8ff777d3c1
Attention slicing ( #407 )
...
uup
2022-09-07 22:48:13 +02:00
1a431ae886
Rename variables from single letter to meaningful name fix ( #395 )
...
Co-authored-by: Rashmi S <rashmis@Rashmis-MacBook-Pro.local>
2022-09-07 18:50:56 +02:00
8d14edf27f
Docs: Stable Diffusion pipeline ( #386 )
...
* Initial description of Stable Diffusion pipeline.
* Placeholder docstrings to test preview.
* Add docstrings to Stable Diffusion pipeline.
* Style
* Docs for all the SD pipelines + attention slicing.
* Style: wrap long lines.
2022-09-07 18:48:49 +02:00
58d627aed6
Small changes to Philosophy ( #403 )
...
Small changes to Philosophy.
2022-09-07 18:47:38 +02:00
65ed5d2845
karras-ve docs ( #401 )
...
* karras-ve docs
for issue #293
* make style
2022-09-07 18:34:54 +02:00
44091d8b2a
Score sde ve doc ( #400 )
...
* initial score_sde_ve docs
* fixed typo
* fix VE term
2022-09-07 18:34:34 +02:00
e0d836c813
[Docs] Finish Intro Section ( #402 )
...
* up
* up
* finish
2022-09-07 18:00:49 +02:00
8603ca6b09
[Docs] Quicktour ( #397 )
...
* uP
* better
* up
* finish
* up
2022-09-07 16:29:34 +02:00
fead3ba386
ddim docs ( #396 )
...
* ddim docs
for issue #293
* space
2022-09-07 16:29:06 +02:00
492f5c9a6c
Docs: optimization / special hardware ( #390 )
...
Add mps documentation.
2022-09-07 16:27:14 +02:00
71d737bfe2
added pndm docs ( #391 )
...
for issue #293
2022-09-07 15:33:17 +02:00
5b4f5951a9
Update text_inversion.mdx ( #393 )
...
* Update text_inversion.mdx
Getting in a bit of background info
* fixed typo mode -> model
* Link SD and re-write a few bits for clarity
* Copied in info from the example script
As suggested by surajpatil :)
* removed an unnecessary heading
2022-09-07 18:48:34 +05:30
3dcc5e9a5a
[Docs] Logging ( #394 )
...
up
2022-09-07 14:58:21 +02:00
9288fb1df8
[Pipeline Docs] ddpm docs for sprint ( #382 )
...
* initial ddpm
for issue #293
* initial ddpm pipeline doc
* added docstrings
* Update docs/source/api/pipelines/ddpm.mdx
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
* make style
* fix docs
* make style
* fix doc strings
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
2022-09-07 14:43:29 +02:00
a0592a13ee
[Pipeline Docs] Unconditional Latent Diffusion ( #388 )
...
* initial description
* add doc strings
2022-09-07 14:42:24 +02:00
cdb371f07b
Docs: Conceptual section ( #392 )
...
Add contribution.mdx by copy/pasting and adapting.
2022-09-07 14:41:17 +02:00
8ef1ee812d
[Pipeline Docs] Latent Diffusion ( #377 )
...
* up
* up
* up
* up
* up
* up
* up
2022-09-07 12:53:03 +02:00
ac84c2fa5a
[textual-inversion] fix saving embeds ( #387 )
...
fix saving embeds
2022-09-07 15:49:16 +05:30
5a38033de4
[Docs] Let's go ( #385 )
2022-09-07 11:31:13 +02:00
7bd50cabaf
Add colab links to textual inversion ( #375 )
2022-09-06 22:23:02 +05:30
5c4ea00de7
Efficient Attention ( #366 )
...
* up
* add tests
* correct
* up
* finish
* better naming
* Update README.md
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>
2022-09-06 18:06:47 +02:00
56c003705f
Use `expand` instead of ones to broadcast tensor ( #373 )
...
Use `expand` instead of ones to broadcast tensor.
As suggested by @bes-dev. According the documentation this shouldn't
take any memory - it just plays with the strides.
2022-09-06 17:36:32 +02:00
7a1229fa29
[Tests] Fix SD slow tests ( #364 )
...
move to fp16, update ddim
2022-09-06 17:01:04 +02:00
f085d2f5c6
[Type Hint] VAE models ( #365 )
...
* [Type Hint] VAE models
* Update src/diffusers/models/vae.py
* apply suggestions from code review
Co-authored-by: Anton Lozhkov <aglozhkov@gmail.com>
2022-09-05 19:09:48 +02:00
be52be7215
[Type hint] scheduling lms discrete ( #360 )
...
* [Type hint] scheduling karras ve
* [Type hint] scheduling lms discrete
2022-09-05 18:28:49 +02:00
3c1cdd3359
[Type hint] scheduling karras ve ( #359 )
2022-09-05 18:20:57 +02:00
07f8ebd543
type hints: models/vae.py ( #346 )
...
* type hints: models/vae.py
* modify typings in vae.py
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: Anton Lozhkov <anton@huggingface.co>
2022-09-05 16:46:12 +02:00
ada09bd3f0
[Type Hints] DDIM pipelines ( #345 )
...
