From 2b13b010e32ae431c8fe9ea9441818f8d990395a Mon Sep 17 00:00:00 2001
From: Eldad Cohen <eldad.cohen@aiola.com>
Date: Thu, 4 Apr 2024 20:03:42 +0000
Subject: [PATCH 1/9] added an example for how to deploy mme with triton for
 custom model based on faster whisper

---
 .../audio/sample.wav                          | Bin 0 -> 87404 bytes
 .../mme-triton-custom-faster-whisper.ipynb    | 589 ++++++++++++++++++
 .../readme.md                                 |  45 ++
 3 files changed, 634 insertions(+)
 create mode 100644 inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/audio/sample.wav
 create mode 100644 inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
 create mode 100644 inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/readme.md

diff --git a/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/audio/sample.wav b/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/audio/sample.wav
new file mode 100644
index 0000000000000000000000000000000000000000..7c135edd19912bba17e011d01682365d2d05c671
GIT binary patch
literal 87404
zcmXt91zb~K^ncslI@##%l9Ep8P_RHnZ0zn%?C$PD3`7h>0hMl$?vBBL!5Hg&|LgDn
z|NB0`-rc$9e9!m9y>suz&WwmyyATD%2QLa=ziqE61A-tFxRPdo7snt74Kbj#iF*@+
zfQ%+|0?L5yK+~W!WG14E%s~DiG0<};8Cna8kqhA7138BLoAwJzgXTbG$Py#~*$kvC
zp)*iA_|b)4B6E;X<O+g?PC`XMY8^y{-XUv|9mp@l6M702LvJ8IC>z;`Y)8@&1X&7w
zgmR(dK<Xv35LtwLM(m-}z~%wS3@QLJvyeSV8!{7m2z`R~K`P`jl878cQV}VV0NsPm
zL2ghHKoAY|*CK(?Dd;M+8M233kW0u;pvwZF_zZmj_?2MJ733k1vIh8HK?fmIs2Dkm
z>_+Y*qhRJvXfHGmA^@zb0k$;YWhk^2+5+^qBR7%#!24W83yhqDmOvD!5V?-*L#`ps
zhz+y?+5yD_?*1UBz{p>O2}MHNp+nGohz$Kj-XU+0cEk}{5ABCGKt7NHDM!8_-;oxC
z1}wxwfe;p|0NB1Eb%+@7I2&3F1p{qe0K;3P9>GJFARhLR2Gk4YJwje1Re)P}AQuK{
zLqp)M5XndS5Pfhz8|Y_1Y@`HST*Mff0b(={3WD^(+@HW^9im3eq3OVLC9t21WFo%-
zw=^(wKF|n3mB8jdWGciI;F<|}0uNe&Tq!^{9hwV812*W;DDW{KDMR>(3ls;&tf48S
z82JH4#fT>42+#*YdQd;Gm=4$`158mM;s(HG513VmbRYzPY$+50uy!ChATOH0Ol`;o
zNZJAvRRF~gq!(cUJ<*UC@REnrfE+7Fc!&w`aVd~fg1hfX9>NCbrvv=KfTMr-KOh;%
zAYui0T>{YK0g7^f|1Tm#oS{X~O2{9Qfmt7c{!YN65wK?mF`*tL1;jHI83ECZ0y3Jw
zVk?+gfK(tH#1MG75?J8@O<8~)7(^ow#5e+s{j-@4ut^XPz_k~|f`mvTu#kr|APT_G
zBEYOE(4PTfSA&QES|i8-;5UajKvx>zO#x=cK)!$<6tK_@^mik2#143{0Lbz{d|E($
z;y`5?fmxOi4fvD~EH;7KvHTA`415{@nb?L5gE4b}!U~Ku0qhNcF$|cY4d$AGtmOi)
zT96^&fdz;f9Z2z!Nn`>U2fxvf35c!(kXC`(Qh_YCgKQuzP?rqwqX0aNfN=uQA_uW)
z1sOL1tl>ca_}4)UAR;sZkn<58$QvNg1rZVgoeaPT8N@*cMAsRDp+<nc7N8;kc?s}q
z80cyQ`Beq-)c`Q>3ZzFtRHs0uVnAiE0N2yN{06{PI>1f@5i<vKWgug?pgMkmnF!(!
zY9<Jbb|TfF)<huZOaO+tAa3qJvI1O_Afp2S@790|5+ndNi~h@3Ymk{{(8PZg(*f*`
zAOcWvCV(F-)CO|39`pfcP^Z3NgavqjKy@_(DQ%E(;b7*}f4oluMrnX66QJJ&a8-c3
z0<|>+aA<*!A_nqo5S<EuZVd3_2e2B0$hIK;fKfD%&;`7E13$5#>T7{tT96aSt$z`N
zASsA#8DOmk<f|d5D+J8y1kn%yzlnfXTfmtekmLi-XyBCth&2u9qyRpoz}^%>1u_l*
z{Wz%qW}saJ_~(L1seo@>q!Q?E1{tRTVub-SWXQkOK@4Q%paQJ{+jwXacpw5kk^y%l
zpqT}_ksRRY1*3F;N(&(W-&I8oW~l)VOV9@_00JSfuL3@*f#g45$^W5Z13tvS9}2*!
z0`LFJPzsQd0^1BweN2#>bnsUX_=o$?x)5MR17^oS#!LeAoggx`Am9FFBnjYFf;#$#
z00Qy;e>-vzfgaEe#z7_u!S8WI38V=?3IVxA2Uaj36DGkZ1$0PVfB*yD)dbf+R4|C7
zE8tBVNXtQ<v%vdGfLaFTk^$C#71#eCb|NGKs09B-g9hG#06Qi?BLch}1b4=O2_`@$
z0-R5Q$h9I}0HY3ww*~NS1l$b(k9a^g8C0$dU>pq%0N+Fa7Zd1W0e&@sm*XH~Yk(I-
za8Ct(jsSgVz^5Ag{a0)MkgGsU+Q2;nWRD4$3jxmlA%Z~dk%1;7Fv<jT|5aTFU{e9I
z9S1Z>01x8;g#h@f{4Y*aAdmlVJ_KGdz$`k@Aq5f=ka=ie=^v+);GPS#wE`|AfF&}(
zE&zIkAd*yolLangP{VeB-+%A?^U)BX_{Tj2vgRMtlmBHY8lcqx_YlY|<i9wf!7C}i
z&H#C!4So{<|Nmk?46ZR?TL6$^0Y(L2;va_+gaoV@02Zi#6#}pz2YmB@k91(s5R9q-
zGdwUR`0o<_moqT%>R)#9zz+ntAc3EM|A^of0%QjPpZy@}1mGL|e|h$g@j(z>>3^CH
zfR9*!LI$`~f$U%cTmQ!Jpu=(h<LDo<|IK9)=y`)+R1HRGKqCp@L;;-Rz#{~#ZeIU&
zZy2l&1Hd*6F#oFu1CSSHK;JOv85~g04v+`P9T-%`#D7*rK=xlG<lv46;#v+mMhkd_
z0KU6Gg;xO&)F5v#|J60`KNdznbxeZFbN_F}V?uqPBbEZ6$RJ)uz@k3*YXEF;Ku6^w
zx<G0<SmA7ej(;(sf{17WANBu}rGxHM3)Z|cfcjsyje@&Uumb(dr+&~4Q^77M9qb2w
zgX=HwlmO})0p0ZsvJ%;VJOH`)FXu#nnO1=HJMsdo$F;!bDzGnH19}M`tTz8vmOyA3
zv<>X6&VgOt8L$_d5B3Kep~qlPSp{~VXTf_%p_fo5^bYJ}<3X481|8oW^ip>y1YnN_
zoHZk-!Co-}?D|qcH!cPHsYJvXAtDp%ZgsC3LcG9^^EeU*MvB$R>hEf<+74NP90mKt
z^I+x~<RaK7ya0YG5htKA7VH7S!4CHY)B=q`!%zp*32`A9Wq}GotwEhcT}PcoC8Eqx
zgHS0{2dN<^R0PTkr2sqLH2@1AFxv(gn*d$vEx=T(-k`Qrqtz<V68WlX)peD>YD}4`
zJf+;GTn(;G$}P%`$^>PEGF-V%`ApfVB&u{(wyH?gS=Ep#UR|q>1uK3e&@O<cq4uH1
zQOW32n8h#}djwmDwZNUkwc}jym+%<Eb3!=rGw~}imRLpDM_57FOo$^W@mulLICtD7
z>}U7_#s`f?=|KmPMs>4li?Uk2M3x{;k?_SbQIRlJpq~0Qb#6*~Y8C&_<kiW6$+LV3
zpD`ujn@-hFk%ZesH^eI?u(VrBkZ)3aRK8K2P@hGlP&Rr$oQd6l%fU^<BX}cXG%1j*
zB;TUAQtPSfX$>@pj-d-_`t+OhIgD;b2P1%yKyRmUX_0gv`UV=BnoD*h8;~1Fmx(L!
z2H5u)UCa$M6&0a2R#ZvZk{6P@VufJa<koTVSk~CgvBx8Vp~Au0gR9slhFphE4vzF=
z21J7ghvAX9k))B)(S~tX{wqO=@RrC!VlBU`S`HD>W|$XnJ1&waBbAWnP}Wef^i7N$
z2E<rEPp7+U++#jr#%TC5#u!vqp5}DTRt*crQ;i|5PQ97>V(lZW=Pb6)E5pY|y?PSO
zldRj?8TuderJ4>@F<~>wm-dqOk>HDHN(aQJWQp?IqEhZ+w)em(w#V?&!DU_T&1Oxp
ztx=ulJCj=t8~qycn!7u`bZd66>m+t<>N_^<$2FU{Gr{AX5}cF{C^sNWpu;EyCLh0z
z+(hl6O{0ILKVbaTc*|^Ou4G=)C}pf<6f$x&zB2DItu@@~_B1}NixJ7Zs<DLbOuNny
zYu(eA>eF=MSqn8&^mm%oo9{7-(B7?CsTXF}YtA)@)7VJH(2i*uYS&Pw5Wc8g<gYx1
zXv=1d&FtCG`Li!&=-7Z)>z68z@~pqBo9kP=>H^Cgi)9rVjq&X!t;^~=YLB&~_s<{O
zJMK4@I(mq=RJsA8z((+X7(+nQ-m@-gS89i9pVw~F>C=_y#_O_l({w7d%$P@Lzo~EN
zpER=R>xo9_lZXK0PG-_}66DAl^>X4n{UcTtX8Sa~sEyih9od0j0$(_5o0DxQ{=7Nr
z7{s&Cn5sv$UF_@bt7HBN{fDoRWa8_|v9i_ep+(;dP3lM6Pt=MsH@?hxb|~%6@2)=q
zpM9QtKjFPE`?ajNGnbxe_q{LwedC;d7j{_x*Z$>WjxsKMgp^M9rRHip(A{VfYUyQD
zXX|HIVB2XUwOMWV&A!dH%zR8gU+cNHyFr1`F5ShH5@?J1DQ*D+rxA&bk&r|@_@-fj
z-I%!(V;{ak%gNm_x^?Dh_e}F1+r+TLEAFq@8ERu;u9sr33K4|gvWrE19$7i5hl`+Z
zQEYA6@iXsxNoiISz52xGfO|`CVqal1le6}Hklg!mv-0`G_oD^<e;hN=Up#*Qth4QB
z4xb&oJJ38DC0mJarE@i2Fjr{z7+<%2;mmYxbJ2CKb-M1-?`rG5d^*E^h3R))k@hCN
zX#Gjnef%Cptc0Uvz|Sxu$)-`<h*<8Z_0TfK_!>DAGgqs{TNo#dL(iyn?C`-YOFg(}
z?~m9xm#}FMLccHXSQ6*qM_4vBsT$RMp_`=`tV#Y>_WgE6cVl|dt5;huT)trOr1(>7
z+TbI{^KEC^?>v1M_<7ZbkFQMMoX9v@n%6kon$dcqqhyfBXRGGn_K__ZA=+1s_F9td
z7dvJ<+f1jqzH+U03-oyCcGdBkCErM8SZpFNvC_LqIggq`lCf0MY}ihsKanT8L{XS8
zv_GyNgNw!a+nk^8pZIm=faN`ts8Hc<^@;h*ho&X#@9<o^B6!1&;P0e#ZiU!LlV@5$
zux{P+E&R*L${D@-O*hk67h6uYKHB_icd<{p{Z%L_@D4dEuIg>x-M3qx`hRY0&>Pnp
zf82?#zt=sf+@k$U?>0q*f{8y(+dU|LcU@A=wM^&P7kE|qMo;%Oi`8DMbKhvSS%ZNC
z#Zx*y9y}F}wBzPU=C^xPULD-VI^!~==Q1H4S))I(<os4t^hN6~$0xDf+nsmiZo0q4
ze#@i9l)zuU&t~T@$q!k?n3}vJCu_%RjZK{TJ@mfhC%y7rRYJzXdoz=uB(u9kp9g;}
zOtrt#pEUi!%{-6Rh^ixBMek(4R`ytlV1W<2r`MJT<7+fO(r3bNP%9X3Z8mtF^$50P
zYYMd+>_q-0eg|wOsRM`!){%al^_x<x+%~G)lQvK%NJX294Vu6IxzMtLmE`YaKQX?k
ztrP!#iTcE+#4~Q*PB-T>cE#^DUi)Op;)S?*zrt>YT%9GF72@@azCkKSVHhG*Y)4cM
zIy1Mpzw%6p(RasZtc%*Gzd!tvb*gwv8s$p)xzd**`CT>lDm6;T#Z#qC4LY2w%A=?q
zGLm4e`j{5SJjCDx1*V2uyz$oyS?5@*nN3YHsqv}_n(mCE>Iz?o=90?vpD>%HV|^F9
zZAVR{)soQucg1t^t*3NcKKoTs_cv^9@G!vdRG&CGCtY7|@Gv-IRnbcQxS!FB!#n(D
zI9~M33vTwGu>1mVfesVJNNnHi%Kpk@O=h)+3pV_!%&bi2UHg28pY^zXUE|8nPwsoa
zMinYMmJFE;9_ePaoM_?>*s04HD2*8Wbc_dagjK4Wtg}Y<is43!c<V^>&3eHa2DBQ6
zrZ!DKlU0UM2-c5hP6en$Xm4c%Z{bi(|IseH#_-N4=$*}4rv_AUb3mV$*}Ihsx2WcA
z_sVgFed0pi&3}|YU2HcS<?UkW<vbi}7JS^Qp8SoNO)`@1>$q6GvaY5PDnFMSn!h7|
zar%neotKZii7KhDKTulnbNHJ_=AYbIZCOHXq*~a`e#@>B4--q7*EEmneA6cw%+P+N
z8LTHTF|Zsq?=(g9IhyBb&6F0(e!?Y8suCw%#joI|jw~H!jIH5b8$H`Cs#{&VYRHr{
zU}<F>ulhUoU6D0y<D!fWsKjk^R(Vf4KMtr^RJvko{I!|UE;g3lZi-+`V4~H1@*!pS
zcpv*(zfu3Arg=H^x235c^0F&VXHUEsyiIt5`E)bayy|m9WQ{y;cFyL~l>=`SBKSD=
zAmpRGr4nMF=v=Xq+4LB{)6~`aW*BX?$a<^EOf58R3vH%Wrrt)@Y6v&ZV{hlqk=|Cy
zMNdZF_pR#Q(oxau(mFhL8F$r4&uExz3td1x*ID8Eady$F{KV84(sV6*?2HYIk1nOm
ze;#_v^P$T$`^%;vMr1u4PPuyWK4)OyW_wea^^ZMY-G0!1Ip@6kc{7Xflk^$>u=W!(
z|8v9J_L0`b&DE{c-ssUzaTR(S{x~)Zqd^YT#hRI#-qSs*xld<?@o!7M^@uq~4@Ea2
z`j9mk_ZXvuMXJXF)a1^|8vebh$;rdq6@$$kS6W^A7l>Yy{8^s#9k@#<b8^1X3wLQ)
zUIfW+y;FkYdcVCfBlE6Dp7N@9*kXsW`(b^?WJo)bdIb}$&=joTa)uA~n|2j6$N%*!
zLl>zE`^t_~-Ke})?2><==s{IL)3UbtZ68}+w;b#=9tz{l7Y+!l_#dWtl3D7@m~FUF
zLN$?2-b)>#J27c2CexR3hc-<0rf#LwlD-lKv2#(8sw;{^%7bbTMUtRv?8fMRQ6-$o
zuw(U-$++Ky3@v+Gw5OGi*q!2%;krK{GBPF_9f=LJ^M2yG-6_C+hV=roGF>OC8o4Lz
zA3r@B&hZ#H*@~`wmOGh~oA<KxdAWa4NlwR)mpRczlU3(yJ!(B_cGPWdV+@9mOZl6o
z{!ZnJA`~ceI3bNXOXCQ01@kzQsAZt1FhC5jy0Oe;sw-tZZIr>JGYE0WA-R>}m}-sk
zswjy2W+;3lP}GX~Nim>Az!L}-yP0WfJ~EB$R6h+nP3)ZM^CZMAd^{+`=Ys1@$9Hx=
zt-CDun0p#rqBo%kl2QJ0o+<Ye+qdI*&6<)Ug)>UGRS~Mcm-^&UexJ|9l*p>Q>hKL!
z^=BL8t-AdSxUK>V(Jc`qaZ}7k=@1Um-ZFJH{aEvvS*%am7`;B-WUV5Gj3OaLfNjfK
zQaQE+xhEf$?vve+1xXtDkH&XQxJ!mHXQ}t-i%3LVAQ7i^+7foz<#uBF0vE1Zg8$dh
zg7BeWN55|O`_AdpF4*)~c$koND`;A<L{`UtJ@#{`vTvl#tfsKYA~z!ccu9D<XGwJK
zu53*9l0R39u9OQaZ&n%o4QssFO&Kwo{5iEr=q9<R5I}!%J=9j#WgQ(obNwHNYfW&b
z-o^sG^P1o3B@|<7GVLN&h<}WbWM9Pl#U7#(UI}MnFp)zSiO`|seBv&&F*J(Xs<qzo
zlB3x9zVi*2Ri6HVK@mYwmqKO!r#$7;y_`=wxY*lT`5Mk*VByvBJHqXgy4+y)&h~eI
z1;x}NeA&0ktcp{`$+=&DC+E&7YAIh-E%+N>^P~>f`neYxIX!WnpDxH2RY>2dF2O`f
zh=wjp#x!9`Gzq$``q73v^p|P5(>IXsl5;5;WK(=FYDD=#-Xgswy2&fx%x4ou_6r`X
zi{P)A7U(qk1Zi0Nk$JFvk7JtigzHw{8==`Vm&~dUzaI42kL`&}4|cS-Ycl_+S4LY4
z2P%7oCF2c4rQKH=-;^)Pf0kQU5K(-k=vCgH-@AUj{hgeDsMN1|c5P-Ir_ri2lKqDp
z#&;G<#gOcT;xuv<dzSo~cAw5+a9MoqgStm`@95yQJ}{P2W|7xYU1@X3@z@$9R+X)6
zQmm7G7yg)V=EhET%V*#wN&5)|>{@(3182;&wR0MG`spZh_V9igqKcA5?hAhJ`_|LM
zZH2SuwEI@;jf*vFiKWN~$&SfAqjv0&?ySZO<?;Ecxo-;ciZlzq{&|~?&X)Y@D>z?%
zu6nY%q<UH1leXmp??z*I+5$a^vtktzguP5IXBe@jSe;tSbrJn}hG9lwMho=oH1E)R
zsok`6S}3Uj6RO^$pelytJu<B5&G_VS)kv{$CF&sI7QP!31N)KxYFk@aIDB+sIPP(1
zo8Ib|8@et$Cb-dWuV>iwTMof?P1aH7V!cK5LadE?uXIen=6xEC=yPczRBkMNSFWs#
zt$b1%TDUN8b?({xWhFN&k}9`XG*?X3U|LgqezSc>kBzV8{o&V%O_2BaBNRQ_QTi%o
zul96<YsME%=_dO68JY%6JgZ1Eg{h`!;EE7fc|^WQ{z@vEdOAuPdOXS&HACI-Ec7{K
zC(4A_q{TMhG|kj$f#WUb&t7`L-4VwkzlXI2&GVn~^mX0lobBLh^TDW8V?QoYr6-=v
zzd62fXjOY^6}E^{NH1AkiY-pcbNcl?+x}N`o_0w^*`qRIS!AWOj@-V!@7u7=IDaxl
z;3Y{_MdRYAYZxlV4rZu!qJe|yT}u<2*H$}Bd-PxHH5q&{%+$qetS3Ie%tCyXNwOlL
zVyvMrr0acO=a{bWsMuGuS9nfhjs%en^l+AGw&!d`_A}fr%*cr7iwcX(4wcMc`*1u_
zo`LR6r{xwObmFLKFi}O5Y~_<iayzAe7nfufu}hwp$_shF>wZ4`QI#E^k12DhJYP{+
zcBSG}ongoMfo{&#F^}<`JS)LfSrMwA&_LNtYozxu19grW9yI%6aoudaAzx>{PLO_-
z;h^4Tri?I*YE<u6|5kO$3iyMZfWa-pi}|LCEW|+lL#~o1ppH=}dLr{RHf);|+ZM-l
zUQI#QA{rxE;eLUJ-W%OYT^-ymJI}H{s#ijB#=t5I*+Sv#F>d$Vx<h5E5~yNv<;}9F
zLi0Z}a?)~s=UprNSZz?vtWZ=0*E2hJ5BiMm9X~$)iCaEF7LOoFI6i&^cMOjuYcs%S
zEPbN>PG&2fgUKZBU_ND?Cz+{Nh)7dLf=A-*QXhem-QHf=Yb#r;tz%<g6J+A8dsX|O
zv2eQ1EaUid^TWc>!PYS$2}hQ#Utk#_bJH`;VI}HJTHLcfsi_dZ?Y6D2s|v3iuTYg#
z{M`L|)%|rZuKlPkbuGR6D==g9TTf1HO$zrR#stq&_z08Lc}zdE1WS&oK!1nkY*w7%
z9MEy*7+=MDi(ZNfArbinV=?i*2<fc)d#Em@L)Gose6ZB&*YBKw(p7DX2jV*=CCE3g
zFNQ^7ynfqh3G7V;2|i&%0UC2cHn<#9pD;LJw9RXE))Jq^m5bKKY<796eUovOxs~<_
zJ5S-<r~jqz43yOWv?ZJJd;RD2A6BK_et-A__j`TC^d^I*v0fQ=sbigw(&>?<g~b-L
zqh|8yfpc<}?wvK^aLmZp&|kY1cT#>u+#oQVSU!GPcw5pm5>?MCZYwQmF70<^=l3&P
zS_?m9Wc+whezZ-aJEL1Y7&QKp7dXW1@abEN)0@%1_0QqyTN4+*kDa!(cvI?@fi)RR
zcgMNKpdu$C<ROsTdToE)Gt@T4K|%7w&mR2GI)JwG;*~G&^16O}Nk91cQF>`w_z&y4
z{*fpA6!CJh%r?x|)c1^AimSrufc;sQgArv5Sy6j!7ip{^b|Tv+1_v&Vx=+Q9T6b^m
z+A?^(^HIsO-#yhfLl-CRjdhLoa`w087BF)e4J(JPacl<$hxnt${hf_D{Vuw@7gX-r
zy?OZx!^EdsuWc<{+p+xSVszZB=wD$Yp@c9ie{<vK&>}^GS|y3(JZe`L89j%<r>>6Y
zYx2JpF3DJ)F3-XiUM}{kvgl5kI3zhrsIbBM&UfdV3ysHZr%(6uDD`<2b}`(}xkqCZ
zc7%)M57=$3W!-CszYad=iye+1mkdlcysHiGxj3FSVK@>qdWL&r@K1AZ{oOwAi4~mH
zz4Lo-^d>iLtgIW2b$+<%+|IWPA4W9IPg#?-LAvTug3E%_vxmb~AsP|CLiX8*qiuyp
z)N=K4?#-6oq9rNHlTgyJht&Mdm08)fpIfp#OLNON)<_56%8sF45Vu)`MPTN&_+7C|
zv6|~z>3zXR5ja0E$<`PjD%v1unE2anQ<?SGsN-BuUFVao!M-J3Ee$%&d1Ic?Oq9Q}
zOHDw{7ikVkJ3<FHjmC2bL+PW%!-_inUpq@O(L19)Y@#nGhph@qo4+A()oS$8@Hng4
zyCPIUo*}+r*8Xp`(BgWTJ%*+<A3fLpH22dzC<%9V^owx=w~YE_N!reWqFO>zLSK@w
z8(E97XIV{e4%+TE&G@d-scBX|>;V0s$)LY3OPO!wfzmya73@>hVdb}*g<ZB?PrCLD
zh7P`LooaL)c0?NR{?H!zQRNCGQl7@I9L?Z#4!ZO_>pwVtXwtshAusROiHR*UZtmZ`
z7d@Zrw>v_ekdwG~*=XE@m;>P{fj@#jhR>fdi?NJ7Z%84UFJ3TK+%+@*#tkSb^8CK^
zm4EAV<?m1*k_y%J4y~7m_sG;LNL{E|5I%kN_Jz;fVe2nWmpuBse)~=H54Y{Y{S-V9
zSBm3$Ba1Truq*F3U1>A!TF*}FcWbR{F5xB-imBUCD)~wj8*^E`L{>XR8M@kz?QWZR
zE;%{zxcODy?mx_)mu8NMjGZ3yR(m}UEm@Ve9<_X3Tun@K<Y3V8z|&zrL%x~EjcjOn
zJz)v$ksj^+Rb&dXwdPdP^UL``g{2?qsre=0!x>Yz`DZ7D<0axxtVG|mxPm$Q-Y@Ko
z9UWYio<-jNUdHxwSQpW8khNm4SNMDPx2u1))hus2*|nl;vdQJ|`<i)dKC+(Dgdc|+
zN#3kdW+HK~qG3eUUf(f;yIQU%%I-@k*U7facIswYXRUp{dsiHNM(k|ty6fwMm(7k7
z&kl%s8Eg}562T9iX|RC3pfgHbuh!!(Z>q@ae7W+p>shCVo?rK*AAVJv{;~Eko7(%a
z+h&w2dZm7+dow(0<==&mX3TXhaQF4R;QKAWJ%HuxuVp}RBVU9)2jAtye7g1{qUc=J
znkJ|21KoRDC@q5ntNHPwN77~R2bQPa4#rG4O;$CrVR#v5l7C1V%7=%ywHlN+{JBwg
zjKyC%zMsFcCj8-?+iPE}KeoJRLCoCYnYTmc1`dRFhODyN%YV>0i*Kbq$?xxIEdTfs
zbrDK>f6d{|wAZB%55Dlq-_;S&N^k1y+A)4^Qd23g$cme{{?%fg$j+#zF>7c333%YS
z$n~Y&G*j5bL8n!+ro1sN=c{kt`GU0tp_L}BhueKx)4NEV%&{e7^@0klfX<`UWAbDJ
z{KVlseM>k7f~!*xMmG(o4l&w!WhJe3`g;=1_p%aBMYJu>+gPy{yR<pZVBWe}xX?8L
zrorl91NT1Y5l1F2SM3=`cWiI`oOk6BloWc|IYp5Y@c@2Q_nlk2t<Iq4ahLvN29L~p
zO4#Q3aE4`|S7`0L&kK8E<^~IA2z?wJVas#2GV6Jy)Sks<!DXSP_p%$mdlZ*8#Wf$T
zyVaE0HM9GAhsJP@;u&@fy9w)x&%$1pZRL)Q?4G(Q-prSeQO5!Yt=iuX{WU}{Z`^fn
z(K-JIQRAyi*6A+Y9eZ{5=9z`TD+1z!6~PxBn<PxO4HSo$4fPh^E;pz=kP2Tsa^2@s
z(^sE|<Xhe;_o|Mxja9qU9pg-jU+~KmR@SMrPRwla*ze{TWxqHgF3%t5G2kV4nXtf`
zY}H=@^|x!6nN)|=jOB;?`dRwDZmim{ysG|1??C_6*5>BrBWICJc)4n|!V_-BqmUNi
z3Hdy@6YVFtJw86Xj=ew-WBhP|YJGha!`8~ReKB#pPW+RZyJz;yycv4b|D8W3_@MVG
za#H^=Peu*L$-6Tv?HiWVI%KSPrTZC^xBTObD_gHD{rbB(qh9t$RU9<nC%}x4@$=~}
z4o@8Jm?Mq@(VT>fbC$Yi+NHbs`)+VdXAu!2(bV`5ThckNd0~C?-}_aL)ujy&yI!%c
z427^uMv}#?sC-nd)Lyh&o`iI%QR*YecX%HLk=TrFAAUPJF^c9@WA57Dj<gEnPcyb?
z@L)&3jt=*4^y&}&81pqs96S_sHi+OZU^=T_p;E~`7&>=OdvRkxv#KHb?~_8i3`w%-
z^VSrH%s2S~#UG0wR!EvKgY$TwM81-G2}LcY95vRqKk1NQ_t@%;MX<#*GcSW^4IKU*
z)T};>GJ~<`21T#%-ITuAMrDDK!mrSu)h6mP%ti75<u;ZjYnzlzj8550|4Ls70>?&&
zqDPV@?@y(V>}-G2u(Gcfi7@<PwcX$s?Hg;MZIDlh&xGR@n=jL<JQ@P=5#q=lL6*)t
zjMiz>^`!cLHRS5k;{)u&9FK{H39rE)f7fP{GKjyolxtO&m73?zE7(>&+_8e)I<&g~
zeD|Ba{k#s;7;TTn8Y)ELFwW|QnBa_rtn<VMj1YH@o~yM*^C`s;c0%=HUs4@47?kg5
zk}4YdNi3i}BgQ~3vc;ko!dlTqNw(Ny^6v1~L4VHdiAxif!`r$qcC77xBUne=p&O#t
zKzAhLbTk~e-gi6_r{Qflj?;Xnho*&J3^McpAAijtvt;wtMpx+C2t}}IB9>>(4;c69
zIaF~oyWl6KFr~_@Qae97`+4rynzy~xoP73Z*SpTs!wIr*f(<R4;!WH@D$<bX>zZ)%
z+-O_iB=jJ@jb6+grWU}VhzB~0$fmhcJm7p~w-SX>5Vc7R%tkpzyhhwAy(l{?Nfk_u
zJ9DV)CBs<m)p3L2(eAw7QQ=X>3{zu$A^2vk)o^&atq<h3!6D7@ft%c$8SpT8Z}4d^
zwG~a*Uvq_ysdg6Wu0%ETzBiE_G8RAF(xlAY_HA?K(}HW2mX)r>jRl-CMoayWGe4N`
zJyycmGUbac(*9*+YcQ$ZpsQ(GYu#r3)>xlO!Y{*Y#b2he7<ULL)m-s)DMuw&KbK(z
z1pZ6GWyvksVfh-_UUBjyW_-qkzo0;5&Ogn82keKAjq`XS&Pd<B;Y#&+{p)t;t>zf6
zH$Lk0bjIfp?-`f;x@Uxiq=#M&KQPlc>X9GCmZO`-VlZP#u8Je0JG+hAwzO_-o!79d
