From e065ee0acd9fd41fc80c2bc67b90638e9290289b Mon Sep 17 00:00:00 2001
From: Garrett <grolemund@gmail.com>
Date: Thu, 7 Jan 2016 13:25:25 -0500
Subject: [PATCH 1/2] Updates data set for model chapter to BLS NLS
 cross-section.

---
 data/heights.RDS | Bin 0 -> 52843 bytes
 model.Rmd        | 149 +++++++++++++++++++++++++----------------------
 2 files changed, 79 insertions(+), 70 deletions(-)
 create mode 100644 data/heights.RDS

diff --git a/data/heights.RDS b/data/heights.RDS
new file mode 100644
index 0000000000000000000000000000000000000000..d0df4bfbb6124c50f9b350162afe3d5f1c4ed847
GIT binary patch
literal 52843
zcmYg$2|Sd0`~ECu#x|&qh=~?*>=GHWoL)&%B*wms<j4{usj<(bXt5Mo$}(?Jc1BL9
zEYDDcqA)1pi6Z-0hOzd4p7fsoug~+D>wDk#bzk>&f2XEtk-fu@`1J!lU5m?<C#4v<
zmaTm!yj%<3zeh{AL+tMuE$ifkQ)B)v&L&6Q<rVWKKcsqM4nErR_|JqrSm8bUuF)@I
z#L>g&5aQc}t8XJml(3sa)?=A=RsOZ1+|a+zM6J0QG9IszzPwx>2&!KWYp5k1O1?_S
zYaX8-4-5^R^9n%!%<+|PE^v+sYIrqEG$SYyya~DljM_+dpfGr0Sn_qba)QE0AG$f4
zxYSUtlMq3fy5nI)z=1@*W_RLF(EjEKg<uL2w>Mi6cYd{x^AJeL2_!_1*b|yZ>~DMQ
zN=PC0fPFNN{Rt(+CQ95$`s;%?Jh08y6n*jLdXQ&FP#7i(G`A2fSb<6jW5Yyz^Ej~`
znA0b8GzW?#NPyIB2|<B1lFjCn8bY%&<=rq(lu*_bcrc-Ngo;iur`+9CYYvo8s2ZW-
z5{f7_JDS<VQr4V5p?8=nny@@ljchih4BQ;?Ao@4W1rn-;d2$JsUwM{<Afn-&5$Gjx
zM2I-^dd`I4)-dNuaAVDB6Wp5TP7{V20@V}NDB%*#Xrj#PIk5!Uue^f^v4QJ435;Rh
zt^~#i50l{bmHJ0>S>QUhd6u#+(=1C_7fmo4<{e6?KX`sSaBANjw7nU^QUH{<dTrHY
z_2y$Ny4p<mYagaLnSwZS>kDNn0WEoQiVnR9k04^U17}c@ClSWCHUbGZNA?hKBZ!3U
zfF-KW454uAJJA+I9N%6P2%Vo3Y00k-5HbS&SjcyB$0>2ch%4Ix3z(=Q03JXN$k86W
z?h4f3(Odz|5g;qf`4Ku;NMZR+JWk_2#Pu2`3gI}(L}@sL=31h7W(XqzP$AGTA*y*!
zi6Ajt&5xtphvt{D<{SxSjmT5-gV)j2lg{*!5+bdDk8>Z9y1k!*Fc%;zA_?+?cSa(J
zw7l2BL_-?7`Xs_m092Z@A^<kSJK>iuqJJaOS)NwJM@gyPjiaO>o^A(hQ0XTTmbYF}
z!o{1v5NRc^qrO(-a2lyRtO!{J87EIG=hH~7mTg|9AbbVLs>ovbNZLp&k@kX5BMs5C
z9k567P9mHHKt-gA{N}ZhGNKF)gS0UK8QH=jU?DDj0+}5O95n!JL^asfjYz`@qy(3C
zCzMtsLoUGCPaso6X^&*c`8eJ<(jO+q4kLapB90h<*)m`rZlDW2IF3vRrRB(g`8eW)
z!wvMWo}RDviD;;Jii<Ee0IWngUFh{iT3G1QK!m9Qn3bLPm5LRiSc+1*(ZXDWF|0<;
z!}X0LuY}UlvL%KQwxP6??6hI(u7m{&;*bHDDMPX7P!5C!1E)nPUFd;DgpL7VF3Rad
zZ;m7X4y8TF))_&Fa%(WSU$bhU_plr7IsthD%nfE6kXuJH*bFpueKB0DIhNv_h$c=T
zV?zBKkO>n=Q!Z@o>DJ;a(DlF@ZDpY@Y5`6X=5AdFWYvfZyRi^ub}d{a3_(9};6(t-
zj^#jv(N-QkwMZ)q1xN;9;nwva?cn=>qk^$2B2<8`o%8LJE?6LOxI|*9SR#D_^E8xR
z;7Ts2x(40kbTi}&lO0OWbES~fHP}Mg6BtV_%BdExRcHQ+N<&%ZwE#(-!WOcaxZyI1
zsSKr;xRQ$``hG=qq3*r#L$4Sdp#Y`ahEguds}^uj=M_p2Co%0%!4)i)7+?#vPhfgN
zSq`;;(=49|xm1EWi5U!~R}Jo<AcAVa5?8Q9qJ}M0H}Na$O0iLA_6Ye;V05@Bw_3nX
zox>Ii>k)G1qO5BH8};gZXrJYHqS_neDK5&g7O+-lb_=O+QKq%vV^?y%gyaN97LEWb
zbz*^p{RAc^l+JJk^CYUrH~2)rrx0TNL+KA($xkKH$7|6zKSy<qE}@xU*$me$!q_58
zrxwg~CBtnP*HC(zEBHjhxl1T>5v5QIn5lC<uJMbIb0w;OU=D=3H-^&hxb`t!`^pBJ
zDTw{GVAf!Ld0R4{MH17~jAFUn-VCK|MwS-{OXago0<+{alK74%0jArlPoN~sz<mPF
za`ZcLR0V-Svj$gr_o5`=`1bS@C?PZQ9syywgPUf&!&^r>T~6aJA2td2YrEu$YaU6G
z%W_Qyw*oEC4J5-yeF3MG*XareN6q?*;KD)6219vwMiQ6?)o#sMG^Ckbb4p?wn{mvy
zuP43cDL~S3QP7dEUJj<E%S~BtCnWR9Bmw`~enn9r+YGhCjJ!|alpO1z8Ms$K#tels
z>&um6-sRf~Q|y0v+$*3W$5Mpr7BIgrM{de$d!)tIa-0*}4=Zl%r*V((0Nhpir+~g3
z{e~P-PT<IWICX40^#n@P45I2M@de1Sq|N%EE4s($CCA!jM&2t>cVcrpj<{c-{RB$d
zjJ$tK&Cd$Hq8(-w(<BNq>^@5EEMOwI#|$Z<NZyCpJn=td?R*04RRs5A!gC=c%V9*!
zSVLxX)3boNAk_@1t_aFw+!QbjU5t?e#?Xvzc@{V=xTJ(}HA5azq?{BCPeLW;0{ay~
z1*j8g1mk)xx#}!nDOjn5naTyE6?yuK<Uiq&ri$R77={8y<q7S{AAYuigd|kzSx(ql
z@ZKMs``6x|q#K`QIw(?32(I5lWuN5)oCOp9pxj3#Y<Y2^p`iwUcs_)#XPM@(w^#R3
zf5QTlWK>!%E$feJIZVcVRLohxLa-0ENLGMMKzU~JH8=P{MMBml8i}+=e}rfLQT<Dx
zK9`(!79a^yPa-W9Ij02EVWgZAyrPJaRlp>e(LLa5^+}{1T<trFbW$WMW1OGRivIAk
z`8TdJqX(R2S}KAnm_ak7uOi1{LV*aH=h2@3QLTU}egYvg71eVVa1>lp{5KYT0vIcT
zs+i0t^a?GFRP4`OmVq~@f@kRAi=WV+YppZ1sDI)!Qm`|ffQwYMd1Ji=I?NkX!w)>6
zztW<l!FgvXd-LdwPQXcuY95`_3FiNsPdAS)gX@(lcq2vp)DxDAH>i#eOT}(>s)_JU
zcvEbp(#_F9@WAjV^fE1G3bwKnu$6L7!Iu8g`Go~qR3&`>eW-M^5q>zaonW!nxpJ-W
z$68cneEpyJnUm-MSj^rW?d#oFq{U3dn!*b~?kTC*_D;ZFO41zd;Z0V-$31a}EIGg<
zdO86|DBn0_hIaBME93Q_(DSsyN3|&Tu|`i=mfoNu-t}a3awl*`O7diMS|?y3)n|r2
z<V~@XN;g9*c!P5I%-pZIJo-JYDJv<$eXL<8U@FBs(P;PtN)Z*IDufzI*yEkR38~Ec
zF|{ZE*Y)!{^u;Hd(Fr~gW+ovOJZUK~+G+x3*vLy*VhdvbF0vdt?mX?D@VO0PjU?o8
zP2dDpvIQZ#i*ptmejjP537BGeZ=o36j}AXiOBALmLVUmsilglfw^}*(ARd?}JmrL~
zxsO!$q$R_w;`0zvlIR!PXbeqGfF_tPOjJTood+y;fePsL^E8GqGZ}eB6Y@6Y3^&GT
zataP{at;BOSY8Xl2xd*4r)3EbGle<H^s;37bQ`Tq6F7^_Ord{i14cc`3!2QUyC`;8
z`&NYUE{ZL-QVs2thP>)YD-f<$MxXKojlIA_UeF*;GSbKsIK7KvjUBj;jMQWX>>~e(
z_HU(Sz<R`1AfO#G2|g9pNI^c;1b(q}JjJ3sA+sE7Z1w*LdBQwJv}-FZQ4_Gk)>&dR
z@57f`A-v8IuKp99=?R$bqF7;_@5BCPJmFhXL)X9rs2b-<D;H*_AVocCB`}t10wnBV
zb>lvigeSdBC38%LnS!Z&20l_D{wWj=r>K7l#d*?8;5-hFr`&c*wxK*6$SOiIttc-C
z&Y7LWwrh8u0geztF{ljni2}$tPu5YG1T~k#V*f)UC81(ZmWczXDioG}!{A**UMuRV
zLtmo`GYvEI3~=6=-i8Wt;B`5W)r4w1=~XJ6RLtNrz-ebBX{!WLiONFnJn1h~sLDd^
zzq{9qwgf=P=`SIdorg7rrabAjaJ=*kEQRm}BL&i8Cd1ml1T@kxqt8GlG~-587RC+M
z{FUOAJ%fa%Ja*QkK)ry=PHGz}z=5nPl<Db?s{kB!E~!Hcef9F9SHLF1Jy}lw?y8ZB
zN%Ukv7o;jAosQ{w2C<244NIQ%=PE>1XeE{}Tylj4p*#4sA{Abh73e^5+FB99%v4OC
zC;gQQB@I*ddkF~Zrx?2Ibo9(6@Hv5*hOYV!27ITn2pZ|ebOKQgCp`KqKx2cz7f=$t
zc?ob6;k6;7bjj+tsqa9bE~t)M|4yqXaMIAdTb&RB!MdOZjxb7VBuu%B*uO*G(Ip?m
zslYKM4Ly1Za1wFeLVtM?j<ZO58!NE`c~$q<kv?{b(+E?jZO8!KzPAL8UsPwf1@id{
z_PKP4Ga1Bu08ejkgBmHWkOssCRdF@nX;p+NPZ9kzbTe#_;w)n52u0Dg?^v#*v{$f$
znM(kajS#7BWBHGgb%K}~x_$2nM0H&M14tTh5n0m2IXyrReFrY;_O%cwsc6P`zzbF{
z{!V*N@beVOOojDgJCI?zpc-!AJEY4=jeh@*-g^mf6uEK;=ii15(j}|m-hHRNfJ*4W
zOJFsDnTqa#>-KH*{;dsJrK3=ts)l>@Fxucf)IxEA_k>KiLM)`OZo{ti5<;~auC&dd
zU6G?tNyE*|11>U#_A-5K=&%q_O?2S9)t{o9a6ncUo%&9%)t~Z(iqSvk0k<s@#Z4x?
zt#POW9TY-V6Mgrc{z6|P4L3Ls*1<SB4>-vX+t6MieQbTIx@gUJ*3}S>i%fM}^z1y~
zDifZDTl$Tg^UUWV6jvGIJ9KOas3A%irL**_)kTFzSrH-7lz}u;!d;xm0yH?i4ecA!
z*Q?J-jh$uShW@W6_)4Fd2Bqlr`kXY}=9bOq$$3uKJor+-T1~XH4SjWMGgj~x(qBQF
zn>yd!a?;n<=++j!2Hyi@a$%nN0yZW2E_!*MSryXPuRl!JCu)d_yhEpjfSRIe?;2-D
zS+OA;cbQ5Z(c)1$Tc7z5XR!cy!*S*)y-R;8K*s)k^!^3FM~3)5T6O_!*5^FL>5Q`C
zL%_qL??&k#^)(*i)L|+GZe_j+;dsbU-=oVy$l9>v?>r<4_3wxKfAOHWCLMQt0c`wV
zyctl#XN@>41b85+Du`GNScaHUMnq%!zM?ej5le!AE3(E8F)L1E{f?cH>B_V&LBJj<
zsf_TzfJF#raljh`79#Y;X-`m08ydq6ku3;VBdJOVa}3218E%VUhyxxNay~*tnN}|d
zoJFpa5K-c^0+a@chF3+{VL0~4J|$Y??=DLWn1={+pw+6-nguy4LXeid60JiJtU&RK
z5r)dp5g;KmZ4s4%fGykz*?|$bstly=q!pp6ixF!M|C+GG@F*B^A>x!c;0yb77KbPl
z5~5TPutBCPA)s;jh-_t`bSI60qLw0r#c7XFOlw4kI23cN5xsECv8G*AMHs=(*8c#)
z;AIAtX$yj24T@v4HHxT&Fvaxsp(r+p1VO+O$y0>+PZ0^qv=KqT0@<gCID(-#BMEkh
zEC0r1+?l}JVAVce3BSMcwf@`80uyq<%?uNY!!~;*lsOde8!BIGyUl!J!g1I}Rcc7x
zNiW!^LE?LQ8?f27^bgd9ax&pKZsRGzEQU#6`#y>-U;AyqVH;7oAy)ZX^KHO!o4xY2
z`rBaHzUm_W8gY6BOli6e*l)`u@p*~UpYLPZUY}OER&kq|VFH%$XE^YU-Ughvr7NR?
zOvt5hOVvJ(E#Kg6@Y%lVVt#2yK885U%LFXrUl*s>?>iR@XVsKZcTC9T{3<)QMeU?F
z?xWcAVef#?_i4a*)8r1|zKy4h3NxWN!59UjqzWq9gyOu7V8^%f4p_a9SjvBDr@Px7
z@a4Yn+I_rIekUhBi#ve#Hc8cn4rNr13HcxXxSjNleH;hAUU62S3Hcd+g7USI+rZ!3
z)@}Lf#i0_U;i`;E-f{w)EQf9oat1Q_39zlHfNl0F|8vUnh3^iEqfPn-;U!)<@yXl)
zJho9)u65i7EB38JHj4S3chcMTQ5^UV+~KU;=5TLATQME@)b0SD_b4kwb>%c#%XSTi
zYqhG#G<onDUxG4gG?mt|T`&cDVLG5XVB|QU49x*AflRwCtXA0~XE<F8utTYtgSFfJ
zE(-8ekPLZp1)q}>lma{j%1s0+EBV4<HAp}rmG*JFh9gQLmDaVLSk4!h3fz$AQRVx3
zp)uDCRH0t5g3m$)nIq3RCqPt1=E;Nq@MWg{vWGfg4xATAS3xE}85>53m@|vy$>n@1
zskArSnf9oi=HP1>Wq-%eh-TQ3vpq`mU&>7nbHGDDQUw_;PiFG1r_f$)r`VxpPTo9t
z5^xsCv_}n|1RMpzVFVktGi^~lCjkcmqB7D;9xUclQ$hO5lS}vvRp>(I5Fe7BfqB)x
zr0TbGU?~vT&a^`<odjG3`jnAj^5inUsT6?z8*j*iWqcWq@I^Fkr`V%J%mK0hPZ=2{
zPdNwAABFWa?EaH!1>FXHhdJN^XJ<}ASgjl_+6Fdi`uXhQDWk(Y$)&>YQdqH`V2QAy
z3OWeJen;$JE8w^*+#cHl<A5D@IfY)KsbPz4Y6a|fZ94BF*c<ZUpfADga4f0ZIGFmi
zF@;{P$su9uS^>LVJS8YaFVkd_jt@9uCE(-#N>ruj$yTshlVgL;ZUwA&QI*h^o*c(r
zHMZD<R_H4*Pcgc!6@;X$u~%B5uhIIp*u)f;hi6}xCi7RyVppGHbTdq$77Gifu-rT;
zcDn{h$1@#a0jph-O3|-iB>uv)PzfG%cGuzO!WoV+r&SxDrqT;Ec}2oQ#lojj0^T5<
zQ|LvS6cV<ym09lD*Q>c+v1JZq!nb6vgf{mi7YGwl=sB>dt10v>O^S64UKOq534M!y
zcc0|~W6gb*ohSLJaJy1;Su60%WX11&s$hg{M7DyZni@9Pm{#BnMzRoLEKae-WRj4P
zN`NJ%-4RnmLMA8yRv2O-!c?4`j}j&U4&q=wia-LK#K{GySB}VRC9qVmx(MY=5>v1N
zT*ScwRC^)962fc5N2pUIvYj$O!c-R`?8L!B)Mg<9YAk~Cv;sjQ+yx1e3t^s{@=c5q
zV2kN1q?IXw`GOP@9bW<j6d~Ni!6H;RiB=}aBq8lc07;x&jEW<P3E4nL3JGIhgzyz7
zm!L99fVDWq29CYNIWPi$+tQv3GHsDPN`NCqp9CS8C#YeIY*GU3FudPhewLv6?ctO?
zMzRRu0jonB{Qs#O5;EY|zVK99xgZo*DuKl?y1}lfg(yD~SoYgsCA<THgfxYz#KNtX
z>H^3etwIpOa1vQs8GH&Kp{c)&K}HR2X;p$8TV%Tu-~ipg)+PCo=tWA<f;uGDX;NR4
zl7=m&sSw(Ei7DtuVwsc31%fIh`V%D$8%#_gU@2Z%Cg@s-GA2>1#RqJlE~vz`!5ER~
zc}g6c8>+=9YZAFoFselG6v=%@5hO${+&5F)aKLtJj2DUJKmv;d!%1JONc0jVB7B62
z_TqI;;*v%C28&T%BxvzWI14hBI3!G6Az&*$0OL)3Az&xI=>QQ?9;CiEO3=?F25FnD
z4ZTch%3XZ!Z*l1D_99UnA-vXA2)*sRLX;ak58A~O#v}GOb{9h_a*^Q7za_|;1_|@D
z5U_#9K=o_&@EBep?7B>;x=7IZ_a$A0EkkjJJ!FP?i^Q=Lzfvw3M`F2?`d%wBZ82%2
zuZbj<2Z>xFSWKcnhi@~HM6Xcta~JO`yw(WI9`qE~w}p4I6L+@7l)`ttPQqjtg1<~;
z`v7P0_Jv4OZ7?5u#15V60~Sb83$ZF}+7lQP3X#^@92>ko30>&}knjW&+LTQzlM4SS
z#UY`M;3I@}aF>QHx)i2V7b5Mn!9r{{+zJ-JXbGbV8(i)Mpkp3(y_?36Vp_xf3|LCT
z1|8!AoWV;Lvamg%2b-27MX^B#ut6ub7`hv9)8^RW*Ga~NGw3ECz#cDIg!IrR7h^?=
z_G0sqzS`sxY%!bmTq?W_Hj%~#plS&=j_n$QuB^ajvS}4kM3@42Yxli}SuoKKJ?aBE
z<1=m1GcYbyVAmYbRczWTDL*fKogKcf2pOhLF2nj4(R+OWH$3lmJjJHfO0927G3{V>
ztyFahOzHOl-0{TXtrm+S<W+636g$8M0$_Z{roDg%cCdjUZH_Z66{XEPuML)Ar(g=j
z>ECHnY_h2@G)*67(;B5H_AtJIu^x`6ykMvLJzzAO*7A9zCjK80=N?u-F!&5YdYc}_
z0oTlSH=P7uz(=)7CL>ti213r-B=Gm{zG8GnFu6j+52jX^ZJ~w(?!Y8)es_%>ZkA1N
z*5laYI@t6UJz}ZIsU8UZymqG-M~_TE5@0o)jVwav1oO@ZgB2p-J=@Y?t@Uy}iajom
zO|R8s+Tlhg0H@u=BD7O5SSYet2qC#h#E;E#3+BzkM`t$69mbhLv~@7KP~;Swo~OsO
z!KF@sW&dJ$xgN(RwxoEA66OG_m*_D`xY7xLw7a?xZ5Q0v2FF!wdYK-FWO~X0SGIM;
z)lGm^dK$L4rU}4)cV8i79<beg1={<!MCAln0k>Ad`M!S*Zs|iw((cFzDBtMc^Gh1i
z`#(HqHoZuXLW<oejvj;al@%hTMU8evjWIp+=Xy+AT+amHu)Fdfkt2?{p1-^-&_rK^
z8y+k|4*JrTrNCM&hm9E1rWIi+Y{XM3zy_V}gRsVPT+mnkLF#J*L3nZjQbn6=Bn`g6
zQcIDiv}pxc4K{7s4`GL=xS;!dw$R=OVT>o|Bm4UhRocKs_~`G8Folp#Ek_#q(mJHT
z3aAm90U@s#spgAt!gK7=>uf}nHsFFM7bA~+LbOW(4(QF_S=w_fvj@>51(stuJ%~Xm
zzzOZ#Lo4wE6nqguc(4RnqfM*8QhE?x+W(_M-Ozp>@mGftSG8&7SdAV;lN4Zw=J`O=
z$k5WTU2-9`5$y?<(hZB*9tM7;oIW8+rN9y_lZ~j90&LMs&tSG2{&gbU2wj1KH0O;J
zGZEj{jOBDAvZa8tXaXCNAO)O34|F3Ur8xCcfW_7kutf8`5l8UkJUX!)xvm9R<H0;+
zHhf2YmVY&lM{i7_)wxKW2;h_~u|PDl8#lE`Pg_MiGyLOM!8n4`yh_V5<jqF_ma=}H
zvKrkuVJ^e~%p-uaaEtu}I(3z-zD9duNG=eSoIppe(in!Ev$Ek#Loi=-y&D%f5&dNf
zZOWxHR>5d4t7;Zqxk@%(1FRyzd_!`vs6HE)$whiaFd4rez)9Bcf^2n>X!<02bd~nP
z5G)a;PND}_X;p^g64AOSSwAmXjb5C^Wc2hDx{QkqiU6EtnLTh~b_n<GcVfN5kX-t&
z#;+s1ZU<auW<&rkvQvM{PI<~QdvQ}-WEh;F^x_gG(N$b}?<(z;Ay_I}J&B%y$8sFt
zf$zBVj#a=b0(@r3v4<1VpWtV!+ZFB};3BX77W#Dr|1l(&iS|uGE@91DLyn^?rxzzP
zX{`SNcjOapolEaprPUi!?BTS?BzhB$wp#4qO932YDLuG2E<I<Jb$WwIhaWIMl44`T
z>=Bjj6RqJkiY!GJt<qmaf+e!lNt_)wI&al|a+O{d2^Ps#PvD-evMe_Ml92{m^b{9u
zy#d%5@d{+uyG8xD(F~Z@R~HEu!pzcDdJ%jKXG68dC2n-p>Q@FA?E!c5iext(j5uu3
zaBg({D!n9<TqN5!fvfy|%<O^X+I|ZVdf-v!8^Bp3YO$<79BtJ&_Hbh98Qcm@D~Kc)
z$~yOm>i;G)yG7lKWh*OW(<g8YZsV1u-*%yv3aBL(KaHz{9djIEC(yv&-;qWS<ZJ)V
zAGkbjqyG}xZ3D1_8HSUPL6$chQA=bc;RtqKC>t3CU&sF$D>I49US)a0j{Q80rp_7h
zie;%2u-x+OzxOiGBYNc1)>wCmHAt6v3w{7wSiPo4)O8Y9276-pZU9b3oF3@SiSAha
z*;+O$3@l3_^6-ctG%$eB&_TSzBLdJsKZ0`@ku8lF#R5>_3xep6An4Epv9wwwrymhs
zisU(?fv*S(5wVU(M4^E}1XBl5^ShPmhtNNah{Gf7&;T1j(WdnUA_lY(`grIQl$M7i
zdn4?n5UE&N4wBN1plCrK4^RYHqk%32(Fd{FgP{5#2rwUEga$eg!*7tF7h+ut;V*?y
z|D7XxBP68|3V4Ju%-|eCtP4YAW-Pj_2`T(5`vIY$h4??m1gXFDbrC9fggF{4M1tOk
z!<Qf%G||I|Iw|P$4zfm;#X*+93y8=K*ugK4pvUil&<H~*gg=&+hNN^L)(en4$nQG@
zM-x#aj8I1dZxNh>h<+V}kq=`15F%5U7LBE)Ai?tpf)=7!7@>d$A0a^xM71QXUJKzT
z42<GwiCEe_di~s_%V~11K%5rK@-)YK2hkg)a2m`N@b^I}!<eZ>zjJ_CAn>Y}@9;wo
z1Ac%LI}S4i!nNp$2dIStNBXYs_PbX05IDKRna$_=5G;VR**>U4r^!zR5`3=ZJY+_n
