运行时出现下面的错误:
C:/cb/pytorch_1000000000000/work/aten/src/ATen/native/cuda/IndexKernel.cu:142: block: [0,0,0], thread: [17,0,0] Assertion `index >= -sizes[i] && index < sizes[i] && "index out of bounds"` failed. C:/cb/pytorch_1000000000000/work/aten/src/ATen/native/cuda/IndexKernel.cu:142: block: [0,0,0], thread: [61,0,0] Assertion `index >= -sizes[i] && index < sizes[i] && "index out of bounds"` failed. C:/cb/pytorch_1000000000000/work/aten/src/ATen/native/cuda/IndexKernel.cu:142: block: [0,0,0], thread: [62,0,0] Assertion `index >= -sizes[i] && index < sizes[i] && "index out of bounds"` failed. C:/cb/pytorch_1000000000000/work/aten/src/ATen/native/cuda/IndexKernel.cu:142: block: [0,0,0], thread: [63,0,0] Assertion `index >= -sizes[i] && index < sizes[i] && "index out of bounds"` failed. ... RuntimeError: CUDA error: device-side assert triggered
这是因为Tensor下标越界,造成CUDA断言错误,不让下标越界就正常了.
CUDA error: device-side assert triggered
原文：https://www.cnblogs.com/mstk/p/14724470.html