z<XmR>NBYN>FOtmRpOx9wc>z^7+VzLFatFq{d1u56(Ux?UUYl{N*)PlewkZzh9MY|~
z7@lRB(=StlC=ZBFaiJIr5-4{UqeRbzQG(69oUyFYRoqdYYO;|><_>bG!(p8C(HUdZ
z5f%G6TWk2{u-OoC)KM|1jdNV=veO2#&hfCDl{{~G6gf~Byeh_OLD+)gd9l%Okgdx@
z)8*Rnv<l28(Y4{U&h<^{btbioD}VjLf3$f%_Kf{D>(i-E37<%rZ6(iJw+>z!QI9_t
zTcM3;Ho7y-uGsH%HTDklUF{2bpPOE4{Y0;p`V&_RPsi@WxT%9AAEstao|@P`dX7!(
zx9OYSpVx0XxRk9LiWpil^m({z^x@cLj(lKG|0cHCur<ey-$tsj*YiK%QEQv;U>W*k
z;r+OY&;$O5!c7)nmV8-gFn1_ovEQ?4hm1tb_rzuD>%7N(*p}dWS)HJ^z4TcI<7LkS
z(u+$U;IzzlHXqLXJW}<n!(ib2h`GRB{SG&WLeRWmGVU<T^O(<+Pov-E8TNirPP+}d
zD4+4`h`GdUc&&1_&}DM^gw<Hc;P3XrhW5JI4Xoyct%Q!)9zwrxfIrwXd~j^;*qfn>
zp1$77$rJPur$>Pu0Y%<^!EOuptO;D<H+OAR=A46z8{_xS+Y~K}$`7q}3pWd8+Tv0q
zb?mmr#bxIT8Vg9p@ABS%ae8+67VeJ5lg8(=XMbM2{Ft4~Z0usc6P7@M1XC)HDK{*&
zOL6!2i<!X;*&pc`GdQauFxBz1{txONJO#!hrDE%e%be_?Swpw_a$3$*-7Q{JtXHaE
zmQlK?f?Dg=e4zbRS5E&Gwj2A>U?A@*G1czyjQkMe(93hm*BbA;w)ew^Ys<ec9bUR%
zF?asY7-1AQbeeCCgR!v_O<%c{<KA?%$T2(ShwabD83*63d(?DW^M3N<(MR9zuYUOC
zb^X_%Lf6I@Lr<l<@Hd&|h6y%vTo?O44RwqxjQlZ^JO@87be3MgGp7)9Po4ARQDmNo
zKel>ce<!Au-Qe~2L&>iEZF#Hm#J_WLeEyJ&e^)xx(wh!+B=w~Y#ql@e`z$W`cm->Q
zh0b+d{e9P>y~aDwZ~D9D(Tdp#%NI1yF`mT?R|oPv*V#MhUPk9l9O)^n#}>c%x%qQ;
zYT26$Plfjgcirx$-nPC?eDw5fQHJrKOC__L5=PvWJIEMab4wTJ{l3!h;<;lBZZDL?
zX~vGvQAHm2?{-p{j<Bro7K$$38}`-Cx&}_AQAt_;-QVt63%=KX-TqZ4Bm9Rlr>`KY
z@@XBh`BrDuIF|U^D#iPLNcilx%a3l)IQa3%(Bbxd1=}k&ND>E^G2`*`uST;Y76tlD
zKVayD8<<+u-&nKf7yi?|x6_gjK5e}7^g49i_4>4HVb_Sa-aKT!)=R(h=VDy~J6bwU
zbkpBwTjq8rpg7{;oSeA7i?+unENq`=A3Z1RmhUvjMMiX55~@N(9|`Mv*=Ss|to&_Z
zZf?r&l{vS5KFbWvoSr$DDbDUK(5R|yT-Ki^U8ou0!0~$<^)fzo{oEayyZ7y-?@!#V
zyQOK()fJM2qPTmrpUr$4u_@rO-7SWjvVDAax47K?$G)_+AGW@k_N4fZ&FyKoW?vh<
zM7_S~5%*0*n&eyTAH7=F!3Oae{B9ko)jGH5Gi)O1vt8$n&$~Hy&TRB7d6<5H!gaRI
zE`2XbziRy?Wf0Z+s_J&p>AZb^rvI9jHU49L*2bLIIrD#&=IWP3R{dzYIoycyHlR2P
ze3dc%iGw@rcL(qMyxV1;+s@Q=OIG+T(@j_&E1p#uiH~RsaI$;CEW-Fmj`rJ@v$A3{
zZl}|ew?F#!Xx;PcPik)`-{`w{?&;ZQ1}{-*#M~C}i85n!1GHCbr}-m$2lr2b(;{z0
zxks`iEF-psMEcu#!qYi+8OEy^@6p>O=eQ9)OKZ0j_5DWWY|Zlgne>yNt^Rf6w?p2U
zlHzK^hS?o!xG&(#`U~x|d~SfpJ#@DA?Woz~aq#|ry={k9&s#Zjg<eAPyu-5^qRxk`
zae1TPPq_ypf-7zDh4wiIeq8&g@$AXNmCvudeERU(t@ishuM{b5$w%K8{bZEO>-Mzs
z#$I3ow75ow)^FV|1nGq<LtH}+ggpoi_h06=(}_JzW%XS@oMr}JRQ#T781Qb*tCdzH
z6`}w9$`Sv5ohQ#F<xb>3EDNpiYw;d3P_{E>o1{1l&Pa=mS@AY;;ky2<#GOw!2Ci7R
z?AR)=6|Qj?XMTvB3cKg|+3XJMCB+{Z8dB8=i@)b5X8uZXe(^QgKIPccH+TO$I`d}y
z?X1__H&z+P3!2In)XePtEH=WqQWLcq)_C_G?;-C&zv`fbAPt|%={)B~rwg{*j8?H+
zD3@RdSrK>Xz|pS3X2M^Sl9IwV#n($}3a{tK6s@Sxs8KZd_Xdj&l0tMhTHJI0Ide_?
z*(J5hyw*8viC!m3kj3k)yt8Wg!u*+6qpr*p2XJlY=pWZwLEa#^+<w01c~wRJ{`7Ni
z4!mFfG3hnxVdmq<DNo*$Umt&kP7nQ^RnSqA*JR6`uRMxLq{JEy*$+AAJL|fkd@_B+
zZevb3=OpJO`}<~d^$oSMs3dfWsEIp&Xm>ZJX`-53wV?WP_43Lq<*dr8zh!m!CZFzP
z-VTgNW3Ay8d)*n-Im88wMe=2hYu~Nj67M#D!=evM2@79E-3(hE`89aVDcne@eM2Kv
zy=w4W<D0syN?G>Cj}Jdi`-)GCdExwYdCExYsW+$>!SB9i2LJy0=S?}SXX9kO_?7BA
zwbEGEYM*7PE!tJ?UhCv<&vH_@zH;7XlVI`DeALjNfq}=A6v>J4Z~b38zO>G3&Td#=
zUs^}4y<JVKM%DbOw`>;-b7V5&Y3*ps8=k)-edn)S=)H9MYSQZJgx<K5@rRdLFWxkJ
zP1M$zZK3;I-<!JV4l(wq53sK^{jI%Kek4o#V_NE1`l?iymyXXfQp`SFNIv><@Wb_=
zhjW({hSt6wED(&y_MqYzd?Q_RH}gj}FP#rfk8o5?Gjxe@O?Bj0*P6!}CuoM?bCnEn
z(}ewycF&uRTWwJ-tmcyDoM!Vzr@Fay%*HP*Y25|f9OMQ4y?%gAluzKyQ*(9aF&A!I
zR=Bi({<oN0aR-)$#Sg_in6-G0WfaE!tyzsethpOm!`|6)zbX4~=%0gMc^M0`IA1j1
zTz@5bxA#5&#p$O(Z@+&V{2BfGQ{~Yf-q@MRbm<&?3RA9WuYK7Fv07}ywp?dp>$JdS
z>$I!ZepU%)_1fP_i!edzcF_`U*Dz%`dLXDXrY)hfsY|ErdgJ@%IURdCdEI}w{uqK*
zq2XyuJNGYP3DGVwP;CEF^71>e*Jo+Z^I5`Acrx#C^i+&q<aYNsGd=L#a5Vbk$cFa2
zE&Cddme}Xq$qxIaoniEr^XmTF-S6(b)O_~&&63X<-ydY_mc_I`V@Hmx;cZp^#qT07
zq|VZ^GkIse(F||($Pwjw)$yrKv(+ApMf&x$Ou`1779<ya=2`HLjvpBM-P_vxvwyT(
z(Av}@=(yND+T+hr$^s}?wJ}EiHW>FCeik!UhWgDG#V(zDXQo&5w)v)UH|7-1x;e8Z
z)ZXK^rHQ^V%LuO$C3EuomvwtIU8=lW$|*_DKm23#^N%mF8MD9UrN@5j$k|#jRs5yu
zb?fH=A;*Q=I>l5@qBg)J!e&}O)0V|xrE0I%f2UWZmBVUh_R_W!lHnvc65Xa;E$fsA
zDPD@>`HT7NsnW?W+!pS_N!4V}1Zq+&xreSL+iU(bJZ&Md4z{y&#Jje64*AXwc<S%x
zo9c7Quf+eE-x427FN`bS=Ct7*%`<c?;ezUt@BnYusBgb}`<mvjjpu8|OZa)Ob4UI-
z{#uel%N;26DY;d)A3PoQr1MtqUG|N!@X4$x1MyV_9km#~kGny1rQD}>(0nvZS^lg$
z8exnNw6o;7gm!!|@c{8V{v0kE_ZkjE9RSZKr-5@uLY0N8P6;VZmG6*?*k`09>PPx3
zrmvQ}Hc6*mk7InzV$!PKdcEy!yCb$2t@c_NSUfU!HObZ+)_kDxg=#{$hY3TG5WE~K
z{LS+mUp7)d@UWZH(b@K-`CUU|!^=j^Ca<PL&E9Rv9m~76_w4B3%a#ql8MT{O&%Y+v
zE^?Nt<rb<v$V&85>|ua#HaVCYK}(~#(&y7()6P;E)E)|va*uR_*h@Hy_r}h~q@d%`
z;i#9$6SYLeQI;!wm9^?Ws0)~3cn;o@R7ftQRM4E6?V5vHP1^5t<MpZf7j<iNG<CCd
z0=3&UO*EreZ!|LLIGPLPGBF-U!@!W8>Zlwgy(H?KS~$6FqL~{!_H@LHqcyB#mk$aD
z@&<$hxWQ?IIfGx=JBBzz!^24<_r{)%zv1=pV}!fJp;BM@6@cL@G#@>HNyOg8RpZAA
zDa1#_Fyc7j7NHog#HHb`;HI!l>~2_%xqy+NE6@$-ZnQCm1iN9QajWsegmjXe{D9&{
zjirI_g&CO|Xx0_YgIb5QbhU11R<kZJmuciP8W?iMcSafAh?Yh^OVGzc=;es5DiJ&$
z&68$G^u%k0Kc~tj?@lBG4-_MZI5xw_hO~yxvd!5`*u(7QLrp^uhu?BqN2119<7A#5
zUtgdS7K%?wSIHMD{ndTQMidtv0w-b9aLxDu!W|-yh$ZoeABYQxhD0-B1aS}XA@L<~
z6Om6?Ncf7kz&GNAxD)sVgh-+lsfV<Le1-gmTtl{?yrC?odeaQ(MRZ$+CxcDjLhq$5
zp|PolsoK;J6f%WLt|Klde8SbhgJ@UOHzZhHp<JtA$UjPLC09gKf?eQ|>svgx33wdK
zH62?p`fbExq=vJLGo3@@FgOc21DtOoKSt$a+s7pnyC%E&{(`$gy!e*HQzn-GR$f#`
zLEWeT%whOFwh&i|uP3w<8%XcTvnXFEU6fkNBg#xlCwV2ggS3#;LYzx1C#)hc2>tkG
z{3zZH#7jxoK^y@88z3QFAe&K2D2db|>Rj3_+FRNUS{w~e`#=q%woo=vIAkyK9FiTe
z6z_^V3V%X(K^R0=WvXzL`AGakA%YG3KfI6$9{2Cqz^L))zL5#eE6z2}6HXb2F|uo9
zU}Wp4<rs}?GQMG=m$#p<C-@^w6r-fcvegQIl^tS+GQz0g6r3DSBg%;%NiO76vOOi4
zqNG?-?WyCG-IQvw5qS%#faphTBJ3uJ@T>4Ya0u2NdjOVVvN2yV%@||&6x<K{V6S1>
zSXbPAoFTp(f12P#Y$7I-zL3gErKBgM1d^0^oH$9CO}L5A!u`ZvgJ)r;P?sPUaz~}7
zd@Xm9<x5tJHAQ~~FQ!iM4^Lj=WlX5YH;*g1+1wZ058N6qaeTq}k8$^j%86^d(~~dx
z15^IOM<TU&nKV;IQ7loWs?_S)&>d7g8o_8_AzTCQ2Hu-cLs(2~AbOGxk&;Q@NhzdL
zqy;1s(lE%jokVY<i13aOPrwp>;P>Lg@RoQ6o{ZPUJK$&F=i@ixPv9TpGw_XgCEk$W
zMhGBy5KIXwd>#HiJ_^snUBwyU?qbo{b#M+wjrK%uLEQ!?55K5?fF}X672o9<vP@~Z
z<c;{1=&kUpplV9Q_vBxjRPfI5Y<QfByov7<<r7mAe!N#a>&ZWpyZDh)b^;@zl_*fW
zPI6CLF2gEfl-E?{Y6x;g#i7?>*243!mbf0=6}&Yeg`h#)O3Wm3i5etZk{ijHWKN=x
zIK&^sgT(2?0m4;+6QL1*7_S4kUW!v<A7W=<d2kB684d;CUm3w#ur_QATf(-m2OJHr
zhIhlK;hXR+_#C_oPJo?Z873cd661u)M~9(*ptMkHpby9>c(OfCbzGUM_$!~3X~`m`
zrzC}9UGWLgq;R_s5+)1c1PlRts(p$*g%^Yho(QnQL*UrXec%yK{8e&Ex>Xh{_g6S6
zjZ_4+5cGjk=p||g+8r~3xdfYH)38psCpbB7KK>>?A3uN><0W`Dz6SpZe-=bT2j7f4
zgLA?)V^3o}u_E{n`~W@(u!h6aVO<yoh!q$#Oandz!i(X(@I5#SZie|V9;=Bpz_PGJ
zEFUg`@4^dV0{jIt8`Fp0iRPg;qPoDTh|h=~azfp$3ROK;iWISm*K&+}qpVpvTUsCq
zmQ;!th(|>?L_s2ys6qHsm@0fH{4Q)1QbaML7b28+gSbb$Trwg_lA6f=$aczY6zz)J
z$^@0Zx>fxEiH0z!&!}khB>EhN4p7mtC$LReJ=}U+3a%74iNoV5coJTL>%#rPUBCt7
zba1uUy;uiqCwvL^gnKZzFn*XG^i^~e8iy_grxDhJ^9Qyl6O<Lo37k;aj!H&#qO8%0
z=uc=mW+f&a!-Q7=X4+vbY$SFg_7L_m_B#0g{9vpGI1>;J^sU7VqXW>lz<CBEXesho
zU8X{*td#DG{UBc(r6$sil22k2@fDGV=!-B!C=>JuvIWlt9|R==l^|A_CUh756#0oq
z#1AA}q|q`%xwb-IIieg;snmH$GV~U;9_@x9!g=sEtP*<^N5=2Kf4~z7QG}I*Lxc;2
z(}V+{r_%{d_`CQR{19#*&J6b*c-svhhB=s17=KI|dILHJMCJ<W2=G(}p-{u%T*zl|
zJq2fq-$Adz+2;397t{>pK@;HOLWU?a6c3U@<4`piyAJIJr%QXmS=?O+6)94mQd@wE
zT&tR{dZhGK5|ud$aLh-3P|+;ACG(Ykl2K)sWs$OUDN9C|K{8X>JlRRv0a=;szC28R
zReoG?N^w^4O=+O2QFf{9)$i4p!TF4Z&=9mA1*5N{oiHCUTu_-LY!o&f%fsSv^<dW{
zz{cX^@us+gI2mpgZZqx`?giEfRNxA1F>C;nU=55Tx)hy?)<-=8=QTHhlS&eGntD3O
z!C9&X)fe?w6-G5nm9NrLVboI9aaD&(sNAYfLK@T;)SJM$tSICv<c79}m?%1$gw_OS
z9+J^C^fmM_It~+x@xfd{2LtZ7==a$3C=WFnoLd>kgd**#>xjAfiu!`;k!qc)TCrN$
zpwLvx6z^pDvI)g``C@sz;*mT}<|a3gn9CNUpQzRgisVPsY-zc=NpVn~1npBER6K*+
z)dB?%V~e1a8}Qobc)1ee2)C)Wf|^~XHbSb1k?2Hd2QeCdU-cZdjj{yOr3xe<*!M~^
zf)=(1%EZK@UqExwkD)~TO$e(LK@F%EaIxwo(u0muxghVcK`JyjOLYaEtGa-yLDebm
zA$u`5Ftv(h@O?~`@;c;$y$8*LmSFoa=ajG2NtkuGl_(#~8FUcRgaV&{P%?Oraxqw$
zkD}%%9?LHh?4?+xFV+sJls2eLr9}BkXc2m?aK4Nt=}_WOYM~~oUuHIy1EG*)ZWF0S
zcu@KZEuPx1o(DN8tW*v-e~}hS4<9Dhg7y-UkT~gfDo=e;DZ^Q!cVPAs4B%wt1?p27
zj(iW+R39V$hF2j`U_E`Wh#(vx9Yd&cH2xO)A+8(JlR)_Qn1kvt)gCMX>6h|IJ@O>U
zLAs4%kthbA0L3F%<puO?=^~s4{)d1|1Si#`p3p&fge(@y)LO*N>O^%Cd;oJ)$-@UL
zq{K0ZBwvg@g1ZPZS(g>4*dm!uJSjgV$zt9YNyQSJ6si_s<i@nE<6o4Kn0#?7a#h(U
z->P5=G8J*mr&GfIPTCALU)C+IR-S@uCbqzP&}(EBQU<y}t*MmDzR+h&VkTpBPY8RZ
zts1YTB-9jqPx2VESVN(lFG5ofs0P#vHP)#iaXTJ?>#--1CW1R(1M;#s!W$Ium^-lb
zvIJ!el?!RftF@Mm&qR+gzlr+^Ta<y=hoo%GGsS83Qf8OzlDts&Hl{}uOEkc)#Mr<)
zl@DoN3O0H<+zSU_&=N<uPqL9N7h}-tRofL=*jr=kP?=D<j4ysH4}~Vwvqo<rMEUT<
zY+M0<tCFWoR14TkM5D}9{#ouD=#NI^&{EMQ@=W5M?iOX6xow}W5-|`;b-3pVEOaC;
zRelR<*3BGg5uUW$ASfEirtTt#VBbS;g+;0@CYDnWkE|u;mm<$>y15Ob+S(^XPZ4{n
zJ1J|lUeRPWB$|QoluL+W^&VlLRv<VZ<cT@1zAkz|rfODD{n6Xt*TnlsjSdHYb-YW*
z4Htt-A#-#aafw3$=mwcZDwJKL`C;mYNSMP|KWMt@3>&3%m7<k5SU1L<hTut$!3yzn
zeIxa{x;*}RaxYhBOhT5ojP>4|wox>Dj0C?OSw9h=PKAO91mo3BBcslq=LP~M9y0ax
zB8L+td{69{CVQ??f*KasJHR~Y=pmC}%rw+q?p>Oz-h;mFf&hoR%!93$sAFhf!BzAc
zxYLn<su_HY$|DV<rBF2WDv3hfX*dfC9DRxR)(9t^)<V<G@gvNQa)$gH^D_~F?oss^
zhe$Eh1OGmi5&TKZB~CIugSXPs-ZY}4`psB`z*^3n2&Y$zM>{g{1>=0SAunda0L3N+
z4&;ecidU_k+(RlSy`&LVe-%|F77F%`8lfM6FZe<!0|pD#31e>t^|=jN`X)LYeYU6j
zLU?uYC6cc`X)qYavbi-fB^cB(7GhA@tapmbf)#{xRw%=SWv<MYSmKr#1S|YBM-75H
zUfLaxS~1$El3G8IrJ!~?h&^ht<CJL@n`PH=Ha6F22gn6lkih}c6SbVN4EiFl6K<xy
z)cCAW%L)V}-a3K71Zi|X@AGhg^sVX+2g5Dyp!VccPKVK&&gE%SuP1t1Yq;rSdUB`1
z8@+1XKkS4Syn%S)QGSX+neuUqvz6J%F*cLDe`G&?gIY=!!-kn1ZqHrr<r@$j#j$9A
zCuC}bgoOflUs>m+e45IWLn7B-!6yZlF@DJV<Q~*thD+UEd59^l^w-KI*lO$Z!3`c)
zn1jr#rgQc9tl4G}+6vZfjWE*(&Vf1(f(rwO)Rzg*7(?nP)GzVjzTLyo{ic#_Lz~N*
zyAX}M!9NwcEh_HE;>RV{_<e0|y<JnGoy2+_UhmM(@^slbbZPJLv9EmX$q7ECCmxA`
z2L|UV(y3`!qp|s;>j?^rbHi`RfAxE(#uRGi>#$7D5{{$ABP>Th!z!0#hVIi)nqr9r
zl9M(Q8>of#ZSh~E)q&rN%AqoSUV6TBPPR02ouS`GGqA|8-eO#Ue{RF|&@(zhJOnLN
zom6ZUOYl_kTlI~pqg_?qrsGB}W#GA$P60H6YNB=CE54Mssx+?Eb>h`vZ^^ll^IaG6
zzI3^_`wessotZi;=8VjmSk)o?Q#~1C?1-u#Nf3CM2M^eE9H3sFI6r0Xyv3zQ_FhVd
zQ2Y$RZo40@r%6xXd*)%L`$$sQPcv-FQhshu(7Gr7#ah}pBmQ6X!#QQEF8HJVhr71`
zi*nohhNqK(3A(!^6_5~7K@kj4vAcWQ?N+zl-7N|>pdcccgh+RH_w+D5^RC5x=5g<R
z_Br4F&ig#?dtL9`7u>bhzgPX|zULR<=f2c~<$l%3(e|CLmG!FNPobe<6CC3CtTH(#
zUu&PCi<wmCZckTLS7mec@O-_GFS8bRTtO}AjZdkLTXg@?%N13OzTDo#vei|w*^M8c
zRQlIgq>UL&VA9a#Lfx9JA6}JJi9#t&`crIYY2)jo$}Bp1Ov4N#85P_@la6s&%~oZ5
zjml+_1J^9|M&%nURR}2O6wy6YImgZ3H*VzC1*dIV+LxQ$3>gu$+R|}4W6zJd3;b!$
zop!yB{&u4R3l=5MKIc1ya#tZxIG`hyUvV$UOJrO5;<gpJb6@s8x$!ZtYH|CPs_kFM
zFFxKpeQ&{6w~{j@H>zth7dDpF8O1&?ZmW7!`>SY_a^K+b_Bs7~zSp(iQx^-Q_*c{c
z?bAIOvQCAyIEXz1ukU?G`vh7?@rFc@O;c!DZLts4Gc($2>emuLGMM~k3N<p%RG&V^
zbN;LqON&O29aSAtvVyQ7X?AOfiF>q*t~Y5)>Ee<Qp^;3oLs4Va${<O@+dAvB+t;>!
ztDExi)&2Gx-_qtb#ujXS5PQ-4uKP39hhe4N6?<QHC){a>sn@LxEZSAl*X`8QM%#=#
z-g25P_)KfQ&q|@*KtJXl!5tK5SK4>3Fias6OV?XV4BXhpxM1O-!8&gDuo;0I)o9cn
zES+8)=Dg)%)M1&2p+QvouF2c2CWkC|I>m2k%<j6?M`lI`ntNtO#xJg(6A&0qpD9nr
z{7Q+}nvIUDJW*6swy2PmA$uBn>*&MTkCQ%~&rVMMe3^F0_`!^VH!Zp8=9#@YlXDmJ
zovq_k#qsZAe(<(7uXJ0c)$$o5@{`*TnT#DRJhedQUQc=VKJ}nre5VI`isC}QkCue6
zTlh-YjC)4gAk8Oz5ga1AM7^BvInGjjUKA8GXSv>rz<HDBQRe<wxNqUbC5nYC-?u(z
z=U-oaa@~ngcN$-79P0#=#kx+qpprM`)*b6UQ@b>|=~BS0n)Jgp(O=d)xOi&*u@hI0
zr8N}?eX)OfI60_cT*b@JjG9c|#L6!?7sFN5+=1|<i#TiiYj(6}1clz>FDt?1_j>S&
zq_e_e8IHQl_ylTe=@#BJ0+Du+vKXD(o2(k=Gkt;}{JjoYzTcR(Xz^B`NxK&8o^8GC
z(oDDMI~JdqgbCgfc4U!oSJqsHg-ZTF#jxqq4>9m2Z(*OWjm%wD?Dehuby>{UH<gvr
zk|jyzS5Sw{?-;xvC=Sfi&uFMTUv82*ruA!=V}_BSkom$;N8Hmjr-R(7E&qzUM5YRR
z$ej{jib|&Z;n&u~#T!qxKPA1FSJXuyMzSgDWvX{w?545O>>$g~>!S{uf8w-gCx%Vh
zIB8kmq7zH^&t@#JUGQ|d-K5%p>7o0VjM<d2)H`T2{efU5^_IPf=|+ipJ4zVX?A_|}
z@%y93=wBZl`^L$cpL_nP={ekQx1T;PPWg(>NqmnlGwGoBERh`lim7kJ*kO8Y1%zuN
z8;);IGI^uvEV+m1SH)=KcXa~nQASYwIOz-<w(_d%rfdWuwSQ{cOEm#g*%IA&#oTjP
z-_-1->$#py*H~qdeak{;*Dvdw+rBVvB6Eu8;=qwNhaDKTbUE!{@4Q~OH^RyNC#adG
zclBE^HZqT1gU;ycO-YNNyT#JqT_|X-!)5l~*miCHov#lQQ^ysFN+JrM<jySj5+2UD
z`ThbaoNF%AbybUV3aCxE!AO1~hvn0xYn&fv*`wE2y@5X+{OpU3Vqb8#Us{8zdzT(p
z<kofE|El?K^&!d?qteIiHxbb@2u%E(pn|oHQwag*1CNfh3zg2;6GinC+th_-j#ov+
zPtUYhpc|!RlL`78SZ?^^Z5EBoYLB)>lz1dMJab95&Y79HJL|~Jr^gOOV?W(#ypqxV
zHZ;>De@=;KWy8R(7K8Q-IhK`zE$vy+T2{o9H4P^-S4t*vYgI9rtcIB3+lecR?5&@%
zI*Fyjbwqv5bJ)`U{GjLVmza&R{Epjwx#Y(_*dS9|2i6$d$#$%WT>)iYL1w!gOE*tk
zb8pFRpTonH`X}s^3vbUH7{1ajo%V}Knzf-$C#Sh?Ir@P>#MSAvC}{o~^Qz(9x{`a@
zp7FZR5)ymTT=KRRHGSOpcFYIAh7;}Q+2aMtnRzYMGIHNt@tD?kyy2+C_Njw?rZ1$g
z3O@}vm^owD@=)3;ewa>{sS!GecnFij9Ikq<eb0KT`7JzdV3^A+%NR}wt(;z{cQ07x
zPS$5T7nyuD=o!9Z;;QiT!*2yo3oISp;+^Qdb@(yMhT$FlHX}Zp`xsKuqm^97D83Ho
zq}X%dR`c`PUCpBGxDQburhmF!B>kq7(DvQ-Q$?I{-jQNVV*Iz*_Sp1J&J=akw<9%d
zJe7T>Y98ZsXL+UEDy&z|UqCx1TirX8^U8y#oP|EY+=jX39^t$V7iqabdty`D@Dq+w
zYl@k(>Vh#wAvI$#?Zhc+9lZ;qqirucU305KuN0kes+n@#fo8pE)Lo}Tj*Wgfb|pG~
zRHkXZf3N3GZ(CD>_>*D-W;J7}V&fngmD5to9;WtfJDqkWGo{$Ia(d<S8foLS@5>X%
zln?Z#e0}`=+E<&#ny#9@t!>6RcZ%<Hi|fwiCZP9sm3Pcn1PUs7>JB@so+7p5zD=fe
zvtU3aWkga_nK$%Iyeg%Z9H$8f$fwjiTeV35uYejiN@_QLAdXh+{seU&58l9W#puPb
z+U;A7Mrpa}vy3yfTpXYIQ>@GNo|}9#wKMpQam7j7B9>-ho^5cInMVI6F;+JWUXV>u
ze^LGr>9;Ivm$a?v+t}(+Hnw1U--hPaqA|RwwNrk)ExuCyoO>>Bc2T*MUmo1-tYkF4
zFDBL6;7Yj_937pqs>X&OOBcFs`8BPFdQW8~3bf8tBNQoxW@O|^k5Iib*hOWtrs*v=
z)zd8}Z#DUadPaVk9H?hROw}2tUqCcd9@Yt0ES5*$Flr3h#-hbJQMgGy!q{J-qxG8g
z#Ps1XBBj~jtYHY_kPaG`$Vjr9ZzQMQG&^P>le`(2N&HSJ;GV?aBRmx=c<z#Ut?}h+
z+lsom!h>D;1IwD?3orIr)vy0_MK;h@*;L+tu-mUTJoh=)q4{3TIrW@+1Ho6!ho%I9
zg7=V8Ee`45N^UWd5MKAam5;X$VHnVl+KPLh&~J@mqS{Cw%+hry)AY@S_66AOv|uMX
zv#ck<u9p?fz#3N&!-)>UAiX@Ay;xW0sq%^Phdvr}1)n1+S5v`{&D_?`Wb)BC(PGvp
z>P>V7Si22V2Vn)ebb6!Q61Nz)sDF|28-53=6~CU(R324NQ#|Q)7PND;dF#YB+>zq?
zuHSe!I`)=d?c0hvC{L(d(qcRq+Z@O0>iJZ&f@@cKR-4QAA!kWl2XFMXm3vx$mi)>V
z8?4lc#LVVQQg5SwSDTq1V=drRnnXI)6I#SuwOfphBoQuSZI3BLq}%EamsHdpv7hTA
zX&UC3X|1j{nNYVDci(nq_iM^BMx{!x7f=3*x2vr*%|l_tEX-O`lIkA0nq$HKY|xM0
zin)%HYcG{24g@eC5tQ=x6#C#9Dw}o%yk`|58;xZN$Kqya&s0Qedv_>$W;sr6Uf0os
zhgqymHPtvR!3@5eRNg(IzN#ltK3CEy=<IjrN0hNKlX@@%-4yqh{lak370eAx0B0(%
zO*qkYf$XtRhQCKVqcho@FJG>3v#uu}#$7a+5LnO@r)6z)Po``0*>;S4FDl)#6Zf3w
z#<Z~%C@%?`bf%albF`ZG4|^v4rLUEl)@DVH)LN@P-cm-2M=`ne!ckg3XeT&{f^Cew
zs%3%<(j&rH#dldYt6MojD@=PoZ@!9bc3ij?o38hk=!6X%oJ%^#G~n;2IVzjPd8*%3
zH;KZYC@M`L>&A#D@V_c1$T-bWir3;FBD1dX^#}BRtvlM8L|#tb(q5byrJbxjuUEgj
z5<5qP!hc7-?a-!Q=HI|B(H+x2!KIwHy>pN42+m16CsMbP<!0L*O_UL1gbTdaFxo^d
z;4Q0LCi<vwoJwuiw%+d{>jNjt&_Vb`w7WB&m@^!Vohp&w3o)0ulNa84OD;9>7yXcb
zW4$J!c(3{fI?lVe>Q=R-;zg9lsyuQBohL}?d}X;)UZ*|PSujw+k%7ESBV`t0i-^Sn
zziK0Xpps*{_19D!kTe?`3)|4q(hdP%k=^;3%g{FMGpE_{cGM8%cf^gTY#y`oK~I;~
zckW^0%7)7FZLE9L3eK8__ZDxvHrBr}%u!<np9xx4-}0Y!S^L&2my%u$K34tE+s!Mo
zCa4$56<9NCsV<8D-Dn<Hpsk=q)5lt0raCozWZD>gW6ySdp|<21$x+4$uJ>!Vi`O%|
zK^A+lsg>ipa$L*WVMhi$T7Q_=G9#N$@k;I6o2!M_>Gk6Is2#jTm|GU&{tiXBlEMF>
