pykan
pykan copied to clipboard
KindXiaoming
Kolmogorov Arnold Networks
when I run the code for deep formular, I have the following issues: grids = [3,5,10,20,50] train_rmse = [] test_rmse = [] for i in range(len(grids)): #model = KAN(width=[4,9,1], grid=grids[i],...
Hello, and thank you for this amazing work! I believe that a valuable feature, particularly for interpretability and inverse problem cases, would be the ability to add “constants” as extra...
I wanted to know if it is possible to add a new functionality when generating a formula (auto_symbolic) with nan values so that all **nan** are replaced with the value...
Enhancement Request: Dynamic Spline Fixing for Critical Inputs During Iterative Training in KAN Models **Description:** The `pyKAN` framework is powerful for iterative training, pruning, and refinement of Kolmogorov-Arnold Networks (KANs)....
Coming from prior versions of PyKAN, I was expecting `model.prune()` to be in-place. I validated my assumption by looking into the code and seeing that `prune()` does return `self`, just...
In the current implementation, the constructor of `MultKAN` sets the seeds for the `torch`, `numpy` and `random` modules on a global level. This can lead to side effects when using...