pykan

pykan copied to clipboard

https://github.com/akaashdash/kansformers

johndpope

FYI: convolutional layer with KAN

17

[https://github.com/StarostinV/convkan](https://github.com/StarostinV/convkan) A basic yet efficient (thanks to [efficient-kan](https://github.com/Blealtan/efficient-kan/)) implementation of the convolutional operation with KAN implemented using [F.unfold](https://pytorch.org/docs/stable/generated/torch.nn.functional.unfold.html).

StarostinV

running example 10 with GPU, error: all tensors on two devices, cuda:0 and cpu!

6

I met the issue that for GPU version the tensors are different devices, cuda:0 and cpu! The detail was shown below. description: 0%| | 0/20 [00:00

AlpsCV

Running Example 2: Deep Formulas on my computer yielded inconsistent results

15

This is the original drawing  This is my drawing  As you can see, the spline of my graph and the result of the original graph seem to be...

GaoLei0

Cut off all the branches

5

When I first demonstrated using Kan for a demo, the actual execution of the code was not the same as in the sample. I found that when constructing nodes, some...

plumv

Odd behavior of training process in almost-machine-precision function fitting.

2

https://gist.github.com/AlessandroFlati/a4f5d23864bca30f2b96c1052efdc140

AlessandroFlati

I know this could look unfeasible on a first look, but do we have some way to prevent the B-Splines to generate "biased" form? Let's say that ```final_model.symbolic_formula(simplify=False)[0][0]``` reports ```...

AlessandroFlati

KAN.initialize_from_another_model() error:Expected input and other to have the same dtype, but got input's dtype Float and other's dtype Double

5

input is tensor([[ 0.0000, -0.2120], [ 0.0000, -0.0247], [ 0.0000, 0.2150], ..., [ 0.0000, 0.7221], [ 0.0000, -0.6781], [ 0.0000, 0.3832]], dtype=torch.float64); model = KAN(width=[2,3,1], grid=3, k=3) train loss: 5.90e-01...

wwz1126

Some Issues in Constructing Physics-Informed Neural Networks Based on KANs

1

1. Nodes cannot directly fit multiplication and division operations; nodes can only use addition and subtraction. 2. When using the model, it is not possible to restrict the activation functions...

hzy24

Classification problems of KAN

2

Notice that you present the classification problem in example3, but treat the problem as a regression problem. My question is, is there any way KAN can limit the output to...

Rome1453