zA0r(_4#k7_{G~C3Kn4+eR@lT-Y9+8@Bv&!kXm65Mg+O*`Vr7$lOP(5y-$>4DR+F`P
z!<!UT4sDODxF|-{M<eu%RUPC$<VVx5W_6kyS>?s$$ghxdIpMmwwqVuN^vdXr%vrgd
za>K5z#*CNjE0vv58WSxBhxLb#|7=j#vP&#5JxsFg#F!1(PgHQTm$;m>TiSe9m1Y&w
zY1(39uWP<d@C)zZ;ACD1!`EmRaaF00lxq=1Hdepq^zff5UgNiGkI;4{<BP-iLBr+{
zhAU5~cHv#5N2R}LKQNf6d^8v?)6#jzJR&yGo;7enbi%?w5yC=oKP$KBZL2Cb;923?
z&((J^-&oJ%FReR8;o$tLpHTK1ZfPs+8%B5`Kq+USjy7@Rhwu>{zv1tb%{YD?M#2~J
zm&D16T-1t=Ae{@8?b&nf^tAMN2Sf^a;)rdWosu&2sZ#uy7UKn+lf*=}o_w^4$}(U2
zwD%3w)!0}f(UBXQ_4U@IlII1HjGv{6F!?kY|EP81V7?$)G>v!7(S$y)wy48MZzbv?
zdX{xs@9T<gs>nF0!;T)H6V00`j8ia-Ua8;WK#0mxO{oL%X7@;+c$rFTRUBh7vxxS+
z%S7&iKIZnm`XYLZZs5QaN{HOr@J#C{b{M|dz+RD4c}`BZ8Q<?zc*p&1qZfxjBnb|y
zFN=;ab-y1~Pc%1XAFbU$FJdw{FM2MiFNrxKcgx@&8uKl4ZR1r=k^HdX?vWe1ndM`x
z$51--s=L=I0?hn4JN08tRP@tTZC%lB^(Lymb6P634r56#s%O^tb|I<00hdF#M$eEX
zC~m8k_s&C2(tf3Nfn86!D}B?LtC;5Tje)8D#-S1~Y6TF=d3L%-wXAApFe(j>q0h=b
zsZba{=|RKa3}b|cW$r>}9EtkP<gt2LbHCDn)<8}ke50-sZIx^F7gI;m^II6w!@Zq@
z6Ls|^&4Y~64x*0wbK9WVx&Hi4X8(3vj&_*vUEgK7w|W@uT<`KeJtsW>F(E;AS4L){
zT3_04q2Cu?l=!ox7C0rJ+ljI<ETF9*x8av)nP7G)W3jW<RV22HwCf(<P`=ZYr@N)e
z9+jnckfYaL>-w6I*<vraZg_^j9-Tm5BD6-G=p`6rX~nAh(W^M4a8zY9lSH1UAQGqI
zR}kJ3>d<oWeU_ot{eA^=u|bslz@WdxT~95lM^T7LeJ^y|(f-|d^?uG8%r4Msl=x8D
zPVWBp0Bj{G{6|=GqisKD1($-IRbs$pX_cY_n$#+PRStg&{%GGJ{e4=+6(4(ExePOy
z3BC>Ds{Jh#3~swnk>~z_aJ<Xda<$bs-Y(1!^cuPyX_{2&`4l~~ax10CI9}YToNus-
zmyJrq^^rMkYZb2cPaAJ5mYO$^Te}GZ!wB^X8mY?oDJ~c#9DI(R%@9ef2=}#vSYOF8
z3<qqWd@++Q)(2zzUIRmNpgfbXQfa9)z`v2L605aS2|fIZz97sxv{zY^@C|V*|6Nc0
zU;yt>&zhczZLTVvj$wV4s!C3J=gCfy>@ois_szgz%u((c?rXw)!C9>wfoyOkV=vZ}
zcNUfG>dI0yUBL655{T=tp14)CM<ze;?-?~j2g4A#rF0T)x31c!rS%s7BI>rua*Ojl
zF+E<?r}k^5xdTs_G}4cjQt@r0MfQ&=F&*KZsn*VBM}<Kuf1{C%FX+e23hgPRG0I>R
zh4jN>qWKopIk8ljs8g(4CduNoDZn%6n7xWc1H0KYiFRvyQvq}1U}^2IWxZ@iba7o;
zTT1uhZhZML;UU!64rb?#+L4^q!gHMSMEx#Nce;F|dWXnVJ(pN24pA24h8uCoSCnUD
zx%!C)NA--*$=c7gI!IBbpHMd`mGrv?+L%G&1${P72eVYmnvkSZf`6z|(Fx?6i~`w9
z$pTCb*5C9u21ZYcWvR;PLCjfthqPTu^*VYk*Bm$`nw&}5%6O*ztJW&B4Ac9VL6IT<
zH`Z;^FWenHQ^i~9x#E4@qV_SpW7@M?n%Zx27xs*4d(?Kf#I|Q`%Yvr(=2MM2?TXHN
z_Q&oXE~@QBQ$)`g-a_>>jD@Ja`=iiPDJR^aQ2QffAPYwo8*3A<;L^p_s$i`zh6_!N
zX<J3jieQ$^=$73%rElK|@=3jsx|<nTo!0~hd<Z5GztdvB-b^`L<ZF5+V4KB0%zf@v
z{WC5}mh*LY88mx(y6n*bKMSYlVspqwLdu|=v~{z`kTd#R2DT7y(ruL=xgA^sg^aVi
zL0uKwdZ^E~_DV@yBdTjm$B+D(d2tm_8YWfjtqJTsJ`mm2P+eI2YZqR!R5;l2+rU?p
zA6g>C^A3tfqwIkv&#9~QytMIX9Qdx{w#h^M3nf<lm5ep;VV))QiJ}#ww2q)J;>PK8
zvdkDqQ3T8i?RN7B`X^3}^q$MBX=lcNrxx{{R9<!W4o&pon5Vd17_H?|$Lc11bZ~Gx
zVV6UDp?}NslILMsLjCEASK>luF!3(?NB5cDlPwzxUw={OUaHwrr=PzvC$<7tUs3oX
z$tOvgzofafA-8T@i*xsncI#?~3b$URVy)~bm>r*$H=z>bmGWNPXY@HrvbL}GR2?_^
zXH<bC5}i)k#fqnorR}5}V;^$Q4!$BNnb);;!QZ;0+cI6~muX*6_c)^@H?1QhuFQ3I
zsOAsvHZsbZTsHHye}a9x$K4=HpITeA!<6Bd-LDy((Hrl<3SS>Ejv%eAt5HY_X$b@u
zg{OE-|L(kt2@6vQ<>}4K3SKAeOAarNY>4^s@wIoXUkWcjt1P+^-Tf>3b#u`Vt1L=0
z9&fCZrfgtu5_=N%QDyi<ycL~GZ^T!sQZTvt&uwoT+2PAY+Ugn9`Q$}}KAZ<(587SA
zlMUeAw3;Yq@v|kVyvvf4n6pN+ywJhpoR6v{0~bm6eZEck61pv5*XYC`mM6nB-#jSb
zROE=U0oHfS=Z{=5ewvem`ZWiOs-Q~HS>k+kCq>2^kr^F3K52S+S;OXn;~y`6smMx6
zoAWa6#oLcH*~dS2zP*(3sOyCIa+_<GOPZiCl%v962-`8KcrvC|)+gUfxn%69Z%4jC
znrRql?`<2Y+kyR#>!<7KF__<UPUtvL58$%Zs|Z(UuW4&ZSJV~q1FCvd9c7*AJJ%GK
z3|6p8g>Es+9=SX4pw9&F556&8t`0J*a@WDpIitf}?^$1Vrg~=#JIXqXjmLxE=ff~W
zDLj_UTNc<6oD=?OduB&IIw$+v@NX^Ok7bSdu=$NRY3<jM8Be~3WYMbvyHK6?D?b(R
zE5GzsN^pShNt{1nIOYqwpB%66Z!%SHEIm*+)$F(>%`jQ}p4KSlL%mkUCCVjy4JK5*
zNnRyuSH4ism!B6+5MC8Ki=VT%_hJS+wY6*xSd(;Dk}p%Q=$e~&+0ONz5xy{zH-<j!
zz3qzOQ^#`0Dtt}6Py4O%>vA==xMx^m7|#epqxmPe{*sZ>-2TP25A)u9t4Mi}NKe_G
z8IoW7y)Wf<oGR|xyBVJreisx6)XcAKtbA7;+%}iHQM6eQ&$AyysV5N&b$az*8+~L+
z$O-r>)L8}*mKHXq=3{khDFK8K$_GZ8?rfb?IE;cQZ&TOf^6&(7lk}NzhG3fLjrfXq
zm~^N51$ibjkabCW3?*7CTz96aq3w9rgT9#2g}y<qm5vo2(*qs`az~bW`H!42Qr~^9
z%}g`C=?KFq)NRVeQiXz{yunAcjRoIgEYG(49G-?rch5bY)B0spGAe0L^862P)7F0D
z<!&t<DEF>hUpKq;0eGr$N5hV8vFr$G1|^$t6+eo&k*uW^&M-3cF?BL|YqZ#eXdZ6%
z(SXBjW5(%q=uKq)LJOySBz#5ZDbtmQKu@Gpe~~4K1l(urcf9+GW7KJSUs*(|9)31~
zPfcd-Grc$LwBLh(5RXeXYpoZ%lz8`y^zvEk<L#qz54KA-Ut)Q}%EgF9-k^M_NKnN~
zm$H-Ey&9jEYiDoxtd-GKu)c(p!~3-7gEIBjr_yhwxzq~x>aJ3UqWP6B9abDKj!FM8
zcAUr?ygS%RI!?P!vm?6^pJ~0(N1J*XJ!eg24zRRYOSDPUAlf1N1#K_VPHZpQ4Yfl4
zT)a*CP`OkYA^8k`F+(Y}lw9Z8aaM`v5XLYTG7r+csA)8F<}7_jvpbGPUZ+OR^XRa@
zXgAlj();m9d#}BoQ#|5^#aewZEY>eGdS^IU2ala8CrR!Je)KiA88j`Z3#dpcd{|&!
zDyukIIx;sb(=^L9H@E11#mu^<Mpc7n-Q4=x*7lyy{rd+d@Q9Ke#XCiq0)u*o7ZFDh
z90?tibM)cNW_k}ji!ohi9@&I&l#<Duu1{rrC%jfQ%Gt{E=mhLI%y#8wNv~+HI9(Pi
ze<1$CT{hq^7>1su6R3BL!6tvfUncSCJ*G>Xz70otA9H`=Fw@r8$#nPyUoAhGSDV`p
zhbT*J6S{G*iONu>vj*!aZ4(4@ul3#Rc;2M0URAvO$MEmWJgt&P#XEAMzXg7=NGE1r
zEzqj`)L_(X*s!ox(SYsD>t}NHgpu+!7%N;H+6luZ8fm-fT&2y_iPBqQ#4y~aXQzL~
z0HYV6vr}ucP6+E5GnF=%tj6=u9g5@9LFqZgL{*+LOI|4n7kviL7W!}(2tyQF#E0}<
zx;Gd?susmWdyT%oWr1_OSEJuM?<>RFY@Hn2JY0Q_c&+o;<od!<Z4qFwS~p4W5;K+1
zA{pR4<bC1LyBST*H4`eI7wTk<{HFYlE!tIlFE=35Kb@0K%DS98yJAf9>GoAE-x{tr
zf&6NJ7x&blr;>@;joV4A(NfUgY8R5&KnAi|nau~Q<t8Y-W*u#sla5q(oM9~MFnKp-
z0cHVi5}F|W${jm!pH1P@M1`Ux{_VaKZKaKeTK4tx6w|Z{S>6mhWf`HGyp3sQHrhGZ
zk1^hHf>$8M>-~rn|B9fcqnrKb`Vac~c!`~Ktc%Pa8qL?*DsSh0;COPs_u^YlSF4LB
zW!WVMzrOaS?bD7dhm2Jp+3z<bZ%Lk)cI(Ia+QQDbZboN(mxApkPEe-eB578-GJR*G
zMDwS1Zymy{8;wg1HXBc{*l3e$mtZr`cnkHWs!X;+wH#A|Ixc(0pU^96Sz2$}u(Nes
zr>t#uV?o{2#sw|5Euqb3o#nzK1SvDg;D%wNp@-2nW4YDm5$y13^RwokkIWt)8oYnP
z>WO)wl!=*P&qFgtJ#@KhvDxUmu0G{}Y{kIpuEXu~8n=|k=Caa#-oJc4=i!>@C$Cq(
zuS{`FN{RQ3+aFI$zLnWub-QyzUsT@(cI#lJI+$omA7?VgHpjl+uF=8Lb&?0qRqB-O
z_}=lY-8YLpCRpP@gM&I_@nOnH$sEDlfeGE~+hvW-mGa_krH5*?8=dMVmb({!D=sff
zs&VX$kXh;MG}ba&z>K2@8e;7WyfY?X=HOO{m!F$sH~CE1lgPx`r)Sqre-zO(IU;1g
zm#=M-(OO0U_B(HS2c|utV@AW&ob<$5G2`#rTwQ(9^hVK>pjW!F*xSpm)!v!>^g-gJ
z4AV-Jj^f^T{jtJ3*x8I4L!r5$Q@Gc9|I?#{W9&i;Cismq@ECB}?8)?Mav$T=Y5l=4
zgLrL_$xa;5>0j55u8S!TD!P$f`Hh{iDQ{z;M-DloIioFiS@~e=Ia#x=%zCT&SJn<)
zl--<wq=;3^j%|Cicf*$JOAgQ7vTVwxZCm!QDPA-<M|Zma_)zb=4z)&IgpE9Qt6LMQ
z>3bFXdrgAHokyo89m+p^=6uC1#q9}~Po8{z%=%2u_3_URe0oxBUmx0P**8J{op#=2
zn@y}Ucf|QX?+LQ0=sD-+d(Z5i_%i5L;O)Sz{x?S?I^QzypvB6k4{U9(s6SVFBAfbU
zV~Xz^+n3W`suB*qvra5~je9Lj`k3X}G)2uf{ovqlYh~7LwaM$&q};{RHb?AN9Xzq~
z=vudxsC8Bw@T<5B@@F?iO`W_YK;kgNKt`C%KiXhiT3q5@x;Xo{H>tNCoJ1Y!KXmGp
z+g0b=!8e^Q%g=wgvhCsKH$O6_Rjg>A%H1jNCfOK%u>0)(aHLt#*wD$5jJf<por~tr
zUKu`bw6AZ$NJk$}w<^oS%vacR!UbI~8YWbn%;#m4q<u=FztMmDYZ5Mz7DsqFIbP>O
zSeCM)pOdO%W53R^+5DpU36J1O4$D%vI)h)zDcE~uYu|eD>W3?4Egm(eJ<@w}XBZ=7
zwwITwKCYcTqh4C{xL{7vlON&lGVd%0GMD})KXdrjnCMB5&7*^3`d(a1*!ICO=U77p
zr$w=rh|$B@?-+4x<Tjr-{!1si%^@$vExk0~ZTg*%Q6qajYKIfN6wV_IU2um)Gy4{|
zmex>9(zB0#5vOfV4R{~@c1pt3_{(wPIM1Y)pSI@O*UjR8p*^s199Cq%)&7Q0+_W<*
zBi52um#(G#(!T!03X4UL=k`bSOc^!d@L01kZGK&jN_w^;q))eILGH-x6F;_nN{LPd
z61N|pb@Bf7H+Q@q`NeL3rJFM8$Eb4Ss?*gi9ewgE+KY@S_EPW7!B%0J;rFK{E|ji1
zxY~Kyv$?OPlm<4t8#~uIIoRg0=&G=O`{wJF&kJOk1*tai2V=~i?MOKJmhkG-)Aujk
z-yTm6OZ%F6wz##HHh|S(`Z!Ms3=bK*KJ4yt#@@l*^bNCDBrTn>XvrL#8T+Tc4%-&|
zByhEVrf-!i(NI?%#;I)#%j|nO=~-|>#q&+z_fxhUHNCj-=BK;p$BSRs#^c_MdOx$E
zsOOV7Qc@=0q;t<^pF3@&WQ=ms<>^17(&sj>2><2ty7q;k5u3(wz5N{<O$T(IptcF8
zvyb(!Zl_fC=gKp4Kc>BUasTC=0}toLj(M^5)vbi`<mfN6vYhf3mqk@KwQNU^9a*)a
zZ>?WsYcO%%!QF2A`I}-_E?+Ww!OK}G(_T+m5oQzoAb>qGz}MRQi1}K@o~E*#!|~#)
zd#(s#lj7oT;*SIzYr6FK4)#%D4CA#|q9l2E>Zq(O4Yx#xFw1lf+ua>Uk4&54Jtuiy
z^nC0h{Z;2TB&=JwlDRl|nu~vj^<H`gcC>=qpH-2bF+F`>!JLBXuglXyKTb~`5&Q67
zQ*_45{Rx-f*e5;wa;PM>{#&!M)wSbM_XX9W;U??W?YqDB;8e@0to6=&R&5Pi6||^(
z?%gQkh`E#Qhph-H^og;)VQ=9x+}#?Ln;rNv?0(ewD<}JIIVX&Kx%~X%lg3x-9*ANZ
zo@K<}O*6_C6|62fR#Vx@5}m-+Ti8rEzNUQ3qF+w0#4Y?~;p$bB)?ZvRcKNyapF-CU
z-((d=%N#u3#c1zo9#a*Nc_AU?X?a|3(%YnysZ-O9K5<`<er}WSBKb_}`fryCN7wD@
zs^NGjE@FnuKMO0h%qFQ0st%TKOpCCZ&RRcuv(Kvg^ZcSJBibkS1uq=8Zp<SuH;ZE0
zRMTn0(o9@Ct&(Ov9(~jJq}PdsPd|UlNou^DeNN}D%QN|l+*f)ZTPh~^EFb6_*v!@9
zqXie$DC>(;Yc{>yv1Rj()xC4N!oEywow;)H@p)boi#*;N$Dv9)ODY{J-D?u7W>yuH
z*MCoXnRV;dqY-aW=?n627j))4%ryVjk|8aiH~94gacYH6rD)|hna5xpKUs0pDfAa8
zxoy}jC#&&W*6!GVUBm(ry*B(qh<otSkoN)AR&A;^IH_&wFp(<!i{lf|hbyj9&uKl^
z%SYwp+;ckTa;xgu_}7^)#BX*KA7#G~w@P@JSfaUn>Y%?qAr!YNY0asn<Yk}dSxpQb
z`7CI}jE)(zgO0eIuoyw=Yi-Yo&F`prTJyL(ztXEr{pI0P;-hs*H?uLN<Vw>jUgg=s
zg9ZGm#hn)h4)v{Nm&&R5BJ?T#rOx!ez53;IRR@1PFmJ*NC)DV;m2K;_7bQ-&neqih
zHFMmHklLUc8(WnRZi#iSvl5+<+4`LFXv5Vlm+!}g=5>AQzkdGQ@ds&djpOoTWv_Ee
zYWa4OwS%*<EUjg@4*Yn#n$QJP;wG$`s<(Q_YR4(-y>z|O#~O@I@i2DMw`Aa{4gGl$
z<pDL53U1_NR$145&+&ebN+acbEBd9luKIfS!oJ~6+p1T!bPi<q|H?k5x<M&dCv-pS
zSjL+p7~t=*zOqhrFgEHAYm(9RDAMNI6$3L;ChJYg4LLvdX)t$eg7Z4?{awD@PX8u-
zN?H5sX)nG%GP?2nDfY+aoL^(tUuwD2`}Tgq^5=!Gf^seT`ouG(FHt)QA(&!Jw$a5=
z5n-FchDCf_%vrc<;xg~IZUrOSyxV<TM{*n<;gt;rg_z1q4f86Eiw>1ds`M}H%wC^w
zRuWveA|F#Te;{74y7NTEU>T?BeUFczQ5A!uDT)R@4_xL?>^af640U~~>fnP-CGLEK
z-TofSf2=W{A3A+rM0hwkI3(o$1ZUrR$^(H9)7tmD!$kIjZ{Lz;#ht&Gbk{Ah?yK*+
zg6GlkJJP%pl#hI({l6x3QKW<N?aDeutjJP)S$mnwn1BhRPKI7sShjTil+T{`>_$7#
zT)VyY`qSL|s1MtID}7sC-uS)zbne`uDYfZ!$z{_DR_3nyF})<b(@157naLTbUt1$<
zyUlJ9d;m|Wx+-rBqJ+zMOFL$jUvD~Py>{Kh-TlE`TII}@Ax3Nc*BUK&7sZRv3V#+R
zowRUF32QZ<M*QNtH+YmzcH!Z;Bd`7Ao<D1k+xTJnhn%-9@5;VS{8Ihw!`%~aw3>CW
ztI65uweqF19=RiFlG$dLubvkIbEg?DoHp6beVbXRg^$B|XO&xn(|FyK!EMcJ8)%KY
z$`!eSVxPvKCPtNO>8A3}#RmB+YNiN$)C}H<hT3}9zGr-APWV8DFh#BuQ-p5@-wZ6O
zqUWDc4ovIb)jjPu`T_0J!DH9^tQt4-$n0fvB-0*Gy%V`;;u?#mvZcg%K7Qk5+Ne_c
z$CB((8Kf7q2L(@^6ZfVSeIH*m<MXV0->%rb{m_~w-8%SPv{bbP$3k;)m-Sv+*xEh!
zdKEdiq-?r{?;Xcc_W8C~Z3gWz4g>?SLfzBYG`^zq$EfVq@)7`iGp{<Uq$lrQ&cfov
zwsh$i)f!P>KW1POAHz@K?G>^IT?WzOdht$qE$={MLB&^j)ad#1zK#huxNhMPHfQ0H
zh}$8tQML2ZXIzRfjSLT2XZ%%OMS19!=MsvU*AN8WM%<e|Io>wjGwJRd>%_E7(?Zvj
z=DUthmStx)O=}+8@o`W@ct_b!6w+9_qxC0QJoLR3IU-_G;3bc6k1H+-HdymIGgAvo
zT?rZ`H0_hL=~NvqMwRbrM0YLdtZA9i5K%k2?tWuR*A>Bi>C(X)+%Wbib{lw4O2S^k
z<8l23V+TL>$JZZd+^M~D3^8<s?RM>lPMabg%@$6Knb0sjc-Hb6+olgp_Hq3{n1?%Q
z+3$RwqSN70yQ;~qa_5(eZ)d%I|28#wQP$29{`b5$*W>t^bL$LR2V1-P)+nz~*g7hm
zh1$J3c7~zO*nrRDI)j{j;@r%ga%{saz8m56%~_qa;n*iabo<#VMtNBE^2S-MFI!X%
zr|JjlHrLS`y*jQ83=@(iBgBQmetsbTy6BMjz9>pS5$7liWFH2|ePKkdvtYyx8+RjT
zhjoF;A&-KB#!jA~47CY2pIjAU=VHb@Ln9kG8SNq32)lY3J3clP6d%ku@Zn?fwlC)j
zhLvV!?@Vh-+x+8N#e)X(wiP{%;M?XB?IVnx+L<~OLmS&s?zg;2-sYamT{bwbwv9I(
zu4|@~OPf#qjM*SE?p@yA+2Yc)y<uykcS}=SOh-s(WV@v0P<vJH6Twxn|KJp%iD;u_
zhj^SQT<9yxkuH)82ETG&4;E^bnN7D481xv{*&g-?81dFK(A&iKhTq=6lECy48?Dg#
z#ri9a#^`#IGgPL7<GAyCPPDjJM;0F_sw^K@yRVv4`k?4U@%@V7b>x<;_M_eQ++DJp
znDe;bu=}taQYbUh_@<em`5p5_i|ZCdQzQK^^z+&mY0Xp{;ysnO_zs`NjT!jTi|NVk
zDeRLBT<4r+iwCZ8jtVj)&*brPUzt?utyrXrRL00WWaf%q<xAxe`FW|8d@1@c*+ZvU
zr%&6Ou~Og8<g{6ld7;IxRt#Hzy9d^@O~2?ZVan+fwVTP8aQUhLS%erZ^x}N#Y3bPB
zF|2EOH=)O@C$c-LJEHeUf5N~IHkx-t$dUxeEahhMR+*8aO*IBhz_?@lu+6xa;9a!E
zcysWs;C%v#*bm-kizTEGJ`kSb#kfGc0=E%Yk1fOO#H3&v(DP7+>KUqd6+zvlT%~v*
zA1TMl`7(lhoxEFq1R$pnm4%Ag%5ADE>J0FBuoANkI}Upadl8p_-%H3P+#*~8?;q_W
zo*_CCUlXnpju0B~DEtEOt{4GF#T~<L!$hI?0?Zsw^%_-|@{F=UQ73PbS;{~TL=r7I
zE7>ktE;%D<m-tIBNIRrcWnW~5^1I-z8I&?ic^;s<YyfC8R5ezuuU-jIdJ+J7ObY56
zK#2)JUq|0YZ$Ym>UqY9od(m;|`Dkaf8QK)B4^U;C&=m0Y)?4%x%qh$X%s$L@OgCmc
z_8#^J_B_@L%f$8p-7C<c;H{^_0NE=_EmNhawx}klB30{DCjgp_Il$;yi=w0N18f))
zc=sm^a|Z*S^TSTVj>LYz44}=?*8$R9Fu+6F1n_S>RPO)+O_k!U;+vvc(XFUa+*U*>
z`~VV7nnJ1|DXB^vc>l{tIb1m&V9+=!Co5}|YgL6REw!t9Ho(dG4p69epjuD|(C(Pi
zK;D&@<(Q`!Q|xst9ycAledhqYwFhgAt-^rc`M?-ro`ZMO&I8mnA9b9{Pt~h@qui^E
zQZ7@*DwRq<@cv4)3a{P{FoGtdo}va(BY<VrqGzK6(f;TWXa>3rbp{~(Fj4IQ8}Bwi
zh+GViBS!#iACOm8KL9v!*#P~l4qzV1h7hRi0Y(!RU>oHDM4~Kkrvym*9-#2t0%eoI
zTUuP;6$L<3LIcz*bAWo}0X7?e*aT6$_(PaW0)RTz0+4$kG8)9e!vnOZR)9)Y3Xr?{
zK>aR&V3Yx{k08!m13+!+2OE0`V@eO;icJEDb@KqiUleK?K)O5z{Phb!Gjax0xPWRU
zsFMxw0b|uy0ixt1;I%qX8{%SN0H&5NK(jmy>g)!{LsJ2o+%WJq9t|L^l>m&v8|p*A
zMhDc_K;4%>pXUH|sT#aPKN7sxwhnBo0Y=hR;FX<#-%Owr1@MDdV1>XEU)6B{+vown
zU#bQePBwsYAV5wE062Qf06OL-fO8oLaGT^nl2(8&RRHY#44`cO0uY>b1HNwoj$96)
zQvvQufQ^j-VwD*{y@J?frU0?Z3*hKkgEK0?*CGs|yFnz<0f6LI0krD^w&@3Uf@qdm
z;D0SRR|9k}0zh3e1@E~{0XR>)0J_t5fSfb~oJ|IAUXBL#q79)a6#-ndGGMhburCJM
zGy~i#I`Engz<q)UP2<5kUlG7o<A5AY5Zwr3SrOos46GCj?DQH$ND=U6C%DH2^5a1{
zQ-IuP0`RwpAX3l(J4_5Ps5*u;&j#Mf1#&lm^Da=Q58T57^fHKs1+lZt0dAf<klGvk
zxq)MIU^j?`#u&m!gQ$xGKtG5@)(heXMvE9&oD6sx0UD4Bg2817xDx<p7;6wqt_4`A
z26(3daOH#eg!o{3L*=ajLKxg^0lu3axC&9-a6>W+0d<IZCI)C{5St4N?!b)*_>n-o
zp@2P<Ahx7n11~oM-62jL5B!H{W-4$^3^p0qE5Mlw{Fi|Z9z#qy;t&liz;=V!c@WJE
z4~|d(-3yil2jBx~$cJb_3_skEHQ{y4kQ7i(XkSDQ$WaL-83a5ffD1f_WmTXh;(^2n
zVj6KsKUfFeff9*_5Zweo>cOE+4E{qDOIV9ABpGF>E)j6UfLgF^i$Q5vUII3VT8KzQ
z1`($NJjDe5f*#WW)`X~qxPR~xgLf!}z&>2C@j-pbkfabx5z>VpG|&e(WCv&!+z?0f
ze`&4;QbDvrXhn!s2oV<%d#V8qEZ7Ku8x8Q)8A6DI8j}GvNC$qPWnoF=^OGm!1+9T>
zKT9B}h{!p#mGXDBAvz%Ru4aRFP=F0u8}b7~-Vjx&8wThFW!G>({EO%Z{SP&S|6y6^
zQB8eFg9J3u{)ZjOz;_rx2}%UfB4Lz4s*o>YJE#eaPwJ4zV0qjSZ)h_Zh42s22O^ci
zNP*rKf(Q_TBS;V4ML0lPL(d}drHMnx5tf27Lj*w>#iSuVP*xZLaD#p)0ZE`NPy*xw
z(J_$^jB6N0aD!fi)Dg)bRp><+H-vv!5NSye4-+CNLZ8D8*3f7U^+qB`V>{R`pjDv_
z;WfBHt7%#VVsEG&Y>65zVB3c>L*F4jLn2%w8xli^UEr~DNDt{hECuO9`@txNJ^-+6
zV1t?<bg|$*v^s1zP<z2pTVN0MF|a+s7=pHd^r1FTC%D&G4aN%mATa~4LT*q;XnEL5
zA>t*(y#&B(VAC4X18R=+5*lAZt7$wfACd+38W7JDT9^ZRSS~o%^jxs7LZVXBVl-Mp
z+rk)t`a#drfDEw3!4Vy{DC7epbr5I=+Xp;Cq7KdlV4oub|M?*Lq(iX=TPV~BY7MEt
zSb%%TAELc#0Uas-;6?{}YJ)3~Bf=M28T@DpV2iYLq`!bsujw<O)uAlNhV+!0UgT$w
z3FGwl8;m2kA>Ke@N#h;FM;ebHwV>Xx4Iv+-)g$c&^7z>g{5(QRBDD~!{qa9SS0e}9
z|6V?L{PUjX&L1iK_nyJnL%1T`{uOno7mOD8AKBpX&(a!MH4*;v4srzV|GN(&!{66k
z)4M?_G@pM*|M&O)D2s;D?@RnVM)XB|uJJPD_p=0Izn}VRju6lOzT6*=H5w!3f8Ntj
z*X)1ZL5_a%hOq)`LyTx-T!i-1e4v-0^qPKBLr=5Ul>7Z2>4^{y8h^v`KY1dh|9lNf
z@yE|!m4NN;XS^fj{~Yh|DpD7bA0BH;B9{6ir=RzcQt(_8p_({F?r8j^*=y)%s3Lm}
z6(rVSzoXeSa=`jO+bz`TC#9d9G)J2KpXG$tVZEPJHAjdo;TiO+rpH8qd8}k;d><U@
z1>x9&Aoapt6Pd-Lhh}teoP#|+8uYGkHVV<lS>QkHVew$x5`tOx02mt(<X<?x(ZGE;
zGl%yf*7g8cy)*$#*6yKl0>BB*v6)~)=7AW1pbN2j;VJ=AL4)2}8=wwbfSEgOC;}mt
zFvKW^^Yw1fPeX)eCRh{LfVsIDSPM}BA0e2}w*gL_Ko$kS<<$rEod9CC6DR@K8)`tk
z5v-*$!CI&Y9I-(uh?cDj?n7j01}FoOu94AB0t}2FlEM=3umZdoKoTL45~4^qfvput
zq#VL3wgqbh2e4Uz6$yeI-3iLnffYq7C=ZddiNJ1JfC3Zzhmu09@ID}KFVI^6MmQ8G
z0q+|Ctsy2e#P)_Zfh!q^Vl4tRbbyT3U|s4twE8gx=Wxwc2Xx6FT7^St;JVHZtl->$
z-gcloL@S3Y%j%)kcrBm@p5F%RAO~>Y5$u_uoeRPJ8bGHR*calL(*TWOV3p+qR@4yL
zodj&s30C0g>Qu0<&jw{Vz<(w~c*!GxbxncwWxy-7fJY%%!Ipp$-5_45pbW%FcL4T<
zwiN^K)q-_o<&f<mCOOo{0hI6n<&1!=5V5%(NYe&N_5r&<bY(ptr7gH`H{?Z#tzHXO
zqxnGVYEViDBw>I%jzA7OP|{$?!e}5doV#*?J~CiA5{M!_Pzt$E9`ZmZSR>Z}548hM
zaGeR)ClHTYA9%nJ#4ao~2x6ogcmPHwj7?}?BcQJVC=1uG5LFw_8N@?v2ev&pqk%IB
zxGseAC^!og1FgD1ywn4|Auch@f8c;j+CXv^h(0DbLgvKK>*67vaOExo83q^;`k(}y
