* Support color correction for img2img and inpainting, avoiding the shift to magenta seen when running images through img2img repeatedly.
* Fix docs for color correction
* add --init_color to prompt reconstruction
* For best results, the --init_color option should point to the *very first* image used in the sequence of img2img operations. Otherwise color correction will skew towards cyan.
Co-authored-by: Lincoln Stein <lincoln.stein@gmail.com>
- use git-revision-date-localized with enabled creation date
- update requirements-mkdocs.txt and pin verisons
- add requirements
- add dev addr
- fix template
- use better icons for repo and edit button
- remove odd extension
disabled toc in order to view those large tables
added linebreaks in long cells to stop multiline arguments/shortcut
added backticks arround arguments to stop interpreting `<...>` as html
added missing identifiers to codeblocks
changed html tags to markdown to insert the png
Fixes:
File "stable-diffusion/ldm/modules/diffusionmodules/model.py", line 37, in nonlinearity
return x*torch.sigmoid(x)
RuntimeError: CUDA out of memory. Tried to allocate 1.56 GiB [..]
Now up to 1536x1280 is possible on 8GB VRAM.
Also remove unused SiLU class.
* Added linux to the workflows
- rename workflow files
Signed-off-by: Ben Alkov <ben.alkov@gmail.com>
* fixes: run on merge to 'main', 'dev';
- reduce dev merge test cases to 1 (1 takes 11 minutes 😯)
- fix model cache name
Signed-off-by: Ben Alkov <ben.alkov@gmail.com>
* add test prompts to workflows
Signed-off-by: Ben Alkov <ben.alkov@gmail.com>
Signed-off-by: Ben Alkov <ben.alkov@gmail.com>
Co-authored-by: James Reynolds <magnsuviri@me.com>
Apply ~6% speedup by moving * self.scale to earlier on a smaller tensor.
When we have enough VRAM don't make a useless zeros tensor.
Switch between cuda/mps/cpu based on q.device.type to allow cleaner per architecture future optimizations.
For cuda and cpu keep VRAM usage and faster slicing consistent.
For cpu use smaller slices. Tested ~20% faster on i7, 9.8 to 7.7 s/it.
Fix = typo to self.mem_total >= 8 in einsum_op_mps_v2 as per #582 discussion.
- fixes no closing quote in pretty-printed dream_prompt string
- removes unecessary -f switch when txt2img used
In addition, this commit does an experimental commenting-out of the
random.seed() call in the variation-generating part of ldm.dream.generator.base.
This fixes the problem of two calls that use the same seed and -v0.1
generating different images (#641). However, it does not fix the issue
of two images generated using the same seed and -VXXXXXX being
different.
