so/10726, 12:06 PM Unite ipyn -Colab start coding o se with AL Snport pandas as pa # changed np to pa npsrandon 2006(333), age = ap-randon.cansint (22, 48, a) # Sxitched the ones of 10 and 4B to gererate ages between 18 and 42 gender = npcrandon.chotce(( Male", “Tenale"), 0) fdueation ='np.randor.choice({‘high School'," ‘college’, “graduate'}, 9) ob_level = np.rancor.chotce(["Juntor’, "sendor’), ) lnse_evolustion = rp.ranaam-univorn(®.¢, 9%, 7) sverage_sonthly_hours = np-randon-randine(i80, 388, ‘Sne_spend_conpany = np-ranéon.rangine(3, 28, n) nunberof projects = np.randon.candine(2, 7,9) ork azctdent = np.randn.chatce((®, 1], 5) pronation = np.rangar.chaice([®, 1], a) Salary = aperandon.chotce(('Jow, “ediun’, “high*], 9) = pa.vatarranet( “age: 366, ender" gender, ‘edueation' edeeation, ob_lovel": Joh level, ‘overage aonthly_hours'? average monthly, hours, "ne_spend_conpany": tine_spend_conpany, umber of projects": nunber_of projects, ‘work accldent': work sccsaent, "salary': salary » a¢-se.csu¢ ploy fata.cw", tndewstalze) Snport seatorn 25 srs nport aatplotlib.pyplot as pit ‘rom skiearn-nodelnselection snport train test split fron skiearn Linear podel Inport Logistickegression fron skiearn-preprocessing Inport LabelEncoder ‘from sklearnonetries Inport classitication_report, confuston_patrsx dats = pé.reae_csvt" employee data. csv') data.neaet) ‘age gender education job level last_evaluation average sonthly hours tine 073 Male calloge——sonor oar9sss 100 112 Male cologe junior ossrre3 oer 2-28 Female graduate junior ossrsas 2 312 Female college senior oarasss oy 416 Fomale graduate sone osa646: 2a data. snut2()- suet) Intps:ilcolab research google. coridrvee?v8ZogsNiUbarVViLATaQnKFZ_BiiKifprintMode 19so/10726, 12:06 PM Unite ipynb -Colab age o gender o {ob oval o last_evaluton 0 average monthly hours 0 time_spend_company 0 ‘umber_of projects 0 work accident 0 promotion oo salary ° type it ae = Laetencoderd) Gatat[" gender", ‘education’, “Joe level’, “salary')] = datal(*gender', ‘education’, prantidaranesaQ)) age gender education jad tevel Last evaluation average_sonthiy_ hours DR 8 i a 8.472353 ue ‘ine_spend_conany nunber_of_projects work accident pronotion salary 4 ° a @ fd X = datedropt‘pronetion’, axise1) Y = aatal ‘promotion’ ) \ joe level Kitrain, x test, y_train, y_test = train test split(t, ¥, test size-0.3, random statens4) print(X train. shape, X test-shape, y_trainshape, y_test.shape) (700, 10) (280, 28) (722,) (200,) nodel = LogtstteRegression() nodel F804 train, y-train) B+ togisescregression Logistictegression() print(nodel.coet_) Drinttnogelsnserceps_) a3ii62 007689883 29729172 0.02595137)}) (e.0:360362} Yopred «= model predict (x test) yetest.hesd() Intps:ilcolab research google.coridrve1e?v8ZogsNiUbarVViLATaQnKFZ_éiiKifprintMode salary’) apoly(le.#it_transtora)United. ipynb ~Colab so/10726, 12:06 PM romotion 0 on a0 120 type eportty_test, Ypred)) print etassitication weighted ave 00025951379] yopred = model predict(X.test) yitest.nead() g t 20 on a0 20 print y_pred) a 3 z = z 5 8 5 i i 3 g : 4 $ 3 8 3 i Esort0rzs, 1208 Pat Unite ipynb -Colab print(classification report(y_test,y_oree)) Gn = confurion-astrix(y.testJ.9768)| 3x = pit sapploe0) ns heetang ea, aronteTru, fats’! samen = (‘Predicted Labels") ™ je" True Labels") aucset_title( contusion matete’) aucsands.set_tieelabels( ("Not Promoted", "Promoted axcyands. set tieklabels("Not Promoted", "Pronoted]) meron MET eae os s wit me 8a ta i Confusion Matrix 88 z 95 z 5. oe 5 3 95 i Leo : “ 6 70 i “0 wot romated srometed : Predicted Labels Snport matplotlib.pyplot as plt Xr datal{ age’, “gender', ‘education’, “Job level", ‘last_evaluation', ‘average sonthly. hours", ‘tine spend company", ‘nunber_of projects") Y = ¢atal ‘promotion’ ) Katrain, Xtest, y_train, y_test = train test spLit(K, ¥, test sizer8.3, random staterse) print(x_train.shape, Xtest-shape, y_trainshape, y_test.shape) nodel = LogisticRegression() nodel #8804 train, y-teain) Nipred’ = node preait(K test) print classification report(y_test, Y.peed)) fn = confusion natrix(y-test,¥_2re@) a= plt-simploe) ‘ns, naataap(en,annot-Toue,fnt~'g"2X-at,cHBp- 61025") auc set lapel Predicted Labels") ax.set_ylasei("True Labs") Sec set tiela(Cantusion matrse’) Su sad. set_tietlabels ("Not Promoted Sucyoxs set tleelabels ("Rat Promoted”, pit. shou() Promoted" }) Promoted") Intps:ilcolab research google. coridrvee?v8ZogsNiUbarVViLATaQnKFZ_BiiKifprintModesort0rzs, 1208 Pat (700, 8) (380, 8) (709,) (29,) Unite ipynb -Colab precision "recall #i-score support 1 eu 59 9.52 135 wittd me 8 tat Confusion Matrix 20 z z os 5 é 0 i Ln b 8 0s 10 wot rometed rrometed - Predicted Labele nport natplotlib.pyplot 95 plt X's data.drop( ‘pronation’, 3835-1) Y = datat 'pramotion’) nodel = Lagistienegression() fodel.