z(LnrdxX!|XSb}X2t_qR4B$ONCe8cF17~)9e^Z^<WlN+w8;4BK6#V7y=conV#fdZg~
zBQs{0ZGrPr)=&)U0gjLZ<Olm$SZ)x^#^4$jT2%}*#RG}pN{&2qANIp4zy;=L+QDBh
zuozs8!<ihM!D;$sWDTbt>T_V!!<jI`3EqJj0VtPd4h1a&&!L`h76~<_1O1T|GHk8<
zp;pobR`Jas9-xol{1C2AVGjeXs|BdR-cGYlK~^sKp%#g(N#K79D5U_hLNs;gIXDxB
zyy0pR&b<+@!ia!<P#@q6TP+z#qYvWT9P}VCE?_PKBAr9I`a!E{23zNlFOhXUv=~H`
zr-PbUAQQ|Fz{~)&DV*!VnLgC__j7ybArz300b*DmYz80(jX;d*gDXT}9|@oj`=mw?
zW9`5iNR9ymBw_x;m(WYdx)%r5rf{_c{Rp!ca5X^&_u#r5S{D3_KRBa<y0GU)Rz;A%
z2uKK5BW&Q==KtaY`*9}F+#a-I8_>RBjsy>^$OEmW6|}8-;FVr*2C>bd2ch;*7U*we
z4i7zptkj^_aG(XrfP4eMX3(Dy(|r)s)f#HOwxEq8eDnb&HIN17Zd!)=3K)@SP!2{f
z^oAw)GX*x%27Hn66v-dJIDmN#xYpnU3ca9qC-~0=R)_Iy0Q7JHSte)Dm+OJLGEk-i
z^o_+JGnEVau2x_J#P1B?0}HU34UHTGAP-zkNdD2bV2sJZF&r7;$kq+GKt6mBTd-Bb
zx^T>}1Ckm7xlt&XjcNy$Z5Zk$`T->@uow$yV*^Urfu7PF#0>%X8?KgMD?!$paFqeq
zqp+WWeG$x~2!|{V#|t=iIRLxC%&ZP*r*JH80sUSX=%vfSz6tnK2rQ>PWH*>)(+9^m
zzy-<Vz?=-!fej=FKg|fVhbuuL-~`u33LqowSq%YobFe{c7y&8aXe<V0VXLeI{dFVI
z9coSjnj`*!<yfFDj5267WP^R?&)7%iN^rgjTdo@Lf}<$x17LrNWH8|D0!r8ic(sEK
zW(|2jG7Kn52a>@IA&h1uvf!8ivxKlOM`E7@?!kT-#w5(M!5Gxcp`fK<4g$&qkKy_p
zu1I0~f>{n6s0+sn1E4W7mqEsDDd55Z-s}W2!5K>{kgFeDQ~s9*+CV0x#!o%ufI934
z_+XFZ6p<_n%%LFJJlH3|jQ}KtGV21)khcyvhgnHjzkg`7Vgqi7jBtd2tA99hz@7r?
z1Iw`iC()2qk&FbiDs25QCkA_a*h-N62GkyAT{P=xEQmD;kP`L<P*#|gPyu_xJR*``
zgMEh@T!-s;n2m*bQm7N0fx_%0)Ca~a91%2gU1X%xWbEMl5AuVO!*x9}4#BYq)`1p)
zc_n1^uUXS0**_>PoCm|+oeIjp87$1bLcc>vxWI$3hlR2rJp#-}K)T524Q>qGLt+5V
zm*Gf*%!i>T;M^GIfMBi-8T+NcuW+1zF$yzRnidGhy`MT$ftO(1!t599D_~^7Y!xhl
zWNaaAs4e6GwfmVRfzb#1U^qjE9FVLk)DudHWNDC=19gI$ARkQz3XU;w6ob5AJ_xoU
z*v}(*Wk>^R0RJPFf^!#0Pm}dP@)&T1qFHSrIV?C*Ama)JSOlR1dm$uigZxM01JXir
zdN7X&r9{>{2u+w}gYkuU`KSCpbwNA@V;`<wU~dPlhUkLCCA1iln?vFOazyg(aP)@y
zAh~#?*V1I^e)58m32XlRAkm4mI1NqBdL8yj2z?~Kk6cHtL0$-bL_26bcm%Bn^M_DM
zBxi-pQ;;)6lAjtwIgqtB;u|Ew;7kqDMp}|4Zedwi6QKc1BDo|@gd?p6_8yv?88Tu(
zc{OF>npATP$F4tq5cv^#kyZ$$gCAr~{m0VKa!?j{_a_Hr?To|}lwU&u9wD|y+8WZb
z5esSj19j6xDN++sL+p*%1IhyL{P}}Oi13DX`y*#WQjO&_b>UTHL#&EO_j_w=WQWv-
zhz|MWuOvVyXej^HUc*DP|En^8zJ~DBX!Ga0|LPp6^{?*!^>vL@8c+OnsegABsR`*o
zn`-F&zQ%td*YE59pFI9|vTAhu<NmLu`}g*QT#%SW_`&~wP3Qm5)&CwUf0yGQHT>_8
z|L=1CyXF3!ZU5-QpSt|4^YiHM^7}~>sqvHQzozh$3q1N+?kBCkL*-|^zdk}}|DM-h
zm;3!yr1bBvXhxxb?msos{;@RDcmDa>e|G*?()?$X{$4GOt^T>RM%TaE|8wcTx&mqc
z)%E}GYZ@Ii`~P10|E&MtrT@RLy@vPS<@o<yPGcju|8r>#w?FRxT<U+~T*L8yg7)ue
z{yXh|U+!NW|CQDLs;ovIxc~DU?61me<oc_7@S5iSU+w=~=l8t-eDD9I^M8)t|6T16
z%l$nne>W1~Sb~fvf28|IPyX@Ve}49R-~ayjKd1Qnntzn}=h^SeAjiMw4Uhl$`B@Vl
z{qY_=)71XwJOBEMrVg_I=W;)J{`0;6yI1}w`OmZeyHfof?oihMsgK5Ania3+yJoo8
z<PDHK0L(-n&o&_|ZcR2#6A5s=jpPx2(m<#p>uw|;3bS3BZ-tONF&pFsVZIn<+u(N$
znr8=*Oc11~iCvfxL1fY7zTkI3NM=-bD6a~i*|Y@tZwK&wqzm{?(;a;G=>@)*^hNoB
z%@;g>I})6Y0N0&Be%ut)(HY9F!Tcor&aDM}15pXSxhw$B&}M-rX+Nve!1KQ;;Q7=<
z@KkTS`W1NAHdY<2zNfwko(aCBKC3>VKBhhbwxi(ajQS$D9<5FQ-+JbQJoW&PoDTSm
z08hToL2Up}3tt217pVZ_q8Y_UVbD5gW3(;W1?>(tH?%9-0X+H5L=(_rfK5??dJoVt
z4uNNW$ABFBAoxDvCEzh1JWEYecdOD>_f-c}D^ycd-YP2<OGQ$lRC4fqHXa<?f~U44
zRclqpR8LetR9uw_c=mc5c%rxvSaKL@E_h-&9#w^sqm0mA=qc!B=&k5O=riam=v(M}
z;ED7H;Qw9l?E5Y7cLy9lM!!VAM`xic(cNeTnuQsL2?lTJZwF7l-vxLjpD^DsIhZU=
zDkcUzS-u40ff+<60YnyE^n28FREK)2S_<S;D-VGu)Q>3`idcEDyi0aj7Ahmks-*9v
zG16FRinLBjkxh_YlJ&`^%d_OciUP#~B}aKgWeA=J4g%Ji37)|Z$0T9&uxkNUM>BYC
z-3&anZUvBHIM@vAS%3sX!DeE9#du;mfXuVd2Iv}840tX(3OqIK2EHo}0@nW(6^H6Y
zxuUnBKcR`3=|K8+j3ahA_6qg`wgg)P5RkI5FR+KOAy^^i3iy@Z=Kzu968KJHn)<Pd
zryQ+3t7wq>04xfs?5LC=y(;mN^oSG1m&Iqr_r>4ESjkk$bBT%cv2>h_Cwn2^teBv5
z0Pj2i3=`Bb^h}H~pnnW!gMW%=64nr&1B8)!LLS&}B^VQ8@z(eo0PAZOHU`5+k3=6q
zRj9qx4^+mg7s_c$va(E(tVmTf1H_$Ql$FYvs$SJK^;DDrcq<_b^Avj=HyckQBoY<?
zgrlRRCXzAPoor05C;du7lU5U75c2R<xNq3~7zcEf`l!lG`AF^}YnK#>GY6AI@xp8Y
zR<N3n=GXI-ykP!aK2EStKogb-3q}0F0LevZyUa&%SXr*JM{Pns!u-Gra5O?c;S7;U
zdPFiIk0-m53FJOf25BMb1Hiu-OZXLk7Z-;u#ZWP$&<jwz)wcjP4Mw?KA(qF<SIUj$
z5}Ap7lRQCQEjLtrP(&-Qs`h|4AuggbF)7%YI1HXe*g$+uIz#rR6j5eSk5P|PpHj=I
zKd6(a&nZTfUF7wk>v@4cf%C#HMxRiRQ28sn<;~J}Vv|7|@E*@3!7uz`o*!>EcRh#B
z*}?JS)^ekG>AWC*0soL7Mi??km+X>mksXxNl*KA{)JXJtj4tjZeiY$6VF@vlluK?U
z?<a33e<eR9Ymu*$E|MmbCJ<}zsW>s#2ur}&0d%NR^(mFDa)lyP;iQO>)8&J*W%3@m
zk7Absqij`Xsg9@*qAV~DxD5Phf{O5+s3Lh$mQb^_e6>5ZcWHy)0;1ib&88jFdQBZg
z-9vdynngH+n}$t9Pf(XDM#^1fITH3DNwipy!6R`Wv8~vf*dN(X*&*!Ofx-bZb~anY
z$>OC8JVni-`GcwAT<JTxp7K0Ek9dVi#wr2Q`vBn^X%~4i`4HKWa+DH4aiYAW^iqN-
zH%Z1sE5c5EE)Ik9$12g&P`gzI%4G_^e1aS;`zle0S>SD%vyxSk(UJ?2LCF>ATiFJM
zziOF!0xAQ&4Cg|0C$9&%Z7;NDYQLcGWSQ$9F&Js!tUpdKTrXQMQvakrSC6GzOJ7af
zKv_V#O<0eML2p%s$!#Q9(PQ3r_LshMJ)*A6PJ_-39W&cwTAf?O%>k_^+Y36=x=ngn
zeeAwF0}0#}!UFL&*=@xuRVgYOV~f)x+$MQZyR{b3v}s?p>a?gdTY&PHuVt-8q5eWX
zO}vSJfi=RIqwj+6&p)Z^6sKjoC4+-I2P+2!;@RL=RTs&sr3a*)(&zHAsyXOkcwh2L
ztrl7%t&H|gCx^AuFwOLpMZ4uWOA`x!Gc(g#(|P8@EuzeKn(*{<8J${bq<i>o%x(2`
z`8M%zffIX4_u<xCjZ+)OH+-m%tvguzTXk8*i?W=u4OJT&j<kk!Jnf`+VE}60ynz+M
z%kuXq0mdCyiFXHBSQ|*OR3^hnzsw+CPss?-UQb;~*+b2s#cFp_O$d5uiRz3}D!nZ#
z5F8NsN<K=a4DR5+Wn1(=>yPEXljop=(E`O~`8-rNDVDAaehaD8<eup@vtO+`9YyY`
zo>Sb4oW;&hT|3-tTw|PW+QnHm8;9we>fU2Kqij^)75VXuxmxT=J%z2m)k(^Z7R=5&
zT_`Fgm2E5x%H5bdrto+vzOuHax51~yqWwenV@~8?h%`geBbBNBNw;Zlv_I3nYi$6C
zc>A=ujPERM`T$WEvjd|5h<P=bG6h?lCj^M`f{DU2LVf;bcKU#t_e89eJ^_*JEcvXw
zLfpeLF^jQGG^;kUF-^B0?PWi@Yg}6Jz}VfRlLKA_WCvazy*ohWeaw^Tw#;FPNvqZ~
z)N4t)U?Uebu(9)ajoFWNDU09m5=9@@d>oy+<ehhdP280?bsxGi9_3#utFNWEzaFp=
zJ`sOb)Z!v_?ipv<)H$f^K3Q}c6&g>lvUMQYQ_WiG-FQ>gI=M_9E0yxzbnb01u3cXf
zSfx|8t=Ok@O|`7iv(>R>WTR!n-H!cAj(Mfu%5k;6?#{IicYNo}O5IR&;P!!<%^MdR
z&$OIrGUxN`>ZtOFV_~*Izqo`l9td{TZ75k*F`{u+)4bvrZ_}>K`wex-@%+I@y|3e5
zzlf>2AN-&vcF(&>*|9Yb`-Czryf<~4_BqCG!(dxauQMUOQ_8|30^?nOvH4=%Z7*|N
zVwY@mfQS(uZeLL!S$(Orrr>*Cd|pZJ!wl0DQsS^wSuUr>q#>jtsQ5|4GVE24oX~_(
zTdhACB)H?|C+$%k+`cDf`=`x$YvjvHm#<r1IeW&0D)8GWjB#%)UUG6ueknGno>KLp
zFgkNr66Q|&5%i&G@TR=!?X!2|UwEXxNL}*g-dkq&!&Vz{3O-qHnJv*_yX73KD36G-
z-zRp2SxwnHd4JF~hfKW-y1}Mb&FuBk@df<6hD-SgS%Z0>OP=LF{<<>d;m5)(Nsd|e
z;%t+`g7VkZQ=4s6uAZ}}JA};F8^EAk=C4*AoVl%F)$*0YmS@iIi7biik2pACuir~w
z@H*zGXaYCuQtHN<(cOal_wip6gh{9F9Rn79ex>Hls~fQoeX{!MqVn*``#x{#enfnv
zHQb=d({H?P(3{~`{pUsc&eD#kn&!BOw?KF7E{kW3PDZgdAL}Mf;+8jDPxp;Cd6D^Q
zc~Zdps6^%4k_3-8xc9fe$#Q4rH|I4~?v}iBd$GWCp~xdw$I$WUO4UJ^ZPhDoElpc8
zZT|I{q-kZ74}{F|=lkvn^7H>dipqPHF{|-H{e<MXG0cxMvQi%{KD6|(?Ip(bUvDJ5
ztgm|0ms~yg^=|%bo}t!A{S~&gK})Cn7S=hzWAdfR0im3!UsuL#<VBrv0LYQpE~(pq
zedqOVqlWp}SChqY_Adp|9S=*MTEDe?XPMZMG%5Azx0%JGYd+QVbr>6cS{A=EB%E#f
z$>GLY)xpJEPcMHsfA#!lQAUxwCvTnjF|g3>js2BjJ;Ta$<Lf+sVCoMS2S#@tU;425
z<E?~KSI1n6zFYk`;+bnQyVSNNt1+u~U+*={OI!^87bj*^;40R-fpytyrp`_bj-1di
zch^GPcxRXYMbcG(Me%*@nc3N<yQE751QA3)LPAgw6$J#bySv5iZUufeC?JXoiiAi=
zmq@2{_jb&D$N%R(@F?JN=bpUhzVA6@sz*@v_?DJKmA(1d1wJJiKlZ+-JX`v5^*hJ+
zuRa!hTK)BIX7jI!@+TGilAgk~U8sX^q|)i~Me8kk{K^h3IB{s#$lAhHu_44#)1}@4
zw>-sWn8}Sf8(hMTVh8fe9IDU#ih78gF*(2G!$?-byN{2bN3VYyoOn3dI#;X3v5(Jl
zS8$(pT2+<pVvyu}W%aQwk-PTq*dBH-V0fuug}-N-`2vexhfTUm`&3gGCpLe7@uTL;
z#dqi5-}!tmiSh1hLR7MI3iVsVkK}T^;e4&y?>qI+&?IkcyXzWxKTH2}+k*}b?y1;p
zzqUJEdqwCH-GDv5imtuZ@fHuvWHr_Z?YkcJkM!a{qAnw6oSuY!9!P+c?|I~#;JA73
zW75tP%KbjnsYN7fJDRjjyQ;ic?~}9jl9%CAt4-IQ*<!yXZ>5uSugMzIM5~2nne;6~
z&W*>a(c<7#GBG~M^NZV8&Cjbp8Yc3RmVdU)Xer35?rq%AyuGt~IFPx`-pr$Zu8vce
zf8Lgooo;KD!d9+{UwtIhXjyQ;exHvn9<vIqz8LRTLx}N-?IRBwf>Rx%*)I$eO+L!M
zdmVc^@l@jeq}$1dQuseZ|9lz^W)&&!mZ7q|HLuLJov-Po@4IBNV1?33_dshmD=RnS
z)y4z*dh$Hcp&{4K<?XL35B~^G^G-is;9Xvqw<c3JccN@Z-9(E|=lhP#wx+Hx<4O``
z<>ThpOeu!DP4PKbygtmo;+*HvxBTI%%HY@qv*uOU)fn$Fx}e{xa-Ci%nJaoY8P=In
z>XBLT1Bdx3({ncD%M}|GZz=y&daL|z&Ebyp@nWG8c^03?o-H@3WTfGvc2~V`TA4|u
zDO2AHptf={r<o~~VzG`uO|(xGFQjsuM-LCjj_6D-<Ct>SaRYeUg-0a)gr;N;uUs$_
zbD<6)-=uqxaQ1b@YtSu~gMVc!EBebnkTH;XBgatLrB<P2DYuBdmi3w4!oJ6B!AZn~
z;!RsYQILf4_#6?hP4J3e&kN!5Ipc!a;#ffg=fl)jzMjMdU=oYOVo9Ks5H|y#I7EC_
z93uHBO_a2W%7wl{d(mN$uSiwANis#WlGRcZX#$+W&JibxPBI<NpB;!FmJFO*7w8u$
z4>LTE;4@*$uMOoqC7<F*8AD$p93&RbJ2Kb;I1B!V$YDL$5||Eo7puV%usax=GMy4l
z`GxPMhEvt32{4JWntGRZibkPzQ$wgLsI8D4mthaFWOM-WC)rXN@ifs2VK9Fe<ok)7
z)tr&3#hfb64Ne^=kxS*-@p^d@{z<`a;bHMk$vWv);t-jL{6G^hb;@1JeLRp_Mzw;e
zowI23XqB{`v?S^gY8gI3Sx<4IOru=I2GCgaCVC3pgi4V2NE_mV%ECNO5mJTxLezj4
zyaCRa-{D*t0zdTwPLs!gFB}LbPh~U>S%5eJFSMCl4CfsiU>hqDwJ<U9iZnnvTj~s)
z<LS~SNr{9HnRPj^o`a;1q^G3`Qe`5PSPy4g8Pc0v2`AzkfRd91)Q$q2ud|S=z^IKw
zI*~-=IPwF*VO2TsLQriuRhJ^&hzd%;*MA`d(hq0xR741W>qZ6;8l1&3lmRmr51~s@
zLv#sDuRI5=W*77Y+K9eFuc6hbES3-FGbcC|Uj_zvD`Es(@Rxx6QvqzuMY4s6A)df_
z{1QmqR?3iGfHT5sm{<8jd_`O#W=VKrDx4C=B??jnX^<2p?!zR?PU1be3HY?f;cR*u
zeF=M1fI4B3SRkee6EtsQJ248@g(|?gb{DK{6Y3AsG#e2v!iOrtWq8(0z!SFzMr{Jp
zf)vA3$)J8vRcJtz(Ydg~3eY4!s4TQ1B`{^P54i&Mj(=FuWO6PrvX7AKNrdbnWXWNe
zD7l_ELR<oV`daw=58^vaseD4z5QW4Ys5G<@V?+s@I~M>;-VG`c8_1hbiwGf?0lPdM
z>K##l*VBi(ivlo{&m%8j?<)}&u;g`7I%tC-aOOFPEb5D{MBQQc{DEKI15-XfBR7CQ
zYzOrXKg0{FGK-M4@UM3A8F>!gTZ1$NRx_70Lp*?UydSD4?}6jI9a!XtpeEyt#K3v}
zDLnr&V2?+Ve$W9xiKM~mo&!p;7+BCR$+hGdagVqPwGu8d9e$<~aD{K+f3G1!;Zz?+
z-UAG*9^wKl?t92H;9@U=bu5QTo(JIHZHPR|Mk%NqFxn?zXRAR&#$n&3P&Ls(hmbE&
zo4E~|wi$^+Dv&ZH8_9*Qod-7bKO86*YDMRO0e+o41-~~6*xxtc_j!|!q&6u7y#~xk
zb*PI>z^+yjvBU!+ig-yx6JKDRR-`NN<<|pud>vSX0XdHhBz>TRz+*t^X+b5!8&+uz
z?_UR~DF<MKUq{}6&L={hXfv$)8m!<nJWC1E1+!GA!A=LGwkQ+*3wwPVxdWfeK{9oq
z^ET*mc&$WNLsx@Fu!9&R60Fe<-hCx7wEcl&z5#gM0;qfaW3q21x03hC1gLH8gQw6U
z2cWw^9nnMd!gSo<L?h7*A1y>H=t%*Q2)!P1;Jf|A0C+u&WP>D~$z34T0<hD6sKQo|
zUn;Qtvw*oR3%~Cb(g6C*MQY$Vt|L!CGycF=rob*oK@PRB<{Tsg`3^hWjEsS2VS&}z
zqWUl|wg|qSi4?%A2!5^zc?ngXrBLT{0iO9{SidgpR4p*duY<;h!ZTa~nO_GD*h#L2
z=g<J%&nG@WKZB?6kx3N8Q*dF$%A_&KU=v7e7fA00)cevvp4Xv<b{(pEBT&CYfdkKn
z-|7Q8ABm*Es@Fjmfm5Il?x4vhV5K6&9yEObBytRihM#y3JNyyJLq37VMuOcw0*g%t
zKeYigW+PbtI*{!&z_aC$p5zoECA2{MEn(ju!SCDxYqcVA@(1xAY*3Zd2hCUo-+2#J
z$n)e$@&r^O3js6wuNwfLw1mp$zk4hP*olA7ZwRw?>tJ_={!N-i-++E1$S5!&3ITPt
z9JKy8=tUvu_Apc&i@-xZf~PWv`sgB9*Gj<W{v>0e9*RTUIu0`041RDK+5w)#0qFvb
zNP#M6H{kcY;0eo+B;+y3;1*QfdI0-e3Vi-;V3|%Ni_9e+fxk-zk75A(e+evkDLEhN
zm~K!X`XVia9uv3WLkFVdL*gcpO~{i0AgeY~9Z-HvpzZ%S8I@!<Fho{D2Zc4@w^yPh
z>W}S#&KaMuC@ct5!-VK0O2MYkICKT-2$jvnP+7HwcvFM^1)n|!dw&<?QwqP|kDN|U
zgO?rb?G5rZyl)|VUQRMVBl=<YX`r<VBub7G?V!~deCIo8ZUA&WI0W*dz+~b-pb-~A
zK6}a2&~;%a*m>grcq|1_BYprHQwzvtTc|?|VD0b79Uw79kl`xu1)Zc3SegysW%~eo
z+YjjNLfGGbT#q5(inu^uh}Yn?jUYbuf~|58E$~ClFfI845&~$vLa?Gf;Nt{<PHBUO
z9R=Uif>6NAHh@-n14eobD%HEe)3|{T`3$;$6}0#b?A|Ppyb*k!5B{SIw8j$lAdEZ)
zwc&qHDI}Cg6X^J{4t&)=R!<J}yEp_Ie-7;a4ZP1Uc#7R%Mc3iG8lV-X(4AyEXowzC
z1|IhZ=!XmVek*vg=Li*gqadLBLWsIQkR^cD?gSk3Jn$>V&}X6-WatPIJPsBhhps_)
zLZ1{0`W>p{Hh>Dpq4xO{cJu^!1@>GG2?me+FD`5W{cHMPz5OZp?^a?r;Ys)stBL*K
zIfcX~@cA<E<OvYdjzDDENt#1UYa&V^k~I=4|LZ*x50?9H=XZd|tOIQyfjIgCR&)XU
zfDAnOT6nf@@+dNa=%a2hx4Ip5$ATdTI|?~i7uf1HkYyS~5LK`a8F(f~@H~!CIll>6
z!fvqjYVeu^|I>I=$ONK5+bakOVGipVB6owY_zU{C86u1eMBoEp&84vN(@>{h0x@cs
z@B&HPg1#SP@V-8<s(<VU3qbK+2AwPht3knv!pO5A&k0f!I&b_3wzwXRM>Ekh^f4L%
zF={Pz;D`ko`GD^JfoE5T-XmI2>z_h=(M!<3CK0^|nU^nQG<2vi*O30?5O}>k#6cp1
z&<7c^p%2M9(7k3v5&Wwb#B2uo33NCIR%=Zbz+St-&)$c;Af9l89AY!rp%&5$R{jZe
zY%6G4GWf;mus7vk3ok%xGC?X7$T+&lXz)|FK&A)5?`uQEwgxn@EBMQY;1#*ZWtgqL
z3@t(S!IN|024ywK?*+v0D$vDakZJJ9Akh6~;8(vvT=)iYzX_soDrg1^$%h*Tf5<{C
zpu(>U`NuMNVpmud19FQ_@VD2%Q(Z?^0ZzIW_zNSj!cN$uZpdj`h{a&NE69zIWgLKP
zNEh}u4m^$(Oer4*pQr-zmO}h<Cg;FfbdX}mHMfA@Q~{g;4qo6g+%E)x1;E5akVg<?
zVei4COcF-mT>`;w8p)ZUz1tuf7eeMmhu%SASc@Do2o`VxwDVs!wTSG62&4mkychPR
z0!c$>V^Zimv;|!VHop|C<Q+tfr{K5FfNdp1R5F2&y`UA*uup*y3I62=&ZHMG6oVm4
zvIPAu1r3gZ_4vYyV~G^$ed&-?DE$VnU7%$bAcD0aI+y_4PsygBcn8It;)d;kS<C~F
zKV2f&#5?GWq$*W{&us97(ICfU(gV&cjqv^lVNUycv>&pABM{;Gp?l3G@Ko{W9`pf{
z3w!v_E*W5%p~Q5!hoTXyh&ALp1jkUyJBlto9c~6*VH4<E@LOug8|Z2jEIlXLByJR4
zfEsoR-1syRp2$HsE7YKY*hjeU8HWtEAFO&VVgWZ#OED$v81@vSVgZnKhe00EN(4&V
zL_>mM{#E`29}|X&j*B}aM&xbOkb>b|_)VG)BcHLDahs-te?=Qe8Nx?mBDNMMiu1+a
z#Ksa?X#nw>d=4uNhFc{zz8r30a0&<2M8nWNbS<_Y?0*$n1b*TyWT7>1^VTN)E9sDM
zU<Rur+{LiK7UiLb6H0kOF~K|WP^uy|hZ2sFh!eS9IxG$lzZA2?slrvl?ZVAMPf@RU
zKw3v$hd6m3GMjuf6T1jqm{5EQKS%AsN1*RjH03KLg3^uUVQXN*`ZdB@YA9JHE)pq=
zABY!-;qFA_Cz>zb3a6oW5>Kf&ah&*1Dlbiw!r>M4MHQmADco!wg&n3-{^AzYr7)*{
zJ>1L9h8wRjN)XkIYKFUD_aIYw0aiE#e(DKnha4kQrDr9xB$p)*rBY%ZIVN2wJ}dMR
zCJ2{`&WZI17J^}(lywwk+>d&S)<gS7^PpGIbn!^^C0Kbr;R1JiD<$m`Wg>)lCDDiT
znjf;0asqG0#rQs|EY+B@16zo_MC%~h)*@R`M`R{hj$~4daWOUvn~elXjm7&#gz&w<
zL*O7P7B&crMa7bvL>Vni&Pyd(t5h{fwuhC<wxni=<fkMfwu3~^s}6@&<Mx3rx8Cz(
z-z8Oe2hELMiQmDu(s?XTSyk2z<}%s)vgwR0d>Z8q`T_fay}%vl8jNaOK)jK#M3zD&
zK1(<xekL6iz2&;{;-ohSj&Kun3wtfvE;z(<5Y3akAa<h-cqhA1<}&LH2J_p|C#-Re
z-D=<1ir7YcT;4|3k~)F?N8_@W$ty5^APYoULvu&735>pid6|wuw)IDnM5I!;Q%_R=
zh|f<237Z7%+#>!>F^7L<^2=0&kcnSpACs3d$FO0{l@-o%#prPRI>5+ddoVg=D%dcE
zRqd4SCBrC#aJejgMrT}uq<K>_wdHqvOIvb%c)6g;s6BkpZAvg2GG$0k!!EOG6=IZv
z<Y(b*>JNETT}gXDO;vG3zLe$0xIw=|bzw%a6y)7l8AK3~C!XZQ@lK0Xu~N!=A%&YV
z(JlCcc}pYU{^W|Zh)@yE;a%j6^U6f~h|6#ea%7CM9<z(&CU7A<^Ep<HoIP_X_5+AN
z?UDvk4{N2qquv48I?{K`8>@CJp1@}j7^;9CA@>N+OsR-?hyryL^(^8n4HWL+wTjY2
zzl5G*D}IgOt-wxtl3GMLPny6S-%x%dvVff;cM2QhrVi-z>rWCeP2EG}fZ$A_+;P=B
z<;ApM-tU1g0x>IHp-oAizMQW;R5Nl*%3$4>ovGxYVlOj;bFXVz=Mv6d^=@NL<AZA2
zlCP7CM&!p;1zX3ijrLFO5Od^W)m@aWW#w>9GF#My8!DH|Z=~jOvjrzGI5iW7;wI@9
z>O1B(*(#aa*lfYcv9^J_-k5<k!yiWMI0>Q`WV6&BZkA?E&Hw`C1P@0KP}Wm(W&e|z
zuQ5$=T+9_a<3S3zMzHK=rWA{q=o&sWv4A)x_gHQX!xH~0zQ#K-ae3-JzFJjV8&fx8
zg^Dvp0?b;;LQ_sLnrSU=7#$j2AUQ~}p?pFnc-JN^1_GMo8V)r$cdrxH5p<S~{2E!N
zPNwb*%}dI!sh_c}wCNa+nBd)*O5-K)B6w!P*`&3=io@d-5>_C+Fyy*eS>!mfV&E#*
zi?T&FMs_vxk^EitP%W0mD%p5Z%}{^8)#NwKPo|F@$~Z>4^PdP(L|d?J>@?ag(erVI
zeqWeaWINh7wN^xD_`zhN_w0?V^VoKxo%CfFv%>iGV}FJj+%9URa-6|g(_(F0*;_GE
zE?4#_&0M-i@*J6<)v+$hEmHE5JI3rrn9{Yv5^k9!A5$W4a`0aF#+5C}?W?-ajYNtt
z_ConLvW5yON};N6l?2Qa)NI6#H0BAqHrM(8iEhgudn`bBk2p9#jB-G}R_?RxN2Zvq
zpiEV0kvYyD!f)Ys(Vawq=+W@6-lv1P69WQ@=o8;Uw1vQ_bOwz|$5W_R+2soVvF~7R
z(iqVI=|uCUTGKqKAE~=o)vRQBUFBdlN=rv?(@o_q6oM2jWQyq()JXaoh6dA!7EM1V
zTdsi0Z)Qg@a%nGUV@Nt*uHU*@z3M{A#j5x^No#bwPn)zmb;y=mEDb`}Qw!;740qaN
zWVQ6D_&0Hy@{;+U#ezHTE@~U)Anw5MlL=(~N3$nGr*;hncOLDuYi_AoQ0-dwzBH!X
z`L|tdQJdz#jd4@%r1Ujoo9cFBd;4V1i$VL=9o#u^1a5pzjvkQil-bH%KRbf6Qa^Y>
zz)T0B_B-@o_h4~e>YLb2PoCX4e;zsW{v3Jv=-o3>$*&!fWKw5kDgHWJ{-)+-)4q1K
zuKE4<hwaClM_Yy!$6bZJ_$|4ws@+<>dK=AIGuv&0ZQW<=Fv`=?R~%$4MEwQWxGzkM
zSK!c?gJz5UPH!bn7oLpW7rO2AMx%9g;kv<*{x+U9u9@>n?DZ8dw$1q@3!TQb&u_XV
zeN-Er_$>XcZ_ddsI6)})>GbL685J2F*Pm1$pr7Zx?pfLTsc~(KL6`i{G5&r^iyW$}
zWbtN}tAneZjM)bDnUs?g?QO5hlr!5>3cr<QFn@Ud)GImFm@&9e^p@44HllMvy@_4L
