From 8c4e6dc61b1885ece3293e9634677ca8098efdd1 Mon Sep 17 00:00:00 2001 From: Justin Yu Date: Thu, 27 Oct 2022 14:46:09 -0700 Subject: [PATCH] [Tune] [PBT] [Doc] Fix and clean up PBT examples (#29060) --- doc/source/_toc.yml | 2 +- .../images/tune_advanced_dcgan_Gloss.png | Bin 70378 -> 0 bytes .../images/tune_advanced_dcgan_inscore.png | Bin 91101 -> 0 bytes doc/source/tune/api_docs/schedulers.rst | 2 +- .../includes/images/pbt_function_results.png | Bin 0 -> 99740 bytes .../tune/examples/includes/pbt_function.rst | 5 + doc/source/tune/examples/pbt_guide.ipynb | 789 ++++++++++++++++++ doc/source/tune/tutorials/overview.rst | 4 +- .../tune/tutorials/tune-advanced-tutorial.rst | 238 ------ .../pytorch/tune_cifar_torch_pbt_example.py | 96 ++- python/ray/tune/BUILD | 2 +- python/ray/tune/examples/mnist_pytorch.py | 6 +- .../tune/examples/pbt_dcgan_mnist/__init__.py | 0 .../tune/examples/pbt_dcgan_mnist/common.py | 5 +- .../pbt_dcgan_mnist/pbt_dcgan_mnist_func.py | 96 ++- .../pbt_dcgan_mnist_trainable.py | 4 +- python/ray/tune/examples/pbt_example.py | 20 +- python/ray/tune/examples/pbt_function.py | 14 +- python/ray/tune/examples/pbt_memnn_example.py | 8 +- .../examples/pbt_tune_cifar10_with_keras.py | 8 +- 20 files changed, 973 insertions(+), 326 deletions(-) delete mode 100644 doc/source/images/tune_advanced_dcgan_Gloss.png delete mode 100644 doc/source/images/tune_advanced_dcgan_inscore.png create mode 100644 doc/source/tune/examples/includes/images/pbt_function_results.png create mode 100644 doc/source/tune/examples/pbt_guide.ipynb delete mode 100644 doc/source/tune/tutorials/tune-advanced-tutorial.rst create mode 100644 python/ray/tune/examples/pbt_dcgan_mnist/__init__.py diff --git a/doc/source/_toc.yml b/doc/source/_toc.yml index cb88dc1400fa..24b8e67a703e 100644 --- a/doc/source/_toc.yml +++ b/doc/source/_toc.yml @@ -101,7 +101,7 @@ parts: - file: tune/tutorials/tune-metrics title: "Using Callbacks and Metrics" - file: tune/examples/tune_analyze_results - - file: tune/tutorials/tune-advanced-tutorial + - file: tune/examples/pbt_guide sections: - file: tune/examples/pbt_visualization/pbt_visualization title: "Visualizing and Understanding PBT" diff --git a/doc/source/images/tune_advanced_dcgan_Gloss.png b/doc/source/images/tune_advanced_dcgan_Gloss.png deleted file mode 100644 index 44d6c1dcbef21c28c932b351eaa22580dedb8221..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 70378 zcmZs@2UJt-vpq~lx=N7_A_6KML3(+qDlJqAU7Cd6dx@wB1Q3y4B}nfOdPjoPNbeA( zgbosV{SLnO-rxP-^<}Lra*~sCo-(s%&z=bp+M22qq>Q9EI5-r~)s%E_aPS~FIJlr& zM8G@W!=W(X4}r%&&-HErkHA~jAAw^MS2ZIK92~N{H&5J=`QQuSO>s}sh}Ljo+EEU?LmvxP=C#d{9Zuv z+%-b!DN+f9jRJ-JK+#k9U*64$$bqqQ{3d-HDu~~;6;M7pKaQ9WXbqVQC@yX)_Ae@K zYAm*0K=r`cOnnLiw8l#tFV|7iFa+Hl;9)N_q;pSR33wFgW^-|HaE#>86hwUd__2nE z5CR->@V$Vh0*7i*I{)kA8bPapd1P{0Vk0xw2G%e~Zxt;RsA#X?+&(EKwEv;f_nuL!-BW|=P^B6uegW50tpfscPE+W}6C%EZcH zrR&sYfzeg{$dTW=7fN=4zZZy>MrX7`(5c^^%yU|)pS=>tNMCYWvb&MG67D&nyV`-l zmz^?;2Md%S!X{?AuzYKGwm1S={-C6ALuMKm(@WD;UzhT|ayXuqh^by5lZt4OD)FF;a--U4Y!-QO(A*G#Ry$n{Fg2?7c`uLKm)_Ho5 z7NOOwXrVBWErvSbeBhTjn=mqPhF%2cf*pJv!!!HL7B=kyJB<{w4XW8*_LZ|Rwps0d zS0;p%rnO10Wwgp`t1ZGfz7ty20DY*kfK9Wo)a@U(0E?+;HSp#D`$=eZg?4t4IziM0 zCSeUbjSWpy%xnfO*6c6nrF@GN_RM;Q4E#x7O*Lgy zd6j2>J`a3vcFmuh(W>V|L$pwxvduJGi=nv?@?!T+S-x672-w7rFrnhmC(w|>golL# z;`CNmF@jx3&S-TsOFJ9>q(;EZEHtPco6{3Q)LaN%x{z?a?TfodOuJc*n@Q44Cr)zm zq9*v>-p1M=LiNGsW{MN`y*p?}G`w9E*q$3e}g{!^Z2e27&2?sgpB2F{j zNm}IgE@z5;!A$wCP}d{>OnIS$EuwT*G{>u9LKp}=@r|9W;zreE1tB~ zK-B|j41PPyu!k&3-~c_gGjiJ;ZwQB<*o;*A!Q)u9^U8@0Rk<^xT((XNBwl$&?Uih& zoTpVge7KRnQ~vk9{rMHEph;yj%L%e3$eGyL9xd)O@rN;!rRBpCeC3{UKeyJb%e_6V z(TN~D-y*1o8FR+cJI2`%9N*E27W`!P0kbk#W{6ufrj2Y0FL%|IZp5)SwzN24u^QM$CIdER zr?<6<`^X|(Xq60^yjc36ScFNDo_ApUa$+Vs=8`qF)}(a-nM~VGgzgzVItAgc6n>k* zEFU711EPh_b5N6m@>*Y%Gc{K1d|VFB>mgDTzM2=!ClYqJ3N{Q*Q^@Ums^WQSFK^JN zK9lp}fxGR>3x{ofhwb=wi|&n+^V8XP3MLm=_0Elkz96#EA*&~ua!E@tL)P!~9(VWn zI$=MVLNR#dkzxXRUDtme>I7o(=jd{!E5Q?4W$K2}(Db9EN&VQw3=bjHbXCzLcE$fY_}tB5o3VrAlPCtlZSb>6c}et>}~8IXHj}t zi0~$>(CZqwh}?6T#r~#6CX;LUw<)Lc`7N6&SwmYWPVVnT6~tYKy>l8JVt1t3!GGn@ z0~2B+l68btw)>E@Riik!gwCGn$H;U^$6^UAZGQ_y3z>5{Hp{et&L}%!!wJaba=cTX zS0}`1BC7%FciTl~co!s=53W@*Y-}l$8y`weUejBZ5ne?JTAlc`Q%uaf6MFBRz_a7k~I&~docYv zKQ!npT2e?|B0}RMlY8J*7esr)sYy07yad~>s|RVc4Fprcks0ueR2>jkBZGXqbu07+x}osZ6H;N>lm zH&nPS@1{q*cuhZOY z*6;btha))pL6Lu%4h8D||7Jb~@P_Er4NqkyrQ8hltd8XZ%CV^?pWQg0)bQ3;*|YQW zG=EBaO4FvJ+|7wr&+VC{viaXr6_#hq5AXl?75@)+0)G*Mlx=3JiE|vdy^FqZIgsAX zGMzM47C)ZvZTD_Jly)iT!!8mu3xF>^Vf?OJ=~n=97I$P(`Y@^g4goH5;7$DE-4cX& zjH0Tl>b#9f$q>Oqw15mMpxtZUrv0*P(-|+At$q~S99v%&!1wE2Ux@DMcb?=XN?(4< zPd;s%jf(|zKLJ+vqw-Nwf82|2YFx%mS=NPm97+Ad)6+(iS5Z2L>wXKeC@kwo)>yn) z+#dXLqTgFnZTRJd`v*z*Y0@Yr*xPE2_LV^a#LbdLvEKsAs^P0~#S&C% zcO$-kPV6`iDr-3cJ2$M*O_ZCX5)93HUz(rkWL_QQYfTGep-{eBNuKS-lU@{ierO>S zR@SU?ovA%7r^f}a)of*b5!Y%YAEq+1}kqnA5t z02$5nJb$$LY`lC`YB9z9N7X9^#67$=GU=_xov`7?V;nYI=}E6e@+}+_oJp@NvgUDY zv)sP4%TKR)Dm=NUOrI8vUs#mhPHN_feoR=)gY@I(F`avW6)S<;uP)E?t&ul_7YP0n z<;UUy!V^VdcfR#0!Q);QZ(Vz!V;?8F5IH+*t}|M}I>3YH{6H!Id0QSx+2k<$mUT4U zmvo3}p(?wlB;1zk%_pI3aShuVX3To@-0@=g} zB#{G4=(^yCsdSnZ^WK^3ZS>kaIh$-7YP;HO8`?}J9b>d z-zBC+wc7;ESBK>on8;A_qR_np#~x-Iz_pdE8(#{ve0eF5F?`W#34u>!SBP}zB>e+? z5@QQo%$l6w1_6|EBWMalf6!M%{?9={^lC6I;8bGY zxqn~ktP5QI77#z4^YH6^X;x*?qnACw9u2?AjuK2JSNhGb_TRGVLcw7$Tdv!{B}q{s zSJ?IKZPTAtkF^~>3E$*RBaW&66e(C{*N%O=Ic44X>=d__VukJ?Ilp8J9h(_2_m;Jq zweMz-@D!4p5f^>|IehwiL3@ptaigsI%V;OZIYIOW;H(0l36}-Tr1}VKXV|BUKplJYlOm+AzPqz2GJ zWRDb`jq$@MZMKr8-`O4^BM->*F-yQW6RY5|@%UvP&vo-)BgHW)BYiuND8sJ8Fe=BvDfGThWn-G47%BrAhzBBn2?9rQBAuW!E~W-?Mj zzH1Xz3I{eDXKT7Te6nT7%Ovnmw~2f2!#U4U4Xmu~6BztxvmFy4=)9haL>ZsdJ^Boqk&V>E&SF=)GNR zS~d^GCsyN15-V9>TSK31mSg5TP&ST^uzvm~*h~I~cOpaLR(-)K>UlfIHMDq$;$)e$ zBMubysJ^U;Zfd!4X9@Jd`2)|#D13?8#l_~Ck%$P_s!7H0v*(;N1&A~t5~QB~Kld}0 z&Ek$z48KuUw8QM1&SvbW}lOe(~`rOjK9V@47cfpr9<%1oPlm*Z4@kE*j|%9nV(RJOmh zFJ)r3%W8+c=AGOBNV`1szYs>Bb@N*Axn+OA1GXtHasLOt5hc|*ng75_3(1h5rlIM1 zOl&MRv99?&RUjIl*SOZv3}gBQmb%$kha-nKEl!de*u5W5yI;rEHLvhBA0f(D`ks`hAO9HuF~6B=2{;zp zMr%3sV%G$?d`}$`hNp}j}>o%pRE{|8uYs*M^OH~M`=qY#Nc%}Xo?M@R0LhlnRQ z#R;MixEOr6i33Jd7=0r==$%cYf{~?l3w=Sy^F206rpTLThR=dTxfQ<`6QcIpUvqwJ z>e@jF*p!_@Y#|IB_E*DDtG+S(n+3UJFGdKV5C-LX)A%57{PAHPfC7Dgz2pC!G%`T& z*L@i?;h=v{cyGer<8nFS5{JV43CdbLiC#pfI+UeM3ef`I`-b$`U1-&n^z>u^mKaDBZ3ULEC*i)y`qoLGn{Wa5w9y>#oS}OL8E_t zc=WUtbu+aFOlH6GJ=5qc7A^fOioeRsNa{`lhLw^UOTcV%OyU2GD**_&?NW=Uns3?JZ{eAp_8GcQ z+awMC#qR&YC(j8ZJGUMdAKo>gaR8J4Mc`=wfnVsV-2hPh{{f(X0oU<=0C&VOYGiZN zYd#HXh(j!clHvv*`?MQ}`mUjmeKU_sAER2BPwd!n=a&5zRO#RYsyTmQ(Z05uF>{Hm zJwH@H)A*VM2)hQG&&xKsLPH4o@xAfoJWP+7GBKA$MueqFe~q9J3B>XG-=q_z#y$7i z#AOar(v%C&04~OAQ31_2!;uCkkZvZ1Hnbe4-0WwksM=rUOV49eE zcc+P=Ordh=C|uLY7RGwUcC7T-+1aWwG8HMTZ1H7h$L8RW&t=9hSK@Q8wXPi72y>(J z_RGn#AVfbx$`y^%yF8**e*mQl4Baeiim9v=f(hyOf9bW}2-Zs4qV>!4VZQ-mOz3rK z)hsLed*ljrju#@v?}4jT&71S{^LyF3;~LiZI4Y{D>$?I^x1~pzgMxydVzKuJzy8Z( z4Vz4w0%f&0;&8GtKd(Ox>>B@EN)p~=Fvj*79Z%QTQx_YQ{@$uat+t-;bb((M>J@<} z*8dyLdhtX7dKHj)uPv&akhQRC78N>bIw|vBP(A>E52^h4cgdK_&FVb9yuip|Vx)G^ zWlcLAStt=n$-ddeEAOin^Q&*atWvw_E5^O%jYFfLvS=Jv)0#k5%W;f1I#vcXi)q{; z>a-LG6pVm}d(S(6NjL-aZg*3~M2^cNat9;)|c3;mvqMahPPwg&|niTzRMIj#;%>xmsVsyEj;h2)^{k0Zh66cjOb_ zy$)hvkTzqh9TNZ8y^R?ruHaJsdOpzoisgvv4p3`MM~joz|n=jVqziuE72T?Z{3 zmihd}lhqFK0a+0!1a}=hlYKMmaQ3Eqp;p^LDF56rnZ^jc} z^IMP@Zj8k%Cm?M%KvMyUQrZ68b~_Im{~ z>{kroZpiidO?0WpWClzDKAE2teAXOO#{*5p_#NYfjZVz}@-uq`z7TTleEr4d{vch! zqZ{Vx?kEv>A}Dv-t5v^IRE(pRKU(st^0iryV}ckYx45@yKJq?FgN0eTZeIr2bfFowL63so^D; zgUK^{`~QP=*7OqxTXP!1QkB^~0$yIi-Xn2Ou33XmEM+f#QF&aQVz3Lxtw~xTXQL(o z+X&tNO8j0j769mc{w{-@|4_OdveAoemKoj+1Gb9D1V@izOd8d6r(OtLb9s45X12Oc z`krHaD87%f@-7b@mR#T;32SG-jEl^k**pDbIQ@yEb>_Cqb>i7GeoQ~zoiZ1z`1_mPo8ZEm%6!n-?$pcw&KeUa=Y|GDVyg7Tb zaXxg(Z{f|siB0=q)T;p_2`>al*A5R@;BL>8cN+>?)upCQi4;s?d&}`!?f`Ri+if}C zKggF0GV3uZ0RRt|pXkhFa-PCOb*f)P@l^)No)rqaumJ!gnhS+3BZ$Fwj;iH`UIRgG zHVU-rKOiREW6dP`H{eSc1nz~gEpNMoNCE6EUQ;q6HsoRnE17AuHfBLq2}CWY4U*tL zOL*{btjR%@jYzExlGU>NNxa{c8$cHDAJ4jlIV)i;Oqk8HXZ|1mOBRJ24COx3zVRm{k9JzP zAl$$T{%=C%r;{y4h5Wku9sjS_#GT$8A8PcK{iS-0aCs+I+imsGB3U50jd`dgd&-$) z>TPs-lPsb=vd-t4V;Yt{f@-U$ZMn2>FnM|EW-Tl2N4anU7cm!D6ejXXS@I;1E~SM7 z=kVB#0HXAbcg_Xuwm8n!$5CsF?c^(4pM@2i-D7rR z`Zq4n8SW*ZEPb4ee_EwdYjyZ_8E?$gY4f-OglKQar!8Osqp#w`3j4ac$E+(1QQZwJ z0-#%X@VNC@Ci8T_uF7^0xadm;0Fo$GGyfsI1#e)$mH@((8ueJ(DJ1dwYWLdg6GTf} z#mUSzDE?FKSkb1V=5218E(zALhp$d_Y4%noI`W-wSB}_D)8IIJ6DFb zJ90Bi`xNwf{6{}Db7@GkKN5p%0EB`MD8;H7C->9n?Be|{T)W0S5#PIyj`xziCTm|A zoiHVv16c2Rd&clS4M@<3Wly?OXuizD)MtgcZ~iHl|8Y?rh^&GO78VUqXnP^Co$W$- zjofd854DW}?d;=bV}Ga}?llmnFi?>Vzp8&@kU<;c$7-ugOzpwn26wxvs|`B zfiGXz{*r(lR&vqu0iGX62N#2IIDOqI(y~a}Y`lkOhY#?5OjB7?`YAU5~Aze}Y+RvxPnFxqaL}f%kuGQMQ8q+N`Cr8xYmr%U6lX7N|nn9;d^p?cmkT=|> zXRE%Q?@@}n*d_KpnbxG!FKa1b7+%V|u1IBHEw8_l?MUj&>?6bWYeLd?7v)Ve+zZe9 zq`&<#y-a<~;Nt+B{%_ulF|RNpP+$y&LKV2&b>riUFF7W9KM|;WAbzxJ_!Pk_{OpEC zpa0i=2)Zmi&=*rvInDWi3h{E%+0AjP;_Jp)> z==Y>|LpK{`-Z9kTobdcg&x$@F+T%h$7~U;6oW@AChd+_L-AwN56NWL?9&*%GGisYl zk-xdl3JrX=wj5adQb+y_5@%e0cn=#sKVBx)g|Up5x9Kq*TD6^_Y;y;AIrt8D#s<&; zaGRfkkzVtd#~dEBvQh;8=4-6gyNgeq}uvRih%;#}j5 zdlO2E{YI)cHwP&Lr_Xel2t#}`eZn^zxdAGHXQ4HgxRtD5Eu!!RGzv%qBUONx+LL)O zWLnAW(ROBLhP6W`y+Q~~fSE0DC6s@Y5M4ck1^6MMss2HcYQ>chU_x}OBPnQ+cE$F` zDC`9O6DTnhB?hLZY`P2F6+frr`wclMGpG12%+ZP$$T6%2$9 z#0W|1?@nWa>u>&b!lJX1UT{>-J&6RHUAyfC5aF1Qvjy(~ z49AGLv;bj*2W~KX#7*Mt%O?G&_#1!3k90}u3+g@7NiR97EgUht5S$Aq5z-0P&RZH8 zBucs4-*HoSt-^4Y23D?*;3nRV8!mAVpCy%35}XkjPz)`)xIZcPyQO2qUDru);GoOwvG- zJri07A{s21%>P(Z&<9JMDb%2~8zJp0=p8ZsD8D_We$UJEHP$WLAc(ZkeVqG7RpHCi zOW$fNs~pRV#J@4is)LwUhwF_yFKwp-2Uhzu@~HRmSm&$t_|xn*Rw%8$SbC ziP~m_$Qzan-K2+rXg%@|AhtXnji?j_*i0SBqw{ZGWB(9jVXJp&WD0T`%k{cRGS}DA znFXBAYJHux;%^nprlo_hVRwmp*jzRwl$Lq!60-xQ#qS^TR0LXs2V|%lTSWzD>9wWO z(yG5dG9WS@__=>OC3Fu^2^l)By_T8&jc$&4z|r%uzOq>uqMv~ObWok0&q7mI?fLtd zA$J@x5$z~Du8$uu1#uvzOrAfLE;|pCNKWK1<-aDM@pmORXdP(pRS)mFHR=?vUIgBe zKz?Va8LEg*<)e!oeq;;zh;L*wmP^QMpAD2x7}^b=7|2lgwqg9TN~SQc(VOkHZ^lCa zpuWtZtx9~9L6YbEn94UQOcW6iw6litFnaxvY6|3h^2>}ey8Kt&keLi7hd`{k@C54o zm_sLglnnTfN~l^1#x|8Uzr?$t+Y`_wGaBnb`Fhp)GhA5 z3tKRt8ara6bV)3i45DqSTbY)fdzCs*l_Nj6k;lQ|l?)WY4977vfZ_4wCd?F=k~#u~ zy&j!=x%{*lBtyjp!Wa<1~DKW_2DlR zMW^z%7(n1hMCe4uSu-G1D7KrS8G+)QY4hvxpTj}{c6jl(*oK7BP9YO^gqF-n?&ROH zD>Vs&X$lf`w*?zAzy?0WMVP`U?NpXPqn?-%|&i!U!Z z|J;s?B0@>gMl&|I)7{f9^(r-%uGjj|fumGo(#djJs)?S+pu`w;`)7jY@aXa}BBRVS z1igfxr{|a^{7hHMv$6)X{YyEWANkA7EUEJh9Y?>WadJMX2@P$2|g;_#sC@%BhaM_<%yz z1nm#KE_!ur0Hn-}e65teoWMR`n%tlGeR^ ze&hKVzoqsK`%PKnKPS5F`E*Sthk_ZVG?SMCT2p)%VC6eqZew?cKjRNHlk70@CE|-N zhU;|>Jb>rPYb?K-5R(39cW9nj9HZ{Y0rJY?Gp@bSEX*!`-?RGN&YSYq`m^b5Cdc`d zRl475O4O^33F}gdu<9KfL0LxI$=|)3%8Bit?Z#$O*0(s;=H9z;-OAGoIs0`0x=rCL zCF0czOiK7HobQ(7*k*MH4180L>#4H-`q7Fn#=I%j+1Hmo)KN_8A>5YwIVBZk6_A{E z^FS20T1}^ZbJ$$|9=DH1~-p5x4t_6Ed-nK~z1*p9xWk*OKCz zbqWw=mx|VL3gzWj))#xMR**SG_Xg(Ef^f{q$;lV_QfqP&CN%1Ew6S39oYr3Du|=I6 z=-PvbN}=%MW$*kHQBP`m{7K9D21$s+Pve`3F&?zzQ&l_^eZg(Y`CK@eDQoTx{_nbp z-xkS!*tuum>}Mh=pqGyUbqA64B)ayK} za~)3vx4ch8*{K`*?5p0Wl8mf|4|4%a+VMDF>?%FO2#mH=9?&gE)}MCo@MegSkA0|) zho%<3fg6-gV%{v+mE5l2rQKOViNFmqH+6yTPN7~BTU3EQ#n|Uxq4V$Cf4^$tfigBk z>>bB&Zw5!QNYls)5}n5fa7{jY(nLo!c!wC$gWe)gu$aD3i)m>ZW2&bG0+P0526C4% zKT4CC8M#j#&-5hk6Y1jbLVeQno=$;at)iR_kvra8$r2ssafGS=yjV`Qy_lXYprE3% zg58hr1<3Kshd`BW`7+yHKdV-+(}49E>DX3G=^>kSfgJPWrDlv}NU(r6N7Qv8)t$iQ z@_Y3)apHoTlHMYaZ?8`zMsB;8Dgcw#S*P)g9<}d!9?cyY?=*Ii6Z@@d5;JyvQA_dO ziM0<3emcqhQyZlPG zoy5uz0b;h`@aSk{eqQFGwv^}EZK#kZXZiyoi1+5wl!>?IO4V$w1=UuD$sNNWBHXP7 zvqhl~XO=c>>DY8ZUq;q<)k9?Ph)6DOGA>GXK*A}8rSjg$(;MRI+@|qSis83fFBp(2 zq~cuJ%RMfoiYt4}6JS<}G$KY_V6#%JBw=JZ|kGM&lyDHtaK#gi$7adO$ZAM z>rjpX%QXtfxVO^OKP-t4-7T2j67P^Rd$|{}S6w*Y=Y9OV|sxBH_NTcCW00Y0&2I}&1GD@m~ z65zu|Q{zgXMNPLw1sDlWLXKJ1Xpt-xjCLKBm?O`3R{k(FS0oDtHaYa$`nINQ?aqvV z=x0$=Q`3+V`UeC68Wu;|;0Fag;ogog;^*F;zX?>|z{PxdN*H`0A{};DUt2ruh$-Jg zt(578R=F((CD)A148!pp_{a~f436HH@2=ZfY5Gap`nlA1{rY8;IiD>a=1m?8rz3{+ z6(ER!;^X6o#J$S_{r$ZK8zRqAA6Qy#1WwZp!{Iz!H{>{RQ3Iqd!FwmDIvb!>2+w3zOkdepb_V!5~JdJ)6vC7zC6o?*cS-4R(;`diLx~l1T80 z4S&o^rx@g0AFj(2VF)pocz_dkjL@hs1XPZaupRp?&$sT0p5>@Th9+^0Fl(&)i4*@R zA&?&zgZ%l)4EX31Vu-?hE!oCNPkv*A%m+2LfO&`=zn#N9j{Vc&Jz5Lc>z(Q(2gnPef{?YM=15-jFbzjf>XCywN0T^;- z=DN~BR#lOlGTr!1MH{GWH(FzS9$zgLPi%Vl`X;aRC&jCAu5FO+%GADMH+&q|Ypv#U z{H4dTn4y4z-qn$xTo;($cZ7k;vv6fU@ ze0KYRkAu;LBDxt~sb8BY-<_M%|fk?u|05P(1H~ zLM*E?z`=OJfN-F(O-V?6{2mjyd?iTl$-HJN;nBA8q%vI9RzUFqKsFL6t8Yfl6rI-t zL4LX)PFk~J1&w+xb^0WGP=Iw5HcxmiOfM69I#Wx#{&87Yj-NIl2 zxeVbrlO2$N(@_k@X};?)e+0=Fla@Bgd}pMTYWMo}Yqvb}qOVmOqnb}t3-w;Rx_%OM z=oB+6)Qi-6OxTUr@H)b4b3#FwfheGw!ryop`I*|IEfWfZmai%{s(omY&j+SrS6VcR&UoK0y_@dx%)1hUxN-4G)$ZAVt=;VK z%=zrWZ}AXBrzuNnsg%e_%bJ83XID*acFVxP%}-es@_}I_^r<|*3 z5l3lJUKizXytMugxF`k`IjX^l%yztnHLH@gMNqP857)gCr*Kb(L z%Q%}}^aCGr*wXb!UarOC{>5rr!w|{fu`VfdDY$J|3KYy0Z+W`Evf?#AKhNnXp96oe zW%7x9_c=ls^5-@~f{egmBeg|s5vt$;Gy+f5eY$r&KS$oFN>M>s3Q;gK+PwG8@)Lom z2!wZ+u&rhf3aY$as}WWp(!pFT+tZWS%)J*_D5Bs1%AP1tqWbVOMud@{u8YY;Rro@O z_&WpQgfBA%gbnJ`&-!B)9Rp8{3eqQrG0SxALJ*|iXg1KzN0TIfs%Py4Lv>ty|~l(2bZqOw&XpXZ@n<*iDTJ*139PY*roAf=~ZO*$u z#g|8$3@>ZAS<9to>?X|HGY%Fc6w7ksBTiz&@kHZRQnnE`t=1LtPbHG~hM`$y6?V zrYWGnz;qsXqS&vbndd4kvhX1St5XiB4UoXZfkd#*a5^8ZA4&L656@`q%S`wUgQXw* zh!B=JYR@E;Fr1?seDyL%kb&mxR2qs{e(|uNK7^ct;@5fb6lH1Wewm>5Mof6K+TcP; zXh*ObXIJ}^+AF1OZ+kx+ARs%uf|0N2pTTPz9iGP)!P}EU0=^{a zZkmVBI7#ug|2e+krR^Ip<4Q=0V<8Ln2Q#`*mrAHOHkFSbYrG0N(sHkk) zbH7)Zr~>N$5uAyF+Ql;|+0sfX-Pec3lkZaM!R6aOg9i%jKEz5%b%{Z!b%7-0FzHqB zYvu5_x|wE_dpPPS_W`!n>LVtd%;X?wvFsBO$a)Hc=y?=d$IPMl?updf<0~QW3OVMo zK!BX{bsYCY!4e;t+1{($TJ}#X1$V0&v5+R3J+T?%|0BKIysiLVCPg2tVIXLcC~(!| z7a$di_$~pBg%XY1#1Inq23Mdzip7|Xn8%0fW7_B35x4T6j3U$*8wy)|&;E&SIK0>o zt#ccMjMy}~RXSebv9qAu^k%jHIOZ$E_IFr(s8^4nQ;40%v|zoIb5MN_{D~vnU^s>S z*&Wy246Ch`+&u}m@VAQRLZEDKB~uy8JeRK`knVrF19ZPd&n#pl$7{YHcjyZFz9be1 zf3U({oeqlXPb?iyEGIN_c$DBc*F#{g=*in)KssiNg`lG{F-9nCF!@1aSyS*MGMX>N ztosDry=fJE%3WT$pK10J*;5(&Vy~1Vie;Dn*nbA83=*M}A2T5EhodW0mc}fQG&=7e zwfsz#u@3!F=?L_HlsYa4TkNLCUhZ{KVc!b`e!YlI(tZ3ZUxy&#`xi>WT0IW)eW4g# z=YMeHSx5|+#eY-<5pzFN8hqLp60y0aZjq-INtt}e-8ob`HBy7)bslelKhpC)MM|=% z0;ub<(Ck|(P|*f-Js$62qYNmaWKJjdj3x%4H;#vbt^+ZHMXlynz?miS^vIo1M4@=7Ui zooD1jJO>7WFSlOAUmvFGoDd?`D`ZqF={1H}=W&*bP$ykkYV^$9u5l-%%U=Y7(;raz zfGi6n0!&?8M4`bB18uRSsg@les4Ak8nqng=rd~?+<@z8F7n=!&g6Eb*G9rj<4ON_a z5lT#==>MnB1gJ{el@Uk0Z7#-@M#0O=RqA#K1ueaW%t*ywsf`$d4pVqmLY@65q@$xI zo${M*i>4zUjy&&rzB&0CW_jRxhyK?(tRnw`g5&92``*ue`Xp!l%jdg~i-6nS190zvPVX%?p;zl>q-~ ze0Y~6aFIJc(<^{EZ)%b)~YasndG3rw~zW zS9{?%E$?=>c^JcDu2P>}9JKvw=CP=Coz+AM@W<0#6c1K(tRYl$vW|SzfYkIIdujNhd1~GpYG{3Ga zW%tA(LLMm8{jaVaB1k|e^^-)cj}sblLw|p)vAw(Nq6x)K>vyYP_OOtCZf>K(C#Q)n zxo=1=?%|g>e4osoou`T;-onBHCv><=cK&%*YbAdA*nIy_Uv3h7_!6~V;%uzSHY*FG zMRVUK0}I);6%!*pQs|LNPq>7%v_$<1@OufM*B4!T{e2<$2|TMoRDGIl3Fr-avx-qn zg^!ob{``4cMn(qc<`R7R^yy(!fQ7lZ!K=LZEnTfoX;uxYs$Je1#&ym%lVzrk11mh9 zUOF|nPd-2LcA7=r2n@uy57ouhZ(DNl1v}p+cqP}aqUeP)rD!R-wx$$Qd3EJ3YT8Fn z$tEkkB0A+{jFYV9{gZ52MU|m2kfppi65TcFdIAmD%W@-Xm?P*&R9%<6(!+6|tp&N> zo%$nm@ki)~r=Xx8R$5rxT`}ZaQ(dub z2Q&ip_2KHO#d_BI&xC1g%r%nzSm^0ruWStD1npk5?;Dyg4mhB|gS(lARtAyqv746n zGV^^c7|;V%w#gsGhA)7hSC@WGv=>me$q4+Mg$m}IS?3ZLC8^v;#LRDW#G?7nn2Hle z!hKMm?u+uTB342~%yOyy@u3Tm0&md9I>XJN^>^LJH==k6vyJq#I|fqngFEw3zTe55 zNO8^#@?h-}v%?e?DyGhE3V&tm1wf1K&w*MMADx+0K6dnM3)xZhT*mt86w*w9Yn|sK zmy&LMey#$i#P81_04fZ*f}dR5-TFL;di1MISy;p;5_^IO05xmLYbmuMEQ>K&yEy~v@jzW~AK<)<_rDsLvka$ItX zGjV|+{rXR8oqgknZO{7yEIoazAnt%dz;#}i0Q?_qEo}3n2uA9$JBEwgI@Pe2Sam}$VmQTx%#w#CEjfF0S z1^}HSw@x1IX#kxE+l@QMz@bA8sxCA?lpBh|bumuO(cHKds1nrQ+V$w8(Rv z7h{Z)@xSt=7{+&{i2R>a)hpmU;-2HqsQ8r(ArV1{DA?dHGma*oD8#*sm`N75e?hxc zpwRsa;ue}Xev!rDVH1XPKBXu^bUm=*?&0hx}NWr-tD@Deo=z74a!+ca`i3{ejVV(kB|s4&21O(fEDhQy(R1 zbE8uj3NAP9uSrqbd5P1diIjTiVD<#(1)70aa9BAQ#P_YZ>8!gl zBesBU9VO*k{Ug&h^sKxcqbN!Yp=p!s76@Fv@}$e#31O@DM#;wRt*`sh$yJlfJ;&sd zxkpq;CaRexUh*+V@4y$?bSShoZ*6cuxwy7*kr?Db!=8w^SqV!)!i$Hkm)kBP;^MD7 zvf#a+4TJnMk7@Qw9&gGS|Cz!IX-r;`yz4T_4!=C_)#N0RgqtOrvWUOu4fdyqSX=y5 zL(!A%NzPkNPc(RZpVn*mg%r?Z;F0Ial{|veV32M5Qk*Nb7?s;ELVc5G4chiA7&qK6 z2XdK)fD;Ts*4gc$b6H#_Exz(0V5L1hKfU;tXFcYK4yWifEP0zy3}}X5p$uInuNm)93pOH|qy!;gKDggEqa?OfEiiLC$;?m2!#r2N2{tYa`ecJAJ9r6m%5 zB8XJk{p&1JG5^(Q+%lUIcON67Kv^Y@OVSa_+dVUSP zi=6S93=#k72K6w{Cm83amC`G-U5L zW_sH(7nFwX09}KuihQFH5t8@5%&==?^S!uUZ9xHn;TK5{87<=Fgk}nn6Cmi^)URv@ z84XIohkwy772HAA!oX0!=#-Fi|$ZaH~g)@0RfVuCv^X*Dm>ok zGyAPR287;YQBfvQZ*lds^0jW}dtTb*!ipVTRB{8)TK`aF2JZJV%#yyxOHu!zt10om zjAxUUlbW1OcEEzar`KUZQ2MD--*YrD^V!kLJ8w6yV3~t$oLOpdF*C^>%P00?t;BjH zV|mRVShPC7i>LdU`Th3E&)KN<@Fl=kFWLIG+1a1{spsm%S!YwL)7$elXFN?~lUn!a z8j!aTH}TQ&6@K~ZqyIT@NH~r{&|$(-Q`|4HSi9Tau5;#aiY?g38I`u`5PN{_JwMX4 zWqQJG6TWv9@Wz}TQBwYfpM`iqDvNY?d|u|~ILpyK3?K3T0Xad&zIDe>25P0&57Z_z zYH`));+A=MBK!7=E2X5!~C$B9p$89hVA3n^nK!I9=4ySuq7PU>Y*)Rxlx*crE zaM?%5IkuK51M%?dg?I!GpyQ`@LTLq?bt zt6*I?FQ(nUo1Hs%^4;%#m(kHt1_lQB!4G}_KsGz`ds!$HNTpJ&UcH*`C}0SZNk&FS z7%1df)6qdD>KK&l$Uq`Ne*5E?LY`l}>i5~wyAF0gHQ9cv{~*C<{}FrhC&J(?9@#&t z9!)VdP_vIc6uyRMoI4q)Risx^<5^0n6o)1z*l*Hgo)*qTkUlPKp9T;l5;)NW=l;N< z8G+i$Rblyu2dj+VRtIAY=E6&1%Ol~e1`Ybv!3iGPpPj6^NKOmXiiNP+lW4;DSkH>E z>UewQ797F(f*CY$;&pME8^%nz3Ct#&TB2Q}pinJU|qfg&)L}6X2J^ zj9Mwms7(}y*`9d{)*TP&^#5b;&4VR7uJg`c=3Q!gRrQvBuXijBkOaV8izU*gD9Ms- z$x>)6Yud6Dw!_OmBkvy1*z$}OF&@p>GjfdWiHVRcOLAB;1rxMIaT9lN0SS@-2m(Ym z8oj^X-&*@}x6GVBGViUbey`CRNRV=$h>q^Ax^?T;T{6G?&Ue0piDUGKAIDriKXVvb zr$`ZS?bXF8dJ|tF4)f*wOOF%v@^|lYl>Q`z}gi~oH)VDFTc$D-uFH}`q7UPh9OeQ8ROhM zk|ZGvLww(7W!{1+?RoTiJ-QgG2I`La3Cd-p=Mh~#i(XuzRjJczE zDvK6UPNSA2wvYe{;0?}nH`l4nIp4VHVuJcSqB>UqweN=1XO7hip!OOwYQy31Rr^p+ zQlFE1q8nSic9dCQwi~0iy}eDR^BQ~VeTT$7pmvAx#s~K?m)JpDMJXZ(k=_WC8P*5t zI!3LA6mZaABOIRrqpU_PXViq=!KkfvzQnWNHy}B%Yz3`JjQPmpm@hv$a~#T~q_N@y zANT-9NtBL@0X_`FsnqAUNTX&9;ow^U{w|)wJo803{qk$S?;AwVrLcSVs*PTkFl$#{ z+v>g7C`=r~hAZruGPupb)vg+~+=R8cXtg!_4q~odqTB7VXU`s%mzU{wyZqRX{TP)> zWyUxcKq-Y%iqHYEl?B_EQ~Ms>ZkMr3g+98^HfZrmC4>;fSI(0hd4wp6=q=gzlJ$!i ztNAbu+GSW;DZpkC1yFOgNVPf0GR)RGG@4MWpGUuzfbZJHNyX$Y zj9TN;Ssp!q3^lDI%^9_ZYkYq2ez+F$!?Jow;)l*{EEj9L!* za-g<2XEmqo!GJi9QB@0g9^3C)7I_|W^9m-7iSB!d@pw!&zl5mHk*r-JzIGADBgEUj zzxWQJ6njpdVd>l|cs`{mjhgx9vyf$oN8e$$odQ6z0BY7QB#vV`jv*mz5>5@kq~ zrXODe*Ks4t>;fof)Ra<%Y@pR@v9xp>)-K;gaSy27Y1A#vUC9zurVv31?c3AYI(-^3 z9FpMKRe`A?g`tcc%!H0->^Xd)r+n1x4VewBF>bV{W(SMB8fs$}s40`8omZRAAX9F@ zER_nNX0mKjjBp#NPop*ZtDi60U+os2GlcZ6_mJ8pUbr^m*PmJE_{Pn3az%>v4pX&3 zbcG{PUzLcD&NX!RBFJb3dvitq5Fh;Te6h$4jaIE?1V z#aB?j`VTO_@oUgrfQKIgC{mx`fSNG{P)m}8jvJpsXP#SV)TG4AEiM2Y8P4jiK<>Qkvt zuq~w;^H8%OJ4lCj%&65jK1glr6L<94+YQ^0T_C+Xx95m63#10g`poOb7~(jl+wI=a zt|qtV02n2jZagW%nNEG?1#`V#k0eRx^?LV!+P#7Z46{*kY9#H$X zN6z@(rG_$RqSn^#Tv)y%E?=;VrNuon#}`Erd&fQG)*9=F9%fH(8~rc-kxi9$S{CTI z$g3Hb;sMQOb7uFur9Q6*YI9+U(Ny2h=KMTqrHt>k)1HLYlY}b=$ozoucx;Ot4n08D z-e7d*7~X@AAj;L^JAB82Z$I=P1OeXZZ}Rgu_Z?u|}B`Fte%8DX?=p#RPf#7}LHn%Q2k> zxmn~uXsv$>jpB1ad55V_p$15kfW8G#=%Vj{WcMDU_U(^+(7wUoeXEfJoNL!!yT&Ev z%BpR`U7ee$`i!E8M>}f}gq(TTpW-tQf3G#RefEo(Zo5c(0+vw&mc~1Q+N&{Yxwl#f zjM9YneEB0h+*zm5XpmJVS}7ezaQQr?Ll2|eBv`3b2oBiqRf7(}8{azf9e$i(fGiC0 z-}9%Ld)vGD`VanDykGe@%s=-1tX$nFW@)K)(e7O;g>+1M&Zs3ykzxoUJ5EhAYH}K= zl`0r*NU`HvqK1}HOU!K>0?(dA2h?t5VYj`G`O1@+usq|raKWaRppJrexctQ{eJNV{|TCPz8kQf3X$Xtq{;~Kq>lD1vl zO&i|2^ye2=y9|;D<$Q0_L<1VcbwF);m3bS$CkJZ0O66tVQbuMs2RC_2@%9+CI{<1r zu^@4fa4K{O(* zSa!;2h5F(rKx@i}A0i${XssHY#N40wNe&!*5TO*+ z?KZ-gVj(89Mmz8#GF2?hbl?}qI&#OaJda%9WIPYy`4i8!a)mTA;MwOgQA5H|pw-WXSm;!2|F;5oS z|A*<|&V1C2Lh1;q24KGR`q^Bg1|Yqg+jF#{^K6b@VQVz;?3<)MXPNEk1<7s|n*CUz z4IF6XKuyt$ui$6<5NU%^+${#P`lG`dW4$J5Gq3i zAz8_myO(r|lpZ=u>BSiWNi8fBxF)wU*Qj}uhc*z}78B$eHQ`MEl8zO&W)q-h98l}W z8Yxtvdq`Z$PI#I{Z;knia*UcAyn=97pk^+e$FycOJ;oTwfm-**K<$ngwc7w{cT!9+ z4bE;As3~n2INzTBy<&oIf7}f)pNbr)^?JQsy~%DCrCuK|PB71X3G>vaFdzOs=y&Ni z7rD5(Mb@9#0W>b0qNGy9yML4*2vAzXdw&8hp0zzMr+^y3pt;0yJfu`A&AeWzR0{1* z9y@`Wg1&tY%UOzf;IPR?NgRb3O%G0H?`-u)cqO&(bTcsDD`1Su!9b$4dR$VvW^$;N2ogEbi%0B>vbB9Mgi0o7Z+!`7yMC*dqC|DLxd$nrAkeu z2&G)rr+x1&{bTsXpjArGr3jX5Rq}f%XVgMDc>qFb2$*QplxJ_)DC<~&l}>>N2h;|M z!V?*p0~pcRPEtZD?i+rXH_u_Bupf)qIz!&ooVo!TNStwO}-2@!Tra zjZOQENs0OB`vGWI8YI$34JJTs-|8!jOI3LEZAFSQ`<{1OP-4supw_B1ScnEYL_cNM z|1xJA!oiS+0}KUFQyC1}7-02{@5FrV3&j88AJY2t6A0f&{?b1o{9C_*@O&Z%P%D)R zQfV4L^^45^#4pSg1ORw(LROkoWUgUogeB^08)#|Ip(q$Ng_fR8Wh#Z|fLdO}s3%o-aR;pDC)QmwYjqJ7P zg%!5m{o{q^ZT=^Jf#qNLJC@CH>hIW{%W}Dl&|n+k4?`j06@>lGp2!BUSK8i@h@l8Jo{XyGmP3@0=04l zGj)ARgO@1xU);G}-vWf3toDhG>0r z&DZrIyOaCa%VyJsd2{k<(9>=ywl_13W5O<7y2P<#$8Kri5{y^bKex(aXw7|Z#;CQ% z=NP41l)QSusO9HszI-3bq>Qu4{4tKklvY~Q9$2H(-C3F4DHt_-Uyd^SteDYWMA{%V zJ$k(!^?Ds41oe8|`U;N6_kh~H*ga%(fpTg+(?k$r9Aj2>EBa%yN!zW^>Qe)v&`Htj zQ@a5^lm_2jjwFgPuH96u`&Of-lr3%Fd$3^Ce)9N-`D33SaAdW|^73+F(@`y3&{UE3 zyW^pKaKz%WooG%Sx9J(Msn9HgLErlR0Kxbo`rmS4e^rf|ybX<7FII#O%-%Hp z`6`Us?;qdh@BiWIt{?mQS4M1K=@4TW8GXx@pE*$5Hy7Ts=XgD$(5T%WqxP|neT?Uy zpHwBMX-cJ1VR?D^h7(hLZJm`@PR@X}YO75+8e;&xWz#j{p+}Ja!!IL#>Mzh29*O7S zr`D&i({58rV*+O!n=@)@nj(Jmr(tnVVHp9oFX(0Lvu-Z0&K%D)qgLD6#>*56__;Se zLg4$8XREaV`w#QTTi%NNU;h>ISAQMvo$tmtD+wcJ%k*+a&3GQ#lSQvjm{fdDGiuuN zIMZ&E20r6O3o@rupK&Tkv@D7VwC|%`1s_mUeyZJ4pP>V4rHb_u^Bq&AUqhgF9it|* z0W#}MJn-_k0i(833sBMSK<#FX+K#B!Y{lp^Pmw8Ommsop~vTx7h%`|G4dS62dj|b-e1c@4=bXw@s zmMRYcv+2)K)JFOm;VDMp*&X}K8MVE00iJ+FtsxMo;W9$iQH<&J+l3ZwTJO1M&z_t9 zzVFt!2h{E&k#as{!nA2zR{yKNq6oke-@gDmy&}t)DJ{#*8MS__II>^|?3<=PUq!QYxtFjuNOu_L zYHdgpgH!_&G!YuI{*KS@7CBHm)C?J>n(W)$i1@$xxC4#aU@+k9*|Rf%EbouGfB$|q zHa2EjBIQvU4Dm9>+=g{KmuX7l>Kbva#wZBLwj6`8vfqNWpZP^bqY;Vp@r-7BZH?7e zUdA(q<{@iVHRelS7 zU3;#V7`6Df-IN5oTNUd?n`5C?paZrvlXB1ZPK7d>dz5o4yfH`L4g=< zxu0Z^=`5p`TU-pCE^$0IX;IIL8w+U?6;j3*6=YZ}k| z)Qy~{GpE_IGuG8x#68mWG4j3xHua-2+jRK)-|q%-!z`nw|NXy2f8}#C+iGJNB#Jle zu>5RQweQ6Z zh_l;V*?*Trk+QlC-@J<#tq;~TSV+^9APCsEZy#Bf?KHmCnqVAZgrK&$g%>9*p1%Ob z&^>&B^{q|D*Dl)%!l^b(qtU2HXRTj2&()K!08n-ne|hRNy@By_x7$TaNw&0i2cwp! zKKDlnQYrfT_Ht%4NnllnE%phl3|@4H=#4jJ3y&ss5FV#l*yl<~x| zErw1Sk^{BTV5&c6XBMdCBK%j2oKY*e>*IM4C7RaIUgt}fF7f!|kMqe-ev*spCm5yM z9Bh0q72h&jk?IvdZJ~TWQUo)VpOLmU8q5Co5H^Qg*`Dnr(TNnzl8+FAS|}0XDs}%L z!{{QN&ME-0?(G0-l}fVh|A+tpAOJ~3K~#lmwaRN5#`PGQ6tpFBO`&$N(IMe0-IyE6Lrw!N{{$byC_bPr$2F=~MP+{d_D zSz`TupXVOgN2#^J{J;*rmD1$qYLpa76fs<`aIrp>3UV!OOwOpS?8i8uX4cjqGx&#= zCxYl*bZU9(NQGhHR<1^XyZYugr>ghlxfT*jo!|mb{xXYVEl(S04^oKkQAqGPuaA zGN(nrFac@-LlQSgN7oI~N*n&w(^omMHalqbuB$fcMc;XqpQpAz&L__P^4qbLtG?XiVq6LkxvroTix0HfmwnYK>fsMk6YV zcA<0m)JX={tV}>S@2nsQXoMk4{lU&u8&ef-haP|{t26I~LBJ0mzMnS;$++1dtyV~q zgpXdnNax^w#)}IKV>@w*{fD?wOfzbgN(Jed5ba4@1TdAu^5O*L`xtNX8w;IXmVppO z;eap(0i;F{XD;1A0Qow-Ri9Z&@3IIoCCG2Q=~pieuKGinSL^P24mV718-XRX^axA_P#iC2%RWn zzGCm6Mwk$ITh#pr8Aj(B5AARF!n6Qrd3TR*1)KK5Gb3Jj=I-o*UaPnV)b1kWR8^M@ zLOiEysJH=hBuC)KK#J$|DaxNm=oqv?3T7EKx359pt~%7%X;%uMR)S>*)P{YY-TMd# zLEH8Ht2?j}hLZugQPw1J%(=If_*(c`bY8vOolSj;?|C0yW1#jOXVWE-m6I zMSt%y(R`EPa7d!84MxkKwa^%iMtIJ4p>ypD73IFu+4KZKzz^&{$OEGYe|%l44GYwS z5Cr=VLAy11HjH7fHLVx1Fa{sZfVLXq~@*cW}4>HKpcl!_D@SNlt zHD5}~<#GYcjOSsz+0>^WClhN97oG!cdB0EXdB{u^m7kdwjMXeq%g=qiUcX^gp_UTe zs4=VyBHgzdEpGy(k{6#Ml-X@DYNOEzrBsnRQ%Z4VyIm9^Ni|uBUC*c$sn2MPQHo)r zsD&O6Zdf}a{f4(pj%`Ytmb2M1vt1d)`BQN59IWqbRVV6AYk=EnIiOZS=0NS%J-F@n z=@ELQQ2h+g6{_WJmimrKn=on`Az@{|&mMVH~t%n`18AJ7apA zi7q0ou(EiVcx#?^_R7p}%^5X-rJ5xcsyslFb?}WfnU`Tky36>~=7D=a?Jh!2`4WjB#B-wFlCY`1m>A(D7?A=2Q6@05 zg3xiH?b5!7!D`feA^53FPjTS+rx8X_irmc~ICdv2+jNXu(LB5NO#pPAN6|tmF&fg+ z1lTmnia3cGI{=bfh?pFBb=%;1h_~Bj&F1r;VD!`~rHAYI^FA~e-S*d|55Dqw%+(9% zEG={{Qf091Gzjq+u3Rj(?>QE%QS#VV4_WKmhi38o1sYec6MaWxVkh()HYAKEgQYR% zhCMlp)Tcigq{nKOImah8 zM+ZLr{rhM=a3B4BHhno542ZJ~({FPn)y2f7wI+%pe8--JTCtF`MspLOcG$gN8Vq)r zc@~vw^GjfK!RiE~F{Qzf&tJRBr?)oH$G?vGum6#U&%8)QPI*FIy#(`1h{}!AJGu12 zbBtOJ%(UlWq-Vz*$3Xc>Ocu@pwHZbYXI86JkupuRB+7&}!qsMmQJb5aE7}9?I4)0; zz3q*B>Mz$BWi66KWBd^n&z`U7gTG18+qf-_+UUBJ-^RuU7dJK$4(NN??v@tD7<=xW z0q^QdZTz8RF7!BX^(1}2%3#lgwVVcOxlXO9bk8f5J2xn{>UBYDQon}wO9)rIaJ!A! zB{7%90$(n@2}uk(p_9u$hO-*A;XqTZNtz2Dp0OadWO_&`8I4Bv+Kz{IeC~Gm(wDx( z@BjYq1CzApD6JBZr#=PspwCLsWqW&@m6a80wHlR5g<-mZFUyp?Iwf~4M@gGe(k1k2 zNFkXE_gRY#lh1d3q-qcZ0Wxi%!~i{I;ycZ#HOm$VhUpt|sA3pVW zxY~c7pZmF=I1A0LfOnMdRBT;G!|)UR$?NP>`oRj~qEtfQ4Ki76bui z<0{KhpY6e5rl24gjY+Ci5Q0=BoaA_6kuw+!P{TfCDZNsSZ+yeH7|Oc_1g;xNs4@=J zof6f7GvD(#>~u4t)#i78=XW^g8mI$oZ*Ma{KVS3+&Y5iART*APB1C%&{oDT<>PrK&{{J zlSsE19SsWqL@jHxj9MJoE-y|(a`AYZgQF62p=8hHlN_HvLT_qfj@1BPx<>MDpLzSN zydc9&jg1#NPt(hrh`ZK%X;4E;AEF@`eM?oJ{dlr?mA+z4H#noMX41<#1*0~0i<&5k z$g+&#KvSEOH0M1^zI|`LoU*dAQk*AR?!M}^+1S`%dwZKK%kUHX`q?d0Fn7L8DEl#W zFD4$gipt7nvq@~)%!PXqLg2}O(3|7zg_pT-`4lD3J`4451!WS(`85bf1lm4FCAk1% zGM*V@=the9vb~Pg;4;Hxk$xuu411UNLRRI_0|(iRUqYLde)8%)eYREf8Zl|i) z7*l8o$HUiOr*;ph-9=0TH9Uzg0u10|7%A7+)3-oP5HLz5T7>k^Ughx11Cq@vwsc-d zt5HiL4i7ER*>}84D1B`g$}*d*xR`}Ja_pMU;pFPf?B?8S;k$)oOo+Kp15{dlc#;(jgSi>p0;^SSk#ZwSPW zQ5)X@qo$4KOBeq&>%*6CynhF*qT6A@cH2*Md;2lQ*wklUmHaC1^10*sg$ozhyLYec zd6{Kst!IE5?QNFFJ+z_d11K^@5XYoWDzMslny+5@by{N!sD{HK^zEcue&8?{E}sS_ zpdfV5`#T?bBL|yxs~L5+I~AZ31SG~#jbfw_oIY(o-|bp9;=cRtn_BqT=g*IAVMFYB zy#1SB0;Q1u)Bj2A1?*qHI5|ISml3a$_Ow6H%IE!pQFDMxdK2SWtqW@obzO&q9jQ;> zvuRGp%%xf~N+v9y4knD+^V^PfbKs?U^>--sziIbJM(8xUqqH^g*%J2ql9C{ZIyat3 z`(4Zje;4!R&lbm*e@_n727>{KJO0aOP9fC*BLcE;1(MDr?Uff2tk^0=Mnjs~-AvURgw0)wA=RB+wC@9x=EaQoJ}Hx>ysRM6Y0oOH7m4W&1Mtj^_dG- ziucQYlPK$w=piNF0X0_*+8(d+%F8W8kWv!%cQh-O$?lEq_%}>!wS5lM0{=4Gqy4ly z_Lco+%3F^sbusAXtZjv1#R*>;D_=_>TFkC77Xc?8l>-&n=NFb;gGCoW@DMpZY7 zg!Om2JZKUG5I^=V))(K76b`7FkWqb+T0AD+TtFCDrOPO-GhfjpnS#J|Ey(ZHI70o? zU$W+#&X`mF%rD~o#lJ-Pp*jXJF^g@?6JGramCly|tXXcp%I>wuw(Wx^m4fXJm|uj4 zA0<71zM%Q1HER3k?4X_Xou1@+1JcoiQ5(+*vhgH!F%6)0)2NmFh*V5o9~p&s=F`x( z;3gm7ls2qQ6*O3&=Hp9SZFD1`cB=JBzH<53x!4);Pd>}$8NYy4yC(=S!X-rcYot7?d>t^>+9Tq|NTWfrsUdX)an_)0kzIH%dR3j z*Qo_ARh8B(Q0vB)p~-s`4h91P#}to`TtVFb4Ybw;P%F*b=VbpvlOP&Rns5WP#6Dl= zw894t*Bh*^uA-DGn6)EEjug+Dan;sA6qA)g%H=Y9I-7|1{5WF&VJRO&O;f+(9z8PgiIGks^a8^;J8Q5v0~ zG(2u!BQAj2IBk(Qwq{DFM=jGA&@ZRby%KxTckS4Rc& zn5e@uPYrDap95-QuLWO0HbNkH!xek)>;!5<>fv7Z?}0rZ?maJG=s89$%6e#{7^SXI ztFz6xCK&Cx3U1p_-MEcZICz<*)nB6@S?y&%xdyE*WW}SKl{i;0YTaVN+30Ta=GGMs z54P!chGZF3=RCraq!Lbm+W!6a_x5@{OnZ|TeyGgJx0Z{=O{>+yPc}Fk!KHndctH)2 zYDg$2DdeSERxoOfW*t!)Q}rj6(Ph7Zk|Xfqf>GO3c>}&IbE)$JCr;UgN5!-Ic<}IJ z_Ih2hi^q|?5h%fh^W3}@ou3k*| z*njQv=@0j4Z7LplYnfV@7KZS*66sVLemHpjvn%(2+FgX0s~37Pq`)@;T7ah&BS-7Y zKw%tEOVW_+stx=4B4$7CLdWtnUK27_ddW<`@|&13!p{Dtfn@szPfi~wtwp}k$CV@|3n zr=2)j1&avx5Qs(tGmI#?3C&!)RGfpXy@h)ADf?Tq470rfODl*YkMPo-pWwMK*nzPh zD|{)a_>#SKs}J#}MAzPQ`g1%YspS2(1fzJ80_sgYY{OkrpUt@uv19%MQDS5ivg*N> zovf!BxhuWNclHxSz3i%uO5P2tKF4VXgW~H~FLQij%-65nxn0I9o8zLW;tx*Drj}1_ zNBsWFExx#F!C9h)80>Gq6Gm;A+Ew)R#TC<$Givc&`dJl>+O4KO9fKwo>;OL8-lE&> za(VPsPF(p6t(HAEowiNQUc9h{F@}|um7>K``1o&9|IJ^sm1{u&ofgexOo(A~)80R! zo8S|t^U70HTVu<74F&_kXaLeb|sifVd@!1&5lhsN_|cj6HIH=ln}&vI`!$g5(r5uok47p6q+#KiO3_y zgr>xN<)GE5Y5UAH9Z*Zkf?#B;0e59RTK3sBy8yMpppQ^P27>{5NQA%uNHesSYgCLAJcb-pE-2 zjZ({`(ca?HpIo8wz-Q2IOuBFFsT*q4!qF+btVQe?-f`N7)-}99&`HazWpfx~t61{| zXKoKwQW@=*{ddjT4wRi2+CnvB@7}#gDJ@WIZE)^wA(tPXGCS^cDRmC8<`!9BSAA64 zB9xQyVtLLK)t1$UKuCkHCX1Dl*91RC1R0*Oiy==2?3;TlYr_{%q1BvKeFxMIS)g{s z&3juzdv2Gie!+5Yj}Fl5w{9^Fwm-+I?N8C|STOz8@2&6y z@2zp)eek^>`P;vCNp+2PF2Zz z2#sA&ypd(pj3)`Dp|*aGwelio4>b{Ubp|g#OIg)WN#0~RWKJf8C4tZ|7dQ=yEVC6h z2!?&ip5l#Og-DLLrZkO`A{K_ncmw;=6+BJh2W1Z`Aqnfmk>-Ohe~GV*ZOfG`+=o;n zFd9*>pavO!bPZlR!VT49j{bG>s#)^t> zZ~H4<$1G^WFwuoRtx@uB$drv!J1B1qtv2GtwVUgyOq!xU`9aL7<1?OcPoD2m_60xr z#(BPSsn1$}at=Oq#!lodcdtwXto#^mCvu<`t6hMaG5AqLG|ri@yJXZ{3hS1O335@k z18SklIQkd=9DXH+AmWK9-p~2-=L=7xjSV{?E-o$-h9U1ha1d|(GK>Z+#zU)7Yi}cf zWfe~twc(ho6c+Xo1yCEE!Wd(L+OUV;T*Q|W8K$(_+XXPF*6UUVFdkZFX3kDhLX8NA zeXb1$4AO+nIOfrnz4*S*@#Du)oi0DR?*O&!7NT6STD)G{YQ>U-tX!t<>}0Y^6{Qpx zYKykF=KFsW^O66I_`x5+9DWe>kN*bd`KNcBLZdaN*TD#@QL|kMW~)B!vD(z9@;pr5 zo6^NjjhaMi&9SSt-)T_siPaQPi^sNrzzfKPYhs=xg(l2PHo;^>x=%@zsh3w6r)@HG z9Z)L?Lf6l7x5`hm?J9Sj5pm9<2_e`cF0p6zUyzMkw4?>#flA57Ewnd>^5>DlPzeTH zy_jGIIiqHQnxBnmnUw1I4BdIfk1So}A&%i??INvK3Fh%-iTFC8hF|9NgFjOMwXtK; z^1o@fIDFvU{P*hTVW|q={~jJZ^I_&**&Q(Cz;+y2rgxOJ5UzbxCo7W~k2Lj>X_tPY zkVa8W?6Xr+DV4#HdXQ0dKrImVy~V1cR;y93*Xj5B42DLqKP6L@8$ zDKm&S>2|wp4DCMm)5QKsU z9xk!6FF^Er_{orRYeFzwA3`Y_SFRTN!y*kYVD#U*M(ya)qlb?kJ^Hz$M~}XI^yty! zM~@!;ySr@j!$*%E?H)aP^ac0lS6+`}y=yrwJF2kr{&uR}g4~9wX;M^e{u3 z1V3JbE0an(HEOf+SN4+{%}ZB$d?`2xOMBt|HxQQ=Nwr6kX!=eA1|WstP$Og+#d%>@ zPF~()4~k-fh44PAUCsQLzlnMBBftcx?Wy{^rao(>38Pk*OH+)R18R|CUp-*48gRt{ zqrC5EJ*iJ&d$W9(Ftc`ByYj%2j%x^rmj(YJK%zyb0=wJI~^e_Jn%$XC! zndZwE`n>CKou7O25>mj2UddPFiZ5NT17te~YOdk2z`Om_r%s4QcGVj1QixD#cUXSq zG%|5o$X&ETr%s*XLm&Fku1^0mU!^7@+N$5_S7oG z(gJhgX_|`_KK}8KGjyrycB_Syl18J!egi-K;1M#HemXR0S(eS|gO;NKTU%SGEJIv6 z&p5KrOq>mfvMzJs3Q^W&m~J|t*2Qlw5fC5*Yby+B9q4rl{XTxZ;Aj%|9d< zM_e6^NYj)v?G^_rRUW+meomb_#ZNqV1R+E*3DC6tbNx86{$%A6RqZB-uuPI9q@K_4 z@DXeO@W!_w{_@{KT)PC@8<^8Ob%y4dFQEVN-#`dSty*R8-n|HAdkGsMFmjUm^yAp- z(~PFt`Yh5_NbY9O0T4QIeR*yI)O0+C3ulPq7%hD=r#Fk^80mnDK;tFb40Vr?kXmUE zqqIea9na>3hmRmF3Bts60Ei7C@g;q^;aH(lKh;h*Nq@Wi)0ur#Vou3mPxkDvVqG&whqbhv?x+ zDjf?p3=q#9;PgX(o?tz z2Hy$eqtxE}FT<^=+$0QxJflwz@oy4Z1GycR}z&jUvRBIBWQ+!{MX~BG1 zA<#4$4bHD^a}hh1q*ZH)(=>!PW6f{xn8XU z3((DIh)4qdyAKEa>_=;mPT0Erb8iPDFwGOFI3V=riudNiWxR62i4!MSAEbB!njsuN z|1>1?pan@fpjNd&Elliw_jekU{YG((z%FJ>-UO%>F~}>dZ)|>EX?K|0+@{=h84Ndu z)f@7jMh)ww|35zQ#1oGJKM4HIqeqWEw##-;J@Ld7?{I(q*|+x)ZxW4p@B;*bQ2A(I zf=Y>q@MD9}h8VZ{&>?XK($Us>)OY$3i8t;N4qKEDggm^QP+2+%K((g|;l2nETZYbmC7E>@vo!5_C=D=rXofI3)FI< zKET`Neab>`MO-E#!o*W4#PN`%B579d9=iKL;C6D_TDn;sZy3is2dpDqFj3IS{oK)M?LP>j^ajCnzf7IjO+~Q*= z+r0dZ6Cfnwul`-=wlUv)mS?XF7^a#ZInrdQ8t_98H2KWgF2^@UJhj><$_$4ZVbS|8 zb`_tE(gDWT_4|9Z2*#@{pZ`_T?pNG7x@vv=roxY71TgQsy5dOBfKOu#U;XM=xp3hE z&p-bHFTXgxVPokyHZ*#xy!GYZ;{1iLa_rbKo_OL3o__l20s!Q(v9ZAiKJWofzF>QG z##K9juQckk$E(G5Z+^!^WLZYb6$SJMZ59?5kWzC0IOgKji^ZgVXxO&QP!80(n<%B& zICGZHCmv_>Bfkqw*t8?_KWvYs{|ZMgpQhB=B3wC0XzYa09$aHI8d0rQ5kf$vO0c!* z7&T|jF&-k2tVLr+qYRPaeLXcdYE?O^Zt?&qHRZRVX*+F})7H)2g1n2!3Sm zp*Wb>jdZ^Mhv6MRfc&LjMg#%osw*zo+@;cO^^Dc}>5K>OyU(5C0x1=NeM|(o?_HuL#g)-%G;kL9G$es(V9pf*jg_b&gJE0 zJY&bShCia83)np3LvAEi{eLR~B7zBPchEo~O0KlTd)-@qi)QdhhoC4m$ylrua} zAlJ?Vu+gs|CCIcxP^mHK_c>O#_qHx;nt^0#X^DB~0a}%F$f(2*cnuO0vo254{c4Bw zg-zy55Y8Ikt1({tHm@>curO#xP2|JbT50tRUD4 z)JjojcdHF!sBT#hC!O6%zSvw?^cf9x(cu2|iq{pWJ@Ld7*PeLdiRazF`@jj{;Oluz zw;Q6~gi_g-?`t0&`j{*uV#uHR-9B&oVnpN^HLXJifbTRslhmh&*s?6ukG#3X!G@1F zXSFQlK|B+(w*YGCEU<{f z2HFfLg%$u@P$}ApL8ThtIqRl0suDyO5IvVl+J8TKG~$`K%fM?h@4>{?&De7oV@P5{ zC9v(Igeg<%YTmBjN&Uee!*gx4daAyg`Q_ zf5={=-O``iqm*x48&V4-gLCIdm-Zsx{U>35iSD&^{>$-gma0B)S+OVg7v8pqJ=K7J z`k9M7{?Zor&zJaPhvsOF?Q!I(&xP{+yE1BCvV~A%Dy=qi>v=!DjQYk0(O!*KVA-n3 zBsA9k^GpSAt`D5|VuY|wn5Qw-rAwFSbUIWj70$i%5^o-^(%YN`dwIl>X5YrM9KQMm z^u|lP{q1k((MKQUx#yna$tNe>D%MWD!WVw$w{6;`ZUNe*b`K>4i2`~Y zfyX{=`1wa3W1tnot#24=2$MF8<|AKJZD;!W` z7ocX`;>n~5!riw*=>nkTRi84;@Ux87Yl6smy-9a(Nhy(1lBFr7!AsOymc_2qKpR7~ z;Q@jy&Cu}}vwE61j!E#z@-$|W;EmQ%gA}6;$2)~nNt>}yMn zK8<;gc&h{T20}_ySS8dWN?99E!FW!B@6xvER^gi#Muu>h5Ss*y1EfH`mhte?aniTc zc;%H>*l15O4a0OBQ;n$8V4E$fr3OL2kC>}-?rlE}7tUbvHhad{Oa&0t-^sv9WumN& zaAvhDuKD;{uS{8O^H^<@l~7GFYW8_ zdX=Seg=WLZe9=u_f?Z;Qy|BYiLBxvWklXde{ zf}?rECRfvE-7&&t9icrtc}g|JFXzP&iv;6Uf{Oj#aIbZPZgver*>h9+bH|gOJ0$( zG+9RS`@d)V*84*uBSDoZZw*;A_5rWX*%etQ`v2K`?;y#t^SbkQ@AEHxmN(ti)xLXr zOpnD(4>2GHfMx)C0!gZLAbpCyGuK@7h?rB%;s;bLZR{G56-`zjn>uLZbrCkNdiP+%%r=qf6<-3{hnsd*0zVjtiY>mA& zhrj*WZxDR)Uw?-gFJyo%N)`ThoxXMfk`lOPiwCC*OqC4YTK|(k?VC3{qzZoF={@Wl zuJg7r&ab?@&P__(zO}+uKjNoOj-%x!rDFwy|NWUcetQ2J|KaE-`S`&eHAg2-75y;9 zk2?r7%GL~V>Jue*vqW`&nfmR2PHpwqDQs_3$m?6WtryUvEqY@=NI$P;5TTHROB9-S zLDL<3lyNJdSjw~azD-|t{``5$a~-DNeS`g_B2#XR>}K~@O7ZDWf0{FA&akw!gzu#| zUiO^FgMh~$f1LNd?|nS_=%ZY@a)pJ31-7@h34Z;57PYeb6;C{hnZ)QfGbV<)gH*#Iv*Tmj zUoArtWBWst@6id?(4~cYYm+Cx5OU!2=U6S&=zaL(2u)|w$;>r}QS&>1nlvrqPA(QN zl@W=LjEgAYA&iJ#uZIGfW20;xJl@h)5O^n_G zGReNzk}44qQ!Hm-%=ZFfP4aeqfq42Ae5H|y%(y!YL$s((+zLrrA-&#pq8vDFE@hxO zwO6rn2F?t)q44V{;%&uZ>?FyZPS(fgr+RMagNI@6~W5~rvzlgtQ z0`=M7MEOI?Ya7&Vd5Cy|f4eltkQAXbV*0y`(MYL^rDDEyYZ8wl|Kt9bn6M(;IQwp$ z+=SPa38X1TGaFswc-qeKr$W`*kOS3 zLabrHxo(?pexD%jx9CS(=xHS@jkbO}ix@;9;eCg*{(1>cv&G1bMVxMrQ9lcm$K z{M7&Rm;A-I)2B}tflmYf-I+6IzFY15CNOvA%$fe_)2BZO{2p-V&v>7|0sIYsGiT21 zrV{@}HC@*kd*VF+s575JR|cBtp!yNRes;pu5|HQC;8Avg+MXP(r`~|kN(Eyuq`9?% zF=pcn0|N)?7C~y50}r?A03O1$P_f7L{_oBL@Aw-;$HTB z?Hi~dMN@s2#4RRs)+%rWdQWAn0GZz6R>h|dOsN{{v~X)9a2rZC#D(lWk}~!y@cQ3Ee8Foqf2k4Z#%*O&_ zWCVm@N#<_nSF=rJ>A1y#^ByCXftN@|iZ)}FB7>zKqobpQm)>IaTi=8yJ_w)q614*< z+%tu2_|#{z(1Sfak0@90CihK^F_S$iLc_9 zs_bts^ZC|zn@e2gl^=-QXqKbmFuJyW^B*R^|rEsB1~L};^YRvEAFg)MMOf~q?T z!46{S6JQk4!X0#f2ufqrXdznPMAaXpxc6~}i~j`ItYyyOg*nRMO`QB$BxlcokkqY+ z$w`N9r_YV+*O{EIGBq{Dty{M^@RXnzuk!Htvsr(=A)}**SY2IZb#)cfG&%SD8L9=1 z{k<$aGMg5t*CwzXjc1>Imi6^@o_p>&#{E9u-^;+{*u*&gNSz*_?P-!Y5;FNLte|0U5dx=r+@Zke(I?w z(B6EDJ4M5)k%yOVabTl@`sDvawP17UF^d<5C-|{3U>RnsM-c;qm>$7k!0hZSqobpA zEC)ZIN0phqSj7~PmQk`xt$24i-M@c7wr#^~79bEiO;}!`UMSF;nqe^PWBNY5eDgHJ zkVFWCTcEX^jsNlS@$Xo}wl`L??9DKw9<8B`%z`z7ibA&X(7p&gjJU;U^*FkeVAE~IY4Iw$! zeV%1y@|-?Parz-1?01Mn8Qrj0T@5%gngg{$g0DFYbqga;7AjV!QR_DSuuRqTIBPouF>$Q z+aUQ&0i`w$a?g62YRw%x{ zm%nxl{7+m)l$(rvpw2|P4>n;&b{GT}69=Y9M!VDpM|uC#A7JA3*GYkPy~OBfop$pY z%tEKUpK8u-e(6SoZqVgq`Vgao0x!jaQrX!Z)BQAN0BC7MWiVuHdXmI-NmGToahdL~ z{sM7x8(pQO&wh;2_dJ0h#&%L>_KX4pq;MG>y_@Czm)+iRpmzH7X@d-;ed5fSGoShH zzkfS$J9FmDmrkEP{cERBpB_DP=1k*Hx%M}J-^hPejmDq2rCCnQL42(D=6&kW=Vz}AjcuZ7S7cDNYpXDS1zrBoQ=c4y{s#a!oQU1;Yh=)6OKc(i8M=iXG$GmE*dR3wdVU7> z%&P-}Nt2B^jRWUl=jtUIeVzH5g-incIAwiv3+?P{ghmO+Z;-AvDOfh*sKe)z6Z`@d zh753PW8AusiLmt~U^WcURf@l}$=XhKzi7~*;Y&o6ZN!V~+Zf^9tfrNX45+Da9_*KB ztlZ{EP3N;w%A0S$#=E>v0&v56n$^w>-+$y9zxVt9kW)%wP89gy$O7G|ZHCJYoNK=b zLT51YK0;?d#m1L0{ag5@2T1fPh1PSF3t9G8b$N-h8t1J-k=V-?zP0UJB#xlA^6SVX zq&54q*xg%H`nhQQ;;)kG2F>yZ=&diHNU2)$48s;}TuR;|jqPnV*4KFe!Ah^kY^g*| z>(KLV@~!7z&c3T$w@=GU`q|&R-2s-Bo0IxO((Mg6 zd=EzNCi>E^5jGCenf|Xc?OK}Vp;#Xgbm%X?NXah|n+C0X&P^uqZV06t>Ug~^a3S#; zH1lh;Bq25}QW+ECqXRkykr>cxH!;$Hj&1R_{RfCdfdtJCnJSsYEuB`og{(#-sSV}? z9sL?Y8IXV;JKTX2W=;rNNCGn^DOE6%A)WS+b}5^G>w1lQdY9(c6O%;VY+PqQ&{mU^`@ps{5zuGc+0vLd`)&119o} zXx*Hqp;mdNGsQFd4O*RbWFpWCRSaDxAR>%PBxVytt}omcJT$HGzC#|rEklC7hAB%7 zMV}F?KupY#4$CYss~yLV8lb_(*eWZPX(rCTPJMJgVetyxevKciP4e%lGc99UfHyZs zHyYw?$BYPv!_z0QoA8xjKTp>eSkDz( zQsM9Hh1pwl?J7w#`~0i^2K(H33_~aO6>}p{?cZglit91uX`Qp`D=3lT$PwbW%l7%R zxJuy_tE`o-a00lxahsES4$_*RWAyq>Ts`4Edo}i29$Uj0h8r}V`wGdy4Ez+rZ7fwK z8o7=pVg98pv||l4DbWDUZ?nA-P~5tTt^%6X!;Gvq!GdpITjas%lo2(G9x385CP{ zM$rdmg}M2294#-Qs~Yo~hE$Tx;T6`D;9JdaaMAQY(?~lWX{wR*VpO?~Rw$vQq_v%3 z$ZQas%`UQDr*QZrNjGA9F{DbBkdSFrK}HcGNic(mhS)|1Q}l8ZO)WPK4b^}`J3(xP zaI~0}ks=dOcY~OUtx+;HD!Px>dKI;Jh0rKbZ)Hr5-HIV`=v41PAh;k!e60w0Zr77mz~GaWc?7+`7n4?^VV|_wd0FevpZY2~OO5fQejr0ZmQ!QRL;_D-%9 z+M9h|D~yTT4bZbqCtALM7q=*F1u(OZvD$qY(FEnAODuFVM$=PIktyYU5XE2iJ+s@H> z*i(D(eri9m2O&}nZxK|y8}<|rJchItAdztk(eGu>_ILV_voHNfBUTbE?dE;zW{g{! zpeGT@_!LH2&{22)H^fJ&6eAw8e)t@r^0@cKte?8$v%hW;3yn1DAtWTSOgw&!HZgjx zYOsWfP}yg406!AgLJ}D%qEg3DK&hkpCMV9-wqMFlNt)4%j4M%ZSUnj~>TC|&euz|jb?`wwF+&T6wp~W z)=_CfJzpnv8RHiv>Gmc{2;ed=^t1J8M$NL<&@q(31NwW>qGKopOz~KXwX=x}2?FI@A?}ozyL=1dH-ByS-WB0znbs(3=73i9FxE~s-%9u# zDI}Pf2%xCN7=}RieJ0+%jO=`yozb9;9)wf|9sEv6dg(02{86r*_;I$!AI4w0iq`Gy zW{wIesAxTct#=l3ywe)BQ^0=(y!YGeH$Hg!^y!~DefspzDC})8~L+0)G6= znKSQvmDKNQN(fBd#ny_bB%;ty@C}WPJr=`pi{Z5k7(#=bq87OoLC{zu-Ofb9J!g)< zPq8TA3!k)HLk&8l{Vu|^5V?*(>x`26RrGSUA=aHt%hEGzh`|v5_3Q-k_Tx6Pm`$c( zt4k6F2;1i7p;I)gnf@m1_t< zW#UFa$J7u)(sjz{{Qz=pN7US+Sywpo$4E7U{-w;edcfESm}JHne8xW zL1J>fmx1h^?k29G@lW4v^F(ElFK(XTufO^zpV~OV4~|{I8(g5glb{SqF!x?!Yl>iB z7CH=rI^D^S(wY1y{rU&knfp8VmB&z*&Z5q}PVbV&sQ(sEmNWYgr4$n#2m945Lo)5W zj5WMUZ|nz%+)N2@CyNl^rL#Mnd;1#CKmR;~yoTS#&FmQs@`d*8*S?0#bzX@(LCcRQ zr!l&8xaa75IC$_NKl-CT%44VQBT*r`GAN0R?WtNOU;gr!@jQ>CM~@=n%*j0I#t0Ot z(6GZcVdFbJ-zu+v3n4Y+M}9hcVP!G9zbPr%(@3X`^aCudjPaE}AgnmN^=uw44?S+( zx`u5R84j|AIE;MEG^XQb%!)U25G@J0J^LQE2QErCu%$uNZJ>`>$dW;_aTAonS#Kfs z%%e#IYwaoGtv&cjmz1o&Wyu~xv$8Six-N$ABfrhsCedPE+0M+K*Vy55@m1`p?LQr; zX_!n^F9T3be$I@aVGm^!`9aij_{{FU8A z0&(&&q$@yaXq^>A9Fg8hKme3~SD%(Bg)4JIYZW0Hr}z(7WS}> z=jcd$&JJAqX~;}E%a|~kReLcoaMI~~jx?BXvc5g)lu7!zbx&mq(v+2UlUNA!ZU)>&8%URVi-d;ufIIp%=-M&o~PjQIjoH~LP4+Q z;EieM1)pMc9PPc|hjro+dNFvhq?Gu()n)4Cn0|2aKvi!h<=5n-ywbIalCz)vB)VoHbdxlSD0EZ$j!7D$xWc^v z03ZNKL_t((3`Y!#%N=Z~VXAE5(`sX+K-64^JO#U1suQo?#!>|W5uhqHlB+!moy(-A znF*GagKHRccb>$W`68MyD404nF+FP(f|&5^BI>ax@k{5aaEKki?I=VTAT0-_cj?=C z$i;QgM1m)LCiODVLwU2r_IZw*Q+yACwu)IAPjMEP=o=OA6WrN)7L)@QXo@Bp_6jh2uacC z$Zg(e4)~#wDs!fyY-z;ncMClKqm|E4+E_$ILlhAGgTK#H`+t{ibdDtbduX}qx)-+* zafRi6kw<~D`671P;c&b_nwq@(w#UKhDud%cLRapmz4|XWzMT+GRB-PwG4;IUN^=|L z4-*vc{my~3kd=4}u52Nm80R0(U!w8Z3kXeP{Ly_F0cc7RdDj_^d=S6#*zUg?et@SG zju6aGPQvsQZ@u*vYRo_hxcJ5j<4?iag#{j-hT+Z)9yszoAPa>xf+G9tl9}=SsN^mn z=>{2NvqwCH5Q11i%`ULA@)ovjGdDMf9xMWoxcB?-&XWso!2OTF z+yT(^T@%yDP?XF`+_H|^Y*VxM!p0j6ADm}>YLy3oIRPJk2pRfY$aI5L52)F5Xfrx0{U(+(WPjb}EurIxyb_5!kJM@mNYUaDtgIGl;yXyI zhbv)csAJ`(t6{B#xzT2r`{&KNhq*bpKx-|gqZRmDdtc|p)*hCuD%R*-9jvDk);lhH z>lw@KSQe?*q#tN3Zgn|a6vSqhXqDtv4>&1WI;2+V(qHUR^Af_c$*dV71A){BpmdIS z`?(n0g2@?N>oO?NQ-xJA7;I#r^rBs&86Tk=Wk9Xob!a9fRJlbVtM z#n|{M%F4trM(GwiRDv~P?+wfXEDp=8gd5D*dstY$&i?UijkYOKD=S7!i7BBMX*^^@ z(##F2N2Vw=mk7KjLsE<=BAFT|4fpYPA34tTmCH2#;}!mSZIKmEap9fz!61HB!bsP#kg>P9*PDhx- z0)ElJ4Kp+3jg1W^`a9I~Br~lS5*Q_3nfnl@HeN!0;|ni7yts53wEB=V^T zFt1+Y`gTA8lL^OTm>a*>Mkx+&x+y)|Ld80MQ6n7=u%*OGvW?8FWueEk(;(ekMi?fY`X;d-;V73-`lOYS%tm3S zjW8UdBo{HeC5ElX5wV5Lv!`k@Vc=#95=n~o%#X1@J%VG`NexM{lS1I3Sp|f)ofRtv z^n%1AB}6fvYitJHlRR7f`TSsNS2MRSQTc(7Ab%k{543M(HBBvjffsN8GRYk?(zJ!CZ2r#NYdAoRq{hR_8CA?XQ|P3Jv)=}TYY{Q2`FNkY_YadW4EpC;V1{{W{> zonm5Qg4>%{NduEd?|&AN>a@KsK`eOJ{rfnwcMl5a_>xxYQqVLWoG_V7chGXSts7)n zag))505r9PvS}cI!-o&^`o_QJTJ!7K(CcQ5vHbYY0-!CNM|pX@(kO(TEyT=RR@I|6 zk`>$rLJNxKByMRIBb1DG4nW|eCZ`z`v%=!@nxJA?JT^7KT>DLuVV^M1`0O|Z!Zf8| znYd<*3exX&zNb;)OeRTgi0hq@yefgwg8E4AfpIM8Bp_wO?<7v zHW60T#d16z-4~IhK2qp-gS*>;tu0bLBsjZ%imUM{lJzZOYYNl5iP~I)$r?!k4!Ls@va85ei0H9{huyLJQL z&E|N;o+W*Qf3 zfI9-$8(DRM$%*4Q!lLD;>`UkQiTw|9y7n}qBAe%pK*PWx5DEVJ5}MH1!!#q-{kVlZ zfn#`qR0)bf$nA863Kd@8Ji=T$&yX%c2o8;g9CmeL1>fALVoQkH**Jz@`Lqv1b9f#`KH&C%6@?rYr0Y|SyXwDe6DB@ZgQb>kD1jPyo33?dgmgmsI2NB)u z9;+jn*+RS>AcGF0t%x*98Ly0!c01@bhmx&0+&loKX$n!ma5!XTWrbcCV(OAO5xD&U zt>P?>QKs|odr>d_Arp%+qTXVpdYftMagHB70cncv`LrX+wr^raA#^i{NP?o%i6@adNYm3~Az^w4Yvs}?gbv+#^vRqW(oZz8=@W~hMQh;}VH z22?n&XPS&wF*8Rcub0T^1oS!tcX;W%@;L#GrX8=H*zEsS;xG(qX& z3fFH5;+=r0Vdm;mKK#7^I31r_!Jwyybd==i^xH%#;eJ1CsJDBh?NAHHPaEn=Gh#aV;U-%u8|MkCTsre$mfA(*}+A`vMKLk?J zy1avV{>|(Zm!?=-LrPDMQxKAlEb*Do{vOw_U+1~!p5s6KyMISPK;k;cLIFLB(6lVu zvRKm?bNBMBo${{qDGnW*L%37uMw{4ck{AZQ5AB>CdaTmqxYee#nQh9wU=2gMl>7_? zZyF7BDR}(BCwSM@}_8uUxYZ!hT_02D4pM7&JvyPa^trCl6 zNK?|HOJoU3Llb2e&;yVcv&~1FnFlHG3N#!*U%Ln;$e-KMOicv7c=@ z4jMb2j+t*kQQ4v;JW{KLjbyZ#tv5~4$DQ$r7jYuUPQs18ZGgI=mJNWf4d_$xPYwij7{-~;^ReSa6<5!AaOoxP7US)JJhi`#LD z)a=k>h#rL4RyIeYfg+YB;m#@gmP@oYq%<)@Wo?7v!c86+uM@Sm39Fg(peStya+G4^ zGZ+k5U)!J;C0Lq8V<4#w+H4o5C<~2CkN+^#>KuQei=_=nmCoabe*#_CF?=5Z7`nx# zm+cjI?0B?lm5R}4z5Xa94odc^Nr#T?6G$Ut)C_^)`$U#W>J3RNRYJRlHfS*%D!zWp zCfr!2GkX#**U{ETG+NUSp`wsIjaM;^6u$O3tm%2=!zU5)KNDfXiQ|KBFD$?n?c`24)h?SeP8dibSwbA=MlIhn91n*;6|IUeHEo^ob?aV zQhmluh5hiODBmM}y~l_ia%1uLKq=mS%%WKz=fvi#sAPl=9rUzJ;IDDKvy2`>qqvWk zdNa&v4!sH(odzeHnMaoK^n=vG9_8y7=wFj~w+AR)$K7`nVL6n$LrS)WXCw>=_^McB zM{aRnaezo;wDuP9zP%X;DpwI>liXYgxH$14?2Rt>ygHzMRi|L=W2^fTcK;l`RI%AR z#815a5ghk0snNuL^#Y_RW6yky#{E+e2E=R2jBXF8G%_Y(r+(KXP6z?@F`|p-X?*p0 z)SG_{i`UqG;|$f?+4J>}ouoT6OZvxOBfjxA+Xs%5*gD~@n-m|JL--n7-3AA3hkx+B zPxJ9-o@H)i6irBGj~yd1O|&R`mgRC8tCTP=j$mskkEAEDJV?!PUK?Cu*jpwtB)*s8 z84isD^ZZdT!>VUv42P(Nx9N6oQPGMVbQb6_U|VEj{ex47Fr1V`MeM{K2F;M51)^vn zbOW!x4?FCjzW({l-|LQ+YI2r^OFpZ?IKElMGm6Bfpt#m06gE;Na61D@#%z`hGbj`> z`YpDG1K7^qC+>HU7hj{}REg6VFNtw94TgOgoly2N6+T#UaskWE7L6A(~qx8gxjUMaD}R3ng6*O{C0LQw-hUkn>H7 z&99;R>kKb<`NrZ8Gr1@0f5YBKpv@piFv~I8Yz;lyp%+yULLybRiN~ELos~D~H5X8E zZnF@;gX7sfZ+7zvByor3^m4W%D0n-HiOe3!oi%}olF=e8v=C7lQ!5e#0f}efT3ynY zzJXy{RPC5>n5}nuF$+siI+@1O#3c5z1;iGTMp8t?xwzfc8ORi6c~pdsZjRv^BWxBz zrlrlmt8%{`Fj*?orON>(`SIi@Ip}|k5xYbuS!cyLi!d!@n6jpKG2={=*GOW5{y_?j z0HntB#4#F!F@)vpE@`@(9iR*B;|Xy;N(mOl4aN%d{L#052U|!Sl>xQpU_dNPx{0Ek zr>`E;pTMQS7Z<1ijzH8>E!BFCkEDk3FzkL%24ovr#^VUaoWe!oOnT z)BhON-l6c&yYVY`D=oYe?XQ5^pI0kn3uD)~D?sn+1w^9L$un@IKiD<9HS7Xu!N%=$ zp&ww>#xaCWrJOJ7!b6Nt?Y@^VZ94rvijLAJb zHbTKU?KK8vNwFUygFe!B(3Q^NnOjVJ@tkVfBD#B%S@?MhZd1{w0FuM@3 z=bBGYduDxm|V(EZjY%O=xT;WutsMqA`uCT-D&QfouXooql^~X<<~O5yOBxW-kY&x zt*r&71}>8u8B@7g?R}>~X?>JoFXpxFgQ#3gKk?darfv%oO>p`8Wv4an zJnS7KHpbYFdz6v_Ep`Zeh3yE!QjySdSR0raL5TXLe?+f$i@HiMg+*UPY{Xd^9p#0) zi3$L>yFKDgj9$q#ePd%d{stYRhkD_wsEyn`e|(Ci8v(Ih!U{Wxa7bhdN>|tLRJM6m z+t*1872Ile5Qr-Ul5P`!dJ2DM6QqWG_UDMq63xOmaU3#?BJA9OpL-rej7`!%_afd} zR=Ge;&ash2OVD~lM6OR;UCA<9j~p9kYSzGy1xSPRT_4LI*8%jPmC3;K4PCk>dgy~H z5t}BdR>aj~KKi47lW~8H8;2qWzx8YU{1MZj8 z2eWzivhH&>?s3m_m8pq2W=mJ`+!O4C&#>^v6}+Ux?3_u_@M+Xf5cO;n0{3K{+RPFP z$&NpY1cWM~1VpjNx6IG;>iAnEx=E^FF-SScZMrJvs`g1Pr7<$$ zOLkDXNIr-Ys`?J8+aw)UxOH<0(_d!Vl4yF2|H2osO_Pcp<3-t8II*t>N>C6Q2`Mr) zh`I&zP*4>*Ju0XisC7j|Dg==cQWiD{NyVI@TU9(ID-7Z)BW8jwH2yd^%k9Ja`I7bp zcC=gEXzy?}Z1AeQL0~FmoX}G~Q5-T<3H1Av#QpSdp2wT5;@Ereg1sCVyE`Y%Vzyb& zTkqzc^j^-33!IbZ(51@#24H$@57sGZ&`SC=)- z+S%N>?j3?|Ck_|_N_#m2R(BuPw^W}xrRQW2MG7(%kyj^MyC?4u9hz4Z#ue(oYF z*XSBUk5tp>DhJ_pS$pL<`24>G+d&Ay$es*1g^|Mt{@^^rSHDF+9xzlHP;+uEc$iC# zD3(Zybque?aF8CZ%r1!0ELyOiXIk+zeOs`O2n3 z=BGYsT5R_`5CKIENIa_c4kF&OTQa7WXQ-|948E_bxE40)YE6b^i5{nDv5&MIOkCbm zZ(@y2A>{-e6;d+C5kinE0f|d%5HTC(LNqU%6wSguoLo0^e{~$MUT1kR;f;MCNAx;QQ1i`XZGNAxA=e9d+%V&uIs$> zxA%$n+#Fuc?{&@zIg$iH1TjDoEhdSQsANUEG*)tz%a&Dk$(pfCGh+p7BvXMBgJ>}U zBtZ})2!Ke9(2Z_%e*IFvd~-Z;&mZTao3J%wR}E)|rQ-glt8U$X_ndvs-uqkM`qsDn zT!KU^)-yQCgWNOqN1R-}4^6A$2#c|L+yuOmb@45M>s5L3>hIF1g`~P?C~U@u)7@O+ z*62JTaOPtC_u77nAJ~&8-7%O`bELv5=I{g+wS`|QQXkN1FP%dVVCMB6ot+bOP4BwH zIDks$FuexTOJ7Bx*d9nTay4M)vcwsl;oUQbIJWOWln~g{yJ(D_VPQ5Ra=W;zWBlOi zDC*DuiY?8fJU4^x`()PQGXKD`2u%YSMTjt@`^Iy;_Ydln7PndL+MLv`GD3-_RzBu zR7i6qW|qr)2T+gy9cpbE@+E3*lU_%WQB?#Ar`;knG|V^7)9Dv66#SB z?s|QF+Un{uLTF4(j#3!V2qHn0D$v>eAVfN0{w2m$%%s^t>@TNT5waH|ltW-AiL@j# zY!mqj#?RSw(~{E#=mm;X>mC00=P%>R2m!d8=h2)Lv_q7=Mp4D~23wBLl`vv@z-Rx? z0UV=EL*LK&XI|h>567Cn;($h}sHo(3<84%_bp}Z6$sqKI)$~W`1PP=N5TywX#nszN z^mLDE+GWe_@zi>O$gN^FS9mU{qx34@@He>FIY-^;GQFB&r%2KfE}#(oA{aMRq>LJ- zy+x<)@wF$uN~wF2+eYGm;I)a{a2jQDiI7&5LrMts3R>i#2|*ZvB@KdRlHG62krWyz zo1kw&Qtj12MtCx~o?^@gMGmG8W66-@U=dr356g|3kLjeiqFv&w@g#rx`s*wOVgI@G zfwc5-MUSqG*xReqs;;8ybuP}j1h|Zw!;EKc9KKfC#1COix(Ea>=ufjO7710zj$8>T z#Py!1+x0q4J%t}d7=#=xHt4D@%lfONtSwS{lP_qW0f#V)2AYIaH{Nf9ev2)U&Jbk^m~NeJ!)37iI^CHgbh^;# z1GRKkqpLLPT87>4dVrx1{R*Y``~=;B|96e^|3}96f!cRz^fGB0BSSRJ8$A#kGKq{# zdTB+ZnM9o?Ru7bAAcQ1Hen*R#d=3+nTsq#kA{u=BObxBy_*obhD-9Qfiz6cPvW%&D zh+vw~fO@2pHbSCEur|NVhzYOtTLZmmZvh~!1WA=7E&9C+!vZzGM%o@HFb44=2M5V$ zzCcP!6d{IiC}uizlC#(UPTNk?93EtI%|}h`qIKvU@jN)=B!xz!IdT2F!SEiQx#hRf zk4|DvIBbpHK|RbOBOfudlRaxsv)$GR2^qALq_zXXVw&5tZ}|TJ03ZNKL_t)C+q^U9 zb8@9Xsc@r1fIc(K&670_-Ljj(ANpC^ci+v$<2D2PZe^o(jrDvtF1z+U*6JE1$KdJ8 zG>(Yr^pT+$T)4JPw&RfNCaHEW5t$pzY(Gh;K{Q~pz9tCzRveMxaO4|Za@8hNYnM5; zl;Q{44`MdU*b_T=T{}g=n4~#7h`zE;^ZZHlwoC8cy8&o zC=~L<%>WEN+hgOe|09|SE-s#m&;y^&n;vaKXEPBXJ8krgj^>bV8~dLcljI) zkFmP!BPFEEm*aO3h_oc4QKz*MKO56;QL|!3*%Ad(`Irg|t7W{^d9;CHwo_+$=)m9d zuDzED!;qHiV!p|!<%^`}4kD9joO(btCpg=@j7qf#GVx)gYGtBN{2`U8Q)K*hjh~DogN59ryOfLJzG$YJPPIt&Bn_Fkgz%V5P zlPMz2;*0ZBJXN0MYn34`Siiuv#w?-g7fD1}LaoOe`$J3wO2DN7$;C#JRCj}lzr@9; zOLzM=9AWWv=R7imiPi|aH4yz4Ls~KxlS>ieq-sblAZn-a+#W@Hn}Uq@tId%y3Y{j7 zsi*~coRp;FZ=?JK9an(xNc6qr+D;6{Gt~y68-I4Kx{a1_L4X+~QA#nqU~-Ro8;SfP zR%U>1pV6wWcTn9931QJzFj9Cg+x@};VaNOLfR~441J4fFsr#%i#sxy#Gd7pxDKdtJ zr{6HMu6tS5n+C!52A*c(x2EugPfcwynjFLr1t}TNMYYqWCvAiP(|4Il^>{_TjDaAT z*rt$LqAAx|Q&p;=&L?7$2;sE2CgN)Ii#40eYZjf9V7FFhztzI*YuI|5u|-KYV{qc} zE}GZDdG7~l{=n@t5*D{VS;x>l+)fS6Nnl#>e4gtx`9`ox*sNhs?jrS(M@an4uVdArQ|o(jG<|ouGZ` z6y;itaebAr(j!7pHyu{%F+dxg=?8qBHkwQ@*p7|H!;LEIi2@U)eg(W6k*+t{*xu$B zvu|hcqrZkRu>(Hw$H)cZ<|OmgUS?;D#E;Q62os1>^T7n%3A?;LLj9ek^Mnf1lK zi+$%ifSR!FSOXWv+61A}1YVP3Vh5f4ovgZt$w|0tG|ww7$!1R@qXLH7^K{a28Vad# zGChnN3aXWmp;4XePyPheshe4Gj*>*uaBpmarI{4By)BKr{{!USJHe)T0syDG31d^3 zeuwEd1e;=mfFjwg5Tg`}S&w8qKHWadNIG?$2~n0!@!VN#Euxa2Ve-H|)bH4hoyv2m zS7tFEGZc+VjjhcHO$thBhpk3@@N3m&d;>IbC*sfv%zBrf_Wv;xg$s->p61$|fx;ry zYVn&(|A@3X%GgVb>^SA|qtAVb?3H;8E5&o`Uk0Eun&UP76p3<@`p|B)P*MB#SJ3h$ zj6{yKN;7(`j$Ns8X8rHDzT6mi)1ZSScv9dg) zeG@}Lb}K;WCWTXLWP(SrHWxsJ%$ScOr3a#csX}C+&|j|;^*UIcDhNRv4@>#9y)8^( zGA$E$(nTnR0F^*$zm_%8{0g1E4o=uyM9N1e#9E zxDmZpjZu2t2wfE@^|fKm*S>|;>G8tQB~FXS(51^{ZW}`?I=+u7HPCcKrhw&obdT=^ zz$@sS?Zr*Tqjdk>eDX^)pZx|sug2g&{P)E-KaUaAkWx}B`xq*Rr7T9&2tv%$){n%` zW7myBl*_Md()h|Z0hmm$vH#rPBD?~u z%%HfAz>_Ya>~UA&0xR_yny+2tY;FgG7_`m!0L+ag36;W-iq-7rId7ll%7Ua)4p4gE z9z~_t2v_h-kGJWuZ(Ac9|NelibJAJEMni18n-ktHGz4Q;BND|twv?2JeZpiSo|i$V zhtOg+J;_b%pBW`#C^A9^1~FugKk~WXh?M0jHO+SFu>4b0AIRx zmBiRQ$Zf%B-zgR$*Ix1<=Te+QdH;H3k zBdMopcGDOtPfzs-m7*st3_~K7!i(UP_A(;KQb?2;Os$ZJ9E3Z-R_L+W>-*+8UFxFF z+1`2DPkaKuQKXX=P}yX2V-34+7OT1*8+&(Sx*j770hdqg;tRu1^JM0Eo*LSVNj-D>deLfmTYiRuU&=@=x2 zsXX0WmUgMg>uQFm+d+3ym_!T+jpWW_gpDd|7tV7yy-9O3WXsMHNy&PNj2Mq@TB z<@d0SBHCDA`m?pkdS;ZNp%kcy2XY~rg3l~AaFw7rGD7Lz_kxPpdghCK>#~K^V5asw z5%4jNa;Z24LLg7Q#7f;`cr#}H!&Z%OtwFu-J!p1&1h$20_ltT`nYhXvC|>wFN=USW zaTR;59FlP2!^L(ap%PqbB-xysr7`yRcMACbdVC+KeV2zcZm|!bPzcduXR?Bpj?<5!ZqpD3p$M_0WHPah2&NDS0&9zg zZ(s_OMq9_|OE1oJ;>GtCM$OhpsWi4KqGX)T8&bmzCil|I+{9vdABS@;c^w{KlWcZU zIJzKt^2;>(KrOP91f>Go8v!U795xWiG~SQ>8dZIm!*+@QNnq$GA<$A5r6GfG$DNc9 z{~pfRGyqOe2hYW`vdmt7gALY^bq&*n^uQdU5)3CiI+>k#n}c*S5xjP_wcuE zi;uSr9AR^BcpniFXG(%bK4ukGt39?ieUws6Bu$np`{~lath8w5G{Wk=Fg^o;kL3#P z-K?YceAW*nsPBItx(ZRX3i;kRiMHa}o&>gTu>Hl4vyn}pZG}Vv&g~3QLUHx#J{F_n zL=uGOlR0z;nNNI*w}Jr2DCYb!8*-h?m9r28WN*8l&@njn>_r-x$8oCPAYnIndf@>gtwz$7 zjLrwpix~^GwM5b=(GqR)MuTqmDk*6(;u*1~4$v|N!r!DFt*|wkVPM6lcK_`Ftf$Jj z*=NW&ufJv8B=r*h@BSt2#L)H6TFEiapY9=QC34x(*hb{3&jV1{@aPGT`Orj{5wj(i zZr~$?K%Tf0Gg1hpIJ>_`FRw8=Ie<_GddH!hEntj0M8Ei(Tq(r;9498mvA3UQ@SY~) z!;9Ejk2{7xg<}+$QZq;)35{(!Td@paL7upNotN^raxooO8}F%qovGSONOugWud#36 zEUkb0y9gPgS~J(zn`^}&A-8VOsKpG5=f$3SW#wbakoo#fYU@?P%nmdmSnnILZghN3 zt2QfxpCz<>(0tY^i)^h2^z60k*L5Mhg7kCD35T?@P>BWRqb{ZKZPrbXgS`@|C`qM~ z<|F11$G_I%g}y=dl<{_U6(2;J{d8aVhOeFz9c8u{gO9eW;|LQHWhTbUe9GJ4YDZF6 zF1oDacXXDzd4%qx1uj{yOTvf*j)kqNP5ssE(MZ~rqbM7=oy2F0GiQBG{+Gyj?l$ekfEz#yr zmKJ!S->OR{l4Ptb!`&W*ghQp*rFHEzizkmFv^eO~{e1-80$tHTOZ0H1O`>m*d$#N1 z3Ky+6#AtGxiPQ#Dty?g!eT0RcOU-xs#=Zdp@SPl`jaL}A6LixOsujY($IL{C2pnga z&@xFq_cFc#=T2Os=C850_(85Eifr0B_MZdORrFFVKHn}A*Zvd^KI*TLv~UbWWO6wv3!>9$ZnE@@w=^54A21VM4yos*WH`GC&u@I+IM+q zcDw=ivN1r@WZ%X@3uB@4KT_i`2E*%UZIFPQOrQ`qa_xl6- zC@i5f>r_y|2%%^r)F#!AhD4{)Ns_fJYJS9;AHRscEx_t)cM_7ZXbKa9NRK0`Q6|XQ zaRhDIg4tA$`JOxNkCHPu$#yP|jvvUv)9rf*4~>&P_I8q! zCukQ7yhxJQy=}JHqCkOlu|_LbrIm~CHT0WwJBm&_Vy+AF(h6?nDtfJhUn~%HiGype;TWrL(Z;bA{pVANgEJ{kkC%;5-qeE>_6`Y00{7`J5_IFLSAa;@Q7Xc( ztkKF!28IvP-1i`s-yrjBd^4A8PXb_$PEmNr5SdEEwcr}_Tc4-B7;v$Ch?>k`VzcM0 zLOAKzv~_KnbdaJc%OtgsUT2YmkC(|jRV7#NH%ZSqG&)yN1hj{4!gL|Jd=`McnLF9% zA7@qlEea5Qo0p84V|7C(_>OdxmmOw#!DEKK2u;UL>m)bYC}8J676ZXWJ5PYYp1~fS zAg->}ZodzsuTj}F%a{+1U`;N#u(M^5XnZ0ifnSpTqGP_!igj^-Uz0Uyygy&5jraA zu+}$smiiu!4u6QS?+>}q?ilx70}TWSauiN9LC7S05b0-nadQJ(R!>eJQvUF#57~l0tHd(3Ri_m zBtT1LDJC55ICcy}(|Gah89Wpx69;*^w}VK(M8TNlxRs+TI#{I|gWfd zieO|CoD^O)Ai^TkGCAyCLh&@y**ke-Q*dUp%U3Vg2>ZTeiIhX5u$#$b53B^%Oq!u< zmoS#A$d=2BGsNKdF;a;gMAzz=QJJpckn&wFDu-jmF5U79LMhb5EM$umf(^#)kgrt* zo@VmSUHSN0hX)Dnd5HDlo7r>uQ;fg<8ED6V5uj{kk-H8OtSu34ZBnW>uqI+JL$j;n zHG8z9c&-D>jx}vUKr)F|jOSgT@OK}_v;_S&+1DAhH~%#)5zA4&N5=Po+IM+qW(?Ha zTzmlOeW2#1Bw?qC)l+Etn?NlFbM=BmQ)xb5ex5hP8J5K|CW52c3R+)8@Ab7H9W)GK zFp}vaq5__DkzR$C7r)4zZib^Eqf{3iP7B(g9_I9_7gHMp-3kqal;d+?`_ z2M=&zR8gN8BW3EC+6u2~aZ~Rlv50Lf;Tt++^Q5}*p|$$_W!~}lZ*%1Af1>#}pGMVd z6h~*dYwA%lMjU}I&QGG2mN2JxvR~ZHdJrGP|J2^#!<_(xV7zpM&fa~9cYgq0dmd}) zGUb6H5dltb0D+`bOmODbJWoBGW7`kN4^OZeZm~45#C+yu+FFln{)T=pF)&VG#f;+} z??~euKN^R4hi@TSirLip<%el49A&dCaZ-{y-jU`|>M}v=V{DJLQdYg!^`)3Bje(ohZ{LIUj=tW4jicG-ne$!$(n?!OqhjPX#sRUy38`(~UWTW{uop zm-N2lX!#V;)g@9dpQolaK$-0NN{4~hyLhK-9Qk&fn)7mlRJ>S=Gjv<5T^xBKj@T!2 zhdJTi#8#@p;!Oo)DrTs1i6kov?puJ`q_e^7E{mIohKZy^ABi=RLO|~3K_;6TOU*OX z#5lVLUAlhXVj*89wf$`Zd<@|+JXXYwviRXm)Cb?mMpLKbCD2-z*w8f&kD4U?%bZ@B z;wP@Wm(OgM5JI4o#-vCw8|(yqjX%6t;wRe&`Gwp673HQNL=f5zM{af)n~c}FG$X8| znX%g>?2kV;{)vViM@)(4;dIlne*sQ%tUF%FfVt8K?DdDku@|NL*V##kw0+{6S;#7yFb9kf}iGu z_YU^RG!;+7diVoq0#21zkw_ZXUPMDDidtMrP0?GtN&-b!#hoLh*}*XiBnokG+Ud2q z&$^SG;Sjv`Doe>hwhFV@LZZ7iWEMDWX!w=}$s{G)X82IdlCE{S9B*&2duEdV^~w&~ z_9eb??Ikw6h;3EJEY>Iz2Nu?B539ie2rmd^G&AFYXNq>_^`ID+5F(i#0Q$DZi$gI_v_Q&AiodOw%S zv2DxK3k_5ss5z+w^~4x6=?=@jj(sqP(RHbOdJz#RmXkx+_I^s40n};@Ar;L;l4RiV za}VxjHrXVo#P8+769*8M#jF!w)R&qT)2%gj4Bu$GRxO9Tw)bc7C-=f*p8+9BgdTbn zu&pOihfaWsc<<#WkV=u<70VHfP$%dGbW}Xofo;V=%}F7ohMv!(-t?C`5O-Ngrs6>4 zdg?O~|7)1vJ0K0d57fTvL$~9NA}qu&p3w(twNh+)Yq(uxU;HVofaRSAD?1F7n;__# zd{uvp$Mi2?2+4lCj8W<{bXF3nBDMpGK$3DSG%VUEw6I3hZ~M~LhnTV@ZPR8tXQTWa zGD@KJQ?at2iFI+g=$&JY6$bhnH9$#p(FANr2YL|o)3c^&F`o4)JopjLKm60Q8WFv& zVsQAa{P5UR{5y6+5}5{}W%Bop(FqfCJkI}HUW4;HeuU2Mn@GL)0}Q|KXQ&wx44t(g z$*Ae`z-R?eYJArC4CjNmYBOa@lI9ARwRu{l0vGkG=rTZx1PqO#zxp(HKmVIdedeEI z_#OW8=AS~m=SOh%9Ajw0XEuKY&5nBnj%{pUc#73ogGrG?X$}|`UTvGr)eXoOxL{nP zQ^+IkdoPSlAyh;)9UF5R;T&4%vh|ri;L7$C*P1)I)aap<;`G{=D7%;N!WN;GAeFKS zbQ=mK3@81TITCl=2e~5ho$n!;ikIhzlvKxd(x`>x@>*>8_~8#TH#5ZQR1vGIP|rTW zH=jApFMsJOzxrB0ghZmpFfBk+>O|Ckm1Wd zGEjVL_XLZo$Ay(1&GYeGuMQtUCPKn$mrQ1sx~x-s<}t=D_Bea+2%9?|fOZS%qF^^gILVHH4-0!=eLgFOOG#ZMP%%VY$9fy617t_!Ymak49gKT5%6Mr7+)s18jS&h z^YtN0##KD!ad5QGmY3w`uH8dmB^k=a*Rb50Vz|AHtyx6c8cV@6$2-fMhyp|tu46cHpU!&zJkvuK3RtQ4@S?vWYj1PM#7{GlPP5S*CrHer4_48PSf3gxg_LbH z5wPAGXC!|hUdOpt5eN?G4%qR0P6aV@-JLzo(d;@cFU#Z0O}sGTk6$bE@5?p^$77&Y z5C*$TuaUF2DT^`Avpw!Z<@1eGv6ag+i>oLUqy5Usr0lS$75I~>2XItCBow;z zh-i>73ON1XW~tSpNRiIv*D-4yJ~eeeo*lEEZli!qRXH~LHI{d41QSyXCK6a-9B7O# zo#Upv-^N`tS=PH6Ux|K{uU={K@7CHpAFN>JnwVw?PLj5_MWAn>wf0f##q+q>PE!^- zp&bX$cTZ=ow`4OTgUgq%@bvgTzSR@->^M*vNfZc0hx|qW#X;0lUx6KaNz5D~3Zc1@-sLAg4?wux`Tb9$R~Z=Ew6ZN6QN?`v6K$Iudp!4aZL z8&?feTx_F-aQMrR(6U@A_s}Jjy8+wlaiLHufn{lS3QH}%OZYIf)rL&{tp!rq&?G1#Tpg0ll`6C6BP9Rz63GQ>Zc;QXgK_1x5&7sL}O&de^ST*_?JW z)~5lcqvKjRI(7+rrr)bI^(K?n_xJ0kKF#j~wSU+{4Uco`ZkyJjIQoARsBMnkc<{oG zKTB9`UU=A{KBy5i6k+E^Wuu9pXjO>vDMo+%*QuxTWZW(b?S3^Dar9OX4I}|+XAMP2 zCN+vJA#MOR@xxa zG<4Tx$o8m{C!LYx3L5{tA;$XY*09aW*gdR2{Lg4j#cW2a z-(tnyikN@<0Fy8sY6yNbfSmPh0=X?w!nE@flEWAWyXuLGF0tntQTEWPdsBP7d+bfiZ zN2pDWf)EH@M}G8|Vb=i~=?Ac)5$K&jtG&+ST}7srkAL&K=&c&$9 zZ_lsuH|FlR`f{nr_P9o6z$YbNr1|R8=o35njm`w~1CvCpfWI;BBHZ4jb^1Bh4;OiU z_D%|S+)4gzkB~WbH?9Vis6n|sz`y+J0?E-BoB!qG001BWNklhL?K9^to^-p`U2SFeuheREKI9LMZ3SMr!CZ1YNGoL@Qj zN&Z9j3U|)N046zllp!tt{#sY^E4R$@^H;yXAzi~v2tFR>`2BUs6Ibf|)#)mYz~o|a zoL=KHmCh(Sf>!mLR8nyeILyV*k^CJgJEY?PFqbjS}Zxa$9%H4?T=F}~wDBbrF3}Mk! z9@AQil+ZDfSq6tS=xbTuGy5WA$#u?DZl&4___t48<7(CAr;ZPjv-2Do{xH+V90!I@ zA_T1KBlt>V#M;3=nPjAX2U;Vpa;^{u3j22)##0&{w~o+E_Ra1kDF@jU=jmKLg^&iy zSmX;=pXU{S8xajsFteoGE-f!$t=*;^HW=#fcR9Pkdv_IhVeBYQKMRnyDkxkWy@1mX zY&)SzD{4@E=?Ti2B+m~VK&AIn8UJ~MeIJ3 z=8sAb(+f4Gav^%7f^^FWBK)lljM8YFPE?9QKFLbkrl17*^c>q80jtX~Bfh!jQz+@^ zxeW4O{66yMejUffBmzSOIt+k8_bIA=h-_!h?dOI@H#kw&y{XJg&B^Zlw!FSgT5WZ%OG0p z?*lO_n`Am~v%36)l$YO+bG$(A-D9XeC}>*>KkBfp#qQx|rwdAv>b6;K+iY*e8-6OQ zjx_18L^=*!*|L$T&!eJ~W7~L$4FD)*K~N z5n_0lmtJ~_t?nvMTubwnjUmju%~<+4y`WCHdxfMCkxy*m7_msZ-EW7JFp`KJ`;Z^{ zImFNY2DNwp5)((_jXtzYnu7y~MnJAyA#j|y_hJ~rzN74(`T*U0mZ5})mLI^XE?{~M zN=NQsDt|*jE@pP3{j-0Aof%=M{4eNkPjl5jPeMh!`@=uR;BbPTE2s|cK&tq+N1U5! zwLCNd(E}eL@z4hV2>vfvN5c@9mf&FC;Y1;33ZC9*;0B6)gAUKEGa;j^q@o@k7Ykf+&+wZLLo&!y!P#k;qdG6kL6@Aaa>z;@*y$cZ)P@=^`ZH8_; zAga|U8>5Vj8vN>AWBlQJ=J@1~?Bc(Ca5o7%!`p_c6!jL5Z|)=+WLfk#2vBrehdIw- zkTL7G>Lu+S*{-;Xbpp#MYDbIEoO%X zur$GZW1U|37d*ec#j^f5Z97imS&8`hCv3qvcLOCl+&VJDiIiZu83TodTC5{&sU{iQ znD!g#`T__=$DcA(|1R|WmZ#ta0 zb{DsvO0d>-Nm>?Wr%T{C$PfMuLenrc$t~k4t~LTZVUQIr-HHu?$=dxflfK?MjY$#V z4U&{Sw*P#dlW$~L?gU(^_n69A%oc3?;LVhNj#8<_ZNo+GSUE=`St62<(iYG{Nv36@ zPQ3`J4C445ZvkqJMuW6z#6T{Sg`fLXB34u)(AeO>ki#t#X|jHkP}32n#az0> zvTyL>mdE98zyGJWjpx7FCe#$)HtGaY@oec3+cll+ObUVkdFe$|qzJY)FpI-+R;k;? zFm#sNHXm#)P))?7#^M!^ZYQEyj~E<@&p#n0LN}l^MA5CY?e`1C#^Y*kIXcCLPl3K!RA8FlThQox>uUZ1 z3`^^ycP+*H-~WCN?mvo%1da5KHbPpmgu)~2#X642(^y`wk!(hwB!niR?3*O^>ELRw zgDWhS+;=V*q6iCV^9z{@nR#D zM*()Nj@4+QMIj3*1XnKfXg4F~cHdxvzH|JKiZF#?!rUIj z?n77!$;yJqYs+K2o|$81dyD6udyW@RevK!te1Qv_Pcc6~&v5b(QfMsIpU0FhB1Oc; zn&jNMb8KzJ_oibL1OZAZ3IjTaZ+5OP%u2P&mo8o4(RPzJe2?=&p>pvdsU5wYx%|D9 zGZ`WQ;nXa`i!~c!YRBJu4^C22*$OB(PveP*2j2GxseGQoK)lhEw<0zQKMG-jX!RDV z)i`R*l~U-ghl1f2V>Y>L5`Aix@rC#rj!zZ1dG9zsdSrlKeQ+Og zaF`=kuaT~VsQ>L>qjh_n`;h}2eAg^@OlLWA%V7XMlc7cF-YC1ts;Jq|(Ez`w-${Cv@xtm=kfJrK$e+s_~kt4nWR6 zfTQn1;nVaMI5P4IQKa$e>f4#l#rGkE;G;*55QGMn-a%@y1x-hDh?HclIZmz+U+Z-4 zPDazm$c+9BUSf{GiUXUyQ$u|2>NH<`WgL6*HnLfr2Y1D+%Fo_3^p^D;WeyW5!BRKR zr_Q!<42hS&o!ZEc;aPDNdMb=HLkYcz6oSc|#p%v4dRAvGp_9~GSjjx%C;ut-k+`Eq zHf535>vZKjbGbZ&x?-_Wqf{K`%2q7vsJrdhyDY5p%gK2XB0gN_3UP`qVHDzneQFN< z{3}epQlVBfaBhDOJBv1h$r!w5Qbj5+?x(Y|o8Gz8Y`QViWb|T7hwmJBPQ{13v>bY! zHtkj$)173w`4VqzeSwYkS@QY;1cGGNAgucg>EMNm<+{go-lpW}1W~-73t8erANmj{ z4j)DUsivfx6zJ_Wa+?8i|8Y2d5^?7{5r#n`k-#vF>x^2aUo64pbEM5tOkuI2TzpGn zL?QWh4J~V+Nt>5f5A)*EN6=&&g(7Rmdg1FGZ#GvE$M0Y~o8^yp{UEM1_{9sKW@p~u z?x_q}uT3wJi7TlETNNQm;m(7ZEANH;emF2*LyL zyVlL6ushp0oeiu?9I&ZZpNDo6(Pz{kX4SrvTo@WPc;J!jbxoRO2{i3S3iNwrd=C|% z{=*pNo4v~V{WFCSm_`a$G!TY~@;!`X38j^xnuWn8P9p|m8MlW}1p?tA?KHRz8m<5f zTAoYUDzMp!UowF~O(kf#eQk^C(UU$#1Z%!P(>3y%B4cSJO-VPjNy{!zzj2l{5zm}i zz&s)m?hu<&FjTGKRBJ@Q%d5{)?+fw&;x7$u9USK2xfrOtdhKyKOKC<1Zf0a6%m33M z|EGTgLSTQRMY|cNQU_*}%)f93Aq4Mx-}`v(>3Qm{%RKt%qoh(PipNK}RCyeLuxGIR z@(N4-21cp_!?3cn&L2Gb2RM$y^z<}?gM)aU$JW*smo8mG(=;s0;+amI(s-!eWE%MF z`2}ic{}a7l4~gKOl*T>)O@s02zxVo`q@?FnX>`u9d-py@W@fK{ZrOtKuXhQ8-Hbf< zQ*7sk(Tplba%~c+B)aRy`n3NWVC$q}AlRr${?GRAJW7tMKJ@taR##W`zD>`xX4Q;l zku;KQS&QWzOJ1;LVw)uwCxGF_iC+$o97D(nA@4m%IFBs>!bwaV60(rsmUTH4C5T0* ztCRz#!8MJ|>{?v~Pj|EOUw$7^9H&3yBYwww3CCL5`?7lxipJdMiEDP?biIlL7;k;g zDX%5AT!;DXuXD>kk5aaB{P^||Rc)K```!I3ry3Bp0YCl$;h{3mZ=K)=)<5$0*oVnQ zBVZVm!ZB(G8^D2d_#KF`9HZ@*Vf^7|$gXH27YGp;`~^k(3LKFmm>i?