From 05267766c07cd6296f272e5e38296dfb639fca5e Mon Sep 17 00:00:00 2001
From: R-icntay <63848664+R-icntay@users.noreply.github.com>
Date: Tue, 31 Aug 2021 22:41:03 +0300
Subject: [PATCH] Add R resources for lesson 12: More classifiers (#329)

* Add artwork for lesson 12: More classifiers

* Delete README.md

* Add R Notebooks for lesson 12: More classifiers

* Update README.md
---
 .../3-Classifiers-2/images/r_learners_sm.jpeg | Bin 0 -> 40334 bytes
 .../3-Classifiers-2/images/svm.png            | Bin 0 -> 38653 bytes
 .../3-Classifiers-2/solution/R/README.md      |   1 -
 .../solution/R/lesson_12-R.ipynb              | 645 ++++++++++++++++++
 .../3-Classifiers-2/solution/R/lesson_12.Rmd  | 452 ++++++++++++
 README.md                                     |   6 +-
 6 files changed, 1100 insertions(+), 4 deletions(-)
 create mode 100644 4-Classification/3-Classifiers-2/images/r_learners_sm.jpeg
 create mode 100644 4-Classification/3-Classifiers-2/images/svm.png
 delete mode 100644 4-Classification/3-Classifiers-2/solution/R/README.md
 create mode 100644 4-Classification/3-Classifiers-2/solution/R/lesson_12-R.ipynb
 create mode 100644 4-Classification/3-Classifiers-2/solution/R/lesson_12.Rmd

diff --git a/4-Classification/3-Classifiers-2/images/r_learners_sm.jpeg b/4-Classification/3-Classifiers-2/images/r_learners_sm.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..ff8d2945dc6ec7751987c5224f1d3139beae8467
GIT binary patch
literal 40334
zcmd422T;>N)Gr#OcTkEnBTYI;??gpGKtMoxlP1j&=`9ci0qFt)0s<l+9i#-1DqTc+
zlhBKZKp>%nKngE@_kQ1<S7z>eZ|2Q?JK1C>+1>0p`#)!Yr|jjg%QXOtzK)&_fQ(cm
z$s_@Q%XL8RgAmVW0Dz$(KpX%7&;Y2&r~u@o78&3Q>Fw%&Z<CSQ+yKh|_#ObTAmjP3
zwgZ{y|L7m*3847zeMocuEC2q5G}nKAl7A%oA2WWu@}IpKlRuLGXPc_&KL!FWkpRiZ
z9*_WtzX#;SO?m0t0F`@sh7`j8@hi&=U(!M<pn>A=OWODQ*n5$eivVrF6*97aum4`i
zDarp8Dhdj6N@^-<>VK7nmY$Y|hK`1snvRi<j-G*3sA-v)85x-Vwf{}>Z}`7aq!$AX
zHO;>s|If<hHvlUQ8P^q6axwwH6;?8GR<g@}0FcB2%71((BQfK@6|yVj6qHn?CDGB7
zI>1>-%O@wlLRu^(1&N7d5v1P%6s(k2`DO1?T{Cf_7Vu@0dzDs1BY3aAhuw4%B_!|c
z7e!0Q!FipFTlkiUsF=8dqLQ+Ts+#uw2RgbB_4Li2m|Iv{S=+d{x;=CE@bvNz2n-4i
z2@Q*W9TOY(CO#oO<6UM}_WPXNPsJsrpUcWCDjOP`;LR<qU)y^7`UeJwz7PMHnx2`R
zoBy?dSl`&(`n~;U=kG51==kImbA~;?_(vCMKmQw9|3%sVfi6}OT~|nZKtcVFF0w1Z
zq)N_8LCG&mb@i?ZwUh5P0l8N+Z1>WN>U(Gf<xNrS&VG}096}1~!svgb{Rd_L_Xvyn
ze?{4U5%xdmLI4=ZNdmx1&I$kl2>4G4qJaOZNt3w%(N78gzo-&@Fdx`rU0>ms)#wgT
zVAaC1SMGe0BIi_q1kHFR-)!$);l9v}r*GcsUh|K29=KzcuKU&U8wm9R8oRny_w8I+
zE){F@wqGYi{e<zkKQVm%YoC>xO2<?z4@(`b-mN<CI<sP0nz1G)sXX1RJs|uK(hX;+
zr?5hfN=^Pem8Z_x7Z%u@>fEDe_qnLTfbQpAO55T5mrr$O1<6M`B5*E|Gr%te^`%}Q
zwt&L~BIQ(|L1s&w<$RrF$AgdONj;`K@oTMLQx-m;G)Chp7;Zl#2wwuCo-RdHVK`BO
zenF+X{>0JK-b;W!1kY}1Gzo`bZ`BQjgTv<NN4Po|t0Z6OCv4jmHVk8&uKQ>lHibGk
zs9t?Oq*B*fF`)2e;X}jswQ`*3Xzx<NvA$hPyU*e5jS}<^_I!3DjS9rvsx9~Gmp@2f
z2+kgQE5g)U<z$>?=a^o~`G?#iPw9yfNo(Pbuhch{#m(oXTKvNwL`Gf$dZ?iM6&{gH
z=-;!83KRQN0exD2jZ3xz>W0yK@x>sNtm8*>G~O}$@wY$hs8&dnlyy*>(dsSakp{ju
z$<tYw27Czsh}NZE0@i9*o9mJ<0WOySI;ctuO8+mW<%asaPryj0=?0mvRMg$FB?r+2
zB`Uco#nyXc*C|#24Cd2&?L4Jj4Y97cMu9d0&1h&~1ET56yZZDNDP5s<TyrN5Qa1hr
ztDlsqXtICMwFk?QVl#0FeQWZ74e-CJTK>5N;CVqDFn3tAyr0%M+S^cr`S5R8H4LX2
z`O>my74;&sE)OS!S8qM<o*;^OFsChswk8%rR$S0^C2Qxl>b>2E#f>tTfY7)~%_<_T
z87L1&gOv<UvDY&uGSsG?))jQGm3KEW{z1y1nIB-Ut<fwx1kO6;S=w4Kw6vg_u~Nzr
ztI{hEIIq^P3q?sboSW7_hu7;CtA-q|VOdH1Z|rNEO-cLw7b!U}^Ay@?QrwzWvLKX~
zct0l5Db3`~rY{%u;VM>m^&2P_#vIAT7v9^&KZ*C;$lJ-%w0zP+$zx2N=E*C~VI_F(
z2OY3{SMCg;i9b94Y|NX=W2-edRA5@6tS_d_wg}euTd+AZ>0DfNhd8pvTG^xzJegfS
zs+;Yez{o79V?r7T*<DmS#B1<0Y3Z7oB`;+!S<J{+)huKCuf63OukTZ<jD|g6v)l@)
zHne--X<2<JrnGplQhmOgDXgkG5)=uA_4J!a0?8w3R=|WUM>Zs~#pR$HaqG+LNU6q?
zy02QyUWJP=S`Df2v#x~_8{<OziDH-ycACh<I!Ro`-N0Tw10K;}i`M@h6DEZk5QS4~
zse_&J8)bu1+G}Tc)(kuJU~0&m+G<=Evq8n*d5N>4j#OPKdS+iy{fo94duTN6^+TjP
z5KnvO;GJ2oHs1J1ml<b35*%|4!?M-A+jx4c=LT~kM*(}Q(#2qD>0eIn50TD|62cv_
zata^iZm(R=Mm+KqSRZ$~1W^1UX@fGXF_DgW{d^ssqAAm#<%JZkY&;LXBFAB0H0zpR
zweU_uj`5qCT(;@Mb7~9&X+Qy9fe;f(^$RcDIC<FdVDnAs)ms^5b`OLH<`0|`C?5*S
z*|i)vr}8*Chm+@?$zVn1_v33r8Zkf6-AJAdX7XjI)Ft3`@0x?{j=V^Q)CtlfYcyzr
zSC49o$<C11%s{@qK@`wE>CVMOO&QzogjO(DG>kkbQ|>m^U+g{DwL!9d>Nr)Gy>9bl
zU35%LxUs9|!m;ji+#-;*OEe^Soin!33athIykq06l8WSPe3SMe>Yj9{`%PZ)tq;6(
zJy>&`F<uM1TCV55?#SuUHgNLL&HG{5q94ms&Z9i0C$3k%J#`IBz(O$UlP55&<alC-
zFPeViQEp1jnt^R>6DljlyJG1t%EoF~@{<_7?e}!a^t%*t54H1;j0Fmphpx&*W24u)
z`(O*UM7n{Zol5}iq2eg8x6P=cXu;ld{sw>BOwGKR!SdOWtzv(}!}7O=V1`$&i`${-
zh6*F-5-0V#*Gz$<ssF`K{xKx`FyA#3AN=TUhS!gSADH$Z1&&6Bli=gh;04d{V34!o
zC4l{JBs&g>`CRJs`4Zr<tn1>5sd^oUw7lx>DtjOpkQ^PuCBjR0bO}IEW6)SfoV<gf
z`$pl_z8e)jQF_m%1ika#u)VyS?Xk#lmm)OjE482^niJ1yU^k<|j&s42&sHBEO1U~R
zhGZGY^OkYUE6s-qjmU|e-;cg|Fc@hvLwAwAri;JEvu+f>2EK(G(xO5grgnISUUBeh
zD@>4_w$4B{Fdi%1+fwTfmNos({sCau=vt^$V357uaKTle<(FRd^BuqIA*cJHdP!FH
zWb<;<Qm;zb?NaVp+u*Lmk@5PS)kbP$+8PLPcE76w<xncMFb!=Cq(Ghf-EV-fcAHjL
zyS(7CN-CnJaBh!A1N-{=lo9iogJeQ#X-EBvV?;Bk(MI8sg?YN+;^X2g)wGlkZ?4U!
zFBAK^D6y(IF{mkuvlmAHmq^}%UMZ~lf!cAZQCeBC)rfSpHF84e?^|K2o$lQENamQo
ziqdh6neT4?)3Zv4h4kkRmdf-wGVf%m5sOO}jV7dv%SS1EMbjqYh7N?DAN+Vfwp_Xp
z^Ylogo6q1{N~!6Sok`y*J$<{{GsJL<utyC?TSLWq`K(!aUXly=H0Tl#_lqEd)x?3J
za@f<&@Kh$OZU5vhcav;W6+>3WVnit?PwuTuHoc^+7@^H=T*q8CQV?erB7-qOse5<H
z%@#MrXCe@jbnsd9BMw%6LH=OrZ6N_F%AC8Skr3iccjIatXsxgtWi|dfQUDuXVjZH}
z2f7;2dLVOxRAqQ<2rFAfEG~_4s#s=?n9yvIbyNf`jr{~iB4AAst1m%oP2A0O5$nlj
zgL9Vv#t!9=JtZNLWAup?J}Wz3rd<57H1261#)TdYs=P0(ANoMS&{Y6@jS8uL!2MoX
zL?__4y54|3D#&-cA)&pcV=lfKqZmt-sW&j)En&C-k5F^;S&8d_@>~L<Uq+Hc9nis;
z`!??|Q=EdMQ3q25=ex7ZH#18S>hE;P=h;tU%W&>c)#)xZPY{!3e~@K?$ZXSbRj5US
z<ivu_d$$7BvbTKo)%88{{36?|8Q*Rez9h<q_-_F5?0bV<to5@XciUC6#gJgDIFr=D
zuA38|rw*G;IE>iqw8n!a!7JB`+|sf%2iQ0rm@&(QLhwl;Ry;BrwW^ov0frX>LL_vX
z3$(xXN*z>{RkhrCQ7)&(%c8?1(WN1ivEj?oSGHfNGWtU{x3c(0=iF8oc<{_KV0rIU
zL*vhNs8m@%{#3){>WNknp%k{(xk?v#1)Jint{$Gs;)#(j?cA=xq&8aXSgq@SNzrcl
z5EDw_f&J0Qz<s)xiWOV!vjVv>%4{qeme;1%uJFkiG#t$=?FM(*rN41<b74#T{XnS8
z>+4-!pZ=)o;sv}2DzQ(C7VFI%eS+jPPV6(W;o+D@P^QyvxFPn$zX%wv<^MjitCvbQ
zgmPX2D0c{jGEiom1XLc~@=625>%fG)R@rgU9oJ;bp1=pvE*LXVmHrqd`%-c?4I>_7
zl~k=?$&I>AFF549^R!s!$#kxfPq1I=vl`beWOD++$OdL!VX81=XMWbTH$Fe&qP%g4
z=b+w<+aE@&_f10|VI@M>Q^~Py=aNl%!~5e4&$1r9N9G=>$8*!iZHazuP7KxIXS<@i
z6W0d=dXiL~7I3Fac*;h1I&e?44w;nLe3$nPFBR{4><e9Inz}rwgzuUoR&suym5{ta
zKZ2EnTR#_?b-dS!Qyl1AuEv~w8HGsmuzq(*c<_|@`)|MNSG7K%>S9hvM`4K1knYrT
z<OxCM(JQGMaFKm|4|kIIt9?p8Uq(9_fxZs8;x}btI{M1LeJz@i?JF0jHp|Pl8g@lK
zx=hx$4#CwPCJ)$;%4q(-M1hul-KZDO-+Fd`kuI!U(Qlqp1EzM!hX?xdFS0jjo5pbK
z`Xm&?gwUfA1J*=wR6*uu9n*@(#dc2F9=A~uSM2VOi)Cr4gU6?q$vqa1B!+2=4MiC9
z7;t(Y>FF=;A9(AOHI|nx6n+3T8so1*)*;Ao9IwX>0klzKWr*eAsOHZ$&J+z*5GJT;
z&S!U0CFia|Rgm55Yt`}f@sFq&9jMXK7;!=wXf1C=@P=?>v?;b`9jr{SOwyqA!4wEg
zZ=CyMM#ZdhSki-eo~0@AyWjm_iB!!0pcYtSfkEq1cF+i>sLY7#e0(ZO{t~c8HdRtr
zS-e5zV!&z91Qs_JS}c<ZQ-Zk7LYod!cD6H%P0ehT8)tzuB=jQ!TdPE*B8O`A))!K;
zH+J@`4QIOer{dBvaVqz_TE4Mrynj)~nr?PQPWd*jt)T<7X2{$d_PHXW)B|CW6JBd1
z4VOqXc(|VZ_&wt%y_`F8S}IEmDm6Ce9eaGS0~`5JVRjuUnLzcb(+H`ZblqFGs>O|S
zKZIJkY=W>dy?rDjgJJ;v46cL*#zOC|*Vr#S{+gfy-(!|Vef%K#eBbvM_sJo7<Txjg
z3$&K&ix)#bQO;#35JyQ8IlHX|fF55A5_MaaznlbGGB&M6-G1S{T3*Jx;`>dPoE<e2
z$kw(pEDXuS<7+B0qo$b1U=TOSz$v^0pv;>|h|ComZi<z_-SucG;6Yo)9VqEBZtXqu
zaecs&muKz1z&Km;E#B2qmqH%I!1WM!j4{Plp=~i$*bL@WxRln}`k!8$K`8%J6HmLH
zLrBbrr*=Qx9;6A{`P^eCcLlG!{(cEC$fQV7<4yIlIrVoC>f7Yq)1|*YS6(C5yw%+3
zn>ZzPHd~t&mNz%3Q~LY+*MdV>pX;jsG3PhZ0(A~k9UCkM3~dduoM`VpHJRQBp95?D
zXjAV!ux;Pm)WakCCsyPS!g+7$9wBsN@9@^5XZmmfupTMA@fw6Xa<pZYF2I{4`d$|n
z^;Mo-;&boG6w-HD79ZutOmj%_HJSaYy(9HSZ}1!_bMfkdr$LIm)ohHlp573y!rx=#
zE21$oW9rb@v^dAoPzi&N2rp!U_1JYgfq;>}fLxJeyDN0{O>K>+&TV+<-r2Pk-79~C
z!eTnEw~6*ER5l=V=ybTo(1Xfygh6ID6GzaSpL_GYlRaZg@4kKw$Bi9a0_3yIOGS0^
z5Xvn-B}fm0pgaEB>HzEubD!gNPc4bBQ`U!YJ;{^22(4$WcR$-MJ7`vq?N`B9)Zc1!
z@I?s=$NR-&+}Ks1KRS68?Sz#N%ztkm{|<J@&ETF&5~ggbfn}8quA)o}e_0M*0^oH>
zW1L1vV(*b>P#f^ovw11JcBsE^vn`F!)ZoL?6v<~U+gAGst{1NN9@-CtfUx3!_JL^I
ze#fSkC=xa_e#K$3w<3ok=5xf<V#ARrD?2Zxi6!_z>2h^OW;BvTPqHWd*`ko~mw>*=
zs7Z$`>iGIrOYH^3%#(COdqw3?v0!83bwq(&i7FTaffSr+i4h9m-H3L~d+e+Ab1lJ;
zpv|?_Xgny8sWiycmc1KIuJbL?Icq-qz7AHv_;!E0G0pio80t%ihI*nciFD^{-HE8-
z5?efI63XS|5~@CCgL3@kl^tx-`{k*do551NvJ<rI``E<lGn4zR87x(}=NkOR%kNB+
zo~NfbhAp>jFSZzx?`3RnCm857*`Lv!9}?;Ga5DIZgsMm?n;jqnk>3W(`J{QqbQiNi
z>|grZB?w1+$5=j~$yFJCiYxZf2fOh{%S^npOp6|v>77pzNdEfipf+PRRFgq$=0*C4
zVBJ?F?V<s48e#rrxt0_OOxZK;nyX6~sW2|H&g0Y6KVa<x``G{V)v4Ao;0Tf&@(_<d
zDvR#+mPxt<G*aU^fHqhsv_U`hzzt3`zTT^VGwXViQew&SG92FF$)vCC>;A$e_FhR&
zv6Eobew|~<DQdNMHMT7q&qj?oa7;qX$KVatE&)_fi7ZDZ52Z+fsnY2{M$>J?-;(gx
z33qtDht7#Xe`Y#Ct^jzADo$0ETuwjeNO0!b_`VbVxs`dId#0+c<yxH;FR`Dpujvyy
zRvNA_-FXS14_&Q46DB><+@U=W*dCMpuc|s;j0R@SlTGJ`GSd7_aCmZQuIC3`-iO7)
zIjI1$VUbskyu>LG>ampe;6+jaJJz8^ym6Z=Fz{aE_^wpXwjz0TP4rCd5;X`oXczbn
zHORPetd`}2|KiA&yWbw1(O~Z|7$~kow~tdwz+USCHd_ur7S2DlpxLB0lDXyBc+Qq@
zX>uU_f`J!ayg17#5Op^M5=T-uKb5Av7}q;ZjkHE8Gh_HVuRcEqXMCfG1ufU*tTS_A
z9p4CEpRBET0(%CQ=oYgdbkI{j7ZBUv7ethoSTM*j-Q}VPF0GqQ?oAN=c;$auN!Hn-
z?&d)$PKr39rG$M|g7x_`&=wcdbR5c*O)q!;$FCw-yxKyvk4xw&E$iSPcxWpA+BA_)
z)&=K?7o37au38VRYr5F9dq><HhbZP3#=q6&X<SfPu;vo-bP6fXvaoDUdnMZ-3%CgT
zpDyM<7yKvk7T$D9>)H-&M5Ug*UZ5zvzM=WeXlTXKZQ-Xa@sXZF{8no(Dyx5*v%iNL
zt=4DvBS|K2-8w8OQWPCnkX(fQBsG^yVOuA2a?SI3d%&*)Iaa|O4$o&t$2?rMW-f)&
zTY!ku|J{S<=G~<-1(gr@;<(F!nSPizrv!Ol`*4g}<vS?zF(k1I2n;&%1nuRQOu=?U
zJ;+NweJ#Y=w#hTU^}Y&gFR{M_fMnQtptkiydPYBE+>5|Vz(oBWK6Cc(Ed#A2!{Xk2
z$%QW6C1A%iK?hqU&z#ZMHbiL3orS#{e5rOJkqn7PBu0!gle*SsPu3lc>O>0dJ&&}f
zHv%5+R)gMAy;S#-h=T#KuZRbEpN395vy2HPaGsFy{_!)NXdt06*O-vwLEiDHHG+9C
zY%`?Ju(2ycFL&j9ONYf~nCJ7ag&u~fFGOA!_ND(JaCWdkNJ=CRmT}#`G9&?OTXn#J
znM^)Yt_4L2eP0UY5{j{l>SyvE1$jXh>$I_KjXM`I3#W(dqT&-iw-${xllIkG2=Zp_
zDf(Av&yA+P5&5x;8*yk*%Hg53@Yf!zQgHhB4{u-TP{=ZT5PumeLaquLCyHaynWzEP
zH58+HKXT=EWDvHLFmr-4v>DH87S?@u;EA`uZf-&rL3ZA5mw<SY*I-vluzBLW65iO%
zR(XSon03n}%HOFE)>6qlyU3=0D#<G6+$5a#9ZL=^VQwU-0PS*T>c7N2YD`&KxsF7x
z+t&|quRjhaGjqXjbdMECZ9W=*<C`(C&2IOhy}Nx_cuIUYKsR4ZRs9Pj-`ZMRf`KKJ
zB8+<jZ;b-pLo5B|fBiRD3Lo?>)D<^$20P5}wn54sUQD*^opYam&=Rkq^*6~#JR*dr
zhn_`iIFhi92WjVD2S3SsGU=p>W2M@i_xM9ab-p?7&k8)WY{%dD=f(>0?=P?~ObWoI
zGOk3-wZBvx%>%xm&beU>uBHyE6=DE}4WJ$S@_~azxx^nz`LUG!Frhvog&KiJ5Yhd7
zEfv_B9e3`jQ<X1_aXZvmPWM4y0$PND+BSiI7<0}zV>N6`vGcmAk0(!g_)G^OUwu90
zMtN>*`lZWhs$0ZFK3{;!qTPG>+~CIK*r=nGNM#}ahyrWCmIU*uU#dGG+_DlcOIqjq
zy#khYLzxYcxRN&4Z{8!K1D$iAj)`X0<K3iaKDK!bHD!cwNU*YgmJ3wV=n}vN5*Ed8
zy5Wbw8m<oM6F8lOt;nnRKvdljqqIRC!yd;)>RYk`hU*7XjUOP%FWsmquIMl^(GCR8
z5*VY0oX`O-@x7x)S7-I56bRgt`FfvgqY^}Gi(=SJ?oj?;R9;S``6v63Y>Xo|?el`5
zUTArUty%M{yFH#arp!&w)9m?H@{8~`DDi;^m$JQFs;^$@a5%Ef&UX3!n%bc*rPU24
z8pRraFdwq%?I^b)!w-|mdB3)_C>qVXd=+g$g^&_Yl#R_@#+sUHZ-(2wuLNV*oXnrR
zU;_ZiUs3`>zhM<OWMVG?9VpI(TPXvo#Bb<pdKFeKxGzf_IP>NgY;N1JitA6=iZO2!
zdthzYn4;Yqlpfd!eGA;PDD3VA@ixcx3xLnma{Lbb@4Lqo+|Me&HJ%eI)=PJTJ^i8{
zK8w98<X5pJ_hl`5S3K;v(-`A|S_*rOh>P8t!BbZ*sb(3}F^)-2g)&>;f{V{(Uj-f-
zGrSHG_$UW8Lou&7GT`q_PpAK|y+6xJU4INd5YqYzH+D+7qbZ$mehIK3?Qb5QANBl9
zDN-CwQE+YU`CHAgoKWgTAAX88MvTq|Cj{e?_TEK^Y}5}%2U}|ZQW9nG!Hp=Y{Bp0X
zu-K0;7gxMk2Ebs6_UG|RISj+wUjlW~6)oHlOdFJ7Zg%drM6C?^o`nNJW=VTqVO=A!
z7HK`d1&Ma0>ivu$%b=Nyh6J)t^;*yWPXV`D_f6m+kQqveg;a(#4sgE?dzM%{ustq+
z^fOG-qtijxnf;01&pYhr+8c#fN%x^c_Q=T=&I0Zh_ynFc!o#NtCxO&b&$BWY60IK*
z80sE#>#IMWR-e&Hs@%^UDU%P<9d+-pe)*&FUJCz;UwebE;RA@ssAx%>l+ViWLgAOp
z?Y8-$iVoX~C68{V=TFfV6%aw&<y{DWs`liTS?o2ta;<>MyHqQy;jv_{w<Tz)|9Jg#
zf<qlQ4Z(2(G8B3n5GKAVLfU$#dD9{-04Q5<wZR~ON9cM<{!{T6ll9wOd01nT$KySg
z6;_7$8yxL`4d<)&giwAi{v_8!S5C*OHL_iKla~PrV4BfkvKewTG?E0q;RVHKOUjxU
z<L2Wwg|asyI8mExO<Jt;&(Iu=<%SzDE(h%`<EmvHK@Rj)t{NmiMGd^`w^_ZO`>$Bb
z%=yq6boWcnLAA4(K;iaqgH(F2IMY<tINO(VEga+5{FvzM<gYL@ym@iw=5M?#p@ay8
zdeCWzY(nTmVlC`f9>kV0Oqby%V^_)+UMen<n>}mMYDf{t^a$6NGqs;KjMyRVA|pv!
z{{wI^F%PO5$CJWgh6<$f487ob3D~IA9GaO~QMK=81sk883=xEliQ*k`10=l#?5&G)
z<UP>h#pVsFL1Llij2)Z?9My)UHU0k5X5bNKA!TswZvP2&4IpfQEv;>o!mn6jK#o>0
zP}Q0(DVDm3!jEQcYJ%)R=pm98_ZSPp*iCnULO0b%MzzNK5!GQ*?@$Y!{m#F8+^p$R
z-?Ce%r(V}Nsk<NGk)PS>rXslVlIP!0#{U><Kc&Dk&S}4DTNe+vIsq$5sH;kEfBhQx
z%$z3rCbV@agz{Z*83`4_IbU7^TvMr}J2x)@EQ=mNW)CM2L-l1}yVG>6`D4g}zX|%2
z!O5@u{1|<QkHCIR-3!D@ob`YbS6LJ;0SvXJ8!~jeHY4JmH5;0uzZO>;Sy+TB4-#A7
zvVAwvyGh>qGP2N3w4XIf<tFAMO?CHH1nQ6wOtQN5w(1_#sWz&h|HQ%eHedvi^iJ-V
zbfTYLitnk-7RCj%CnE3dt_=Q|Wyr9zdnlN|&S(984WMZVb;s#MuBvtSFY6oALuGA8
z!aZAbr7K?!1qP@AMO1DN?Fl-aZ$6WoAsP3@YTNA_G6?bqsEENDg&V+jIJ~u`*#uXb
zuIHNceOGq-w%Fk25oS<N5z4b>sQxV%varvO7*ADlKQH}oQBS<{lv$rBg5DgejwENd
zjq@vSv#S|2MVK~Yyw_8Ux3fd4ipTq%`OXE4e&Xe^!tQI^DY(ZThj6<C>S~Ko;&)JJ
zHv;sM&p1R2atW|kq4V@2au2vjID4sO=C7XC4^qls0(L)s7G*G$4y^^>F>HE%2~a6B
zy}i`*67oY6re>V=p^_uk@MP$t%=AemVl2jOLju%D@-MGw&%0K%Ls~WOo?Tab&Oj%c
z^qX^(^O>Lsk#@6EZ*3d_8o1DeXU=YcUDT={cR0O_{mpLV8c6-yf2knybJwJnWG?75
z_f{hie?^1@?_3qp%==#>S9l2B{4GH7-FSc26ZIO6O8|jZ;xOzIU`k9rkDF^%qCXcA
zJ}V!;1o&u53KACHtf;pn65LW+fYXLKmw*#?g>kIaMG$Or)!xznlf$d?bMwvzk(j{i
zAXLLiAnuHl<q|M&t5-ZkIEf;K^tE^$yK6mP{QE8@^%8JTpr!mHgJ_p1_E}JWq`<_7
z3n{$rCE!_AGf{fV=MpgMl?yFAup>nR>o&&OY_WbS_a_KNpG9b7CjvEyz<igQvAoSh
zy_$U_7+e>1;Hb60BPgi(hS)JbxQWFH^=@AqCNyM8i|;0Q8X0Hj=xB_|`lZ7&(w7|u
zC;LW+3J}3+-+hX|a-}@n0`7lN6b^+99NvIRVQI@e8bU6N>2^kI?Gou#r2|eSf_;<r
z?f84Nu<WIqVEVo;jw~7^Ub!fTmlJ9u6k;;Ob7$)UXIaq?eGvWKuMSy_IN7Q4+E{G&
zQqprkDrLZtJ04$D;;xUU9gX2@K4&yzAi6%yzfR~Ra=TH$xRX3uX`v=1bbtH%0fn^9
zzwj7SwOdhbbY6^FUm+BQftHpXI?l8_cdWxuFoOAH(s;wZu(ip6Ws>l9wPaPHS%oS8
zV1!t^{a=5Y=?qJQ@B8JoJiUm>!VTN_;#deP;{BcH&Go)FZ=T@BLNPV<YZk_*VcVTR
zLPr$|)a!fs@m>PR@}CyJcgtkgH*GW3t6FF}@zmt@ep+GlI9swsxx5;P)jk|rM(cDG
zPmeT$#;$lmsUre|y~}k21BLgjS4SYcQvwnVwJYuHmZ~Bj+4os|CP}e<%MlpK(S~Q6
z8pk}qCVH<ClzYw!Zll@fa&z-4S8Gb))eynKqE(mX{Xrw8S5Y0uj;59y-AHz{JdraS
z>4}yY7#0|H_pMUODQhdttf{&4iVL<7nC!Sa*qM|nlOlg@{}1XE^c3YD^#ZD$_r&_0
z>W*ogyrtI>SoHJ8G||1-k@oM376~m>LqoBzx-_LBv{<Hv)F02QV}2u2?;Ym|f*vP2
z2$IwwTS)gN7X)jl5xjB8rpvzsu07ns-PD@c8tg`~hzlOo!2=?duy;37V=r_v%e@gy
zMV~z{wrG8wf`vk({_>I6&YN2Lnlkljlg-c(m{JU0ew;F~40jABx??zO`#L}wV{9m^
zuqW;^aSxAwAlmg-{F)CfUE-&qe5E_PpH_O8gHD8l)3D86$!C3xOqT$g8ThRR`f{&H
zA?0hdT+tLfPhJas1Q$K|^4Q)~(fQs@J|<Zm@5wBVcg4e^5<v}``xYE2V#4uJxkh_A
zBJS|n!<szHW36Hm>Ci}ogXDaNH3YRe2)q`UEaULL+JU=cR!U0dr1QyB(dKtep8lAq
zJHHIagvNf-DxG)x(nd;qh;t5OeK&rdW{h>BFsoRm#r;dbs<zfFcCii{w%Xt3Z#zbF
z$eVcf$)_I{6ERA1<Pu1KR$`-DG`yCY%zc`bQAG++OU1>XO1xrhu6dsL!C2xg#cuIo
zJz<BJh~m2h6#V%B62FFdAF$;3pc4li(UJ+#=GAa7*it}J*64deHUCiJyGMbIZD&5k
zr@q+sO+yJ<7^<w_GToi>1nl_e>+A6Cjg1A8`2ZM)`~C9}7p;&H@=mS?GQXJ!$@_*{
zme=*lP05<-+)9u$<@a%8(&f5|w2OL3t%yED29rR{pA2j7aV9*)ROGL3?@Q5<jXo~$
zxbj<b3*S7iyRsYu$Q(~i#i+W-(3*2KXpzBr2T^oZO9i&O|Ez=y&SzvM|B{xU<pwLK
zh}TQ@Fj13%h&TqUS@NJw*Ry5rh_1zRFff1h%=iyN)yx4Sd~{BV->*SWkF>UcC_hXt
z%(HtRh_$dOM;T*0x`zviA_aHf56URPhPsxjWYV_XlUIW8gI+)dNNi2Oi(rq749Y`5
z*7F^DjOE7gTRH1siz9<uzL|O!5<$O)sH5kpx$623&pq3LO0w;|o6`y*x~6ko93d;K
z16s86a1b>%cbs5O%$LdTaJ`x?vK-({we{4zJKnZ5A`7y9<7x-!*XN31a7iv~!^hZB
zrIdK>(`Tw=L&1ZfWx@qf;s6w8hABnu?_2NzH}0ngnXU2pNnQfJl{A^-zn3Ww)z}>8
zE3e1(8|usXZeKsav1tBSw(rMYh-uBDR`+2fZf_@uRpXV>dj_fZ+n#GPCkB*r8)|KB
z0+zmg4Y68t+gy5GFDzi<SnR-t_Q~ArW?I<ezEM`Tr)E@iJ$LS6a9vue_VI~p8m)!d
zE-gDYGPExz*2HqbrJ@xRPCEL!Y-=lS7>Q6@L=!fqi+2cI{e>KO(|P6#&M3UZryh|T
zJ~bO+LalL2J&@-+t9M^$!4o%;>iq@RP|r#?n1N86xIQ(eRC2o&gLNbapPIs(z^j^*
z9cl5n;vceKg}^;GEvvk9e;39E*p3>QwzqtlKEXO{?{070<yD##$O`4lBQt+;RZj)L
zjgKdETmry*hl_bbjtgNL*%lzW0QwILy*-~Y`mfaiJu^eZqx0W{KK8pqI`nSE`y04I
zLDzkQfC9bY*0@>W62axOh)-?Z>9YmlxCeOqI8wx+5;5CZTkpMw6fThBV01BSoq^O*
zMS90?73k4rYrJ{;X7l<3U_>D&SV_TZ)7+e{@;2Z)Ak3`lXN~nt=0$++QIN~~4Owi?
