Update app.py

app.py

@@ -1,157 +1,72 @@
-import gradio as gr
-import numpy as np
-import random
-import peft
-
-# import spaces #[uncomment to use ZeroGPU]
-from diffusers import DiffusionPipeline
 import torch
-
-device = "cuda" if torch.cuda.is_available() else "cpu"
-model_repo_id = "stabilityai/sdxl-turbo"  # Replace to the model you would like to use
-
-if torch.cuda.is_available():
-    torch_dtype = torch.float16
-else:
-    torch_dtype = torch.float32
-
-pipe = DiffusionPipeline.from_pretrained(model_repo_id, torch_dtype=torch_dtype)
-pipe.load_lora_weights("fofr/sdxl-emoji", weight_name="lora.safetensors", adapter_name="toy")
-
-pipe = pipe.to(device)
-
-MAX_SEED = np.iinfo(np.int32).max
-MAX_IMAGE_SIZE = 1024
-
-
-# @spaces.GPU #[uncomment to use ZeroGPU]
-def infer(
-    prompt,
-    negative_prompt,
-    seed,
-    randomize_seed,
-    width,
-    height,
-    guidance_scale,
-    num_inference_steps,
-    progress=gr.Progress(track_tqdm=True),
-):
-    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 image, seed
-
-
-examples = [
-    "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k",
-    "An astronaut riding a green horse",
-    "A delicious ceviche cheesecake slice",
-]
-
-css = """
-#col-container {
-    margin: 0 auto;
-    max-width: 640px;
-}
-"""
-
-with gr.Blocks(css=css) as demo:
-    with gr.Column(elem_id="col-container"):
-        gr.Markdown(" # Text-to-Image Gradio Template")
-
-        with gr.Row():
-            prompt = gr.Text(
-                label="Prompt",
-                show_label=False,
-                max_lines=1,
-                placeholder="Enter your prompt",
-                container=False,
-            )
-
-            run_button = gr.Button("Run", scale=0, variant="primary")
-
-        result = gr.Image(label="Result", show_label=False)
-
-        with gr.Accordion("Advanced Settings", open=False):
-            negative_prompt = gr.Text(
-                label="Negative prompt",
-                max_lines=1,
-                placeholder="Enter a negative prompt",
-                visible=False,
-            )
-
-            seed = gr.Slider(
-                label="Seed",
-                minimum=0,
-                maximum=MAX_SEED,
-                step=1,
-                value=0,
-            )
-
-            randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
-
-            with gr.Row():
-                width = gr.Slider(
-                    label="Width",
-                    minimum=256,
-                    maximum=MAX_IMAGE_SIZE,
-                    step=32,
-                    value=1024,  # Replace with defaults that work for your model
-                )
-
-                height = gr.Slider(
-                    label="Height",
-                    minimum=256,
-                    maximum=MAX_IMAGE_SIZE,
-                    step=32,
-                    value=1024,  # Replace with defaults that work for your model
-                )
-
-            with gr.Row():
-                guidance_scale = gr.Slider(
-                    label="Guidance scale",
-                    minimum=0.0,
-                    maximum=10.0,
-                    step=0.1,
-                    value=0.0,  # Replace with defaults that work for your model
-                )
-
-                num_inference_steps = gr.Slider(
-                    label="Number of inference steps",
-                    minimum=1,
-                    maximum=50,
-                    step=1,
-                    value=2,  # Replace with defaults that work for your model
-                )
-
-        gr.Examples(examples=examples, inputs=[prompt])
-    gr.on(
-        triggers=[run_button.click, prompt.submit],
-        fn=infer,
-        inputs=[
-            prompt,
-            negative_prompt,
-            seed,
-            randomize_seed,
-            width,
-            height,
-            guidance_scale,
-            num_inference_steps,
-        ],
-        outputs=[result, seed],
-    )
-
-if __name__ == "__main__":
-    demo.launch()
+import gradio as gr
+from diffusers import StableDiffusionXLPipeline
+
+# --- Settings and paths ---
+# Base SDXL model – change this to the base model you want to use.
+BASE_MODEL = "stabilityai/stable-diffusion-xl-base-1.0"
+# Path to your LoRA weights (assumed to be in a format that Diffusers can use)
+LORA_PATH = "fofr/sdxl-emoji"
+
+# --- Load the base pipeline ---
+pipe = StableDiffusionXLPipeline.from_pretrained(
+    BASE_MODEL,
+    torch_dtype=torch.float16,
+    variant="fp16",           # Use FP16 variant if available for speed
+    safety_checker=None,      # (Optional) disable safety checker to speed things up
+)
+pipe.to("cuda")
+
+# --- Enable fast attention if available ---
+try:
+    pipe.enable_xformers_memory_efficient_attention()
+except Exception as e:
+    print("xFormers not enabled:", e)
+
+# --- Apply the LoRA weights ---
+# Diffusers v0.18+ supports applying LoRA weights to parts of the pipeline.
+# Here we assume the LoRA affects the UNet (and, if needed, the text encoder).
+try:
+    # For the UNet:
+    pipe.unet.load_attn_procs(LORA_PATH)
+    # If you also have LoRA weights for the text encoder, you might do:
+    # pipe.text_encoder.load_attn_procs(LORA_PATH)
+except Exception as e:
+    print("Error applying LoRA weights:", e)
+
+# --- Define the image generation function ---
+def generate_image(prompt: str, steps: int = 30, guidance: float = 7.5):
+    """
+    Generate an image from a text prompt.
+    
+    Args:
+        prompt (str): The text prompt.
+        steps (int): Number of inference steps.
+        guidance (float): Guidance scale (higher values encourage the image to follow the prompt).
+        
+    Returns:
+        A generated PIL image.
+    """
+    # Use autocast for faster FP16 inference on CUDA
+    with torch.cuda.amp.autocast():
+        result = pipe(prompt, num_inference_steps=steps, guidance_scale=guidance)
+    return result.images[0]
+
+# --- Build the Gradio interface ---
+demo = gr.Interface(
+    fn=generate_image,
+    inputs=[
+        gr.Textbox(lines=2, placeholder="Enter your prompt here...", label="Prompt"),
+        gr.Slider(minimum=10, maximum=100, step=5, value=30, label="Inference Steps"),
+        gr.Slider(minimum=1.0, maximum=15.0, step=0.5, value=7.5, label="Guidance Scale")
+    ],
+    outputs=gr.Image(type="pil", label="Generated Image"),
+    title="Super Fast SDXL-Emoji Generator",
+    description=(
+        "This demo uses a Stable Diffusion XL model enhanced with a custom LoRA "
+        "to generate images quickly. Adjust the prompt and settings below, then hit 'Submit'!"
+    ),
+)
+
+# --- Launch the demo ---
+demo.launch()