app.py

import random
import os
import uuid
from datetime import datetime
import gradio as gr
import numpy as np
import spaces
import torch
from diffusers import AutoPipelineForText2Image
from PIL import Image

# Create permanent storage directory
SAVE_DIR = "saved_images"  # Gradio will handle the persistence
if not os.path.exists(SAVE_DIR):
    os.makedirs(SAVE_DIR, exist_ok=True)

device = "cuda" if torch.cuda.is_available() else "cpu"
repo_id = "black-forest-labs/FLUX.1-dev"
lora_id = "seawolf2357/nsfw-detection"  # LoRA model

print("Loading pipeline...")
# Use AutoPipelineForText2Image which has better compatibility with LoRA loading
pipeline = AutoPipelineForText2Image.from_pretrained(
    repo_id, 
    torch_dtype=torch.bfloat16,
    use_safetensors=True
)
pipeline = pipeline.to(device)

# Try to load the LoRA with direct method (simpler approach)
print("Loading LoRA weights...")
try:
    pipeline.load_lora_weights(lora_id)
    print("LoRA weights loaded successfully!")
    lora_loaded = True
except Exception as e:
    print(f"Could not load LoRA weights using standard method: {e}")
    print("Continuing without LoRA functionality.")
    lora_loaded = False

MAX_SEED = np.iinfo(np.int32).max
MAX_IMAGE_SIZE = 1024

def save_generated_image(image, prompt):
    # Generate unique filename with timestamp
    timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
    unique_id = str(uuid.uuid4())[:8]
    filename = f"{timestamp}_{unique_id}.png"
    filepath = os.path.join(SAVE_DIR, filename)
    
    # Save the image
    image.save(filepath)
    
    # Save metadata
    metadata_file = os.path.join(SAVE_DIR, "metadata.txt")
    with open(metadata_file, "a", encoding="utf-8") as f:
        f.write(f"{filename}|{prompt}|{timestamp}\n")
    
    return filepath

# Function to ensure "nsfw" and "[trigger]" are in the prompt
def process_prompt(prompt):
    # Add "nsfw" prefix if not already present
    if not prompt.lower().startswith("nsfw "):
        prompt = "nsfw " + prompt
    
    # Add "[trigger]" suffix if not already present
    if not prompt.lower().endswith("[trigger]"):
        if prompt.endswith(" "):
            prompt = prompt + "[trigger]"
        else:
            prompt = prompt + " [trigger]"
    
    return prompt

@spaces.GPU(duration=120)
def inference(
    prompt: str,
    seed: int,
    randomize_seed: bool,
    width: int,
    height: int,
    guidance_scale: float,
    num_inference_steps: int,
    lora_scale: float,
    progress: gr.Progress = gr.Progress(track_tqdm=True),
):
    # Process the prompt to ensure it has the required format
    processed_prompt = process_prompt(prompt)
    
    if randomize_seed:
        seed = random.randint(0, MAX_SEED)
    generator = torch.Generator(device=device).manual_seed(seed)
    
    try:
        # Try with cross_attention_kwargs if LoRA was loaded successfully
        if lora_loaded:
            image = pipeline(
                prompt=processed_prompt,
                guidance_scale=guidance_scale,
                num_inference_steps=num_inference_steps,
                width=width,
                height=height,
                generator=generator,
                cross_attention_kwargs={"scale": lora_scale}
            ).images[0]
        else:
            # Fall back to standard generation if LoRA wasn't loaded
            image = pipeline(
                prompt=processed_prompt,
                guidance_scale=guidance_scale,
                num_inference_steps=num_inference_steps,
                width=width,
                height=height,
                generator=generator,
            ).images[0]
    except Exception as e:
        print(f"Error during inference with cross_attention_kwargs: {e}")
        # Fall back to standard generation without LoRA parameters
        image = pipeline(
            prompt=processed_prompt,
            guidance_scale=guidance_scale,
            num_inference_steps=num_inference_steps,
            width=width,
            height=height,
            generator=generator,
        ).images[0]
    
    # Save the generated image
    filepath = save_generated_image(image, processed_prompt)
    
    # Return the image, seed, and processed prompt
    return image, seed, processed_prompt

examples = [
    "A young couple, their bodies glistening with sweat, make love in the rain, the woman"
]

# Brighter custom CSS with vibrant colors
custom_css = """
:root {
    --color-primary: #FF9E6C;
    --color-secondary: #FFD8A9;
}
footer {
    visibility: hidden;
}
.gradio-container {
    background: linear-gradient(to right, #FFF4E0, #FFEDDB);
}
.title {
    color: #E25822 !important;
    font-size: 2.5rem !important;
    font-weight: 700 !important;
    text-align: center;
    margin: 1rem 0;
    text-shadow: 2px 2px 4px rgba(0,0,0,0.1);
}
.subtitle {
    color: #2B3A67 !important;
    font-size: 1.2rem !important;
    text-align: center;
    margin-bottom: 2rem;
}
.model-description {
    background-color: rgba(255, 255, 255, 0.7);
    border-radius: 10px;
    padding: 20px;
    margin: 20px 0;
    box-shadow: 0 4px 6px rgba(0, 0, 0, 0.1);
    border-left: 5px solid #E25822;
}
button.primary {
    background-color: #E25822 !important;
}
button:hover {
    transform: translateY(-2px);
    box-shadow: 0 5px 15px rgba(0,0,0,0.1);
}
"""

with gr.Blocks(css=custom_css, analytics_enabled=False) as demo:
    gr.HTML('<div class="title">NSFW Detection STUDIO</div>')
    
    # Main generation interface
    with gr.Column(elem_id="col-container"):
        with gr.Row():
            prompt = gr.Text(
                label="Prompt",
                show_label=False,
                max_lines=1,
                placeholder="Enter your prompt (nsfw and [trigger] will be added automatically)",
                container=False,
            )
            run_button = gr.Button("Generate", variant="primary", scale=0)

        result = gr.Image(label="Result", show_label=False)
        processed_prompt_display = gr.Textbox(label="Processed Prompt", show_label=True)

        with gr.Accordion("Advanced Settings", open=False):
            seed = gr.Slider(
                label="Seed",
                minimum=0,
                maximum=MAX_SEED,
                step=1,
                value=42,
            )
            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,
                )
                height = gr.Slider(
                    label="Height",
                    minimum=256,
                    maximum=MAX_IMAGE_SIZE,
                    step=32,
                    value=768,
                )

            with gr.Row():
                guidance_scale = gr.Slider(
                    label="Guidance scale",
                    minimum=0.0,
                    maximum=10.0,
                    step=0.1,
                    value=3.5,
                )
                num_inference_steps = gr.Slider(
                    label="Number of inference steps",
                    minimum=1,
                    maximum=50,
                    step=1,
                    value=30,
                )
                lora_scale = gr.Slider(
                    label="LoRA scale",
                    minimum=0.0,
                    maximum=1.0,
                    step=0.1,
                    value=1.0,
                )

        gr.Examples(
            examples=examples,
            inputs=[prompt],
            outputs=[result, seed, processed_prompt_display],
        )

    gr.on(
        triggers=[run_button.click, prompt.submit],
        fn=inference,
        inputs=[
            prompt,
            seed,
            randomize_seed,
            width,
            height,
            guidance_scale,
            num_inference_steps,
            lora_scale,
        ],
        outputs=[result, seed, processed_prompt_display],
    )

demo.queue()
demo.launch()