From 718524bf7dddcf14f7bb66e363d351a4c6e5e496 Mon Sep 17 00:00:00 2001 From: "hy.kim" Date: Fri, 8 Sep 2023 15:23:38 +0900 Subject: [PATCH] update1 --- README.md | 192 +++++ images/11_2.png | Bin 0 -> 63966 bytes 앙상블과제_2022254026김홍열.ipynb | 919 ++++++++++++++++++++++ 3 files changed, 1111 insertions(+) create mode 100644 images/11_2.png create mode 100644 앙상블과제_2022254026김홍열.ipynb diff --git a/README.md b/README.md index e69de29..91c028c 100644 --- a/README.md +++ b/README.md @@ -0,0 +1,192 @@ +# 앙상블(Enemble)이란? +앙상블(Ensemble) 학습은 여러 개의 머신러닝 모델을 결합하여 단일 모델보다 더 나은 성능을 얻기 위한 기법을 말한다 +앙상블 기법은 여러 개의 약한 학습기(weak learner)를 결합하여 강한 학습기(strong learner)를 생성하는 아이디어에 기반한다. +앙상블 기법은 다양한 머신러닝 문제에서 높은 성능을 달성하는 데 매우 효과적이다. +다양한 모델의 특징과 장점을 결합하므로, 오버피팅(과적합)을 줄이고 일반화 성능을 향상시킬 수 있다. + + +# 앙상블 학습의 주요 기법: + +1. **배깅(Bagging, Bootstrap Aggregating)**: + - 동일한 알고리즘에 대해 훈련 데이터의 서로 다른 부분 집합(subset)을 사용하여 여러 모델을 훈련한다. + - 모든 모델의 예측을 집계하여 최종 예측을 생성한다. + - ex) **랜덤 포레스트(Random Forest)**. + +2. **부스팅(Boosting)**: + - 연속적으로 모델을 훈련시키면서, 이전 모델의 오류를 다음 모델이 보정한다. + - 모든 모델의 예측을 조합하여 최종 예측을 생성한다. + - ex) **AdaBoost, Gradient Boosting, XGBoost, LightGBM, CatBoost** + +3. **스태킹(Stacking)**: + - 여러 다른 모델로부터의 예측을 취합하여, 그 예측들을 입력으로 사용하는 새로운 모델(메타 모델)을 훈련시킨다. + - 이 메타 모델이 최종 예측을 생성한다. + + + +### 배깅(Bagging) 예제 코드[¶]() + +
+Code View + +Bootstrap Aggregating +
+ +````python + import numpy as np + import matplotlib as mpl + import matplotlib.pyplot as plt + from sklearn.datasets import load_iris + from sklearn.tree import DecisionTreeClassifier + from sklearn.ensemble import BaggingClassifier + + iris = load_iris() + X, y = iris.data[:, [0,2]], iris.target + + model1 = DecisionTreeClassifier(max_depth =10, random_state=0).fit(X, y) + model2 = BaggingClassifier(DecisionTreeClassifier(max_depth=4), n_estimators=50, random_state=0).fit(X, y) + + x_min, x_max = X[:,0].min() - 1, X[:,0].max() + 1 + y_min, y_max = X[:,1].min() - 1, X[:,1].max() + 1 + xx, yy = np.meshgrid(np.arange(x_min, x_max, 0.1), np.arange(y_min, y_max, 0.1)) + + plt.subplot(121) + Z1 = model1.predict(np.c_[xx.ravel(), yy.ravel()]).reshape(xx.shape) + plt.contourf(xx, yy, Z1, alpha=0.6, cmap=mpl.cm.jet) + plt.scatter(X[:,0], X[:,1], c=y, alpha=1, s=50, cmap=mpl.cm.jet, edgecolors="k") + plt.title("Decision tree") + plt.subplot(122) + + Z2 = model2.predict(np.c_[xx.ravel(), yy.ravel()]).reshape(xx.shape) + plt.contourf(xx, yy, Z2, alpha=0.6, cmap=mpl.cm.jet) + plt.scatter(X[:,0], X[:,1],c=y,alpha=1,s=50,cmap=mpl.cm.jet,edgecolors="k") + plt.title("Bagging of decision trees") + plt.tight_layout() + plt.show() +```` + +![result](./images/11_2.png) + +
+ + +랜덤 포리스트 (Random Forest) +
+ +````python + import pandas as pd + from sklearn import datasets + from sklearn.model_selection import train_test_split + from sklearn.ensemble import RandomForestClassifier + from sklearn import metrics + + iris = datasets.load_iris() + print('Class names :', iris.target_names) + print('target : [0:setosa, 1:versicolor, 2:virginical]') + print('No. of Data :', len(iris.data)) + print('Featrue names :', iris.feature_names) + + data = pd.DataFrame({ + 'sepal length': iris.data[:,0], 'sepal width': iris.data[:,1], 'petal length': iris.data[:,2], + 'petal width':iris.data[:,3], 'species':iris.target + }) + print(data.head()) # 일부 데이터 출력 + + x = data[['sepal length', 'sepal width', 'petal length', 'petal width']] # 입력 + y = data['species'] # 출력 + x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.3) # 테스트 데이터 30% + print('No. of traing data: ', len(x_train)) + print('No. of test data:', len(y_test)) + + forest = RandomForestClassifier(n_estimators=100) # 모델 생성 + forest.fit(x_train, y_train) + + y_pred = forest.predict(x_test) # 추론 (예측) + print('Accuracy :', metrics.accuracy_score(y_test, y_pred)) +```` + +
+ +Result +
+ +````planetext + Class names : ['setosa' 'versicolor' 'virginica'] + target : [0:setosa, 1:versicolor, 2:virginical] + No. of Data : 150 + Featrue names : ['sepal length (cm)', 'sepal width (cm)', 'petal length (cm)', 'petal width (cm)'] + sepal length sepal width petal length petal width species + 0 5.1 3.5 1.4 0.2 0 + 1 4.9 3.0 1.4 0.2 0 + 2 4.7 3.2 1.3 0.2 0 + 3 4.6 3.1 1.5 0.2 0 + 4 5.0 3.6 1.4 0.2 0 + No. of traing data: 105 + No. of test data: 45 + Accuracy : 0.9333333333333333 +```` + +
+ + +배깅 회귀 (Bagging Regression) +
+ +````python + import numpy as np + import pandas as pd + from sklearn.datasets import load_boston # scikit-leanr < 1.2 + # from sklearn.datasets import fetch_california_housing # replace dataset + from sklearn.metrics import mean_squared_error + from sklearn.model_selection import train_test_split + from sklearn.ensemble import BaggingRegressor + from sklearn.tree import DecisionTreeRegressor + import matplotlib.pyplot as plt + + boston = load_boston() # < 1.2 + data = pd.DataFrame(boston.data) + data.columns = boston.feature_names + data['PRICE'] = boston.target + print(data.head()) + + # replace dataset + # california = fetch_california_housing() + # data = pd.DataFrame(california.data) + # data.columns = california.feature_names + # data['PRICE'] = california.target + # print(data.head()) + + X, y = data.iloc[:,:-1],data.iloc[:,-1] + X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=123) + bag = BaggingRegressor(base_estimator = DecisionTreeRegressor( ), n_estimators = 10, + max_features=1.0, bootstrap_features=False, random_state=0) + bag.fit(X_train,y_train) + preds = bag.predict(X_test) + rmse = np.sqrt(mean_squared_error(y_test, preds)) + print("RMSE: %f" % (rmse)) + +```` + +
+ +Result +
+ +````planetext + CRIM ZN INDUS CHAS NOX RM AGE DIS RAD TAX + 0 0.00632 18.0 2.31 0.0 0.538 6.575 65.2 4.0900 1.0 296.0 + 1 0.02731 0.0 7.07 0.0 0.469 6.421 78.9 4.9671 2.0 242.0 + 2 0.02729 0.0 7.07 0.0 0.469 7.185 61.1 4.9671 2.0 242.0 + 3 0.03237 0.0 2.18 0.0 0.458 6.998 45.8 6.0622 3.0 222.0 + 4 0.06905 0.0 2.18 0.0 0.458 7.147 54.2 6.0622 3.0 222.0 + + PTRATIO B LSTAT PRICE + 0 15.3 396.90 4.98 24.0 + 1 17.8 396.90 9.14 21.6 + 2 17.8 392.83 4.03 34.7 + 3 18.7 394.63 2.94 33.4 + 4 18.7 396.90 5.33 36.2 + RMSE: 4.594919 +```` + +
+
diff --git a/images/11_2.png b/images/11_2.png new file mode 100644 index 0000000000000000000000000000000000000000..99370589c8e7af6b42011840ebc822f91efe6d52 GIT binary patch literal 63966 zcmd?QWmJ{l7d1+QfOMxwcPi2)-5?>|-Ca`BsUY3yp+mYG>F)0C2GMu(`;YOC`|W$|3pWcp(;EvH7YAovR#w~p`vw+! zCv(;%WTa5A$SVh_&(2U#=!P$Updms(Euo+uWTiifeDX*?UUAp_B)QbFv9?g)j)74* z-h;q!yXt)=ktY$KoOw^pE?~d9PmDcfT+A*X-PZ#NpP2VPBNd@XrVB)pVjVgiz3YcI zDKZjIe`;z$H8;N-F5g(#OHc8Xz!XKIl8ytv^Pe(z_1uMD{!1m@Bw%7h``^o-#D4cE z{P$u}$hYu}|9-0SngW~ee;* zyorekx`?i>V3d;U7BT^^@jS^`zDaPQQ*uE;LAeax8$I_tQLom_IX&-aubac12@3*t zqmK#tTMF%CX`at1oaVW7gBb_kT;>LP%0o0894I1kah|KHtGkAVBL6-=UCn6dfB*j7 zv7%YCEGQ&|!+BROH!ttZRfFmvhMo`~N4#@oB_mV7yW02p(R-!!X_U47PZw>LPggR# zv1W}0CTr{UWb5hV7iK=bG;pzMl z{idFKfe!Z-&)+?ve;d)w*jY5pzy$yk(p`oU>_S)?g-YRt|0}4vy(rVI@Jd%i=-6`qAGSz_{9U(rM-ycV- zx$=>-_J4Y|SbVawL*6UFGSoCQo7>wi70t>yX%8eL!^3QxC4Ucu)z$Ip>gw*Ah-dU} z+W6z_BK2CmE=KvzFtD+^)ye>2VXB za;|tJ9+Oth?|k`<2f^oSRp0v!#6Qo^RW&vFwo7#+f<6shW5yd@!BM1qc>ds+@m?$A zq9NYj-?N$g#&~bHES2`&wz96y?C+lk@vQDh);6hDubWEnX_2o8KGzeH1kn=l&fu@z z(F|DwgE#8x>eJ=AX+JZ)BCO}i)#kaSn~2lX)4>WXH>Q6Ner8P;snS}_6bIMVzAMwN zH|veSEy&D@kMF)dm~wP=?Ok2XH0%z+#K#ZKr5jG?naV$jA?34}DtrwVzJGe+^!xjV zW6vcaoml-tREztWo}TZ6>GJY&Wo@nL_V4dHtzI+RjS_j^K_G7K>;y+gzZUSmrQ+hc z4%tgHH8o`iJDj_ABoR$?0dD1O5L9`8+R0MA^X_np3VR-tD3aM&ru<5iYviK~c!^_4 zox0rTVJC^pdKMKN3b+mdn*o$$3<>?6^YGZ1`E2QDI5;>1Q&aIWt=j#`0+cjvdloCz z+4fMqioMpmRmO-iaBeL(v)bD`J1I4cQAysn$2q#qZknGzZ|v>$SG51t+At?0C5=u? zBZdNj$?zzL^6y^?uPcL_o7+cmadwOGH^(O@DIXk-CiWciE?l1f{^yu45x+q;Jk~P!RKD6kipw>(MLSLxOlmj z<;P*vgQQ-rGw}O6CG_v%;e2pVLBYWn^Tx3g#cGTowlrJ4-X*cnHachA8KK@cZl2*xdy-+W7JMz~^R0<7JPyEGOZz_wAKJ zIcemQJ5&_~da)JxMe-z~e~*q9df%P)^!8p%t12qD`+m7SS?W)-ZY#pyO*AI%##$2T{g)uEIXuqSbKO*bD(ryAGr8OzaM>RZ0Q;HB zueT4DMkQc_QdCp~I2GJ!uaU{f7r!B(F*+j{Y~fhi~|T9pO&V5aD+nf0z6Cg_Inro$M$5?=&*QucRmh%~{|^A{aHc>S$i|mo0m&lK?)yA`;6ig! zjHU4cQFtsmzdHh8Q{LMW0hU;4c6VOyfR5*@?j1?zd0B?dc9Cnb)>_ibix))p#(b6W zHAso&zlv$^qXBlXn++3~4JUVYenkXHg$97}54Z&yTS`hwLQXT(SfRfkR14)cz#bYN zP8ZeKt)zBzba3dejf_NlUhQ5CbB?h&ZA*cSgE!#Nm8zJVYWMSNayjVwndKV+($it< z!u4XaS1yGk_=SrmB+0*LWQ+h_cuA-JSaQ}!vZj=?Vk>#6cU^(-C~$Ak$OU!mh$ST@ zC#p?{aq9rpC)6<_Qf-nFFcMT8AMG>%f zZQq!Urqe^k&oK=ZtChgL0nlgb9^-< zGcz$kqNSz1=)qP@u920J!ib595%0IZZa(yV8)`)U27sh^iLt7R%6y$I@m60nh{><{ z`BVVGq!bhiL4tk!_z?piAF*RFfzfgzSA2rBWphybt-Y;nbZ{^{pW88H@_ehw_2~Tk z+@LGym3aT~$jA$P0Ax5`Zk)=07DE>Z`t{3sVhmt5{Q94VD{i0B>ifsXNf5;G#zRT0 zKLK`r;N!#pq9-ja4JZupYiCEfZga#U4=?Xzy}f?C8z8y9(a{)?Qfn$}R7J&Y&$rs| z?N^6)N76o5o4f+_WCFk)@?20Qm$Uq;N^sDh*+w#sr^A2b}di zHsIs4UL(M2*V!2MgkgGkdg7AWk7q-}BBHfyvNqXfzfuU1x8mhrK4$eLE# zUAEcnL@wkIkd5$ABD3e;hdn|L6RC-}AaI_O7JivKZ}z;3Z~WAY9h#g>z(piqwDuk0 zHBs`P^yFj^UJ6B9rnO~KrtMq7>^a-kO$`mnI@bxbG&BgTtgL{X8`Rx4qx@UPg9UcX*(&b>%dv=cRhEGp-*+QrB zGl)e>28IZ~zYn4EY40zl6oo+S>7N5q;j~?>LC*5d6OTNbZvS3TU|JOUhd8<-c+C+I z9}A}lW1GXiosY(fJV%wcFq2*jK{j3Z)zua3wBe}Wm&Z@{I(~kBqN1X9hb@2q{slnx z6$A|iCZ>pk16$UY+s{QajlnIgtvs%WI1+gi+xU;i3%|r;*U-3CO;^^|vWAC;i62|) zY!*ZT4VZ@G6fI63-S~Ual27Ld0`Ui=u!RbTac3Yr&DhoLZJz~~(?%D3Jf8}vooG&$ zS5|=5W^ipOt*g&vqxmQyk-97{r=ZZ5!zZ!2x(diYI4TM?zS*rF%NaQf8xvC*#~Qj^ zuT`s9Q7xR2nA?tunz~@n{RF_d2hcDtI%WVUGwpgiAcB7__mrQneR%t}%Ov1;0;7hM zynO4P2^bBa*NQ1iZLt6*HM?v5*Jb-wutAvedpESz9Cb_I^?s+KD9QvUuSM8phRYFM$`K z1Yp+XY_)Zh+xsiBHAO*T;p(|_Uo>$)Anb((N8_S?6;;*9HoE?3;!vQ2UUct_LdBfM zL3*4E?e^Z@FUu(!CMKqcwuV|Q=7?MaHiC#naY@NvRx=Fodw6;zzIQ9`H2v*=AFboF zz&pM?UK;|o)oJr--1jgs*|uG|$#p+lO-@N+w_oLt-vX7|$ZG*VPyiC3hNzsh-*2Me zh!;)zbcljac7mV-s(O)10#AQKYhPk(5YQ*4C=Tv5b9pYG@xZ_UjnLAUO$b3@6`pT64a&1tv%A@#j& z1W0=TNZcCbI);DlFLrMK4bWzpfNBhs{t3Vf*xcIcj=*J{Z+2&Wk;3oLcg^4t5%XrO z9zbO%B`ezv02I`k34Hc>=gu#}UMvKaUTZKbA))tR>gVSwBY0piG&D6$0u_|$bx05D zp6+mL+Wol-^4;BCfcup`4^C}=E@c64IBTzlQmuq_H9#Gb z{|Hjfb2r620t=Fz{YD}EgQNDU_Q_)HL;JrcA;G6(P5>=^{?M>5TX-~A(GLof$!5_P z6>UD2+46k5^ifiBDpwo@EY&0Q@6St_s_C-#lL1t09yFV>J_TvDLq|yDMR1!vyHN0GuO;ac=HsUA3~6 zUpEbtwrbU-0l~q+fbuYa+QbF0YpR^`EiN}Vw}|1Cd1e^6s%BSiIQ3;~Bpm}Zqx zuyphGw#fyE?nqwOLnIUw6lGu%u$vCzeE;!-Du_<^l!~5Sn*VOuIoENkuWNQz{R1y= z1SqQ_!S+pUyUHf=VTy(^YLtgbl7IOG!UGvBF##Nx+&Sh?9&Oc$5Y^GkhdAFq7m}cm zrLEB_lRm(*(#MCr=FVkD0p`ihZkB^TXl(vq*41x{=XEUCEGFzrC4k!-nUGVwc z9lp@>bv&@GLMp4PIV?5w8sp=r)9dWl^ol=y*Q>xkYq^Sf*M2C|?3 z`FbkBR{>P9n*~XFSbBPTUCYZlKy`#E3O?jd?>Xn~C(Mn>dONj@l%2zb>%Dk^|ShJSN7gAMA)LTS&^uB!xG)}e?K(}jBbEQ^(< z+RDmGw(u2Dk6{9{X-Du3h%j)b;{X0W2j}JGfv3g-%xDDav`tX^i-r;Z1I_`R==3*2 z2rwik1zr@7R-MfNFr*?1eSVk&k*UVgsjOC1R0KpWr}wSRs_)YYpBtHUvE4Swi=9~{ z1Drb<(lkDI7NY17yOk!j`S&21K(?>U%C8L_q5mG)DL{-(H-2$O)dpVb|A{=wm((*e zGkXEko1?j_?RZ6%3O#}NI$vL3fNHoxLP9{lCUe`9fvws}ylWr?NBH7Vp%QVu9P;h) zLN}<+6+of8zub6b%NI+vg8o1ARZaS*LdJZHC+7*FxV^pQuFVLel2-W~gaJ}ueP>6_%5R`qypJ1LwL!