From 0950bf140da219d6660710f7c928326003b43bec Mon Sep 17 00:00:00 2001 From: Khushi Kalra Date: Tue, 25 Jun 2024 23:15:55 +0530 Subject: [PATCH 1/5] Add files via upload --- Sucide & Depression Detection/Dataset/README.md | 5 +++++ 1 file changed, 5 insertions(+) create mode 100644 Sucide & Depression Detection/Dataset/README.md diff --git a/Sucide & Depression Detection/Dataset/README.md b/Sucide & Depression Detection/Dataset/README.md new file mode 100644 index 000000000..5a9572a1f --- /dev/null +++ b/Sucide & Depression Detection/Dataset/README.md @@ -0,0 +1,5 @@ +# **Sucide & Depression Detection** + +The dataset is a collection of posts from the "SuicideWatch" and "depression" subreddits of the Reddit platform. The posts are collected using Pushshift API. All posts that were made to "SuicideWatch" from Dec 16, 2008(creation) till Jan 2, 2021, were collected while "depression" posts were collected from Jan 1, 2009, to Jan 2, 2021. All posts collected from SuicideWatch are labeled as suicide, While posts collected from the depression subreddit are labeled as depression. Non-suicide posts are collected from r/teenagers. + +### Dataset Link : https://www.kaggle.com/datasets/nikhileswarkomati/suicide-watch/data \ No newline at end of file From 6b3bec073e64c3fd584951a625c05a8628e5ec1a Mon Sep 17 00:00:00 2001 From: Khushi Kalra Date: Fri, 12 Jul 2024 10:45:48 +0530 Subject: [PATCH 2/5] Update README.md --- Sucide & Depression Detection/Dataset/README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Sucide & Depression Detection/Dataset/README.md b/Sucide & Depression Detection/Dataset/README.md index 5a9572a1f..43f701866 100644 --- a/Sucide & Depression Detection/Dataset/README.md +++ b/Sucide & Depression Detection/Dataset/README.md @@ -1,5 +1,5 @@ -# **Sucide & Depression Detection** +# **Suicide & Depression Detection** The dataset is a collection of posts from the "SuicideWatch" and "depression" subreddits of the Reddit platform. The posts are collected using Pushshift API. All posts that were made to "SuicideWatch" from Dec 16, 2008(creation) till Jan 2, 2021, were collected while "depression" posts were collected from Jan 1, 2009, to Jan 2, 2021. All posts collected from SuicideWatch are labeled as suicide, While posts collected from the depression subreddit are labeled as depression. Non-suicide posts are collected from r/teenagers. -### Dataset Link : https://www.kaggle.com/datasets/nikhileswarkomati/suicide-watch/data \ No newline at end of file +### Dataset Link : https://www.kaggle.com/datasets/nikhileswarkomati/suicide-watch/data From 9c08d30cd8541fefd56c7c13f334faa53a77c89d Mon Sep 17 00:00:00 2001 From: Khushi Kalra Date: Fri, 12 Jul 2024 10:46:44 +0530 Subject: [PATCH 3/5] added images --- .../Images/GRU_Accuracy_Plot.png | Bin 0 -> 19407 bytes .../Images/Model_LSTMs_Accuracy.png | Bin 0 -> 49315 bytes .../Images/Top 50 words.png | Bin 0 -> 33892 bytes .../Images/WordCloud.png | Bin 0 -> 203023 bytes 4 files changed, 0 insertions(+), 0 deletions(-) create mode 100644 Sucide & Depression Detection/Images/GRU_Accuracy_Plot.png create mode 100644 Sucide & Depression Detection/Images/Model_LSTMs_Accuracy.png create mode 100644 Sucide & Depression Detection/Images/Top 50 words.png create mode 100644 Sucide & Depression Detection/Images/WordCloud.png diff --git a/Sucide & Depression Detection/Images/GRU_Accuracy_Plot.png b/Sucide & Depression Detection/Images/GRU_Accuracy_Plot.png new file mode 100644 index 0000000000000000000000000000000000000000..1d34faf404c3d960c4f636d2d68054216657e6a9 GIT binary patch literal 19407 zcmb`vby!s2yFNT9D1wTTDpD$4Dy1S43J6H|z|h^D!XScxNGaiGkdTIp{hjyy<2_!Nm)o`X?6p_j&;8ubvsQ$fiu?tNYZNdT?1IA6CmJx= zDSa4>q?G&&C_x#x#DafF+%@E7V8uVLFM$7$T1zWS!(e4Gl*guI;Q!}bp6a{9VD#+J zUlIk4Tk9~Gl7+$(X)Pb4m9euQXlG8^Hdt5@q-4`lm%_*J!A(h`=7 z3MS;SeLzoC)!A1mYmd6td~;;@euB?XE8sb-2LE!&hVa2)LBgySBrsSO2^%F0#(IGqGzn%a z)qV;FdwN=i9tM-TcBvc`^#sL&Cd8ik-~5Rh8~xmb{OIn=$jV$AvI>r|(+{PfJ1zp> z@q^ZSpitjD>Y7OTYr2;1S5Mj%%Wb-E*!N^bG=c8Uu$XKP9jI|NNAg#Y>nrFK!7V)( z`&^M(61#&f6JUgz9d;J`i{mV2z9s0?c{w8OtuB|posY?qjBcg}4Gpp(|8tPs?ESyj z{7+;V?_6Hkz+hhEiTqZ)I`8?7opzfD*^Lb|^c4ZS8%^o)Asfdj-LWD*>n2BSl6ccd z2EnSARF4Lk{pQ)0da@%4Hi0n`yYD3q=QDR(s3Zp)k5=tZpS@T)-r#=}gNnLxr&*w3 zTLbABc7ah;4`bJVjA_0MM&|ZNRj0$ux8Ro=F3I)$f!FwZa??t0jxrCsL|)ygGl?Ol;zLg~Y-1yMe|4|D)xKuI1Of#id{{ z#<8ADv#FV}o+mrFkq`IXNBvkiIoq0ag~60y9hrZ|B3U$qBp9r}z5ao=)Qg?1FW+3J zx_HRQ-?utqPq6k&p}r6yB0D5D;dlGZn93^05$rmx%}{x^=i2z#Dt1W#`Tb(b15Y>8 zXZ*%50#6Qy;<+`K|HMdoEm7zXS82$9Rk;BsWwXZGMtkCM2uNsawzkBTtSsxj6Vq}!l9*sRZUVjW~S{cSv z^r}jZ;Wv|-_6>fVIe*1uZvxp=WKjK+N4I!cMrWciz%eagx6E;*NYp3 zN{XHJ2iI5p2y9=2QOm0fjy+ilyJOxHLm$JK7rmg#;9&msP1#i;t3i^p*MB!@7*$UM z){VKgG2w?Vf(7}h_}7qUC#wDh_x&gA#meBEwcE`KiFj=JZX z!Q|Pej@PMZI$3ev{UOvf2uA;p^N@7w63);)%HL4Z?d6#Zh1OlEUTY0|U#WhwSM81a zA9anI`ahiIjb(EGac38d=DxvJ20JB8jhaG5`W_FTur0T(i-wZs@phuM0$pNdAKF@{ zwww^kYC-0eQ+K=6@_XX`i@Glu!;`Hvf=BCm!RG?KJ3H~lJkiFLHgQo*;sX;JHMj9` zFqozJiC;x1<(9yP=F`-jecOyDZ(qLqYDQOSRFOlF<+>xIbBeAJ{ZlcM@^ld#Q(p1& z0k7qxKTjL8f1&;5Uh)>LZM5h7ug{(=7a2u=Ou+AfscOOL3BPrggFj+K>zUDY(%f}6 z3wH7x7oX#eHbl}#c($-a%a4DAifnsxit|o#(*n!E-(d;Dh;x%?uPE1tBE7O^W}n|A zjFN%%=uW%YnA6&h+HKKHh?4}bDlB<_L*B4pXweh^P^oQ)afd1&*YWi%_B&Gih(*a- zatbwN)HLFn*iV1so2!=uLJDu$wQ49QaNLxbj(xoR`7Ddaj%H7;DvQ{{{8^UgzXw3a zJ2D_V;&@-Y0(>1?cGlIbY>|`As_-lXj|5hKxXY>1s$JIrNzey&k7g&9}9s z6zNiFOBMaRG%ov?=}{zpaYQ8s!4gu)^)CbL59;%%`ek!JlpT>QP{Y66NFU0f{3?GS z|E<~;4X&Q2*oM8`#Nbu?ZjUo$Z(hjUmui0vlc5iCsCLcNZPLsMg)U@1pMB zVn54K#A~url}regy8QR;vC6`5r?2NLYnc2IX))JqyWjkVYm8Aj zy@X}^rWEMx7?W~Y{{8gu8x8LS^-6C(C957zd`g zfQ@2Z)*8v8G{5zVd#pfQFk!Z{)8iaY)X{5nwSr3e+)!vN9SK-5rWuiEkO!9Iw|Dw= z$G@ODw36Jy8Om%TgZOOF zraX-$e*Goxb$re9qm|;QW?t_8r%XO3 zT<=Oyc#-Mw@-IjkYBHvNVJ1JD zaan}to95HaF~nxG!uwM8PWBF3;;Zr*Z~t@Z-QquQTl(-vLUs2p{aY_vJ|=mJ`>YwM zXx{{x+Q#kc{j&Q4VCkjskv$*j0Bj2k3R#;B#Z4J^4&6plRG@LQI z(XYh)ko_zV#n%^`FaOqwF(FOm_^S23rzml)QE7}{m%1RVK)r=hiENC+xi=ee@j*MK z<|01R3ghmyGE#Gl zpM9lxQS%gp*y}QPb`HgXv2GJRkZ! z0@6h6>m~1Q>9D2<_sC|N3R9Lkz0CG@uE2Vpa3);*`x-Lrc#6RF>mm8Sy?$O_li;W8 zmsohqA7QGvmn@piQDFGX;3S^gN>##g_jt>wotlh4Ww~-0gNr^2Em~Ua+onpAeDG>& zu08%Ex8B=o1`qDgj(438Jl1s;&$98JBG)H>M>r8gqT`+Bx&(cmq+z#cbA&gGEx$8W z!?i4;-Q<+tcRgAP<=vr4I>Tfk{hNp9{Q~xh^lRJ#hc9>M3eUwp5>*)A)y{b|tkFh& zJcyQ0+NEdZ=Z}Vr0S#q11K%}n_wW4U^}N~P#KDUf31z24aaJoX=&ae3MX=;Fh2(yb z>aX`b;~%umQAkr7c+VpGVuAO>PH_Kxx46n16z`u=_s+vzk)_cgM!AApC9L#(vPI~t zB0q%dcPe{-s7Q`2htSmHU9xh1o%|3o`aY{FdB7Mc-SU#=vF4=@Z>z~hb^_JTd`==( zpY`t1Z9ATj2%1igfb!19RQ=D52UQu#A1YWMSl>F!&3~1pJD2pHXTjdNdo}%>rrlJ- zWtl^9@_yBaUldnV5A#`__g$P1Hsq1!qmyCDh``z*Sq5Kzr8=3D?ud`*Jhhzs15ebs zi#lT*0@)a4;Wz^`a;@Kzv+tQcDiK_R=Mr@z8Il{;gBgkeOo@HXAJKap_U^RClwYYb z8*j+pDQK(l_=O@BZ9yg9fk!G+!z#ac=hxwl-Bj(;_nl?*qPa3kN6rn2RSUZOMT$#_ zI)uEFl~0e+@JM>L1J+{tf-h3pvYgvfJ;`oIyW%oQPK+X(Ea$n|%|!0lMN?v67{{gl z8=+CJ2aWwZ?J<5wD@+J$JBgZoGeN@zYe6bnnB@K^hh@ErJLdJ2dK?-g8O@0@xQ(`? zv?bnrfyD{JZgBtphPd4YTA8kUErq&*YdUSdA(tVXq~l)#ByV70np$`3ty1ADxk8I@ zJg4pM7OR@=0e%rI67!LqEMi zhh%aM(XaCwFM>bodz)@?-}bHd(UA3F%nVZclt?%GD}e?%eTJxom>l~1Wckyw$Y*5l zgWyt?1UmYc*`7;3Qr`F7Dlj+ZR5|H9)r7||6#l#)cH?2xMfE|Ad5wL=V}HAzYV(+} z%UBOSYfavyN4M$;jo#1ST(k7GqJFfu68aho)%4CIhllb7?l& zac_b!ZmcBexxxLC^u=PYYQved6UX8dJ0tsk*WDNWrA#d+W{Y>l3*KqplfdT$F6NJO z+R_=b3w~}d(ds)Df4#nPdOH62*2m(MX{l_1BrRbDolAL@Tx$84EUv@8!W$K=9t%1o zY_)F2M{*z8@AxeECw`4kJ<{YH)iUM5BO9a+vzjIXjg+JcuEKX}UfFV-5%jFhTDm4l zfxap4H>r4ZH{Co#m8Iglz*eBqbZ*nf;6$0E#OXhVEOx<8t$(5^9h3CpC<-ci)^bhy zE3A)|?T|=wFE$9%cpoIDc_upQa6MH!`2DhIyv7v|U(l#kH{?U@DTy zRI&P6&imKnAuk7j4Xw++_oAR?J0CtFNsfRE8`#~A((1jv23p=&m7OuQ>bs?JHoti{ zS>^*auJ7$f+0OzsB`eYyr9WSD z;kn5u+l~J^nbG~$rpY;3BhrLx?!ng893ljP@5#I!QW~{c2vIp{6wN`A}!}DK% z!)T>w&+so*?mR}9OMGOMDPlxmeVl($=<*y26p4w^`La_<_hq=mjozhI^G>ADd6}_o zVRt|d&!sJ(z4XnGL!;q>?OD!O0w<1|Q5Kr}>hnEs@(;6^N%Na$F|iGM5&!r`OaKyNaGOrCSYfw@S(3FrwBvC+h{GZcg?4jlQ>GhtXz_wyZ z%Dc4m$J8!G^0-*;_(E=uWSt8ivs)Fet8B;f+VJAy{oRSNYr|b5?hg2zSEAmY(E*sk zj{yyQjbn@YKQxrRnJY4P-&-b$U9)d<_Wh0$cyB^ied`0)%Zwpby--h^n(_FA>qz|2 z(bY`+Ec%o)T@SAZ_m4cEM|YmBJVU^-+$Q-d9Cv&=Km3F3SDt)W#;l6SqR^qxt7ox% zg=a>&^N}OSFyx$?w^c40MO5ck(P^E{xh99vTL>uo(666BN;J?{(;utx)7zD4Af(M+ zl&PQ?3{AWyL&xbNwW}+9${Z2#_UJjjs+}!2$ED_09ar08ZG+muUA?-&0!KlMp8Q(| zMQ`~_`abFOx0hYmkg6&&itZnlu9 zb$&t`X@OYB5{=FT^9#%lyR*sp{yOn5`f3vF)Fa~0?9jox3RRO)Hx+KWAJgCUMw{`` z6^CViqLcFKwbk3v&ToGIjScI)2d`4&BH~4lRNA62h1}SSYh;-Z*t%uSwQm)hc97kUaV^3(gLAxV5AB>&6|f{EFm`Fk3(&cfXJRP3w0KdgD1l>@ z^2=Vz$*h=Bsm5fDY(B>lSi;%b@wBHZ_3FmWPFFFa*qQ)`u>Bhxd( zNzM7JuU!i9Sb6LPYLV5{r|b;PVgV~Kth;|=#ziq2ws;i`d|_b$nam;n~?jT|a~vG0|Ki9{K%>6(Xd)P4kHrh zdMJJ*HoHZtBH+wNo`ZsIJExxIX6e!g$1ctkuSmaH@^os{D1JzvpzOwn_Fu;nWd39U?|AbQzT((Rg%S`BVYQJCql( zM-L0~0KR~fP8gqdntgFdIz|zvtn@)qq6Yi?7d+dRf#T}y-bR~M_1J8J4VlNxKNBZE zlP46%UV`r=12tM8+BKHqPbJWIR;AwaPMAMcDQ%x1z^*y(Y4b2y^q(<5^pGTea_4RV zPEGfVI=Wwhb7o1F6vhg9lCgAbO4@Uoa=!dJShMFh5T6kWPd_R~3asm%;u%>Sa+mK+ zfEPM7jVrX&&V$yH^TMJ9#u5fw9E>F zU1j*XCDt!AWCnQSctb;w=L%+=R2vK15;(($2wwm7xFIOpjUVwvU#Oa<6o|yv*?1YB zZ${wUFrH}d`^jM}yV!Hy(#NM?AlxEwS90J7UMoLxl(+nD)E0xc%igq*9eZ=t%KC$8 zmFdf&ga;$gejpmuD>3t<{_DYNLNd zKN?=TQD``XES(nqd=x}7?wNwz|6J(n?mV^Egzqm@q$4Q6VCj0~`szDdUE^Bkl@fhA zRZ5zibTVHNOgcJK9S491Fy`%ViQw_T313JNKT-wSHM>mv>l=uRbJOn{W;qbH`dg;9 zcyvm-l!#5N{z#QkXTsvNaX>$##Ff^oKoBR7eDkXC^UJdaTeF=&XR;e8Mu;b;o2AP< z+#NI@D7REX8#Q1tHr>ssI^Ak-b^o0EsEyXq!Psegi~DWc5aRegqHm+mC{+ZD+D%_L z`rfg@!gp2VRf-pC$P@REo3q(Zx5o{X7)vtVb4gcD7VX4( zEKo~Mrdd(a^R1`F;Q%r@dD%odn@{%W-qM121$#r6P!j1X7@`s4;lzO!le#3Ry>?q;*HS`_9{S_V8VL)(O=8 zh4w*3JcaGRh7;0TRit}S+Umj z+@S1^_OCqr`zjJUS|jdV5-ViNefsA=ro7Q&@NQg}DXD3&Qm*ix=vUch;d`Xq>2k6& z&ids3u?sNty^VVVB(U%s95;6y$Loh6?UNH-%=Kb`)2=v(R@8BL8cRo$KeArY^{9g1 zpsL0bW5M1u%boi{A#>;BoED6f4Ef#*O7G*{X z22jyhP6mwAcZZQG$ANU|)5u8qrP|tBJlu#v>2cUa&&g*|&UT-#i8{Qe)bo?*#wAXR z$QD^CHhu=OD-9v;Quh;W!m;&n2_3;*f>1A3P)aH+^3v$Wd(=mLA!oZ+O@x&Es~|d( z^oG7s_bKXu;Uu@O!S6%^Csz>zCULLIZwb63`qn9y0f*bJG1>)M9b<|LMw(i4**UEy z`ZYkOc=~h{+4MHuo3PQ=8Mwu`7I-IMD+&v5kjjpHXix9mX34ug0&tge$3L3Eclo*Dhv1@WZaZV8%$*&CN}58ia(>~8LA|fq)`9>-AevJLvS8|Lm!-5bx+o@s<|R4;J&DmmKVlI5QleK4xLjjR6fuV&gVm zTb<)4W@UUuJK}2I#VS^4)@m9;JL-v zMV`GH?1tao>dNYOAY|KjJvu7vOcBS?)fu~dCfnV?8IMf#Q_a@l50?;m*tt~y<^8P$0vIe&L9#{^-kqMhp5x& zyxfI?@3+(Gc`;5GSRRdD6kop0c*iUxIYy~h_Q}_|v}DhTPA6YP*|8*k>exwYy!FKr#f0Mwjky$UEDeAuHTnX+RZHpW)N5AFRvvzXS{Val<0? zb5FL>r^5wKXDc;2buj9ZdUvtRA6;!ye@6zBx_7DkF57#;;tsB%qtFB0QREfpmi(NZv?P5 zyFE%)ZP))u*4QWVZ5SB|qQ~Vy(C~gpVfiku2_TS$Gy*$uaXG}pYPk)z$8wUHQ zHNS>W;xKQhsM>FjuF`LJdo_vr^*9NPoJvMV-xu$@c1_G}#L)_em>Uz6rENG3yTlCM zSL&3a8^(?^S_&-`Hn{;}vCS`Iwdw5!g#FhfD)1)_PR-PY>$SJ$1%Od~NUm?0t<76; zmN|TFV|&F73TsYEw|~3Q2elr&jZa4T--@V{#gb3rKjS638m zfUBbdEe?xKV5hF5_-BO)49Xvm5A0liXJM>{9YE^HQ-=ZH1Q zw11-})&xQTeA2|4pdt^3Q!pqhz*lPfZfJ96Jb?JHRQtD6(h@oeJXZcIGh=5#|Hy)p z2-!6{9-loDm=uU(mcwM%V6YG>HeO1#?Tav}{Y#(-3TlGrh9wEd4Kf((bMQSu&s4ac)}7KN4Cxzvh*A_D^uKQ1(FdF8c`oiw}&6kYjF zxi?o#>3Fo-s~tvbx{-1~O#xHD=qjcDP5J0bC=&VU#c9XkO51^KB?j@8%3g=h!DM)$ z&BOgaF-6{M<6Z|dNh2U+v@(-qT-7B>7(LuwS(0q2070n1sn*CXJ_^^NaHx-bEcq-X z)<2yVxXpbF<1o|%MR}p<^@XcS~GVdXgCk;eL={(gPtF50R;#3_^gx?`O0*S zd|Q9s`*WUfH5FrzgZJhr7rFlVI)Jg)#VP8%2kc^Ku^+v~r+$eb@<$Twod$)#n>lZ? z*$2W}2zjFX^+j*g~9< zf2WbpBpFV5u-4r=S=_sr@a7nb?xGQOyMmGM-V9co)&l+BLLL6KM}Szx0E>yX7mS1F z0ihSPeg;e7+_0mAqu3oVa2`57*jgz-7lYRil$pyf4OQgwM(al5lmr_r&NBO$q_QSL zG^h&LJLAMqN`}Lj5Cby_9g@29Fqnm!oQzI`-%g=Mw&DT2C{X}`D)Hn^r{}~dj`{qF z@)|S^8v+z!N{$l>4boNTYLuDD+JmU?M}B=Ub39u&Ba~>>pgCRxIy=8kRa5j0i{$vk zR5un&rm5c)*g&ZqLEHC)-z4AWXnUb|%!6v^6>$Mnzwu-kcWJo*h1GG?BL2G$OA}2^ z$&GCDculd?90zibrGZYlsB1HFEI*UQx*i=&)8GOIKR!NdKP@&l3L?0`-hp|uYFI<2iaq@Bd>nn?8Fj)8* zaEmx@!T{7>K`;NW7tk2}eF$1f#CxkUAbsQoaLRt{f+7#_JBxmviYUg0BO)L>foS77 zc-dm#9kkIHg!s)7;8_nkfYnUk77;G~%Y`Xqd@=xrvPmn^-=qYfC@G^8@(V%!_vKLD z0I?RCiUv){oKmArKBaLmI+N z;zhSO4Au>tnecS>$bU8M39+h5@YZ0cD&hr%{25TF6Sb1^@7Ldxk{ySp#;mKrl7~8g zM4A_@87pnHA7oc*z~}@)BbN~dZoT0H1K|himLgUO`Wprr#BgRX(9i@RUZ_ERx(zBO zS{mqgZX?ybZ^1{~YA=JN5Dc?~pB~7NarpY?>MDrIbyat~<2>pLXO^P}L#|9`=88YL>p2}K z4+q|aV|y$cp1vZH-vGY1I<`_=x4K!q>@pFsS34I-IN1ZYq+gyH0C4+#+5``XG zx1_4fybbGlf2U8=5rp$A8H{{BPiWi=2mnbbj-V^5faTaX5Sq)+CECM8?!I}`&uHC66sf5OEBz^|2j zHbu<=k-$sJSI_BxPvozF(p&(LAM-O8&CSjNL&9y5&!JSL4V;y-4b_NVY zk@)9ziv{RoYahi;)LoNLFkIyoK53k}9ueHSW6sBY~*N6V0z!O6Qu%GmH;B^6Vo}68=>? zYZDX76PG*)`HTScNkP_=Ri<2&+s*THoypc|K=)XBwEXMi}mCvy>-8ZaC`T^dmT>GCcqGn{aHNq?1KRCN%ue zg*g$epnF6{3KeD60smqi_!n0oD*;`^bFd-Gzx?HD00h`D9{EGND;tt($AXo6<9Io+ z_CP*KECrUr^F9;d!5|=TGT<>;2|%K|<=QE@Atp<_AQHA=aF4Fi;?8Bj_Pd{e+gN~- z;i57+PoexZ;sw#us065Cv!r9?5A4k%_)S(2U8L{3*nyHYXHpm}PNtpWV>&Z0Cnx^w zlprZg>Xs`76&vIzgg=teAy7a;+W`89y3ZUqjR-IZ zMB`KUKf(9rQg3dIcAgsBgxmV(vqxQ!ul(fOoP(COZ#!|khuHtG`j%NP(8KukNh=Lmc?p6dc%XWV~ zrqTyo+5rM)q0ItBUDeF=^l|yKf9|&G6zZ8K0rit`J_cS)otDc^K}|UUEZz41#HU03 z7xQzMztE^w<|rp08a%Ye6?={VEyMw+Af81Qp+g`Wu*j(4LD$=J(9p#q?oyWnQ(y%Q zL&cd3*Y*t_!n-bTyx|GK!a$DL68tkYlgBUVVu2;lrV0r6J!!mFM~7yArYuR`Y^qzu ze#g8e11$OO3uu0Nb=XYSVRxxp&B~sSXsE{)02GF@B`2m92Fo&cmWO)b)lMsR^+tXR z?Bbvs9Dk)tgC;ZHHVFbMmnUxOEA<4MexD{PF*d6aNzoKoz6{lo;s%-3MuJ zN5`vE2@-G_hTm!f2~(FZttKARif@ z)xaBc^(?;AabE3gVfzMNH9*<~Zkem7^Y}xCV%jerjP+J}L|(buq1V3Hof(!~HThF~ ztrEzDgYql^kmGY;S(*NNlpDj|jE&55`X0iGt-;dR5p@(Mpx2jJ7MAZ%=h zp2{TF0sYYKUq$eWuLdDRU;)1J%i!YXF>f)n_IC)dAoO`jq>lcV%kPHA0kdZT_~jWQ zKJJ1VE)i>hiikBBff_Q1H2|G~l~@B*6wyi)4U)k-o(MIR7|f03f}GEwWzf{vbY^?t zwqJ$T5^$d_$q`T;V&IdBYJ}HKP(-W)DkAn2;MC`dG`Sr36OW0|SO;Dl0#R|O==9&d z(fyka4jj}Xzz>0=$20O+C^Sdn?An^8nk#v&HKsD8MA|Gv&g z88VO>3@ANUq-0Os8Y;yn1CS%D3abM~5)%e~qxwOzLcruk?E3QJHlL6_iOcwQxU=NA z(334-Bl7Hy-5WrYEjQ%-X0L)UN~g-b&&w&^$5BA`?33MJsh+?FX;(WTki4;oL_y=# z;DHxsnctTz-G0Ls*}Qi?eV$3mmeQ`YFPO{%{d+%Jv6Btzg>YIa>@V4H2YSq(b>NFw_7{;Qh? z@HSd=rPmFropx{(x=n#6g}ivdj751);*370( z^-KBqIBa;xI&GyH2eOn+N*Z=i1RFa@+&yXp>FRDY#zr39vQEQ6PlT+0Rto*R*P=|9 zKIgg6(|Yp!1rTY5Kt}9xpjYRD2QUKImBW*5At;BtC~{V`SM+4OoVSSpaCtc=*{PgE z8C6r#yLQy4sal&}MZh0HA2TP}y9KmkgVxt{TgS^>5+Q|m={@LHYjgE6cEHGSIogTf zEbH!;&HFQiHTSE_Gg;~yx`rk{)sAeImNK0LWWIc;!g>U-Pb+}@7${+!ED~e}L&|~K z7}m3bjFNueIBuB9d(;%9b~gq9JG8na*tl3_rrQu^>mu?uLqfCg4%+IBbdP6H-2*)To zQ>}_JjRzVyOO_1+OUwvMr*#-CWfQd@;Pd9S$N8;^{G+$s5VabMg9o$qXHp!zyNqtz&{Z5=x?W>|Ns* zKRW9AN2o1eY=7g=m|uAs5`SMOUn6^Py5rp#HV_XMR~r!LCI>QE5ALbPuWs%AOlL>G zfpDuev$*`p(m;`Wk@$wmq(D&yJqj@Joc2UM%qcRvfx-ps-rG6?b$-IrF}ICRv#IE* z=jR$@u!ntznUwULy-QzP-lU`}`7LL>#p4fag(>|W)?jxm7OS)va^WS@QDWnzQH)~4 zzH=#a1L}uBD}DnAw52p=!X9U3#a)DTC(C#uHtv(zaJY!xt!_H$?C0Xvy6!jsvDm-l zc8BRxg!M?Z^B!(fBlbSnNr4DbcTAMyjb@e@mY{#02GFc-M2o`znHJ&J9bEe?x6SS* z_zc+q=Y^5#k|mYVQqz}950<+#MZ-OkLz^Tq0s8o$p7L^^8v^qB+)=jA9(Dtnrnw?%t;!k+@0b`^31ywq6#dFWeC zNl76_hPfT_x=kJ^S27W=dqfsXO6S(zfK`SgYl3{-dfzR>ZtUydG`3mk0&>qklQ9f} zEgTN3qqW411<;O>Kg3N6NWw|+zF|xN@M~6go)jsA2$=_)6PP=@-SR9$X&}(Hr>Fk< z@B||$SUXM4@F6%jAWG{w&})ya-!G|XsX_e4U*f_;sl!X<-yjdgYrlmB|6Ap_KuCjE z)pSRj>d}5Wr1uWwYjTPesAMh;S2-?8!XIo4y#ZThi~sPlh^~J zJTGu7d1A0%*LVd!DeR_+NRJR`I(EagKIPT9y1qCe^as$@$kun>kX})?|AIP_(~l5$ zJax8+Q6#qB#zHm&NVDuvSW@r!U3gtoZXh9UN?~t4WWR(Fd7|eq^#pku79ZR!haib~ zgx~)&=v22t|Gl$wr~1j+Ov*@G5F=jIpaA=UJZ|3a+Y1=B&86GtmfOlF-O9817jgEL zlYM@;kXsyH0{C4qw{|s(QC7LzG~Gv53GCXT5msODo(Qn69(9Wp7qipQ%+e)e=O+aI z^I)_6CPF9|C3+f6r^?l_y1EYB>kRy?`_KDrJNND17jc>7rx~_Bf1#-=(JxZcxZX^O zPi$UT0x0RYackLXlV2HVOa17%F@kWVJoTXU`jESI+)#11=WTYQNoOqDv^=8nkt^tgA{XqlPZ6A{`%0gvq0n#6VPWJE zYJtAetiMR<(gV;;fs&AY`PSiMPBCpe2{AF8G4FdUWj{Ro;1TtL zz0x7d?veJPvSAPVl`T(cwPHc{wrMn8YivSVbvy-5HrPBi4eVJ!mDMz9It??Eh1_Ik6K-F5Ln5Ith_y}>T z&70<9p<<&hFE21X=))-fD^i-{Tmv+e(LZ*ki(_@wb!kt@)~@~I^Oq?9%5tL6-Wzp~ zsS^%%n(&=t64QQi1)^G9l6UFLg~x)ezBB-JV$4MY0XO;a{!Y^ydM#J~%@6vEKYLs6 zXkyGF;DsKON{DR@T%yKG;_M#{F3QssfY5>8i>m7AWxJ(bu+89N12Op`;|t7&3wu&0fR)s_p^DLmW}!S(>sH#0<|`f$>Ud7?N6Pimd&n}cxcU|+?9ZCuXQDV~YR zudM;tRRc$+jzl?B-TR|J1N}a$_dAFlB*cZbKu4n*QxF0s-~?<_A-)80I8l~6XXimAdmmBI07St&$Ma-x;X-* zNO<)si`6cz@nkLRP;MQ}rHv?XrcCEZYIqLt1!9enrm7P6?9V zt%*$d)5`CcRCvoYL;n?w(~)Ho!V2El10#&Iw_aXqI@m zc#42M)qb?rwB~%rdx}NO=>y{VMynGfJ$eD0s51e6oy#u}{C<8qlcHPzzcv4)7?%FM zgX`Y$Bo%UhaD${ z6v0APKU{FL{oy`jZ=X;&Q1HCtRILJh2X_Iclp4s@jS{gpod0m!G*}rW+;Q>VZ`fWy zC1QIE?XQ&->klZ!*qRF%#JWULb@GRkAvA`+x@YD96q;94(Y;lisv2=}XIla;o*PMi z-U-L>S!mB1WzgXRgpe!HUcF-$3iv@ZDRCr?;h+y#GkdW0L&$uCrmBfP651&Vq|gt8Ac%Ov ze*ow!kEqQrRwO??$R4nT_HJ$Cwr}b=09ksq(a0Ha-Pf%xi3Umg5fc)e%)`h02Z(^3 z{xXmngENk7onqspr@DeiS>ulne2qAnZ<}r0W{3=hWBgol+6c8X0bm(VF6;9Kr_X|13QjX zfjGbv8ODR`9LYU7+}XlWB-T${6S1R(l8TxV)9{H9T!5AVNkL)&Y&t{oH$89j4faTB z@+RZgw}iaOc2#~I6iOZ!mjH4uehSq54mO6H_EBdSWA-}+5h}`vSE#dPHT?Rhush6G z48`=NXmmL>b>CyDK|;z}V;k7$sK2~+`PW{LJ=&iC5|vo7rB!!^ES}z7^E|r-1FBd> z;}Y0Xbc^cY`X^%nl3d*Rz(*LZd|8EoMIwBpk*MGlKzZa>x=4_tNPTZl-2tjvz<6o3_bOQJR=}|1lhOzmC@OQZP^)l zlJQfG zi9fSMPJr5_CDlnznyqzu=IL4Rs8A_M0H4_{4_2adk`NoE)IYxz5S*>g_!OiwZ$G%41`KL3*-|Kj zkxS7X>@r;Hvm)Cc08(C)in852_XGtnLHUpD!2VA>eW&)lj;6WpJ+-KQh;dL%YYSxs zA3ecdH&hpSc4$N)9q~eT3395)@Zw66X|HFr!{}&Z(TV)&iFYvZ0U%WjpIE=WU$@q{ zR5FvRllKM7)hU);oA+9yFR*v5=$F>uo{nB$-xZJtmJyq%40SG;08haX+jaDWmw(aW z>yGg6%Tq5-cw)Mc5h=h^+0Xe;YDwa?s9;UkDhl*k-P$99dQ4**Abp0P8-l;_SC3A~ zZL@3W?Qeb;bo}|@zQFG+y9M<@HT%&uB*7n-o!=eKLyxnZ-WFQsJ9*yqaO=%dv{qUD zuC5txoZJOwu(#9;M1ruV{K&GPfJug1x8>sY(fnNRT6&r?*@Cp1I3Pu1OT5AXU`KLs z`=j|{{%5Yc$!D&O0)`1~OC21obsx)X2jui1#;?yZ?=$T1Bv?IrC>R(LkOZ|%z;{ih z$sAvR!Q?r8m95iZnQHO#U?F>GUd(HKf{46^7_4$MAarNstIPv8(wju7h##U`IE-?Z zA15Yf!OaUzyr~=!`>aTt!WDH_#?!E1W)887B%|wzETR-|H9k;Lu~H%=a&fH4~=2XIvKzYcK2s5IFE3Bhu^3vRC#+_ zhH{R$x^VXfa1Wg`1{H`=>j@Y z8d|(MqDHAuNEY_(!x@@9GliFAissBEx1qh)|M@_GSXsrpVYHV-)WsYBX=Xd8c@O08 zr92lV#B#*-Jybz`e0V@DD5b?+Fz+S-b~z)*j*ZV=cads5XVl_@=dfT0G1G9Kln@VJ z+D>s&i<^jf1Dd48$4~D5jU_($XS}D{X$(H?xf_(H&Rdk@%as?J2*edg;m+WEGv^Eu zlmB-9KspC+#G%_#U;dw*gIo?kqHNTZbs!xWtm0x(-t#~Q0*MjR_@^K=;n(-H=ig5X z5dJs;vNjwbK<|$@4h~VKyn1xwQ{Z|jiZrNpzdXwM?dSy*$o|(E3ewple~yoy3L&U}11-}%iHvJ)kO90h?41`WO`ydI9|lg{1z6qZYJZpP^!Q0N7Oxya}+sYQ>>BBFNNi{4qoRpPAZb(+!sX4QJwD8DKu&C& zCXj8mS@;Km3ATgwUl(Yju_M55J&Fm@G(9o}fkluFC^iE2tlNp$4|sGfK(uvH&J;NH zadZ3RiNMD`>>^Lghz@%ca5#Mw@W5+%UJ_oK$3&&k*Z6dt>hop{^0);l7icQML9U-| z@&C+<{m)T@t+}phNTQKY?IQH1;Vp0o@024d8cPa_2QO3rzW{2rGf=)Tz$y&^BR(gY zX+mKRj$FW7-v1qR`dB_=0kEl`TSw2q4q*Jv1Iwk2b->`}hR#0Vn6BUY>0V9N@ycAf z?Kxn4E5K9Tsh;)(o8pI0Q$766j0@Btmt4qtH)<(OvAK|&29IHufFyNM^RMT<{eciS z0yK>i>kw7YEx;J>wlWwGLICLm+-b{+z@wrR-`RigLKHdwG||Ld=06Og){$kRas$za z*@FNOhyEk^7ET*;mfOzw*6L|~U=ZA2ER4E$?WD!Qxnj;KPX-}yKx-S+273)kd@nwj zPak9MUQa>$Qna}=a_4FS8XsvFp5XdnT)B)9F zUIu)4j;Rv5qWbGpfL;VRG@a~5I!-i>O}y0=Wv0xSu!Vx|@E<4m4K=#Dqt&_-{xZE| z8t+0N&Xz%v)WlokW72DpY2ha#T|gtFuKOKKd5m#U8mG#C4k<*wlKB11CiW(9(m`DS zop%0i3wx3wwfItD#W7|o;&6kt^*%?XakM1f2g#2qWu~dLNC*Z~exxs76A`I=qe%6` zy>8?0N@-GW(^v>yrXV-URe63$28Vjvusi&HjD8l7`|4L07>CbselXd1SJWheg<>nH zmj2M|$leNImk(m5b(y2jo=EzF?^!ATF5sS~NcqmOP3A7fa2EVINJiJq0Y!{|UE(pt z0~LxRR9?6LtLT^qo{Rei8U}SV?f znMohLa*K$DgV|&#G<9S^LI2epSRIQ078}Ab+VTviE*>lz^)6Yy^a-zKbgKCbWqyf+ z;{c9}4i+p9yO8)^&mblFJy)#-x3qddqrUjb9B3S@B&fw{bZ9|z}Vl+S!mI_hdi zKfF#=P^VYxmW`I+ehS7cpLC5fA3sahVEt6OO~KTdhP@!|;13JI1{^D({*mF-qxtlg zh8KR?mo!)j)~^RIQA4MK9GTE)^9cF^>$?o}?h4>8On9;Z|6`-dwwHr&^}zCIgB2V` z&}103c7N`}A7=&r!ZuPiG%Y*wfo?n3GL3)1oRWIlX6DqeeF;Xe>9~h!J+wPQD{UAE zG4}FIp7D%QE?=)&4-zHXn||A`@()Sqwo`qkB39qMpTCgFIu!WuAxMvKYRdC0VIA_} zqvhzIsLyNK`}tg%Hm_BOzY)l5-pIY?Jl1+}L_%?~yyqLihV8fZCs&1SVhL~N)g2R} zJp)*^)%OR>%nOuB2MOj=1#Q^I z_`lGBSmC!@oir;+`d(U0ZL2MQYm;z|Czrr+ikKnF+hsqmYHZZqHT ziSymLTo{vlw0!4yf&B2*MI~JFvnsWBgyo?M%46ay&x0K0e$@OsKLyK?GS)-nM;eQ% z2}2S4Ybgf5{9n;}e4yPGF=Zg-OF2*X#e4Q~=(u-|YFesxTZAGdy?