src/data_loader.py

# src/data_loader.py
import pandas as pd

def load_data(file_path: str) -> pd.DataFrame | None:
    try:
        if file_path.endswith('.csv'):
            df = pd.read_csv(file_path)
        elif file_path.endswith(('.xls', '.xlsx')):
            df = pd.read_excel(file_path)
        else:
            print("Unsupported file format.")
            return None

        # Basic validation: Ensure 'Department' column exists.
        # Subject-specific columns will be handled by the preprocessor.
        if 'Department'  not in df.columns:
            print("Missing required column: 'Department'")
            return None

        print("Data loaded successfully (wide format).")
        return df
    except Exception as e:
        print(f"Error loading data: {e}")
        return None