From ebf4070d88fa65957f1540c06708acef591d5e97 Mon Sep 17 00:00:00 2001
From: Crystal <crystalzyan@berkeley.edu>
Date: Thu, 12 Jul 2018 16:57:39 -0700
Subject: [PATCH] Documentation- Basic Profiling for Ray Users (#2326)

* Ray documentation - created new section 'Profiling for Ray Users', opposed to current Profiling section for Ray developers. Completed three sections 'A Basic Profiling Example', 'Timing Performance Using Python's Timestamps', and 'Profiling Using An External Profiler (Line_Profiler).' Left to-do two sections on CProfile and Ray Timeline Visualization.'

* Ray documentation - Fixed rst codeblock linebreaks in 'User Profiling'

* Ray documentation - For User Profiling, added section on cProfile

* Ray documentation - For User Profiling, completed Ray Timeline Visualization section, including graphical images

* Ray documentation - made User Profiling timeline image larger, minor wording edits

* Ray documentation - minor wording edits to User Profiling

* Ray documentation - User Profiling- fixed broken link

* Minor wording changes requested by Philipp Moritz addressed. Still need to address (1) compressing the image files, (2) correcting ex 3 to not be remote, and (3) using cProfile on an actor

* Ray documentation - For user-profiling.rst, revised example 3 to show a semi-parallelized example. Compressed timeline example image to be under 50 KB, removed view timeline GUI image. Updated timeline example image to reflect revised example 3. cProfile actor example left

* Ray documentation - in user-profiling.rst, added a new example including actors in the cProfile section

* Ray documentation - For user-profiling.rst, added section header for the Ray actor cProfile example

* Update user-profiling.rst

* Update user-profiling.rst

* 4 space indentation

* Update user-profiling.rst

* Update user-profiling.rst

* Update user-profiling.rst

* corrections
---
 doc/source/index.rst                   |   1 +
 doc/source/profiling.rst               |   4 +-
 doc/source/user-profiling-timeline.gif | Bin 0 -> 49184 bytes
 doc/source/user-profiling.rst          | 649 +++++++++++++++++++++++++
 4 files changed, 652 insertions(+), 2 deletions(-)
 create mode 100644 doc/source/user-profiling-timeline.gif
 create mode 100644 doc/source/user-profiling.rst

diff --git a/doc/source/index.rst b/doc/source/index.rst
index 14865b002..5f0e0c215 100644
--- a/doc/source/index.rst
+++ b/doc/source/index.rst
@@ -132,6 +132,7 @@ Ray comes with libraries that accelerate deep learning and reinforcement learnin
    :caption: Help
 
    troubleshooting.rst
+   user-profiling.rst
    development.rst
    profiling.rst
    contact.rst
diff --git a/doc/source/profiling.rst b/doc/source/profiling.rst
index 6f76c7eb1..59d12d635 100644
--- a/doc/source/profiling.rst
+++ b/doc/source/profiling.rst
@@ -1,5 +1,5 @@
-Profiling Ray
-=============
+Profiling for Ray Developers
+============================
 
