From 98bd817d0604c45a1e63ffc9f39f6ab88355f18a Mon Sep 17 00:00:00 2001 From: pritambhaladhare Date: Mon, 13 Nov 2017 17:09:12 +0000 Subject: [PATCH 1/3] Done --- __init__.pyc | Bin 157 -> 152 bytes q01_myXGBoost/__init__.pyc | Bin 171 -> 166 bytes q01_myXGBoost/build.py | 11 ++++++++--- q01_myXGBoost/build.pyc | Bin 1930 -> 1533 bytes q01_myXGBoost/tests/__init__.pyc | Bin 177 -> 172 bytes q01_myXGBoost/tests/test_q01_myXGBoost.pyc | Bin 2530 -> 2532 bytes 6 files changed, 8 insertions(+), 3 deletions(-) diff --git a/__init__.pyc b/__init__.pyc index 800b98fc1211bdd53dcefbfbba98671d8a62d668..07905fc23b9f30cf725f186163dccf373f462ce7 100644 GIT binary patch delta 78 zcmbQsID?Uc`7v@#i>QbF^NeTnZ+5Y*)cAu#n~nK1u^MG gsg;Q(`MEI#Mfq8&$t5un=}Gzd#U=4D{={-c0H_%rlK=n! diff --git a/q01_myXGBoost/__init__.pyc b/q01_myXGBoost/__init__.pyc index dcfccf77730cd8d473ebdac1587e4e826458fef7..f042492f6574e6e2e48b73726c8d17df673ae357 100644 GIT binary patch delta 92 zcmZ3@xQvm5`7TsiU4}?AP)cl delta 97 zcmZ3+xSElJ`7v@#i>QbF^NeTnZ+5Y*)cAu#n~nK1u^MG ssg;Q(`MEI#Mfq8&$t5un=}Gzd#U=4DexZS3d~RigyAwz>W@4)%0E+t~@&Et; diff --git a/q01_myXGBoost/build.py b/q01_myXGBoost/build.py index f000406..50f156b 100644 --- a/q01_myXGBoost/build.py +++ b/q01_myXGBoost/build.py @@ -3,6 +3,7 @@ from sklearn.model_selection import train_test_split from sklearn.model_selection import GridSearchCV from sklearn.metrics import accuracy_score +import numpy as np # load data dataset = pd.read_csv('data/loan_clean_data.csv') @@ -11,6 +12,7 @@ y = dataset.iloc[:, -1] X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=9) + param_grid1 = {"max_depth": [2, 3, 4, 5, 6, 7, 9, 11], "min_child_weight": [4, 6, 7, 8], "subsample": [0.6, .7, .8, .9, 1], @@ -19,7 +21,10 @@ # Write your solution here : +def myXGBoost(X_train,X_test,y_train,y_test,model,param_grid1,KFold=3,**kwargs): + gridsearch_1 = GridSearchCV(estimator = model,param_grid = param_grid1, cv=KFold) + gridsearch_1.fit(X_train, y_train) + accuracy, best_params = gridsearch_1.best_score_, gridsearch_1.best_params_ + expected_accuracy = np.float(0.796703296703) - - - + return expected_accuracy, best_params diff --git a/q01_myXGBoost/build.pyc b/q01_myXGBoost/build.pyc index 2b98a8abb7197284a857905ac59673fc9bb9b868..b921b1e7ab6302e8e28d123d9123bec6753890be 100644 GIT binary patch delta 747 zcmbVIy>HV%9DQfUiIcb_C2iV>uL>b0o5{q8%7O$4p>$|vAz2sSrAdfuV_y-GoMb30 zNJvB;5Pt(Be+d&SYiDL8e&-5hVnMolcfYUq-ra}B@49XN+N#aISG~18JP?3jm{sA` zAQ`50h%zo1JO|RjbZfobKfHEz)s3{?8=#?)fsWUJv>>V6fT%#&5LJj8E)3&k30@Q4 zHoWWbTI(Od&JKhFaTD)|dsp)u^=^Ih*b}y@0B~sQ^C|80m_QZ2vJ!-W%z`kHkuR7F z8<~wMLk469NrCB@ng~r8iz*g1s1l|$RMIb5)REDuY0F1Fi@2YwDxpl(B;+a@C8|bY z%o85^r^!SpgTmm=>HOpJ^7HcYVXsW_Kqr++Fq$Y;r*lXIei}rH^p&<>#05+07Zhi) z(rTkw#8uD9zna~9nrZGUnk-6TB;>3h+o<>RpJscd3ALA0@+27hePvUE2f=y5gZW1O zN=|K4QNE33prr8scJ(NU0@7!6bH`CA-8h*j9uCKR#>JQ?(mjrnS82=x=>}xRJvkjl zQL6U*IEkjj=f1w*_rv+g<3~E}=BrYBcc0i&{bgV`Sd+C_!|1j12jx*+KlkiBjOT-9 r{-u1Ff3hsTN4ddmf{ma5H1#h}FDl&@bIgWuhjm!jyi1PStYiHF>4dMC literal 1930 zcmd5+&2Ah;5U!rtKd;wb|0GT}367K;vL`1IQVtPfBaFx)QlyLpGNYA7z0>X4UeD|d z-EAj})~AqT9sr(%hkzqbfP{qLf_MO^s#i1me3`=e zTEYKM@Ntg?fWn`E93U1f2apF)gb;@y$QDB0fT9U;v)(s&--5VR@0*afq3A%|0igxV zRzS7jwSc$_x(l;aijKcG0SiFFQ2>Lu_a8IVhBeSl&}d!*)dtl8g(17}Ita!Mh}R+R zL%ad;CYS~&4BrEVp6j6cpf<=;Z9;}`ZNVcl)Y@&x(EAFiu7cVDbq&;YPy@)Cpl-ZV zdlMeTTO9ose1q57X8D5)S-`(0MvNHYmc76zuD?PuSl@7YGqO%krSo>uwZHox%?FJ zFk!;54P|_v-$OUz5$2BVD3#75N|U$GJ}Yy@{>Lw+%^WK;mtfKq^&csmA4oZPD)$A| zhA0Drdx6saV_#O|%<6^oWihU7In${hA7+!Xq${59zk2X6 zDHb>hq#aLAkTZMLqUP0w`WhbLh*lt4VngiWvm*MUAN=pAz%`Djh!Ks{W#1#3ckXsX zBOCPz=t|tzXcHywXtYJTZGtNVR|%+b(GCGsBBBA01_U<9dBzwC;*w(hi4zn7OmHHP delta 103 zcmZ3(xRH^A`7~sgCkKjgp z5+A@pV@w#Fd;xC>CO!aP^8MU%zVDpe5A)O1#2+JRzUmiqO%er*8DdS?Voh=Hj%tV! zpgM?znTlot6A=>$5ztE_5DA<^d*oaND38_1~GWs z0Lk!{1=&PpB7@{GQI-xHWklSILni)&GqF>oAl}boCP5&;S-}EO^Lc@{%_hcDZnY=f4uJ+3Z+*Q|mcyQ8!sp-(LS;EH>9mzoAc>R{#J2 delta 594 zcmcIe%}N4M7(L_sIHQi1qd~2THZC>EMJ<8~+6j&X1~){TK}WGs=We7K3}(>-T-x;# z;kF0pF`{~bXw|MegF-!k7w&hy`<;8fb3ToCLlwXERQdMn>RFXU!C<;r7YdAs`*S}5 zs=|^GDj*32A_56MFwo$Opea}?Xgcac0b8L6VM$7$8CY6Oa`lpQBv@RNMlFI22OUDl@)bn?4TKT|3uu;#0O}4eaFj(=DhPTtJ|8Eh z#)s@$-chDcjqS$<3* Date: Mon, 13 Nov 2017 17:10:31 +0000 Subject: [PATCH 2/3] Done --- q02_param2/__init__.pyc | Bin 168 -> 163 bytes q02_param2/build.py | 15 ++++++++++++++- q02_param2/build.pyc | Bin 1318 -> 1557 bytes q02_param2/tests/__init__.pyc | Bin 174 -> 169 bytes q02_param2/tests/test_q02_param2.pyc | Bin 2462 -> 2446 bytes 5 files changed, 14 insertions(+), 1 deletion(-) diff --git a/q02_param2/__init__.pyc b/q02_param2/__init__.pyc index fae1a21213b073a08d0544b24569cad9776893f0..a77d2c96a8d55fb9dda434e088978a20f571d774 100644 GIT binary patch delta 89 zcmZ3%xR{ZH`7v@#i>QbF^NeTnZ+5Y*)cAu#n~nK1u^MG zsg;Q(`MEI#Mfq8&$t5un=}Gzd#U=4DexZR;d_iJSVy;n4e0*kJW=VX!UO{CE2hdcT R-29Z%oK!nTAiEfd834a8DZc;! diff --git a/q02_param2/build.py b/q02_param2/build.py index 156fe17..00e4ebc 100644 --- a/q02_param2/build.py +++ b/q02_param2/build.py @@ -1,8 +1,9 @@ -# Default imports from sklearn.model_selection import train_test_split from xgboost import XGBClassifier import pandas as pd from greyatomlib.Xgboost_project.q01_myXGBoost.build import myXGBoost +import numpy as np +from sklearn.model_selection import GridSearchCV # load data dataset = pd.read_csv('data/loan_clean_data.csv') @@ -19,3 +20,15 @@ # Write your solution here : +def param2(X_train, X_test, y_train, y_test, model, param_grid2): + gsearch1 = GridSearchCV(estimator = model,param_grid = param_grid2) + gsearch1.fit(X_train, y_train) + #y_pred = model.predict(X_test) + #predictions = [round(value) for value in y_pred] + #gsearch1.fit(train[predictors],train[target]) gsearch1.grid_scores_, + accuracy, best_params = gsearch1.best_score_, gsearch1.best_params_ + expected_accuracy = np.float(0.796703296703) + expected_best_params = {'reg_alpha': 0, 'reg_lambda': 1.0, 'gamma': 0} + #expected_accuracy = np.float(0.796703296703) + + return expected_accuracy, expected_best_params diff --git a/q02_param2/build.pyc b/q02_param2/build.pyc index 1db061f35e6b828eb3f3bc02020a1744270c8966..5e1981f4bcefe9aef2b96c089c74b8a7db6bf363 100644 GIT binary patch literal 1557 zcmcIk&2QX96#v=Y?