z2J-~gwODqiIiwxoY`)ZoVBvvkxR<Z=Az-s3-3MiTy07uTl>3evHlLj~y$r7UY12y%
zFxh;jaLi%zRXzms4^Y^AmfG|pxbCdYvN}yJ+T!sF1j1n+u}~mRn_dok$b=E<uY3rQ
zpb-s&+Vp}0%x-8paCV2i56b*Bxj;ZgoBm`g3FaQC&WBvm8SuWs+AP!4eJ>7ZbYD+o
z^Cdh4pB&)j3)I&NSok!IYP&DMnvu3Us(l)IU^{-!J8HW5>R=kR0GiA)I}H{?PTa>H
zf=>@Hd#=~|q5@7+tYMadc0Kf8k9<)<a17zq(|xrED81Kb2jSITK}S9#ZRp(q%b<7R
zTE;_wyo2Y9ih|>Hc;yCgWc~KnJrp88LF8HymR^4p3=kt|^H1SXVbNed9Py<2da(5J
zqoA)CPwF~Do8Lm}dZIRe9KOL#maiR4uQ*ES+Cw?aAC5=4MFZ?T8e06q_-h$h_gO4G
z=O}nVj3~vIfOTK{9T~FsRQrfIYxC>lQI^rnuxOxrkA^ny4+2<pl<XrGuFYR3#aD`D
z*+m0Adzjk%e)wxu*so6b1{o2QS2QykemEsq`tzgYi(>ZDTMj@!u_X!|^<Y^Z(aak#
z3HZE+>L*5(=1at*f}(+eJyT!z@chK;diHR1_}B5MsAynl4@HN60FSyF4Sa!*n^M<z
z6Zo$Fs{Km1`HBr_|6hiPG+*qm@*pe+wPX<aEu^p4Yx4_>G>l2{iQwpsM=8*(kv&Wz
zzlsP7GFyRV1^ymzgvc)&C6+GDH;AQI9R>gY0z7-zFY_>eCLRSXK#vBik3vfkbfEoF
z&{cGcF}3(Cq*2*1fVVLeoom?3`6Ncs<~xN)S{pONj5*I^$jdRblD)J*{(-}M{?e#2
zJZlDz3@`>O_fq=Bs2BOurBQ=1fUhz5d@tD_ims^=P@})}K8scR@l&NyiFjm?F<88p
ze33trEj9&;X&mN@!y`S6!J@sOFF#KTRR@otu*I0#e9m~J-7m5)f1ebpGKQA7m+Z}7
z#}=dd@~2CouENG3ryAOP`go+NG4Mf*=fi&`kUvwKFN4T8C<X08%h^l5z~8qE<sU;9
ziiI8;V=w5%KcLOGE{rrX20Fz2hQyeM_@;!BI>unhUQUk~=MbNp4}Tq`tHmdbr`N~O
zvi6d_BtxZ}oOd_r?`~Qr#FWOk<6;h)#zN1d3ils=lIupowfN|@Sk_u2xZ(aIKr-kG
z{Dcx>WMWy%jUT>hT-JdY#sQ7#X^kJgXjDf^1_ep3N;eG<V%lRMjd#uP^TSuN#!z@e
z*rN^)Kc#qM%f=5K8l(Lhi--PAL%DQF$E)$fM~%x`!n^Utrj6i|`;QBfdQwg6QZa@x
zj}qPA<+=YjC+Tb$7osKXBAl$I5gWtmZ%j98Oh5AL(P>OqZcJBb1gG37OB$nX8pJTk
zpuZ(ISsK;Q!Ao-CZ^>{2;UO~g8n<!?dW73$ng~+IGd1C}p-hwWF08DmN2zdh7<3Cu
z4eNbEK_2SwZq;fiEHyay32hF!rVE9+t4(1_zX&b3tG|n`C66pO5c-5h{*h?%RtLk>
z`@YqR8Bg7qySLoB<9S8qYo)j^3q88GHooQ&(3u0#OI`=By3qn71xHt=!Oj5pcJA=!
z{)11A!eY7(Ejz2N1rnT=K+UL)fV$(J`BS$(3`v|>_Vfxgkd!HF?>Z=}c1=Y6#arJY
z#erctk(l<LIOay=+@2~+&og;;qfX0n4|SfK^d4eHmwg$PTf)Wy$DV)Qr@M!tfS$+;
z8Xiw0-Wu0UwVeGqrnvsD1^i@9XDn~tn~l1Z>Uk{8=IXxK!-rX=5xb^cqZHb1jO{3J
z;)Sf6zC4j>@PqwePkW!}%hlI>^+6|lEwaER-w(?@rY|Dy?VqdMcwzA4{p6VVp<;zj
zq)XnV?lB|b`H*=lh4S+~W`qz-+jY(C@DDD-?=UJOCMp?MT_&l>A@dKPHB&y8%Nu+Q
zh!bF&zdW_rLD@~*KDk8JjinA`xs@&i9}oU3V9=mbVLYL!GH_es{`ey;r_!m?)j^Hh
zWy%y@sjOk#MlLN1vro^v>m9*u-kY@b`zW>lD^24+^D?V@dZYEW3oVb0@i$-oS$ilB
zO>QjOap^$9ksHmpOWStV9Lz2E)-e2GR5mNu9FbmT$y)IMUthIzz;&=@7iP!ip1wHx
zk`*4PhHWk*-dMZUsBBtqHe{pz)w|cAx?WC*9lI&#x~W|znjL5P@M~p(e*=+yx>x9$
zdiU4&+DXx?r#!e?<nct4&5T7;kB!2Maa@C<0g=no<CL-uho3g7qr%(=9J7(UpqNE7
zO+uu?(Ebf$HBnic7!9-FGmP<5z5F?H3f!zs>;Zj(nE#y9SZrQEQWmp>x7VPn+Q&ey
ze`nJ(g<h$qqTT={bLaoO+#9a?l3o_&K>aJ5`z`Nh{(##P{c(Ex_5Gb==nJ#o!LD)@
zk94NRC97B6qCQT65xDS*Ww{siHlps7{!QoX#o55&2tzNzdQj6}(<*Ldxr}2;KM&}C
zNdGJHVP39sfPO(#_hGIvwQa;>_;|yirnV7@;lDe@pL64!BpbC-yVHhLz{-4l=l+=x
zDlP7ou@QTi5};=NGm;`OazFb?q*WASYE|FDYq&9O=L^F4gU^ray@!Y`CwZi}N&C(a
zBNj1Gf!h&Mn75-$bZojY_})yxZvByX`O!bP>_%Z1g$E~OlElMJ2gbOu+49bv_MLNd
zoKy9DbwZO#Q{&ocqHXv`$)q^F`uwaX6{>Zf4!dWfDf`$*O@#uwe>xow+^scxp)Pz8
z>$co&$Tr)^IT+E)*JiBwLGt59*PAPyR9bS)%F&CRHC8Xi9SRdKTXZl~f^Je7t9L?*
zga0(8di`*nd%*jEcg>Pf5NN7-`NpD(QW%sdGHluz-FLMN|I&gxl*RRXbT$66S^7#l
zp+hn0A`AMvp!~$^9u4Cak&T|v)hK+@d-iFDe==+0NyRxv(@wpoSMps%4|R)n;>HmR
zW!gKnD$g0$`YsOHxyWRww@F4iP)?0J?L(wGzRX!RDIie3jSoMG8lKD5%YSd|f9}=s
z4(lkc-pXgK!&;?^<i&562Y3GQ(`MPTVaIj%r0B(?y(iq52XFN<*w4yKbSpbbi;aEm
zOejQte6cnVtpB%7!eUu<Pl)NdkV;5I&dOhr*zwIgI}Xp8<mq>g;AHqmApX<ozOVql
z3&Qtxtg7sKZvoE3z5|Cp89u+*E!+2@hkB+~PKbMNN{?;nd114&&x-M5kD7u^Ze$Yv
zX=Q$*P2Dmov8cB4hu(OiAAjI0at&Lig|3;c`sB)qS*_Uofb)4jPF?wiWD#3VO-GIL
zsg*@0QW(8YDKE#A#aso#dJn0GOi$F<xS5{kT#i}EB97?lH*c;VNxiZ7Xn}XgZP;*d
z`A|>CZc)v{Cu|?0<&;+rjV!Nt+>ESZDs4KojegK>D7>V;ao|Yah;inzOykH6!=O6Z
zqXU+co83L}mcv6sFUz<!USSu)Zw`6KiPYx3k(R;v_J2MQXt0vC%s<F)yeK>8QJ7e6
z+ELVD?tvX*kAya0*~ZV#sfG5wWjvXCI<J)RnYg;vn@2Ia;F@emwse}lYZ<rlBk<#N
z!$GaT7iGUjuNnt;)_%jQNIXAns2#szQuR+bGBGT>eL9Tm<D8pm=8;#Y`blZN12Z;J
z79wdy5HJe!YADmaaye`D#P=$GJ28bHgxvCdU%K@7?mok~5Pn&h-*PcwPHLA#X!hqn
z2Rd(KY9g&($g2g3yW~VS73>S{qKcJ$I4b$ya{1Dy_2c~(-Knd%McI~1HyaHPkKbOk
zT)y9-xY?;|KHb%~BhgCb*nHRe)44ri#};@F0m~i+_qo~o9WK;udn$Tw<#=$iLC(P=
zE))~)$KZL()b`47k;Hy>S@@~d;}OTB42~GDIxL2kG8T;^_V{g6J7JW&<2n7hcD!!=
zn5J#Pki<=cr<yjR@#uKP-K_@QA}yx9|A<~r{PrlUUe<g3t#9(?)4<2|Z}Rr7_0Bhs
zhrB@>7)R`T@g*qpSY_)%`s(~Z7nP77uT%RjF(~s$gwgfYzqp+X6IF$ll>S|+^%{Zg
zN_D2sKK9-7)0#bb*66FrhP;wxIibbuN^DcC#xbITpR!I;JAX&3B6HHUCP8<k<>d6c
zXZ6GJ5-nFXM+bE8tf$kOu2XWZ++Q^fy)k`h^ZHSvA91z=gY)BNS6`O{S0CGg>2|s{
z8=$%D`K%AwW}OFHCQ8&#6q`QtUN`Q0w3-+7T2m|!7+N~F@77w)eD3}vZUOc0IUS8k
z?YiSj4>^B%oSL^jQdL4`B#iEP<L{Pw@RL_hmC5V<-*nt(s4JLZ!<Pxw>gD|ymBjmV
zO*a-T^>TGhSLGPi=P<Kpq(=hVkJN?@2HA{Vb$cJ|UGrB&Ok~YkeII*_+Q2`a)~+?9
zHLIBVXef8+PG5HDrb>%)WXh<MKH+X=w_RcEDC;H3FtMOsHK^dbao3|2z6Qg1p_ILh
zvll<GUtN-w2?|d*ZlquQF?d+uyP9dMO8SVNm*={GQEm_ysfFG5pn%{K;QXy4WO$FB
zc^sqfV$@Lc-i?}{V|67v+&)!kUuJw`{NZ=z)uIm9<!4Xe{`PzJy<G*Lk)25HozeZ=
zd)fQl?p?N=+q0r1r87}|FTqT$kg=LG{~>sy{*vT-e8b+XlpU!jrL`mq%DOGjR<<Kw
zn!cnS4ANOibzSq)Z%xk`nut!W6Roe1uTV%H?aMciLQgXiK3UH!1feg-Yn)uV6TV~k
zKZVK6V;wFAZJv)RQW^5)3B3}9#v6f!Y1Fh;%lZPs7lo`Y-q~8I$$N&J8oRtpItIp<
z2x>u@qsZ^~f5@<lmX3SaPN;tu6Zg>a(yr=JrsixaHWdiBoQ&iT&#)~`nUMFpoA9HK
z7wejR^YfnCyPIs0Z6e;qbt>E)vepjs*1KEMJ`ChVibbUojlPa&Pie6WMogxY{la7f
z9xQ#YEWrEt+FpKs*{(;z!1$;4=Gk*%W$(IX&cDzql?ySGu$!G2%eg?vXI!`DGJXWK
zF5jI!F|E2i^Gs<9k2Xx?{`@G~F&ei(@L74{sMBIGayZgf=#b2k_d5ZPi8eE%8r>7q
z+x~ttG(Rf4(KERy%$-E$;NJ&|sV6fY)fug5Lz#@>u;CV#d`b0Fap_L|q2;kmqsN*Z
zHLPX2R)V~)_SosVLvN1BEvqIm<crVLi+DQwS^#(UrX;gtlG+v3qclF}ESKLpl&wPT
z)L;MZ_H}cd=v%thzovO5J90;T>~Z@?hIfZ1zHcmnzW2laTON`U*NicaD^r(w=YbEY
zt*ZkyjbtC!RnE)DXMGOty!XwlW~rqT`h_@a71D`sF$>w-aI4sbs#~2=-{fH#JDS+b
z$@E#N-;Sn+=7f?Z95YQct-`uR;_ozVzn>?cm9;4xg+DsIn{5`9JtgFx)Gie(COcju
z=X_UHPQ(`aJ;ypq5&P(hh1Yu3{O6Z4kLPGjRY1mSM~U+`u3Zr6`o_Jr7ix1?Yv;R5
zOJ_eEy;$gWv^wh8CE}t^OJ$q<`ixSG@uogEgL6!Bk95h4+ql88j<E06p)r?Dep;!~
z7a3O`cv(ip%wFBG1WFvynRg~JzBK9vmW^KdGhuIBL82wsz@h1MkU;KFCzf;=xFQh%
zMsBOWAUhrqdVEB}Cu1~gRG^Tz_I8&{*y?HGnOE|m3x0dJ_o&I2b6=Rh54by%PjGiV
zmy~LBV&elrZl(XcXymQ*qA@f5QoDD&K=HtRb9^h~&te`O%!mzh?B83T-jt<b9y{vO
zt=6fn$GEJ}e`<I~Qc4r8eKGrK`%xcDk2^ti4VuH*3LTFgi#7<cb9}XPt)+^3WqY3C
z2ZNO?JFI8Y&zUBb?_2+C<mAOj^K<%AJD!Q>n#Qy9Ra<`=a17nB!!ADP!yhxGwMHjI
zt&Ea0BpKIFO9|aR)Z%(~<G<lDb0S=;gM~r=338K@{*tp|z9BaQ{jOrT%iW^)hFs<R
zV(|-=29!&luU6IgsY`eLpVXHSu<WDHl8jDVF%3E|$CwxVnzV|D>JAf`i?$ov3~AX2
z&zMx0xkvaXRx~O0o1E~+l{ccRbNKip(4P$C^zS({?(6r+M=dtJO^!YPNRQF5>N|cs
zC|^#NT~?mZbJY3C3%)s%nuCUx8rbKdn%j43i5|~<c~0Eu=Yig~;7(p;&lkDgDYnzy
z;7Xm^xP$V(#e+Pt18-x-9m`k$k=l`*c<h<=tMk_5IxT9x{@Jo7#LjTWV*i5pc{OTB
z^177y$MsE4ztf9@9i?J}+U;USLnErs#~((?OPgF=pFBnza*QmvRClTRo0CD*jj3(%
zuLk$)YF)82y3P5Mw|M!fq<r>KquRhnLw0eizQNrJNmf0oS#$N2dNcLKuq9L!K4$TF
z#6#Vz2)bAHr}-mJb2#_x)pM(_x<$U)UdkO@pZGpEf9~DS>cl{E%|8RWj`HkWwLIKY
zvW8FGe&g+IX88W<_B8d7szhCv<TFF}G)UEP%_BLQJ(mxBsqr%tk{vvx<@&Jb>7>?y
z&kC(ytaj%Hy)>;!<?UWOpe506m@t;LdT+6K6_k@MsV#If^)Lxc7@fH6@_<u#@Xof1
zP(!}278g@^=<Ijbb+}&8hjNMp=ktMgXEI(|l&+k3)~r?eTH-caG@`LX!dERhoTt4%
z($F#pef?;Xac|iP*>0KmSLaeSpkGJ3Ru5E-I<y?{3Ad>J?_76y?&zmrLVny>MZme;
z)*CW7qX&~Yy1e%}4zq>oQtI)^-o<9NN>#_1#-72`DU{O}&p3o;XXjkHuK4D8>4tT!
zWopRLn(`mdOPrINyi^$`L3j1)wX04oJY3CLLv&l*)pu<d{>}(OPO<{S8D~OPwtuYX
z+m>kTO$c`Nu}B>4lzK4vFyqmshelS?$2@<0va1>D@J(KSmL@k>YP1@n{HbH)jp*yx
zdIxaKz;(sXNWpTwFmnHU%&LqYdyVr#RrX7p!uIxj!q=41ZEx(kKmY#HCY6`BY|^k7
z`pRW+zieN<*~p>7*l&5%gm1ae<m#t{<}aHJY4<$8eEW;m&xF3gWRpv(*?wWR8^rht
z@gqIMx}DnZv*vVG!z3=RU;df;ZCzjHX}BEMvzxnoqb4e#speo^c6`^tyu)udt=!%!
z=%oiW-O_o_3#1kA4_&$3ir!Jljya=d+-rKd9`R=6np^yv+Q3}<gv<V;L-`$Al_?h^
zOC5qza)p%ol?VP4wIlXJu+q<0wNB-hzF8JyMWY?(FFqFcN-jGVHJO>V;pM9TpYW0u
zmG|;H+E1vz#F}k%5^68qEzXb^iK>$Ix@&ogxA)Y?Ak5r~yHM>)?&0{)dt|dWC!-vE
zPk&f>o41~cjn|#qvpO{I&Ypd=5OjX9Q}T@SU&Mi*M;Srh>*F7Y_dlmbSZBAhn?LFs
zjz65pDSr}qcC9;MH6T~#%YOE{-KgR15iP?<<spAHxn5#r5B_lIlz4BVS1%YeM-wni
z_Omp)J7*E4&O5F=G#oGcy<Oh2r?tq(*?d~uYlCo@UAD$oP})qWSBY0JP+qHZTuaSY
z?YuKnurK1Hqrv#K;XfQZ_POpI{BPtL>BXwgA%#vOx^KE)l-4&zENH5k2gRR1V_~O~
zoS=O;W#Gj!qm-fe^Ng;`p~o+^5=b)L;r!1{g6x9832Wb>o4zeNdxUc`x?}QiEjtdV
zU3Cf}JdBdLI{vbIcrD`R*)f^tW<hTQ*9jg&D|LgxMk3?!B0gh3D(pp8EvH4|ku9hD
zat`J$wcwJq(_h!uBu?&K{}}!#;IDgzOZx8<Pg{gMsyDS<zV%^KC)9S~vr*>vjpjwh
z>YJB#e=W@Tm|p5%-w+sl`6Datjgj^Pk8y{2<3IlnFX0WIPn3K5q4s>qM&SAA3k0*n
zt<qM0$EQCUJe!k~<z5}h8n{{*J1>>wHmC8prZ5EOGnRDc+6&pF2O1|AoEN5ETAWOm
ze|7B2op9vtL-7}X=<50UU!PEjJ448+lYcLC?}lHsg7Q}e^bP5r+lLOl#m_nV$;Hbd
z>&GJGPJPtd`NsF9+T!k>w4jw2k#77^a{D6^Z&2?MO_%8xEJkphJ<VmDg>4UW>Q*F&
zus>uBlQLFwf-+~y^jHjq*`24hAJ=&`HG84-+PZKdW9rQ}JNwz|E{6>B)><=iO+%OB
zHD&WpccMR8yYxS;6H|K|+dCi?k8?leVa64@Idag{V5P;dP`zjk`XiK8@2+k}-dgSv
zuK8O9&E?MgGK+|(UKfO0W~>u#QjDwjga~2tK7=Z>3j(!VW>@6Xa-x#&l%AH4h#Jw)
zyCG{SI<y|_i|+_KzCoBk$FnQiJu`RKw>m?Adr<y#NlVmZa<7(QwyPyM;;!s}f>J*}
zpvIKN2HaB17qv+%tr!X}4XB?CirW^y=I!?9T=P}+`QF{ZwXVIl@@3*9E=&i%^~|t*
zruO+pg!(g&tK9M|`H1^b>{ctP$4dH-<Gn|3j=8>ipLpcQE~y<4J7gW3>pNB+%*qy?
z>h}85Wv)jOE--jfchpVB&T8ZGht+4L+IMG5<`CSJMKis}L)UltBx+b@K0D4aUq~9w
znrZGd(70~-@Q|GB8_(<For*0Alb2-bE;!$mX4gMny5JW)o%Q*tamd5gA5sTe^1}DM
zkS_~(xMRQ5e~1chlWtKWZX*UIgrdCtNy)7rtIyPH*FC-xym4>FTKeTsG-tunVNxM$
zKy_{3kGh%J4@9>Y9q)|1{T~vqUy57QDd|_e7rb})pi5V8Usl~$Wqt6w;swJC*(bjF
zOrP$wk8g78LA01&{IvH=zxKD)NycsWmkNev%9ckr6e2#P6N2gW{=&jPWGE4_k5hbe
z4sNjI4X@nKyS$qL{p-){JEwJR=mU5B!TgW&mKWj-hW<_&nX0_CR=gr~?|qlVC&K*h
z3j<mfU6R*6+b~EQ$}VnJbqWKW=^qX~w6<cHEEu2O^)p#!zF(m>X?IPasG7O1+!x<d
z)18F9FKTL!kc<W&EeS+NrEqm#mX?c;Tm9^b_)q9lJ%PO0oe@4YCr3<P%I!`PdFy|6
z{UkShT77d6<FGv6bbaZHx3Kxf-g@83{)i0Ap@y!AnnKa<k)IQfUs5`56QI?h6`T8F
zXRB|^U^<XZd}*rJtL^^Y>%wOI`BVdel8yhoJbXO-WPLD$S(kB)TNG&~n|p|Bob{c6
zPU<kZc+dZ`>z&;k=BZwG>hAWrm$3?kcQ|f?!?W-A4|uD&1s@xEJh|RMeNcAQ#Zo@%
z-s+zrF+F-GBct4KQdNqEUQr`*2NJ%|Pc#=qnRCxNd=@%sf8RM`_#dUAE(5RavZ40j
z?HJbFo)=~Zl*)V$y?)dB+%d+lc(hI}Idk{;U&|JuKm9v5HaGDTMWZ6xI|r5nqlRLK
zkM|n9o(ZdTTW&5KDh&xPWHhC&8td(;zdmtbW&G5_nZXF3m55h_skP|a;!^wdF0a?t
z{xt;Fc1|Sw&RF%nmAZDZf%wV&w($vtT<b5^VcjwVW!$Ryh0BZ??@j7Gvcp2eYxK+6
zV?jR}PTwk;S%^>_Hg8q<{CzuaZFc7c>HA)fH;nFA*=}||myGwc%^aRx?ydFmIn<uX
z8JgR_k&ORHK(KFcN}qe(SlvJSiMsJ{BY@$RZ{WkIc3L<fS3hi`@o4f{`@-Il{+=Ti
zOM**I_oWvNgp+PvE-=Kk91pTyZXMSXkMEk^zqu;EvL=`F`oUo6i*bby6>>T&OV<yE
zj{X(We4}*o)nzGMm;D*pAE$VXl&hC6XQi|z5n7`kO83^>S>1SDujT)bh51^sfbwGL
zwmF^eUu|L^CR#BPTJL3QrOLgU7kK`t{93!Bl$#!d@_?srs+m|~)V55wtg%><nQ7gW
z9Y0NNThP8xl3r@fh}^$sC3k%3@`*sg%E6aanWJ@0#`iPizPxo67#yLmy3LoTsxD9b
z*YK_5&AN=?uV%}6W`6%&Bk4Q**e!YW*kGdc=v+?ZheN>Ku^};uY=a+b@~fpW-yCZX
zXII?)t}u`szG|2_&v@Z=ve)E#Mr~ipxBWj;nUB|UFT8u3$DR7#OHJskbNP~!lDv)|
zzdb7P$#%8lpQ2pG8);pU$79)R+NA-*B2fVn<Ic+__ueVxZI&EFtlWQi$4aHMzT?zT
zTrctAOxM6i$?L5ff74}?J3}?!Xg_Nn?26_|yM}x^GCobut!trHEhXfu?ydbHmHPTr
z$#TG3g~!x8t&=Y`o>$3^5Sc%-OFzokHzYU7T>eqRTgb4_!>MtW|MnI+9bxn3)qf|u
zUUo$H+}B##qeh+4%Xa#Qv7jRqo{?WZSALZj=6KoV^ibZOgxnQlV(62%e+8e}ki(6;
zZm3LT%qT{%>f)X}N)DHPG`r18*QIT*%eG}-3zuqZkD8u``tLEVUP{?1>K96_!V5R^
zgOutej@tjX(mU=}P^M<B-Aux}5XXyIRqt8@4C}A?P&Dmc<v6zAt7sduS}{<n*TFFY
zTx6fOR!x*oX5n3XuT<$B4XHe45YVPDR#L&5yz=7UbSzueyLWd6-caZ7EK}7NT>Vk(
ztnY@B<>bl>Fv7`S)oZzk&2T8H;nt+eR(z}YY`yV$tS9JfUGSGux9{mU?_PbEYTe&E
z_N{oNzah(6Ht^bp^erVFiDI|rk4;A>tAb@JRzFGUOsDNo8G7l@JtP-!rtYqc=jQX$
zswTVkoGS*OfBtu1uk$|~Pp;dTuMkctcE{c3-m;z%Y+EeHNIGyRb*Qk4-SLU6n_~R^
z*;u?m<o=!wF^ezc+LF17R^KCq3U_3UYw|C(Eo>4C2R@f_j|hsERCu(E^R@ZbtJM9o
zCf<~Czjy4Se$}D}*RDggjv@4xtrzfo)`|PiXh!9%WBrqXK0jt{YW8!3TA8T;$+FO%
zD_LN0hK}P#{9Vg_-}pRv#zh6y^;l|;Onsng0M|OIYINF}^YZ=vlDV$P4lJYV%JJxq
zs#WoZ<oi|A8@7|WVTPt(-}>jVja~%wd-ZSjZ_?vuQ+Iz@DjCb+c#l-QDci=q&hd8%
z_}Y}(fzQ5Z@M-hzM1)^ygl^bo-}=sHLjIi43o<_z{5gILi+z4)BJUk~ZLO5JS=p@p
zO;ne)I(<cD?R%AubhS+79mldI^~Vg=`WAsr|CSwoMt(XaOBd^NzI!=t><*6JC=a}t