%__05^{x`5NB zi|N54OFNsrk<(?`cbVzh|GguhINaV{po(!B%WAG0-PqjFv0jl?O>)Egf*cw3NDkRN ztb{L>H%Wa{^iAoi>@-n!4%6p1YD}s>^OhzH>TQ=A(nK_7yzol4iv zQc_3&-%WuysBbj}De0aTlB`F3$z~A zR<23-G8Fnj;(8oc^(*1*xZQ%SJ;;pz?;6Spr}Mb=KyLw|2FyjP;FHVC%ilvojE_75 z0?%fO^wN{^4c}4+^*1(pOwEi6W+uz@1iYT^4a&b;|738YWLE$&+Ejmuitt{yp<`7x zX`C>6+TFpxr8{KW*`Z-hx7pl|x9SOyCgBH%Hvk}U>(T(_0zBcN~)5&+LyTT6wf7&SGk^ zJW{Ua&Cw^Va^v=%Cj~n^+yyIl`N4!4%3OToZhX9Xq1GyAFnk46_F>V{J5~K;&DhW1 zsSe&FwR+(SQ9vx1&0&S1Ht6Zyi!|;KLF5#9@o8ZEBKkBg0eg=uVxcmrd!Xv7vNJRR zdOZ|D4cK2^K0HB(+$Xb$BUt{ufM4!60j>0mw&6n#akwM8k&pLzVx-h0?3?-SuVFHA z&AHWX_ zVIxF^$jO&&Y);MQKt$Vx*XL1Y98DWmWh{3;HbFgx(Rr>~X(24^BS5`bB zj>doC>!5>&D5E-f403X_t18!LZcolWlT(8gW|uN}zvsAbymd6?S58s|Sf%d~tJOXgs8Gju_2YWV-+ zY)2;_Wn8MO+FpEDs)U^?*P&Iiv!h<-!TaHU)$Nk(JLvDtC}~9_pp@a2<-SpZC4P0vdVUW;{bKj@SFqO;16T z7;yaR=XYIGKf^tsj$8&oyV5E<7M-jIYhQ7pecRRlR#;O7OL}3!z1AL_Wc{Y76$AUh zflIFoUzMnq0fyD_lfl2g&5QMi!l0M=seAkICWw|AGpxx~(^D(Eoa%{@|Iew59Jnst zD+*1AM=lr(JrO?QY)psE!Gz|iDyxrxSo4YAGV!EjDQ22HKV0nxXIG~<2gjQ4ySO)5 zzg+X9Ia?(YQxr7?MY^IQVnYsrIh%T&koXAYOB6uYNl0FGB{I*Wh;UQP!|Rl9^909o zJ2^IXw}R$rlV_usfMTisSkApiYW8XoAAAq&4TFasiGjJLCwG%(*-L-+)H#WRBe~5K z<#m*VMBZPOpHTh&9|tqPzV(NN++-p=Jgl^i$^g+ak^cO|llrZL(jVObYBamA@)E0A zbGqj&MI!oChP(uur~hRwSuwE*;YMVIvhul~Q>_G?1I?~+Hd@sy| ze}Y8!B_?GhoA>!-I^r3rtjg>7#3V^2!v5BPqyTwg0te{Yx8AV$_+~Y>vUiU+XJ*_W zZHk$gx$#lNI~BrIV&xG;15O|Vr2f&3Jm}mO9gHVSLY2)&|J%0F#Kipl8&g3u*eGFU zr9XU?Zo|olPO3J5jTA@K)wf@XAQ6S6kObk+xCgh;8yz+4>xgduyCzyh#EhXlU2prevXba2TcU z1QhxB+D_K8%(?BhiSxc2P(WUrK-1y|f1m&go35&w`7k-bgMe z80GjRi}jk^bsO7}SXe|O8jX9{|FSm6dZp00Y#ORk!mOCz1FzJ{ew>T9|n6Gp41tsdI^{Qjb$@6~gfYbK;&#A%62)qDw;SxBP&(}>3Dd=-ok63b12EvZi;5omDdMad)>n5t#w{5QmI-bIRaX# zGJfTM4=zzMv91|x>$1Ao82>`7h^Z(dh78G+a7zIx-gidkh7_Sr)Y8RxHeksHTTLg9 zso9GDAqx`*B~5H$BoWke4H;xr;HxI&Nf_#iS*;F0TS|NRXUIw$x|&dZ^-mPOc=Dd@ z5S~_n$s*yu7q%;8mY#e*Nz0S0;v^zuh3iETpoc8Q%i{$8p8ljR`EfZ7Z58_E#t`SL zE7?eJg7J3v%b|4r85b6g+guJlp zn7B`0gJ`I$fQQF%c?mH!?pE^iYat~)``tMDvry&d{JHh#7ARTq4IiswCfYaBlgtC7 zW9GJ>aLiRz#-W(4=L%zE(eb$1$Qz1{rl#1D1Kg=keEbD0E$g5E+(m$pElO>OOXwZa zu6LrNfYui8)Kiwi4rGlb$!U*8clNq&IMLqLwBXv-nCych2s zO*fofZLNW6f*FN(cB_C|a&~?Rgtnu(*wuRWMid9PzKO0tjEtD3of=GGNbYzJLT`O) ze6Gyom*bzRKc$9;O(HP3b{^j%IA`M?!mVwEhnFIwx0mgag180sU>O9uJfFlMDR;cd z&6cRCVR_bBHtyO$YG`Py`}xTd1r;`49JMyUQ5bf8MZkM}txsGl(abE$dNOD$qAS5^ zkpvzq7@|y_@e=xX$)g7AW0h=LNFts^sshum+S)UCz@D7wu)#Tz+>?fh$e&+B+>!m1 zahyOGXQgb(R-l4P8a<}HyL*B-+J$w`@%1>Y>-X8w`S~TUQ**7lp1FZ=@rW+p=}0wy zDg)AU)^mq=ns~!m11um*aT)dGaPk}#tIC`bAy)H zi0QCo(wl=5he8qqNi%NfGq{MWd?p&ax zMt;-ruFiIU{pWJ{(Z`I25FmJIX=Q!Q_S$bLiNDY8%FQ$^EOIDlW`NjmGxsb+Vb)0b z`c7q&B52sW=jYkz>B1z_-ZfJ$-Alz$iR(4@)%V9#ZT#`PhDUiz!puZ4m;WlWuI}T{ zpLxnSKBT1K4Q|mX-nZY3*`N%{W;mGM`0s&MWA|*`vNMole;hX9+CxL)ex#v5y1%uy zw0f4SmGJTESMa`^$BHRJ-giQvYES=o2Ko;`up=s``@qEQy*>$VPe8+m zU37vs@xU9hE7{U0Ki^-h)g(f_#)HwHwrw5VcVXgl4)KbKVIqAW<&qH-`tKpMbXjNd zmB>bQh{M-_+P%>#W}1+-Q1t$#&YWqU>Z_}iCO2jZjfzQ8q!5_I4)Jvd8zyw!mfv^J zPgAvetB`5$3Cw1$nxPG|qjd9{xXdQdlzL7$jh5fhEO58f}y9Kz}C@w4$eVW;-^f# zF-~IcO)IwZN(@ar?^3dH86>=df{V44=EmgA7(wnN+Td)D!(I3qyg>2BM!B_?6tz5w zgjBD~gB-CSz1Be-3b{IzD=01G&MeXMKf5ExKKF*{rKP9yPh~ONABm-#BBMkr_owLt z+2jwAE&ZYWGjiGf^}IS{h>%v%(2>l}%soFkm?|LCcDrSYz(DQfDq`G5tU*Z3B zJP!rNUD=|;2FyTXBP#RvXmpW(nh}F* zJ~o;2QAOi7s_RiTe`{vD{nh)ZJ6N!b??NyFby~@~ZI`Xj-6BDJMW&?GFu4sQ?KkN4 zsG@LT2Hux4X?;>D<Y5qNlCIvwD?(=t3eZ1-X0zRCYQ=91X(I zJwNR1r23!>z6E;DsQ)rxGbwgNc`G9$S80ODP^d#%c^-!pOdaU!knkXUl_p^xKyxMT zi^ga9lymp)E5PAX?^mBcGoigQd|3FUGc+Uy>r$>7Ixvztk*C*cRjVeaE&^kQr~`#2 zs_co)m<>NpR>@+aug9(^!=G6;msa|AXz~E^=(G*T6Td%SZTm8FcsSc?5Im<~fcXvP8d@R;EjkmS4be#1j|ZF4v-M+#XkO>U zCAEDh92;8idmvW&u2$7`0G76ld!P}kfPvxoSYvH?j1u)HAOsR z23t=H!x!D$sGx+h`*_Hw1<|3I{M!SlA@1iY!P&VrvWWReA}3gk#wf<>rWY5*xt|(} z>k5k}Nt;36VWQF5-0Svu;PK|j(b<{u?OPc0#;kl7bd!A-IqiD1r$0Vqc!8S-(%uvw zP?-ds5rk-IY1HCLi4PZ-w)Xaar_^wH!K|>YP}cEPqJBV!`@rYS|1Mz^K21F-A2Eg= z`70PYWU8um{gfxs;(fzF?=Q>qg$M4Q=fn6hH1jJZJi@K~!|ncf+Rub!qf%@si4!pU z6mac3I~Vs&t~O*sWF=PBdk5zTuQhSWVkpl^4rTs|+A2xX{S|IC{t2s)yV}b*ME^zu zDFhKNndEGJC%C!xU8s{2f%2&HRsaEo#{=}dXlXvfO4SqSi$OFh$WZW@vcXu#%QQ!Y zz7QCRJp1&Isz}E#G!zs|{qtDTx;b?aXC7Z($6T$EW==e&f2=9f0bb&KcsCIxMJ0?2 z5nVELu4*N)e_=RqgZoptSzTicp`oYisti+dzo}pTfoM#c2bgQs7ORWR+@(f2s<`HG z65Q`>OKauPI8iS^8wb6i;7hAjR(5Z6-x=a2!i3~unChavgKL6`$Ytvoiyes&{oE>q zZljI^s!`?kkB!)z(IG#z>bBt={Ex2UQ3{|Kap3h45g+gGmluhf(ADUxv_XRpbkPlg zgV@mAJXLKf3tBY&d^6nA#p6O78(eTCj22S}VMBwPAB{}8PtIbJYq)Z!rYt27^2?Pr zbeg-9Sg%~y+CExx4XfeI<{?zwbm)jvsN`)N34XEADN!$)t9-|8-nrgsVA?we(Y`%K ztb<;oLtYBNU#c+-TFxB~dU;OA^-EKg%S#JYku)uTp>xHLJ-UZ#JF_1pHgo7ceG({t zZGaIt>UuanI~3X50V8h6T8Xa-q5Feeb9ue}^>k&(Cwd?tfPb}xhuHrNCzuq%ZpU-Z zY+!6XH1-ui;F~LNiObn77{(b+RPzN~ly`AK5S(3A!ib&vzCbmD*Mgz( z$d#V*t58>$CNZJqWJSayXV4YeGGhGRQhs(lCDHix%dxV?B)eV6nD||>OF+n~g^9Hs>X4`y4+(i}7RHWVL!j1R zkBYL?+`+x&y6}5R$t!IsB=7h*sFzwTDKKzL7-}E7UI5r!)W{W_sI#n`DG^1dbU*i ztcWUM-M~^>OIHjmJ@*PlQ#)efE^xNKc)f1p#1UJjEvce_7OjNOpKoq@;VOfC6+^mp)b`~- zuiZCgOnzX*6qGvY3NpEExdx0MCm7~0O<+Am*G(+LQzW{Dg@h0uA67oCEqQrO&Q4tw zW@@XdS<=l;O~n_=%MTU5x10=`pPzTiy4p>%k4sEczup5~8SXh%X$C`wuL$X4ud#M` z5T?NlbwX&reAeEQalfY1E-GxmfhZf2>&}U)C>zxrC7&Ej|rVsEC*c#%#7!)j36u-FW2MPzDexlv(?vj*b%kyGDqc#O=e*cj%^26_6 zFF$wEKe3gjN>(-^k5>?Tw)5jc`E>e*H1%|4q)e^$Oq78Eegu9RxK{rmVeuJuMJTff z(F!cSU8#D>1gz%K3_fSj_>4E+*x2yUIFB{9v<(0H_3Hq+bQ5VGmsLb19eb?QRzVwhVFnw@;(esKocQ%RfF9Y)7tR8Hz!OFwcN4;kC!v}p*od7h(@yh6yZEgGuxux$+xy^vLxJ`swQ;{vtIgNbC2(p%*9s~?S z0!H0O-Ku7Ud)!*L75M33%r!>joQN*-R`0DX&53G^1BGvZZxssb?d12oZZso^#=4Y@ zmldYrVZ+VLhRNC16L;g7QBWwQ!ki+XQ}a%jnt;lgj2NMX*YrIl$&4`KnEQ+(C)}EL z&snEzN2A2+x>yHr}t+7eyWDDZ95lbj)a=TxDuB{6!#Y^X@t2n#}yG zom26k>Y5N`v1y<+iHa}AMkz}?d%D|;tz^T?5*(=x?Bz)$dfg2JlRTRX@2j97esUv0 zxnyKCk3}jE3b9&S8-j^S!y=EjLF*kmL!~iM+cIu$!T9CbeQd25RA#0r!+Xs?5<@fy zQK+q86=9YFzI2=hZpGl?@z0eKg0c|u-u2EZm%*?4WkTckHwyP$);Jec;1--qk7W)Y zPX^~7+msFuJJeUP7mb;*5v4e<$iooYpL zpvdhGoWI+}ybzdC;RG8!+qZD}G$(4lH2$y9+}75h?3Z86ke1n8h$e1KPY5>ed>cX; z(G@qY#OvjGp(Y~Nn4U5)Mf@HkF;0|u8f9?M3IRvZ?lVVUN?Ou`4kfs1yg&7Nw3gCy zoI-yrt>Wg%{%<9o(EFJ0^uInC&!?stYmc7qywtQrL^9;rWf(ZCS+a*jQS&(VDQ;2tZI-io_C{yuMv1&^iB38))!A?M# zb0!R4CpowG;g7A*?w{P~mOY`Pke>W1U1Z>0x?jUZ_W2ircWdTPsIkDAyI$i#{#HtFk)Gr${Bd&{y7tz#J*K%Xvs$4Sg&uv#uZM6_sQrXkXIvjDGPBSPKicC1#wruDi z5@8=k2e@zgIyy6}r7mLIN~NE+#Pjo&^pS=cEmZMc(J~@)*EQMx*`Zr~gGCZ0ty>8q zf}{g5 zpg5TEXh^0m8c_FWl@~o%ppi5QMEF5)hGl%gQ7Qc%gAzIE1Fofe>BEEmb4qJ zrnkCEgYrIv%fdfC0oNIw+-zZ{#9!+&Ki5MW-E}Yg{0C0cbeZ>-y?PS`FmbK;qjL=) z`O$Ru#&B-g2>h^zZDrYsloPpV+UaUITGM4iV$sB$cfT<~4;jT;yi-S6ikrx>u4a43 z?Sx?MV@86xwcV&6uj^((o$Y7OTc)#xA>B(VP5rTRHS0=#!;-9ccP{u~#%BG2}uhtF$;SvJ+mLRp&h*Z0vt)oL;OHd-SK_N-BdW?u?XjQ|1fSazW zG6XFH;M(s;G(mN>8BkH;%zHg>Yyo!#mr^nnhtO_&Qxy1#Z3+rBKLc|%kfznD%S@x1 zDZx-nQ6KRV1d|S#PYG5>ns-`eKRP3QJX_&7BaD|{+e6?pq6dLR(4*`xK?Y>nyzlg) za(z$n-nccc>1Zk!au%R+otrsSW}hDcPxU zYJ1{J%9thAF|%RkB&kZ3_MR1T)<-iXx>F%^64`p|c;&(GU{hgLgV(jx@zzL~Tm44o zC89BuEL&a|t3{gG*{OYLo_r0yO8IH4adH0KtUre2W0_B;ro(brliJu@@r_goG5-=q ziEMP!e?}I_WSlUD^z<%_>hBBkPa$IUewhN=hgEx8wOu2lF~JdCugdL16T`hU&g~7t z+J#nRe<*&Zl4clkF4Nl3Q;=H-Xu6pfCMO5@VSy^XVR$ zpp#V~gCZ68>59tC+?kghlM(|M$EWIn_)}94jQr_3t(jm?Q7`ZIWYqB(M>8lXSz{$r z-x*8Fp_XiJMi|KyPK13HkxRiL`2DzV!LSt2M1Keo`7&5xjc7%fpz9@o3Cx1(I?HJv zDmo-C)ZdqxPfL@_b9pJd@E_B&Od9SDOZj->_uL0@Stc>P(Ua2!lpz$ z8Q%u?_M-DEli-(BoxG}jATtFbx^7-|XCo9n-rn+072^`}>P{pYm08V6DC4+toV~s- zV=F5yEyH`a!4#Pm{aOf}=%b4kk+!kXZ(#4(S`R@93B@yeQ%M)6fEkJNjR07in3(rP zd$T~=Xe`zL^z3|Zz;0>D-)IPnKRGEE@?L+nsdhT>m27kf`)gvX^-WQ;@MNs`M1*C{ z<-myerp=SV5^-+ea04%)@*Pc|2}*yL2jdD4fU4qx6!M<)lRqA*(6P|v!Uzs>KaE$b zjME#v63!m)_GUm&pPQi=f)|f}h+LrxVQjWh-i!Bdit@9UK*%Eb&>>{pH{P+Zu$>;Q zt8-6F9n3YjM*N<%LF;_23ZlL9Lma(_0T~Q|Gx%f)#>DXJ*#rbrNx##V(;`AI(+U>( znb(fNB_jVL*Z=v|EB+R*cVH)inxkKfX_P|ceJ6ulErA&Uh<9aQwgS3kFmI?``^4{v zbBD*L$4mLOHk!RsWBXV_2QrAotsRclWs;^yT9=zygvu$YgwkCT>AuEiN7@6ocxP~Y zd+W0fZ5iZaPzGFw-~3-6VjKuh+cvoAwfqJG<(``%!mWcUTSTN;3xe%SE~dNKPVTB8 zrLq1>m=P|vLSO%|rJ}|cFkn z&@no%8Sk8G$#wsM#}TG>TKcm}z!s<+taIr}Xz7{pKa|p4-`%|XjLWEo{&H*?LT!t| zq3XH?YKc&w`2ePSda?Fp7rGJ1AQg@8bgTJ#;4#UKyiKcCvxjL!>jgiNRU1QG7L(!AGVhSX_BDf}nUFqo8GWhQYZXvJR zSV}XxyZ=?f88y(NWF^zgrQ5I3Dy*ls7a~8;(j_6h4 zCi0%Vzi!7u;2s^kpFNx}uWJO)N^>4_a!Eay7fQzob`!0{eb-r{ah@It>8wCMTxpOc zBm@I4`2SXFBUx|2xHs~ysKaOF-?)K5>t4G3RH#3f2Ec)R(&qAtQ9MT({wcpHXEmca zgs5WBIJ`+#p5QQpS9Y59#;jxD2M&h|?=kaq?GtEQr%OxcpdW6A=_|9kWGJyJ#4EFg zEJ3BIsyktnsLv1^)|KgI3?xNrmEL2hekPV`y&EqCIINR=-?~5+7*4KP^nTTrw{9tkT}7D~m^r9?~L{)<=DN`$Bj`)wND&rT9i#<+Je&3DW3u z1O(VNNN^OiM46U2I4wQfxtO3*rg9TgLu)$)a&mGnPjBPG@36iz2`?BZo~L{c*^<)H z(P~VS`kLzSYF&jeRu?W23J-F1c$kOQ8}a!-4@UqJFug>ukVQej(z)-B{vw~USM8zj+>hlj(ajw}OHdma)UPa6XF z&|nmFTL)>w!}%w2;n!}D+^_7lno3Ltnbr6xvyP?0cD)Q3>B&Fj(t&}$Hwf>taR*_2 zC3P(>NDhOH^{3^F`m4L$6<7QM}Bm}dAEQ2F=Od23L|V; zCv&wjXW8^07_|rncw}U2R%oZF88Zbbc^bhDeoj7JuOxS+b7f+KzG2B=9e?G4=W20_ zY?U!`+o39oK?ess77=oNeFdE32Q4t}@!-`7!LLva$`(4`8i+5AjA&})W?FImFV(9Q z-TNj6q6)MMrVq<(40yhk(+wZYz6NN&){wc3i_oL2wChk#ZUr@+O=FR*u6;%WmzJAH zI0M1tQwrsYsPK4~ds@(d>xvd3*6HYN5Qg!wprDGI<#czg2BW+bhmIrM;7(uPml>2j{}Bj&4Z<|MPUjEZ>K+n;a4p_~bGvN3o42!A;|$0VDTM8e z$b{6HY0yl}9~484`uSjdBf0`re||ef2OZ90H=X5^&kvHSWT^z5a~jST{=7Bc(9&Ut zYqWN6jwH2NS$)X`(_evaIQZm!e}uRFqeKgr^b@y|xE4*-kUh@8;1>p5Q&Su{Mh8fk zV`Us7q<}!)Gb(sjl}mrD3Jr1x4V?GsLW*-uuJ%LbU?geeMrSB7e|NW=6m$5N>0%je9GI?TdX8m52E8*SR3@eyhSuIof`|vBFA9$Novn@RefhDo@*dqjzj=c+TOHa9ynQ3Rm1|BYC+=ouRbmxO@@ zwyWZHt@WFajm}rdJTUl!9}>%_CaT?R>*SMPwzm|_@d&G6#XwYL1KD*;zlZZaZ_`;V z6JPITUEf_9g9*oga#LKx@5L{jwVA|>G9E=&r@}j2bBpEZZzUR@E+<-`SN{qt@XZ_9 zpT_2&a4bd70RSJzL{smoe3oS>XDrpz_|wD*`x_TVgP0gcJSQO(5mXxY@2G-!o^IKZ zru@y}X0y>2W+gI)%+Y%C^UqH+1^y6Wb>&is0k@=4J_}S2@8q*f7@eEk5vR)(vK)mMW#^iP?1qT4vW0N)?9_B& zvnY7uTtZnf)QkPl?Kx;82swR?rGI?tf#K(*n`(CZ{SD+7M^Z5_KxMaaP32#;cn1$FZjGAp=HYw1y>8xf?i)= ziP0hV%eM5fOy}FdopT)y*2F-Xjrd`8WkwI7aw|jr=%T+^Q6_J&3RLMl6 z>7@@2d>rrDZjO%`bk-9BwC2h`R~a`7hvl-h7p#X!XHsZr>gVJU;&Q1PUE zaS{On?*EWkGyCkl*IIX-LVIcK4Yt(=2`pOq|VDyl!yF*<}F%O;35^#;MDgG zRUcPZu#YEBde*4u;v&MUw$^hURh;0b$EYSkrlH}{nB3QxrtVU&6eJ}0GO}Vw*44$0 z@Vk{ztIboiVgYTvT=I|P(AI)=^&SW&iW9;2zDnlKy;ZUR1_nTw9TD4l@R^+jXv5Q& znwpNHqqlzGEM#VKmZ;GH=+MTo>T_i~fL~Nna)$L&Qv-S?2E7`#sM4KsgR|S@qi^jN z&NsHw^q^_K9suZ2?CAW%6Z-MLYQqTt{FIyBcM>-k`i1=A;&RdJ`~wXV(!xR&nXYu} zhP^W|oXPOo_U*~U-qA3}yqv5lVeL5DXfEU~(E$oIxdi#&oV>C|wTDX59CYa1CBD9A z??tnPBERcKERbF+ueRFFw&s0lV z>a0IslDA{t1faRNqW%n_EfwXR^e@a0!b-;fp@oIAPY>1vp?Cxha^fr|K7%>KoXdP3Yy>bT5Q3aFGiTIi1zGb|!^QNB)(q=@{9e{Xyzw;lc z`VHbmJMftj8(A4K*1$k+WWoyLKqRWaH-muv6)FN@c!)7q%C&s5Yjd5+k6fxY{9p7YmR$4QoB!`Xj6tO{NjNrBk_^m z2p!$VcL5 zA%3+p?gG1%{C^?>5qyWWSDffSMgl}N&?#(cvx9NAcQvo>(qv-Ke$lQCx4F+I0^KfL zzEr?{4WO^f;=2R2}i&;Yhf!x3QUnvWqeno5SLt zpgYea0u+Y#n0Nb-k4s*jXBN34ErdsD8bWJ^L*oNYQtipV7=U7ZOQ#~(_Sx<;x3l|> zrvy@KndY@rRXV-N4O|0=u_0pU#H#3xHC-;Dk#=*nrrTeRSeT;>yV^aOzN%b^{Q9r3 zeX)+l+ZcVm;&&-hn;NnSl_CRqTVZwDHQ{$1pPRF1K-(#{Dp7hT_6v_4qw&)tu&_hp zY@LYlfMzHgW@9CbY-NRopcPF4)j7NU4^oATep;H2TorgO;q+c4N5|(DH#=k4oZ5Q$ z+RIxv@9kLxj=rLbmGhUs6aj40*?3%w)e5j`}QY)O(d$^gw^f}Ndxov#=HiH;!