z2HJ1?L;fOx)3w>)6NA~aSoQaqhw%czmSk~T-o;o;RG9;IakCx77~(Ywfrqald5z`n
z6`2^wbNB(s?rN3sUJ0qQ#$EzQ(*7GgzrTt41mW}!Ug7I%fEQf%ai0F1YuR#DX~54x
zfhYK#gL-TACYNmv&${2END)R2H#*Zy6~|mXkLudfBb)Z;skv}qT&{HaVo80BwUM^&
zJMoD6p}&sSBzuHE66EbbL#T9I$JA^{g+D=E0umKGdGz2jG9_PArjAzSknwhzN!RcH
zZhqD7<*su5{KE^eHpkS1NCx~pRAEZ>oNg=1ryf$?z~5&`*LGeX!RHsZkR0UXJvwTZ
z(G~_vYhm1;8j%-Q`Q9!%>~C8gl#Z*7JTx&siLhC8m;ytpw^O#qj)V$mi&d>%o^n|M
z@6jkax1}Q4rY-N1Ja4x@C5V#0!M;?&;tW8m)3x1t;BIW?@py7!!)o6+!o|=Q$QvS~
zjn-isVw>l{L2jBMB;NcOzXAIdSa%<rHk1+|xaStujV06*Gmoi}+Bmil9HtK2@*+2T
z(3mv@w645x^L2=KpEo9a^qyhaxcqpGeRV%e1MPfjTNByQ8Ip0`n3lvRIcbOxDe=L`
zF$821rQ?D+HUpv3tw+REW#50Y<cMj56bSqQ?cMFEs|T$WTmqVy`(+TuYdj%|8<tV2
zQGxIoZUlng-Pip_g%^L!9FqS-_Je1kY|x)IzkeGsQKjxL<F4QZ2^s0MP|meZxoL}+
zzSRzOi-8(v5*j(O?rnX)>u`$uk^^}n>7hbrA9Y8%QP##Wjwm>6t)NRmmOp%)e%O11
z=Xs#<t;ZQk;m!!}=t8gcZf_V?tT0iFgGhpvlz786dkF(HRm)}_Gbr)&m-L!vOl`im
zCq>KqMS<rqIm^d?ZeF*0I{&K+gi<UUU<P=vaL(M8)G+*P_m@*B=@omw41FQ)_h=-P
z#1E3Ld<J6=xfmIkgj%Bx>WhD2{d@eNGSf~o)ov}yO2IAJ*}-E`=48$1Y5hVbQ879p
zs+2P)#)z}Wjf7&*TA5=5$=!K!o^_@@0rt@nFc3KAeM|Cc!h<fF3qHc;VcHJ5=sz34
zHyWg<I^`!c#T+Rja7dJc^_EW{ryrp9ZC+b=J{_ej8HKKRsVNN~+P!KC+J4oDD;+W+
z$;@l4%;p{hp*!dxfKC47g>d`R*G)^BiRmN<^88jy?7?-{qn}wwfmZ@C^``c;*IRkl
zzan8?hcR91giMf6A`Aigro}oFdD9~dlhOuEZjzER*wy1MF{$Yn{n#iy5^iyB-VASq
zAQ~Wdm~1gC==njJMF=Kx7J98Za@au<-qz~cI<07pH*pWXmEl*-oSV2+p`IP^oSbo3
zkl|$L7qB4_TTcABa>ba>)*cZuxT+0rnG<Y#dPcRaz}ooz+gp1~oHOmPi=)OeiQBFx
zz^e#&;j2hSs2@5mfzOvC9jet_Sw~oKOVL}Pu9Ebj7Mh<cV4SIg7gFt<<(8&m94bPc
zC<l;P$W@p8O^bBivtPfnH8B%%zsvBc1tb(&7vFu7YjfgtTxEQIqeJ-;;3yCKs{r3=
zV)*B2!B!FJ!kq|l=;&{`f0mD>ay3^^57qQY0lrEp+b8e|WJv*L7csUXADgvdo}hTq
z$U+$iL|HIWBef_|DP=80O~{|d@v=M?7@_FG@+XWG9^r?AaqQlfw)c~LBo*1dk6VAz
z0iE?%d*kOX?z!Hs+F+jjeC}fN*m7kF{kaZ;wnO_K1uSj1Nf4<!AM1E$V8NS<e8%gb
z=u_K0oP^GNK)A&FuE`f^Tm=sWojHAZQf%L<&zt*;FFX*+kn7_WchE_TlJug|a}EJ8
z>1ogr-TR#?P0IfcKS!K4Hp(YuSl*#R9LroMPFUVxBUEY#qL5ARZH5N8pU)SYdPm(*
zW%aTB!HOKhe{nE?jTn9KPbow_c6@CbQWNlM1KTkOlR@f?1vV~!fYpe#KyWaB-h&~<
zACbR)YAE82pycSQnZFP-xvo>>N3XMN-5es6M#L^GXSX`k?On8Lavr~nC{_FL;fKlD
zJKhXihfyhwQ(dX&_d`|x>ZvP+2eoM$^f)#J$7zJ4Tlx#g>QkR^|A|>YK3MU7ku~*l
zvMLnZS}#8J@jUbHutY1vJgmtw+Yah)FM(-~TVM3xc-vL;iSe=5-tFUqW@>t#TRpvR
zwgB0hDpxwnHiD668{@oCFSK$DUU1`(9I8Hn=Mih_nA6Z!5HQ$#`?EH!y66?z<UCyt
z89fKjq?mTh-lq2hC#U_Q{bG}^207Naj+I@=N+yym&%PaKlO%Ayyo{KH6}rt_<Uu;c
zi{+gx{eecGyR5Kg=<b14xNct|FIE6_b-}Znr9+xy(`X(F^8an)O>@0mJ@$6eFVk<L
zS3DNI8dySB<GbNmWuy#^`;92rVGZ$?SMG943|w9Cy${#q@gEjdN7>C3WnL$Q`5<82
z{07;OQTOa`(G~e5ckeE%VSn*_8LGNjh1vWZxu0U?$I_#9N^N0<9^MFKYK%g~A>4LR
zQ+jabS$N*?9E}s#(9ivQDIr^Pj!{}-A?_QlmW=DGEFNt;@Sy@p8JAWM|A{ZA`q97A
z#=Xv!F%zhg6)tQl%6)w_C50%A+RlC&?^ztt>WSX7e7ItC!>qTSE>5IyRfU2}-e^LL
zJ*0U34^m{!k;6U&vVPcu3s3AR({sAf6^(P|xM{%hK#IqpzLng={pk1@thk^Zh4TNV
z%POhcZ7Tmxiy}8M$Vl1_Ijn;mP~=%y`I+N#tJ1H0C)YT#dCxoWyim_$eyDe~dHmHO
z=gk!d>!CtNMZ_;+Y}fX5O&v7IM}{P$GEF$eob`r;iBG>x61aF9@&|JC$UoZol5(iX
z(R1f`2{ccy+Kl^1`j~m9RTA>{&vYiu7R%)VHqyz}rv-q|-kK@SHipBGBZt}FTfTk!
zB5&k|0gL>r@lu#a#8otL{$~Gr9ZH}15^%)`gmXH%u3W6^>(2Jpx<A9<&yP3kTL{jd
zPgS2YQH8BN$sDmC*@%#Mlwq0`A-7(nE!pOGXXNNrdWH0U_Ru{`3JiG~WlO(0(%+$d
z`lL!dJT}?6K=pG_is*VsTAwN)ec=7ej{y|vvg4<(eg(1fT7DV#l@ybND5xB>T5+Um
zB&8H<om$@Q?C5BOv@F!*_xn7q)E7#*BO9mPgYchxc{6&*KYP7hQ&9t4rw4h8Ioox%
zjbAttVQ7D4lFk-}lJmH7mY|^s*)X`7-jXd8D*HpNtZpqrqHwB%WJJ(dwa!UPQ}!#}
zHjsTNxb*W%y6?CMfVAr!;I*pMs}8yf@`FX!oKhYkdH)D<5$$7?zun*?_I_<0J?%KU
zO|~9$CJr<^T>Mzq<1hI+OFgyRS$09>+q|^IV)?P3ZmwLw?AIlm-+C=VaIYUD6J=U9
z+{UxEr#n!Sf4K*gIC<iNmN-f)dK1Qkzi*2(^jWc|l~LGBWaZmWNh@+;`*?fl2hH;Y
zT%iNXt^184Pk4Op1CJM6I<f+PR6u_^p&-)H8GQ*bB2<H1H8ryXf(ukB<H|WN^e+MX
z+g(^i0@q*%DD&A@BaRgDjC`(RgM~4>Nh5Xz?}vy3<p@l$F2dJMLs6qnYG6&Xgx^BZ
z=5=NG1NtW~pk4dmpPRv-iE#-b<CseTR&gE39<taMc9vA5K+yhXv8UC|zMT9+*DLk+
zYkK}v*RjDT^t_M8`cY-nyk)(PSW)U8;NrD`tS)wxax!uYO<^Fo@N3!gOi2gL8hkr$
z>^{@I3UH-CjE?XNTA(6v1V-MWjy@*RgV38+Xr(;2S{tmKlV*PD(mn6M4v09-w&;R|
z$CT>518Q#-r&x1AQk0jJ*>eeafOW&^M~)H|PzO0KCD>dDXP>v?XlYyjeA<w&2Rc6J
zRkx=k6)e`0k4a);9IC!vy(@F`r;}>iiSuHco|d_}?){Rf^tPtV{HCy$Q??Lg{uDP|
z;n#Z+&NVc%X4`Q>BH}mZGl$ZT5)Vo_GW5sBj1TEOPY&a_{rV(h4usEFy?$?LweG7k
zTmm%awsR^TRQ($ArT0m>`}W~bi+)XYh{cdYtDbbxu}~72Le)W2MN!?(dM6F>ZR)O9
zqe>LTGy2Xi56w?iOkZd+MjH2p8iBK2#GlT1OpjCy$`;Xx;U0L~k9G%lB7>Imf;%q(
z+!ZSN8i?b-eVf|g^`c{iPJQF^zg4TS%@(FYoRp72`C?BvW8)~@lAVNt2`AYWeR4lj
z85h^kqdCsr=3@;J!Aronp~2&f<IWEx#|X}ay!apd{&@*_UYF#)M!gT;EaVuH$p|g=
zG)?3@Y0qX^yaa4{3f2v#Qh1WzN98AX^S{2ksOh(Ke7=dQ>*`)>?Y_nUV$z^T5ou{M
z25j2BloM*^<|cK_<BHNfeucd^E~<+~JAQl-$uaZC{&a#kcL8!Oe6l|5vemhNpEP+3
z|H05D;MO0OBlv}2O^5{uVN9e|EK!g>N5n>d9Psz>{n|lL4lpGnPM64)5t<awmy;}a
zTJTqy2hrHnHO@C0%IcEiA9sI_uyWBue7W1AG`FnxCN_*b9>?C|nNEkzj|~MU7c4Sn
zt~XXI55?5b;^D@J+2SW4HH$~J-cuh585&@beOxeF&Raqq$Y2}#ywJP#$j4_R_=)wW
zd$%XE+{h*Ax7ZkHM0V;H+=;Zw!t;an{*Te6dCb&Ucgt4zRJ91{DoTnD>&Q*e(P5L=
z5WY#3`ZqO<ltxS`NvPh$4s5<g8OPPKc8INeelM_s&F(5E(51`HsWh1Q_Bj7}%R)lF
zh_Nnl6)0b>g<tmlb{S=C1m*-Zf4;r2%v%_XR2iJEF{v74nLLtE3>^Uz<|pCOj_FX}
z4Qf<N^r@7-%A6-DVc1Pa#p0%Kp%>ay@~}%lv#V@FVz7(pla-VpVY!3sr6B#@^lFNM
zWi{UA5|D0N2b<5k1c)CyGQqyLmyx{4)s7c{y7>@gX8k8+)&u?`t~d$7HE7Ui{ky7(
zo~iO#Rh<K|1?AxGh%S42lm0J*^@CNs9=9v2nnW>d`>+otjvrT1hA^zKP8?c(9aD(D
zwh@y*Kq&a~Y4gFvqf`bNpHPFms1o_}4{|y*hOgxzt8mS+n;Vih3}Y@<WZNBvrR)k_
zoA5RdiQpO=g?WY4ajQQm&i6W=>)P>d6t?_5w<KL-Na+TTzXp-AguNp;*QL(-VAOlv
zj4mHSoB~G@-NMTcX|0^n=kERt+PldLVf-3>{wmXw`8P>VkPHT_a5b!vIqoxh{#9+A
zOb2&kby#eEz@Tk1=^!iL6S01}bdPUU?P^IjWsZg&HMd!TK*}}QVo6Q?VHJOcncDr#
z*%}dijnvd~i97GTwxrp-U)Z*gtGiSwPf{K(Ii+R$eK}*Fi5xF%TJ4AYineZ@>*5JX
zFpyGwzvOk}#tgZ=r&{c7t0rqoyBKn)o6xsGl#0geV>D@D3GSq~I#1BaHAmZ|o{0g|
z0NLAS(aVYf!6q5qg0q^_q=bBu+bRohfmTLRk}RJo67HhnXq~`=HYyZI5_D|`;}3~1
z{`y+)#^0b=kIna@bayCA?%fqFfjUuk1idZstt|0PF0Y!rb5b`&GRYh%IWcjh85Kg5
zdkSpyd~2a9(^KQ2y)_7%A*n{7TaF;4w_66vSlL{+?yYJk_oL$X!|K~VMtZ1MU&6SF
z0^{X_`C)I%N?cThI%<*URzz)dp;RUV(%JmnOJ`)bXQ6b#x@>n*Q5Ytz1UX2W?;uXD
zeyJ_Of$&VIx|Cux#SMY!(?cP98p)qx&d(BmzKNAyapsjf;8ztJ8hy1+o#t4wS|2f%
zk8^?bXdZM>S3lC4ELrT1bN5qLY{_mm5S#VC1W3*g4~9t5Gxb}h>WEt!`~MD+7w+Aj
z>VN80x2gu73ig^x>}+Z5wp|%kN*!N~?!HzVNsqTg9VSXa`$XanH`@cXV{>o%O1coR
ziDi>j7@kegtC~o4JT;mrhF;s3IzAxY8J=AU2^o$xIf+PN{!%w+%!$=CK`|%54XLeK
zaMBgUOT+HLw5kgNN^?C2Wn2_5?ozyX5ZMIl(SrIAwFoIt+E59f+_7(eQ^!F&4VquZ
zzgoQ`D%P?eQblyX4#^gO-%-fY^v-AcFT~GaJJ)yuAN#sKIaor%B~vg{v!^`JmU1>=
zB=H%pg|#X4<og<tI`;tN0b$l%2O;Q(a-T~y&;wb$Snm~MT{z{Q1JA}JQftA^1m?kd
z1wQVEVyO)TE<bg~7%K!Dw7_=aSVErh<S^uBM>`lZtdQe@&xX=LD6X~P)6oc7#~c{y
z0W6^#6%#qRqKx9}&jUm4%XDk4Vj=JBZ0q&Rb<gsJB_nEvHZMFK`g0*UfiuLWHHO8#
zb9TCh;CGfnoi3ldz}EQQ1knnu@BggJ4dN|^kh(jN3-^G+_emr$0ylzd-TIrIt&Gr|
zElnOcLJW&$-W2Q9*14nd>~Sc0XXxP|Fj>ubABQ@AjD{pT&EVD7YeYtHK1G_XSrShV
z>MNYCsRQgY{gYf1Xg_Syd^KE<T3p}@WZwcc#3AKFaTZtuEA;1#^-gCX+4AnK@!poJ
zRtmvkHBF*bnp;JMbg6!BDGGEnR?4!s?o{l$n|w7HORi1D`H7r!U+elDAaM__Xi-of
zuC>=)yd280BWYV>s@7C>)|4MkQsWJUTa5hk`lr)@>z9BRc^UZhXlJzZY7Ag)mbJFx
z5|GUo8ri0}dI{hu(GrLWTio3vVlo3jn0$ceaG+!byd*7qqF#|N6o?eTxnlh>79<C0
zVL}tZkD1mxEYUMJr7KI0kk{8B$MNdm@#kaSS}x^}>O5S)+qH|mj$9fNn@ADl1yg6)
zUY0t(yXo(kPZ2ZR)-$BXe$m=U{=;p}0g0+#gDMH8<vSJ99tUlEjVH&8fo<-0NlD9s
z^Q6?2y<G_5R&FPENB8|X@M-D|Y$!Ot)T8!UFkz7~5#y<2bm!m>-g`zY?_YWiPz%=c
zC;)_nER-y{fEL2w@L|ky7m!fpC;@#q4O&+sBtY+aasCWK!M`WLLf2vvm~p%Oj^4yY
z1V5U3ziLzT&Ul7LHte;$t}H1!DzRl8#X7J7jIMG~9;J`?Ja$L*+^!x}l8L82sV+J-
z66$@%O0@AiA2cJd8Os2fp3Mh9aX2yJKo>n0htu#N3ZTQi5Zg7?w!`k=7cyJM``#hK
z=_{g6cY>gwG=HU8`wb-p_M#vO9R}r*y&$d*eqsDY36ZRW__L**A{|4oMq5ms@w?i@
z=#J2aGX2>TvIV>tA+AdlDvE^$L~w4PP!}uTN5jGHWjuL-Kgy49ru`V^{qS@jVZ?So
zVcAU5>5X%VICVVfp4)4GLr+D@*0cjf`l=H9i`TV9Rt-m2z-0IQuA~@T{kfX(hu0*w
zRBhHl*x9oA_1}9_a7Ef<O*gi5-!sj{75sR`@yUw@PiY6oz|SXhdlz9BKyj?&dP_n_
zFpj%N@ej$5w*zEEK7T!bWzj%cc~}VAk~{V>v3HX~@Zi~V83syE;#4wa_C7tT&T;D0
zJC#Ly-QH$q(d(Mku#3>drzL?@v0*pFWdF;pSeolsY3iKwh_s`q6c`)dBdpJhYNAtv
z?RnM&(2cQG2FKsLndFjiO3>}e)oT3MS3<g^VYr>)BWP9a-k4%w_4+AI@8Gv`IspBv
zK`L3h))loLNkj2pk)Aq=_PRG7-WA%|n4NxCdaIxTEaG<1mC3-cWakF1X@~uG+X`Vu
za`^&D_q;xWC{D2pk8EZ*ZA@=p0<sdI1HcO#<_(EA*ff;d-dY(_b^~(F5h|Xs(fwM(
zUVgP#3T(JMmhCT?BSdwxrYFevSbEOl-gUW#`Cq8i#*bE)fS}6uE9HnQ7ZIA}b@@H$
z(nkuO(=qCU3d>^QQ${9#{1c6oe0@8cd@j!Fr!}}Uz+Eji;L^5(5aa!vkJz?91Ei;k
zsbY=5xjr5PaW;tH>HFeoF*jeo>GRhuWK$2IZe(gyeNI<I7T$G>@U{tQiD@Sp8U0m|
zTcTi>tO%o-R4lghA;xyBcz%0nS<8dv$i52Ut0V6K_7NHAuzzI1Ck{mgAd-<kHb<(!
z%@@Z?i&-r@USB4*OtPATN$FPhOyrno!Ure_r;Bx%>$CPAQLm9R^DXmI^0~S_sGYp1
zN5KxR`f6-nd6$BUH@naA*XN-=AWIV|9XO<HU_~4oQ?t{3#1P@N)}bUdqgE_m|00y4
zd16&*xyLHpjeR6;L-I!~=k60PxUo4{!E{-~yX2x|DndW&#)s0exw35HOppU4_M|<P
zl+TNXt$G#`(!2g*)ky%b*9E7s!|Hgg#bT=XT@(12GA+26d@t5)-+N;facdk2Vid|C
zM)(wE|4lc$1i07L!CYW$x*N<3sXeA##d6x8Yd$ac2@Xp!8{^y^7l7LMzO&Y!d}mQG
zEJHW0Y$MD{`A{K{(*_hD2-a*`%&U=B4}CJAc)~x>|0YD;MbJx@@9Z|k7Sn8a-~Yzd
zdq=bR|Np}}RBM*nE3HwR+SEu@?V`1}x1vf)sI4I_MeU*}N~-ou?VZ$KMeRLeZ;25>
z<o&xopYQ!U_kDl=NKQ`9xz5SSb-iA%=ku|iA<U%TQ?+`ZEQ}WStn(=I4ou@!zJU$q
zGiHCH8xx+0WMcnR$;Y=m8H5Bfdb74PI)$ey%88FfD09OlZTT6m>2<Oj?^@ZUj?Al6
zCbRjnt&ZW4%7psWU?sc(!MY@1=ujJH7;W~ugKBz9PQ*W0Fx4qRvfWaR(n6YHUN}jG
zE*By%ZMp5O17*p2Tww%0WFN%ZM&!7=%68|-B*+fjqOW%82N{GZZmE<md(DIx8pmJ*
zVl}V(mUh!zTK5^}-&{zUTig7esiP_G`hKX77foR$voDM~xn~}?w}vcf%`~?S0;Qts
z1@;Gjl1A)ndiE|?b|r4Wi~zT9_{H%~P}2^I>(QTHkFtl{M!5}+G3LGeLGg_WmuS2h
zP`E|Gz*oOooQvP6-iz$~>t>~on5*EOkg2P=uCfcdnRI8qBjNdldH$9?J>R%Elt`d=
zs0}aFPhSXEp(|G8{?=9=1G*=kgGn1tWtlD7Uq4fN2hI<EP1SX9Z-tutv1JgjqAWF!
zQpO(`FD1uB6zf{KW$IhL8Lk^9M2dGikG~@<JL~_AiBo*<;ncXXqw2JMIoo%Y@2&*F
z4ig!6G$!H8x%Rc|J6|0kE4Rxniieg?nooO3YFC25Z|_bS2fInk@anl<2izb5%VdS<
z3y;)vqVn)0Y81=T$%up9k+v=azY(o$s$0I)Sw5o(^?d(2bCquI62Y(yG)i{dVMdo6
zQSK7_^_JO_&%>IOi-pxp6>oyRa<?}<3Xya&y-H&mOTK9N>WcIW(iy{hZJbFk9^?mV
zXbwVNXHIfD3*z{^i(e32i{5#Qe}n&nkZr!4tUrT}@0214IY79=^81bP0>~Tp@#o3i
z;J}+DYH`U|t<%2`43s9}(KSkg(&q68122+5^#+wG{z|6928!gtpI5iX^2TeFY$1CQ
zM@(5|1=io{U8s0MZ#bf)yuL3lpOQ@tuPJMM?pgH;;51`4agIzfe;1pp(0?EmV{nm|
zX&^jRJ7=(zHC~X*#CPbONKkz}=}{|YXaQhUcX(228b6M?O#dX4;)zA19JgeDN&-jJ
zqUe{2s+A6-W*IgY2PtzGlkHn_QAoLhU;C(}YaXMwc!O><T$c_D`_3RG6$}9_AA%`c
zl$VYt+u6W}`7~FMomhh;K&eO94=0FdY}CL#2>DZG9-=Y<C#<3^@}8?#pqDS6CpSG(
zpso6YtNJd-ETzdt6<~p!PMGz_J7IhF9k2z*?I?jdG3v;n>Xy&fybEl|o*Ghd$4aSD
zL>Tq|F6Ke=ki>}(aPoU-oYCrUl1yxH-6N+lSs5d-+EN(CITo?9*Kc6jptxMW#VL_#
z<TJ?-q?+x5KMN)Lei)R>uoTwKf+-uVZz~|{L8{fZ92-_DQoH+50xvRDSs$`u&lnKp
zJ_)*QZ&e;M@IZJf&8x2wmNSwn>#8of%tprcAe4x{d<>HrK}o<~kND8t4G10Xcq}^~
zUV#)ycMvZfPUevB=TAN;_PD0lahD$b9Q+0xugY>28A4_a&|fn%Pycq{L9|=AI_8Zh
zE&ju)jZ3`A*+FT(6xE+`NnRF|FE_Fbo2G~-=N?1$Yz^^cTOFUUjJCRjW*@(o8%@<S
z{X1u~LUQ{#oAy{E<eUJBap_ZLn;l1s(8^%Y4SX>MiqNj{pmM3HdRH88;sFc22hNOQ
zjk{j-+m5Up%tAn$5W-ekwQMJm{Nu9jR(_354asHk!I6|d$a1uRHz20THz?h}lA4L!
z;BkF3U<K>*Q9bS;-RZy%nR9X}>E_iIOKha2*;(u}&DqW=Cd9VDM5oZZgS&$*O8=7a
z93cKBv#W$}U2o-GRqJUV^d|u|;!?-bkp!^PPWM?zA4E<osDM+qm?SX*iVgq{FkVi1
zKv}ur1E*^Nup}yq6_&^raBD6*snoTAh0Az2oo5(>LPG&S3C$t@8!<iB6I2k5o;ljO
zxLG&rKhr&2r6iDe$Hr%RDDT?UEKwdi93`&Iwh}CDS!;V{rLrcEt?+7GNpyO7UWA-O
z&n7=fz?WTx6OBRP>9zbg7oeD}@$q6kM{>NHf0Db<|MSuV8HLjYV{_6B>51GfAxn%B
z@U*qzO-}!wTX^(2D8e3+{H&gZx#F+>>zlutu#Qga+81|4F31V%nD_Rd!PkhkxjVOi
zPFeh|*e|3ZQt{E|U1@q}cba3)_Hwb05Ai8iIWVE+OY6XgVaBaGvJAJ#TVvmHZ)lPZ
zc&T$02C5{?m@WW}fn4J1M>*K#aOPf@JdJ1KE;nCo^3%~*3sHR87&f*kQlq9*nn1$B
z$MFx|I9)>7n8a@kI>zs0!CO}dj1oJyTIqQ|_Q`B>ML#dX+G6_D=_bk4unT@9vI9Qd
z7a)bYU5Iy4KEC$RMA;m;Z>U`0z5)J|hv>KOvui=|szq1Rz~<^*u{2G$jJ}Gp0(iVW
z?b2fwd<Vf42tVP)PZHuVy{ktxSGp$wG5Q>S#Sw>R@r-a^ewSGIl<?JmbX|TG!_yqg
z$g3!JMc`_|(sg@+xOqEruhj*E?ko*xFZ1rh#2FZ@^U=uG!Do!>l`on|;Lh~UK-{;_
z{$Ek24~IKo4*jUwlwo?N!<Pv8K9psMD_VV({R~MEFx-~_?QP@mtfxu|)dZI=w+z3$
zz3KRwvwP$ne(g$3uMC*@{q{uKblM7<0%PKH$0kv)q+Az3gNkO())L*u-I%3T3g68I
z@WTp3GWaQ={wYFN>Q5!CK%-ReRlm=A)lHVM{#B$&Z;Rfu{x*depPO9Ok>&N(uapas
zL<aT)J;V&hhEh&~b+ZNCix;2g65RhxU_ey~L?(|-jC`m&On$Xhzy!yXopv{($LLqU
z?1|<rW<?`S)8seHIxTW!Hzfx+hvSy8CVH)o?eZiJf+OZLj3VF7P$o?teZMuafIt7u
zsN{MFujDJIfl5(r!D+W${MYf;pO*=<f;bQj<`{IE6mHHoL}QQ?v*GPGvsX8<IJo=1
z4k(uIt?0p7L;i*+O^7*JmP-)T;j%5!@2WaIK#p{;8o+U$8Qn-|j_atctu}AOgCs+^
z!=YhQQ>6B<zyjxp+vcM0frC}~+JD6!Qh-_@8Rj-2GaLh=_rWK$<M+g25;-=7#802s
zp%sk(k_~W2$bffgIj$`1<sUN!=QZZn+aGDIog2K29j<24qHf);7g9~sU#MS0?kvU_
z0~~(nOH$k8)U=9e*c;qDi&aR`Uj-~G-ZB@$5$NayDek%1$LG=J=)q)DP1E%Z4u3fs
zeZsK^xA4TPMf71`wmJA&)t!CKJ__+z;&=WFfCE|$j4w6F7-yFCimViwoZ_FDrbL*2
zTn<$VGx+)8KT_XNT{m+Owk?F_?_q(I8~3e-@s<akWpf!4dxK9``eju~CX$>%^!D0-
zA`)wm26SagD`)$P%Yj*O-7z@bUAt!jSGFjm3k3F!`$%~ops<hyr?ys0>SN*(P>=i7
z)zwi2*(?6^T{qpRq^mi}UtHr${y?q|{&alQ2(Nl7u~z{sF^@~+k?2P3)Y|ZV3b44r
zgs97>0hfA);UPRtSeA|j;S*c_anEn3hV-|ytc08>XqZ~Ubu6Uvdcx$6rLEgmKn%+8
z6=a;WzCfI^^UA+uEWtT3eYE+dgLtW)VrzwHtDcj1e(s53Q4`~;apJ#Z@;Gaec9Rrg
z`V_O8rmC3p-E-+P)mai%PohJHkQMVRhEuL%%j@s>1fld&fHcl6eigkZs|}+nBv7)+
zc7NWvXTNopM)QtmX}yB^oay>`Lfj7*f29UwiRvwQ8}#|2^sf<;aDd!YN07?+&qYHP
z*V)i4xsh)eyNyNl4eO5YHzd;UKVPpa5*23srsrMFhm#QOE*e9EU5sqGW|uSuoluvM
zCGRc3&YJ^lD4`xi4>8Vnj%~A1FSb}k0>x@SfIG`0D2ndW_aY4mRM?(&^lcL3n^|xu
zQ7MmL|HRrSn}bGf#vP;16Y=^%$Z@5j%iC(sB%O^G1c?$6ObhfMHwOuWUhXpf%xBzm
zElEO-|7;7(SW@SCs!`mJuiZmXh=<H!W7@0X!|vQ8fX@5HBpSLF{58-{)nN8W%l<M*
zOGkszH2Xg>Sl9j#-~65H{FjUZW?4`)KN}I}|CXY;eFvNcqq>x^4_u@je=vN}JS@G%
zEepC7LXMoN@h}0xq<_NZ%{P#H+iw=RaG<D~Kxx#_v)HiRdwm>mNqfMdptnbk{U&O)
zFMFm-6Y>_*EoQ7@2PDNEL1*XiVJ!qI43%_&<yEffWX7BF!c-MJevav#^#(<VlTm4C
zA!L(vlh7Qdpf0uYeV6WwGrbYgww7@=dPGrfZbnn6eyv77TioFruwL?`JC!saqfquS
ze^0-U*qpT^Kg*#RTgV+MEYGtwrUTAo-rs2P7}4&R3VFIUr?bi0OA);V=^m7tH?WN@
zO<UzzHT#->M+4xi#*>frWrm_<bIiWGjrqqWxt^=M03ssV-~4w0zAzApC1m0<I)2l)
zw!D|{eGV_8IZ``(uus}ryS$OH)#2c_A}s7<I(-iYZLW>PWal$=jy#<bETybG&-eBH
zr{r;R4`#6ys5Uk+gB0CM)uxEK6I-5(RsC`Cax3c?g*CrF*>48!>iqsnH`+3Whh0VZ
z9>xXbsR|QP_!r_+*JRi&|Hzx`jNiI(>vqDA&dB5=&p~`-BuvQxBgQbNKe`7*VW<u&
zH;C=8()Tc2kSUy9Wc*9^1Qi#;w5dp=<6;?kuxdmjD080wXXOxD4f3_RXx&T6JZK+D
zg6(xd!|d)k9ufSt5=f1+dGL$djFHPr5j_iIv!IK0Z&COK73KC8a(A_1!yLmL2fC=6
z%mLI%K@%&EN2!S@rr8^pt>&G$E7{Ui0L1z}WvD^+4qy_>7QhpP<Su%^EW3gxeTw*w
zx<1YWYR18%SgzKp;hRufS;lL9UD!*r8dr7@qivYH1lpiRiPY?9GnU*Q!dDkI5ZRd;
zV!`oLHL2kS>H4m%Dbp~g7^}F28h=Hf*uv7{Y4Bq1hRY^R50M$UoUpfEMyOW9`A3T@
zOBBIWaKZ%Z7omp++15STO6pz!$jft;zt=e+hR)=C%T`x^QSA8TYM+A0fAl*^D-iR(
zXzfM+fV-5)jlW?V@=RC+kx`v=hXZTAy18$Oom!b}b$lPtv=p`PW5{UT<1+{;CmUP~
z{QaHWEN1V>F*Od5SF(mg-$=C246ttt5a(zgZ(lAMdYKfq{&jqni=%}-)+2AOSx3Ox
zsqAnm{%b+5k>2~=?WMnKvK&Z8jkVxXL+sU}v%SbbDm2v)^ZJ~20<^fv{e7akgwVvl
zsg5Ng4j@tqf9K#<+k(guK?g2n4XXq5&GpeftNyui+p{T;NUD@kElG*bu;)C^P#Wvh
z`yXCSF#UorYuDy3bFo8d16VZ14&-8DIBturQNFvu=T=CuBVrZ3&-Wh`T4VqXv7b@u
z0M`QWMqi8K%I(?34F;FdQ*9HDtcR<t3<4$Dv7azg@wBW#aZG@Jb0dE9wf(nOo(kU_
zA0>Mh-}>%UuMf9NaLFOS0a7b13hjc9n##V3>nfSsHFNw^*Z4fs-EKaKA;98l{Vm8-
zUFcfMhXS^iqA_n)%uUV4b4Q9Kvi;<3`wIIyiM?P9dX5+4{U&JWtuKUOie0>#(xYM2
z<2Fd@NM3XL1up5v2U^n!E;twl6xKKMuYM@27Y#ZG(}fbj<lhJB@NpYIU06!WKgtME
zHg?c`yTdHkwkEV3@-YP`E1)@D?<&zlqC4Has%%r0a+)cy-meb&)iG7*EGytFacFGa
zd>Y*xbL5&0`aYqeZeKmBb)i4e!si!A1|2959$W?y9t>q*oN&XPXa<N@1jL<vR!olb
z%*cLFROPq&2@*!I2Y%ELfeZURFBMGZW2?FC88T89Il=jhbYGmR1aZa`Txt4wm6Com
z*7}r$Hb%(dj>EBO@hr3E@P!1?kn<ySlPEEKq9)66wJ^Naou!Q%*kxmCsXjkflgW~^
z{t=nbV#1X>zij@kQyj)$+K0*VHXMh6S5&x+W)Gi_m(R}bFjGFx{Se5R@Iq}n%zgxw
zSzx_|m$aI-?tr*o--{P~p~dO=S8$mwKhS>P{*sO;dsU+>+T<VKqC&Gi(ZpCeQC<lx
z-p<?n=5oW+CaYU+f#a?`@2l2$7M#km@MDW6^Fo35Ju9|yL9*+)$6Ig}+*FhaA(M^r
zxTEQ1=(<^&^g=V*Po!dLj3|uE13MytD5M0ilNdP+%v7EbP2}Y)7qxOGBB|>%qFk7d
zD-Ojhqh6K8H}<{AtAmiH?6ErN=%p%&*;*)F{#2LYHT5)kIl0YC$n#kE@O$UJL=(_y
zq6Y;;T2!wKHjxt$G2EH@>b=P4&F4-ZLYyz+m-^8janBK-v1T^4#j07xM1Gbtr9kP`
zbzP<hf64{V?2#tJO++T3Sb<`p?a+Gx$(?re1d9AE(nFk2_ogWz9{3cOfDl?k5bz2{
zON!5?%euZ1|5y;mE)7E_7NxE#THbTFu3S#r@4flEeG&YVb+2Z6I&^@6!C>P)vmbw^
zTc}@&@OkUYDVUZErWr1E<ecEYU{LWdnc78=sdI}+$};2%LZ^3@S}&6?BvTF%c(`AX
zVXZGGT^Jdi)=+ynFo6hUnu@LE$D+atcU14=9{wtax)yAw4*GhF-1R=XFYvMN<c2_O
z#zhPs=wCvg!OX5oRR>B5WwtN@tZ1i#yyvFvIMDeNnC&Or=cGru=Q+`HmOxH9@TON$
zAsivyAxw{<ABalO!g-tf)(b}Pr5&q8)(zLW4Tr=Ap8Yr-rI+5(=XMWq(&7V9Evn0%
z*5ote<1~QQpZ4`K`E_VM$UpM7)qV?kBu?t)zBz$n3F)<`g2^H+>#`~W;O#+nSw&Xq
z4C?M)OnEwd&R?qPCXJ->!s72R`QtOLzMp_*>z5Y`mk|s)!O#LXD0RU~*NAv|wvd@d
zz|6=vX9cavx@Bb0VG(lUxh9V(M7O0r8EUXUfBx2)J)p=SE1+1@U*E^|AYsTJ1aG5l
zG0YcDB1o*{$qyhFW(bkb+x{i{X!Gn3TrfNUnSfr-vaez3c$ip#6-$VAf4~Mo(z$v4
zp&J#V{Im4Q_wtU0gJ8Qoqxt%|1#t<4QC0su8J!|<HDen*x<M2Tp(0dWTw-dEW(n0D
zfZQ!^jPv$smN?|;lK##eVsYxCz!>trD=(_X>19KrV1I&*FLKdV-JpJ=EU&I9lp`}R
z?QNp&$B~{K<2H~L4*nU9)j#|QXDmGSg+_$1$G@7J)t7rX3fSyKC(8jR5VN;Mi<rPd
zh(OmWk)^Z8hWVk?(RUj-)8IKT#-1P`uty`m@4BL7&m90n$dK5!jZ?>D&ZnS6zOkG(