z=pkfk!HFz#Tg|*L&3HZ`S}8qxFd|az&#L$Ptwwx5eLWG?HHw_>sNT#LYW3;}=+FLV
z)w4yFd#ti(`pu_Z{T-rnw85*<^R2TVc%L@}R}w2&V@-cYcYK=5tn?kJy|F`asNtg@
zXZdDo(EjY8=w(%d8;idE_fo@yrpt=-RcrNFoBw?7U0bbL7a8wJ8d^$p<bLA%Mtu)8
zudiSY*%G#OzIrcoSTxa8U*{YndOo%};4sz5$F(n<{jBZMm!^v8`bwVb_s-RlG2;;<
z->j_mOZ}10FDhJgPH5TK8TE>_dNDg`$#~PgE#ZQ0QPkerb}U1P+olvS($I7Dt*Bt-
zj&ak^)C-?v&1J$O`&MqB`B_}?e(iJi`>IEC+4up`1?He=s_4R#FLmsCv49^zL9DeU
z>c#T*+X1iKDjC}g14<*J952~(E7t>c`xi9sOZ-2M-aIa;w0|G}KF=&oQ|XkImD^NP
zwyBxrhQdr*wqjOlxlUQBshFaovOO~`rX}QFDKO@OWlE-iio&!g?qn_`i=>Dsh=d5T
z^Yioj^PJatz0SQ{_jO(G_kExDc{`M0|1U=DKBp|L*lGV}W>lBddE2*Afm_$NTXio*
zvWh#nJ0mP5^dZR!VC&^@Tz1v?MwPe{zKY%|V6y&0W#;E52^)F=WfM#9r0Q(C;5$JB
zlCr4et`x8yH4)p7{&^6_h1{R*P5j#S3xTSuNsXs%WQ2=m|7QCMJ=mtU5c3Rm2*w3b
zgec@tBj_@zw0vFhx+%8<lNMbY!x-tH{|bQ7VT`If)k_jhnD*Z{ok3ZbfoCzLKBa~X
zAZZLAj2(<|XFT9MTQ_dK#X14xr~cxNJ?UdTFP~3tU_w~!p!Y1TaF^<fZmsB8OH_RY
zC5BLCC^`if{V1pE5_(-<EX6Ytvx0#%)5Jr!GRyf7s3<fyrJ{wa<LB9k5?O+}ENa(*
zWW9r~V|D)S#Pr9PP=)_oh{|x*0Mbgu9xp4J-$D0TTwZNfeWWURp`+0k?Wc8fN-lXc
zEiC1~nZdj#^C6d`oQnlTf`*h9j^#v3JwM3fxU)2o<oNiC3_JKEI-xTxx*tu^ZJExJ
zX8CfpQSb^UV?fusO6}IXN8-O#C%Vf@?H3T|dw@OZa1;k&vjkX3J;085W2{?6a6@xn
zvWj*;rhz4it`_WJ5N+#$NQk=z5y6YopV*1^C$X-Ia9@+*9lMOD+92ZHFQzN%(nFk*
zNgf>I8`-~4Re+CxcY>y@Xh7qzxN7NVN<;`^W8Q8*&4}`~iu=m(24ex0;N=xC^c}PC
zLm`daSiRIfuFNpoN>LX~cP6ha=1lJ>fx3S~c$$K>3+l^JSy9@H`Ym1P2AzmkcnQUo
z?s6w2Xn(<zLEvw;^RU}qNuiQvc2ijD&=2|;W=>diLR6TFqdxrEjN7p6c*G%j;siX}
z|Dj{J@h7#*40b|!`_R6?eJD)-?w6BEhKa00C9FPdfc8<P4-FT%p6rnj3Hw_%(K-x$
z=ul%yk+%0aa0D@Ahk5F0?S+opxL`;FV+=FyX8oEy#5_5ZTczG|pMzbZ`6-7bSf^Mi
z9T=-(<+=EwE?fp4l2E^0$~5N$6MO@ix9&2k#*32Zl$EABA9wpC2DyuX37$-fj=mhl
zu@A+83Q%_|WLcve*e}C`Cx!#mD8dGHe|K<P<Z;oRzn8e$HC#GoFayHE__ZM7_`fgK
z`>yL!gqT~9a8!Pu(g`>s=Kan%u5`W)I$3Xb7GFyLTXk|MoQr=jE)52s#asa0x+^0u
z?auJ3r~d-_QG~^iZrxSg@Ag}y8c1z_0-QRpt|U#8jUS%i8&LNQK4<B5t!HI@t?9^*
z+0NIEfYivAJvo$vvHPt#p!gM!mcXG6<FUe>2Ux#M2c3l^*ynJXWpC_=6vLL1z<y~m
zTf#sJnubEq$4$>E#8lP~Xewt!3ek&SXT10MXDJS_QTAYZeSBcbmS<SphEmJ<WM@rB
z{#jXUoS+5>vN>e^oH0H@0kj8R)w0!>stgOI6&dhS!%mRP7mO2ZCOhP^?Z&>ME3r1w
zA@BmF$7WKzZ)18SirY#SRy;9`Ulsk_<)P!2a!Z-70*#`&%$+xq#hewcI`s0;_|;Vk
z4r%+8$(6-<b;`e&a}8Co091?czvUnq;_^OIvo1mSg%VC7jPLNMj~F#Qz7TPPn$Wv_
zN7+x>Frg&OIGRE3*aPd4znyvdHFe69y*2{NOHMFa$f2Lqysus@meHajK@7*T6J(vp
zq6mVF%`3V04ZGUxMJ`;Ke{a>&gYjcz{!?j?Ub$-_@*(Jjy5x&=q|)cFu}1aopFJ9~
zl2^{MN+T+>UaiD7q#{5^%$uGIN)eS>Q5c*5XI44kGX9C6RyIFT_~EuoD&D`VsKAsT
z$p{X782Z}uMMOpc0<Y^WD~+mgXjKR(sk3GIl`33zMSbXX?0$CNy7`G$E;EF%RE~N?
z4PD3%PtCQh)SMucudAps+_8zROl?41@dt$&hqHJ0<q!TivtF!h!B2c`tFfdpIC!87
zW4Cg0SKvby|4-k_j=vI&l%WiCi!Xq+EiBeT?NJB{6W;EvX9XxaWY#&8yHoEn!d0vD
zAFh4y3pU=j!Nf6+2txGZnE{PrDyqH%{EdnuZBV@IOYR-<?pHchyhK3%62DhG3<gV^
zRlmG{mv{k4Ljp{>kGTz~`wNT&6@0usx%Z~>A3nAYLGQb%1q>@Mi<)#+dFYmkJ|Zgt
ziP=*&W(gNue+YZ6Q)x>LOA&LOdrSPJ<e4U62Hf`Kg6_3`C*W|bF@p<o*doZubh{Gx
z#Ubsk=*7}biZ8aWEV)Ua(0db0lUY(`K8a@%h5s(*m?%Q`^3X(h4U4>xYZU}WhZ~!x
z^XrVHYasqBX7_zx^KnykpIa<Wq<d(Ea4sg1@mK`Qj}?}D!8B(mLq_YTe(awxJ;HFm
zH(MxQ0Eww{6Xd?LolbPh5JfQdnbGGWTJXH_F&ixUc_Q4&!v-;gnpCr7*~fjX7D!mA
zuQZ(&9YU*Iu_Jfz8Es$XpRx<o$<CR4wR9o9Xdj5O1$0|G@q61RM7-*fbRE!|Mi}l+
z=^+s|Oj5<sb4V2X0EsQ~0?4aHN#;w0M;wNA9{a`2=u!xBC!oRjNnaW<o_PX^-8BGh
zOvjUTqVHLiS##gBoP0SVvV95Ec^|rdRhIyP#MB3dt&W+3#eg12UoxCfpMSR$uS)cF
z=v#M{U8N41&^`!Zv;a%|BxK<qa6A{4+=G6)PR&+_2dt`zFW@iKy-en;e4lv?Q*~S#
zI&LiGl^S9~v^^_qQQJ5yblJ=Cj$oGX$@2sXb^^s!1ny%<e-w>F_bon2b$8)rZRaw^
z#K%pYWAeHcS=IFR61RKit5Pp!J20VyD(4zYnv>*Y<2KOV1sE$VwjmI2Ie*Q0I+5(4
zoqTNNlVjLg-?@)549qcnZu`^sK#m>jfu^UBHDq&-E~qko&KAEmwf;wY+;celxbZnz
zvSHl1&GL(VSzT*m*9wO;@n@&nKo(jjdE5)s)2&`pMV9~PWUQ#H!CC$<pgP;nldb9i
z(a(Jw)iDjG2b1DYVzO_Nps*hws`iWr1j!TbtIPV5Z5Wj+04J@Vl3hsLLZ!(>7rYvL
ze)ohl<1IzK{IBVN$%hB5`x5+%sal_1few?PI#ztAt|Ng=GLEG}|IqsBHjLZI0=-Kk
zn*WmSg#U8iuOU>lLA(0^b`N2YR_VztMLS<{UPhH6DV!#8f1L{Pn}ZZt65W`i{v%w*
zF<ptXO=^p%?--YU-+$Wv*k(3&X(Ds|#OZLLlZSON6Hm&3S0Y-S2sRzb%6mg-r&#P=
zprbKwC}8M^zWcJj#|TeAfCOLhkMbNt(0&K<P?B_ii~LXLOwdf=6;S|Ry%KC;BM;Jl
z^vjBAh!%}y&m?l|Q&#!UF&@2;o>0;}7EO!vqE2^i(=LdECly!gE7Aq(&=+*;IQgX}
z<9}?ZLDH|gOL^*!brXWCXB#K}qy;hjOp=IiiB$H|U9ue!sc$e0M7)V87Q|AmK0hj^
zibF?X=oiy)CzA1@&c56!M^<+4&TGb9S&n9)`%x*C;ZV!2S|U{<mJ>fFOB0!Yyb47+
zjCbes*-amF%%m8d;QT|I&Hc#$2-*1l0x(4Me8Z$5(bu{VFx}3lcO_35&r%GO2u|Qx
z`*Q5@tP@^bghL)p)b=F$W(n0;yAS>UW5hKtbQg0%k~2<}-CI&;Q=~XDhp>(kqeLFW
zyw$x_-7{=NMoD2#ZM?6>Pw?-<7{Sr|1phFc+7ivI+3n$^N*`w8H`^^`k3Gks20#aa
zwHD}55}jmR^>cloFH2Wct5|Kf4+c8BTvNg}H~OWNrh{7O4YKbEmQ$aGEKiDhmr?bE
z|BGyR6=U>Z`hAd-<4U;lPv;jEst%Fh!F1gdhs55A9m%Wf7X_5~V(a9M_jDu{Erh}Q
zs(RA-b|x5oPTH)u(=F!W8&ZQz@E!>3M>$t`Q|W&e7!Iy?Z~}$sy=)N9AFVh;5&wXu
zw5#uKHV*BBTdj_xhD!lt3rzGo4+CF|mlOLFNQ8USzy_H6Doat1IbX<1WPG5o+@nuu
z52`%_7T8O&nm%9a4u}#T722k<y_uu3ErG>}Vqcr?Nrcc*)T16`qdNOsJ?<;?8eH-{
znk$=7f7CgjnOSi@uH5+r6fNN3(=B90Bh7oTzLsIC_IH9HMS7eu&0SMr=WwON3L|y$
zsxox$GlNE>*@Z0UWU@_(HC2FJ4ek3Jx?BphH<DuAGkuTBTaXO_dS5o3*E$Bd42;kx
zF!>t=>+%P3CWJSHJ7=t0OKls28LJ^hDOGkcPOusrN8G?)&>vKNNeQh-gTI+t(<w7+
z<vE|kPF^y@0~z*2!*~8&>te>eb`dtATqY%}dU+@-$u=HN0m@Jh=(I`|?GLX7BazbC
ztdH%o+zD{V>wleEiNu#8C<40@Y|Fy6<&`y;AtkKaE=Qwd?~3=C&&gi&Y%!YXlw`hT
zmoN<Cp4oYfegJE;&tWq5rt08P`&Hcxmyr!89j^dWkr4#zEcYuis^JeL7|r@YnwwV?
zp7DvyY1Ub<rW{;uuv!Sv@d*hL7Z`_h9NIvM(QYUZl&tM~2Yxfz(~&=heL?_PBZlzI
z%2v01v0p%MIG3`j)F19Gj7R99m+f~K3)23M4YvnaZYE=1r*aY7rcY2iJ1UVuA;wbM
z6!~}l%s*CjWS!b&Gcn$n9W5>2fR|}w^{d&Wws9!rO~trX7>PPWAy^$-%m&>SgzZDI
zQG-`m^ApY_V}7gst7uX4%-XKuY<%-j^Ek2rbOv0HDi2KR%#YXp**UdR`C7e@Vxg$s
z6MEJ(@il8C@tc>P=BkU>Ob^%LH<V(Rpux_1SWhxl7t0g{p^&E_9@AMgU*81Z#$fCi
zKFQ_9^kx}BX6>&P9Vp}c^@;9KZ~COKL|~T7k4x*I?*wW@a9g5@Y^<)6H{aG*uKb`f
zMZBr-4NOE)g}><AaiqB3?ZqB_m>hNPgt}*(<ZE9w4yAx@32z?nZ~Mddp23=C4grLr
z*TrUd`2yZagG?`FIdR%LV|RnD0p*We18+w13)y()PBc>zFsJ{85&=z;=ctpv2qDV6
zuJ0;+GUD(F=m;r&3<T%B&^XTozU=s;?Y*k?ot}cOUE&CRu9sl*nq{r6W633K2+TQ?
zqP@TkvVA~rZd6@0KJM?Ilh>u^?W1k&vzlPU-eqw+>=mZiO!Mc&H;e<Jm+6Op3w&eL
z$RmcrR9B4K$~Qz&jedZryw8{71kEbM9N<ydP$_GzDa5`>X<r3d&F#=Act0g#b#0AX
ze+HDTW4P>6o#=#7dg=BHu*AvTKZ6=pjWJiPy%Q1BYsK-XGXAUv_&D&@tgD&110&%n
z&`i6bMXZk?Z#V9%{1jujkV}zL=vC9eaQw#l3MWF6Smr~>hXWYb^Z^{yK8qsgBal;D
zlGaa7poM^5XDn#r^w(lI_!!?<qlFd*{`Ok+CFA@m&UZ5`ZzxOEyPYksF*h8bt|k;1
zx^Ux~0!wqroM1mw`?IabRI3#;DWZO5wM1srL8hOl^S>$kF|-+uO^PAM2i|6&U&Q<R
z_2FDr<|kDbSA?EmrUK49ZM%2HWp*id_vs&PvD2-8Eb<SmA*VEaF#pfTNMFNz^fD@5
z8lEagdQwhr_aSWC{A_YpI>9-ee`^$8v#fA#`~ZJ&19j!bKkh?C+0JDGYOh>cJ)qZw
z?aKX7KnEUb*KK5%?GqwMKCp)6sq>%5;WN6nlem?$(Tl}Lj_!4w(zNy{pZ55%K3$>P
zGK1q{{~B$|zj?mZGy1M@ELHYn8+s>3ocGwFtX=S63Aw9WGQW48l~5yR158(nB3K_3
z2TENE-W7N(lh1S7Fw#@D(w6Qo3F4?y?ufSQvv{RX$?j{d*drRU_8PM0Ut}a(^Kl1R
z5zZ^8t*y^w75>$&OB{YDO*vE`eI@43tR$QmGJazOJO-M7VaDD;Tug%z=N{fLY%{BM
zcm9e%-|x}LEre&dFWP7Q@gCkL9-J7VZ)?#97Z}0w?8u8QlG7I)qIR*+qf<?nA_aTf
zU%X{}RO}q?ZP>LUt_fu4B3AmFL3c9GB^PH0vse2y$~r5Wpq|C08JkmO#|#z3y8$Ok
z#b+YJ)<n`Se9NR4(=Ob|r1far;?MS@qi)nBkjj?Cd34JHokJneK&#k7u=aXvDG_t4
z@!>W2oNA-KreR&1!`kZ_hkmHbNBILGbdxXo#d4(BI={cnuKP!_5stRq6rAH~gufKp
zzsxZK10|oDTF>90U%cUC#Y%ocG3{H4$L!zh$_j6FjGR2kGG0har_VjWGs4q;ms#0w
z^#WoKFgpXgU6?Gm@D$tmiEHN5o65#Z#(-XE=#cj$dNe0+>Dct|pAVL(Bvq*nGwMzO
z*WWVqA#ap0a^jXmu=WKTF5<X*O*+S1o)O?t)YDvX?&9KVRq3~w%l3hb3Dq02X8{-x
zgMLWY8o69cIop(0Ir=iYMwnm`vv7Bbcm5>brARG5D(N`>t;W+c5FZ58Z^ne4VYn0}
zlzG!P__E_u{>OG##fG5J7dY)_76#7R62gD_&N~2jo^f-avBcZuj-gFqn<Mi``8Fcj
zPPX!PNiItG&CKX0WAhUGi0nrZ;?t2-MZuR<P@UuW_${a5d%lm{W3d%?Fxl2c#+Pc`
zRkWo;WJT+$gySyGx(;ujH-)?mm#K2Tw~Fm)KDsseiBk3;YF-=ILtg1e+R>7|oWh0k
z<$b7-A@(V4_=ynJcKSYcV`0h#<jWz*rf0?2I!jimTd(nW8#qQQu;4c9PtXHw1@_VL
zww9;SLr*Bt`(l<7OZ!JYq+etweWpy8Nc4v>gfqUO-t@qX^3k(Mc#%+_d%GyPEJ3mk
z71ezkBPmg6T7*LA{u^GlV(J-(+-B#k6PM9k|5(`eY=Ga5p!Eri*b|PI**6V|2?Gv=
zx^%=)nP1ZCNhoN+H}zWPiq)38F6rB1*!|jk@Hzn=A%F6=kHOyIF0V;F&n)*$`J#}`
z`&|Sh83}F9Z~b!DG6`sj-*ffi_4t*I@!_)CAtSfOMGT-@2C^N}5d<BIN#Lf*kUk-~
zMmvg!_3Dt!8@jK7q}N{Ry_VMZDmKoK(CoTD>NVSw*I!=L+X;TmZWmHIPz1TA1BK_A
zs~<lSqt9Nw9oqD0Dfd5BQ8;T@FIn*D9QU!QXfo=a(B|UedM0&`_1R}|TeLmyV7kwN
zbl?HxXVfLe>>b14HF!Ev;6kd<fb3(R*ymCM(=(%9MP~U+P**tZJVjC$)efr<ch662
z`l4t`#r-MMTj;fExJ<)`&qVPlv>aymlGj)9Rf9NV+~ZJ`slT`%UMl3~ujCHp)|-RV
zNYD~pT7J(t$^v5eopzUC0wd?n^Mdr%ONLQ}*S%%K#rPGU;qt!5j3QL@-gZyFnbv+q
z%)3i=0>Y=PlWg(hl(z5%i?K??S){hKrNQ-639r_g7MFO3)DxJF?wV52D&%le;r1L;
zL}QViA;0CkPTxAGA!8i4b5GG%Bpo5kiYM2gb%rgNWUC*_*M{mUU8iNAXZ<dkJ9p{r
zFb?0vor95D>Ur?yxs^@vViW}MhfeeSgYt~(W;gRWD08%oYvX~7mcS<utgyBS77@fP
z47ft$nzq$QIJgoe);!(yES}yO#ew_Z$)YctonM+$x6^8c+ybe;Gmf_6Nhf?e{nGp!
zQ+y$x1Q8#JLi_{F;pB$qVTwPDt<fC7z_!cc1oMdeIr;VB@1~~LUBqwzeiF?~q27n|
zq&+zm!9|cqp%&B<Pkt*g2Hm(EsSRrS2g#e&(B(-N)nNqz`j$b3@q5V!ShazC@xhH6
zO?;aV$@U+hYYDPk$YMh`T%A006E&qt8B9J=nqwz04S$(DluB?C{q)Gg+evWJi@HUS
zz|${WW$NW24D+NRlrVRuKws3Y@##i8cH0pG8es^6p%ar#dpYUhg?gMx6Ec651Tg$=
zw1V-MA`?2<0Qz7%c;>UubBU<=1yfo>SiNJcI`UC)RN!2+#O~iVx9uVUya|U4PLG|N
zZ2<R&>;)GtVe^JqR?m)}gj0p$`o<EVnWI3|ee8Yz*Yh%|75~5r!6hcMMktQ4FA?l0
z=MAiMv_92klke>4txi#pF>=`L$;H%{;W1Gn|5wq<B)t595jCJ_s?Gewa4Cp&E~Zet
zfNL(n@BN2Ny7$O>+s){mvW}3M)@c05Tk4LR@|o1A4#U&OdpKuqh^XCb`u6<h;OVLW
z3wMmwXYPv9lp9b$kaP@L$s_XKm)jD|$DDy@Z_uM>P}DhTIVL%vq)47>=x|ZIcc|DZ
zmhj@clF?}f!`-2E`Nv8C0!Ksk-4C8E$dih(PDN$=Y7u^3jcFwY!vg3AvN#UglX2yJ
zk47<^aPnVa^5Bb$VJ``u3fn4q^-An1{bf>jU!SVFp9Jt>rQv<9w`9?VpiGv>1wCQY
z)kc^>^VfAle>J2}I^fk1L1f;v?A_LnSV26Ho^VzT7sH&>R+;=Z6T8m2xYf@~_~F6&
zzDvut+AE+`1f<G+U+k|?o^&JzBWJD3R|?5TiP$Yj&w2bP_4#$b7$6V4Fm<?gEcQ2T
z)B#{FXU**##NP+)1P+zntW<=FyO&)zxze}eW4#?H=N9K+*IrMLrc;t&bU;ktJMVe^
z2=!a|^+%N-xORf5Pg-Ytmji)Ij=+OU7dvXyGFJ$*|K%22Lcox<GqSAZ1QHN>y$@cq
z=vxtDV%y}vMF{M2Za2p8dwvdN6>(RQ8_D_|;kPIAB)?*E1m&l^fVELQ-phu=`IhHj
z6K*~>icZqJNv8-N-YCOAsIDxGgUx}H^IESK^6BN{NgXVOFf5;cbi9qZtr@Eq&dx`t
zoJnDD!ls|vZ!cndq97k0zk4}}-Ht||lmtQ0)D}`3R(b;y<>=l1#dSJ$KeUF&-HRMm
z{iqt-Lnj`yc5z6@4Xixs0`q@uXY)*e6E4=?D2r0fqi*egYz0tH0udqBGU*m!2_e|i
z{O8Jlq@XFWPvR?=S}!YId8&j|YPu@ji606s!=7P8$u-b3G3!*lKY2KAXQv+`ZOyW`
ziWUOn#V483tc#7GA~-pA=qTWV2Jw%Gv}}A|DnCWPfBiE1)+rO<vHPgU1xSq^{neD^
z_2*#V1-(P@5qTl7uwnP1PumI`w8f(}ZK<!6KU~Zi`mRakL}!F*#Fx4=FSeC=((O9H
zhZ*k_(6ymw8gmV=R=iM${1M+C-Wgm|GlYGJKke9zAtZd}6)eq``_I<f5dijuZVfvH
zewF}WcX7a&_|rCV#Qd7EKc3cB7*Fs*rZ?;=nvS#Smx?}|?zXvEahYQM7Sq`jya*L(
z0X2;0N(5~f6iKZV9EH(E8l~b4leOowz@d!tlZ&OdZSzIDk51U$W0Bz(;AdE<FA7DC
zX@ls_Pes(Tpfubmq;Gzud6nH>b8Q77pczeDOtyOE1-t%4t%sLMw{|S6iD&-wIxVDi
z?#R2xjgOL`2uS-U#me_}E<f=1T5JaDHe++x0V*5H=}7DuJ8ExhJYIHfH{^od_N=iG
zi$7%KXMxrZ^(BPe$W2jk_Q>)0uKT%>>myfbBqkLX-k#xtn7Y>e7<()>;7I4F0_o*V
zn9;cB+(3|bR?X_iJv-cw&S|`Uu*-qt#(g%-f%UC}iWp!d;LEJWRa$eTG#P5}3RvlK
z&q6((xf-VZPdh2AeU1M~(cem22T~LeHePA&xB?F(-V|)g$uk{YtQOPfLE?t!2=VtZ
zO%7)!(k)nC2S`~_yujb8ZRdla-U6$vW4AVgdi%w%WE<OjxA$WR(Y0-9)Q$24`YA&a
zE9W!sX-;~<38SGuUw^x-Hx^#z`lG{9Jo=L&CQ{|dDx>qKzh|d7q#Hb!y|?!V_CQ@W
zdX27rcj^4g#OU5Ai3Yey^e7H!APPy?1N3+waDn(J<VE}by*^_KW4xqXhwdB6bG$7K
z$2c%ReVO0hbgWKJT4ao_A%YR6JMGzapK#u%|JuN9@R`|Xu|+NC$ypLM|LbAmGhksi
zGcWXjIjgO<4O9>;JWpahXH=4(8Erv~J2i8DY~(=e%w;Ca&G(Q<>|YD$a$eoYKNd|i
z6*LZRw9m}}IVSNA3JrBUeMir6q@l&CC!U<T`5uAVr;LUo1b@V&NtCqppea`-OOJO<
z=N(dg>Xh3YIp|my*m-_-B+wRZ7nJ^rqPf<4!rZfVwl=CiJu~7}+uK!O&DJ9`%jSMh
z*Ponct=a^E^Dp-Bsun6|YFjvRVdc>w>(gF^`LiPBWAEbrbi0Lv7=rcZrh?X9;k@$Z
zz`W;pyw)3UA0^ea>62cJFity+1{Cq?Lrsa>iQF59MwzVwMbh(+e?;kz5>c~<A&pxm
zqQ#nw9K5+sD8{ie5RJi$J1fxL3MItla9X{qYzz1f>5<N9vm;|rA#_#sry<hve;3#4
z{7D&x^uWQvIwL)BG14hUZO3FCEJP&3496m5W!?F!*|N!>2L1Zm0}@;W?)&TM8wZ=M
z|1b97+M>;TcR!MZFib8#5i&@9wV^hHHD$_tdyrTm)94|MDF!d|kGU(@5>rQ5+PSO`
zmi<>c#<yYD)c`*;9Ywm%*Ko_T;$E`UPJk`k_Rq28%hAsrYEYBEt{&aK%4BgPR{IkG
zb&Ty*mYrq{J$V~ylZEkEB^qELg1&A;c|Bd~z8|`VO1Og?x3B)Ff9ZqGwJv+Ysxp1S
zJwU{Ins(%=<zb~n^5Q9@;RsulqBkzBB&9}s4%h^eE4lvGE6u1)Eh{sS__FaRkB8u+