-SE7{5MZ`n` zvz%b8D#dUnPHfjh_*WE(uOG)b@ENkvD+sie$op60c*mJ>=dWWq4t)i%g7rnTAyz2;0UfxHnZssf~k;1B{Rv}x2@*U|Jn;Y4`qu)sFrwA@Ze7^{f=F7 zqZ!ysO{571plAhnZ139`n5brxH$~OT_|>e8UrhlRjI5q|zOSQ`8~^kpu;#Lz1EcS&01RZ^xo1OuH&8eJ=(8{}&cNZ9aH3Ies!MTGTWab()TH7J0rlnt zgBbKpOiVpCJ5bwL8)j{FkhDL}$jAue^}Z=0II_q7y-g75@7miNKAK# zS(&gALf|(NN^em0TgStga5T3cpXCR#Smh#4EQDhO`RTDyHrKCad-I2R>BOVdguJo;W=>7G`C3*) z{0!Pr{<`%pKCtIs5I_1h>092yUT2JQG(o;BundR#jv79(3Oa+eoOouCbvxD|Oq1Sy zd5U}C(LMdtm_;JhZJ1UUqP30Up4XVjYz4qd*W$|;2o4TYC>1DAjFGouq{VSMI@>t7 zFGttjJb|Df8lROz+OUPH-Nz_7Fqkdzf~aC_=STSMo3`M4<(uS_Nd(8l_px_ifUZz0O@qe( zciR(I5G!O;@T848uM)LzC!yB=Kkg0d5O{Sm>c_DDjwG{HN(AE7do_ulx+qZ8&r_W1x z*^6Md_7%VUF_byo9>@RurnEAqLr@FRt}~u z{^0sLKKj@z-0tilldh-p>Z>_<@+ABB@8{*0-I1|v+h+6T&0KNC6*M(9(c9ZgcXu~~ zPXyU}>@bnqI;w1&WHQOf$OyZC`~x>ot7}47HbX;0y!hga96frJY&J`&R3a2e(LXeV zVOU&${Wa6?O9)ccCdUpJsB1J?`^q=?uh@s^Eu_d6Mls6{W-fn9`?UFlBL+tgIt=!g zh$Vb;vXw#zf{mRVZyCfM&G5>mUYvEe(muYH=YBp$Z+DRu_5P_&A7b;hIA47T5l>>& ztT=UB*WZfs#J4%o7UtQjvgAU7HKFMXx??5r*96EPK8j<&(KeHy6=JwydQixGI7DI~ zz=rs3q(%gVK!Ay4J0dy#I)z}sZ4YGmDFmXNZ2LIY=s5QNF7m+uL-n;_7z_->;m9EV zdmkV;{2amIXAn-2Y}L*=pMhmrWV2bfJ*79#>Fm$;@~t6!K6tG=$l%~0FB&m6uTD`) zUIBauvDzKStn72YlV&!Ha2#^M3`V&?wr`ZtLdR!Jkbl;EL+x&6Yj{+b8H2p`Ce8;)?@j?Bm#}R z+V@=yQR3B;TL_rORCZ|6+Um9JIk}grR2#+d0{L*5?vWM{5KEq+y=Y4Zm?k1I-KofD zh&e&p)aPA)C!O{1@y=71$Gvu`)uA=?I?<_f;L3!}u5yaeBRMv$*o#j%Q#}$f{FE`N z4vg^hU;}^Nbv5q|@8yQZCXS4ZaO3JWo_Y5BJbv^Dub%8?eG3@H0WkejRiA^|LkI)` z)2$^CLO>`?YW0=#?tf!VlD$AWP-f5y)0dqQsLlG$_Y39y6i{M5~D*g{YpGWkKxMs6E zX1+k5Stc}ooV*pIDSQIU?x428&(T8#cU;a{=Q-o*?Nm38z>`NgF;?K{aGuLMS`dbh zm5FQcR|SEWXzczzrFa_u?Qf&}Gqh(d+B%Z9iu&DC(XqgL`aqeF8;3>=e zz7PR3jxTCc8aRPrhnU!U2X@+k{Sfh`xaPH=fN3&Z`yukBQOe%58$Csvt|M+^^h4o# zZr=AT9H-H}(`t#beO;LHu_V6I2szV&9E_M2llM#yTvJEcaR?4Iz(@}pw%tnEpQNLs zgTB5#N~IE}X)-V{z`-L&Fym1!i(JQGzmJAyKUZuEM#t-UNr+nNBOuz}+^^!E1B);3+;Xcz{qt*x}SwlX+)BiU@0OeVu) z`v&>Zv%S2~Cm1x-G(_LQ+L|`1ymk-ea+$HQF@nJ$=ByD6eLkNX2)^sL+U1FLQq#%M|E{IhGEd%-Akc3$f~C62zk@5OsZ@2Gdf&i{pJu> z=q;@L?&EZ=+{t*hNH~|n%;m^=0~#-6vO~e}V2PIY`Kw6N!FIlX*K;K7I78zT1YUY0 z{iQL40qd>^($a2Co%lpFj@b4Fw?XkNBAvnb^M@Jg`xd$0uVDxuso?aqVP#(gyE;Jr z@Hn}Z0SW#TQLwIFf@!&E|Cw1K^Wx2Ax3Jd8BeER zvW?h1Zn>0z(dsubp1gsSd6JydJZGCzftu-qVv&+*y2FpmkV<=FS%pH3jEvwK(xJyuH@AdjT&c6WF;yve^;?i4IoGlT8C~yo_2t=zeeGb|V9hz?f(x zveF z4zM%+XPAZ?q)cZ2o)TNR-P-MaR@28y$zDOh7v;#}?L;EM-o1OL_CvIFx^w5|a-BR zo55b$W^?pVflU4UQ#Pf&_Sbt(jFT@pG{oG`c5B^z93MLgJc7SCPOjrx{LSrvqi_F(SZ8 zN`{|UVVvzf&*7AB1D{DT6vXEfl*}Z44v`Oq*w+Um$?2(4V%}YGA>V{Q+eO)&uCA=A zn!bZhrBbxEx{bbH_|X`H{e`LOvXbQ|9}JRB#Od4^#4v&+(>Ku7^%~DU`z&vM^PA84 zGYSL^0^ZHe;NT#+T#gkh<^~hMR0CUp&2xZC{9;T8U`bE^OyIe zvKfQ)DmZy^&SREkVHgIE&jfM{UT^_|-n2T+&CTxHvTVeC(;0`!K+W$}p%xNJ0{KJq zg<2^1Bcy8Q?TA1pJYvXdLZS=>Ycld}&=>$Ut zx85~*ewlvV$^HFe-v(~3e+j=)rff`SHU=7g8!&jQGseJZk^Q?uj28MyM62BIV0+< zEfR^$`}d?w-Ko!8nXs8CxYO4GY-@}Xv#+Fe#+=2*4R55J&k-Dan4e!Cq~*%@A$pF$ zhRvsJYv%bb8jT{_J8}Nw8_@L&V*mLQu5V6Jebe25JFxGg6YWIyC1|hmO`ZQ|Gvs}z z{9hNf_<7GbO^r1efAm?LuRetHqwhhgn&x#8(&;oO=Q-{G(qTV_fclu_mbDO+tZK|u zFGf|Aa^DCvt)dXH8BZk{Pgj8nYxX`ud)L3?+wuV52KqPM5nZs_F&%QhQ$IcC9+#7W z8i19IQ!qpLi{m(JH@R&UH*Uju;y(~GGHT0}4%9A!)m66v;BOeF)P0arB!c7fVH06z z+n*8;M}Yr=_;meH*`LG_CM7co9ALx<^Y#OeW0#AR1Yp}!8H18{l8Pp3_|JTXs~hJA zCfV3lvazj;+F#4NxJTpOK`sO*SG#xnD`r$-ZQi_@r=EI>W5^-!k000r_NklQ2Yd5PbOsLK_-0f=d6}>+^${G$;ouWt^2L5)QMc`2O^w^GtE#Gsk?QS)hJJ`O z_6th>6t;kgV*gb8w8d&{Y-God9X$T{<0N7+vauMZZ7!g0z#*|Hfm zLyV3V*uHA6|FkV{Bgwk z%i+nVpsAG;YdWT%6iwhS=kVty+^Sg5WDUZ6Vzv*zA$ZxF-*a-v8UQQkV?0a3Y>FL4m_mfN}&%82jR_*XmKdVxU7!u?>6GAMi-2;H= zScmz*M=4d+!7+EtuZ0E#o@Chq({x7%{LjSExg?|Yt|5p z#R!Mz4|p6yxYJuP9qSfn!Vv5Yw+4_Bwt^{KsQLYVY}-bNfZG}2m7ha6ir9P&Q9v40z#1n4S zWOQ-YEFlC{A(Nib0&y?+pJWFF2dXzS6la{^7<3#Ek4)E^-X5G&Eb7 zelcf^wAn1nV$-HgJpJ_3JoL~*1Ofqaxg3t;kW41oxpU_!)vJru)6+vHlff_+mmS+w z7v|sh^s_2?ezG$clIYw>F_(j5y*NTFSk*GMz7T{WKAf_H&$n<}qo(*(=XF1Q#{3tW zvNr&1$*N{u`u2I(8BZn=r4rfL0#(jJP_%7#IE3>QsMXZC1CTn4$fz|mG|3j(U0@$`q zBodi!Yuet4bMzqM>KzEno|@-Zq-$u3+w@NqX{%a%pzimHQ!-TNDi^(sT=X&oVEo0` z+_BJRi>J;iPy@I6vuCt$O0_4rW27BieHW!7G|z}?E`=Zsbf9)oP2TMn!ts*tBWWqW5=w zO^A1ItY*hqr23aq$+B>SaD$XZj5)XVvdYMe0F=BojwNg13`8LuVlX^+j@MJx2?p@x z^Yd1X{xYt*$_-S@z3w<>r)#6O7>;4Uj6lr}8Q?%FFuv$*1cHK8wU3(GbI)=vcXEB2 zyEZlv@#QFU`Zl9=nGTFnBT+AK5JIqR+qQYvogB9^mfK-zMAX z&CKM9^;>X$xC`+Hw^@xad=N_@GIhLbV>P*A8KXib6`waq1GimfU$2{8TO_wb!2}Jl zxyQF{8{~ykQIn~kFH0V%Nhxmwz5w`uzg0?oYS!a6@D*S)&e;aC9CD!n1gw{ZEZ^jZOuo$6UCyMA2Z%hQcWXVV$L6JGJ(O95%e~4< zZo1sS3L7c6E_3gXX2#p2`}0?R{>9l~zLu7sZ&Yp5{8r2m{&C=?yfJz6i&2H~G47|E z(&ByYTb|@ezr0#!+&@5m$ew|gReW(?W%pv#Z1DPoyRE0sT4i!l*uSx@**^%J0Fh%Z zXv+eg0$*w<4oje`{J=>Zo>vJkzIdwM@C)w5ZJX^2k zSj=-Q(3erQNbqWHXeY?)4m6nDH@*?!mHjYax{|IlkhYbkdq(JZ9xbfQDv~wScjh+< zE_p=Ya4muMh~Iu!@S{u(g{ApXje?p7`U;o58^Se7Y$v^%(+@6N_gFL>i4Wv@C!J_` zb7(Jt)v_<{_@pDqX$2#A_td8WI7BTQ#3A?^ub1=02;9iR!U};c40!+|Y2xtI9|oKa z{#aD=pc_UFDSjWO{PP;(h@4L35F>BO4vOO6X`^RfHSVaPzjnH`t9R=rIqP(Av-YWP z*kYFK>poulVXI$ryjGWu5M2shywBaxPt%^`iZz1>g%1Dnhz&%Nipmy{RY_{i2PiUHzZ<_HreN(gHS}sfa)7lmiZECaX$S zQ>&mk-bzVM+*+_rjjiR6xPq^+2HRFi?Qa?HRPlm&Y;dAvBnMnXNjC}@U?hwF%Rp@o zsKXb=V7BTu?E$7*z|ZuCqrEB1XRHMm>3w@3aB$j8zFUCf3%|tYf?33`1>M=810qy) zJL73^iZ?K&k-`tqE!dCs0r6<5c?|Zg_B;)$Jwy72CtRGsS6o%jU_gP^j05043UNTu z*C9s=8V+*|t*ZJdagx#Vf*P9+T%=5n9LYnmM_DJLg_-7e4H(FvQP<+o<+7(W?_1W;Wk@R=astg)OOb-N9+Sv%hR#Uka^N zomQFpB)oV0ll#QyF17TC`P65G(f}Le%suFfve@LTFk_gvO(HSl*EI*;Ftpm7vc*PE zLIoF(;Ai8(wmWoRDp4Z`bWl5_!o8Knyi%-YKQSe6hIt$L(1QF90&Ft`;Iw}8>3pB2eSL^K&@n*y5^=O3P~gemoxqw1#SLAhZQFFm%~6VPYfSV&iaPg7)4{8+BKg^imR49T2@FJHR}wt>C1o z%J~Dp)J$69ioKn6HT0@H|F<8kqKB7pdrcLd*|T zTb^a_W}kI;lagTGV1j&h+bx4v!+vG4S_yP;{G4RcP364N@2VfzNx2}?hP|hh)VKY% z@~T`ZQKy!TA=d!#I(!9(B~p;Fp8HK=G{i@0nw^B_`pG>pWozYw49Wn8;(XLv{IL_Q z=8tDBb7ah|me}KVaL>CSOeYH!4gv$P4Yk9pxs!mQU?YY@5njL{p=LfI7u+wn$~a&{ zqD?!O9avBT=G=jd`M$v(fW2y_H{hnn*PE}W2PpnV-0)A_^;RVN!uB6M2xHcR5v>4<6m8>3;2p_#vEM)A$)#thfs3L+#I&c z1(l@K6NF2;y#cfut{fT<`FK(}=H;atp0~nvWk-}{LUi`;w3`6mE#t3)?I18r zd3pe%cnTK}C@@=f>Qq+XD=GE$$V opB}gPAIALj{|n-Ome7UNUvC?E?3vluTr&w*RhD^!Nrs7lfbd36R#F`S0SSzNfT)K43Vx>C1-TCX zf#M93(?o~Ae9+Cp;KvvavbxR)2w3?4-iVXSevj~z0xnXYTr})0T-=SF%n@vjT^zpH zyL_=SdFN*C+c64LTY=&i z-`Wi-O@74Rprc#EDh)#E{zASu0{>lg^p~M{dG*qr^(b%f!E^Ps=&okz4^8@+cS8rh zlp=-BV=FdQ?bnqHofi)sSL_pD)9O!b#F2d_y1tpw#{qpN{|<+nBZ$ENPQnjtg8qMg z9R@~)LGbSGqK^k^-{N_?dQprm?sa5x&^X0}`FI3BJm!W8H)#9Td>V7!plpGr7 z0}xQL@!gcZF&CwALUWL`dAjyIIYB6@pRI7+>)}LtL33b!*M+yH2R_J{yZ=-CHLQ)S ziPg6>P>>77hv0S~s2HO2Hnv4Uo3+^~sW)T@36*d8YKhbD%gC?K0Th+|kZWvk5fl!U`}# z8$<5kr14g1G{!rtpEdTPm{l|@Y$N|Oe|3U>Ds!>&FoxvoB4wsKj#P2AB`>MnqB0u6 zHeCCbwT>-WLsC+dgT_5mE6{T1cIGb}2wb5C@vYS$nlcGKrg{yqrz|3Z<#hkQ8O7nWC9g zT=Y|$+tp1yV8FYRUf3Ow+8R)P4-iB25x03VuDI&BT-~jkJxS#!o(mt;nRG4*A8e;N zpM*b;TEvivMn-b$MyfZ1pV*(3rjwDTr)&Mv*GSS&MlxG&;0%rCoRzUXETmc!odO2- zhY$E_Z8ab!T-Ipec~Ock+nr-~l>Fe!;mSLjRs1;O)8Pw=&|wNwyk>?sGpID@iD7N@ z?P7j5xd|}vc)_ltzKt(HHUNABw@EIn?eM|z$U)I}2-jhQ3Q}7i@ZJ){{Rs0>k+e&l zD+cEFTwcLNbC)vHlw_DZ(D=Jxp}hC=;MN{5&UfGUln>8ytj)B&;=s#KL(-3iqLIY7 zI#cRXu6cDPCFd(c#zCW01v%~H1jl4%R=c2paikPvG3+)JQbaqFxIoX07mU+RE{^e= z0Dt_rXZn6)QjfSWVfX4cl3Si#`~l2s^<8{jFADeWvXUzZj})gEYb+?ZkSKHVY>*OVa+L(NYa6nBY!$EH|6~(ZZ~`o>k!1ywzzM0^a2|Td<`N* zx)uX-ly4R>eb^pUB@MBC0GqLfZyw*C77q*aM{bXRY;0fV{jXQwKN~-vu5zDTF)_T* zd7tnHtXIyh_6mqZ)_daK&Ul^}#0*}TYQO1B!M9a7j9T>02 zyib-t5)8UV_jMe9;K~z6Oc`Ii@khd8pV&RrAJMhA6<4id_g*vrIhS{4E*88G z)CY9ONRj!G`My`#WDHN)`q`x29Q+DyKz+sy3(OjZY=(Auw;HY@`*^Gzc<&xv+;kaS zKyUuE%^~hTWp&(OeU$WyLC!(6%b7cVNy~nTRK9Zm^IZG#VI#*C?w3=Wauu$yt9#O& z+%!iDheeQ3#Gv*F|AUxP4C&EHRm4AqY@fdi6s&c>C<9YrMLTE78Rb{$YWk>%aH zuC@<`L2@0P{wo5{3=G^anz4t|%EqhSDb363rattiL9<>(HW`e8`x<-Wpx0Ivp{=X9 zaTGM?b2R4?+|o^pr@(f^Vmza=Y!7;oh_Knl5S;) zA0Dw4aL2qW_Js>nq_3rRmt8D6U7fcyh3YX!#5g}m{|KDD$5_KytCHHf;X0o3M-**} z@GawDU8K)vr%b z68N;|zGimw8}!h1pViq{WA{mB*hE*462Y($hE^KMjiO&v@8=chzg&xI#748sb8862 z=sZ+03eEHhS;Yxpr*SgB4hh@h_B6Fo+ueI7eQO4tMecf2bNjlH>T`F|avLUr;FABw zjxD9Ezv0g=QjYpn^+%j2Zx@++FG2_NW64k(mFlDIQvnC5y1{ zJ7E@}F4NS8G1`i4JSilH|aPBZ2Q2l3V=A6HgUZe-86p z`qxMTb5K_1iCS7oN|5}SXi_2u{d>0pDu^0Qoa&yr+wBc8AK9f{DqE)V>t{)C&MIim zh4Gq+Z!fK>EKz;XX@C2(Qk>)v z-mbtw&)COa!gP@c46Ku1L54WhgKV?TZ}9G16`@PK{dmuvfqE5#HJP^m_MA^_mJE7a z##75(bVUUNdOJx3z!bRN`My)tB6YEsA0KK(nj8GgmG~c)O6Kck=+XNRZdI4#{ab|p zKvp@Pr4uwc+6w&NAuvP0*gTd;7vgyMr<%{!|MTa7v$H97Dzka?!+Ig1>baxCKEu{p z67QqgP=i)C_S@$c^on-x8^c$R8gqIih%e3#4yWU@o45Z5z>)ugU;p*%7ZnmG%AZ#E zs>sqe7xo8^U;d9s(I&L<>-GrOIM~>f3jwD8S2}X?xcn=g)GL_s=aZH8liqUO{mN*x1laGiMR8w;l6eHXcN?8wKxS7j0Suwr?E)1X6f?CXQN* z^&@=@?%7#?4&1Iys+Q*?+uaT%(qH?39j;l)?V&>M=zJYl|4zPCi$K>Vc-NDu@Iwpv z|E4>JFslc7vLwbJ=LoH(2e>y}PW=zkF4Wk8Y~WKdUt?G^nmb6f`DNg~+{bEkCvSKb zp>w*9}TOmzL~u@u$x>ojEGLe>D!5niJ+apRd}EU8^>m z$9<2b-ADg7{M#^t0|!*$+qG}{9_zt}^R=P+90M6Emkb^EGgWpjjsL^gByHMQj8Sl5 z|G8jAz~(^AICFKU^R63due#N&;*ieIAg^2Tc~ZOTT% zOqqfm_yQCVbH3Pj`|rWY&iHcpy<_D@c(W5=L?O4s)LPip=_wcd+n+AV|Lm*AFKl*D zaCTLgM8JlkV#Xm`8~pP!nA3-PT=?R3C=Qjy0F=V#*g0pDmg;WM>Za%4e3EMsCw>cp zS+~lGz~q6uo{!HbFQQ@(wPKO#5joHMIXgAO9E*E){~KSF=?(nhaC5GRCAVt|gujjR zTaDqP5D&MQN`$fw?ED`?2Li#gpMz&Ht3kv6Q<@xJ5@tC^ZehK2BN_R0el0RQKL59j znE>1PwZ*SkYY(g9Pez@PzFGYLc^h%$c0rv)$TD>HEFYIQ*KqN>VbqHY0*&t8Pd&&?1x)bmKRJya>Ww!4mR0T4SNR@EyZ^iC1EA9r zmoa@|&+gOgH@Cf?8b38Hq1te9O+Mw;gRYO4hbtOaS|*l%!nMQmbtO(kBfG^LK2Ke= zh1{NFvs>xqM}2b{1skP`^m6MA@?avk_RKlG-)4NU{Wm6D8E*qii1W%b&c3|EdW-3@ zRKbTyF{a3{bXCY=I*_kr>rVUge|jm7DnY+&UKjqP405;AzJB3#Dbdk|)PswjXwiE9 z-6vp^vevx)jF+8eouy4q^YysJF?-5rEybwr$)(i-uK00mm#^P(DePTT*bYA4m`vvJ zk~aBEp)gF)MC`-tyyDRYQJ~biN1< zlmhz4)$cfQ;VfNgiHm0M2yvcNXXKwL8Ju{L*!ygf&v*KZ%^A2;a zdtQQ{`5T+9yZ!F^YF2qQ@F`#xl=%^Z-G6lV{Fs}FFpWmfbT@|fpn;^JQim(GPCCQk zriqQ!>bpy~wvGSRF;lzTV>|`>ydIf{d^t%gKOx`Tl0uX`H6y=pc4@*%h}p% z#(7iYav0n@lPxi#W$~Ct6%Us1=vkFQ{CCnU1>mj*b{nb*8Ek{)+Laf=kmiUoH496& zra__qh4|sxpm*BNe%V;%ylcL;g$X zZ>#^G6Kv0mr~1*_XKnCb@Wdqvf*k=<)>(#662tx6j4mZD@$3AyKy#U4oBCK(ixnk*M$R0)g7gY<#Q1T!2Uodz4bil6=VVx`}lqcSC_ z=t6AgUtWt0$f$UTgv2k1 z0}{#D?4`xZ9NUlS_Vv_=j~IO|R{N%o>U5`p56&aBBixi`B&LftB@je$Q08{oD^N~* zLy;{75b;Xn@4OprgNMW&zADc*jZ03$((XfW8>9~k_f)TocOIPh4GJwuh+wbaQzOJf zL9|9y87=oycRC)S{^V*x^nOD4=gt5y_q|WM#ZtKjAy^r5Ol27unL$T3cO(EVj(>du z&UL+B_oSq{?LupMIBvar+!Y%i5PNpKK3Z&?Scb2mC8GFWvn;*b9uCG0IuPr=SL!R) zlIFzVDCPRk+dg!14&Q41b2sd;mHi{DC~hgzCwMzt*ICQD?M zt#ptFu{zh7QRfm(jZFv|$PDj5vZTJOc4X~}z1$&avJi2!U#kBL?7!OoJr-F!ejAb{ zME!GQ;{A(}ph}r{b8S)(PJ+Ngv*6~evrl4R@!ZPOzC{C(2A0Zb{1o_N>yz`4E%=O( zOo@+oJfBX^<8NuPfYa6>Gles^q>YQVv4}#5STxL-p)M0uf8`jKdw?ll@p!)oTh|?f z!Ej?lu1V+gok#zRZ<()pWI)+oK1Y__=0NbTW z(64N1aG8AidnL%f>cd0zxQ4dBzYnYsh5K0a_8-;WtxqF`I>fV=0=rw~63Y~V`l#^O zdXMu)jq1qkCmz5my;{V}Ghhp??k5au#A_@ckxxhgfX8Dm3jg}Bys!-WjG_W`&)GswGkYuL!e&P;&6!E(s~GR;JFI=WX4DP~C$p~?VS@Hhs&`J4qk zbk^+RT_e@4tzFyRw=c|wyf>t!FsJ-00`Lr1E~9^5nbSef<}_`e1E0PogNS+2#-*N`m}B8W)pz83`h zWBZuHVd3b1AdImQn*13!HZ^oAY%djIWHai$`e4Y>yoK+#t>d>{7h@0K4Jj)we*_mM z(|D^^O6@sJor>WmR()@{iY{BZR(E0SsgmAzo1G{WTx|MYH8@7zDw?_zCJrV zIqMOum1{*Ac1UZ!C*D;2OAbA+ZbsnhXOT}U36|%Dwm$wGkl!B>)b~|mltX6J6*Tr@ z|8xY)CCEyxox-5t@LwXENSxQnx=nS|9GDO!>Mnqu5aQ_?3jDw`NAk9NvG?_eia&vc zDH9qm*9Fhw=hjM_vDD3d-Go$F@sU~NoT)k03vTQt&&<6R5&i0Sj+aT*Y&G_fC)g)O zPw=A<6J@`hg})+bq{z>H zaCpjzK#8y-A|UX|sCcsqlhboZLBQ*BPud+0%gSkVm0bWfPBu(+?`2rE(=6dl_!RV$-^)l%`^QEh;I+ z%ir!Z_MkSob@8w=8tJ|~g6hj@plwt$A5y`=ebsZuuu4^V(LiFiO+`hfH$8!A(kI!I zd8=Nj!&8}CC!zR}HJh4@4{Eq%J?Z_0)ZGb_LjVV%f~ejhG5ZlJwK%ofzBW?6V zQ_$G?)PX@E2*IO^-X@9=x!nE0eYmTgh|4LzPKQyEjcHf^K}WxMPEgF1uxYOjz^=tn z<v*M6O5zX&ciVG4w!q2`l^nqO$VAB$AdL{Vq0dhR?v>4 zFbeNvADoj?PmM%%%szjPkc{dz#piu{nZGH}9Gvs^VL-@Gvg4DX136D_*)JDo z@(hIm$4K_eX%vL&*;@p;$(5V6-kaO=sm|f?&Mo_W_Aw$o8Jsx`iBV#TDHJb@en)Gb z+wo3pykE}Q{mg{lwHb5#RQC3d2C^#Gudb&wSl9)N7*+WzGgen|X2^WE-wQ75FtRtq ztc-S1_$#kcJ-sQM)Mog!lIs5EwQ{EF-jS|f&TQBfK3FDz7M=&KDj2i(Ze8slL)AeJ?5s)o!7XPS2{~5$$k!9DD=9B zlmXQ|%O6Vli7ASoe?rMsoVNz?EG#VkbSw)5C5)OPjA$|vFYDNhIvi0q5LIm?>pdqQ zkr-iop$#MG@9)ROV<@sjglJn1Hg~(!cYOAaDF*5zXw=q5#pHO5OUeY@@F*LtS8ay(-u1sR@*IVe0+sU@G&53mL=I^5PwoiM)HST>a=#n(KbL2*mrr$I6m@9jXltriGU!E0P;fzgDakBf#fMo7#lI7OFl?oA88G=;;smYG zMt(E?)oaXf+{dJ_sMeQDSY38nm6Hgq=UUI9BInP&;7tS&De5L2J5*-_EDw6~cl!3Y z0hZg5@;S_X^2WYc$qP>;u5j1C6PB~iIlr5q!_dv#&wK?6rIi0~U@PUQcAu;d0 z8~GlARMj4?IM3#i$_ZZ#ei0Il{9^R!T%a+-zsBbvIK7mY(2KCWK11p&(6iEs`TfK~U${0!-m z_f{6%=iksK5;nuMA$SS-NfG>xQCNrCUh8x`Kq>+7W&%#k#k3?=NX++o8jh`rkt~FD zF$c!4(kb2(v;&IJSMbAp$Sz-BLVFV`5C zI8;4P?J?Ouea$axQId8Ek~j=&mfgaGe~KpxvKy=SI47r!*8#4Te1ef5PPVR6e9!}| z>wj-;(UssHk>&Oc;@(e~MmFFLa05F(2JtYy^=rE4LM!e>{mAh_iWqg%fzELN>r=_> z{K*oqS8Gw>F(oBcKmXA9$dvNA`tM<42`&-28a@%`mW`{t2F}D@YX~EblJkTVNxich zeYiMQIvKPW62fOVo}cIv9lDF z@0u3@i&+^z9U*ozI;^#S|7mGLeJAri!4UBFyS5l;gEVE(vG!u05G7jcFC-XC030W6 z6Q)-^bQD3;v}1m{CbD<3g0X90Wo-d8Sr{Q1DspK`5_S1wQp!xJ`=+^i({%dz#-5t< zMXhNE6*6aQBvgSPf#VllJ*rdfse|F(D*%RUiV6*9N=_1kpZl@cTTx>AOgyy{g0bh8 zCgcTi<{I+(cu=PZ812dlwXPvdXKViW4WNWSn`JQiS zwA=XZpiq|gN!=vWP*b0|KKCG|h_KC6E-Q!C>dyQP`+*7U4yT2wvn;wao+*di1D=^E z;4G2hFlD7DoWkHZiQPljNjr4KYJ(1r0a>Vn&7d|}&rl}UI>V6< zE;UK%*>tj~+P_W|j|Ysu%0CukX2p>Ud5+lU%cFdNlZ z30SVE7wJ`&&?5E4{DdPceF51>hM$~iPU`*KQMk;h#Q?D8z{~7`gOt1jvpMHEIpej(!jr2B_062Q)g29Y|FruHhY z-Sfe<@tXR(b_51jFw~{*dAg++K!=9Tvk_!R?pW}7i?Tjv!y@oU7hkTdIy_6fx{rsc zoqxa2ov^NVklYNDZY@+c%7`Wyqn9WI0u5V6e-{f^E=qhH++gC>mvf>q zf8FY`eNc{?qQ10ZTMpQTnL4>%n48fLp9(le^>1z3^8TYjeMHCM{IdI!BUg8H{s{T*IB{&_e?84oWu0h(@un{L?GG_%$K~`Ka zKA9R(a~7K8?esS~|5GtoDJm=qzesN=9*cleDszO06pmP0cXCsJ-~t-^=jOVHwCvhY zyGFdCTmjA%IjJ;N%f|eqZO9!E_-g_aKYEgDB5Xm~U~z&52fQN)X0aABSQmRlB?~UW zmrE!4AS|p{p^Z!w$nlNiD{1cR>s`!>LQv?u<-#|oN;z?NN!|T8`@E9}2JuAI%jh-z zkqI(r6nXs=#wmqv+pIRYTr>W?V^iN{ZcN<~G-Gqbxek0blBwTeM@>%OaTkm$E*%<2 z()K56aSyMK_8o`@SoE8uVev>g1jmHWz54Jws%CO#i)*_+Fr*$c;NgV zN#8{^Xv`s(v!pXm5yu~SKbb56Zo65`Swnr+f3{Bp0QPw^(C-jc$Cs6NA)f@uh(F^mUhueJYkHP_~PYAkRnA?uu5gLd}? zc2zKmsjP0so4*NpYfOlO*paZ42Gk|xI69zNm&hv6TAgCcwy2eh7rornhDN*(<`#l# zgW|i!Od>PfppK3m3zYyIXV%KOsTWI}gEo$LYGUCLw_RBhOa5lBlqTI;rVj}hFg~Wh z9A(gtRy(OUA@2mxIZ(Mlg?z5Bufk*S&GXz!0Tq>U$n^dZHGkQVw_@Iq*Tbx1-GzW) ztY$3xZkDBOz|_PtMBQWiiZ=kK5CgNPBm_cY@#)|GOtmm znll=&pEGQoFsS*Yjrz2QTw?Fzn&0a97h+3H2i37qs)rNx#Ndj{K=MKm>%>uT@NN(` z;$LTj?@2+}^sZ(g!6zKc$yoDLj0#@>e!&rhovrS-vX?S?B|1FveL@x3z;6KjZXwy*QUS{24WvhUrDqyJ&L^Mc8pNI@ z&TFmS{3xMSW*~C}NAa>}d}obK4AUBv=y9bKv>T&W$!5Zstig7bOq!_1i`jMp5`ii3 zwB(js)*jq*JI{#n7DyLyr1$E4P(J!uBu{!Uj&sZ++ClmmbgX}t&Rv$y;$|{W;dFkG zm{4)96V`RmEnjJ`sR1I86=x9_K-wc4UM-Z@PRIp)jYWx8UGalyhz*5W z`4sn?$#hrbeP25GJ~GxCyEllKNyk05^@pn>`^Z9kUnzq+WgzFn&4s)LWs91N^p@qX zdk&WNA8@jI_4Ai6=G0&B!Ui+9U=@0ePES{JoXs%@ElYk6M=P2x> z0gI6!yxq|jpZE}L1w)Y5osT&L%f7`Y{rQ_jl#v&AbfxN?B<%*U5NfYFEfRJeHWiMP zF#1YFDuHg5lDssKETA{hr)eXC)nhFX4R$qF&lEl+hBf)KymIXH_xJ~oTQ8W76~T6D zYOhTL*EaWmDg=K7-uF=)hWq|#o zj_K4uoiO5|3F`*ychb1T#RqL?&b2;wL{lhz!#e`td>JH4Am&*gV$zb^k9Dpq2!KNQ z>ugYhN`jX2)rLzHM<*GjKdv-1Fo!2k5l|Pfi7qd!8u91mG>?>|p2;-~P8*U+NSlifh5AQSQQha8+dgb1s5qUp-(+)r ztEA}u)IEVeV{)-BmapoIo2m!0Dh)Gz$)&w~Q}HEh>QGPEShig8*KcTkAxyqrBGfAm zP+=a7Jq6SG!$)aMtz;$p00$J6<)ZQeQ+G~(>E1{NGZG#V61}6#nvZf30vqJ^G~{o8 zYYH=~|Fsh#UC8V1Z=Ot9oWOJ>q`WwBo+(^Wl$STm3rYK$>f%%vVF#k4OkmFm7x9Kc zdv=h~!oNXH->-!221Q-8ApN#y>`~#<+kfx`O-wZ3kqK9@0VIDkEu ziHeWXKja66p4glYK1D0v3d=J+U<*Og)Cok|e;)W;j?f80XlRRsWy{*K+`Q(mvpfiP zo)Sg{r@^%+X92~OHXD81n=?ET8=3Xc*`dth?VpqM9t;c&r@}Gtbx$8t~PT_ppHAaJPvK>iMrr7m}ug)<1Y6l%6Pz zHBX%_c1&36z82-_GpiNbIR)sW$q_mt(R1c*7s(w$2ezcVN&htF7|bX`ob&M3>9PiN z4N1fwm`0k$3cfZZ5<)x?z4TCzc;OsRQjsdtPuBzO^E+W>()=fuJxqEQgxfx1zd~7; z=^%FRBEmT9bwOaJK$=D|OFX)q{>D$)I?o7$JHw(43PnZNv6pk{!M!LI!bYauXK1Fk zi0){`yV6{>-rePh#uwi@q`vJ%X^$VxSr!s;oai@8&N#-T=lxVKz{JJxZ;8wYc zk*CuQl%8zKgc7Jari>#Z*B{IKIk>(Q+1H+8gw=r17|ElmV^{Y z-@aXExA(;2U-6)|_vj&SdcHqs-Y?4d(4v>j#~=^(@xP@?$+U7&jt%x9{&7&(&zqk{ zpTf6aGyxg8ON#@7b*qPi2jfSwcrjXqn%zoaxpdZj%|^DlL~apEJNnco5pp5+u30Dh zLD*AVD#`g;*-#9HoxquLsimuJ-5*m$Px_y1=1k*xs|}`?nD3PJ;OOhSOOSlR4v$Qk zkPK^-_@)a+g=ZoCq)(wz@YtHD8+*({L=5=F#WL z0x7m$8E4v%h*l|)o$_n8foWZAgHb<4zNn_1k){bo zYzmKk+*W;P36hJhai$dH7jN3+*@~=fv)}D>v^r*T_p?gQ4+V8|70g&W!sF=I#qVUh z7%-20krs2B?@Ln(E6#G0OU=i+p_K{whd8P+{iQLYGn4~9Tyhg{&JJ83S4Q{R&^EfJ zQR0X5H=EiYm8~J`NYP9TnCVTGtHfY<={6)?TY{i(X9e9vYM$g?^Wz~o3Dkfj)iDDQ zJYnnrNWWumDutOWa!4}Ek5reZ~I3!fpQL zEv!{-@r!n?(L%>f--J%u5^madMaY-whXuJNd?Jb&>@D*=Kj*jBSU{3oC}TAaI`^De zvuU&u<>{Lkb^A@#a%YIp^c4x~Q5ElN7bha}g2KatrHb;W<(xZ}zq7ie&fxs^#`fei zSgf3xd6`C?oFbq*xdUEK@G3R~a+5F%xkO&h;S80flbb~u$X0tI^~XgBt7TIi@BCP< z7V;xt5pmQD+ zSj6UOZE`&Wwdr-rJEc(CFbA=USQv3;X!T+n@u;4odwOscE}X*;s_gR+eHH#Jz14=2Ngeee~oq_HeInzcc_^~fm@0fJs4;B_C=RtsJQadZk>?`$hY7#IEdbKp)-0-%9Cd3~j~`FB{-dn28;@j( z*jSy8IC?IrcbXq4PrZwahmXHe1QMQL(AOJ#^OI*qy1GcomhF4$Yc=cWkAO{h4Fo_H z`iuxKsBG*mY*T2G(l9-04u*u&pd&Bo=N>_VtF2%Bw4C*nRtY?mImCkzr{ZO7q6M&P zHih$s@Gk#2wylin>z6y*I(9!QEi_;SsGDB)n%&mD`$;+a_evFr zBZmT-FLE+A)iC27Na2RKdw8vdGiKp55|T1?DSA@U^qzchBKEiI@|w2YRHUTpzl5x) zgZP8Brj3H0>VG4@l~g@1=7kQSsFm>fZNx?f!MFo^>9V%#i@8uf0*eJiMZ@#S3?9ba z>n24DsE6!iSdV{C{H-VevC6_)2n46E2%PIaq)11f%9UX zs4UUVb-roTED{^V`%>5^eUa^0S5uhKEgc6=obCOccHk}U7?rt514*9?cEa&R)$nGE z!dink8Gy>TK1lv@hyP~>vR=fB2Eo@Ks!g+P1Kz!Dm}M=wP`s@qr6|Tt#k@1L7yC65 z#azwBXBWSoKzfAEf@!tlPR`kKf-`ca1){}txF~GMFBD|98X+D5CyFC^3kPHBMJWUh zS+sCnja`H@%BXJA-XVQYwV01%Ak-ZT#qWv(1E$PZjSH@g(KD zBd$5d+&{o5v@RZCwhqfJpe{UKYL?}bjfb8Jd$>m*3w&6TDE0Npe&_Pq9y?a~zjLjN`i_auqkRXjOHVr3EirLtnSqA& zsbqMCSS0EuSK}T+^ulr~vUEQRQ>S1~{bfLV+2h$VNRb7BFwW=^2M!N_p7*_BbtX#d zcFRg-bcElyf5)Y*?_SyUW6?|(v2%%pdjM<86VX|sRCx$AttB88;qgJzr0Y9ayi@bv zm;E_x=rw-i6&g1ZBuG-ba*jLjX_>a(#7>vNnWuxJ-IXCxY1i}h5W+sU-ALWzgdqT!q4E)*7_)uogDtR#J471up7@q2spb$-2o#eHJt ze!4XF1)yR&5W_-I*8Bsza@xd^3DFFbop5l%u^7LM&AJUyv8Nhz3>0b=59fAjLs#Jq zXbG81%tC=oU&@-ot)rz~L!8tZPk>vKFyrK|;hhQ!8CGAF#u~F*CIcMDvv5*S6sFR9 zpGug-bGi<4QSY}1BHufV2EwVlOwYc<5kh61tMd|3zG4wZy-E6~uW_TiBb&Gkqt~~I zk{w?*(@!DKSTPHkhJ+(yGe?V=QW|mdfa4s7D8^Cbvv)C^kOP{*ohX zzI3nreoaf~Y!M;~@M=ODzX17As9ihrTH(nzyjDS^-q`g@*Ecv{H~>T#ebxq-cm**0 zBFXU)jnR%+aaoqgB-7xr7-7*whbab|-evIH5F@pxOo?=z@Bxs%&`*Jvqcda5F3E7ssw|#h1WUdtY4MNnziaLuP-uM(5LGvqZhaa7@6Z2o911@qKvm%C5WJyf8IX=brm9Nvp}a zOF$d6U>w?8!t{Ngn6$Le$a-vXic+S^ohP&mHQTY_UBK!W@mxrd_I;Uv6Whi<8pDEr zqhPMYg{NpZx`A`&JS?}*i8yv};r+VTmYFIR2C4b5Sv(fPlgG#3^2j;ye`MD;2vf18zrqTwAh zwnK?`*gWD&u-O9M1lfbkO^or$f@#+vEIaL;PQ|)p z*}+fzu_WSn^^F2*>DgPVs)Ikmk^5WYIdE*S=Y;b{Hm+B_ce~vGm#85 zeLiI}>89ooAIE^N22kpJG2H+5^k7S$@eM_~+3mi5xM{myv?;<-PD@gj2E4PF9>6Y> zkDr1Y&1t9y2^R5;N4dTf89kdvLMt<8@s#?Mk^5me&8kQ%ava{~b##z0DxFxzp|dl2 z8K#iP3n@HKKf!dQ>2;JX*Y9=y?!OkK8_HrGcLU1nrBzl?-4a3Mm6v%v=;w8Ap|8h# zmKpowy}981QV4U;yR}wC#{xN%QTO9=P>-?y)?`kIs6vN|lFFul^lPFvVUg4qqdnrm z0~aOf&W}+HQ7B14P4Sz_XpDD%o(A^M_|?plC4q|kqrNS{qICh+mV41k*eN_elQtAT z;zaEHVk~`H6aA4ujfRu@P){VJ`CG~fN$IHTET7l=74E<<_Vk0P0j9noIfPe5>5|4D zMPxf&O7N$u(a7){fu&>TPRZ~&%fXao;xC0FwlAxqNP1TT@O zB$ZDTdxFJI-R_oSaj~p*QKw-psBBuMns{xFTy~s6dRu^=f4tiI8;tuXO*a%oN?#(N zR0`1yuix>=SKeGp@bMVgTrH-kWMOQ|o1Zjo$oMqL$@QO?wbV2utNFnel&+}|0V&szFM~aj<3|mbK`jUa9b**(3}$8i&>%Z zP7dGjre*G|p|RaIztuz+4c<}q7l02=YZ@ZHD*S*2MQ$An6_7?9m9jj}VmYJPC?!(b z43o(@w&~P4^8pO#PEVY-D+vfX*G%H-{H^2N^f)`mE`^-Zp;RmkoyxldveIsGhlR;KTc$|CWj3;s}4Ow5czdbY>_1Xsh9A6%> zUv?x-0=nsDr^=7K=qU(*E2Z}{9qUd9EkR`)>#|)9ztyqcxj+i07%4Cd`@DOeIZ6i-t;Nhs|!h5!;Abru?0Yr zGbs0Sk9sztc!PqnwH#pWb$8!}&3HqAAW|v)+a41>Z!5(bO9i zN&TAoz}N_;p(%pdGuoU#R(O4$(mzP>qj~4mxaK(MVs7`JNCguku%(J4nn3ld)I$$@ zamFADd>GAJv5+zv7@5wwfcQsVn;=6Dqr9}$C=#os@VA=w!MfbO`@`F$tk>@Xpr-N( zM!7#r!=WQ0>Acok5mJBUxu5$~PQ$s^Cj?|x0<-!8*y(u~YLl=$Y`J58Uk{U%7DJQ6 z)-F%FDlWP6M);gXsLy;ee{Mp|fu9@;nE59;v2+Yg{h*YL;kg4UB$Z>^gXW-NL9iti zyn`Go#n5XX-XK|{TqhB_s|wZLfk>snIG^~N3+VqWChEXtU z^vmrFeTQGDnq^upr1d?i(bC^Ke1b2$z0L;pb9e(q4PKf3n4z;9%Zf{-HjuKELOrq{ zRGY*Di6U&RZLXw+_6^@I`B!E}DU!miIl0{5B`atbiZZ-0% zZT~#R+4z~MXJ4_uQD3Xe8IaD4_f~fCYF{ZhnGsuP`uNhTU}oS6N$n4U)0Xh4p8ITx zYA%CT-n$|O3~K83QrPcp3#Njh(`|N!>B3GQRzXNSe9ILOFUEp5VVNs1cgciFeODV` zB`%e*Tbx+fpe%ZC4j9R1?o;Xb{FQy1SrwR9@@rJZH(BS=J4qI;q(srRVenRr^7CA^ z41#PXjq=|WS++|;it}kamBB~_Uw5&yLACC8)Rp56&PZAC78zy2v<0n?W8+gJXQ*wPC^Voj+Z)!5$|pu?U%xWiJ;Xi?l`R^#N$QG+n#Jw*Dd?zY0Rnd#c)Q; zU|B>&Vf^*nI!$<u8Z{2(D zxyyO(^FHtM<~R=rNlx$Jp;1jnLs9Ruf8{^rjhUdX!t?6$;xu*Wy@x0MG~@4nKd&5x z6tC6-nopt+;dN4j=vKNBQ6f zKlpkN<$ipF!hRcy;ZK0p38UptN}y-KSny8*Xq_-xZF{y@hq25Vr#nm{qqW<0OtXw@ z*RCOfOV2*fa6gEI>$>L8KKuwDeB&lc^}J;3Q$}kM(=e;P$Ho1tuA?~DyH0T)Xf>xV z?k?)Hq?B|n0Ij+U78AE_1O~x#mXh>UiTjOTK}7vkwk}Cm=fK4|P1R&DL0A7dZyhDU zM@xm&p|5Lto#25*5?Brstr2|ZaB76_&VXHU^WU<0^Y6ZPQo9zB2%Jjka3n&Ks<5wa zph(j&j-yIZyYTx_=E#-7HV)r8vS{Xcl9oMX3YCCzMQ>|Tp-JO%QMEDMBpt>poNMV_ zg3#iub<;VE2XDjmkwW9z2C!&2Wwa6mg)*3!p2}3mBHnsZNL=*5D$*8fJf@tpHQeQ) zVwVFM7Ui8hd?3`KM){pEKJ;no5a2-{z)>14&Feg;YijTrTSLYBhsOv6rC4ElN0&&} ztQq;RNN=Skh{1vELI5T9-Yw86m>${I5E(6R6q`kd76M1G@<{UmrBT={T!*PVWO~#i zk`!g3n>RSwv-67JM}PBg^1TONVCz=PcUSM{P3rAja);m`iyFE_XoPO@R+2f(uux=F zpd|w#Szk`*k9O(G4zYJu;Gzt?^uqPvydN2@TqHE=!P9NbIHeJhZOe0(v#Ydf1l}WU z#j@0=jFzib$VAh6Sn0s{y=na_FCBF}IBcM87_V<)(h1|VXIX~pIWBL~%tP-@V_ z<8Pcc)Vv?x{&6OT|Mv0GI@M^MFg#^+PmYX)Eth}vmNQ=?&i55PVYQrd)aIC6Z?M58 zoiN;9F37d!!eq+D!4TyfNqy!9f3YiZH~xS!H~g|WWYz6qtz|V2`_$txDT4WY&T#L5 zjaP4hb6Dd^L_==_D-nQJcyYJg3MC7&tiXy`eb}$Umv$VtO_S5AL;ulV!2jR>CtMHh zOS}+_@9mxy5S(MQ21#I&9?T}_|M#Dh)hnEbzG>(qyRu4ye#po3bP;H|_FUTX_H!t+ zcx~N$xk6}3uL598Q7D<9Wky#>vb4cE$-0%)Hh_*n>L^l=sCsxOoC7>l@GCLjDNt_fbC=~u^A23KjCSk4;B#Cg? zQ=ru{@6Y+0MQv3c8Ld8w?T50(Cz{fGMk@F;^}fTz^HeA%enRaHoo%_q#jyYQ>aGVl zZQFoPxiZy=E<6jRSkqaFNhE8rqKKxYTZxW7QVKPi;Iu&KgycPcCLFI`oibX=X-2BV zIgJ!SXP0INR8Y2{$R~%h*0T|smM(rP+&Zn}T-~aSeSnDvW1xSTbnyfejT8tCC zXzA(M%yX9GONb~=_ttS-4bB*?YK_ndt@A9CA%$u9!2kEZv3mOB%u<7DdoJ!jMN-~j zly;|_K*Id+kfOEA7N{2i+8-^ z9en)bAOB+=`+j`u#EA(0+l$c(8ZC2PqqThrrW+sxcD?))W3(Uut(~16e)i{n2A(J> z^o%Jfm1?+icnBA^*%%DCJPM1Pj^glcxj=vZH#zDX^8e-?;df@u(VaVp=nOu)c#-_* zn5RGcSss5NEFQ`jR<&YNwUphMS#gAPEhml;O|wF)5lLdOvP1!0^f)@H8MkSQGzRZI z{*%7}fPeL6c<{}vFK#j3-6vbGX~$#yIq`ZHE9C}ai{sz?2V~#>0C1JI?dc*TQ|VZd z7lBq_w1&uYK+7N8gPja%E1V7shlqmhE-r#u?B|xM5nbMrXAM<9z$56Q_9e*^HUhGDqw7qTHqzfWT`^1cNh8doo@Vw@y+sc3Hoz*s!` z1mz2#qup&uW|GOg_bsDb zpnXe?1Uum9Vg$0gbCUK<>!=L8>q>(V5_it?Y;!WjNr7~pWU5JppcHF1vYapEQ-qAp z^1b4?DcCs&w5(Ee&aq5}ylHs@=`7tcdW2cWLwA0IL%C+-yMCPOb;kf`r6TQCeEw&C ziU%Kh2aJD=tsTwoo$#K}N?>Jz@E(z7kJ4g^zvgWBj$h_SatTVcm~E;W!h+e>(xK zFK4vGaEwq2Ih_27r4EZCuse&zg86*T{ydnM7RijUpRlClC^9SKB;j&=LFzUvR_}W8 zv$*vf-?bFcwYzr}ue|aKiHcQ9gV_3HcQ<(O0J1lfD?_SU)?G+Zo=+Xt^@>D|Np#9Y zEB1R*?+HZ3QY z*c>W+6ZS8CM^O}1RduRiTeK=3^9W;W3LoZl+i%nMYqBguxE|*dyzeniBBX#K9LqO*)JL*F{E4wP^j1xv9Qz zy6spuIMdTP$htLSKc=o1%w0%PH3+n?$dS~kU=dTZ4-W(<-lHqAq9(yBScnA$DG#^@ z*%240%nGN%xR2iZ-Hf03Ueq8gqpDruWsP@&i=zpuKcF-PX|(FdvW&_oTI=!o0COkI zb-H~SvoWII?by5?w*A=VTTY*qs+#lM@yn<-Tz}Pfcm&h&O%{03ig6#e!E~eH9hkizqu}>@~OCJr;3cO04DTG=L)M z9PamihsXC{MGi)|YE5AaNJ69I>({UI#3PS_l(+~wE-HO$`5=v!%rW+ItxF9pYAf3m{00gA9_)q>;C`G^e2=?}MIy&xDUt}Z6(RCdx2V&<(8w`Yu&y>$=KoloD!bFmdCzZK0Aa#;AVAe}f> zrD{l*9#CX8N7aRJ9!#+)#;HZpVHz7!c=QYg1FEV*#)`vvr2=_;!TVTaq{UX@!24Er$DX-927^$2gBkrzV7{DweR7_?SDZ+%3V{TrK0mamqh@scH?TeU;7d( z7I-Up@WKe`YL@+g&FJJVvz*E(YUREn>X|%PZmmi_M zdmUv!lLbc0v@qS^#%G`6(RaTaGupw+6?wWQMX(fe9?!4RppZHM%)TzZ1Zb`4NU+|s zNQS&+aUFm`x8z2f@wPz^QZk6LgysY=ZeQiu)h4s{kY^)XwzZ=X61s(3Ga5Db4W#!)D1avn+}nAqAOTW3n+?oQ?POi2IDz{kR`rBhH#ie|rEe z=M8PFvNNs;=9s2G)iJ2i6tiBU{sc5yNs=&~PFXA#+_-TAnZ-VR#|Dk&$R5!|aHRU4 zOM}6w5u*R_&r=O1Oup;;0m%FCm0EXe=JPqT$t3h)i;Tl#lQX*rWek0+9_%>*TA}LD zM21MKF-j$DtFV9;v3H;H9zfpvG@n!tE)Zr=%KvR5%SYM^CLgvQPm%h`(>8oyq6)U_^Ja8ev z>_tByZKFR~T2L9q3Wl3scU9aG=w<)8m+GL_+*K}qh)O2VtVkBJdYJ*U9^Ny4org7mVs!2ScH-$ zIF*sAn1YLyeL}}}H+9A_M>Nzdj)rWdFVdKvtvrB_wQZQI1w95UC1rU)DhgV+V!sdf zM~)*^Ii*-*@U#YqEKBQiKIDFmkIE0z^c7Ah=C)=dGFm1H3%O`YoLJ+HVlocbs>sGwj>Y!=7_Eu&^&b_9?w}&_cwesi zZ~iy|SlBZp&r*W%0a^o8m^`Rn+@GdJ{a!Lg|+sS(Wh>7uGX^XGt;>haz)uLVx#C@1i0 zL?JSSAK@B{?<_*v)70S!qqU40t-Hq1yM{ET4ma4u`;I)%5x&Q`@caIx{gJm}e$Fvk zPA|a+I|4u1z}p^yB#&+JbM_Mh!&|Mk=KrbLOzvbY)}9nY&*Nk@`PX~}Vw)0id45$q~!$<%-~5g@uV!&Yog zX*xcqo%P5X}PcjJCSVlXHWHY2m zn}Oh98~|dPB>21_FFl=bh^Xyzt>>|;7HrN#GH2D!kRnzD8ZbiP+6k#oNj5IhOGPzK zk%RDjCq`iG5Fbu79_cN0ThX@_7k93(T@09)b7+p~b;=}<#SqR>OHXfm{^6(Y@zA3w*WQ?+G@&pj z(Wv+=($*N`$=hT4;U<~%7^(2r-iY4VrhNLhd7&DfUYoRE)2H`=*8R92-}LeAWLSFq z$W{wc_a}_j^h9?hB|p%=mA`2J64&Av>rx%_@WT%W;dxxV@0M#?tr6CcB`H;O_)dC- z@`~Q|T)TEHR2`mUv;dAxk8u{+8(d$5A2AYuxq2#wH+F@Tl0*+kwP6$uDhmpVhuhDA!>_uP3Uux0BejPXzb&Z5ciVvaT8 z`O_*pZwj1=EQ|3tp*cDSwCvrxh$6!^6^U{vo1j^8r=9S)o1OwKkrb?Y$#L(=BGA$s zQRqL~qAdh>9!(I&k>Ze9jxTbu%HgDA*q`Y48vfky``BE+6L;8g_qk{dg2fw$m6E<3 zlaeB&WItJ4K|B{J9STDdyF11O2hu7P*@ym+wkhVVYaDx z+jka-$VS^FK}R9~Th%=_j}J)uHEu8_RgT_Ey6J?;yWUTKxJ!2DtT;kds3H>Lt^C%P8H!d5c@c{%R*0tyrPs+8SXDH{lo^j>EO3jW!a_Mx7Tqjtn2g;2*mk7RuF8wEc*wPnEb+A$%`mWmq~U&k_U{ zV`JlE({tOk{Njzj&F9`(@zT`=ht(m!`urRt*XRhe%xD~wn;Jxi0?ICq$**xnB9!Iw zu)&Lxqc+6~PvcUO=n$Wq07#99&hiKbQnB4Xh+f4)iU_nyN-7OK7TY_d3D=-Vqpm8c z@$|?ZQkhdmMyvJy)KJy-8Qz>!?17ajhp`W+2mE(@4`eB>F3FTa*#hO}yj)Hwx`H$U ztu85O@vJQ*vkN!@+17*cUd#CCPmK87_va+NWk!e03Ph2Sm!4Es3`~Pn61->7rTl<+ zAM20pQT)!Q@ff_?2WzoHV#^&;QXH0p?|hgqeCNXupNZ6ZygT#3(yGHe@)o@7$Xvr9 z4d*T7CRSM1re$Z4qSEm9?&b|deAnp6Xi-++Jr5t<;@XX}5&&lvFx8Cut^Q@;AMcZoI?Q zXZ}q{S9(vZJEv-!k*qZNvVuLocHtsOk7GFFiw}VntOkLgBa? zffl0xioEf-&Qj+Nr$jJr1c=o;#qlu}d-khaX zM7oKOBIdhwWW-cSX9PO7w#ig?Dul0HjiKjqHe--!ME7Q}1t73}%kP>c&RgEL3c910 z4qoBaqmEwIB(Vmw9_o+=Mga{8lG?^Y@1sUb z(0A$-XesOIWyXF?=~^XVGQ%Y~uBphiB5?zhopZaMAzT_5tu&(%2_2r&L2>QPeE82l z%5WS_R7FZPm6%kJA=nWWN-5N+AgeuNQzL=pW;oaD$FFiF*SON5`kv0uL97DMk_o2S zWY|B7r)Otu@HU~3t(HOtFh6Zml_ec!wu8fw+m<{Mt;3CA$FQ;$mxdWK_K57O=CtRh zo&=yJt2LW#Np5P^n{T2U1>TVrD(O`^l9vQ{v zt_eWPuLGl%{Me5|mg9fvLx`9<_|o*DjeQ4Ov*JtRM59#{1%Kf$`~}{ywT;E##zQ`} zzso0&UW(^Yp8_pE8ZsV_`P|DdfpZ9{SQ1{GuJ4%oDMomtl!$1<00iN>F$@l(sqy@P zR5hoJmLfx#9&0_lT``-@&|0varPNfE+EMs|v`ER6qJ8NZn5l>xCT1F{pPI5})&$nbl{J6;!QbVb zZ;$D=DAv7DN74TTqz#o4YAmu?)JzKbgDjOw5r_Ur9>c+w6WeW7A&}9Y#*xqwnjb2{g~cM4(gtH zmmq0*b^qtkxyN`-7j$&xPdSPb4uW<_jdJq~>jW!mFhi;R6B z8y^^z)`L_aeS+!0MKGu=^r~PlesdWGOlFvLfNM%}rO8!-z;m~mK?GVdE2yoa$FsHq zAtd5LK^{SDCXy5P4ssM%Yc9BBhN{KKXN9_VH*LY)4$p|3? zJH_KT@7Wpka1Lmt7NpNlo3Q7Fqg*ufN|L#TBr;9KG|W-WnrSwK2(%hKKt4&^$Menm114160aM-hU154?{)pWuJ*w}J^X(B4lS-jDn7%^&V{6vOYA&vIw= zSzcXy@@zaG{(0Ode*N_46QFg{xA3b)5-Zk*ONo}@#p%~`CWhLO z`>`M6sBJ?zx>77fV6<#sGi62}Q%J%)d}KG8$(Wb-Zj(m#hvgxuZt?!C=b7&XO}oT< z!*n`DYr$5Uv(EuNf}Dc)+c-~CINT_An*tu)IOHy15a1?lz`05^((xg?%q;C-U0KDgg{SlWhhWo0EuI71j(f~KSmVG+Fww5mS^a=Pr zmg#qKVObuNM8UQdEy|||Jj=xq*7no*YpPktwL(+5H0cc* zL9PN2l6JufX*?_phfTOvYvWkk1g(0kE77uH;U~0Fk2a73??JbY#JgZFdk_}(m0!^- zhsPj%#?~;Sw}LE*ReouYa}6RgZYSWa7tw&#Cv=^|#kNh#dW_1s7q3B!MNDRNYJ{yT zGNBpg;kv%Oz635KZzU?t+3Rz9JXL&TCdmLLB#nt_=xEMSNWo@ggp+9iVjIf<09PAv z?(CA4Epe3rYk2Np#X7RSc{V|5xL4Km&PPT|)9WZ;x1dPrA{#0afotZpN|O4HBJnsR zIti>)CFdxikTQHUbdSCj*_ZKeQ^-eLl?{)T_c*+w=tejBRlC3;NSg(7>-og3n|yR` z(aABogFclw-*WMu_q>N6_<Jf`M?K0@OlsDetZMM`LC}=%X|M-8LIQK>Vikm z-SyLFfB$R0g#XyDp8niNlS$*hI-}*SC8@&;e6aTCfYvfpiV=X5%+G*Up&dy~Y00Dm zoq*{PaY6c0M(YGNd{!8w0hYN$p_)~xCPx0RG`Mz*IuBICe znP0s^Go7%pvBCHK@DG8rcp<6NAck-HB~w2Oi=%h1VYJqlE^=@0*4eR_#|*VW_z`tn zRKHNKxvM_}aGE;YQbYFb5uFFMZg~5vF9qi7(I=4q-Tz2!9k;Jv=kCD~Gv(2<3pgd& zzHM1=XS`y=*Ylt`KonE5Nm!g7im5EI8Pc^!$BJRJHz?JKxIeN4;|QeE9Mf7%Dllp2 ze`vjA(1b;Q)Hq}`e08!zMq&E%FTaF~K&W(Kq3&dgP=Gyi-j;(YJ_4;te&Q%!Q}>c0 z3uDSfd&;`3`hvAMLE*o1p~Z#>1WYFyi} zWQEZN1Sm#q3?#>u!u1ibq$Zel#%Nozr`FT5URu4VYK)vCT-y5EL4+Xl_G5^TWkL9AB6oC*^qoHsTxCVC3)gW@pL+-`;14E(wP&HzGRdxshlGh zmO<+2C`jUcxsE*+r4HAvpIt)l5$)+d}M}IY|mA7XSzyfJ=Ftm zf#U;w>_gh>(P=a4*DLM=t^a`d*`NJcKJt-|@GHOaE3fxZ?#DMEPPAEHuhijw^(?>e z{9k9??S0K{Pqx{wo~5@nTjnv`Meqeood}YzmCV}QC)GzMaRRhDyT(T_<_~tUANy6TwSm!EE`tBm^Pj^#^*M&g zMW)j!+nYfX(%Tw@pxU|s5jagYH}THV%M_KRc%^9j6;nT@!y$b48b%9Xb9q276^y5t z)iKz6;0l(qzTuZ(I8@hpp695hcW|_LpOB{+T^7Lp zR8NtqW#Sa3KOh&9nvNveK&*(Suyf6L1+tt02oc zXre|dI(AHA3#mmE%co>yNHXooBlD|<;U1aH!@V(|ga*{PYw_j=C;7&5z2^E}_$Vv# zAxmW#Drlmi>`?dgq6ekW2DH$pDYp}#Ra!?XJcHEHQlMqHFNZ~#S0NQ%NNm$0yAlux zrE%E=6ek{T2FpgSlX5ETop_xff>m_Mkk#wFugZ_BQRGtG`<2|-VXN>Df*jO0j z(}KQjadiM%CthS_d(Nfac)4v92WgYxk^TSyC zHGtM`cp1*OHF?>iWmr^1S)N)u22tIeh*Z<{B~xaB*|5!P7%kQ9Gugk%ZZW~lwy?`N1DzsV&aLA;8rL!N7jVfTZ`^4Z zb}zA&hOhL!c0t+HwD(@cDn)YPGIig@>(MeyQZBal*b&hHb&@h%dTMvX(m=lq%t~Ku zlgS)LBW8VxLvTC^jFw(GiqxT{=g=g0AwfVVHCaqOk)+T$wvz{#Bp0x~!|NHciNf{D zV$zb{3cP9%5zGwR10K5)z((UV@A6=ULh9ghUsADRg%}KQF{L<545J4Rc~LESAv+`= zMgz=*Q$}mvjNdh@lGf7|3C0L|Jf#c3U_zLO z(E(N|u1%*r>O7;DgXn$m{vXCn2Dsn)cf9;u z%X2Y1`jvndv1)18jL?aIG@(t zAq217yUwyN+4L7cS_Y%Ah-@cefmqum-!c6H-Q62B7q$^fQFoyy`5$CodgEGG~1pfL1fSfba!w5EiUOTJZGXU7$2s8ICy{3Kp$9t>|iWMj{<35t_Yq zjyxTcCPRGZL2p15G3(0EBo(#wc)dCWTA4IlyR?A-CfE0P!^eM`H!p6WwIIP#iITNg zBGMcmZ6C&IOM3Yh64=jIEF-Jdr-GauPe;|o=Fh_Nnd*63(U9tjL6+c(hKmfy7Y%lI zo22$6ZAGeMd!y(qU-|>=DBf#mu$n?ijw8Dg>i{@t1YO^wHLN>_ja7apRfG4s5uJA^ zuP72QE@d)*g>^b&kro)O@yE~e@U!9gdFjd2h&-uj+>Ga+f1Y~n$Ocf!$c%_JS1*RK z=>&}E5iV9EmK)f0js%=cNuoHuH6@2-$J|FV)dqM)pq=H){3aV(3BtNzJktF2$m0A z#-s!2+t3IbPYehWGw8 z&hXy*QzO0q&^Sx4eRXGb<2q_{%p|>n9roN?e-7V&=h+7HBe*mme|3+$AO1UtzxpI> zzjhzv!(YZDm|y+Np}W1*9AAu;W{V2dd(?ahJ3-Xl_u<87G7-e^+R=9&O zHb@Gc2t=xI(b;;yklqA~h1Ea%U-7dG*zX&I^Xv~N%#)m=T#)+-2&hpUHA^n%*T{U% zT6AntFjNbAH)1olB&Mc2xQl(_od8r_N87e6nueiL+$^rKtvh-;#AVyUb=^oWy zG#;Id5op@Hpa72ZA;+CX4<`(&3P_m^1dZ|NVt@yXPQi=NSQn49ppqPyq=-&n%n08$ z2%Vrz&q{WDf+Nr+!{ax?=4a5-b8#aj-4krw3fizqU!t@i6P6=NYEko$egtF&lz4p| zp&R~;-r%F%oag25^1Vps{Q5CN9 zMw`>6`#3<`08DyXsx;mpUV%#(Xw7~^AAjyGYknR$XcgA9C?%+RhZjA2uH(+5&r$k| z9JCn^_yGKh#AC=AwZ|L`w?PPwMjK==&DmZ?rm1wNb^Ysvfk&Cd>deD>1r%hYbWTB<@tBBcOmw(WfoxV(I;HJfoRZ|GLA7UV*Pw4v zv4CMC)i@7@^mMsqZ!ckE(6SUuHg(FVi+vW2!K#F1qsg*47K`_uQDV92GDbomeM09Q zM@`8Cg9~_-AiECD8PIBn6UIVt``Qp_AYGH{8J!+6?svg^F0YRGZ*Tq#2Nz?%#Js1k zdWxjPISs4F5j)Ru=jd4;+4+tyfBpfV4G_!vjM2&mux_b485pbhURDZt)7#3m>yV!2l3D~0B+BJFN`gZ8ZBv&O@)kK)j|RN zS&CA(`)u6&`)7Rw+m?6O@8F`}p*2go)-z|H7vI{G#I#-=+s53>kI|bC!2F25^KjbE zrA8jTg#4fXYvg|D3wXk48LwjMP~!U&lg4oS9olY>czG4f6y-Vs zz40b{(Mn;1+?fDT7$9KZmroyV2&9HQ1sHI7$g$s@n>5EpL7s?DT8+a;+NY~2axIW#p%BFLqq6eVj`Br-wd z11cjqR?j2b32&AICi5Lue1U!M=rb9CR>$%B`l)7VfFu(eWTY^>hs=A%evDBItO?q- zje3X}hITIM6lnF7{-iyT#@n1iYi_nF|7!2O+&m0(b)!s4EB}>;tF=z;{@zo(d4CP92=O|{T1H*?sqd8Zd0`jy4de< zE!GqwMF3hklr+d_i_r;%j%jiy6RzqB-YWJE_Ic{*Pthl_y0mua-Ee=W>q8z)LtoK# ztU;J<|0dt_@Xz6JjH4gi!HIX2nUUI*B-1?P`>1iWh)H0IN(R`b!X^sQ^`YW0#;|Kn z$G4elet7E8-Xn(@hgHHx+Vg4oyI2Kb*l3@9iE%Z zaKAd0Ao>n1t3dk!^J{O#{!jlqWo%g~q$e30Pc!ygtt%aNwRH5FbZD5dfI|Lnbauw~g*=J(sf8Sc5$yKl;Qs<}!6 zm4qe)5|YsX7H$CzWr#MIFoEq3j4=V>hz^IrVWPv1Xh*pJfNlqLcXZPR+vV=Y(6kvO z5C{QMNfeT*N<&SVSy`F+=KJn=#=W=y*!$l1GHVdUAT?I5h>Z6#^WHO`y}$jfZ+#1D zCn!s|6uc`fJfj7;pf#+&${MY&B%}4sk3Rb7^+z6gWE1%CBab}t#YZ1~^kc_A5CGVJ z4)ExskFvhLe#-{7x3`z&;cq?nLtl}DJ|45#%p?9zhuLh#$3FHke)1=Oa(PUPNRs3w z$9^aNYC~)Nny&G)yMKrjio4d|@tXEih5mD~Zj;v)E318C)q2V2p1ko1{`BTQ=hDfC zInlfSRks<90?F;^~v-9XB-$nHzPM&RZZO;$xW;`S6!TOm?>^$!mNULs*19!mxyLUI=(}%H`bE@U2Z-na1 z(3E$c=eXIF7bl+ad-I7t?dpJ>oFt7r=MqZ z(!n=^iCZL$=?^8@++XWSm65Tx2rl2ALxLF%8= z;ra!*`B@GxC5-xx!IKTWpnW?S_gGt7qqVbv5mS+vm~uWlgGP{L8g*DBgB3+7 z6e^^mjEuM^KEb3>G;V@c87pgV2l1QeNr@i&lyWlJWmAsW0jl7tuA>=+$*5^ah{>Cp zQQ?8smECvn_CZdTHfSB;Obr6EjWd+bUuPyVL{s{-Xlb#t<3MYMvY{Dz6!9}>47T5+ z{m}gkKb3HwvKZTN7|n>nn5L>3LWt#a6T3xaz#!3YU+O0lmrhqa1ewXzJPP5xvL5FcGOFOKucVBk9 zUy1mN0j))kNhg|R^Vf|edZT{UkHD7dwYAgY@;kolzORH?k^fzYns=P zyxrQ`;)j3uhXHTC_?_SRoxJN^?|R8L%ChA4W4{xBwXwdw{+h1wC!haiv`px6-)q{> z!gy1K3ccOkEfW3G7xrM}*{7$k|1qjsd(~}DW_}VsobJ8qIo#ayY)&&z$V(4Is@-j` z-7yW57q2qi-9;USrpRBq!tTyCym$rsOFxf%*LykfjRxD~?Cxx1w{9@G=PvTLz#*7j zdy&De#iTI@Li4$2eudbD^7a&LJ|_|mTUf*?&lJ?Az*R?#Ry`jP1M|^7aK<-OjwG>* zLHXmF^lml9EDrZETF`2rXP&gc8tiDFq%|P7DYMmTbiZ#}G(mc(d9G>BO zdWso)bdXfnXXur6vW}m;#-l^7|N5is)1okfH(1ExghCvWSc%;1a&mu4%Y1>(d`j6n zy-YJ-)*Vji7Tc{Auk-8l;qsWqoFdkNl}jSDHFSfwqKH~FhX)9ti3JBbXV5f62Oh-8 z6>l5yb+eNsCPKL_(#bZkiHQ>2{3aV$kJ#6ClM@Fs&Y74o!51XVWz2O~B81?)T?4Lh zI6cAs{tuwme}Q3Ck<w8gdk6+;YXZN}A;?q3yZolwp${FK%&U&z=h_e~CjzI6BsA#5B4ue)QcZT7# zroz&fiby%kzGsKG;w}!!!MXe43D|!_i+r~rs-a`2SSJ}5ZEoJ!Ww@G2j(gY_ zd${}p5faSSLg28i4#{+9+4OrR z4Omcm#;91wWi2pM)I7c}%QjL}YU6e2z#n%GnGZ&TD69NEQ}0c|K)-zk+Z3 z+fj1ikNA_(d9Ud-lrI_i~$m z&eY+g+NLX4Ir+jq$`~TmLQSTyegcvf93Eg?g_z6`Ns25=c>ZbJ?+4qF0O*9RV1L(x zvlWGNJ*v2e81B*EbL1!c2q732SLrA(@+uQ>MUHb8n;Trm|GSjNS3QOhft?af5FIa? z&647@2U@d!FvAQ$%go1A39Pm*AqIZ+8tX7wg#Z8`07*naR899ebv8ny5O$kF`BZ`k z#)u0ws}l+}+C?M+SshhtSyNzw!@caMr1$nn@9VMIshC9xP6*88kW$vPS}9g^7;?a> z8*oU?lhbvSYcUry>as<3eU3VLnvMt_QpPaObA*s==MkrLO9Tx&T`JnL{4Wq|JdV>4+Am z+Y@G$U{tQLcW{CCr5C|LD~RDEwT5n1c#G~~Jixj7sZTn|9)O@XbUG5H!j9}`kg(B6Dv$tB&Dt) zt`JR^5-2Xeo^0!Vykev+bW>pRIkMG88nbM*RaFiXQioE7P6O}z5b|5U!yn_npVHJMlFh&RMS)`U?Q91+ZGUmcNo|X%#Bv_U2E2SsRGP zEyVDz4z#A#HW%dmw8mrN*&HoX)OZg40j#XTaG$!IASP4V_rDP*1n#jvz@6g|Cn4j>2b5_Q-VHt0w7^-5EoW4TAkjStlSrHoNm`Ojd3Ww2y4P!I%;=9f1(YF#8(SV>QjOLgLzOn(>fQH7F@S4VZ{AEA(hY zgOn8!Ef6`wn>}1%;N%HZ7$Pfc*>?&p+kCN*tfn1T1YozQr~+tGy98Qh>I(_< zs%IE=ULuB`VU*6&98|1jHL5A;y*Ow4T#HJ@befz@`+co_`Wo%0Lcu}QVk}BlMT?De zmluXHT6M?^$A%klseF)>6idScXa7S?n1CB4mQG zlGL4Hzr0E__JE|b@9(cH4v|fXkqPOue~|m5nCGGDT%6+fDJLouOVvcqaq@jVl#OOy z8HCbH4b+x^tR;z67pG>NIUmv0b6nz&w}@9LPv3`dmdXBp_&0yKm6}sF87^dI7G9gB zwX~&U*2JW;L`E@=u&C?%Mb2C&$U3Kxfzg^7_S>PSAP(mK9n8M&Ub4Xv>t5In$_1dM z1q8j4wiUz)hq*tN^T!yigDRX$6Qg8FlBWFD&Fh#bMY=hvafl$kuhp2M_KO@-#MmmP za)%^g(a>p1%FVN+*@SVHFnRaeXc9r0G}x#@H9ZO}oz4oA$xB;zEyAT4(i!Y*gbF}w zaUJW*Ertk=C%yDyard=Qb@=F`k3J8);VZGPJ8>s&kJ3zW_-4n)!#Rr(ufEo>FJZJ^ z9z-qJGUF=t^Jm#?KlqBrjw1lPv1#f&e7Q6|;`YicM9>O>IB3CB3D#6JlM$l7io$=V%I+pJf5w|$_nkER zpQV`$sdeLf*y#ZivCBp4n+v~T?Tf%07|4bfwWsC+7W)@=Xht_+<8A(#87umy8d_R%F1iAh zp>Be=S#)IF=n{(wI?!g~ftG3~h%`eN6>=o79_G;OmE3*ekm^N08E#!r44&6?O-@Vt zN%X`MLrm8(d;JU%X(~~%DFj_T=H}=ut@t#tvuw{D=hP|!fpLbk6H^-htbBYs;|)$g z3sRS1=P4xM%p5{WR~0qNX)>wMQJdNcVta<2>QgkK!FJ~$RJoQDdPZYag0L-8T1@i< zj9@k0Lbs1*&zZHPHpd7{5`eHh7(;J`?!W`klbQc~GmVyvRy3P2>-+8lSD!5rO`k;e z*-vXSV|?#DR}%?MV@s55m{*FTw$$wn?!9;sp8FJ)jOfIQShcClj5oeL!#?%`_VW=E z81*h8G7o@?18?40rxrJJNI9o$S`gloCREB*&e2hD)x<=?;GzU4Ye2&Jdq79*w|Xd3 zG8d6w{0PCcJ-`gFqMY>}d*J=ojzy=}XI#caQem_}G;rT@|B8s3I!-t&0?=yb%mdKc zEfcJB_NWjf5NU38~|9S@N>rf2S^9Y0Aj z(hMn7iLXi&jh~zvTVTVgbZ`C)zxu_$%ijE%SN*QBw=r7w6;g+GPB%V@K(O;=(ubxg zSij!91o+*)-+8@D>ii14&MUM?i6Y$n*i`b93Kv4T4BgH=5W6;W9St9wWp08dyxy2i69!VKvT z6y30#RMU#~q(UD%$4E*4y{e>61(9k)r-xITv>c)%LFx>JgL>C90iu2>qm`8V*jOMP z)J^RfEn8!zBgD$8F8~1QmQO#P=tdMhjhP&hD}z)5p?XZ1ph<{b%o{iSJ?nMvN9-3k zDKM*^-D@mZ=O~q?b^Z$K%s*eAQ%TNVy~2w{Nuw0;g*J^_V|1b6-_*N|(gS#MgLnQ> z7dJ~t?|YnP!>6p=#Ic&zJh0KC+iZaYnY6iDvP{bWxVS;aenOc~y%tcl6e{Z==M`#{ zbE7@L0?l5@z1fiR@)RotyJsDRis>{3$(Gm5#DyU#EE6$Cg>>Rc5qc(;ZzK{iiIB`1 z$?28{c*WK0T>R`Wkz3D{JUQ-h=pP^6JbEL^i@qA$R!55mo6i_9po1b#TGSSFeU^0Y zmJ~y8bVyU=jLMR>nPF9Evg}&m1a_LzOZOS928VPMb1bWIC{hV2G5g{TICFuueui+C zQ{=d&p&lvn0JIL~Z@{iRh!Qo$!Gx?M8DtkYj3*THnnVhOt(&gEJ>R3 z#@Bnt@dr-#dH7VDM0F^`vnKm1nAE50M!icA`|D))y?0%mT97$HSPxc1%F8Yo8K5Is zI^fGBL8uB7Zy?U!gZxMTigDIQ0COU2VX<>V-R@Ba!9F2Pc{|YR^?K}0C6SO=4K$8> zF8?|>3rR|CBnMT>nP{6VS^%wxd1a|hf)E8sl2Dc2Xt5yz?Dz=iTpqH?I^TJt{)dY35Uh${W1a zm0PxNx?RStwfLGz9o~sM@l}fj&;p2ud#qf&#;CkbZRh;Si$71%inL0 z+T>)>8iUrQrA}-8OV5(+AH3wV{bzQ0_y0X3+Btm1{m$z>9FE!C4=$5gNorzLZ8@mB zOjRE{nPY~Yz0ti@%DYd4l!$NrThQ*{=EwY~#zSnY1rpMtO&m-kofM4MG?kg)a6~Sl zae^xJzN^PR<>;~wR&|lW11lFMN)^&MiJCE3)pUaarr7cmj50?n)K(;riAB^kRU(L0 z8zCjGw@O+Li6TLQrE(543iqro`R_h_m4EWd>;C!+McjXDVz7=h zGaQnt<;B5He1W_tG1DxjD_;Qc2R z|NZCwDZOF@WsWGIP6S4QECoUc9=iOm`KHHyh7%_htLa6?@_=dGkP3k=Z=i+t79+xt z*OGZ{XvYe|qT)Y!^J(r{O^H

uQXvpu0ky`g0idE|P`<46*OupDJl^!MxSjqtxL- zW7Ob#Hco>LhKO4TUC_in%`Aj~0I23N2J3J|!^N|k=y(CN0`h%1pfyv(IE)gA#`(rQ z)ZVj3%{6TdGLfR>FjjP4<2 z@Z?$ot)R)$bGm&+Cjc#X(*q-Iisd1}$6wmCeEhUpI)b3ZKNm#8#9%KW$;BLtb2cJ%{={ zr#CEjtw6sy&wPJCgoE)Ra^zEyITK{NhZ_2{nXYz;LWzHrwK;4YhvFEc70pRg4?MD| zrHLhy0@Z3$Zx={wIY`Xn0=C+vrw>%tmyKrXv)kFu2NJ zuNXmnvjFE1cRfI7(C6XdGcZ1cB1bje$g;=;(cGuE2E(Um=R2J2%^0*U;&AMDJ!n&A zj2019V5;VU(P~FWzHJL26KnqtRb~pk6^fn)sc3DUWJNWItcx%WO{_uJxKM?gMZIOK zC=md(k&?D_Oq&>-MCu5k3Yz%j@;MN}x2n)b&&Jxxb>g6fS{A89M~%b51Fij;CPrX1 z2xDo4@&y_A&+hU7mOU z)w?*gb{7suu@dvy!M%K9{q@YXf1lfhLR1xzNwKLPTW;1d>+LVEViMUA(2}h-0+x(c zRUV1q)$Kq_i?5wthj-#me6^xzj+i4|RtV>)nkiB!-f;5!xG{O08`D2~+3(&ihJU51 z!^+HwWX6g1LpThFGjGix%}WhVb?(86s%n~7Xaqg2ytYGxRv(cuv_e*Zt|LYMkKlWWa?I{2 zirXm?FS?+nQ7?%z0Si-{ABmka7`SU2v6xtpB^z~!<-~EHU_wI(xsG!L_ur_Ig zlNBQNFYu4P_3kC3r2^2hQqoZ{H4(4?N;OS#f<-KV&0HPn_e8f3!4P*Dw=r7%exLoh zB1U4gKw3+o5-<(BjR!cV+KR^yy!A~<3+v!u-cW}|-P!c`@48G-VPusu>X+qZ(lmkZH4N*$`y zUwf+*aUxz(3=jTx-ANt(4U5re#E<;QkMOQ{y^C-8mT%$HpT5%{```B{8o!doK^!ar zj46m@n+yGK1Ylmhe1qo&gQZ`A(Q+?d#y$NcwegHrFY!JqMd<;Gcp0Fj=3$$adx^zE zXM2bKwd+i(ZJdkfh7?j_4e88pt4#WoNdQ_09%zMwuIN8u(79Uf^zG z`e~c>H4nsE>*p}0TwWvPGzNhSgPIxn2`6^Ki{Zm+(S@?~v4cM{BkE{MN z@1{@JScvi!Br#e@Ts21kCef&HZgtx;TB(>2%^;Fj{3z zLW-LBLEC=f_Ynuzc)Z%-u^Yd@zR1{d6=LlifBOC|?aSM!zyIU(#zP)|Pl9OoX+8KK z0WiAy3|#kyuzA0Ylwq5-{}~U);3SDld0Kvv%i>Fn%>;+V0iD(+`_%=c62uC&^B&gL z=upZpWQ;Rj|HC(Z8_}7&7=8FA>X|>mh37I(T2#()Z6St_V>E_dax@W{I08|jg1-`Y z*ZEOv4J^zaKcG$%<}04j$|Fr;YIG|lFM|aJ4qd`?@>xDyKaOx6^74!aTG~>T4ksdn zRWzf0pJu`rEf|&JKJ^JI>!**gi|NFX?>ATY`!-Gz-+CqQaaN{8cn9ygar7n{Ivv_4 zd}_DOC$vJN;Pe`W_Q+*Me3*{!h+FKfzfKFSQtdb{0uy=1hEm~r^Idl-+8Z8-KV zz@2au@)|TfqDat=LsRdL<9o0EE|t_=$!}n+qH8o%CGmpjcak6crt^H?*ME8Q>Dw7C z0H*_^b4A;f1D3BTQhl<_*LyMvVh78;?+;Wa>(+ zGkiRMfnD2>>DRIKx)rX!?{#pn!}as$*}5xYS}W3xlh{^=*$dC&uK5(=w9`h5MRjP2 z5E1S5F{fn2c_p|}ewN?Wzd;!o$Eog-iiEK{&*JgBF-dUl$X#5>1lvezlT*Z+jSqYu zGSiHIM4zXt*to^0>+K4for0}X z=WxNry@)-al|oSGM`?61^UuZ?$Dsti8X#Rm969V{gtZc374`ns@;j|!A9LRuC_ego z7$>MKTv01zae|@37XChsyME$q7co+%NM|XnZ-vqt*3j#s2kYK^S4jG0&Zm2CA=&ij zAS1=YyH|Ml>P>7&QT9qr7PMTthhr5($qMvM41pk-!DQM{N;`RLAoKinR$t360^t;7SZ zICL&h-DT>qy)xi~AN(Nw^)*rnNl`8PD_Rx~SuPL1j+RzXm84^TEtTu?;E4|F*_UUu`vHjKn0~+iRX(aav4|VjYM%N`v21X> zh@v?pRviZJs9fKel2~P^poM8nf&FLyEAH2R8TWtwBKDvCL;tx~V6^Q2^S|T%<8URIxL+n>H z8aD4rXq`G*0qU+lFBwj|%q=qYul95l;~EQP$vkV2)jET=rZR&0=?!EQ(=~0%W`qK&8L4=jKGYw8!-xwIEg(3a?V*t!=UIAcu-mf~7qKB|CN~ooQ zqs#oH@$;YD;1>G>W zC5h^?RXu|fj)B`mkEh&t^9kfn{EvL*ZEwUm!K_ha!XQsw!d-tBd*vD2>ISmYUjnU> za3nE^EN9)d5$+^YGi8QhUi$&RADtws1S)RPq-PkH5xe68hea+@J*tH>imI72lb~8{ zR{!>!P`!xavp!WU;{*|Da^f@ZFEiKB_#MeMZrM zEHHOb(s2K9i+e`fnB23oy)vQ2t&CQ5XmEK0cR#Q^FM9nWpcNK6MHYMegXg~lTEjyR zFG2MpeW;StSlL2_ls0%q3!oh-xJ{##1fW%#=~B}b=N79JpFiv@87&tU6~sfN&}f-b zWf@hgg-K%WzWZ)eNWJ!qK?Fw2x)_ex&!~8jI=l^NwT_M#`SP7pdi?V)Rm5>ZD>UEUnbnMnO+I+7pZn6q<+D{}wOC zSGoF7#C_-g7-Kb0Y<&wlj$pV?HuB?MEBf=#pE3)JM#M73>{Ex5rtuaIQt}c}z0%^2 zy~E$6c;ST?SX*1;Cw}55c>nv~&rki-Pcfgrb`#7y@z)b|t1fV^(L_yr0>OQs! ztx|3ST8%A9ieTV*Tk0^LhE*|;<`a6?1u2$bw2c~;6m3a&yUiy^Pb z9WwFzn2+_+Kyucu5Y+}<<=8GX=T?2g-oEXTHzQKrLph5L*Q5@WhbGF1tO46v4%F5f z5O%rO0!tJ$U~cOsda!}%hEn}1L)Zv~#)Q<82&M4j(l7}PdF9SE+F^UjhUT-z((YVd zHYzTKBJcKTLh8_!4Vsu}TzRpuw9J*k;dp4{L&U=0^Ywj)p4T+<5#phTsS}5@5z|_e z($G48FS07&vyUOp-_7eEehbPB{_y|+AOJ~3K~zfmRd?nvh>(KS4b5cZL;T^^eH@N_I$AC!M%BU@NoD3tg3Sp13#|RXUBn-JH@h3> zQ0*==ZLp^#Hy#tWe< zK%_bxod0jI$2{;%s7B3Vd6YT~8m&dju`<&oqZJnp9TrDb=%%np!&^NJV~(K7TI}6_ z&r+-9LiORmS{53ug?a5F7S)BbXz@HL$lB04VtQUe=Y&{YU1ejs!UIW2xxLr3R;7y3 zQAAzWR8>WiB(y?B?cTgbOGvdM4*OQe*t2>tV79Rajo+pSsY8GVH$KFMIM3t%L5sah zf=IrPv%Ph1Q?j>BHucxC(|Qd+tJT6GsG`pDJY5jOgU@K(E`$ePb$Ex-`U^(y#T%?W z^ZZ}x&&n5{x#jP@7jL}kK8-Ov@x&AS;1B*FzwPaOHMK1IMAw=1EmrP@s=Ox#reD7#Az>losN&}9TpXG3o-nBhB$EsvElU$+1Ppm zPN1b%mOx7w1IJRCjV+0f0j=YvrOEW@_j{kH_@R$H21U+|#=oY~cuvt47+0{ip%Cp* zk=Ps12(xrCFLJwNK&q+4&i(I28+y$NVr$te42T>^Kd^GsUF?JHHyX*OzWDwsc~i%AqDaS-^~keeLq5JqBJ7P{KV}- zilh@OOkf&hw!vecDA2tDPDc#;_mNvmfev zhi|7<*ECHvp=Ig*v77|AlFZjmIpMAqdavl(lw~zHusnjGEUi@ z8MgOh3Z>a<`S-$9FQChd2evbhJJRo73j&AEG^Dp{!(6?u|xes7nP?h4vjTBatx zvcf&j#KbNm4bQkvD?ngwR35aSJi9zEdMn;zuE^1Ku;}Q7d$4l@aqbf0>?QwRUcbtE z>t3XY=q4wzx`$Fnz~zY13SxL2Eg2IfW0Z)g%#^A*4zxn0W;Xz>kn(czkVL(FW8+0C z@$DSo>Y!bey<0vL2l08n0PY>%I_tQ<)GGn4m6a6^hZ&g*%>2YZ2c?WilK2>-6)zdB z-C09BQV1bPf@r_Lv9a7{Jy>OSViVc;dnuY8q4*RjByaD1Ke}>cW5@SA^xgb}4_sK< zNi4bsr0O*QEhT7FN)>mP{$mRDNda_ z#hc#rCICM8!4LAp6Hk1VkLpe=qInrc%Q{18rex6?tL~`TvpWW~v`mP%!b<=2dBj>6 zO!jt`z-q~8+1hKs<}=7T>{-ty-Ogw&oXiDa>uSl29S2(aIM8yAcsj!b-=hA$MLCC= zPuUJsMv{A=Rok4wnnrXMKx>s!8WakLp)4#b`~Fy(++pS(cvLOHN#WBO<6;g13N>NX ztum{3NL3Fd9CHJQGqYs03f;j0swgR4F!8JnbzMU!|1X?HY@TM~#^|y_W*wqv(Sie9 zqfuy@B4{%b%Q#fp&2%4!z#SeSX9ndWbkM_8LEuU-F7-nFh%)x8ei70+(WDGO%P|c= zs~xQ&#t*VG`~mc2=$+(k!S1Za$Ql_KEdj*sF2jMJti`#za3UcJRfnp#!*gQDn(bnp zrmec9K|AA2n}7vbZ5@@UAD0ve-NFG*W;i3MZ9&~MY|rKh+s7fPlhbs=WR|8eamP=D zbpWerWN?9LF!4J3L+9-se)J!(z40c#l`Y=3GG+Qr@4)`Y-$+}+AS~{ypz||jjw=e9 zz)n=-0iV3~9@->SrG%mP@wrRq?MFngT!-I&)_=#&d=;RLVv_h)pp`}b zScZAy&l||HjC1GCapvM(M;i_MmBt^ZcHLXjcCGFi-uo+4-t+GSO2+>G!ZV)C1JKg9 zg2`74DFuyAspBp-bV{(t8Lhc}i4Y#3(^af*et8YpUuS$pTJZdJ#Q5yXcgcP&p|z%I z{`^glHa9mpefl)dKKm^9+;a~f|MX=0P$*pY~Yyes&0IfJ`m=yjT1rc@H zC!I`jlNM>#LpPpb3|eMvXc+A<=maXQF%n#8*`cO6>fCuSJ4^zoZRs_}!;0RDqIu?! zVRwbtRebdF2q!DF)l8eCCGgda`&d1=2Iud>$stlYeq*-9XC^%!xwMWcJd@R~=jh}# zH=9F0>4#WJUmj>Vm0@M^|>G=Co2Ypwf zN_5hK=>!w?se(W}iCZY?I4LU5q>ew?dOsUyB;UKb!v{Ba`9;&?W*|IjQxWHud4G-G z?hc(g$2e#=IZZJ~bk^v!BlONyhW&e3ONLB8_c{OEudSmuPoPbSoy{nVgtQak;(@O$ zjWqMRq1%*{S(lO15P3G$q&cJ0ceC~EJ{RWxz4%D=R-U}l;sf8{8J?5H0YADq<}GLZ zc-lJLIl8XJ7)A$F4&-%WDk7bwd)k!;Z%>>}UrFt<#vy%m^FU_K)NH{`^2|Fc@I0U>+DcG4;<^lC_9q`?8Fd z13GbFQKjU&zU#YipZ^R?|K0%nlv$wxB5q@++8wf~TY{_O-$`W$a3uP*X`HKMg(_Re zYK0R5gdkp~4jrA|tDByFwc-w=^%~=6fA(kju^;;}-uvG7^3+pL@nb*sV_)q<`kDym z9PY}ix666C)ZxNIt0hgX=Z+hWh&{42rDI8_;YB)0X_O? z7m=wxNcM;*3Ge=E$sz$}|&a zh-HRw3Db$CKTv!oPkB1K#FmKo*H7&uM2U1U)B0#)`h!dF;t${RpCQ&ZabdFk&7tJ6 z;a%LNQjAT55y5XM7E0?>(K za%1SE1v0KEwWLTr10uwcKBkZv#`*q+%38XSzeYW6F-=U8WSov1Qou=jmU;oSnt~*E zRIM&1PU(a+YTH)SbN^m+dnxkf6-K>F$h+=e8jv=tn?#$wicSo7DRN7aO5C|isCEx2 z6|=e~H8pk8Wm;Ki#n{~7n&Ca z?Jjk8_cGnonR|-@w|SmhV&SE=w?z~*UC~DOs3Py|CC{#$y|@&_ACKdps~vz=76*F= zeJeBK`yG0qr9@10|HI3VM3LbCWADv_B+Jh0zTaKmmP=J-E!|VSO!sV<0cOPv2MY;M z07P&BKoBsAA_S8ThfT^1Q8t6Y2u0AQXj-E9he=7MZEB+wvTTVYW24^s1N1JIo!3O^(9hg6P z_K)aB6+;76B%mrV1OFMzb_Yo5V`#+(=x+r8vY>deu+%o9CC1Dr8qFA7vWwB$sU1(u zALAW>4U26{)#kAa%Q4q?A}|u_~ih8B#mUTb^~k*(yX8T^PjkAk$mPO+Z!0H})-WIRcnjJ`n{*;c zC#<)t96}^7fRI$xm^gQg<~z7nhfZk6wy7va1FVCN8zSnOow?gcj#=_g{}CvGnO~j) zt*s3TyUOwbflU&M)6b#j7Ac)0F%@|{$FMZGG{KGxE_hTr+N;ZAXy!VD&M4?483nwHk=>m~Xcj8@jhjt5wo z?*=G9+O+DFM*iggG-a>S8AdDG8)#*j2UK;KVw4Tg>a>XUm_fMFOOk|Cd*JICt;@X< zX7(v*hFxxFTr87_C9bG;cQ@k}jw$XP;{A93V z^~6|H3hgypBfsctrKwBHsFTv(vFPCDo#_SQ8K9MtQko&7(lQtPwVc*O)zn8T9t2 z&t@1LsKSVSPYjulg3W{Sckb=s7R!J74Bz+b5#2@ag5Aq@-?K_7>2^b~uxwN73@J4( z0K2Lhvl*t?sAc~PmN>vO!qZN0d-J#nsTFzk0n(b1Pj6K*HE#F3W;Sg%Bi(XD}Dn>5*R+iY@ zpeh|nCNcSljgS5sTmR|zvBD8s!yuZ+8n8)<9oG{^D~%+b@deJOi?pPgw0t~OLZ22H|X3GIeG^b2fzJN%xn>;_sB)6xrPQDG24i5QOEiFm_9)9sEN_rdBL z=XO1P78uP1)}|dxTvc8HWEzcDX$BMhRWo(iFfFsK$i@=l^0EQwC20qxSE)DqSF|Zx z?G{hA@1gD;MWn5XQ@lv74YY(1%+JqLhxbu|!R)j>0IPzQEsnjGs$pl_@nsNUcAo+W z&FVLhJ^eotN4tTce}V3vj22;t=tfH_fsO=KyT!N{v|*Y_V#DP*hp&~o3AAq3H%9FQ zfK}xCZW7yyZ&})=nn?j}fL1F+O|0Vqn`gBAg4JCQqvE@O)&aP90qZJc5X34GVQmFN z4|-Ib65EPlCm||92CI)$&l9;A6Q&JQmttJWxXS4Ue=WDTG_iLO#-OW;Dob(BG1{I2 zEju0}m7?f}K-yNyTp1IMM&N2@{RCD0^6 zupZg*#I}KYOqmJ>qkH(Z;=kaZpL&QR-Bakc#f}ZNO=y{EE5w;Fr5M)KgT*Fux@2q} zRjd$IPI2x2y1}O6a-aoMot6=q2U;Q})*-f~9#IHEVFXc; z)8974NCs8Kpg50n0+YvRH(f++SjdJp04*a>)-WGc;rj`hvZO+?5_abROB|93GRZLg zODF@SnnSd^bgIyT7BzNUuvJE=?Q_uT&{$Q)Dk#77 z8L)=(+zZIy7`Z8XLc}6P$t(L0kjLz{w`xas=NWwIvSUw2WP8coc9E^uN+x5c< zsSSfA?cmE6fL3d}_ABhtbP^7LCMlsTP?7&GDH$Pu{C`0F(BGJBCRRO2CggH9)wt3F ztaMtmW9i#zRe1%Ym8cFyJ)o%fV6>t=7_EI;he(MNf|iQGIkx}!_ja>KK&&LE(_1I* z`O_DjX&qj+f3r0=H-`;ld9rC0cUu$d24M2{%@7^DrRp>#ueT|-y{D zAt?jUsxmkJ_TM-Sf#!0EFmNs^e0CdXD!^9RrU9Wrg{7h?wH zdFrYIP7*O)RBM%b#xtCsyp||uCX5z9OB*6R%mf&TP5_C zCSl8ynwxP`4?`?Uc&28TndPbna_-!zT~BYAX@K{=ZiQcdPtbJj0kmS(#$gy$8xyvw zVYHgop_&N@->-EDFkD_@C)J$$^Avmc`sCsUh?#@}rZpd}u76mjnxuh>`d5*e+#?@zho?(~YEPtNtt*5S>% zS$~1r4)zPhc+ZFH%Ap%=(u_%HUT>4?c~n_3QYCqq4w}_ywJr=9^z`lv_TmcUV;GN6 z0ca^1VO@;~phqMr^HI%cF8msW_GYX)#3YPeN#!hQ09w)xS#5IGmJgx}Ls<{e0X$Y| z!l1f<-M)m(I=D`cdNf4!CG}_kQn8wN@u&c5*Crvwi4q|U8ik`|u@!O)lNPER;Kn6+ z2JLzoAp+2fD)Op@RKb(Y-<+5{gkwL^qp7>;Ya5R;RWl43hA0^S+kP$Po#c2owh zrqB)>%Y@r|8N?cGa@vgsrpPH1K{d|kmxk@K!|9#lbUT3wb16xPwW;C+X&fse@~vH< z6F`N*V&jOY8d9#GAvt)ILf7mBlhMp|+1WPCFDTgFVi2`Z#$Sc&i>Gn2MufoJi>)4m zgDWg(%O(lVInHbhDXrw-oT3+6hZ{vq%a%SX@7SG)L$}fmcl?X#l|bu`JMQ4%${kds z9wY%rRztp@852bj7vwn(ut-w2No7pYUZ-E4! z`MBnVKv1@|K?IY|uMIrNxKG{=Nrt-QS>Zw~K)Xw~_PoP#IOQW?vqxJB^5A*)-{{aA!!hfy{a~-0YcH@;m>vA!CIF=c3*R<>QVze$Vao;0| z_kRSAzqIJOLq}tlR<8VAizh(qCZlz;Zq`dKH`6*i+y5eY)!+4aRk?{aX)mB9tieh_ z;rv~^MiYgojqJh>TgzgSkvOil$QS$tCE6Sy!yTyyN~62bbx%#)A+HIRFsDp%6W0+>u&NNW%~BZ^GY!i!Z+91-b?!lsC5iFL6~k403_ z4{13hLa?)}_{Tr+H`zLnF(!ogX6Hd^oT)IvAcFrwRE3f703>xQLpzH|B9tlVG-BVP zph^T~(WYM*mJ`j>7w<>5{KeC_j3nf*m8n8nSe6mefD=r%v%*3e(;1xuQxhLNN+oJe zD-QzPu#Gid2+#H=1D&CR<94-i8mnrs{{Mqkm*L7Xy+lz;&30}$cj_D?6VY30F|TuK zBDPCKJA@63x4sL218q&L1jl<1z&!3d3Y(?w^8W?#2cjW9@-y2&*dt$oDsGtAl5 z7_G}ov^&V}{qPmR*MHH9lADay&AM6FUzY_!Pbx54^=Ok%zVJ&tdHxS}JzSNJNQMkS zI|H;#1GF4YN-P$KBN7QpYFs`d-WsyKBqye#pFj7n`C|2l07P4ZU1E6aYQ$)Oh!cdA z%*Pc~R(YrLjj8ufOiC#Va&p?jUpx;f5cwFAgy&uiel~fI8t2qW0zxn-&ZD-5==nv` zb_X}^W5!^V>4yLSAOJ~3K~!KzJE(S-gZ2O?qqAO9WS5XgjIiMP-af$=bmK6g7wiCp zv6Pvh6XIx06cI%=Wu4&EG83Qq8~l&|N%yXzvBIVKqjX*Ac~W7WS*n z!omWZ`T}1w7r1BMe`dp~@ZYhr4I-L2lLO>ML=imL78Vtyl_b(41w1yA-V;k3RQL?~ zmF9vA8IY~%{#9`olXlR~keZUD8e^p9V}IqJGW@Y$Wbn@K;>+Ml(6meGo@Y^&|w;gHr~D+(3SYel;lt;PXIC9j{+dZj83`+1Ym z`X;Dno_U6Mzx&;jKX>1KH^2I;zxrzL);CoxKlKcq^B1msATV0R(Cavgy%?>~I@|@c z8b-@mAGj*OU`RyJzN~u|YS2^Ewg}JJVzV7=7q*5Ijg}G-##S&IBIArwNao`jAqR}Q z*>o}zh+Ir5#^hp1J52K(Qkps+5v3iHV6mu6ht9{;slUJthNp;zC3=2|q|=3Bgs=w2 zJE&F%iRAS5W0-18WJ8{usnNk1zP2ST?O8iH6U|X%f>vl-lv)55LFvAubxmm!#%7UC z;m@fHCXtaU5EETlP9;YK6L&I~A(X~CLFJ5}#DWPpngs7ADZ&~=lHw}Iawc(gg*1*b zk&Mke{ld_THO57Jxw;SVY@iI@S);f_8ppDTJx~@PdqEH#c}5FBwCB;Cc?v5Ts{uI4 zXwXAS=*&sj@r+iI_=T|Sp9AS8;k^yey7itrdDF(@{QECo;A9n1TSY7?7NUZo%@~%J z_R=EqC;u*@JGG#Q)&7^!Njyllf)I{4Ou>EY+aBfZZ+|=JA&+yOuvW2gasd!9>LO%J zIpfEc7m6YTUoI?a3Nu|)P8P6b#4eyUtQAk^zC9>AQ>|ASb+O@GwCgdt7-M9LlZxWV z?NoDJMy)P|nJ)H1h4A&cWu!B_>2v=xv|7mT`k?>L9GN%?ochE3k#=?ggLh%TI@ z-?^3d91pf1L8ImERKoskUi3gqNIs%Ya`@0PNQwOGe`{jF(2iDkBzlkEevj56Fx6?z z#?$OotA~Bq^k4^`e;Vov@z!_I-myIV``f#>t*Skz4k!QK$Mqe2`%sB^R~vEP>daj+ zaR^sw9nOm3lWPDirE%ffQ>hp&#V$sx0bTntTHrz845KA)fYExTYSKFVW)Z`$t*za& zwzl@S*Vfj4W#<16y!^ZN+N(S7ypxZA{NsH5;~(eue((3lvW)Nfp6_|Jck7#{oFmE$ zbW!g9aaH+?AyADyv(?xlH02qs0JJdv4%OM<#@s4%FaYs;x zAmh@E80ZlsDe=~L62qZHYeHmjQ_8mylWT`KtQpd>a+o3TMjasCPC<>{2-P zidY%XXqkdUhQ&3RF0Mr;=%|Fs!VmsG&mdJzX;My~?=WJU+8GuDBbcZ$BAAUvvLxdY zmGu{}j#&smOS4lMuO|Vb5*!W@vgJrKgt3sMxYE)~1+Mhx*Jwpai*aFShY7Yn>_0O3 z4v_oLwn#x5=#U@{SS^tY{+z_h;F?7x7*-BLwqHC_bmE%%o#YgUI`#lMlEh8IK*&)v zVu(B>D8j`>jI;dAHJ8J&QYF!@IOcX{6E2kC>wp$ z11%>canj;I7I7xJ582D8zwiWXZ=yy!>~xM&hhu0oTEGM_xy^Ol`dPqncUrTsvgD_H zAAaj(Qy~Pm8Sfx3F1L1Qwz*EAC7O)HvoGSF{<1eu{lOpM`8O@n-l}mIrwI?fb+{WS z%|_~6tu3fTd`&iMQk}UQYaPzC3&pjLR|x8$MXiHfNCU2-C_)A>+qxFenq{;u)1ib~|< zNmeW%q@pI{OXu}O{H`mgDA6@o1TsyL*5Z;DuCi#Uuz_7M;d79cAiLnLEH-lwJX~Et zkz=Kxjs?O9hHf5Z1ujcb_uY$?iT`ZF0^?SKyR`5NSu#|7+mWyF=-WJ)aNAo9bVe3j z)(aDYl`m@@0^I%RL)^dE7(sRsHm9cxc1+5kGPI)`yOK98jZrR$A~Bh0tIZhW3aqP$ zb(fuk2iPfLtH{A>*d3I>MehdOAo%+x6y>a34uC% z9CU<9I%p(ouU}$R&tcyDKHO7J;J)$&0Q%iq8IJ;>&t?|pgNFctJ97%op7MY@Yk9G{ zsbReP_kF@1UqhpHZAQzt6y3Q~xTn7YhmRrJ9kz}KcljqDn{K-Yqt)mS`d|;YmI8)t3@f~-}^TyTP0m(wDDKP6?J3G)Qe+;HE4-JggEVH#=>(FnQykw2m zD^=fYV)(xUy!*t76F&wt{h2QU55BB>^$k$J^E?Usj3mN3wxP1B3L_s zH}C|eVa%b*a@smfl%=Zdq!p+|;sIO;%B`tUp>1S0FXE&iZ;8qFRS3AXi;JY&#YA5r zg`YY!!#*-eIl1MTkpa%_ zz-aBY@LZ)DM#~!S|1$wvHlKWlW8I-#oag^|Dy6bMcg|ELk%+PJ^mD0g#Ll)QNrO#+ z_AA`bMu4D?q^%9HtrO8%O0EI4ddt2|clrf*_DRHD_X1EHoTqGQ*aftTE0`uF z_2e_X8lx4dj8t7;fU^9dzk&RbpSt4bN(%nlx2$r!w?})h0hId~md=(5pt(7gy@1wz z2U|RPxHDn4cGc>UDHF(a>QKLmZM)aHy~B-0@Eib|Z=Rd^dmjb!D_8b6es#~BsvCL@ zF`YC6A2upK{G-==URjoW>|-C}r+@mV|3dG7vv!vZRsq88!)PUGFf48F!)W>ErOGlY z9V*%lv_x|!M~w`3>tb^qJ2wXsYFpr(r3!7GdZA1DbUC$m5ZY_B#zVx?Eu0S1hdtNh z@uRnJ@3SY-|Lhl0KmCgUl!O?a&c8gLA zM!HQF7_Ix<=KwgHF0gs#PpI#f)WfNZ_SVosICknG!uds1&KMiUElpcmo*Y)Zb^Pak ze@u!K3D!y<`+{%5b@T{V_=T!&MHpRDRQSMS77$7?9`rGyo`~OdX(=dCr6gCHC{kVt z5*RI*0xh8>RE7g0B1TXHThcQv7cXw}cv0|eNraUKWgUGM&=!mYVMf#eq>kQjJ7>Oj zn9kDez>q|y==HzC7l(_Kx}aH(21DPnY)2V86%?(oAOuEB$lVLX{Ri9p%m3_7=K7B# z2Ys}xSuYau+Fohkc#Vo@+9^5`NTE<7qOv(f9b!*Z2O-3yg;P}(MNyDN3C9mi7w_6e z^tTOJE5xXTeHjy>WyMe^QdJ^dg$|L-3~_y5CJ;X2;>6|0G1={E_Z^;xDZ!=^q= z6UOzIq(1S9PjLVJ_p`FHa_x<8)^(SdFj~8TmJ1VyW_bx6T;KQ7XvJ!l(Q*@pDhh^) zvY9#)HL3wx+m~<)UYuLog368wT2acml&G-ix|tpd8m&A>CK>&*WKh-g+#1Gt(wC41%D24@Cn`px9rC)St$GNhDe^H!RueJ2tSp5nP$iTqMg+D+ z#4T)P!RZMbDI!fPKh1HG5+SiT264xG2Mu<}>H-@lp3!oUGl#ZdrA68yWe^mL&ODdC z<%8rVw1kZyKKBDWHN2Nk_os{6aP0T76Bd>{fCMqoXmLgLGK~`nGU)bB5B;|7z3%X@ zQa&=o|GNPeTBd=~^5>)RbBlwMeo+)ul}{vC>`qs&+G@(eO+@4}4M0mqykR9HZHHD6 z1{3TUv=+R6`FrWB;BnRR=b#Kgi%aW}`G_EYE`nZ72;xLyoF9Bpp{?8;!^CV9D7sxt znjnteMtz$HD*4z=K$&>-?Eu(SpS&;{+FfiG%xJBof=B&Cr{kK?}Z zr#?1kAAJ@8w;T?x@dx(brg`txqm!v-an~CFc;wK3$W3Ml*FCXvc+ZRI-2T@J4EJMRQ_R=@q*zkRc9_zIK^ zpT&Ny!=l$lpN6^Q@D+?!RZYe4iL_LuLs)-toB*w`_|(+s;7_D>F5+%G2y9|(L1m_f zsAXu!B|{Hll!$PqLe?cBPN7s>Du(pp9?B)`bdPc5!#|4qzkY@Mt~XNMQK4scbM(Sj z{oiqd3%*LODp_}D`Hv@m)4$ht#ZJ1!W+gb%2^Y^TfBfrXOQhR?rIICM24r+NTWDU` zRNT=9QIBz=O>QJ@Z4q%wX-CXQa|jFlsw52|$;Y1hcSO&f!@jqRjr_!6Wi6>P53JNzsGnna^&; zc8=f_46-?R-Mz%mK2Msa)U6b&0u#tQT?@6eNHe00eG5j(m?#pA!^C4{1wz2e=mNhl z=Xn?4>Y72YeCRfgDfhk$b~6)V>!siJMj0*eI<300Gm$JAwaqCeK&unB^Y1Yn z273+>6UP!)X@sjW;XB)a6q!Ys97ZzX?|#PtlAS(0=h@$3@I+Huuo^$ev)xtT8C=JM zfm+0VQLaju?_mSmJ1Zvkf_RM;a;n2=O#UZ#e?zP9a7E&thdQ=!D;soo#sfE1ntqN59}j>v!M3?>QHTk0F2QUrj~vSF5HC zc3>5;usD5juBT>I3* zqUO!tex3TWSE#a<X0V}T{;%C+xKL&8jY40!#nCAZl44^ zE9Kb)Y82XQsRkDjJs(R^+k!IavjB1xtP4hdzogVK4i?-=N*c|xS@oH7^CTi6A3MZ@ zZvwy!wted`4~tEbVuFBo7$?Xq#UXHIfoLx=t~Xgok0D26ie!OJ6LdIAgQ+n>aL|elnn=V67zajR#sjJt zBGsbqpTs&x(LRh&kQ;ETx6&&5WTv3fEo^857|}%;Fv23ukU9V$En{>nC_-R$ReP<* zcb)$;p0pjlwwXhJn_*yUx=DnzlA>AU_t9uIWoLlaR+wnKQAP{6B8N@Ogv#cW#skkL z|1L>9iWv+-#FLDP5`hyME4&U1rfiaG$Nc(pt6U6iKQq(TI(5!*U{#}({~jGb?5|f1 z(5fmoF#u(01uGkK%_#u#f_ROwdS`4~la6;Sqr7~JK>)LhQjMm!PngX5z-f-f?i4(8f zF1@y^)oStheI<{t^>{ICDEVds2eha5MNDC&iU`xu0 zW-c&VVZyF7f;bkudFyfh4d2B_pWfof4i4x?p2_OO674ia)-b162U@d?)@B2=UIL@l z2z48_CsG-;9bds{MX?6NWsH`+TL-0k7`(~B6rMq zq#_=C6LMq$e}OylHp!>7S6ti_|Sia_>T9(D*Sw_Q)vWmXTZi4ilxH*EY7ag15UZ;% zEr0N4UAwOP_Uy!o69zwNc#ScRn{~6MCBm?QF!mZ-f;3lY9nLKCa*aPeZP1ZY$8O2 z3N{3?-62v7 z!WRxuJ?j^*Jn8Y-LNi?KO)j7#mC&;rJbOB5v$6|(PW?Gw-0(vAZU}RBPO(#JTv_7E zobjXI#^>La(Q;wBk#eHZvNAz@;D>O@0!ad&{_11=e`kDq)86`ev`iIh>9sjF#L5v<4I{4|e+^;`!iO-moQSLxA#fvBgNmSmoO|k%&+h ztb!6pJz-xw&}#god=#Cas|9Z~8$5Py1lyYvMynkww1_DiM$1MZt`x&JH4C)1>IP`N zbVf^0T8F7;PwN3%rjz7>IF?AMcQIOVB0*@Jj3=+twB6br-_mQfL;R7F80R0GJD4!v zn}Q(~CA9zEzoz@>d(n}go?*1a(c7?#2N7a-8#T!!#wu#7C#^#%1jUlE3(Bm1x}IFAvz>s~T&D8H-?5to+_*;Te*u1T zZEfurfK$v&8=g3E;<=Y|$8Od)Nd zz#X4?9Qpq5rj8ro95HattF61fpR8 zS5^?0{wrU_4!LrQQuz0nbB+tSV|{&tR4Q8I0V+|LV6d7TK8{rqX)K$ng^mQJjMzDU zo_oIg9jvdfBN{DQ(&K9=wr@QG?KbReG8TT7y8VH-v&C1bGe7PwFZlhL6aL(76fMq` znpHa>txBS$6}IOTiR)mTrK}3lID95Q_s~ckvL(s-sZEzgVe!8OX5u!wMak&QNE4$WwT%k zv@DW=X~gg^MWeMFXhkw*SZ+|5!LCU)Kb06wftHAfVu1^7!#I)TV>1CMNhwD#gx&CRf}5GXd%eLcT5E^n}NrmH` z?|dhnP6rjdf(G$CNyD53F>oI4RJW~P9IJH19EyCyvu<+)@Z=;dCHQKcF%XJWYh0~R zu0vrBMUj(ghjl<;dVcc>f6Pm=oGg@1W)puPfLI)&4q{)S6xtZ18`Iotxw4?7MAr_d z6jttT9bVj)*wK)@C5d#0zHSp&1MXSgpm*4B`*bl~BnHG30cfSd1fV6zouvxb zR3nxT!okCcRvT;F6+;0@DltK$b-j+;R|Q(LjMg4vdJ)AQREm$wP7fBm3Gp zyiz5EK)!d)o2J5R7-)q?)Ai`Anq#;Lw7zb2LyXpm6DMBP*-5uw;P7j#zESq`>rqV` z@VZ)u^$gHz419(ID00enL>+7v;xJKF*)SovehyuYaL!G1Dx$720cfePU^^*L_|~Bk zkmj2>1gay^Mj@4%Oe(fbo6U4g-je75v}|c6K&yx?)NM>M00TWdNP70UXSpxQu%jJx zrf{Qe&=FEcY#h=gmtaK7sN5kJmX1zwLXoB^Q3&PTNxD;`z+gR*r!6dkuMJupO8@~Y zGG^2tpci^XQJ6@Rf`f_Tj9uqqI_98GF?j@8huoDEML`-_@=7o&Ya*>U7lr{Bx{CF! za=tzQ03ZNKL_t)bK{6tl04-ZfQhf_XO&Kk1YhoLo7)98s42Go^E5(G-k_|xIvUG|u zc`M_PUZx*eI`xn{-;i>)9m8kg$$Um*L`)nQw8R;N6<7s1cEYSF5hEgjL-0bnLO&m| z7)wY}R`ZL{g>(IiLDHRQt8~6oe_HfnJC#e^gHD_L|$iMR=WbgfuKSpz<*<+FlK6m!pkP;rg zPJlBDv{ZB@qa}p@)!s{^<#%qy`$1*d1XML%TH?NMxvqV9+3SHv;mB=p>iO#uo4+b` z1B})LjNJrUU$=Tmz-eu5?bfxmwYRRVt?eC&{tA_p@+H|Zd_AkK>q%bJU&h)Te~nMS z_`gkkox%j6+^cnHE$QZ#zw$Em(Sj*J%UWtdeC>-(lB30T&NP$Kmltn6UQIP5$d1fhE0zL2yLtPl+0JO>k;GwA$8ZD_MVzh%uGK{l4_GZZ^KHQ;pF{jTyK`AxEvEdWXZ<8bm$#{T~ z5gR%t3oRh!Y@4CZ z$@82nhP;-nZx7fQ)ssQW`8$sjrD5_dcfQRAxGEyj^SGu*r8LSqqA-vkJ3c0*vN_r~ zj1W_eRsdSj*n_e!4c|*AIY>Wh(XB?Xe2`9Hi7u+CICl$89s-pUEW(17P&iA~1aixG zA~Y{@hp9v6S(0*a?58EqjV*(8Zj#xi;|4To1nRn4hXl{5R$#?$v>JdE&9n|1dxv<& zdr50CiF*P_L(q8MeTcmK5UsW(PQS`Qy3r@!5!(v4`)PMB;sKgNBDQ^ducJDX)?rkEb`f`-EU4^P*a;xc7Gq4fh-ho6U5S=4vSApdpk3dK z%`YGlh=-S`GQZGxpk+|DBB}&QJdcy7DO!U0tufC&nN#HzS?-y(GPcO#AaiFAKo2&` z8Y55n{p&8_sxM>mF>2JuB^jIfm}(BHL}9lS=g&OFz0n+5agmc-1rMYtX)$J;EV4n( z!6v-+j@JV)j8jIk#bz$BP9lY6nV6x9X(b8I#0%U5Xdyi45f!70^Q21|_SCJ2TVssK zsj7;0s>mzB##W!m#*71aRCnA#bn> zc<6H*@#n}&i3}PoJ@jndDbdCy8U3i@XuXYyV%ouIcBE2_R48lMauG`bXh~aRt-vb5 z7)x`j)wB))Mg#x-ierIDQkKr2MN2qWDF#V<0_GZ)PKm2e9KwvfgKiTm@sjje-~*uB z9Z6>{fIXSvaEz-<+|r&}3Pkag(UMY=B*Bn$;oAVX^B%|1+v30QwQp|0n>$|Gs5yKr zXz0uy8m-1g;s#a;A;d%jwpaF@lma6hNwSZl`{K&fL#z4OyFZ9{%R3Rtt9M-$54{=r zzy8Lnx(`(&baMijAw7%|jqhk1fjL$ZF&xsQ!hHGnUZQvJxe*(P!T2BrZ z5q}PxK5^p2`>wV9&X?Yukyd{w@j`+$Kb2_Vz zxxyinj7n69a)o1o1^8+S+oR`5qGfdK8L;isFQW1hWow=@+uH<(1zN?Jx-Jm$3aL56 znVph1Ev2-pf;>6EIahN$3I>GtJb?SF-^&4^o~JHqtb?*HNVRV(4h92c+h33q-QmEhrrndU z2~KOI^_C6*vYF!_`c`f}bG}8OHPq0zRxhd!eVsDU^#uGVDQHEKQDv#JE^4^NNG5Cq z>msTOjKxXFg~EhUdL`lsgVN;Z#X+N`iZKV;G3T~YhC1Uw!)S@A2N&#bR_gbE+B)ne z>Luy3z`Itz>%G&Qy9U4-R}$;)^yiV72{4a|lHKp8BY}R2U~jgh6!exfmo}!qlh(Q` z1n+SyK-_q&$pmQa(Uw&zv=H|@?~5UAV)SX0 zE%6c!N^jPU*Nt8|-Uj^Ci4!M=0O9Yy1$^suKL00xCvNaPZ`RkjKK~~}PM<9IeY`9j z=T28#ID4gcOk)So=%<<%;9g>QRRr6FvP2k*V8#n5JkJGzcqdupK>wZa zL1_$w(!bwoCyen+Kh4&P9eE|#?hjZWR-`s(SW5D8Kr1qgE6HdyA`jD+?WD(>zrDl! zEe*rt!9b%|n?wbT{AWIDj!