Scribd
Upload
8 views

Decision Tree

Decision tree

Uploaded by

2023aa05902
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
8 views

Decision Tree

Decision tree

Uploaded by

2023aa05902
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
You are on page 1/ 9
sav eae import pandas 35 aypd Amport nunsy 25 nye from sklearn.tree import DectsionTreeclassi fen from sklearn nodel_selection dnport train test split from sklearn.netrics import confusion nacrix from sklearninctrics dnpert accuracy_score from sklearn.netrics import classification report import natplotisb.ayplot as ayplot from sklearn import tree from skiearn.ensenole Inport RandenForest¢iassifter ssydatannypd.read_csv("C:/Users/hp/Oounoads /data csv") sydata, info) Rangelndex: S69 entries, @ to S6B Data colunns (total 33 colums) clue Hon-mull Count type ie 59 nonnull inte agnosis Se9 nonsnll object, radius nean Ses on-ruldFlstea ectre nean Sco non-null Floated perinetar-nean 565 non-null floated brea aes 5s non-null Floated Snoothness_nean 565 non-null floated Comoctnessinean ——-SeD non-null Floated concavity. naan 565 non-null floats Concave polnts_tean—SeD non-null Floated symetryaean 565 non-null floats Stactaldnension_pean 569 non-null floated radius. se 565 non-null floates texture se 565 non-null floats perinetarse 565 non-null floated tease 565 non-null floates snocthness_se Se5 non-nail Floatet Conpactnesi se 565 non-null flootet toneaity 58 Fister floates Syme se ‘Homes savnaa. sae a 2 2 2 23 28 2 28 2 Fr a 32 evr: fractal_dinension_se rodiue_poret texture worst perineter worst area_worst Enoathness worst compactness uorst concavity worst concave paints worst symetry worst fractal dinension worst Unnared: 32 memory usage: 146.84 XB sydata.deop("id", syaata. info) 569 369 569 569 569 569 569 569 569 569 569 non-null fon-nuld ron-nuld fon-nuld ron-rull fon-nuld non-null fon-nuld ron-nuld fon-nuld oa-ruld @ non-null floate4(31), intsa(2), ebject(2) "uonaned: 32"), axiseL, inpla Rangetndex: loatsa floatet Floatsa Floats Floatsa ‘floatet floats floated floats #loatet Hloatss Floats Non-Mull Count type 569 entries, @ to 568 Data coluns (total 31 colums): elu diagnosis radius_nean texture ean perineter_nean snoothness. ean compactness mean concavity. nen concave paints mean symetry scan ‘ractal_dinension mean redius_ se texture_se perinetér_se snoathness_ se compactness se concavity se concave points se 569 569 569 569 569 569 569 569 569 569 559 569 569 569 559 569 569 569 569 on-nul ron-nuld fon-nuld ron-ruld fon-null ogra fon-nul fon-rall mon-nul ron-rull fon-rull fon-nul fon-null fon-null fon-rull fon-rull roa-rull ron-ruld fon-nuld ail (sig ain on RATE [gst object Floatsa Floats Floatsa Floats ‘loatoa floated loatoa floated Floated Aloatss Floatea Floated Floated Aloatss Floated Aoatss Floatsa floatet Floated floatet sav eae con Ye 21 radius worst 569 non-null fleatsa 22 texture worst 569 non-null floated 23 perineter worst 569 non-null fleatsa 24 breaworst 5369 non-null fleatsa 25 Snoothness worst 569 non-null fleatsa 25 compactness worst 569 non-null floated 27 concavity worst 568 non-null floats 2B concave paints worst 569 non-null floatsd 29 symetry worst 568 non-null floats 5 fractal. dinension worst $69 non-null floatéd types: #loatea(se), object(1) memory usage: 137.94 KB 4 Separating the target vartable X = mydata.values[:, 2] Y= mydatavalues{:, &] Ktrain, xzest,¥ train, Y_test = train sest_split(x, ¥, test_size Yirainisie 455 4 Function t0 perform training with giniindex. fee train using gial(%,¥) ¥ creating the clossifter object ‘clf_gind = DecistontreeClassifier(criterson ni" randon_state = 100,max dept 4 