z^pf*5MO`}<cSgi*^Vup{{WMfR=y+i6!tFkDJVKog+nurF+Y2qr_~Bn&pldsFX5MAz
z+ZP_zJoyl_<n`C=@}A|$8F^dPaIL#qhqZQVq$_*K24cG<b5UKHI<*@H<I|IDs~uOm
z7PuJNmza&qZx!z!aBBKoxFKycO*8v;{+YtWqWa?Y(veDWy=u#`HrK8<J+fWl&BvQ`
z+VA$z2K~C<wRsG7*%TavBfx3D6T=7Cd&upo8?{$&3ZX8gE!gPx%7eZjY2MGNTZx<Q
z+eiNwf9U;{uh%jva=d@FmMyC<sqe|Zo4zbF>{o928<^a(rv|BR;eVR<`y_E%^T4&O
zJJ#P^?G@g>>flP;`u0%h=ahKUv?o7v^6vdSmn+Su)LD1G;NGEMXFrwRYTjPdQ)W~%
z(y*-lTD?PkYs1OrqkqSHCVE!2gf!79%0d5DxMAO}Em|83*CvEhgWCd47k%`8?H=uv
zVE@g&+Tr<hCE-}nf8TDWyh$v4A&4?~b}g2jcrjTmb#HoZwnF~-U+;=5DqVVxP3fT(
z8l5xdd&DmLz0P!Z?GEjT)!sBaZ>xRg__QN*&B?1>+uApEF6%tdA^v-~am#ONx!RwA
z_PL#H^>2$p3O1C+RQcD;tesXDQNN?Xv#F{jyu+*SZns9$+@=GH(Qpq&oSxXXZrk}S
z)f=|0jtqUXtjqtTw~^~5=STDOT;h#=IKLWA+Yb%v5BRiMR!k<@-&=C`M9lP9qeTDY
zoRp&}&1szKOwvR5fLZg5cGC+wP0BA}=I#Q6C#r*#RO&JH6LwiX;|p!w_RYMb*+hOD
zu&wngj7YCedy%2@!=t#dTTd8Fc#5rg2M5)wHA}q9Db>OnlbW2`oejoqy*;geU)B6>
z=r;JbO5$|M9?`bX+YW8mxtbrGw{)*R=`+{kkjsU+V#ijyevN3(q4u+7?^Bn)Fp8P}
z7=L;>=G^n}gp`u~&FMAXmCAo}#Am4$44m?aT1upkhw~;`TTH&rakl?B^NqtJkJs}k
zvr7!OD*mAM2zmy>+Zi2G-7$a1YYM9VwtVM5#{P<;28vn?>r{&y3YeuJnd-yU^x8#r
zS@i);(1Ex6Z<j}dsE)zB`oHe$IfuJ;_U$y<vS~dtylMINz!txqUa#G>;11M%4$FFy
zxv8hFIQG-x7h?}j+?)O=@zK3#nXkb;NusSItzE;Nb4R-d2kKMGoGYU1VycZht>rVE
z?}rM4;yh^{(_J6fj_Tf2+o3eb*d+ZXtfz&lq^Y>j6DHN#k!oIbZS9YS)pcuX-qhY}
zL>i;3zEsRFeOP)1I+>rZMr(%~13K5W>s8nNiBtOjJ?+?UwL^L5*^RwxR)vNJ?O43Q
z_w#(}ym=0_4&jdZHp|5}IdM-Yciz1iNj~>EJn8nQp<kSdJhf{oN#h5zLt;CURZ0|F
z(EcsLYm72x5l+zEbOOz(s`G^S*i-5e{c`Jcd#U>=|0_N(tQX43@GFLlI@9V_Rot)9
z8Hh$n#ijD6(9mJ8KR(&vX(zJ2=Dq**s^EBWK}lBm!ry~+`t5U?T`F!@@6rDMwJzRM
zxa}^;-(+=th*{9DfZYpP+*6$YI3Aw+b<S+n@b<0oJFj%zE|2d`cSuZ)ANb|dwV%IP
zU^48{8kDJ#sPMI?eH!l1+Rt=UJ8OJT%Z2EeJVW(X=%f~)KB}vozAfYW-J5G*8*OAu
z`>)%i{&v^NsRF^`30tlj9Z?=)-{)2}u}epCiqb!1%H$oXKq{^kA1eJ(wy!Gi52c~4
z{Z30%Wqh@_+W)ULbC2q_nr*K)e2&NsE?(9YFk^wM+iR!I&TE`IZR#*#%lrHrao^(t
zYU?Q~-0I9-FBfNEO=Zo(H0ozwnQxnG+u`<wldq|5!d10#nSZLOg0=KpOpbz<*)4lh
zn@{@DCKv3F+u9qhRJ2hmFx_r!uHwa@(BmZKJzrXThDvB>m@_8~I&8bL2Hfk#S*2eB
ze-`FM|JeIeGgqgu7P@{ftaqz+DjceMXY&86J+fDKr{Ye^7TdLrD<_uQFX>ot#*OdV
z<<;qWO65wecSS|}(m!{9(#jrBXb(Mmhn%@~y{TZqNMXa(Y+>Qw&eDPLR;2aiaP#De
zc88Kftuji+b_*@+6?QXjYdINt>iOxtaX7YM<6M^3cA5FpR+&4h4s*z1J*j{;Md%O*
zX<_Wke4ECPMbg6E<p~vDCAz;%zW@DJpXvOwH-AGRv(l@^vix2C?M6e#f2&<{z%ZP(
zHE=g=+t)QwAvVh|F0NQ`-JR)T=)T5v{fsQ`qrdowa@X(Q=gTeYWAi(1A!pXywfkcv
z4(~r+O)DMz<6N*UWhk$8Vy<G11Z%y}I)}-%{;pporE{OEr`cSv4VZP*_0x*PMP<5Y
zP*;YT{v%zA>?F>>eUMs?>%bJh0jnNaUZtL4{PRwwMcI+;duf`P@;NfUiVJ*-HOjtL
zZD=5BREji9ixk)Y&ud-WU9fd(yUnK6Ys{b=GZApycZuf)_gWXrIe{}Ztv}inTVLlf
z<I8T(P1egRdW4)&you%B>*p0ti<Nydkaj2i;-}1Sk%QN#eNYV<_Z>*awG>~YouauK
z<ow2kQMU3{77haU>vI<C?a=yS5n*@1I7b83@KNT7F1N_HZ>B8Ja={H-&R0d%{8#Dw
zb@_+3Z(Z3H**mju<(@1ci>H;(tHtWNO1Sw(W9e)E?e1FN%WmJd#P0LnL4-e8QM2@p
zKWo9j`~~jIJksZMSsYRx)BDfFhkGEs;3*}>_}-lh?_Qg>5Q1gDdC@M9oIfZ032d5I
z)B4wSDoErw*}*ktM9M8iGCBLHn$~hYe1~V#R;sdeF6bs|Fm-UNtFsnpe-iAODrQ~O
zV96wLuW(1m8-yHZa?G21YD}@#C)=P<sU_-fNd+_0BJ)6ATainJeoal4K|yTEG{u;m
z#OYm&x2sH<$ZlA4=+ma!V1p&g0*}H(5vJQ7H^Dr0`^Q>2G+F!tXMOdxk~>W~E!E$z
zM{z!K>$94Szt~6Z{pQ!TcC4@sY0U0)?uu*D?QUgMn+dgXenjtO$qVf+i&5)ecFSkg
z=&Ydy%dW9r?-^>DkC796$`6fsO7i4w!DZMm#zb;|;MgeFb<nVWTSZZWZ^z}vJEcz9
zMH!sT$Q-2TUJa#seZEz8`{=-O;`G;0W%eY!W~tY)1AG5O++Jb7!YnY!L+tU$Yps{l
zyo8xL>hsmEuu~fSe=cpV7j#vzZl+y&k@2c#Zqk{11JBJW>!)7T@KSYiHvSE55|xOC
z580ded#PKtE05~WTouqY+e5$C=7!JY96g;O1JpCgJxn2JVxQc(xo$HJxjNMs`i{^o
z5N6M=Hpa-WUfV*|l(2%o?TiNbto&53tX=s>3#*H`mCx%{s+Z;I7CPwt(+7JKtKkz?
z=uzUd%Fcb^mqXivhnEiePxm$OnR2tTud&3ep3dH($Y`po@Ew^us`=UC!teWrDU03>
zK3f0eS^S4;y$UQ<<?Dlz6O9=~^3^iRK1&|EJ>xU$J?R_$riGrK_jy*tya@+Cjr;P{
znHv|LwT&llPdTa|wDM3sGI+QrpV-UXD!e>=bz=YIye6&ex?h^T>66)YXo`8VR{EM8
z-FzcJT70Y;DU#2{S_HGM?Xe3?VSXcCd)+#PoL;|uPefMO(NK$^H38;+8~v8gx1PON
zeL}my5GAHpPgD+$?CW3r_0FT2@1u%qQ;**k-ddHbIjYslNU{5HqvB7;jSi2o7aFf;
z9ni8I^B9y-@twzXpJrn}quZv(oT;`%nQM00{-9Z(%otTwrAq4nqk(gWe}$IAY!vAa
zTJ%@+-)@nuyjtDfGkx$$&5AVbWScD4(#fi!y6epre`2!7)3RHaO%DiXdqvasPiPwn
z;52Y*pV>O^u(Z&=<r|koEi&<a=kdeQ*E(w^F;lFB58iKe8TmLEQP39uE~T^J$CtM2
z4OiBtIyWc(rlfRztjIlAax(uy{VnEMi(uul9;FUnnSnX0-Fxh(9B}88Hf+5P!@Vx$
zzE@^#QMgU}sT!htM*)*;5ZI9k(y$4Wq5YE<lkZxd6+SOL*ZF49q55=c*GHWU`JAgi
z<n#7da4U}H{QREXAf0i36W?c?h#eJLw}VzF?cTiRK^PVmzdUt`tpA4vWgZ1ipXXRO
zy>-wxO2%uEUre*%=gEo>c}WWjcfI$$mUxx_694J>Jt^Vshl#YHOxdp<`3j;iBc`Hv
z=c3wCWXN`@N8-G>ZihWwZKb-^Gnh*f*2a04=;$jxGu&)-LAx7WE!{z#N8#~qPY8Lb
z<I`)je-@Rz?e8DTEd7!6<^!3tHs3qrLgw}2Pqotv8qx!6t8}s86!&QMbn3|5ngbE1
zB6q5;CBw{CE(zKc_<P|!pD%8Y=boMG=`5TztZ+xv%N|#coKXK*|48#QQmFf}>RQy5
zgD+H)PQ`V_d4Da+d;a}(vTk0F*w>(t>CyGMy`1{S{DVV`o4o&X--nhnHB)CctrTuq
zvS3)>QLV&~ZA#awQ!r9o!;0o}dx*g#-nz+q&6P#fMJK8+RI8V6DN-q^%@d}){*>{n
zs`YNmy}XOx4>pG?&Yb0M`boA$4znosKetwM-{65u+mlx7dA;|^TsE@QauLnbbw1Cj
z+@y~cPH|GL(^xb4INRc5Y@Ta{^;gR$x1QuBG$!qR{`PrHN_bBBmwR#3vn~&Pqxurd
zM^8=Yqt|8pG*8+l`h8!JKO;h8wt1GX{EBLy+SwK}39E49R~k{8Df-tn*Ps={(%}H!
z;>pT3TBTjZ>))PbDY>?}o<$o9?tNeU#VlubU0&1j(&D_b=6dQb<4EH!jV$d47V@q?
z0wXsqIoPtNd>zA&J>PkWVUS(G7jFg6F_+h-H)M|C+A2aFIg(j9=iBd}{w3V+JDxi|
zt$UUC%JIq3sN4^mGm_r7zp{Jp@TZmYko#@;;zT8~k!HY7GiqMAIl{z$zwM;mM*n@`
zPnN$~)bIZ-pv1L&db`1Tv&BZaG8r7+=(#DQk>?GRLX8|+_KTE=ggx)MDQ4eN->bg6
z^WpOM!XLEnooT;*he&5@*JvD+=}-)vzInmywa1RGiVQk@YUjF5`1;ckf{@asw-;sl
z8ZVgRe#_#f`coY_lOu}fdgBXs{lW_#eclzteCYOE?m2Pi&&>@_7QKC#F#nx2aWr+j
z>|s-7XZ}PC%BBvGHtbzin-{q(zTjTv#t1yL!g}FHkGh4E!EXaY9DB_DW+=_ruc?U@
z3O&ds@w`66>X6c-rDH$)KRCQ?{ZO6u?CZ6U!ymQ2p=mom-%Q)x^qbbC>!|ryR!a#n
zMjZqEYd1eRy717Ut&3JV27g<5K16ouIzNjAq25Pa&&}}A>DMmM%*Ug8)$8Xp>esE$
z-Je9p&3qgGEa*Y!)7}p+(ocV^eD>#gW3pdfYQ>zE;&B(!OKQj~B+ImB*!`Tl*754R
zI-f<pq`RwYo)^pSipLy#C0jo`4Qq<d0cH$hN2u}329eI1c7x`f<;*Pe)Q4#bKN7Qt
zv-@)^f33;i_H$Rhu>SAl9GG7j&%PzUTg%H#YFF>|CuG_B#Sx5UnSsAUHitPakM(!;
zQ(y3O-pd(5dPjAYv}2g(#`N0PwOX`%uM+=!lBV;u_rsId$6g#s_?o&XH909fspV^K
zPC`-IZ|jcDV>sZkrW0+7RHICDm+6DnM;y1#vv;X<ahYG`o$A(R_tY|F#+~Ur{ZO@0
zxoM1ciSyV<uX*p?ZhG^=@-x3|3S^303o`Qe7F{fNEZb2quZY_CeXLG;7pX;;Gao3M
zYdaaep5^WnwsbsTulG@J?&5}E)!>pPFZ~ZLxae|e=7Z^63k~xL^>XCWc+gNzzjnvX
zx>*&Q3s$9nc>gqRPvXGmi(kf*P9|(kI`-x3cf(wRip7mH+mkz&4q1rrvBy>SX$KhW
zHos@J(MHAIXCBMzi04WdRr`mwUuUXX=9)SiBx$KDj#Dp5KL`_f%44jqvU<zv?%&Km
z8~!l=*jJyZd|Gn1aH4cZBX!V}zejqDj6uIsYh|=F9-5UpsJXp$o#DLE<(}v8!r%Ui
zehkkPmq3Tvwi{=L%ygL1rW+_*E{))*ja?jE)x&MuS39rhc}7%f?pGp>_toRGYjVwJ
zr*HfW?L4*eGqpDw{xlB#eK@8=_AxZ&9w>`76?8N8t{Xj{VPt1M`{vC1Gd9d<osl_R
z-jr!{NUs16TAgeM_5<c8EL|{TV(ZAG!7Y6+dfa=hdiQpgwq!JFHg~q&?_v+yPi_&U
z3flz-#l6^7wuh3A#tq#RBNr1#lY{2EGxO{&%u=&Xn2|MuJ>#7D71J~mQ^SuMD`Y$<
z?&L+uV?H*qV^Fb6p{e$_QN^2z=nA#cm;(D>=5V?RDO*%ER(qhysHLR!NSFM`3f@x5
z9ApI^$FP*Ck>9F1UFVa&n}NH&tif#~t})ekk>RlZA6>e(iaJ%5rJ}D$VgIEZC8Qz&
zf0A>4(s}&bFui|ew_{gaS4!9Wu1h_a2DXjt8_OS0oH``fA^n7|qxr~KD!HnqXxz~*
z*3&S4Y3gb=&8*LKv*{s|)5b52stjtU)u^vm{KYP&#iCY{=e()$uEF~5f%X$E*^S9{
zL4T5}wpJdl{8r`o=V4t<V`xiJ+m+6)zR{7wDKFuDVhiO8{RZ1uK}l6zbG`Nvot@Jj
z>t55N>eKYM>fM^QTT4&<ql!?eSYf#=jrk7Wjm(m03-UQOleuG4!-|8vzUtm-y&HQ5
zyN~qj>f1c{V?=)<ma|CkQ!FL^!`9FxGS*7!YBMzV>lo<oH2PuUWoB;PY}R6;ZscGP
zt$$a~Q>RUBP{D;AOSMN5#VkSl)Zhr#ccJ59o2Z4`<lazRt5HM$Bdq>e6VhPTlH9iT
zFRklFpVdhG<Td`N_$Ij;^Pu`LP2`M}7%Ep(+*C8wmTNrJyr{*|nyqE1(XHC5Qm(>Q
ziB$Y7+sO0=UPK>RBpndT2vM%zgzMOq(UcMY;opM~2X6My>2DlpA9*>M&lU631rJ2?
ziPxACBT=qcWsSxi&0MX_X_pL)ja5yIOx7544VLI0(|)4Wsd-gxp2Bt}hG&rRViA9s
zqdP$xdfnyKhPJx2m9`yj4Q~F~C~SD$*x0hSlh@ZYxMS$hu=RKoH$wc1jK-YsPxxNy
zQiiSUM#W=FJql{_@p6&!e1!z1Qsn`qWO-C(39FX1h&{qort9NbXcDPQoRn%y1H}9I
z;ZsK^G$(gY{hX4W`a7XBt~>gDq-y*gFGnl~_k9Lr7`g$sVmy*HRngX*r`4~~t&yX>
zLC?;BX`rNUr+Z(^Qmt5dzRGTuvkI|{IK&?iyid51Q<W1rqj&q!j+mAot%)7oUA>)!
ze?PVJ+rM>m^;iy{o#>ogIpI8E&fOsf79^z!UjU7W{i%iws_X*=9fd}jNY)UmO2$R*
zuUwg&mh4vMZ|XWKgF$2YGZ^?Gq=@(icgxz6y%IZdGw;hp`dIly8YhsOJ=HxiV>}RM
zDz!~~gSL!EDe>qZvK={qUt=y*plGa_<~Ge&Q&pX+bwe-DP}?w0uUBWY=5tlCvZ3lC
z)na)Y>SnPvKb)^9XyaX<`ZnU!!)@tp%xKjE9@Dh$XYH=7e_KcYM)X>bXie^)G#C#Y
z597K>5)lR9N+wWl<A15)%xiML6`v{S$x7JCa;{33m0v2&P+TRSFY}Rwvb$tX$VAe&
zA}7V#B7c#NXoK)DpFWj6e5&7jfEczJGa8K@`Z$o?j}G>YjR+gC`&4(zBzldSC1av`
zLidx|5zCtC{~7%<FgLTX*0rgyTxD)!d{{S4y-N9^@@~ZeIx1Q}-ZJVr-ZNs<cdC7M
zLrj%?>5a0vwFg^cTbcFwzXz&Q8?xGO^bL-^=g#3}!t{4l<Obs>dpA2>MnUn8>RpZX
z+BDrY)08x>DP2-Jp#DMYt$Mt|3f5cdJTzMxF1<@AN^+)d4}R{g?sMz6?iuKibQE=^
zc5Ug5>{9Kk>A%^xY@l=M8-**Ut9VwPt1zm{)jnpdHfy(gvA2oaZu=Q_dXBhDywmbI
zF|#da?l5l9II1*XzL&O&uQedot=V(Av#ViS`LX<*%*@ov6j9b_3AIY6<YjJRo}?_Z
zp4qW{D2x{=T}fu4GORFF109)Zp1PLC_h$51f3oVem^6*nSJpb9p`$Z6ZHamW>ol>8
zKRDSvS}@!;ynX0*cYHHh8&m60Pu52_9%xPKFz8C`Q2e{QBdcGNKg1|fi`IUqd0*q<
zG<oxH_I6%Xfy|)DMXNn-xtsVE2No}q_Xu!|pGCF!I?Y-cVY7+K{+!x@s>0f=>VThj
z<FjJ+M|DT_JhMuC@x49WH);2q=byfmlbr>Fb`#Uco3f`=j%n^S4Vfcx{o_*XHn{k7
zICuGwyNzAMtXR8k*1t@T=})TuM%V*`tq%Z#;8iaw%=uZ6^En%eM;{83Lo(EUZ7<nh
zX;_<Ad$GEsA&2`@&)XSwJZaizj5>>g2ew%4eRS0C<j?)x8}6*Vu+?|#r{GwZbN1P;
z6emONWd5h#Gp&B5l<x)KoAPYa4?jiElwI8O^l7xkos|3L2~X4JeBS&)CNH&fWP-*S
zLYL^RoXxfSU}NZNA2hl$X2tWBQJXS0=q^rm?6H68)M!iS#L5oPS4p*c7yhm<4Je)U
z^K^1xlKa;y-+sQGA65L4pPH8wldE5NvG)7l8y$s%I&v9v=Phb*$(i|WeiNJ?bt82S
zh3>t+Gk5dpHP6DjLaLU0o3A^2hW)rh)U@)66E&}^=2YlrlOLyjXh?Vpypj5|Umo~8
zkBCl;&Uq)w;g!72A1HC@dLRnHeHC9>zF#mFQ0{YO;mMVc*56)zXI1IOTU(BXOrO_g
zS!S|T8)X}j^931$UNtT`mfv@LZBJ5(ogXWon3pgayY{XA=gaAF>62M)mACp1k0tfX
zj2+X=T^P4wo%2d-i-6T|Zv43a%T~9I6>FWt&M%i+Y8DXhdCcCx_KjnVtv9}<+O+U;
zwQFfXQdx}Wv*pppt|XsNx>@nG{Xfm;w-es{a;yz4j>*scdzgMu(@jU&a=GuKm8z>!
z*T!!5+Hx`MSIDXjp_|1k7ti}>yh|fmwuCGim+h{oonE*sJ1_lVa`to11C6M}x8uo|
zK41P4_pL3nH^;wJwmD^Z%T(|9EbeioyB=G^cQ5vJuynb#@$D(IqcPjm*IkbI9y+`%
zC}5ReugBgwan>&#NXLZ=yQ<~B9xB>d?UeH2-m~iwkEcH>yS3>qJu2hH`d2^V19R#c
zTkFmh?`$5Xsu*83)ix5(dK-lA-*{ZI%PcfH;A!Z?&097*gg83gG^|&3k-Lme47Qg0
zWOV#6FRL$g|GecP?~cKLt3OiG&wqdLmHR!rV0T?;i(kEMQ_s*#Ndpqg+GW|k;!3!a
zhiG=ulA*&lkDKm1wQgWl;!2&MIRV@KW_hicdu+x<%Wi9yl3u(0&xv1l75d-mAI-fh
zcy%u6<I6Ai$L|qOEnf7*zt34;=idCgNz{{!{8T!x^+?xz#>)8>>sB60-yv9W!q;qZ
z;;NTx9xh$u^2owhM?p~u4IOw{p_F|r?_HU15&HG$E64vn#7=!sOTPV?osseD)bG%i
zkyf-}u(^9Enarl$!c%4Gvt9*xFAH+!*xD>QxJUjdXUB<k%U4N5hL$3WTm8;^ys&XK
zN30)NTw^cqtSlR^7%r>(=Kbc_i|*K^?~>pAdOr13ct7aLsgG|<e%8MDGu~i48c8vd
z@s@w1CTp(femLl3*xlfpi*9>)EK*tW-p|+dm93$fyFOpVjV$WsRgxt;|14`TuBgc>
zPTlnVNtSoUvUI0x#}e<l1AoK2PyH3N{^>Y2eo5RWGUK0NUb8#3Fn|6Y^FBjQuQzLk
zA{H;DE#I<gN7(%3a~E}cWY|O*^ctO=AvH3mGWv7dFL!Tf&iomXv?O65^<vhl?{hw0
zd$sPrfw-28!=>frHD#jO(9sM$hPeospWbTCGn>5Z7uI-|Id67O@Er6$<NAKqX0xk?
zr}UIG<yg}>;cXtZ9xa<YV%xLp-dFTjEvjv<iT^G8$FAvbm(wtR^upl5p03{XQ8PgS
zH+cBYC<pJ;yKiz?qgCd(nyU2~XE`Tw_8G@@?z9CS3rbu!**TkCGn#GDWhG-|E2mE;
zOFT(U{$#gu?bnL_-*0M7{~Rt>&9=%|ox?2SR^F*Js|v0U>b^Y6oigX1;oTO0qWUT7
zs7<NVDx0hC(CyK0(kfA|lMhy?P(G#dpFBzHK}bZ2Mq{p|tds~Ox6l<-cZg|GM1$mw
zWEsrv+btgA=#0;qxWS3w){nm#VhxE#A9GKNY$bZ)Kyfc(PcL9)vErE1Wg-+E)Q)NF
z(r{9rt|nDGr_ryzUR9u|ukc>cTFFhG&Du!)gS8`l#5Bo6p$h=tE2b(Y4^N<DR>N6C
z9mA60Ps1|97l+!0!iMh-Ge$f|cZ~&2NT&RG8w4~FPke{4MH49$+Bp3SGnoBQW{<3@
zthUU2b}nlPi^7bjSJUdKpD3*en>+=Kqbg~>BwV5*i4>m@ofQ&-69P5CRd}Biaj|5M
z^dw*cm61ajooYufW$3V&GPqoeoPyk#jFW7W+%5%61v@!QnLX^=>?GD|Mm%*E9)_3T
zHaHXWCo3c};#Hy?;d4PH&xb4I7;|em@>6OP*Tx0oVH5r1`^KM-`%XNZ*fp_j@;E1i
zx1GO7U?)<Sgh5|~dt@$}K<UTzXqV~#F<h7~%tpoyhB0FV`U*@yAAx$RGxa=vhGLDO
zXfz;AS;z-KPM;#(0i!uaz5wK*F`9wqVQ(no6f68WZcjZ=Jxtw4eMpU>hQa*FINTR+
zq@1G|Q!25Om@ZZb%o!=*HCq64{)FI4mjGTZMU0BKh#H0Sg^2=vK>~jRe*xc*AILw<
zujKm+`UJOxb3{F&7vgOa2Ps1QB#uDuj0TvYD36`RIx$PiM#>YI_t;Arp|k*!G8QI3
z?xEO2=Yu6!4r&8@bTP1FJ`sk{!(oAxBl#)GlKhqsFzK>RdY1426tN<zjcHOwDWCCE
z)VZ`mngM+?{WZOwE}}CTIHQmLiN2LCPk%wvqurqr_-?$5vVd|It3<~EIZhJ&(hSLQ
zv9m}?DB*LUiMJ_t7w7j>*p%WFK4m*~Z)!S6!lCn=_@@PBLL2cj2}1aiyO49}Z9p6@
z#E0-mDv#<#+dzw;dC}x)8Nejt<BxDFJdUDFIf2zc2aP4jUBFkm0fN*+Dut<&eUecL
zE}ai3{rSWYkxX7gPNIjgJ(MlD7gb2T0gY;x)9=8lKhbZ|m(V4&i?kl9Ep<O&K6`-6
z6oMuITKEZ|>Q4~8(sj_&!a?#>jEGN)6h!ZYfkKAxx8S8<w_vHjU$9KDSCA&K5f%wg
zihRUcl0nH&$SW6u<RcL_pq({A0xg&$$Txu!Pq{={LSa)fv8{j(%||bwL8u9^jlKgm
z)e2@%z5yPG1F1u*0KV4|xMF7kBV7V$cn?4y=Kz=IHY$hRz?hUsN-ISVUxFXNBk}!s
z5IzGYj5Y&mbqjPBc>pspp97-u9dwoX4oGJ+;J+jjhlrU(xAZA==+Kt-N|GfPBpW14
zB+k$k!dPM{v6t9M+$FmtAHbJ*0tUBF>I_VzYT&HR2DaNOKscWReD-^oeE4rpWgj4u
zhX8%v4~XMwfPy{<7-Rv=Al-l-K_k(#z^>Z{$ZZqA$#)_7z_8d4D1H^_|K1Kfgev&|
zY(T?iBhz7TSb&i2#GEK+DH)V@3Lh}7YPb?~ulNPnQZGtBe3ya609w5TP|*R<PrVwr
z8168i^gVQv=>m40iL_dB4v?Uu;xu5bg^6v%@?s(MWf>PS#IwZ*#d%^Q$t?*g-6^es
zS)eflk5~#E5H;W){RU3aHDJo=0vF>L_83dWYOx`VhlwygHi|W1Nr0v`#j4<0jL<@0
zIcNfN;4H|551gSG@KieiT^<EHWB}-9DR~gMKU)EVAB^Qd@04wnFJLJe_)Odpw}q}R
z6);QJm{JInOexqaV7&2x4Y3zc^R0juw+Hn5PJ&6i0bOmC9F!<Yvcwz3RB@{40Cb(P
z5*fnW+|{DjA`bMLNfB!U=5+|HvQX*-OduUVRb#+JC;<+@F7ziVgLy!og9NM{V^NGK
zmhjT2V3ay64%-WpaEAf;ycswqd_Z`|03v!Va8uR*L*zcNlK99x;14#Ude{!^GseX%
zD2qUsB0(-EDZ4;pEh${+ofC;!WA(r_&_#0qNlrx)0gp~0KNBknI*|!`tSv2)9G6&2
zx?pB)pqL@96g?Lm7i|zN7p)SV5XFlAij2g^#Vz6h$uEhMG(+l1lo9^G)tCjGBQY@l
zuA_M<2mFj9a0%C7+u*euJAmy1Jn$lHE@q6OSQDBA8s-oEdRl;owGa3S#=t!I*R`Gm
zq`Mv9$WH@XQVJY`CulF~h+V`gF-3|G<scxQvp}ngDW&jHKuM#-P>#U-S_4Wm^Z}Xy
zmKcI|f>u@m_acj!O~gx;rH3SK;$ZP_QK)E8cunXd)Pfe{6VPcyS?DP|D=ZdjiB5}f
z@guPfa6b1)O^I3}5?F;<NC+UTw_(+o9pyIodvE+O{t$nQzrr7YSDXzUfMO^f;CDl@
zPOy?;cm^RbbEc8$z+^Nh2<XYuByE>sgd=ePrtr!GdcKi#0bIQfXsZBS2{Vc{DElet
z;Ky}w4~TE;@eOz=J|EY_+bB=LW~Q*0uvfo8%klwJUkFT||3JGAgHPWs=@TD?o;G!&
z`=Yg=TTUV`QH1ECC|5)WX}uI<k}Z-lh#s$`axne(2mJ1hWEwCNf`Io>0ZdIVu<kgt
z7@b1ZFiXr8bAkRvR+t4anYd^na7?!VPjV92aPNTGvlcirnh*zBFxSBw_zc^CGx`E}
zavI?8ZUDQZ9>p+2K*%4)o<lFD3YaEb1s3=c*up2FThm;ylXCElvw;Qm5x8aqaI56W
zD8h)ykcLRdBv&K~lDA@iu_4T67K;A;F&D2Be-_iggLO%kNxw>2L>LiAu;2+kLXRvG
z_z)Jr=m`R=djNclQs5gBz(;aI*P*9@5pxvq^K(#Dh`AZyt7gKypkKh}aU`X{5G(|q
zV;u1f7%G~;<!mOG0_!XYBpiv!!hW0Lzwp)6De6&LAK<y~(JSdAbVY^+W03xa?nm#S
z?V<Hjmr`#+oJzsoqp9FSs)%gqCrPHbS~Mo43rz(}_*Z#<xdGe;&K1r&jy=bK<H|Y0
zDdXsI&vE5>pLm=2)&hN@wJ1OwD%k>YC5;pybnsK9Fmrk-RYHACQ>5>qKclD7Khr<J
z=Xdmn^ab=5+G<)pWQ5U>R}5oAz_-B>8*)F<Aa#<a!ZhX?;PKZ>K1t}($1qcRCwUBc
zgw|o|xC_;pHjh4sp}<UHl1wMoe%5tX1Z#-7k=e$W$yh<(Oglz>g8xU^hZ&=zWD#*e
zY9PrI9T(aNq`W`e0gfZ*;uLrC%;c;|tx5Gsv&nUnS(6@9`BUMXnOrM~@6G~u;T}<x
zxJhC_Y#>t*KB_~p#7(IrwTR|TkB4l+gfWXD&#0xJq>E_5v_!~`9P!<hd)ODyt+~ie
zh{!3#Qi#cirS8&&(zVhE=>ch$bUN6I8?c=oArYuL_8WUid4?BIQ=m)IUwR<p2LokV
zGYyzLMl8dS@tN*OZ=yxg20$_;xFvoW=3m=kx6oW9i;N>~Ne@W8#iPO%f)U<#ZWQOi
zRMcesMCb%%LNe|%kw4)#**+OJ^^7ydUBJI6=n&dM#`PQ+^BH6gas@TQo&Yj_4dhuN
zcrmU)b%ZKkJn-F{sQJ``R8#6W4h(giq@0HN)z^Um_z5&m1Rm=jpVA!IHbU~>?1FlT
