Finegrain/refiners
1
0
Fork 0
mirror of https://github.com/finegrain-ai/refiners.git synced 2024-11-24 23:28:45 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
579 commits 35 branches 5 tags 149 MiB
Commit graph

579 commits

This branch
This branch
All branches
Author SHA1 Message Date
Pierre Chapuis b1c200c63a Update src/refiners/foundationals/latent_diffusion/lora.py 
Co-authored-by: Laureηt <laurent@lagon.tech>
 2024-02-07 17:47:14 +01:00
Pierre Chapuis eb9abefe07 add a few comments in SDLoraManager 2024-02-07 17:47:14 +01:00
Benjamin Trom bbe0759151 fix docstring 2024-02-07 16:13:01 +01:00
limiteinductive 2e526d35d1 Make Dataset part of the trainer 2024-02-07 16:13:01 +01:00
Laurent 9883f24f9a (fluxion/layers) remove View layer 
+ replace existing `View` layers by `Reshape`
 2024-02-07 12:06:07 +01:00
limiteinductive 2ef4982e04 remove wandb from base config 2024-02-07 11:06:59 +01:00
Pierre Chapuis 11da76f7df fix sdxl structural copy 2024-02-07 10:51:26 +01:00
Pierre Chapuis ca9e89b22a cosmetics 2024-02-07 10:51:26 +01:00
limiteinductive ea05f3d327 make device and dtype work in Trainer class 2024-02-06 23:10:10 +01:00
Hugues Pouillot 1fa5266f56 fix github actions triggers 2024-02-06 16:48:18 +01:00
Cédric Deltheil 907a6becbc
bump library version to v0.3.1 2024-02-06 15:22:15 +01:00
Pierre Chapuis 98fce82853 fix 37425fb609 
Things to understand:

- subscripted generic basic types (e.g. `list[int]`) are types.GenericAlias;
- subscripted generic classes are `typing._GenericAlias`;
- neither can be used with `isinstance()`;
- get_origin is the cleanest way to check for this.
 2024-02-06 13:49:37 +01:00
Cédric Deltheil f9305aa416 pyproject.toml: add some PyPI classifiers 
In particuler, the Python versions (3.10, etc) used to be included with
builds created with Poetry (they got removed after the switch to Rye:
see #141). This commit should fix the broken pypi badge.
 2024-02-06 11:40:01 +01:00
Pierre Chapuis 37425fb609 make LoRA generic 2024-02-06 11:32:18 +01:00
Pierre Chapuis 471ef91d1c make __getattr__ on Module return object, not Any 
PyTorch chose to make it Any because they expect its users' code
to be "highly dynamic": https://github.com/pytorch/pytorch/pull/104321

It is not the case for us, in Refiners having untyped code
goes contrary to one of our core principles.

Note that there is currently an open PR in PyTorch to
return `Module | Tensor`, but in practice this is not always
correct either: https://github.com/pytorch/pytorch/pull/115074