 This document details, for Ray developers, how to use ``pprof`` to profile Ray
 binaries.
diff --git a/doc/source/user-profiling-timeline.gif b/doc/source/user-profiling-timeline.gif
new file mode 100644
index 0000000000000000000000000000000000000000..455f1a9af3d9567cda0f8c5e7b5db1890c263118
GIT binary patch
literal 49184
zcmbTc2UHVZ_wPMPNTCPNP=$b~C@nyMP^C!~h=6o8R1pgiR1g$GLI)8k(joLBLXaX&
z2^|43fFezeh=7PlQ>x_pd;ZV!zjwXs|K4}4cUD$1b7t?e&pxwHX4YrECT7Ms4Yvpg
z2{aD?-$q$a6}|yY!1VO=#>NH@H+OP!@*4}i$?nyW(b2q5No8eaK|w*?-Q9X`r|K%+
zzhkj>cXur<EgPq1x_|r}Wz8^|%+}V{0b{lFmZCo_*2s9{o1(n;Dws~%_(J{rg@uLB
z+NUFAkGCwX*oX>_v%Y`-#Nsa-xj!@8=X0Zbb>Lmow`-GUjcqFi)_;FKIH;Ry9r@7p
zYinm~XD@M@`DN;3S;mV;krl7MExp+v@B6v-;=`xy9|wDX4k9BXziofDY?&VU@UcBU
z?eW~Vr8-u%hEnUb`JKJ}s86wDKi7LUM`p&y%VvH=6;a1$#u#J6zxLNYfBW!edFbcv
zkMZ&GZ=XI*ej49fXRWTTz8{^Gm6grS&0YBReP?a$V=OE0C4Kty_|VW$b#?U(hx0@0
z6JO_-zI^$zzr*_a^(&Cv`+jp|_s{MK<Hal1LR-|#pFe*tU%veP%MV8<r#EjHI|o}N
zPxo)%zMVdO`s3E5rH*!8UERj7t<gWM<>lqAtt}RdCCTD+w8W#0+-|x3{nb7lM0E4G
z>U9h4?sfUf4IQbUt?g3iE1o)14yu+IOQNBd&lS^Pe=pl$D?5+is~#GjQm1v%+CiE@
z*NNA?Zn>d@u3x(mpc$ki<>u+<?xp#6|DR?>DfGWgZe7)p(*NfK`k$DFe*Ru)RRtA!
z4~&u$T1`zs39G86rX+_}#waN(V*Y;A<dv|Rs%o0b>ga!OQh%fQdwOfy7#aUN*54x?
zDW6-nh?<Iufq{Vvfhr1q{+AV%G&D5+=|fpr{;!36z}*|S+=Aq91RVWO4@O=A9{yK|
zx32izK>yRDo4eobTRKvINBUnWTqjyu{&&YW0u=t0kb;Nbb;TezqN0)lM)CUffAad5
zb-*nfum3%a|Is?Y?k>?w(Z(yl@3y~(*I)Ca|8f31cmMZ5{}}!qji$B#mA@Co?V6FF
z$L;H0H*T33=}7%Oqu_bPQ`10MO%1DIY=lu)#;D<xlyJ(*`o^jn>T1d=MoMQ4G5_h~
zfAUpR!5XNm8L6rp8lO>8I)gLNz!(`DpD|F<S5;M4H&FkNui1@&TW&Wzy#5pS%HO#E
z_Qn2ReKigJz1(j3`P=#VUHeZ3;C=jV`33m+5z&UWYG@ftH;*ef{%JV*&jtFge2u*P
zuiWwSH1_wqj{esSYhL*uM)3cg`G5I({{NXZ#lM45{HHwrk8=4>)888WXZOF0|L?(n
zFCVWPf7guv-^Br99sJqf+uhmT`u%HjV|{J)=gN=e?@Qkn7v{gteVP5t{Pgj|%=Fac
z#QX7gZ^uSQ7;lDO4-F2y>hJ6A>F(<6ptrw#(bn41-1NM$p}y`}Ev=^dY1Na;it@73
zlH#Jm0&0F<ZccVqW=8tswA7U3M@bJ8A0))b#l}QOMMhA_;iRz8`}abE?*;|lxqU0Z
z|0dDz#`SAgeXsaj_V)7haCakIx_H6W#o5Wx!T!9Rt<5>SwUy;r3v)A56XP>Rh6eh2
zr*(C-wKO$w>T0T36=fxiqJq4f?5UG7CypPJJ}M=NmJmN8CMqH<B#1hENPwRY$;-pd
z#mT{rfWx3{5HJV;BzM?=e>S;)zW({w2jDDH7?~0g85JE98yBDOAn{?+qvVv-w8!Zg
znOWI6xq122g2JNWlG3vBipnR+eBLNp?X$Z2hQ{Yj%`L5MFJ89OJ370%dwTo&UkwZn
zy&itU7#SUV`)<6gDivjS2mSFA^YiSNxv%pJi{F;MFaKEixw^K#vH9!w*7nZs-u|Bh
z769Q8GsRqJw6~)fIrh^Ex}uRs&2vm^3wz>)RD5UpYm53Gq7CE3%$^m$N|CXxbba-#
zWH24$aZy|7WG4_A>c2Ges;=x!z8;0+h<SbaNResEsS5-36=P-iBJ*7HhRSzO9P4~P
z3^Y7>Ut@H|;MwE}@I;-@SmlMm#;4QIiL5UJxfai>FVsHl&uSQaUh}D)1mP4v+eG`^
z<+i3a8sAj=r7!8IMc&!wXJ7N<1bJm#Hecsg`C*qYh<tvtSR{gN`&RLJ<Li5x$6%gi
z>z(PL1R>scddCv-$8-#3A25Zt-;B~fk$gxSJ`v;9x7SPE+ho>r!|bO&4!>yK`pNtx
zqUkP$NXO|TOq1OtV-gH_uWbrN9J9zgnx(<{i_3&lgj=s!S?~6{ubU<k#2NiFxBA>3
z0idF^#xo#pOi+$h7+x|Q#f7BiUgkGlohI64fT$8|5w-XrxyHrWcuenNwB-GILeT?W
zQ+LV(FC>2iN?uy2Ura4H=~4t4dBT)sRr1YEFF^N2Z6q7Hs?p>MIcG8Ya^zW=Ui8$j
zO%q69L~U`n-8Lco3WM8UDt+>GOm8ZbaMS&Y`!yx^?3?p!3qE14)E;R589n+D-HhX3
z3KP_p7K=iUBuBvIXOP|EI@-(*9`y-oo|M{3_$RdbI;NisSD*9+t?W4xE!lE~5Y6k7
z@QNf<63(RXLSkTVMRk2f;Gp~0zV(5LbbX&IsBHlL6up15T>Op@(-4dj3DO4zx4HHC
zg<R{<gOs?{MmFiX0)1dIl1UNcs@id(c830r7S<Iag!}N}CIJXB!%gupvA<x!*Kb*}
z<4DQZ;&2d@D@nfiki)e#R*%q$$bND833NZ7`i##PDao$)V6go1xtPj`37~i&V)4$;
zBhFk@6OePcHvI^c{@E?num2n{tmC0nisB5|i$3B!(?LQxFV@zxaqDT9M{@39HW`6N
z7JwK|(Uio7yQ)-=dYSn_V(Rx&LFAF&$LT5%7hmdakoKp7XF@>~`RHg56hZG+m?Iv5
z#Ccb3&nBpUT<kr<tE@k#Rh?=2ib5Re<j|@o)Mv}3_<mWSEUF#-bU69%?dRtd-~8hu
zKJ7(lKx!*VXu5OFL<o9KiQazIy_SYrepx*x>ciH0Mu=MvS42I0@Q8$3F-LYBarZeA
zB}}&YmMQd0c?Zsd_~Y)Ci?s@3;HaSNk?8T|1JIh-h;VS<kuaZc+XwqwKf}C|T;55X
zWsuqTKZE2EWGT!EL^?>)Q#}K;_!5s`LsTG3_qxFKNiAAcR-yi+=`{pF?z9hqb9*rP
z0|y#ZAi2UDsFg$iNLP_3aLR1PT`cgIiNW}QBEpjB;0R-ttlOr@8!B1b$qh-TsroTt
zKsfhnQk|5syq2|qWXe86;?4fgcmh|E&`<vuVIwP~BW@DpLeMb6tn$jVj&K_vi(#<4
zgVlq8BPsR5dh%2tLYdAB|14LO=@qydoF);BHr2;>fV~-=%#6}jkPsSVw_UU%g$|eB
zG-0{FD-bLgVv6b`2F`{F(RuIdm1w!@@i+q=fE+T)W>nV95iIn0nLv|4F%Rq1#X?00
zQry~>kIu&Fg^8&3dTVZEGV0?bf3K2~ZeN4@O^k3K5Fzr~fOm9I5xSai^dWa4{LY*=
z$e94<`psv^PLS*l-im@^2shUFyFZ-TY|B$(68-c<WN)egABE3erBECriFYtc=L)WL
zv_L=y$FM=K^{=`hJ%v_)XxQd8&JF`t2#f;`8|vVN1&Ku8V@7A2F}abHF3|(CCkIIL
zP-v563XRyVduF3dkGm3o*F8p1595iz`1YLzA`)NpJ&?z+l?|Jo0(I7U@Us#owR%p-
z7^NL@bWWm(t)Ro9f82QZSHq<o>LN>(02l5Fq!<?wuCpBu>Jk-0@z)=3>vXdSVVb4*
zz~DMYz&-i4eDRNI3L02?%KavKR8Ew<PLisApb!1RBQ<*sf*VNeU78pZ8LqQU^5zdp
zzO1c@<Km_Qu_Be%E?c-@M5NpllR7vtzoLc3K0kNBb#cu@-W_obM_xw0<BMsHmMh-i
z)j?O=|2nEGC&O>xNQ1CHjhBpCjKJLqdMBDkfnw|9BQ%sCGxn57z8w%}KKWe&0wB+}
zIY_ekD1v@pE&r6X%IioYjr4wx)~y%h#OV{QxDuqccinjRB9mA!Jwmu}<}o*#-NXF6
z<Fu*Z?Jjin#EGiOqh}&IjLTgkzI9*DQb)^aJsru9hh|HE%b3?xsIF!I#72N@Z}-6_
z3?%HwvR<SbtD14V4(i(zi~jnie~0ryjkZ7*F3nr@%?Quqi-FjR@b5fNxob1}&U&xj
zGk}?4!f)|OLIk8gSDgA->AI3u<q4`P9@!l<=xh}PXA}&Pc<QN_DA_pM=L%5p*;%gC
zMSZQ^k)&#ig2T15rY^gq8N>B9$8%%MukMafd482>JV0dEJG?EoYN&j5cFE8PExBXy
zpz5;acjxKd@uuO1>cH3EFYfKWr}H$@qAi!*MfWE9ts0-DvtBQIYwb-k5*zEwEPwdA
z>`k@Z_$AqRF6l<d-ZYcvc~hU|ihuUr%%auvmiMn$e6F-2Luqt5IOb>Y;Pl7Mm(O2r
zy!N}d7xIa<qE3e^tdimTOg59I4uRp-NG;dTNWY+N35B&d+x=OAH%&dNiE9r++P<I{
zn)*-ct|z7J&mFhE`^q8F|M8Qyub8mrp$k?U_uKaxm0vXv2M%xK?d>n<@wPCctu_lq
z|16sJjlNDF-YnJn^9}#7<!zbOuS%CcOO9_^#-9)Wst)<{oxs~V(P#DhS@xf0AM4hs
z_rt#%TmSqZK5U&?u-a;w{<9MFruF02@Yc({KR-#lZA`fJHmg(gU^OTVqQp58eTYJ~
z&NStot+_}zBdtfs#_uN}Yr-8Y^HyO&Cr`}dX;69vlADyyqrV6VT)1G)MVi(xUD#eb
zck)p3^`pmEPT894ISTLvrGsW=*|pd^!CVvIIkN}%B#`YE{;T23Cm-3wO3psi#cN8I
zZk1SGBW`4D%gsYaNQuhe9T&#IQB#(*xirZ)6Ag@3vF_HG0FmxTu0B=PclhN(4#al>
zOgjYG23S-<SQlYKwyf3DxINAz&<rgdX?R04JGn3<eH8|+yVtyCjW={oM}lZZk|}z}
z@wN^&x8P%C*YDF+Cddf(iTkOm))We&j*Jz;v2<07tQ9@kovB(C#d42<(-b_3BL<<R
zg4HDkKVYhPABE=$vX9jTr>(-=X_9NUVXp7tl7*^78iq`RyAeUf5Lq7pTnzzV#vNUs
z4<-X(L8=qr4GV)2PyqOi38KSafWr-DOo6IDlD<#0j?j~HriEFI0)o_w6wEm{2E4`q
zb?z;&f&#fw;VmYBAQ6-o4fS62Z9K#dA%g7iVkkWW2#xGWi_Le3`!L}FNZ45)1kn__
zbKBk-4MSnzc}i{&%u#l{78&VO5&`{47lSt34WY)|KzKy0I@@hy^QB}ibsi&?i?<hv
zi;$q)bg>A!4up2e561$0=*L3vZr%*w9D>kp3ZR%__YTF73GjTO_`1d8o=0KgCD0;w
zP%+{}C?-nQ6j0{_p{UAydJmMwfMCMKSmwDyi;?RA@K(pzq)nKBDqv>}6(rpE!2u@@
z!#1N~HxUmb=np>PqHBy52vURqx_&GZK4wdCN(SOh5%(Dnx-Y}QQ8IB<u_I21%2b)j
zwa7^~krfrVx_<1IJ66yM5hfj?ayK=!QS3hEtj0OF5SsAxWwjVnd-N*Sbrzl-u3fW^
zg<vF6ByBVUo?CXYniS1PPf;<j4_CqYEW)`7_OS%`j<M#GiND!B{xSL>e+-_V0mx0p
z40*7UolS3R)j@qoYB!GRcz;QYA%J{~Fm7Mq+5sdo!`?dpE)%IT(W!3Y7-L|ob{MQG
z(rxfe-S)W&qFV)_HHHoHF!v5ndt3qSAghD|U;_^|7|j_EQ4Kf3!i3mejj;YSxLrgx
z?Yzjef6n3Aa6?GM`{$~IS755QaHSwMs0y}0Su*{Kic4KKWh1%oq^iy#d*%rQduLE`
z>Jb5(GyJQk!sf8q?4%$M`wc8skxi}GEHph<B_lek&nzqwqeF`%4NQhaV<gp8;aqqT
zf`=MERoveY+jAG|Wk@zTO8LmHN=`v=Lew3R_rs`EDB<|AU~5Bx#4h2pgaDQr&hYen
ztU6a*@bOCMRsVeJa`F3^9Q+13(<$G_NX?gGbbLvF_m-M@C1OxrlAi*55*wM9nI~6Y
zIO?PuK|fbA0;JzPeTOEGa|Sq()>|rA)omERp03f3b3~`s;q~P6Y%ZWkTu4&<qqizy
zOm!c<yf%obzR@wVjqKT_Ql-oS(o}Z7SuXsPs<TI7m<K{^JHrkRjw~?`c?Fylk|!8O
zQdUha`KvUn!|pRoOW9n;-(l^fJ*`J2Vpd^pG|6C6Y|YN8j&<_0D~N`Z2$qq;&t!8;
z3QXP+@X*9wZGnaW7W@z;n=S(YrZ&;3a-Xig@(m$$Scar2DapoC;$*-{r?lG-f$$Jp
zDPKf0B)(ZAlOL)yuWGB^Qrbqr-O*EVb+ji3Sl>6#LaY~(G1e_6XOzpWoiT=jtC`1s
zot&?T;lvYQbZzF2NlbtWya~<}h;%**yT3{ahrmNttz%*;yp!ZgAgZBy1u+bXtM2eS
zROJumrD29D{6(sQbZj84G8%G3H#KtCISYO@+qMO6tRhJfu<uh>e~f(=HK49b*X3l`
zbGAb57+RjO*tMMpFd}HJSSu8b^<kb9WaJ!qrE+-tfvvfk<}MY*Xb?}65tzM)m0^Fh
zY4CF+Jn~nT8(`si3g++xEu@9GXL$PZ86yA!*1@ps9kLZ`i&eSkP{PKs>&LWJ7Y#WL
z${Ybz2wmNo3IFIsdAS2OwP=aT(h_(U7C=)iyo_7ya*HLw*lC5&x3Ck*a()Dy=q}vA
zHV(VtR8dhuSTG#<a_(fW<kmLa{AwzN;DkH{myf`HF_hK#B^QSbW^N^eA<j{IPa=_3
zCSzECJ(=>2C{DbHqFG7^N#&BMT1LMCV^HPrsF^U>J0lK}Qjv=^RLtj>njSUqI1h08
z!<T0b4Yd$|!tdIx9^?IW%;u0z2l=iD&NSQ*%0-u@C?E3-hOb>MxAFwsiNSyL@2#7p
z!lp=b3mrQ37==SdP|f4qrqBz9rg$Ni(c)3le6mhF0hro>|K6*U|1OcUNeM=SrY7y%
zj#5$&TdZFJ7;DeFVpJFwYR9w?W41k*TX3*p415;5*&&ILlXGsPh5O3GgkBtqMLZuk
zN2&ykKWXE{lp8g-T+u>gIZ?7sQZ|Gg#P+H3SVWO1yLwl8xMsHel<FJ=97uy^#hTyy
zi!TpfQcj2o>;Zg8I?Rp6lZT921nik#oJ{J_c5i{N7A>c5^^q|x+L5~Q^>^)VMMySa
zI7iDk4N9Y)Ac9J=XI&&k=aSYJs%~aUGG*HcTlk(QBoVq~W{ZBTIl@z}ismKnPAxhu
zWY6LMV%#>unOOfm=SY<&3!%9mHA!)R0$}wod1~iGL>rziA2q7?2e}8~#-Bky%D|WT
zGbwsd7gA>=9ZTJ@4r8jVj2lcRE9%k!IVT5?LPqqe^*B7HqPS~)i#+_y%X+L0e6OST
z8sLw#=EA%=84C?&0M|RQ=jBYFi!sExaGv~;p^OrVY#G|72FSZlQSLO6fGzTB@h&un
zyCP1WLE!7po#r-_?(aGzWgpS4AC2heA~q~boxy%HToDc@0yY;~6DiEhomk@$&91qP
zsET(n4+mP^@0D)XDF~%0aw6k=WkFZL&w8a>$UD@JnxWB1v%Sl3{%v9Rukh18@a$|E
z$z*BUC4@YJl|=?#Sn$g@pVduo1A4`>^eRVX?)A6{fw+*0%}x4s$p+B|&YZS}hu58L
zzMP|n*mi&kV#nk;7S(7G<|4#MtBsSq!O!yxdQ3bt`kq1WW(<m808XY{d#n72K=|-L
zp_c92?u6TZgbp2#<2#x%6(?vhL>am6W85UU?mA#*LZ!WKve#VIG6msBqb1h^BmNCs
z1Sj*l57+=L>C_WC6ozK`M}hSid=b3<2M4QOfqNBZ0;nC)FuGzG9*|^1L)wcFRrpU~
zk82l13%~^YYvrV2DSNd<WP&a-M7LNg;7Knm2qYSzdyj$(!{QcSK-~#gSV;^m7OOkQ
z@=zT!;7$TgLa8coqyv2f=>vtl+ORaMQhWl(iB?i!sIbWD58kT?qD^B`%!F>wNtHv{
zL8`qzM&5>%X8s^s1G+m6?(jqlH5i_4q3yS4h!TY%0}2c%j3RSDp<AUP`6;U#Fd1&G
zPBohkM1)~q%0%Fbmb%zm;Yg1+f_M?6O0CeDQWs3W`{CztcRRdH>xoF1a}n95zM$re
zGht(~`&Bp_8W#O=Jp2^S4N$vz4`=bR;CeOua9VU6P5$Kdx51cWR$$YuO+43XZL;*E
zXy>R$Myhj1yS%<;i>3ksV7Nm$)|<Ngwq>Gu!q(<iOqxi^%#Wl2&-XteESI*UApU^U
z*A&9lQ}Di1l^s_@O4qPMYu5MiKW_TeDViNSv6Aun!tKpUwY*0WxVop)nUUf_nR6Cc
z%OmA2Us7TS)#*->J0Cki57jA7hTfFfzHiwZl3lJ{3iBFkCFhEbAS<mKxs#Vw{r2)F
zL8{K&eo98_o60$IXB55aB3SixDxVUjw6Td^Z?#0f$G1@GUS#tV;CmUt{B$jogWTG(
zZ2nb^M~6O_T14^&sKFRHt&{h+UiCJto8Tm#N)E{L_Q>yTAA^t_$aZf->BZ?4vOz_^
zTka!746rc*AkftJEdw=N9q@Ebw*u-d!lWK|mPO%<h>W5LVm>T`1(@~sdl$hSu2zWr
zwl>G<%B7Z*-!`+=+>0Ups=SzMhc@l(jKqDyy2v6@gbQQI(4<fAH{Kd7@a%Y|**>Jd
ziz0yTScfaH>3%%{hag`>dt97Oj*JM&)YU7c*kE9;vw`{Nr*9aNnYkKErk(;))E8L?
zJWA>f4>_6pl(B+nTam*Jg)m1E<)Qt^q`}gwg)odn09^%VD``)G*;C$cRo`y6Wie{w
zT3Qe%!J%DjhyXioZpW}NNHDi|beOkKOw@yvFh9EhNCaDSXhbN)E)2!ZPI2So3*feM
z<3@!!R>h^<5fn^H<c7EjLi}<Bg9Abzpc`L?`LMIIp;&Hq*@945>bo?%;Ie3U{usOV
zXrEv~lt>{tbl9$YJVuZ^o3F@D2gMHIW6$7rTNXhza{X$KfO4lH?9jh7U@IaDM!G_l
zEi^Cd26guSOi@q$v|K~Kpzu|J`LtFHibX<W_0CIxD`^BX<h)0bE_AbgEQyp_f$*7^
zo+wJpZHSyy`!k~V)gU9vQDjkHjXeV8%eL+1OAv^E%4R$blG}tIcL1mF_B*DnO&8b&
z^AG9S#nAzz<Gkawfrbl3Y{8681O;Z1;pQE2W1Zs-dD^$4vtGI{)}j1)+wLv$A<2Mz
zjpTi+bq{KIgR=~a_6$e?5Lh>9e62ufAFBYPt7rE;{71GM!<$EWLvrEaoLR-=NO80`
z@DvuR;B)dZ_tFuZdY3d~#`d|461Pp`p^N-8K#HHL5o45lyqc3;M4^p&;Rs0OO=sN4
zH|^~#766lBlGso_%y7i<awhrDH%Wa7uru`x$U({Hb2zu9`*0+T2i+0DFVj;HexI&X
zfEs>KJ{u>;pCL(d?AMnRGk#H9K;k}xryp@SWZH8?)5ew?MO@rw#&O{t>0a_`-{w;A
z#wWj~+WlUDgVYZ$_n}<2iTaPvM+qMXac-lHecTO7qj@<eQs=XY$}v6Q3pe=m$q?c<
zLm{X$0Z12g{L%*wQveF$*!Ni{&*l7!*d`g9Q(**i--__Gi<RF>6TO3}WvuY2e#Jt6
z-=&K3(#oaEs>Y?IC$!#E->d4!eZN0#ny>s`-S&Iwdkq~XyG-jwU0ts2KmKI-*^t`z
z<vND3?2r1l_E&#2On5)}(KvJG`;X_$DA|>!xyM&mnior-th6jQeqU)_?Unu6wmE+F
z=Zo$6CqG~A|Nj27s-#h5wOx)!E~|rXj2`RcQeR%}LYl~}b^k+bdxS1mt@VlpF0b{W
zqvh87rPHsi_pxodnMlYzUtS-?^vP`uVc(zK7+@njs(Y=ywY>304=%sSP@0xZVHkVV
zDl?21a@I$6zf`H9Ot)C!k)gCu_l|Mbz#qTf6QbpRPk5$Z|2^qb_VoAEwdX&6PZRs(
zw`Oj=zrOY18imk-f^GiT`b2^&Y%?PSZ)|_Q?-$++#Hg=qe@QY?*qKXZnQ44VQD;O8
zr7K=+7I$pVN)#?mzp?wRysUb6sp|R4?sr<B!rpTI`x|>dnii_}R@$~!_I}dei^ucc
