mirror of
https://github.com/huggingface/diffusers.git
synced 2026-01-27 17:22:53 +03:00
fix link in the docs (#10058)
* fix link in the docs * fix same issue for ko
This commit is contained in:
@@ -1,6 +1,6 @@
|
||||
# Create a dataset for training
|
||||
|
||||
There are many datasets on the [Hub](https://huggingface.co/datasets?task_categories=task_categories:text-to-image&sort=downloads) to train a model on, but if you can't find one you're interested in or want to use your own, you can create a dataset with the ๐ค [Datasets](hf.co/docs/datasets) library. The dataset structure depends on the task you want to train your model on. The most basic dataset structure is a directory of images for tasks like unconditional image generation. Another dataset structure may be a directory of images and a text file containing their corresponding text captions for tasks like text-to-image generation.
|
||||
There are many datasets on the [Hub](https://huggingface.co/datasets?task_categories=task_categories:text-to-image&sort=downloads) to train a model on, but if you can't find one you're interested in or want to use your own, you can create a dataset with the ๐ค [Datasets](https://huggingface.co/docs/datasets) library. The dataset structure depends on the task you want to train your model on. The most basic dataset structure is a directory of images for tasks like unconditional image generation. Another dataset structure may be a directory of images and a text file containing their corresponding text captions for tasks like text-to-image generation.
|
||||
|
||||
This guide will show you two ways to create a dataset to finetune on:
|
||||
|
||||
@@ -87,4 +87,4 @@ accelerate launch --mixed_precision="fp16" train_text_to_image.py \
|
||||
|
||||
Now that you've created a dataset, you can plug it into the `train_data_dir` (if your dataset is local) or `dataset_name` (if your dataset is on the Hub) arguments of a training script.
|
||||
|
||||
For your next steps, feel free to try and use your dataset to train a model for [unconditional generation](unconditional_training) or [text-to-image generation](text2image)!
|
||||
For your next steps, feel free to try and use your dataset to train a model for [unconditional generation](unconditional_training) or [text-to-image generation](text2image)!
|
||||
|
||||
@@ -121,7 +121,7 @@ image = pipe(prompt=prompt, image=init_image, mask_image=mask_image, num_inferen
|
||||
|
||||
### ์ด๋ฏธ์ง ๊ฒฐ๊ณผ๋ฌผ์ ์ ์ ํ๊ธฐ
|
||||
|
||||
[base ๋ชจ๋ธ ์ฒดํฌํฌ์ธํธ](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0)์์, StableDiffusion-XL ๋ํ ๊ณ ์ฃผํ ํ์ง์ ํฅ์์ํค๋ ์ด๋ฏธ์ง๋ฅผ ์์ฑํ๊ธฐ ์ํด ๋ฎ์ ๋
ธ์ด์ฆ ๋จ๊ณ ์ด๋ฏธ์ง๋ฅผ ์ ๊ฑฐํ๋๋ฐ ํนํ๋ [refiner ์ฒดํฌํฌ์ธํธ](huggingface.co/stabilityai/stable-diffusion-xl-refiner-1.0)๋ฅผ ํฌํจํ๊ณ ์์ต๋๋ค. ์ด refiner ์ฒดํฌํฌ์ธํธ๋ ์ด๋ฏธ์ง ํ์ง์ ํฅ์์ํค๊ธฐ ์ํด base ์ฒดํฌํฌ์ธํธ๋ฅผ ์คํํ ํ "๋ ๋ฒ์งธ ๋จ๊ณ" ํ์ดํ๋ผ์ธ์ ์ฌ์ฉ๋ ์ ์์ต๋๋ค.
