pykan

pykan copied to clipboard

something wrong

2

ElevenMorning

Irreproducible Example_1_function_fitting.ipynb

1

Example_1_function_fitting.ipynb runs, but it's plots/results no longer match the saved ones on github. They also don't match the ones given in the paper. Results obtained by running Example_1_function_fitting.ipynb are very...

tyagi-ankit

Trying a coef2curve example

1

Hi. Thanks for making the KAN code available. I've been trying to understand how it all works, and was trying to run coef2curve to get an idea of how the...

drdozer

I use model.prune() to prune my KAN network. After that, I use model.auto_symbolic(lib=lib) to get a formula. The problem is that when I use lib = ['x'] to get a...

niuniu666-ignore

Tutorial hellokan doesn't work at `symbolic_formula`! 示例教程 hellokan的symbolic_formula有问题！

3

I am new to this repository, following readme.md, I am trying the hellokan.ipynb example. However, although the training and plotting seams successful, which correctly reproduce the result of the notebook...

2catycm

What is the exact KAN width of this model?

prokash122

What are the possible reasons for the big gap between the use of LBFGS and Adam optimizer?（什么可能的原因造成使用LBFGS和Adam优化器差距很大？）

2

# Data  The function is , $$ y=exp(0.1x_1+0.12x_2+exp(0.2x_3+0.5x_4)) $$ # KAN define    When I continue to train, there is still a big gap in the subsequent...

glotm

I used KAN to try the symbolic regression, then i got that:  This is very complex. Is there any way to reduce the complexity of the final symbolic formula?...

Dash-RSer

The implemented following code snippet for binary classification on tabular data, using stratified K-fold cross-validation (K=10). The performance results seem exceptionally good. Can someone help review and suggest improvements to...

SaranDS

Hi, I'm a bit confused about the L1 norm as defined in the paper vs. how it's implemented in the code. From the paper, L1 norm seems to be defined...

MasayukiNagai