%^jM=n#5^{-@QzFYtSnoV zmJ5ykWu>=fb}}=Xosva{ONtmhN>JJ7$uJa z&of-r)l`34eS0=(&8JaYhV=eoi5qKh5SSUyU0n*yvIShQqG*NRyS+w!S#C?!(5f*l zVT?gVHFi=t(dKy|<0NwA{GN-TTEdbxzoFu_$ zpK0X;NNY0N>DCR#pwo94=CljQ|A(=zjlI z=@S&!v!VkzpwLjb0$!hR&8iQ}#t0BWGnl&i9uTU8cA#3$e=smp(}gndB9%9fK3Y{c zG%>~5E08g`yl~mz#Kij%55+>ElixdOdqsa#O)KQ1*u=y2TNe*l{G_T>525MKyQDSM zwv{G{JOs7!^7b}6qBnTBs%UGulx22SgLV`G?2|RWl_mi{MXqp%GtIJ@DSH7{z}s~< zb==caX#6{FVEWN@;<42Ief^rt@<1$I&9UgG9?O~y^yiJC;Z6<>U|#|1+*q|js^X@i zy`7!mx7CdLj~~CaIOR{@x%Zw|?TNm+RLNQ&#AEm}vma&JsHsN02O9*6*%mEP)K>*y zoI8p5Yd23^=R#*G7Vw_w0;{SX>gqJChIi_5DzYHn**n@{_e`=g7KvqV1JF% zKfVGuGh7Qf#A;M=dp*4Y0B*DJ7b#SOm0AZWbgzvnmN+i{d9w=UXmZ~A3DV}R!qD1Y}qt`ecL;bft)I@1Q_3QV5K-Z%$snJd? za8Ok%Ul%!CHFZ<;J1HtU5eHvBt1U6QRAuxBH)&!JCCCl#s70xJivw4$*YUskR_>am zAh!Xm)m|ugA>T>ERv{1bEDfi~!U``tCk|^tPub|Y8w;>M_4lk~Q6%}fGodXlrf1f3 z@(_*y(lEU5mSu7k-$_@fB$RZmP8KCy=-@2C)%WsNNq*DGSE#26>ZiC1 zNXA(viMa{AiBVpZ4Up09Zg0(|<03t8uXY<(1&7$o7_x3}k0nzxN6aK9akp^50gtXiK>jQ9_XrDuBem6hQh$wa1aysU85V=L0X43Le@=n zNuiX2AR4Mvx749NW$3w+;_Ik-L;eXnBu_hnIA3=^CUU5u2p8#e zV84~*JG$LY7EH{3akV}D@k4&Z<4|L(Z_H@6fc>GY7!l3jwLkKvYV(E{AhD8At*Bri zuuFaQ7%CulKa|44@4bVz8p7z>>wv&j>wFbDsy1XsmHwDnkKO`mgaTb0!u}H62PTlX1YiF>SpJvkPH}p`$~#U zq$u9;4_@1MUjznse}yDbv`IalhZIOv4~3LS z(HH;7u+mL&g%&GFyQd*k=VO%{ZhCj4AzewRl5ww7zS%{B5f=Q`pqu9^i?sIordqRrO<1wO3RpYXL^u=C%ggVK!D!q- zd#hW8mUD=_U`?R>5YK!YX%PXkqgFw2~4g`KvC2?#MsB-+64_hfg9YH6j6z9o#0C=GyY6rea= zXhm{LP+TKN@KB3jFjpSB;_sS3YYH5iX6c-sZfpkS+N)8L1Q%CCcM~aU$sXt~z^xA> ziFK<~`S);lr(@N)_YD{r1buZWlW=4=jw9Uz6}+C4Xe#TiV?pv%a+N8m5Q%s zl$!JeWPU`ue~k=KfB!D@7sH4TLwWS6y0H??57*Ym8$Bs|)(uHrT4vO)@1X$ynM^Bp z8w`v|zJA-NZGdje$xplCN6_$ACXzMg_e69Ryo+U$3=v4u$;)9_YZX6OXhu7qEy4q7 z0fSt!gGT!=RpX=j2ERd|O<5tfmD=gtKT0jGtXS^}_=;FSMtH^-c47E8*#`Hkc@T+VlOSz73C^3qd+4B7kO*`CB8eEASf=ClI+fMkJrva~{r;|*BQ^$5uLKITXJ9qRC_$Lj%tpi^ ztMo^$owAj0Y%wY|23pM!jF>n@bh{FA*@`DUwGpsI1yQX67^<^XcJ3ax+OOwX$7ON` z=G7$?I|agPgr>@f`mSq2By2rVSc;bm(9WxdJ&ur&kr#nvx@#i#MBXP#w|^4)vFC0D>LlQXt=4JBwTw?16tw750@wpw$lwlN2ZT_4n~ zEm}5BDJfQmPQ`TCcn_$7IJRcWdhTW;mAM$1hUJ?Lyjc%;PgoK-ls~buu*CyDX_NqJ z;^2p38{|hsdJZ~bcjsA?VU>uRdg) zTBWDEbO&BD()Ecxx`zjrQHQ(&wfw5wD|a<1zO!$&+|U!H{1+`hn_G$Hf@oo8^fK`0?0Hi~BdL~_~un0#0 zu5*8E%wp~mkImP4tne31;vlfFmKEgfKy=X0K)t#f=Q4D{BX*@uK(2kD^r;Zb0PuF~ zmQ^$jXE5thxf>3>?a0Z0ey)srDsoF9mTA6o+ZF?){Bk zCEWR4xf@bBIeB5@%(3zoA>#PQ!srJEg8|$>OsaQQRUX9iRv6&bWfHy>E~Z59-UX$m z5&@y9K#Ji{K+&lRgk`ua!*+IlYJHpb@%HvY|6L59${$H>1)QOmxPzGm^d4kEY-$Xc z%Y|4peTA%sW$~mKiD+@HNQtU7VbSQX=kz<|fkCP<3Z>J2YU-8xc^qy}w@610pn0~& z4+T`Li}{QYqVK-zLbhjckemH@tDUouS+r5^#hYk9Z|^OAXiibqE0N5tL9$c*^&+)( ztN+3?*lelC5DH(<1K}tJa=UIQ9f=O57!DH$?Z5i?NT9Yn!Bc++E@9_|S0Dow zvY{-tJSNX?$Or94q7Z^#YGod+-fG@JDP?@X4lQHFHC^!ic`+|V zi{EPp?H@{A4@93OD{D2>Zv86w0Z1|qw10d4VhK7eWxq>4A?SqdYUtJOo;Q*&k&J3g z9Nt~K5jHliTLykZVMTSSo&9|^C<8%{VMxD2Gg1Ky|H6d=sSJ&d(40_oqUka<8X$-% zo!gEOkdyhf|4>j95fQgvda1x40v5(daJOeQn<-;bpg@*h8mYT(dm50H!rAEMD&Xn$Gdb7M zWxj?v^mg?E2pc$_jkaOY{b!brNWvd=nIW-u+`hJP=Lmqro>Lvvy37_@4;!JEFrsNX zG-Z7i+B+K-8_=P&JT8Uvd+_6-us5?84i>%lo;Vqb`+V{s=b;SpraTb-nW;4kq!;Lo zxdC!$=~9({peq0fBDz=W2|}VZQLevVNJHA51ks=85P+u4tMe<EYs8Pya(70oS>5uV9oVQTzd2L)Ee9JPRCbAR7(ePuphBnQA4LjZO6 z7qV|{Ho6bVo1ck@^`ShFXTVH@x?9 z{od|BGoF>n(vPcHUn3o(7W_*VDlQTeWT<^m4SCtTyT0%88G%{fK!@Mg#hl9wwmInuIb!id-!e95a?v(-u>6#e$tFY2!hbk{xT6 z&rWO=YtX^y)ZD@8&Lemrn3=-0qca~Cofs)hD!%~s<09TxNpx0XZ|0z}lw~d@3^m@` z?!H|1?-tH`Y+IZ@PEO1jI*eNdA})G+dN!KcEiC{YadQUjE}2QF(}6;!BgdFJiqJe; zf@1DlsK6`T#|t*fH0(zqr}wJ|1~m|Vz)938;+vKl%1)v)nXjx7DNQkF|D9BD0=bj` z1@J~2I?TrrH9PD$57#`sJ>y*0YGxJ1B>nH9r@qaa9hZ1DHsmq!j!RSquC+AdXW8sx zKY;rDCUiynA9!XX)(HW5Y_-shB%o%I-Ff&f7j;^XK0yJSizD+Pa!t+Vra`ySaQlU21AEh~J|O@VKN`l}z8-c4v%V zd(nO@tLV2(UN9U{EHv@#4XYINNg3~_e=?L-Iv_0^?`L`zwGpIV!1DFp!zn~U;sXNN zwDLy;Ba)U$v{2kIviW8kr6bS5&tf_rG&d<_!iUQ4y&pCFzybqi*15IyYd1ZFu5K5|!zBa4XorfsER^;AnMy{3}RPEcn=-TUyW%hK|q z_k2s+hKrnwbVTWhwb1ZC(PKTeD#Ya0)-YwfV7`m3j{+TY?_4vdQIsxRFZg+0UWxaD z>zp1lJw0kh(g+F_mzsbw@TQZmZ#R`_4os(R2Skm?G=ft_73dj-h+ ziUf)#VUCEB-Y_fIw_q@fHvN3KN)2KFBN`J2E5mvfTJuOn(6{rP7!IVNEA!WH=4WrPIA4bpURFPF1Z6Ven_2FpJQ(essGO(+`l11Pl^LAZ$^gC zoIbk?45WNHKtmyVT#Og+7>@v}p1(R%tk5jhJn>mtdlI3^@4>E|BB;oxM!p%>&})YP2-1yeZsrA|eE%gj zOIOT(Hu)1GI4fFC%8x@d*x26wSs@JHMTRA-^^H6675TcQ{n#IB$y#nUiuaFMS$BY; zKWDM!CtHjRQs_F@!DIWyGweUr=K6cR(n65~oa6a2kGpo&a3y6Taz}J>!n6Kqd&NxK zhaG*Pawe{mW!}Ha%4UnTJD!c_AB?q;@lpQFOF7@Gk}JLOC{fG3A%8K$tIGK%=Ov`) zwLUu+UD)MV)c@Czz}HmW{e#&`u)UJhtJu61mUc0NV{at);(ZHgY8})ZG%QQgnG%bp z+{H$wdB7hg%#$Jz_1(QE4y%VCs~4NQN8uwMR1>1ECazy&Mic}dkroDq+at^l(VBs3 zp}GTCf68k#*F#^)^$>bKxL~69GpUt7cCU5d;LAhO&cqn?*+z-BKvDoN_J)3<2W7KSL$9oy!Q zOP3!Xjp`|7umf~#;=K&u^m`;B99T+22RL8I@21UsdL>jqZ+(@_L>#h?f1v8XBwr)L zv=%M}QAgG+lQZW!c!406$z>xSxXrAl=Rm5oz*k?+i?F&^h7tD*hQbPR8D zf!LdY9BWrsZ*Lz?fVxO&v?CIB!2h02#P*wuIjI(c&&Al_RNlWAhPg;WrXV3sZV;N$ zaIXO-Ecp&r=Rx@*!&2@>5Ow3>{yHpg?qCIjmej$w-+Zf+!V`XEbOhgEi+k?Osb~67 zSGDt{lwr`+#B8j~(NF21m}&bdF|4W$6L|QMlB?6M;H;O_lUx?7&E&?u*^ z76P37mXd8#-^v7f7OdH8kJ9-oy)Mc0k!+_n>q@Ly8nU3CIile;-8D6AW-p|%q1pT9 z_YicOmbv8X$rFZV?g6=53V^jxumAn&n}%mO4vv#>O;M4<}f}4fr@nXV}bwNVkMkOOwAzvw)v5ibuo^keM64!dcya zZH?#YsBl`(%^>tqVs633*?zdPWXlwiZtzyp?5BR_yX!Z(32^zL77&v+brwcn^V6G} z0D&xZxn5Nc?02$Q>F{!%?OdBuF>u1*etF%_WA;|o)eVEOROu7WTVKHjZcOhqETyan zrOkWnmmZHFqP_kUQOSh_AAf!%zuSHvenuVRV=Sd%#@RVP7%sopE~N8yP=^PhWPNNZAWaL`?;pPb6{2 z81RGl__dVja6Q=L{A(1u_Adbl%>(l->iW#K`R9*5MF^0zO$U3W!_QILU3GR0ZxV2c zyFlcauOA8-(GGK014Z;Eh#`Pcwav97kR3wYOz3`_CBaeVaY!j1YgJd~8kvfy5#1s4 z$+n$B_jFeaTH0>}Ft;$Lt9rAgbGZc{1FDFVTQ90Pq0W@#t!K-FHQ6&;S3iM~Ix6M$ zcYD!Rc`r)2dSK3zRhN#+4&xslM<)!L-RK6#UI7}PMao>4qDBy()w}z3d;9X3ex#dU z+=agkY^YA;s0GCJjjBHxovyF+rX60z@J6YJhS>*(tHLIZWjkmXR2gt(iG*Gs{5V#H zm@FFTs{mOy4q2;iVphzR!x?A>tyv8ipwXvTWk99at)WArft5|f3}|v zKNJ!_j2I)m=0u7_J|nhABZUP$9vUNV5r2ou&;^L{e+Zc9Hbk5JG%A z7N!&)aiVW3E_E;q^F5t=U3y{BqpwL2l4}`d7CvbozkLMVzetUv-ITz7ODyZH5LOLm zN3&5S(tivE%d;T0AV4mzO_xM-SFhm0oxbP~`V{uLOaVh*)+Ln|FAjK`y6Vx10<4su zUC=lc>~f$Fm`k4tV*pM~m|sWd{3#GosLo#)3K*ww|Cozw(%QaFv|*7yeRDqF>bd~d z#=CMdo~oPa$`)rDUMmEvDge`$o?0`#P>{EpAmvP}2V&XGf1mx^=Eh>t2(~X7-$0Iu3TBa{uB}s zBJq1R5z;fF>inxTCY_pXv?~(P?^U2oY#1&ub!P%;b*Cff9H;MI`IB{%`1)eGsno zU?-F>(Q>`g6~-3KU9=H&2c{7-Wo=&?WUIWiTn68wa1UgQQ`D86u6F}u!_;`iZ&dm6 zD$Ll|$oYr2E?2j)MFz?rp3_iZn&GK!IHUEA41eA)@cD6Kf}2hE!h7X9aRzULOU+E| zr4xg)t%XexQkv$9bC_2BZkg-sK>N%$7jdgf%c>K%VMNF)AYJR51?%sW1m;Exvbhz# zbF*(b(XUVb8_IA&;BSiT7W$p&5*0@rYz zk`>N=_s4ZQ+g333E+0Q;3TD1%E!4J79htKQ5k*}-JajUWN80k?@wJYOAD!Jk4rjT# z&-x?0##K}6*3~FxNLdPum;5|Ba@+dcPs*Il+zfhU8efLVYVpfFULoelcGzj_?w~ma>CB&^L0Ew z*S^PP@1N;o=ZViuL`cqJYjtfTLs^#H{Z-$}w7~jF)081Id`0Pl9GIwIM#6UnE2tiKKh$|PPj2B6mM-a@@wvh*T;na2vFn< zf`@h-L)~kX_$bG6q4po1J&|%)W`9No_Fyi&oFlTTdRqkK6GOZ}2-Ia38N&&$rI^^y zo+w|5u6^em`J?@^EwDB|yp?MOPmcST-p1J4raU@J={sZhA*wiM5KA#%8Ga~ep)cZ_ z{x!bRM6L$v`FK{i;H7UQ;&L#LBZTCq10VO$7xE698j+p$G%qY@^ONi43TVf4rlRh#&O`VMo&3!l&$f>B2d!U^w$hj3BvaV`P*-iY;UO)ZYmp)F6VZ;QHd*Gx=4 zu!+dcGh{*M5pD}Q&!jiDag=+izIJjv7!GqCQsj9q{>aQ&LC@SLI63TDL>5*mQ*X6~M(>ajXk;`~E;in&9P?y{k1oiv z3w`k&LDz#p|1f-9{RNaP=_*8iTp#occL?1^b&A z)_g*N+%Hl=fQAoVHufYa~7fEpq3y&ZT;RaW}#)AG%Twa_^rx7Mm5U_@O>pS zcJtwg|Ej@+_293_mmkEG!9J3wXmAhUwj;i{*uS$4cUg8zo5~3`qTHR-lC8aM*jgtF zA8JYJ->We)+b^v5v>ww=EUIpyFC^WhAdI+LZ=f zMj3S{m8Wa+rXR#byr*(H;a%?nk?)+Cnt5m4cP#*dH==og2o()ENSBv*m=9VasLW<; z9C2PD?Ur5cV_6ea+u~E|a-b=|4;98p<84FiB&|Bzla$u&+~u!JJ=K9I;Kkb+C|3Y& z?P%*Cgd4+~hP0)aEzz9!#!kq}ECOSZM#|HsZjRLkirw{%R^>o8OL{TMv5~{bFS!c2 zn23Ji^<9P9FU`Aa#hRFc!J$Civ$66h_aUin4@eMuY>cqp^pHSTe= zEwT3PVTRC_5>UAQf{EQSeCKH9FgrY4u587or+#FZYyS_Cca&p{@Wfcs^V`@sk@usY5C3#-e-I%9Kn|V1hhsFnNO;7e0#c9X^ zY%#?4zi0zak3ws%S6KUhwnlPxF3KamU*0Fd^`#5fXDUSAh8O(0YL%NtE?>w7Z+hhX zYV$!b-h)qwU)4}?kVU*&SbNwty@o%shG5ulJte&zk9G^lnKszI;4=} z!<|^8vs(D+3H7zL@|9Ct|nP%ttJBpz`*$Y&AAi(z)0=XlIbC zfjd)v$mHe0y~|aaUaHZiRBnZ=9c#cb?Y;~bF$}S=O}yvV_<{d9nlT=##`tj9(tVJi!+fSs^;yOeDekF=@@@=8xy?aT&>VP_?%%*G!q zqNM_z@Wy+zMhA-h6l@>p_EG@Ai0;`*C@TFA@b(dDLGdKIZ(hW@=$xilp|vJ+@eSUe z)0AsG`W7eBw?uQ-p*DxsLg04>(0z%EMe~$+UAaGzOrEdiSud|~C2{3MGW#ZVkfRd> z5TMFqYFwm$G_#6l6D_docVXLjb@WkK8Wk_8Krm?YKzW>?vipis96zL5I$Soq&gFsO zk<$iqP56Klz==E?ucg%R6UOQ4Z-f25d75ZUxw2BlvGLZ8Objvjy3-tRZtKM1>lVP zj{_UB?)-%Z@6i)o=9!1QnZBBfoHWpn;$QK@lULq+kBE$smV9;h~rd%ipVI6jov+4+E}y zn`>3kBmIgO-@h4(^dR-d#Ke>!6%|KcpHyMFX3ao!mQoZm2yDi6fVQa}9D6-s|DIo> zDz9{6*7ah3hWV(Sn3L1a_TIakPJXS^aN^`D0q3gK8$5BKiQmm-NgDS2Ic?9L<8kiP zqo7%qCIm>Vyow|CKYV!u<26{{$c3`}TGU!8W3x2=8y1?KigPhYf+4`o-%jeC%pG52 zar<(^e%E2;WT$!x3NQz>K@1<7sd~~Ji(p45)?*4P)N1(BRmtUQXQ$*b)UqZfrTrm~ zh5{u=8I9r3r;1xw2iXnYC3JDuvlkj+zQ}_Ce9D)y85?tbjeCv3RCYlsY>xg^4Kcto zVfv+#nw>P(8Gl>v)}4Fjj#;vWS1e0#(D+U9REx@Fgh@~5<#FKi<}vQ*XeG%_{t6Y` zFj2DZ_8s-B!5aMyQ<>MAmXcq- z{2eGAFb_6XStGI_7#~37;;Gruzz*>hY!)kQtvd|Qx>5xB-V1WPk=%Rv3PoGyAzh4+ zG@{5)sUZqp^U0tvdK~z98fUU&uH@f@6+AAs^-Qt%u);Uj0T&kOI56As^m|?@Ms|kS z7`+J>C_$%tgx{Z^4V-`T^rwmTVd)=T!jLZfk((UvW9FD+cuNoqPwB9oo2_AGJJIQ; zQt2Fo;ys(h?&ywIjsd`L&-iLOaBV8jTMe#GT2aek7i?#d146z;D|S^;;Wq&$UmZ-K znHH0Wkt@zgS0Uz==ejt=fNAUxy#-Wji;l%OZ^$(?1Pl$@$ud`;Dd|9AvI;j3l+$}HK~h=l|mydK!~Cc zXWPPj#zl$=FwlnK@Q`!*ri)g>1@Fi=WEgmOW*N2 zi%j;tYL${kclY+JCLd>Bpa zJBhMbJWKX$wo&eCrlQG{&6Gz+Z{sc-L|^uQFLEWcp@)?K44kuFtm2Yw$zG{|W}iDh zvpzudu%O^C8=+5ze$%>N+k+dK(;*B;qIPQpx_GnyVR|zmDo&4oY5IO>3d8#}y&zs(_DDaGSeZ`i`?~zJy~d;1TEj`C&+lamH*HhFuO9e1 z9N`RHS{NC;KPXS#tuC4L_#5}Qwst$jqo@38Jt{m=zpCMTcU_7PTrgxKGu?=y?bZwKL4 zGZ(gFKu8B8KSTIG87n_$BBjJ6MlBL65**Jx#@mH6H()Oue#jeuMg5AYBw+N!@tO9n zj8Iv+Bl#wlLCJDJ+xqNyv>z3f@uCJ5yZa(J>8<^QmT=B{{s-J_=!saN3F7051GO4i zL6qS{eh#wcxTaK9{GaAjm2?=;RZ)vFi>rEZx`+xP0bG!Y^51Q2n z9P6j2%d)b@LI;##gn=}Z1ZlP*(NnFD=L1c80uu?cR;I_6JzL==lLwJ03y|ME(`yO9 zM8oNRYwqjgb9va;taBegmFUhADtjv`F{mYAd}!~X`6Gic|C=GJK zx^A1Z8jVtO^J>GZFnWm^q5fGs4xfMVlOGQClzHO#9t6EA1#x=(gWmwbRivoqO->BS zc{=&aIp0_**I-p7*{T)D?4s#yUp9u-btd#|FtOg|T-%j$jh*u4+5g`LYPInqf8PBv zWyh5}jq&OTzAcvi@1XkJ7#<1hTl2IJviLP6&?Y;MKR?l0SzeUkxV;VAJ+U41WpgHB=h6|H)Igr0wj;}rGf?nP# zY)ok2NDtlmy%Gn$!}ewN+q8SI!3pmaUR;=W}ghjyDo}+@*qZ zti`YlU1LGV*E6NvKW=-$Dr z97cX`Pq8L^h{t#GaB;UA&(&L){6+!(qP#F%Pkb#SrqpB#8mMi7;R4ORK!?9y>f4$o zLja!13PyhxpZaw5)Zn2;f^Q{iB`|J1F~{3ff7l@mtf1+kr9yMM!6Jb1G_`kDS>tku zE%m$22{olmTA0M#UD5G}e3J9{0tkSBD|(OdmBrY98iAcIS7 z#%|Vn7Djw=P5y9g1PoiZ8+R$!(fP#sVG1*xfTBj59Ke|IsjhQSRf=5QovtJGv6KRS86R#>x#lM z0s_q-GE-Ed606^Qc&mDgJcTM=Z^;&^GicR6 zv1$LU+;)k@l&=eR;ucUj6j%6iy67uS0hYSuBY8PXNvJy@cvi%6W>IBu$nfmGtT2SJ7~s@Aa8j zlV+fKG*}}y@zKOS9W@v6QT*KRN59cbB{b9d(N+B=BH7@eXM|Tu80Nj`nGp-lAx(I1 z2pWZ3UC2O&Ajsi2uxi2n0zGjGvG&W6j6_OKKFG^$Bl_9%Vc@Wo(&R;;_-EU%!VFFRRvBTy0B2ND17DBir-g^+p#n!5Y|j+jONDU zRet}hjnwwsoOu@T!O3oSpr{IaYi4GdR&0mzv7JqdD?nSnVJVX7}FEMQ?*#R2B#hEuT!p1p{pughtvxAD_{icz- zUiaTdwNb|#n&Xn@;(vq}%48LqVX0>fQ~M8QlC3K*fl6}b5kp^|QoQ%tKU?HARSZ7eEziE}rVM}|=IJ;^f+2hVC2o0QY8W_of5?mb-QkD(*6SLnQvstrN`ra+ zW?@noSUGg0a89c^CMz}DM@ofD@+7~J>UtMRN)_%Fp+*~Z@Km*bO0G?XE{5#+TE(?7e>M#&es%M*DrIEt#FISd(@R2_%xbq2l&3 z0o6%%ZtLyugpeGnB`ny9ovOCRT1_`cH(F)WrYs2GTH)@!_=(b=5D)$>?gcs(lf^A2 z?awDQk0d2t7dczLV@^qX6bET`*?)Xm!E@JiCy?34I>#4F6D#q;cR}Vm^i}TOv#d!F zi&87j<`-X{cJR9{<<1yGcTv2i_`<(g8oSfxGgf!