Compare commits

..

56 Commits

Author SHA1 Message Date
sayakpaul
971d0cd6e6 replace with actual fix. 2024-09-23 08:18:50 +05:30
Sayak Paul
aa73072f1f [CI] fix nightly model tests (#9483)
* check if default attn procs fix it.

* print

* print

* replace

* style./

* replace revision with variant.

* replace with stable-diffusion-v1-5/stable-diffusion-inpainting.

* replace with stable-diffusion-v1-5/stable-diffusion-v1-5.

* fix
2024-09-21 07:44:47 +05:30
Aryan
e5d0a328d6 [refactor] LoRA tests (#9481)
* refactor scheduler class usage

* reorder to make tests more readable

* remove pipeline specific checks and skip tests directly

* rewrite denoiser conditions cleaner

* bump tolerance for cog test
2024-09-21 07:10:36 +05:30
Vladimir Mandic
14a1b86fc7 Several fixes to Flux ControlNet pipelines (#9472)
* fix flux controlnet pipelines

---------

Co-authored-by: yiyixuxu <yixu310@gmail.com>
2024-09-19 15:49:36 -10:00
Aryan
2b443a5d62 [training] CogVideoX Lora (#9302)
* cogvideox lora training draft

* update

* update

* update

* update

* update

* make fix-copies

* update

* update

* apply suggestions from review

* apply suggestions from reveiw

* fix typo

* Update examples/cogvideo/train_cogvideox_lora.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* fix lora alpha

* use correct lora scaling for final test pipeline

* Update examples/cogvideo/train_cogvideox_lora.py

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* apply suggestions from review; prodigy optimizer

YiYi Xu <yixu310@gmail.com>

* add tests

* make style

* add README

* update

* update

* make style

* fix

* update

* add test skeleton

* revert lora utils changes

* add cleaner modifications to lora testing utils

* update lora tests

* deepspeed stuff

* add requirements.txt

* deepspeed refactor

* add lora stuff to img2vid pipeline to fix tests

* fight tests

* add co-authors

Co-Authored-By: Fu-Yun Wang <1697256461@qq.com>

Co-Authored-By: zR <2448370773@qq.com>

* fight lora runner tests

* import Dummy optim and scheduler only wheh required

* update docs

* add coauthors

Co-Authored-By: Fu-Yun Wang <1697256461@qq.com>

* remove option to train text encoder

Co-Authored-By: bghira <bghira@users.github.com>

* update tests

* fight more tests

* update

* fix vid2vid

* fix typo

* remove lora tests; todo in follow-up PR

* undo img2vid changes

* remove text encoder related changes in lora loader mixin

* Revert "remove text encoder related changes in lora loader mixin"

This reverts commit f8a8444487.

* update

* round 1 of fighting tests

* round 2 of fighting tests

* fix copied from comment

* fix typo in lora test

* update styling

Co-Authored-By: YiYi Xu <yixu310@gmail.com>

---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: zR <2448370773@qq.com>
Co-authored-by: Fu-Yun Wang <1697256461@qq.com>
Co-authored-by: bghira <bghira@users.github.com>
2024-09-19 14:37:57 +05:30
Sayak Paul
d13b0d63c0 [Flux] add lora integration tests. (#9353)
* add lora integration tests.

* internal note

* add a skip marker.
2024-09-19 09:21:28 +05:30
Anatoly Belikov
5d476f57c5 adapt masked im2im pipeline for SDXL (#7790)
* adapt masked im2im pipeline for SDXL

* usage for masked im2im stable diffusion XL pipeline

* style

* style

* style

---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
2024-09-17 16:55:49 -10:00
Aryan
da18fbd54c set max_shard_size to None for pipeline save_pretrained (#9447)
* update default max_shard_size

* add None check to fix tests

---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
2024-09-17 10:15:18 -10:00
Aryan
ba06124e4a Remove CogVideoX mentions from single file docs; Test updates (#9444)
* remove mentions from single file

* update tests

* update
2024-09-17 10:05:45 -10:00
Subho Ghosh
bb1b0fa1f9 Feature flux controlnet img2img and inpaint pipeline (#9408)
* Implemented FLUX controlnet support to Img2Img pipeline
2024-09-17 09:43:54 -10:00
Linoy Tsaban
8fcfb2a456 [Flux with CFG] add flux pipeline with cfg support (#9445)
* true_cfg

* add check negative prompt/embeds inputs

* move to community pipelines

* move to community pipelines

* revert true cfg changes to the orig pipline

* style

---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
2024-09-16 12:09:34 -10:00
Sayak Paul
5440cbd34e [CI] updates to the CI report naming, and accelerate installation (#9429)
* chore: id accordingly to avoid duplicates.

* update properly.

* updates

* updates

* empty

* updates

* changing order helps?
2024-09-16 11:29:07 -10:00
suzukimain
b52119ae92 [docs] Replace runwayml/stable-diffusion-v1-5 with Lykon/dreamshaper-8 (#9428)
* [docs] Replace runwayml/stable-diffusion-v1-5 with Lykon/dreamshaper-8

Updated documentation as runwayml/stable-diffusion-v1-5 has been removed from Huggingface.

* Update docs/source/en/using-diffusers/inpaint.md

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Replace with stable-diffusion-v1-5/stable-diffusion-v1-5

* Update inpaint.md

---------

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
2024-09-16 10:18:45 -07:00
Yuxuan.Zhang
8336405e50 CogVideoX-5b-I2V support (#9418)
* draft Init

* draft

* vae encode image

* make style

* image latents preparation

* remove image encoder from conversion script

* fix minor bugs

* make pipeline work

* make style

* remove debug prints

* fix imports

* update example

* make fix-copies

* add fast tests

* fix import

* update vae

* update docs

* update image link

* apply suggestions from review

* apply suggestions from review

* add slow test

* make use of learned positional embeddings

* apply suggestions from review

* doc change

* Update convert_cogvideox_to_diffusers.py

* make style

* final changes

* make style

* fix tests

---------

Co-authored-by: Aryan <aryan@huggingface.co>
2024-09-16 14:46:24 +05:30
Sayak Paul
2171f77ac5 [CI] make runner_type restricted. (#9441)
make runner_type restricted.
2024-09-16 12:09:31 +05:30
Aryan
2454b98af4 Allow max shard size to be specified when saving pipeline (#9440)
allow max shard size to be specified when saving pipeline
2024-09-16 08:36:07 +05:30
Linoy Tsaban
37e3603c4a [Flux Dreambooth lora] add latent caching (#9160)
* add ostris trainer to README & add cache latents of vae

* add ostris trainer to README & add cache latents of vae

* style

* readme

* add test for latent caching

* add ostris noise scheduler
9ee1ef2a0a/toolkit/samplers/custom_flowmatch_sampler.py (L95)

* style

* fix import

* style

* fix tests

* style

* --change upcasting of transformer?

* update readme according to main

* keep only latent caching

* add configurable param for final saving of trained layers- --upcast_before_saving

* style

* Update examples/dreambooth/README_flux.md

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

* Update examples/dreambooth/README_flux.md

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

* use clear_objs_and_retain_memory from utilities

* style

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
2024-09-15 15:30:31 +03:00
Leo Jiang
e2ead7cdcc Fix the issue on sd3 dreambooth w./w.t. lora training (#9419)
* Fix dtype error

* [bugfix] Fixed the issue on sd3 dreambooth training

* [bugfix] Fixed the issue on sd3 dreambooth training

---------

Co-authored-by: 蒋硕 <jiangshuo9@h-partners.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
2024-09-14 16:29:38 +05:30
Benjamin Bossan
48e36353d8 MAINT Permission for GH token in stale.yml (#9427)
* MAINT Permission for GH token in stale.yml

See https://github.com/huggingface/peft/pull/2061 for the equivalent PR
in PEFT.

This restores the functionality of the stale bot after permissions for
the token have been limited. The action still shows errors for PEFT but
the bot appears to work fine.

* Also add write permissions for PRs

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
2024-09-13 21:00:57 +05:30
Sayak Paul
6dc6486565 [LoRA] fix adapter movement when using DoRA. (#9411)
fix adapter movement when using DoRA.
2024-09-13 07:31:53 +05:30
Dhruv Nair
1e8cf2763d [CI] Nightly Test Updates (#9380)
* update

* update

* update

* update

* update

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
2024-09-12 20:21:28 +05:30
Sayak Paul
6cf8d98ce1 [CI] update artifact uploader version (#9426)
update artifact uploader version
2024-09-12 19:26:09 +05:30
Juan Acevedo
45aa8bb187 Ptxla sd training (#9381)
* enable pxla training of stable diffusion 2.x models.

* run linter/style and run pipeline test for stable diffusion and fix issues.

* update xla libraries

* fix read me newline.

* move files to research folder.

* update per comments.

* rename readme.

---------

Co-authored-by: Juan Acevedo <jfacevedo@google.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
2024-09-12 08:35:06 +05:30
Aryan
5e1427a7da [docs] AnimateDiff FreeNoise (#9414)
* update docs

* apply suggestions from review

* Update docs/source/en/api/pipelines/animatediff.md

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update docs/source/en/api/pipelines/animatediff.md

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update docs/source/en/api/pipelines/animatediff.md

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* apply suggestions from review

---------

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
2024-09-11 12:59:58 -07:00
asfiyab-nvidia
b9e2f886cd FluxPosEmbed: Remove Squeeze No-op (#9409)
Remove Squeeze op

Signed-off-by: Asfiya Baig <asfiyab@nvidia.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
2024-09-10 19:12:36 -10:00
dianyo
b19827f6b4 Migrate the BrownianTree to BrownianInterval in DPM solver (#9335)
migrate the BrownianTree to BrownianInterval

Co-authored-by: YiYi Xu <yixu310@gmail.com>
2024-09-10 18:29:15 -10:00
Yu Zheng
c002731d93 [examples] add controlnet sd3 example (#9249)
* add controlnet sd3 example

* add controlnet sd3 example

* update controlnet sd3 example

* add controlnet sd3 example test

* fix quality and style

* update test

* update test

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
2024-09-11 07:04:37 +05:30
Sayak Paul
adf1f911f0 [Tests] fix some fast gpu tests. (#9379)
fix some fast gpu tests.
2024-09-11 06:50:02 +05:30
captainzz
f28a8c257a fix from_transformer() with extra conditioning channels (#9364)
* fix from_transformer() with extra conditioning channels

* style fix

---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: Álvaro Somoza <somoza.alvaro@gmail.com>
2024-09-09 07:51:48 -10:00
Jinzhe Pan
2c6a6c97b3 [docs] Add xDiT in section optimization (#9365)
* docs: add xDiT to optimization methods

* fix: picture layout problem

* docs: add more introduction about xdit & apply suggestions

* Apply suggestions from code review

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

---------

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
2024-09-09 10:31:07 -07:00
Igor Filippov
a7361dccdc [Pipeline] animatediff + vid2vid + controlnet (#9337)
* add animatediff + vid2vide + controlnet

* post tests fixes

* PR discussion fixes

* update docs

* change input video to links on HF + update an example

* make quality fix

* fix ip adapter test

* fix ip adapter test input

* update ip adapter test
2024-09-09 22:48:21 +05:30
YiYi Xu
485b8bb000 refactor get_timesteps for SDXL img2img + add set_begin_index (#9375)
* refator + add begin_index

* add kolors img2img to doc
2024-09-09 06:38:22 -10:00
Sayak Paul
d08ad65819 modify benchmarks to replace sdv1.5 with dreamshaper. (#9334) 2024-09-09 20:54:56 +05:30
YiYi Xu
8cdcdd9e32 add flux inpaint + img2img + controlnet to auto pipeline (#9367) 2024-09-06 07:14:48 -10:00
Dhruv Nair
d269cc8a4e [CI] Quick fix for Cog Video Test (#9373)
update
2024-09-06 15:25:53 +05:30
Aryan
6dfa49963c [core] Freenoise memory improvements (#9262)
* update

* implement prompt interpolation

* make style

* resnet memory optimizations

* more memory optimizations; todo: refactor

* update

* update animatediff controlnet with latest changes

* refactor chunked inference changes

* remove print statements

* update

* chunk -> split

* remove changes from incorrect conflict resolution

* remove changes from incorrect conflict resolution

* add explanation of SplitInferenceModule

* update docs

* Revert "update docs"

This reverts commit c55a50a271.

* update docstring for freenoise split inference

* apply suggestions from review

* add tests

* apply suggestions from review
2024-09-06 12:51:20 +05:30
Haruya Ishikawa
5249a2666e fix one uncaught deprecation warning for accessing vae_latent_channels in VaeImagePreprocessor (#9372)
deprecation warning vae_latent_channels
2024-09-05 07:32:27 -10:00
Linoy Tsaban
55ac421f7b improve README for flux dreambooth lora (#9290)
* improve readme

* improve readme

* improve readme

* improve readme
2024-09-05 17:53:23 +05:30
Dhruv Nair
53051cf282 [CI] Update Single file Nightly Tests (#9357)
* update

* update
2024-09-05 14:33:44 +05:30
Tolga Cangöz
3000551729 Update UNet2DConditionModel's error messages (#9230)
* refactor
2024-09-04 10:49:56 -10:00
Vishnu V Jaddipal
249a9e48e8 Add Flux inpainting and Flux Img2Img (#9135)
---------

Co-authored-by: yiyixuxu <yixu310@gmail.com>
2024-09-04 10:31:43 -10:00
Fanli Lin
2ee3215949 [tests] make 2 tests device-agnostic (#9347)
* enabel on xpu

* fix style
2024-09-03 16:34:03 -10:00
Eduardo Escobar
8ecf499d8b Enable load_lora_weights for StableDiffusion3InpaintPipeline (#9330)
Enable load_lora_weights for StableDiffusion3InpaintPipeline

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
2024-09-03 15:19:37 -10:00
YiYi Xu
dcf320f293 small update on rotary embedding (#9354)
* update

* fix

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
2024-09-03 07:18:33 -10:00
Sayak Paul
8ba90aa706 chore: add a cleaning utility to be useful during training. (#9240) 2024-09-03 15:00:17 +05:30
Aryan
9d49b45b19 [refactor] move positional embeddings to patch embed layer for CogVideoX (#9263)
* remove frame limit in cogvideox

* remove debug prints

* Update src/diffusers/models/transformers/cogvideox_transformer_3d.py

* revert pipeline; remove frame limitation

* revert transformer changes

* address review comments

* add error message

* apply suggestions from review
2024-09-03 14:45:12 +05:30
Dhruv Nair
81da2e1c95 [CI] Add option to dispatch Fast GPU tests on main (#9355)
update
2024-09-03 14:35:13 +05:30
Aryan
24053832b5 [tests] remove/speedup some low signal tests (#9285)
* remove 2 shapes from SDFunctionTesterMixin::test_vae_tiling

* combine freeu enable/disable test to reduce many inference runs

* remove low signal unet test for signature

* remove low signal embeddings test

* remove low signal progress bar test from PipelineTesterMixin

* combine ip-adapter single and multi tests to save many inferences

* fix broken tests

* Update tests/pipelines/test_pipelines_common.py

* Update tests/pipelines/test_pipelines_common.py

* add progress bar tests
2024-09-03 13:59:18 +05:30
Dhruv Nair
f6f16a0c11 [CI] More Fast GPU Test Fixes (#9346)
* update

* update

* update

* update
2024-09-03 13:22:38 +05:30
Vishnu V Jaddipal
1c1ccaa03f Xlabs lora fix (#9348)
* Fix ```from_single_file``` for xl_inpaint

* Add basic flux inpaint pipeline

* style, quality, stray print

* Fix stray changes

* Add inpainting model support

* Change lora conversion for xlabs

* Fix stray changes

* Apply suggestions from code review

* style

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
2024-09-03 10:43:43 +05:30
Dhruv Nair
007ad0e2aa [CI] More fixes for Fast GPU Tests on main (#9300)
update
2024-09-02 17:51:48 +05:30
Aryan
0e6a8403f6 [core] Support VideoToVideo with CogVideoX (#9333)
* add vid2vid pipeline for cogvideox

* make fix-copies

* update docs

* fake context parallel cache, vae encode tiling

* add test for cog vid2vid

* use video link from HF docs repo

* add copied from comments; correctly rename test class
2024-09-02 16:54:58 +05:30
Aryan
af6c0fb766 [core] CogVideoX memory optimizations in VAE encode (#9340)
fake context parallel cache, vae encode tiling

(cherry picked from commit bf890bca0e)
2024-09-02 15:48:37 +05:30
YiYi Xu
d8a16635f4 update runway repo for single_file (#9323)
update to a place holder
2024-08-30 08:51:21 -10:00
Aryan
e417d02811 [docs] Add a note on torchao/quanto benchmarks for CogVideoX and memory-efficient inference (#9296)
* add a note on torchao/quanto benchmarks and memory-efficient inference

* apply suggestions from review

* update

* Update docs/source/en/api/pipelines/cogvideox.md

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update docs/source/en/api/pipelines/cogvideox.md

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* add note on enable sequential cpu offload

---------

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
2024-08-30 13:53:25 +05:30
Dhruv Nair
1d4d71875b [CI] Update Hub Token on nightly tests (#9318)
update
2024-08-30 10:23:50 +05:30
265 changed files with 18487 additions and 2392 deletions

View File

@@ -50,7 +50,7 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: benchmark_test_reports
path: benchmarks/benchmark_outputs

View File

@@ -43,7 +43,7 @@ jobs:
- name: Pipeline Tests Artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: test-pipelines.json
path: reports
@@ -72,14 +72,14 @@ jobs:
run: |
python -m venv /opt/venv && export PATH="/opt/venv/bin:$PATH"
python -m uv pip install -e [quality,test]
python -m uv pip install accelerate@git+https://github.com/huggingface/accelerate.git
pip uninstall accelerate -y && python -m uv pip install -U accelerate@git+https://github.com/huggingface/accelerate.git
python -m uv pip install pytest-reportlog
- name: Environment
run: |
python utils/print_env.py
- name: Pipeline CUDA Test
env:
HF_TOKEN: ${{ secrets.HF_TOKEN }}
HF_TOKEN: ${{ secrets.DIFFUSERS_HF_HUB_READ_TOKEN }}
# https://pytorch.org/docs/stable/notes/randomness.html#avoiding-nondeterministic-algorithms
CUBLAS_WORKSPACE_CONFIG: :16:8
run: |
@@ -95,7 +95,7 @@ jobs:
cat reports/tests_pipeline_${{ matrix.module }}_cuda_failures_short.txt
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: pipeline_${{ matrix.module }}_test_reports
path: reports
@@ -130,8 +130,8 @@ jobs:
run: |
python -m venv /opt/venv && export PATH="/opt/venv/bin:$PATH"
python -m uv pip install -e [quality,test]
python -m uv pip install accelerate@git+https://github.com/huggingface/accelerate.git
python -m uv pip install peft@git+https://github.com/huggingface/peft.git
pip uninstall accelerate -y && python -m uv pip install -U accelerate@git+https://github.com/huggingface/accelerate.git
python -m uv pip install pytest-reportlog
- name: Environment
run: python utils/print_env.py
@@ -139,7 +139,7 @@ jobs:
- name: Run nightly PyTorch CUDA tests for non-pipeline modules
if: ${{ matrix.module != 'examples'}}
env:
HF_TOKEN: ${{ secrets.HF_TOKEN }}
HF_TOKEN: ${{ secrets.DIFFUSERS_HF_HUB_READ_TOKEN }}
# https://pytorch.org/docs/stable/notes/randomness.html#avoiding-nondeterministic-algorithms
CUBLAS_WORKSPACE_CONFIG: :16:8
run: |
@@ -152,7 +152,7 @@ jobs:
- name: Run nightly example tests with Torch
if: ${{ matrix.module == 'examples' }}
env:
HF_TOKEN: ${{ secrets.HF_TOKEN }}
HF_TOKEN: ${{ secrets.DIFFUSERS_HF_HUB_READ_TOKEN }}
# https://pytorch.org/docs/stable/notes/randomness.html#avoiding-nondeterministic-algorithms
CUBLAS_WORKSPACE_CONFIG: :16:8
run: |
@@ -169,7 +169,7 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: torch_${{ matrix.module }}_cuda_test_reports
path: reports
@@ -201,7 +201,7 @@ jobs:
run: |
python -m venv /opt/venv && export PATH="/opt/venv/bin:$PATH"
python -m uv pip install -e [quality,test]
python -m uv pip install accelerate@git+https://github.com/huggingface/accelerate.git
pip uninstall accelerate -y && python -m uv pip install -U accelerate@git+https://github.com/huggingface/accelerate.git
python -m uv pip install pytest-reportlog
- name: Environment
@@ -209,7 +209,7 @@ jobs:
- name: Run nightly Flax TPU tests
env:
HF_TOKEN: ${{ secrets.HF_TOKEN }}
HF_TOKEN: ${{ secrets.DIFFUSERS_HF_HUB_READ_TOKEN }}
run: |
python -m pytest -n 0 \
-s -v -k "Flax" \
@@ -225,7 +225,7 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: flax_tpu_test_reports
path: reports
@@ -257,14 +257,14 @@ jobs:
run: |
python -m venv /opt/venv && export PATH="/opt/venv/bin:$PATH"
python -m uv pip install -e [quality,test]
python -m uv pip install accelerate@git+https://github.com/huggingface/accelerate.git
pip uninstall accelerate -y && python -m uv pip install -U accelerate@git+https://github.com/huggingface/accelerate.git
python -m uv pip install pytest-reportlog
- name: Environment
run: python utils/print_env.py
- name: Run Nightly ONNXRuntime CUDA tests
env:
HF_TOKEN: ${{ secrets.HF_TOKEN }}
HF_TOKEN: ${{ secrets.DIFFUSERS_HF_HUB_READ_TOKEN }}
run: |
python -m pytest -n 1 --max-worker-restart=0 --dist=loadfile \
-s -v -k "Onnx" \
@@ -280,9 +280,9 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: ${{ matrix.config.report }}_test_reports
name: tests_onnx_cuda_reports
path: reports
- name: Generate Report and Notify Channel
@@ -340,7 +340,7 @@ jobs:
#
# - name: Test suite reports artifacts
# if: ${{ always() }}
# uses: actions/upload-artifact@v2
# uses: actions/upload-artifact@v4
# with:
# name: torch_mps_test_reports
# path: reports
@@ -396,7 +396,7 @@ jobs:
#
# - name: Test suite reports artifacts
# if: ${{ always() }}
# uses: actions/upload-artifact@v2
# uses: actions/upload-artifact@v4
# with:
# name: torch_mps_test_reports
# path: reports

