csubiogroup / deeplncloc Goto Github PK

Hi CSU BioGroup,
I have issues when using gpu to train this model. When I changed device to cpu, everything works fine. Could you tell me the version of pytorch, cuda and gensim you are currently using? I thought it may be because I use a newer version. Thank you!

Here are the error messages I got after I ran with command line:
python train.py --k 3 --d 64 --s 64 --f 128 --metrics MaF --device "cuda:0"

Traceback (most recent call last):
File "train.py", line 27, in
model.cv_train(dataClass, trainSize=1, batchSize=16, stopRounds=-1, earlyStop=10,
File "/home/shenyang/github/DeepLncLoc/model/DL_ClassifierModel.py", line 24, in cv_train
res = self.train(dataClass,trainSize,batchSize,epoch,stopRounds,earlyStop,saveRounds,optimType,lr,weightDecay,
File "/home/shenyang/github/DeepLncLoc/model/DL_ClassifierModel.py", line 50, in train
loss = self._train_step(X,Y, optimizer)
File "/home/shenyang/github/DeepLncLoc/model/DL_ClassifierModel.py", line 168, in _train_step
loss.backward()
File "/home/shenyang/anaconda3/envs/envdeeplog/lib/python3.8/site-packages/torch/tensor.py", line 221, in backward
torch.autograd.backward(self, gradient, retain_graph, create_graph)
File "/home/shenyang/anaconda3/envs/envdeeplog/lib/python3.8/site-packages/torch/autograd/init.py", line 130, in backward
Variable._execution_engine.run_backward(
RuntimeError: shmem_size <= sharedMemPerBlock INTERNAL ASSERT FAILED at "/opt/conda/conda-bld/pytorch_1607370131125/work/aten/src/ATen/native/cuda/AdaptiveAveragePooling.cu":669, pleas
e report a bug to PyTorch.