app.py

#srlsy bruh... checkin the code??

import gradio as gr
import torch
import torch.nn.functional as F
from torchvision import transforms
from PIL import Image
from transformers import ViTForImageClassification, ViTImageProcessor

processor = ViTImageProcessor.from_pretrained("google/vit-base-patch16-224-in21k")

human_model = ViTForImageClassification.from_pretrained(
    "google/vit-base-patch16-224-in21k", num_labels=2
)
human_model.load_state_dict(torch.load("humanNsfw_Swf.pth", map_location="cpu"))
human_model.eval()

anime_model = ViTForImageClassification.from_pretrained(
    "google/vit-base-patch16-224-in21k", num_labels=2
)
anime_model.load_state_dict(torch.load("animeCartoonNsfw_Sfw.pth", map_location="cpu"))
anime_model.eval()


def preprocess(image: Image.Image):
    inputs = processor(images=image, return_tensors="pt")
    return inputs["pixel_values"]

def predict(image, model_type):
    if image is None:
        return "<div class='result-box'>pls upload an img...</div>"

    inputs = preprocess(image)
    model = human_model if model_type == "Human" else anime_model

    with torch.no_grad():
        outputs = model(pixel_values=inputs)
        logits = outputs.logits
        probs = F.softmax(logits, dim=1)
        pred_class = torch.argmax(probs, dim=1).item()
        confidence = probs[0][pred_class].item()

    label = "NSFW" if pred_class == 0 else "SFW"
    return f"""
    <div class='result-box'>
        <strong>Model:</strong> {model_type}<br>
        <strong>Prediction:</strong> {label}<br>
        <strong>Confidence:</strong> {confidence:.2%}
    </div>
    """
custom_css = """
.result-box {
    position: relative;
    background-color: black;
    padding: 20px;
    border-radius: 12px;
    color: white;
    font-size: 1.2rem;
    text-align: center;
    font-weight: bold;
    width: 100%;
    z-index: 1;
    overflow: hidden;
    box-shadow: 0 0 15px oklch(0.718 0.202 349.761);
}

.result-box::before {
    content: "";
    position: absolute;
    top: -4px;
    left: -4px;
    right: -4px;
    bottom: -4px;
    background: conic-gradient(from 0deg, oklch(0.718 0.202 349.761), transparent 40%, oklch(0.718 0.202 349.761));
    border-radius: 16px;
    animation: spin 3s linear infinite;
    z-index: -1;
    filter: blur(8px);
}

@keyframes spin {
    0% { transform: rotate(0deg); }
    100% { transform: rotate(360deg); }
}

.disclaimer {
    color: white;
    font-size: 0.9rem;
    text-align: center;
    margin-top: 40px;
    text-shadow: 0 0 10px oklch(0.718 0.202 349.761);
}

.gradio-container {
    max-width: 900px;
    margin: auto;
}
"""


# ui
with gr.Blocks(css=custom_css) as demo:
    gr.Markdown("## NSFW Detector (Human and Anime/Cartoon)")
    gr.Markdown(
        "Upload an img and select the appropriate model. The system will detect whether the content is NSFW or SFW."
    )

    with gr.Row():
        with gr.Column(scale=1):
            model_choice = gr.Radio(["Human", "Anime"], label="Select Model Type", value="Human")
            image_input = gr.Image(type="pil", label="Upload Image")
        with gr.Column(scale=1):
            output_box = gr.HTML("<div class='result-box'>Awaiting input...</div>")

    image_input.change(fn=predict, inputs=[image_input, model_choice], outputs=output_box)
    model_choice.change(fn=predict, inputs=[image_input, model_choice], outputs=output_box)

    # Disclaimer with glow
    gr.Markdown(
        "<div class='disclaimer'>This is a side project. Results are not guaranteed. No images are stored.For more info, pls check readme file</div>"
    )

if __name__ == "__main__":
    demo.launch()