3f13dd3ae8
prompt parsing is now much more robust
2022-10-20 21:05:36 +02:00
79b4afeae7
parser working with basic escapes
2022-10-20 16:56:34 +02:00
da223dfe81
wip re-writing parts of prompt parser
2022-10-20 15:56:46 +02:00
c9d27634b4
bring in prompt parser from fix-prompts branch
...
attention is parsed but ignored, blends old syntax doesn't work,
conjunctions are parsed but ignored, the only part that's used
here is the new .blend() syntax and cross-attention control
using .swap()
2022-10-20 12:01:48 +02:00
42883545f9
add prompt language support for cross-attention .swap
2022-10-20 01:42:04 +02:00
1ffd4a9e06
refactored single diffusion path seems to be working for all samplers
2022-10-19 21:08:03 +02:00
147d39cb7c
wip refactoring shared InvokeAI diffuser mixin to component
2022-10-19 21:08:03 +02:00
824cb201b1
pass img2img ddim/plms edited conditioning through kwargs
2022-10-19 21:08:03 +02:00
582880b314
add cross-attention support to im2img; prevent inpainting from crashing
2022-10-19 21:08:03 +02:00
2b79a716aa
wip hi-res fix
2022-10-19 21:08:03 +02:00
d572af2acf
fix cross-attention on k* samplers
2022-10-19 21:08:03 +02:00
54e6a68acb
wip bringing cross-attention to PLMS and DDIM
2022-10-19 21:08:03 +02:00
09f62032ec
cleanup and clarify comments
2022-10-19 21:08:03 +02:00
711ffd238f
cleanup
2022-10-19 21:08:03 +02:00
056cb0d8a8
sliced cross-attention wrangler works
2022-10-19 21:08:03 +02:00
37a204324b
go back to using InvokeAI attention
2022-10-19 21:08:03 +02:00
1fc1f8bf05
cross-attention working with placeholder {} syntax
2022-10-19 21:06:42 +02:00
8ff507b03b
runs but doesn't work properly - see below for test prompt
...
test prompt:
"a cat sitting on a car {a dog sitting on a car}" -W 384 -H 256 -s 10 -S 12346 -A k_euler
note that substition of dog for cat is currently hard-coded (ksampler.py
line 43-44)
2022-10-19 21:06:42 +02:00
33d6603fef
cleanup initial experiments
2022-10-19 21:06:42 +02:00
b0b1993918
initial experiments
2022-10-19 21:06:42 +02:00
07a3df6001
DRAFT: Cross-Attention Control
...
Signed-off-by: Ben Alkov <ben.alkov@gmail.com >
2022-10-19 21:06:42 +02:00
92d4dfaabf
Merge branch 'asymmetric-tiling' of https://github.com/carson-katri/InvokeAI into carson-katri-asymmetric-tiling
2022-10-19 13:46:07 -04:00
bc626af6ca
Skips normalizing prompts for web UI metadata
2022-10-19 13:38:16 -04:00
a45786ca2e
Builds fresh bundle
2022-10-19 13:27:43 -04:00
2926c8299c
Fixes lingering references to GFPGAN vs Facetool
2022-10-19 13:27:43 -04:00
32a5ffe436
Adds Codeformer support
2022-10-19 13:27:43 -04:00
62dd3b7d7d
resolve models.clipseg vs clipseg ambiguity
2022-10-18 23:09:26 -04:00
15aa7593f6
Merge branch 'development' into asymmetric-tiling
2022-10-18 22:37:18 -04:00
9b3ac92c24
fix incorrect import of clipseg
2022-10-18 19:28:30 -04:00
66f6ef1b35
fix syntax errors in preload
2022-10-18 19:25:18 -04:00
d93cd10b0d
Merge branch 'development' into asymmetric-tiling
2022-10-18 17:27:29 -04:00
a488b14373
prevent preload warning message
2022-10-18 17:09:17 -04:00
0147dd6431
update requirements to address #1149
2022-10-18 16:28:58 -04:00
9d19213b8a
Merge branch 'development' of github.com:lstein/stable-diffusion into asymmetric-tiling
2022-10-18 13:34:10 -04:00
71c3835f3e
yarn built
2022-10-18 13:22:58 -04:00
0fbd26e9bf
simpler socketio setup URL handling
2022-10-18 13:22:58 -04:00
2a78eb96d0
Merge branch 'development' of github.com:invoke-ai/InvokeAI into development
2022-10-18 08:30:02 -04:00
3a1003f702
Fix typo
...
Taken from `main` PR #1147
Author: eltociear
2022-10-18 08:29:26 -04:00
329a9d0b11
Merge branch 'text-masking' of github.com:invoke-ai/InvokeAI into text-masking
2022-10-18 08:28:56 -04:00
17d75f3da8
update environment/requirements for clipseg dependency
2022-10-18 08:27:49 -04:00
20551857da
add clipseg support for creating inpaint masks from text
...
On the command line, the new option is --text_mask or -tm.
Example:
```
invoke> a baseball -I /path/to/still_life.png -tm orange
```
This will find the orange fruit in the still life painting and replace
it with an image of a baseball.
2022-10-18 08:27:48 -04:00
32122e0312
clipseg library and environment in place
2022-10-18 08:27:48 -04:00
e6fc8af249
Fix typo
...
Taken from `main` PR #1147
Author: eltociear
2022-10-18 08:08:58 -04:00
c974c95e2b
Merge branch 'development' of github.com:invoke-ai/InvokeAI into development
2022-10-17 23:14:55 -04:00
3b2590243c
^C at invoke> cmd line exits gracefully
2022-10-17 23:14:32 -04:00
1c2bd275fe
Fix img2img DDIM index out of bound
...
Added a [community solution](https://github.com/CompVis/stable-diffusion/issues/111#issuecomment-1229483511 ) to fix index out of bound when doing img2img generation with `ddim` sampler. Also, restored `steps_out` to be `ddim_timesteps + 1` since the removal was meant to fix the [1000 steps issue](https://github.com/CompVis/stable-diffusion/issues/111 )
2022-10-17 22:32:15 -04:00
0cf11ce488
add option to CLI and pngwriter that allows user to set PNG compression level
...
- In CLI: the argument is --png_compression <0..9> (-z<0..9>)
- In API, pass `compress_level` to PngWriter.save_image_and_prompt_to_png()
Compression ranges from 0 (no compression) to 9 (maximum compression).
Default value is 6 (as specified by Pillow package).
This addresses an issue first raised in #652 .
2022-10-17 22:27:47 -04:00
d6195522aa
Add seamless_axes docs to CLI.md
2022-10-17 20:17:34 -04:00
3b79b935a3
Merge branch 'development' into asymmetric-tiling
2022-10-17 20:15:42 -04:00
4079333e29
Document the seamless_axes argument
2022-10-17 19:33:17 -04:00
