RuntimeError: CUDNN_STATUS_EXECUTION_FAILED about fewshot_detection

[whsun21]

same as me, and when I set torch.backends.cudnn.enabled = False in the front of train_meta.py, it return another error at the same place in the code, as follwing. Have you solved this error, if so, can you provide me some advises? Thanks a lot.

File "/home/sun/projects/Fewshot_Detection/darknet_meta.py", line 199, in forward
dynamic_weights = self.meta_forward(metax, mask)
File "/home/sun/projects/Fewshot_Detection/darknet_meta.py", line 122, in meta_forward
metax = model(metax)
File "/home/sun/anaconda3/envs/FR/lib/python2.7/site-packages/torch/nn/modules/module.py", line 357, in call
result = self.forward(*input, **kwargs)
File "/home/sun/anaconda3/envs/FR/lib/python2.7/site-packages/torch/nn/modules/container.py", line 67, in forward
input = module(input)
File "/home/sun/anaconda3/envs/FR/lib/python2.7/site-packages/torch/nn/modules/module.py", line 357, in call
result = self.forward(*input, **kwargs)
File "/home/sun/anaconda3/envs/FR/lib/python2.7/site-packages/torch/nn/modules/conv.py", line 282, in forward
self.padding, self.dilation, self.groups)
File "/home/sun/anaconda3/envs/FR/lib/python2.7/site-packages/torch/nn/functional.py", line 90, in conv2d
return f(input, weight, bias)
RuntimeError: cublas runtime error : the GPU program failed to execute at /opt/conda/conda-bld/pytorch_1518238441757/work/torch/lib/THC/THCBlas.cu:247

[linsongxue]

minimize the batch size of .cfg file