#Kca2yu{|2b~Q9sd5)j49Wbd!cn3M`#$(gN7OL_{5}P$!3( zOVmp;E8VNKnu3U3V)#fPg4Kf*(XO1tm%A316+7F;+dKfqTzM{p6eNkbVW1T?qW?XB zmQ-TziNoeSgT=!vqjj@h=DMlT`nuPRUO6^_H(d6!yMe81J#lSq?FjH*;Gh5h?7e%m zWZ7Nc_u2bA>s|NRefu%p(>-s^=rKr)gph;)gOETL3sMFF&LVM0jIl$&$x7A=#!eis zWgLer$2b9uo!81@-589q1u-B5;xT|8Xv9dG(aiKr_w?i5zVCXT*WNpS>|Is2Z{OS9 zchudZ)|~HJy>4~YsdG-9sygRe-{1H5{T*_fCGIk=T*_HnJ=hC3J~=0&>rMZ;+NEf* zD@SWQm~|KYESvltcN|X4>|<4AV99~1gq-mo2B5XN#T9XxwaVyPqLq%&Qj%`1BL|x_ zUc6$7mKiH7f?|7%=<0KrEJ5Tcre;s*p#aaj!@6*6h@Hv0b#4nmz2GDM{bmIk_y(1y>H3oC3(X7q+(Ao zDkOu!fVDwRos?b{pz=O-KWCKN$JW>EcH0e0I`w1@iyK0l*!R}2xFOaYMJ-e>=zx_e3x@!GWYpM{8Jyf!3kKM=%S#|5DC_pKslngK^omzjDGsOpim?~N> z-hnLPNY!H{nqfdN^=OS#<;DI3rLTbyY@YfFlIGj_z9;7So;S@MyzC(m?C$2=L$stT z9SWwoS$6_6^FA2Z+OXeqtqgJ>LS>28;nqDd{qIvc^t>DRv+RxmZq8weJA<-xc-L{* z(&5hlztib-ehD}aoC1Cr_^DfX?*EKczZwTE(CKu31o#nv<>h7O=kM%T1K;=W_%`k} z7-LYyHiF>bW1|zGhgR1gKXb{dAaTt;wmW=uzC8os)T|kN;W+>*212L0gv|zYyUaA3 zn0AYSwm-FBt*ml7Im;ESDf(mHuk)ec2a($w7%xDm9Gxb#=jIr=AqbVnkz%h zEiAC~7r)5ti@(A8iL-VmTuQQ!!1EPG3shF+$nghZ6;d~9B+R<>3kWl-Xqh}f=z_-L zJjNIvkcuDsu^;1)KDJ2^)hU3%m5anGXX)V+TwV8>D}0C`%2R^+JVoDik*Nfa7QWG> z?%H%@t~EKfFTeaU07F^hp@ll%{p3I3Pyg*tqrxq;kWihWz@yL^v9PZ-9}3QSw%l^8 z-awh0&`SZIP!r7+%w-oz$_@bG;fKaTOVw!XV`ga|_Zix0e+zgZMqefkfXd&`ILT-kA^b_KRl z8jB0$FGAu)6t0P(QK_(_3jDCixqKbg{~du>nH--=b;Qthh8{UGhnStCuM}r$9>3w+ z_jhro!TkI@&EXuZLlo7>v14y;sP7l2=I|222oKUI3Pu7@h@#k1c}4m>vN z>y}J)cdlP%kXYqnb9JR8{Y%m85R6!Hi+1es2 za>6u4cs|dF&vP_-pWSY~O6n^H#en+G2HE^cqSbTgEMaqJ2XACmwaVo#Mth6~ed@sa z<|bZm4dVr*7j3?-s9VwyjJpmc1fwR!hP`l)2M|Zu4r!Vbqy{-tp%{7u`GBGHXsZIW zRkWJ^HbJ#U0W^j^YYvZVq{2sM5^4}<8LD0-a|!L0BIW9ZOWX$}PCZbg0qg5_+ZQih z#PfYp8F8ZSvAx+P3IYlz5MJtKK!DK&6_*ngZx(ENA)q)~tC8pd5~l>2l%c9K>~*;+ zHaMFN7*eC<4T)vJRUPrgFMbixE{Fk@%E|kD)RX9OlDQ*VW3p82*khS_9_P=Wr)>n` zD4~!ZfmVoGgQ6%{UtdQ)_y}9W1ctUgSoJguRiDBU#gRAtwcXuV2V+>8eJ^icI?Yz{ z3e{kCvQI)xy+-}P_9P~xxpM`a!Y6Y3SxrYysa0-ljV|u`IbblHHsB-!&APGMFLJZX z3?GmeL)O=C9`yTK(P&97Ug&c2j3!NU{6JrKERLVHf7jR7Z~1&V9!E1AbXi}&c2t(D zGPAxu*!Q_cYX~C}KO`&9&9@D^-#ZgV;Q4H8fhdphZN;%;`_JDztfIy8^78KjzkA51 zy&L#&r_=c$_RQDpbUOd{^78Tz9(MaB?ktpY$>-_8SzzPrzWqvv2YIyY?de5h?9VE~ z058jsF1BND=};6I_yH?iypI%(`T` zB1jA}ic$T5pco+PP4uupx|XdOHi^O-!ysffvByHy+oUo>=dOIEG$XId$g{7xsuI#Q zdu#?Sf2z965463#O}o|NTh6q3-~HCZ6a|vP%|qNQd}`8QR6*=a^zlwkzvcGb^XTRy zLUJNU?nXMINQam$+bnG2S!VoAe4)7_nq*l<)fvYle_Gzso{tEwRkX%t_*QDmyoL1W z^?In;dA#i%@-fk>PYvOF@zFOg(90A&ZA*tLjHly$QkT#lHaI)`{_CFWJ1>)vwnNT1 z)AG^&Dpqvox~@ZC+En-(nc?pgc9#x!A6w7o8bvFNZjUGIAhc(Fu3WUoc)uL2*NSnB z#oipPTY-=g`4j&=EWG~a_pezz`c~vW`j?2t>pM2+jf6velRKTxuXZ|}_w2dvJ)KVH zU)|_=%gf9E&GPc{iRI5uTG4ktn&O|qBZttWvQ7EvoUhLhiI9pFKf3U ze(cXm8+D{HlMciQ(Q@7{QxxF&Hr&x~Sfg_?BJ`@H$qpt<$y-YpFF>aWTiHdFyWmR$ z^^F{%4Mqx*L=&ak(43)i--ARS{p&;(O=6mOMqoURkcO715u`a)RV8ty!zk;LSQ89=k_9Ml;lbtGj!_Hf>$6mZTvyc zbxl^oYdu=+cALNTt|cB_aEACmq8!m$O+8Ljz^I(q*mX9Bn%*u;+(-@aML`_cJ4aM7 z*@$+!hF_@@)#hmW8C9>qIMe-!BeuL+4HhQr$!b+})%n`eYZa|_W<6kq^exc}L)+ls z@&Rr{w2n1H9-glt9(@Zu{&jfG7R5MkAa%jC2jmQlFYICHQqrCG(pcN6V8D z(z}Tn{tm%Ez@sIEz;~WDDW(}=cM8>-B+m>*VQh}pAx-ZGVdBxcK^W|fAkIF5Xx^EU zX$dzhyChk66^A{|ybt)}J@-8gT>mq(#65{|=`hP~K(vbKV`(zIZHy*Uwp6g!3=h{T zTE0`XQs>PIT*sPrElZPmhk1_heGJNmJ4O{qt)f+IT?Jqm%wnPlm5tCsW3v7v6sWc` zLKtH^4u_U#iPp?y-y%OFNZvxEra^=acOg8fPoe@)UO<@bK$anuL54ATHiw@Nq253n zjdb00;b4-+o$z+*v`M>tUYIAO+cW7ta*v znc2xUs?sCYCz&%3fPWs(c+@=WBRj?ko)^!PhIPVho5gKQB!b1`B>e_av0@c1Km-9s zD;V|=o{#i>#yqDN)JWxs(Zxk7!$&Y|Gm;hhp8eV!^$HrpZH9h0nQ1s4_0sG0&{|(d zv?8hL8V?}^>zT)iW(dQA*m;uTK#~|H0kEH2QM6}Uy!*5zbjoW08+d5$J9UbBH3c_O zXYR;J-tv~?96x@1|0xHqYbi|ZZQ%zX1T&dEEyS%By0s>4C5RBa%JygY!rH&)>5Ko2ZuiTmD5P+iE}l|o-oV*W7$GUvHUN;bQ=%2R3vH-u zS*y;plyih70Mf8Ym~M05_7>Wg#K%rEk~M;21aSo|Y$qaFiWpf^Aa)UyK~N);c0l3g z<`yQZVT7a~Hz~Y`r#}B1EOak0@?#2@`d#tu^S8F_Q0kePYfCAiRHQl}h-)AO$7iaT zU2mR@V(O7kYV7*f`y+n+#RoXIZC_^*I4=~Ehwr}+b_HkFTMNAJeeXMXl$0Zbaq1z0 za8j;|wf$gJ+B3Fv=tzSfIz##)iB@r)qP4$?XsnzBV`4H6mRw)us4HdHqia1{o{VVE z`ZU|O$(VEqNa5R@tX*ZpF~acfl@2SeEIt})l)il)t!hnj>TDSuP@=>bhb3A+iOsj( z>~uO;flZ(V{P^wOu@bKjc6+Y!1BjN|rD&bqxk&5M6@W?E5Fl69C$(GAk_L`9-p-qk z{1uwI&dApwCCVuk+Ve4v_z0uHScA9m9ntdZcBWUF5P3Dc?Jly~pt5ik5mhkG_hm){ zytv9}KBl$3gU(Wn6zDBG&c~adbK5Wkj@U%nUi9k55{@hgm}dbDWMP9S+u_vCCdtV; zgdZ>xH3C^6gg^^{3?oPqgd=-m{IWfdt_;a{4GWePyA4yV@R_gs+f;^K+Pzm8c>x6w z`x4I`hs{m9&$F}F&R7eCW=A$q54{(FlXG=sGG)M~QlTCD3?0$hN)pyLclh|lN7y{^ zqok4Tj@b(5n5{*V!N~wx-emfMd%tRVjBsj|bY#j%k4Afje!q_f5@QIXG0|%8ySn`_ zQ}HIlqe*U}XiYp??sOKp{m(O=V2d*|(S*3}Hp9RD-8CM1?Da|e4rIJ9(W=*Gh=O^l z)%iQFw#5}&Ivox)M!!DlUWpQK1RRQJEiW%`EiW&B2pETto>*R9{;uWa<(meUmbf>t zYp(NwM5`zaVI+_eVx3Kl?BlRfAA!nU>%Gxvr;PYo5cv(3YERIhNl(~P(>U@$`*xm+ z!4fp7HK+pPE|QT~w-?c5MBrBmwnzA_Spt8A7uPUO`4GbqGOmy(N{1~hC~$%1_PqFLJ^;&-D*(?pQLrdlEqyYedFCEEOMOQ zKkQFB5Xa+5L>$v-G%&`{@Av6zMLZ@x&BG~Lc3z@~xtSS0yw;<|uF)dz3%B{%58gkS z4><8?iPu+g@nw8X$Ke>2O3bm@4_@P$yHgNCP^n7#J^NX`U5ZwT5+$z1VY?0&I-O2^ zd3kvS_$=@polfVkb~>Hc!BSe{PGGlMm1WnD%>mnW=m&zhBJho#XQH*HOI)p zHqwVLs? z{0X!t@r_R)D}XSKkQ392L!JAzqBE*Yo z=wyT%1PDOJRhXH<2+$iVP z2pdt2otcvioLCjMEh*Ekq$NfpJZHq#V-=>}V3ccOUr?)A)vV^cO3j+*jyVL?9D%?$ z1^>~LcI1sQhRw}Q=H{*+{o+xOx#aW{t?|r27m1+?M59SN_W1d$lApUWC0b);tPyz_ zAz7$8(j>&Bd}zE|Bw8>j9g4By7R5B1mPmHHUAEG6YKA{V%l=Y5m~Jb%wRneMKcdyT zmfZO=ecF_74Kz0emwy8>__E0Ez<6)Q7ag{02nJX5=(0uD!n~z?g|gt04kXpt|!%)+jFp zS_&dU0y8QrL^iKnvD~FQA3NPgbaG{~1Orlm~&V*x7b*1=11zCEFrr zX6K1aj0md;qltE0;A5pm=vOea0ma5DSpoiFo21g>r#}9wEl>8l2s6DXH)+u@bu>df zq1Z@=%tnh09kDpyT%_g+`cAA0zx6xpK8&-!wm0$IP-LArzgoS?AX8KVkE*c@t>rHM zv-K%)_^XST`Ob@fguw9lyd@_6exE$g_f@onQ5bP8(K4<)X7U{BP1>=~pY#NOJeY20 zOthMA$nrvM`k40WXc2WVyt>LglO95f0 z(|H2eSzcbgfE_4(z1g6|Jqe`@?>)N7lXKl0bRF8#Va;7z1t693veSY903ZNKL_t&{ z(*q=eiU=vtc%$=3HNpsq77!BR%L;y)p)u^}O?v4SE?VA2R}22&!v92~2L#C#qQOg; zqM-2f)Hv)EE&q}&1FE?e5i#0#^9=j0Yca0c0n3XK#ri5~DhQJ;dQpqduMW}1ApZ1s z5hkW^-lnE$(xAcG(3;BEM(Z4@)yQiSEhMkZ+(%u)Xsl>ee5zZYq$2*pZqpGhRpgUy zOHpq!%r&(@GIRN5YI(w=Ga(Pp$J{q(NlLRNiIYtt*EN_p^Y_Mv?ULM&XkipmTEcUp zY1cC*TDr)g(W2!BUKb*m5RWm@ny&>cHG)Y1DuD57H&?S@S~G=7l!6%hu%Z~>_o-H^ z^m;wk`hBWrA7}9+f9=pFpcBo=lB=6_9r}W&-dm^MbkQI}OiG#eTJYs{>f9a1xbqMu z(Q_qAl(@|}EYbQ+U>W#dfnRp_JqBF5ojX4uYp$37yupSxN3zUj0L_~MO9Aw|#L5!9G0{4x;}AfY0?)-CoM`S*w2Ew^XxW7Y zfg?m?k5=qU{-CIO!qG;W<(&2PFW$Ca)i93VCezpD@@G;<*fPdibn+N{xZQQ34 zuOo~$2m!4ir#Q$AUl@E}PziTMHxvcA8W2mLP>hL|GqejqArzqr31p0yWtgynFC(@_ z=Xkka@GZ~zJeOOKS7A~hKqv*OAeYK^9Kz&69$vUech+M>kI+l3){l04OtjpbL{sEYYfOn25IZ8Yw|j9nhO_UdccgRR z;}iJ%C|YBp6;6rP#>NJl!{NlFZ(d6byYElq9S-nl?Tc|q^AHC7uLevNt-Hf|#$m%y zmJUmlxKlV(GW_!Lay)%wc%0uCZPKK%(V(##+nm_e#GbgZZQE&Vr z_rK4*-{-@e^KPuO_S$p z_HjUtm_ZS$pR+PnQJ6B#slsd-TsQZ2Q?Rl#S>unO=k0w|&!3?&F}DE{pJ26nVy%#v z1^PXJpeof^GV}A$$%8>*>CR$L8TXmkxS&b>+AVZuiOw6(-{Nr`r)wV;HrH(4LzEoZ z3f7&b$oj)Y_;HuVia+VXZHSX-w!HL@ECexi-J-L+#@i3it^t>qyXl$#3^SIJ2G>jb zLq@})IOa238RsMQUkBUn%`}28{eV}g2-?-S-U52qk|zv(4t*Rnobf0~t?!mie_~j= z5k%=>+3Gc?ugekJBglcOc8E^9TafX~ac!$4wQUskXPc>|{aDoJIOhYCWwRc}Kd8^K z2I?5gI%Mm8*6__HE`!MK%f=iGWdB4%W0DktB^dfke<6!8BLOG?n7g>2=}~yE0Qn%5 zOD2~0FHm6{PLm9QP_YTB(HfTSf5~Uhxr0os{!(HhasdR(D3Nu-EkF5AQq~we7hkUZ*Lx!Q_t$U0RwRgAVjgR^U}njyiwKkC>kc$K z9D;A zDnpSFgljKqoyH@`gM-si0h(z@oY|QhXxiTQjion}6eRwi{tDW;6f$=?p9P(HMA_bR z(a5qN@vgC~Kud$~-QlyH`nqpyKIfHuSqadT-f@U2`mCP6;OVy;4|w=km#F?e*3DhC ze-ZFv8OiXFD}qcGOLg%ik7^V5OB!Q%zuzY|C7sS`vH&d7{k!ppenCcU!;d8ehRSj| z`LWdqP`jx5B%Y^Kl8tN;Rg+>!lK8PLpDAtW%6ticBDSm7l0uxKa_^q9Tte)-+rT_F zNsyYa=bbgc1RezkC%q2OeYub4R|i_Tf9ZLybhPIB9&gr~s1_z<8jHlhXgF1-QMRBJ zhmCU!n%_1h%rEABX}dT`JChjs68@v-46BiWNnPI6ONQ{6yTe$3l$9s6Ips?LlV9vK zEC=%ahsi&jTILL&dwi1Bc7%jZ0;hY-UucK3pIGz#2(_58##wf}NMw`O8g^CMW(qME zvCwztNSov{LBqlmBfH%%=ewuNO^*@e3_QWB=~X;3CtR-Um<-aM!@58y$=Yb7SL;2< z^g_Kseu>4R*K<;U|E-9?>-|O`z6VKqP${+C8bPYW2`bt?9P|Tw^`i_=&iBwsh0Nl+ z=1&&ek?9x6=7lOV;$FhFRd9|H|M+nk{1l^M&35AaoynKKFRP=E3nBN8=?V8Z*M%dAge0?U!w1XgnPzHE29I1>*-wGi z-iSMw32%Qv;@KnUAlbeeG{FS0{X#T7qa5(*?dAhRNhI>|@&S98O*Z-adna`_4?VG= zG7qE91N<)S2mH@UoFn`So$kj3`O$l5R_Atuj-hjri5LfMhtd2UbK z%-xWq>vw`4WO1|*#|$(=tPY9XQ-br~VyVsEo}GMkOP9EqrP{7FzaWDb`}-*wcA|>X zKI6Xf9*l$HT=4=FI=@Y;iHTMlSlF5``nccAbGzI<{=z?gIm*Vh*_MQ?x5N>ktCzxL z3ZlCtT8@m!rx7lsjb2D9itxC_FA}nSQ?w!->b#V^{diJl``EGCE*wr0d4Ir|=U2tK zIid zoOkT!8Yc)%VfAN&A|t0a3^4uPRrbX4 z2x@j;)DlZTo8m^i=-r`Nf!m1k6_(z9^fE}!p?RE(g+f8xn*PVT_0HstWB0bhmGIId z1FxOeIM6w`x#ch1-#{ul<57i^1FB1aJlX=E%&^r!{gEdQ)M+ErAdD# zr}`b`Ri}ZMRsTfP{*3iUVwy0?&VE)IJC(bFZ0r5y;_z$t6;sL5&vlnoriF06Q95pE zd5sWgCBH~?U-X@;qu1^xt4ZgB zaVFS;4FtnV(7H7O`Q=-(c^aKl`F9=36Zk-@^|PLkA0Ce;W!t6H=iIS{UPnlG)wZp# zsTGUGe%lmx&8QTIm4xQDAoVSisN||4bGh%_+=5M}AriyID!~v!1-V3xj6GK762)^U zAEUO-3PXwv1D{+LY!JZ;mr|>yuSl*f5lY8LaRspHisiG8`V#o9Cc9ARcR33|JlChZ z&icU%-%C;_ldF@8f4F_V{HdZEwFwN>2x@EHa`M7JQm?NcV`iI>+G}}>Gs$kJl@x7> zwssPz8{8#1lbziblFro{Kp`|XsuJ>7hnNUv%iKP#csuZ}CQL-k9=xFft z3?o4@wbb_y@!7ZBAJXr%B2m09U5-cV7ydk(FI}6z0$R5ppCBdYgM_6X>3Gt|wbgLh z#X?)arQqJfYXkxh>({Ny^1L0$?;p;V(TM6!AP*BP>TbF35P}P$ejtIcb9`IlkdjV4 z1Vw+KeLqG!)r-;#WW-Uk4OVr&34>;l&GLHsL=TQ0Ff~;igPS1J>z6K$C5xoa(e2Gz zx;)gbqXwF63_pn3uu{O3lW0~_xJZMav-_HRzQIzKMe8};eP#{Umk}gRi3;UJsl4Mv z?QDc6NtNe+5X{q^?z}OFA&1oze5T8;?in|-e$P3Iafwc^8cJ~(=wK*+@*nEB&O+ym zg@f%oFQ29i&tTGh88{{^UbdzRM1s6lQ>OiV8*LNmx1Bu@?O})e@DNH4oPqr#_`9iM z1Bvfzt;>sEL-eM@=XwAcM1JXmRUY4fl{4;`m>3>?OF8hp>vo)G*@^+aP)Dn)XE`JZ z(_0Pv=f5NaCX@O`=dS!BP_Ouxt5A?oNTP30F<=mJG1TlS88jY~!-!fdpTP!baJu*Z zDK|dtS|xh>6IYN^4k1{yfj!!@pvcqT%~+guSc2gMGujD7V(9U#KAYZIjb8(_h}YO& ziQhiw5okB>&RaLgfv`X7hnircj9m#;x;8kUwcWf4Aowky3{-=3CWfY}*k4z^%W}FK zb@rY|)M?3t2{9{5(|r~P6lP9#K&5cH)k>jgpVjn%lrs+--k9t3eIw5}CCCw?G27#p z2$`l9T5K=Vg6_fyGDON^iz1baBvM8ZPn3b)uWV4qUwi(XR(}kHxA{}!=PcesI=6Hq zJSaV#Yna1~yr7Pm?1vxhDa9a0Be!PuA3>nHxi;^QQf;ph_$JbwB+nu{!ug}MUU*E!7U zH&~})PFAy{SvtC8VMu_Ra=kXMIryDt8h8R|{wrsZ2hc5I&&HKOHJPMjSsT<`K74RZ zX^)-^adFUBw29nNguT32RG#J0GC*9nb5XBWmWzmU(k6;_PrTH6vOe(t=fAU*Mjb<5 z9Ow4(Dl}xjJCA9ZG~8Z~uBU&TxsrCbY#GdjnDqh)4n$T0*5iX{v~ z9DV}GP~~JwaP;xv_6RrD$DdC3dosPjb2I_|M;WI`l5h}jWpf+hl9GePKw~1ur)H&p z1|5EkF2JieUZ0NZM)y@|bVEDTE>?Pc?Ze1Er}&RB<9 z=7eap)^c*epf~i@yX7`>aZ)XYOId62$}b;uNu)-5F7(f^=LZbgjlHGe8=rYMSGzJ% zDhdm=-a~$NYS;x8&3d4R{4{;ke+7bb)wSrNyGi%;cj~SDUp>D*-fmWVY^fbS`)8|4 zV2(Z@<`!u#pHnnTRR~q3v&q;a%-})M?L_9Cx)?bHdwYf&wgAYufHFw-_=Ih_(w*gL z$;SbJq$NzgLLlwXL>uS~phz}@)utlPbryZf)#-Lek-Okp6XvN1~z zE^DU=#?~<4_k{d7!voA^9IZW`>Dn9PP?}noI7Th#k~;X7V*7GzRsdZlyr&xB(@9px zEZiO2Z+;1S-~ZgrmutJs>JJ1l41271fBZdHit6uER*R+s;Kzv&lBd!T>4eE1H-?*& z$28WDrqzXR1qWnF|Afy(jCXOm^!j8yHADH0^ILX|?D(%@Bqlsai!hpBPYLf(I64!}7)kY^^S2Kx;8UJZGyHSDNeLTDKc&v*qUw1vr+6 zc@*iW$EFPUTtBLvX`bO4!9_EsYSuaEN0N@jOxo_K%Zq$6!H)znaT|7<8S3=n$Gqb7 z8hne2m(^YhKgtmz-N;?PQDdjD=+Q@X@5%4Xt+39tdc9joBiv9KJ_H|*t%~#7z9ub$ z-^lI`JWY7}Bdp$Ir?;T{wr@(Bc`(uQB&+mjfnwLuZ#jmhYvem(qekL#lc*lOwt$Gj+cIZ2E~=Gx7| z6};>EVp($zuhIW-!b@K>c0<$N&fLtpgAhoN)ViC_D=(L^p~PaI8g zeth2AWjVV%^!uG)boq4MvVPEBo0{+Yu16;kwvhG#6J*JIBmCkWy>Yj?j6Q@BYuLD| zz;f2{5orGSD{6Qn+H?3lsHzpWv`7;|Quq*2Pj^IYTk!SaYv3eln})K@E%4AJAVV}w zia52~d)+~e)Zlb0>;)KQliZ>-lOGGE0Su(|Q^CG$y|$Fb5> z-QcLHuX-Jn+c74S^eI<6P{c+YTS*U{8XeqY**X{#HTMkx||Z?iXz82tzf z)N;_BP9EgWJ1*)X^zYy&*MUb(>&{7ppwa7y_n{w~Q+0I^z6-*3BRoTt!A#=B>4bVIu>A)0fExGYGSzs7%j6kvwbZ%BT8uwJ)6VVr6!wCO%Mxs%+IZilwc6 z15S5Gr=Lii+GDqo&69!an@GIp0J3E$$|TiSWrOq?9R&UQS7FuS@e447ag=r&57>63 zFDAWGnrl_kw))&eT={^4%XPugknu~S6uMs!8WobYim8APtnS#g6wx&65ydSGmEYVZ zuGvYkYk&V|4HDG=T!_>#kb&3||DX>yFJlr5YkGts#eN_pHZ}A|(OAdde^DE*e@!$f zv6(ikzs#0L10fuwVOC*UR1ZAx9 zhR0Z^JRRTRZA~>r>+D1V3%S0^j${i5uiz)rs?WZ3ME@yWfA6Mr94hVKhT3$qjgR`$ zJKaJ3>4JZG>(nk5`mcSNueJSvSxg=5u_kL?Owi@m)-^Z$ctf$V%o6M26@}4=vR*Q= zPU!A#i$# z>`%KsC1Eri*@MT_%rWwvqtyfL(rT@w1bjqE(Wz&?F>Wcd4^oliSbUq+DFaM>-AwDH z@k^g!;*QYUiAc8#0c^imL+Eqe0q(a{A zd`-p}iZor}(jOjO4n9eQp>LLQiz7RH!4M5x_&6q$>yF~T#5cas#XKT|)Q@n)4db7mWVU~bo(vA1olX!U!ex-SVzA3MFWV3&9u=B@^ zRN|4gI9IYz+&76yjtPt15)25mqe(1rzig{OetXlp{-;0W=O`202d(OkZ=y2nZwGXy zS5VJ*lB&=To@1NVF>C_!YmSS`b2+v;f65 zy1e3_pyv(BL_zOu)b$rZB?Utjx}>_3&{QT#is8x^T~bHgfF`tAk_(>z^sJI{P1f*u zCnoV%uzFAQt7W}00I+=gWG8)xJ5iRcLV8gNwbBe`?b$J+AUOdX2R;EC1ArvWg|WF7 zYv10D4c|8W0k)QmQl3lIPNdGTJ08>Ri}2eo&Ys6g$R(Wecr zVlEnBAS=PB&W^h0ePeXvb|GwwD=+rh)@Q9aLL4Rq!>z?yYeU-4G&LzFurPCdG+yVC z=`~_wXy#fW%tM$=E?i;mIeo1IB|O)nG9}vEZ~ie_V^4+Li*$6XRicl{;Tg$g;dIW? zS}jVAI}}*lb>}|5Gn^KCatZBhk2F?8rGx5~iQF6G*&i5(DKs#LOgCus%6$1+7zla3 z+G#I;iE=xInMs8^be4I{Q^;NYfKem(FxoCY6&a?BM+wsWL{e7M~}#y(J!j z@|I2#PF>w=)p1st>D!nDs->M;2che?PI9%EiV2)X*W});%@>1T4riKI65$9`Zw+vI zU+KWqz`PQAwb*dhFUFky-f5?O{NO`lADE^Hl}}W%fSvK%%Am+kB2HM^`jVerYQ{gH zqR@IlC_EGp_rMP&>S4*HvwhjFp#&zJ?M{B0>L%uMg`AFMQ_2VRJ8_bi>)JWsnjZgd zNEO^UgJSew;{NKfe+r@EOt(@*_gbFn(=##xo*nF7Y(MK(9l3G2Am1~6jYv55^FJ~= zf=nzR-#D38gUhX=bd7h8mfuE{aO|xNc*!x}yfi2&w~~pDgg>rESwrbrfrJ>nNV!lY zxJMHhI1CCIDPekIESRX8-HBzlYX6MB%>fSRDx+SQn9z7CR}V(98B>0i`intMDH!ln z0a9%00CNw0G!-BL15kJabrp2c_j=9JcTY3UtbYTX<+R|W0gj@1&@`?m6q(AWHsRuh znA98|^`_o5k_&e}LG2GGXgp2zb97#~x@-UBuop!Zv(pUT=P@#fV*lB#9lSO%w^vp2?4?tT7)5T3D$Q zV1lJ&>5YX$U~#teOz732rjPw>Y7CdyffNyPwP zQSFsMXrA)RMjB8%e&Q-~@}` z4mzL5IT6&&JXC@nB`_4;r^$h!Pu%D3YX(}J)E9w8Iv|wnjXel+r*h+KxvS0rDD6LD zX4v2c-S!4`;*6ZS^%c;D$l4PNZKntcLpDROD}mn&IbT#eaH5m(@m|@U+&~OP$hV3>abq z3Ypk3ILn~oHRtA_^$bG?5ofV}V-%F=8I9VExFI#0cU9 zkY%h87myAlbsFOk5eblf`19~4&>gq>Su-{;7}z|q<{R2-7BzjkQ*_3;_IcNyZEoxL z?H%g+5^*^wOJBuh=nV8|J=Y6EUXkmCeY|hQhi$Nh{+wXqD?jvwGQ;j%Qo|B&+UOa_ z#l@ww;px+XMz4S}*RmQPKLVePR+f=LA!tIyPA^%*#RVOHsLpCzBW5h0G_Xz$--nG! zV?=3cYw0_33e`V=iFqKYuCv$a(x+n^G(vN~ys+Kj_7yv9^bXF4&IF*Xb4ISn%z}?a z>VZw^AAyZ$t&U>Mo{yKF$;NWwz8FyFSKlx0Kzpo$6&v*))O@@~HuR&R-}zm2nrFu@ z52}tAx@FJMkFWVy67xO>O~1ZIy#4)#b0Vr=ZS+8r(e|q?<8KCao#fL-S9jZq2zEwB zMk_=wKpTh-HF@rJ`hOd~PEJldT#XJ658G!~+YIn;nYMNHRtF*jj~8e(&3GS188~wm z3F}*5#KVz7>1pa&@KJQ3*zFMl&+hCY>++1$N-BZ?8iaEj*UVBZLh@IyGy#GiWb*H0 zfvD1-52vHW&Dr-}pJHO2!zjoI&I_av*)Mss@~z%djHs%(r!oxTe5>L?_9en|g}_O% z-hKJixd^^;EAmO|Rk0KmT$CBuTBKCcE%@vyZ94~h40|K$s-Q}+u94C3_&7($E&X!d zLt|@K+=$wH3Cm-$Ew`_U1uhdcvUI}b)+y1X$cNy&@dP~(K=PIJAyUuJkMg}P9%idX$gD$&rCRgfoISV0@I3EIUCaj) zS$xlrPSRUT`#X^* zR3DeZQdDM1u#bC|LS9}(kSZX0gd%AltY-eQWZj@=0OUHf@~WVcfQKH{Y%@!_9B*yB zmUQijh~`@M57ml}=cXjDqK@$WT{J7FV_p7cnPfUxy+zHIAozg8P#mh0)m;(nC?#FN z`t~dFy{Bn~sz#qz-gfS6YY_8eE3{u5NJIDIU-kf3lNs&24er>_@6yC}Ka9J8=u_Sa z3tyXEbq3_T3yNd>*U=_k2pt{5H%o-xif@NEKEiXy+$NKLp@m`awA3C?XG3=b>lgE3 zN&0U=Z%i{#aLuh15dBT<#iTXSeKYv(%;NYMn4K1xANH-FH&lE0wNX;_1v*qlr;4axM z+9vsXHSWw=ju{`)jz7GkH(rv`zeC*8eW=aGUXp@j#^vVivTn!h?VxiT>*Bz`OWAs8g5BBxr*2g2$o1{an3dRtFj?;^0E%@lm*8-rz%Pml?Y+p4s| z+wkJ$_xO@pq{5+D-&hdxV}q4-V#wK`R;zf`nq%9R2B{vaXyFjl1r1r`yp(ULX|k*4Xx&R96lSBuq($yco7cXYdlN(AzO6AG~tevDWJUh6tL*jir1=@mg? zxgKF8Jj6comxl}KLBhF6;2#5YpctWO^tmqhhmx2PGEh|^a9PD$0x*Pn^bkQOdZ+l< za3i@xWvQcL9vo=LN!AEg!5f1lO*Lu?T2Bem(g@4P#FNn! zqWe;(F}QzXyyF9=;;QmY7?nKK(|`(O0Kf1_&DM<)WJloe9mfCRTXg>g)D*f> zVd`Ms`maW;Df)usR$U;lXx#z3OoVoq6HzAPd7-R7LfrdSb;Y#8j6Ew+$uY8)X-5gN zB##;#kH0q|5+fH7CV@Jo`bCHkOHUXkO|%dLd%72P7wG^>2SFQqspoMcKk!|NBrPSv;e*x(^1Wa;5sv)R`?)uWk0 z+?0Ww_vJctrga41TNy!uN?rD-p8ecFDG?yvm;kn%4$taO zUDzisio@(gS4i4|gdlFj*@F46*%s~J!J{gunUNkT1%JMLicW)-{+7HUB^g6yn#qG( zEN2xZaYsIyj1oT3qRhfJ=D#rPA8Zp47LoUioScy$Ci0^&kpbh0LfeM}Rp>ya_=zXkF?=)iVG*Xn-k#qwh}3oJYzN=X>z?`|SZ&-=bSg-RB1XSS;pi)2Dr(At5`QhchMVfVra-L4-#) zdN=Kc>pFfBT{d4$s1>m^IEWV~55?g5f(Tp;V*PQy;V?uMRfd;6RWL2c&nqXz<#cKh z+#ufG;M3JeEXC_B+OEGdS2=6TgTLcd4e{UYsZcs%G(-Zm=`u`##_-f3QzEkSY*x!d z`GQY1oruI?8|-Xl(hSFFk@oq2R~AGfE_T|l4L96wPpDYBLrF^H(*`LT54T|{QxDCa z2>QHc12^szMAAr@ikmG>Oo9w2HEOP(1nG0i@w4-1XKfXy2;hRks>g2_wTthD*z?tp zY*G$2!CzL0f2xX_*ktDiJSc6f@co7#BO%T8fo5;NJJd}d)amGav=NhUXI_wgx1IaC zmRNg6BzFGg^?Wvw8GBWg5L{A39g5+IgaxjHu}!O>V*ASU0K=i`o|#E>IiiMbq$h%Xcty!U>%?i@X5H&F<82}CBYD((V{Qg88^y%5X;d1 zx-)vA&v;;Z^#bo z9YsL!f*TnMXFHgss+n+BkpbwT_`&;m|N5*#oVqF6g0G+na@@KzJzn*(S%+2`*Vr_! zAckHIn_%PK>$Tm18>{WCjPJ7qN@DC()8a&>_$^p9oe64R(s5N9gs?s@2O;`G1`#s+ z^wa7?sO5_3inL;QIa($tnjctminJz5m=&grQH^Oj`D7&FoRvgdqy zF@af37jcwqYLRU655o(QpUC9t3;)?Juy|;QeEpafMOF(IfS24SFWQAM1A;ot@Jso#SB3Z31IkbU;iZc~h>6Ms290oI^65snGZ z5Od~-iN{%>?-twWw@%|_BHF)25oJa5il%U?eWsI)JW}b3>^%x=iXn!kf*lszr>jj7;67d6c%O{Yr`YYjmY zZ(>4^M)a8d5jDupScZC-by8haT8kofj^Te{oxA;w{l^7{rMvxDI|3w>@5)YfpoInv z-Qet3iQ96OkuoQH(5lEKO|#A+&fB36J4FHFj%Plw3G8~i<{&RzvGZKp4vu)%Ot;0QK( zKWy-;9URugzsyQeYZ3W9M_8t`c%vyk^i)$E=47VD%&e@!d7H-0&dzV3QgRtx1+!qS zmfS*+u6b9ex^zC5$k?C!54iMMi{K}{(ME%7samcKbsLi1o%%i`nK_*CN8ioqQYB%yH)^`^uQ?F)Su+o@Ocm zpq?Fc(+9)h4uMQ5CK=JcG(knP+F<%H#7>mDw64f|R)uZKT_AvQ+joa27F|Vaq2g|$ z#JMA_x-$b1$TxJbz%9B&S!W0a+NzZ+8n1lbzp?lnKc$75H%y6cklHaie#QiTel7ov z>??3=Oe^HvjwLonr6C{tgF`;1_%AzOwQON=@6SI(bi3WRj4dtrm^qe@2&bxnn&6{^ z36o{}l|tpQ$?b>P?rPo`;sh1L@TlB=QsL%tQyydAw&<01JlhM0jk#2pbz( zs1E%iJSG^#6m@k4C8fF^cw1a1i)Blr-wIz=Xh7R-pw_@cQ4!WAFxCg|e0g)UmTlP6 zlNZg4y+&o{)!j~6f|2UV_X15SxCfMmXpO2=UkAquRS{R3Bx+88CC8O#IyVyR7TGSq|ia zw)Q*1P$7L&&0K*#x~8jZHT<1Dp2gaVpGz98mDWE!Lsh&_t);D<^{=&{?OQA^x1VoP zftwJ*bwt$b^8b(+h9fH?aMjKp9CGpv$uPp*yLGLyI3`!xZ3sp9AB&R;jUa?@@ zd8_s>zO*#0)4iZLeskq`mSKs+vc%aJW)gqeK6)Id`>fHw<5!uOh>d7hJ_nClm4Voq zmN%CF8jW-I+Z2p;w|1$Z>2vnqUPi~i2=&1ZNh{8sI@IXRh~=dN+AW{zb23$KPa8A7 z`J0Pu3tX6SWm1BjW3^|tQni}7;h9dvNu?G4+I5P>%kSje_(wQC_r70WbdUW>pi@}-1VB}&~(BI&n&phTdO6yFO!uUZg|BsJA956OmMIul`Rj8h?zetcIGn`8J^iQ z95VdJF^6b^we2yp|69%4PsftOHM2=^>CXd`?^tf}HOEr=4z@=;OlI`LxYCjb@)FX% zF35sj$J!qD*Y+tKZ)~`yv2v1fLf&@QZ$u8- zvB}-D5fiU3z$}yUAFeZe@%=m`?*b$r2Ru6Nw&NM-CVG}f7nj@pf{$Av3JMCjA7yri z8!kew!av;7crNBQApPMxF)Zx(D7@acbU@pUwA)9C&c5%l9-ck z(6w6M(J0AtmeW6O0#d5P0hQY*Kuk~FMwAxo{V9fHVxkE9R#MdJQ7QIr2u65R(OK5_ zlneUKAc9~UrhIXa6&<;&e%hX)fs*zURXyg?2W8dg_pH0<1v#Rt9A*EEACA9Lra=;= zYTGjV%t-G;+}zz4E)VM-MoW=-t3+V^409q<7{V_^n;i5<8k=g9y`t(LuRl4#!O*M%fw;2RZ=Lc zu<+mUag@!MjR#(r1C$6f;=ZbQG274lwYPT;Ri^v1B(>vr%LpuxoH-#Myr3CP5x0=t zx09b_E#kC%Sly=Vb&iaaKAz`)A|Fu~yf2v=ek`|wod&y9{POhs(ahCUafmRD(W?aWZ(5E~Ya7k{`&hh`|(rgd-o z2s(?)zpu3xqyjv~b1T#~LlNa8DoRFS(L%t7mDwDqC*p9O4QkXa3Z5hMnYk>}%!3+Z zgum=QO@hvOp~8SMX_`%QJe|8t5&>_#4p1?z>3^O`Vkk-VBWEXa<((pMMU;EyX{xf# z_ysH*Htq|NjZ)dWbmbx-*K7i!8Ft)QIUp5o2w@=9Zf=(zW2>h`#BG6*wWBw6^gGde zn#t&vsMIj^^(8V;%3zz!4FnwZ?T?tsJ~u(4jvZem$RKQ*jy%3#T|q@Q5EtF>?unix z(OyWw)Iya%l;sV&Dd5evEw3HLX^iuSwhs-4Ib+z2uAsiY1xfWYO)M(@_n(@(QJ=Bt z*p(7ieje=45Vx7BhZ3NphQ_!UFY9tt`uk!SRVEab7L0pi=mfYx#kthqq1 z_?8R{=P66f+2Va%W?~KS^-RxfX+P_L+xpF?gbN+dd)(KQRbQ(DTYraSt)H&^84UT8 zH@z_Emk=>Ld?`+qY4DeFBjJGXh3r4N6^e$BZIEx~gbIPLNKX9fE}XHAu7Psl_0O+% zCnh!K5>`8AP!)f%l&1QdX1b7PY&7&d&eGj|g9(X}hP1l;ok*%HJUDR*)N<6t$oZ*MIEQ#gO0h|{!sB!SUyJ*Q zsgQ+jjA=>_UGeZkT;+ZpClny)pImmCA~N%GPzd%7KYx(UTd<%l_YZ*(k^ zBWjq7dHZ`Ox6wLKns5ybyV9(*{A}Dq4^yKACHKB1<3?OnjK%!m?B$H0(gBG~{szx* zukxTCBv~{$O>R9gYfgI`EO=Kkj@@cQ1ARnAi`el#^5op!#NCpj(>s)C`;Z}T2bYxS z1)#x!_<*O62G zjtrszP}pdNk!W70Ph!QSo)JBq(tRHu$8T&IX`c8IBpogAhnl#LYOuL zq>dR>S~#-&_}&~?dL|EB3m~Yu&3DP50yaO46KP8@#%1mX#dthJk>OD zZ-nzumNCMm-RWnCG9bxtNF$uTyYimV+iU0!V{!ur3>bVIvD@-x7ltw!PZo(XqBHY* zovko>%#WI3Lu2TYe<-lt&$7{4Npa&6ljo~DnphzM+hKQO9qxFb(T@eI0q5xr+g$1X zb76Qypteb^Gy~#za^VbSNy_?S#Q-Bbe?K{)Uri_!oR=7?Z`T;3e*US#O!~_@o^x=aDbR0v^7Xbh z@~SPu%tFUjby2(mRdEVvgC42s+ARLJNyP`vn7D)ZCFKKES>e>AbiQNHr;zEPlMP*4 zbNZ9|yfe>pqbo3M)?r(16WSK_3eOfomqVaH5NvHV#&B7YMY-M_RK??K)dEm*3=Ipb z94Q?U-NcgC+Z>RI*K+kNlq{C8#Dxi}x7r2jQdBE5AbwLDh*ucVQ9(}4ZM-vsRfV%{ zo(Q)66n;S?ktAMiOs`u0Vs(U;z{ZtrCit%)wNF_*>0B~4K_L)$43|5u%+a^AwKRnF z!)5GD&}U4jhC3(2mj!&yO}NA60*2VfzKJ@AAA9#0k(KX5X%8(bP%a%(+W_d(LI}>V z=2G!Rq(xGn{p-SL#mt`+2_x(HEQ!3D$<1dTQub?NqMWB}gghSiXAR?v*{TV6*_JX# z692HVr;215DyanipBCWH#-r|)Gi=lW$O)bsF%SNVjC|>eOqsMZCPkgJ*Nv8Lz~1JF z-Gs&-sC|ZMCiWG1LXBtZ>DmwqO*uCpVtZ)`zq+~_^0)g=N&3b&;wCr$MQcrW8vdv^ zJQh&dzqQ3{KT(0a{`PM$95wx!mPZQyj`rC{u zthMAg0k{F4+?$gMpJyK57DTvz7-3+j!?kTs@1&zTrojl0dl2=}I+^5q7b9_~fbn8@ zhv7B+`dT6&K`ZPi)Q}IbWZ^Z#DPo>?-2{lB0<~EB77#UYf6k$Mb$4cDV{>GQh)`)( zYeF6_vNNSq+JqV;1wwCTC;3KhQ=|KnC(}qxFGm@hika=clJfsiN&tNp^Ui+FNJyKz zb=!3IMV&8V++f!R>|kjR*>``-x2H+LCK;V6 zTI#K!fU?;=4Upd;0Pd6jO4?z2*dt0x5kd#q3F4w_Q~N+33; z?96!4F2%aTqVLad5^Y@;$+tIjdi|sA16^oQHrNBvyegb*(AFFL!>Rk2x$GkOLhy~+ zTHk5dJ-dq5ykBvyBfKegPgaO!Ih$tAqc5sGGaUCy3XvN+N{6+ zq$%b1yBXsuGgm>C8Wu&EKJ@RO>$!3!lK+_3RTMS16oO%y{=@kKkjP6O&(%_V*%f(M z42}3*`CSQ$Uo~ec2U#-F4waNjeV zrAKFgw7L*Ua>z#>YHdY0$SLJrZ z>#5bJKMfOJ0%pumG6>3Jt?>R05L0|Ga}ub?*Pw{s?|WdCiOHAaGX3I|AF=MUn01*S zlsY7Gg)DH8qE$itwdYOKsgSP}>G9Wu8q@nwn7=j?!gbT1C;EEBmo(o%F^Yz#!*{D}&V$}!s@>fZ`aZp=SF6S6ef!U16R#442)V)`Eln`R zRtDrx+V}eP9vsazd&$9r+^qOLE_Tq)O*peaW{Cn(vjjR zprc)nI=9}n=NRSx>FcYbqWYqLhelCQQd$KO0ci;-0f_+tk?xQfrCYj`kaUnvML=Re zItPZ99zYnnkr-j9p<&*I?{B@g-ao%t>&~oo4RGG)%OS*45s?oE zVzX{o6jRY^*`{kx@e%Md%&RuihuYQ@6iyLvZIi})G7|Ed!&Y)Y{5P|1uw)3UhN!yP z(mzeRX<*|+yr*zj%Cbf<$A!tniF{z(FBU1*Exr}r9bj>CrZ&R>6S88lO(8OSmgwYRY3vEfV!|s5Jw)R#B{Gq1 z$tGj@b3 zAiO<69Tm8X@%|lqLxQ=)%@%a*IH_IWY7#T(4W#jcmTY`cX}Alw=+Jgbtbg*qs8ZhAo-v5rdaoC@*#7I(sCo=dgv9PeAH zzU|M-ESTLI^<-A@^1FAOe-bclxnB1%PmneraD{sGiu}1B{6M$)!(qaC$3Y8++cx1v zuXNIy8U^a#;LX#{kN%F~KhxR|RbCCf0yz-t2!Z|IH-urf7Np$x_Vlr{8O$$Nlv6TN zt*d?|a4fi|cLan*Mn*<8&l$oLITF52Z#8e8c^>{V0C8||lukZtnzDOb6K-6~m~7m5 zb!=5z%NCoKHKxR48v)M&4c@~H#oS%*h5d~JpLg#z(y~U6kHz%dqWNFxzss`babg&I zx+O1(U`dVKRfXwvaSBDh&9EM&!k(v6&D?em;b7Af(uF5X5OWi^-3{Iv*{jeF{=o=E zA2_?Wm4B~~`qhFF5nNN}Pn&5UX(I($V#->P{?#e>*s+*$GTw&h|i#!_Dh}(k* zwiL+KY4k5gk{1bBt>3-7!JO*g@%@wALS5NuR&-2^%isP~J^#U}5zE?XxOv+-!&uhi za;mu_B@+&o^zLd)S8xas;kO%xcg@RE0C=6Dd_*RaQIT_=kg51J$b2EaMNd9^v)e1z zI10{9=5EZIIH#~aA3!%==7;BnF$LXJAme6$9?0=L?ouI!^bv|a-Wqi^vF1`u<>4eE zv$b#~w(&}(OW?C_`qf7W118YDF!$$~sx6{|N~*vr4XS}`{2GS%5!8v;DNnPD7=+l^ z1*b$zxwdgqnH%MtsZuA(Q>Pjj^zI|~yI;S^l2+Z+biDn{an9aOzJ`=XKriE^I14xR zuII`{Lu$+!bh^2IT>1Ao4~OvZ=xzBw1Tb(>boP0h~MYYUT0_^(0nU zQ_`Ftl5jx46Em4Z|2mgQMK_FHSZ&WyL|!oPCgVZDGu_l48HR` zG^gN%?}_0P_TkGmekFO2rr+3OG4f#t=6J*ceToADrtE!rl-a&ih=F^%nRWCIe*4QYzH;8SCoh zzhxbk1mF?D+Eu2tVj-G!5g0D_?vG5q6H0b!<1B9~R3_vjpdTw@HC@_zd9w?b*$&DX z9FHZf)b2mrdiPyW9rff5H{@PWjYyFyUGW{YfwXR@8qHJlT&2G0^6E6{^)Kt&4d?AQ*P-aBu2AVbw?V0u6#oA(CN2S;Hk zwR1st84dW;l-`HL#Kn!;i=fiy%)tL?9S>yD9ydcH?i<^@|NeP)l9b7%;iU5HL;9xE z1yTFuP18|A*fByDw3jQw6)9f->vwLPbgw9P<@vxjUvUYSlnSQ=?5IleDr{bMr$|Zz zpfRrP^LW?D;woQdLtCCaKh0AZ_EGK97nUo@xkMb)VQ;5 zz1On8ck(dtQGX+uc8;29g7$urAzaIbo`#d}R`U6UQm{;QPfGve^hV2qjij^koR>(6 zFI2N%mB~ZJ)`gSap>@(0`BtA1!m1@Gu4+^bmnmto%U-eHi;!`2O`R|W!V2~g1-_^A zst!EzS-0wak$#VxMgn8zH=}ar{j&)BWLI-JEz!@Zg-H)$zdaw^@I(w&C_#DY-}iO8 znar?sXMy`p@0V?n7Wl5R>NGB12e(?~x;m_sZSolL04F{wAPc<&}1R zMb-%DgT5ZN3Ed>W@rR?}dt3K}%+P6{lV;H&4)VpFv7O=#H0fnh$#2(be@4GT-sB0V zpC^Q>zY96!w6X0m!wih{cJE6zv4^>nWk(Ie$yYwJd(Qb~5trJJ8_(Ha-sIh%LXL3% zggnVRURRRo-Kw1*T#l0&%#F)xk*T}ZbFaxQDTGs6&z%3xPkvx{EUzXe6kKE9R$93I zArvUiA%Sd;%6w6)zqHbKgRSi46aM|`s%#6MOqRedQP}|=W+4B%WRpFuXyh<=RI>Cp zQ@TAON(m=Lxt}rgMi`v(m-5{{3UZwj_v4?wCpkps(;`JFB#bw8f86}U*1at zHEohc{NdIy5c`m#-f3C2EaGP^uvd(hv9tFiDdCaj=u3T2mt^_S317wQoP?ehv$}OO z(+&qyB#q`^x0H}AXXteqocz9g^BE^khQh25N$7~gx*m&OA(-o5)*{U zYo)3}fwUmjO`%uP1^7Ew*3QuhFr|qCFHKGIX1co`X`s4915O=eX#8SPp?-ZH*V>H1 z&i`A-&pL_)gfWbwKlhlSUr&i^#8BEZ$#S|jXIaD%gGj<85OI6!$q#HrYXvr@2szoy5``e@|=C<)$nI+Pww zpf${(iV@>6X-fOi)g_uluE7$zkr~e&f8K{9N}~G}aK5IgS07wi;ZTCx6#D6k-P4%+ zVfbvwNknbIn9-8!no9wxkaiHFus6q^YNf1DKOxpJOUq@JBl+5fqE+_APx5%qPnmqf zBO;C5)}I}-C+nRAfB&c~1)cp7&^LL=kZ#q?Sg}kg8wI$=mgeM=s>7ojRy|8JF8xUf z{z%M6HLS+Me5p*oX>(Z^Bxzq^6cJHCj|>NGkM_33vKZdzkQ=zZoaM;nN#tKB&C@Q97muC)d9Qm} zprk*XUFhR3WRxye0vkVJ5TER)^R>Ca*MglFaI>*PGWIV@yG1wAX^rq2WU~D$8F|vh zdAro^uikAlpR6AcpX7hWK6K{OgkPpTH9-8XmCh>$Aa$DomIw(Ix|U^)m{iR0{ImU{ zKcyF(d9TCg!+NdwPNj?r4Pm8x%qrLYh$tf)lpFCLTRWcX;orFnV5UWMhW-r#`wcSLGJX<7PXD3Kd+99?0r_q{wYgZm7~a z2uH2!G)vpba)T+a9OWK#Z~%})(ENL$vvZfZ^rq}~As~l^y&+s`h2-W^%CB`~R%%zR zNLy!y9dW#_YJRUmpLW%sj$4syqm};60^W400jb?aBJ+VcA$De#b27$_iM@87fd9Qj{bd;Ir}A^`8#_3ACG=o3H{V1<;Wjz zSvxfR;(60F1_JY9J}3P z9Wjt(U*k!NZ#?U6amQMS1^i{D<(V@k@yCy2VC@7(`l zgrIhN5imWy^&gRd>2Nv_7M0lT9?iY6HXB>psyyk-3O6i+m;}nIK2fMPv<6)8-dFdA z%bYKAPn&51i`$0;z*t}{mU}m}Q(SPCrE`uvtR8QYJ?@iIUBe5fvM4%Glv@Vlw$(Ua z2v@FYy||k%jmNRWn$$t0#N*?ROe#Hhmd#d*)?@0T4;!YXTf*Yv;DXV`T-FQD=D<}( ztt<83$%e9uw|)CCqA?!>*;H0`R?05ng{n5NU+{W5q)jG*GR3>aIlO)58|!>+UFU|7 zt)!=LY+gG`erqCTs{E|Tw6FRADNo^oyEsz)eWhKAe93FqLte1#mR(XEOeXvdJ`oDy z(B$83dlXWp_EojS0+xt>5~Uj)Buy$&bf;_<+8`$U)>xhoqQ~7lk~YUV|TC#Jo^qSEMsaM0auAp3jP2u+SQQUzR4DztMbUr1@AUu8dTn zAx`6EUdkf^X@Bc?6EgTdezTay`C)RQgmFb}q8f6uM*r!n_#>XbmT|bcGq{+FRwdhq zO_x8@^D$Gtr0T5UiH|rFHIv}5va5M`$4E7V_RdQwjwJ@Z&hC_K953MREN4@uHsdhd zYh*=di^u8jOw8@XrL|uV4%plVAJRP?wC+y*^|quSJBa_q>^|t9ILIO>VDSkjIPm(d zE5GCr9i;OuF1q~Fj$5)+*5{UGa#<&$Lw+s7tR%r@@Pk&TCopH4Gm!>=yAF4dbCfax zD0a~3!07t?iQa4aQ2B68U-C2p9zNy&8l-C22pvB+Rz9i%lCM9n2+CV*JbDRwIV3}X z=$F8e9WpgVO^76xw07X-qqB1*rDVIx>%8<*k33Bmnm;|(#Q5DoCts!-aAc$r&D(Rq z<~oF~rI(9~DAwwjQ{{NB_O!{iz(%K9tj?AEcK_DVI!&e=Q1bnk(Vn=>SF+q0lWv=XZV~ApoTG z`K@WJBd^;^kJ|^;$0@15U88A=2=sfXA=3Mx;?C|KuhV!e5Yrojp_r7xlA8uhu5LKC z`CF%O+JeKyjO%I_r_56T_J1^{12Y|0<>Y%Znlp%mCR3dEg7-EQ*mzPm1 zVqoe;*5*(KD$uPX-qFkHdGiC;5p>CJb$rcKit=}GvEYJlhu0-x#Y~?gQ_b9247Q{r}PNKO&KkMCNWwm|x?p>pXe3Rn;kKXVE z`K&IBI;2hb4pH>TNCQkFXRIZ9!F09#l8dmjRt?dGo8tTz?kgn&G0N9E z$j3TG_G^7x&X~=p2DONGp3O?Rl5!mi#=lUDSmWV0s8bENX`m%{EL-5pR1a&tqv$DX zy{G6l@MUzmhL3XlpN78ur#%5JIv(a=Pqw7-b{hbq3H1Cw;whrsgqp5fzrAHs_CNg! z4SHrHJb$9lti%Dge+AFm3p3y>(>1_NpMf`KS_cW-dBzaqi^iVb+X+XQE{MH$5@sl# z_K0l*9{InsUzxRWy&4#moBO`UcXKGtwE@B$dS=wttyG^CAcpXwHbyU9fOD3o)wmnS z^!=DsJ`skPmZ`l2+CONl34$)@Nx%M&AyI}HvxNCZbhnxwP;vqdFhDRX$dN!=vkc`l zJy%Jow{QVKh^30TE>5Q>GWg9S5MgpWGq4QZG^RF~qGhRE)ufmeM4lHYO?G}to0(fz zH~+WIj%25J12bomE;sIgh{LNbnBT|!z&X$u@!<#54?|{%XGr{b;nI7)UYT_pA+$T{ z3(C2}V&f2D`~`|2*15lKlx4FN1T6^x=MsqSnqA|UMZ=JDTk};K3bRea3;vx?in-qa zWPctn5pV*id}hmKx+#dH3`4X{%%Tv0DPx7YQsMhCrqjj8VitCuZymb=@ju!}`p2ek_@H z5M~I3O7paMp(4`-AX3Ckz<8c*VjIHWt)EvqNI!PDf%TCxD?>%uu@bNnW_1}?Y7}Uk z3Gd(>hv+|ZI~UqvZ4m{){lH^KcrZ0p=B4L5pXV}JThF(H>pf+3%TO0S@Melp{RG^Pco_=z2z38$Tk#ZX7S8K`OJ4=S~ zfb}B+EeFV{0z@kSty7_P?sW%e0|Q{ztlJHUv4N{4U^oAD87fbX52)^6F*EvIEH2e3-Uz!ENR6`&TK9H+R zR=WcETMm=f`RhIzncBlA?>UbcI@xJEL^=a10r@60&6K%O{JO25 zSDkx8T*uv0aDfdEJ_}Yu2s)cVkCcvk*)PKNhoQo^b%Ti(V%Tjt>Kn2H6UUcU-z+c~ zmh}PQt2G2>NW*7RV6yj72uWumnD;hI5PKtJ5u#+ayz#7b*<)c^8b@yat;ms~DZq%e zGId;Nn<%KLA%m8-Ohb@lM$XQFZ6k2-P2-P3ZJo6dE`#(0CW3(CB7X!Nq3S>33SB$f zbuq-%gkGF17WpKNS2{tlygAsI#%DcTJKj2Qx#A>$WVrv=l}zc&XXaiqIFes?Cv8h2HJY6KEvPea$3W~6S_<_Xt2Ik zs#5Vrr*<@O&dBy0Dd?=LGjO|Y{1d_}!PpOafw(N1dRbhzMdYa_s5b&c?ts&1Wk8~O zV^6@BuUvl?%@Jf37Ig=i1;Y?0U}@tuslP$Fh~c6se95XVhUAiG0zGeZ{PB5o$3hX- zh}o*!2am^&AP#ua`CFnA#2Kd=`|JbKqV1|_v?Qm?8ah}5nPw`nOi!(0HbgqY`>}Ui zcd*?%3$(+O41G9luu}2n8cV)<6;)k|?yOJLAw@TMyT1DFhnvS=4)>SD5o}cb3Bt=w za3KG~h;j)83}uh$$aw^Y3R8wAU7fmR2xSMe3HVjqUMDK(ptBV>Do#>tDiy&cjB6=s$YlP@QbT)b~QKR=3tls~d z=Xt*GeXsX>|GU?{_w1QDbI!~?_nA4rbHZP~P$0mk#z#X#BT!P5(?CN5Frdm!xY(#) ze`wzz8XCHSoviFjC0SX9moR5*I|nN?G`2VkQ&UnUF18*sGgH%^K~8pjn5V}3_mLW= zeqC*yZ3AtCZ9UoXsfLD2RK!aFtq!z2_4Yb@DGwTQ>W2@FiBl_wzD67%f*y? zx=E=;VO408$OgfXW!G*OG?O0C0>M ztT*-?J9Pgm5J*}_Y|in3it56EimEn)SCLCO_^JO9jf(5XN|NBg)gNtbZx8P8hhGDM zdkG+_yB8Pt_vg9y_xC*5AN|4tmZj*>YOyhfYD4PeLy{7)IJ2~~$CttqRdgt}{5z^tsC+-#lQQ%j0Jp$K>`iu!J7XcUkC zTYq51 zt-HI67%#7MdRFs!bfLB0(8`Xl_&D+V{%!}K}jrs3R{-+-~D>n<6 zor}AjvlGLge$C9CJ>12anEnX*-_PHAT6x+1OOlh@Kf^)|koQjwFFy|-?|)-+x3m8L zu>Gm|o9(Z0{Vfjqr!%pac3xHvdUAG-sHsM&CLzEl!Uz3J&i|?Um!^NSYP(s%WSt#R zobD3;dRYH3|9jhp)&{$(MZRHtY#IRQ-g^Njw8ylM$Yn%=V&#>-xJy3P3YWR7- zDKe(sC32utu2j4Bbn$#)AVbt`W8k}>%~0B#@rDVdEg8aKFs>*ujA2cxCmJ|s8&tEjVqgU8J6%JX(V_v`eK?}C)SgT*5d5FGhf%mLK zD*+^mhGIKJ-rVJ=+Y&OkyOS$ z;BCqK%V8|Xy0lVJqNt#)Uy5>IA%NUp$}gdy{rDj9hf> zrbS-umN(kcNAtOZ;92ti;Y7*LnwC3VFhn{qZPsawcmz=LT)?olgQq8yy9y^H>8G+C z$%K)okFoc=>D~>jU$@?0ONxX`(IM;s<;&FZq zyeE-e^OqF-p!vM->U7s?zD+0lOj5yI` zgcVGRl2S_9^tw7aV)p?!bS}3)EJIEOuSlz-P7bQ71iA}8Ik)I9Tn>4PT_RDFd9LyD zB?HRL#?d}~gB#?iq!ex->bqT#VklXd?t6Xqy`|3*fMn(5>_64C#yM!chvP_^bzp{N z3B)n;w%$}3dd_=Pxoy6fsj*AQI}V{VCblH!Oh2FZ%-Y?ACg7Y;XsUgu7h(jEJvm*Eu=X)bxK`)WH%TS90PI0==^S6Q|O=JW(JWJd}t8A-mB)Jcj=jF zYJ!6J&4Uv%4r#tSe2C*@LoNOhBAn8waSG78?Sn8vTA1T$C#8vd-;-7L+hn+6tC>%s z(bz_hiI&zH`Qg)pCe4N0-=6(>F%lf~H%mC{`3WyzS2$j-At51U2O8LppWY1Dzwj$A zD|=(5w`hKYO?0J%vZDvVk9^U5m($&bTp@7}PgE_5s{HvEw&bGKv3)7tIyax{TrLy& zv7xrJMCByk6V5AODBf*s!|3oj1`X+eLTI>#fudcS8Qxf_g` zffp1=h|Idb%nA&Qxt>3OZ(<8|xXWYP1EZ6ZDW-~vF>d_t2c6-7RJTRcVn~;gZewuZ zvz*R6N-Fp1EXecnxt?2m(*aB$u_9a!C6;2V{JKiBco9nTnf>_oz^^qA0Kj4lvmm~o ztZP~CDD4}_l3ZjLhc|Oas)>JQ6#}RfDd+|BthF6=q3=?bn)~c%X2ErqGh$9^lO02Q$40O)YjI1ahWCI+dwlNM*y4z4%CN+Vqb>&UEtJV zAP2Q&0oDcjaTlc1XbNHn~GmngwD$Mh`>6ueOUC`N$skFj;f*gUh~e zP!}nh^7(A5n85H!{n;qjo-PNIt!3S9XQc}S0NWI0loX0F=oG~HKJrDULZl_CUcc^> z;OoK9R1dyeH@WLP!nFwBr+}=DO|_goDT=&BJ;B(ln>ALRJKff+r0n-Ori|S$YqS5@ z&sHNbzs2i?bFiO!=4TqKrJe{3Ouw9m%)staZh7(#Bp+x)iIj4;|XO4d=Tqvre`gXZ|;^F`PZ+PDw`kpK-7)YJ=9I3~`-A(tgCSc}2Zhk3Et|-r9&8|Xmy^Mh?3ld0)sZJH z7oEHf0`$6wTa)`!6C-E>KHE(Xcqy&g3WPhJ2AOpk?&wHN?%z+Pr##6Khd=W zdfczl-)|k>tx&ec;?VoNEPRo{aeonWzqRu`D~OWCf;Rg+3|y&Au{KZHa)>o`PO#v^ zt3m}7bs2zV&jAm@pqRJ-CfcMl0x^+lili@xi}}|i>T(juxEM&@YNX(h3oQtN+1=KM z*2BdlNsev}l#ilb$F{PD434I4`@1tm#?`f)eub?k>GrLsnEbZJ0=-WY{eIZ-fOg2U zX_b~R6oi$`y?9nzC#nT(b2t}R?&5kZNDW*OR`ExxdPpt6$i_~ja9P+?k9$Cb?oS$$ z-GLr)h-cIjuo?##z=q#nFM0y_0SA6J-G24@Qsj8pifbKV5+{sx8daw47$RPo9+Udf z+Q5yT2jD8Ju1HXoWiP2s%}7x!CFCl%bv8UV;JMWHJu;!Lk|IX!Jrl&vxn=fiv%7|ZH_BP}g+ z#;@6{jEj#cflU@V)wpoV6Gm2$nWjYkKoj@5JcD4IV-HiJ5HLf{XN7DC2M5~mUJBhD z`(bC)en}4>5XV+dK7J8gMfE+@q2pzBLx)J@;|nj89cf52+(B?WP$u?rs0GnUO({rT zsn_M2d!QGMI<1GmdjjqtQjIn%?X@p*d@On_PR4lY;TL9?g&fE&Y_|PU!;!PJrZz?z z+XTqc%&AmXLB@EY`km_}rS0w(dDiu>&2&SCjNVWc-sXu@D{{X`3>_Bu?XJmFS$|lX zR_asFGK2?$K%}Qg^43nj-X*6$OrkxQ)Az?3UQKpF#cO=uEm|s8$d3GXD%zz=i60c> z$3stt+ohQk$__VHB6#0a0d5j|U*kggR(lu;Zq(J_5KD|IV8#ddhiaj7;=2$qq)QZa zf|AJ-?#v(@-UHF5NE5er0I_JpylS8DNiplW3>4;QrCLw~fCg;|+n?D8tON9FaZT6w z1pNd3T@iv_Gs11kK)Y(2;WxHv56y_XrD-q#VO9QPfq5?`?YxBpu#I=9+$Ss5owhC@ z9lm%G4AF<8Yi%woXuc0+7d?T$fjoQfk&uN{FWWEg83=5=nXKP#g zcBh=xZMMee#jJV3XwOIC-?|pBl0J;tv(}qHtGae%1*!@G>F*Bb=lq&}&Ds^tena(m zf-=beCz*?eCz3%hRvofNP*O3`={Z7Z)B^-kvS`E_Ja3SCicjDQvE)c#WK%gLz4Dh1 z=&1ZclBmui(rX_rd+y*TembFo<4t##DE;eKIr0jSaR`U`-SS9WiDx$jGAAEVvgj<nGy_`I;HImeV-b=^Y#o=iU9x)05ng{#(!e~;KyfLf0^>>UrVPpX~jU{PVUrfV;E&9p(M6N4cPmDAJ zy$fUX@8!<2V@DR*hAjcZ!Rjtl?`K|wfEmg`0s}Py_IA963VkC)dbF-ANl>2zCxoHpPn_U;v;EBjY%?0*%&T^X(7}nE3qb=Z1%5#2P?tvQegY;P*x|@L zG;k>7B|jo~sP5UhHnaOQtW$@5Ngc!K`#J?>Om;TdZ6qMutT65?nL7;I5nCU~N{*Lo z$sg-7;@n3lGmII|i={FHJb)&aZ5p7ot_+%r`nVW3)D49!eW%#+^KL@DVeUXZ7U^w31J`a6ZFQ| zha~zS-Ckgc+MZOa5~Nc-cPLqc;$c(Iy=b`{bZ+=c_l7VuO#@k(6y!$NBeTZhp1K@T zC2xGz&8(~b=A{eYXFXAC8XAMjHb;DN7+K5A4!?(N06uDS^AeSs%b`adOHZR75_dTd zaph<(@!_ZP{W#p{t{>z`|N0ya@q_Vp3uTnqW1Je}qgEZFs*X}0O7430;9d!QxMh{R zRPyW!YjgM}>%K&0al3(l6e-n|uJ^dCS!Sm}(IpHy>k)rn|Pf+B-^b=EMu@Hh(~ zJD5Ch9ZAVdJ|F&uwiu_tCQF*apFzjPIz&6F^g>3Mpb$gEGR-~@(;v_G2ZVS$4A+s< z-%)TW4b+gJCNbkdzOvl|9CFu6#1*3kaM4ESi-?g2@nxh%58T;Cl3;7pDot~ty=w*y z@XK&`2qzI@FL2C{bxJ7}S!`gm!j@A%C$6f)_zBVz9fx?s)tBvjoR>8!5-V9cDfN^L zn~I!j>w@yjf*CSeV7qQ4#0=rBwTAx|;}oIJb4Bu&^p*@H=NtFrM9oF}3ouY*0^oCE zw}vAyT3ZpN)oNpCR(m5&7=pplRcF7>DjGf9hz? z%bP*0r{Q>%DnwV6ZG!&H{#*Ij&Mh(JtyMsZ_JAds05l@W;Eyr%(~4G9(IY5ytI3t& zl^;VUBD)k)b^AUbIUwb(hr#}GK%@mLi+kb5)Q?aC5JoE3TEM&NlMaWZQsC(lhi+yc zyXFR)Hl+es$=Ikc;?lFG#_~Gj=#;PxByJSFuRkc5S?GWY!hdGjLJ#|K5sWxYt_;&> zfy2o-v;jY{2~*wlX!BAeRQyi6=WH(t)JD$doNAh=c(+*KlK36ebUN&MomNR>hQyOA z*T`Eae0td8C^71WkNPmDeJC6R_P(u%;RJ?(A(MX{c2UfHbAcv7J`gqat{JiWb! zxl`=&Eky)JxTl-YSuHXBL1pk(OTiuTjr0mo!lrg)gJ$$i(Jwu-m;sPP2$-q`TmhFR zxV?LpIwPA1w5CaGw$Wz7b@u%=yr^q(e*Gnh5^v$kC-xW)9}7GoRxm~k3i4=xD4Z)K zB14R<4~9*fKfkY!x?^Zs4slu}ajw9&7{;c=W94gNwNBKZS#Izkr(FpkLfIlYQQ|#= z8g}pz709q82Ihx??ixV13{8u7rOns-zltKfOamREq)qJ?m)s8P(}I`!0Y6+v2}|Ce zQ05IT-68`D%6=@b6g=?i-zE0)TYBNPbnW zg?|Im4-s{1$XD?tQo85xkgh3SKodn{1pt`#>4b zapQ>=EPCM`WI>q@zp5P=P+JcFT#3Wj!?>cz+NUUM@pMgO~om?mv3a zTeB|Bs&KL_h9EzLp>XCAja+VQjG!eel)k5sHGB@Hf)!V(4Nz zUFKg?Q&UgGcy>w=pk#f9u^Pmigj|0kD0jud`-<8#;wM4h!x1Fx9<^RJWI{j%f@W}YE{c&JLq)5JkNSdigER_!WCMT z$^U^jfi+KoDolfa15~CJDY7J>Z|{|Ggt+P)YSwCQ)!LGhe{w(s%21xV4PVkShxXMI z(*bSYo$a}rh|vmKNAAkSm7&d}K*PkOr0=`}-72+7$3Mu_dbURM)|-xpB3``^`V|^f zUWLTcYl(Mo9YtcU5@bq;iRtd76ehSn_57Wvl)&;mM!74Z+jiL=`Rv&<0oYnk@NSS< z*dZq`uNoCjP;USL#}c{L>$+OeFdnds=c6irv^t-doJ@Cg>dcuZxp<*$bfIXdq@?ug zlpow%^X+a6{N6q9ld!g?=4wmqbCVo8GM49CPlWwfWk0u#PHmq9tlsPC*NleIy;bCC zxiFua#A$%G=)9OY!NUgZ^tyea-A`|+^Syr4ZBeE}{~i|Ubj?P4Q}X$gA(nQm2HOvaS?jY*rr+8ozBIDTMd5H#IexY8Vbr5dFcAT-T2SzQQAidjz`ZB950= z@MOvM8AL=x7!7qK$oi;ywxq(v5LlUVr!lrxL{9-fIH)6DVzN*`5TITKy!V#BbbZoF!I9|u;oof=T0Y^;&{usPN_5ZWoPA;I zlgk4fNbJ0(*=iW2OB)48He4PMtcM6E?3~Mk4hd3T|2m6~PNsME^)K?v@9Q*x*T6JG zaBhk_ure(k>IzA8k!5{p01az~9OxldkA*zMySv|lL%hEiziscvmU2<<-1KF%=PJMz`3IrR8mnr1E%~(e?o#15MtHsqg6bdtZS*Xm+|bSI?)FP5XK+3mKb#=XHqFx zqB^0Mx3Q@%0vBI_lIeUMR06NO{jZj(s#t&m>AT5%n$a-%cjr`>!AGtX^x;qyLCIuj z+~LK#KjL6n1e9p_AY81x)PI8Z>IJDgzbk!I+gTwwd75> zKKvHgnzdjujl2*#O<{fY_K_h4CAV7XQ4TsOXOT?(AJZP~O*oAvK^07BtskVGRS`J> z^FNFG5+L?1F=Tpg-X#r-GI$!qIh5jhDdV@9*`r41)Nb)DJm@|=^E*1_x@!-m;Z3|K z&a+Re#DbgA`1tsw9ayTRgfK#2L;TDa+tX?q0NGc1uTsgQp(T7> z%7$s`-aa#?p`C6lwnMxYR+$8oHyk@S+%?XG_nNv}&wuFOmS(+dwcRJvz^=YTkM)YdnsGyFZ1xKP>s%{D--1D^l*t(t+|Y#g1_^*w`FA6f zmSoIiK`fH>r0o_~*$(Dj{;c`{Oav}7reV04rfGXne3rwlmnYlLiKycj;najEeU1rp z*wvwDcl8-@Z+`4&Q8DyFhKi&$1+Xa&2#sCM@anS#No-CLsR1Sv<>e|bm+VVk-cj?L!upbJW6(>{yfOo)hxuqeatqr#`GTgo|n zyjtHt*ZFW=P#?!=#FEnU(TA8DD)}XqNp;D@8?B!)=j6N3-wDkY(}$9DJ+F$sGV{-` zLTxbYTREy`{*QJ))8BM$sU)eCb?1NhkoUg3ROTkHcnd_U5cjZKYqWSn;md+^u2)Zr zZ3Zd#aq8O~M*hDi5rQphicHeRA-D7)Z*5t7Vn;U^hIqi0xJ4_qihtgLjRH?2RB)B$7Lxa+}>xy5xB57Fp(3F#L)Ho7%ik{rDUV;(Bt zQvN-+%YEnxVbZPTSJb8Lhv3NS8a*G$a{B1U&ct2FRRaj__+s_Cs*vP`>d_O!AD3pC zs~+$>OWQoderI@7&Bb9T!bg%jlkbusj6NC%ry`V@DwBsU>7u*H}dmJYrfC$gEh>u7GSe>vxdIRIH##**M#B z&5ItAw}mf)2rn!G5o!m#d*c`mOs4_mBg@B6=rT?Higw2=Qif`&Oa>2P;F1{vIay5? z6Wk=3{y1QH*(PWJ%POb0Ywr@D*nlT4^Nn=fpqHE?VctJE2cN-GcOQY#k3KmJNWD2i z=w~V(`jV(gAVw(-Qz>8!JpmQ_n4+YY-qTbJxS4~WM4om6v67q9xg=rcSpY%lkdK>2 zjc>lY6JDh*eHbVd=^krq8%AeAbQ<@XiaHO?&<2_2Jx?b7?j7X&eo2n~QItz*v;h03 zISv=~W9Hb0&Ni|c@FuC0ar}r!+}q;i3!)%DK&5Gj(INroY=b{Ik?8+J$WfnCrKQAI*)C~R6eiVao2lC zes2oTT;{BE7%|?fFn?6vveMSREIvy;6q#m9w-q9PHVuA)2xw<1?2d>2I{!ecI4xzZM!25~k2B{(+{V6FsD97i3Ap7CfbYl*3(a3l}X80%i= z8OcY|a&CKH@(f`LDQ^ekk^Lf|UyAMqabJL6$p&C$Ek|QfdM}q4MTqFwMLrB86p2qO z47HMN-YY4IQ)fn6oqRS2l|#4p`1j0n+D)_PoZck_FkADjmU^iEn~QK z;IY=*toSd#DbWJw(+!A)S)tsn-(V_XY+PP^tSA${NOQX0oGqlR;xg+BpO3?2r%Pu< z_ZEXmt~(WBiJv?5gsD0_KW~e({BtA4?1LBZ3oDNu)%VP%HL;J=uBMfxn;HetZJETu*5OHB{8!FN$}`*OP}g9>hY*jH_b3+!^6w>Kc?y1u33In+DKxIb<0j~^D9UN z69IYyw2p3~PWuA#-w?ab>CjMFMop|RAYxb@?6B?OkK(zxQ-@>f1en(Ago4id&k)#@ z1YucD^*$eS3&@6*rHaawmpr#7g>Xu$uQq(35>6;zA6vF5q&OlZ2+r3kPLqDr-fiaE zMwPyfsDm7ntq}{L}A*$zm5i4Sp zEtfk&oJ!aY;yF1uEPLZ0JEdThbagk-*nWR-_oB-mjhTvN@Ior>d|T?N%kW1vKtOuX z`PR5xar>qh+tKl>=^L*@ASRdlv@^pYy$WP8^-3c7vvMFsb?_jLc1=Z=HQ$B;Q~TIJ z4Cd-;Wa975D2m@i3NCD!&C7T-7-im|p1x1i8%OS9anod)Er(wIiq9@}Z+EKuUch4x_Q|yLun8q4imZ+neE?dzz7>kW!x=8?)op_O(s)l%roUF%PrWl_-eTL1nhk zm?V}UfJv46v)tap?%nJ9$j!;xmClF3^lz{gD*~bCkA0ZhiFP4^QihKkZ3Hi39wml~ z(F>eF6~AoSN6Q;1d!z~X5S&~;X;9Uh5nPx5ASk_wAP|aTyn~1?^%K#ud7SK1N%YDI-cwVW(gJ&J|<>;N3O1+SkQ2>ofcGLgy#&d&E1lLwS0r+mn7HE z+G|pInaAKX#;~X3j}^r7RXrcn6k7F{oPO7cZ!vK)Xx1vrdiw#qWBYTSlvco4px~Kz zY60a`K6~^~X?ne`%o1zUig@L#hucP7uS_mfZl~i@daz`42z@YcLcc5;bBXI@op-<4 zXnuFz@Fs{qec(FJu#tgXM5NAhQ?B%c3Jl%cU8njzGOhcr78>VcpJmu3D=sSg+QuU2 zunp#vQvIc}URqG96d5OGV`3M_f$sAN8fS{AXBKVySu;!eM1#gMCU(~(1VY;V(o2TE zcj=(V)rKurg3Go>qR+G%j`=@ zcLtSQw)+=-TlCc>{zpKzIVPiY39_LcNLALWxd6au7xtRoxyOnWJ9v2hlj1Kcn)%A& zdxd~pQsv6Iz77uiE$lEr&jWG}@IJ<4qE3=9j2Lsv;kKC&t}T3u1OLHWs@>+`r9CK< zRdv)T^4-rm@HJwI=8Uh)&}28-Hxf|aS&P~IxE`=!=$-#fmPf3@;6qAz3Xy|T0(aV6 zjY;^;yH2gN7W>2juAbOl*5`3&qpMpk^9jl7S1l%LwlbhlLa&duw`~>&QCCNqdpb%H z=l%Pf4n7@vZA*ZZ-MNqKB!*3rhcDG!j_AnRW$A3~9UR=fY&FAI0x4$M2s493NE9S? z!YZ*Pc5L<~-Vd7w9gSwFfZr#5044Q^lGY=f>7$bkHpsHJ-bN({#p+jE_mZhW8B+{l;f-edyndGv zaFqOKTIE?$} ztvKY2uU#6v?zLv{Q%15A+?HK`e|K%QF_2be-iZ%FrCf0a%$V02=D0N9UF|w?Qj}2t z$$&8wJHQPYYA!)jWz*_m0&lbV9Q{fTQJ156Or-ouU?>tRGS`yUJz&_yvGi*Sy=0R# z{Cov|xtFRATg-eu^KzWYncE>wa)P2Kr-nGT$SQK=`JUtGskMSB0FaRe7S2d4rdIy?B{h@Z6j zB1YUllVWP9Z-+(*?%AqVZ04C2j?$R2Zi{>Nyt%AS%Ajo&WZc91PH~%OwSDggguMHI z7^Xo5G>v~%m}L4kzdBDY>JMR!O0Zjv&^q$Ah(I%VQtLRk(FH71%7S_$X;wo7z~4<8 z-xlm$>_?3?m^h1%8N{)e+^hF;7sn1E2ID2F-rYCgH;CYdALd4vT418__RTIN2VJif zpW5d?Bgh)(DC+B7Wff&pIj;(CCwFag$Etmk{_gj8{YFH;U^<9H5f+1Pj_ez|HV zP!VjEgyz#^9r_yPKN0%Y!<#jiBSniGA3W7C9?9oqpslK9>31yNnK%d-oG+TI9cx%y z0F6Hue5gc=p+55Oe(0-Yz>W|Dbh{SR?qeLNjmil=t}o+w0&_CkY{S-Skc@ z1P}kAO~ppS-qi5QAd#iy%Bbl8(F*$X<|;`VGt?i)1}Z{~aBk^TKoWGFNAUoOB{N}H zFljye;p8xPLq9M7?FT8}X1opq1h=p!&?@oyINOQ32`GGQFoJ>!`|pIT6^e;vFhj5` zEKhfSqUNBy=clg8elP*hqLoGP%RsELOFNe9lSI?bn5j*kCsOH4Bcam;@5EevMYeOg zi{Ch^yn4F*vY_jiOG)0(UA+_4`V8-lv={HCndd6_h5RRU2l^HygAhxhBcOlrkP z{Hp?3<1N~^RXtIQq?aiMavFn5pXQy+xU+N`nRavfEvq{;LpHrCsBB z>*C?Clc~cKG6CC&XbWNs~ckfIt;E-3M#qaB?uuGb>M!*jUcY?U2LJtMQfN7F`Eh@}S zwxdicFxNHNaEpQ`#x~HCW31I&ahz3OFoiJHarikRE|a_b;M)79Lb9s->*QDahV|Dz z$$bj6^eW#jC%3ZoQgRmiUOOi>A-*A_i+(356b{9P6=rf&yLT}178y2vrnV4M_up-} zCyKu-I{0w2gcpak>A6d?@VZcqY?;|^8&25uEHEOpdst;U^UtyO9*kF84v}=u=kvbh z?=xjuuBpkU4_=>vOO&=>y0*Lc@!b|&Ut9%#6FR>2ISR%ZK;uUlY<%m-3#jxo6kgbP z&|*-Ais;(-NNuggNdB(19Z2Pub!xe^7Mbf5Ai@rxYyA8H_Z>)I-n=$ZpGecR6j;JH zOc8d_xFi@U=(QBM4PQ_+TM8ODKYG?j7AXjZOLQDT5Ylw?!=GGEM*2&`NPpLI@$K4P zE?}0EgpaL^#{miR*mb~#Z0@H1Te43}#KjfU=~jy6PotyMQ| z7`!TN$r>h7=%;sB9bIDT45EJ=lSEuAXXta(>15uX6^_{j4-J~h%3k$1QB{rl^wTm+ zI28B+tD?(uu!hQlD86@5IA6F* znAc6I%*g*-y>Z{UG_$<82YFhTD5n8RtTp_iOOxq3I5&h+EV~v(J zDxp4hj?n{sN-6xWAG_~R#1MQNM23V-_Iz#vvV$_1L-hBNRb{b0n?&V3DZG<3ku|Nt z!a}d%&C4tbbwO+dhu-5GgYgAI9*v9=D0Nd$ZrcNv* z@Z=Q^#&4+2b8-K~DY&#q+2$%AEB5y3zRnZxkfpKD{YGZ%V)tRSF=oC*<#bS>v3{bn zxv&n)=z7Zt>HelAaRM18>pCT5B$TZ6NByi)8jjqf$uStOf)&WGM}eK=j80{yiOBwULfhfp(0ZZCb=oxnto1s)MD{+GmP08I<(a8c=5GK z?}&=f%fg+i%*z$!v2T7to=@<~IwO{xKEIfkiH)0S9vqY$P@;`Y{#?Hh-1M|#saGw- zrFWux!M7Chb}@UB-=KeZ{PxRN?gnS=jL{9cm5XXmivwX47zt++o4ms8h3dFqZhts& z_<-S?aed|Z?x9CFq>V9O-Fok^a5dXhwU>0rac$Hs%wE@b;IG%hwbZpI^P+TGUA4P{E+%q`%S|Huhui`w@xX0yUHn1gLVRp z_py;YWcu1A0utp?EEF4kzMLk>0M#(d7cbIy;! z0KZCih`Qe8JqNn)$oqkC0l@-P-LBFt-+Rz+@`lwVT14~OeO7MWGjD}5V{p1biuE<_ zx+OQ3DnaDYlMmO^fz4PdL`L>mKAVsB5h_!?M0LV_PhSZXcx_W$_PnEcWG|n5b1aBs zsj6S_oX$1S!MQ9hx)8h9cHGUx+v1z-6dlptgIx$Nn92nZEDb6^3(FP>a2OW;)Qx7p zO{jNIJj(*%6HB#gLZz5mT_V?=o`h70A_Yo9+Mbb^(esz^>a^hT1Z{!<-Gp1Ql5{pQ zRAy=!gmlm?@)A=~i5IkC>6(|@JI~T4XdV+l*PcAaz8M8W8RvbFGaj%BT1s}wQ2Fj0 zz}Q8H+nCqXLBX2JVqChN;C#!rP=?2kcAK4q04o8Mqx1ZwESXe(u3;|;jLxgHZt3t-FfWY70Do+ z%#q0%SR$@kWI0>Qq|J?hIbIN8ecstV4k^tm1p6HANmafo-p<+AupCOWHfvw*;Ts3} zePq-Bc#i$jh01AXq8OR))VTcm*%g$GjEWM1T)GKzcKDXvoQIjlwB~;jBZQPw7EIN; zOvEZb#L0us5<2nq<%0Y22y7pF4A%{XM(eESeUrD$znYcES0#u&_fdecYGIVAkj~7v z+^NybGqPIO^khc!G|y_;A8#$qE5Xe=ubbnj8*JC>gHCQ*&POy|wuY}fEeD!e~ zwtiy#rtf~8vTr&aYB!OTT=?sG@$c6uRmm*HixoQX84XUi#Z!sPw(;TZnZy^MGztBW zvqSMWLq2vr`!3hhD%BoO@=f!hU*wgw<)gKt)K$9O@@kBq&DP%bQu2Nmnf^@nWTs)! zw~6C;Tt2C%`udHMt6AlmJ&i#*Cqw5Y7EC6D&VOhq9^>{ItcGaq=8rvHZ5~|=%gY-k z2&E~7Y3NgSN!90367_%BF{+Yw{`R&uM_Eo+THdDQW)pw@isG_^{K*83{d#-1$d_1d ztvsKG7j#fof{arEj%FhYCw$9-XRFP~Yd=F5ks=OcQ-zQN63H|ldCT)1BhYs_KS1Xqa*W&?*9D?nnFJ9 zUU?x+H%s9lE$M07;U)j9wapBwn4kk*4LGpS@ZpEkw=Yi=c%eKdS#E{tg0&iT&J1kK z>swEw1-$7;IvV*Ux@96_KwgMAHZ?XHYM>g^K4weT?BtU8VK1bXEyI0GX$Z18Yovoe zT|Sr+PCxZEg{{8U=K2F?nG4y2iQ_LNWw6dMipZ(poN)J9@>{S1m@gkewt{zU_nR@q zPd;oEO&p-VU^+QCEobst*SL1~i+thGiRb<_lilz_t^r@+_JaIZXM-SjZLX0FP(i#= zNdoJ$%ce*tiLG~Rs`g0-Ngh4xZ|8>C)3}kd%P&Nm^IuBb50R}E&Y8sPHVUkF+wWM_ zI;DN8*z{)CJ8jU~I`DHJSNzDKyqePb2#Z5<_T;;|?(6gW-*oz}htKRrWH+;VMhwck z29gUSu|YHZ#t(Vs$CBW==vFvPP!t=eezizB@?VEgW*iR;+%3=Kh7> zT`!4MM+o1&lSxu!c>afsZ``jM@T}$o4=w5DZ)G}P z+?k^NYTYs#WDTkJ!cQRpp|HLB;*C z;_C%9gb1`5b!Qye;ILjhDt@i_=Gl>4&w^+O6gh1Is!o`pp(YK`SL9rI0Q)PhX zm^kZ#hZkYq%*)%)4-z*r(;6^nT@{{e!*Bm#vs1M}nb?0#IXR1t@? zA4xr$JZBY64tS}5M*^t883M)mTJpO(!RnnHY-H}Zq}Fq{wMn=fv((jBRA}L1Iuf?d zy)|TsCzf-s{7%L*muc!KkB(`NPORV?Uc)_jPL=$WK zHO#6rd)u)pzft&Dvzm(&IxUZe zoo5)H?hy9Y{Cq6%s`cWg*o{I-Kfsto*+h4B<>&lUD`owVy-7D-Nxu#Lw6Cx=giZQ^ zQ&YogE~iy*7MyWW^x9lsz$P;t>XeHZRQ{mK1tQw_y*%h6qA}#~MbH=TX(f}b>};UH zIP^FB8?=^!GR5DjcYOUcxic+dbgkE-wL2X*=dI`|yoKrVuHo0+-YRTLIS$#RqhIBh z@(WU>TpKxZ;A6XhuuoAs+cF2aO*L1&WOgz>Vceds*VL84UmGQ{LjLCdGCc1V(c_Vo z$TKe*XRYggj=qMGImp@+mtGWu|0lMgzRO_ z+NY|%C`_FLa{mZ_^mh0Wf)ZIfS7-5n zz6*~4I&1R0W0AK6?XvKTKy*?fHDF0hP`#zgbmMk8PvzmStp`_YSgj`5t`tO2JE(!A zfL_ppmB{rOmme6PxCTs?{^j` zM6IT%=9$JAPI-48wrF}N=xK&nl1C@p7BE}xg}$mgeXXtDa3j4*!$IVn!riY^Ec}zr zPp?kH`OC+m5>VLM{{zxMEx+TB+@)Jk^xmwI#qywTr;^$OGH*hqT+wrr)z11l`S62( zmiJAYD=+QYA%EJWbab@Ig;QqAtnzWXdaUP{ls6wed!dw$z-U3fz1jGH-`2k4=6#r9 zP&t6S!6Aq2$Fs!~4mwL?Ou&)P_uWH`e|}!WXZT}8(Fn9 z$yOZ`^V?h8%~#)eR&TJYpCcb0aY){^X0_Z{G*xyLmdbRkMqTGf$*60Qwn`mF>vZyA z_ajnXw?Yr4pDNpm^vdn7ABsGDqjWBLkFI=bXVdoImou8Ip1@jMU-^P*ejaco!z;rD z9;3_Rg5~!2Z23}wi?YBs5cTV?9?{YxsM^&4002M$Nkl6AAm}IOSK7H0kW%=EhDb&Q8;VT{>$JKAD=Nm(oc4Gk+5K*taF^2OpEPKmSr>%`=jgqua4uKJWz)^z-P`VxaxeWN_wN zz0D^+RveaR=ONvj4nJIG4R#~pKW=9@gMPe-{5tsIc<}i7{BS%tfF-XvKWuTS@}tpc zWtvZU(O^?WONgg`4j@Ws8Y{QfIKMF2mv*dFdGUb8J0{Qu~Psk~Et&lylGUZnjcE|(U z^h{A!m(1S3Thb4w$zS$2%475LVKU?d-5&L$#aBqR(||nl19uGqOdi%j(7l zW}PBG)LVY8zT-E-=iNr?YRcCZzg2#{a)qqcn|;RUw?l4y$-)?JJkd*UGPw_0_v|*;rT2+5fH%Df32ea#7*m5a5964xB|} zM+IA3nV8{{8}t;@FQQe2{N&-RDf!r9%IY*>U#>;V@5h6S@+g~0Dt3P0$BV~sLjT77 z)p{Uqiewj_C|&h>nS+k|YB@++d!yvZ2r0`bku4e7GBZmrEYKUTbhb`bm(7r~tE!}Z z=U(Z|&>bqefLYw0t!YY2{cF-(I#Ft}i>0pOy;8Vug*4y)N7;McmG1J1zG$`j!7FsW zWa~#+yfbztpihD`5r-1ZZVl)noG2cvpWVR7?i-urON1`+a04NHO7tQLqlsNdoW-PW z#3kYkmz{;Bew@pV^v;3ha~iSk^cYYB#3~C>9E`&ehHyGB4#v>=%=oZ$m)P?Myv6qg z=jZXmme-Fv8ey%)5!W=Pk_PJ7zf&0Hx7#kD>}N7hPG6||smt}0uWpOb<WYLa3zcPZu(?typMOmz>9;|4 zP05l=3M=IHof~CYi=M5}6^9~~MR}cRa+`h&>V0aIjMrInQ9AW< ziT{1!G5NDz1H7?(l)O-xC-+>wL?-K&%kM3HP_2)0`H!8)vI zY*b@MSNzHk?34G^u9sg64d)EaT=#nnmq$*_xtEh7$X0)A#G2MT90RS22)g72$h2&^6o-1k3I)_L`2=; zc3HhkubWk4;_DCVb?17}yKK)gS)F^mjN^h0Ya=JfqfKgj=vfb`7$f(VoFUh&-Yg^a zn*(!Jb;z%0Zj)ttNOp?e(sNqrWcj^rld4skUfY!^=iTyWnYwJFy!!Pea_ywadb~?N zT03iod@esz{(R#P-Z`q@I6VQJ7${80ryllVXn&8k4T@Iiu*?CeFG!!;=SeU-iU(HZ}fh7;h zSu*EI!G#i=3=Zg}D<9C04vMZQwrHK&yVB%D{Q%+07PbKEmNBI_x4B4aGdpGS?m9_p zIV7|7J0PMfAUkxkR@F{bfIobmr0uMf5zjv@*LN(|HIpv=22Zve&|&qljxF+uT)iM- zZ;d>%rCHuNvqy*Mhgt9l9%H!0Izt`}C5kzd+R5TU|Ag9}jFa97i+0$JQEL5_*&irY9J@yPEaZ z8eLv()9stun!0Jx1D8@fuKSC|dPlhX#PHx!nN(?3#ag}H=A}m6x2m&1W~?TKILsV! z;JDyGcMsZss7dbLQYYsvD3zxUw8*w6R!NyEMQgjp$d9IF%ipapmV5PJ>!Pw-W$&7$ za-e0qe78i6lR4+h+_rjIm0K#gITsWm)pc!W~juI8`=e=?bfEt9qZF&1g-Z zE#Iq}DKC8XQ*zz9HL~Q+$K{e&>g2BZ1#(K!Sh=ETycFoQp}%NqmcP>(*s7YXvgXPY zGitPa?yN=f&bjlYM9*X3$_q!zmg_$whwJ4;H9*?S@@4wrb&{*+M(!JXhFo;xFC@FGRk~W$SXp(8%z5Dvy@fkV zzw6T}3-v1Qb1SFFQyZR^#fSIEyPn%C-)hQ}FQ1$%Il4N^2M#$us@;Wd=6Gjx8Gj+( z;(}qGgU(>hCp}fX3gQ7b@69iZxM&VI9C^TH@7(g6A1y!e2IKqiiXRHUA8+`h;S0h- z6F+mLi!Xz7Xp-$oT2y;}m2~Pr+&zbf z#Grgqx*I7^L&dp{r9xmcGoY;Iu$^q^Rb!~IJEU`%aiCjS(3w-cbxKcK%e9wO%I=r; zNcqZ#Wao%8MfB*ymF)-RAIeMQQwJ;L?z$qGyiRZ1Ii%MsFU*h@{XpNU%mOLtY?TA) zS@LQD-wM$$*5{PUiu3|GY0nlpYyVogc+YcEeBUl9n}3$%Wa+`!j3!w%X1AR7%Q(!lX&G|O#9W!F{o^x@dd@>{u-n#rP>Q>> zZ)OzAKkwQm6YCqKZcLt(HN7lnr8moiO=IMPnG>YEs8-5GjF-cwo-UOe@0E;si)FRq zt5k*fbEls#4Z0H8cEj7{Un?5rA9l3KhC|KrgR}MKtX$oe*)#5H*E_;NXOLC^BE|U` zE!}*}W6wd~&l?O!g6%6Vm_PVTUZ;Bm5+dOYfb>2FVR^P+FOUWd&W{U*`-AB(-N)A# ztfiT6`GU_DAI%R9kZ7D4j`3K+rfF#X+UqvEJ*%vNbhS2hpP3kFFs1gxLFeF<{yd5mr^2mQQ%1i~_A&#NPfpvRy`<3>`1$R9uZ_ydlBK-*2 z{b`p-&hBQZxcxD?scMG2Gk=n7D4ZebRrkpgQ_q)tRpJloHjfLNc1lxLzMS5wv#dG+ z@Ymc5S<^Z~W)(Ecf*BQZ(bgAb*6a65Q}w-a_4B&ex=HynLyexQv!!-cspL0j%ATx| zQa|oYY1fbb~LU?!Q%5?c1Z142`mV|30~(YP#Eho~bn8 zkS_Fl`i%7(X6nlFbrVTXyusP_DZY%(SEHw@k-`mtE*)QS5Rc2!tyU8~!^UvC#PDxW zIe<~z4d-OIs z(SzlM(j=Sm56ejhbL7I^8>O>FZ)ntyerBuje|q{Mc}#DXTUDPXg~gdtGQL6%o_wyz zv)WlRv!yntL}r#&=plC9-8D(Cy4FdF^9pn_;Oqr5SGQ|E|HwhP^Q!Jhn)#M! z`7PdjOZR<)wR)Iu@dn!$4qzn!Q&owf4K$F|7A zye=90(pH(P=R0!8q{+diN3P1> zFV`>skvv>>xtv?~suVY@mYUPPDTnmCPG#dt!{ZZ9P|Jrgg}py2AOk zvWbp8U3%kJjxMAe*s)X2IY$>r7njHaJ(q&u}7 z7XKDD6!d9N8`FNR0OHY4e7KSm5`{Qvwt(Mn( z2N2=U(#^NJ*t4I-;wTeEJzeXZrn?k6Ht3kD`wY7s1aMA8GeJcXX>S;yXXyi{b0G?@n7B6q90w# zc+>lJklh&^eDLoLoZa}WE=Erm zo!Z>_)F#=mu|clOZj{312W9p+{lK5bm*|K7W?g==be5|8`t&d5g1Xn`JEQbwoWjec za9fVl7VGyg#udupRV8wE=SmrKXq%M1v`OA`=63m9#l><+w}dp*t(JeTxmW&qfi8LI zwkv5@Wv8QA-mX(2KmL!c`Ze-)dGM>{vbk`AjOoaf5nbt0p&wAo%qWr${T=Mekw^48 z+(mlud{TL_oS=uuf4uf(xn8f$om-(>MhZCt5?xMFzhqpuh#5Hi$pAnsolR9^)z#IYn>D*?Xu(GL+9b86C2J@)Q~K z>IB876AXGGLqU#Sqm9$EQrV^RrE}LNsZrb&1)1`@3ybC7meyOkp~7i@mS;Jo&vkkzTU;H+7o-XM4nf3KZro{1m;*ofu{z4* zwgd1Z6mj4Rmcyikp3Rb`)-Gw*D=3@wYROhr^!1yBZbpqmM$o|zJm8Cm4(f}ej(j5p zx{8bSlU~XLTdLMbB_c!G8yclmX>3dejKU#;9jlEVMj|IS8KwjINmR{rA*$5wG-;r%hUeZ z(F*|{+NQOSBY4sa6+Hk?xUGf!pjU9hALlz-J8yg$t_SzI?g7{q1k>g&(e*Y~g-=)ibj6s_YC^KDzV@YwbEx zz`h4fTTq`Vqh4u|LrcmedqSCXtzPa}QFh9yGVW>p21Tt-u%a{R`Ow1?D`aZx25FIU zxmZ6iI#zKWQgx!Jph$0lm?K%6UzTn9G0$S%zBnQ?O+GYhg#62n7QJ;&$J=bxi}ZUx zGH!A=tMxsGF64mEfnjBUAZSp`P}cbP3d$A^CDIp8#9=5fGlHPv$}>~sNsGmi4<#SP zlG&9#d-k|6cuDgeSMMHDAyCtzv*1~sl2xzwVJir5D0nqBHI7#BVF<sM=o1b!QHQ45!cfb z`WcPxgO;(gumj<+lyZ!mw5hlbnY?b*i>h6jcNCaD@44$G5u?egP8yXDm7uS@%Pz9xnBTP5%D zcA59V6QsRnt>mwJST)~iE1+}6-GHA33; zYwVr++^Qc+LpC((S+&C&-=s4Ag`L~~L_%(BNcYDtHx?b*aD^ z3l91*I~VXD*YDa$My@;0q({Q_LxF9wrnW==-?>%t`ID;T2sv?`BieC>MH5l@7k4jt+jd~L-sy-t?LZgfA8IrwZBH%KJaN7k(VV| zI?$lL$x5nH~hlg+1rIiIDC!OxbclU8m074gXKshjM3=o6~ zvmqu}P_~#U=3_M$#~b$r1o3=S1p~y7@2_P5%a1ZC14XT(qT+~h42Od0z^E&ht)0@= z&>>^>LqHXI-Fuc9Pz?u)DIY|NhYrd>nMFZnS(?04rc9B_%E}{Y=^>)^V6L?3EX7_u zOxc9yuC7S0#vHGLyQ{T9^2e0P_;KT9V(A!R1teZt3?4;*<#AggcZz{AqD>E;H0VuL zI=jM_ukxy?-5)|!UMMqq%d{1o^78WDxNeHy{NJGpN`@LQZG1sl%b7V=myEZrlhLD! zUENAlu_4~Qh&N~NGscVsX5DyKR>wXwMMxxHB5e&+Rwiv4!vQ7D;vy6kG0Nb%TRh&j zmaJmu=Qlbc5$VH4N)O{w%mHnaT{?N8dV}mfq_d{_MdX>gbaFy8Jadv{uG}iQlPhHG z*c_>U>P30&jBE7@-gz>&qfRCjACgngm@X^wi{&#j@?`7?{ScRa=qzhAib$G_KlKdB znKs)Q9wT%SZA3C;I&H#)TK4Q!`jYF23Gr5CrKhiAT;Cw%rX3ysPu>FNCkNLp+;r{ab>6>d0 z$?ZEf>gvO((ya9<)T?&qu6$U>+_PoesF8A|p0Ub(a+kDfr_ZfaUcB^}wBPx%q`&{| zdS)_9$_sRsKzBi;>7^1K4Runvzg|XWZk5f=1(NgI8+6OnI%#;@2c=e5K*mg*Cet3; zAX{{zuiI(w)-qN3(7vnva_#FPw|-t^))J8^H|g!0s#EB8OkAS%>nr#6=RZc~7f7*QmB~yCyy-2!_eBKo^z#^`qZeO4`GUzlo_KK>y!qhv!@;s5 zJsUi}AC3f@ia6iUVD*QOebGKCn^{R#3~|J({&0OC}ORH-H>aUh9jHR{g^G z;q5zRWNW@W|FV9(v%N#gCXSH&N53U)lk{u1&#P=G7%eaEZjoP?UM>%-Vwas)Bp2NG zOWAzNHL_yN1bKUUlS~`cJ&Z8_nx;lFZo=eBDOH)o7t?WFs?cQWRxjU?LUMYhN+J}N z9s`71sjQ9CE+RRMjRL;-++%=*#z!PB4X-#R`|9fSe8IYPvVHq@IqR&m^O-{@`N8A3q+40iLgudTaOYmbHhqPwFIRt&V^B8ZOzhM*G*< za$ukK_f0xpzP?S`E*>MTFD;XiPv0sXx*4HOcZ=+)u9oaHodnSBTY5*ewCZk!llC8! zhMWa*XZm=#?$i^dxnYm&TenufCo)y$r{~Bx-5I%b+a}pwTp=0l&9bji`*Ge#{YH?M z`PlCzt#qEG>9yMXwrrN|dZsHww_0O@xNYFthZAwghF!aMIh`k&FNK@##Lf}kHxtr{ zI`Xp;htJNNc;8>JTHmh zqQ}omL&X4LwTBt*pflJ`_D(~LHcTD6-M0o?bqkb!JxC>x3OxM^u{KjjSG4*Bar@6! zj^#Dq@!J@|H#+q2DcAaGFYRHDLRxwcLZ2r+QO>h6+8Z0BNe`fA>Wpi?9*)DU*XudY z(Zg}Y`r$EWRB0UW&IRyyM@Ij*a)NYNeDs`=Q+zz};xKr8a~5aM{rLuuZ+I-Ozr69L z8(xIp(@#Gw&ph*teBc8gaI16RdiddoogiZ+hk+LZ4xp-HtO#PcvC2iqpZgQei z4Ti%VxpGr=rrcXPQ$`h)%l_I;QqZ_vD!cZ}q{D~gf=ZnU)w};4PABNP3uB;S05s<% zS(Kxpw?%8aG=5@CSh!nw_GUREfWmbwyWTn+%R~+#+8%rCF*)Kctb z$T(Z2N0B8DxRNnID2x1MWo2u<&$C6ClfiOmQ8DzET+0sUh*rYdTwW&@~y-v#> zEmEVm;;GKlqAQlN>IErp8YyiPrf6Q(M-p)KE0kc{QIM#D&U-Us0iGv{hqgd_!RVkJ zCG!4)Y+zE^+7x)>l_k(b`=?Dp6FW(gtsnh^dQ!h=1_< zAXp#7p@itL-)(^C&M48C5Lo62FIlprca{VtfpbuoUw*j*x%uXs-E1c9nRX2xoEbLmA<7>xgXIRZtW?#B zZJ%xxRQXi@;&Q3os29?W-zazIy+wXDP7i;dJJ%JZy;JIRi)407wT#t6<6BCLWkKU! zdCv&>@EMf4+%Y#_( z9+)ZL2M^N0@YGsadA8gga$RM5bkvMxVjF&cPhnxbZ zJsXa`e70|c3m@ey)|uZ(IKjbKo(zt?y5d2v7_YRvsG-98Ly|s1w=L>Xk5T#^knEZU zX}o>6WL;e=IV)}v`S+#Lxom}`J$Q>`6^xRyB3;tZC5raOcFEeQTc}!VWnJ!NigJWB z>DAl!PrgQWzS=3XFH|FDY?bImSu#>DjmXPT8J&?Yd3C{rjAcxRJkoMci+_G-ZplGX&d=L=L z!R&yb7LhV?Q5V#)3e%y&iRI2*-F{^TNSu7>-b<=0P-=*<-Kt&O*dX? z86CRaO^vOdV!I0u{|Ac$v`c1R7@UJMFD&Jv7+{E8dg-OE-7=V%Luj%!0t3WX63r07 zHw7Z-lV)(1W_VNO8~&g_P~{n3$}@ggT)cI1#)oF=Q2sXScQo$4^zUR#$e2zvht*|ZKup_*ddG8@07Pa{URlovErQJ+_`h*?z``HiXSjIn~mnyD-Ni$ZHtwUi4!Nfb_S2k*(8TeDZpvcu3NU zN^~_MLk?@b4sY6}>mm)3b?{j|_wik6pE^elTy>6&`NH=UZJLZJRFyEhQ}*uAoeFhZ z)u_-L?DSfD4vp^39Va_S=?!H2GG)WsjWTO$iRgDnB}2zuH8u1jhblKvbpAhk-vMCP zQJww0_qO-mv@7jOtKKEcvfR7n-Z0n(V@wI5LkLMAke`GQ5);4y@)JnNZ--)HjExJ% z6&FdiC0nvptloCDTJ1`!?R{VQzwf?1($(|Z-M3AW?H%du+&g#X%$alN&YUxI<{W*? zhia%*!PFJEz_Tfds~r2_JyJ`nSucSS3_|B!4a}GyUWg>V!wI6V=rmGVSzUsocRgA- z?)PV~hx`7loW<6@vFKAa|Be(Vcl(T1gZ1;_^#7A4fLPa0V)K zI;3k-PvDHnV<2M^L@)b*q=%}GszK8|upX;Om9t*$fUuCm`+Lba&~Oy`XciZZ_1sNh z91!O+ zG$=dujA=tAfW<-q^-T>{V)E>pwQ_gdafc(ENJZ<3C!WCa<;#yTgUVwArsY>`yChnH zkvU==q*tF;9jZ(3T>W49^bKg$N0ZBVj1$67bxZGD*V0E9BQ#$1S=*$mtZS)Tj;5|4 z)l7Vcs&S;P5C1#qEF4ZN!9vu+`|#ILbJIBpl$4>ndJiUTdVyTUd|bNs5fnW91X|C! z0#iPE1#;%h#J=V>#C37pAcG>4;=}uTkH1-Ko9WU^1!gpDI}Ul^E7d6-YYD*WPm%p| z;F808qI%T-wYz8SI+<=+=3}J3@sX{jpOl+xiN6V*SQaO(7TfgmVOqD z?uuoSUI)=FM8ntSOsgj{wzARU$h^9@v}pMUON%TUD;!O&rtH|jGZtraVK13;YgR8! zws$llzG52n@-wi1`(gBW+ToqXvv~e}h`;P+)bngi4`qeotFL!Z%gd$~L^6G?bWs4| zxvYbji@j5(W5(VPo=GZ2E6;uK;1xFQI*dyT`PENaiN3}xSu5$y4VHPSY^ z%ek+k(PF4HcX?gA?}sWo(DRcbfVe+;_qC9+`^d&vx*=}&HdxrbJwM&!KF-p+v;#86 zK`X^Nn5W4gl7`2s>i@tUt6RJdaL2^UH}B=RpaC4cQi)>j5ex*!>WKA115VEYZV49x z)45tT%+(LWQ9B?$n(=jZLU^v^hz303KZzE;>a_|q(k0ng+cA-END5egbgynp=nw-V zbqD=qK~DYG@(QTFZQISWbTRnKFUDVo>NE~WN7fiRm5~4i8{UGUS~bw1T$Jl!ly{KP(x4du1-UsXE6jYs6QGN^4z*LzbN)xOBK zg&~FW+O@%ojP=~oEU!Dwo~bSgv8u~0fY@`3t*m0h)g3F{9ju%?R^L|{JELds&<#iV zEhycB!HK-Lbay%BF_>zb)RpwTm;hpB4Awxg<+E7%-1S%;k>BS)@Wwh_@6K>gtZ&eb zpXyV(#RENaGW7p(w>6e@cb>7lj>I)q*oMDMR&P37he=eWj!#O&ytYmxw$vag{~Rpq z%fNw0_oJtJ2NxcD;hA|6LOd^d@W1~7IUm0b>C-Bm>eAV7&M8dAO74!*6Cs}HosF`} zV$7w%iig_^QRJga5P64=y)E!B=lPC;DafVjv&JKSOz^PIMcnB~KEgRZV-baoatbLG zz%xHV+`1|7wAKR`-OOiM>R5VNu{QG^mCxSEI|}Jjfb2;TKr|p|^-hzhCTe>*7Hc(A zOh|EY@i7vpj5@Uq;JVjfs+B7-9vWP&{INo(`Gr2UvF3|KR~wWy7!MjWc*pbnhmI8xCxsX{Df$GLuXXg#=pNa>57(B0m|~zUHGy;%cqbMK+TuLKY9C zQ!OMQ4~yg1k}%E_e}|-izE1_I2pZ!h#^wI|@5kkrU+%05YZpQ1L4Na_-#BU#V$@Vl z43uK))B4`^Q_wh|e1c_+kn@0lq`$2(Bib{|Ggh(c(zvAfWYlLWX``*He$6Tz{p#Il zxbgEOg%A9K&@zM0KQAmt{Ez+-HT6BXuw@^g(f3BzL8^4uqu|2pXzQP<7^#a8KYbqJ zWwk1b!Hbd7R&7%s8ExBQtzYjcZC%bL>4&H6S-j}{R3+`)6vv{y) zmgk)w8ztH}557fBfJ-t{k(tbSu&0@-mJKv>pNDDfNqE_tj6hK)LeIU0gioH27Fz#m z{rRsj`0BrDZ(Y z^#saipz@)Af#+ZqRjTG7NVC1^JsrsCqySPhJxtn+OiC=<;d%2pVBQV2#mK zfAka>f@2+B1;&89_~MI@N7Cn>dkz~nZgeJXO*R@Fs;a6S{Y{z4d-TyqoeH(8|Mb&O z@FD_vQmgXy1sJ3a-UZ7kvYPWLmG`VvrmHYLyd7VZTXZLDR8l z+i>x)J~C2T##-O0Xq)OBO+SgXZO#Wa@#nOmKohkXc2J#qD;;_+D5-!yp6BQIFF-#% z{+j(rsH?%NX({OOap9(tiwd-cRsYPBNL*Hq+=bk!P%}CAW8NGl9i@O_fZg(xJ+Bv>@sE##b0kZ24BW?8)4hT>EQ514p| zFuC~RiyfWPhaP$ed-m*coCWD%nN~-o%dN0w%NER?JJ&gIrWGX#GScmgWTvcM#TKHa zSvkvRv6UGrHki@!87wxRj1P10A|MHbzvE{>!c0*l9tO`~*uBNduSrJCog5H~M!L42 z@}JyTf+-+|tr+9C^Bj5~L3L&KAgBj?r+;y3jN@Xe*?(boWJ!Dmho zjQ)zF;WSh`+;~`di-#)fer|BAtYS-Z=Zki2`7CzBh?YN=G=rhM?(4|1s&h2KaF&Bj z3#w06YGj$lg8Fl0n)12p*R=^Wy0&NTYr%8hyRWUTv54JuTb&l$v$3cr5{4W0u-n%E zE?d^Th>oqmlec{Xzs#MFdk$2g#Crhi>%CY%_iQX}OUH`fYV`U_@Pmc7;q~$b_(1K` z$d78D10ke#OUr*^lHXXwJi zROw1M(1?zCCDcNoHM|$^L-4X2xq#V=0)Gc`n>ff6bGJi5e!6I#E0A0PZ+_Wv-(TUU z6Mx^s6d2bX5Ybfvhz_hNh=GvsAeAcx>8dJ7ElWXON|j&d8O)i9F&(maF=x&kC!f@{ zq`#{W$v|0t#qMk6wZT}if}Yzpwfk8t*9Kd8?b>21&l+h&B3zeGX6kt9=|A%zCfV{U zcH<%33`S&Zcq^7Vnk)@Tl|ZX{wi6PqjCKt4vVGFiWPL(Iu0PUq^RSSbElXeg5ekzN z&`8?)rO(L_2`83&l~Y@+Z7QSS)TxVhsz>>>^TBkg)adE4XS}v@ZWxy4)UP~Zg1Cj_ zNl8c3Prri$Z*77OrOz+qN-f)#ASWmNu)O`kR4lkh<`E3T!~K3D^9^-xa16!??IC>z zE3dM89u3A|<)S?k4AmWL{Z6H${3AT{joNo=_YykvQ@Zd`tUj!tu9a^j!Ri@F+2NKs znRN{pw-fS+FOs^~@2zn-g$)58S+C`p0|VT@2R5fySDF(g%YznrH6%+}A_F^l>uZ zuzxqcw(1GYFRj3t<(25+8OcO`r*g+}|jME?-Fp zS`-%k?a!&AQO&C8JAwj7n>Pz>bRJrnMJC~upQ2SkzjL6d$)pW-fGij?jizfOS7Z`v<>+_DWj5l1Fqk>NFTf zhC}C}-h=!3c)@koaVoCIx?SV817gI{AS??Y3Sv0a0BeMvX}efVhk|0;8L@lk0sp9s z7!`RO-M@c7WJyQCUtK_Y zo-5y>Lx-F;>RxTsebm6`QlaRT$Jg#2n)Yr+kdBJ{X~oVtiYBIKCuAZkF&>B8yRdER zRyr0cj7qnt@Lh>wvbEzSAyEg1?cItNLB5x~<|YnKToK{+ZF4)%{t#y`-0C78cI~otPjD4!69O~-HEd}| z-{3I&?$DS{# zHczFcnZ9n$`&?g$^N_#c9NL*m%ad7?(Du|b$o=YHAazOw@are&KyW+Nscr?R34w+K zTpZ5Af+fW$>E^kwH}}Jv+ld?E_T%i%`%#_pw^XH~)5RC6;H4V$_MQ@y(ORBo!)u6p zGaF~;a3`jSJ1)hgj+IK!ufGFa^$!TnSpq-jZf7U;R1>(r{*mwnv7jEy`-=t6>6T}# z={4>Gh=c)KnHPiMW_pyaG%-~QV(^S%)xDnEz@U7xpri97Qs+{-7Dx~40ag!sXNv~IE~!G36=9~?4y6gXDg?pa8fCf zIUr~9eX=kGOYRQ!(_|zcj@qrhBOz1$M163ia6}Wq&<@~G?>Gu+ox0V()Zb2^NwmxZ zqzQj_UyJr#SW&f80=7b|@88N!#HXGDdVk*j-Pgbie}u1q7Qym%`wrEiZ0lNt=6)0n7u|@w$A6A^3Lr?Ij7yglqWj=9B)(Nd6|6(JBXAIDoLW6S z+}$8Qb2u#>bMJc=3wo!)zo>)<(RU)%Q;Kx&PVThK$L{(D6mH*uto<+I$Q|G02Frd5 zB(y+RbC3!QZ(kH1h(4BX?@af4XU5|V_MWV4WGqB&eMFr~G%R&#Qd{@i5A3+Doz2HSPk44xhe2 zPo+AqASO)2l-WvsY*M2Vll#USZ#YhZrb3$%*GXCcpYO_;M(Vf<4Dn&QptK(L+M|F@gb50~3n*Yrrg(%Y zQ>HlcuXqhHR#F8w0Ypu;yt>wWpnJuda}~tAO3hqjs>Qt6<_)^j(qcb^wA-+=RjAIc%{Rb z)LZN5UTYmNv8x%q8R>{SvIoKQKZKmKE`)F0GgQf<3GM~gS@GV-b(CFnv9v<&mcJE0%Xsg zhtQtw2&CsAkL?S{^gr9{3^x6wGTIpJ-1kO9OA~WmNG%QV7{=@yT?o_(k@rfpG{vH~ z_()yro$Ea)SY3AQB=V1ywR^=%Q&4{~n8M5b{`A~nTI^19U)ysDp~6Xh%H2NYRZzZY z!EjV8o>IJv`j%iRO+O0;!^&8?#g?wvz6ZflUl)GX&kbjb?Yl5IgT><}fEY4jm}GKu z_3G6Q!dG2&m4lCj4GAEZTyhCsdg&!R{`ljLLo&TQbp$m|ht7?umvpJatP)&y?b_vh zq>C1DbvyiRiEm12kn5waV4^`i%d7K zGy5qXaMUBhxC)2Ca5Z>Vr(V8R04gdfoQ|Qg>TDX2tQC>nGZ04!KY^<{RI`F7gucWy z9tz{f%asU!hz_}ugebKyaubl`6FWx2#bBt^aA6o-W9@5(S>$(P=@aqk zNddiQy|en}E}Zj|4Y;$t8uxnjdmlC@EWw;2G9Xk{yzbC;B+ag4-Zpg9@gQl+IXJgv z9eOYNGvM{9n0lDwa~!SJbhXo0)3enRSWQr1WE5~O{D}|HFY3bC0N}Mkt?x}` z#H0;}`HBQ-0*v737i;L#+{s;#R5JI8)DCFbwbSuRp7!$Z$S+k>>j2oBJQ+`fHetzu zC#ecu#3{H0-ks}_)P55e>!`Vr&9X9!KY0f5ET?CBKTnABkzt`@-|k(=Sh@nKXPl3^ zIZKhZhdU&k=l@VJ3DdSU@|+17)KDC#%RpLC-7~%qf689UVTQa=E_Y75jv%hD%USFU z(Ijs?&%?$KuqYgBn{5G2^d1e(_3i{~Gz)`E&udETcM)eioFbL**>>{J3oUFEmnONBCp-| z?mU)mu{|3}EEth+3?sA5WU-UCzY=#j_w_(Hi1`#vtP|+o#{tXJ$yJ4xP6X-2G*DB| zfrF+q+dAm2m^&UFt>`&g$5jeC4XUP3mQJo%#L@c2*3IbZOGc=T`kBwXfz)2AJ+b1N z-FwkoM?Wl7-4Nr;hDAq*J3EOh=21tKOiDkG?{0(p+5sRqK{7T;8St<(A4`jXuW}ZS z!{|>_`?-SR>FY*MXB+O%OvWdex$8fFh}`)LIUo_POz|=p9E+Vq*B_xwQl1d{HCLQ8 ztJ}(3d_sX3ZVZmHy02p>dMt~(ocsD%hCwcrU%_3@eLYCgiAV213as8*$J3|nc%Hj} z>$=ZC1HXx9GfMGns0f#Fe8{fkAywaggt98pw)i%zuCB)wJHNnf(9a|IXLK6&4SGDE zb|yl!+3%46;+VE3LoyL2DDW<$faXtIIMcNjX~aZojvXlInaVhGBQseUmAA2Wpt6za zZde8{7C2UJm(;Jg7?(rGf?YM$bTaxhJhQ1)QT-OO=&d}Idn>LDwc^UqTF$++JefjG zjmbB_b9f&@wN>NF2wH1X;`*@KhJY^VN2;J@TI9| z07*naROoxe8)ySnDnR##Xb*TD{z6WXNvKFt#B?wEicUtW!QRtg00qarAU9Gt_8tr$ z)j3jFhOf0x!c{DN!OB_e#>wy+DNfN~4TjTj`v15JAR>m4Q3EWFCC#13eQj`*rVwjG zBg=(LpWMGa+6ozXJR31G>InWmI`E^d1wCKG{0SOWc#!p1XgHa|jukI~vuyO-Ev$CtEKN|}+%xtxKH1@TXR4Jl|SIexWNaLB!IJV83UyKxLZLUl>id`XE+oExc zbUG<4?O^HFZnXEFi~1WbL%Gk7!$+ucU%#8UaO~-7M~^oRM*?xEAY2Fc;Y_>^Hpl0@*_{xtThv4W!-#}n2_+c?%Z8D-sYEo~(6xCbi^cvX@H>D5$qS3j?x*M?*06u5@l)J0zXlcZU86~$rBVfeJDUp2*=M* zO?@-|-;NG^x}XCIokwsF%25>Pz>+`%HuPj-;ffsORCQuY$+gI(S94E$5U-Zc!p-$d zk^kDi(a%mXJY`E5bC>at_W?I>e^CO+M3|t!X+Qy6eT{`5u*DYTH32=c3Byf~EPpEQ zrM}!p?SoG4D>OIva<{V-S)7wYoZ~~>aM*X@?I^EWk5rn&Jwi)&Nalhg=fglt4dP}| zbzM6hG(1C-k(`gYSeKa_ZU&wC;?KXyfh0UPix2Q}j@4XBt^-~QIq9sf+)IImG(A#~ ziPE=f(An0B*XAui@r74o*2OFE=T%+!^)9M+KK~#x<@ZY(3p8XAul3OPKp3b`y$6~; ztR|im0c2Phy~{FlUBC$zk45uL$53ZMP6$%9s;Y-~VG%u?&ZVD><2wvK@%yP>#O3Bt z^@%E@aTHbx^AK0N9l;i8xOcoF3RNi_4ck3*nS9c z=S=6SsI|`G6R~KgNUX z_hQX?UqXF;J{Fa7DB*nW%*e;5bx*N3>i||5S5T9Pvgnfs*&~e^n!25oR&^6zsPk{O8C@M$j zb5y~SgHJg!)8M6rfQc|cfs;%Dw}7#I`*z22rglOUY)3?S?7Oi1(h^v|e!XM9U%t9b z2sz2-j3~Vo;^|(Eg+7+3^Mv=AMh~2?eN2_8-Fx;pK3J4)!GEBRG78nNrf_vc z!8K@w69)L}oXy}ft+>QsVJi|!uwNA{$lae{{hLV0=HSVR$eWbnSZ64?`UVtnfV^hM zHXeXu`}+AMsapv}F=?&b?f9=P9r)7jE?k`E!~OG9@z+Z;kjHImttKlBBukx&zBJZ# zjy1e6#RmRsOo2Oxn^~~*XleFrB92ahRhw&Z`Q5MLxs3;%3`g6$@!eGoRH^VGZR0CQ ze`_6XJMt#g4Y%TVpST=bXm+#p_0{O6UibUk=i;JO%}9&$;IUsY{OM3QF`QPNYsy;1~r~ty+bB`}R3jFl`*uj)(~%O4GHCIS)Vl zuyfFO+O%nS^wCEhUtP*$&yN8f;YB2z$ot`f9jd%N_i}D)>gvQVH?Bh!H#d+@C&#?# z`~5@IfLMW|-bN%y^~yIH1)fHf_O9c$MJW<5xRQbh`2)?3GZu59;t?;fLBuFaZ^!{mvPcd)jrkgo0C;xMq-{<_Q9wPL%D4a)fBGCWQP zTn1|4SmAb@qa3KFcTl>$IHBw&kfpn0%VYW6&tkofB+WqOtS-e_@nP0B%n*mUT`On# z>^@p-c@!(iii4lK9Etr7Yl3y{)V1V*!eHffEyi7|4ou|@&r#@#13;*!i9b5d5re@G z(7}#>1Xp1)&wxnPOMXWbIJ7&le|t6a=E9qo?cBTTigxX!IXMYj!WbDF2M1nSev6e& z*B0FAL9XP6*bZ$ChdS9L2~KQ}PbP(F8@yo3ROJ54e`3o2Jb~1`{D;I;vACjFhK%ap9yydW~h9DIF5>F?^=VeeohvD#;h zft9y5iwSTu0!oievoaI$+bN*_tj}omTipHNisIk=_APW!t?-3Shp@4>51(w?jhr>V z0yeM4O_oxmt^247v|uaiX^1+3|UC3k*0#d zZ-)axc%eW$juT;m0;hliHa5r+njEy9efHUo<6n&z3R|{p!Eb)^8~ou9e{h_)Ng&Zr zeN<`Vl6z{;$u902E}n_H14g{NbR72!$Tdf?r7m|h8L4*;Np8b zH#tI49@UvINsU8wa}URmy?Cy^70-#K;7&{J5%LmWSpd&74?9&l6La`#(7%S0GU<9K zS`7xwU^(q0EQ4w3gO!bZW@RE_Mdpk3-r$UN`(mv>GGDaz8E%nzhkNhz$Ke(_UGs}j zMi|*D)=xoc_EIfYx}BxawWY@zTRn>HTJ=SiH?;^k{o^tuG7`f$$v23V6CQ)%YB1gT zm8PIH?QS@2p<^KQ@_;rO2?eJdpDXy7KaNa<_t;7Y%d=k4Y@XCBLe%91pC^!^muX}R zG6N?#W@}sKeii&X==|`Oa>VxwATssq#!vVtD9!M9Fc#h#`?+J`5o}&M102(p&kf@Q znpitXLz&ri?1($O5g|x$AkJQH4|9D6{_`&a*Im%{*h9RNSwx<->ki;w-)P5~YuYh| zI;u~dm5u4iUgRhFP{^M$>75+IRfQO=j~FiQ=doT3j><=+3G0akjvE(UpFr0bU<^pt zAE80nV{1EA=PYd`^}c7ykZ$cXv%RT--)~MLS3?8% zpBH!I8&6l^jJRgpy8Rx+bvGknZw>ZbdN*E}^%2~fI0yG^xR+j%pF`=HOYyaZ8JLx$ z@04E&ze#`P;q%2f--UB;>EWjneb%Hp4K{$_$j|hULDFmS4WU6dK#x@IN~@a`z$^9liX9JwY6xwW+=-+zg>+!lIU4 zbW+%hdRoG4M+zrw;zo5Teck$$ow#HQLQA>3Nprd(&c{xD90(MjfQNRJ^i#zg&*Sn7 zk=Rg$vIH(_a-7&jt&2_`h`)Zu9L&xy!r!f1g@&d>IFGxqO;q95d9=#Jata?5l&CUx z@ba6vtNL}|mtUtcchCSN zhDVN7LTRX(is=fNDPSxE)h9tMC2YqCy+q!w?gScsc5QE>{M3$V4doq1z zguRe*XMh$h=z|cUmIjVbm1u77bl}C&QK7I?XOlO22_3>QmAedmlxMuH+%b7&FR*+X zbv`-3#IbE^D1-1)UdO@|?`22ap?p96)2P~|-QPZ*LFwi@3ekZPM>nUe{)n!kiB_sq zzZ^ViB2l|#d8_9r)k|cCm*-bPR5$JIWdnRcw32=egqg@QCN(Vo(nqgD`M17@<_lM1 zm#+vny?hi$yZre1bT5{b#-S_Nh4tM*T#@5qU$g$L?`OC{Wz{}?8*jYv2G*@xhnsG? z30j3zkmFH#XjPwIwrm;7%ggcFYp*%ZHTBMC&YbBuA=R}p3MyxCbv+R{ulEG7s;(2C zKdTt8RW;+&|GkL^OtWxL`az_1?MG|IJanH^j>o4h#uujM;G!cuvqhmiaM9K z2J4f0rZRGjta1lvtVR5Qt_|N5P$Jc9ZuSEK%f?i>)4_|?MIiFV?U zL>sgL9$E=dSn!kCm*>(}!b=scXf4aL>j08g)gy58GH#x58O(f1G@LjB zNIwp|MKT9DlAbh4Kx+rh`w0QI#e&+UfnD!L??e;EvG+$feNByy3EqH*gC*hg@%#!> z;?U0beT2r()A{^xuosV2R^UUI7GUpBe~D|}KOdhjO2Q}0xFJnpX+PW9M5`TjK?iv#X08u{?Z3JgDF!XJxJ%SM}2w&l;iCP1o z+8o)hqm_$H(|1u&P~Zq2>e~`PG}((0xbx0C9fo1enl;F$WfZMye&iz`aVC{t{pwfF z%B9-rt~(YS+NHlV_b9z3~i*wM_xeHZ^oB$GNoRgpb zBVYajHfE+{^^vD=@9Z-$jlzB(*CBj&d@1Z)kl&Y2ALD5dP6aK;yV>u?HjOvU27{lT z&htwgw-husXlznEZ{9pdz!1ZxxnFa;(loxUUAxv%!IB^%264xZ9nPYMjpJH;tf;7P z#&*p!8uzqIA~>3V26~4ouew#2enmw^&SKd}#%00MxUM=?e{pf~faALakR;}9=vGQklV7dCIgPqNR$wmDbh%nZ&EG_hNoQi1>V&cLEBE=@$7@(yUqfvVFD|WGPt~tEnAWfs*YEos9x1vIEuK6w0(3;&+lIn|LU`yMo!qvw zyW+-SpliJoy(`TvTA&;*?`ClI-qb$DYTroV*!vS+qLuI(Nk1^W6svq;A%!#wSn3l4 z^+R_V_qA|Rf717&erh{^?mV$xTl{Z#Ec#*;+>qCByNcZ~p|PNee+7WBe2xM<)R zDCotCm=YvL8dOJSW~ReL4F)5zfzp3p2MY6ZX;MD{!pe9FeMI^YG>F9-cF+o1LN6e`^bak~NJ~Nxb7ww-P;dXNY+u7cL?A%=B z6cnR~riYX34*|uO;+2aR;)bt%6ffNI5uAU{g`{^TnkctqbLS(O!Wy- z^4p-~?ptnrV!hT!thS7%uW4`-U8KuA(xCpT9gyMlwc!qamR}5mgcu1OmL@?&6TX5? z+WVZXZU^jEtT42EL&^+3;@LgVmfu)4>gx)a*UZgecXRYL{tBrl$7 z_F&$me0=}rZd|pUyOrDGIVJ(mm(RoI_8piV*NRl`mf+Rr5r4*wKuP~N$uV=FU^KS2 z4wP@S=?35G6Ff^BE$o5HYpyt%^__~kjD{8uM2mM)x^_M8zyE$`PTxbrC{w0P!GZ+~ z@QYvk!eQ2=cBOGd2hNQF(@OI5&p+>sCAvQAth1aZs}Cq>Y_ai0v6waqHnGx8 zuh9f+-)P|o2faUgcZx?7-0d3;Zfc+K8jesf5Z(qy-;S=wB8=|sjRgi{3_}Bt1_UuK z#yl#v0mogP()GbBSULCefv%NTLG|d`?j!3lW`Ws`t?~!^Rh1NL3^80|!NZNCj~`$s z8U0=Y?_+286PDJqxv^^6P|l8oQ#I3Nem0JsrmwXT-P<-GwR{1Zk`mEdRDqVA2a&;@ zncimZ;_%>hZ(6FeyAsl>gvwGF>EY^oh*G(qP}^VM3jcJf@TBq3D$V#h{zkZCBC6

