aljaz/generative-models
1
0
Fork 0
mirror of https://github.com/Stability-AI/generative-models.git synced 2026-02-02 20:34:27 +01:00
Code Issues Packages Projects Releases Wiki Activity
71 Commits 6 Branches 2 Tags
d245e2002fa6b2b0eb6826a954d738a6481c9505
Commit Graph

46 Commits

Author SHA1 Message Date
Stephan Auerhahn
d245e2002f more types 2023-08-09 13:46:06 -07:00
Stephan Auerhahn
a009aa8a9f adding some typing 2023-08-09 13:27:30 -07:00
Stephan Auerhahn
f86ffac274 context manager 2023-08-09 12:38:44 -07:00
Stephan Auerhahn
c4b7baf896 Streamlit refactor (#105) 
* initial streamlit refactoring pass

* cleanup and fixes

* fix refiner strength

* Modify params correctly

* fix exception
 2023-08-06 19:58:52 -07:00
Stephan Auerhahn
6c18c8443a rename ModelOnDevice to SwapToDevice 2023-08-06 23:46:20 +00:00
Stephan Auerhahn
ced97f0e84 update defaults 2023-08-06 23:24:14 +00:00
Stephan Auerhahn
76ca428422 fix path resolution bug 2023-08-06 21:39:18 +00:00
Stephan Auerhahn
8f8757b4ff version bump for changes to inference helpers 2023-08-06 21:09:09 +00:00
Stephan Auerhahn
f2fba1dfa2 fix noisy latent handling 2023-08-06 21:08:19 +00:00
Stephan Auerhahn
451c76ada1 format 2023-08-06 12:26:16 +00:00
Stephan Auerhahn
0c2c5c66a2 fix device check 2023-08-06 12:26:01 +00:00
Stephan Auerhahn
ea5f232d5d move conditioner to device 2023-08-06 11:42:39 +00:00
Stephan Auerhahn
f06c67c206 formatting, remove reference 2023-08-06 11:30:40 +00:00
Stephan Auerhahn
b216934b7e align with streamlit helpers and re-de-deuplicate 2023-08-06 11:20:22 +00:00
Stephan Auerhahn
77d0e27747 format 2023-08-03 17:57:55 -07:00
Stephan Auerhahn
4aea6fa2a4 Fix checkpoint loading too 2023-08-03 17:56:24 -07:00
Stephan Auerhahn
84d3a7f6f5 fix fallback logic for config path 2023-08-03 17:50:10 -07:00
Stephan Auerhahn
19fa4da3de run black again 2023-08-04 00:16:29 +00:00
Stephan Auerhahn
4e2236f67d Fix path logic for development installs 2023-08-04 00:15:22 +00:00
Stephan Auerhahn
baf79d2d79 black 2023-08-04 00:00:51 +00:00
Stephan Auerhahn
44943df4f2 Allow loading custom models and improve path logic 2023-08-03 23:59:42 +00:00
Stephan Auerhahn
73287ec3a3 Extract method for img2img wrapper 2023-08-03 23:42:11 +00:00
Stephan Auerhahn
853adb4022 Add defaults to refiner function 2023-08-03 12:50:23 -07:00
Stephan Auerhahn
45feb6cb9c Use wrapper correctly in refiner helper 2023-08-02 23:14:30 +00:00
Stephan Auerhahn
299abbcd90 Use final v1 filename (#67) 2023-07-26 19:53:19 +02:00
Stephan Auerhahn
931d7a389a Add inference helpers & tests (#57) 
* Add inference helpers & tests

