Bryce
69af07ab67
style: fix lint issues
2 years ago
Bryce
38c7f88950
feature: boolean logic masks
...
Specify advanced text based masks using boolean logic and strength modifiers. Mask descriptions must be lowercase. Keywords uppercase.
Valid symbols: `AND`, `OR`, `NOT`, `()`, and mask strength modifier `{*1.5}` where `+` can be any of `+ - * /`. Single-character boolean
operators also work. When writing strength modifies know that pixel values are between 0 and 1.
- feature: apply mask edits to original files
- feature: auto-rotate images if exif data specifies to do so
- fix: accept mask images in command line
2 years ago
Bryce Drennan
d090f9d072
fix: unneeded argument
2 years ago
Bryce Drennan
eb54b9ca7f
Merge pull request #22 from brycedrennan/pylama_lint_updates
...
refactor: implements changes to comply with pylama
2 years ago
Bryce
a3a0de08e9
autoformat
2 years ago
Christian Käser
9fdd38fc11
Fix loading mask image from command line
2 years ago
Jay Drennan
df28bf8805
refactor: implements changes to comply with pylama
2 years ago
Bryce
cc014685bd
fix: leave sample in float32 for all final processing
2 years ago
Bryce
9614a82f12
fix: use float32 at image render time for wider cpu compatibility
...
If the x_sample was a bfloat on the gpu but the cpu doesn't support bfloat, that can cause a TypeError
```
File "/home/stdiff/.local/lib/python3.10/site-packages/imaginairy/api.py", line 292, in imagine
x_sample.cpu().numpy(), "c h w -> h w c"
TypeError: Got unsupported ScalarType BFloat16`
```
2 years ago
Bryce
cdfeaa4c6f
tests: add docker image for testing environment. minor test improvements
2 years ago
Bryce
09bc1c70e6
fix: try to address #13 BFloat16 issue
...
Seems to be caused by incompatible types in group_norm when we use autocast.
Patch group_norm to cast the weights to the same type as the inputs
From what I can understand all the other repos just switch to full precision instead
of addressing this. I think this would make things slower but I'm not sure. So maybe
the patching solution I'm doing is better?
https://github.com/pytorch/pytorch/pull/81852
2 years ago
Bryce
e23e363bf5
lint
2 years ago
Bryce
ff7fc0dab7
feature: tile mode can now be specified per-prompt
2 years ago
Bryce
6f77c23c1d
version 1.5.4
...
minor changes to help with google colab compat
2 years ago
Bryce
a46424c673
feature: img2img now supported with PLMS (instead of just DDIM)
...
Kinda hacky copy/pasting from ddim. Need to cleanup
2 years ago
Bryce
4705d182d5
feature: generate captions for images
...
- add wip functionality for negative masks
- ci: add code linter that removes unused imports
- add instructions to install rust on osx
2 years ago
Bryce
f040e3bffe
feature: allow multiple additive targets for text masking targets
2 years ago
Bryce
930295d840
feature: txt2mask - automated text replacement
...
from https://github.com/timojl/clipseg
2 years ago
Bryce
7087c4a680
feature: automatic mask generation
...
mask generation working but not integrated
2 years ago
Bryce
8238e59067
tests: add more tests
2 years ago
Bryce
d7cbf6e416
refactor: cleanup ddim
...
- delete more unused code
- fix some lints
2 years ago
Bryce
6307a0daf5
refactor: cleanup ddim
2 years ago
Bryce
89908603cf
fix: don't repeatedly download the same url
2 years ago
Bryce
59648dbe61
test: add some tests/logging
...
add some experiments
2 years ago
Bryce
19d0b563ac
style: lint fixes. remove unused code
2 years ago
Bryce
c7a822d701
feature: urls as init images
...
- --init-image accepts urls
- cleanup command line code
2 years ago
Bryce
23f0e2be12
fix: select ddim for img2img
2 years ago
wseagar
56fc414724
fix: is_nsfw_img is None rather than false
2 years ago
Bryce
fa3673ef56
fix: use context manager syntax compatible with python 3.8
...
parentheses supported in python 3.10
2 years ago
Bryce
51a8b1b8d8
fix: filter logic was wrong
2 years ago
Bryce
967eb76365
refactor/test: logging suppression + hashed image test
...
- simpler logging suppression for `transformers` library
- suppress logging noise for running tests
- get test running for all samplers on mps and cuda platforms
- refactor safety model env variable to allow classification
2 years ago
Bryce
bb665b9eb6
fix: make k-diffusion samplers deterministic
...
- add test for hashes on mps. images look same on CUDA but are slightly different.
2 years ago
Bryce
b4a3b8c2b3
feature: k-diffusion samplers
...
- improved image logging functionality. can just stick log_latent wherever you want
- improved some variable naming
- moved all the samplers together
- vendored k-diffusion library
2 years ago
Bryce
bb7eaa47f6
docs: small doc updates
2 years ago
Bryce
ff7455034d
docs and lint
2 years ago
Bryce
541ecb9701
feature: face enhancement and upscaling!!
2 years ago
Bryce
14a06e160d
build: vendor clip
...
it's not on pypi
https://github.com/openai/CLIP/issues/141
2 years ago
Bryce
6ac3d40009
docs: update readme. add docs to package
...
Clean up logging some more
2 years ago
Bryce
292d1bcab5
perf: get "half" mode working when running on cuda
2 years ago
Bryce
b9c00dd8de
fix: util functions in wrong place
...
add tile example in readme
2 years ago
Bryce
84a73cb5a2
feature: (wip) better image to image
...
I tried it with the DDIM sampler and it didn't work.
Probably need to use the k-diffusion sampler with it
from a846393251/find_noise.py
needs https://github.com/crowsonkb/k-diffusion
2 years ago
Bryce
438c2868ad
refactor: run import sorter
2 years ago
Bryce
3bdf2dc3ec
feature: tile mode
...
from https://github.com/replicate/cog-stable-diffusion/compare/main...TomMoore515:material_stable_diffusion:main
2 years ago
Bryce
0bb5b6b345
perf: performance optimizations from Doggettx
...
https://github.com/CompVis/stable-diffusion/compare/main...Doggettx:stable-diffusion:autocast-improvements#
https://www.reddit.com/r/StableDiffusion/comments/xalaws/test_update_for_less_memory_usage_and_higher/
2 years ago
Bryce
910b7b4180
refactor: simplify structure
2 years ago
Bryce
f1b1e1c1c8
feature: add nsfw image filter
2 years ago
Bryce
7a33ee2480
feature: cleaned up logging
...
- cleans up all the logging. hide most of it
- create better readme. show example images
- save metadata into image
2 years ago
Bryce
6d1d0622eb
feature: Mac M1 Support out of box
...
- auto-downloads checkpoints
- fixes bug in torch
- autoset environment variable
2 years ago
Bryce
66c640ce7b
feature: add ImageResult. step output option
...
remove verbose args
2 years ago
Bryce
47c6bcee59
refactor: move commands to entrypoints
2 years ago
Bryce
a26e002a86
refactor: use logging instead of prints
2 years ago
Bryce
f782fac570
feature: image prompts
2 years ago
Bryce
0835b2db16
first commit
2 years ago