ze9hOp)_<aAe|<>(=l%vmE1wV2y3HW1ja;tz^Lr-n=btU+3F|YN(Ki(j6{O#?js$Tc
z38lI-)|dYL`tU}7F}I#Sa_b*p`?mtQr~urQ4i-teny*AJbccc=T<)AwqYjW|EEOh4
z(hGJ3P+UocfGTS>oNwP+hzqIdaOKAh??uX#>4K^X^(n7{HkIqLX_v^<Y80+2`^Y?j
z<X}VsLW3w0<ZT#uxr3KasE|Ls-{{!IB1N-ng%9?L-7;!xagMI;%3>1$1WAu``b%v4
zOna5H))IWq774L1`qFtyU=bQqf{yFeitWJj>_bGJ;}D9YH)H0J?gC#-JDqbGSJ6IE
z(|V5@b^5z`(YS}g2vQX7+z)M+RudleSFfys*HfqoqSu3iVl?_o`9nz*kkINe+)K&W
zay==zr!g_S_xs^%L@zdjSutFFvQ1rt$DyCad8%3AlneWXBI&P&{sFdwo?lQiKE}9<
zOfoBstLVs@P@Imqia@C6Y_2`H;P|uJq@6C$_xA1pb^^qWCb?<<@&+M?0<NCyCJXP=
zBDH3Ec#0t%1Fnl!T5Tx+ECVUFWCm!_^|?;f(+8F(qvc(xe_v%lAJUN4q`b>scP2!%
zolsr)tj%H>%L?;pHPR7(;z>3JnU`k-kdF!HNC@g0xX5mFtZ+6altGeMVZJ$itpMQC
zD;ePd!6l67htu$r3XhuLuiHMSX}vBL7@?ox_dT3N;C|xR=R3^45}w%8EhYAK`<+(!
z&FuUAU&NcIOniCPtb);%Z1dE<WBkt|o=$ZEnAlgk+j^XK_$dXBXNVVWk^(DOmZ6;p
zRTOfF4$e2#<5<AJ58LRcP)?D=?%{#uMK=Q%IvFZ13TvCt=OhMI3BzbWPzFeMH{pz;
z-BsAswLWmf2WMiq?943xR#TT{_06>SnXj79n;Bl!xAL_<EZS_fjK98d`+59@72oHr
z%s<sZFDyQ;J`QY~w^a1k$8<U%yWoMi2wrJS+4klJkDw>X>K<l{6T`&2FvvHCK$m)z
zm^}PGv8Z|mDchL&(#jHmFp0SX=1iW8x4)jR$ucrQtK@ic*T$I4oV_xYCgh_>fvN4A
z=le3V0ya;^vw$wtr(e8*^%U34?SYD{jqF!!9?I3p12+)wVlSM1RmECXc4RhUxi}6}
z#13&t%V&g(3al@P$JHN2vQ3mb%Zwd4-VK2JeLA=@ex^w?xM5WL<0c=@W!<hP=?F!D
zih)rq@t%|plGLeh^W$#T*+qTPsaG)yw+eZmE*h)sj>IR{7aimNX6Di|no(L`s`%pD
z*?`@#Jf7QSCfvR@@hxx5bsH)zbA5ilOck<Y7ziQI;hE-i6xT%(N&$eww%t0yGFEL4
z0QmAPoZ9FkJ;)py#7WoZxu4Xf)hU+?jVJ)TN6BY|?}cmK3J{Q07<Ag1=n+WS{tIl?
zlVuGokfyEta(9$myxQE}^@m$`b>GasvutaKudQ(|zS<nA%>pF!5HH(TuQp%pnd?xS
zoP^T3ZOmlhgTJ{eTB5qnSx|0XKia^{FGEMgrINHzCmTJ%$+`P_&_}1w*Oz@z4S~^|
z<QEG}zo)2$h^!WX<R<iEddj>0Lu{>EM*nAGo6RF?0~_IONW4vDG60w<n(OW{aftiT
z=w}`&6J|d!Z+@Yp@%ugzI%GqkVEIe-1D|8||2ztPukRsIJ_^60`&MbM{OoP5gEdLD
zn}@E5t(UH+gMt!kRckejj0kz{Ftoh;UdwA<aNC9ptr4{Uw5JE2*%zIZXcdv;8D(3r
z$hNI~Po5AJ!<zogql?iy$158?%a(1T8CP@Od&)o8bdiUDvNM`Vh)#;x=VtdZ`W>Fn
zt0m3k3C8)_zIOm#Hr@jX9Z;$jTYmj+W&K%w*uKAD&Dx#5^!-9054yw;&|phj=0d|G
z+yOTNtY=)^AAo<@gs<L4%%krGpy3iq!1ed)-Qz&zEyNRP7<Lw(Q6}Pj0)Foo>?TUO
zrv8kM8*mwVS8jqlENv-k%7u=$D+s(>3bMG_eOKR(xPSOwp$XK)K6oEy>PB+wi86Qb
z`U{IFx{(Ow!yX(9XBq=Ts2ImPL`Of&rE|4c2QLHsc&Fv?BU~}}l>I@Dz&*dDqp7Oi
zEuUEsmsQTkULqRqC!)`Av8w`s%&SOc_*4|srWEQ=BqR_(P#VC`ILcn4@!TXb0q6L@
zF*0d8CV4R?RVX$M7yCFPHlryveIhn%F*avAwooXp7!g;Biz{=CD{qRcoQSK~j;lh%
zR}010$Jkz&yVy-Q>{e=WUHwub^-{Mzpc8IaUMxauV1KV>`z;;sxEdY|zjQjnZBWhV
z$~d4)v=8~sIf@6|ksjiY$qyFDv$#YL>AO~*@arb*<On0hQi{G2JBsLsZ?cRJ4Ay^l
zeKg(fyD3*6TVhzZ@l6MN8&g@IwV0^$fWvmAMj?DK(2f%=S=1nXA8-LhK?Ubol3Ym8
zC@Fj<NL+c;S*Oo<3<=`IaC>9CPn?8HHv^}i0WXY`Cw@oZo+m){oa%y7o@Jyol&9dD
zQ*<X&^j@VH?4%e9ryAW&HI7ZeMT$1Xa2)67;A7nJrh6mi#n_>Shi8q-Zo_S({Ebw-
z-XaNYjR?VBZ?s<E+eJu_42XyZC6J7E=RI7CIZkb#p9@N#MNkYLySLW6+1J~zOr$%M
zfB-KK;$4$p{YFbCVA`I1aj(wbdKF<6!Cl*J6miIaW|I^j6mo|OFL@VsA8#v2Oukll
zRKP$eI7;ep=26h1-KZq&gbl#IdSBBf#$F{Fb}U%SMhZ&5%7BNOG_VN%j0E#%(EIeG
zgx<q;v>Yle^Yu>7kZ|ts$y|m;?zmI#gjw#?-Q4NS+?k!+kHUE$H1e2cd9zM=i+A&u
zuH-Fe=KW~STM^D*)yV&OGS{0J50u%$!#N;;q&E%Va$<j&W+dE@*y;)5#JixB(!MZ&
zcI06c35EoN**VxHBRO0s2=?8_>afWD@`#$~bg8)X`9+88?-IqQjBYd-ZQ>#fQ!Vx;
z?*0)fbdCtsKbDlSbBYbmF*sR7@#5^Q%PFR@#S;ncG-v?Q=I*SXRz1ho4o?OKp~J`J
zt|xJqTnqtmJi%~WG>~?tsZOqP)p`TgU!wxo@-HeQi!m%fr#4e(x0EPS=C4_H%em}M
za9L1RS#V3)y{WR$-7=C$Ia#wj!nr&uuB?c8+uW1qoZZ=?WFz*#6G{U-?W>&poet%j
zoSmMB4AO}Dx5=sQ57Q>^CYkYl#ub)Nk%LcKO{G#!4phz-`-P5L&PH%dV$(}^P57t=
z^C-Iu5yxwv6$LCAEY#+7=tEIR(D6=@#G4^tFW#QGs?LU-?(>?7BQ}!1O6wXp5Apjq
z^A%t(Khgl<?;y>#kWOz0O0j|E%Tv`WyVa{AHS3x+o6a@MS>*|YC%Y{*`%^UsQ*NGx
zqAX1s+hEP7AW*0sjiZ&uHBIB$qaj6Wcg?x3Z?b4IyR>InCOjcD_STx8i!}>uRhh4f
zLO76ePBp`f$1X@SU32J-iK^csb;??GSeH7rkUHF6S!|p*ml3ISO7Xg#z91GIICtU*
z8=AYD6NyD@OFz>*g}A2)Kj~SM)nHMp_w05!N;<pFb-Ll=UIRh2QDd+~kihXdpwVYq
zTc-oCtUYmMuaPMF+@DkT%NNLm!0CI20!v-G-h|`cPjs$MH-1V1?xhq}g9QeYn&MiU
z;`W?M=uJtY&B>z8>2#-u(@h4Sn^IbvbEcc~_L`}pErnVw#V#$SAuZ+EEtRb;Rnsj6
zt!xo48qc&^qw1j7%v+nXTU%ONlPMaHx>(0j4S8}@G#ayIQX%IW)<vsZ+lHpwhWFYS
zPk0A~U%YjBF+Savf9%kE&X&yYP5HlGJmh>iEBbOy>*c)5%f*nFOW7|!wX%KRd-+YY
zJs-+32(<r}ZQl-Q-_35{mwoxk0!o{1|El$3Z?6<DOFwAky0-gbGrNqPd8E!=C_t;@
zkZT9(UWd>=S8%p=08>bKzXL7SDV$IhDBJm~=*2YWi(X3SsVXP=N1Yh4uF}@d5{s@u
z^r85!&J+uwC_5OM4%HLuHqh=ia_u(0*KL~9ZQj;>cBb2EzZ);sW7F2nIl{|T2S?M7
z`%zUi0+cWA_YlN-1qF_9(X4Hc_WI<odVQ}c=Tk3<pam|p_4;e~-EvhT_W+z2&8Ky$
zR^LzgVIU;2ezJCdglm7)z5bY-{<yaOgqi-t{r)7eSII;DI#f`d9y|JkG+Th=BD^EJ
z4IMObH2<1pv6+Wmz<_<qKslFV)hHxnb$})|_)L4S-gU6?-e6PCU`yLz+sxq0{Xx3e
zP^&gXhxjT5dHpRioKuxuO?d#^$*t&i$Ax&*;p-5{9TXjeh<c|l8-*EPM$ot5_y862
z<FbC}p~alxrMBVanc<b?;qGfg>XL9B@Vz(V2sgU<_Qvb|TA;`6PR(5ecm1>d8HowG
zfis;aEY98GX1*SF9r&WHatF^EuJs@Od2d80cSPjHh^P+4WR>R~@;WEM+(+-~@Cfi8
z1UGSE{9wMp6sY%`Nwb^k%l!}(A3b{1A+Q$iGUz%M8V>s`HlnOD(%?QKcK@wu?pyN*
zZ`ZDgWXO@U)%pFIVuuiL&Q+xD1@5SF(GP02UixAo>rzhl5fgGScIx%XeK<QB#6>*r
zah>~w8Mna&>>+Mh;}>s(a^DBPc;B=<g8uWi_b%M#CL9$`1*yIEvt&O=xbNl-&!)kJ
zrQuzf6Dj!dkaTH(0?*J*M@{|72f&E7;oExu_xCPLmENB^a$(5o{?OsTdsBBr!-KBo
zeLx(DB?x+pyxzX{Qf<00i8anfH?I|p`PM&8y{=NIBU}7nYT(7p(2v2;7gHN&xkE8L
zDTLE&H_>-zUx~S2=aEM@4oW%Fpq%v|HXD$fXxK*(`24`YAEgs#m8D$#Wlq_Y$QdMa
zJe7jq7uepg<D`v3+&SfyN9CVEwFf1G84u*ZZ&g&^V*+O8XJ(M%pZRqM#dEn03vbH#
zJO6ww$&aa*^@oQ)lR^>U8I@9kz;#*WG(@%3>Df=H&Olj%t6ADoYm1+cu7}GRypadb
z%K5*M<9|OZHg)^YsaHCmjYH=`<ROaQfIDN(GL%WX4YJbxYIpIgedyP58GoD5udW}z
z+UoM0Tlwn#@vE@3t&1*e-sj*eyBBadZ~n^4rML6*H>~)a>E`~r3s*nR2d;emaQu2m
z;ym8!PMGfet;EZem9JljZg#<ot~~Rh!;24AGB1L^xef=A|17vLfW2kGo(n2G3bLvE
zONF{i#TS=KLzl|)mMUK^RefBlK3JlOe}A^J<aN{Xe4c*l@K>O6p;`AUCz5O7!*}~o
zW{1MET`05n#@C*_Wh;fRgEy8JESHBq&H-`X#}4MkhnG7JmS0uF`$K=YGk-8Y&c*Bg
zXg~N7nD;IFBTuf?a05RGc32g5_@|u8dvuBNcIeNw>Yp1Qe;yqC1SD2Lium82)-bQt
zgNv)|ysHS-Pq6Aws03>bp}2PF(i$pkO(=g&L~)H@6Y?jRjenY^1FTD4qUqq*IYZWE
z+Sg@&uF1Wr5m&T$X4g-fgR2nFs4E&5nY(LPyWe5AH#xNF&^>IJI&6Gskd0tumcMCN
zziGwV#7q3LY2OUN{IU=Gb?(wHN5x+bV-g(VbG<rq$0MxeHdVbI3KGT69ds!p4%xVk
z+4w4|tr`U=d^s1yx3zF{%hfPo)6O+Ha;v*UJp*fP86`*#yO1`1CF4$*p#ucM+)iR`
zCrj+4p5A%<b2~YICxds#g1SRF53vAt3ng}oPXoE1cBp>4m0`P2{C3OU>{=lAYJTq4
zEABOh?KS1^wY0PL+CJ^QWbM7ohc+ticek_kgzfj{?+>t?ob;gG`A~+$pV8BQ-d_4M
z9`<J<|IbuDz<Be=iTQ{5rYT4D;7bkfob>_wDx2LZ+tQnZ@1G7<S!`CT2dmbs<uKNE
zK5Ms~wa<&N{%?tG06Bq@kYxQRB{?nik=p}nxBrsZazp&W{z`01y#rxkckcg{*w)?&
zsra|Vwk_1#+o!9er^BZMjrvP;-5l9^N5{H*`o~A#d)xWk?}R`)P^eKfWM*~@b%*WF
zTn)|9hw(Xdz#FxRrq4Ov|1$d<x<QE&(^YY#G3yvN^g2OoG`tzE`X7lc>%TO%CJE;a
z{tu1qm1j|=>E{Vi5oQ6UuSFCBmJ*r;PQ1z2r7SJS*8M*<wx@3EJvgRX<^DSO@PfuE
zLRZ`oi*XJi*?%;)@u(rEM&g%&wC}~$0<w43rat^{iS2|RS%aNl_To@e?JNdmzUKV3
z>s{3#&39*NsJZU49X;awpc*m!@fcK^YXb5>@8xMeJ+?0JUY-Q^h}`A~^>cpabiwOQ
zkfgprKEitrvnr?$d3+NsweN=3mkd8$&8^(sNj;yT#XQ#EOo9%7lb_5V)Q~TIn0wlt
zxu|AJr+<U;aU}1t1YlMaT{bK_9K|n9|8xY}&)0ieLUoZzL0+`$;lbMZe2x_IF8>@Q
zc4zT(G&)LVHb(lf&upxWcYn74NBzGvwlZH5u;V^o9^g7NdK^4|;T|7$(q{H18KJJs
zJu*FBF_(Py-^8{gnRiG((!<d^Q%?xu!{GFK?4}m)ou77PKC^%u(wTLQjDH{jJ4B;H
zZz|K^`T<6aJ}7$|s8gJ6wc;`x%9Rl<0jT{2M2_1^bSMg){!rNY5MH>%-&7EG_qP$o
z<gPQ4s$X!&G>!q~KLqsseUDHeMi#^nD#W0!3?5MZuHIF!Pk%5RQJq=a2d*3q?ht2n
zX1HAfaqol3s4C<(EXYM~&^qEDVjCfYWB$JpTP+2Cxwro(v26#wJkdj0Z2J%$ZwT?}
z0UYW>t6e-MTx;F@4rkYTP?r<ddPM?<*80SwwQmA9hCW9;mN|`bQuLWQC#>y0>OPo>
zeLrK7<xRrJrH5$;K9JCzxfvG(_LY<%I}0(Rp$KP=OCadXU*$7VPacjE%qi^w(}My`
zO6)e>HYIcq0rH1MQtxN6g?e894ekG9e;vftRv-*=#t@=WX@4F)&qJBgdL-;HT6&f3
zx?9<721Rc5Is=T#GeWGeiS!Ampj`+Ekr`j~V}TlC2j%9W5;VEXvJ=5e70+L`e6MQf
zX<e#)Z?*TMZb4yhrEw#%^=Au2QERE2w<h>|)rp7u>%HnVZ5u<fm-aVnY^`-($e|1V
zLu{iJ54Pvh+wZ?nS`H#Uw6CJBrmprWvi3LM^Okgq<JOD@WCC7oC_M3j4!O_gWlk=W
z5ZE)sspB9V4iDy9q!<h#!*M%wpx&7z&Tm(*(O-Ff$eo1hraxypWWwd|OGo@fhaU*_
zNhG=eoJWapm@5^4^-l#Y-Z9kRLi^b{@*4DyM6)Hlw)8_5Xp5}oJ-c~b(i{1?=Axv5
zN|QIs^6nl*B8~zL#3wllW+YA)dPm78oa5x~w^|(U?~Su7aFIcD^30p|X-=&r1`QR7
zZ}s=-PTd6GVw>Y()9BX|@lPb@=t&7U_3MMzA4NtKOCLRHcE))<Ib*2!_`iv5)(l8s
zF)~7k&P**oTXM?b)qs7=dKxXEM9$htQRpq<{_+NiFM$~j3u30zxk@q7W<%~g8=&5^
zrON5eXFbKoApHr)v1KP;``T<|jSn4HD~KK9M4SY6zFcD2s|lw^#*(0CshZbOVsc2h
zFJ8d#8krBILSFFu9jhlaSA6E>+7I6QLVA--0;%TPXRKNR@N|wA$u<F!Ol3LK#W-lt
zuk-@VFw6rnOz?;}D%EK}sc&Nb91haf!4+l-kRMZcvHJxMY0h^14Rj=D|M7S>nZzS)
z8RxuXN8q0x=}ijJIOHevg|DAIwP?Hmh-|M_gjIc!E}JsdA+J|DzOJ-?UTJFL|LX|>
z>Lp~H`NApdSCx<DRk<zzvdPDl?g65}g?<B<lsBq_UO&0GHSqrB?(K~2v3y+JT)Sds
z07>cSn5N*KaE?C3ekhotCdGsl9q;B%8#CY)!OKp}bZAp3kyT7O0NV!G^s7>Zey@%Q
z1|Pk`mvnw~4v7Aqgx4;i@^F}O!X>!qqAdk)EAgicwzl0Ufj;9bE88UqWas*MyD_i<
z0|}B#co5}}Cglsc_hPd8P$Dn7fIO61NtU|H*)!pi&zIiOTDD%SKB&33VDZT!`7pch
z>zdFlQH$V7{_30Wjw!Otnc#(YnTUfDmHf%G1O>-T&O{pJL=^%4Q#psy!|y3w7ytML
zS8k>f+QVcy9xDkEAtJjZ67}B<Om&FMJQx!Xm5I(PT;egIS$p-6pb)wnCqE);M4JBP
z88>h))fb9DdXjLF9b7%JDdPKj=;5W$+^{4)-QKk+5eqzLW-w29IRVs<)#GKy9|rw)
z%GAl|<MK15qjp!l-LHc<T`~;iTc&N5q@()}{pt!HXWjZ}a5|>Ks-)O{+U1k+KN{P_
z%@aQdKbgA(w=c>Hk-DVbU(lVPQ+XQ7AM9N5HgD=I&nv}1THCqQOKD#Zp<~9(W}R8H
zLfwmyJG8ZIa-pOM@ndHVO}lrh<-v`xa_5pp0<13gIOt{JwD`p}IfxM&+`Zm`%KZSy
z6)$0ltCVZ<?-B|7U+VZ)yTlrM%Wprd1>DO~IlV#f&qxv|kVcodF+sZQbp5sS1X%)s
ze+sj6@P{%CpM7q2Hq_cb`t8Ec!ef?eaW4C_*%9H#RI5!BWs^tKd7Aq@oCaNPh{*YS
zO_Vzf`=2}A^tN03Hs+G<#-$}mTJ@cwDDJ&yEaRQe3wkF=a)+*dFD#S7jb*@mMU$#>
z&V_#)xdK4gLd!ARgoaQd6eOzwD}f>KoW2A=P?+l!;uW>Hb;=R0zjDW?AAkTyWTQZ(
z1O<x*-4M}()Q<s5as&`}N$31Wq$~vA%@r`ws$u*&i&J+Fyco0#jOzon(URb0WJsR4
zc)=^J{tc=`-yGj-pM>m#&GNpcz8a48N393H>OJ*$trGn~U2R{R{D_rTUWRZync8Yk
zrwu*5`+C-JY|J95TkP?hd);ECJB&*Qk4-x^@|^U}4f_~FI*+Rqg-9IsLzA}qB>2`J
zN^p*z3^Ou+l=MA=tXhALP;vHi&X*h(m=k7Xx1)-`;NvrL5Vi_CAUTx*=rZ&z_D3Y=
zIo?9D%mG{^6ir&&3gdq(b|J#a;DoU6dlHR!#NaXsWO^0&$}26Se|(YGolMRsAlHqP
z6>yvv^fYBi5S$)J&4+T71UX8gX!cXIOhFn=6n#X54lcsDoMIvqVcHaN6d7UFA7MQa
zfyYJKI7V76MraWtohBk&aNrBuk(P^*Zip!N{z&hPNFSM~fVoJYi74NUC;@uZHKFL6
zIMa)c(LqM0f!$G$-bN>3O}36fF;sAwQ4GZ~21789GL4CgiHY8hiC>I~HjPQe#U>#_
zw5Z@k0eBCUj}wIaj%2~O>cDO^aA{Cnc}AQ#A?`^=oGc@b)*tt*DXv~7zQ!lM!8E?P
zJiewszWr8wD<Yv|JN{);d^h6k3z>ug$FncW<A+V-w*{amfak5_gE7Yk6Q&RBs48+E
zR($B&rH2zsp>mSH@6S)bC(;v_ni7{M5?8hpSA`y~;~sJoK|g#RZe=8HJ3ic<c(~{D
zaL@E1SR;w8IZ=?#4mwG>L`&kDOyb!|LJB|P|0@D<dW5?BNGS7>Nb@7H$w%Vfl8TD~
zEXiDo9dhhsvJ5)}O9f6<0GEj7Cp40kt|TjKBx905xXcvI<`nJ86y2Q^Ehmc%l~g0A
zRO20>^%g{p#!5XqnQC<f!n>MkqmgFklxBZ7%`r30xjD^sGS!6uY4wC;6v2auAg{ZR
zy)z&CG(Q%k0qK_=pE9l7+<<^aK$mdlt54!b2V<Ajz|fs^l5hrDBO}5oBT6IV{@sk|
z%#83?8F4!qNy3@Q8ks34VV3jhE;J~y9<XF&=Ivxsg|iAZvWlIuO7CWsXJ%D4XH`vR
zJ=uX)3uhO_>Tu!VCAYI1?q)YuWH(J_H}7P(HfOhggFJJ}$-e?v>M4?*=L}_L4DSRE
zzrv^!0Cjxs_}$!z=G2Mi+?mPTk2|?c;k;Rmyg8@5PiDCj6&N*ACSx-1^|!o)OvBN0
zRyqqT__szV$9q7Rj=Woue#YSf8xpjdPGU2svH>#CU@D&*72!<f*`>0VQ<1w=PV)kR
zEb3ujs<38(=v3&DsZa*8a6z!}DX5UonHqwINKA#QU<y&e1sIV6HvHp+WTeuB9gYf7
zD=*SE2kGn<>53HVX%-te7aIi^8)p@pwiKIH7F!JzYa&Z*mWs}a6sreOY_dwu2bZ``
zl{i)wIdVi8w3J+QF7*~E^~z#9N&?FzfsQahLO_{+9O$NbS%7cZZRav#T-n{0vd}4z
z3l;otU>jT>lT}WxWXtLY`{F~7w@{Ng3fV|NYdshs0a+~-Ia3t~8Vj5!Qdy{3S?pX{
z8eCbPRax0mnVnU^kpfLIFRayk!WJz04z8lp2<+WSmoU9`VCH_YP)>05bQu+-Lab7u
zg1iVaVWcV%`d?8w^mTC6@J*0HWYt90)2ZO9FyLux_){GcrwQ$;3X$!#?-`H`<Z~Q{
z2(^0R9Bfpnx-XdRN8Ho3xT^VGm=_*)(cmfI1)D!r^D3B48B?u{=NQ32!Wa<v9=xhU
