RuntimeError: CUDA error: unknown error

[SonwYang]

The details of error are as follows:
RUNDIR: runs\seg_pascal\59030 Traceback (most recent call last): File "train.py", line 226, in <module> train(cfg, writer, logger) File "train.py", line 72, in train model = get_model(cfg["model"], n_classes).to(device) File "C:\Users\yp\Anaconda3\lib\site-packages\torch\nn\modules\module.py", line 386, in to return self._apply(convert) File "C:\Users\yp\Anaconda3\lib\site-packages\torch\nn\modules\module.py", line 193, in _apply module._apply(fn) File "C:\Users\yp\Anaconda3\lib\site-packages\torch\nn\modules\module.py", line 193, in _apply module._apply(fn) File "C:\Users\yp\Anaconda3\lib\site-packages\torch\nn\modules\module.py", line 199, in _apply param.data = fn(param.data) File "C:\Users\yp\Anaconda3\lib\site-packages\torch\nn\modules\module.py", line 384, in convert return t.to(device, dtype if t.is_floating_point() else None, non_blocking) File "C:\Users\yp\Anaconda3\lib\site-packages\torch\cuda\__init__.py", line 163, in _lazy_init torch._C._cuda_init() RuntimeError: CUDA error: unknown error
My config is:
model: arch: fcn8s data: dataset: pascal train_split: train val_split: val img_rows: 256 img_cols: 256 path: ./datasets/ training: train_iters: 1000 batch_size: 1 val_interval: 10 n_workers: 4 print_interval: 1 optimizer: name: 'sgd' lr: 1.0e-10 weight_decay: 0.0005 momentum: 0.99 loss: name: 'cross_entropy' size_average: True lr_schedule: name: 'multi_step' milestones: [300,600] gamma: 0.1 resume: null
How can I solve this problem? Thank you

[SonwYang]

I can run this code by cpu, but it reports mistakes when I choose gpu. I think my cuda is normal because I have do some regular methods to verify it and it works.

[Aurevious]

I can run this code by cpu, but it reports mistakes when I choose gpu. I think my cuda is normal because I have do some regular methods to verify it and it works.

Hi,have you solve this problem?thank you!