Create app.py

app.py

@@ -0,0 +1,63 @@
+import json
+import torch
+import gradio as gr
+import numpy as np
+from PIL import Image
+from torchvision import transforms
+
+from model import FoodIngredientClassifier
+
+DEVICE = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+# Load classes
+with open("mlb_classes.json", "r") as f:
+    classes = json.load(f)
+
+NUM_CLASSES = len(classes)
+
+# Load model
+model = FoodIngredientClassifier(NUM_CLASSES)
+checkpoint = torch.load("best_model.pth", map_location=DEVICE)
+model.load_state_dict(checkpoint["model_state_dict"])
+model.to(DEVICE)
+model.eval()
+
+# Image preprocessing
+transform = transforms.Compose([
+    transforms.Resize((224, 224)),
+    transforms.ToTensor(),
+    transforms.Normalize(
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]
+    )
+])
+
+def predict(image, threshold=0.5):
+    image = image.convert("RGB")
+    x = transform(image).unsqueeze(0).to(DEVICE)
+
+    with torch.no_grad():
+        logits = model(x)
+        probs = torch.sigmoid(logits).cpu().numpy()[0]
+
+    results = [
+        (classes[i], float(probs[i]))
+        for i in np.where(probs > threshold)[0]
+    ]
+
+    results.sort(key=lambda x: x[1], reverse=True)
+    return results[:10]
+
+demo = gr.Interface(
+    fn=predict,
+    inputs=[
+        gr.Image(type="pil", label="Upload Food Image"),
+        gr.Slider(0.1, 0.9, value=0.5, label="Confidence Threshold")
+    ],
+    outputs=gr.Label(label="Detected Ingredients"),
+    title="🍽️ Food Ingredient Detector",
+    description="Upload a food image and detect likely ingredients using a ResNet50 multi-label model."
+)
+
+if __name__ == "__main__":
+    demo.launch()