* Support testing with hatch

* fixes to hatch script

* add inference test action

* change workflow trigger

* widen trigger to test

* revert changes to workflow triggers

* Install local python in action

* Trigger on push again

* fix python version

* add CODEOWNERS and change triggers

* Report tests results

* update action versions

* format

* Fix typo and add refiner helper

* use a shared path loaded from a secret for checkpoints source

* typo fix

* Use device from input and remove duplicated code

* PR feedback

* fix call to load_model_from_config

* Move model to gpu

* Refactor helpers

* cleanup

* test refiner, prep for 1.0, align with metadata

* fix paths on second load

* deduplicate streamlit code

* filenames

* fixes

* add pydantic to requirements

* fix usage of `msg` in demo script

* remove double text

* run black

* fix streamlit sampling when returning latents

* extract function for streamlit output

* another fix for streamlit outputs

* fix img2img in streamlit

* Make fp16 optional and fix device param

* PR feedback

* fix dict cast for dataclass

* run black, update ci script

* cache pip dependencies on hosted runners, remove extra runs

* install package in ci env

* fix cache path

* PR cleanup

* one more cleanup

* don't cache, it filled up
 2023-07-26 04:37:24 -07:00
Benjamin Aubin
e596332148 Pre release changes for production (#59) 
* clean requirements

* rm taming deps

* isort, black

* mv lipips, license

* clean vq, fix path

* fix loss path, gitignore

* tested requirements pt13

* fix numpy req for python3.8, add tests

* fix name

* fix dep scipy 3.8 pt2

* add black test formatter
 2023-07-26 12:09:28 +02:00
Jonas Müller
4a3f0f546e Revert "Replace most print()s with logging calls (#42)" (#65) 
This reverts commit 6f6d3f8716.
 2023-07-26 10:30:21 +02:00
Tim Dockhorn
7934245835 Revert "Minimize re-exports from __init__ files (#44)" (#63) 
This reverts commit 57862fb4c7.
 2023-07-26 10:26:28 +02:00
Tim Dockhorn
1da250906d Revert "Dead code removal (#48)" (#62) 
This reverts commit b5b5680150.
 2023-07-26 10:26:00 +02:00
Tim Dockhorn
a4ceca6d03 Revert "fall back to vanilla if xformers is not available (#51)" (#61) 
This reverts commit ef520df1db.
 2023-07-26 10:25:17 +02:00
ablattmann
68f3f89bd3 Fix crashing line in logging in sgm/models/diffusion.py (#64) 2023-07-26 10:13:42 +02:00
Aarni Koskela
57862fb4c7 Minimize re-exports from __init__ files (#44) 
This allows importing parts of the package without having to
import practically everything (since importing a package will
import its parents' __init__s, etc).
 2023-07-25 16:24:09 +02:00
Aarni Koskela
ef520df1db fall back to vanilla if xformers is not available (#51) 2023-07-25 16:21:51 +02:00
Aarni Koskela
2897fdc99a Move attention testing/benchmarking code out of package (#47) 2023-07-25 15:40:22 +02:00
Aarni Koskela
b5b5680150 Dead code removal (#48) 
* Remove old commented-out attention code

* Mark two functions as likely unused

* Use exists() and default() from sgm.util
 2023-07-25 15:24:24 +02:00
Aarni Koskela
6f6d3f8716 Replace most print()s with logging calls (#42) 2023-07-25 15:21:30 +02:00
Jonas Müller
e25e4c0df1 Merge pull request #43 from akx/fix-safetensors-load 
Fix loading safetensors with load_model_from_config
 2023-07-21 18:00:53 +02:00
Aarni Koskela
e5dc9669ed Set up Python packaging (#17) 
* Sort .gitignore; add dist and *.py[cod]

* Use pyproject.toml + Hatch instead of setup.py

Sibling of https://github.com/Stability-AI/stablediffusion/pull/269

* Add packaging documentation
 2023-07-18 13:06:05 +02:00
Aarni Koskela
48904a692d Fix loading safetensors with load_model_from_config 
Previously, the `sd` from the safetensors if branch was not used at all, and `pl_sd` would have not been assigned.
 2023-07-17 09:56:35 +03:00
Tim
89f5413e6d Getting rid of unnecessary error message 2023-07-09 10:40:18 -07:00
Tim
ba3e7fed5a Fixing additional GPU memory on device 0 due to discretization 2023-07-09 10:40:09 -07:00
Tim
2796c81a5f Making spacing function slimmer 2023-06-30 11:17:57 -07:00
Tim
e9869d7822 Changed LegacyDDPMDiscretization for sampling 2023-06-30 11:17:46 -07:00
Tim
a1af4ac4f1 Adapting txt logging for python 3.8 2023-06-25 16:42:45 -07:00
Andreas Blattmann
081e0d4629 soon is now 2023-06-22 09:53:12 -07:00