Stable WarpFusion v0.5.12 (Patreon)

Rushing towards a stable stable (sic!) version, all thank to your support and constant feedback, guys!

Here's this version's changelog:

News stuff:
- added vram usage fix, now supports up to 1536x1536 images on 16gig gpus
- added frame range (start-end frame) for video inits
- added pseudo-inpainting by diffusing only inconsistent areas

Fixes:
- fixed changing width height not working correctly
- removed LAMA inpainting to reduce load and installation bugs
- hiden intermediate saves (unusable for now)
- fixed multiple image operations being applied during intermediate previews (even though the previews were not shown)
- moved Stable model loading to a later stage to allow processings optical flow for larger frame sizes
- fixed videoframes being saved correctly without google drive\locally
- fixed PIL module error for colab to work without restarting

Upd:
- fix RAFT models download error x2

By popular demand, here's a shor guide on running locally: 

To make it work you'll need to install conda:

https://conda.io/projects/conda/en/latest/user-guide/install/index.html

then follow these steps:

conda create -name env_stable -python=3.7

conda activate env_stable

conda install ipykernel

python -m ipykernel install --user

conda install pytorch==1.11.0 torchvision==0.12.0 torchaudio==0.11.0 cudatoolkit=11.3 -c pytorch

cudatoolkit versions might differ depending on your GPU, but this should work in most cases.

Then manually download ffmpeg from here https://www.gyan.dev/ffmpeg/builds/ffmpeg-git-essentials.7z

unzip and put ffmpeg.exe in the root dir where your ipynb file is.

I'd suggest installing VSCode to run the notebook if you are not familiar with jupyter.