deep learning
文章平均质量分 57
_Little_Coder_
学习中,修行中,励志中。
展开
-
用pytorch for C++ 编译的问题
报错:Cannot initialize CUDA without ATen_cuda library. PyTorch splits its backend into two shared libraries: a CPU library and a CUDA library; this error has occurred because you are trying to use some CUDA functionality, but the CUDA library has not been lo原创 2021-06-08 20:06:14 · 753 阅读 · 0 评论 -
【pytorch】one of the variables needed for gradient computation has been modified by......
运行 loss.backward() 的时候报错RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation…经过网上提示,发现没有用。于是自己找了原因。慢慢发现是自己的每次网络输入出的问题。源代码 ite[j, n - 2 : n] = temp # my_step=torch.randn(1,1,n).to(device)原创 2020-12-23 09:32:28 · 463 阅读 · 0 评论