z7q0!pj}+BkjK6@uB_q#%Ju8YX`Fy?P<A)!9Otgu4RsnFG+F!qF>nkfOqr~>h%U9_;
zpzLHNpG(M0P07e^k+GICp;Fi5d#_x=1MF!MmnXU7tAmDY0W0dWCh=_dx_-f*i6@x<
zn*gECNhuO@l|zyT#0J%oZBT}QRl5|&F_@ww|JZ}za_{@B@i&(@bex3PKim;rCVL}Y
zYR0&{W8$Y{TR*~k<WHiETyY^d0&5Vcaa;hdFq&KJ?<t7=jTP6)g?(hL_8z7I6)z-~
zI(nhK9=B>*uf1`5K)I~m?b+LKKZ1VYhnmY~pET>r5#P9%ACC{##zwMjF4E=)d)NfB
z8@NRB3Hhru(AC54=7gw_M@b-p?x~1hgf&FlF?y(5Zl;dWlnp9m?vn9m=$!vsTuBW-
zf@#eZAyYF=(prf<QXu;3e)GG^OooJ%H_z0i^)Kwd^UL(wJe9d4xl^mR&waAi&=b^q
z{_bdyq@fsU5N;tNLe0c`KOx@)SdMi0x_rZ}H>Zbj{s}NsY$V<vu%5y|6~p-f=A9Y(
zgB7Ki8Tk9WdA^coNnLkVW|v(&leltbo+R4%TonO;nrr}cb$G9OxfKXtir!)V<?444
zJjx_F=<G`^nA9ff9-+gC=Owqa)Z-Nabs?Cyc)J$wjn~wDYi|m#o;%O+XRW_3@5rmL
z8ykCR{&xxcd}%fq?X;ATAs1O>S49w5p~g2cT`0Nh**-5(cMPH^O|f>Bj4l1$8D3ve
zrUFrtJ_TxpCX44zo&kM|PYate-H-ky6TjkcmmNI=+X2PGNy7Gj^PiDlLIC)qO{GPL
zBK8!@6hP5QIc1eHKUL*=MuQJz%Jb&sdzX+n<$ag=$qzEAZWL-z{VErVU#bqoUgdJM
zl&^-Ucuk@VV-<LXR^IZoZH&+7ax~P;dO<a(^qXX+60>x@GHMcr0C+K4Rgj=xR*|wg
zOuJ-h<6Wmtm(2IW(-OY^J$K<Y$Z93zfhB;#)xo<s;8e3a#+C&8cDa=^<Ax8ycD8nZ
z9;H@7DLi^Vx_Htq>`11Xfh<E7mas1n;ow^+?tn10WFss(PWf$#yJ(9{fBnc+Lsn7R
zvm0Cu&(C8Yz8UI#`%WVe1nel&<``rgigoJ(`Y}X|0Jp^LmlrinW=t7H;KiPvM{IA>
z=iGZ*@H|}MfujkBffPIo63lU|%9!QBhY^mqNMT`T(UmG#D}Nk)m}C^o5vvLcf^B;n
zo1as+E7<t)UdkFs#UGnn&0PLk@|&35O36;r(#YAW3t=C5d$ORqD56B8%2;0)D)}iw
z=9;P2-)mDDwKjCIEz&r1Jj4$;urjriEk+y^E`uZ+!rrW^2F3MdFTM-gH*ev^^{C;@
zre{|#A8(`AQ${FR|0M&k3Lg^Ry>0sr8=ZE=Mf2$I(vOWz#pgB*)9ai4CzjstWS_Yp
z`}z)MdmJ?WA-=>V!;}f`NXP)AcVFJLo-d=poRNzuKV#xd=9}JAq!NMY(o~`c6m3<P
z?CJ=G<ju~A;*p{|^~R5XF+7U>Ud7PCVZL^oinEs~=*XVdn>feYg?mVuAa{=n?kX>s
z&)g_8QLr{XrsFp1*KDRduL%v$iHe}jLkFs)WH;|_7p}c$k_%j|`?!`ks<3J_bkJr>
zwxgp=rUH@fT9h6m(&t~5zCaA9C$bAJ@C|Oo%#*^D6K(1{62k4|u4;S-Q(tD$wc)GD
zvyYxwG2K7mqCSdm)?^Ckvk9G77HVrm+;w~qHc+hCZ}{_)Sk_cGYHOvU-PJe4SPP9}
zpX#E!*MmtQo3V9cIddvRP-3Qi`0xCVyQS&$j%?WD&IJ^2aTVr+2s4Bt0}_8swOEw#
zYqmjb6)!d(bx1r6JZ5$Bs*&I$`G>TFu%4iff62nLCEZy%LDcr`i<FZdj6EyXB0b~2
z3Oqx+Nf#2LY)_0EQNwZR|9nB|hqMutx~&X5B`CKqHo}T6cG<<tJ<|N-Cyjx0M9&85
zB+VV&sxN9cIe~z==gO)VQbx;f#=16ZzT(6O-32wyRq|yQh~U>ED{!^Xy_E{LP_<O@
z^`>VP1t~z3R6XneP6^>=x(t}z0@I)!(4RB%PsXOV7n#PKahyh4>HO9)I^^@h=?A0T
zpKk@psX&I&%PC;2F%^35xPzN-6lRDq!zOe}T&H?jVLgY=EL^xl{F^jax8Y2lle{Ln
z>Z)nUPm_3p=%x#Q4C=8DSLMUnN0o;x02@EEYdyO`E}1un&P`m62UDc*oCPL=@;iMf
zIqV})NDC6Y@___kCLGVeTYo!o>{3E4m>hDEsa3afdHEo9a#~wW*zdP##NQI-v9`X~
zQhW6j;T`%8TX|m{ncNjk>>HhZ8sBbY$1Su^>SV7|eicZah%=dA<y*P5>$}>99I37V
zvQFjySgZcuQ2YP;Id<d;h1EG252y}9)qo<o99hw#Q4pmHB0l++4z92HtT`}4$G0%I
zZUoy*Bj;Bj>osvlmI+o6>t(vuPg8j*UbgF%5Hn_w0zU3q^hRGBx0}JchhV0YC4iRM
zXDfnP!zt!nQEZiC4xLE>y^B#YQ=gK<IG+R#%HIrpElak_Z&Hf*3!knAbe`BAymI!E
z4a5akpTOGz{(SmvQs*BSb*4q2p4Sw0bAAVt)T0$t&QBCV8FxcrP_RtJwWxj4jomwN
zz--~t6~tSDa<<9~H2v8$r*{D;3>E=LnR#b+z<(n!P2Lg>3wGGA@X633i{TR2$}ZXU
z@7|?0R(Z|Dg2Zd{(IlHpJu(hQM+JpeQ^AI29k`*%H4R3S5C1zr(|luz$Wm&Cn~Ftr
zgn+O^nO=Z!eE8$Z87>mgWcQ}awE2pz4CsAQl@x1342e)4QEBcU-9U`$63$3e-mA9&
z3BjIM;kIydWOR%3&uNeY&G6ju`Aou4bs6#&Y?#&f)6rY48h3u?lU;YC6OX=gP<lN3
zZBus5^snTuw<&Et_WfRVsi638|E{+x&|`QbWDLboiDe=3z1Y%++}QrRUHAHZu+ds|
zV5uBq6f37h>L%q!>l=ZNXYNy=^s<;4E17@F`sr@Eln0uu_a96QuP<^B7OxY~WhS}o
zC4=t<+=KT1B@?S?>AjCiT#hS@S8#DM=JP_G#{H5PRQat)cE-eGw^Ci0>{{VR7nSi=
z-aoY|{4_fy_|;Kh>X|<k)i-sn<cbw}&F$KHaTA7KZKr?kuSKw%-t#oChwehb2AjTI
z|B|(OBiK|Gw!ma4uEl`D&SaapZ8-_fcd{+FT;;VfjO5&##rH!LZY6}l-l3#Ik9H>S
zU1f6hq&Nqng^D%lsSH_vUo9SOmG^qF`SP=|o|x4~sst2Pk27=2K=?n$xXSuuWi^$~
z%$&?=H8nk^g5=1!4Sat@eaEPgB3o@On2x77U}fAsK-$P&oXsizp0bd*A6qVuc6BJN
za+lpF&{Lte8;c>e6bW1$zu8Coc%D^R=(D=qC;Jm_j%4HeK=JNLp4ZczpI(a>AoK-u
z;Z^NV%f&z_-7H3@rT%i{;Kj7+szRx0`tpKl?J%XQf0bsX%KCTXD#7*~C$sy((pJ6)
zpt%u$op46;1kxP3CXFZ!ZWF2%;iU>lwrOD33}eg{_E^(`%SPh(B(+5Uw0PXzyUBd3
zE=SY1L^DO!U%<(GwX(+ZTi{A1?cY#$G=K5&5|PmxKN|_8WUFD}?F?>FMHW4HD5D#z
z_MVaA{?J?Pfb>l<&XR)%N%KWQQBg}=Ftr~rCTv=yOsRDM4E(c?KciwU_17mGW>C)t
zIZg9C=jD{1ayiSlQGG98<x%D1c^hq~j2-h=qk6pAK-!Mwx1&Fu*|jegU<(kuO`%Zk
zhupo@()Xx6lHy9XNVm(kp3j-<p&@JL8e?`ybavN*fMNZJ525_6L&VVJwK@g&ReCMf
z6Uv4g*xZ+Y_hC;RB90~}ab{0OVm?o5#8r3DbcCvC@id?l>^W-=XNH&4qLjT$trzXK
zNgOqo&r|U~*a}5=*?BfUrumX5P)Y@Mi=<ygf!}fo;1>yei&CQ}@AoP8n!e8JFT{-x
zC`JUVMypq8P_<I(v-O$YKVV(Hx7A|GWa*H3meJZ$r_8!;(eo+QnHB8JCfMkjm(f~}
z9-5r9HLQX!14>AfX8VLxe><j$Q#vMiyPPZ}j-sQUBoV58DqZyS_kc5B#eq{)$HyDG
z&pxiEGAj{_c6Y(E@_2cIc?AJ#jH4|QgVBoSKr+kChMv{d$35X|sYEQ$XX}tjlJNQ=
zuXcr%T}yT!o36H`%E?WZCI7atnYG<;y`ytY$vKX^CYp@8dHjg;8s2^ngGpe9x0$x6
zV9@MDlcavSaI1%%zpx7*d6v3^;u<A~g9BPMiqb*OjRJnR`MUEfzz<fl$@?o$rtnz`
zCV7WGD>Z3n%-iUP<)w{RG{0f4)?Au_P=IgdP_S*ZqOm*9O6w@7Yi`o9H{P6s&YZ14
z_irq7lxwy7q4>vDX|4*dn0-IZE(R3v{bLs3PQ~U}`|29`yCJgsv}`gYAgZSwO)gV9
z?L&97BbV?I2bD<>t!GfBTS&4@FZ}%vT}ngj8AZmGY3QPo1%twfW4tV7NiO9NPcp6#
zWx?i{`p8-|qvdW|26u{Ynd_$9)aY!kjoRG$J9%oC_yFXFodWi8w(F+|^Q2ds;cW%q
ze{B!U^exzVuYa}GX%-kX&)Ujmdnzmxk{F4(FxlMSvGNV-%ZF<N@?i64DEGR#a}Ipq
zRhTa#IwLPcn2>#zI$7m_8D;%eA<k07uzCxq8MiYqLk5118-U?(<SkJt-+Fo*Ux$}7
zE8>k<T$JQ|QM6S8(CjBndM|Jr2?YVE9a@<$YHf{;Qda6~>6*-)KKIYOO!{$u$oj^{
z%kNcTy)cifTp#o&dCzSTn(q7+ku7Tcq?bS+XP_{D`p|W)21l~ASF{rf3YP|)7!>>R
z+md%bR&TfW=j&Lfw~A)9nu{{*{%mf(C_@W%nisRPx_KBSHI4aBELBxE&sZV`>GmD5
zs?eE2{13v%Wwx9oa%@4Xud`dlqhZ$#mfB>{{#=W=%6IcIG*-)yAi{sP$Gh4~<XROe
zEgk-(dcWbJGhgo_OCzA427QS)O!@;4zI!V5W0C72pd5fAXMXPDTL?43tApM~k?J_P
zaz;<(N#-cI%s@`wfu|oHz5eIN!(-mNcdtt|AU}slh755F5jY@FFQ7J(1%xZN58R!l
z0J$nz(k&{o&g<9CIX<?5M`28>0_0wc42rs){GXF_MvJ~bTuYv088y_L5^)#fO99TE
zAJyCp-N8@6cA;N6P1jleAe~Df5Y_VFEmFL)^`(nk%<3)|uRb<wva%q*#;UHeOgx`j
zF32NI7ziGhQ@qqZ<|XUl=5TP~_;9T?<|&RJ)l$a0z&Ub7`0kKK)M`L>0g<>p`%yra
zSmYU~8W~$lBd21tIBQAbbi1g?q%EprC{g0D7j~#qXOViAS1cLzllJ4Q8z=9~Bk@L8
z2w%vwZ=7X(n2UEX|8Z5Hm+YT1+kVlHiLk!;{n*eFr=HfpnI5WOd-_ua5rb{9t2KPE
z|LlNEp+21ZkCfn8(J$<%f7Q|=YWSq;O7<c&I)3y(X;*tQd{X=M6I|2x2qUT67tu*5
z#`7ot-)apRt9nOw=|BAR(hBsW2YLi1d8zLDh;Z2<aL_LMp3mCWYC@@|B3pkh`-8%T
z`3ShyexEm&)Q`AXo51202UQttc}>HspX3*z@BPBr5$eoll31@j5ZWJCd84=El_cm3
z`0jzS&ZT}p(h|UwvB*M@jF2x2eeazM_5AF0uq0vv&m=kJM?ZISKKb-t9%3ILdkyB1
zA{)B7Xfe4)>)<o9!`I0{<uW_LkyO}ISeJ?G(HJD9e#sZA`9EJ@UAhB(v_Xs3r7+QJ
zq(up{6nSCaNdq#fk{<APUm>qgn)H25$C%^zMdRbI&}9)1kX!%@=`L{1D-su>TcZwF
z5VL}e@XwSQ+nK~Iy2dVs9*nG;a7TDC8<6$g2MzDiNKNU6_1D(d*M9cVcC;7|A$wJJ
zBmR4^$O#=G@*L99{q=Att+N8T`Fbs(<zKQFRhy9}n;||Z{(R!zoYpoq0&9l+;Oe`E
zjP=!N>`ABlbef(6#*esDIxZ<74G8>~<}lf-Jk`G_avVB>sPFwc+j$~CWk+^BW0Ngo
zK2g*LcU*+*qZ#~u@b(H~j}InvcwI6Q)q%EkuKmI=%jIsM>>xvL59Ka%v(6XIve;tJ
zM|?)vmWjD0{*;{wAW7^-tmjcG?K^KhjfZ8#8ojkZ#mLC7j!%u+h74Ta<V_$_qhJX$
z86qIeW<m`jI{W$vO#jECR?(a62bGaVznT&b=GBDn@Oouyntp6sGwTY`7p&Pk0}Fq(
z@_P?-LE1nq{NIdEiY|0j+J0INp^!xBjVt^e6jEF4#%|DpU2sK1yw}u=x3R!qB8Q3M
z=X#}elZ<XA|2~s&&%Q8$4iAKzg<82GZo%Ks7~;bS)<_#Ww~SjqNTTptI)z~m%O$>z
z+F8xn@*-0<+sAl#)gLT9qzVU1%mLA6s(b*3#){(HvJ=IguDG#-qC_8zmx(hv4n~;J
zev1FG92?fF2$(&Bdl;okN@>11Rlh6P9{V(=uVDI#(JD3S{fJIMO|o&dyON7xrYWyq
zw8*fkV8|qxHiX)(a+;9Vb)lIVEWRREpI*S>No(}zPisIg+a_CEPbbFH7bmNd%1--k
zpK<OowII_Yh>Hmmgd_qCAU+LB4bXTgJt{3px7U5@6Yb`?V_$hzHC=og!Hs`Vtmq)I
zqcM9Ie;x-XM9b`3t-KjEnuTx-1knE$95ZBn)XvJ8AUnmI_&xfd{L6-q=?-R->nt{P
z?qu5+ZaSaBhgV^U?)$6kpF^m4ouD`AlclS{5k0%mBYkZ*{9$eBz*l|>R}rbuoN-@{
zU|z%QgPk|!`|gge1*v9~i9E5@lFDRce96Y<gJ|r_CDvU#VXS5X*1YZA6O<fb=y?07
zrP`aCavqt5qQ!m+tSH~pTPD@>_l+}7q+qqHxUbj}n-ast<XK7GqvWGl<K&~KU&|^2
zU5E_1_)hHMzWk*+(Gvx)vnC+B*uPt$5S)Fre1F|@<3}e<yz8`MJW>p<L4KKoRaMuK
z(SX7x1%TP0cS)Tj37mZt5mFKWHJS3Ywm4~h^k&qM&#x|%*cVe%%(lRr2NE=5R6on~
zW7*feNTA<wW*9DkI}B%s#wiqzEqro(vj8xJwL4igm5hOn!Taseg2-^0|Af{3|6bT7
zf}qx(tIMTV$na5s!Kw*r<WN+?+<fdnOgG~4nU_9mZ0c4;^<#!dntNf1cI#?vj2TVU
ziZ}}^iwvn8OFJpEH>6;q2nKWy1|F?YmR}WtR}xA){NHUiW!0S|?^Sib<C*`I>u&b4
zQRgOg#6M~xa5{S|vK78u4*Es2p4Ahrstn}(G(szTphW?N9Z}93fJC*68Uj_>Gx8wk
z7BP+2j1^iGQ+V(#f?(R|UpF#Hoj6F{&lj&f2gHSb`F(biGt5rk&4ZyDoE%Ucxx04$
zso2DBEz>i2HCPo=qS)NqD~kwprrFYHh!9GZU(9}W>-J)<3SFy@c}B>_wu$ptd=Pst
z&UJhz`dSQXFV%7UVq#8mP&DYh%Rxpn#G&ToNBX##n1zI_58>hPN<<}Q>9)OS97UY$
zmGl$I?m2G5n#GQL<Q{%m`SAzS!*1o?fsh~DqqcVQ>MjtE=DZA-+n-hXNTz2~M21V!
zosqIFl<cVLBiPe=<~)mjm?8#p$(3fMob>qSxcH5CX<`hjz0)31;*Z++7H`$KHum?P
zDd<Jw*ieqQjS?D}zayR{6PKQagsQQ3mK2o~<x-OUzVWMAh~6*S>U4Pb2Xfe+AhPuN
zpO&jaRrYNLeV?Q4Mb7a<OxTTH&5ZYN{-B!io!fo9bidhN!Jn3{p$Q^%zuzO-iHd)T
z!TCKd_=LzR8ymMlo)z`HKKe^FGqU7$DJ}8QC$kmaiM6d2T;-IOaYEy>_{)ieI2;AJ
zHB$#6Lq=6Zz+*>D4z&imS3FoV37!456jW0iV8H&Xzy_>{oEDZ#Cq_OM>TC{%IIoED
z>>7@{i}}*1Ua-_R)i#B{4&4wSx0BZ-x3plO1Q4iDVep**DmAS9C-)Ps8#+~3QrsJz
zh+!SWUmEm$)l7Hy+xU_UjSLL*)jif^-&pB(zT`GLn9SETjt%5}m~sf&_MMs<_SVy`
zsUIyHU7WlPuoHY`UV^4Klm2)rCFuwh^Cc%i4(JGcbTJaObT^5bBjRQ&*0utjk~Xsj
zBl~e?E8X)o!V6R)-9uV09@?A^!oY4)OXC@r&7|9>%K$?WGBi6C%M2tZP4#5#P2sOU
zF*TaJ89zu_yh)AKUH&4)<|!E(0*!ShPkYxopMnPcy<%yn^eyb9&4W7M@djH%k1J2}
z0o3d(Vtlh7O^Kom@Lv|#cfrlA?9C1p7HdW^g?HY2R$w8JAt>P-D%X1~#aFHUHJsP;
zsWn*w6QL)s4+m`v5gmi+6xjzOdzj7rtn$}mN&1~4UiaEHTE3~;r*Q1YF7Hnl-rQ7O
zzkExS9U&^M0wFDgNQ2}TW4?w=cA3F$vy<;qAMaROD}T}RrdtTa1YnNVyRo^hqD*oD
z3DGYz)VJ+>>>j;N5#0FICunsyU%+$*9puZmmWj6L?(=h>Q@^+u<~h>+!eai}_k1Jz
zh++}VjJ0*Mt+mW!oAR9mz%v;dLh85x?Fn2e>tQzrY{pyc3+LU=*jwt^^R#p9OylIV
z*w-|pe-(G@h>yx??&qQiF_%JHaC#ZVa%K^w80%Q+X*bDp{E%92+x?XLM|Y^z4utjn
z+n{BHT#bpG$=Y(BRV`Qgq(01}r8eGYM{3I6T0AL^H{p+OrrcUiy6;c3;A&Ozw>g1(
z;@r>Wo1^f`r0YJSK6&Gv-QL&ak%DrUY6H3Ylmt|}adDc_1?tYbvG^#DjCzy9ZRBEz
z18D;O>tC{1v@-xJg5#AXxVjMq+boi9sa2NJ!MV>;73q`S%!zUt=_fRB76d*i%>a<F
zZZOtJ?y4UAdYEd?dIs*iC@d0C!FTSM{30eUks5L0-MPadaF&R)h=Mv2WQzjktgu71
zbe%C^iu|>v)-@~G<9oxhHBELzt8^j`QNuKZMGHK=zE(n@`;1=BW&s+*yc0_jdk?27
z7a<bPomjf9`l`g$sF7bBGyg;o(IT4lG;Q`yJK_L&jg&YIdkv=RnD<p^8lR2h)W4ch
z;>$HXO(C%(+{`v=TDM^p@;4%oMhbwfq<1w?w#wF8XWxDA5Mrwux!bcx9FzZTf0H9_
z>CI#E$jt0!)G)U<VaoQ!_T=4dD;RU#R^{BBX63W6pTe=%2-l3wXI=@v*R3jQsw3gy
zvXjNb@~Zp4enyMj`!5>$<2$8L)pyM|H;(B@l!>?`aEyH&bqJ__-Ule#G6Jjo=jWU!
zWN%{{5WGr;$#3eS%f~qyr*@A|>PsSOQ1V+|Z@{%_J$F)fF7S{gF1wZ_+lM`BRSzw`
zx{%)sG<7)pDx{26SkN*S?8zW~#mw_PI(Mh^oJ#>`PM}*6BP}Z7;nYoW3hw=g!JVbW
zes<)C!&>{e41t(jod4&tlLZ0&>~<)vx*+@OT)s~!4Jq$?(^t{G--K^O?&q{D*%`0G
zsS;RsLqOxNCQelVe7vI`>UDDxLUrE9_gc}xH$&d2qLAYQZQk`V#bF@pTFgtTB-`@C
z4b|Emqx=COORIyk*2ABdb+vUn^)7|F2+@3JNb}+uD&}Bzat>+c`|2MJMbP)+wUCr!
z0J6-zgb}pxjX1#cU{5m(Nv-E*@)Z-iYU<(Id%hR?Y_=2m2kra&tyFf%Kd3{RZotJ{
z?h@YhvUAL|NJm%v0x94^NXK}Yc*#t-F&M?5O8aI=0%)MV7e>2UBWEsGuj<=vb+_k!
zgXy_9HalMSE;av^O`-8p$qU5XZ&DL2`G9fJqHO#&wcS}&CMjJlToTur{DGnZf>Y9H
zsvHH31!5Dj<i`(2NcRJR_Fw3Zq(b*uU-jM(^(!m843Zlx422vl&3cn4d4CY!V*TS$
z%rJ^kUm#cPZY|ftL+wDAB;&g`cHG$kTnu{q2Ildj%Ff?3t7>8$MLrm4ttopY<)>@r
z-=5L8Gjz(@n|m*zG-odtqSu`+Wf|-C!pzt_dF#JlBkbjZuVIbxbp-#*XTzK!G&tG1
zdcG#-x?d$<^y%g5CEleTh2U7YU1r4DOo)FH4{tb_sm9r-0)yOuoqw|nIWxDKM0`%j
zs|#UaH~pI7m-R#68(QjiFYS}sMoVSO7i*qpi9cCw^PM^FL_g}2-_eJ8c!hU3;KWlI
z3TqPJs#7x;xogsI#X0Cc`(SGoSC@us@gYS3rFk^`=)Wo`VBjG!1du?5dcgDvDcCf#
z>ZaC_4OeAn?9kB72OeTbr^E`M!8YCBsb`@^R4(*-4JW2x=+r8j$at?)2<Nhjh(bLK
z(DLcSygv<otj^m%+j5tu%g<k0?!iCEO5WN07k6{AVnT8-A$o@LUFHLoX?TWc=ucxK
zyJ5Sy#g|a?KB3vU3T3|sFAIl#gHtZ6Csg@DCYSDo$XMA~L$dFa`bG~Ee<thvP1~y@
z8B;R`8o0budw>pL#J{vx1jfx^BLP&wf#8Oc!e%2@b@(iJM@CKK4HX+K^ggafW!}FX
z6Zq-N^&|~{D}mTDqE3#bo1fGjF^Qj+7}ZV}ig_Q^<*5yD=P=>HW^9YsrHdO$G_re^
zQ<ZA6MMJ!me<szdY>uxweUaZ#H1~N~mnrdg;kBVrKf1|kw}APAl*fwV3}T8_FZEIf
zh9;*9rCXa24-5=h6odV8&E{&a=*FX*s443v&6bPTKt&}&m_dG!=a|ew=O*~I9XdtB
zza%p%s|osQnycDo4a@=qKg?8F*=@3)arf025rEee9Z6(3T%T<a5Rpttd3u*MmGhqs
zRJAT-F->A<jvMAWG#ND?AP8@7#vLG566jho3~CDpIZOF#MxUnL&SQ>H;}C!4N#@Mj
ziyQ^bB;zaoqFwdbee2?C5cIA^<wdUdbh)*!E$eVB2mVUBO-rVTN#R&W4oG~?7Mzly
zi<rGS3qSehd&pqvdaR$nU2Skb{$l;bw-pyYa_Zl-xhhi9i4eRsArZ&W)24b0SJzs&
zO`;<p92geuT=Cw@sb1PP2X!#T*AW)&fr*E&@;}q+-U8Eg+eoWFc8fhI)fDUYgVmus
z1Zp$4Tl}qT!S-T*N*qo~4yopX%`<i#@#-(f-B03F-<LLGYX4{OiVdF^&<9C#U;l82
z4@-v3L_3HGp;QUOUFaIQa;S{^>K|8b>WF~;zCyz5S<NF>A+l0C{S8}%K5{Es&z}w8
zSz~-$GGKQw$M#?h!v;4JOJM$5nVqkNq~$(hE?q>5>#&dednYTK#b@N}pGRMUI#Aq|
z*o4SEpDgi74leoU_3@*lvrKbbEs{mgG`Lbs$S8gwJ8-O4>8iH>nGy_B_kxtUJ~e|A
zxU}b*OcrtJc<?>({|sK^|KH3b`TsMk;C2TOit_3Qdik@J9ePzM&IMy0vr2gH94Imj
zjyO0$?$a;}aUM9nE#-SRC*%REab!>O<3SGC2kAey?zYmh825dxaJ^C%4P2h=HdKFi
zQ9JJ%r_?RSVWfeu^X)z>t+Y(12GOCk${u55J=29C*^(>6Uc-3AFS;(|huOm!4<e|P
z(4p^3bE^`gcgI)o;klag`!^jn?K|%tjeK<ot<Vuh-IM<;(`@T^tlQj38l0#(qhD~n
zh&8fUw3cMMhDkTlP`JO0^xCy=x<_<0gDz~Rp9SJB+)1^^X&p!wwdG-g;9=JsTg|}x
zq7Yt!p&IU8n^NGHsJ*qu(F4xws<^nfYAPS>8Y?U@yd#Lzjm3Na*gnIi<m>C2RqN`!
zS@JnEn`j*#fS}n4{sw4ko!SLvrM}yDJfM!!iYnK;X;%>rcKcWk{7pjJr8+P^FL`+m
zNx3CYGt|72{}{$?0S*x{BwS|lrT{aLRza;VyzSlKZ3;v;6wP0hOb-nrovWwYkf~oQ
ztNyje)H<tO*bijh4?@qa!$sF)iHaF@_#?px2BkXDw00QvSIEk)uIq_j+Mz%6puTto
z2n``J#9+|B_02VK%$3xO%DBlGp@zE2+9yVf{l?WBwN=#re9E!Oi{;Dm2Q6_+o-$}v
zH8xAcu&WFycQ*jrZbc#-r&tl%VO&3P|5&UbIsjn8IBs?MY6=VYUO}&=Is7t7q2^=c
zY2t^Xn~#H%foDnUACt%NS<uq$<Jko@V0PNLJLBK}kE<!H4k=%;CZTE+p)~p$*HE#<
z^s7p+%QP@y8W{rwjvTqvW6c`M7MaHR6Z6CApA2P*R_NXNNxHt4&NJ;V6Z$-`YbbgS
zwQb*yc9o6e9_xIpKM{Ua-g+8$A_4V=dfiRs_&m?#<nxcZh|vAeM`}DM)vIdwau%9a
znfy{IV6|J7u5Ln^r8L0s$#`v1vR=!AYXG)QbeHp&&7D_w8pv<SYMD@e3>7ZAx*@be
zIMR$47i(pank1E8G!lULD7}}H#c};^nv(v>pDM~Hj%t0~_cF|;2@XENwXdfP-P=sx
zdE2#O-{ZM({+$W4@yD1U<jSS3fZ<=5SZ+YUnQ`?9{Q+`tJDtBKR(X)~Jv(z}f!y%G
zWs3{=`5OT72EvQy2nfUGN#&QT$a}74Zu;E`!CtO3*?H%^$D~2%y&A`s_o2U7gZnM5
zXOUf&!D5*nQzP(nTmy*8n`l<9DU5gTQ_0E(HU&{q@4sY~6KDJN6@k?{z7ANlUW=sL
z#dn51=v^K9&Wzwn3jZ`js%uX_*_*#<i?6PXDl36eGtvJuYDSIIOEbqv_Qo|6TCy9l
zlwlnx&M_kIPpj=F{kZJ(LQ74a`r4Y04iPuGnCPFFa{qo+F|`?pf|`S@XAPg#*`}SJ
zfpZk}K4BYa@9mtZZ0lL-912|Uj{y6LozAINCTW8ES~G5{9TG7%QZVkLm}Xl4(WKfo
z&}~KgE8S{9wuaqY|7Pq|Z+$6ktpn3UO_T~S^}r&}&g_}P7p=;h>{r*3DJWIaa8qFi
z;_lAl$z(TQx;snHr;Ie99J1YsB9*Rpp3v^T>#SSf5bLC0kLx3@uOe=&0I&>rV0lwe
z+$zfL+;ld{excc@uFiEY{Q?;Y(=Z4>x;WCaZbOwHg3NOhg3b1M+r;<S1-A9v*bGGS
zqN!B;COkwB#13*^J?*<0`b=TBw>K}LV}MO`fhFnvHT!`9Hs)He`vuY6^}u9|*WY2X
zvX+h(*@c*;XeZJC-E;S7afVQ5<oQcO7`6r+@0#DX+PE%m10^TMyyGHP`D^Cn8w_`_
z>rX}37M-B!rVT6HjNBiJ*s$)Wd=%k3y6Sbj&%hRMeN)i8D;R`r#!F(;!ez7Qy4W|8
z3wBkNN6uP;y1UKlL|Fr0#vU`=a{t)YA&t7B#C|2+PNL{>3CvRV=Z)u1*6-WO8?)Q<
z@%L-J5HGj2Uk%8N_B?kN@&+&JO1;$cnYS`>+b%j^kRQdncwvoy2NiZmL_9W#DkmU}
zaPh^CY%GskPavFA9u}47nx45zCeljeWYUcT+$GT>mqBrYW1)pKNje@Uc2M(U@b_J_
z6ssg4zho|mzx2pVZ$t!m1gzM?nTBBoxHUL+i{R`+r>Y0SMVaS3%yDn>kSBNe@46#R
zYvdVyWiJ^(tw5~)C37uLn|*NU*y&>7pCFlgQ^@o2{>_gFBN4DWt-rxVDHw&n+#{_t
z@(&D-5$lNqjD5hc0QO*NDUqj!UFg}z{kvL#3|GqaZ1L5S8Zb;F)$$s(LxXmO*tAQ@
zx4b8}SE=GBHXgS<$UH?h1Jhx^*PQ>>mb0ufWdOMebsQEj*%icv-={zO4nc-a@viF;
z>f65e7?>X-Kg$lvhCP~#^-9nseZkmtXG@FaD?k~4PPs5ad*GVzXP(^-)UT}#JQsyD
zDh5U1Z6_`gVja-pGNc7~RnNDOctyyst#)4!*-H2!Mi6Lkk`IT<QpE{qj`}oby@J;4
z+_SXQfbtVtNJp<VZfTuoZoEZ&yU0M}I9BQq*xZgpHx}H2t&Z=b+M3$B*p?~pUrLVl
z-U-NmSiewbXp{<1Ug&eDJM(vr^2y=1#`Ji<i@N;)1leR#5N=b}S$4LgqVAEu%K)F=
zVSn{QRTL&)Nr)qbVGxV~!<wV^?PkU@@HchY`^*l$kMCZ_@~D5hkEEu&YlaRYaS;^(
zMM%5hEmsm378R3E@K+mMJt{ldmzxlCcTeNP@IGO5N;06}OCFyBoTbEjXLzeA@EgO~
zrS_0V0SBVWbl3u?$%5?nmdq;eE><Q@Ya<ro8PA3Llp5yAQe`<CMJ{R-m<Wi^7Skr(
z-QkM%iI9pkc=)~}E(F~Dv_MWqsGnGB>(fk>fB?^vJpys-zkVRVl;=o%3VDX5ZJ4_F
zG^_u4Y*SwT*Dp)a)uTC|67Bm^6&4J1ygA6;VcOeeFi8*h)D;d(mchDce>%QLh*RyZ
zCq`{@5%<Zgy?zggob=!irfz+_<RWT(0mwz2k=K<R29G?*K7_D+aq?$lI<GF76RSn0
zu+DM1F=_p{aUC6mjih_a6*TOsFCpOzm<>Eg6~mb;!WmBy@6k<5fMqh(uZiZ9oEdW+
z-p_Bj?8Vbig_WDh?eVvC1a@58F&{6-ll}!v(^k4yBO*J)2m$NzjtUWqmjriAuyGIb
z|LN_#qMBN_J&vM+f+B2sRjJZMI#Qx2K}4hogpSgrBhqU?MWhG<8;~y2O9;|?BoIRH
zf`lH56axuJ2&C-0_PFDmea3ydPv;>S-&o1h8gqSXeslik@82cfhM$e~?l5r3w{7|=
zn{_TbrVc0dwnb`F0`u_O^KRiz4{Us2DR1tt(zN=n&RLTpu_hVoMauWa(xrX+nbzvH
z9Vy7eH2F(otOl<Z`ii}4B6I*9S3I6GnWaxCKKqcVDlc+o{z+|9NREGK#M=?ym*d@i
z#!fAeAfv#H2dWGr)J6Lnz~T3%OSZny`%9KxLjvDFn*9=yYxO&lbfm;Y1lSKo3eV=*
zU9VXu8d)#9>e2+4^M5WmANX9pK)wbPFjhyu{bz^pU+Wqu@aW|9I6J)V^QaJ%-+au<
z@0c<`4;N6yjrem|Lnfci`k%;nn|8<pJ7DJ(UEzgGbcBR?^mL8JfWAU#eLbyOkLs=g
zDFStOswEEvM0U-81WI%Q<v&RZF3)`=4{%QCL3<Agt(7yBE&4%<BecTJ?&|lN23svn
zd&|bJ`CeK`{b=vZSt&k6qlVb`JuPOk!A5f$mrmeBd1_vliVejUkO#IB=1~}r;B12%
zKXs-$ZWdvzsX5^KmTaDUwHjgWl(!$j+SfnIE@Z7s-zDFHzO|~>-AElZo@bLUm^Kw(
z^{y{Ck^@W=R!!FYGMcV%xFd__j{_bNA$yaHfKOhamMnfBc`F%O55>hP1V8d(>03Sr
zc!7S|x1W`fHOBz831phxP}k<&UqjM62r019K+8>Vyh5Dy4#Cze87cDwknBZ>HnyGa
z$r`SJ`Wj?tOcJ23>8#FF0Yrt^`#vP{*;*_haIFuj57Hk9&>y75GC>sitatPDy|cy0
zCKW!5$}{u}>%RTFb}`TK1$qsJ%hIz3yCQER6Ow0|6-Rid!0$_79m+e)VGdE&J7F{>
zUCZQ^HLOUccjZS#ZqRM#LsKHTL+EhC-}G;nGoz74zwIpv-~a?&sYdWCLf<UcqLZov
znBusM2*f<!l|@1BFY=1B&BQ4kZqJwT><>3SMZfRoaj0Bej@P<j7yND`LrtW<$wa$1
z(d-kn>E1foUkDZsH0I9pkQA^!Nka7;qd6;!$})~9Uc1Ip4|8Q7Ux9WTrOrupTnNuq
zxy06!tXF5~)}ulJ80Zo|**5Y;K~9)%bai>vP=<M3`}~(3vS3UZV5>I!2DDOPcYVS!
