generic-github-user/xkcd-Generator

Issues

• Change name from xkcd-Generator to xkcd-generator

  #26 opened 6 years ago by generic-github-user
  0

• Move trainableVars list outside of training loop

  #27 opened 6 years ago by generic-github-user
  0

• Replace high-level tf.model with low-level tensor operations

  #28 opened 6 years ago by generic-github-user
  0

• Neural networks not training

  #8 opened 6 years ago by generic-github-user
  8

• Generated image canvas briefly displays a training image before the training process begins

  #25 opened 6 years ago by generic-github-user
  0

• Calculate numLayers from imageSize

  #22 opened 6 years ago by generic-github-user
  0

• Generator output pixel values exploding

  #21 opened 6 years ago by generic-github-user
  0

• Write README.md

  #24 opened 6 years ago by generic-github-user
  0

• Increase number of random latent parameters

  #23 opened 6 years ago by generic-github-user
  1

• Reduce input/output image color channels from 3 (RGB) to 1 (B/W)

  #13 opened 6 years ago by generic-github-user
  0

• Remove old code comments

  #20 opened 6 years ago by generic-github-user
  0

• Both models may be training at once where only one is supposed to

  #19 opened 6 years ago by generic-github-user
  0

• Discriminator network not training

  #18 opened 6 years ago by generic-github-user
  3

• Add more console logs

  #5 opened 6 years ago by generic-github-user
  0

• Reduce number of discriminator model outputs to 1

  #16 opened 6 years ago by generic-github-user
  0

• Add comic title text

  #17 opened 6 years ago by generic-github-user
  0

• Output values of generator network appear to be limited to a very low range, causing a darkened image to be generated

  #15 opened 6 years ago by generic-github-user
  1

• Add image data buffer

  #9 opened 6 years ago by generic-github-user
  0

• Only regenerate training data every several iterations instead of every iteration

  #14 opened 6 years ago by generic-github-user
  0

• Only one training image is added to the trainingData object

  #12 opened 6 years ago by generic-github-user
  1

• Old training data not deleted from trainingData.pixels object after being regenerated each iteration

  #11 opened 6 years ago by generic-github-user
  0

• Memory leak adds 2 tensors to memory each iteration

  #10 opened 6 years ago by generic-github-user
  1

• Add training data regeneration

  #7 opened 6 years ago by generic-github-user
  0

• Undefined errors occur when not using tf.tidy() to define calculateLoss functions

  #6 opened 6 years ago by generic-github-user
  0

• Add probability distribution type option for random latent parameters

  #3 opened 6 years ago by generic-github-user
  0

• Add different optimizers for each neural network

  #4 opened 6 years ago by generic-github-user
  0

• Add thumbnail display canvases

  #2 opened 6 years ago by generic-github-user
  0

• Update code comments

  #1 opened 6 years ago by generic-github-user
  0