View File

@@ -171,7 +171,7 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: pr_${{ matrix.config.report }}_test_reports
path: reports

View File

@@ -95,7 +95,7 @@ jobs:
if [ "${{ matrix.lib-versions }}" == "main" ]; then
python -m pip install -U peft@git+https://github.com/huggingface/peft.git
python -m uv pip install -U transformers@git+https://github.com/huggingface/transformers.git
python -m uv pip install -U accelerate@git+https://github.com/huggingface/accelerate.git
pip uninstall accelerate -y && python -m uv pip install -U accelerate@git+https://github.com/huggingface/accelerate.git
else
python -m uv pip install -U peft transformers accelerate
fi
@@ -110,23 +110,23 @@ jobs:
python -m venv /opt/venv && export PATH="/opt/venv/bin:$PATH"
python -m pytest -n 4 --max-worker-restart=0 --dist=loadfile \
-s -v \
--make-reports=tests_${{ matrix.config.report }} \
--make-reports=tests_${{ matrix.lib-versions }} \
tests/lora/
python -m pytest -n 4 --max-worker-restart=0 --dist=loadfile \
-s -v \
--make-reports=tests_models_lora_${{ matrix.config.report }} \
--make-reports=tests_models_lora_${{ matrix.lib-versions }} \
tests/models/ -k "lora"
- name: Failure short reports
if: ${{ failure() }}
run: |
cat reports/tests_${{ matrix.config.report }}_failures_short.txt
cat reports/tests_models_lora_${{ matrix.config.report }}_failures_short.txt
cat reports/tests_${{ matrix.lib-versions }}_failures_short.txt
cat reports/tests_models_lora_${{ matrix.lib-versions }}_failures_short.txt
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: pr_${{ matrix.config.report }}_test_reports
name: pr_${{ matrix.lib-versions }}_test_reports
path: reports

View File

@@ -168,9 +168,9 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: pr_${{ matrix.config.report }}_test_reports
name: pr_${{ matrix.config.framework }}_${{ matrix.config.report }}_test_reports
path: reports
run_staging_tests:
@@ -229,7 +229,7 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: pr_${{ matrix.config.report }}_test_reports
path: reports

View File

@@ -1,6 +1,7 @@
name: Fast GPU Tests on main
on:
workflow_dispatch:
push:
branches:
- main
@@ -45,7 +46,7 @@ jobs:
echo "pipeline_test_matrix=$matrix" >> $GITHUB_OUTPUT
- name: Pipeline Tests Artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: test-pipelines.json
path: reports
@@ -75,7 +76,7 @@ jobs:
run: |
python -m venv /opt/venv && export PATH="/opt/venv/bin:$PATH"
python -m uv pip install -e [quality,test]
python -m uv pip install accelerate@git+https://github.com/huggingface/accelerate.git
pip uninstall accelerate -y && python -m uv pip install -U accelerate@git+https://github.com/huggingface/accelerate.git
- name: Environment
run: |
python utils/print_env.py
@@ -96,7 +97,7 @@ jobs:
cat reports/tests_pipeline_${{ matrix.module }}_cuda_failures_short.txt
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: pipeline_${{ matrix.module }}_test_reports
path: reports
@@ -126,8 +127,8 @@ jobs:
run: |
python -m venv /opt/venv && export PATH="/opt/venv/bin:$PATH"
python -m uv pip install -e [quality,test]
python -m uv pip install accelerate@git+https://github.com/huggingface/accelerate.git
python -m uv pip install peft@git+https://github.com/huggingface/peft.git
pip uninstall accelerate -y && python -m uv pip install -U accelerate@git+https://github.com/huggingface/accelerate.git
- name: Environment
run: |
@@ -141,20 +142,20 @@ jobs:
run: |
python -m pytest -n 1 --max-worker-restart=0 --dist=loadfile \
-s -v -k "not Flax and not Onnx" \
--make-reports=tests_torch_cuda \
--make-reports=tests_torch_cuda_${{ matrix.module }} \
tests/${{ matrix.module }}
- name: Failure short reports
if: ${{ failure() }}
run: |
cat reports/tests_torch_cuda_stats.txt
cat reports/tests_torch_cuda_failures_short.txt
cat reports/tests_torch_cuda_${{ matrix.module }}_stats.txt
cat reports/tests_torch_cuda_${{ matrix.module }}_failures_short.txt
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: torch_cuda_test_reports
name: torch_cuda_test_reports_${{ matrix.module }}
path: reports
flax_tpu_tests:
@@ -176,7 +177,7 @@ jobs:
run: |
python -m venv /opt/venv && export PATH="/opt/venv/bin:$PATH"
python -m uv pip install -e [quality,test]
python -m uv pip install accelerate@git+https://github.com/huggingface/accelerate.git
pip uninstall accelerate -y && python -m uv pip install -U accelerate@git+https://github.com/huggingface/accelerate.git
- name: Environment
run: |
@@ -199,7 +200,7 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: flax_tpu_test_reports
path: reports
@@ -224,7 +225,7 @@ jobs:
run: |
python -m venv /opt/venv && export PATH="/opt/venv/bin:$PATH"
python -m uv pip install -e [quality,test]
python -m uv pip install accelerate@git+https://github.com/huggingface/accelerate.git
pip uninstall accelerate -y && python -m uv pip install -U accelerate@git+https://github.com/huggingface/accelerate.git
- name: Environment
run: |
@@ -247,7 +248,7 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: onnx_cuda_test_reports
path: reports
@@ -290,7 +291,7 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: torch_compile_test_reports
path: reports
@@ -332,7 +333,7 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: torch_xformers_test_reports
path: reports
@@ -383,7 +384,7 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: examples_test_reports
path: reports

View File

@@ -119,7 +119,7 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: pr_${{ matrix.config.report }}_test_reports
path: reports

View File

@@ -69,7 +69,7 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: pr_torch_mps_test_reports
path: reports

View File

@@ -45,7 +45,7 @@ jobs:
echo "pipeline_test_matrix=$matrix" >> $GITHUB_OUTPUT
- name: Pipeline Tests Artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: test-pipelines.json
path: reports
@@ -75,7 +75,7 @@ jobs:
run: |
python -m venv /opt/venv && export PATH="/opt/venv/bin:$PATH"
python -m uv pip install -e [quality,test]
python -m uv pip install accelerate@git+https://github.com/huggingface/accelerate.git
pip uninstall accelerate -y && python -m uv pip install -U accelerate@git+https://github.com/huggingface/accelerate.git
- name: Environment
run: |
python utils/print_env.py
@@ -96,7 +96,7 @@ jobs:
cat reports/tests_pipeline_${{ matrix.module }}_cuda_failures_short.txt
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: pipeline_${{ matrix.module }}_test_reports
path: reports
@@ -126,8 +126,8 @@ jobs:
run: |
python -m venv /opt/venv && export PATH="/opt/venv/bin:$PATH"
python -m uv pip install -e [quality,test]
python -m uv pip install accelerate@git+https://github.com/huggingface/accelerate.git
python -m uv pip install peft@git+https://github.com/huggingface/peft.git
pip uninstall accelerate -y && python -m uv pip install -U accelerate@git+https://github.com/huggingface/accelerate.git
- name: Environment
run: |
@@ -141,20 +141,20 @@ jobs:
run: |
python -m pytest -n 1 --max-worker-restart=0 --dist=loadfile \
-s -v -k "not Flax and not Onnx" \
--make-reports=tests_torch_cuda \
--make-reports=tests_torch_${{ matrix.module }}_cuda \
tests/${{ matrix.module }}
- name: Failure short reports
if: ${{ failure() }}
run: |
cat reports/tests_torch_cuda_stats.txt
cat reports/tests_torch_cuda_failures_short.txt
cat reports/tests_torch_${{ matrix.module }}_cuda_stats.txt
cat reports/tests_torch_${{ matrix.module }}_cuda_failures_short.txt
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: torch_cuda_test_reports
name: torch_cuda_${{ matrix.module }}_test_reports
path: reports
flax_tpu_tests:
@@ -176,7 +176,7 @@ jobs:
run: |
python -m venv /opt/venv && export PATH="/opt/venv/bin:$PATH"
python -m uv pip install -e [quality,test]
python -m uv pip install accelerate@git+https://github.com/huggingface/accelerate.git
pip uninstall accelerate -y && python -m uv pip install -U accelerate@git+https://github.com/huggingface/accelerate.git
- name: Environment
run: |
@@ -199,7 +199,7 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: flax_tpu_test_reports
path: reports
@@ -224,7 +224,7 @@ jobs:
run: |
python -m venv /opt/venv && export PATH="/opt/venv/bin:$PATH"
python -m uv pip install -e [quality,test]
python -m uv pip install accelerate@git+https://github.com/huggingface/accelerate.git
pip uninstall accelerate -y && python -m uv pip install -U accelerate@git+https://github.com/huggingface/accelerate.git
- name: Environment
run: |
@@ -247,7 +247,7 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: onnx_cuda_test_reports
path: reports
@@ -290,7 +290,7 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: torch_compile_test_reports
path: reports
@@ -332,7 +332,7 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: torch_xformers_test_reports
path: reports
@@ -383,7 +383,7 @@ jobs:
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v2
uses: actions/upload-artifact@v4
with:
name: examples_test_reports
path: reports

View File

@@ -4,8 +4,12 @@ on:
workflow_dispatch:
inputs:
runner_type:
description: 'Type of runner to test (a10 or t4)'
description: 'Type of runner to test (aws-g6-4xlarge-plus: a10 or aws-g4dn-2xlarge: t4)'
type: choice
required: true
options:
- aws-g6-4xlarge-plus
- aws-g4dn-2xlarge
docker_image:
description: 'Name of the Docker image'
required: true

View File

@@ -9,6 +9,9 @@ jobs:
name: Close Stale Issues
if: github.repository == 'huggingface/diffusers'
runs-on: ubuntu-latest
permissions:
issues: write
pull-requests: write
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
steps:

View File

@@ -65,7 +65,7 @@ Pipelines are designed to be easy to use (therefore do not follow [*Simple over
The following design principles are followed:
- Pipelines follow the single-file policy. All pipelines can be found in individual directories under src/diffusers/pipelines. One pipeline folder corresponds to one diffusion paper/project/release. Multiple pipeline files can be gathered in one pipeline folder, as its done for [`src/diffusers/pipelines/stable-diffusion`](https://github.com/huggingface/diffusers/tree/main/src/diffusers/pipelines/stable_diffusion). If pipelines share similar functionality, one can make use of the [# Copied from mechanism](https://github.com/huggingface/diffusers/blob/125d783076e5bd9785beb05367a2d2566843a271/src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_img2img.py#L251).
- Pipelines all inherit from [`DiffusionPipeline`].
- Every pipeline consists of different model and scheduler components, that are documented in the [`model_index.json` file](https://huggingface.co/runwayml/stable-diffusion-v1-5/blob/main/model_index.json), are accessible under the same name as attributes of the pipeline and can be shared between pipelines with [`DiffusionPipeline.components`](https://huggingface.co/docs/diffusers/main/en/api/diffusion_pipeline#diffusers.DiffusionPipeline.components) function.
- Every pipeline consists of different model and scheduler components, that are documented in the [`model_index.json` file](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5/blob/main/model_index.json), are accessible under the same name as attributes of the pipeline and can be shared between pipelines with [`DiffusionPipeline.components`](https://huggingface.co/docs/diffusers/main/en/api/diffusion_pipeline#diffusers.DiffusionPipeline.components) function.
- Every pipeline should be loadable via the [`DiffusionPipeline.from_pretrained`](https://huggingface.co/docs/diffusers/main/en/api/diffusion_pipeline#diffusers.DiffusionPipeline.from_pretrained) function.
- Pipelines should be used **only** for inference.
- Pipelines should be very readable, self-explanatory, and easy to tweak.

View File

@@ -73,7 +73,7 @@ Generating outputs is super easy with 🤗 Diffusers. To generate an image from
from diffusers import DiffusionPipeline
import torch
pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16)
pipeline = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16)
pipeline.to("cuda")
pipeline("An image of a squirrel in Picasso style").images[0]
```
@@ -144,7 +144,7 @@ Also, say 👋 in our public Discord channel <a href="https://discord.gg/G7tWnz9
<tr style="border-top: 2px solid black">
<td>Text-to-Image</td>
<td><a href="https://huggingface.co/docs/diffusers/api/pipelines/stable_diffusion/text2img">Stable Diffusion Text-to-Image</a></td>
<td><a href="https://huggingface.co/runwayml/stable-diffusion-v1-5"> runwayml/stable-diffusion-v1-5 </a></td>
<td><a href="https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5"> stable-diffusion-v1-5/stable-diffusion-v1-5 </a></td>
</tr>
<tr>
<td>Text-to-Image</td>
@@ -174,7 +174,7 @@ Also, say 👋 in our public Discord channel <a href="https://discord.gg/G7tWnz9
<tr>
<td>Text-guided Image-to-Image</td>
<td><a href="https://huggingface.co/docs/diffusers/api/pipelines/stable_diffusion/img2img">Stable Diffusion Image-to-Image</a></td>
<td><a href="https://huggingface.co/runwayml/stable-diffusion-v1-5"> runwayml/stable-diffusion-v1-5 </a></td>
<td><a href="https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5"> stable-diffusion-v1-5/stable-diffusion-v1-5 </a></td>
</tr>
<tr style="border-top: 2px solid black">
<td>Text-guided Image Inpainting</td>

View File

@@ -34,7 +34,7 @@ from utils import ( # noqa: E402
RESOLUTION_MAPPING = {
"runwayml/stable-diffusion-v1-5": (512, 512),
"Lykon/DreamShaper": (512, 512),
"lllyasviel/sd-controlnet-canny": (512, 512),
"diffusers/controlnet-canny-sdxl-1.0": (1024, 1024),
"TencentARC/t2iadapter_canny_sd14v1": (512, 512),
@@ -268,7 +268,7 @@ class IPAdapterTextToImageBenchmark(TextToImageBenchmark):
class ControlNetBenchmark(TextToImageBenchmark):
pipeline_class = StableDiffusionControlNetPipeline
aux_network_class = ControlNetModel
root_ckpt = "runwayml/stable-diffusion-v1-5"
root_ckpt = "Lykon/DreamShaper"
url = "https://huggingface.co/datasets/diffusers/docs-images/resolve/main/benchmarking/canny_image_condition.png"
image = load_image(url).convert("RGB")
@@ -311,7 +311,7 @@ class ControlNetSDXLBenchmark(ControlNetBenchmark):
class T2IAdapterBenchmark(ControlNetBenchmark):
pipeline_class = StableDiffusionAdapterPipeline
aux_network_class = T2IAdapter
root_ckpt = "CompVis/stable-diffusion-v1-4"
root_ckpt = "Lykon/DreamShaper"
url = "https://huggingface.co/datasets/diffusers/docs-images/resolve/main/benchmarking/canny_for_adapter.png"
image = load_image(url).convert("L")

View File

@@ -7,7 +7,8 @@ from base_classes import IPAdapterTextToImageBenchmark # noqa: E402
IP_ADAPTER_CKPTS = {
"runwayml/stable-diffusion-v1-5": ("h94/IP-Adapter", "ip-adapter_sd15.bin"),
# because original SD v1.5 has been taken down.
"Lykon/DreamShaper": ("h94/IP-Adapter", "ip-adapter_sd15.bin"),
"stabilityai/stable-diffusion-xl-base-1.0": ("h94/IP-Adapter", "ip-adapter_sdxl.bin"),
}
@@ -17,7 +18,7 @@ if __name__ == "__main__":
parser.add_argument(
"--ckpt",
type=str,
default="runwayml/stable-diffusion-v1-5",
default="rstabilityai/stable-diffusion-xl-base-1.0",
choices=list(IP_ADAPTER_CKPTS.keys()),
)
parser.add_argument("--batch_size", type=int, default=1)

View File

@@ -11,9 +11,9 @@ if __name__ == "__main__":
parser.add_argument(
"--ckpt",
type=str,
default="runwayml/stable-diffusion-v1-5",
default="Lykon/DreamShaper",
choices=[
"runwayml/stable-diffusion-v1-5",
"Lykon/DreamShaper",
"stabilityai/stable-diffusion-2-1",
"stabilityai/stable-diffusion-xl-refiner-1.0",
"stabilityai/sdxl-turbo",

View File

@@ -11,9 +11,9 @@ if __name__ == "__main__":
parser.add_argument(
"--ckpt",
type=str,
default="runwayml/stable-diffusion-v1-5",
default="Lykon/DreamShaper",
choices=[
"runwayml/stable-diffusion-v1-5",
"Lykon/DreamShaper",
"stabilityai/stable-diffusion-2-1",
"stabilityai/stable-diffusion-xl-base-1.0",
],

View File

@@ -7,7 +7,7 @@ from base_classes import TextToImageBenchmark, TurboTextToImageBenchmark # noqa
ALL_T2I_CKPTS = [
"runwayml/stable-diffusion-v1-5",
"Lykon/DreamShaper",
"segmind/SSD-1B",
"stabilityai/stable-diffusion-xl-base-1.0",
"kandinsky-community/kandinsky-2-2-decoder",
@@ -21,7 +21,7 @@ if __name__ == "__main__":
parser.add_argument(
"--ckpt",
type=str,
default="runwayml/stable-diffusion-v1-5",
default="Lykon/DreamShaper",
choices=ALL_T2I_CKPTS,
)
parser.add_argument("--batch_size", type=int, default=1)

View File

@@ -161,6 +161,8 @@
title: DeepCache
- local: optimization/tgate
title: TGATE
- local: optimization/xdit
title: xDiT
- sections:
- local: using-diffusers/stable_diffusion_jax_how_to
title: JAX/Flax

View File

@@ -22,7 +22,6 @@ The [`~loaders.FromSingleFileMixin.from_single_file`] method allows you to load:
## Supported pipelines
- [`CogVideoXPipeline`]
- [`StableDiffusionPipeline`]
- [`StableDiffusionImg2ImgPipeline`]
- [`StableDiffusionInpaintPipeline`]
@@ -50,7 +49,6 @@ The [`~loaders.FromSingleFileMixin.from_single_file`] method allows you to load:
- [`UNet2DConditionModel`]
- [`StableCascadeUNet`]
- [`AutoencoderKL`]
- [`AutoencoderKLCogVideoX`]
- [`ControlNetModel`]
- [`SD3Transformer2DModel`]
- [`FluxTransformer2DModel`]

View File

@@ -29,7 +29,7 @@ from diffusers import StableDiffusionControlNetPipeline, ControlNetModel
url = "https://huggingface.co/lllyasviel/ControlNet-v1-1/blob/main/control_v11p_sd15_canny.pth" # can also be a local path
controlnet = ControlNetModel.from_single_file(url)
url = "https://huggingface.co/runwayml/stable-diffusion-v1-5/blob/main/v1-5-pruned.safetensors" # can also be a local path
url = "https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5/blob/main/v1-5-pruned.safetensors" # can also be a local path
pipe = StableDiffusionControlNetPipeline.from_single_file(url, controlnet=controlnet)
```