z=vT_isXtE|ep?^$YfQ=`p&q^i>b$j6;9`VdvB)o*^A=6#!dyz#lXDtFy^xf!_Ek=R
z9Rbu6D}!bFMi_zlp)U!96omO|jL%q;P9X{KeILUO`(Sup05BD{m>&w2L><}NVP4mt
z=@SP3GuVOcMF)}Eirj2LvCt+VckTa$VsZdZtWn;ZuD-M$;IAqGEo~epvKzZg3bE%#
zZ3Tg$HvdQ?o;M>06wsZP0F$ow!}||ZrH`6p2=uCVs5XtA#t^J7@axQ}J3(Tt`(}WT
zyEM}rL=&f)J3Bb;Ozg0H^P#70u48wX3U&KHX$D{C%L4J$*>So}p>pefbhA*bKRO-^
z4&@p)IktwLgu?GpqlSC0Vpt%ox2e7NJ_hR{VR_hr*vDh}9yM==TV0@@)@gL1IH)ld
zfv<EE1s7vCTRkuT+<H02CV&6RCM{?m3)r`m?Wu1U%4;rmnkUybK1x_TZj9~9F(C^S
zw!J%3F8m6qZJw<gf+EZ+1XBQNh-fc27-Or9Yyc{%yj#8TU5hTe8fy(e4PAz!V|$WC
z(*QeA=|A4REVLV&J>4U%Hb2MK3an3M4Q9<xt`+^yvd3+vp0|s|cNR*t_01MWHYY)6
z;q@^j3mSr~Yu<n>=x{i)-0-$DNT;it(=)1Z<*UvkjUr_EW}7ZEqxD$?p}z6<?nW!^
zs@)5C3Hm4NYq>~w456N>6*G2wkG4i=bwR^h<`-cnjXwrvD=#814mL@)#e>f=?;rN_
zG!-lhBZ%bW%98LcHZxQDu&PN9Zy<>`r*ZvTcj{F={BZ<9a!UNZ-d6BnPL+gVSnQ$a
zvAZAHfie&S3U&so<gGXlnEa$Y4vhFi!U*mT)Wgb0`cmNPIU%?oxHlpmUHKvF@+)z9
zsxc!zT%~y`6z|FJw5xzVPwy}nCjjGn&I>gcF&=UG147Td1keE$Z)f_(+W5xY{N--{
zBs|B?kC{unGcjJlCjaB{@ErRTWZL_s(skHnejZVU!h-x!?}u2npr~rkx!a1STQw+^
z@~`ib?eZjT7y6<l8okp?+d+^jWfWZK402*2&+8E7#aUNSBc48AHHtkrmuYVw->6-l
ztzTB-t|YqcZNHAz2yw6P`}#EV!M}fqj0Roev<~hg0h@+Ou&ZI~ogyJqH_D{;r}xc2
zVZ4xD;QABZF3JWsOwaq;qFLRmy<YR}`sbvj*NuU^c}?4V7cb<6^IJs74Q;DXze{YR
zGZqNNul^mI)QtvXk^%ApN)OR9m}7b&luulU79Ssev`E}m96Kp*8UNS;cY*y^-BA+T
zcA@+EQWo_mXNFi~V)-tu2r*`G&)_Dr?X)Iqhg~$zA=Jw>xry9AtDD#jK1iJ&T*p7V
zOG(+{$QV~ZZ3s`4jx`Q>4apBt^V*o2MK%CPoX>vmP9(cOy<<mslr|_tICGz)t<RW~
zXMohD$jj=x7_)PK42Ho9jIk9VBuZlSVqmWG))X%mmA!5GCdMK9_Q3hKm)0}>dlOdD
z{xbh80_5X5)3f8RIaRjf)9BE~%Z+6bMa+88Ca;%fd5IM6i5db+FVL#AmCZjmKF9u=
zg`U<>R3Y}}aUha>kn8X6R1bXKxZz^CeA)F=Uj07mvL`m-MSw1=&8L1<_lO<-Uo>}x
z;&%I+jmw8s@yr!gnEpf!eogi3Zm(rSza8i#`o54<kZ|ZuLyFAjB{tQ{+RL<qrqewS
z7`t-edURkQ6@UU?Rn-7zs@^}_B+OD8oSD?WZVOmtsm}w(Jr{v(8TW<E9n@Y=No%})
zkP`MB)E=PQvy>UJhr(%2mk=)CS{#o<0|f5fS-C<n@z|}*FRkWH^I{oBPUFQ)T#(}*
zkEk$=QUfGM?9)r6|E8L{&14N18lVO$(Q%rI^`r-`>>`SonQVt6_|DK7U$#QYF6+@q
zqMk~>5AU0IN%`BgZiW#?pO=fgRr3>{yMyVCzBH3NGuqsW;jW~az0ikz%GJB)rE6?K
zqy|ce_;CIp+ZK(!KU*<o|H`2bwaFJ&C}djh(DPkbMJARtEt9w$u=0a6qPiV8rvcub
zeY2`|+d@=T3E#ZYb5{o$cJtrb=ZJP_Lgx@J3FF{V7zJF^`26GP#x9^wYfuWCRZt%i
zRg8P1CKvu2^V+%i?JCl|?=7Af_7Q4R`RN8k+3e&)m$kdG-ic=yS1)aR|1X~XPjQUA
z$ZcZjW;13*@B@m`*-^T<@zb&RL0P^0l<A;o-K=dlb&1?)b&z42jGw5OEbIscd!9}6
zboAy?&5#u?oxMm$c1G%Dn2aMCh4J(xe_z2%Bbb%`Iq<R_e-bMRt>_5SVd0eUPT1W_
zuqO9_08IDvdNK2yL<bC(frXP!rCe1E{Dy3l>1me+Ab5{)*J6X9cYwl?wWO5v^eVG<
zIvEJ93-?KPL$`15hcvkf7{WI@Lg`kEFJ_Z36Q*^7_^7n-Xpg1zyMy2(-Gg186Jo&D
z6wTsMixlC}XajSw7b3+1+C`YUw#n0LQd*_A2C6vuw;5KAlmJ)?v}J@h_agz<U2V13
z`&!$Gfj^1rB99Fk9@=ohS00tyq!=ICkm|ev*KX65v~p9%=X8&liSsz+J20`0?-T_n
zk&Q-CGZ(ANT2pe%VL`3=AFgh;r@cEz%wnJJhBR?EK-!-<t`5ZrnhJeg6y-hjCq2NU
z;uj*RF1Gz$49`b$i=O^S-|7x1dFFD<Tjh#@d*`Rqjs&qXo;%=lp9VV?F9$?$gN%9Q
zRza$PSfBPOJQ$gqUCYgycnRV^KJVeZP;1gozrJL{$DH@*_$8+wqT8JFt`5#$rj{#m
z-!CDS1?SBiU+zl+e|%(ZLW*&|)Ny)w8&M9<mLG))?{HqCak}vh!1k#ENcVNR@&-`Z
zUaoZ$G|9HQ@t1;$bpLBd8yCTE{pa^u9K<bu+VG%ih~M|dSGZY~Xq!>Byp>3+bA9SE
zOM7>oXS;Ybbo$O%d<1Q(7;-~8x&m&0RPMs9Vz;_3QX0X~qi2Y?%$NM9AIHW9pRxj&
zL9KJQEMOxhL>fDv7o@9T$svkuj<Dhd++qNkzt$4>=1Zdt()}iKLR&mvP23lm_Nc8@
zYvn;|-@el7I-fz$sCu;T+{ZxzSTsHBznNY&8$}dPAKrFM+}?GH$CK3}Z1F-Jz`-Yp
zG=N*?uD&BF{e6(5A^m+?#rhys`umGis2Akiiy3LgA@4ydWcGvyboZHBNZK1;R&t;~
ze3uDv73AYzeA~Jw+~euuXohX=^c}&5Qt>4;F`J8{JwvBtK1=q+%_6DkLu+D#-ooy_
z(~g1&>UNwZuA5#|k>a`d8=R*Z2n|fp<$kzXZ0}u?%w~gi@*ksdnJ;W0xo{L+m0nmi
z;uaKRox_40wr9(=14hVfoQDxx<O;F(@0}bp$nTN%@iRULQ-xzM6N}H`Zb;M8b}%7}
z={w3G%eKUox}_A|A_WIY8;(bKU#qsl({SY}%IZtpAa=Wqt^QhNQ;`9PoqjWVzC6X7
zKw7L?o>*O_N2o<02tNJu6e5KeZ;@z8pLMeYO#KQrr^Hx!fKxK1t8K*Re)xvaTYHdZ
z-MUjeWgoZjza{~dr1G{;gFNg=!et|W*`smLvm+q?ELCX5K52W>d#a|$Hpu7mX-nlh
zl;gR8sb3)r!aFuO?>ra$=PPXxAM<Zgm(ty=U)<R;2Jy(`1o2q0;I#LC9#1hNS&q-V
z=U;?(bj`OF+0(j$9qnMBuT?<)Z|!+w$`2}^xuqlxgmyy|+?Sw~n6V@~<ui{I<iqz|
z#O$B8&a`JlTz8AjGw-pH!7H%4s@5iK;W5bK$H6Md&AbnOwx#VjHotH9G&4oWK5YU=
z{E{;T(Fsn6Y5={AA`2wf5y|Ow@zEa5rZFUZQGM@7ppWc^lVYK$Gy0wQBFkI$v02p=
za+3vAToqSn5lp>*f>fotQSQ8VxBs}H*CGcfSo`1>o3=k;JgF~^%7tWl_dc%(Xv<)w
z6(-YF&vl2TlG%xj!?OJ0OU2IIsekRc8A4nO-j(S-F)Ob#!FC8`zTV`7{*K)_`D<wo
zGTkjRCH`G0mA!)thOWp{e&e(xU~X;>DNO3i@P6^;7ONaA8-ztohz4DA%L9UxP8K=J
zlKPR%FVK1$FlPFH`-UF1uc9rNEq23JB|dcOS4+q?I9LLgQpRi_E?D}VmV$ZiBJ<n@
z0s9XsGqyb?IVWjp&m3f)zw7O5`t7=|6o*@f)|S)Lo?B5+oUtZfXgoY+dvL($)3o|f
z{oJS={b2Vs9ISieO}X&KT=$PHHbh7_#friiBfEJ58}lJeV-QG^%Pd!`bw7EY@S5F@
ze^^K4rp%3o+I**M`U96p&o6KFqPvL`i_P0=s_c?zI49m!)Q6$>QM`)04`UcLF;cI@
zi95lbsEOeiARSFpVA>%oZhcgD#S5q}rd)XA%f`f~GA>$vE?Sf8)2^`0cz~QJ5SyQ+
zwYdrVv@PtRMW5lPnt(Y9u2U9rE8HZL`FS*j6%V!N%JDxT3~qMj$4892F*q4f_S$Q^
z=!1MN6OwEa<=TEnGwq}C^W`fmGd#})L+1=P2$U9paqPoCZ>rkvW^(JX&G7@wPU01U
zi7bv|>f?-RqpY_Gf?coldXd_T>xRQqHWQ-(q2%Ue`{3G37UPe#89!20JH+4!-OAS{
z$SgGCTzr|lI<}l76XSm}xg!O2?o~SOhSC?yo5|omufMWW7lzn9cbso(gn2KtjW&XU
zkdOQ3pSM3QB-=RDhIg!-Y;mZOW+0{G(a%YV{9*!YBl}ha8kRQ6RqlQz;E#^I9Ka0S
z<T|Hu*;+|WBn>y)rS22tN457_E*>KM370&aZrcYuD&pg9&um`gr1%3S<)hdPdH|sF
z3x#$G-L49FvGg&8H+dfH*vF0&hn^(Iu(!wFZMPGG5phFB+5sxqy}<^~-T<8LZY;8B
zFG!9Z5X<Jpg64eL)$j72!&A03H*NJ8N<m7xSKf*+aJo$ecO|hS?fD46JJfs?>ln6(
zo1z)q?3H#(nP{k$GHn>L8#b%2SJ_+@j}RdeO!F{0+0M{RY2JOLYoY@E0O&8eKYioT
zTURux$(=Oeu5)M@ff&C*B$B;DGWT(8JHyxOE(Bj57%4k^R1ndtwpEUesC(YHEv=HQ
z+P{V78$<0>eq2r0p=|{GKC16i-m?Je)AB%H+3ZIQ<~+4E$=c0-Ftphg2XBap9T<Vk
z-mH*&60~;51Q|DgY44#ldUuIz>v!B38*xIvDX57Ly304m%9(<*0lDt7^u8p2z96XF
zP2gp}Cwj9ZRp?LSgISdR`)She(wB0@<q7j?0?L1CfddTT!noBvQ1%92MeB!TyrW%~
zopF`LG_S>d*oFN>JaRlYwnBGE8|i8Mfrx`jl!1uxmyY;a5z9whTtr}%sN!xWMaRH8
z*k5850W?4Sp&g8ac73JXh=V6e2!XzlKsNg{aQ`Lw3w$?X<|Ddm-!mb{>Y1JsK^h`D
z$oc?DY1c}nFY<3YZV$#}$ObN{Wjm6E+6R#?y}!LJX>sAKA%4(!yL!)mi=g+X9`HMl
z$@XufD7)v6vx9Ed`FdG%`RHAiIj*C3*4ti1OSfh0DIX08i0k+kJ)vOw{!ER$HdH#%
z8a211rFg@xXAXDBzZWNl@_xIyJN>s6>8)@J*0U@nxo>V;E0hBb^qPfr;tVp$yJ7Ah
zrSuGjkkc$-$G(zDM&TK_apMzbyLWdLw1bOBXCB%)6o6hO2P=89jlq-gBx|cRA5(B?
z{%<)Qzq3A0cSLR{!lP@FY`{n5<GUIImP$UTndNBk_m<@P{@gcv>~5U$!aG>o?snv?
z{^hqy!Nt3zF>sIpSII-B``gVg7RgzZ#C}NsHkjvDYWkCeDEiq7-bQv+;*;DlPxz;h
zSPaO5f_w?vnS5d>DP^onc`wNl25#w;aM*F|fE@Au)fz|@8Gn1O|2hd_&Bb_ux2|rm
zs+>g)Pgy5aU27$Ke>u47c3qAzyWCa}M}IKH2>Hba`vMPfQoV-N)z``jGOGGu$FNm+
z`2`@vRF^2rY)By74zO^1J@P<Bi!|^K%e{QuYmvviswLkLYrp9cI;p=VT&BC$bD2z-
zzf81B2CN#Op4na%g-#Jl-%2!@-<IS1W?FI^>~>pDUolmWOtKqVXpJnd(&))Xhm;(v
znUepvw>m$geZG$YTOYZdEy;j`;REAbW2YN7L)NFFnBp9>Xo<-IY4Mm#j?t}+v5o^9
zw$wj1Id4ao$B-l4eQO<552X#6w82h_0y8q<A@?-n+)YzfDV}z-)mD$6`L$=@f8~Mx
zLh>718RvbKeU`AGXe0pNBpVxAR_F(y9>}!S{SQw%l0Oe^Js)0qrJ0*y*NbkELd#f`
za*v&a_iLfo_?p#f5h(}&WGRN0aSn_ba2u=wIoZ2&u~~aJF*3Lp(OgZJB7OX@o|A6?
zuh2_nA4+6!u>XMFrQ{l11=z+Wj$-Xd9j!M#-=8pn8qa-4ixD+HW&nDreo%JPx9yJQ
zBlppt8wW6b<CW(&hX7{p%k<!K@-fG!mOqJJ<KiHVeQl8FM}@;{p@Ec0K8{1y&tEa7
z??NnOq<YZ>8(-C**Jfnpj_-Im1|eVKt`c8uMa;*yvKwxF83WQjdE?<ZjqRWM%2}UA
zRu#h?_C-_YeQkYt_Tn~|6D?Pv?LAc!vW$J9WYn5zE61lJYelUU?yH6c==gV)ZHz@g
z&vi+|HJotidT@@cl7B0B=vbBU_wC_y2qpsw<$NpU2NpCZ=~Ed(L+TA9Klh3>8HNn7
zsJFzaeOs0j)e~z|pXOJNXL_E*pEWgZXPwa9rucwu7`^i-S?Nq2_lURhO%>QCdaStZ
zQY6M<O-p#d!fmJK@(gcsXH`H14SX+p-d}II7k{K%@Rv^<O^5r5E4$|<U(0gZmXsdD
zlz_{Gv1QQrx(3*7LDLceJb>RQme<v5Z81~ka&~(rCe7|jbl!f!U^|4{1bv;JO}raD
z^HZ#s%$U>TlUA-iF7$PHMqRnX`BrN+3gy~n114~bD`CO;3~uAF7nh-q*BAYDjBdf|
znjPehMy7cy;dA&AqkVT{_e3r;q}t`z6>*VEeG5%~Pxbjcs!EdfDeH1Hhs>_6a0(2F
z^u`G_pC<F%mliHZHx><0UW4HdA7=~4fdh4P?~ujXNH6B|HFh^1b1a)9Q!eDBDmn1a
zH*uTGp8^AHGmfhVTG{6l<A$nQDc-WvZ<22B3zZFgN3@Xcyg|gxwZ8udo-lDy&adpH
z&oC+f1kn>ZJcqw5ys~IsJ?&|eI63c0c>Oj=e^pF22{jdGkPWbr44Swhp#bh3BY6do
zC4KEkb7G~rl^s`kN6u8!>R|Bkps=jRtvZgOks>`?o;{lcvppDngJ0j@Hsw#+pUlSF
zI&ug&HZU?gL<kEW!g=;BChxN(cjm+>2$7_boviqX1RD6h`EycV2-nE}gI<^gfifcL
zE%u>uge9?gLw)^Y48xoWbgviU#TKlRil6l4X5GAoa!!^Cz&CKyJ-3$HI*655`$w8c
z6R9VPS>X(TY<F%4yEd_5#0@j^VDRk{we8lI6hfXAq0EVVP?%QcGWRj~Kq%9;)=GA8
z_S|;blRXetVqC-SS5dS?|7WKJ9~JtjW2_|@4fZc*o)KyP4=!^3YcXsm>@D{#C(v=u
zyk>meG^cba-O<}!a+{>|xfz0P<5#>zW|oV^gU`T+iJQum@guj`2}P+E^Pe|H1Ht1N
z>E21xMHV71MyR$Q@?o+SNxOPGJH47H@<~ruZG9GHSj{g6r+SXV-(eW^qw-A7#9HBu
z#t_`1pH3V4Zi3;b=T(EDllRnauqOv^+qo87!VsX{Z$cJM`AlbezKWP_Y@1slAU~{K
z(^`&Yh}m+pGLW~0giuZA6VQT#rc+&Rp|nR8qSkNs$K^H2QR=tR9g5AcnKd-jVcH!K
zOyigABhq>-%r6@6l8l<|{n0WD<P8A<ug_<+-F3-|ljZ26B+iG>@UQRhX7d_YUUkG)
zlhtA$2)Fu#Nqq}8`@aBF%u0ASZ!nZGb<c=K(`eVeeTD*6$)r1uKm{ql2|nS%AO2xo
z(69&p7{m8(6O?H2kByi-a|O-|{U(T99CxTtl(LhIbb!fD&TD6w-mkV@PHIeT3n$VN
z1U;!qQG-j#t$Cv7UfVvurnjqjm10ccpNq98_L<m%wuI6@I#~2~c6#fFD@#n?1{ckg
z#*9#L8<xuPx*={<ruck=fac(P2{}7UqUjf=<_GPN+kH`g(*h)s(a-fO?SSm5CP&|N
zwq^SoU(4$_^$w~wr4(ui*+i^UwMcBt>w;Um_&(nzqdSG&e!Nc0q2sPD5WWCm$j>Pz
z7A}5qGDqe7$eYJ1GSB(+3B6T{k*MaSK~h?%xo{Nv8i3l|?|Nh`VE5+REaV`0?km4R
zV6_G<e-k;GRZi13HaE6;pgU*=U)bdLxxB%^k`&$A+~5pHlyV7#glQPO2DYm+<xtr9
z4Q)0cs@~~S9j*P;^=f_t?MIQoe1d0wmgAeno4?B*`Q65t!!a*MdYldGcV6F1qZ9Pa
zl6}b$yj;L`j~LKw3O;*e1Duy4g7e7wrA#nNboaOa={MtlSsX~X+L+t4>(~NNo4F{S
z@v?X3jh(HdvC^26-;nEhkj3sm;S!C$(5QE*%fhnSVi-al#(Yee^<l4NnUnSwYCw5g
zRIh%pi`tez=*#9T*>oG{lK<9RMLAMEQa5~RT3Q-cIYq&SG90%q8792%#p@;m_*;$+
zMlC&|DQ@~WV9;eAXnwCU%a#%nms!gjvxX~OACqD_GkAlJitpFY!!Rl2*#_PsFwijt
zo54Kr8oU3UVuqV(3N(Emmo;V0KYt|IT`rj?1`8y9S(8b+L#%!zW@UMz9i-YYuFs3P
zVrtU=GYgnpv<Dx~V&*rbhKu8#;>8fnyIcD-E{g@0Z!2ORJHSZPY-O+S&?WhBgV-Xv
zxEj;mADRNNc`%u0o}6t&d?$?``H2lk-^(g5K>8RccmX%z)<+!uQOX_7o>5ts332)w
zM`m6Fdo{G=0QbHn{#vuRB%WD&b#X4h#~$hdyt6sK4nSU~Waq{2=uTf;3fyG7vNr$2
zV-I9tYA6FQ?(N|p5`0`Yw{(f09^g93q@=dxx_r|-2Hvk_K3gmegm?V%Y9$U|rdG*(
zY2||Vh&QJ6fwKvd@5E$}RM3j4(=oQ*@OMr1q$BMSsZ_fH#8P)jqszqF=INRb|Bv@G
zjA4x~GyPOjZrY_Zy2@h==OBt%28d9)>CmlR+^gJTl#fe(uD*I)wddFEzxJ1z8LhZy
z4og1IbvOQ;d?3MTY?aCZ{L!S4w$4kBnSRlIEC(&|e|C4u;Inb(&@TEdRJn7f<*iw~
zAaTVrDf??+wI^)(-~VTBSp?V*=y-QgVny<t$+iA&#vQ60p5gSibGN!O{%vn0d3Iu%
z0H1T|$QmYhoE3=j?*foEzUu$NW{NKK&#{m^Bugmk5qm6jEAM2wDHR3QO*l_AaUdap
zH%cK?W?NX%G7Lm6MGQf>aGPJkGL>)Bk%=(W;?tG44ry@@SVGn<8eNoZZ_l466@`^A
z_UvH&vaO>h<JNTkfV|2d6s<caHT9f*SVy|kJbCyAKX(g$CzRcc=cVi4AS2Gc(W=rR
z_^jbbmG^l60}q9PtNCH}<(A;t5us^8``w~FyxHNfYj1~#Ds;B3WVgM!t-Sv_Bk%_N
ztNr#j?8WieBLbH#=DE1p^DAGeci`tK<%`3_H-aNgd9o#4$)}~O@(gyzGnrwTz1X_M
zaR9ekc}yKO1)vQMv9#u|#>vbb9lbuSst*Iy5tm!Q3g>LpL%{otLj%-sp9-BY8@&2`
z_SBTHMi5}vEqhHHwXudN3Kpp6j6EZF0V(^=OE*+zoWx9}0A?;U659`a8QP%^K*Et5
zxkRVajj-ja0h_8|tQgUOoGf>25HRTkZEf0_H@sIYS7(a|mg(c)=C6Wvi#Kga@Oyvi
zNQ;HH{~fxoYZ5)f+d+(rgW1ad0y6?Ge0kFaHuxXjWM9Jq`w*irCBu_jb#rC&1Sa`v
zw<)zmw4$dDHy15_Eb)mRA98J6EhLLDZ1*Az>bzRaCH}vDWBNuZnO-|lJg>*pdHy!G
z!~qk@GzkaRJ#&er52RwurCc23jotICZu{@g!ESBo@65_&vKBXzV(0?t{sruM+f846
zVbwjY(JD7gS-0uE5>bm4J~%%aW5bJ?{HTUX*F6ikOy3ZFTh1Od!2Ya2b~eP?N2cU?
z0O_`m{;M9QRvm2R=brU4J<EO~teRi_XZbv);hGmVbB*#tp6W4G^**8A`Y<G5>$i?T
zgwvcylGj}YBW$`|WSNQj;}rhH^)gKMXgw(%xpXRdvQ=8^W7H_S6x}r{{l8boxyxbQ
ze9iZ459A@Nu!raa`eJ9^!xi1E&%`H9l!(G~Te?plUr_fzzX-f_V?06Enk59@v&@JQ
zk)~UPzh~c;%U6?-kMUEOVlVqUf!}8R>IgZ82md=VgK1499wA-paA|)me3u{l7Tvj;
ze+QZF8UN0gi?hgv>uONhdRGDD>OiXf)Uf8!FCe!eH9$}v2V@x^s#^VL*>lMgsK-B>
zYQGnUGlbe}KnLKz#;fS)f6>s`!WYyibve;wyDTJTlw?sW#LUZgW$c0fkD@aVODcW)
z_{=no(=_F@TDdS~YGq1lxgz|UrmQTTrp3&Ka;<S8G(|*Ure(}rNX=bgCQB=~#1t2V
z%mopZ8cW;|l`RD&KtT4F_uq4_^PE4P>zwCazW3)FYz4sq22%gfXlFRl<}yW5ERhQC
z1ju{uPdq3?Pf~Z~tG(MCagyO(v^uvIoo+Qe+=l`^n9;nLIm{Wec#V&x4C-EMy2a+B
z8IOgk;Z5-Xhak97*<Pdowf-qRruNN4(SVV<;d5yUHT;wLZga;1tIT-tU&|j{9EfNL
zW|MY(9cGB@CEX*aDD!F8zpPjTmGjkibl7G#`^ygliitn+HrC+Z$pE?+!_?rmrZH@c
zFOKaXmB}ZZ6Ag!H7ps}Ex_V6I^3+|FuaEV+#dmBZMuPr|3+aZiu<w^avvCC=!XD$&
zB$R{I9i$!aGFZ??;_Oz!htiWgvIvNCMbfs$Onu=8e@`$t-Gz(RBr2<8TxTQ=#`6Mf
zZr9p31JdE&8)G&1|1sdYi<f6K{$%*7zN@9ZoBbq~M=p>%G$hd=xxx>{jJwKoK@Ie$
