Benchmarking

app.py

@@ -4,6 +4,7 @@ import onnx
 import onnxruntime
 from PIL import Image, ImageDraw, ImageFont
 import gradio as gr
+import time  # Import time for benchmarking
 
 # Constants
 PROB_THRESHOLD = 0.5  # Minimum probability to show results
@@ -44,9 +45,15 @@ class Model:
         if not self.is_range255:
             input_array = input_array / 255.0  # Normalize to [0,1]
 
-        # Run inference
+        # Run inference with benchmarking
+        start_time = time.time()  # Start timing
         outputs = self.session.run(self.output_names, {self.input_name: input_array.astype(self.input_type)})
-        return {name: outputs[i] for i, name in enumerate(self.output_names)}
+        end_time = time.time()  # End timing
+
+        execution_time = (end_time - start_time) * 1000  # Convert to milliseconds
+        print(f"Inference time: {execution_time:.2f} ms")
+
+        return {name: outputs[i] for i, name in enumerate(self.output_names)}, execution_time
 
 def draw_boxes(image: Image.Image, outputs: dict):
     draw = ImageDraw.Draw(image, "RGBA")  # Use RGBA for transparency
@@ -115,7 +122,7 @@ def draw_boxes(image: Image.Image, outputs: dict):
 model = Model(MODEL_PATH)
 
 def detect_objects(image):
-    outputs = model.predict(image)
+    outputs, execution_time = model.predict(image)
     annotated_image = draw_boxes(image.copy(), outputs)
 
     # Prepare detection summary
@@ -131,6 +138,7 @@ def detect_objects(image):
         detections.append(f"{label}: {score:.2f}")
 
     detection_summary = "\n".join(detections) if detections else "No objects detected."
+    detection_summary += f"\n\nInference Time: {execution_time:.2f} ms"
 
     return annotated_image, detection_summary
 
@@ -153,4 +161,4 @@ iface = gr.Interface(
 )
 
 if __name__ == "__main__":
-    iface.launch()
+    iface.launch()