Fix internal server error: Add memory management and improve error handling

Dockerfile

@@ -6,6 +6,7 @@ WORKDIR /code
 RUN apt-get update && apt-get install -y \
     libgl1-mesa-glx \
     libglib2.0-0 \
+    ffmpeg \
     && rm -rf /var/lib/apt/lists/*
 
 # Copy requirements first to leverage Docker cache
@@ -16,9 +17,11 @@ RUN pip install --no-cache-dir -r requirements.txt
 RUN mkdir -p /code/static/uploads \
     && mkdir -p /code/logs \
     && mkdir -p /code/external/BodybuildingPoseClassifier \
+    && mkdir -p /code/templates \
     && chmod -R 777 /code/static/uploads \
     && chmod -R 777 /code/logs \
-    && chmod -R 777 /code/external
+    && chmod -R 777 /code/external \
+    && chmod -R 777 /code/templates
 
 # Copy the model file first
 COPY external/BodybuildingPoseClassifier/bodybuilding_pose_classifier.h5 /code/external/BodybuildingPoseClassifier/
@@ -29,6 +32,7 @@ COPY . .
 # Set environment variables
 ENV PYTHONUNBUFFERED=1
 ENV FLASK_APP=app.py
+ENV TF_CPP_MIN_LOG_LEVEL=2
 
 # Expose the port
 EXPOSE 7860

app.py

@@ -42,6 +42,23 @@ logging.basicConfig(
 )
 logger = logging.getLogger(__name__)
 
+def log_memory_usage():
+    """Log current memory usage."""
+    try:
+        process = psutil.Process()
+        memory_info = process.memory_info()
+        logger.info(f"Memory usage: {memory_info.rss / 1024 / 1024:.2f} MB")
+    except Exception as e:
+        logger.error(f"Error logging memory usage: {e}")
+
+def cleanup_memory():
+    """Force garbage collection and log memory usage."""
+    try:
+        gc.collect()
+        log_memory_usage()
+    except Exception as e:
+        logger.error(f"Error in cleanup_memory: {e}")
+
 # Add file handler for persistent logging
 log_dir = 'logs'
 os.makedirs(log_dir, exist_ok=True)
@@ -83,6 +100,19 @@ except Exception as e:
     logger.error(traceback.format_exc())
     raise
 
+# Initialize TensorFlow session with memory growth
+try:
+    gpus = tf.config.list_physical_devices('GPU')
+    if gpus:
+        for gpu in gpus:
+            tf.config.experimental.set_memory_growth(gpu, True)
+        logger.info("GPU memory growth enabled")
+    else:
+        logger.info("No GPU found, using CPU")
+except Exception as e:
+    logger.error(f"Error configuring GPU: {e}")
+    logger.error(traceback.format_exc())
+
 cnn_class_labels = ['side_chest', 'front_double_biceps', 'back_double_biceps', 'front_lat_spread', 'back_lat_spread']
 
 def predict_pose_cnn(img_path):
@@ -371,6 +401,21 @@ def process_video_mediapipe(video_path):
 def index():
     return render_template('index.html')
 
+# Add error handling for video processing
+def safe_video_processing(video_path, model_choice):
+    """Wrapper function to handle video processing with proper cleanup."""
+    try:
+        if model_choice == 'movenet':
+            return process_video_movenet(video_path)
+        else:
+            return process_video_mediapipe(video_path)
+    except Exception as e:
+        logger.error(f"Error in video processing: {e}")
+        logger.error(traceback.format_exc())
+        raise
+    finally:
+        cleanup_memory()
+
 @app.route('/upload', methods=['POST'])
 def upload_file():
     try:
@@ -416,13 +461,7 @@ def upload_file():
                 model_choice = request.form.get('model_choice', 'Gladiator SupaDot')
                 logger.info(f"[UPLOAD] Processing with model: {model_choice}")
                 
-                if model_choice == 'movenet':
-                    movenet_variant = request.form.get('movenet_variant', 'lightning')
-                    logger.info(f"[UPLOAD] Using MoveNet variant: {movenet_variant}")
-                    output_path_url = process_video_movenet(filepath)
-                else:
-                    output_path_url = process_video_mediapipe(filepath)
-                    
+                output_path_url = safe_video_processing(filepath, model_choice)
                 logger.info(f"[UPLOAD] Processing complete. Output URL: {output_path_url}")
                 
                 output_path = os.path.join(app.config['UPLOAD_FOLDER'], os.path.basename(output_path_url))