62de446a92
ci: add mypy github action
5 months ago
012cc648d3
style: fix all the mypy typing issues
...
...or ignore them.
5 months ago
5a636e45c5
feature: skip composition at sizes slightly larger than model is expecting
5 months ago
203747b14f
refactor: simplify model_weights/architecture
5 months ago
37ecd1e5e0
fix: videogen. track gpu tests
5 months ago
eae4f20ae2
ci: add type checker
...
fix some typehint issues
5 months ago
e898e3a799
fix: several cli commands, edit demo, negative prompt
...
- fix colorize cmd. add test
- fix describe cmd. add test
- fix model-list cmd. add test
- fix stable studio
- hide stack grace for ValueErrors in cli
- set controlnet scale
- fix negative prompt to allow emptystring instead of replacing it with default
- adjust edit-demo parameters
- arg scheduler that works at click level (but disable it). works but not ideal experience.
5 months ago
c299cfffd9
fix: cache the controlnet models
5 months ago
9b95e8b0b6
perf: improve cli startup time
...
- do not provide automatically imported api functions and objects in `imaginairy` root module
- horrible hack to overcome horrible design choices by easy_install/setuptools
The hack modifies the installed script to remove the __import__ pkg_resources line
If we don't do this then the scripts will be slow to start up because of
pkg_resources.require() which is called by setuptools to ensure the
"correct" version of the package is installed.
before modification example:
```
__requires__ = 'imaginAIry==14.0.0b5'
__import__('pkg_resources').require('imaginAIry==14.0.0b5')
__file__ = '/home/user/projects/imaginairy/imaginairy/bin/aimg'
with open(__file__) as f:
exec(compile(f.read(), __file__, 'exec'))
```
5 months ago
2bd6cb264b
feature: large refactor
...
- add type hints
- size parameter
- ControlNetInput => ControlInput
- simplify imagineresult
5 months ago
db85f0898a
feature: remove training feature
5 months ago
ef0f44646e
feature: adds --control-strength as parameter for cli
5 months ago
24f4af3482
feature: better torch installation experience
5 months ago
71d4992dca
feature: added `--size` parameter to allow using named sizes
5 months ago
14ecf93c6a
feature: add sliced self-attention for future use with video generation
5 months ago
1136fdb939
version: 14.0.0b5
5 months ago
6fcf0da331
test: try local runner
5 months ago
0fe3733933
fix: memory management issue
...
the dtype being used as a cache key wasn't consistent. this caused the model to be loaded twice
5 months ago
82c30024c9
feature: DDIM now default sampler
...
better output quality
5 months ago
ba57393022
feature: patch refiners ScaledDotProductAttention for sliced attention
5 months ago
1b15d6dcd4
feature: sliced image encoding for SD1Autoencoder
5 months ago
b61d06651c
tests: fix tests
...
- disable details mode. needs more work done to support
5 months ago
cff17ef6f4
feature: adds text to video generation flag ( #404 )
...
Co-authored-by: jaydrennan
6 months ago
8fec56970e
version: 14.0.0b4
6 months ago
8267482aad
fix: remove padding approach
...
- padding didnt make animations better
- some changes to better support CPU generation (not yet working)
- better log output coloring
- better log messages when cuda not found
6 months ago
6e1c44dae7
fix: add missing data files
...
fixes #402
6 months ago
b9e245eb7d
fix: remove cuda memory tracking
6 months ago
3b2265f82d
feature: add video progress bar
6 months ago
b5a0e65f35
fix: edit mode and some controlnet tests
6 months ago
b7fad562d0
feature: improved logging
...
- clean up some error messages
- add color
- indent tqdm bar
6 months ago
07b097e001
version: 14.0.0b3
6 months ago
a126e1c1d7
feature: better error message when file not found
6 months ago
aa91d0a9c9
feature: autoresize and crop videos.
...
This means you can just stick any image into the video generator without worrying about the size.
- better generated video filenames
- output h264 video as well
6 months ago
f6c9927d0c
version: 14.0.0b2
6 months ago
c24ed1f33d
feature: videogen improvements
...
- better filenames
- allow urls as image inputs
- better memory efficiency
- add timing information
6 months ago
ab75c49b15
docs: smaller animations
6 months ago
ed23c7e1ca
version: 14.0.0b1
6 months ago
1da7043081
docs: update documentation
6 months ago
e8fe8d7d6c
feature: stable diffusion video (SVD)
6 months ago
80ff006604
fix: updates weights_url's for controlnet
6 months ago
e91a041a78
refactor: removes unused model configs
6 months ago
76df1210f4
refactor: removes outdated info and spelling error
6 months ago
d6d2f4b9dd
refactor: updates README for 14.0.0
...
adds notes under currently broken features, removes deprecated features descriptions, and adds 14.0.0 changelog
6 months ago
f97f6a3b4b
feature: use refiners library for generation
...
BREAKING CHANGE
- stable diffusion 1.5 + inpainting working
- self-attention guidance working. improves image generation quality
- tile-mode working
- inpainting self-attention guidance working
disable/broken features:
- sd 1.4, 2.0, 2.1
- most of the samplers
- pix2pix edit
- most of the controlnets
- memory management
- python 3.8 support
wip
6 months ago
6cd519cdb2
refactor: move code to avoid conflicts with "http" namespace
7 months ago
703fb6e331
ci: faster pip install in github actions
8 months ago
2273c9144d
ci: smarter model caching in github actions
8 months ago
db4f040536
tests: split test suite for faster overall runtime
8 months ago
558d3388e5
style: speed up linting and autoformatting. fix lints
8 months ago
460add16b8
version: 13.2.1
8 months ago
Bryce
jaydrennan
jaydrennan
jaydrennan