$>FI5&>Nb#^# zZ)0X4-23Uk<#Z)~=7kuxy@f^pS z6z!%CXm~(^7p6Q!thwjhBfoT4$Qfi{p6YD#_jY-ORwgl9-sb2P^|Mod3R?7o z{rs8Mvq$w34lK~_jh+~=OAwCW&Z6LI&b`#UQqe+T$ ziiRU~{ew0l?YtbaANEA6utD5MQ7ea}>+wc_-lHU6V!N) z{nm6QQ*YsNb?q_IIFFv=-_Id4Tn$mFe~ppxfFEv2Vfj=+HY7jtfs;ScZ{~Jp_kIr8 zr}ba%flio2$n3q&M;-DsXB@_e<~5~5XOIeUBY3`GR)bplWR@#7hsu4hIES;XYbi0H zvN!1l*k_hAz`Ln0G6|D|V(KckM=@Xfcfu$s4Lh@PYxH}xX3pl^rvtY#!sKS|A6^bN zJXM4@9X4FO+xb!&?r7)QFVF==S7QP4!{xK9HBa5P$yhaF^_}~b#SB^I;{Bq7=}o^e zJFoGUNiRTCI+>hGkK2oedpP;!Xg`0gt4N=>oe6p4?-#9ux3*m&099{xq|iTTb-IRp zRj%d>?HTQU5^a|;hTbUZK2bogeWLcq0{(Q>i`H$_agom<*ue@oy)L;jiAPUrZrB~W zkywx7sA)bX!aKbq(Zf|XNY}%R1{2z9$X4fn5U5mdFBF)ybQCnvR4fCzyK7kF+ATS{ z_8CbVWP}pG za%r%R4LQ-ty!08z6Vg1+*G<^PMsfALdAEy|Hjy5yW?pKwwSM?F(t?5iGh57-g7S}_ zNkbh*onJ=O`fCavk5cx1A--HpIr{x6&CR&%Wg;2kTX)lVwAm^H!ri&BmwOU)KqFO6 zdv=>t^Y+PKl(NUcoYr#kD1RMy#0q93+u?KY`)P|JWDFa&a^rf(RWXrfw6Jp!DJ&%I+(WWMPu zyihIoOav*ssa_`nG9D9iw}j zFTTnPQB_&1xc|r8TL#6|bkV{P0)#-&;1+^~;O>wRyF+ld;O_43 z?lS1i?bKV}uUp^$OBGcE9L|BB?!DJuYwhlciRDy02t>CZCK#o7ipR;s+2Ufckid}s zL?@waO-s+|m*eH1lj5KzOzLahoUMM%bN45y8taJCw)kYIVvH<%Dj)rOk8jT1q4+Zv@;h4HDBQmbGZrF|DXIHE=8JRILt(CHJjR#2NL zs5XbxT=*%jP}Ueean4JKk<9$_RgiDk@GUZ~pL`jQn79bnh{UyCKkZm-QjAjJ6Lo7w zIRY;s@3z72HpF#%B{u!{`w&fBYVfIR#4?_v#OONxC(f%xi~ikDc4eu*#m9e%W;R;O zwX)^Ulc-wB-SVm?R0$bxcK6R!eCQz0-UA z+%THzSrRB&MuRx|t_{Mv8^x4x?IF^juDd$=G~YvGCEq&vK5 zesn$Ut^geMO_rU+(6l#bX71XP_f*FMs2E0(QNev#ZNuTXn@*Q?)%teDN-|b)%|p_6 zMu(g)tg`Y>*BDAlO$43bO8d~)SK4*E$ruG)djpHFq8OmRiQbsOzf~K}We^d+q0aSJ z5`qu9nKQ}#+9x`^-vC{8Nl2s^i`ze?;B|=zj#1ZL<9&p<1PK7rIhyTR) zVCynVBWJPPb$3smX-ZTnp+SXOt%=@k-}s?V3aW!)v;$qFDMES&V}D2iWk%-OoQKK9fA)MW zi$1hbf^vOxd=;lVZ%Cfm|Bib!w=IN83mQLoPJ{*Bh8wSsJue7(ur^83((y&>OGhT$ zB|Ulxd8g;#fU_MhwkfYV{|l3*HjruKzbz99<}bhTLwJ?qRWrW1JvZr?Cb;ImiI1I_ zF})&5znUt(zx)F&h1)-5sc-W3ezwNF1hH!uZR3J6Dkd$x>&XZ8^~hw4J4ZY_@z^*t z?)DniN-i%Z;N#cE$u2IN_4$nJ_}R)UTD9e8)npaV`4xP!z&uae-g{d5pm9@V6DMu? z;nK4DR}qCJRFybARW-HwL+m;87XxzWWAuw}u$|c=#c~ohS8B+7JeJljE}z&vuvASC z!c`osqgfr3Kj#1uxoD*t59VH2_iDcl7;OBv!ofk>x7lIXIb#m`379!-7ixdA~dzzkStWy=T0?%G-WMR7=W~|Y;bPd zAImuZsQIBwhw4FpSNcdxN2_a9g4$>AfUtG28Ge(dNt;O+7u7m&Sr3?1EV zTS14_CSCi#6-l>H*yLQSY>Bpy)HjShch50ySMkKMMW=& zP~QZdJ|pYQ5+mj%CJLmZMV3}4R*NG24=4m(oqu%r%WC|ooJQ_aJiJPPJ^aNj=khB; zdkjYL+s@AXJHoOMR%H`xl?OZI(}-ob`I{aqA~Y1RKwFAtN|mCImS4;`CKVkBH1E(l z!Rk^=*BEtn6wSxch&Y#CPCzR$M^8|VSG>63)|ay3bG)GBjN&(pPEO^R*@s0dbtsDH zj5?T*LBY_mV6}k=JZ6jrt)&GG;X^k?-(giMlv^M z3d#J0J>#25QD^6;9Mk&Kf);L{$y1leBaK+THd`w43E7gb%7$v2bFl8d1{4r=q%qQu zpN`hYa`R6)02t>Dsi&DjZEPxWmY0bev8%6mPH513+YH~Ow55-oGSB@=|K*$3jML%g zmX+cF(ro91aA&1cUBS+?$Nu&7e;@}m?wtMj&LDX#TEronbFf~&Hp}@PD)0NY8iBH<dRO?5ME$8)W*Nv0>)Gu~i?@Q{KW%6B{DG zBHY(GoB9ZCNLUm^t>Zj~Sdw$K)`g^Hly5-_gsJ!`!=AlL^v|vROxXB7cx5c4mXJ7k z)dq1g5BDw(q_>s}f7`eBdvNYH(d`AxdGb`)GVC|%`rrk9>WV8Ky#)X5vqNTgHxnhR zUrrl&&f|8a*eGs7J2-{vx!?NxGK=Qi{o{>wp|iZ^gXan|D*p_1SV6NV+>Rk5WIze=YMOF!GhC5i|4*9XFM57U5GsZKL*-7IS3^bpzSi^s$^ipu8yYgRa+9EN z;IlC4`kgjys769;QkOOcW7qG}snh7J`Xh6!AK^9gA;fr%TbTG=?+;)2H#N4Ub_8qXAMATBip>&+^Qj&N;riPN|)Bj-b2$1l4iGQL1d5AmJnvc zX6Iq!4)3z>ELzmY|sD-#BMs`e8*+|>Dou1f#4F^2f~>PSYVZAHGADN#9R03sAAf3Hr)cZ0H z^#l3*`2j92YtdN1M@T&y#N zKROCl-J2@KJ$uu08Vu_<|{|*Qi8JuN97}Or*g5n(}g6Ar1<27PUoBW zR$7_Ad{+g4fmo?rS~`5!h$P+K--)>fJ`-@#B_2NorD>l6jM?Q+AN2{e7+c_KpJ%gF zgH&&hiDYiE=fCCxP@|U%4s6Dk4~XW^GlQbJt3@qfIfk)wBf<6z@#JuHz3D0}5Pu^p)}nT`z1HXaM9;$WuM{bVo>%-aI}23d4wR=B z&KL8avd9?8Oas$%rp}gxYTUA9J-;EZulA=!O9yQ&OyaPcv;Y0_*QV%bxTO(?5%66J zZF$PB?tDkutHIV+aAx)9Y7}R*r|AF2-wCYp7-r2L6T{x}6XS|!n3Ij?86sOOn`)w$cM715ggpD-5L}a?pe#5un_IvsUdI5 zb=z7Du-?Mly)%}E-(6UI#25y*QNVWo?a&iST;&Y?IP3L945ztzC_3Hy0npeXslz02 zhqsJBm0!tD6$)pPWv!iUt(SCh5NMD~yGbA36c05e%(1sD$n&@~PC_CiY_4y0P9)eX z93689*2@xA6Kz5Iz}7F**)ovd@&RT>C*}laK;)9fF!fYE0g{t(Ez~)M$icl9!L4VwfXYX#Az0>cGjn8@sKw+bHL8kTlQD4g$21wSoQU&SO3U5G{T}mZmpq!kx;N zwC*!b?&=)5S_$!hMfC4KL1x|AUSG*2d*fOgSG&{A7~L)R`>;LxW;p#E=>VmOfH6FA zDXIE68h8G2e(EZgELTHD_o#Vk+&;97(V0=PE(DO5@!^`G0cnV(87LjgvoF}CQG3Ty z2BX{RD~t}Aw+#&ScTB_p!3dxW24F(ymi*&laS0^}pPol&lTh|yx_1mny#EOdgcV<_r@3~VcbiR7HLbWkw|5l(+-E69#-shlp0Tv7GY<6|| zN>iJILrri#BhX{9>ft%#&hi*GZ6}{BZe)kF%M}meIH=NnItrtsfBQ&VG6#CHC9P+z z3~kY4&WtnuKoz=MN8pg%YTQe2mhHic8|a`WzBlY1n*XWJ{xs(HlX^!?OVTx3;#SaK zTS8k&u!RVydT6^yk&HrGoAwTxd-I!H90b_y?8fQuw+G@aQkv4zGETCUnrSScW@OD? zk13Gn^cqsG78aw?ACC zL-~{U08vXOak!gw%jQkh&YqjL-q~{8&drZtv)z20=9Ok=cFRSMWUjc+nKLuTDAwy5 zb0H{8eG@h&eotku#*8=#xSU`8bL;fZabvM$3h@gsi4Erc*2}QB*`?Tw?8CY=kS2L1 z*}wSAjWj~=-{H>p;nEVGRx54@GqGHd@}mBc)zE}+7jNtKO9&Hz)Leb@_SrYUeE{2! z`o((SM7MvM-ZO3W;7K#4Gk%Em%v$f*o(-!&47~92mvY#VkD{!^Cq~uD&FI~K$_=24 z24qIxb8iMTNa$k_p}p2wjz_lrv*jO0q{T!i#@lP8Al1w%s2ZwA|K?DY60vfG}1LhT(Thw*#rv4wyyvCw#CSKko zR{dTUaG+mmT6x>Hf1xWngPQ!}j6u`imBxNG`^?@E$DhsYNbCr5jwA!4!y5_c?7Vf? zE}r$GgL!YcT51E@$kfEo@AB`QKPJYn8OUIg$qMkaIFjEWPZucFtKIM2dJ>0%2}gu( z`PE`UAY~?V!0)rgqb>#S0`**WKm#H6+9^7*a7u(Z0)DdK_ov)fa}P{4?tc;c64EnA zwr5K+s-kCN=VENQu)Ydn70ijVR39;5k5Y|4UTNJ!)PM|dvchCQQxJ*T*6>lUnE>Qzsg}AwVs`pmxj=9;?vTe`qZCQ{axPvKss>tr4xW^BFHV-><^31cfCz9bpmAH6nNwC1J@M2K4w^JN2{Bt^GZ3Yd38<>jjAd+Q7AS?sxv*cDt=s z|Nis`8;H0;PKld552uU%-pr~SG9@;6Fp==~td)BVZRDj)$o}CRG0_#j!fGKnU!~G` zY6)GAjZKO$nwCmKLDDg@6qgf6Pjq-%;>ezvhfhTn9K1aUGlfaCT z4wshSk9}1}h`z!@X1uVsAKNq7`|X;BX-IKVddQk2WAS9cmC~9zG*cLM-zP|PH?@^_ zQIp0}IKM4BcS9V^{wI!+o)=Zxjd)Rpz+d8%ESEbUDtkRgAnYeIH0>$WqqSq$dty0w z;;#)Fr{RuH4TqNV01i7ndx-UB{_i6C+vA05fyd+O;dDV_`r_;r`<^5=!`CP%DED5Y z>4I*Gh`8K)!>5$)vz`JI66h0WDXRjc7gVuwLZ9m)%{CTYTtb%okwXQcP&u z?nbZBQ`1CRK>QWTDsT`kCF))U{mSdBJTJm1TF7S=w`aEyw2+oLgnK_L4*nWSJxv2B zE1))J$K!%;qN!JrF)Nv{?TMu|o&=>^JPLu{>0D1811082ZaWRkztRX z+wsp>R`kVLbkrN4#o^z6&_1N%Q;!*MmcBHO}E4;c{}dZrUZ?fR*ZAH#~IE8t6? zWQa7?iPXnO&t>i7mZq|5$3tP2c{M#F!w!H~wlmCGc;M7Ah(S?7>*XBgwFIDk5}hiL zscjb8-rCO5E9%YVNMN9o@^wR>AFp8k{{AdR!vwcrgHHu6mGbYA3`PqWz5q!iamv{z;@Y3rsqjZaVSI=mBP!=OeBNOjlr}Y6@4)uplIZ0dRvfFN~J4F(J4THj`q?=;> zQ;fXldwg(8!u9c>V5X~iD>b?2oqJ4Cp^r! z(Vz2gC*;DUhuXO5Im1pejt#oJxF>HK6?h&cqx?TKvjNGazP`YAt2ZEWVR3Qu@K9~L zq0mKls1CHM`zVLS_EXR#_gXbt&li<&DX#8VU3u)5?jjHtrV+g)p$f}pGn>MN1CTDW~gn@%@R>vxE3#4+h#JYiv|w#UWeb4)ZG@tP9Id73s19n z?)-PrY4JTkGc`eq@Lapqy{R_0y}ZeUHOp6RMw~6&+>@C(-5|dKF3b7n8!{2EdHGT+ zD1geTD|V6BcMeSg{GlD>f^L1S{0c1tqc}f)+!?*{ci8maG_E(w&DGsfx+LKKir`+F zTGB5i%Y6Ifq221AQ_p`Sd>;{98x2Rm81XX7T58U@NvGz+VyLdJF}$}Itygk}mH1so z-z#m0aKJ%6@jtNYJP6)7U+T}{HxJ0j_O+Cemy>9;PyAR~;CIFyNZ2~JRtxGlyXoy5 z^cA#MiXp%WGd@wf{&_)|jDG4n@pqpMUL@G;y}LcMO=|E(6q2AK#4FWE! z*+e@TKvn2gd|bBQP-vr(FNOKsN451Or3Fdy#KQ44aM@%{!T2_pT1|E)dPk5rDoS;w zVWyDr+W@ZPl-7(P-AVv6Cr+Nr!2jYKQTztb2=D0`{e;SeL{lyh7NA>t*|+}i<5xRY z9E6Jz=W#I1i!pea1sXnqS)bb|v^c5a2I~6MdCTR7i(M;$);Ghs=3DuOJ-;Fm=^(R) zdD<7!emswEb`CQ}RqR!))eKPOG&{P0?rIbs+^Ox1xbzMTzx1u{2f>$shMx!yOk7ID zsig~j7AD``4pkJWu;L*5900M3i24ui9Suhyk*t7;z(-(-@C_A!iI|{`dUC?C5JbFO znTpPI-&_^THQdrodkNDA$r@v!i9a`JZuuw%6u|Wj15&F2tNhLe?zS=|i>4 z7zyxVzKK`Q$W&o_IeJI^{$LfqvQ+mAaI1WyocEiSE+VHAu zY8Gqe7nyxAhuBDSZ+bytCL+HrIZbLo3etjQg1zmn&u!dfjCN_-)iDXbu{VvrtAuH; z_^jLL%p`i=}= zM)w5b#`4u|Ecexe$agcaU6v-FZMUwb#W&1Zuasy}CZrQn1+hs$4xFpppQIfOX2GwU1JHwy@~6Yd7g zN)$e*#27$Yzcy z%<&w3-qQ!7a&RI7{{7afm<#fpG0?oPG|wcT0H^fadXmG{IvR%=*Z%a`OJ{5ILv^d@ z4b{a-eBtR_Ol*??kkb1KEa$3iw)K$42)>#mq;7`|JlUST=NJhG{L%-zGvO&{w@w+W z!)&HrY-fB_{rdIdD<}7+SR)KX0R6fGmAP&0dgywY@4Qc@VB6&=LltTkVk4u_dq|R< z%<;Vwb93u*S9_7Xt}Y{C>j!Pz`1u}Ne+I%$=c%%U=^etWE8-`IHR)mE3b=ASt?3hG zV#T`A${yMVJBIk#EhHA3Lq)Mu{!0j6)g&8`iL0?QG;>0bg)ILrT<_xD$_jt#UVQa8 zd3q-Vh$s6eqke5IT_v{GHtqu`L=hIjz_O5luDN>`gEQ~ebiRFP9Y*zANOXqCWthD{m^fIWyb`I4Q4v%YA;1+^UqIr)2_|3E~_NW{} z4zpU*-hETb!nsDVN>qz{jgY?OMG|0<%&to}9OHK5t-e6xRi#-bYMOkS*nJs+)?E~d z(}$PQJ+P92Y}6`ywIRqlMw+>7_g&kp)_kze6qUEZMDRe;7;rMi^LwzYnR%!R|NP4B znP~9fr_H6(=IA>l%;`297=NX|S3dK5hmMc$mBjUdEdl;aP;3dsPpU*9UqiCJ$O+C= zd5Y(J=rj70>#uHmNnpsr1Xdfy42uv&MhUo4t0vK*=;wpGZW{pva|FNyjZG1IuAfhi zSftz{2YK>Fb`1abJbH4#;mlLJ$kE8S=7(vRcR*eG%O!}4hGuhrzdr~iWt;g{$^vq{ z^K{#O#NqZK5y%AXplE__hX8{YPYd6RF}(DBK9`v{dYfnDKOCZaG?^G#lM+aqfBpLP zRTmXe+P#=>wM_UG#K{%tl*e+oRo@I5y{%W5RP=epV&ve*xM#>1d2EusEUq^PBBNfZ~+Q zC>{@_S$`Qb(2q`WFR4jZj+c&>BuF4A(KlIkU3=+|nO((NU-^Rc6y_$H1vpI;-Q2T; zmJ7>0zoa~|kjIYK3WSn+w4u0l?!-xSgg{})?KU@q6b7hMIy;T2X+=`why<%U+f2B= z2;ydlnckT|BeG1l9@)`X01SOysx)o)2vFnR9vvTpt6AAWFoa7NYm=U#qD#;g~`=2u{lc2l=w ztCE!e;8|QLK*vf)Q0mw@XT4kPw3WnluF4jpI8^joVfHc+PJX|6hF{^MP;0NM8~Fl8LG?` z`5i$83@MSTryR}Y$SG@!;ubCz(O0Ubstj%ph_j=!9Au>aM&qyk)>C{j!r0s&JV~9e zXnrBZ=#R;r4+wTJV{}KnW^vWA1RF2kA{&zdk_?Yv7;WBd$^Pt#!lm|l`)sbY$Qv8g zRv*YQ)I`vmA)e{;^@KWhRmdO`({b1akpDofA4RSXm6@KcUVeNW7g-9JA^(a25kIwf z+!%Y!(1azv@BFm+@2Ch zO(O@4k0HE73O;%G*#=dSnz#wBoY8{S$~O0+Z^1MJ-Y_vgElXAcD?#IWN5{>^(zbs_ z)8i*kI+%L{>DTI>LyW?{U#fl=W(pNQqClyz;)I@)=d*R>R0)k;dqp^ffh=FI&J+e@ z!JH$$g8Gxa{;E=g{M?OiQ%7~Zzayj8f~wnSi`O2b1G?}*QF&flFtn6r3d!ICI^ccZ zR9QZj^WjH0uV;w)KcW7OYh5Gbg;Ma6m{WRTwhRpp{aX4cP@HrIM|-K;;{YJU&fc^w z-W!@%JzJe+v}F4>`A-dshn=~R@t$L*%c73aGy_YHSPq^&u%&<`2wGogSU*OmGT@Ms zwI%3g$n~07c`gl7cDDh@IwL(i@yaMXUE~^yGyiizf6J_EM$y)pAzyk^`DFS!+w(M+tslz_R(X^Pf{w2!LS?bC8{mH7o1CxSMMRdm+v3lL@Ul zfXmxw$rCKic+*i9NDAy=R&E^TukCgI-d6yT#YT(`UzkXAQe6y)%avwm)0U^GUh zcCsxfW2?Tho}aBS96khmodW7eX0#J$#5nnF^}$U*u7;$}@-f69OP-?Ju_s-qFF}*6 zcd;w`tBhonJneOmhlRT(1+d$Mr(8iT!|2Fvo>30hLY~8va0%_}Z+je%?=0|1weFSw zKJSs3YcIDeuaIvwG6L7Z=54n+g;hWex$H0E4UNa@y&cb5B{I8R0m`o3`6l zfeNR$W$>d<_sf0tH}<40E2exk#4z`Y*xW0`BVJ1vN6OtgyzrH4I3NL!uH}I}S(4G2 z>-ERO;r&N7sd0cOkbC<~xvj&bl-+4+Uxc-1a3J)SxmJIC7+jz}a&c47o$ugo$1}hx zdEX`X=|=qc&H-Og_*8*o)#-dFsvb4tv295WkHRXR%IU}W>>0k^mX&mILD(8@=W=f< zI?S@~pcI+M>u%ff`8W%_%if4>HL!^1HWDD}rJo1%nOwQkTWDB@w$ScF#TyWts&ATw zt;|({H9~$ej|ZJWL)%q`aN$4M$p`3d@7uH9oBhY$91C|3c4=DZNM2XMt7oCNpng{q zoT-|31Abpq;XKBI`s2OMXvO+k0g+|2gV0jnY$x)3tF^^M)%d3)RD$oqkVTfi#B(++B3be&n7q`yg%RY@Re_KUif3ac=yn5YgbiurLSTO ztR&D(SCTaQ{W>pC0E&B7yOQG0WcJ#bky#Uo0f;Z0c3FDI4HY{P@>Tx2vZO^rF({&T z{O6tKTseQ)wp^s4{RL$6!chBiF+|Ma=Pq=4jQ%cyipAIZ;jW8Y+9mX&S^ekNrF4I;YENn8k1TQqbws_d2Dc6`{qoAclqdH;wtZg zARI}KiMbI~WX@5Ji9f*Fo=}Cs|3jK{tZO4{-nK3knDm!tig$G;4t;~V9NYy1nM^Nl zk6czzrpmV_bDqyuv_K>!1Inc{0;DWs;oAAG6&XLKE%D*=!;sIbZ;pb5jA{ zG}*x4@-i1z-!jnUm;3%0&sOZTSc+{~wk^9n9-sotoXGMk)y#%a$?%XJv+ro)sP?zR zYNPqgds&1XMRho;MGI7^A$87bB_oTaHCm#4h!o>sKpGa1;UA#5f6L2jycxrKhwApJy;}sV=_yuZS@&cV{ zwVjz#tD`oH9wKuT=4@^a2-6DBLGfV=nQQ1@T0rg&$PBLc|7Dqlt)X?!9j_O0uIZX8 zKk_^@fc?;FJ$(3}Z`>vk7Tq#bTY{c$1q{j>jh^|KhQJ?8onW)U#xC9_ z)xPeEmc~wy(-lzfd5`@R2Az4$&;pYB z=^4!P%XB&=sMgZujS1+C ztDm~u5P%V4P}s@h!uIf{G|Vr`BLu>kLLj50hO|-a{9A$*pxj|Oi}^3&-#(7eAnlQd zfhLme#NQ%)(2>qr-^@p1JDrK9%MbaOKSKYtuc+yjm;TwGiwXCU~`BW_yEQOwFlBPQd8H<1^Z`90NLzbj#Uf z^iU(Dd9AVvg`&IE=GqD2!`rdrmLKxnc_7GrNuX{;Q9zSC{@BR1g=A;eL$@i{yF-pj zHs`;ew#VDHClUfVxp&35O=sfRpQ#fFrSr!${O62FeKxN+Di*x&174u@iD^GU?X$dG zLnWMSY#t!vS@yp9Sf2e_!SFvl<$+p~8Dv@*5d*>BU;fo*??v0&9~zJwD{Z>pm!$^N zJG7J?9X|{G23dS+OIV+^`U791$SaDE2!0d)jVaUE?ALc4Ke^*KmF1Wck>=NI0auQN zt37H{k&Rt~14R;|AHLq%fk-j(Sy5&Ts>06!JKDI^fCx_cvZrKgSK$Hfn^sC2moM&W zZi>yymY`E=tGV5jgJ2R4Cf5p#qjmFvUz;*8v%G3&G+@Gy22p0ZNpP8|- zcMgl5zT(`SgA>?M`D7M9-RO}$dH?x=<{K7lbgj{JucGv}Hv40XTRRpjw^0m4mP^5$ zcwxp-GD45MTm4+nV*xW|-9Y8mwX&4lg4eXYEbXYQ<)5!