koshian2/OctConv-TFKeras

Issues

• ValueError: tf.function-decorated function tried to create variables on non-first call. While running on Tensorflow 2.0 and above

  #16 opened 2 years ago by rytisss
  1

• Issue in loading model network

  #17 opened 2 years ago by ritikajha
  0

• Octave Conv1D Layer Request

  #10 opened 3 years ago by praneet195
  1

• I want to insert OctConv into my network, is it just to replace Conv2d with OctConv2D?

  #15 opened 4 years ago by woodcss
  0

• Question about models.py and the result

  #13 opened 5 years ago by supermark1995
  2

• I found that the initial parameters(kernel_initializer and kernel_regularizer) are more import than octave

  #11 opened 5 years ago by gogo03
  6

• Is there a limit on input_shape when using OctConv?

  #12 opened 5 years ago by a5372935
  1

• Octave Wide-ResNet implementation not according to paper

  #7 opened 5 years ago by miguelvr
  8

• About GPU_Util

  #9 opened 5 years ago by a5372935
  2

• How can I run it on GPU mode?

  #6 opened 5 years ago by BlueAnthony
  1

• Why Octave Convolution is slower than vanilla?

  #8 opened 5 years ago by BlueAnthony
  2

• 怎么插入到resnet中，

  #1 opened 5 years ago by jiafengshen
  7

• wide-resnet network structure is not standard or exits errors.

  #4 opened 5 years ago by gogo03
  4

• How to divide low frequency and high frequency？

  #3 opened 5 years ago by ys-dpc
  1

• Can I access your colab notebook?

  #2 opened 5 years ago by ryoppippi
  2