Ziming Liu

Thanks @Jim137 , is now good?

Thank you, gotcha! Have released 0.0.4 including new changes.

Thank you for the great insights! I really need to read this carefully this weekend, looking like treasure troves, but for now need to study for a deadline due Friday...

Hope this is relevant. Kolmogorov-Arnold Networks (KANs) for Time Series Analysis: https://arxiv.org/abs/2405.08790

This is not necessarily a bug, but I believe future variants will be better than this.

thanks for reporting this! do you have any idea why this is the case? (I can check it out later but I don't have a windows machine right now and...

Hi it seems like in your first input tensor, the first inputs are zero for all samples. This can lead to a singular problem. You could remove the first column...

Your drawing is consistent with my plot (up to node permutation and sign). Not sure about the training loss result. Note that you may need to remove different edges because...

Hi, the results of example 2 is produced before code release. After a few merge, it could be that random seeds have a slightly different effect. To avoid removing edges,...

From my experience, this example can be very seed dependent. Please try other seeding KAN(... , seed=42)