Mirrored_Repos/InvokeAI
1
0
Fork 0
mirror of https://github.com/invoke-ai/InvokeAI synced 2024-08-30 20:32:17 +00:00
Code Issues Packages Projects Releases Wiki Activity
2,803 Commits 274 Branches 239 Tags 397 MiB
6ab364b16a
Commit Graph

2803 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Jonathan
28b40bebbe
Refactor CUDA cache clearing to add statistical reporting. (#2553) 2023-02-06 12:53:30 -05:00
Lincoln Stein
1c9fd00f98 this is likely the penultimate rc 2023-02-06 12:03:08 -05:00
Lincoln Stein
8ab66a211c
force torch reinstall (#2532) 
For the torch and torchvision libraries **only**, the installer will now
pass the pip `--force-reinstall` option. This is intended to fix issues
with the user getting a CPU-only version of torch and then not being
able to replace it.
 2023-02-06 11:58:57 -05:00
Lincoln Stein
bc03ff8b30
Merge branch 'main' into install/force-torch-reinstall 2023-02-06 11:31:57 -05:00
blessedcoolant
0247d63511 Build (negative-prompt-box) 2023-02-07 05:21:09 +13:00
blessedcoolant
7604b36577 Add Negative Prompts Box 2023-02-07 05:21:09 +13:00
blessedcoolant
4a026bd46e Organize language picker items alphabetically 2023-02-07 05:21:09 +13:00
blessedcoolant
6241fc19e0 Fix the model manager edit placeholder not being full height 2023-02-07 05:21:09 +13:00
blessedcoolant
25d7d71dd8 Slightly decrease the size of the tab list icons 2023-02-07 05:21:09 +13:00
Jonathan
2432adb38f
In exception handlers, clear the torch CUDA cache (if we're using CUDA) to free up memory for other programs using the GPU and to reduce fragmentation. (#2549) 2023-02-06 10:33:24 -05:00
mauwii
0402766f4d add author label 2023-02-06 14:05:27 +01:00
mauwii
a9ef5d1532 update tags 2023-02-06 14:05:27 +01:00
Matthias Wild
a485d45400
Update test-invoke-pip.yml (#2524) 
test-invoke-pip.yml:
- enable caching of pip dependencies in `actions/setup-python@v4`
- add workflow_dispatch trigger
- fix indentation in concurrency
- set env `PIP_USE_PEP517: '1'`
- cache python dependencies
- remove models cache (since we currently use 190.96 GB of 10 GB while I
am writing this)
- add step to set `INVOKEAI_OUTDIR`
- add outdir arg to invokeai
- fix path in archive results

model_manager.py:
- read files in chunks when calculating sha (windows runner is crashing
otherwise)
 2023-02-06 12:56:15 +01:00
mauwii
a40bdef29f update model_manager.py 
- read files in chunks when calculating sha
  - windows runner is crashing without
 2023-02-06 12:30:10 +01:00
mauwii
fc2670b4d6 update test-invoke-pip.yml 
- add workflow_dispatch trigger
- fix indentation in concurrency
- set env `PIP_USE_PEP517: '1'`
- cache python dependencies
- remove models cache (since currently 183.59 GB of 10 GB are Used)
- add step to set `INVOKEAI_OUTDIR`
- add outdir arg to invokeai
- fix path in archive results
 2023-02-06 12:30:10 +01:00
Lincoln Stein
c3807b044d
Merge branch 'main' into install/force-torch-reinstall 2023-02-06 00:18:38 -05:00
Jonathan
b7ab025f40
Update base.py (#2543) 
Free up CUDA cache right after each image is generated. VRAM usage drops down to pre-generation levels.
 2023-02-06 05:14:35 +00:00
Lincoln Stein
633f702b39
fix crash in txt2img and img2img w/ inpainting models and perlin > 0 (#2544) 
- get_perlin_noise() was returning 9 channels; fixed code to return
noise for just the 4 image channels and not the mask ones.

- Closes Issue #2541
 2023-02-05 23:50:32 -05:00
Lincoln Stein
0240656361 fix crash in txt2img and img2img w/ inpainting models and perlin > 0 
- get_perlin_noise() was returning 9 channels; fixed code to return
  noise for just the 4 image channels and not the mask ones.

- Closes Issue #2541
 2023-02-05 22:55:08 -05:00
Matthias Wild
05bb9e444b
update pypi_helper.py (#2533) 
- dont rename requests
- remove dash in verison (`2.3.0-rc3` becomes `2.3.0rc3`)
- read package_name instead of hardcode it
 2023-02-06 03:34:52 +01:00
Lincoln Stein
0076757767
Merge branch 'main' into dev/ci/update-pypi-helper 2023-02-05 21:10:49 -05:00
Lincoln Stein
6ab03c4d08
fix crash in both textual_inversion and merge front ends when not enough models defined (#2540) 
- Issue is that if insufficient diffusers models are defined in
models.yaml the frontend would ungraciously crash.

