pykan
pykan copied to clipboard
KindXiaoming
Kolmogorov Arnold Networks
Why do I get unexpected results when I run the code as per the example provided? 
Testing on a in-house dataset which we have in our research group as a benchmark. It has lots of inputs to one output but to keep it simple now we...
I don't know where to post this so I will post it here. I have run some benchmarking on the different repos that were created after this to see how...
class PhysicsInformedNN: # Initialize the class def __init__(self, x, y, u0_real, u0_imag, u0_real_xx, u0_real_yy, u0_imag_xx,u0_imag_yy, m, m0, eta, delta, layers, omega): X = np.concatenate([x, y], 1) self.lb = X.min(0) self.ub...
I am encountering a `RuntimeError` related to the `einsum()` function during my project where I'm integrating the KAN model with the MNIST dataset. The error message is as follows: Steps...
hello, this work is great, but I'm running into a little problem. when the function I want to fit is f(x,y)=x^-1+y^2 , I found that y^2 is easy to fit...
Upon attempting to run tutorials on the CPU device, I encountered the following error: `RuntimeError: Expected all tensors to be on the same device, but found at least two devices,...
Hello, first of all, congratulations for the great work! As far as you know, is it possible to extend the Kolmogorov-Arnold theorem to images? Something like _"If IM is a...
I'm trying to fit a vanilla Black-Scholes model as follows: ```python import numpy as np import pandas as pd import plotnine as pn import torch from kan import create_dataset device...
~/.local/lib/python3.10/site-packages/kan/KANLayer.py --> 126 self.scale_base = torch.nn.Parameter(torch.FloatTensor(scale_base).cuda()).requires_grad_(sb_trainable) when runing device='cpu' and without cuda card, this line is always executed