View File

@@ -29,6 +29,7 @@ The abstract of the paper is the following:
| [AnimateDiffSparseControlNetPipeline](https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/animatediff/pipeline_animatediff_sparsectrl.py) | *Controlled Video-to-Video Generation with AnimateDiff using SparseCtrl* |
| [AnimateDiffSDXLPipeline](https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/animatediff/pipeline_animatediff_sdxl.py) | *Video-to-Video Generation with AnimateDiff* |
| [AnimateDiffVideoToVideoPipeline](https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/animatediff/pipeline_animatediff_video2video.py) | *Video-to-Video Generation with AnimateDiff* |
| [AnimateDiffVideoToVideoControlNetPipeline](https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/animatediff/pipeline_animatediff_video2video_controlnet.py) | *Video-to-Video Generation with AnimateDiff using ControlNet* |
## Available checkpoints
@@ -518,6 +519,97 @@ Here are some sample outputs:
</tr>
</table>
### AnimateDiffVideoToVideoControlNetPipeline
AnimateDiff can be used together with ControlNets to enhance video-to-video generation by allowing for precise control over the output. ControlNet was introduced in [Adding Conditional Control to Text-to-Image Diffusion Models](https://huggingface.co/papers/2302.05543) by Lvmin Zhang, Anyi Rao, and Maneesh Agrawala, and allows you to condition Stable Diffusion with an additional control image to ensure that the spatial information is preserved throughout the video.
This pipeline allows you to condition your generation both on the original video and on a sequence of control images.
```python
import torch
from PIL import Image
from tqdm.auto import tqdm
from controlnet_aux.processor import OpenposeDetector
from diffusers import AnimateDiffVideoToVideoControlNetPipeline
from diffusers.utils import export_to_gif, load_video
from diffusers import AutoencoderKL, ControlNetModel, MotionAdapter, LCMScheduler
# Load the ControlNet
controlnet = ControlNetModel.from_pretrained("lllyasviel/sd-controlnet-openpose", torch_dtype=torch.float16)
# Load the motion adapter
motion_adapter = MotionAdapter.from_pretrained("wangfuyun/AnimateLCM")
# Load SD 1.5 based finetuned model
vae = AutoencoderKL.from_pretrained("stabilityai/sd-vae-ft-mse", torch_dtype=torch.float16)
pipe = AnimateDiffVideoToVideoControlNetPipeline.from_pretrained(
"SG161222/Realistic_Vision_V5.1_noVAE",
motion_adapter=motion_adapter,
controlnet=controlnet,
vae=vae,
).to(device="cuda", dtype=torch.float16)
# Enable LCM to speed up inference
pipe.scheduler = LCMScheduler.from_config(pipe.scheduler.config, beta_schedule="linear")
pipe.load_lora_weights("wangfuyun/AnimateLCM", weight_name="AnimateLCM_sd15_t2v_lora.safetensors", adapter_name="lcm-lora")
pipe.set_adapters(["lcm-lora"], [0.8])
video = load_video("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/dance.gif")
video = [frame.convert("RGB") for frame in video]
prompt = "astronaut in space, dancing"
negative_prompt = "bad quality, worst quality, jpeg artifacts, ugly"
# Create controlnet preprocessor
open_pose = OpenposeDetector.from_pretrained("lllyasviel/Annotators").to("cuda")
# Preprocess controlnet images
conditioning_frames = []
for frame in tqdm(video):
conditioning_frames.append(open_pose(frame))
strength = 0.8
with torch.inference_mode():
video = pipe(
video=video,
prompt=prompt,
negative_prompt=negative_prompt,
num_inference_steps=10,
guidance_scale=2.0,
controlnet_conditioning_scale=0.75,
conditioning_frames=conditioning_frames,
strength=strength,
generator=torch.Generator().manual_seed(42),
).frames[0]
video = [frame.resize(conditioning_frames[0].size) for frame in video]
export_to_gif(video, f"animatediff_vid2vid_controlnet.gif", fps=8)
```
Here are some sample outputs:
<table align="center">
<tr>
<th align="center">Source Video</th>
<th align="center">Output Video</th>
</tr>
<tr>
<td align="center">
anime girl, dancing
<br />
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/dance.gif" alt="anime girl, dancing" />
</td>
<td align="center">
astronaut in space, dancing
<br/>
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/animatediff_vid2vid_controlnet.gif" alt="astronaut in space, dancing" />
</td>
</tr>
</table>
**The lights and composition were transferred from the Source Video.**
## Using Motion LoRAs
Motion LoRAs are a collection of LoRAs that work with the `guoyww/animatediff-motion-adapter-v1-5-2` checkpoint. These LoRAs are responsible for adding specific types of motion to the animations.
@@ -822,6 +914,89 @@ export_to_gif(frames, "animatelcm-motion-lora.gif")
</tr>
</table>
## Using FreeNoise
[FreeNoise: Tuning-Free Longer Video Diffusion via Noise Rescheduling](https://arxiv.org/abs/2310.15169) by Haonan Qiu, Menghan Xia, Yong Zhang, Yingqing He, Xintao Wang, Ying Shan, Ziwei Liu.
FreeNoise is a sampling mechanism that can generate longer videos with short-video generation models by employing noise-rescheduling, temporal attention over sliding windows, and weighted averaging of latent frames. It also can be used with multiple prompts to allow for interpolated video generations. More details are available in the paper.
The currently supported AnimateDiff pipelines that can be used with FreeNoise are:
- [`AnimateDiffPipeline`]
- [`AnimateDiffControlNetPipeline`]
- [`AnimateDiffVideoToVideoPipeline`]
- [`AnimateDiffVideoToVideoControlNetPipeline`]
In order to use FreeNoise, a single line needs to be added to the inference code after loading your pipelines.
```diff
+ pipe.enable_free_noise()
```
After this, either a single prompt could be used, or multiple prompts can be passed as a dictionary of integer-string pairs. The integer keys of the dictionary correspond to the frame index at which the influence of that prompt would be maximum. Each frame index should map to a single string prompt. The prompts for intermediate frame indices, that are not passed in the dictionary, are created by interpolating between the frame prompts that are passed. By default, simple linear interpolation is used. However, you can customize this behaviour with a callback to the `prompt_interpolation_callback` parameter when enabling FreeNoise.
Full example:
```python
import torch
from diffusers import AutoencoderKL, AnimateDiffPipeline, LCMScheduler, MotionAdapter
from diffusers.utils import export_to_video, load_image
# Load pipeline
dtype = torch.float16
motion_adapter = MotionAdapter.from_pretrained("wangfuyun/AnimateLCM", torch_dtype=dtype)
vae = AutoencoderKL.from_pretrained("stabilityai/sd-vae-ft-mse", torch_dtype=dtype)
pipe = AnimateDiffPipeline.from_pretrained("emilianJR/epiCRealism", motion_adapter=motion_adapter, vae=vae, torch_dtype=dtype)
pipe.scheduler = LCMScheduler.from_config(pipe.scheduler.config, beta_schedule="linear")
pipe.load_lora_weights(
"wangfuyun/AnimateLCM", weight_name="AnimateLCM_sd15_t2v_lora.safetensors", adapter_name="lcm_lora"
)
pipe.set_adapters(["lcm_lora"], [0.8])
# Enable FreeNoise for long prompt generation
pipe.enable_free_noise(context_length=16, context_stride=4)
pipe.to("cuda")
# Can be a single prompt, or a dictionary with frame timesteps
prompt = {
0: "A caterpillar on a leaf, high quality, photorealistic",
40: "A caterpillar transforming into a cocoon, on a leaf, near flowers, photorealistic",
80: "A cocoon on a leaf, flowers in the backgrond, photorealistic",
120: "A cocoon maturing and a butterfly being born, flowers and leaves visible in the background, photorealistic",
160: "A beautiful butterfly, vibrant colors, sitting on a leaf, flowers in the background, photorealistic",
200: "A beautiful butterfly, flying away in a forest, photorealistic",
240: "A cyberpunk butterfly, neon lights, glowing",
}
negative_prompt = "bad quality, worst quality, jpeg artifacts"
# Run inference
output = pipe(
prompt=prompt,
negative_prompt=negative_prompt,
num_frames=256,
guidance_scale=2.5,
num_inference_steps=10,
generator=torch.Generator("cpu").manual_seed(0),
)
# Save video
frames = output.frames[0]
export_to_video(frames, "output.mp4", fps=16)
```
### FreeNoise memory savings
Since FreeNoise processes multiple frames together, there are parts in the modeling where the memory required exceeds that available on normal consumer GPUs. The main memory bottlenecks that we identified are spatial and temporal attention blocks, upsampling and downsampling blocks, resnet blocks and feed-forward layers. Since most of these blocks operate effectively only on the channel/embedding dimension, one can perform chunked inference across the batch dimensions. The batch dimension in AnimateDiff are either spatial (`[B x F, H x W, C]`) or temporal (`B x H x W, F, C`) in nature (note that it may seem counter-intuitive, but the batch dimension here are correct, because spatial blocks process across the `B x F` dimension while the temporal blocks process across the `B x H x W` dimension). We introduce a `SplitInferenceModule` that makes it easier to chunk across any dimension and perform inference. This saves a lot of memory but comes at the cost of requiring more time for inference.
```diff
# Load pipeline and adapters
# ...
+ pipe.enable_free_noise_split_inference()
+ pipe.unet.enable_forward_chunking(16)
```
The call to `pipe.enable_free_noise_split_inference` method accepts two parameters: `spatial_split_size` (defaults to `256`) and `temporal_split_size` (defaults to `16`). These can be configured based on how much VRAM you have available. A lower split size results in lower memory usage but slower inference, whereas a larger split size results in faster inference at the cost of more memory.
## Using `from_single_file` with the MotionAdapter
@@ -866,6 +1041,12 @@ pipe = AnimateDiffPipeline.from_pretrained("emilianJR/epiCRealism", motion_adapt
- all
- __call__
## AnimateDiffVideoToVideoControlNetPipeline
[[autodoc]] AnimateDiffVideoToVideoControlNetPipeline
- all
- __call__
## AnimateDiffPipelineOutput
[[autodoc]] pipelines.animatediff.AnimateDiffPipelineOutput

View File

@@ -29,9 +29,12 @@ Make sure to check out the Schedulers [guide](../../using-diffusers/schedulers.m
This pipeline was contributed by [zRzRzRzRzRzRzR](https://github.com/zRzRzRzRzRzRzR). The original codebase can be found [here](https://huggingface.co/THUDM). The original weights can be found under [hf.co/THUDM](https://huggingface.co/THUDM).
There are two models available that can be used with the CogVideoX pipeline:
- [`THUDM/CogVideoX-2b`](https://huggingface.co/THUDM/CogVideoX-2b)
- [`THUDM/CogVideoX-5b`](https://huggingface.co/THUDM/CogVideoX-5b)
There are two models available that can be used with the text-to-video and video-to-video CogVideoX pipelines:
- [`THUDM/CogVideoX-2b`](https://huggingface.co/THUDM/CogVideoX-2b): The recommended dtype for running this model is `fp16`.
- [`THUDM/CogVideoX-5b`](https://huggingface.co/THUDM/CogVideoX-5b): The recommended dtype for running this model is `bf16`.
There is one model available that can be used with the image-to-video CogVideoX pipeline:
- [`THUDM/CogVideoX-5b-I2V`](https://huggingface.co/THUDM/CogVideoX-5b-I2V): The recommended dtype for running this model is `bf16`.
## Inference
@@ -41,10 +44,15 @@ First, load the pipeline:
```python
import torch
from diffusers import CogVideoXPipeline
from diffusers.utils import export_to_video
from diffusers import CogVideoXPipeline, CogVideoXImageToVideoPipeline
from diffusers.utils import export_to_video,load_image
pipe = CogVideoXPipeline.from_pretrained("THUDM/CogVideoX-5b").to("cuda") # or "THUDM/CogVideoX-2b"
```
pipe = CogVideoXPipeline.from_pretrained("THUDM/CogVideoX-2b").to("cuda")
If you are using the image-to-video pipeline, load it as follows:
```python
pipe = CogVideoXImageToVideoPipeline.from_pretrained("THUDM/CogVideoX-5b-I2V").to("cuda")
```
Then change the memory layout of the pipelines `transformer` component to `torch.channels_last`:
@@ -53,7 +61,7 @@ Then change the memory layout of the pipelines `transformer` component to `torch
pipe.transformer.to(memory_format=torch.channels_last)
```
Finally, compile the components and run inference:
Compile the components and run inference:
```python
pipe.transformer = torch.compile(pipeline.transformer, mode="max-autotune", fullgraph=True)
@@ -63,7 +71,7 @@ prompt = "A panda, dressed in a small, red jacket and a tiny hat, sits on a wood
video = pipe(prompt=prompt, guidance_scale=6, num_inference_steps=50).frames[0]
```
The [benchmark](https://gist.github.com/a-r-r-o-w/5183d75e452a368fd17448fcc810bd3f) results on an 80GB A100 machine are:
The [T2V benchmark](https://gist.github.com/a-r-r-o-w/5183d75e452a368fd17448fcc810bd3f) results on an 80GB A100 machine are:
```
Without torch.compile(): Average inference time: 96.89 seconds.
@@ -77,16 +85,39 @@ CogVideoX-2b requires about 19 GB of GPU memory to decode 49 frames (6 seconds o
- `pipe.enable_model_cpu_offload()`:
- Without enabling cpu offloading, memory usage is `33 GB`
- With enabling cpu offloading, memory usage is `19 GB`
- `pipe.enable_sequential_cpu_offload()`:
- Similar to `enable_model_cpu_offload` but can significantly reduce memory usage at the cost of slow inference
- When enabled, memory usage is under `4 GB`
- `pipe.vae.enable_tiling()`:
- With enabling cpu offloading and tiling, memory usage is `11 GB`
- `pipe.vae.enable_slicing()`
### Quantized inference
[torchao](https://github.com/pytorch/ao) and [optimum-quanto](https://github.com/huggingface/optimum-quanto/) can be used to quantize the text encoder, transformer and VAE modules to lower the memory requirements. This makes it possible to run the model on a free-tier T4 Colab or lower VRAM GPUs!
It is also worth noting that torchao quantization is fully compatible with [torch.compile](/optimization/torch2.0#torchcompile), which allows for much faster inference speed. Additionally, models can be serialized and stored in a quantized datatype to save disk space with torchao. Find examples and benchmarks in the gists below.
- [torchao](https://gist.github.com/a-r-r-o-w/4d9732d17412888c885480c6521a9897)
- [quanto](https://gist.github.com/a-r-r-o-w/31be62828b00a9292821b85c1017effa)
## CogVideoXPipeline
[[autodoc]] CogVideoXPipeline
- all
- __call__
## CogVideoXImageToVideoPipeline
[[autodoc]] CogVideoXImageToVideoPipeline
- all
- __call__
## CogVideoXVideoToVideoPipeline
[[autodoc]] CogVideoXVideoToVideoPipeline
- all
- __call__
## CogVideoXPipelineOutput
[[autodoc]] pipelines.cogvideo.pipeline_cogvideox.CogVideoXPipelineOutput
[[autodoc]] pipelines.cogvideo.pipeline_output.CogVideoXPipelineOutput

View File

@@ -163,3 +163,28 @@ image.save("flux-fp8-dev.png")
[[autodoc]] FluxPipeline
- all
- __call__
## FluxImg2ImgPipeline
[[autodoc]] FluxImg2ImgPipeline
- all
- __call__
## FluxInpaintPipeline
[[autodoc]] FluxInpaintPipeline
- all
- __call__
## FluxControlNetInpaintPipeline
[[autodoc]] FluxControlNetInpaintPipeline
- all
- __call__
## FluxControlNetImg2ImgPipeline
[[autodoc]] FluxControlNetImg2ImgPipeline
- all
- __call__

View File

@@ -105,3 +105,11 @@ image.save("kolors_ipa_sample.png")
- all
- __call__
## KolorsImg2ImgPipeline
[[autodoc]] KolorsImg2ImgPipeline
- all
- __call__

View File

@@ -19,7 +19,7 @@ The Stable Diffusion model can also be applied to inpainting which lets you edit
It is recommended to use this pipeline with checkpoints that have been specifically fine-tuned for inpainting, such
as [runwayml/stable-diffusion-inpainting](https://huggingface.co/runwayml/stable-diffusion-inpainting). Default
text-to-image Stable Diffusion checkpoints, such as
[runwayml/stable-diffusion-v1-5](https://huggingface.co/runwayml/stable-diffusion-v1-5) are also compatible but they might be less performant.
[stable-diffusion-v1-5/stable-diffusion-v1-5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) are also compatible but they might be less performant.
<Tip>

View File

@@ -203,7 +203,7 @@ from diffusers import StableDiffusionImg2ImgPipeline
import gradio as gr
pipe = StableDiffusionImg2ImgPipeline.from_pretrained("runwayml/stable-diffusion-v1-5")
pipe = StableDiffusionImg2ImgPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5")
gr.Interface.from_pipeline(pipe).launch()
```

View File

@@ -41,7 +41,7 @@ To generate a video from prompt, run the following Python code:
import torch
from diffusers import TextToVideoZeroPipeline
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
pipe = TextToVideoZeroPipeline.from_pretrained(model_id, torch_dtype=torch.float16).to("cuda")
prompt = "A panda is playing guitar on times square"
@@ -63,7 +63,7 @@ import torch
from diffusers import TextToVideoZeroPipeline
import numpy as np
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
pipe = TextToVideoZeroPipeline.from_pretrained(model_id, torch_dtype=torch.float16).to("cuda")
seed = 0
video_length = 24 #24 ÷ 4fps = 6 seconds
@@ -137,7 +137,7 @@ To generate a video from prompt with additional pose control
from diffusers import StableDiffusionControlNetPipeline, ControlNetModel
from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_zero import CrossFrameAttnProcessor
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
controlnet = ControlNetModel.from_pretrained("lllyasviel/sd-controlnet-openpose", torch_dtype=torch.float16)
pipe = StableDiffusionControlNetPipeline.from_pretrained(
model_id, controlnet=controlnet, torch_dtype=torch.float16

View File

@@ -92,7 +92,7 @@ images = sd_pipeline(sample_prompts, num_images_per_prompt=1, generator=generato
![parti-prompts-14](https://huggingface.co/datasets/diffusers/docs-images/resolve/main/evaluation_diffusion_models/parti-prompts-14.png)
We can also set `num_images_per_prompt` accordingly to compare different images for the same prompt. Running the same pipeline but with a different checkpoint ([v1-5](https://huggingface.co/runwayml/stable-diffusion-v1-5)), yields:
We can also set `num_images_per_prompt` accordingly to compare different images for the same prompt. Running the same pipeline but with a different checkpoint ([v1-5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5)), yields:
![parti-prompts-15](https://huggingface.co/datasets/diffusers/docs-images/resolve/main/evaluation_diffusion_models/parti-prompts-15.png)
@@ -177,10 +177,10 @@ generator = torch.manual_seed(seed)
images = sd_pipeline(prompts, num_images_per_prompt=1, generator=generator, output_type="np").images
```
Then we load the [v1-5 checkpoint](https://huggingface.co/runwayml/stable-diffusion-v1-5) to generate images:
Then we load the [v1-5 checkpoint](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) to generate images:
```python
model_ckpt_1_5 = "runwayml/stable-diffusion-v1-5"
model_ckpt_1_5 = "stable-diffusion-v1-5/stable-diffusion-v1-5"
sd_pipeline_1_5 = StableDiffusionPipeline.from_pretrained(model_ckpt_1_5, torch_dtype=weight_dtype).to(device)
images_1_5 = sd_pipeline_1_5(prompts, num_images_per_prompt=1, generator=generator, output_type="np").images
@@ -198,7 +198,7 @@ print(f"CLIP Score with v-1-5: {sd_clip_score_1_5}")
# CLIP Score with v-1-5: 36.2137
```
It seems like the [v1-5](https://huggingface.co/runwayml/stable-diffusion-v1-5) checkpoint performs better than its predecessor. Note, however, that the number of prompts we used to compute the CLIP scores is quite low. For a more practical evaluation, this number should be way higher, and the prompts should be diverse.
It seems like the [v1-5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) checkpoint performs better than its predecessor. Note, however, that the number of prompts we used to compute the CLIP scores is quite low. For a more practical evaluation, this number should be way higher, and the prompts should be diverse.
<Tip warning={true}>

View File

@@ -65,7 +65,7 @@ Pipelines are designed to be easy to use (therefore do not follow [*Simple over
The following design principles are followed:
- Pipelines follow the single-file policy. All pipelines can be found in individual directories under src/diffusers/pipelines. One pipeline folder corresponds to one diffusion paper/project/release. Multiple pipeline files can be gathered in one pipeline folder, as its done for [`src/diffusers/pipelines/stable-diffusion`](https://github.com/huggingface/diffusers/tree/main/src/diffusers/pipelines/stable_diffusion). If pipelines share similar functionality, one can make use of the [# Copied from mechanism](https://github.com/huggingface/diffusers/blob/125d783076e5bd9785beb05367a2d2566843a271/src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_img2img.py#L251).
- Pipelines all inherit from [`DiffusionPipeline`].
- Every pipeline consists of different model and scheduler components, that are documented in the [`model_index.json` file](https://huggingface.co/runwayml/stable-diffusion-v1-5/blob/main/model_index.json), are accessible under the same name as attributes of the pipeline and can be shared between pipelines with [`DiffusionPipeline.components`](https://huggingface.co/docs/diffusers/main/en/api/diffusion_pipeline#diffusers.DiffusionPipeline.components) function.
- Every pipeline consists of different model and scheduler components, that are documented in the [`model_index.json` file](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5/blob/main/model_index.json), are accessible under the same name as attributes of the pipeline and can be shared between pipelines with [`DiffusionPipeline.components`](https://huggingface.co/docs/diffusers/main/en/api/diffusion_pipeline#diffusers.DiffusionPipeline.components) function.
- Every pipeline should be loadable via the [`DiffusionPipeline.from_pretrained`](https://huggingface.co/docs/diffusers/main/en/api/diffusion_pipeline#diffusers.DiffusionPipeline.from_pretrained) function.
- Pipelines should be used **only** for inference.
- Pipelines should be very readable, self-explanatory, and easy to tweak.

