pandas简单的数据筛选，欢迎大家指导

#  coding=utf-8-sig
# ！/usr/bin/python3
# @Time   : 2021.9.22
# @Author : Coly
# @version: V1.0
# @Des    : data analysis.  spark有三大引擎，spark core、sparkSQL、sparkStreaming，
#           spark core 的关键抽象是 SparkContext、RDD；
#           SparkSQL 的关键抽象是 SparkSession、DataFrame；
#           sparkStreaming 的关键抽象是 StreamingContext、DStream

import findspark
findspark.init()
import os
import numpy as np
import pandas as pd

os.environ['JAVA_HOME'] = '/usr/lib/jdk8/jdk1.8.0_301'
os.environ["PYSPARK_PYTHON"] = "/usr/bin/python3"


if __name__ == "__main__":

    data_frame = pd.read_csv("data2")    #local reading file 
    CS_DF = pd.read_csv("cs.csv", delimiter=",",names=['a','b','c'])

    data_frame = data_frame.iloc[:,[0,3,4,5,7,8]]
    data_frame.columns = ["EV_ID", "longitude", "latitude", "date_time", "speed","direction"]
    CS_DF = CS_DF.iloc[:,[0,0,1,1,2]]
    CS_DF.columns = ["cs_lon_min", "cs_lon_max", "cs_lat_min", "cs_lat_max",'range']

    Area_covered = 100000
    CS_DF['cs_lon_min'] = CS_DF['cs_lon_min'] - CS_DF['range']/Area_covered
    CS_DF['cs_lon_max'] = CS_DF['cs_lon_max'] + CS_DF['range']/Area_covered
    CS_DF['cs_lat_min'] = CS_DF['cs_lat_min'] - CS_DF['range']/Area_covered
    CS_DF['cs_lat_max'] = CS_DF['cs_lat_max'] + CS_DF['range']/Area_covered
    print(CS_DF)
    for i in range(np.size(CS_DF, 0)):
        location_data = data_frame.loc[data_frame["longitude"] > CS_DF.loc[i,'cs_lon_min']]
        location_data = location_data.loc[data_frame["longitude"] < CS_DF.loc[i,'cs_lon_max']]
        location_data = location_data.loc[data_frame["latitude"] > CS_DF.loc[i,'cs_lat_min']]
        location_data = location_data.loc[data_frame["latitude"] < CS_DF.loc[i,'cs_lat_max']]
    location_data = data_frame.loc[data_frame["speed"] == 0]
    print(location_data)
    print('The runing is ok!')


    
	

标签：DF,欢迎,cs,location,frame,CS,筛选,data,pandas
来源： https://blog.csdn.net/qq_38833931/article/details/120517718