Blealtan efficient-kan issues

Blealtan / efficient-kan

An efficient pure-PyTorch implementation of Kolmogorov-Arnold Network (KAN).

MIT License

3.46k stars 303 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Constant loss, network is not learning

#48 SuperKogito opened 1 week ago
2
How to cite

#47 dgkoenigsberg opened 1 week ago
1
Questions

#46 jindeok opened 1 week ago
0
flops and params

#45 songxinghui opened 3 weeks ago
0
I would like to add a train test function to the KAN class

#44 riteshshergill opened 3 weeks ago
0
which types of images could be better recognised using KAN approach?

#43 matsevytyi opened 4 weeks ago
0
refactor: utilize Einops in appropriate areas to enhance readability

#42 juvi21 opened 4 weeks ago
0
block: [0,0,0], thread: [1,0,0] Assertion `input_val >= zero && input_val <= one` failed. when using BCEloss

#41 Treakun opened 1 month ago
0
How to pip install?

#40 piwawa closed 1 month ago
3
ONNX

#39 alonejxy opened 1 month ago
0
Something different from the official results for KAN.

#38 ASCIIJK opened 1 month ago
0
Fix issue with view on previously transformed data

#37 akaashdash opened 1 month ago
0
Layer fails on inputs that have been transposed previously

#36 tRosenflanz opened 1 month ago
1
Train Multi-KANs simultaneously! Coding realization

#30 dwtum closed 1 month ago
2
Can still plot model and prune model in efficient kan?

#29 yhaddbb opened 1 month ago
11
Can configure efficient-kan model for continual learning?

#28 lukmanulhakeem97 opened 1 month ago
9
Symbolic Formulation from efficient-KAN

#27 MinSiThu closed 1 month ago
1
AssertionError

#26 leedrake5 opened 1 month ago
1
please report a bug to PyTorch. torch.linalg.lstsq: (Batch element 0): Argument 6 has illegal value. Most certainly there is a bug in the implementation calling the backend library.

#25 maomaocun closed 1 month ago
1
CUDA out of memory.

#24 ybu-lxd closed 1 month ago
2
Difference in memory usage of mlp and kan

#23 c-pupil opened 1 month ago
1
RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [800, 256]], which is output 0 of AsStridedBackward0, is at version 2; expected version 1 instead. Hint: enable anomaly detection to find the operation that failed to compute its gradient, with torch.autograd.set_detect_anomaly(True).

#22 boostingboosting opened 1 month ago
1
Symbolic expression support

#21 maramos874 closed 1 month ago
1
Intel oneMKL ERROR: Parameter 6 was incorrect on entry to SGELSY.

#20 wza13 opened 1 month ago
7
Supports inputs more than 2 dimensions For efficient-KAN

#19 WhatMelonGua closed 1 month ago
5
Are the "efficient-kan" and "official-kan" equivalent in terms of algorithms?

#18 yuedajiong closed 1 month ago
12
Is there anyway that I can save the model for later use

#17 binhnguyen7tiki closed 1 month ago
7
What is the mathematic equation behind the model?

#16 hoangthangta closed 1 month ago
4
KAN has more learnable parameters?

#15 ningwanyi closed 1 month ago
2
KAN intuition for setting the hidden layers

#14 marcopeix closed 1 month ago
1
Can MLP replace transformer?

#13 taotangSeek closed 1 month ago
1
Update test_simple_math.py

#12 ODamilare opened 1 month ago
0
WIP: 2d convolution with KANs

#11 kabachuha closed 1 month ago
3
fix bug: false INTERNAL ASSERT FAILED

#10 AmbroseX opened 1 month ago
0
Can anyone put kan(attention and mlp) into llama2.c?

#9 win10ogod closed 1 month ago
1
KANs are MLPs? MLP Equivalent

#8 hypnopump closed 1 month ago
2
time complexity

#7 Aniosss closed 1 month ago
2
Allow interchangeability of KANLinear with nn.Linear

#6 akaashdash closed 1 month ago
1
Contributing?

#5 dschonholtz closed 1 month ago
3
RuntimeError: false INTERNAL ASSERT FAILED

#4 csinva opened 1 month ago
3
Update README.md

#3 eltociear closed 1 month ago
3
Missing license

#2 simenkva closed 1 month ago
2
Necessity of the scale parameter?

#1 KindXiaoming closed 1 month ago
2