基于列组条件的DataFrame样式设置：分组高亮不等值

To achieve the desired conditional formatting in your pandas DataFrame, we can use pandas.Styler with a custom function that applies highlighting based on your criteria. Here's a step-by-step implementation:

Step 1: Import pandas and create the DataFrame #

import pandas as pd
from pandas import DataFrame

# Create the DataFrame as specified
df = DataFrame({
    'A': ['Bob', 'Rob', 'Dob'],
    'B': ['Bob', 'Rob', 'Dob'],
    'C': ['Bob', 'Dob', 'Dob'],
    'D': ['Ben', 'Ten', 'Zen'],
    'E': ['Ben', 'Ten', 'Zu']
})

Step 2: Define the custom styling function #

This function checks each row to apply targeted highlighting:

• For columns A, B, C: Highlights cells that differ from the most common value (mode) in the row (catching any cell that doesn’t match the majority of values in these three columns).
• For columns D, E: Highlights both cells if their values are not equal.

def highlight_differences(row):
    # Initialize empty list to store styles for each cell
    cell_styles = [''] * len(row)
    
    # Handle A/B/C columns: Highlight non-matching values
    abc_values = row[['A', 'B', 'C']]
    if abc_values.nunique() != 1:
        # Get the most frequent value in the row's A/B/C columns
        most_common = abc_values.mode()[0]
        # Apply light red background to cells that don't match the most common value
        for col_idx, col_name in enumerate(['A', 'B', 'C']):
            if row[col_name] != most_common:
                cell_styles[col_idx] = 'background-color: #ffcccc'
    
    # Handle D/E columns: Highlight if values are unequal
    if row['D'] != row['E']:
        # Get indices of D and E columns
        d_col_idx = row.index.get_loc('D')
        e_col_idx = row.index.get_loc('E')
        # Apply light green background to both cells
        cell_styles[d_col_idx] = 'background-color: #ccffcc'
        cell_styles[e_col_idx] = 'background-color: #ccffcc'
    
    return cell_styles

Step 3: Apply the styling to the DataFrame #

# Apply the custom style function to each row
styled_df = df.style.apply(highlight_differences, axis=1)

# Display the styled DataFrame (works in Jupyter/IPython environments)
styled_df

Expected Result #

When you run this code, the styled DataFrame will have:

• Row 1: Column C (value 'Dob') with a light red background (since it differs from A and B's 'Rob').
• Row 2: Columns D ('Zen') and E ('Zu') with a light green background (since their values don’t match).
• All other cells remain unhighlighted as they meet the matching criteria.

You can tweak the background colors by changing the hex codes in the background-color properties (e.g., #ff9999 for darker red, #99ff99 for darker green).

内容的提问来源于stack exchange，提问作者Sharvari Gc