#8¢(% train, y_teata) ipred’ = wodel.preatt(x test) print(elassi¢ieation.report(y test, Yared) en = confusion_natrix(y_test,¥_2re@) 3x = ple rapploe0) {rss heataap(enannot=T-ue fats" g" xen, cnape blues") aucset_daen( ‘Predicted Labels") sxcset_ylanel ("True Labels") Sucset title ‘contusion matese’) axcxands.set_ticelabels({ Not Promoted", "Promoted" aucyands.set_tieelabels( (‘Not Promoted", Pronated}) prt. shou() Intps:ilcolab research google. coridrvee?v8ZogsNiUbarVViLATaQnKFZ_BiiKifprintMode seen, Y, kest_siz000.2, randon_seatens4)so/10726, 12:06 PM Unite ipynb -Colab precision recall f2-sc0re support wihted we 8a baat Confusion Matrix 88 z 95 z 5. oe 5 3 95 i Leo : “ 6 70 i “0 wot romated srometed : Predicted Labels ta{L'age", "gender", "education", “jos tevel"]} ta("pronetion") Ktrain, Xtest, y_train, y test = train test split, ¥, test sizes8.3, random statensd) rint(X train. shape, X test-shave, y_train.shape, y_test-stape) nodel = Logisttcregression() nodel F804 train, y-teain) “Epred = node preaict(x cert) print classification report(y_test, Y.pred)) fm = contusion_natrix(y_test,¥_2re@) ax = put sumplaet) ‘ns, eataap (en annot=True,fat='g"axea4,cma9-" blues") faxsset_xdavel( ‘Predicted Labels") axcset_ylabel(‘True Labels") axcset_title( ‘contusion matrix’) ductors. ret_ttetlabels("hot Promoted" "Promoted" 1) Suiyanis.set_tietlabels( {hot Promoted", "Pronated”) pit shou() Intps:ilcolab research google.coridrve1e?v8ZogsNiUbarVViLATaQnKFZ_éiiKifprintModeso/10726, 12:06 PM Unite ipynb -Colab (700, 4) (380, 4) (700,) (38,) precision” © recall #-score support 1 eas 62 est 15, come 0. wind eat ‘Confusion Matrix z 100 z 5. “ é %0 3 0 & : - a wot romated romoted Predicted Labels ‘port sepsis np a8 Xenia, “genders aueaion', “Joplevel’, “astevaletion', “average tonthiypoors 1 dati promotion print (X_train.shape, XLtest-shave, y_train.s ABEL, Y, test sizest.2, random states24) 1 yatest shape) rnodel = Lopisticregression() odel.F1t04 train, y_teain) pred = nodel.precict(x test) print(elassification_roport(y_sest, Yared) Gn © confurion_aatrix(y.test,¥_are@) 3x = plt-sabploe() snasheataag(enarnoteTeuefats'g!atear enap aucset_dasen( ‘Predicted Labels") fax.set_ylabel( "True Labels") aucset_title( ‘contusion matetx’) axcxands.set_ticelabels( "Not Promoted", "Promoted axcyanis.sef_tictlabels({'Not Promoted" "Promoted" pit. shout) Intps:ilcolab research google.coridrve1e?v8ZogsNiUbarVViLATaQnKFZ_éiiKifprintMode 19so/10726, 12:06 PM Unite ipynb -Colab (700, 6) (380, 6) (709,) (229,) precision recall #-score support cro ave 858 weighed me 38 100 z 3 z4 ® n02 0 s g 7 b 2 2 not rooted rome Predicted Labels snport natplotlib.pyplot a5 pit X's datalL ‘age’, “gender’, ‘ecueation', “Joh level’, ‘last_evaluation', ‘average.vonthly_hours", ‘tine_spend_company', ‘nunber_of projects", Y = datat ‘promotion’ 7 préne(X/train.shape, Xtestishape, y_train.st BLO, Y, test_aizest.2, random states24) + ytest shape) rnodel = Lopisticregression() rodel.F2€04 train, y_teain) ‘pred = nodel.precict(x test) print(elnssification_roport(y_sest, Yared) Gm © confurion_aatrix(y,test,¥_are@) ax = pt subplot) snasheataag(en,armoteTeue fats’ g!atear enap aucset_daen( ‘Predicted Labels") fx.set_ylabel( True Labels") ausset title contusion matrix’) axcxands.set_ticelabels( "Not Promoted", "Promoted ax.yanis.sef_sictlabels({'Not Promoted", "Pronated") pit. shou() Intps:ilcolab research google. coridrvee?v8ZogsNiUbarVViLATaQnKFZ_BiiKifprintModeso/10726, 12:06 PM Unite ipynb -Colab (700, 8) (380, 9) (709,) (29,) precision © retail support ‘ase /ocat/Lib/nythond.1/ist-packages/sklearn/Linear_nodel/ Intps:ilcolab research google.coridrve1e?v8ZogsNiUbarVViLATaQnKFZ_éiiKifprintMode