|
||||
[base ๋ชจ๋ธ ์ฒดํฌํฌ์ธํธ](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0)์์, StableDiffusion-XL ๋ํ ๊ณ ์ฃผํ ํ์ง์ ํฅ์์ํค๋ ์ด๋ฏธ์ง๋ฅผ ์์ฑํ๊ธฐ ์ํด ๋ฎ์ ๋
ธ์ด์ฆ ๋จ๊ณ ์ด๋ฏธ์ง๋ฅผ ์ ๊ฑฐํ๋๋ฐ ํนํ๋ [refiner ์ฒดํฌํฌ์ธํธ](https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-1.0)๋ฅผ ํฌํจํ๊ณ ์์ต๋๋ค. ์ด refiner ์ฒดํฌํฌ์ธํธ๋ ์ด๋ฏธ์ง ํ์ง์ ํฅ์์ํค๊ธฐ ์ํด base ์ฒดํฌํฌ์ธํธ๋ฅผ ์คํํ ํ "๋ ๋ฒ์งธ ๋จ๊ณ" ํ์ดํ๋ผ์ธ์ ์ฌ์ฉ๋ ์ ์์ต๋๋ค.
|
||||
|
||||
refiner๋ฅผ ์ฌ์ฉํ ๋, ์ฝ๊ฒ ์ฌ์ฉํ ์ ์์ต๋๋ค
|
||||
- 1.) base ๋ชจ๋ธ๊ณผ refiner์ ์ฌ์ฉํ๋๋ฐ, ์ด๋ *Denoisers์ ์์๋ธ*์ ์ํ ์ฒซ ๋ฒ์งธ ์ ์๋ [eDiff-I](https://research.nvidia.com/labs/dir/eDiff-I/)๋ฅผ ์ฌ์ฉํ๊ฑฐ๋
|
||||
@@ -215,7 +215,7 @@ image = refiner(
|
||||
|
||||
#### 2.) ๋
ธ์ด์ฆ๊ฐ ์์ ํ ์ ๊ฑฐ๋ ๊ธฐ๋ณธ ์ด๋ฏธ์ง์์ ์ด๋ฏธ์ง ์ถ๋ ฅ์ ์ ์ ํ๊ธฐ
|
||||
|
||||
์ผ๋ฐ์ ์ธ [`StableDiffusionImg2ImgPipeline`] ๋ฐฉ์์์, ๊ธฐ๋ณธ ๋ชจ๋ธ์์ ์์ฑ๋ ์์ ํ ๋
ธ์ด์ฆ๊ฐ ์ ๊ฑฐ๋ ์ด๋ฏธ์ง๋ [refiner checkpoint](huggingface.co/stabilityai/stable-diffusion-xl-refiner-1.0)๋ฅผ ์ฌ์ฉํด ๋ ํฅ์์ํฌ ์ ์์ต๋๋ค.
|
||||
์ผ๋ฐ์ ์ธ [`StableDiffusionImg2ImgPipeline`] ๋ฐฉ์์์, ๊ธฐ๋ณธ ๋ชจ๋ธ์์ ์์ฑ๋ ์์ ํ ๋
ธ์ด์ฆ๊ฐ ์ ๊ฑฐ๋ ์ด๋ฏธ์ง๋ [refiner checkpoint](https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-1.0)๋ฅผ ์ฌ์ฉํด ๋ ํฅ์์ํฌ ์ ์์ต๋๋ค.
|
||||
|
||||
์ด๋ฅผ ์ํด, ๋ณดํต์ "base" text-to-image ํ์ดํ๋ผ์ธ์ ์ํ ํ์ image-to-image ํ์ดํ๋ผ์ธ์ผ๋ก์จ refiner๋ฅผ ์คํ์ํฌ ์ ์์ต๋๋ค. base ๋ชจ๋ธ์ ์ถ๋ ฅ์ ์ ์ฌ ๊ณต๊ฐ์ ๋จ๊ฒจ๋ ์ ์์ต๋๋ค.
|
||||
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
# ํ์ต์ ์ํ ๋ฐ์ดํฐ์
๋ง๋ค๊ธฐ
|
||||
|
||||
[Hub](https://huggingface.co/datasets?task_categories=task_categories:text-to-image&sort=downloads) ์๋ ๋ชจ๋ธ ๊ต์ก์ ์ํ ๋ง์ ๋ฐ์ดํฐ์
์ด ์์ง๋ง,
|
||||
๊ด์ฌ์ด ์๊ฑฐ๋ ์ฌ์ฉํ๊ณ ์ถ์ ๋ฐ์ดํฐ์
์ ์ฐพ์ ์ ์๋ ๊ฒฝ์ฐ ๐ค [Datasets](hf.co/docs/datasets) ๋ผ์ด๋ธ๋ฌ๋ฆฌ๋ฅผ ์ฌ์ฉํ์ฌ ๋ฐ์ดํฐ์
์ ๋ง๋ค ์ ์์ต๋๋ค.