T4!P4P(S?q8I83Cj4pq&e;qa8<-B^@Fie-ENZ*!$%SkT!8;tTM4!|zh` zvGY1Y(}HlYpZbvd9ax@NVznohcN`6t)#;{*rA5mhEzR;XyYHzzWf#(t)s6Xt{~bs=dRcKgkq_*h<ozw_@wot&ZSPUS95Gd*A!s2bt#8Vxxo-cewG!8=YMb?Lg~V>R6FD z+r8SQYrQAy^KMvnZTKiwUIlk-^MvIeDps8;BN)O#!EjP+LG@`tP`j3GRHM4-a4ibD z+VCIU379l12P-S`@$DztaD>hV(>N})rcZ^Zk_XS9%16)k26(FvBW_B81N#`v{)^+a z9H+On&)Q}(}GM7tY6PS$U{FP=yluBf5~jxqeH&V1ayv9NCEt zXWWJ5eN~v=OtmmNM6K%$lE2{iLoJ9P$1!I%P`ml95II9FYAl=-B54&h^AL&A+BZaz z0UsMI!_WQR1}qfify%3_wNLYm(S5iu1xx+O{r;jJ0o=!k-Jc^3D%S_m<#1IsUQ5Qvb0$ZB5ANYC+nj zEy!N~M_f>M5J_9o$kS6j>Xt+Bt-T*;c$s>Y6iHa24wj~{j7-(aL|)~x`S5k`Mb~Ef z{U{&NO`>iN;27)+Hv~UJ~2?>{WwdeH$Jz%8_QET@CDc7(s@gf_o=_c!E3HZ)(4+M(WO_? z1aFY4WZX%R6hr{={Vvbl^;vb%;#l)VmQ}gYloh;@mXC%@WZHP#8@`&%EojgZGp1mZ ztyWWPU{ku7ZFd=^iNP1MsGk^r1u-0gp?XwagSCVm)nU(^jAQu6>JlCby4Q}4e#TtJ zG&T$aC&5$OwC&wYRm=Q>e5~HpfJb&v&nhnssa!}n%!&LLo)VM=ni1dAf{fY*e0tpr z$UZL@XBJi!P+R5Khbe8W-MCySktT?wN13syHtIH zZSThld!T$)rv%vS>}+Qx&)xS1DmT(}qn-Mt+NO4nS5Vv3SJXDu75QC_MZHEpTLIPm z$q3dqsiZ7jy7X9;dUw(wSh9qwKdxZ;6>I0jg6K04&XMVn_XfiqN9M76i`{u+xsEIs zd2cW}sUDrt%6X2?LbE$tkzJCHIpw+3F4%_r6si>=6_xCB|CdgM>$yu*_Wxc+_l+Bn zcGt?N?H}wd8V(l6r=}z8s$0);Y|wG%8}n;% z$o#q#N#keKfrF<$RQc^FJ-8EjFWrsc^ zn2?&mgWo*}HuK$ z%)_G!d-v{j&P`QTR$}$))ehsNoeK#VrjjLLLP4va`Y8yOV2T-u3{v}2nP|@<^N#jj zb!h-rY`BdU=3wR1(=)KUvl*{6b>e4lc3|tlE=*rh=6qWP?9Ur{_US`AUPt-%4anq< zNb;GLI43Uwb0=3|F&&3_S}AbT30b_<9^8({2B=i4#S;8ZMi32c)&AtrGTl*~wf;4p zZB*MNu%3)|sU7y-B8B+zY~#s*t#3_Zij5^|pWdH4P4~tWMdsK2aNf7z8myDizTwp2 zrvMU9gDb6>X_&@Ehm^P=7M4++cna03k5J1Yr4*giJkUJ8@7l{T<@xRC`uTH6x$zu$ zvuKS<^Qh+S;RcSqBqCDhhXuYMhmzDL?!?mBU@y;ydbz;4pu7?{Q$ykMSAI(^jmhZY z4s1_X9wu`g>gHD-LSj32j1RoZl_nlOZ|7oCI(Kc~W#M%0JEs$=-xC7J=|J#+KKRjw z{_#R<|C#ms&_3%1w0`27So``LSg>>vlIm#H>&&~6_S=sm<$V`&jFGwvw{|J8wC~)C zv}rTY^z>ucd*kg`coWaQeB|%(YI+HLFF%KK+V-R4V_!#IQwye-PjQ(0&dv_5M#tm% z7hb?rZf(z*I~RRCfZjagt4MqJ&yl!z7QBh3!c&} z)-$c5yVIP9WBf;WiTSY|99z{^y0v*Ms|@lx_K}&Wr_T3Zu6hn{=B@+|aRoGk1|@hl z%h%hAME@>)eRCDA`omK=x_%?_e*T~EUVAB$p8O*UsF&?|^%aDs%!4NZ9XD%kk*v zqSsJC^+pO-(b5dgV9zYiJ2CcBwKms{vrwYT~dgY`>NUB3a}%W z=O@_*Iw=7A&gGo`%k2m@(NXbe1dx8AB`!6^(WDMgt17L83!YqMZ1N}JrR|Nl{OYri zI%_fFzJDKD-+vA;cM+1k+kjn#C~H4}S(~@>V0u16y`AvzyPpW}T?(910mMD<$1Z4l=`u3M^A@lADTnLHoCgk*U65sMw~mrr;YB+OhpnOQ{V z9h_&ZyqgE{5~gtR!9x>zG{ZZW2ligxx*1129mtXjl?E98mpYe~cg|eU!i9o=XEh(w z6l@t!p)r`Jb?5!Pz6x>Qz8qa=d<20T?m~IUi@e+JLf!BF7n?7>9H|#9K=Cbipdh{s z$+FrtOk()C10umgJ0P2m^g@a}dwI4Wb!&0?^bnR7&f`uKSIl`BpSFfcbLw5wj9>}{ zuf*@)jotTr9i{XUf-)Ymml_b;Npkj_37Do)MxXR8+K&9k$8>v-s*D4c0F3y@3r!$ zGu>J#fTYm=RR_n0qPPH(X`IFD@4>}MjY#UI84;QI1kf!(oV;2+H^Jbihc$^)W6E0xTmHH$=p8mrIRV- z&PY*O3QF>FP?eGgy27A8Bd#EGmqFm7AQvzi2R;)KhV#bJ)ANF!E(cJU#pJDGFER3 z=k5?akJF+*60*t2cXIKeW-oj?7s4}^jmb1TK~?H$94j`^%x(P!`f#Wqi}1rI5IDkd zque+VnVb{EZhH^Hf=g~ z=unJ@HAdGKtc;ultzW<1aWZ7j-U+eZl^osC9XSs@_p{Z;U5}&zwErEX4ecIUk#JtP zV3DSDx}l5X%!Sp#K}b6l5W0$1@V=cDLhEdW>B39zGPYeq998` zJKo+Cc~(H@4k#e?GW}4iAIXy@q2zDx!;wGx6mW1Kg7L(keJoLe$V3=d3Pdg#*v^Q3 z*Q(r|=HB@jSH#`{z{4~G{RVa8mBWolPs~6s&ujU6+K@{1!R>T#`d^znaKo!@98-Io zZ!2Z?Y^P;OJnNOK-4FJ6G#{mD>eBue>uZ__6BL-BfGA+&lg1rg+r3>66sxR_RaV9g zW1w=8>24VAYs1Z*-<}7#yDy~#PK8J+%%(4o48+~{9e7qhP8H`Ow54Wa7uBaTxNvxo zzJLPPanW%fEk`~>+x4wlk`t$(f8qAK>#%Dl59#65DgL5Auj07WG$c=+ilix(NI0+^ zmDD&`%>_{;Cm^l59>J&YL&CE2Fmu5|ELgY%_4E^Rz?X{h&O^xQ=3%ttG}Q0di9P*Z>I`wLB(JX zCw3V4;S?A~UYq6}0mPQ^2Pi)*Jjh(jr${$djU4)L7;);d?^D8?!@Y+SBzSn3x39So z{yncEN&ZH-su~`CXt}eo4#iD}u3B z&#r03pYu-q6$OLISeFJ4=psiy$( zSaTQh;*(H`HY^Txqm7;Vy7mCBS<{733crDr4yP9Fw?mQS%g zZ|iEu>lw?iBQXtmshn?<=$-t?9(WIOH=<}6S}M5+!8TN;cu_+OIH6BnkA!=!1IP2=_9J}1A{mnPubUa1NO32esKkZn?Iq$sl zj#0I{<2+j?i8lsy*MK^b6sCRCyl+!>Dtzt+3 z={Jj_Ooz059$7#!as`V?MpRzPL06D=2`( z6ZjDdLBc&jen418*S>yL6`!0Pa(5*_hevI}I83A2-FDg?&o0VA<|Dtuq3dr(-5l=J zq|SC!%)(V0#^F2nU-vt4U%T@Rb**xvDLWMWp`N?zwrh)rDm(G~)KEZuT56CA&+gbn z)rwx8fuhy7-ow}(u3ExfpouI&bjsnJu$kep__-Kw3vn4GHnS~z8@Pf)KpZTw;si|?y9*Y?p?zk#g zs!z^WRbF$h;(>zvRn_q@`Lj98IY7iAH_eBR#7g)RXHiZ0HEL%kp*|g0T;j(JFc3NP_oCj`>U!!ss5Ep15R6r4q? zniiU=0U^k}+<~p7&!G4xN%SZ0inK<%nH8y_k<0$*b@4-uc zq>tm9hvoa383TDjEc(iik9a~^JJfp{Nl=?)by5D6?EQ}fj=ewmpb|`3XdLYQ#Zrgb zC*Dpx@krWdILZHw{AY<~N*hTqSoQT(<6}Qn`;PIQ5df7}MrX`SeZ=Fp-JVnd#E2l| z7A{ z6PaKHS3KN7Are+B_mTDJeQU07r~1`V?SSw)%qpJ<&G1xeB#aYs5(G7h9Lo_Ns$ukU z@bU5k@<5HpyTBMcok6- z83+3ZYSG%5wKyKG7tA1cKT_kpNGB={ogw5gPuArt(Rkyz2>turnEJW@K+wayLxh;$ z&k%4%`gqVj5hGAH3~P@&Ek-aC*-j<}`p2zK?lwNTXDg-cBJN~*uz-sN^>uCB(wmM; zs(22pdM{FMNI~xfe+m5VL&v}CW?HBD<@af%focvHaP?C`leO!63nxzeouI&bivl*! z$YilRr;BkEQzgc+w6t_chE!vx_!u$iDqB)g;xNPFRSb5l6csGZvHGcg!E)db0H-V` zXl|=&ZFXdak_@U;$?O!zg_6V;GVU}{zJ4!)TPXz0O+{^O4qAgPD4ZX_+#=?y?V;sxoJGZnM5H(v-7|~ZIjbiF`Z&6)h=i5U`#~eC6CbFQjVikILPg5=OIl_ zv^M`_v=_}9L~^CXmzDY2o-8#995&5TU*h!FAaakosX*suW?^4 z4pvtqj;pwdwCK^nl@$pfjrE5~=nxv4n^4Dhklfmi2JWnsNyqe14SH#ca38I1)bbr9 zHu}+#&hsnp|0E{eb_Z^};x6o&xg06<0N)wn01`3i>0Ro5Xtz%9TYWX!2d8&ESP&fb z5A|7j4<7U#4**DYYXw3Bfco4>VQ4i(eNErXK!c?4)iXKQlS4DL+1;-PKye~@j47b; zLw&rSAIrl>>##h3CJuAwv*_pn6a;;k(ilMb9vYsQRtw*<_oKeL))CGVQ~Gx?pZFE9 zn!6t#`d47y8DZ==R!xKn3QSNSOaaZ^wpyz(PP~Keb*(XXs9=nl%D7<-RW90d?GpI8 zvZlHPH!^6Cq0`?-Bah72O@4wJ1@~N(jy#&f$2T*E=U!B9bX*zbhA{j^Eoe8#4u#M6fxTB|vaQoDu1PWI>SL}o=L z=Tt6uGEazeUn|wGGA7U9A@mH?#uwwarOPq>{ApM;|3YkEwj4+1Eke;3z7|&PKB2O= z-v}oY0AqPiZ-)?05e&f+jM2amG=rmlpxha}wDhlwaxj)K3&sDBP45OE4x>0*Ls`^A*hFg@~bcpvtB{1IIAz_&2>j;BzNT}}b(#6F^T ztUe>*OU%CBjry+oSFHVn=sCgN*VI>rD?A7nt6P0reRd=;G`XtJ4%*lHaaNtSk}=%& z_Jf#+Mo9rJEXdsE6WccP6nYCPr=(K=p8%0`UoQPRyd*k(oEZ<#tS;wfkUy4Bv9-;eU78#e zmEhxH^kDG}Os{Ht#arld}kUcYF>$YO*+fyC_t}lb)L2li0wFhFv&Zc^UTP zmEqquypE4wb}eTA^!wQH$-h9{{AI|y;#x;F+B7<>Uc*WE!pZP*r(52kVk_f@GgO&@ zo*OL1`YBl2KxKwXSNXAMpBty4>Ky2~wZ)B-a2swvj|Tsw3Lu8C7+b9d>ma!paOt~U zeDTFtwQ3b~*u0!fn4k{u2_) zDo{DO7=foAfakocX_|8(LVtP`9$Lx?U3&)_+B=X*CRxIpaFKAPm2jy#X(u3;40#|m z-q{h6kBuO=T!TRmx>|eDwdiUDbLXJzaJ@5t6dl@yl5G!SL$e=;-`q)0?Oogjp?>T2 zUx#-E9VbzU@RiTv<_w)GC3EXCmAhe7?TY6h+S1m6KkjV9q6dD6#QC$3%E2s|hd+Gu z1})Ncy!Hx`Z@2{6bdcD`-5Ceh6xupxe+nd|bEM1UPD{IE6)VVdEqSz-m6#~0C>(qr z`)kClScb$m)=(I`YQR3CzG3~u%8n$~JJxT=cRUDil~-RI4GgsIb`NP1h9)!;BLt^hP)c*QGlz;d#6{zmsO{mW6{>ccyrrqzG#Xe* zqRQrB8eB=G5TJVrGQ$ng+9csaL1#;S>G>$w^jkC)Q{V|MrIoE98iHvk({m>m*CIJ@ zdB`hC^&!-=AA7P&v8|(%yx(3Vt=)i|DIk^2^^7F!aVz2!O_{>*%SfUwIifJW_cv;(YK1Vw|YhPQ3*#~M-z(W=5-&l{T`U0%? zPR7rBsb?DC3X3m|TNjJT@bEe=84or+jGMiFe5h#}c6Kt4tX?G%C;Bo#t{5T?Pud3N z8JQ1d#o=emDG)jsb32oR=4*f0frTxffxqY?XCM=8yb>_FyU_UKe@4l-e@vezN$4S5 z(Njm?+NtK%%8jCOo&)jIlY4t-51F+9Qk)%aTH2t1BNeYrd2`TQc0Q&({RpxtfH(uC z`V0Go1@#f_G^k#^BU=@-=ds3-eJ$!c9s<~@fT21k!pWzAGZu;V+_89lGu0+}c@=jp zXZ+!Bked8}O#Z_^8~9CNJ!WU_L|m^I3$ObiQgSJ1)Y78NCEoznJxX(&6~Hvk>Jwpt z0`D>k*ccki7-e(3^1A2ep-7_dFc$bj!5R9wm;>V0OWz*dR7rm2b()o$0W4ZUUq8+x z^wOwGyL@JFG2!=IxLwUTbr}VYl;&P|I>hIYm*9FsUxy#*bIR$AbQ%(ARVs;RAXCY6 z&|>$v1TS}4yxF;21fg^D08R4t@IVL`C=ajQ2P|hBILAA5BtM~rv);o8(da9}o|Ig? z$%V@~DVcb@lV?OzO0e>(JCXR^`_RipP)AWre9t>{4mx>#XIx(ENdtOP1rW7LAGBUh zHZ3v9(v<}bC|Ye^wrrUtIr(hBP#GN-)6X5;1H@q0?q^nBX_juWJsU~vhNtqzzNK*x z$kXbBYd~}Ix+<~yBVj0u#vlcc803yNGW#5eQn-sj1|`TH36%+uQ4Uc(gW?^{;PG5a zS2qVFGW#qeCxTu4sJqwmFxodt35Wt3Wd#SG0CNza=Q5hD2p_`hWqvz^E~Z-o$daW? zQAtKWu;U>z@rP(p<03S0wdLo#y!d`)66R2V@Uwk!9B3LEo8V1Nz}egOqc3|7ez?4l zLJ&Q+Q}7HmHWzQCfdBwN07*naRKWA(gKPtLLg=V4;j$~aS`vp3Zf?O@9K}Afh^G1& z`nbC?pND7Ds{J@xJQ+UfzEc9Ag`Xs}?%0N&1{(EQvdp1zH`^V@fi6JL>-9a{W#Xuv z7DvIA_ubs0){aUFdwP&3_~lcix-Eb&@9xImPb;Dk5n2jjea^cg9U})k_j?gucCYK< z1owM&U!N=(`hFDLb-S;Jg8+9xh`+}~-tnb?-j&p^df9h(AB@A5-1VEXWfK}V`%!uB zAtWy3?q^aLE(<tST(amQmHg_Y&+gXij7X*TOyz=BID@b`=K>e~}xf&%XX z3OL^f=NScEYaELf?3v~;3zla0r$^k6Kj)%2Ejaw|7l1qd!x2FGG^g^{#|1E{LW`l_ zRDY2A%caOl2qLACzwJlSa#j{P(+?qI(o!5)z8srt`cOb~yF4C(j5J#K>SrRNQ)&cM z=%=}lJ2lDFqGZ{R^R)oEw9xMy^ zuqU~aXV*)yI&c8-bYyroQHiIvh7;KCiSX{Gz)2TCL{1GD8VDqG*u4oF1Fcr5jPlz6 zV)-2aL;SHaLu4Db$CN+dX#b4xdpb5~I3CcsLLyt*w1A~8&G1%eco5b#*>K7iOlMu; za(u=?OD1r2pP0qn57N=g6$=dtP7rfX>rz~zaFWFQSPe5*SOrD1;rgQg0;czv*34RZFgK8lvnz<0TYo$)9 zVQ)(}KD)06EApDLq%@1o<00E4TOHvm#B&`!?NrfS^zmB{Uh#RvZG4{AuGZ1a+oyoSGT+t09_%63iEJwbudQozPS zoe9;sJX2q`F~gp#9^*5lHPO=2;%H51;nSUGFnm=0WCWd|)Wn-W73KIBf6Z@r4m?|4 zMquUpv7NgYX+%RtCu^v(ENy|*&i~KecL2s!o#}ovqmf3v_hLyda<{?89Wc0&KtkWp zNtT3!%_c7)*-bXt>~6~LrflA(XUP&09uP>V2^~|6DaH*~xyb6hcWFj5nwj@~S2|`8 zl96oLPKb^@y7!)Y`aR|EKcyCUkQGwmz+T!8Z+R8^m(av~u$yZ?9d)UZXz8#cA(H3i zaM`D=E8JJYD5kCB+!>5T1c{$g;H>E+(zN#|oJ2sZHt@aRkZ?+I$dcs|FAc?nC8ImB zkP7ooJkZ&V?EY5JurOMi8i&NUPFtf-|5rUzbLaH;v>P<`aq&g=w~($ zB^oWxWLCP7>CL-8=c~XyEzk);x=gN3M?sv0KHc|q!%u2wFgBA9bd*HELO+#lt`h7b znnnf#o%$%RuFUUgZQ~R7hSW3!X<^jVZ>XKG_HBXM$9bNFj5o1ZDkJezN2;{RH%!^@qH zGXwbUYtQ0uAKzk$GS+s+;!y8QT#-3|`H=)NY)-@scEO)WqZ`+M1+KrnlYqxGdOYVy zrKArH*W7afu01RK1s~)?9hmw%H3HnDc?N0!t)?}^JMX;H@^7GV(Y|4V-iwf7h#5~j z@q}e`BR@YMk3II7#YIq>@$4}tlV&`5Ok0m%dG07UdwbxJFD;5E`KorIyuKRL<{-=y zhB^|{4v|Q=EVd8>y?GoA`W2-Qh;|H_zs(`0<9&Jmy|kb3fj4 z#!7c8<(PJyN*QCdYqH`et8Hh612S2I@RwAMwFI?@_-{Nx>#ot`r(X^~9xUlYEy)VM zO72IxjjRhhIUod5!il`OY&09l1sqB7HF-QI9{{d!iIxp*%F>x2;Q>A>hD$FsGKr#A zRpoFQyajClLrvBZ!@}!)>7;$64C5KB;{ZIZLR3N>pskQ0_PKG-qByL`h`^D=Txz`?vdVF2lN{bCtqALb zQ9Xv^F%>3n1oRzbeNILtw26tkCR4wPZ=f*>JopB?x@{-{897LqkKg?iz|^&?NP&`9sfKGu9f@Q|P=N zdOP)ZY6SkGMnDLKh>tQ}vuoEbTz~!b);^+695`^m`iZ--Zr!@$Mq#wS$YQ4aj48y$ z#aY%jr9-NB+H({_ueuL?wi1l;$D}dH&&cOdvWy)ye9k($TB)GTb940#DBb-Wo^5TR zZtxjGTHDhEB!cO%hB?g?Rw8am(Ymr)k0AFA@* zeH^@sM@SU9=$pSrq%8&Zx2rI>+l7}n*4>S5aB-Z?7=`is1$sJ> zoizhD&nd)XFI6FrIxn4J>|6Y7^d+oBu=6fN(_Yr$_5q4_d+_<9C@i2AFIDzVjYHqe zlsjts-qqOmKA1eA{WEdzO?~1RlW#K0Qcb3u$?o5YdolLg(fViFHg>M@y_tfp#y3u< zoXSEwxQHx3cO0dO^4>#*>U)TM{HB$%7OrS)=<(r3Asv1%!fEBHg&R#HSs87EUexyW z(nkk%JjexTl}J^07lIT34f5XK(T=W?gRq@*AzH~12-X}!Fn2xe-m+W<8-0Y>!sEC( zbfJ;{JnTI#IHj*aZixPA;WsmBI|2GMe-Gl&g%)m#hq(|-NUh96>Ljf zcm&ExMQQEsfUCP5ZX#{pc$*xX*dWf&^0EgmdPxruq4IJZIVKYEV*A-J4arbY8<}P* zOwI@hh|tFM$}6u}2#Bn!$=aH9!DIoiTbwz52Z>fl7tEwLwTA)?ChmiDHM6gW^YONl z3Y1|9io)u#IS`MTF-eFTY`}eqp8)*@Hvh^msd!Efh~_cPc%Dko zScitwJ@zVpWSyb)pN{)8UEk^G!(U681_;`+WeZlWTxt0XlQA0+GWPG^j~j2i(Q;D! z^2;w{!-fr(A2C^Q6!IaC$qO&MU>UD5$cF)-GzYZj2)Qu^M|71ws^OxHHpa>)Vu{i! z7}0S(o3@V3$K%90U5q}4-iu}z~nPok)27SDiXy4cu9baCkG@*LgmmoKC(3@ zeh%%Q`eyn!GM7x{25AT;YW4zhZ^Gc5QHWK?+VMYxTk1&=Nu#H9UwjPDaGRya;-uau zT4@23V$!#f7J4s5O&#R;Op8c`KcmwNqDK!hr7$&SfA`50-|pty||&;HNI+( z##+|3PV}8dwm0-ybHMaX{XN-o%vgnvQRsbWo#VYX{nI?u{aABqtny5oG)BrlS^YEX z=0x*cX~x^$zX}9oyn`{;$2{5i5KwkV35*KRE>hzaLf?&AJ{MVsCxB8q4V>1q6IHYn z@w4a|_+0ubZ0>&r)5*##eC)T}!NL$p(I@v4NAUIX0c={e2;XX`#fuO9FRfsymH-L~ z*nP9|5X!Qq;oqKkNOD;~{xaYJ0xaT`2*{py32@)fhk%o`H zJMpEAc---a-{J{6Kg)A>;g>s~Wjlyea1mR98`964%5g0bgpbwL;FY|^+*KOM{m8et zEX1CM>XgL~Db|&~&EH<+9!RxxGY+M1!ac9;#jPyvDD^-7t7Z=_IoON?doyrL@j=Yk zumE4({t~)(ze-Mw4W)r7)Yp~Z?T7Zm%{&fnx=oawjiRW^r(S=S8}#e&WNt)@Jq!6= zui?tFY8-S#;|C?X5KH7_^{;*~)E=w-EI?O{3jx8)^Je0QD>uCV_ZsT?Y56-D^Eb4N z(EHQU;4um{^$KXX;)*LQ%XdmDr=c>xeakJkSR3WN_uh;7^XFUg-P$-!K64$Tcq@85 zX-<@XBJU=%gwD+Tw&cpuRUycjbuE)7=?M9Gtm_J{9E&8e7G~*CCf4BV> zju)~JJ~Ny7CivH167SNW&Aj**p~?)px5`6K)DL47uRKyh)g z<&0GX3K22#^71U7Vl6E#*0T@|S;~}k&6=7T3sI31V<94o7caIP56cIVo=ur@c&xMr z;gEkaqw^?$bgXZv{L0hXN^TcjRhWg=@8rX_ z<6epr)8|<%#h6%67yV#$_V%Hwvcn=F4sR{vo+9d>kTUrJ)0m7k4%S&ofKwqr;%0>Q zp+Q>fj0;CwaW9$=97gx7Tyks&(9HJa$r2{_x3;cMe8g6R{3Dh4dB!Jkz!r(~sLxhU zp@c01Y4{m+Sn}f>IN02al@ZORi(!G8PiZhs-;j(|8G ziZ>x(kw1oX3uK1Z#s&MXb|?=Mdntp}PE)i;X$fbh-G|$99e8oT1zY?%$hEyl>P9$z zJa-8?xiFh*su5d|heiF=rKsA0TVpfuHwCldO^u>_We}0a4kM}MXRt575Iu%SHFOS`JPkKdf8z?WmTjv(jG&9G){;GN zR8Z_GieA!}uY;Rj&5MtgAT@Iu_P0|%y@4|BJq_q;v*Cs1U&i)8K3ze#;yDi~x`7rf zEa;bPXA$**hZy~$^;Bi|{0GK1MmM%H7VPqr>ouA7HqnLP)! zM4;p2o%pDC53+*W(Gar&u`$uu9+!ZftIor;=YEO)d!Ix2cWxu+Bkg3X9W|f&Vt%70 zn50NEW}$VNxJjzwJ;i98&t!e7%Y56@-8a)OQ;)g6XAu4ZBeW-I9L3Gpy?eK{KS=Cq z*|KGj<6;4%+AH*1v}ln91itdbF`*ITlMaH(`M6O^-Cge>Z8U)A+-N` zHfgPQ{39`|{C*M}!yC{}^T0_p9Y|QT0O2VV%cj(XyuE{Ger`7P*ScGg`-d-M;gT76 zjv^KhN7He!vxt^Gy_ims@G@HEY_~@vGdI-|DbpFO`mDNmFMo0xb0HBz2Foy z&kE*Sn5)_%1Y2e3`zf7CYx0`3rk|l@n|obVmez#&XdEG%@eZw9_lh5H_&^~b8bl40 zf*CB`n>gK@f!4hS(_m^78MM$cl+ToJt}0u>h+Boz3X{b?YCQsdI}lNKH|C^Y z46Is5CDb9HuAe~m+XsO)-@X8G%T@s|P=6wS9wV4W!A`*V;m-o!_$8X$KBUW^00SjW z(ZoGDoiM!sbSp$0Mh_mus<|uB)fkVfan_J+L}2WC z_kM0@M=||NKf-@6T|##wkz`%R=Mm<>&Sf3zuzGV|0^Pg#`8(6;Ch9~nelQhHH@MAhll zVaD+Y-JG@Nl}>dHQ0#9kD$U5UC;F^|O5-TYI=Zi`tFw@%ao0C|U<647AZp6JfygD} zWJEibh;LneE4D7%3BnftVpKF0tGGypV*7~j~&4v zl{#$$TxS#5{7=+xeJfGnG?p4yJ)e%Ca#g;eBx?-EJ&q^YAbm6SO>;+cNY7@Con(=t zW>ep^Z-~PuYNqNt-F-9PO?AB=MSj#qk0VlL3kV72B#a~5NhdLT$C@>3pr3*O$0alk zFOjO>{N^_nuqIB3fWN}R!s7^r92#rC5Z6Q;lB=$|%F1u{8#BKZsd=b)eKWSYO z#vfCTNo(?smRw8IZzEC9YeTv#h^H!h@z2M+xOP?ol5N!AEZd908Ry_FIU}r%Lw-gS z-U_Qk5s7ch=v>g_4WORNkhv6rY|)bgTH+Ro_)0l|JwD`zq#6z_A zSn!XVPL%D=1kBz>-$wQx{BqjI_?%2B1{=QC^cG6u^RdetXCZ46^$P8VX_wZn_ImT} zPSn29-kH9Q_j`|4hKVz6(zEVQC75}5s%4orsek&NN-$-aHl1pDD(3@(fN0}2hnRo^ zQA@N*DQNQ_IavC@+OU*Qz>p^LM1t~ZgI68GGW8QCu5$HGL4S!y5Ro82M2?2A5VbLh zaDto2h0O1%tvaAoj>_{67XcSJPoP4!e}4@Y}Pr=@Emc~M7@({7M@A*{cR%JH3) z@BGn@N;)`n;d_@A;-MB7l3#ri=fCzVJa_dsuqxY*=MFn?s6G)#(_*P-y9m=W3pjo( zpMzCErF%O!t)+=+^!Un+irkyc&S>nTiQqRpY1B#a;hQwMTS&c-0rJ7+u+Z)ys^g}U zLi$VT6d*~!N8e4~%zQu18O_YaQwUF+uGNeOA0EmFl~G zS~Ch#UKcQ=pz;;3^0fv<88z>$1e5qjb*UW+>XUR?^c!nD-%~ZYjxs}M_w>Ut@Uv&( zsBA{Wycr1gQuKj+KhDoC`u2x>rrSTr(jlCH*5`i;9I&DyU8^`Qw}PVoAF71AIe*%qr*FN z{&2(*HrR)TndhO?#=Ld47>Fi75$MHUcLH|xyYUnI5lUaT({Ujs4u->6?K1C2UQK#k z^&3wxc@pIHm(B<~C1FtA1<&`lB6!E82xcWA=%8^74-pU&UF`gRfw(kyOSd3;^{psO zNg=JWladNvtSGc2EjtR|Yw@!m0mQq8h&0AUfN7K2H$uPXc-m^xm~zbX$y`mpbv5Og zxRWVwWS)~L%jD5jzf%dLm1W{irCjBG=hQ#dlFo8|4TzbHYKnjl{Y;#I5(P7GI&{o? zlXkq5SY;@vY~8D@k@fJ!klqL7JqO8l4F&`>QF)30?I5+J2Z@vrRUr_~c98X}-yjbu zITRFF=%$4!KNYK`r4|>?`>r-5?|T}zO(S5)G@$sCjSJl0-D6Qu)gCFE5znQ`+5sw6 zvu)H3;m(ACno2m+`LQX!;g^F#+S>-r^@-}rL0~@l!Wf#UQo63`^`V2JVbi1S$l!j_ zOqtYIX}b2FMmJs`h{QDnAb&swxRvu`kl^nH;d>4Sjovm?A&|gBHh~l*wSn!lSxV=_ zid1Z$6yNRQitt^4bqm<&^8``pvElaRshCeoSjjZw8^b>5b?>Ev^#H1Tjo7$wF}_fp z0_Vlw!g*9cFY|>_N#BEMqzEO1IoO6l+DoU`SN^;N&?0#*9;4XRf4bxG*t9mxqBr#R zK94ot`Zi*T`#46TI(s zVLS=dFC!W^p7f`ZMq{b9p>dqhSe|61rf)(JPIoZK^O@+E`l9;u-OfbcPWRY_mT&ed z?T13<6towpujih7?(w8%k2Y6h1xpD`-0{RGxH0veAjx}Pjh08yg5u>A_k}tS9_k2U zW<@*ddFGMQf=Ut&RtK6<+tZHL@HD(x#dsIjbAJSGBWI(3;1IqYz8G7`@h7zcGYPh} zrslv#k>(61&*qY5;kodaZbN!zK8g~<(c4*y*uG8Y>8l3oj+m zO>wF8{7fWr)I(E8BqpPmKvzma{P7}$I@sw%58iqd6fHz1fs>wE%CCO-lStiPhhSGM z60`C!5G2q~Bqp1E_a;>kdFiz_( z&w*UFk}OpPTDIT^0CH#2%2qXzi+)njxnU9kLt+XtQ@EL!KriJqjm7*aym&0uGFd2C zlAnZW=}CyLrymekDqPLXN_o*A)-~ah;0q|U&%);T^Kd;aW<^nVJ_RQ~+5&Xp7q zM!8tpB4-`KhqBUU(h`^-|B@MvO&K)l%Q22OUZIEE(GeGqJ|b}$?_>5=hg|5~GToTS zbu#9f>J#PhRDePO<^u5bt-`!{Xcgzek6ThD9PuC>kF-VOJX)xGsc8VcwpfZc+p&!XC#Kg?rz7hc?C~X{m}cnH zT{NW9lSp0ATwIK&;pF$cAZ$Hu9gHA<)Qi_i#JD&)1zUT_1*s{=jlOow%ZVk=RA*f} z{InXWGe>h_BsY1ic1~8jlYU7GY2~UhA&!>zh2VP-XreMaw``k}=TFUkbk}ttV>W%7 zBxlOEB4LF%%ubA>%z_tL9c6f~;bN>@LZVK1KMqk6WL4ICTc>X$1Vo(LGl6WA{Alg3 zy6TA03Q!viSH?vN6Y0QAh4GKT2L=IAm$hN&V9{jM2CN0BphXx8$S6(|NAE>R6#}As zIOL-RN9sZmR)kEG@0unAPyo8 zQml4c)-l51?Ia78-oJgjw*eLV;2=;Qkv{{8HM>wtrL*G7F3ffc5VKkRwfaEnh%HEJ zJpp_h#~?^*L;ta2M9!Q;Eo&}lA_{_mNVZjzhXO5d*jT-r%5BxbL<~?w=fF`LzERCc zaxK0&EdoDZM3aVTF4!w5PDL${zNT)7!D|6moh?LcD9)CVXCVNC6h*Via+b{7_EGdI zhe(NP;&+44n$CtDuQDY?IZ=8-y?!V6Cc0Rf6thQu3gV# zQC%ax=l?J^Y_GP8e1 z5XoT9>9Wh?lpB)2#cJK*Jp3Stu(c{km8HG8d4|+oq>X05+R6GU(#5I zKyEq$ITUL|18Nstje|7mkiu;wGm6L$$B}$A4U{!jl99H8PQ^(GrVpm5kDs=>25(eklU|A^ zS9J|il$m;`RIqNgCBsEg%v9=r>}niDbz>5`Gj2djB^N&>B>Dx%(z0ZAAe(DE0_O*I z;68FjU#0$M1*IcirAT*0Lo^;Ii1YAZB}JM%=#8e~6`Jawl|>KneNm9NcxYm@1jLb+1wR3au&8)slQ>n( z?hjI~{8Nut;`fA6ZY7a)3l&jsnM1KxZl*&gzEzD?o)8cbwTwvY(3_L{t24)fQ%1=y zB*W?-yJ)T^nerUzXvN{mGGwyND@jQHbC`FgeSgl^4-^8Tk8S|8(0j#=HX%l)8SjTu zK26q<_ONs^rZ1=p`4$yYG3pZv9tyf^r1ZPcv7`s0}alfpzldpvY2J_BuS<;DI(T=JmYI`LKk2CE`VSdj$(n8~%1S zA{0TgsjYZ@yP+HF+59oz)O=Snzs&pbUX|bEHDyjDuhNF*3%wstIZAKJIN5fLr=Isq zqjm`4Fl{jR=BoUm_wQHU*?d3R9-#JUdTF?O=;`;V;s#95$;AHVZhWT9hL@K2rEkP$df-`a z^U=5k_Y1BU&3#Q;^KH-OSWMMDH3A>B5tz(%<>Q&%;iCdMMT*ntEVzTp*A9v)yFHDx zHIYP}%P71_pH!&o!|MES+;LGMZu(;_?DVMx=}l0K`@9tT{p#LFe}s1I&Y}owAQHc| z?Zb-J?c4+?`W(pvlV;_^$R3sChpc;@QC%JGMcNZJZUyn-_zBPB-fP#vdH?GOv{J8y z@xDGUV$$cq^U4w6>WdMckOT3+9)OdzA0aSOiVP9(_Dq5RSJkXCa8V`8xqc>am%{Yu&*i_{45%@qMAjf+(d}9EFGd>vv zfruhW=Ye?rNdy*r8t$BFw0}*JhC6>mA>IGxtRG&i1hjj}c@UtTl0gwBQsup*aELG2 zUB_#44+cK_1Ii*L(oAd|{8Xe?n((M_9B~KeHQa|yL`Hl>F6z1{%5mqv;>qmg@K#Y- zoE(g+8tZWD74JghWoOyk1~5CxhW!aHc)4fAMa1GuwUPMhlaHWx-KTLtvc3l^ zaK$qX2;}E-tmrL1Y8|;d0i5%ehj!IzDa22VF02W~-TP^ZcQu^!BI}_u#^YfaLUdAA z+F!XJ0U~Mk5#m|L(~h_~<8z7iPez@gW$6BF1|c7&4F(YzE$(c#@l@*8;U*46OG`6; zUg^hMjeS@W^dgf;|Mzz`r69kS^D0wiy@hF-yma@bkezh=wd6k6T557pY=uE=wca7T=q(0H4_3hj}S3im!FSN6+aF8Ug8} zv4uX0n%QaChPxtN3lG(;c3LMK8)LyP!{;4`kfS?4$evEH>;4F3Th}mb^Q+zpu8s9X$h%-6I3fDB>0jt;GtYThV>*!IDkXk z3wrSeDfxq>pdy8fn0W@fdM#>5LTV(g&2Zqe<*`^wYR4j4%c>w3;_sR64*LM+(E8PN zX;Fyn?Ln}S&Q)im;+`yG{Rc`c#dWM&V*&j(aVJ{nVCK!&vG~%}h@#(t!W;tdT&;dm z8!XnN7A}7ZL8f?;D9f+gFo%V*Dh6;4ZHT8DVYFjJ1BeHQW2?id@Rk`Rh zr-}v_?(i<=8t`GU%Zt6FV*aMH8k=l?;u^@nomF1!o3{jaIp*MjChi?{J{Xmmh<^?% zMan=k+&wm0IigdtB(66uq7TZLorcdaRpitNjB5nU8q(GLl+nbGR)~2w@@fbwq4@(W zx3aAT9qw?%QZy-Mpbe$oG+fqJiJKEUkwr_APrYyidnlF#id2`-det`$@PJH9MJy2& zG!r%mOGkWfCC|e;0`krHW@;CbmV6fNG(wZyL2kmF3#dQAvz)+jG7*m!T3wXrd1yOE z>&IAeq5U%VW2H6m7KktmD5NH`-LyFQ?PreRLR$KZEv^Bc--WJvilq~gsGzv@0L_Lw zi{>LDXD%GI4Tvdy3%=NCXdnWzFlrEu$xgg^xC`IDAc%wAk+^zS4_+(kM_wGo%V{L1 zn{a^7O@;Qzc={`&=XW+_kjJ@j3t1m!BHa$b9J&q%nw&vbOFP;QSD`kH{ze8mFnDY~ zb~o3fxUCsGxOWRiv9K2+OKa#0EIlR;H%^<$x)YE{PxJ*uoO0t6kxSz1F>*le`hP#h z($vi3t{28TW~NOeugY)AFxQc3PV{W@gqCxn^e1{ZT3Mm<{6u-*@7-BPKqPK4-a#^= z83rRa1M_~JJoE1@{Ci03$6dkIppAjZh!!USX3N5b%<>8fmktOQk$z+&3~@jRbcE4f zc(~FqkDsDkgT4fWQ`wr#j$v*DaA|Hh{=L?PUG{0XlF0E!{^MQD7F8b*Jxr6SHXE_vW4~ zoZobWGW8CWw{>8kwi9_YRft9pEr#_G+>58YD4i!>MI>(JJc_zBlG8y8TowYt`b|*3 zLb_?9r>;{YFwqgvtkP7nmO0t=TS{B#Y%CnPT-(9)L@c2b&%wcDl#$z+O>+??c?D9u z!pxc$uKNV8tsvIN*I`pzH|8IWLV9Nz*H;1}Nq~qBco3VKjYM)6E#nSWJyR2RqTgsL z{nQAAj(}haqa!Mj9^FgVPI?o1H&ze~BHdA?)xECf*}x^*4`goEO|$lXI@Pn&gnb{) z+?NvZKx7hbYC4FIoPif3JA^0JZNhXp88cXq&)2?!FFc!r8|q51F+L4_=WRd>^(^9Pcqf7)NotRAFL0uMjMYEI8(mtJ zrJwp`V6UqtGxt6|eLY0V|M0+7$f5Qw_Mav>u%q_Vbl@uy#qea$gqMho zv$cU6Za6km9KAk1Kt*;3rCR+ch?S2h9~yeQ=@FhD*}F*)YHdOT5y3Zm+7Ly_lFRmX z;3xSul*I}japGd?qkJT{5T8v-rAgp)q{hUepF4VxaF9ST7bQ6o=!zhLhHIi}>VIgj zw_!*5Q6gJjoSU6bIw$84=Z$ntPbQc#)_7|^YEBtMa-!vhem~8DlbHuNRU%Vcv7lA(o>27$5Ohgfd=UPS*C%T`LV1W105`ogJ*McCMM9QdNPVwW; z7AkA?QYJDYoSwglAo$(%bWQ}oJ&4X)`r%*}2N#uuErWeTgwhr;h6SktwTViwGRqa@ zRU1gClSC$bC#doCNE;u?_XLb@2Y|%ZTi#BQ@qp*#7Ye zVEyG}>2M6#7u1zgQ;ou&8M%06%U;~`lf(Gw$L1n?-6cdC9K>M)IQrz{h`seb*rP~I zj;D9gxtAd6+jqcD1+PE}75RTyj<>&bGZrmP!_NC3LDVC=k^Q?)7zIqW3b0hjr z-!c5-@_ekMl`?fzqE;aFVk%6H!08wPaSI1YfehnX*i&1HAHB`_Pf<$0eGv6c?fBc4 zqsZ#<<7paYsN})&^~$X{w|N(m<65mlRNKK!BH|dV#I+@N!$rcwz>MqgHcfaZ^DxXW z$hUS3$vls>N1l#;O_ez{0w1&yFzZJ?Q{-q?0G8kp`Q*~Q5D=vqE9jll$-k5G)z;Qp z`W6BfkHaU-EN8?WGzcTE<$ygK`$)A(g+Xz}Ww07>~LYinye4yTQlHZ)E{CS6WC+`@N0oP~eh-j2OK zI~JuzAy`hy1!oL$Xe5U;MpTv5z&0ZW?nZKRe!Cx$S@EcdC$Z4$z{~)hO6F1WBQ*lQ zeEAsu53PItlw$1Lx74BZ&KuDGAKymb%JUIbUx98)L;Rg{4bHRs@F9w=HPPJhzWVK` z@N}U`XTHq-6OIBo8CQ^_F+DjMn~J+}&F?x<`2D5mSXqSgEzv1@lxMnLA``77R+hK4 zS*H4WDyWEmim4I z^0n88&MEW${jTQw$>OCP_-am`Xg;5)Jmu4RXlkOQ46pKcH`bg|S{aiWuTg-rhJYC4 zqq4Hn8t9obXIcwcNQN0Kvyjc;$f1xn3)ujd7u2!Rne=D02EimXdZw<3umORjJ%k7!1WKLhtyM$<81 zBxX|FBZi9=l)<#q1g@WekaNWh?u0~0IH7|bO9$PyLaF7uwU zo@*^4Aq)y?qmU=bJ~q@0#ienBFz+oihmVw%OnBvSL^P5u9G!%~SBQM9A)=x{%7_)j z5u&L+xG?kCz$=Dl84qQlgn1%_aMs0;g>htqPIBbJT)3{~5H>#k1>`TH$l&br&`-Pl1Ej(w+PrYv6Oc};ZUGW7 zb`+dT;)4wf=gzU#rhtm6Ff{^GBXDXXVD?n`Oqnxhjs;)~;8V!Y&p*xqG3iFuCO}KD zij=$ssORP79S0T_ulL#)9K3f>Vfl@nTTtT9LwIi(dRub2H@NUaM;XnZNV86T~yML~KN zX0KR;R1zNJX#O~s->ajOK>6L@CvnGxKYDxcHxw7ok4r^O7{$H_v(ZsK;@IYb$TTEH zMiJ47pnHMpU zU4QbEpJ4Us)er)r4yj}M34_vm-3#m5zkk2wHTk0-{U|nX-i)0)cVgKxP33oj((9)I z&~;=8jWdH};>>kqp0S>l-jp#ijj1`*;bm_nk&)fXZ1BNO<8j%-(nzoTB`T_1pFweRX_tsL50Yw}KSLgn$rOilOpXMC7oZ1eMy} z30n3HJrv(^@jj!F55sX*JgK-ZJ_BnqQBqGv01=#QBy$L9d& z4-*c}qv3Ssopl|4&jcZ*gxs0}L(5UU!|~%PL&Yimh%zEYI%1GDFH^fI*LpF*KPSpN z*8HAe6(8K>!{1*AghFX^13vruW?VhJ0IxT6GbRxYSt(eW?7)o&_Th1NEZi|3e5ZFO zt{Zp}VdqarYPKAT0rl1GrL{-T;#5Pp9>oJ5W~Hh`E^=h@+lYb93{UdydtB zsXtRAFf{@b8UgKbW}g%MVSXxPEC5UIOdfMJX>~Pan!UA+VnFv)RU>*Jmk3BW8r`WV z2=n29BOVV0&cmndJJ2tyls-4scl&EDYWH|UXunLK#*^3NizIB3lN5yu=@Gr2 zh=DDKFab3*TWS4^kR283eJCbpWJz-;d@t`K(UFF9d@fW55-5VY9~s;`BcsBwban=4 zrx>06ly0CtYhq^;ZHQ6u;@Cc{-?=j{zyI?=fG|J#!naT#~+Rz!5!xi26gu#pCVgM5_x@s^BBy0upTD( z$9ykShPh6p81o%;HTg~4M9NWGlTYQCvUD}^6D=<^{aHakOaoqe=_Qb8kM5%+M;>h=3rP)DW1V&{g%HEHQNd;*o2LAP}%~t+JVf~~4xKTsSNjK|$0i-s(CQ54rHilh@%Sm-; z-qQeoDOYRw7byZ24clHK67oGFod}AD1atlmO$N82sV<7tf^H&xWJ$A5tA(sOsD2Jo z`HM1?L5fcWYwCvpYGTATdTF=4@C4eDy}t+wh< zfL>c835Xr+W_w!PSh|t&?+gyLg$RhKMv-xG=*nA!*q2^LU{)m3Z+w6_QY3axg5+c6VQy*Tj*7({dUT1f>)V!c?jcQ0|1X%sKuPo_6LK5{wfWbi{D1NJ_P#MyJux#4e+v3MEI zPoY>vh=bA-H)!hZ)Cf$Cz(hwth=O$%@Tz-*IGFg+$}{gm(<+^b_Yx5GQ(7RBXIf7` z&u?-v655GaFiuX+>j;#SOpZH~3b7)smVQ4dy};*S{0cPx{$7N4k$5>LnbOfZ$Lu&j zN5IqRfHcO@OQIU}19;X)&0dXjuKPR&jvhl7jnm|j$}LB;rXAMJN&YeY(AD&DB1J_| zRy;c{0`n=UP}1#%BO#VMqaAhO(QuGU;EJ^4kcTkFW3Rw>&tAAL%%UZ$FdU@G{A(IZ z;22mWsDb(xJgez5ErFNadk!IO+b%>Ek|s%1YzD`|J%<7rei9vz)xbA1lV*Hf$f9qR z8>qbgryV6Y_uj4876Z(_a0UKt$Md+!6N$x-HKLYTLuUX0KmbWZK~(wL>F6WDb{_Rz zk1mbHPG=rk_B+tn8$>BNQN>=u5*%X*c3c?gK(b`t$z{4Kov;rr8a{dSARali2XT@} zV1MRQe}peA&WFayjNe4Y%%nAAZpQs&$_mY6$}#0n^lHk|)s$u8CR*0W^k*Ca5w%c2 zP)SLNg{@1JYR{fMR=b{k_E~J)xY6<~e)rvXL+nw5fMjH3Sn`OEJ@yzXDk^Zt9d}rH zLI*-}X)0gb6QjhVL~;OUkOM0d|7gPz%T+(6H&jgX{{611SL`U|6PHHecs-0`p|G)D z2mkqLu1z9UoR(zKk54qE2~!Z!-+_qxUqeiIBFe8@ik&-lVOmxOIumE$lEZ`e?2%&J+`Sc>ue<@jdZ`vK zL@&h)RcznCzfS!JLgw)s$f=+h$KzYMudz++hqWUnZ313p+dRkEIJRGlZ>Y&bSDDf! zpoC~{i18HJ>aXO&W4cyuu3VVN<4%yZNJI``YY@dOH-!x1%l~U9>)?B9J7Iy+i*d0X z%FD-sY+9^}oiMz@HS1pLTqezda!vknoUW7|upa7JxBbWnb^om`?GTB-=v*HXWQ--W=Qow_Q0 z6lbwH$R(jO#Bd^2qXFbj&Q&LoM#nJXI9eKZ?tO!ko3GR2fN9t@uoDQw>G#u0$HkBS zdI)J`yMDO+W&96;)=#gz$YTATsBN0B8duFxBWj@QiSmRZD-&q#UrkbdMcI@WvMvVV zv(@Fez0!;GDpT=%PA-yzKFl4Yf0~9)__vkgwhL2`>VKQxZhMlP6N#7q9Eo-ifBLZ)cdS2yTxE=(74Ns=w1Ks^euK37Sl45pIn48A|T1sZ7{*K!#tnL^;F7AqcZqI zZ9aI27!2Gz3p;3x!(Ht~bW8wcwcY66y%XUdo`$H8Uxc>1w-J8uBQtRjyLP;ch)b_U z8cp2B)ACLS{fE4I(1Y7MxqZeH{Gh0+y}ber%WlAL;tCLRqzZfKn{4jF8Mv`18Q-Td zn()eQBvg^pbg&fJB{H+=K*L#=579@Dj_JOuamfxA~pQcRQVV zPi0Pow(E={AUZq+)JQSrfd?MIr#|&53lS0LLyA*!DzkU*UQ2mO2uBwul$62}saPcr zhyeaIYt~rLM~)o9x%3HP211BHRaKP*TnW%pSwbufz@>Ua4{tp;Ha1%CMKw4X1VrWc zQ*20O)swm@d2sjwR}8bc~DC(=dcnHc|dK z!pX`_>Gfb!h8?RfN@eFGB{EwjE@lS~?)?kHo&x8T*%XGu*ai6c1zUw!)(U!Rw=|Uvmg&T0( z`7Jv+4-?A|TimOlkSW1{OlRg^=z<#k-iCJidWn&R5{ZIwu5y^hAA}3CPxqIS(-Reg zS)5w~TuUAzO`m!5MJy&_noTa7sBS`TLg#sCC6ZN^LYn59+Abu0w4irFs93a z-`CMf{9wk3E2#P4qLrK7wasYW@&>;5M?b!~p#X2hy71Aa9-7YfAg|xUYdcbAO@rg8 z9|12F89ASWJm_pUehq%EdEdSN3fmw4FM`vl+tyh{&LBCI8M&5e>ma}Vgpt+{=8#YI z>aTnRLie`u>`9^NHFV?2V}|%t_C0P5G=6EfY|!lEIYVbZlyD~UF+Cd_V*FTTtHH8{ z4jk%^!yj_<@s=|ZUu>#GPInnR5@Dt|W^z(2^-FB@NkJlM(;*~X`X!q6-v#Go|ABdS18h{AG`kc#5zCoR#5G&*6u-O1~yQT&o;>G+)s{5kg|%A z1f;#>n8eW@wVi+>j`qXNXJ>RWb-pr*?~u|zi-0@ZLa2xK#7TV(Y(d`AB5HwAwvtGR zwc^_dz&}l7C(etw1^LKLO`+}zSG@&kb4TbRrM5qkvW}F#-10c^$k%{Xw*i0mpR_-| z2s14<@jD~eNNz*Sv2^+u*^LY5X0u#wsPxfcDZIDT(j@O*E@qauas#&Ya^X@}BcI<+ z)*YX@!CRagaj*Ej*g%%a@@3Ua&*KS7r}8;(PAYJ0WI1dGk&lvLy_N!f zSKdbxQOL>89d(|E)+Gc)N{J!>E2_B0`eQ_h~8gxKlju-DU?9Tn2+EB7Mjx+~$7 ze$=dm$mF~jESd#8_!HQ+{fG!PxgQj>Pj!(P;0+8#mc~ z*ADL%cbV5#nt$C#ky+4!QY0zrMR}=6i3;KzM+XuQl%v2CjtA(35AJSUR`&*pKKZC0 z=%!FY5(;xEO3ky--AhYb=?f80fEq_PBaoWIGnISHaL6UxB;uANMoch!sd+xrR~>WO zyOIdhqox;uxdm{@huLG5=p%xa7$j`NGmSO_;YvAc*?p}QBjd)((c6s{z3qw)>#qN@Ndg-}GsiWvT*_vW2R8~T1c_n~#F3~PfN7T1x}J^}Ia zOiD-PTyPZiv8*!DrZ-1k83o#K3}-ZR5?CkFaqKWXx<5xpLf@sB)iMOXz8P3?ks7X2s8?ZKn10A{p-**b zlk7T%0I3I7KV@yJ^bqi!8z^T+HnYD2_9tiqeZv>gmpk23K6kJW;&N#WW#TTDGMaW; zx|p9DzzlmIn#$|Ywj_Y9gKWQ_btgt)`iuf(lJhi|&WPqwj<&e690UDqGk3sn%HIm$ zP9s7ULnZuZI(8H%$d;3D>4bDW`~Z3%egp@8eFwtmz%eMme)!v|N1{4QWUi4My3F(p z`XV9XM-j4MFF6Rb*kz*?vVePld8{59> zIOYwJUCyB&Z+#v=Ke!X~iHK@Y5B7GG`4fj6+J;v!b9u~S_`;fN!{6;>=lqGvKbd!C z4O?sHWC}P-c_bFxI6xcp5B~|Zndf0a_GNgTL=th}Hnr`;hPFr0U%nJ&ap#kBMdeW* zRz4zsZJtzQe&iN7X;xBw*E|cS{dwH-W!k3P&vp0;*GSJ$uQZw7nO{>8?Aso-{M-~;BLFtAK6uruze~PSG z1!yUI1#w}ED8ZqU1}PCjFr5}9?XmGZTqu?$abW_y0^qgR2=R~<$9VStvHE(t;#1>@ zfOPvTebO$8UHj>uBO;Ls`9wx|)&{7{VV{wSZ8hEaJ`syss9zDoQyL8|xPUq$Rf+Ky zTw&S1B{!nl6_2K2V&2}!6l4)TsV5wCdF@^_(uDc;f;4RKlwdPm6U0^T!bQLPGQw6| z4PRmko|?A`kJ61m0_%3s)44k`7U@-;h-rEYYc^)%vKSiy} zL>kc~r=NgQdokp)shN2%&CJ6;tosPoK6)>fWaU{~w8U|TAR=~cLg)L$M&G~}<2e@8 zra$K}1ofW57MjgLn}q-%aUjG&P>?k&+0)i$AwDO;PKxqnv~s#g<`6BRK8Y+V%b^7Ec^IMVN7 zQUd%|+eFbB=FAY0P)mGlPb3kSLB6l+=(aW|r}QI;h`31=jU(_H)W7B_827Z6F1k zh}JklBFziy(6o`Fb+X}(C2Q~~?ZO}IW~23OsN?eR#w(Ga`NiiGiRqYmKk9rv*#=EQ zKGQB0Ra#nV9isX9`IAuc=`Vs{4oU}Nk-BIFDiass2mQ-X$aT7@ry5_~_eErKy|-@Dn9+6z(1`3-h7UllM9ECeC<(Yjh+5( zP1P_p0)JK`VD<fUcA^sKGa~H9STZs!k_;1CkyG2NcW?U zK8hP|xB)U`W6Ct?M!SwD&B%O$A^LeY&W-@$ z$-#&rA`wrKY|M_wYWl2-_IJWXA0M-(FQVgIA{aJu3CKzHHbUY_oYzh_8W!AH=-;1_x66=VB`7;YKt?c*s%wPk0W|EUb z0%}WXrzmKcPv0m>6hH6INF*U~kRs6S;7D*Bcy3aPVuv%9`meFnuNy=c_fEHnyd+wF zc*j$SJp4;+%Dx(l+Wcr)HV?(*IF~P`+G_u!G_(Vpx9=`I^bGgp{Eg_1O{as%PI{VW zvBx6N^ZyztQIdu{ia5tmMC&rb5MPc?!#(}+xW7LfpF8>t-gdTOS5qUlv@~E7wqZz61G+GehmeAJmSITRC$0I{-Ov;gUgd{nBE59Zt{qe|2 zxEW8~!W?(i1C6^eM389@7q7YBr1pO33-wes6C2}JqfKxyKViIS9-9w58y>we;glv?WfP;L)wTeiRR^8O^s3WyBS&HHW z!kzS+;gQ%6De`u5Kq4uI)=2*VQgjp7KnS`J3Kzwd2Hz&jvxmAV6wR@yBz$Ed3-TJ8 z$nlkagcUS3S)oLW#G2eS8|lBk-ior6$=Q4wo&r98H?Z+)vMz@Z#p8&F#o@6EdiNi3 zZm2wlOTGpC>h_^Bh<_l909EpAA|pNQKbh@_DSKfKqIdls>yGR}O;0*r?Ci##hCX;6E=3wQqr|jy#3b4gMLA4IJCTe!>fJQA z!b60ti*qJI7Pd$wayQl@vWGUv;}hudeZWF;Jmj)yL;}|LvIf#xw7T;WG3=*VVsASY z=5GByyrf#$sa{WsQ`o=#IueflE1a|P$Qj|cVLLUPBZKxm%(BIIh^b7;6Q z9j@v0*Yd_ypFpL%2kkUTxHPX2`5I@2iIz89>{-8txrY^7Y+xJ>7Mw?V)v@iTYHhsb}Q9o+mSyzNt)8r@2mMJu1trwaG5`MCHk+ z!6lbm0wEJ?*RI8Due}B#AV-fLwM4W<;JN6ci;lw<>bv|OELy}h#URNEaZm(f=$+|< zxoVHqy+LG5{Ag`5@66TYQCbhr(^j3SEU&DI)~s|c_^GE!W$G$-5HrXXh~c>sEN{l* zSe_d^ZRYh8$w-t=rk|Yh82IcYz_E=UQsa4qkxMX>ggc(yi`UV}%Wr{eKS@I1(0T0Q zxu#!eFnv*+X=CX7_q#WFR8DB$-Y=hdFHv-baEYN)Y`n3@Yw5DUMN#tAE8wVq1oOJ< z&@-Jd3c34bEwno22x18>XGYDq^FNb$`^Z2f7DDt5f${RZFBg+~f%%`|;{>BV^)lHbA zifxVIJUW$N>JM$x$;^q-+Hh77kkE#h4bpT{K!@VgK@+F+dN%JAKUo71+LlSb*Zj~% z$BT;#vR&K`u^veU4N{5RFd(dYnMlT`l$Mf5mXRnur_g4foePvk8Hki{v*P%m_W*5V z%d8zmgb38gJgWuHaGu3FsHD!P!SY5#EG1?xaUTVF1Z}P02DWMFS$|cT{={%2o3jM? z#&6NIZx>QBGPx&k;}F6l>^^IFk*RKlu@H};x0cD?U{(P|O6~;KT{DE>2r{RPJC450 zP7w)F-R;#Zdp3%eEl1U^y*PB}FqV?@rEhBx5Fx7es%AvwWTE%g4AFR5pp`};iqZe7%t@gW`4IlEr?jdgMees%m}~%M+zMz6=w4w8V=U_VZCEiYm)t0UJVc+Z^HWB zc;k&$JMX*iK6(!BwZ!K}A|a+DX8oJ18PCwY!MrnhChKai$xY6sTr%kl(OJ}QN}(cobqN}Iu(o#6GjV{j+Ix?}v2Q!D{_>%DV9lH1FDTus zul^dy&L3*7zU{Ecocd$-%hT02Ge#Q6(6Kb*Yu-=R)s$=MQ@N+RZz|vX&QyJ-pJ$?P zs>{@4u9Nk>&PF-vkLLZ{xpR+KziZbnOJYKXYUC5_rkifE5E7a8mnBW@AF@m&8xWh3OqU;o z4S38tqq`~fU_+fJ0_mikWl$&7%l!82LWFhEkV-ps8gy>c0EQ)n5Yj;PP3_T7ZB(De z>PJYncipN_8WW9$%F)2c`k?8UB6%*S1Xi@usa2N$Gg6SUe(DHV!V*50+8yfVMe)+5ksHFVn#2sj1hrT~-ELsaGNU%`Ij(4W z_SQrvmamOMlf;_2Q4+`U(lj4IaZw=*0zx1f^bQVrB~?)Rp(e4gVUet47kY}$K@vF; zMwan-S{2U4Fs{IKL3%m0JgiTj^G=qJeGd4KEfk{M%W+{p273^^{6~mP&A`0rg?O75 z(54j>SR#3v;Ndit>mah@Aq54C^O5k=&%yrhy>L*Rsc3fte)?=R4u0z*gi~?su5EU- z(PsE(2AXl%PI4(Ox&&cgx(2=B8W;7|hCXfrIx{V4tfSz_kH|+Qj`%ALWd&(a$HV#(6`k((Iz7#L_wS{ z6L-4nKi%calALsPOpsv*86!}fHLr(XnDu{rzBAs8P3YC+F>$8cNxKR{;MxwSVTap% zyC@>K9Lw!{K=n`}w|-oJL$GJg!O`4>`04*|?>*q`uC9CEb*4%)Dx=<|QSYKjLWn9l zm~I0m4gr6H3(1S~ocl<**Y^|OBu*YTO)<<$nrY4g0u2j}5! z(F*1{T+gpF^C~uN1!Gc%fY5ccchvp)^XCTxkIuxrdGo?)naqYXG&F?05bwVGZn!6G z_Uzf=Xj;;6=mw`vVo=9Nrg9($Cvajk0iRO>H*VbcVVjx;KgyIe44lAg(p1XW$H-i# zQ=b}W<5iVsZ9ZZjf3L}2oFV+HXuD-puGwb$e)bDXjM6|<53`##Jg1>|El*;(3`;aQ z;quT%Yn*wmRVhQ}z}}mKW_rRoQ@VSj+^`(-1>Sl{lwY(TAarDj2f=(yphMy#+u z(jLqq2L{_Wr)nj_$m{HTHM_0nySuINPiAN{{yuAX_eQPi8mj4+T6=NeB{reEY!OXw z+`4zCUH#JQR{7(1&Ccp$2YQ!TgH{l{)w0iCuh9W7+DzU!Xm}W)JlwaNZR%NRe_gGW zA=2ZwegDtxGw-}@3$@>7fR@Vk(Z<@fx?HkRZ2|h5<=(5b&$B{2<2LlrX%NFy!gffV zk7avzSw9Va4%WbNNYlnl%IX!Z%J}#9-mrhY^kY`7!R98xqx_I7WEI~#7`tQ_J*+|E z=hZ>GYGA!i?CNe?biwj)?FD6J1)ZVyc5bt`v`mau8BcFmt)=SQZI5D+Kjyb;V~uKkQ|YaJ0}VVdGh? zy6PUUjljcd57}oPoWAXAuCV%^>Rd=(s+F!$ON|<(aF`*f(p?%Y62>O)fap{$T!PCA zr^zs0En`|6q!ZrJQRl_BzptfH+M5v^@!K+!&;(rMmrHRb0^Sm!U}!0GaL!U_Lt64t zr3gvMZsOTyfB6fuUp*?LO&y^bbIdfNfg3eujIGkrn?;Kj84U*&vqio57ImVUYg&c5 zOo}9-A6=*+dj#;wwf5Y8T};`p+8*ndu)o>0*=8?YWp7zOoAGo)2BwttN}UE(E* z^yODywFB$cOTjeSI$cp;w{&4xU)sENxB5dG@!GrJ%C*1b$g7L2@^@!judjVkGfYQq z{Z<{hx4+r;PX23a5&IIOWGriaY728Q=V~?w2M@F5QAn8GpYQVs3Mol`vV(ML#YS#7EJ>Y8_Kvl`usDYt4lR+)`&_=z=-mguh1jEIgRuTbOiA?0ld zs*y;ZW$ZNCA|5dcmuW*1CCmEpWYLC{!NP?LZR*si zVb3VDBfQU;G2<9KHgIzQ4s^jc=z%Vu0U?u^KBpI-G?Xbg(7|CTIdL~>qqyxlWFPC& zTj#uJHf39r4c4vSy*KT)gK9TwWD?Xo^nx9_Pj{Y8o@6ca|5N%zMlv(=+8C`f^>t^} zJgd6)lh*hB<+lA>vVm7Wt|NN?z*=;oCd(b@g!CCaNbRd;H#A6>j2HZ3e=R3$Vhp=1 zfL|OpFfpPr{ze*3EJc=B%I41F_|2#QX=rfBh#K3eD+JaH_j=h&Gj$a41zV2TtGd!) zjE*%Q{rWQ7cJ2ALVqcBj|I-(2qb>++tR8IV4Q#Os=l;Ovec}okQ&sk}K9g*FA2r&V zjhVK1tyR9GOCQheWj%B^$S!-QU*k+G1aLXRb|Ay&Wyn z=bbLY?DA2wZ1k#EZP>^W_Q}g{uqk851-`m$z^j!9H8?6>g%c*le;EZ~0CmTK{kBL~ z4b%@CVFR?s9((1Vb5({Z`Hkf*>LSk4NJ6@dcc4Ip+ilFX0Ha2YbD!j*RQn| zI>ueSUF%WT!-fnRW{0+|I<+g+0r_D6Zo6sPdB@2D zaUFjzH~0DG&tCVa6UYLGNX0Tm+}Od3OkJr1k}0P$ji-vIQf-Z`qdCntY!fa;)vR^= zc8FzI#~T?8sOv;e&2+S;-E*YOmh3LK3r8uwHegjLm`dHut*UYfl%1y98Z}zN+Cy|+ z2(OaiHB=kX-&$uapJa3%b6t*3kNQJeUlm@1)zgONF?>-@C^P*+Slh*UtqF4cM;M(S zfEX_8>)Jasn;?azd>B1I&=Fc1in(*=hJy$=)gz1|@J$_&;OIojCFBp$30SwE)7QTG z@+Q?(&EPa1vDK!J?8O1A|WQ%1R%mao;3#oXLprplyU{o%AP z+rq80gREkBA1OdB^;^Ehsy;DOg`wFeO~(pDtI{!o-G6nz?ft@ctmPH$Vd(D0aKqOXB&XNVz@q>e)kDdn2vLsSr~-$3gODW-wm^P7 zy~8eqPoCUWvqJ|T{BrYpyL#qL)~CFiMz40;4+hS*QT4rSgzdN5hNbpf+R(VEwx6|J zG~LR&R9la2_gS?zs2YSj~GGmfp72w+US7cxSyq1w0Wb0Pp1zf27Y>Y zj9ww0oi>)zKoh4O7K=8=m-1s&m%NZCmm&XNDBo#gHVpg!?!5L&UZ}kL$k2!S}FtHKJ|X_S^s7Q)gQbcd_Tk^s%Kn zy>iw82ICs7-+|3CvbBUthoh9&_0-CQ4c1txeGh8H%Q(V&*m!N4)+qf@tp-S_WAa!3 zq6PhHE%D1ASmN$4N&$?qgzSLOe7CjLp11qpa&mjTwGOPR~Z9JRVfU$&_qSG%pN z3zi<1uB_`9W*fd+(bXnsmf&p7ew2Nnwtl!yLX!+H@26#3TK%zC_kh^;ZT7=8AK0xk zuCgN)(#e~<+dz9XBgk5<@DRws3~W@d2gP4(G?4LY^(VJ+9<=KJY7 z2#1*nckKWC_;Zihyz6ea0i(vt9%++3MK~(ietmsIIQ4M*-rb=Cg1zE=2L|Q$rUp*m z-MG#cZd|Jilx5eccbJD|l=Ey|*K!vSkKpdN=C?=pJAx+;$EYsZ0ROhP+RbNfWctCvP3oHwG$%ZJT{ zjI^++aO~Kz;dnu4X1P&tbk3yMsN7@vcD28_ZM-(>>rNRR&wJVS9kzYWNZUWUk6pM| zNAs=yz*cWy&1J&YXw(e9T1$HxHKW0WgH<}ws-{Vu906__(%TY4S`}l{CMm%ec3b)H zX`L!gCy%iqr$uS1=bUR5FO1ewt*yGaP?t3h>DL;E{!tN`=KN{REVus{M*N%`w8Erv zJh7+=|b%;K{_u+F_)_lg={!SON#1JHuh5Q|=hf zo#7aez$onuhXKs`>G-DVV&;qNcLr_9qSxVK(&MlhfxYXlyFwjcsT7?$=)C@V9b%%6 z2J(t7!O^Lt?YQHPJHqC3PR}GBbtf5bf5-P9FnGqrG{$iWCm$R-3AbB%?C+PqXaBUl z%tpW1U{75z%^n!4K{qYU?5|Pqox61i%79uK*5_LFIU_U(rlms~kle9zqaB@oy>-9& zw{#tsM!0p?mqyF9k4$%B{PulXkF?NQ{%D3JW?iiVDkiI;KM?Z6zTh%pAsg_F@pIgq z7yhj^Vd{U66Z<^7i=Sh@#NP?`^a-t^xaW#7wqV14n=vqC{LDPXF_N z%PW`Jj1j_7ZjDY~(*<$1M87{BJJqJB6S7#dAasOk)Ink(+gDPA1Zn=f3}fGMie-xA zsbnDy*yDoFs169siJ9e;I@A0e-x=oL+>hh5_w4zc5JwWplBwWi?6znL7O!>E5ur1H zVqr;>?r><45QV5p!)EfM>MV3subxWt_!@OIYAWod^_y(z=8ZO{f9o`HvwGlR<&=)_ zE7uaMX2H?~tLSbWK|ZQYu%_EOTi(&yPgTl_@jCHC@DkiBpieET5W$I8O4Rxs{X>`Y z!u{xTMvrm2_&Fq21z`Fxr&m8Ylnh4dQQ$NuD| zDYosuy=_-Yaa=KUj=i;Dk3IgoXWI?ePO}}`ciK)Vp3!5+Xz#~LTlmJ?)d?s z2`rC5;F8B7jYO5JF-ypBrcOdjEft!oC9hfrQ!4(pMQM%LssY-?#`A+VhgU4xfDb=` z&yp1C9<+LRyNKrS5x8D%Ue7y=Hu%Y`S~2r;+QVo)tXj3rB23lwe!Pl+u}FPZWv?LW2UNo zlrkBARzR)74^Ywa<-{eQH+$hH8rf(a;SLd`^T0t&3a>GK{+#TGh-v;jZu~h8i=X2- z{_Nk~XW(yW?;88`oRM}O<8^HvHS5{Os>(Omz!r_>YvgrKeGlt3`?u`ScbCa{);_t> z*UDzX-nxXiOoz)fjqGEUI$5lFL!<4}^0YxR;0I~h+je#O z9@R3e2iGZoxd}4#E3|~`k7r*Y+e@n>)EQaw*pIB21{1c(mKv{0virMbUQ-h8jL&H9dGR)zK> z*c|d#?AT@x>-hb#J^R`RiaTET4AIJ(zkB;-du`V?$&l#OCUzX%;00SYSh@K7h4EAD z7F~4v+WYU>w>PZRiH!$B2c)&z)G99_QJOz5!&IOBoBOHaWD+JxKXV++pkQ$*9)38& z_!j>N=R2b+xcFNp&z=u){WEWzAo)wQhgRAa8o>$G^!$`0$8Tr!rc z6b_AA?!?qvFB#TLbx!z0oA=q2qs(?3wO3azwF-4Kx~X@~#eBM2SunJeNXw3x4Zxrd zv`XPtXbG0-#0=&_TC}I(*d!sRV09U&`#JQ}SJtqnqB9?!01pRG)AF9Y`N5~?=RD+? z<%ts~hNUz#qQvoU_?|}@GbnO5O&(y^-#62CFFb6Ymcsefz0>Wct93TK_}PQY2YYna zbD53p-rpva_0aMw%^In**V4PI^_Vs$RPqF)2y%@Zx7p#>7U}7*RraUh5H0!!ErA~J z4WX@nG5+m66FcUs)6rg<3Kmc|m;e45#Nrr)CVtIbv|v0j?d+uEbXUe|`|3ws^1zv{WchHcwnA9QErUbaA& zjXL)JA>ESrh6X9S+Wy5d9{Ue#wFy|JjqTtLqn$SDhI_QY|0V4&S)(Qj+dxivpUMv2 z^wmJ>-51tdeSdAh=9Xf)=3f0Z4%uYq53AM)zV^#pKFKQYdBB=4o2`qFb;rlQEwtu; zdBS>^Yk*+uerudI+^VmcE(2Su6Q#8ek$614{a*ymG9m2C; z_D}6q=h-d8YjjeW;A`W5uPqzHtVGvtU99o*m)W;J)!jCCtF->I5x75Wgm`|DmcNZ; z&#rJjM|YEbdCb{1dB9++ksdG8U0YSd#@NRB&)QEH>!@FKhHzX;xLC6ozh1puJq9gX zZ9Hms&-=N3=GjN>$wlf^$~9`{0om9jdu_kIyW5~8R6mq$w~tj8Y9}_UlZ2yw(Q&Ji zjYBTKX%K9s`#^Y@mXBXPa-7W>J1O9~T`<^ZyizM;{_*AK1B|`G=^k}{J~R7r8?F1; zs#zr?91gN{3)it&x5n#)|5nSZ6EZ4x76`;fokYPX8f!mU7oESZ0=?DflYtMthEV!YUFM&-=z0mqT`hZn@=kIOr*}oTh!5cSh z(G;N8J8#>pr9~2KEy>!eBlXtFFfKn@q07CG+LP$_Jhw?o~# z*?|qKb!gR)fcs$y8iOS()^AbX5@MaOvriJ+sJ#m{{UvZq*DBxj)?A?xfOQR4-F>)i z-lAK`Yjl&hDnr;1pF||Z#Iwx-FK`Py zL6$OToE%r^hp)&Qn*y`piNhdo_)57jnwl#NmMF2Y5O^o>vn@05rIBZYCd)aXIny$9 ze6#U^P5ndeT;_7k&(o8CvGI=ca(a)~G9Fs;PbChdDz{?bS8Vw!2F|B4OmqTzE7_$B zCtiEslHug?PJO4G(Or!iKrCxmVc&i3CsvcV*5+yX%hE30?7^eU?CQ~4^0T+nUhdw< z?%%E@H=EaL5OIRp!slf#s1Xkf6C%rywoNdDbAN)CZ;qN{w)p#MwA7}BLt@(EKk!vsDC1-1iTDjHg*wE2h9{rH^*xjko{tFj8uft@@?cm62>pNzs&Dp=l z9<11Bvv)Mx`O99nnmX;rt2kocsoY^-(H@e|Yp=-Q!9#5Odf5va^u2x9ID7BlF8k(+ z_w0F{Qus05lfVEkOTa7C8K~50orl&h(P1|8?6-!EwywIsc!v!CZ$9&ktyB9hqs7)} zhGUv;x<4q>?SH=UYy16+uCT4)M6`rXUOr$yS@4E^Zq}t?Aam~WrC}w^%rTQC0%i8% zmi4w#vny;=-zd23^XZbPwdd4Hp#%Dzi?0dGx+W^$qbmEE1BTgrbx4?9c}GjEo}T-p z?7=b(S-0r=1}!g_y`=jutuL!yG=tOJWjtaR&=TA3(X|CCyT&7Lr7ivx7EKS=T(&=!$tWnc# zsF|1v+ z3`n@Mk*d4=qZ%cslZ?$#M`O399}~T_j=JX>YuUIEdz9E zk!CH9?9&m4eS2G#MnTG?oSL=Q=J4(#Ry{z*w)!TGI)8nj?ngQD{A*^VS@$b1x5|4jv9fw8#QwePXcdZG4W9S~FDPGV$;Ehqeuv?3vf&Bzk~d}J<&!O6 z9R~SRNAOeC3*zCR6Q1L;#Wf_#@()~aosn0TsFk9ZI8friN00-V=%N-Gf)#4R_bp#$ ztN#6;ty=d*4c3P9?PVM6&4K;JwcR#&=W5$GX_{5`UT+VsnP=xN)|EJQ?;P4=uWsF7W7M{?*K@d*HTCJ+M;(yGwnZycez|R(eM*>NtiPu{ zw41eTyI0S_cGe^RVjqk-%LZOI#5V5UWB>H-E7r4Xon1R-gqBDhv<2(cIb?-K)lple zj>!h?SEBresf_M_l*@ZpVLK& zPs;B3m^w9+d+81o4K~)&0nrT4qRkt$ENy5wKt&m!67G!ZfW+%VIVzA%Y3*%BWt5)H zt{8Q>IyyZV$o@M4&iSfbU-(z+x^$gJFUqV>|Gv5nTd}l1L|syy4ACvzXj$91$A%2( zX^VBj!@|RR?4h&k>|b6=*ag-7>_GQ{_Nt~3uhj;+8g4J_);rw%UAB6i_0TQdJ+&da zr;hHdv(35?t-BqrpJu&|{?b~ujIbV0d|!7G$ogLRoZ0AEN;mKX6ypGR!pS$8DBn0; z{LBbDvmeYX*{|Ypv@2 z1}odN*UIMVS(hCp2KKPZYqe4F&#$rOLE4b`$>**1|7a8OMJU144tkK2d>k`n`N|q= zyOC9Ro2#SaLfvSuUgQe9l&-{q5(i2gI9)lAJh+tn9@FJUHT?okp**-&os94N)Ly;y z&+O9vx`BJ^Q}$qOnf>mT_ibp?GOJzKWWzVUWq0gZqCE{tbX3bU_q}V)b@yw)Yfx*) zBkpu4v*Yq7BfihN!fc#I&^N6L$MCavpf$0+w>fXSJ-z$>Jrn;!`qU{>-tl`T=~KpO zgTuG5banW-*7EZ|x5N#1TS6y(nT+l-?f)$At}%Qa&)cNVRrA{oIsvRX98pWU>dVi= zlljzW1Nv0i&AqE_;lBO$q?VT7a`qS-{=`b%mR_qH=PPVrH`yP()KS}h*w%H|p(aOL zbnk~u-dw_&^(*axRV!@D$kF!K zbDy@c4U6siT^sC*Wtza6^^EP(9@XY&p0Gb%y4aRYzR+%c?KL~7qmg@T*2#K0S@Z9- zB4hd2e{7q&tD~ra(F&#`v=r)pJo}hV=W4N+H?6f#YOl{lI#p|)240!v_}cO%)}-a* zzt*0PO$YYcf3E%@oM1XfvmXQ10WH&h7VZkcT{rn0`;IzIBYO3*8Cr7n_nL8eTl+su zGe3Q`w2UrkmEs->gVr_n#a};T-_x?NKbU!mT|RE|@$wZQrvCG?kG~R@&bj=h`yPk4 zQWA%p#?M|+7)@oA*Eg#2>?5pB9!z2h3UM&%+GD7viGAV&lM#ELJxR`|Mc=|+i|aq7 zQ?3?hFhnO_?9k~I19W_-?(Z=5s>>uChlO7wzB=qbUAfn8(I)b<2Y0h^^>y~MSGU{e zt{7>5wdgs!XVZE+hjo12`f8d@m(pn|PEXxbt|PQMDdiN!dAVh2V+904BW2Mj)N9q~$~tHluUwYDzd%0+m4v zGd!1$vV9LdZwH=#%Z`3+f#}>N4r!DBHAAd<#we@)#06G4Wwccc){&^v14p*$7ISsB z7%7ozP zys^|xGYsI;t0G~udM4~2`t-G*YX9ILKk>5d>wl@ue0{(DYDzcz+u7&Y1CQ^vO=VSG zk(VR@06+jqL_t(`sAq+3)T$Des;@26;BT1WZ?RXkjOlad%?_9OJ@U@{0Dl=ByT-%z zFTeSf-L6iPFQc9 zl8W1@)iYmPpcNn5JFGsLJ+^kGJ=XBPz4WQyw+ly(4RJd9^D>X?MlLdyDxWi|1LBN= zP@erAX8Ci>1j3w=slLZ?GkFHaad1!o7|C7K8=b1Y{ns>APPwxu4YjS0t+OHJs(d;<>o@8hpCe(knw~m(uR@*a zBi2Kk+C`p3j&*;P_VAWQ+uWnCO;x>TR1*!Dz%4LfEvbd;F`r9e-5B<}Brm*7< z5Bp)~xpE$>Mvje-6oy+-aTeF(AGQ!PUclZm9tTH56z|8`!H1Weu*E!}le$$0&C^yZD(BxonP!-?YZBjv@i zgukCwzS*pfir1RUHpMSbuQ}W+uhZzV?TcTmHo}-~WXM3p8}m*1xs6zf|wq4u#O1RL{TG%t;^q(?0ZHTXD4FujC-bQprfl3U#oxEdAO^&wnJ#l3RZbE2a~urEO8u*Zx3Ys-~RC18IP_k!B?7Z zJLOa}QmQL8+DX4D3H{m)^Gp;>^BCQJ=7D*F0c)Zj)H5`=Y*}4iUqDZCq8DrYJu~0# zoNY{%a2&o87@e&!`yE$^VW?wds4rQ*9Fg-NX+b7vm4KznF^hD^^kxoq@5%l6lXV5w z)S!eIhKCHXbu)dR z9Q%Ad%8IkKv6%0ik(~B)xAxiG^vtIxOZCq@@xPCEf<#EFCk^yRtBbv2t8V0&es?+R zIrUpB)_HgAPaAajRQ$c!MK$QAO{yn5Og*f&tB0~WjdbSB<`v4!AiuQKokYErD}#Km z(C{n={<0F%n#EN~+?cO#7ZY3|Vsv@24?jUCZk%HnDE9(W*mR6q{mIu4QNu;j z+x76-rkhI1_eNQcH(6lPPja#V`=EXi&pYBH5e^oE==y2vdg&a_shth_Cqizzy@P>q zf?4y;&I_}y_r-1195@rSCUvlMA2|n$t|^4`eu@8+vPOcR%!=k;OPYAFAHUyuY^A79 zFfqOUoY>YMb+}N9ypNE6vXPagT7HM*qt4mUVJO~QXL8^^l!A^UbMd`K$W6j}@WTn1 zoB147yY1-C`+t|(JLt6cl*rd}yxij|_<%2WuMRGhPZY`D%;Y+k_ezv_mzznJRz8LW zC7K0{SYFP4Idwdpoh~10P6e8%pkjy3Y?!ukDgMffeWstPG8u~C{b_v8aQ)q(9?Zpl zRFqhKU%7*h;n#rG^FPgBiOk0DUL_$1OHA~kluqiXdRG~ejO|m6^OBVnBPym$!P^y@ zSsfB@dYof^6a+&uU4TQ!APPh9E3rgo?4CKk*-M6sjkX4EeW zPJTeE0&Cv=3an?vIW2Rl&@|PiCjKHpemrhQsp_CKsWg5rt5u9zO+;Er1rYXv%0ZF+ zDORThpe1@{Dd?3UBc%L^iXBC0bKx?R!IPtZ)!%}y=$)SKy|?`^XL1qnTf}01K@l^5Ealta|s6Wy52P79ZD~)*ouvV z89m?9eb8}!vlrNKl^4_)N_SKHORJ;DD=EP82VF&7icIRpeay#l=nCkE^`P|b{w0VqQ&PI- zoO)iC>HD#Xi9P2`&Ma{;yZ<$rR={+|IkpC|)`?Ac5F+!QCITqK<^2vvcknR>so#9( z&&I*Etl2!WTRwnBXuLQ*KDha$dR9K@!-;p3RbN9WRv$|y3tNnc@1m;VQ zHbLlvzw9TvU<1y&Uz_+lkqJfBJ-pNglDd8j@MIniz-0?~T?UC%Ju-{#v|ZCcXqOh{e@&$&-IX8gO@YS0+UxL4B^ak6K}Gz_sYar&T1or!Foh0=E) z$^P4NaZD$i*!E*Xs|3m!y7Rb7lckzvFYIEDkzv~wuSD6!LBRXOEtHzwueS!H)$OZu zC!Xy28p9=T&~_fSqGhZgO(w<>iXMXvO=n31?@)2?dOT!HI+ytRdqcU)!O4oA zY2j)#NYek>x7Ovl{v`A&9VoEneAMt{?pOVqba@B(q1(t;!&QE-sh0v}*I(YxwPKt( zAf`;r-)i#`Qb@jc>9tZl?tSngTrl6De9)(&s`nE4Wsx`q2%egoic~7-l+^bU_ zXjT%u!_Bdoc{3|$vHqzbdWo@Elc6nU)f25*N8-}K`p&;_&GuJgyA^HDzmPYvj%1fN zaoIH~svHe@Ipd400WRf#xQ#a||F;VAL$Ok8!E^V{ZbI+V>AWzwdYreWV8xk7GbASQ zT{=Q~1vLHsc|!znDa18NG~*kpold1>?cG_q!{?j0(A4Ad$889&+e17m?ZsXA)Hn^Y zE#?czKAZf+w$Q&q;PhS?OMFjH?ORc|sa$va>E+UuW@7f$-lG?xWlpb7-AwAtPFozT zFOG>f?;CvX+ooqi))oG3+iCkH_uc#SL&g)U;1gtB#iv}L;Jd?OL8&Kf>mEXvt3i@KyrX8uUo~7>2U#jk++UVKuQ#|0#-`kX)(|^) zTkbGzOpl}lRjFl^@NR|tJb3H;>+37uhN}7#OMxCiPlay3U$yFw!~PuouT9WxHiG^d(j!Jjjh1Snme>I2|x(+;E-KZmaQcsiR6zfb+gq zt%_p9HQKQ5rxWR%Vvmid{$T1{-@mhOw3EUwW%=TC6;|HZHQCI&4@gGbRm9&aODg(5 z&&n(iR{A;A)j)21ijj0do$X8J1k~xpp&=`a>hJF%^lh9S(QV|7z_u6hS(=?tN)8LSt6IO0+$EMei1oA*1?V6V-bK<00lIC zhglldq)KabfGtSn*ib@02S&BH+r3>)qePPr!1ZMPUY?OX^Hd+) zy~zXxezQZ5bP%t`OYhN*I}ZAoSu$t3QvR-+*n`#{xOY8fxYBDUxuhTx`NTp%P|-xtj2Oe2@Li?&*e3`# zTZPB7ykym4?30lU1~c|Tp3j#Ydmm5B4xcoBalXQr?ZgsKZyEe7mA2Cq%1m+75+(mM znKjvuW;t7^7(_|YLs#CaUh*EKMA}^gSX7bdX+EhPg zNmSi>!Kb}F|HucWZ!CtJOUlZAE$eej6h`a#4YbjXJRs%C9!+Bp)P}tQW0yPm2&i~C znsVcj{!`Ocph@BJk2H=)dfkO$; zr!{#)wzTZ7GX;+V1^vJz~@1jo_vXnQJ&M}*$c6+wj>oV?6bODN=`apixyt68WlBDKzI3$lEOTR2*e5@<4- zpmRN^C$kU0yyaPUoW4&-{9{ZBAnZM2rQ%^Rtlc`Fsig$=R%fRZ!!+C)ieVHzcg?FT z;aTpZvQ~)eOj-BswNwjfFjut_d_qK+`3d`Sltv>wUF#JG;;b1ob+q%rs8F)=a?T3* zQhE51=NFrg$zJEC25)N{CPXlOB>~Z%-+mnD&E*v*6!A!_=J<^_AS6fYm2k<)`vOGQTHwsmskXzMFU~XM24yK5q}5a+xMB z6PVV|kuYuZN=@;1Vvt_xiU8&vK57m93CME<9np(srtUn~IWOgcy_u<6jvRXE?wM4M z`TPy`1Hj^Z<(Dmw+3xrk?g#!7^)N@wiqkM{V(wo zN-_W@ZA)Z7nZWiqDA`5;ksd#N5In!Umqb5rnz$KFdBx3EV8eoxYPq}a7Qt~1qUBb6 z2b>UznZ1b_{_I2G9VYS^-o2NWC1=e)N5$@IY{_EYMm^+-#RKKM!sd$gvO;N1k=(0k z;1~%%M0P4l6OSoN=B=eN3--30=EykbJT++yT7s9lL{6=Ry;b^a^+80mLs_VFw`Ah2 zQdUgfn^y+qVm5Uz`0OE7f$hm1)iYnD^p*z*DK$$^>~CIV{X9m1-{ix(I38*Kjvf13 z6Kdf*|KZ^xV#gtiuY#JP{+GkbBW0&SK9@Rg)!3Tmz0H)7vFm*4XTWNC7m@lArkMIL z>3oV@&aYnG7;@kOd574|vV%18ENSG=NCwHCtQ9USo#63$E7WbVJ;1*|$NJv&!|#C4 z9GBZZx()s(+iHrTO686`;vS>?cLts{OP9!)yVnAJ4$BN`ebj>e_iIYMa4SNNS;)Um zlLJNvN!!&i7zmf4Fefg>7tGzfPuea){j%9r2vz}sb7ner%P#J85usaJTH=|mhU=UQ zte)NW0?B#r))CACNz;C|L03II@(xYv>|gO;4!r&FGLWM~K8^3lgxkkW;y`rGY^5`O zka!xnKA&ptC$QeCmdHzHwz`yg;_(+`)>JHXPGHPeiHcqfcn-t=t8s}~D&nP_YdLi? z)O4&}Os32dOwvQ;EM?Ft*y{Hqlsdy{m{^QN;X;Rd1buzISTbsf$UX(JB8Cr$CiA|En04UlC6`9i!?31XR$sdYXVW z^yMCCVN zr}|F*g%4O2;*pMyymp+^My&ZJ^6l4zuQq`nhj&sk-?*R6TxeU+An>LeJZ2K#9ia~t zV1lIV#iGquZe0VIhpWktE7|$5``e0$oY=jpcg&qy2zuDa&n&_n`6)s-=B12**lUEf*aPo?-)6w4k_R>dN$9MCE*7li5JJ;^KV8XILrQD=l zidJR-9H>iC5aat)#zy)3lp00Iip5z0(T;56&Snq$Y{!@i4e+abx5DY(Ql=)cqA zX&{>I+5wn>G2lyF880OpIM8|wA?P~sd*=+H?KZQJv3Iz&s#Qp| z8rylVukTH9)ZU@;j}8r{Ht{1kARPq#T5j4bA-twi*`(lh9TM9WXZ$`B(Wsg_$32W& zVSn%MFd0CSQtYkx-wfrzM+K)B>lC6l_XwkR#mb3TTm-KjuZ3n{D74QzxdTiC!~Z)* z30ZPc|D1+x3A>&xA9e{4*TR8@y(0Q;>A4`_Dx|^`%AHm0*+Cl}(?%d}GsTHsJR}?|t{ywJ`FwfsSf!S# zYq0oMV#koxR8t+d7FcI1uWd#;Cnf$c|Q0b}TE*wMi;G|<_|&nWLG>gZvd&Fqzb z9t@oW8`Yl$hsK$6?AJ;j9l>;7EXlD9Z|3Xy;AOo$UNhf$cGHTgft??YQ*yTt*=0u9 z*(4*QmO$F9hwJvbd@M-RSEr|)2bwme!#9&(bGW9cE@Z3${EAiz0Bq|1AMfB~?%4Vc7{? za3G)M9-$Y^5hun3%RY~YMq#>+Tvra==V4!`;=HKyah*R!XSQ8-{OTg8p06`W!6`%w z66gX;CnNSQqOJU3lX4tANAs92hgkO&#eHlf8|i2ny>&M8N)z@7<@-CbS`Xm^a|NY~ zB{efIyahnO=DWpx3IY7bxCbAf9z&?R=q(STmm^XiT*)r-cM>ApA9+ZUOdk2RccsbwCyY4f9YbVG6#XL(6?#cj?>fd$GsOu8w4)rx>$Mcy% z+J@B*M{YcdDvE7^9{PY87zJ;mW35~8RTMYV3>ITt`MRtD*5q*Ndk)cF;%{Fk7hM$V8pd~IZ-5k8|b|0#p6TzKxFaEm%Kcg`tE>eU=A zIlz;owNrr8`x~aw=U}DPch_^ysTg)f_FXB3vnHofvtE^RjT&aNhLqdA?U(G`J?DO| z_QpS+`fMc9dV^BYl($EgN}PVhuKn`IWp2b%Au@u06hj-M`loot%mZTYiSVtMKRJoq9BNajXP>+PF>}Med{^ zvB-fuhoL?NrzV-%jgu>GWF{I`@EigsG&nl@7?c;~20QZiu%yU!Y8W}y#X6ENh1|lm zi;P<~LKe{0?eGQULAloR>L1Xv1s>g=hw2`T`LWv~G%T$v3iKlS(#2WTANP!Llc?pc zqT%A_(bOv?3z4c$%9$4yk|mPwdOp$cs{wmANC2Kz?wOW`rmS8l~BapN$fF> zh!#S1F)Ftwwfnbg!4*ZDXTD0`cgp6{D*gyg$XCA~qkkCVzRG>0k`_cwM|P!5(d$bL z&JGu_G(bOXw!)q7cR^j>#=Q{f48^?Uz^SF*7vjWz<~#=l;7o+gmW=Cs#e2^08qg^8 z`4TS;(W1T?oimidGE-my$E0;}9P(gcrq!qSri#3AnL#2)8|x3VQy+^hH($=M_J|83 zR;`Hb8H2I`Ajm32m|6@yNrZ+D@&>R^AH`Yu8nIXA<3U<^mXA=x2oKwAfFhdOD0>Pu zS?C~|SvTM#JAX3Sy9D{itbHY)f!D7J&M40;9z3ceaBB~8?;HF#fEH6w=ziswRGFHg zUIEyqu>VmXTg5gOhh4L^D|#pKMq~}u;#ULha1<6Ru$5N#JZ`fuHcbP28G1rEe^~V1!TG4#$8eARM8m8!3z}rJ}FFZBhwFm!RBw`~mGZ;Vxq2wyvJ4w3ahOnMoy~OEt;*3L618SMpta!)=u^8> z%vu8b?!mz-9_LfMB}cWip>A+9uy;Pb?hA^+d3RH1Kt1(i=^oq1NZ3}j^?XQmJ9Koz0jT39; zfsM~giI^fk?<+_->hx6KVMR=b6LIcdBnAlt8_jg9MW|j7YWX%)?hq(L=yp^$B(Afm zA+8Uy{;jmS!BbuQ&}iSa{m$MF-?YmCs5~S&sL$i65q@B1AhToml~rcxNS{HsURF9? zA}ICC8T%N1>$#q|z#C?=&V;X-6`7as}DvX}-+(b~TF{I>16;K0^u6%&J6M zj?UB8Wy-7IF{GaVIbRS7a)?zTU4sQhk0uGkQK8oAc#K1}k%u-%yjR01-UASRoZMNH zdu!|BD;0vWXZ}>``~r~Ey5jy%p7Zo*1dTzT%kFR(thcT}21r<;y@aT+&pLJ(72rC} zEY`YuNa)U_fi~`q<1C~@azOQF+&{e^CfvAAq?4#2i>_(~_@rIV9LlD^VbH^4Y&^9> zHaW++JB_5t>?a*Rd^2j83?Rmy3L{ke@L)2IPPGbDGuH6-MJ6u{-;7ub(KFSiq`X3u z`(CX{`A5*uXy3n0h8}4#&Fl^!3VS5SjDW{LsR;U2lJKg4%aeFc{95rKpX?IOd;ck% zu*|Ft6Y(JmJvIf9#K71g#N>5rQPg7mp{8JQtoGu(O>BN*st^bB#y%rUKLfcfl}Kms zSabJ);Z#2x1WPNCA;gj1|7GYQ+5i9qOW|Jp9`G99hG7@uaX%EWv_N0e3fpYTSS)-r z7s7mu{Val$O1;QY3+#z>C@U=$_+XBt4;AuAi#9tMSGf)jYNjEVZJ#--W7rorT*z@S z7pb(PavBdce<36GblFovFEDfPAeXEyMCUSmDf`f-!jkBgH+7>lLK>&T@8=BUXQqU4IkZ2yO<5l|#IJ8-XGU4c-&;M|e- zzVj;ZykB3F6Q(mgm4IkOh|6zekEH=#3f{7Ufs0V%Z^{h9O@QtW$HCG21dDTS=CJwk z?4c4Vs*urp5%j4KxyAHjG=V<3XICm?FlXUMIzAy2>Cp*WkmZ!uDx=K~Aj532=#Y4K zA;$&?95Bd45(RiinCxb$l;=H=Zd`lyAEzVKlRuJ>WJW|paD8kZ4i4t@#f-lanScI; zS|^xTQ6k}qs0Qvx-@~Kf+}DiQ!nIK0^~w37%G})%oJMiZRs4jhZGJxNY_|?d@IT22 za~Jp*4?-y(K7T)L>u16(rJx*nNmO&iBJ4^6ueL^ezvN1T4mK)RWcnlaO|u1F3xb z^v2N4$+chRKHN6t@@H9pV+@ii4Q!3O1=NiJ_jrzEzzJW+KKIRF>kzK9-0oWPvpWI^bTC)biR`=|#7ySmP-EFW|c z`?)~@p>^l`J4XA=`R4RImB4U*oFiPJIfzX>mM9zp%2Z#HbmOaqmWw({8mAxH# z|LkxqCCF!nXxgeV_20(+h%Z&#*)qWd0reM*U)x*`|C5SgB-nT|fJKXm&uF4DBkP{& z*^U2c-5z%188ey&0m4%fJn*%nb{97i^75X4ECuep<3M<-RsggBF#Z&O`Cq;J5p z7Bxm*f#RkG_3ju+yMEfLGG+2fL)ejwIl!fhMVUa^d=5v2W6%)Q^OPLapkYyF`)|>I zHKy#YMT%(PeX1|n+-(NkwVnmUn_o7+cCIOlIk~8y)G@oo6?n{kaQ;_!KTvg zn(z$4Kye4cW4Fkv)OF;MFfgFGbCX}|TkGqIxx1N@r^6uKmy3UH3B6$@oV~;)$^2c^ zntD1R!cCMxyJv)~k}yMbZ|QdmFg#9=DEulQE6ng62TE^ey2t1nXanYIM?OxUP>99F zc;>hUX=~GW_+Nm87bN#TTO9&oK#Fnnh{8}7gk6*dP0n%a<(QQZ`xUMVG^Rhsq4=v& zvm2s=?eU411`LrJC^Y#j0#Hjp#lzDoq$`ve*h^wggULC7cZmYreK`K??VP8tOmf7; zUC7`Yk0zI>yO;u7daNM$&Chde#iE8jP_U~OVUHItTO=E(gEaIT%OVkx8)zK`t{~P@C}TX%e+Y-B zq7THdenz~$chDO{QVIwCJfr9%oxaE>DDz%=lOt5U4*>yp53JgR&_Uh01hkoshOB(k zR@Gs2tFrX?Dv3xoGLe|2;5Nr_MAuoSvDwS_{(<6} zbkWd#F&Oew^s3zM`5AzdIrB*@58XfLyAXu0>Hfamm6F^Ne=jfJ7$R1WJ)g$hT*Q(2 z$W1_s1Bdth9Ct*d_$Ly9whHg*9U{*5H`i8lqpEzD-6^@x`#s(#Y*-^mKo%=8h;9Nf zMEy&8o#@`RKTC&skBMS$2xo9`;DAV`SYG|}L_Q{&Mp2t4cAaXP8ph)?hPei5ZOF<2 zd}QOl#k(a6FhXxc3`OS720(Wsw4VWdc>~0OzWXTIijB>t#*HjET~n^V>nOqq4H~Tu z<~r#~U;tlXj_aM`Dk z;J-U%@ZHF2uk}c~(_y72QrfDUG*qXu{o(^Z)0xL^GoB+ffllWowNqu=$=i}+cG4=h zwL#t8L_VDcG>5-1wkx7hsG{jX(#+AABIk+{D%sGSMVbpNf4-vRhl`WARjDD`CbiEV z=&DL>g(=o)^jmN`?4scOO|Z{WPD$j|qJn*eBqpnHy9##MuLY+{Mx7C0e(G2jwiv>H z*yPH%k!K3@eT&NktD-y>eeq;jFN{Us*6{}yD(H$5jFui}(3p9HVlNkW!3OowdsekX}#!4-8^Iy1cpEoAcjm zW75r*u#JjklM$Jh4@N!++N}c-ER}0VKN3RX$bnexwzJ=81;))jjE*8uyDTqT*V*yP zb3qldDV4DY#B2I)9K2*|l)<*UcC=&`Y)sQkCh36C=Ig!Dk2K;QM9353=79#t z+?c>$!?B#h!pC2cmQ|C;P&5IHb3EN$$SZFm2(XpYt*u>k$ejOkGX%uGt7kwi(FL8M zEcGU&9~0cwOYpczO+R39xiUv{qFrAMi|lms&rGmFl3%rD*R4>#c^915owIjArasHz zdCqx~<9d!*RYk0v7Fv0DXT<8&`}GIXUKoL|BKgtd^a^KDnpTOq_+|>K8#CBJFR=w8 zk&SUfPuA2`X^cybupBZ_?e(eQ*5$bE!fG)h?PiEYp5&IbD>LDCFz2f+INg2MNXfTj$(1G32@e;(=Y% z#6%RCm{o_`zw+`7cS*>@!&grXSv6xgSOo-m$VGi7S%~QN)8)LKKkd18Yh(xUzW@Vq z&uoiV66l)l0z~ayH*V)~Ik){aAPE}fMO21&l92%%V+3ge@86=jOY#Z81Lax>6*l0{1_6^ufN!uEfg()I)dRM1Kp=;T-O0DIikkl*pM8p%XX$vJGG9i#DzB`> zOVm@@3F2pJt8Q!Z8}W7`Z3Uwf%w|OZ#fwi7MCZldDe>u5Oct<%wB1~uxS02PeDSh2 zoCtQzz_Y-`z{?#X&RH@0^|SG_1D+U-AN_22H#T#LGak<4eLjVQ$no>ePaz!ZUzKl} zT0U4MT)tl8(afSaTBr2=X{b^o@+a*tnR!>`;K*#{oWli&{MAsviiA=3=C5i#;OMf= z(<$gHSkfJEbJ!iGRBEX^+D6i&2M2dea7YRCxeQh!mtQ`&h?*ND zPJM<;85MU!SrWTNZ3!=a{l{{rAgiasQY@;**7BDvOIujXe(XsoG3}mIGoH`#eX$@r3o?WbL~WYHT_gfH-@LHY7%z5R_hi)Od&4*I3xN=HdRI|4gy6=qy9& zv)mm;_g_%t(yfzrD#DHhE0{seB4MJqaOYcej$DWM2X>nhwc<&fJ`(%t6s3}xW8LHA zO(U$SbW7FYx7`)&qNiCFh~eTOzbfYO0H^Tw!Px% zj4_>Te)ijy|FOz!)4@F%ukGLC%#&4?M61k?_YPb0X{pSqB6Y@foJ7a-gW1M|O82B`B3~6TVUbN; z$ec@QZn^Gy&gur2Ksd9nb;^-HkhBHiNMySc6qXK&0pz4v%It%8Gx;e})xd~;ni_Td zZPQ3pd1J*xD!A&?++*)s6sEj$M@1r(<86S$BrY|=UcnlK$Sr!{|)gR<^ zV&-BVpZ3-`A0*vK{0Sa{Tug0I?vOZZmXgvNjK z4)P)#kSB7BkL{eAX!e|h(T&_mtdrN1JwU}j9BC~!Hs}5Hox5cyHmXaJ z#%J>Z37fe#RwDqNpy(4Cxm@Af9PVhE4=++--NgXnXU1r+OUg`HMeT#n=)5^BbW#bW;D4m6(Oj4)ry(rAYM&5&483HEji1*(6-Nig@ru;nUq1Cl5xsHw7&EiUn*>twb?KeD% z*$bWhGC?{z=tNt1KY6X2$0xCenO-Xg!*N#&+!i9%90rHcXn5K`2wsRH3FsN6GDG{z z;%bG-D(ntR%uvb&e0f)bG8ytCyT8Es9>PyMG`edH=G3v+_c;go{AI zCDxt1qLmtF;tIgEiAC=-N*KFG7v3d^u>M2&NJT}{(BJiQhiu10^367MoPjyNsW{tD zFedGy5y9cd53`ni;RcS=JDkH4!Q5G%^S05Vy_-L9DlNB9Hb1Wrd(9%Cx+s1U5R4gO zPUqu5itT&(in2>{0#qztyf7=TnxqKBY6E>-aCs_p?Bqtgp-Kkx_D60(V(Vm1FPhR zBxr?S%+=^`d9K^{X@N*Pb~1c8=dEM>lD&_B4UCSbg1Q7fHcioh@_w-_R?4qW6*j!8 zOF7SV;B$U)TdGzh=nN6WoT8Z9s{4Kso2c0O2^#M0oEaY^!CBiId5l9PeU(cB@$lSk zmRs4^mJKEWw95uT%oW;E?@38l5yniJhT!aFS|44G3Yz_{F*5}S_v3T%?k*0x6zyOp zp80XMAuHjsY#9e@n0zaa2|<$7Hbm^U;mMaX|L(lY$HoKcr_LuOVshBs|D3T+`%Yiy zEs+S!T?NI9iMVc7D7Z~{x>HQ02uMCwLEkn?02R^K9=>onC^M{iyJiR(3zoR-9KO51 zz>Qy*J9M3W$wu&J+3B}SSv#3?Y4l?O3Jch}!ol0F1F3e0ZtOlH&y7P&zAM6qCYfOt z7p$ocs&xeyfZc&sZ}VE2V3yaMT7MrY#e7-C79VjlY`-@80O)kVi$e+^P!Ww%JqCrO z)&$Yn;hZZRpVa`MS+_`}1Igz{YzB8vclV`(lbyN4#0rdO!*<_UFkDar@6Czr$N1eA zJty>EnT~V}MCC-c@;P8e?_-bJ+OfFSTXcsAMf0Mwo9$K_BC2CV;JZ?WT6sik^GGr0OqKN)krT$& zrS!-(#@pa4%bkY+^Cl}IPU_|0qfpK(t|XHE-1H3v7+4MaB43MntbLda5)jGi^Jgow zl9^xF$rX`+hka#duD7Mi*f3>ZGCB+vBGqB1D+&xy=INtXpD&ILaF>u&-ACE|WF8<_ z%tTR&7IuWV1%tzsMST`>F~hS${iL`w( zvrd_M_r!eU%M&5tSWC+LehHjlBD6gdIgRJ|bd%n{ngsvK{C;u2uojxf{i3eMUQd~` zf;R7i+01oMq5CpS@|p{-U^q(7b-WgoB3?6}cx4od3~dw^?|hX4y||Os%ej$9P2nV5 z>Vt(-Y$vF4P_C&EvTs|*wR1b5*-U3&(YeXj??m+K;6dl9K`iUf4Gu@-Y%cCzA4w=L zJ;(rUk`FOM&-*or#=AIe)w7hC#DY5~)hao$*d6*BO$hkwPdRocAvJYSqe`Tr&r|@IEu?oVDr(?CQ$&g3#5qj%(gRz`Q zX+(M)fpi_Z*;%@I9L8gEO`&X%?9c;gxdg%|5Tz>ZsXyCgATKJ!2JQDV;cDfYeO z()-Jvt7_GP&oA*JuB0+*XaE?*(Dc4Hp^fW)4t*&nigEVql z#4P#|7is~|>9}<2)_&5@KALs3NpAqMJY$^E)nZdcCw1wh<||o}Ds)=M-!oIn-X}Hd zj%(TTN}h%uBd4L@hyCxRJt6e1St`d*e0_m%9UPs`cni(2VE$8&|K?F;ErjyLJ*nDcHv!uWNd+I?=+q zB6?It7*fOTH2^jW2d`-W?h)KJSN6<=qz|?jFY6yhj;`!5#*o1+tKROeWo3FQ>KFIr z{MK%>jjl?3Aa<^>p?Y%{sE{S$a|aU_e+2AZTf288(ARut3>l4zTC&bQI|T_Nx0a3$ z5>$P*f!v6=d)@X}JOr)-$={YYSMW`p^csezjY-{WpO69UF{^Jf147+A5$u*em-c?# zTcqdB8Wl^`G5(pLVd!5M;5mmP1F*FFQ%)JjNo!{$`TbCSFKDx`k3yYn3=&q^*b@!i z<~LJaf`_{=!~nQO7blnoY}r;P?ru+>(#zgSH51l=o<){M(JL+F!P>qLJW3YUOSR`4 zf7IKJS-Z1DL2Wwq`WLi)qJjZ473I*I!d(dx{*#=lTRkK-+QD;;gLVPs>eJTn$Hj+0 zy7oeU?c|EKsyIgq(juSUPdV0oeM@uJ?-Z6+AXm-5K|TAveUD^xIndAtNW_6DKNE_; zb`GF&7UED;;mTl;cw0K8I5hF25BQ7xd;k_b*8Im%Jyn(>o+5LaEeHUAaTYx^owLCA z<4ksG+j=&5t!6KfJ;ljfJo{!_q3TL@bO=>5A{5ajIXX0SjT2j3^=M_9Wnc^RKg%9& zokMMTTrnPfwqtkmtTB@?$KIdLJ8Q0WjXgTXGwaX^F)BdK2|wd|A*mo@tQgjL>z{8n zU}|ZBZ$8KR#wV><&Nz^J|Lmpr)7Ibv>t15(M3mCggqzoEse<3fXNvFTdRX=1R0ZEk z(h5oGyzbIKCmOwfT{!eam|>?=ogrILcp9d$QGlU{~xUU z-yVSwh5C;-I@6zpmd*a#EBL>GXMX2ABoPtOW3}hc^#8Nbap;G?roc%F*WDZZzd!cB jqS!}(%nJ6if5blxd9p^D%?b!#BvN~!