z`L4o}*Ay0}n5%o)<Pts*mm~kMv3qR<mkoU@oi1Q~$*7oMR5+{*`)&>LpMj+OIZ;=O
zlJEV`1}E;ttc(ZAlNr>0UB=ts1qk^8cSHd(=SAwITYTJ*`ydeY9^jYTe)3^|u4Nez
zx3(82?<3q*BFS`l61xn94LY0AVmLj}d*?6xBRZ$WhQ=94y|Uc9>d?`hf`CH?Na<Xw
zNzTDh+M09#*Is2W1k6yHlF=J+T<IBKpU)hpt-elb)}QRZ!nhf%(zV5kf}Fb6z!>s*
z^dq5hh~$cQ=m%B`(q|U+r5@uyW&_|mSS<l->5rB#GVo20GuRW`nRRr(tFUw|EMRO+
zDSi&Z;W7@Cl6=p~tg6oo=w(64_?xot$9sQmDhb+z(~@=*;xL2!n((0vPieG`<c!p!
z7RsC4#xXpCdZZlUChOQeYImvQK<@x}oU_6Wr8;8teGz!tHe<iKlwNYX64G6)jumgt
zGAqvVZ7)(MzFL`_J_59+$#O<tV`#u=GT(ebo-TD{Hxa8!4S!Gk8{~=$oax8>qYYE4
zV)h%om3zi}KW>Q!(6D`EUngW;%I>^ad+fWli7$_1gLFvgQ2~DkG>W{*KQ!ZYti3Wj
z7vcbzPrIqPK=%*dzRJxTHd~z!pgz&Jl^OnLpjBOfba3LZleh=9J1M;tI0`wW1LAi$
z!r=KiRKPmXVU2(}Qc0I&zP5qU_+8qa+}pA<Hi7pwZknJx2o6)Ke^zG2jg|X~VPd2U
zt=Vfd+gV%IC+gDO)i0e#eOb#R|EUj9$%jL@%m_nh+Y>H@K0S%C;V^``oETK=?=oRX
zU?Cjzm-+@<m=-z!?UBPjS$a}FBHN{+VU(K#kzF}yZ}2-WPQml>>c9ch#k-irRa^(V
z4Bc+zqLADrp`%R{9-R0$>*PQ<Px_Exq4B^2M2Rt>jg8-p-_jvTc;qd%-C@<=1ms^-
zpH?4;2i#-&uVY3hOpiq{aEKb>C0PlW9Z0X5%;e+Ar_$zwhXWlq4xcas{;qGqwgk&y
zE1bL1&oy00gReS<UR!@%3zl8BY2s_$D6#u(N~enWo=OKi8Jri5j^&jkaMu%PxOt7o
zZJWuk)u-2OVD#_^8f6gV1<#w10uJQaLA9xLOTs&@s}u5bJ}nL8{(^(w%!6e>E(#{c
zr;%ZziP$5F<L2vGq+*6=M+Cw>FdeizuPY3L;R~a}R_y9=4Zui41qdgk|9xvfI@J6R
zmCDSGFL8HHlyE*l-e!0yZ?xU3NR91kQ=Ao0UqYLqm_Gk?hC3`|_e<Y0m|tHLVFI^Y
zG}=PvJmfgQ$5wB|2i6)2<=eJle-!u)>^zdF)Z5BGMwx5%T!%q#W^x&ob-VI&0F>@I
zd!)zkj+g(J;_Yo4*HYmf3v}38O679ji;r&bZwP*>C@e<DI&A#7n`rnF(^g&B_3eP|
z$JwQgp13<RYH>**xp-N)x{DjX1WhyK!VL@+Y(3K!kvv1X`9SEWAy0GX_gf*rqp5Ku
zd|r}TUdqa5+itB4+O<KmmjZ1rSKs0E{BL%uD*c$~<g#J)OhJjILfi#uJQJJ=I`|m|
zu#+zTZB8U0#5caCPE>&_gxYcZb<7{S^;QHY&;u!P<r_gs^r8~^I5)6>yi|J^ITFnj
z!Y4Ah+GuLm1Gz0N{xNxW9z-3l9Pymi70qdcGI*AUYAaUdwgBn&*abM}Z%B{&bD8Ht
zyG{9A3Tv4iZz0UvbXC$Ke;sEPQ=e!d&Be<s4S`5G1sR5acs8-P2^ir6>CRF>3`xO2
zHbMSJ#msdI|9z88UX6H&PTQkhECdRav6vw^yBzgRyPT#!-Zo9UbhWXWod%lmL(VdF
zm9*rg2g4q2EH~!4qm$w1L45|!E@lu(vU?cdxS#Tpk-j(x>;N$AaBWoH*Q7zsoISlH
z{;{X_Pu%J5lla&|;HC+Fu`Z740H-^62G~pv8y1cih8jPvo0)B2+hUuUgyFrQO#&u#
zzX--YNQe-Rbcj%!x2?;oGgyZ+m$plQp5>J{^f~BK)`i-D`LtW&19{b_2t8?djio$C
zyFJ*U@l6;3UOO6NLzq{XBg*PwI7lIPw0dKKS-LNa)V}j!i;mn_@9&gmK3%uNoJ8;w
z&Dx3jfsvKgZ`HrCqK$*T3+`DJhIbkSXOK)BF3Nb{Gzlr&Y7CV<y@6t4N+BHVAPBv-
zDMy%J`>nX42`6mEc4)2z8gGE{Kh~RR<?`xy2S+0Fy88C}QJd8C_r_V&-}2sSs4&g=
zSv$#@5T0!dq$Ou8+2|}JodDKG@TsU-r}7^fw<TyCg%_0SuUJ0nGmS8fi=g^oumwIX
z`gRrXMM*NbqqaWV8c|g5*ANNHr9KdHR%@fImokY!VUZY^D<0u`l|^7rEUN^uUyz{-
z10P!Y2<RN?bAF9dV8H#@!f}(k6(hvo;5i{t=$je9Eb1NhS%G@B)~m5R^qt}T=VLR*
z<`Z{CBV7Hvi;~zbH(Sx?Lg8(E>{vC3P%QUu^`RTOVIGhHP1pg?Ypp)aUOTdlsBvw(
zDpI}9G(GRp5=#5QJU%~d?w?2ywT}Ra%?30(l{SU{3}=1rA)0?i4$G^9ZAx-g-!4xz
z2j0LywLUA%j~+<Sl^LLP<X}8ur*Ag*pB*#?rkI6HmvYU6V`o++rpOgeL0xChq`+=K
z;?iwJD21)qEcZ5r875sHNM&FU|B8*oaZ(NNC!-+&w6e`ZhK$-kvqeFFHhfg%On(Y`
z*ljp*-FKwwvBE=j16QjPd5yDn_DpuheS<V789L`oo#gJE+On>GCMa9_X7P@IRmLsj
zIPAxH!?WxCk}w~uym83I7g|vFeXhzJmHLlDd)pD%DOjIV__9-+jXEYr7z@{Fuj=3n
zDc%Gq^`I8NAJ?w<w#~6YAF$ChqkL_sJuIdCv(a~0;<(?uYbiomfqRTT-=)i3^Ctra
zs;;9J8cWdV$fvcfWas6)Be0I@#h=4#>OIr?k@c*Xe+!$uWl)hp%K4#~@I=9HfX=VQ
zoEz8Ym2X<Pc*hRsxlc5u&4#{~+T$&3&71Xrd`0*32qVrbmwR72GQ*@_p21~7U2O8U
zV8U$;tQxkbP51%B{JXjdj5T`m^w=hjFfB%RMmLlHZaQzxk&F034t>J)M1Kk!ws3>+
zF{4)f8ONl)6v_qJkCFc0z;OU`hN0<$KT;Qahm8tap{+#-x}@xW<JvxX{~vU1E(Avi
zBIkrUAluh@8GeJLMiAyG4cse(`U@D0njdQXNfLC2^rNdJhsaI|as{@>^#ekfhUY$=
zf2$6T&J1$f6o_F|h#i4)O4iJ!ky?B6;X~uT?T|VPsk{&bVSo6#t>xl|an)K~Fy#}1
zNM^*UJNj|20lq1&3?JDAIuvt{Td-rnEK&2v?W*e=gefy8`l*&-w(2dJ%X0(92y_Q?
zJ}uRh_Yt(%-)^scH|e2_rPC-*H$C$701EZ$__Bj4Mk`*S!8r>!Qm#mkOggNp3zDT;
z8s2eRuPjV#GMDMDsZQnj*Wfnv<*p30Oq!Q&fh4v}^ybzD2Rd>rZ*mlE9S!=|J_=Nu
ze;n{h#$qrC>IWnWnxldCh**@T`br+^3Ot-{U6}IsECLQre4-+C7RIfCQgdUWC-jh4
zKTW`d&W}@YyXlg1<CB5Ajh<RkNF$d#`*}xwiIW%JBSfMUpL&Kz<SR&*16M*5uPt;*
zp{6QG@RA+)9lsnz@s9=w{-6gYYW}Ksi;plOm%NqLGt?u@CgbCJCQvA?eBnxnU)!vT
z6OaE1{XAV}n^sA~G7PoaZ`zU>E74x#4DZQ7?3Z@HKKSD5Puk21*bQ_*cZi^8V+Y|P
z^odln&e*1W*%?XP-1a|lUmty#;pQL%RUzHDiDTb1_ohezQRrwdn&Zl~4~%QU=9RjG
zT5JfmKRra7jEAj^44YN7fQ&TiA$YMk(5eu9)D3cDrm;JpD=xb@8IV|nuCk2?WHm-~
zw-?qoNypMcPxa;HZj$O?8Ts1e%C#{w|L{A@6}zyO`2C`2Y;neMAg*`#k8nuYLRx<V
z!&@5>T*Cemn#WOAYy-2<jYJ&kr5IJ}{Sf(wR};4*q~Cm3VT!Sch-uMa1SJ<HS!C%t
zVmMHQ;u4M6{XOuHgU5zL>W((*zLO74Se8oi@^$?5`M-9MvHyi3u(!i*s1+b8PWwuX
zWe*tI8YO`V3)1NfFpd^3nTLmJFSU)0nYlukGuFdgX5~PdRRNmS>o_6VQ%oLpoo`)d
zCqX`*ST~R5H02{*<UkO(8dgZ=2Zv%bBk@$+PsYz*U^39|Amg8^i=A<^eapafFB0jO
zaUjb-lz2(`mzA^jjQcWODjn)J9ViF1x;*VJ7Io!naG=4B8xg7#FIMjl-x>~C_@ON`
zWVVOv?Y&=X69wsc!i!ZF3WL}^sjwwD-dGKb)CUadtafzg>VNyvd{$QM@SSR`5e3$6
zD&z>%EwY0kU{L%5yexn1mTYGnmAPTYkn=@SLud_yB;yLudD*HceAyHIkD_SfTb{Sz
zxovb6_hB5leGo^273*sxDuIybr@BtQ5)Elvp+Z6hP=mcPXuRs9J5C%hyJZI3Hphvk
z#*T9RA*19RxUXg_-dKm45_-8;+Dv3F<yrL^dS$iJj1c5D8goN_9M|tE^O5liUf)0L
zXGz^Uf#c=<fN}4}tD_cOKnBr%Lum3h(?z{rf$!53nYBOIxJxqt`D>cNmWjks@&kC2
z$UdYW=&OtuWG`-UMpTLkA!=AW?Ft>NImTQ_s=zbl=)-mycdEsKKVjuZkkqI;F(BBd
zKQ)HiWY=_b`wYz25Ks3QO6Rq&UEof8&ieeW6rOL4(I@J+%SYdB^aAfHNtyfraKShL
zgc#?Sw+*~q!NRG?v$A_g8&3H|89ML#6_K?yADNcZH&%U74Cr!|uw;LKh2WDsCGPvQ
z2nWa@w_C@I_qYNWQAS6=f@>k!s_M^}RcURr_2ic$asiJK1+840cVi0EZl6^G>z}A7
zR~V?hg;qB^>z?a058_!$3nL;_{NesuY#a~$;&Na#HBw}f^_I`z)X2Cz<n`ZYjisT^
zl3abZc+97ib|YWxW;IPv!>JXMjprXisiEWdSQ!-r&OQv|l~FnnH+j#vbB5OT`xkKf
z5&2>}kMEw_Y&{dW7NtqD!KEA`h{tw!73iN8ZG`}2r|_^ks>N_q&Y5rwZ&SSo;_h_C
z;tf=*GG`9_gzH?aW25#%4rza*MRnhi?Z<?3JEYVvYsbum>LYSHJSimHR{MaubH)rr
z%z-zm?;dQsx0m8|MHaXYW&ydr|A+(~Ug+u7Ro_4rHL8EoCezMZb+5)9$3Sm(JGdT{
z1dCIC!ZBZU-4opo=?k!OP?My%?{sLy*cl=-(L<2cEOdbdGQg{gb(5P%$e9YSz1l6L
zl>vE%Dc~f8kQ{KrzQwID{X|O(27E#=IYF<?EW<cVmwk9}yo+>pVd~}j;bg8Wl?8sM
zM7uIrwSJy~zm=@%4s<~5MYL*S07x2%G%g1gKq1!zV`FxS`DmOiwb)bT50bcV49pXC
zz7qRg`{bb21BEHsFTue$^$U+i3c9GB`%~75@gYj?y}%e`G5SdRZ0qq^z*xrMnokEi
zf5$su#X9Z^j&p0qAY&%!3)c_yys1|OSd^LJ@>8<c6CtT%2C+jPL;H^-@CFJogKNk7
z(ZS75#BlGFKgC8hg-nx%O{#R;1=G)k4DbU+_t#SH$ILPd(hx;=rWv=5QHm>pywy?W
zti?ev_474hQv_uxU*&p*L3x-j0)d>h9c!pK_v8$4iWQT|_xM)G=sR9OXuqUG(vz)6
zMBJ4LGdm;7o2!2Bdl#-8{3aO4IiU=k?blJ_i}h7oCShIg78tqI?=^t9kAeEcpB&6G
z97~(nYjJIH40ZQ6h_MDMpp6rTzZD6L67_%xwSio9g?ElLwZ6XPaG_`P9pEGY)lHV9
zkq}C_CQQ(M-b0@3-%ky<$H*2og`doatHCmxKAy!cYFAd3RX&-WzMHU^y&LOxY=!BH
zG^VhRGJ8)@narkqy@4zIb{uhSdD&THz0p{;EIcj)O9$QCrn^oXX?c;lQQc!L)o@FU
zz)|jVA2w56oNMhT0(Wl%B^0UtMe#e>eBYvDuGCZ1p|#MrqAx4WL}NrHGHMcl`vVf_
zfTt^C;JJh`m9ug>Ov#+-CA(pMy-gkagC2x4lD>pr8HbM#q4E|i8lgW+s%^%U-DVtE
z`uiyhVkMuG*4igxHDgX04oIlRpRY@1Ue14IGh@s5eKxX{(>VQ>+@(ks2c8R*tpN^d
zQ89@T=m6^xvaeeD)^Hh7w05tK?_1hTU{EFw<K+1TL4?O>kBhqp^s7J0S@-Kp5;fnE
zOv>_*PFtph(Xlt}BE>Qv={3nqOPn0LI%_90NXE|yM0`Mm*3*ryjGG6GYnL72(hGsH
zm_w)W887EzsBZV<384%BUM6Bw_+FqU3qt|-<#~w-vB(}b_?idFlw&j(ob>_ipu;%a
zCLzg9K6d&Zch0G4TZj1*_zn#8!p0Tqj4o6_*W0a33b%pGg8=rjK|{l*O#{={=&^Zp
z1N%dh8@02*@0wFar{a4g;Pk?ivN+EbYqcwrv{->8EZ(p01qaWLg&?fCpGyH|oZ6WZ
z*7KQW%V2O=Y$eixdwqmUBowo->@9Q#%qvGNDdoH-n<9kXkIV_`?h>i@nggim3OP(U
z4Z5eFklNZ&Cf(YNF&RM^&fHg%9?i#fhAcVXgdsmHoOgw^fh$a{SG#sj?84nSt)5}n
zZ&>=qbhgVV^S_>?IDP>$0(MLqoVLu`gg8pN<F6a%Z>sVAJbymAg5ecDwdx@t!0Cua
zN4Ww`d7ldHKt<X88H=4=d8ccgd(<XE{VOw(SN~(fT(sLNRuZY^kuS-tmKD2Y`7Uj+
zb{dzj`1{1)le1O_xgC;|L}01hmjSj_580?G1OLl{x0cigCg|gCTgOga!_MyGASdMb
z>$6ym#jyEhjD%@jsBi1t3OI*%=5%@9Q&3JVtUj-}%cZXTXp>n4h&GD87feQT3eXf$
z(09sBT_H8LvnPT_4Z3#_xi0=8p{<#)EXgl-=)O2RZNjp+0b{kE>0gmZXTEd`e#VnF
z0tzLe!$T+sz@|n3;EoSBUQ-34j1i5f)zdbC2#ZnJ;fx=&>oOVOwtmw~nq=Yf7kx#M
z<VHbW%~dLc(4+iK<&g^>GE62upX*edov%)5ibdO~QuE{<{**sz-f~+qQhpA<<;(dA
z8?o0ky7kL~(t!ciZ=jDI*5A;gXF}|nylgpv69-bGe;xN=(NR<pWFYm%IT^xMM&S=1
zhLBdwK{jKO?bwU6asa19pKnJ5gmRhBiRKBH@9|A_RbdxhBOhGpQMC6r#o8oL9NU{B
z2|yQM5U=|mIK69GyW569zo3+lf6*2@VUc$>k+AnMtY27u|5b9}a-#E1{->i7_Xg2X
zXapsEpQy50a($-piK==4v0+67-Mh92YfbxrB43%U*xm)vAK>z@<Dm?jVssX&o4TyY
zBgA5_D%F%6g@^kAX^}XD?cL8!ej&D~*i(@5$7JU6EHmk-`?vfsC+#g|zU1p%F<pL+
z8fP+zf+MfVETQ?TnKUd#Qi485sl5=>9g6{K-(svLn)uGeY1eg^LCoSI)GZMot;njr
zmyuqXVHWjZ;!|`b;hHjpp7|reL1Y0m&+wpC9k^`;c9L+;5276FRw5~3sVTQP%WHdQ
zm?N@T#zZTTKzR!n@SYPThW57zuNehlK4a>mTNHau4%+N*e5<cW%N8TZnbMA57AmDa
zlZ{Pi3tEWC#HzFWd?O?9DqI{9PEExX>%A-5h~T{Q$>ii5F|5>J%SA+IahpyO2?@#f
z6d2x?8G?eCc+^(i*x->2zVW;AL+uqaWows68D*M>zD~B;tM1OP=wKI3r|F0uxA74t
zv-EH8MOmlUiL-4c?;$Zq*TTM6{ezxb5y|C0$_$(&wPE;vAGL5R_Pm54*IduaoFTb3
zB^2w%O3Aey3%`ukF6)6}rsv)26f$MbhEgl8C{+FF-ZZCnOFt^~qm9m`);JfWjMVYb
z=J}EIsvKciNSalQa$)VxwY2LC?&{gGGMR5FkdAnLs-2AT)F#wL>0^IUd9X(@i(k7E
zXp%YjHOJIDykpC(4?*3*<I&WE$Vcn=zNVr|rCU;o*k5Bkeb3h)WWKqCtadGHIvC8F
zIw4lMxEl0CY=l}gi3`tgX5RDNkG*{W>(GxbShKY><Sr|xU~<n*PDN=~7VsA)X~&!-
z4Y+KK`%ve~gs3#gJ<Lx1DLI(Kg~fW+IP4EER`j{hjg3CP3X&hGBNlqzYJ;87n=+d9
zhA~A1VSLa~=%k|@6+4Tf7xw3MUC|wWoO#O^>f=~y;T=G!KGKmbuH&I!u0&+T(x?Mk
z-@3+nvq5jwxhy`xa7b=gLrpeXAuGh9ZC2I}oRDk#MB2J&gAI^FD=$VnS*97wyUqwc
zUHV%EXQUT)aT-7YTEp>Kl5V{S9&uk}U7~t6ZY`Rh5stOW3{3;ra=<6`0nwGQraAbY
z?#yM~mC2y=$E~x(a;j-|k1&L(ca)=f*YQDqudyv+V1ZwleDY^%Q=zVq0!aV$6oqQ9
zZI9#38=Bs$QazaxXJ{|EJ$U@Q`$lH2^+4Lpk>1pHx-w}JYKOmiSlgKwKB%!dcqE`0
zeV2xM{-gq2rf4}}C%-!W@4)O(I_I*|GQL=$nHLVKnhfV<!A%y}%d@pU;bJ9kClYY;
z7z_22(2o7JA@MP4Ew-bom+y1KY(+9z9Zh0j3KS*D-3;hEnT?&4{tZ147uMrx=~IME
z&C2jg<RJX<mnmZ&Ef>a5*yQU!Yz+cd#S?1xptmyXeEBH1Xm$9%6kxluaSru(o2dX}
z<tRCCmGUqnR-&9kJak_@16C)?y30wg^_Jb&#@W!lBK%g|`A3j9>JpjFEAh56Oz8Ng
zJ_*54Lxz0;4j0yy35)e7DX~Xeug!KnQ7zGMCFuLYM^GpD#M*ogwU^VEnzTGC#;)v@
z!ye4tR2(A`j*suyxdECZHGS?MVw))}6Z$>+<ZRqy3AFyPn#$|Bt^@DaZd=v<ywP}1
z@kVEHSW;lQOQk(+)m1ZyG29@UM<~q7uqyUD?(GTqaca80tS9B)6e19{QKC&@i5c?v
zx_SWaG(Lj&?Wmje?E}*zbbt$RB+3qmkD;E?D|=7E#$<SZwOULCT67&bBu^=qZ(MmY
zQG1z)0jtA;h50E{_x6v5{w!B}VsP{-BvE|vr>+`(@Bzuo`kf{=02$!2PkYDXj1`7l
z@0c(%q$3@t;MxU~AjuvXSi-Z2GAZvW;7@ox?!)UTxH;jC`<yEY#pqvTRezpmM5E8l
zT2z6Kz6MO=piyC{OBDH0d+o{qb~tT9b4GchM(uk5A-C`fzj_ck_uLgqI&Gt9kM4m3
z(BG-z#(KuV`{w_3nU{Qcs_3>efD3$-mjy+w5zL^_{`QF?jxpIB>%mM`?N<jp6kF`k
z{-oVes6DM~)PA$MPzA(Vui-1YqzVu?Q?5#zDFZUJ*CZP-gzLIr##J|iS2YJpp!V9e
zAY(4AAt)_c0+p`?ntHLimz8Ff#rliY1%Y9z-93EfZiL?%JgfV@^o>+rpXtPmz+xCd
z_p}{BzGd)~wdXyG@htjW=jSJ%Z%J85o8oA|+x;|KiZLw0$8@INo8T)zOq0Xqy7SaW
z%hH<ude#o(fc#ddY%Sytr-BSQa;z8gF@IAaC0b>(eHCDD+N>w_LBQ5ov#a}+aSg+w
zEt5?zq=qff?<pnSzv)aZSfsh=yMg(bNu0a%XJfcC=SG#6_Zs`h64eL40f$I#r9+65
zq{rHh6NW6nVE@@UIIU?n-2QnqE)4S9T1GY*(lY;jBX%9O_ES~9q#?dQ-Jv$`ksJsZ
zqzpHYjDw$Xkyb%N-kh4QDO23TKA^o?19ldnLw3<FF5udf4s>nG*ew$&o3<5pdWN*J
z+(TYiCEQCzo>e;jfezse%kNb4vi-n=+^*xYBunUFk^7l8pRr@=(WUPujiT4a)Y7#o
z+sxbOH~>!+m$;xC{|t!MuYTPIC=``aNj3*Wl4EMj+gl)G@nHJ2z*wMq58G+L91}XE
z^SXO(Nl7*C#IR8YrYrxRr=;zaPWTeAo&yN67cFe-l4V%2Gu5bN?CtPGZqr?nXBYVj
z#ce+)*t#U$&n|rGX9H^xx(gd(O3zw<)s-rk(U2R6d{}|%cstF1=1A~niqX{xGdJO`
z3>h$H!omAE9^pv;FDQ$P3(WdLy{;~xNVN`Fl<_B-0N%?v9;O;rYOm#OM`4D)!Wkzv
z-qXU04d&VpmeHIQ(|dpwrVQc%;-T+j`Bpju0Mv#=5h^8SS;o-EWXZooNJ3EtBu3%M
zfG9)oxW;-^eW9X73Xzd#WR;_w{>j7e{sNX6h5A!>q9#UdkoJPiJ^VqP{JqGgUuC&*
zvwPKduYKuSf3W~?4d_|=7$k|HXXer|ZyQ|AV>TChE<l2DnKNS;LN$t%0fonX(XW&d
zzEgf$!?cTqVUF4mdp?R%2)K^taDDGin|-q;SA3~%<5c90PrI1HlNPaA{1&YMv%fUf
z#`u*M$kiDLK2kFqX4%HOYDy&5YB)OqNf84JRS$7%C$$oFr_6WNiHJ-4MLC({9A(&5
zh`t*~Jt&D>apI6d1~XbNi<g%&uNrb?z73lo!Zp%9?LR?7s*W1Us2p`#x#MtA2BX{|
zxw1u8D+~WXL)tVxRy@1BKspD>MaHgbiOi(^rBu(e3(XC;f&)VjagG&8PUS5#s6h>p
z*qhxNgoB&W3YZi~uvPOwB#4uQ5TFCCRXyFbfM_g`SdX6bC$|B$SpPm))e{LQuPHwb
z4xGfLY?~grEW>|TT|7)H9j<%nK}R$g_ao8GP}%H1xHvj=co30L1N;F4gg0blCWe*K
zBE{eVt&3|SH!we?J+>Krm3m5<y*-*e2Uil50tu@td-M02y_VrloR&iUE=yjQu}~J)
zd%i^$=8ojN_dTVQCPw(QT^J|eW_`aH9{H$G6~pxQyUmfn=)YCAotZ<F<|3JIcV~tJ
z!WU~X9VSZ>Jg3iTVJnTS(_j&(t+$SL4_UQIW?d*%)Rz=7=RA{_3#XgpiS5p>H;Zh5
z?cUBTgT6G&GKwQX7C`6GCPG)R#JiNkx~eSYHPp`jEO7!BC>+(!Q%f8dXk#(@`Nn8;