zU+)$?lHE-~gjJ_U*Q?6ic>Ol=E0rjBQT6d}j(sLVj-mKwV;B9L5l~zDPGG?Ofgi)V
zv=f?vN_JlX1L@9;A$xF`LHkt4(^a>|A=i!6nJq9tVbCwV!NV*&dskIemQZVQhU0gC
zfD4jF-)Sx>IiBsd{i1RL29i=*7?VkRTBV07F7d~>DTx%AwFMb?!)kt(cM}|SbN2N#
zK>ONHK+-cy=~eT)fgZjdk|RuPDs;NS(}Fr7f@!M}_pS3Bn;UaBX(}dP;jND7BlttS
z?(4tkR<Aba1oVu_^f5&#_qF^M>^RbVc?WWRhp)ldAm6z;X5DMv(F9HR^&64D02w8q
z6GDR#Qr&DbkTa@N9|fUV$&_<~AO(v|(ACDOlI!qAQ{n?NH1>k)S)qBBRpo>#fG>6R
zC?Sl`R;AaRCIl#xzT<9OUAnKKe4z2f$%gCFiU2%alkr-L4dk2QTo^JRfFOLxv1;s^
z_59sV=2KVq0-w6AL<vX^V(-^>&B}=ODBB1WuLee)cc6H7lF|t`_I8ok`j!?Xh()bO
zguZ20yIb*`gj1>n?0dHiz!l=g1VyZXB@1FZJb*dncL<>)F9H7u_gy4SGM{PJ%W7oZ
zsXj%zusNVb7$5>18y|Y9N`il-+YcSuz@=fix?J(!%Fe5HkD5O1$gY4R*Vl$uT5&^R
zmUGGo@Q^Qezq5SLILfN%ZRwXv%knaP{Y6n(GnIiz4>4F^+G!awl6g{tp-8OY%XVmF
znj;^Vb(ng<J2`X(<}-8t&kER`19j+o-tV<WZEq8~>c2EW*%Yj4T%>v_+hX2SDQQ@v
z%ZCt?(6hjB6U!jNpzH}dw$MN%BTk0R?jhx%MDNKHKd<KY(59dUV;xIC+i7$mD4}5l
zWP|T2d=4D9-7<-jR*SmVkpt}U#(z@pvyx_ky#FGQGwDOm@7kT8I-)+}%d_sz&M#RB
z+s#Pl!eu1W<uC1~A)9~TP7lsSrKQE6`mIHLAI-$inoi5WFocGj5TESxsv*6`>c8xX
zada%O$jI<>fH$#F;^~#b3Q}_Sod<W`6}X8~kBLenBW+Ao21V}ZO-GNL%&>e%Cx&aY
zfbf4a*G!kkq}2Ew6~(Q4oRB_f^5;~8qngB^+=lJ4LgO!tW_44kpxnzE+N~b_$eW07
z<iQ?d9$u5sO#r-F{d*R5$EKJ)YR5GU%SfeTINpGb&R=SMBpW#n0vt7>vbv89m>kZQ
z=_|(M#GTcFstkBDE?zr1QB^fgY5DmP0KM0XqdfRnFxa(KT3R7y`xHdfA%4kxxc_rN
zv!ThTt^cBWO3c*NSBwX-$F*8PJ$9hAj-8TAEgj4+-d2Q7H`fQ682_1x5W#qnJK_d(
z=C=Kq&G}f);bg*mPK6ADhmcl5A1DeHuBV+JYD$WCh;?&z7@IzkcRLra7N^|F=i#xa
zU?>+u+yp#~N#ZjIGN+|}STPza#2c>kp*WRk3zmIAMvlMOXlj43C_N5F1ma9r?^$Yl
zT8dJAgl`mP%!c&u{z&>xwfso9A}&>hli20zup6TkxCA(b--{cPJe-)h72H{HqBTt6
zoACaP;ITvp6uWcsmUI-|6fHqMhLI{XL664Y^rYVWi_Ua8Jz@^nH@m6H;Vf%HKlrej
zi8A50R$(e=j9|f)GB>Qgtt{h_hwtSDd3`dzaL!6Jh0)>O6_04@-(B=vdj{H#@F&y~
zSh!B%{E){N%H0ki6&c_8<Gxso8#Tv$1JKEYeCV#xuc~!`B}PoxR~oJTQp6)H^KMEh
zXB%j?Mu4&+#dj1{_gSP>W?mRn{c~~fPFH6!Npg6>9h06>z}u2=JQ?HSmv{rd^g@l>
z?0gdV4BUa5ig>mNrCnl(YYdZtc0(C)-Vn0(0c`~4(_y>OukvI50m><UVI@x1;_SnN
z35zUV(D?owf_Cu`Z%?#{$M~G3XyQjR%dHb-YlzrGQy&xW&1Dre4F>wOC@D<UW_>cN
zyZM`v=vYzO)be_heVCSq3gJ?T>m3V;SIam}n!fXKodD>`8@;dZW-#0^9InWsFka*$
zV6O~F0iytfji#i#1bi4CVR_PY3N&i|tyrgzrIvKD4G`v6RN^&6eS9yM%rrWT&yUE7
zscxRraQ-^FxYK%#jlppeAr~NYc~=V{flY+#lrC+DZf5i`eX9B5I!v}KI$Z1L!v-Z8
z)U=voUK&KUL!xO5z^I5Kz>!goF5Ar)N>zt)2Rn-@Pjqo=+FjuBP~*D>+=-+)MpR2r
zDo7uHoG6@OcA9jv1YCqrIT)Nl6xA6vQ3WU+Q)iB{?e=?{04C1i8<PfC7qQBp@!R4S
zM@%_YA;Q_1)4uzc7fX6wq;k~mNlpyLra$}kn#I>te4P6$1|&(#DvC32NdB<t8(u$=
z6;h-Wy!DD@-zZz;?E!BSaeR-zzl7ZdObLPIX(`q>r-o!Xz7vt%L!V)1PK|tBTsxu6
znqv2z%o#a)CmC&+kR4eVR0<s27+GT1$!13j6On6XtsnlP3-Vy~7BNfllRaZ24E`Zw
zCmWDE-u?VVd<~AfYTtvXKhnvzzu{-2dC|dL(&(hC{}dnwF5)e_uF^&WgED;=bgF|#
z9kG0*(mt=m{x1@Dze&|`Byy--ANt)-zgYeSkzD^6Sp(vF%43bq@tGuxu4g0+2tm2a
zQ97Ql0n662z}Nr7ditOLF#7n<M}Woj-#cVkJc~67uKtY|ODZ^IX`(=sU^;Y0vB)-O
z#KP1s$im%L3IX!`wR*rjue8Tj5&O<RaU*VymraK)oXm2@yLeY8PGp?ogvA$TOaUN7
zL#vZ?y}dSK(ULv-qyhvB@Hb-M0{}S=Qy>C4;N%v`RClE#3_xCAmt8iLcmWI{0%QL#
zDFRcDecSuRN9U(mCZI6St)(kd{uecdtWEW9C*O{9p^W;H%=6gk{xvr+=U|C-T{rW8
z36M?h(@(S7e~e@uS4w^ZIo*KAIyKN{CO{1ue^gp6zdKN}4znz70K4$G{$vFwrqM0P
zF;}<DpPsH)vT^yM>1VnwP>Z3?aKW-ZI42-xp>f&jg9&EqR%vsn@Qh<~$XN`xDE%la
za%+=SY+qGiQdLq_oF6Qh9_szUwL)hvf7snXy{JE6jL#109qh7fJ{}%^3^7-Zv|+QL
zA|!_IJ``yV5x#vi-}>i4kYsN3Zv^F!=5KYuv1l!IQV;x|Sei(hjccfR!EN*G!SBUM
zJ6Owj&Zi<ZGP4exj51LRs3Z*)FiD)i19uG8V5+fLkc!p2Qc+R`?)f;EL`b@~*Qx|)
znA7o7M)=Hbi*Oa}Wm$<5H|Mt245nCD3^?tTg!KDi?Q;Y4JSOtsi)i|o;84~tn82=~
zbe{;0#V*1iB<PN<$J+Vn$+~Skn%S0qP(?j#?enRh4Y3eA^3*h9%6_4_P9+18x&pC;
z;+>t^Qeg0oPuA(W2_g2L>%O#*>O9?5ttuC&-};g)D2AxA{zC{B$oBV${dowrg~mZL
zBnFZy@S|ju@yObU_5Ga32J$nbVVSS7!GHGF4RQ@?2Z;~bLx;6Yf@_3;5PF?Kx<$LV
zP3XZA-w=d*wAAZoG~BSogm*dooaSFqnjiWy%X;|&P`wZ8hJ55(jJj<fu?g5h_5|(|
zHSqv*ons0O0WoU`gOYAI`4gC2Ia7Daw+dNNGd&W{M(lE)F?46?Pytr4$aXlpE^`ou
zIcVEQJm=Dqb>HqEz<VNml%1T&d%U%2amU{=)<5^l-bRQO?f~cf<v4V4w{>q~xc1L2
z^p$NjU5M1+Nb^u*M#AD8@B=FRA|ThGM38h=#{`03x?kO*N!00+Z#DK1!=m}vE^M3w
z6nRzhwLMbu>5j|W`KsR=GB2pl72EGU4o7J%9DvSB5=geN$DH6Kfyoyy-Rg#pa3?Me
zzYE@&s4YeWmm?OUoC-*Pux#6L=yHjzRJY0e>I}^n1=iUAUFYPVCA0@OsZF|?-^3A9
zpQ>B%(`@fRbS!}yNM}LJN-)ki`s3@pk=6nMh2KQjrBv;IZyke#<T%3^MQ9^aV9N>e
zneuq+n8?bCZ*m34fgy#xZWnuuHk><M?g24bsjK6xEndsm!j7(&4N(g=hB|n%rcWY$
zJ~~;@0u$w9UJjQ{54<8d5~gbd-1`z;CjD|69NL;FnI%h^4+hO4T0<ioc%RqMl`MNk
z3ePH8U3W5IMEi7!UGIp6q!*yCC%bWp{*ejOrwJAQyvona(U>L?>YEOxtEi+xg4Cnm
zWbx$2l;l!!2K-Se%eAMz?_{fNsc%gpp=cj`i=8b(h;-qY!a)~)4R0QO0fuFYxYs4=
zanG|u!Fjd#Pze|MN*m4ck^W-gIl8_FUPY3z!%j~R@^gSf!w~5Nn=wIgTNAj_xBKh4
zPj$zwlAQ@%tvt!xH>#fC1K+#~sZdVz-tO+LsDgTi7|;uQ0e3;kNiOplz`*!H(^w5D
ziFy?Ah$(>ks*d8^ZS)RT#tNHuxxH%wb_ve*43?XZW9_GWg3PXHfGbi~i=qc|D%(aY
zXGX%=u_fddS$Ni<q1j+g+>kZHllsl_#(4wU;USvF5u0Qlay@Mo-W$A`xaLC<4ti1@
zk;h(mx`(pE&wd+wFDQ@TY}4&Q@EEG?_$Z>MuG`xT1#G|t@Onh}_`^s!_mqOA`;Irr
z!a$}qUX@EaCfnPFvyc?vz~jc?t|iej@ys&QV5(HiBeB2SkP&4}&lPwHWh~H?>(I#$
zinpMcSv;@(;Topb$XzD>LTnP!Ze`z*-c~>I7aas&W!)YM`08o-PlHx+xXeM917z&6
z&|Vm5buZ4xQ?xw?oMNKS+7dyYk75PO6-(P>eDjRhXCgs?JAMjk(xK)hBXNG%Hi|^u
zvobUQ=*laab};edOexjnKeMJL)b;7y23|#MW`UZ3aAh(i-OR*YjHn2Ang|%<aa-bv
zKDw;%sNoXOhAOdS6;85#1)cQHvh8<UJka<ijERojHi$(TaHKd6&s2VveD@Y-{3E;-
z6>4)Ib+XTl6^-tl-bA@#9!aLF!_-WUD<3VooS0Xox>X3i6Ji42Zj_jdbb0Y>o~%v^
zBAStuib!0@st9UwQQRfH5Kw`yuSlw5>X7iW%cFb4+A+qtl9f?KZIu}BdXMTps~Id`
zj9oi78^U;=P^Cl)l;W`d!zwuOL2jAyhS-r;bKv~mQuUsdEAalVru$n&@ZbJ0_E$&v
ze9Lxe7Zho>a2gHoHx;Id^V5{R>d^T@gwEpzs3Vjub)Q6)1C(jbO2-Z;sJxdpa?zgp
zY4)@6T|wI*e3eC$PpW!Y^e%r35LsW1Sl+d*0`4NoWG^9L7q&~r3Yax)nHj%Uc&(os
zo{(-wc_H1APdD(BTdBHHKj9fnI*8Bty!NA>%z<lfqV7sCDX*gGnAeRME=2O|KKqLf
z4b-=n*8xKgh8_doD;teANuRP*sUBT|Vk$SjuUb%A<IEXS88cK#XwTakhf`%Yiqm>)
z=RSpxDHd<JSn{9e3-M$zq1#eJ!xhi`tPckf&i&FD(i1^z&oTnjg~?_K_HWx7Ho|*6
z828(>A5!Q*qwR4T*1Vkd^$Q{KHK;S@*Cb3r&Lt(X*sV>Q!JZHz(Hg!ZybFq+R~h)4
z9TSZ%fMRHT)+86IY{t%xb&N^b$FH@*xzPgxctayisp9IaaC;wqCb0pnX9Dybwp-x)
z%qg=B^cCpGemb-98Iu@2s|%({Pg!4hRi&gIZK=}_5Jl)bj7<R|eOyhAxuMihyvdm>
z`p=>uiC22fcK++}@t^Ss4{{`C38y#(J11Yad>O3Te~qXrNO_upvihN`a{A;7*4gNl
zU;ZbwK&89H*_)n)1%X>dUULIdW2$2ZLV%IM&8vK75COS6Ps=}Fkm+zOG*>jEi#q$8
z0sixD^M@x7+*!)mPbJP&@b(s(nVEs<=(&Mp^Yevj_r&4K!kmn20jE>t|3}IUNSi@_
G&HWE+c~2Ms

literal 0
HcmV?d00001

diff --git a/4-Classification/3-Classifiers-2/images/svm.png b/4-Classification/3-Classifiers-2/images/svm.png
new file mode 100644
index 0000000000000000000000000000000000000000..f4f0424619e635206c5239cc43b6d94b5a2f6dd7
GIT binary patch
literal 38653
zcmaHScRbZ!{Qn)eS@(90$hhieWTt_OYm_UiR5C*umCd!ux;K?;L}XP)ic;2<Y~@Ob
ziYTM3D_L3B-h9vP^Zh;^zrTLBKXUKmeb(!|=JWM@M_)A7<K__JfIuMJ2KodO2n2oz
z0)f5V#|D1cVpv6k|JWRi^azkW=D(-!UOt3C4nYhEr%i)0W(ET@Y|M7{7J45Iiq}fB
zm>hmf@yjhB=^Gq+$79EutFLz{X3UK&_DtUbx)O^$3Ntx(#z6V7HtAw+G{sOa87rT8
zNBqef^@+8Xi8&RcH$PAKPGp5R4Ry3Qian}*q&o53{pN&khiW#8^8f#sQe5pWP97pK
zKOg^`L0oijBz-@FIE-a}QtV|IlJql(e_ya;MaQTd44MGE<-hOAYX=I|p1LisS3MyA
zTSR{&gvWY^LDSRr=p+7{G1q36_Uq5{jRFN{!oNW#71q@E18r;Z7w{Zt=0`e2l`h|N
z4Qd5<7gCV>?<YM(Y4~F2w!bH>>c3gI25I+rAJntP8U6d-b}rD4UD#L~Pqn5_{`*D7
zrdsq&_CEI??Y{BnvHSm<CZfzaWTg>44>iky@iBi&T%@XgBW$r#5NBCpE-?q@K1=z2
zpz!(B`d-$k;lckTm0T|{)%Q0FVGeoMct`r*uB|E&q8BtD*Bf*2ykb72&nlZHZ+X}K
zRhRO=EhqAAXu4T&J|^v`EnGq1dw#EY!8AsZiY9l>DqORsgXt<X9r^cb2N<&Sl0uFA
zAr^P5XX*bop8N21Y5T}mvESF&YWx}H!}N#!nMiqEQX;zVQT#P%f93k_UHS2dXH9De
z*f4zjY=q%dX5+8D1N{N?jfu*j-4_yZM&h2ERsxBJ7Z>LJG0Lmor2Iszx_b^3o}I}c
zIQqTgC7wn!i*VWe`n?%Qi@eLydm2v4B$vQT0_hvR<j<s8(ioV-B>$%C-a}W(wQMWj
z37Uyy6L!iwuRcz8+uvqNi&u>go#;ci=e0vJ?nIr=8Et*G+D_UQt*UEU*@J{sEIk$`
zVLC_G$X@&hu0##q_9*>Zm`6J{%B|bg9()$p;=K4?!A}kw%?Le+<L)re(LtkI)llWM
zOiO642UU2v%Rp51PB@xjxLhS<-I+f`^SuB^x_F1NQFG~&1c&j=0t9h=!NyWqSCDyU
zAD{hkWK(3ZWR3F;7`Ie=bwrG24wZs$Oy0WSE^*z~8e&mV^<Fgpn2cIty8qXbI)Q$3
z9Ch_1TuVR=1GK<E%WLB4JoKJJ_L2AihIU{6n}X+oq-<Dt=kHCJU5kH~uPF7>a*!s`
z+&d@wqnFEImo6<EDzf2CZe;3E@>0dpM1UPGKxbDR5g2t}reCd$bqr6}>Kyg#GcO$P
z&4$EzE(~}<IAh|CbNMo|k#vnwohYUrO^wZ#&E2Qj2eiIwZx=H{t34zYLX17UcUa{*
z`}nF}H?7S>cTGc`v3Vo5-ESvHij@M91Cq{)U}%!z(B`i}h=_35v<RXeC1~=q@0<N1
zcd%epR(bTNCK{no<RHy~Ua)XS@`fww={5C^pU;w^h4|392-g?OA*~|;-Wx9#Xx(`k
z7i7#C*(((Ip{NUX3g}mOm*5iFy9H$4=Pd$N7chDcuzt!==sL1u(ABqaV6H)ackT%-
zswqn1hP&Y;1maUQg$W=J;F3xDF1f#*Y{6}`Exv2F=I<L{`jVzro0`sNa?g+5{A!@>
z8a(2tj8Nj8p!sJU9=o)m%GIIVS*KWFJ$Vd9SYe1dtwUt{&~^M+{x{8id-fht%QpoC
z4#t|^o2AMh_6XalH$ikV7VDg--8Fx^ZENCWRzgh@O%}=&uqF;SR<2Hr^wT6d$Gl22
zwRp<T@H`L+fH)1#T0hzt;TbuzIP{qPbo(ozRIu9JC88Qp*DqL@VHVZik#goR`zKYJ
zW>IK_AS313GqlUK3v@l2oZaEoX2!!aY;D0_ww)T!)6ezxVfmt46li0#VR8a_tu%XU
z=M0+!mZwaV^n_$lx>h4WTk4zm>S@3_GV+;L-WT(gj~-N6!Cv{Xu#ElF%Z)#WGu%qw
z)P4<T<C0T8ny9?oC}cl(``u}NcUGk`GCxN7Z!;<K_da9svs$orwjpS7!;sq=YZfGa
zpI{nAKO+FT%9hpq`kbv;*q#hMUL3}_7$tF9)lQOF82Lhcyx27J)$K9rc;w*o=|Ivv
z^d0H2w%1b>!)Qh;S6@05!WFYSfDX!|C;lDsuxleB8WK8GwUFte7bw`$Ds(@RN^M#z
zX4!?>aGc{tCsXvl%*DLF(q>=(P|BfxY4}+8A562|ebxe!9_ff^p+a}4gIXEPhmH76
zt@mp`?ET>HEj<`-4C;KYSpoIWkpn~47Jfut#-)0eM8EV_sfQ?%3R#crK~@kIX$dc0
z`qd(beU#`l%iZiBi)m`^=U$f9wxD;=ED_v!k1zE6b%wo>s1T_B`L*xrpPMRn=Q*C(
z#uia1xmW!Ci!q-fhhMi0!UH%(6Z8?wEj$mEa^Prm^d&JnMXD@OlE_mb>ULFHa)}5r
zk9%<-2x+95n&zNjbxn`=!LWmM5o8=TQL&iuoy6Zcdb-b?T;U$oZ56&RJz?iWX2J@u
z>g!1FXucaibE}s~PLRh!OINBNz7>w)hgM7b9l~Od*FvnvcBCpY_tN^DKKBKy`<f((
z$_B9uPR%fC#dD9ae`2S7I?#I!aiDy$F2Yd@|J8Wpp>3D-FGP@NK^dM;-Eu&1M^&Us
zv>(;iDbI#wU0RjA<LhZNX-W8{xUc`^wq$@du%V1?B#ujcE8rMI>utoz+V6Fux9U;-
zua<;}lt@<Zjw4M!`HYl9m;Ig8J4#@?UB5T>KBQLktmwLz|G6axbqR03X}L2t-Yf9k
z`p<_0qxef7aY<LzO$Y~vZfd;rtTCER-hCfnxEDH`8CF4&I+bX-!g`fv{xop%hinOB
z-uv#F&4^yb-gR{1k1F;p(Rw!<_^rOqMjO;n<CCFyL4C+<5y^(rok!zp+-$3x5ijwC
z+nS_TtORWO7WX<IQ6?ZDcZ1e}ux#GypWOTrFwQIe>&+3(P<Er#%%yHeot7hS;)nVM
zYvf7jzO9!O{c1ObhFUL?$YAwrix5SK?9)?U8X`sot<_43SaOlp9B!-U>5a&^4H(fX
zcG?RXQnZw@(U)1r?V@n3TH@k&vZ`*(sEvm9P>8>7+Nr5%58Xbr;C&?tKJ+V=uTBZE
z#q<de&T01<-!U7z{797;Q45aB9G2W;0qZY+!f9e^$LE>~<+IK2nzjBt*O53bR?%(w
z(=0J*QPXQe+wSQ1%@x=A4J>;;^o64eQ4(^2|5S{{!K-M4a#6c?C)SlLkC>zI=3nST
zLf5t8X4|Y-;)quake)gQ4N84H{InBl*Y4W!SMX1FSSFQ8F+>Z4&Gk+NZiL+tJaJ80
z-3cmeugghm=gSn+c+VPCTLv#)=ERiv?+C!c+<O1qQ+XRPWRUwK{C;<S1k3uF<?b#e
z%VU!0M7gblsUmpxx~LQ(v8bXe+aj`8O!s@`h0V7lWh9V~w?D3qDRjD4h)qQx<s33N
zI}b1WtSGqc^%rmX6LTVajHd$sslZn9ZUp7<*+d785BDP4H3B4a!_v0asy~D%EFCfA
zeL>N`Bx%2c8)^!8A9_d0X2<7v<gLlEk*P`A#tGDxfY|Ggc_;gyB~e+;!z(#zYb}1f
zd6HbjWA@tTJK0eq?gp}(OWJnLED?#r`9ZK)_A1%UnEO!}%f^{7s;z32>R_C4O1v`m
z+!?Kz!q%IXlH0S?5&b+D#+MI6`I|j*ll9jogP76X7w-q3Yc|IVk)LOXYNzoT7{T7K
zuQrzQ_8i>N|2F7kU^%#bV`ym~NvE3%o`3VPzdcKxWN+z4c`iD@t1RPkUWn(xkmX`2
zBEUHeMm^z7Fe&v>osL0zI`ru?7kFOhF<dX2?wpqKyQjkcK{9(L$^DL*RG;wlQ%yEY
z%qi?E=kW92{bugF^%q~>niEvRBtEOoue)K)&#Ie?58lFuz}*V+vu4$LHP7%II02m^
zEA;_;#|W~^aSJBw<JLR2dF)bIrOdTF4@+jy59DT~OfCe^J$yGKe)o*Tq4K`dOFLwN
z(s#FF;Z4{@_V(BI<eFz7F6KV2yn6Urg2FLxqQR%2Tu1QisFk&Yk77@vPXpWgiu0QI
zSI*US5{Ke<ZVf2g+0mVu$Kl7c!`tsGl=ub6AKWhk^_Iu1v20c>q*Pp$Fc~*LcZGqE
z--A`ZPzd+?C*MC0@u7%y1HTPndM&gu(U>EJG(eU(EV^+^>0+G8D-n!WH&rB#Q`DGK
z9p>bKn9hRk4*s4#M*B)rA!Wjwq_9{MA51$$AwgkpU&hqMocnT}e{bu{`ebd^jk?i@
zo+37l(O!3DxCi-hLhC||4dtv#ZiXtoYI0MR-rFa<^xA~&C`f1cP*U(-2bEWQETK)O
zHQi0Sta3gxVsg6{v))bM3fg*}ha}vHV&QOvF^;)6Jumw6Tq(p+f_?AnuQzv09=U%m
z=p8(bZdFEIgB3Rp_1=WEsJh{f9(>M1$W?kiY&lk>$Gu^rFy(5gL|d_3>nVy6ar3p~
zAo6~D_o)=f>^4L3KozSS4u{D1rAoS+njDolgcYX0EAu@xl+YOQjCe+xn+1a~jt(9e
zE%iNO-Ha~=d4xgmG=HGfH`Xmt)t=Kjefhb9hS~(>r?Rq6u=})|<(D_Z70A*djs*(Y
z@}2A1(arPL&lYqE1c$K^ZMls}j%|Ly2|Had%#u0G?q6ET3uzxSwB;_~NavWj`a3uP
zHrG6djRu;KS}E+a(s$su0QnL!_0E06-%r8O<cBUG7y*aslG_nT`2~@$NJx0bHyfa0
znd2Va;RmSbo}`iePTa5axQ;^6E;^doPY!4dl;2#WT}W|$JWHNWbDrVeIIudd0OT(-
zWIinh+k|?-8b!Wv3a5NDV5jH{n~N&V4<px3kd9n??P{sTR+9pTOnvx;d;?Z_MIX6L
zgVP}NFu9E&6?an}6(F@)Mtg<*NT_r>*&Ql!^|rUVg9u|v+t<ja@V2#am^MLt5YGj>
z=~yqU#c~ff?IdyIELbPxnNJ-ZITPkKf8GJJgL|qreq3E9DrJHnjb?9ad|?@pZ~-1m
zLL;xDgj14Zc=jg-oT@uD?=oY2F0Fg8CN3^`-QajaTKR@9A@@V@WMoS8FlJp)aF(8J
z|2)aW1cvR(;%rrKIR~kWT@ukhX4fIuBKuLH@30og-aiI!+2T0+xGj-!emyvgwsbIi
zJfqrK_y9BIOFuh!aC#4&^v|`3Lr@yzEm8+umK=@D;u1~~KMnSdC}h}TXlf;^#jk{d
z?9MfM@_)xQ85)TIyEu&hVRzrV9#RO6_hH!e<%>&$n^t=^H^_!ZL90SFi(M@>9EDz5
z3g1O)69_U<l_rdc4?+xOn4Wri&uT$#2DrAVL%lZf0$Nb96lHmoOvja>87HNZy>={5
zR%Kg4C47{OC&omoH=dO+xD3vDy889&DQ(O^kH#E^YR#piaNj--mn(q;WwqmwX5bL*
z*RF+=YL&E|J`V*P6nRD2We$&zfI7Rz<jHPEI0zm=x-N=iu?Fj`#52q5Hi$jp6N2ng
zrtnjV=N98{A-2@!o1p^yssqz~L%k37kV`aK=)}BSlo@IZYyQStxMz2-lFjJFRmd(^
zod{MNjkd~;Dj=PF{yA2}M)U-R-3SOw%G6@)g^JSHvF3%Jztp$fJBjPh0%HX|RT%<>
zF#V$uhy#d>`sbUFmoO=Kzdu8%1X>R-Mrg`h0-D{0L<vA92x%6$+Ek7g@|M_4s(aUf
z067BMKIZG_An0?J%kg^f?&=SS7HkNr@O-otK||8`X%XZ-gkP+N8XE2LFlydmeiTyA
z_Fh0y#DE>BZ07h)$e+=vm~hSt|D$&#>EqhIfgg=&zI~Z*FuHNydDo9K<fvg>-qGN)
zgVu$JE`Gsq;)}~;y3Ff;xnc7jLsLcCU#Y*Ui(FF}ks%Ej-$Fe_A+j3ULq6WHaR1zY
z_;~yyR`Yu*FRyK4lSEK2A@v`IBIh9O7Z(qIO(s$Ny4o~3F)i)q{R(U|g%3175OB2Z
zdH9yYnqU8UX<rZ3;bCphGTR*M*;G4sG#ZK*Vcljk()>#v%`B>kg)mjhYu|nP94oQC
zrIUY0p8i3wc`7cpv`o&WOrS@!(ujRxlH**{YiF?kIDA}V#hN*^jC91Ns;D{11dQfD
z`|GY{d4Gw{dpyuVLO2M%Jc40WFmNtzpCp6CDT{7X&i+LLYl6ndeeh)XK{1U<3EjZ3
zP65L*A>B(kzecv^UM{lw7Ev#F?+e_YrsW`5AYWo9>ioTmFZ^IDhG+QTy20(GXiBm?
z3Sr->XHhuWf=k<8b!goOq<<F|&}kAC!Q#+=kCn=TBXt$s!6Ro-a&n#PO@dq*XuF{j
zxHU_<1tHfxXkA26)Vd)YBRFU00?dgiJibpMe`in2jRidk<2pxc<8Pk2IYalq@&uWs
zlir+fOD=<dk;R%EsLLD-f%NUOz9Ba9W83Uk4)C6K_rI4W-B$N|GT;>@h`1W<zk`W<
z_efyWUoJ<LysU7y7j}&uw-?O}+1O0FhwOA1t(g`<vPX_b-8X!r4ep*S^!Gr$j^E8A
zdJ64k4{okGypxS*-HprSo!W=OM@xv-+I{}*60qMKqmul-ujj4LjWF%S*Y(G+3ky{4
zip5T`AYsAZ^mo)t2IhDA77v*yi+Db6nI|uel4Xq)b$N+y%jK%(_J3k6(-`6MS(W>P
z_>D|&LiglP%jNkHCn2O9TE@526v}mSR8i^Lm#<`0X-Y!ihqDq^&%o&fHLoT3YO$<N
zs|J7H{rp!`Zmer5r0?g0s*b2T6?&5peLXJV!6)ob+9^}{1D4#XjGV4X8yxp&;F10i
zb3|a9J4hhkHP_|bWY7s$=3*C^K1zJH&co`RSRYc>)h9nvfVoyKb1h`T(h8&<ZJ|13
z$h)|$S9a>-A#BnWj4~ASv0YLD-i%uh0oPT)ye{dqz^_qtdc;xs<EIVRFAEJtb?uL&
z=*t98vVZwlQBcexsTL&TZklvsGf&necT6LsqhT{}d2-0{=V1UdJx$O-XsbEwM;nNu
zlxP~XaGDH=Jx9Uw{y>vsls501^<`NuD!lZt%v=iq+pYuv<{75xov6chsbXsETU!om
zA0fuYh7U*c>3}=r!ne1R=`EwrQK4Oznv|{MbcC4Gb9<$jc)`!wh2<Lu5G_EEwmAw-
z^f8N6GH~4$y4coQvzc;|TC4NKzrJAJ$w<}W;4L{n2<=D?9E|1k%0@+?&b=`*_9yy+
zp}>tN`gDp;3x=O$uu(<U17l07X;}B+zZ;h94SjM&);0S&9zy;Yb@9K;&|oP<;v(SS
zFa!vr!ei+cNR+H6)Qrv&96&Pj#?y%IMBAdt_X&@`blGtV;Vm3p(WSXrsy19I)&0Hg
z2SFKXt2rIMC`Q~3AxS_d1keT#3rG*2wFszHNnTep$pSEq`54K|F=VaYU}6i0o4Gjn
zuCv#?i7nZyZID-Epy=kJwo=7(M(p8Fho?eMRZ{eumKaCqGn0!aoO-dPAjVkzKosbO
zOME+WZ{aL7q7nIv-@W_N((Y`A=*@uYy4Uvp?Mt-ffNcJHb#fXE*j|zh<~_n@zwPco
z!<tU>k+Jjal<QwpX`{4H=`FLIU+iYz_@1t-M0G06r;z-HRK@!f9N8;wna>r<JYm*Z
zE}2nY(6qB@y^iW!1VfN4TsP1HD(f5FT+%!glpr0tZ#rf!Q$2sJ=$%TmRAJ1*0piWu
zOPR>JC-hLD;G-j_-1VTeS(=4yx_qh0Y3BJ&?N?d?k1pU<=_uO!yLD9~dBGRpVlZ3t
zz-gW}ne!4=tn2o$rESRiW!Tbxsh4Mcv$^3*xD?KrLpa40CUGc<-wBEkKV>0657h28
zu3tQ}*p4iZw6Iu)*|OM}d{~J1@~emqi{)=^OthTZ6j=?K)<?Jjb4fV^4$yEYspMnn
z$pFcmrHoj^w$09S3WuW6s0>5~Wo44`*{9Ir9FMFV*!-CNDTbbVV8am_TESgQ@|B{Y
zwm4@vWSQgCd~;@4XN+ez3L;AyQu4GjDNr7pMA3$5@Z)j*UZ;ABj>k!0ldLdEwigZs
z+YwO7svxgb%3BbAb8U~&Wy#}n_Xm&9XL!$-Krn2?Ki$bxT#4OSVw3~xzQxQS4@{L9
zZogopQZ}feEivyYW}$_wS@3)6mO{!huLxilv+;+ru8Q4Nd;dV7I$u_8o!y)5a>F>J
zk?c`g<b9dkOKv{UVFj<_IM*QL;t`qR7LM=K*Cm*I>X%_~%>@4aOgri{_DPrLB@3a;
z^BjFci_^X5r$BXg^?$j`3uK@CE>))I2~DYQq*au98>+DRmt%(I2eckCgyeJlWhwR`
zqs=wORK{yU$jM%Is|2be`UT{MrYFGF1Z=Tp4PVkW7#eH&@`Q8OiX=H-&JFqmw;2It
z<MvOrdc0iHBdbqGCdDw9FH|vy%Dxrl3{G4#=l(om8gD@$@J!e%284SaVn9<vw^f`k
zVu1ns^`G*CBpQ1)!19aR&zI@#`j!W-{=~MEK>6_N_iSLVyPc68G=RQ+pL!`;5NzY-
zVN`ad=|y(>jfdm!Der!&r*+)b{0w3V{|6nd!aExk0G+*vICk&8;dOcD<1SO5YF<-%
z+aZi!_as2M%%HL}4lxeW9qrymMu3d}&qJWQkL8h~+PQRGU5B1+r)sjx&Y4q}0Tj<M
z;?cBw)h@8GnodAm0O9F+7`UXbLyc-NQRa-SpD8;6n)f6W2VE`3koS*?CnC#E)CH%E
z@Jl-lJ?0#WH1o8%D<Osg<8px;75|E7oS8gy^ybv9efM!HoHW%EX5Gr8jvA!?aXi}d
zbmgd%lQzKw6bQMtleH?bq={Ie!%UDg?#2l-Zx*n57(GX*3NDEJL*~ttID&nZ2LwIW
zLt7e#p5i~L-T3!~72O=Z<2EH$Q~HcTNoM!T9`$6A{3Y=cx9-@NPx!^Wm^lghBhu2|
z`$=@X?b{zCVyxQzpu87n8)^^y5B7F<YNhD^m4+tDRpGwXm60ZX3x3iu<^Fs>!4Z!}
zzY6MwNS#gS2xNSI&_MCK$Fsln_N`~(LhM)AGx+<R-+K#H;NLq|s&9IT^;dyh-5bBn
zvVPVtIGQu?xpD3xAO_iXXsRS`IFIjAc%Ut}f9&w#*Dq2E+FbYvNx2}J-1yjtA}u}p
zoLy6%pnC#@lIu<N<jaQ>QI}9Aw`&dBtEW~!$wH+go<I?>sI%8}NDZf?;8-jxgW=J$
z8(xE4&nAsf_|AcGt=HWpa+3`{T_ZP2zgaY!-Yt>c6u330oPTY5)JR*2cAS<;!_&j)
zm+S(TR}>kN6)`HDINm;S{akF40mc|nOi}P+eLtR5EiJ{P4pd309%2kLPqUlZqDmS|
z{QO(~o59)vorjsZ)f6AR9HT1k{lqS$bht@#|2f!Z<-&FrO`mQ-3!!;YQ$Rt)y~wyh
zzv9mCe}Z_sEK3ag2t@cdlc>Jc)Jg}T5_Hu*LC#Q-@9ERNK?&JV`eMF_*QzM8?)Gf`
zoHcFQtyKN-G3fA_{xT*G@t5?3+TnhZJZfWLKBb$)3}Al`+Bwlf?Yzm3L4`Fvb2L_Q
z#=Wc~YKY&04fiA63^lTo3H(HUMSelS#-KrRHtaj;>hU-!?2QkNg7gD3VNOV5t}Li?
