在R中批量合并对应列：无需逐一操作生成新组合列

Hey there! I totally get the pain of manually merging 20 column pairs—let's fix that with a scalable, loop-based approach that'll handle all your pairs in one go.

First, Let's Set Up a Sample DataFrame #

To make this concrete, here's a small example matching your structure (you can scale this to 20 pairs easily):

import pandas as pd

# Sample DataFrame with 2 A-C pairs (replace with your actual data)
df = pd.DataFrame({
    'A1': ['B1U', 'B2U'],
    'A2': ['B1F', 'B2F'],
    'C1': ['sweet', 'sour'],
    'C2': ['dip', 'spread']
})

Method 1: Simple Loop for Explicit Column Pairs #

If your columns follow a strict A1/A2/.../A20 and C1/C2/.../C20 naming pattern, this loop will directly pair each A column with its corresponding C column, then create a new combined column:

total_pairs = 20  # Match this to your actual number of column pairs

for i in range(1, total_pairs + 1):
    # Define the current pair of columns
    a_col = f"A{i}"
    c_col = f"C{i}"
    # Name your new column (customize this to your preference)
    new_col_name = f"Combined_{i}"
    # Merge the values with an underscore (like your example: B1U_sweet)
    df[new_col_name] = df[a_col] + "_" + df[c_col]

Method 2: Flexible Pairing for Non-Standard Column Names #

If your column names don't follow a strict numbering pattern (but still group into A and C categories), you can filter and sort the columns to pair them automatically:

# Get all A-starting columns, sorted to maintain order
a_columns = sorted([col for col in df.columns if col.startswith('A')])
# Get all C-starting columns, sorted to match A column order
c_columns = sorted([col for col in df.columns if col.startswith('C')])

# Iterate through each paired column set
for a_col, c_col in zip(a_columns, c_columns):
    # Create a descriptive new column name (or use a custom pattern)
    new_col_name = f"{a_col}_{c_col}"
    # Combine the values
    df[new_col_name] = df[a_col] + "_" + df[c_col]

Method 3: Concise List Comprehension (Advanced) #

For a more concise approach, you can generate all combined columns at once using a list comprehension, then add them to your DataFrame:

total_pairs = 20
# Generate all combined columns
combined_columns = [df[f"A{i}"] + "_" + df[f"C{i}"] for i in range(1, total_pairs + 1)]
# Add the new columns to the original DataFrame
df = pd.concat([df] + combined_columns, axis=1)
# Rename the new columns (optional but recommended)
df.columns = list(df.columns[:-total_pairs]) + [f"Combined_{i}" for i in range(1, total_pairs + 1)]

All these methods will produce exactly what you need: new columns where each value is the concatenation of the corresponding A and C column values (e.g., B1U_sweet, B2F_dip). No more manual merging for each pair!

内容的提问来源于stack exchange，提问作者melbez