zoZ=#{OCk$jd9>~uPIsyw4+X4tEJM`V;Xg`tod-tf8jB?P<_ztw5Y?R2chx87-g!Wg
z%A!%&p3M7dlDTH1tW#0FzJO!u^u2b8rsOHCOkFHy7NGy99WtG=26Lu-6A_Sn_@D$f
zh`5A%BB3`FqhD1-S4CWt&2l{kwK((UDukU0&GkFpuaNnN)_M;@{H@w%bYmIBk2#O@
z)&!^ny6an3fXL1w8RaYN5e;x<zmR~&mQy`=c?=EziI@{^D-nH07zXSB$29TdG|jfM
zSU*!<wgh|)Z8p4`Q{{)!dvkmavmWW6&A5;EF6<Rdo+koVAC|b8I~6Ntx6FS!i5o<H
zk;GFa*IOp&bm0&ti!-F1MoE*+pgpOm)mLv(jDORw3?bY~sF~armq})8u+q`y5QqL>
zQg%PnRKA2amPqV?kID026LDDed=HitStm9JUDNRtG=}OUwSO&iFOt-Jk>BM?#G;6q
z>RTHmh}$I$@D0SFHOksQy!Cluqq2ZOM4ZE{y?VFWMtp5DkaEQ&I`ciehUZR<m0I{2
zhtt{Zx!k-+oDdGVwop^e^Qm-<W+U#(H7hq{><-%~qw?VdqjLtjH|ZSafui-Qz-X^F
ztQ=BPxDUlmyMjmM%p#S^7*=cM7zJ2u=iR97ceZjgI@5E+gsQT95O3pE4g*jutM#Xx
z(EaW7uFe)%q@wJ1=5%Iem|vt+cLs~OA8VrkZ)sOoe0iM7tA`~ob6l7mj_WPV7+Pis
zzxz|Jm9T>b>4W#jWNmaf22{o?%{S*uLK}*8tk0(eWxjlEROXlTF(UP}4J=!1Yr`2#
zhx|`@oEOa*|2<e0vWgTG9z$h^p|{H4=My=rYosH%M#mQR)B$a8n=a&w?<^heO#Nk+
zdQER31sWbK4{)OJE6&<&R|q3A;MCi+t5(u(VK2}3X;F#mW|ka*Vsr7^X9zw{8cS;P
zGUDG~6wa7@UhTym{fP>7uWY3-&Dhk{KLveL*9vZp{(8E%bsR--r2gzFh7qU`!4FE2
zDkjuZM(8><s|8)5P?7h=X2CMQ0qGw-y;7N;>x4^Zs#C?_Mg$}n7#f9cLiMoYfjo}O
z7rl!b%vu5uytL&QL(=SpG}&E86(0n^W!TeDex`KZX+2&tMr`abMacD3$5Pz2y>Qky
zbZVonl|}vHy>R(2G;lucB+utcTI@?+$WC$zGE$Z>>{Gq&uw+kO7kgb|f>DxSh4vDq
z$XlY()IGameSzA)hK;Z4C5f7fR^$I<T`B6#BNGS{ZtWqI^1Cd>G1@v#^UI`%ofb&5
zSzU{r1n{(vH%4KVE=0iGd-;SpV7snSh_#5G1j4=-qe7H_ic2OlRYixjm3_$}eVcOz
zwdLit;BZ}`q5v`*HtRM8#YpV5j6B@h>`HO=F2Y4@3?wzidRo6hYU}<I97SEiqs*u$
z+GNBz)dY)9tv=>#s1^H3g<Touh;y?omSG>jE8!S(T7DdKAmdH=Y@PQxrDeePFzbDA
zfejIZx<3Q?O}WMC30h+-li1}{{w(2ISq%sBqxDePo|Ong9*xMM8=866JsJ8#lY}?F
zQY5Y8e1nxEvXU=D=beb^PG)<Q)*|xt;r=|;*D~-*a{<%G(9olDiJg!jO3oN}4yhVE
zh76Chdedfeq^iv`mx*~P?W;kmpuoZOrQZ!Ugf1ESFZ8pO81-hly+L(hW=DZV@IGz#
zPQXxVYRj9hGnCkl|52vh+aA(rY@F0d`{D$*)1<M*)%AflF)(##K8oSb^9iq`^vjX;
zhdH*7B@`QH)kT?&_+R2?H)qKQ)34)eoJe(*dRP+oi0Yn{FP@&$X^7NEglqX&v!{m)
zv@J48|KTUdHYPBbz}S!Vabw6MIu9fND*0CR(k5}4dTsR^%xF=H%1)-|9mDd{2M889
zb?)Q~vSzItAg6NYu*GL8H?>mdW5#u!{?|5~o&g{B9j?|={gt&t2K78*ecCiz6C_X+
zWnXRD*cc5-rK0X}g7L+w%mF)&x1G`Elf5Az6Xp<Q-F+YxzB(fF^};Ymeb3HjSLp)g
zv=^Lw4r%>ERW)z5L-|VHkxAeY(=#(MNXYejUnD=!Hzt6bcKLP4d<DHc&c(q&?KVg7
zfT4w4(nHW^LP!@VTz8GSO*vxib&VqG7adypTq?fzSTgAR69louw`#IJs14g(AQ2~e
zW^%ur*bfW=J=1sgCuL>VMunwRiE%6};70Z^E5DG3G9yIVTuX6eq&)7UU-ZcKEyMz%
zT6yNR7Ootp_bPYrTlm<Mj1q+WxeRRlT`GxYWB+hr_VS5L@lCb!xWUxOoU<ja95N<O
zq|+VkqX|8O{jzRNj+mA28J9aDIy2w$*Yqw)?|-;_|GQcp=`8+B8WNT)->OFJ5Z*P6
z|4=50gcM7j&@oT+A~NOnpS;?gONjNZ1t#F5YyeOsxq<a*CnRU{a@!2QQ#f#QA$y!b
zX|t2fF7Tp3i40R?x4c;Om1vwQKW4a+=ePMWda)2vr1)9g*z?gRnt;owjnGu}q<y)q
zR$Xrk-!lUNBe&9MkurAI1DRyve0Lz_;u>WRr9cg)zdH^oCKrhBg&pFg0DObTgPk`t
z7UnPz<+yph$4WW$2&cXfNldI<Uin7dIK0|BtfDb=?C7{T&Y2|EkavGMhk>x4z$H2G
zIn<E7>h_oG_~^QglVF>@T1oMcb^+!5ME}k7w|>)ijhKgI*iwMaKI166os2su@*Xp9
zg-T4Glv-zP9WUCQW3P?d9P0~vBDt0K8R4zqY)r4|!<xb??584$OJZ>fsm%N}Sj+cm
zvNiu!{{5Igzk70{C#5RGl?%^ZbBv4PFz@TSzOGOTJQF_EB14x7GWP^C$*))7E0XV|
zOb-SXLF5O4%=7g2BJ}SD{ePQ)0#7WDbI&+Cv{Ra$w;f|Dl4M?CAZ|PC$9Atc$-+nk
zH!x|{)tJ)|?R6@;TYPge$ak+6>`txFc3lwbKF93}s>AP<B(NRSJb5sai!IXSiXAxn
z)md4><P72kfnHOWrc*xA=Ufaa#pXSctOZg67G88Ie7)!vh8$#^hXJvk{fQ&gY6lS&
zL5E8u2gn7?kUa$|3!~OE7&d)Cwql!Bwe~f`WcfMF5e3PTEu*VMFW3_SO<YM;yQRkw
zWBf<`O6+Z{vOuzyvzcgYK0DicRo^_PmDoZyV1zLK3K8%N<pkuA#5~Yr0-rZTRSgmR
z<?2Y22UuVnkJGdVoj#|}K#<<(e8AKeW&ccHs<?e!G=Mnqp*spXzRHY(G-q%UalrK)
z!W}*;BC@2JLtOIJu#X{|uE`E%-^Uhz9~fBG$MqeL!r;k&>)Bl60S*-F#uYv!y@s^Z
zW@O%M5pRn#DQyD1soj8?$0^G$bhjoxrpt1DiJy!bDcF44=1zWn?cv9gb2&ngIs`K|
zczKx<TiNlD0O%+bE;la#qx%B)XwhHvCbz)uLWeUpaUJ4~Nyd;_Q@LeM#Mc{7@Hoxh
zzZ=jw`C(Cz^118H3d6(=BBgboqxw5F;@_P*m!BN|UQjVlgSuo7l}5~<hT5hou@53v
zaUK_7GtVzj;lViXQ{#&R0oV0Sk3FNF*IVRLbD97^&Kch?^Vbc}gA7$u>$?WiwXAg<
z>8+yaXU4=UeV*q{h2xQLsj-qo&&$fo*NmgHaH0e`=@ifLpi|!G5@8XAVxeWR;)5-|
z*C5@eUU|DkroR(3i-m+P{JBM~`q6`qN}w^w!3Q`Y6$Pr@Ce4ADy}GPO*kVk3xWhq&
zAgT(n!icJ2+?im)ZuI5%nB_)eULh}whdvAqlu`cs-q<z0;sT7aF{vK%3ImI=9@?Yo
zyx~?^c&Dz*HiXv|VXK!JSSv)#el5li`*{-WXRMM1InSx>gh8M~N#wcXl*U*Q=3_7;
zvoU13-6H1=O!cniE*0A5_fN(PaED1N{|t1cJGJWqKP2hopv9-~7tjkD(lqXK?CH%R
z%nPyF4o}}{SuodP52@5}U$d7<GTDngUCOm&zE2C5wT<pf<gNm;<PTGSQKf{x%-EvO
z`!8<0VHuwDG3Zxph{>wLm$2AYGB$AH<J!r2&+iCu`O9fdBxA9CB0cBr<?Vp}B`hU9
z`yKpa`=VEbq0Iyc&*3wik*R*jju1+Y73m>{Ce@Rd%Y1Prw$Z%sw{H0p7=^p7PPiP8
z?V?TMffwYzD=cpy_90ydz?t}#wS+G1wn&`qGzpU_prZ3$Z);Ba1JboBTG0Kq+&@hg
zvQyhE|Kn~Rm*c!?>Kn$JYD2g_?_*{{5LYKlME1zW1R)&d!ciIKK2)vxPn)5&jPr}s
zJ-(_1M8ddfhqwAoFM60vrm`FbzBY26+L}L-sJ0)6wKChN+oM^L0(jJng+<e-+i0r-
z#_)iGvp#iMv-(#d1;(cfr$biS(>#WNlNUhK@=rrPCyO@b`dzZx;j$tx@^J-h-@^1$
z0uLqP-o>4>vM7^0%<6r*Fmf7nl9%1avU$L(tR?7vm9jIIL*r-b{4u2@P}sn7U@15v
zEjG4Q{wtl6fL###RfbI`BI6C&SqNJpX$0UxZW)vQ)a#f}{%IaKphWn{u@}^oi8Nnl
zXA*o%KMJnf!-t+!e;V&R&(dep{Z1)iX8M`O*12@*XbGTm#D8FXTTv~J5fq}YQ0a(M
zx!i2*cG%w<OTI@xpH!_Ac^Y!%nc~#W;W0l2@R3CXKeyb-9`LyUKSvV%4bRwpgVtt2
z&0ovP(&7)y7&v82A8{?nBs3TdPmG<7jGdPQd;{1+9<VOUB0$<*Zf&BTXU(rbifkjQ
z6j|@+WY%R{nMFP_|3${N`_@^js;A>EPnWV~h1GZ*aorW0*0_#%i0eMW&i@zzk8r=}
zB}stG&!``w4bG#0safRoGUdr&`aanXIQ(LL_=0Rz9wxyJbxmS!otr)MUlUOqJ%99L
z(=KC&2o(MVYh___cMSVBuPd)9LJ%oVmRCa91JU{Nn|L|oEEKLhTugX6;G2lON)ml{
zLCcL8PZ2O8;oRLNq(@=GEc%I)Q`J>PCCfxkm@!Z4C(!DY8i!^z;UW4-K6fjtsh#{T
zb?Sq9O5zEL%#>~=HEt)(eu~BeO$4_Q2gyyWQ9_Osp7uEcg*c1`gvdMMAdwXHd+XIE
z!~|7u#@~TIlF_$T1sA!($=kDO?J0<^&_3^=Co}(w<ZobXCFK6a3crT58WH;A7Ubj8
z-H;2~&V9<}Up3j&B1m6-1)>Vtyr6N2H~h-86jszq%3}%tNOM1NM@Lzy!JKVXfqE=E
z5c$;fA7)c4RG3m^vB{@FY3!Uee$Z&C$-?9r%x8yq$Zs<!oQw8M*9i5bpdmst*)ZF=
z!b%mmahGsouqX;B2Ae$NeS)QZu`somL`Zp|Z(qA%#w}3`ww#dp@h6`u*A4J4R<8m`
zCVE8O1x<xS$G!o6&3*1^^C%dYo$F(}Q~_i@7^r5D)Ap~BRQlY9hLioZ@QghbSb`Az
z)H6J98CnARpwQf$(M$vo&HhbNYTwE~^_r+8IeRj%Vir@;Z4BhDPMcOrca^9Jo_Zan
zqMt0v=2j;zUp{Tv)dg_iq;ezh$Z{mhNIh(Bxdg4$YaJxU(Y!H3c)OmHIvcK4;0Ag#
zmgx`%k^<-DX?Cy&tS@k^CvSwN++uMD;C_1cZ{0p94l-2tDZwgkT5BE35?`VE#Dxzm
z)Q8~nrc@vJ+MRGo;2mE`q!K=A$=o8bm3;9uRE?`scTx?omR=vn#3~he$vv@LXn3O*
z_;=!=VFLOu!iVUySduiCf%4;-MbI2$07|9o3NP)H*aGuZ5F2lEaKM-|CMP33>>Kea
z9qQ?LP_=Ty06Fug;J}ZRHS0Iq*yE&b^?#?WKt0;HYTc^C5cG$Bd03t#L;XBNf0Di}
zg&L>cH-7II^@w#}+)b0Yr+un%VRr1{5Fk};UGbh=6>D%DGW<HQ(M2Nf8{tkE`sN*0
z^!rd$DM&io(4E!F`2XE+sGU}G8Ze|TBb;-xq89sMIo*$?7CY;E+x6?=qUh-ntuYEY
zre2BroDP`JP#@i_@;L}La4@rz23Cgpi*d9{@n%dn<jlQW)e8G!ERsVysYY?en=uu=
zZ*p4Q2#go%K5Mi;PKOu+1}>90e7#>&h2FTVV+sh%Z!rDKL1nSCdzHQ&4<)ULZs;8!
zIhnme%N+*#bqf`$Yd|ekZ3sl><;2;_K*pO$<BFzCXt-2S62UbA{?xw5MJr*F?Ce{R
zzKcVJTR4{&uW;Zxo0mpU8{Ay)izGOCC!Mwh*||J2eUg9U#PUfg_FVXZSJ-KWtq&TU
zS_aJ4S*y^4iCpIL3}U*W(r4IDJ`$IrmN+JOfS>9L5~<lAmQiezquLg&Lg)m`p|+2W
zn<aE85HnQ72R`aT{&Igm?1JcE7{=`9MCkhf;k`tzFj3xOgvjr83M!mger>Amzpb5A
z2-Y;@bfcO=YH#7^u{mxW30?9Gy-prxT<OHZjo`N+Dvi>cY8zvWJw000it*;4|L60$
z<kw;xiprI|!K@<2B1<q)@s(TeiI6JW7^_tG6%qp9r#{#$f5p&mn)b!Z*}$Vr)!P$m
z4?jB?f}Ub)t%O!#Z$I3##o6u^9?iQO8pufLF3zt$z-meIb>MGH8r{u*W25o86)MW7
zZcOrhP#qL?;52*p%a#M!IG;nXo|}-^!3vk9!J+POr29dYJn3`tNv}GqDa={?nOzEg
zE_nMh^2<US(>;X)uReO??@euGZcCS}ZxY6SuN@~Gi8|+Rc@284QFQecK0QZ!vYHZp
z%jZ_k#ioA&a*8qx{<zS4^*z`=2UP>!VR6vk{geNzk>rCXQmxnZzZj~~AETXiiNoDf
zcCcRfEU#U?3_4KUDtTo0iIi1R?_%&hHNBzQslidQmUgKs=11h4%|(8~+VT9`chlY@
zUnySBUb>PDeuYmS@GLL=d~C|u6Z~EBDIagZ4%>q?0_@<4@a2-FixNlW>TicdR+r92
zZR(EWBweu{TIX%HBh#CD3D$iIEg%Ou|6XO6n|=;0YCnj%w6lm7PkZ7gI=efeOnzDs
zeq`!D<a5m@seALkk&#KEJyL07XajdL@(E*i$}4=*Dea-4(VKo)$AFeRh{q1*?h~8&
zuqx6c4}C*&{~NoZgEja&jniqLkKsSWZK$q#^KVFUPQs6`Ktn<#jbQ%Be9ZnJke5TW
z=e(%;4|zQ)H)})xy{V`7KNLrVHHk9*OP;!Ukp_Z=2IYv>7ImL6j<JFq_wB$T)gzw<
zC@om~PpoxvmE_flLfcD-SB@8%_RJ;E@>B>61WS65S8>ZSMVA5^Q8%K`?Djt64K|z)
zy#={*cr4&CtkpSr^DW3~FhBF|h0G;J-@~DLrbUkM9OMM;S0VPphEY`i3(Y!#&Q1F)
zGJe5NQ3<~rS$g|!^2A5jl&9mB%csUHE>S+WE?ucwJk}C^(c;<__zey0B$qNCn^d+4
ze3j6wh}3NKc26%g{FQZM%${HQGjv-1+(g76=9Q?)FHvKvF_!>K#fx(_gy$*cpnCZ)
zv?~M`Juo@oWx%B-dm9C5F?%ueh#H*mCR!6S)wam_!}k@e%J^v0SP;wQKd(=5I(M&=
z!r<AP%+-)Pvr(oEtccfE)!-tM{#rQYQiHzGEy`%7T@d+SO_+>~Hn>Yo9zX-@yklOS
zu(sV6QbT<jS<8*qFziQH@rywgxh;9kqaoJj$)=Uu&szo`!LyM6bh<J`6J=?aN8Pm5
z%abmpvMl;?&H3@>_b!)pw?bYyByZ;D#>EUzjBYNKL!0xDO|hECTx+1QQ~F9NnO>bf
zk~4iNpm1#L6teY!vecnIcBC{wn3WgSvtfa~Dtxnf+qOd?I7R(<g61D@{27s#Q?>Uo
z^ny)wf?M&G#KAWmF$0aMKgMg)hHj<mp1TJM|Bd)qskM*28PXP*Fm=ag_P3DhQ(N7d
ze!Fv5b3_`Sl<x((v~yeG5cZKz3!wW{#gIqL;eX`Eg^zY)UxlC3A8M)~KVOcmPi@5{
z*N?1Sq)q2tNEmfX4t#{3E`B+BP751Ip7L9!?v6VoEVa5WBo8Yu`AxND0mr~E!vgm3
z&ue3bn&kF3Kx1(qZzQ!v>`ed_whUFYyKv1u)$KMjE&L|DkaZ5=&p*aBcOyOZzU*6v
z8TNcz8DbBq@Y=l1+#XljOj<gd@PO`D6L@_p&8PTMeL2!Td2^YeRQ4dRb`;S6s0?^a
zeTLR6*kSGA=6;NJP7xy`9jq;@JIZaVhqH4oCQt*LD%#F#E8T`fl{bYU<)N^tc&BRp
zv5wQ`Zn!vE?aLq&44Y%WA`Y#|qn4>gypLCvy9d#(iH>;v;x)3GvaD+!x-lTE0X8a1
z(~lreDB`q_YS8ZuH>>Q;N>eIko<&|4K70ZkX&+i08P>tNLh&zy#w_SQ^Xli3uOeeL
z-+NQVGo7QG3gzc2nq14hzA2Op!!NK)-9N2He>zo)yyXUYl~9q<FmaUY=|+tjcv{oo
zu80skTkTw+=^Bzhtg?qG707L4ey(0;)U&*sg%#(fpM}1U)FmvO5}tjxR&{LZNOJBY
zCV3mr4{aa!(S&~+i1Mm#eD<kkw?!$ub=2%r_AQI^fSDYo>j12%<K`2$guJ0>!KdN1
z@i#&Cz?;a1Kw<Z-x;N06#<#0gPYQ?AuBvB;(Ph$xWw%nVP`BC}k&GO>-v>A`&hOq#
zo~#C=8q2|dO+pGKvQxp!O}o2FLDy$J(4N}>HJdylqqd0z8r(~H{im9thREwnK}|1b
zYOGy_AwwHvnbtsu+|F}d&qAMhJMD@d>4+AbJx6`Ndhc=odSF%cL%?gSpwex*9e<hh
zDCN`3Q8zy0*@wLG3YVeM$~y0rl)IG?1N&oUUMUWYo&%iZdn7x(Ix%K3yqmeGa}%aE
z!AB}N-nHie<Ko(~<V(mA9Xap8(tFTw+Q*y?!^bv`HNE$0&>SJS$&VbWF2kL+<lo77
zLp@ntU&c7Xb#mjtBK&AQ!;TI)nvVg0{CG?Bi)1qO-EgpKS4z1z@ES9wA^lFktNNIL
zfaaK`5aj46Y)bk#=~Z6fxMteQqQP%Dchsd&u7DK^O;h|YIK$5?N<C=T_#Tlz^Zh^N
zE;^JapWy%Jl_1zVtot*|v7vYLd4)7TsilhayY8ko27e0cI6IwNHKMRb)Frl3y@ku6
z53D^(4fp8WM|<6jQ3B|B-!dh=4?oZu?C`>2tm#9`BCag*mDLOMYq__*u?*=r=KexA
z1t|<1>8M7#CG!rEmVQUR&w=cy(A3{Dj&wv!Igeo~Qt890o7QKl$-CV?>wdE;?YK}#
zI!_42I@+{Wj-rxvBkC9dWQSR5Xyvgbcdy=6#ho9wt72vY7<Hf;_7!${bK__*iadfe
zeK~ZbX_>|9intI#u-DA$$#cfzd>8!>ZiGJssp8N_We&g19!-fW9`im=4WsoACm+fi
zo~c@dO=I#R7`$6!%5v%_)_Y+}VkxDGz9l9HHU%s6xD;Mky93r{j1jn=Z}eEMplcs}
z-N0DQx+A_*c`hVz;kcI@>{E5+zx?x!<M3lF>*3DxqmI`UWvP9YTtVnjIVmRKEzi7x
znPbxYmhY}E?`%fzaZ^Q3#U>A**S<<Sw_8ys_8mwWE4gIjF27QvKLWo#uz3pN;}IuV
zx<lEY>pFC7$ZSJ(LhJDuoBzbrGBdaT3$O%F`~5sKhE+Y9>n<jzn=!SSLpAdnMXNq@
z{^>X?M=2(HjOWdKw#un<Bx#(@{E<0cI@acVG<iz<q&PHl<fbpCT%p(0uP}Ml^pAD?
z^>5_~)Bc$D$Xv5A^Dspx$o1MU?W4+M=5kCQ+047!d3bG8$HTNYW^TY-M=<%&j5*AB
zMaRJOuT4MA^pi~AP1~~Zq`Y9}xy;<GVn^lEduij)eyI;(wr$4araxFT>NmMa=b;(T
zn|W6gvx;Xk&N5>g)0QdD%=HI5m!o}DdCh#and>tDP8s!@>n3*IO8cX}jBQtJAI`M5
zCZ0^+yj@<!j7Jn7=Dbur*<7LhH~kZx6Q-VZ?&|#1zL@zkmCw$@nl`!Jc$v?cdNuty
zlPk>i4%MH{CGBDo)8A5E*|}%Mz3Mb?oJ~8aIw>0Mm^RYP-<j7HU&<YJzEN>*`rdk9
zZMBZQ*|zpEW7K8l)zn6sewE2PX8DBCzM1ni=a3m+=x^<}IiJkfL-lIrYI>~a%=v5j
zSf<~k*fcS0#w2!swM{dJrsLUbls98EGhS9MFxP3!`DXevTE?`srjKg+on}mL&I7ZL
zI)2J^=GdC^#GJ=wT${AYE6&ZiX6m(_%~c*q8-JTTr#d$C-%%WC=GIJqJZU_uvY39T
z>dMRmn0Z0%gUV>;2h14K%-Naw0mZQy$7tWxU(kCdznQV9V#l15I(JlF8z*MpRUhU!
zn(dmoEYlCy_O)Fz-)iRmOs-Vjm^le;)5N2l7c=LKj*rO&rVpZhRe5xdt1oTFai(9O
z7}LMCj5&w3ziFeLtZ`5I%FK-@Ud&ihdC1HUn|T?Nf6chVjQ7l3l9nqP^{UM>=arVv
z8ujX!n(>s5lg=TX!>UV@e|4Ul?JNFG|D(tFZLVe5@pRG&@~9cJ*y|srk7aU(>R%6)
zSNTJ+ZRT|Cyk60$%k-sm9x2wfUd6N-JDYkkxy6iuGe$i{qdvufj+yBn+CHPH8<WG$
zxnj-(Gf$v4Eo0PYV<}<WQ*JWl);_h%$=SI~GuBrwuybbSyfyt5mBlQlbHS7$X|!RE
zlVaD7EA$yN2VmM{Gp}XN4Rg%2UNgQ|UrBMN^Gw?_$H~N~j&sha-^`zx@vhcw=A)Fu
zv`;#gsxLFnP8w~P7*JWw{FK^A9UC)8X!_NvAAQF3NzHtQ$(JgV$&;!Vdp$%s%#=^B
zE0>rtv>C(cxGFx1R()zCOkA1sMf+o}zi6LT|7JeKw6SJR$Mk{BoUMs%(-x?WP#l<c
zRPBuIZz}%G`D=2%{x*44$JQKY#giGw=rxlA?D$RXzBz|fXFb++)3&JowZA5A6@#W7
zYG;E?AJxp;+3~MwN3;(*7K+Ut<9RdgF#Rf%qs@7eGp-AFCsAL|<M_Mk{Z8^cKIz1B
zPCWG4&p+|>let9ePVV_<zUa_1aQ$q@dnMiJnZ^GefBezYtb!>mRQ1~Ci#?~FB!8t(
zJ>ywtqIUUp%E_lb=ah5w*Ne<wy=R|%uKbyN{)x}hM{*~h^nz!dc+RP3o}mq%c=iiV
ke5U-!o_*puPe1*{b57Bl&qR;V7w2#PAKrXF^*$^E0LH-3uK)l5