~+K{91fe+v6LU zz@^awaJ&OW=bg(0X*CKX;3n!FTU?1`p)J#|uxbU`ECU)`EQ1JvON@QIDM*+Z4fsdQ z*{Bfsux0p@qq@1JrR|-P`pwsbH#Uc3Yk_leK9Db6)*8t`C@%!G1Sy9P*Z`P|?PiA> zxeW9*ub8D@5dNN6Kj|WzOb}`V(wnxj6yBKu$2JP&#-K-;sLDaI3Um|>V;xvuC;Z|T zTV;ra)Eo@_W$xGY{+_Fu1Mumb;vFAKwE~DqD->tW$|EqA z1A4KSde5T$5>m-X{XS&Q<{^0ypouVuyC&#?2-em!gY?N-=NT}#?1~&P{ zSlmWt2W3JR27$OlwU&TVpr6G|cYs>%)^Wi1*?_vV)&Jr#r<@siZKzfM2 z>5(*x%#QfHclU&LQ*sqJm*W$`PF3_eFy({`zEcsLW}xq5ia;_VFeX~j5fq1<`0s%9 qq|-u~9{k8eqKHAb0QSvB#;pZ5Wv8|kiSTZDuIDPh=75C_rt+}j$CrR_JTgZos}en!79gz z4naSl%!Ooyz`$za5MK*D$%Dx~6}cajNhKs2`; zx?!`bYhBXHJ1Lx-s8nv@$RK#;n! zZF=2?GD`lB%*@QS>v76`=Nej4xIoa$A<>8x_~#2LoCF0~D3lNyln_@bI`hXrM|u=v z!aw{uf);BfiUj}n|0?80l7Ef_=M$id{5_FqE*uZz&xu>b;(hV|hGPH!d??3Ayr<); zPKN<}bJ9Fz}9=5^&qx8tv``P|Ri`9k~H#eejB>Wds5`Z7trJmPgoalsvh>VPk z(Ior?v})x{jenPqlnWNOwI{=#xz=o~XS+X&+2_F_!+D(h;qCdH+4V@p*;;pddz;Ci z3rxrV^|tBhsEpZuop06e`Sjs-LEmwh8hGGqHl7hN!q`c~YBoA4e2}EB*S(_}rP4hL$!VF%in&-=BW^@BUakE2RD2-L)i-8gqlo(7+NI4-Qi?8qz7v_C9MK zT<`Svcm8FtTxaP<5j;xK>N{Jkc$aS1@=NFE3M~!I$x+^7w%?Pb^H{(aNy*85a@jOC z%jtqI*8VZC54l~`$sZ&v0=VN%EnXfjm&L@bX9Dy}dkTpj^uYDiRXg z0tp{V6yDKX8E$3UmDH;L`|BRTX5)rmuO%=uw5Lbd#MCrbBn(C4rwRN(EV1<&_^R^+ zzebb8hh#>5yULPZ`Ede~gCyRgI;Ej`lK-$)he{Pn6^Nqx?kj5gT~X(7WJA7+CG zxaf?iXlM|qsj2tAfv1$bwY9gS)z*EJ8^;o$@JO|y~vz7IXRyP z3I3WCDHkFzJU59}-RNL4w*?&me_Zfth};Xb;LTEa36;97N6aA>$OL>IOnN?-Vr}nl z&#J~q36?WOTOqi5lY@9HCgC{hwdNDzCME^VVSnvU_}I~lLsNTOK_VaVkBv@xeN316 zW3^CQXk`CS_A7x-flV<)GcIf?R?=Mbn5dfk0r(O2%j1_PjZ@3mT9rx9Lt=P zffB{r>HeT3Yho#=??rTRPH@}IIjE|Y(PzzPYv;^(G(Bb|ayT+xKRGV&C&XI0Cjw2> zF|2-Edg!VN19zCnnt#+^hZ5{2E@lcn9W|Yxm0G&;*7K@*?d^O|#=wc_3X4(tl3sTU~3-9^aLsN#D zuE)T+H$pRvM4J%C__F$1N+&+gNB`xF!~J?LBP#p@<==JPI>>8k(`a?Nl16ry)6*nD zkV3wXtT%#`f!aC*$iCL0*;txd*uqHQR<%l4*!*BFWN&8@(IVi(DGk z2kW`KLl|mH{d+3)Pdf5g3|OR!9g3=Ic-B9dS=AJnja|6v=@jO@NNQ{S%}=xPE^=EV zjY7|ua{WY+xY7ScnlQk?(8u4O4AjRNPDa;D`cZ){#EvTg7w-2GP4Z1{@RljbDO=M{ zF1DJ-T=JJQ7`feN@el(T5-ur<4;Q!ijpRf;|97mp$gjgetj8|o*c|&53{2tNm-JO4%*_Q7pM!>xQ?#asuQqqow}rd%dD>t&XbUDkD-P} zqv!6K3i)sMgY&VaA8f`K%U6n`Wf#+$Rka0(4AJbu)x&YbqlIdUut<@xiXOHcsVuZ0 z#pLh1&JIXaYQ>(#qgo}4gZ?*01In7{s7&5&pHxYa^netwVU=}tz*l-$T0upz09ZFa z$yQ_}HH#tU zOb>@1fA(DDt8*|sVhj0hOv4T&ik2Tcnzfeal+Wz$+thnQ2I@5I9d7>&MqN{QAr!K( z|KHs?1Pi(orGzqk8mY(!7V(BDYYA`_hI7ShhYcDlT!!8twzE1 zN7IW~N3oAb7ic1$CJjsR4%5&kNxGkTNRWkQgo6?ouBT0Tg4#lhm?=$pGq7s*2}k`j zgdKM(qjK?k3}d76-PLIwVsB+aGYe!m!mmPxK{tsz!%PaP{^W3{@P64p5l7H5lApwd($oY(+O-_Wc7=oLkO4eCG1YPPltUWQZhP_^@sVvV) zXw#5(Z(`|5F?}T+pQar>%g1iIh`KvA&eat8%znd!@4_fg;JEG?W5hq4#5oG$5n;8h+lXQ9taTMptOv)HI(`7W7B)- z7|k3Vwem}BNh(PW@Mo3fp}Ut%uoI0WtU9bb7Zbk4HQhj%;3ip48N&nQXww&?ys3mt z;64%#)0_;c4BFG6zhxa@Z3oG9ukX#m5{t~XrHR9Ob&t(8+v%Z;$ccg?bsc>w#5kLi znHmM|M|IgB?V{bidm3Wt!o|!Z$Q69i&4<{li0vX?$JDw?Si)zOu)wqi2Nx0R;jCx~ zdO0d3saYqPcut1DOto2kij=^NQ{=F<04!i3@OCc zy0Hi_LI7x4ZnGjNI)W+7#K*Pbh_z5NqyIh)DDP3Cu#>7||0EKTfyBPs0!bV-cz}Nx zmWA$jKyntQK%wKque${lMGc8#zg09UY#@xjc)$`DD|M7gAplldZAeml4QyLGer@dq zrQ&~M3<`?Nhtu(;3Ts)#3x$Wi{k~&MU&vQ(BkXpQ&yIEaY<0o-`H#rfZVYRkc)!cMu-q@b7 zA8KSa%eZZ@PEo@(f{hxDSPdCQ#6HsMGKI@}xddwh*rgIvAdJe)TuId(<8>8O^ z*oSJ0dhnd?ggQ&JHD!CM&dMVhUXRVaf(53i_!J>hDrciY)3G)mZ6|s*BQhtgO%j{u zxc_rwxoK_FeC6`mRb?Tp^P$EkOqjz2%7Zk8U7j!)tgP0?1LE|7N74WwG*H3;mEHpn z*js`l2%c#LMTKaa{p|{2&2pxvdtla*K^-12M^Kx{7RE#YhWB-4$p)!w*GW;6d;V-} zAT8L2e@r>Ah{HuNsU)^TqaLz3g&YKa=(sV)0B$I zSt%(;hF+KhSah+~(l&~*sO`Yd*S3bX+>L8U49c*Sul?W@N6w_ed6W|Mtq=fnZf4}Z z`W;F`d*Wi9#`(-!KRfom{V&&&pncJDX6mZ9w66K&8k2Q4y1Zn;*YM92 z9Xa^$895epZTK7xq#97+>ugy$qkz0I5`04OkahHliUxjaXR`)5)K%8=_iks5WyW6 zbW`kP%f?9(3b#9|DNJA=DiG<3e^P_lMf%XilGQaWpjA;yvRaI#RWc%5BvbJ>oUZJb zqjt|`eB*tM?XaDFB7{cTDrRU=aib0#*!jA=)!0oc0F=ZtnP~l{1ptbgS?FHHSVc`G zj}(aIiQhjoLMLwLm4NfvvUIm-?mW=#y4n&Q1Ftd-Lm*&l{ivSnHb6c{qP7w;L~;I2 z_*HogyBH?Gbu=sG0Wvu*;);3iS{;Q`2NbYM)6YO6&9^Dn3wEVM?yM~#U^C2mrQ z5Is_T(2Up36O-?-%5Z9~Szq8B!Zgf$sS6o!GFE)KScLr^nDMRjUN3Lm-o%D9&l{#^ z<$O{ADdKM$988WL+w$_9wQ;#}w&6lghZvb%!)(k<<^&ZgiZmrGY2=Q=PXrZ5{>WmH zn*UH#MorHV8qVB@*dK?(Tv}Vg$Wc`2(X{MBCu5U}Y9!)Yf3cM~g2gvp32^<`Ww3b@ z>{9@OnQhh7Pi2H-Y?@?5YPJE~55#!JJ7x5rHGhQ+t?K6#K@X`^lsEOfL=Dt5H(T6) z^A882n3;{r72^o{h3WXzvHwd;dlH1*964}4(oN#zU_Yx{aN-#5%?}kPa8(>za+QpH z@uDNiC3tI7Otq04@5h;I__S}h^2TQyV91BKqvaAE#Xa$WW4$iRZuZzzmV||WW|gIE ztY_=&DkfH_INpUOFp!9oncUTc+fsV)w3Qfh$tbfnuUx;rF`}#O5a`2y^T0GJYD6So z-8|Z9DL+rFWq^kyb7|=8DkKv}>hatt12vhKP)&hZ;*FdsZS|I5|J1^4LPto~~f< zFO7yuQ7cqPwnkd@E+(S12Tj$eSI5d4B6#>FQcwM8BiBYGZHVqG!riPuhL-Z6^bdBw z6G_c_0^FqH;;_YgaUF`_a4J2)g1)8M^JG$S-}8&+HMoxi6U$=g zOuICQ<{fG1t0qS>%;8}Mo-Tf51%}1^P#s7lhXO1??k2IuQa9F>V(;QtFg1>5XdMru z;F*im^nTr8xm}t=XFGzJArmp2&2^8D&0lJyM}$vjO=x44fCX32y^B#G3x$ZBC-m!J z%VF8g$Yuo2v+%ZeF`d&~?i2=>Og8*iH``+So#LHe_s=uByr8nS5N%P7$ru8N1T-3j zwt~=3ux;%4pK=toio42osC|AWf(`c-73jFGS-trW_Gi#W*QIb-V@)9Z`5o&5tRhv_?Dqs9K_#GdgFi(pq z)E!>voXVenAOixcVOZ#ifUv5lgvlZGa8&@DglBwka6?!AwoOmH{p56i zJ#P_j0NgQ0kUn-cF^{grJRV0dGYKv)=NnQJIp)7icT!DqSiWb}mi2Ni_PkWw`L5b? zbG8z=ohN{M*o_qTk;*9fbi7M{Wu}IoLOw+}=ODMslKSLYqsl56LjdbDxr%YAhjxfy zMaiY79T70_tov)WEj_iWYi#J-z(E=YxZ(rlyI0u8y5cE{tIfrmsKd zvqM&P0@RdN@Z3<$y@zLAiOdVr>OZXWQoh zK~%c=X5Na=^*$#`%35FDfcwWSYO$RpXQlZhxxuK=lkn~2leE+4;V$`1-)RI}p&m6U zTI^=zC<2LUa}KPISV*z>#YIWo4DeR@DLk~%(*EONF|=|YV42?t-u;=f|3 zThj!~)IjcqC=xJ)Sl-@QhansF~zCw1Nq!qU}!QMdd&TK0CXt>|$9H z4RI{*N{kU}SNcnXJ#drnKYwd&03Y4#y>A6Bs-B7sCadv~m@gQ2&P0+4vAhlzUiU;7 zH_}B$|8=^Co?>EE2{So8CkOue`g3tw=#v-e=tim=v$eJ&oMUH@&{GGW%^NQ~{ugMk z(bTU@_WT46pb9W|y~RezH=rZ>JN2k?;6S%gzF15eCln}bg$&eD7;if`2xg%|$7O{U zQ!kWUZ}G?jeidK@2SxnL11;dKl^3=%Z_f=hr&$tL%6O;2b0KdICa{D+yw=`wgNU%% z2fwxW@(RKCKGQk8^sv{v9FzdO1y&;C)-bDk9I?2n!Q3j-{_YpIa+&T=6oE?jztgOb z!+-;6a^O;Uj;jRx7ul9$hLeuYImg2z3$pO|?wO{h@Fa4#ZQR(o zJgT{>X}~SYnJ&fGWh?mTSMGif-#3)e7s}j$@KNRfP`Mj@j2#NPcED{73S~Kh9PMNS z_H$XO7pOX6_yj|weDK>0i<~iE5?HWb@<_0YVRcH@9b~K3Z75m8?sKMM(B07gHRZi(ZkaAe9z$1<4%{R0+NkBM-jIpB z*!=40u>HYfJ(AD`XMdBAZ6VXEMhc?ZCH}*)vZu|V@q^hjXE<_rKc~4Lq`bO3XyDE3 zUGK{tz)HNP!ECxwOeHVnZBKz;LYNK|vns!(If`4=m)`X;{Gl)OERA>$eZN6j%k9vJ z4VDN-JRO|y|0*z9!`-)}rmK6}+ZR}%&({P+@>?J6t=ip7kBjFRudI@@M){M78Q6ZJ z>7x^DDKHAVu^tABuWnVcWy}|(Yq|+F;8G8G5L9{AU;C~=eF7R3*Z0#p{c_F?AIzGX znv1Keo`b3U8l!#$XKCHx;o*yiow$7O`!lci?jTs>yOX6owIB(AuZo~7PlJH(Q>r#} zH+P~+qD2`Rch!QYtsETvVp?RFKU44|GNsF42JgU5`3GW@2sESDa{l8P=JTDghl`D< zi|ga#YEQ*)IKrcoN8U1Qa_xSkm0pfS?ZRS%Eg=~h*=T!d^ZmN_2Shuy{ z@lH18*ZisTD-I~KMiWGzM_4d8wr!7ZMaN=~*SqV_1hD7(O7)O}Wji*KyK6iDS zMdxu#Pz$c+>RCGl$^Ivrk&XOtmQHzk%GafHyDo`)`T-y+WB+Pg!vT6W>-W+SG|0?X zkv6pXU-Gbg?&^8KjhZPuo)Xr9tTo-%j!KUsyy)zCL}vaVF%S|Q`lhW$6qJ-Ocw7#0 zbajjeW7!uV?pK{raTs*McSlmvtxKC9wxeq7)_80m4Ob`IJg*(*HO)%^R|xbLL*<&a z@t`K&db>9{H#dG9{~aC%CT5M}j`+j#^Rcg$h&9H2S&T+#@Ul6Jwuoq5oN1 zJ5?x++jcW8-E>r1Hd@b~8~?KpYpcJYRjx07Q&Pzfqv(wbTWnX%!(+6!Dd!$~FS!@7 zGxFZ2mPEg585V%5?&G?qfpmaoVP~5WgLrSag}ZXgC#|T_6GopW4YoRa3F-qbBZg@q z)jv(egO62KR!V;P!n}azz)e&7W|X|OwY5y6qM-2g;laaZtp(f9&u^J{9!*F{=m6BW zbvs+NtSpHF0)ZVCv9f4HxtktFOPE@4$WeGjtV7EP_M;mmA!zN2_BwoZzl5nAy>Z^M znnfn!1ZoK@y$p%uIxU$(YOTWrfef{}t!^K9c$Rg@;l|9>R_W-ZbfDbRd5Zw7y6|xgTQpC8V*+hl zgViT|Zu4Lni$F^|-q<(3=e3v?B=HG!bUZ&RTx#bJApa(lF^$PYpJjOkYFjfm*enGS z+BFGj`&PD1;|;}H-h;eFXBMSPrtkfl`(nE{=iA$}y-XILCL&eKJl$Exv)!rLO09XE zTX&}K-O@6I>y#Mo4{tc(8R{~+USPUl|IP@>!pH(j3(gt@VS^}%RPx#ImQQbIZW0xKniT2Vqh2$LFEtjdwsG@5III;{dIs)yVz)-jU#}R)3F(dh(cXP=Q(ttS5C0n~te^|7dnzfMa8$e#Q4hMblOKzo#v{aA_`3)rtS8JNMEU*=7i?ui zFD=pZwe-bM9O30k=&RQc)qTNcE4H@PbT>U(|ES)p>}{8##aZ6aql9db5+D|9Sr@FH zw2ocFXeZP4d~*~qJ}VsVD1>|0f!3Mvia37m`eukN>~r^E>WlHI{UrrQn6AumB!d6q za84FixExvhgXn<%qI57wUm{HV$6fbzX?t#Ugm}31*Dj`X%2yc=EJEf3KA*?0dp$aQ zOE%XCGP7kX7xK+}x$JoiO3z>yKfrmuYgXmav)H^OW_ru(eeVDnLdIC`S*I8hyabds zOT%V8d)1)~GsYHi0I$Sq1hp=Zs4izaZyLd)c?HG8$FJW{FFX2B3J%2Pb&u@@5>noD zjg_(k`(PjoJN@q1;R^B-lT7W4x1~wr?!Du5b6U16(@3 z`H(K%b9)*=a8YGHtzJ4RG%rwZdmiuOzBXouHLHSJ`$NKg@tEk$F29A**^JkaM%Q@N z8C#3}m0M)Hk%R0%r+h6cENTqS$YasD|0PAYl4)6KN#w4Mn@KHhJvl~s*A+v_HbEdN z1!Ce4(uo&;a^c%$RdU^L^rFvCO{~&@v_ilC^?iA+78B|r?W?>AowxT7Oe1rC=~C`I zbuR#J9E0vmv5#5|d6llW_76iB7zrX5neZGl`7vW^bOg|`y?owB#%pZ*RcJPt!gNgcYvFQ;&$ zGCX=|-zXnMwf=r{WlMS6cZ%uVuqgaJTeUnSj^ZnVv(y+pOqPJ%fdC(b2&J32!tZW; zl+S2Tx%~gIWA%)cgbU&|e|Y%sJP%R-nmT?8y_q9BSWGp}JR;k!=-2tJ^f{^s&Y1H@lDR|J%`B2TSBPLY{00CFLHS zN7TgHx5$OP)2>ia5eMRMP!F+<#TIVh9)WX#Eheu`sr=m;z)`HGWYEW>I1%CwKiLEP zZlx#%*UVmn@&hE4`>*3wd@AI=h^yT>c~*V$&8EFFNFO$n*v+eP<`!;xJn!dpEVs5& z9o%waV%dTxEjmk${zu`YMfe*x2~f{4ODkps+=}322mIbFv1RBBZQpV;Jp1$(WKTP@ zixK4w9y&au{ywzspLlWjuVh(9aS8D4*md^#xr_NbY46ZA^ObieO=w}W&JNhUw%3P5 zPZ8+1Q3t4^L9cV|ywvI-scd>l)zI(h2% z-Q)Dk+!kGfs(ORvFf$17boCS?01KbC#-Mp?c-JHc@9&*@--%aL8$V~yH?M2p1Z{`W zd#a<1f>*QC#R+k0ygL$a0=7*nvHuMJkG?vX>}kqLQE!j7clRkv>22?I+$er_a7y`} zdBg!Cbo=hD?+(GyHA$109v(|~TDBvNjc$FU4KJ>X#J=#*3-EVWvuHBmghLds#5+)g z`i#~8k!#hW|CKTk6L=uLqQApt>AkVBM8*uUoiG@^DAgOiN6o)IBkmL)wi?M&!O zj&kVO%LH_q>2T_BG(^RsT2a$Q)!2F>yggE&c>d`3BEBWu@u%`@K|#Sq1O?EZ5B~*< zizWSPfINdY2I4*a>$=mrwLO)& zwKL^}-nZg^pT zCbaa98D*8%=R^{uz=4!M0OLf($5I5vI){Jm#7V$3$uyNv5Eo$gf~L*Vc;G5R9%}To zcz8fEtINm>ncU}LcgNw=!lxIv(JO>;&ee=N4wypEva7(Pd3N8YLa#%NScwq=tnukG z|1OZ@As?Mt@ho}!Glh|VE}R<`MuL?o4U)15tLVVh@64~s_<@)URNTsqVeEX3nDJj# z*~Q@X_mJT8(AxM^IJ820N;S+VA|w~V5U%At>_Iw$VdCuE-E~RKY~zIc02R&)65{1n z{8P_Z@n1GyKDb;19~>9rRCl05yh4n$vMp2Ry$OWdgl~Av$2X@SN^{?CWww^4AZp-IG= z8Pu&2835HC7w|HH;c(y6TElHz+YIxE($#{Bi6odix(xJaHN{Q-Kn;ZZwbU;; zPCp=Tjd?jil3(7x85`iSw<>f%sC#`)LtE0dE{nfd{|Wh$kX1p;L?xIK6kNiI@rsl; z$&FsOh&RC(U?R(L$~9wFf4@4M&L9>RQXFxGP|?$gMp$Ju7MW7?OSvQ+t7+<)$mrCEVCOWt$hx2{DB$R|)xu4-dxZp=kG=YYb# zZ&2Zgfs=0k5X~N!smbejnqi%wmI!0dX`iw5Dx))fkCDl+fiZtF*7H(PBAEx$!2Mz_ z>Nit$m+;}!al9{B_nH({58Ey(?BCU3w?>^GTPYI_Z&vEY%dV7qaZ#r)a#sl29!$uz zZwnRNdyOPR-B*hTc&$oCckbymc@jgVX#3Hib>c1NFTWv&9#9XnadO1vA#k4jLKQdj z1T%WR6&nqf-kdfj+nk4>)i;Hfx_g1VpQc+JMx2A~SIbc;vtr*#j&8zv; zT4gsEF^r-_pzHfro%T!v+;gif05l_Ry@T`MzHxwKR+}@1;moia*l{;cMoG= zY!YA~sCRyf)yQnrhtC!(0#j0s%;qV+myMFwV4U+Ro1aNVRln8?I2w?jRwMIPk8u2U2nPoFT9(@jhr~@N7>U6rp^#U%){*<^|J2vzb#H1>G4c^53LR+4`)C58WJZzau!hq$ zhpqSe)NTG-@&SnuD!ZdbN+StUOtZ9Pyo?D_QpXf7CpvQvksFO0ESz61A}JyC0RA*r zu1<@KM@da5x;=f~@czgjt*Ck2cv>gt@R)taH`{}wln8L*lLL-7 zq$ZShg?4fG`A^^&YB>h4bT;}mSSxo)WHsR%`7IEI+6Bm%p)en)`t&o?0h^WM6C&`R zIdE?^)k}?46A1=q$I{f5?W8xrPjsbrtw`g+W-m6Hes7WCKuUFB8OZWRe@R5c4Zi!Qh&t#R z-QWeh>pj>fswEq_+*q`6Y_UIFz&4W%fBRnVG4%S1idCU!`x4KIzHq_xfP{~5{y}7u z--W8}={{&o*;Ve--IdeFH!f8Fw>BKhUFuT*0xBjRaa4CyQMbdu;LOrq6(}s6od>_5 z$(@bH(l-&;x;upWGOHGH)l6bu**Wl(-QQdH-hdtsytn$Snl*>z-ngDQ&Zbi30*&GK zV@CSutM?a7w|bi`_vtTit#RLGpz#N>E=A-WG9gm$`%Sat3fXdd|?cG#9d@^}XGnHPqfn?kpWe7&;p z05>&qUH2s+4P=O#Z<166OdMjtWS^!{+a&-&m6S65MV+|Y+1VYcDV;=iN6rMUcaM0U z1IK5%)z$$RQghkRKIC7|Fp+r#3Cmc+bL{)|8!!17JtNW9duN3xdy_;dYwdF?zHw`_bv3tG(cq!j z^##>nhw$DB321RQzL!>|dzhFv+_C)cMmI#R(Jkv^&6_<5|8Q39=(m2$N}~5teI!cb zU#Ygezx9zM>Ox_Uv3*XRk~;Cq4S za}0AAs{|<0F#*J$H%hL= zQYzQ3&H+Fmu8z;F>-AMYR7%?YjY6QJ>|C4d*o+c3mn#h)kNV)8$*FwLPy3FpWHH3K zLxOE)Qa+5e(3sDSx#Z}?_(3=3*~-~l=U(aZ3Z#6rw2BE?u|}?I(!M^Mj=R{BNSFP4 zoqM@CU3dby98=7O2mE$qUa^AE;D%d!T961ep6|Yd-kg;)BSMN(T!-&=pHDW`5>0Y7 zmbfR-#8>8;Q+}rr9#NK&q1Dc|C+Otx=zaejrxfgsUHnXiyOjkEr#hDWJ8-nZPfEFj zuzMjk5X}Df!!!Bd3@ZMGs9S_CJmNLjCqF-daCk(af=))VIsQ{!mB7dh9~G1R67Nr_ z4A^20)Qk!-*H%*Wl~9bX!BH{Ft<&S!M*{R=tyXq=Nv{1#jTi?nFD}tZA`TglSXSN$ zfi>)$5Up3j3U=y@1{h~Q;sgRL4>03HS6Sp=G6|vz$a-6j4-Z2VfLWo2=K4i0Uv6(p z0-5!A_aU_Xl#E;&v^WN~c+f2S1s0P4kw2x>>4_e*EL&&Jb?H}bW;Mo@_Y-5O=-I9j ztxWyr^F*@!xVCdnC?VcTz`b>lVMe(4pV z2{Wt150B50h3Ag2809=;2T@k!U)IT#ZTpBcQSKo=!K(07V@xtb!S2DEIB+_JL+X~( zpI%lw{9wGp?}G>NNj7k!FXDl_8AEWykdf?V4pn*D%3wFFtHWR6 zM?TMb0UuUjb4i#dDprY_;&!C)xr5EwEOE&?5*!Cq45H3dFDIcfnH#CFt4aMY+UqVx zN8k7Y9zl#B>2cqA2grYx@xa~*97%AoE@2Dyxk@C*aoRg%V6|Ze<=-E_mTi7NN~$HO zYe_s=kBPotml&TAEYp|BcfN|e`fPFkdI8`>U`Q#WG@r3mZgQfyd*7!{q?$@OKa6#4 z=OG`wIY1&}2Ak65$;&Agc7>oAnxGZ!z##KHw4_p!ADT~z+s9AS4=z6Oxy?e@D(O=Z zd59*4T=@09nwcLzeGZkwI8X8(J)e0m0}*|xB1~^wbw4X_yi9oQ-DM6Ml>qK}!~fV8 zM=UFg%JKf|LXJ?l*A@K7T~p(o4E<{JUY=la2^w%V`1XNV5d+bkOXD{C2KrY*|F?5r zEE9b2_4WaxFieI{g121cIL!V|T1INcyzcpG_L4R%il|PZt&Qu=nNC8!)_XPq&TNIn zc99y(IhfJ+jW8a_$=(%0620{y&WT(OtSP_#LMrFl-1vZ-TNekrqh;9SL~2EAk%lL} z(|V&BzL8<{JONriJ;Tz+NtqcW8>x&YiLmZkXNe~FQcvKBi@?#4O|?$1oFcapg&N+u9s}l-N2n_ zVF41Hn{f90mzFEaEut$jy_pm#w%HSzZ#r!cj7Jxxl&{^gQF)pvcSEvgrq7%&{-jWe z8R*F(APd@YLxI6gsVvoS)#D20at$U-wg<~o3<0=2r*hqqXn>DMIP_pCXFr|nb(DW% z*K@s#Jm0FY$R>%g##_ceAy6^HOVfttJIcDwbCI4|!=f&KrP$Qf?vQ>P9<+3F4j;9q z{Y)`q<3zFIMYapeOA?q`EFa&+t9W@hD^JGRR)>et5g5^o;;Vz*i+Ds1>Q|Fq zDTQ>hMwCTGH;4wa7-6MxJr%5~$>X(EVP=QV=0x6sM<)=f66XB3hm@ahe@m!?ufZJE zq)(*Nn)yZc3;I>`L-7KQy^#AR2j7GG^!>Xh-o4dbT}pcu3FSL?Hx4+?JZ{O0{kk5= zhii5oj6dG^nS$t{Z`6@ z6(YN4yw$i0@qRtt|INO$YgoU7)6a#lTdW-*4*+1{stf`z;MsUu$c=d2MN*1UF4V~C ze-@)bN8ynB_B)+K!}WBa`4dBmwcMPA?Nm0wC}P1Qw6J~$_+mjLzE6*2`972F3Cfeb zSGFoN2)WXsEVj6cxG6Hz%*CZeVUTY%Zp6mZGN`xwTt1p4ikx%A@qa^~r+*OHu(q>1 zJD73@J9&E>V1k)ffI(4?#G@2#**&Y1@+@M={O@I==c2LUo{Rl>Oz@wFOm%@TON7W8^NZ?KUe7@v?3>BERh??7~Uc_Us8 z#D$ov*vi(2d#^)kI+vf@PE;wj&LAfGT0Ti?^5j8^nN)P$khU0GgSTu>8tC&xQgaG2wv^4c+A4g5?h=`1X){)l?F&G$Re(@Nb=M+y!j5>l~Tw9?%YpG^09ik$|M*ykFph@@Oc z_s9?Tk$GmhfrFNy_4?7kM92IS3=-0Wm|rt&gUN&SF|K1{k~Qa5{HDoiWCeH#d02Ih zcH%bGVk(ZeSZ)uelqDlY^74nI{{Ex&wKCBO5?@qaIa(H$8-cCHwa98hyUFYEKU5v} zavz%D@PJHnA$FmM>uuSC&+Fhp`$ViiHj3DJW`F4I;1L}* zkH2hI2)b{b3cBZ&_n_FjtMfB$&+#K}cI=TGysFQAa%)+6l81)L2y@ED>+5eeer^#y zSpz-zI5*?zIR&ATSPbMB%ibC9r{>hDOLZ|<=6;xN$bU(499rFuE$25ZrgI55 z{cw~9o9W+yfzha{ODo(s6$$86%nU6%LK5CAOJhc!Q|8F-B5mhqrB-x3imEa0$0#^) zgUeGJj6IygZ5IawmH!sCNqv`4HOI=*4$q?g0jIPN3g}#c`(6E_${(2qzt0sxe^~dH^gA%@WiP%U;Xq=Rf0bz61<8g=&ru=wMrK%N z&6k&l-Q*DUals zcL&cU;d(POE}`a^g_DCWB{Q4i+W$JDteoUlsUY8$Tm49ojIE@m75MvfrP_W3=A+Ge z4rKDJyF{k@IrE_5i6^c#t+0rNW3_$uRBoTDJB`lZMDgeZ?xmhxcp05gKu~MBhy~Qs z%|X84_4Y39?8d+kZJF5(FH{mJeTrxu$QQ*rN6bi0Tb8>W^I|_73}J|Bau&S8Gxp>2 zslwy@zMwdaFLh3jD$D4_P_gENv645{hVoi0kx^w8ED38~dGD8$c@s5*y(^u2=QO-> z1DH5;+OdJVBMRl?BEDOe!up7IbbRN$ zdG31rbKip_hQ2ISRlbU!C@*Bsyo?^DHikMVFHp8P!-~3}MyFQRz{RMbx|SsyuT8TL zsclc^IV-@2LY;}YpMTVV>HdbGZZjOw$3v0IB3%Y;KKRuL=};k5u7i6>c_VC9$(>@f ztuCEq0rcoV-LwblhquZD4il=DOa~65r<;YP66~cVDxG$ikTSJAlV=oP)Wx>;`f5|> z%aW`{&$2x;byVH%>c_|4enUayR!*H+=5)l)oyVirjCw{AO>K2j+g>Y^PalsjHzQ9u zno2jk)G2Iq=X|@7R@1KCGCc{sPdt$PF3_*T$A$jbu~MJH_(Uh^eoD5Xy8H#wh3RCK z?(2a{DZs4NtA%w>`gVq@Q-5*IBn|)J7V_8Pw-4WEl?@Sdtxe8q<&CaCiD)FC-PZZP zMcMZ-^7ZH_qmHgWJ;>~YrOFrLzR1HjdHr<9E4%3aX#9RgnQL@3IP`WzYfE1$DZS9W znSjaIKqCsm!|N30VxHR{FZ9~N>*Dclkb3Q{dT zaoo1=Y*4~JlF{-b+ffug4bF)CAFkdqtg0^T7FLlEknRo)9An#NH zooHiuuDO_^bBo+k>C9#^UybI@Q!MGEiG+HR0NqZP`UO!Wt_$j>O1Bnu`4xmN?*-9v`#lqHf^ty-dt=U?_RBN8P4qX9K&9_ z(yo3zy4&%-c}rB_6?AIS#aarEDym?!rjaj-cIN@}^T_Vi2OV|%EY2W*|vrgKV z=!LR;ZvFe@543-JiP`J_$UCy`>Ung5nxf(tapT`hW$VA8a^aRb%s)5G*_C=+Hq^>Z znlxv>tOGryHB*+SB_lNMxNoHp#5hXiEWJ8+d3Nwpc?G}V;RNIr!?$X$0O6eu8BAhmbumMBq=t>_!}S zoqN@Z{@6;(tb>q&;mze(KgO~ik}A|F)7*TxDCW;{_T%3@Tm$0WO$dF3NJfl#J(qWa zhTbQI1Z>#gzF(4ME1VoxIc!B{xH#~JH7r(6wrzwkoi<2qHNASqN)Jkph^hw}Ho00_H#H#^*k+hNVnA`Emo~vChOIvD(Uc{fSND{2|T8<&F zQ<)2i1yk-Yd9wB;mA-nb6udXhVZHK{?r>@^sMdKslRWR{nX-T|r1m83Th_Kc$8!7U zabkCeoxmRV1D}J7BV-sC&8H(utw-Ma`Rhki z4Q5|gd1|Lw4`a?J#m<3yu|xWws~Y&sD!#8gtr~da3%7*zJ|~<_rl?3p6A&^UC8akl z_G;P~9>_~fEV!)YezFyRNHgC!M9qaMF1Wfh(lM~e!im!a=`wZBKG;+g_>t){>y+7i zTZP?Y?ZkgaFaK&PTm}!emmf=-uH~6*?tL$1BFl*zP7t%1?1?p6i7aF?=rcI;6omBL z#!jWNJnk-`x_QmA^3un&qfxxxP1Mh>&Lob$q)ny7rTX2q*E@_))*h77k5Q`YU{-ki z*k1_mdrDgJ3C6km=Dk=c&b#elT#;=uk8y;U;(FwpIpO7eir+U4ACwI;E$%5gn1*X~ zE#=V;hl6oda(A#-*8SoJv=uY`b(~a5?2bo^87|U8MFUZcfQNxKp{vx&&q3{kOb#wcF(?~Mdzf5 zL%S(=D7p9c^?pPJ(Oc*wZ*R@_@hR=_Y#ek{*L$^Wci=Oy{0<7bvV2lvAD@KRU>87-d@Ox*}v>{nfa>5YVMwP9r?I4NE9Pnf1Ghd~OXNpy&=FFLFr$#(CFJjKB>0p& z^uMdyS3M=-f6X>J9t|5;>$%)_w|tr|^!LwPfhBq`Da6;}VXuCG2 z7`oWE#Ibi;+h?Wscr_gda(0-1`LQejxe8r6LFY8XS)OQ%#q3VG-igAQ{`Q^1fWf$7 zfs=3h;bp`Da^iq#=07lz2{`#|YPimN7JDvv?3$xvdiccv&733L^I)gtz^(T{qKH#)vo zk`|1*AGBJ}*+PGqm4%ic9dF0NJDmEs#Xd_vh}g4X?s6mwS`{D*0tv0rOzMk{>NDlt z8>rL`1!2I-?`%h^FxOymP82kNd{C$h8!NxK9*oeL8bbpwv>k=}c}c|w)-Y%vApSuvuU91`o5lIO=6RpfD< zWqFKw#(2tmc@q`^B)m~gJlk$^n@FBOVLjv;uYHSxyFF=3m1R%+-%qP|+>3b}@VPP~ z38>Y^#VOvO9n7RjVIwy?_!DGMjh6;EM{sLesB!P1-j^Km(0|5wEGpggox?WuZBha5OMrvlCDvaW#dbcRI;qVLfrbpLSjVjGm?m-xvrvvCNzc3m&h-#7@ z?hzQY^lS9D-ouDF+OUK_<;^_OO-{&$j*ekk7ZUku4jr!#7cvReJNZeUTY*s4I=<%a z>>rt@e=;Q>G&UA!38%7WhaQ;QIHE|X3YF}mt+CzX*!BjzVLT2>aepGwD1$YU#p{d2 zxIF6qjYDnaZy@Op|;YUB)XEN!bV-+;WA?wa&y@@(EYjO zp@3Y|+h+x0mxK3RZ7^4rEIL%-+mbeG29I&vo;VDh&7!NAZ#P1Cc{{zsR-tzG3Pnw) z+_rx8YHAMTJgD6|?25v@M37;dd+U^hi0FC2)6s+Jb2;g5k%7=+TN0)P1h4wpxRle( zD;P5pK`x{BJ1ZsmOJ=JVeB@3_@hRR#$u(S99$R#W6nsOV<)==j?AsZ2fpV;|b3+hr zm&cw?InKl7WzAFenmWd!;*5ZA^R_(-27$Iw5$SOy-_MlNdw?8F&Tv`WO%LtNBN_eS zqe5O%M+5J`f#bGWW1pr0 zW~3`oJ58dMW$DIq>YW^9{l%AP&Ck!oHG_{3b#eJnWPQ8!v8x5G~zak{D9=Bdbb3#8I;oud50&#)p{h5ABGW+}|kfpv?KA#zL zXs+^d9KMLzGfr2c5#Qg64c1Hf=M$tsp7W1`LVn3T9sF&0HA#gDNA%T-Y>&%P)5i31dZ~7LVrA=z*bXR_oCy+?HgqO|jsi1RQ8&AH)bqulZu3qq6^v7iv z)o-HUaP#(wGWdN=d>&qhIqYTpHaL9o5n3`&0da+yO#{BLQOi%J-Cvz1`0a)4h|4h= zapld(RqG_gh6|;h?69hOY+U6%)D?9xcV1(?lNG~`{Jh4I8aX1&F^J3KD{|cM+($t!}4kSrDO#(Xn>Caep(G9K0E9p7$O%gfE0 zg@rC@{at1+0+j$kBIa5crbTwXJTrf%%SkS}f2C->vo!heJhG0m+xeGM;?|eD$v~1( zgs)m_K`FS`J7okOWdcq!MnsO#S`~Q>I;nL57O$OehImlbZbT_ZG_gepqyp>s(Cn^7 z2VQ$N^x^LxV+d52HroiM-+DQOQQRCj*yg-a=xE*A-}&^O zsce+vGiT`oj$r|Ak!RU%csg)_HNLpNA4m~BtMN==WYJ9)6I#6!Lu#|3`#KO8EiZZW zgiKf$A={*hLn4oXPLK5tb<;7V_F+#CTWhn(4e~6GlC2C>q611?D#V=XdR)o7gKhP8 z;_lH6Y~5*h)=d3}FO4TVv;fo(ks(dKer*M>jS&B{m|~ME$^1*m(ql) zLdzlf9EKyG1SWDCDjwwfs zQDba7Ew2t5Y%=DUQ{y!=BIpV=kuiDCW&|i~+)>w1BCEOZ$9x1=& zr(wH<5)nQl%%b99gjzeCaz_Q9o`$XD06tK&6Z%}mlw_r6oL z+4MFJP4YZ&aJjmI%-=lrxZC91)l5PnOhiMH-}MirBoE(jr>Us?r!x2Y6?)~v7sqH{ zFPDSd?lPhf>t-&{x|=^2`DnxH1WMW}jqpkA5wOOdcw5PKLW;eExz0QS4oA>kkRPu! z1SPk`*=JMJ<~>*xsmBRzn(-<7hMM%;yA^@fX3kZZPEH5@b+&C2h}`~vnvZ?HE-*z^ z)xd5_Z!)0tH+hXoAw&dZ_jsyRM}I=9Ov;)rv~tGi10Kj;)h+E~JFmuJjN`{~;ou$) z4le2s$EN7r1s&*}d0TZxnrj`wn$j*Dte!74!T@u;F@7hL{8ZkZG2I>3zCPHLosCd( zdYuHK=`nA@2bJ5p@w;{BNH^w$$@IxFt*K?cNv{=dy|+Yw9d2aV?| zqw=Di;2%MtJ#Xp%-wl@nr6?Dfk>nwkj$d$Ec}yipSe#-dIW@hp{nG7^WoP6@yI_%| zs@*poD|D}~?e7IYK1pW!S|Lp6?bW2xjTp5r+0J35$wp55S1r^s-YH&FzLe8$?M(yd zbaSgME5k_=aBO3#p7Xulua_^je+HyBL8lXyRo&Ke`=>VhwHs}-vmZmBjm$wNq}{P3 zgAbYl=3;fa@2HkCf?h8*D&M1Fs$tsakG#45ezZ7Spm*r#9F-0QZ3p1Yd2L!rBqJ{`EhU9WLrdGwjij|O*yaYa0eYZI zS+P285#)juohAc6`7c0+;Mbo&3#!5Etbwk^37HYIif5H7s6Uun_e0tI<-LekXtivP z$PZ}71jpty-w2~tXo9GflJu4795sgsx zmO||hh$D_h-CEl}KFa(%KO7jcJk$BbO(Ds`?%OWZn3$Mz9w+@Z$G@2>VL9Azz0R3T z<4S-3$Dn_e1_a1T&u4a)lob5?Wr}Y-7kf8JIa=m4!?mAuK9)syM0qsfmA%2=veFlG zU1H--A~a`FBk~OIW9;rEgf$|5=IMqgJC33@Lb&HYsnBTE(6kcviDEk`8JUpUuFroI z6&0aMa&7IP9TQU&2xdMwIywT$Gp3uHo1wywyp4@?-OrMflV_?or)F@_2W3D!qFVNjkGP@hAX?(8Mz`1n*1?THt{Q3qQhOvD*f&7SK9 zO6~c}Ug=$|sprxE)dTHM9dCTjBG?{;*$sLJ`?z11aK&~5Mbe@?%C4u5n?vGTzIGnh zd)0h9%#wH&&^XJO-y=%TZm%f}8!S6eVsxAcbmEjytq&f+TQVApcvKq{x><*9=7rTf zHTQFcYz6i&jfGKVm~71S9=uRB+V%u(uK2eRp~ENeJDBPvc}^Tw9Kqnr%Sh2XqK{P; ztxtL9QVT^KJh8KTMgxpWJudCYX)D~%t!{457vqFn%LoKP`(tEnw@>7Ba&FJS>$ZFC zfH_^ShjVjSw<+{TCvUy#eoO@?$rs?hwA2qy8#vo@3yD%&{=QeFUMGk!zh^}htzLAp zbsZVn`h~HyOXS~cz=cWw6B=!yjTG{wmJHO}o2hiel45$`20B1hT8A@hYKSm}PqD6G zJ3-}xA|F2n#l0oK!!z8sU4m2*%HZOZY@~-Of~o}zFR#s;eZfQF>vuP$8<>rrGLwvM znu#IDcDu6;ii6Iqi>dzKnO3eBKj)ZOimL~p4KSXLA*$A6V6@EgA`xV}x3z{XW3OV>& z4PVFojtq-`!xc<&H&)k-`juDcwVb^CzMb7I{uaJFFA;0m`#VGd*@me=VIZjjv;YhmCe1*PI$&iTjY{*jy4>J1_-LgH9I#Yi&pOBxoJ|W z97z-%mDJa)G(AnSb@D-$c%gNgJ7O}&Nu<-X??wq9J(gy~moz^ykao56x=|bqns=bJ zXlimx2JMjkpf%&w)hh#2f5o6e9wV7obO%$}8-mxykiuDouz@G9yHVAIzt9AH^?uWj z-SBV25z_=>JmLpMp}%iDYMi&w7d0gx_Jt&7<$QXH0&9N?fiFQeNRA%z9G1GCFBv_3*#4ItUiyL zwk?M_*iH66jX>)-Ce1slhj|2;%j5lUd|ew#ZlX0&;sQT_AE&Luk>zH!r&u?VRGMbqiq$H+M&nz_Mbx9?JQA_JKQ5y#q=pXo`Ex5t40oLOUI2&5?I2Yc{ z%5u}9E*AIet3*0E8<2E0;&($je|4U^51pw~Ka$X)Nb4yM6DyLX8%UUtF2;rbax0vJ z;GF%IG3@`Dx8cWHR}w+m_nXGN@5L)-j%c4%*h{(zQtgZzSOa_|q6~lzweJk-4Udi7 z&)c8}ku9U|xh6?639}%bcld61(NoH&6(V+{%26>tGdK3E=A%|L1E1u)d8i6sE`Q-P zj;es?