I also moved Residuals-related code from SD1 to latent_diffusion
because SDXL should not depend on SD1.
 2024-02-06 11:32:18 +01:00
Cédric Deltheil ec401133f1
bump library version to v0.3.0 2024-02-05 10:27:22 +01:00
Pierre Chapuis 3de1508b65 increase tolerance on Euler test 2024-02-04 08:58:22 +01:00
Pierre Chapuis 83b478c0ff fix test failure caused by Diffusers 0.26.0 2024-02-04 08:58:22 +01:00
Laurent 8d190e4256 (fluxion/layers/activations) replace ApproximateGeLU by GeLUApproximation 2024-02-02 19:41:18 +01:00
hugojarkoff 2bdb42e88d Change image preprocessing resizing to use Pillow 2024-02-02 18:21:04 +01:00
hugojarkoff 75830e2179 Add T2I-Adapter subsection to SDXL Adaptation guide 2024-02-02 18:21:04 +01:00
Pierre Chapuis fbb1fcb8ff Chain#pop does not return tuples 2024-02-02 18:11:51 +01:00
Cédric Deltheil a779f86941
README: add link to Discussions 2024-02-02 17:58:04 +01:00
Laurent 1dcb36e1e0 (doc/foundationals) add IPAdapter, related docstrings 2024-02-02 17:35:03 +01:00
Laurent 6b35f1cc84 (doc/foundationals) add SDLoraManager, related docstrings 2024-02-02 17:35:03 +01:00
Laurent 7406d8e01f (mkdocs) fix cross-reference typo 2024-02-02 17:35:03 +01:00
Laurent 093527a7de apply @deltheil suggestions 2024-02-02 17:35:03 +01:00
Laurent f62e71da1c (doc/foundationals) add SegmentAnything, related docstrings 2024-02-02 17:35:03 +01:00
Laurent a926696141 (doc/foundationals) add CLIP, related docstrings 2024-02-02 17:35:03 +01:00
Laurent 3910845e29 (doc/foundationals) add DINOv2, related docstrings 2024-02-02 17:35:03 +01:00
Laurent fc7b4dd62d (doc/fluxion/ld) add DDPM, DDIM, DPM++ and Euleur docstrings 2024-02-02 17:35:03 +01:00
Laurent 6d8016190c (docs) modify which docstrings are displayed in docs/reference/latent_diffusion.md 2024-02-02 17:35:03 +01:00
Laurent a1a00998ea (doc/fluxion/ld) add StableDiffusion_1 docstrings 2024-02-02 17:35:03 +01:00
Laurent f2bcb7f45e (mkdocstrings) export SDXLAutoencoder in src/refiners/foundationals/latent_diffusion/stable_diffusion_xl/__init__.py 2024-02-02 17:35:03 +01:00
Laurent 2a7b86ac02 (doc/fluxion/ld) add LatentDiffusionAutoencoder docstrings 2024-02-02 17:35:03 +01:00
Laurent effd95a1bd (doc/fluxion/ld) add SDXLAutoencoder docstrings 2024-02-02 17:35:03 +01:00
Laurent 0c5a7a8269 (doc/fluxion/ld) add Solver docstrings 2024-02-02 17:35:03 +01:00
Laurent 289261f2fb (doc/fluxion/ld) add SD1UNet docstrings 2024-02-02 17:35:03 +01:00
Laurent fae08c058e (doc/fluxion/ld) add SDXLUNet docstrings 2024-02-02 17:35:03 +01:00
Laurent 270357ed29 (docs) ignore CLIP related docstrings in latent_diffusion.md 2024-02-02 17:35:03 +01:00
Pierre Chapuis c02a7b378f better example model for Chain 2024-02-02 17:30:14 +01:00
Cédric Deltheil 17b085341a brush up main README 2024-02-02 16:31:02 +01:00
Pierre Chapuis 4e80d295bd ruff 0.2.0 2024-02-02 14:46:37 +01:00
Pierre Chapuis add4f37d97 add footnote for PyTorch paper and link to walk 2024-02-02 14:38:36 +01:00
Cédric Deltheil 2f9e30bf63 deprecate outdated notebooks/basics.ipynb 
To be re-introduced later
 2024-02-02 14:12:59 +01:00
Cédric Deltheil c53f5c499b
Update LICENSE 2024-02-02 14:08:09 +01:00
Cédric Deltheil 17046e0717 docs/getting-started: tweak toc and update snippet 
`encode_image` has been deprecated part of #213
 2024-02-02 14:05:06 +01:00
Cédric Deltheil 7b8e9e3c67 docs: wording tweaks (homepage and banner) 2024-02-02 12:24:46 +01:00
Pierre Chapuis 2afb7cb638 add some links to API reference 2024-02-02 12:01:06 +01:00
Cédric Deltheil f7cc6e577c docs/index.md: add badges 2024-02-02 11:26:13 +01:00