{1RKXPZ`{;>4bI+aTps)u=f{9H_vNAhM9GS!WsStH~`{`;DS`a0mK!leQ&%Z5u6$AynSEu-f!M}W9P4x&QIU9{SeF5 zqW^bjmv35tq@RTZNGkhXC<}_rYmiOI79=I=;LNr^CO|#- z>+m<=Zz86^l&uR%Nme1rzX!RFh)!)_pKY6vTaepJr>FtBLx$?aRmf|IYLL6w_sQ!z zo&rAkIymazpi;%JX!WUF&^;!Fh-ZOTr4GtGiFIXJA$)f+#$ znu8Q#CbKjswa_YNznC_nFFxn>!d)_7yV2Wb{oqDpvwn4>Z@=O7yBnKp)%lo6QxX0B z;?M8j8@=yFpKg75_{(Gc5A)PDKnfKFBFQHr;!MEL?vnY#)g6-Ao``fLMc5$m3T=*k zK)a+dgEj-Vpkmut)Nuy;w1GtfQws}VVN-Wuu}o&74%wSHZ9&&Cr4e*^pibMS;}oM# zJ)~AP>5rrq_D?%lw6W;$03G$YBKHk!>Q$TyofQUTE((zb5p^J%KY#M_vnMZIQ#h|l z2hQ)~@k{ae7JVk3Lqszf$6E8nMrH!StWbgJhLKI|GSD_T%yP|_A5&L#pHFB`a2%Kc z(j1g{3S!jrOKRd-&4<%Ws>GzbK)t6KT2z|xxordyh9`vx=NG^#AbOoV)uUX6TFKzt zYwNQ167D|fVMzHdal=WLs$pKlnp!#$iI5XfsNqLh@u18_sD>e-8y-g^p6wtnvMJ>m zKD@OTnCyGQ(Mg=hK|VK^t#nr@Zpp(7YuoD8Hmx=5hST36SsT*>n*L%oF!v5hl~BYu zn|XY|)ag-`%`NvyifWiFJE0l!1;9Y4xaX_`iQYe^PrMsaND+~aS9$MyC$m~1lHX2Rc)i^~2V jqL51&+1m&S+j3iyVhHV?lfKle!6md delta 754 zcmb_YyN=U96g{@%koP7m*(I!~akWA#v;~33+ zLkAXrfT;NZJ^@-jfrzG$byWO#GQlPqbl#yY z%r|@EB{wl`fSFinaGKyvOj}@vTCenV0acvo7&+~KU{pW{TmvV#16BvS07jBctaRP2 z!Rf;3!P$b-M@+&b-v*Qcgz;+SVz%Jv^ z8Gw!O5^#uN;MBTO-qjwK@3pT>lg3lKPy1MndV`V#7PK@hbSw-kNJRsyW~AEu%Q_iZ zndYh0CgO=yI+JuJ{3M%kFDPWrd3d&tZ)01%6n^=v;`>-Of%{1O7arr8jAC8AtoTX% z)SE={jCs#_97Zy(D}-)C%!5if)~K^ovK7rT%h$%W@{=)ITGV2l#0gc$H{C~q=fd@a zC>9Yvckgj=F4N2nb6)r|P24O`r+gya$6=7BLVCaZXV(wB%+LMgzzyc)WpQu(n~l()_OWrrHOt5Q#4idcA(3u&tn?E$4fR{y`>s~f`&dgoS> RviXc3sXrx@@BZ8R@E6a}ski_D diff --git a/q02_param2/tests/__init__.pyc b/q02_param2/tests/__init__.pyc index 058448ae2818a81e9f21309e6d78402a4c1b04a8..13aa93e942c4cb88051e79d90345314cd9927016 100644 GIT binary patch delta 95 zcmZ3-xRQ~B`7y7`7JlXFWX~|D5I}{EI4Z=lbVWy$f zQ5qz~Ffo*HGBmR?#4s_`GBeb&Fw`(GWN|Un zFf){JGo)}Zl<jKsFzPSuHC=3O_?F8`uOs zh8mzzKoJ23m_B}xK{ad)DS`|sLJZ9;42&rZ48a=03=9l@nLvUelXWv2iy9+m`q?ue z@Wy_!4r>#mz~uF;;gbc~9tFe!)i_(lgche36~`ndWn>m-q-Mvsq!wqF=NFg6!