- Now it emits appropriate error messages telling user what the problem
is.
 2023-02-05 19:34:07 -05:00
Lincoln Stein
142016827f fix formatting bugs in both textual_inversion and merge front ends 
- Issue is that if insufficient diffusers models are defined in
  models.yaml the frontend would ungraciously crash.

- Now it emits appropriate error messages telling user what the problem
  is.
 2023-02-05 18:35:01 -05:00
Lincoln Stein
466a82bcc2
Updates frontend README.md (#2539) 2023-02-05 17:25:25 -05:00
Lincoln Stein
05349f6cdc
Merge branch 'main' into dev/ci/update-pypi-helper 2023-02-05 17:13:09 -05:00
psychedelicious
ab585aefae
Update README.md 2023-02-06 09:07:44 +11:00
Matthias Wild
083ce9358b
hotfix build-container.yml (#2537) 
fix broken tag
 2023-02-05 22:30:23 +01:00
Lincoln Stein
f56cf2400a
Merge branch 'main' into install/force-torch-reinstall 2023-02-05 15:40:35 -05:00
mauwii
fc53f6d47c
hotfix build-container.yml 2023-02-05 21:25:44 +01:00
Matthias Wild
2f70daef8f
Issue/2487/address docker issues (#2517) 
Address issues of #2487
 2023-02-05 21:20:13 +01:00
mauwii
fc2a136eb0
add requested change 2023-02-05 21:15:39 +01:00
Lincoln Stein
ce3da40434 Merge branch 'main' into install/force-torch-reinstall 2023-02-05 15:01:56 -05:00
mauwii
7933f27a72
update pypi_helper.py` 
- dont rename requests
- remove dash in verison (`2.3.0-rc3` becomes `2.3.0rc3`)
- read package_name instead of hardcode it
 2023-02-05 20:45:31 +01:00
mauwii
1c197c602f update Dockerfile, .dockerignore and workflow 
- dont build frontend since complications with QEMU
- set pip cache dir
- add pip cache to all pip related build steps
- dont lock pip cache
- update dockerignore to exclude uneeded files
 2023-02-05 20:20:50 +01:00
mauwii
90656aa7bf update Dockerfile 
- add build arg `FRONTEND_DIR`
 2023-02-05 20:20:50 +01:00
mauwii
394b4a771e update Dockerfile 
- remove yarn install args `--prefer-offline` and `--production=false`
 2023-02-05 20:20:50 +01:00
mauwii
9c3f548900 update settings output in build.sh 2023-02-05 20:20:50 +01:00
mauwii
5662d2daa8 add invokeai/frontend/dist/** to .dockerignore 2023-02-05 20:20:50 +01:00
mauwii
fc0f966ad2 fix docs 2023-02-05 20:20:50 +01:00
mauwii
eb702a5049 fix env.sh, update Dockerfile, update build.sh 
env.sh:
- move check for torch to CONVTAINER_FLAVOR detection

Dockerfile
- only mount `/var/cache/apt` for apt related steps
- remove `docker-clean` from `/etc/apt/apt.conf.d` for BuildKit cache
- remove apt-get clean for BuildKit cache
- only copy frontend to frontend-builder
- mount `/usr/local/share/.cache/yarn` in frountend-builder
- separate steps for yarn install and yarn build
- build pytorch in pyproject-builder

build.sh
- prepare for installation with extras
 2023-02-05 20:20:50 +01:00
mauwii
1386d73302 fix env.sh 
only try to auto-detect CUDA/ROCm if torch is installed
 2023-02-05 20:20:50 +01:00
mauwii
6089f33e54 fix HUGGING_FACE_HUB_TOKEN 2023-02-05 20:20:50 +01:00
mauwii
3a260cf54f update directory from docker-build to docker 2023-02-05 20:20:50 +01:00
mauwii
9949a438f4 update docs with newly added variables 
also remove outdated information
 2023-02-05 20:20:50 +01:00
mauwii
84c1122208 fix build.sh and env.sh 2023-02-05 20:20:50 +01:00
Lincoln Stein
cc3d431928
2.3.0rc4 (#2514) 
This will bring main up to date with v2.3.0-rc4
 2023-02-05 14:05:15 -05:00
Lincoln Stein
c44b060a2e
Merge branch 'main' into 2.3.0rc4 2023-02-05 13:40:56 -05:00
Lincoln Stein
eff7fb89d8 installer will --force-reinstall torch 2023-02-05 13:39:46 -05:00
Lincoln Stein
cd5c112fcd
Allow multiple models to be imported by passing a directory. (#2529) 
This change allows passing a directory with multiple models in it to be
imported.

Ensures that diffusers directories will still work.

Fixed up some minor type issues.
 2023-02-05 13:36:00 -05:00
Lincoln Stein
563867fa99
Merge branch 'main' into main 2023-02-05 12:51:03 -05:00