zv`?fu<nP1B6Pg{v_4z7{Llz%^hjA0xbVyI<ML=G}Ba#-?GMv>SBn`i$+iVP|Azl;5
z1pgb@X4h%9)@e`I>F(9(iPjrv)f<V{Yhu8fcvihucD=q9c*3v@FWO+E)nMn+VBZQg
zPlBAHm%Fw$TpVmLr<ZFW8$B%=FNHK-$ZkB91ku1N+7My$g3tXe;7)W5BB`P>?s?E)
zWuRzfNbB>Q-RDm=o5Ba5-<_@?U9BX?H$}t)swChB5<uv|{au<LPs?;?af*8j|0vTB
zqr$o9`BK|Y8X7qs+c;f<hzDYDDoKU+=V{7l$X>7Y$y2R+SX>CcRXM4(S+uovu=V9t
zPSLk=wX$txk6Q9e{92}4^IcBf5~xPe#4o8Bgb~@iJk(rPVJDZ4>Hk&u(%=`Rt_`$4
z8<xFTst2{C7>D6$e*qgCq539Rl>%XjKI2#Q31HiTS;LUkE5~6b=obrX@YB*KeoMCY
zxX`@WRQZZQZPRp~eL8&_l3`oqNG!tQDVMW4MA|yUW;(?GMQo+CJGC&K$FEU*<8(rN
z8|1Vb;6y;4tHJR}Io7pH?JVrUDX0UHtnEtH<$^x;Ws^;UR&rEEh5Vb?7F1#a0I2Rv
zmyK8tklfsa)T?HQd=&(d;b57?C&hz~Ip>9KZdqG-aE$89aR}1to;+DQMaR9Za=s;M
z!fXNW!nn9zPHE+fV&TqqFJ5!XGxd$bs2~%P)09>#DE$=IYO{A2E|5yRDFqtjH#P{w
zI|JY!0~%#YtcxN^-uG-(f8AkAvvF2};_dLNEvH&j<gV>Y%lcE*TI^L*6VPdH#w`O*
zY+&&gb2&~1WS)=2Aqx>98B`^m`fT5aY#ZWsr3suRt7}pS!JThiUyt8=J(2SoSN{6Q
zGT3Sm>;g1k5<nQ&;kmQJiaEoI3B!u6Y*NJGd9gRjVvTD5i>p6zhcbTq$Ny_~V;#bb
zee6rr6tWaXvZRe{V=Rq*&)9dyzH7=ZGImOgHCq}YLX5OX+fX5Ggp$;J^?u)<`?!C<
ze_^icIOcjiU(fS#I?nt!(Ww2o`vhuqZ|c5<#hsJoJ+;0gXC0aQjdzaQO`IKIi&rze
zYNOQ`<Kn5nxo7S>$|^>Rt}X+Fqrjw7Fz1<YL|>X08Hn7ojxzA{A__{u>4vvaplB>2
zm=1iKRA~t&Zu!FX0mJy5lSL0L`dwNYERJ!SMsj1>2sq9;fvU3GZLDc)KV8SkTuNkq
zM+jRCVwvjG2c7CIGz*N5A@3OW>GtP3-rIX}+v@cooUVllGm0q1K0X4mw1Rg&9)DWx
z2P0^5Tfk-B8wS%5kFTn1m>zT*o=U5z7Kl|vdJnyNHJ|rqo*FkSFaE-3_mL`nc*U{n
z)4+>QvC}dZ(08wf6DJyfIxc9G&!}QyzfX)d^NbXiG3gR7+bfuy`GCdL0-sZojbmI4
z9yWd2BInwy*g02W+C4pn7{7%kyxPJNa^cMf{|z~lMnlU&$m?6}7aqkS{!*~#?jtRd
zMPJz5U7oPdb$rcqJ;o*}hfpfjQ7tl^<=@)a7jpeuI>Ab%zciIJRq^#Oa|w-;0Vey6
z<zJqhT2ID#7~x*$R9%^X15-7DGFeZ7HO(TDLO2`R3@!43<DUCf--}vM&C<=+g^xL%
zy*Qziw^6b4sLxYq)GHy?WF7MtNA=%nyqPy3ygTmuBGh+!{na}eJQ#)t;}3xmz<W6h
zC}w^7`s;U9xgf%sg?fn(Y36Wq+Dxz7DEmYfSdZE9`h&Cn%i;xQkKQu3_vjsZZ$dC<
zAgL6Pd39gURm@sMZ5QF@&uLV5q3{~euL-e6S(VF!pcZfr8Tf%z%g$kqJ*U^vtW$d~
z%05Mbrr3kGv5?zbYLD9xtRM-kPBtucP6O!**8}V=YCIrRQ3c*(4J1k7^}F@hlh*V>
z#G1Sdbw$su<(r1K^F{4VzYFV6#)AY*)LDp29fRFex0@GVtJrF;hay@Jd64d^#8I%*
zngE%Cz16%l+1yg1P7}@<(*FB}CHYn4#EZ8r_5N%1OXV*te-9f;zBh1&n)1G1=Uq7R
z`MvYO4?B4?Is*%K&Pgq5Bd2n1LjQb#URpOFU|KQ1xedN-Tjq?Rt#G%=hcl8$d&?Z5
zMz`sRp}(4+b~V3kuwc{)812cP?XgQ0p3w}igWWKtC*xlhST<keT`4;H#@b^ynqAL*
z7N)IW<P=lgly^(954$yOz1V9t+m?F)<7d>IxL>_lKH>~Twcrn25Jk1<;k5Lbkg&5W
zPn=$Bhk|T6U_p;n#adwvSZou&TLp|5VsT7>X;t7R$$4wFo^+D)Gh!m=3-8p@A<0K?
z4t=Fa?YxbDrybvUq5oZ}#0)O>y>T9B^Y6Qpyx%@IPwSm;xNzfJ$%PNL=a>JUKoFfr
z;8MHQl{f9_;H#5N<j^lZiP^0tW638HXlA%u_3VB;ecov5djRKAz!Ra{&tvfgN~8M-
zA0#qhV?5w>PZPMnY`La3fz)!8{27Yv*Lx~&N`a}?(nZOlk5u~WH(m!HwItK%{5Y%!
zi?0EHa5ds?juk+lv9}ZGlETcbK$2wZWv;hZMO)+P`|l2nh*sB&iAmsrIV2>&vvb2?
z!49zK1k)H|L;}A)3<h`1y;2-klMqOT33C;Rl8F_WRCV>3xLd{P|4@XZdT4YE;enF|
zuZH0NK@pzMzgT$r>gDm*EaquhGgVay3<9%{A|McQG8j4IO}VX~##?J&K7QK6bT?X5
zSJQ@)IS!~ej#G0uLoP;=KB~9LjAjcfo-`zljB`T&p$M5Q+5aCDAx}eq#9NSq30dl#
zPp71r;FH0cA!DWuFgSdt2m8QO$}z5b!BDRAe^7**HW%292I_yCkJc<^q#Y;j?uBY9
z3coNwb876Is<Q;2DHPqCr84mHkX*W|8rCK7irW46$EvGg=WB4V9GNnsonfHWq4CrM
z=$T3JBM`=47n%niqXH4#ny)o{aMeJdlloe{Cd8X3hszO5w}TWN&;G2@mufZ(zSzg!
z!qSY-JDMcW&+cv_rzw&#O`nQfDE{*ogjJkY4Hqu#-z{W;Z->TeC+Y?M9oO#s`CR(>
z{XOeb%K(g5ViEe^iY-=4R=g*d>ph(gJO1sIl1s|QlvRqpj3D^}NQPx~Dt$9!{k(yL
zk|D}SslL(G&|L|1^_1?Y_pG2jl*JF=O6Aydc_)6XzPoa?m!H9@^+oR8RsDd}v}k@u
z5<`-Uu$(Lse@b4+#?hncLy;p`sMs<AWdSIfj^l_R*HZ{_p({9nMDgIvwo*co(h|lW
zh6^{9%3JL=2Y2p}pI21k$MB$#y~K?2oPi)ES%&?3IxMBMDixaXDr_vU<nK^pb?d{b
zRG28UK1(>Pq6sbRxc<wo#L~sEQil046XpoRVeczoy&Dj9B|lJpm1tU!4*`<Xlz0&B
zjVR$}B<^>LW2b=zXC4gGk1yKa5MRorN`?mAg{q%DGA3sP;pO~K#dh%bxq&A+R=ix&
zwIPgjnXONO_s}g?eDLkqNcJKFhW<P=4C3irx{nfj&&)QleKlqxahp1NMi>0?U&nSp
z;M_5hklIn97guX-Pl1-xVca{|4n7BxA*C?x?9&g^Cw{)lu%pZlQ>rt#aRL4`RImoQ
zh!usH%V|z}=2_@}5(DZzjpY!cBEjX*<v$=t=>dE4R%a4<meEO8={66ct!_V*Ae*m$
zk(?hr=0Ldhqz<$|qX;~xN=yb50)iya&t!`^lT0dYCpXgVQV*DXE+uRI8IMf4@F=4^
z<=iA58oJhnD6ev<N)?VNSmTOEubT2F{GIjtSFxQwo78u_#wb}8KmP3=JFjs%W!$v!
zo57cYNY@c&%{*IXRO!?EG4W$96xlJEe;wQ7o^46gQ)P~KM-O@#7g)wSvW$8b%YxLW
zb$|P_|Jd^|D9$}caOt0n4+x2u3yK~ZImsK1188)-C{p|MVRalsVT7FWd=;do&ltYY
zh-EORh5&TNIryXur0ik3zPSHcscv$5UyAgx;KoTmg4rFmkSo_MH=y#(fg&{wjjxo$
z#41ufSq%s9lC@>TC&-C%y0bdpKPLVaFs9HQ=|>!yc!Ds+@QoZRgH$^r1xJ?|r~+x3
zqOy=iY>pb)lB-yISZoV*{kSSa_?0U|NrMiPioKNrUo}*4R^)cYK)CkoO}+^~95;!+
zf0g}11)-Z)d;<9kg53kg<<UhC!#LhY&2q_~RCth-QGQqvvuT#c7~4;5ugPgV92T64
zXwPr@tf*?m2NGZcC<?%%bN@x*y6Kc_`0&{~jWP<ciDzh26kr}X!9OTp@F8Wi6pA}w
z0T-z>!7OP$9=H$}J$vp0cS9IH=(VPp<xT&z4s>fN-{E+f<e#V1m4+I#q(^if{nnU6
zz3^XFdxk8~`n;$)(|peGC;pMwn-2c%C-v%l8~I*KUs625tWSB=%0F`V4S?WUGgIA*
zZ;pJA_z;PbKj`zY@K9@Aac8aKIk|~4N%!cwlv!oPlQo|MX5=Me>MYXNj+7MyHkkRc
z6>N(}2BAAxPg6$#w5)a3jo9S{@49lhT_-vEyUcZu=o#rZ2W&5eK5c&AcY!FfCVG4D
zP-zU+%W-mo%OSE5&99X2@Gr4_X487NWC;6bg6sMCQ~mB$7~#1&;UsS|aNwhn%nRek
z6GY(%*%GvD5HP??gGgWiwc~gPzR&_&Qk6XB_<0BDTzoeKPH*zJ-=datZg%%nfK5-d
zAE60s-(h^=e@LUeToQfGEBeZOvf3?8M~=cNwIys5l2&(%Xm`dwpgH``<m<1|Kl@oc
zt&w9fQh5))Qaft^kHeF1en^^iWV<u(?i#+`UHCcpGwhAOUzU~CGXX`5#@Sr9t;j2|
z7)D-fl5q8}!{nBWM_dBFMQRV<*Gr+9+unQ|rZrIb`P$F>Sl<1P?OE5a!YBEnqmr|p
z<DedLAaE=znalNro5<YGgJaV{xe+;TlAn6U@PFH=acAFS(NW_j=Z{`1j(x9iu=k;E
z#zua_lsonw1NOlJ5IcrS#*kl3l9t;``p+&ON+_LjJJeC8MF%OrXFVe4b(F81UD18{
z>ybQ}pKF~468imXhVpk)3Sja>O8_9_d}lS9=bG_<0^5I&`J?%NPy~&0DAeltnB5tE
zrH0)nA$p}EtR^nH$e){&i?(*7_vQ1t{(~aO7S%+9%=w6C*${mO2K@i8BK!}E;CxJ~
zdhe6B!|{27CrH5ABI$oo1ii1ECI5>egkSG}xy*pfK)ESK4x*sa={tizAAO7eEjFvr
zI8^T>Ay{wX1uP2O2V-c67ak=~p^Dng0sl>GQ}lvb?!Isp93V_!`Fh7q&wsymeeb=}
zh29Cf>F-7TuRd7D_f8r81?v|;kYA_&2SpeF%a0olM45QYVnpY5(mIimq@+5nijU{7
z=ss!qyLI!->osdFknra}tG8nGK9R=y=358DTG+pt8p>`X#au~SI!LHRlkdoieozkk
zpNde~MLJ9m`mnz6T>ZZ)!m!wz4}!qnq782%T$!~(y)y7#`ce`ncbwPOvEQ}%ydU|<
zEC`M7?yBj9-N*Zh?Lv$O`25W7v;O$4a)aYQN(!Zy4H_o&OyORw6!cg<{}FO2&+h?j
zKG_$aEvD!1G!&e%;2orYR4JSAb4}ENz-}I5;`!eD2!9cF{r$+G%C_)y%bH)8>bm%6
z$1AOSMq3P8t=~QF!ME|hPu=_bVGV>Fy|qF6+-D$lClahpff6WSObDD1tQirK|D3|R
zL;0o<8e<YFQWJGeI3_FHTbX4qKA$MLlZeG6$*kHdlS4&zK(ZxCDn0BZ2@*tjK2XCF
z(v5J5N5S<;fVMXff^)iuh=ty{kpARAP#77)I`CZ%a+IV#5drme1I?}=AS6&c&6vOh
zgc$?|+3W$%1|j4Ep}+%~5Oq9^hk7o~Lm`F)@<xGP+QE642tuTVuN!>C@&bVtyB#Vz
zpU(BF0Thk{JQ-=8g@6mzn1`JHp<2lolNKuv;U^*DC<&39a8Dd?M<!Dw-Oq5yb%GX&
z+{vV2vZ_iVaSB-%pG9ifMWK4Knj1q3BchxxLXKSwG<zNf&=Vz4;NDgHzMkyC`E15`
z_AmzgFSDHoDX)76t%LdqInzox54>|mIC6*P+1U~-(BZ~FkQO<_i#(nRb74fpt;6rs
zx#-XPCT(b7o^bFmGz3th=rBVUyhf0aa1aRNN$1Q*iF(sT`02LlDA4A{zYf!ynHHI<
z<VraPCrP$pmAz~TZ0!pW<}q&a!C<gqx@JN?1lbA#3y=ig3kZA4>KH5{%;cnkQ;cZ1
zliwCa17)~EgU)Up0f+}g2spE+5rm#$j_-ohP0%^$I?}BgT{3U?r@>DGO{t$RIRC2v
zhbWAtQx=|O?FY6Wg+uA%^2T0Cg=vYI^@+V^IqvpFo>4{KrA0oOIfg861xAirQ&F%^
zai|EGkDB`{EgXbR4!Ps_fX*g!Gl0{HE_@Mr&|*!vk6_u=gMn@cQylQI3bQztm(7GQ
z5>L4u6}mLb0pFH<Bq-&ARf!-`^WF&&835ym2A_^)3|<${ANTkoK&EMOGa<o8ujJy)
zK5Z%chA0_ro)xgUmUR?8w;`HyA0WQN1Z#*_S4%WKMvCGE+mVV!_pkQIUC0NE>)SiM
zJE<6q;)3C>u(~4R#xn2PW?d}3u~vGccBi0m9L!0}BIs1!+Q~YQ5p`^-@KRIG9ZYf9
z>0${i_-|8@1Q{|hp2IGsNqw)BAQ!XY<PJ2@^}P6OFzEL?6YWOdRG*UDMcRLVB7K$)
z#?Y=TB30t)0GY)`7Xc!M_a#kfFb!D*H%ip=yD=FjPYSxQKfT_#iHN7`)>Oe<Q8uKi
z{niNVXf!fmUF8H3XtLZt`9p$900C<jPlqWJkvS|*HDxJJ*DCJkm`;H6T8$DT=pccn
z8D<F62N15za(x<%3xW7DRO>4$0bj2Wlq2xduB6WySoy*esHz^EB6<Aj@r4EA2+C^j
zjlA8)$_&c+(}miml|{Q0VpOHm<HE>~-lQgPu@V|vnHH9r<Af-#Hm8mB(CXG`d|BQ{
zw3E-a!yL%fA&h$Kc3<%YU#=)$I96O8TX$zk*&-V1N~z7_k4s5D#>3*oQY6Y2A%{8H
zoQy=3Oq|+WFyh){$h?cPN3>c3O>yCfLfX5UKnSM;NsYL!JrWIykmi^JBQogCG_ttV
z7Sd=_l$Lxf&jqAHtarrKMb&{Safl)rjwgM!G`j9}X%!DS#Rxgg?P$rXRWQQ?@OfPY
zM?eIPPlempy*0jZeJ3K9vq>SQJ@xsG^GMJO+os~(_N!i%-_t5h@Y%&bnp~A{N>oeB
zF{Gub(pxC790gQYmXp_=^Jtv*limrVWuN(EFGHoZsKj)7#lX=Z9tPlozoo5h)7N)P
z><34pI!6LyAN7!c0OFJ+@kiC5EG)v21Sf`ooU3Ys7BO=I@S_$;qhz*ls*+-Q7RN2D
zasm$FRf8FPf$(tv)Uk^CC=lq38ocxZRT^IQIQ$kFR3?px!&Z#78MeRRh@<NwnTYgY
zMVDaZ5-h}%CM@v~d5aFju{hh;Imir{kArA18Nt*r@eaNd&w$N=Q+s73K6mI=E%rF#
z6isBsM7x}_)*^DU+V?BA>WWS0qCB+H5xAFk^^2R-`<sL!JIw7M%G8^`XwnkHhxJ~9
z47q5^bjamQ8gc<7{<uRz2E1zl6|d&w*6sRRUmV*T9syvE8QpTFRzvV5!H90VFeNx1
z7Tn(QS+3iGx&K}xd&8AH-g9!%4+<ab?gQ5(8fy8yV;k^L+OAh8L|9hz97WE7G2(mb
zeL+fSvS=mhWCY{(VwOS{t+pW8x0%TysVW5(>EjfuE~_dx#r36HUYNhim4pKxf)Rlh
zDs5E~e!s>09XNUt6>1H8&~#Ybx<)=2nN}v67K1oS)6~en?Y%zmrXdsWa^r^g-N^DW
z`G)p$58Js{?@~Tz6*V+D%4BQ4Y`Wo_T~$Wgz;^bU@f|+{{(79|=R?bT0BhW>{QK$N
zpQXW_PxshJ#-Hw70hL=O8@YugfQnTfYK5`02`7-?3)Y;AoZ=C<E6fIoN5eHl#{CEo
z#>A-p>MSIiX6zb-h^Dsr=Q}3H#CqhYh*H?_HffAcRYF-CnIro)6034z?|v4b<QiGl
zMnQJ6^7UdBui#+|UWm|S<V+{>dz<mbuJ-XY(ON@^x}aD?Zp1*gMAmve(Sl>WicZFt
z@1F^Ru0G0N7Ywj6WeT>xdK|%C36j4xCVYEr-Y=^_bX@%PW!9NQll1YdA5EFv<29mP
zhGKlzUP*s-fV!W4csQUl`$tjJ+MvN1u)#}dZ3#Y_V-fB(O(xf;&Ah7NEWZQBe%L*)
z<?Px0iTF%cg%&38AW7kK!KCObj$0&z!G?GY4nZKNN^;gL+(8Ji#H%PULkqYI6Ubm9
z<`Oy1h^KkJuMIL+=?Ml^yTUz4Y|z<LK%0e0Aqp=3k^_t4Oke7&yjWwl$I<JlF`)xF
zuPPe+_~eTX`AKL))XCClTt4S@ppEMIh^R?P##imzHmtNj3_R#ok85AKzveTeo(p##
zM999(6pEYeId?bKKkJ)FrNp0?s|p>PMU{pX-r6rKs~5(%ScCZNd)KDsj!SfT`rIro
ze;C%pC&765r(Jr3{`#a?7YyI&r9810?rrljcMX(DY)rk>|7287?U^~iusrE~67VF!
zg~{n6<>Dbc3VrL7SztwLI>Kk}sy<y*04tv6fC-5c9{_>9Ez#jv73ew#TwF|Vog*t`
zKIn+XlqDdZI+88`cXnY5(gCDr`$#TcB$#}#aDCo&n03=i)SL2F=iMDs$NsB}V&tO`
z`5S$@(;Vu51s!n^>Jg1UN+8dqHtzN6;B^U$ETk(9F+_p6;yFYu;9N`~_*E?bmxMD>
zFMIWtuawPJ_slXXmZk2hmzxhdc@09cKqH%nZ*ONGlAxily^fXyA9_vndRVlXU36c4
zC7w4Iq(Te3@jCXa{h>cA@9zeJRc}!?az6ze6<d2VYOTq#SJ$%FE+#1)%R&0+BDJE#
zi;?2+0Xq4jD0TB#uL7&CU7H5nFfr7w{w_2$HB;q>ASP>!4b{~sU+)Bo2V1O}oE6DS
zlt>Ro>~bJH_}SDVs~$=RJq|ZCxuBQ=5;qxw-PKncldiaP2tM}nK*4Nv5?(vAoii9v
zB$Fi7acau($^#qSOJ262uQ<Y0m)!@yUF>~XJ^#|DWjR@V(D7PS*J{!Dm${I?hnsYZ
zVY}WX<(=-m>wh1w^0XFRX~_lKO`MWY@ZTRk=mEx`GD`9c>_QO^bnz%o#3$#^E*1*p
zBQ=v%pkA_a+{gz{dOu9H4^Cw~k~TWCs}T$sijsW^)U&qk)Z;U1+B+IK>!OhTPlgka
zjYRmxdoWWb#6Q}kp-Ff!2p?yiF+_oLHz>Vb+=62dne`yzXE}tib!vw}+*sw0pc=EQ
z9NOM&DrH0TR}XSYKsfHo6f0fJn=a@v3kj4%WL{Kgj4BhABOG-TuJ;34T|i_S!s)!&
zrIv!Z$!|CA-SCvm@|IkexVyaciGMU#_FDKL`t9zV+MMF*Zae$2$h|<La}1NjE;9bs
z0NXmi^t=39b+ED^r0IgIkZ|y}#iWM#JgD6KY3cJ6iFOI8^ssX+A1y{M*xm&RFC@f%
zpY1oidUoRN1*7>3Pu?e7suQ;|wwRW)e4~-F`1a3}Z}B@@Hsd3<FW;V@mwsLfa6n^3
z6Jnr86GYQY6ZDc|s8?guP0>6FB?*2;yc)5n*#D#0Hs5Y(rMI;RR<?K7pxFC4ET_=_
zWo(7yg;Z4|OZOSu5>1c^pa8sl_4*CK@%9CvEFt|rjO`Z*<t0a{ultPc8sEaw-k*1u
zK`09`GdfE{BO-P$lgTix+;=hee7qC!mHOd7F#lm}^8&FFAA?T8<wTUM%5qXJK9=J#
z{m~iw!svl)zWSRfs$7lfl?qk6z7y3|h%=@7PSZ*%1qw5Ir>-B$)!B5PyJ6z-UGU~b
z`C7@qvsr&cZ67~wzFObcInkaDrzyrce=v^-Sdff9<gqq{G#0S91OAU<>q5}wy&pZ~
zg%<Gmr`YB+pW^jm>U)AoBOFO_e|u67u)V5HdU+eI(s5(a+iOq$F}6I%NAm&|^qJwW
z%QU~dWmtaVI#+)Em@LTG+xgI=DG3PI;T<fifv+f2e~K{Dc(wQEw@IS~i3@`RzcyFr
zZe5xgy!-p-`m9y^{$D@?F>CYF+Zh7`HO8OPYuRp77M`p`4uSjJo^L#Y9(j#snv@}1
z7u^<98%H<KiEovJET;e9KoC>@6r`2oSw73<C;6v_@R-t5Bnlsa6l7_lyxb_Oxiq@+
z?g}2R{GNKMTMObG?J`^pi3ed=`zQj^!}a2|;Sx8bXH#8b(LQwRbt4zE;Qc3-v|Q?4
z-?Uuj-~9_z`nE@Tr94yyP|W<x#x}gl6TiJ<nhR&N;UQ?~0!8Xr@YqVV&JV)-8mfuP
z$6CYb;#!6B$R+o}^6<5f2iZ}&*u${PK5I>!$2QL<KhH5*s(4d^{a+M8#k0BZABqqu
z(7=ONI%<wO$(e|{8p@4h-LJ}2@%`Lcrepr`ABqrg)pS|t(KRJJ_K?#1XBWTrF4t#&