{k%fSGUWdO0&!4e literal 0 HcmV?d00001 diff --git a/doc/source/tune/examples/includes/pbt_function.rst b/doc/source/tune/examples/includes/pbt_function.rst index d70be66e70cb..dce2a0deee88 100644 --- a/doc/source/tune/examples/includes/pbt_function.rst +++ b/doc/source/tune/examples/includes/pbt_function.rst @@ -3,4 +3,9 @@ PBT Function Example ~~~~~~~~~~~~~~~~~~~~ +The following script produces the following results. For a population of 8 trials, +the PBT learning rate schedule roughly matches the optimal learning rate schedule. + +.. image:: images/pbt_function_results.png + .. literalinclude:: /../../python/ray/tune/examples/pbt_function.py diff --git a/doc/source/tune/examples/pbt_guide.ipynb b/doc/source/tune/examples/pbt_guide.ipynb new file mode 100644 index 000000000000..dddfd2402e3f --- /dev/null +++ b/doc/source/tune/examples/pbt_guide.ipynb @@ -0,0 +1,789 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "eacef74e", + "metadata": {}, + "source": [ + "(pbt-guide-ref)=\n", + "\n", + "# A Guide to Population Based Training\n", + "\n", + "Tune includes a distributed implementation of [Population Based Training (PBT)](https://www.deepmind.com/blog/population-based-training-of-neural-networks) as\n", + "a [scheduler](tune-scheduler-pbt).\n", + "\n", + "![Paper figure](/images/tune_advanced_paper1.png)\n", + "\n", + "\n", + "PBT starts by training many neural networks in parallel with random hyperparameters, using information from the rest of the population to refine these\n", + "hyperparameters and allocate resources to promising models. Let's walk through how to use this algorithm.\n", + "\n", + "```{contents}\n", + " :local:\n", + " :backlinks: none\n", + "```\n", + "\n", + "## Function API with Population Based Training\n", + "\n", + "PBT takes its inspiration from genetic algorithms where poor performing members of the population\n", + "can exploit information from the top performers the population. In our case, the *population*\n", + "is the set of Tune trials running in parallel, where trial performance is determined by a user-specified\n", + "metric such as `mean_accuracy`.\n", + "\n", + "PBT has two main steps: **exploitation** and **exploration**.\n", + "One example of exploitation is a trial copying the model parameters from a better performing trial.\n", + "One example of exploration is generating a new hyperparameter configuration by perturbing the current values randomly.\n", + "\n", + "As the training of the population of neural networks progresses, this process of exploiting and exploring\n", + "is performed periodically, ensuring that all the workers in the population have a good base level of performance\n", + "and also consistently exploring new hyperparameters configurations.\n", + "This means that PBT can quickly exploit good hyperparameters, dedicate more training time to\n", + "promising models and, crucially, mutate the hyperparameter values throughout training,\n", + "leading to learning the best *adaptive* hyperparameter schedules.\n", + "\n", + "Here, we will walk through how to use PBT using a MNIST ConvNet training example. First, we define a training function that trains a ConvNet model using SGD." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "051603e1", + "metadata": {}, + "outputs": [], + "source": [ + "!pip install \"ray[tune]\"" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "ecf2f3f3", + "metadata": {}, + "outputs": [], + "source": [ + "import torch\n", + "import torch.optim as optim\n", + "\n", + "import ray\n", + "from ray import tune, air\n", + "from ray.air import session, Checkpoint\n", + "from ray.tune.examples.mnist_pytorch import ConvNet, get_data_loaders, train, test\n", + "from ray.tune.schedulers import PopulationBasedTraining\n", + "\n", + "\n", + "def train_convnet(config):\n", + " # Create our data loaders, model, and optmizer.\n", + " step = 1\n", + " train_loader, test_loader = get_data_loaders()\n", + " model = ConvNet()\n", + " optimizer = optim.SGD(\n", + " model.parameters(),\n", + " lr=config.get(\"lr\", 0.01),\n", + " momentum=config.get(\"momentum\", 0.9),\n", + " )\n", + "\n", + " # If `session.get_checkpoint()` is not None, then we are resuming from a checkpoint.\n", + " if session.get_checkpoint():\n", + " # Load model state and iteration step from checkpoint.\n", + " checkpoint_dict = session.get_checkpoint().to_dict()\n", + " model.load_state_dict(checkpoint_dict[\"model_state_dict\"])\n", + " # Load optimizer state (needed since we're using momentum),\n", + " # then set the `lr` and `momentum` according to the config.\n", + " optimizer.load_state_dict(checkpoint_dict[\"optimizer_state_dict\"])\n", + " for param_group in optimizer.param_groups:\n", + " if \"lr\" in config:\n", + " param_group[\"lr\"] = config[\"lr\"]\n", + " if \"momentum\" in config:\n", + " param_group[\"momentum\"] = config[\"momentum\"]\n", + "\n", + " # Note: Make sure to increment the checkpointed step by 1 to get the current step.\n", + " last_step = checkpoint_dict[\"step\"]\n", + " step = last_step + 1\n", + "\n", + " while True:\n", + " train(model, optimizer, train_loader)\n", + " acc = test(model, test_loader)\n", + " checkpoint = None\n", + " if step % config[\"checkpoint_interval\"] == 0:\n", + " # Every `checkpoint_interval` steps, checkpoint our current state.\n", + " checkpoint = Checkpoint.from_dict({\n", + " \"step\": step,\n", + " \"model_state_dict\": model.state_dict(),\n", + " \"optimizer_state_dict\": optimizer.state_dict(),\n", + " })\n", + "\n", + " session.report(\n", + " {\"mean_accuracy\": acc, \"lr\": config[\"lr\"]},\n", + " checkpoint=checkpoint\n", + " )\n", + " step += 1" + ] + }, + { + "cell_type": "markdown", + "id": "ec07bf8e", + "metadata": {}, + "source": [ + "The example reuses some of the functions in `ray/tune/examples/mnist_pytorch.py`: this is also a good\n", + "demo for how to decouple the tuning logic and original training code.\n", + "\n", + "**Checkpointing saving and loading is required for PBT**, so we have to both load in the checkpoint if one is provided via `session.get_checkpoint()`, and periodically save our\n", + "model state in a checkpoint via `session.report(...)` - in this case every `checkpoint_interval` iterations, which is a config that we set later.\n", + "\n", + "Then, we define a PBT scheduler:" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "4608918c", + "metadata": {}, + "outputs": [], + "source": [ + "perturbation_interval = 5\n", + "scheduler = PopulationBasedTraining(\n", + " time_attr=\"training_iteration\",\n", + " perturbation_interval=perturbation_interval,\n", + " metric=\"mean_accuracy\",\n", + " mode=\"max\",\n", + " hyperparam_mutations={\n", + " # distribution for resampling\n", + " \"lr\": tune.uniform(0.0001, 1),\n", + " # allow perturbations within this set of categorical values\n", + " \"momentum\": [0.8, 0.9, 0.99],\n", + " },\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "c36162cc", + "metadata": {}, + "source": [ + "Some of the most important parameters are:\n", + "\n", + "- ``hyperparam_mutations`` and ``custom_explore_fn`` are used to mutate the hyperparameters.\n", + " ``hyperparam_mutations`` is a dictionary where each key/value pair specifies the candidates\n", + " or function for a hyperparameter. custom_explore_fn is applied after built-in perturbations\n", + " from hyperparam_mutations are applied, and should return config updated as needed.\n", + "\n", + "- ``resample_probability``: The probability of resampling from the original distribution\n", + " when applying hyperparam_mutations. If not resampled, the value will be perturbed by a\n", + " factor of 1.2 or 0.8 if continuous, or changed to an adjacent value if discrete. Note that\n", + " ``resample_probability`` by default is 0.25, thus hyperparameter with a distribution\n", + " may go out of the specific range.\n", + "\n", + "Now we can kick off the tuning process by invoking ``Tuner.fit()``:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "c39d7c74", + "metadata": {}, + "outputs": [], + "source": [ + "if ray.is_initialized():\n", + " ray.shutdown()\n", + "ray.init()\n", + "\n", + "tuner = tune.Tuner(\n", + " train_convnet,\n", + " run_config=air.RunConfig(\n", + " name=\"pbt_test\",\n", + " # Stop when we've reached a threshold accuracy, or a maximum\n", + " # training_iteration, whichever comes first\n", + " stop={\"mean_accuracy\": 0.96, \"training_iteration\": 50},\n", + " verbose=1,\n", + " checkpoint_config=air.CheckpointConfig(\n", + " checkpoint_score_attribute=\"mean_accuracy\",\n", + " num_to_keep=4,\n", + " ),\n", + " local_dir=\"/tmp/ray_results\",\n", + " ),\n", + " tune_config=tune.TuneConfig(\n", + " scheduler=scheduler,\n", + " num_samples=4,\n", + " ),\n", + " param_space={\n", + " \"lr\": tune.uniform(0.001, 1),\n", + " \"momentum\": tune.uniform(0.001, 1),\n", + " \"checkpoint_interval\": perturbation_interval,\n", + " },\n", + ")\n", + "\n", + "results_grid = tuner.fit()" + ] + }, + { + "cell_type": "markdown", + "id": "9c4172c9", + "metadata": {}, + "source": [ + "```{note}\n", + "We recommend matching `checkpoint_interval` with `perturbation_interval` from the PBT config.\n", + "This ensures that the PBT algorithm actually exploits the trials in the most recent iteration.\n", + "\n", + "If your `perturbation_interval` is large and want to checkpoint more frequently, set `perturbation_interval` to be a multiple of `checkpoint_interval` (e.g. checkpoint every 2 steps and perturb every 4 steps).\n", + "```" + ] + }, + { + "cell_type": "markdown", + "id": "ad0aac25", + "metadata": {}, + "source": [ + "In `{LOG_DIR}/{MY_EXPERIMENT_NAME}/`, all mutations are logged in `pbt_global.txt` and individual policy perturbations are recorded in `pbt_policy_{i}.txt`. Tune logs the following information on each perturbation step: target trial tag, clone trial tag, target trial iteration, clone trial iteration, old config, new config.\n", + "\n", + "Checking the accuracy:" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "c49dc1bb", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Best result logdir: /tmp/ray_results/pbt_test/train_convnet_69158_00000_0_lr=0.0701,momentum=0.1774_2022-10-20_11-31-32\n", + "Best final iteration hyperparameter config:\n", + " {'lr': 0.07008752890101211, 'momentum': 0.17736213114751204, 'checkpoint_interval': 5}\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "

" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "import matplotlib.pyplot as plt\n", + "import os\n", + "\n", + "# Get the best trial result\n", + "best_result = results_grid.get_best_result(metric=\"mean_accuracy\", mode=\"max\")\n", + "\n", + "# Print `log_dir` where checkpoints are stored\n", + "print('Best result logdir:', best_result.log_dir)\n", + "\n", + "# Print the best trial `config` reported at the last iteration\n", + "# NOTE: This config is just what the trial ended up with at the last iteration.\n", + "# See the next section for replaying the entire history of configs.\n", + "print('Best final iteration hyperparameter config:\\n', best_result.config)\n", + "\n", + "# Plot the learning curve for the best trial\n", + "df = best_result.metrics_dataframe\n", + "# Deduplicate, since PBT might introduce duplicate data\n", + "df = df.drop_duplicates(subset=\"training_iteration\", keep=\"last\")\n", + "df.plot(\"training_iteration\", \"mean_accuracy\")\n", + "plt.xlabel(\"Training Iterations\")\n", + "plt.ylabel(\"Test Accuracy\")\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "id": "72475e5c", + "metadata": {}, + "source": [ + "(tune-advanced-tutorial-pbt-replay)=\n", + "\n", + "## Replaying a PBT run\n", + "\n", + "A run of Population Based Training ends with fully trained models. However, sometimes\n", + "you might like to train the model from scratch, but use the same hyperparameter\n", + "schedule as obtained from PBT. Ray Tune offers a replay utility for this.\n", + "\n", + "All you need to do is pass the policy log file for the trial you want to replay.\n", + "This is usually stored in the experiment directory, for instance\n", + "``~/ray_results/pbt_test/pbt_policy_ba982_00000.txt``.\n", + "\n", + "The replay utility reads the original configuration for the trial and updates it\n", + "each time when it was originally perturbed. You can (and should)\n", + "thus just use the same ``Trainable`` for the replay run. Note that the end result\n", + "will not be exactly the same, since only the hyperparameter config changes are replayed,\n", + "not the checkpoint loading from other samples." + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "7fbf4857", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "
\n", + "
\n", + "

Tune Status

\n", + " \n", + "\n", + "\n", + "\n", + "\n", + "\n", + "
Current time:2022-10-20 11:32:49
Running for: 00:00:30.39
Memory: 3.8/62.0 GiB
\n", + "
\n", + "
\n", + "
\n", + "

System Info

\n", + " PopulationBasedTraining replay: Step 39, perturb 2
Resources requested: 0/16 CPUs, 0/0 GPUs, 0.0/34.21 GiB heap, 0.0/17.1 GiB objects\n", + "
\n", + " \n", + "
\n", + "
\n", + "
\n", + "

Trial Status

\n", + " \n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "
Trial name status loc acc iter total time (s) lr
train_convnet_87836_00000TERMINATED172.31.111.100:180210.93125 100 21.09940.00720379
\n", + "
\n", + "
\n", + "\n" + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "

Trial Progress

\n", + " \n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "
Trial name date done episodes_total experiment_id hostname iterations_since_restore lr mean_accuracynode_ip pid time_since_restore time_this_iter_s time_total_s timestamp timesteps_since_restoretimesteps_total training_iteration trial_id warmup_time
train_convnet_87836_000002022-10-20_11-32-49True 2a88b6f21b54451aa81c935c77ffbce5ip-172-31-111-100 610.00720379 0.93125172.31.111.10018021 12.787 0.196162 21.0994 1666290769 0 10087836_00000 0.00894547
\n", + "
\n", + "\n" + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2022-10-20 11:32:28,900\tINFO pbt.py:1085 -- Population Based Training replay is now at step 32. Configuration will be changed to {'lr': 0.08410503468121452, 'momentum': 0.99, 'checkpoint_interval': 5}.\n", + "(train_convnet pid=17974) 2022-10-20 11:32:32,098\tINFO trainable.py:772 -- Restored on 172.31.111.100 from checkpoint: /home/ray/ray_results/train_convnet_2022-10-20_11-32-19/train_convnet_87836_00000_0_2022-10-20_11-32-19/checkpoint_tmp4ab367\n", + "(train_convnet pid=17974) 2022-10-20 11:32:32,098\tINFO trainable.py:781 -- Current state after restoring: {'_iteration': 32, '_timesteps_total': None, '_time_total': 6.83707332611084, '_episodes_total': None}\n", + "2022-10-20 11:32:33,575\tINFO pbt.py:1085 -- Population Based Training replay is now at step 39. Configuration will be changed to {'lr': 0.007203792764253441, 'momentum': 0.9, 'checkpoint_interval': 5}.\n", + "(train_convnet pid=18021) 2022-10-20 11:32:36,764\tINFO trainable.py:772 -- Restored on 172.31.111.100 from checkpoint: /home/ray/ray_results/train_convnet_2022-10-20_11-32-19/train_convnet_87836_00000_0_2022-10-20_11-32-19/checkpoint_tmpb82652\n", + "(train_convnet pid=18021) 2022-10-20 11:32:36,765\tINFO trainable.py:781 -- Current state after restoring: {'_iteration': 39, '_timesteps_total': None, '_time_total': 8.312420129776001, '_episodes_total': None}\n", + "2022-10-20 11:32:49,668\tINFO tune.py:787 -- Total run time: 30.50 seconds (30.38 seconds for the tuning loop).\n" + ] + } + ], + "source": [ + "import glob\n", + "\n", + "from ray import tune\n", + "from ray.tune.schedulers import PopulationBasedTrainingReplay\n", + "\n", + "# Get a random replay policy from the experiment we just ran\n", + "sample_pbt_trial_log = glob.glob(\n", + " os.path.expanduser(\"/tmp/ray_results/pbt_test/pbt_policy*.txt\")\n", + ")[0]\n", + "replay = PopulationBasedTrainingReplay(sample_pbt_trial_log)\n", + "\n", + "tuner = tune.Tuner(\n", + " train_convnet,\n", + " tune_config=tune.TuneConfig(scheduler=replay),\n", + " run_config=air.RunConfig(stop={\"training_iteration\": 50}),\n", + ")\n", + "results_grid = tuner.fit()" + ] + }, + { + "cell_type": "markdown", + "id": "81b983c9", + "metadata": {}, + "source": [ + "## Example: DCGAN with PBT\n", + "\n", + "Let's take a look at a more involved example: training Generative Adversarial Networks (GAN) (Goodfellow et al., 2014).\n", + "The GAN framework learns generative\n", + "models via a training paradigm consisting of two competing modules – a generator and a\n", + "discriminator. **GAN training can be remarkably brittle and unstable in the face of suboptimal\n", + "hyperparameter selection** with generators often collapsing to a single mode or diverging entirely.\n", + "\n", + "As presented in [Population Based Training (PBT)](https://www.deepmind.com/blog/population-based-training-of-neural-networks),\n", + "PBT can help with the DCGAN training. We will now walk through how to do this in Tune.\n", + "The complete code example is on [Github](https://github.com/ray-project/ray/tree/master/python/ray/tune/examples/pbt_dcgan_mnist).\n", + "\n", + "We define the Generator and Discriminator with standard Pytorch API:\n", + "\n", + "```{literalinclude} /../../python/ray/tune/examples/pbt_dcgan_mnist/common.py\n", + " :language: python\n", + " :start-after: __GANmodel_begin__\n", + " :end-before: __GANmodel_end__\n", + "```" + ] + }, + { + "cell_type": "markdown", + "id": "c38af0e5", + "metadata": {}, + "source": [ + "To train the model with PBT, we need to define a metric for the scheduler to evaluate\n", + "the model candidates. For a GAN network, inception score is arguably the most\n", + "commonly used metric. We trained a mnist classification model (LeNet) and use\n", + "it to perform inference on the generated images and evaluate the image quality.\n", + "\n", + "```{tip}\n", + "The inception score uses a trained classification model, which we save in the object\n", + "store and pass as an object reference into the `inception_score` function.\n", + "```\n", + "\n", + "```{literalinclude} /../../python/ray/tune/examples/pbt_dcgan_mnist/common.py\n", + " :language: python\n", + " :start-after: __INCEPTION_SCORE_begin__\n", + " :end-before: __INCEPTION_SCORE_end__\n", + "```" + ] + }, + { + "cell_type": "markdown", + "id": "1b7cb163", + "metadata": {}, + "source": [ + "We define a training function that includes a Generator and a Discriminator,\n", + "each with an independent learning rate and optimizer. We make sure to implement checkpointing for our training.\n", + "In particular, note that we need to set the optimizer learning rates after loading from a checkpoint, since we want to use the perturbed config passed to us in `config` rather than the exact same config as the trial we are exploiting.\n", + "\n", + "```{literalinclude} /../../python/ray/tune/examples/pbt_dcgan_mnist/pbt_dcgan_mnist_func.py\n", + " :language: python\n", + " :start-after: __Train_begin__\n", + " :end-before: __Train_end__\n", + "```" + ] + }, + { + "cell_type": "markdown", + "id": "2a97813d", + "metadata": {}, + "source": [ + "We specify inception score as the metric and start the tuning:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "41158082", + "metadata": { + "scrolled": true + }, + "outputs": [], + "source": [ + "import torch\n", + "import ray\n", + "from ray import air, tune\n", + "from ray.tune.schedulers import PopulationBasedTraining\n", + "\n", + "from ray.tune.examples.pbt_dcgan_mnist.common import Net\n", + "from ray.tune.examples.pbt_dcgan_mnist.pbt_dcgan_mnist_func import (\n", + " dcgan_train, download_mnist_cnn\n", + ")\n", + "\n", + "# Load the pretrained mnist classification model for inception_score\n", + "mnist_cnn = Net()\n", + "model_path = download_mnist_cnn()\n", + "mnist_cnn.load_state_dict(torch.load(model_path))\n", + "mnist_cnn.eval()\n", + "# Put the model in Ray object store.\n", + "mnist_model_ref = ray.put(mnist_cnn)\n", + "\n", + "perturbation_interval = 5\n", + "scheduler = PopulationBasedTraining(\n", + " perturbation_interval=perturbation_interval,\n", + " hyperparam_mutations={\n", + " # Distribution for resampling\n", + " \"netG_lr\": tune.uniform(1e-2, 1e-5),\n", + " \"netD_lr\": tune.uniform(1e-2, 1e-5),\n", + " },\n", + ")\n", + "\n", + "smoke_test = True # For testing purposes: set this to False to run the full experiment\n", + "tuner = tune.Tuner(\n", + " dcgan_train,\n", + " run_config=air.RunConfig(\n", + " name=\"pbt_dcgan_mnist_tutorial\",\n", + " stop={\"training_iteration\": 5 if smoke_test else 150},\n", + " verbose=1,\n", + " ),\n", + " tune_config=tune.TuneConfig(\n", + " metric=\"is_score\",\n", + " mode=\"max\",\n", + " num_samples=2 if smoke_test else 8,\n", + " scheduler=scheduler,\n", + " ),\n", + " param_space={\n", + " # Define how initial values of the learning rates should be chosen.\n", + " \"netG_lr\": tune.choice([0.0001, 0.0002, 0.0005]),\n", + " \"netD_lr\": tune.choice([0.0001, 0.0002, 0.0005]),\n", + " \"mnist_model_ref\": mnist_model_ref,\n", + " \"checkpoint_interval\": perturbation_interval,\n", + " },\n", + ")\n", + "results_grid = tuner.fit()" + ] + }, + { + "cell_type": "markdown", + "id": "b101f746", + "metadata": {}, + "source": [ + "The trained Generator models can be loaded from checkpoints to generate images of digits from noise signals." + ] + }, + { + "cell_type": "markdown", + "id": "52a53db7", + "metadata": {}, + "source": [ + "## Visualization\n", + "\n", + "Below, we visualize the increasing inception score from the training logs." + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "0aef4d29", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "import matplotlib.pyplot as plt\n", + "\n", + "# Uncomment to apply plotting styles\n", + "# !pip install seaborn\n", + "# import seaborn as sns\n", + "# sns.set_style(\"darkgrid\")\n", + "\n", + "result_dfs = [result.metrics_dataframe for result in results_grid]\n", + "best_result = results_grid.get_best_result(metric=\"is_score\", mode=\"max\")\n", + "\n", + "plt.figure(figsize=(7, 4))\n", + "for i, df in enumerate(result_dfs):\n", + " plt.plot(df[\"is_score\"], label=i)\n", + "plt.legend()\n", + "plt.title(\"Inception Score During Training\")\n", + "plt.xlabel(\"Training Iterations\")\n", + "plt.ylabel(\"Inception Score\")\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "id": "71129ee3", + "metadata": {}, + "source": [ + "Next, let's take a look at the Generator and Discriminator losses:" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "id": "8cb4d1b6", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAA+QAAAGJCAYAAAADqPm8AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/av/WaAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdd3gU1foH8O+Z2U0PIYReBOkqXSwgiqAiiAVQ8YLClYuNYuHaQP150YsiXNGLoIKIeMWCgoqiqAhWEBURRaSpdEIgvW2dmfP7Y3dmZ7Yk2WR3Zzd5P8/jI9lseXeyyZl3znvewzjnHIQQQgghhBBCCIkpwewACCGEEEIIIYSQhogSckIIIYQQQgghxASUkBNCCCGEEEIIISaghJwQQgghhBBCCDEBJeSEEEIIIYQQQogJKCEnhBBCCCGEEEJMQAk5IYQQQgghhBBiAkrICSGEEEIIIYQQE1BCTgghhBBCCCGEmIASckJIwpk5cyaGDh1qdhhR061bNyxatKhWjx06dChmzpwZ4YgIIQ3JokWL0K1bt5i+5rFjx9CtWze89957EXvOH374Ad26dcMPP/wQseck8S8an6V4Upffz/feew/dunXDsWPHIhwVqQuL2QGQxHT06FGsWLECW7ZsQV5eHgCgTZs2OO+883DDDTege/fuJkcYOV9//TV27tyJO++8M2avOWHCBBQXF+Ojjz6K2WvWxtChQ3H8+HEAAGMMGRkZaNWqFfr06YPrrrsOvXv3NjnCyHnvvfcwa9asau/Xpk0bfPHFFzGIiBBCquf/tyspKQlZWVno1q0bBg8ejDFjxiAjI8PECBOb3W7Hyy+/jHPPPRfnnXdeTF7zhx9+wMSJE7Fw4UIMHz48Jq9ZGw3ts6c/J6rK3LlzMWbMmBhERBIF45xzs4MgieXLL7/EjBkzIIoirrrqKnTv3h2CIODAgQPYsGEDcnNzsWnTJrRp08bsUCPi8ccfxxtvvIF9+/bF7DUTKSFv1KgRJk2aBACorKzEgQMH8OmnnyI/Px8333xzjZLYcLndbnDOkZSUFPHnDuXo0aP4+eefDbc98sgj6NWrF8aOHavdlp6ejksvvbROr+V0OiGKIiyW8K+ZulwuMMZgtVrrFAMhpH5Qk6K77roLbdu2hSRJKCgowI8//ogtW7agdevWeOGFFwwX0iVJgizLSE5OjlmcnHO4XC5YLBaIohiR51QUBW63G1arFYIQnaLQoqIiDBgwANOnT4/ZhftES8jD+exFQjQ+SzWxceNGVFZWal9/8803+OijjzBr1ixkZ2drt/fr1w/t2rWr9evU5fdTlmVIkoSkpCQwxmodA4ksmiEnYTly5Aj++c9/onXr1nj11VfRvHlzw/fvu+8+vPnmm1Eb+CLBZrMhLS3N1BjUk4RYnuxES4sWLXDNNdcYbrvvvvtw77334tVXX0X79u0xfvz4iLyW+rMzI9ls165dwAA6e/ZstGvXLuD960mSBEVRwrp4UJfPRSwvUhBCEsdFF12Enj17al/ffvvt2Lp1K+644w5MnToV69evR0pKCgDAYrHU6oJgbej/RkZ6TBQEIWHH2Xg4V4mUcD57dRHNz1JN+F+MLygowEcffYRLL70Ubdu2Dfm4cH/Wdfn9FEUxphcpSM3Eb9ZE4tLLL78Mm82GuXPnBiTjgOePxMSJE9GqVSvD7X/99RfuuusunHvuuejZsyfGjBmDTZs2Ge6jrmvZvn075s6di/PPPx99+vTBtGnTUFRUFPBaX3/9NcaPH48+ffqgb9++uO222/DHH38Y7jNz5kz07dsXR44cwa233oq+ffvivvvuAwD89NNPuOuuu3DxxRejR48eGDx4MJ588kk4HA7D49944w0AnnW96n8qm82Gp556CoMHD0aPHj1w+eWXY/ny5fAvPOnWrRsef/xxfPjhhxg5ciR69uyJb7/9tiaHvEpvvPEGRo4ciR49emDQoEF47LHHUFZWZrjPoUOHcOedd+KCCy5Az549cdFFF2HGjBkoLy/X7rNlyxaMGzcO/fv3R9++fXH55ZfjmWeeqXVcKSkpmD9/Pho3bowlS5ZoxyPUer5g672q+tn5ryFXH798+XK8/fbbuPTSS9GjRw9ce+212LlzZ0B8n3zyCa644gr07NkTV155JT7//POIrEvXx/Hqq6/i0ksvRc+ePfHXX3/B5XJh4cKFGDNmDM4++2z06dMH48ePx/fffx/wPP5ryNX1YocPH8bMmTPRv39/nH322Zg1axbsdrvhsf5ryMP5vVIUBYsWLcKgQYPQu3dvTJgwAX/++SetSyeknhowYACmTp2K48eP48MPP9RuD7ZGtSbjhNPpxKJFi3D55ZejZ8+eGDRoEKZPn44jR44AqPpvZFXjQG5uLm6//Xb07dsXF154oTYu79u3DxMnTkSfPn0wZMgQrFu3zhBPsDFnwoQJuPLKK/Hnn39iwoQJ6N27Ny688EIsW7bM8Nia/M0+duwYBgwYAABYvHixdo6g//u9detW7Vylf//+mDJlCv766y/Da6nH+88//8S9996Lc845JyIXso8ePaqde/Xu3Rtjx47FV199FXC/lStXYuTIkejduzfOOeccjBkzxnAsKyoq8MQTT2Do0KHo0aMHBgwYgEmTJuH333+vdWyhPnsTJkzAhAkTAu5f1bgfzmfp5MmTmDp1Kvr27Yvzzz8f8+bNgyzLhtcqLi7G/fffj379+qF///548MEHsXfv3oisS6/reSkQ/PdTPc/cuHEjrrzySvTo0QMjR47EN998Y7hfsDXkQ4cOxe23346ffvoJ1113HXr27IlLLrkEa9euDYh/7969uOmmm9CrVy9cdNFFeOGFF/Duu+/SuvQ6ohlyEpYvv/wS7du3D2tt8B9//IFx48ahRYsWuPXWW5GWloZPPvkE06ZNw6JFi3DZZZcZ7j9nzhw0atQI06dPx/Hjx/G///0Pjz/+OP773/9q91m7di1mzpyJQYMG4b777oPdbsdbb72F8ePH4/333zdciZQkCZMnT8bZZ5+NBx98ULsK++mnn8LhcGDcuHFo3Lgxdu7ciddffx15eXl47rnnAAA33HADTp06hS1btmD+/PmGODnnmDJlCn744Qdcd911OOOMM/Dtt99i/vz5OHnyJB566CHD/b///nt88sknuPHGG5GdnV3nkv5FixZh8eLFGDhwIMaNG4eDBw/irbfewm+//Ya33noLVqsVLpcLkydPhsvlwk033YSmTZvi5MmT+Oqrr1BWVobMzEz88ccfuP3229GtWzfcddddSEpKwuHDhwPKs8Ollm6vWbMGf/75J7p06RL2c4T62YXy0UcfobKyEjfccAMYY3j55Zdx5513YuPGjdqs+ldffYUZM2aga9euuPfee1FaWoqHH34YLVq0qNX7DOa9996D0+nE2LFjtTVzFRUVWL16Na688kpcf/31qKysxJo1a3DLLbdg9erVOOOMM6p93nvuuQdt27bFP//5T+zevRurV69GkyZNcP/991f72Jr8Xi1YsAAvv/wyhgwZggsvvBB79+7F5MmT4XQ663I4CCFx7JprrsEzzzyDzZs3G5bf6NVknJBlWZv5HDlyJCZOnIjKykps2bIF+/fvx2mnnabdN9jfSEVRgr62LMu49dZb0b9/f9x3331Yt24dHn/8caSmpuLZZ5/FVVddhWHDhmHVqlV48MEH0adPn2rLgUtLS3HLLbfgsssuw4gRI/DZZ5/h6aefRteuXTF48GAAqNHf7CZNmmD27NmYPXs2LrvsMu18Rk2WvvvuO9x6661o27Ytpk+fDofDgddffx3jxo3De++9FzBrevfdd6N9+/aYMWNGwIX9cBUUFOBvf/sb7HY7JkyYgOzsbLz//vuYMmUKnnvuOS3Wd955B3PmzMHll1+OiRMnwul0Yt++ffj1119x1VVXAQD+9a9/4bPPPsNNN92ETp06oaSkBNu3b8dff/2Fs846q9Yx1uSzV51wP0uTJ09Gr1698MADD2Dr1q145ZVX0K5dO+0CiKIomDJlCnbu3Ilx48ahY8eO2LRpEx588MFav09/dTkvrcr27duxYcMGjB8/Hunp6Vi5ciXuuusufPnll4aS+WAOHz6Mu+++G9dddx1Gjx6Nd999FzNnzsRZZ52lnb+dPHkSf//73wEAt912G9LS0rB69WqqzIsASshJjVVUVODUqVNB18eWlZVBkiTt67S0NO0PzBNPPIFWrVrh3Xff1X5px48fj3HjxuHpp58OSMgbN26MV155RVvboigKVq5cifLycmRmZqKyshJPPPEErr/+evz73//WHjd69GgMHz4cS5cuNdzucrkwfPhw3HvvvYbXue+++wwJ3g033ID27dvjmWeeQW5uLlq3bo2+ffuiQ4cO2LJlS0BZ8qZNm/D999/jnnvuwZQpUwAAN954I+666y689tpruOmmmwwnIAcPHsS6devQuXPnGhztqhUVFWHp0qUYNGgQli1bpi0R6NixozYTf+2112pXiv3XmU2fPl3795YtW+B2u7Fs2TI0adKkzrHpqX/Ejxw5UquEPNTPLpTc3Fxs2LABWVlZAIDTTz8dU6dOxebNmzFkyBAAnqSzRYsWeOutt5Ceng7Ac6V+woQJEet7kJeXh88//9xwPGVZxhdffGEYuMaOHYsRI0Zg5cqVePLJJ6t93jPOOMNwv5KSEqxZs6ZGCXl1v1cFBQXaLMPzzz+vPW7x4sW17vhOCIl/LVu2RGZmJo4ePRryPjUZJ9auXYutW7di1qxZuPnmm7Xbb7vttoDkMtjfyFCza06nE1dffTVuv/12AMBVV12FCy+8EA899BCeeeYZXHHFFQCAgQMHYsSIEVi7dm21a7lPnTqFefPmYdSoUQCA6667DkOHDsW7776rJeRZWVnV/s1OS0vD5ZdfjtmzZ6Nbt24B5wnz589HVlYW3n77bTRu3BiAp6x59OjRWLRoEebNm2e4f/fu3bFgwYIqY6+pl156CQUFBXjjjTfQv39/AMD111+Pq6++GnPnzsUll1wCQRDw1VdfoUuXLlUmfF9//TXGjh1rqJS69dZb6xxjTT571Qn3szRixAhMmzYNADBu3DiMHj0aa9as0RLyjRs3YseOHXjooYe05HPcuHFar5xIqMt5aVX++usvrF+/Xjv3PO+883DNNdfg448/xk033VTlYw8ePGj4rIwYMQKDBw/Ge++9p12MWLZsGUpLS/H+++9rkwhjxozB5ZdfHt4BIAGoZJ3UWEVFBQAEXecyYcIEDBgwQPtPLScrKSnB999/jxEjRqCiogJFRUUoKipCcXExBg0ahEOHDuHkyZOG5xo7dqyh0UT//v0hy7LWufK7775DWVkZRo4cqT1fUVERBEFA7969g25vMm7cuIDb9H/0bDYbioqK0LdvX3DOsXv37mqPxzfffANRFANKq/7xj3+Acx5QJnTOOedEJBkHPMfA7XZj4sSJhvX6119/PTIyMvD1118DgNa9dPPmzQGlzapGjRoB8FxgCHVVubbUhFff5CRcwX52oVxxxRVaMg5AG1jUwf7kyZPYv38/Ro0apcUGAOeeey66du1a6xj9DRs2LOCkVRRF7cROURSUlJRAkiT06NGjRp83APjb3/5m+Lp///4oKSnRfjerUt3v1datWyFJUkCZZHWDOCEk8aWlpVX5d7om48SGDRuQnZ0d9G+Gf/OoYH8jq3L99dcbYjn99NORmpqKESNGaLd37NgRjRo1qlFyl5aWZkiek5KS0LNnT8Nj6/o3+9SpU9izZw9Gjx6tJeOAJ+keOHCgNk7r+f+Nr4uvv/4avXr10sZBwDMm33DDDTh+/Dj+/PNPAJ7jmZeXF3R5l6pRo0b49ddfA87XIqG6z151wv0s+Z9TnH322YYE/ttvv4XVajXM2AuCgBtvvLHWMdYkDqDu56UDBw40TAR1794dGRkZNfqd6Ny5s+Gz0qRJE5x++umGx3777bfo06ePoaKvcePGWiUFqT2aISc1piYwNpst4HuPP/44KisrUVBQYJitO3LkCDjnWLhwIRYuXBj0eQsLCw3lwv5XANUTAXVt9KFDhwBAu3Lpz38LDYvFgpYtWwbcLzc3F8899xy++OILlJaWGr5XkwTn+PHjaN68ecDrderUSfu+XlUNPcKVm5sLwHMCopeUlIR27dppr92uXTtMmjQJK1aswLp169C/f38MHToUV199NTIzMwF4ktjVq1fjkUcewYIFCzBgwABcdtllGD58eJ2b86mDrD75DUeon10o/r0L1ORc/eyox00/YKnat29f48S4OqF+1u+//z5eeeUVHDx4EG63u9r7+wv1u1FaWlrt1jHV/V6FOjaNGzc2XOQghNQ/NpsNOTk5Ib9fk3HiyJEjOP3002vUbCqc8TA5OTkg4crMzETLli0DEv3MzMyAPirBBHtsVlZWwG4qdfmbrf5NPf300wO+16lTJ2zevDmgmVekzxOCLS9Uzxtyc3PRtWtX3Hrrrfjuu+9w/fXXo3379rjgggtw5ZVX4uyzz9Yec99992HmzJm4+OKLcdZZZ2Hw4MEYNWpUnTqFq6r77FWnrp+lrKwswzlgbm4umjVrhtTUVMP9gp031Fa0zkv9z4EAz/urye9EqMfq4zh+/Dj69OkTcL9IHpuGihJyUmOZmZlo1qxZQOM0ANofff8yIfVK+j/+8Q9ceOGFQZ/X/xc5VBKolryp/58/fz6aNWsWcD//7pFJSUkBzynLMiZNmqStI+vYsSPS0tJw8uRJzJw5M+IzxQAi0kG0NmbOnInRo0dj06ZN2LJlC+bMmYOlS5finXfeQcuWLZGSkoI33ngDP/zwA7766it8++23WL9+Pd5++2288sorderGqX5W2rdvDyBwlkQV6ngH+9lVJVSssd7dMdjP+oMPPsDMmTNx6aWXYvLkycjJyYEoili6dGmNy/Wq+92I1mMJIfVXXl4eysvLqzypjvQ4Ec54GOq56/L3vibxRuJvdrjM6AzeqVMnfPrpp9rPdcOGDXjzzTcxbdo03HXXXQA8F2T69++Pzz//HFu2bMHy5cuxbNkyLFq0SCvxr42afPZU/o3XVJH4LMVatM5Lo/07QaKHEnISlosvvhirV6/Gzp070atXr2rvr149tVqtGDhwYERiUJ8zJyen1s+5f/9+HDp0yLCGDPCsk/MXKols06YNtm7dioqKCsPs5IEDB7TvR4s623ngwAHDFWqXy4Vjx44FHBe18+vUqVPx888/Y9y4cXjrrbcwY8YMAJ5kTV1uMGvWLCxZsgTPPvssfvjhh1of48rKSmzcuBGtWrXSqgbUWVl9h3cgsJogWtTjpnb81Tt8+HBUX/uzzz5Du3btsHjxYsNnqiaNWmJBf2z0n6ni4uKAK/WEkPrjgw8+AAAMGjSoyvtVN06cdtpp+PXXX7V9vxNdTf9mhzpHUP+mHjx4MOB7Bw4cQHZ2dlS3NWvdunXI19bHB3jKxq+44gpcccUVcLlcuPPOO7FkyRLcfvvt2kWC5s2b48Ybb8SNN96IwsJCjB49GkuWLKlTQh7ss5eVlRX0godacRBtrVu3xg8//AC73W6YJQ923hBJ4ZyXmqVNmzZBz5WifWwaAlpDTsJyyy23IDU1FQ899BAKCgoCvu9/FS4nJwfnnnsu3n77bZw6dSrg/sG2M6vOhRdeiIyMDCxdutRQQhbOc6pXJvXxcs7x2muvBdxX/YPsX/Jz0UUXQZZlbb286tVXXwVjDBdddFH1b6aWBg4cCKvVipUrVxrew5o1a1BeXm7oEqtvtgcAXbt2hSAIcLlcADzr/P2p64PU+4TL4XDggQceQElJCe644w7thKVNmzYQRRHbtm0z3P+tt96q1euEq0WLFujatSvWrl1rWLP2448/Yv/+/VF9bfXqs/7n9euvv+KXX36J6uvW1IABA2CxWAJ+Fv6fb0JI/bF161a88MILaNu2La6++uqQ96vJODFs2DAUFxcH/ZuRiJU4Nf2bHeocoXnz5jjjjDOwdu1aw/f279+PLVu21CmRrYnBgwdj586d2LFjh3abzWbDO++8gzZt2mg9bYqLiw2PS0pKQqdOncA5h9vthizLARfRc3Jy0Lx581qfIwChP3vt2rXDgQMHDOdye/furfPOLzU1aNAguN1uvPPOO9ptiqJEfSwM57zULIMGDcIvv/yCPXv2aLeVlJQEbDdIwkcz5CQsHTp0wNNPP417770Xw4cPx1VXXYXu3buDc45jx47ho48+giAIhrUx//rXvzB+/HhcddVVGDt2LNq1a4eCggL88ssvyMvLM+w/WRMZGRmYPXs2HnjgAYwZMwZXXHEFmjRpgtzcXHz99dfo168fHn300Sqfo2PHjjjttNMwb948nDx5EhkZGfjss8+CrrNRt/SYM2cOBg0aBFEUMXLkSAwdOhTnnXcenn32WRw/fhzdunXDli1bsGnTJvz973+v85qaoqIivPDCCwG3q4PX7bffjsWLF+OWW27B0KFDcfDgQbz55pvo2bOnNrh9//33ePzxxzF8+HB06NABsizjgw8+gCiKWlfM559/Hj/99BMGDx6MNm3aoLCwEG+++SZatmxpWEMWysmTJ7Wr3DabDX/99Rc+/fRT5Ofn4x//+IehSU1mZiaGDx+O119/HYwxtGvXDl999RUKCwvrdKzCMWPGDEydOhXjxo3DmDFjUFZWhjfeeANdu3atU2OZ6lx88cXYsGEDpk2bhosvvhjHjh3DqlWr0Llz56B9GWKtadOmmDhxIl555RXccccduPDCC7Fv3z588803yM7ODjkLRAhJDN988w0OHDgAWZZRUFCAH374AVu2bEHr1q3x4osvVlkuXZNxYtSoUVi7di3mzp2LnTt34uyzz4bdbsfWrVsxbty4oDu0xLOa/s1OSUlB586d8cknn6BDhw5o3LgxunTpgq5du+KBBx7ArbfeihtuuAHXXXedtu1ZZmamYbeT2tqwYYM24603evRo3Hbbbfj4449x6623YsKECcjKysLatWtx7NgxLFq0SEsAJ0+ejKZNm6Jfv37IycnBgQMH8Prrr2Pw4MHIyMhAWVkZBg8ejMsvvxzdu3dHWloavvvuO/z222+GrutVCeezd9111+HVV1/F5MmTcd1116GwsFA77tEco1WXXnopevXqhXnz5uHIkSPo2LGjYU13tMbCcM5LzXLLLbfgww8/xKRJk3DTTTdp2561atUKJSUldJ5QB5SQk7BdeumlWLduHV555RVs2bIF7777LhhjaN26NQYPHoxx48ahe/fu2v07d+6Md999F4sXL8b777+PkpISNGnSBGeeeaa29US4rrrqKjRv3hwvvfQSli9fDpfLhRYtWqB///4YM2ZMtY+3Wq1YsmSJtp46OTkZl112GW688caAbUuGDRuGCRMm4OOPP8aHH34IzjlGjhwJQRDw4osv4rnnnsP69evx3nvvoU2bNnjggQfwj3/8o1bvS6+wsDBoI7wBAwbg6quvxp133okmTZrg9ddfx9y5c5GVlYWxY8fin//8p1Yu2K1bNwwaNAhffvklTp48idTUVHTr1g3Lli3TGnMMHToUx48fx7vvvovi4mJkZ2fj3HPPxZ133qk1fqvKnj178MADD4AxhvT0dLRq1QpDhgzB9ddfH3RZwyOPPAJJkrBq1SokJSVh+PDheOCBB3DllVfW7YDV0NChQ/HMM89g0aJFWLBgATp06IC5c+di7dq1QfsjRMqYMWNQUFCAt99+G5s3b0bnzp3xn//8B59++il+/PHHqL1uONQtV1avXo2tW7eiT58+WL58OcaPH0/7jBKS4NRSa6vVisaNG6Nr16546KGHMGbMmGqbQtZknBBFEcuWLcOLL76Ijz76CBs2bEDjxo3Rr18/bV/uRBLO3+w5c+bg3//+N+bOnQu3243p06eja9euGDhwIF5++WU899xzeO6552CxWHDOOefg/vvvj0hDtI8//jjo7eeeey769++PVatW4T//+Q9ef/11OJ1OdOvWDUuWLMHFF1+s3feGG27AunXrsGLFCthsNrRs2RITJkzA1KlTAXguOIwbNw5btmzBhg0bwDnHaaedpk221EQ4n71OnTph3rx5eO655zB37lx07twZ8+fPx0cffRSTsVLtE/DEE0/g/fffhyAIuOyyyzBt2jSMGzcuauv8wzkvNUurVq3w2muvaTE2adIEN954I1JTUzFnzhxTeiDUF4wnYh0RIYRE2DXXXIMmTZpgxYoVZocSV8rKynDOOefgnnvuwZQpU8wOhxBCCIm5jRs3Ytq0aXjzzTdrVD3YkDzxxBN4++23sWPHDmoOV0u0hpwQ0qC43e6AdfU//PAD9u7di3PPPdekqOKDw+EIuO1///sfADT4Y0MIIaRh8B8LZVnGypUrkZGRoS1jbKj8j01xcTE+/PBDnH322ZSM1wGVrBNCGpSTJ09i0qRJuPrqq9G8eXMcOHAAq1atQrNmzQzr3Rui9evX4/3338dFF12EtLQ0/Pzzz/joo48waNAgmhEghBDSIPz73/+Gw+FA37594XK5sGHDBuzYsQP//Oc/TdvCNl7ccMMNOPfcc9GpUycUFBTg3XffRUVFhbbEgdQOJeSEkAYlKysLZ511FlavXo2ioiKkpaVh8ODBuO+++5CdnW12eKbq1q0bRFHEyy+/jMrKSuTk5GDixIm45557zA6NEEIIiYnzzz8fK1aswFdffQWn04n27dvj//7v/3DTTTeZHZrpBg8ejM8++wzvvPMOGGM488wz8cQTT+Ccc84xO7SERmvICSGEEEIIIYQQE9AackIIIYQQQgghxASUkBNCCCGEEEIIISaghJwQQgghhBBCCDEBJeSEEEIIIYQQQogJGkyX9cLCctS1fR1jQE5OZkSeK9YSOXYgseOn2M2TyPFT7OaJRfzqa5DIauhjPZDY8VPs5knk+Cl28yRy/PE01jeYhJxzROxgR/K5Yi2RYwcSO36K3TyJHD/Fbp5Ej78horHeJ5Hjp9jNk8jxU+zmSeT44yF2KlknhBBCCCGEEEJMQAk5IYQQQgghhBBiAkrICSGEEEIIIYQQEzSYNeSEEELiB+cciiJDURSzQwnAGOBwOOB2u2q9rkwQBAiCCMZYZIMjhBBCEgSN9TVDCTkhhJCYkiQ3SkuL4HY7zA4lpKIioc4nEElJKWjUqAksFmuEoiKEEEISA431NUcJOSGEkJjhnKOwMA+CICArqylE0RKXs8iiyCDLtbtkzjmHLEuoqChBYWEemjdvG5fvMVa2bduG5cuXY9euXcjPz8fzzz+PSy+9VPs+5xzPPfccVq9ejbKyMvTr1w+zZ89Ghw4dzAuaEEJIrdFYHx5KyAkhhMSMJLnBuYKsrGZISkoxO5yQLBYBklSXq+bJEEURRUUnIUluWK1JEYst0dhsNnTr1g3XXnstpk+fHvD9ZcuWYeXKlXjqqafQtm1bLFy4EJMnT8b69euRnJxsQsSEEELqgsb6MOOoQwSEEEJIrTBW/3uKNoT3WBODBw/G4MGDg36Pc47XXnsNU6ZM0WbN58+fj4EDB2Ljxo0YOXJkLEMlhBASQQ1hHIzEe6SEnBBCCCGmOHbsGPLz8zFw4EDttszMTPTu3Rs7duwIOyGPREWk+hxxWF1ZI4kcP8VunkSOn2I3T6j4E/X91AVjtT8OlJCHwV5WgkJHCZDS2OxQCCGEkISXn58PAMjJyTHcnpOTg4KCgrCfLycns+5BuR3AsZ+Q07ofICTu7E5EjoVJKHbzJHL8FLt5/ON3OBwoKhIgigwWS3z/Ha1rfIrCIAgCsrPTkZJSu/J8SsjD8PmiJ1F68gSuf/J5JKc3MjscQgghhOgUFpbXevsaVfqWfyN1x1KUj1gCZ6crIxNYDDHmOTmOxLGINYrdPIkcP8VunlDxu90uKIoCWeZ1XKMdXXVfQw7IMoeiKCguroTV6jZ8Tz0+1cZRpwgaGFtpMRRZgr2slBJyQghpgN599x289dZKFBUVolOnLpgx436ceWYPs8NKWM2aNQMAFBYWonnz5trthYWF6N69e9jPxznqfFLLKvI8/y/PS8gTZFUkjoVZKHbzJHL8FLt5/ONP5PcC1G6sr8vPML5rCOKMtmg/0T9lhBBCwrZp0wYsXvwsJk26FcuXv47Onbvin/+8E8XFRWaHlrDatm2LZs2aYevWrdptFRUV+PXXX9G3b19zgtLG+vid1SGEEBIdZoz1NEMeBuZdS8YpISeEkIjhnMMR45K2FIsQ9n6hq1a9gauuGoWRI68GANx//yxs3boZH330ISZMuDkKUdYPlZWVOHLkiPb1sWPHsGfPHmRlZaF169aYOHEiXnzxRbRv317b9qx58+aGvcpjypuQM0rICSEkomI93ifKWE8JeTi8P1Cu0CBNCCGRwDnHLat+xc7cspi+bu/WjbDsb71rPFC73W7s378XEyZM0m4TBAH9+5+L33/fGa0w64Vdu3Zh4sSJ2tdz584FAIwePRpPPfUUbr31Vtjtdjz66KMoKyvD2WefjZdfftm8Pci1GXLZnNcnhJB6yIzxPlHGekrIwyBoM+SUkBNCSKQkwu4opaUlkGUZTZo0MdzepEkTHD58yJygEsR5552Hffv2hfw+Ywx333037r777hhGVQVankYIIVER7+O9WWM9JeThUGfIaZAmhJCIYIxh2d96J0TJOmkg1IQcdPGdEEIixYzxPlHGekrIw6A2daOSdUIIiRzGGFKtotlhVCkrqzFEUURRkbGpS1FRUcAe2iSxcWrqRgghURHv471ZYz11WQ+D2tSNytgIIaRhsVqt6Nq1O7Zv/1G7TVEUbN++DWed1cvEyEjEUUJOCCENklljPc2Qh4FRyTohhDRYf/vbjXjiidno3v1MnHHGWXjnnTdht9sxcuRVZodGIom6rBNCSINlxlhPCXkYGHVZJ4SQBuuSS4ahpKQYL7+8BEVFhejcuSsWLFiEJk2oZL1+8a43pIvvhBDS4Jgx1lNCHgZGXdYJIaRBu/baG3DttTeYHQaJJipZJ4SQBi3WYz2tIQ8DlawTQggh9RztQ04IISSGKCEPB3VZJ4QQQuo1zrwdgGmGnBBCSAxQQh4GX8k6zZATQggh9RKjNeSEEEJihxLyMGgby9NVc0IIIaR+oi7rhBBCYogS8jCoCblCJeuEEEJI/aSuIQeN9YQQQqKPEvIwqCXrVMZGCCGE1FPUZZ0QQkgMmZqQb9u2DXfccQcGDRqEbt26YePGjSHv++ijj6Jbt2549dVXYxegH0ZN3QghhJD6jZanEUIIiSFTE3KbzYZu3brhX//6V5X3+/zzz/Hrr7+iefPmMYosOCbQtmeEEEJI/UbVcIQQQmLHYuaLDx48GIMHD67yPidPnsS///1vLF++HLfffnutX0u94F0XjPkG6Ug8Xyyp8SZa3KpEjp9iN08ix19fY0/E91NXjAW+74Z4HBIFp33ICSGExJCpCXl1FEXB/fffj8mTJ6NLly51eq6cnMw6x5OU7Dlc6elJaNq07s9nhkgcBzMlcvwUu3kSOf76FrvD4UBRkQBRZLBY4ruNSV3jUxQGQRCQnZ2OlJSUCEVFok7w7ENOXdYJIYTEQlwn5MuWLYPFYsHEiRPr/FyFheV1rj6TJM8TlJfZUFBQXueYYokxz8lxJI6DGRI5fordPIkcf32N3e12QVEUyDKHJMVvwmOxCAHx/fLLz3jzzZXYt28PCgsL8OSTT+Oiiy4O+RyyzKEoCoqLK2G1ug3fU48RiUfU1I0QQhqqcMf6SIjbhHzXrl147bXX8N577/n2/64DziOwHIz51pAn2gmyKiLHwUSJHD/Fbp5Ejr++xZ6o7wUA7HY7OnfugpEjr8bDD99f48cl8s+wQdKautEPjRBCGprajvV1EbcJ+U8//YTCwkIMGTJEu02WZcybNw+vvfYavvjii5jHpK4hp33ICSEkgjgHJHtsX9OSGvZC7gEDLsCAARdEKSASN2jbM0IIiY5Yj/cJMtbHbUJ+zTXXYODAgYbbJk+ejGuuuQZjxowxJSbah5wQQiKMczR+bzSseT/F9GXdrc5Byej3qLsaCaQm5KCEnBBCIsaE8T5RxnpTE/LKykocOXJE+/rYsWPYs2cPsrKy0Lp1a2RnZxvub7Va0bRpU3Ts2DHWoQKAVjrP6ao5IYRETpwPlKRh4TRDTggh0UHjfVCmJuS7du0yNGybO3cuAGD06NF46qmnzAorNHUfcipZJ4SQyGDMc/U6AUrWSQPh/VwwqoYjhJDIMWO8T5Cx3tSE/LzzzsO+fftqfH8z1o3rCd6r5pwGaUIIiRzGAGua2VEQ4kX7kBNCSFTQeB9UfG8CG2/UknWFEnJCCCGkXvLuQ04l64QQQmIhbpu6xSO1qRutISeEkIbHZrPh+PGj2tcnThzHH3/sQ2ZmFlq2bGliZCSiaA05IYQ0WGaM9ZSQh4FRyTohhDRYe/fuxl133aF9vWjRswCAESOuxMMPzzYpKhJpHLQPOSGENFRmjPWUkIdB7bIOaupGCCENTr9+/bF5c2y3ZyMmoBlyQghpsMwY62kNeRioZJ0QQgip57wJOaOxnhBCSAxQQh4G3z7kVMZGCCGE1EvqDDkoISeEEBJ9lJCHQZshp5J1QgghpH6iknVCCCExRAl5GKipGyGEEFLPMWrqRgghJHYoIQ+Htg85XTUnhBBC6iPO1H3IZXMDIYQQ0iBQQh4GJtAackIIIaReo5J1QgghMUQJeRh8Jes0SBNCCCH1EnVZJ4QQEkOUkIdBnSGHQjPkhBBCSP1Ea8gJIYTEDiXkYaAZckIIIaSeo5J1QgghMUQJeRi0hJxmyAkhhJD6iRJyQgghMWQxO4BE4mvqRoM0IYQ0JCtXrsDXX3+Jw4cPITk5GT179sKUKXfitNM6mB0aiTQ1IQeN9YQQ0tCYMd7TDHkYqGSdEEIaph07fsaYMddj6dIVePbZ5yFJEmbMmA673W52aCTCuDZDTtVwhBDS0Jgx3tMMeTjUGXIqWSeEkIjhnMMhO2L6miliChhjNb7/M88sMnz90EOzcdVVl2Hfvj3o06dfpMMjZtK6rNM+5IQQEkmxHu/DHesBc8Z7SsjD4Jshp4ScEEIigXOOu76/A78X/xbT1+2R3QsLz38x7IFaVVlZAQBo1KhRJMMi8YDWkBNCSMSZMd7XdawHYjPeU8l6GNQfJldokCaEkEhhqP1AaQZFUfDccwvQs2dvdOzY2exwSKRRQk4IIVFB431wNEMeBibQujJCCIkkxhgWnv9i3Jes6z3zzDwcOPAXXnjh5QhHReICrSEnhJCIM2O8r8tYD8RuvKeEPAzaDDldNSeEkIhhjCHVkmp2GDXyzDPz8N13m7F48Uto3ryF2eGQqPCevNFYTwghEUXjfXCUkIdBnSGnknVCCGlYOOd49tn5+Oabr7Bo0VK0bt3G7JBIlHAqWSeEkAbLjPGeEvIw+GbIqYyNEEIakgUL5mHjxk8xd+4CpKWlobCwAACQkZGB5OQUk6MjEaV1WaeEnBBCGhozxntKyMPBaIacEEIaorVr1wAA7rzzdsPtDz30L1xxxVVmhFSvyLKMRYsW4cMPP0RBQQGaN2+O0aNHY+rUqXVa/1crTPT+g8Z6QghpaMwY7ykhD4NWsk4z5IQQ0qBs3vyT2SHUa8uWLcNbb72FefPmoXPnzti1axdmzZqFzMxMTJw4MbbBeC8AVFZeCNf/9sJ6bSewDGtsYyCEEGIKM8Z7SsjDQE3dCCGEkMjbsWMHLrnkElx88cUAgLZt2+Ljjz/Gzp07Yx+MtxrO7jgbvMIBJbcSYtfGsY+DEEJIg2BqQr5t2zYsX74cu3btQn5+Pp5//nlceumlAAC3243//ve/+Oabb3D06FFkZGRg4MCBuPfee9GihTmdbWkfckIIISTy+vbti3feeQcHDx7E6aefjr1792L79u2YOXNmWM8Tkep2dYtTb7d1xnlknjdG1FgTKWYVxW6eRI6fYjdPqPgT9f3UBWO1Pw6mJuQ2mw3dunXDtddei+nTpxu+53A4sHv3bkyZMgXdu3dHWVkZnnjiCUyZMgXvvfeeKfHSPuSEEEJI5N12222oqKjAiBEjIIoiZFnGjBkzcPXVV4f1PDk5mXUPxp3h/YdnzM/MSEFa0wg8b4xF5FiYhGI3TyLHT7Gbxz9+h8OBoiIBoshgsQghHhUf6hqfojAIgoDs7HSkpNSu6ZupCfngwYMxePDgoN/LzMzEihUrDLf93//9H66//nrk5uaidevWsQjRgLqsE0IIIZH3ySefYN26dViwYAE6d+6MPXv2YO7cuVpzt5oqLCyv8zVzS6kDjQGAe07SysvssBWU1+1JY4gxz8lxJI5FrFHs5knk+Cl284SK3+12QVEUyDKHJMVvZbHFItQ5PlnmUBQFxcWVsFrdhu+px6faOOoUQYxVVFSAMYZGjRqF/dhIlE4IWlM3JeFKMeprSUwioNjNk8jx19fYE/H91FVdytgaivnz5+O2227DyJEjAQDdunVDbm4uli5dGlZCznndi9gqyytwtKQFmqWpy9QSszAuEsfCLBS7eRI5fordPP7xJ/J7qa26/AwTJiF3Op14+umnMXLkSGRkZFT/AD+RKAU5kZkGALBaBDRNwPI1oP6VxCQSit08iRx/fYudythIMA6HI2B7M1EUTalI++Xrb7H3RFdc28ECC0PDPLMkhBASMwmRkLvdbtx9993gnOOxxx6r1XNEohSk0uYEALicbhQkUPkaUH9LYhIBxW6eRI6/vsZOZWweNS1jayiGDBmCJUuWoHXr1lrJ+ooVK3DttdfGPBaX0wUA4Fzw9HVLsN8/QgghiSXuE3K324177rkHubm5+N///ler2XEgMqUgzLsViqIoCXeCrKpvJTGJhGI3TyLHX99iT9T3UheJ/DOMlUceeQQLFy7EY489hsLCQjRv3hw33HADpk2bFvNY1LFe7bJOCTkhhJBoiuuEXE3GDx8+jNdeew3Z2dnmBkRN3QghhJCIy8jIwMMPP4yHH37Y7FB8O6poCTmN+YQQQqLH1IS8srISR44c0b4+duwY9uzZg6ysLDRr1gx33XUXdu/ejaVLl0KWZeTn5wMAsrKykJSUFPN4tavmNDgTQggh9RITvPuP0ww5IYSQGDA1Id+1axcmTpyofT137lwAwOjRozF9+nR88cUXAIBrrrnG8LjXXnsN5513XuwC9VKvmnMlftc9EkIIibz331+DtWvX4MSJEwCA00/viJtvvgUDBlxgcmQk0pggev9FF+EJIaShMWO8NzUhP++887Bv376Q36/qe2bw7UNOCTkhhDQkzZo1xx13TEfbtqeBc45PPvkIs2bdi1deeQMdO3YyOzwSSdoacnWrU/NCIYQQEltmjPdxvYY83jBaQ04IIRHHOQccjti+aEpKwDZbVRk06CLD17ffPg1r176L3bt/o4S8nhFoDTkhhERFzMf7MMd6wJzxnhLyMFDJOiGERBbnHKVTb4W0a2dMX9fSszeynn8p7IEaAGRZxpdfboTDYcdZZ/WKQnTEVH4z5KAhnxBC6syM8b4uYz0Qu/GeEvJw0Aw5IYREXi0Hylj7668/cccdk+ByuZCamoonn/wPTj+9o9lhkQhjorqGnGbICSEkomi8D4oS8jCoXdZphpwQQiKDMYas51+K+5J1ADjttPZYseJNVFRU4KuvNuGJJ2Zj0aKXKCmvZ9SxnrqsE0JI5Jgy3tdirAdiP95TQh4GrWSdrpYTQkjEMMaA1FSzw6iW1WpF27btAADdu5+BPXt2Y/Xqt/DAA+bvnU0ix7cPOXVZJ4SQSKLxPjih+rsQlXaFhbqsE0JIg8e5ArfbbXYYJMLUGXKttJLycUIIadCiPd7TDHkYqKkbIYQ0TEuWLMb55w9EixYtYbPZ8Pnnn2LHju145plFZodGIkzdh5zRDDkhhDQ4Zoz3lJCHgbY9I4SQhqm4uAhz5vwLhYUFSE/PQKdOXfDMM4twzjnnmx0aiTAWsO2ZaaEQQgiJMTPGe0rIw6A1daOEnBBCGpRZsx41OwQSK4LaZV2dITctEkIIITFmxnhPa8jDQCXrhBBCSP3GBM/MOJWsE0IIiQVKyMNAJeuEEEJI/cYEtXjQW7JO1+AJIYREESXk4dASchqdCSGEkPpIrYbTuq3TRXhCCCFRRAl5GKhknRBCSEPgcDhgt9u1r48fP45XX30VmzdvNjGq2GDaGnIvyscJIYREESXkYfDtQ06jMyGEkPpr6tSpWLt2LQCgrKwMY8eOxYoVKzB16lS8+eab5gYXZYwJvvXjAI35hBBCoooS8jBQl3VCCCENwe+//47+/fsDAD777DPk5OTgyy+/xLx587By5UqTo4suJgi+C/CgfJwQQkh0UUIeBipZJ4QQ0hA4HA6kp6cDADZv3oxhw4ZBEAT06dMHubm5JkcXZYIABl9CThk5IYSQaKKEPAyMmroRQghpAE477TRs3LgRJ06cwObNm3HBBRcAAAoLC5GRkWFydNElBJSsmxcLIYSQ+o8S8nDQtmeEEEIagGnTpmH+/PkYOnQoevfujb59+wIAtmzZgjPOOMPk6KKMGebHaYacEEJIVFmqvwtRCVSyTgghpAEYPnw4zj77bOTn56N79+7a7QMGDMCll15qYmTR51lDTjPkhBBCYoNmyMNBTd0IIYQAWLnyVQwa1B8LFy4wO5SoadasGc4880wIgoCKigps3LgR6enp6NSpk9mhRRVjjNaQE0IIidlYTwl5GKipGyGEkD17fseHH76HTp26mB1K1Nx99914/fXXAXgavF177bW45557cPXVV+Ozzz4zObro8oz1+oTctFAIIYSYJJZjPSXkYaB9yAkhJPI455Bcckz/q22lk81mw2OP/R8eeOBhZGZmRvhIxI+ffvpJ2/bs888/B+cc27Ztw8MPP4wXX3zR5OiiizFmLFmna/CEEBIRsR7vE2WspzXkYaAu64QQElmcc3zx8l4UHqmI6es2PS0DQ27pbthvuiaeeWYeBg68AOeccx7+97/lUYrOfOXl5cjKygIAfPvttxg2bBhSU1Nx8cUX4z//+Y/J0UUXo23PCCEk4swY7xNlrKcZ8jBQybp5FFk2OwRCSJSEN0yaZ+PGz7B//17cfvt0s0OJulatWmHHjh2w2Wz49ttvtW3PysrKkJSUZHJ00cWYYDx5o3ycEEIiIhHGezPG+rBnyE+cOAHGGFq2bAkA2LlzJ9atW4fOnTvjhhtuiHiA8YTRtmemsJUU4cMnHsCZFw1B76tvNDscQkgEMcYw5JbukN2xvdApWoWwrpifPJmHhQsX4Nlnn0dycnIUI4sPEydOxP3334+0tDS0bt0a5513HgBg27Zt6Nq1q8nRRRfNkBNCSOSZMd4nylgfdkJ+7733YuzYsRg1ahTy8/MxadIkdOnSBevWrUN+fj6mT6+/MweMuqybojj3CFz2Shzbswu9rzY7GkJIpDHGYEkSzQ6jSvv27UVxcREmT75Ju02WZfz66w689947+OKL7yCK8f0ewnHjjTeiV69eyMvLw8CBA7VtP9u1a4d77rnH3OCiLDAhNy8WQgipT+J9vDdrrA87If/jjz/Qq1cvAMAnn3yCLl26YNWqVdi8eTP+9a9/hZWQb9u2DcuXL8euXbuQn5+P559/3rC/Kecczz33HFavXo2ysjL069cPs2fPRocOHcINOyLUknVwDs552OsRSO1w2XMljZYKEELM0r//OXjttVWG25588nG0b98eN97493qVjKt69uyJnj17guvGvIsvvtjssKIuoKkbXYQnhJAGwayxPuyEXJIkbf3Yd999h6FDhwIAOnbsiPz8/LCey2azoVu3brj22muDJvLLli3DypUr8dRTT6Ft27ZYuHAhJk+ejPXr15tTMqhLwCkhjx1F8awfp3XkhBCzpKWlo2PHzobbUlJS0KhR44Db64u1a9di+fLlOHToEACgQ4cOmDx5MkaNGmVqXNFG254RQkjDZNZYH3ZC3rlzZ6xatQoXX3wxvvvuO6107dSpU2jcuHFYzzV48GAMHjw46Pc453jttdcwZcoUbdZ8/vz5GDhwIDZu3IiRI0eGG3qdMUE/QCugnnixoc6Mq4k5IYSQ6FqxYgUWLlyIG2+8URvnt2/fjtmzZ6OkpAQ333yzqfFFE2MMAq0hJ4QQEiNhJ+T33Xcfpk+fjuXLl2PUqFHo3r07AOCLL77QStkj4dixY8jPz8fAgQO12zIzM9G7d2/s2LEj7IQ8EpPZ6ho6AABXIvKcsaLGmkgxq7g3EedKYh1zVSIf+0SOHUjs+Otr7In4fkJZvPilGt2PscD3He/HYeXKlZg9e7ZhNvySSy5Bly5dsGjRovqdkAvGJkCUjxNCSMNV07G+LsJOyM877zx8//33qKio0PYoBYCxY8ciNTU1YoGp5e85OTmG23NyclBQUBD28+Xk1H1Td5fDd7iaNEmHNTmlzs8Za5E4DrF2Kt2zREKR5YSMX0WxmyeR469vsTscDhQVCRBFBoslvquM6hqfojAIgoDs7HSkpCTWeJGfn4++ffsG3N63b9+wl6clGk8yrrtiolBGTgghJHrCTsgdDgc451oyfvz4cXz++efo1KkTLrzwwogHGCmFheV1vsotu53avwvyy2FNcRu+L1TkQUlrBgjx19yHMc/JcSSOQ6yVlVQC8CTkNYlfLPoDQmUe3O3i4/OYyMc+kWMHEjv++hq72+2CoiiQZQ5Jit9GjRaLUOf4ZJlDURQUF1fCajWOF+oxilft27fHJ598gjvuuMNw+/r1601rrBorjAlgtCSNEEJIjISdkE+dOhWXXXYZxo0bh7KyMowdOxYWiwXFxcWYOXMmxo8fH5HAmjVrBgAoLCxE8+bNtdsLCwu1MvlwcB6JsjPfAK0oiuH5LCd3IHvNVbD3+DsqBj9R1xeKmsgch9hSdCXrNYk/85NbIRb/haK//wglo1UMIqyZRDz2qkSOHUjs+Otb7In6XuoiEX+Gd955J2bMmIFt27ahX79+AICff/4Z33//Pf773/+aG1y0+ZWs0ww5IYSQaAr7EvDvv/+O/v37AwA+++wz5OTk4Msvv8S8efOwcuXKiAXWtm1bNGvWDFu3btVuq6iowK+//hq0jC4W9E3dODfOnIglf3n+X/xnTGNqCNTu6jVt6ibYCsDAwRzF0QyLEELqrcsvvxzvvPMOsrOzsWnTJmzatAnZ2dlYvXo1LrvsMrPDiyqB0T7khBBCYqdWJevp6ekAgM2bN2PYsGEQBAF9+vRBbm5uWM9VWVmJI0eOaF8fO3YMe/bsQVZWFlq3bo2JEyfixRdfRPv27bVtz5o3b27YqzyW9PuSBuyJrSaLnDqBR5rWZV2uYfmo92IJo58FIYTUWo8ePfD0008bbissLMSSJUsCStnrFYFRQk4IISRmwp4hP+2007Bx40acOHECmzdvxgUXXADAM0hnZGSE9Vy7du3CqFGjtC6uc+fOxahRo/Dcc88BAG699VbcdNNNePTRR3HdddfBZrPh5ZdfNmcPciBgH3LDtxTJ839KAiNOX7JeE+rPAur/CSGERER+fj4WLlxodhhRxZhfyXqirTcghBCSUMKeIZ82bRruu+8+zJ07F+eff75WPr5lyxacccYZYT3Xeeedh3379oX8PmMMd999N+6+++5ww4wKpu5dE3RhpDcRp72yI47L6gx5DY+t+rPg8dswihBCSHxijBmbulE+TgghJIrCTsiHDx+Os88+G/n5+YbmagMGDDCtlDyWBEGAIstBSta9s7E0Qx5x6gx5TdeQ08URQgghtcUE/zXklJETQgiJnrATcsDTAb1Zs2bIy8sDALRs2RK9evWKaGDxSi1jCyxZp1nZaOHaseWeCyGsmpUW3vszTiXrhBBCwsOYYFiiRjPkhBBCoinshFxRFLzwwgtYsWIFbDYbACA9PR2TJk3ClClTIAj1e+9OtbHb0V2F6H5hju8b6hpyWrcccfpmbpwrhuZ6AbgCpp491XDNOSGEEI+5c+dW+f2ioqIYRWIeJviXrFNGTgghJHrCTsifffZZrFmzBvfee6+2N+n27duxePFiuFwuzJgxI+JBxhPmveDw62eH0fn8zrBYvYM2rVuOGq4rPeeKUvUEub5MnZYPEEIiZPnypVixYpnhttNOa48333zXpIiiY/fu3dXeR936tL5iTNAXrNMMOSGENCBmjPdhJ+Tvv/8+5syZg0suuUS7rXv37mjRogUee+yxep+Qq2VsXOGQ3YqWkDNaQx41im6mW5EVCFV9anXHn6oVCCGRdPrpHfHf/76gfS2KtVr1FddWrlxpdgimY4JgrMRSKCMnhJCGJNbjfdjPXlpaio4dOwbc3rFjR5SWlkYkqHim3wqF6wdp2morarisnyGv5oKHYYacqhUISQScc0guZ0xf05KUbNzaqgZE0YKcnKZRioicPHkS//nPf/Dtt9/Cbrejffv2ePLJJ9GzZ8+YxuHpsk5ryAkhJNJiPd7XZqwHYj/eh52Qd+/eHW+88QYeeeQRw+1vvPGGoet6fcW0NfIKFH1CztVGYpQERpo+CVeqWRdu2AeemroREvc45/j02dnIP7A/pq/brGM3DJ/xr7AG6mPHjuCaa4YjKSkZPXr0xO23T0fLli2jGGXDUVpainHjxuG8887DsmXLkJ2djcOHDyMrKyvmsVCXdUIIiTwzxvvajPVA7Mf7sBPy+++/H7fffju+++479OnTBwDwyy+/4MSJE1i2bFnVD64HfD9QDi77BmmtPJq22oo4fRJe7Qy5PiGnnwUhCSL8q9exduaZPfDQQ7Nx2mntUVhYgBUrlmHatFuwcuXbSEtLNzu8hLds2TK0bNnS0FSuXbt2psTCGDOUrFM+TgghkULjfTBhJ+TnnnsuPv30U7z55ps4cOAAAOCyyy7D+PHj0aJFi4gHGH/UQZr7zZB7k0ZaQx5x+pL16mbI9Uk4VSsQEv8YYxg+419xX7I+YMAF2r87d+6CM8/sgeuuuxJffPE5rrxyVBQibFi++OILDBo0CHfddRe2bduGFi1aYPz48Rg7dmxYz1OLysQAgigAfjPkkXjeWFFjTaSYVRS7eRI5fordPKHiD/Z+zBjva1OyXtvxnrGaHYegcYYVoVeLFi0Cmrfl5eXh//7v//Dvf/+7Nk+ZMLSSdW6cIfdte0YJeaQZZ8jDKFmn9fyEJATGGKzJKWaHEZbMzEy0a9cex44dMzuUiJMkCUuWLMF1110Xs5L8o0eP4q233sKkSZNwxx134LfffsOcOXNgtVoxevToGj9PTk5mnWOxSBXQ91kXmYCmTev+vLEWiWNhFordPIkcP8VuHv/4HQ4HiooEiCKDxWLcHslqTYtlaNXyj89fdnYWTjutPXJzjwe9r6IwCIKA7Ox0pKTU7lwmYi3jSkpKsGbNmvqfkOtK1vUz5IxTl/VoMW57Rk3dCCHms9lsOH78GC6//AqzQ4k4i8WC5cuXY9SoUTF7Tc45evTogX/+858AgDPPPBN//PEHVq1aFVZCXlhYXucS84oyu2FGRZYVFBSU1+1JY4gxz8lxJI5FrFHs5knk+Cl284SK3+12QVEUyDKHJMXv+bjFIlQbn81mw7FjxzBs2BVB7yvLHIqioLi4Elar2/A99fhUG0d4YRPoE3LDDLm6Dzkl5JGm6Ju6ydX8UhuautHPghASGYsX/xcXXHAhWrZshYKCfCxfvhSiKODSSy83O7SoOP/887Ft2za0bds2Jq/XrFkzdOrUyXBbx44d8dlnn4X1PJxHYs03g3/JeiKeKEfmWJiDYjdPIsdPsZvHP/5Efi+1He/r8jOkhDxMvs6rPPi2ZzQrG3Fcl4Tz6o6vrkyd9iEnhERKfv5JzJ79MMrKStG4cTZ69eqNpUtfRXZ2ttmhRcVFF12EBQsWYP/+/TjrrLOQmppq+P4ll1wS0dfr168fDh48aLjt0KFDaNOmTURfpyY8XdZ1ZYkJfGJJCCEkPGaM95SQh4sFb+qmrl2mJDDy9DPk+gZvwRgaudHFEUJIhDz22Nzq71SPPPbYYwCAFStWBHyPMYY9e/ZE9PX+/ve/Y9y4cViyZAlGjBiBnTt34p133sHjjz8e0depCU+Xdd0MuUIZOSGENBRmjPc1TsinT59e5ffLysrqHEwiCLXtGc2QR4++kZsSzrZnVLJOCCG1snfv3pi+Xq9evbB48WI888wzeP7559G2bVs89NBDuPrqq2MaBxBkH3JCCCEkimqckGdmVr0gPTMz05TSstjzDtLcb9szWkMeNYqsb+pGJeuEEFIfDRkyBEOGDDE7DM8MOWiGnBBCSGzUOCGfO7dhleuF5hmkud8acq3LOm17FnHGLuvVbXtGJeuEEBIJP/74I1555RX89ddfAIBOnTrhlltuQf/+/U2OLLqYIBhL1ikfJ4QQEkVVb7xGAjD9GvIgXdYZOCWCEabfh1ypZg25Ye9xmiEnhJBa+eCDDzBp0iSkpKRgwoQJmDBhAlJSUnDzzTdj3bp1ZocXVYz5N3WjjJwQQkj0UFO3sIXa9kyfCMqA6BnMma0A1tzv4Tp9GCAmxS7MekTfyK3aLuuc9iEnhJC6WrJkCe6//37cfPPN2m0TJ07EihUr8MILL+Cqq64yL7go81x4pxlyQgghsUEz5OFiwbc900rWAUNSmP7DPGR9dgeSD3waowDrH/0MOa9uH3JdeTuj9fyEEFIrR48eDbqee+jQoTh27JgJEcUOE4xd1jnn+L34NxMjIoQQUp9RQh42dZBWgjd18/u3YCvwPMpeGIPY6if9GvLquqwbknAqWSeEkFpp1aoVtm7dGnD7d999h1atWpkQUexwm83Q1I0rCu7cejtybcdNjIoQQkh9FVbJutvtxqOPPoqpU6eiXbt20Yopzqld1hGyZJ1xWatwY4pbu43UDg+nyzo1dSOEkDqbNGkS5syZgz179qBv374AgJ9//hnvv/8+Hn74YZOjiy77ipfB4FtiJnjnLoochWid1hB2kyGEEBJLYSXkVqsVGzZswNSpU6MVTwLwlazLsgzOORhjUOTgJeuQ3d7bKDmsLSWcfcgN257RRRBCCKmN8ePHo1mzZnjllVfw6aeeJVcdO3bEs88+i0svvdTk6KKLl5aCsRaG2xhnkEFjCiGEkMgLu6nbpZdeik2bNhkavTQs3m3PmIxvfl2LvSeb46qrxqDC4USKehdd8q3OkFP5dO2Ft+2ZvqkbHXNCCKmtyy67DJdddpnZYcSeKIK5meEmBga5gVzk5TYbyp+YjaTBQ5EybLjZ4RBCSL0XdkLevn17PP/88/j5559x1llnITU11fD9iRMnRiy4+OQZpGWLGw5XJXJzjwIwJo1MkXxNWRWaIa8rJZySdcNafjrmhJDIyc8/hRdfXITvv/8ODocDbdu2xUMP/Qvdu59pdmgRd8kll2DNmjXIzs423F5WVobRo0dj06ZNJkUWfYJoMawhBwCRC5AbyNIz92+/wvXNV1BOnaSEnBDSIMV6vA87IV+zZg0yMzOxa9cu7Nq1y/A9xliDSci54Em5JUkC5xyCUnXJOq0hrz0ljKZuxm3P6JgTQiKjrKwMU6ZMRr9+/fH00wvRuHE2jh07iszMRmaHFhXHjx83LBdSuVwunDx50oSIYkgUAhJyBtZgEnLu9py3cImqzAghDY8Z433YCfkXX3wRjTiCkmUZixYtwocffoiCggI0b94co0ePxtSpUw1bksSWmpDrk0QFIg8+M0sl63Wn3+qs+hlyfXM9OuaEJALOOSDFuKLFIoQ1jrzxxv/QvHkLPPTQv7TbWreufw2+9DPf3377LTIzM7WvFUXB1q1b0aZN/XvfBqII/0+GAAFKQ6l0Uy98U5UZISTCYj7ehznWA+aM92En5Hqce2aJo5UcL1u2DG+99RbmzZuHzp07Y9euXZg1axYyMzNNnIlX15D7PkyyLCNJlwhKstu3nxyVrNeZfla8un3Imf4408kEIXGPcw73qj/Bcytj+rqsdTqsf+tc4/Fry5ZvcO655+ORRx7EL7/8jGbNmmH06Otx9dWjoxxpbE2bNg2AZ1yfOXOm4XsWiwVt2rQJuL3eEUSts7p2E284M+RQl4nJDeT9EkJiwozxPtyxHjBnvK9VQr527VosX74chw4dAgB06NABkydPxqhRoyIYGrBjxw5ccskluPjiiwEAbdu2xccff4ydO3dG9HXCo86Q6xNyyVCSXmF3opF32R3zJurU8bv2wtmHHNTUjRASBbm5x7F27bu44YYbMXHiJOzZsxv//e/TsFqtGDHiSrPDi5i9e/cCAIYOHYo1a9agSZMmJkcUe0wUA0rWhQZUsg7vhW9OCTkhpAEyY7wPOyFfsWIFFi5ciBtvvBH33HMPAGD79u2YPXs2SkpKItp9vW/fvnjnnXdw8OBBnH766di7dy+2b99eq6vzkZjEZwzwFgVUmZDbnC5kqa8nu9R/RCSG2lJf28wYastQps6VKt+D/ufAqrlvrCTysU/k2IHEjr++xu5/G2MM1r91jvuSdUVR0L37mbj9ds8Mcteu3XHw4F9Yu/bdagdoxoK977AjjqlYLk+LO0LgZ4NxAUoDSci1MZeqzAghEWTKeF+LkvW6jPe1FXZCvnLlSsyePdswG37JJZegS5cuWLRoUUQT8ttuuw0VFRUYMWIERFGELMuYMWMGrr766rCfKycns/o71Yjnh6owrY86GjVKgajbn1RIEtC0qff1vLO0acki0ppGKobai9xxiB19Qp6aavUd22DSrdo/U5IEpMTBMVcl4rFXJXLsQGLHX99idzgcKCoSIIoMFouuLNgqxjCymtHH17RpU5x+ekfDbaef3hFff/2F8X3oKAqDIAjIzk5HSkpK0PvEM5vNhm3btiE3Nxdub6MvVb1u4NrQZ8gVKlknhEQHYywux3u9nJym6NDhdMNt7dufjq++it6F6rAT8vz8fPTt2zfg9r59+yI/Pz8iQak++eQTrFu3DgsWLEDnzp2xZ88ezJ07V2vuFo7CwnJtdru2PBdYAkvWCwrK0Fq3hvxkfilaFJQDAJp415PbbXZUem8zA2Oek+NIHIdY0297VlFmQ0EVxzG5rAJqCuCw21Fh4jFXJfKxT+TYgcSOv77G7na7oCgKZJlDivWseBgsFsEQX48evXH48CHDbYcPH0KLFq1Cvg9Z5lAUBcXFlbBajQmteozi1e7du3HbbbfBbrfDbrcjKysLxcXFSE1NRZMmTRpgQt5wtj1TS9ZphpwQ0hD17NkbR44cNtx29OhhtGzZKmqvGfyyfhXat2+PTz75JOD29evXo0OHDpGISTN//nzcdtttGDlyJLp164ZRo0bh73//O5YuXRr2c3Eeqf+M254Bnq3PBN1AXelwa/dnasm6Ikcwhtr9F9njELv/jNueKVW/R8XY1M3s2BP92Cd67Ikef32NPRHdcMN4/P77b3jttVdw7NhRbNjwKT788H2MGXN9tY9NxOMwd+5cDBkyBNu2bUNycjLeeecdfPnllzjrrLPw4IMPmh1eVAVbQ844azhd1r0XwWkNOSGkIarLeF9bYc+Q33nnnZgxYwa2bduGfv36AQB+/vlnfP/99/jvf/8b0eAcDkdA3b8oiuBmnsmoJ5qCscu6ISF3On33V2fOqalbrelPCngYTd1o73dCSKScccZZePLJp7F06WK8+urLaNWqNe66614MGzbC7NCiYs+ePXjssccgCAJEUYTL5UK7du1w//3348EHH8SwYcPMDjF6BCGwy3oDmiHnVLJOCGnAzBjvw07IL7/8crzzzjt49dVXtf1KO3bsiNWrV+PMM8+MaHBDhgzBkiVL0Lp1a61kfcWKFbj22msj+jrh4Ag+Q25s6qZudcZ9+5A3kIE80riiGC7AKNVte6bf7532fieERNAFF1yICy640OwwYsJisUAQPElpTk4OcnNz0alTJ2RkZCAvL8/k6KIs1BryhnJhnZq6EUIauFiP97Xa9qxHjx54+umnIx1LgEceeQQLFy7EY489hsLCQjRv3hw33HCDtk+qKYKUrMuyDEHX1M3hVMvUdQkhJeS1wv1KBP2/DnyAEvzfhBBCauzMM8/Eb7/9hg4dOuCcc87Bc889h+LiYnzwwQfo0qWL2eFFFRPEgC74nn3IG8hFXi0hp/MWQgiJhbDXkJ9xxhkoLCwMuL24uBhnnHFGRIJSZWRk4OGHH8aXX36JnTt3YuPGjZgxYwaSkpIi+jrh0NZGGhJyCaIu4bZrCbmviQ+j5LBW/GfEq13TRjPkhBBSZzNmzECzZs20fzdq1AizZ89GcXEx/v3vf5scXZSJQuAacgiodLlDPKCekTxjJ60hJ4SQ2Ah7hjzU+m2XywWr1Rr0e/UJ5wwcANddypAlybDtmd3lSci1hm4AJYe15L9mXKmuhE534YMughBCSO307NlT+3dOTg6WL19uYjQxJopgMI49AmcosDlMCii2aB9yQgiJrRon5K+99hoAz/5xq1evRlpamvY9RVGwbds2dOzYMfIRxhvOACZAf/FckozJtkO9im4oWaeBrTYUv4ScV3OCwGiZACGEkLoQRDBmHGvUpm6cc3z3+lIkpaXjnGsnmBRglKnjLM2QE0JITNQ4IX/11VcBeGbIV61apTV7AQCr1Yq2bdvisccei3iA8YZzgIvGDe1lyWX42qnOkCv6GXIa2GqD+5Ws+yfoQR6guzNVJRBCSG0UFBRg3rx52Lp1K4qKigKq4/bs2WNSZNHn2fbMOH4I3NPUzVlRjr9++BoAcPboG7VzIaXQAemDgxDPawHxrCYxjzmiqMs6IYTEVI0T8i+++AIAMGHCBCxevBhZWVlRCyqecQ5wwT8hNw7cTnfgDDlrKM1gIixwhry6hJxK1gkhpK5mzpyJEydOYOrUqWjevLnZ4cSWIID5tdhhECBxGYrsG8sVSYLg7WnDj1WAFzuh/FGS+Am5YatRBUwIu90QIYSQMIS9hnzlypXRiCNhcM7AReNhk2W/GXK3GwrnEGVdAxhai1Ur/iXqVLJOCCHRt337drz55psRb9aaECyiX0s3QPRue6a/SOxJzr1NZhVvBUE9GOoNlWmKAlBCTgghUVWrbc/y8vKwadMmnDhxAm63sevorFmzIhJYvOKcAQEz5MZjwDhHpVNGY33JOiWHtaL4lcxVtw85lawTQkjdtWrVKmQT1/rOs+2Z3ww5Z5415LoxSD/2azuwKPXgmOkr0WQZsNTqVJEQQkgNhf1XduvWrZgyZQratWuHAwcOoEuXLjh+/Dg45zjzzDOjEWN8UQBu9UvI3cbEzwIZpQ43sg0l65SQ14Z/iXq1Jev671PJOiGE1MpDDz2EBQsW4LHHHkPbtm3NDie2RDFg2zO1qZthhly/XE2bIa8PCblu7KR15IQQEnVhJ+QLFizAP/7xD9x1113o27cvFi1ahCZNmuC+++7DhRdeGI0Y4wZXuGfLM/+mbrJxhlyAgjKHBICautWV/zZn1ZWs62fIGR1zQkiEXHfdVcjLOxFw++jR1+Peex80IaLomjFjBux2Oy677DKkpKQEbGv6448/mhRZDAiB+5ALasm67F+yrn5RjxJy/zXkJoZCCCGxZsZ4H3ZC/tdff+GZZ57xPNhigcPhQHp6Ou6++25MnToV48ePj3iQ8UJROAAhsKmbbJwhF6Gg3CGBJdF65rri/iXr1a4h18+Q0zEnhETGsmWvGWZHDxz4CzNmTMOQIZeYGFX0PPTQQ2aHYJrCsiQoYrLhNsY9Td30VVqGhq5qHl4PEnLDuEsXtgkhDYwZ433YCXlaWpq2brxZs2Y4cuQIunTpAgAoLi6ObHRxRpE5gMCmbv77kFuYp2Qdom7mnJLDWgm/y3r0E3K3w47v3liK9n3PR4d+50flNQgh8SU7O9vw9euv/w9t2rRF375nmxRRdI0ePdrsEExzKC8ZbZjdcJsABoUrxhlyKlknhJB6x4zxPuyEvHfv3ti+fTs6deqEwYMHY968edi/fz8+//xz9O7dOxoxxo2ioiJIyb6mbgK3QmHugMZjArinZD3Vl5BT+XTthNtlPRZN3U7s/x2Hd/yAyuJCSsgJiQDOecCFzWizWCxgrHbFuG63Gxs2rMcNN9xY6+eIRxUVFcjIyND+XRX1fvUTC2jqJnABMpcMs8fBS9ZjEV+UGRLy+vCGCCHxItbjfV3GeiB2433YCfmsWbNQWVkJALjzzjtRWVmJ9evXo0OHDpg5c2bEA4wnGzd9jIrWDIIzHQAg8iQozA0pSMl6mUMCU2iGvK7qUrIerX3IJYdn5kRyOaPy/IQ0JJxzvPfeKuTl5cb0dVu1ao3Ro/9WqwH2m2++QkVFBa644qooRGaec845B5s3b0ZOTg769+8f9NhwzsEYw549e0yIMDYEkQVdQ+7msmEMUuppybpxDTmduxBCIsOM8b4uYz0Qu/E+rIRclmXk5eWhW7duADzl648//nhUAotH6WkZKC4phJKSBgAQeBKASuM6MniaupU63IB+f/IqkkPOOfZ9swHZbU5Di84NcM/XKgQ0dauufI5Hf+2b2+nwPH2MZ/QIqa8SbZL5448/wHnnDUTTps3MDiWi/ve//yErKwsA8Nprr5kcjXkEgQWcvDFvybpxDbnuons9Klk3jLNUsk4IiSAa74MLKyEXRRH/+Mc/sH79ejRq1ChaMcWtdo2b4VjuYe1rQUkCRED2S/wskFHokMD0JdNVJIclJ47hx9WvIqtFa1zzfwsiHnciC9j2rLpZb/0x59FJmNWZcdntruaehJDqMMYwevTfEqZkPS/vBH766Uc88cT8KERlrnPPPTfovxsaJgSZIefebc+qKVmvF3u36y+EV7dMjBBCasiM8b4uJeuxHO/DLlnv0qULjh07hnbt2kUjnriW9dYaoFcHqOO0qCQBAGTJmDSKUOCQFEBXss6qSA7ddhsAwOX9P/HxX5/v/3UAXcLOonQioSXkkquaexJCaoIxFrCtVrz6+OMPkZ2djQEDBpkdStQ5nU7s27cPhYWFAdVKl1xSP7vLAwALUbJe5T7kah4u14eEnGbICSHRQeN9cGEn5Pfccw/mzZuHu+++G2eddRbS0tIM36/PjV7EsgpYHAxSqmfAZbLnAyX5NT0RoMAlKcY15FUkh+pVdkWmEmh/AU3dqpkhZzHosi45aYackIZIURSsX78Ow4dfCYsl7OEzoXzzzTd48MEHg+6eUt/XkDMheFM3hSuGcm79cjWulqqHyMc553ArbiSJSRGPN+L0a8gpISeENECxHu/DfoXbbrsNADBlyhRDCUBDaPTCLclIqgxMyP1L1kUocEoKINesqZs66+u/Fp0E2fasuo6vMdiH3DdDTj8vQhqSn376ESdP5mHkyKvNDiXq5syZg+HDh2PatGlo2rSp2eHElCAKATPknjXkVZSsq6XqIWbIF+x6Cl+f+BKvXvQmclLi+3hyKlknhDRwsR7vw07IG3KjF25NhtUmetYmcwZB8pas+yWJIhS0OroXpz5YDUtXK1KbuI3ryf3QDHlo/jPk/gl64AOi39RNnSFXJLd2IYoQUv+de+752Lz5J7PDiImCggJMmjSpwSXjQIg15BA8CXmoLuvqzSHWkO8q2olKqQKHKg7GfUJu2OqMEnJCSAMU6/E+7IS8ITd6UazJEGQBqUf/hGDpACQFL1kXoaDH3h/gOnQK5akpSG3irrLLunrFXZEkSvD8+K8Zr24fcv2FDxa1GXKH9m9FckO0JkAJIiGEhOHyyy/HDz/8gNNOO83sUGJOEAO7rAucQUHtu6xL3j4yUhUX5+OGoi9ZT4B4CSEkwdWqKP6nn37CqlWrcOzYMSxcuBAtWrTA2rVr0bZtW/Tv3z/SMcYNbk0GGIOlsgxikgxm9QzYkl+SKDIFyQ5PgzaueAf1KkvW9evQZDCxfq9NDId/l/Xq9iE3XPiI8gw54Clbp4ScEFLfPProo7j77ruxfft2dO3aNWAN3cSJE02KLPoEUQCClKzLXPIrWdcvkao6IZe95wBSlHb/iCTjtmc0Q04IIdEWdub32Wef4YEHHsBVV12F33//HS6Xp9N0RUUFli5dWr8TcksS1EGag0OAp+mLbHfi5G+N0KJvGQBAhIwkp5qQqw+ufg054EnwBErINQH7kFeXZBu2PYv+DLnsdgGpaVXcmxBCEs9HH32ELVu2ICkpCT/++KPhe4yxep2QM90acgUyBIgQuadknYfqsq4OVaFmyBV1hjwBmoHSGnJCCImpsDO/F198EY899hhGjRqFjz/+WLu9X79+ePHFFyMaXLzRJ+QAB7g3IRcEVOSm6BJyBSlO4ww5qyKR1A/qiiQByZGPPVGpV+oFi9WzZrvaGXLfcWZc9sxaRHgJgHGGPAFOrgghJEz//e9/ceedd+K2226DIAjVP6Ae0a8h51AAiJ6mbqhBUzcFQZeeKQk0Q27Y6oxK1gkhJOrCHmUPHjwYdBY8MzMTZWVlEQkqXimWJPgOGQfzJuRcECApvsFXgAIluSV+OOcRFFi7eO9esxlyauxmpDZxE717Fla3DznzX6tfzTZptaF2WQdo6zNCSP3kdrtxxRVXNLhkHACYRdASagXei8LwbHtmrGgLsoYcCLr1mVqy7r8rS1zSXfiu9iI4IYSQOgt7pG3atCmOHDkScPv27dvRrl27iAQVr7hu/1AGDqY7fJKu2ECEAkdmZ1Smt8LxlLO9D66+qRtAW2n5U08GRO/6xWpPDvwb5kShbN2tS8gVmiEnpFZ4FC6WxZtEfo+jRo3C+vXrzQ7DFIIg+M2Qe5q6cciGMchYss6D/9tLnRl38/gfMwxryCkhJ4TUQSKPgzUVifcYdsn62LFj8cQTT+DJJ58EYwwnT57Ejh07MG/ePEydOrXOAcUzLloBbSz1lawDgMxE7d8iFDTNbIbzsiw47PJ2qK3Btmeeu8X/YB1L6sUK0eK5GFJ9ybrf9xUZEIPftbYMJes0Q05IWCwWKxgTUFpaiIyMxhBFS1zuLKEoDHKIPaWrwzmHLEsoLy8BYwIsFmuEo4s+RVHw8ssvY/PmzejWrVtAU7dZs2aZFFn0CRbRsIYc0M2QK6FK1nVPECQhV2fGE63LOqqpSiOEkGBorA9P2An5bbfdBkVRcPPNN8Nut+Omm25CUlIS/vGPf2DChAm1DiQRKBYrwLyDE+NgYGCcgzMGSZf1WRQZ2elNYGUMSanNvXevWZd1Klk30mbI1ZL1cJq6wXPca/drFpqhqRtdQCEkLIwx5OS0RGlpEUpLC8wOJyRBEKrf1aEaSUkpaNSoSVyehFRn3759OOOMMwAA+/fvN3wv2u/npZdewoIFCzBx4kQ8/PDDUX2tYJiF6UrWPZ8Bxhk4FMPssb6ijddwhlxOhDXk+s99NQk5VxSwBrisgRBSNRrrwxN2Qs4Yw5QpUzB58mQcOXIENpsNnTp1Qnp6eq2DqMrJkyfxn//8B99++y3sdjvat2+PJ598Ej179ozK61WFCxaorVSZN80TFA5ZZJB1h9LqdkMQrd7HeDu0VfHDppL10LQ15DUsWQ+48BHhknVFlg1lijRDTkj4LBYrmjRpDkWR6zwQRgNjQHZ2OoqLK7VeXeESBAGCICZkMg4AK1euNOV1d+7ciVWrVqFbt26mvD4ACKKotW9V4Pl7L4BB4bJxhlyq2Qw559y37VkizJDLNVtDXvHcM3Bu3IDsV9+A0CQnFpERQhIIjfU1V+v9tZKSktC5c+c6vXh1SktLMW7cOJx33nlYtmwZsrOzcfjwYWRlZUX1dUNRBCsAzzZv6mgtKApkUYAsiFpDb6tbguhNyMG8685r3NSNysP0tC7r1hqWrPvPoEe4gY6+oRtAM+SE1BZjDKJogRjhJSWRwBiQkpICq9Vd60GahK+yshL3338/5syZY+quLfp9yBWuAMxTss4hg+uSVUNFm2GG3Ph8im78T4SE3LiGPPQY6t72A3hxEaQ//0DSuZSQE0IC0VhfM2En5DabDS+99BK+//57FBYWBlzx2LRpU8SCW7ZsGVq2bIm5c+dqt5nZOI6LFui7rAOAKCtwWwFZFMEVgIlAiiRBtHg+eUzwJOZVbcFFa8hD8zV1q2HJepRnyCWnw/A1zZATQuqL6dOn46mnnkJGRgamT59e5X0XL14c8dd//PHHMXjwYAwcOLDWCXkkChIEqxjQZZ1xBgUKFF1CLUtu3+vpzuYY54Y41OcAAAlSpHfiDKA+f61fRzfOMlkO/TxqhYAcufdU59hNlMixA4kdP8VunkSOPxax1/S5w07IH3nkEfz444+45ppr0KxZs6iW433xxRcYNGgQ7rrrLmzbtg0tWrTA+PHjMXbs2LCfKxJhcsEK9ao5Y57Bl3mvliuC4NlzXORIkmRYki3e+1mgcAECU8Dgzdj9n9dv27NoHNJE/YXh2rZn3pJ1WanyPfiXrAtcghLB9yy7jTPkiv6ELFRMCXrsgcSOHUjs+Cl288TTIB1LmZmZQf8dCx9//DF2796NNWvW1Ol5cnLqHndRVpquy7qvqRuYgpRk3fI0kaFpU8/rnRJFtX4O2VlpsOSkaverdPvWWCelCNpjoq22x6KMQbuEkJGehKwQ8ZZwz6WGRmlWZEb4PUXi52iWRI4dSOz4KXbzJHL88RB72An5N998g6VLl+Lss8+ORjwGR48exVtvvYVJkybhjjvuwG+//YY5c+bAarVi9OjRYT1XJA62YE3SBmn1ZEr0zuDKogguM8DKkSzJEL1NTgTG4OTpSGXlaNokDbAkBTxvcpJvsM5It0Z1sI6HD104UlI8H9HUNPXkRqnm+BhrTpo0TgUaR+49KxX5hq9Twzi5SrRjr5fIsQOJHT/Fbp5Ejz9cajUa5xx33nknmjRpgpSUlKi/7okTJ/DEE0/glVdeQXJycp2eq7CwvM6lhzanC1kBJeueNeSVFXbtfnabAwUF5QAAt9M3c15UWAFB17yt3F2m/busolJ7TLQw5vns1vZYSG5d7KU2uEPEK7s8FWJlReVwRug91TV2MyVy7EBix0+xmyeR449F7OprVCfshLxRo0Zo3LhxbWIKG+ccPXr0wD//+U8AwJlnnok//vgDq1atCjshj8TBdisM6gy5uucc864bU0RB23HL4pZgYWpCDjiUTKQK5SjILwGsqf5Pi8oKm/bvkqLyqAzWifoLo578yN4t5mRJrvL4ZMtuwy5nRUVlUKTIHc/8vELD16XFFdX+vBL12AOJHTuQ2PFT7OaJp0HaDJxzDBs2DB999BE6dOgQ9df7/fffUVhYiDFjxmi3ybKMbdu24Y033sBvv/0GsYYLEDlHnX9mTAiy7RkXwLkS0IRVey3di3KZG2KQdI+RFCkmvxOK3Q7b2vdhHXABxGbNw3os13ePl0LHq95PqeI+tRWJn6NZEjl2ILHjp9jNk8jxx0PsYSfkd999NxYuXIh58+YhNTUwuYykZs2aoVOnTobbOnbsiM8++yzs54rEwVaYqE2NqzPkgjcJlwXRU7IOwOIGLN4ZchGehNzzBHLQGGT9YO2O7mAdDx+6cPj2IVe7rAc/hroHGL+WI3s83U6/pm7umjeCSLRjr5cosW85+Q2ShCSc0+x8w+2JEn8wFLt5Ej3+2hIEAe3bt0dJSUlMXu/888/HunXrDLfNmjULHTt2xK233lrjZDxSmEXQxni1IRtD4LZnhp4vVWx7Julmy6UI9zUJpeyTT1Hxn7lIuXo0Mu4Pc894fW+gqhqpqv1vaHcYQgipk7AT8hUrVuDIkSMYOHAg2rZtC4vF+BTvv/9+xILr168fDh48aLjt0KFDaNOmTcReIxwKEwHdZigAIHjP1mRRhKwIAGRYZd8xERhgUxp5vggxEPuvISc+irYPec26rAduexbZbRYCu6y7QtyTxJpNqsTsnx+GVUjCR8M+h8Bob1xCauvee+/F/PnzMXv2bHTt2jWqr5WRkRHwGmlpaWjcuHHUXzsYJgq6GXJ1rPd0Wdc3sjWM14Ztz4zPJ+vGJTlGXdbl4mJPKKWl4T/YsA95FWOo5H1flJATQkidhJ2QX3rppdGII6i///3vGDduHJYsWYIRI0Zg586deOedd/D444/HLAY9Q0LuTfQEXcm6S7YiFW6Iim8NnACgFI0Mj/Gn33tcoYHNQD3hEbxd1jnncG75FskXXBjqAVV/XUcBCTl1WY8bdskOmcuQZTskxY0ksW5rUQlpyB588EHY7XZcc801sFqtAWvJf/zxR5Miiz5Bn5BztambZ4bcuCtKqG3P/GbIdUm4m8dmzNAu9NfmIr9u3ORVPF4tWed03kIIIXUSdkJe3VYokdSrVy8sXrwYzzzzDJ5//nm0bdsWDz30EK6++uqYxaDH4dubVJsh15WsS9yTNIqyPiFnKOPehDzElXGu0Ax5KOqer6KuEsP2/prQCXmstz2jE5G4oT/RdSsSJeSE1MFDDz1k6uuvXLnStNcWBBaw7ZlnhlwxVGkZ/v4bZsiNCblswj7k3FtOX6tkWQ6zZF2OTRk+IYTUV2En5ABQVlaGzz77DEeOHMHkyZPRuHFj/P7772jatClatGgR0QCHDBmCIUOGRPQ5a8tYsu7d9kxNyEURkuLd6gxp2mMEBpR7E3LGZQRbjqgY1qSZk+Cl7HoNTHbB3vsWU14/FEXb9szXnV5x2EPdHcxvRjyghL2OAmfIqWQ9XuhPdPVrNgkh4Qu3cWp9YthDXG3gCuYtWQ9+AZ1XMUMu6/4eybH62yTVPlnWL6MLlZBzzrXnphlyQgipm7AT8r1792LSpEnIzMzE8ePHMXbsWDRu3BgbNmzAiRMnMH/+/GjEGRc493VZV2fIGfeVrEtcPZzp2mMEADatqVvwgU2fhMuSCSXQshMZ3/wfwBU4zvgbeFJG7GMIgWtryK2+2/xmqY0PiHLJesAMOZWsxwu34vtZSAr9XAipi6+//hqCIODCC43VSJs3b4Ysyxg8eLBJkUWfPiHnajUcGMCU0D1fqkzI9TPksZlN5urWZbVJlg0l6yHi1d9OCTkhhNRJ2F2PnnrqKYwePRobNmxAUpJv1nLw4MH46aefIhpcvFEg6EZqz4CrL1l3y96kkflmyEWm67IeYrbWsCbNhJJ15raBcRkMHHDbqn9ADPl3WQcAxVnFrLT3GHO1XDlaM+RqOSOtIY8b+iQ8Vus0Camvnn76aUMDM5WiKFiwYIEJEcWOIPgycjWZFtStN/Xbnun//utycP92MWauIa/V7LWhqVuIMVT/vJSQE0JInYSdkP/222/429/+FnB7ixYtkJ+fH5Gg4pUC3Qy5d2accc/XzuRM7LH+B79W3mRIyAUALlmdIQ8+aBlL1mO/FotJvllfJjuruGfsqevrRYuuZN0VOkbmPcZc9N4/4jPkntdOSvVUQdAMefyQTFinSUh9dfjw4YBtRwHP1qNHjhwxIaLYMcyQ67Y9AwBFCXEBvYYz5LHqsq6uIa9VslyDNeT+e5UTQgipvbAT8qSkJFRUVATcfujQITRp0iQiQcUrheuaumlXzT1fyjkd0TM9EwIfAQjGNeSSd4acheiyrl+TJpvR1C2uE3JvuaDFtw+t4gwRo/74ehNyFuH1euoMeXK6p6yfuqzHD8MMOZWsE1InmZmZOHr0aMDtR44cQWpqqgkRxY6gy8jVbc9E7+mSfomZYlhrXcOS9RivIa9NsmwoU6cZckIIibqwE/KhQ4fi+eefh1uXiOTm5uLpp5/GsGHDIhpcvFF4kBly7/es6dkAgBQhFUw0zpArcjUl64Ztz2KfSDBZl5BLVazPNoF6wsMEUet6y0M1UtOve9NmyKOzD3lyuudnSjPk8YPWkBMSOZdccgmefPJJw2z44cOH8dRTT2Ho0KEmRhZ9xqZu3jGIqzPkugvohi7rNdv2LGZd1tU15HXd9ixUlZn+eWl3GEIIqZOwE/KZM2fCZrNh4MCBcDqdmDBhAoYNG4b09HTMmDEjGjHGDc9467ftmfpN7z7ZScwCWH2zByJj4Eq657E1Klk3YQ25PgmP1xlyQfQ01UEVJev6mQfBm5BHfNszmiGPV/qZJ+qyTkjd3H///UhLS8OIESMwdOhQDB06FFdccQUaN26MBx980OzwosowQ87Vsd47Q24Yr92ebuOAekqgPsjwfGbMkGuz3HVdQx6yZF2XtNMMOSGE1EnYXdYzMzOxYsUK/PTTT9i3bx9sNhvOOussDBw4MBrxxZVsztAkJQt7K6CVR6vrymTv/mciE2BJzTY8ToQFLp4WumRdd3XZjJJ1wxryGM+QO7d8C+cnHyPjgVkQGmUFfF+djWCi4Jm14J4TDS7LYKLof2ftn+oMOYvwbITk8hwfNSE3o6KBBGdonEQz5ITUSWZmJlatWoUtW7Zg7969SElJQbdu3XDOOeeYHVrUBdv2TFBnyP3GaK7IYKKlyhly/VZn8b4POefcr6lbiCoz/XGgC9OEEFIntdqHHAD69++P/v37RzKWuNcNDKmZTXG0JAtu7xVvkXmSQhkKOOdgjMGa3tTwOAHeTushu6ybPUPu29c71mvI8996DSePHcQZ321B2vArAr6vlssJgqhd/OBggNMJpKUZ7qtfL66VrIe4CFJbbv8ZckrI44ZkKFmnGRtC6ooxhkGDBmHQoEEAgLKyMpMjig0GX0bu2/bMO0PuV8ItSxIE0eK3htz4fIambrHehzzccwr/NeMhJgmoqRshhEROjRNyh8OBrVu3YsiQIQCABQsWwOXyreUVBAH33HMPkpOTIx9lnFDL0y0sCW6lEgCQltIEgB0uxYVSmaOxhcGSZEwUBQY4eCaSQqzFMmx7ZsbAZmLJ+u9SBU62bYasw3+gc5DvK96r80wQtVMkzjx7kTO/hDxYU7dobXtGJevxR7+dEM2QE1I3L730Etq2bYsrrvBcKL377ruxYcMGNG3aFMuWLUP37t1NjjB6GPMk1wpXwP2q4RS/GWNFkoBkVFmybkb1jrqGPOxk2a9EnYfqw6J/XlpDTgghdVLjNeTvv/8+3n77be3r119/HTt27MDu3buxe/durFu3Dm+99VZUgowX6hArMAHc+1VmanMAgNNthy3EuBXWDLnpJeuxTcgd3osU5aXFQb+vzZCLgm/1PmPgwTqt62dFBWvgbREgOdWSdW9TN0rI44ahcRLtQ05InaxatQotW7YEAGzZsgXfffcdli1bhosuugjz5883Obro0hqIgmuV6AIX0Dt3CBxlxqaiiiz51pFrN8bRGvJwzyn8E/BQCbl+Jp1myAkhpE5qPEO+bt063HLLLYbbFixYgHbt2gEAPvjgA7z55pu4+eabIxpgPLGmWQC7DIGJAOdoJACp1kYAABuvhN1/UPYSGOBQGiErREIuyxIczdtCtFcau7bGiL7LeqgZcntZCVIys7QTlUiRuAJAhN1WHvT7iuKbIddfPeKOwLXuzHt8ORPBBdF7W7S6rFPJerzRn+i6qWSdkDopKChAq1atAABffvklRowYgUGDBqFNmzYYO3asydFFF2PeC/CcaxffBTCcc2QkJPvrhvvKkhRQos79Z8h1f5vkUF3LI0zbhzzUtmWhHucfX4jHU8k6IYRETo1nyI8cOYKuXbtqXycnJxuSs169euHPP/+MbHRxxpLiSfIEJgBcQftkARbvIXQJbthD5H4iAKeSZthKRM8tWuHOaQln87amzJCjmqZux37ditUPTcHv6yNbAcHdbsjez5DdYQ9+H1ldQ26cIUfQGXLvD0CwAN61/dErWadtz+KNRNue1YhbceM/O5/EptwNZodC4lijRo1w4sQJAMC3336LAQMGAPA0/ZLDTPISjaD1K1G02W/GGSzcCv/sW5HcATPiATPkSuxnyGu9D3nAGnLah5wQQqKtxjPkZWVlhjXj33//veH7iqIYvl8vCZ7km0EEYwLaJQk45U3IORNgDzIIi4IIgQEuWEOXrHv/z0URign7gBtK1oPMkJf9+pnn/79/CYwcH7HX5RXlkERvQu4M/r4VrWTdAu+yPm0NeeATek8KmAB4Z8hDXQSpDUWWtDX+tIY8/ripy3qN7C7ZhU+OfYTdJbtwSethZodD4tSwYcNw3333oX379igpKcFFF10EANizZw/at29vcnTRpc41eErWPQOPyD1jCverulJkydhhHai6ZD3Ga8jDb+oW/hpy/RZohBBCwlfjGfKWLVvijz/+CPn9ffv2aevN6i1v8igwAWlCBqyMgamDFUPADLlb8VygEMBgQ1LI5FDWFqeLkEwY2KpbQy7bPJ11FVdkLxbIZWWQvBc57FLwiznqyQDTrSHnCLWG3FuyLljAazlDXpJnw/7v8qDIgcsPJN1rqjPkhn1oiakMM+S0D3lIdu+uCjbJZnIkJJ7NmjULN954Izp16oQVK1YgPT0dAJCfn4/x4yN3YTaecc615jEi1G02jQO9p2TdPyE3fmnY9syEfcjDGqNqWrKur+ajpm6EEFInNZ4hv+iii/Dcc8/h4osvDuik7nA48Pzzz2Pw4MERDzCuCN6EHAKYei3DO9Bxxgwz5BLnWpMpAYADSUHXM3POtWZxYMykfch15eJyYNItuT3JshzhiwXu4kJtKsLBZW3bOD1FK1kXwbRjjaAJubqGHEzQStbD3Yf810+P4uRfZWjULBUtuxj3RVfL1ZkgwJqS6otRkiBarWG9Dok84xpymiEPxeWtgnF6/+9wOLBr1y/o2vUMNGqUVdVDSQNitVoxefLkgNvrc58Yle/ir24NuaLOX/jPkMsBCbj/jLlxhjxW+5DrE2YZsNTsdC9gRjxUlRk1dSOEkIipcUJ+xx134NNPP8Xw4cNx4403okOHDgCAgwcP4o033oAkSbjjjjuiFWd88CbkjImedeTwDdwQBDgUriWVEgcULqOQleNUxnFYeXLQ2VquyOCCr1BBliPbhKxGqilZdzu9CXmE1w26i4q0f3MAzopypGQ2MtyH65q6Mc4B5rn4gSBN3bQTByb6StbDbOrmsntOLJyVgQmdOkNuSUqGaPEl4LLkpoQ8Dhi6rFNCHpJTURNyz+/Qvn278cMPW1BRUYGLL77UzNCIyTZt2oSLLroIVqsVmzZtqvK+l1xySYyiMoF68Vdfsu4/Q84YwLlnDbn/DLRfhZXxb1OsEnLd30BZqnFCHriGPPgYSk3dCCEkcmqckDdt2hRvvfUWZs+ejQULFvganTCGgQMHYvbs2WjatGnUAo0L3rxZYAKYmpB7xyrOGDgkOLgVqQxwc0CBjF8tR1EknkKSMzXoFlyKLPuSRwBSqPVaUWTosh6kZN3tXYsW6YsF7jLjVme20qKAhFzRb3vG4UnIEXyGXL3gwevQ1E2WPJ9ryR34Xt3ekn1LUjIE3cmN7HYBqWkB9yex5TY0daMTxFDUmXGn4gTnHA5vQ0WXK7ZbHpL4M23aNGzZsgU5OTmYNm1ayPsxxrBnz54YRhZjar8Srmgz5CJXq+I8Y4PFmgTJ5QxRsh56hlyBAoUr2kX9qHEbE2aWXMV99fwSck5N3QghJOpqnJADQLt27bB8+XKUlJTgyJEjAIDTTjsNjRs3jkZscYeJDBzGknVBHbkZA7gEu8KRKjC4OQe4BBc8g5mbWYLO1iqSZJwhj9GWKHrVNXWTJDUhj+xaaVdpqeFrW0kRmrTtYLhNPRlghpJ1FrSpm75kndeyZF2RPD8jyRXkZ+Vt4CYmJYExBtFqhex2U6f1OKHfe5zWkIfmUpy6f7vg9i5JUUz420Piy969e4P+u8FRfDPk6hAvcOMMuSU5GZLL6W3q5vd4vxlz/79HkiIhSUyKdNTGEGpbUl7TknWaISeEkIgJKyFXNW7cuMEk4QbexFlgIgS1fE0ddxkDIGmN3SQOCFyC7E3IXRB9CaOOIsueNc9esv+V9hhg1Wx75nZ74pYiHJur3C8hL/XNmBcdOwzGmK5kXfCuwRdCriHXThwES61L1tVmbnKQGXJ1Lb3FmuR9GU9CrrjpZCQeSNRlvUYcuooYp+yE23uhqb5vZUVIjXHfP7Seq9y4htyS5JlyVoLMkPsP9f4X2j1N3qKckNdyBjtwDXmIknX93wtq6kYIIXVSq4S8wfLmeIwJEAR1XxR1hlyAwl1aYzc3ByyQIHtr2h1MDFGyLoHrStblgEvtMVDtDLnnPUT6fN1dUWH42lbiWVMuuVz49NnZYILg24dc1M2QI8Q+5Fy3hpzVbtszWZshD3ycOhOurh8XLVa4AcghOsST2DKs06QZ8pBcsu/z6pQduhlyE/pXkLikKAree+89fP755zh+/DgYY2jTpg2GDx+Oa665JqD5Zr2jjjXct4ZcgOD9t+drNSGXJSlgCXlgUzfj3yO3IiEV0aVfQx7WtmT+yTWVrBNCgjhY/heapTRHZlKm2aHUC1FexFTPBOmyzvTjLnejyDvDWiZzKNwN2Xs13cnE4CXrsrGpmwkT5H5d1oOsIVeT1Bqer1c+vxC2la9Wez+3zS8hLywAADjKSyE5HXDbbbrO5iKYWkbIAF5FUzcuiLXe9kydIQ+2hlzdc1zwzpCrjdxoL/L44NaVrNMMeWhOXcm6U9HPkFNCTjxJ6JQpU/DII4/g5MmT6Nq1Kzp37ozc3FzMnDmzyrXl9Ya6hhy+BNyzD7lvgNZmyOUga8hl/4TcOA7F5IKhvnIrnBls/33IQ/1doJJ1QhqsIxWHMfnbCfj3jkfNDqXeoBnycKgJORMhsMCSaM4knHBzfFbqhoMDZye5IXkTcgUCKtw2+PfiVmRJK4UHzJkhr24fcvXiusw9XWVRxeyIXJAP+6o3AFFE6k1/r3ImxW037oNsK/Yk5M7K8oD7CoKg7fnOGQMP0oCKqSc5ui7rwZYJVEVy5ENyHoXkHAIAcNltkCU3UjOztMTb4k3ERW9iTmvI44NhH3JKyENy6i66eWbIPceK1pATAHjvvfewbds2vPrqqzj//PMN39u6dSumTZuGtWvXYtSoUeYEGAMnC/Pwq/V3nCk19c2QcxGA73dE1Jes+0+R6xN0rkD2Wwomx6DpZLTXkBv2IaeEnJAG5aT9BAAg13bc5EjqD5ohD4e27ZkApibkugSaw1P66fDepMCtlawzAHucJwKe0jND7itZV8C0EwDVlpPf4o7N/8AP37yHT599DPayksi8Hy99l/Wg2555x2NJEQzl7cFwtVGbLAPVdG2WvN2dk7wZv927htxZWRFwXyaKukY7oWbI1e1oBF3Jes1PFBwVZbCXvAPJtgllp/4AAHw8/2F88Pi9nm663tJeNREXvZ3WaYY8Pkgm7PWbiFyGhNyplazTGnICAB9//DHuuOOOgGQcAAYMGIDbbrsN69atMyGy2NlzaC/+FPNwyOLbmlMIMUMuS+4gM+S+36X0rU/Cuus1w7f11TzRYtiWLIwZcu6fgNegZN2sGXLOOZw2+ltPSKy5vJMe+p40pG4oIQ8DE30l6+qWJZwruqvjxoFB5i6tZJ2BYXfQhNw4Qw5BDBgQPz7yAfaX7cX+7zbh1F97kbv71wi9IzVs3S9UFQm5zAVjeXsQvMI3ux00adY/rzdhT3d4EgJbmSeZd1RUP0Ne1RpyXsumbj+u/h/APbP2bnsZZElCeX4eXPZK2MtKtcRbLVVX15LTDHl80Jepx+KEN1HpS9YdNENO/Ozbtw8XXnhhyO9fdNFF9b4De7I1BQBgYy7PGA9vUzfdeGJJ1pes+z2BLgFOOvi5dh6gisUFQ2NTt3DWkPvdN46buu39Ng8fzN2BE3+UVn9nQkjEqOdb+l1bIk3mMo5VHg2YpKyvKCEPR5CSdYXL2iDN/RJyhbkheUvcGIDdzpMBT+m/hpwLgmdfU51jtmOe73nXhAVLWOuiqi7riixD9naX5WDgjqpfW6moQEWyFTarBdxR9S+q2+VJxNOdnl9sp8MG2e0OWrIOpk/IEXzbM8VXsu7b9qxmJyJHdv6EQ9u/076WXA64Hb6LD7LbpTVv05q60RryuEIl6zWjL1l3KfoZclpDToDS0lLk5OSE/H5OTg5KS+t3ApSRkg4AsAsuXVM3EfrMW91tI2jJupqgSg6IpYcg+S3d8m/yFhVufVO3Oqwhr8G2Z2aVrJecqPT+31bNPQkhkaSeYzmDTOJFyqv7X8bEr2/AlpPfRO014gkl5OEwlKx7Dp0nIVdLqY1JgAynboYc2O3KD7jSo8iyrwEZPA3JFN3gJisSTnjXaCjeq9yOijIAwCn7SRwsP1Dnt6VPwl0FDsgF+drXkl/ZuVJNQu4qLsLmrm3xXZc2kG2VVd5X8s4spzEBgjfZtpUWBy1ZFzg3dFkPuu2ZOnsh6LqshzjxqSjKxxdLn0bunp2QXC5sW/0/73c8jwuakPuVrAsWWkMeT4zbnlEZYyguwwy5s97OkFMJfu3IsgyLJXR7GVEU6/2xTU9JAwDYmNtv2zPfsijBe2E22LZn6iyzWHIAjMvw/2sUkxnyWq8hr0XJuklbf8qS57grNe04SwiJCN8MuQtKmNsL19SRisOG/9d3CdXU7aWXXsKCBQswceJEPPzww7EPQNdlXfBey5C5DMbVXqzGQUliTijMc4gZOIoUO0468tAytZV2n2Al64qu/CvPnufr0CobE/IZP0xDgaMAay75EJnWRrV+W+oactnFcOQDF4StU5D95hrPe/CbiVac5VVexbEXFUARBLgEAYVHD6LV6R2D3o/LMiTvwG9Nz0Cy2w17sgB7aXHQGXKmKFDnGBSGECXrgU3dQpXbHdq+Fcd+2468fbvQ4ewBqCwuQGpWE7jsbSC7foPsdhoScsnlClKy7vnZKpSQxwV952KaIQ/N0NRNctTLfciPHTuCjz56DxdccDF69uxjdjgJhXOOmTNnIikp+D7ZLlf93+bRN0Pu1nYi1yfkjAm+HiKyBP9erMw7y2wp2gcAATPk0e6yzhXFOPbVZR/yEJUzhll3k0rW1URcpoSckJhyKa6g/44kdXmdXW4YFTAJM0O+c+dOrFq1Ct26dTMvCLHqknUwv71G4fuQqsPxX2V/Gu6jSJJfybpoKFk/7i1XB3x7iToryuGSnThhy4VbcSHf7pvRDhvn2gy52yaCywxy7nFtJl+ylxnjtVczQ17mK2U8eXB/6JetrIAket63NbMRUrxX2G2lRSESclmbIQdjVW97pt+HPESXdXUWXnI58efWrwAAZ116HSB4ZkZkyQG303+G3C8hD1GyXnS8Er9tPBZ0L3MSPbSGvGb0CbnD7dD2H69P+5Dn5eVClmXk5h6r/s7EYPTo0cjJyUFmZmbQ/3Jycup1h3UASE/yjAMOJvnWkEPU/s2YAEFUL8gGzpCrs9NikWcMlMRkw/ejPkPul4CH09QtcA15DUrWZdmUdZ6yd3s5daacEBIb+kkPZ5Qau6kNaO1y1b2r6ouEmCGvrKzE/fffjzlz5uDFF180LxAhsKmbwvXrx/yauglOAGner7z7k7uMa+9kWfLN5gLgomCYIT9WeVR3Z8/JgKOiHEUuX/fXSimwxLvGdL9IisTUoACXC0hOhmQzJuSys+oydGeF7/3lHTsY8n68vByS90JEUloGUk56yvJtJcFL1pmsaHu+KwC40wmlogLubd8jaeAgsOQU3xZngqh1rmchZiKcfnugN23fCW3OPBe/rN/jeQ3Jb4bc7dJK07Uu6+q2Z34J+e9fHMeJ/aVo3DIN7Xo0CXkMSGTpT3Kpy3pohpJ1t+8zXp9K1tVZf4mqV8I2d+5cs0MwXWpSKgTOoDAOt3d4FjiDsWRdVyEV0NRNnSH3JOTOxh0Ap+/iUNT3IfefEa/DtmcBM+Yq/8RdkgCr/8au0aXOkFPJOiGx5TYk5NFZR67NkFfTTLq+SIgZ8scffxyDBw/GwIEDa/0cjEXgP9G3htw3Qy5pM7eKbo9SBTLcTHcy6N3+rNRdYnhOWZKM+3oLIhTJrX3/uM2XkDPvVWBnRRlKXMXa7ZVSRY3iD3YcBG9CvispCX9CV6LosIMxQPGfIXdWVvka7kpfwl5wMheK5Ap6P15RBtk7Q25JTUWyd/a/yR9p6OM4X1sSoL13WQbzXtTgjIEVH4Xzpfkof/QhOD/6wPO83oT8y70i3l2/F27F0xU32Ou7vOvbuw66BKf3H4gLJk4B5wxgaudcp6Fcv9Jm19aQW6xWMOabIVdkt/G57Z6TH5ddqvLYJ8p/iRK7xI1N3eIt/kqpHI9un4lv8r4w9dgbmrq5fJ9xWQ7+u5KInxs1EZckqdbPEYv4SXxiHEjjnrHA7b1wbFhDDtFQsu4/O8zVhLzQ041eSjYuKYv2BUPulyyH19StZmvIA2bdTWjsps6M0ww5IbEVk4ScZsjjy8cff4zdu3djzZo1dXqenJzMOsdSllGCMgAiBDBv8y8Fum3PmG4fZMENSfc1907vui12NG3qiyU3yXhWxgURv5+qwKhenvuccvu2SlPXpTkryyEn+z6gLFUyPGdV1OPwzTff4LfffsPN142AG8AtrZqjXynHnd6kt3GKADEZOPiqcf/UFFGu8rVkXYM4RZHhKjqOdmf1CrhfBZO0GfLM7MZIcUtgYGhckQXGGNKt2UAmUF5UCMaB7K9najPknAFw2CDs/wkAkFxR4onpWBLcioDfjwEcNuRnpKN1koDUpplQ3G58+cAMtOnXH90n3Azu8hy/Lmefje4DLwIA5B0oBWOeJJvLToi6n59dcsMieAJo1DgTTZtmIj3DU/2QbGXGY8I9P9OUJKvh9kh8Bs2SCLEr+t83UYm7Y7/90HfYfPIbFEuFuLbnNTV+XKRj1y+lgeCbWVKUqn+3a8Os4y6qK1Z43d5TPHxuiAk4kI5kVMAByeL5ey6AacvTjCXrcpCSdQVw2yGUHQEAuC2phu/HumQ9ZGO2IGq6hjygLF6SEOtrTLSGnBBz6JcFRr1kXWoYa8jjOiE/ceIEnnjiCbzyyitITk6u/gFVKCwsD9iZJFySd79sEcxvH3J1MPCuJ2cCZMENWdQlCFAADuSV5KOgwLdGurjYuF6aCwJWffcXBvU5AwBwsPiQ+gTqJDtcdhsOnTiiPeZEUQEKGlW9tpsxz8mlehx++eVXFBUVYvfu/egiCKgUBFjcCtTS+qLcfEhfb0H53j+ADi215ykvLoalIPRr2SqMJe17t/2E1Bana19XlhRh71efooOYCsm7BEARk5HilpEiZoB5h3SrkIT0xo28CbkC/ufXYPxM76FgUCQGV5EnjsqiMhQUlCO5rAIVzjStv85JZws0sTtQUVCOExs/wS+5h/Dn0YNoOuJaVHi37XHJovbzKCqsALxVApy7cCrPtywg71Qp0io9SbzdqaCgoBxqY9ny0krDz9Rp9/yhKi224/iRE/jlo9UQuIRzxk4GWEIUpWj8PzfxzCX5Ek2Hy4mCgvK4ij+36BQA4ERFnuHzEkq0Yne4fYNnWbnv91WWZeTnl4FFYOo2nNhlLuPtv95A75y+OCu7Z51fGwDKve/LbnfW6Fj7i8XnRn0NEoc4kM5TAJTCrfaOgQDJaoOUmgFRYr6SdTlIl3VFgaXkTzBwKCnZkPz+7ke9qVtdZq9ruIY8YNbdxBlyKlknJLYMM+RR2otcfV4bJeTm+/3331FYWIgxY8Zot8myjG3btuGNN97Ab7/9BlEUq3gGH84DtwoNmzeBFLmvZF3tsg4AnMmwum1wWzMgCW7Iuhk7MAWCIqLMXWqIw+3fsVYQUVxpB+eeD3yezTNDLurGG84YCk/laV9Xuitq/N7U46CusbTbKlHhfV8pulAUmw1KRQVkwXhyLrscVb6Wy+UErIBVkuG2iDix73f0GXm99v3dmz7G7i/Ww3FaZ19Tt/QMpLglpFl8J6dWlowWRTtwFGlg3HudQ3v/AJcZlArPLyu32TwxKTJO2n3PUeLOBBQZnAOOAk8yZGee96+uU09Ky9DejyxxMOYt2+cu2Mp968wrbA4ku3zbnnEOCN6SdcntMhwT2e35YZWc+AsfzHkL9lLP8oKuF49Eo+atQx+8OBaR358ocxu2PXMb4o2H+CvcniSx2FkEtyzBItTsz2+kY9eXl7ndblh1y0MUhUMQIjfPVZPYdxb+imX7lqBb1hl48YLlEXldt/dqmSS563Ts4uFzQ0zAOdK9JeuSNyEHB8qaHgcXusFy5AhE77ZnsuQO6LLOFQ5R7bDepBtkGJPaqO9DHmT2usbUBNxiASQpoPxd418Wb0Kndd8MOf2SEhJLbl1ndSpZj4y4TsjPP/98rFu3znDbrFmz0LFjR9x66601TsYjxnuiatFte8ah34dchtVdqUvIdVk0A5KUZJT6NXVzqY3CwCHDs7DQ6XDCJSnIc+R6SuIBiIrvJNnetjPknyqR1jYNNqsNFbVo6qausXTYKlHpvXqfpvud4nY7uM0GWTBe2Vdcvl8Ml92G0rzjaHZ6F+02t+QGrBY0LbfhRHYmSnKPgHOuzboVHvHsm15aUgTF+9zWRp4u66lihvY8ViEJTSqOAegKxjkUtwBBu/DBwBUGycYBMHC1+RqXkefwrdWrkJIAb5LmLvfMkskCg+RyamvIk9J8r6lIHGC+GXKHzfdeK212NPbOwKonYqJ+H1odNSE/9tsHcJT51vrrm8TFk9y9vyE9OwdZLRLzYoFK3/UzHpu6Vbo9v6ccHEXOQjRPbRHzGDjnhqvZktsNK3zVR7IsQxBiW8VR6ioBABQ46rBbhB+3t9+D201N3UgtcOgSckG7kYsyAAZ3eoZfl3VdlRxEQPE1dJObdIOs+KqtrG4GR0kJ4Nv9NKhCRwGaJOfUrmIlWMO1Gj/W+16sVs/jQjV1i4sZcipZJ/XP4cMHUVRUiD59zo5IxVo0uOXoryFXG9A6qKmb+TIyMtC1a1fDf2lpaWjcuDG6du0a+4BE31oyX5d1/RpyBRbvLJgsuCH7XTa3KEkBXdbVE0YLY9rzWCEhv9KJn074tkjTJ+RKSioAhky3ZzZYPdEPhzZD7rDDps2Q++LlNht4ZWVAQi57m0C5nQ6seXgqPlnwKF74cg4mf3MT7JINbu9V8iy75xfJ7bBr+6ZzRUHRscMAgJJKX7M4a2YWkt0yUnUz5OkWEVmSE4xzpEgyZDczriGHZ5Yc8Fw88BwYGScdvueolKzacgK3rnN7RVEBJJcnvuS0dN97kxQtIQd3w2nzlcnY7Y4g+5DrZkh0JG9C7rYXG2532+Pvj0p5wSlsXPwkvn75v2aHUmeGGfI43PasUtI1PIxg8hkOt99+oZLbeBJtRqd1h3f9mad6KDIzXb6mbvH3OSAJQOHeknVAsnjGQEk3yy2lp2kl65IkaftwM3g/bwoglOcCAOSs9oYZ8RHft8Sppe8H7PSh903eV7j+i6ux6sDrtQo/YEY8jNlrdQ05Uzumx3HJuiJTyTqpf7766nN8993XKC4uqv7OJjGsIY9CyTrnvMHNkMd1Qh53vEdLhGDYh5x5kz4ODqs3IRdsf8BdYtzyKllORqnbmJCre44LuoQ8CTJOlbuw6eAe79eNIcq+hJx7S10tiuf/4W57xjn3nEQAsDscqPQm3am6c3XPDHmlts5bpXgT8m3vvqYltbsPb8PBigP4q2Q/3N5jkeyWkWb1zDCU558EAFQU5muzxGrizhiDJbMRRM6RrpshTxWtSHHJGPDHcfQ/cAKyS/B1WfdrHcPtnsRZkdwocqZot9skQeu8Ltl8yVDZ8aPaa1tTfM12FFlXsg4XXHZfQu5wOnUJubrtmeeEpfRkrvZzVBQOReLgnEP2bimV3qSp5z074++PSmVxAQDP/u+JTr8uUz9bHi/0v6dmJeT+A6fkX90RqoFTFKlbmrgVN+xyZNaK+bY9C+yATUh1uL5k3ZuQC7oZISklBeVOz+/OntxiMO+FH6Y2TOQcgrMEADxryL1JrVXhaFxhBSQZFYWh/wb8VfaH9/9/hrxPlSJQss684xyXQpWs1+E1IkSbIXfT7zipP5xO78xwnFZVAtHvsu5W3ODec/5InRfEu4RLyFeuXImHH37YlNdm6hpyzxw5AGOXdc5kWL2zYKLzJBRmHCSschLKXWWeWXUvt3cg1yfkVibjVLkTp+yeAVtwN9dmyDljgFrqzb1rmE8Uo/Bo6D2//elPwu1OFyq9JTGphpJ1G7gtcIZccjtx5Ndt+PO7L7XbnG7PL4u9vAhudSszWUG6xTOgl+d71rsXHTsUEIvVYgXzzlJnCL6EPEW0QnIKaGx3Is0tQXEL2gy5/6eW2z0XCYoKy6DoknWHzLSSdUmXXJfmehriJaWlg+nen2GGHIBLt6e6y+nUtj1TE/GW3XpAsFhRePgvfPPKQsiSpJWrgzugLizMyGkGAHA7otOJsi7c3uOivrdEZixZj789tfWVLAVOkxJyv4FT8TvZNmOGXD/YlnjL1+tKTcg555DD6DBNCICAknUFHFy/7psxHCsuAQA4nW5wrRLD+3/OwBye7/PkxpC9F4bT3ep+5r6/vcGoTYxsUmXI+1QZfl1mr9XfF22GPD5L1hWZa/10qWSd1BeeCTPP3xGXf4+pOGJcQx75c1uXbvLALtkbxIX1hEvIzZR0cD0AT7dV/Qy5b8DisHi3/RLkyoCSdauSBAUKJq36Hg63d+bWO/iJjGmNYazMjVMVTpS7PCfwdkeqLyEXLbrns0KUGTp8VogNz82BUsMTT/26SrvTXcUMuS2gqVuFy4ED2zYbbhO8sTnLCrWE3CorSPfO5Ksz5EVHDwXEYk1NA0vzbB+WppshTxatkJ26ZNnFfM3zAhJyz8lLQYGnDJ4JnnXkTpmDexMMfbl4ad5xAMb144B3DTksUNvHSTZfQu52ObXSdHWGvHHLNhh6+30QrVYc3fkTdn/xsZaQc+75HFhTUpGc7imjj8c15C4tIXcHbneTQDjnxhlyKlkPKjAhN/7MFRM+Aw7dYO7fY6O29H/jqGydhI0DqUj2XARmDA644BKMvxsl5d5lV4oMplV86RJybYa8sVaynqH7KLqqSMjVbX5stZ0ZqtM+5N736S3JD1myHrBfeawTct2WjQmckHPOUZJ33NOtnzR4iqJoyac7jidK9EsEozFDrn9ODh61Tu7xhBLyMCQd/QqAMSHnXIHobbwGpkD0fogEtw2y3wx5uuK54r4nPw+/HTyOgoJ8bbZaFARw71X01PT9OF7qgE0t5XSnaSXrXPA1srMqVqS4BIiy52p7TUui9SeodpesJeQphoQ8+BryF6UTyCs+YrhNjc1VVgzJ22jPKstI92bOZQEz5L4k35qeAZbumSFP0SXkSUKSISFX3IKvy7r/p9abbOft86zZE6wdPY/hHE71wodLd9J/ytO5PjndmJDLsuJpoOGdJVfsvnXuksvlK1n3rh0HgNZn9EKfK8cCAPIP7NfNkNu117Ame8ro47FkXX9SKHn/+OuvRHKFo+SULe6vTvpvI+SmkvWgXH6DGvfrTlzVbPLCXU9j9s8Pa2u77tp6B1bsX1bnmPR7jJa6iqu4Z83pE3Jq7EbCxjkEMCQrnvGskjnh8u6aopanV1SUggMQFQmyy79knUHQzZBL3rE9w+0b+6pKyNVEvNbb/dRh9lq9iM2SvNViof4m1KUsPgL0ndUTucv6sV0/48M59+HnD1aZHQqJA/oK1oY8Q+6fgDeErc/iust6vOHemVGBiVqXdQWKdlWDMw6rVAnO3RAVd0DJeobsSciZWIkdX6/Dz4qE5qmeZE0UBChchgjAIrjwy/FSCEkOtM5PwRElFYIl+Ay51e3LTt0OB5L9Zn2DMcyQu2VUpnqeO03X1A3eGXL/NeSiwmC3l0OEZ3bbbbdpW7K5y4qNM+TevVnLC7wz5N6EvFnzVsg/5UmeralpcCV7TnqSrb4O6UliEmSXfoZc0GbI/ZaQa13W80+UAFYrBEtryK49AHfCZpNhgSehhvdaRlmRJxnyP1aKOqizJIA7tXJ3z/uWIHn/AFmsSYbH5bTzXAAozTuuNXTjiucPVHJ6prZOvSYl65JbQeGRcjRtnwnRUvfrZYqsQBCDP8+SPYuRdsy3fZ7scqGySMHX/9uHrgMbA8o+uBynY9+WMgwY2xHteubUOZ5o8e+qLnMZClcgxtG+78YZ8gJTYvC/ks1l49+oUCXrLtmFD468BwA46cjDKftJ7CreiRO2XEzqemudYtI3bCl1leJg+QEs3v0s/t5lMno16RP288mybHgfNENOwuYd05IVEQ5RRiVzwMW8F89t5ZAys6EoMrjFipa2AjidTlgBVLgZUizwzpB7qj30M+Rp+vG6BiXr9lqehNalqZtW8ee98Byqcsrspm76WfFELlkvO+mZICg9mWtyJCQe6Mer+E7Io9vUzf9cxS7ZgOQmIe5dP8TP2WoCcIqe5FlgApjWZV2GRUu8OVqc/AlJ7i883/NLyFNlzwBnESvhdjkgSRIqvL9wFkEE954FWMHwV4ENZ51wYNi2FuiZa/N1Wddt9WZRLEiSfD9CyVmzq1T6X3hJASrhSTBTqlhDbvFeJRcUhqRiz0lCsrdpmxqbXFrqWeMOb0LuHSTL8/NgLyuBvawEYAxtUnyd0POkUxi9ZTRgTYMg+hJdC0sOUrLujS2gZN2zvsTu/ZkwIQuMeZLtCofifc++P2xu73FKSk83PI86wDMYE24AsHDJk9TDt/+4KqulZ7uwisJTcNk8z831M+RaQl79DPm+b0/g61f34+D2uids+7eexPtzfkb+4fKA7xU48vHOwTfxe94O7TbJ7cKpQ+VwVkr4Y8smbFvzPxz6+XMAQNHx2q1ljJVgJerxtvVZpdu8kvU//9yPP/7YGzhw+ledhmjqpm9GWe4q03aLcETgyrhd0ifkJfj8+KfYUbgdnxz9yHC/b/O+xtM758JVTXmcfwLudsfX54AkALXBquIZU+zM5ZshlyUw71jBLUmwKhJcDs9nskLyjH+MM18T0uQsyN4LRKluBg4GLghwOWpQsl7LNeR1K1lXZ8i941yoRo9mz5DrLiYmcsm65PbtSEOIcYY8fsu0o93UzX+cdzSATuuUkIfhU4u6Tsy4htw3Qw4kuUph4Z96vuf3+GTFM8BZRd9AbPOWuomioDV+Eb1NXxo5vGVuTu5bQ64rWc9gGYaE3F3DhNy/hLMSnoTRsIbc5l1D7t3qzSr59kNnLk9cSd7xUF1DLpV4TtKZwiFwjlTve3PZKnFi3++e99S8FTKKfSf3paiAi0vgWc0MMVlYsmGGXNF1WQ/41CoKuNMBl/fiAWPpYN4GcZUOdZ/SwJMF/xlybYBnwRNy7p1lEP0S8pTMLCSlpYNzjjJvObxWsp6RCWuKt2S9BgNuZYn3xK647olO/sEyyBJH/qHAhLzIWQgAsEr6ix5OSN6frcu7ft5V6bmfvTy+ZxndQZJvM9aRf3D4Pcz7dY72u6zinBtL1p0FMVsG4HK58PnnH+Pzz9fD5vCc4FsFz2eY+SXkoWbI9aXkZe4ylLs9nymHXPdmK/qBtsRVglN2T0VNscvY+f9/fyzH+mPr8Fvxziqfz//vG82Qk7B5P9JJ3rHYDpc2Q85kGYJ3PFSsSeAMkMo9v9u+33vv+YE1HRCTtNtT3AJs7buhonMv2CtD747ia+oWqRnyMBobqgm5WgkWanbd//YIJeSVLy9F3uP/rvbviuLWz5DzuF9WFYp6oT8el7SR2NMn5PG83CraCbn/5IG9AexFTgl5GLZYPAOoAF3JOlcgqjXUjEFhgFNdg+zfZV3ynAQnCb6BWPIOIhZRhOydrrJ4Z3qTvFemLTLT1mkLqWnaY/+fvf+Ot/Q667vh77rbrqef6U0z6pIly5ILrphiAyaAY0LgiWNIqE8IvG/yJCFPIE8e7ATHhJiENxAM2JZtsA0GN1mWLVuyeu+aoulzZubM6WX3ve+21vvHWnfZp0yRRrKN59LnfEZnn73vtu97rfW7fr/rd1Wo4uZq0vKAL/R7PPvVv2P5TH+9N6xuddRZC5CvqCH3EvO5WKSJhoJJDqTJgqZZ7EvdmMzu+ZSHtcTkwLe+CsD4rsspnsyOKbD1dtXgeN8xWaKYNRwH4pzLesLC5yNYWkQm8nqrnAPkeqKO1pDdrTZ1Mwz5GoC8kBscVkrWhRAMbd4OQMNI8TOGPCdZXydhknc4j3zZ9+9LiSjQ2/DbqxdKy74GWF7u/onCgMg3NfdmcRD6Gnh1G99ZE8Oh2gt84vBHCWJ97RKH9SRRBmuD9Jc7PnXkY9x55g6O1A/1ve5Lvw+k9+Jun4T95YxGo5YaxbTaep+D7hAAQvY/S+vVkNeMQRVAM2ykjHms4pdcr5+XrDeCOnM9U+Jikkbp38w+zyXjXbmI+U5e1FyK79BYkyFPAHmUAnLlekS2TWzmvqzfuP6cKgzj3/stfvNTiwy1FMXIQhbLYDvps7hWJMy4L33iFzOOvYSWZKlEPTV1W0+yvmKsuAiAXMUxnU9+nOXPfAa5uHjW966Uqcv4uxOQx8ElhvxSZPFKMORKKU6ePEGrtZqsOd/oqyF/JSTr3wOtzy4B8guIwUgDMSHsjCEnTsE5QiAti1DpiUyuAI5OrD/z1jP9i3UA27bTydxWNhDjmQnGkQo71tu0y5nM2pMuVZn13c5L1k899wTPf+3zPPzFW/kPT/wbjt73d0z87M8RHTm8aoEaqCJCKoq5l1Wng+pmNeQJQ+5IkbYW88y/qZy+bZzFk/P2ewyMbwJgybRlu+KWN+JOTmKbZINvmf7dVV2f3DMPnSDrDw4gz+KyDtCZ1CBfKBshHDCAvOUrI71f/ZmVpm7yLAx5MTc4rGTIIZOtN+dNHVhSQ16u4hbWl6wvnj7BZ//dL/PM7Z8DSBnqsPfSWzWFBlz77dWAJGEf+0oeAj8F8ZHpoR4bJrTb/M6qZfrooY/wqaMf54mFx4DM1M2zvPTZfKV7kSulqIfaCDBhkJNIWp5ZWFQcfd+9UrL1ej1TpPQ6CSDXfg2W7H8w1mXIw1r2/0E9lazDS5et90nWwxxD7vcbvLXMNQ3k2e/FlYz4JYb8UlxwJB1PzMTRE2EKyJExlvEbka5HZFvEZmErc4BcKQtZHKb35S9w3fGQm44rvNBO25b2zqJoy7urvyin9ZdS320AeGrqplQK0r3jX2P0U2/EmXn65elDHgRZG9mzKAhgNQC/0Dpyqz7BwDf+Jfb8/gs7xosc0fcQIJ+YOM6dd37lO7q/9rc78vPVy+WyPjs7ze23f4F77vnGi95G9DK7rK80oL3EkF+KvqgGuvZZWDaWYbGVkqCM47oQhJZFt7gFZdl5ghcAV9rsmFe8+7l+KSaA4zjEpmWKjQOWj2cmfVcqbAPmKRTSz9ixTVlmv+cH9F5DL5jnJo/z2PwjTH3pU3Sfew7//ntXPeQhpT6HdQC5vARKrWLIC4ENBnAXjGlbwt7T1RvxEjOYXo+BDZvTbY7t3MNorM3ZEsO3rq3PubdZA/dGoLPiQvQD8jjnst5n6mZo8860bmVmCWOcZ+nvqtUD1WyucosHKJT7a8iTCV1Y6zPkQggiJfjThyZ47kwGSoYNQ95a1JL1hCF3S5VUsr6yxj8KYuZPHEFGIVP7nzWv6WNIwPRLibMx5LWUIc9L1oM0IaCUPl8Vt3V9fiN8WSWBcSS552MHefDTR/ra2awXy0ZC3TQAOAHfruXgCp28Cl9hyXon6ug2iPQ7qud/LzsVNhR1ecYrB8gzYNvt6ntwwNOA3JH9vp7r1pDn+oM3w0Z63eGlZ67zkvXF3mJ6XZaD5VRVEMowBf7nAuSrJeuXasi/G+LP/uzP+Omf/mle85rX8MY3vpFf//Vf5/jx49+egzHzUypZF/2SdRGasdDRc0XUNgnMvlIVC1UYRnX081HpgRNlc0vgr7+ITaTqRd/ihXu+ht85OzhdGatakl1QDbl5bz7xbAB54fid2M3TeKfuzbaZJOAvBkOeuybqLAoCfZgrGPILdFovHv4ixSNfprTvUxf0uYsdGSC/+E7V32nxzDNPcPToYU6ePPHtPpTv2HglXNYTZrzZbJzjnetH8HK7rK9iyC8B8kuRix/YdoX+H9EvWVeYiUsITl62i+6mG/HHt6ySVrvS5pfujDWDuyIc2yHCyG5xEJaPp4r0Nu+k4thYBpDna8hFJKjE2QSfl0QHhgmzWgF2LIiXdRJAdrvUW/0TXaSKfXJ1ALmgTcXilCE3NXBJXbdS6WsJQy56eiOup0Gx8nsMbNiUbvPaH3gX8bEjAFSLGgx3bP2Z3qhmyK1uAlJWMOShtZohF+CU9DF0Te12wqwnpm7tQKCazVVu8QDeSobcTOi2U1z1XstQJrbr8cSpOh9/9BR//EA2qSQMeXt5xlwePXg4hUrKkOfb3LRrPl/+0LMcf0InEpqLc0COITeAfLG3wNMLT74oMNxrniFsf5NuY/WguyZDHgaEiVReJvdSDKpHHMqLwtqvF7WZDvMTTaYO1tj/rXO7zSYAtxfp40zk6Y5wcUx99Ctt6pYHqasYciNBrbgVxhNA7r/yDLmfAHLDkDtGzWOZhNV6DHktB8gbQZ1GkJ3rS2bIcxPtROs40hTFSBWnTHyiMADSMoX14pJk/bszHn/8cd773vfyuc99jltvvZUoivilX/olOp1vg1QxYcjlWjXkEZZ5TJQB5En9b5xrv6iwUcXhtAtIKQBHZiA3yN2XsYz4yAt/zGNzjxDJKJWD3nhsiIO3fYnDD959Ycf/UgzXEoY8D8iTunKTKBBRN9umSThfDIb8QgD5SgB+oQy5MOVY1golzisdSQ25jMI1vW7+PkXXrIH88/Q7+l6M/hrylweQJ3PiSwH8L3sN+Yptfi+0PbsEyC8grt2wC9AMuU0mWZ+Tpv5ZWHSMc7fyCqk0LQkbm+0LEDmrAbltW0QpQ+5iOy2csasJRzZSGKimkvU4B/JVrCjG2aSZZ2CDrp7MBDDYdii19IP30PGD7Lvjc337VnI1IO9120SWyDHkerIrBvq8HSkRSV/uFJDrB8grJIDcZ2iTBqrl4VEuu/kNyCkNtq7efjm7X/dmjm7WC/u4oAFCdfJwei2xM/YfwDLzb89K5PsutmuY9qUF87GK+dcAct9CNhtrMuRjj/6/eBPZQieZ0B0vA+TxikfEdl2WjRKg3s0GzqHN2/SxNeZ0P3kDaF2vilvSgDz//SydbhOHkvayPv+g08bvtDLJuAG///W5D/BvH///cLD+wqrjP1d0Fu8nDvbSmH181d+W12DIZ+qTBH7Sizy3MFIaDL2cxm616WywfeGB6TWN6PKROJb3zHXOGHIX19wfr3Qv8kYOkLdWAPKWAZQVp5oB8leIIW80aun/ByZpVnbKOMJJGXLbM+PLugx5zmU9bPYlH3ovUUqWl6KtBPfJfdrKKQ7OzZCvUABdAuTfFfGxj32M97znPVx55ZVcc801fOhDH2Jqaor9+78NkmKTAHUN8O6JgAADyGWMZW4paWTdST21VHlA5SALw6iuvr/LvsLKJ9Bz0tTnl5/jcyc+w58f/JO+hed4Tb+/m3uGzyteQtszlYwBzmqGXIQZIE9BeqG49j5fTOTmSNW5MIY8Di8QkJtxR5h+8d+uiHJ1wn/fZetJcs0/izrkez1eibZnGSB/8d/Dy932bJVk/VIN+aXIhygNmf/JJOtSSSwyybpvssrKclIpl2sk7bayKIRrA/JISHyhF/EFe5h3MgcFDS4tIbDipO95LhQUomzSzA/meTZ2sO0y2IHFSpFZv8bKRt5KFSnl7v2O53DPtTt5cnfG8qcMuZ+0QZNYSRuwBJCbYywmh9Trse1VN3PTP/gZ3v7L/xrLdpCmLmxs4xZu+ic/T8cw3JVQg1bZngOzqBFuP1M91uoilGLRrrBULmKNjmFvfxs4Bbr1mn6TrYF4Asg7oSCq11KmX+SY5mr9BQoH/zb9PalJcwrZfjt2P1Nvux4tw153woxNrAyP4XgFlJIoWU8Zctst4xYSl/VssdExNdlxkH1nrYW5PoZcKcWh+kEApjqT6fuUVDz46SM8c8dqw758xKFm3YPO1CqGfTlYAtXPkH/swP/m5LLZpsomAiX1wqjbePnqyGszRlHgWaDgua+fXve9SqnU9CgBg0kNuW3ZOYb8lQVizT5AvlKybhhyp3JOQB488RidT3xs3f6/Fxr15NkAop6+JgWrQMEupAz5YqwTWufjsl4P66nBGmiGuxk2+MgLf8zxxrELPr6zSdESY7e84iA/UUspeeyxhzhwYG/62iWX9b8f0Wzq73xoaOiCPifES/9ZyZD7IqJnEuYijrBNezPpuCjLQhlpe16y3hROH0Ne9sGSGSCP4jjd37K5z5eDJbqxAaIKRpqGge92znq8kd9DRmF2/Gs4oJ/3+csVbc8AlDnWBMRG3RSAC8OQE1/APta77kE/Q362966sIZexurD9mcSH5dcvyj2TnsMF3oNx7pwjv3tRj+WVOP7z/ZEyTpnxIPC/q479lbzucdwvWX859pXMiWEYopR8cdtQeYa8d9Gv/UpA3otfvmfjlbhvzidWI8NLsW7I0hAQgGWnAFmpGAtHz99C0CuYjLltgzJ1V7iExFjKohCtDch7KqBrLePNTRJs3M7GHPQWZDXkcgWYdqNsW2tJ1gGG2g7VLjyxRwMBlbDeyiEQEZYsUgr05NYuwFKlhLQslqoZGE0Y8kKUMOQKyw+glGPIZQLIs0WJFYXc+KPvSX9PjFpEpZoyAQNRhU0d7cYu65OouItwBhBOCUUNzDmXwojL1RJHxRhHN4+wfds7sMfegtur0Gsbqbg9aC5aKf2cv7iQMuSlIKJT0AuNoh1i5aS3ict64ooO0LHLDMTZtbQdl5qvB8xuDpALy2Jo01YWT59AxQug9HdhuSWc1GVdt4gSQtCtG0AeZd9Zc2GOKHDNe2OWg+VUmp2XCLdrPlMHawgLbvqxHYg1nvb28jJKtsx5TRN0IwrlbIG17C/jxAIrZ3Rgx4LYV9oXgdWAvPciGfLadIcX7p/mijdsZMNlA2u+pz6j74XLX7eRQw/N0Kmvn3Htxd1U1pwwqok83RVu8rWnIP2Vij5Avk4NecWtMhwO84NnfpCGU4dXrd5O6799EDkzzfJVV1G//iqGhzetftN5RhzHfU6qsR9BATy7QNEqGQNJCKwgff9asVKyXu8zdetyz9TdfO7EZ1j0F/idm373/I8vJ89dK5LSirziIC9Zn5g4zpNPPorjuFx33Q3AJUD+9yGklHzwgx/k5ptv5qqrrrqgz46NrT3GXEjUS0s0AVuh2XIh0q4pIo4RwsIJIyLXQboeCD22ShXrJL2w2Fco8eaxTWASsWUfuiqnaFOK8XF9rPG8Hu+aYRPPHH6166SmckKG6XtXRuj3+Ohv/wuqI2O87/f/CIClokNeo+PZYt3PrwxZsOkApcEqyew0NlTCHh5I57WiFabA3amUCYBK0Wb0PPexXnQnbZLUX5norNtbKPWPsQPV4nmfIwBmzHPCxoV97jzigu7BXGlVpXj+39PLGRfjGVoZSYINQAj5sp3ny3Hsr1SMjQ1QKGRlqVG0/nP/UsJ1s3XfwIBHqVQ6y7vXjjxDnpTbXsxrb59Z8YIbv6zPxnfCfXMJkF9AqNIgsICybWTOZd1WjhazCYvAZJWlbWObCaugHFqCFPysBcg7ykfaEd7iDLVBRbm4Q2e5bQcEODJh5Ps/Z+fsw/sZ8gxE7vE3MjMU0yqa7LwBp1VVZEm0cGQhlawvV6Fe7JeK20Jir2DrnFgiwh6UipnLuinuLuX6rKteL5O0AaplAHm1mkpVb25fg4WAAYHq1VBBD+EMgFvCKUmibjZA3aBmOKZGWRgoE5Y3YAOiMEjPP6HvZsOMuy74ogCqR2dpIWX6S0GYA+QRsZ8Bi6QPeR6Qd1cw5Jbr0vJjLG+WdjiSAmyAoS3bWTx9AhlNkWgZbLvMscf0EkNJSRwGOF6BOdPSReWygM35WaJAS/wjP+Z062T6t7ypVlLLraRuj+YWs+uTxMLJiewX1WV5aobNV+xIX6oFy31yddDt9URsg1oBhg2w77xIhvz4U/Oc3rfE5IElbnjHdq5+8+a+JIKSipoB5OO7qhx6iKyWfY1o5dqFJYZgyeTgWC7K0FuvuGQ9yDPk/XX7SQ101algL1qMBCPUZmurtiFbLeTMNAr4+vNP4T/7BP/0n/4ig4PDL+qYms1GnzpCmutasAqUcsaJodHgynVY+X7J+gpTt6ib3p8rpfrnim5Ool60i6sk60u+BuTtPsl6dn/u3fsMoBcucRxh284qAH6hknWlFIcPH8SyrgDcc77/Ulz8eP/738+RI0f4zGc+c8GfXVxs8lL9J6O2uceUwopjZDJnK0DGIATFbkjLdVBOgch2zJ8lCgloQH5DWEQZQF7yQamMIZcIZqcXsV2P6WWtZopVzNFpPe6PNrJ7r1VvsLCw9rNVm56kU6/RqdeYn6sjLItuvV/u7Xe6635+ZXRa+nh7uRrtxfkGVmQz3G3hAEG7gTTPlTTS9la9jTzPfawXwVymxGnNL551e/XlfmXN0kILZ+D8RZ+D7QYeIDvLLL3E405CCL2wv5B70O9m5zE/s4hV3XBRjuXFxIs5/vONhYW59P8bjdZ534/nGy/nsb/ckT/2ej2b63zfZ36+sSbp8lKinhsfZmaWGBgYvKDPxyruUwN1jNLzbNf+4MH9zM/P8aY3vQ3bXr1mXRnLKwznllr1i37PwCtz3yT7OFdckqxfQKjSMKBbF6Vtz5TENllvJQRBIlm3bZSRtReSrLgBrGsB8lrUQFq6f3evc4LgmjbFGS0fVmSy8JWA3MqbcuUYcj/HkG/uDnJmVN8MtoywTG12VWmgXIgLqWS9VhU0S/0u464VE6/oqe5IiTCg346FqbnT16SkllKHVtXrB3d5hjypCbmlfZ0+l6vGsEaGUaGRvzhF3HI/YzdoBVxbNQZoBdNL2Sngx0ZOKMoIGVEsxAihz6+Tc5kuB6Y9lq2wBFi5+rGEIffKGVAJykv9bvm2y7R/lMrl/wN3y18T5mRzYzv36O2EE8lVQuFw8OHMVT9JmnQaxsU8B34b87PZdVJwup7J1PP1yUmdOUDQXZsFXjzV72I6d/Rwdp5KUgtqfXJ10IDciby+Y9LH8tIY8uQYlYTn75xk9ugKsFrT7dYsRzC6TZdpxKFErrzZk/fn5OCJ8Ucin3ItB8eYJr7Spm55MLq+ZL2aqlriMF5VShAb99nQdenF+u+nT5+9NOFskcjVE9M2jBKmaBcpGQNEiSQy7QfXY8jzkvX53nxfHXcv7qXnd6EGb4lc3RI244VsIZq0hls2gLxfsq73vbi4wORkdm2SeruXauo2OXmKb37zDm6//fYL+tyluDjxgQ98gHvvvZdPfvKTbN68+dwfWBFKXYSf3NgjcrXRQiYCHJuiKf+QrqcVcehkjjLKuDnbI7YqaT162VeZASygbAe/00Gp/oTXdEcblCZyddCS9fWOtZtbuIZBoF8PV5i6hfH5n3syBlhZRxUV688L491BuIapWxi99OueWy/Idvus711dQ64uaF/CtPa0whYqCi/OfaMu/B4Mc/XUQa970Y7jpR5/r+dz8OABTiyf4NNHPkUnXP8ePJ+fvDmj7/sv67F/N/4kxx7mnl0pJVF0Ac/uef7kfVZ8P7jwz68oiUlqyM/2mYcffoDnnnuaI0cOndc+krWdZ7oedaOXdv99u++b84lLgPxCImf25aQSNYlrgCg5QC4sJ2WiC+kkbCTUxkDJyw3ENblMbFzLLAnNYIFE8q6EhRdaKATKbEMYFkjIDC3mJevddjZJBws1GqUCQikGuy1sc4MP5AB5OdDbaZSgsYohj5FW/x1lxxLLGER5yubdj5EmIKpqJmXFlTmm2nSH2z/8HJPs1sdd1ZJ1oQS3tDQgZ/cAxbe+PgPkbglnBSCPLZdrRzUgF8q0OLM9/KQOxCrjRF0WozlIAHmzpq+rUhTMYFd0jElNjtWMDSNQqGTt0OLKNKGdTfzKclgMdX2zUz5BO8gGpvFdl+v3SAPARYluM0QnEhMpupFYJ3hN9TPk+TiznLmN5/s+51uirQfIl88k7Lq+NxdPZ+2DmmEDqeJVDLktBW5cYCVDbhnFw4utIc8nEADqc/3MRmLoNrSxhFtycp9bGxx2cgx5AugS8G0LJ60h/7aaup1Fso551qzYWuXGHp/Q31M3JyE7c+b8AflCb567p75BbK5HAsjHxzfqNwSAMpJ1A8gjK0IK/R2tBcilkitc1fu/v27czVzvLxCQJx4AJbvEUGE4ff2aoWuBrIZ8Lcn63r3P9m0rMQpKALhjEp8X2vYsaRNXq9Uu6HOX4qWFUooPfOADfPOb3+STn/wkO3bsOPeHXraDSf6RWHEekCfzoKBgxmHleoik6woSaebtZdtDquw5LvugcqJEZducmNZzRX7smOkaQJ5jyIOzmH35piRFAXc98gQvnDiV1pCHhji4EFO3xMANJ+uZrlKX9VwN+ctg6qZyJlYX2vbsQl3WCTOAmF8HvNIR5875O6n12bPPPsldd32Nzz/4WT56+CPcNfXie1YDdDrZPXzJ1G39WDlfvRxO6/kk9YsxdltZZnYul3WlFD1zb6+ct9cL3+xj2BsBLvUhvxQrI9c6yzUASxJjYTLZwupzVk+kXJ4xTkqmiwSQD+cWfEvRPEaVjqUEzXYNkUyMlq49VzmZhzAPab5MNj+Y5+XriVHTaKtLyVdYtj6uqlksuMqlGurXup5LbFtYUmIlzqq2Irb7AbkjJbapK6924efuiQltfQJlt4fwzLGagXf6SJ1OLWCyootmrXKFbtRhj7+N0XiIrujR2wjlN78akgfPKeFWMoDQGNjJ54f+nDPxjyCEwE1aozkFfCdJilRwog5B3CTpZd42agFbCAqJOZ2pHxN+I01fJQx5sVpO9xk5Mk2UAEjLph01zHXpMtnKQPTo9l0gct+RVaKzbAYq4aXfy8TEMQJfLwbybHRrcQ6lAiJ/P0qFzDWMxEvRZ6IV5QF5LyYKYp6+/SRzJ7JFRW1agzjLu9L8fiL9WyIDXsmQu5GNrRxWMuS2re+rbjPkwL1T3P0XL1xQn/QEWHsl21yD/s8mhm5Dm8vYjkgNMKJ1ZOt5+XJaQ24ehLzLesKaqzAmemoeVbuwiedsreYO1V7g3zz2mzx99Am+8Y3b6XY7K0zdVrQ9CzNTt55ZmLjSZdYsvpOI1gTkk2c9lnz86Qv/i9979nd5aPYBIHNY32za8gkEnvQoWAWK6IV0JKK0Jj+IVi8Im2Ez/fta0Yu6aZLkQvuRJgmVklNiyB1KX796WAPyrIa832U9jmMOHTrQt62MIdf/lkr6Ob7QGvK2AQLflnZb38Px/ve/n9tuu40Pf/jDVCoV5ufnmZ+fTxdyr2iY500pVgDy5H8sPDM+SddDiMRTRhnJOtQsF0mWxC8GIHMtT5Xl8Lu3PUsQyb6EazImjOYZ8t7692KvpccdWShxfP/j3P71r6YA2jdz/QUB8pQhtyBZc6xyWe9kpm6mq8p6++g2Q+752EFOPb94zl2/om3Pcm721rfRaT0K8y7r5z/mTE1NUqstnfuNLzKSZG7Q1uPpdGdlUe+FRTdnNPxS3L3/vsfK+erlcFp/qU7uwQqyw497Z12jxHGU4pDZ2Wnm5mbXfW+6DwPyh7xhADqXXNYvRV/Yq+s4pJIIk/VWQvT1CcfUlSUMuTRIIzB9rkeXljQLHkUsycWUhbaloNdppJOgsrQ7O2bbESEimfxyz0CU9EINAzDAM84ZRGyutykEtmbvgYoqpK7j5djIV80EPtgNGG33zGuSaAVD7sQSy3y20lFYQNMAroIVYRkA7+39GyBjV1vFTUhha4Y87vDa1vUAPFs5REu18TZVsCzTss3tl6wvjVyLEg6z0Y2MbdyAZxsW3ikgTSIkYciF7IGl/9419S2O5TBgWP1RVwMmoWKEWewnrq2lasaQB44ksvOA3KEbZ2DraONI+v+26+GVMwMuIYq0DQgUBpDXa0t89atfoj5kwEQO/Hbqi4TtbxB17iT2n2O+ucj1S9fzrtPvotPWg1HQ7dBt5sBJJ2L6UJ2jj81x4B7NqId+j/aSHvAc70Z93RdOI809UzPy45UMuWcM+7JjMotMU0PeXOhx4N4pFk+1WDzVzwCfLRKGvDSor0HQ61+4JfXjw5vLLPmL9Cz9e7QO6G+ftYY8z5Dr/XT2LRLdc4bowWnON1QvIrz1IOGXT6z597898dc8s/gUDz96P0eOHGL//r1nb3uWMORONQV6rnSZ6c70vS+e0PvLA/JOp33eC6/Euf2MceVPFlUjIyMUjTdCMS5SsAsU0QvpPEPeClYvghO5esWpUlzhqQAJQ/5iJev6WhTtUpoJB7jaMOSJZL21QrLebNaJohDHcRgZ0YaQySIvyf4ngPxCJettU1bT6/XWlfBfiosfn/3sZ2k2m7zvfe/jLW95S/pzxx13vPIHkzLkql+ynpYoWRRM+YdyPSwS87U4ZchrwiPOtSX1QmvF+sAm8rssdoK+hOtMdwYnEgx0coat58OQm57oTtxL2xv65jUZXjhDLiy7H5DHISJZiPudNGkhztGHfOZInfmJJseePI82jy+h7dlKgH6uEHmG3K9d0GcvVsg4Rub7Tp9n27Nms8EXv/g33H77l16mI8sAtDL3+fxLbNOZB+SvJEPea4V9JSjf6bGSIX85APlLZchXdrBRqLMqEn2//xz27Xv2nPtIZPDDBpC/XAz5wsI809PnvzZ8OeMSIL+QsNYC5DEkNeJCpLLtfCQ15Mn0ETp6IVzpdhk9+gLl00doxMsp8HMjQdjtpi6mCA3IlQH4Uc7hND/BJwx50vJMopAbM3C5qd6mGEbpZ1xsSkYC7zJoTsHI2bs+400D8O1+UArGZT0tvjBmdYlE346wQ50Ndw98CcgAubRcWpWtiEoVaynk3Us/CMBT1QO0wiaW38Armzpzp4hbyQanbmUzlxcslBpn27bL09eVuZ6WBCFc3KiDHfeyGnKTDbQdl5F2j7fWerxj89H088I3vdAThnwou2ahrYhzkvVIOPgqA14TzUwKDuCVtma/iBLtmhmIzHVtNQybYfmmZ60+P4kuNJGhrvVW8SKNdpOtna2U4hI0FEpKbv/Q/81TX/gvKBWhlOL08w8yf1KfS9DV90TtzClAgaggnG0gCigZsTylpfYzx15g23yRquwvTXBMr/uEIReWBkhhTxuDhb04TVpcEENu3lseSlr49H92fnaByO4wvLnE8eYxAtsAq3Uk63lA7q/hsu6acpJEth0bx3Zl7kHVCIgem0X11l+kxgeWUcs+8lh9zcn8WPMoQglUQ98b8/Ozq1zW8xnjTLJeodPR/+9Kl+nOFPmIJ/T91Cv1t/w7c2b9NnD5SPazaKTe9bpe6A8ODlMu6/u6GBfxbA/XGExFIgPk3Wh1Fjqpbx32hhl0V5u/dONuymCfiyGfbJ/msGnlB9Axi+KSXWLQ0wz5oDvE1rJ+jlLJepQBcj/2qdV0kmBoaATP0/fxeoD8QiXrCSAHvj3s7PdoHDp0aM2f97znPef+8MWOhCFHIeJsoZkw5AILz5SeSLeAMPOoI6OUIVfCppGTm7pxPyBXtk2l+ghL7SAtCbGlTXWizI7aBgSCwIGoOkQURcTr3Me9djPdHoAnJLPGtCkwycn4ApJSKseQi0TxF0V9jLIKcs/FOSTroRlnV477a+77FWXIswX+2RjyublZZmdn1v37WtFe9jm9b+mcyqZoBRg6X8l6kmit15fXNeJ8qZEAaNPtj/ne3Fneff7bAz1Wn6/qa71QSp1TKr081ea2//YsT33l5Fnf950Ur7xk/cK3n4DvpL4bzp6MXwn6jxw5eE7lWvLdDheMZP0s7VFfbMRxzBe+8Nfceuut3xHdWC4B8gsI7XTYP/gpFSMSybol+iTrSRRSBl3/Hrr6d7dQxGvVsXttpKXoFPWDWO45uCEIlTHkxUimE25ghYh4NSBPTN2SlmehKxHGzG2k3aUYxZSiIG175iibDUmywBvX2zOs2VA3YPtSgx3UCbfViOz+83ZiiWWAihICReYEX7BiRPJ+I8Hq5gzBmoO7IPZ47f1bGI4HOFI8xd1Dj9OO2oigQaGqFyfd0T04pWy/ldHdvKpkc1lhIxu2XpEdjOlX7pgWTk7UxQ27KSDvmuvoeB4CGGq28excyzLjtJ6AzfJgNf1bbIs+dUCETUC2YD/V6QfkdmFLtl2rSHu5nyHvJZOSAEU2QTUq/YOwkg3syMMzfWujIKTbqNFanCfyW7rXeTzL0Yc/y9GHPmPeo89pydSPW/YGhBAIW3+3jdlplFLMf+4b/NATG9naG9b7IkkEJfeuOWbbtKKLIwT9g+2FSdb7GfK8ZD0KY6adJ6iNPUNpxKER1AmNRD7orT1A5k3dkkkgmSBsy8ExCpDQSNZlx5jKmX+jx2eJH5wmfmZhze0rpZB7jbxSASuAexD7nG6fYjgYxjJGjfPzszRzdYihDPuMz9phG1c6XP7EYDoZWljMtbKFnmy1kHN60dM1bHZSB33+gFw/+ymQNXLWwcFBymUNUItRUfchTwB5jiHvBqsBedLybMgbYtBbDch7cSZZ751jgfRbj/8rfuPhX+V06xRhGLLv68/w2vnXasm6yYRvLG1ipDAG6GRALKMVknU/re8eHh6hUEgAub7eycSanO+LZcgBeufJWF2Kv2dhph2l+hlyS2YMuZcYsTtu2sXDkVkNuY1No11LPxvaVqpyA0BYVIvHONVYoBHWGW667GpsY+fyTq5saoVIY+tGujuuJBzZSNhd/WwC9IypW76kbaqh79ueKZt7MQw5tp0er5Kyn1EOs+fiXAx5Mt6HZ0uAJuuZHCCXawByudgj+MRB4kPLa5i6XSAgz5/DOgx5HEd8+cuf4wtf+GuaK1yfzxZP3jbBI39zjPmJs7tCxyvA1vky5J2ceqBzDiXBi41EyWUb5dzCRWTItVnZS/Mc+MN9v88/vOvHmDqLlL420wUFS2denmv0csRqyfrFVxO8VECerG2Kdik1uO6tUe6WRKKIqFYHcF2PKIr62uCtuY94JUN+8SXrQeATBAFBEHxHlKhdAuQXGMJaIZNCkRhnKawU7OYjZciT2lizyPbK5VTGroSiXdSTUimwKQZ2NjFaFoVYpROubwe69Qr9k3BoJOsJIA8cSeH6y6jYHlfMaEapEEVpYsDBZoPSrJlfGkcBsWFNB3s+Xix5h3OUmY2KyStvJi5kclVHSma3bKG7bQ+RbeN7ufp6K04l6yoIIQ77DMEaw3uIn16k1HM5UTjD7+z4X/Qsn1bYRPgNlG2k1cPXIZ2c47txuS9bDsOjl2WvG4Y8cbt3oi5e2E1N3XrmK3ENY6ZWZKEtMxknGfdCtUhiwCdtu69+PhQ2cQ6QT3X7Jc22kwFyRClzJjeAvG9SEvr/Y8uiUVkhAZINvKiIF5vzD6G2MJP7exMVazmv39aZ+AQkN+a0/EbYGtQkiYmw19FMSS/EQjA2py9Mt6A/55pe9zjGuMcqp9fQK/ZPCusZrq0V4QqGPC9Z77R7KDtEWTFB1KEWLBMahrzeWVsWn2fIExlTvoY8kawnsirZMdc2AeamB7yaXnuSVtMd1EJOOtnuXzhMtE4gVcxYbyx9rdls9H230N+XvB21eEPrBsrH+seP0qRCmYVk4rAuRsfoVvS9eqUxtjrfOvJODpBHUdjHFucZ8oJdwDW+ET27l/V1X2NSTSTrQ94IA2sw5L0oc1k/Wy1ZL+4x050mUhFfm/wKi4sLBC2fre2tlOwS2yv6XPcMXM6QN4SFhUJRC2r9LutxkBqvDQ0N43n6vlpp6vbia8iz+27ld3opvkciZcgzA1XImboJCyu2ETICIeiZdYEDKUPuKJtmM2MVI3uFZB0oRi4nm5MUFkN+8oEtXHdKjylFU0IWGYPVuFQh7HXp9bp97aMA/JQhzyTuCx39LCQ15PJCklIJQ273S9bzjDJBDswmNeTrAXIz/ge9teeMBx+8l4997H9r1jc4O0MuTzRQiz3kwVqaQE8P+0Ik6zJCGFARtGzqf/115BqAu9PpEAQBUsbs3//8eW++Y+aYAy88xze+8dU1S1/8TsRDn32h77VkDXfO7edAePscSoIXE1LKNBnpRXp8ne/NvyRWu9vtP7e1gOaFbP/pxSfpxT2eXnhi3feEvr4n/fa3n/0833glJOv9NeQvxtQt6WrjUrD08+9H628n2UehUKBa1YRXfp5dKxLJepKoX2kmezEiXzrxciW2LiQuAfILjRVXTIpSWj+mGfLV/fUShjwB5LEB5G65kmbWYwt8N6vVHm66mambMA7rhvkL7ADIWpP0XFMvHobIOE4l674rKW/dzGvVEBtapo5akrYwc7AZRctBepVBlqtlpHJAKQa6ehCwXEXQuYHapj0Eo1l9tCMlR665hmhwlGBgmE45d94jl0FJP3QqFii/Sa+ZDSrNgV1gfv/i6N00Hf0gtKIWImjgSyOpF4KWpdnd0Kmk7drKFhBnPf1Sybph952oQyHoIUwNeeIS5hhAolbUs6SS9Thh0m0QelvKsvvk+j1loazswV0MJvtrZ8QIGMWEyPV5TgC5n0sGSNE1x+/SKieDsGkXJ1sUomJ6b3nSY3E+c9tWsoGShhmRIRCmTHRgMn3CKlGoOOm5BN1OX396t66Pu2MAuROb5FDChosCwvR1dworGPLe+bERSqmzMuTdZjbIdnsd6mE9BeTN9toZ1DVN3ZIacuHgGkAedCV3/fkL1E6Z7YQSFcZgJmc501lzARAn7Hi6w/7JPPENGPVH+1632/3Pfp7VbUdtNoQjdOmftN518PXEz2jmIXFYd/ZcTs/0Bd01pGXc3W6H+BzGTEop2iaLvNRbTBdAlmXheYWUMS7EBTy7gBPo8aTjdFKGfG1AnpesZ8ZrSWuybo4hl8hVDqxJJPXgAHdOfo12R38vrnIpWiXeuPHN/MHr/4h/ed3/F1vY6US8HCz1udYHMkgl68PDmWQ9kfa9lBryOI76ZOqXGPLvzQhMkkxBn2TdytWQK8slQt/3XdO1w7VEypBbyqLVzFQ4oW2h7P4FRCHyONk6ymXTZSwEVd+ovcwcIt1EwVYi6HW4++6v8zd/85fMzWXJ2cTULVkfADTMPZ9I1uWFsJEm2S+snEltHCPCDidaI/zl8dcw38nmtnOZuqUMuR+vWf5z6tQJwjBgZma6X7K+1gI5SJj0eBUjfiGS9Tzbv7B/gNZdz9H91CdWvS+/SD9wYO95e0ok53z49HMcOXKQ2dnVNaqzx+osnFxe8bnzG2/yIPxcwObFRK/XTedGV7lYyiKU/V4HFxork5sJGEr2E4Yhn/70x7nzzq+cc1tKqXQ+Ob6ibDAfyffgt6OXLJF/peKVqSHPtvlith/lJOsF26g/z0OyrtchCSA/OwBOJes5l/WL/R3mr8Mlhvy7MFaWiLvVnyLpL57v25mPQtqnXN9McVLrVR1MQboUCgT0ivqFkZaXMeQAlpWy4aEVpt+csmw6xWySCP1u2oM8cCUVpwKm3gjAc6pEaQbfYkN3GK/bBGHx7NV7zOsd7MSwxZUUe9v1vpzMoMaJJV0DcKXr0ksAuSOovfc+ot1v1X+LBM37HiK02yyPPU2vOE+ruBFpWMeGnT2UrbCF5TfoGSmtI6CFbtfUKW/ENdfKFgK5kDPaMUB95+CreEPFxpMh5V7GkCfhVowUPZamFZn5vF/XMuVIEVs+k1MTYOnFfGw7fS7rrUggnBzLTcypVlafFEcKyzGO1lbGJiaSdT+XjZRWwkR7nNrYBcvGKb8Nzc7HVIKcIVDssTSf1Rsr2UwBuf69nfbuTp1aRYHKSAHRB8hXDzoJQ24bQC4xxyUK2K6+ZtXRGMsRjO/Uv58vQy4jlS7C0hryHCDvtLNBvNNpUw/qBEay3lonY5k4lsMaNeSWm/UhP+OweKpFZz63yOlEGePdjWFFKzcVK+ShGgANWy90VjLkxwwgTxhyu6j3V+3pa5MAyQREKqVohy1GoyE6on9/ARHKlDUkDuv27j1pDflAvnPCCmAZBEHfBNWLe9rTAlgKMkBeLJYQQvQz5FYBy9eDSMfpMFTQQNsPV2e5+yTrOYZ8U0kn6LpRh05OTraebD0PyJeDJfbP701/L4kSlrC4Zfx1KQs/amTrS/4S7RxDHq4C5CsZ8gt3WW+1mnQ67VWLhJWszqX43oiDs0krMZl2NIF82zMLaTm0zfzVs8z449opQ759YYxnpmLuedtbObZnD6Fjr0rYF2SBqd5xdswZgGtYbmFrGbxlxm5ZKNBrNTl9Widl7/rLP+OOP/iPzM/N0DJjRF8XFrM+SBjylX3JzxYqp8wTKUMeQ9jhUGMDc36VY22TjLRtMOuCc0nWUatbYEK2EO502qvanq1agCef92NknJjPmXnrAgB52GuhgAVG6Jr2csFTj69xbDk1VrfD8eNHVr1n7e3HKCShYQ3ruTVYEn4n6m+Tw4uVrF98QL4SnCRKvZciW0+2aZkkj+/3+M/P/D/8yoO/QBD7LCzMUa/XOHbsyDkTH924kwLAE81j674vUWfIWK3bteU7LRJAXjLGrt+Zbc9yDLkxVz4fyXo/Q34OyfoKhlwi+8oAL0bkkxHfCWq4S4D8QiNn7BYR6zpdVoPwfCRtzxKpuDRZ8l6hkknWDejzS6YlWddBoCBXR55I0gIrIOmeomyboCBS0Bj2eikLGjiSslNGmNZHAE5xhNiwYS42RVVix6R25K6NjWMhsayMIZwqbaJg5HN9hjSWTWgWwsp26ZWMI7cjQAiEqYFtTReZ/1+fpDl4lMhtERROg7CIjKSraWUPQSJZ75j99QPyTTj5yzyfPfxJT/jLhl/FZtdisDxEpZeZuqXvSwA5IONsY5ZfT+VvrcEj3HXP7Vib3oJT+gG6Ra+vDKERKIRZhMlAL0qONzODuDiUuJV3sPM1/weWuzu3dzPp5wbCnqOzzcIuMDPew/2B/wOn8GowrHS5l+3Xkx7NpUyquBKQI/UxRUGmkBCiQHWkkLLzjbk6J55aXW/VNQkd22SHlMwYcregwdHYdsW7f/tmNl2ufz9bDbnv+3zjG1/l9OmJPuBeGjTXICdZ77WzBUi73aaek6y311mc9Jm6SR+pZFov7lhuypAnCicrzvWRb4WpdB1AzqzYRy/STDqKpypaThi2+t9zrHmUUlSiHJe11HuLSaL5I1jCZmNRA9XEGTyQAZGKGFsLkIuIyCQl4pMTANi7LqNjmLFSvYGdJOJyi+rZ2Wn+4i/+Fw8+eE/6Wr4/eyfqUG/V9DYMMF0pWRc9fdwdp8N4aYPeR7x6wstL1vM15Ml5LvmLqQ8BrG/slrQwS2L/bA6Qq+LKtzNizFxOHj3OzSduphxmjHfLOEuvZeqW1ZCX0vefLbPe63X567/+FH/3d59Jt5vEJUD+vRky8UcB6GPIM9WatFxarmHIjWeK8NyUIR/ujiK9EnNbt/Lk61/H3MbN6RyavMeLXXqtEzoBTz/LHY9WKcpiur9TU5OpSqbV7TJ/8jhf+tLnWBrbqs1kc4A8mSx9w16tB5bXjOQcc5J1FUtE1CEw6rXkX2wHjOLvXJJ10GO/M/ssg195H/biQeI4wjfeN51OK22Tmm4vt2Cempqk1jAJkHaYStS9oj6WC5GsH59e4Fmu44/5Z+zdqDu9xEcOI2s14jOT+Pffa46pf5F+Pj2UpdSqMGln59JorGaWg06EUv3JwvM1dXu5JesrwUkh1mPsizV2C8MwHZcHBozqq9fh3ulvcbx5lInWBA1jeKuUWjOBkY+lXHL3RPPYuuN7Pvnfewmy9VfSpT25TqWSnq8vNkOulOpj4V8M4E+AsZOTrJ8vQ16p6PM6X4Y8AeTQv8a5GJFPRlxiyL8LQ+Ran0UiyfyufxmFUjhJjblQSCFS2Vq3UO6TrAPE5RVthXLZ6iS7HlohlunzrSybnrAJbV3NHvm91PwlcCW92S5f+/43M7Frlz6eSlb36mCDU2DTrAbkcWWQ6zcG+F72sO4rZ27m5GrUQi9bQCvbISxYSMclMvK1xOilM+cxtXUbkWdk4Uaenjhcb1y+mhFvlKHuRoIJDxE0WJb6gXWFYFltBKHolDbi5NQHTj17kGzLQSBwDVPuWYJKp7eaIS+W0gWGjLJtCb+e1o/Hhp2Ny2Wc4msoRqNZn3l0b2xhEhpRRysKjpsMrVKKOJQIa4DNV70u7U0LpCx1X2bS1Zlty0jrA1//LWHWi7lj9KRHdyknpc5J1vW+k1ZhMmPBhWcYcr39My/McvCBCVZGJ2XIzbYMIBeiiFfWoKi9vITjWtgFC4Uk6sV9k2Bzfpapg3uZnzjKoRf2ceTIQR599CFCw2jYroVXdtJjTBa83U5+QGxRD+qElul7vs7iJC9ZBw0AE4bcEXZaQy67Ru6fe29tfz8oVDP9A7wyE3jPCVh0anr7zew9SimONY6mcvW6V2e+qBcpw8EwA+4AA64up0gAeXK8o9EwXdGfjQ6ICDumveCMljVGmzaliTpvYR7HMFB5pndqSidWnn/+GRYX581+zHGar2WhpV9PMu0JIC/EBTyrgDJlBx2nw8byZn080VqAPJOs52vINxqGfNHvN8dbb2JOGPLtlZ0AzDYy2W2B1e3UEoZ84dgsw8EwO1v6c7ZvFEZegVKplAPka0vWgbMyLidPTuD7PZrNxirzvO+ErPmleOXDS+YaBSiJawxDs+SeZsiXiwkgz3pyq1Sybthuc+91qtUUkIdCPyOu9Bib0wmvWKg+ltvatBFbZb+fns4ppEplZKGoE7yWjXS9vhpyHD32Ra4Z/S6oD7k5F9teJVkPjXFraAC5cByEAeQqWvsZywPyxcYShRf+hsKpeyge/kLfIrjT6aD8/rEjka1PTp7ii1/8G740+U2aokvYCVNG3E0B+fkzoIemZjnAlQAsjI6nrwdPPEr9X/8Gzd/5LYInHkuB7/bteuyZmZk6J3sbdE13Dysb68+fIV893iilVjmp9wPyl58hzwD5i2PIk3HUtp2UIV1qZ4nchd58n2leooBaLxLTUoBG2Fg1ByWRB+R++wKegVyouk/wv/cR3TP5oj5/oZGA5cyU9OIBcqUU3Tv7fY9enGQ9UySeD0Oe7KNQyEvWz1FDbgB5yS5SNPu42E7r+XO/VEP+3Ri5GrDY1HELtT5DLmQGyMHI1c02mt5ACsilkbOLylD/5xNzGZFJ1gM7wC6Yida28S2X0BV09lzPg489nJOsw8TTx4hch1O7duIPb0FVNROGAhsLISwGTdZZFkpc8bP/Fd9NTOrgWCHXXiy3WAhybZmU4xC50LnsWnqbr9F1lwaYy9ji6JWZ0VnkKhAxrjnvweZ2NpY28iOHfpHhB65lfnmQJaUBjSNgOd6E5Sg65Y19DLlYMffuqPVwTQsGz3Yo+d1VDHll6j6ElwHy2PQMt/x6Wj8uLb2YD40sfSDYRMHOmPViUmctXWRXT9IHawfMuarEC0jXbufDsPhhLjMZmwy6ZWkwEhgDEsvWoMfLzfte7OHXNDBSQFCWhIUc85tjyMMcQ+5VHGzjQq9kD9TqwbeTMuT64KWRCiEKFAcSQK4nvENnHmdx4yPsm3+eX3zgvfixT6/V4LYP/jvu+uMP8rX//v+w94FvArC4OE9gvAgcz8It5DsCGOCbA92dTodaUEvbnvnruKyvzJJ2415q6qYZcrNANIDczd03cyt64cqZFQsgs5iqWy2WHb1AiFrZMc5PTfKfjv0y17b0Ym6psMRx9ARXiSqMiFGqRuaf1JAn/45Hw3RYzZDLrmZw4zmdGOsN6Pvf832YncM1bHk+mdPNeQE88sgD6XW5unY1P3nyJxnyh6i1NABeyZCX4hIiFqgkCeXFjJf0ojSKV1/zfpf1bHxKAHl7xfexnvnKsq8XWTeOvJqKU80MCwFPuqvev62sS2VCw2yMBPpedHr6+x0eHkYIkXNZ9/vce/OA/Gyy9ZMnsxrEEyeO9v3tUg3592Y4VuKnYTxdjPmpEyaDskVkO9QMIO85+rnWgNx4upj5smhUF71SlhD2jaGno1y2zet7f+ryfmM2ZzgDigAL9YxlVcUyspjd38px+hRswgDygjGHJI45Uj/Ef9/7X6mb53m9UH0kgFnvyLiPIU8AeWDFfPnMl/V7ziVZB/7Dw7/Fp7qH9DH2aisAebuvhhzgA198moWWz+nTEwD4qsdXvaeJVJd213SrMCXsX983y5H51Yv8WCr+4uGTfP2FjN09PjXPSbbp/Zaz69j5yB8jTeIjuO+edJG+adMWHMdBKXVOt/UEkEs7Z2S7FkPejoAkCZ8kq1eDmrvu+hof//if9gGYlx+Q94/pBfnSGPIEkJdKJQqmTd5yJ0uOa0CeXaNarT9xvjLyDDlkpMjKiPw8IH9xDLmc6YIfE+9dQsUvP1O+cv66mAy5mu3i7+9fA53P9qWUfeuPxCdG15AbU7ezdFhJVDCed/6mbolk3bMKlGy9Rr7YvcgvSda/y0PkQGkksmz5uqEkTu7vfmKAAiw4FWQy8RvJuTsw1vfxNFFv2elkHVohbikzHfOtAlGpjCyUODU1iW9qisreNnzDvtXGNzP+9vdjXfXD+ogVYLJcT15bSrPigecSF/RAsDg2RlcMpA7MynZSYapfzBgtZTuEroNyPbBspqbOpAz5me3b6FRdhLSxzWB2k3cvAFIpRlo72SIvY7SrQftccyPNaBgAT8CQuBp7bA+d0sY+YLUybphtYhtG2nJcnHg1Q74wX2RpSCcYZGQRj2iGO2HIFQplagEDI6UfiDZSsDJAPpiUH8QVorb+/P7lvXSjTp/JTKGyAmAY2XgkswkitpL2V0ZaG+h9e8ZN3s5lxT3pEddbWhGx/XK627bS3b4n/btSBpD7MUGuhtzxbFyzcFMqSHuM5yNhyNNSeSNbFqLA0GZdMtBe0oB8sTGDsmLavRonWxOcbp+kMTtNHIZYjoOwLJpGihTHMYuL+nNuwcZ2LGxXf0eJ467fy7UGa7dpBLVUsp4kKFZGvoYcNEOer2lKShjo2dhoz4EkhhKFizkONdvtk6MlDHndaqaAPDgdcep5nZHv7J3ixs5VbPe1T0Ddq3PaP4VX1c/jhnCcqmMYctM7O3FbH42G6BrJelJHFxBh+QrVaoKRRyctz4rdHuHsTArI86Ayv5A9efIEZ86cph212dbehqtcNvQ20GjrBU7KkFcrKJQuf1jUf/Mtn/HKBioFfczRGkxaPQXk/S7rm4qbV70X1p+Yl4xkfaQwyrby9pR1gbUB+eWDV+LFXtp2Z8TXgLzg62dpaEj/ntSQB4HfJ8UrFApYCSO5jrGblJJTpybS3+fn9YKzaL6DS5L1781wzfOZjAw/7L+Kn/Jfh2vuL4FFz3NpG8l6W8R6/igWQemSl8iMNZ6plezmSqZ6tnGwli5bFvU8VR16LdLJg+rs/fljAYgRRJXsWVS22wfmlWMRuC7tHSM0BgYQccQnjnyMO05/hc8d/+zZTz4F5DbCbFO3PetmQNz821UBRzuGcVsn6ZUH5F5U5NlYjz2WX1/VvmslID94Yo5f+PQzHD+llStCCVqix3PuBE3DeB5o69KXOJL88QP97B/AnQfn+OgzD/L+ux7m6HybIJLMLy0RGBPYbqmEEVWlbScBgocfTI+vUqkQuzopcy45dVKSJXMMeSNXNpiE34lQSSLZ0991sGK8UUpx/PgRfL/H5KT2D4jjuC+RfaGSdaUUH977If7b87+3rtR7Pcn6i60hT8bRcrmcJlDrnQyAz/fmXjRDDusD8tB/6Qx5YiRIKNftzHIxI2PIL75kXXXCTNlr4nxqyG+77e/4q7/6WPo8JOstx3LOU7JuALxXoFK5MIa8YBcoOmY+vugM+SXJ+nnHn/3Zn/HTP/3TvOY1r+GNb3wjv/7rv87x4+s7Kr4i4WSTXiSSyVn0z5b5kBKBwDIsegLIlYJpK8vMXndST4KVof5Frp1kqK2sZUpohRRNHYaybAJRRCb13ErR7rRRls2IldUwdz2LLgFhyupLpJGYHN8xiGUm065tExiG7+A11wBwpmLqjkXmIt8r5QG5S5BriTY9PYkwWdAjV2omsdTZSsVMVEFBccKa428Lj+Ii2D55ffrZhe4mupFhwoTgqsIGvFf/Gt1yv2R9ZYjCoOkTD7LgYkc9/bvIFv1T4U0c3vXT+pgjQTyswbnwG8hYokQMifGecRqvRCMUc4B8SCY9VyuocIyCGidSEc8tPUNkALlliz42WF86A8hz9czJ6Qj0NmMj73aMSsJWOUAeexArOjuvJDKsNY6jawchrSEP/H6G3PUsigP6XvGKEZa3etBMTN0sKXXfXcOSft/PXsvVb9HfX2tpQbdWMyY1SZKp5tfomBZUYzsvZ/ctb0LmyhkWFvQE7nhGPm7khaFhEYLeCsl62Ej7kEd+zMLCPH/1Vx/n8OGsPcxKyXov7q5wWTfPaNdelcQZNItksa2iZZ2hTE3V9Mb0cTXtdgrIna5g3936GQiXTLLBLGSiYoRUMUFZTziDwSDVVLKuj3OyfZpSXKAkC3SMZD0Bk74IcX2beEbLt8XQEJ3ElKzXJZqZTXuR9zPkCeOgx5CDB/fTCltUQ+M/EJVpm8mzWCzx/qf/I7/9zL+l4epzOjOhF7gdp8Plg1cyaAC5jGUqRwNd65rIAccKYwy5qxnylbG+ZF0vokYKo2yrbO9jyB3lrXr/5YNXpOcDmtkvRkWKgb6/hocTQG4y9H7Qd41s28F1DfO0DliYnZ1Os/f5GB/XSqJLDPn3ZkQVPU61Iv28lJXHBjWofV0AhIXvFugYJVUoFD4RlIogY23WaJLtjlGz+EmXDyQ9M8Z5sYcjLTpFweZjbwM7ew58065RqRVyMAOYo4Hh9CXluH0KNuVanNi9m8XREntvvAERxxxtHAbgqbO0iQJWtD1bKVk347cp0wstmZbbrWvqlgNFXlxixiSFxRqAnBWAfIyQhVaPxQWtHtoT6ST0omgSGIY8MG06HSV4+MQyz52pc3Kpw+G5FrFUfPSxp3lr9VGuH/k6v/+tI9x3bJFyLjHdLZWo7JGpekGUK1AqIefnaC/qMaurXE609IkuLJ2dvV2LIe92u6uAVdCNQCUMeQXpFqiPbEqT2KCZ9QSgJe3uVrLXF2rqdqj+AseeP8zk8yc50zq95ntWgpOS1Ou7szHkuj3c2mUD+fkqGa9b3cyvY6Vk/ZnTT/Klic+vua12u8X8iRlQpF1o1jN2uxg15Cp3/8qTZzcie6mh67sTD5SLL1mnE6UYIAm/1kaeWf8eiuOYM2dO0+m0OXTI+OqohADJGPLzNXXLu6yvlxCKVZyqHgt2gZKtr8XF7kV+iSG/gHj88cd573vfy+c+9zluvfVWoijil37pl76tmQzh5uqoVV6WvM6lVDFShqlsPQHkIRbHRbbY/DdfjPmlr8cMDm/r+3hi6LTS1K1UHUjegE8F5WbAc//UIuHgCA4ug6pHtakftkWrSfuglpfJOCQ28o+hsIowg37X7xHYVWpDQ5zZsR2F4oXhFxCJuZxtg1I0BvrlcqGb/Z4w5LWhIeY2bQIFxc4WKsYcq+E67HVO0bQ69EpzVI5k53ymey2BLNCKFUopDttTzJRBWi7OGlmPhL2nlCkL/IEiTpw4hedbj7kEbgI8MkBu+XXiSKFEDogYaXE1GsKjkr5eUkWq/gjXzb2O8diiIq9lW/0qTnxUcXqvnqht10oBaLZvvdDKD4WWuSeEMbRKuqdZBkAnyQEAGxvleshSVWdzVKZaAFDS+Aa0e8hkQSUKOAWbq95kam+dCDFkyhNy5lzdYv6oIoRh8TdfvpHBDRqYRH6PXquZSpptk2CqBzW6JvtfHhrmqre9E+Vli8qlmp7AHZOgSOr9kokyn6Fsd9pIFacMuQy0nLheX+bIkUPp+1pGIm0Jkxzqk6w7aQ256Dl4KwB5ktQRAy5iozErnMtAV8KQN+0Oy7ZeIBQFdBra0dyrg0TRM1/W0KAGhbOWXjCW/XJOsq4n75OtCbb0NuETpqZuIyO6Bj0kwkIgp/V1sjZsTDPHpW6XaH4+BeT9DLm+Bpddpheoy8tLNNsNXNPRoRSV0hZ7vu1z38y3eG7pGWqFmj6mCc0kXbHxKv7Vq/4tFXPMFha1IGMnakGNWMUIBKOFsX5Tt3UB+dkl66MGkCcySAA7Xj1+bixuYoPc0PfaiD9CJdTPR8aQJ/4Mfrp4cV0PIURafx+u4zJ90vR+37Ch/1wSQH6JIf/ejMlrh/j3vUWmu7qDRgKKRQqOBYFdIrZifMN2N0UXUa6CitPEG3EEpnNB7CVzQIRvwFoyftdH9+jmGjkD0W5bj+k9cnLnOKKYlJXkJOrSqNOy320aQ/pZXR4ZwZJRCqaONA5R89dmIOtBnUNL+/U2sDg5eAuP3/J/M3lKm7qFKyTrsQVRYsSebyEUxASRRMayXzkWlZi2tN/NSkDe6/WIVnR5+N3v3851gzE2iq5yuCzWMv666GCZMTgw13/rgB4Hfuu2A/zMrU/y3r98mn/zpf1YvePsbl3GjY097J07wm/f/gIlK99mRRCPgXOdJgaK//Cn8V7/ffo8zNx2sH2EptLbPz51dtm2301KFvrPZaVs3c9J1qUsEYxvJhoc4fHHHkrfkwfnSXI7uWbJ2Nbtds9d1x77af33tw5+g+tr13N1/Woeeui+NUFRAk6EmZ/HLOPn0ZvnjtNf4Y/2f7gvcXv06GFuvfUjfOUra4PoPCBPGPKun2s915un2cyB3Q58fuJvVm1HSslXvvIFogMBu5u7uWpIk0brAvKLyZADcuLlBeT57/HlYcijtMtSsjQKZUR4x6l1jevyTPbBg/s1MdPnsn5uyXpm6ualpm5SxusmvIPctjyrQNnR+KIVXtzrf4khv4D42Mc+xnve8x6uvPJKrrnmGj70oQ8xNTXF/v37v30HlashD+OMCRQ5Y7ck+w2AkkRBK2UUe0V980bYzObcphXwI88otjb6b2onSQCI/rZnA0NZD+RIVNNe3DpkWl+2oTXNqOn1vECNYMnUb8k4ZchHe9WUFe10OnStAfZfryen5YEOTa+Z1m8r28GRilY1A6nKdgidDPguLMwRuR5Hr7gCgHK7ii2LVJf1w7RgCeaFqc912ogwW0i05RgK+Epvgb/tTXO/+wJ3e/twhMIVqweMReNULisbaIouE9YcUbGEE/VwwmafbF3gEtplFELXkA/nJesyrR8HEKYOsBxV+gA52Lzz0C/yA1M/wA93XLzgGnYtXY/dLXDc1CfbbibNBtM6xwBymWP5LZGAaaOaCE35gllYsOJ8Y/OdilimCRTlJHL0xH02WeAIwMXxLDbuNr3cex16BZ3xd6092K6HKrp0c8XqSrbTgdotlXlo8WGCgj6O5bnMVCh5Tz3MGPLS0AiF0Q19vQGX63pB4Xj9gDyRrAf5fpi+jyWttIZchSIFRMmEHsRBWr80WtDPQDfqZqZulotrJOuW7+JZa6sqRMVFDBjX91aHybZhClJA3qY0aGq5LV1v3WsHjHYH6OAjUViWxZUbtILguNQLArfrrpKsTy6c5qa5W7jNe5LeCkCeLNzjWf29WBs3pRLEku+DlDjmNsizv8nksW3bDgCWlhZpNTKGoRyVCY1J4HycLSBrXs18Xu9j2+h2BtxBbAMMLNUPyBOJ4khhFMdy2FDcRMWpsKW0Ne0PujLWm5gTU7cRb5Tt5R19knU7tmn5Ef/j3mMcmNHXTQjBNrG9bxvjvXGGAs3SDw8PA+TangVp0iKR+a+U+9dqS7zwwr50EZrUj99442tSYA/9gPy7pX/tpbh4USm6nEaSjXTG4yUdk21Cq4w9/2bGlZ6DmqKLqJRBxmnizYpC5IoykJIs8mP1d+itWjYzoz1G4zeixDqeGaKestZ2t03JcVa9J69KAohdm1ZSq1mtEueA/pb2Fm7/xhfXNDH6y6O30g1bLIzt5FOHjvLM9l20BnYwcUogwg5Bwowbs7nIygxpmz09bpypd/nJv3iMX/j0MzRb/efkxSU6AhqWhfDr1Fv9zFx7hTHbQBzwniv08z0vq4yrEpYSxELimXWWbxjyrdUCri1Y6oTsspa4zp7hoROL7DT1+rayGSodR6AoJmy4YXT9QZfqb/wmpZ//55R//hfx3qTbtnYTouL2P+GNE88AMLuYyaV7vS4PPXRvX81zypBb/UBqpWw9b+qmZImorOeN05MnU2C2tNQPyJVS6fc2OjqaluScy5Dq9559P//knvfw+PyjzB7OeqIvHJvj6acf12ZfUYc/P/i/eXbuWdrme7F8fW0HlR5zpzpT/OG+/8aXT36eh2bvB+Dpp5/gzju/QhSFTE6eWrP+O5mv8iacvZwyqdZaTjsIABRlkeX2Ekop/Njn3ulvUQ/qfSam29rbuGX8dQCcbJ0klqsBd7+p24t0Wc+16lMzHVT3RQL784h8ydXFqiFXSmVzWDciNMRTyQD+UMTIho88uHaSLq9cWFpaYH5+Ll2DuZaLl0jWz4shL2LbdlpGl6x1emHMH9x9lIdP6Hsnv4Yo2AV2V015aG0vFzPy19b3e+dMbL3c8R0NyFdGkkEbGho6xztXhxAX6Se3wO97vHPGbl07l/WRijBspW6pvpFyh8qiVs8mowdepf/deceX0hZmkSXx3KQdStbDNLADhkc2pNK1WFSx7GxCHi0OQ1EP7mPNBUZDvY1FFokSWZuSREZaOhSUEVECyNt0GWZyp17oHxnWDJJnEhHKsnFiSS9nmIQQ9AzzDDqLOSlDTl6mnd0LXc2AD9X1wzZpiawnu3Fdj0VEz+7gF+bplWZoFJdYLurJyBcRwumkRnDpdbZ8VGTY2PI497r7ucvby77CMkrAdQf/KnU318fpgrCI7QKxKiGrujwgaXuW1I8D2AiUiClGBWyZZ9kdNrY14zwkBVbvSiqhvh+bC/p6Oq6Fm2PIW14dhKcNe3Ng1Up71xWwpYNjJLz+kDHys/sfT2lq+60IRDJ5ebrGOzF1880AJyzNELoFm0Iy8Pa6dCM96AoxxPf/0m9T+rm3oiyyXutS35O25+G4DndPfYNGQQ9ai/PZRJ44zdeDGr2G3mZ5aIS6AeeJJ0Gzs4xC4hYshACvZOTXvQghIFwhJS7GRSLDHonITjOonU4bIaAbZwuPEU+D2kD6RKmEysG1HSxpYQdeKllfCapExUWYY3lg4m5+4b6f42TrBGGnS110qIsON269hdhkkwsCTk8cx1MuDeMvMDg4xLv3/CN9nq4BkR2omFrDVthCCOjMthBY1K2klECkYLJrFm3N6UkOX3klX92+lclJzcqVzfPuGsVCFIWmzEGlCYqxmWmElIRhQGs+yx6X4zLKLCQmQ51sePuWH2LTxsxgEWBgYBAhwDb3mqUsloOldLxb9PXiZ7y4ASGg7Jb4xPd/hj99y8co2N6aLR/D5klEkoXPjZ1J27