zv`@$)U~8-gj<h41kDkEhQ5g6jo=#mt`$}`A3+Sng4a|IIKXZs##7twJWvVexFz(Yo
z&<3fCsUPrin9F?{^+VKv&;LLYC^i+z3WxdSye6(6x0*AXvtjD(WW?n4Ny^mbsfSbP
zQ*2HQXBl^h8^+7zvH2JH<^qo3yD&l|BhC@8kQ7O_N_Bx{D1`kr0}n04P++S(rK-~&
z)8>Q5Uj^1m7ULCk)iPy_)6df<AWuu9&ZE}iPWUe1(6m9s2nRmUW{4Mafy12%T<vng
z6S#(ZA@UqYS7E0pck%nw`LskD6Y|PrdJDaeK1y$-pP|zrr?jA5qt@WP6dpDoGMipt
zf~+9Dp{o^5LKUwN-4>Pz>;#Ya&U_y432zxsn-|Y@=SnzyjyZQT_XIbR`;ptmoy&W}
zbLLY8)q)i06!l)5E9sLC5H;jC<PCZmJ4jiDJ5#09n*g>xL9e2tjA42lJ&ZmL&L_)g
zEx<!rizia3lqZ-oRt+-i0siO}h(0*vY_-4*P6Gzw81TMoVSmEF7Y?J#u~uvwrH0~*
zU&B9x<mOU$QRhQVxCpNVmI*@T;b{;VEbv0gb%?Ls5LvGR2hkmQ4ElW!@aJ2^N}{)d
zmwY*XG_RVAao2K$Qv*|jkSi-p@h2sd)2FhgHgbM(=5xPseR$jWM+Ge5PT?*QDp@U^
zNiaxLu;_5gTYM|^7EKj&MV8$rb6xhdtc+Z!{1*j|LZf_}T&mn`c_;a$vJ=2QNoIJ_
zuTrxpe&_&kTH+~E5eoT^JYSB@RO-}K&H>I=P8pZrcM6w-mra!3L1>h#xCiw#ZHV5;
z7-s}9gIT6<p1Tb^8#(%J+G6}3_6|LTbP{`|Ws-*ysrZ5Di?B+_5NU{Z3!Q`xq9T!k
zc%e8+yiQ^u{R~;DJLC_yU~aszgemzVUM+ekd@I~2d@YC;C<-qMX`-v5&tiM&BBF|z
zLna_Om>PZ>@1};)57Mom|C%q<qT5-Yn7bHznUOMQSpm2<x&!}6OTp-5h}4A$BF81;
z!V|(IaiFA3bWqqLd?$_;XF>dvirz|^q_c=t=+l6Z^MQqaA6bH4q5OxV)KUC8ZbluY
zhSPr0%b24q71nB|C+iyfx{Rt!6>FT4!qSy5VN>u3EQF>-{X@1&eu@S~*F<wAkpkmU
z=Mn#@Xuj{{>e1U{E5<A*uS_nPP#bj`37tB@JH)NyEaL_9^0<Z2`|Bd#ivO4UkRL1B
zBsn1}5-bpHmiAHdnAPl8j3!2>;*`R3#$u+r><3v1gU-xk7t&jCP4p#^DTXWoSCqQ*
z7W3kGaopUA<KyxZAE%so3}Fg+3ph*{(DV2Yc>~Q5T~CAeMwd*^SR9(^X4zzJHhrDd
zhnX7IDHb=4613(h2ddPo%#%GsT}f)fAQK7q<*-F(QFB^Tdh4Rrk~)5sdFA2qqSCU`
z=Ov%ZJge2~b6ZkcZ#KVZ-`6{RWO&L!)P}BMzf#hbbyirbd0E9y>AkA5_9tyWb&1kG
z8CS+v`f=t{Dh3@grb(ajkB-;$K5v~<>sPh;kLn**$-?}71$xDiC647Us=GSA@S3T~
z%IghIn<v<=@|zvLX(O`j&yHV*R-ZV0TKVYe9XHk`t@4eSTK3XC+DcuoR;3JkH1woK
zD}OvgkfNP<KBo5Oyo<>Bp7Rx#hOREVn;umaTa>ao_k8J_8lzr&$#Zt5X0q81hZWwo
z3tjwYE?pgRA);o@&ItN)*F_%l&)d(QF0W-Sw*r++ruU^bVdZ<WM^ciL;^I!dd>a+<
zxZ}~1Cmzu@(aJAd6Kqqr=9*OnwZwMEPe$NtweLF|_r9^DJM8bK8Had)YY#j%>&TgX
z&v)$I9JeMbWNOLvMOHrEj(d%0a)zRY{>`<#ADsB5F@+C*-43`Kc@8<#aW3p4>$>y~
z|4G}sh)lJr_wDY(ONBb@eFiJ+yxgRI^}(Mv?A{){BV`9~tNO<Fbq~W9FRpMew>Hof
z$t6kG4X<k&`@i<yGOVqo`vZ+Ygg~(1?p~xw6)6Qu3pJqbE~j>S>fUqe?n<3gms2;S
zP^7rKyCgt>KnNkZze)eky`S&*xBCguo!K+X*P2-~d!^Ch^gf?|O?lLO1IO>Kox9Dt
z7xmEK$?KOh-!**wn3(fRoI|fz$X_WdR-R3dRhd1qd7bP)@YcH3mv_kyjN1Kblfmk!
z<;NFA&$Eu`nmpjY#Cy40iCfHwKJD{O6EgqI+5c<VJ#zNI?8F}z)vx;A$cx_bVCoa;
zE00glzv;xLCyuEvW|x>1Tj@E4g|6IqXTR328ymCNY+H3?*{%hTBXz>dC%q5x54`U1
zg*8X?Z`;?xdtW1-Oo~>%sdmTkk@3rhr<zZXy!!HP*%z;GkK>XP?kB6|tJLVU>Nbt9
z%jS10#T%UGUY4{_>_=q*+MuOPQ$_1icD>!)w&K6}&U0@?L`-W5RStSJuGmA(Wu5EP
z5y#Xn7GM7~lrS}E>n#HH&J!0dx7~Od{olhwPuIO>zc2ZkmUJQ8qjH4&vXzF*4O4rQ
z_hVNt%3kBMuqi5G-uWfFmh6}j=y%Tbv_rk~AGWi#UP#M27S-)8rxwo3pPYLu**?xJ
zxh*q0_1bs$xCz-N6-P_rbD5=c>%=vlmF<mEk!LThIlGNP)m1$-SluvLH0WuxM0uL7
zi-}`t)cQZO)>+LuJcAW!HEs8#yMf$5eP2u0`zARir%YUF&vT<=|I9Ya>Ag=m*mEfO
ziOLu2hl5uF?%F)Jdj9Qo;;;02^OmBHM+^h|S<dacFN`AnBny`=y*q32?5~SG7B)}c
z5xmg1VWjy8j(v;KId-s+RwKx-&G?v+mg$mK^|k(;+E<^SaX*LRXurH`7VwXC%xmba
zb8VQ^xT7mjT1g2Joog!QKaobtjk~-W`rBjGCk1v!X!^HWzIR-(?Ax}@YyY079^n=7
zPxzfs*~B>$zKqTGV7dg1s`U9s&VSz+qn@-YUhgR7VESR5dz<fyuiU%k^04hm%F{O=
z))s7PxSKz*NMH8Wc82W^!`06GIj8=dzj0^{W8K^3f6k#zoiX#jnGKVoynk3c(R85w
zBaoC|%W2E?F9^;_`bK-Y>DiBuv*NeL=ca^|sW+ag-d`43!>`*~^R6zV`w6vGT+--L
zO=*4Bdr@?oKfn1>r=!}(iA$Ha1=pJSIxJedW3T=nP75wahDPO16HYM=nKJR%_|0C+
zJoos@e6*#eAAFzZXHLn|M|LLUc<V!ryJN0a-AsM3^51EXdGDKwYFZ|g4;6hEX`0V>
zcCldF{hDaKc*hFC?B*GTGY&?sobzPf)!9cwdR!+OQ&sa=63W4juENBul?AN4<Zrr<
z>Cb0>UzR%f%RH;DgkNb{WziTST;0d(NN;x^vSP2J=r&&}*;r%QsUw_(JoC=R@Txh&
zlg>)>nkT2)#yQw7Ro{Mi-StKOb0TNYm=P7`JSi}^z%R<%&bh|xiswqkpQ$zP=H~bm
zn?0o*^gj6Z4kzZ@)xDP{Mf=CheEQ$3{^W?Vgo3*{yINJa(@l-^V)gW0)n-mzv3~yA
z@JZoX^SG;gSJuovJjF8LgWGA#YuZYz4>HS+F{Rh@R^{JJ-T6N8dBFRViSooZ3AJf2
z^Bf8u<*q5OY-?-pt^3?u+s_;vZb~lRRr5x4O16IRl$b3J?=BGa+r%&GUUhfkV;9w#
z6Sr7z(pt`&yJFV68O-o=lbV8?1Fm|1GdpPe&&Uei#@rK6{o>YS_r8z1P<VRxooDxs
zTp76hF`5_C_juvU&xxChx^g;xsaKoI#hQzFU-j2JM)`$>(Wj4{5j8`7X472dmBy>a
zEciaTepH3aJ=+x)9!7<%DP3+g(~FIB)}<uHPyKl$qwr_Pm+5i)GEEE8v-9%SRvm5J
zRIOfeq*k-dvT1cidim?-;EtnR=3>fV9iLlc(tpgSch#eL9Zvt61Wqp6n!A;;V#+-I
zIc`w})9j}(C(#2hkI5gA=o0E+qpX`Z;Zsf8n)2w>rpH@vvZ51jd_UWB{_LGEG5wEB
zpZLCC`@<n&C@!>6CLE%$7)vx(S_Zp)@~HMP3lYt7U(&H+>I&NYml2-R^1`Bn)yJrL
zmN<oK?dV<FJd^*elV4Mt{UIYDz4iOaPh*qP^8E9xQ!-Ls<c-VUllC-CSdq)0)~r^-
zEz@tA)wx4>W$-gyN0!pnI6TU;YRSe(Kc^Op3*pcH%-+0U>HOL9NZE|!Fk#5HpnCs*
zN3on%J5@Lx(zI?om&q*u*mx>0{>_>DzV|b3a4vkfV)gJ$j3heYF7pZZ745(5PpY}!
zo9B1<v}^XLu%b*#oQuW;PY#>$SA@&_J4>pUd9FITR%1=xoT<Tm{+#i9?JlZ6Vx%(v
zk}j((PxeWzOW}Q1e!b)4`?$3KX5R06yzkqRgu+iBzAnsGuiRI_|5Z~I-+rYxPkf$w
zm-7#^U|^qok#o`FscVD6M|vg%B(62sZnl2WBFj1NA|0pIPks{oB_L~zs+W=Xmr*>Y
zB+A9=bM^oAEN|J6bm-;v=N%6^&c&QHyFdKw*G;p_A8)DtyZ1%n>*&vgzcv@$EBn~M
z=!;_9<5^kJN0yJN_B#;veNOB`y#+%{4A*>EzAmzSaz^M~za%HR`8l&LooaDp&D_fP
z+WEzMleWj7NLGp;@%qM#dtVF^hQ1bl^G>^)ZSm_%>ZoFkmcy7_GlbQ{;Yxo}hGfqf
z&W73E&B395ads!nT*uyCvU<JK!hO>_ri#N1f@%VdLgr7@9(&y3tW}wLpqgjD&A@ra
z7+HV0N#ew$*FUYkntTXLU?wHLbiF<pZTfb}m(ovm->)XWPo1CClK#1Nyy&`2Ln)Xu
zS|dfL!s57l*mzg}l@k^Pubs#qe{rm3;MicvxH`9Z2R92V%@53Ml}D;^a>tHOb<VX)
z<)@Ix5|j?-w<rA_$4|}8Y0ey%Wm&qc?m)Fs*`7KTAqVTF2Zm46r>Z>BwXn@`(Y588
z>~%Obam9>9lOB(Y@;yD_qVM5x4nglGPZ{HC`BJ~jSkI_Stwl*+Y2wh_W=_e|67$NE
zBJ+&<DTjW(`nKpj^~;Y$jpStsyA#q<9Mk)M337a@PPJ+YespPw^oMFFmsOnfdn{_K
zcvkamTpe~h^W0;6Mvjhgue7^kWo6Z1e#|7-=#JiXj=`|7|Ax43Km(_i``hiC8q3Rz
z7FAd^cr+fYwW@Zj;#B%qr?fok)f2n-8FrWUJe1b48+gz3t9jQo-{~E+S?(6)G0xS=
zCDtX*b@@m`pH45;5eIC1ELU0b&F&b;wWg>^l^+e?5GV9!^;EXkH0-XsSUbJ)L9tQc
z^TMBnKMVF0q!o1JSLV4F^2*QDe5q5bm(;5^M|4c;sS)3k&ca&VKa@cwbCn)$gXT%i
zuj)6|2Q>4vE44@P&geYjm8*wyE^$tCFL3v8RM>}@V)}BqfovPzh3!53TKsNc9!gdI
z>>KJ^G4Mq^S^Te9D6t;y9=;*}LC<6bss?f=YM5)eYRB{T>$6O3%&wU1GW676t9Mt=
zMZZALR%e7pu1YrJhte11S>{ldN=*kw_e|{a>U8eR5{NoBwZ=65s9)Uhxp86BgC^}}
zgXRrQdzyZ>T<^%}OzgJ7s?69vz5dUGPO?!{4$XnqMLUT1Pn5BeRi1Hnaz=7KaY9wM
zt3FkYRC}RzNYz5+6l*nW4$GIhMfo=U4t0dwc6gOEb;w4-5$E=Y_MaA&^qms@8hAZu
zhP6d83T8&pJ}UpgUah(d?~h)m{zO|zzsBH<!8QHUdIxo<>i(fOUVpOgDb3@mA?!k?
zHZxJVo)#*z9^5S|?se`}>2c`(A{cJp(yH0Ar=_&@LYt^nrInBK+;7^S@LzQv=*|`v
z^mO;$#f&sg5;$BZdm!IL>7i!RS1Cs`N}2lXU+ieM6MHnfo1Lu^3~V;3$XRokOBjb3
zZp_`xEesQ-5b9Jp8*kWg#X9MG*)Uc}c#=Cqsl)T|&Nf?Fth7oxO|C<q&uCyKFq;|o
z88)n~92qxD!(Ky`Yp<Hh5vo4m?ogl2oy7UfoUWXJcUSO~Pt&hcilv@|{Qi~w=lT;w
zmc8q{==_GZkhX%hUG4hq>syyMZ*0<N_H4E5$QIaisdrU&ZtY&xw`}m}aFe`_vX&Z8
z%TgYpvYYFqS*7`xrmbd<W{kF;?t497y=2}~%{A&5xQDrHZn!F!-J~=~nJ;r0-XYZ)
z`Yz`7r}eJs85a8YWcGyg#`j+As}KbbB#NI({D)`a-4=d~$E?3qDmkOo7O5}OnxRvx
zyH_`r=cJvebxZpl?-egb>nHbbm2J$4jBtiBqlexkua#WK?7y(rr0<_znlO`pxOJeZ
zv+3^^No!YYa0|Dos3Ea2wq<jNXXo0kt6lY7&cg0q!GN7K3hQ9)$R;$>nw6Ka?N$F$
z^Mm|sR^6+nr9MyNt%klv2G>hX!13T@s`#jMvIZE{N`tfr>Hyvswh3>jd?<M^*oZiS
zKH$_}A`0wZF;Fj7mn@a|Nd%JV!z}6>C12)s)^xUwN(V>IP0%XWq3H$b=J4d&hj>Q1
zH}#(C?bErY`9{r`vtGqiMZ~gWWKy3?RmED6tg*eB!uOr)+HITHG<4J(H*Rj)-{jSJ
zx8A6JO+#cewSBrEyvqvt;HI82{eMZG$nvS%@wSt*N<Iujc7W<4?k_H#%TkL4PA9n)
z+-GWAIDfJOSpP7uGuX-kS~8_ec47DtY6dt-=8B*8UlHvTIrQrdtRHyaAA`Tg4t$Xu
zkx}qInLSiLn!B=?ZLZ#<wUalB_eJ{}@3`JqBTKU^vo$7f42E@g0+aLlTl7w8cdKn?
z{fqaGOr(`io(;bhYxPd;+}N?ao!36DP1by|L9I5S>O_@S&9OS;h7k<`^*`!#8=g1s
z>S*ce?s?v)+;1+99tx#cDI2oSbMjR`sWo!HY1n8V)M4sr=-tt|kD}OPHElGjHCCzL
zQ&nL{DKDozpxl*Dk~Iz85|0&~=xIPn<cO}vo%)^61(AZ$f^7op&hoC_p5TE8Lz`u1
z<$EdH={wjXG;8(N7+o}4V))XKZhFh=q5W9L{dN|XJ|<m;^No%gml!?QTc_c~+ChCh
zY#_~*#12S=JKM)M7*+RIjI30uT2pzq>`u|y{H^(Kihh*;QR9b`VXqsOH*Lpj3~vnt
z%GS|FDJ@q1%=A>9sI^tU&{)rOhcU&_#^9yl0uvZM6R|;+cD7nQyOtHgUdsAg$x-Gr
z&>-B`rQKaFln70`X0<mp={9yX{nfs*OSLCb__}ku;6OKjz>ZqSdawFY^*>HF$~<p0
zi?-Kxx#axG;ho(X`<>3#?iM5c+>9OWn<pBi>4>$bYpviuWz-J|yZNoB8t2waYqBbK
z6&Yl|PhIk}^yi(_k?H;!ni>7+>oSk#2A9mO`Pl5rzuLV(G-+rQErT7TS)g~?_?+30
zMXfd4KH5>m+0(hp(aL_B^=k7}<5GPe-e>M?))MLliC$k~=RliklU=Q}qPsMuSW@_~
zz^0(DKw7k;j8=W8zOcEaJ*msQf0}FpOQOv)bFl5P|K#-F$PfNw!tEkmqT*-zO}icD
z8+Kyqj%g{OTgN?h_A%d~_ouo#VE@^@zu|cyH*H43#BX;$_`HsP_T!1>V`hwTOxV+=
z*Ka?5kNuKTk^8hPx0ccNPoEMkgL~27hGnEfnoF8{w$DtzqKVYttO@SpCEjP;GhEI%
zj<@+@^i_R@(%PZ=-s61tCcEnOCDyqGX*r3J@hjqN;tLWUC93}%ldO~)o{?R^ukqqv
z6YU;aAb+H^PQBBzU~J6PjF}U{JSM-7y18c0_PAZO+gvxMF8?)0KjKV8&h)ei9A_oH
zZOkv?3oYIy(=!$&+<aT`u<%CM<*GAwM=5`o9=E&r?AqlUwRfgG=YJje#VWhs=012$
zdAnw*d4Wf3z|<+8Gt_5iEnqL*w*2by+{FR2Ux%3mwvSoso@Je;6Ux{(AZ$BPEiH`B
z=uWWvSn$&Isn26^Okm8_n1IJ=Ppe)Te$GxTOdpZcSoFN1N_1Q4j_O;@zw|bnzjHMY
z{d?v1zxp;VUAB3-#a7vY%loqae7*MIvV(JJGaiLIPYxI>GJB|eNW^RHDp(c!<8{PS
zqkA3~XP+2*vi#IZ%#W3i?7bq5S^awKr?o#z3Xe5>5?U*5*Z*Xv>V9F&K**^`<9TUw
zz31&*xL|SQg5^=vFt-02_i0vF^`vU&=qLJ{T0WIM%PPh!-tpVc&ksLN{-FE*$-Ai^
zu6=nQe<>w6d#Jdk8XKlstveeB0+l>87VEz=GBlcJ_|7cBZC&X7`OyoMr?&*p4qH0^
z{;GXzELO<ob%wng|I<sw)7Hhn(3d_a+SHxftX`O!KzV=knc&`kw^rTLeT-KPUHj{z
zMRdyR>2Z%znsR1VrnH{u^QW)kb(mA_j9mssFA43KSr+lf^d(_YA*B<p`rq|CIeN2`
zK%dR(8>;CU+YwuDS{9l!KFvJ&XVR{uh@Tfzmu0NVX)XLwv9Hd&VQa(lCjE{{-N$-%
z^+X7}h2Q(uOB^Y-^Z?pL*?%$%_Hm2Wala<73OYXKjPKkT$5tF$uetu-T9p-gb0$w=
zguD&h;Q32mK#3Bs>9cB?k$v*x=6~1R+Hk4%{Fy70Z+Kqaf3EnP=w|8DDIY&2lw`lG
zjp&(9U9bM#B*mfBqi*cgNdr+4iy{_h&A&S9#Wa3MtH0aW86$P=67@E-4odR*<u#T?
zkF!M?W$CqP+mrp1CjPvbHa)wqu&6S$ab4T%4i!O;V58uo;5zoOtBD`<`}X7gS=e`B
zCe0YSApgdY=s3FC1v^jK?0d$0#svF0UMu*^DwoifzMbC^w$X3rXtsNeQ9MN^Z0>Fo
z*jCaK#V_*is$UDcn0EEvz4S-9civtvxwGVz&3FBz3mFv^w%xf@bB#Hs*$yGDEcd^C
zwgl~s9I=qQuzc2su)imGjD7F@)NR=2vH_i&N|W`eH7_qS%?<peot%(-;8#Y5NqTH*
zMp{-@ND;p(yM^Da)4xt+Dop826Yd%i4wel#4s4ZfqbDd?P);D9UpBmm=BnlBGBt2n
zaL3q2pPK<wrml<lD>8GA@q*S_!BcuBXpavZk)^$1D5&?1$hXJ4er<L`oa28#9;Dvb
z5Z(CL`UU&(tB1#*pZ+3ElqM&pZz}%Wbgl1*QmO7wn><IUeXZjfuha=<Q>fw7rg(+;
z1oVvY_PXLyZn2hU#GXUBJJ8Z;-twh-sG$2-amvT^`#F2^YVu7=j#L`fr8dR#n|mqZ
zAj#$7U(}=Y1+*E|3ThE5$2iCi$S=_@(v@g#*m*L8I!9@`>U@KfHY*(pY!a>3J0y6V
z_u={M^A`AQ@-ZH%?0VGY4+m|NZ>lS4uPE>22gM5n9Sv71eG5x6(o;>-CZ%d88pTrM
z-BLTU)AQEndl!8xe_MCHUA=!gwUo(JnX9I+bI^ooQD@d``q;eM@~jo#($C_J@p+!V
zY9gXX6~+#l#;}PvuJ2`!hfvUE-Zj1JQa8Uxy+3epK;j`eg8f5RhP>p*sVDIs?5R|$
zd^g^e_Er8;wr0o%^+0||en@;J>XKq{qKMhISp1G^#BO7UGU=@EYLj`v`lk%FO#;kD
znRi(P*(NzwIA~k;>qqkfbb@r(@Y>XWv3+P<$=TkgUH|Y?TGHwcReh)os1B_uuUcFF
zN6Cqj?23Z=Pi_4IlddIQw}g-S=Si(l;l_m)LFXw`ne{9sl_%_GrV``0^52XSW)N#5
z(+NAUo+2kgWqo6sup}&7wlSNf(!$YHHDIs78|6CiUhpXDb9w*pM~O_dqG!09+dWTE
z(dORzu=QZuk+#uo+gf=o<;_3ZKlBz12P&;*II^ax&D2?KqHevzA;smm+X;^tug1~#
z<Elp$xn<jbw?1#hwD{LBNYk0QWZ1Xw6<?+0ZQZo;4S8cS{zxy*{F1dHBQ>oyO*eB*
z-j>p}wLvYHJJh?*_eDsr(`T@@aE@@xHQjZR^(96k<L8Eny1RIFykK2lT~*#94PVt#
zRuO~2uv8vL|3Rsd?UUJ}ij$)}e7HwqFgQp2MBE{E67%|Z^==f_b-8tZ?wH+H*pk~U
zZ)$0}+hWxIwqrcsLEzN&OSr7xOMZsqsH<;OYpiL(x4-Om$g648@o}5|ruaws-}e{z
z#`xTK+G~EyFiWpQYZCkP@Xa2J*4WCjyc6k}DQZc#VqbkZ|8;EK_=L9j`h*3^+cVb}
zxz+V{QY2I5GU_+x1<g>Szb!`FOt)`w)OG&mWaN0$mS%CxaEy+s);IMej#TNalr7Te
zJ|uX_R}sVtHVY2&U-0L5RrM72Z4{;T`$)2e1@ifnKjfanA(Dk+O>xCQp6G=TO3?9!
zZ_vG<Cst_Q)x+m?hKsJyukhBIFSim|=Ga{vdD?G6aLlBr$r@py;r~vpnA$eUY4msN
z>pB)1pSibL2@=&-e&N4A4Zqxb+yCm#vo7o%cDTRovFmG>_g_Db_{K^8v(T-jY%qdV
zrM^Nt!(guM56^SHAroX1v;xmh81SzepW-8P>9FiIn5Mm-bBEH~+tPBnI=HN_<aP<K
zXklJ&?qJcas<#cxTK;IC*?FedN)k&st7NFuLE*?I4%^_(>W<P){YBkDT@Qrf-i^JF
zyQ2kH`BlQzR6o6Ac0G>I>_Q!d-a*01)1^_b=UkX;Gv`gzkD2ta<Ks^_IOzweo2c$r
z_7$mCE2Yi&yz!O)6U&&ShxGf?ZZEu9c$fZ6{q6pbXJS`pZm3<+t3&&sx`+47DBb49
z$O`{0lU{|gCus#=2z=?+H`>u7&VI7lb=?Y8Z;Dz^W7FSN0j2guu?08t@8p{2wB)K3
zCs(Yh-O==h-`%%N8c54jHd4My`6W#qmQXLyJLI~Om!e6%@j|LFt-G}ohDLBm8m?30
zP&MMD{YTrw9(_R*W=&swb@|uj)r+EMThC;NfAJ4?8rCuA>}9!A>4Lo6*PmWJ_Ph7`
zj_~f=yB}{pyrg^4`DVmpt+&IU^%L*r-)WsQ{Dm`0+fHwziOkW_k3BPMp4Oa0)Bl{p
znJk;68x%7>b7YbIMH45@9;#b+Pqjz><BXv6xO739UXsK2$gfut5;Am(&8u4Lwc5}0
zW)6ob|H)WIy)P}4xiKTuPN=M*7Kr!s342%cFOYnZl=YqLP8s6qZ}BMcVLFG{zV{p%
zZnsEp`LpGdmtC0uY({2Sc=(>66)p`L8MH`xzwA?MdfL4=+A-|A?l)VnZM&v)Y0XKW
zV@ek`J=pg8#0SZDwXAOqF@s-KcIvJ*Jz)D~q*sW+Jnt2wmU}E1H)}~GElNC-J$Yb^
zl9R~fj^;et!ETe<l!AR3TT;)b7$@(KPx!q0b5nwP)~n*5RkNDj3yz9jNe|Em8C2!l
zR5ki2wyx??#w)2>|L(qPeJA?t`z8x}+b_3vh+MRuIoghRWx3bvliRJyW9N4+9=9}q
z>7#{RQ3Eq>OfL!==W>WQk~2Z|D?O=oIMw@Y+0*!c&pljsZ|QB<D=*GOpZWKC*b|?3
zPd`6Th|Y^@^_7lbE#P+Xip;w_T7nlvjhH(%a^2MU@WV59MI4{16?n=k$05<olozPP
z7fS1Plmz9uWiCir6u<8Ci}xL$wk2-JzEb2~+EsD7(WCpnp-IXioW&d;=CJZ^PK&04
z#(jnb-WmRa7D{^{`z-12yCi@+P>!`V8hv!ckb#3?#>l(Vw#-YMvvKy8Sq0O=Lq-La
zP5k7eYr0M4Dl184{ZLN1Qmob+i&rCFjD2GFC@T8xg|R2}&Mv;Q?PdOZ-EY;Y-KCdX
z=J(diAFH_;O>tBk&zSloQfbDwu)?WxBFD@=6eXLgJHgd$(5ziYjmx54?OEAWT@jQ2
zEG;7;@%!s<CUJ|CHF7+Pb`*6L#*{s2H0|9-<#KLuILZ`CDQ!RdK4(5_G{cbfmK94k
z9CnakMXvXXuutx9_Gr{AuV9mV+!zycU)$N|7Oz^QJEv{Ru%C|~H)KrE0{6*A13bE3
zkV;hB<TT1xn{WBC)bDy9)1S(&s2zE9B<A|TS69BQ`c@GenergNvPMNPSHfT)HB@u4
zA2VhAAl_gl^wkW#F{3;pAY94!w$ldFHiKjOU7EH^2YMSC{mTC-7|hv}<(QtH%1pnT
z>s;Jj>{F~$?%C)pWDXB0=`!l&XT%o=|5KW$;h}X_C4eTEt(8xqo>ZE~7)do3uaR8V
z+UeooyTsXxcU`;2#V&m7Le~Y2)3*EV7+LO_=szJS(`%OTQPoK*L5#^FQI%fSk7UcF
z@Wk5*3qNQ6`~Awe%Qqk0`&{@VJRu@(WzzB-VYOP9NSefCXepb1btrQS8F|7hf6S7A
zb3u24Zuo0^r`cPWoic7WD&*Z#e$aQXb#A?KO?Ktn@(slo^1O4zc`YS%mAy4j8uquT
zci$GB8(cRuVaNrucx&Yjb|kY(mN~F;V5M{>?TNC6lDD*2XxZgGOx0WM&~96$YsCGb
zKh5<{!0wQ<fy{Az-t18U<1>B#@V2q1o9;6TG4NNf81C$dt;sK^SL`W2o$rxU`99`N
z>c@rgl}V$2<|J)Tu1<F-tgAlQ@>yUcIwf~dnW<@^BhZgF-EGt5bja<Kd*X=o4s@Gw
zR^KcW&F2`6)$(BO9}XYz?t3C!CYaqaz3xj@VpT|ON4>PctMOaYlD0fSq;OR)>Vk--
zidCg@*<#r`$(4aagZE_KG#Pcb+(>$R&{DdX`H$Wfi;HG$`gQshmYHsb<Gu&f<85#~
zBU3&4y@I@>T&t}^jBRvgscxm->3_-BXbi1(sGybH$}al(GQKB4HTg_hFlxT=a>{eJ
z6>YCFXkISx>Wd!or@dq>R{5cRPxqSf5%WyTQ&vkY>&zNW^i86SJq&re67?-CO={dw
zn|R)UnJB%xqWx2oTH~~4llHZIPrhD<az`5fTi4y5vwejkw}JYB^ZlK@r9DUb9`qmX
z=l4bTx%C@~4F<Ok$|doX8>-t3%Pl^cWf(bG6uYetFrEA|v^3Cr{O2)ieJcZhOz8I6
z?7%ja>0DQPK)*Ry-=)@^T`4N`&f4>n7Q5i<#qT(EmoOYx82dKfF2x|1QU0a&TI2gx
zuddpG6Vyj4b2JBdwt9aV`k7s_QnmeHtBv<sF)gFal#O%s4r;q-_^DO0?$9Pk?+?&K
z!rr>xSG^U&MV)~iYugJt0y>{`KNnhd*YZ!d%iBi^vQYcaU9hynvHf9ts6b2T*@rc}
zfe}OR=w|Bs^z)6kn5I~2I(zsu`iTSA1}6BO7$+L{%x||}^C+JY0;`ios=UjpsY-Q2
zxxz)QdUaYAZwfbOd!`>qYf1ehH6V3D>ipD!w3MvA!sQi_HEQ*0&3pL#KA!BYvXuRf
zW5GSG{m{_doM)+RQD*95qG|HPq{7(QV7qpjS|EEs*_6(v*h;Ss{MUO|nA*Ly%S2$=
z@veP!M}5b}j?wLLt(4XYZIS#Zowb7McDJ_L_SVkOUgLo;gX<(uB;#dHOo8TVgB0T{
zX78;}I%RvTAN_IsHecTOv~e8Y5?GRtBS*WevpsKmP`8!)kx?!^-ILk2zMfOTDy+)h
zmN6;KC;4d7x+H#*bMlL{u<VY4u+r<5U3GaaQo&M@hAdI}lL{5*Z1?M$8o8UjvV39v
zm-TF`V#}wNJ{EIK*BX7*pQ+QU5v}@>HAiWje3#@_e@9P8r)P&>>+B}A`Z+ZXRZ3N6
zl^GS!E1j!1*TmO$*9)2sv^5B3^jZz>mle{Vvy;?!>6jWFvaq%<adGfi;q}Pp+^GAb
zIHMnr)*P+kUGBEh(bGz9bX=!GJ&Wy58$W2<UER`L`>>+A<YeLZ+}O;-Uz^e{rM3Ms
z&7|Zk${$mFwR}O%&Bhb$hQc=k*N1=6%9!?Q$8o}Txe>+mp;@sx$8wC-Pb*KWWfnV4
