Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Nov 12, 2016

• Fix exception when training with JPEG compression.

  alexjc

  committed
  559e66b

• Fix histogram colors overflowing in certain images. Closes #73.

  alexjc

  committed
  9d2aa3c

• Fix memory error in Docker by changing default tile size. Closes #74.

  alexjc

  committed
  904071d

Commits on Nov 11, 2016

• Moved demo website to its own URL on nucl.ai rather than IP.

  alexjc

  committed
  9fae68b

• Fix filenames used on stdout to relative paths.

  alexjc

  committed
  4000bd3

• Fix default parameters to zoom=2 so it works without arguments. Closes #71.

  alexjc

  committed
  881e909

• Rename folder with training script so they are not used for enhancing by users accidentally.

  alexjc

  committed
  a22d7ce

• Fix bug in handling of --train-jpeg parameter.

  alexjc

  committed
  fabe04f

Commits on Nov 10, 2016

• Add training scripts, they take a while to run!

  alexjc

  committed
  7c8cc56

• Merge pull request alexjc#69 from alexjc/v0.3

  Show description for 9d8ab0e
  alexjc

  authored
  9d8ab0e

• Prepare build scripts for merge and update README.

  alexjc

  committed
  80d7b57

• Fix alias for bash and zsh, Closes #54. Add standalone alias-free docker example, Closes #46.

  alexjc

  committed
  b706599

• Fix for trained models loaded from script directory. Closes #67.

  alexjc

  committed
  c03a0b7

• Add histogram color matching, enabled with --rendering-histogram.

  alexjc

  committed
  a69d2ae

• Add new model selection (by image type and 'enhance' operation) and parameters and documentation in README.

  alexjc

  committed
  965bd55

Commits on Nov 9, 2016

• Minor rework of the training parameters.

  alexjc

  committed
  e612087

Commits on Nov 8, 2016

• Simplify architecture outside of convolution blocks for speed, improve training code.

  alexjc

  committed
  0d4621d

• Fix for padding in convolution.

  alexjc

  committed
  c8e8056

• Buffer chunks of images proportionally to size.

  alexjc

  committed
  2a29fc9

Commits on Nov 4, 2016

• Remove reflection layer, important but not the bottleneck right now and quite slower.

  alexjc

  committed
  da354f3

• Improve pre-processing steps to include more randomness: scales, blur...

  alexjc

  committed
  9bba985

Commits on Nov 3, 2016

• Update training parameters.

  alexjc

  committed
  149ae4e

• Fix for images that are not multiples of 2 or 4, depending on the rescaling.

  alexjc

  committed
  de047d6

• Merged documentation changes.

  alexjc

  committed
  29bfe58

• Merge pull request alexjc#44 from wieczorek1990/master

  Show description for 9a228c1
  alexjc

  authored
  9a228c1

• Merge pull request alexjc#47 from alexjc/v0.2

  Show description for b24fa74
  alexjc

  authored
  b24fa74

• Fix formatting, minor tweaks to Docker build files for release.

  alexjc

  committed
  03914db

• Fix progress output when tiled rendering, changed default zoom level.

  alexjc

  committed
  448e7b9

• Add training scripts for networks currently being trained, for release v0.2.

  alexjc

  committed
  cabaaee

• Merge branch 'master' into v0.2

  alexjc

  committed
  1ad40b6

• Add tiled rendering with padding, no feather-blending but looks good enough.

  alexjc

  committed
  ac49676

• Fixed typo

  

  Łukasz Wieczorek

  committed
  ed20dce

• Add tiled rendering, currently with no padding for each tile.

  alexjc

  committed
  095fe42

Commits on Nov 2, 2016

• Integrated reflection padding instead of zero padding for extra quality during training and inference.

  alexjc

  committed
  d18c08f

• Fix padding code, more reliable for specific upscale/downscale combinations.

  alexjc

  committed
  90c0b7e

Pagination

PreviousNext