main.py

import gradio as gr
import torch
import torch.nn as nn
from PIL import Image
import torchvision.transforms as T

# Define your model class
class UNetClassifier(nn.Module):
    def __init__(self, num_classes=1):
        super().__init__()
        def conv_block(in_c, out_c):
            return nn.Sequential(
                nn.Conv2d(in_c, out_c, 3, padding=1),
                nn.ReLU(),
                nn.Conv2d(out_c, out_c, 3, padding=1),
                nn.ReLU()
            )
        self.enc1 = conv_block(3, 64)
        self.enc2 = conv_block(64, 128)
        self.pool = nn.MaxPool2d(2)
        self.global_pool = nn.AdaptiveAvgPool2d((1, 1))
        self.fc = nn.Linear(128, num_classes)

    def forward(self, x):
        x = self.enc1(x)
        x = self.pool(x)
        x = self.enc2(x)
        x = self.pool(x)
        x = self.global_pool(x)
        x = torch.flatten(x, 1)
        x = self.fc(x)
        return torch.sigmoid(x)

# Load model
model = UNetClassifier()
model.load_state_dict(torch.load("model_weights.pth", map_location="cpu"))
model.eval()

# Image transform
transform = T.Compose([
    T.Resize((128, 128)),
    T.ToTensor()
])

# Prediction function
def classify_fire(image):
    image = image.convert("RGB")
    img = transform(image).unsqueeze(0)
    with torch.no_grad():
        output = model(img)
    return "🔥 FIRE" if output.item() > 0.5 else "✅ NO FIRE"

# Gradio UI
gr.Interface(fn=classify_fire, inputs="image", outputs="text", title="Fire Classifier").launch()