全Python數(shù)據(jù)分析,工資情況

Python數(shù)據(jù)分析，工資情況以美國城市巴爾的摩2016年公務員的工資數(shù)據(jù)集為例，數(shù)據(jù)來源于網(wǎng)絡，通過數(shù)據(jù)分組統(tǒng)計，并且可視化，分析其工資情況。.數(shù)據(jù)來源importnumpyasnpimportpandasaspdimportmatplotlib.pyplotasplt%matplotlibinline#該數(shù)據(jù)為美國政府公開的公職人員的薪資數(shù)據(jù)salary=pd.read_csv(open('D:\python數(shù)據(jù)分析\數(shù)據(jù)\Baltimore_City_Employee_Salaries_FY2016.csv'))salary.head()其中數(shù)據(jù)集字段Name其中數(shù)據(jù)集字段Name、JobTitle、AgencyID、Agency、HireDate、NginejDbTNfcAgenicylDAgencyHkeDarleAnnualSalaryGrowPay0.AarcmPaUicisGFacllitie^CMfkeServicesIIA03D31OED-Employment□ev(031)1200:00AMS5G70S.0Q$541354J1Aarani.PetraLASSISTANTSTATE'SATTORNEYA29045S1at&sAttameyaQffics(Q4S)口鼻2卬？口配1?:00:00A始$75500.00$72445.672AbbeyErnfflanuelCONTRACTSERVSPECIIA400Q1M-RInfoTechnalon^i(001)O5.?101.Qai315-00:0flAMS￡oosa.ooSE9BD2583.Abbott-ColeMichelleOperaHiNnsOific&rIIIA90005TnAN￡-Trafl:ic(005)11/2^01412:00:QDAM$70000.00$59517.214Abdsl-IRshimNaimAEMTFirefighterSuppre喇口mAM120FireDepartment(120)0^300011000:00AMM436S.D0S74770&2AnnualSalary、GrossPay分別代表的含義為姓名、職位名稱、工號、單位、入職日期、年薪、總薪資。.問題探索年薪的分布情況公務人員入職日期情況年薪最高的職務人數(shù)最高的職務.數(shù)據(jù)清洗salary.shape(13818,7)本例含有13818條數(shù)據(jù)，7列。salary.isnull().sum()Name0JobTitie0AgencyID0Agency0Hir&Date0AimualSalary0GrossPay272dtype:int64查看缺失值，GrossPay列有272個缺失值salary=salary.dropna()salary.isnull().sum()ffameJobTitieAgencyIDAgencyHireiDateAnnualSalaryGrossFaydtype:int64刪除缺失值后的結(jié)果。dtype('O')salary['AnnualSalary']=salary['AnnualSalary'].astype(float)salary['GrossPay']=salary['GrossPay'].astype(float)salary['GrossPay'].dtypedtype('float64')AnnuallS-alary GrossPay0 567JSO。 5413S.441 75600.OC 72445.372 60060.0(3 59Mz5m3 700D0,DC 5%17.2154365.0(] 74770.82字符串處理，去掉“$”符號，轉(zhuǎn)換為浮點類型。dtype('O')salary['AnnualSalary']=salary['AnnualSalary'].astype(float)salary['GrossPay']=salary['GrossPay'].astype(float)salary['GrossPay'].dtypedtype('float64')salary['month']=salary['HireDate'].str.split('/').str[0]salary[['HireDate','month']].head()轉(zhuǎn)換數(shù)據(jù)類型為浮點類型。salary['month']=salary['HireDate'].str.split('/').str[0]salary[['HireDate','month']].head()HineDatenwiithTOC\o"1-5"\h\z0 1W24/1975 12:OG:COAM 1。1 093笈口儂 12:OC:COAM 092 05/C1^013 12:OC:-3OAM 053 11/28/2014 12:0C:C0AM 114 03/30^011 12:00:-30AM 03對于入職日期，新增一列，存放入職月份，把日期數(shù)據(jù)當作字符串來處理4.數(shù)據(jù)探索salary['AnnualSalary'].hist(bins=20)做出年薪工資的直方圖可以發(fā)現(xiàn)，年薪基本呈正態(tài)分布，但向左略有傾斜，說明高工資的職務還是較少的。month_count.plot(kind='barh')

0614680812830912270711981011851110990110971210350510230310180496702946Name:montl%dtype:int64monthcount.plot(kind='barh')agg_salary=salary.groupby('JobTitle')['AnnualSalary'].agg(['mean','count'])agg_salary#聚合運算，計算各職位的年薪平均值和職位個數(shù)m&ancountm&ancountsort_salarysort_salary['mean'].plot(kind='bar')m&ancountm&ancountsort_salarysort_salary['mean'].plot(kind='bar')JobTitle911LEADOPERATOR43616.7500004911OPERATOR44&29.4615J8骷911OPERATORSUPIERVISOR572035M0004ACCOUNTEXECUTIVE57200OCN0C04ACCOUNTANT1490&5.36&66715ACCOUNTANTII5G172.64900026ACCOUNTANTSUPV&7417.1420571ACCOUNTANTTRAINEE366B1OOOOCO1ACCOUNTINGASST129226.33J3336ACCOUNTINGASSTII34281.5^33315ACCOUNTINGASSTIII43K37.S1&18233ACCOUNTINGMANAGER781555000002ACCOUNTINGOPERATIONSOFFICER7&100ODOOOO1ACCOUNTINGSYSTEMSADMINISTRAT924000900003ACCOUNTINGSYSTEMSANALYSTt3S57142657214DMCOORDINATOR46005.50000C2ADMINISTRATIVEAIDE,SHERIFF13212.6JE-3&411ADMINISTRATIVEANALYST143Gs7MM008ADMINISTRATIVEANALYSTII58S33333J333ADMINISTRATIVECOORDINATOR&0036.50000C1CADMINISTRATIVEPOLICYANALYST77MJOODOOOO1ALCOHOLASSESSMENTCOUNSELOR13S7310000001ALCOHOLASSESSMENTDIRECTORCO64M00000001ALCOHOLASSESSMTCOUNSELORII3S331OO00001ALCOHOLASSESSM1COUNSELORIII39701OQOOCO1sortcountsortcountJobTitlemeancountSTATE'SATTORNEY23877201PoliceCommissioner2DG00001ExecutIveDirectorV182500.01MAYOR17163601ExecutiveDirectorIII1713065對年薪平均值降序排列，并取前5。sort_salary['count'].plot(kind='bar')cdoJWW■=」cdoJWW■=」a?J5wnEML&JWbTi小做出柱狀圖可以發(fā)現(xiàn)，STATE'SATTORNEY(州檢查官)的年薪最高。sort_count=aggsalary.sortvalues(by='count',ascending=False)[:5]

JobTitlemeanCOUIHPOLICEOFFICER703969738041756LA日。RER(Hourly)31274.6185