z?OW@f{M!HZ0oS*I@6XPByZd7~rX_#p$MwTfcI}i?NGRHWeTYM`dHugA0`yY%T5*Lm
z<9=)V<>Vs|<S&12+jm=4gImjCIG_ungK(095>RO(p4==`DEeuR?1Z+M`<L!>0oeiO
zhiqU?Og}bA!so)9LH;2zQ+Hj25b~7T_7iKnfbFM*ceL&Qha%h#LlOR=2v!Cl&3zO>
zA`Eh&UoGOr8QkJLJaXms&o{9@{~wBQM0Gq#l*^d*kFjm(f%A$QDwau}y{yyxvYGU2
z`MMh~utJ*=na)g5`m*z}@!(&RO5`6yKuDj_W^CkQZP(GMv!b6XdlXd-vb)9RpR-8V
znsSy+Q!V6h{Is3&aGk1j$x1GVLrUZ9%J$#@RzK|UbIDr=Wgf20#uyB22DVx25ySZ{
z?wXYp$j@zFIk7OeA8xzI3Og}bczELw0N2Ih!Mt<|#|avkXPf~&w$7$N1;Gdfvms96
zdJ<-e2G_CR-LLy3O2yJR_{|uQl=T$s*+mYkjA5Y`i&W_oj6(zz9<#tqRow3y1<nsk
z-lX%oTO_I1GLm14Y-DIAgTcB7$`y2vXdk2J3x0tD(kJ#&gwG6+B#@`P@DZ@*b|a_I
zM}R31r`g#WQG4Nmq#b227mfkMk`rM}@8eo)d0&6*oH#XK1@erXly*d?EeIt=Lop0#
zmX(=E^Rx#R<@|@4%%Pf+$jPrQM}RXapZ0x3ehdTfB%58GACJ7ModW;s4f=!}=8LCh
zN)??<eKGWb!<ys1G!~qhL3@9J$NK|6bR$ht4$B?KR(ca@E|aDvrw$$~;4sf)Qq;I9
zai^{s%74FO04B>oIA2w1kDZ<lF54`}%+qwAKRnO7hN@W4S<yMCG!wnBd1Gfl@dTBg
z9kIMw39ePdBZv&a9K7o(M>o+bR4?atnieW`)ksgnGPC*t+^Jkf$3k{C&$hE#vdf$E
z1j|?@=}4`V@kgDDYu9->5|I`5LxR<H7Mcr*Knjkd{Vyw>%Rdp|>S2gKBK4p6zR{S8
zN*3m%!64$&`8+*FAWm#L$QzHU-AWWz#p;*h4o6@5;v}`N*fxlgnuznm+^nO?Hy+Y%
zEB?9<t0e;31cn-6eaQM78N}mF<Kr$-VT*&2g$CSsu!F_&X;u=)5oXwaU1xvfO(k(+
zLK;M%s)P=HN9cGD@rvRo(EaW_+)VDKl_nq@y`BzbxU}NTxhzmFl%}sBo>n!6geYTy
z>+H5bzEb8ze{kjRpbnEiAq+G)D-lw{?0pp15_<j2>yPHod!OgFg!3_{zeYzSDRX#-
zSDksYaunVFbo6WF?d&%{iX!@#a|1)4a$VXP{5kmfc3|ws?6;tLQrcg+M*JeuDXA!g
zvEFPXmNE{O`8uP!FS4fmrj1OO3!cKnr;C}3@=Ck!-2cRsLF+j3qOM13x}Tcnq~R8Y
zO?O5!hd-KFKd}1u!l{2x1j(?2`>#veM7WY}f@;g=p}41r#6SK}jq3EKi&dzM@e%);
zBU6SX*Jgp8WM_F6L<XJIRk9<oH=L7!6aUo}K}vD{K!+*Lu)4?`%}_cGq^g1Eg;DYd
zjj=@OGuWA!lFt%fDJx(N8Yt?)@YSP<UZ`<WQ!wT>-w%9hohUHF`OqI!Hu?lAc(@^f
z@JULJmZ~$(nt4Q2$=O<<$i8<!oHp*RU~&3kfXsw$%uape>q)=uUE*CCRATFX135H4
zqT(dRDB^zq8)x?Ja}3Imf9JP9cP;7OqxkObWS%dVy?*yC#E0{zeEGZ@PUG#LALjk|
z@bdDsUH^!rR@lnxuWYBem#?>5e)QKlv7)o*XdE9Bvk~VP({4mcynB$YgCjmx3EpC@
zQoDz(=wM!1+6d}W6UPa`YB4N;J=nsZV|tz}BKAHP-LoOgO^2bF&12DfN2clcH66hv
zUgC~X0Gfily9PP{!5<Wg|Cx^4ryIED?_YRBOP{zb4Nf7ei66p(boM|i55*~$k1VF2
zQD-dcWio`{0|?9?X}T+8#P2sg73^-`LPV+9#5hwrpC9X*(D*!SlZ|YuUgYDrnk?IF
z^a!jCoguGlh+?qm$NfOiI3>KRg!I+t&PyDTk__<ooJ_1bZ~(0xz3`SErM5bBr;>et
z`HqFp!#nK9T}aR+zCAh&Dd*hw`AExO>a9<k-9KM19{Sc3<@9<@@;XEc9}mmF5-a`0
z<;4<ga4Y$WQ~cYA{*go9B0IhXd`dok_x|6zJIPY77t<@}e}|eNH*y2N)ZFR)e(c-b
zTVij*Ix63MW$WGhc8TP7FmZ_2Bp6I^F(JE%6q7m8#yl%7BIqDv8rh-1(kdm0rzeQB
z*cC*DsmoKKKS`<(U+AJD0c&o)iSR_h(}0kp#mBvI5%xAf9M1ju2Ewq2K)~5YdxV(M
zqpWP~o(Dyo*^c6&*jmXWOc{XBPsG(?kGOR&@9hw+acis55C;Guuz_gRgk1lr;jkai
z;@Ln~7^sBg&P|58VByhS_IecWi%K9Y1r}ujbR@&PX`#(e5kxW~_-V2$06x{|5QjtT
ziwInFSU596XN(vS4vHotu4?dyZ6g8{QY(uc_K(1K6f<EKN>*Jqo(icED3E23b=V#C
zt~7&;i?VOF41T~b586WIEC4xFwj3FtD`g`X0Z)$kKUj8hqrr$&L=-7foQhz00}tVG
zi|ETT*z67hF#Ifsi3yk!zPu0?z(U8~*~#hZ$?Y|}ykqSA!fb!KA$x`Z{g9E}r=7>_
zy3kKb*fe#LA?3kv0i6Ws##q8qLEzW4gw53)W=VFN68On{;GZ5NeKOm^NRBMQ;kN&F
zS1gL6UGv(!c{}Gp92gmC2HE*NS*ktvkr`N;oGpU_3q<g(TLkfVQFd@wIxyLMm}|yN
z5VVvcg$AUbQ+a%XVQp3f030+DY_<&Pb5}odCuz<Vk^D>|G8|-z_oCdfzG3U|)Ec<)
zEL>ncDcS;lSrHOP@e*G53Lzo7<Ux8s7@B1dBZahf0h!PoUB#$)S^?1lZt4o>#@cXJ
z*@YH*9R&~$g&dI_`;V6+<405&JSPqxbFCq)_Lz+TIfJ?mZ!Y2p=mrSuL?63o0@Gcu
zC$qu^P{$As8kt)v!G$gC9GQsdLPC@o%oKH|tAXu+#X)jCLE31zIx5Z36UYGPI5KV4
zs$w*U4a(_`@3iBLfm|Fim)=wU&Nh1i5&tO6e~5s9;oxpm#7QjZBsS+H4JI&hep{>J
z?y>C2)8I|*%d@>VmQ4K_UQVNSaoh9d?-T;Ik$GQ7<2H5T))BdnR<F%8-Pl|Wcw88-
zp^@-8&8ZzjTU8ENJq><3e`Q|b5~rxsqoy0%82`=rs-=ts*meR}Q~qHS3IdP;Cd%cQ
zpo6;Eo8HnAEVzJ~vE+lGpFjofG;If&EuVSK!97SBka;BHuV$XV{fnZMnP|^&6`!v?
z0#8!j6tBw<j-~^yB!HWAZg$Dd>@-`=&olI%kY|1gA)_AErCFHF0fvw~^-=Dw^F-S+
zuM}3g0M_pWA>v9nLYNA8FaRPBxM2*lwt#mPra<s^WEAK$-0Y$dJbpJKv$QdT<-pmN
zt`C5%QK7-ujI=RWA`D@$|1HBxD#O6Z-eDQ@h=5~0Xu36o<<KugxR#Dory&d*J;J;|
z)|6<w2!w?)r2QuB0=%Rs3rW&O)Ovy-3|O>$+Lx+xjq~U3-pOep&^AlJYf-?AS?znX
z3u6x|M#n*KKgQl4uOQN5KaKcr*jK$!slw1;v!yjp2?49Axzj)Vo~ih+A+vRERxUiq
z`{|qgnvlIo$b;CkX<G|<>{0Mq{gtT?EfBE^9+ir1&GX+(tG<i{2;98A5na0!*~ZaC
z<xwD?Csy-#)cgRlG0N9eeqTY|B#S+Ck&Kc)Eb1?ftx_dkUDF9Vh^tv-T>FWw(&}&5
z_;pR?7Wpy*iXC&6BL-WT^T_Pn%-Vv-aPq6n*J;<-^~u{@2yg5QL#nTPaiQQP4`30`
zGm{(bhE4;`T~V&uka(OulmherjqsqRr*D@=X(LjfWd_+voC*qu(GbO%$fhVa7|;~Y
zfjADzESyNL^^SBO53Bv*-oq&?&=$qRDCDf-r@Sz~Y{-$?4|vOWCy)^5OH$gix|v73
z(@BWwcWj4XJTO8`zEqzQ_s}}%=kc&1VZ%AdPDjjxCdb*8tky=Dg7URf#B8WaMYs3K
z%?0PTyRj>~6%R`MmvwI}-8775`2YIp|0OF=)Vy{E=``2v_j@#MYJ6Z;2~5ORt(D$p
z9=|7)6}Q0A`fT^ww~5=}?ST)ev_Y>+3UXES4A34Cn1$b49&!GVnee&p_O8kxXrkst
zA7ir!ETh|gj>|=1S6Vge%EYZIRrA_ayya&mNW96C+pI&Tw`1dF^@%KA4E>_?j~Y3J
zTKUrqQPGMS`(c45-a|GfXZnT?{YX@QrVAy(6Wd{_GU_HHb%cJUqbRsA>9i>UP$z-T
zjIu)<SM0A|_sX6}xZr?jsvV4cG=Y9A^(Dkj9$}3I1>W=^qTuc}a0nF$VFGS-4KZZ^
zdVfH($jXuGb-8c<Dg<%vM+ns$khW~TS?q<~-=jeW3FB^Z8k)CmyPE>VXX-O~m`Rmm
z#9B`T8j#JPM{%)^#!%{uDItY7!_z!LA=s|FS+M=qK`E=R)i}9h_c#|05j75jP|rj&
zV2Q^eBcWy)SkDtKevc=z*7^gWw_1l3vJY@&qa2|`X64=S^AC&nOVpsHUln0Am>a{I
zn~Kn9RD@wcqpJ{1ZOhwFelIv%1^05I_T`mm;Mbn>GvU0)+(W3nYt!TG+X7kicj!m=
z(jNU1yLP~V_PeibE$nhXaQ^pBtFAxe@ClnwQA|0Dyy3LAO_A6y=G4O{sJo5~j)UhV
zs8o>Y2rPnUzsDe90T^E&m=w7tp*JkWJtFq8gEzf>d%gP5=W9wlRjSc*_!AfLdUM;f
z8>jya0XkQt&OXsA<5h~y*RgLu-_L&>W~z@vgm%IrLc1)dlr(-eP{#pL6ameIl$yf`
z3|L<uB5^yUU?D2`WjJ|%DU~ohYTn!sj))K|y<M2DNAbE5o*cdUtO{NdvraTXG+$TB
zAk%eIb#L1LfEU?u<YgmcojlWiA{g`L848Ge-G!v}@n{_6%r-kBqaIfF!#y5b)Eah_
z=oJ>w4U<)B9N6ped3p=q<x|3RC$AS1aS&nz;-O$#G)y)gD8a77dFX&6ULYC^w@8F>
zF&FGW17n&4!WGWq#5dwuSGpqy=Y;)tGV)e)ugpFRz`TzQvv|vG@pk(Q<ekn9I0F=h
zgP)|m6@bowra`+BGpjL`%a5u+sd+QM%fD2-=~Y^s<haD;{N{jez{=#CZ+G7ae$5h&
zUlc!tKd_to<=oOoiML-;Y4YybGShG5E;Bx<1t^}UO2oHmUp{&G@;i9W(uwG$bv6H0
zwS>P2id_GYL^ChvAWJ=pmv7us{Dvr>cK)owlOrk}2g}*KhxPLZhs1_V@-OON!(`SH
zBoein-#dJjKBxadWBS8o{cG;d!{!sUxG#w(_SeQNSQ%(4(T%J<<DO+d_N*JV_Qcut
z9ZU^U?K$Rk{qj#Dkuuq;X<y^y5J7dYMH31Kt=;>q$+TkQv5yz!Q<Ap5>_uK*PWKW(
zoe36R8{S5ot4I0PJLJV9&xMcQOhYDyA#X{fwM<)69$I4=Y4Q6Hsu6ZS9vKPGsVyNy
zDj*`m5$W3AXla{ieH&bomqDXL<7EP^Wg)U2eI`sIVCB{@@-~{96?*(O<Hr)`xwx;2
zc#J!+qxbDsg`pG!zFPKe*V*t*w$qtG*4M9JN4?mspAId7F2o~mJG1i^5m(kpTbI6>
zIdLVt;b{#VR7^PZcS-18<+~@3WyC0E@$KG+mV)BR80zpvjSk}9C7vEpLB$UInS8mo
zl*X%_XATC3@w!Av)>Zhq9c*yBnQGbtS&eN^){ck?=0F5>Zy|9ff=Bny>2mQ*k2)Lp
zu==)p7tFRBmi!=?`ZO$A0pVk6ZenVC!qnWF`7uNO!~tn<UQ0oewl_BuVSLrj=^n9C
z1K?&g7g)p;{&lY{C?e9ED$~*ktgS;QVRF*&+GWVW2#Kl&SVk97logqqj^tth=J$TN
zngP}`z;GvQ;15EW`7z-whj#{t9~}suVZ&}gVlrStM@go(Zz1aK#HpLx%w)kKhjJ%!
zIr`ql@I5Edg4hT5zKx}`z8u;^NL5bG?6Lc@nFf1&e56k%@jss59__!pjozF6#(1J~
z37k4)!aH~R&3Bw=4c@6{m9NIr`J&;%U!3lWPT<Og<2><?t)`F4!T}Csd|XOOcud0o
z**lhCs-6JFnkK|#<eFAg=p`Jj7A3^#iPl4*h0yCzLZ)6!TinqEb?8yiqjB-Ach%!s
zOa}@b38J^4{3T6Y-3=Z5RWZHKD%2a|Ob5qJ<BIEIYSf>vjQ0m-&L`**hE_9Tuo*|8
zH952|gpS9@!ggyCne6(DQ{S(Kq#`d#om!3lC_^`eG(<dZd8i${m?ULQ;NT9Tf*@&-
z|F?LoQ1M)F$?1;zN7uDO*2+tl-h(O+Cv&D|K6cNlHp*9kt#vd#Zcw`(ZkCubDx7~L
zp?zvXXn*mTL2sIb-^EXptw;LnM@66cA*5u3=_Kh?#JGOM=F(`k#{{RW$;nU-!+(3n
zG7cPc_J<_C8{+^tL<V}=KTIZBHB<I54A4Z0QTA8n{5gt{#osSaC!@EW!NT%Yz3<-f
z73uI;@Gn?3GNsKK$y`!{t31bSXx)zp(*ur*BI@OntFy^(j;%dzGrITYUB;H|>x!4d
zlD@KGpT^j?E?qZtb_>mYl5c=GoG|m<Z~jd~aogPkJCZMa09MG9=2%78PYJ?}__-$Z
zM8e~b#k9Cpz2`=zF?-3zXxYLiAcAFA>NBEjQ6vR*yl$Vd)k}m4TFO6AlFjtjDHO<c
z_R)No=W#c~I!EUF{1AwDYFqk>nSzM>wX+Yj3pvfFZ0!83e}G_M6X<%Wx1;kjkPko>
ze&L=J=HuyVq>rt4Q>MTdN*#HhC87v$s<Cg2%?sj@He=TZ67W5j^b@x`A72l=fofQ-
z&K6C7``^H}z}EL;U5V$uV*4+!jm0F%37p1oH^~uY4k^FOYVP*6x_&OtOgQ)E66O~B
zW>+=w-R(m^cAr|CSqpulH!hp2=1bsE#RcIKYt0?;Acc~0k3oBk#}iM7FWn0h$oi6N
zaSsDLUKi<H?hUSI8J3iT=wG_Bt75(m%)Oydd=8$idfBn3>xb~q-?M*7`y(^4I>ceM
zwA%2$jBWkg$61C{p!=))W=a<PBmc|T2BX~N5|bxXTR*o?s*UZnPvD-5hX@=p3;Moa
zvCX;9tA|k2U>csqZq4d_!!dYml<7n8Bd63fpIDsv7ub5`c5$kBJ9RxHrRQosxoC|=
zb49v0?<=+)e-Ho&vjGB2J=s@m{}b589N-rLC<tDF%^&-Zv5okx*24;w-DhmWAvZn%
zm^%T#KKNP5&}lHS^7R$k!S?l$|1!1)Cm^QigY9h)BRl|->fA>WKKE=MiTTpIc{%25
z|I5F>*W1y7d;5%S;K2Hw1FgR|8AgZoHXk7_-2ME-?r_}J0^;rQZ_m8m{`oPLemH*P
zkcQLnuDmt#@9%GRtiio6Gl2<zcIH~|?d@$!v3LK>y=DI>z6bi-8gJ0Za9pP=sgLjg
zcDO{S;(lNYru`?dRqW9o3Rq9Za!bRHW3>6p)>CAs1i&XvGhc8whoqSaa(Oj2&enZR
z)1Ip0ReT_st+o`VL{9uQVmcxtx{<*)JNu8Z9Z}G|8YLK6E$H(TtZxlNex6*~U7(E~
zir&b+#7+AzV>_GR<wmOp01UK*l0pvU7GL>4C_>|ymS+=OW9tA-8*%?F^gDH^ZIKtf
z-?7yV)@%N~Xtsazul)Z|1p9s^-mCc^#D;`k;LVTuQwkb1<W=mNCT#1!7dEHJD^G?$
zATB)4`QpH;hNHB!mNx=p{|80*x{+INFD)b1rgNM@=Sy>}*O1zPXZa7Kt1zP~a$S6W
zD3*ZGf{9GG^0c{y3Oa!akI!3&uWe!7_D(ArV`0<hWAlYKxl<{`S!_i2oWYk^wqkUT
zp>lZNX2q}hu{Chnkx+&Vf7ZwqZt%hlKNQp;5Zh26J+Dh7t0F6*b!MjKkFQBuA{2x!
z3Edh_l9liyNX(xUyL@`CtnYAcwrdT4Y^Etm^uLS8KJAS89I7sz$*H$2_C3#Y3gGF*
zn~88^(~wTGHNrT^gFGM!POCaO!%D)ME#?>$xBxRH1#kgYGR|92EKYX~_SvHH`xXEt
z;K#+LaFjE(Zs@;3(1fMvoB(m769TWt__$mtFlaJ=mv_2I5#>s3OFsYV;swkyR*GjH
z1YPWWvUZ3$YtZa+t?+D<uiYf^e6>ccTq<RMSxGIg;G$9Y=$8QYnXZ?wzT4hm{JBjI
zIcRlV^k+`*sh8Eh>kO^G!Hvw^meAzPDL^Tt>5!cJO2NbvpKe)h0VRqsHyNoqUImb~
zPdbGJC9eA#@>=gCOLAi27gLt_u#{xsk$%c4QX**S?)_2KTNhM5Ygv5y)hhgzK|-40
zxp|ZrutU{X4g3wc&<8Wbf5fEgke39)srZ1uOaO`N849K|Bw6I8<69B%FZ;Q3AEe0*
zM;hV@-;e^h)746PhGg`UV8<LO$IoKgAfUkP1Ng<jSBAW;N*Vh%1(oN5tgrFnL1qq<
z9}dOXXDV%fH@ByPiU&KD%@&4(o}YanzzWR`2>mJ%r^|l5-?5#J=LWE)yA61q4(id*
zUp09CRh39w31)cEdsO03o;o_^2S_|i^(}Zyi~p9xM)hIUMGk<#3s=H}rbyzW&zAUH
zwKYN)syJY84Y>s7zX{7YzCER_IQ#o<iPZEYAeB)ZtPX%EBGi9t|HdMAG=)ZfzuT1r
zacw>YhbdLhr>y>b+IC2TL--E@B7S}7w|sbHjM|&*FNsh4##37_sDc!(8}V~J8<kh!
zNjGuOxt67U8={1KdHDD5w>RSgiajPDOJ;sxna%Q>XJL~%mC8?Xl_+0VY%y?BQQ^8<
z&>`_5awRV&wo#!mk+VPB8qZbMG&r~w?}75wVRMrN(G&<|n92cslCBu6ABL&k1$6vD
z{QPM~GJ-tmAVK{8t;u>u69drRVSp2G_oL#N=>`M}D3}(lMW<Fp&e-wkU1FH1xH|6_
zf>tfM*Q?lcz!FWF@8>`Y=b?O01y%u&aSOA*_*8ea$OHPvWRUyUXR&)VLK}lI{d|E}
z2+5=seL(llf!CpWpN)7U=y>ImU>_+Q?OOx|@zVH_x18PDM9XjQd!XsvXMRVgoFqfe
zM0m&)YZya~m9jF=R(obyy?P|9n0G#EVeX0;duLQtrP8P7byCpJ_vs_Uw|<lOFMV>k
z7syU-ZM!4}83#dY$A5{x8oX3{3D4%*4qK~^c=OhBBxYqvs8wY7psyZ*bZ=(`tZWkG
zJTYYOH<-qEydJjS-qT7C&t&;+8^gFzzGQ?%G0od2!p(&aNLF!T=`dHa|F<m!gpyF5
z<~@6aGanB2ro@o}v0{|3)*XPza!J6&Tfh^3!#vd2-Gwl4eiA>wa6&#B9{AHeo)leh
zlvr06xVIc_-jmW&NW5LBZ|g$l^Ag@47PK`3!?0ik_TSW?F@ww-K_>7A4Qd5FtTP!&
z3^E}b&ZNd(-bo9$c6#n%{bJkerI0rhZT2%g^g7koh2}j{<jcJdSI7C>o<r!9oll}&
zh(P?Ht(apva?q0vyfgKeB)A_8M-b~mMadE91J3BVaE>2{<7lN?f;&sg=Z{U+yNHlL
z`sGD=5P_~6I~K4-;T402CE$E6M<5Dh`Ne+lyjF1OBm1wjIO|F{-YJICX!{8&sGZ6c
zhx2iuz_^OIsyqQxRz$Ule>l~@VACgoiQxW$h`@q&uPQgLSlV}coZUweR!`B#JPre?
z9bV+oq*O;Uunhs~?44>%gY|kHbe@gz=<#GUIFdb~2WdfJr@Y9RtNf*2G2Pa2&tm_4
zkB!ncnuL4I&_j5Dl&&CW;REohW3C?50E=Un{)BNBq#Z$0{RPO-w*+q<hKnWo+P)MU
z$Yh6G0~y`R5G^gvMOb`Igikod*^>r4sdWjR49kB&GX4>qLC@*Kq0CKD33L|+a+oO^
z)FAA~GYZ=;RiUZART0n`lrxx+`Qw96WID2#4CSHvpDfPxDD@4d=bD(n7B&4+>kCcq
zxWVut-$o^&;n6!q9A>L28#^V`Qa$jP2Z{;r90hB-3fpfSaE=7zNCM7;RHxM|*P~>0
zT?O1DsbgVjlTRI^p;rY&u0KEr%if7S>>ezZX%^GSRr50)W+C=f-g!&`uD5<}1DDAk
z9umVsv_3$-T!9F<fEL3-3*gSx-Iw&4nJ`*NVkUnJ?o#44uBPxKGmL@=MRvgnVR(~5
zs4oHmT+BX+=I5uo5bOBj7_b^LG~Xm71LgOY2Kp6NQ2-BlEt33VO!B2?aw3BCu8;zb
z_V2Rf{z?v2J<XZGBAG-WND=-HBzTpk@+S_yo9I$GVs)4LmD@H~Y%D$H1mu*4u3%WH
ze=}0m*RDXC4#JwsIDUB!qrC!@uH)CzCcLgrd8JRMr}v8Jb_5}~K_qYNwPgYjO)m1(