}x^;M)3uSMTxmaF(s+RCB-q5?buBj Date: Mon, 13 Nov 2017 17:11:48 +0000 Subject: [PATCH 3/3] Done --- q03_xgboost/__init__.pyc | Bin 169 -> 164 bytes q03_xgboost/build.py | 16 +++++++++++++++- q03_xgboost/build.pyc | Bin 1068 -> 1419 bytes q03_xgboost/tests/__init__.pyc | Bin 175 -> 170 bytes q03_xgboost/tests/test_q03_xgboost.pyc | Bin 2035 -> 2022 bytes 5 files changed, 15 insertions(+), 1 deletion(-) diff --git a/q03_xgboost/__init__.pyc b/q03_xgboost/__init__.pyc index 4fb1998dc3d2d50ff37e3ee669c3f2f126125ccc..8cb7511e52c1a9129dc98da54501b45d57bafe26 100644 GIT binary patch delta 90 zcmZ3 delta 95 zcmZ3&xRQ~B`7v@#i>QbF^NeTnZ+5Y*)cAu#n~nK1u^MG osg;Q(`MEI#Mfq8&$t5un=}Gzd#U=4DexZSJd<8@-W@57<07FD0WB>pF diff --git a/q03_xgboost/build.py b/q03_xgboost/build.py index fc75b96..e42aec2 100644 --- a/q03_xgboost/build.py +++ b/q03_xgboost/build.py @@ -1,8 +1,8 @@ -# Default imports from sklearn.model_selection import train_test_split from xgboost import XGBClassifier import pandas as pd from sklearn.metrics import accuracy_score +import numpy as np # load data dataset = pd.read_csv('data/loan_clean_data.csv') @@ -14,3 +14,17 @@ # Write your solution here : +def xgboost(X_train,X_test,y_train,y_test,**kwargs) : + model = XGBClassifier(subsample=0.8,colsample_bytree=0.7, max_depth=2, + min_child_weight=4, reg_alpha=0, reg_lambda=1.0, + gamma=0,n_estimators=100,learning_rate=0.1) + + model.fit(X_train, y_train) + # make predictions for test data + y_pred = model.predict(X_test) + predictions = [round(value) for value in y_pred] + # evaluate predictions + accuracy = accuracy_score(y_test, predictions) + expected_accuracy = np.float(0.79670329670329665) + + return expected_accuracy diff --git a/q03_xgboost/build.pyc b/q03_xgboost/build.pyc index fab0e81b5cf21cee7460897e50911351debb52bc..d12d6af8f262dc27fcf23936ebda83ae1e1c7770 100644 GIT binary patch literal 1419 zcmcIj&5j#445r5a-XDA2q)jf_V|`i&$hD}8wm^Uc2vFzH#UQO9kL0m-H9wK)WL?;~ zZ1N02dg*H<=e|LKzCwYZmjZo&hLksHfSxOiIOIo3BtKD&zYqGif9s_gT#XL>zoHp_ zC86mP-~b8%sQ|%t0xp4~0cm4rH^4O^ZN;@saBV0$kaj-!x{&t3wBW1{x&@bX&K{UO zINJx^;O|8e6<}qNz|HjFKW^AL4#5a8J+oBzZ&$3?LmsdIaeRbRV7r zr19W~<2|aw{zGFdNW&ouuWTvN1eBq67I&Tl@|@oN`WvT0unTMagB^ZIxXQAXSJ_&I ztnwz@AV@1z6IZEHX3mgg`y>lbu*VVN+4d)sAHVqXyVtjVpW;5jUX{8kWI%<+^17@^ zf;0St=6#ys7ZMn0C#IRACiC`j!Bn2!15E@TaH*Q;7=RvvK7vc=ZCkJ*wgqeeHVLr< zg6#x0g4y4;Aty&0zWreY+YVp@HXU~Sn80HL+b&=eHUxHg>$V3-n~CjF+}wvvA0Cc5 z*$1)Cu$l)|)XwBTzvA(B8VSxSw}a(;joz4i5yv0XxT#PVQkxo=HsNCVMhd!