Web Demo Support running on Mac M1/2 (#17)

* support MPS

* Update README.md

* Make requirements more clearly

* resolve conflict

* Update: support MacOS

---------
Co-authored-by: Zhen Li <zhenli1031@gmail.com>

.gitignore

@@ -3,5 +3,6 @@ civitai_models/
 gradio_cached_examples/
 outputs/
 __pycache__/
+.idea/
 *.bin
 *.safetensors
\ No newline at end of file

MacGPUEnv.md

+# How to use GPU on Mac M1/2
+
+1. Install xcode tools
+
+```
+xcode-select --install
+```
+
+2. Install llvm
+
+```
+brew install llvm libomp
+```
+
+3. Install torch 2.1.2
+
+```
+pip install torch==2.1.2
+```
\ No newline at end of file

README.md

@@ -189,6 +189,8 @@ python gradio_demo/app.py
 
 You could customize this script in [this file](gradio_demo/app.py).
 
+If you want to run it on MAC, you should follow [this Instruction](MacGPUEnv.md) and then run the app.py.
+
 ## Usage Tips:
 - Upload more photos of the person to be customized to improve ID fidelty. If the input is Asian face(s), maybe consider adding 'asian' before the class word, e.g., `asian woman img`
 - When stylizing, does the generated face look too realistic? Adjust the Style strength to 30-50, the larger the number, the less ID fidelty, but the stylization ability will be better. You could also try out other base models or LoRAs with good stylization effects.

gradio_demo/app.py

@@ -2,6 +2,7 @@ import torch
 import numpy as np
 import random
 import os
+import sys
 
 from diffusers.utils import load_image
 from diffusers import EulerDiscreteScheduler
@@ -15,7 +16,16 @@ from style_template import styles
 
 # global variable
 base_model_path = 'SG161222/RealVisXL_V3.0'
-device = "cuda" if torch.cuda.is_available() else "cpu"
+try:
+    if torch.cuda.is_available():
+        device = "cuda"
+    elif sys.platform == "darwin" and torch.backends.mps.is_available():
+        device = "mps"
+    else:
+        device = "cpu"
+except:
+    device = "cpu"
+
 MAX_SEED = np.iinfo(np.int32).max
 STYLE_NAMES = list(styles.keys())
 DEFAULT_STYLE_NAME = "Photographic (Default)"
@@ -23,9 +33,13 @@ DEFAULT_STYLE_NAME = "Photographic (Default)"
 # download PhotoMaker checkpoint to cache
 photomaker_ckpt = hf_hub_download(repo_id="TencentARC/PhotoMaker", filename="photomaker-v1.bin", repo_type="model")
 
+if device == "mps":
+    torch_dtype = torch.float16
+else:
+    torch_dtype = torch.bfloat16
 pipe = PhotoMakerStableDiffusionXLPipeline.from_pretrained(
     base_model_path, 
-    torch_dtype=torch.bfloat16, 
+    torch_dtype=torch_dtype,
     use_safetensors=True, 
     variant="fp16",
     # local_files_only=True,
@@ -101,7 +115,7 @@ def randomize_seed_fn(seed: int, randomize_seed: bool) -> int:
         seed = random.randint(0, MAX_SEED)
     return seed
 
-def apply_style(style_name: str, positive: str, negative: str = "") -> tuple[str, str]:
+def apply_style(style_name: str, positive: str, negative: str = ""):
     p, n = styles.get(style_name, styles[DEFAULT_STYLE_NAME])
     return p.replace("{prompt}", positive), n + ' ' + negative
 
@@ -284,4 +298,4 @@ with gr.Blocks(css=css) as demo:
     
     gr.Markdown(article)
     
-demo.launch(share=False)
+demo.launch(share=False)
\ No newline at end of file