Python數(shù)據(jù)分析基礎(chǔ)與應(yīng)用電子活頁4-13實(shí)現(xiàn)Pandas的排序操作

Python數(shù)據(jù)分析基礎(chǔ)與應(yīng)用模塊電子活頁4-13實(shí)現(xiàn)pandas的排序操作【技能訓(xùn)練4-17】實(shí)現(xiàn)pandas的排序操作【訓(xùn)練要求】在JupyterNotebook開發(fā)環(huán)境中創(chuàng)建j4-17.ipynb，然后編寫代碼實(shí)現(xiàn)pandas的排序操作?！緦?shí)施過程】1．使用sort_index()函數(shù)按索引標(biāo)簽進(jìn)行排序（1）創(chuàng)建指定行索引的DataFrame對(duì)象代碼如下：importpandasaspddata={'date':pd.Series(['2022/1/1','2022/1/2','2022/1/3','2022/1/4','2022/1/5']),'highT':pd.Series([12,15,12,12,7]),'lowT':pd.Series([1,4,8,6,5]),'AQI':pd.Series([167,145,123,212,104])}df1=pd.DataFrame(data,index=[4,1,2,0,3])print(df1)輸出結(jié)果：datehighTlowTAQI42022/1/57510412022/1/215414522022/1/312812302022/1/112116732022/1/4126212（2）按行索引順序排列代碼如下：df2=df1.sort_index()print(df2)輸出結(jié)果：datehighTlowTAQI02022/1/112116712022/1/215414522022/1/312812332022/1/412621242022/1/575104（3）通過ascending參數(shù)控制排序方式代碼如下：df3=df1.sort_index(ascending=False)print(df3)輸出結(jié)果：datehighTlowTAQI42022/1/57510432022/1/412621222022/1/312812312022/1/215414502022/1/1121167（4）按列索引順序排列代碼如下：df4=df1.sort_index(axis=1)print(df4)輸出結(jié)果：AQIdatehighTlowT41042022/1/57511452022/1/215421232022/1/312801672022/1/112132122022/1/41262．使用sort_values()函數(shù)按數(shù)值進(jìn)行排序（1）按1列的數(shù)值進(jìn)行排序代碼如下：df5=df1.sort_values(by='highT')#輸出結(jié)果與df1.sort_values(by='highT',axis=0)的相同print(df5)輸出結(jié)果：datehighTlowTAQI42022/1/57510422022/1/312812302022/1/112116732022/1/412621212022/1/2154145【注意】當(dāng)按highT列排序時(shí)，相應(yīng)的其他列（date、lowT、AQI）的元素值和行索引也會(huì)隨highT列一起改變。（2）按多列數(shù)值的升序進(jìn)行排列代碼如下：#先按highT列升序排列，highT列數(shù)值相同的按lowT列升序排列df6=df1.sort_values(by=['highT','lowT'])print(df6)輸出結(jié)果：datehighTlowTAQI42022/1/57510402022/1/112116732022/1/412621222022/1/312812312022/1/2154145（3）指定不同的排序方式對(duì)多列數(shù)值進(jìn)行排序代碼如下：#先按highT列降序排列，highT列數(shù)值相同的按lowT列升序排列df7=df1.sort_values(by=['highT','lowT'],ascending=[False,True])print(df7)輸出結(jié)果：datehighTlowTAQI12022/1/215414502022/1/112116732022/1/412621222022/1/312812342022/1/575104（4）指定排序算法代碼如下：df8=df1.sort_values(by='highT',kind='mergesort')print(df8)輸出結(jié)果：datehighTlowTAQI42022/1/57510422022/1/312812302022/1/112116732022/1/412621212022/1/2154145（5）按1行的數(shù)值進(jìn)行排序代碼如下：df9=pd.DataFrame({'b':[1,2,3,4],'a':[4,3,2,1],'c':[5,3,8,6]},inde