* type hints
* Apply suggestions from code review
Co-authored-by: Anton Lozhkov <anton@huggingface.co>
2022-09-05 16:07:37 +02:00
cc59b05635
[ModelOutputs] Replace dict outputs with Dict/Dataclass and allow to return tuples ( #334 )
...
* add outputs for models
* add for pipelines
* finish schedulers
* better naming
* adapt tests as well
* replace dict access with . access
* make schedulers works
* finish
* correct readme
* make bcp compatible
* up
* small fix
* finish
* more fixes
* more fixes
* Apply suggestions from code review
Co-authored-by: Suraj Patil <surajp815@gmail.com>
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>
* Update src/diffusers/models/vae.py
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>
* Adapt model outputs
* Apply more suggestions
* finish examples
* correct
Co-authored-by: Suraj Patil <surajp815@gmail.com>
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>
2022-09-05 14:49:26 +02:00
daddd98b88
package `version` on main should have `.dev0` suffix ( #354 )
...
* package `version` on main should have `.dev0` suffix
package `version` on main should have `.dev0` suffix, which is the convention followed in transformers [here](https://github.com/huggingface/transformers/blob/main/setup.py#L403 )
which will also make the docs built into `main` folder in [doc-build diffusers](https://github.com/huggingface/doc-build/tree/main/diffusers )
* dev version should be incremented
* Update version in `__init__.py`
2022-09-05 11:26:23 +02:00
55d6453fce
[textual_inversion] use tokenizer.add_tokens to add placeholder_token ( #357 )
...
use add_tokens
2022-09-05 13:12:49 +05:30
9ea9c6d1c2
[Type hint] scheduling ddim ( #343 )
...
* [Type hint] scheduling ddim
* apply suggestions from code review
apply suggestions to also return the return type
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
2022-09-04 18:07:54 +02:00
5791f4acde
[Type Hints] VAE models ( #344 )
...
* [Type Hints] VAE models
* apply suggestions from code review
apply suggestions to also return the return type
2022-09-04 18:06:16 +02:00
878af0e113
[Type Hint] DDPM schedulers ( #349 )
2022-09-04 18:05:13 +02:00
dea5ec508f
[Type hint] PNDM schedulers ( #335 )
...
* [Type hint] PNDM Schedulers
* ran make style
* updated timesteps type hint
* apply suggestions from code review
* ran make style
* removed unused import
2022-09-04 18:01:57 +02:00
6c0ca5efa6
Fix typo in unet_blocks.py ( #353 )
...
Update unet_blocks.py
fix typo
2022-09-04 18:01:14 +02:00
cab7650524
Update bug-report.yml
2022-09-04 17:52:56 +02:00
ed8ef6226d
Update bug-report.yml
2022-09-04 17:50:59 +02:00
59c1af77e8
[Commands] Add env command ( #352 )
...
* [Commands] Add env command
* Apply suggestions from code review
2022-09-04 17:43:51 +02:00
fd76845651
Add transformers and scipy to dependency table ( #348 )
...
uP
2022-09-04 09:46:20 +02:00
b1fe170642
[Type Hint] Unet Models ( #330 )
...
* add void check
* remove void, add types for params
2022-09-03 12:31:38 +02:00
9b704f7688
[Img2Img2] Re-add K LMS scheduler ( #340 )
2022-09-03 12:19:58 +02:00
e49dd03d2d
Use ONNX / Core ML compatible method to broadcast ( #310 )
...
* Use ONNX / Core ML compatible method to broadcast.
Unfortunately `tile` could not be used either, it's still not compatible
with ONNX.
See #284 .
* Add comment about why broadcast_to is not used.
Also, apply style to changed files.
* Make sure broadcast remains in same device.
2022-09-02 18:22:57 +02:00
7b628a225a
[Type hint] PNDM pipeline ( #327 )
...
* [Type hint] PNDM pipeline
* ran make style
* Revert "ran make style" wrong black version
2022-09-02 17:45:33 +02:00
033b77ebc4
[Type hint] Latent Diffusion Uncond pipeline ( #333 )
2022-09-02 16:39:34 +02:00
e54206d095
Update README.md
...
Remove joke
2022-09-02 13:20:00 +02:00
6b5baa9332
Add contributions to README and re-order a bit ( #316 )
...
* up
* thanks Clau
* finish
* finish
* up
2022-09-02 13:19:13 +02:00
66fd3ec70d
[CI] try to fix GPU OOMs between tests and excessive tqdm logging ( #323 )
...
* Fix tqdm and OOM
* tqdm auto
* tqdm is still spamming try to disable it altogether
* rather just set the pipe config, to keep the global tqdm clean
* style
2022-09-02 13:18:49 +02:00
3a536ac8f1
README: stable diffusion version v1-3 -> v1-4 ( #331 )
...
Prose: stable diffusion version v1-3 -> v1-4
The code examples use `v1-4`, but the license text was referring to
`v1-3`.
2022-09-02 13:18:09 +02:00
30e7c78ac3
Update README.md
2022-09-02 14:29:27 +05:30
Patrick von Platen