python - 运行时错误 : cuda runtime error (30) : unknown error at torch/csrc/cuda/Module. cpp:32

Question

using config:

       'CONFIG_NAME': 'stageI',
       'CUDA': True,
       'DATASET_NAME': 'coco',
       'DATA_DIR': '/home/anwar/Documents/StackGAN-Pytorch- 
        master/data/coco',
       'EMBEDDING_TYPE': 'cnn-rnn',
       'GAN': {'CONDITION_DIM': 128, 'DF_DIM': 96, 'GF_DIM': 192, 'R_NUM': 
        4},
       'GPU_ID': '1',
       'IMSIZE': 64,
       'NET_D': '',
       'NET_G': '',
       'STAGE': 1,
       'STAGE1_G': '',
       'TEXT': {'DIMENSION': 1024},
       'TRAIN': {'BATCH_SIZE': 128,
       'COEFF': {'KL': 2.0},
       'DISCRIMINATOR_LR': 0.0002,
       'FLAG': True,
       'GENERATOR_LR': 0.0002,
       'LR_DECAY_EPOCH': 20,
       'MAX_EPOCH': 120,
       'PRETRAINED_EPOCH': 600,
       'PRETRAINED_MODEL': '',
       'SNAPSHOT_INTERVAL': 10},
       'VIS_COUNT': 64,
       'WORKERS': 4,
       'Z_DIM': 100
Load filenames from: /home/anwar/Documents/StackGAN-Pytorch- 
master/data/coco/train/filenames.pickle (82783)
embeddings:  (82783, 5, 1024)

THCudaCheck FAIL file=torch/csrc/cuda/Module.cpp line=32 error=30 : unknown error Traceback (most recent call last): File "main.py", line 74, in algo = GANTrainer(output_dir) File "/home/anwar/Documents/StackGAN-Pytorch-master/code/trainer.py", line 45, in init torch.cuda.set_device(self.gpus[0]) File "/home/anwar/.conda/envs/tensorflow/lib/python2.7/site-packages/torch/cuda/init.py", line 262, in set_device torch._C._cuda_setDevice(device)

RuntimeError: cuda runtime error (30) : unknown error at torch/csrc/cuda/Module.cpp:32

Answer 1

可能你没有cudnn，记得安装正确的版本! 其次，CUDA的版本可能与pytorch所需的版本不同。 我遇到了同样的错误，现在一切正常了。 祝你好运!