View File

@@ -102,10 +102,10 @@ Pass the path of the downloaded checkpoint with `-i` flag to the script. `--comp
The inference script assumes you're using the original version of the Stable Diffusion model, `CompVis/stable-diffusion-v1-4`. If you use another model, you *have* to specify its Hub id in the inference command line, using the `--model-version` option. This works for models already supported and custom models you trained or fine-tuned yourself.
For example, if you want to use [`runwayml/stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5):
For example, if you want to use [`stable-diffusion-v1-5/stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5):
```shell
python -m python_coreml_stable_diffusion.pipeline --prompt "a photo of an astronaut riding a horse on mars" --compute-unit ALL -o output --seed 93 -i models/coreml-stable-diffusion-v1-5_original_packages --model-version runwayml/stable-diffusion-v1-5
python -m python_coreml_stable_diffusion.pipeline --prompt "a photo of an astronaut riding a horse on mars" --compute-unit ALL -o output --seed 93 -i models/coreml-stable-diffusion-v1-5_original_packages --model-version stable-diffusion-v1-5/stable-diffusion-v1-5
```
## Core ML inference in Swift

View File

@@ -23,7 +23,7 @@ Then load and enable the [`DeepCacheSDHelper`](https://github.com/horseee/DeepCa
```diff
import torch
from diffusers import StableDiffusionPipeline
pipe = StableDiffusionPipeline.from_pretrained('runwayml/stable-diffusion-v1-5', torch_dtype=torch.float16).to("cuda")
pipe = StableDiffusionPipeline.from_pretrained('stable-diffusion-v1-5/stable-diffusion-v1-5', torch_dtype=torch.float16).to("cuda")
+ from DeepCache import DeepCacheSDHelper
+ helper = DeepCacheSDHelper(pipe=pipe)

View File

@@ -47,7 +47,7 @@ import torch
from diffusers import DiffusionPipeline
pipe = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
use_safetensors=True,
)

View File

@@ -61,7 +61,7 @@ For more information, check out 🤗 Optimum Habana's [documentation](https://hu
We benchmarked Habana's first-generation Gaudi and Gaudi2 with the [Habana/stable-diffusion](https://huggingface.co/Habana/stable-diffusion) and [Habana/stable-diffusion-2](https://huggingface.co/Habana/stable-diffusion-2) Gaudi configurations (mixed precision bf16/fp32) to demonstrate their performance.
For [Stable Diffusion v1.5](https://huggingface.co/runwayml/stable-diffusion-v1-5) on 512x512 images:
For [Stable Diffusion v1.5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) on 512x512 images:
| | Latency (batch size = 1) | Throughput |
| ---------------------- |:------------------------:|:---------------------------:|

View File

@@ -41,7 +41,7 @@ import torch
from diffusers import StableDiffusionPipeline
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
use_safetensors=True,
)
@@ -66,7 +66,7 @@ import torch
from diffusers import StableDiffusionPipeline, UniPCMultistepScheduler
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
use_safetensors=True,
)
@@ -92,7 +92,7 @@ import torch
from diffusers import StableDiffusionPipeline
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
use_safetensors=True,
)
@@ -140,7 +140,7 @@ import torch
from diffusers import StableDiffusionPipeline
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
use_safetensors=True,
)
@@ -201,7 +201,7 @@ def generate_inputs():
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
use_safetensors=True,
).to("cuda")
@@ -265,7 +265,7 @@ class UNet2DConditionOutput:
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
use_safetensors=True,
).to("cuda")
@@ -315,7 +315,7 @@ from diffusers import DiffusionPipeline
import torch
pipe = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
use_safetensors=True,
).to("cuda")

View File

@@ -24,7 +24,7 @@ The `mps` backend uses PyTorch's `.to()` interface to move the Stable Diffusion
```python
from diffusers import DiffusionPipeline
pipe = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5")
pipe = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5")
pipe = pipe.to("mps")
# Recommended if your computer has < 64 GB of RAM
@@ -46,7 +46,7 @@ If you're using **PyTorch 1.13**, you need to "prime" the pipeline with an addit
```diff
from diffusers import DiffusionPipeline
pipe = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5").to("mps")
pipe = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5").to("mps")
pipe.enable_attention_slicing()
prompt = "a photo of an astronaut riding a horse on mars"
@@ -67,7 +67,7 @@ To prevent this from happening, we recommend *attention slicing* to reduce memor
from diffusers import DiffusionPipeline
import torch
pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True).to("mps")
pipeline = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True).to("mps")
pipeline.enable_attention_slicing()
```

View File

@@ -27,7 +27,7 @@ To load and run inference, use the [`~optimum.onnxruntime.ORTStableDiffusionPipe
```python
from optimum.onnxruntime import ORTStableDiffusionPipeline
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
pipeline = ORTStableDiffusionPipeline.from_pretrained(model_id, export=True)
prompt = "sailing ship in storm by Leonardo da Vinci"
image = pipeline(prompt).images[0]
@@ -44,7 +44,7 @@ To export the pipeline in the ONNX format offline and use it later for inference
use the [`optimum-cli export`](https://huggingface.co/docs/optimum/main/en/exporters/onnx/usage_guides/export_a_model#exporting-a-model-to-onnx-using-the-cli) command:
```bash
optimum-cli export onnx --model runwayml/stable-diffusion-v1-5 sd_v15_onnx/
optimum-cli export onnx --model stable-diffusion-v1-5/stable-diffusion-v1-5 sd_v15_onnx/
```
Then to perform inference (you don't have to specify `export=True` again):

View File

@@ -29,7 +29,7 @@ To load and run inference, use the [`~optimum.intel.OVStableDiffusionPipeline`].
```python
from optimum.intel import OVStableDiffusionPipeline
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
pipeline = OVStableDiffusionPipeline.from_pretrained(model_id, export=True)
prompt = "sailing ship in storm by Rembrandt"
image = pipeline(prompt).images[0]

View File

@@ -28,7 +28,7 @@ You can use ToMe from the [`tomesd`](https://github.com/dbolya/tomesd) library w
import tomesd
pipeline = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True,
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True,
).to("cuda")
+ tomesd.apply_patch(pipeline, ratio=0.5)

View File

@@ -34,7 +34,7 @@ However, if you want to explicitly enable it, you can set a [`DiffusionPipeline`
from diffusers import DiffusionPipeline
+ from diffusers.models.attention_processor import AttnProcessor2_0
pipe = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
pipe = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
+ pipe.unet.set_attn_processor(AttnProcessor2_0())
prompt = "a photo of an astronaut riding a horse on mars"
@@ -49,7 +49,7 @@ In some cases - such as making the pipeline more deterministic or converting it
import torch
from diffusers import DiffusionPipeline
pipe = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
pipe = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
+ pipe.unet.set_default_attn_processor()
prompt = "a photo of an astronaut riding a horse on mars"
@@ -64,7 +64,7 @@ The `torch.compile` function can often provide an additional speed-up to your Py
from diffusers import DiffusionPipeline
import torch
pipe = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
pipe = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
pipe.unet = torch.compile(pipe.unet, mode="reduce-overhead", fullgraph=True)
images = pipe(prompt, num_inference_steps=steps, num_images_per_prompt=batch_size).images[0]
```
@@ -92,7 +92,7 @@ Expand the dropdown below to find the code used to benchmark each pipeline:
from diffusers import DiffusionPipeline
import torch
path = "runwayml/stable-diffusion-v1-5"
path = "stable-diffusion-v1-5/stable-diffusion-v1-5"
run_compile = True # Set True / False
@@ -122,7 +122,7 @@ url = "https://raw.githubusercontent.com/CompVis/stable-diffusion/main/assets/st
init_image = load_image(url)
init_image = init_image.resize((512, 512))
path = "runwayml/stable-diffusion-v1-5"
path = "stable-diffusion-v1-5/stable-diffusion-v1-5"
run_compile = True # Set True / False
@@ -183,7 +183,7 @@ url = "https://raw.githubusercontent.com/CompVis/stable-diffusion/main/assets/st
init_image = load_image(url)
init_image = init_image.resize((512, 512))
path = "runwayml/stable-diffusion-v1-5"
path = "stable-diffusion-v1-5/stable-diffusion-v1-5"
run_compile = True # Set True / False
controlnet = ControlNetModel.from_pretrained("lllyasviel/sd-controlnet-canny", torch_dtype=torch.float16, use_safetensors=True)

View File

@@ -0,0 +1,122 @@
# xDiT
[xDiT](https://github.com/xdit-project/xDiT) is an inference engine designed for the large scale parallel deployment of Diffusion Transformers (DiTs). xDiT provides a suite of efficient parallel approaches for Diffusion Models, as well as GPU kernel accelerations.
There are four parallel methods supported in xDiT, including [Unified Sequence Parallelism](https://arxiv.org/abs/2405.07719), [PipeFusion](https://arxiv.org/abs/2405.14430), CFG parallelism and data parallelism. The four parallel methods in xDiT can be configured in a hybrid manner, optimizing communication patterns to best suit the underlying network hardware.
Optimization orthogonal to parallelization focuses on accelerating single GPU performance. In addition to utilizing well-known Attention optimization libraries, we leverage compilation acceleration technologies such as torch.compile and onediff.
The overview of xDiT is shown as follows.
<div class="flex justify-center">
<img src="https://github.com/xdit-project/xDiT/raw/main/assets/methods/xdit_overview.png">
</div>
You can install xDiT using the following command:
```bash
pip install xfuser
```
Here's an example of using xDiT to accelerate inference of a Diffusers model.
```diff
import torch
from diffusers import StableDiffusion3Pipeline
from xfuser import xFuserArgs, xDiTParallel
from xfuser.config import FlexibleArgumentParser
from xfuser.core.distributed import get_world_group
def main():
+ parser = FlexibleArgumentParser(description="xFuser Arguments")
+ args = xFuserArgs.add_cli_args(parser).parse_args()
+ engine_args = xFuserArgs.from_cli_args(args)
+ engine_config, input_config = engine_args.create_config()
local_rank = get_world_group().local_rank
pipe = StableDiffusion3Pipeline.from_pretrained(
pretrained_model_name_or_path=engine_config.model_config.model,
torch_dtype=torch.float16,
).to(f"cuda:{local_rank}")
# do anything you want with pipeline here
+ pipe = xDiTParallel(pipe, engine_config, input_config)
pipe(
height=input_config.height,
width=input_config.height,
prompt=input_config.prompt,
num_inference_steps=input_config.num_inference_steps,
output_type=input_config.output_type,
generator=torch.Generator(device="cuda").manual_seed(input_config.seed),
)
+ if input_config.output_type == "pil":
+ pipe.save("results", "stable_diffusion_3")
if __name__ == "__main__":
main()
```
As you can see, we only need to use xFuserArgs from xDiT to get configuration parameters, and pass these parameters along with the pipeline object from the Diffusers library into xDiTParallel to complete the parallelization of a specific pipeline in Diffusers.
xDiT runtime parameters can be viewed in the command line using `-h`, and you can refer to this [usage](https://github.com/xdit-project/xDiT?tab=readme-ov-file#2-usage) example for more details.
xDiT needs to be launched using torchrun to support its multi-node, multi-GPU parallel capabilities. For example, the following command can be used for 8-GPU parallel inference:
```bash
torchrun --nproc_per_node=8 ./inference.py --model models/FLUX.1-dev --data_parallel_degree 2 --ulysses_degree 2 --ring_degree 2 --prompt "A snowy mountain" "A small dog" --num_inference_steps 50
```
## Supported models
A subset of Diffusers models are supported in xDiT, such as Flux.1, Stable Diffusion 3, etc. The latest supported models can be found [here](https://github.com/xdit-project/xDiT?tab=readme-ov-file#-supported-dits).
## Benchmark
We tested different models on various machines, and here is some of the benchmark data.
### Flux.1-schnell
<div class="flex justify-center">
<img src="https://github.com/xdit-project/xDiT/raw/main/assets/performance/flux/Flux-2k-L40.png">
</div>
<div class="flex justify-center">
<img src="https://github.com/xdit-project/xDiT/raw/main/assets/performance/flux/Flux-2K-A100.png">
</div>
### Stable Diffusion 3
<div class="flex justify-center">
<img src="https://github.com/xdit-project/xDiT/raw/main/assets/performance/sd3/L40-SD3.png">
</div>
<div class="flex justify-center">
<img src="https://github.com/xdit-project/xDiT/raw/main/assets/performance/sd3/A100-SD3.png">
</div>
### HunyuanDiT
<div class="flex justify-center">
<img src="https://github.com/xdit-project/xDiT/raw/main/assets/performance/hunuyuandit/L40-HunyuanDiT.png">
</div>
<div class="flex justify-center">
<img src="https://github.com/xdit-project/xDiT/raw/main/assets/performance/hunuyuandit/A100-HunyuanDiT.png">
</div>
<div class="flex justify-center">
<img src="https://github.com/xdit-project/xDiT/raw/main/assets/performance/hunuyuandit/T4-HunyuanDiT.png">
</div>
More detailed performance metric can be found on our [github page](https://github.com/xdit-project/xDiT?tab=readme-ov-file#perf).
## Reference
[xDiT-project](https://github.com/xdit-project/xDiT)
[USP: A Unified Sequence Parallelism Approach for Long Context Generative AI](https://arxiv.org/abs/2405.07719)
[PipeFusion: Displaced Patch Pipeline Parallelism for Inference of Diffusion Transformer Models](https://arxiv.org/abs/2405.14430)

View File

@@ -54,7 +54,7 @@ The [`DiffusionPipeline`] is the easiest way to use a pretrained diffusion syste
Start by creating an instance of a [`DiffusionPipeline`] and specify which pipeline checkpoint you would like to download.
You can use the [`DiffusionPipeline`] for any [checkpoint](https://huggingface.co/models?library=diffusers&sort=downloads) stored on the Hugging Face Hub.
In this quicktour, you'll load the [`stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5) checkpoint for text-to-image generation.
In this quicktour, you'll load the [`stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) checkpoint for text-to-image generation.
<Tip warning={true}>
@@ -67,7 +67,7 @@ Load the model with the [`~DiffusionPipeline.from_pretrained`] method:
```python
>>> from diffusers import DiffusionPipeline
>>> pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", use_safetensors=True)
>>> pipeline = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", use_safetensors=True)
```
The [`DiffusionPipeline`] downloads and caches all modeling, tokenization, and scheduling components. You'll see that the Stable Diffusion pipeline is composed of the [`UNet2DConditionModel`] and [`PNDMScheduler`] among other things:
@@ -124,7 +124,7 @@ You can also use the pipeline locally. The only difference is you need to downlo
```bash
!git lfs install
!git clone https://huggingface.co/runwayml/stable-diffusion-v1-5
!git clone https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5
```
Then load the saved weights into the pipeline:
@@ -142,7 +142,7 @@ Different schedulers come with different denoising speeds and quality trade-offs
```py
>>> from diffusers import EulerDiscreteScheduler
>>> pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", use_safetensors=True)
>>> pipeline = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", use_safetensors=True)
>>> pipeline.scheduler = EulerDiscreteScheduler.from_config(pipeline.scheduler.config)
```

View File

@@ -20,12 +20,12 @@ This is why it's important to get the most *computational* (speed) and *memory*
This tutorial walks you through how to generate faster and better with the [`DiffusionPipeline`].
Begin by loading the [`runwayml/stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5) model:
Begin by loading the [`stable-diffusion-v1-5/stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) model:
```python
from diffusers import DiffusionPipeline
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
pipeline = DiffusionPipeline.from_pretrained(model_id, use_safetensors=True)
```

View File

@@ -6,12 +6,12 @@ This guide will show you how to adapt a pretrained text-to-image model for inpai
## Configure UNet2DConditionModel parameters
A [`UNet2DConditionModel`] by default accepts 4 channels in the [input sample](https://huggingface.co/docs/diffusers/v0.16.0/en/api/models#diffusers.UNet2DConditionModel.in_channels). For example, load a pretrained text-to-image model like [`runwayml/stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5) and take a look at the number of `in_channels`:
A [`UNet2DConditionModel`] by default accepts 4 channels in the [input sample](https://huggingface.co/docs/diffusers/v0.16.0/en/api/models#diffusers.UNet2DConditionModel.in_channels). For example, load a pretrained text-to-image model like [`stable-diffusion-v1-5/stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) and take a look at the number of `in_channels`:
```py
from diffusers import StableDiffusionPipeline
pipeline = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", use_safetensors=True)
pipeline = StableDiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", use_safetensors=True)
pipeline.unet.config["in_channels"]
4
```
@@ -33,7 +33,7 @@ Initialize a [`UNet2DConditionModel`] with the pretrained text-to-image model we
```py
from diffusers import UNet2DConditionModel
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
unet = UNet2DConditionModel.from_pretrained(
model_id,
subfolder="unet",

View File

@@ -276,7 +276,7 @@ That's it! You don't need to add any additional parameters to your training comm
<hfoption id="PyTorch">
```bash
export MODEL_DIR="runwayml/stable-diffusion-v1-5"
export MODEL_DIR="stable-diffusion-v1-5/stable-diffusion-v1-5"
export OUTPUT_DIR="path/to/save/model"
accelerate launch train_controlnet.py \

View File

@@ -78,7 +78,7 @@ Now the dataset is available for training by passing the dataset name to the `--
```bash
accelerate launch --mixed_precision="fp16" train_text_to_image.py \
--pretrained_model_name_or_path="runwayml/stable-diffusion-v1-5" \
--pretrained_model_name_or_path="stable-diffusion-v1-5/stable-diffusion-v1-5" \
--dataset_name="name_of_your_dataset" \
<other-arguments>
```

View File

@@ -30,7 +30,7 @@ from accelerate import PartialState
from diffusers import DiffusionPipeline
pipeline = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True
)
distributed_state = PartialState()
pipeline.to(distributed_state.device)
@@ -66,7 +66,7 @@ import torch.multiprocessing as mp
from diffusers import DiffusionPipeline
sd = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True
)
```

View File

@@ -315,7 +315,7 @@ That's it! You don't need to add any additional parameters to your training comm
<hfoption id="PyTorch">
```bash
export MODEL_NAME="runwayml/stable-diffusion-v1-5"
export MODEL_NAME="stable-diffusion-v1-5/stable-diffusion-v1-5"
export INSTANCE_DIR="./dog"
export OUTPUT_DIR="path_to_saved_model"
@@ -374,7 +374,7 @@ unet = UNet2DConditionModel.from_pretrained("path/to/model/checkpoint-100/unet")
text_encoder = CLIPTextModel.from_pretrained("path/to/model/checkpoint-100/checkpoint-100/text_encoder")
pipeline = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", unet=unet, text_encoder=text_encoder, dtype=torch.float16,
"stable-diffusion-v1-5/stable-diffusion-v1-5", unet=unet, text_encoder=text_encoder, dtype=torch.float16,
).to("cuda")
image = pipeline("A photo of sks dog in a bucket", num_inference_steps=50, guidance_scale=7.5).images[0]

View File

@@ -193,7 +193,7 @@ Now you're ready to launch the training script and start distilling!
For this guide, you'll use the `--train_shards_path_or_url` to specify the path to the [Conceptual Captions 12M](https://github.com/google-research-datasets/conceptual-12m) dataset stored on the Hub [here](https://huggingface.co/datasets/laion/conceptual-captions-12m-webdataset). Set the `MODEL_DIR` environment variable to the name of the teacher model and `OUTPUT_DIR` to where you want to save the model.
```bash
export MODEL_DIR="runwayml/stable-diffusion-v1-5"
export MODEL_DIR="stable-diffusion-v1-5/stable-diffusion-v1-5"
export OUTPUT_DIR="path/to/saved/model"
accelerate launch train_lcm_distill_sd_wds.py \
@@ -225,7 +225,7 @@ from diffusers import UNet2DConditionModel, DiffusionPipeline, LCMScheduler
import torch
unet = UNet2DConditionModel.from_pretrained("your-username/your-model", torch_dtype=torch.float16, variant="fp16")
pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", unet=unet, torch_dtype=torch.float16, variant="fp16")
pipeline = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", unet=unet, torch_dtype=torch.float16, variant="fp16")
pipeline.scheduler = LCMScheduler.from_config(pipe.scheduler.config)
pipeline.to("cuda")

View File

@@ -184,7 +184,7 @@ A full training run takes ~5 hours on a 2080 Ti GPU with 11GB of VRAM.
</Tip>
```bash
export MODEL_NAME="runwayml/stable-diffusion-v1-5"
export MODEL_NAME="stable-diffusion-v1-5/stable-diffusion-v1-5"
export OUTPUT_DIR="/sddata/finetune/lora/naruto"
export HUB_MODEL_ID="naruto-lora"
export DATASET_NAME="lambdalabs/naruto-blip-captions"
@@ -218,7 +218,7 @@ Once training has been completed, you can use your model for inference:
from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
pipeline = AutoPipelineForText2Image.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
pipeline.load_lora_weights("path/to/lora/model", weight_name="pytorch_lora_weights.safetensors")
image = pipeline("A naruto with blue eyes").images[0]
```

View File

@@ -167,7 +167,7 @@ To train on a local dataset, set the `TRAIN_DIR` and `OUTPUT_DIR` environment va
</Tip>
```bash
export MODEL_NAME="runwayml/stable-diffusion-v1-5"
export MODEL_NAME="stable-diffusion-v1-5/stable-diffusion-v1-5"
export dataset_name="lambdalabs/naruto-blip-captions"
accelerate launch --mixed_precision="fp16" train_text_to_image.py \
@@ -201,7 +201,7 @@ To train on a local dataset, set the `TRAIN_DIR` and `OUTPUT_DIR` environment va
</Tip>
```bash
export MODEL_NAME="runwayml/stable-diffusion-v1-5"
export MODEL_NAME="stable-diffusion-v1-5/stable-diffusion-v1-5"
export dataset_name="lambdalabs/naruto-blip-captions"
python train_text_to_image_flax.py \

View File

@@ -193,7 +193,7 @@ One more thing before you launch the script. If you're interested in following a
<hfoption id="PyTorch">
```bash
export MODEL_NAME="runwayml/stable-diffusion-v1-5"
export MODEL_NAME="stable-diffusion-v1-5/stable-diffusion-v1-5"
export DATA_DIR="./cat"
accelerate launch textual_inversion.py \
@@ -248,7 +248,7 @@ After training is complete, you can use your newly trained model for inference l
from diffusers import StableDiffusionPipeline
import torch
pipeline = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
pipeline = StableDiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
pipeline.load_textual_inversion("sd-concepts-library/cat-toy")
image = pipeline("A <cat-toy> train", num_inference_steps=50).images[0]
image.save("cat-train.png")

View File

@@ -90,8 +90,8 @@ from diffusers import DiffusionPipeline
import torch
pipeline = DiffusionPipeline.from_pretrained(
- "runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True,
+ "runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True, device_map="balanced"
- "stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True,
+ "stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True, device_map="balanced"
)
image = pipeline("a dog").images[0]
image
@@ -105,7 +105,7 @@ import torch
max_memory = {0:"1GB", 1:"1GB"}
pipeline = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
use_safetensors=True,
device_map="balanced",

View File

@@ -109,7 +109,7 @@ Now, you can pass the callback function to the `callback_on_step_end` parameter
import torch
from diffusers import StableDiffusionPipeline
pipeline = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16)
pipeline = StableDiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16)
pipeline = pipeline.to("cuda")
prompt = "a photo of an astronaut riding a horse on mars"
@@ -139,7 +139,7 @@ In this example, the diffusion process is stopped after 10 steps even though `nu
```python
from diffusers import StableDiffusionPipeline
pipeline = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5")
pipeline = StableDiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5")
pipeline.enable_model_cpu_offload()
num_inference_steps = 50

