Skip to content

Releases: rlaphoenix/VSGAN

v1.0.6

23 Jan 11:38
@rlaphoenix rlaphoenix
v1.0.6
a339d98
Compare
Choose a tag to compare
v1.0.6

Added

  • Detect ESRGAN old/new arch models via archaic trial-and-error.
Assets 2
Loading

v1.0.5

23 Jan 11:38
@rlaphoenix rlaphoenix
v1.0.5
b44c0ea
Compare
Choose a tag to compare
v1.0.5

Changed

  • Reworked code from Functional to Object-oriented Programming.
  • Improve code readability, project starting to get serious.
Assets 2
Loading

v1.0.4

23 Jan 11:38
@rlaphoenix rlaphoenix
v1.0.4
3ff4c9b
Compare
Choose a tag to compare
v1.0.4

Added

  • Add ability to tile the input to reduce VRAM (does not hide seams).
Assets 2
Loading

v1.0.3

23 Jan 11:37
@rlaphoenix rlaphoenix
v1.0.3
46779ae
Compare
Choose a tag to compare
v1.0.3

Added

  • VapourSynth to requirements.

Changed

  • Convert back to original color-space after applying the model.
Assets 2
Loading

v1.0.2

23 Jan 11:37
@rlaphoenix rlaphoenix
v1.0.2
46998a2
Compare
Choose a tag to compare
v1.0.2

Added

  • Ability to select device via argument.
Assets 2
Loading

v1.0.1

23 Jan 11:37
@rlaphoenix rlaphoenix
v1.0.1
1b00746
Compare
Choose a tag to compare
v1.0.1

Added

  • README file with some basic information.

Changed

  • Improved RGB conversion by using mvsfunc instead of core.resize.Point.
Assets 2
Loading

v1.0.0

23 Jan 11:36
@rlaphoenix rlaphoenix
v1.0.0
b9521d1
Compare
Choose a tag to compare
v1.0.0

Initial Release.

Assets 2
Loading

v1.4.0

13 Dec 23:50
@github-actions github-actions
v1.4.0
3a894f3
Compare
Choose a tag to compare
v1.4.0

Added

  • Added support for all RGB formats including float.

Changed

  • Heavily improved main model execution code.
  • Replace current chunk system with a seamless chunk system using overlap.
  • Add self-chaining system, calls can be made directly after another.
  • Made torch dependency optional and pointed directly to torch+cuda.
    This is due to conflicting kinds of torch installation methods.

Removed

  • Remove JetBrains .idea folder, added to gitignore.

Fixed

  • Only transpose C for RGB if it's 3-channels.
Assets 3
Loading

v1.3.1

25 Oct 20:21
@github-actions github-actions
v1.3.1
3b545bf
Compare
Choose a tag to compare
v1.3.1

Fixed

  • Fix type annotations on Python versions older than 3.9.
  • Use Python version 3.9.x for Dist workflow as 3.10 is not yet supported.
Assets 3
Loading

v1.3.0

07 Oct 02:44
@github-actions github-actions
v1.3.0
d700baa
Compare
Choose a tag to compare
v1.3.0

Added

  • Allow specification of the input array dimension order.
  • Add Jekyll Documentation in gh-pages branch.
  • Added a VSGAN Jupyter Notebook (Colab), with an Open in Colab Badge on the README.

Changed

  • Drop support for Python versions older than 3.6.2, due to bugs discovered in NumPy.
  • Replace setup.py/setuptools with Poetry.
  • Rename cv2_imread to frame_to_np, don't reverse to BGR as it's unnecessary.
  • More efficiently write an array to a VapourSynth VideoFrame.
  • Inherit output clip properties from input clip.
  • Moved README's information to the docs.
  • Reworked the CD GitHub Workflow to auto-create a GitHub Release and push to PyPI.

Removed

  • Remove the need for plane_count, now gets it from the input frame.
  • Don't define the transposes, it's unnecessary.

Fixed

  • Fixed a bug with frame plane access on VapourSynth API 4.
Assets 3
Loading