`M-@k9f8knVw5?28xK@gr zod3gN0!)(c$@#2zUB<5t*P?+-$9S%Qzh7#syEc+rGo4uWn+$_n+k%{t8PcdH*|po1 z?V>Q-TXp4vP^Ko#xEPT6{UfRo1flHlWH3Tdj$w%eFO_NS;Q{Yac6JzrXdJEnu|Fml z+&?Ii|JMsbedRYjVwyUA^3uU5{1F|);Z14DSeNaBpvKjN8G%}+8|W0gfJgvw=07lv zK{-%Ew3Lxt>iebqTp(ZQhd}%Qag6{@VH<1!zk2l69P?D`n_mK7-Yk(9*?hH$1Hvm zQ@`#s{-?eFQ!jxc4{VZfzg(E2Zxp%M?1#-0)m5S^pa-fye69zs7v z)}FZBMTNN3p1*AWt-lExUl`mB5dXFJ7<0q?8}q~v6O@8O74z|$eGriC#sneYzzUBS zFJ8EuwLmJWtM&ULNuB?=Zz5>|BiD%bXilHRiYe?{^FLor$DB_GJJL5V9S;!Q6lG`> z5#zzbrNP5%4_C-pt2tSWMxO5_NxeHvGB?JN!qr%P3pI5C_WR+J_cnWT<$60LHI-y{ zBt=3>>h;NDZF8C-Hn%SD4yy2~VqJ<}9DOoHiim_+%4>8ZMwUjm<@z^}f$@|$oDmh- zEB$BM0}5+cOK)==x8viq;St2^xBY>Cp~j%V&&~0yFE!?Kk=NG_$H&JiH6~c#WSx%} z58>fxY;0_oUUD)re;Z>$zoiK~V1q8r#HHVVxl81UtwFmvkJmnxkH=zyKDZU%Aodnj zYh2V2?O04&v{>3|t#nJ7T*jh>tA{ynoK)~&(J-Ua8zK%J81(sgxiOYD-NhFu4_J73 zN}D%|1#&$T6RfbP46uz6uFBgn9gbhINs5P@O_F`TEjz(!n~Ft1VE^AEF#Hsre;78& zg%1r3540%-H~&b2VN_I9WNNjz)>tgW6e$*9laLJKYr89eJLk(#fDhQ%Lbb7VQG2#a z?Gm!R(H5`^OiKOTo&GZgydmG3J*Mm!k}!WrtLmk8*Iz17R&W8EC&}}bO3_>Zvw--? zZO6{{j}#vZ)XNqVW%5Ndv0z4~rb%rMpH|6jA8XTi2taq_!3g7ilh(Gok{Vvh+H&YW zKU1BR20ZoKY_pGyy^`+qOH=P7O3F=Rl7}>Q(dD)J4d~js_pa#{?kA zqyRDX`QfH(Z7m}=muh!7sjD}FShZHlh~fta*a_>Ob455E8VnQ$zQ2>`p#0sZNdqz{ zC~nvjFtj(4A`~Y$O0&c0Mf0dw%PtSE4n=`DHXJ@$eE|Ke>?{5fNN(c5esC+d)Hmx~ zH(}u-?~CN{XCDOXtij*&noSNNmHK^l8=Z(Gq@<=`Ef*`5W-FI#fq9*c#wW}PgjRZJ z@+EGNXG&yRfgms8LawH4SFtT=`YMa8+L8Z%hhd!)( zZFVewSIA)UFSM${rf+(QNYx4qQw3Gs^{=ImQ;j*hItR0|g|uENj2PrArw_|*reJP5 zeUx^OEmNTYfj`(RyHo9$2MY!77Wei7d`ZJzlr8=B)Bi5};vS2~|G&#bu>YTUku4Ll zGF_x-I`aqj_V(7r#f8P7A2X7a-w9+Cf6N^zy;RKXKR_C;2huu?Y;rgFkxI4E8&y@+ z>(iB#-@kwJ{}2N895*)bW&cQ1Rb|p@1T8ihUe7nm0s;c_&Cbke{LZbVh=A_9q+tb= z*v!f#L9@UAOE?V;4J&C~KqRt76VUn>nhj3tIWxu%blZ;!N(7;knk+ZihRhin82pNk zj)vLm>1~d9Zfky-b7az7?5osAm6Q{T3J4(OG{4Q1OYg_4;_%f5Z0ksI=bd zR{xam;NYS}^@<9g>G;!>#)Km`2V=xkK|-a+-=G6O2VGLAagB=}VX{?499Hj0#8_~-XVr_^+*cdodOpcuVLeS^$*$3izJa6p}*2!!$O27X$m)0ck?N85fE zhwnED+uHV_lFG^m_>~#M&mB?!k#+yu&28&UKeZi_LJ&zi_-QJ<{iaE>y1cQ@d&0yM z0g~)Zf)<#P&Q2*>OT&6P%1RJkNMP_HJWnG@lS&--Db*uQen5PRsx_oJ<2_b3Ksk*4_WE6uCkiKj36HJGq81TH1BSi*l87-8 z2SH?q4&1DwyI!JCzGdf3Ez0SsthIGV0Ols2hQ^CRg&f@EI>P^$cVCmvzuF$J*ytBA z9v)^uK_dytKS5x^w`XqGb$Ax|{260)R1Z~Q2NYi+E1h_ z`dB+&#Rcb^Ivieu6MZ;&T+V5$s_XZdI8x!a*Z1bqM$^{%2LIfhM7%&a>swwn@O$n|xJe<50G5OH|e;sSCO==Vi( zDD!PaWsy&W?3>9+rf=gbS-RF|etCI{MUZs<`iPsO3|c`!F4+7=SDT)Zk;GD6`FP>P zKK9wju#8?`BnFS?le|p6FQWR%i8&VbCECpqMx)h20Om|am^4Blx~8bbG1h3tn$N)R zfLKM%r{`upmfu{(Ccq?*b@<0>jVdd7R2X4I1gY?W3kRPW!<+MZB_ zkB^tzI$MjtpzP9`-}ye83GzF84$l<#T}GKE6FMH<7FaI9^z^qp{noty_mBQ=Y_v>@ z&Fb&31u5la(#nJn&Ztn;x;74PB*TlXdrio2<4FLl=bS)$j zd^iL!b)YNerGc?=g~fsk$gfdl_3+6#c-n2EYKwApRvu~F+a_xDo<;pFp!rM~M)Irq z+?IV>umAMZ`--TcgnD*kIc4#vRsVnh&FIG{a#(Bxv9>5u|RDLdMahzFF#G;Hj#s%_Ap4uWt5K~lpE3O?nWO;8p>LcSC(S-M+jt$D_aRPR1- z%yzt)xTRMvuL`?PS2BL3YdOST3Azbw9nQRR@E{qdhVbHjV$^v#9-YQ7{Go~~4pU^g z&=1*b9ZY72XZTFyb>C3-*7lZ9*0*P{^u)v!vD1|{So3mjnJ~-C^MRJRH~WYJpNxzI zh`9u`w^yg9hb&eF5aTd~ycZ8Auz#RfMU}}@QB(U^E*XsJDDZ;CE}IH6^YQWbOy-_) zEW@BO=Z(&Q6EVCGLbiG+3k!6GaK?(mEk>V&+q=s&$?^DNpb-F#KOXnM45oO zK~xNFZ?)A5?jBg&L3o)n6`LI{9UZ>)zK4H9ETBTDCMWem!cb{>dHG-}?oX!g-z(W{ zbXtJ|I0OR9H8$@3M!4BV33hMkPYj6cV?^*qq7v4O4zLiMu&?||FMod%;cgGJ->ZAe z=X!tXvz%K|L;Zbt{N_`3Hdnk|1ZH*$*F>U{#6yh{hAh;M4tFmd@{{t;; z(DS1g72^q;*%HfF>6~r^d^(lpDep4w-Q$H3*ndoOnuAtY@koK-^#GE-TMieU*nNJV?09SDfyb-=~mU=&;UqLQj+_mkK~8z zczB&-K8JCotSfA1fCpC5CnqN&MDsaWF-O+KuUki-90_Y)E@qkT8W`itSi#}y%n0U{ zB4Ox6419NiH*jPwEDM=biH?o^nPv{E6-lr-kkMs8N@XG>REEPHDk(|*Pc|Y?2ip&X z6j?qecU0{)xkgC-#hf-}gB^9M@8H2&yWRIOkxR~u@8tX5q3i)$DWL&;)MuV9_1I(o0 z#6-bmFBqTG>C>+-L7w6fSw}vH>A{~%y^LQj+MaBq%)ci2JU10OTm~{7So}=uAJ;WR zOsjpa<`?m-X6O@xZP9HQE;KnA9|DbTUQ_%Dy1p)E(BupC_h;XCRh+)#!l9Y^_DzvF zOKqGe#}#$-S7Jl#&JgW*-VpM1*-AzdySg4*UQRD6_m|qlQt=)l=1jz&v9vneEC}bl ztF1PvtOUd39QF>+41avpW2)iAXdHO>X5TYeIiEV&Q`g?5A$)MsOUB%h#cqC$P3-2+ zsIG;qmUZNz(6VV;2siVRKxPB4Ca|3%CKkU+uSYaHrykz3kWekYkmy9@yWQQpfWE;Q zu?WU{T)z(AWKOrYAEp26H|XrbhvWaMe-LU9_g{hGP^Vv5TI%8E#)PkQf>ZVj8{3vg z*8K;%7YM?1H~d&4-^o+BT0}%dIGUiCX70r8*wbt>7+Ewd+ZPfe-=8C7Q}Qv19vn}r zBqL36r#ibRV7Pt>Pks4t`yo(EQRP z9^X{O#w)oDC@~M>H2tm?>3n2b8o8F1W$5Ea{}0M+y=A^6LIMb~_M);*{g0f-;*KHc zxp!6*G@oLZbLEf&6~@IdBh`F7UiHX(pTB6*XETgVC)3^!kBCsWg7nFK|3B@0RajKf z+b<#A-6Be-l!Sz&bm)LIDALj;B@NOFk_rqUFbpz?W8etb9i8nNGu7wDJKY(bR7%9NyCK#Z0vS zsVgod4obqnu&*)%H}?~>;&hzbuTBp2dxEtyLt`%9+FdN0lT$WYB_BFY{p*J?)kN&u zPm^U_{jaw3pMAI?t8lm$3XDzPKh5>l?~h$3z-|m>ESxLFC3~uh z;PHZ>y)TZ=DSS_DVP$@DDjW{M*p`^VR@O^h-(LpPT|z7u6211Wd2p5Jvrc&6D{fL* zC!!BoxVUsLPIE&ur;ocz08Sd-(=!1h&ErHf%Z6?m+gtQyBVhEl7ijManCvUf1~jp~ zs&YrxeLH3rr}5(9Qy2lAWkJ1jdI4pz6ldn9jf7P!KLgUY$w(kGV7Uwd*-t1C>D$gi~8uFsU0y0?1f4W+M z75`@TL>??ge6l1fkfD$sEBMf*MauVan)g5v91?^)rFZvq(pqSix3)1jINX|GiWs$T z1DT0}l@)j0J`1PXsnlTF**VUX5HN%08Cu4mQjZ6bed%E33MTAMi#Ge1c){0QlhNP} zaheiDWk_)|50Ax388YLEh#(Hy$&TGU;0Yg|q=ZAFva`tqOk`A^v^Jr6(!q06Hh54n zAhSR|RX_`Pm~NgttI}*GHQ>9d)r1K~c%iJ8h&s*e@RSR%UaMmVQt6xHCG-p>{{DCF z3`<1?V7jgJxuYa0`B227n!l+lDxxzV*Alfy3A9Zk>x4PK+YPTpM7@hCLKe{S)U>^p z)>ZCc1dDvUrN zs=h>)J#waijNSP-R=h76^qrQ$3BBfNg%plvmYlu4(bH8M9D){ePm6W-)LwjpQL|U@ z;kmp20R@%Uu8EqGzImDSch+@Pg|Zr+?+n%CwiacNurpFG53ON^nxxplP}??t3(q~w zdz7i)XzYY=tWP~c_F?mE2UVE)A=5DS{dZ7tawVN1Vp^f*B=;pnVR}Nm@}+J%3dReL z5f<~kj_Q~~gfEw}?H&cnv88r-tM^i-8o5LEAq( zakhMM-gow(n}!?eHKMVXkR?^{C3y`KhqQ#419ZK+-Ed8=|b z*@X;;7__=W!n{bup}tbQXQ%?VC&Kw@QoRb=E=G4ERxQ$T1H(1wPocKSSMh}Y-uen- zVPn3r_=?_k%#Eu?L9KGWJB?T`5X@|Bf2eW9cNQfvNbt!ljrznD-^16JdpYd0I8>}P zSRn(MB>!R^Q!Af#hRcVrhS0#+?w8$oyNW_o+y(v$z@KHs&(QfW{l)oCKyzw{k&C^i zFZ_@;^-U^@%|^M*&Nm`VqZ-p*TeWLq4(B5g7-X)jRoylOU8e)+S#0((+b;r!KSmS4 z^tIS>wpa?EapWSV`Tj5T4=k+LkA{x_WJf zq>OSj7loRvVtJN7PH^<|RPbJk!e;WwwR>_bX0-AvCDM`QO$^-Zu5u@OtNX#kW_n_5 z`yUwlXls^uiv5(^9TFze?!Lar<>k2b`dROs;-TU~8no$s<@4l@UD|Bf25>w#=Zed? z0jnE38(Zx9`r7x}aupENp<$!K5P^vZcxQb+E(Pnb?7|(%C%JM9l^!bv$X=)rJS`(^ zbL?SWJ^5x{GJ*x!J*VF$_~6v@r=B?x8?~2n2;nO!7zrUgE_O~5*%C3JAU#-k9u%js@1VFkUmE%SYC&u~tllcxV>;3}wK9{~& z^IRVpj)>6enwz6d;h~s#?#7U6FRt2|Z4J}NS|Zll2L-7q#-LJi3qRnUbKS@Mz#Eh+E>tes_7pP?&N$%*N_Sw0JXbNb7 zK1~wx?8dPadii{~^7{vRF5Vn=pW5FCD@xx?j=4FX3!u`h99Pp2*c~6f{KMlj)asaT z;sHRLvH(g7%YbT-DoNvvba9)xHI^m`x@FiZiON9Al)zFalJdei+4K*Khi@VIkG=cq5bgj~^*C_Tv)a zA`Ub9n<&4!h(pEqefI59H*P1}wpMrQ@2XjR89nB=;tMzl)NuQ24a-Q@P>2*MAA3vY zm^`%*qNAvI$8Ol=_kyqr358Um3AQLxYR`^BXMKQP1^=)VY9!gX^#L`zj}(kni3>ue z_A8@laa-v)jlZX9PoJ=RmfFK3t-^LNMa)Gek%;b{f&hl)SWV33L&mcG;~k@)er$b7 z@f8g?d3g)aEgjhF!Cgx`+0o-sb}q|%JRAIQL+P}i>MXE(M;KNr&&G_sWT~n-GEQ3% zl?oj0cVYW)DPV1F$K1?~{Q3>XV#?l@%OoO+3LJq6`sS5<+!=4(%apB*ch^{-a6~n6 z&v}<(l8hdQetBH`5kN%+VW$>-eZNEf`V@Zw2FSoitb|Y zM8#jPMnn$w%c60_r50y$LK^qq%wu|_baYX(K4*m1Jd;Ue{IM9GHR10JoCRFgVvcpa zALpz&AvTY2#2*RaPH=pCjxvOJtVc20ibUn_Ptw-Wxvll378n^#6*y>e)2H~Baf|J+ z0@yk+GIHR8jz-sH+vS;w3BTRYHx&+jj-Uj=CrV@EN%@XRozBNN@Xw-P(_98qF8jLA zrV9CO7DlLXtXZdnJyXD`y?qA(r4({PL6^0xc>c2Ru0QH&T9b5?!qf{5G~M;opnPA1 zv=Xt=ue9|Y`jVL&U^zbL+Z0l?gNYac==0Qc>Z3CweErObXrU+L%%jI$g=P~u-@XMc zYs+yP9UAObZ;I3~zWs@PS`Y&UizqrmyT0f5XkqsG~ zuU27Ke(SkO7_0aLx}#ol#$6AEgI}EZ2YhK3XmYw z32=^!JgojlsGDVD*w-TKU3n@55+U{uHe=&p8n$9?*K=j3K2HpFtcIGLzR?#5O4pyK zgC?ufp+f!yV>}eC?3|m+sjA{b?pZ72jPRq*(tGB7-Bp}-tdkG{DIdr?Loh4xi9UDc zDT%6SWTjj(;QqS`7->AGdiz$aVJLurjW;DCLd|QQiUvS!0c zpnpF}94^ABG&)(~#GvQS6SLK=6^#1&qYzfN5ev^)HPDgO>wVGoCeYN>b!@z>r&lWN z(zp$2Cs^RmO{p&sL!zf-24VPCHcv0@MR9o=++Fg#-?JgVt*a{z3SF}wTV|UU(T6Z5 zPtCIz2$wVcVTarB6HSykCNbJKV9;P3*4v8$yG!HYw1iu>Mbz3f=5RrnCntOoo)=bZ zyz8AyMK4JV>g^imfXAojColr)m#}@|5*TLxDzf_#mUF1rzpccJb|BBmLVZ`t;ht zDJ#O04+&4T-+wQ(5<#n4Jy43i?gS_e1xwI%mPj%5l{Y>7{%%?Tzw6JZbQtlfng9=8 z|8*rQ?e#WgdHE2qE|QX{pH|!slv(67N|X(sm$%Kxh=`twh)O_{zlOpXzeW*iy@XXb z4#6;MqxGElQkU<+aT%gaZc4rizE^@-&bmV|E>F)=oXW@bdZ%e1I}>5~0%3NcVjyEr zfr1s4QNIhjcg5jIv3vo;NjNO}S6oo@PhXS#>>{xzIIGA(QaY{b5C~Y{e~)+6UN3LX z<<>)aJ^c1LNhiYS5;w2s-fLVtrv0t5hZItx>grZG%+lkt&K;v-U&?CK3A*M-NWnd_ zPi2bdpQPDNYp@Clt%X!g+#NeCQ1^@04FY=VNBURdpyQ%alw6lT#?5!yf^Q-QGD&uG)pgP@)a*o9hugGl$R>@$ufJLHDIa z#ww^nN~dsREfYH;7^Pn+cb2ih1b;8SQ9QnnQZb0qO>>)n=Os&Mx8`q#^q@Cq4L1py zf6r)|9%;he%z(4Q%m^p)VYZ4aE8|iv@T!c3-@e&TUjjqpci#EYpd9&-i4`bXHf=$Xl;6=_vGm83BPBeY2t(i|k z=H`-CDJ3LaQW!iRdQyF~qcr~R=Kz0Z6v@-%oOdxlO~bMWn`($S%6-v<+LeEfHf#!m zjY0D!F0ms%NSx7UO2k9xD^UG82^K64U_!QgKbt*dnZTN)OI0q!7}uk^`%JH5>&`=L zgY%uRSieWGj9Lb9Ln$HyeTxqP1A?a(mR6*^74d^jg1F)luL(#P>n!68QPTH0vDjm5g(uV2~r z3)cpKRtPOBi6O@kT$U@Xaxw7qYN7=d1kVa%>Yb-5Wn7}745qV;DEI`C=iwNx2@xBw z3CQ1@H#cJ@rT;vJ?ebpDIihHz(LYoZ!kyAC5AQ%cYQh@lxv8;vixB-+e$>2%E~=eK z89^9gyiwhP^NG^9Pr=sBvZam9it2cg&}h|0`@b`5Y!qVvx;_McmR5APzBvmVUia4j zI-rEN^L5{!Ipm;OlRss68?jLNWE~R6d=061+tYBKUAy0@lNZ5vWWu`+gHGU7KU`yk zguTF<$aZ(8i^iN_k;b3Ms*T1aQ62fQGCno+7BHl5PIj+ikGxnWyk{`P->ndCqBoKp zubx-d1O`fzQ;K9Vif2*L`F^RZD+s|PODY0UWgsZs02boE^Lwo^*uwQmRbmlMu$=4_ zuW3!A>g#1*FLR|^21N`|4i00lu>?%z_^^{AVO3%X)&)cYQx2?Jx*Te59h@2RrKcwqCfpXTrzCe@Pk4}mW{J3!F<#Y4Q89;; z+g|2_lGl+upvSG8ccPMX9~0w(rYoJ?)!#1$8!dlTt7OZZ(>?YHeB!u?BQ{8AK8*Rp z&6ljmbM4gKo1}5*FtYB7y`n)~PL4&d4Pd2(DzQ{Le^?j>LQ3fq9%X1jw>Y%lJ;WaP z{@v}Xb$9JS_uD%-vTCmu8Aw^O;t0J~k-0Z}BCv#Sv$Z#@Lql+b(M}Ewy7XW?^X1pbr$PD(dv|#K_;qlzXcZC|=i_pE)n0FmU8<`|CLvC7!bZI{@8xW7&u`D3 zlJxejZOj}mQBoTDCFeQgcli!TJenb{IARBI&mC)tpPQ!FGTn&02 zE8@%rqwuLZhP!gPgP)6`{yMnOEjgGvOLt7d{6I3t1A3Xaq3pi-bG`HppI=)qrCrKj zy2LKDVWI{mDq+Ufe)%C(s8TemxqQM*bE)Kuk*X_D*YS(Ep%+^4c&B!`|NJ7h5~Zcp z0WK^DFoCS>){+>z(^CRak57j=4@_=iXQB{xl=N?dKZ;(S1ZgV%H1-OTaSR=4^mS95 zXpII8b}pzLSL#=&6hF22>{zIDh9_UM4ytw%lM(Vpa9UwMdX$3{wxPliUNMpW&JfyspVe*sE zd|EB&ni5y$Fl`vEnnuAW6faL)e=m(wXn&5@7&&mPn{H(jbH)W~4CLiQT3VLu+P8nD zQT#t5=0v68xU$%F8UQ4ViP1*~k!gzpR6D?ah5>bN*%N8Eiz&Dww?~XKPq`0MzxSDd zz>oBC!_R;YTRS_qqxC?qCMk~c&0o~ohl@QQGtmOM~2TUXtU!G}!=-&M5LQz-!|`E{j%0L8~{8^@W~h!gjuS!_6z36|;3uQ4vaYokRr# z?qG#7#t&$ltS}8)I&50A#=5j(uGXZQi6#U^DfQVY9Z#Ue!s_5~* z2b)h$cXMe8{4v&s_mnx}a$qw@q!3LbPTs%Bf|J~RMjea11US96C1?Tu55i>AC{R(0)&5s>M7B&GJJB^cZ`h*sCVR$X;BC~ zAaL6SypAFyqN9VGl4^LQYD0;hSoQt`bbgwN$CJ()YDhx}LHJ98dKX3XQboo7SI;%p zzo&b+Pg>r|(@;G}Sh?(~&DQME&Cakg1#tTgZ{q=iX+WjZz=@Ez0$#ZLGtyjTR|xR= zI=OP=Q$JRMBNxkhKSgLTqA{6->A?n(+Dn$k_cZNjeV?JGM#UHVS*~f^nC~n;#sHwT zC0twzo%=aEc7S}I;g^$5VciG#g(K$I84p?hY04ghg z>tP)*t#!FkEHkyWH3S+K;WE>f){^JJfKnwOBEoY-?>={=gZF-ts6kM{KOa8CN#SxV zz)IZ6a&nC?cjZz|nCU=zetsV_9fw%^y*T#4r24{bf6?Mr>Y<}U3zSGOQ(uhL)y};j zCy^D7>Q(f)^5I+usJs|DI<{Y)?heu7js)z#}3Nl@svx-NN}znt_>Kn zz;j7}vPL6npX><39@Qr$%&WZFKE2fj@pC7KPt&Q|D8Q!W(i`Dd8?3TP6zUxGC=(`f?U*v z3)#VVdfo_NRTo5%MRM73XVJ#QX-+RPePVX~ESBlkKJ4a$cY5ibLYyU~ zaDM;MADfi*O-xM{1w17vK%O7GOA>A+2n$n9x9*J~@9gbSv)Y-l8#%sMzfgvBP7tIn zA3s5JefZ#(<2?JTe#xey_Ouh6Ws+5slFa3zlE0-d<_HnqfAG+ax;_UhBF!z5^i6wL z7a%2vR*Xd0K(b|dm$B z9fqU$z+v`eh0s*vt4dt)L&ovZDk+#s>fw4TkC^kLp&?9EsYIP(A6Epo+xFDn^789^ zPDSe4ny+~Eze1lEr+rMLH61vgEj1|ZwKrjQ#G09K8p;T96LZD{db1Ma$~6`tO?M!s z8-cdrqn4}9Z)qi*>Ma#|=hU7QtPem>skeXE z<~k)jF|e<%=j~^X@|Y}=c#<+RIb;Lg^``Ral7nIaY}n&fi2?)f-UH>wNAGlA#RSGd z22oq0V%_q40MO3NRG*r8bpPIOi^j&8o}V3x5K3d?5;OT`<24J@!1(?FEl&Pz2~yS^ zZO2VN?