Update app.py

app.py

@@ -1,60 +1,32 @@
 import gradio as gr
-import numpy as np
-import random
-from diffusers import StableDiffusionPipeline
 import torch
+from diffusers import StableDiffusionPipeline
 from safetensors.torch import load_file
-import os
 
-# CUDA 사용 여부 확인
-device = "cuda" if torch.cuda.is_available() else "cpu"
-
-# 모델 로드 함수
+# safetensors 모델 로드 함수
 def load_safetensors_model(model_id, model_file):
-    try:
-        # safetensors 파일 로드
-        state_dict = load_file(model_file)
-        # 모델 초기화
-        print("Loading model from pretrained...")
-        model = StableDiffusionPipeline.from_pretrained(model_id)
-        print("Model loaded from pretrained.")
-        # 모델에 state_dict 로드
-        model.unet.load_state_dict(state_dict)
-        model = model.to(device)
-        print("Model state_dict loaded and moved to device.")
-        return model
-    except Exception as e:
-        print(f"Error loading model: {e}")
-        return None
+    # safetensors 파일 로드
+    state_dict = load_file(model_file)
+    # 모델 초기화
+    model = StableDiffusionPipeline.from_pretrained(model_id)
+    # 모델에 state_dict 로드
+    model.model.load_state_dict(state_dict)
+    return model
 
-# 모델 ID와 파일 경로 설정
-model_id = "stabilityai/stable-diffusion-2"
-model_file = "./Traditional_Korean_Painting_Model_2.safetensors"
+model_id = "gagong/Traditional-Korean-Painting-Model-v2.0"
+model_file = "./Traditional_Korean_Painting_Model_2.safetensors"  # 루트 디렉토리에 있는 파일 경로 설정
 
-pipe = load_safetensors_model(model_id, model_file)
+try:
+    pipe = load_safetensors_model(model_id, model_file)
+    pipe = pipe.to("cuda" if torch.cuda.is_available() else "cpu")
+except Exception as e:
+    print(f"Error loading model: {e}")
+    pipe = None
 
-MAX_SEED = np.iinfo(np.int32).max
-MAX_IMAGE_SIZE = 1024
-
-def infer(prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps):
+def generate_image(prompt):
     if not pipe:
-        return "Model not loaded properly"
-
-    if randomize_seed:
-        seed = random.randint(0, MAX_SEED)
-        
-    generator = torch.Generator().manual_seed(seed)
-    
-    image = pipe(
-        prompt=prompt, 
-        negative_prompt=negative_prompt,
-        guidance_scale=guidance_scale, 
-        num_inference_steps=num_inference_steps, 
-        width=width, 
-        height=height,
-        generator=generator
-    ).images[0] 
-    
+        return None
+    image = pipe(prompt).images[0]
     return image
 
 examples = [