performing training ce gin EK, return clf_gink # Function to perform training with entropy. def Carin using. entropy (X,Y) 1 bectston tree with entropy cLfLentropy * Decistontreeclassifier(criterion » "entropy", randon_state = 18¢ymax_depth = 28, nin_sanples_lsaf = 3) 4 performing training eLf_entropy- 4806) _eading Nat afaoutg/ConmontTMl fone ont sav eae # Function to perform training with random forest classifier, def tarin_using efe(A¥yn est) rfc = RanconForestClassirien(n_estinatorsen_est, randon_stater fee fitOK ¥) return rf 1+ Function to make predictions det prediction(x test, clf object) # Predicton on test Yyapred = elf object.predict(X test) Dprint( "Predicted vaiues:") print(y pred) return y_pred det cal_aceuracy(y_test, y_pred) prlat("confusian Matcie: "confusion matebx(y_test, y_preé)) print ("Accuracy : ",accunacy_score(y.test,y- pred) +108) Drint("Repart : ",classifieation, renert(y_ test, y-pred)) ‘ranyaatacoluans.values ene['h, 8") fet gen tree(ci#) Fig, sx08 » yplot.subplots(nrous = Lyncols = 2, flgsize = (747), spie500) ‘Eree.plot_tree(elf, feature nanes ~ fn,class_sanee-cn, filled ~ true) nypiot. show) ‘Fipssaveftgt"snagenane.prg') cle gin » erotn_using pint (X train, V_tratn) Prediction using gint print("Results Using Gini Tadex:") Yopred_gint = predietion(X test, elf_gint) cal_aceuracy(¥_test, y_pred_gini) fenctrea(el ins) _eading Mat liopoiConmeniTM foneiTRonests Predicted values c 3° tm ae me ae aoe 18 Ww Confusion Matrix: (433]] Accuracy Report: 8 " accuracy cro ave weighted avg. oe ee % Bas a m] (1683) 93.85964912280701 precision 0.96 ean 0.93 2.94 recall _eading ataaoupuiConmeniTMt oneness aa ane mon OW an Fi-seore support n 8 aa ua ua clt_entropy = tarin_using_entropy( train, ¥_train) # prediction using entropy print(*Results Using tntropy:*) Yuprea_entropy = preaiction(X test, elf_entropy) al sccuracy(_test, y_pres_entrony) enstree(elt_entrony) Results Using Entropy Predicted values (at kw a ae on ow sow oa! ON Ee woe ae a eee aN aN om ~ a ~ a ~ Confusion Matrix: [178 2] T3491] Accuracy : 96.49122807017544 Report: ‘precision recall. fi-score support 8 0.96 0990.97 n Meme 083 019s a accuracy 0.96 macro avg 0.97 0.96 0.96 td weighted avg 0.970.950.9518 _eading Mat aop ConmanITM faneiTRonsts clé_rfe = tarin_using efc(X train, Y_train, 28) 4 Prediction using entropy prine(Results Using Fntrony:*) Ypres rfe = prediction(x test, clf_rfe) cal aceuracy(Y_test, y predr#e) Results Using Entropy: Predicted values (at nw a at ow Bee ee ee eae See see WB Bt Ow) Confusion Matrix: [[78 1] T3491] Accuracy : 96.49122807017544 Report precision recall. fl-score 8 0.96 0990.97 n M098 0.93 8.95 8 aa support sav eae con Ye macroyg 8.57, 096 us weighted avg 0.97. 4 view the feature scores colefois"] feature scorescrypd.Sentes(clf_rfc. Feature teportances.index=col).sort_values(ascendingsFalse) Feature scores, area_se 0.182502 concave points_nean 2.169849, Concave points.worst 0.112208 radius worst 000198 concavity-nean 0.083746, perineter_worst 074382 Fadlus_se 0.045875 area_vorst 0.033503, perineter_mean 0.022796, kexture worst 0.022662 radius. ean 0.016896, fractal dinension worst 0.016442 fractal_dimension_ce 0.013438 symetry worst 0.012875 concavity worst ecena3e conpactness_se 2.011388 farea.nean 010982 texture mean 2.010007 penineter-se 0009439, compactness worst 2.006927 swoothness se 2.006619 ‘fractal dinension 2.005590 texture se 2085804 symetry_nean 2.002044 snoothness.nean 0.003662 symetry £2 2.003557, concavity. se 2.003287 conpactness_nean 2002886, concave poiats se 0.002658, snoothness worst 0.002522 type: floates _eading at anoupuiConmeniTMl faneiTX ont

You might also like