Buckets:
| # κ°μ | |
| π€ Diffusersλ μ체 diffusion λͺ¨λΈμ νμ΅ν μ μλλ‘ λ€μν νμ΅ μ€ν¬λ¦½νΈλ₯Ό μ 곡ν©λλ€. λͺ¨λ νμ΅ μ€ν¬λ¦½νΈλ [diffusers/examples](https://github.com/huggingface/diffusers/tree/main/examples)μμ μ°Ύμ μ μμ΅λλ€. | |
| κ° νμ΅ μ€ν¬λ¦½νΈλ λ€μκ³Ό κ°μ΅λλ€: | |
| - **μκΈ° μκ²°ν**: νμ΅ μ€ν¬λ¦½νΈλ λ‘컬 νμΌμ μμ‘΄νμ§ μμΌλ©°, μ€ν¬λ¦½νΈ μ€νμ νμν λͺ¨λ ν¨ν€μ§λ `requirements.txt` νμΌμμ μ€μΉλ©λλ€. | |
| - **μμ½κ² μ‘°μ κ°λ₯**: νμ΅ μ€ν¬λ¦½νΈλ νΉμ μμ μ μν΄ diffusion λͺ¨λΈμ νμ΅νλ λ°©λ²μ 보μ¬μ£Όλ μμμ΄λ©°, λͺ¨λ νμ΅ μλ리μ€μμ λ°λ‘ μ¬μ©ν μ μμ΅λλ€. μ¬μ©μμ νΉμ μ¬μ© μ¬λ‘μ λ§κ² νμ΅ μ€ν¬λ¦½νΈλ₯Ό μμ ν΄μΌ ν κ°λ₯μ±μ΄ λμ΅λλ€. μ΄λ₯Ό λκΈ° μν΄ λ°μ΄ν° μ μ²λ¦¬ μ½λμ νμ΅ λ£¨νλ₯Ό μμ ν 곡κ°νμ¬ μ¬μ©μκ° μ§μ μμ ν μ μλλ‘ νμ΅λλ€. | |
| - **μ λ¬Έμ μΉνμ **: νμ΅ μ€ν¬λ¦½νΈλ μ΅μ μ΅μ²¨λ¨ κΈ°λ²μ ν¬ν¨νμ¬ κ°μ₯ κ²½μλ ₯ μλ κ²°κ³Όλ₯Ό λ΄λ κ²λ³΄λ€λ, μ λ¬Έμκ° μ΄ν΄νκΈ° μ½κ³ μ κ·ΌνκΈ° μ½λλ‘ μ€κ³λμμ΅λλ€. λ무 볡μ‘νλ€κ³ νλ¨λλ νμ΅ κΈ°λ²μ μλμ μΌλ‘ μ μΈλ©λλ€. | |
| - **λ¨μΌ λͺ©μ **: κ° νμ΅ μ€ν¬λ¦½νΈλ κ°λ μ±κ³Ό μ΄ν΄λλ₯Ό λμ΄κΈ° μν΄ μ€μ§ νλμ μμ λ§μ μννλλ‘ λͺ μμ μΌλ‘ μ€κ³λμ΄ μμ΅λλ€. | |
| νμ¬ μ 곡λλ νμ΅ μ€ν¬λ¦½νΈ λͺ©λ‘μ λ€μκ³Ό κ°μ΅λλ€: | |
| | νμ΅ | SDXL μ§μ | LoRA μ§μ | | |
| |---|---|---| | |
| | [무쑰건μ μ΄λ―Έμ§ μμ±](https://github.com/huggingface/diffusers/tree/main/examples/unconditional_image_generation) [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/training_example.ipynb) | | | | |
| | [ν μ€νΈ-μ΄λ―Έμ§](https://github.com/huggingface/diffusers/tree/main/examples/text_to_image) | π | π | | |
| | [ν μ€μΆμΌ μΈλ²μ ](https://github.com/huggingface/diffusers/tree/main/examples/textual_inversion) [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/sd_textual_inversion_training.ipynb) | | | | |
| | [DreamBooth](https://github.com/huggingface/diffusers/tree/main/examples/dreambooth) [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/sd_dreambooth_training.ipynb) | π | π | | |
| | [ControlNet](https://github.com/huggingface/diffusers/tree/main/examples/controlnet) | π | | | |
| | [InstructPix2Pix](https://github.com/huggingface/diffusers/tree/main/examples/instruct_pix2pix) | π | | | |
