Home Creators Posts Import Register Favorites Logout
haven't archived this post yet. have a subscription? use the importer!

Downloads

  • stable_warpfusion_v0_32_0_speedup_fixes.ipynb

Missing 1 file, as this post has not yet been imported.

Content

Local install guides:
Windows: https://github.com/Sxela/WarpFusion#local-installation-guide-for-windows-venv
Linux:
https://github.com/Sxela/WarpFusion/blob/main/README.md#local-installation-guide-for-linux-ubuntu-2204-venv

If something's not working, make a new env and grab a fresh install.bat from that repo first, ask questions later :D

Changelog:

  • add optimization fixes by #infinitydeltax from discord, thanks to #romustach for pushing

  • add offload_model to loaded settings

  • remove duplicate temporal controlnet

  • move optical flow definitions to the "Import dependencies, define functions" cell

  • move optical flow default settings to Render settings -> non-gui -> advanced (non-gui settings are used to initialize the GUI)

  • remove reference controlnet cell leftovers

  • remove clip guidance cell (as deprecated)

  • hide use torch v2 and force torch reinstall switches from the "Prepare folders & Install" cell as deprecated

  • move tiled vae to GUI (kudos to foxpara)

  • remove tiled vae cell

  • check for existing masks when extracting background mask

  • add force overwrite checkbox for background mask extraction

  • add mask extraction to GUI

  • remove mask extraction cell

  • move logs to a logs subfolder

  • add auto brightness correction to GUI

  • remove auto brightness correction cell

  • move color_video_source to GUI colormatch tab

  • add include, python.zip, libs to github (required for some local python package build)

  • add include, libs to install.bat

  • fix gui settings not being saved before the initial run

  • add torch and xformers downgrade to colab installer

  • turn SDXL VAE black image exception into a warning

  • fix running facefusion cell breaking the main notebook

  • add insightface install for ipadater faceid

  • add raw frame to preview images

  • fix tiled vae not working with SDXL

  • add freeu setting to GUI

11.4.2024:

  • move to L

Speedup fixes

Render settings -> Non-Gui -> Advanced

Please note that most of the fixes increase VRAM usage. So if you are running out of ram, switch some of them. 

These values were used in previous notebooks:

These are applied each frame: 

gc_collect_offload = True #garbage collect during model offload (slower)

cuda_empty_cache = True #empty cuda cache (slower)

gc_collect = True #garbage collect (slower)

do_run_cast = 'cpu' #['cuda', 'cpu', 'off'] #move models to gpu at the beginning of every frame, faster / needs more vram. cpu - slower, less vram. off - do nothing

save_img_format = 'png' #['png', 'jpg', 'tiff'] #image save format, jpg is faster, png is slower, tiff is the slowest. 

These are applied every diffusion step (so the slowdown may happen each step), so enable those the last:

sample_gc_collect = True

sample_cuda_empty_cache = True

GUI updates

Tiled vae moved to misc -> tiled vae

Mask source and extraction moved to gui mask tab

FreeU settings moved to mics -> freeu

Color video source moved to colormatch tab -> color_video_source


Local install guide:
https://github.com/Sxela/WarpFusion/blob/main/README.md

Guides made by users:

YouTube playlist with settings:
https://www.youtube.com/watch?v=wvvcWm4Snmc&list=PL2cEnissQhlCUgjnGrdvYMwUaDkGemLGq

For tech support and other questions please join our discord server:
https://discord.gg/58Rd73uAe7

Discord is the preferred method because it is nearly impossible to provide any decent help or tech support via Patreon due to its limited text formatting and inability to add screenshots or videos to comments or DMs.
Due to the recent Patreon comments update it's impossible to reply to comments from notifications anymore, so if your comment hasn't been replied to for a while, DM me.



Comments

AI Mastery

what's the latest version of wrap fusion that i need to download for Google colab?