`Ut5u&}O?y8S7UZg$pOQklQ4@_-bg)7RhX9nxtZLhz6<5_h*3_(gV;g|~ef z$}S!QRWY-ybW-UHU8{!~ODo>2;4g%HRIu6llrpIr20u$2?sv0R`tE_)RM!eLcl08! zdsw(lR`O{3?^M@9t$AK_fyaIMG-=4DKMg%1=Ihl&=h)Kzb|{C0gV#|ENQf+o+8X@K z%&~z2YEWe1g$3ACZ=zWhuP&5$>$w%p(DU>Q7~4BRbsj5Ec56&IIJAT@I=kwQjwG)y zoQPkP=U_2nvtTAvRZ(rvU(#^t@Z0`wMNUn9szTW8gJ9X#A25=K zA>JmM!m&w$OVzp9-HjO&>S_ zvF074&>70D8Sxne)WNf&`k2Q@tFtX!KJOOvm5*H=V;mvUj{E()E%)z7(ZT?n)Ugh=SGSnwmQmsF2kD3APbG%6~ls$WgSZOf-N`mI{)vHx1 zrT`<_Et*Brh9&E5;4bLt;Wj$BFD$&T6THja502~pp85HPA2KuJBUK!*Ga|`TRlH+dvur_Oz-rta zDC6!f4|VV!@s__wrTsV-(LJ8uIerJ1*4$zlRbpeGMN)4uqa%uGpM}k(do2)?P3}9z zqPtzBlig?PI)7fB)OSlC?4P)38cn?cmBA|M>T%J_c3+rfu-nJ4q$ucXas5SqY=>;j z?12GgAQOk(hd&RjtZ=tmACZ>|^2Y(XN_92-cB_Zo!{rh{mG>#*gUxGpr~~P`ze8{~ zJ?2wDtKN!O7JcfLV&*^bJkOBv^Y~8_y|eT4`HiF|>)Wu^e6>~OdUHWy?&{)_(m>RH zjAg=X#a#t&00O&Gm;D`E&d0{aQh8ehutGVF3qB6K_kdqmnIz(g;GSTygiei(Q3A!Q zV(>1fYJtkGAn|3-Y)dsah~-JQs)pH3z<+(&JvqbGtKEI#C7Y1!`x8?#3c;mcyW4ue zn?~*{FzuJmEao60>`V0BhFW`9tCiR5>b~}8ImmV7ZKH(W%Md*ImRq>-0x-lmG#a65!oZMGD##N%&wDx9cuZ8uVpo_F)NeV8~P6y+u zo7Xq(;T%hd8D?Z%+$X7u*-AUI`Fff+2B{LwuYpMW-^EsWZJE`Cc6-zVH+O>HBYBpR z;KkJEBor}}7?etAXKA&Li+;_IqE9abr60K<_)U|9l^f7_5uc3|3C#af;A;G6DZ7BR10i?0|Eb|Cb*WUL~WkK zc(en+AOT&E!9qk7O}KEO`Mx0m=t?m`cm+CWzIuA#I-GULiYi>0mxuw4L=-u7@oz== z5US)Ug8%>F|J@)+q)r7L4oaZl2H->^zL_eAcJNwVe33x(|5GiVN)cjUqq7YZ%EQaO zDSk@{68Iz}93XPb1zospj1@lCzjYS8_>Z%|7)%LW`KwP)@A^d$P;rjH^dB_eTnD^) z^XBY)Eq71(99+^&{q~YJwl+vfQPGs9fKw5hzE8Kxu~5)5@kh|*$*aBIo}POT9z3{N z1MZeNJB|e{KY*qh`3b9_k4@jIP6K0-1R%+|H3Na;fsTm@$nQzT9Z}p~jj9QZ5z*0f z0mJ_-`j2JK6)jdWmNUU;b5VQtj50AXQNr8R*!sUNQ2+7Fu&86#Rx48^vNwfm?dFY0 z=KHv~_K67^?eX2C?sQ(@#gUhn2fZBQ!o!s!(^d?78z6EME3#0>#yG?lkR1Nm*)ey? zHO^dOf+Q@=Ccp&uFb@P%^$p%80Vj$GhmVVk3w?sBphct6`dMcj-Tu)L|H)mThUXU$ zAY$?K@Zjz5{^brlf~hCz#)jZOb-n$_VE^_!Dh0g#`WTnsih<#MHvtX$;5(oEDincl z@3|N3muaPN@$>WBPygW9nk?&=2c2<^YF$`>V}Q?;0E|@ff1DbdeRDLA{SICO&(u>j zHGE)(SUnt;=i=aq1aD?8=74}r8JT=*>FDeGGL^tAmE4~$bl7rpl?3GQS>jGzeW`qL z#E%e!;8t&6daPF=MCE_qp~aquEaeh!?&iVWdwb6q4AM_~3N~tgJPNt>5=FqoUYGy3 z%jW-LpIJocZvRiS;nMZ|Pt%`Wn`4R`BCo~XOnuLf^B&JHDDe9ois$C--8nOp1$-?) z$&wQ*XqY(xHz@7dQyCWnzk`gb@TCAlphfVNw<8*nP+cwVy*0U#?fVN^*kq(~o2Ms1DjHvIBfHeyv9IshDnAo|vL_zb$ zBHi4fhP`If(N3fTXpza`vSyeX2rdY6-+sSP%Q9=Jd4ms5+WIQ8p@Ht=Je*7p3U1B| zEWR)eFVE}qHCEvG)B&_6Kn8USos zUnDD@y&<%u-I=Ymo3Hm~mVd__f{jP|-FwUM;&|&t8wZtyBCvBlW3e_fNOaXpu>SIn zF6*%`;#Zx!wXA@EHrm^KZ>%uxdb@Ji@J*h}&x4izV;mydbv#n$krxQFl3FLU>nX znu>~wuioAU_zO!+O-wd{e{obvN$J>OylBJF(D0(X<=SLarGevcZ5W-PT5y+z#e7tS zrxMJCtCe)iNpbD*!GQrU?eU}2h2WO_I1ZN(V8|<$?y7?`FL*Uwbk#KoR>Z}|j>*c% z9D&;h!wgs0TR>Pe!5>RXuFGGa0S{f_z`)?(F_;`Q(kDmbx?9e(HDyrB`MV%)P2v&L z>y38S@p!HcO#}r69V0#0Hu1>rqU?sglr+@UK{i_6@8W`GQoKpO1osFWKi*_o+O6+! z8O>LbhB75|{gV$tIApnqF&+c6X{_L)0O%Y$YU#I<8Y`%pd474hMJ9WsmuJ23?y_tb z#2E#SdT+1;t8cR!p%bz&8&yH_G&eV2pe=*V*QXxuH5N$Nk3hpT*j*yPW0ur+*1_-o z_hY&oC@0;OmGoK%SEJ(8g1&5zYW#Q5PoL05$U9vI!sjZ8mojI+!@-;o*&4~i2JT%x zukEw_CDb?LN-)Q^-kq<|0i)Of-m0N^Nl6Jkab)x##st+{^KPoz96qquKKG7uH;`>s z5KnZc$u>5`HK_i)7)fU_BH=Ozd#bFwK3n5F9&oZ<(tt#6qzjq@n`P=WNHbodqM}z# zr@zK^zewf{=g6AYR#c4I4Q3gD7s~?{T764P%k@FO&MC|ZU6FYGbrC;$Ke literal 0 HcmV?d00001 diff --git a/앙상블과제_2022254026김홍열.ipynb b/앙상블과제_2022254026김홍열.ipynb new file mode 100644 index 0000000..13322fe --- /dev/null +++ b/앙상블과제_2022254026김홍열.ipynb @@ -0,0 +1,919 @@ +{ + "cells": [ + { + "attachments": {}, + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### [실습] 배깅(bagging, bootstrap aggregating)" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [], + "source": [ + "import numpy as np\n", + "import matplotlib as mpl\n", + "import matplotlib.pyplot as plt\n", + "from sklearn.datasets import load_iris\n", + "from sklearn.tree import DecisionTreeClassifier\n", + "from sklearn.ensemble import BaggingClassifier" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [], + "source": [ + "iris = load_iris()\n", + "X, y = iris.data[:, [0,2]], iris.target\n", + "\n", + "model1 = DecisionTreeClassifier(max_depth =10, random_state=0).fit(X, y)\n", + "model2 = BaggingClassifier(DecisionTreeClassifier(max_depth=4), n_estimators=50, random_state=0).fit(X, y)" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "x_min, x_max = X[:,0].min() - 1, X[:,0].max() + 1\n", + "y_min, y_max = X[:,1].min() - 1, X[:,1].max() + 1\n", + "xx, yy = np.meshgrid(np.arange(x_min, x_max, 0.1), np.arange(y_min, y_max, 0.1))\n", + "\n", + "plt.subplot(121)\n", + "Z1 = model1.predict(np.c_[xx.ravel(), yy.ravel()]).reshape(xx.shape)\n", + "plt.contourf(xx, yy, Z1, alpha=0.6, cmap=mpl.cm.jet)\n", + "plt.scatter(X[:,0], X[:,1], c=y, alpha=1, s=50, cmap=mpl.cm.jet, edgecolors=\"k\")\n", + "plt.title(\"Decision tree\")\n", + "plt.subplot(122)\n", + "\n", + "Z2 = model2.predict(np.c_[xx.ravel(), yy.ravel()]).reshape(xx.shape)\n", + "plt.contourf(xx, yy, Z2, alpha=0.6, cmap=mpl.cm.jet)\n", + "plt.scatter(X[:,0], X[:,1],c=y,alpha=1,s=50,cmap=mpl.cm.jet,edgecolors=\"k\")\n", + "plt.title(\"Bagging of decision trees\")\n", + "plt.tight_layout()\n", + "plt.show()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### [실습] 랜덤 포리스트 (random forest)" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": {}, + "outputs": [], + "source": [ + "import pandas as pd\n", + "from sklearn import datasets\n", + "from sklearn.model_selection import train_test_split\n", + "from sklearn.ensemble import RandomForestClassifier\n", + "from sklearn import metrics" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Class names : ['setosa' 'versicolor' 'virginica']\n", + "target : [0:setosa, 1:versicolor, 2:virginical]\n", + "No. of Data : 150\n", + "Featrue names : ['sepal length (cm)', 'sepal width (cm)', 'petal length (cm)', 'petal width (cm)']\n", + " sepal length sepal width petal length petal width species\n", + "0 5.1 3.5 1.4 0.2 0\n", + "1 4.9 3.0 1.4 0.2 0\n", + "2 4.7 3.2 1.3 0.2 0\n", + "3 4.6 3.1 1.5 0.2 0\n", + "4 5.0 3.6 1.4 0.2 0\n", + "No. of traing data: 105\n", + "No. of test data: 45\n", + "Accuracy : 0.9333333333333333\n" + ] + } + ], + "source": [ + "iris = datasets.load_iris()\n", + "print('Class names :', iris.target_names)\n", + "print('target : [0:setosa, 1:versicolor, 2:virginical]')\n", + "print('No. of Data :', len(iris.data))\n", + "print('Featrue names :', iris.feature_names)\n", + "\n", + "data = pd.DataFrame({\n", + " 'sepal length': iris.data[:,0], 'sepal width': iris.data[:,1], 'petal length': iris.data[:,2],\n", + " 'petal width':iris.data[:,3], 'species':iris.target\n", + "})\n", + "print(data.head()) # 일부 데이터 출력\n", + "\n", + "x = data[['sepal length', 'sepal width', 'petal length', 'petal width']] # 입력\n", + "y = data['species'] # 출력\n", + "x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.3) # 테스트 데이터 30%\n", + "print('No. of traing data: ', len(x_train))\n", + "print('No. of test data:', len(y_test))\n", + "\n", + "forest = RandomForestClassifier(n_estimators=100) # 모델 생성\n", + "forest.fit(x_train, y_train)\n", + "\n", + "y_pred = forest.predict(x_test) # 추론 (예측)\n", + "print('Accuracy :', metrics.accuracy_score(y_test, y_pred))" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### [실습] 배깅 회귀 (Bagging Regression)" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [], + "source": [ + "import numpy as np\n", + "import pandas as pd\n", + "from sklearn.datasets import load_boston # scikit-leanr < 1.2\n", + "# from sklearn.datasets import fetch_california_housing # replace dataset\n", + "from sklearn.metrics import mean_squared_error\n", + "from sklearn.model_selection import train_test_split\n", + "from sklearn.ensemble import BaggingRegressor\n", + "from sklearn.tree import DecisionTreeRegressor\n", + "import matplotlib.pyplot as plt\n" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + " CRIM ZN INDUS CHAS NOX RM AGE DIS RAD TAX \n", + "0 0.00632 18.0 2.31 0.0 0.538 6.575 65.2 4.0900 1.0 296.0 \\\n", + "1 0.02731 0.0 7.07 0.0 0.469 6.421 78.9 4.9671 2.0 242.0 \n", + "2 0.02729 0.0 7.07 0.0 0.469 7.185 61.1 4.9671 2.0 242.0 \n", + "3 0.03237 0.0 2.18 0.0 0.458 6.998 45.8 6.0622 3.0 222.0 \n", + "4 0.06905 0.0 2.18 0.0 0.458 7.147 54.2 6.0622 3.0 222.0 \n", + "\n", + " PTRATIO B LSTAT PRICE \n", + "0 15.3 396.90 4.98 24.0 \n", + "1 17.8 396.90 9.14 21.6 \n", + "2 17.8 392.83 4.03 34.7 \n", + "3 18.7 394.63 2.94 33.4 \n", + "4 18.7 396.90 5.33 36.2 \n", + "RMSE: 4.594919\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "C:\\Users\\pinb\\AppData\\Local\\Packages\\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\\LocalCache\\local-packages\\Python310\\site-packages\\sklearn\\utils\\deprecation.py:87: FutureWarning: Function load_boston is deprecated; `load_boston` is deprecated in 1.0 and will be removed in 1.2.\n", + "\n", + " The Boston housing prices dataset has an ethical problem. You can refer to\n", + " the documentation of this function for further details.\n", + "\n", + " The scikit-learn maintainers therefore strongly discourage the use of this\n", + " dataset unless the purpose of the code is to study and educate about\n", + " ethical issues in data science and machine learning.\n", + "\n", + " In this special case, you can fetch the dataset from the original\n", + " source::\n", + "\n", + " import pandas as pd\n", + " import numpy as np\n", + "\n", + " data_url = \"http://lib.stat.cmu.edu/datasets/boston\"\n", + " raw_df = pd.read_csv(data_url, sep=\"\\s+\", skiprows=22, header=None)\n", + " data = np.hstack([raw_df.values[::2, :], raw_df.values[1::2, :2]])\n", + " target = raw_df.values[1::2, 2]\n", + "\n", + " Alternative datasets include the California housing dataset (i.e.\n", + " :func:`~sklearn.datasets.fetch_california_housing`) and the Ames housing\n", + " dataset. You can load the datasets as follows::\n", + "\n", + " from sklearn.datasets import fetch_california_housing\n", + " housing = fetch_california_housing()\n", + "\n", + " for the California housing dataset and::\n", + "\n", + " from sklearn.datasets import fetch_openml\n", + " housing = fetch_openml(name=\"house_prices\", as_frame=True)\n", + "\n", + " for the Ames housing dataset.