头歌实训--Pandas合并数据集--第3关：案例：美国各州的统计数据

任务描述
本关为练习关卡，请按照编程要求完成任务，获取美国各州2010年的人口密度排名。

import pandas as pd
import numpy as np

def task3():
    #********** Begin **********#
    #读取三个csv文件
    pop = pd.DataFrame(pd.read_csv("./step3/state-population.csv"))
    ares = pd.DataFrame(pd.read_csv("./step3/state-areas.csv"))
    abbrevs = pd.DataFrame(pd.read_csv("./step3/state-abbrevs.csv"))
    # 合并pop和abbrevs并删除重复列
    df1 = pd.merge(pop, abbrevs, how='outer', left_on='state/region', right_on='abbreviation')
    df1 = df1.drop('abbreviation', axis=1)
    # 填充对应的全称

    df1.loc[df1['state/region'] == 'PR', 'state'] = 'Puerto Rico'
    df1.loc[df1['state/region'] == 'USA', 'state'] = 'United States'

    # 合并面积数据
    df1 = pd.merge(df1, ares, on='state', how='left')

    # 删掉这些缺失值
    df1 = df1.dropna()

    # 取year为2010年的数据，并将索引设为state列
    df1 = df1.loc[df1['year'] == 2010]

    df1.set_index('state')

    # 计算人口密度
    df1['population'] = df1['population'] / df1['area (sq. mi)']

    # 对密度求和

    a = df1.loc[df1['ages'] == 'under18']['population']
    b = df1.loc[df1['ages'] == 'total']['population']
    all = a.values + b.values

    all = pd.DataFrame(all, index=df1.loc[df1['ages'] == 'under18']['state'])

    # 对值进行排序

    all = all.sort_values(0, ascending=False)
    # 输出人口密度前5名和倒数5名
    print("前5名：")
    front = str(all.iloc[:5, 0])[:-24]
    print(front)
    print('dtype: float64')
    print("后5名：")
    back = str(all.iloc[-5:, 0])[:-24]
    print(back)
    print('dtype: float64')
    # ********** End **********#