Fix: solve some environment and demo issues

README.md

@@ -25,7 +25,6 @@ Official implementation of **[PhotoMaker: Customizing Realistic Human Photos via
 2. Ensures impressive ID fidelity, offering diversity, promising text controllability, and high-quality generation.
 3. Can serve as an **Adapter** to collaborate with other Base Models alongside LoRA modules in community.
 
-
 ---
 
 <div align="center">
@@ -75,7 +74,19 @@ Note: only change the base model and add the LoRA modules for better stylization
 - Python >= 3.8 (Recommend to use [Anaconda](https://www.anaconda.com/download/#linux) or [Miniconda](https://docs.conda.io/en/latest/miniconda.html))
 - [PyTorch >= 2.0.0](https://pytorch.org/)
 ```bash
+conda create --name photomaker python=3.10
+pip install -U pip
+
+# Install requirements
 pip install -r requirements.txt
+
+# Install photomaker
+pip install git+https://github.com/TencentARC/PhotoMaker.git
+```
+
+Then you can run the following command to use it
+```python
+from photomaker import PhotoMakerStableDiffusionXLPipeline
 ```
 
 # ⏬ Download Models 
@@ -98,7 +109,7 @@ import torch
 import os
 from diffusers.utils import load_image
 from diffusers import EulerDiscreteScheduler
-from photomaker.pipeline import PhotoMakerStableDiffusionXLPipeline
+from photomaker import PhotoMakerStableDiffusionXLPipeline
 
 ### Load base model
 pipe = PhotoMakerStableDiffusionXLPipeline.from_pretrained(
@@ -181,7 +192,7 @@ You could customize this script in [this file](gradio_demo/app.py).
 - For faster speed, reduce the number of generated images and sampling steps. However, please note that reducing the sampling steps may compromise the ID fidelity.
 
 # 🤗 Acknowledgements
-- T2I-Adapter is co-hosted by Tencent ARC Lab and Nankai University [MCG-NKU](https://mmcheng.net/cmm/).
+- PhotoMaker is co-hosted by Tencent ARC Lab and Nankai University [MCG-NKU](https://mmcheng.net/cmm/).
 - Inspired from many excellent demos and repos, including [IP-Adapter](https://github.com/tencent-ailab/IP-Adapter), [multimodalart/Ip-Adapter-FaceID](https://huggingface.co/spaces/multimodalart/Ip-Adapter-FaceID), [FastComposer](https://github.com/mit-han-lab/fastcomposer), and [T2I-Adapter](https://github.com/TencentARC/T2I-Adapter). Thanks for their great works!
 - Thanks for Venus team in Tencent PCG for their feedback and suggestions.
 

gradio_demo/app.py

@@ -10,7 +10,7 @@
 import spaces
 import gradio as gr
 
-from photomaker.pipeline import PhotoMakerStableDiffusionXLPipeline
+from photomaker import PhotoMakerStableDiffusionXLPipeline
 from style_template import styles
 
 # global variable
@@ -28,14 +28,16 @@
     torch_dtype=torch.bfloat16, 
     use_safetensors=True, 
     variant="fp16",
+    # local_files_only=True,
 ).to(device)
 
 pipe.load_photomaker_adapter(
     os.path.dirname(photomaker_ckpt),
     subfolder="",
     weight_name=os.path.basename(photomaker_ckpt),
     trigger_word="img"
-)     
+)
+pipe.id_encoder.to(device)
 
 pipe.scheduler = EulerDiscreteScheduler.from_config(pipe.scheduler.config)
 # pipe.set_adapters(["photomaker"], adapter_weights=[1.0])
@@ -282,4 +284,4 @@ def get_example():
 
     gr.Markdown(article)
 
-demo.launch()
+demo.launch(share=False)

photomaker/__init__.py

@@ -0,0 +1,7 @@
+from .model import PhotoMakerIDEncoder
+from .pipeline import PhotoMakerStableDiffusionXLPipeline
+
+__all__ = [
+    "PhotoMakerIDEncoder",
+    "PhotoMakerStableDiffusionXLPipeline",
+]

photomaker/pipeline.py

@@ -18,7 +18,7 @@
     logging,
 )
 
-from photomaker.model import PhotoMakerIDEncoder
+from . import PhotoMakerIDEncoder
 
 PipelineImageInput = Union[
     PIL.Image.Image,

photomaker_demo.ipynb

@@ -26,7 +26,7 @@
     "from diffusers import EulerDiscreteScheduler\n",
     "from huggingface_hub import hf_hub_download\n",
     "\n",
-    "from photomaker.pipeline import PhotoMakerStableDiffusionXLPipeline"
+    "from photomaker import PhotoMakerStableDiffusionXLPipeline"
    ]
   },
   {
@@ -256,4 +256,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 4
-}
+}

photomaker_style_demo.ipynb

@@ -26,7 +26,7 @@
     "from diffusers import DDIMScheduler\n",
     "from huggingface_hub import hf_hub_download\n",
     "\n",
-    "from photomaker.pipeline import PhotoMakerStableDiffusionXLPipeline"
+    "from photomaker import PhotoMakerStableDiffusionXLPipeline"
    ]
   },
   {

pyproject.toml

@@ -0,0 +1,43 @@
+[tool.poetry]
+name = "photomaker"
+version = "0.1.0"
+description = "PhotoMaker: Customizing Realistic Human Photos via Stacked ID Embedding"
+authors = ["Li, Zhen", "Cao, Mingdeng", "Wang, Xintao", "Qi, Zhongang", "Cheng, Ming-Ming", "Shan, Ying"]
+license = "Apache-2.0"
+readme = "README.md"
+packages = [{ include = "photomaker" }]
+
+[tool.poetry.dependencies]
+python = ">=3.7"
+
+[tool.ruff]
+line-length = 119
+# Deprecation of Cuda 11.6, Python 3.7 support for PyTorch 2.0
+target-version = "py38"
+
+# A list of file patterns to omit from linting, in addition to those specified by exclude.
+extend-exclude = ["__pycache__", "*.pyc", "*.egg-info", ".cache"]
+
+select = ["E", "F", "W", "C90", "I", "UP", "B", "C4", "RET", "RUF", "SIM"]
+
+
+ignore = [
+    "UP006",    # UP006: Use list instead of typing.List for type annotations
+    "UP007",    # UP007: Use X | Y for type annotations
+    "UP009",
+    "UP035",
+    "UP038",
+    "E402",
+    "RET504",
+]
+
+[tool.isort]
+profile = "black"
+
+[tool.black]
+line-length = 119
+skip-string-normalization = 1
+
+[build-system]
+requires = ["poetry-core"]
+build-backend = "poetry.core.masonry.api"

requirements.txt

@@ -1,6 +1,5 @@
 torch==2.0.1
 torchvision==0.15.2
-pytorch-cuda==11.8
 diffusers==0.25.0
 transformers==4.36.2
 huggingface-hub==0.20.2