z0rpGE+-?43AZ6P_Jo%v;(HP}O26-}p-YVGs(Qv5?hNOAFh>Ac`b9{SY2fDLi4EJ7U
zxRxnrV1&#3^V}-o`eh~1PiV@9J-;v`Kr9i?MS(x9ii45Zm(eN#&?8}J9Pp@=V+9tL
z0i=+%at^?`Oc{P2G!DQM;K9{J<My*K_!ERAS9eIpa{QqP{{+#dEGpk!lK424Kh2n!
z%`Ca)RXwsBnE!$Z)k6DA3*v~trSU8FjaS&?<jRLvI+8$5O*NXMW$5CdR*_oPT2S!3
zQ&@?$Sy?UZonJ<H0+B=;@eF~s!6NXm{9`Vn)Z8qEY}3=B<xB*YZhOx@B-JPMvwegw
z8vblPqtI6Dj56pA!#UeF?qVOpLkD<4fUlH>u6~eMk1F)FyVSCt7_rX%GXlY0s3g*n
z?;{Z%C>JwDCHk>!n@nfoFJD|5N7FK7)Q64O)w;w@;&)-WZfd&f`@v1MxHeNmA*6bE
zIKpN9oGTp=YD4U|+DN5|Jd|_K7mb>p=iMhZlO{csCW!4nrMVQr-q*W*ey1uNp&C;_
ze!sgh>)rmpQ?-(W;7zwR3aQ@O;`djmq0sA#x;-b2ugYnenUq~!$!b9wdA~U3!oQsR
zPyx|q=2~0qUxi3m(Jra7b?=@4^vKQy+0Gjl$?**^VP-fwB;uRA^KNlRE%rurcQ2@s
z&~(ftAvknpJ85tokydsYO$5BxUBY+We?PsnvE_T6)F421u6}kYptz!r)LT^2Xoi*u
z_ucPdD&;>LSW1OsvHAVS2R<#c2M%)u3Q)N^s$lmgVTXG|rZAi)nC8aqE|-{l%3;mT
zh0S)B!rTqrz@DsEmB86H;TJdWbdTKjSuHa#@8K1<?(}g=kGUQ(-lL(x38A})noD>x
zfUt!ju4Tw14a7qm;>Hai@it-)0HRRHIRpZi;~htKgRb)E;ecIvkWeH7$q4OhM6^8^
z9__Ac>$la%!VSwge~yMCah%vaXGvR5JqC>1;?mwM1Wp2YgQ4-*R{H)AH=7A3Nx_2L
zBCv2)sy7dS2%($<h@+Xzp$X+1ILEPIi%apW0B>`hzyXr6orE5%VSmNak5NfD9roG~
zfTN<i7a<M|Hn;WwAWFgNF}glCdi=<~SEhQ=t)J-OQ08DL>}JTgh)L7QV#G|=#B*VI
zUNHFfCJZ!dO>BaN{D)MZ<nS}X<~1yNx8e6FurDdxQlP24qnrbah+A~=coc}Ag_s=R
zS7)7lGK;)L;ldbU>sWRjJ?GTv@TDUXOh9ty2;_$?|7!-%TNeq`s`xIRrx$0jyK$Zk
z@WtTyk20o2heo-AdA6#GrgC^$(p*pv-oSOf<Aj-4c%X?~_`wDF{kJ5uisz@J^yc-z
z?f!?D*B-n5#D`g>Iv0gGf68_T-)?vzjL4bjsl{`i7N6=$qlop;#pgyeTF%<E^!QBD
z-m#UK&HNu#CBzmbJ`7254@;eL@|dSNzx~WK33&2r<I(v?QusegXaCHZJ$N#2@syeP
z^sA-EnOuU^pC`snyq`WxevEuF>+;l2;)(Ac@eHAA{QEo+iRZt5dmOY8@(lMpEnuY-
zcP93_l|C`{sKk7%d(Y&G*K*ce)d9;Zo%1=b=l{RH-oq{F|MA~_0RkeZU=B0~&Ri9m
z<;a1h=EzZ5xkEEEvo!6U=ES{o<{oJ7%FxWMp}BM5$lRfonQ38}xjFRye9rIt`<-*H
z^9RVKmlqd2@7Lpb-{?yt@efD5w5Ft9jXZ@$z!=iLgHrXE?fet$<mc);3L`sJZ-xd=
z`|`GXtC`jY6i4TcJz_}_v}Uec=_~m?k#&BmDNir&nT>ou*pQ)8rZQWqH{1Q(_|4JT
z-qzWkeU;jcsfObIj`jMg_~{AGb2S6g`hJry^6;^yGjx&pYEI@<sq|EH;CzPjl-%QI
zh5(4TG0zbB@EbBGClc_cFw)&>?tSYAl?3~lz_GbO*$+qO*&OJnKTQa~fnP4xM9tH=
z4(R`CY!`SP7FdXO@|O*^ZQ**PzBGManhO6WxhSr^DC4jwZMG;Mw5V{f@XPeW_3`&e
z>uK)B52xDH!cB`KIq4vWv#^7YgdlqI1`NZ>o8gU&Nb|<p`o{>MWAqznoUtMqD6#yj
zv9-$oY@==Bk_vG+_)HXCa?)OMc3A2wf$0qmxV0_0S1kn-rjLu93-$k4uH|)We#tlg
zi++{*z@PK9+6FB%L=OegCjX96j2Xa*p?ifl$C=2~c(F>0<s72OdttdO0~}OBd;qXY
zy({+vB{wFH6H30YaQ`c}lUX7-eaiCdB>nBL3%L2}<+qr3kE({!DH=CSl?1#%f}sNT
z1c-GcgqHx4B&>Ejto8=2_UEq-w5<-!uhI@yM@84hwbv#b)+U42rt{ZkwO8M?LEg--
zEw-&0;XeOqm@;&-wI#FJjeY-q@SP$0<EQoyyR$HPs^aeAm#ZE9w@3f0u?6ce*sd{P
z!3>~4nx#w$npX!T5jkO3eYM&14=u<3<a~g?4f!d%M&>wl@9w_eAa@Bn9gU2p4nD?n
zZ9YbVQ_#u;sAbZ(@n>WAB5v_HG7rBbBAjE8`q5BF3mCBkiA<nz7J{CPA%7<?^NWkV
zRWk2VzdgCZ`I&4bW&S&swXuexf48(W?#La+$hUE=st`aO-`8A%H(d)h-Cl3Hf7ras
z-1HRt<E8WGzRn-I!yn(^O%2@VEPMH&MB8AoEt1Yw=(Vi|zDuiSKVn~R#hJq{oDtBZ
zF(O!DTbc><H8>}Pk^Oo*_rrERH-n`^*eW`n!c=5l{Pifw-iidfGre=i7}{Ex%+D>1
z!{1G}7>9`ev#5Wg67T%-t#k3#b@Nj#ZNc{~@qH&~jUV5M2+>3JPdNtBiH~uj{0o5(
z2pozJ9+}I4RIg?f!-MV!FuRCHA4;hGW&?<OU8pa|-PXDh%j5UGqu2$K+?Q`&tE1fJ
z@!UU$c34HQ2mARzf-x`~9v&Yr9$TlFxL9p1D^Dnt$1*sWGbJX%zvxMEN$G#w2hUmV
z16eiR+E;b|)!6=bV%u@LtGD4_`2m*3n%IuNgTI@!H8z%?oBu#xSp4|OS$^oErd8m#
zwbdWrzge&w7>EMw+dI2^`v*(_!Xa!(`71@(7epjMjWvDIVol%95+G%9*yGua*Qpi|
z818I`5-IZE(XF#xUrou|4E*(OcMD*sKu`<Sx*gU(oCi+ez;BIxIg)E)!?UN$^gA+Q
z^u(}nZ~X%O5%*E3Mywne$~D1u?)&WaFPE}-EoBO=1+q9Sim+KcP$ZXeUx9g54p~_c
zFNN!z+eE9KN%tH=LhFEovFnkMCOlXopZBK}VVK>TD~>=gmr)Eq<oWn~Mp?WLta5t$
zx`C`Y7(&_UQWQw#Zu6?{U_CaLxrn^fc&hbfU)uUd<wxZ|$G$&?2)#@(Dlltp`u(lb
zfGAdaosHw><TDYhxhP~CqOX_9ecbeL;~&PyV96(Jc}>w<bGm&zRQ6Cs`12&%f4XCA
zO)Q%Mmr>e>FutzZWbR-h2+U`RA};`rQl6tma(z;NEw<4YxGme6BT=wGMJkD#bAZvq
zZ#{QE1{(0s2HffF94Ja(;Ew)+Sin(QHk<q!Mu6WlYrRMK<a#N>5MgBryK?z?13W3X
zQ>I5O^CT5<+0S=^)vZya^c>dOx(70JB90hc@xxOhEFFQzW62Pp3t=h}6B#3HwSjJz
zb_ykTi3FkzE~R@E!_FB$bmTEM6Zm}Dz(93VeLNnj$mo&4=KV8UQA{r4P^h9xynXO*
zcdSPAtitiPDyh0~n%#-~hix$b`G?ff`uP{VI~)n8W_!P^zMRXvUJy3EsbsDN7D_uB
z`fqm(ax{$)#(ibxkqD+k#W4a>zIcoeqsI_8md|3kvchAVF)^U?S6=-nhpKJ&TtO<5
zyJYMqHlp}Zb`+{`E~<#P+|RSHD+wv~jH0je`+{7$z+QU<C>mecE_&{h9!c1S7u+NA
zn;#)9({wM?a*&f4^O#L5yfC6XS8*i^8-G;uiXhBRIwUv9@^_T}!#m&(5gy-na6YSx
z4H2N$h|o)BmN7BLW-77rLJ)q((jGBORSE%uX$e`Gb(Ql0d9SL$=bryNv8{E5HyoWi
z`D?(h)xdkuPa=t~hJf<kBe6r1EW)efZsz-{AXSiSUl=2*cN(|W-hbfOaz%b;b;lgY
zp{TkuO^v+wzVZlN*<a`p*53FI5w=A$DML}>i?{dImG7M7ZP$xAS;?Ppko?6Wuy^@h
zd9|aHEw{Qzb(i$@L8^YMO2^2Nh_cV`dZf=vqXh)`FPBDG7NK#53CfhzLwq{YlnLNO
zLnoLQ3Kwhvoj{PmSI-zcE+kQd&XHref2v49-3ncwG==5bTu&VfzawWu?YK;%JOOVJ
zVagnjF0TBEdb!<hX>Y8j|A-`r_@vLpmIO+7KT8s&tyxyu=qpg@xmP*AN3(#fQd*|M
zO~#5<YyYgTJiM`oFN7`OWiCIuUVYG!WKBkXR4eWZy-)it{$9DGj=jd<edBt9=RnUH
zAuof`cV=gVx5Gpzbhu6nYgx*i1oNLi)8Z%oBPqHi`lg0cB;G<7CGGF9;To!>wC60V
zo=Or`v<8aSKk-}^1d_@DZ8WY_QofN*!0<-03BjGkF{a$~Mj*}cpV_S@`@|_l*fzp>
zKMoTrutAAv82}l_-isSk!$h_y4uW-*sQeE~LAe72MK6uR=AkDBM)9Nzh)eOZ)Nz6O
zRb7J&D{-4%!5YC;@#9Vv90Klgk;n}LrP3Woxs;c=OH?NhV^lc-F0UrV{|IVJj$BI9
z2meX}w9*fOK2s)Xj{C83*EVvoszT3q($=;ANrMNETZI8#>&3FH9ds|>#l7Nh(%+^f
zZ*HPPR70rTzvLa55<Y${mw2Ejrc+5$H}p~q$sjpf+<>TNCjz-uk<CUwv)y|}vQRB1
zWyZ3Aa}HdYB+|~U_$^HRN0jknDYB{$`cOH^n@vCPR`#?46>yc1XWgtm7HOD(xLOQC
zZpa+U@-)dQm${-MI}iqK0Io!D9}*)MnwRS-c6Q5Ui#1-7u1=<+*xv}iCXb(VeG+Qv
z)KE)Z_AaPEJ3V#6<b^bCz6yHD$CJ|=iYAN*VJWkGN}l>XzaP}cr5bn&zzrUGrNz7C
zis+3_8ic5*mau@W?$R>)^ObLpR{H0@2N0CSCXMjwC#B4LZWrkvlUCr&3={575eVT!
z5VZj=q7c(Fgk8;rn_x{6Hta=|jTnL*x8A0n>OQrcEPo_3Wb$gOvPpBt4ShZ($Q6DZ
zUmW9xQ~nv0@UbL7NE&Azt3Fi}3W3Y7>OovxsdkqK@b#95a)NQBG4d6Xx{7pF%;z%+
z6ERtfxKqwWYNl>_kNZjY&KR-i?({_ipwf|raivJ{=Wk=d*r)=ZUthd!kTb{p1y4h@
z7InGdigD@ZkQuj^`HpZjcH1jyads8esFhv#7<u8o*qsW^CC{dDzE6##oVDaX)fcC}
z3Q8RRr?2S%2vxQxaC2fmwJht;8)x*E=~Q1U=<X2Ua4eJyBqu*gIV4OTF+|<?$d<TD
zk!7k6;?xW7PVEN(6X58WceK+TSKq8Sl?8o$_k}N&+x(*-3@vj~z%U`x1;_wJ8-SgD
zpAQJ}5)#aXJ#EK~r3cigoE5K(IGG$GqT&onpEFfS<{WbNryf2^ZjX|M+`iZ#?{@h}
z_lZ7BhxiZOTE`S|C@mTlsKfOul%-$RyU&;LdVmpm=xp6NyKbKcmlb1-RkS1whd+}q
zu+1BPUljb&;CC!av~N)T$v<@`YLk!`zuClzy0Os;;HJN_zlAkL1+hPH?8lsTU$`TH
z4ii0w{<;ui*phD6WHMmFCw9wRsnuF9fzWssVg)GZ3Yu(&@^qYAvnY=qIpZ)ZVp2t6
zK2zyS2E8>bPZ4gsaYSdTkY9))TM3;wVc*UJsX6kXY&}9@It=nMBt<|?kFF@Rb^jKV
zbdpSjI@$f?KoshUiwY1UAn9)opI2a<FLpQKf7>|;NAr((m!Qq~+z^74L-p6D?)sW+
zrHh#DPBSG2!VY^c;_t4{#}Q&(PMx5fs(+UDbkExN)RX)__>1FN<4d%RyM^A5TxVW&
zqIg&<ZNtr?qkDIe^rOY2JwH-F?wLwgN5JhrNT*CXr8{>XhwUA%t^-lx9U0$7_y5eC
zQ5$%TNyD+<69<{`fNbf3m#a_R1ewggKWpp7y1&J%r2^4r1fSA_B0LY9g5X}tO#D3%
z+=U(tDZA$Mo}ZH*{3|0+FN9ayLbk}04MVlRSjyu_y5dMQ`=TOwv{%TJd`BJ>as!P8
z!{Ho;CgN6CKXyVLKpeRgZ80nMS8STO#Gw3fz{@iXRxf|nQV)64-(fidt*Tc0Tki&c
zfW2j;0H0#dqJGyu0R?ZoqPp#d1p@C!UsU_z#Z}`Sn|}KAA|!hOo+Nc^@mx^2FXo)5
z&^g=<fkF`fJ>~FwnCN<*m}t3NqD@*aRy`AQMACs#cPK8?+@E#n{sc$(J?n3^JUNds
zF}^|q$%hi|#jK{E!7fB&9b;YFwYE2+Gp#s4N3fSqai}MoD}z0rcnUpHV{+v8A{el^
zP;tp4Pogcmu~!~&P|9uqzau7H{TtwF@W&Q0k^%iZk2p@=T*TVteT?*3y3qITq6PJ)
z+ozlEmQg?1&y7Vz9lH^zSbfZc0_@cs5};eDos5n!jP5avJwd!N;gYm-pRH~yR11hD
zY$TOZlUTmez6%_(QdoqbbCjWKpQDYaTr!_}N>8#fIa8R^C`C8jro)@Zt{ka`2P>(k
zDqE+jJWM^6^*1`EF_o&hovJOEMo>>XWBqS;%sRCJ4|a4#8g8e}U{fQ<K+e6QEGfb$
z4Wv&OUEYfP!nlyipYEg%#n*$C1rmQ{raym)W}Tg!w$nWundi0|c)l&dEm`((GW=K#
za8!=v_Y8tor2JQ7`!F*)D^n-@Ms-ibh*IhlHp`eT>lJV>4WW=x;cwNLmD8tSH}$x;
z?2#sr;W_naKN(FbO|Q&CvyLti2H6yZM9c(F!&&hMbU5Uk@^5|M0oozhC@EGjw?jR*
z(>k~NmarP=mNQ26HlA&G0USVu(**Oz)bmDI8r!VA>Hn9;c1bR8yaGbImG|vo{$gYP
zaXNUA47gKahlvH7ji8;zf<F%nwx$a9whMN)3w|TW?l`D=5}BikjGF#yu|;Q-rRf)Z
zDG%4Z3bEOR0y|7NV8P8!1Ow<I35_DDY;GyPBH8RB`KBVp>7rvhMXVW%vc|c$6@{n#
zo~)<DYre-UJAo2QIUE^WLEAa${Fyxui&+fQ*ey3l|NfE1Too!DjW0ptN-T{_{%UNS
zN^GV}>~=~Vgi47TrA{`b&VHpX*`=;crEb%u?tXb!WAm+q%G_<rtU1cA$``QC`O>tq
zKpRkyP`NZv9=1~!>Q)|N%nFc!?=Z^v80ASq6)74OX*Lz<JLQ2ZC|}Rxl28D*Q^byY
z^1m#$;v7Xt5CE~x@aeN>9ks&m-`CR!TCFQ?$bQPQ1H?UK`@K<Ioy2x&CBM^#`K;UT
zS#S2UE}Ifw^0Otuf<Y2nVm5@2@O;AU`8$p0Q`yhoS3XBnKA&xRo}p3o$)<{w4axR9
zE>dYwuu@bQTlt^xm||rmYNWF95C}=PdO(7?&~j2sFTJw4^~U86%%8%Z^M8cLcurUI
z+E(-XSECPuv}nMgC-D}3WkW(W5~pjVY-<9@5ZvvWX#~ix5?I=)!5@C9Y*H1LU9~w{
zv9?mR(ezhi`-G#2Bb4<Q=7?ghd7zJjzUDxhYtPTrn(fy9xyVn|V0%+pbH)FaRn9A$
zr+Kt~@TkU1R(H(lc7-&lN|92rSyLsp^3r{#?(S`7CGrAfYZ`+|Vo4h_q+SA{hEFYK
z>VtOcNzJuF0P-vYe7WgWbk5)K*r&dl5smWetniq^%ai_fe=b7GvR-=p8y=IffFQ+d
zBN~oT3ORB%PA&A*hcy2Uk5%~}*8)l|zip_sZLWLz%6|7%YIAdoNu5hvolH?(_Do}^
zZOiG~FT74SHHIDI@8I&r!60ag;h(0z-LdAT!LLAse?$FD>-=tO4+m&~Qr2qQwiH*v
zM|>$0+IXPU(!JZpxLv7xx_0_^B}>)sGXn(5QQptIKG=QzTCjO~7aWNMK?A6lVyUoP
zDyoIb^^A(1rDFD|Ja+B80quu!neF_~+OaL|qO<L|fp&?#cFEcH<5YN}f4O{1hvIC<
zvAw_HF+K+P_0y_-5THq~z`21>&%V*vd!vc!)E4g4)9f@*>@>9NGz#c6p6xup*J&c$
zWv1Ar<qp1gy54y`@xA~oX&i7R-rqw~oiw|h0}QnY4Q!gv+*(T9-AnGabbHKpdusN0
z+4bBD=(!)?^B}h;prt41Sr2KpCuFZD)UG#Nv8Qnt@};>qbr&K{Zj*Kbsm@kq+xAJh
z^-0F|NmTY#`1R%I_7$}B70&jN?fQ${`-=nmD|7pw?KKw0y~uX~@q5;K(BQ1TUW<D1
zjeomi#|lA1G2jkO=0K<2KzG1EZ|*>U%fP_wz|h_RO?YrrbC9L6?Op|j$bcb<ZTi{Z
z?B3vq)xo)z!H>d2pHM@ebBC5&h872gzRV6W0*1bz`lOt?Ky=vFD$Ia9yxlUq8$Wz7
zJG{s8!3+*F-D%)J8pxE!VNYZ6B~%`bM`VPTbA*4NhFzlx&C!^%BOu(haO|b9@f_tF
zaNucETkfd*+*^M-aBP1RFEXaAHKt-eb}Dd8EpJStbxd<^OnZNfATq9UM{USr+#vAZ
z?pW*i`MGhk{c(fmz-`at_6z`t0j%=g*|fg1n|tT5|Bfg!;iNU;Y(K#oZ(iF6ZW3Pe
zf|=le@pGPQ@R!E76ztzWINGg@2eZOs0dwzz_TQ64CPTF*!|f*{11F>NCSzOw65F6W
zHVZP0gA7i!pGpgyO3$0hY@N!Uo66mv$`_d?YfYy=hutS?-XsAElxgtj^z-M_mHX3A
zIcKU&XKF-d>P%-E>}P7`W}38STJmPvwPx!%XPb}C_L$DTUYqH<lYW(vad-qHEB<!e
zbk5lS{bb(U_*<>nxw-lMITo>9)SCZfKfe?>znnL}(mKC7H~)QZF5Va8L4a@Af7lHC
zu$A{=r}e|$+=qj?4=mOiT?-@Bz&r@_U?QF4E(V!ThX>J7Tnk+Dblw9xy909pE4m=4
zy}<KmLF8cJa1|X_wIDgPAi%Y7Lwpf-0P|K4`bQ8ZF3zLPAjm$SJH=9(=6}>^`=~kp
zaliQ(hJpFY2h+=UI3v`po&U+W?bG@BPi6<7rcG;uXu#2Okbqz9d(O``ZGXdKe~E37
z;m93!rYlR-G3pfa$&p2D=a=ptEO`#SCFHv|>H!$y{U+frer;a@=D!3Te2J+&_6-L^
z;sG2U2+aCIYFmz<Uw(A3>~QR}V04B@$_;C-ubKH@v)jJr9{cjJ_{Ehkki;cOAO)T=
z|CQAptIKASeX~+|SK<<xMNA=lJ1YwJU~2K-hSxs*ljZJD0R&k6OKl>|$mD%p&}y&f
zAuU{LQzk~m6U>W)90AyB%2zHCrtl>ArB)st9H7X6bKn5^q_3I;wxzpZjaZBW8qj0F
zty(8c9WwYC5=I{(?I+&5<*)8MS~c<nAF)^ysD>X5AykrJ=NT(P=<i?(WBvdlxDH$3
zf&}BC^IXs*Gd8<BU<F68RqLT|^WRO)YB6MZ{RO-e4gS+>Ww-5Tode?tZB-xy3>!x5
zbFqy*{mL=SKn{bsf-$_u0f7)UFmC<Pqjme|>noyw0~Mi3dZ&s0@zD7#KUJ{dA?S?F
zN5z8Qxk0ewc<_-V*zwmq;9&5xEjS?wCYiM2Mq&Qq;)W_x7#t+b8L>aPIOu%Pbo~5=
zl_?u8`<+QJ2B2>URR6Is7iZlfn165YiPwFZ{lkX2W%vDT2KK}2J-89TcC!FROiYAw
zvtgCMw}`*)aY6YLF<*yxIJjGs8JmP*gaw76HH`2u->^G?X#zhaoi>8a`M-$*=EI2A
zL`;Yn&|LLt$Lg!zm%s?FZdLB>A?7D#0wj~0m8FV153tGN93%<*;nk4r7uyG143`3!
z5eajX@%!_SrE6bNHA4s;8md<3+aMWyg0X3sguzo7YX!^?T<pQ2!5m=4u;~Ap*hXtQ
z7=R2MG{f|=WTNvNf*qJ3(ADT-P(f&kUbIXJkouBEY*So)w1Wd75Pold90bGmAh7p=
zfBpZdv3+!$aBF^^U=(_OUN+3=$Ss0Lu)m7R`MD!nF}$~qoQRn_zcGEljNw3F)ht4A
zvWl=p!Ek3XZ#XzfDy7K6;!I5RI4T!Rd7>g(awdIxBg#-tGK~0GC_*mT=yE8~X?!11
z(vwp=Fsnp;n}~%GI7y7KXXVPdhN<%Q{k0WiWha_SLe4>EF}goH_miXaF|qzvG0Bio
z587mn`Piw8N=Z-@6ag(qcTObX5-`m3eIz|oYDf?hl^bVhaHhj8uMP3FYqakYj(PK-
zk(YRgTS5wIudXipSYkNl@yaaU0yqo?<HzDj5rh%#b7$Y?n>N?2&7|~}v@5xW*JDCy
z6}u-SL{jyynY+?5U|m^yZR-nW$Ygz9-9Vn2t@SUaXF(SaI)6(>bLHt>s@3Uvt2NT~
zRrOjdPnug>KM?E;%QWv}1N!&g75ADvrA7_92J2C%{A}(R7@u~gZrX*n*#C6L9HnmF