(pX-}<;A8e8i7$^g_;47pmaDsFf1wzjgf+`kRTEg$}^)mwKCpjoC?5y4+{gX zXep*)$9YW1Jap1Z_ z#LyFlub}ZHMKNP(y3yFi0d$W}WES~^UQNb@hWT-K73$<|%+IGEWyBI@hg(dA{QHhT z%wSAr*T|^n4oWIa2}F+Fw!Z)nht2&6TjVKbLL5mmGF`x>*#Lg)!zW)h$pV^&A9lU| z_ogH?ftqg@8eDVV$Of6a5Z%ZN^lE(!hp?1&soy0jX1C7^Ke~5O(VOWRaMOSf&EGHd zOPcC8SO7L3xx<31jji+K34txYEs!lmv|_d5tNqgpJ|5m$G?mpJi7@3(qHptj;A&Xw z@c;gGg)(zO)L1q(wc)@eG*|m8*$ARM^*9XR-uW6O_jYIKa!@F2h9CqY*&SzYF40YmQ72Nmzv z;wIiF6n93_PBR9)nM)4;eeBYxwYC?;B@MGGNpJk0<*0y`n%m2RkmHvl=NDQCHDYFC zl3WNWbX(}0MF0ixaWJTC+~sEDoB8sO#q^+QW32SI9-I8wpL|cGm7n_L|E=pFBy35R zRF`%nJnIE?LQKrZ$@W#XRS9yT(DuSy97LVe7YB~YpDnQPSMaI~O=H=x4X zSlTB5&iLtZ+>9L_lM<6 zL_}9V?_$_^zFJ~pz?IEY6tSN5*w%U=<}Syb@5RO0=j71W{e{)Byr&MG#P?5%H#zxr zO3@F#A78OQkiR_Bs+oCa16qH;?&Yr1Zx39qoYav73JKZrDG>nnQ_@ODbbO_j1lc9Z zvEI)(B+ntbk5(oMJ{)&Nu#~zIk#o<0fLdZcsZ8i~E7zg!CWS)8w;{cso+}LYf7u;M z5Y9&p4dSM&ZKj{D(D<62qI!0N4iuIz1H$K8LWeB~HEZ8w`?LcPO0TLZVGh=Y*h1~T zL)<0#Iz@74cM@yHKibEk0X?}w%m?mI1p+RBR0wcFPpaSF-NN-3m^CX<)O1|w zD7-H8`;r6+7rpNrYx9SQ0&4l)n5oP1%voPcg@a&BYGX60kwCg|)zaND6GxMCrVhd1 zBWMWX4lo&z)A&ND)5T#sW{A>B8Ss_lV@ApDKxg`0xy;x5?Tw6Sh50npm$|Gf5*bEB#l07uJ?;g^u>Vl&L{EDOus8a|7QYhAUh*jXwToprIIRdogXYl66$&k&oRFY z&FP-(lq2q%DQa9;+)@e)|$$dzF1Q(a^wYex22B>&*vY0Xu zlQIGb^5c{8Scv7kPwMNJklYj}ASCf)#t^l9p}MP^wqc$*R-E!=_k!F8CN>?-8&%O> z(q>r|OWpG4siYv{)_*fgS!G__NB|^cvMC&nz@cDm(?^~5e1a+zl*F_M@{r7ySM9)P zY32SXhzqdqP(bedufN9O`-B|JOnGWW>6b&a25o9;C!S3Ig(1sW5D0W?ce?xtu=H_L ziFTW>ot54X4)6Azh0molfCMlj(Rub(NACC!9%gUDsaYA$;5$z#{qf*c~rRM1` z^lc@7^r~xVyuMQh7DQjI(z*OSIc6SLjg5V{xVX?8&6c|W{umD9SEBbfqjSt-(q6_$ zTJ!T94#&hcj`sNxgk>EHR-7fy@f?fi@76L2sCgUJwJoU@3UPo=0TwY8BMklph6qXm zK(gAne9LLhwtE>6*e+m=U}i7kSMP3HZkPGB7f}E!O*WTV*+W+Azc};v;@|dX!g&3VXU0 z`XKUwaHL$9SlA}7iu_puYuVG7EH^jb|0XF>JTz@ML)wq=ER>L~3z#uPSk2dMg}3=! zn%s-JX|B`!nJpe0RDv`u#xpP#imjYkG}iR$pnA*mqVLeCY{Sp1w8GR*EJOt@EwZj~ z(x64J>!-zyX9)`oWN`wkXqG(MhUC(Ze8{6}SpK}pJLgf_j@-jWa(~)&$+4`QOffvo3)*L%?Tkh+~{z`4bc(;U2%giy^hQEd? z6)a|P3=$F&OKk!lCBBywNFcN!wRF!T_sBk6iQRPEbpDb2A!{u5jm+O`xieGejIXQ`aKyJ?9N6Z+N@4H@qPe8ksDHoE^AaOt5-YMa*Qvd;U#$ev z(%}=MzkCR;Nm)7ny!c?4Jr=6+H|#b;Q3Yg*PWR_Y+XJ8G5Bqlo{|T4WR3s6ypdcdS zAnK7<*){=+Q3vsEO1xP*zs5kDI#`zNPWQLxt1F>wT^c4yDjD2_1n%MC;Q_I+vG2!q zOWR5kUUo1QH>69&RaaLRee>=ncTTOTnVLfty4uVT@aipvZH;!%xq*a~gd|O1%jo#1 zP1H1#4-wrnzxvb$Rhh=(Af`as|3=MkF`6SU?ODj*_mdyjdBHX3t{z;bki(@`zV&Ej z2V+f7kp17NIcT~*y@OL~@B3?A&oE>PmzKHqRPXz{H}BsUnZc!wfTq%R^im0u=1Xlu zMt3_!MdQq|oJ2+#g%wUZ9UK@~KFv8DU~4?Zfvfmvxvs~+j~L^=xU`)lPjJCPDY!L& zgGq$&nYv|EIyc;1rDiidbX)!V9H8U)_$uRkN=_eBM|3o}r&fU6p_rFuWoTa?4O1Vt z7Gzomg_sexLIYI6Bj$TG$NKyXgrfCC+;X$DY z?LH*HMFxwQ_)`(b((&Exb$?dE5KV_-u?Eq^{5&~`jCMbv|EbU}*Hr`DTCPw)gtTkc zx>=iB!bwLbxmH?R-H?Z!z2p}M!IoZ$kZWx+MkpDQbE12areS6eo}?73(`m@5tCdf#Q3N8g^}cK_jbxw+g(aoNtxIh^=ty+uSDQmX5Hi!U#9$?(gnYaX*I}Y!>P1>FJB@{Ng_0?KZu`pFcsF%Ga-YR<^8n6nC4rBgl~St_O%) zEy>N{PqRbA)PJN&&;>@swiW!{9maXMTrZ9Li24$F2eky)!&cKive40WYC&3zx-Q4M zL0l^A4{JR-xLUZ8Fvz~Z10bQanohv7r@zq*2M^C*r_oJ+V;~8dQ0L<2MngrF`dX0a z=lqFD+MCImcd8_p?H>?`mvEK15quECeD6|7!>`=%X@ zt>LpYh3jLVAiX&*nI==Xkv?_@!`964D;@yQDNFC-HqJH4wX zaD0^8e0OEOw~vD%rvJReJ&i+`Nhi0VLJ5s}WpAVCpP5#L-1e8WFJDlSrs>EY{E;Hy z$3oe&DtcMM+51~RD5dzHtCs1XF>FcgItTuxH-oAXAt7fg*fiI9LxMPiG zt;LMsUAQPa`b7oVzwm~b0bmEs`$&Z`dPty%^vdlc5n*T#jcy_P#7K^u0@Yct5IH^vPONr1OVpa=gm7@Fa! zg2#@1b8}LF-b6A%!j~M;C2~X7v5r!M#+01@Gpr}6#^Y`wv}x;jf_R@`v2lMbn@mT+ zgybZ+$;oq&8oktruH$M+!pOW~)*vG{pi6h4H_b^TtF!&O) zrXi$a%CE|!5zMr-#~Psd_p)Yz)B~3)2NupMxWAUX$mKIQv#>88(o)Z?%xuY3KiJ+w zB_t&LNB1OVJF4tA-p{$LhQry-*SNS`WS!mNjD0a*iBSfoqR)*Z3?tjmSvlbd*YsqH zfK4aN*Z0xxLL(`mmd!)BJQzBK&=i2EgiJ1Vcup4=Z=M5f7!Xf%wCE$t$Y%!G54NU@ zbP6B0=`tgzaa!#g?SIeGo#M+@5EFYKBO?R2hUikfNpdTY3+(TguxI>50zo;9eMtCs zK>ZL|8JRW!%&b^Z3$yC*g46@yD>F@4&)zX+%-UpS^Z9p|JP2` zRMs&mO}fwjV4eUEMp@6&77!xri|0aib#;Y4Pow*w5$K@J`QP&R*j;5hOk*)sA|A=r zv9twNGMx0m-+8R(`S%BA|E6Hmjtm%JO^k1!@685*p-EU{X}(tBiN{DK;QeppW@FDG z?69#sWjMl(H}{7sSrQxcef{sf3zc&e@`nU|DeYJ~UHs1q5(2>|6!UN6WImezefkNw z_JT~vZ+%{~ije=-*%Nq!y%GKkAM%)gUUO;Ma7_($1XfO+FK_z(!6r|5F57^sFgZis zy}rJ!Uxl^!-v`O8cA^0mR*WM+)`|*c4@4v0XBt0`|8IF>!h?e<245iPZ#DlRDm3_m zXv|xr2&u%F|JvRSvBLF1$mR{dz5Sap*h2u%==lBo`+rMW6#fjb9@FZ6DTRO{44}D( z%<2@t@7)oVm>Am9BDS`+Mny%{qUC<*azD_lprjsge?_=j_Ey-YyBCO>`VM*J<>rcM z#8aPwPyf@WzFtsETf0B24xTs$LBZUD+>Tk^+q{iM6oxi!lhe-e)jB{@Od}6-^Z}jj z85rQ11Y_fiOiiVFB_-9`+1aV34EGSRV_-AZw_+4&Fu``=&_!o%E8lq9hDa+&Op2IB z{te`s0u)}odhFrhv3=VLNQxR2-f4&j!IP1d~?-`bsV*ff$-ql2bJF$SlKyy zW%>q{%XKcR61gOB^lQJYgZ%7SzGiYrtVW5oY$2~Y6V--26PsGd47lPrj;LK~=(W~9hW7HOLoVb8n(nABxGMy-!doR1G4 zdhImUE|VanrI?oot~6u>cd%mbPiYWtq|frnLc}|h#k}Va1RS?B%h!?6iS&x{K%C_+ z@4hK54KePiJ;1VX@l&9RD&gY77a@A*as%{GRSfb=&wx*$O589G&4M10Qq2>QMrJ+fEBiXydA6 zAT8}|Gb1v=od;7V2vE%lEi-amkD~yLvBtmZpuT$W@qU=3L(sBhJ#?~mVP^ypAJ*L- zMyDxgeW3ZBM_sKyXle)NMg$-q24I5t4jN7z0>kL4 z3Y;djoznpN<(^P1Q3nE;tJeq3ys_wa*NgY*KqVEbw8}mk*!}L^yAI&f7Ye@$z}R@J zRV9bdz^zQqt`{?Dv_Ds?UZUqy3vJ$9a0rlJ;P zN)J8Oyv3+=8q3z@tL&EZHps{ehdz$(rC<3)Z6NUG;8sRUsO=C{b=UwTW1$IGM=;p< zf!}0d9vDRStB)HUuS?&w@|sUDdnR36!b4N_PJ3;D#su{AyzvZw9YK~S6-Qe+Zz!}6 zrY!O=^vi_beS<4v9i5%}z2E4rggIgoC8~IC&>1vcy&8T1Ro(L1&XO=~=Xu8~Eyfq0 zRBpgJIOWl7bV=b?nq7Gvx(+`nQ|@n&IXE~r=j*F=?tAG|^g6y^UtJw)YI_`Y3e3+% z0j0=K9P&QBw}6A@6wd?bn(G-Bpj7zuWLC?8(n~Lu>$qDpN!I} z@`aty1V_PyVu5TVA1W$pUnQ&^!wc}h3uvNsA9rDp#2Wm{0@jA?rbAS>(+clK+^UWH zNi2s2SHrlrYc9-ArXZ2azG$pNMe=vIbg^2#eSPoj&2K#}mV@~ADn=_5>Ve*AWlT>L zc%zT?gPY%0ud`cLge3uxwgq_J;Ao}uN@u3QwcLxcviQJf={j@V3Z%zXBtlf{=#yO@ z0gMDL&YRQ#fq?Ibg`TV3k&}-s?`|&DE6rjEepM9MdIH_riqx{DBa!P}uExkAbiU1% zD4F-~F}%fLr+O#N$b_?tHCD-jmM`o6d0YPyw7 ziW$|sf!k{^5szXK5fQiKJwSE1&)NCzE>`!d=baK80a0E6O0o4guiS0{zA1=haS@^; zI&HvVI|Hm^S_L_OFsgfPSlyc(d_9q2dh6w<+A-)`5;#- zs{Q0eqt!$Kc;UJaA3jV80_#%P)u&INoInY{_5OT4j59bmxRfc)ly!U!M3bv$EqLm2 zIKlA7&MzmLiW=BxX=(QW#{}C5Ff;GdYekxmY#@Aldj|X~XF$aT%+4~|^Gw5ivB|R^ z{^vf`rvSMQR?DO-FB4`AqkFjy&b4REyu=8(MlDj}d<_g7Y1(=g4unZ|zZtK3!2~SK zWjuEXAChbDx=nbL@>|8^SIQYN(Ljd`3f_;qN)7P&kp_eG zjh3#{$2Y}(P;168Z;oVi=2w~bVrv?hKFPoj_74bHUi&Uz(y5~?x1pt`LAFJN6olvBY zhfPYE3V@qVOv;r`CY0&agQN3nO|L&K)^kJ)ubb%hk%YvrEfJqtF?hPt*t`6b^d(C< z9h@~*g0Fe-ofLL+8A!ldz+~ZM2-L(Olf~{*r|!Bgu`d|RGD)H()&1kEVdznehYz?j zeODNge*t>Dt19xUN~#s-$NG< z9l6e{GT`Crm(Aa1D0Lm1uOUAPAk|YB@!mH_-B$->tJT}9VzaKddzD(XFvPq`uloji z#xX|d=8OUKC;l62Z9jxY9$Dz2R~jih9i3FSQE+I%C7&F6BXEzM>F#eS1EH;to|pDy@{4)lOU;5MM^xR zA1>QNX@0d;m!~_E^4)e7&y<77;qVJQ=%^ca#Z6bgYP($7k%E3aMUqs0%z?2ZwlS&|H{ZRCLG*Q>;W=?ebWBW^ z6f+CwuDR)YKBi_moNVsE#*w)eh&px|zFd$X`}(X4iFQ#cemDnvIU*pmST1c!9cc+{ z-$Ky?=S5F7Oqk7XX?Ma$|7>6!+kxpLPYXg+le$f}6gvI_G`rBBLnWUtNbUo7tqR3G=_+8|spI_qivvwwR?mi`(87V71EJ4sS9z1C{81h;`+Gv##SKu4-Q4JlBuN!9w;*&)4#vbH#(!>?ov;Dur zc#IKsT$Tf7p8mF9bvK}@3NX@6&pPJ#678%A^KQ^xId0VGDnDK z&BWM^HipUCn%3 z^62d$a8urSpI=*oOD5o8ssb_N(?@>Ai{{fwLy|31h%kmPeUcFUOM@;S@cy@87?VuFzLzm)NVk@bb2% z4e^^h6gajSTwAC)dQ3~F3>-{$sV+}vE6vp`i2+5lY***f%t_#cDt$~VT_h&q;zMt1 z-EU^mv?NYX&n8(H`w(GMdnTqpP=Ug6aZs}wF}nRj@QvhVCek%R+gWQKbR z3O}tM*TuxCDL={uyu@Gf;k5vU-dkFP!tWA)Y(f$XYT3oUP9wZz8ex-7;pnerk$YRI z_nQGf6C#Mmhf(vULk`ZC$WZxwd9^Fa%zWArQCKp@4Zhr2Mzr-ti%8e>^?kl_0kEJ_ zHVpf)k1AC(cIC_TQ}kr>UJgupV0^q)I`GfSnD6CKO^ZIJj@PA;efSWLCQBgEVfSa& z?&w#jB&?C-f)iTGxL!8s7j_*a6F;fnvG7BeJIc`46PMML;6?F(k72&I{my}yJiOOe zphv=le7&Nlq#8n*F-trMX)SmdVclv=NhH+yK}s{UW&uYES-|*X8;J=+{YVcp)_?St z@+VDTwp?QN=tKT&M5c4QK9uY#&4)u|szUty8iU`1DBMs0)Qe5#O!}&L_%6?Z>|jN# z+d@{|s*IVdU+F(~>t*U_=J^q-kOlR;X)#zc`LTrCY_RKG4%wrnr|roxz6Fjf@^8!l z5k*d|@AIGHxMFro=Mo^<;N!q))Dw7fE=C#(#;K6+fcX78pRRfIx zEeqDAp->#QXDF7G6F3l8q%Im%OEioYc(ycAc8D%f{doq*g~lBSK0uZJ*iP8`p9O&` z34>s2UTZa&91d`U4xN2l+ zJxd~eD~|pWe!lH(0u;sZv<(^x0hvEU+WE}ibE5!C8K&ZV}c$n;1(F#Ff46!xTnjfhy=}(bPD}CKDM*p+I-caz{L21 zefwMMXifA(17aNib_~InP8Jpw59Yv~Qm0Cdrh#Kp)2oj49fp_lmkp6I{HRR3JGj3q zma)+29~_iNL+^H6SXh8Yh0%-^6j_PqKV;$4TEx51ae1d1p#Ctq%|B)|K57+(sc5vV zonwxqeR5K*8B?b^TcY%nioUb|_=HAG(#ET_9(&-K;YYw%Aa>YEs2k@#A#ce`%+wo6 zDXN1eh|w?cP&E;h785qQ!A6Q(&92G12XHMVnz};YHZ#bLV$WF8pNS%6yS0Ah6mS^>|xF2k%Q|_#xEOiQAD>4wx zHnQUWb)@|6NN2-*5!9l(d3v5ZS>`7jMg8P9|IuXaiax=+dZe-7CpmSPY!u@aVjykW zxdO{t+E|2b4ojwpBI0kDtIcn?9#;OfF8GuPiqz?T80C)4*F>*8$Bj_uLzVv2y8Nk- z=%IhoB#2u6^3I*s;71FIglG5Qv9W|7?Z>Yi=roF0G@mj&3VceM9h)d=X7;O!Q}L^Y z-;`zv1fKM9OtZi+m?d(cQ$z=^$qP;BN}F1JP>)9tL0hGOQhLZEg{AGY>!A^9x7ro1 zsdd<>T*Hm1O2fDD)7~Kl$R@->07$EY6g}-k7=)Za`&~dgW93EpZxeu1eUh0s09&V5PjH`4snA=grN6eeMjxNv^t{mV z^zb+pnuVvH_w{W6ex0?h-vb2Z*%_;L-Bb+K$WF|gp+5vdI(mAk$!bYijUZBWR#wL8 z86m_vSWEq3L)|3BcC~97r2LfZ^H_6|oD#HyaqaW2gjigxgrtuFZ=Q|WEMqE7jqL^I zBB$oZn%o@#+F7g7^l6!0oY)8rfcT5KnnX|0V_%);G9A;FJc*tlO6L}TPBJJTjMw;X z030XvIBHO(TX-$a)kziKPC@x8zXrVK`hPf7KR5SZ7N%&iNwL35gL4*kC1XmOtf=xx`$Hu*rU$K-0T5Mijy-)E~cadwpER! zP5d{@CVjG`%*}!-^t=!$R?9B9Hj8A3Y?`RtP+S&eap!jo(-BDdnuU&dgSJl>a{8L3M|b%thL>PQf6uF$(a?3 zCe3EDqCtz6pWBWK;QF}T6z&C83rE~U^y6cbB}?Xvt(n?^1mG2v!+Oh>8dptyGX8{2 z)CMkbry>ULbD#~Jly8(i5tUt#l`-8hl6KkskFUWuG#{%pfL9=+ISk6L9wQ=dnRoB# zX$vM3E;8XFZu_N#0%e+%dvL}V%$vbFiz|UeXC;m-x@{OW4FW5wH1obl3&PvKtNEcI zd&2rO(fQ%s^?`Y{^#7)4s=F2cL(!(D^XU!jBeNypEDaU8?)kR;mc~X;*JPO<0r4V} zAjgCw8JBTb(Z&?ku1EdY!9uAZvQf(H3`6uv9cI9lyq4+&$Ps z-S>CrqPIFOx4kQ!p^l$fJBbQe(XAGmJnI)G@EFdBPzL!ZLxJx}ygica59Z+%df=>1 zDeQh$009no*eox@TPF>^RzfWB2U89T!9s>6K-={oA(Ybkf4Ur1skng;{OfI8j*3T| z$$dEQEA_DiZ5v(-c_}YlLliEiO;QS_uvp9^&XFe-H(5X^M@yM7L?MKE;bUKk#iovx z`M=wQVbRK?^bs7}29YN~L9tRXY5&dc(AJ~W&MW|k$4f^Ikl*rIVfslE&;7>mo1im} z9hRCL(UhrpG0Y-m{?@3hGjkYEAK9nsrEc2#7mHV86ylZlvh_06UUAeBwO(ra+Ca%icciBEWxX!4{ zC3Z)3M{|reE=3K2cn7+ku$bB)4huw4d#^ho9Z##isxJUhD*d#M!?f=s?UDyc6q~7K zTt*1*)YS6FX)!$Z{(w?(kmxQn>kFj8V-CBGd->^G6-`OTnm~6%TJ4ZQtM~=8tc*Ee zCV~zVs*rfiL6K{V?*;0+4?cB%df^abWMs6pvrM$u{yzTboz~!>?+@hcY+ig#BL)l8 zWm0alXs6z>h;V_EJ}#JhPU=z#DxS`}Lg%$P`@!*eCA^}F%(2)7`AH0IO+0#sN>0`= zTLJ5aqrh6}leDZb)$dm;o;48?f$@DI(US)c&s1#9c zb5Qe`Rhzv>tFrU)A=`Y|$GcZXwdD5o^#*dUMMP@N@Va0x`=Wh^bpF4--a4$R_4^l9 z5s;7uNs;cBE@_lhkS=LKI#fz&kdg*zkWjiCBwWZw5NSbLx{(qD@0jlI_ulh6_dI9+ z;q%B^+clYMzT+L^Q|}|pZ<>OSVd_FqUv6AzZ{6%Lb6qthXqL>4Ahy)Xd<}WGH^5dbSNT+}NDLscH3eFaJ^#<@z{(3FM%0XM=ofm8wICI zzhtCr0Vn-GS5mKcTuJ!tjdx3z7D%opa;xL*X%Du`V-4ms@8F22%6po($D}}u^nTn- zvS?DUQ%|FRvRGV!ChK9xGbBs)ma{jDh2gZMtZj)alP0qqs7 z=js#aT+LXWF))HquS=}CY2i_ z?nXWTY?EDFTzIe-|2b~|qX7C5;QcC63KsXEzYCz;NsBtut8li~az@owW}8_ao&YMN zbOT@V`|U%@Yf7|m9oHgiz|{GRTk)e8RyJ8s9VFitZV$rY#J}6lwucoMhKGg@JZ%~R zfGT1^8jcixD{ONM3y)xDow7gez^cII2ue#!b8%CTSIRb`7rz#Ng|1^&7&dnrVitTh z+2iIjD`5S3FmW0Ic`1W=*(L$L6M>!daWRF$0)V~;LJ!U+^~hLX{R6Yls@*n;{#2dA z@Dx?xST-DCwYKpG5SXfN8hzR5(#Z(epKR*aD1`!jbMg33LGQ!${8^mhz3TVA`l)-~j(f0YY}g&PIKs(( zB4?Y`ho@lJY%wXO*r}cz-3vYn49g*VV4#iHEV}9L7|ayX&Cf{PxO+LfqGY$?cAKY4 z@0Tw<$P^%M;CV3+5U}_^DUzn4V-T1h?!!-u#rHvGy|$E81(>ccr$Gg$ix!-kCce`` z%%NL0a-|1UkI`*aLnk!$VF-7EQ2!U=*y69+&}1IDt>q`LuP=hu;l1Z|G1=HoFCA#M zDUlP&5JV40DGo%;dy75xvA17b_!uwvm+PB2CP>+_BPWDm;NOgdPF*{k6q_Tn4_LoF zkT)Uc9d;o!Kz%F_LJv)ev5hZi%87%S=0$gn)=)KbQtn+~5KUaB%xRs^^|Lg%;21{r z_V$~(Z#$#slcV1rJ9$bTVIF5nz$;?FfUU-|7EnzC} zqbpTaRW+pfLN)G2uDwjKbA8R=D&=yIr}5sbT)sJ4@#+}Z8_gb~r4!0cmp^?6$CW+J4K)e|&bd1?-wIxMu@l`+W1_;q_9- zn&_`rmbOjC*VX$7zU(CGn!u5GNpyZXNvpZpKLks+0!pr<6PrfJ#tiN&L=f)eomQ0i zBY2=m+u@ku;h`ZApeAH*6n)zT$S|w_l;Kz6))Mq76hzb{k!!HSq#0CqS=(m>}n+S&t$I{=ALc zQ~JPxjlikrk{iUb{r+XGlOT@NiqJ7QK1r267?cy7kv5$o@;h|t7aX^ZAQN{((fLAa zsaBPMHb%WH=sBuR;P{vPkp->I@sQy7&`KQXh zgF}zPS+=LNC+=5wS>yf#?>0;xSGzp#Dok&ehJvHDo zEh{Ux`*CMwll7kMH@fSMV^E?%nLD=4+%z&lp4By8Yo14KQGjp_Ycs_Te8n`!a(1R@ zDz*>OzZxjxkQ={@-uZKStQb~iGrE6v}r-sX($RkX)mOkN$J5^>HZ)55vIef!UQ<{Ejv==W0oa<+TssC0Dn z3E*Y>^Fu%b8h+8j_R>~x{advSO$4238d-bDoXy9%BjjPP2ff?w%PaS6r(hqXi*~HC z>jU=ZY7W;2&U~QhHc)1|*3bD~ztK2=&gfP&&1sXz#(@{Agm*64zcsQt=aN}Ejyg(iJ9DTEylvI+D zneTZK&AAk&}tsCixT)8ly8xYeu>S6QT9oiaQX}qnx(-XSc%G1>!b(g@YUazP>z^NG^nz z2;kD1>k~43Wx);ys}QFY`grEwh7Y&1mQrgK3mwN;CZ1Vsh>Q%IyR2I5XVM4v=ctLA zF`_Yon&|siH9y=>2mLtBGeoUeI^chKiG+0})4Qy(fRe+l|6bM$zyxZ&*Wh$j`79}4 zcTT+N)S%ob^FaC}7loE>_bvTxF>HcYYzD&(8GM=~xtI$8dyMR)nMYins393u4`1=^ zJPYn(ai+!QL1l`{-v4;VK2vyyFUObRn5WHYFs5x@``{Yi`0SynY&Zo7XhimNIDSS6&kQ>1Q5dd% zf))#>vRgC1*#)b0+IW7MJy{DZUdk-=^s+X1{70``^kmPJymrXUt>P1#soH$QZy^7M zSaw9@SWW~_+k_EA%~R-+2>ST7*V)bgmFNk;j(^80pXOt52syha-@=!3#?pTk^Kg>4 z!5ysN*QtTJmIBsSV6*$HcY>1%xjwkYK7ZCs*Ipz0CE`DtDi(P6OA9uWn(Ajl8#`LCuiWg+F@wWxojp886)e#^2;n~5 zVw}9us-#5Mq@?*UcI=W%JeF*l|3YX6=M*@<=a!d4fH<7~jy+AMsGzWR$hCC3QQiYM zY$)_E#=2@YE_!4%{hvM+5s~OLgKBWF{ly>34FEB%Z{C$u{dG3}E`7kfAVd7AQ1W_b zZW*AQY*N2><-pTLP4e12LP52(vq0@5#dyKHUCG^AN+O?uknsd`k3S%QWMpDWwa3RN zsg|9Rfp1||OoW=ad^6Oq=c9IWFVeQmR|J-Fa!PQPHXgo}l$7*f=|jC=1mO)W5fRZr zQAan0Bfrk=7jyoYyjK5^bfZw~@7FhN_0^`~QW3q(Lg#<nX>?GKV2rz`$y2fp5JUs_Is~fylnV z4-a1$AK}nIqvH`RjkOI`!n+-xs>m?|Mt2|pA^h+8tRbvq_CKMov-t5Tc5+pB;QvI&3`wOlF)_^&X)wRFAdxLeltL zU(khlrG12tKRCROh2>vf&Z8Jf?hp9uhP`d|w+h##du|$mFpz*kNh`qn}|Ek$a|~yRI*gBQr4;P8l|`=C%6yZu=kNGC^+bp8U*8TH(UZ~ z1D(*5iEQ+4P{SB0HE7v-v zCcU0NSGk-9LRs_^*qxleNFJ|lwJ0&Gr|b2IeX_#Vys2VKziusJ-rL(dpHWtb8DWu` zX|v4iG24 ztboDg*vs4Ea|&NqnV3~Gk+7Gju<-cy;-Ure3J6x%+1Vj`{o)8DfZvSrNd5h|V5IpA z$RZQW^9Z0m>c85{sf%yjJ_AXY75BHT_j2{Ko>LCXaWY%>pw_xE1)^h1NvFM?7k`>j z8&B7UK9+SyHd@wm+&7B}3#({{(gc#3`wN2?z>2T_0?gT4PcOyK7l&zV^DnTfP0Unt z@O&Uupa)7Dlhmfazd!Pie~CW6w37&CEjP9ExB2;N?ONuVJRD#1rOP8Daa)bjQUz=# z0|NuM?!8{OnpxrI=00_MvN3WjZsEdu>ZBwuR7b>73m8BU1I-O5rvsDLb!7H&%jO8x zn0Zp8o(`fL5x@ZnX=$Cd?-Ms!og{n#*-huk-eN7sD~+F#q?0v|i4pY~TC(AlX{nl* zyD7K_4{53!^IcY@?q5+dl)7dtQ*%v^?=Bk~+a$eQ1i`fB<;#~RC(ZP8@UE8L0*p%t zL^4Z&R(&~EQz11LZHaFGbKxCt+7(dlfk-A)AQGk0BGm7CgiIhts+%G+b*cM3%M|-{ z*R--J%<- zU9`ui5z{suKh>l+$Vwx1LwScCw|h|MJQjKQSC4Xvmd%r3db3a7fc7)zoSIRysN%2j z0qjQ>W|oFB;l1p_sR@e&(e5Yb)xS=tk(G}ZS)6Sc8Rf``dOS$8H{-pN>OB&)&8;Hg zGF#9&XNi3a6$dN>?KAkJmeuyIGp7Q%MKJk_T(Y*7vDfAuHWa5WR-wC~J~wsKr<1nT z_Xf}^X0B9bP<%K8b>+t5PccrtiaYHX=&K#rL}fH217~@aC*-d(fg$Yf?zW}tK7yy* zKKY@#dh|>WUr4b~lmi1v)+RpnJD!M`V=IRfeH&}*b?P!$-)Fk|D3Jd;PmwhR%oF$Hj|#26F94^jhAO_yW1D`3Zr`tZjO*^fB!xZ zdF2(KO#AQs2r>zyI=z->e2%S$Iy5*I=*$2`q#5F?f)Z>)VuJtZ6JZ--4$HT2IhoSn zz(i}Nc~rP%JN*v7jVxyAS~5jT|3wbd?$49iKNp3Ik(?cUs0pf94864mP+B4v$kD+>r(x0B3g zBKy-BVxC|UuVFia?++Rb zgDS@yMMcY)hGy(Z;!2{0OJu!UgFbBm?RIz@UX9q`x#unWKn7~rmu2Z-Oq*?5ELh6% zydz*5*gj8U?vnwng?jS}bz)9h%9M= zj`e26!;Xtc9Wm=Jz6LK4PfixiT{E`bJX^)f<9$n&NXFIlfQvO3Xv#b;x9{zuS>ZSv|zg4Sw;*kqn#$V8Aqh<;;SpT<%p z`!I-Yy8m2q{$jJ}+uC>?`(AzGMrtYwm&YLdByCbVrDbPdOd?*(7i7M_0QQ)!LjsTOH@?U9BBEHA9Fk4zM*)0*%kN1{u;*8UhS%3DL(4PeHpR3!0^0(CkQ-2 zWR~=t9_nY|u1Z>jrz4A58n;K;2cyTO@#q;zZH|6SrPOCx5d!MCbUwLAG+P|Ke39Q) zk$i8IyF()rf!7wMPegzGiBK(V)je&29v0}>79bp}ZRr)-flA{#L z+vj}V{%q-=rLrr`*O>*F-=sZusd)YBU*quwB)BcuX2OUa5Pf08D5a&D`jwD^;$6Lf z+snnPqMf(ZqNDC-MMPpJ`Jpyy{ z5BzILd4a&&Fk%D@>damHt!hGe6$z!ZLr)^#Cq{;33Oz{nH|A8*vrW{0Uvk);twuI) zDf9%?24o~9KS%W#e69~vhdp*c64Y8tbG!`3W`#zcNpbntwo|@unlaLw&N{3U`NY*z zK?r+00QK3E{EUwLGPd{?^uO=2WV{tW1;ExgCj4kVl$M^ESurL;QKp0LPRSTZl#N+O zTrw2E9)Ga489oTz@H^#4mC)&A?rsnTKWgjy>_CPPC7Muf3 zx7cuMZmOs38kso$9>ebJkcAuY+LgNV>s9gms<>BOL+OH7J|XPgdqgQ;&M_#E|Fkm* z>ltvBhZ5-MD%ao{%=mS7Hm+yWfxST@KkIv((TL%dDI3<~c&z|uT*^dKQ*(4598n%$ z)$p|Y5*EN1p|j%#mc@0xalMu|`QeDKD96tegFIPT|C67n%pUr$KZS(u(8y9wQcCk6>s4ZjccnCgUjCFlyp z(dY^<1=s!T%?c%9BS!l5h?}xsJBD!5gq_C5n4(WK2*=5B$T$b-1GIE7=Vq$~G5$ZX zhsdCVk55Y@sWLb;kp65UMf!yl%mex5fmW^05nyCBH*Bin5gG#ZbIHHy15kUQbuL*ZMmIr`aG$b!4weAGDLH{GbS@nyho) z3?tT%AOzy_M$^-sGAcxQ?hJm6)fDPHDaLuOIQJ-w{biYZO^JVm5ZN$8*=n zgjx0U{HFvfUU~Dj7%n;QC7^{z#snPmCg`W?oyNxlSCAdieuKg|HxiSfCZTe@1AHL| z{9KmqkMFr?$>e25N9#e3#mQ=6;efS)5N3_5u9G9}vze+Bn&?zB8)NX}gjES{dWXMVaX#{TFr_h_)pl6&CS zHw(+l1yf*uEh;rhQs@PDtHwrNrGtF<4atYTF`028<|zOk8+5MD_<*aVW+Qq}3Jnvl zqzMCBWLLmKbsKDA!&v{&&`{5M-n0@XUe90}esPbc1Uy7)Ad}2( zTyrVc{*@zrkH}4%4n$OdeZA*@6%|C<=Teo^=^8zI){tqS?;h0SgzYXfG^%ieT!CP* zo&2>!hX;ulf+3HO1e?dvN4ThQGIDeCBLx)aU6~#a=obga%usq;c#?ED}}C+OK^0b z%M)Bx>sP+k)&&E&7KCcpZxcCm%^}b51gs1PwTz$bykS;76m^|iA1&U4=VZg7W`@_X z4?D8J`4OkK)|{YO-A!qgXwr`jJW3ol6v)41TnSzI=Snu1cp{4zR^((npqgzgMuF<~ zJ3aLl0%+@8@^}d@&z$}a_19}#(OZ>0H|5&>&T9;VF1M-5+4OuKh{ zZ5gw=N(eK&1?i*+=AREgu(V1QjreoE^J3DjGkbB;nI`x%!pa#(ex|qZ9EaJee0VnX zn|0$4R%Imiv;{aAvDQrgDG~)TTxaWLia*i*qqq%@*dW=8!&|o|5Jmg+ICyavBYJSb zZRYvPJYdBmGK=FzZfZ0Uvp`Mcb3GsUbV?jeZA30&IY8^OyB4gE?{m-n2!FIk<;d(* zBMsb*Br{Ex4d4ul#ua&XE2~8G2v6xTSr7wy(+TUgEgML@JLmH%XHx^ey%+~)u7Pr5 zn(7b|8!{&b*2kId0t>u<{3tIIdy*t*vUz zNzf7n)#0NRLfXPX*f$z6ciwUA0DoXF<9Ocv?QxvtZ$U1Y9-rK0V3!a!-xG*-vIk4d z#C`uZ8}qULHJXg5N2j($X)Hck_((? zVdIE9Sg5x#ia*wq=mf8LKXTDlGVwip$c4fq1&?Azn>Fy7wT+4d>5$+?W|lwNJZUty z<>&lf8z*4SM=v$Q;~gZV>_W}l_-BpQ?}X0W+}vG81N#*iZQmp9jRd%51jqE<*GKN^ z#`3VU^N*9)CQ1GI?dHDy{Tt|du=&Wu&%w{xFF1-oZWE1eh1|@OMvyjo;DV1{La}6< zD3(G(Mz+4;g8*Ih<)Kt$&q<>t;4WY4xqPFP1$;|GIo=Jb9!H^hqQ#j9NXs*?r1Q>( z$(t-J9Y|0_t1hV_Lx@Cx*hcLhAD34S1dtB1g~9JF?sok^BA6g%nxJHR0Q=XWzRtwA zn46LYy52U4M}VTP-`n@0(J{BlKM5(?VqUcEtKq$0qse$*b)+pC$=CpTM7I{7Pc%UD(pI9Ns9)yWlOS2IJxuDjU?~R2`IE&dux82XYF(~deYs# zv#roD`{i@2_kj1@_p|RCl3#phS0u+NqR(MX+D3x$_$|cpfb=M9qLf}Kn)+)`Fwa@H z+c(g#8paUDV9uuR;KuYd>f%d=LJ_RPr54frgMwr;ckyWbc0XTVPLK7GSdb3%n!d2| z)AZDHTcjBiFG%798bvrf9d zzaK{-V7=N)-KuO@?>HL7Tn~cO_gFW!hru?q1SuJ(4H-GYxLCsg3Jc!v;B)C508jK0 zMCqP4LQ2;bshj%m^m)HWC-}z#e0=wb<0>k6B@3$gEKsZGoqP8Z+4J|#c$A+YWZPEM z`pCQC_Bdrmwv!pYZfr?kkeCd=CF3Re*5qNY7DpiMLy8$4K7Pl<2Xx60K12Mi-f6bf z5+qV$DHC%w@cbXPlW8pRHvM2olP@{6*t`=GsmdauV{{s z;-AnLWqjR_59Jklp?jl42yY3HP@UeTVHs!BD)Bqfgxj7wEPit#gxR~ofvPf=5+8DS^NPqcEbY2 zOXc-;Fh0|Q=5c(b zdM7jVmu0HnKDNys`?tr!O`Oo8oidycJ-xh|wwoQMgp8lA!gH83q1PK! zsXkEvGW!DAU~I1!yWU9H$kC4noSx1?PW|G-f_oO& zvW-K*3o&iY6nm8PqMC?#7@S7l{V8Cfr{bU!Kt`8CM3x!Hr1FMJ*m2d%Q^(aBGIDIo z&(jipQi_ULy}7iGR`AL%a}d^!Up!*cL){CBgr*VI59)Qj-n?nM$V6O9j=R<47h9ky z85yU!h?#fM$12pXv5lQxr*QSka&aF0b>=xCkbqy$-SFuV|NgC_U#wdjhO81Xx{t&54A{-6Gen7};DYpqzTmu1T1EW&WB8#GsdG0W}JAfJ)p z!6dP~%;R;NHz7DP9U;B0l(7>jwk8BFihAJoT7|x0=_ct?BY)5pa7z8%R4IniD{K2l z`y?NyZ3gC6t?IW8ze^s518_r`&@gZZ=_+dN$JYsNgj~v4JTmQ3+~o5K%~2U4n{R;J zj5@>K=@i^0wD!u(qLbaDqp=)(+~Xfg(3TxChQ(8aNV(z&vM7IYHt!+83USu_j-%}N znlZZ#qwBG)3J7TGzJD0hG&7qk{ehv$64<`63xm(&%K{5yaG7-3zbSRO(R88=q3I>Sj1B?emW70q{G`M^R)+>w0{Wmj5G%R z0Y$kh^O}vWR$Fut;o)D`ezWop`1vw?#_PhOtRB$R4Q{XwHHn8|>G@XexVv=sJj}Br zY# z1NYN_3Ab0nv=fq*ldHX8W@&s!@z33(rg(8IJiM`{pD2$`=%ReZIHo!4FZMiz_RIFK`Yuu&}V81_{bVpWzkBbqb9zStY-DV~1=( z9>$`Tab!O~F~qDoPIXsdhhpAQfM87lwYIb59-lbZ_6T>38 zDkABhkQBA7Cy+2~|KZ+6%>d@1(fR4&uy1-6{t1x7ssPIz`q2p$@$}&x7v9fPJbGi( zID8ga{AKoA;s@!m%MnaI^mm@}N%ha2638crr_2Xxd}|oMEsK?vN5?NTJ>bt7=a$x` z;aT54kXqdSDID4x78dp$S640zEJAM!3ic;uk1sLH$;-<-hInxrg;NR|7R}~%ezb(Q zed_0x51FAex|D!T8&3z2^}elYNZ4xKtG(aY^z4?$LuWYfk%vQH}V3A)NbSFT)f{hj^Mvgi^Y1$~qENn62`Z#^;LeLpkXO?6Y#u76*kr)T&2 z?t5WrSS5}zK{;}J`Px%p`t$_gSJRfywl4BITo+3tt(7a& zDT&WPWPpKp;UxupF@c*tyE5MW9MVxvEYXflD}E{<7$L&+Wxg3{q`-jXvV6mX;IV^ELy|9ivX zi^zj}@%N|5*AgZ{8K2{mBI#W>^8TTUkFRh3^PkIcklR*2IE|c%@D`dZuaEbQ_$@FB9okR&;lewl0jdkMbGA#>WhnC zg}78gs_%75*@65|hQx$kq?gh8Cj!hf%F4=NChmADkOpW8F&Ca37F%{8R2e%?1l;7? zOlQgsMpacMpNwSi_Dwjp9$WNVvc>>m&RpKI|4Lp-NyXee_x9r<`OeNxMlP=YnZW%G zn5-!^7}G?b!=!8l917C#d~5ZtPvFp1w}|;;neY_R#|jsMC=l7qkqKl77L@NzYDgMk zvmYyo1>xZ8#feuMWQ7~Irx^7ELPqP8wx;T*xD;-Ur5I>5j+f{QSY+JU6tr-hX?YwF5D!Q zewn9#VnPpga$u5)2?)#}aK+q-k6BW38iGMnm{gJ`3|AfXuv$^mUU@?v-SzHkLnyDG zLmnbt2SwY}_T${%IMY#J0ZIeCPD-U`PG2%l8a$U29P`2slQn#Bs6$f*ACKPGuU}E8 zy!Pc*4@u1XGjk*wToyz8-goUrqP5+@+ZZFZbE4CQM8>3%krB?yo~P4hDRGrN$? ziXP;@y6Cp2X;m2DQ>uR&MJY)Bv)R%&EDlcbu;ZZ)=_9uvm)Wrabf&7CjQnRnXcc;b zD8#BLL7>kqq-1X?