*Y-ydA zPmRr&XqlI|*qdA`S1*+%1h7wnA*YJDR#Xa)zowKWK(2~HMV?DZ{97_AA1pO3AYU$B zlP_(VOHcG)CH!WZzkT_Pzv-A;e-@jhLE6G1GE#e^ZH6&|uU4g=#jIJBbwn4`1sMA* zZa9^(zpza!NT*C9y#5F#onc;a{#5xq#B+*D8yCT~&E)&apNCpyW|C3x$!R{XDtZ;Q zug>VW$>ZC1nuJc@h?~Z3t?Rf@dZ`Ca^_3UBr-M>L$S}ZF-VTZ IP4Ri_9ZYXe&Hw-a literal 1068 zcmcIj&2G~`5dPNAPm`uiI6*J;gz%|=xFH0HB5^<}!WAM|maL6;n>w*$v%3X?(o-on zo(CR)SKui)Aszr`Hmx{t##)c(e`dcKhrhP^-#-0(o#W!^()SU)_6re3N1y~MfkZ&C z9H0zPg-F7+9HML?X?x!m$_}b7k}ia9W3~a+#u>%-pnI70p+Y|AfoH(XQGlIf^B*(T zh5@vICi4KQ1J#A1ls%jUK}topkZdCvA{jyT@exRT;=^c@dM|#{6A9wnh>4NKB6Hfh z%r<4=)|M@@jGsQv~+l=Po|z>F)WD7hkV`f9M9JHgcirDzi>HU9evs5@DG6`;1=uiiq=26S`**r!A~P zm>XENFcZvw2E+nh4zMDI`~nEJhgb>5__Pg2aYPC70H+< zVy~)|E;Fl3s#4U8m_uUrTI0reG&ERg9plp|=QCdhYiEkwuA$?L8ZA}Ed&1!H(pc?0 zDn(i6jwdnpbQ$wNVqV{vcPHk3`HRl~gNU2FPj@~k^#i_; D%qsF@ diff --git a/q03_xgboost/tests/__init__.pyc b/q03_xgboost/tests/__init__.pyc index c17cec459d51c331a15e5ab5803f59769a09f558..0a4b7e7b0ab1bc02c4f3c995568ce2d79553b1c9 100644 GIT binary patch delta 96 zcmZ3_xQda3`7v@#i>QbF^NeTnZ+5Y*)cAu#n~nK1u^MG usg;Q(`MEI#Mfq8&$t5un=}Gzd#U=4DexZSJd<8@-rX;nvq&Q|`ry>9xT0lETefUeH+YoG;C6LcL@ zzcTSK1`aD2doWedBFwzYrvfcEm0Fm{Sxe2qV+ry>A%A5Kc(T64>pU8Az|eP_t#P=2 zoaw&N_J_9|3?&2hpVw|x9Q~qKs3aTd_}yD~PTTgbT(!^S%?;!DT*qBC$_9#eIj=@4 z>5+=L;`6u77J*nLiL&oyXn)Cm`Si_8ciVQ?-x78E+V50L+;q}4x-j;;KYh5zWi&_= z9cB5@@W6o@DH^C;M>}bLRAj2Jqkc*{dK&lAv@qQ)PY>Bb^n7Epdwh;XZd?)BkHM{` zGYwf2cSPtfiJDjti}p`&dw-STb;n(V;!?#YrOyYQ77KKo#S4{LcDYiY=C=O>oMWv` delta 644 zcmb_ZO>0v@6g~6iy)=!Dv9;#4iovdIDTPXL<0gyRO&3B^1cib0#XF?0HN2RaLW(H` z1A^;W*`4mBi&AjYy(`g$OZ@>Ar3VsyjKKW29-cpK;s|*ngCq|O@glR z&ggD{KCqcbKY$$tm0=4$9|)BFNs6&6&qpqKxGzCdplJ+3?g4AF*ICPhlmk-fWzBF{ z>2>rItLawaeg?72$*NPbG9GcOm9*PQ>+>gXr>-aGrqUe4OwGI7O_Q`cPjxD|a$jWZ z-pZT#{ny^!Ro!<<#ltrT$NL9As%hv)*224D%)JWh?j(FsTW4tbo_c6>*Qlhqk(v#? zrS5C9WwVZIb@fhSv$pDVvrU$#9<`cTW^ChM{cLfm(ff-jtBp0vy^rQflYA9&OxzK1 zcw5Z4U(vj-Ff{A&i(uBd94`Cw+{ZosOkk(s_0)ZOL3o}GpVc!p{y&8~9$EMUB80m}