头歌 6. Pandas合并数据集

第1关：Concat与Append操作

import pandas as pd

def task1():
    #********** Begin **********#
    df1 = pd.read_csv("step1/data.csv")
    df2 = pd.read_csv("step1/data1.csv")
    result = pd.concat([df1, df2], axis=1).set_index('Ladder').fillna(0)
    
    #********** End **********#
    return result

第2关：合并与连接

import pandas as pd


def task2(dataset1,dataset2,dataset3):

    # ********** Begin **********#
    df1 = pd.DataFrame(dataset1)
    df2 = pd.DataFrame(dataset2)
    df3 = pd.DataFrame(dataset3)

    df4 = pd.merge(df1, df2, how='outer')
    df4 = pd.merge(df4, df3, how='outer')
    df4 = df4.fillna(0, axis=1)
    df4['user_id'] = df4['user_id'] + df4['id']
    df4['user_id'] =df4['user_id'].astype('int')
    df4.drop(df4.index[3], inplace=True)
    result = df4.drop('id', axis=1)
    result = result.sort_values(by='user_id', ascending=True)

    result.index = [6, 7, 8, 0, 1, 2, 3, 4, 5]
    
    # ********** End **********#
    return result

第3关：案例：美国各州的统计数据

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 **********#