需求：两张含经纬度的表A,B，需要找到距离A表内各点2km范围的B表的经纬度清单，如A表中point1，找到B表中距离point1 2km范围内的点
遇到的问题：处理时间太长，现在A表有10万、B表有25万。我处理一遍得17分钟，希望能缩短处理时间
编程能力：只会简单的python

如下是我丑陋的代码：

import pandas as pd
import numpy as np
from datetime import datetime
from math import radians, cos, sin, asin, sqrt

df5 = pd.read_csv('cell.csv',encoding='gbk')
df4 = pd.read_csv('cell_all.csv',encoding='gbk')

def fun3(df5,df4,group=20,interval=0.15,dis=2000):
    result1 = pd.DataFrame()
    
    df4 = df4.sort_values('Longitude')
    df5 = df5.sort_values('Longitude')
    df4['t'] = 1
    df5['t'] = 1
    
    n = 0
    nmax = df5.shape[0]
    for i in range(nmax//group+1):
        start = n
        end = n+group
        if end > nmax:
            end = nmax
        df51 = df5.iloc[start:end]

        df4_lon_max = df51.Longitude.max()+0.15
        df4_lon_min = df51.Longitude.min()-0.15
        df4_lat_max = df51.Latitude.max()+0.15
        df4_lat_min = df51.Latitude.min()-0.15 

        df41 = df4[(df4['Longitude']>df4_lon_min) & (df4['Longitude']<df4_lon_max) & (df4['Latitude']<df4_lat_max) & (df4['Latitude']>df4_lat_min)]
        df45 = pd.merge(df51,df41,how='left',on='t')
        df_array = df45.values
        df_np = np.insert(df_array, -1, values=list(map(lambda x: haversine(x[1], x[2], x[5], x[6]), df_array)), axis=1)
        df = pd.DataFrame(df_np)
        df.columns = ['point_x', 'Longitude_x', 'Latitude_x', 't', 'point_y', 'Longitude_y','dis', 'Latitude_y']

        df = df[(df['dis'] < dis)]
        
        if n==0:
            result1 = df
        else:
            result1 = pd.concat([result1,df])
    return result1

fun3(df5,df4,20,0.15,2000)

Wall time: 16min 45s
1775184 rows × 8 columns

尝试过将group设为10/20/30/40/50，但貌似越大越慢，所以目前就定在20.麻烦大神帮忙指点指点