Blealtan/efficient-kan

Issues

• need long time to init efficient-KAN

  #53 opened 2 months ago by TuuSiwei
  1

• ModuleNotFoundError: No module named 'efficient_kan'

  #52 opened 3 months ago by pascal-maker
  2

• torch._C._LinAlgError: torch.linalg.lstsq: (Batch element 1): The least squares solution could not be computed because the input matrix does not have full rank (error code: 4).

  #50 opened 4 months ago by black-yt
  2

• Kolmogorov-Arnold Transformer

  #51 opened 4 months ago by Adamdad
  0

• flops and params

  #45 opened 7 months ago by songxinghui
  1

• Can still plot model and prune model in efficient kan?

  #29 opened 8 months ago by yhaddbb
  12

• Something different from the official results for KAN.

  #38 opened 8 months ago by ASCIIJK
  1

• Constant loss, network is not learning

  #48 opened 6 months ago by SuperKogito
  3

• Layer fails on inputs that have been transposed previously

  #36 opened 6 months ago by tRosenflanz
  1

• How to cite

  #47 opened 7 months ago by dgkoenigsberg
  1

• Questions

  #46 opened 7 months ago by jindeok
  0

• I would like to add a train test function to the KAN class

  #44 opened 7 months ago by riteshshergill
  0

• which types of images could be better recognised using KAN approach?

  #43 opened 7 months ago by matsevytyi
  0

• block: [0,0,0], thread: [1,0,0] Assertion `input_val >= zero && input_val <= one` failed. when using BCEloss

  #41 opened 7 months ago by Treakun
  0

• How to pip install?

  #40 opened 7 months ago by piwawa
  3

• Can configure efficient-kan model for continual learning?

  #28 opened 8 months ago by lukmanulhakeem97
  9

• ONNX

  #39 opened 8 months ago by alonejxy
  0

• Is there anyway that I can save the model for later use

  #17 opened 8 months ago by binhnguyen7tiki
  7

• Are the "efficient-kan" and "official-kan" equivalent in terms of algorithms?

  #18 opened 8 months ago by yuedajiong
  12

• What is the mathematic equation behind the model?

  #16 opened 8 months ago by hoangthangta
  4

• KAN intuition for setting the hidden layers

  #14 opened 8 months ago by marcopeix
  1

• Can MLP replace transformer?

  #13 opened 8 months ago by taotangSeek
  1

• KANs are MLPs? MLP Equivalent

  #8 opened 8 months ago by hypnopump
  2

• Train Multi-KANs simultaneously! Coding realization

  #30 opened 8 months ago by dwtum
  2

• Symbolic Formulation from efficient-KAN

  #27 opened 8 months ago by MinSiThu
  1

• AssertionError

  #26 opened 8 months ago by leedrake5
  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 opened 8 months ago by maomaocun
  1

• CUDA out of memory.

  #24 opened 8 months ago by ybu-lxd
  2

• Difference in memory usage of mlp and kan

  #23 opened 8 months ago by c-pupil
  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 opened 8 months ago by boostingboosting
  1

• Symbolic expression support

  #21 opened 8 months ago by maramos874
  1

• KAN has more learnable parameters?

  #15 opened 8 months ago by ningwanyi
  2

• Can anyone put kan(attention and mlp) into llama2.c?

  #9 opened 8 months ago by win10ogod
  1

• time complexity

  #7 opened 8 months ago by Aniosss
  2

• Contributing?

  #5 opened 8 months ago by dschonholtz
  3

• RuntimeError: false INTERNAL ASSERT FAILED

  #4 opened 8 months ago by csinva
  3

• Missing license

  #2 opened 8 months ago by simenkva
  2

• Intel oneMKL ERROR: Parameter 6 was incorrect on entry to SGELSY.

  #20 opened 8 months ago by wza13
  7

• Supports inputs more than 2 dimensions For efficient-KAN

  #19 opened 8 months ago by WhatMelonGua
  5

• Necessity of the scale parameter?

  #1 opened 8 months ago by KindXiaoming
  2