PR4gibC5txcgojSwq+/sIcn3nqDH/60ET6mZFYg/6mo8/tisYVeNKjOjDAhg0bEQKKxYQhD3IMuYsQ/YBcCLj77q/zrW/dycTEMYKgl7JOu3btZuPGjCVPALmUMVEUXLz5Y8XPpfjOjErBJhSCBJAn40disooQRFaRYPld7JZ6zGuKLqI6gJAxHYwsOwyJhExVRwADlNmlNOOobJsDeyRDwUaktfZCv2d1sQI9Dtq9DtXCGoC80A/IpevSHMjm5MbwEEIqxnpjvGHuDdRPLXPXXV9DKcVkrcvRhTZn2pN8+eTngTJPvPYniO0evZIupel0FYTtVZJ1aUFslhStbg2lFL9/11HqvYijC20++1i/JNqL9Fw65diE7SW+sbf/75Ntfd2aBSPv77RxenrcmJNVylgMKjMPmnKBhCG3FfzHd17FT147yg8UTvB69zR73Abbcxj/VWN1/usPbUEqKKoeo4t6291KGW/HBiq/8i+Qp3ys6GrCYjVdL/3kwx1eO6XXQrKWgeQnn3yUZ599iscfexjh68RMJlnX51I1asY8Qx6FWjmQtD2TjocySZUwDJma0o7eeUDe63Vpt1spiCmXK7kWUnrenpqa5NOf/jhf+9ptHD78AtKUoT21+Dixivn9x/4LG5p6bDs+oJORjz76IHfe+RU+ffCTfPbYX/FHT/9RKoO3DSAvmTUkoWJzaxNbOlu4f+ZeFhfneeQRDcwTs7a8mi2J5WVznfF4aioxoM2VYBk1SLVaRbpmzg0KNMI6t5/6En/wxO/x77/+r3j0sQfSz2zobeCq8jWUnTKhDHhkLlMWQNZ+Lon1GHKlFJ84/FH+n6f+fdpRpO/vK1q8ylMX30QviSRxLIRIfUxeqmT9d576LX7lwZ8niP0+hjy5d0Ar9eLH59ZMPq9sBXrwhX0EOYY8cUA/v7Zn+j5a6bT+0IklPvfsFP/Xl/bz2MQyvjF0cy0XS1hp4uWc5TYXGCuTHd9u2frqkf07NKSUfPCDH+Tmm2/mqquuuuDPj40NnPtN5xHz+bZnaBdWYf5LbuWu000dgYWShGEzx5DryaQUdfjHxz8JaIOWv32LxVv3x5T2P4d6/R7wdduy0VIJWh2U46YruNAK2blzF1ZwP7JYxhUVLCtLAgyVNjJn1YGYza1Z2sXvA2KWrRabTHZdSElsslADUQkhDNMsA2rFDWxHUWwuMeOeAgVlr0C766NsB0spYs+AVKVACPyclBXgvpkpYtdlqFbHjjVoHF2aAW7I5y6QdoAUEbMDR7ELy1i2ftBbUREHlXLT0msixDAAvlQULIEvwYlK4IAqDrNkzmHSWubBd/wQb/nm3QxssKgbTG4ZF/bQLYMdMbJ5C74sc3DxTQzaTh9DDhBbAaXAw0aQDDUi98hUlCCMioyrfpBTKDls3JIljRrFBaq9TVqOmFuFJ86qiAJDjOJKc3yDPRCZI296PMUEkNvIxL/ANSyl6pj6bzPAGGXAhk0DDI5nEnLZzAzftuy6kq3xaZjTyR9b2inT7pXLjI8PsBjO45Rixhuw78wTkHQMEDGWtOhZbcKOkV+GZSJTBuF0msjqEBKInQ6VwW2Mjw9QHSoCdTzbYXx8gDheDcjHBof1NiKX0GRde70u4+MDdBLw75QZKg1AA9wyJJey6DvEQvCm+i0MWKSS9Y6ESu5yDm0ZIMCiCfjtDqqiOB0dY8jfxG3eE/gdjz3OdjqEDFCgaAlqM/PsYpCG0Oe4YcM4N1/2Kt5+7O3cd/o+YmLs2GbMG+SfzP8Ym5qbGPgxD7fZP8xWq1U2bNDfW2gMVlqNGs/d9Gpt8mgWL1XTySBZahcKNuPjA321VwPPPU211aI5NEg4l2WVi3Exrec7GeqF5I9d8U5+cOsP8qEPfSh937ZtmxgfH6BeN5J1ZRF7PcbH9XjZW9ST8daBzelr42Rjackp0VlR12U98WHGbBfGfzMdd/3YT93YL9+yk+XlAR7IfaZguxxt6uM/udxN91UMC/Tocbp6mutq16U1g69/6+vYtGkYgCjKEmaWkaIWiwXGxwdSsF4s2pwOjjIzq5NKy8tzjI3pMWtkZISdOzexa9eO1Dhp9+7tuK5LGIYUixajoxdn/rgU3x1R9RxiYZFxFgaQp++wiC0XYXtU0fd1Q3SxKwNEMpOsiygAIbDCLnFBP2MuNgUzj7QrHlWp1zNqPUBu93DrS/heEaexhLtJADqZLMIA5XppyZqMAyzbQ1kW3Zx7eG14mErY4fvm3pw+Q6dPn+SZZ57kdx6Pafohb33jV4mlpLb5V2kP6IVyMj+3uxB1O+kVSBlyGyrFYWCJdq/BJx8/zSMTy9gCYgV3PDfDPyVLihfiBJA7XB20KYr+BXFomPyBLRth4gSy1WJ6WY+3xZ1PY5+4mmFVpkYbZdpjBo5pVRpJ3nXdJq4pNvjaCT32vbNyml4vl9j264zIGgDbw2mCjsUi47TdCsKvoUqbie46Db7E+ge/CRxHxCFP3byJ12/+/wBP4oiQ2pkZhrZu4uhRXZe/MHWM8Y/+nzTe+Sf43begUClDvmnDVlqtQywZ5lgpxe1HbgO2oVeREJUs8kZEExPH2bFjF4s5Nh40S56w4Z7jUCwWaTYb7J/dy4YNG/nWt+6kXq9Rqy1z/PgRut0um67cQifqUIgLXLVwFQLBcmWZZ8aeoVAtsGN2B8eOHaF5uom9xeZ04zS7e5cBGUNOKLix/mquWLo8TcI+aT3Fc7WnAdiz5woG7E08d+Qhjhw5xOte98bskvt+mmT+5IGAVrPGD3kgc6UMoqe3OTAwyLSaYSCsUg2rzPfmOHn8BO86/S79HRNTGRyi0WhgYyEXY/7BjnfzuROf4cN7f5/Xbr2JR+55HLvi8In6J/gRfiM7jnUY8r86+gk+dfTj+nz5fX735t9L/YiATLLuas8ZebKJffXwmtt6qZEw5I7j4nlGwfcSGPJQhjxqEhVHGoe5qlsmNPdcsVjCsmykjAk9SWGxR/TVk4hBD3mmjWqHuD+8ncaUvgcrqkBb+Bzet5/CZj0fupaLl0jWo3NL1pNyhWq1yuLiPO12i6nOGSYbeg0RS8Vv3XaAf/x6fe4J+/6a8VuwsDjZmmC+O8eG0sYXfU36rk+4GpAnCbRvR3zXMOTvf//7OXLkCP/jf/yPF/X5xcUmCwsv7WdxsQl9fchzLHlOsh5YQWr4hpKEYSutIU/M0EpRBwtFyy7zxNh1LAxa3HmDvhEGOolZicpkzQnTIyKkkAiqWMZNu+pY2DkmeNaqE4oYoQTjbZ/xomZ067bMXN6lJDJZqIIoIo3Z2MLiMl2hJ/vjG1pEykyGjmHqbRtsU6MrQZg2MIFxCbUCw3wriR1FvPrZwwgEdtRlqLGcSsTyETttVHk6BeMAPa9O4GZsaOToyUwiaRi39jhyMJ2bCCpVc85gK4vp0TGOXHUlw40z6TZcA3Aip0xImcWOwxOtn+XB+i/w6G1H+2rIQWe4la9Q7eyhjS1YLs6iUNgIgnZENRzu+5wSUG9mIKVZWAI8lLUi/2XpqnghCoxZG3CNhHc2nkZYg5lrrrnlpMmYWrGXMuQqVSYoUF0ai0mGV39f9Xab5VoP2ygt7GaSgfWYn2mAr7/L2DasvGHaledy5MhJzjSmaJf0vg5PHsyujRXgxkVmG/Msz2hJ9KHHmpw5Y8zJAh/HDMKh2yRWMQsLTaRh1muLHRYWmmsy5EMlncxw4yLLxtFWswaLTBoX7LJTwTbO3HO1ZTq9HgPBABN3T3DmnjP8h5lf4A0Vh4J5Xpsr6qOaMqJryhKKgd7OgZlDLC4v4YsIEJy6/xQL6O+xIKB1Rl/bBJCXSlUWFpq8e/vPoISi6RmW/IEO71v4Cd45+Xqe37uXsZ5mw6rGELBYLOH7+jokgPyQAGnblG0nXQxUk0V2R++vVmuxsNBkclJfg0KhSPfgYYYM8yIb/c+WZcakF5raiGW7s4f60VMMDeTcmZXLwkKTVstIbLE4vTiVjnkTC5qpGbRG0teOvzDHySML1J/5GsXE0T4XXWERPfVpIBt3j5zRCzLXcgkagtZC/0TYa4fsO10DYLre4+TUMvPzDYKmft9kZTJ9Dhpug8ENY+nx1Grd1GtjZkYzSpZls7DQJJnivrj/y/zb2/5tuubde2QfR47oRMX4+EYWFpoMGLNE1/VotYLUVGdqaoHJyfmXPH+sOZ9ciu/IKHs2kZVjyJM/pJJ1C2U5FBwrBZVN0cUplXCjDJAnZqlKZglzTzl4Rh3iCI/t9auBTOKcMFhJdJ0e3vIcA4efxfa73B8fSP/mtMx4nwCIOLf2yEVteJjrl2+kGLs03AbPjz4PaHbU79Rh/Is8tnA/O+pX0R7cQexk81dstxHK4XSzRzg4SuuKG+iaTiCxBb2uKf2KI/7m6TvYIJr8ypt28fYrxnBWKEAzhtzBQlExXi2WWV9I4zVS3qzVKnPtJt1uh5iYWftZAIZMC7ooaY9p62ONzFok3zUhUVjFZpwtdAscPaVB9NbWNGUjpV5SlyF6NeSxBpixuSf1eNAqBFx92a8xaA1iKwtlWZz+oz9k6tmjKcNX64QEOBSO3UHQjVBWqO8VBU5Zj/sPTtxHN+ry/NKzfHq/Hh+F6V8fF/U4JQKdXJiYOEYcx6n8e8sWrTL88v7P0zL7PPXEgyyf0Sz3HUe/wrPPPkW9XqNcrnD55TrJMzl5iuPNY1y/dD0/furH2d7R27lMDHLDsUGeLD3Ju9/zjxGeYCAY4JaFW1huLSNNwjcB5HEQs7t2GQKBdPT1uXrhKg4d1vfirs3XMPmgAiVYXl7sq30/efI4UkoKlSFeqAsCk8whd5tWIv2dVquDLNoaAA6EA8x15whn9T3Sc3qcKZ+Bhc1UW7qUqT61zC9e9avsGbgCq27xkT/7M/bvf57nHn+KhUZ/MmNmaZ4vTHyu77W7z3yDW4/8hT5XLB6YvZc7z9zR956EIY/GjaHv4vpM8EuNDJA7OW+U4EWXTeW9Io43j2mG3DwLjuNSMO2T42v0mkAeqhE/MYeaakM9IPzicRoHdKniFfZ2inj4hPhL+hq4IqshX48hV0r1uawDqdP6TG2an7/v5/jq4gcBsAR0wphPPanvaz+0CSLJgDuY+gU8tXjxWPKVDvPf7jry7wpA/oEPfIB7772XT37yk2zevHarnXOFUhfnJy9Zj1CYttSI3KWMrIjIsK1CKsKoScUYgtSN0QpSEmNx646f54nBt9GZ+HVu2/g+AIabiQlMJrmUZsIKrRCBRbUyimMMUDa4xlE8cX42D9yIqmDt/HdUqjspKhclBItjY+a4ZFpL4loFOiaL3Wy1qRrrsUbBSHPdQUqGEVe2Awm4iwpYSf9uA9K9VgY6b3xuH4Wefr0Q1LGUopqrGUsYAVU6iYetQW5o6macNnFO+t91avpfy2fSXuLThQeYsOZTk7R2Se9nQJV4Y6Qno7033IDMmd25xQSQl/DHX490q5wJdK1AbbqzSjIoLR+FQ6Cyev2vXvNxPn/jfydyjTN5T2J3C32fc1wLYYm0lLrrNgldwNNMusjVySnbAVFghAyQL4hZhDWcAvKKqQVMWBBbljNAbnuaTUfXkQdJf2pTs/6P7/tJ/uC5/4pjzr1oMtFKFHj8ufuZ36fNyBLZoTKSdb8yyKc/fStb5jbTLhq3yyD7bqVo80NPjFDvLqemblFYZHlZTwBW0EN0muZ6t3BcG6XALRpjwl6MUqzJkFdNOYQbF/pMtTqdTupcXnGqFHO9KUMZcm3tWlAgI0mHgIotGDay9LZUfZOaKjlgjmUw1tf3VOsk7U62v8iPeMo9os/dgmJTX7uGSCTrwygFN4zcxA0jr04dzMOcpG1q4qTu+w38aHgT2wY3c+ONN6cTUyQkHXxODerfv/+q6/iZn/mn/OiP/gTjA8Zt39RfhWGIUtmkUS6XiSdOMJirHV8VQtAVXX7oSInSv/qPLP2jn2LwaGaAU60OolTWhs1SFkv+MkrBiWcWEHdtw4tKjBXHUQp67Yhv/ukB7vrT56l86X1UerXsuwsrCCXoCoGz+ALR/EQ6bi6ZhYGuOxerzNJkKDm6kCXgJha7NJtNZCyRSFpuC3tEf1/7RvYRqLBvXE6uZyKNtD2Xjx/6ixScTLem0sQIQHOpycyMXmhs3LgZpWDr1h2USmV27boMpTKX26effpyPfvRPePrpJy7aPJL8XIrvzKh4NpHIFE2pZN38XWCBcPBsi7IBVfNWA8txcKRMa8htA7BimSWgXBwKuXQbRwAAwkRJREFU6PnKkYJR6dArzjBU0M+lVDYqB8pDqx+gzxYCOomzeKvW9zcrjgis1Yza/IYNbOlqEPPc8BxHBo9QGi+jlOS6wYfwRh4FBD/Z+wf4hYUVn9bP1D2tEH/jdpRboLNpFzEakC+09PgZ2wV+VDX48cJBNi8+w2983yauHS33balggNeUae9WNM9nGJrrVLBwtr8Be8MtdItF7g71WDdVmaIS688OS7NGMAy5Z0qBEuCSGDWOjmZ93CeqE/qzwTALszqhOVqLcDZq87YF+VoOPzTHscee1B9wBF2TLNgcjbIz2IxAMKD0nGM3Ig7f/3S6fYXgZPQe5Jl5gm5EbFqeWdLj62e+ro8zcHlm8SmeXHicorkOoVFXhEV9fxUWppHENBp1Tpw4hpQS1/XYZkoIZ+anODinPZRUr4toGYO3ps0TTz4CwJve9DZuuukW/f6ZKR6d3M/lDc1sO2GVSm0X8RPPcMvhEbyOol6o89imx5BIdrR3cP2y6XoTR1iGQVQKHOkQ2CE/9tM/iXI0YJaRZGh4BBpVLOXg+XqtdP/9d/PII/ezvLzEsWN6Dj0lRxB2i7isVUi2KX/YXNpCOTL+OCVBw/gCVcMqc91ZvKZZ2716mGdH91Ja2kDB12P5zOQZbCz+z02/wQ9M/yCOb7wBEAyZNpnKpNPcsMQf7/+ffH3yq8w1fT7+5HP84b7fB+Bn97yXX7zqVwH44wP/o8/NO6jrazAxaciK5fWZ4PVCSsmjjz7I6dMnz/q+PCB3zTpbStlXX38hsZzzgznWONInWXddF9esKaPrBnB+fBf2923CunGML1/5CPcMPgES2sakdeTNu9l9rV5bJ2o813IpWkayvk4NuV636O8gmacTw7qZ2jRSxSzFxwHJP33tdn79LZdx807zjIc2//qL++gE8csiW88SBfpaf7sl69/RgFwpxQc+8AG++c1v8slPfpIdO3Z8uw+pz9QtggyQ53TYsYgJE/mzkvSiDsMmq9sY1JOXkDFduwhCMFR0kL0dnC7thuFhyqb/cuwJXHMDKwPIAyvAE2Usy+Lqm/VkMmRqbt0wwlWZLneAIlR2Y1kO40YqtrBh3BxvnC5WXatAg6Tvboeqye63jKxuY2lTKjVRtuk3DliyhIj7V5Vh7FPp7KTS3M22ySVi48bu+Ro0DDQ0SBuUJXqevvm7ZT1BLnt1HDNRxXavT8LXdiN6hHSsLhPuGboi4KQ7nTLkLaPKHlRlro63smFkjNhxOH31tSm7EZV9YrtL6JSJC5sJA8VipOvcpRQ6q50/F7PoUWSS73aprmucjY64FEgtlciF7VoIIVJjt67bQrkxdvkN+trHPlJlrWWEKDDMKI4xd1uIZogLm1MTv8EVEhpLVnOA3EIIY+wmOwS9xF3eZC2tDndMfoV5pRdVidxMORanZ48wdfC0bj/lGjd/pcFkaAbqKxpXEFX1+Y+JbIGjbNhQcxBTNSI/qY2uMr+kAb4V+GAUHNL2U5f1tO2ZMTSMjcakYgxtinGRigFCjnJXtEVr004BeTmtXfLjHqINO9rZ+JAwVMPm90CCn7tV44KNKOljGY4G2bF8Dafbp1LjNNcwOTW7SZeAohCMBPq4EoZ8aHDInLfg9177B/zIxh8GYNlq07T05L0wra/HkCgzrCr8xOvfxTXXXI9n6gUjFXPAmSS2BKOLi+y8+lo2bNjI5ZdfhWWSZ44Br1HO5wGg5HqoZmMVIHfKOSWGq/ihZyW/9ndNogN6IReHetEklJXW/SUMs1CCeqBrQZ//xiTl6Q1cPf96xgu67nB5qk0USHodOOW/hrI5pu21q/lnT36Q151+Fz1L8Ez7p/jz/3Qs7d+eGLolRnwr3VW7vZBuImFUcOx4jcUF/dmgEKCEYuMtW3j2smeZrkwTxP2gI5noZ40kfcGa5y+P3sop45Ld8TuM97L7V8iMSdu4cTOnWidZkkv8s3/2a7zznf8AyAD58eNHAXjssYdXOSVfir+fUfFskqamOszgkdaQWyAsPMeiaCs8ZRGKmHp3GWlZtM344/r6Po1VDpArGyHtdOthcZHm0BEq5va04wKRlS1ux1tX9B3bxu61bLLaHK4PYbf7VRYijvFz7t4bDBPVHBzExmZZFpn0yyAg2KrPcE9YxJIWO/1fwZ3Zgl/UvgqR0segrBoAx7oj6dwvi2VO7tpFZMMbd+9m/7W/wJEbfhnHrD8mJo7x1Tv+kl9+7XZz9fR1KwcDDHbHCWffy1K8CccoDt4gngUg9EoUb/4FYvsGHnrr2/Ath4bbYPvNu3i9q2XQQ6aGvGeY8apq0S1NseAc4NSpCVqtJrZt86M/+hM4joNv+Rwe0qx4JaqgIoWHYHjz77B5QM//beGzbfIyts/r8cn9qd10zbpqLB4kRrK/9AKDBpB3Nl/OhD2Xfp8Ap9QPc7r272kd3gvGcM6KC7QNUKmEFZ6Ye4xnFp+iaDpFhKqDtB2kp8c+u1WnKfRY+dBD9+r9j43RK5l2mcEQCw39/YgoTL1/drZ3Ekcxg2Ob+IOnfE60PSxLe7A8s/cYrnIRwmZ48TUUm1Z6V480XT7y9B/y2odimoEeK/c0jeIhigisEHJt904FW/DDLex+1eXpa/dZ93DnM/cAUOhpKfHU1CRPP/0En//8Zzh1SidIHquVKW3/DPaWLwDgSY+KM8CGwtYUkPecXurHMuKP8NjRY3iRRyxifuiaH2FH7RoEFk44gBV7hGHI3/zNX/HUtx4BoXD9EZqOXiPslq8GoOVpUGorGy8u8d/3fohfvOOP+MTxP6Qbd3nVyI38ytX/gp+9/L2MFsboRJ2+vuYJQ75o3PPpRn115bPdGb4x+TXkyvaE6FKt33nyt/jzR/+Ep556jPvuu3vVe/KRzPGO4+K6bpooz/edP1dIKVPlRjLvAhxrHNWmbmZkc103BaJhFGJfM4Lz5i2479jBHUP38wdbP8Hz10zRLuhjGhgeYvdu/b3LhRiUbnvmRA43LdxEsLA6EQgZC21ZVtoxJmHIG8s13jD3Bi5r7EA4LbYNl/jnb9jJr7xpKwBCuTx+qsZv/N3zXDNwMwBPLzzBzMxUmnh7KZEA8tFRY7J7DnPElzu+owH5+9//fm677TY+/OEPU6lUmJ+fZ35+/oJuzosdIidZ3zJcZLBsJJs5yXokIiIna5PSjTsMy6RveNI3WtIzDN/VG037MyFwb7yJak/fJEHFTh8YaZjy0ApxDCP62nf+RN+xVWQhNTwBODZ8AmXqxLcYOUknMXKQksDKAHnNtGAL/C4DZjFRq98C0uP1G95IwTCs2DaxARNrAfLl4iwDahfl9g4KP/uPKF9pgJgBUsO1GgDb5CgtM/AGBvBOVk6hzHW0HXRf8FzMW3U6Vo8FWy+Ka1Yr7QfdckwbKlVCIHjTdd+HENAcHUcWikSlKlPiIPXhA0ROGdXrsnymjSJLYCQMeWI4F9kJ0MwATmJ+5lRNsiJa4xGyY2ZmphCufm/odXVfcsdcwzjGDhNA7YDwGJSjqfHfXDxDWB4nodiDhX4Q4MhxML3RlRUjkj7psk3Uy2rEYytECcWgO4i/Qjsoc0rjX9v+L9k8oKVsKtaTV2wGzoIssKFwmd5vnF0rhEDZDpUzyeLPRgmXjjFnsaMwza5L20/bwKUu690YpVSamBg2LayKUZGhigbnijgtpQCdLErqlTVDnpmJDE8N951fW+j7vWSk6oHKAHkgFXGswLDnG/wN/PjBf0EwJwiMU6sXVSkbY8R5q0HBgnE5gEKlgHywnEm/K22X7Sf0s7dc6vKNYc1WNJY1WN6MAcEVoxhJpGJKctrSwPPKI0ewN2S1UdaoBuR2J2GAsk4IAEXjCDpcyGqohZRsO3g8/T32W/zKnZLArdL9iV9i4df/J+3ya/T2oyK1h++muPcTDD78X/RryqI9cSe9Rz5Pz5Q37Fp6FeNFjRaWp7IM8tHemyiZ52/Pol78XDP3RjpWgX2dH0UpwZO3naTTCFhsz1MNquyZvZkD902l55Cwed1eNpnfGNh0vzbDQ1/YB8DA4CAVp8KNm2/Cr5pyHrkSkBvjSiOVTGSPbeOJEAQ+o4a96ZjnWkqJEAJZUfzag/+MX7jv5/jo4T9NHWQLpX7lSxxHPPTQfVyKv//h2BaeZ5NJ1g1Dnk7/FjYCz7bw3SG2mvl9bmma2LLoGlO3gpH85nt0K+lwwp3j5nA3I8GYxvhCseTpBJ8XlbCMnwjKYs/S6/uO7dUzP0bx5Lt5Y3ObbsOWcwa2Y0mQ639tdcawo+zvR+INyFA/yxPeJG1sCrLA9sYVjJ0eZKryLLHbQcSSZaGZzdjpoFBUpEnmG3fsA6+6nnZ1hB+84grmNtxCe0Cf65HBIwRWgOpITs1OAND29BxWisvcMv12hNzI/YFOfDkqJDCmjUFlAGE5POkcY3FsBCsKeHTTo/yjK36OayytZkvIjcAKiK0em3q7aA0dxfcWuP12DfY2bN7EyMgob/rx7+eerfdgl228YpZY3xSP4Vge0pjCLloNFIopa5k7i09x76lH6O7U48REcZJ/ddnv89nxu1NAfmhU0BEBNhZXxlr5tmyd5AH3AE+K/XSG9BhsSY8yQ4RCS6L9h5rUT0xRjCpIERKODNDdqVlHy4+w4ogFcRwlVGqmNToyytFIm99VwyqDoZ6XrChMSyIApKe4vbGD56ab/N7dxygP6fnj+p6epzx/REvO40zKPdzyaB07yVDbZdPxJSa9bK1hxSFzI73MQBY4FQ9zx4E5fvj1P0an0KFttzlZPclofStx8AJOY4FxdTWve90b2bhxM77vE0URLeVRc5vY5eMpWWUrm06jyr4JUkC+bNVYKC4QiYhqVCU4qdcjy4Uau4f2cF1bPwvLpRmqjSsQymF5eZEojnD9YYaWr6dn1o3hgk4Gdd0WgUmQ/PDojyFVTGfgSzjVIwjl8u9u+G0sYWELm90DOhkx0dJg756pu1CBcdpXCt+sJ1RNP98LvXl+85Ff40PP/2funvoGK+ORuQd5ZO5BHpvU64F6fXmVTDofeYZcCJEmhf/jw/+e33nyt9b9XD7uvfebfOITf8b8/GxqpAowVTsNkSIyZYO6Tt0QfbljUkox15tDCcV9O56jZdQekRexffsu3ZGlB4PhIA4OjWeWubx5Oe5+d02X8sRh3fMKWTmeUX3GtYjt7e1ct3wdlrPIaMklCAJ8g1d2Dg8yVHQ4MF3nz+9sUbCKLPvLfPG2v+GrX/0i7XaLIAj467/+FHff/fXzuj75SM47A+SXGPJ147Of/SzNZpP3ve99vOUtb0l/7rjjjnN/+OWKnGR922iZgQSQ5+xeYismNjI2pKIjuwzngDIYhtxIPa7ZpG9OxxJ4N93MllqL4c4MU68u4SXuqUa+HFohlpkUisUSbZVNMmNiOJ0wAPyBefyn/5S4fpobwodX7F+m9auuVaBdHNELA6UYMmCm4e/EP/r/8ktX/RpFIyNWlkNs2Cg7LtCnplOKmYFTqII591e/nrikJ0/X9A+8+tAhro+3cUt0OcveMvmYrkxzeOwZAIbdgbTOrmTOcc6q07Z6YNrDSDugaxjWxME6yZ5vcEYZ3qInpKg6QjSoH7jYbdMrVFC9Lgun+90yk7Yzbc9Ik1KmYTUgr5jWZJvW6J98qvMMn//8Z6kXj6JQFCsehZKbMv6WcnIMdwEhHKrGBBCgRYMg+VqlpHnmaN/2bVnEHrlO/9kKwcpan4V+xpAn/bz/+VW/QuD0Z2+lkyVS9lh7qBaH9TbMZB3nDOVGfZ2tDHOLSn3sDmOz+r4XVhXp+FgIIhFhuUIbGgGxHeiEBOAVE5f1mCiQqTnd8IgB5HGRoeIw0opXmex1Oh1akWHI3SpFJ5Gsd6jW9H1WHNDPS1v4KBQnrXla9AiUSv0efAVREKcu66fsWRpDLzDa3Jxmj0XsUXX0/TNn1SlbggEcugRExqugbGXPdHTvGUZ8/XstbhEM6OvSM06y20Nj8lg1pR25FluLln42qt0mItfDPWHIbePqG3a7xKdOsjytGZKiYc5HNm5MmbtKq81APVtUbZpvI60Cj775/TzSvJnnD7i4wRBDS9sYaFzFmfufYuD+/0hxUjMcFhb1qM3Sg1kmf0vzcoaFPpbaTDZhTfivp2ySNJuauwEohwMI/2dpxLq0KOzFPPXlCc48eop3nnknW/dez767zrBwxiR+jLIoqPe4NrCxBbwqMB0pjKP+ducKvvSOr3PF4FV4xphxNSA3Ch5zHSaVXsDWTQ/nnfWdOErX5TXHsud+ZGSML535O3zpI5H89fFP85+e/m19ft2J9H29UR8hBMePH2F6OvOluBR/f6NUcEjm9dXcl4WtwHMsAneIbVLP4WcWzoBrpZL1YreHLRy2kRl/BrHFsrefm+M9/LT0EMYfZcmUG3lRFdvMB3ZcYCjY2r9nMQLCQtgjgJO2LAWwoiit8QXoLJewjepMojgejyEDrXY5Vp/gpFkgv275BvbYJ4idLlYMr3nmGdqWPq7QixgsL+EIF0sqyqcOIaKQ9sAAx1/9Tr7+zH5aA2eI3DYKxfHR48xWtTJoYlb7jrQKeltlVWBYuoSFGvsLOgExINrMuyYxVypzylpgr6NlzduOPs7NlUF2zB9kW6wBlodD2awJamPP4kQbQAmsXBvF+fI8KlZsuSfm1uO/yx8c+78YDzKl2SZKPL84zXOmxC5GcV/hOe70nuU0NV7Yu4+9ZzQo2185ypw3wQul4wwoPb/UTDndoeoL3D+s11ZH3UkmTXI1SCTrcYHL5kbYsO8kIvKpxAN839L3c3N3N40NTyGHhtMWtW7b+IrEDfaOZOzf3JmH+dzRx1n2lhEIXHPuIgqxTHLEt3we3PAMx01Z1WzT56klPY4OGHWk2x2hJxSxzEoSxuslNiybvyvFpKylCUsRhcyO+mmLVWFZTMlB7j26gOd4uFe8hzNb3sy/2XYrhV6PsP01ovbX8A99gdpCg5/4yZ+h4eh117F4jO07tW9BaIVpguvHD+3gPScHqIb6+ZlVM0RWxKTpsLPHqJo61S5KwviCnmce2v0FpHAYmbmB3VtuYMTexVDtegQWYUtL7gtG4TFcHaRcLaJkl/DQLRSPvwsZ6CRFofHj7KjuTK/HZaa91snmCY7UD/Gfn/1PiDjpZiRo5QB5y2/xe/f+v8gFl5vO/BCPn3mMlXHftJ5XvThjQJLOHmtFHpADVAw5MV+f5ZG5B9Ne8meLuTn97M3OzvQx5F6QtE02637XTefNvHFcO2rTMzL1ufZsClr/r2d/g1pcY8cOfb12tXYRHfTTenI7sDl0KPO3SCJzWM+ez4QhT6IgCxTdJazOAn/xF/+L03snABgulvnIz76aHyyd5HXtp3l74SephlVkqLsHLC0tMDs7xeLiPAcP7qfRqBPEPl89fRuz3ZmzXqc4jtIEwphZa12SrJ8lDh06tObPe97znm/bMeUZciyBZfoqiRUMeWybTJGSdOJuChTT7UhJ1y5RcCx2mTqrkmvj3ngTFnDTqTZeKQfITQR2kPatjmLJosy2O26NMZhr0TXsNgnPHKBzz3+m5O5ney+3kFQS39RIeVaR4U1b8Q3wTFiAtvIIYptuKCmaTJ2yHcLUXKyEiLPrIeKQemmZyDNmK2GBrjRZf2kMr2TILdHlFHGZLy6mtXKjLLFdLjFX1SZSA3Iw7VO8K9aD8qyo03T8vnZBDVsvrutJXa+5zrYPl+/WPaKjgWGigeH0M7XhMrX6DIsrAHnCkC+ZBZIwkyo546rE/GxwSO9nowHkvpHfSyug5uuBoO2cYW7saQYqFl6JFHxaYmMKyIVRG4waJ/rIClFCIh3j0htHmXM6YAkbodxU8iytECGM+kJ2iE17HEQB38ge37r57XilfC91iJ1sEbewMI/lJOdoWrbk7me37aHKA6mEPgnluIw2DYAUFSJXX8+m26Rt97AMo6usEGGbuqWcZD3sRSgzOYyYDGUxLjLkDYErVwHybrdD2wDycl6y3vNTKf7QRr2QbQufGVHjm97z3O8eIFQgDUvfk4rQl1C0CYl4zD2CX5pnq78plZoSF7B7etKYEw1GbYEQgiVzn1VUEdnJssFytsuAKuLYDnEc8+bdP0iPAN+4HGyOTMLFMORCiDQ7DeBJQU0scfupL6WvWcP62XGWtAy7+/gjHP+Xv82xfXpSri24hE6FSrHGiOkwcGq0hUWOMafKgevfSkQRxwopVQVXlR7ijU9/CTcc5GjvzXS2vp3gpl/R+1QWS8UBTvuvTrdhKxuOGNXMdDJhSUJVYrx5E15UYrSbeXsMn3gLABucY1iWYupwjbgWIxBERhUzd1o/Y8MdDTaUDHlHx+Utm4bYap6pisEhm3duwhYmoWMA+cpe54niIIljsU5i7avso1QtUzBj5pYt29i8KQNHI+Oj3HH6KwC89/Kfx8Li8flHmO5MMdGbMGcqubtyF/FOo5xZkZi6FH8/o1zIGPI4MUpL/7VwgIItCL0htpp5bmZhBlFxiA0LVeoGvH3zz/Gm0g+k2+3GNj8R3Kt/sa6gFej5LfI1ILekm9YYK+kBDsocR2QJdre+wvc99rtsbzyOW303lsqVNMUxrpfNV3ZUwjKtAxeCAq+/fDPX9rbxvrl/wFuf/yneMnsLSdsTEbtU61fwticOc+XRo+yQ+jnxC9Cq6Hn56mgLdhRRmD7J4PIyduhrV/mBCQDGvDG++KrfYtseXT7Uai+giOkUFlAiJlYxlmHwQ5PctkWXU56+rq1yiftdvai/eqrJppkz/PzJpxm49/+m2M3WQkNGkSDtAIHP8OJNbFx8DVgxoLg/uJfe8QUG5mwGZIVd7U2MBtla6YFNX+YUVYR0QekuOUeFBtOzRQ1oIrMesi2fbUGJjt2jYGXlQdcvwGsnW1wx8VMA9OQoakUvQ1t6vH5flx/Z+yg3P/QQdrsBlkWHDqEVMCTLXLHs4C5UKNX0Z71IcGToGQ4MH6DpNpmy7iG0J3hw84P07Exh2aKA3WvTjgt8a+u3mPdO4A4/zuZr/3/Y1QOcCfvXnF4wwmE3phdngHz33A6umt6V/v4aUebJ8cexeh2c+hKzoxlDvm3bTjzXY64V8MnHT/PppxZ5+KjLZ78+g4pzQFF1OfWNT/Of7tzHF1u7+VZ8Hf/y3T/IoP8MKG24mbDkheEdMDiGjY20JPvbWio+YfWbhe6cfAOf+OCniOc+gh/tZbp6lEb9C4S1j3PmqQ721C6IQ5TssqGg7/ey6Txki00sNk7g1z/Ctmf+nJ87uJ/3TP8M7WP/mvmp76PWzdYZu6qXAXCyNcHe5eewpMAzKsrW0NO0k6qqZZ+/vPUPuOLY5fzYoR/m+079JEP3XYfvB3QbNRpz0/TiHo/N62RNMc7u3YWFefYt7+W99/wjPnroI0S5/umZZD0B5Po+L5mkygvL+zhXJKCy3W5Ry9WQD8d6nIiStZibZ8izNeZCDvTXjT9QYAU0ZIN7p+9OZetX1a8iMArJ6Ypeozz6xENpd5d2u82hQwdSRXNS9graf8dxHCIrwjfr7BGnQXteJ9IXjsxRjIp4tsf2qsV2dCvW7y+9lh8dfle6nSPTh1haypIOR48e4g/3/Tc+vPdD/PnBPznrdcqfc8KQXwLk321h9QNy2zHumDmGPLIiAk8/WFYU0lI9XJzU2A0AA8gHiw5bB/XDWnIt7CuuJCq6lH3YMScpFHM9sNHZxTgyBnF1H1tlQGtAFQm6+ub0LZ+xjp48FDBd3s6ruplLtpCSjpGFuFaB7bt20VXZIKicYjooLHcDisaEwZMQGLbcico4YTaYiDAktBWBqwfCbuDQkToTWVT6WDrlAp7Zz7LTQNqGOecEv7O4lPZwd0M3BeS7pXZcnbFqNHO1dQBNp4FE0jQy4opZeNONuf7KG3X7kVIlrX0DaA0WWT51kMWTet9OaBhxM0nUjWwr+X2lZF1Jj+ERfQ0qZjHTGV0mFpGpv1NpD0nLbbPh2Bineo+mJnWWqtAqGEBhgPVwpFmLMEkCeBrIaUCeXeOiW0YgKHmJYiFEWGbxptpEYcKQe4S2jyVshr0Rto9cll0020HmZI0akGfXR5GVjSUqBjG+OXN9T97nZL8Lq0JQ0AuaudIcDbsLMk69FUIj4Uwl670YvxPqYl5gZCQHyN1hhKtW1fTrGnL9XeVN3QIjiQqsgKqtr0tb+CnzXLc6BErhDJu6egWRHyMtOGbNpSaIg1Ela1ckPcJF/d3MW3XMY868yhI/QcuoHCIJrRCBYHxMf4+xbTNnFm+B41PEBSeieOI2RE9f03zGeDwssTQo+J/7P8yxxhFQksHndHsyYWrXIsvi1I4fRpluBI1wnAPX/gLV1oNsMMZLjULA1NZ3pNvtFrcTbv5+AN5Y+QT/fOhneefQh9m5/QyF3hIRRZ4c+wOmRn8G0IC8jsNUqFmG00Paob19z9fh6D00jcPs1UUt3R5qvYWNrV19ppZ25BJbPeZGn+Kyrc/q7gWJ4aRJ2iSmbrZZaCgrpoDgDfN6LJ21JZFZUG3YlJkqevbZGXKAQqlI14wHPavL6Os2ptL4rVu3ccNlWbLhhDhOL+5xxeBV/OJVv8arx7Sc/wsTf8vhSNedDuwcpOf0eLD8AL/8y/+SHTuyBeyl+PsblaKDMi2yUlO3xKlTgYWgIgWhN8ywKlNWHrGMOVk0JQ/KYteWH2S8uA03l7huxfCH8c34tgLp0bO29e3XUk4qWT9l2cxssHANk1Pthmx64QHK3XnetOlrbL3mVdgqK50RcUQpV8JixyXKne0M1spMNTby7sEB/ku8iX+y+C7eFV6DLQsM1q6l0tjD6MLrKPfG2by4F/eyt/Grvf9Tq9SEYM6ug4Ib1R62lq/AbdV4/YP3c93Be9g2uD3d39hSi01f+zVes+tG2nYbUCyPP82A06M2+hzd0uyq6zznBhw2xqG+Z9MTISOywqvnBYNxkdf1fOzWGWLTGkkplSoOhbS4ovQAl1uD/NhAmV/svINf6L2d75+5kvmHngPg7sHHuPvm/WzaZcwxRciTaoHYLiIQWHE29+26bA9nrphKDToBSoMbOdx7Ez+x/zeIwwFuiHby9uB6Xv3ss+w5fYP2vclV7u0Kc2VHcYHQqVAcCWh6FuVTR3lb53K2xiNcFW7n3cHreVvxrXguJK1Kt9fLWNKmboUcLk/zrcEQvGWumigyvu9xipPH8Bp1XjWs59uFmWWasVlnbfkCbaYY2fwIxXYGxkQYY8dFrrxxDDsnWSdewg5r6a9bmyE9Vady4gB2r8XSYJh287ly9+W8abcGu3/y4AQA126qskvZSKnnn4XylSCKeFHEVY8/zz9se/zc0Fb2fez3+YH7B7lleivv2PojhCLxWFK4S7P0Wsc5tvsEp00S9LKtN9GxTT2vgrHaVty5+4CYuH0P15ysMNSpATHF1n2oqIbfuBW//lFuMG3KylaHKDzJgaU2cXs/oFAmsbvh1OPc7BURSPZP15k9+gITTz/CrsplgAbkxw48wU89mrHn/tH7aQaabDn99DMIc4vE1mmU7LKxcRl3/sUD3PZ7v8VtH/wtHjzwtdR5vJBTb+x/9D7u+/onmGtN85ljn+I3H/lVFlqzTB89RGhAomMIkoRJLkb63thfOzsgV0ql/izNZoPl3iJb54tsmy8xFOltJYC8V1/GMSW07XaLF17Yx9TUJDOt6XR73Zb+7jum68L9M/ewe/cVqQrHqbr80A/9KO6riviWT6fZ5uhR3Yf+vvu+yV13fY3nnnsKWDFHFwr8w3/0s9y57U6WjHpmSPToGGd8JRVXNq6kYBU4fvwoyixI5+dneU3hlnQ7+55/hImDe9PfnzrwON848zVQED43we0f+g+ceLJfHZxEAshd12VgQCcrLknWv8uijyG3M0CeryGPRczc+BzDw1twl+fpERHLKK0jh0yyPlh0uHHrIG/ZM8rP3bwNYdt0r9WymetOQ6HUD8gDK+Cty9cTfPwFqrce4hdk5hzcaE1ycu4hlCWZrEwy0tCL357ncbxyI1eHx3AMa4mUNHwNPD2nxJXbx/sAuV2oMGKcy2udkIKRzgSlEqGXuKyXKHYzlllEIZGj6Jp6sL8+9Hn2G9lvyYAj3wwwvgjwRYAqTFDA5yb2c5Mf8M/NQxJ2wrRf6hY5zKAsIYViOtLbSSRPLadJS/hIobCVRcf0JFWdiGplgHY5M2mwzGKqVbUZXqoQ9CRChowv6gc6YbAbXsP8LvV+RAnHcnAGK0hhI/1NDA3315dO9LrUi/P0itro5bWv/T56VwfMlmbBEUgVExSTidBidlifR3lcf9e1+iJL44/TM1Kt9ka9cBFKIXJ1OZ6jFyIDZc3ISCvCKZp6ctlChj2UZeNXBaHdYcQbwRIWl41flX23XpHYyUy1lpYW+5IOWHbKmkyXzeBcqqSA3Er8jexcFtuqEhT0AiAaiQk8iQAssxANTO13xpDHdFtZcmVsTLNEpbgELbA9sQZD3qVjellXnSpF024jNBlu3/ZThUiDbqqa6BLw/b96LRvfvBkFLESSKIiZOVbnsJ1NPgXp0iGTGtItgdRmTS8EDRqx4jT6extWZaK6udcaBhy6Fpu2aFp3buk083YNgOs6ZsyonWDgzt9g8Ou/pveXU7+MywHcsVGkirl3+m6KBz5LaUnXndmBnowCr8ji2KvS9khW7LI4dj2d6kau2DhAjOT7Z3+Ea71N6XalGKQejOKKLleNPo8wzHLh//g1Ns1pR+G935zkkc9qiaSFxUBtK5Eq4lo1ntl2FwCneq8h+Nr/BAVla5kbK7cDUF68issXTE368H6ipOxjYIIFazN/WX6SKNdCKXJabLpiMP1uB40xHkKikKhFfW5H3IiGkd4Pmr70QE6y3s+Q59UGib9DEkejwzyw+QEam5pcf/2ruXnL62m4umb06y1d/vSze/4JQgjetlkzmV848bfMlmY5du1x/vGPvg/P8pjvzTHpT3Ip/v5HEPuI0hSWrZ+lbqyf+yFXj1O2GVuGA4gLIwhstkidODpsTKVKssiOzW8FYP/CPWmWsyvhVvUjzA7re/bV5SzhBCCkS6G7kU44yD65ga3xPjxTBz7gB1Q6PSxXMl7dy03L78/qzdGmbgOFarod367jhC6F3mv54Wg3lYfnGTfz6g7PYnD8JI2dj3LZrnEeLimuG/4Iju1SuO4fIhBpvTbA5niIAVXiikH9vEeWRcEtcvnADVSal1Fqb2U0vJJe8Gpu8TYyadRuyVwTuS06Fc2AOTn5+Gkv5GihX3Xy5vBq5jZtZf+GTSnYjc180VWSa+JtbInG/v/svXWYJOd19v17irqau2d6mGdnZ5aZV1rtSitmlmW2DDHGMSQOOXGcxAGHHGPMdswgW5YsZlppmXlnh3mmpxkKvj+qpntGK8mWE9vv+357rkuXdrqrq6sLnue5z7nPfROLt9EWjLLUKyMLwaQyjYrC64cvpXrE+Y57K54k3+ahq3qcSiaJ+49TGy+DZtkqz+Vrli3hlvY7OBw9XHpNUcNUT8ynITGfeKqB9cZ8au0AueRp0nI9AgnZnEkQWDSmW1liNKFZMkw9y4D/DJE3L2Us7GNecCmdUiuX5BeRmGhnNHUKSUhcWmim0dfG1to7uCR3MX/04j/wjhPv511n3sL6scsIZT2sPh5BAEu7WnnfxREahbNWuFrayVbmc9OBDzF/dDWRhErV9BDXJvYhXLqwkopj2yZXrwyh2Hkc5ocMmDh+QU4YA+ME0874qYcDmLKNSAzi6z5CbTTC1o6yMGZIKvKPl7WyPugvac9UdXQi685aQ8/00V6A/N7vYk46QHbZqVouqN3CuD6ObZnoA2fQR/qI9U1Rf6CDG5+q5dJdrdwir8EXd9cbuQzxwmPgUtAly2DdsfIzY5vDFJLfBztLSA2xyapBsoUjXFh4EMPKo7uCdQu2vJtgVS3JQJQlxgluzJ/lzGc+xYP//jc89bXPMHn/MwTTCoueM6n5ZT91SeceMm0TG5PpjDNnKlMG1gxLUVOJB3+CLqUIDxwgn05iGQYnH3bmlmq9pgSoAeKJBMFnB7n5iXrmDQboHj/OL/7lL/jun3+YXT//PgAT3Sd45pufw3QFc/1FH7IJR34FID+967lShfrMgZ1U/OwMl+2s4dKd1bRPuC2cLiA/+sgv6N3jrDH279/NY489yN13/4A9P9vB4snFSJaEWnTWeDOFssNTB0mRYnTFOPc030PDtmYWLFjMVS3XcDrksBsOHzlAsVikt9c55wMDzjOPWeT0C09hu8eX0bJzRPwitsXkLLu89kQ7uqWXAD7A6OgwExPlCr42aTM0S9ytMJ0nmgly6c5qOnYZTPaf5ZlvfpaefS+ec67KgFwrMRHS6eRvbDH3vxHnAflrDDFLZX1OhXw2ZV0ykFSJUKgSYVuYkkTOTM+lrdsWRclLWFfRFIl/u3EJb1zr0Lyq110CwJqnBlFeImDnt3TeP3Fh2Xph1j6/WvVjHl7Vy4nlp9kX20do2lkYpDQvj1e9Ef3Wz9F54gTCspCzKYruolaTPHRWBchSBliaL0DU7Y+fyhbR3QxSSexL9iOQUPPl45OMIoZskZqhdeUVcm6F3OdaWRRn1BXlNAjQvXv5aOAX1OCA1RtdulHZws1AQqLFciqPectZyI+4mfaMnJ5FV/cy4tLNybrU6/L8QWvdQgCSAehudGg3oewAdbEz2JQrsjOLdVsIbFFECIVVtQ1s3baITPf7sYbehD88F5CnkYn7BjC0JCDo6OiiP9DPM7XPEGlyftNMVToiCxblHeVcn9eZWCbyZzGVHIZbZVYibm91rBrJjpSvi+QAzqjrmYyw8MWcgdY2pzDNHIWKajKxDLJvhJaCU83zzhIg0/UAply+bpZlUpzlkZ7xzdAjjXJyQvWUALniCvnNrpBbXi+2ZGBiceOyW8mpzqArud6wecMZ0DWvgkdAyLDIJN1qPhKqrjle08CZAyfweb0l0Ga5bQ2ZTJrsUIbt/dvx5DylCrnpKrHn5ByRuAPYUyLPlHtfWMKGgEDUKzxnpTlbLHLy+Uc4uPMooy5oBpCxS5R12dKcfj3Due+PFQZ4PGkw4W5fa0VIffXrTL3tDeS+f7fzGWWCeae/CsD40ScZE05yptp0Fij53sOMHQ6hDTyHNHGilHACiEnVBBXnWu0cfQ7fc//IIB/Bu/VjeKLOvZrzuGI+buW4cvIsAEeVaxnIvY6L4tu4VGmgQy0n8SRXvW9hq0Fq27cxIu3kWy7G2vh62qJThBJn8aWHkWf1oa7vc8SWPL5DDIXOYMhpcnaInXGHmhlTzhCt91PVGkBYMgvHHLcHfWkL/RGHhVPQ3WSf5aE/Uu4rM9QUK69uxHKr/MsubC+9V5zln9yv5bAsE0mSCLq+6cmcgWG4DItXUFkHyHvmvnc4fpC4J47a6cHj8eBX/agrPDxV/xRKQGVL7Va21l4MOO0dEpJzzwlY3bIOn+pjZaWTlX9h9OWz7efj/6349qlvcEb7OxTvZhR9Eyl3uq3SnTlaUl3V67xNLtCCjUyz6YzDU64Gix8dSUj0po7Slz6KcJPhwm0rmqpyFulL7Ln9lJKloJg+ns0vZNgKsXzf4yhuH6vHZaSpXSEkBdrU51BmqTsL0yDidZ4XxfBRpzxHzYQDIlQhmO/6XY+71pvbjA7ePnYjG4eCfErz0s4VqMveg9D8pM1hPNlyMrg45dgX1XrbiGjVmLKEpvnIpm3mFVq42l7AUm8tE8WPoz5WJFgdwLYFctGPFXYBiQAsidD0ghLQTisFDNkqLYQ7zFpq7SiaEiSjKuRdnYoZm2PLzFBpB7mksBy/oVObvwZFCKYMi8/Yn6Vg51C9MYTiIWXEiRckar11BIvjvJ9vsk7aQXWyvDCY6T2vNLO8+O04q+zNpH2jDPgGmFanKSoRWtIOi2HYsHnQt4cv136D8Bf/nWzUuR8UVzW9anSUqaKHDUYnN2dWQu4seWuCe3s7sCSVxZHNAByPH8AEDk4+hWVbxIL1bAi3UONtoSu2me1hPxsCKiv8Xj46dRPfPftpLq9/K+uqrmFV7wTj3z3Eidy7EAgKOYsLjjdRnW7h8lMd/POJ9/L5kU9yY81bqTV8ThV6epSQMo5+zy9ZVbEdTfajKeX7TtEaUXUfVqHAqtPOPGTnnXvF9AvkXIb08DgX+HQUSeA1M7xx8Ic88U9/THJ0Gtt01l0XX7CEbW+/AYCCeYRR88tYRi+WcNo/7NQEtUMxToQOETyxDzU5hZDCCGxq+w4RSls0jNqceWgHgaEzqJMj6MM9+NIOKOu66PLSMXurYnRuduYq7Ayy6qMhsAoJqbTWtlSJjd69SHbeKax461l6+Q2YXue3VxqHsY1BhHBE1Hp2PMPNTzbQOuzHwma82k36A8daUqSKTuJBV0JlxwFNx9s/yjpPgQ2xbVR5nfnadyZJ7YSHm04voTZePteW7iOpm/jyChfuq+TGJxuQRl13G1cINxuf4MzOZzjzxIMAtI1GuP7pevqGT1K0iqSnJujZ+wIHH/o5/Yf3Eh/q54kv/yvPfPfLpe8xkfBNGFjug7Z00ElC5YozD5JFIVVmggjbRtc0bMtmwfQCrj57OTVZ5zMzFXLFEDw1/DgFUaAoF5EL0H94L549g/jcvvPBgX5Onjx2jlXb6PHDPPvtL7DjB1/DtiwGM04LbcrVbKorBigU8khCIEkmiq1Q+ZSgz3VDAUdtvvdsWbTW0nQst7BhucWGrd1LaBj3YkgW1R0LsG2bp7/+GSb7y/uZfa41TaOyshJN08jlciVXld9HKL96k/MxJ6SXVMjVsjPpTBjCQBYKyoyHuAQZMzkn2ywsk4VWkCHP3L5cAO81N5C/9x6sgX4KX/gsLF5Yeu/C5EoUJKyWAHsWBvn0g8/wV+xEweSpeTBWzDOZcBbFgekMoJP1h7l0YR1SuJIlBw/RcvIEz3Q1UXTxgIxCc8RLYVavdCAQIpx1AXmmiKcuPOcYVfe3+BvLdDthFDBkm4QrbKIbfryuWIdPjTs2ca691Uzvt8+yMRo3ohxzsmhqqAZ1XC0pShddENtqVpWEXgB6Aj3UZmvJihxx4TzQYdvHoDbKqlwntgvIw/VRzN48kiLR2NTOmYHDIFs8uKaaFUMQ27IaX+Mq7HtGS5ewSJYc4AXSgV4CyXYCSpKaQgYrv4Zw0IM3NLfHKYXKeOQ4wXQT9XVNZOVMKZvZ3NBKvLvc59IhBzCyHexQTyJPp/FJMIlLDXd/b4AABgWqm9tITa0izdMAKOgUAZ/XhyUsJFuie+kR1HwnanwckU5iqa6AmLBoO9NKT0/3nB5y2RsoCcxVVlYxMTFGblZSMOtX8OCoUafcak/etks95EqxQEHxYM2iuRd9rhAXsMm/gYAyxKTnMAOGDQhyuQSFXS9inD7L5sB8grLg2DHnekoojJ3eS692mIZ0A309PXT4mksJkqSaJFwMkxmbpG6kgqxloOyJ03HkcT6ebSZbF+BMCPzpHBU9E1DjMDDis65P8pFvI757HxMLrmSypshEygCcDK7P9pAReQpyDsul0M/QGJVCiKI2Tdo7TTBVxFSce63WijClBjEPn6Bo1CMvX4gxcJrGqt2grmCEGIq7MK+WqsGG3PhJdtV/CGsoRPbzI0z7iiX7vBo7xDH5jVQnP8tJ+xSDmTcgjItQIuCPxoCdGKqHGgUmXIrCYkmwW8Cx3DaW9hm0u4J5MhJeWyMrCuiaipa3aU1WkrsnzcRtv0RqcJ7JyC3Xs+7vP42tWeTf/sfc7VqkVrn2cUZwD7awSMUeIjJyIwOFpQCkIyE+3/RvXFxVx9jZE6Vz/FwiRXfDI4TzNRhSFtmWmJ/1kvP1Qd7tMRc2Z6cnKUgFFECPhbGFjLBNpiugagIKHkHBzcgHAqGS9ctf/PIoewpp1NCrU9bjSnzOez2pswDE9KrSax9Z92fYtl1SfZ2JCk8FyypWsG/S8RjeXOP0w2+o3sQLY8+zY+w57ux4E+fj/+2o9taAAEmqQvLWMlnsoQ2o1pvxFA2EX4ck+DM2qcrl5Oij3fIRX+xh72Hn3jELcGzyGIeT9wOgTQ5jRRbQNtiLCNWRrffCsSS1UzLodkm8RVgqcjFD2kjTnElR0X0UpcOZZz31TVijCbwf/S8mqn1I6RHqv/sAJ1ynEmGZLK1fyN5ne5HNOjYpd5PpSxCaOkth8w1oxSAJ0+a5lElHxTjNBDju72F1YDXKUAbsxVABtm0h6/fjn46CP4hs6BipPEPyaep9HVxa/yZSnqOo4Qq04SKy3xl7DNtGESbmgMIHp27jh+k+KlKtNGySOL3jFEm5Hz1XjWx60fIVFPRJp0JmgzdTi0dPs67oJKt1xZmzphM6SrQAlivAlopDNIgiwCuZaGMLADiWs2gfUzih72JJ1HluB4p+rjvyPgLNNvbEIAB3JpLI01Gm3Udfz9YiqeN4EovIFeDoIyNcF1H5dvUOEDA//QeszpXXP8n8CO+JGmjNq8lKe8B29mEoKXwNJhMDAsMy8Cthor5VTGX2MHr6GE3+hXiVIBkjyYnEHtTIShLGJGeS++kIrcS2bU4n9xPWYlTpjSQK46SNBFGtFl3xEfXUEPXUYLIYzQ+XAIlQDc+P/IJcZj0+e4ILgqvxzgLaV0oXcXZoD97Q1dT5GiHfyPwwFMUpLGMHxxJNtAeXMi+0kaySYGf/z4m5ziCW3Qb2XsyQDhNptCenUTjDvZFd7Dn5BENZH0VAlfvBjgMQqq4nUBHDH42RnhonnMxhKYL5b7qNobt3k5k6wf6fPcZNS7aCfRQh16EGOyjEnwa7zNozMvcjbAMrnqRg23iRCcRqWHvTG/H4/Bx9/H62vP69VDS1cmbX0xj5aTa87g/wPu7MFRHbzxRpLI+XeK9THZXUNlLjBZZsX4u1y/GytlUNWVuK6t/I8stVXvzRF7Esm6GKNC8ummJ7bhOMQ0FYrLjpTlqOLoBjUFTLyXRbUbFlhZDmJORqg9s5rYwRSia44oVacpyCjmXlwUUIntmQY+Gon9ajAm9BJqeZ3PEX/8CLz+3leM9Z6uYvItrUzPHD+0vfEcqoXPJ8BQ+M/ANTh84VTwOwfeVrb+oeDi/OcahmnLed2Ygv4TxPmXwKVKhu7WBkBqTaFnrvSTxmkezqRsSoD0310ZBxEhuxCXjD7mYUUzD19M/xrfOzeCTE2EM/5jGXvbMUSLemsbx+nnri4XMPzu2PP/nsowhJ0L/WBzbUus9WwFXaJ5vCM9JPtmk+lptclHIZsN3201lz9uxWVO/YCPnaZhQpSlHu5/4Nw/zndf/KyR/8hLN7nic5PkpFY2tp+9lic5qmsWzZKnbt2sGuXc/T3t5xztrgdxHnK+SvMc4RdXuFCrkiKWgeB7TZksXZ1KE5SuvCsqi3/DRPnUuPEOEwwT/+F+T6DqTxuT6jXfkWCpiMr4kxYdk0ijEu5Rm28TztzO2v9LmiUx2drWxorUB4fQgo0d+Ks2ifStHCP+thDofDZcp6toiu62UPVkAuOlU4f3PZ+1kYRQzZZsqlUnmLAXT3IZO9TsXMcL/DcKvoPtui0LCptA8zUD9HgXGGKlNlh9FdSr2NzZBviKJw1Dq7XT/QsO2jT3cpyBkHcDZVNfNU3VOMLpxgmKGS4mzYcAYBf8yLr6GuBP4k0yacsTiGs13OP0i84gCWnKEie5Y3rW3igxe1owfmAvKk5SWWd/bZtXABP+r+HkWryNLocla0r5qzrcdW6dGdxUGBIov8ZonWZ0tFLq+9ijbdqRzqupfKWckQxVV51XQF0+3j6evrwfSHKESrsc3pUiXbFQRl375daLNaH4Qr0CdZKnUuxXo2IFddC7KcnEOYTmbUsKySDZuUcq6vrZVpWEWvW9EvBBHfOcsqsZILam5EcSsx0/f/nMQfvQ/rocME3WcoPOxSrwsGnrvez398IU7NsIMKR4sn0FyngoTq3DuZYoGCWymami4QPjTBhoYPEfI5ivPLuov43X5MW9jkpPL9PfrdezkRDjFal8WWDGSjXFVaW3S9NV0wrloKkkurUl2Lmbw3S1GbBgFBS8eHB988P5V3bkKtd87hmBHl+MSl6LoHC5mC20ZRYfshoNKz9ibGY8uZFG1kCzqyO2Z4LQ0/OorQuPHYO/mDodchjMuwbJuebArJbVkQkkWbr+BU/IGq+nVsDwhW+mTaPTIWNp+v/zHPhyVqrDDChsTiBjZcUo8wbLBg/EenePrQCLZpI2Vb8V/yCXK3/xPv6C/nZvc2PAiLn2c8sg8APfJLUnK5Uva1VDP/9twovxyLY1c5921ByvHY8DQjwbPc3fmV0m8TZpg1L1GG/fpju1Hcc/3d/WPkLGfbznUVRGp9GJ0Bgu61C4ede9+2bfb2T5cW5a8m6jaIk3nvDC2Ys82MddtMvNKEO0Nbbwg0lFR311c7Y9SRqUNMF6Zf9nPn4/+duLzhSjwignBFGafMIpZt41NDbDubQlQ496U3Y1Ed9OB0lAs2rtpMk3cpsqETPXUSdd8XMWxnbtGmxghNdFI/2c264aN4ojp4DFShEsiVGSqyJWgYeoY/2fXf/OnObwOgu2Ny1Y23UPXAYyjtHViBeoyalTSsK9uiCdOg5940nnwHWl6hTjmKbqdpHHyaBa6at9EZQWrI85dt/8qtXR/hvxc/iufOTtQ3zSes/BfG0E7yB75PjX6GZTI09/QQS7chbINd4w8xlOlGEjKh2BK8aj0ykLFsTuUMHk4YFPkyQssTyuisxaH8z4/Np0ZeTHhyKYHkPCQtRXC6i+aMRL+/H38hTDDZxTX5dfhwkmte2ZmDp+I+UmYVivu82hmXRSYELVoALIlx02CkaOEpSJxM7MEy89i2TX/RMag787MEPd/pw7ZBBQKu0nyASbRClNDkcooFZy0z2p3k0oEIiltZzIxEiVjl9V2Veg2hy79CaiIPNmhemfq6Jiom1lLX8D4sYNh1aFgR3YLqcxIM7UEHlHUnD2Da09hWFlvA/skn6J+f4+mUyZ74AR4b+g6/HL2b+we+ylMjP+Lnff/JPb2f59mRn3EimyeRniJn5DFsm5BWydb621npFVwYCuNVAiSzoySf+Fset36GhKC9YjV1vjZs2yLviq81V26n2iNzZeNdrKq8lLAaoFbUc2XD27mg5mZWVd3G/MBFvGvPWgLeRcT0ViK47YeTm5jOlUU8dd8ZbNtEVlX8FTGEJNG2xhkvZVXl8nf/GZtXXcfSy7cDkJrYj2eXs4aQtXksuujC0r6WXncnqteHbTtz/4F5cR5bPUpl+3w23P42JFlmxdW38rpPf42ajgWoHp0bPv4prvvzf6J9yXIq5BlA7lbIZ61RZLWdRE+SkcdPlpJftqcO4d8OBMgmm1HWXUxqwUpyTQvwS9U02Q5bQNE83Nb+OhpXdJK37BIrcyZWjxWR3ba/Ci1ETL6u9F5FU1vJtlh3hc1uGb+S4IUruez9f8FgG9y/foSz3nG8UeccVza2csGb3sPJi9x1ZsBP0S8TzqglMF7R2ErLyvVYrlBsoTFI5VVl8UhhS+xrmiSrm2x88x/gcW1cZ2zPVl1zK0tWrEWZnkAf6sbUspj5HKHnTuHvPoKULydIasYEitt+qGVt2p5MsfZYFAyTYKyG1tUbCdfWoiRm3FNczY1Ztnwej4d1t74FhODE04+Q+PnTXLQ3RufuufOpnMti1SmcUl8EV0S5obqakK+MnzQhSuMBgKkYaMUk2BaW7uXYao3JcJGJwiQXvvX93PK3n6NlxVz7yBnK+ky72/Llq1BVlfHxMc7OqsL/LuM8IH+tMQuQC0kglzJls24OYaIIBdWt2kjCojt5ELUwi8JhWSD5qO7OMTU09+G29k9gPT6Nb90fE1r73jnveWyFH3t6GZVhOlekUZQXux2pybnbZp1jmvEzlnwOKFNcsGFhYbgUcPImleFyX1csGi1T1jNFRxW6WH647Iwz0Pn8OljOomWGsj4DyMO5qpLY07MLMzy/tZJjHc53+G3nIZxfKFKcBcitQN0cQJ6ZEUJD0GI6k2hSTWJKZolOPSo5+4pYfgY8DpV9pkLe4GtkQp+gx+7mdP4E6kzF3qWpeQISw1NnS325smHTlYiSyFdwtfkswlIwtAQPebo4NV7gfZsb2d5VhaxIGLPIDbKwCBohZMNAObGLX7hq2XfOexPBYAhdKw8mBSx+VunYSqWkPIVZAjIAb2m4C9ml6em6TmVDqKRRIFvOeVc0Gcm996pdWtEMhXzm/yn3luzv76Uw6/7E6/rIGzqVFc45zRoGtd42dNlPJOyo6xZFjo//wMA7W+jCMrFmBmrFOYeW6sHUimDDjYWlzOjF6bKfOtf2JIuNVNmEtsChmCXMWX6Yps2Me9zKva5NTnGMoObsaFV85njMknpxTjZJXfxuhKKRc90Cgk0XoQXrS4mb2ZHzeck2OQs8TzqMNXSKisEO1qRWMd+qQ5vVh+mxPQTSTmJHK0SRDQsh2aRdJeEa996xLYlq68f4a5y/hwOtTB23OZma1RtuB5GQsJsD9JjO5D7/5I/YeOpTtKcd1kOVK2y2xJfmGn+E66bdvtOcxb68h9MFNxElLLSZe84WKMYIHlmh2VWP/0LND/ilvpf/HOtmVXERr8tfQO+RDEf2lIWUwia0PzjA2BcPYh1wFrZTQxlG08VSAkdT04xNFJnOO4mkvBXkfr08do26k+QXnuvhydBpLEzORk7QVhkBoGJW60sBnVHhAOGiq1bbnHfOY9zS+fGhMfK28/rCZj+XvXcx4QVhgm7rwEz/+EgyT7ZoYbvX9ujIXMtEVS1XyLstZzJdV71hzjYxvZpfJ65suoZb2m7nrzb+VQm013rraAu0Y2Gx16UAn4//d0OTPSzyX+v4fANFoRB3W3W0yvkkK515TMtZLKzwoc4sDiVBkGYqxtdRMzZVEmObCa8iiMaPc/OpJwjpKkbcoUdGRKS0zVUiyXw7zorxU7QmR7CjFay48+0svfwGWldvQqhzx7doeznxJEyTxJiTxI+NH0RWQbiMGk/BGSdql2ksvuAIGXf+nFGWlgN5gso9FPZ+iWL3E8jFBLqQ2fj8DiKWBnaRrJnkqZEf8sjgtxk4+0uK1j4eTxV5OGHQP3KKgg1xU8Hf5DwjS+UgAcnRDvF4FbRCFGHLzK9PIdkqvkwjG3JZwrlqBOCZNU3pLiCPJ70kzGqUmfey5XVOzNVUOVnMAEUky6ZgZTl94AucGTpAys275vUKDlXeTCHhzI1JyQGUVX6HAWZZ5ZYugJ741XxqdAJj5FJ8o84clnCniPhIjkIBkuMOAyoY02lY5CRMhk+lsG2b/ZOPYVhFqjUPbeFLCWgV1HpbsW2bM2nnmlum01dt2AVaLl9JQoCkuBThQu/swyFrJhks5Dmal3gmCw9nbB6eLjCRG8Qje2n2KOiyl3hhnL5jj0G8l2W//CX5o/eQL+Y5Nr2b+/r/i4cGf4xt2wSKlcT8b8OvhMkYSQ6mU4wViyiSSoOvg/mBNpb7VG7wvZULpzazsMKZty3bQpY8rItdhVCc5Hd60gGIwaraEptp0SXX0Ln5Era/98+o63L84+etW4es6sjkaJVaWVV5BXWhZSzcspD1t7+Ndbe+lZWXXcuKbTeyuvIy6sNdnGpMkalSueqPPkH9wmW8XPjCUSJ1TVhnkgghSJl2SYUftyVQSDLB6k7mq4KJ4+VzaykWj3jdyu3OXibjU4CgKl/FutF1RAoRAAqSYCyVJ1LrJWNzDiDPVpQTFJWqhCSq6bro7Vzw5vey7X1/XhJ9DrqJgop8BSsDK6jtXIR++Sqmg0V2Du+cY3uWN/McsJxzaxk2ys0bGIvkybT4ueZj/8A1H/sUF931QfZc7+U7l/YyvqCCilNzXZk8hgeBoCLSSHvnWudYVWdM0jQN+YL5PFf3InevPshP13RTdPVX4r4MxfSJkgaBlMvQftWV3HtFnAPzprGETUGxqLn2Ym74q3/jord9gLf/x1dYcuvVpe+2bRM7Udbomb/+QhZcdDnB7W8ASUY7PUXbsN/R1jDKjLeuNRvZu77I/uYB2rcvYOvWS7nqdXexatsVpW0aWudRVVu2g2ytm8dNf/xPBN21b8DvJNfGc2OOn3tkrk4HzK6QOw+21+tl6VJHH2P37nMt7H4XcR6Qv8YQL6Wsv1yFXBjIkjxHaMjG4vTUzpK3tmJJnFU1hA0v/vhMqX/Ktm3M3aOlzynS3F7lJ/0H+IY6yES6wHTOmAPI5yfmKpgqWcFA3Saez63nyJODmK4gkmyVy6Fp0wGz5vE4DbFI6fXayigRb7mHHMqVdQCp4IAIT0B3vDBt28moqTI51aH1+orOJJeXM3ynysf3LpJpEk4/bVdxlId7B9hmqFjBeizd7bMO1OHzlenVWSVLzl2cLzAbkIRU6jWeUWdUbZnVxXbmWTUMau75yJnYlk1LoBWBoDfdw+MTD6K4Fc+Z7Hfv+FEee/x+0hFHkEI1LFZMVlIrJlkijhKdWIlSDJAXKj80L+UH3/kyPa6IhOEpX/Nqt3e9fmCQkXu+Sc7MMT/UxbqqDQghqIqUhbZ2Kr2M6mNIksS0SM+h4gNMTIyTSDjXRde9ROt9KEU/wpYcRVdA9UioLgNjxhJjRmRtpkJendeocSeA/vGyWMZMtlY2dfQ9+wAomAYX1d7GJXVvIJp2rvOCniyNw0UCqbJwn8dW2Fjh2E6obj97MeIArnqrgqDiJ1tIsy/pDMRtmjMw5iNBgpd9BCGpTBhZnkoaJNy+8oDi54N/4OEXW3xEigoew8TCYliKA7CUlSW19tnRq8SZlKc56HX6y2Z8wf2cC8jlD/0JqaUO0JWSfQRSCazcUbyGc48GZgFyzVLwus+FsGXktHMcpttHVWM5AHKSOsbtEDnLAdppy8b01hGbJUxS7dohnZ4uYBk246JAzdgzePv7CZ6Mg21TU4w450+NoAjBtGmzK21wJm8hCiMM5meslyyy7rMwbXu4ORhhV3WGrChwd9UR7q14igv3+7l+4PskQmP48HAnGivdbEf36gpyMsSQCOesUm9Z0BY0hHUsd9EQcj1alewiPIaHBW0bWL+ulke8BZ7Ui/zh1fO5c1U9kWKcY9lRvrPqb3ih4yH+7ipHvdxbnHv+B3AWKz2m8zs1V9U+odUCogTUZ3q6KnzaLEAe5sADP+WJf/kz6nOD4I6fL/aNcWTqKEfjzoJlxlJFVmRSUgq/4mdRZMmc46iaRVl/tfDIHt676A/ZWL9xzus3tN6CV/b92sD+fPzfHSvDV5SE2CxJZcJwnpeBzkuYrowx7SYH40OZktAlAopuMlgpZlBnMXFsTebq9y9FMnMsneimZteTFA46idmo4oyhqi2jhRoILH89Ssfl7KxdSPQb3yG2bhMrr70dzTvXygooz5e2DZbJ0s5+Vu/5NIuOfQuzaoGr36WBO44ED/09i5/+29LnWzzO/Sy57g8zRDipEAcXYHmlDLab+BRCMJEfZHDkKbIiQcIAYRlE4k77StyoR1f3ILWHkIRgkVdG1eWSoCdAQ7tS2vafRyb5yEQ1HgGSEKW1kCp5UITKRCHGUGFhCZCLXBzD3UZ37bGSVhrbKoOkeH6YofouAOpHHOGqseqVTA2FsG1ByhXrq6uf2zK4OPoskmQzUFhK9eRKspOX0Fp0thmtkAlUeMCGid4UiXFn/gpWeamod65BajIPdo6UMcXh+DMALPEE2b7EsZUcM2zykgOabBeQCyGQNY1AhY6QnWthun2+4brNzHSWyprTulhUAliWjBLReS51P3smHuHg5DM8N/pznpg8QabaYfVoBhSO38sz+RyHswbZwjQZI8FwdkbAcxG2bXNg4nsE1ft5Lg1PJw32ZUxOZsYYKlpYtk2D7qHe4/y+vVOPYlgmNd4WVlc5uhuWm3TyTyfJPXCfc10CQTa87u3UdDjJIuO5hyn89ftZv+RSrmp8JysqL2Z+aDmbfEGs7x5m/tptdDTNI/Gnf0nb0Uo6QivZWHkNQYLUo0KyiG3kkcePzGFqzg7rZByA/qJF2O8UoYoeGUvz0FJRpNJ/mKhiMy7K6xlbLnJMzaNlx8m7INvj18nIGSQkjIKz9j2byfO6b7zAsd0PUtQlpiVnW8k9lkQ4UtqnCvgkSE7U0bZmM+nnnPVdXsrTkHHWA/3SBLHuBvKPPMSdf/gzlnZb/OL4U2RcPRxFUTgxfYycyGG68+X86uXct2mYR1eNQkUthmlh2Rank6cxFcGdvZeipebez17TC5aPi/5zB2cG5rJtU1aKZ0afoq8mQyhaTV6zeHzjJAcXZXho3Qj2lna83YfwnTnEUHiKtZfdTkWkjj1dcX6ydYAfbx0gtnrFHKbZpmXbUcPOWioeSvDw0rIgW7Smnq/t6OXvTwW4u/oqCqpNWjc4e8GljM3SfWldtIST087nltevZPHiZSiKQnV1OekRi1WVnHnAcekJVMRY74L2aDwCNgyPDZa8xl8a5Qp5GWMtX74aSZIZGRliZGToZT/324zzgPy1xjmibu7NOAswmJJTIZ8NyAF6U4dZmquj06hjgb6IX/olEDA9miUz7dwc1tkkdrwAHgntHYvILZnbu91jBsikO5nIFJnOFktKmwDzZlWwg7kKXqj8BMe7Xs9UPsjBhwf4zt/uZbRqlTM/u1XyfvUsAPlnhgjsd2nTNtRURkuU9QFXTVqbdWPLhg89oKBoKt7+U/jOHkOxbSKeCnJKWdkcIK+kGFAV5o+3sim1HFvY+OVHqTVNrIBT/cx13YIZaqFYu3pOhTwn50jKzuBXZYe46/I3cTzqPKwnoifY0LyM2/ObWWm2IWzBqDpRUmAnZ1Cpx3hjx1sBiIsJ5Jk+FbeHOpmNA1B0e9qVokXrmEUtk+gihcdWiEysYHUkh0aBiWSWxx9/yDlPLiA3sGhz7cFaenqYdyaDpwhvmf/20mBVnZ+lspnppyAXaOlwJs6ZCr/s9i2PjAwxPOxQ2uvrG4nU+QlPLSU6toZCwtmfKsCvz/UWR5YdIToXkF9tLmOp2493dqDPEWYDCuMT7vd5kb//I/f32+QoElAjMO0MUO0DWUTNYipry8DGL/xUyM5kZ8o2K/0XY0adTOUis4FJu8jJapnuXI6xXD9Bl/qf8cewsl4E4zws92ECJ1yBIl3S+dTYX3FHxb8SvOxTVAtnUVJ0JyIdDX0WZRDLuefPSqN8V/0+CZfd8IPah5DfuQBTn1vtAEcQLpFxac4559yahdPkbOc7wrMsCVXJR7/HtZWzi6RyfXOUN31uAmTCinGb/BkMnHs4Y9kkgs1cdejp0rYVZpCUbbP/iHN/7AoKRt/8Nl7saGAIwdhxA1OUbbR2WUUen07SlxvjlG+QvqnniHV/tvT+zL3iMzNERg/xH9nP85bGj/Klys9RNaXROu5MyHsO/gBTWLQj40Ng+RW6LmpCe30Tz68I8NdkeD+uUjSCz968BE117puZBI9kKyyML6S28xY+cnEH733zEt78+k66po/SseMrvLH/e7z58D4279NYmQkQ1nTaB/ws7S0nn5xw7tllszxUpWyaJcfvxWtmSlY0J3Y8xcM/+CZaeqQEyKdOHWHfvT+iMDHENcP3s9BdnKaKaT684/18cMe7GU4OEI1EURQFUeGMqS2BNmq8tXOOYnYPeXyoj/6De3gtcW3zDdx72cMsiS59TZ87H/93RtQbRMxQLyVfCZD7w83kDavEFInPZrhJgoIrhKYa6TkVcuHzoFRVo9xwCwD2p/8Oc/gwxvRJQmln/lFzGewKp3LtXXIzqzd+EOtnoxTvPoPx7BB28dzFZSgUZlkFRAZ6EEBlYRfhRDeSbWJF2xGSjeR3k0i6jDa5g6hlEXTbfzriToJb5Jyksu2uZSbSRQxXUNMn0uBS7z3u/FyUFbLCGSf1/BT+tFMQiBsNyJlhpA3OOBBTBKoZR9Pcc4lJ1fxqZPKYaFCsQpWX4HWLHTnLxHTb7jyynwm5ht78ihIgl/LTFGdjsqCKbuXn9CDnVYWplDNWLPmL1+NVC9hCpj+xhKQVw5JUhGUQmxejOMuzbLHyMzpanPNwIHM9fguaDWfuKcQ0qlqdhP7Y2STJMbdCXqkTqfPNsKCx3ePoLhxm0rBQhcA1IGEyqFLV2upu6AByWfMghCBQ6UGS5yb7GhavQ/VfiexZg+Lror6zzGKs74rgq4xyMrGbI9PP0pc+hq20kmxZj1nlgBVjQSdpy4+ir+LCk1NssnU8qTJIyvY8RceuPdQeP0TUn2fStBkoJAk8+zleTJvsy8wS9RvaR+cLP+HsqAPo2z21NOgN7jUVtGYbMR+bovAP9yN/6T9RT92PPHEc+7ufYPpP/4ziYCWNUwvwKUHyEgwYFrZZhEmb7Cd/SPIjH0cNbEfgMjlthdvHL+edZy6k8OUjyF/9KhU/uAzv/q9gGxbGCyMU7+vBThQw+6cxTznXbbhoIXxhms0YCKhqiuKtVNhjTrBfP1GyQ52Ju8IT1I3uwpjRh6muY8oXByBecP6fweCz1t9z0c53YOVPlyrk1aNO8Ww6NFeYsUIRTA6kSR2cIH3IBXbCZpnRgtfWSEhZTh06Se8vX+DAgney8WQTCauPg4Pj7m+X6ek7AQJsF0bEdmq8b/R1dI01cM03fsiVX3qBv3n0ebJmhgsTK6kpVpITc7VVvIYXoxDAssFn2NjYpeLh8eQx9k44lmRX1rwNj+RhUJ1id+sYOY/Fgnnr2N01wWBokpPrFDyKh0qPM6emvY6TjjJLDHgmLtl4OR6Pzhu2vg1/OEDOtSre0ZfiC8+eBaDfW8cPtw7yk60DTPvmMzWLeTylxTFsg6gWpc5broJHoxUlO7hYrIpIZDYgd9akbW3zkGUFOSezbXAbE0+P8N///RX27dt1jsjcSynrAD6fj/nznUTegQN7z/ltv+04D8hfY7y0h7zsQ/6SCrmQ5/Q1gtP7nBs7xhZjEQvC61FkgVLhbDPR61Kn9joVXnlJJSKkkZqVeQNYfVKnMpNjPFUgMatCXqxeTrsLyCVLYvvJN5PyNKEU03Q0ZQlUesimihxafBdHFrwR4Vbe05UWaVmgAF1pD95imArRgPXTs2x/YZKfEeDN/UWO3HcGj7s40QyLoFDwhjRkVaNJa+XiyDVc3/Be/uvAn/EXw29mqVeiyyPR6ZFYr3p41/CtvHf4DgCU5Rqa5AzqVsB9wC/4Kybf+Cy2Hi1ZEIBjZTUjAAegej1U6M7D5/cG6JrX4Xg841DBTWGRdy1UrFPOwubN8+/iqqZrKUp5hFsJteQ8tmySzsyli3vyObyJaWrFFELYbF34LOtubOfCi7byAb7mHGs6RS6XRXgdIJHyxNHsAlo+T+3wMJoBnytcw8aazaX9Vo+WJ7aVJxPIps2KxYvnfPcS1zrlxImjWJZFOBQhEoni8SlEoh5kSyc/XaBLlwjfd4aG3nOrJRc3vQ1cQO6VfDRblfhsjXyhQLpjKbmOlaRcAFSNF+/CbfgyzgIi7maOi34nCaT7a/FveC/BfBngemyVkxNOxaEoTKhtxJRtfLaGr1DBi+lxcgMnQNLZOf4AxbxzfrMUkPRT/FfXD9hZ4Qiq5NxeaM2WqSvGHGo3JlXG3ETDKWUHaat8nRTlKFgWSSmHOTFd8vj0VPuRAx6MWVniggvshrpPkUw6+5CKMw4FKTCdRWRYLQ/uRjbOuH0U2y5iZJ+icrKIOu0kMYRpcVJ2es5jmsKXbt2EglPVyVgw7fdRNPLIpokAkjVZ7skcR7EFo2S40n6YAy8+yrhfZzKiUC2dRHEnd8NjkZEeID/9BQqJb1E3+CMqtW76Z+Xk9hQcgRpvNs6l449xzQGTm56qZ9veSlYfd1gm4doGilaeU/Hd5XM2P0J8qJ+7P/Vhhh/8R950WZArO50JSUdw8EufQmZmwVwe49qT7Tz2sx18+Uv/wbFHvs+Ru7/Io08/znAujyTLCKB+wkvH0xke+cSfs2V/DH1GNfolmWlpfAjcSdE7MYxdyPGR2iE6fM52Zwb6OTE+zoFHf0rIFcTrfeFJAIq+KB67wMbdfaw7HKVzaozGPsH6vSEe+ss/5rH/+Bs2X7KBnwZ+CsB1LTdSpVXR0edn+85qNh6JkTzRjW3bGIU8D33mb3nsS/9MemqC1xK/D6GX8/H7iYCuggtIbUllwrSxbJs6SyDnTUaVcoW8FJKgMFMhN7KoZrlCjt9V837Xu5Fq65znwzIQHSmqa/Jg24Snp9BuXcWztSomNmFbYE/lsc4kMHeMYL5QZs/NhBCCbRtWUzHg6G+Yo2dL75mxDoQEUsABxyKiIKdHEMDNwSUszufZ0OfQy6UZKrgtmPZqfH94OTvHnepevdxTAuSaK7Q0WLuWflesUc9N4Ms6Y+mU2QDJUYyAimXbqEJQ9YNr8I88AUBIHkFEG4koTmJ0ymhkymrCFYAna0vkXO0Sr+wn6Y0yanSU2gKkfILirASpVO9Hli1sqwzIs8EolmGjeWXCjRXUNDhzwjCLmMo5INKbG0epqyPpEnq8UpyYdJqF+t0ITNKF+Vyf1lARDMkWnmgZkA+fTJBwKeuhKh3VIxOscnVaXECuBYLsyNvszRhM2zZjRQt9eYzWVU7V2Lac86W4gCBS50PIlSWtFtXro2lpJ7I2H9W3herWCM0ryknFus4w/llVQo9pI+QYyYkC/re+H3Qd9fb3YFs2siKo/YdPMu8/v0rN33wEW8liFeOI/nvxFC2Kp4dY+MjfEBs/wILD38OfHsRjjdNXtDmYLSBZ/Rin74a8Rc2Ln6bY9zxCSKyLXU9Eq2ZV5XYaW65GqVmCrdaRSV1I4htJJt/6Tia+eC9K++XoS29HCAlP4UGqWr9JePffkHn6n7DNIkrFInybP4qkhzGn+8ju+goA101exJL0lc41TW0iby1A2vkkha8fxXxmCOvYFIX/PoHxk9MISSWVGGbahIliihVGKwA9WiMvSk771UlpiEl3nTPDVt2U6qZ6bF9JsLUgh1jfdhkAaTcBvtmzmwvlQ9g2RA99tQTIG9zWqSmvio2N4SqNzzjCnnnsLFmXWdKYr0JDodN2nsVRqYcj+nomKxah6O9EL/oppN1+6h1jXPrIfK6aupCI6fZ+98e5euJC/nLgXXyxWE911uTBM/vAhpsnL3XOEXMBedAI8MGp63i4pZ5KIWFhl+b3B888wUB6ANsWfOZBhUyyrAely17agu0caUvy6NoxqiubODqS5LmTc3ZPKss5oVY0key4gsNJP5fX38ixyDESviTfPuasv25eXocWewJDNdDVAKmMn2k362B6TE6knbbFRdElc+ZbSZJYs2Y9TU0tNDW1Eo1GS+/NVMs1TaOlpQ2AioLzWiqV4tlnn+See35CbpZr1Usp6zOxbJlDWz916jiZzNzi4m87zqusv9Y4R2XdGTxnpgdbskGAIinnAHKA3vRRFhU2EtaquB3w13qZnigw0Z+msc6H1e1k7+TlDoUtGy86JWshHKVmTx0fOfUsZ8J1JKsUauwEP5v8BJPTi9GLQ1ydGyYjWdSkWlHMHGt3/QM1t/wNK1ct4fTz4+x5oIfh2g2Y00fBmqLviIzlL7IhoNDmUWgz1zjWlH0pVBx6awwJjiXo9jQACSJ2iO0hlSEBakpjU/UNJY9vLFiR6YI5TPsYnVOO2ES/d5T2iy7B6g4j5acx/XXnnKNXqpADoMlU6dWM58aIaFG0ygA2ziBWcG2TejsSdB6uxHi8H1HrRar28UeLP8pSaRGZfV7itgBhUwwkySbmCkoEU2N4U2lqhLM4aWoxqVhVRdGqIOgLEcokSBBiamoSyQXkij5AEWjs7y8xD5rv+w65k8PkhgoEPvAhagqVVKpB0ibUjSe4/WmL+Yevo827nW6lhYCt05QKs78iXuojapwMUPjMLhRtjM3WDvar17LYK+Nz70FdzO0XAsi5E4GwYVwMUTueZnPVAvaK40wqBQxVxnBtsRarfqINl1DpeYEMKXqMXjKZk2QqHYBU0XolwlKIUW6FyOQnODW9B2pXgyR4VnUGz2orzIspGyQfKakaIXSSxQmeHPhvCK4FYfMWv2BQHEWv1NnScyu2WwEfz/SwM9FPxp9h7evfROX3joCnfA8MTaVQPTKmiqMGOuAnG0ph+kNEC5WkXUBeH+ogMZrFyqnMsNZ90xmMkIeRkWFsjw625QqNOB6seWsEaCJsarjMaZT8BBX5ExSMOLbpLICbo1F6MinUyVFaIidAXoxPwOmf/JwlLCRnFzGts+TT97C/uYoWv45hGJx8/ofON/mvpzJ8hmzvMSRFpWP9FgaOHoDJMaaGH2Ky8kJ2DTzIVMGpmiiaB9wJI5jNk7JMkGSkosACKmrqSdsTJOPjBLMKrcNOEsOWBJe852P0HdjF0V/8gvmh1UhCYlqP8+J3f4BRyEMhz+6v/COWaWK0fAhFUskMj2HoPpilWjrkHaIuW1e6H8dTKdC8oHkxA2Fu+8CfIdk2p55/nKNPPEA+lcSQLHJ+1Tn9IgU42QTJNLjkDz5CxgJLkvBbW3n8S58mtf8pcrEGCEWxXCcAw+PF44ppScU8q264k88MVtG57ztUF8ZZ1BNiEUVgRqTNYnpkkBe++B/ULpZp6FxP11glT3//n7mg291mDB774j+x8ro7UDQPuWSCQKwa7yyP8/NxPmZHwCMz7lLWbRRMG5LYhBHoWZMR1883PlBetFmWhVGYEYRMO2l62xHElPzOeC18PgIf+mMSf/xHAHi2X0at18t1f/kxgitXIwUCDHWFuX64m9s6q7lreT1WbwrzhRHMvWPIa6oQ+tylm1m3vGR/VswUZ8wbsCrnI4RdAuSyLw9xsLwx3rzuH/jQN1Yj7BEm42eQShVymPY6Y+p0waVO22PoIk4GSCdmxFUN+rPOfvX8FL7MCBJFslaEH/f9Me1HJwlZEJTBturwJI4DKwkyhqVHiClnmTDaGDKXkEgHqHYX31nLxrLS+NUIuuxnUsmjIaFhACoYWQzTLLEVRb0f+ZDAtsvrhIQeQQEqmwIISVDbanP2LExFOumeyoECvvQwUvU2ctokFCGknkIIm8ap+5ind3EqdyENpiOW+ZC3wFVeldr5YWRVIj5c/q5gzDnbFQ0+EqPZEnVeDwRB1uidyNFbMBCS4OqlFRRz851rVnCtpFzKbOemWrwhjUMPNBIf6qVm3gIqm4LO0s+Gmnkh6uZH8PgVJElQ1Rai/2Bl6ThatmxnrF8jmyiSXbqFqoevpv/IFHCKULUXz2qnACA8Ep4PbHCU7aWtmMND5J94DOknP2TZoS8hx2IE/vUztD9/gKNjMTTpMWqj32b6/e8klW1F27AZc2Qc42f9aJ4Yl9a/CUnIgIUS/xHJMzn0pbchRTtRl/0R5p6v41l8AwD5wz8lP/IL7CcFdlpB8RsEla+Tst8JgBY6hXX078j0qxgtF6BUOckLmX5MGhnL/CXkgyCKEFARuow9ngMkzHgPI8cegQVvYuLMUSxPJ/VKlEHZAc2aYVNQBDY2ii1TY4XpkydInjxCLNWHJTtrwSf3p2gcOIKv00NSSWCZFiFrADPazImeBvL9QxTWOXNis9zKbiAnG/xC7GKKBDcVNlEpy/QgaCkITrtaOAHhrLtWL1tP3+E4kyTRY9O0ZCrpFhFu2/8xrOBZslocxcXV7x9+HY+ph0jJGZ4nzaA9zRrJS5dRyRe9Mm+vmKYh08X8XDN5UaBHcSrsHlslL4pcNrmZyworIe4kIXKUk+Rrn7mBhZ6tPNn0KJMeP+Gsn1G3FpLPBvmzHw/hkv/wS7V84EcH2Z7cjBht4ZH53wJh8+CxKVZWmfxk/yAbu2oIYPHuH+5nNOVWn5UaQl1DnLZOY1c9SrTYTrB+JZ78I7RNLKPRvJSThsGwUckiZYiBwAC9Yw5j56UtZwCrV69n9er1AC9bIQdYsmQ5Z7pPMqVOkZiX4s01b+XZ555icLCfH//4OzQ2NlNRESOff3lAXl1dS01NHSMjQxw5cpA1a+Zq0fw24zwgf43xSirrM2G5zWSyUEp9jS+NQ1PPsLnmRu5EI27Z7BGQ6ElSdHs8pI4wIup8NjWZQ1hgyyAsgaEGmWy6gsioTVcqzynzJgYKjthFlkaaJhtL37Og/2d485NI4QiSLLHxhnnw5b9nSG5lPOQnaSeQ1CZGLZtsewhfqkg+WSCZNRnImCRMGxOokAVdXgmfEgISVApnkV2XLmIf1BFC0Js6Rp98ip5tYcbP9rKt7waClo4NROReeqKH2KtbTLfb/IlyGUb1CrS+J0uU9dnxUkCekssLHqHJVOlVHAXCWhglHGCGqJ8Tzr+GlxRZkAlhdSco/rwb7dYO6E2x9eEWUjr8yPRgKjm8WpLpwtyMoiefR8/kqBWuWqRbwUeSyXdcQ9WBSRKESBx/At2/0qmfqHEAaoZHyIVU9ESR+BkfnN4BQPJPPoFv80e5KreGBxIGWvERbnjepocA7ZUn6b24kQ6rmgpfKzNWXABNVgzbUigW69C4kbXuYJmxbEKFLxJWPaCU6c4Aw5YDIGVb4SMdn+WHvSot1t/RQhVF0+RZ+Qin1FGwQTK9IEHErKBPSdFvDaLmnyQpnOywz9ZINXtprUvAfue+z+TH0QEtYVAIq47Img0P+w+waWwhWD6mbR/CBdsCR/xMyCYZ6Qm8BUjm2xA1XuykM6llinG6UwPYSZv7P/+PXPrBv4S7f1hWQh3tRtQ5v1PKZhBxC0nLYPpD2Lof1RX62v3gEA1t+yE3BC6zT84kMUJRB4wDoujI2yneTRjZp0nleiC0hkBO4LLrMUxA1kpgvPPCS9lw+9sYPnmEx7/0aRJTAxBbjJE+Q3LkJFQvJJkfJj99d+k6jO19fs51KWbug7SBkGQu/+DHqWrtwCwWOfzovRx44Kc83P9NVN3Lymtvp33dhfjCUZLjIwTjo4zf9Q5+ssTE1Jx+OICWpSv555YHmciPc5G9mprHhvHlFXKLKglUxFi49QrqFizl2H8/APECB7//tONHq3upau9k8Ihjp2KpNpiwcOPljAweLR2vMIqEuk+jiCk8SRN/TT1SNIbi9TNVtEnn0kwlE7S2zmPltbczf8ul7HjsQb5R+A4bxl1Ng2UhhMtSr6prpHHxytL+bdsm2tDC1EAPdnHuMzj7GDquvIPFl1zDic8+x4G6G7mrazdnjz5JMKM5lmxeg97aDJtO1uKP21yyuxp2D/Mk/+ZcSwUOtcZptKqInSmy/5c/LlkALr7kGiT5XNvJ83E+AAIetSTqZgsZbJi2bcIC9ILFoGxhC0iN5yHijEHFfLkirsy4mAgTkJD8ZacLbeNmAn/y5yDLyPVOxbbps/9Vev/6pbXIkuDi+TGkoAfRFMA6PY09nsN4agh7MgdZA/X2+Qifgq1HUdy1R76oEHQrZWasE2ZVyGXNAd1GZB62L0ahaQvZJ3aQ+/zfoV2x2vlyGzKuPknWdEnrRgGPmCSDjj3jFjPLpUUvTKGYeS7SP89z+buYNFqZ/EUfa30yQVlQtJsJDp5EDSaITe8H+e00+Q5xPLcNy7yMDbJKSnWOP2uZWKYDIHQ5gG05LBbVZSsIr4plFZnJukr1fnwRD0yUQbJh5R1A3uysJeoqJoFqksFmTlgOLbxu/EWE/+2MVSr4sgUsvZx4nud7iFM5R3Nkj8dkVLEJ6Qp6QGXhljoOPeo4OQhJOH3lQLTBz9m9EyXKuscfZPk1bQwdn0bxSFQ2BvCFPRCup2vLZRx/yml9mwHkqkemfXUVE92LiA/10rh0FYomUzMvxNjZJPULIqi6zGXvXYwQAkWV8M8CJY1LVmHafgYSccZ7klS1BkmMuX3usbnJeyFESYdYrq3Dd8fr8d5yO8bBfdSsW8mUIbNUnCL62Nfp6LKY2v4olr+2lOiRKmMo7R1kfnwMxT1tAekrRGrvIdzUxeTSt2DvtJGjrfgv/wQYIOo9FO5/HPKuho1u0njRFNk770KZanGyDl1LiFb+EE9wH5mp/8YMfZTi2WcxE3vxLv9DhOasPQtnn8YavodgcwHmvZ3sWUFu99epaHaFAIv1jAF1mRaGAnGazCgrUhXcE3EE9artCYK2c+5ytcuRTh7FUJzk0015HxP6Ooatw1iSwYRIwWiEybFNiP37SQadxUXA1onMW41h70URFqOuc9BZeYylNNAvZ/HKAdLus+h1q8C5Q8/TrM1jsrCPSXWSrd751CiwNxsiKVtkgawpeF7pYaPZgs9tpxOFPJMpH49KBdZHM1RlI7w330HLhKMefsJMYxe8oCWpsoL0y5MIWyKl5wgvacbqSdAtLEjgiMLaGpFcNdeffB1rOo+i5B/io6k1rO29mppkO4ql8QtvJ8OREzxwwGZJwmZ+thoy1eyvf5yxQC+PnZiif+QAB4eSfOapbvyaTLpg0hjR0WSJMxOw7NT7qMsWeLTj66Q9J/nJ2ZNUp5q5/MRdAHRhs9OziIObHmcgdxZc0tqi6LmAfHaEwxHmzZuPqmp4Z7kINTW1sOHGC/nong9SJ9Vz0n+KX1bdx7bRi5mejjM9HXeeO5f+PruHfCaaF7UxMjLE0eHDrOE8IP8/N2YDclkgv4TCaEkOEFHEuRVy2bQwZYn+zAmO5wfp8tRTPZTlirBKMV0AIRARDeXSMnUkNZlHoGJjoMquwnYxjSl7qMsonOBisC0WH/0Gmk/jaOcykoVqhmInufjp5wAQkUhpf5X2CJETz+G59gY87/8ERkHGNGyCla6dFhCwbDwDaYyihWVY7H98gMxIlo5ALXGRYYHZwEDBokGTEJYgnh/lxfH7iDQ2YcVi3D/9LP7eTcSyTnJglf8F1msP8czK27iuybGDyC57G6KQID/vmnNO8WxRt7ycn1sh90hU6c7iIqxFEB695CdccPtnfKof5apmit85gR0vUPjOCcg51yUABGwv0+SotIsMA7qtoiCTEjk8wouaNWl0s6qWv9yHmmu5keDex0G2SB26j9cF/5ufh25jXLYAQZ2vi+a6AwwfqEPyNWBOnkb4JOQqp+d01AAQ9HfGUc+aBDIeqvJ5rnvkcaw1C9D8CgFLJyXlkA2TwL2fJK1XYC1Yh79yM0LROV4QnMkXeGPdc1TYNYADVDVboSAMThdOge4nL0yEmKaqYYi+7sdRmzZzdPo5RhMvsq01Tb95GTtSMhe3ZYi6/pRZLcAlzf3scUX/3msXGegd4ebY9fi4x7mXXMqxb8qDv7COtHcQbzHMUPWLpfMk2YCQ3bup4FwfYOOUl0jvZoZMlfvVETZKbvIAwfoPfJL7vvBvVGYnuPuf/hqpuQ1L94FpIbCRDac+Imfc/i9XOMzwu8jbstg0dJiJocPYviDQhW3bPF0TZLbZhVTII5QGZG0pRvZZsnknoxyToghbYAubU77V3Hnr9Tz7X/+K5ouy+oY7Aaidv4hL3//ndH/nAcAZPAOqU4FNmy4dXltA8/KNDBz8PpZpogWvITf9FLbpfM/yK2+iqtXp65dVlWVX3EjLyvUMHNlH2+pNcyq24Zo6wlURJgBPoUBG08h7nerzqDnKRH6csBbh/Rd+grcUbyI2rrBuQ1mILFLbwNIPvJEnv/Jv4Norrr7h9XRs3MqZF58mVFOP5zkbeyRL19qLeeGxATJu1hijQO2UDmTwhiJc854/xhuKYOwZ44mnH+G4kqb/ycM0hxuRoh5Onz3L8cFBVkubEQiKokhVfRXxQ3Hnt0TK9DJwFoRLL7+Bp772HyXQA84kabhWKdO2jnfphYylCqQLJpKsEFs6j+8pd/PSuDvaw5rjFSyJ12JMp9CDYeatv5CfR3ezL91DtHYNK3wh+g/tIZecxhuK0LHhonP2cz7Ox0wEdKUkIGW7fZfTLhfOW7DJSiA1+xC95fmpkHf7x3UZpaEBs68XS1hINkgB75z969dc/8rf7VG4Y1VD6W8hBPL6Goz7erAOltsszMOTKGsdgKm4yaWCK0KGLGOFW13KujOPKTjVJzPq6Jfk2q9l6NPHwDqIf1HM+bk2ZLSyMGNelSn2DpMXEigghDNf6LlhZsTJvcU4AK3+Y7TUfZJdAxdyxLiapDuXFO1W/MPPcMHBP0UJOOeyMdSNmDSpU4J4JIHfXVtlzRy26SThdcVfqnzLbquVsnw5ZqEA+LAEiCovq25fBf/yJY5m3WSBncG2bWIuIA9JQwSwSYkairIfPTtGXf4gQghyVRpfTiSBi/lGbjUGMim8XB60aVFVnnFpyGHdOSddm2vp3jNOeipPoMKD5FbqZ4TdcKnzejBEZWOAysa5/cUAa258PaOnjzE10Iv8EkCw4trbaViyqqROvvGODop5E1/IVYMOzup5dam6sqpS27WYTHKKgaNxDj02gFGw3Ao5hKrn3nsvF0JR0FatQY4EYTwJa95A0+KryHnPVagGELqC946FnP3G/aipozSqzhrB2PxWtMXLseoSFH9yBmfiFqhXtBMIfJjcz35IRXA/0bYEubXvwqhdjTxL7iN5xecIej6K0nQRqWAjmfc/jj09Tdb4Ovqq27F77iF/9AWwBZPDIHZ9BtsS+GvyVC/r49mxPEXVOedDqQB3JOrwVS5A6IIWc5oeeYwqfwV6vAcUH5loNfaSRVgyyLbEOl+Ms7LNVNZDzm8wLMUJ9o1RPPsUAOmVTgEsbKnIRGiVsvTbntL6od8eZinNrHdFDOO2A/S9toZpWTw62OJY1lbDpJQiW4xToUbYpgi+7/Zbn8woSMV6Hgo/jzYtQyVYchG1kKCohTiWKlLlhfXDnVhY9FmD9CZjWJXO3BlK5SEM/eoYz8a+xxuGVPbuXMZosBMaHLHa7uhBsmqSRaOb2HWii82hrVw98Eb0WS2D63uv5+fhf8aTqeOiQvm+a5lczFigl0JR4uBQEp8qU7Qs0gWTmqCHL962HF2ReOsXd7JypB4JwfrDH6PplkFOpfcxr89JdhWw0RBszGto2XfxGfGnzv0sZLrCC1/9fhWCK6647mXfqw83goDx3Dj39P6UhJbg/rpf0phpZJt2CfHBydI6oz/Xx3ODz9GpOet027b5XuK/6anvZkvjtpfd/28rzgPy1xhCkmb9W5Q13lxgXnCrtKqk4tXnDrShXJ4pN0u+23+K4y3zuD4jsIcyqEJgywY9dVHYO05NLEGoo90B5GEVMAhVhKifV83dDz/Am08f4diCNwDQdvY+avOnsEen2XTWqcwlAnKph1Oa1YcudFewKRpF8/rQXmacFpKgsqk8idTODzPZnyb9ree40rOCE7kChyyLjFJkpW1wdvRuTNtA8XiIehzqyIzSOoBXShDVQnx46cdKrxVaL6HQeknp72++2Meu3jgfv6KTaCCIonooUnAB+aw+Dk1mWcUKfnL2Byx2e0wsLGRkVFFkXnA+dVoX3ek8bbfPp/jT09iu+Iqo82EPZagTXqaZYkByJisPCiuL7exRzjCv7Q68/k3oeZthohi/qEWEj6Fc3EDhOT8BqxPkYwyLxURSX2CDF7rFdoKWl4qFt2GaOv5LtiLUEFJ6FyH9l0wZToZtuOgAj+1tA/xXyxX8wHML2aJFpmiyGYW/x0bPmaR8EJiM87OmNdx8+il4sZ+0+Dl76xbhWxWgyWtzSfYTvE+9F2QQtkWFrTIsDLJulaEoFckVK7nH08544gjrHvgpu5Z24cfmrwq3sspqZR7wNyOCRW71uKgaTFzwSXjqDJIk0zm/ht6T4/xwzxC3elT8ooiY8cQWARRLI5xuBWBF0xrsoxbCFcCRdBlkL5gFbJfG71dqKEYdonF19jlMywRC+Ovn8/Nem0eqLuGOgR+jm1ly2bQDyL0hXui4jjp7jOb4PKREDosBvPY8isRLXujCNLC0WuTCGFJBYNkwLYJsW3E96X33lG6foiVQ9A0ISUdSW8ibTh+jKjSWmE2MkSJX30D7kmU0f/pLCElGVsrDZKxlHqHr78B+oA9VqSQoOaIj9evXsjBTw5k9IYxiiOv/8t959rvHmRqEaGMFqdHvEm1sZcll5y7CwzX1hGvqz3kdQK2uRr/pVjRNJoON6vaCDBnOcW+qvoAKTwXLm9bzrPY0l/vmem0rmsbWd36Y/ff9CIRg/qZtCEmiY+NW5znc67gLkDHQ/YESIB8JJUlUZmlIh7jorg+iqwGMx/ox945TI0U4ziDDiVGKPzqFdtdC+vrOOo+nq9GQVJMsji4mraYpFosEg+cK7bWsWMeq61/HaL7I0R7n821t7RQKRXp6zjAmRZhMFym6YlqNES8+dW6lp1lfRG/uCJYqccUbP8yFNReRS07j8QeRZJnnDo9AGmp8dWy44w5+/nfHKGYzLNx2JbKqvfSQzsf5KEXAo1DqIXc1YuIuQPe6Y7mnK4Q1G5C7iV/NqyC3z8Ps68WWbTBACb5EhPM1htQZQTw/jD2Zh4AKqSLWwQnslTGM+3tZ1P4Oqs0EduJpYA9CUZxmVs2HFHBAu2o4gl5mxNHByGqLwXLWLtbhF8F9TDNaeczLqQr5Iz0Yi1oBWLJ9Pvvv24lkpAmqCRJGiKjuaoVMR/BFdGofu49o9RHsdZcBa8n7N5Kb/CECMFMW1uQEHp9Gm3YWj1S2bQPImGlwK+Re2Q92FsnMISTXam7TeuxHnHOe8ch4ZYEcrcVQQzjoD8AkWidof/GNyPt1jOqlNOrjHMs5yfym/idQgw5gCnqc/3dW+UkX9JKI7XCLh5aWKMVnnSRwyG0TkFWJVdc088x3TlEzrzyuhWt9CEmUK+SBc8e8mZBVjS1v+0Oe+dbnz0kMqh6d+gVLZ/0to3penslT07EQX7SS1pXrUTQP89ZVMzmQpvfAJEefcsTEZEVQ2xF+2c//qrBfAYzPhKTItL/9GtTeIPwCLC1ErvMm573WEPKWesynBpE31yGiHvRrb0C/9gaUM/eTG95Net1Hzv1OXxWJq78BOAAl/K+fJfPNr+LZth3tkksJPfYA6oFhplJLmNxtYE44VHPtpjvR7/oAS390in0vONdQMXKonjMIaRFYNpvTdYTPPELbmTOM1NTA5k2kRJZ48zZgjIjtQ0Ki3QMZa5IdeBiR4izatAHt+jXY6RTJqhh0nyISjcIorM0tIioXabZ07tNOMCzHKRbzHNKHEQiEPw4FwNI4VbCxJIVo6gx5f4ak30dP70/panszeckkIwpg28w78QjdbTeRja9iyjsOHCPAGbac+iZPzv8Djlf3YxoNbLU62Sf3sEc/g99sw5Yc/DGVlCAMKZEjkj7Lswdu4WjkevLaJDCOZEs8Ov+b2IrFdabOqYlVPJv4A3RgJHCW6Lw9eA9cQ126kTdMX0ZHbAPDIwlkCphotE4tYVfTAyzMBsh4FD55+1Ja6iL8cEc3l3ZVUxP0YNs2t+BDcjmsnXkfF3AlN7XeyAM/P4gNfD9UoLkgsTWnUnjBx5pVW9mlPcG84Hx0+dyWzF83Kj2u1apV4GyqGwmJZTUr2DX+Ilk5x2axqSTU+83urzE8MsxnN32JRZGlPDzwAPsn9+Lxerij8w2/8TH8JnEekL/GOEfUbaan3AXkeZyBYH646xzKeihbKAHyVfPrWX6Vo+b3wpdfJDcokbdSxJ8vi5W0VfwIs9hRUkX0+T0svLCOPzq8kE0Du1l49FvkPWG6jKcJ/PB+sj/8Hsnek2SeeYxQygVOfj9iVo+EmPGgfknF6lV/s3AAusZzpO57nParr2DZR/6UHWcn+cR3HuYWIw6AoulUeFwakDIbkE9j+V6qvFyOwekcX3imG9OGD//sMFcsrOYHqS7k0B5sYZOYAeSqxEAix/rYhfz80gcIqiGKpoWFgYyM185xaeBTvPlbJ8gbFv9yw2K8m6ro+flpJrCZqvJx+5BN0HauQVxy9nsocILns53clF5OVPihehHGzGUwwJ7IUfyRA1wirjfquB1hMv86jhrtEMhQ42YVs/J1uIl8LP8a4qxByJDSIFbVS8jK8wbjPzg0rUHGGagEULG4Eumidvw70owffoGHo2s5EqygNqixaugo4be/C7N5JR996ARF08arSjxZ+xYaxl6gLqwSap7H8KFDWG6GJSvnaK2ax47wH3K3MYBYYvK2doNmMclofyPxhHOQsmlzPKPS4QFTshgOLAPO4Pf7+ftrF3LRsVEePT1JRSJGfmoITXE+J6RywqaqNcitGz7Cfc8eKLkFhCIerFSAXHIas1hA0hTkbAbJLGD6IqV+YYDnrWr2HBomp1USu/B6pp6+G9UoUgTqYxFuueMG7nz8ZgKGQC9U86muv0ZRK/jFk98q7UOyfPj9d2L7LCYiCs2bKti2oI6Qz8Nn9wmEW9k66VlEjeqwClTfZcQrJjCxnSq9MZ89GYP6LW416WWoTACyT3G6GeUQfleYzt9cRWdlM917DxEfyrD7nmGmBp3F1Ja3bMIfuRAhKyWf1tcSwQ99FO2H34HR4dJrvQWntWFeyOlH/MPFH6Ej1MlVTecyTmRFYdX1r3vZfQufYwBoZ00kqbzoy2hZ2trWc51xBd4dgsLEEXBBSP2q+XDoCGNSAiOZQxyeYHDQqbzZrmhMSk1R5atiKhJnbGyUUOjcxamQJJZceh1HjhwsAfKmplbq6ho4OprmSLySDZkCY2nnnmqv9KFJc0H09uYL6Aq/Hb8SKFHcZrMMbmq9DRtH5M3nreDid32UoWMHWbj1Cs7H+Xi1CHjKFfIZfm/cHUf8rsy3t1pHb/RDsuA4WJQAuYz3ltuxczm6F/RTnEyyubPlpV/xmkJIAvXWDqyRDFKDn8J/HcGeylP86RnsvhQBbxMBwPTOI6McRbjtdLa/BknzY9s2Lxw+wXqfVgLkxmBZJM7oH8J2i1LZWRXynKoAeQx37ROqcl02JMEFNbuxbvkAnv/ykDgJqV6BYSlYhoQ62EPtxHdJV67FzASwzfLayThxHNsTptXjVPtzlo3urqUyRgLJrZBXeuq4tuk9FDMpZvqQ1LVrKN57L1BDijzer38Z/aZbSWY8lAE5XNz0Hbx9jrCllBmjwVPPsdwlqFKBDvsJwqtjGMAlnVUcHErykYs7uO/ICAMHnXG2LuShLlyeA0Le8nK5rjPCdX+yAm2WlZuiSoRrvIwm3Qr5DHvrFSJcU8/VH/3bV93mV4U3FOGWT5ZdOGRFYv0t7UTqfJx+YZSGxVG63N7032YUm7aQ3PbPmJF2UMtis8raauSlFedoHhTar6TQfuWvtW+ls4vQ3/1T6e/k1n9EXv0+7HAbkWyWzH9/AzuTQb79gxCI0nmNTv/RZxhP6Ch2Af8H34N1NI/5/DC+jmqWDuiwcDGRClf0S+QYlyqBMaJ2gELfDrSmDVTJG4E9DElx1EuvRe2IYpomp7/xJQDa1iyHX6aptFupNED2j+NTPGTyeV6c/AVHm5w1tua21MkN09Sc+iHR3ePoyTH2rlxJsquT0Xl1tOz8GuMLVoEOvnSK2MijFLQYAw1bkFz3Hdtr0rHpJM8k9oAUo1vuY/l0hCMR5xkuaOPYLkM35+kCdqEbXiYGP8Ow3zkWn3mcBFAfH8CUi2wuKlyi/APT6j8yVmxDVcY42f55/naymyGvwv7MdYSOXsYwTsJte+TfeSD+EWKZRjb0XMuKVAS5IOja8XdUt1Zz1/xrMcIOxjizawxlsoiBzXHVZHFRYdfPu/FFPNi2U+i7uEnme7sHaBQyHVmJrYkbOVTzApc1/nr3xiuFJmuEtQjTrkr+4uhSPrn6H3nH02+iP9NHpiKLd8ItRLlJjH879Gn+dNnH+eKx/wTgTfPfRq3v3Jba32acB+SvNV5KWXcnPcnNoBuusNia2Fo0RcZEQnaz7KFsuedqdva0IjLG3t5qwI9MngbtEL2FlXRPutRWbAwgOPwsNcMh/ELi88tu5JuP/y3eYoHKd6+l4PPhe8td+IDEd79B4QufB0C8BHgrHZ0Ud+9C6Zqblf51Qng8YGTBrahuaK1gWVMluK4Oz/en2PHwKMQgp5SV0X9oracv3swFR0dY1Rih0qeizLKP+/bOPkx33XN0JMXRkRTgQbF1vFCirI8XDW74yk48isTKhjAXzqvk0FCCPxXTQDUZe4p/fuxUab8fv/8YshBMz0zSB4aRshpbvR5mW1Wn5TwvZlR2iwIP7fgnbO8CKjsGEVuuJr/oGsydY1jHpp1+61YLBpxBfJrbGfHuBzJEjz2HUSNQIsuQV8SQuqIY9xzBzijIS8PMv2ERsewKbBu+ZttMZw2Gkjl8qkxN0IOuOhP7NdsuwN66mXcDecPCq5az51cAC2qCTGWLLKkLosoSudxaVFVj3z5HKReXtphWk7SEavjj5R28blUDPVMZNrVVIAnBt6eyHH18gOm9U2xrqeD0SArJ8GApefr7HVVdn8+PEIIrF9Xwxi0dHD9eRXf3aey+bvoP7sQwm7BMhwq36XUdCCEcn1lXI88bUilWhMklB5iSevCPjFA3qdNy2XZalm7lvl/+rPS7BtI2Ocuis8rP1bfczKnGSkLV9ZwZG6OtbR41QQ+LKts4PL2bttp5tK90ElnhfVGmp13BFm+QsOpl8cX1NCyKltQ5hYBQMFhSWL95XQcjD7q9iJKfyy9egrxjGDIGGQEDuswNS17dr1pobjIDSgJ7IqQRrNJRNAmjYDF0YhoErL+1nVDVr6YL/qpQ1bne3ofTB0GGeSFnjIjpVbx5/l2veb9iZpGZMZBn9VNvTi3nytxaAOxp5/kTVTrypjoq5oXwnfGTyaQZFwmkF09SLBbxeHROVp2kZrCaAf8AVd4qfCtCHD58kNbWea94DLNFVZqbW/H7A9gtG5iYGmQiXWAq60yY7ZU+NHnuwrLSE2Nt1Sv3eDX4G/nA4g+V/q7pWFDyxj0f5+PVIqAr5GRlziJp0p3Lg27GVpMFbSsq4ekhbMoe5JpXQV2xmPCKVUw//24OTu1nq/6bV3xmQgRU5IArlNgVwTo0id3nzLUj2T145RZCWiVa2xaM8d0YBycQFY7CdM6cZt+gwpC+iMsCzU612k2EAeSnVbAFRVmiqJTHgqzmJO0sN5nocedqU5JQpTxapY7SESABZM+mKE6XhaOmjyVQNltgSAg9gp2LA2AcP4pVFSWiaJg2nM1bROUJghMjTDGFLBxAHtZcG7CQA25tq4jk85KPH+Rpfxs1O7+Df+BZiocPkU0lYZamj3n8IXCxoTJ1gg69m77qt1JfOUF79SD55i4SwIXzKrlwnpNkGE3m+bkLyGtCOnXB8jUL6XPHYI/v3OVz85IKxk66gDzw6oD8txVCCBZcUMeCC36HYEIIcoteIemr/y/DDFl1gD+OQKL/ne+ZOQT3/4L171jHc1/dT+28BqRoFLHRRmoJIKq9eG51tBq0TBq+/kXSIl+yQosUJIwT38Xb1k6VEUOzFfKiyHhmkjqi9PWdJZfL4vX6aOrqwDzRg3XKXfTU1tIcNDh27HAJjAMUXPejpi1bWGB9lcRYHyO7I9SMjnCyq5ORYABjYCfDgQRUraBC9/L3H2lBGvkF7xmYwhQaz+MnTphhvYpBj7NvU4ZT3Z8nt9ahfxc1t53PtlGLMtgCIWzi8jQexWRt3910r/QwTCvtVRl+0j9EjWkgKxpbrlU4NlnD8jN/wTtHT2IjCG1r59ADFqapoooMW0JfZl5gH0l7mNB0PSuGHIarWbB5ZN9alp64n6GChdyykpTcUmqXqFkTI1cp4985TXoyTy7ljJELLqhlRb2X3qksTfUhVkk61