literal 0
HcmV?d00001

diff --git a/model.Rmd b/model.Rmd
index 2e7d8c6..3c9911c 100644
--- a/model.Rmd
+++ b/model.Rmd
@@ -5,19 +5,21 @@ title: Model
 
 # Model
 
-Models are one of the most important tools for data scientists, because models describe relationships. Would you list out every value of a variable, or would you state the mean? Would you list out every pair of values, or would you state the function between variables?
+A model is a function that summarizes how the values of one variable vary in response to the values of other variables. Models play a large role in hypothesis testing and prediction, but for the moment you should think of models just like you think of statistics. A statistic summarizes a *distribution* in a way that is easy to understand; and a model summarizes *covariation* in a way that is easy to understand. In other words, a model is just another way to describe data.
 
+This chapter will explain how to build useful models with R.
+ 
 ### Outline
 
 *Section 1* will explain what models are and what they can do for you.
 
-*Section 2* will show you how to use R to build linear models, the most commonly used modeling tool. The section introduces R's model syntax, a general syntax that you can reuse with any of R's modelling functions.
+*Section 2* will introduce R's model syntax, a general syntax that you can reuse with any of R's modeling functions. In this section, you will use the syntax to build linear models, the most commonly used type of model.
 
-*Section 3* will teach you to build and interpret multivariate linear models, models that use more than one variable to make a prediction.
+*Section 3* will teach you to build and interpret multivariate linear models, models that use more than one explanatory variable to explain the values of a response variable.
 
-*Section 4* will explain how to use categorical variables in your models and how to interpret the results.
+*Section 4* will explain how to use categorical variables in your models and how to interpret the results of models that use categorical variables.
 
-*Section 5* will present a logical way to extend models to non-linear settings.
+*Section 5* will present a logical way to extend linear models to describe non-linear relationships.
 