z6A8?FccURk!#HXqw;d4IDp}lymaXhW7(NEi+?rxA@5GwTt!@R(mxnr=EE&4jI7GLo
zN(hFCFa#B!vN*=-HwVdHJ<b|394sO@J+qdjKW-L<)D~wmB4z7cizFR`s6(ya2DP7{
zE761LUU#m%$LuuANUlXdT1JBBTcB%f(K;N7a8EYLRARgaH9I=8G}{XON*jZunNPZv
z{ysV?Qn*hzML!hSY;~|$txo0@O}%$ZB6Gyig;4I7GLi=ubm&IxU&>U7!iwY|nSs8a
zNg3hogg3PQzo8lx8cPy($z{SG*c&?5R02<rM6W+v03+<;n)=qPs2XPK!8`qmpS?JG
zAIk)9A|HdwY5uxag(0K(qv1C2J|8cDyyk;u*7DF9`~?`jn_uHd_+3%h`K}c0ggX;_
ziQ4!X$_m%ts1<IZ@xqb^`adthuspYYS0F<LvbWf{;H#X9^V;+R=s17konYA)Cxm6s
zENG^N)(I=iBz7^M)gIakfe>e|o(!hHghd_kP(MN+c$u?%2z%qJ^+NJb%aP=2+J3|y
z5(vOW8j5DOGaKAiq3*oHrHKjSM1E4wQ`u5-+z~Up#Bj9qVK3QCUlOyvANg0`>rXL<
zG+#gv^JQ<va5uR!qE5*$wy`EK%q9c{>37`8Ijg?z1l&Xt$l}G5Z9LUTQIRU|Lq_lG
zL>(zIZ2a5!UiaNEUF)hvnMT>91DIR(8W92TRL8u4Y<93CACQHzbTK~Ey`=8Ew>4KX
zPv8<wc<UPV4f-HnkRUEO^t%7+2Tx)l3(}WCV6Fn3mv}N~Ywfu4Br~s^I(5ho(ueUz
zOT#g)q+0S0S&4S39r-lYl+XW+0Q;v9i-aYq-#dcE&pBU~N&1~zWFOQ)kizUX<nIRP
zdR_T`*Cx#p&Xd@_^o_L&E;~EOA~I@s<^7|tSO1R-0M`6)5lH!>j>pPQxomGC<W{eH
z*BU8BE)>U{jyneqF}EKJ;=J5FCWs&W&8bIF`)qyWdq0U?8-x4V@B^s8o#&u7K72nU
zg0*MkK-y)CWxgwXY7jg7NeCgAIHc*egT!GsdwwE4IRumRi6GmA`408M={p};UKBtz
z;WMIPv(YOx4|?7mP=j=#BPo1yLi9V63o4@9GS?NcE&zNIe=GABit?PsJVeY1Elt{%
zp1suz1vv!f8e~@}Y=^t{^B#sFMHf$g@OC-v8}y>Pi?LYBNRi3GtBJY6>#B^E$wet-
zF~`9K{VM?FV0UjC;nakz$!YU$R8`n_Lxw^5{<ribi~5_9h^0u9n)?%w_8udDb%>Gp
zjCua8ad;tEoT(HL)C7FuEWiWSBMy(E4R}F8DNPUi!9TcxYg@p?`G4P$zMDcUYMASL
z2MgtXraqZ6rSa^f8aWh#=VNLgDQXX@5Tg*m9eT_ThkXTHlsM~aYVjmpvv;RG{BAKG
zN0;hD^vpW!D^Ow7)8Zz*^J+Bd11hllAh}L7USycj-%KtB(jF;n8@|?m4US?!9`pCT
z1{q@UId$oij6QPfX((o4uNHH^2)2Abc%U#nrie{q*75B%8}ZBdjN5}%-j*E+-%~;<
zg-o-(f_niEn&2?dE`}M$t318X$C=K)d~`&_rJE(gcu@@74nH7{e+jDI#!J9ogSI!Z
zu51m3wu%C1PBKH&LQDgH=LsTUJaV^}TxPhxt7a8sU}nD`J!l~d^JL7kyAM}Uu9Kp?
zX<scH8@5P|qJ>PGxCKnkAUY*>T$4?X&9ydgtx;|LjRz=01-frF!><_=-=9)N$lBVt
zia}bQmVuZ+x96)cE8{d*ohKp+0Qi*T33RCIw_gu^Kn_wj@2omhQ2N@;U{m^=?1<`j
zJJtZqcpgvtaWTQ@lne2(MV4dbC|hw)4Dw@ALNTq3hI5+d$9A!S%X9e+DoIT9iAS#k
zvfgPazF0n8<aDUjeZ%ni1`SHf;ECH&W3U$9c|*y~i0ZHpORa%-z#B#7xWOTmnZi#L
z8a~B2;QBPeqni}gAG|qb@bO3UK+5KBsh}Xz<_tchPJ<}BYCnEou$ecJXNfAsfKbuq
z5?;BH2_QXB5>45Jo)J$u&?|rScj)eM$fAd`L?J8ENIbvCBf5xY`uc34VRv!gYHTQZ
z6d&&~f;%v%%HPU@Qira=g&oJfN2#X^059ndykyb|&56IzjORtZhY%2rhtbZFXAxOi
z+*jf8ECJAPWZfXS_e+Zq*Ia^xScHO=AC^ZJ)ek8n@AlQNMT-9OwKK1fq9jUaQegM@
z46nLhApw5_^y+6VM9b+b=2<;MtA*k-0%|0_VE%RU<#K+iaL@hsZp&-u`!c)IsJXg`
z54tCs8+-aTN&qrIwpSRsjmGjbE8ez3{B~reo7CE#RQ#Z?S@nWt_bdO|bU&LSeH|{B
zW#r~Xtcf|^J*FtsomK8*h3$^4yD9c(9&jkqe=?pRNyBv22jo`WveWQyA^Ha_4-RFR
z+Y&fNa*hSiL)B-VZYUy-fpk^8e|hAz_<f)AJ^X{E5heX>D{f;42Oo0ho5lNmJd8Er
z`G*d07|Q{4K(<;Is|#FS(ivGv`EA(%YN_k_wB{aUNQogXm)q)~8U&qa!Z4q7l%yh7
zuLh<^N&%HK!2m~|o7fmKa1PRY%VN2SUGK-^JWJIp=)^a_bJ(g7vK6Il>q%uP`_a*4
zrggNbf`*>Rkt%QESG9;8*F|*EMEN%;h4S&Mj*8p{j-|HMCpmJ;{yC5IF`&#QuDWYO
zD5w}*qjt?k^Ll3et~KdF&x#H3poHIliqs}OX!{BqCR>qnpd{4aU|zU6=rr6-3Hw;P
z^4k7u!!4sbr__Oq!j>+qKAH@qiP!!SMtU-1VlV*Ck{pH{yQy_gNC|DP9<$y#57C6p
zapUU!GIj?3nBB9uu2Z+8UX8(NZuY2%+6XI(C%$<BTvJnxX<$*y>LJu$&Aj2O)7vZa
z@H;<dkfu;0LR*U&&7gBsOtnf4c<3=>aM3Yg_^EeyX_p~?e5TDCYdC`>rqW5>v@sCa
z(b1VvoaD0SpFd~X#_F>(&0l@~FLZko;Ao_ihZc&I(x6@vz|rW>;6qXC_I|<gj2)1#
znsPB6Y?{=t&8~OC<9-#<y7p^Y_hzk)+q6nq2n6x(_)y-eQ#Ey%^%|qyd-3|Z%%l1g
z{5?>2_J?nynbLiaF=;1whxq=sGj!t9!{llL!LYgU745?F&;8^QnT`Dh@!+cDD35v6
z^xSMaEnuVMUL(N<AG#v_K2JMP8y`$OTljsm_m7bo^bcuD0<VsOyLxkI9(YY$&}gm7
zP*sZ47x*e7ORix?45CA7F;c9c=sVO67FM(8!)`+L)GR?uFtLs4=;1=?dwhywlW%~w
zp@oBf!RUwR*xy@sZ>dn#8Sdvs4HgSg`Y9D@`Ztyoh0<XF`DBeJns;9y@G$R!^RKf{
z{38gi($iu&Rfx<<dPmxdNnf~OhwVWJg`-C)6qbn*J3E<g{Og>GLk%%cPe+?70_hzb
zc@CN#dMBg)24wAG6*<=Bh};Zg#EbQP>T#-=KK^ub=Rz#jWn^TjZ_7%ap<;I@G7l)_
zEBDzj3LqOfI}_J>dp1*Ol1j0gqSPc-Up2;?iDxCgBwLaOX_oBSIXqFCg*oOB$SOP<
zZ%q*A^D1aSL4@sPYma0Ox}y_Gz-+EJE%>IL%-hqU38n>(lH<h>$-m<Z0|2cCO5M@=
zQV4txdQaU$n%{sG;DN3CqY^6Y%dapiV+bm{*<5oI?giHosRN!4?ZOm=k<hD_$2UB_
zFBOxs*+t_i`oDq%<BIls#iyXf#?=Cl#snS!Ns8;3i@c+U;Uid`-P_#w32GnO7|O4>
zN|LouPoQOFdGg&p@>0az`-X%bX3FJQ*I4sxG=)T=5b(=Q6}BFna%@O{#u<<v<tl+o
zKk=f?3Xn{)r^B7<zkWD-1aw5A&XW-24#;QHapW!rZ0_fNro79xDZzUzARY%b-;38H
zh>L?RPDtsAQCTuOG7A_l<*hC-7UgdEdgv3|PF+IV<I3^d-$`L)Gt0#VB6B$n=5n7V
ze9Yr}y^pG(X3#hhX5+{${}xJ-&+XZ*w`C^*-s!`9%6aT|b%H~boLg<Lk2ABsn)IhA
zadDPLJcP>lL!SewnZ6#dw$^9gt>AN-wHH;dx;*?ZEs{nJY1R(x6S<j|@t2;;e9#F5
zv{sq4O5a0@sja4Wk$QyOA9WHxR3I&9ht*md=Hd6mv8+tDc}Z#GMMy&t8{3Z>=++(P
z8KD+%Q2ZVvGGamTt8dS|@*lrznv4$Z;UMh=fCZJTgc5){>81gO)`3JWrj_2fbomT|
z_YB~rnE=ENl_6%k`=tSOruhiAiRs+zhs-X(8CjV%gPmSm-xio|v$2wF^z8F6Y?xgz
zp7N0ih~Fz>h8A)J?9c0OfCwoq`@hvG{QcjHx+;l<06t9eM!wKX02|W($zT?mg^S&Q
zCgr4|MGzllZZl)eJYHbkff+D3h2nA(SD>xa#iiGyqSWY;Grn!-=$44{z#I~*n8K;L
zcSJywvyRiW=#{4kKOplZIs^gO>Odh31C95issZSRM;7RS)dT~h!V7?m;z=edcF(TA
zs0J3n)~x0XiZsec07>c}{8Th}Pbo<l(aXYpmXHg<AR#v)<uqTw8^DA3XTkw)w~ZE)
zC9JhClxumW({ZDLN&lPTLxsRsv;*l9cDa!f%#E1<X_+9;)|YYw?LRcn#FO4MHY(7<
zw7ozrxgWs^(hO`K6E!~<G;9-}#=f4vm<W7za34BIa1v3+%H^xbgq%#6>eNO^`&+zX
z`Kaw{?S-jIi^(8yCAovl9NItt6-h>rptLQZqr<?p=wpHKlA{x(-hn;=2PEw~5Wj&u
zDw6b*mf9E7auMMLVlp8Km|<>V)h$Ru+d_=KTCgC+qxRPUOTbwPSk)oO+yTD#MAx6k
zk2}aSR=!&=B6XUv*f$EOFwz!Tl-d^ND6)M7$Rg<!5b2HUD`?JZ5;YIgr@Kx`5^-4U
zUW?8#IunGdZ#nu8gd&^)%4y=6QE9dxIFM+4nV?DN7uYJ)92%t>gas4vFk87NM#Gj=
z_uq35U27JO0=g^nB+8RENZSP{eR3Im5^IvfJlUg^@ZE=&yd??+k88M4q6sWcH-es+
z-m^%W_=hyzOJUdg)u6$Y|2O92Pt+cZd|{!5q>S<>9>SYKE8)DeNtZ>=fjp6t%ghE&
z{PDh|R5;)~eo<6e2>t5Rkmj`khzL4t+8q(GPau(!8?BkHulo9fA%2c!*<)enE=|@l
zCI1D)TjE7d!{9aRX^Y$Qm&@{pp|VzrQhw*KCYOq#?|C#dI~tC`fzYHud}-6Qu)`BY
zh>Fxp!X36SJZymL2i(n2#CkD4yLRBh+#c08E?{E-2MmWI{-LphT%wnjhN}|4j>jfx
z{<<w;Wx<-kQHMO&ko2G;2}K&BvCxg(mHzy3&Ulmb*-A-{!TaGXU}S%245n3gj-|90
z>k>3UNAmj5EZP#yi(JGhGJ3gBb8Lb-ZaL<7K3pw;;rk@(vdW-@9}a6W^#77R(20CZ
zVBcSkAx?EWH#&>Nvx@M34{9jF-D=pfbkKe7WpeopTKXW$36iFl7Qxc)m%m&_c?)uk
z>zyj4OTtRW6Rcx*e&eOay(1SaE&N^<Po{j?B>h>jzi8OluiWESnq961z2^M@K(v(H
zM*xa@%uCMh+Rl*)N2+TFhekJFFf}u4Z(C5)9Cq~CI}eXKg1yn)m`0kWuC*Vec^7{v
z;EK^-09Z=uOyjR>etCbWz7ip+L0L)Gnl3|HfLoXm814?dZEk3gAm_ks{)Ke-k7}Pn
znWO)8EH2#@b)8B27CK0pmQMI<c@lzG_16_hv<woAHshIr;gNCv-}Zxe(aHhPGya#*
z5OMs(5Y2$S<DhzWEMPN4UDC=A>2oHCPvb=>5@NRGdf0#X4&{0@6IU#zt--(Uz>}=~
zrK3|g$2B=ZS$+g~S%xtKfOsuF^ne@vmuL`LaICPpfTAD8gs#%}xgT_%YtXK!RLE=>
zqH98Yi~)No_Yg)C{^i5tckekwe%qY}u}3@>occhhG-KPJJaH^*ZDkcPT6J-6PBxC;
zBOf>d>5QmMQWPoN(mob6CYizWjiDIm)4e9?z*-dJ;`Xdx<+w<iXWrDFZHKS2izZNj
zE4=_kLExmd(m$hv%s7+EguZe`v)OJT;&V`|oa@8n(W8CLaN+urD@bJf;>0$taJrVy
zN!U)-?&MPeDo;A9j=K(8h!l9Nk7v?3;MNc^n2yc8VNP9fQLy1(v-{OZpODvg(u&u<
zdvmd>K-F!IWsZ^}OXbOqRMd6h#pA0v9x{sDX8V%37OB!b!=vA6Sdi)Tg2Wy>cIbP*
zVJ>zVC8{kh@6Gqc!fY5$Ur-Nx@++eD+rsrP;&d@uBW>z!W+<tMat`JRAA;>7^5yMA
zv89U@iy>N6jns<?0gMjj*{XpHa<Sl5Bk7NCIsGeq$N~_b;OyD9tF1qUYT=<$R5!BY
zGC~1y6>7?6Wzt~{)A(OD2;&?x_A{GzZ;gzc2FG&Xlt|xm37Df$m*LZZ9Z*q#bHAUW
zvuk{AH2Pz^&V`wY<)yyr7ynU1D9OS=mjYZ~9nVh1#r$5{09_Kqa*f;4`_qvImFPZE
zdexrRYgjQvAwGCuRSBd)Z@>1(EDh2GsIP<dRD%?mF(;lWN&vx-^soFrYS@^_naw%`
zf8Fz^H;zN}g5hkeN43n@qx)izVVRNj<`9cT)m*utN!T`fzhvJ&k=a7hZIHe*r`X|?
z^$_Yp&>~<h`%zJA?O$1l0P5S^RBqAqThX^G27DIwFGsp0z4(uNA)4XAbUr78D!^}I
z9(~DjEaw3x{`XN%)$Ec4@W+rxEi1QVSU6e9+UPcXSD^1`&IX}pbk>)~O&jT~Z1*8v
zrad6!!oXm;eLJ5S<Fus~*SjUb`_xT`%UWXc5D28yd1RB}grv)UV)zH&n77NEOl+yT
zJ;?!`m;_Q0D@Ea;livVg9>w~&_7vbN)3$y>f=hhr<Zbpl50zYE3%e;Eg12GKIB2`u
z*U$}cpTsr*1-^NTE)e1SFs6O^X|++adan()8x|vKdQXE4hkYG{<D|b0nB>$+3;_B_
zn)<r^C*UsLfK3)wVb()F_-Y~s(is`i`Kx9psUKaJ<^osxPx|4d8F{p&-jvXLuNPHY
z#G4ivwsgJDv2tMV_l{us^??J_<y}4O;c{k49vc6Q%R&p1+8u@y>-Y}&<NNv^Ya^_q
zUMjEe-pY|8&sM46{>4BhcZA_@aaLEEATA8<CUu(=cTkqs%J4ibx^E+4aV#IfP(yU_
z_sJ8vkbUYzE|YP525g&SE4UVp(`k*O+yU6zo0~F3p*-tH&Md}FHPEzj$+SvI$9N4l
zulceYbmEy`Y}1kZe<5?=4;IHfbk|fY7cZei=G*-4S`fMb#&o@Np6c_;+pFcuZKZp%
zpMAxquY`U4VBuYd#97$dwiX<6d8M6_ip%{HBp8*#Tf!`waLl3!{pvdCiWIznXT!6i
zUFu$vI$$Rqg^mf=Wg{Py%oI-o8aBlbB}z&bLp`Uaj4IHcgWO4tsmD{ygT6zCLe&Kp
zuaF(<ZfCBoW5>eBpY#8NNPKnH9kUH*KMgXS#^N`qe((Zue`=^7YZhy>x+NN{j0Uz8
z7EE8GwOdL}KOwFC3jP{F)`IAeZJ)K|Y4&9q%03y_g33XL%PXwADc=@Y-T;g?@$JXM
zH)#DF3x4i5X1!>{G#mcY=tSkb5qh4!$f9M-;&}1c6R0{g%o^`>xZ?N1={GrfmZIl9
zIo59BFcU4%%V(naXc5%k@SI^3r)|@8x}=%<Yd*{>3<FVyU1d9~whnj)SghOK-Mg|T
zGGb2&&voj2zc1_tg?e)N=HJ|RKgZFq&y`+6dk=MVzMX%}C?lYG!F=YqVN9?!_j^>I
zpiwb^i^Rb35-%_laM4IoCm=MW(q3qz+%K%z&_!o1?!N0&E<?I}+S+mxm2FsiW%X@7
zR@RE4KR|Y(>D)Q5ouw0)pQk@$XsAUpGYtWBl3&9~GT0;>NYJLtSwq=z@V;#B?EcMj
z%M;O-j-@Zj(<JMhO#b^d8=tq6cYQ_Td=ylWLAmSHlM@NFVCY1>Jbl1~4e!qV1*+U)
z9ucKlD}5N7q^5R9A?Ydty7w`O0-s#3w}>iA+J*wsd(RN3&k$g%W`EmI{QRMSHcB49
z3JQ4X+m$E1%E(33GS4F-+f4WfE6l7D!(2&j)Y?bt0;)Z$eW`KN&IKKW%i%)wh0vPm
zKblSLkIb}?r{>F@6W;TEvd+1`MV=)!(Q=SinMt~SKgjEHjjqiYZ(H>lp0zQC?{aZD
z&mOa2)|4|MFf&s#*UKLe<XWDrr7*e_dxtdW8A<ka%ofv5*u_QGgXSh77j#7lUC-4O
zxu-u}<p_b5QM)px7l!(k<@E9-4?$j@aqxdGM0V7Ft)f%j#z^G_h2kR`hGuB@Eb=8Y
zZSOkH9N%E?q_KKz0UrW==rT4O&hL3^nV5}vxaD=_H?GSSO+3}=C$Xo=a5iuXQtdH2
zWx(#$e`>y={Z`S3q=cwn;ovB@m`G)mFYvlwm2R8MXv&v7*;~tZBNVeYi}M)?X>f}m
zPbCxMhaD03%%U%<HT}b<?l|**{dsUdI`In#X*@$cG+p+NU`&8^nC(eOhQDsf^)>i&
z${sQ32t$;99lT#wLPaHGXBShcL-z&=pWsr}e>&u&DSXUYCrB%&h{G(ele@}&&})3@
zYd>{>_!l7|difWgW5zf+xrwe+_;&Ha^XW8l0qGi~{P@?yH`@B=r-$JWXOb?94)8E_
z_5(NsyW9RAXT>}0v<mt9QrCq#J!<?w`Jk}5m=HEUc>^Md5IH-NcYXps+?4c4s&}Kd
zmt;h`27AXLa;+_+@}xsAyhV3dt=H-V7Bn(}Xg{bSCwZLbM2Oo`?|s$Wh!xrRd7C`1
zM%O8f99b~NBh$5{wLjEO#(d0Gk;ocgS*7F!pKyG4ZS&*)g5>OfJQpp$)iN-f6>~O@
zox>3`HN(ydEz@>*fxjBhSQC_@wq*jsUx@YO><ZHT=qz^PsYEW8cot8Dxu4w(iac-M
zU$)sIQt4uHa7b?Pe~B*w#rA*;?`}%g70;NXO~6~I>H)oBmzGp&imv+3|1Gu^CPT|>
zk=vng-)!Uo+Zy?gkJX!RU>N5CSBI4rqMJ~4FkV780ZR@tGCW36h!`*QE2HR-PtCp1
z$El+|x6H~!dUQAT<Te(Qnu0J0;zp>1iUill5xN}xjNKDzj=nzTJ*$n7?e2ksEJ?s+
z1BP&?%zqMsjqKfuU-}MR43aq1Y8~%Uy`ez0rCS1;>Px7_*|bXNHGgN$OiD5ec!Lpl
zK9EMt``3^7ntg|gc&2$?-kR1SE}BDzA?@~={4&yBL+yM!LIt5pthgbo4irF9K};p|
zlr1u<_ZG4Qpg!v>$k#?#bI1NtWASaP!v>_v53DMV^-tzm<Nd$!F;!xn*|~GF0y{IZ
z-n!Xtf1Z6wrbYN7Z@>Z2m%U3X)n^YD<laW-GploA>YW%$vM6(FbqB%L3cQ2T%2+;n
z+m+NR&p&Y{bS1UO9KO$Tk2PM`kk;5>Ih<Z>d851(DRi3+a1qZi?>Cs+nM))Ran*D7
zn9$XXEx8`{I@j~zU-Jpill+yW4%b3>IfZfzqtd^BY7`(J)Rt$6q=w2dOOXG%l1>L1
z?sUPcw3>jD`7?_c6*w&z5FMt7O6KJ=6r{!$+(tSw@pn+4u#x61I~CYj)gRRW9Lfg8
zBlTu(@NULul|Wsw*6THryh(3IUF~=amujvjM`e<&Ect7CNjaW6DsTd!EspuTr-V(4
zaDwQq!KeHij}hcC@UGW5n@EPfX&1T|?nWxPizKLZG$Lt@+7WblyMOEy9<;#&=FlNF
zPt>V34VtWn=5s(CW}skyek3b;cs9<ZXN#)spTc=rPu_&XOHZ0Niu~iMVA(+}$1|i$
zdwplZoh7_4QR>#j;{yBpEz{>tx1ECyK@_+|qWnlvr$Ka<ph5)aX6myk`nmwM;2D~x
zNw9Zp_|8CX)xP$vKB&!`C0{%IGa{kuo&;-p^veAzNb;c<DV6*QQQ%!JrfYUvmByO5
zFg<f)Q(tDM`1V0u{qqY~nXF>^K~i0>a$)i<mY{uAaoq1pR9<m;9R)qyHAg2cO*YWO
z)dTr8;R6#~LY}83(uR%j9R)P|KInM-5}}*=gI}X53$lBs$kPn|O-~k^6Fhmh$YeHN
zNg!jAJ_T}n${i+5sF4a#UMZtfyVpoNrSB#(yOKLD#A4>mLPE0T9RW{%S%t1f7tpSx
zm73zcb~J86w;U$wB-Q%w0WIoeZv3+zpg*Lhw`9WNAP4vte|PK2hqAUzlYG0oudo_a
zT99PP+1|a?<vEeBOow=AwWYcBXk1$Z9AFJ-0|YbBE*&T0w<A{P@S3mxZjY`{oP&wm
zUxRVX)%INR=&rFay|~{7-RnASQa;!7YtyahN6l}~?jF~2SAK~qZNA<E`~@C^R|1P-
zo;{Ph?QTH2-ael&(R^_KnN)S(xi6u^B^JQRDEf9hcet4q(jE+OsTYBaD<?^vogEgU
zHA#Wx5SJhP?;z*k&5PUe7E#DtzF=!u;4wuOeYjSD`RWw{fe8^_=Ov#fU3Lj6&j=)y
zQDfAVyjezfdTl!qCrhub;fkeIl`_jK4)Ur|Oh$9Kre}RgnLc&a=Nm78#>-_4=)bsh
z^3a<nole@J>_Y5(3twtmJVm-WuPQUJ70ugUtKcqn?Y=3d9neiP1@DmF#yeFl57IML
z{Uy0O7y8TJMK+tj>QYX>=NIJH@V^(?&qOY<!OCp4TYpObDgxSi3h(BimDS~0S*hjP
zlYVkt9KBl#aVuG+LuGUEN|s$Rs=r5Wp)@x)%>F^DhDM-NUg^4I;g|g@P5SZa^L5o$
zH-b>_7qE1g-HzP?-*K;Zp~kVs$Gii;s|~VDKX#=EgMkat)?TF#dQE&=&==<V^#X$v
z1vDBQ@sw^;<DR`1#6s<BIEC^SWc8G{Lz6#a>--0stCL^b^JhuWI!XzD`pJ&c97DTd
z${hg%Tp7ZO00h1bvT?#M@w+LbO&f4n6n3_Z?+L3ao2>Jp**_!V^u@_#Px{ODRnZuV
zJ||E%{U<@gr_hWb<(W{g;jiFDxY@2dgVn<VebPiZ*4D;Lmi}spxJyAJ+yBP}K*W3Z
zEa?yk=Ady$d26sUc}xp_6?)VtZxtmH|2DhjH7k;Jl|$Ijet~$DAuw}t8BLFBPh!&C
zS%GGD-NcB=H=fzj=CJnVc_Z=k`Wt?H9Axf;5YiZm^sXp8599(^T$Hy~ZMtJY_EE*7
z1-<)(;@%oxjt${0W@UpQCZr$V{$sCLG4PrfaB@bTf){hJ*t>fF#ti2FEcv~@PhTld
z5Tk~%x{I+2(>9;K0C3Jp5#7t*@G~5EE>qYs=n>H$Ok5&Y5k!Z?)df6^UN?Nq?<mg`
zYw`_RYYyCi2h|TN*XCzPDt>9$W*Ro10(0{UE|1gft8;!J3ps@Ee{>dMNE2NrwUcQ5
zyXku5ESfqEgB)jqnz>Zw!~s?(0(<!AvQOW*wsvU%(U!hu)r2VQ69v&z8NSQ541WRD
z5&6Qz`}IeFhs4hMVn36lAYE5&J7d2+x9r_o$~hlm5OX_fg)BwhrE$q^te;|@wh3Tr
zl($a;K_7Eu%`q-Kn=i+S)Iy3qn$y9Ff|verl(&;q?E4w_06^kTuWMc;h=*r>=YB{&
zYo|Sy3a<cfslI(WcA`!`7@ZEWVNGrkxOgJi5<Wd&dRS3F@Gq184Gt{8Jg_I{*<<Ok
zPBNq=7qdpg$E>CDnYZ|>4hqJu(2g1c$C&_s0^uYBmJE13hoZj#94A)v;hW+~mjs>%
zzgLk&4P{=FgSgbb;7Q~;whGt*z<Al?XU65<0B$CzeXM;G4QVgEk+#q+d^cXt=kF1^
zAzf{TGvpHE8-M@wV%SqC?puQvGto7vV=8OZDG~gVPC^9lp|NdHROz>$Z;^4djbCX#
z>Wq;Api^eox8=hk^X~8(3gr)Yodi8aHTAuGX8tW#!Kp7S;jA_>E1!<)53QCBaL%xv
z3gc+;*cSqe{BRcJFa!VQ-!N_C+22yEF+cZHorp^kTOuuyjHqwQS(~hNOS}ODEQIs4
zt%BpY*X60FX$%4(=JCH(HhZ0Rh1X$6uXR;_XM9`uWSCrD%jIP>4tH{h{`5<=D%?0|
zYo?YThr9Kc%c&L7$F}^Pd<`ay*mr@DJIPd*cwwDijg63Y00Ul#DGr{pyNcMB+muem
zvVfG7(Bh<CWq((z1a8dQ{^7MlG<a=-xm`JvA%IYJC^mmr^}{Dw7?FKR*^ptRh*2!w
z2A;u_dG*gs=#5l|&!ypUW-UDE#B`7#CtCivv>;CH(@9WR-a8HO`+1?#JWD-zKO+7{
zUY*&z+DSFh>6to-^L=0157XMK<IWRulbPd!;&$I+wl70oC0i&JcKu+<;N(ICaEN^N
zscZ7L_uE9cC+NB5lC$Ak5FMZWk16`COl>G#80!8-?nh;(?I4oy7(w{V?UYLH_QqD)
zjn1iT@?{Z^U&-^OKa;$mh>s#!|2Ihb*oAJG3uD@^=+&+<*kXPuE!FZ-za3r#HLwp?