(2c2v?9cuF_4AZ@d1uGTsW-?Yy{OlNde3Up7=kS1?AZ+Jp0Gtx z36D916+@(q_IvHS<22+ye~Y`Xz8+J7`|t^xQzjuH9ZI1mpY)oq5veB#WJ__2qgu@7 z1J|J2w?a&Qw4wMFr3SS_PMJK|mDaF8PS<|z&Ubn3-x`N{o&+L4`+h@9%eSy_l%X%5 zzK6aBd0a%Kq)f1TkM(l@Fo5Gz;^5pCTVpai(eZLq9sm@ppsDtb@urH;k=O+h&yH@d z$>o^CmoG1QM;Ubxnxfj~uG2~xjg=X5xVX3&{MdNeU~Wt9WIF&?-&ME>Dvf_EVk8S% zNmF=tfdb$k<9?Y#q59-GqL(?VsK|D{5zp&P)k+lExRHIM2y_mv`#*Z=_fD<(6EL4K zbfrzeQ@+WS+J&Rn(;(fb4u?S;N=LbE@<85TRt3e`Z$6YQeNfIDh#9hr%$Tn=fvuL4 zuV=)2VlZ8O1`(&t0D22aG1O zsU9Na~u6X0;g)bwf5v>n>NnG|`2dg<)#7on29f4{9%^v#Z8PwXH zo_a9Ke$Z<+IkF7n;}LVr>B0P^m0G{(Hrl%(Mr2PbkgI6#)#Bc<^Pu3{}$H& zcboe^Ro(yn#a#1p`Ea5tI55XMDvXr_cA*~+z!%rT%L!`e!o4Bo`^JFetu-GX0=XokqXWHNQ?2G8>d( zG%VhNh|@$kW8@PI)lj41S$KEbZbUhTl$@Lun$UVf{tW3CCU8^eV6sSrqS6c0lvRD0 zQOmRY$m)i@VAvwt7d&yE9E7A#VIwaZgjK8#?_|QnqaJ+w7U(M1gVp;E&0dG^AQ|ZvKbdI}RjCTJw411S^oEhnMK)Kf-qE;OaJW01INxcu&2@K9X6<`p_01Jn z-IF1UEKE$jU#O$(mh0C+ec~j3 z+iB*zyrSaz7L0EeOtQ;RBb%Vol?s@cm@o?pJ{%7W2mmpX#et`~b0q9illF>#$m;Za zW6GXxjP>JE-d1)Zj$L{5g@CZ}J7^NQ&W?A!ROw;8o@O7)RlqsZVJ;E(3fZlO)C_Yb z%_cl9!@2}0;i8_KDWDL#3uo%o*pCl5_Upsm=Bskx^bi`w7dJ|}Rg$=k*}-&?hzJ@H zrPoyWCjnmAi<9N{{U33x7!LT@reAGqom3)Hk!97}{j$vhOM@H$1!* zC2M19YO1DF_Jce?)pe*G70&*4>}3ECKdZV<=$bm1B+RAqhRekUfFmlL^f! zUeXK5P#a4%&B@6bYJTcOgMx`DsSb^Z>0sFhDyUd%oBep%8g^0I{XQ;}rh>yPGrg<$$Tp}k?}625i%VX9`(x(+SaU8Zk$?Li?+~Ot{r`1E|Kq*? id&~5{z8Dz(Tl(mVb~al!L*^y;@2-Nne3`6S!2biXt`iyn literal 0 HcmV?d00001 diff --git a/Sucide & Depression Detection/Images/Top 50 words.png b/Sucide & Depression Detection/Images/Top 50 words.png new file mode 100644 index 0000000000000000000000000000000000000000..791f15008264691fbd7a0149e99aaf900c840a8d GIT binary patch literal 33892 zcma&Oby!qu_Xa$QU~n){KtVthM5L4s0fUg1?ow*#?i2yBke0?^$N_|*Q&165YUm*a zq`UiD8_(~3-|zeDJJ;pmVwk;W&wkdk*1higUOVKGf)v>)>Qg8bicI>UxH1ZL)Cz?n zI(dQwUa2BjxWSJjj>=N^Q8~?*r{I@kX7}Xpp-`VgNq3Ei;rEkv549XosMB@Gza#aw z>BcBjz?QW5JykdTxq;(eYHNE3OJ{mLBSaV&jDi?sYy9(xZx>l$JJ-97x~mI|)VH&` z+WNM$uwDK#m(?zJW>ScY$L4QDS27p{=P~ev_=?guOI%RB!f$8WJkwi0RXO3=BJ!o4 zX(X{_BVi$}Ev^0}T!ojq-^vO2NiShYghDOaALB!Qk^2Al%LG30w0SKYu&eh`@^ymQ zM2i`bj|u#75-YdIu{GcIKEUCkIr>C|{kD%8`BO-&|sL1c2taF;H--m?`-&&es0BD10% zyLt1b6&o+F+K0r#uE&obkG){cdHW^cxU84g)7M8(s1mJ8lA<3!9Qn5weZ5z1+?d>Sq=ApIEA$|@*NX@MsQ&$xtbRh9W5g*9r7tPRq!yf0`!rkC2YGx5&x!)myJurCTB+do;~1K)ejiG$@ABLu;x}g z?kx7+S??Wvb&1<*>U81|*@+W*L5uZoXd_BTzsvf_Ws3&}3L3yCb`(o~`9hu$#_>g3 zy7@$N3}q^gHqFI%Io7i0^kso7_%y{Bf^D^4PU#=&UK~0iTdP3QXD!P>T1hEPtsh6+ znHr;iv$V97auFw@mMrT139h1RvGT)*`-46^*6x2|&5Y}tX`5r=llq&o{INwvT{oZH zzI{8Iu~1oEMa40{w7iOvlJfe}(h_Aq51NEtamgdQDrcwfpZ&6#oCx~;#o#VaxOFvi z71TXE>CRK_5J(`b^Se`pUgRvISDak zzE_C8R-4@ZCX zdM9D$q4SMZCGPr=plYVP%J5$zs`s#~4L_qJO9cL1$wj-6$yWW^Q}M`3=9tT?X=~qP zVqz*9I}XuR{hpTse#mY*1zPv0C^ii5?=iCTe!E1Bs;TBZbut2_cQ(ngSnmsk%7@T{ z?bB0;OU|8Tr$Z)c-vq%Js71YYn8 z>pn_o`zU47k|g3V_TyY{nR8ZQY+sdk`E1YU2VGblW+XR!9w>L!QE?M79jfx?GO9gB zSQUh06(*u;FdOXVSMW(GO);D?#e*JAPa!3Y8vlyp<+ASMi4}IuAuh~NNo)KXa^9pd zk|p11N?9w<&|fZDbl_=1`*5*E=S{5=yZ4txoql|WGxq;v^_O4pIC1ji=UaT9>sBnR ztPxSH8jU%6<=L5(moKx)MzekXoG9!nV`_@&=;-K~k&ow#YT~T!r~bx@ZAJgU-*VkN zp^^%x(vK(a^m$L+2T2Io@m!`qFG?#YL_>zw^24WKy0gC3*UP%N6g>+JOk7&;Dk%|| z>dsB>_geWC9{FwzN$e#t-A8rI&C{JAT)4QnKBcEWeERh1C-_7pVBjkj$P*>xJRm=J zjzL;lI{e)`bL+HK3{mC2Jo-6)F>OJt>d6zh5Su@RfsU4zmWqQ;Zz1jI6j_B`mygr@ zCZB7I_rtvBxN_z0)oa(pT+PzehuiUZ_Mh%~b{wb_hQ}kiu1lqXj=ay&H}EgwNVu^1 zQx`Jyh0IE>13IiHear>U1)wi6FDzCG9p_6_8i1k&)vK|{nqa&-*|xmDv(WVX+_f>- zuSKU%(C}Hk)iCf@!^p)x)+57}mcK#&Kq^=RW)V&ykf?zX=Z(bb6e|Rsp|?pS<;jzf zl4|NJv^422v8U{K{N#zrOc%B&cXYHqw0M-KW6Cp>PQ+rO9hKgltEQ$VJ(pI<2X}Ya zRR5$EbLihSv6yDSqgZ)dG`mO3^&ZSn;Lt1)bh_5w3SQpe8DPRd1ftTxlm#zM?^ zQ;g0>S8J_Q+(Nqg6z*!VXSWRvqyszd++#&~dACYOf@rcmFuw)^0s^q1=_I%3Xfi`1 zm=%jMR4QffvznKgn3^{F6CIx|rAxz+cV}sEZvEk}MjttY|ph#Xm*|JN=izU zKPfr&ttg6`6NQnNW+~P8$Vn;L-a@rR4^>qX7weU~XmM}O5sGz~1N!@QCtfk9lhK(oN zZ^j34wUn3%<4fok+>gPf#JV%(b+VE<4wJ!ViuSJVlX_q8LlS@zp3?hbpAPVaKBZj8_Cp&4#mz1%WEy2DnFp87drP6XEq7h z&3X^PS_wIWN8^JEKR>B&?8s2D_E>4M8m!QYX485sA`aEWuFT`4m9`QAfiX!UI@(26 z=sgLITzz&E6O%4sRULKoit9jkm#R*&O(DrL-S&QNW+NO{B3*fmFL5wCJ3G>vP^5+5 zsHgYU4LVJG^I*HX+PDvgUjF@(vv1!rx6<^>ysD5yroRoOi)2!JxFQ=hPJea*0Zk;M zXcd^Yf090*k)Hl`ch@T;(G>?kBhFL^uNIPnWEF>k*MHlIM9fwbbi-cUc*H>e?FVUf>$3bn|4 zZ^)#YbtHJm6MumUx`8DtDO+4;9|D5V8VRZCr}7~Hj`2kR zKrN3NsHj|{;kSu=`t)d4nfpx}kDn_oVw}Faqx9@!NT>`DlUjM4zX#m~Z(uj+!utGx zCH_;g*K){ZK?m{s_wUb^Af;!hy1L(?i9I|r()h4uxO}k{Y87U5l7+Pf1e}#uXiexs zDkg?>{Li01GUhidiQw!Vo`(1ru_XDjoRz9VuJdexfr0H%9;Vt;U+TJ#MLS;J5ndgnfF0fBcfsNtr?EKVsLR}^umUR~(YR;FPV__#<*omai7 zQGdQnkyZ?EZ?Dy9) z$WVrY_nxYmXLfO@3cO!LL1e#R^aH8K!@IWzjQzHMzvxo^zpBK5EGvz+3UX(kDQfft ziEyjt(DoQWXaOJ>@O*vrDdUwJtDB=CGp0v|?>ho~Ht=zbaFQP4l>V%1aAdgun{iPD zPNDdru>-f+F%$m20+gV|ded43Lf$Xe^$QUph>XT~u)^J8arhp*6@@Yw zPaq-y6fs6RFEsu-0l~IVJ0gC?^AjwC&(KE$Ztb-ccAdv{sq_td!!1?T*XN+1pvZTg zQK#vaq8%!-j#pJv6EwJ&y=_V%FjP?iH+%T$sfq)Ei7l;5ycUraYh5~B2aaPu!qzvZ zvvRd>=#;x)d2M}1U-Pq7VD&4tp}`t!lcJA$5PZ6Krf=$QJ zGBWurcv@i>%^sawb3_fVbHkqJU75%ts_7+tEgdoWijSN;cJdXfCXmX5SpgzVqH33GSj-UGPcI&c-eB8rGmJXLH`m5)=xdm9nW=l9ZO0FJw_F zH2p=5be>~>{`C3Q0D!dCB_be&$GPPq+`srl4MThi-;?kEWIab;wCv4KMY=DWk`WRb zd}Ex#%kTB|gboZzvBXQtUMKYZ+7Lb!NDW&?|cg4FMw` zNAq`Bf&STwqw#~cZu#zeS<*P;<@vQFOB7 z1P2?3-Un`i>_gRX?xQHGXjYGq=;-J^XNR(X(R>Ub^FuU0*A_e<&MIUqU^PDellLFdd2-^#!mAN+GR;gwVVkqE3FlBHoA1aE)ACUo zIA7?sJP6=Tfn4IExHxJ{ZeFXNTQfmwHt<9~Sn@rVDM zizF{S{YF;O_{4-x6uXFs-kZyUSE09!N)qvuk&}D%EI3#eKo66!uug4l?W`Luextjq ztFP3FjhR{M6e%gw?&fV+G@FS+mbOoy7`J}e`aJ2zqBm~>ylbay>kpC)9yL50Uu#}1OLQXj^2625Q1sqX`thg&8x|}0vfhgT3HklyE?=oNw~=@ zWj5AItLQFYw#uDaN*YxlvqXaPm0~f*KQ9Y{nKZGAf zfNq;C4u@0CDW4aTR+M#=JTlDMF%ryx>Np;=O5Vw~m9cVQWo1R?U9j_N^?2Ww4XX9- zmQ&!|0o`V)ll6Jd*XKV%x+K#YodsCG>VUp*cI*|(&lgFTM6Lrjum$VVkFCpHU30hNeHEwmZf5d&jI>O zTT6>YDwMA9mdj)-!wgksrG&G#3RQdOT};avdQq#_!-dinK}D6G8;#p*Gvh-C`^5-R zCCCcrP*6*&S^rLBtBGF+Sl>xq-SY1C1B_xaM{ebMT*Y!DYra)4mnDf9#4`&gXG|=& znRbav-v#{u!jG_Pb6Xia$CA=(w#LSt;apUziAUqtk0o-Oy~&2;35}Prj+Ry|qr2Qvt~{cHET(-PCSjPPfx0Uz-q6Zi1%}y?eRs(9@IvriXmkC z_==tR^7d>kZp-N$m4vXcu&&ktypIg4NfKIV;LaWP42WVes-9I%Y$N)_9f#HDu)|Tq z`p&1UAh#9HbmC;X0P~?CC0?MQXo%yrw2I>v7A73k6&HV|Hpo-K#K_na9TjCVv8Y>) zsduLncAjR`&~uLZ`u%I)VQTF})AD)Gz`kV^6C28r$7gEk>L#fDz$s%G6ynQfRMXYF z!^6X~mofE7sZYn#oa6a4#goa5W9c#?>JN7ALGkd2em))VGeXI({qW(#=bt`(dg$Pg zFT#1_MxKR^n_J1Vpdc9=n;e|YfOdO(I~xuyfL}^TxQrgG*lLiuj4Y8xS`G9>e%ljx zY{ve2tL-PLU?+6K*Fq~l8pcmlvdZer5K~ZQAf_ZZ+x`N-&lw2%_kF;L+)MxvrwG@4 zOLaV)8ck&`n5R6Bh?5?DwN8HF^871kN`Jp#X%pLBmqaHOYLNfd>>@IMo*%as{o!=6 zw7J_;$dY8Fhk?jp{{!*ow8W9IhPdcbe?Q>gQ^7N zg6*QGFZA$ydQC5;zYkEoF__-xhN9RWu3}@*6Um0!*Ea1(zc&C^#SX-9zPeiQY1(jL zKzl}i#&Vl9CkS>uaRPWxIeBNIW2QSZGxO8B%6(u`?rkPT z^(g~=eVZ0$`8Y;s1ziX2Kcq&H6hj}1bRs6uFLvpuqHWut0yeBp;m3#l1OB#|NDJJr z7hh5T_3IA`4P6GYo!*wQH^TFI!>MWoK6~3*58MXCqSmQ9KfAB|(b8Nw#y58xu1N*m znHfcV4{mMoN+Yiob$(y5E$8twbdlW#zIsRr`10k8OTd@0hKSU%vTdAY|LVGWBgsStyM)!vL}t}%GSbZ6{xg#TT#1)pvW@&&Xtmf4 z4ZW^yw-Q+=$s;_BVYjEqBYISi52V|35;(0L4-wO2VuGG3Py{)nOu=cDz#()L*k*ioK4&~;QF%#kU- zt@byUj8QmhHWm_4cn5)r4FMmk~PEJsz}8R!7@_YuE7nH3T& zXGAyen*MypfV2diSsG0UG78h3d!CXz^D!+qozHIwK-24;$)7-_*GN#g9oRRo2;pz7 z_dE%0*xg#ewqw3nL2MkB<_{7tjUnN1d2c(>;DHG0d42KreZZWCu1_sIS+UJ?{iP}{ zbA3$Qm`cuukWO+s$}uluP<8YRM9I%J{O(0j=OD@PBg(pMin}`xJv=>7UhK8tITQgz z3KZ*7RaI3@*~~w`;+r51jfKl{eD+&!7GBW2MRUh3Z+4(u1Gpy3iPjec=wz>lif*S{ z5vm3baoEo+fBA@y>%^Ti-FH}&&jU??2`qZ`>K{)!a`MMttbp}jZeULCvF!6zO~qgu z6I}<<{_M=cg#SC3{l<+SPh-?uSL1jp19^pnw15Bp%~ry6{kpa$-kbdlpJmt2H{EB` ze*KOjsr2(hSzb4YX=gaBx@#V@w0s=T=V>j#4Qd@9K73%yMl-RCiVoH$lJJV~ zadIjGgh=3q1BO*a2d0KYH}*993Wexq$v-4O#gry+h=si?xoma@MNOxH!%%N`JZZ09 zpa>85_#Jc%MJB82lzEJ#&Exdly73nfd559rA_6%R*xdi(zk~e+R|j-V#b^c(wCBUD zPXn$oGGR4;kH!mJL-}zctpD3`+qV_7G9fQVsi|tZ#U=Tof3FuyM_?);=YXiAm`Jr^2Ya#r22 zt==mNECig`^z7t$_dt)XcKZUe`ST#}-`KJ8vmkcheGXycNS zID}nh-);=~8i;NV6KPereMWY1pE5dN6N;!blrg>MY4X22!>&mkc1ng{jtV+WzWrC& zkqfbilJg89Id<#^!ok(}9Z`Pt=n6tI4}Jg%VEy97i{}39EG!KkZ>I;&gYp(08_R;Q zym`0gPXp{-v@itP^z{JmGqOuuUorQbUpW2#8ltX*zmoQTwu7Kun2+f!M&mKiro28z zOe_t^laZNO_K{tlBajY$p6=eXMbpz3@s!Zzo205Ojc(B094No z`3G?dF%~8{90w4qC^EH5-8v{bsD-7eE>$Qq&vdy{ia%j6z-D+!eDp(|JQ)(6l5z`r z%SDNtNha;T(Zcr)!3(yNe^*#k&uC}vZ9F5hZSq%#m>a{qrM&~6yq8=V`S_YH9YJ+a zX~=~P@{W--G#Y80sfl-RERzi`8j2Et-k?MHZ*N;`=`YXvxXe7q1^}6}q#`b({WOWe zr~0YirG36|Pf!NUz`6&-@VO;(nY;iwPd3SC8$Cm#t)^!3YfuBLZQ`#$P?;O59y&UJ zq$uYoKELjug>4}?RN-$1Tfz9xybq>3Bw9|>i?Sg^3o$9Brn5ZZ5xZ9Y*GwzU%GGeF zq4iG@Mm>9r6$jZX^= zEL)rD*2&wM{1^(=M`C$pWlk3f-1iTH8$p%zgx&J;Z->#HV}x(TsPSME&_j>rsHo4K zvtRZ==-KiI_v>QmKB@_l84Kk+Ns3o1+?cG~7-Ve`JE-76w}1Vc%~<_`7p*BfXU;2c zd$lz~p3DT~0QIA*cPz%`1N5w{vPI|x>{5828+?8oC?9c?or;8nq|e2x{_oM2GkItx z>(TG8$$`!vVG|PSJDQ;0>vXC^#Tsvq&6(2s0Q}(>g}S&Z#ViN9>^hQ;GZ2)1r1cYB zM)-_hA3j_~I2_OiZtvBh-ByqfwY}eZDqgf(szr4w!wAYMDtQUmnvDV9a-2EE^^+lC z%{TnYWaMqwf1TQvQ8ms7G?mT?;uEjaq}V5e16C$pCJ>n(X?*!IQuImc?%lk)|pE`n0bP-0);t!c=%z1jOD~EP_kIILfGeOQAmx{V8fo-gsAN9RtPuI)#VLaC!QKXYHb9n!}gIy`f z*VIf18)LfoQ7Sar_h4rzAL33J)ZD*~O4HEyH?*vvXBJq*kraaHscG8=yh?); z9uw4{yv*MARM}M9p{g#7MMF3f zHf2-S5spUJp@FF#BTv)t>sO&nY!GgC)(R4D6fmO=EiJ|I|9pQf^0FgG4bp1(^XDhBYS1b840&wIl;glELTj@#rY&^( z)vq!GUW$by1EDto2E*DzCGCDobdV7CJGxLl?U2kj6&;A1&Ukf4VTcMfJo9hMEF&Xh#c~!R;C~yD#Oi7-*ttmh2jlgdjPD93(chW{(E!acV>*x&2Q+XMSZ zm+AE4NS!jI($GFmRRjF@`MzWN`~%&liyp7$XY9*z7uH&}wCNKwptO|?A6*b@cK2!K zdA>H=i!Y<9*!=yn>tF>kYyS5Iq*pDB_t+S4#pX^T^%`szTrRVkm&AN3pm$c~v+iTG zt=ht5>y>x)sZ1R$qweR~MbBg|sYpIl{eAl+*%^nbtHTPQ-#v7Z(P=R6Z-@G^82|`T z2GF*+q@;eoWe<<}0MO9wSRMXLw;w-a)34A(z%0UGT6S|-DiuQK-3xWIFgZ@Z{sSW8 ztLNylK<69*BuNvn08Anx1I=ks{A6ArjfH=ddh>6~DwyZ);Rkt7Mp-%BYbAaNcasos zM)Zb=$7d)fKSnphB>+vqy!8lb74WW~&(jK|^FYH2Bkw+8mQ0 zQ^O_V-Iwa$sOybo#m%!JknGNboNBN)qK@w0ttO8+JOAqFK9k7%SSQxqAmi(Nu2AW0 ze!r+$FF2^Y_hE&VDGtQBwEPHdnHKe8X&Z;nho`V5=8Il~8?k6Vak&S!m)UDVR#uj^ z0jU#`=_br9ft+?hnmDh@nw?O`Bws9mrv{OY{wKeBdGovhmMiP%usZ(v8CF|+AJO88 zg%qVctXHeNwj2<~aP5L#rAMLH?ovHcQsb99qSc>#e-#EyjdoL}ijL}(yV$$M=x5h2 zN3||qF}twH`T1N!cuc?g@2e__uodEeCJ5rS9X;~fwuI77A;n&;1Bc#$J%AM~4}yp> zZ2(gNd*0M-#1sN5M8oFNDEm9Cc46aw4Jj1|ZZ9+QCPFL+3eGWmO#6NXzI_brA>ju* z)dvNRh%7%KNLq1w>c7!J~?_TqGJL7K6dnNFY;h-{QaP6kBps&YahwG|nrsTBqs zKO+MT6%$Z_M~R5QTk)+q>x*1Ke0xnA*?8e&D!S%@_i(o+obmn&CX$f%oaK{eqCw18 zN!`AK^aas2Mn*>DK+)At^Gx)YI;Li4|4f<@1bmoHT@1~!uq(6@x;|Uu4s(UweeQ6s z{`rI)BNghJI6iBO&H@mI;u||A02&uAMu7nU8j`NWyF@YSV>~73PPyDCcosEm2XLDN zdu?xg_nmqc5$5i^_YA|-d?-79TfI7$=9xTN5sy1}1B3cLG8?*oztf%Oe$%2XHr2mS7$51Dl=bd|DNOE99%nDtfH&XW_3N?+Z>Ieuuo&hkR=X#Zy!$G zt@@$1gzh}5{|5y&BZ+oB`yaYH50&vH`^WcLQj}_z*cUhXr`!jIh)|1;|KlUL1t;C4 z@i;UoA&QO9cCZd(rbLwZw*o(oT}4x`DqTId03LaZxmR%KzJr%VtYU)XBu5Wx7~9Md zvB-!=jC(qR*8>~okW5SXSM#M}&qF`i3^~s%!Qo%q7l;Iyo%+z z|CE|O;;ctV?U`?Z%+mK9bfjMxYf+=@SZ&wHv5tZA54zZk`>zh|uNOlgN|2nPvl_94 zl+}Fh@ez55+Wi3f-Md0Av*uJoPyke*S{p<69pCP!p{{-%5t=@#Nm>5`ndVo816h2U zefc_x8g&g$EIsHJnT_^6{1co<5hzWB?!gab=oY2%QY|%BM<4;)tGqpBK|)2@XiU^l zQc;+ab?@%j|uu zM*U2u=p!g&8A1*Ao&r3qiT!`Iz7w4_7?B5sYaNh9OrU{9%9K{VaY(jyp$r%z*nj|Y zdV{Fiz01%y(!J-bSbTbCd-5ZfMaK;-$a8GEC6CmzH5mZXg@e@%L675W-3G<>WKIAw zA`nN@8>q}RQ^5JG?h}K z_m`VEn+y)NHF}Gjv;Gw_h;%eH-yt$RLd{K0O)2Z?%~d$m!ksn&5(B|wAt52_WqiI1 zgO%lD|`Nfzi>ZtE)@R%xuhYM3B?FcNa;m%G}?C zzI+)W>b?C#o`1j?5qu9FZ%*3Wz!cv!o$D)%0F>~9DMt4yB;0(6JYgg^Vz)7P3^6e= zbXaWAx-?%>*vzj17Ij!s|BMxAXOyixucmm+3QT`7^7D7ot?`oCZ{OkOR&j7}_ym3` zbu?gg@PK&BsM_9WooVH%nEK>@IhI(?$b6Hk6&I=#VT9yG|~zU?+KG z=q%j@23kDQ=5AU*_q#otXExsVC4iBa{nsU8uK*uB@$qBAu)7mtS&!=cwUIoe_J?+6 zlL5+TvZzzAxXZ)nbVeJQnr|z2V+>>rujLdRLnR2qU*q=|MS2Yw$*Ji_ULBZ~ z8zqCIphd2q@t04#jaWm5_U2v30bfLb$ok`2=$)%W+g-)RU~WU|DrDqkP-hzyeYaKO zyE`XDJy?Jqu91LKWB5tUBQ~IE=`OqU>f-HMkKjS?jR8a;2K-box>{P4(Y^!{SyR4A zla(&f0C-r&0Z~!T!pHYF!^i79nkqy0(W?>^umHRexCg9B6c9TJvozF~F0p{26@umQ z4kEkMBVu+s(z>Y9`a&EGyTDTR$@1L1*|;4r&7R$d+6S>j-8m~W6KLPFSrups>n~4< z5!$iO6L%_YwGbm8FN26`zc-N#3l}hxH^qZdFtcEq>&WbdCYtRBnpmNF3@CK-Tm$d5 zPI6jWYqAl=RFy3L!5zI`9n~fwtCDm{GBy(cT{Z4Y2!EoDQAyM`!k@*&&zl(=KZM?R zZpJB{qzuY=Jo(;ZO(c-aD(HHeA96{Noy+3u#mRq8+Idm zg*yYQpo1n*f%k;{Ul=w6bOyk%4+ztS@~Ws^PxIYP##L`COUt2WUmYEcxlG3_csqlf z`r<59n_*f2_LWOX1QP?Ll4swgAb)MF{j^{%=3BVdaBc}{RtCTP%~)OqCxh>_$GaIa|SBp3~QJBGJ{jGgFPrtogF#qUhtVClDJ6 zVkuF#rvZ^)F6oYzX>$S#xQvkXN0JXh!Xb$OnEZ9!o!1`CNOLL7N?-ZcApp{ou-jq^ zF?Z?_OSlTkD%|JtpdzrPG=Z3ao!SBZ7U9|`=k-0NnB3<}C$h`dCA6_w;7|f?RLc%- ztX;VfF<+&Ne0(X@g$oaXiVrN>+;l_t;W9)l&Tj${gSbY3DkDUqB_slQfPwW0v3Rh# zxM(M=P9L>?g^Y|W3OEtZ!)S7r5%2q|7hf#4yI9A!moXyufKbpP$_gq%SHIqa?u_Ek z^S}w-G^cLGDS3g!^ooYRIn!RI9w%LqB-;XDdDz8&gMQ+Um>G2rPaZmNy65vpoJD8j zO>62gCxc{56ZrrgRaLOQj6H-F7}`-;r02tXV~gOPNrT?yXc`8jq`T+u9p$JCZ1+YO zpon<8e?}|!3m9AtP*in4f?A}k45}_E>B&DwI&svw2IPFd?OOQ%zYetH^q!i1tFvq& zEYira8DS^e8Ts^TkH&f&N4dp-crPk!^-kAwpfEf-`VYhoRqKtKqLl-_qp0DZz+cG7 z%BFt$^sEM2iQRA@I~3`*6UB>dZz@SA4~U#{dm!yYg^o!;xR6I;@OGQBje~~$TDOxg zLtP|~ENrUn__uSH7eYn5yJz=J??W7ur5HAfnvajuw(xgEbsDUfB;X&oQKNs2yGSP= zp(q+grSEp{4V}o%h3bPUNF(=s_GS#4_^Y?%wW_>I*>sA2{;{b(*jpg1G(<2r+4}D2 zxG#P=jy*m_#@~OXdUu@D1lq=C=x$JUnZQ^8IE4oA&1F$9O(2CuNNqqZs=TvULqs|J z;>C-x(ut%9f=#)+X-g72q9s9JU%IzWz;;i%R`2M!kA+Dg8e}tK<3%h?YqWYL=I{R! zy@^Eee0~e9?*o9D{vKBx1Jn!%4JObSx2Gx6$MT@V`<=V>5IzR!I(VHG<>YXfB}C}7 z>cchMv;p&C-UHOG(L1 zRkZ>qA;G!zCeo5LboN>&fJ8JM+>%{~iQ=k64@c|ReD)^O10V&~E+?+MDb)-h_YCS{ zgLqSSghZ$*DAxYt-AxNv5LvX`!a2u@Gkz8N=;!K{JFez{_IUp{NMOF;eT(R|z2kb< zGSC?iva}=j7|1W59s9o!`x}wspxA;^fkGXOyf_>(SX??qNT&By6-x@Gb7q`%sM!R0+6I;4*8BI*-+6b`@tt+Mo|DAFHl>V|4{S%$g}{&TDYhN zs4Y>*IUEXhpHo^8uU;r41H-*jr%uh@p`L^n5!$Dw}O*+Zo@|JUI=fPv525Bhhm1suNgBAcEkt#z)1g)BASSnSkhN^RQEhPXKb z$CW;oS^e%gAL;!|nTbub>k@I5b`0)LY4HpS!DugEvK(3jt{gg>N=PYtifjgMA{1__ zYzz{zC;=KoC7<167TXFzEc*_?iZzKHY$=v5eWyVNqqv+VAC+!TXN@D&0r;W>Jyz45 zh23`Mijf!y)6jR9Ks1Eh>ix1x$l!i>U=};Ox*B0c#t>`=czVY0mhq3XS(v@y<6>_o zl6?oiPXGgKWc~H|Im%_+M!Jir8@Q>g3=A~-{?Yf0u$ajN#p6G8*A(rUzC{r(E@krb z=h@A_eg95(Mu&9F_h7@ zqeqcZ6b=pm?th&k49LP)j{y}{%$)p|ZqGw@H1YzRts=z_C4DM1vZvJ!F=P&nyr^+m zsjQBA5U52?*}`ijz)fbsUp#rG(jiU|!esL_oRMZ>>-8+_4`Zz}UKu5${&T|Y(DPJW z{d+vj)M?HNqyy$j zV<7w6;fxX1pLlAld-K0xq!-E;M3L=*{ysS{E$ z^G#%tx-Jg25afVOkoK%7isMu~-XJ_&r`B=4R9Aiw*Lrbj3ceQjjmb%KQbR;I$V>2>&fI8+ zdixH8JbF!m)rNZeC{YWGB3vfVL$k`$@pPDYlAC0fbpHPSe6KU&t1u(7p?Im5cV{?q~|7= zE%Z8xv9Za)#13}53)ZkDkQY|e>(7T@f*9~qf%kqELxg-cKkgNff6s`nyP^pzT+(t1 zkVsodPyZ8H(IYM$`obPq-$mO0%D;cnCIs)9|KZG*>vxzw=&^>#$U)~E_G9?$mUlNR zvT3|xz$=Kt>$!T45r7)}79RfoU^krmck3 z6q&WqSzdw8)(=lS=$o)op}^xbU|&zWzKh7Xb-`zv5RZJh>q0abSYTp=f5=M(G}vp< zs^rssp!Z%4lZ)e#hHQBqp_jnVW@(n1JGCPv7t2^5{+sE>ouAF5>3JzvN~^S35ri;= zxQ%CWa|9!`+MWQEP%GAU1(vl12vVhy|%7yJc!07k90xu&1Du~ zBbE!2_kO-vY4yCa8WaG#ry;FS)Ljgrbcn}c{47K%Vsx<5)7l`7|G|R?)&?}+EWgQt zsWRxc_p3K=ic(x5xgyD5M5E7_hmTL4oSdBT){TFM%`E-duV0r$XbjlI%1TNMFbufv z1`?O{xHT9PD8~?gC8_y?pnt+Zooli#9=Hw^b>#4{;e-|Qq~OdWLAVa{TNOp8bs%}d z1PKAdW#B$eaf#3B-Lt^NgF>_Hx**OvAuFK1MjWAmjx^ht9~eibU4eGYQ24JnTRgZ? zy*%`1i#+LWAX#$A$&*pjYdQ3yURJZ#GVqS$i{F2rWauD z^yXg6sXO~;xM>I(G;qpcK1A8rfq}vCy=@r>L}l(Otj(_7ATkFBI7!0jUR=kYY;}*L!bnNv`2LCVt;%B%;~~j835? z7ojh%sn=Vq!OUegHa2GKMb#snYj=*`q5DJuS(fg2c`%(whSISk;?UO0MAN(5PmBQ@ z=0o6Vl{%yrruG3j!Rw!V`$-UQ@WEJwqVe)r!!p=De*piHP=EjsRiwz{^oZg}ILRJU zSMZd`zq|G2J2I1qxLVX&an4aee5q=MFy7Jg9p+uFpJamT1?gjEqVp`f%|;>EZPX@e z)S=29kzepD0iZeeVG%}cwxD%jgHoB_*cXdvXRwRMgT;2=^S<6_{k{m&ResQ-!MZ*i z&!6FFh}Th#h=?!&FbXkZ4AT7q`nuII%di-^z1f;2?!+N+@Jm(Dii=L)e5P zBS8+1Vx$iLUIgWZ3ycU}Yt%M0Oi}B`J;wNF!|z2ypPxvDym}Q0&{!I-=o9z>+e;nI zVDvd}W(a0`#57C*?beuid5!clQ~3hIA#Sx^~~6G6v9#;O`XGn!ihmGCW$tCj9=&tRD+XJqb`y%n`7I+7*s#?6EP>5poOC>L1M4K ztR))IzAR*T#4_bD+jE zXgeKf@xjsGlxd*NfS=YtoVwtxfAr|BS{u(RH4P0}50A16DC_xlBlq0_>&{)ohNHjD z0n{ z)&BZjcb|!ar=yn$RkMM#8JMA=YdfSbN@OqmA+X0c$^R-1f)pvH%YEptLTFK!U%gZ> zTtv0M!aiFhg(=UzAA0d6mM?EwU&x)yzJp7QdINtM-aUpC%3Ve>Lxit`sbpna> zHcrej5`JQc&*O!eW`u#{wWWt}CoTeUwU-KCgGfM#H)E2pc4lcLm|lCh{cf>UFIQH{ zlxNgBvl`-Fs%%klMx0UL?Y`CVatY_ZTjjlzeQ7sA(9yVWnS%UXtb}0z0>gJ-7at{D zN)z8$7~(`~*6wPWSQPM%I>d4`B}1-5-jo=R^;wql*X7!3$pFXj-_>-t}Ow}Y<}hdn37En{P&P~Wj{|y7wCuh z%y;S?NJ^*p44^VA1-SN~80iSGwLj)_TQrQP6R{ii1J?dK{f^)5{8U{02hgXO#rs3;Qa&tA~UFGE&)$-N-+}% zB$;Zq9O(z}>%R1jYCOM*BLbYLwUD^7o5E40o{YmQvfNu)7y znxN>x-CXak-dBUR_8K@$!+~qjRBSs==XMvNDUp7Mes5B0IoEej7Ma^YY%E^z@P%k# z8V53M5pMy);UT;=;w)KX6#<^x(FOU;X&CleKg-U^Sxh>Gn@A_gt=^lnU9O|wZ^iHx zhE8r&ZHNdf5EPNoZXfWJXmTp+>B!gnF1+^l4H$l>ZflOmi-&{1F@p*M$U!oqwNqYhF%*4j_ z^QJ>ud-tmMw)S1xc??{h`Hg1Qyr~O+Ag^a{?Etqq4o{3gs4s+Rf+t7lB^_#X?2+g^ z{Ko410O5YWilPV;f`^_xYvJsDzwr9iE1VtG#f!+~E2^a19b1RNCkZ)cy#L8u&^6wG4uSyI=)*xxK~Duuv6x8vl_n%myB$@gM)vsfuieZU3GacCOi z?uI8(t8d^gk~?plofczEDiB?7#QnkeHxUHjGa+EqhMA40X|d{sF_?o zyKa|YT6gW@p-4Lot^iE}(8Cd-dLe?H?%g$yaG>d21QBOk%K1TG1e&i`S z)16r+J$Xj?Fw2Mt_=Q>RXZgVNl9iVjQspyTb0nS*rdzU)BO6PnQwJ_2Ix><`P*5=6 zn!rbD*e9l1oLeS)-}437Mr1*g0OPj-e5Vu`IahDQv3P+;eRinYpvZQ}M&%Qzr};B| zJnzuyD|bBZCy08))y;q=4v*8BiA+}rfUKhq!8b-2+Ajj>SF2Dy%Z7RkWF1o-Bcu0f zqY!X*LX^#cittzn7(PSPjDK~E1Me}Vh6SI%ul&$ij!^E!o64v@%tLC1ufu<4Z~i-U z{jY%n3TOEWr*F(^bBwf{@)KB{MJ%OY_*uH&K+p|W>)vMC#(M5E2A}G4j7e|0GniFr zTyGbA166bIul}x2YNeeHb~dnGH2=`RE1b}l`1nN_R#am}+YpfzK^bmtL?xg^6Chj=M`Gz*5dkT`gYFZw zSyhEV(%w&4rx+-8OnU|6vhdKML*?SL{ix*DH%=c0aCbyRWvTz=@7#IwKn~Y>vH{4G zR3FJ$$g`8)5&+HUjbiR$;j57YU5q@+17QC{7=93!kmwb33hOVQFOyAj`+c(ES3}`n z=FZ|!xZB+sI*10LLl(@Hh+FFLu&GVv46EzTB6ob=@64$E%egskS_AIdMN7I6lP zJp;4Pn#H+u=g7hDflnv{3bIcVxIevIjK6$xaBwi8#0J=6WJ$hf9VY%oQJA$s9zfHr z=bHE_HBI=VUBC+GMIzL&jxIh<=3sPHEcBNMptj9B$!DyXZRA>|ih|0HfVb`s9-crW zy7o%9?SNWZLEcfa?QE*}vxnk#DG-Kt>qf9FO3A0K_=i@^1a(`rYJW#0DDj9_m-85K zNWXLhplvRo>qrJC_9N)6))e_3_z;}z_29}9n!&nGqydxn@+L_oO2aCiIZu#@Nt>kZ zwJk&?00yqB3)oHIG-M33u-HQVl)m#$;AYK#{Oy$GY!%21mfQM>#15ns3}-h0QG<-s ztZ&03t>(6X@q<_r=mw4h`%BGQ9{YDxPrr031x6r{@rihT+i3ktj|O-c65_(>;#c*2 zZqanpTDlPIa+Xq2ad8&x*|?)=Mhcu?%eVf-V*{h!%neo&ZrW7SZ7XJ=oiFjz^4Rq| z!FNpen#T~$o$)|sR>0S)7l9eEf?|;Gx}cvGhc&1k%tV9Fvg??>X9y()g)|_kIkDOY zo-eM$P)s^3xV-1FW{5!!M;PUNlfpGE!4vHa#vf+ zQ`G?a;zc_%0`LCCh z6aSyCzB?Z4{eNFe!;zEIIB955AsR+SW|NG{$fi{2CS;YFM0FZc8QFVglp94x8k*K^ zB_&C=WRLK>p6YYX_woC$9=PxKeZODt=eVBNbv*-x3`QS{9@Zi?W)R3ZCeLUQ;0CqRhDHkj1Q=(!8$ERdpq$GNDD#4 zv&dwON68L^JZLsv6Nj83C{GQb4VfAM`z1ImGBT1x&_s<>2J|v|2_d_AFcZCim3suH zWG}zF4M~kuXXo<+Kl>H$ILa#HS1St{=B1=Ld?ugWY*U zhCqgafZ{n2lJmT+Du2?{^bOGJ6T>}?5-SqbgLQ>j^YqEVZ`Q2$`QDG{$z5jkcr`u6 zke%1wmt2vwXos|+&-fbYOQ*xmM0hia9NUpraJr@E8p8}|QzFKBN=0jne$niBAP(wL zH|SHIaa;9Dd-Q39QwPU*Lq?q61H!r)`)(}1e_l&R$9&Ot*R+*QiGk*|eKU84wpW^v zhmfXZN?luY{=NeL?ha8F znP4z_2Ha?izR_kgn!t+Sc_GF$mqRa-wK=YZWfRI%HPjK9zfZ{2>0Y<9Ld2)fJ)Ie) zN-(cmxNzZicRn2QkoG9ykmPM!>d&==(jbgbhgm!62PN@(>e8S!{K4ab2p=mIZDz)7 zbN>8eQfhJy(pCN&P&=pJn7rgQlYMzZzi-Ow)AARQBJ%y!W@pu@?Md>PRLOB2{Yt) z0=vHy_k^r4UGzluvbD&|_B#*O-7}C{Z00!(Ua*ddqvc?#Wim6^!$GRdM%oCqFeHgJ zg99OTE9Aow`ab5b!yEioAl+tmRg$Fbd;VCez4+fqtl};;u3d-IF`|}Q2fdc}A4}Ju zr}8rzZ$DCf)efVx#ho$}!zQoZyvY~>J=&I1Mu*F0)PH&#k4UX(wV9k4DiH28-=yBf zjOa>Z`uVXJ9lG7}kJ#xoPa1to!vEk@DZV!6)#an*$MGbJ{c_XDQaxx`&?aZK9Aol` zpO6@5osVtod@YoA2SGpsCLJ4GNBeyt5-pq_;Jvl$@J)aJb(md-d!+=Q&QWw4>b8`( z0@%q!>K#=}_v~Vc2k1xS@xjSlkdD}h4xkHWdF$uP^ zSJaLvF;^>Ind=|bzxm45Id*&ThJpBVH;2+DZiBOayNlzuN1M&*Rab~Lz52JJPg9*Z zj>b1k8DBH;Rx+fs(^Fmn*g|rx3w%}{AjmDN5xgtmU<7_x7okk`v&^FHExBT^HlFIK zSbR8VL#ns1<|Zj2On6{XD3Cy}2gpt@(K2-fl7+|G5rD9Ib1KWS(F~D=&+3U&X+R-# z98WWO9Z98Wbs~oHuUb`D5*w*H`57Z6U^TW6MsnRO$0d1I&Y0n8f!?<8-SpE1;|?Pq zzq_rM9e%=fFNuUL9EOs9JK!5i{vc}$hM#ZZ6A>v%GUkbcg-FUV2@G)`!F&St49^q- zyzgTOq=>ufN0H zgat}~6cdGmcnc1Hd8nQ{wVib{cC~WyF(-YP(eS|wq9|0=tKHsW@@I=|%GS8zc7G`I zr>PW!m)R_*CazjglBAjZl}rYVyAd|gE$XmqrS2xdtHt1 zD^U%Ch`xbZ0k0iU95NniUjlBrc!