\n", + " warnings.warn(msg, category=FutureWarning)\n" + ] + } + ], + "source": [ + "boston = load_boston() # < 1.2\n", + "data = pd.DataFrame(boston.data)\n", + "data.columns = boston.feature_names\n", + "data['PRICE'] = boston.target\n", + "print(data.head())\n", + "\n", + "# replace dataset\n", + "# california = fetch_california_housing()\n", + "# data = pd.DataFrame(california.data)\n", + "# data.columns = california.feature_names\n", + "# data['PRICE'] = california.target\n", + "# print(data.head())\n", + "\n", + "X, y = data.iloc[:,:-1],data.iloc[:,-1]\n", + "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=123)\n", + "bag = BaggingRegressor(base_estimator = DecisionTreeRegressor( ), n_estimators = 10,\n", + "max_features=1.0, bootstrap_features=False, random_state=0)\n", + "bag.fit(X_train,y_train)\n", + "preds = bag.predict(X_test)\n", + "rmse = np.sqrt(mean_squared_error(y_test, preds))\n", + "print(\"RMSE: %f\" % (rmse))" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### [실습] AdaBoost - 회귀" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": {}, + "outputs": [], + "source": [ + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "from sklearn.tree import DecisionTreeRegressor\n", + "from sklearn.ensemble import AdaBoostRegressor" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "rng = np.random.RandomState(1)\n", + "X = np.linspace(0, 6, 100)[:, np.newaxis]\n", + "y = np.sin(X).ravel() + np.sin(6*X).ravel() + rng.normal(0, 0.1, X.shape[0])\n", + "\n", + "regr_1 = DecisionTreeRegressor(max_depth=4)\n", + "regr_2 = AdaBoostRegressor(DecisionTreeRegressor(max_depth=4), n_estimators=100, random_state=rng)\n", + "\n", + "regr_1.fit(X, y)\n", + "regr_2.fit(X, y)\n", + "y_1 = regr_1.predict(X)\n", + "y_2 = regr_2.predict(X)\n", + "\n", + "plt.figure()\n", + "plt.scatter(X, y, c=\"k\", label=\"training samples\")\n", + "plt.plot(X, y_1, c=\"g\", label=\"n_estimators=1\", linewidth=2)\n", + "plt.plot(X, y_2, c=\"r\", label=\"n_estimators=100\", linewidth=2)\n", + "plt.xlabel(\"data\")\n", + "plt.ylabel(\"target\")\n", + "plt.title(\"AdaBoost Regression\")\n", + "plt.legend()\n", + "plt.show()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### [실습] Gradient Boosting 기반 회귀" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [], + "source": [ + "import numpy as np\n", + "import pandas as pd\n", + "from sklearn import datasets\n", + "import matplotlib.pyplot as plt\n", + "from sklearn.model_selection import train_test_split\n", + "from sklearn.metrics import mean_squared_error\n", + "from sklearn import ensemble\n", + "from sklearn.metrics import mean_squared_error, r2_score\n", + "from sklearn.model_selection import cross_val_predict" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "(506, 13) (506,)\n", + "['CRIM' 'ZN' 'INDUS' 'CHAS' 'NOX' 'RM' 'AGE' 'DIS' 'RAD' 'TAX' 'PTRATIO'\n", + " 'B' 'LSTAT']\n", + " CRIM ZN INDUS CHAS NOX RM AGE DIS RAD TAX \n", + "0 0.00632 18.0 2.31 0.0 0.538 6.575 65.2 4.0900 1.0 296.0 \\\n", + "1 0.02731 0.0 7.07 0.0 0.469 6.421 78.9 4.9671 2.0 242.0 \n", + "2 0.02729 0.0 7.07 0.0 0.469 7.185 61.1 4.9671 2.0 242.0 \n", + "3 0.03237 0.0 2.18 0.0 0.458 6.998 45.8 6.0622 3.0 222.0 \n", + "4 0.06905 0.0 2.18 0.0 0.458 7.147 54.2 6.0622 3.0 222.0 \n", + "\n", + " PTRATIO B LSTAT MEDV \n", + "0 15.3 396.90 4.98 24.0 \n", + "1 17.8 396.90 9.14 21.6 \n", + "2 17.8 392.83 4.03 34.7 \n", + "3 18.7 394.63 2.94 33.4 \n", + "4 18.7 396.90 5.33 36.2 \n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "C:\\Users\\pinb\\AppData\\Local\\Packages\\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\\LocalCache\\local-packages\\Python310\\site-packages\\sklearn\\utils\\deprecation.py:87: FutureWarning: Function load_boston is deprecated; `load_boston` is deprecated in 1.0 and will be removed in 1.2.\n", + "\n", + " The Boston housing prices dataset has an ethical problem. You can refer to\n", + " the documentation of this function for further details.\n", + "\n", + " The scikit-learn maintainers therefore strongly discourage the use of this\n", + " dataset unless the purpose of the code is to study and educate about\n", + " ethical issues in data science and machine learning.\n", + "\n", + " In this special case, you can fetch the dataset from the original\n", + " source::\n", + "\n", + " import pandas as pd\n", + " import numpy as np\n", + "\n", + " data_url = \"http://lib.stat.cmu.edu/datasets/boston\"\n", + " raw_df = pd.read_csv(data_url, sep=\"\\s+\", skiprows=22, header=None)\n", + " data = np.hstack([raw_df.values[::2, :], raw_df.values[1::2, :2]])\n", + " target = raw_df.values[1::2, 2]\n", + "\n", + " Alternative datasets include the California housing dataset (i.e.\n", + " :func:`~sklearn.datasets.fetch_california_housing`) and the Ames housing\n", + " dataset. You can load the datasets as follows::\n", + "\n", + " from sklearn.datasets import fetch_california_housing\n", + " housing = fetch_california_housing()\n", + "\n", + " for the California housing dataset and::\n", + "\n", + " from sklearn.datasets import fetch_openml\n", + " housing = fetch_openml(name=\"house_prices\", as_frame=True)\n", + "\n", + " for the Ames housing dataset.\n", + " warnings.warn(msg, category=FutureWarning)\n" + ] + } + ], + "source": [ + "boston = datasets.load_boston() # Boston 집값 데이터, 13개 속성, 마지막 중간값 정보\n", + "print(boston.data.shape, boston.target.shape)\n", + "print(boston.feature_names)\n", + "\n", + "data = pd.DataFrame(boston.data, columns=boston.feature_names)\n", + "data = pd.concat([data, pd.Series(boston.target, name='MEDV')], axis=1)\n", + "print(data.head())\n", + "X = data.iloc[:,:-1]\n", + "y = data.iloc[:,-1]\n", + "x_training_set, x_test_set, y_training_set, y_test_set = train_test_split(X, y, test_size=0.10, random_state=42, shuffle=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "C:\\Users\\pinb\\AppData\\Local\\Packages\\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\\LocalCache\\local-packages\\Python310\\site-packages\\sklearn\\ensemble\\_gb.py:294: FutureWarning: The loss 'ls' was deprecated in v1.0 and will be removed in version 1.2. Use 'squared_error' which is equivalent.\n", + " warnings.warn(\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "R2 sq: 0.9800347273281852\n", + "Mean squared error: 5.88\n", + "Test Variance score: 0.91\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "params = {'n_estimators':500, 'max_depth':4, 'min_samples_split':2, 'learning_rate':0.01, 'loss':'ls'}\n", + "model = ensemble.GradientBoostingRegressor(**params)\n", + "model.fit(x_training_set, y_training_set)\n", + "model_score = model.score(x_training_set, y_training_set)\n", + "print('R2 sq: ', model_score)\n", + "\n", + "y_predicted = model.predict(x_test_set)\n", + "print('Mean squared error: %.2f'% mean_squared_error(y_test_set, y_predicted))\n", + "print('Test Variance score: %.2f' % r2_score(y_test_set, y_predicted))\n", + "\n", + "fig, ax = plt.subplots()\n", + "ax.scatter(y_test_set, y_predicted, edgecolors=(0,0,0))\n", + "ax.plot([y_test_set.min(), y_test_set.max()], [y_test_set.min(), y_test_set.max()], 'k--', lw=4)\n", + "ax.set_xlabel('Actual')\n", + "ax.set_ylabel('Predicted')\n", + "ax.set_title('Ground Truth vs Predicted')\n", + "plt.show()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### [실습] Gradient Boosting 기반 분류" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": {}, + "outputs": [], + "source": [ + "from sklearn.datasets import make_hastie_10_2\n", + "from sklearn.ensemble import GradientBoostingClassifier\n", + "import matplotlib.pyplot as plt" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "(12000, 10) (12000,)\n", + "[[ 1.76405235 0.40015721 0.97873798 2.2408932 1.86755799 -0.97727788\n", + " 0.95008842 -0.15135721 -0.10321885 0.4105985 ]\n", + " [ 0.14404357 1.45427351 0.76103773 0.12167502 0.44386323 0.33367433\n", + " 1.49407907 -0.20515826 0.3130677 -0.85409574]\n", + " [-2.55298982 0.6536186 0.8644362 -0.74216502 2.26975462 -1.45436567\n", + " 0.04575852 -0.18718385 1.53277921 1.46935877]\n", + " [ 0.15494743 0.37816252 -0.88778575 -1.98079647 -0.34791215 0.15634897\n", + " 1.23029068 1.20237985 -0.38732682 -0.30230275]\n", + " [-1.04855297 -1.42001794 -1.70627019 1.9507754 -0.50965218 -0.4380743\n", + " -1.25279536 0.77749036 -1.61389785 -0.21274028]]\n", + "[ 1. -1. 1. -1. 1.]\n", + "Accuracy score (training): 0.879\n", + "Accuracy score (testing): 0.819\n" + ] + } + ], + "source": [ + "X, y = make_hastie_10_2(random_state=0)\n", + "X_train, X_test = X[:2000], X[2000:]\n", + "y_train, y_test = y[:2000], y[2000:]\n", + "print(X.shape, y.shape)\n", + "print(X[0:5,:])\n", + "print(y[0:5])\n", + "\n", + "clf = GradientBoostingClassifier(n_estimators=100, learning_rate=0.1, max_depth=1, random_state=0)\n", + "clf.fit(X_train, y_train)\n", + "print('Accuracy score (training): {0:.3f}'.format(clf.score(X_train, y_train)))\n", + "print('Accuracy score (testing): {0:.3f}'.format(clf.score(X_test, y_test)))" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### [실습] XGBoosting 기반 회귀" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "metadata": {}, + "outputs": [], + "source": [ + "import numpy as np\n", + "import pandas as pd\n", + "from sklearn.datasets import load_boston\n", + "from sklearn.metrics import mean_squared_error\n", + "from sklearn.model_selection import train_test_split\n", + "import xgboost as xgb" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + " CRIM ZN INDUS CHAS NOX RM AGE DIS RAD TAX \n", + "0 0.00632 18.0 2.31 0.0 0.538 6.575 65.2 4.0900 1.0 296.0 \\\n", + "1 0.02731 0.0 7.07 0.0 0.469 6.421 78.9 4.9671 2.0 242.0 \n", + "2 0.02729 0.0 7.07 0.0 0.469 7.185 61.1 4.9671 2.0 242.0 \n", + "3 0.03237 0.0 2.18 0.0 0.458 6.998 45.8 6.0622 3.0 222.0 \n", + "4 0.06905 0.0 2.18 0.0 0.458 7.147 54.2 6.0622 3.0 222.0 \n", + "\n", + " PTRATIO B LSTAT PRICE \n", + "0 15.3 396.90 4.98 24.0 \n", + "1 17.8 396.90 9.14 21.6 \n", + "2 17.8 392.83 4.03 34.7 \n", + "3 18.7 394.63 2.94 33.4 \n", + "4 18.7 396.90 5.33 36.2 \n", + "RMSE: 10.423243\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "C:\\Users\\pinb\\AppData\\Local\\Packages\\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\\LocalCache\\local-packages\\Python310\\site-packages\\sklearn\\utils\\deprecation.py:87: FutureWarning: Function load_boston is deprecated; `load_boston` is deprecated in 1.0 and will be removed in 1.2.\n", + "\n", + " The Boston housing prices dataset has an ethical problem. You can refer to\n", + " the documentation of this function for further details.\n", + "\n", + " The scikit-learn maintainers therefore strongly discourage the use of this\n", + " dataset unless the purpose of the code is to study and educate about\n", + " ethical issues in data science and machine learning.\n", + "\n", + " In this special case, you can fetch the dataset from the original\n", + " source::\n", + "\n", + " import pandas as pd\n", + " import numpy as np\n", + "\n", + " data_url = \"http://lib.stat.cmu.edu/datasets/boston\"\n", + " raw_df = pd.read_csv(data_url, sep=\"\\s+\", skiprows=22, header=None)\n", + " data = np.hstack([raw_df.values[::2, :], raw_df.values[1::2, :2]])\n", + " target = raw_df.values[1::2, 2]\n", + "\n", + " Alternative datasets include the California housing dataset (i.e.