View File

@@ -33,7 +33,7 @@ from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16"
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16"
).to("cuda")
```
@@ -52,18 +52,18 @@ image
## Popular models
The most common text-to-image models are [Stable Diffusion v1.5](https://huggingface.co/runwayml/stable-diffusion-v1-5), [Stable Diffusion XL (SDXL)](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0), and [Kandinsky 2.2](https://huggingface.co/kandinsky-community/kandinsky-2-2-decoder). There are also ControlNet models or adapters that can be used with text-to-image models for more direct control in generating images. The results from each model are slightly different because of their architecture and training process, but no matter which model you choose, their usage is more or less the same. Let's use the same prompt for each model and compare their results.
The most common text-to-image models are [Stable Diffusion v1.5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5), [Stable Diffusion XL (SDXL)](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0), and [Kandinsky 2.2](https://huggingface.co/kandinsky-community/kandinsky-2-2-decoder). There are also ControlNet models or adapters that can be used with text-to-image models for more direct control in generating images. The results from each model are slightly different because of their architecture and training process, but no matter which model you choose, their usage is more or less the same. Let's use the same prompt for each model and compare their results.
### Stable Diffusion v1.5
[Stable Diffusion v1.5](https://huggingface.co/runwayml/stable-diffusion-v1-5) is a latent diffusion model initialized from [Stable Diffusion v1-4](https://huggingface.co/CompVis/stable-diffusion-v1-4), and finetuned for 595K steps on 512x512 images from the LAION-Aesthetics V2 dataset. You can use this model like:
[Stable Diffusion v1.5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) is a latent diffusion model initialized from [Stable Diffusion v1-4](https://huggingface.co/CompVis/stable-diffusion-v1-4), and finetuned for 595K steps on 512x512 images from the LAION-Aesthetics V2 dataset. You can use this model like:
```py
from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16"
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16"
).to("cuda")
generator = torch.Generator("cuda").manual_seed(31)
image = pipeline("Astronaut in a jungle, cold color palette, muted colors, detailed, 8k", generator=generator).images[0]
@@ -106,7 +106,7 @@ image
### ControlNet
ControlNet models are auxiliary models or adapters that are finetuned on top of text-to-image models, such as [Stable Diffusion v1.5](https://huggingface.co/runwayml/stable-diffusion-v1-5). Using ControlNet models in combination with text-to-image models offers diverse options for more explicit control over how to generate an image. With ControlNet, you add an additional conditioning input image to the model. For example, if you provide an image of a human pose (usually represented as multiple keypoints that are connected into a skeleton) as a conditioning input, the model generates an image that follows the pose of the image. Check out the more in-depth [ControlNet](controlnet) guide to learn more about other conditioning inputs and how to use them.
ControlNet models are auxiliary models or adapters that are finetuned on top of text-to-image models, such as [Stable Diffusion v1.5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5). Using ControlNet models in combination with text-to-image models offers diverse options for more explicit control over how to generate an image. With ControlNet, you add an additional conditioning input image to the model. For example, if you provide an image of a human pose (usually represented as multiple keypoints that are connected into a skeleton) as a conditioning input, the model generates an image that follows the pose of the image. Check out the more in-depth [ControlNet](controlnet) guide to learn more about other conditioning inputs and how to use them.
In this example, let's condition the ControlNet with a human pose estimation image. Load the ControlNet model pretrained on human pose estimations:
@@ -125,7 +125,7 @@ Pass the `controlnet` to the [`AutoPipelineForText2Image`], and provide the prom
```py
pipeline = AutoPipelineForText2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", controlnet=controlnet, torch_dtype=torch.float16, variant="fp16"
"stable-diffusion-v1-5/stable-diffusion-v1-5", controlnet=controlnet, torch_dtype=torch.float16, variant="fp16"
).to("cuda")
generator = torch.Generator("cuda").manual_seed(31)
image = pipeline("Astronaut in a jungle, cold color palette, muted colors, detailed, 8k", image=pose_image, generator=generator).images[0]
@@ -164,7 +164,7 @@ from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16"
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16"
).to("cuda")
image = pipeline(
"Astronaut in a jungle, cold color palette, muted colors, detailed, 8k", height=768, width=512
@@ -191,7 +191,7 @@ from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16
).to("cuda")
image = pipeline(
"Astronaut in a jungle, cold color palette, muted colors, detailed, 8k", guidance_scale=3.5
@@ -223,7 +223,7 @@ from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16
).to("cuda")
image = pipeline(
prompt="Astronaut in a jungle, cold color palette, muted colors, detailed, 8k",
@@ -254,7 +254,7 @@ from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16
).to("cuda")
generator = torch.Generator(device="cuda").manual_seed(30)
image = pipeline(
@@ -285,7 +285,7 @@ from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16
).to("cuda")
image = pipeline(
prompt_embeds=prompt_embeds, # generated from Compel
@@ -309,7 +309,7 @@ PyTorch 2.0 also supports a more memory-efficient attention mechanism called [*s
from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16").to("cuda")
pipeline = AutoPipelineForText2Image.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16").to("cuda")
pipeline.unet = torch.compile(pipeline.unet, mode="reduce-overhead", fullgraph=True)
```

View File

@@ -84,7 +84,7 @@ import torch
controlnet = ControlNetModel.from_pretrained("lllyasviel/sd-controlnet-canny", torch_dtype=torch.float16, use_safetensors=True)
pipe = StableDiffusionControlNetPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", controlnet=controlnet, torch_dtype=torch.float16, use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", controlnet=controlnet, torch_dtype=torch.float16, use_safetensors=True
)
pipe.scheduler = UniPCMultistepScheduler.from_config(pipe.scheduler.config)
@@ -144,7 +144,7 @@ import torch
controlnet = ControlNetModel.from_pretrained("lllyasviel/control_v11f1p_sd15_depth", torch_dtype=torch.float16, use_safetensors=True)
pipe = StableDiffusionControlNetImg2ImgPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", controlnet=controlnet, torch_dtype=torch.float16, use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", controlnet=controlnet, torch_dtype=torch.float16, use_safetensors=True
)
pipe.scheduler = UniPCMultistepScheduler.from_config(pipe.scheduler.config)
@@ -229,7 +229,7 @@ from diffusers import StableDiffusionControlNetInpaintPipeline, ControlNetModel,
controlnet = ControlNetModel.from_pretrained("lllyasviel/control_v11p_sd15_inpaint", torch_dtype=torch.float16, use_safetensors=True)
pipe = StableDiffusionControlNetInpaintPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", controlnet=controlnet, torch_dtype=torch.float16, use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", controlnet=controlnet, torch_dtype=torch.float16, use_safetensors=True
)
pipe.scheduler = UniPCMultistepScheduler.from_config(pipe.scheduler.config)
@@ -277,7 +277,7 @@ from PIL import Image
import cv2
controlnet = ControlNetModel.from_pretrained("lllyasviel/sd-controlnet-canny", use_safetensors=True)
pipe = StableDiffusionControlNetPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", controlnet=controlnet, use_safetensors=True).to("cuda")
pipe = StableDiffusionControlNetPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", controlnet=controlnet, use_safetensors=True).to("cuda")
original_image = load_image("https://huggingface.co/takuma104/controlnet_dev/resolve/main/bird_512x512.png")
@@ -454,7 +454,7 @@ image = base(
<Tip>
Replace the SDXL model with a model like [runwayml/stable-diffusion-v1-5](https://huggingface.co/runwayml/stable-diffusion-v1-5) to use multiple conditioning inputs with Stable Diffusion models.
Replace the SDXL model with a model like [stable-diffusion-v1-5/stable-diffusion-v1-5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) to use multiple conditioning inputs with Stable Diffusion models.
</Tip>

View File

@@ -61,7 +61,7 @@ feature_extractor = CLIPImageProcessor.from_pretrained(clip_model_id)
clip_model = CLIPModel.from_pretrained(clip_model_id)
pipeline = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
custom_pipeline="clip_guided_stable_diffusion",
clip_model=clip_model,
feature_extractor=feature_extractor,
@@ -78,7 +78,7 @@ Community pipelines can also be loaded from a local file if you pass a file path
```py
pipeline = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
custom_pipeline="./path/to/pipeline_directory/",
clip_model=clip_model,
feature_extractor=feature_extractor,
@@ -97,7 +97,7 @@ For example, to load from the main branch:
```py
pipeline = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
custom_pipeline="clip_guided_stable_diffusion",
custom_revision="main",
clip_model=clip_model,
@@ -113,7 +113,7 @@ For example, to load from a previous version of Diffusers like v0.25.0:
```py
pipeline = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
custom_pipeline="clip_guided_stable_diffusion",
custom_revision="v0.25.0",
clip_model=clip_model,
@@ -235,7 +235,7 @@ from diffusers import DiffusionPipeline, DDIMScheduler
from diffusers.utils import load_image
pipeline = DiffusionPipeline.from_pretrained(
"Lykon/dreamshaper-8-inpainting",
"stable-diffusion-v1-5/stable-diffusion-v1-5-inpainting",
custom_pipeline="hd_painter"
)
pipeline.scheduler = DDIMScheduler.from_config(pipeline.scheduler.config)

View File

@@ -30,7 +30,7 @@ import torch
from diffusers import DiffusionPipeline
pipeline = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, safety_checker=None
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, safety_checker=None
).to("cuda")
pipeline.enable_freeu(s1=0.9, s2=0.2, b1=1.5, b2=1.6)
generator = torch.Generator(device="cpu").manual_seed(33)

View File

@@ -66,7 +66,7 @@ make_image_grid([init_image, image], rows=1, cols=2)
## Popular models
The most popular image-to-image models are [Stable Diffusion v1.5](https://huggingface.co/runwayml/stable-diffusion-v1-5), [Stable Diffusion XL (SDXL)](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0), and [Kandinsky 2.2](https://huggingface.co/kandinsky-community/kandinsky-2-2-decoder). The results from the Stable Diffusion and Kandinsky models vary due to their architecture differences and training process; you can generally expect SDXL to produce higher quality images than Stable Diffusion v1.5. Let's take a quick look at how to use each of these models and compare their results.
The most popular image-to-image models are [Stable Diffusion v1.5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5), [Stable Diffusion XL (SDXL)](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0), and [Kandinsky 2.2](https://huggingface.co/kandinsky-community/kandinsky-2-2-decoder). The results from the Stable Diffusion and Kandinsky models vary due to their architecture differences and training process; you can generally expect SDXL to produce higher quality images than Stable Diffusion v1.5. Let's take a quick look at how to use each of these models and compare their results.
### Stable Diffusion v1.5
@@ -78,7 +78,7 @@ from diffusers import AutoPipelineForImage2Image
from diffusers.utils import make_image_grid, load_image
pipeline = AutoPipelineForImage2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
)
pipeline.enable_model_cpu_offload()
# remove following line if xFormers is not installed or you have PyTorch 2.0 or higher installed
@@ -203,7 +203,7 @@ from diffusers import AutoPipelineForImage2Image
from diffusers.utils import make_image_grid, load_image
pipeline = AutoPipelineForImage2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
)
pipeline.enable_model_cpu_offload()
# remove following line if xFormers is not installed or you have PyTorch 2.0 or higher installed
@@ -247,7 +247,7 @@ from diffusers import AutoPipelineForImage2Image
from diffusers.utils import make_image_grid, load_image
pipeline = AutoPipelineForImage2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
)
pipeline.enable_model_cpu_offload()
# remove following line if xFormers is not installed or you have PyTorch 2.0 or higher installed
@@ -334,7 +334,7 @@ import torch
from diffusers.utils import make_image_grid
pipeline = AutoPipelineForText2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
)
pipeline.enable_model_cpu_offload()
# remove following line if xFormers is not installed or you have PyTorch 2.0 or higher installed
@@ -370,7 +370,7 @@ from diffusers import AutoPipelineForImage2Image
from diffusers.utils import make_image_grid, load_image
pipeline = AutoPipelineForImage2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
)
pipeline.enable_model_cpu_offload()
# remove following line if xFormers is not installed or you have PyTorch 2.0 or higher installed
@@ -433,7 +433,7 @@ from diffusers import AutoPipelineForImage2Image
from diffusers.utils import make_image_grid, load_image
pipeline = AutoPipelineForImage2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
)
pipeline.enable_model_cpu_offload()
# remove following line if xFormers is not installed or you have PyTorch 2.0 or higher installed
@@ -499,7 +499,7 @@ from diffusers import AutoPipelineForImage2Image
import torch
pipeline = AutoPipelineForImage2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
)
pipeline.enable_model_cpu_offload()
# remove following line if xFormers is not installed or you have PyTorch 2.0 or higher installed
@@ -536,7 +536,7 @@ import torch
controlnet = ControlNetModel.from_pretrained("lllyasviel/control_v11f1p_sd15_depth", torch_dtype=torch.float16, variant="fp16", use_safetensors=True)
pipeline = AutoPipelineForImage2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", controlnet=controlnet, torch_dtype=torch.float16, variant="fp16", use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", controlnet=controlnet, torch_dtype=torch.float16, variant="fp16", use_safetensors=True
)
pipeline.enable_model_cpu_offload()
# remove following line if xFormers is not installed or you have PyTorch 2.0 or higher installed

View File

@@ -419,7 +419,7 @@ canny_image = Image.fromarray(image)
controlnet = ControlNetModel.from_pretrained("lllyasviel/sd-controlnet-canny", torch_dtype=torch.float16)
pipe = StableDiffusionControlNetPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
controlnet=controlnet,
torch_dtype=torch.float16,
safety_checker=None,

View File

@@ -35,7 +35,7 @@ This guide will show you how to perform inference with TCD-LoRAs for a variety o
| Base model | TCD-LoRA checkpoint |
|-------------------------------------------------------------------------------------------------|----------------------------------------------------------------|
| [stable-diffusion-v1-5](https://huggingface.co/runwayml/stable-diffusion-v1-5) | [TCD-SD15](https://huggingface.co/h1t/TCD-SD15-LoRA) |
| [stable-diffusion-v1-5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) | [TCD-SD15](https://huggingface.co/h1t/TCD-SD15-LoRA) |
| [stable-diffusion-2-1-base](https://huggingface.co/stabilityai/stable-diffusion-2-1-base) | [TCD-SD21-base](https://huggingface.co/h1t/TCD-SD21-base-LoRA) |
| [stable-diffusion-xl-base-1.0](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0) | [TCD-SDXL](https://huggingface.co/h1t/TCD-SDXL-LoRA) |

View File

@@ -95,7 +95,7 @@ from diffusers import AutoPipelineForInpainting
from diffusers.utils import load_image
from PIL import Image
pipeline = AutoPipelineForInpainting.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16).to('cuda')
pipeline = AutoPipelineForInpainting.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16).to('cuda')
mask = load_image("https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/seashore_mask.png")
blurred_mask = pipeline.mask_processor.blur(mask, blur_factor=33)
@@ -216,12 +216,13 @@ make_image_grid([init_image, mask_image, image], rows=1, cols=3)
## Non-inpaint specific checkpoints
So far, this guide has used inpaint specific checkpoints such as [runwayml/stable-diffusion-inpainting](https://huggingface.co/runwayml/stable-diffusion-inpainting). But you can also use regular checkpoints like [runwayml/stable-diffusion-v1-5](https://huggingface.co/runwayml/stable-diffusion-v1-5). Let's compare the results of the two checkpoints.
So far, this guide has used inpaint specific checkpoints such as [stable-diffusion-v1-5/stable-diffusion-inpainting](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-inpainting). But you can also use regular checkpoints like [stable-diffusion-v1-5/stable-diffusion-v1-5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5). Let's compare the results of the two checkpoints.
The image on the left is generated from a regular checkpoint, and the image on the right is from an inpaint checkpoint. You'll immediately notice the image on the left is not as clean, and you can still see the outline of the area the model is supposed to inpaint. The image on the right is much cleaner and the inpainted area appears more natural.
<hfoptions id="regular-specific">
<hfoption id="runwayml/stable-diffusion-v1-5">
<hfoption id="stable-diffusion-v1-5/stable-diffusion-v1-5">
```py
import torch
@@ -229,7 +230,7 @@ from diffusers import AutoPipelineForInpainting
from diffusers.utils import load_image, make_image_grid
pipeline = AutoPipelineForInpainting.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16"
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16"
)
pipeline.enable_model_cpu_offload()
# remove following line if xFormers is not installed or you have PyTorch 2.0 or higher installed
@@ -276,7 +277,7 @@ make_image_grid([init_image, image], rows=1, cols=2)
<div class="flex gap-4">
<div>
<img class="rounded-xl" src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/non-inpaint-specific.png"/>
<figcaption class="mt-2 text-center text-sm text-gray-500">runwayml/stable-diffusion-v1-5</figcaption>
<figcaption class="mt-2 text-center text-sm text-gray-500">stable-diffusion-v1-5/stable-diffusion-v1-5</figcaption>
</div>
<div>
<img class="rounded-xl" src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/inpaint-specific.png"/>
@@ -287,7 +288,7 @@ make_image_grid([init_image, image], rows=1, cols=2)
However, for more basic tasks like erasing an object from an image (like the rocks in the road for example), a regular checkpoint yields pretty good results. There isn't as noticeable of difference between the regular and inpaint checkpoint.
<hfoptions id="inpaint">
<hfoption id="runwayml/stable-diffusion-v1-5">
<hfoption id="stable-diffusion-v1-5/stable-diffusion-v1-5">
```py
import torch
@@ -295,7 +296,7 @@ from diffusers import AutoPipelineForInpainting
from diffusers.utils import load_image, make_image_grid
pipeline = AutoPipelineForInpainting.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16"
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16"
)
pipeline.enable_model_cpu_offload()
# remove following line if xFormers is not installed or you have PyTorch 2.0 or higher installed
@@ -338,7 +339,7 @@ make_image_grid([init_image, image], rows=1, cols=2)
<div class="flex gap-4">
<div>
<img class="rounded-xl" src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/regular-inpaint-basic.png"/>
<figcaption class="mt-2 text-center text-sm text-gray-500">runwayml/stable-diffusion-v1-5</figcaption>
<figcaption class="mt-2 text-center text-sm text-gray-500">stable-diffusion-v1-5/stable-diffusion-v1-5</figcaption>
</div>
<div>
<img class="rounded-xl" src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/specific-inpaint-basic.png"/>
@@ -518,7 +519,7 @@ from diffusers.utils import load_image
from PIL import Image
generator = torch.Generator(device='cuda').manual_seed(0)
pipeline = AutoPipelineForInpainting.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16).to('cuda')
pipeline = AutoPipelineForInpainting.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16).to('cuda')
base = load_image("https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/seashore.png")
mask = load_image("https://huggingface.co/datasets/YiYiXu/testing-images/resolve/main/seashore_mask.png")
@@ -554,7 +555,7 @@ from diffusers import AutoPipelineForText2Image, AutoPipelineForInpainting
from diffusers.utils import load_image, make_image_grid
pipeline = AutoPipelineForText2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
)
pipeline.enable_model_cpu_offload()
# remove following line if xFormers is not installed or you have PyTorch 2.0 or higher installed

View File

@@ -380,7 +380,7 @@ from diffusers import StableDiffusionPipeline, DDIMScheduler
from diffusers.utils import load_image
pipeline = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
).to("cuda")
pipeline.scheduler = DDIMScheduler.from_config(pipeline.scheduler.config)
@@ -421,7 +421,7 @@ from diffusers.utils import load_image
from insightface.app import FaceAnalysis
pipeline = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
).to("cuda")
pipeline.scheduler = DDIMScheduler.from_config(pipeline.scheduler.config)
@@ -617,7 +617,7 @@ controlnet_model_path = "lllyasviel/control_v11f1p_sd15_depth"
controlnet = ControlNetModel.from_pretrained(controlnet_model_path, torch_dtype=torch.float16)
pipeline = StableDiffusionControlNetPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", controlnet=controlnet, torch_dtype=torch.float16)
"stable-diffusion-v1-5/stable-diffusion-v1-5", controlnet=controlnet, torch_dtype=torch.float16)
pipeline.to("cuda")
pipeline.load_ip_adapter("h94/IP-Adapter", subfolder="models", weight_name="ip-adapter_sd15.bin")
```

