如何对DataFrame每行执行条件运算？求解行级if语句正确实现方式

Hey there! Let's get that DataFrame logic working properly. The issue with your current code comes from two main things: using and instead of pandas-compatible boolean operators, and trying to apply a standard Python if statement to an entire Series instead of leveraging pandas' vectorized operations (which are way faster and cleaner).

Let's break down the correct approaches: #

1. Recommended: Use numpy.where (Vectorized, Fast)

This is the best method for most cases, especially with larger datasets, since it operates on the entire Series at once instead of looping row-by-row.

import numpy as np
import pandas as pd

# Assuming your DataFrame is already defined
df['720-1080'] = np.where(
    (df['position'] > 720) & (df['position'] < 1080),  # The condition
    df['position'] - 360,  # Value if condition is True
    np.nan  # Value if condition is False (replace with 0/df['position'] if needed)
)

Note: We use & instead of and here because we're working with boolean Series—and expects single boolean values, not arrays of booleans.

2. Use pandas.loc for Explicit Indexing

If you prefer a more step-by-step approach, you can create a boolean mask and update only the rows that meet your condition:

# Initialize the new column with NaN (or another default value)
df['720-1080'] = np.nan

# Create a mask for rows where position is between 720 and 1080
position_mask = (df['position'] > 720) & (df['position'] < 1080)

# Update only the rows that match the mask
df.loc[position_mask, '720-1080'] = df.loc[position_mask, 'position'] - 360

3. Row-wise apply (Not Recommended for Large Data)

If you really need to use a row-by-row if statement (this is slower for big DataFrames), you can use apply with a custom function:

def adjust_position(row):
    if 720 < row['position'] < 1080:
        return row['position'] - 360
    else:
        return np.nan  # Replace with your preferred default value

df['720-1080'] = df.apply(adjust_position, axis=1)

Remember: axis=1 tells pandas to apply the function to each row instead of each column.

Why Your Original Code Didn't Work: #

• You used and instead of &—pandas requires & for element-wise boolean operations on Series.
• The if row in (...) syntax doesn't work here because df['position']>720 and df['position']<1080 returns a boolean Series, not a list of rows to check against.

内容的提问来源于stack exchange，提问作者1lk4