yolov5-6.1 RuntimeError: Unable to find a valid cuDNN algorithm to run convolution

错误信息:
RuntimeError: CUDA out of memory. Tried to allocate 26.00 MiB (GPU 0; 2.00 GiB total capacity; 745.25 MiB already allocated; 0 bytes free; 788.00 MiB reserved in total by PyTorch) 

yolov5-6.1 RuntimeError: Unable to find a valid cuDNN algorithm to run convolution_第1张图片

原因:
运行时错误:CUDA显卡内存不足

解决方法:

 尝试减小输入的Batch-size,我多次尝试减少到了2

你可能感兴趣的:(python,深度学习,开发语言)