zhouch

> 看你有跑通这篇代码，所以想请教你个问题！我是python2.7,pytorch1.0跑的segmentation的代码！ > > 看到有报这个错误，RuntimeError: cuda runtime error (59) : device-side assert triggered at ... > > 网上说是计算损失的时候label不是从0开始的，pytorch必须是（0，n-1）的label才不会报错，然后这篇论文在设置label的时候，类别个数设置为20，有一个类别作者说是背景。但是我现在想，我的这个报错会不会与作者设置的20个label有关！ > 请问你在跑这篇代码的时候有遇到这个问题吗？ > > 或者你知道这是哪里出错了吗？多有打扰，期待你的答复！！！ 你好，我对作者关于这两个数据"synthia", "city16"设置的16个类别这里有疑问， 0-15是15个实际的类别，255是背景，那在代码里args.n_class就等于16， 那么这句话： weight[args.n_class - 1]...

> For the hard bc you need to modify the output of the network, see 'apply_output_transform'. Briefly, imagine you have a domain (0,1)^2 and DirichletBC = 1.0 for u on...

@forxltk Thanks for your reply. I have used soft bcs, but the results seem very strange. I referred to this link [https://github.com/lululxvi/deepxde/issues/564](url) to simulate the 2D steady-state heat distribution. But...

@forxltk Thanks a lot. It works for me. In addition, when I simulate the time-dependent heat equations, I have a strange result. Is it due to the conflicting adjacent bc,...

@forxltk Thank you very much. The problem is what you said. Now I got the correct result. Many thanks!

@forxltk Sorry to bother you again. For the following initial and boundary conditions, the results seem strange, compared to the results of the finite difference method. import deepxde as dde...

@forxltk Thanks for your suggestions. I not only sample points in (x, 0.4), but also sample more points in t=0. But the result is not very good. Can you have...

@cjfuture Was the problem solved? Can you tell me how to solve it ?

@dutran I meet the problem: Multi-GPU execution not available - rebuild wiyh USE_NCCL ? How can I use multi-GPU ?