View File

@@ -41,7 +41,7 @@ The [`DiffusionPipeline`] class is a simple and generic way to load the latest t
```python
from diffusers import DiffusionPipeline
pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", use_safetensors=True)
pipeline = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", use_safetensors=True)
```
This same checkpoint can also be used for an image-to-image task. The [`DiffusionPipeline`] class can handle any task as long as you provide the appropriate inputs. For example, for an image-to-image task, you need to pass an initial image to the pipeline.
@@ -49,7 +49,7 @@ This same checkpoint can also be used for an image-to-image task. The [`Diffusio
```py
from diffusers import DiffusionPipeline
pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", use_safetensors=True)
pipeline = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", use_safetensors=True)
init_image = load_image("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/img2img-init.png")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
@@ -64,7 +64,7 @@ Checkpoints can be loaded by their specific pipeline class if you already know i
```python
from diffusers import StableDiffusionPipeline
pipeline = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", use_safetensors=True)
pipeline = StableDiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", use_safetensors=True)
```
This same checkpoint may also be used for another task like image-to-image. To differentiate what task you want to use the checkpoint for, you have to use the corresponding task-specific pipeline class. For example, to use the same checkpoint for image-to-image, use the [`StableDiffusionImg2ImgPipeline`] class.
@@ -72,7 +72,7 @@ This same checkpoint may also be used for another task like image-to-image. To d
```py
from diffusers import StableDiffusionImg2ImgPipeline
pipeline = StableDiffusionImg2ImgPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", use_safetensors=True)
pipeline = StableDiffusionImg2ImgPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", use_safetensors=True)
```
</hfoption>
@@ -101,7 +101,7 @@ To load a pipeline locally, use [git-lfs](https://git-lfs.github.com/) to manual
```bash
git-lfs install
git clone https://huggingface.co/runwayml/stable-diffusion-v1-5
git clone https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5
```
This creates a local folder, ./stable-diffusion-v1-5, on your disk and you should pass its path to [`~DiffusionPipeline.from_pretrained`].
@@ -311,7 +311,7 @@ Diffusers implements a [safety checker](https://github.com/huggingface/diffusers
```python
from diffusers import DiffusionPipeline
pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", safety_checker=None, use_safetensors=True)
pipeline = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", safety_checker=None, use_safetensors=True)
"""
You have disabled the safety checker for <class 'diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline'> by passing `safety_checker=None`. Ensure that you abide by the conditions of the Stable Diffusion license and do not expose unfiltered results in services or applications open to the public. Both the diffusers team and Hugging Face strongly recommend keeping the safety filter enabled in all public-facing circumstances, disabling it only for use cases that involve analyzing network behavior or auditing its results. For more information, please have a look at https://github.com/huggingface/diffusers/pull/254 .
"""
@@ -341,7 +341,7 @@ There are two important arguments for loading variants:
If you only set `torch_dtype=torch.float16`, the default fp32 weights are downloaded first and then converted to fp16.
- `variant` specifies which files should be loaded from the repository. For example, if you want to load a non-EMA variant of a UNet from [runwayml/stable-diffusion-v1-5](https://hf.co/runwayml/stable-diffusion-v1-5/tree/main/unet), set `variant="non_ema"` to download the `non_ema` file.
- `variant` specifies which files should be loaded from the repository. For example, if you want to load a non-EMA variant of a UNet from [stable-diffusion-v1-5/stable-diffusion-v1-5](https://hf.co/stable-diffusion-v1-5/stable-diffusion-v1-5/tree/main/unet), set `variant="non_ema"` to download the `non_ema` file.
<hfoptions id="variants">
<hfoption id="fp16">
@@ -351,7 +351,7 @@ from diffusers import DiffusionPipeline
import torch
pipeline = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", variant="fp16", torch_dtype=torch.float16, use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", variant="fp16", torch_dtype=torch.float16, use_safetensors=True
)
```
@@ -360,7 +360,7 @@ pipeline = DiffusionPipeline.from_pretrained(
```py
pipeline = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", variant="non_ema", use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", variant="non_ema", use_safetensors=True
)
```
@@ -375,14 +375,14 @@ Use the `variant` parameter in the [`DiffusionPipeline.save_pretrained`] method
```python
from diffusers import DiffusionPipeline
pipeline.save_pretrained("runwayml/stable-diffusion-v1-5", variant="fp16")
pipeline.save_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", variant="fp16")
```
</hfoption>
<hfoption id="non_ema">
```py
pipeline.save_pretrained("runwayml/stable-diffusion-v1-5", variant="non_ema")
pipeline.save_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", variant="non_ema")
```
</hfoption>
@@ -408,12 +408,12 @@ As a class method, [`DiffusionPipeline.from_pretrained`] is responsible for two
- Download the latest version of the folder structure required for inference and cache it. If the latest folder structure is available in the local cache, [`DiffusionPipeline.from_pretrained`] reuses the cache and won't redownload the files.
- Load the cached weights into the correct pipeline [class](../api/pipelines/overview#diffusers-summary) - retrieved from the `model_index.json` file - and return an instance of it.
The pipelines' underlying folder structure corresponds directly with their class instances. For example, the [`StableDiffusionPipeline`] corresponds to the folder structure in [`runwayml/stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5).
The pipelines' underlying folder structure corresponds directly with their class instances. For example, the [`StableDiffusionPipeline`] corresponds to the folder structure in [`stable-diffusion-v1-5/stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5).
```python
from diffusers import DiffusionPipeline
repo_id = "runwayml/stable-diffusion-v1-5"
repo_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
pipeline = DiffusionPipeline.from_pretrained(repo_id, use_safetensors=True)
print(pipeline)
```
@@ -461,7 +461,7 @@ StableDiffusionPipeline {
}
```
Compare the components of the pipeline instance to the [`runwayml/stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5/tree/main) folder structure, and you'll see there is a separate folder for each of the components in the repository:
Compare the components of the pipeline instance to the [`stable-diffusion-v1-5/stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5/tree/main) folder structure, and you'll see there is a separate folder for each of the components in the repository:
```
.
@@ -524,7 +524,7 @@ CLIPTokenizer(
)
```
Every pipeline expects a [`model_index.json`](https://huggingface.co/runwayml/stable-diffusion-v1-5/blob/main/model_index.json) file that tells the [`DiffusionPipeline`]:
Every pipeline expects a [`model_index.json`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5/blob/main/model_index.json) file that tells the [`DiffusionPipeline`]:
- which pipeline class to load from `_class_name`
- which version of 🧨 Diffusers was used to create the model in `_diffusers_version`

View File

@@ -54,7 +54,7 @@ Because textual inversion creates embeddings, it cannot be used on its own like
from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
pipeline = AutoPipelineForText2Image.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
```
Now you can load the textual inversion embeddings with the [`~loaders.TextualInversionLoaderMixin.load_textual_inversion`] method and generate some images. Let's load the [sd-concepts-library/gta5-artwork](https://huggingface.co/sd-concepts-library/gta5-artwork) embeddings and you'll need to include the special word `<gta5-artwork>` in your prompt to trigger it:
@@ -269,7 +269,7 @@ from diffusers import AutoPipelineForText2Image
import torch
from diffusers.utils import load_image
pipeline = AutoPipelineForText2Image.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
pipeline = AutoPipelineForText2Image.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
```
Then load the IP-Adapter weights and add it to the pipeline with the [`~loaders.IPAdapterMixin.load_ip_adapter`] method.
@@ -350,7 +350,7 @@ image_encoder = CLIPVisionModelWithProjection.from_pretrained(
)
pipeline = AutoPipelineForText2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
image_encoder=image_encoder,
torch_dtype=torch.float16
).to("cuda")

View File

@@ -37,7 +37,7 @@ Make sure you have the [Safetensors](https://hf.co/docs/safetensors) library ins
Safetensors stores weights in a safetensors file. Diffusers loads safetensors files by default if they're available and the Safetensors library is installed. There are two ways safetensors files can be organized:
1. Diffusers-multifolder layout: there may be several separate safetensors files, one for each pipeline component (text encoder, UNet, VAE), organized in subfolders (check out the [runwayml/stable-diffusion-v1-5](https://hf.co/runwayml/stable-diffusion-v1-5/tree/main) repository as an example)
1. Diffusers-multifolder layout: there may be several separate safetensors files, one for each pipeline component (text encoder, UNet, VAE), organized in subfolders (check out the [stable-diffusion-v1-5/stable-diffusion-v1-5](https://hf.co/stable-diffusion-v1-5/stable-diffusion-v1-5/tree/main) repository as an example)
2. single-file layout: all the model weights may be saved in a single file (check out the [WarriorMama777/OrangeMixs](https://hf.co/WarriorMama777/OrangeMixs/tree/main/Models/AbyssOrangeMix) repository as an example)
<hfoptions id="safetensors">
@@ -49,7 +49,7 @@ Use the [`~DiffusionPipeline.from_pretrained`] method to load a model with safet
from diffusers import DiffusionPipeline
pipeline = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
use_safetensors=True
)
```
@@ -118,7 +118,7 @@ Use the [`~loaders.FromSingleFileMixin.from_single_file`] method to directly loa
from diffusers import StableDiffusionPipeline
pipeline = StableDiffusionPipeline.from_single_file(
"https://huggingface.co/runwayml/stable-diffusion-v1-5/blob/main/v1-5-pruned.ckpt"
"https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5/blob/main/v1-5-pruned.ckpt"
)
```
@@ -314,7 +314,7 @@ Or you could use a ControlNet model in the pipeline.
```py
from diffusers import StableDiffusionControlNetPipeline, ControlNetModel
ckpt_path = "https://huggingface.co/runwayml/stable-diffusion-v1-5/blob/main/v1-5-pruned-emaonly.safetensors"
ckpt_path = "https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5/blob/main/v1-5-pruned-emaonly.safetensors"
controlnet = ControlNetModel.from_pretrained("lllyasviel/control_v11p_sd15_canny")
pipeline = StableDiffusionControlNetPipeline.from_single_file(ckpt_path, controlnet=controlnet)
```

View File

@@ -121,7 +121,7 @@ Now when you run the same pipeline twice, you'll get identical results.
import torch
from diffusers import DDIMScheduler, StableDiffusionPipeline
pipe = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", use_safetensors=True).to("cuda")
pipe = StableDiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", use_safetensors=True).to("cuda")
pipe.scheduler = DDIMScheduler.from_config(pipe.scheduler.config)
g = torch.Generator(device="cuda")
@@ -141,7 +141,7 @@ print("L_inf dist =", abs(result1 - result2).max())
A practical application of creating reproducible pipelines is *deterministic batch generation*. You generate a batch of images and select one image to improve with a more detailed prompt. The main idea is to pass a list of [Generator's](https://pytorch.org/docs/stable/generated/torch.Generator.html) to the pipeline and tie each `Generator` to a seed so you can reuse it.
Let's use the [runwayml/stable-diffusion-v1-5](https://huggingface.co/runwayml/stable-diffusion-v1-5) checkpoint and generate a batch of images.
Let's use the [stable-diffusion-v1-5/stable-diffusion-v1-5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) checkpoint and generate a batch of images.
```py
import torch
@@ -149,7 +149,7 @@ from diffusers import DiffusionPipeline
from diffusers.utils import make_image_grid
pipeline = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True
)
pipeline = pipeline.to("cuda")
```

View File

@@ -16,14 +16,14 @@ specific language governing permissions and limitations under the License.
Diffusion pipelines are a collection of interchangeable schedulers and models that can be mixed and matched to tailor a pipeline to a specific use case. The scheduler encapsulates the entire denoising process such as the number of denoising steps and the algorithm for finding the denoised sample. A scheduler is not parameterized or trained so they don't take very much memory. The model is usually only concerned with the forward pass of going from a noisy input to a less noisy sample.
This guide will show you how to load schedulers and models to customize a pipeline. You'll use the [runwayml/stable-diffusion-v1-5](https://hf.co/runwayml/stable-diffusion-v1-5) checkpoint throughout this guide, so let's load it first.
This guide will show you how to load schedulers and models to customize a pipeline. You'll use the [stable-diffusion-v1-5/stable-diffusion-v1-5](https://hf.co/stable-diffusion-v1-5/stable-diffusion-v1-5) checkpoint throughout this guide, so let's load it first.
```py
import torch
from diffusers import DiffusionPipeline
pipeline = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True
).to("cuda")
```
@@ -56,14 +56,14 @@ For example, to load the [`DDIMScheduler`]:
```py
from diffusers import DDIMScheduler, DiffusionPipeline
ddim = DDIMScheduler.from_pretrained("runwayml/stable-diffusion-v1-5", subfolder="scheduler")
ddim = DDIMScheduler.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", subfolder="scheduler")
```
Then you can pass the newly loaded scheduler to the pipeline.
```python
pipeline = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", scheduler=ddim, torch_dtype=torch.float16, use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", scheduler=ddim, torch_dtype=torch.float16, use_safetensors=True
).to("cuda")
```
@@ -78,7 +78,7 @@ import torch
from diffusers import DiffusionPipeline
pipeline = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True
).to("cuda")
prompt = "A photograph of an astronaut riding a horse on Mars, high resolution, high definition."
@@ -180,11 +180,11 @@ from flax.training.common_utils import shard
from diffusers import FlaxStableDiffusionPipeline, FlaxDPMSolverMultistepScheduler
scheduler, scheduler_state = FlaxDPMSolverMultistepScheduler.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
subfolder="scheduler"
)
pipeline, params = FlaxStableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
scheduler=scheduler,
variant="bf16",
dtype=jax.numpy.bfloat16,
@@ -216,12 +216,12 @@ images = pipeline.numpy_to_pil(np.asarray(images.reshape((num_samples,) + images
Models are loaded from the [`ModelMixin.from_pretrained`] method, which downloads and caches the latest version of the model weights and configurations. If the latest files are available in the local cache, [`~ModelMixin.from_pretrained`] reuses files in the cache instead of re-downloading them.
Models can be loaded from a subfolder with the `subfolder` argument. For example, the model weights for [runwayml/stable-diffusion-v1-5](https://hf.co/runwayml/stable-diffusion-v1-5) are stored in the [unet](https://hf.co/runwayml/stable-diffusion-v1-5/tree/main/unet) subfolder.
Models can be loaded from a subfolder with the `subfolder` argument. For example, the model weights for [stable-diffusion-v1-5/stable-diffusion-v1-5](https://hf.co/stable-diffusion-v1-5/stable-diffusion-v1-5) are stored in the [unet](https://hf.co/stable-diffusion-v1-5/stable-diffusion-v1-5/tree/main/unet) subfolder.
```python
from diffusers import UNet2DConditionModel
unet = UNet2DConditionModel.from_pretrained("runwayml/stable-diffusion-v1-5", subfolder="unet", use_safetensors=True)
unet = UNet2DConditionModel.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", subfolder="unet", use_safetensors=True)
```
They can also be directly loaded from a [repository](https://huggingface.co/google/ddpm-cifar10-32/tree/main).
@@ -238,7 +238,7 @@ To load and save model variants, specify the `variant` argument in [`ModelMixin.
from diffusers import UNet2DConditionModel
unet = UNet2DConditionModel.from_pretrained(
"runwayml/stable-diffusion-v1-5", subfolder="unet", variant="non_ema", use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", subfolder="unet", variant="non_ema", use_safetensors=True
)
unet.save_pretrained("./local-unet", variant="non_ema")
```

View File

@@ -73,7 +73,7 @@ from diffusers import StableDiffusionAdapterPipeline, T2IAdapter
adapter = T2IAdapter.from_pretrained("TencentARC/t2iadapter_canny_sd15v2", torch_dtype=torch.float16)
pipeline = StableDiffusionAdapterPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
adapter=adapter,
torch_dtype=torch.float16,
)

View File

@@ -365,7 +365,7 @@ import torch
from diffusers import StableDiffusionControlNetPipeline, ControlNetModel
from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_zero import CrossFrameAttnProcessor
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
controlnet = ControlNetModel.from_pretrained("lllyasviel/sd-controlnet-openpose", torch_dtype=torch.float16)
pipeline = StableDiffusionControlNetPipeline.from_pretrained(
model_id, controlnet=controlnet, torch_dtype=torch.float16
@@ -411,7 +411,7 @@ import torch
from diffusers import StableDiffusionControlNetPipeline, ControlNetModel
from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_zero import CrossFrameAttnProcessor
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
controlnet = ControlNetModel.from_pretrained("lllyasviel/sd-controlnet-canny", torch_dtype=torch.float16)
pipeline = StableDiffusionControlNetPipeline.from_pretrained(
model_id, controlnet=controlnet, torch_dtype=torch.float16

View File

@@ -31,7 +31,7 @@ from diffusers.utils import make_image_grid
Pick a Stable Diffusion checkpoint and a pre-learned concept from the [Stable Diffusion Conceptualizer](https://huggingface.co/spaces/sd-concepts-library/stable-diffusion-conceptualizer):
```py
pretrained_model_name_or_path = "runwayml/stable-diffusion-v1-5"
pretrained_model_name_or_path = "stable-diffusion-v1-5/stable-diffusion-v1-5"
repo_id_embeds = "sd-concepts-library/cat-toy"
```

View File

@@ -366,7 +366,7 @@ from diffusers import StableDiffusionPipeline
from compel import Compel, DiffusersTextualInversionManager
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16,
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16,
use_safetensors=True, variant="fp16").to("cuda")
pipe.load_textual_inversion("sd-concepts-library/midjourney-style")
```

View File

@@ -116,7 +116,7 @@ As you can see, this is already more complex than the DDPM pipeline which only c
</Tip>
Now that you know what you need for the Stable Diffusion pipeline, load all these components with the [`~ModelMixin.from_pretrained`] method. You can find them in the pretrained [`runwayml/stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5) checkpoint, and each component is stored in a separate subfolder:
Now that you know what you need for the Stable Diffusion pipeline, load all these components with the [`~ModelMixin.from_pretrained`] method. You can find them in the pretrained [`stable-diffusion-v1-5/stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) checkpoint, and each component is stored in a separate subfolder:
```py
>>> from PIL import Image

View File

@@ -54,7 +54,7 @@ specific language governing permissions and limitations under the License.
まず、[`DiffusionPipeline`]のインスタンスを作成し、ダウンロードしたいパイプラインのチェックポイントを指定します。
この[`DiffusionPipeline`]はHugging Face Hubに保存されている任意の[チェックポイント](https://huggingface.co/models?library=diffusers&sort=downloads)を使用することができます。
この案内では、[`stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5)チェックポイントでテキストから画像へ生成します。
この案内では、[`stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5)チェックポイントでテキストから画像へ生成します。
<Tip warning={true}>
@@ -67,7 +67,7 @@ specific language governing permissions and limitations under the License.
```python
>>> from diffusers import DiffusionPipeline
>>> pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", use_safetensors=True)
>>> pipeline = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", use_safetensors=True)
```
[`DiffusionPipeline`]は全てのモデリング、トークン化、スケジューリングコンポーネントをダウンロードしてキャッシュします。Stable Diffusionパイプラインは[`UNet2DConditionModel`]と[`PNDMScheduler`]などで構成されています:
@@ -123,7 +123,7 @@ PyTorchと同じように、ジェネレータオブジェクトをGPUに移す
```bash
!git lfs install
!git clone https://huggingface.co/runwayml/stable-diffusion-v1-5
!git clone https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5
```
保存したウェイトをパイプラインにロードします:
@@ -141,7 +141,7 @@ PyTorchと同じように、ジェネレータオブジェクトをGPUに移す
```py
>>> from diffusers import EulerDiscreteScheduler
>>> pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", use_safetensors=True)
>>> pipeline = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", use_safetensors=True)
>>> pipeline.scheduler = EulerDiscreteScheduler.from_config(pipeline.scheduler.config)
```

View File

@@ -20,12 +20,12 @@ specific language governing permissions and limitations under the License.
このチュートリアルでは、[`DiffusionPipeline`]を用いて、より速く、より良い計算を行う方法を説明します。
まず、[`runwayml/stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5)モデルをロードします:
まず、[`stable-diffusion-v1-5/stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5)モデルをロードします:
```python
from diffusers import DiffusionPipeline
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
pipeline = DiffusionPipeline.from_pretrained(model_id, use_safetensors=True)
```

View File

@@ -12,7 +12,7 @@ specific language governing permissions and limitations under the License.
# AutoPipeline
Diffusersは様々なタスクをこなすことができ、テキストから画像、画像から画像、画像の修復など、複数のタスクに対して同じように事前学習された重みを再利用することができます。しかし、ライブラリや拡散モデルに慣れていない場合、どのタスクにどのパイプラインを使えばいいのかがわかりにくいかもしれません。例えば、 [runwayml/stable-diffusion-v1-5](https://huggingface.co/runwayml/stable-diffusion-v1-5) チェックポイントをテキストから画像に変換するために使用している場合、それぞれ[`StableDiffusionImg2ImgPipeline`]クラスと[`StableDiffusionInpaintPipeline`]クラスでチェックポイントをロードすることで、画像から画像や画像の修復にも使えることを知らない可能性もあります。
Diffusersは様々なタスクをこなすことができ、テキストから画像、画像から画像、画像の修復など、複数のタスクに対して同じように事前学習された重みを再利用することができます。しかし、ライブラリや拡散モデルに慣れていない場合、どのタスクにどのパイプラインを使えばいいのかがわかりにくいかもしれません。例えば、 [stable-diffusion-v1-5/stable-diffusion-v1-5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) チェックポイントをテキストから画像に変換するために使用している場合、それぞれ[`StableDiffusionImg2ImgPipeline`]クラスと[`StableDiffusionInpaintPipeline`]クラスでチェックポイントをロードすることで、画像から画像や画像の修復にも使えることを知らない可能性もあります。
`AutoPipeline` クラスは、🤗 Diffusers の様々なパイプラインをよりシンプルするために設計されています。この汎用的でタスク重視のパイプラインによってタスクそのものに集中することができます。`AutoPipeline` は、使用するべき正しいパイプラインクラスを自動的に検出するため、特定のパイプラインクラス名を知らなくても、タスクのチェックポイントを簡単にロードできます。
@@ -25,14 +25,14 @@ Diffusersは様々なタスクをこなすことができ、テキストから
このチュートリアルでは、`AutoPipeline` を使用して、事前に学習された重みが与えられたときに、特定のタスクを読み込むためのパイプラインクラスを自動的に推測する方法を示します。
## タスクに合わせてAutoPipeline を選択する
まずはチェックポイントを選ぶことから始めましょう。例えば、 [runwayml/stable-diffusion-v1-5](https://huggingface.co/runwayml/stable-diffusion-v1-5) チェックポイントでテキストから画像への変換したいなら、[`AutoPipelineForText2Image`]を使います:
まずはチェックポイントを選ぶことから始めましょう。例えば、 [stable-diffusion-v1-5/stable-diffusion-v1-5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) チェックポイントでテキストから画像への変換したいなら、[`AutoPipelineForText2Image`]を使います:
```py
from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True
).to("cuda")
prompt = "peasant and dragon combat, wood cutting style, viking era, bevel with rune"
@@ -46,7 +46,7 @@ image
[`AutoPipelineForText2Image`] を具体的に見ていきましょう:
1. [`model_index.json`](https://huggingface.co/runwayml/stable-diffusion-v1-5/blob/main/model_index.json) ファイルから `"stable-diffusion"` クラスを自動的に検出します。
1. [`model_index.json`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5/blob/main/model_index.json) ファイルから `"stable-diffusion"` クラスを自動的に検出します。
2. `"stable-diffusion"` のクラス名に基づいて、テキストから画像へ変換する [`StableDiffusionPipeline`] を読み込みます。
同様に、画像から画像へ変換する場合、[`AutoPipelineForImage2Image`] は `model_index.json` ファイルから `"stable-diffusion"` チェックポイントを検出し、対応する [`StableDiffusionImg2ImgPipeline`] を読み込みます。また、入力画像にノイズの量やバリエーションの追加を決めるための強さなど、パイプラインクラスに固有の追加引数を渡すこともできます:
@@ -59,7 +59,7 @@ from PIL import Image
from io import BytesIO
pipeline = AutoPipelineForImage2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
use_safetensors=True,
).to("cuda")
@@ -128,7 +128,7 @@ from diffusers import AutoPipelineForText2Image, AutoPipelineForImage2Image
import torch
pipeline_text2img = AutoPipelineForText2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True
)
print(type(pipeline_text2img))
"<class 'diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline'>"
@@ -148,7 +148,7 @@ from diffusers import AutoPipelineForText2Image, AutoPipelineForImage2Image
import torch
pipeline_text2img = AutoPipelineForText2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
use_safetensors=True,
requires_safety_checker=False,

