df_sample = df.head(100)  # Reduce sample size for faster processing