a1Brqx65H/FzSTmqSoB8vXVG/HIHj667M/44I73sEN7nm04dPTN9RfyZPpxTidO8o5n3oSNTVugnVva7vgfH8NrjfOA/DWGkF5SIXcnAMUO0+fvoyfQQ0yvosnvZMNNISPb5wJyfRYgr/WcBKoRWGxdfZSGzhAv7klx7LizjUaBPOArThB94C4OegR71Q70G6LMSx+kMG/BHLOD4I23M/m972HHp5Bm9Y8D+N/7Abx3vhGpopLXGlK1KwxTXa5237muhV1u+3Dakhmd1gjEIKeWaXwPyqvpnzB56JcOXU6VBX95eSdXLqxhPF3gnkPOBPjuza18ZUcPRdPmzeua6M6OsNeAlFvJzmCjyoK8YbGjZ4odPc4D/zFRABtMBWRT8K5NLTxzZpIDgw4IW1Ad4D0XtvKjvYP4RkxOjKZh1tooZct0VgVoinoRu3spdA8xOWXgu+kDJD/4FszRETybr8fs68b46U48N1xPXtcZmz7CaLUzGMf6T8Hq5WhvW1aqTKjvWA15EymgIvtVyLrUeSGI+FQivrkT/EwI4RhTeNVzaWqtlT5aZ/2tuy0IXu9cC7S0lmCR31FVbYp6aYqWQWFz1Eu2NsB+pqj2qATCfk4n/BSUPMeOOSpcL6UYV1bGqKiIwer1rLnhDvbc10N8KMPG2+aVFibarAqCN6RhGxEmesCnpaibdAT+Vl9wPf5oJdXVNYyOOmowBVfe/I1rm5Bkmc7NzkBfO7/cQ7Qg2sHh6d10RMq/vrIyVgLkzfPr2L795UVAgsFACZC3NVQxMks13hNQEX4VO2MQ2tbAzStiv3oycAG5Kpz/wAHkQhIsu7yJwaNThGq8NCyMUtXyv7MomxEgmYmEcH5Pe/CVge6vFe41s7NFpFk2nW3FejJSjuD2DoTXUbMVDf7Suamra+D06RMMawlI2aBCQ0MT+yMHuFe5FwRUeitp7IrQ2bnoVQ8h4I6F1dU1JUHHCvfZmMgUOTPujCVtlf5zKuQVntc+jp2P8/HrhE+VGfOGqZvl3jflVsyD7muaIhGKuRUs2yafLVfIZ+KKxqvJmhmWV5ZFDf83Ql5SiXXIEWmTVsSI7+rh7NQg66uuQZ23HaVhBcaDfUgxp89yuuiMlSO5IM/c+yAXvKVzDiAvJBRsey5dHZwKuTFr3aO63suGLBBugkKP5FC8JkYWiiMpx+YDyI7JBH0WZCTkaCsiEKZ4dhfG8WMk19yOecgDMkyYNp7BZ/EcfhRr9ZXkeHm7oZlkaGN4msonPoRiOnNq8YXnKSxsBiQkYWPZgoyhYkQ7UKacJL0iiqzdqmBGNlDMLCe/4JZz9r+6OVL6d13QQ124vFAI/xrAcuFFdZzdqTDaDZ7fEyA/H+CPeLj0w2X1GCEEomFuL7/X60OWFUzToE9ymBpVy+cTec8DGM9PIu0bp8GqoFsepWe8jzrmcfz4EQA6Oxc4bLfVVWVAXh2jKSZz7NhhABRFKTmUAPj8QVJbPklk6g3Ex8JUDY0hbJvpYoGsrhN31+o1Ta18bvsHyRoZmh7/Eyik6TGuYXBokG9xC5Yojy0HLt7izL9QEunTbZMbm3/At6bryfo1MoFesj6LpjaLZ5NOsaHDep5GBJk1H2ZiyZuQvRUsBuT6t2E9/ylSm/8Cum5hXWCS8eefYUPmrwkpo6RXfgijO1UqvNSoR0mZ1UwZzTyVeJfz4nGAKYQECy6oY/HFDUiyoLDaYOBonMmBFL6wh6q2IEII/v2mVxdv+00jpsc4nXR82jZUORbESyuW864F7+WJwccoFIsEigHeteZdrEwu5xPPfwIbm9WVa/nTFR9HlV5+ff7bjPOA/LXGSyrkFY1+Khr9DMUGedHrINPLK68qLVwtSQbTyZD580VkScK0rDmAvMF+kc3BE+irLqNy+zvIAYs6TAb+9TmSaR1NlQEDpWYxVuIAUnaC1ZJrCOiHbLAs/gIgvF68r38Tmc/9B3Jj09z3JOk3AuMA3tvuROlcgLpydem1xsogbm2W9fOqaVywjB9NxMhRXjh//vUr+OXZce49PMJQIk/RtPn3J86wZV4l33yxj4Jps7QuxFvXN7GpLcpkpsimtgoOTEyz9wU4pfdhSdCxpIbnLm3g1HiG57snefzkBCfGUiiY2Aa01wV55LqNBDwK1yyu4R3f30/RtPjH6xZRH9bZ2FrBsaeH2DeYmvO7IuFK/uby5RiWhXHMgzxapBBXKLz7vaVt8o/90Dl/kSgRw2QEOBUbxRQC3aNT/6d/hWf12hIYB5x/K6+dovybxEsBuexRuKHl5lfcXnMXF4WcSXoqj2z7gUkKhQKKorBq1dpX/b5VV59Lv5y9CPUGVcy8q3B72pkwpKYYgQqnt2f16g3cf//PnX83V5JSwmzvjPFKcXPrbdhY3Nx6W+m1iopKzpxxngOf71xP9pkIBMoTcaQiCrMAue5XUC5uwBpMIy37NcA4IFyrQ4+7SLUFTk8n0LGumo511b9yH681VLU8VBdFEUuyqNZrCKqv3Kf464SYqfJkTKRMGXn0ega5u/MJ/nrpP7/s52YA+VAwiT3tjG8NDU3oWb20MIh5Y6Tzxst+fnbU1NRy1VXXU1lZZiZU+J3xYyJd4NS487x2VvtJzvVTpFI/D8jPx28nJEnw0uFgwqWsh10dQk2W8PoVLBzr0/SUa6UzKzl5ZdM1XPkyrST/0xD1PqTFUciaKFvqUY8G6Onbx/LIheh6GPSwY9PqCqslChPU6EnG8gG6dz9P88oNhHvPlvaXTyrYppgj6AYOIDdnWm1sG7nXYVFZkoTa+yjy5J3I6RECdTniZ5xWJH9tHlm1SfR6MaeHkdVG9NVvRygaauMWcke/RvwLRfS2t2BbJv7uh6k48yCSbRHwTJFJT2O753rfxOOsrLzY+f6AA5DlpmaUF55HRCvwXHYFuR98l4LL8AnXNTM12Eciuorpa/+Wim9vLAlHWr4qrFAz8Vvve9lzGvNrdFUHOD6aYl7MT1VAo63Ch2FZRH2/Hu07m3TQiv4qPeTn4/cfQgiCwSDx+BSmsKhQQzResATh1ZGaC1j7xmmyKumWR+kd6WVlPk93t9O62NXlJJlFg99xjBnJIqp0GhvK66LVqzfQ13e21M7l8/kohjYx8c4j+G6YgG99nVg4ylgizkhtTQmQVy9ehiZ70GQPiSucav62+BTf//63KLhTdH19I4OD/dgzYHxW+CqrkW/7MJvf/kbOhsMMNzeR0nVeaLmD9KH9SJhU+WXiV/wYo3bVnM/mF95OfuHtpb+bllTQtPga1CeeoZAaILPyD/BJP8U6ZGFKRbaH/42cFeHB/N8SrKuistlPsWBRzJm0r45RMUvQUPMqtK2K0bbqldd5/5sR0531RLVeQ1uwvfT6be13clv7naW/hYCb5t/E0NQYfiXA1U3XlW2cf8dxHpC/xhCzqNZIAtUjs/1di/je6d1uZgjWxMqZOVM4p1iyLWTLpi4WZSydJ9rQXNpGiZ9khX+IqcVvLXVAKZrMxjev5sgT/STD7Ux0H0XpvJyJZR9Djp8h/LNbkdMOsDCDc0E3gPe21yHXN6AuWXrOe7/xb9d1tA2b5rwmz6rcNVWFueHC+dyc/D6TRxMc6u8DoKpC567aFu7a0ELRtLj9G7voi+f4k3uO8EJPHIC7NjQ7VKuaclY54lJz8gEL73uXEqsLMzGZoiPmpyPm541rnd9tff5nFA0QuuoK8UBVwMOP3roGywbPLFCs6jLCVsGWQDiT9Lr2DoIuQE0vv5xw+scMPx4Ey0JZvAT/u95Hcd8eRDCIfvV1VL34LCOH9nPCTc40NDahX7Dlf+08/yYx2/YB4OKWS18VrM30vuWSBbLJIopeBrQXX3w5VVWv3PP/SqG+pEJezDrXUpgWQpa55Pp3lt5va5tHQ0MTExNjfOyaVXg8r07nrPXV8b5FfzTntdkA7qUJidkxA8hlWSEQ9CMkxxoYwONXkcIa0sso4b5iaHMH65nq+G8zZlfI87Kz6J/pH/8fhW+mQm4giuXJ/Wu1P2VDbPMrfqyx0Rm/BlIjJf/2xsZmPGccwOxX/HgVL2mSv/IQhBC0tXXMea3SrZAfHEyQLVp4FImWqI8TiZdWyH83k/v5+P9nFBQxuy2ZyZcB5LIsYeHg3viQw+aYPRb+tkIIgXpFGQAogSA2FmfHn2dB3RXYZh7PnUuY/ta3GEx7OJHYzbUNZzhur2Rvr8Xxpx5idU/ZZhNLUEgqJUCuSQYFSyE7C5ArloV9+FDpI2O5FNnPvYHlFSMEGuQSIA83Z9FCBoleL8bpfcgLGhGK8+zK4Sa0xqswp7oBsKUE807Pck6wU+StLMk1EsNnjnKieyddTRvxZbwIlzWm33gL9nQc7+2vR+6YT+7YEUx3rInUO4B8qulqWoINGLWrUYd2Oj/R96uTpe+NDdA3tIeuyGoE8F6exzQLSPZqSoPdq0Q24bCXzlfI/8+PxsZmpqfjrFy5hrVrN5bm2Zn1QKNZCSqMTY7x3HNPYpom0WgFsZhzHwkhUK9rw+pNInWE8QnB8uWricenWL58FRUVFQwO9uPx6OU5XNGR6xsIfuwvaN7xDGO7X+DMwoUk3XVKrK7hnOOMRKKsX7+J5557Co9HZ9u2y/jOd75Wer+ysoqJiTEAvD4/QlGY9/G/peX0SU7V1PDk049x5MjB0rbJWx4H5ddspRMSqW3lHv7aukq+tehzFNQsH0iA3riIqy/bRKwqxPh4siy78XuO9qCzpthSu/VXFlokIfG6eW/4vR/7eUD+WuMlFfKZ8CllQLAqtqb0b9sVSvLINlVLElz2umvIt16C7PaEikIKOeU0YZvRuYvSSJ2PTa/rJJGoobK6spSVMyPtTF/zbSJ334wwC5gV5y7MhSTh2bL1f/Zbf42YDchnRLBqgz6ocKpmsiqhaOVJTJUl3nthGx/7xdESGL99ZT2b289V82zyN/PW+e+gKdCC8MivCHqkkAFZsOvn9v6q8rlZrmCljkBgCwmB5fSLVJVpv/7LPoGx/eOErnwR48RxvLfdgfDoqCvLmcRotFyVC4XCbNz4+wXjcC4gXVq9/FW3V91qdsJVoPfZVUSacrS2tjN//m/WYztb4MYbUonWrWPw6AFqOhawcOsV+GZ5pgohuPZap4Iv/4Ze0JWV5esw2yrvpTEDyEOhEJLk3I9FV3zJ4/8NhsCXtBKI37JgDsztIc/JzjX7H9PVKfeQ25kiUpFSdTun5GgOvLIIVWVljC1bLuHppx/Dtm28Xh/RaAW67Ezy/1MqeaVbIZ/OOWioI+ZHlgSaVK6QS0ImokX+R99zPs7Hq8X+GsHgcJGNwoNsw1lXNElFEMShrLsYHZsyIJ/NFvpdhRpyesvPTu6kdSyOUBN46z/P2cxpDkwOELVzVBbTVLw4BA3VDJ84TMos4JdllJZmjDPdZJN+Mm5Su6axnr7eUXKaUhJ0ky0b69A+xLJ2bCF4pNBFcQTShsrmugGkaBTbtvE3jyMFQvj/6MPkvuv4bVv5ONp1Synefxa1cS1qo8PCkhc3ws/KvyOTd3qw1c4KzJRzHge9/XRk5iNcP22luYXgxz9Z+oznzz8Of/dRJEUlGHOSybmUU6nOt25HHdqJLeRfqRpu2zZ9T95LLpWgZ9+LVLd30rPbsY89s/OZX2mVaBTyGAUnYXq+Qv5/fmzZcgkbN26Zo2MCzrwoqr34RiFmBRmXkiVAu2bNhjkAT4Q05CXl+e6CC7aW/t3W1sH69RcQnuV0NDsWL17G3r07GQs7z64mJILBl0/kLF++GlmWicWqiUSiRKMVTE1NIkkSq1at5eGHfwmU2xiV9nko7fOYl83w1DOPY7luN1U1jb8+GH+ZqPPWMxg+hS57mbhlr3MOfrv1iN8ormm+nhpvLWtir872/D8pzgPy1xizVdZnA8SA4jxE84LziXrKg74lOadY8UhUdaVIJbsxZi2u5bhDgbG8MWz95ZXPQ6Ewa9asn/OaGVvE1OseRRRSr/i530VIsyyhFK1c5fRHnYVzoOJcpeqL58dYUhfk0FCSNU1hPnhR+znbgAPa3jj/rb/yGOw7rscz3I/duOJXblvVFmTb2xfw7acewZPQyMpZ2sIv+X5JRlu/EW39xpfdR0NDk6M8X1nFNdfchN//P7Oz+d+IlwLyl1bMXxozlErLVdNrXVrDmuvWv9pHfmW8tIc8UtvMVR/5m1fc/jcF4jMRDkeRZRnTNF+1Qj6j8B1y1bdVjwPIVV0uaUC8plAczfaZp/93Ach/WxXyEmU9UURWhWOjJwTLa1ZySf1lr/rZpUtXEIlEeeqpR1m40LEgnLEq+Z8C8oqX0EO7qp2kyuwe8gpPxe+NWnY+/v8Rik/haW+Wq67uYE1DhE/91w4S2IQQxJDwyFKJEm4DxXzZ9ux3HapbtSvKUDz5GNo8J2E3kHGAaUPapPexGJ58iqpQgLGQn77KEIvVIHLXYowz3aSC28loZwCoWnkJfb3fI6/IFN2xWrZsBCCbFoYiU3QptDsnmohe/mGqX78FbJtksRdbj+CNtCP8YTJf/xzaxeuRl2wl+7MfI4sFCNWDiAZR1jYhNzVj9vWSVWWKxQJCSAQqqkqOCaOFXha+5VrGxrrZ+Y9/xuJLrqFtTZmtly+B4CB60AE3ObdSXWi/EvuFf8Go6IRfMV6kp8bJpZzPjZw6im2V23gOPPBT2tduRpJf+drmXK0SSVFRfgXr63z8/kMIcQ4YnwmpOYA5mqWJGOMu+2LFijV0dr66P/ZL9//StfvsCAZDdHR0ceLEUQBitXWvWM2VJIlly8qFoebmVqamJqmpqaOxsZw8f+naz+v10djYTF+fw4apqXnt7MfZ0R7sIKJFaA28/Lr9/5RQJZVNNRf8vg/jNcV5QP5aQ3r5Cvn66o1c3XQd2+q2z9ncdgF5UXcF2nofI7vq3eVdTDk9sMZLquO/TliBl/cu/l3GbC9fZZbNW7BS56K3dOGLnDvYCSH4+2sW8uDRUW5aXjdHcf03CaGo0Nj2620rBFUtQeSADAnIqjmq9dc2QFVWxnjLW/4AXdd/Ixur30aoqloSKIFXp3CDQ92fieq2ICuvbn6VrX/NY5jdQ/47AKmSJNHU1Ep/f+8c+vpLY/HixfT09JcYJopLOf+NquO4AjGaBAVXrf33VCGfF3ztY8Y5MQs4SG6Kwe8P8A/r/vXX+nhTUwuvf/3bSn97ZGcMqPC8eiXqV0XFSwQPO6udpNdslfXK84Ju5+O3HDcsq0UgWNUcKVk+TmARQqYS4VTIczNe5eX4fQDy8JLl8APIeDQm/Dr1ikIulWAi62gwVPWOYxZl5PYOmieGGAv56a8IsTjajNzmLK4Lw9Mly7NYSweScGTbZmjsqgteFMSs9joPRiHPU/c8wHXLtqMHghiUW0n0y6/Ac8l2xxcd8Ny4icyXv4jvD96LusAZk9U16zD7eku2sFWtbai6XgLkidEhhkdP8ORX/51iLsuun36b5uVrSuuPGRCtB0J4XabATC+3GWln6vYHsD2/umI90XOm9O+Rk0exzDIgT42PcvrFp5m/cVvptXw6xannn6Bl1Yb/r717D4ui7P8H/p5ddmE5CiygiKCCu6Cc1UjCYx7zkEraQaWItEwze6woOyhmgh1NLX+mRZb5lKX2qJnfR+vRIg07UJZaapYKapxFQDnszu+PZUdWUEBhZ1ffr+vyupjZ2dnP3u7OZz9z33MPXL20qKo4VxeHW6vMFE3yEQLdgB8KEOLQEb8qchEU1AV9+vRt9deJiel1sSD3af78M5GRPVFSUoyYmN5wdnaGu7sHysrOSj3k9YWE6KWC3Ne3/TXF66Jywb8HboKDYP1RQNc726gm7Mil15CbuarcMCfiKYvh6gAg1p1NPe9s+hKoTn8PoapMetyh2DQDqMGzFa4HlYGiXi/npfdt9gt2h5t342eJO7g74b64wAa3ErEWp3amg5boKl5V4nR2draZYhwwFYn1z4w2dlCuT+OuhjbQFdogV8TfHXJ1PcWXME8Up3AQrDZk87bbbkdy8kNXHKWg0WjQv/+t8PMz3QbGwbHuOkSXa/jsqepN3udh/R5yR4Uj/F0aXmvWYmqFdGJRIZrek3mm86vhUTeE3E9zbUnfSaWES71LXXQ+jfWQ8/pxaltjIzrg3Xui4eWshrIu35sndvOGAmqlYDFk3UyOIeuuXloEdzHdEvJgRy1EiMj9zTSktJ2LO7yn3A+X2Y/DY8Uq+IdFwam6FtUOSpx0dYJDZ9MJ7eo/j+J83Szrbj5+0NTNgVNel6dV5lFG9Ybg3jJlOtp1CMCF8jL8/Pn6RmMT6o2kU/WIgMeSN6ViHDAV5ABQ7GL6vRAQZpp52aVuEtCzZ/Kw88101FwwDWc/X1aKo9/tlp5fdc583bY7nOruEHKhriAHAIOXDkaXpo9JhScuFuQVJYXI/e0nAEDHHqYZ8n/d/hmMdSe9z5eV4v+WLMCPn32IPWtX1L2m+cQArx+3d4pANyjCveAdH4wHHpiBYcNGtclvPq3WF0FBpu9fh0auH78cd3d3jB6dKM3n0rGjaU6lxobHd+0aAkdHJ7i6upnulnONHJWOUCrYn9vabKeisBP1h6yjORM51X1oa120qG0XDMFYC1XuN9LD5h7yS68ftxeCIEjXw19akNuyoOCu+MbvG3iGXj+9bOZecYVC2eC+1ZdSKAQMmhqGgSmhrdabo64b/qxxU1utd+BKQ84uR1VXkDu6Xn1BLtQrFgX3tj+pdGkPeXfPcChb4Qy1IAhSL7myXg/51RodOBYP6B/CHV3ubHrjJpivI1cIQDefuh7yeteQs4ecrMmc+gvrbvUl9ZCbh6zXq8jl6CEHgNiUWXCoNeCcxhFHCs/g5P4fAQCB/QbDOfkBaBInQuHsApdxdyA433QbtENFeTAGBMAI4G83R4iCAIVRhLOHJzR1hWW5xvS9UwcEwnPjVqi1phFJKo0zAiJ64qaJpkvLjmR9ieLcv1sctyqmJ6BQXCzIQ0237PQK6Ized9wrTYLr01WHmDGm+wMf2LlV6sG+UF7XM+12ccj6+bMlyHr/Lez75D2IddfPNqXo+J8Wy7VVF6BQKnHLlOlwcnNHeVE+/sz+xlSMv/ECSk+bJq49c/gg8o8dluLghG72T1AKUA0LhDJKC6VS2aa/aYYNG4l77rkHISG6q95HfHw/jBgxptE5gJycNLjrriRMmDDJpjqSyBJPcbTUZYasX1ZdD7lS44rqoEFwKP0T6uNfoTp4pGl93T0yaxuZmM1eKBxUMNTUWAxZt3VDAobBW6NFD8+2uQeiHMwFuUajaXbyaM0kow10hW9XN3QMk29Og+YwD1l3usoh6wAse8itPGT9nrB7ERPa6wpbt4zg7ACxvAYKXHsPuaejF+4JTmqVSV68nFU4UXIegZ4aONVNpGcxZN2JPeRkPYoGPeQC1JdcQ24mV0Gu8fRCmKs3fr1QigMdtVAcMPWQd4q0PF6o4vogSOOBv6pqUAlg7xefoiSqG8rrTjZ0qLgAQaGAW4gehT/tRaV/B6CsFConJyh9fKGqy/WBkb2hdHBA+27d0Tm2D/7+aS+y12di2KPPW4yea4rCzQ2KUWNQftw0cVbH0O6orDHlp7ABwxE2YDgqSoqgcfeA0WDAoa+2obwoH3/9uAfBN/W1HLJeV5DXVlfh2D5T54dH+47Q9x3S6GuLooiqinI4OrugqK6HvGP3aOQd/BkA4B0UAidXN4QPHoMfNq3Fr/+3CUe/24Wyf07B2dMbXh2DkPvbT/h1+yZ0CIuQ4iBqLrXaETqd7ppmKXdy0qBr18vXEa48SWTzeKqkhVraQ25U18146NoO1UGme2mqj/8PQkU+VLnfQllmuq7DXnvIgYszrdtTD7lS4YDePnFwdpB/QrbWYr4Xd1PD1duKylGJAcmh6HbztU0a0tZUdUPrHVtjyLoAwNUaQ9Yv/sDv7tvjmq/RtlBXPAQYveGicUGXLtc+e3trME/sZp7QDQAcBAcIdT357CEna1IIpk+euSDXQtFglnUzOYasm0WmvYIgDx9AEGCsrYVzOy94BXS22EZQKuG5dAViRk0EAJzc/wPKYYSq1oDwkwWIPHYKAOBSNzdHZVkpgIsTt7bv1gNKlRr6fheL3J7jJsFB7YiCY4dNvdKieeSAiPxjh6Vrui+nYvgwAIC7nz+cGxl26+LpDYXSAQ5qR4QNug0AkLPlY9RWV1kU5CqNs/RbxNxT/eNn61BeXNjo6+7/YgPWPzUN2Z9koubCeShVKuj6XpwLqH2IaRIvXd/BcHLzQHlRAQqOHYZK44whM+eid2ISBEFA3sGfkVs3IoEFORG1FAvylrrMLOuXFTkYezzj0Cn6ZtT43wTRwRnKynxo34tFu//cCcFYC4NLexhdOrRh0G3LxdMbEAS4eLLHSk7mQlyugtxedI7xhk8XN3SKuPqiVjDfi9xVZXmSro3Uv4b8Srd4uxrmmdYD4YN7730QnTpd/nZn1hSsNZ1gig3wkNYJglBv4jgeb8i6HB0UKKrrRb7J2Qm1Hx1FzRd19/KuOwwolAKUKvl+WildXND/xTcQN85UbAff3L/RkVBKH190HX0HgmJvRrsOndArcQqG+gYjsLhMuoOEq9ZykikXL9NJsMgR43H3K+9AG1TvlqGe3ki4dwYgCDictRP7v9gAURTx03/+je2vzcPG5x9B9vpMVJQUAQBKTp3E4awvcfYfU/Gf/+fvAAC/YH2T7zFswAi4eGlRWVKEn7euR0neCQCmAlwQBNx8VwqibkvE+LSl8OmqR23VBexZuwKG2lqL/Zwr/Ae//vc/AIDD3+wEYBom7xcSJrWZXzdTQe6gdkT40DEATMeh/vc/Cg8/f7j5+KFLb9NszmeOHJTiICJqCbsYsv7hhx/inXfeQUFBAUJDQ/Hcc88hMjJSlliE+tdfNOOH+IyRvXG8T7jpx6Ug4Fz/F+H88yooiw4BSjUudL8LlTEzbPNGfs00YOq/UHm2BK5XmOma2p55YjNzTzk1zreLO3y7XGMPRt0QamtM6AZYDllv9YK8rodc8HS0ysmF5kqOC8TNnb3Qo73lj1u1whEXDBfYQ05W99jAYBhPlAN/VMCt0gCxskJ6TFH33VFr2vZ60+YQBAEJdyWh8823Qu18+eJQUCjQ//5HpWWxzyCUa5zhEGoqQrv0ugVl/5yGQqlEuw4BCIy+eAunxm7/FRjVG70Tk/D9p2vwy7YNOJ6zT7rO2lBTgz++/i+O7v0fOugjkHcgR+pFd9X64vzZUgCAX0jTt5VyUKvRa9xk7H5nCQ5+tU16L75dTcV815suzoYdP2kaPl88F2cOH0TWmuWIHD4eRSeOwaN9R9N16LW1UKrUMNRUAwC8g4Kh1jgjcsR4lJ7Jg1+3i5PP6ROG4PzZEmiDQuAfdvE36E0T7oODSo0je3dBNBr4W4iIWszmC/Jt27YhPT0daWlpiIqKwpo1a5CSkoLt27fD21uGH2QtHLKuUioQ4nPxB3RV6ARUhU6AUFVmuiWayv6LJxdPb1MvOcmqW7cwFBcXITw8Su5Qrn91PWDWuH4cuFiQOziompywr8XqrqUXtLZ131yVUoFI/4YnTiZ2uRtHyg4jxOPqJ8Ah22VLJ+AvNT6yA4wBF1Dzh6k3F64qoLwGAGD+9sh1/XhjnFzdW3RNquDkBLenn5OWVY5O6DV+coteM2zAcAiCAj9sWisV47Fj74E2sCt+/vxT5P/5uzR7uWdAEEpPnUR5Yb7p9Zw08A+NaNbrBEbfhPa67jhz+CBctb64ZbJptvdLefj5o/8Dj+F/K1/G8ZxsHM/JtnzPgoBhs5/H3n+vQknucbTXmSaUi7rtjgb7UqpU6Dl2UoP1ao0zbr77AYQPHY3qktNo1+X6mZuGiKzDdjLHZWRmZmLixIlITEwEAKSlpWHXrl3YsGEDpk2bZvV4LHqQrqE3qTn3xCRqCRcXFwwaNEzuMG4IQt0M7YKXdYpYT08veHtr4efn3+r7Vnb3glhWA2WUfZxUuyckSe4QqI3Y3An4RgjualMhDkA9MQTi2SrUfH4cJc4qoKTGpgpyuYT2H4r2ujDkbFkPv+BQdL/VNIntsG7dkfvbTzjxy/cIjuuH9t264/y5szh7OhcqjTPctH5wbOYIL0EQMGDqHOQd/AUB4TFQOV7+WNyxexT6Js/CN5lLISiU8A7sguLc46itugB9v6HQBgVj+Ox5KDrxlzRE/Wq4af2gDQ25psm5iOjGZNOZo7q6GgcOHMCDDz4orVMoFIiPj0dOTk6L9tUaI8gEwfI+5IKDYDcjzc1x2ku8l7Ln+Bm7fNoqfofePlB4O0IR4tFmbVM/dpVKhbvvvrdtXsdNBfWQhj1L17RPfm6a/Rp0ka2dgG+M4KCA+n5T0SaoFBA8HaGeHo6K7Hzg2Dnp9o83unYdOmHgtDkW6wRBQKeInugU0VNap3HzkGZGbym1xhldevZp1rZB0Teh/aL/Bwe1WhqiXno6F551E96pnDRor+t+5Z0QEbURm84cJSUlMBgMDc6Me3t749ixYy3al7d3602y4T4kCKJRhId/u1bbp7W0ZjvIwZ7jZ+zyaZP4A6xzezd7bnt7jh2w//jtSWudgG+tk+9X2pc0qaN5WSkgMMILRSfOIfgmX9lPttjzCbG2jN3Jtd4dG9RqaIO6tur+7bndAfuOn7HLx57jt6WT7zZdkLemoqJrH0IkCKYfaDXRnhBFoLDwXOsEZwXm2FujHeRgz/EzdvnYc/yMXT7WiN/8GmTSWifgW7NNW7QvLdBpZivejrAV2PPni7HLx57jZ+zysef4bSF2my7IPT09oVQqUVRUZLG+qKgIWm3Lbnkjimi1H1atuS9rs+fYAfuOn7HLx57jZ+zysff4b0StefKdJ5Ssj7HLx57jZ+zysef4benku00X5Gq1Gj169MDevXsxePBgAIDRaMTevXsxeXLLZv4kIiIi29RaJ+B58v0ie46fscvHnuNn7PKx5/htIXZF05vIKzk5GevXr8emTZvw559/Yv78+Th//jzGjx8vd2hERETUCuqfgDczn4CPiYmRMTIiIqK2ZdM95ABw2223obi4GEuXLkVBQQHCwsKwevXqFg9ZJyIiItuVnJyM1NRUhIeHIzIyEmvWrOEJeCIiuu7ZfEEOAJMnT+YQdSIiousYT8ATEdGNyC4KciIiIrr+8QQ8ERHdaGz+GnIiIiIiIiKi6xELciIiIiIiIiIZsCAnIiIiIiIikgELciIiIiIiIiIZsCAnIiIiIiIiksENM8u6ILTePlpjX9Zmz7ED9h0/Y5ePPcfP2OVjjfjttW1s3Y2e6wH7jp+xy8ee42fs8rHn+G0p1wuiKIptFwYRERERERERNYZD1omIiIiIiIhkwIKciIiIiIiISAYsyImIiIiIiIhkwIKciIiIiIiISAYsyImIiIiIiIhkwIKciIiIiIiISAYsyImIiIiIiIhkwIKciIiIiIiISAYsyImIiIiIiIhkwIKciIiIiIiISAYsyFvgww8/xKBBgxAREYEJEyZg//79cofUwMqVK5GYmIiYmBj06dMHDz/8MI4dO2axzZQpU6DX6y3+Pf/88zJFfNGyZcsaxDV8+HDp8aqqKqSlpSEuLg4xMTF45JFHUFhYKGPEFw0aNKhB7Hq9HmlpaQBsr82///57PPTQQ0hISIBer8fOnTstHhdFEW+88QYSEhIQGRmJ++67D3///bfFNqWlpZgzZw5iY2PRq1cvzJ07FxUVFbLGXlNTg5dffhmjR49GdHQ0EhIS8OSTT+Kff/6x2Edj/19vv/22rLEDwFNPPdUgrpSUFItt5Gr35sTf2HdAr9dj9erV0jZytX1zjo3NOcacOnUK06ZNQ1RUFPr06YPFixejtra2zeMn62Gub1v2nOsB+8r3zPXy5Jum4gdsO98z11s/1zu02Z6vM9u2bUN6ejrS0tIQFRWFNWvWICUlBdu3b4e3t7fc4Un27duHSZMmISIiAgaDAa+99hpSUlLw+eefw9nZWdpu4sSJmDVrlrSs0WjkCLeBbt26ITMzU1pWKpXS34sWLcLu3buxZMkSuLm54YUXXsDMmTPx0UcfyRGqhU8//RQGg0FaPnLkCJKTky1+ZNhSm1dWVkKv1yMxMREzZ85s8PiqVavwwQcfICMjAwEBAXjjjTeQkpKCbdu2wdHREQDw+OOPo6CgAJmZmaipqcHcuXPx/PPP49VXX5Ut9gsXLuDgwYOYPn06QkNDUVZWhhdffBHTp0/Hxo0bLbadNWsWJk6cKC27uLi0adxNxW7Wt29fpKenS8tqtdricbnaHWg6/qysLIvlr7/+Gs888wyGDRtmsV6Otm/OsbGpY4zBYMCDDz4IrVaLjz76CPn5+UhNTYVKpcK//vWvNn8P1PaY663DXnM9YF/5nrlennzTVPxmtprvmetlyPUiNcsdd9whpqWlScsGg0FMSEgQV65cKWNUTSsqKhJ1Op24b98+ad3kyZPFhQsXyhhV45YuXSqOGTOm0cfKysrEHj16iF988YW07ujRo6JOpxNzcnKsFGHzLVy4UBw8eLBoNBpFUbTdNhdFUdTpdOKOHTukZaPRKN5yyy3i6tWrpXVlZWVieHi4uHXrVlEUL7b9/v37pW12794t6vV68cyZM7LF3phffvlF1Ol0Yl5enrRu4MCBYmZmZhtHd2WNxZ6amipOnz79ss+xlXYXxea1/fTp08WkpCSLdbbQ9qLY8NjYnGPMrl27xNDQULGgoEDaZt26dWJsbKxYVVVl1fipbTDXt73rKdeLov3ke+Z6+dhzvmeuN2nrXM8h681QXV2NAwcOID4+XlqnUCgQHx+PnJwcGSNr2rlz5wAAHh4eFuu3bNmCuLg4jBo1Cq+++irOnz8vR3gNHD9+HAkJCbj11lsxZ84cnDp1CgDw22+/oaamxuL/IDg4GP7+/vj5559lirZx1dXV2Lx5MxITEyEIgrTeVtv8Urm5uSgoKLBoazc3N0RFRUmf95ycHLi7uyMiIkLaJj4+HgqFwuaGd5aXl0MQBLi7u1usX7VqFeLi4jB27FisXr3aZoYd79u3D3369MGwYcMwb948lJSUSI/ZU7sXFhZi9+7duOOOOxo8Zgttf+mxsTnHmJ9//hk6nQ5arVbaJiEhAeXl5Th69Kj1gqc2wVxvPddDrgfsO98z18vvesj3zPWtg0PWm6GkpAQGg6HBcDVvb+8G1yXYEqPRiEWLFiE2NhY6nU5aP2rUKPj7+8PX1xd//PEHXnnlFfz1119Yvny5jNECkZGRSE9PR5cuXVBQUIA333wTkyZNwpYtW1BYWAiVStXgQOvt7Y2CggKZIm7czp07ce7cOYwbN05aZ6tt3hhzezb2eTdfY1NYWAgvLy+Lxx0cHODh4WFT/x9VVVV45ZVXMHLkSLi6ukrrp0yZgu7du8PDwwM5OTl47bXXUFBQgKefflrGaE3D14YMGYKAgACcPHkSr732GqZOnYqPP/4YSqXSbtodADZt2gQXFxcMHTrUYr0ttH1jx8bmHGMKCwstEjQAadnW2p9ajrneOq6XXA/Yd75nrpfX9ZLvmetbBwvy61haWhqOHDmCdevWWay/8847pb/1ej18fHxw33334cSJEwgMDLR2mJL+/ftLf4eGhiIqKgoDBw7EF198AScnJ9niaqkNGzagX79+8PPzk9bZaptfz2pqavDoo49CFEVpsh2z5ORk6e/Q0FCoVCrMmzcPc+bMaXANlzWNHDlS+ts8CcrgwYOls+j2ZMOGDRg9erR0HaKZLbT95Y6NRPaIuV4+zPfys8dcD1w/+Z65vnVwyHozeHp6QqlUoqioyGJ9UVFRgzMotmLBggXYtWsX1qxZg/bt219x26ioKACmIWS2xN3dHZ07d8aJEyeg1WpRU1ODsrIyi22Kiorg4+MjU4QN5eXlYc+ePY0O3anPVtscgNSeV/q8a7VaFBcXWzxeW1uLs2fP2sT/R01NDWbPno1Tp07h3XfftThj3pioqCjU1tYiNzfXShE2T6dOneDp6Sl9Tmy93c1++OEH/PXXX5gwYUKT21q77S93bGzOMUar1TaYidW8bEvtT1eHuV4e9pjrAfvP98z1tsUe8z1zfethQd4MarUaPXr0wN69e6V1RqMRe/fuRUxMjIyRNSSKIhYsWIAdO3ZgzZo16NSpU5PPOXToEADb+0FZUVGBkydPwsfHB+Hh4VCpVBb/B8eOHcOpU6cQHR0tX5CX2LhxI7y9vTFgwIArbmerbQ4AAQEB8PHxsWjr8vJy/PLLL9LnPSYmBmVlZfjtt9+kbb777jsYjUZERkZaPeb6zAn6+PHjeO+99+Dp6dnkcw4dOgSFQmFTsygDwJkzZ1BaWip9Tmy53ev79NNP0aNHD4SGhja5rbXavqljY3OOMdHR0Th8+LDFD9g9e/bA1dUVISEhbRo/tT3mennYY64H7D/fM9fbFnvM98z1rYdD1pspOTkZqampCA8PR2RkJNasWYPz589j/PjxcodmIS0tDVu3bsVbb70FFxcX6VoHNzc3ODk54cSJE9iyZQv69++Pdu3a4Y8//kB6ejp69+7drC9UW1q8eDEGDhwIf39/5OfnY9myZVAoFBg1ahTc3NyQmJiIjIwMeHh4wNXVFQsXLkRMTIzNJGmj0YiNGzdi7NixcHC4+NWyxTavqKjAiRMnpOXc3FwcOnQIHh4e8Pf3R1JSElasWIGgoCDpVii+vr4YPHgwANMEGH379sVzzz2HtLQ01NTU4IUXXsDIkSMthu5ZO3YfHx/MmjULBw8exMqVK2EwGKTvgIeHB9RqNXJycvDLL7/g5ptvhouLC3JycpCeno4xY8Y0mBDJmrF7eHhg+fLlGDZsGLRaLU6ePImXX34ZQUFB6Nu3LwB5272p+P39/QGYftBt374dqampDZ4vZ9s3dWxszjEmISEBISEhePLJJ/HEE0+goKAAS5YswaRJk2Qf/kitg7m+7dl7rgfsJ98z18uTb5qK39bzPXO99XO9IIqi2CZ7vg6tXbsW77zzDgoKChAWFoZnn31WGopkK/R6faPr09PTMX78eJw+fRpPPPEEjhw5gsrKSnTo0AGDBw/Gww8/3ORQn7b22GOP4fvvv0dpaSm8vLzQs2dPPPbYY9I1V1VVVcjIyMDnn3+O6upqJCQkYN68eTZz1jkrK0u6X22XLl2k9bbY5tnZ2UhKSmqwfty4ccjIyIAoili6dCnWr1+PsrIy9OzZE/PmzbN4X6WlpXjhhRfw1VdfQaFQYOjQoXj22Wfb/D6TV4p95syZuPXWWxt93vvvv4+4uDgcOHAAaWlpOHbsGKqrqxEQEIDbb78dycnJbV5UXSn2+fPnY8aMGTh48CDOnTsHX19f3HLLLXj00UcthsvK1e5NxZ+RkQEA+Pjjj7Fo0SJkZWXBzc3NYjs5276pYyPQvGNMXl4e5s+fj3379kGj0WDcuHGYM2eOxY9ysm/M9W3L3nM9YD/5nrlennwD2He+Z663fq5nQU5EREREREQkA15DTkRERERERCQDFuREREREREREMmBBTkRERERERCQDFuREREREREREMmBBTkRERERERCQDFuREREREREREMmBBTkRERERERCQDFuREREREREREMmBBTmRDBg0ahPfee6/Z22dnZ0Ov16OsrKztgrJhy5Ytw+233y53GERERM3GXN8yzPV0vRNEURTlDoLI3uj1+is+PnPmTDzyyCMt3m9xcTE0Gg00Gk2ztq+ursbZs2eh1WohCEKLX6+5srOzkZSUhO+//x7u7u7YuHEjFi1ahB9++KHNXvNSer0eb775JgYPHiytq6ioQHV1NTw9Pa0WBxER3RiY65nriazBQe4AiOxRVlaW9Pe2bduwdOlSbN++XVrn7Ows/S2KIgwGAxwcmv66eXl5tSgOtVoNHx+fFj3HlhgMBgiCAIXi6gbruLi4wMXFpZWjIiIiYq5vLcz1RFfGIetEV8HHx0f65+bmBkEQpOVjx44hNjYWu3fvxvjx4xEREYEff/wRJ06cwPTp0xEfH4+YmBgkJiZiz549Fvu9dBibXq/HJ598ghkzZiAqKgpDhw7Fl19+KT1+6TC2jRs3olevXvjmm28wYsQIxMTEICUlBfn5+dJzamtrsXDhQvTq1QtxcXF4+eWXkZqaiocffrhZ7z07OxtPP/00zp07B71eD71ej2XLlgEwncVfvHgx+vbti+joaEyYMAHZ2dnSc83xffnll7jtttsQERGBU6dOYf/+/UhOTkZcXBx69uyJyZMn48CBAxbtAgAzZsyAXq+Xli8dxmY0GrF8+XL069cP4eHhuP322/H1119Lj+fm5kKv1+O///0vpkyZgqioKIwZMwY5OTnSNnl5eXjooYfQu3dvREdHY+TIkdi9e3ez2oaIiK4fzPXM9UTWwIKcqI28+uqrmDNnDrZt2wa9Xo/Kykr0798f7733HjZt2oS+ffvioYcewqlTp664n+XLl2PEiBHYvHkz+vXrh8cffxylpaWX3f7ChQt499138dJLL2Ht2rU4ffo0Fi9eLD2+atUqbNmyBenp6Vi3bh3Ky8uxc+fOZr+vmJgYzJ07F66ursjKykJWVhbuv/9+AMCCBQuQk5OD119/HZs3b8bw4cPxwAMP4O+//7aIb9WqVVi4cCG2bt0Kb29vVFRUYOzYsVi3bh3Wr1+PoKAgTJs2DeXl5QCATz/9FACQnp6OrKwsaflS77//PjIzM5GamorNmzcjISEBDz/8sMXrA8Drr7+OlJQUfPbZZ+jcuTPmzJmD2tpa6T1UV1dj7dq12LJlCx5//HGLXhAiIiIz5nrmeqJrJhLRNdmwYYPYs2dPafm7774TdTqduGPHjiafO3LkSPGDDz6QlgcOHChmZmZKyzqdTnz99del5YqKClGn04m7d++2eK2zZ89Kseh0OvH48ePSc9auXSvGx8dLy/Hx8eLq1aul5draWnHAgAHi9OnTLxtnY69T/z2Loijm5eWJYWFh4pkzZyzW33vvveKrr75qEd+hQ4eu2C4Gg0GMiYkRv/rqK4u2uLRNly5dKo4ZM0ZaTkhIEFesWGGxTWJiojh//nxRFEXx5MmTok6nE9evXy89fuTIEVGn04lHjx4VRVEUR40aJS5btuyK8RER0Y2Fud6EuZ6o9fEacqI2EhERYbFcUVGB5cuXY9euXSgoKIDBYMCFCxeaPGtef1IZZ2dnuLq6ori4+LLbazQaBAYGSsu+vr4oKioCAJw7dw6FhYWIjIyUHlcqlejRoweMRmOL3t+lDh8+DIPBgOHDh1usr66uRrt27aRllUrVYKKcwsJCLFmyBPv27UNRURGMRiPOnz/fZNvUV15ejvz8fMTGxlqsj42Nxe+//26xrv7rm6/LKy4uRnBwMJKSkjB//nxkZWUhPj4eQ4cORWhoaLPjICKiGwdzvQlzPdHVY0FO1EYunT118eLF2LNnD1JTUxEYGAgnJyfMmjULNTU1V9yPSqWyWBYE4YoJ9dIJZQRBgGiFmylUVlZCqVRiw4YNUCqVFo/VHwbm5OTUYJbY1NRUlJaW4plnnoG/vz/UajXuvPPOJtvmatVvU3Ms5jadMGECEhISsGvXLnz77bd4++23kZqaiilTprRJLEREZL+Y6y9irie6OryGnMhKcnJyMG7cOAwZMgR6vR5arRZ5eXlWjcHNzQ1arRa//vqrtM5gMODgwYMt2o9KpYLBYLBYFxYWBoPBgOLiYgQFBVn8a2p22J9++glTpkxB//790a1bN6jVapSUlDT5mvW5urrC19cXP/30U4N9h4SEtOj9dejQAXfffTeWL1+O5ORkrF+/vkXPJyKiGxNz/eUx1xM1jj3kRFYSFBSEHTt2YNCgQRAEAUuWLLnmoWNXY/LkyVi5ciUCAwPRtWtXrF27FmfPnm3RvU07duyIyspK7N27F3q9HhqNBl26dMHo0aPx5JNP4qmnnkJYWBhKSkqkbQYMGHDZ/XXu3BmbN29GREQEysvL8dJLL8HJyanBa+7duxexsbFQq9Xw8PBosJ+UlBQsW7YMgYGBCA0NxcaNG/H777/jlVdeafZ7e/HFF9GvXz907twZZWVlyM7ORnBwcLOfT0RENy7m+gGX3R9zPVHjWJATWclTTz2FuXPn4q677oKnpyemTp2KiooKq8cxdepUFBYWIjU1FUqlEhMnTkRCQkKDoWdXEhsbi7vuuguzZ89GaWkpZs6ciUceeQTp6elYsWIFMjIykJ+fj3bt2iE6OvqKCRowJcbnnnsO48aNQ4cOHfDYY4/hpZdestgmNTUVGRkZ+OSTT+Dn54evvvqqwX6SkpJQXl6OjIwM6Tqxt956C507d272ezMajViwYAHOnDkDV1dX9O3bF08//XSzn09ERDcu5vrLY64napwgWuOCEyKyWUajESNGjMCIESMwe/ZsucMhIiKiVsZcT2S72ENOdIPJy8vDt99+i969e6O6uhoffvgh8vLyMHr0aLlDIyIiolbAXE9kP1iQE91gFAoFNm7ciMWLF0MUReh0OmRmZvLaKSIiousEcz2R/eCQdSIiIiIiIiIZ8LZnRERERERERDJgQU5EREREREQkAxbkRERERERERDJgQU5EREREREQkAxbkRERERERERDJgQU5EREREREQkAxbkRERERERERDJgQU5EREREREQkg/8PolTQ0IsQRp0AAAAASUVORK5CYII=", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "fig, axs = plt.subplots(1, 2, figsize=(12, 4))\n", + "\n", + "for i, df in enumerate(result_dfs):\n", + " axs[0].plot(df[\"lossg\"], label=i)\n", + "axs[0].legend()\n", + "axs[0].set_title(\"Generator Loss During Training\")\n", + "axs[0].set_xlabel(\"Training Iterations\")\n", + "axs[0].set_ylabel(\"Generator Loss\")\n", + "\n", + "for i, df in enumerate(result_dfs):\n", + " axs[1].plot(df[\"lossd\"], label=i)\n", + "axs[1].legend()\n", + "axs[1].set_title(\"Discriminator Loss During Training\")\n", + "axs[1].set_xlabel(\"Training Iterations\")\n", + "axs[1].set_ylabel(\"Discriminator Loss\")\n", + "\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "73571f59", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from ray.tune.examples.pbt_dcgan_mnist.common import demo_gan\n", + "\n", + "with best_result.checkpoint.as_directory() as best_checkpoint:\n", + " demo_gan([best_checkpoint])" + ] + }, + { + "cell_type": "markdown", + "id": "aaff207d", + "metadata": {}, + "source": [ + "Training of the MNist Generator should take a couple of minutes. The example can be easily altered to generate images for other datasets, e.g. cifar10 or LSUN." + ] + }, + { + "cell_type": "markdown", + "id": "35b48117", + "metadata": {}, + "source": [ + "## Summary\n" + ] + }, + { + "cell_type": "markdown", + "id": "c4444bcd", + "metadata": {}, + "source": [ + "This tutorial covered:\n", + "1. **Two examples** of using Population Based Training to tune deep learning hyperparmeters (CNN and GAN training)\n", + "2. **Saving and loading checkpoints** and making sure that all hyperparameters are used (ex: optimizer state)\n", + "3. **Visualizing reported metrics** after training\n", + "\n", + "To learn more, check out the next tutorial {doc}`/tune/examples/pbt_visualization/pbt_visualization` for a visual guide to understanding PBT and its underlying behavior.\n", + "\n", + "If you have any questions, suggestions, or run into any problems please reach out on [Discuss](https://discuss.ray.io/), [GitHub](https://github.com/ray-project/ray) or the [Ray Slack](https://forms.gle/9TSdDYUgxYs8SA9e8)!\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "2ceb6a7a", + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3.8.9 64-bit", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.9" + }, + "vscode": { + "interpreter": { + "hash": "31f2aee4e71d21fbe5cf8b01ff0e069b9275f58929596ceb00d14d90e3e16cd6" + } + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/doc/source/tune/tutorials/overview.rst b/doc/source/tune/tutorials/overview.rst index 6d4350326200..efbfa458f982 100644 --- a/doc/source/tune/tutorials/overview.rst +++ b/doc/source/tune/tutorials/overview.rst @@ -87,9 +87,9 @@ Tune Feature Guides --- :img-top: /images/tune.png - .. link-button:: tune-advanced-tutorial + .. link-button:: ../examples/pbt_guide :type: ref - :text: Using Population-Based Training + :text: A Guide to Population-Based Training :classes: btn-link btn-block stretched-link --- diff --git a/doc/source/tune/tutorials/tune-advanced-tutorial.rst b/doc/source/tune/tutorials/tune-advanced-tutorial.rst deleted file mode 100644 index 9e3f89d1bbe8..000000000000 --- a/doc/source/tune/tutorials/tune-advanced-tutorial.rst +++ /dev/null @@ -1,238 +0,0 @@ -.. _pbt-guide-ref: - -A Guide to Population Based Training -==================================== - -Tune includes a distributed implementation of `Population Based Training (PBT) `__ as -a :ref:`scheduler `. - -.. image:: /images/tune_advanced_paper1.png - - -PBT starts by training many neural networks in parallel with random hyperparameters, using information from the rest of the population to refine these -hyperparameters and allocate resources to promising models. Let's walk through how to use this algorithm. - -.. contents:: - :local: - :backlinks: none - -Function API with Population Based Training -------------------------------------------- - -PBT takes its inspiration from genetic algorithms where each member of the population -can exploit information from the remainder of the population. For example, a worker might -copy the model parameters from a better performing worker. It can also explore new hyperparameters by -changing the current values randomly. - -As the training of the population of neural networks progresses, this process of exploiting and exploring -is performed periodically, ensuring that all the workers in the population have a good base level of performance -and also that new hyperparameters are consistently explored. - -This means that PBT can quickly exploit good hyperparameters, can dedicate more training time to -promising models and, crucially, can adapt the hyperparameter values throughout training, -leading to automatic learning of the best configurations. - -First we define a training function that trains a ConvNet model using SGD. - -.. literalinclude:: /../../python/ray/tune/examples/pbt_convnet_function_example.py - :language: python - :start-after: __train_begin__ - :end-before: __train_end__ - -The example reuses some of the functions in ray/tune/examples/mnist_pytorch.py, and is also a good -demo for how to decouple the tuning logic and original training code. - -Here, we also need to take in a ``checkpoint_dir`` arg since checkpointing is required for the exploitation process in PBT. -We have to both load in the checkpoint if one is provided, and periodically save our -model state in a checkpoint- in this case every 5 iterations. With SGD, there's no need to checkpoint the optimizer -since it does not depend on previous states, but this is necessary with other optimizers like Adam. - -Then, we define a PBT scheduler: - -.. literalinclude:: /../../python/ray/tune/examples/pbt_convnet_function_example.py - :language: python - :start-after: __pbt_begin__ - :end-before: __pbt_end__ - -Some of the most important parameters are: - -- ``hyperparam_mutations`` and ``custom_explore_fn`` are used to mutate the hyperparameters. - ``hyperparam_mutations`` is a dictionary where each key/value pair specifies the candidates - or function for a hyperparameter. custom_explore_fn is applied after built-in perturbations - from hyperparam_mutations are applied, and should return config updated as needed. - -- ``resample_probability``: The probability of resampling from the original distribution - when applying hyperparam_mutations. If not resampled, the value will be perturbed by a - factor of 1.2 or 0.8 if continuous, or changed to an adjacent value if discrete. Note that - ``resample_probability`` by default is 0.25, thus hyperparameter with a distribution - may go out of the specific range. - -Now we can kick off the tuning process by invoking ``Tuner.fit()``: - -.. literalinclude:: /../../python/ray/tune/examples/pbt_convnet_function_example.py - :language: python - :start-after: __tune_begin__ - :end-before: __tune_end__ - -During the training, we can constantly check the status of the models from console log: - -.. code-block:: bash - - == Status == - Memory usage on this node: 11.2/16.0 GiB - PopulationBasedTraining: 12 checkpoints, 5 perturbs - Resources requested: 0/16 CPUs, 0/0 GPUs, 0.0/4.83 GiB heap, 0.0/1.66 GiB objects - Result logdir: /Users/foo/ray_results/pbt_test - Number of trials: 4 (4 TERMINATED) - +---------------------------+------------+-------+-----------+------------+----------+--------+------------------+ - | Trial name | status | loc | lr | momentum | acc | iter | total time (s) | - |---------------------------+------------+-------+-----------+------------+----------+--------+------------------| - | train_convnet_b2732_00000 | TERMINATED | | 0.221776 | 0.608416 | 0.95625 | 59 | 13.0862 | - | train_convnet_b2732_00001 | TERMINATED | | 0.0734679 | 0.1484 | 0.934375 | 59 | 13.1084 | - | train_convnet_b2732_00002 | TERMINATED | | 0.0376862 | 0.8 | 0.971875 | 46 | 10.2909 | - | train_convnet_b2732_00003 | TERMINATED | | 0.0471078 | 0.8 | 0.95 | 51 | 11.3355 | - +---------------------------+------------+-------+-----------+------------+----------+--------+------------------+ - -In {LOG_DIR}/{MY_EXPERIMENT_NAME}/, all mutations are logged in pbt_global.txt -and individual policy perturbations are recorded in pbt_policy_{i}.txt. Tune logs: -[target trial tag, clone trial tag, target trial iteration, clone trial iteration, -old config, new config] on each perturbation step. - -Checking the accuracy: - -.. code-block:: python - - # Plot by wall-clock time - dfs = analysis.fetch_trial_dataframes() - # This plots everything on the same plot - ax = None - for d in dfs.values(): - ax = d.plot("training_iteration", "mean_accuracy", ax=ax, legend=False) - - plt.xlabel("iterations") - plt.ylabel("Test Accuracy") - - print('best config:', analysis.get_best_config("mean_accuracy")) - -.. image:: /images/tune_advanced_plot1.png - -.. _tune-advanced-tutorial-pbt-replay: - -Replaying a PBT run -------------------- -A run of Population Based Training ends with fully trained models. However, sometimes -you might like to train the model from scratch, but use the same hyperparameter -schedule as obtained from PBT. Ray Tune offers a replay utility for this. - -All you need to do is pass the policy log file for the trial you want to replay. -This is usually stored in the experiment directory, for instance -``~/ray_results/pbt_test/pbt_policy_ba982_00000.txt``. - -The replay utility reads the original configuration for the trial and updates it -each time when it was originally perturbed. You can (and should) -thus just use the same ``Trainable`` for the replay run. - -.. code-block:: python - - from ray import tune - - from ray.tune.examples.pbt_convnet_example import PytorchTrainable - from ray.tune.schedulers import PopulationBasedTrainingReplay - - replay = PopulationBasedTrainingReplay( - "~/ray_results/pbt_test/pbt_policy_ba982_00003.txt") - - tuner = tune.Tuner( - PytorchTrainable, - tune_config=tune.TuneConfig(scheduler=replay), - run_config=air.RunConfig(stop={"training_iteration": 100}), - ) - -DCGAN with PBT --------------- - -The Generative Adversarial Networks (GAN) (Goodfellow et al., 2014) framework learns generative -models via a training paradigm consisting of two competing modules – a generator and a -discriminator. GAN training can be remarkably brittle and unstable in the face of suboptimal -hyperparameter selection with generators often collapsing to a single mode or diverging entirely. - -As presented in `Population Based Training (PBT) `__, -PBT can help with the DCGAN training. We will now walk through how to do this in Tune. -Complete code example at `github `__ - -We define the Generator and Discriminator with standard Pytorch API: - -.. literalinclude:: /../../python/ray/tune/examples/pbt_dcgan_mnist/common.py - :language: python - :start-after: __GANmodel_begin__ - :end-before: __GANmodel_end__ - -To train the model with PBT, we need to define a metric for the scheduler to evaluate -the model candidates. For a GAN network, inception score is arguably the most -commonly used metric. We trained a mnist classification model (LeNet) and use -it to inference the generated images and evaluate the image quality. - -.. literalinclude:: /../../python/ray/tune/examples/pbt_dcgan_mnist/common.py - :language: python - :start-after: __INCEPTION_SCORE_begin__ - :end-before: __INCEPTION_SCORE_end__ - -We define a training function that includes a Generator and a Discriminator, each with an -independent learning rate and optimizer. We make sure to implement checkpointing for our training. - -.. literalinclude:: /../../python/ray/tune/examples/pbt_dcgan_mnist/pbt_dcgan_mnist_func.py - :language: python - :start-after: __Train_begin__ - :end-before: __Train_end__ - -We specify inception score as the metric and start the tuning: - -.. literalinclude:: /../../python/ray/tune/examples/pbt_dcgan_mnist/pbt_dcgan_mnist_func.py - :language: python - :start-after: __tune_begin__ - :end-before: __tune_end__ - -The trained Generator models can be loaded from log directory, and generate images -from noise signals. - -Visualization -~~~~~~~~~~~~~ - -Below, we visualize the increasing inception score from the training logs. - -.. code-block:: python - - lossG = [df['is_score'].tolist() for df in list(analysis.trial_dataframes.values())] - - plt.figure(figsize=(10,5)) - plt.title("Inception Score During Training") - for i, lossg in enumerate(lossG): - plt.plot(lossg,label=i) - - plt.xlabel("iterations") - plt.ylabel("is_score") - plt.legend() - plt.show() - -.. image:: /images/tune_advanced_dcgan_inscore.png - -And the Generator loss: - -.. code-block:: python - - lossG = [df['lossg'].tolist() for df in list(analysis.trial_dataframes.values())] - - plt.figure(figsize=(10,5)) - plt.title("Generator Loss During Training") - for i, lossg in enumerate(lossG): - plt.plot(lossg,label=i) - - plt.xlabel("iterations") - plt.ylabel("LossG") - plt.legend() - plt.show() - -.. image:: /images/tune_advanced_dcgan_Gloss.png - -Training of the MNist Generator takes a couple of minutes. The example can be easily -altered to generate images for other datasets, e.g. cifar10 or LSUN. diff --git a/python/ray/train/examples/pytorch/tune_cifar_torch_pbt_example.py b/python/ray/train/examples/pytorch/tune_cifar_torch_pbt_example.py index 6d57ad30d59d..71472227a249 100644 --- a/python/ray/train/examples/pytorch/tune_cifar_torch_pbt_example.py +++ b/python/ray/train/examples/pytorch/tune_cifar_torch_pbt_example.py @@ -1,18 +1,19 @@ import argparse +import os -import numpy as np -from ray.air import session import torch import torch.nn as nn import torchvision.transforms as transforms -from torchvision.models import resnet18 from filelock import FileLock from torch.utils.data import DataLoader, Subset from torchvision.datasets import CIFAR10 +from torchvision.models import resnet18 import ray import ray.train as train from ray import tune +from ray.air import session +from ray.air.checkpoint import Checkpoint from ray.air.config import FailureConfig, RunConfig, ScalingConfig from ray.train.torch import TorchTrainer from ray.tune.schedulers import PopulationBasedTraining @@ -58,17 +59,44 @@ def validate_epoch(dataloader, model, loss_fn): return {"loss": test_loss} +def update_optimizer_config(optimizer, config): + for param_group in optimizer.param_groups: + for param, val in config.items(): + param_group[param] = val + + def train_func(config): - epochs = config.pop("epochs", 3) + epochs = config.get("epochs", 3) + model = resnet18() model = train.torch.prepare_model(model) # Create optimizer. - optimizer = torch.optim.SGD( - model.parameters(), - lr=config.get("lr", 0.1), - momentum=config.get("momentum", 0.9), - ) + optimizer_config = { + "lr": config.get("lr"), + "momentum": config.get("momentum"), + } + optimizer = torch.optim.SGD(model.parameters(), **optimizer_config) + + starting_epoch = 0 + if session.get_checkpoint(): + checkpoint_dict = session.get_checkpoint().to_dict() + + # Load in model + model_state = checkpoint_dict["model_state_dict"] + model.load_state_dict(model_state) + + # Load in optimizer + optimizer_state = checkpoint_dict["optimizer_state_dict"] + optimizer.load_state_dict(optimizer_state) + + # Optimizer configs (`lr`, `momentum`) are being mutated by PBT and passed in + # through config, so we need to update the optimizer loaded from the checkpoint + update_optimizer_config(optimizer, optimizer_config) + + # The current epoch increments the loaded epoch by 1 + checkpoint_epoch = checkpoint_dict["epoch"] + starting_epoch = checkpoint_epoch + 1 # Load in training and validation data. transform_train = transforms.Compose( @@ -87,12 +115,14 @@ def train_func(config): ] ) - with FileLock(".ray.lock"): + data_dir = config.get("data_dir", os.path.expanduser("~/data")) + os.makedirs(data_dir, exist_ok=True) + with FileLock(os.path.join(data_dir, ".ray.lock")): train_dataset = CIFAR10( - root="~/data", train=True, download=True, transform=transform_train + root=data_dir, train=True, download=True, transform=transform_train ) validation_dataset = CIFAR10( - root="~/data", train=False, download=False, transform=transform_test + root=data_dir, train=False, download=False, transform=transform_test ) if config.get("test_mode"): @@ -110,22 +140,24 @@ def train_func(config): # Create loss. criterion = nn.CrossEntropyLoss() - results = [] - for _ in range(epochs): + for epoch in range(starting_epoch, epochs): train_epoch(train_loader, model, criterion, optimizer) result = validate_epoch(validation_loader, model, criterion) - session.report(result) - results.append(result) + checkpoint = Checkpoint.from_dict( + { + "epoch": epoch, + "model_state_dict": model.state_dict(), + "optimizer_state_dict": optimizer.state_dict(), + } + ) - # return required for backwards compatibility with the old API - # TODO(team-ml) clean up and remove return - return results + session.report(result, checkpoint=checkpoint) if __name__ == "__main__": parser = argparse.ArgumentParser() parser.add_argument( - "--address", required=False, type=str, help="the address to use for Redis" + "--address", required=False, type=str, help="The address to use for Redis." ) parser.add_argument( "--num-workers", @@ -144,7 +176,17 @@ def train_func(config): help="Finish quickly for testing.", ) parser.add_argument( - "--use-gpu", action="store_true", default=False, help="Enables GPU training" + "--use-gpu", action="store_true", default=False, help="Enables GPU training." + ) + parser.add_argument( + "--data-dir", + required=False, + type=str, + default="~/data", + help="Root directory for storing downloaded dataset.", + ) + parser.add_argument( + "--synch", action="store_true", default=False, help="Use synchronous PBT." ) args, _ = parser.parse_known_args() @@ -165,29 +207,31 @@ def train_func(config): hyperparam_mutations={ "train_loop_config": { # distribution for resampling - "lr": lambda: np.random.uniform(0.001, 1), + "lr": tune.loguniform(0.001, 0.1), # allow perturbations within this set of categorical values "momentum": [0.8, 0.9, 0.99], } }, + synch=args.synch, ) tuner = Tuner( trainer, param_space={ "train_loop_config": { - "lr": tune.choice([0.001, 0.01, 0.1]), + "lr": tune.grid_search([0.001, 0.01, 0.05, 0.1]), "momentum": 0.8, "batch_size": 128 * args.num_workers, - "epochs": args.num_epochs, "test_mode": args.smoke_test, # whether to to subset the data + "data_dir": args.data_dir, + "epochs": args.num_epochs, } }, tune_config=TuneConfig( - num_samples=4, metric="loss", mode="min", scheduler=pbt_scheduler + num_samples=1, metric="loss", mode="min", scheduler=pbt_scheduler ), run_config=RunConfig( - stop={"training_iteration": 2 if args.smoke_test else 100}, + stop={"training_iteration": 3 if args.smoke_test else args.num_epochs}, failure_config=FailureConfig(max_failures=3), # used for fault tolerance ), ) diff --git a/python/ray/tune/BUILD b/python/ray/tune/BUILD index 178751096778..e87f47a66197 100644 --- a/python/ray/tune/BUILD +++ b/python/ray/tune/BUILD @@ -796,7 +796,7 @@ py_test( py_test( name = "pbt_function", - size = "medium", + size = "small", srcs = ["examples/pbt_function.py"], deps = [":tune_lib"], tags = ["team:ml", "exclusive", "example"], diff --git a/python/ray/tune/examples/mnist_pytorch.py b/python/ray/tune/examples/mnist_pytorch.py index d9428e0a200b..ce83440efb34 100644 --- a/python/ray/tune/examples/mnist_pytorch.py +++ b/python/ray/tune/examples/mnist_pytorch.py @@ -65,7 +65,7 @@ def test(model, data_loader, device=None): return correct / total -def get_data_loaders(): +def get_data_loaders(batch_size=64): mnist_transforms = transforms.Compose( [transforms.ToTensor(), transforms.Normalize((0.1307,), (0.3081,))] ) @@ -78,14 +78,14 @@ def get_data_loaders(): datasets.MNIST( "~/data", train=True, download=True, transform=mnist_transforms ), - batch_size=64, + batch_size=batch_size, shuffle=True, ) test_loader = torch.utils.data.DataLoader( datasets.MNIST( "~/data", train=False, download=True, transform=mnist_transforms ), - batch_size=64, + batch_size=batch_size, shuffle=True, ) return train_loader, test_loader diff --git a/python/ray/tune/examples/pbt_dcgan_mnist/__init__.py b/python/ray/tune/examples/pbt_dcgan_mnist/__init__.py new file mode 100644 index 000000000000..e69de29bb2d1 diff --git a/python/ray/tune/examples/pbt_dcgan_mnist/common.py b/python/ray/tune/examples/pbt_dcgan_mnist/common.py index 32aef4c12622..470811d15757 100644 --- a/python/ray/tune/examples/pbt_dcgan_mnist/common.py +++ b/python/ray/tune/examples/pbt_dcgan_mnist/common.py @@ -1,4 +1,5 @@ import ray +from ray.air import Checkpoint import os import torch @@ -268,9 +269,9 @@ def demo_gan(checkpoint_paths): img_list = [] fixed_noise = torch.randn(64, nz, 1, 1) for path in checkpoint_paths: - netG_path = os.path.join(path, "checkpoint.pt") + checkpoint_dict = Checkpoint.from_directory(path).to_dict() loadedG = Generator() - loadedG.load_state_dict(torch.load(netG_path)["netGmodel"]) + loadedG.load_state_dict(checkpoint_dict["netGmodel"]) with torch.no_grad(): fake = loadedG(fixed_noise).detach().cpu() img_list.append(vutils.make_grid(fake, padding=2, normalize=True)) diff --git a/python/ray/tune/examples/pbt_dcgan_mnist/pbt_dcgan_mnist_func.py b/python/ray/tune/examples/pbt_dcgan_mnist/pbt_dcgan_mnist_func.py index ba108095a438..5bab1e28860a 100644 --- a/python/ray/tune/examples/pbt_dcgan_mnist/pbt_dcgan_mnist_func.py +++ b/python/ray/tune/examples/pbt_dcgan_mnist/pbt_dcgan_mnist_func.py @@ -18,14 +18,22 @@ import torch.utils.data import numpy as np -from common import beta1, MODEL_PATH -from common import demo_gan, get_data_loader, plot_images, train, weights_init -from common import Discriminator, Generator, Net +from ray.tune.examples.pbt_dcgan_mnist.common import ( + beta1, + MODEL_PATH, + demo_gan, + get_data_loader, + plot_images, + train, + weights_init, + Discriminator, + Generator, + Net, +) # __Train_begin__ def dcgan_train(config): - step = 0 use_cuda = config.get("use_gpu") and torch.cuda.is_available() device = torch.device("cuda" if use_cuda else "cpu") netD = Discriminator().to(device) @@ -39,20 +47,25 @@ def dcgan_train(config): optimizerG = optim.Adam( netG.parameters(), lr=config.get("lr", 0.01), betas=(beta1, 0.999) ) - with FileLock(os.path.expanduser("~/.data.lock")): + with FileLock(os.path.expanduser("~/ray_results/.data.lock")): dataloader = get_data_loader() + step = 1 if session.get_checkpoint(): - loaded_checkpoint = session.get_checkpoint() - with loaded_checkpoint.as_directory() as loaded_checkpoint_dir: - path = os.path.join(loaded_checkpoint_dir, "checkpoint.pt") - checkpoint = torch.load(path) - netD.load_state_dict(checkpoint["netDmodel"]) - netG.load_state_dict(checkpoint["netGmodel"]) - optimizerD.load_state_dict(checkpoint["optimD"]) - optimizerG.load_state_dict(checkpoint["optimG"]) - step = checkpoint["step"] - + checkpoint_dict = session.get_checkpoint().to_dict() + netD.load_state_dict(checkpoint_dict["netDmodel"]) + netG.load_state_dict(checkpoint_dict["netGmodel"]) + optimizerD.load_state_dict(checkpoint_dict["optimD"]) + optimizerG.load_state_dict(checkpoint_dict["optimG"]) + # Note: Make sure to increment the loaded step by 1 to get the + # current step. + last_step = checkpoint_dict["step"] + step = last_step + 1 + + # NOTE: It's important to set the optimizer learning rates + # again, since we want to explore the parameters passed in by PBT. + # Without this, we would continue using the exact same + # configuration as the trial whose checkpoint we are exploiting. if "netD_lr" in config: for param_group in optimizerD.param_groups: param_group["lr"] = config["netD_lr"] @@ -72,38 +85,28 @@ def dcgan_train(config): device, config["mnist_model_ref"], ) - step += 1 - os.makedirs("my_model", exist_ok=True) - torch.save( - { - "netDmodel": netD.state_dict(), - "netGmodel": netG.state_dict(), - "optimD": optimizerD.state_dict(), - "optimG": optimizerG.state_dict(), - "step": step, - }, - "my_model/checkpoint.pt", - ) - + checkpoint = None + if step % config["checkpoint_interval"] == 0: + checkpoint = Checkpoint.from_dict( + { + "netDmodel": netD.state_dict(), + "netGmodel": netG.state_dict(), + "optimD": optimizerD.state_dict(), + "optimG": optimizerG.state_dict(), + "step": step, + } + ) session.report( {"lossg": lossG, "lossd": lossD, "is_score": is_score}, - checkpoint=Checkpoint.from_directory("my_model"), + checkpoint=checkpoint, ) + step += 1 # __Train_end__ -if __name__ == "__main__": - parser = argparse.ArgumentParser() - parser.add_argument( - "--smoke-test", action="store_true", help="Finish quickly for testing" - ) - parser.add_argument( - "--data-dir", type=str, default="~/data/", help="Set the path of the dataset." - ) - args, _ = parser.parse_known_args() - ray.init() +def download_mnist_cnn(): import urllib.request # Download a pre-trained MNIST model for inception score calculation. @@ -116,6 +119,21 @@ def dcgan_train(config): "examples/pbt_dcgan_mnist/mnist_cnn.pt", MODEL_PATH, ) + return MODEL_PATH + + +if __name__ == "__main__": + parser = argparse.ArgumentParser() + parser.add_argument( + "--smoke-test", action="store_true", help="Finish quickly for testing" + ) + parser.add_argument( + "--data-dir", type=str, default="~/data/", help="Set the path of the dataset." + ) + args, _ = parser.parse_known_args() + ray.init() + + download_mnist_cnn() dataloader = get_data_loader(args.data_dir) if not args.smoke_test: diff --git a/python/ray/tune/examples/pbt_dcgan_mnist/pbt_dcgan_mnist_trainable.py b/python/ray/tune/examples/pbt_dcgan_mnist/pbt_dcgan_mnist_trainable.py index dedbe8d8f7dd..87ae77ab18e6 100644 --- a/python/ray/tune/examples/pbt_dcgan_mnist/pbt_dcgan_mnist_trainable.py +++ b/python/ray/tune/examples/pbt_dcgan_mnist/pbt_dcgan_mnist_trainable.py @@ -72,7 +72,7 @@ def save_checkpoint(self, checkpoint_dir): return checkpoint_dir def load_checkpoint(self, checkpoint_dir): - path = os.path.join(checkpoint_dir, "checkpoint") + path = os.path.join(checkpoint_dir, "checkpoint.pt") checkpoint = torch.load(path) self.netD.load_state_dict(checkpoint["netDmodel"]) self.netG.load_state_dict(checkpoint["netGmodel"]) @@ -138,7 +138,7 @@ def reset_config(self, new_config): }, ) - tune_iter = 5 if args.smoke_test else 300 + tune_iter = 10 if args.smoke_test else 300 tuner = tune.Tuner( PytorchTrainable, run_config=air.RunConfig( diff --git a/python/ray/tune/examples/pbt_example.py b/python/ray/tune/examples/pbt_example.py index b93681ceeff1..44badf3d02db 100755 --- a/python/ray/tune/examples/pbt_example.py +++ b/python/ray/tune/examples/pbt_example.py @@ -107,9 +107,10 @@ def reset_config(self, new_config): else: ray.init() + perturbation_interval = 5 pbt = PopulationBasedTraining( time_attr="training_iteration", - perturbation_interval=20, + perturbation_interval=perturbation_interval, hyperparam_mutations={ # distribution for resampling "lr": lambda: random.uniform(0.0001, 0.02), @@ -121,13 +122,24 @@ def reset_config(self, new_config): tuner = tune.Tuner( PBTBenchmarkExample, run_config=air.RunConfig( - name="pbt_test", + name="pbt_class_api_example", + # Stop when done = True or at some # of train steps (whichever comes first) stop={ - "training_iteration": 200, + "done": True, + "training_iteration": 10 if args.smoke_test else 1000, }, verbose=0, + # We recommend matching `perturbation_interval` and `checkpoint_interval` + # (e.g. checkpoint every 4 steps, and perturb on those same steps) + # or making `perturbation_interval` a multiple of `checkpoint_interval` + # (e.g. checkpoint every 2 steps, and perturb every 4 steps). + # This is to ensure that the lastest checkpoints are being used by PBT + # when trials decide to exploit. If checkpointing and perturbing are not + # aligned, then PBT may use a stale checkpoint to resume from. checkpoint_config=air.CheckpointConfig( - checkpoint_frequency=20, + checkpoint_frequency=perturbation_interval, + checkpoint_score_attribute="mean_accuracy", + num_to_keep=2, ), ), tune_config=tune.TuneConfig( diff --git a/python/ray/tune/examples/pbt_function.py b/python/ray/tune/examples/pbt_function.py index d2fd59e5508b..4307b14aba16 100644 --- a/python/ray/tune/examples/pbt_function.py +++ b/python/ray/tune/examples/pbt_function.py @@ -94,7 +94,7 @@ def pbt_function(config): step += 1 -def run_tune_pbt(): +def run_tune_pbt(smoke_test=False): perturbation_interval = 5 pbt = PopulationBasedTraining( time_attr="training_iteration", @@ -113,9 +113,10 @@ def run_tune_pbt(): name="pbt_function_api_example", verbose=False, stop={ - # Stop when done = True or at 1000 train steps (whichever comes first) + # Stop when done = True or at some # of train steps + # (whichever comes first) "done": True, - "training_iteration": 1000, + "training_iteration": 10 if smoke_test else 1000, }, failure_config=air.FailureConfig( fail_fast=True, @@ -158,7 +159,10 @@ def run_tune_pbt(): if __name__ == "__main__": parser = argparse.ArgumentParser() parser.add_argument( - "--smoke-test", action="store_true", help="Finish quickly for testing" + "--smoke-test", + action="store_true", + default=False, + help="Finish quickly for testing", ) parser.add_argument( "--server-address", @@ -176,4 +180,4 @@ def run_tune_pbt(): else: ray.init() - run_tune_pbt() + run_tune_pbt(smoke_test=args.smoke_test) diff --git a/python/ray/tune/examples/pbt_memnn_example.py b/python/ray/tune/examples/pbt_memnn_example.py index 3ffe1298c55c..6c140649f8fc 100644 --- a/python/ray/tune/examples/pbt_memnn_example.py +++ b/python/ray/tune/examples/pbt_memnn_example.py @@ -280,8 +280,9 @@ def load_checkpoint(self, path): elif args.server_address: ray.init(f"ray://{args.server_address}") + perturbation_interval = 2 pbt = PopulationBasedTraining( - perturbation_interval=2, + perturbation_interval=perturbation_interval, hyperparam_mutations={ "dropout": lambda: np.random.uniform(0, 1), "lr": lambda: 10 ** np.random.randint(-10, 0), @@ -294,6 +295,11 @@ def load_checkpoint(self, path): run_config=air.RunConfig( name="pbt_babi_memnn", stop={"training_iteration": 4 if args.smoke_test else 100}, + checkpoint_config=air.CheckpointConfig( + checkpoint_frequency=perturbation_interval, + checkpoint_score_attribute="mean_accuracy", + num_to_keep=2, + ), ), tune_config=tune.TuneConfig( scheduler=pbt, diff --git a/python/ray/tune/examples/pbt_tune_cifar10_with_keras.py b/python/ray/tune/examples/pbt_tune_cifar10_with_keras.py index 7ca672e51198..93338324c563 100755 --- a/python/ray/tune/examples/pbt_tune_cifar10_with_keras.py +++ b/python/ray/tune/examples/pbt_tune_cifar10_with_keras.py @@ -198,9 +198,10 @@ def cleanup(self): space["lr"] = 10**-4 space["dropout"] = 0.5 + perturbation_interval = 10 pbt = PopulationBasedTraining( time_attr="training_iteration", - perturbation_interval=10, + perturbation_interval=perturbation_interval, hyperparam_mutations={ "dropout": lambda _: np.random.uniform(0, 1), }, @@ -217,6 +218,11 @@ def cleanup(self): "mean_accuracy": 0.80, "training_iteration": 30, }, + checkpoint_config=air.CheckpointConfig( + checkpoint_frequency=perturbation_interval, + checkpoint_score_attribute="mean_accuracy", + num_to_keep=2, + ), ), tune_config=tune.TuneConfig( scheduler=pbt,