_`$g&B8G^KgrQb^UQ{=XAZN zWK;I=`5uGqN8*Ll4<9a(x+jqG(0q?d8iVzZzQM#=oI{?ir09Q}@$~rTOfh#WGvPu9u2Jwsg|);oT*56Fu{kq`n$%*Qe&5f(=~7Ag0*}11FitMvi6}?GxN&Nr(CiYku!}Dh5R_c(Agq0#zTLp6q?D2A-iSqrayexC=>M#Oho# zqyiW{f>zFl;kr|Q>rq^@uCGaBz4sM@N>4WqakR9DMiMbTJltLPKY(-w7nk1`)p7p& zHc6uYct@-fReo@57qRhpM65JxT_k^7@V)UEnv4ZbFBQ1{m-&HNY}T_=kKzzoUjg%0xFiqE$!dSMyEkdJtB zo)v0qOh}g{4I|LMeh6gGz>uC*^d2{JBbvvI8nKI{hedhZOhYtuzd9L?oMwG3r2zoN zUIXdxbmjTHvqhIoqK)$PZ!BN;3^61G^G@Rd)C+H@=cP&ib@S&IxF(W=%>_6?5-yzV zd#@!BiHU0boBPf%0oI#&Gmi5s$aur7c_H)Q7ObU7oVtb2K3>GfhG30()EUgo6EF;^ zE;Infq6k>5(+M7jqdIedjbyf|avC4exA7uNCxGpfAQ$kd^85jHkAW8DPB)R+|| z`7UvqYMKr4<+i|-)dr)%K{rmwsmR|fIf45BA{_`FXUS6FJUbOQC6-vO$U@FYz6m87 z&FQ31nX!I=PwQM+wiuontZ+&H;9y3iYTAKin~l%w+rATb2+^tsH=#37{V5<@f?Gr9 zPxKV3S3FRWRoov_RQ)I43w(ibjoj`(7hqCt4Th*`*wS^L>;~)25nBJ`G4C6a3o4kw zR{wO7N6IfD4v~oqH=zcS9;nQyNGmm_l4m5NFk&yF4LoA%2RYcRmw-3r>^>PIwyAaP zk9z+08gA|ju?Z(G^2J&^0s=G;sWXUL67*bAqTrKE<|GiHKJY(PIBaEfLvY;94uZ9a zFjloVPx>#;Gbmx;&y%GdM(|%qZr^+h8WzD6Sl3y=Qa1MEAIxMA?}wx>^c;fb;mw!l zSDU?-7~Ui_%!W)T3XZ~h~F8qQ!RTPa)Xf&EN+_^vu=sf*|T@@ zmP`ojTqAeu)-JxSf@L%%7Z+_I&7Fw@mf5Jtw}5E_!jdhv}Hug=TJ`pxX- z9$@^qB7ZnWdiSFPFhSiaV-t|zEy`(7H(SH`QZ;7gYqA@De%qE3IY!*HV{zpDGEHsl z#;K;WoM%vBb1z!N4>`BPwTk>G-F}u}$lykR71s0}*L^k+I!OkwWC)vCV<<$=eQ)oX z+*I<`NM#_)J$aRjMo{tk+toW{O?# z*@0*!Z^ADrwH#pIfB0~c=sfor$4e)Vdn?&(bcDXI{`sQ3-3JaF=qgdq>!xhd@E_Lf z-05rj_4Ue{{YoZ+788Pe_k5p;&-YK?AGd27TXz%aD6_qHu1*Q=_uY*zMQ_Vo`3mOt z!`v+GoLOw2WqvuFUj%aFAX$qUk#Jx#^KAUbKbck-7_yb9fokq8)}_DMv0I-pJ?;7@ zxC%8rc3Q(6X}t$OvGYcA9vdxlbh}PYOEhVOn168@)31`MT^iwLn61%8Zh zFId@pt*GIm_w4-&r!Mc_ZOg(Ta{YaXfEepc)}&j4-R|T8%grfbJAezyl1IDVZEGN- zPl#F{2%~&;@_qsPC+g@BF2R4bcz^8O=gjO@=+i7&Z7c*5jk4civBlkx-Ajn;ii zG`SgJ-0wlDJso%IV@EuW^FDT}jweyj6eQfw8IPdz#WsxgSeNR419ij|;U#y%&{CUKGw^L@o{l^SSZasWWjmI> zmbBoULQm>DFQh(=GurKWSL;R4HmlQ5NL>tli(y)vdLTU=yu4QG@jNB{>=%#)>H@Xi z^p88;JB{vFS3J*GewP90HWn~8fEPsHgOx2paH;|@wlVvoL$JYQoO%p?vOSbq&AxFp z#bT{-J^>GbL+*=~{>689)=}=@-K%D^J@g@+NZ;GdOYPp--}2wv=jdLwLC%WLoNt;( z%VYrBXZQbXfc|x$@|M}6JUW!Gby!^{*LmRt|K5D&KF!#!wo3+e8=U$a5A4c1|t-MIbo%{s~-v-0R}ST1x6w*~zInuY>YBG~ih&$Z76$Eq^_9 z1P>lOI1ZL!FOrsi_rs?7jFoBq_R`r}-8)-=(OYL^sP0A=kwTgt6gKTg(259an9u2e zOlw87O1(Ynb)XiDIJQ6)NmE0t*)76zzkU6xJ^ZG$G=OMvak~fVw&4OpN`z213)t!Q zQ~0t5_bk<}vHlW*ZkMfHqkd4qSaQO%Rj*Dj5wK~aXm8CMuDkunBpXz!z0thmFNOX` zRjk(V*|}ZAiP;4YNi&t^w4!6vV{8lqGZUt>6$8q!W*Hv?`9LGTO{MbWGiyL#pZi?vS@eBCp?jG0N-DIrT@u$F+ zyHEh*dVYBsT_{BH`+p=+WVAt>PDWZ}Cj}_o(Y!a>v1@J8yN>}l$i#KAmoHoQopAV; zFz1pRMoXFx~z;D-K+Omd-1u<~Obir%YeVbS@@Ukk-7 zIb7RB1N)8P3tW72X}8jRxT^ zTI_=>Hzp>6{rXcy?&f~$-4QdrH#c+QSo|mc3|6A^wv9BQzb(uAd`wTsmCD}=Qucbs zJ8KX7elvxeG#8&bY~KscDL5S;^#A%0%7O_#%wl5nOLU!J^GWV&@tNVT<0r>8PWjv) zX&MyhGKQRR-c&Q!V*|<<8mnH|>b$}zHznA^R6SnC&1UL)tMt5xZb#;K+TrD>x5zfl zULpbWaL`m<0$mjI*qR(*cPyT^y*pP;Ge!M^mj$H2^f2ZohQ#tCNnk9p{`v-9u-{sGE`ryUuHwX?4I>x_;Auj0D&5&kZ_oAqn!h*#oDXztHGRz*CEO;JV3es80i z!tD$Ef}5rof%MaVWpZW{VCVnzfp#-J^6Bv&aZi3>zrt^h(|Hu-_OCTxXgA@$19#mK zUKVQ7+cicP4F%kp8Mk{}#aQF~#cgwMe~wuN2!L&fyy!b`6vvt#-lTY?F>Z!?w7hVZ z9`j-gjkF8%FniR#yBKG=*x)AZk1D_IDxY~KZd@U=IHBdo=O>=kITRn`b60t&ID3_V zLmtN9J;Z5BD+l)Wz}9xzb6y0QDWYr(UP23VIr7#ab4k>P!V#9W>GBkyF!d%eG2F3A z%GWD|p8j(CNDWE?eeETD=*KGxB_Kvlj=?D`QK-L5+ca=ary8x$T!*QpMBVnVQq^Kn z;De8R^{RbrJ~sp9l;P~6Ls`<-#>lKqPvYW)(mrm05Tsy$g>s0r&qO?pRRZ8SQJKBwHjpp zsp5zEb9!TGbF&s37zPUPkK zH^2H2$a9~?9mZGilk-2YNA1%9S6k>gznEw1q@sIf%iBERwo$xx@tc>_T5zy#LruXu z*&JOV^g~JU%Ifry+YuLyue7gRf7HvDM$O&Ne(NAX20GbI{0{;K!2_H!2o6cck_8|o z84?<_bpWIz+H=6xk_1AKAy51V)GuE>X0N0iHRS>O;@Z+F!aFGwyK}$$6Aw;iv~_Ln z$#=UN;yIge>DC_brsL>hN>_nObFUXnc473FSq>W1Z<@0ZV5D%r@_LEpnUGv-aMT2E zc-lc8F6oOFpgTmMqO8@yn{-rIR@R99kH3wtHk~|vycYNAKKWk1*H%z?{3}o8&B}tS zE4b*)h{dGkZ&F`fkaM_qF0j10U^4HWwuFy_Hp61o zM?Guq!ySZeJ$Z!uwmQEe#e4fxqO`e6v+49EPyQs^dl$d$+y3jz)>W4tFH_$du}k*X zqsleCXa+x@kn#m?Pk0Hj)xVm8Eh@VgtB_iZ?j)!KAJXH){ho}c`7t$68k;RF+t1_0 zzCU`Eh|f&6LcnpYo<48Re=)puEAY#;qM9~4Gqc-fb5-Zt?!~vc3BX42_u=IJQ}qUi zkMZCH*n8Nl=Kd(GQ#8*6{xUc_dLGx$kqb5N5i5@_F!~N3vT@pUu!}zjgrp9kBp>qf zHN>S9-mN@foQ$YS$`IT+erkPCJ${HDvyH8>%oW#AY6TzP?tastm>D^y0oVMzAZB)EPt+CwoE87F_HWo z)WfN16;O|R30NvLdN(5Y$)nJ*ya;;c-0J#g){05gy`Yupc~K+T?OoPZf>@z<9`qPL zpv02(h?Ff{_tHO#)TA=kX?wIFO1%nY~P*xXWoZnx+1jM#j2r_ zijYVhM~Ei=8&1H^lrlhgD;n;v#wR zrl6Frs0e|l(z$IN=_5k0F(i>`!F)Y%Q;2R$MK#jMU?5iR!HEU~3fYG;r;Ojcef#q_ z6BYzN`s$^K5u4#7E$o@Pe3khtG54WmPpSsV+g`6HUlBb$Mib9tD78-+NF~WGF6raN7TN_*e%B2s;36FGgzo@WibaTZaaE!p z?+h-hYSF#hr@?~j>DTOY-CrhdG;PJZ}b(wQlz_!2Kz!I zk;E&nN;bGCMyId3(tfK8Bu9EtV`tpzTb~5FbgG~3o3;0nQpKb`5g^ZR`U3YYxRn2* zFKmZ(a2H38>zIRc>bU&CqpT2J>_g}T&T9zy&L!XuFJ9Q!Wkf|48P>S+hK8a7eV72p zAR<8j#@#-UZj*Z7SxgLwUZQE<_^@rF$crPTxE0e5+O}A|3T~H!f0H60v;~uwfA`&y zS#nUm&r;F7{#abp*hvazJklQ~Fb3SngtvJ;Pd;Nagrv#OUc3l<{I}Y!V5|^Q*!P@3 zxHlduPj%(xJv2+1|5U>V3Q@!tkFKb^WL)MgJ;K!g(88m5KoPZ}`t<318&=oZxI?VQamKIM|%;MjzWV~?^&PV<=q@&vfuKP#7+TE;t2##}p)FJDLi`j2( z^(?M$#uYsLknh5!PuzgPsD0F~y zi42u9hVJ51YBL0Rh1>D7;TWu`LLXS|)P9v&&`MoLK|;mep0#_grTN0TLbFHmMbaRpl;=9gl%4DWvm~Qhb7W+sTHOx3)9Rn&`(#IbGgB*MZ?H>R zE}VR8HMz#M;-5#}?`GQ&AJeFGecDk_5B(XbkM_Ft-VP^3bYM{z7Ugv@#soM`K<>De z`#oO|lr3Vv)7H(APeb{pM^#YUtH9>0;1x->0JJedFG5dnYKnL`JUo1R?%^3hHv~#e z;HLr9r*=zN;vwl0^;XF0NNyP)?xuUjw(r-l9>;)Xj7h(Jj_BI8yQt+khlV-!@#A{K z8kH%c?l8zJui%Lp18=?;_`-V54dl!ZHmo5j4EdsbH0~i8X;J-qsM;G|&<+?E=>qLb zw77y==;SEZrezJ=JJP3)(Xay}DVO(F{<(ntO8U1%=NXl6*ie|nIQq`~E!Wg7or*IfjJf^6t1l?u0_0Me=3tO${UY=)*moB_ZFvk_C@2H4` zu?_DDvWwZmzgPJh1S#&$T}LNk`(OW=HjjC>_pYd$(XWwXgn-$31?}P+3__aWh(f>0 zmj6%9^x%?v0D5aiJ;Rhd@7{2xmXV+dj&_=EcZ34a5DXHzD20ZTmceG;skl$3w(Do4 zq*F~2l;6Ri?h+$$(C{r0gh;DSh0E0TM+_@ynzB(l{j=@V`l#?7@r7|!85q;(^;B37 z7IAFCpRs|+ic*#~>?{Qr-zy8K!+UI0BbOs20^_49LJZv#tWQx+A%i@wK-$TIW>B2Q znv1qe$9v1-qhw3rg&nu)JG?(ZUTrjD@gh@26RuJH zT_27B`4TQ`8?g8tKK5fR!HOg5>LxW0%~#z^{GYK0UUr%gCLVEe@z@|jO9=)N7j5wY zlJe^9+dCMX>hY^S4VcHpl??~zhZr@%tnMGL9h_%TfH;W6K1spJTKq?JwAw{Nh_*?! z%p+eS!fD8@^if+}Fdt=JIMC3uK+g0;*RK|fpy4Jam}q-w|3Rlc=c!S*6|8+JS?1Pw zqGU$I$;u(mUMCGbWJDyE#Ds{Fmh*EbZGZf|^s_b3-Pi4%l#-GQox-vW5xy72OK+CN zng2Q|z&^0c@!+E@HTPfPY#CbD;|Vh_UfMo}yRZCzo85%3xctc`SSoZzPu}zhFc3C> zg@h=Q)8t-rf%_Q$W|7xR@f${jJ-=E#bfBSLYpOnyX&N@lx|}tO`;%blzrTPJ?EA_5 z9UuSTv#6Y&X~mNd_Sm0ZyPsWKkmCh9(+W;r3FktPxa6?HvOs&DC?xtm^pVsIk<2V%O?D&?nT9W9DhLW zT6e+S82HN|%i=B}2oxHmK4sVyzU*pl5nfch&Mo@ub|eX$}_3%5PgJQq`k&Q`zP3 zEk-mw72b@D43fX!q|HX?B&Ql>j7sY&&5-S|FGle&cOHeD{69L<8|$aqEYs`1%^0)w z)IYk0_rEe%MBa!hSh1VQSl5_B8wJ=tzY!a*)_oQ;YbcO!^_@nCA|9M z+4rkv&=%NcWQ+&7$|fczp@)Wr45HO0liGTP?MFB`8X^|;(2l09J6*yi(g0F{9;4af ze(>`$+Gv+7fi&W*BJF0^iN9a401atf8SA`?uA{ql>;x+D$t#uAKI`SIM7Y z&W6I{6hpY(=?Zup!1VmwtWX|3JyR+@R9m6cW0j+{3+QDGSfDOpGIH*#=gpm)*k{_9 z;gOP)Q>Uv%LoJE#9FwHg8QT{L`N+&HcDW@*)q5TQv31fruwX((C=M?c@^iRY_8ado zewbXdr%jys^{sLem+@jBKhs@5wCiBuhk6ybi`vjCEWM(VC*;mcU5u zs}Czn9cgXJf~{t*&Tc7#R!%Hr~lJN_PuG5tbEx`?g6{5So9ir^+s^X9@ki z#NQ#fJUmvHv+Dc4ahLuK5Jqp3gQN=ks80J(!o)Sp@+hwrX*C<#HFN>5eCP`SA`BG*Z zXOm=AFXW^jCL+s9HFgyu4+_s|9TR)m)-Y>)pBrk-#WMp>%M$vV4p1AtDz#4|e)OvZ|XaKi+(xKvw)_M@!mdYtxtV(ngw*gh%RH zm6vL|AFaNbQlQZ`j9Rl6R_Db6!5a)vnLKpqd-<|C%qx>nX;>J!XRvtN$wE^2Y#SHX_`Yb6W~vWpg?GqK zIBmHn!j~ZHLaY1==0ZQ7GWc}2``TyEqsPw=bX$ziQcTliaRj>eNAv2a8IQ}StyZ2a z??x@Qe7kkwg&?=m_LjtH#*jY^a=n7h?6IS&1pziTW?t&P_&!F=#I9QxUo@vqoIpCn@1ko*tBhu8@=n4zfHQ~YED{sN=rs%srHa`gFF5c z>y|&NUo4&0_@&IqdZ2d!EOB3!2`2Xbd_3UdHtL!Xe!s;urNX2ie{!dkB|GxRl+OV=hAC)6Bj)t8!N8(ujcnI^lN*&D?&9@PtZG%#PseP^|GA^yG}v2HYHW3>mlEgpT}J`}H2-QUYS{+4Fp z>Y<}L9i{4d6M6+p2F|9Wn#XFU?W!!*iZ33}*H!d#?|D@imuNKkq`D)#pJ(3qCC}jb zE&&sAWmax^)d%Ax1{AbIb<5qCGnJylBPYJaCx7YSS}48O{SE6r9_>2W9=tji*T)*( z4Pgy`+cdUu*^bxuXIpdD^E|ZERZlZudHD={IXPbZ#MU>*@NsHdTcFI)<0GD7Zq=iA zGaeT{kM#R{+)K$b)6hGjV^w=gOap_o<*qWgO9f}!OcS^HwQkZJ9ZxRMHh|(JE3Hwk zz(@RZ+jf!p<1!WB_i-AI>X=7VcerLgj0{gxb?zp?p2r!D4fn));l>~FloBvWXFqtde}tP z30ko6@Nt^d#mWcVDbLjgnb+ye$O3~$?x<+iFIipX*EGj2eFGw zTCqK>1xCfgXYg7fAik`dBd6X&L?yBmo&cvAH`)&KgQSz0VyDAn@J((;kb@1@`)1vk z@AOzCJc8jnt?>D7;~8bE+MG0aitNe{8YpYDIce5SYAjh+HSpEWr1 zddNl=kve(#Drv4P!)rO|_f$hqU7Bq$yc8?%eJ8o980UI9UOmr8PGnQuqq;j)!T7@e E122|EdjJ3c literal 0 HcmV?d00001 diff --git a/Sucide & Depression Detection/Images/WordCloud.png b/Sucide & Depression Detection/Images/WordCloud.png new file mode 100644 index 0000000000000000000000000000000000000000..c933b66583514a7d76ef12b95332a408c04c5a37 GIT binary patch literal 203023 zcmagFWmKIp^Dm0CfdU(Ep|}?>?y#}q?heJ>-QC?Cin~K`cX!>myF0h<`JZ#wd+wL} z;aSNflT22g$*jrDFJW>rV#tWNh!7AE$l||*6(As>K@bp-4FI@*Erf_E>i-Hz2L&+@ zMAhW?<9`H|X>NfU7lojcjd&~dC2}nmy5Q59z8hb?k|XikA8%g?d&!>t(Q#>vaL?_Uw-QM1|nck zNTo9VtAs6~(TbY?R|5h@HD0QJ?ElN~ucG6^2uu8b2K=X^{X!2B^uH(gUyQCBJ(z+2 zGvWUXy#5ADAO8PzRCE9TFGP-g zjFE4u52x!%M50h(F!_Ir`=5EW_*$M1GR*pdkp~w+iI{;uyB$K=X1Ogr9p(9lY3o?ktiibofZAfV+ltq8sgqjOVY_n zrB2Oqa5E@$fYQuyC)6{Va)NE5e>HD&gW8Odjm(UAJsM8jaKub}tjF3(GJT|oiaR5D zDxydKHg;hrc1p5!kk*JgOZ=p#O zPa@o(z&yBl58KroQMLZO%TrTdZ+pb@K;Q3Z>KmQoCBPR#G60`TAyom#h372E#qu8h zCkWvBmpm&BMM49`T1;S8iDJ-SLV=F~E~Y?!6;+RVDK%?Cfd}j7!+$iPe?P$9KlU4K z&wp=DHFtXz3kW+S0GI8o==bmR-aeD4j%u%ETA8|e^`j=~u3w8>cwUG7Ik@1|mUTVz zz=IlW+&pP^m)J3zV%09Mo@Hi2KZ0sceme@`6-Si^BW96|c44(xld7XT?N#EdRVc+W4+=It1 zNxrrwoBiyn=%SBXygK}s?YVgD{CfSAKPuJ?kf}iwb7(Wc<2N}=EB(c_do0wZs#(tb zLA!>S(lYV|%`xADOEsvND8THl?t9;#2u$y7;#lHP#yCl@?U>B^m;^N6*~H-~P$W*A z*d$L}nj^wAIQV`%CKy3^O~8|7MP?T0S*D(OO0O5J%}~A8udoa#%q5OcP$z{cc4UlI zcj$EXvD#LE7w8Y{t;Ki;6a^)|rW0nNw!nFFbKJ*UhCFA7U25-6dCP9>YB#%?Udk9b ztA3FsjGXI>89KVvMV>1k8ndXDYX1Pbo@(!{RG)0$Ep0ZkFoUkKx<#bisN($wOyemg)iQ-YM(?bHNo)k;=+&PM_Cg&I5 z93})G#7%TYo{?&=*pUvIYb&d@&rre*#h)#9ZbGUsuok11=#taLIKBwKYX7myapBW1 zeJ^a8-|Zp)FvicHG(bxRwO&fFknRymlJ1=JH5UazhuQb*psq%clwFd87#}8Ciuz-d&bC?n?OJCOU!p zECTb*Bg%|1i8#9j?5HdSw}i+d>TfdgQ&0yZfer>6TgRA&YTiCk5k{ zT*Z|mlg4CLHKn&O%oy>1>G;en&YVs<`s>X_8B_h({Dv(MV&`@WO`Qiha$X3A*jKrz z`VZ42Vy+II?Y|xFObP_w#f`HR8e_y`3uQ%BuWd?Xb=}eo5(F*B8OBF{oKup~^$#xM zk@0drB>q9pww}2O+#$`!_y3G8oRsKe7Tfn|TOj9SPqWl$;94GhgzzU27-QyeQfp#m zYu=zeKAstB<*y5Eg4IZk`{rbWr%{!3_x4mD54WxzziMU8**vsDb-A5+{hld$#1mvd zI)4JDO)oGLvIu$|eLOeYzHCfJSdzmfzL0%p=Tj{=ZSF<5k8VgtT>2FuYWh)cnu?n@ zhT-H~MTyVDX3sftqRRX_gQJt|SXo2Q3gW48c64|>-14T_DxRMPILtrQaja)>|2|u_ zuP)0m)K%6v;wm>M)XG>L*=bOG8xOTK!7wky^{CG|vx1a}l2W9I-*R51g!YIKYb;nX z^;a+ya0ZVNd>A>`aZ?p3H+Z53*2 z8{S43Cp~$VfQCW>5Ca-GD?!LeNRsY&;E;o_cz&s zgzq{Tsl)14>!t3NAeR8euLM!`dRVHxn9Nu-zk^{5Z_s*F(6M`kFWuzRaRaxVM16uTHQ6taS zZSk=mep8dZnE=PPZ{vZzy$aop0>u+F4K==5x-R{~(PoBuc*5R9zx;w;zoljfQuVWi zQdzrNr))69ty*G-$nDR*eZW5x_qNBrHt?KS2m%zma;{0Cm#a+^z^9&bOEb5bPgd6! zTkH_g6E}95^@FUe^NkD7htaxxMDsgY9#+WXqBm;F-62VVQLvUZ3Pk*NJev~+zSQ|+ zCs>PXnrU=zY2a27Wb7r@7AZAPg1Q#BF9RSnQv>4ZXh}Y|^&c@Uohd?Wlm3WEcnT&#s=amx+aHQBT%!J!E1jon=_TiipZ)S2*d{TBC`lw4ARfs_|A+-_Ac)M2a1D7PVR;udn74M-M0X+ z@G;jmb0orPE$uZ%5mB4&F%eaqdd0kLZZ*n9E%*1UrfVu~H7YIl*u^!qwXTOHU6&`c zn{t2E1$y5bnrORLVfzADFmQ9@uKPfNEpU1F5Tr9`d5e30X6?`-BlY9#l9L3%2Wvqp z8ImRP;1)mJjAl8HM@W6@7F@2DPG^PP!S~yvG8?Ao5?-Qt2%u88BNRw1*wczW5AMzT z?Oe3P0NYuH*IpbTjdrA`59s~Uz&leKm_A^bH+0b?4{ofSZ!gVOpxqNN^*08baZi)* zd`uH%eQYyIXCvEr41EXC=*VRfL{5)ytv;QZa?sMxEC$Xd^x7IrEdRblH&&y9>cA#g z|H(8R1c0_eeK6Fc4sZf6Ndb7Nv8~s!`JXGLNsfG}vFnH;O}EnNBT^{lgjkYsx9;LI z3121j+4?}rnXtukCPDQhWT==5nEcKXYKZQ+;2kiD;~Ml19<)9vP@qa|$7I(WWA$nI z_(%jhoo7ckQR?a0uu7nBES$M>2MJ&yQ%GB@yu8{zo9m9Do=eJ0+-cM8m1(xEl`~q1+i-|>B@@5_gZgA?!Cn_>KSRWcg3V&G!f zX6ni)M7v6`8DdyxcHbMHCm!c}y{3-sn(1vle1@{pR4WJLz@?g-^fQcVpg2jwzI3|Y zhS$#v)7HrrfBSVmx@Ms*j625J+}7pR+gk1EtK*+k-XJ&sbme+EpSi~QqYGkgRN8!M zAy#$Yix=L#TK3^ft@WPP=lZ1{d{P&Uqv3R3ygd!JRO4mk;$W1@fj0T7tYKjzQHCt# zOVq~*&l8&#jSGDe`{b$9!(F>6{2M$Tr0539M%xM$o|%XuUdigi2tI8U?iF=e91IRQX2 z+R5H`pjLVY-TX`*iwp`M{KM;)2b&ssJi;zFW>ht`)U@>8?$DCrwb8@+8;CG&nN6R4 zES_b?20;Q;_H0jn9hvK2DDXM;o1l#6&U~i)>o8iUn|pzE9EJ6ag`#X?0tWhqEqo(a zfnY&(9;S&*v^>i6)IYy0Bd2&p^|1qcXXFt@f~miM%h&rI3nCHF&UM$t0Y|4F_q0kt z9%dtEVB6#;J3D6R>=i6r53Uv2?EV?2jb3u z2VahfBlGImvtMmx#{_|8ND?V*>LmjUEZvhqq)dp&LsZkp+s;GukSM~?{X?WjkTA3a ziVER#!c-z7zceJc9(FymDN+(kx%hM}64|uY2A_NG?;T^$=;=*7E$Q^`-He>)IxM)= z-yO0<*)WTS9T;Y4aZCXy&%QNK-DD`~nO~@plfa9#Ios_qD-PJ~L_5D$6rd*F>?65Q zh)06i!n#5j{VQDH5u?4!w=bqORp8h9B?`;_%}G;OTnoLS@Ptm;a~oXuSm z6$*9}Ik*MF6iuk`P!x@eT)!T{*)-21!wAXwp)1t{3RCnt`Q}8d8Jc=b^HGp%HU3Bd zqS(d&!-=FIT6p+&T>z23x1q_&Yhso{a=rK@{J=~Ia6v~%Y(O_+Tz?BdYB^*V!_4VR zRi$@zHog&I4jPbn08CwpiIn6SX&UhCMl`6uW>pzv?6Xm({{pGpAY<-=zh2Z9W ztHrbZ+8Jx!P_Zj(wUv%OC_>1Li#@gogR+8!Q=QTuZZgUtnmqCORU$7U2m_#}I3hWi zJ%CozEFVx1Ft(qV*r*`D+4mS`Dck^>)-b>;%GhrBiJ?tB&xx;Y;v`h79TNJUp{hISi&` zKoQp;w8`D5^4YE2`<+K*_mY`e4a_coj7zA_i5e)35;Wn-$fRMwE;#S;a;;u1uBxw< zl~T9z^nwRBPXLE^hGfTF>3Ju2xjLI>wN^$tY;acER`>a?Dqg(pWHPxi4(uaFklOy5 zfG3tl?MzQ;t+9`qZAeUf7&tknV!8gho&|W;+N6blhN!!yDA+o8qmptpQ*a~~QidgB~t8xETE38`lt87j6d4E9TBw0-c-1qOxPTQO{2 z{}&{K)Pd0-l`(xo(2m&YXY_IZ z18~?san%J(jr9MF%+b6OX9_F}9)^MQFuIBkz2{?2ISPQEQ5P{SG>*bA7_7C`s!6jMdACbOYDuLWFrn;VRQ;iS;w^r1+Bhe(p{+s> zt|t+&(uXYW*pR?ut0C$?+@pjrl9%pRaENx6RzSBne#b zwY>_36jHmk)z;x*YB^$}Y`ro)hETX+1JrTb6Su)@`>VjZm)FYq!N=Hxp& zqSyd4B2wPvQ+n^(j&l4k1RTrIjn9lEd?@1?(%U0x|88*Ad@sqH!{;MR!flR?;qRLc z0G$(<$>z5D2|!0hm1pC*IEuEP8w!prolr0!{ooLVXRPsP(YkS&M)BLOZK;t1q7#R} zQe`4_iqw|=0Z0Rbukgn=SDV>GY-if&YxbL7lwiUp8jPF`Pix z*zA3o1z-?_siWK@lSA|k=7`LgfE){`U7Mp8GJ&`)j}G*09)l`ur6d#DANb}^6Zl>Z4q?CVR_ z%z6uaZ`)_H|{vVuPQQEs9*RqSWc--2@NLEC== zn`i8n(=wHW9oY0-g@MuDlX^4oBSgi7zmbVD9%tZtJ!bAgq)AIAW0pz`00cZz6#~|| z^ZvHjY1&<-8FAPrsG8CoX|%#)sQd7YMMoaZX{pKX=Cbj#4f^Y;t7M9rdJ z#X{DUbdVu3^E})j8E4ldKnq1aU55doNDl z2^$8~5(ehY!(wmhIU57m>G+?$Ijbf|{LyoUJT{c$RWCNy$h*#Xv+T20Fq=6Gc4-A# zo|@@zy0$5vvmBrgM~aZ`ZH6g5OfmrJt)2$K1YQz12#07y`iq3iy-RzwgN5RJlt^my z-h@CtVc%N09jImt+AD$79`CntD8-gAU}3Q<&mgqNBq*FfmCr{5*!$|M7r+Gf!oWl6 zJ%p7wON+vT2-7-@bo)L!iU2AU`@e$o|3bN(4%867FpRT>7n2_oc+WIYd`xW?o6<0-i4n+@hP-_qO~xKRzmiI&?g}jk)ni z-<;aE{Q}|7+rE2F|NO2fOL;tDV?VPrLcFS(6vN%^?L6MBhdg1qa?s)Sj zh|R4UeS-Lg&dgNd44YkKju~?d=GfpY^$+ee%7eSJWP@Oiv`EYBSINwT+sugX@m8ij zxzw9yZ5mB$nRw(`Qe|mQQGIh@oO2R?HV8gVjK(%L`_=Kol43n`sm`TPW8)*WD}-mQ zr%pQ?s+Ftah#N1|fpAUFJ9p=>2Mky zCmf(68+Y@yPE4wY*<6cfH@I+H+a~Vv=UiD7s`sV5I1(f>f(h0A2be$h{FeU_?_Zwa zz3Zbv%i*Nl{45S3xKKNe>1+MH2_KQMdH1|(C;3uzmpjNTKqH;{Dw%t))pw@rnSJK0 zKRSy<%!h&1XNzxz|N66&%7|hG*)Re$g4Ouqzmzqijg~shmNE{~nK9pt4q7VKm=TrTc*<9`f;0 z9&f&vT}YI}N`L9}ek9`fb;v3d+xS9K&JgSaodm<>1Idb10|>^H$ZiCegvG!bdsMk;$8qS9?u+zHUZYT^#M- zqmB?d4Yn!Ze(Hp1(X-q?C0b`=(d^(Ig-2Y%I|PGH8_|z0JCzkoHOsHjZk|?;M)7>Y z{=VAibSF1y&yP2(W}-9Z6})!0U&2rET4x&ZZQZ7b5B4|5w0t+7qaGzAS?`nz)BShK zw@LpHW7rO)rz=Ry^>`!X%ZaggQwRryP*4*{UBBNl?^&dVXTei(sG<};sD|x70Xhie zM&XMvr&aQ^7Og%{vBvJv{0UKL(WXDT81LzD!*lX`mgNz+rY}}$S8*WHiA;UFoexFR z#iM3vdc+36Vh$0P5UrK>`GUO~s%Rf{P+ublr)GY=(cZf{k2+LSU?YMY%pk?VeGPnH$(X)~sVMd0w64AJb{bY@v_3-r~XgE=ng7nxwGE@#wL zr@;prh;3&PtSx{}%s%kea#?+8#qJ4z1?cBOFA_(x@fp$qLo`9El4WM9Ck;!uxdaSK zajK{#^tVdcks7-CoW2M|nIV#B_~NxaFc%%Go)Y$Lfu?o<3JBrWco8+ztxkcGQz#Hv z=&N5%pf5Dh_+~sUjNq^@g7$LmnaNK3GF(e#f)#KG@tXTUijgMpDgeAqU@Ig2&aFi_ z{aS-e@#NA%c-D8x{6-5nMm?w+pv7Xc=7l_M=w7d62@rXpPWGpDAD=(_Rjc##N5!r}pDYZb!yifAKbKk<+VLuZ5iyzfhD1p6$3E&jjifpuS# zd=>v(4kM?l7TFCNa%Vb$nukEv)j?;`TzPYaC4PoAS=)Tjv@kFv8Cg?(hgVMrlL2AN zbz3IFpR^bnU#HuHYZRuG8e^uD*?;J!+ILG`u88qlSZa4t-J!Xko$=!!2O#qkZW&w| zg_BWIr+$qS{j5Ut475J~RG-PtOmi1)_<4oa3vyYlBKY`~D*xB6Pptp*Xn9$vu4{9F z+RR2T=MRIX%`QgRnVa8Bt|^bYx5MT3dF#}{^~ih1fKDX0 z=aSJdhRUfsF7EQ`lV%R648@{??idRRS%!YJOI))Do$%Widonqk8=Tvq(%touF&mGi zQ?RIpOqjQy?{>Qoc=fzXnH^_U+!f|&g-89<+iwRjp*yX#Fe`R+5986cO?p?*c#HP) z`enf0E6{zt4G^(gRENNY?zzknUeec?rjdnb`qPaXrxqC(9RLJ=-D+lz=b(9IZF;Fn zV98dNhs@IkeFlRenwz`@X2pg+a{d(h=`_hSFmGJo-|gJF?71i9X2!%mAgn6;;63`T zWDb@CO?t~dqy7{>yNznj2;2rlBd_8Ecym7wPVGSoch7y8US{4L`g&%3p!LohR`<^v zRHN6osn4Q6{_W1KyX^cwa1X;i<@mm@iv0@wi9M;occT;b1o!1F{v_gN=4!Hy9EyPP ze&BU~y|fk~%zEi8l*679}>P1 zL{Re*-7Cc^b~0ZubxUfK)v82Ie>(eZB{(c4>n6V^%-rT*_kdilDtn+xRU*=*+o}%V zF%Q7s@!vnfO#^k%D!=3pmIAnC%4PGrj!W!hvtR$5g_J^ihSYS^y!lZW9zJ^qpj*-F zW(3|E$_omo&LJV1w>{q%p=ro=Ylv85<;q}(Ya>_u5e1xo}K8|Zptya zYQ8)T((YNQ$OqLBs8J;49`%ca8%EVYwQsV59;*IDyC_jxno5oY-0cK9YV_Vp($fZa z@@s{vV+T&L@Z87kPaGhU%maZNBOdR{H=I|4YvftpTORRngi6e#sZj&GPnRtSumL=6 zF2mjy1>y67BJYd^8EGiuT|i!YrH8%;zMF=0G~F+1bA62tZ8u@dtrN>_9R8@h6vub$ zO7r9U`R6!3AcZGY;!g#P1beWeuD5`3Lb>2>OsBTO_z|?3VUhO^vu=KWY5ff!4)ohM zMy&G=`R5vg%5N1(2G_1+KXK-l&GGprPvufizcyc+z*F%uN!RvOOY~Pr)!b=F4%BEx zO|yaTXLCTOx9fZCkp-bIJRZLx=b7Vh`c2?%=Q#&dONQzr1~G7sJnw4GmRkN^>rAGr zbB%RD`X4i0SI>AI;9LFpd)`-&6Yt1W4)OXagT2w_A~Jm3Or1T8iOO3XU1Ly_y_z9>qwc_&h3xb?dM zx{l%fX*ZSf{x}^hhWa{W8Z~{STRvPjO;$)*wC7G`QOW;FY^}S_lHau?zY8pl93raZ z&%j~5h`?B z8sX>7AdDbcp{q5D4sYiUx@=^P1uVzyjt8B}_e7q*w7uOt&&Rc-{?g^De(7!YQKS3y z-Y6?0@e=F;LpA-ouo`zQ2sdvHtRsup#ASBm89tk)ezQOmHG3rxpY5e)7PpAFU>=ye z?LMT*vuCGS($*(h%A!S}#Oczeg#VP#rNL1I$L0N?BOda(pApaHq0!1;R($YkeZV5X z3|90S>Q_B{C9ZA}r5ejpBiEuP{^bjmu9~()iOptO7Pt5qMR#A^rA|$2n|K&8q1`cQ z>b;Qj*G8n)X7W>;;EJ0wE-XB5KN$U>s+x&;(@1C(kBhNS6@XuS9TaB3xXO!=goJC% zbku?sW%b^n4^nhu9iV>F0#WM1vosy%ULb>FD?B^5Hd)+qvP!3l&*jD-rzW8J+~~bf z_gl6IQsC8V3Hm#3-RN3;rU8%JdXj;TZ*DQLY&d%ziKyCaWj?{9 zeimj2Lia+5WtTHPCmBJHR36WphtU-%NY!$@2NzNsSq~OJ{_x4*w*kmbpZIr;w{f6) zAQQf*$*DET=lpO>E$sA29BTg=);^_OzU;PF`^+x~J#?62;f0!x`{Dv8xwyM-6Lp>> z@Zp6;5)vBQ0v7h zGr~_&s>>y>{iOX`zBW~uZypsJyl_FLnj2+<7=&4#8zaY6MWzZP-@PV`Nu_BB-P&CK zjb}LuYLyR**}^pOywV&$Zm*D-`iX1P<8EJ-%<0Z?46#c*d0ij=bR#0|YyQ@*#Olz= z$P=>Vb|nF7VeoUym&}UU#IJI-hba;J7J+b2@^$FP$R#dK;=eh6^M4rc2%-A{e1qZQ z7J`L_%*)WYrr!x6|V4xsAJ>08I(jc*9$3}BT@c^-G!GWn z2z_7rZ2uFI{UP=bt#@!2+qXm}Pwk``AG1_0I^1Z1$NnN?{%p4YD0jT2vnpWCV90=< z_Y}sA={lD6+d`Q+PGA6+_#^6a&|;5y5;BlL-pb}F#9X>Q`1NA@=W1pCtn#$Skd6v2 zaYfkAdo0$q=C6zaXGa&YSYUj-L=gy23S76g ztJo2n*Hg8s1y@~;j2`Z}8$Us00ZO^h#;E9onmxGr%e6i8An2}+OgK3TNBN;dF_FIH zLFN!mDSL=$8dw2ri!H`Wu06&_xwv*Qy*|h#>^o$%apl4LkkF#0E?JuF;`6K!tKW(6 z4qm26pYlF0tBo}KJhEnC*ad{3a?fsa`Q5J3jfCQ~yioZn7BxoA0>22nl31WKrSm_L z+S9%;M?xwJ6(|HVmB4~xc*q}tyEgzoeECD)qDQLA(vE6L1#OpE#W{2`{WD!Yn$k_( zN>RW$JjuqtL{Q1c1^1r!J+A~GQ}D+T!@!}KMIZ_<78<0fxfC_T#orh(st|k7#JMv< z@Kv8psr`L+j;hm9|27A{8@&V_KHChVn-|Z;?!jsQ*yt<;Pr`^<+9G1(4cf_Z-3+gb z+fh&-g8K(K>%;rBhth}Ff#i?RxRB?gS6zvE?kjGI`n9_zykVCKXWaL@$Ch?k`A3_7 z*^LjKzB{!4M$ra_K`B+Do&1%U_#!~MXRU;bp!alzO2y)-!Dr)jJ0&sSGKcCV!NQrX zJqpL-VkUIfCnd;+TR6@9kF*`KB8e)dOu{E8^M-hs3>6~ecF(N4=2!q%8GHE0yvZp% zCQs~bwg$xH#IH!FX&cJ(Wc-qQIv!QAV$(O^&Hs>I8vBx|^SrvFk{-JyR5nSQ3Zm}* zcd8z=U+d!uydQGN>4Zx!^TeCDwD!zS4$RF*CiYgoTiZY3n(f>PBI+pA&p*3r|AY8( z@TV_CI;@HLDDa)Ci<`>(j!c#Rl3q@3{Q0GlPDK-*vEaDl;YXT!5Sz~U^c^O70d5q~ zSP}Ax79fYx>V^8TWryJMl#>i4C%ZMU&3l{*y3TtTVZlZTdfo7q;WP7P z+zUmHCMuS#!}fARWP{@V_TlMy9beqPD=l7{XsD(nWIQ{0+ zrHPUg>IibSc9FAbj$*qjsU1FQai5m;;gcmz3F+ilN>(YzQQ|Aj?s(rjvnaS`-~pA^ zrma6hcfFd4vr0w%Ux;lf%v^IpKz-{DUVr|@aGV$w%kK&*9`3PmL?dTO3!kqi+=jZF zAC|NfDBh{$FnQ-aoqJok)AcURz=YK0!N~l^&TPx5s9P$HKgVWiN&RPE4Z4blhN$~W z5inAiXGt*<(-7k;!^U_W*wPg0&tRR*=pb~Nr;S~ZSS#2lS1jqxNW)>*e*MF}0i@+D z?PK#X0x-Yx5NCelkzvM(Q6RBg)RZFs>tchWv2Hyh5h}#O{PaE)7>WoFDj-EV>7fio zKt>BalInpc%@e zwa6ikqJ_EhvBXVglQuo8#~U}^P?uTb;uQWM&0>4>=cd6b0umc(k`6BjSKM1Scaouf z$iVh;=Vy=b9T!r5?=`EPP0uAS$rt{I#P-X1KuQGFW3{+9zi4Z$cN|#oD2|fq=1ubE z+tTcnR)=g$qKR8&&xHk%Np6s5$J+2xb5y%_#;`nCicR%6%@%{=UvPD!;Y$w9Jb`w! zKzZ*(l5jLMe8_CC!Y{3_7;Npw=EUo#y$snqd<^8+DCv?S3;wbHvYh#OnXRsoJIsGc znKz#dY&y0)I2F!tq2EFQ%3asfO@O-dG$ygIY>I%pxf+P_%OtzaCEk^Y% z`ya47++eORT(J)`>wEmAlwR%zF=he>%L71gBC^9>3jnVR>}49KEDY)K-2j4!z6|Ja z${b%Kd}v!)(UB!GD5ebcAJ=2@-i4$0iVpq;(o2zHx^*Mp(s58wGJKH$=-8I#Wh}wO zH`mv78<rn*=wyycbUkfcr-;VC z%z4JRerf<__d%PGj}btX1QUi~6OI)N4)!~M?vy|`cE*rw)d7k(iW=haNbGcdv2G}n_s&}hF0?qA&;?EkF+lm8 zwBekti-;PI4k!Bqm-XSYa1oz<_9`#rz%%gT!