This is caused by the unmatching of gpu device number when loading a saved model.
torch.load('my_file.pt', map_location=lambda storage, loc: storage)
ref:
http://blog.csdn.net/shincling/article/details/78919282
RuntimeError: cuda runtime error (10) : invalid device ordinal
免责声明！
本站转载的文章为个人学习借鉴使用，本站对版权不负任何法律责任。如果侵犯了您的隐私权益，请联系本站邮箱yoyou2525@163.com删除。
猜您在找
RuntimeError: CUDA error: invalid device ordinal
指定用第几块GPU、CUDA error: invalid device ordinal、
Check failed: error == cudaSuccess (10 vs. 0) invalid device ordinal
RuntimeError: cuda runtime error (59) : device-side assert triggered（已解决）
RuntimeError: cuda runtime error (30) : unknown errorr
RuntimeError: cuda runtime error (710) : device-side assert triggered at /pytorch/aten/src/THC/generic/THCTensorMath.cu:26
完美解决-RuntimeError: CUDA error: device-side assert triggered
RuntimeError: CUDA error: device-side assert triggered的解决
Cuda runtime error (999)
Pytorch: cuda runtime error (59) : device-side assert triggered at /pytorch/aten/src/THC/generic/THCTensorMa