zn+*N+oOow6UvM*3vXm!Mgp%3)cY98DUF5H88`+F`f9;}buPT=+<LdXw=KpH=)WmP$
zb=>KS5`7qQrZHI*wG*1Nb*qfNS}e93>3qr6#ogDv%<ZMyUvBGMnJ$42*4Aw%tMnFV
zc5?1A3Mfk?hkKv#uQ%(|tJZ9-TwZ2eyuCm@UnSqApt~@$IJ%VVYI<5Xxp_jzgl?v2
z>tMm~TAC-bl`}=djkis&&)}C)sL3tUzs-J{O)}FlH87rNAk&G{+OOWj5i;-7mE@+9
zNRf{4FM(crPIGa?;kwr~_SK`RR#l}`tJJkL*teW%7YiEkj+{xuE=t8JaT+B$6AcfW
zx>-%MA98kf+vx7$?(9}OVwKBLr(F9w>o~J=gE;LtwH{^!EmkV)pC}Y`c(v#>Y^&K>
z$t@pOYFH9qY+phz9Z^<NZdS#u-O(_s<#tC`7r*z>z>%S7`B(ZQ=48$uu2dsiYYDGS
z$40L}uUqf3-VfcQI!xYTtv@tmYHpmFtbddr(T>X_rMiP<qSrm?UEu=vj=62GT4ptG
zZqjL5-E^#ZcdJ?lrE|2fvQKZYeprvbhV@YGpjNAHo8cwXT8pjLR<=@GhTS|{wvDdU
zesfC`TZ1IbpNv(TnfK@(@~4tX{mXjHx_0o#wU2Gx*qqyVzk%Ce((t(<y75s{ese&p
zO1n4zL+5&-V_$jy{y|SEL*9Ya{PE0f>=hhGwM=BYmT78gP0^BSF4lakk*WTbJ5g;r
zXFi+Fe5JIMhAJ|{_l7(Mi~7Iy)%ExZ^Sa)2<_Ww6J$zF^ydbud1v~V(S1h_IzA%(6
zGo$^Z+`zI@4O5?|HBKi_ugBoB(Oct4lbt3`CYy{MjAj`y_5S2N(7eJO!@0+7rF&Cu
zOPvS#qWGRmU61%C?K-WynmwB4H8wS5H5fEDG+H<RYWdVA>o@`W4E2`x*ABKzYvnI#
z%ax_f#VY$$r*VtaX_^_D0XSuTSu0%Yp5`G94fVBZPdK$~F|$T_AHAC5D~li6Jvep1
zQRL9OMA+8#p|e_WOmJ5~N5xcWSFEtI*S-I<_~6h^*%fM<k_IcBGm@L4!PRE#Jk_nz
z`=LMCV2MGaevaN$-EF*CT3n6$*tgllJga0+tr`9*Nf^lM8xqdy>f;Z#uWj>a^>4Y+
zys<gG`B96o^?dtQ{;|&T?mv5H_HP)xBz+*iK#Nf3GlNv7s+x0OsJCl;)zs1wXwK98
zN8^C{Om4lZ4abdb!DJ}c&=M)}s8eb<bac>0Eb1Q=d5E6(ZR$IWYFw*D*8S4{?g7SN
zu;j~7_;9=IFG@Fc9{rJ0xw0D5ljXx6W*<{&QL*5R;BYy4Dw|cBkdvFm+Qi((Sgt%;
zNsq2flc2Usi>!NCER_#w4NXR!G<}@Va~Bs6>>ltMFdVQLm^83=AOZV!SBTrhhX$P_
zRg#B8yHLx_L#BaRB&R8!)JE*ZXX2zvIo(((Olhgo8l_dZW-3`Lb<$tqE%0>u9n`1d
z(b7=6LzVgzbuYxI)o}!M7+CUevV*c&IPam4n$PlKO_>MkO5Ku`pvKf>yvN@VRfwdN
z1=JVR4pg~{rm^S|^zAs+@d)Qf(&*S@O+P{pqRVk6V>OLQJBE`rn<+gw_eYT*k*Q)Q
zxfRaC?2)?QJj}hJHK-&#5*3+!hr;mp^`T!w!$V`GNAVW=5bS@|!0C^-GF_YpLS-)0
zm{>^tL3N;=qP5V*(s$DDqt<gD-3S$=S=a^rT~R@rO;19NwKnQrROC)WEsPrZMmblW
zDm#VU@myKn@blq^!(WCgheg9`z-k*#Eith7dK)T43}Tn-ahxU-pu)~*oY}gBv5HVH
zyO3Id(;XjB&)^^`ExF-M=pvlhcuLs=irJ#VM*(WiY(%vP8@V#-&LrYg#6i4IW0cHR
zrX!=t29UWHp`y2vOam3YC(2f$I@dF_`h^n|IkFy^Gpa=G$9biDs9w{9vxzG}+s8Ns
z)=9BMl`TK&1nisMiZdT4aSG@hejcTsrC!AUms5kOUQ~C~_h>`~mTjoxVnC6gu0;$g
zc!bIWQIld0&Zo^ot(Og`5_1~WSni_!O9AQ&IN|)u9Gu2Hj0!4gs0dL{DN)qE%0Y#u
zZ>apU3A?TJfR_N3#_yq~Mi?s0uu+4n5EVw=%93QQC<Z?)6U!tx1!sYCaKXUf5L*5~
z9WIGn8P(Xw<HXQ<oPOJd(|IRR<KzQQ|2#&GiFK&;5r|#thS<wb!^z44oR-T$b&w<S
z?Wj(57Em4ptS7O~lz|#Z)%cc+dNGU8%PCYucz}MAa3b&_{&x^{bS~oL%x(0#3VoUZ
z_blLk0zI5St*iU;+xWd3CsY1}+?~Lwfv4zm0OycM75VXaViane*rVFVM99cyROX1r
zNu0^314W^<pw3MvPENI=?nxrf$y`A7qcN!XWh@_ux>?bveUgB)hb6$4j~Y^P)Ha#`
zOkd*kOd0mMrvk$Jz<MXn(jAACBvXC@$`v?|)q#@>pMcwA)K!VY?)Y+4&bk2p_@gpO
zC~6B`K+jcBH&xU>p@I74IN{WUswvv2!xRaQO~+|RTU0=y<Gh~-YLff~+>hdfggq+J
z)S`08WAOM6PVwcU#u!Q)LKa@4&kLwA@)A^*px%!ys;}t)!zNTIYLrXm7O3TO5YRrw
z>7OmYUKduU4tNjYB%&c&GEg_Y6cjiEO72E=me1e_f)1QJYQcHOkD%!rR6S}!br@$*
z)DLHxXrNvePDAE_GWIy_bQCpA&f>I&J$kBD)JCEMig278+=3b}I;b&o2~~pr2J97b
zO;iz_iL*%KDO#xfl7i}D8Q_8wFgynCp9Y`pQ1vASCy;L78yZe7jYhu`v^#+F6W);0
z5S)U{1jW>Vi5wjI4>j$cgEF0{2D4RBe`N!z$}lL|sCJi$nsin;=@$+Dyaz39DRrpZ
zL+W72<e?bxHFV)FAlJd!%4dKt31>S^ak60%YS@g$`K=_t9f3+R4<H3@Xtx3UF+?9v
zAj?YuQ7WwDLYyXC1sUrEr8cAWTu6F0YSk@;bS#37B|-D&pia&?oB(yh39IX%)^?oM
zYecU{A+x!Fa67p367n>O(ha#@jOzk86a*c3PKiN39;jk<73WabLjx(Ouk;l4iT;6<
zSAnuC<qI)F98LhPSLn(@@aHk03&EK8Vee)^mlAO<EebW4ZlRAqQ1hx(Zi8MfL1zC#
z1-$Rr^)XRC3lx{2^;uBs1SC-l=jTpCQ}094#-I+`WwafEx+(`@cdSv3Ya4nQqo`na
z22`O_;&D!m)aR;_k3j2Bum-0AcMjUmf$g{ixo}1$lMA4F7*4}p1$2w?M7`V-TsseY
zxE7o*!AaXm@II&H-B_!AiJDqpFoGAbzl8HMJE2>ykhbrz-)C@U=ojiW1)x&b6j(hq
zV1E&uumc<mVWo|5di)abSp#d*0j=Fap;DqT<^r7L70VZ+?|j(BGmwb?<kRI=IPbO}
zx;_ouxdU!$AP*9O3NmcyYyu?ZJx;OLfD_^1&tuT!FwV$%p@yIeYA;pc3~Cq7!A^qx
zSctPr#=yH6m6@(VUw*)rMB)_kLugwR_>cxG9ziJu6o#k=v<SAr2-wF!J6Gd0%_X@L
z>d8rD{?Ir}Sm-0rEmg`BR1MNa)vHl}T$QpLXAILpMJLpHTL&B`K@#&&_5CF*d^RM7
zV7UtQv96-_)I!kD7AF_E)bqH11s3iIG};4t){e@W6r9K#Px%J#F&c2cfrZ<HvsfvR
zIeW?{jH!v*R8wICLm`6&(391uS+xh0qJy@}A+<q}ldq6jYt&K-Ks84tSh!2DluIFn
z_aPB^&{qadD_f(IRHZxwni-FCL(3Gsj04T@%NgK#0;LBvkUqgCRDr)CfaEK*<T~ir
zL#e`V57dA=0F2H-GZ#Xp_>@M#GYxk3CcM^Ocsn*Adj&3Eh1Ta|tnsLAG##=$B%cdj
zzJ+X!hb>qK+c6gUnToSPe*)@7(2}b-f7~bQM9$+Ryz&G{&q4Go1eKf-#c<&Bo<XZt
zf)BgEc@{MJ1bodI*la)escKY;`Uvf(P(DCjk3%lnasLeb*Hh@TD{3LV0zDqV!#Sbn
zm%#KT_+v*YhW$PRd)qDdLBFryOa6sSh|&5ObmcDe-XFSs3p%(6x|xl8k)ZbjoUro1
zS<)5oLlRhl`QX|JNZbZQ6kPZ@XK*+Zwqq3}b{42bMqLkG{u{LwDU{E+a~U?YSMCn%
zUx07t(2pDz_5pP89sKz+XwPMgy95zNEcm<^K4w_%3we8j_B$y?lw|a=5z?|595M&z
zwxHIVE;REoXfhj6mckb-L=V@Yv38XAfMh=Uxr=DT1%3ZQ91;$nk_5}P2K~eVl7Ari
z&mcu>VXf~$j&DFtc;LWoRF_PJzh**zF2P<cM%2(NpAINCz;;-```m`rTLUj$jp}+n
z7<DbYW*<C3H1zK~Du4-4V{A5L&=Rz&1J7fiJO4qyRS>s0AT}6*^V|9G#9!evYoRHs
zc>ll{e8WY&(g~YX2p$jMy$^P{tkH`h>{}PUk)ow8AaI8t9E%ElDxfVC32%9zz-E}E
ztr32+a90TL$%pl&0Shj?fFVU49=r>Xbpsa>Xv;=RRcO5ozKoACYhh>m;g_hm$5!+q
z!ie4Qv81*c1Fd!NQAYbgwClko#;7c`W#S16PEU5D_a4YS74%d?{GhFXPYS3p!9dT-
z7*8GFuoR<;08J12qTqi_^rDIn4KQNS4dYQT5{IIRi^L)_JU@U@2ch4}c%F%Gsc74W
zK7{z@H`FYQrwm%)6&JWidZD5v2X`50Ed#v0xJY$KCi+lOd_(XV!X^Jdyt$wjDp;Wx
z3EKCgpCQ0P$3=y_4g-Qd^wAHDC}>4sVJTV<D&Qt}3H3Pm5D0(MY5>pCG3M`!P+SZM
z#ppv0SQr?OR63O@MjTL#NW&d91w|O3>=4EyJRoh!eJ;kNKx0J;_#_yUp}<865R$e+
zaAE*=S-49s(s}@Q3CHAsgbgkbx)PrC1D`&$mP0l*ai0UYN#EVzeIMwmr07KrFj64l
zow$U!PiV*id`#T^JtApM!5uAdT}vT@GTi9}RR||lu@68G_Zi?Bk)3XZoDiAP05l{@
z7sBo|BJ%4;dnRzy1eSEXTR;HW>{iGt6VK}aLIx<+55AN8Qb;km%fo#pdLeg(fKCZd
z>!Ly=8Bu~eo#;o1)+)GbfYu!FW&rov@C}(waQ|QU#JJmzyM3TH8$BBU9~Pvc7qGS9
zBSvd=^s9wul+XvM%E?zi%~ABB4LHb%d<7%~K7u)3@unE@ck2Pj8lg5%p-;czAY+og
zwa{7z_$z@@q?fM$gHH$8=-?@$!>#`}CJoQ)0V6^wG1@lb>I5g%(AET&Mhlef0o0Aa
zvlnA(V+Lb}UX%fMo1%(n3wr0Gr6p{NKJE(@@pA)4rUE)EL?_1Jj|e@K!M-+to0_=m
zfEb*#?nFFY49g*g6|_b)=!0lk15#3hyM>6uM3_~$;J!U@9fTds!0asp)eDsrR?r#t
zN?B3+H6C$z8OG$G7uWx5-Hbbhc!mNP0x)X`hL5p;d{-g1Edl45=)*~2N!2NQ*!T=s
z*$j-V1dBNm@Nq$@Iz&nDU}MX`OHIX?E~xt`!(5^sTpj{%Y%rsUK)me=I<f#`6L1_t
z?Cgb}-2uBL_)vqnMiHK91Lb9aQ5n?_Q{bK6C?fXH@EOGRav|g7rwp)(6?l^wU<kZl
z0N^GvPNZB6xGjL89@+^oS_!=8d-T=_T1f%BF8Id5tgIH?6hQ`iz)c=(Hv`tc1??*^
zqsRgUmBC9*aMB#TX`nADW=uR>#-MfxXf+ae@*qVLJk<{PH1RjlRN`F<G3HOuQ3z;Q
zpq4*8u`B!rxwvSh1m0TWZzA;~w2OyNdx<Et0Ca4HTxjAOAv_<c<ST+iXu%iFQOq1C
z<82dOz>U}=9-t%Ef!LT*Q0fZ&|1Df(j>Cr@5-lb^K?J(i<0lWem_pl3Km#hJ9FcZ0
z@I<jRRBCqtC7l#H!2oo`A88>j^98Psprs|+SA!~L(9SM=`oJR=;<HrDt8;;E4&*=^
zGn<8|c^Hh}L7<KaptOU$dO(`3aj7F(N>rR9Pr~ZdPdqUU$~WQZ0$fRm^5Q^A;_HZa
z)Br33^j!+OsR~Y*qNmC5$6CO37#tn~hQo+Pf<PH7+|>rI`RF|rd?nVU8~S1i7%RZ}
zJkYTjG++W2f51e1fhsgB4Xv>+0TOBi9*>1Q>Y(3#jQ5+a`p^d#$W$dHwheL=i>QNq
zONB;}nfV~}<SQT~Hko(_7f87$`lf)p_MkP<%mtVag+gX@AYp9G3X%aG>Ae<Q7sF4K
zV%~5ca1=prNi|#IT^oT3A5zc*2@oL0n~t>&Q_#~D81*YwEQlmG0#hpP>R^stj{j>y
z8*3pi7NETr?8y+gN<)PH8ITiSN4%anc;k%`$jsUla5lp>*@EvBR6!jC2U&ot0x<Xh
za^f#-@LLKBW<sXhfh~ca$R-)F54`FH=c+*wCrA|spV5#ab9`?I8Fc^;$@)ehM&w~k
zqDxdt8T!XUB%VqJmTkZy3s(oEO$X4*6*bkhA^SbhKjIHX;F=q1^qN9`Okk&Uai<kB
zt&SdvPvu~gVf0xD`0ByYbo40#^(xSN6J(W0Z8@ZY=!gImC%Q+0&MJfF#(>ELapFkG
zgdXILj7Q=cVh>DV#noZ+i8hx4^J+ju{J0Dj%@9u+ffu$4o=WkA3>22(IfA(suuw*S
z#EX(Bg~TZmJc-T$!vJ`$4N4LJ(2G8aN2x^b#50jdh{!*QF+_lw_*oLcXh1q`01we!
zyr&FM$q_4(ID&#vl|a=Y^wf(HsF3P1@Vf!Hla(PA1*8Vhz7hBkdm~4C3M8u!c$lC^
zUC@sMNJY?)CQ!B=^sC2{#Fvx5+i{QB#d`doL}!D*m5j{;7HYuN2t5!R#06w3_)8xg
z_X0i5K^<58WCA7zu0im)2Yeqw&n=3vD&b4g04vUg;1U9J4&Z9XUnFX6gJg2hw*mTa
zMo%N~(;VXwO&dT@gpw%rgTIsji6-uos8_7O;CKJ6=!e*FJ-}q7;0qI!BT=&|+G{EL
z{=Irhs7PWl5_NR}f@aXK9q^J0?&LR#1#n0JkT(L-R*XpntVDL~A^G;;jU(VA9I}B&
z8w;H=0^I80dK>y`fmT*3xKgOFb$DSUtX(o}NilkqqAyKgW(k_vLlXU<>w%E>39!8$
zxJ#@knJEw)iFYAhAswFL6RtS)lZjqhK?@Rz<MmDWO=7=6@JJ6FaK_~V=&T{@T<A#;
zMyvwOYXMgcewP8VR<tJ3)NeeAeHY+EIBJL<h}`fLK1+%b=zziskj=mxav7{WSp_D3
zjo3so56}cv)$u<S$S~n_1vDoXD|aziH+zNI*;~X4-#`Uo{mStz;d}=qcL36e;sd}t
z3>GQ?RImrvxUdMl;0c+faL}(VE)r{Y<EbL_{0q8W4af%p7Z*@iLY|3-H3RkaLC?|X
zDFA&=M1&9ssdNGrXs{Z5VBd;93Ke?Yh(5~DUjcec#mw?8=8QiSUZWYbBoPA}awEYr
z<!Dm`84yD<^+46p@O2>=!4tO5Ofj-K{<2a;AKnVzVF9>^1?hl9m7v!r@Y3(W`)u?>
zJTe{dNPs;ZG}FP8Ug&We)=3sqmIG!-$m%d8kkG0Y@HXM^Zg4?_-g4p9Dgb#4dMC4s
z-;uN-Feh?C<^xn{1K}i@kNlo35~(LzDo6X@u`{t-1Q%=2#undEK<_GOWdb~PGT=?a
zC?qB&I#CDA$vQcSx6RPr9hgr;OtJuV$0M*#K;jZ7Xn+oQ-;1^+w#-9hP>8=u6?)!<
zr-vYGLy%5GVCM*a8Dm@`p+wSkz+ZjvnarffB>;^ofIAtV%*^}1k1jycgo{w58a6%)
zV-Sm13rZ0BbSYXAZ61K<AQ8PW=xT`%@nt@cXJXlZYZVQACAGte1tpQv@A;@0Pm-vK
zTvTYDI$*T{g^bXf9&~{R3X>TGp|!4}cM?C4ITjs|596s}(45R0dlkA*JW(oiHBRA2
za{%#gn2F@~gEwT3tAq~&GC=S*KyMbHiaGkE!21$^hsxTpB^?TT(vGKy9VM|V@m3;0
zsez|((hm|D2x*@HnRf-2Y+yy^q{N#NjL58%%xyW)RAR~H@S<kG%mJ2WG@c>R4zaOP
zP@Kec8HmziAql_mziLpp8254@6Uq3d98eMY<)Q~iSQ%?Tpalw$S?%xHp*nEk0Vguo
z?FZ!Lc$Q#JW-Y`Y6W_=MElIRbX6s}Ys{wl2L$bz0p2C0$i5+!8n|?e&JQ59@a|N74
zue~s)I`HVl{Xy6WqCX??<ap?#C8V6pAYNb{?mAY|-U8-)z+a)T>7~Hv6Rgq;*rI&M
zlN9qaTiEUY5C3{-I*DuR&^r^nCi5{1K-vc?<bon^ku%8#Oxl1Z5RyM0FdKo3d|1_L
z$e{?H*BW=jp&7HF$>RW#61<xL98(5HPKZMzvFf-A`L$5c-WJp#u>kSm+W0mYD_isN
zv=gMF9@_W_Ymd8;n~R2h!rSpddqU+J%o`qIt?eJIo*u(0Tr4z!*arrrwjY?18C#ly
zdo^gyfaH&Yb)AWk2}Gs}ZV`V@=7fHLa|NtU1irCW*d3yM9e{@i>HqDQNUUuS_=psd
zxixw7-72hm{)JV@1)!fMaHzm|iSSC5;J64+6DiQfNS)w+2H;J=&r0yi08wEe@FYF5
zC=Iy(dj?buNmT<aZNO<3W@K52j(1^Y_fM?HUxp9Kfo7BFs1lTj22JK7GY}-7fz`&#
zxYG#RWC6=>0-0)u)yaY8<$xw8h@tml?ej8LTz7)DUKoMQB~9>G6fzl4klic8s&h56
z2(PfV84kXV041hlh4Q>&UH3EI5x5eP#XxkP1HT)KPY$3U@k%G8A{F%CAos<(IYmAs
z(?%|E6IL{DW4-eX)>&s`)n65>%@%S$taKiO7c7G;$T0UHc8`IXR34-?2Najd9Wc%V
ztn(KmFLDMMrU2-dH8?OEE3BE6ZUlK$yzMR*E6RHSrvs=?){l$<?L6dV7lS7b;7>Iq
zV>_~1n=wAg7I!Ob8C5~i8(3Ej!y57&w73S!_aHkJh;eKXrDb3~att|_U68ovfJYbJ
zJ`%DK3_htSRwHUbeSJvnVXV9Vq<jWVS3{~?!DSI7vI*QVgPaAyH$>o?3>{|Rt&`Uk
zV@}2@_bJRs>Y&LQ@Qf)KD-<+iAP?gKXrDnp+|ipCB(NTHtAD}cd*Ev;G8g0FYt|tT
zGY_k{hOiu5v?00eR@gzZQsIs7_kl~du_C@6Jy4KoN`@r1Vb-%6S?g@%biZTu-v?M`
zq3=XsM{<~9h-7Bs`AV#l@5K8Q=OVLm2pnpG9J)Y0M+38b@NzZQx)&gKu>mlZgD(D%
zoJD|!U}yjy48o3Zz)cdxjR*d|kVG;=B&!%dASb4Xnzlo>b^;5sYTbnUSCHZO1UVgm
zB{o4XWUZLk8xlLa;bR9*7J(ywBj>RRGQJbKej1+*Sfh`^s`(@Ebqumgry#?#fI|uP
zA1suIW99q?teqyfIuCsCfK|5wE&+(z4kK^#3D*_;Z3nq~i`>9Xz}E{M@q@G-LJsF&
za5fhCgBYydFGqyq0?R%KDJSa??y$`BkZqj^nIx-lL!coC6w(Hl`Oxp9$X0RW<+5~H
z6W-ps2wFkrlVp8e75tW9oEmr)LX#)RUmb!5UxWk>0*k%SyzdxuJ18zke3Sr7D}^*H
z2hUCb$5F88cObDOYoUSGRe0z0SiC3o6WS+$9;T3^Nx;7fd9^6Y8tBUiP_r3Ym4cp{
z;bHnQb}XoQ7t!czthc-c?cc+uyu!~|wCP9oWiK**vCuR><f`@|+hq*7+zwg>LefpC
z3$a_n4cWj`psFJ*BUwW=1l=_hey9z0pcdLhX1zg>iwU3>4;ohhDa?fbzX>Zg4Ovwa
zxi+$^3jhNHG8>J&$uDHK{2_t+u?s^TW6l6o$SwyHXth5iiUv+~AriHQZmdJT=O6r>
z2(GGt10-%Is}h#bfDmwI4fc@y4K4VNd|3>#DK^;GvK#M6?w3hr6M(f9MwH?`r7e&d
zCA@FaTHb(nf08$eW+0<No(_lZyhXN$$lFbD!U)#=DD2=N$c!!4>h8ckY=%9)1Ae#5
z?Vyv7k<Dqw=LTe~2_9(%a`<M*W^M=Gc31(qj6C!jJlhP9`wIG>4{q2%3)bT9G~iJW
zOSuZ~=UxIy{|}bY2l==k$PX5v=jHI1x!~&@<ihljCG~*Ze}%_hjJ#4EGKe+E70w65
zuR*=-fT<I_`Wt!J+pq%K$dXTjT)3lOiTnWYj>I0R(bR14=PNR+-LeeqhEYZq@Gzhf
zfZhR+?lpjGF(~Z>?uucn@ZLLEPdjM!C}?RPM&E_}xGVh5K1fC=@~IBgDD3YMf_5#i
zaCCVQc34rt>sPQLi$J*>px*{aCCMilAYyU`U#CE0=Riid$Z?-T)_*Q!csslxA1gBR
zfWa+f+4h1uW{8G!v8uNinM^xm!h_-M3z28v2JXgz%QrwFCwKs|#zST$1WG2b`wJP;
z8KCMVL?oW@6?foOuEIjjhg~5twg{T}4Hot$Jj6j{@h-umkySGXL<id-VShoEf?$QX
z(7QbNw;$ll9q@@{-_jMH)(M(73YOLf7TF8h=8C_@0yi$C^%roz4e#<8Z95SMkO*r6
zunB@r=p(|cf>!=OziqG}9K<4S&>1U*jrhIpL4}9*04`(kjS=F*Iz%#Gaiv0sWr!Xe
zaFG>MY{rG1B`Y~{=w6&+Crb+ASF(n+5S)yJ_6RZ3BjEV}@+yEQ@dPBH(9dz;1s^@W
z1YS1*X)2(hf;Q&BDjZLe^+sp(uL3C`IWV$5L-K50;QD)bw3E=%H-O9pG(7^^o&%NV
z!djBG(j-KPZvf4Gv^$S>&*AYk;A;*;qTeEW{wK6`2>5@4ohyUSv;gOH5VL#(B-daW
zDuCk>XzCASNgpFJ(?k}3D{}Qwuzz1bGkefx9B8AfSc8iNhII;mH5%SW8#?$I)Q^T$
z>i~4Ypu}9@OkxE&G>_zrhG9*l82KssTLHUt4;*zzEO!^y{|0^s!*bPOCVU@0ISq25
zhW}HL*Zvo>vI~~G7!r5@KB5Hs4<5qSO3>3L*tE5f<s?u@j!zzBh3wa=g<ZWy>7%&7
zc2cSLz^$utOL-pNz}$_umH(2DhO}J+WuibkLvYmtqpyS?*n;@M9Cr8?bSeN5)pFpJ
z3t6&(w2cLa*5bJbumV@%-_5a;Wr<uxUV`_oie>XK<{z*h+Ef-b9qp8n*M9+z-;R|%
zH%RLUXh<^jXezA9DD0z&h2=XAyPpnv-$y<?6nZ`k>U;pE`(f)gf<pU&3)$J{itkC}
z^G>nDU@Th7(J#pw;5;k5T8v^v_c}cATWFp-;5>spNUyLXXg4&=2CXZ>^Z!8cbFh>*
z;f-P-De3TmV<A6}u?yo3@Ct!O=RnF{qn9|$Z%Lk+tfkdMLY{ytWJkbCSiXEjf;T`X
zCi4GeAJY`@;0$O=2ZhKQ%{0iev0@G;14nAWlVn6c0?48*G|>utD@Ckx40~7(Ku0Qp
z-9PX~-Jso3=;<9;&JJ0#%p5%LlrI5=m{beuFe2g^h`i1sB8mWY9)a4rkgsL1niCba
zs7W!-Px#&@=!+Kk6$qLxfTfxWeQJV)FUNi%B6U63k8=gPMp9w#ra{8)!m{ehwXl<8
zFQ_+CA!8&eIE0=h;L}~i3|nE%@8eqs*v~)E;vbA_0Zk^mEB3&WKSQtM;2|%8n#Una
z!{8%{nYQ8SUcBG>EcW_*1oy@Oho``N9wcBtw5lJFlDz*INT4I^E(iXR<QZx}qXO96
zUU*ru%1Fm^WWD|!IMa^3K>qMFCjj9xSc?W|-JgiTmLh)6#&;_f5$Qfydo}p?Wbo|=
z<ca}`EP<6>121C-iO2<n?;#bXh^^Hioy0dxh9)y8>5!5Qh@JNWt0wr&m4N;N{6Y|H
z(tF^$6}~+huxY`ztOEpr@Buv-p%nD##B9hK*aSir91+6_!H;+7=N0H!0(mk;lrtO8
zECW}p;Ma<A|35&I4_|44@5maY8*FkjB7=CyR4ugHKoOZ*z$SKM)Ng427IL2s8g;^^
zFcAsqfQC-6HRNLoTSW3*U4Xp}mPZTnKLUL^K(>D8>PR-9tnF!`M-m(6K{|_IwHTm3
z$${!YvIi02)PmAv)m{@8pJZL7h~7!ws0y~8tm{hw8`(F|jdo<knhMKk4_Ms+hdz8=
zH7HUCxlqAfPf*_)-m(|Z7GcCfM2GztQ4eF<pm(w+PI6~tO-&EjSimZgtT<VxAnQ-H
z=!@hr$XakOMkc!k$Swu4OG6z`=)tcT;HRo$^^UA{qJSCTV#2nK0aeL*fem!I5&a~9
z9}g6^sTw`$fI8m5o2;P?D#jxDxqe_yvX5k6i#wnv`3g1QMApH`iq|mc>WH4lL;f_t
z{al50xd&dC!^)3<uCBm5eHOmghcAzXz7m`B4i<}Ko7G^!$ZDQD@MJ?q$ZB;BM&;ni
zsfvuAH+a|xKA(eboq_Z;Vn!7K54H!9T^RV$3_W-T8<qfCHls%{9GJ^tp)3I_StXnS
zoXKjY04?Hik@X{zH6<(ct&j{dvv7j6O@&6!hfD^;I?2G7kDv`%zxfJk_k(`g-~{0X
z$^ZS%-IBTpWbK&DKsXpx0NAS_8=c?@k>N00K??b20rx`4b_sZ10SO~n1{~>v%#gJx
zbwyl8=3qLw$Q}Tad6ht7$exJ?T-6G?@WBZNq>x1Y>WIwhfk`#m5H69-4%tDVuZT&=
zDm2Mo5V**y8yy-`iPnvZQM(bv5Q$O7M-Ox&QfCBCXkg{025=Sv$`VKzS&3K0X!g(=
zqBln9jjSlQ!7lyYZ9>Prk%$GnAX5eyhh*4E=9rIGEx<G%d`^b%`@L4KgfZOFdKCC!
zgt3SI56|DbWylVj9?(VrsuE4JfL*WxK4jMd$%>G<HVt@_Y{c*VF+G5g?5Q9*B(gHy
zrWldbP$9E&0=E}NBe5OX5%PN%N)KpAR@I3=8w3{JpxqGek-bDD&QOK?kX6`Tj7ipZ
z>(Lh#Jhp{y_=B4yKjH*B%RtF0;1!Q(=L@2aLO@Oj4an}V0LYLZp5VZ1v_qy#KnFt6
zLTE=GeiMx+D{*9IYz2O4fa5LTP8q1*hOt@bUr&K0;Sky1#8O~L=uibqnTt;gct|9{
z4E(bMm6*^;vJ-^lQaQNeff&>iF^w(yC;31%v=|01o#0J7xYU5YdKCO6>p*11$r=!;
zf!bu<SFE^*MI-rxX7o$oq5~3D!1FsNLH3jpc*#BsU4=}NRcf-rPdG_d_X)*&F&-a0
zB|CITo{_8rvmlSe!Vq{H(7yzhk;GnRpfe5M45H89Kgn(}vNMNhKo#I1>*^$rKvwIu
z0c$_L<Kr&LB=A6ef+<;B(}VmFi8h05leL}(SnEu93Q`|O2-@r58*_}w1vF$Q8_CiU
zN|N0(6wDyVzL!#Pj_k-I`?tsphu}hXD-oSlL9d$ls4Hke_BoLZkPWzL2uKD&;d+cz
zjd8pF-yZ0oKanf4n~e_G>9DgTA4)RDB)9T=m)q}sY-9(0u|jI9@h#Z_Mm&NHnnC^|
z`!9&6v&D$SKagxF*>gulZ%P<n3$)<jr#kQ>nPsvUic}P;14gy@O(JA<$OZ9}WPb(8
z7ZM3$qdl=^B%?|8tM!3HWq_g(^k{*8kz9@(_z-;|kPu#y%sRP<=OO!=XwV)3@Fw}H
zY}ndNNPeS&X21P2@yRS;PcD{XJTe*?T^W7%LY~OJMp6NZ*qB~$07(7s-~TA^KMMSh
P0{^4H|0wYPn*#q2LVnOg