>v#3D1juE$k9apNOLBD+)$GX3J%vp zfJ@1GzA>sTnlB||V(0xP>zbDf0y0N-O)R6O9YPw4WWjBnN3Z7Z$f?;ru?xN%2W$;o zLf@c&&A?q+dmfM_oXrV9xW6D{pE{}T6Dt{psux`+(b8&l4uh}ilQt%NdSiqv104ZYx6 z)>Gv{GDp>KI)caka-+0Gtn&GfPe`bm+DE?9{NHe|!0ebgRDzfpoKBT>F~9(ZwzE&Q zG>ODH`20_>uYLS-}rEf>3&GLrbpm#54 zd(P96av`tF$8yZ)?Jdu4Tv{skI4WA(y(C4B%>8rqzbx^|JbqnpSN-)r_@nJVv@RI6 z#E%b7l`#Kyv}MlE>lL1BP7}_{yA*SEAIf_b);fNByyU?i9-n;*{6QnX$!EtULTdPp z%ZEv1NQPoS)u`0SsIt}}Bj8mSX~VN4LLVu(9zeqi)CTmz4PgLb&R7Zs~|H~B#$&)0hY^gjf~Ao$y|nj*0c*%#_%B+*hTiA zP*jN`JsQ7oH3C#nYC<7euj8#!$#*smUV0J59A+7I)pMm;h!T=>@lU@bWwVsBGPVrm zRyM9k;9G?Qqlti)8^4RKGR}W4PaGykIj@cdVT0N)9`YGZRx*|(|p~#ySNoS;ev}z!Y|pHoFGm~wn}6C!DxQz6{+x| z$#$|e+|^`Mwuuy^hNkUB=q~CtHE{7)xL~FBm!7yoYIMz(A0G-cYG8%{7rmte7jX>~ z27JZ(dgeSN+Le^-?Mh59ZFYt3B}uxdHDo`1zW7q0BC=x+pOPnCn*B5%nU%D8M`1K} zp^+R&2E%76>jn`eIqzHV-(TI+sXwZV;pe%ek0aBW)Fq#6I^vOt*$KYBG5#b@0>>36 zItxrhCJz2{8(HSLD@a2WD-j^J#h#I{2@`u7u!|hdJuYKe*a}HPx0&-tq;*}BdJ7Jf z3{yhxOVD~PnQ9t8>1Zq3_BU74mDP2=lsZO=QtB3_Z9QwXX0ZOntw@88lA9Xx5o=vBd}9N?P4;BbE3M1W+n^6e2rx#LXW%yVQvl_stdgi|0k_oREhz z$C1!wWpS`?5Y&AcVF}h`rsC4;JHINLDWTU-nUvr$*+?!tje##!`fFJ2%9Iww6h(tn zOuVAC@{Pvx3XsI`(AM~Nr`nIMMps_%3uR|t`|_~{#;PgIJbwy(UX6Ui{VFxrBQ$zq zTzHMdFE5Lyo(t>!d>G}_d2==11CzyBy}1y=_t*BEbgQ$u%=5J#tXTBt>)8@|f#J8A z;J5p(@##IsVDu67mr>{lsjW&#qIr<1p9&lZ<*H~csp*e~KmwxYE;rhL(0W}@A=Kf6 z#%{E^50`6|Cr{NM2GH-zUS%;1!S=YsT~j3fybT>Cuie|K%ACT#M+UZ5^R<4@lSKvX z6f?-PC6h%6B?Va;RCz&;m)QOH14Bl6-!fJZE_V69XFtBC0hZSMg(P!3FvvtCn8slP z@ZARF5z-TzT&GFGgzb)6vr388s;ZRCJrLiq{1R)++BZs#p=L)?O*8g5_y{b%LcaV9 zRK#5FFq0}tr1YdGZ#O4_1s6}3`8?JkAsu04q6u#Kgp0;E5LLR;>8k@T=@!DJeN5>h};ov@B2_^q7s!2xyRUSfuD?cV& zLZ^Tb^(-*{U!TW(R4B~dC`@nx?lS|IN-7iVZ@K}fWo#ZNV`c5hP*n(yZ;L*0j^s@D zIgb%xZcpNdm32@`WoR!ShD2_p5fk~?CHPh(5i<(gxuKbKNuk2jW>Vd%N~=b>LTuR zx9MkW&WFvYJ1id=jo&}*F*R(s9G%R7BNJhVsvP=vSzL(IWtm-oVj=r{Sjl}-ejhXO zH;`{v8MRLR(Cz?>UwReCTzDm<@L0o8{@9#aL==mmbZZy7dq6GgoU=X51HU$kzKozI z5)>!R0R&Hxx+sLh+8EW+)+EMWFJtFBv>SbJNk4C_{@GiF%ga)y11!H>Kil~j9NIsXAE#wK#ujUi}UPE~LhhW>T*$Mg%^O40G)t!qq z+K#7v#hf%-ampKhMi57rR-bvwZTYLlHvvfqbM6li2R>h(2m`Vf`I&PJ$3s982p+xr z%!kz%98fs*=7bC2HP9wEI@=hPSw)grtfd!vy%HxV4<%&M#`GVZX;yk5w+{29yf_N+; zC0JRHfb^H&o!!67_g`*@8JSy1x}?guS0>tLy(cV4cizG}0X(gI{KOZ z^!f&ylz>go5HU}Ks@7Q%)id??mkz6Xp(?Mjd+WdV!L6)Qh<>4VDEurqKQI^v?Vj*P zPHrSRs$sbB@bI>ldb(WZ?R<0!q$(k8WnY~+gC5uS(WdiI7&)9-G?*8{FS{I+eVpBnk>eU1P$H zQjp-6{F04nSn)3L=KUBXdNv2^U>pv%g?jf25$<8Zp%s_Gj^EgoxubwxKs3H53L?NZ zb07f_1ROueHNX}Og-6K6cC?SEvnLH0z|kjx)*zY>C7NfLheCzPW34ep0FOyvp?iy_*7+8WdwzG$snH-7hCVQm#0u2YfulEuQbxP7zV0LfUU^+5 zUk`Zx5dkVZ*D3wN)e;OFeSU|3$F3;4-h7LQp64HQS=!AoV2)7${(oc?&rll#`>2ANI|QJd0}k8;Cja*T)L)_ux{jswlZ`V6a5b!NW*?*sR`6P0#+$ zqbTw|`KGn&RYi!Kk&p|t7??U-L|mweEl$dt%sWoF2E@)wS#DeU+GZ(5K^J5P+R;-D8%>q1N)_5IKG zwa|^*Pgj*$35XuVPO0aTGale+-oQ^r%qU5-Yp%9*H@=y2%15=T9(Vc_euZ0Uqm+3O?H@`$u_FCN7Hg&I zyUMJX>6Bki?A^}FkTCG*qD#DKC>TYNA}e=s(yPwvuJALq0Vs`E9Xk>mHV1FCc5IW} zMD&x5=)*fmdO_m_aGciiQKdp5`WCy#}Ed^R9PwqVqPDO{kp`}S7w zX?^zpP<57VZFS+g#R(qV-5rWM1Z#13cX!tSfl?^$?heJ>X>lzs#i3X!UK~!|b3W{| ze?hL~nwe|OdEa9^>O}9ig!FNu+P{2D@W^_Naj0u|on2=hTVS*w0m>|)Uk2Z=-aq~K z45OsxK-WMf-EtRL6WOrNu-$ec__~)W=r2dWk0DlF9PL1L(K)BF^H+rKfRCmZR+{pi z^pmIKb{aUu)j0H$G}(d5z{p1U<9PEN*D{hGknkr{8%$O!fDRYL`5CQ#QOw zv3paGEq2vgSNqk*fUC#P^4cO;9ohPc-h$&u=3_9YAr=9-7A14Pw6`ikF6Ff)ZpQob zZzfrhHmA&ubF$3jio2vze6la_GqW}BGZf50KIu0=+%$!`p@G$r4FGoBm_^`2uw)_l zUm=cBPwBU-e^{84e!F8o*S`s9&B!xR_TPQRWrF};r#^GR8ZjDtu^=MWz7M*!iYd4K z;C0H@ugO6fV6Jk0fl!3{IJa#}+uj}T(*G%Lqwq=!q1hwRJ1c4ddQk~E@if&==?%ch zPlKBf*g7G6uaj=W)(BqV*B<@KQOTsnQ5@7i@I9fXjm#S`?43NBFg#=(euF0|1p95h zVy33m1-N>H$WG4p*V9yeHe?!~W4&>jyEEwugYv?_&o$M%`dmG+|}5 z6O`z3K!nEBZ8)Hc{ou7&_Ct#x%OZzB&v*wAgqLX(hLU=7c|Z52NA5yQ74{_Rw;}jm zOkbv8wu6zLNc%XiBTIp&U6!jDIJX|&rrGFUW*Vqp zv%Og=+$wecYaJ8{%9vXwqcTKWHB$>V!D*9!AD8(^lq;;dkYbyfIKhU< z?R@xD8`c%!tUqdrf7A70C*)M-86Knv29ae4^ zQ9EdMhDwVt$MAkTX&CauN|*?VqjO05qG)`W%TWOYZc771vEHZN5mYc~(29KS?}Qitb2psK z;`hZg|H1%4;9;&JcQ{3_Ni^{DYRh20VfW~C#9@#X4!6Q6%|WHB6@SvE;jHD?)E=HC zPwjY5cxH2w6wrm%-LyArNGygzM3wO#5_}lMdq!0sZ0&@MRtaf24l6(J;D({{exI zMu7ni0rUjhyt_{!x_`Y<4YS$tSDG!r4J5J5QWR2}l9}US^hK=l)TJW)A8d$U4VFlX z&hLY?G-{DTuhv}0j*2l5E{6VeqEobHR&hF)daz!Tod))1^Vob@LQabg7NLv{!znBv zjj)uoaeO4Ze%?#a#lJ)kG0h0tYJYv*`>cstf!wFcvn!R2nty1h4~n%KICr9}uZZLK z+LVdABaQ3zl*~W5Nhg8y2AtM-U?r_Gxw4B%?A*Yl6y(Rle?T(@1Y_B9isu?iFgMF6 ze<5J;aSG|SgBXysjr-74wL~VZ0wVP*QRFzj?7$Dop-i2w#o<75vYvZjMzp`el^}M< zFTW=H<^*0OH|eVEXlmP}oA}+Od}tUIgQZBwE^8fGYp?8l zh;uw``i1}zZrU(P_N~Gr(3}cHDw$c$Ia~VUq&D4Pr>yQFDSO8HIc*Ve-mN36A^P%1 zVzRUS{8fET-184CLG)0|(ZFZx=dqiDMjE`F9=aq(v_=9F z>eoIW=a7n0a`tEXw5HaN@aVfib}wAYo_$EP6Sq|s&NKyt5vPA8)GuaE?>_niB;LL{ zsRsR4Ox8iY?=wDF?@8)eF&3J2np^~EBM0gR&TLog3y3o?Bb1H}O7q@KBU|%`l~1_e zkA@=GYcEGDU?gZ9KPQa8C0?ma4ly~3QRI}~cf0lS>h zLr`rKf98VVz5ZCm^gY})pwDyJMMxsV5=vtIZl3=)BefzVMp=kl(f=<ALa^ z$-N5U7P`N)S&`d>p+0HUSToBWTWO<7 zj4E~&^QRz0O+U@9HfT}s0CF&rD~H6xfR-6%fC+Cy(Q{3jB~Ls@QbssR^vw4=e%SR< za9{TH)xeVO`SAXf8&7ljkFjhtQ5C71_znRK+~axVuta(cA{=;f3+x=SWZK z$1eMWqm-b6Nsc_<$VXi#iFco^HrUj_R7aOch<%7$B5QwbZBKj=;%MZoFy)h&dAM}a zjSsQ4Js%7Pd*v37-cu(`s3NDehBDMz-ppc0gDFGgc@gH zgb`3bRW(yA>`Cwi^M_Ig#YbJZsh$WR#qYr})68eqYuQH5jNLTLt==PVcLyrLinX~} zm|fs+y;Vm4fm!%dTe;?fP?bD+L8Q@9i&Ifxc_A?zDWF6VI^2E?%*>M#0W^ z>lj&tb-mB)%lpM;7DnC^PATqcd|_0~Y4{xvXeExo zf;4U*0Gywh8BtA&Sy7Pt4Se+{R+O|arAJfJKP78bHl%so?Xo+3L?b6J9UdZyfF-pS zsc*z*?aQC9mdgM6taY-C7kC(u5aecJFlXNpx1Ci?Eu>4ONhQqBTJePmVFof`M^9qRd>yA>42Ij3^$jCnJT*si%1hB;?aD zmi!WuoujeuY9Ak=3y9OJJey%f4Qp*3q>)CLIyww9 zD3d{$if=qTrzpjWV6lf58W30oPK;~<)M*!jpzuxvLxpzdsO`b1$0;xy$IXd`R@{=u zD5OQFieNx&@SbyMhybCE3+-i&BFun^HByO=LrFCXk3PUfus!eaVgTAnTIiT|Rhv<~ z%Hr<2o%elYend3;_r!;ZZWVTRADsH@o16<<44O#Bh)E)%AaYc`M)(5G&z)__I%Sc{ zGIvt{O|(3=-Lnt;-b&*4K1Q7O)H7IO%=h3(?F`ylszfQn1s4ZYr*vC(1PB1YO8B{8 z_bB-1YPTpAI`=hBY#ZFuT8G(HFK3C4E&xC^jbS)^66{%z3R9fjHc!-XOD9lDEX;Q) zw&voX0ZVxs{1im~>!a|P^?|;qxGW*3xII9O!olI-#jEBBT^@)|rMP>UI5B(N@rM&~ zQU2PoCbT=+p>w=@M9=;?jW=$i>qz{QYVVV`czYWEd53Nr{XErn^Y#V`NHqP zSEkZl9fxovkn;PQmph|gQ=}y>b9j!|t%6V4S>q{amJ;+;2wH-{@gr~g?D2+9yGz4!ZWlRp-WO|9 z@oy&D3?1*}PbvNBufVYn3zO(XrSb#TPd5t6`>0`Sw!EfpTW>p@U*`}ixDsmU`!4Gz z#$TW}&oDw|R6N`0Y!6RAx5+9}>%!)B3K|Pp(qGi>5Qeb-MRMkrR#~RCc1CtsBxvT7 zW|QnJ>g(~0jpC0ozRSf?i$DAo(yXxn#!1Wqlq@dT(@d4D1@pzPTpH$=I&|lpT4@XNOzi{a1BaV z`aCcXW}rPJnom4saWZQzX;{x}6^$sj$rWUd(94&1oF5oba=2);JCwWgXJn8(->Vn; z3M~0L(}ogavD_Rx3Lpub;SnXCO4(aWhD8bUaTt8nXwYx?hlzzSiC(P;8zG7x6}Vd~ zp){ls6l|wBpwVDx{2^tUQ{h~XhRaUpXG4tcM}$w{3$(u#2FEXpgScMer2DV<2#e!m z@(3Dwq7r;$daN6hypbL9g%NF{N=pWz zNCTL5Df|-#9}X5rvWHD63ZViS@^g%7m}Z}sZ!iMchWXg-LHMHrl9YO(J#X=Q9fq3i z>DaylKBX8I+Uey6D~7g-?GT}Url{(|AZ*N!5t$7=47E4Si0>d`HqWyGK8BPcj6pq@|(DQFV#-D0Fn*g&2R%5q`bZq$-6W*m_0<9g-?O!L#pA8sI5IKg>lx|?P()>`#vEA~6AzH!JkR;yR|~an zzLB8-sxGSA!)H}YBq7s=c9fz4`;uaGsga{F+OdT{Jv_dYZ|>$;4`)7M7S& z3P}v~qmE6uZr7qJ1iW~fiMJ7%$yK-Jt4dI2m;qW5u6Iw4aBsg?>n`*Od`!7-3c3-=nfdQVcL?=m}KZQptlX0o+S)gV%YY5Qt2D3%Zo)E7jzNiijFlApj9zvS>bmKAv9jHZchJa%|DaEF{ky{VA6; z3FeVyl1-B}GHH+yQc38S->ArmV#_cyhmQ6!V@86R4UsMX^pa15g#d>}EnFr{CZShu zwt9+)oekkmQ}vl|QLR6W^lv1rG_MxCxFYSo{PFxTnwR|_dX&^+X&X+`pqbU0(MYbM z@Irmtc9gS=F6HRNiNaK@zh!*TQVVFMO<4mh z_qN9xZL}vsv7-*QO%}+qs^gmu+oz9c|oC$m3cB&e5H3bqIXPR$lgsQnJ@$`E+hvHWoTYF_m4nQafZ1oQpb4HVmp22wZU+J0shBThX(9C~_2-@g$JdfBHuB4y_^Hz%j8J%Jx7 zk7DK*`q-F-z4efC@QwMhWb4`wcwuP6ByGmPMwrV`lM+j? z*=VJ~!yB)D{x<=N(=%n3$RUS|cPU9gc+;GGz=0u^J6%iTN zu+S-C7xdIlNg=aMRv2qEHpE{dNSAw|I`)EIn=#W7*OZF%v^01zlySumS0uvzo1}#Pvc8x2|caMi7W^p2>Ria zyaA{s))C!K9S|4tG+3S^$qUk+LcRLy$Sp;+vqkJUV+PD2$)mC@AW@TshuH=~i{L^6 z_pt-C%};^8+`h?yNBkaFA@ZgPwVz6rTmQ`A5UJ0AotTu8j!ZNnuC-3WqK41LF@75x zrj=7Nm`ENMQ7Uvq8F7 z!}IiY?@L>M2H8s>BO1#Lh8PsdQ#0UOY#f%8iG}SIL=*n|+FeR!-_JI6?Kn;DZVV86 zK`3Knc*q`B7G{C_%biDXp1gE{D?-B^c;FFsG$0hO^}!a zikzdyDb-O~yl>lDZAc$2GfLM2iO(eXYEGUz@AI2Ql}+q0wTaQbz7(^0M18 z1Z!Gu>)+AHmDFllK(-xn*Q={SADv|neRQl>E;*BWSp+QSY#odSt}4vQ;O}9bZ54QG zab|z-PQTy%wQ`!oA{ zKPv*opTcgfBRwT0^-eVvgl7(7QUKl;MO7IIx#7}>A#B}XQf|7%DMAof(TvJ2$8foS z)()-wF=Rs+JVKdOSKRR;^7ym0n|StzrZgbZK5HSyl z3ch*BwE%YN;mk*m&Ve8zxIC49fan7S94a)U`OfqIL9fZkCU83t~5EBp+gQZM%yLQ*uJ;dSvJXiY1Szmmf(cgGnk z9Fam_hVw%rFDxRf&8=r*B?-WwM-M3hLLEv9f)L3ci`f(asiZNcNKgbGx(N^jlQ2BF z@h=9K-+&&Ek@~$|@K@esPNU?yVDDO4EQgr3kid-$X|s`b^hfHc2RsB7fDMEgiTF!mg}KgR79GWifi9Mw$D3^#(I_#C zkd*o*XSJUp(7#C}#|y>K0ww`!W-&WFH;X<4RcdL)!)A%^=fa60TBdL!&4CLJz7G@d zJ1hjRa)tv+#w6*!F}^wGVFW{z(f1hJ)PN9S*Paz4X068t-9t(=Ibbn0KuVF$e2My! z^Mihq5fu$Sq)1a#P}p6V`K+rJPwSwy7QKAX157K5R1hFWjiA`j8GHb^^b4~eExsB@ zm@appHG0=;_~{>Kr$h8*-GFrXmWxl2W-5vXM11(cIFQ|QpI7fG9eh=60s zv=+zihT&lNxC4iY06zCl%I@u5M1z2fax~3m`?td=`d{Lg)~o>o zcyV1ho{@bmy@k{en+t72NUPjkcp`xhgA_d;zGYDLzi{Ba)-M%9AzExWA!ZPKB>}~- zl9U3{vAAnaX0sVr#0VN()Vzx%s&-KZw+vjjFqWF;f(~jFhej^4>mg8CNlkD0h#B+Y zzJm+A2W!ksnPUNNn-~Sd z=4`QCC)O?9FsCFP&fhmd2mn}htXQcWeifyHV%(!XWob*!iKDV_C9QaLD!Bx4Gpqd} zIz*v&9bBZpVF_{Jnb0Q+cgL^6EKft`=9h!=_P3N0%0==8ICW|Ps62B)=r>KUEs*KB z;$1M+!LO!IQ+<-P`|t)i9lg#w_mMvhw*WJHUZ}gA@D84QE9D_&pEibvfSNS*+~Xz& zlX{q$)1!s_A-4!+Q622u3~Vh}D9e~Qm~k2-m-N+0I6L9n`-?06()$I_Cvxyf$nKOt zzv`7`1(W|(a_Z+b1gt&(AFwu$JR!)tDEHAG_#vnM*-)HuD9crXu9h#~pIXKAl^|m> z(?hu4>(WN5fath5Dt&r!*)whK>;jk`LBWg`h!s?`5GKa0eU zia@2!ouMIQ9)Y?brqyK8ugM@kH*%cit6x$Y{g^k;cI7X~PCtbQf330k43c7_^l3_6iq`*$#xg2hNK@YIxF7q~XP);(C zz;F%|Z2(Ar*x>`Q*@jb#hva?`HA~mGCkF0kPS$Kqp!aNYFvtP*T5kF?2fr}qdQ=+2 zl*}9x2W8TtSZjnuQc8c664J76jOmKQCcUP(K>Wu~GSYc8qW@rgyPEMrg2B>uibj+%}$7mMd z4xVJSEldrwuvYBP(1vLNU7;d9Md2t?@KO3?Xxp{wis$QLCXzF;KK3zZz922l<9TZR z#hPsH8%NHTOHETSL7YGZjQLgVsEz<$beVztq~#Wu;~;Kkls=(fl&2bH)IV!v>*^zp zB9I8q45?2|#1+lUo@%NU8(pHk{KhT7nttS-t>6&_JEx2&cO#RMQo{Yz!O+H$YM=%c zt9_4~M4B#uOyAEna?INSsi7B|2aA|;SME=~N7X~URv0@%%O}k{9fXVEnN5fFBlE)pBKO?z z&R682CxL5E^r1WMIn15*>&>ChbM^4#ypzP8*Nh}cbAqbWU)@8MsmqJ}7=q79crxnc z#Mwy&<9bNb`fn$vrRvH$F03j1PZqoj*?g zoJzl6ud!p4#-|b z!JP{YR0xSr4<6YHVmcpxSVR%_xRy9NJUT?RZ~F>+N&1bGMBWM(%8(_{Eve+NqP&wM z|G476p2A-Hw;2?v^>Vq-Io{U77HdYog1zp!!$>D~{lZ8;J!qHtt!p=fE_nA->&PeS zYys0@jGQ%oU zV|gA|{?G#_9m({AF zru7Ga1_8YwRBm2F2ZhqV|8)EPkTV9EZ=Q%uVd3U8&h7Xt;z`bSY;wC}ih<-(=GZiy z1s;WH+{h;gm5ZRq`9{yrfHyGu=%Mrl=Ymc>o%oX`tb=3g{QE^i{xZy9@*@#XE?RLslhaLM0OnhG7vjTO+Re0B97y&){%d6Fd)T8N$X1Sf-@ zWbMWWXPGn9aoe@uuhA1}Dzd}wHNs3lETpel1)E6@~m>@g+m+BF6yL>LV7b?2$wsWi6<%2hAve$N6e)wpPuG{&KML z3Fwbq=*s-4v`{Y*4H5-Np)AfUx#;FWLhhbZt3J!we%4-Ha4zekAoXrBtlDg7H08fW z&KoZ5=aj{*fdhyU%{g#%uI*8lYPzd@V1p(%Gt}0`X}-C&w7bWYe`LwW2C*JouZUov}4M>kY5U-={`j<8l4)mE>@q2_*@ztc`$acz$l=tyi5nE{NIE%bZ(W4=P z!5k8EVTxv+qtLF7w7^(R*zE(J`7YOm zcXwn^M76xM!C>z)DvjuaNL867r}O$Ff6?qV{ja|F;G2%K7)r|!aPcm-IdkP*MEI|w zHdcP0#ELA$+Uj*oFU3pwk0@xcVKe48Kg;t)AQ*4rs1FMDjQp_yGe<&vmxXF*GMQem zRF@rOuns6rMau&|zX`6p48|`07BalcIYdSAHE2gaVHAf*eY|!v{@!bMyuY5~gLV}N zJO_L$n&~&5!D3iYFS^Dx3eX3SRK!N61_%DRUGCoWTEfR%ZyIOF&D2WjD~&$PBZ0&9 zyC>ZoayE1x@*!}`)=OMpb*)f-35@Ih>xQ!ZWFydf%L_q^Qlf{b054S;t}9Glz>f zE4`tkCT<>s_Mf8;vNuMpYB8@;*-nFHgQh>gG4s@ndf9@64E+ZeJccuRi#o{{Ej8#{ z?UWi1E8<^R_?BcF`IkR>Y6Luc3_953y!0w2KnBiujq4#)87nsUsfiIwO}t!+!lc+) zw2~z5A`?OF8)!lEWUrHQ{$SPJ{1Xv8L zK;HMob7kN$#UrG_z|8owv*8YvfBi*@& z|Go=m7Uk7*pT(@1WKNGKizyqr z7L-EFY&A@|1r3VIaKGu1n)J)HTnVlO-0QaC8EZ@Y;QSR@uMh)0x0cS@V)7Uj*2zo* zSZhSI$aFY3!?3|Dw3`mSKvNVJ4p#J3{7tsPSD!@1%yPQ2O#QvK9DU|O%$nhCpS1At_mG) zu-J4i{D=+$&W9(sU7RmA-eVTwKv-^gD)|(&r`EOg_;3+5g&Xjw5QNCnW}GYukcw)k z0n_b#h+Jc!4z~x9gJ&2s9fMgV!iY(I&ZAm9A}=AoLYrp1f}fHR5of&wLpG)yl9p(r zzaPH9ntPb}DI-v?p2Sk|ddEj`iZgzVl8=lyq9_cyM&wl7Tg3%|vwke4Z1ay9{Zp%DwO z(p);VSAS;QO!@e^6IDj9NsDYH>);qYRLxDRrfgSw(B8p_v1Su9PKqYYrLL`2?)5+( zeG{=XbXZ+-z9D$Q?tS}$2B^;|FWUXufCs!oM*rMAv_+A_HF<&g8<)m#3K!zHGYye| z71ju9U+u!CAMB;0`EE{5fQ-gdh{?>kTQ9UU7amtqBh9qIhJM)A=!P7>e*L%t8J$O)q~eFB-O2I!eunYj~MrAsw@4MZOR z7%F@N*~+7;^uwbkCrA53Cs|+r?~)1n<0fmwlFkMl0QNsK)1WiC;;TzWoO{mUlNrJ- zdA*LGuZc5#*Y`5UsOQ=h=fAOT)Adi5rb^#kDq?-{qNLDKFJ0zoyWgg@cddwVPzT8D z4mI~&I!MXX(OeGs4lJkywpTH^MJ6JW@}LL@fnWt8X@IOL(m&x(XtdHQqUeahlt@4vsHlo-7jHRN%bl+{Fdgig*`Np?a+$Xlh~T!vQ+EXwU6OU4EAv!|wYp%lYUs&3AHqw^{=Sp@vex~~5bu20tC<6liDZ@bzqb=VUBhF0wEM&oOz#%dyDKZj zse6?+eKebt)hvlTD~oJ%_4x=)3GUY3^6{?pA%*?4qX>5tu&7*syl&@TR=g>@1YToV z?&MC*Z~9oFu;dlksPBuRRnh+pad>{ebuIrb8`u5&m6U#(t{I0L^S3)!ib&AAEO!cP z8Lwk@sGib%qka6{y6(NxFeTMZAA+3TUCzi88xuii#HUYCHK|i2V z`2_yxp0)bLt!;rsgO6S5G2~;P(+3(!!E$!nZTFK$76VR?H}Y@#WWBht2r-pN(WQtx z`h~Cdm#&G=H(iERYWuA^Op5^fUvC6he7yQ6js@Dyq;bUZekl5;yUSS*I_j*^JcS0=t?rAf9XsZFQ7I%&;ww5sY1cnt673ONa$@T(cS zo+AkbMN^BQP_DR1DcaKAN>`j|Ah19!`;$Ia1uB$>47hR&&ky3@aJ(f8us%6cpa#cq z&dK^z6#O#bTgX?a4zab`@&BY~zD82?o1;lE_rOIri4I$qLtLD$s3oQ7uxj^3p`-LyQT2D>Ra&ft@MTHQ-Vf|Rgh5KBA( z1tU!VLkjFuo+*28Tgr&(+X8y=HTTej(YRwnOeCffxcSyWcO^FXLzn zf2#-`j_eRDz?Q|Eq`(PD0aaf|&+B`omyuoPG3=y210XZXnUektXH{_M_gg*|WcA#F zFTh?ZQ1HfPP>5*?Sxx;HXRf0$V;FZk6B+Uv8cF6O+u_b576hdh0@Bj-Zk^7H)@EbKjRg@IlOXu~}=|3_IYp^=$EW zV&Ve5FY=tg2J^Hj%Z`P7d=kkTn)hG1BhfXWhA+2agP*8{n@|h*a6rs7)8wo1qssp zu>CXXOd1Y;iAT{C<{}*wdK3j@^vE>?c5Yen5 zz-dU;_ZpD1`N}7RG-)`yRV`Ah7U|WI2kTBSF`m+j*&dS1u{y*bSh=F1@AS1{xnZ3h z^cCI=HA5v_-{aA_DPN5p+|;D^2T@FAup1T|CzO2@>4U%J3p;-1M;x68O}@8e)u!I; z0TI}mx=}xqOqX)MN@2c~yitt?LI0Ogr3B>2b-u#&Am~P9#Ydt>jAwbEiv%4Eiipma zv+2ELNO$M!oWzSwmho#?7@ud5mUDI`amUYf z0|^SorQx6c1p6BXprG*f8K(tlZem@av9`ag~juN$?d_fXJ0dyB+pnDxUVeWqkWGzu07q!*Jiu6I%oW#5xBQ;HwZi20g^mX8 zqVMcyx;r0N8d5YFMMMVN;tZ%>c#4(x@mll1$u4bNZeW35peFmE&*k^j{P!6+sHrnw zh^Rt7Klcmttn-p(yCrw>C$I8I| zinW(Qn_Tbk(9&|h>(W;Q zMO)Ft9;-C3#hMtmP3@RTiX>jom=(%xPO#p6-KPzG>*se)LkM4Co)UJ82(4&{&fh(t z9}HYZJ7qW&Nn;Qr(3v$+p>1)Gm`EvjGZT9q86}t{+?jp%$$mRo;JH?*j@xR}D(r#l zSZVL=ye+j5iUYPt&N1~qcz}YCS*5lqtGGu7-nA^hjs(^-ZihMZUaMlMdDtZfgqpTy z5Qe;O(=Txa|8GrtF^o1E$Z+d+VJP2m+vSfw*`6X^E-*jvws~mm*nc4(Tb%_Pj@M@h9xwF!LHeO ze$3R!#KchJs^!%l=ECixVneLW?O#28%p^x@4OO8YYBWccLt?tW!(Q>vbY)2s+Kdsu1#9{wRn+o%HGTX_e-2QlQ7*I1y%7y7%gr+SHfJ`Pr@wD zx#6Uv$R&{0^O`>G;lxPmG-X5LLiB1gnP9-i`035u?W+XdZ=3og3B{=r?Fe^Q4&Rgp zpif3YZXbighC5H|MOU1EY=iev++x!qOebjv2iyLdnvI#x8s4gsh(ND&hnCqnWkdZ9 z1N1Braj+kt?v!MS>xhu$!C$G~Va%NV+&u~4O7Z0L28?39bB$&-PgGmbhvnj@khFGM zNk%{|$4V@6gsKuO2L|8lTDV)~f-ItorH{1o`Fvv%v>C7~oKRywgZ9-P7t9#QgZ>8g z-uAx#Nni7F>s+U@gJUTF;PpEwCO-e;JnrldS?Q~>6sW1dm*zIjRgSJ{wL+Y-IP~$p z@RIY*+yJT{E52R48D`gDfnCOg%QGSEJR@HWl*M)Ku8ZKH$S++`jnnY#S3Pz$3^A-y zZcs3{dFq)vI^@R31t$Or`c1y^R|-#&oFP{YH{QHYz@k-!$48nF$|fnyYOsA9A*jKq zMMfGbbru5K9|vxqln8A8C3r6s*xoJLe&UQ9N_f?h)WZqj($?F?*Z^h9hyP~-Q*$Xa z7@XvQGJ6k{q_naBIZEpBa+pn%#{VxgPeLF+t ztJv{%i^ZepMS5cqirboAgDW}QTJa=w^we}rk!iEbFLZ^R;`SOy{+u31^=mH*?iswl zwctt=QBQ7z!982X&b!B&jxuR66ZUpiK;YV3f~!>|$;R<4kY=$>X=ME0FF_4TO&+4z zaLoz-E2AaRoG&Z9*OPtJ->3E+N*pMQrd=PRORuujNkkvmAA7W%`V`k*F?w%MsrY+)tQ(yr_nkDK_ z&+`qs$bRggFS+8Y7Z#T#TMAPhj1Q>0qJ#y#rfq2McYmQ2S!18+JSbQ2-uY8eUS?)~ zEPH~!ZE&3-)=PTnr=BoD@e6gClM9vx)(z7&% zHESHejvNq~>*Mh?dKO7emu99c68U=tJ!V{Cdk1@kP!vSYN?@7oeD?20ZMkJis_wf#rlCKyX6nvhY;vgRiPbxv2;$c_-%dX2=(T`ifVL3%Z;{$2=<)?{? z#a7lOm}!LO4>xZSlydd~42s)9YH94XKd^SX-K_l2XX{@;@U3lnCo3V1EfS~* z5!B#rwdJBrj5G>NqYs||S7TwKr7sslr4wZtsS-gD0#ia;&G+1dx%$wSo`kuA*Xfqp zQ9k~pq}FsD-1f~%xcZkpRfOZdz>Evp7Bb$RPj#d^p_3+`oMAcjMO743Bk2J%xA&>_ zw^QG5HsBZJ|JNvb22{0Hxys;K=TrCshMZ|N7Gdcxa}cK^c%F)~7H;U7W$qNlShr7q zK8ZF#-OY`|Vtu(A8to}^2B~WNf^U%)iknl!nz4u2#MEo#E&%c*ig`gF9NWC&=9V#h z6=R5pEasBdxtfFQy*TCxU&3212JuBgjOK#|IGG4^d1&kLYLiU?+{_lgY^L7M>{v)9 zUSf}vY~Mf54*^&OjEa`2hc1j}BDv2K^52BZ=G7lJA8t`BFXjtD*to{+?}U%mh9o6i-hRG|W?M{+~|eh1z8EWs5*d?{C9K}jg% z`Fk0?=E6{m7NHKAx7}G;#LBpwmQkB@alyT=v^P(|5vzcWDn(lyPrBr@;PhNo7_xOJ z^YLAt(F~IK5#O~CWeSgNN3VJk|6MB~JgD^EySh>XYVLX1Iy?SC_*U=a%1vY5#&?mr zXyHC~&vXXEy9IM?$$(%wH9UE)!3&~jpSLpVHN5jb=5Z%am6mM3B7uJUna$6z(`0B5fZwGmqnu%#jk?~!m*DNY{*bMc0j$Bfoy0^?CHkQ z3m7xxq{r^FLQy(NQ6EP2w3@n;w**f&14&WfMGp1<;p!ZtBWt5J9ox1$wrzB5+qP}H zW81dvq?1%^vy+aKj+3c()|xf*&7b;xR#ly<^Xz@y*UiUMS>e!P7ir1jk^e&~Z#rkG zMP@7e5*{8feghVie$o+Wevny%^8HL)@yAdIy)-tXF+nLFUJAGO33$%_Sca~S#nVLm zOIZD}f!F2mww+3lRlH%Qvx-I=k+ue7Z7Nj04tO9DZ9r#F#6)0}5#{29awd#EacsQx zJ;(-KGx+;vMbK9H3;HEEKPxr@Zl&_(wkqceP|x zhiY09Hg-rq`PbTxHJ;<$v+11}F$4H-&A3>Q4(k?^Pu7)zlV)F3BLH&wQq1j@b^~BO zj5mMww=+k~tqRt$6wXUVS4D>XHYS6%P@J}`f_{EZ@YvdXZF<16$X|VyP)(C0MeM}ygF*qg~ZSMo)jGPtLDr$7^&yit@~Ms zl`{zz3qypHTiCpB^e}4M?u-jyq1!Mj%Qms5mlI~*+4u!lTm367#jyORUG8D@J>JZT z@R?%U?V2t%z^ZIVlQ*1DUxW8vG&h2YF|gf6;OKBeD!dD{T*SJN)~a+n7|VsLb>Vzr z;*5op$55SC3!F&c0g-eSc9{O7DmXH%Bn7yGy{NpKSv+^>(E)Z};ox#8>)YCsjSrqJE&}Mp zxCqBPaeo}HbyPhg85^F(i&|w3E@20d5afADd^|5^Wm*+agW~a`2u4C1zCsL?C@vhB zev@uewVJOCT0Bg4+iCv&a5gUNkKEHtJcj0Geu2m%{xGip-0Hup6Sb=1?TUBd!Hzaj z11)r3zWP${5hM$)Nmg5)QFbmnP+X0@gk}WkI zz5T%iMZY2pY^*lK8S;F?SZQeWaEGXIb4y>2sU;%j;7Z8(`!jJX+}J5n*vmHCb{X+M zKcp_g^R76Jc|Nuq;^4mjI#z_w_oXIHI?<6LPaW%FHgvnE7~&AbEVO((b4tAz-z_)NElk4Ol9`?3l@%?Uw9nOvSGluTE3k5Kb8x4#Wo`6A4i6jQP}x3L^N*5lapQ9-A{j<#lo=6j=;3cDntg{ypZ};oepY7s^h#}3-EmnH` z)u=M~k6?M>6z<@}7O1X@0XuAc2kw!Kf_>-MDTuj|{X)^InAs?j|E9;LM_diN2tRT; znOtz^5&OjnWduaFf#l>?pPGxH@C%lf&0RyoG{t`w#_@|p3d$iFXj&^2rF4+pGekrPXf8Ex6 zBOYPfLUw3P6|qmrJ9%w;c1}AiC_jN=CpWd#ao+KH-^WL;Q6|)0=0}XY{#%qh?q@P`&$Cr9I0i3{F4kz`LT2k6U zS2Wo;8&?3f3_jT%`QC0c{Z1`6vOAa$e~G*Q{qhf z^{)A3lKg%l?s?dBq_EnEY>X}we_C#Sf7HY=|l za`as5i?=GVE4aJTPzx#Q14`0V8!n=ySpi^TRoU^M*m~=Fg;<=OT`T-H(7@j9<8Uji z4DiL~h{=cvJcCzhRg#q-Z5O2l30iQbP;!f=gAwo!xzW?dme^c;h7StnVuRv1?T*dm zDTs&L+^G{K;lkHhs5lpG7NkoUIP1|+zyDh-S1b!dsB}}8V5}F7B=qib#WuNqIZTrw zIL&%9SQc(`6C<0}CrUk6sK`!$KFAny0rV5#v<}5tH+gE&S*iQB1-)nQC{NrB-Xb3Q zuN%O-|FbF7Bcwx~BV{zyh}1&zVkPXwb>&I8q=)f^$wSryfXMHI{`C6Q(bCt4bX!b(93yt#u6)ZD|}wJ z5OpX}=s{~cl25x1#)8Z^p^1w0oMZ9FQH?6+8#gMPZ=i-9O7AdT`th-nj%U5zPu)F=-s^;K#@@U5K$6}+@qyIafuBr3u>E4&IGg{043W@Bi96uK z`tDO!Qga-;7Kix-3YqXdP6;`iGzvQui$W*IV{zC1rU;mv=6|tH@vjL#jbIRszpF}d zbAI8uR$AfJ36B~dw|04JZB=y*{hmfmR9xs9B8GF0S6S0`>|}myZA@jZg&2i7o6khC z@7P)ywRd}k#Rqu>0mXRq{ogrs{k}7*s+v!kyEQQ6{(EUs_RD=6a@^eF_4Puh^D1bo zOiP{<$QHRovOV2e^UJq8E0~2 zHDpnJ4GQVC!$r@Zoz~~R0kQYJZ)mqM=&z{ymwn>1&d>bBg>=5a>6fvKp$=m{j;p?4 zo%-%Po%()L(x&g~t+2L;fAchP>*b&=%)dFwSn=7#6e5`{1(T686@}r!0|~6efIE#W zSbdcAZs>sq*~(A6#)!y?y45ghsAxUxar*6hdDKFGu(oI;9;3b=D5qRCpqWiphOss* zPLvW<2GWZ{AAd@3AHI3z&lG-G+xXdA)N0_x*KLK(3*I~Xd^LT`n~Ft-K@)$=`n(-{ zC;kUwk?I9=-nai#zWj$@X+I&q4CHvm^gEaxDHT=ke#X$j*3o$XS+@)`Rh^KaQ>*@> zL8qp&zvv3BU%wMss4X@av2!R>K0gcPB+n1p8$GlS=XKEO#1J;cijIe6qB@@pJ}LAE zhI^T3YTe55E!;NM?x*#M%ky&fNH+tm?iBOC>y>QdH1ItV3;3__D~&JHmYdHgjW54+ ztK9eMmYZ*|;Ow?d3AzLi9RHRG&sbdi9SH8`+!f!+aX|I3@7C@#2r6D&a3~ocpSGMa z^y86+My_+n;&uv;{r6X|pV)3st`Qtkj z?e^z%&ZQ>;G;Q6KnA%jZd8U5+tsrbo_D~_V$JBtUHFeLNKwujh;}G@V@#y{vioLkh zG|T|ntM%8HJDb<)UDqdpa`pA+cj9uyz>k#UogU@R*MhGteuvM|hQ-yuca9js0ct(k z|F5jTz_U^Z#bvbv?tp^dhb{Sv z*)!|sk2dnmyQm-K=2#w8fG?zJLu=zOE&*Ry zp&xpNuS{wAL+Me@A95=%f3!xcdU_uUBLX7Q+@fZoSE~oJ74}^gfOG;~z3T$z+T2;d zEgrTz9-ZyLDszVxO$JK&>RYs;B&CD~1cnh7Sg=~-KTSWMFq7Z6liJ2G1cH6lf}-Jo z?4mzeE$pmctYrt!9Y0?q%dwwe`b7!rG=I_th<|jr40b)tIUaw@fX+AMeF47|_gyMZ zd^s-3LPY|JR2H3|3epq`eGWQhA|^J75MR>a6*NH4@7)wC>FOA4XlMmBr?NNuVw*78 z_%l&nv6YpYUua0mSs2?Y-h}DB zs5lOutP)wsAild11I9np%(iTM6e z%<6fi9kP`izVLfI!Wwt|v~7bbJ^&Ir9nZf4M?CJI9LycGg&mR6X<&`;fE#`tXmB<+mg1!9r^X!iet@TWsMY1sPpSB}-lDpxZ~>&u%Q4UG65 z TUMaBnVySJu!OG-UxfnP*`x_aeMk6B<)QBK5nTW?iM19VW>TMYjDZZt~jqO8Q zEeB3F1@gW(TjD;fVE2mh(6ew$U<->SBfs2#9yjzQD%TVRUC!+vwJGNC8iAq^c#g3< zG`-tFlPyRUiM_#$Hdpi^cL%o+oz{+zmj$pq$gB<2D zQNYUvb6$W+s=?>=4;d;AM;GxmQhI%mw+})Lt99Wb05fDLcBjx83!-aVp`@xt5nPUa z2^3bqAc;I&-Z3iKupWyZp2AYf{MeW#bykSS>`O5SXU!t7LED;i$KE+sJ$t>-J_eZ` zS0`3*VJJO9Ft3goqYVXXHJrTJQ=-Xs5Rq94ZwdIgk@lWhUK+eznr$#WBvwvdE#rK( zK?}-eaziVi5I6?J6|D(qEf$J)Kn-ot2$y`3k8p zJ%$EBQ>VWF9Vb8so=fPRbaUIQC*DP#n>`6QXLh9)NF>i##GpNb@$M3Z7xdEf(tp$M z5nwoeYMbwyK}(v#D2EEjkvIKie)S;BM2Wz*qSgCR+z+T1WrHY!5M3}qM_3WH4d8e@ z3^@t%3_haeSrzvdHmx5qk!6lP&xDa`L=eV@w^{01Vc*JvhBwTZnb7$czcM&%3Y(J$ z>Cc7ilBLlj3;g@cXUP-s1#;M+s{}fAZ71D)_QN)@`HG4bhq{n1~2I+ABqKXsUAAXoc9WwRF7wO)q01$<7|bYd#{dxt(lc9D)aIX%QG z1QuK*z z@v!U8NEGF3xhIEv96$<0hvd8|NvCrFDq`8%{`N&@w;PIm<=fk`HEV+@oH*VqUd<~OSizkxsNkQ@ zD+&3i9iJJvuQsb5|KTM;WfG)wjBA(s8;PR`!Iv%|~S z?