z59{y>y+1jNyPQYWS0(h4gslcffJODMuCEHu;Vo4R6TpfaJ=_p+-DJB2B1QmppOY;@
zVtUZmRL+6&L@;&uGLMO28BK~OI2f-^b~e^dMcyinsHTQKYcz1oVE$`G01!PnRx6fr
zXElx#|0=!uhs<|bs7JcyoQ4J^%T!zuO)~T&aD-mBU6U&^u!~|pvn-LyF&%GgRb!cj
z`*UkM%MjG?GYXb3;bl;fZdP>2{P~?~64nz-Y!AO#>h|q<!?dxkhAe)aOH;OuHlw<P
zPqDKKB_?xu#t2utSB<)s0X$=7wlthhr!;}MjF2MfhFR`hEv6YuT)UpoS}PYAubveh
ztXB+XH5uAQjT;kju5f-=3#D@qSK_*jGT-9uzfuI3VO5m>^+lY2@0-=wU$~zgFN2(N
z@kc#eO=vw>x#{%gFnZw~-{|O^p9Con6h}G!t%RX7v+#_>2eITEJ3SS<jq&WELE#Hw
z+)uYDgxfE=mVS+%SLICL`a@=Xd(pZi8SODoqxDut?`3s&hHVgY$CB*CkG^tMAOrQ$
z;B@vxEd#RQ*-uq=K4TTmqnYywT1TRr9JT^sfS#p#bWqXqRS)Ao%KM+itgJkBA)L$Y
zRDOSUJB6uSpYIPPEDAhX$JsH&DT%&9zI#HaWfxCE3r-Z-yTtq6j(+FIX}wEGo9=2{
zOr*{FI>#}qtxc>qbJ>SAj<l?ZofQuiV)vJ__gtO4SwQg&+<XA$G)vE9i*0x^w1`t)
zeG=`d?8b!83F{7kks4=4o)DrD;a)!!pDES<U^@F7t)J&ndk`Tnb>YlAF?<)tXK}Fa
z`Lh<Ej^P_&sOQ}8W^#8=_3E1ZCIy7iyDdZCJk?n)^<R35mC&?o(^*|%>sTb(NAp&N
z*!7&g8;3lN10OS>N7#lLUi>0^y&*U6jI?ay)ZvTbFJC`2Cd82Fqa>NycUUT%lM%MW
zKGA;6Ps!BkEJ$nAhx>W&bwPiP_TOynC-p2N6l1W+np0Q-k$rYa-*)y#D6pzq#aiNc
zY{A(iDL}}pO8dl<c*QDmFYK<ohg5RL2Sdk0md}cBC7y-}2@!PBdPE|{`lG-(qyDb}
zpG66V@Ph8-SB@upJi-h`?tdStc>%P?=_oT<!K@RN4Z?jA9PMv=B8_0gO$)RqL}E%7
zQn@`l5Yu~IrhG-8tVfawJsl#<&kyjzwupe>I8lxzMWES%pp`A>{?HI?R{0@$@;q-5
zbl1wewH^x1+eN&Z2Ctj39ZA6!x@U+C*|mXT^z=w_KbV6sY%~^<R%owx5B4%!=@?+@
zPF%N(B#+{8s0Uc8TapY*wy;rbVifwq06nZDyns4-xbyJRR`d(kJLtmOw*>ad;8&mv
zcN2Wio)-kE{$x`+J_!MuvbQ1?uy-F)EZv&;b&dQT$8U*TP7wR#qKmNRE+KL%<ak;f
zQGI0bY{b=&`vX47pzxhbi7!TunO$GqqmM(OtN;+M7MkL<{+z_A7|^s{DmWg447~#r
z1+3VisFT%UYHy^s#u`e+qXGX05@-((he%P+OGi9&Jr|dzh3W~vDZQpR#Cp%K7`{5D
zJ8!3+frMm=qrBospHQ@`v`C&hDreQHwiGtb`;h96gia^!_Vgs_Kb#q!`Zta$gdAn@
zMN3*3H<n0(5gp@Rn^yVu%%BwU;s|A8$8v8od%ln!2Z!fvFH<_y+Sl>PrC9{14he@X
zx2F~eXO2#q(7~reMG{X^aFgp2n=;s(X6hM89ljPC4q{_DaYI!H7P$Ny;u7D4ifvM6
ztT5Di$N?qx>Hu?*|LnJiJ<AYKY%&SI0cw$ZQ2oef;G^gVgKLmqkmqN<eUv!7-D{|w
zbnF0q-Gaj^&FE20TG{!}2F{~><bivS>jyp``-Sbc3aZKIIlrXrgY9*csmXdjuyhK-
zKX_|`lJi$%oA|5$Ui!<tKLcO1eYgf5PSmjQ<?@*i97DJ2!U|D?%Lb>LOfu%eSQ^_~
z#*lwZgD51yQ-fbw8r!kI-LZ3~D;INi#>>rKRbCxjxzfDzj`ClP?aa<Z!_ccLpTTbq
zyF2fx4zEO($Z=YG9pGNGx|(Dyk~Vuks!Axew#;<R&31Qsa;UDZesJx7S!_+edogqV
zM`PQ(JI8ob-}1%uhu@6<Y>RzTn|;@hK#z9a(4TKgbqS2n!+Ugg^M($AAx7lsy#;h)
zL+25*f5T(g;f8M2q4h`pdmsM~jqP^lASqG@O+|6lL)Zr>%O{~$M77t|L*7Hb@^)Lk
z5SWd$Ftd%q!>!API!&XgLmMThTfbTaH;q3r`(0t%x?=UZY2xM3@8|xl--yhn$v0-3
z)j6%Jt~$-rV`jgvJF5*^Tp1NDr^hw@wb%|f&;J?v)3&?+gY<t}Y!4XG<}Hf?!&}{_
z57z%$Y>$|4_uC%)%pBHSs-U}{TD#2r^}lIs%gtL?uKv~79{jeXw~*rlSEo<xZt_zC
zzu$YYJOAt8PcwJhj|lF)g~PA6dXKdIOnklf=>&6oC^6_)8F%1{$?KiT*KM26hxb=~
zf_LY+b+&q#`#<A8?0wRCy)!l(w9y>A|HDIP@5_%p7oF47e}-Nk{268*>@t}Emh>Nu
zEzFk$Z-9E^WFfR`a~mWyHiSnlgx4yB-!}xC86wmW@;@3|wNNRmP#ND)*}odwP{qm6
zV_TtkY?!ian5QQw^bCMN`)f3WX-<Y|Z-o)C;b+vs^{m1Te8UYh!&wvC^ONCbTj3U1
zX2eCc2urJoE4~p{nGrS(5q6Ui4wDfGB0RJRgt3Tp@r`uNjC5;=bf1i5ZEQWUQC@0M
z_pPEH_(u6=M)@^F1x!W-ZAFo=(V=S5;a1U+tPj}C=-BdT?ox4<#@00bW{O%&npI4?
zZ;Ul8Cc7afcQPjbUyW^EH0!cq?0>prl`M_zWNg(|ECm}^qZU_d6<6mQ_d?eDWSLB2
zbX=QkTo4P<8b)~#QLlaDdox*sruc!$_@OP95g}pJDkk)lWUE@jWHc)ItorO^!u(bO
z9s6ie?a`bX9F&@{ocX_X$1-EzHHg2*CiYGWRTjcu)G!mBsEG%YiNJOuSTKoAJ&Ebc
zIu$u@HvCU`?7LdxBWw(hVBCii5<)rNbBm==OcD@Gkx);OvQCjvPvOlH|LvPnxn++`
z0&3Tj7dlf^tW!@tOl5_~N-|ThH7RLZ_7F)YkN2@`Wc^BCr;Ozkq}pL#l~j%GGz-DU
ztnL`HFI8YMHC0xX9-(4|i|02=aICw0HcKUP5w5iPI1uM-GzCBYmEHdaoYMu#@*hZU
zvMU=s_RY%hYs?6+j=OS8+)D6fq?3>(QKqC2inI{&CZZ%<WC*v?{E6`fO4qv^j<aHl
z7$V?hktuO1)hz}J@nl`A&%}_J&>qTh4*)#>$u^}tEg=cuVHqf>^4dhuz8caJfSWu7
z;bX2s=<KFb8UH1=6*paZG9Bx#+LEDA8sOo1opoZfA;a0otd$;WZXG5>gA+bWy~PlF
zah4b|XfDe(3<rl2!w^NFktx1eId&%t0GWi_%*u{i`fJw6x)cK*OOGRhcKb}B3{x8%
zOaKRaNd<g7uOlr0O%T{*g5Bp~E~_cVBDNCS8L*;Uvk5+13qWF;Wo(CYT7;o+MSc`G
zic;XmfX}kQn<xNlS?!G`V({>4IgkJYW(BacL{Ear`J4>6HlB=dV!b%r7FP%*Goc+g
zxD5`DzXez3fJ1R0gYDFr+O${(nPUfz0^qI|dHiTtFb&}Gg#XUUV!3Lg@$i@1a2T%0
z7RNql1WiEa3Qd;<?Ua$^6NEX$M0W~Vngkz4ZoxPocV(8u&LQiS+{W!<iIrUIX`$|<
za=A*DmbGN1ufj?LkQ2&*;^DvKil9`od1a}!4d6qE%f(iDQA$#`3h<5W2u88hCpa4s
z<U_Wu<14nB1`_;Ozip&%CDhN)7A?<?U_I0E1DOeykJ~(-@O%Cf>oXCZe#=1AYWj&P
z><QEj4h7(WjN)`ip_LF!zye@h_K(rv4I)4#{Pg)FtE}9ZfSey_2vLE;fuLv@1P6L?
zk(K%Zs-}UMCTO<?>qdM@$=3q=X?9D1V(tfGO~d(Xp(qW7sv@W-1HQIX6iOu<{D#|5
zkjGRD<EUkm{~@+dT{e23zf~`e6{-@SE}k?39?Rz-$@x~as%GLd9to#Gxgt+AOvfKi
zsmULb%W0w%<EYL-7G+K(in+}bad~*sq2eGi$i@1p`@^TP1gIk+62kgHj%Qu~{L?zj
zn^EkA1MyfCODlk)Y0rY`a33l#qVfF1>4tFIhA#E;u?yv9F0w0#s(6k<7mKRor%*Q1
zRRlSopF|<eP+GD9PftpW8?ZaYgokfGADV)n@dg2>)zLeYxyx1VUm<&y&$#8`fkZ&T
z=INnkc5MrwUcRY~;~B~grsG$v83KEy&W>tw^`O>UQEOu9HIb)VS;Y22OhZFq1ItJ}
zyqzD6W?kL_4K+gliLLY|=erPUgOkc%D!&*)Q2K?Qnrl=bEZQV@$mPcBTgDYOBxq|~
zr6-BvcRF3yx-K3M)O>|l^#fo(kX{(oi}f_YqRhAwdRc)TNrpPukVDZhT0c9VTPv$O
zCKaDDZSyi+)@jx<vlp#$lKly*d3AzZ>}63r%LZ<I2*Ti12y*o-8Slw~r`kBsNz@Y*
zBgdA|d0Yr&q{u8jPyjgsPkjKm-^%MII>d<x$~gP<)>3vn0ji@2T8LvAm2LGao6J@r
zguR?F0@Pcc9Y#qjdfR<e;kk4`4@+aqh;FSfY_&EJwJml%_8z`U)>r<V*FxiS>En8_
zE-_;5U7#nr&_qntluxxwQ+{<k+9Dg_^b(39Ptl6_li1rTYjEy8e!2aXE&YSNJ#`B`
zw;!U|2vq9{p;E+Q<#flIRa--AUhvD~F9Wg(MzIdLoHnZ^Lw=pqnEq$~0$a**|0~d2
z-&;?a_>7!@nli-(+U(%`-rzl%!CKux%dF(yT&y^1Xr*Q7o8r*R3#cRf!#_2LH(1@V
z0Os&k?(j~_@Ls@hGNQjwSGQ|&n9cs*@E9w+)k;In(YW?$Xps>ftr1@P5&pmtY~F~_
z(cz3z87nl*8$BwaH7aF4Dib&=n>Q-oI;uD~dL$1(kl~gj*xM`rCAMl>0FGL9usF&h
zwr8})_3Zzj#MVOO-9@c;miF(i936F}dAWzeSRbkEvatQFcaCfC{zGhC^CsL{C)@+?
z3T?Z3ioEyIdVik<Yy;o>=DlYD+lG4}PXK;)?3z_4vSRA0vy6(TAMyYRw;$ov!(;@A
zOx-AjWA=)J*C38Wm=_7=<yL1OdG$#E6UpvhW(HD737o!Ns&7TgOF9d^VnY^vR;JPq
zB+NM#W_5a#n{s}`^Gzt{)Ab+!dhaKXU?Y_v2~fl~+zHQ0D8XR_$|}%(#u#c#cUM_~
zG<}BPZ|SyB$x^pL|J)AqqmXbon1*h_EqRb<(q#GE{N?UxHmbc<Wlq}6D_xrzg^8-&
z##c}rZ1Hs0z6gv5m{0bDZ>_Og0hAqixcxn{ck>5F3jE?>$}gk2?;0f-V*UGD&D|W7
z<MKJxMpMl+fMw{2@q1BS0bO02SA68m*l)ZWZ!bMG%^h2>xJlXXXy+l!*m{;W5f?Qu
zHEV7klRQB_xKbV(-P)E3_*FuAOQGk}uLkB+Yt!JrD&XLiN;U@BkpX+)U))Pj@xm|K
zP}BWPIw|e#N86SRx~GGHk2WOx*YY|4Jgq)f2l##K{@wV|idHmq%Ms#WA53qTI*c$l
zR*GtCvsM7D;@~g<?vR5sc>I*mQhMBN*2AI<quGUVw|w;*?&Y@RL0h=kf5<&(MRRkJ
z3tVz{;Dx4AyTE{x(ZGyZTiYjFz=a4u?YPK~>%LV9UH!_Av3M0VMDDf7bCXwf8X(Wv
z7bJ(xL+B8-y*4&n)5IKbR47ZE0`&qKxMQKit!?7AYp$^?OY=?@(_D{&s$leXexO*z
zt^L?UE_t5$tO;Kr{H==Ruqw$d-RDGf8Z_CEai6Cf_WHSkO<r7Y$~U+RBlNHFqdEnf
z=Pz-z;Vi1;+;q_^xr43wgN^ne|L+nnoN97#NQ>gcV;A*6#TIr%H8nUs55RMe9fa2T
zaDdkiT2D<DReK2`X>eA|(uFi-iT<Um1J_Tio7A7-r&yoX?1C|nwp6Z(l39~V@(~RM
z<)$po=u#vA$B-#<GkkuO(%;+PgPWlxUzhI-6+6+QCB^>043xVjM?OKjF{rP8%(|68
zdXNPBjRy8FyLI8_Tb`7U&Fo6t{Ge$?_PH3Ml%(oE>g@5yFJVYTWS#H{nLn(FZTdxK
zBhG&GWIFJ`94RKU<Bf(D766<CXvvRlir8+A&Tj3s-MZl2hJxLu*SjqrcH5Y{RI$B|
z;U(vmwL-1E-h#br&M+XSyuXs2rPFVpzhU}et3P<(h<{P{P_f-B;Puz;_yd+BdOr*g
zIFYi0aTh;-&tr~0ByaDO)WZG9@D4Tj<(wz>$-t-M?A`#ot))<)YcK{U1cM4haUuhQ
zQ&Q6&r)OklW#{DP<rk3uUmDw*m$gqVp~zGOGRPC^W8>%qg<1;OKy93Y-}Lc3;I!$5
z1#6G`@T8!;f?myk$PV=Cob~j2XTy)|webr^vY!8P+J3=#+VVhWQJp@P^#9q|{tb_z
ziD6EU%MEFeGhpEpKG;k-`kT;`r)}tZSXG=<x;`Yp-WOY1oEVAMZAZ8iQq!ZnYO#bT
zW8kpUF12{{ksfNE4o)C?MERyYM8&6}E7$ndf5T(_kH;gx-ZH-tAIm_w0VsAuJmd1v
zsXQN{y++TQd09G^X-cHCRP*i^j!kC2C3R(W)Pn1H37RObunz6TgY~f^KMS;dItNC;
z9p^wWRPAhxR7368vGPk#9R@YX(4*QauFc-(hoA@)<=MMSRBHNB<OFo$$&I``9qF#s
zHJ5R_Z(&vxs46bw15@3%t=<lg`;iEh*RBb%;D2uJs$`3b{<_)nS7ZC}Z2>dkLnvr+
z!5&L~k=WIuow*PujO#131xdF%!gT^O54DTo&>y)1KTt6Oe6{V@O-Wj7&=i9UoT*d^
zCn*a8Hg?c4(w4Vwc`**(Eqnb6+g2#bNkJo(OVg1dGsqvza^BtS6i-D$wC|6_@v{xE
z$3Y{$UWeaZwg4rnG*VLsfDvi$bHvH5c5`K>+O7X;Y%{~F3B;7iLO7p?6e&W?TlnT@
zfc51i{v-}}Bw8wvD2|AXuFm2BIoUBH#Vjf3A_|)s56eYO4?-eD`Oj~{g&ZG8R1_7b
zyD?WP%Da8Y$8CEf5&WZU*(=X#C;k%K?3gaF_VPMW#(y^L8>N%~_G)#n%+uAHfm7dB
zU(yT|)@sME++O>uv0bbCcVkPBWPK;}-_sEEFVrJNhy+r0yQSFBbhjfI73(S)4=a2R
z@I!tjs^ue0kwJZ<OrbUhVF?McPKI<(%N`>q68DnVC;&m4`5Lj=;>KoB87M+_oYkIW
zB)8<DXFvNDp9qeIx!v#<<p0G(flGVLD#jcqw=aYh))f9BH>J4Z4uh5+{Cao$_=0Ai
zgy-^tASA_N^P33r>Fp|5<V7&J>B|-6<L7brTd5Zora%;_b^iUH!?)`dnSW;QXUL?v
z2ld_?;-^}?j5_Jl;9V_j%3MfqbivZEKpx3hOj;nAQXzIcTBr^YZTIT<OfPYPQ7Op5
z2*J^7W8sM)r<*vafWJ2x({B|n6Y|8*U9z2j&i#V#Sr6>xs~NV-n9Bw~Vn=-1v@^#%
z<_1)MXr4XOYC;nBlmc>`Ebq5a&tEwh-4;8c4(kz?se38#%!^_ATcow(;GE-Chqf;3
zV8+Du*|&x~AtE@_W8*G76jH<n9p6DU5Ly~BGIOfOUJKVpeu?fI*YpN){4AA?pF}|#
z-c&e$T6~B@6|%nu<-kAR3D{wO)28<lY4=R>#$E#izk!p&g)Au`nVDzJ6g}ZMK@X7x
zIv00X8T5ozv<mkvHfg<AT%8*f8G!}3Pe)A}Y{OI8#&GyTw`62OS_Nz6IxVK`)TQJ3
zMJ&^{6B;fRMGo)?_l$ocY>H{sS=IIMb}Y~-cH~k+kvtyLa-ala^mP@m5d(kj0y&MO
z8TsYd2zx9jw4*)(+9#9hI=O+{YHVVSzV#O#6?V|xOc_ZeU^)|SIHly~wHL<2k_coe
zI%Ig`0(VZ#q_9ulTL*^@f#l4y9zys&vO7dcyP9+X+?Qb_d|gmgkp$f%i86cjeN94K
zK!l$>!>$W)O(PxmXh#MN=t^;WpiyY;B@(QqT_~rpn|V|S-L!*&7%IhZ3D2WJlX`FB
zYXYtv{lVG?Z?UAMU(=2Y^joO6l>*xh5Pnf73?-lhuWu(Ky^&tA^Xen_Of(mp|1Ve{
z8jftEmW$w3GrBCF1#A;9hEtMJAmgzU>I;!Kos3EPpY*`3(R8t-<eEq-<Ozlya*iJt
z((LRksmx|_=CJX*O6}nz;c=PP?vgPx)OK`txgKOz9f6K6yyTdK1c|01b?k6Ft_{F@
zSaB3=%h9apjk7d?frTx*-xpD$3U3)luy_8v>EW5`@tt}k08i!ia0PO^WkX`yKY_=;
zjf)1HAmL}-6M9E%iJ#a`J<Z;%E!WpK`p(2XwcK)4>OQn0-8nuqKD}Aj@K4J6d}i#-
zEQsqlr}c&O3|4sTPeX+tuUE#~*(HrXjen&GWvs@O&7Y>p!Kw#kJfOFC|Axmrj?R<e
ztw*Wh8jvfux)fA(P$UjZpuL$!q6K9+CWu75w?AN2vklX=4h1Us!=J_K)9!m{8~%{S
zlFY$x?_g9z(|Sl7Xs1<g9t{4CdDDuL+SUTW*^3f=4s_TEdmN;mYy#z^$)Ou=$@$Pe
zN5kvjYB(2wKbnMMb_*hA9bwmQ;Uj|z<-}&Z_1J~Xi&Wl8w)2c#)n$2Z6;FNyy`=&9
zN0G{%3^Y<iUunm%+xe+5g&F(o26Ejge7R;hj7kiVOtmO=Rrx11yyUu*M=HcZdL4ah
z4=l(jYh)H{=(0H^F810rJmQo-r?0Qx;OnCs)X<`);ql^2z~Tu!N;igf9$eCg<K*vN
zZ%>b0;Qw~XT!}y=85r+0uXS`lGh@jhM{#K?E%YHu#Xzae#xCYgvD3sl_kl&27?0;a
zWv>{gg6<xP6T+E+Ht@$kLMzn3d_BkO=yTVustVmu#=IC>z^mSq(M;}*uJ8KYlOy+7
ztr0I3)3o*#<ZQrm29y+^buUe*EfK_@^z7!(`n#(&=H3w(WUMGk8U}hxYSc-e_E$F~
z)Y&~$PFP2URH1~BvrL04CmIJ(+sVMS^C&JSPriExxbry2J0jk6(~2Vw_IKh+F%yTv
zRW>^SB`npVa|AnB|4;li#x1|s6u|g~Y=Zi%j1v<55pd!t?!~uyI{F$Je_A|}W_;b}
zZ2eufbk!Ht3j?c;J?XG{e4XgP?b}-`$A-EO5IXlyul@fMTN4F`nvj_n+UIBe^y$~C
zMyXM{Kxp$A6507r%ksHveBWn3^CMrfPnK$PSGc+fQ!^jPt|lSD7qd1P)e?Bo-N>|`
zw_v>oi(*%#gb+kW?!9HRcZ9Ir&q4afwDq}UM(}{vD@kSM>Ozcg+lfa8qyz3#5FP&S
zi}RrSVXBDbvk=scC=i?cO*RVQQBJTBXSmEmTi>&;0ap{|f$I;y0YY=*ha(*21oRNU
zM}vM#D4z?v{!Ksl9^2uEzmPZvAA+W&xFN(R1=4tPC<3BuNi)FR05J3$0gq9ho;WW8
z6h;6>aGU}R)eEEYXj$wc6@YjOU}%R$r8UP%Fet%uhGfy(elaLXKbWW$r;Bi;YjCEa
zT#{6KsX*b?Fv(B&kGg~--f%ZZz!;`5qic~uJWqZC?O|~9SBJUCYFZ}g?TBk)lik%F
z;p!J0SY_`kzI<2=H!lbiJu1Rx;dD9`C0xzCs5)^tyq#F{o9{5&g&wqK7lF|7I7-}^
z*94^PAA)tnE6L~{UMM{AapAU(qkh05zdtcD+e>+m5@dr0IVEZPp!tm!PTYRrpXSPE
zg}Y(u8u=ky-~tAfEy0_L{%1z%PRqjeu?eM}o`_V6*MWhFyOh;s-?*Vvfr2rVfTwzR
zDkcOQxr*TxlZBu05-5$1HDxoG)-}}^2QdK!TPo4J5L#!La47_un{GBIE;WG@pw!?u
zw{T^?T(92fZqtzk>k6FScL<?}F`i0_dWn*?H%5;nGHCFSUe5PD@n{|%90A&f2e=F2
zge@x<EQo9DqJYIA2n|0%;GCc^CB_+izu1#GF`VZv0tgZ(u1kt*>=G1^b7B$2hbScJ
zW~>tHKLhHJ<gRkMGeSwyrf5C9VWjd^a}O=s1hNdoOZDSVSEtG03{Z`9u|D!44^r$K
zoKVq-&gUC&n9HF>vRU|*$CoZ#4>xr5$+|j5G_av4ghykqc7&x;IMZ;Nic#UW?;+ol
zE$&ZguB#$kaSAxX*#}7gjs_QVfoHu>MkE0)booFEj|&0tKu7H9vsx=qmm~m#2M~1l
z1N?(TG|Ynmb2}lJ_dYXho!y&pOo)`-Du>v@z_iG65C*P%3NZZ%Gi9v{!k}`HoK&T3
cL){ZAWVmnn;hg`EjqSty<rv=(066`B0p|Y0*Z=?k

