fix_preprocessing.py

import pickle\nimport pandas as pd\nimport numpy as np\nimport logging\n\n# Configure logging\nlogging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')\n\n# Load model and scaler\nwith open('vehicle_load_model.pkl', 'rb') as f:\n    model = pickle.load(f)\n\nwith open('vehicle_load_scaler.pkl', 'rb') as f:\n    scaler = pickle.load(f)\n\n# Print model features\nprint('Model expects these features:')\nfor feature in model.feature_names_in_:\n    print(f'- {feature}')\n\n# Create test data\ntest_data = {\n    'vehicle_type': '2-wheeler',\n    'weight': 100,\n    'max_load_capacity': 150,\n    'passenger_count': 2,\n    'cargo_weight': 20,\n    'region': 'Urban',\n    'road_condition': 'Good',\n    'weather': 'Clear'\n}\n\n# Fixed preprocessing function\ndef fixed_preprocess_data(data):\n    \
\\Fixed
preprocessing
function
that
ensures
feature
names
match
training
data.\\\\n    # Convert to DataFrame if dict\n    if isinstance(data, dict):\n        df = pd.DataFrame([data])\n    else:\n        df = pd.DataFrame(data)\n    \n    # Rename 'weather' to 'weather_condition' if needed\n    if 'weather' in df.columns and 'weather_condition' not in df.columns:\n        df['weather_condition'] = df['weather']\n        df = df.drop('weather', axis=1)\n    \n    # Define numerical features (these need to be scaled)\n    numerical_features = ['weight', 'max_load_capacity', 'passenger_count', 'cargo_weight']\n    \n    # Ensure all numerical features exist\n    for feature in numerical_features:\n        if feature not in df.columns:\n            df[feature] = 0\n    \n    # Scale numerical features\n    numerical_data = df[numerical_features].values\n    scaled_data = scaler.transform(numerical_data)\n    \n    # Create a new DataFrame with just the features the model expects\n    result_df = pd.DataFrame()\n    \n    # Add scaled numerical features\n    for i, feature in enumerate(numerical_features):\n        result_df[f\
feature
_scaled\] = scaled_data[:, i]\n    \n    # Handle one-hot encoding for vehicle_type\n    vehicle_types = ['2-wheeler', '4-wheeler 5-seater', '4-wheeler 7-seater', 'delivery vehicle', 'heavy vehicle']\n    current_type = df['vehicle_type'].iloc[0] if 'vehicle_type' in df.columns else None\n    \n    for vtype in vehicle_types:\n        result_df[f\vehicle_type_
vtype
\] = 1 if current_type == vtype else 0\n    \n    # Make sure columns are in the same order as the model expects\n    result_df = result_df[model.feature_names_in_]\n    \n    return result_df\n\n# Process test data\ntry:\n    processed_data = fixed_preprocess_data(test_data)\n    \n    # Print processed data\n    print('\\nProcessed data:')\n    print(processed_data)\n    \n    # Make prediction\n    prediction = model.predict(processed_data)\n    print(f'\\nPrediction: {prediction[0]}')\n    \n    print('\\nThis fixed preprocessing function works! Update app.py to use this approach.')\nexcept Exception as e:\n    print(f'Error: {e}')