Update app.py

app.py

@@ -54,6 +54,7 @@ def generate_image_embeddings(zip_file):
     try:
         # Extract images from zip
         images = []
+        print(f"Extracting images from zip file: {zip_file.name}")
         with zipfile.ZipFile(zip_file.name, "r") as zip_ref:
             for file_info in zip_ref.filelist:
                 if file_info.filename.lower().endswith(
@@ -63,13 +64,17 @@ def generate_image_embeddings(zip_file):
                         img = Image.open(io.BytesIO(img_file.read())).convert("RGB")
                         images.append(img)
 
+        print(f"Extracted {len(images)} images from zip file")
+
         if len(images) == 0:
             return None, "❌ No valid images found in the zip file"
 
         # Generate embeddings
         embeddings = []
+        print(f"Generating embeddings for {len(images)} images...")
         with torch.no_grad():
             for i, image in enumerate(images):
+                print(f"Processing image {i + 1}/{len(images)}")
                 image_input = image_processor(
                     images=image,
                     max_num_patches=determine_max_value(image),
@@ -84,6 +89,7 @@ def generate_image_embeddings(zip_file):
                 embeddings.append(normalized_features.cpu().numpy())
 
         embeddings = np.vstack(embeddings)
+        print(f"Embeddings shape: {embeddings.shape}")
 
         # Create JSON output
         result = json.dumps(
@@ -96,10 +102,13 @@ def generate_image_embeddings(zip_file):
         )
 
         message = f"✅ Successfully generated embeddings for {len(images)} images\nShape: {embeddings.shape}"
+        print(message)
         return result, message
 
     except Exception as e:
-        return None, f"❌ Error: {str(e)}"
+        error_msg = f"❌ Error: {str(e)}"
+        print(error_msg)
+        return None, error_msg
 
 
 def extract_frames(video_path: str, fps: int = 4):
@@ -151,15 +160,19 @@ def generate_video_embeddings(video_file, fps):
     """
     try:
         # Extract frames
+        print(f"Extracting frames from video: {video_file.name} at {fps} fps")
         frames = extract_frames(video_file.name, fps)
+        print(f"Extracted {len(frames)} frames from video")
 
         if len(frames) == 0:
             return None, "❌ No frames could be extracted from the video"
 
         # Generate embeddings
         embeddings = []
+        print(f"Generating embeddings for {len(frames)} frames...")
         with torch.no_grad():
             for i, frame in enumerate(frames):
+                print(f"Processing frame {i + 1}/{len(frames)}")
                 image_input = image_processor(
                     images=frame,
                     max_num_patches=determine_max_value(frame),
@@ -174,6 +187,7 @@ def generate_video_embeddings(video_file, fps):
                 embeddings.append(normalized_features.cpu().numpy())
 
         embeddings = np.vstack(embeddings)
+        print(f"Embeddings shape: {embeddings.shape}")
 
         # Create JSON output
         result = json.dumps(
@@ -187,10 +201,13 @@ def generate_video_embeddings(video_file, fps):
         )
 
         message = f"✅ Successfully generated embeddings for {len(frames)} frames (extracted at {fps} fps)\nShape: {embeddings.shape}"
+        print(message)
         return result, message
 
     except Exception as e:
-        return None, f"❌ Error: {str(e)}"
+        error_msg = f"❌ Error: {str(e)}"
+        print(error_msg)
+        return None, error_msg
 
 
 # Create Gradio interface