Bryce
929eff353e
fix: edit was broken by recent changes
2 years ago
Bryce
5b1cded94f
fix: compose phase for edits probably doesn't make sense
2 years ago
Bryce
08f6014941
version: 10.0.0
2 years ago
Bryce
6f749ac4b6
fix: create compilations dir as needed
2 years ago
Bryce
882cc7e0f1
feature: generate large images with coherent composition
2 years ago
Bryce
c3a88c44cd
perf: tiled encoding of images
...
The fold/unfold/split_input_params images didn't look good
2 years ago
Bryce
2aef6089e0
feature: generate large images
...
Added a composition stage so large images are more coherent
2 years ago
Bryce
b93b6a4d7c
perf: use silu instead of nonlinearity for speedup
2 years ago
Bryce
68e7fd73c5
perf: add xformers dependency
2 years ago
Bryce
003a512dc8
perf: mps perf improvement
...
attention slicing wasn't working
2 years ago
Bryce
66b28c80fc
fix: images came out wrong if processed on mps
2 years ago
Bryce
8db73707ff
perf: load model to gpu after making it "half"
2 years ago
Bryce
1c8893cecc
fix: fix image gen retry
2 years ago
Bryce
ea1d4baafe
refactor: cleanup image generation code
2 years ago
Bryce
8a97213622
feature: sliced latent decoding
...
allows generation of bigger images. tile seams can be noticeable occasionally despite the feathering
2 years ago
Bryce
ad62496557
version: 9.0.2
2 years ago
Bryce
da70b310d5
fix: edit interface was broken
2 years ago
Bryce
75e3c32d54
version: 9.0.1
2 years ago
Bryce
d1dbd500eb
fix: use entry_points for windows
...
python converts entry_points to exe files so they run on windows. The scripts that work on linux/macos don't work on windows.
2 years ago
Bryce
5bedaebe45
version: 9.0.0
2 years ago
Bryce
eeb80f81aa
docs: update changelog
2 years ago
Bryce
76d40560ef
style: lintfix
2 years ago
Bryce
6c1e3b2ae7
feature: allow specifying output format
...
closes #228
2 years ago
Bryce
d220819377
feature: accept multiple input images to allow batch editing
...
- move "surprise me" to "edit-demo"
2 years ago
Bryce
63f32846bb
docs: no mps training; how change cache dir
2 years ago
Bryce
d9a70a55c6
feature/docs: cleanup cli
...
- add colors to help text
- add suggestion to use `aimg` shell
- lazy imports for faster response time
2 years ago
Deadolus
8df10576f3
fix: removing stray space from readme at inopportune location ( #237 )
...
In the multiline example with the "girl with the pearl earring" was a stray space that led to an error.
As the command line was interpreted as two separate commands.
I remove the stray space in this pull-request, to spare others the headache of finding the whitespace.
2 years ago
Bryce Drennan
9eacf5e7ed
perf: improve startup time by doing some imports lazily ( #233 )
...
just running `aimg --help` or `aimg --version` was very slow due to all the imports being brought in eagerly
Before changes `aimg --help`
`2.24s user 4.05s system 184% cpu 3.416 total`
After changes:
`0.04s user 0.02s system 8% cpu 0.625 total`
Used `PYTHONPROFILEIMPORTTIME=1 aimg --help` to find time consuming imports.
Also switched to using `scripts` instead of `entrypoints` since the scripts are much faster.
Made duplicate SAMPLER_TYPE_OPTIONS that can be loaded without loading all the samplers themselves.
Likely a breaking change - not sure.
2 years ago
Bryce
b611a92b49
version: 8.3.1
2 years ago
Bryce
19444d8e22
fix: first frame should be the interesting one
2 years ago
Bryce Drennan
e619dc2867
fix: init image strength ( #225 )
2 years ago
Bryce
d7eaac2c5e
version: 8.3.0
2 years ago
Bryce Drennan
9ee09ac842
feature: add compilation animations ( #224 )
...
- add generation/compare gifs
2 years ago
Bryce Drennan
a67683d318
feature: add upscale command line function ( #222 )
2 years ago
Bryce Drennan
542e4fbd55
fix: tile mode was broken since latest perf improvements ( #220 )
2 years ago
Pavel Yakovlev
e1c9f14489
feature: add openjourney models
2 years ago
Bryce Drennan
9e06013ade
feature: save gifs that show image generation process ( #218 )
2 years ago
Bryce
8791e15bec
version: 8.2.0
2 years ago
Bryce Drennan
6d733ed3ff
feature: added `aimg system-info` command to help debug issues ( #216 )
2 years ago
Bryce Drennan
7deabaae62
tests: update expected images for tests ( #215 )
2 years ago
Bryce
72517f28c6
version: 8.1.0
2 years ago
Bryce Drennan
7bdde559cc
performance: memory management improvements ( #210 )
...
- tile mode made more efficient. especially when not being used
- add script to iteratively make bigger images
2 years ago
Jiaming Sun
a338902ab5
fix: add png handling for LazyLoadingImage ( #203 )
...
Loading png images for captioning will cause the following error since PIL loads png images in RGBA mode.
```python
File "./miniconda3/lib/python3.8/site-packages/torchvision/transforms/functional_tensor.py", line 940, in normalize
return tensor.sub_(mean).div_(std)
RuntimeError: The size of tensor a (4) must match the size of tensor b (3) at non-singleton dimension 0
```
2 years ago
Mahmoud Al-Qudsi
0b7692df07
fix: make model cached paths compatible with windows ( #205 )
...
If a model has a remote filename with characters that might not be supported on
the local cache filesystem, replace them during translation of URL to cached
file path.
Closes #202 .
2 years ago
Bryce
92baff890e
feature: surprise me improvements
2 years ago
Bryce
c7ed86a36f
feature: auto-retry generation unsafe image is created
2 years ago
Bryce
f60fa880a7
feature: less visually confusing shell output
2 years ago
Bryce
16cef16c3e
feature: facilitate references to previous image generation
2 years ago
Bryce
248679d8de
feature: image sizes can now be multiples of 8 instead of 64
...
from https://github.com/CompVis/stable-diffusion/issues/60#issuecomment-1240294667
2 years ago
Bryce
9bef5a151a
version: 8.0.5
2 years ago