Create app.py

app.py

@@ -0,0 +1,43 @@
+import gradio as gr
+from ultralytics import YOLO
+import torch
+import numpy as np
+import cv2
+
+# Load the trained YOLO model from your Hugging Face Model Hub
+MODEL_REPO = "Alaaeldin/yolo-demo"  # Replace with your actual model repo
+model = YOLO("yolov8n.pt")  # Change to your trained model file if needed
+
+# Define the function for image prediction
+def predict(image):
+    results = model(image)
+
+    # Extract bounding boxes, labels, and confidence scores
+    detections = results[0].boxes.xyxy.cpu().numpy()
+    labels = results[0].boxes.cls.cpu().numpy()
+    confidences = results[0].boxes.conf.cpu().numpy()
+
+    # Load class names
+    class_names = ["apples", "banana", "cherry", "grapes", "oranges"]
+
+    # Draw bounding boxes on the image
+    for i, (box, label, conf) in enumerate(zip(detections, labels, confidences)):
+        x1, y1, x2, y2 = map(int, box)
+        label_text = f"{class_names[int(label)]}: {conf:.2f}"
+        cv2.rectangle(image, (x1, y1), (x2, y2), (0, 255, 0), 2)
+        cv2.putText(image, label_text, (x1, y1 - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 0), 2)
+
+    return image
+
+# Create Gradio Interface
+demo = gr.Interface(
+    fn=predict,
+    inputs=gr.Image(type="numpy"),
+    outputs=gr.Image(type="numpy"),
+    title="YOLOv8 Fruit Detection",
+    description="Upload an image to detect fruits using YOLOv8.",
+)
+
+# Run the app
+if __name__ == "__main__":
+    demo.launch()