~({LSl^#6A?x@Fz-z(<3zT!u1RH2iJA7TYx9>J4WHOiB-W()}&O!~+f4{R*s z3!%ZcxOQc?-foJ@2QB8H^(Mp_3gz3&sT7xv*I8tKkPokcCFJ!Pa`U)KMo}Bb4qZBw zF!*~$k1O-dt-+b5NQG|-jRBqe_5W(DfnXfqi%>`s$V{R+{O!tXxA%j;+Vvs6+WHb? zSqW*!(WBGq{5r{rDNhV*F?%P%l%a(oz3H#Igb3TbayiGgV5&!G@JiT zZtBbU( zr_&H8uX9LaZgz24A|}M#z__`*Pm%g>Z3Z(MF}6|nI3Dq{aOW7f`9)>Y1W@I4?2%`p zghinM8_ADZEHTB3_;!b8`0qpHoIhNG8omXUhVNG7pVti>3&6AMfll$|zz&t&cUK3T z?rtC-N8iynq#o^RdE5Uxo(zlT5TAu0{XIB{J$d>sc?Z96gLhgYUN!kd!+=#Pl^bxV zr0i^z-?5qmSdsX!LAioR@SkSc>l3zzLquU`q2(5F4@yKHwzM2$w?04sFP1dkA>h`= z25(gb*sQo9Cdg>Et|`~?*JHuG_>8^w;u-zZ=4s|LAIA+#Ot&PQr(T1iHk4hdyt|YA zoVtXnoNqxxML>{>EK5jR+V1Yk7uf>c?9vNu8wO?K0@)|t$t1@6i()-8TyUm(S-pIhG#L`vDpq`l`hUn6MI zk?;~k-EN?tjbPx#&z!Bmr0+Yugux7g1GRNA0co@?oh#v?xgTd_4)%ypr_NQGnbV8= zO4)PzMo4PyZg8y}K;8cND*m=gemcGee!nEi^nCnU=lR8_!huiF*cD1d0S{#y0l#!| zMkf>9zKSv*P>8?a(NRFK3*`cdGS{52co&6Yi$fALEHiTYMAyu0;ojt=Y+^K!9fX3R zICop|iU?7h)k&epafXYm02hu@>1G-|RUJZMwNGFpUvgM&l2IZeH;gh=pGjy-Nr+u! zPGj6K1GbAZg$jP?69j0$7{gGyDERh4hzs|D;$Oa- zFDQz$;S+f0m%n5xnf>h4P`a#qwKAQzAO6bMt!lWWytwKO_TjBhgT-r=wxU|$6`1|l zdAb(}QGE2=BO3r^g0>pM9kL~9Ims`25`jGN>2;^q9<5 z5RUzGVyDk-x*8wB6!nq%}5ZU3-necqC=m^9)5S?g=zr6%=8x{J! z&KMz;|86Rc+TN1xbOOvk1z{=0?%k*0#`n!}^(j>`n!XUtCpxuwRq)atZ2#`ULrDX> z$J0ULSLKb(5@aNo)L&q$N2ykhXe(EO-vE$+vAWV#HAm-XDksk_Hg`Hc1>FXm%n@gJ z_N~8(Yki-+Y(O8o)lR^kge5^Y1c(6(bemk-~ zox3VQqG;h#)z{z(J1+y%?vyi^Reh2@l2UceEV)>F!0S#VMSoFXN7A1un=f5rBR`p{xHZ`Hfh;{51*uGS-bH`GVr~k zZh4`6TAPOdU#gF#=N?yZdb$?|{g(BY*kx&(2)Uc`V%cGx@?lqc^Z@DliBOw)0d*owCr_#-ZXqyA%i7wx`vVjoed5`&KIvbDUTYB%nU1ne`8Tf~sc!bRf&KpT#oE86fAbbb*cJY?t8<#fZYkQI5 zW+83%?fl8%(xk~s%k{L-EwFv!kcfd?T(w6-owwubyPefJbU$Dr_uYAL5gLj7zND$gLk^p zXwihBkx}}ETal?0K8vKBX%RLvc2EM#>xE-~T-&()riO$SW^4m%dpEDWrKDfGgSSP;P;sbIyS*|{*J$J_xtC{#8)!QfVk6QR&3%V ziRiqS2E0D!8?WQxJl22dl?_H{)a;i&Yy0^Ag6-dp&ISnF9Lpq20JOWd7)0K?cQ3yf z(cBx$S8dR`w#}|TSHIGSf$TYow%mN`NUQYVGgEXx|L_?@bKSEVc-p`?RMo0u{U4<& zg3wl{fu_&H1w~o7ltLXN0jn@}HShMu!c7L9Ui(i+%y{^F!sLY*kxdyWxeXGh0{(Bp zDLNRoHgThz#nck1&~icefmyW}l>^*t`5J98O<&|d1yvi15PBr+jctSK7FU0k3mIO# zU(LsM*GYS&xn{(+8p2dO#5Sg1;?32v)hNmR14S%SeIjY#>B1)0|g2&bL!i>G~Plhbav>^vsH zrqN3-@-9Rq_>h0zzA@byVZbvR@@w|$*R@H)*uoAaVhs)-St;_@3D6dAZ;&?!R%tQB zCym)q85YJqv<34&*_d!(X`DDd9>xOSKpWdI0KvJ0%J+F7_3T)~m#8^vAM+qYA*hUV zSEz<~gt}x)b;gBAi^%~ZbcR4G9gKigq?0|2^jJ_Ybij${L{-O(WWh@frwxo2#hVr> za~$0=j4`UZ(a6elH5n^tJV0*%8{P!eZs&_hoPk{wk0_BASg=iZe%C&Ad@%abRJ{o} zu?oBHy+P~v(zwURBS8)norjO-XZ~qPp-zVo`q?IO3b@h({V1}Hdw?|haw*lZnAAoAP| zhSCrxf85(+sv4_#dw(>`WHk>y&l2n>dD%hG4 ziytJ&+ZO@P%|w2o>_D*-Ie#I^k*80i8oW`tkM}d!6!A0bK%+W1$G#YOgWN7{`0CqW z1IRKSdPj>gyU;JzrvN3w%agf~8Dmt^pJG`^L+p$c;)O9f5H(tiM%kghPz!_b59~&w zVlyFUXk>v1@rg>=3*qt0Dx~qeYt9m7>eKqF^@;C;pSOK@&(u0Rln`JnBdphd+F%$4 zjU-%C^SXR=yXVe72E}2gEg5!IS+sH)i!j|tGviQ|#%)S?L($A41|qXCTA?nyItUs9 zMkd%Uy++k7k31c6(I?;^=E%y8pESjRDq){ z382CF?y^}7)sWk^8%!&3oj(L(g?Q3XHQ=ngg@133e87Z^lmlxx{Du9XngYd;{Q{Q} zgrFL6lG4#aK z#9?%#LK#++nA9^x46OOKHc#+>_osU>UG&ZENL^h^fP-Pd44|Fps=V8#g5N`wlLA<) zf>18BSsLq}{`N+%aqhE)--o$Cv}jy0wc)5y)rT4UnbgE^eUc5s6AL=X!ICMR^%jk( zhKG}!kK!bq1y?C|aK+Ar4n-Fzo5Ub);h%QCM0i_B+1?1%?w86+<{$#^K-d6%7p6m> z7}0~AIBNM)Xcf;r=a?9{;#9Gl0o&m?$wL*%UhgE(gb5b2VTnP{K3nHk-q=E420k&3{+?MIs;+{a( z{YMBaX~3cue}&QHAN$oLoAMO?{h;HL6@>=hi-g#NU!IpmFQy%t9Gw|<4NZiNeO0Yv zZ(RSYc23ae_Q>XT@!{#JQ_bTWxIKTdn-V>PzOWwl5|~T$A_(M=7roKqA1LJ@cITGY2y`1|dIH9HJpgdC9|j>{E~i_c!an z=SicJ4Zm1VgNu@FWE#I9yu6}?l?(msd+sm;-z%SBz`mSJ0y4YyvDfV?m-nBT%WaGU zk=PkkDP2#VFa|=-gJf(px8pGgLV?`Wy#_3!JY(1H@ZI|OW9u7q!=ja_3mpakF{$YG z=wuDM)5rU8bP@URQ{2Go8t%dOyUEYa>_L*Jsk6h!jBj>^kGgLTQV0?pfe&#xInaz4 zGj@hnt;@-!<3dE}Z0obUJTq6}Mgzi^OW&~BI=*lu?VlN2@?yd|_KiO!iK?*;id)>0 zf#pOD{jZ;ux9H!f#fC3WEI#`~l`tzPk{WnSPWVss9hx`qS;nNIBcW$ZrGLuErDMc3 z9qrh?zn6v&1_lhdzrv-gE^_7p%N+&gf*D6D%Kp|Y78#8ZNe!mx-nv{#6=jjV_W0)b z;uSFo%S+pAiESaA69BN}d|WMLK)JIli)`qvD!kOOsPIu!YLq^TU#cNM2AERB0K%3l zy!fHsc*Qy+nd)Z4+Ij>tKT@wGyft5kiR4ZOb$B9jygo^2PSZqWC_+@~(W}Owh0i;U z$)Hu|Fm+YSh4VAa5z+JpvaGmRtfogxyhBZPm2dG|>)f}~w?66jna380PBpyGK%E_O zWxuu7x}FfMyaqY39E3B+%YM-lbTg;{!$CXs?1Da$7&{0Fkp(CT6C?-|p(sIxkU<47 z7V=_j+C@-*ABzxm?6r(mtulEiEyl;VCPy#!PfDaohe?Ue56Fj3AZj_3IE$We@uVL> zI2qT(oFJHqsY98QLnDl+&(oCg&4zsxRapagb9-9eD?M(~_WMs3vBW6%O-U|gF zI85&7<)K?#{@ReDM>9=tAUv~e^4P!v2k(ntVD)br7d>Bs7lOAM7weu>PkOFQPdiQp zfYZmM5%m2}&)=qX`&M@7tE6UJ)E8e*pa_LO;IwHqnmnLGAG9E2;~6n=k|YkNwY$)a zULFkKFzpmhdMbz}q5bLfeqqTK$zc0CMYpeuqb%=%2tF@=h=7 zF!5R+-h+d&8`&34Wi6p~A#TNj1{eUlH~#}-=T<+5vqciOT}K6r)wq97Wl0ykT;0x% z9#fZ*)3=@^h}rmA7^%#h-X(jEa2f*(ADO@TNkT~X_3m|{*rj=aXW#G{b*kP|iw=}w zd2Vi5`8th@y$huPF@wz4y9Z_l{=<>GPhA2R{xAvxWF-BcOQTjgAHjzU+Pt2$KbC*q z2SJ^>)gabvQGuclW-UcppZ=x0fJZRUw1s%*9TW0%k$mYW$ShLy7fgLvW2yc*n=~~y z4e}aodGz=mC@3BH(6IEl!4w|$I#~D?J_4)V>RJU{wxD}lBPi{q$sb$qJ7Z@Q@P?_Z zE)Zzo+`Yu1Bp(in`(Fj2y4s0Z&gKZ$kr|YsKwl0d*X%{{(ih+d#oAY zQrHWmrxihPBMH?J&T$g_7{T$srX!^h8+f2iJ?}eIrTHt zV6@4V^nB;|Ma(JtI_Zp}LBcM7CtRIAXarx$ZbJ|TmN^~`VqWXowCKc#FW$In>wPW( zl{V5B*%$$`IvOkl>Ep=#b54Xq!mA;ZW|0jLtrjL!L6e+WR0^DWo#>q1$o-snVKOLc zg2LEut(jy9=N5VlD{X66pV*{;d9M?!;ZSC@)H5kUr^Zg`R3U+Z5)GBzhq9m{A`R=>+#Ig*vIb7*?UGTHZ^7{ zk~WGq)q`e3d^l*>#>a^r>mlSqPmYout~#omyfG@jgfENxo`{=fUGA4_@G_@ep%E)1HEBhZ^Lf*UV zl&OFfP2v(G?vA~ZJFn;eLG} zQIl9Om9Nj2^*Ua&e*AZOxm!uvxG&hDYZYW5b9kTiBL^9Uda5ZEuWvIaS0nuCXEb*!*@WE(+>mgMb7!`Xe6;zy&tNIe#&cEy&)xM*f$}r0z zw;O_vd9~({?6@?mnPANy`GxGk<9NdR;J~tzC;It;(tG@DajWcnU(}RW_}SZTqth+B zw-@tO0(o@-I2 zDLW)7x=dv2<>GcB3MNwhloW@=Cwt`J$f8`K_?LAQ&S*Rto(KYUn-N+}0&t@Izy;$` zG}-vKVHQuGQcmv+@Kk5=z)>eaeJ*%9$F|mP!00uvNGozATcGkuL`3?1XXK7}4k~gZ z$+zPanS=BYbTvw6)0gSJPe1H<%#UedOKQ_!Xd@NKnA#H{X4)P7@E z>cuRrdWP*;&vl0 z;QB3R6R%?n%q)A=6_%)G)r!GG3>IK#7dAC%Qt00dY0J?g#jZ)jI|`B%Pz~H+%$Bn) zza4o_dCRq~lK<`1*Z^cm=`k_EDEp? zqJhbOJ5sQ6nNrEMC3L@K(?J6ROV4;h(4l6A&JRLoK(I=1@9?Ux+sI52DCvfiiSZ|# zOqr;WSTuMuc=?5X|Lo8*rGhOV6&2}6xap6ZUr(wpJ+-K(nt=m&x9mJq7mbVQTisSr zApqnwlI8vpC%b>uXK@OYW{v#Lk}IrHe!1DC%99U4SB=~Y zy6cwqi$k#cQQat2AbdKkq3^y9G@SVsoOq^{%dyH%t&1+vyMZNsoV*!uH{-Z{j+)fc zs+$L)SYV%kxntd6@9u+b(0*OR@Adep-53bo^J!Y^{IiPuqaC(ja=a?FIM9dCSPwyZ zlt$PH%d#6$(*M}9&Eyq5Ku9tvrK??YO$uC6VCji*jqASM$5HRh<;JeQ zPhZ+f^Ee@cExQQR&28+KUh0M|%`2m02VZ z_CR=)?T6(!H72Y~MM%MgQcex5nuS~ASA35d=}P_9p1EHOd!>|S;ZDJ};kV?I)9P#mq$XeXI~EnneqK0+Z5a+-Q`%Y1LQsB)%yJ2&c@nc0S6(dz`n z<|_yFOImLH6#?U*NJ*R&1}{sm0g!+aUYqMw?*S|~H`5qbdPit-6( zkRRpa$9a?$6J~*aauTCDii9W7g8qC1p)mdr9U1*(|K7r@edorpygjcuF`@Q$$~P%< zX6TAKl`HNN>;*Bq@g|BxC3n(62`-p-1;TJk`sT`)LA2M*^@)B-`YRL(xigbOpXvDK}{1;i&P|63H0f-TggLRgekU zL*IpeQykpt(^m1c?O0~=^$+dU%+Y0di+-l6B4xJz`Q`e#bR+hhw>~~R)twjbOenUa z8{gj|2%xeS6B{V?;pg(Yvu3nOL+&(89OTPr&pNo=p{!~tQ&rJyfs;-6fni4f<0Cdo&`-(s8^4@|3pW+qF;ZT7YI1) zQvE3#c#A2;<3JBXu~}MLr6T2&fEoUc_&p~1gh#&uE{aH7U-~WNdjm?c;hYP-rZ8vh z+P^b{8aYA={fh(^Dj5TPltMYWxYq33rM=cj`^|nxy4g-PHNZRzLpqx5$EjQrNmSg( zNiOFZGtZ5o;nU^DMpxF(M%G4O--utN=)CX4=@tBry1qW_7Ck*^tg3w6W4qe;X>>}c zA%R8~c+XAh=OwfT+j2#dW1yt}dT|i*%-W%r7|K~9#oE{pYPFbzXk-OM@qc+g6X z;pTJ_^nexWaI;F02AT4uS(pA{^l~r%R$RXDY|W7aWkbXR=i?7k(Q_(GnNNJ$tV84wG5K$�(n z46I=p63_`xL>z{m&De8}_5OV0dtlF(E7>KXToIMlh#7V*egtDxc+Q+f@0 zw*3?48zZ0!3r1q-KoI3Q#H6$W|YmgS_t-Dep&{Gum-tMDf*6UYmH&5*kiO<$AhS}B9%e*A?ulivfdi1nfz zCU}Df?8;^zClo4E==;}Q6%I47gq~@y049|u`yxt)M5~tlE>-wBR6VHGc^lK}N~Lz7hTsKr;#2Gmpo;+{BVVCOoKJgB zOfDV#*7duOK>?RxsO16C+Xjg=jE|Ji@B{fknqC3GtER?SRN8l#xU55qTcTlP#Vvg( zh5b6fGYKQPjlZkWOz#bS7pLxeS69}{8IYjt&;|7Y!|eCb*n~VaL|I)x)$i>T6!j7C zdcN6JG!DStd+kf1hc@ck{;;jY8}E5`*dniUJR@A}=Ixk*kfWR$XW_#4xg+B#!D2T# zlF5|pk|2)KM-nLxhnPhf6f;u&mB!#65iQBhy_8x*SeHcRh zPJmje`=<{DM{M75`-uLAcsga9P_pv955Jq_W;_@c@{naw+?yxy0c@S&ZWeVVq-*N~O&7sIIX%0mWSK@FLUIFi1%%WOq3@&`%z_+!s6`2Vg0DRks z;VXAICSYQ!H6&5@0swKT0 zW^pn{eJ~WBRpI}Afw`Q)!?q%?;i4`;XESY%!OBqySQ}Ha;An&U8wQ^(eZ2aysMT}OB*6R5xWMQ7vN@#a)6fyhh!E_NukX9$ zy)3W4@#n!5&eQgWs76M3{rn=OTSgM-C%MMnV+_ zx|I0<`F3nXx_yiNzJc$;PF7B2FY5lJVnCUc>S{wJO zZiZtE$Zk$cuQxEMN$V$7imG1f(M8i{RR^Bg^iFgCJ;p^ua1|1|8wAc{wB z_+2{s6#c)*`ljf}!Y=LD>e%Yo=-BB_(y=QQ+fF*RZ6_Vuw#|;MifucS@4xtG*33<< zQx|nn>zwoMy`TL6wx4(Z#|zED;-<(x@3Qk0LBWPqxn5x?k}l>4luY*(qxor6uvp1| z-~9h~1(#$Rh1itP?80R^HS!d%xR_BymCn~m=&Oz^yy-a{Q0IUI^<~j&4Y`ny9WmX* z+!eOo&U<{;{a!#JK2ptsi=BJ%xFO#Q>pZjn`2sB!X>BE3jM8G+eJMNJwDIn4&NY77 zku`lywF`RfXt+JS>jlB})PZpa{LZ#6oHh%rJ6;dR%@6PKg1m%UB=&HmesCf@bHHA{ZwSV=*-koSsWOs0#BTenMniovgm+|$qM%xW{(y*|R`#0Qy- zd8JhWDC#E|#~IHxiBR}KVb7sRA;YpKLAKR8(^rRoBj9=^c($F}eP40CpTiD}@_Kaa zh|6zPZoJm3?#yiSd^Klfm5bYp1ct^+PFg6HAJ;&TVUJii|Mpa%;P^QjS%*!6q z|MZQT`k3FSduXK3eplIvCOLC5TJYxXTg# zb`Zz>X3P;&6E}Z4G8(HAwX6{?O6cBgn{RYQ@>-z;>bD7mrPpEqWvmfLb9j>X#XS2D z53NDn{+F(ACD~%3l(nf`#uLZcSSWW+at-*)Kejr_TqpJ#;#Q0kcz=?5!sHf7*xRXZ z=(AYDuN$E+Nfv41lY6Qp*Nd(qgn_fzPqqtJ5h(bpU~@?-`FXI-0-0mp*9cmC;Q1Hd zu`d4vP!T97r67r^2E)e;cOZWM>DAr4TaW8WhMRrzUM>h0|4wj6*o0qGQ@pNk>$DPl z+I>yDm$^d}fJ2aQrg^_$ppT7bbn8$!M&Ne?(cHWqKNcR*lse6fAcP|X>N7+vwb;^0 z4L3dn#IiSndrlhN{bb~`jrQ~t0Q+phq&x2Sj7%P%6}n@usQa|#Wl`LA3%}C9ljefO z(hNky@Wou6Y!WP`4g8Y~B-3tbVUx`i?7Uhaa)((rO9=iB#r;V-xJ>Hq(s5IDvt>P)S^ z0WRvtS&^D62n9$$uP;;bV8zGIeaXX)h#f^?cgNs)Q3U0LSyj^cbToYwn0!x!AEr&| z88h`#Yd`x(q?B5H${}QBMNdCpX}j3{jW-wy4=XMR32#K9#WCMNniFYVq6Pw4_m5_d zDnoBIz)V`0?uYaWFi^bv9^MwkwZ!V7Fol{t} zS6kP|>7XwqToA6nyGKDZj%x>zU8XySu&LV5#iGH8;Z_WqEHg`oP~)6V$;y%8 zWbz;FDjFXPh#Z=MeYZax-(C|_3F8)w=iO#my{#p5hd*uOAHCa!acus&9?-{UT3x@f zu5@AfJw~~y9roL%JjJZ3JBPn|GtoB&D|o;OAN;@tDVVu@x4T0Alv$NwlS0HejKs}- z8KPCS{FW~83^}UN>Uv>3I;n+pWdcOtBX*52+JwGg6=@FDB&-t>ugb;|)THyS%_?aZ zgG*?$q9e<6{;u^4zeLDE(%bYJ)A&357z>#`1YEYridj6h8*Q4^0)gHt6{3H+far1; z5gt#@`}sf4;+FWY>|8NVC&XS1q3e}SEbZ45q!#?pcE5$LH@Q;=vm7Y(OzgT=x4+!3 z9jR^(V}aIWyRZLzBh|`fhsyA-;VQs7(-TR-*icl~KXakBv(p$vi!=vwA$C{n4*S}J z+RIxy$eDIImHfeoBm!O&1Lc}oYRtl}&1&0iJXYX|HEH-naP0|rN|zjgBfMc=8=}|M zedd5xV#*Y>;qj^QW9)cSrwAGmhzZuMj~3{=P!2{>&gwCBMvLGr zu}Gm;%4Zz~yRxs!)KxM_4&@EO{iu9ZV;PUSfk;$ArD&ooeYt1l*Qo%-lkhhqFAb-gF&GwE z5nKJ(oaYS<`f(#4`_N`4sO)K;Mpp0^ONvp(_UXxL0FL9o&x8M5Vd`h9@21ldX=6rA zYtXNaa2l~U=$T4%H{8RNt^Oo;4Z7=NT3P`F6 zfST+~)2YxFe>CgY+n5q|%FE`qhqm zLJO_RFo?EBmOYZ2+umvuFu{nsOGZe+Ben3VcK^x8hV+S%F<)KRlAp+W@@oKQX1X2o z-8=(CPBUW^^gtU*rchJW*hk?z*Q5Fia#6n{4*gY`)v9!Ica`v7&d zJ1O25x*WB7dT3WfyX)Og*;NOcCOGJf)zPJSnS@p{so0Zt$RaO;V|A@`Ec&=YNdYx3 zx56MzqBAiyN7K=ny}XAa!4;uIKITgCbEm*Q20ooQQ%D}GIzWPRT6Llu9l#EJRB+f( z$@Ec*Y2r8HX>$_=O$kcH?nIB!){BwtajZ+Q$81%~RwsAYX6cI4;7fV-`*-tZwjYxa8c&ni*fAIk89P09oKdv)q~-VKPP2C-=qqLq0^J>lw# zA^yI8WaDK6@-!mQHs$H~<#$2vu-gf%|K>K@e;N!i*my4vwF8-Z)V>|YOiT%T&c^nW zsTQMj;DIp*yXSz49Mtv}x0+A9>UGe};!f)cYNdKKux0L;CpM~lu9>l_VY5NPKNv*+ zXF2f8}K@?2cEji#o2uwY0$)4Hmp_a^ZU{Mi_U` zmHT|Mnm5xKL35OPirtU77aqMHFU48bWSVq&Ar#g=^B3OUe&w$#b2bTDCcpr0v;@{h zje(!5mud^u<=Qi|m(c1)zVkash1e#>_pB+RVip$}o=66ZU==ibvvfV`0Q+#pWKXX$ z8k&S!B`c1X!a;)A(_sW20QcW>3#Y}6Zsk>t0GmciRP312!!D5(*IzrRHqfrHkK=Th z3=Byck8E*bZa$-KstIH~Lf^}!feVm-K7*i9g@f!nnyttuHTkKc7Do01k?Sa5eLLiT zbhFo9Nyb(N$h=)#;n|V1Cy!PDW;1OZa2vT!_D=;rI;W)hZXM3=Q8#Z$E)25L3y;V3 zrH0uPfhy|$p4t}BRfe{{4*WLV*V(#p=Nh?y?SJUZd1`cu?7xS8XBvJhK*4x~p05Q* zXUCY`$uv|&FVAFiMy7A>a0LEg{kT?Nr8FMgI)$9)=9G-6E!|B+Yex93$Kz96PqgZ9 zJ-Zs!q$QNa_gvAC527RbEUX*qB`^3=S4YHVcH-Ly+r0u^>cw|*k_aB zs>(UNrn?+BBCDUHy#5J@(e0GzqKcF+U!w<Em3vzJf>3r*FLgmKo>J<&bNUA}{1-hDgQ9xCn;X2<@Xh>=pItJHh>d3h zgi}T?Bkh3|Ar(D*i*?FTbNB|pC&Ab>ZdN|pS{vWw;zonVcami6spXTa__WJ^R!%qd zzi-ZO*+=^b^wlpag6BYr@8!4)8i89Jems36OG1PEY-2Z%Oe>dW=XqPbug~ggSx3LR zUzeX$@A=bUzXvm6=+cJF=oWC)&6n18Sbei#BVhFz0tGp(47?6#gqoUNsZtF47ozdQ z8sfh;a!9`Tcg(?eWu2nG>YQEIeglq^t7MW7VDLyd7I#E0@5R~=96%=gNR4^jyvZNR zx)<4KYkODj{;V$8DD%~uwxM4QRHtove>2?MToF0+-dA2nSE1s#PH(-K+T!KbDMP_lMrx$}$fL^`elsOzCt-^xy{`)oFilvp~u!GD< z7nro7crSb$H}B0DzwrLO_n_)r-|vb>M|mu2=jg@5>*5d-crwjtK``-PP!GgO?+bDa z;k9-;Sn@0G=HG)xNmTG8ZeHX1#G*W9a`bmrF@WMN5v%XPbp zonSiAM!nvxPUSC1g6d!1ybV~6XMg9|%@?LssKwqHyf=dA`Q^YSfy3%X6FT=({eIop)NYFE0I$C#blLQc`2Oqgu zS?(esn8G~4rxb}~k>)M3Ct|dou+WI+_1l1GBlrxzHd;|>Ta1QHCIYMh7!xo|vd9#7V3#`j3&iQ@~U3-$nSlp`71er64T1{7?7A#_YlEof`Z)0ia1VC7NB?UA)q>f2TGG zoGjRT^>z96b_%lG2d%-CdZeuIRrBbE*1U0Fq8c-TG6A$APFqu$Lxm&DA1$ymS3%va zjau#w047y^*(f@FFn=U8xW=HT0H8b0PL?r9BQ_Xl>-~-@G;#?91gWRwmvG9k zx-vsX`V$yICeUB1#E^6s3pxwd&z$RMZMG-~0K^Id^`Gs?EDDVsAz%X@OtiR{S_!CNmx8 zOs0Mv)hPB@OxCm>-H?Ji>y#eKPG)Z$lasM4)_E$bxHGJyb(pJZsfye#`_6)D5XzA( zWU0Pwb^2`|E+R~$3IMtOvrl2?Pe)3Vo(f`$4^L4cU;tZs38NhF&0lNqJ<8`Zf?T4x z^PSpQ=pd$CK%(|WtUr%CIAlIgYCbeUnt)B(LuWqY^Pm$RNS5rztV=({QAeByy}st^ zJ>r8?2OiFgnfFil-5}64$3Uq7n#$%&eAqp)WozsUpPN4l6%Jc11BD3pD&jgLTG!^j z>)pv{@&#w7M*58&KGx?2z2Brgj$dBMkRd(s5ysr+d4~y-$V8HQ5kB-&{(eWUZo@7j zhz*K0+?SwmdcGk%9v7>5(ZqgNx9krGaIMZf@+jADS1K@|SLM~7es z*Vh$*g2ROb{Fy+**28b^iB%sf81RCBj0sclW`hzOyM&*>x5*#{`UgfH!kARyy_nF6 z_{y9CX?#}<;>dXT$S4q`i4x_%oyoYyh;l@hU#a{hPmoz3!bwP!Jq+yYKn;Aq|C+aI zFY1+S>-6J-sUMxbzBcN`C9=}L>qTe3t`{&sjT*#uq#!#7Z8`4R3WLLzdUI!Ard%H@ zW|2l?-OFDb4sj$0)SVi^2dRjR7UJ3(54fvX<&~sl#YU+AiP5l2dC1zgf#C-&?>O$sF{B)M`X9zf4M4(~5&vXBIF{9u<@2PSR z;z*J-MESm(`pg7pW0u43qq&OvQT|FY#5lxu098ZY-}nuI=eO%vUi#E=&}^k|IHM@r z_5qHcyJ(GYPQY$3PIZ}TeuiIE4IK0X5ms)f2txvYzfax$k!9gm_utNDZ^qg*nyR8N zH0>V427ii$yRZ#pA&7Bvo9$iekiW$7GfbtcMW7O#-}D_A`4z$(7s6p~Gzej$`;kRM zVvX9N{vo$&mo&G+f8|By?z6+Ohi`@9PmbAM;Mp00SW?AhNhq_uia&pU zKgOD&MylM9CC4`)Oj4Z6q^uF4^&iv@0{DD0goP@^RvVb!POaia^gl>`zg;zc7cZ|` zzd>Ki`tZlL@sU|oKD~2rDxVxJ!Xn+LrUr*zt3HMUARgZ_gsskaUaS>F#8DhI+jf< z-){0`sO!)#E}y^9jhkGi4%Jv8PTM4)*J>BEawpfO;(Q`)^?8N;{YEwe-@(9zNx0gXrT(pKL* zR2QQ=2K!yDUaqk&Lo^8#p18n(ikQijB47a7CyE=Ml+e*Bv!=jDT-`c*s+unRmVY0vPj7H9t*1e(HnFp_x>z>i3p!b(yN)sshubk(G_L0c68o>r03D!P#G+r|_M8pSy0r<>7~K z*hSm0IJk}Gs?6PaHTBV0;VOXhlG`AF?63EUsVY^S?nmoAWyVVDuvD)JM(2%E_-v_h zRwv%_?}*s&Tj}Zhsm~xF)>fYq+@VfVPj72|MzRu~l$IiBnAuU2w4D=St9o_+&r--| zR7JQYy;Nh%ORG>aXRpJ(iAEPo_6eJT|64l>^AyPvLnw#5(ev-dT2o#$9X}`P>+G z2h}S6$)h-@r`j_${|A8Reak@Fy+z=^9>7SHVS?rU&Vsv0is6Y>iw`x%s5mhq?ZXcC zssFU_;`|xTS*H88S>t$ z3I|<=od4{5zjaz+zN+SQ{mppnKg1|^YGJC_lzfvB^u=mayJ|K#A&iS1z}s#S)oV1T zO`=TL7#rjMtS4Wt37f6|Hb$v;KQ@i7Of}X`A5K6A-I!;UkXn$;M@&KCApBVZ^ZOFs z=V0)bSz5SrlBJ2VcM55W=(wO3-Sj>WMLbF5=8bP4mkbX)3*@epm%%{g(B$?cU}{{r zwjwO$xNq3 zK8P#+rqSTF@?i5v%!xAcLC)80$ZNYa#7A8A<257{Z-~v5E#nk{rm z3>@k+Y`t%LRYBW^Bxt~N#M8=A3m*y0l5+%%%}g#iOS_-X>6TTgfU_Owi!Cl}{>CSTo!jM4!l>V#5calP`Ur7o(&i(f)T9rKl#$vR>`_KtGf8s(XqJ@ zHdYc(pBiD_bUvd^e@lt%Dtu2tBSOvUGY05+fx1?Ry&D&$950;bRhZGWS@|qI=2i&p zI+DFRRyWO21IO!Z+j!0djm(q=V#V6MRQ3got=_HH?eQ-cJPz^>Ts(6wL#-<4`hPN; zIzk+mmiZU3uOn-R)EPL~oa~f13Tt$dwQ0EH<1L=zf6;cbt~;X+Sw4D6#l-vBvL%B~ z6O4|phe?V9U{%!e@WxX++pWN!J2cN(l$`<%UVC3`m#4qZ9%uAhGGO&}GbN zM10S1l*SMeqxZeIE|&1ZV&7qoB{-5crpA1!o;%Mn5+sA)yt=3ghGD|$`U^gYKehjU zpFqIqPEnd58v80}q6O-$!J(o97`e+yv@57C+=TUsRSX%)m?H+ckp}ALj~yb&H*}!S z^{`?$I~nQpGsa3N);q@#(*Anqdb=tM(a|Y{$R8Mpv@WQ&zIhdBfN(1Ic$MX6d^Y~p z`!Ulm@5lyMH!n#l{{ZisT2aTR0FTl`Gd?FmtHNY$a0eJd%#5g8(1;RUo5RDBSAlk7 z)Pt9Z2{`{KQ5O#r6-sv|4DoT?>E`p0cb$i|l*sA*?DBZ|(s*oM6Fj4d z5Z<34GpPHrqgf5JBrQRiU(TvxPc5s%lB1#!uJXJ1NJe!xSh>H^qyeSTLBUO2H=Zj8 zmWnn}dR#IwTvz|Q(S7@9XK@59YqvnwZ^)%jpsEGRdV9b+AD{V61hcZcw{?W*7nV;k z1co834eim1W}I%nfBDg@#W~Bjh>?s7>IXf{0pw&cWNtflkjarX&!Sr!B=N~J3g5bc`df}c{_Y& z54TNbs%ynoAjC-(x@VB>E@3Z9k{I3uYrIgXC0^~jefo6!o%J{}Uif;vJzR^QmEZj` zy?Uq%KZpYZx>q#+X6O&mx&Q$HACMzQZtYrXtmEr?d}pcyW>5^l(BOQ(o3>n_0PS+j zd;|c}v&Uy9RFZYP$tV$2S##FcU%}|Rs){<|7uNF<)tlSH`nD-whElL%YFFnUKkszQ z?ndPWN?*#Ss;g)lT7IpZp?mExV_u%_IYL6NF*X>c1Z_S6sI3Wp%$W)iY38PCcm9pT zrP9kk$hnXHHk$?IJ8!t{f}ahjEgZBX0Z4^7nxcMENu-H2Zhd>%lsrbv;JFmt)&&M! z@w31e(Yal`;bEC_Ii7jM$Eo%&7Ke**tuR!)@sz+vd|7W(%IW4>wR9mUpK4bp3^rKo zVl5c^rWImYGu@a)ym6E6#Mo!hy_s;tQ@G!lBKGW~Vfc4*f9s+Lqdk zT%i00Em7iDb-k&Z=Nl`TE%AxNU1GB;)ePiMs(GH_8}GFF*zEXfL1*cZY&=UpeytQK zGp6LegF3|#agzQ{*9z01`tbyYVNtj?Yyr1=BviI=l>qxx z66D}4$}?k@G6HMDoxb1--e}8w5PO7X|B9!@Z=hl)5>g2TL~e-Bj1RQbU*rn5gE%I! zOEzV}1MmXFj2vqJfcIjHcbe z)!1yCGY-_B(|Eq3v`r7q@Ahfa52mi4_mHK~&@#n0GE(0TOlU=fIg<*5a#V*m`#SvR))%^topN!l{J z0HX?rT19j8X-O+`D)Q+oX0}9(C~#Gi8NF_uzwuhJcb|leLX~#q1=GJyvb@6C5l}%h^6mbL`4iq&y-68T>_yw{65H({N2r<~PsBwI>y7WvV zcD@#;8vh+CY9H7>d(w~v3AqUQXeVHYLkIaiK?);{QNk_vGnX^lt}X8Cy?y%MCNQ_N z${W1lWXMhZESGswK0gcNVwiYxUn{1p&jytEOifHHd2v(PHWu-?R{bI}1a^Hs;>Lsg!j5L}+AB`f&I zVP2fIo%}TosFkxXHi5Z(H80gC6=W7g`sT*q3-~z`cm6FFq5l24gVDMG5UhX^$x*=s zgS^#GMTfI<>I6+%kzCu(f$X%_OOh=}*=GigV^)s1gaz3y7)u;P>O1&;F_xP?9&=k? z*4Ja7(~v>BlMQd-rX^$+4#OwywGep2v?h>DiOhj>G(r5d*G->ZKAjBXRLx z2{gQ+&pe+=mY5l>GnMf6qS`Judws^nR25X*mg<2&wzB*o@+)10EY8$P3NBd`TD@D?M-iXiUh2DAneTZy$oTiw zMwA;z?WfRB{Mfx!o(HJU8cI8xn(%auaINbnZRwgacUP9@dk^VAn6xA5&z{ToGpQvR zqg=kR@R9CkClarVOM|H0J`#L_1>kroD1(Fm|gkB;XjMl6`tN{nX;2;EzJ49*BmiSe`Bq!d3G{yblXjfX*Bw8s}-DS|7n-WWYcU3L0&$Yv#k&+ ztySQgZ6dC~`%?G5S1Fx&9ThLYc2V)y2aSFaMc$1CnG?M(Znw~frWHC8+_-M%4zdD7 zA$DKuL7&R^xp4PQaJkdheL-q8ynp=i4AsV?H!|Rjf<1Og-dSWj`NfyI?IqW@$qfsi zy;++&)&JpLyeBOw{Nb0;{n!GHI#-q$E^_~YqWXh&K|`mrb3iMCtP)KHU{L6xtH&+A zGe9=+4hZwh{Ru8Svt&V*EJlT_l^?rNW{~QN-r---0n*1hcAdfr)xrj&RU({rFCcfy@SXHf~lz8QV%cTG|4J-Vessr~R2!qy&BPNYp@@ab+r zuILw$6uI#adj5-M5Fwlz(CbK<21K9ju@&Zaj)S+CAzjCRFw8<2vvUC(G@u*(8D90(v1AZw3u(LhOmWGZxvEI_tI!`KNH&Oq4J7^Ey+r)Y;rBtU%8 zaXY9sg89G(Jz7i^Ys^bKGwmCrUWieC@7W{z+#+8ZXh3p80>#EM?SeORzo85V#9o0F z#Ef5Z)&4cBtGP);_a`ra?_Oez!89N5bB!V0FH(oRZ-eq9q=a_g&8LU2>P*L@d7lGx@_K&bJ{_CpF)IJ~6bfl}VKEtKDtc)9dn^mS$} zg=Ntf@M1c-IFMcm*qeR;e) zN)1vee0!b}ljGp@82)5bH?$gG>_BO+(DIXxZqu7r(SWED4a&~0zM_Ah$;d^UDBv?P zX)vAivF72HIjAwqlzd={LhD(o7-oo$O8efI!Si}R_&%g@ct{)~#a9g`0%@!|_W=#= z!jhDRr7#}fXmh#6w{eTe zKY1&&4h6alpyHWgLeHk$7`L$`%4y54u1zeeSaS~nZaOJkJbfPOTPkZN?FSoUkbabE zfs;dBaC9YVc?X25aDR1)$e=){R7n|x%k&5n#gT_IH-oWqU`ue~yBQ0&Vi(~y#_lsR zM~pR8tFBuGG8ln28I6w3l8raC;;A1!2|t1Xf56~4-Z*ih6`wqYgRCGkMHl5Rrieto zU9_qM*VPtGYxtV;vgK*rS_Bo_;MS=rZ9Qw7F5vXVs%Qx#x+cblL@BaT7P zej9Rnx$trGtMWDtESj2}k@m4q&wAa(xSJp$F1Gy`V!s=so(xlhXyyjHus!Uu6FZs^ z(#oTkugW4relK%t*F7o}5@Pf@9sj)X5dGZ*hH&Q*6bnNTYx%Tn|8mm(-ZicEaj$$= z=<}f+E<0$pWcY3rf@Mx5(}+%C@`N`jPoSwPj4LpN_JW_-c(1!P>eC^I)2-a)yWae{ zX77IEB9E_`)X!H_?B4v3oOHxjRN%*b z^-JK#hcfF+dK^tI8qSny;A{nw-;cfs=?0dR?xuw{M|g57YB z5Z=YUUZLbJKa+)Lb`5#5(=@32a?lG3Xw+(ciQiud;V=$Gd-m}I2YBrw*m=(Omg_64 zl9@;3Vn-&+5b&!LGe_`YNlk_?&y;HU<8gZ@B}xsA3e@zb(Y!Kmnp*@VRCVa_EeA&=;6Dt7}1LCe0((G!lUN@leIG3VC7Y#r!t} zBkGYe1*V0hKB6P+y5)@@T<-A(#^}VD--*);R^k#wc^erUrb+{xLsjjh4p3{1Nm}=*=;om>+AT6w=dSSe%epy1BEzYSS&_BB>Sp z(ay9{0L3wcT}`uls(<+U=Js$_rF{9cEO=$IAUpR2=H*RNWJg>iy3yAAzQ|wbF8RC( z+i_I|Bm{|X!CEHqeqJO)t(WwaJJ3cBvn@W6?-@vvdCj$Qf4%Au>(xM3(Tbvi#Z88* z@qWSrZ2AtJ`>tLKzaUX{2BWL@19Zt`R()flG7J`MOS~9YpP)Kr&A|Ay5*pg;yF4Q| zX}gFDn`OEBn}i8*+BkT6DSH1TSR4BjuJJ%C+Z>>~2weBl;cpXi%-wFt+1L6CBk(6# z3Le>GEhVi(I=_>c#`cx)Gb4@Wd-5r?)Pfi_?v`g@ZI>hon|F7wa6CTRiFEj2RkV*1 z^;>P@l~y?XSd2nVT7{oU*VH|$z*mtI;MB0~&OQ@?ee?iSL`~QY3NORd4f%PlL5Pxl zco`K4me$jY3<|)S2p7C?tXRd1ug%PgO(x%z0yh}FFwgi(N*;R<5biE;+K6V%SE(A3 z(AqhFZnJ-%fB%3jdF)4Zq<%!W{}`|goRA|3l0&MbokzVF?7l?eA8nI;D> z0!|IPX=E9)?%}G(sP4xrMQ{iz`ax0$uiU6~{+`!WPHbhnSW}7;o?tSxoThKZxro@; z3Aes8tJsIh9FCgEoUNHimdR+&Sb(GYP~k3Yj9$hM8U2tO66C z8%va5Mn5vb+3)B2H@NFx3V$Bnrb*9z4pDgDdIdyXI()c@JTK)GgVE*!adYq4_8W6{ zC@r!l!s3fe$DPDC$C~AO{57pL?EJE5Sc0lElOTdMp+ont-DUrrO9gk`#n|U|KkbH2 z-%e2+@6mI0DW?hHI&OsPtqVqu(V#`REy8qs4g3|ZBP_|$K{i{R`>|QuVLuXpJkKJe z-;hUGrzNA$HxknhYqdpMgWSOpvE%n^VIm5$?yhC87jMQBp}EoMaezqU6Co7(Q%ZH- z;lx7*jJt}ansb-(!*IO@>_n}<9BawaO-0L_4UgRrsEf@61?&9TRwxSDSb22b587fs zygA)v;kseBZ5gPB_B%*Bb&EH$VvwbGg1EoPA zEsT091#TvZLPw-I#`Cgc8b#r*et1X`4PGXW)bLNGC@#FT{N*JP5_%diCA7hMebJTeMn|Q z#N#EOUP9tR=R5Yv$$|H_Svm8sNXooZ@#9~(1*@1ffyGGEO;zU&e z3%U)XeixIVooW!5-mg)0Azf0Q?K2aWIU3G(lHnJvz+j!H`1KkA%pMGQ-R2M*$>8ry zYk{H7lN#4+tj17dt#&W9Y*%z)6tFGw-+v+9Ak!338#dl)dMnpYDkn2pcs*+y&MF{6 zFD3Dk^D^KAwZ2Gps&wW_HyCvo7r}A4h**!