View File

@@ -88,7 +88,7 @@ images = sd_pipeline(sample_prompts, num_images_per_prompt=1, generator=generato
![parti-prompts-14](https://huggingface.co/datasets/diffusers/docs-images/resolve/main/evaluation_diffusion_models/parti-prompts-14.png)
`num_images_per_prompt`를 설정하여 동일한 프롬프트에 대해 다른 이미지를 비교할 수도 있습니다. 다른 체크포인트([v1-5](https://huggingface.co/runwayml/stable-diffusion-v1-5))로 동일한 파이프라인을 실행하면 다음과 같은 결과가 나옵니다:
`num_images_per_prompt`를 설정하여 동일한 프롬프트에 대해 다른 이미지를 비교할 수도 있습니다. 다른 체크포인트([v1-5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5))로 동일한 파이프라인을 실행하면 다음과 같은 결과가 나옵니다:
![parti-prompts-15](https://huggingface.co/datasets/diffusers/docs-images/resolve/main/evaluation_diffusion_models/parti-prompts-15.png)
@@ -170,10 +170,10 @@ generator = torch.manual_seed(seed)
images = sd_pipeline(prompts, num_images_per_prompt=1, generator=generator, output_type="np").images
```
그런 다음 [v1-5 checkpoint](https://huggingface.co/runwayml/stable-diffusion-v1-5)를 로드하여 이미지를 생성합니다:
그런 다음 [v1-5 checkpoint](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5)를 로드하여 이미지를 생성합니다:
```python
model_ckpt_1_5 = "runwayml/stable-diffusion-v1-5"
model_ckpt_1_5 = "stable-diffusion-v1-5/stable-diffusion-v1-5"
sd_pipeline_1_5 = StableDiffusionPipeline.from_pretrained(model_ckpt_1_5, torch_dtype=weight_dtype).to(device)
images_1_5 = sd_pipeline_1_5(prompts, num_images_per_prompt=1, generator=generator, output_type="np").images
@@ -191,7 +191,7 @@ print(f"CLIP Score with v-1-5: {sd_clip_score_1_5}")
# CLIP Score with v-1-5: 36.2137
```
[v1-5](https://huggingface.co/runwayml/stable-diffusion-v1-5) 체크포인트가 이전 버전보다 더 나은 성능을 보이는 것 같습니다. 그러나 CLIP 점수를 계산하기 위해 사용한 프롬프트의 수가 상당히 적습니다. 보다 실용적인 평가를 위해서는 이 수를 훨씬 높게 설정하고, 프롬프트를 다양하게 사용해야 합니다.
[v1-5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) 체크포인트가 이전 버전보다 더 나은 성능을 보이는 것 같습니다. 그러나 CLIP 점수를 계산하기 위해 사용한 프롬프트의 수가 상당히 적습니다. 보다 실용적인 평가를 위해서는 이 수를 훨씬 높게 설정하고, 프롬프트를 다양하게 사용해야 합니다.
<Tip warning={true}>

View File

@@ -59,7 +59,7 @@ Diffusers에서는 이러한 철학을 파이프라인과 스케줄러에 모두
다음과 같은 설계 원칙을 따릅니다:
- 파이프라인은 단일 파일 정책을 따릅니다. 모든 파이프라인은 src/diffusers/pipelines의 개별 디렉토리에 있습니다. 하나의 파이프라인 폴더는 하나의 diffusion 논문/프로젝트/릴리스에 해당합니다. 여러 파이프라인 파일은 하나의 파이프라인 폴더에 모을 수 있습니다. 예를 들어 [`src/diffusers/pipelines/stable-diffusion`](https://github.com/huggingface/diffusers/tree/main/src/diffusers/pipelines/stable_diffusion)에서 그렇게 하고 있습니다. 파이프라인이 유사한 기능을 공유하는 경우, [# Copied from mechanism](https://github.com/huggingface/diffusers/blob/125d783076e5bd9785beb05367a2d2566843a271/src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_img2img.py#L251)을 사용할 수 있습니다.
- 파이프라인은 모두 [`DiffusionPipeline`]을 상속합니다.
- 각 파이프라인은 서로 다른 모델 및 스케줄러 구성 요소로 구성되어 있으며, 이는 [`model_index.json` 파일](https://huggingface.co/runwayml/stable-diffusion-v1-5/blob/main/model_index.json)에 문서화되어 있으며, 파이프라인의 속성 이름과 동일한 이름으로 액세스할 수 있으며, [`DiffusionPipeline.components`](https://huggingface.co/docs/diffusers/main/en/api/diffusion_pipeline#diffusers.DiffusionPipeline.components) 함수를 통해 파이프라인 간에 공유할 수 있습니다.
- 각 파이프라인은 서로 다른 모델 및 스케줄러 구성 요소로 구성되어 있으며, 이는 [`model_index.json` 파일](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5/blob/main/model_index.json)에 문서화되어 있으며, 파이프라인의 속성 이름과 동일한 이름으로 액세스할 수 있으며, [`DiffusionPipeline.components`](https://huggingface.co/docs/diffusers/main/en/api/diffusion_pipeline#diffusers.DiffusionPipeline.components) 함수를 통해 파이프라인 간에 공유할 수 있습니다.
- 각 파이프라인은 [`DiffusionPipeline.from_pretrained`](https://huggingface.co/docs/diffusers/main/en/api/diffusion_pipeline#diffusers.DiffusionPipeline.from_pretrained) 함수를 통해 로드할 수 있어야 합니다.
- 파이프라인은 추론에**만** 사용되어야 합니다.
- 파이프라인은 매우 가독성이 좋고, 이해하기 쉽고, 쉽게 조정할 수 있도록 설계되어야 합니다.

View File

@@ -103,10 +103,10 @@ python -m python_coreml_stable_diffusion.pipeline --prompt "a photo of an astron
추론 스크립트에서는 Stable Diffusion 모델의 원래 버전인 `CompVis/stable-diffusion-v1-4`를 사용한다고 가정합니다. 다른 모델을 사용하는 경우 추론 명령줄에서 `--model-version` 옵션을 사용하여 해당 허브 ID를 *지정*해야 합니다. 이는 이미 지원되는 모델과 사용자가 직접 학습하거나 파인튜닝한 사용자 지정 모델에 적용됩니다.
예를 들어, [`runwayml/stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5)를 사용하려는 경우입니다:
예를 들어, [`stable-diffusion-v1-5/stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5)를 사용하려는 경우입니다:
```shell
python -m python_coreml_stable_diffusion.pipeline --prompt "a photo of an astronaut riding a horse on mars" --compute-unit ALL -o output --seed 93 -i models/coreml-stable-diffusion-v1-5_original_packages --model-version runwayml/stable-diffusion-v1-5
python -m python_coreml_stable_diffusion.pipeline --prompt "a photo of an astronaut riding a horse on mars" --compute-unit ALL -o output --seed 93 -i models/coreml-stable-diffusion-v1-5_original_packages --model-version stable-diffusion-v1-5/stable-diffusion-v1-5
```

View File

@@ -64,7 +64,7 @@ torch.backends.cuda.matmul.allow_tf32 = True
```Python
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
)
@@ -94,7 +94,7 @@ import torch
from diffusers import StableDiffusionPipeline
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
)
@@ -121,7 +121,7 @@ import torch
from diffusers import StableDiffusionPipeline
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
)
@@ -147,7 +147,7 @@ import torch
from diffusers import StableDiffusionPipeline
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
)
@@ -173,7 +173,7 @@ import torch
from diffusers import StableDiffusionPipeline
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
)
@@ -204,7 +204,7 @@ import torch
from diffusers import StableDiffusionPipeline
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
)
@@ -220,7 +220,7 @@ import torch
from diffusers import StableDiffusionPipeline
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
)
@@ -280,7 +280,7 @@ def generate_inputs():
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
).to("cuda")
unet = pipe.unet
@@ -343,7 +343,7 @@ class UNet2DConditionOutput:
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
).to("cuda")
@@ -396,7 +396,7 @@ from diffusers import StableDiffusionPipeline
import torch
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
torch_dtype=torch.float16,
).to("cuda")

View File

@@ -40,7 +40,7 @@ Diffusers는 Stable Diffusion 추론을 위해 PyTorch `mps`를 사용해 Apple
# `huggingface-cli login`에 로그인되어 있음을 확인
from diffusers import DiffusionPipeline
pipe = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5")
pipe = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5")
pipe = pipe.to("mps")
# 컴퓨터가 64GB 이하의 RAM 램일 때 추천

View File

@@ -31,7 +31,7 @@ PyTorch 모델을 불러오고 즉시 ONNX 형식으로 변환하려는 경우 `
```python
from optimum.onnxruntime import ORTStableDiffusionPipeline
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
pipe = ORTStableDiffusionPipeline.from_pretrained(model_id, export=True)
prompt = "a photo of an astronaut riding a horse on mars"
images = pipe(prompt).images[0]
@@ -42,7 +42,7 @@ pipe.save_pretrained("./onnx-stable-diffusion-v1-5")
[`optimum-cli export`](https://huggingface.co/docs/optimum/main/en/exporters/onnx/usage_guides/export_a_model#exporting-a-model-to-onnx-using-the-cli) 명령어를 사용할 수 있습니다:
```bash
optimum-cli export onnx --model runwayml/stable-diffusion-v1-5 sd_v15_onnx/
optimum-cli export onnx --model stable-diffusion-v1-5/stable-diffusion-v1-5 sd_v15_onnx/
```
그 다음 추론을 수행합니다:

View File

@@ -30,7 +30,7 @@ OpenVINO 모델을 불러오고 OpenVINO 런타임으로 추론을 실행하려
```python
from optimum.intel.openvino import OVStableDiffusionPipeline
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
pipe = OVStableDiffusionPipeline.from_pretrained(model_id, export=True)
prompt = "a photo of an astronaut riding a horse on mars"
images = pipe(prompt).images[0]

View File

@@ -25,7 +25,7 @@ from diffusers import StableDiffusionPipeline
import tomesd
pipeline = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16
"stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16
).to("cuda")
+ tomesd.apply_patch(pipeline, ratio=0.5)

View File

@@ -37,7 +37,7 @@ pip install --upgrade torch diffusers
import torch
from diffusers import DiffusionPipeline
pipe = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16)
pipe = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16)
pipe = pipe.to("cuda")
prompt = "a photo of an astronaut riding a horse on mars"
@@ -51,7 +51,7 @@ pip install --upgrade torch diffusers
from diffusers import DiffusionPipeline
+ from diffusers.models.attention_processor import AttnProcessor2_0
pipe = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
pipe = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
+ pipe.unet.set_attn_processor(AttnProcessor2_0())
prompt = "a photo of an astronaut riding a horse on mars"
@@ -67,7 +67,7 @@ pip install --upgrade torch diffusers
from diffusers import DiffusionPipeline
from diffusers.models.attention_processor import AttnProcessor
pipe = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
pipe = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
pipe.unet.set_default_attn_processor()
prompt = "a photo of an astronaut riding a horse on mars"
@@ -100,7 +100,7 @@ PyTorch 2.0의 효율적인 어텐션 구현과 `torch.compile`을 사용하여
from diffusers import DiffusionPipeline
import torch
path = "runwayml/stable-diffusion-v1-5"
path = "stable-diffusion-v1-5/stable-diffusion-v1-5"
run_compile = True # Set True / False
@@ -133,7 +133,7 @@ response = requests.get(url)
init_image = Image.open(BytesIO(response.content)).convert("RGB")
init_image = init_image.resize((512, 512))
path = "runwayml/stable-diffusion-v1-5"
path = "stable-diffusion-v1-5/stable-diffusion-v1-5"
run_compile = True # Set True / False
@@ -206,7 +206,7 @@ response = requests.get(url)
init_image = Image.open(BytesIO(response.content)).convert("RGB")
init_image = init_image.resize((512, 512))
path = "runwayml/stable-diffusion-v1-5"
path = "stable-diffusion-v1-5/stable-diffusion-v1-5"
run_compile = True # Set True / False
controlnet = ControlNetModel.from_pretrained("lllyasviel/sd-controlnet-canny", torch_dtype=torch.float16)

View File

@@ -53,7 +53,7 @@ Diffusion 모델은 이미지나 오디오와 같은 관심 샘플들을 생성
먼저 [`DiffusionPipeline`]의 인스턴스를 생성하고 다운로드할 파이프라인 체크포인트를 지정합니다.
허깅페이스 허브에 저장된 모든 [checkpoint](https://huggingface.co/models?library=diffusers&sort=downloads)에 대해 [`DiffusionPipeline`]을 사용할 수 있습니다.
이 훑어보기에서는 text-to-image 생성을 위한 [`stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5) 체크포인트를 로드합니다.
이 훑어보기에서는 text-to-image 생성을 위한 [`stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) 체크포인트를 로드합니다.
<Tip warning={true}>
@@ -66,7 +66,7 @@ Diffusion 모델은 이미지나 오디오와 같은 관심 샘플들을 생성
```python
>>> from diffusers import DiffusionPipeline
>>> pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5")
>>> pipeline = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5")
```
The [`DiffusionPipeline`]은 모든 모델링, 토큰화, 스케줄링 컴포넌트를 다운로드하고 캐시합니다. Stable Diffusion Pipeline은 무엇보다도 [`UNet2DConditionModel`]과 [`PNDMScheduler`]로 구성되어 있음을 알 수 있습니다:
@@ -123,7 +123,7 @@ PyTorch에서와 마찬가지로 제너레이터 객체를 GPU로 이동할 수
```bash
!git lfs install
!git clone https://huggingface.co/runwayml/stable-diffusion-v1-5
!git clone https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5
```
그런 다음 저장된 가중치를 파이프라인에 로드합니다:
@@ -141,7 +141,7 @@ PyTorch에서와 마찬가지로 제너레이터 객체를 GPU로 이동할 수
```py
>>> from diffusers import EulerDiscreteScheduler
>>> pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5")
>>> pipeline = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5")
>>> pipeline.scheduler = EulerDiscreteScheduler.from_config(pipeline.scheduler.config)
```

View File

@@ -20,12 +20,12 @@ specific language governing permissions and limitations under the License.
이 튜토리얼에서는 [`DiffusionPipeline`]을 사용하여 더 빠르고 효과적으로 생성하는 방법을 안내합니다.
[`runwayml/stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5) 모델을 불러와서 시작합니다:
[`stable-diffusion-v1-5/stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) 모델을 불러와서 시작합니다:
```python
from diffusers import DiffusionPipeline
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
pipeline = DiffusionPipeline.from_pretrained(model_id)
```

View File

@@ -18,12 +18,12 @@ specific language governing permissions and limitations under the License.
## UNet2DConditionModel 파라미터 구성
[`UNet2DConditionModel`]은 [input sample](https://huggingface.co/docs/diffusers/v0.16.0/en/api/models#diffusers.UNet2DConditionModel.in_channels)에서 4개의 채널을 기본적으로 허용합니다. 예를 들어, [`runwayml/stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5)와 같은 사전학습된 text-to-image 모델을 불러오고 `in_channels`의 수를 확인합니다:
[`UNet2DConditionModel`]은 [input sample](https://huggingface.co/docs/diffusers/v0.16.0/en/api/models#diffusers.UNet2DConditionModel.in_channels)에서 4개의 채널을 기본적으로 허용합니다. 예를 들어, [`stable-diffusion-v1-5/stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5)와 같은 사전학습된 text-to-image 모델을 불러오고 `in_channels`의 수를 확인합니다:
```py
from diffusers import StableDiffusionPipeline
pipeline = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5")
pipeline = StableDiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5")
pipeline.unet.config["in_channels"]
4
```
@@ -45,7 +45,7 @@ pipeline.unet.config["in_channels"]
```py
from diffusers import UNet2DConditionModel
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
unet = UNet2DConditionModel.from_pretrained(
model_id, subfolder="unet", in_channels=9, low_cpu_mem_usage=False, ignore_mismatched_sizes=True
)

View File

@@ -70,7 +70,7 @@ write_basic_config()
원본 데이터셋은 ControlNet [repo](https://huggingface.co/lllyasviel/ControlNet/blob/main/training/fill50k.zip)에 올라와있지만, 우리는 [여기](https://huggingface.co/datasets/fusing/fill50k)에 새롭게 다시 올려서 🤗 Datasets 과 호환가능합니다. 그래서 학습 스크립트 상에서 데이터 불러오기를 다룰 수 있습니다.
우리의 학습 예시는 원래 ControlNet의 학습에 쓰였던 [`runwayml/stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5)을 사용합니다. 그렇지만 ControlNet은 대응되는 어느 Stable Diffusion 모델([`CompVis/stable-diffusion-v1-4`](https://huggingface.co/CompVis/stable-diffusion-v1-4)) 혹은 [`stabilityai/stable-diffusion-2-1`](https://huggingface.co/stabilityai/stable-diffusion-2-1)의 증가를 위해 학습될 수 있습니다.
우리의 학습 예시는 원래 ControlNet의 학습에 쓰였던 [`stable-diffusion-v1-5/stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5)을 사용합니다. 그렇지만 ControlNet은 대응되는 어느 Stable Diffusion 모델([`CompVis/stable-diffusion-v1-4`](https://huggingface.co/CompVis/stable-diffusion-v1-4)) 혹은 [`stabilityai/stable-diffusion-2-1`](https://huggingface.co/stabilityai/stable-diffusion-2-1)의 증가를 위해 학습될 수 있습니다.
자체 데이터셋을 사용하기 위해서는 [학습을 위한 데이터셋 생성하기](create_dataset) 가이드를 확인하세요.
@@ -89,7 +89,7 @@ wget https://huggingface.co/datasets/huggingface/documentation-images/resolve/ma
학습 스크립트는 당신의 리포지토리에 `diffusion_pytorch_model.bin` 파일을 생성하고 저장합니다.
```bash
export MODEL_DIR="runwayml/stable-diffusion-v1-5"
export MODEL_DIR="stable-diffusion-v1-5/stable-diffusion-v1-5"
export OUTPUT_DIR="path to save model"
accelerate launch train_controlnet.py \
@@ -111,7 +111,7 @@ accelerate launch train_controlnet.py \
더 작은 batch(배치) 크기로 gradient accumulation(기울기 누적)을 하면 학습 요구사항을 ~20 GB VRAM으로 줄일 수 있습니다.
```bash
export MODEL_DIR="runwayml/stable-diffusion-v1-5"
export MODEL_DIR="stable-diffusion-v1-5/stable-diffusion-v1-5"
export OUTPUT_DIR="path to save model"
accelerate launch train_controlnet.py \
@@ -133,7 +133,7 @@ accelerate launch train_controlnet.py \
의 설명을 확인하세요. 아래는 예시 명령어입니다:
```bash
export MODEL_DIR="runwayml/stable-diffusion-v1-5"
export MODEL_DIR="stable-diffusion-v1-5/stable-diffusion-v1-5"
export OUTPUT_DIR="path to save model"
accelerate launch --mixed_precision="fp16" --multi_gpu train_controlnet.py \
@@ -181,7 +181,7 @@ accelerate launch --mixed_precision="fp16" --multi_gpu train_controlnet.py \
이제 학습 스크립트를 시작할 수 있습니다:
```bash
export MODEL_DIR="runwayml/stable-diffusion-v1-5"
export MODEL_DIR="stable-diffusion-v1-5/stable-diffusion-v1-5"
export OUTPUT_DIR="path to save model"
accelerate launch train_controlnet.py \
@@ -209,7 +209,7 @@ accelerate launch train_controlnet.py \
- 기울기를 `None`으로 설정
```bash
export MODEL_DIR="runwayml/stable-diffusion-v1-5"
export MODEL_DIR="stable-diffusion-v1-5/stable-diffusion-v1-5"
export OUTPUT_DIR="path to save model"
accelerate launch train_controlnet.py \
@@ -275,7 +275,7 @@ Pytorch와 같은 버전의 CUDA toolchain이 필요합니다. 8-비트 optimize
호환되지 않는 것 같습니다.
```bash
export MODEL_DIR="runwayml/stable-diffusion-v1-5"
export MODEL_DIR="stable-diffusion-v1-5/stable-diffusion-v1-5"
export OUTPUT_DIR="path to save model"
accelerate launch train_controlnet.py \

View File

@@ -86,7 +86,7 @@ dataset.push_to_hub("name_of_your_dataset", private=True)
```bash
accelerate launch --mixed_precision="fp16" train_text_to_image.py \
--pretrained_model_name_or_path="runwayml/stable-diffusion-v1-5" \
--pretrained_model_name_or_path="stable-diffusion-v1-5/stable-diffusion-v1-5" \
--dataset_name="name_of_your_dataset" \
<other-arguments>
```

View File

@@ -17,7 +17,7 @@
from accelerate import PartialState
from diffusers import DiffusionPipeline
pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16)
pipeline = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16)
distributed_state = PartialState()
pipeline.to(distributed_state.device)
@@ -51,7 +51,7 @@ import torch.multiprocessing as mp
from diffusers import DiffusionPipeline
sd = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16)
sd = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16)
```
사용할 백엔드 유형, 현재 프로세스의 `rank`, `world_size` 또는 참여하는 프로세스 수로 분산 환경 생성을 처리하는 함수[`init_process_group`]를 만들어 추론을 실행해야 합니다.

