如何在Pandas DataFrame中基于行列值新增坐标距离计算列？

原始DataFrame #

预期结果 #

实现方案 #

使用Haversine公式计算球面两点间的距离，以下提供两种实现方式：

方式1：逐行计算（直观易读） #

通过apply遍历DataFrame每行，调用距离计算函数并处理NaN值：

import pandas as pd
import numpy as np

def haversine_distance(lat1, lon1, lat2, lon2):
    # 角度转弧度
    lat1_rad = np.radians(lat1)
    lon1_rad = np.radians(lon1)
    lat2_rad = np.radians(lat2)
    lon2_rad = np.radians(lon2)
    
    # Haversine公式核心计算
    dlat = lat2_rad - lat1_rad
    dlon = lon2_rad - lon1_rad
    a = np.sin(dlat/2)**2 + np.cos(lat1_rad) * np.cos(lat2_rad) * np.sin(dlon/2)**2
    c = 2 * np.arcsin(np.sqrt(a))
    return c * 6371  # 6371为地球平均半径（公里）

# 构造目标DataFrame
df = pd.DataFrame({
    'H_Lat': [18.447259, 18.523069, 18.511014],
    'H_Lon': [73.896742, 73.842460, 73.864071],
    'W_Lat': [18.534579, 18.491357, np.nan],
    'W_Lon': [73.819043, 73.851985, np.nan]
})

# 新增Distance列
df['Distance'] = df.apply(
    lambda row: haversine_distance(row['H_Lat'], row['H_Lon'], row['W_Lat'], row['W_Lon'])
    if pd.notna(row['W_Lat']) and pd.notna(row['W_Lon']) else np.nan,
    axis=1
)

方式2：向量化运算（高效推荐） #

利用numpy向量化操作批量计算整列数据，避免逐行循环，处理大数据量时效率更高，且自动兼容NaN值：

import pandas as pd
import numpy as np

# 构造目标DataFrame
df = pd.DataFrame({
    'H_Lat': [18.447259, 18.523069, 18.511014],
    'H_Lon': [73.896742, 73.842460, 73.864071],
    'W_Lat': [18.534579, 18.491357, np.nan],
    'W_Lon': [73.819043, 73.851985, np.nan]
})

# 批量转换为弧度
lat1_rad = np.radians(df['H_Lat'])
lon1_rad = np.radians(df['H_Lon'])
lat2_rad = np.radians(df['W_Lat'])
lon2_rad = np.radians(df['W_Lon'])

# 批量计算距离
dlat = lat2_rad - lat1_rad
dlon = lon2_rad - lon1_rad
a = np.sin(dlat/2)**2 + np.cos(lat1_rad) * np.cos(lat2_rad) * np.sin(dlon/2)**2
c = 2 * np.arcsin(np.sqrt(a))

# 赋值给Distance列
df['Distance'] = c * 6371

两种方式运行后，均可得到与预期一致的结果，第三行因缺失坐标值，Distance自动为NaN。

内容的提问来源于stack exchange，提问作者Bhavesh Neekhra