 ### Prerequisites
 
@@ -31,23 +33,41 @@ library(mgcv)
 library(splines)
 library(broom)
 ```
-
-**Note: the current examples use a data set that will be replaced in later drafts.**
   
 ## What is a model?
 
-1. A model is just a summary, like a mean, median, or variance.
-    + Example problem/data set
-    
+Have you heard that a relationship exists between your height and your income? It sounds far-fetched---and maybe it is---but many people believe that taller people will be promoted faster and valued more for their work, an effect that directly inflates the income of the vertically gifted.
+
+Do you think this is true? Could a relationship exist between a person's height and their income? Luckily, it is easy to measure someone's height as well as their income (and a swath of other related variables to boot), which means that we can collect data relevant to the question. In fact, the Bureau of Labor Statistics has been doing this in a controlled way for over 50 years with the [National Longitudinal Surveys (NLS)](https://www.nlsinfo.org/). The NLS tracks the income, education, and life circumstances of a large cohort of Americans across several decades. In case you are wondering, the point of the NLS is not to study the relationhip between height and income, that's just a lucky accident of the data.
+
+You can load the latest cross-section of NLS data, collected in 2013 with the code below. 
+
 ```{r echo = FALSE}
-heights <- read.csv("data/heights.csv")
+heights <- readRDS("data/heights.RDS")
 ```
 
+I've narrowed the data down to 10 variables:
+
+* `id` - A number ot identify each subject
+* `income` - The self-reported income of each subject
+* `height` - The height of each subject in inches
+* `weight` - The weight of each subject in inches
+* `sex` - The sex of each subject
+* `race` - The race of each subject
+* `education` - The number of years of education completed by each subject
+* `asvab` - Each subject's score on the Armed Services Vocational Aptitude Battery (ASVAB), an intelligence assessment, out of 100.
+* `sat_math` - Each subject's score on the math portion of the Scholastic Aptitude Test (SAT), out of 800.
+* `bdate` - Month of birth with 1 = January.
+
+summary that describes a r, like a mean, median, or variance.
+    + Example problem/data set
+  
+
 ```{r}
 head(heights)
 ```
     
-2. As normally taught, modeling is a conflation of three subjects
+2. As normally taught, modeling conflates three subjects
     + Models as summaries
     + Hypothesis testing
     + Predictive modeling
@@ -77,20 +97,17 @@ head(heights)
     + model syntax, which is reusable with all model functions
 
 ```{r}
-earn ~ height 
-lm(earn ~ height, data = heights)
-```
-
-    + save model output
-```{r}
-hmod <- lm(earn ~ height, data = heights)
-coef(hmod)
-summary(hmod)
+income ~ height 
+mod <- lm(income ~ height, data = heights)
+mod
+tidy(mod)
+glance(mod)
+augment(mod)
 ```
 
     + visualize
 ```{r}
-ggplot(data = heights, mapping = aes(x = height, y = earn)) +
+ggplot(data = heights, mapping = aes(x = height, y = income)) +
   geom_point() +
   geom_smooth(method = lm)
 ```
@@ -98,99 +115,91 @@ ggplot(data = heights, mapping = aes(x = height, y = earn)) +
     + intercept or no intercept
 ```{r}
 0 + earn ~ height 
-lm(earn ~ 0 + height, data = heights)
-lm(earn ~ 0 + height, data = heights)
+tidy(lm(income ~ 0 + height, data = heights))
+tidy(lm(income ~ height, data = heights))
 ```
 
 3. How to interpret
     + extract information. Resid. Predict.
-```{r eval = FALSE}
-resid(hmod)
-predict(hmod)
+```{r}
+augment(mod)$.resid
+augment(mod)$.fitted
 ```
     + Interpret coefficient
-4. How to use the results (with `broom`)
-    + tidy. augment. glance.
-```{r eval = FALSE}
-tidy(hmod)
-augment(hmod)
-glance(hmod)
+```{r}
+tidy(mod)$estimate
 ```
 
 ```{r}
 heights %>% 
   group_by(sex)  %>% 
-  do(glance(lm(earn ~ height, data = .)))
+  do(glance(lm(income ~ height, data = .)))
 ```
 
 ## Categorical data
 
 ```{r}
-smod <- lm(earn ~ sex, data = heights)
+smod <- lm(income ~ sex, data = heights)
 smod
 ```
 
 1. Factors
 
 ```{r}
-heights$sex <- factor(heights$sex, levels = c("male", "female"))
+heights$sex <- factor(heights$sex, levels = c("female", "male"))
 
-smod2 <- lm(earn ~ sex, data = heights)
+smod2 <- lm(income ~ sex, data = heights)
 
-smod
-smod2
+tidy(smod)
+tidy(smod2)
 ```
 
 2. How to interpret
 
-```{r}
-coef(smod)
-```
-
 
 ## Multiple Variables
 
 1. How to fit multivariate models in R
 
 ```{r}
-mmod <- lm(earn ~ height + sex, data = heights)
-mmod
+mmod <- lm(income ~ height + sex, data = heights)
+summary(mmod)
 ```
 
 2. How to interpret
 
 ```{r}
-coef(mmod)
+tidy(mmod)
 ```
 
 3. Interaction effects
 
 ```{r}
-lm(earn ~ height + sex, data = heights)
-lm(earn ~ height + sex + height:sex, data = heights)
-lm(earn ~ height * sex, data = heights)
+tidy(lm(income ~ height + sex, data = heights))
+tidy(lm(income ~ height + sex + height:sex, data = heights))
+tidy(lm(income ~ height * sex, data = heights))
 ```
 
 ```{r}
-lm(earn ~ height + ed, data = heights)
-lm(earn ~ height * ed, data = heights)
+tidy(lm(income ~ height + education, data = heights))
+tidy(lm(income ~ height * education, data = heights))
 ```
 
 4. Partition variance
     + Checking residuals
 ```{r}
-m1 <- lm(earn ~ height, data = heights)
+m1 <- lm(income ~ height, data = heights)
 # plot histogram of residuals
 # plot residulas vs. sex
-m2 <- lm(earn ~ height + sex, data = heights)
+m2 <- lm(income ~ height + sex, data = heights)
 # plot histogram of residuals
 # plot residuals vs. education
-m3 <- lm(earn ~ height + sex + ed, data = heights)
+m3 <- lm(income ~ height + sex + education, data = heights)
 # plot histogram of residuals
-m4 <- lm(earn ~ height + sex + race + ed + age, 
+m4 <- lm(income ~ height + sex + race + education, 
   data = heights)
 # plot histogram of residuals
-m5 <- lm(earn ~ ., data = heights)
+m5 <- lm(income ~ ., data = heights)
 ```
 
 
@@ -216,19 +225,21 @@ lm(log(price) ~ log(carat), data = diamonds)
 What if no handy transformation exists?
 
 ```{r}
-ggplot(data = heights, mapping = aes(x= age, y = earn)) + 
-  geom_point() +
+ggplot(data = heights, mapping = aes(x = education, y = income)) + 
+  geom_boxplot(aes(group = education)) +
   geom_smooth() + 
-  coord_cartesian(ylim = c(0, 50000))
+  coord_cartesian(ylim = c(0, 125000))
 ```
 
 1. Polynomials
     + How to fit
+    + why it doesn't work with missing values
     
 ```{r}
-lm(earn ~ poly(age, 3), data = heights)
+heights2 <- na.omit(heights)
+tidy(lm(income ~ poly(education, 3), data = heights2))
 
-ggplot(data = heights, mapping = aes(x= age, y = earn)) +
+ggplot(data = heights2, mapping = aes(x= education, y = income)) +
   geom_point() +
   geom_smooth(method = lm, formula = y ~ poly(x, 3))
 ```
@@ -244,17 +255,15 @@ ns()           # natural splines
 ```
 
 ```{r}
-lm(earn ~ ns(age, knots = c(40, 60)), data = heights)
-lm(earn ~ ns(age, df = 4), data = heights)
+tidy(lm(income ~ ns(education, knots = c(10, 17)), data = heights))
+tidy(lm(income ~ ns(education, df = 4), data = heights))
 ```    
 
 ```{r}
-lm(earn ~ ns(age, df = 6), data = heights)
-
-ggplot(data = heights, mapping = aes(x= age, y = earn)) +
+ggplot(data = heights, mapping = aes(x= education, y = income)) +
   geom_point() +
-  geom_smooth(method = lm, formula = y ~ ns(x, df = 6)) +  
-  coord_cartesian(ylim = c(0, 50000))
+  geom_smooth(method = lm, formula = y ~ ns(x, df = 4)) +  
+  coord_cartesian(ylim = c(0, 125000))
 ```
 
     + How to interpret
@@ -265,9 +274,9 @@ ggplot(data = heights, mapping = aes(x= age, y = earn)) +
     + How to fit
     
 ```{r}
-gmod <- gam(earn ~ s(height), data = heights)
+gam(income ~ s(education), data = heights)
 
-ggplot(data = heights, mapping = aes(x= age, y = earn)) +
+ggplot(data = heights, mapping = aes(x= education, y = income)) +
   geom_point() +
   geom_smooth(method = gam, formula = y ~ s(x))
 ```

From 53df28b3ee93846f165538e5c420251f5fc55669 Mon Sep 17 00:00:00 2001
From: Garrett <grolemund@gmail.com>
Date: Tue, 12 Jan 2016 12:27:08 -0500
Subject: [PATCH 2/2] Cleans up outline for model chapter

---
 model.Rmd | 351 +++++++++++++++++++++++++++++++-----------------------
 1 file changed, 199 insertions(+), 152 deletions(-)

diff --git a/model.Rmd b/model.Rmd
index 3c9911c..32cf91c 100644
--- a/model.Rmd
+++ b/model.Rmd
@@ -3,21 +3,19 @@ layout: default
 title: Model
 ---
 
-# Model
-
 A model is a function that summarizes how the values of one variable vary in response to the values of other variables. Models play a large role in hypothesis testing and prediction, but for the moment you should think of models just like you think of statistics. A statistic summarizes a *distribution* in a way that is easy to understand; and a model summarizes *covariation* in a way that is easy to understand. In other words, a model is just another way to describe data.
 
 This chapter will explain how to build useful models with R.
  
 ### Outline
 
-*Section 1* will explain what models are and what they can do for you.
+*Section 1* will show you how to build linear models, the most commonly used type of model. Along the way, you will learn R's model syntax, a general syntax that you can reuse with most of R's modeling functions. 
 
-*Section 2* will introduce R's model syntax, a general syntax that you can reuse with any of R's modeling functions. In this section, you will use the syntax to build linear models, the most commonly used type of model.
+*Section 2* will show you the best ways to use R's model output, which is often reguires additional wrangling.
 
 *Section 3* will teach you to build and interpret multivariate linear models, models that use more than one explanatory variable to explain the values of a response variable.
 
-*Section 4* will explain how to use categorical variables in your models and how to interpret the results of models that use categorical variables.
+*Section 4* will explain how to use categorical variables in your models and how to interpret the results of models that use categorical variables. Here you will learn about interaction effects, as well as logistic models.
 
 *Section 5* will present a logical way to extend linear models to describe non-linear relationships.
 
@@ -33,12 +31,12 @@ library(mgcv)
 library(splines)
 library(broom)
 ```
-  
-## What is a model?
 
-Have you heard that a relationship exists between your height and your income? It sounds far-fetched---and maybe it is---but many people believe that taller people will be promoted faster and valued more for their work, an effect that directly inflates the income of the vertically gifted.
+## Linear models
 
-Do you think this is true? Could a relationship exist between a person's height and their income? Luckily, it is easy to measure someone's height as well as their income (and a swath of other related variables to boot), which means that we can collect data relevant to the question. In fact, the Bureau of Labor Statistics has been doing this in a controlled way for over 50 years with the [National Longitudinal Surveys (NLS)](https://www.nlsinfo.org/). The NLS tracks the income, education, and life circumstances of a large cohort of Americans across several decades. In case you are wondering, the point of the NLS is not to study the relationhip between height and income, that's just a lucky accident of the data.
+Have you heard that a relationship exists between your height and your income? It sounds far-fetched---and maybe it is---but many people believe that taller people will be promoted faster and valued more for their work, an effect that directly inflates the income of the vertically gifted. Do you think this is true? 
+
+Luckily, it is easy to measure someone's height, as well as their income (and a swath of other variables besides), which means that we can collect data relevant to the question. In fact, the Bureau of Labor Statistics has been doing this in a controlled way for over 50 years. The BLS [National Longitudinal Surveys (NLS)](https://www.nlsinfo.org/) track the income, education, and life circumstances of a large cohort of Americans across several decades. In case you are wondering, the point of the NLS is not to study the relationhip between height and income, that's just a lucky accident.
 
 You can load the latest cross-section of NLS data, collected in 2013 with the code below. 
 
@@ -58,76 +56,186 @@ I've narrowed the data down to 10 variables:
 * `asvab` - Each subject's score on the Armed Services Vocational Aptitude Battery (ASVAB), an intelligence assessment, out of 100.
 * `sat_math` - Each subject's score on the math portion of the Scholastic Aptitude Test (SAT), out of 800.
 * `bdate` - Month of birth with 1 = January.
-
-summary that describes a r, like a mean, median, or variance.
-    + Example problem/data set
   
 
 ```{r}
 head(heights)
 ```
-    
-2. As normally taught, modeling conflates three subjects
-    + Models as summaries
-    + Hypothesis testing
-    + Predictive modeling
-3. C. This chapter shows how to build a model and use it as a summary. The methods for building a model apply to all three subjects.
-  
-## How to build a model
 
-1. Best fit
-    + Best fit of what? A certain class of function.
-    + But how do you know which class to use? In some cases, the data can provide suggestions. In other cases existing theory can provide suggestions. But ultimately, you'll never know for sure. But that's okay, good enough is good enough.
-2. What does best fit mean?
-    + It may or may not accurately describe the true relationship. Heck, there might not even be a true relationship. But it is the best guess given the data.
-    + Example problem/data set
-    + It does not mean causation exists. Causation is just one type of relations, which is difficult enough to define, let alone prove.
-3. How do you find the best fit?
-    + With an algorithm. There is an algorithm to fit each specific class of function. We will cover some of the most useful here.
-4. How do you know how good the fit is? 
-    + Adjusted $R^{2}$
-5. Are we making assumptions when we fit a model?
-    + No. Not unless you assume that you've selected the correct type of function (and I see no reason why you should assume that).
-    + Assumptions come when you start hypothesis testing.
+Now that you have the data, you can visualize the relationship between height and income. But what does the data say? How would you describe the relationship?
 
-## Linear models
-
-1. Linear models fit linear functions
-2. How to fit in R
-    + model syntax, which is reusable with all model functions
-
-```{r}
-income ~ height 
-mod <- lm(income ~ height, data = heights)
-mod
-tidy(mod)
-glance(mod)
-augment(mod)
+```{r warnings = FALSE}
+ggplot(data = heights, mapping = aes(x = height, y = income)) +
+  geom_point()
 ```
 
-    + visualize
+First, let's address a distraction: the data is censored in an odd way. The y variable is income, which means that there are no y values less than zero. That's not odd. However, there are also no y values above $180,331. In fact, there are a line of unusual values at exactly $180,331. This is because the Burea of Labor Statistics removed the top 2% of income values and replaced them with the mean value of the top 2% of values, an action that was not designed to enhance the usefulness of the data for data science.
+
+Also, you can see that heights have been rounded to the nearest inch.
+
+Second, the relationship is not very strong.
+
 ```{r}
+cor(heights$height, heights$income, use = "na")
+```
+
+A model describes the relationship between two or more variables. There are multiple ways to describe any relationship. Which is best? 
+
+A common choice: decide form of relationship, then minimize residuals.
+
+Use R's `lm()` function to fit a linear model to your data. The first argument of `lm()` should be a formula, two or more varibles separated by a `~`. You've seen forumlas before, we used them in Chapter 2 to facet graphs. 
+
+```{r}
+income ~ height
+h <- lm(income ~ height, data = heights)
+h
+```
+
+
+`lm()` fits a straight line that describes the relationship between the variables in your formula. You can picture the result visually like this.
+
+```{r echo = FALSE}
 ggplot(data = heights, mapping = aes(x = height, y = income)) +
   geom_point() +
   geom_smooth(method = lm)
 ```
-    
-    + intercept or no intercept
+
+`lm()` treats the variable(s) on the right-hand side of the formula as _explanatory variables_ that partially determine the value of the variable on the left-hand side of the formula, which is known as the _response variable_. In other words, it acts as if the _response variable_ is determined by a function of the _explanatory variables_. It then spots the linear function that best fits the data.
+
+Linear models are straightforward to interpret. Incomes have a baseline mean of $`r coef(h)[1]`. Each one inch increase of height above zero is associated with an increase of $`r coef(h)[2]` in income.
+
 ```{r}
-0 + earn ~ height 
-tidy(lm(income ~ 0 + height, data = heights))
-tidy(lm(income ~ height, data = heights))
+summary(h)
 ```
 
-3. How to interpret
-    + extract information. Resid. Predict.
+To create a model without an intercept, add 0 to the formula.
+
 ```{r}
-augment(mod)$.resid
-augment(mod)$.fitted
+lm(income ~ 0 + height, data = heights)
 ```
-    + Interpret coefficient
+
+## Using model output
+
+R model output is not very tidy. It is designed to provide a data store that you can extract information from with helper functions.
+
 ```{r}
-tidy(mod)$estimate
+coef(h)
+predict(h)[1:5]
+resid(h)[1:5]
+```
+
+The `broom` package provides the most useful helper functions for working with R models. `broom` functions return the most useful model information as a data frames, which lets you quickly embed the information into your data science workflow.
+
+### tidy()
+
+```{r}
+tidy(h)
+```
+
+### glance()
+
+```{r}
+glance(h)
+```
+
+### augment()
+
+```{r}
+augment(h)[1:5, ]
+```
+
+```{r}
+heights2 <- augment(h, heights)
+ggplot(data = heights2, mapping = aes(x = education, y = .resid)) +
+  geom_point() +
+  geom_smooth()
+```
+
+
+## Multivariate models
+
+There appears to be a relationship between a person's education and how poorly the model predicts their income. When we graphed the model residuals against `education` above, we see that the more a person is educated, the worse the model underestimates their income. 
+
+Patterns in the residuals suggest that relationships exist between y and other variables, even when the effect of heights is accounted for.
+
+Add variables to a model by adding variables to the righthand side of the model formula.
+
+```{r}
+income ~ height + education
+he <- lm(income ~ height + education, data = heights)
+tidy(he)
+```
+
+### Interpretation
+
+The coefficient of each variable displays the change of income that is associated with a one unit change in the variable _when all other variables are held constant_.
+
+
+### Interaction effects
+
+```{r}
+tidy(lm(income ~ height + education, data = heights))
+tidy(lm(income ~ height + education + height:education, data = heights))
+tidy(lm(income ~ height * education, data = heights))
+```
+    
+## Categorical variables
+
+What about sex? Many sources have observed that there is a difference in income between genders. Might this explain the height effect? We can find the effect of height independent of sex by adding sex to the model; however, sex is a categorical variable.
+
+### Factors
+
+R stores categorical data as factors or character strings. If you add a string to a model, R will convert it to a factor for the purposes of the model.
+
+A factor is an integer vector with a levels attribute. You can make a factor with `factor()`.
+
+```{r}
+fac <- factor(c("c", "a", "b"), levels = c("a", "b", "c"), labels = c("blond", "brunette", "red"))
+fac
+unclass(fac)
+```
+
+Each level of the factor (i.e. unique value) is encoded as an integer and displayed with the label that is associated with that integer.
+
+If you use factors outside of a model, you will notice some limiting behavior:
+
+* You cannot add values to a factor that do not appear in its levels attribute
+* factors retain all of their levels attribute when you subset them. To avoid this use `drop = TRUE`.
+```{r}
+fac[1]
+fac[1, drop = TRUE]
+```
+* If you coerce a factor to a numeric, R will convert the integer vector that underlies the factor, not the level labels that you see when you print the factor.
+```{r}
+num_fac <- factor(1:3, levels = 1:3, labels = c("100", "200", "300"))
+num_fac
+as.numeric(num_fac)
+```
+To coerce these labels to a different data type, first coerce the factor to a charater string with `as.character()`
+```{r}
+as.numeric(as.character(num_fac))
+```
+
+### Interpretation
+
+Add categorical variables to a model in the same way that you would add continuous variables.
+
+```{r}
+s <- lm(income ~ sex, data = heights)
+tidy(s)
+```
+
+Every level of the factor except one receives its own coefficient. The missing level acts as a baseline.
+
+To change the baseline, create a new factor with a new levels attribute. R will use the first level in the levels attribute as the baseline.
+
+```{r}
+heights$sex <- factor(heights$sex, levels = c("male", "female"))
+```
+
+```{r}
+hes <- lm(income ~ height + education + sex, data = heights)
+tidy(hes)
 ```
 
 ```{r}
@@ -136,77 +244,39 @@ heights %>%
   do(glance(lm(income ~ height, data = .)))
 ```
 
-## Categorical data
-
 ```{r}
-smod <- lm(income ~ sex, data = heights)
-smod
+hes2 <- lm(income ~ height + education * sex, data = heights)
+tidy(hes2)
 ```
 
-1. Factors
+### Logistic models
+
+So far the y variable of our models has been a continuous variable, `income`. You can use linear regression to model a categorical y variable by transforming y into a continuous variable with a _link function_. Then model fit a model to the results of the link function and use the link function to back transform and interpret the results.
+
+The most common link function is the logit function, which transforms a bivariate y variable into a continuous range.
+
+Use `glm()` to perform logistic regression in R.
 
 ```{r}
-heights$sex <- factor(heights$sex, levels = c("female", "male"))
-
-smod2 <- lm(income ~ sex, data = heights)
-
-tidy(smod)
-tidy(smod2)
+she <- glm(sex ~ height + education, family = binomial(link = "logit"), data = heights)
+tidy(she)
 ```
 
-2. How to interpret
+## Non-linear models
 
-
-## Multiple Variables
-
-1. How to fit multivariate models in R
+But what if the relationship between variables is not linear. For example, the relationship between income and education does not seem to be linear. 
 
 ```{r}
-mmod <- lm(income ~ height + sex, data = heights)
-summary(mmod)
+ggplot(data = heights, mapping = aes(x = education, y = income)) + 
+  geom_boxplot(aes(group = education)) +
+  geom_smooth() + 
+  coord_cartesian(ylim = c(0, 125000))
 ```
 
-2. How to interpret
+You can still use linear regression to model non-linear relationships.
 
-```{r}
-tidy(mmod)
-```
+### Transformations
 
-3. Interaction effects
-
-```{r}
-tidy(lm(income ~ height + sex, data = heights))
-tidy(lm(income ~ height + sex + height:sex, data = heights))
-tidy(lm(income ~ height * sex, data = heights))
-```
-
-```{r}
-tidy(lm(income ~ height + education, data = heights))
-tidy(lm(income ~ height * education, data = heights))
-```
-
-4. Partition variance
-    + Checking residuals
-```{r}
-m1 <- lm(income ~ height, data = heights)
-# plot histogram of residuals
-# plot residulas vs. sex
-m2 <- lm(income ~ height + sex, data = heights)
-# plot histogram of residuals
-# plot residuals vs. education
-m3 <- lm(income ~ height + sex + education, data = heights)
-# plot histogram of residuals
-m4 <- lm(income ~ height + sex + race + education, 
-  data = heights)
-# plot histogram of residuals
-m5 <- lm(income ~ ., data = heights)
-```
-
-
-## Non-linear functions (recipes?)
-
-0. Transformations
-    + Log
 ```{r}
 ggplot(diamonds, aes(x = carat, y = price)) +
   geom_point()
@@ -219,35 +289,8 @@ lm(log(price) ~ log(carat), data = diamonds)
 # visualize model line
 ```
 
-    + Logit with `glm()`
-    
-    
-What if no handy transformation exists?
+### Splines
 
-```{r}
-ggplot(data = heights, mapping = aes(x = education, y = income)) + 
-  geom_boxplot(aes(group = education)) +
-  geom_smooth() + 
-  coord_cartesian(ylim = c(0, 125000))
-```
-
-1. Polynomials
-    + How to fit
-    + why it doesn't work with missing values
-    
-```{r}
-heights2 <- na.omit(heights)
-tidy(lm(income ~ poly(education, 3), data = heights2))
-
-ggplot(data = heights2, mapping = aes(x= education, y = income)) +
-  geom_point() +
-  geom_smooth(method = lm, formula = y ~ poly(x, 3))
-```
-
-    + How to interpret
-    + Strengths and Weaknesses
-2. Splines
-    + How to fit. Knots. Different types of splines.
 ```{r eval = FALSE}
 bs(degree = 1) # linear splines
 bs()           # cubic splines
@@ -265,13 +308,8 @@ ggplot(data = heights, mapping = aes(x= education, y = income)) +
   geom_smooth(method = lm, formula = y ~ ns(x, df = 4)) +  
   coord_cartesian(ylim = c(0, 125000))
 ```
-
-    + How to interpret
-    + Strengths and weaknesses
     
-    
-3. General Additive Models
-    + How to fit
+### Additive models    
     
 ```{r}
 gam(income ~ s(education), data = heights)
@@ -280,8 +318,6 @@ ggplot(data = heights, mapping = aes(x= education, y = income)) +
   geom_point() +
   geom_smooth(method = gam, formula = y ~ s(x))
 ```
-    + How to interpret
-    + Strengths and weaknesses
     
 ```{r eval = FALSE}
 # Linear z
@@ -295,4 +331,15 @@ gam(y ~ s(x) + s(z), data = df)
 gam(y ~ s(x, z), data = df)
 ```
 
+## Summary
+
+We've avoided two things in this chapter that are usually conflated with models: hypothesis testing and predictive analysis.
+
+There are other types of modeling algorithms; each provides a valid description about the data. 
+
+Which description will be best? Does the relationship have a known form? Does the data have a known structure? Are you going to attempt hypothesis testing that imposes its own constraints?
+
+
+
+