| | [컀μ€ν λν¨μ ](https://github.com/huggingface/diffusers/tree/main/examples/custom_diffusion) | | | | |
| | [T2I-Adapters](https://github.com/huggingface/diffusers/tree/main/examples/t2i_adapter) | π | | | |
| | [Kandinsky 2.2](https://github.com/huggingface/diffusers/tree/main/examples/kandinsky2_2/text_to_image) | | π | | |
| μ΄ μμ λ€μ **μ κ·Ήμ μΌλ‘** μ μ§ κ΄λ¦¬λκ³ μμΌλ―λ‘, μμλλ‘ μλνμ§ μλ κ²½μ° μ΄μλ₯Ό μ΄μ΄ μ£ΌμκΈ° λ°λλλ€. λ€λ₯Έ νμ΅ μμ κ° ν¬ν¨λμ΄μΌ νλ€κ³ μκ°νμ λ€λ©΄, [Feature Request](https://github.com/huggingface/diffusers/issues/new?assignees=&labels=&template=feature_request.md&title=)λ₯Ό μμνμ¬ κΈ°λ₯ μμ΄λμ΄λ₯Ό λ Όμνκ³ , λ 립μ μ΄λ©°, μ½κ² μμ ν μ μμΌλ©°, μ΄λ³΄μμκ² μΉνμ μ΄λ©°, λ¨μΌ λͺ©μ μ΄λΌλ μ°λ¦¬μ κΈ°μ€μ μΆ©μ‘±νλμ§ νμΈν΄ 보μκΈ° λ°λλλ€. | |
| ## μ€μΉ | |
| μ κ°μ νκ²½μμ μμ€μμ λΌμ΄λΈλ¬λ¦¬λ₯Ό μ€μΉνμ¬ μ΅μ λ²μ μ μμ μ€ν¬λ¦½νΈλ₯Ό μ±κ³΅μ μΌλ‘ μ€νν μ μλμ§ νμΈνμΈμ: | |
| ```bash | |
| git clone https://github.com/huggingface/diffusers | |
| cd diffusers | |
| pip install . | |
| ``` | |
| μ΄ν νμ΅ μ€ν¬λ¦½νΈ ν΄λ(μ: [DreamBooth](https://github.com/huggingface/diffusers/tree/main/examples/dreambooth))λ‘ μ΄λνμ¬ `requirements.txt` νμΌμ μ€μΉν©λλ€. μΌλΆ νμ΅ μ€ν¬λ¦½νΈλ SDXL λλ LoRAλ₯Ό μν μ μ© μꡬ μ¬ν νμΌμ΄ μμ΅λλ€. μ΄λ¬ν μ€ν¬λ¦½νΈ μ€ νλλ₯Ό μ¬μ©νλ κ²½μ°, ν΄λΉ μ€ν¬λ¦½νΈμ λμνλ μꡬ μ¬ν νμΌμ μ€μΉνλμ§ νμΈνμΈμ. | |
| ```bash | |
| cd examples/dreambooth | |
| pip install -r requirements.txt | |
| # to train SDXL with DreamBooth | |
| pip install -r requirements_sdxl.txt | |
| ``` | |
| νμ΅ μλλ₯Ό λμ΄κ³ λ©λͺ¨λ¦¬ μ¬μ©λμ μ€μ΄κΈ° μν΄ λ€μμ κΆμ₯ν©λλ€: | |
| - PyTorch 2.0 μ΄μμ μ¬μ©νμ¬ νμ΅ μ€μ [scaled dot product attention](../optimization/fp16#scaled-dot-product-attention)μ μλμΌλ‘ μ¬μ©νλλ‘ μ€μ ν©λλ€(νμ΅ μ½λλ₯Ό μμ ν νμκ° μμ΅λλ€). | |
| - λ©λͺ¨λ¦¬ ν¨μ¨μ μΈ μ΄ν μ μ νμ±ννκΈ° μν΄ [xFormers](../optimization/attention_backends) μ€μΉ | |
Xet Storage Details
- Size:
- 5.05 kB
- Xet hash:
- 88b7359af16b0a2b4ad9ee7f66ae903df752e2ee53ba63cd285cfa6be40fbe47
Β·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.