|
||||
๊ด์ฌ์ด ์๊ฑฐ๋ ์ฌ์ฉํ๊ณ ์ถ์ ๋ฐ์ดํฐ์
์ ์ฐพ์ ์ ์๋ ๊ฒฝ์ฐ ๐ค [Datasets](https://huggingface.co/docs/datasets) ๋ผ์ด๋ธ๋ฌ๋ฆฌ๋ฅผ ์ฌ์ฉํ์ฌ ๋ฐ์ดํฐ์
์ ๋ง๋ค ์ ์์ต๋๋ค.
|
||||
๋ฐ์ดํฐ์
๊ตฌ์กฐ๋ ๋ชจ๋ธ์ ํ์ตํ๋ ค๋ ์์
์ ๋ฐ๋ผ ๋ฌ๋ผ์ง๋๋ค.
|
||||
๊ฐ์ฅ ๊ธฐ๋ณธ์ ์ธ ๋ฐ์ดํฐ์
๊ตฌ์กฐ๋ unconditional ์ด๋ฏธ์ง ์์ฑ๊ณผ ๊ฐ์ ์์
์ ์ํ ์ด๋ฏธ์ง ๋๋ ํ ๋ฆฌ์
๋๋ค.
|
||||
๋ ๋ค๋ฅธ ๋ฐ์ดํฐ์
๊ตฌ์กฐ๋ ์ด๋ฏธ์ง ๋๋ ํ ๋ฆฌ์ text-to-image ์์ฑ๊ณผ ๊ฐ์ ์์
์ ํด๋นํ๋ ํ
์คํธ ์บก์
์ด ํฌํจ๋ ํ
์คํธ ํ์ผ์ผ ์ ์์ต๋๋ค.
|
||||
|
||||
@@ -36,7 +36,7 @@ specific language governing permissions and limitations under the License.
|
||||
|
||||
[cloneofsimo](https://github.com/cloneofsimo)๋ ์ธ๊ธฐ ์๋ [lora](https://github.com/cloneofsimo/lora) GitHub ๋ฆฌํฌ์งํ ๋ฆฌ์์ Stable Diffusion์ ์ํ LoRA ํ์ต์ ์ต์ด๋ก ์๋ํ์ต๋๋ค. ๐งจ Diffusers๋ [text-to-image ์์ฑ](https://github.com/huggingface/diffusers/tree/main/examples/text_to_image#training-with-lora) ๋ฐ [DreamBooth](https://github.com/huggingface/diffusers/tree/main/examples/dreambooth#training-with-low-rank-adaptation-of-large-language-models-lora)์ ์ง์ํฉ๋๋ค. ์ด ๊ฐ์ด๋๋ ๋ ๊ฐ์ง๋ฅผ ๋ชจ๋ ์ํํ๋ ๋ฐฉ๋ฒ์ ๋ณด์ฌ์ค๋๋ค.
|
||||
|
||||
๋ชจ๋ธ์ ์ ์ฅํ๊ฑฐ๋ ์ปค๋ฎค๋ํฐ์ ๊ณต์ ํ๋ ค๋ฉด Hugging Face ๊ณ์ ์ ๋ก๊ทธ์ธํ์ธ์(์์ง ๊ณ์ ์ด ์๋ ๊ฒฝ์ฐ [์์ฑ](hf.co/join)ํ์ธ์):
|
||||
๋ชจ๋ธ์ ์ ์ฅํ๊ฑฐ๋ ์ปค๋ฎค๋ํฐ์ ๊ณต์ ํ๋ ค๋ฉด Hugging Face ๊ณ์ ์ ๋ก๊ทธ์ธํ์ธ์(์์ง ๊ณ์ ์ด ์๋ ๊ฒฝ์ฐ [์์ฑ](https://huggingface.co/join)ํ์ธ์):
|
||||
|
||||
```bash
|
||||
huggingface-cli login
|
||||
|
||||
Reference in New Issue
Block a user