zIMEXAaZdA%EoWhZd@DFs0H{^5rB;65zF5)fA&<YEkQ=%3REv0#(x4&2wv9H@#IVH&
zRGce3QVhKX*ZNEM2QA)IaBb0PPCO*vzcE=y&#O+syM!ts@2Fp=f=BJeiBcEY@N6P8
zc2?)o_ILw|E}O5NL8CV$flcrfNIsU-jT$bhtM9mSI#Sv{mbVNwp=`%aG|^;RQshd9
z_`$N8k`nTcqkL5qqk6n<3ZN62su<nf)#*D7a6w$JY4M9YT{D{W%RYEFx;<$1Pjc1m
z12217DhkZOueK-;#!~cUnOd=~T5vBV;3_~!LIl?#Rfhxw)~KQvIZdG@Fdb33PfeCq
z7-$xo!rhNwaM2WTrYc5fuDO60NkbGt5CsPE49~VLy6&GLutMc#uzZ6D3HP-${0~!a
z9T!y>c7e_?z(`B?C<r3mA+2;P-Hn6@NH;U$OShDyk`lrI5>gHlBOs~5C@>6?(ji^<
zjNkX&d;jIf>~nUUXRp1UwVrMEJZ0iVC9r+?{{6_u*=TvD#>?68bl0!KabvN=SV)J0
zf<2XM)L4M}V;qxWth^^BpeqS1ju{F6y1o`Lvv$As>rS(BND-QWnirqz=87zJ|9q+e
zE-<@qYf+U0e~zhr(D4Zwh5-fBydzd54h(9U=6Y#{6xxsm*%0&tdC6e>ucS-r)?m^v
zc-!x=S%)C;;fS`4Ugx1~^Wg|=il_nvNFL!Floe<ehnX%4YYJ;BW8xGwRjzdW!q@*E
z1|!Ob?-+@;hE|U6WtI)Sz|dU^?S|pB-;^|a?<0w&4jL$Txmz~ZqwdEex3CbQfvb-n
zpxUa=ME{<xK4!b5U;t_RQ0<_Vs7N5xn~~;K6V?#hjfpu(Zb)8HDah*8%lLy)Ki@zG
z15X9ARI0vrm#x2Nj<esCqTvM<Ll_<7HOEZ4zNmQPcyTiTS8V6X9rS9A49A9}@k~vK
zi^FfkzF*EO;^9!Kb@{hDPu?!3_vQaL!eUME*J_WdYa~UUz>`7(1eFa|tfy|7v&J!+
zJ4)X30d)59|9)E$`8TI{VQbs}wZYacSs3pPI|n<AFo;-6>CdyMFDj@Mlo;-jt2oDh
zIDVY}X0dxTssTvx14?C8-yt=di9NSlaUZ5Q1xQZ0ku9wmEmYtzzy<0+d6y`uUV#Bp
z33hc10}|-}K83;e#=pAn$p&tQ8eD@BrMx!D@dCXw8M3$nq1~pCY|l-KGQ=XA{7tD3
zfmScJNbW;0eE9e5!3TzdYB{&j8W}~fp<dwy5d1PTKi})^(1x)vs?1|WzU$u~=_!1>
zxc;WxoWPfS_wwIM;sKEF@WNh;wYtqjc{8AC%NCO%wdc5)?9oH?Cg{*g(KZSr{cz^0
zs{8tL@oHSFhNC|!x|E1kV+`ex3F+eN57AulK@>io{ueVa5h-Vp`?v<ZKTdGI+Nu)G
zrft++RE7CrD_slWbLy$i0|k2rpbC#H9!D+B{{CCbYf=}Ah=rL@=p4U`j#WW(4K8o#
zoH9@RSsooteC$W#dl%S$+=R<(lzg~+WRMhsQ5sJPY{A>$0kF6qtWY!Q1EG$pueXGb
zf;?qfAg@VphRMF^>H3Sczjtw+@P(z}W>h<h<1Hlp_7v0$stBngZgD!4jvFqSwt<9`
zUU0XkXg&ZiFlWd@Y2DiBdHTs`ESuoRg#}OC<l0{boUula$&N?;29k1DiIgpN$>4sr
z&nLuXkhi`p3*gFjUNXli+iU&}_WXOHGqI6Z??=A-#9!gP7d43+`BBd@BI+fDnrdj$
z&O~KnMW}^~UUNO$d=QW#&wRNNK{XIMX1??Bdp)CUL3XfAkIJv;W1rMkxz6%6A6(}6
z1I^L*V0n#|umtoI<;WDlevjm<i47&I1=;JkI^hiSJYd2*%|$|ctT!QyIoee$rriPB
z+qy4Wh3BGBe0F*SDbHI5y6BQ2ir`TS_PC|51DKu9E2M6R7`<;42!6Q5`_9qYwL@0N
zVTO&;)u7KAX?nj&%(x=)Uu9HheKRx@zNBKapVJ<Kp=yKV?*-Gz^z}3%9}`KF1Q&j_
zW#WZx5EvoLz_@3)4j`o%6KLB6Zo4_tc4hi^mKgQT%`T2B69|KY`Tnw!qOuo({U_iv
zdx7V$7>J}Y4~`koFsny64r0EcxYI6F4K2%QbcCR4t7DK3&`zKKWl-sc0%Ku4iEiJJ
zd?FW+3kDkEjvYBeEno?J$|T7QKg(}Y>Vjk;*ThGDZ>~g!!-#`#wlPK6GDbf|yf0b2
zwkJ86%|++NO16bd%3cRb)M=KiG3LcBy>7L*xta;jyhk<T6!e6@!In5o&itlSbzdlT
zK6g8@i8P<H9E3*|?blEqd*CNA^<wR8WN>89`6xe*>jnNb_;+lc`G2q)>j*odWKCYm
zZ#~>L04X33*Ux260CQUWF<*<;^|5jeKxv#?KX}4BS7rVdVSSdVD=7(08rcixe1;EU
z`_)(PEv^Z=K|?u#>k<M0Xi$Q+fMC66HsLqO14QIJjfoc7H+7awFkjQ@QsxlsrHiXb
zZP!yDVG&7LED2LV<#RPtz_Js!G|e<Qx_1LKTE!pQEfFKwyYYT~j8<84S5yQ8Jixun
z-JHf(VO+sSU@{c#%5Imv+8{hAtLTtbPctYE?4lV_?n=C}ECa!^(B$&m04?=HE~*WO
zat$cB7|`+PQn45><WXL*%+OBq_y{zHW|>3%kbUP9@~PLV5<)1j+mS8};9H7YvSFt;
zaHLNSHO-{J2m;-a!)hZGNopy=<Dbb~+hN|3Vs&dFNBx+j46bQ?=OG-Hs_Z~Mk|p@I
zY@KTU1UCnMGr39zfX%ECNa#HPWrkGjHYqpN%|ph!3M4VFQu3h}G;Nu>b|hLwJFe_X
z?Dt|ep9kHbi~fQaZEN04@dxZ3Td4*Eh8*#!u~?K5K$hcof-Wj9#RJ4Q#JqiBJ$H!I
zh-{?zBl3C|cJKBfv-qWSU`9zeO-!XoG^6b*H=-|FEw@<9?1R^c0hCP40>5PM1lC50
z3ABK?0PM*o!*E)0HBrjb#CT`E;+4gj<n2e~oWpo{^*}(@ZAC!-n%K=_Qmoc-;aTCS
zs%I6N1~@4I_QnLPA;9+;eDVL4?YzdIFa^;=IBGgIrM+2q&$x;I;FxxD32~3i{uXRx
zo{M#2^Oh6{H^CAFv2rsI&X3b^TO_9(!#i?mNS0FcxXq@VTI%o%bq*T*gL^VN8drQf
z=WG5K1*ub<&R8vOkxj|i-wQ+^nnS5po`!=rDbGKmM2xrY@iM5V77l)mS+b&kz289w
zfPh9eJeK-e^4P93KFc{^G$-{2a4bn}9m|+$h=jsqdU|XeO8G4Ch>?Jtgf#(l3m7~)
z{OBu@{@>`^Z#%&Ke*q2w@IDT1g(xkl9QF6HLZ$UtQ-+g0RpeQ4HtYT+^oabY&i+G_
z-!v9L9sn@!ziNrNoucp>W5CKqHr9+vEnS?KX;X9-X#{E^ssC5{5Ni0jXhasqaxbQ<
zDW#b5{~q>;B68HYRwXcJsGL`jnfEp<GWPH-sO|gDk6goo%2c@AaYW|3Rs>baBDHd{
zeCj+aMq2htBte4<pZTDfa?3+0k>Yl%V@z(#7eFs$WxIxYuK!EYu=Ho0N?ope3*mhn
zq@GHT(h7l{XS!LBcME9YrD8Nie$615x(IjhM{NrVs#Q>qYGo8bBus?Oz3?0v<6rq7
zWf|ez3vM{b)v<}@be#3+BQ1%mkq{O$=Vtr@^*N-_OGQ|b{(YZo)aA4moM7lAAE7K?
za3d*Me+R4|xYUh~>-+6cyFN<<yn<-4Q@kvusDOW8V%z+Uo&QDUJ<4=ES;M)&E*XUA
zRieNZR6F!n=yU>Yu#wEZH0iDervw6g04dtZ-1SW7c2tDpOHowq+KCJ<jx9Av#wXlg
zUXI0%OBDYyVyXZr*+!w#zi0<DQOqJDyyOYkmTUJ4>58g=stx^XVanZKdyk?d#u%`~
zn^~m|N2s&?o5?|j65X1PYmBKtD72fBAUQ|QzB_o?oe!KC#X3zUK9b#oRDfb-@J_%#
zZ*t%r4KA|3g0gN=5tyl?ZQxqw2ct4uRNvL7p=~TWQff4Sv@VY8;pKv#68H-8Xl2-#
zf}ca4d?4E@U|_o-)#nkA{)!*W#}NP7_!91rHG9VW+QT>tb%`>Wc&=frJ{@xNT|!X@
zt4wS3H2omntDnI{={k%YS4`vDMe?^i|40+e%=G7t5Kw%gDX76BjeqSV$KIk#0$}d*
z)@H%3ka|e$_|c9FZ|%g=#za8g@u4+@E&0ez>meTSFYW&MwdRKQTAY4I$)Xc~u&dz7
zt3e-F8GZ|T00<H8?~D5Iw#dH=!<lP8V9{YnfiS3zJ-d;FrujS%p6}0=uhX^oyZR~7
z){eHTcr6sIHf^@gL4#uQRQL$JqL!kgcBfCRDR`vN)mZV%EUEtL@tO}Q_D46BQ<rP)
zvGo?T`REy}d)c7{y0h?%>e7$7f=fNhQ_HlwBOs?S03Q1#+-~sdHB~1_cOc`{;KZp}
zP4H#kS}13HC6EQuqKriw0qFM^bn-F9GvTIn2k~0nr%Jdx_V0!>sj_}ckv-)Lp`NOL
z=<6-(144>tJ4-lUT2R?T?U`ESpPKt)#Wy4=Ce8x~q}~{eMSS68#{*d!ZhG$9s7GFQ
z^_*@Z+ylL#C|XS6E;HceDJG=Q-rwREAo&P#JFI=}k2yfS#e6#GMT)fxL&!{n=U%Lx
z&Xi75Bw>%kv7Ger2<@GC%@Tp_DL?Jy-=}1VZ4(>DEelr#9xileM0yB~IcqM-<^t{#
z@V?(pY*^PabIAXuM9!+O^_py!@ls;r#a4f{>nSr$`ipXWXR{Y}guyVVW(UB=EhA#V
zn8uo9q_A?B+{`Rrc@3ouKoI~uo<O-5$&}`!!B{#9{7R9*THUveou`^ViXfj1#Y(X$
zQF@XJ5XQ)zcmrm1j%<Qe5E^0?uxqJ+c;B$@F{76H)PFaR)9?-c>-p_pu|WNX+PtZ;
zVkhwJBCjb{yH20DJ6f=Vc?D+&R6ls>PB02dA?TiuISxS6<ix^-kzL&Nj_jCgc7@0k
zlh8A{sW%2;`l{N1sE%5zfT)!5D>q;9)(3Ka*zoSpN?1EVs-qoW4sagH63EHYfqViT
z4O9y@J40||=DyF#OkcYzlhyBl$p_BT!jmowt>>6H4RIexY*cB6reB;RVlUwbDjG&<
zvWuou314R_rl1YNp}=`#RO(G)Z30xN@93rR-pD6mdLR!tvQcPu(!^|jo#2j1(zr8D
zmcNH#;J+lFviB}tb7h;--AT;7z7HG`Jjgv%?-?T7XJ?{%hw?nU^+=$#w`7Z6_?Pb-
z2i*BQe5nkgT86=r^2nt^ndoJ8d1YLIk_UOuz@4M}dV5doi+l4w{0@b{jo{O!Z3CY6
zs@4w@HGSPkvg5cWGkTvy;Mlw&ZN`?vNJrcj<$=uRuoH_6w&x|V2*B+OPmS-BV(5b@
zF=A&;{TSc(6b*z+U8dIFcYW@(F#dUj`?hu(iGgLja!RmL{Rc<5vSZ@|0e4$h^x5|@
zJ#cHr>phv?9=p$KTtwTZ0uCpv<8q;1e8~GE3>y{`rm&6mF6NdfLIL+~l9~Z-PBlED
z;mQdDNaWUkK+o&6e=_4bdG6#PAo2JESfg$*rO<UJH(cH>EI9}R)ZGS-ZAXrP8wi%R
zVx#MXh<0piiQ|IfXSeoD+Vzrl=H)MU_Z)pdfLr;<#wJWEVtwUGMz#ynBl!T5@o<{R
zs(o5qwwUySc7v6s@+rJn4NcibQPT+cw9TEy@A{{|Ff%c_be&1UUKfow&=v4Ue6rh$
zNKch22=fj~SCd^27b~v6?yI)lqP&pRSe#3f#4fQVb8ylmiUQ|C-e40+$WdywLqz|H
zcjj>>!`IEJC|$pO(faCXc341G4y6_CyIw^r#-@v;J|R?p-}P}VLGY`uuI*1$)VLZ3
zfFp3(Zn9lXBXEX!jz7b+MWJlp_VUuf^!L+_aNcuQGIjx{zY%YpGD>`iUXM*>jRc1?
z>OA~y`@0KzrpuU4AVPfENW)8-wOGTW)^vONzm)Qu!7_FCPuq^=V$T|fSrn#EBvsJd
zMD=95+(#eJ=%@O_U8K-7S#A032HXc9y92&XGo;H;-u0)}lF!+HEGh6Q3Zt1ZW_9bo
zeDL%m+r2(2s*c18V~aIpnG8AfOcfnOKxVqv-r6A6KJBQbW9>C0t0*?Gk)T<G5AA(^
zI!3SH5^Vt;`fHzntU<<*)%02i5=)Z0_55N!)y8znU2=!Uy%|6q==UqlI}U180|Z&3
z$x7+EQR!V)p>N~3^6^vK*{|_6kK{@Xepiy)n%A4vZwGxQCQrt8@mW@Gz=pv@dq!wt
zkpS0k+8HV9!<+rmKh@zv7Wdy;u+AGZ@@cws5I&eZ*lp7Sjg}HR6Tt$3;whsj9h9c4
z-IVHY!xYI|guxIu&*k%OkPt}dPE&Zz9orNbJ5r5&n)ZKxd{0<Mj{w#1Z22-`ij+;?
zF&M+}4)CZ6$Mnt)O|1y`uHKj66}rT2h+%Gh*z58`*nt@dey}Gk_SQO?1k&<80<dS<
zZ9H^kMV-U_{(SF#yQoFrJ^TpPUrpUbvH8b-bXs_=tos&C{wBBa!OY%OJ(6pQz|=;e
zrg_@J0+6;U<NLr7!h~8>kD=s}@GjTCj+uL1L=2o^0BwgM=dXOsC^T}Qy2iCx{YbZn
z#xtD&$jqsd#u%JxO)S{zZkja)DJj%a$%FqwpMX@@!U62S>zmk^5*CP))sbb=*r{Di
z;S%Gd0TQ0nyi61kE4oB0U>m-iikO!vJcQmG*kZz|&r&CpfzRw-2~V0fec_^FZT<V1
zZ5Vd`%KkAxRO5+}&!V~NKCi$92i&dX!@VphI(rpQrycC*nPA+KC2b3o%=6T4)@Iad
zLvErJilv~7j`RQbJOONbN!Xa?fRUJBcUP831RI9XJkF#eR1oNzV^Jb33^YT;^oa(;
zD;rt3EM1nU3C2TmqVh=hZ>6|wf(ZU9tf;^IDenERu~L*N#_}S9?V>}E@iDXQpEh(;
z3Kf;e71+7GT8Y{|oH`zmi&FCnXpGq4cE><zd{+P|*A7kv&p;?p2+AVwANe`e+G%|r
zVl98EaBi}u`mGB3!?N9SV}N1#9|zsD2djx$wt38R8(`p}4K6&xa20lWfaK?O(1FjC
zEEb{ugs_F(7?)o>mESFvPZ8u4#pgPR-aV&jAqn^<w0QoEMK;fRcbue+3Pa{bNEswj
zwHy?`H2Xt2!ap2_d;lA$1=WqnUfaM1=+fEnR4-9w<loLes3@>uS%McspVfbFpQ9oH
zW**yf38X;5I6r_zX(Mfet`#hb))ThGrQGVz=1R~6|9kG>eb*+i8sM9l*;%$iM3+|T
z2l;3K@<3CGY+a(39Y|1m#}2J4xKb5;fH3ltNQwJ*TI1FI<A2~=R@k0wS1Ri!u-6Wg
zxhtzU+>^`^O>Gg+jk+gR_(j7>OLJD70|4T;>~(}vi4>DU$f-&2T57toc9oS*f^gDI
z$;l7iYr(&93oK$mVZ#2y^<g`?qkt*6;p~<vLo0`@5<zfG@X<?u^pndvjst~dMxcC#
zHL$z<Yvpb%+QR7J>b(;ebC?in`-H%Hd$ZF?7Eo#~eW$CHHSu%DvLnkaRx|!gPcTsC
z>xGc<ACOHUE}d}+lxlpO@puM+H2a`VN_IAXX1eAEi!L;{qeJelY!^w55SQVNB1j++
zHg2?oAje6kT&O3{wQkfB_0xRc*Uq0(1e))1>XZa%^a(NnZLVrr*LwTWMcDsBsX4}-
z$F9kfqzSRb8C87#{65tx!{udfI*B}?7o_=@Yvi>E?#We$tfnGpg~PczOM@b3H2fK2
z#MF1>xw+xKT8a}Rz>hO+s?0*JZ?GoBEom>GtdnF({^%~Li0YZ+KJ<AJ`vJP+P!QW_
z4rSVywy!+~a<Tm3*XH$tjlQgaae^V^l-z5};Lh$1qz<-}A-!{w25@`0De|E=cFz~c
zT*sngdl5?Z8`!6a(!%#ajJ(c(aTE+`@bEe&5Vs7zG&8qL6^94FBN|+6I1T&o{2r;L
z>-_Gn*mOC7xz-WgGgEuZiEo`O*}W-<cTi8{VMOZp%^3jamboC|b`qeHvwlzZxBm(|
zMx+RNLL9H3QA?VywUIP;&%o&%QBk^FrH0R#(tW_=uJ>$qm+~odnmNgYG==bBCgSg)
zLS$okDce8W=;#Q~x6VggR*d#sNI4|5g~!dM>1QIswSl?u91(OtSG)wzwvX3b@B_+&
zgaI`vA6fohfT9RDN3n^hNXqqnu@kYm`*i85RRc#GkuT;l?-OAPvUqOU(@({K86i8M
zsKimSCx6~MyAcEH!S>gAtYHejPk5!{1^kdzeXM{%wt_wc1Iv9WC>%`0v@TXy=2uFq
zPu8Kz<zQ&;os-y59`bX=k-Ko7`<q<_oB1saoz9~?@1h;*%L|2eY}ox|OUHk~7Mn5A
zG+&?h=(6S5bKCGwI_mJP)6mL72WnB7s@D0B;;7U-pts40MA;#_nu==BM}3#{!Qh#L
z2EGXRV?cYlV**5ddH~?_uLN^z;AbS^#j!b{12Zr-^QW@1H;D{&82K3^o6C;RD0a@!
zok+i4Dy42?sfi2}n-Y6lz69|!T{1u&!#M=zK?nNw0+tyjH}N~ib4|5#<%07E_0X_G
zQg%vsECrkg**tm+5?yF=1r(;^-mnjono1qLhHl=tAg}TH0W)xDx}B4uWgSUrJ#i(%
zZHI5=@w=#qDKc}dF?}7D;bTC^-kMQrg!CE%7_5S<bUL7!E+ngHbjd*!L=Gg}@3ZBC
zx&~`x@VZ=y*VZR-0cHKDc`2PV<#IiFs^0uS%GZ*}$Il&}(_m(PvMT{92S2iw3sTQi
zG|X8l7rkuj>8<JHA4#-(nTt+i!UKxmK$SG8Qo)`-CH0F|^{1&2R>%Roi>(vJnZ4|9
z3KS|D>`-aw19_hQ0h=qOTIaqTiGd$lw6lK>r~CpGDWOg;c|2WyK~h_<K1M2SKOLUB
z^H#b===5JI1A09e#AatzJ(9~wqsy2*pZZ)tgdgo>8!z|;gsk{KEkFbj|IaYP6!fcq
z8MuC@r6k$fyYi#u2bL7WD$+P=`=>wx7xmK@0`$gT0vYHp51xT$JcY?*eKoB0Yf#ms
z0jej((b2C)2!gP;(&|MOj6t*w)j~SRrA6;k7CTiggnkO_8;U4|?N|`Ey%u~E83jXU
zo8mVCwSLKpy9Dk&8-eUfxS5Zzv6;E%c+9|%I_`#Rm@<&o_|Xb~R<g*`P<Ff6>k9zH
z#+-Ec@w;+K<4K%_c+aOB<EpZiz_4dR7U6F(Hr-C_@@d1OYO18DwA#hz;DM-k-Q}Jh
z%#O!DBF50av$f*o=*I%?+t|n`EHRr~WdROXXPwLgppn2)*`mfFvt`o*G0Zl~6yca=
zf+tPXDwB-;?qj$3(8uiG_iJWGqkBDOu0=xE1zZO;eFaVMQ<N`Qg1-OqDe6pnYOfXG
z#cJ@GpM=rh6XJHz3h?N{(MZqDX-95uu#k_B4duXEn?A5xa6YJ|SBPX1iJ@vtnYLW6
z;lfv~Gb~_x1?u!`$X-13;qVFd4}IO_M@=nC$*)5xqGZoz=fZw`gW<ngJUh;-tyq|n
z?a;szM7hQ?EV&?$yIA|n=ei15t9&_fBA=wb|A~8v#}sN+Z_kEZ5h2PhF8|&7HL($L
z!tA5vf5wAbW$yw6h}8S2-)k~UUsdXB7m0dG(z5Uq)9)^lEOUGh?$@m{j08&9v~kLI
zL1lja*VV)Wh<=JyQ(1|UIKnm!Oi%ep@FyNR;o>@?U6B`inavwCK7_kaPts4sGQaSG
zKWdAQL8a(ObLst?jXFEswD*WDi10oAz!&}x>HNrYxeRjKI9FgJn2o=zD<EfEe#5&W
zqNbWK;%9*lBTjhxk@@#SoYMb0Ay2VuAqVVN%%I6<K#eX3J6ydo^R9%*P|swW)7N}b
zL1x<Y&JfOxlX1|ZH-#q)y8$?xDDHvfzd6Z^wUqOS03uI^)9`y)r=WdxnRjCkf<HFv
zW&sN|P8vR4y7(3q69T&~t6e<bmv8xiC+>N@x}S5lID|I|9Usj&Hw(wwz+(~GI2VCa
z^V$6lRUjWJt(Lls^4^!(CPt-}ERG?d`!X#bGVnk3I4oV&Lm#jxR}1|3Ef5#GgRfMG
zSL2b*XzHkzJ(>f5jZ2BTv6&q!%I*Uh=a&6M>h`6#ZBiXDh5#u8$<<k|)N4v?Dq)@V
z&!_U?WWnTX@g_b$dwVwJV|Y0yi7-qU^0m}U#cFhW=%?6AuQM@o|7S*21r)Q1+{8iP
z8IYf!%!)e-xpwjlcYL2sNA@2IQOYTL8ML~H`FO4XITw_794OdSI%=`l^EMy+k_u#O
zpAg3xt7;qEvZrgY2-EO9b=AQV+$C82ve~3reGb2aNJq(H$Pa9lfz|wrDt`|ISO&_{
z$YSa43--{YmzV|Z7UUod5REQ{Ia~RcuCOUc5h1bt#(%LglJ9BKXEz|$#Np(#<eNR~
zKnZt2K%Oy;+{eu<^PzItqUSJR9pv4n@SJ@awwK&;Bhph*6uQYF<negIoi~zlXO9Zf
zCSez(b}VvJM{$aH1*F2>dF$>MHmee#d=iPTz^)Qrn74L6Rgm@Uj%mO9t=gRan>ZgF
zCyqJ0{);sE31iEANthymcWekoN#`JA>J~`Ye%MxykEnnu3CdTp>IEJc@OhRV$Q*q+
zw)sDKxQ8`)j}$ubgxi#{@Y2yCpFH{QZVCWq>SA|Zriccboi;OXC^&Vk8qUiBKDdWL
zo_{jO&Me>~FFn5xJ@xOU$(1&38$Cz0h@x35M;czQtpu|oqqY^dI<fXRwe0I4ABA@p
zZ7(J9wJIh(_tkCqo>ns&PZ^)j32lh~0!jrhIU!=Dg0zHh0LQM3GOAqzv3Wff#D?27
z<1U@4&3aSqQ-GT>=lWrZnzvAN&!w`CmZY+_*&m0@7>Nb^_w+lMXZ5agdJA(jFo=VO
z_q#(MUdx@I9#7vp=hRkQ?|k1Oq>pd8gtf)E`ZR3NbCWWX?aV<(E5F{xAW|FKVn!;%
zn18y*e-TbBPiUPYXmCyhTry(1*)cnr)#=KIG_k!VQMDAozkh~HZ~pn^(&>oo>Ah38
z-Kn>&!N5kPMF7>keNgDkg8WtS?%8*e8R(?^09)<4amFFE4)x@jsV<-8!2u0Y$<mjb
zL>%7^&E9zV+gJI&dxZIW4Sv-P-v^M#ec<m}KBTXA?%0i=b-|vQ4D82`9?+>bysg9B
zJr6@_zpT-g$pfe<M#wxRjr&y=nAK7IPz9O)B+RkPk+eNa{Q*>%bCLxcIXySy@_Jrj
z3)fd1xQe}H9O(MJcKmI;=5s6BHZEDOzO#s>>PcJL79MBXO|r8TlmW8ZQmch&CRa&F
zb|*TbXS8npki!%fVG*f9=J?FBcv#51AL&&SY{TAi;mMcXsW+_S-)QMHP7df{4j!$%
zy=xhm)&0D9=)A<kygnvHaJA7@xo+?I^c-XQkMldPcQ|ua+`qm@3a`VSzO&4#d|vJp
z2--%|47LaXS!)l3%>gVxj{fuaf;i2NU@O*e{=jK!TvAi&)=|w`@J7NOWQvN`qDbJ|
zFSm5~6k#MC`xq!Pr}DYo9r2BcTQVF5i`wVLkRV5$6-v&7IeCf;#;3?&j@{ukS8^ZW
zUB2w#n_aGiyPki~BoAoYYrir{KsXI2<*jdxga><P8H3){Rz=F?5#1Mb4!<At+D-%T
zrS}x*tb}QH9~V$?ij(cD7ooGKW?nCcl?K;7SpI{8n|6L;IpLi`iJh~EYX$$OF@nch
z`z?<f?)S9y{R}xM)+g|B##bZn+l*U#&KwvkSF8wurKWsR`{Vgbh)csk;W4q=syOnK
zc)lYAy=%*FmL^0#@_L{Y{%7Lgvf=tlDP9z7{to~xE?gsT!TD{bHe<U`Kh*1Ix~8MM
zB!E|ZzF|-y-7l^jm%zB#loCtGWM{f{8z-#1y4Fdo!#UufQ~!%EdF$5C4BbX6AW1IQ
zeh0CFPZ#s@!6Nmav@AaHWbX4_pU=CL4-BgMmf=!(H0l)0X~@#T^R3uX`wdegdXD1}
z`M#bUB*=+75L`z{Nz|RN@QV!L1ScywScyoB*HH!v4QNj6`p`^0VBBQ9FerQ<YlCH)
zq`V7B^F3b{RHo*o&v#+}1iKew01uhis{l5+<-`6b_xVZ=5`jfKe^IcWyP~&YugAAT
zFclY)MEP$+@0khx#o8dJxQKAFCNuh}Xc_b0@wu;f;XX}STxP<R1^S!WY4z+=nA~GK
zMI8*ic`+3{9_8&TGqr3UbrVUkHNEOnEA{UnJvQ_A&<;88o(fX(f4%>b-(((F1Y&^~
z(RZCQ=x+TDw4ym@m!h_4-duT}_(IJ=$JhSH$R<}*3x`p52&Ip}-dK3;8dvRYFp2}E
ze;-!=;~y<`V0wNb<f*Vl2l&!;%+3c&U4@9Rh;i1{GBplaHB@n$w|{<$maX46`8X{f
z$m@U8p#9*>CYdZ%ZhovNqoYn|-lZ&+Q_H`ZFozN2H)^mEf|_#O>O6AwJ)7{G2p~38
z{%3*iidF+hl%P++PB*Ox0+jvbDe@b}a?Zc}een}&E}ZIA|8E1<gas18i1`pX@N~-c
zU;gQ09$SW<<68)@)c_0_uIkjr6loGV6<0AksY;@R2VdXr9@_RQfs#Ym<k*u*q#%C5
zFtjZ<5O)lH<UPs06PTd#Q6keRa^V5cu~W9$L{M__pUcH$Lsf0_z~j9wf_@~$6d3{T
zy|QTBPw|bDZqQ;Xk5Q&eJ!mlsAAEW5VOazfNdFRI!ACwVZ|e)E51K)vr@v>RoSYn;
z51$-_yb})psgP0xBrS7SM&@tbtl;~_Cbs-0=r(x%IvNV>50+SujcbavSV3N<-?A8o
zwyz%nk<d)t`c>r9KnZnfYR_-SMiYB?mp)Oyfutm*<UHx;JT*&>cmnQZ5lJyCLG#&P
z!k>fi{GD>}`G2Qergl-p+2P>;G!%kKZb%Be)>>emnAqSp%^KQGSo+CsfmRwmz}y+!
zR0w%_VQig|#hMww<}0j!go!pit(M0x$@o==+u1-h&##2Ua`{6kI)d<HwJG|D*FXb=
ze??vkM?k?=%kHyu!_sXXLI#UNLoUbm7Ht69wmniD-0mrf!-IO&gQ=b%QY1$pqxZY$
z<KOIeepsgj8EmtSOO8HG)Ex966pBwVLX9}{aV$@|qmD}!D-o{yA{AVAM-){!@!vp)
zxEeZtAfwo<w2?QNwsFmEX!l&X0}FfDBMXz|*uIDeYF>+0u&_E>4wSPR+_?jBDqU2^
zC9$RQ)Z{2^Oo*G)jq`5=KS18&#J82+sYh)vkS!=MpuKg)&l6qvMa(DlC#o8d_%~=l
zG}5c$0vLR>QwDzB$W^U4Kdm3^p<s1yN^MG!rex^fcZ%u+x`!v`wAiR@5NKpa2%e0)
zO8@r2rd^69jBt^6_>VcKp*9wh!u>6F;TR?C2y<e^S=4GmQPC)R(NvDD1NvPx|4u!W
zOiK!k5nN1^1i_z3NEY7`1mCF1?<SY|`jhx?LA9o|9&sz(n%$qXD5dGeculP;uzqR^
zyASg&$h4&Sjv=NfbrrE9_Ir##BTXxqt8=vk-HgzKBz<B)s`<-j5njyUn~h(17QNDC
zzF2n`cLPd$AbQj}aj+s+94VDh)>iER{vJE;f6wr@7ok4Ns}{s|qm!Z_VJ_Ic#-`!*
z!bTHTiG(R;giXg+K06@X4ZcUk-9&P^_rFVCi;!v=C1W48x&8Gy<=rL8n9RiX5+gPe
z&5ywH;QIO=we1|rbl9Km96_gW1zMZ@d@fBiIv~}eq*Kx3K<lHJ{r!7CTa*7R1eLLN
z!t>!`UdGD>6+iKuR*IFe02pAB(V3l+e>C8z&#?5~kO5m=I4Atr6T6I!%&*-(1Vq`p
z@<?u}JCk$@(x7~j5SAJ7U9rH*Q2D9KHRf9Aq#Viw)nYLnMvZ~B#Ssd;PtVGO&a$&p
zalX2_)GO*N^=3X*azBfWCHlwxqovS;>Nx+nSXTLywbk0R1^}d8xBdf$e}{usf1lVa
zOap0HfY=(2*c~w9+^5I~MWMRioJn0sfDP=$VR?T9$nyzw8m=Sl(u=*8exPdIzN(9{
zPDFnr2XueS0H^4}E9@HxOKOt*{>T$39K#ilh3FF#6WoEbBJ(Sn%z|wcc8Ps%wY+q-
zvA<`!rG4H(+S}o*I3G$7#oe0BYlP8$vGNcq4Y?j%2&Q3<exj^OGtzr%A%5$l*Z2=U
zljGyL(I?8)tR3^>NZ=+ESyf=8<kXs^=a(P9xcsZ{D;*u6Hzo7Uyqns=2d@(Yln!Lp
zOodnA6Bm?Ew`tILj<5r(>Tq^|2OA*I2Y42iM1;y0iHZx3UB=&qB)kwWg3+W=R`>*_
zvAz$#u0mFy4fgzxP}tfqn_8w5tw6(G%@2sX`~FbW;<WutYD*E7NS9!C6AJl+h=CVd
z=Q1?>#%cdlsrB+~sU}hQQg*boufkOg0a6`u?sFcxRW(*DzYf$r&7d=|4~t~bSnY7o
zX^KP$e)Fi&u7wYGpwiYp660>y68ikchnCJgAEsRmpr$vI%rBRrM|ah2T5S!$WC$;t
z4k4_9kIdHeyLF(}2+1Mc6nZ&%trOZJW^eiTq|nrQ(F(N#r$W>quoDM61ReI=1+-Q2
zz{N9$Eb>WxQnOhOHV#yQuTxNiVi%X(-;I?>*fRA`FU`C07fWffVNn@r^B`Ay3euW}
z*J*b43UXK!O~MNCD|*xuz$i6EI&BrRb$vpCyp3)M!V8QC?^E?6>sAVp+g(t(n%E+9
ztPyY#+rr#cuzrs_$tz&cyhO7-X9){pw@9jzt~~Akiu+4Wn-k1JtjWXE@<kTQxy?B<
zjIdKJ`XE^+)&=$UuZ0DD>&DoR3|25HWOc=S5MW{eNP#UU9G5*P`{iu1iL-UqSVKgn
zQVP~jfG8sz4lEfiKrtNq{=s_!TcViumj-0gR>9-0cBWItpFo>0bdm{eDcON{p##c*
z*-TJjGHUd`WUfE!V=<pvRr+v(rzCwp_ez<Ur^3twQX{NCucy^#2fYHxM_!M=#out^
z0kZA&deKFGkwajhh(C}n!DB}d=!VgnZ^+lF77nEz9TI<lB+nH6;*`0tvjF>)Nf@9a
z<zmL;i&}Y4kbJz!{s|LiuDLTsvLQRE=s?Icg3xSAg40LlnOOyJ4&=;`$H#D~|Gx@3
zHn}>wpqLQHww?g^cW5B^GX@#eQ+x4S8z4K!;kqkEsp^yi><@xqrrC`i+pdTBfihgf
zKQlIDE%BMmJkyMuVqedOLurCWbn&IU2<zI1vZ`<E9u2^)I$u2S3ypGboHdyAL<)HW
zGi@<A_gf_I?rJb0Xy=HGEg9%}&p(UE%3D@|;+8b!+XX?#fSdgyM~0CoD%XTFp7^*p
ze|%=&ozZ(F9YzpHm~B6Zp8u?##d>(S{am1D-A1%^zV7X!drm)|LgF06vl_p?sxS(i
zVrVypLSHs@-|GLf_Byd^bq6F=$XDQDuJ=Fx2vm+y{ne*(--ik*okpDEhrf`}SC~F)
z5fB-x0X7a*U=}{)wazm~5mPoA>HGKqrBL_F(k(ZOL!!N?$T4ETu@>;?XygAa0ScWZ
zU}IMZ$c~;Nh{-eyh#h;4SmlNfQ~*+JpbH;f;k)quDIC~eMJemZn00SpPc1>KKj>{#
zfusX4pkF*fm>>nzg|;iWR&Xy#MXi9#xiZ^K5W8KMIu3uOCn+Agxu?h45iqJ&n#sus
zJIiwjD!NTYA>5UR<diu5<?(NW*8<mO#5N&xO21jb{Rm=S(KUm*$J<MeAKd~j=hVU<
zBO482`k=ZbeKeRy?l1RLr6Im@AS0eMlPSo~s>0t}%9RQ{?39=)7V>}cSHO5$?d+B-
ze!z^k{#0lD?^<xfzq+sdCUh2@34Vo8VG|%@HF;6+MGPkjvbOwpPt-5`JB5$*`n(}t
zBb28|@<mFxJ<_7X--VH$R4MR3qlh1D5Y8dUo?x8$L>l=i`s~dEZez7}cd@o|<!kNf
z>+`pZz#WjhT)-M8q?&=y0&EjKaLX8xL#cLMNa^ZG>|bM0!L%4~DQLPB5M&L1%ZD=t
z6Rd$di8dGQBLT@I98e(qCE!JODARxa-w!8K1mN3QtHTIx0V}?#SKWju+~QOP1pB0x
zL-GTtk>px#9MhU<0-1-jKw5}ez-NTXn*T)eWvU^VoB5}J4S;}H?nJ=t5mDI!ZcK^T
z7OGj^Os?}{gQ|dFmh3Cx6R@zu`A6HnY}3g2OQlH0snyD$G)=X#CE%FG`v4$;@P2?X
z1P7%e35zk<x5b;-8X?e|;rKx^`%yRnP$wyY`%aD6O&QV?T9G@UvjS+*ggAjpe6Y>1
zgG~_@5l#(k#{sq4@cnyr{KwLIeHtY+x5JF_UZ56Vd6}Z6t$xd1z(*k5scKFFH@8;F
z5n4y8BO%p$H4lA<dh#%m`~n7q+Iy9c6Y;wKva;Z{M8OUKzCkyaaX@`_`noff?(hcR
zhlj4HH!jP`y-2wU&W|k+ks@7AuoPq;>Pa&7;YVkbI8Zjw*|VjLO@?d(B0+koe>0u!
