search

KindXiaoming / pykan

Kolmogorov Arnold Networks
MIT License
14.76k stars 1.35k forks source link

关于symbolic regression报错的问题 #287

Open GuoMinghui07 opened 3 months ago

GuoMinghui07 commented 3 months ago

对于tutorials中的example 4的symbolic regression问题，我尝试将原本的表达式： f = lambda x: torch.exp(torch.sin(torch.pi*x[:,[0]]) + x[:,[1]]**2)中的x2前面的系数变成3，即：f = lambda x: torch.exp(torch.sin(torch.pi*x[:,[0]]) + 3*x[:,[1]]**2) 以下文件为训练过程和结果，在最后model.auto_symbolic(lib=lib)后便报错，不知道是什么问题，请大神指教。 报错.pdf

截屏2024-06-26 21 32 29
HUSTlwr commented 3 months ago

I have the same question.

shuaizengMU commented 2 months ago

I got a very similar error on a classification task. I fixed it by removing a feature that contained the same values for all samples.

menggedu commented 2 months ago

I got a very similar error on a classification task. I fixed it by removing a feature that contained the same values for all samples.

could you explain how you fix it in details