View File

@@ -84,7 +84,7 @@ write_basic_config()
```bash
export MODEL_NAME="runwayml/stable-diffusion-v1-5"
export MODEL_NAME="stable-diffusion-v1-5/stable-diffusion-v1-5"
export DATASET_ID="fusing/instructpix2pix-1000-samples"
```
@@ -141,7 +141,7 @@ accelerate launch --mixed_precision="fp16" train_instruct_pix2pix.py \
```bash
accelerate launch --mixed_precision="fp16" --multi_gpu train_instruct_pix2pix.py \
--pretrained_model_name_or_path=runwayml/stable-diffusion-v1-5 \
--pretrained_model_name_or_path=stable-diffusion-v1-5/stable-diffusion-v1-5 \
--dataset_name=sayakpaul/instructpix2pix-1000-samples \
--use_ema \
--enable_xformers_memory_efficient_attention \

View File

@@ -49,12 +49,12 @@ huggingface-cli login
### 학습[[dreambooth-training]]
[Naruto BLIP 캡션](https://huggingface.co/datasets/lambdalabs/naruto-blip-captions) 데이터셋으로 [`stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5)를 파인튜닝해 나만의 포켓몬을 생성해 보겠습니다.
[Naruto BLIP 캡션](https://huggingface.co/datasets/lambdalabs/naruto-blip-captions) 데이터셋으로 [`stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5)를 파인튜닝해 나만의 포켓몬을 생성해 보겠습니다.
시작하려면 `MODEL_NAME``DATASET_NAME` 환경 변수가 설정되어 있는지 확인하십시오. `OUTPUT_DIR``HUB_MODEL_ID` 변수는 선택 사항이며 허브에서 모델을 저장할 위치를 지정합니다.
```bash
export MODEL_NAME="runwayml/stable-diffusion-v1-5"
export MODEL_NAME="stable-diffusion-v1-5/stable-diffusion-v1-5"
export OUTPUT_DIR="/sddata/finetune/lora/naruto"
export HUB_MODEL_ID="naruto-lora"
export DATASET_NAME="lambdalabs/naruto-blip-captions"
@@ -97,7 +97,7 @@ accelerate launch train_dreambooth_lora.py \
>>> import torch
>>> from diffusers import StableDiffusionPipeline
>>> model_base = "runwayml/stable-diffusion-v1-5"
>>> model_base = "stable-diffusion-v1-5/stable-diffusion-v1-5"
>>> pipe = StableDiffusionPipeline.from_pretrained(model_base, torch_dtype=torch.float16)
```

View File

@@ -135,7 +135,7 @@ pip install -U -r requirements_flax.txt
그러면 다음과 같이 [Flax 학습 스크립트](https://github.com/huggingface/diffusers/blob/main/examples/text_to_image/train_text_to_image_flax.py)를 실행할 수 있습니다.
```bash
export MODEL_NAME="runwayml/stable-diffusion-v1-5"
export MODEL_NAME="stable-diffusion-v1-5/stable-diffusion-v1-5"
export dataset_name="lambdalabs/naruto-blip-captions"
python train_text_to_image_flax.py \

View File

@@ -21,7 +21,7 @@ specific language governing permissions and limitations under the License.
![Textual Inversion example](https://textual-inversion.github.io/static/images/editing/colorful_teapot.JPG)
<small>By using just 3-5 images you can teach new concepts to a model such as Stable Diffusion for personalized image generation <a href="https://github.com/rinongal/textual_inversion">(image source)</a>.</small>
이 가이드에서는 textual-inversion으로 [`runwayml/stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5) 모델을 학습하는 방법을 설명합니다. 이 가이드에서 사용된 모든 textual-inversion 학습 스크립트는 [여기](https://github.com/huggingface/diffusers/tree/main/examples/textual_inversion)에서 확인할 수 있습니다. 내부적으로 어떻게 작동하는지 자세히 살펴보고 싶으시다면 해당 링크를 참조해주시기 바랍니다.
이 가이드에서는 textual-inversion으로 [`stable-diffusion-v1-5/stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) 모델을 학습하는 방법을 설명합니다. 이 가이드에서 사용된 모든 textual-inversion 학습 스크립트는 [여기](https://github.com/huggingface/diffusers/tree/main/examples/textual_inversion)에서 확인할 수 있습니다. 내부적으로 어떻게 작동하는지 자세히 살펴보고 싶으시다면 해당 링크를 참조해주시기 바랍니다.
<Tip>
@@ -109,7 +109,7 @@ snapshot_download(
<frameworkcontent>
<pt>
```bash
export MODEL_NAME="runwayml/stable-diffusion-v1-5"
export MODEL_NAME="stable-diffusion-v1-5/stable-diffusion-v1-5"
export DATA_DIR="./cat"
accelerate launch textual_inversion.py \
@@ -199,13 +199,13 @@ textual-inversion 스크립트는 기본적으로 textual-inversion을 통해
</Tip>
textual-inversion 임베딩 벡터을 불러오기 위해서는, 먼저 해당 임베딩 벡터를 학습할 때 사용한 모델을 불러와야 합니다. 여기서는 [`runwayml/stable-diffusion-v1-5`](https://huggingface.co/docs/diffusers/training/runwayml/stable-diffusion-v1-5) 모델이 사용되었다고 가정하고 불러오겠습니다.
textual-inversion 임베딩 벡터을 불러오기 위해서는, 먼저 해당 임베딩 벡터를 학습할 때 사용한 모델을 불러와야 합니다. 여기서는 [`stable-diffusion-v1-5/stable-diffusion-v1-5`](https://huggingface.co/docs/diffusers/training/stable-diffusion-v1-5/stable-diffusion-v1-5) 모델이 사용되었다고 가정하고 불러오겠습니다.
```python
from diffusers import StableDiffusionPipeline
import torch
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
pipe = StableDiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16).to("cuda")
```

View File

@@ -46,7 +46,7 @@ feature_extractor = CLIPImageProcessor.from_pretrained(clip_model_id)
clip_model = CLIPModel.from_pretrained(clip_model_id)
pipeline = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
custom_pipeline="clip_guided_stable_diffusion",
clip_model=clip_model,
feature_extractor=feature_extractor,

View File

@@ -41,7 +41,7 @@ diffusion 모델의 훈련과 추론에 필요한 모든 것은 [`DiffusionPipel
```python
from diffusers import DiffusionPipeline
repo_id = "runwayml/stable-diffusion-v1-5"
repo_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
pipe = DiffusionPipeline.from_pretrained(repo_id)
```
@@ -50,16 +50,16 @@ pipe = DiffusionPipeline.from_pretrained(repo_id)
```python
from diffusers import StableDiffusionPipeline
repo_id = "runwayml/stable-diffusion-v1-5"
repo_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
pipe = StableDiffusionPipeline.from_pretrained(repo_id)
```
[CompVis/stable-diffusion-v1-4](https://huggingface.co/CompVis/stable-diffusion-v1-4)이나 [runwayml/stable-diffusion-v1-5](https://huggingface.co/runwayml/stable-diffusion-v1-5) 같은 체크포인트들의 경우, 하나 이상의 다양한 태스크에 활용될 수 있습니다. (예를 들어 위의 두 체크포인트의 경우, text-to-image와 image-to-image에 모두 활용될 수 있습니다.) 만약 이러한 체크포인트들을 기본 설정 태스크가 아닌 다른 태스크에 활용하고자 한다면, 해당 태스크에 대응되는 파이프라인(task-specific pipeline)을 사용해야 합니다.
[CompVis/stable-diffusion-v1-4](https://huggingface.co/CompVis/stable-diffusion-v1-4)이나 [stable-diffusion-v1-5/stable-diffusion-v1-5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) 같은 체크포인트들의 경우, 하나 이상의 다양한 태스크에 활용될 수 있습니다. (예를 들어 위의 두 체크포인트의 경우, text-to-image와 image-to-image에 모두 활용될 수 있습니다.) 만약 이러한 체크포인트들을 기본 설정 태스크가 아닌 다른 태스크에 활용하고자 한다면, 해당 태스크에 대응되는 파이프라인(task-specific pipeline)을 사용해야 합니다.
```python
from diffusers import StableDiffusionImg2ImgPipeline
repo_id = "runwayml/stable-diffusion-v1-5"
repo_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
pipe = StableDiffusionImg2ImgPipeline.from_pretrained(repo_id)
```
@@ -71,7 +71,7 @@ pipe = StableDiffusionImg2ImgPipeline.from_pretrained(repo_id)
```bash
git lfs install
git clone https://huggingface.co/runwayml/stable-diffusion-v1-5
git clone https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5
```
그런 다음 해당 로컬 경로를 [`~DiffusionPipeline.from_pretrained`] 메서드에 전달합니다.
@@ -100,19 +100,19 @@ stable_diffusion = DiffusionPipeline.from_pretrained(repo_id)
```python
from diffusers import DiffusionPipeline
repo_id = "runwayml/stable-diffusion-v1-5"
repo_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
stable_diffusion = DiffusionPipeline.from_pretrained(repo_id)
stable_diffusion.scheduler.compatibles
```
이번에는 [`SchedulerMixin.from_pretrained`] 메서드를 사용해서, 기존 기본 스케줄러였던 [`PNDMScheduler`]를 보다 우수한 성능의 [`EulerDiscreteScheduler`]로 바꿔봅시다. 스케줄러를 로드할 때는 `subfolder` 인자를 통해, 해당 파이프라인의 리포지토리에서 [스케줄러에 관한 하위폴더](https://huggingface.co/runwayml/stable-diffusion-v1-5/tree/main/scheduler)를 명시해주어야 합니다.
이번에는 [`SchedulerMixin.from_pretrained`] 메서드를 사용해서, 기존 기본 스케줄러였던 [`PNDMScheduler`]를 보다 우수한 성능의 [`EulerDiscreteScheduler`]로 바꿔봅시다. 스케줄러를 로드할 때는 `subfolder` 인자를 통해, 해당 파이프라인의 리포지토리에서 [스케줄러에 관한 하위폴더](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5/tree/main/scheduler)를 명시해주어야 합니다.
그 다음 새롭게 생성한 [`EulerDiscreteScheduler`] 인스턴스를 [`DiffusionPipeline`]의 `scheduler` 인자에 전달합니다.
```python
from diffusers import DiffusionPipeline, EulerDiscreteScheduler, DPMSolverMultistepScheduler
repo_id = "runwayml/stable-diffusion-v1-5"
repo_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
scheduler = EulerDiscreteScheduler.from_pretrained(repo_id, subfolder="scheduler")
@@ -126,7 +126,7 @@ stable_diffusion = DiffusionPipeline.from_pretrained(repo_id, scheduler=schedule
```python
from diffusers import DiffusionPipeline
repo_id = "runwayml/stable-diffusion-v1-5"
repo_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
stable_diffusion = DiffusionPipeline.from_pretrained(repo_id, safety_checker=None)
```
@@ -137,7 +137,7 @@ stable_diffusion = DiffusionPipeline.from_pretrained(repo_id, safety_checker=Non
```python
from diffusers import StableDiffusionPipeline, StableDiffusionImg2ImgPipeline
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
stable_diffusion_txt2img = StableDiffusionPipeline.from_pretrained(model_id)
components = stable_diffusion_txt2img.components
@@ -154,7 +154,7 @@ stable_diffusion_img2img = StableDiffusionImg2ImgPipeline(**components)
```python
from diffusers import StableDiffusionPipeline, StableDiffusionImg2ImgPipeline
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
stable_diffusion_txt2img = StableDiffusionPipeline.from_pretrained(model_id)
stable_diffusion_img2img = StableDiffusionImg2ImgPipeline(
vae=stable_diffusion_txt2img.vae,
@@ -197,10 +197,10 @@ from diffusers import DiffusionPipeline
# load fp16 variant
stable_diffusion = DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", variant="fp16", torch_dtype=torch.float16
"stable-diffusion-v1-5/stable-diffusion-v1-5", variant="fp16", torch_dtype=torch.float16
)
# load non_ema variant
stable_diffusion = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", variant="non_ema")
stable_diffusion = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", variant="non_ema")
```
다른 부동소수점 타입의 가중치 혹은 non-EMA 가중치를 사용하는 체크포인트를 저장하기 위해서는, [`DiffusionPipeline.save_pretrained`] 메서드를 사용해야 하며, 이 때 `variant` 인자를 명시해줘야 합니다. 원래의 체크포인트와 동일한 폴더에 variant를 저장해야 하며, 이렇게 하면 동일한 폴더에서 오리지널 체크포인트과 variant를 모두 불러올 수 있습니다.
@@ -209,9 +209,9 @@ stable_diffusion = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-
from diffusers import DiffusionPipeline
# save as fp16 variant
stable_diffusion.save_pretrained("runwayml/stable-diffusion-v1-5", variant="fp16")
stable_diffusion.save_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", variant="fp16")
# save as non-ema variant
stable_diffusion.save_pretrained("runwayml/stable-diffusion-v1-5", variant="non_ema")
stable_diffusion.save_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", variant="non_ema")
```
만약 variant를 기존 폴더에 저장하지 않을 경우, `variant` 인자를 반드시 명시해야 합니다. 그렇게 하지 않을 경우 원래의 오리지널 체크포인트를 찾을 수 없게 되기 때문에 에러가 발생합니다.
@@ -229,12 +229,12 @@ stable_diffusion = DiffusionPipeline.from_pretrained(
모델들은 [`ModelMixin.from_pretrained`] 메서드를 통해 불러올 수 있습니다. 해당 메서드는 최신 버전의 모델 가중치 파일과 설정 파일(configurations)을 다운로드하고 캐싱합니다. 만약 이러한 파일들이 최신 버전으로 로컬 캐시에 저장되어 있다면, [`ModelMixin.from_pretrained`]는 굳이 해당 파일들을 다시 다운로드하지 않으며, 그저 캐시에 있는 최신 파일들을 재사용합니다.
모델은 `subfolder` 인자에 명시된 하위 폴더로부터 로드됩니다. 예를 들어 `runwayml/stable-diffusion-v1-5`의 UNet 모델의 가중치는 [`unet`](https://huggingface.co/runwayml/stable-diffusion-v1-5/tree/main/unet) 폴더에 저장되어 있습니다.
모델은 `subfolder` 인자에 명시된 하위 폴더로부터 로드됩니다. 예를 들어 `stable-diffusion-v1-5/stable-diffusion-v1-5`의 UNet 모델의 가중치는 [`unet`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5/tree/main/unet) 폴더에 저장되어 있습니다.
```python
from diffusers import UNet2DConditionModel
repo_id = "runwayml/stable-diffusion-v1-5"
repo_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
model = UNet2DConditionModel.from_pretrained(repo_id, subfolder="unet")
```
@@ -252,7 +252,7 @@ model = UNet2DModel.from_pretrained(repo_id)
```python
from diffusers import UNet2DConditionModel
model = UNet2DConditionModel.from_pretrained("runwayml/stable-diffusion-v1-5", subfolder="unet", variant="non-ema")
model = UNet2DConditionModel.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", subfolder="unet", variant="non-ema")
model.save_pretrained("./local-unet", variant="non-ema")
```
@@ -274,7 +274,7 @@ from diffusers import (
DPMSolverMultistepScheduler,
)
repo_id = "runwayml/stable-diffusion-v1-5"
repo_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
ddpm = DDPMScheduler.from_pretrained(repo_id, subfolder="scheduler")
ddim = DDIMScheduler.from_pretrained(repo_id, subfolder="scheduler")
@@ -295,12 +295,12 @@ pipeline = StableDiffusionPipeline.from_pretrained(repo_id, scheduler=dpm)
- 첫째로, `from_pretrained` 메서드는 최신 버전의 파이프라인을 다운로드하고, 캐시에 저장합니다. 이미 로컬 캐시에 최신 버전의 파이프라인이 저장되어 있다면, [`DiffusionPipeline.from_pretrained`]은 해당 파일들을 다시 다운로드하지 않고, 로컬 캐시에 저장되어 있는 파이프라인을 불러옵니다.
- `model_index.json` 파일을 통해 체크포인트에 대응되는 적합한 파이프라인 클래스로 불러옵니다.
파이프라인의 폴더 구조는 해당 파이프라인 클래스의 구조와 직접적으로 일치합니다. 예를 들어 [`StableDiffusionPipeline`] 클래스는 [`runwayml/stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5) 리포지토리와 대응되는 구조를 갖습니다.
파이프라인의 폴더 구조는 해당 파이프라인 클래스의 구조와 직접적으로 일치합니다. 예를 들어 [`StableDiffusionPipeline`] 클래스는 [`stable-diffusion-v1-5/stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5) 리포지토리와 대응되는 구조를 갖습니다.
```python
from diffusers import DiffusionPipeline
repo_id = "runwayml/stable-diffusion-v1-5"
repo_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
pipeline = DiffusionPipeline.from_pretrained(repo_id)
print(pipeline)
```
@@ -348,7 +348,7 @@ StableDiffusionPipeline {
}
```
파이프라인 인스턴스의 컴포넌트들을 [`runwayml/stable-diffusion-v1-5`](https://huggingface.co/runwayml/stable-diffusion-v1-5)의 폴더 구조와 비교해볼 경우, 각각의 컴포넌트마다 별도의 폴더가 있음을 확인할 수 있습니다.
파이프라인 인스턴스의 컴포넌트들을 [`stable-diffusion-v1-5/stable-diffusion-v1-5`](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5)의 폴더 구조와 비교해볼 경우, 각각의 컴포넌트마다 별도의 폴더가 있음을 확인할 수 있습니다.
```
.

View File

@@ -54,7 +54,7 @@ Textual inversion은 임베딩을 생성하기 때문에 DreamBooth처럼 단독
from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
pipeline = AutoPipelineForText2Image.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
```
이제 [`~loaders.TextualInversionLoaderMixin.load_textual_inversion`] 메서드를 사용하여 textual inversion 임베딩을 불러와 이미지를 생성할 수 있습니다. [sd-concepts-library/gta5-artwork](https://huggingface.co/sd-concepts-library/gta5-artwork) 임베딩을 불러와 보겠습니다. 이를 트리거하려면 프롬프트에 특수 단어 `<gta5-artwork>`를 포함시켜야 합니다:
@@ -269,7 +269,7 @@ from diffusers import AutoPipelineForText2Image
import torch
from diffusers.utils import load_image
pipeline = AutoPipelineForText2Image.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
pipeline = AutoPipelineForText2Image.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
```
그런 다음 IP-Adapter 가중치를 불러와 [`~loaders.IPAdapterMixin.load_ip_adapter`] 메서드를 사용하여 파이프라인에 추가합니다.
@@ -350,7 +350,7 @@ image_encoder = CLIPVisionModelWithProjection.from_pretrained(
)
pipeline = AutoPipelineForText2Image.from_pretrained(
"runwayml/stable-diffusion-v1-5",
"stable-diffusion-v1-5/stable-diffusion-v1-5",
image_encoder=image_encoder,
torch_dtype=torch.float16
).to("cuda")

View File

@@ -30,7 +30,7 @@ diffusion 파이프라인은 diffusion 모델, 스케줄러 등의 컴포넌트
## 파이프라인 불러오기
먼저 스테이블 diffusion 파이프라인을 불러오도록 해보겠습니다. 물론 스테이블 diffusion을 사용하기 위해서는, 허깅페이스 허브에 등록된 사용자여야 하며, 관련 [라이센스](https://huggingface.co/runwayml/stable-diffusion-v1-5)에 동의해야 한다는 점을 잊지 말아주세요.
먼저 스테이블 diffusion 파이프라인을 불러오도록 해보겠습니다. 물론 스테이블 diffusion을 사용하기 위해서는, 허깅페이스 허브에 등록된 사용자여야 하며, 관련 [라이센스](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5)에 동의해야 한다는 점을 잊지 말아주세요.
*역자 주: 다만, 현재 신규로 생성한 허깅페이스 계정에 대해서는 라이센스 동의를 요구하지 않는 것으로 보입니다!*
@@ -43,7 +43,7 @@ import torch
login()
# Now we can download the pipeline
pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16)
pipeline = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16)
```
다음으로, GPU로 이동합니다.
@@ -288,7 +288,7 @@ from flax.training.common_utils import shard
from diffusers import FlaxStableDiffusionPipeline, FlaxDPMSolverMultistepScheduler
model_id = "runwayml/stable-diffusion-v1-5"
model_id = "stable-diffusion-v1-5/stable-diffusion-v1-5"
scheduler, scheduler_state = FlaxDPMSolverMultistepScheduler.from_pretrained(
model_id,
subfolder="scheduler"

View File

@@ -42,7 +42,7 @@ def image_grid(imgs, rows, cols):
Stable Diffusion과 [Stable Diffusion Conceptualizer](https://huggingface.co/spaces/sd-concepts-library/stable-diffusion-conceptualizer)에서 사전학습된 컨셉을 선택합니다:
```py
pretrained_model_name_or_path = "runwayml/stable-diffusion-v1-5"
pretrained_model_name_or_path = "stable-diffusion-v1-5/stable-diffusion-v1-5"
repo_id_embeds = "sd-concepts-library/cat-toy"
```

Some files were not shown because too many files have changed in this diff Show More