z8;w9Z&BC*;V?aELu^C;UyrPw=CRp9Ye$)UJ4DD%0vVrAQ?w;oUH}b;shV>Ah_X>iG
z0s`|tD6@%!e1hvf8elF904!}t09;IV*J-1l6?r%Mnw-_-xn}vdS0%j;v)#68<Mx)5
z8ZN<!DBywztJSrO2elb<Td2lB>w2(N=%L`|3L$XeA%}fEK2<mgTqwY)HfrV96OU<3
z<_0Qy9Rqz#ZTIRbB48kxYY(*P48!?kPrnoERCAWbd5IOzm^qLP=7?H{5hJO5to*Sg
z!Jrd;@@WR(+rT#kgqcmR6&O%KtF1}lHcbwop5kZ^D+qLB&R%*b=|gpNEsjz{-M_g}
zYf9@Q-;zrBM$YG!R2bj?6iZVZni-gUj(T*P;sU5jO%eY(_Sc$EROb_Ru0VhrZ|Zq+
zcXVt*1|~xhEGs)tcVd$=K$DKIh*)+b<UHy;5T`7+91M-3U^Us}J9eRR3QB{TD2#jO
z1OAxDKfyBcvR4E$Lg+<Fum%Be_%=c$`^i_tWisL4d%(&7Zq@L_3FHb~c)_K?kbLDR
zrTFW6v3?cr+rB?$&S&8TIu(v4LL;C^R`wjo>av3_P(K>@^b*FJ&7DNy2kh3j#Bwth
zX1PGA!pnqZJf`kjVW_}~#k>c1_&hh#c;nA0TSye_ygt)VBbjrtA#7Hv2>>un=luPw
z>*&wKpYl<u8f}=w#|l7gJ7b(=<<yndu=H5E7TOQN<ou|In*oQ@D2-Hc<g&aG{2DSv
zO5^SzoAGt^F9pW7F5DgD6`#V7Ot1p{QTT72t5fjRNq~Q7d6RHh29J(L#3^OUG03nm
z?}&nZO!{kbv*Tnasu=jtl$^qnE%W}YQEpNK;NYc+<BZ@yaKewYHPK{~vZ2;6GaA66
z6q|+51j~+EX^rZXe9LcW{+;2CH6RGmmmSf>4U8{5ib*^z3Sq?MOpGPyF)BfAL5h7(
z<g<Z`KoJ<jTEXs=q3*AgU=O(Xpuk~zl9(q%7WEWW021wlgvXfResQBtI1S5yut_IF
z#Y5m!`mF;|+#Y-m(c}-A<$|uttP&qgD=Nr%Y`^&Nsm9(jW{!HyOJABcpX4L-yzy7Y
z73u*>%VJuF5JOwD`S(#l4+G+`;cf}2o-k?AMF7kkw*`64C<1txXe?iF8=^YjQ-30<
z$q<Lh3@ir~wmG}+HWfox;5*ho%RP6-_xjv4aQDUHRoGklydgO0TE$aPF3|-whIqzc
z_xa!OTewKEu$(^Uovc$4bGqTw_m^@2Kmb+{(NM1&aF5vaq6YkyY|Vo3+oX!YpHXD@
ziagBaoC>k$>Z<^OTqV>`lisqAsW=owE-`QJ>;|~a8@6m)pbw(e8p40hZ`Xta_DWU&
zB=9Q3j)=BDPsJMmc3C9g1|`510w<!%81Zsk&d0jcfROrx(Wmep=O@ClC6qEm8=Z6V
zPlKp^+{mZ1Pn?BU{r|s~M6lhUDNrGz3Uvc|2*?MY9J*i2jDxxj`U@ie>wMlSE`rd#
zj)fv4Z|XS0R*3~jz~?W*$dFoJo#;x`EYJ!7r_*JVCrHllACSSr_q4CK4%LfU`JRJF
zuZtMA=Q6{ate%;Yt{UD^i9KkRoUzUY>PK`;><y!+m%QTV^-z)%&x6EtnzVoNSfE*n
z6A7e-WQ$+|qyUTdU$?w+ne#<dSP+cI%z``o;@IDH1N1^G50r3{@J>MJo~+2)jW1px
z>LdX^@f+_-+q&1^>~oadi(3>+=oFEmEO|!~_cDUkIQIb^o89d@zd?LD4{tk={Ol+`
zU7W4p<0!N>4DdEN+&Mq}Ub*wW`&f~u(02WMM3J9wrWkuJrJpo1syfiCjd&t!a<<?A
zoJIv!d~9($qU^DL-X9`;p_UxlL5<R#@Y=27xCHZhK1CS$SDPK?!Y%Ocr;6!oLUvh*
z&yWKPKLt&b3eXH#m`D3T5Vg?>1LPP4kP{A7%dg>@&u;gopnmlE$0>or%012q++<7-
z$6Tz>C|^I5?siLL&tX%%-mX?*d7DbWN<ZWJP{Q1^F{>{x&5r(@N4$kL?FZ^TDnFjk
znQu3EP_&aAQ-SIJxYV%m48o?8`3#$WSg&UBAbd{fcE!!JyUE{Q6ujVWZezP-jHJ}P
zgQcI7R?{+SepjxccqAhFEdIyG`nhU{WM~S1OAcuM-t*1;_?-Zte5-V(Odx-I%L#^v
z&9>*{kO*8?Wq)Btd+D$ZRixVN8z)=fq0yQNQT6*;PaE`?tNAAF&8q*F-#zqapXl3i
zurL0wGV9a)uHe80L}Ka6{$+PlC;Za`i56?){gcBDr4*r%P1A!r9wzS;l|P!i8=$4&
z<k;4?G5n2rg#fNtBcgnJv}5~Vmg8J(cl)>B`OOFyeb6XpMJ1)N$p**v$|YE)r3}NJ
z{Jbg7ZkD;N{MVzh^kJk)jFQV2=9SglgIcME2?Ob<fUyGXtaxRscj?bAD`T2I=8&-N
z3~{@`*M9pa4n>E;-y?&zCq7_^rPJTFWHQ@E_v#06H(#grYO@_*1}qlY{cfuVt^7zX
zjBjS&aXR^dvenJ>x9yw?5PB=s{3s*Ok>T!b4<qIZ*S1PcMc7#Le{VeQDENi$Y`p%K
z&oi<eu`{kA-LpoMCKorKuM_Y5hScow`DNGXO>|(pH1*D!Yss}XGj~WO?S`HhGQo|-
zB4#eG`5EN;I5AF6X|A*r`$YBi>W_fw&g7G0t?D$u`o!?6bZEG_k6^qK0inI<{`F2m
zI|ovhS#1a1|AHLa4W!;tR?&5F4_vP4nYC@TnM#~DJ>+-fV?ZUsoATvtIk$s#d5=En
z&8EAx*u(|gHA=k9k0)An6+hb_GXQGlU8R^)5$uNILOm{@ahae9RXsTFhu*f<%H)h7
zmbu+iUz4m|`lA@yUa}g|_LPaj00d%qp{4r3*t2GEL#dslU8D1fy|^P*9u+~6@`P(0
zjje9HXNl&46ZIcGM!Dw8OywXh6SmIbMtL$*@IR{kO?nBJiNM&Vl7|j_xqv<oS}t;o
z;=w%(r{FRf5bHEd{30rS%7^og3o4@0)4%2Ul%GK<)MzXd%L5;|D{S@gc#>~;C^F<*
ztH@uu!9yb>DdHn6oH|NNz!sK#7D<~=9{L?4TC6Ew$8qHqm7WU=A|>A#o$~Rgr5{r?
z-W28;!YclFvq5zitORNiII1T|;i`y~oGgwnTVb$&Y9EDrg<__FX-tabUg`+m)=+t`
z1D}gO9nvU!7g-*7q-dI667C+Qy=6*KqkSk`JzRV7eJE02pbvKPt3bCi@zZliokEtI
zuWVX)?OmE8IToL-w`_WX6J0s?E@#%3oB-*IN)`7&<xR3F?ka~T3!MJV2=Yboi&}G^
zZdo&<9Xq{A@vvWy)g%FWl<!opxPFDb$mXRYKh<i=v9ZS|>8QzB+|&l9T_&YkB*oO9
z{Haj&6~7N*sm}Ly#A1B+L&&t1y8m0@UP7F3e_I5qjyw6>!s8lCrp;nx$i4&Dq#xQN
zz&$J}ics>gPT1~Rzxisd<3+GW&_NNq)OCLKBJ*d?lGEF$7W#q5N-YuIQ~rv8avzC_
z;%Lyk^ujrPuMm4bCWX>cjXn3Pl3YLMNAfJ|(T6jNvXgRBw>6hN|D-Ju@Si5u)^M^d
zgY@ogLp9%vi-??G?5U|-C&%lYhT;N6A5XM2KM`p1ww_7U>gq-#-o;41Es+VVo9T8P
zFCD1cfKT5{=7d%_KQoZn35$D|7sMgjzxJ(v@u1J<eMn#8j_~8Kg*j+}O8DHcHo@)f
zZsiNsA?w737n%=mPkp~oiFeXj_i6GW&9*d;@uvS)=bI8F^y6*litf1^5Eta#M?Z9E
zvj;C3A?-)hUwV7l`b6hx3WSfMqRd1hJ3`ofc+*vvs1;3VgAA(g1zcpb%Xrn=U+Bf#
zN;97r5iROC<pn91kuC*S`r7xbx|4L<y2yLfb%xH4foR;(@r;LFW?DDp$U7`2B?D$O
zr0!&Le0>A>M1PSEcuTj^8p*VRuY~s`ty+o}b3kSw9vbeH)UpD_TTg51ky4c}_KM4n
zE}I4Qdo#3M2spC@dRO1g?a7Gucgte-@2o0;W=Y1?iXQ($)o3(}oLV|Q7VEn>(PL#C
zM_xG19Z~7PItmqTT!>HfIyvEpP`0Xnen!M~SXfaH&XJe&;7T_-4ew(gwMH1JiGEL=
zS>I?XOK}9Aa<8`%E=<>Potm5fUi5akql32>qHb-Yl~kr4gQYtOj(7T$9y~$5YlUl}
zL2q+iqI&zjS8ODrh25>4KB)&=>^-M%4$yGvX!1IDiH31kE$AwjWjLRJd_o(}ZE&}Y
z7W3k&`d!)+>ef2lUln)TdzAN(Z?U4p`;RK`t@i&U#MQQ63X=!Ek^RijkA}M2&ZoYW
zf7II5Zx7Y-kO#IGLT0Q^XPdSvci!>q@Fc<EZOdKO-|TDKCYLpo?_`TO1j@2Ox}1D`
zdLH|^m5_SA7VkGhb$fih%8-8U4Tb8D>m8lmVr!2xSVn=+*2s3SHpY>WI*bUad}m(T
z;LLFO+W)@tkf2=(V~NOTWuDDxYAwy6R|99an*$yjG_A<q45;|gk6ti!eHybDFk>?J
zU-|$S@K*6+Y=q&(!3%+sKIZ#WbGbj%u2CIlP~T-89#fXD<EAtU9t@1<u{2tmrHfa$
zAD=P)uEUY2M#~4ZR3@ex@g=>D+^mEuXdCtdz(6m<A*!V2G}sZm=D=(F-95=Nk7rJ~
z_B|%JwA^6p@jC86U2=IR5@%mxM}X3GbU2Wrh;tJY72KQoe#KBoe$fV=Hr)t&b0WhY
zYzxe7PG9$@5~ecIL7=}c$fKavx&pPAf<BL`yW+l?OCI#NEQPZxM=^gxwriNg1U-|c
zRbYkj9|hexQxA{P<W%Pa$I?$T<$C;}3TKje7R_lQ_p!q(EGNP7-L2#82dNp~eSq&e
z+9g=Qhk5V!rVtMPVZ~7++2JEU5VZPwmSVVWxSSZ2sBZ5hUJj}@`Y8GE?It)_JNe-*
zQ9E&R(U73W4Adlptt?FSmh-~vEyp$YVi1lu`4Lup2@hCaRa;+h2NaOY(I(FpK`*3t
zvyH=6%^j`Hrk5ikE+d><io&1oJk4(04KhJX=pNr3`E^wltjn2^@G21<j#GjwJ<@68
zx%tgH5Zq#imRtqh;tn-+^E@RIWq*DLcMr$)Ti@IwkTcRwMK?X-i2_=C5;?-jAn_#J
zRfYQtG9reE**69-Mk3d9zvqQMImW{XX@flRt;`Ba|0me^`lYXCzE4cOme_Z$As76}
z7r%CXrptTXuHg0Xlq>#q+&7~GJD+WDm^a$TOy;iA2~>f7J8ySn2Wy-F*ocmzrEFE(
znmEn!6aSh0HZ+ANua#ZJ7#-@SfL7AF2oGlq=G|rZZR48!?C{&RzRO5s+sFs@=CVZx
z`#`NlZN6MMa3Vlk0}f$1*a1R;_QIh2K~?$TA^mbNt=#d2+t3t!Y!e3qtxHAf-R^7Y
zJ9#AAB%3O_UB4QGBb7>8<CIugfR8xAF-!yZiaY5iQGwq3g1^kDeVTzR3H7g^dCff<
z$*r7EoJ6<Q_pCRT1m2sCsdXLy{rYF*)v719vm)%!2<hvUL$t$U)pCaht}8$s<@5@D
z&el9l9At>KVblufgm5Ms##Hrv$k>XzYYOE$x&EPE-jZI*VydJkfS#F_GHH<roFfZw
z!N91#ax=t9_louz3&$PmpdFpet7z`tp1ik|Yv{|qt8Gt$jCwIsovgjZ6}0>jL~^tQ
zHLCXLyW9Mto;u<csK`K~W45`+Dg5S%kbHh-6#0o>G&)T}7ii&z3nn+WwNqR)05?vz
ztWoY@g>BtsC{aD=D~g{pIed_2`G6R!#{~m@OW2}<%&`V<t<uJ$TQ}+gSQ;o*4b_0?
zMg3*s97lumj<-7A(e;UFeq1FuQYPG!1uCOyFVT7bjCq1-x5P1u2B@w#t@`6bdLF1i
z`EU9CmS7>sEsiOKW_21#oP_TE6`pHC468W>+iS<}MhU+A8ux8|r>#Ct3?Nw!OhFnq
zNj{@~B~4R%@6S>e+}K1G9iSl{5Ie$c$S8+)DMX(1E9k0x;js>j=rvBOH*L{%8I9^u
zZN)AcJ{tv^8^*1}%qc^L%KrNnXgBKg>7DG|D#wUEx~h+T5><1Ft9DlMAQ*I;GlMiv
z+NhNSu<t#!slRms+vT-?{H-Kgf)fi)k4alxUF!HQdl)IB2DNkV{c7FDr9s7PXlCY@
zPg9YZ9~x2Bu+62tKYBaQ^k_Q5i*NS{j=g_Tq-0e$V!wg>gyS*GrHrm(SvIhbyw^xc
z;GHj-l00?-ZjTq?3dpT1NN^87O^dPXoDu8v2s67mIFK0E%*IIY37TCOFuMpGXM*4=
z@IR;yc!y4+@g(v-k&fl3dN3sUx4zzPGGFv#==#Q|%;$du#Aft!6<lJo#2H6wWVxz{
z6jXJ~&ur)XUd5-Z=pc(@zncOpkAkz87p0LnCvLc@L(%;A3mQhe4O*+{`1lQ3HrhyM
zo-`QyJrJ1PfG@l5_NEQz?6|V^lE;<9OYdT`kZko_Rf|9<emW?+yguB&9fFEA{GOVV
zS0Y!blQ?1{oj2E^R^FqT@!c5+eud;{T6RC$$AQO}cmJQtu00;g^o!30V>Wl%42m>}
zu~<@)ZQ3F;#$}Bdm)tL{%V;LXU1n%SG-VW{&@OTbW2kYdU5&L8(xfI@r6xkjN|8JL
zo>#x$|G%Bj`}xf0eV_B3?|II1&hxzI{d}Htv?H2Ngqf_Fswl#IQ)dTnjOAf#iU7*t
z%w5^H=g_Tk2`=ZZ%#$%DC_<cFA(mmKs)W8j5HOqat7hvrRyJ*XSG_5b;UMec!gy2t
zV05KEk=$(tAK(_gh6N~??O2k|OR1~dYjoZc_w5f!yFf-4i^Q(pmOXFEQbh}Jy2fwV
znoutDXUDz)$&_vYVpdZk0cjB^<hegCxC;fa_5EIK3P;@NcxX>pXadr#@aHfFihnY_
z3V2qqm%d1vKr_3!wr)(Ea@=4&;a++z5Lyc7n2Ev-!^*pa%SElHo?aG41k9CEoYRgx
z{RqY{r>~<CV?GT#E*~ZH!v4jwrziE8icR8teooKYH#(meedI|TaUQ^{3`szLFKXYm
zWWhg?eiq=}c(T=ZVAX<$%)>Ik4tgGgX@*Bj`N3=u)A<t9r$_bb4Xb8IaB@)9=4%wg
zhJ&@dg3h{kM$QK@1UE(OKrN7HCk3uVkzkE4y*U2B7v}p0qZ!+1(l_hC@xFl0vi1XO
zPx>5`vGpZ-252pu=9}FHn>QF@5jfUtK;WG@i5x;UNEN_NI?eh+nEo?U;LRj^A9bbE
z>6$;?3V>b$gdUmaSvckptdG^Aw<L-v=VUfdKa_5P&#~w@{aieMDgjH30Ha`3$mKIh
ztP$o`+fVc(K)B95IU90?NbK}E6XKMyzRr{tMqj0;+t1dmrJP&`&l^5UNovNeY&-}%
zys6LKfMhr39vqKYkX~U0tFGM|uCB=Q8OFa$k`&2_u5$OjD&GWC17NvYuS8#Dv^4H$
zU6Z5(p^k=6Wc2CyD?YmX=`E;w>jP>Lh?5)-iIvT<5`R|wHkVXmJ9Aw@eX-lI&fa$c
zpKkV?!4j$cQnn?dn!?7e1w(YZSEw0zSFYv|1M>*AY^T}(E|sw<mn;Z!lQGWH2aive
zmInDpB$HmYOI#&Guv9~B@}`>3ptFX(GxM4GIU{M#jiwCt0}%AzDI;;LGeC<+rVGs9
zC=Q&B@CvjdyUh>mYk8JJ6?M7!d4`(-B{KmHIa8Hsku$P|rHI9?S0n3-+kW0bGrdTD
zW%VJQz!@$DB+i-T%$wQ0lo;k>#oOhbj7qhq{1vxm+-+>lKSj*y`j7<FZ0cz8qJzX2
zy#kCZg7@MT{?XP{7n73a9=^-;O~ov1&1WvNni`UTn?(YrLtEK&^g3j49QeiE(K0)b
z`U`c0Rd<m8h^YCelr42<=Z?Q(Yth1fVXjo4k-URZpW_N)*@$_GwjFS)2bklb3A<?f
zA0b~jW3@%cK*86hGF*~=zvSpW_w8&fu{;z}R#5PoVf(33*-Q}x(D}g3E}!(8FbdfJ
z#xVce&l^-o!F#-cg1s<tQaqWx(lhV{2R<K6{5fa^?jANEr!{F5U~8xVPA@cJKMiRn
zp8$Ji(2))VHyX5a?9?4^ao0Kr(82&({_8FwLl`Sj)?HqTy#6EcJDtgP63a2{IT~Mj
zd;Ow8tk?exgKQpUht>8`4Ygl%nymsB<lWg+bOCQW5f2*~@da@QR-mV<B*@i@w*lqB
z)I8;;OC&sqjO3gc-58G0B}R~d+-OWCY)B~*Un-;EIJ*k9yHFgE==(g){+W@o#{pQP
zhP0}ZyGf_MaI3I3dm{ZLi53Z*qdxrI>n&;aqwrsA8*-4fY<o^co;lq0`l<QFR&emv
z^GB>Uz5$f??F!|=2<w0I-f}9MgArxDAzazxh-rm788(4Nh4?h!<S&gmt)BYZBOj*#
zxF~-Qchxh`MV^+MM$RL@T$G#uN*A0xv7-IQ_Z@D+H$8on8Oyh<)PmQSv4S0{N|#NJ
z>;vcXsf~Sf^y#0d$H<yRJ-u=i<b3dcKO#99aH|0>WFwg!tpS0n-!D6fRKsmg_$tc*
zVas=pbHoSN%C9%2sxGYtt;V3+-2;uu&#yeXyXFw|k9fYcr1EY%<gywj07$<xT$tT8
zQwTOCjbEz54hm#jF**Y$Escrf=VowML-Ul(1Xzk?$BEip`zcvF%KXM@g&J!$A}6y?
z%GRkc5GC2tH6tyo!pYfBsjt`OoYu)M(f9p*e6M@=F5+Vvu<Um#50GqBnCf`kDo%ko
zC5vMxOkc>J@8CUn*!ODVzJF<q(0pQ`%S0GGZf`(*{LCU}_s1(NpBwUa94t{K0c8vR
z$k~pEutW~_@;#}|+wgyJ`?!|*(p<(Io(CArOVJc*7&BF_xQ_Le#k#cB7A|~E%2V!v
zi~{1GyE?_ENcznvoh0~;0D!+cNeb917<z=fu*x!pnKqc*S}@z9ssuy{$MVM7(1i7|
z=uyT(j{O0q8y}KTzLlOO`Tzs@=#^X5tcsO++c3Dfw!O#&p0zA!uIX}8pP-MiGIF|A
z|4Pc+l2v4%m}*cdpUvy)*KM)@6vXia^;TIc?bcvll$A6I8hm+_bg^uT{If(~RO~0U
zx^dfOGl;O$Lg87+rOAzKZszAy3G1jdgkKZ0y?>=PpnjdJSgbJZX~zeIoZ;t1awV~K
zGCbHzqUwQ4i0M>HTF3~|9#6BNAzTAJAgaCh-zP?oHo}&yGmLi^f(I1?x48uC0%Qy`
zBkQy_L?ZqP1@^iRup}ckG&|(8$$n4ake>dpOYH27IQ8)@GETFj4ZRdPvJ46W?z27w
zbuJ0kqM-!XpdDs(?qsQFyr0ZM^id|n)w(Z|=`L2uev(Oln8VUzOcoE-NgzN*sVSMH
z?*TucNMHNR=*CM$4U6AyHeAa7Yrd9MF{%)hG<dsMc*U(j!D!c4n9zMeJ68?`eJo;V
z%^jBX+zvJ|{>~Z_VEg2IF(+qSh#Op4+se3>rpEP%h~32R4Uwc(>tR}8zFm{-!2TM_
zhdSN=GBX`?+;b2S1vkEtJj3ZN%LnLU`orr5QO5A-y64~!1{UHVR=SE%Y?>&!$m)JY
z*;W9aR5)8^OzLY>C>oGaR<H%g$T)&Xc-81Uq)W2>6e4=OE&dR%-hB6_a^8*AC9j{2
z+;jm;akJ1utiryT)pQ5BdI9+am*Wg|>Lv)C#GLbZkO0#M?jrsHf2V8-Fvl`3D)J{{
zYmrk2<6n4RJ5VwIRY$q?tg31~^~r`YuIALiFG>TZkYjO<P8&t^bfpCJTC=^9Q}4GO
zIUW-!@svbK;sQeHFYV5wy&$Xu)|r-7l<n=M!FbpB<9L2|5_?@&R9^-@3IW9Pfj+RD
zYCY68MS99Un$O6`MZ^yoTB~%BN^{Qk#OD~ZO;@~k>l&CoDAfMRk4Uw4mwBi@;aL^@
zm4lZBCN;mq?fpX^`QPKzCqZNhV1vcS0^kDd=oRHsyfT*(zp-7|y7OU;G5noT^5%p@
zF6r|V8~>293OqB}0DDO?wnaU|EQJIgh9;&$esVkHx5N_BzgR<{4&=^hDI1SPPltK9
zSL5Iu%JRU)-@I=@P85+g#;kdtaQaVgk&Cf`7DNnP70#dT4iCpiQLkESz`&1a=WNS6
H9FYDWq0kBX

literal 0
HcmV?d00001

diff --git a/4-Classification/3-Classifiers-2/solution/R/README.md b/4-Classification/3-Classifiers-2/solution/R/README.md
deleted file mode 100644
index f59c07cc0..000000000
--- a/4-Classification/3-Classifiers-2/solution/R/README.md
+++ /dev/null
@@ -1 +0,0 @@
-this is a temporary placeholder
\ No newline at end of file
diff --git a/4-Classification/3-Classifiers-2/solution/R/lesson_12-R.ipynb b/4-Classification/3-Classifiers-2/solution/R/lesson_12-R.ipynb
new file mode 100644
index 000000000..d1a6fbf2b
--- /dev/null
+++ b/4-Classification/3-Classifiers-2/solution/R/lesson_12-R.ipynb
@@ -0,0 +1,645 @@
+{
+  "nbformat": 4,
+  "nbformat_minor": 0,
+  "metadata": {
+    "colab": {
+      "name": "lesson_12-R.ipynb",
+      "provenance": [],
+      "collapsed_sections": []
+    },
+    "kernelspec": {
+      "name": "ir",
+      "display_name": "R"
+    },
+    "language_info": {
+      "name": "R"
+    }
+  },
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "jsFutf_ygqSx"
+      },
+      "source": [
+        "# Build a classification model: Delicious Asian and Indian Cuisines"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "HD54bEefgtNO"
+      },
+      "source": [
+        "## Cuisine classifiers 2\n",
+        "\n",
+        "In this second classification lesson, we will explore `more ways` to classify categorical data. We will also learn about the ramifications for choosing one classifier over the other.\n",
+        "\n",
+        "### [**Pre-lecture quiz**](https://white-water-09ec41f0f.azurestaticapps.net/quiz/23/)\n",
+        "\n",
+        "### **Prerequisite**\n",
+        "\n",
+        "We assume that you have completed the previous lessons since we will be carrying forward some concepts we learned before.\n",
+        "\n",
+        "For this lesson, we'll require the following packages:\n",
+        "\n",
+        "-   `tidyverse`: The [tidyverse](https://www.tidyverse.org/) is a [collection of R packages](https://www.tidyverse.org/packages) designed to makes data science faster, easier and more fun!\n",
+        "\n",
+        "-   `tidymodels`: The [tidymodels](https://www.tidymodels.org/) framework is a [collection of packages](https://www.tidymodels.org/packages/) for modeling and machine learning.\n",
+        "\n",
+        "-   `themis`: The [themis package](https://themis.tidymodels.org/) provides Extra Recipes Steps for Dealing with Unbalanced Data.\n",
+        "\n",
+        "You can have them installed as:\n",
+        "\n",
+        "`install.packages(c(\"tidyverse\", \"tidymodels\", \"kernlab\", \"themis\", \"ranger\", \"xgboost\", \"kknn\"))`\n",
+        "\n",
+        "Alternatively, the script below checks whether you have the packages required to complete this module and installs them for you in case they are missing."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "vZ57IuUxgyQt"
+      },
+      "source": [
+        "suppressWarnings(if (!require(\"pacman\"))install.packages(\"pacman\"))\n",
+        "\n",
+        "pacman::p_load(tidyverse, tidymodels, themis, kernlab, ranger, xgboost, kknn)"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "z22M-pj4g07x"
+      },
+      "source": [
+        "Now, let's hit the ground running!\n",
+        "\n",
+        "## **1. A classification map**\n",
+        "\n",
+        "In our [previous lesson](https://github.com/microsoft/ML-For-Beginners/tree/main/4-Classification/2-Classifiers-1), we tried to address the question: how do we choose between multiple models? To a great extent, it depends on the characteristics of the data and the type of problem we want to solve (for instance classification or regression?)\n",
+        "\n",
+        "Previously, we learned about the various options you have when classifying data using Microsoft's cheat sheet. Python's Machine Learning framework, Scikit-learn, offers a similar but more granular cheat sheet that can further help narrow down your estimators (another term for classifiers):\n",
+        "\n",
+        "<p >\n",
+        "   <img src=\"../../images/map.png\"\n",
+        "   width=\"700\"/>\n",
+        "   <figcaption></figcaption>\n",
+        "\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "u1i3xRIVg7vG"
+      },
+      "source": [
+        "> Tip: [visit this map online](https://scikit-learn.org/stable/tutorial/machine_learning_map/) and click along the path to read documentation.\n",
+        ">\n",
+        "> The [Tidymodels reference site](https://www.tidymodels.org/find/parsnip/#models) also provides an excellent documentation about different types of model.\n",
+        "\n",
+        "### **The plan** 🗺️\n",
+        "\n",
+        "This map is very helpful once you have a clear grasp of your data, as you can 'walk' along its paths to a decision:\n",
+        "\n",
+        "-   We have \\>50 samples\n",
+        "\n",
+        "-   We want to predict a category\n",
+        "\n",
+        "-   We have labeled data\n",
+        "\n",
+        "-   We have fewer than 100K samples\n",
+        "\n",
+        "-   ✨ We can choose a Linear SVC\n",
+        "\n",
+        "-   If that doesn't work, since we have numeric data\n",
+        "\n",
+        "    -   We can try a ✨ KNeighbors Classifier\n",
+        "\n",
+        "        -   If that doesn't work, try ✨ SVC and ✨ Ensemble Classifiers\n",
+        "\n",
+        "This is a very helpful trail to follow. Now, let's get right into it using the [tidymodels](https://www.tidymodels.org/) modelling framework: a consistent and flexible collection of R packages developed to encourage good statistical practice 😊.\n",
+        "\n",
+        "## 2. Split the data and deal with imbalanced data set.\n",
+        "\n",
+        "From our previous lessons, we learnt that there were a set of common ingredients across our cuisines. Also, there was quite an unequal distribution in the number of cuisines.\n",
+        "\n",
+        "We'll deal with these by\n",
+        "\n",
+        "-   Dropping the most common ingredients that create confusion between distinct cuisines, using `dplyr::select()`.\n",
+        "\n",
+        "-   Use a `recipe` that preprocesses the data to get it ready for modelling by applying an `over-sampling` algorithm.\n",
+        "\n",
+        "We already looked at the above in the previous lesson so this should be a breeze 🥳!"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "6tj_rN00hClA"
+      },
+      "source": [
+        "# Load the core Tidyverse and Tidymodels packages\n",
+        "library(tidyverse)\n",
+        "library(tidymodels)\n",
+        "\n",
+        "# Load the original cuisines data\n",
+        "df <- read_csv(file = \"https://raw.githubusercontent.com/microsoft/ML-For-Beginners/main/4-Classification/data/cuisines.csv\")\n",
+        "\n",
+        "# Drop id column, rice, garlic and ginger from our original data set\n",
+        "df_select <- df %>% \n",
+        "  select(-c(1, rice, garlic, ginger)) %>%\n",
+        "  # Encode cuisine column as categorical\n",
+        "  mutate(cuisine = factor(cuisine))\n",
+        "\n",
+        "\n",
+        "# Create data split specification\n",
+        "set.seed(2056)\n",
+        "cuisines_split <- initial_split(data = df_select,\n",
+        "                                strata = cuisine,\n",
+        "                                prop = 0.7)\n",
+        "\n",
+        "# Extract the data in each split\n",
+        "cuisines_train <- training(cuisines_split)\n",
+        "cuisines_test <- testing(cuisines_split)\n",
+        "\n",
+        "# Display distribution of cuisines in the training set\n",
+        "cuisines_train %>% \n",
+        "  count(cuisine) %>% \n",
+        "  arrange(desc(n))"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "zFin5yw3hHb1"
+      },
+      "source": [
+        "### Deal with imbalanced data\n",
+        "\n",
+        "Imbalanced data often has negative effects on the model performance. Many models perform best when the number of observations is equal and, thus, tend to struggle with unbalanced data.\n",
+        "\n",
+        "There are majorly two ways of dealing with imbalanced data sets:\n",
+        "\n",
+        "-   adding observations to the minority class: `Over-sampling` e.g using a SMOTE algorithm which synthetically generates new examples of the minority class using nearest neighbors of these cases.\n",
+        "\n",
+        "-   removing observations from majority class: `Under-sampling`\n",
+        "\n",
+        "In our previous lesson, we demonstrated how to deal with imbalanced data sets using a `recipe`. A recipe can be thought of as a blueprint that describes what steps should be applied to a data set in order to get it ready for data analysis. In our case, we want to have an equal distribution in the number of our cuisines for our `training set`. Let's get right into it."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "cRzTnHolhLWd"
+      },
+      "source": [
+        "# Load themis package for dealing with imbalanced data\n",
+        "library(themis)\n",
+        "\n",
+        "# Create a recipe for preprocessing training data\n",
+        "cuisines_recipe <- recipe(cuisine ~ ., data = cuisines_train) %>%\n",
+        "  step_smote(cuisine) \n",
+        "\n",
+        "# Print recipe\n",
+        "cuisines_recipe"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "KxOQ2ORhhO81"
+      },
+      "source": [
+        "Now we are ready to train models 👩‍💻👨‍💻!\n",
+        "\n",
+        "## 3. Beyond multinomial regression models\n",
+        "\n",
+        "In our previous lesson, we looked at multinomial regression models. Let's explore some more flexible models for classification.\n",
+        "\n",
+        "### Support Vector Machines.\n",
+        "\n",
+        "In the context of classification, `Support Vector Machines` is a machine learning technique that tries to find a *hyperplane* that \"best\" separates the classes. Let's look at a simple example:\n",
+        "\n",
+        "<p >\n",
+        "   <img src=\"../../images/svm.png\"\n",
+        "   width=\"300\"/>\n",
+        "   <figcaption>https://commons.wikimedia.org/w/index.php?curid=22877598</figcaption>\n",
+        "\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "C4Wsd0vZhXYu"
+      },
+      "source": [
+        "H1~ does not separate the classes. H2~ does, but only with a small margin. H3~ separates them with the maximal margin.\n",
+        "\n",
+        "#### Linear Support Vector Classifier\n",
+        "\n",
+        "Support-Vector clustering (SVC) is a child of the Support-Vector machines family of ML techniques. In SVC, the hyperplane is chosen to correctly separate `most` of the training observations, but `may misclassify` a few observations. By allowing some points to be on the wrong side, the SVM becomes more robust to outliers hence better generalization to new data. The parameter that regulates this violation is referred to as `cost` which has a default value of 1 (see `help(\"svm_poly\")`).\n",
+        "\n",
+        "Let's create a linear SVC by setting `degree = 1` in a polynomial SVM model."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "vJpp6nuChlBz"
+      },
+      "source": [
+        "# Make a linear SVC specification\n",
+        "svc_linear_spec <- svm_poly(degree = 1) %>% \n",
+        "  set_engine(\"kernlab\") %>% \n",
+        "  set_mode(\"classification\")\n",
+        "\n",
+        "# Bundle specification and recipe into a worklow\n",
+        "svc_linear_wf <- workflow() %>% \n",
+        "  add_recipe(cuisines_recipe) %>% \n",
+        "  add_model(svc_linear_spec)\n",
+        "\n",
+        "# Print out workflow\n",
+        "svc_linear_wf"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "rDs8cWNkhoqu"
+      },
+      "source": [
+        "Now that we have captured the preprocessing steps and model specification into a *workflow*, we can go ahead and train the linear SVC and evaluate results while at it. For performance metrics, let's create a metric set that will evaluate: `accuracy`, `sensitivity`, `Positive Predicted Value` and `F Measure`\n",
+        "\n",
+        "> `augment()` will add column(s) for predictions to the given data."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "81wiqcwuhrnq"
+      },
+      "source": [
+        "# Train a linear SVC model\n",
+        "svc_linear_fit <- svc_linear_wf %>% \n",
+        "  fit(data = cuisines_train)\n",
+        "\n",
+        "# Create a metric set\n",
+        "eval_metrics <- metric_set(ppv, sens, accuracy, f_meas)\n",
+        "\n",
+        "\n",
+        "# Make predictions and Evaluate model performance\n",
+        "svc_linear_fit %>% \n",
+        "  augment(new_data = cuisines_test) %>% \n",
+        "  eval_metrics(truth = cuisine, estimate = .pred_class)"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "0UFQvHf-huo3"
+      },
+      "source": [
+        "#### Support Vector Machine\n",
+        "\n",
+        "The support vector machine (SVM) is an extension of the support vector classifier in order to accommodate a non-linear boundary between the classes. In essence, SVMs use the *kernel trick* to enlarge the feature space to adapt to nonlinear relationships between classes. One popular and extremely flexible kernel function used by SVMs is the *Radial basis function.* Let's see how it will perform on our data."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "-KX4S8mzhzmp"
+      },
+      "source": [
+        "set.seed(2056)\n",
+        "\n",
+        "# Make an RBF SVM specification\n",
+        "svm_rbf_spec <- svm_rbf() %>% \n",
+        "  set_engine(\"kernlab\") %>% \n",
+        "  set_mode(\"classification\")\n",
+        "\n",
+        "# Bundle specification and recipe into a worklow\n",
+        "svm_rbf_wf <- workflow() %>% \n",
+        "  add_recipe(cuisines_recipe) %>% \n",
+        "  add_model(svm_rbf_spec)\n",
+        "\n",
+        "\n",
+        "# Train an RBF model\n",
+        "svm_rbf_fit <- svm_rbf_wf %>% \n",
+        "  fit(data = cuisines_train)\n",
+        "\n",
+        "\n",
+        "# Make predictions and Evaluate model performance\n",
+        "svm_rbf_fit %>% \n",
+        "  augment(new_data = cuisines_test) %>% \n",
+        "  eval_metrics(truth = cuisine, estimate = .pred_class)"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "QBFSa7WSh4HQ"
+      },
+      "source": [
+        "Much better 🤩!\n",
+        "\n",
+        "> ✅ Please see:\n",
+        ">\n",
+        "> -   [*Support Vector Machines*](https://bradleyboehmke.github.io/HOML/svm.html), Hands-on Machine Learning with R\n",
+        ">\n",
+        "> -   [*Support Vector Machines*](https://www.statlearning.com/), An Introduction to Statistical Learning with Applications in R\n",
+        ">\n",
+        "> for further reading.\n",
+        "\n",
+        "### Nearest Neighbor classifiers\n",
+        "\n",
+        "*K*-nearest neighbor (KNN) is an algorithm in which each observation is predicted based on its *similarity* to other observations.\n",
+        "\n",
+        "Let's fit one to our data."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "k4BxxBcdh9Ka"
+      },
+      "source": [
+        "# Make a KNN specification\n",
+        "knn_spec <- nearest_neighbor() %>% \n",
+        "  set_engine(\"kknn\") %>% \n",
+        "  set_mode(\"classification\")\n",
+        "\n",
+        "# Bundle recipe and model specification into a workflow\n",
+        "knn_wf <- workflow() %>% \n",
+        "  add_recipe(cuisines_recipe) %>% \n",
+        "  add_model(knn_spec)\n",
+        "\n",
+        "# Train a boosted tree model\n",
+        "knn_wf_fit <- knn_wf %>% \n",
+        "  fit(data = cuisines_train)\n",
+        "\n",
+        "\n",
+        "# Make predictions and Evaluate model performance\n",
+        "knn_wf_fit %>% \n",
+        "  augment(new_data = cuisines_test) %>% \n",
+        "  eval_metrics(truth = cuisine, estimate = .pred_class)"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "HaegQseriAcj"
+      },
+      "source": [
+        "It appears that this model is not performing that well. Probably changing the model's arguments (see `help(\"nearest_neighbor\")` will improve model performance. Be sure to try it out.\n",
+        "\n",
+        "> ✅ Please see:\n",
+        ">\n",
+        "> -   [Hands-on Machine Learning with R](https://bradleyboehmke.github.io/HOML/)\n",
+        ">\n",
+        "> -   [An Introduction to Statistical Learning with Applications in R](https://www.statlearning.com/)\n",
+        ">\n",
+        "> to learn more about *K*-Nearest Neighbors classifiers.\n",
+        "\n",
+        "### Ensemble classifiers\n",
+        "\n",
+        "Ensemble algorithms work by combining multiple base estimators to produce an optimal model either by:\n",
+        "\n",
+        "`bagging`: applying an *averaging function* to a collection of base models\n",
+        "\n",
+        "`boosting`: building a sequence of models that build on one another to improve predictive performance.\n",
+        "\n",
+        "Let's start by trying out a Random Forest model, which builds a large collection of decision trees then applies an averaging function to for a better overall model."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "49DPoVs6iK1M"
+      },
+      "source": [
+        "# Make a random forest specification\n",
+        "rf_spec <- rand_forest() %>% \n",
+        "  set_engine(\"ranger\") %>% \n",
+        "  set_mode(\"classification\")\n",
+        "\n",
+        "# Bundle recipe and model specification into a workflow\n",
+        "rf_wf <- workflow() %>% \n",
+        "  add_recipe(cuisines_recipe) %>% \n",
+        "  add_model(rf_spec)\n",
+        "\n",
+        "# Train a random forest model\n",
+        "rf_wf_fit <- rf_wf %>% \n",
+        "  fit(data = cuisines_train)\n",
+        "\n",
+        "\n",
+        "# Make predictions and Evaluate model performance\n",
+        "rf_wf_fit %>% \n",
+        "  augment(new_data = cuisines_test) %>% \n",
+        "  eval_metrics(truth = cuisine, estimate = .pred_class)"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "RGVYwC_aiUWc"
+      },
+      "source": [
+        "Good job 👏!\n",
+        "\n",
+        "Let's also experiment with a Boosted Tree model.\n",
+        "\n",
+        "Boosted Tree defines an ensemble method that creates a series of sequential decision trees where each tree depends on the results of previous trees in an attempt to incrementally reduce the error. It focuses on the weights of incorrectly classified items and adjusts the fit for the next classifier to correct.\n",
+        "\n",
+        "There are different ways to fit this model (see `help(\"boost_tree\")`). In this example, we'll fit Boosted trees via `xgboost` engine."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "Py1YWo-micWs"
+      },
+      "source": [
+        "# Make a boosted tree specification\n",
+        "boost_spec <- boost_tree(trees = 200) %>% \n",
+        "  set_engine(\"xgboost\") %>% \n",
+        "  set_mode(\"classification\")\n",
+        "\n",
+        "# Bundle recipe and model specification into a workflow\n",
+        "boost_wf <- workflow() %>% \n",
+        "  add_recipe(cuisines_recipe) %>% \n",
+        "  add_model(boost_spec)\n",
+        "\n",
+        "# Train a boosted tree model\n",
+        "boost_wf_fit <- boost_wf %>% \n",
+        "  fit(data = cuisines_train)\n",
+        "\n",
+        "\n",
+        "# Make predictions and Evaluate model performance\n",
+        "boost_wf_fit %>% \n",
+        "  augment(new_data = cuisines_test) %>% \n",
+        "  eval_metrics(truth = cuisine, estimate = .pred_class)"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "zNQnbuejigZM"
