jpsml
commented
2 years ago I was using 7 views, I have changed to use only 4 views and the error was gone.
Closed jpsml closed 2 years ago
jpsml
commented
2 years ago I was using 7 views, I have changed to use only 4 views and the error was gone.
Hello, I have tried to run your code using a RTX 2080 Ti GPU, but got a CUDA out of memory error, as shown below:
model: model loss: pairwise triplewise lr: 1e-05 network: resnet 1%|█▊ | 2/180 [00:05<07:57, 2.68s/it] Traceback (most recent call last): File "train.py", line 66, in
epoch_loss = train(epoch_i)
File "train.py", line 22, in train
feature = model(img.squeeze(0).cuda())
File "/home/voxarlabs/anaconda3/envs/MVMHAT/lib/python3.6/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, kwargs)
File "/home/voxarlabs/anaconda3/envs/MVMHAT/lib/python3.6/site-packages/torchvision/models/resnet.py", line 220, in forward
return self._forward_impl(x)
File "/home/voxarlabs/anaconda3/envs/MVMHAT/lib/python3.6/site-packages/torchvision/models/resnet.py", line 210, in _forward_impl
x = self.layer3(x)
File "/home/voxarlabs/anaconda3/envs/MVMHAT/lib/python3.6/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, *kwargs)
File "/home/voxarlabs/anaconda3/envs/MVMHAT/lib/python3.6/site-packages/torch/nn/modules/container.py", line 117, in forward
input = module(input)
File "/home/voxarlabs/anaconda3/envs/MVMHAT/lib/python3.6/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(input, kwargs)
File "/home/voxarlabs/anaconda3/envs/MVMHAT/lib/python3.6/site-packages/torchvision/models/resnet.py", line 116, in forward
identity = self.downsample(x)
File "/home/voxarlabs/anaconda3/envs/MVMHAT/lib/python3.6/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, *kwargs)
File "/home/voxarlabs/anaconda3/envs/MVMHAT/lib/python3.6/site-packages/torch/nn/modules/container.py", line 117, in forward
input = module(input)
File "/home/voxarlabs/anaconda3/envs/MVMHAT/lib/python3.6/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(input, **kwargs)
File "/home/voxarlabs/anaconda3/envs/MVMHAT/lib/python3.6/site-packages/torch/nn/modules/batchnorm.py", line 136, in forward
self.weight, self.bias, bn_training, exponential_average_factor, self.eps)
File "/home/voxarlabs/anaconda3/envs/MVMHAT/lib/python3.6/site-packages/torch/nn/functional.py", line 2058, in batch_norm
training, momentum, eps, torch.backends.cudnn.enabled
RuntimeError: CUDA out of memory. Tried to allocate 20.00 MiB (GPU 0; 10.76 GiB total capacity; 9.11 GiB already allocated; 62.00 MiB free; 9.18 GiB reserved in total by PyTorch)
I am using Ubuntu 18.04. Do you have any suggestion for solving this problem? Best regards.