\n", + " :func:`~sklearn.datasets.fetch_california_housing`) and the Ames housing\n", + " dataset. You can load the datasets as follows::\n", + "\n", + " from sklearn.datasets import fetch_california_housing\n", + " housing = fetch_california_housing()\n", + "\n", + " for the California housing dataset and::\n", + "\n", + " from sklearn.datasets import fetch_openml\n", + " housing = fetch_openml(name=\"house_prices\", as_frame=True)\n", + "\n", + " for the Ames housing dataset.\n", + " warnings.warn(msg, category=FutureWarning)\n" + ] + } + ], + "source": [ + "boston = load_boston()\n", + "data = pd.DataFrame(boston.data)\n", + "data.columns = boston.feature_names\n", + "data['PRICE'] = boston.target\n", + "print(data.head())\n", + "X, y = data.iloc[:,:-1], data.iloc[:,-1]\n", + "\n", + "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=123)\n", + "xg_reg = xgb.XGBRegressor(objective='reg:squarederror', colsample_bytree=0.3, learning_rate=0.1, max_depth=5, alpha=10, n_estimators=10)\n", + "xg_reg.fit(X_train, y_train)\n", + "preds = xg_reg.predict(X_test)\n", + "rmse = np.sqrt(mean_squared_error(y_test, preds))\n", + "print('RMSE: %f' % (rmse))" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### [실습] LightGBM" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [], + "source": [ + "from lightgbm import LGBMClassifier, LGBMRegressor\n", + "from lightgbm import plot_importance, plot_metric, plot_tree\n", + "from sklearn.datasets import load_iris\n", + "from sklearn.model_selection import train_test_split\n", + "from sklearn.model_selection import cross_validate" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "C:\\Users\\pinb\\AppData\\Local\\Packages\\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\\LocalCache\\local-packages\\Python310\\site-packages\\lightgbm\\sklearn.py:726: UserWarning: 'early_stopping_rounds' argument is deprecated and will be removed in a future release of LightGBM. Pass 'early_stopping()' callback via 'callbacks' argument instead.\n", + " _log_warning(\"'early_stopping_rounds' argument is deprecated and will be removed in a future release of LightGBM. \"\n", + "C:\\Users\\pinb\\AppData\\Local\\Packages\\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\\LocalCache\\local-packages\\Python310\\site-packages\\lightgbm\\sklearn.py:736: UserWarning: 'verbose' argument is deprecated and will be removed in a future release of LightGBM. Pass 'log_evaluation()' callback via 'callbacks' argument instead.\n", + " _log_warning(\"'verbose' argument is deprecated and will be removed in a future release of LightGBM. \"\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[1]\tvalid_0's multi_logloss: 0.95847\n", + "[2]\tvalid_0's multi_logloss: 0.832184\n", + "[3]\tvalid_0's multi_logloss: 0.731164\n", + "[4]\tvalid_0's multi_logloss: 0.641056\n", + "[5]\tvalid_0's multi_logloss: 0.571726\n", + "[6]\tvalid_0's multi_logloss: 0.507286\n", + "[7]\tvalid_0's multi_logloss: 0.454933\n", + "[8]\tvalid_0's multi_logloss: 0.410205\n", + "[9]\tvalid_0's multi_logloss: 0.372194\n", + "[10]\tvalid_0's multi_logloss: 0.333919\n", + "[11]\tvalid_0's multi_logloss: 0.310212\n", + "[12]\tvalid_0's multi_logloss: 0.282326\n", + "[13]\tvalid_0's multi_logloss: 0.257165\n", + "[14]\tvalid_0's multi_logloss: 0.240836\n", + "[15]\tvalid_0's multi_logloss: 0.225383\n", + "[16]\tvalid_0's multi_logloss: 0.211583\n", + "[17]\tvalid_0's multi_logloss: 0.199289\n", + "[18]\tvalid_0's multi_logloss: 0.186269\n", + "[19]\tvalid_0's multi_logloss: 0.171556\n", + "[20]\tvalid_0's multi_logloss: 0.168245\n", + "[21]\tvalid_0's multi_logloss: 0.161065\n", + "[22]\tvalid_0's multi_logloss: 0.151371\n", + "[23]\tvalid_0's multi_logloss: 0.148081\n", + "[24]\tvalid_0's multi_logloss: 0.143843\n", + "[25]\tvalid_0's multi_logloss: 0.140169\n", + "[26]\tvalid_0's multi_logloss: 0.138303\n", + "[27]\tvalid_0's multi_logloss: 0.134058\n", + "[28]\tvalid_0's multi_logloss: 0.130884\n", + "[29]\tvalid_0's multi_logloss: 0.128082\n", + "[30]\tvalid_0's multi_logloss: 0.124975\n", + "[31]\tvalid_0's multi_logloss: 0.122225\n", + "[32]\tvalid_0's multi_logloss: 0.120298\n", + "[33]\tvalid_0's multi_logloss: 0.117257\n", + "[34]\tvalid_0's multi_logloss: 0.115021\n", + "[35]\tvalid_0's multi_logloss: 0.115037\n", + "[36]\tvalid_0's multi_logloss: 0.115831\n", + "[37]\tvalid_0's multi_logloss: 0.113318\n", + "[38]\tvalid_0's multi_logloss: 0.115651\n", + "[39]\tvalid_0's multi_logloss: 0.115772\n", + "[40]\tvalid_0's multi_logloss: 0.114953\n", + "[41]\tvalid_0's multi_logloss: 0.117056\n", + "[42]\tvalid_0's multi_logloss: 0.115412\n", + "[43]\tvalid_0's multi_logloss: 0.118359\n", + "[44]\tvalid_0's multi_logloss: 0.117129\n", + "[45]\tvalid_0's multi_logloss: 0.119174\n", + "[46]\tvalid_0's multi_logloss: 0.117789\n", + "[47]\tvalid_0's multi_logloss: 0.121333\n", + "[48]\tvalid_0's multi_logloss: 0.120375\n", + "[49]\tvalid_0's multi_logloss: 0.124128\n", + "[50]\tvalid_0's multi_logloss: 0.123394\n", + "[51]\tvalid_0's multi_logloss: 0.126631\n", + "[52]\tvalid_0's multi_logloss: 0.129833\n", + "[53]\tvalid_0's multi_logloss: 0.129069\n", + "[54]\tvalid_0's multi_logloss: 0.135166\n", + "[55]\tvalid_0's multi_logloss: 0.134996\n", + "[56]\tvalid_0's multi_logloss: 0.13912\n", + "[57]\tvalid_0's multi_logloss: 0.138818\n", + "[58]\tvalid_0's multi_logloss: 0.142758\n", + "[59]\tvalid_0's multi_logloss: 0.142228\n", + "[60]\tvalid_0's multi_logloss: 0.142928\n", + "[61]\tvalid_0's multi_logloss: 0.142513\n", + "[62]\tvalid_0's multi_logloss: 0.143485\n", + "[63]\tvalid_0's multi_logloss: 0.143408\n", + "[64]\tvalid_0's multi_logloss: 0.148199\n", + "[65]\tvalid_0's multi_logloss: 0.148074\n", + "[66]\tvalid_0's multi_logloss: 0.156199\n", + "[67]\tvalid_0's multi_logloss: 0.15898\n", + "[68]\tvalid_0's multi_logloss: 0.157612\n", + "[69]\tvalid_0's multi_logloss: 0.162526\n", + "[70]\tvalid_0's multi_logloss: 0.166269\n", + "[71]\tvalid_0's multi_logloss: 0.168114\n", + "[72]\tvalid_0's multi_logloss: 0.173203\n", + "[73]\tvalid_0's multi_logloss: 0.181871\n", + "[74]\tvalid_0's multi_logloss: 0.181307\n", + "[75]\tvalid_0's multi_logloss: 0.186251\n", + "[76]\tvalid_0's multi_logloss: 0.185765\n", + "[77]\tvalid_0's multi_logloss: 0.190847\n", + "[78]\tvalid_0's multi_logloss: 0.190228\n", + "[79]\tvalid_0's multi_logloss: 0.195371\n", + "[80]\tvalid_0's multi_logloss: 0.199459\n", + "[81]\tvalid_0's multi_logloss: 0.198517\n", + "[82]\tvalid_0's multi_logloss: 0.203972\n", + "[83]\tvalid_0's multi_logloss: 0.213262\n", + "[84]\tvalid_0's multi_logloss: 0.212185\n", + "[85]\tvalid_0's multi_logloss: 0.217603\n", + "[86]\tvalid_0's multi_logloss: 0.227068\n", + "[87]\tvalid_0's multi_logloss: 0.225914\n", + "[88]\tvalid_0's multi_logloss: 0.230099\n", + "[89]\tvalid_0's multi_logloss: 0.229018\n", + "[90]\tvalid_0's multi_logloss: 0.23464\n", + "[91]\tvalid_0's multi_logloss: 0.24434\n", + "[92]\tvalid_0's multi_logloss: 0.243782\n", + "[93]\tvalid_0's multi_logloss: 0.24814\n", + "[94]\tvalid_0's multi_logloss: 0.25793\n", + "[95]\tvalid_0's multi_logloss: 0.257366\n", + "[96]\tvalid_0's multi_logloss: 0.261762\n", + "[97]\tvalid_0's multi_logloss: 0.260774\n", + "[98]\tvalid_0's multi_logloss: 0.270632\n", + "[99]\tvalid_0's multi_logloss: 0.269316\n", + "[100]\tvalid_0's multi_logloss: 0.269535\n", + "[101]\tvalid_0's multi_logloss: 0.279374\n", + "[102]\tvalid_0's multi_logloss: 0.278105\n", + "[103]\tvalid_0's multi_logloss: 0.279826\n", + "[104]\tvalid_0's multi_logloss: 0.282811\n", + "[105]\tvalid_0's multi_logloss: 0.29269\n", + "[106]\tvalid_0's multi_logloss: 0.297696\n", + "[107]\tvalid_0's multi_logloss: 0.297028\n", + "[108]\tvalid_0's multi_logloss: 0.29694\n", + "[109]\tvalid_0's multi_logloss: 0.30682\n", + "[110]\tvalid_0's multi_logloss: 0.306206\n", + "[111]\tvalid_0's multi_logloss: 0.303895\n", + "[112]\tvalid_0's multi_logloss: 0.300907\n", + "[113]\tvalid_0's multi_logloss: 0.304274\n", + "[114]\tvalid_0's multi_logloss: 0.314218\n", + "[115]\tvalid_0's multi_logloss: 0.312988\n", + "[116]\tvalid_0's multi_logloss: 0.317589\n", + "[117]\tvalid_0's multi_logloss: 0.323073\n", + "[118]\tvalid_0's multi_logloss: 0.333026\n", + "[119]\tvalid_0's multi_logloss: 0.332652\n", + "[120]\tvalid_0's multi_logloss: 0.337212\n", + "[121]\tvalid_0's multi_logloss: 0.334481\n", + "[122]\tvalid_0's multi_logloss: 0.340022\n", + "[123]\tvalid_0's multi_logloss: 0.350061\n", + "[124]\tvalid_0's multi_logloss: 0.351676\n", + "[125]\tvalid_0's multi_logloss: 0.348515\n", + "[126]\tvalid_0's multi_logloss: 0.358595\n", + "[127]\tvalid_0's multi_logloss: 0.356737\n", + "[128]\tvalid_0's multi_logloss: 0.351512\n", + "[129]\tvalid_0's multi_logloss: 0.361591\n", + "[130]\tvalid_0's multi_logloss: 0.35978\n", + "[131]\tvalid_0's multi_logloss: 0.357317\n", + "[132]\tvalid_0's multi_logloss: 0.367439\n", + "[133]\tvalid_0's multi_logloss: 0.365665\n", + "[134]\tvalid_0's multi_logloss: 0.365745\n", + "[135]\tvalid_0's multi_logloss: 0.375832\n", + "[136]\tvalid_0's multi_logloss: 0.374115\n", + "[137]\tvalid_0's multi_logloss: 0.376748\n", + "avg fit time: 0.5514350891113281 (+/- 0.3701610138582717)\n", + "avg fit time: 0.010002517700195312 (+/- 0.009552237668971902)\n", + "avg fit time: 0.9600000000000002 (+/- 0.04898979485566355)\n" + ] + }, + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 2, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "iris = load_iris()\n", + "X_train, X_test, y_train, y_test = train_test_split(iris.data, iris.target, test_size=0.2, random_state=123)\n", + "lgbmc = LGBMClassifier(n_estimators=400)\n", + "evals = [(X_test, y_test)]\n", + "lgbmc.fit(X_train, y_train, early_stopping_rounds=100, eval_metric='logloss', eval_set=evals, verbose=True)\n", + "preds = lgbmc.predict(X_test)\n", + "\n", + "cross_val = cross_validate(\n", + " estimator=lgbmc,\n", + " X=iris.data, y=iris.target,\n", + " cv=5\n", + ")\n", + "\n", + "print('avg fit time: {} (+/- {})'.format(cross_val['fit_time'].mean(), cross_val['fit_time'].std()))\n", + "print('avg fit time: {} (+/- {})'.format(cross_val['score_time'].mean(), cross_val['score_time'].std()))\n", + "print('avg fit time: {} (+/- {})'.format(cross_val['test_score'].mean(), cross_val['test_score'].std()))\n", + "\n", + "plot_metric(lgbmc)\n", + "plot_importance(lgbmc, figsize=(10,12))\n", + "plot_tree(lgbmc, figsize=(28,14))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.10.11" + }, + "orig_nbformat": 4 + }, + "nbformat": 4, + "nbformat_minor": 2 +}