+      },
+      "source": [
+        "\n",
+        "> ✅ Please see:\n",
+        ">\n",
+        "> -   [Machine Learning for Social Scientists](https://cimentadaj.github.io/ml_socsci/tree-based-methods.html#random-forests)\n",
+        ">\n",
+        "> -   [Hands-on Machine Learning with R](https://bradleyboehmke.github.io/HOML/)\n",
+        ">\n",
+        "> -   [An Introduction to Statistical Learning with Applications in R](https://www.statlearning.com/)\n",
+        ">\n",
+        "> -   <https://algotech.netlify.app/blog/xgboost/> - Explores the AdaBoost model which is a good alternative to xgboost.\n",
+        ">\n",
+        "> to learn more about Ensemble classifiers.\n",
+        "\n",
+        "## 4. Extra - comparing multiple models\n",
+        "\n",
+        "We have fitted quite a number of models in this lab 🙌. It can become tedious or onerous to create a lot of workflows from different sets of preprocessors and/or model specifications and then calculate the performance metrics one by one.\n",
+        "\n",
+        "Let's see if we can address this by creating a function that fits a list of workflows on the training set then returns the performance metrics based on the test set. We'll get to use `map()` and `map_dfr()` from the [purrr](https://purrr.tidyverse.org/) package to apply functions to each element in list.\n",
+        "\n",
+        "> [`map()`](https://purrr.tidyverse.org/reference/map.html) functions allow you to replace many for loops with code that is both more succinct and easier to read. The best place to learn about the [`map()`](https://purrr.tidyverse.org/reference/map.html) functions is the [iteration chapter](http://r4ds.had.co.nz/iteration.html) in R for data science."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "Qzb7LyZnimd2"
+      },
+      "source": [
+        "set.seed(2056)\n",
+        "\n",
+        "# Create a metric set\n",
+        "eval_metrics <- metric_set(ppv, sens, accuracy, f_meas)\n",
+        "\n",
+        "# Define a function that returns performance metrics\n",
+        "compare_models <- function(workflow_list, train_set, test_set){\n",
+        "  \n",
+        "  suppressWarnings(\n",
+        "    # Fit each model to the train_set\n",
+        "    map(workflow_list, fit, data = train_set) %>% \n",
+        "    # Make predictions on the test set\n",
+        "      map_dfr(augment, new_data = test_set, .id = \"model\") %>%\n",
+        "    # Select desired columns\n",
+        "      select(model, cuisine, .pred_class) %>% \n",
+        "    # Evaluate model performance\n",
+        "      group_by(model) %>% \n",
+        "      eval_metrics(truth = cuisine, estimate = .pred_class) %>% \n",
+        "      ungroup()\n",
+        "  )\n",
+        "  \n",
+        "} # End of function"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "Fwa712sNisDA"
+      },
+      "source": [
+        "Let's call our function and compare the accuracy across the models."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "3i4VJOi2iu-a"
+      },
+      "source": [
+        "# Make a list of workflows\n",
+        "workflow_list <- list(\n",
+        "  \"svc\" = svc_linear_wf,\n",
+        "  \"svm\" = svm_rbf_wf,\n",
+        "  \"knn\" = knn_wf,\n",
+        "  \"random_forest\" = rf_wf,\n",
+        "  \"xgboost\" = boost_wf)\n",
+        "\n",
+        "# Call the function\n",
+        "set.seed(2056)\n",
+        "perf_metrics <- compare_models(workflow_list = workflow_list, train_set = cuisines_train, test_set = cuisines_test)\n",
+        "\n",
+        "# Print out performance metrics\n",
+        "perf_metrics %>% \n",
+        "  group_by(.metric) %>% \n",
+        "  arrange(desc(.estimate)) %>% \n",
+        "  slice_head(n=7)\n",
+        "\n",
+        "# Compare accuracy\n",
+        "perf_metrics %>% \n",
+        "  filter(.metric == \"accuracy\") %>% \n",
+        "  arrange(desc(.estimate))\n"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "KuWK_lEli4nW"
+      },
+      "source": [
+        "\n",
+        "[**workflowset**](https://workflowsets.tidymodels.org/) package allow users to create and easily fit a large number of models but is mostly designed to work with resampling techniques such as `cross-validation`, an approach we are yet to cover.\n",
+        "\n",
+        "## **🚀Challenge**\n",
+        "\n",
+        "Each of these techniques has a large number of parameters that you can tweak for instance `cost` in SVMs, `neighbors` in KNN, `mtry` (Randomly Selected Predictors) in Random Forest.\n",
+        "\n",
+        "Research each one's default parameters and think about what tweaking these parameters would mean for the model's quality.\n",
+        "\n",
+        "To find out more about a particular model and its parameters, use: `help(\"model\")` e.g `help(\"rand_forest\")`\n",
+        "\n",
+        "> In practice, we usually *estimate* the *best values* for these by training many models on a `simulated data set` and measuring how well all these models perform. This process is called **tuning**.\n",
+        "\n",
+        "### [**Post-lecture quiz**](https://white-water-09ec41f0f.azurestaticapps.net/quiz/24/)\n",
+        "\n",
+        "### **Review & Self Study**\n",
+        "\n",
+        "There's a lot of jargon in these lessons, so take a minute to review [this list](https://docs.microsoft.com/dotnet/machine-learning/resources/glossary?WT.mc_id=academic-15963-cxa) of useful terminology!\n",
+        "\n",
+        "#### THANK YOU TO:\n",
+        "\n",
+        "[`Allison Horst`](https://twitter.com/allison_horst/) for creating the amazing illustrations that make R more welcoming and engaging. Find more illustrations at her [gallery](https://www.google.com/url?q=https://github.com/allisonhorst/stats-illustrations&sa=D&source=editors&ust=1626380772530000&usg=AOvVaw3zcfyCizFQZpkSLzxiiQEM).\n",
+        "\n",
+        "[Cassie Breviu](https://www.twitter.com/cassieview) and [Jen Looper](https://www.twitter.com/jenlooper) for creating the original Python version of this module ♥️\n",
+        "\n",
+        "Happy Learning,\n",
+        "\n",
+        "[Eric](https://twitter.com/ericntay), Gold Microsoft Learn Student Ambassador.\n",
+        "\n",
+        "<p >\n",
+        "   <img src=\"../../images/r_learners_sm.jpeg\"\n",
+        "   width=\"569\"/>\n",
+        "   <figcaption>Artwork by @allison_horst</figcaption>\n"
+      ]
+    }
+  ]
+}
\ No newline at end of file
diff --git a/4-Classification/3-Classifiers-2/solution/R/lesson_12.Rmd b/4-Classification/3-Classifiers-2/solution/R/lesson_12.Rmd
new file mode 100644
index 000000000..3a6f6ba43
--- /dev/null
+++ b/4-Classification/3-Classifiers-2/solution/R/lesson_12.Rmd
@@ -0,0 +1,452 @@
+---
+title: 'Build a classification model: Delicious Asian and Indian Cuisines'
+output:
+  html_document:
+    df_print: paged
+    theme: flatly
+    highlight: breezedark
+    toc: yes
+    toc_float: yes
+    code_download: yes
+---
+
+## Cuisine classifiers 2
+
+In this second classification lesson, we will explore `more ways` to classify categorical data. We will also learn about the ramifications for choosing one classifier over the other.
+
+### [**Pre-lecture quiz**](https://white-water-09ec41f0f.azurestaticapps.net/quiz/23/)
+
+### **Prerequisite**
+
+We assume that you have completed the previous lessons since we will be carrying forward some concepts we learned before.
+
+For this lesson, we'll require the following packages:
+
+-   `tidyverse`: The [tidyverse](https://www.tidyverse.org/) is a [collection of R packages](https://www.tidyverse.org/packages) designed to makes data science faster, easier and more fun!
+
+-   `tidymodels`: The [tidymodels](https://www.tidymodels.org/) framework is a [collection of packages](https://www.tidymodels.org/packages/) for modeling and machine learning.
+
+-   `themis`: The [themis package](https://themis.tidymodels.org/) provides Extra Recipes Steps for Dealing with Unbalanced Data.
+
+You can have them installed as:
+
+`install.packages(c("tidyverse", "tidymodels", "kernlab", "themis", "ranger", "xgboost", "kknn"))`
+
+Alternatively, the script below checks whether you have the packages required to complete this module and installs them for you in case they are missing.
+
+```{r, message=F, warning=F}
+suppressWarnings(if (!require("pacman"))install.packages("pacman"))
+
+pacman::p_load(tidyverse, tidymodels, themis, kernlab, ranger, xgboost, kknn)
+```
+
+Now, let's hit the ground running!
+
+## **1. A classification map**
+
+In our [previous lesson](https://github.com/microsoft/ML-For-Beginners/tree/main/4-Classification/2-Classifiers-1), we tried to address the question: how do we choose between multiple models? To a great extent, it depends on the characteristics of the data and the type of problem we want to solve (for instance classification or regression?)
+
+Previously, we learned about the various options you have when classifying data using Microsoft's cheat sheet. Python's Machine Learning framework, Scikit-learn, offers a similar but more granular cheat sheet that can further help narrow down your estimators (another term for classifiers):
+
+![](../../images/map.png){width="650"}\
+
+> Tip: [visit this map online](https://scikit-learn.org/stable/tutorial/machine_learning_map/) and click along the path to read documentation.
+>
+> The [Tidymodels reference site](https://www.tidymodels.org/find/parsnip/#models) also provides an excellent documentation about different types of model.
+
+### **The plan** 🗺️
+
+This map is very helpful once you have a clear grasp of your data, as you can 'walk' along its paths to a decision:
+
+-   We have \>50 samples
+
+-   We want to predict a category
+
+-   We have labeled data
+
+-   We have fewer than 100K samples
+
+-   ✨ We can choose a Linear SVC
+
+-   If that doesn't work, since we have numeric data
+
+    -   We can try a ✨ KNeighbors Classifier
+
+        -   If that doesn't work, try ✨ SVC and ✨ Ensemble Classifiers
+
+This is a very helpful trail to follow. Now, let's get right into it using the [tidymodels](https://www.tidymodels.org/) modelling framework: a consistent and flexible collection of R packages developed to encourage good statistical practice 😊.
+
+## 2. Split the data and deal with imbalanced data set.
+
+From our previous lessons, we learnt that there were a set of common ingredients across our cuisines. Also, there was quite an unequal distribution in the number of cuisines.
+
+We'll deal with these by
+
+-   Dropping the most common ingredients that create confusion between distinct cuisines, using `dplyr::select()`.
+
+-   Use a `recipe` that preprocesses the data to get it ready for modelling by applying an `over-sampling` algorithm.
+
+We already looked at the above in the previous lesson so this should be a breeze 🥳!
+
+```{r clean_imbalance}
+# Load the core Tidyverse and Tidymodels packages
+library(tidyverse)
+library(tidymodels)
+
+# Load the original cuisines data
+df <- read_csv(file = "https://raw.githubusercontent.com/microsoft/ML-For-Beginners/main/4-Classification/data/cuisines.csv")
+
+# Drop id column, rice, garlic and ginger from our original data set
+df_select <- df %>% 
+  select(-c(1, rice, garlic, ginger)) %>%
+  # Encode cuisine column as categorical
+  mutate(cuisine = factor(cuisine))
+
+
+# Create data split specification
+set.seed(2056)
+cuisines_split <- initial_split(data = df_select,
+                                strata = cuisine,
+                                prop = 0.7)
+
+# Extract the data in each split
+cuisines_train <- training(cuisines_split)
+cuisines_test <- testing(cuisines_split)
+
+# Display distribution of cuisines in the training set
+cuisines_train %>% 
+  count(cuisine) %>% 
+  arrange(desc(n))
+
+
+```
+
+### Deal with imbalanced data
+
+Imbalanced data often has negative effects on the model performance. Many models perform best when the number of observations is equal and, thus, tend to struggle with unbalanced data.
+
+There are majorly two ways of dealing with imbalanced data sets:
+
+-   adding observations to the minority class: `Over-sampling` e.g using a SMOTE algorithm which synthetically generates new examples of the minority class using nearest neighbors of these cases.
+
+-   removing observations from majority class: `Under-sampling`
+
+In our previous lesson, we demonstrated how to deal with imbalanced data sets using a `recipe`. A recipe can be thought of as a blueprint that describes what steps should be applied to a data set in order to get it ready for data analysis. In our case, we want to have an equal distribution in the number of our cuisines for our `training set`. Let's get right into it.
+
+```{r recap_balance}
+# Load themis package for dealing with imbalanced data
+library(themis)
+
+# Create a recipe for preprocessing training data
+cuisines_recipe <- recipe(cuisine ~ ., data = cuisines_train) %>%
+  step_smote(cuisine) 
+
+# Print recipe
+cuisines_recipe
+
+```
+
+Now we are ready to train models 👩‍💻👨‍💻!
+
+## 3. Beyond multinomial regression models
+
+In our previous lesson, we looked at multinomial regression models. Let's explore some more flexible models for classification.
+
+### Support Vector Machines.
+
+In the context of classification, `Support Vector Machines` is a machine learning technique that tries to find a *hyperplane* that "best" separates the classes. Let's look at a simple example:
+
+![By User:ZackWeinberg:This file was derived from: <https://commons.wikimedia.org/w/index.php?curid=22877598>](../../images/svm.png){width="300"}
+
+H~1~ does not separate the classes. H~2~ does, but only with a small margin. H~3~ separates them with the maximal margin.
+
+#### Linear Support Vector Classifier
+
+Support-Vector clustering (SVC) is a child of the Support-Vector machines family of ML techniques. In SVC, the hyperplane is chosen to correctly separate `most` of the training observations, but `may misclassify` a few observations. By allowing some points to be on the wrong side, the SVM becomes more robust to outliers hence better generalization to new data. The parameter that regulates this violation is referred to as `cost` which has a default value of 1 (see `help("svm_poly")`).
+
+Let's create a linear SVC by setting `degree = 1` in a polynomial SVM model.
+
+```{r svc_spec}
+# Make a linear SVC specification
+svc_linear_spec <- svm_poly(degree = 1) %>% 
+  set_engine("kernlab") %>% 
+  set_mode("classification")
+
+# Bundle specification and recipe into a worklow
+svc_linear_wf <- workflow() %>% 
+  add_recipe(cuisines_recipe) %>% 
+  add_model(svc_linear_spec)
+
+# Print out workflow
+svc_linear_wf
+```
+
+Now that we have captured the preprocessing steps and model specification into a *workflow*, we can go ahead and train the linear SVC and evaluate results while at it. For performance metrics, let's create a metric set that will evaluate: `accuracy`, `sensitivity`, `Positive Predicted Value` and `F Measure`
+
+> `augment()` will add column(s) for predictions to the given data.
+
+```{r svc_train}
+# Train a linear SVC model
+svc_linear_fit <- svc_linear_wf %>% 
+  fit(data = cuisines_train)
+
+# Create a metric set
+eval_metrics <- metric_set(ppv, sens, accuracy, f_meas)
+
+
+# Make predictions and Evaluate model performance
+svc_linear_fit %>% 
+  augment(new_data = cuisines_test) %>% 
+  eval_metrics(truth = cuisine, estimate = .pred_class)
+  
+
+
+```
+
+#### 
+
+#### Support Vector Machine
+
+The support vector machine (SVM) is an extension of the support vector classifier in order to accommodate a non-linear boundary between the classes. In essence, SVMs use the *kernel trick* to enlarge the feature space to adapt to nonlinear relationships between classes. One popular and extremely flexible kernel function used by SVMs is the *Radial basis function.* Let's see how it will perform on our data.
+
+```{r svm_rbf}
+set.seed(2056)
+
+# Make an RBF SVM specification
+svm_rbf_spec <- svm_rbf() %>% 
+  set_engine("kernlab") %>% 
+  set_mode("classification")
+
+# Bundle specification and recipe into a worklow
+svm_rbf_wf <- workflow() %>% 
+  add_recipe(cuisines_recipe) %>% 
+  add_model(svm_rbf_spec)
+
+
+# Train an RBF model
+svm_rbf_fit <- svm_rbf_wf %>% 
+  fit(data = cuisines_train)
+
+
+# Make predictions and Evaluate model performance
+svm_rbf_fit %>% 
+  augment(new_data = cuisines_test) %>% 
+  eval_metrics(truth = cuisine, estimate = .pred_class)
+```
+
+Much better 🤩!
+
+> ✅ Please see:
+>
+> -   [*Support Vector Machines*](https://bradleyboehmke.github.io/HOML/svm.html), Hands-on Machine Learning with R
+>
+> -   [*Support Vector Machines*](https://www.statlearning.com/), An Introduction to Statistical Learning with Applications in R
+>
+> for further reading.
+
+### Nearest Neighbor classifiers
+
+*K*-nearest neighbor (KNN) is an algorithm in which each observation is predicted based on its *similarity* to other observations.
+
+Let's fit one to our data.
+
+```{r knn}
+# Make a KNN specification
+knn_spec <- nearest_neighbor() %>% 
+  set_engine("kknn") %>% 
+  set_mode("classification")
+
+# Bundle recipe and model specification into a workflow
+knn_wf <- workflow() %>% 
+  add_recipe(cuisines_recipe) %>% 
+  add_model(knn_spec)
+
+# Train a boosted tree model
+knn_wf_fit <- knn_wf %>% 
+  fit(data = cuisines_train)
+
+
+# Make predictions and Evaluate model performance
+knn_wf_fit %>% 
+  augment(new_data = cuisines_test) %>% 
+  eval_metrics(truth = cuisine, estimate = .pred_class)
+```
+
+It appears that this model is not performing that well. Probably changing the model's arguments (see `help("nearest_neighbor")` will improve model performance. Be sure to try it out.
+
+> ✅ Please see:
+>
+> -   [Hands-on Machine Learning with R](https://bradleyboehmke.github.io/HOML/)
+>
+> -   [An Introduction to Statistical Learning with Applications in R](https://www.statlearning.com/)
+>
+> to learn more about *K*-Nearest Neighbors classifiers.
+
+### Ensemble classifiers
+
+Ensemble algorithms work by combining multiple base estimators to produce an optimal model either by:
+
+`bagging`: applying an *averaging function* to a collection of base models
+
+`boosting`: building a sequence of models that build on one another to improve predictive performance.
+
+Let's start by trying out a Random Forest model, which builds a large collection of decision trees then applies an averaging function to for a better overall model.
+
+```{r rf}
+# Make a random forest specification
+rf_spec <- rand_forest() %>% 
+  set_engine("ranger") %>% 
+  set_mode("classification")
+
+# Bundle recipe and model specification into a workflow
+rf_wf <- workflow() %>% 
+  add_recipe(cuisines_recipe) %>% 
+  add_model(rf_spec)
+
+# Train a random forest model
+rf_wf_fit <- rf_wf %>% 
+  fit(data = cuisines_train)
+
+
+# Make predictions and Evaluate model performance
+rf_wf_fit %>% 
+  augment(new_data = cuisines_test) %>% 
+  eval_metrics(truth = cuisine, estimate = .pred_class)
+  
+
+```
+
+Good job 👏!
+
+Let's also experiment with a Boosted Tree model.
+
+Boosted Tree defines an ensemble method that creates a series of sequential decision trees where each tree depends on the results of previous trees in an attempt to incrementally reduce the error. It focuses on the weights of incorrectly classified items and adjusts the fit for the next classifier to correct.
+
+There are different ways to fit this model (see `help("boost_tree")`). In this example, we'll fit Boosted trees via `xgboost` engine.
+
+```{r boosted_tree}
+# Make a boosted tree specification
+boost_spec <- boost_tree(trees = 200) %>% 
+  set_engine("xgboost") %>% 
+  set_mode("classification")
+
+# Bundle recipe and model specification into a workflow
+boost_wf <- workflow() %>% 
+  add_recipe(cuisines_recipe) %>% 
+  add_model(boost_spec)
+
+# Train a boosted tree model
+boost_wf_fit <- boost_wf %>% 
+  fit(data = cuisines_train)
+
+
+# Make predictions and Evaluate model performance
+boost_wf_fit %>% 
+  augment(new_data = cuisines_test) %>% 
+  eval_metrics(truth = cuisine, estimate = .pred_class)
+```
+
+> ✅ Please see:
+>
+> -   [Machine Learning for Social Scientists](https://cimentadaj.github.io/ml_socsci/tree-based-methods.html#random-forests)
+>
+> -   [Hands-on Machine Learning with R](https://bradleyboehmke.github.io/HOML/)
+>
+> -   [An Introduction to Statistical Learning with Applications in R](https://www.statlearning.com/)
+>
+> -   <https://algotech.netlify.app/blog/xgboost/> - Explores the AdaBoost model which is a good alternative to xgboost.
+>
+> to learn more about Ensemble classifiers.
+
+## 4. Extra - comparing multiple models
+
+We have fitted quite a number of models in this lab 🙌. It can become tedious or onerous to create a lot of workflows from different sets of preprocessors and/or model specifications and then calculate the performance metrics one by one.
+
+Let's see if we can address this by creating a function that fits a list of workflows on the training set then returns the performance metrics based on the test set. We'll get to use `map()` and `map_dfr()` from the [purrr](https://purrr.tidyverse.org/) package to apply functions to each element in list.
+
+> [`map()`](https://purrr.tidyverse.org/reference/map.html) functions allow you to replace many for loops with code that is both more succinct and easier to read. The best place to learn about the [`map()`](https://purrr.tidyverse.org/reference/map.html) functions is the [iteration chapter](http://r4ds.had.co.nz/iteration.html) in R for data science.
+
+```{r compare_models}
+set.seed(2056)
+
+# Create a metric set
+eval_metrics <- metric_set(ppv, sens, accuracy, f_meas)
+
+# Define a function that returns performance metrics
+compare_models <- function(workflow_list, train_set, test_set){
+  
+  suppressWarnings(
+    # Fit each model to the train_set
+    map(workflow_list, fit, data = train_set) %>% 
+    # Make predictions on the test set
+      map_dfr(augment, new_data = test_set, .id = "model") %>%
+    # Select desired columns
+      select(model, cuisine, .pred_class) %>% 
+    # Evaluate model performance
+      group_by(model) %>% 
+      eval_metrics(truth = cuisine, estimate = .pred_class) %>% 
+      ungroup()
+  )
+  
+} # End of function
+
+
+```
+
+Let's call our function and compare the accuracy across the models.
+
+```{r call_fn}
+# Make a list of workflows
+workflow_list <- list(
+  "svc" = svc_linear_wf,
+  "svm" = svm_rbf_wf,
+  "knn" = knn_wf,
+  "random_forest" = rf_wf,
+  "xgboost" = boost_wf)
+
+# Call the function
+set.seed(2056)
+perf_metrics <- compare_models(workflow_list = workflow_list, train_set = cuisines_train, test_set = cuisines_test)
+
+# Print out performance metrics
+perf_metrics %>% 
+  group_by(.metric) %>% 
+  arrange(desc(.estimate)) %>% 
+  slice_head(n=7)
+
+# Compare accuracy
+perf_metrics %>% 
+  filter(.metric == "accuracy") %>% 
+  arrange(desc(.estimate))
+
+```
+
+[**workflowset**](https://workflowsets.tidymodels.org/) package allow users to create and easily fit a large number of models but is mostly designed to work with resampling techniques such as `cross-validation`, an approach we are yet to cover.
+
+## **🚀Challenge**
+
+Each of these techniques has a large number of parameters that you can tweak for instance `cost` in SVMs, `neighbors` in KNN, `mtry` (Randomly Selected Predictors) in Random Forest.
+
+Research each one's default parameters and think about what tweaking these parameters would mean for the model's quality.
+
+To find out more about a particular model and its parameters, use: `help("model")` e.g `help("rand_forest")`
+
+> In practice, we usually *estimate* the *best values* for these by training many models on a `simulated data set` and measuring how well all these models perform. This process is called **tuning**.
+
+### [**Post-lecture quiz**](https://white-water-09ec41f0f.azurestaticapps.net/quiz/24/)
+
+### **Review & Self Study**
+
+There's a lot of jargon in these lessons, so take a minute to review [this list](https://docs.microsoft.com/dotnet/machine-learning/resources/glossary?WT.mc_id=academic-15963-cxa) of useful terminology!
+
+#### THANK YOU TO:
+
+[`Allison Horst`](https://twitter.com/allison_horst/) for creating the amazing illustrations that make R more welcoming and engaging. Find more illustrations at her [gallery](https://www.google.com/url?q=https://github.com/allisonhorst/stats-illustrations&sa=D&source=editors&ust=1626380772530000&usg=AOvVaw3zcfyCizFQZpkSLzxiiQEM).
+
+[Cassie Breviu](https://www.twitter.com/cassieview) and [Jen Looper](https://www.twitter.com/jenlooper) for creating the original Python version of this module ♥️
+
+Happy Learning,
+
+[Eric](https://twitter.com/ericntay), Gold Microsoft Learn Student Ambassador.
+
+![Artwork by \@allison_horst](../../images/r_learners_sm.jpeg)
diff --git a/README.md b/README.md
index 148ddd824..50fe2399d 100644
--- a/README.md
+++ b/README.md
@@ -86,9 +86,9 @@ By ensuring that the content aligns with projects, the process is made more enga
 |       07      |               North American pumpkin prices 🎃              |         [Regression](2-Regression/README.md)        | Build linear and polynomial regression models                                                                                   |     <ul><li>[Python](2-Regression/3-Linear/README.md)</li><li>[R](2-Regression/3-Linear/solution/R/lesson_3-R.ipynb)</li></ul>       |       <ul><li>Jen</li><li>Eric Wanjau</li></ul>     |
 |       08      |               North American pumpkin prices 🎃              |         [Regression](2-Regression/README.md)        | Build a logistic regression model                                                                                               |     <ul><li>[Python](2-Regression/4-Logistic/README.md)  </li><li>[R](2-Regression/4-Logistic/solution/R/lesson_4-R.ipynb)</li></ul>     |       <ul><li>Jen</li><li>Eric Wanjau</li></ul>     |
 |       09      |                         A Web App 🔌                        |            [Web App](3-Web-App/README.md)           | Build a web app to use your trained model                                                                                       |        [Python](3-Web-App/1-Web-App/README.md)       |       Jen      |
-|       10      |               Introduction to classification               |     [Classification](4-Classification/README.md)    | Clean, prep, and visualize your data; introduction to classification                                                            |  [Python](4-Classification/1-Introduction/README.md)  | Jen and Cassie |
-|       11      |            Delicious Asian and Indian cuisines 🍜           |     [Classification](4-Classification/README.md)    | Introduction to classifiers                                                                                                     |  [Python](4-Classification/2-Classifiers-1/README.md)  | Jen and Cassie |
-|       12      |            Delicious Asian and Indian cuisines 🍜           |     [Classification](4-Classification/README.md)    | More classifiers                                                                                                                |  [Python](4-Classification/3-Classifiers-2/README.md)  | Jen and Cassie |
+|       10     |               Introduction to classification               |     [Classification](4-Classification/README.md)    | Clean, prep, and visualize your data; introduction to classification                                                            |<ul><li>  [Python](4-Classification/1-Introduction/README.md) </li><li>[R](4-Classification/1-Introduction/solution/R/lesson_10-R.ipynb)  | <ul><li>Jen and Cassie</li><li>Eric Wanjau</li></ul> |
+|       11      |            Delicious Asian and Indian cuisines 🍜           |     [Classification](4-Classification/README.md)    | Introduction to classifiers                                                                                                     |<ul><li> [Python](4-Classification/2-Classifiers-1/README.md)</li><li>[R](4-Classification/2-Classifiers-1/solution/R/lesson_11-R.ipynb) | <ul><li>Jen and Cassie</li><li>Eric Wanjau</li></ul> |
+|       12      |            Delicious Asian and Indian cuisines 🍜           |     [Classification](4-Classification/README.md)    | More classifiers                                                                                                                |<ul><li>  [Python](4-Classification/3-Classifiers-2/README.md)</li><li>[R](4-Classification/3-Classifiers-2/solution/R/lesson_12-R.ipynb)  | <ul><li>Jen and Cassie</li><li>Eric Wanjau</li></ul> |
 |       13      |            Delicious Asian and Indian cuisines 🍜           |     [Classification](4-Classification/README.md)    | Build a recommender web app using your model                                                                                    |     [Python](4-Classification/4-Applied/README.md)     |       Jen      |
 |       14      |                 Introduction to clustering                 |         [Clustering](5-Clustering/README.md)        | Clean, prep, and visualize your data; Introduction to clustering                                                                |      [Python](5-Clustering/1-Visualize/README.md)     |       Jen      |
 |       15      |             Exploring Nigerian Musical Tastes 🎧            |         [Clustering](5-Clustering/README.md)        | Explore the K-Means clustering method                                                                                           |       [Python](5-Clustering/2-K-Means/README.md)       |       Jen      |