=var.dtype) RuntimeError: CUDA error: invalid device function

it caused in here ：ML_Decoder/src_files/models/tresnet/tresnet.py line:210 def forward(self, x):
x = self.body(x) #the shape of x is right(b,48,112,112) print(x)

mrT23 commented 2 years ago

triple-check that your GPU works well. are you able to train with ResNet50 ?

also try to reinstall inplace-abn