literal 0
HcmV?d00001

diff --git a/doc/source/user-profiling.rst b/doc/source/user-profiling.rst
new file mode 100644
index 000000000..e7c18dd5e
--- /dev/null
+++ b/doc/source/user-profiling.rst
@@ -0,0 +1,649 @@
+Profiling for Ray Users
+=======================
+
+This document is intended for users of Ray who want to know how to evaluate 
+the performance of their code while running on Ray. Profiling the 
+performance of your code can be very helpful to determine performance 
+bottlenecks or to find out where your code may not be parallelized properly. 
+If you are interested in pinpointing why your Ray application may not be 
+achieving the expected speedup, read on!
+
+
+A Basic Example to Profile
+--------------------------
+
+Let's try to profile a simple example, and compare how different ways to
+write a simple loop can affect performance.
+
+As a proxy for a computationally intensive and possibly slower function,
+let's define our remote function to just sleep for 0.5 seconds:
+
+.. code-block:: python
+
+  import ray
+  import time
+
+  # Our time-consuming remote function
+  @ray.remote
+  def func():
+      time.sleep(0.5)
+
+In our example setup, we wish to call our remote function ``func()`` five 
+times, and store the result of each call into a list. To compare the 
+performance of different ways of looping our calls to our remote function, 
+we can define each loop version as a separate function on the driver script.
+
+For the first version **ex1**, each iteration of the loop calls the remote 
+function, then calls ``ray.get`` in an attempt to store the current result 
+into the list, as follows:
+
+.. code-block:: python
+
+  # This loop is suboptimal in Ray, and should only be used for the sake of this example
+  def ex1():  
+      list1 = []
+      for i in range(5):
+          list1.append(ray.get(func.remote()))
+
+For the second version **ex2**, each iteration of the loop calls the remote 
+function, and stores it into the list **without** calling ``ray.get`` each time. 
+``ray.get`` is used after the loop has finished, in preparation for processing 
+``func()``'s results:
+
+.. code-block:: python
+
+  # This loop is more proper in Ray
+  def ex2():
+      list2 = []
+      for i in range(5):
+          list2.append(func.remote())
+      ray.get(list2)
+
+Finally, for an example that's not so parallelizable, let's create a 
+third version **ex3** where the driver has to call a local 
+function in between each call to the remote function ``func()``:
+
+.. code-block:: python
+
+  # A local function executed on the driver, not on Ray
+  def other_func():
+      time.sleep(0.3)
+
+  def ex3():
+      list3 = []
+      for i in range(5):
+          other_func()
+          list3.append(func.remote())
+      ray.get(list3)
+
+
+Timing Performance Using Python's Timestamps
+--------------------------------------------
+
+One way to sanity-check the performance of the three loops is simply to
+time how long it takes to complete each loop version. We can do this using 
+python's built-in ``time`` `module`_.
+
+.. _`module`: https://docs.python.org/3/library/time.html
+
+The ``time`` module contains a useful ``time()`` function that returns the 
+current timestamp in unix time whenever it's called. We can create a generic 
+function wrapper to call ``time()`` right before and right after each loop 
+function to print out how long each loop takes overall:
+
+.. code-block:: python
+
+  # This is a generic wrapper for any driver function you want to time
+  def time_this(f):
+      def timed_wrapper(*args, **kw):
+          start_time = time.time()
+          result = f(*args, **kw)
+          end_time = time.time()
+
+          # Time taken = end_time - start_time
+          print('| func:%r args:[%r, %r] took: %2.4f seconds |' % \
+                (f.__name__, args, kw, end_time - start_time))
+          return result
+      return timed_wrapper
+
+To always print out how long the loop takes to run each time the loop 
+function ``ex1()`` is called, we can evoke our ``time_this`` wrapper with 
+a function decorator. This can similarly be done to functions ``ex2()``
+and ``ex3()``:
+
+.. code-block:: python
+
+  @time_this  # Added decorator
+  def ex1():
+      list1 = []
+      for i in range(5):
+          list1.append(ray.get(func.remote()))
+
+  def main():
+      ray.init()
+      ex1()
+      ex2()
+      ex3()
+
+  if __name__ == "__main__":
+      main()
+
+Then, running the three timed loops should yield output similar to this:
+
+.. code-block:: bash
+
+  | func:'ex1' args:[(), {}] took: 2.5083 seconds |
+  | func:'ex2' args:[(), {}] took: 1.0032 seconds |
+  | func:'ex3' args:[(), {}] took: 2.0039 seconds |
+
+Let's interpret these results. 
+
+Here, ``ex1()`` took substantially more time than ``ex2()``, where 
+their only difference is that ``ex1()`` calls ``ray.get`` on the remote
+function before adding it to the list, while ``ex2()`` waits to fetch the
+entire list with ``ray.get`` at once.
+
+.. code-block:: python
+
+  @ray.remote
+  def func(): # A single call takes 0.5 seconds
+      time.sleep(0.5)
+
+  def ex1():  # Took Ray 2.5 seconds
+      list1 = []
+      for i in range(5):
+          list1.append(ray.get(func.remote()))
+
+  def ex2():  # Took Ray 1 second
+      list2 = []
+      for i in range(5):
+          list2.append(func.remote())
+      ray.get(list2)
+
+Notice how ``ex1()`` took 2.5 seconds, exactly five times 0.5 seconds, or 
+the time it would take to wait for our remote function five times in a row. 
+
+By calling ``ray.get`` after each call to the remote function, ``ex1()`` 
+removes all ability to parallelize work, by forcing the driver to wait for 
+each ``func()``'s result in succession. We are not taking advantage of Ray 
+parallelization here! 
+
+Meanwhile, ``ex2()`` takes about 1 second, much faster than it would normally 
+take to call ``func()`` five times iteratively. Ray is running each call to 
+``func()`` in parallel, saving us time. 
+
+``ex1()`` is actually a common user mistake in Ray. ``ray.get`` is not 
+necessary to do before adding the result of ``func()`` to the list. Instead, 
+the driver should send out all parallelizable calls to the remote function 
+to Ray before waiting to receive their results with ``ray.get``. ``ex1()``'s
+suboptimal behavior can be noticed just using this simple timing test.
+
+Realistically, however, many applications are not as highly parallelizable 
+as ``ex2()``, and the application includes sections where the code must run in 
+serial. ``ex3()`` is such an example, where the local function ``other_func()``
+must run first before each call to ``func()`` can be submitted to Ray. 
+
+.. code-block:: python
+
+  # A local function that must run in serial
+  def other_func():
+      time.sleep(0.3)
+
+  def ex3():  # Took Ray 2 seconds, vs. ex1 taking 2.5 seconds
+      list3 = []
+      for i in range(5):
+          other_func()
+          list2.append(func.remote())
+      ray.get(list3)
+
+What results is that while ``ex3()`` still gained 0.5 seconds of speedup 
+compared to the completely serialized ``ex1()`` version, this speedup is
+still nowhere near the ideal speedup of ``ex2()``. 
+
+The dramatic speedup of ``ex2()`` is possible because ``ex2()`` is 
+theoretically completely parallelizable: if we were given 5 CPUs, all 5 calls 
+to ``func()`` can be run in parallel. What is happening with ``ex3()``, 
+however, is that each parallelized call to ``func()`` is staggered by a wait 
+of 0.3 seconds for the local ``other_func()`` to finish.
+
+``ex3()`` is thus a manifestation of `Amdahls Law`_: the fastest theoretically 
+possible execution time from parallelizing an application is limited to be 
+no better than the time it takes to run all serial parts in serial. 
+
+.. _`Amdahls Law`: https://en.wikipedia.org/wiki/Amdahl%27s_law
+
+Due to Amdahl's Law, ``ex3()`` must take at least 1.5 
+seconds -- the time it takes for 5 serial calls to ``other_func()`` to finish!
+After an additional 0.5 seconds to execute func and get the result, the
+computation is done.
+
+
+Profiling Using An External Profiler (Line Profiler)
+----------------------------------------------------
+
+One way to profile the performance of our code using Ray is to use a third-party
+profiler such as `Line_profiler`_. Line_profiler is a useful line-by-line
+profiler for pure Python applications that formats its output side-by-side with
+the profiled code itself. 
+
+Alternatively, another third-party profiler (not covered in this documentation)
+that you could use is `Pyflame`_, which can generate profiling graphs.
+
+.. _`Line_profiler`: https://github.com/rkern/line_profiler
+.. _`Pyflame`: https://github.com/uber/pyflame
+
+First install ``line_profiler`` with pip:
+
+.. code-block:: bash
+
+  pip install line_profiler
+
+``line_profiler`` requires each section of driver code that you want to profile as 
+its own independent function. Conveniently, we have already done so by defining 
+each loop version as its own function. To tell ``line_profiler`` which functions
+to profile, just add the ``@profile`` decorator to ``ex1()``, ``ex2()`` and 
+``ex3()``. Note that you do not need to import ``line_profiler`` into your Ray 
+application:
+
+.. code-block:: python
+
+  @profile  # Added decorator
+  def ex1():
+      list1 = []
+      for i in range(5):
+          list1.append(ray.get(func.remote()))
+
+  def main():
+      ray.init()
+      ex1()
+      ex2()
+      ex3()
+
+  if __name__ == "__main__":
+      main()
+
+Then, when we want to execute our Python script from the command line, instead 
+of ``python your_script_here.py``, we use the following shell command to run the 
+script with ``line_profiler`` enabled:
+
+.. code-block:: bash
+
+  kernprof -l your_script_here.py 
+
+This command runs your script and prints only your script's output as usual. 
+``Line_profiler`` instead outputs its profiling results to a corresponding 
+binary file called ``your_script_here.py.lprof``.
+
+To read ``line_profiler``'s results to terminal, use this shell command:
+
+.. code-block:: bash
+
+  python -m line_profiler your_script_here.py.lprof
+
+In our loop example, this command outputs results for ``ex1()`` as follows.
+Note that execution time is given in units of 1e-06 seconds:
+
+.. code-block:: bash
+
+  Timer unit: 1e-06 s
+
+  Total time: 2.50883 s
+  File: your_script_here.py
+  Function: ex1 at line 28
+
+  Line #      Hits         Time  Per Hit   % Time  Line Contents
+  ==============================================================
+      29                                           @profile
+      30                                           def ex1():
+      31         1          3.0      3.0      0.0   list1 = []
+      32         6         18.0      3.0      0.0   for i in range(5):
+      33         5    2508805.0 501761.0    100.0     list1.append(ray.get(func.remote()))
+
+
+Notice that each hit to ``list1.append(ray.get(func.remote()))`` at line 33 
+takes the full 0.5 seconds waiting for ``func()`` to finish. Meanwhile, in 
+``ex2()`` below, each call of ``func.remote()`` at line 40 only takes 0.127 ms, 
+and the majority of the time (about 1 second) is spent on waiting for ``ray.get()`` 
+at the end:
+
+
+.. code-block:: bash
+
+  Total time: 1.00357 s
+  File: your_script_here.py
+  Function: ex2 at line 35
+
+  Line #      Hits         Time  Per Hit   % Time  Line Contents
+  ==============================================================
+      36                                           @profile
+      37                                           def ex2():
+      38         1          2.0      2.0      0.0   list2 = []
+      39         6         13.0      2.2      0.0   for i in range(5):
+      40         5        637.0    127.4      0.1     list2.append(func.remote())
+      41         1    1002919.0 1002919.0     99.9    ray.get(list2)
+
+
+And finally, ``line_profiler``'s output for ``ex3()``. Each call to 
+``func.remote()`` at line 50 still take magnitudes faster than 0.5 seconds, 
+showing that Ray is successfully parallelizing the remote calls. However, each 
+call to the local function ``other_func()`` takes the full 0.3 seconds, 
+totalling up to the guaranteed minimum application execution time of 1.5 
+seconds:
+
+.. code-block:: bash
+
+  Total time: 2.00446 s
+  File: basic_kernprof.py
+  Function: ex3 at line 44
+
+  Line #      Hits         Time  Per Hit   % Time  Line Contents
+  ==============================================================
+      44                                           @profile
+      45                                           #@time_this
+      46                                           def ex3():
+      47         1          2.0      2.0      0.0   list3 = []
+      48         6         13.0      2.2      0.0   for i in range(5):
+      49         5    1501934.0 300386.8     74.9     other_func()
+      50         5        917.0    183.4      0.0     list3.append(func.remote())
+      51         1     501589.0 501589.0     25.0   ray.get(list3)
+
+
+Profiling Using Python's CProfile
+---------------------------------
+
+A second way to profile the performance of your Ray application is to 
+use Python's native cProfile `profiling module`_. Rather than tracking 
+line-by-line of your application code, cProfile can give the total runtime
+of each loop function, as well as list the number of calls made and
+execution time of all function calls made within the profiled code. 
+
+.. _`profiling module`: https://docs.python.org/3/library/profile.html#module-cProfile
+
+Unlike ``line_profiler`` above, this detailed list of profiled function calls 
+**includes** internal function calls and function calls made within Ray! 
+
+However, similar to ``line_profiler``, cProfile can be enabled with minimal 
+changes to your application code (given that each section of the code you want 
+to profile is defined as its own function). To use cProfile, add an import 
+statement, then replace calls to the loop functions as follows:
+
+.. code-block:: python
+
+  import cProfile  # Added import statement
+
+  def ex1():
+      list1 = []
+      for i in range(5):
+          list1.append(ray.get(func.remote()))
+
+  def main():
+      ray.init()
+      cProfile.run('ex1()')  # Modified call to ex1
+      cProfile.run('ex2()')
+      cProfile.run('ex3()')
+
+  if __name__ == "__main__":
+      main()
+
+Now, when executing your Python script, a cProfile list of profiled function 
+calls will be outputted to terminal for each call made to ``cProfile.run()``.
+At the very top of cProfile's output gives the total execution time for 
+``'ex1()'``:
+
+.. code-block:: bash
+
+  601 function calls (595 primitive calls) in 2.509 seconds
+
+Following is a snippet of profiled function calls for ``'ex1()'``. Most of 
+these calls are quick and take around 0.000 seconds, so the functions of 
+interest are the ones with non-zero execution times:
+
+.. code-block:: bash
+
+  ncalls  tottime  percall  cumtime  percall filename:lineno(function)
+  ...
+      1    0.000    0.000    2.509    2.509 your_script_here.py:31(ex1)
+      5    0.000    0.000    0.001    0.000 remote_function.py:103(remote)
+      5    0.000    0.000    0.001    0.000 remote_function.py:107(_submit)
+  ...  
+     10    0.000    0.000    0.000    0.000 worker.py:2459(__init__)
+      5    0.000    0.000    2.508    0.502 worker.py:2535(get)
+      5    0.000    0.000    0.000    0.000 worker.py:2695(get_global_worker)
+     10    0.000    0.000    2.507    0.251 worker.py:374(retrieve_and_deserialize)
+      5    0.000    0.000    2.508    0.502 worker.py:424(get_object)
+      5    0.000    0.000    0.000    0.000 worker.py:514(submit_task)
+  ...
+
+The 5 separate calls to Ray's ``get``, taking the full 0.502 seconds each call, 
+can be noticed at ``worker.py:2535(get)``. Meanwhile, the act of calling the 
+remote function itself at ``remote_function.py:103(remote)`` only takes 0.001 
+seconds over 5 calls, and thus is not the source of the slow performance of 
+``ex1()``.
+
+
+Profiling Ray Actors with cProfile
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Considering that the detailed output of cProfile can be quite different depending 
+on what Ray functionalities we use, let us see what cProfile's output might look 
+like if our example involved Actors (for an introduction to Ray actors, see our 
+`Actor documentation here`_). 
+
+.. _`Actor documentation here`: http://ray.readthedocs.io/en/latest/actors.html
+
+Now, instead of looping over five calls to a remote function like in ``ex1``,
+let's create a new example and loop over five calls to a remote function 
+**inside an actor**. Our actor's remote function again just sleeps for 0.5
+seconds:
+
+.. code-block:: python
+
+  # Our actor
+  @ray.remote
+  class Sleeper(object):  
+      def __init__(self):
+          self.sleepValue = 0.5
+
+      # Equivalent to func(), but defined within an actor
+      def actor_func(self):
+          time.sleep(self.sleepValue)
+
+Recalling the suboptimality of ``ex1``, let's first see what happens if we 
+attempt to perform all five ``actor_func()`` calls within a single actor:
+
+.. code-block:: python
+
+  def ex4():
+      # This is suboptimal in Ray, and should only be used for the sake of this example
+      actor_example = Sleeper.remote()
+
+      five_results = []
+      for i in range(5):
+          five_results.append(actor_example.actor_func.remote())
+
+      # Wait until the end to call ray.get()
+      ray.get(five_results)
+
+We enable cProfile on this example as follows:
+
+.. code-block:: python
+
+  def main():
+      ray.init()
+      cProfile.run('ex4()') 
+
+  if __name__ == "__main__":
+      main()
+
+Running our new Actor example, cProfile's abbreviated output is as follows:
+
+.. code-block:: bash
+
+  12519 function calls (11956 primitive calls) in 2.525 seconds
+
+  ncalls  tottime  percall  cumtime  percall filename:lineno(function)
+  ...
+  1    0.000    0.000    0.015    0.015 actor.py:546(remote)
+  1    0.000    0.000    0.015    0.015 actor.py:560(_submit)
+  1    0.000    0.000    0.000    0.000 actor.py:697(__init__)
+  ...
+  1    0.000    0.000    2.525    2.525 your_script_here.py:63(ex4)
+  ...
+  9    0.000    0.000    0.000    0.000 worker.py:2459(__init__)
+  1    0.000    0.000    2.509    2.509 worker.py:2535(get)
+  9    0.000    0.000    0.000    0.000 worker.py:2695(get_global_worker)
+  4    0.000    0.000    2.508    0.627 worker.py:374(retrieve_and_deserialize)
+  1    0.000    0.000    2.509    2.509 worker.py:424(get_object)
+  8    0.000    0.000    0.001    0.000 worker.py:514(submit_task)
+  ...
+
+It turns out that the entire example still took 2.5 seconds to execute, or the 
+time for five calls to ``actor_func()`` to run in serial. We remember in ``ex1`` 
+that this behavior was because we did not wait until after submitting all five 
+remote function tasks to call ``ray.get()``, but we can verify on cProfile's
+output line ``worker.py:2535(get)`` that ``ray.get()`` was only called once at 
+the end, for 2.509 seconds. What happened? 
+
+It turns out Ray cannot parallelize this example, because we have only 
+initialized a single ``Sleeper`` actor. Because each actor is a single, 
+stateful worker, our entire code is submitted and ran on a single worker the 
+whole time.
+
+To better parallelize the actors in ``ex4``, we can take advantage
+that each call to ``actor_func()`` is independent, and instead
+create five ``Sleeper`` actors. That way, we are creating five workers
+that can run in parallel, instead of creating a single worker that 
+can only handle one call to ``actor_func()`` at a time.
+
+.. code-block:: python
+
+  def ex4():
+      # Modified to create five separate Sleepers
+      five_actors = [Sleeper.remote() for i in range(5)]
+
+      # Each call to actor_func now goes to a different Sleeper
+      five_results = []
+      for actor_example in five_actors:
+          five_results.append(actor_example.actor_func.remote())
+
+      ray.get(five_results)
+
+Our example in total now takes only 1.5 seconds to run:
+
+.. code-block:: bash 
+
+  1378 function calls (1363 primitive calls) in 1.567 seconds
+
+  ncalls  tottime  percall  cumtime  percall filename:lineno(function)
+  ...
+  5    0.000    0.000    0.002    0.000 actor.py:546(remote)
+  5    0.000    0.000    0.002    0.000 actor.py:560(_submit)
+  5    0.000    0.000    0.000    0.000 actor.py:697(__init__)
+  ...
+  1    0.000    0.000    1.566    1.566 your_script_here.py:71(ex4)
+  ...
+  21    0.000    0.000    0.000    0.000 worker.py:2459(__init__)
+  1    0.000    0.000    1.564    1.564 worker.py:2535(get)
+  25    0.000    0.000    0.000    0.000 worker.py:2695(get_global_worker)
+  3    0.000    0.000    1.564    0.521 worker.py:374(retrieve_and_deserialize)
+  1    0.000    0.000    1.564    1.564 worker.py:424(get_object)
+  20    0.001    0.000    0.001    0.000 worker.py:514(submit_task)
+  ...
+
+
+Visualizing Tasks in the Ray Timeline
+-------------------------------------
+Profiling the performance of your Ray application doesn't need to be 
+an eye-straining endeavor of interpreting numbers among hundreds of 
+lines of text. Ray comes with its own visual web UI to visualize the 
+parallelization (or lack thereof) of user tasks submitted to Ray!
+
+This method does have its own limitations, however. The Ray Timeline 
+can only show timing info about Ray tasks, and not timing for normal
+Python functions. This can be an issue especially for debugging slow
+Python code that is running on the driver, and not running as a task on 
+one of the workers. The other profiling techniques above are options that 
+do cover profiling normal Python functions.
+
+Currently, whenever initializing Ray, a URL is generated and printed
+in the terminal. This URL can be used to view Ray's web UI as a Jupyter 
+notebook:
+
+.. code-block:: bash
+
+  ~$: python your_script_here.py
+
+  Process STDOUT and STDERR is being redirected to /tmp/raylogs/.
+  Waiting for redis server at 127.0.0.1:61150 to respond...
+  Waiting for redis server at 127.0.0.1:21607 to respond...
+  Starting local scheduler with the following resources: {'CPU': 4, 'GPU': 0}.
+
+  ======================================================================
+  View the web UI at http://localhost:8897/notebooks/ray_ui84907.ipynb?token=025e8ab295270a57fac209204b37349fdf34e037671a13ff
+  ======================================================================
+
+Ray's web UI attempts to run on localhost at port 8888, and if it fails 
+it tries successive ports until it finds an open port. In this above 
+example, it has opened on port 8897.
+
+Because this web UI is only available as long as your Ray application 
+is currently running, you may need to add a user prompt to prevent 
+your Ray application from exiting once it has finished executing,  
+such as below. You can then browse the web UI for as long as you like:
+
+.. code-block:: python
+
+  def main():
+      ray.init()
+      ex1()
+      ex2()
+      ex3()
+
+      # Require user input confirmation before exiting
+      hang = input('Examples finished executing. Press enter to exit:')
+
+  if __name__ == "__main__":
+      main()
+
+Now, when executing your python script, you can access the Ray timeline
+by copying the web UI URL into your web browser on the Ray machine. To 
+load the web UI in the jupyter notebook, select **Kernel -> Restart and 
+Run All** in the jupyter menu.
+
+The Ray timeline can be viewed in the fourth cell of the UI notebook by 
+using the task filter options, then clicking on the **View task timeline** 
+button.
+
+For example, here are the results of executing ``ex1()``, ``ex2()``, and 
+``ex3()`` visualized in the Ray timeline. Each red block is a call to one 
+of our user-defined remote functions, namely ``func()``, which sleeps for 
+0.5 seconds:
+
+.. image:: user-profiling-timeline.gif
+
+(highlighted color boxes for ``ex1()``, ``ex2()``, and ``ex3()`` added for 
+the sake of this example)
+
+Note how ``ex1()`` executes all five calls to ``func()`` in serial, 
+while ``ex2()`` and ``ex3()`` are able to parallelize their remote
+function calls. 
+
+Because we have 4 CPUs available on our machine, we can only able to 
+execute up to 4 remote functions in parallel. So, the fifth call to the 
+remote function in ``ex2()`` must wait until the first batch of ``func()`` 
+calls is finished.
+
+In ``ex3()``, because of the serial dependency on ``other_func()``, we 
+aren't even able to use all 4 of our cores to parallelize calls to ``func()``.
+The time gaps between the ``func()`` blocks are a result of staggering the
+calls to ``func()`` in between waiting 0.3 seconds for ``other_func()``. 
+
+Also, notice that due to the aforementioned limitation of the Ray timeline, 
+``other_func()``, as a driver function and not a Ray task, is never 
+visualized on the Ray timeline.
+
+**For more on Ray's Web UI,** such as how to access the UI on a remote
+node over ssh, or for troubleshooting installation, please see our 
+`Web UI documentation section`_.
+
+.. _`Web UI documentation section`: http://ray.readthedocs.io/en/latest/webui.html