literal 0
HcmV?d00001

diff --git a/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb b/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
new file mode 100644
index 0000000000..b766e1bda9
--- /dev/null
+++ b/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
@@ -0,0 +1,589 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Deploying Pre-trained Faster-Whisper-Large-v3 Model on SageMaker with Multi-Model Endpoint and Triton Serve as custome model\n",
+    "\n",
+    "Welcome to this notebook, where we will explore the process of deploying a pre-trained Faster-Whisper-Large-v3 model from Hugging Face on Amazon SageMaker. This deployment will be facilitated through the use of a Multi-Model Endpoint, allowing us to serve multiple models on a single SageMaker instance efficiently.\n",
+    "\n",
+    "## Overview\n",
+    "\n",
+    "In this demonstration, we'll cover the following key aspects:\n",
+    "\n",
+    "1. **Model Selection:** We will leverage the Hugging Face model hub to choose the Faster-Whisper-Large-v3 model, a powerful pre-trained model for various natural language processing tasks.\n",
+    "\n",
+    "2. **SageMaker Deployment:** Learn how to deploy the selected model on Amazon SageMaker, a fully managed service that enables the training and deployment of machine learning models at scale.\n",
+    "\n",
+    "3. **Multi-Model Endpoint:** Explore the advantages of using a Multi-Model Endpoint on SageMaker. This approach allows us to host and serve multiple models on a single instance, optimizing resource utilization and cost efficiency.\n",
+    "\n",
+    "4. **Triton Serve Integration:** Understand how Triton Serve, an open-source model serving platform, can be employed to dynamically load and unload models from the GPU. This capability is crucial for managing resources effectively and reducing operational costs.\n",
+    "\n",
+    "## Cost Optimization\n",
+    "\n",
+    "By utilizing a Multi-Model Endpoint on SageMaker, we aim to showcase how you can achieve significant cost savings. Serving multiple models on a single instance helps maximize the utilization of resources, leading to more efficient and economical deployment.\n",
+    "\n",
+    "## Note\n",
+    "Ensure that you have [git-lfs](https://git-lfs.com/) installed \n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Installs\n",
+    "Installs the dependencies required to package the model and run inferences using Triton server. Update SageMaker, boto3, awscli etc\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%pip install -qU pip awscli boto3 sagemaker\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Imports and variables"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Setting up SageMaker Execution Role\n",
+    "\n",
+    "1. **Create Role:**\n",
+    "   - Go to IAM Console.\n",
+    "   - Choose \"Roles\" -> \"Create role.\"\n",
+    "   - Select \"AWS service\" -> \"SageMaker.\"\n",
+    "   - Attach necessary policies.\n",
+    "   - Name the role: `AmazonSageMaker-ExecutionRole-mme-FasterWishper`.\n",
+    "\n",
+    "2. **Assign Permissions:**\n",
+    "   - Go to SageMaker Console.\n",
+    "   - Select your instance or job.\n",
+    "   - In settings, choose the created role.\n",
+    "   - Save changes.\n",
+    "\n",
+    "Now, your SageMaker instance/job is set up with the required execution role.\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import boto3, json, sagemaker, time\n",
+    "from sagemaker import get_execution_role\n",
+    "import numpy as np\n",
+    "import os\n",
+    "\n",
+    "os.environ[\"TOKENIZERS_PARALLELISM\"] = \"false\"\n",
+    "\n",
+    "# sagemaker variables\n",
+    "region_name = \"eu-west-1\" # change to your region\n",
+    "sm_client = boto3.client(service_name=\"sagemaker\", region_name=\"eu-west-1\")\n",
+    "runtime_sm_client = boto3.client(\"sagemaker-runtime\")\n",
+    "sagemaker_session = sagemaker.Session(boto_session=boto3.Session())\n",
+    "role = f\"arn:aws:iam::{boto3.client('sts').get_caller_identity().get('Account')}:role/AmazonSageMaker-ExecutionRole-mme-FasterWishper\"\n",
+    "s3_client = boto3.client(\"s3\")\n",
+    "bucket = sagemaker.Session().default_bucket()\n",
+    "print(\"S3 Bucket: {}\".format(bucket))\n",
+    "prefix = \"faster-whisper-large-v3-mme-gpu\"\n",
+    "\n",
+    "# account mapping for SageMaker MME Triton Image\n",
+    "account_id_map = {\n",
+    "    \"us-east-1\": \"785573368785\",\n",
+    "    \"us-east-2\": \"007439368137\",\n",
+    "    \"us-west-1\": \"710691900526\",\n",
+    "    \"us-west-2\": \"301217895009\",\n",
+    "    \"eu-west-1\": \"802834080501\",\n",
+    "    \"eu-west-2\": \"205493899709\",\n",
+    "    \"eu-west-3\": \"254080097072\",\n",
+    "    \"eu-north-1\": \"601324751636\",\n",
+    "    \"eu-south-1\": \"966458181534\",\n",
+    "    \"eu-central-1\": \"746233611703\",\n",
+    "    \"ap-east-1\": \"110948597952\",\n",
+    "    \"ap-south-1\": \"763008648453\",\n",
+    "    \"ap-northeast-1\": \"941853720454\",\n",
+    "    \"ap-northeast-2\": \"151534178276\",\n",
+    "    \"ap-southeast-1\": \"324986816169\",\n",
+    "    \"ap-southeast-2\": \"355873309152\",\n",
+    "    \"cn-northwest-1\": \"474822919863\",\n",
+    "    \"cn-north-1\": \"472730292857\",\n",
+    "    \"sa-east-1\": \"756306329178\",\n",
+    "    \"ca-central-1\": \"464438896020\",\n",
+    "    \"me-south-1\": \"836785723513\",\n",
+    "    \"af-south-1\": \"774647643957\",\n",
+    "}\n",
+    "\n",
+    "region = boto3.Session(region_name=region_name).region_name\n",
+    "if region not in account_id_map.keys():\n",
+    "    raise (\"UNSUPPORTED REGION\")\n",
+    "\n",
+    "base = \"amazonaws.com.cn\" if region.startswith(\"cn-\") else \"amazonaws.com\"\n",
+    "mme_triton_image_uri = (\n",
+    "    \"{account_id}.dkr.ecr.{region}.{base}/sagemaker-tritonserver:23.05-py3\".format(\n",
+    "        account_id=account_id_map[region], region=region, base=base\n",
+    "    )\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Generate model artifact for faster-whisper-large-v3 from huggingface"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We will clone the model from huggingface"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "!git clone https://huggingface.co/Systran/faster-whisper-large-v3"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Now we need to arrange the model as the triton server expects it\n",
+    "\n",
+    "Model repository structure for Faster Whisper large v3 Model.\n",
+    "\n",
+    "```\n",
+    "faster-whisper-large-v3\n",
+    "├── 1\n",
+    "│   └── model.py\n",
+    "└── config.pbtxt\n",
+    "```"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "!mkdir ./faster-whisper-large-v3/1\n",
+    "!mv ./faster-whisper-large-v3/*.* ./faster-whisper-large-v3/1/\n",
+    "!rm -f ./faster-whisper-large-v3/.gitattributes ./faster-whisper-large-v3/.git"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Create config.pbtxt"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%writefile faster-whisper-large-v3/config.pbtxt\n",
+    "name: \"faster-whisper-large-v3\"\n",
+    "backend: \"python\"\n",
+    "max_batch_size: 1\n",
+    "\n",
+    "input [\n",
+    "    {\n",
+    "        name: \"audio_array\"\n",
+    "        data_type: TYPE_STRING\n",
+    "        dims: [1]\n",
+    "    },\n",
+    "    {\n",
+    "        name: \"audio_lang\"\n",
+    "        data_type: TYPE_STRING\n",
+    "        dims: [1]\n",
+    "    }\n",
+    "]\n",
+    "\n",
+    "output [\n",
+    "    {\n",
+    "        name: \"transcript\"\n",
+    "        data_type: TYPE_STRING\n",
+    "        dims: [ 1 ]\n",
+    "    }\n",
+    "]\n",
+    "\n",
+    "instance_group [\n",
+    "  {\n",
+    "    kind: KIND_GPU\n",
+    "  }\n",
+    "]\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Package models and upload to S3\n",
+    "Next, we will package our model as `*.tar.gz` files for uploading to S3. \n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "!tar -C ./ -czf faster-whisper-large-v3.tar.gz faster-whisper-large-v3\n",
+    "model_uri_faster_whisper_large = sagemaker_session.upload_data(path=\"faster-whisper-large-v3.tar.gz\", key_prefix=prefix)\n",
+    "print(\"Model URI: {}\".format(model_uri_faster_whisper_large))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Build and Push docker image\n",
+    "Now we will build our docker file that will based on [Sagemaker deep learning images](https://github.com/aws/deep-learning-containers/blob/master/available_images.md)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Create the docker file\n",
+    "In the dokcer image we will add all our requirements"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from IPython.core.magic import register_line_cell_magic\n",
+    "\n",
+    "@register_line_cell_magic\n",
+    "def writetemplate(line, cell):\n",
+    "    with open(line, 'w') as f:\n",
+    "        f.write(cell.format(**globals()))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%writetemplate Dockerfile\n",
+    "FROM {mme_triton_image_uri}\n",
+    "\n",
+    "RUN apt-get update && apt-get install -y \\\n",
+    "    ffmpeg \\\n",
+    "    libcublas11\n",
+    "\n",
+    "RUN pip3 install faster-whisper ffmpeg ffmpeg-python \n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Build the docker image and push to ECR"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "docker_version=\"0.0.1\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "!docker build -t {sagemaker_session.account_id()}.dkr.ecr.{region}.amazonaws.com/triton/faster-whisper:{docker_version} ."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "!aws ecr create-repository --repository-name triton/faster-whisper --region {region}\n",
+    "!aws ecr get-login-password --region {region} | docker login --username AWS --password-stdin {sagemaker_session.account_id()}.dkr.ecr.{region}.amazonaws.com\n",
+    "!docker push {sagemaker_session.account_id()}.dkr.ecr.{region}.amazonaws.com/triton/faster-whisper:{docker_version}\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Create SageMaker Endpoint\n",
+    "\n",
+    "Now that we have uploaded the model artifacts to S3, we can create a SageMaker multi-model endpoint."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Define the serving container\n",
+    "In the container definition, define the `ModelDataUrl` to specify the S3 directory that contains all the models that SageMaker multi-model endpoint will use to load and serve predictions. Set `Mode` to `MultiModel` to indicate SageMaker would create the endpoint with MME container specifications. We set the container with an image that supports deploying multi-model endpoints with GPU"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "model_data_url = f\"s3://{bucket}/{prefix}/\"\n",
+    "\n",
+    "container = {\n",
+    "    \"Image\": f\"{sagemaker_session.account_id()}.dkr.ecr.{region}.amazonaws.com/triton/faster-whisper:{docker_version}\",\n",
+    "    \"ModelDataUrl\": model_data_url,\n",
+    "    \"Mode\": \"MultiModel\",\n",
+    "}\n",
+    "print(\"Container: {}\".format(container))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Create a multi-model object"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Once the image, data location are set we create the model using `create_model` by specifying the `ModelName` and the Container definition"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "ts = time.strftime(\"%Y-%m-%d-%H-%M-%S\", time.gmtime())\n",
+    "sm_model_name = f\"{prefix}-mdl-{ts}\"\n",
+    "sagemaker_execution_role = role\n",
+    "\n",
+    "create_model_response = sm_client.create_model(\n",
+    "    ModelName=sm_model_name, ExecutionRoleArn=sagemaker_execution_role, PrimaryContainer=container\n",
+    ")\n",
+    "\n",
+    "print(\"Model Arn: \" + create_model_response[\"ModelArn\"])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Define configuration for the multi-model endpoint"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Using the model above, we create an [endpoint configuration](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateEndpointConfig.html) where we can specify the type and number of instances we want in the endpoint. Here we are deploying to `g4dn.2xlarge` NVIDIA GPU instance."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "endpoint_config_name = f\"{prefix}-epc-{ts}-2xl\"\n",
+    "\n",
+    "create_endpoint_config_response = sm_client.create_endpoint_config(\n",
+    "    EndpointConfigName=endpoint_config_name,\n",
+    "    ProductionVariants=[\n",
+    "        {\n",
+    "            \"InstanceType\": \"ml.g4dn.2xlarge\",\n",
+    "            \"InitialVariantWeight\": 1,\n",
+    "            \"InitialInstanceCount\": 1,\n",
+    "            \"ModelName\": sm_model_name,\n",
+    "            \"VariantName\": \"AllTraffic\",\n",
+    "        }\n",
+    "    ],\n",
+    ")\n",
+    "\n",
+    "print(\"Endpoint Config Arn: \" + create_endpoint_config_response[\"EndpointConfigArn\"])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Create Multi-Model Endpoint"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Using the above endpoint configuration we create a new sagemaker endpoint and wait for the deployment to finish. The status will change to **InService** once the deployment is successful."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "endpoint_name = f\"{prefix}-ep-{ts}-2xl\"\n",
+    "\n",
+    "create_endpoint_response = sm_client.create_endpoint(\n",
+    "    EndpointName=endpoint_name, EndpointConfigName=endpoint_config_name\n",
+    ")\n",
+    "\n",
+    "print(\"Endpoint Arn: \" + create_endpoint_response[\"EndpointArn\"])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "resp = sm_client.describe_endpoint(EndpointName=endpoint_name)\n",
+    "status = resp[\"EndpointStatus\"]\n",
+    "print(\"Status: \" + status)\n",
+    "\n",
+    "while status == \"Creating\":\n",
+    "    time.sleep(60)\n",
+    "    resp = sm_client.describe_endpoint(EndpointName=endpoint_name)\n",
+    "    status = resp[\"EndpointStatus\"]\n",
+    "    print(\"Status: \" + status)\n",
+    "\n",
+    "print(\"Arn: \" + resp[\"EndpointArn\"])\n",
+    "print(\"Status: \" + status)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Run Inference"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Once we have the endpoint running we can use some sample raw data to do an inference using JSON as the payload format. For the inference request format, Triton uses the KFServing community standard [inference protocols](https://github.com/triton-inference-server/server/blob/main/docs/protocol/README.md)."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import io\n",
+    "import base64\n",
+    "\n",
+    "def get_payload(wav_file_path, language):\n",
+    "    with open(wav_file_path, \"rb\") as f:\n",
+    "        wav = base64.b64encode(f.read()).decode('ascii')\n",
+    "         \n",
+    "    payload = {}\n",
+    "    payload[\"inputs\"] = []\n",
+    "    payload[\"inputs\"].append(\n",
+    "        {\n",
+    "            \"name\": \"audio_array\",\n",
+    "            \"shape\": [1, 1],\n",
+    "            \"datatype\": \"BYTES\",\n",
+    "            \"data\": [wav],\n",
+    "        }\n",
+    "    )\n",
+    "    payload[\"inputs\"].append(\n",
+    "        {\n",
+    "            \"name\": \"audio_lang\",\n",
+    "            \"shape\": [1, 1],\n",
+    "            \"datatype\": \"BYTES\",\n",
+    "            \"data\": [language],\n",
+    "        }\n",
+    "    )\n",
+    "    return payload"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "payload = get_payload(\"./audio/sample.wav\", \"fr\")\n",
+    "\n",
+    "response = runtime_sm_client.invoke_endpoint(\n",
+    "    EndpointName=endpoint_name,\n",
+    "    ContentType=\"application/json\",\n",
+    "    Body=json.dumps(payload),\n",
+    "    TargetModel=\"faster-whisper-large-v3.tar.gz\",\n",
+    ")\n",
+    "\n",
+    "response_body = json.loads(response[\"Body\"].read().decode(\"utf-8\"))\n",
+    "print(response_body)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "sm_client.delete_model(ModelName=sm_model_name)\n",
+    "sm_client.delete_endpoint_config(EndpointConfigName=endpoint_config_name)\n",
+    "sm_client.delete_endpoint(EndpointName=endpoint_name)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": ".venv",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/readme.md b/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/readme.md
new file mode 100644
index 0000000000..78c2a3296a
--- /dev/null
+++ b/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/readme.md
@@ -0,0 +1,45 @@
+# Deploying Pre-trained Faster-Whisper-Large-v3 Model on SageMaker with Multi-Model Endpoint and Triton Serve
+
+This guide provides an overview of deploying a pre-trained [Faster-Whisper-Large-v3 model](https://huggingface.co/Systran/faster-whisper-large-v3) from Hugging Face on Amazon SageMaker. The deployment utilizes a Multi-Model Endpoint, allowing efficient serving of multiple models on a single SageMaker instance.
+
+## Overview
+
+1. **Model Selection:** Choose the Faster-Whisper-Large-v3 model from Hugging Face's model hub.
+2. **SageMaker Deployment:** Deploy the selected model on Amazon SageMaker, a managed service for ML model training and deployment.
+3. **Multi-Model Endpoint:** Utilize a Multi-Model Endpoint on SageMaker for hosting and serving multiple models on a single instance.
+4. **Triton Serve Integration:** Integrate Triton Serve, an open-source model serving platform, for dynamic loading and unloading of models from the GPU.
+
+## Cost Optimization
+
+By using a Multi-Model Endpoint on SageMaker, achieve cost savings by maximizing resource utilization and cost efficiency.
+
+## Prerequisites
+
+Ensure [git-lfs](https://git-lfs.com/) is installed.
+
+### Install Dependencies
+
+```bash
+%pip install -qU pip awscli boto3 sagemaker
+```
+
+### Setting up SageMaker Execution Role
+
+1. Create a SageMaker execution role `AmazonSageMaker-ExecutionRole-mme-FasterWishper` with necessary policies.
+2. Assign the created role to your SageMaker instance/job.
+
+## Usage
+
+1. Clone the Faster-Whisper-Large-v3 model from Hugging Face.
+2. Arrange the model as Triton server expects it.
+3. Create the Triton server configuration file (`config.pbtxt`).
+4. Package the model as `*.tar.gz` for uploading to S3.
+5. Build and push the Docker image to ECR.
+6. Creat SageMaker Multi Model EndPoint base on Triton server
+7. Test your endpoint
+
+---
+
+This summarizes the key steps and commands involved in deploying the Faster-Whisper-Large-v3 model on SageMaker with a Multi-Model Endpoint and Triton Serve. Adjustments can be made based on specific requirements and configurations.
+
+Link to the [example notebook](sagemaker-mme-triton-fasterwhisper.ipynb)
\ No newline at end of file

From d912e1c509c95ee0e569820fbdf45eab004f1dcf Mon Sep 17 00:00:00 2001
From: Eldad Cohen <eldad.cohen@aiola.com>
Date: Fri, 5 Apr 2024 15:07:55 +0000
Subject: [PATCH 2/9] fixed grammar test

---
 .../mme-triton-custom-faster-whisper.ipynb    | 20 +++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb b/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
index b766e1bda9..5be9e2ee29 100644
--- a/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
+++ b/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
@@ -4,7 +4,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Deploying Pre-trained Faster-Whisper-Large-v3 Model on SageMaker with Multi-Model Endpoint and Triton Serve as custome model\n",
+    "# Deploying Pre-trained Faster-Whisper-Large-v3 Model on SageMaker with Multi-Model Endpoint and Triton Serve as custom model\n",
     "\n",
     "Welcome to this notebook, where we will explore the process of deploying a pre-trained Faster-Whisper-Large-v3 model from Hugging Face on Amazon SageMaker. This deployment will be facilitated through the use of a Multi-Model Endpoint, allowing us to serve multiple models on a single SageMaker instance efficiently.\n",
     "\n",
@@ -34,7 +34,7 @@
    "metadata": {},
    "source": [
     "### Installs\n",
-    "Installs the dependencies required to package the model and run inferences using Triton server. Update SageMaker, boto3, awscli etc\n",
+    "Installs the dependencies required to package the model and run inferences using Triton server.</br>Update SageMaker, boto3, AWS CLI etc.\n",
     "\n"
    ]
   },
@@ -62,8 +62,8 @@
     "\n",
     "1. **Create Role:**\n",
     "   - Go to IAM Console.\n",
-    "   - Choose \"Roles\" -> \"Create role.\"\n",
-    "   - Select \"AWS service\" -> \"SageMaker.\"\n",
+    "   - Choose \"Roles\" &rarr; \"Create role.\"\n",
+    "   - Select \"AWS service\" &rarr; \"SageMaker\"\n",
     "   - Attach necessary policies.\n",
     "   - Name the role: `AmazonSageMaker-ExecutionRole-mme-FasterWishper`.\n",
     "\n",
@@ -142,14 +142,14 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Generate model artifact for faster-whisper-large-v3 from huggingface"
+    "### Generate model artifact for faster-whisper-large-v3 from hugging face"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "We will clone the model from huggingface"
+    "We will clone the model from hugging face"
    ]
   },
   {
@@ -165,7 +165,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Now we need to arrange the model as the triton server expects it\n",
+    "Now we need to arrange the model as the Triton server expects it\n",
     "\n",
     "Model repository structure for Faster Whisper large v3 Model.\n",
     "\n",
@@ -258,7 +258,7 @@
    "metadata": {},
    "source": [
     "### Build and Push docker image\n",
-    "Now we will build our docker file that will based on [Sagemaker deep learning images](https://github.com/aws/deep-learning-containers/blob/master/available_images.md)"
+    "Now we will build our docker file that will be based on [SageMaker deep learning images](https://github.com/aws/deep-learning-containers/blob/master/available_images.md)"
    ]
   },
   {
@@ -266,7 +266,7 @@
    "metadata": {},
    "source": [
     "#### Create the docker file\n",
-    "In the dokcer image we will add all our requirements"
+    "In the docker image we will add all our requirements"
    ]
   },
   {
@@ -448,7 +448,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Using the above endpoint configuration we create a new sagemaker endpoint and wait for the deployment to finish. The status will change to **InService** once the deployment is successful."
+    "Using the above endpoint configuration we create a new sagemaker endpoint and wait for the deployment to finish. The status will change to **In-service** once the deployment is successful."
    ]
   },
   {

From 7bcdd1c691e6702b268aaa12b25b7ba05345575b Mon Sep 17 00:00:00 2001
From: Eldad Cohen <eldad.cohen@aiola.com>
Date: Thu, 25 Apr 2024 07:27:36 +0000
Subject: [PATCH 3/9] update format with black

---
 .../mme-triton-custom-faster-whisper.ipynb    | 20 +++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)

diff --git a/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb b/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
index 5be9e2ee29..bd8e58cc0c 100644
--- a/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
+++ b/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
@@ -44,7 +44,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "%pip install -qU pip awscli boto3 sagemaker\n"
+    "%pip install -qU pip awscli boto3 sagemaker"
    ]
   },
   {
@@ -90,7 +90,7 @@
     "os.environ[\"TOKENIZERS_PARALLELISM\"] = \"false\"\n",
     "\n",
     "# sagemaker variables\n",
-    "region_name = \"eu-west-1\" # change to your region\n",
+    "region_name = \"eu-west-1\"  # change to your region\n",
     "sm_client = boto3.client(service_name=\"sagemaker\", region_name=\"eu-west-1\")\n",
     "runtime_sm_client = boto3.client(\"sagemaker-runtime\")\n",
     "sagemaker_session = sagemaker.Session(boto_session=boto3.Session())\n",
@@ -249,7 +249,9 @@
    "outputs": [],
    "source": [
     "!tar -C ./ -czf faster-whisper-large-v3.tar.gz faster-whisper-large-v3\n",
-    "model_uri_faster_whisper_large = sagemaker_session.upload_data(path=\"faster-whisper-large-v3.tar.gz\", key_prefix=prefix)\n",
+    "model_uri_faster_whisper_large = sagemaker_session.upload_data(\n",
+    "    path=\"faster-whisper-large-v3.tar.gz\", key_prefix=prefix\n",
+    ")\n",
     "print(\"Model URI: {}\".format(model_uri_faster_whisper_large))"
    ]
   },
@@ -277,9 +279,10 @@
    "source": [
     "from IPython.core.magic import register_line_cell_magic\n",
     "\n",
+    "\n",
     "@register_line_cell_magic\n",
     "def writetemplate(line, cell):\n",
-    "    with open(line, 'w') as f:\n",
+    "    with open(line, \"w\") as f:\n",
     "        f.write(cell.format(**globals()))"
    ]
   },
@@ -312,7 +315,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "docker_version=\"0.0.1\""
+    "docker_version = \"0.0.1\""
    ]
   },
   {
@@ -332,7 +335,7 @@
    "source": [
     "!aws ecr create-repository --repository-name triton/faster-whisper --region {region}\n",
     "!aws ecr get-login-password --region {region} | docker login --username AWS --password-stdin {sagemaker_session.account_id()}.dkr.ecr.{region}.amazonaws.com\n",
-    "!docker push {sagemaker_session.account_id()}.dkr.ecr.{region}.amazonaws.com/triton/faster-whisper:{docker_version}\n"
+    "!docker push {sagemaker_session.account_id()}.dkr.ecr.{region}.amazonaws.com/triton/faster-whisper:{docker_version}"
    ]
   },
   {
@@ -509,10 +512,11 @@
     "import io\n",
     "import base64\n",
     "\n",
+    "\n",
     "def get_payload(wav_file_path, language):\n",
     "    with open(wav_file_path, \"rb\") as f:\n",
-    "        wav = base64.b64encode(f.read()).decode('ascii')\n",
-    "         \n",
+    "        wav = base64.b64encode(f.read()).decode(\"ascii\")\n",
+    "\n",
     "    payload = {}\n",
     "    payload[\"inputs\"] = []\n",
     "    payload[\"inputs\"].append(\n",

From 56e95eda7b7a002914ba9007b33bb958dc7e453b Mon Sep 17 00:00:00 2001
From: Eldad Cohen <eldad.cohen@aiola.com>
Date: Thu, 25 Apr 2024 08:37:39 +0000
Subject: [PATCH 4/9] added ci test result

---
 .../mme-triton-custom-faster-whisper.ipynb    | 60 ++++++++++++++++++-
 1 file changed, 59 insertions(+), 1 deletion(-)

diff --git a/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb b/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
index bd8e58cc0c..92a7498396 100644
--- a/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
+++ b/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
@@ -4,8 +4,26 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Deploying Pre-trained Faster-Whisper-Large-v3 Model on SageMaker with Multi-Model Endpoint and Triton Serve as custom model\n",
+    "# Deploying Pre-trained Faster-Whisper-Large-v3 Model on SageMaker with Multi-Model Endpoint and Triton Serve as custom model"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "---\n",
+    "\n",
+    "This notebook's CI test result for us-west-2 is as follows. CI test results in other regions can be found at the end of the notebook.\n",
+    "\n",
+    "![This us-west-2 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/us-west-2/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
     "\n",
+    "---"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
     "Welcome to this notebook, where we will explore the process of deploying a pre-trained Faster-Whisper-Large-v3 model from Hugging Face on Amazon SageMaker. This deployment will be facilitated through the use of a Multi-Model Endpoint, allowing us to serve multiple models on a single SageMaker instance efficiently.\n",
     "\n",
     "## Overview\n",
@@ -567,6 +585,46 @@
     "sm_client.delete_endpoint_config(EndpointConfigName=endpoint_config_name)\n",
     "sm_client.delete_endpoint(EndpointName=endpoint_name)"
    ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Notebook CI Test Results\n",
+    "\n",
+    "This notebook was tested in multiple regions. The test results are as follows, except for us-west-2 which is shown at the top of the notebook.\n",
+    "\n",
+    "\n",
+    "![This us-east-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/us-east-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "\n",
+    "![This us-east-2 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/us-east-2/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "\n",
+    "![This us-west-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/us-west-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "\n",
+    "![This ca-central-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ca-central-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "\n",
+    "![This sa-east-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/sa-east-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "\n",
+    "![This eu-west-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/eu-west-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "\n",
+    "![This eu-west-2 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/eu-west-2/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "\n",
+    "![This eu-west-3 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/eu-west-3/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "\n",
+    "![This eu-central-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/eu-central-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "\n",
+    "![This eu-north-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/eu-north-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "\n",
+    "![This ap-southeast-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ap-southeast-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "\n",
+    "![This ap-southeast-2 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ap-southeast-2/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "\n",
+    "![This ap-northeast-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ap-northeast-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "\n",
+    "![This ap-northeast-2 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ap-northeast-2/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "\n",
+    "![This ap-south-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ap-south-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n"
+   ]
   }
  ],
  "metadata": {

From 727a7f18e349fd452434cd91ed3652a8b72e065e Mon Sep 17 00:00:00 2001
From: Eldad Cohen <eldad.cohen@aiola.com>
Date: Thu, 25 Apr 2024 10:58:45 +0000
Subject: [PATCH 5/9] added example to index.rst + move to triton examples

---
 inference/index.rst                                 |   1 +
 .../audio/sample.wav                                | Bin
 .../mme-triton-custom-faster-whisper.ipynb          |   0
 .../mme-triton-custom-faster-whisper/readme.md      |   0
 4 files changed, 1 insertion(+)
 rename {inference/nlp/realtime/triton/multi-model => sagemaker-triton}/mme-triton-custom-faster-whisper/audio/sample.wav (100%)
 rename {inference/nlp/realtime/triton/multi-model => sagemaker-triton}/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb (100%)
 rename {inference/nlp/realtime/triton/multi-model => sagemaker-triton}/mme-triton-custom-faster-whisper/readme.md (100%)

diff --git a/inference/index.rst b/inference/index.rst
index e92a747cb6..591432cf3b 100644
--- a/inference/index.rst
+++ b/inference/index.rst
@@ -109,3 +109,4 @@ Nvidia Triton Inference
 
    ../sagemaker-triton/resnet50/triton_resnet50
    ../sagemaker-triton/nlp_bert/triton_nlp_bert
+   ../sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper
diff --git a/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/audio/sample.wav b/sagemaker-triton/mme-triton-custom-faster-whisper/audio/sample.wav
similarity index 100%
rename from inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/audio/sample.wav
rename to sagemaker-triton/mme-triton-custom-faster-whisper/audio/sample.wav
diff --git a/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb b/sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
similarity index 100%
rename from inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
rename to sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
diff --git a/inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/readme.md b/sagemaker-triton/mme-triton-custom-faster-whisper/readme.md
similarity index 100%
rename from inference/nlp/realtime/triton/multi-model/mme-triton-custom-faster-whisper/readme.md
rename to sagemaker-triton/mme-triton-custom-faster-whisper/readme.md

From 1b5f263b53c9e412fe50b6940863d201e1e1f66c Mon Sep 17 00:00:00 2001
From: Eldad Cohen <eldad.cohen@aiola.com>
Date: Thu, 25 Apr 2024 11:02:36 +0000
Subject: [PATCH 6/9] update notebook path to ci

---
 .../mme-triton-custom-faster-whisper.ipynb                      | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb b/sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
index 92a7498396..c98f1608bb 100644
--- a/sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
+++ b/sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
@@ -15,7 +15,7 @@
     "\n",
     "This notebook's CI test result for us-west-2 is as follows. CI test results in other regions can be found at the end of the notebook.\n",
     "\n",
-    "![This us-west-2 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/us-west-2/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "![This us-west-2 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/us-west-2/sagemaker-triton|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
     "\n",
     "---"
    ]

From 4a6a421cf1728e2bf7595dbff188e7ad157edf4d Mon Sep 17 00:00:00 2001
From: Eldad Cohen <eldad.cohen@aiola.com>
Date: Thu, 25 Apr 2024 11:04:52 +0000
Subject: [PATCH 7/9] update notebook path to ci

---
 .../mme-triton-custom-faster-whisper.ipynb    | 35 +++++++++++--------
 1 file changed, 20 insertions(+), 15 deletions(-)

diff --git a/sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb b/sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
index c98f1608bb..ed67d23102 100644
--- a/sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
+++ b/sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
@@ -595,36 +595,41 @@
     "This notebook was tested in multiple regions. The test results are as follows, except for us-west-2 which is shown at the top of the notebook.\n",
     "\n",
     "\n",
-    "![This us-east-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/us-east-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "![This us-east-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/us-east-1/sagemaker-triton|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
     "\n",
-    "![This us-east-2 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/us-east-2/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "![This us-east-2 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/us-east-2/sagemaker-triton|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
     "\n",
-    "![This us-west-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/us-west-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "![This us-west-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/us-west-1/sagemaker-triton|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
     "\n",
-    "![This ca-central-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ca-central-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "![This ca-central-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ca-central-1/sagemaker-triton|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
     "\n",
-    "![This sa-east-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/sa-east-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "![This sa-east-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/sa-east-1/sagemaker-triton|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
     "\n",
-    "![This eu-west-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/eu-west-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "![This eu-west-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/eu-west-1/sagemaker-triton|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
     "\n",
-    "![This eu-west-2 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/eu-west-2/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "![This eu-west-2 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/eu-west-2/sagemaker-triton|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
     "\n",
-    "![This eu-west-3 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/eu-west-3/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "![This eu-west-3 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/eu-west-3/sagemaker-triton|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
     "\n",
-    "![This eu-central-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/eu-central-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "![This eu-central-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/eu-central-1/sagemaker-triton|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
     "\n",
-    "![This eu-north-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/eu-north-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "![This eu-north-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/eu-north-1/sagemaker-triton|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
     "\n",
-    "![This ap-southeast-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ap-southeast-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "![This ap-southeast-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ap-southeast-1/sagemaker-triton|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
     "\n",
-    "![This ap-southeast-2 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ap-southeast-2/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "![This ap-southeast-2 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ap-southeast-2/sagemaker-triton|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
     "\n",
-    "![This ap-northeast-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ap-northeast-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "![This ap-northeast-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ap-northeast-1/sagemaker-triton|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
     "\n",
-    "![This ap-northeast-2 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ap-northeast-2/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
+    "![This ap-northeast-2 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ap-northeast-2/sagemaker-triton|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n",
     "\n",
-    "![This ap-south-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ap-south-1/inference|nlp|realtime|triton|multi-model|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n"
+    "![This ap-south-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ap-south-1/sagemaker-triton|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n"
    ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": []
   }
  ],
  "metadata": {

From 7fcce8bf4ec2a50ca525258ea2d521faf23b67c5 Mon Sep 17 00:00:00 2001
From: Eldad Cohen <eldad.cohen@aiola.com>
Date: Thu, 25 Apr 2024 11:25:52 +0000
Subject: [PATCH 8/9] removed empty cell

---
 .../mme-triton-custom-faster-whisper.ipynb                   | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb b/sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
index ed67d23102..c9ce67194c 100644
--- a/sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
+++ b/sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
@@ -625,11 +625,6 @@
     "\n",
     "![This ap-south-1 badge failed to load. Check your device's internet connectivity, otherwise the service is currently unavailable](https://prod.us-west-2.tcx-beacon.docs.aws.dev/sagemaker-nb/ap-south-1/sagemaker-triton|mme-triton-custom-faster-whisper|mme-triton-custom-faster-whisper.ipynb)\n"
    ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": []
   }
  ],
  "metadata": {

From a453bea0a00a6a4ae76c76c679ef51d8bee7b009 Mon Sep 17 00:00:00 2001
From: eldadcohen1 <eldad@boomnow.com>
Date: Sun, 14 Jun 2026 15:55:46 +0300
Subject: [PATCH 9/9] Fix grammar check: use arrow char and backtick config
 filename

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---
 .../mme-triton-custom-faster-whisper.ipynb                  | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb b/sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
index c9ce67194c..37717cf4c8 100644
--- a/sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
+++ b/sagemaker-triton/mme-triton-custom-faster-whisper/mme-triton-custom-faster-whisper.ipynb
@@ -80,8 +80,8 @@
     "\n",
     "1. **Create Role:**\n",
     "   - Go to IAM Console.\n",
-    "   - Choose \"Roles\" &rarr; \"Create role.\"\n",
-    "   - Select \"AWS service\" &rarr; \"SageMaker\"\n",
+    "   - Choose \"Roles\" → \"Create role.\"\n",
+    "   - Select \"AWS service\" → \"SageMaker\"\n",
     "   - Attach necessary policies.\n",
     "   - Name the role: `AmazonSageMaker-ExecutionRole-mme-FasterWishper`.\n",
     "\n",
@@ -210,7 +210,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Create config.pbtxt"
+    "### Create `config.pbtxt`"
    ]
   },
   {