pytorch-semseg
pytorch-semseg copied to clipboard
meetps
RuntimeError: CUDA error: out of memory
When I try to train fcn8s_pascal.py, it always appears this issue. But if I train segnet_pacal.py, it works normally. Can anyone help me? I didn't change the .yml file provided by author.
Hi，Have you solved the problem？
Some others had also met the same mistake when they were trying to train fcn on pytorch. I found two solutions. The first one is to modify the input to 224*224 which is same with the original paper, and the training process works normally. However, I can't test any images because CUDA error appeared again. Here is the link of the second solution:https://github.com/zijundeng/pytorch-semantic-segmentation/issues/37, and I am still trying to change my code.
Hi，Have you solved the problem？
Some others had also met the same mistake when they were trying to train fcn on pytorch. I found two solutions. The first one is to modify the input to 224*224 which is same with the original paper, and the training process works normally. However, I can't test any images because CUDA error appeared again. Here is the link of the second solution:zijundeng/pytorch-semantic-segmentation#37, and I am still trying to change my code.
Have you solved this problem?
