Update app.py

app.py

@@ -57,41 +57,23 @@ def randomize_seed_fn(seed: int, randomize_seed: bool) -> int:
 
 
 @spaces.GPU(enable_queue=True)
-def generate(
-    prompt: str,
-    negative_prompt: str = "",
-    use_negative_prompt: bool = False,
-    seed: int = 0,
-    width: int = 1024,
-    height: int = 1024,
-    guidance_scale: float = 3,
-    randomize_seed: bool = False,
-    use_resolution_binning: bool = True,
-    progress=gr.Progress(track_tqdm=True),
-):
-    pipe.to(device)
-    seed = int(randomize_seed_fn(seed, randomize_seed))
-    generator = torch.Generator().manual_seed(seed)
-
-    if not use_negative_prompt:
-        negative_prompt = None  # type: ignore
+def generate(prompt, negative_prompt, guidance_scale, num_inference_steps):
+    # Ensure consistent data type
+    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+    dtype = torch.float32  # or torch.float16 if you're using half precision
+    
+    pipe = pipe.to(device)
+    pipe.text_encoder = pipe.text_encoder.to(dtype)
+    pipe.unet = pipe.unet.to(dtype)
     
     images = pipe(
         prompt=prompt,
         negative_prompt=negative_prompt,
-        width=width,
-        height=height,
         guidance_scale=guidance_scale,
-        num_inference_steps=25,
-        generator=generator,
-        num_images_per_prompt=NUM_IMAGES_PER_PROMPT,
-        use_resolution_binning=use_resolution_binning,
-        output_type="pil",
+        num_inference_steps=num_inference_steps,
     ).images
+    return images[0]
 
-    image_paths = [save_image(img) for img in images]
-    print(image_paths)
-    return image_paths, seed
 
 
 examples = [