-Ifa*#o(50@hk3HU013~sd2g=iuQbzL z(zDivI+4$3dr78|hJZXBXjR1?jVF7vtXJ#I@u(|YUnT20KG&>~@Ux@XhBRp6T}SrW zgUUiIFz8Lb%kwyQQ5~v`DUn`n#(8H6YG{|Qcs*_H`Ht?W)r+%&=ys329k2je7Pb`*IJ4qGHYRn|0O7Va>{}$2ADqCHP(__n2s@mLA`?Uow`&j+z=tMy z=!YgHJ$ZuXrux2r6zF1e|6mSqu_odQk`;@k6G`VuTZY0h%#PCR@jc9?5*gR0#MjZG zgc++i$61(a!XpShYXaOajSAy@EFCpzr^mxd=Hec$aZbAOgW6D3vzUO1d;s6+MK-rd zyU(HUN;zrGuzjRaL}3Mgom+Rs>ke@REiN8c}HLMZz$ zTYm+8gy6itK z{*&JvDZ-{RPWo#;+Wig_g)kUWt@wtI{kYh?9Kg_UVZ+RmUQ2448=b$=@J{3r=-g#0 zbnf}+z53C#Vez{0YWB+aIQ!6i`zXUrJU?-pjt%$&^W3s9vZR~Mk{Esp*K0kz!82kQ zmWK8Tbxcxw!|(CnwhJfTTZ);WNQ7`Lye#ey^(Nj9kkzGoV(fqK(QZhTQoOm5qNHtC zc-gmAAtD2+p#U%PM`p?Y9Z(tRT1i&^Wl#io=Qhh4CHw8_8AG?My}+33%2ZcCkZ6h) z9#RU*RU$nBXHM8qXJu?6u^p$cQeZ)H<783{gJu*SVUIVudd6^4F`$CfqtQF|8?I}j z3S|N8q>#@)FFfWL()4Q4jDp1jOGe)<3PA%z$Rm^sP}8etUz1}mQFaoUp)Oe>pI@f? z<`uUa>_c_NHm1RzW%k}D^UT`{&O)@o(Xc|DnhjFL?*=fckU#5sW8$6)?rm%^&6oL~ ziW*Sc07N*~il@Jyx1WKPb$!HQ%5fr<&pAp%im5U~-`CY}Ys2m&AszS??WX<9@&3fD3*e1iruE zgM!pRFS;ag56hnd^o%Fi#&hcK=iQwW7uGKo4qS;UbJq-w7sygdLAGx%sC%}#O=!N$ zWG!>8nuL3&*J!iOr=$Ny5HTF&v@kB<)la3s$Ed>zE}nXr2|UXB&#sHjPoAp@nCMTH zD>ff%L(J?M#C2cK4Gq9~-!{Qle0E5xq&*&CT@*O$e|=t6XC<9CZk^!Ab}hdap^izJ zHwy1Nv>1RT2t9>2SoZ3j&rgO&^ye~ggC!D?5-URz9GAyY5rm=k1m~V;?*hQUKJO%d ziKtA2xq8vKhkJ)GMa?h7_Rl4#o9o-V|2sW*xJU-DYqusAovJO|F76yc3SAWa@(lYb zP*J<4*#NAawWysk1q+0kHPhrT*R=VkCM^d8_6o>S|z5#+z%!ef62xu#N^>$XoGb)?65ykI`G| zP-R_)a@t1 zS94!`Hqtq8j>)vCanV%r6dk2m5o#2*3VKe7S$8+eR)lk zC%JZ|p0_ZqQ_hs^3b~eByJ7>nV3}X)@uJtX1^-}Fkbq84Of+4X-@+Lf;drD~K|7Z{3n;@> zv_leobjG{3^W6r*@S+bSlC2Yc3>b1>+b1|s#>T@N#SM&J(tuS#NrebJ1O?ZpMR~ld z%!#5ke&RS0Jz8+h5|e9^=;N(Mj+X56%2*Gx(32zWg*6N8)DTj*6e6@)r<8(N&;6B; zoy(m1X3mi?ya*$y3&ADm0ArR_=oH!@G$&BR&7Up5{vY&<*WS5!uxNeDqIp^+UU5i%MA;c?3Z0{Yj%C9GiIpa%Ll6%tDLBvD{h%0@t&sa)BxfZ@sM% zNu+v4Ky^<$nbK9^`Z*qays_D!dh=FJudFY;`2_xo8eT?M%mk-@;08A^aV3@{z5)hZ zxzUM!6vE<~*Vqrm=%J=U+psUc=Te+kX8%!vc*}C!N157R_!>C+J3K7y;xBzqf`KGS zXY@6Yl+Q9QJ4wTsUPMsvfC}=KJT99`K@i(0T-rZ=g5Bp*>Sbvr zfHPqhz@a5WT~aB~`fT`d2R&pKO6G--04D$QxWazCG5-XXCTD>zm$J7Sj15OhmMS=& zkTwhlLZ5&lXjOfs*Az2o3AD=&XY)eZGEymiUxNaPMgPAgRn3TW&QSpoL4*4k7*two ze=9fYPPo0^+K0aF&vdg^F*=}vjqc7C@)0G=nMQWw%#f%t}g1*MFWB0P`C$ocX#)~ z-QC>@6@pt~!7aFZ@Zj#j-GaMY*LQAC|J{AP>uNo_)}AuPI|>{RIq!md$|f3Rer`5e%xg) zq)vEOzA`DA&&ima9SSmz%~uEJOoUei)4-#aTM*@huFo1hO6)7Q1CI9zc!HmOeRGrQ z@j$K&Atxd7b`)}0Na%jYnub6#A^+dS9v&d>uhClO$+t{KEbH=BTIaxPd z9=;@5ujT}x`!)S{0V@C`E9-GThI14W;N>K$s_^#> z_5-AuY8*vgF>EUa2I4=wZJlYBZPA)FK30Rp%?PT_S9XM?yWZ)d;wbbp+;D5KYfpQ2 zb{&_cd#eNYu04|$wo~3l|1?VKl5iYZfD3LY8+knw#Wmz72m2~$G|LybnE80901eTk zAI^h0&y+5keX}e6FU3+>s^kD;pYMmUcP)~R)W-(|5vb6MbY;NFFOfVkEPtiBvrXg= zl=3hWfIAQAqOu!79y^{Md(Y-_1{-;KHjBJx&j$|IdnpC#A>xfc6_6&|75U`V4c6A8 zJ~R8^cfk-L$X2S#Ew=o64GP>aD>^Jxry1Tjr}4KIH4FR7jClT8@dvSd3|T_+Egb5F zJhkR`QfsCJZ-cb0TP8TfKfhb?O=2=Ev#O9tq5&mZ!v_t>`=Q#JFZOmOSc#lSi=Up? zEIc#l{>Wn>vg;&28(Ss_Ykq?%f=Ww?X%*8}ceiSQstak&9~dAq#K3KG*r-x&Uq?Z+ zv@whEC2tU)t{{1x?$H?Fd$^D4FNQ}X$Ye(=s1z*9W1{rOUzAnj12K%D>WLGR)@iOnLve>Te zBg&$C)@L_D8M*8i`XZ0nR&$mABxx}+^8fANN$G53{3Ns@z<{T1W~JWjOHnrSJ>|WUz95bAFR_=lTWRn%eoF<|Tam%Ts+Er#1X2bXK=(_IP$|=%WLYtA*IZ&FcRx z^y&eIVbfWZqtNwOnhYMFhE|a(;l#YE z7o&`j5;E19O|t9Jc43hw*Z$G{_IgG3ct;49D#?A7;D5{h=(|OT%6Y}R_)s9=f3o~Z z%-04%-+h{$k&@R+Un|lJS0UB**s=T;SC!}2ky#ZNWw=|J6S8-nhV!;|{Vg)ah+8-% zf?Yc9mBX{?cZ?XbkOH@U1}@aYf`ox3a=fHxWmE`0Io2o)^*SHW(-U^va!r(hU1Ug z_a1_#gv_KO6o^~jGgHTwtt>1vjeTgJ_N&G;*Y}Wy4BR46yCFGxf?a)j**hH#oCLi0~ebhLo(?uCJ@RAf1@{0@JjgWuTwqNt9jhLw#AQD#q*pU(I6#t^zGt3J&1~L6E$*bKd{=N+5 zQHoS^IQ(5+wM%Rf9aP=}wTKbn8vsVo&&ya>r}w)4Thh2J8{Mjk8-bs$I8^mNl~uRD>0p2yG7J#1+`@IYp5ZQl zg#2Xw(3U>7l+RM=sBBcHlPE$`xpmY^O}7pTUhEQ7`WoE zV;{=Jx!WHsDhVdbVq%*8NM%d8fBfw+A6k(!%nNw)XM^zetiX{ zYC480WMIvvezWFb@E*Ty`MB1%eA#1p2Q~hA)js-%E=a3)aRY-SgG&pF*%y z1}}s7+`c(`!e=GBPi(jL5(M%l?~0Zn$rhN#_8$Xhxcdlo{Hhj}(gbd;-lhS&0<@Bt z`oo)FRi$S|u2tZV!ISvCR@AB)=f$cV5Q-}@{#87&KN3PJ|*Q`*+NiM{@wxg=g z%+Wyh!nuWtAcZKO(6^Jqo39D9)G|<1me-HT;d%YhI1xdT&>R{I{iqe-mR>-a))G^% zQ$Z2FB&_2;l=GTZXMJPh=wxX*ELIj-=gE+opgRe~^?iPwU8&vbnMQ6I^zhRn^IzK! z*0dcV4tyJjQUtd`;9-l=&jbIb_hgM)XO04U$urHNqSDpK9LGk^GG}{gZb)0tQ$9a% z@h#k>F@xc|WyzPyY)74+DO*xKUGAdIea6A^&Czvg)vgs{nfHw&8{v+Bo$fR$2Q2fj zx1K+ps704?{Ty& zI`<8((A&H zN$3+1+G5y|Ky6a}Y&X))rq#g^T)n3s!|#r`gK=J|FR{LC5{$?a}`AY-C&poRa)Pi{wYC zEB>$l&eBvb9Ae3DyHLZ|wCt!=R6$Wuw{nXca`PIv>%p8qTA@0>lKmM1Q{iHX@V#%8 zQesRaHq`q)rZK&Sv(`0afJ2yzV&EGCgsDXu5ZyRzCCK3xc34V#TE&i2^T`4%p{CWd zuf42SQj?am`qvKQC*Qr$C}b2rI20RJs6Aro?_JZFR2puI(kvU8f5Z|5Tp+7JKWAnk ze$8kfZebE-LRiu2BrQy^yj(gx?g+$FY zV$!MUH?6B-fY1-`z9_!{%_lFFzzc<0v*5TFf416VYcIdyh4$=Mc0&MJ>+0m;<+&gkkW}A=ufd8Mka+Wbl7uoDWw}z5tkpYhl=rtTgrXye z&V%vWBk9G)(zBGC;4gqhM1T-$3>Z|R6Y6zw)K$hNP9U3La^^t3xNW$%v>?+E{DvbR z%RFTG2WGFCJ9&(NEVn8fGlR9*y__^&W(v}nW=qGiTg&7&qBwl{mGkm@UI5uxsBhQH zF?t$Sq~Mr+05Ehkdy@<17b|O($F3#F+VFCF*4#-r^X}AM!R=P{6I^t_@Vi=_l_|;j z_*G}UtD42wyMxAhQLko~`gFq2*wZV-bo)KN^P825%FI6v3C^omL{RlY>e)0K@9?)d z$<&Z?w(>+Gf`Z%IDjKCl0i@ixlv)RRShd=V#XC zaq;rzP}lgiVeaJb${LmNxZqY?l+QZ{n-XO(N1LJOx6HY^YV0H@D5^&UCO@VwyTPhW zJq=n9-(DE~sx9AzO$1Tiom0LM_$-JHc4fRYtW!bUdHTJNaPpB0GUgkOvyGi>=FkGW_d&;I0max&A1Q3O`eQaV?M)i z+_Tv!RcYwM9tvo%BE+5yZi^?c5`k;70B|Omg2UcPoJhzF2|q8=EDq-;YKlH=28b}z zY{uAgnSQzKJP9Y1g{`bkzA!!)$o^QBoz2Jc?9hXLGOri=L_g~$Ymq3ZvD=dgWk@Mk z4cZp~Sh&yl7-4YN!89rcOt5>3lO70<^n*lrCv?M!ghLie;=v-|W)=l@V}qMd)$ zAMqmJ?FuJl&#l4}v~Y8Y7-MG3tiHLKnIhc9)XdM)Lq6EFz{}O%iloWOE831g(CHL& z^LECyM;~5oS5T)jZ<4ELADKw)qP6fJwMt@~e+$n#!~ffX!GUe8dX&j!;7tti`-mN3 zG;Mk=lC$%?(0O`Ft_RF0t$uUF^bf&H)ibR#27kfz(y_?`MyW{k^Y3jgi^o~vf|x7D z9f%(uZXUSZ8rz1~cMDnVe`na}1x2r-8LADLUWc7Xd9ir6#IMBNs=7D}@ae7Z9bF;& zzQF7e+EA+mhZInf*NK;1H3j6VhN#Zjkn>2)wXx;fZEq}-gEWbS#OhmKX!#%$S^9-e zNTAS|{Ql`bDwPY=oJW3(5YAe*9*Zx!zqb~1<0?DZJsU!FNQ-7K0{%HBeJ(pYU(|5< z=-fO0hb!)YL|tD)E!$i$Gj@%ttuJ)aUXgO&Q5W$Wjcab04JZELlI66dtFX$_cD2)S z4{7CSsU#)XgHO~%%=hx5E%}sdo@u0C9ITg~Iu4?GA@}(NS3Dy5jr2WuxaC0^NNYwR zX`S;#EmdQfcUi1~-Q^QCi`)RAwvqeWm%LhS-$wfm6ediGe%^)Q0%b zx!o~>n6uEM4EU91@{!GK8~xtq-EO(urc@hMdCI2QE$aD(8v=6!ydKWT^z%T1t6s_0 zCTJ6+{y&rEb{k!e;kzd17#p#`s1q6-xo7|;TK!g31JZk32}aGeZrAtomLO4}Kbxr$ z?-dbfvh&YkK%7}$TJd|5_n}xSj$uHTQ9Vz~4WHI#dZ4N06UgoIf}fFT=cpWCLlzY0 zFFUCsPuiK`a?73kaLMgFtd~jiv8PeL}?CNRzSGK8}d=UV!aj-br4lN?Y*TmTaZ9+hiIxC<$J8h zs!wduykt4<{bfUex*-~lhTSZv6!Y=(kFoe@Rj*(I3;dwK9>p^ue`5-OC3c6@*@aCPJ z&AMwmk$^}*RHZVh@9KixRT`{k~OJ&=;(T%+u%2&qO`3LG0)5A66^4+y{ z`U=@qLm@NxD5dbv8Qz%g0L>L@j(1^)D5}|*;%WM|y?ZqB0_J7KNhp0%I`#&6Jhm zE*A)a@o_=KNF&ohu&eBK7s#8%%l}I(jD%X6vK_st{&yYi4|^WSpMElRO@pk?05zkb&NY0E?0Fq#BW|9Uu5@|P|?kH)r!YT{iLeIKw^XO{(10 zPnzq)XeRX^8}pZM;G+`t)kJeUL48>?b1hd zT{&7EtRawAf~`UH@DGX@I#L6DUu_gQ7S%hkV!T4zH5%0WmMY+9rQnlU(oLG{VmVLG zqv-W)3~tG8DEL|xu4Add94S6|6E`6KWja_Qy)ww_X?3;rg;Q5|*$m!2qqg&Ohx$sM zxpe{DI`xTQD=3C7J`xeJ?AG&(jKpU>(%MIgl$+T3yO4e;wyBvHXEIYB-mz1nb_moQ<3L|AK6qJtB7G`gR_c2eVMdjUUCb~bAZ8m>F=H*&$?vL6BH{TE4Ii=@$OHvhBq#(k5Mi&WgC-$uVj-=*g< zW=BI<542Q|T=ya4)zm)2H8u2H^MW#5X<8h1&U2}SzsRdW=dpVmGh&DDZzZ~0);611 zqzP}X7}xDG8(#ExPyLEQQ(eg_-UdG-1#UXRS7YMmlYe z2oOM(w`#2-C2bZ2U_?+b0TX=wRB9g=zvHh2bRt?k1{lWywa^Om=99S2tIgawr@^fN5F0*gnl~by3 z+#v_=Aw;g9Ir;-(3?0B^k?DK=`hde?pwRV%@a?w8x!cI>%+ZLD$so%-qrmJe$KSkBj0rJJkcE9&k zuoFlff>k;k+h=DTSN$HN=H)XT9y}ufyf4V{frY z%2)!}Z0E9#_WrGLm`KY8!>!Tt`Xlq-Z0>8ELz=|YY6_7U3Z=>U{zY0N;@RU-FeTI4 z3Azer@GCx2wg~f!d@qD}xRbMY!;PbVY1fR114~7m1$zc5glKjDP>PvTGKt5%eXA#B z8?~BptLk{Xihde*E#7MM8L@L4K2$CD7;Ik`DPv5$4=LAgGUBwbwoDJYS{-*D-0KAB zg>%NcwT4-E`i!jBD;?wuy=>6FaFp?atgG3BCsq>??oBtaX}c#|C6g9O5u<$g$N{#Hms^_)zg%+#Iiq_?mr_*2|OM}(~ODw&i9XI z^njKSX{@=a0R`C?Ov2JRJ6nsW^}Q!hVC3&k{=woY6?T)mN5w+qJlL;0Ke-Q^G^ZG3 z&7w{nU>>0E@*y`ic9h&5jlhzt{mndHFGVTKxYKJC7#5fo z#njWJ@6T_$-C$fi-dN7i1n7ttVuXJ)cp&<9YL>2^R(;cH{A*82D>1mxbbvk}S*|`5 zrQ@cso;|(@@ea-`Hw1J;&&}$w4{rQNXeKxEcQHx*b90y!qLc=PeY!GZjTsV{{D*JZ zwTwzQ&=}+ImH(Ubhuk%4DFOci86S__OkD*hWHA{PtG)Q)FyRP+cDH}T&ZY?p%-F8F!xNcNT7K0SiX9+s<8Br zef)Zx$*$YZ1@rLci$W_1WMF24=?qwlcz2gUMbq=oMyr|Cvcmrqa zVVBHso{)W{tt!?mjoRk=WFozO`zMRu z*E42|M(sXKHID#up=V2(cxCE~MvOinQ!~+s7$a}0DrP<>1*1RiEK8Ow}#_Op&U7FK;X9!j+8|173OF}TMmEW`~$6YZwXP0wOKzSIw7FnZOu zY2AY__aJ#94zQg`Ra8Rdjr^4fuuIxV`Xdw`ml(zQIey23c$s%FhfyF7kNp%vZN2i) zWydrvo<&gxk$Gyj`wG0O{@QYf+$>ZzYH-5nMsWZmtp9ldBtav&vn_TUwWY4b&7@ew zQq!jUa#2OrSO>XEq-5i0oHS5;MIYKXdioqDdRoyEK%Ts)M4g#pfJQV?BjEH)s`RAk z0$dmFz}uDWJ-lOGRaV#Fy00%1d`=}U-EnED<^F|h7)p!u(;IAALYs=_G|O>pl?jI5 z%f*_+b|i1$!$S(F63_YNter~uA47MU9XJJa)z#d&ThdPfmwRj%67}+9?q}dWT_jHV zCkWiX;nx5m_`PG7lD=^Du|=51(cvE^m36B21^?zFiEUyACL33@wW`d&_M%xmZL7MT zI+$}A$ZYkPy&A_=wYuw!^X9b0=DoiGtnd|R#Bigh6C3byUUVO^QQ(DtRl_qzF~2;r zxT{5U5E$Q1o1s_G{qlmR%c!*H#g&-WedDLu-6mXgPL*Ork=YdKb*6s(IGPgkWnn;y ziXm>Tj4@S1Ls#z$lNxixss8b6&yYyjqbTi)Byv7={U-zUL2|5vx<5lvY2qpzV|jQ3 zHEB~&P5HXJM&4;xC=yHPj~EXc{fHp*y(ZBv9yKRAnWMG(#qhpU@(Q_uiAidCeL=5| zi(I*rW8k%Er)od;%$W z@gja^l|qp-)s_CdV(0S2Bz^}qxqC?)s*5Jtt9B$O5v6klo$o_WD@4i_uFmIPEYG0( zCYVA4myXSU-I=)f^gnJR(JJCa(W|q!dR0_?x-aHJoE-k~9(?Dept*Mg!Ha{&_ zLf6SV6t!?|24yfJsz*LeTX|KO?A}9+I2!xz#H82t5DPNE#d@wPr-3VuzbKjCWe1Z7 zZt@Uvcfpl3$dz`kOFC9nhV#xhudesYCq;&dNihRsM3vu392FYm5Qj`xowVo2Uqp@G ztB8zUg3u??>N)pm)Ar=lVh-UFAVgikVqBe1S6!JPYQkaZ7-M~r6?278-O4`wGpp1WIs2-lwQ0crHTA0yjaY-2aE1WDMAkSvu5vdMIfRBdej@gl(|EpB;ly@&>jO0_WFi9&8eaIl5$NQxNkhL#$!kRa;81MU%_L zF<2yKrwo!0J&Fjb6iP~@93U=}^zLy-J?)HMTVJ)j3s~QBKmN^BB645B^N!L*ZO;RA zY(wj~7iMoW+~ao>ZN@qT>Y?CvUn%H@~ISoV$TEM>&?qRh;Hn5aqgIxr_q z0~YR6Uf^!yzpp&Zqe?`2UpMWx@BtmZ8s$p~H5f2X2Kt95{rKRs)NEma0&=gR5#wz7 zemQr8+X4qL-M1{_cs}fkA?UT7TbKX z{e}-Vk3oU{X^+nRH$ffMHBJWI1NECZ^<>#+^haD$doSJ*fP)6oXDrsY+r=Vy?pcbW zh86Prk{gtZ)c4p^*V4f7L+UO20r%^(Hi>H7MM_rce#lK|(E;d{2^v+3+{1iZFdphx zzrB>;CLkP}eQfa6G2|30p==zKE!qb2smhv$>G|Z7wvVCUt8V!Xq zdYL05QBVxMO2_XkOyJ4U#Rs5PE>aH8f+%D5nzQo9b~ohY!kZCMR?rfgM&hVwD$LemrlOXdKM}MjOt90$ryRzHbtV2KffulA>MvW20)ie3?Mtv0TgogfiEwBqmp@}_|BsNaCG zSS}0ovZkAFEl%C~srih+^xAr`9K&L6n|+OJ#l+yf(5j{9@ae;7v-E973Ih}Yq#n3U z%~e~N=zcCK=bqIUE^ppmjTgDu+lOrj&a9s=(%`FK zQNliLoC6)=9u?PA8S;5^HFh4M)_XsXG_+)HwbZztFKPzyWZxlPt?;lQ;S; zeY8!ejrjAFyZ+8op^DxSeg{mh>(Ib58v***E323%Q3n=H%AhqJRFnsxT{p9Q z&sXca4T#B6Tqzp_h8H+Qb9y`X{+;63M=kH0V*J@v@R7-K8y7Lr693iuxa9d%)hrJU zsbOYA28K3fl{c~xh8J3r9eROvdQcGPJfl>!bvzx9SfAa(f1^StP^D|_Et&K`_Yd^6 zrM$CMCXY8iYEnW@uHwCG|BZd@=2JGa2<04kx#qT)bG3)xCjCf}?O9ck@An*|6yZ0p zxzGBRYI059>h+l~k-Zh1U!_aCQ_OH_Y_3U-S?@P~K$geL&tMsG=G46LQ;HiMw2U?7 z(Q(EfY4Ut<#Or(YrFEbPXWuDZL&MXH3}@5RKOO-lz%a? ze0Jw{OGq3>n5R|d8I++?PSMP*R3AlQ&Q#|)X@MlOymr#izBAL6w@|a;tNtA`(nYVm zQ@+r>|K80&(I{hUy4DSaBL>oy7et{-N%xu^rqh}EL%UGQhH|U(kR&5FJSXt)QiJgS zBbtEqgSzjr14c#g3XUpSrwVwo1C0Jg6i2pP=|qjjcN$c`8i1`*I&NOkdvKlnReAUq zZKkq1P}rIl-^lhzlo1~IHf|Je6Wrz^32fj?4c8p z-MPSrSD_0?2)}X`kgEVzW>TY0!MgtRq_^q-mNO^S;2)YYk90sye*M(r{ex=s+S`3s zE(%m{QD;hcsLtSgBDM98jPG|?<4XkurIm`pWuBJ9U>7eJI4i`O%RXQ)39?cDv6WCFQ^0`E1K&AcGcp_lYYQ z?t7C`B{vB!FwIg#^wVFRUzh?m^U4ttxP=~-1mW})Tuyg{D;fBL}dXPA}CMG{(G0Jx@_orxF2nn^<{|0a4<+Y`@hpDzdRP9TI$co ze_Zm)@oHvvih~^xk+G*|(k3j!EM^9z{j@-4EPNz=ASQ~JM3#GZ3<%JfqymsR?XO|= zxjFSyjkICGL!bN`3y7OP1*@95eRbxeN6;x$cb^02FsQzrFzxaU|UpS@%U9kRAF8vrbOAIHVj-2qI7Q z0&J+XwLS5dI3Xgiw9_9-+#EM*$b1(|a;a9NQ$jkzybcz&&eRGOxlhhE@bTPS86&p) zZSez2%E&EN!d-Gy4~QjJ{nct;to7q#aCGYJde@Gu2MF{T1x!)I1rw%zjD=AZ$~jkuLMZ>*fLb$c!AG+KUq_PE}BYh z4$VoR!?TQ-pL4#QLKZDAk9Zu(AgL0zp2g2uSltNnRM2$@*_@Bf>kv<674@|v^7J1n z99*jtP0xxRSf)_2cnuca*Pz5Eg`a9wVMCN%Lg8$qC2I9-e|dxL}&q#W#DzU&d78K`56 zYg4fEJ0B1vO_mI;fMMynb(nDSRA9{TB1F}&EfSh23QT+rcMCAQX~-TdpYkz9t*EzM^BFv!);zwj7+Zc{ZX!P}UJL0nAFr#gnB z{~I2x?%EHJY3G#xl1Ynflx0TY27qOVmO9X3g!tfe_3AH5UINPqep0WFMN^8I^VWehjMDD<@}I6VyUkNr&=Wy3rJ4_`McH~uGQ4~6f%uYrvU zplVkfTJcuthAD>gK$kHNo|h*cp?YHmF;Oc&U=AoI8*>x}Ox7{ilT*0!S2G2(U$h&< zfJ2=P3A!PoI;7S_7j&+HKdfH=GtEw}$eo(or^o6aSOB-fS9R!JY~p-QhRh(|MH~?$ zw2%*wLahHf)UorPyCRYO5>^1`Nox1&y|8NDlT4N+a0inqPSV!Y<1Wt~B}SuRT`Vqj zvK*WIHu#m9-z|gY(fyD4wpj|o2H%als+pZ8Ew`H#t>0S#I#NMcUP#E-{E&P&XvffX zMStHvMEfE`ugp!1$kVwl4t8-kctBanI)dMlGvvZ7Esp|Mb@tcZ+aw6^HXzLY9{&(C ziqX&DUtEyPFeGBxcak|fdGwHz@Mxn^)hw58wKipgLiO;xqGE|kL=Y~hwp*PFaY}Luot*EM7-XocmS32Z%6C``e2^QeaDNa-?hJ)r6 z0h48k03sOo9tf0F&A=(ESD<~yUmT_f5p=cE}3g>h3+NTA1KptBfA|eHt)*r)sD#(t$|Rb^5>!qd?U?p+yPA z+j9_NfQ`YPKpA$5elh5+MVIhOx7y|4naXlp@4PZ(SM(y zMObQh#Ugnsy3O{TkfEl%Z+2?&=P$$yo@6P5I~V{kTu~mwjnh}B%^qV`9=$dp6Vo@= zAG(o(GdCOTAL_pb?6qPLS5?>3yO%}IaH`{_OH`fe?0h5RBN7IvypptS0x|!pBXMVH zV>4%2+a3+Jfm#-2WIh%oATDWZrSM8u?e5YfdEM1jsq-a&U(}VgN^ElAkNO!zOG^9A zA-$;SpX%z={EFy0`NN0|GUJVp&_x`s@>jxN1m9|iOg^hep{uArwQxshm*S;z|Gk9~ z&LN@>XLKBoS<#ndMRqjpxGc6M?|%)M6fQ($6UH^xDZ2S&NK1AqQ*j@0{SqJSuN;IN zJ^gP0FZ<9v+^S-31Ab88tCbz-nf_ChV#uOTS2~=S&A)NyCa8+}y!swbcClWlDo;m- zkj+#HY6)E+BkM_{S0mE?6v|E{=`%qk_wq74IjtrcqIAM{T6)C;03eHr=lPwYB9rz8 z2;hVdhFx_NWZ)YPZ`C_h=Ce5T5g(8@=Kh1F@N0l2AQ~X@p``a3QN!>I6KoYlDk4)h z^I6xH3BDpZrHmU`?(@e0nZ1rxg}@hcylBo0w{K%rCNA`BZ>3Xa`0rn4i~G7BbGI*F zO5cdHE$lD0E>^-%g*x|Enq>6sk-Ita0oO#r6nh&j#`KE%nQbapkWBe+Zi0wIdOja| ztgDafK(^jWnqMqTv(wRYw+owo8C`dqD$^Xne#-ysKQ7qvM{bZgJ+}uB^-z6i@M@Z@ zjvM@*)@#|}?Y8c}_1bAeF^n+dG7u;ZhB^}M_CmI5g#xmnFJ9`G zy+X4q0YRie_Zr)J@1Lz6Xm!wn_0*PYs532nEPdg#NjSK5ldGypV8W|njETRu5bQ03g9Tw|ZEFvT+}PACMAMRo!kMXzQx#o*Xv0X=--z=-1zW{- zYy9{Rh}T>Fwr5ZLHKM&<(OTJ)A1?Yr;D5}o)K%NzlHB2!f~j4ty2slc{A<2H zT$S!3*G{$9E&up&*+7p|R#4IvVarjKDn#`A0irjrqNFJX6i(mXo?K)fHN!obBCXid z$R1#@f1wo}jGNCUb%UjtOt*lv-5|zr&l1cdl$GHAwLmv&ykx&!v*Wv5ENcNK5lpCQ_duTc4=D@ur4Rt<8|NgK9UW*t+n6=a-LJUkc>GV*6M;ncf*A*h6+ zbwX@L4B4Jw?O_X0frGWitbeDxBvP*){1?#VHP+TrA`Cy%dB3Z#m7@Wc^yfP1wq~D_W zcJ;c$r(U(!SYHJgB)Od?m=B8YSA5jCrUN)RxVpn;%{jmX8D>WMoI~3$Vgy$EnEf4w zPO2KVqF&XtrNGE0i<08VO+&03&%W70XwblHe?aPIx=OLyev2nzle$o@UwBLLyRJl~ zkN1s9R8uBOmG1#iwCPD^nK%@*8Du&?+j5m=I4Fx`RHZN>={YdF`RHDm z)VH^yaX13Ldh|PBReS*xR-{-Kfrcs)`pCZW)X9^=tS{dNRiT?=NVd(p=Jsrc z9O7QE2nQRP?%1N@ezn}e&oQqvbKGvj7;U`__>7GYk=?p2 z&%)Luj#Er={3-#t(dLX6DbUPEQRk4Qp5_;yAZ}AE4A||t*|*y_eIe>4{MsyMp4ZjMdJgi|r<%y8O>^2Q04eJ+m&}5sE>W zaFQxHdAF}bzVSpj_8I58N6Yw8#%+wLB>L94QYzV7-h|Yf-_XV$GF;LNhT=!O<3h#w z0Nv054??tx?6;@KIDc2;gz2%~^f>qzlDJT7mnOz(Ix7P>6H`QB`-bM}>j&(Ty3*x< z0JMFZ6Dz7kr8no7kx>BgC>g&ONFDi@0ODwrvt5ktLJfKt7x~#a8c3C)_GiK>w5Mrq z`0K-*M>>FFG-b%?PThKp+$nI+t+Wnp9D;*LwL}4I4qBz|i}lPo_d`5%onzrQYtIV3q-oL{&0Ow( zohb_aogW*%HLZ933m{KZdE_oyE*ri+GT6MtXqCj4n0lDKi4@UVt79&usJsk8su&%C zXw;K<6cbFX9w)i3F%q3z}3P zDqYQ%F+z(*@(D>FXX8Cq_`y}Yf2CIz^SA0#nODeM<_2C|L>r9bRmUIP&f5a zG}#)D#=B0P=%u^)zFZf}Cp_2-=db}-pz^M_YEhmjq(%$DVjbz^JEgyu`hf!@Y!npV z4SJ+Za8*h&384#ClKScds(@CMim%h_nSugHsOsW9l-B50vcd<72BGnY6&eGASYGh! zC#JOYk3_GDrjIYS8w?Jd!SK~gEN0Hkihy&3weBW=QIg{;EmZT4zetKn3_)?0A(Dtc zDr)PY5Oj5(piNVLUd-~^;n(ksbgK=q1+ya__5@XW0AS=(p9vo)uCI=H?1I^%x+BED zI|jyE0ZIZJC?amRkU?+$I={Iz(aj;mL62Yb%Y&pXDO$=WT=mCf7;$n- zEgM+6GMobn2^in8`7>v>3pmE01EG#9NXNTN(!$hRcH(%G3zN}ZN&-73GS ztc?6=mm@e*efalv@Bic|m^#J^)ICoeEYnNLxye&)?B_b6GDbtA84S) z>^C=NbYc3Y!jBjoA#x3+k5BGvDuwgcK&>>`(4?6LYcO&)DLze_FLDqpJcFOO?!&0* zmLchWmjk8u1?vB2u&1~kW;v&+=wrm+LAq+`zHDr~_$Z99WZRt+Y4O`ZN6BEka$mZ^ z>ZbTxt?o3OB8sgkkI%nJaShQN|MU(9rqPG(QuNP#q)O zMxh21w7Bj^qco9QoOrY!b%TJgN%h&E%)L5xx2?2m=9D-uM7!#sJHWJUGOh1mxBIQ{ z`e}5@Tgrq*I%Vo9(NNxKoY4sEWp-ZF-{B-Rze=&k@%jv1dB1~CaX^;xWeka>3jvx^ zqP*L2^A4DUZb?U?F#=V4{N*1=wMJ#+K{7yj?XaXoS&h1r-B*hIRA>T|W^ZGG4w*as z(E?DI#7GVkt%~nZBsKi-Nvn`rh)?E9V@%E?DsTynXls|-^udJPrSn$uVGQ%f7`}KPm5G;KwOa_nQV=|^+Kqd(vgB)qOYV& zKJp*q0D#6~(=J3IYp&v$?z~IoD(O(T#7+>Wa0Bqa$U3X2xSBT1;_fbi1a}Ya4Fq?0 zhv4oG4Fq=x!GpWId*klz8r*gIyPNsvt}lD7uG6RLt!M9tY~Z+|=^{x9H(FTBNj$<7 z<<_%up4_aym5x9yTeF58a;IiDj_WXMc6H~)qHub*=8YLg#nfduwpl9b-%nSWC#gbX zb{peurqe%Y@8=sh8IiNi{ujt~sLdWAw3cDWM1i0L)bK&opG@CR|Ch(Z$up5aAR_y& zRB_SEUZDC@6J~MWsIQJspd79sf_K&W_c=G@(C)q+dPT=`97?>6C5>u$3n)lQ7k=I> z^cl0rcn{t@UV1=HqXz46S|)Fuu7$-D8qnBxQIFiZs)w^qNa+!tVM|K9^4=kw3tBp( zP)w72;0;D|ZsH5|^%EkqM0VijA31G`R3BNIdQHE(ye0vyo#FOfFUOf4c22_lFF z=#uP0L35`P!38-#E0rgfSIR}2lO^Tdt)I*M$;O&>gP7$&0A~Ijf78px;4cuid2HWb zE>W_!mYK!x4=#vXDDbZzm7%)G7}+(+-dAH-=SHrwAdTbV&=0VWw5SHgbZ)nXy+VJe z{-}-2^{*FC=Mhd@zIk3mZIS6w|21;xV6aq1y${oYz`g>UW_P5Ze@3&0?=L%L&+YSK;21ea_Wxp- z1Yn>8rw9dTqP!LOw1d&ICkvDx$b0)GS|jc_}V{JVqPyUYbRpvYe(ef*5bdwQT6asEBg_OSo3#{)bl`#`&KaHOUE_mXge*_0tG*Z9K)3LhT zWy3T3UW6KW)BJtI|EFKS=;`yc*!<1Q`}|GYyP3^3Qm-JSnZ+Sk1yT8`80y(07dJou zJdOUx6Kd!59NamYZ6BvJwu!g;aLi@Bu|z)_-#pBKgi&()lcsTNMq$%_a~Mg%QBtYo zESov6=KKE_`R|`wCnzqeZo7Vj!Q;DlAQFnkvKl$Ily-j45zP>0g z-IW{;?{%co`J47=7~ro!%8R^ni|;=I!O|Ldui3qBHuWouzvdX&w#b6FmPaQyk7@cG zsdb)1I)pm#IU&KLrZ|nJiS-D!)o3@-W*s`9g?c~zM%@}MI881}K;RiM4&HOPe?O2s zU_|YgFoQwx?)tX@sKLbBid>Pa-E3T{r)8c?N#sY~mad(rFJn|+(CG#It>`F*Lma>O zt49Ac`B5?pM3@x8XQ-rSgARQ0Ty}q2GiN`t`bXg~>`s<`Ra;T-6}*qc&T_KKRDeq! z-W~EK2u;Wx-fOk#;L+jx?zIa(4J~S1VG1BMEZD~}M!&Q zR+~W@ynYy~uoNm zMRhb>tAn}bdqq6>#f_wnRoG#^K|1j4`-i`_ArQ1O_4W7zoK*zN z!raB{L3url%xH)cx2+=HqR?Pn$P+P-@DN`zc}!_3HXv{M*A?tU|5=b6IR!XX1Uhs{ zY|Ixc8M(li4P1v;R{>kQ)!q)-o@z(o0|4zYG_zsfzy(cCHN3v`Nhlsf6@fuFStAhk zNTrOoS+c6U`ym+Ru#=Ti)vmTU6$TcmUBvLb54`+-W>^wtrmT4g} zCY8K{I0LA+vetlnn@-6u{x5E3HjC+|GSIMa*mI^>G)mMGjXtgK5|I+1W zDZ|L3;(VQ+AE84+4HdV#cJ;nVCH0^|7NoAdcp|BzkHbueECM#bO2JX7dQvjt5XLB( zF_LRg`wF}H%19-uu#ODeYxG!~+MsTeCcM`s3dR2+DBk~-fNI@2RWyJ7A~3j>vStYN zkg7n2IP@L(I1#LZ_9QlQHvNXF?OEof8L@(x?HNO;+X$%*x8cYCq~jT=#%dhc2(Ljz zD+Ypre^!1%M?|vK;ND8*!^feOWn~Hi0@sO%{3TT`< z3DBm*f@-;Q`?~%@mMi8hz?CoI!E4kel(|de{2&pC!PQ(R#-XAWZ~N6UL4pyC6)J7W zF3|s7b_A3;Jp}zGJ+pjO|6S zprih;=b+Fste(ijWS%^(4?ne^^?7DL4#1^k$^7{4DVv7f-KZj6CwD8A;e5b&5i*jo z_?07U=ksOJ$t?bnR#U2-c5frQh$1q;3L_p_`5@JaqTBSoj>vHBa@P7zkJ_JPX*Iqq zJUygDI+R5?gm(ZvG~38R0RoelW$`lY?<}sm7JC$(At9gos6SlOOb~R^N7+J(d8SK1 z_5@(OjbymeRj>X$xGlqsF0TX^fNwqEqlQxdVj1_%PpMLhnG%?l4kYx(F zQmc3Cw5LYL+_n;Ro#7EOx|cIPhk|F5#`j%X7d-5sAO)SnxrfHTG z%jC7EetPa-9gwHpzXW7Dy$*VK+V8(HC^lI4CD`JnAR|1qU`CtV9YSQk;823n0l&s~ z-fO;FrE!Bd5~ji>81d@f-R|Y(xVsk?x`zqJ{L72Z&;hEt(`J9irbc#baLuI^SsTy$ zTjK`+!4rFAGQ%Tz!a>|GR6-~JZaXIeEHDTWT6%G|kJ=S+2*%dGU zujwgB$kdNYvkBkGlK&Wp+KA$7q;x4kZZX0$@!PmD;SUZXqLci6>T8L2_LwZ~ngn)f zFALYxM#ll^wjt%lDj;$&KenlG&K9G1lpfgNrgyt099tO)pb5l6j||H%=dqXs1cFp;}nrk+3<7s+th3ko<qIb(x%N5Omoh;AwPJW>!D$A{m93!ce18V>>Pysbe!JH`C>G#pbPfA@ByR zDSpI)TCzQSb_Y!H?+IAB#IHEEvh&X)nfa0rj7mqEn3f|?FG@bjZonaY)oafNQCh0I zS4Eq}$f_JaxsCqN)J_l$KM2@x@MJLzxxs%xJLl-q`|2FVuyCa*G>L8+0nk(!d1y0c zHEC%>?QjY#Qk0+TJAk#`?8nyEaDc8RN`0dy)|M1lWn6eD_L!7k zsJeO3)_aM_xSHTBkg(mZbRGO0o2>W>6uIh;5E|C)u z6xi*~*+qw1FbvndymAYdN>af}P$f3fu&Z7(Z^UNj`a-pdpXVso3e$H`S$9EquSu z6dQ3#aO$wLe@Iw4E1F))$sZqwYUEQVfT$bts+jnTkY^A}UsZM}xHq$4Go338;pgz` z)P=9R&>=ad5ZxPE&!X(>TvE17akrKF(f`+1gbZT44~~*N-jFek26sv{2mUjx5&l@{ zs-z)d#W+&V5c>duSw#?jllIc;DTQU^QK@0uyZEA0ntj|DWz0i~QY%c{GQ34Ts?dsK z#%FtCcJ-Xp`b1ECrxv%_`f~K`D|h$!g`NI@ z4F|T$P{RgSse`xq&$tVr6vSX zQBjOErfr-$9y_zcv|sPve5R>j3Jc@UC+wXY>fCK$f_iZk#)C@gCbsaa>UMo=a8}+F zvd5PXe2^74cYeSpc28`TcDFwa3=tW{v~5T}py7$j@o59Nf<ALB1txPJ5nN-BY&? z>KTofjWh@Sv&97jlVISGq>B>`i%E}Qo52QLx?ee<^wnMkOyQ8yBq#5Qh7MDtqsQX2 z$5!I%*!K#e0aJ_xdl}wHfxA1qXLBht@F%jDiY#z7MSgGFL%6IFK7I8p1dFx!z-}8EM1jLVd(~zdVqJMHMY#xz8H}x zSV1DsGt90lu7DWGHn-)zSz<~cWYkD$H7 zj{P=sTsi(ch%PA-k+TM6y!&k}BxNqv0_ofuv(DQqdo1~Pl&VSgt0HBc+zx9qLn#(j z308$F<}9?CS8fH`mu6=r>2Q{%Yh*Yh^OB=&ZLdmPG5?`Y4k^XdgIOfRPu<#1q-fb0 z?2Ch2;#VjC_*#7qna-#MqYmM>azgfWgJvOrXrM8^fX@murLU%Aw2(q9)3Qgu1JVQT zUf9<0KW~9FI4(GEMmw_o*BJLo##Xy89i+!XQF!n5r0?ugQS`pC?9Y4MjQ?T=;0X$H zo1)K{1kfsLy9t(AK9Rsb*axmyK+bSXcyieF7i8&c*pSGsxz~Lnq{qr-IM-MDK&P$a zIr9GKF6Ms|o+wWlFZwvxRZ*X+;X5I5c6JG&3ae;aIhaOs} zX>{|xu;|QkqAI}=2UjXxw_lQyr@k|jXR=m}(zU#|lb)V|Iq~a4aM*|ZhZkoJmM7)e zxA!l|=GoE0r~-SazLPd>L#4PZ(??F2NP1n)~k@AYL^Ovh`wgz5uham0KqOd&&&GjPS6|$B<|Gix^+5tx=_Ab%Ek|De-`Nd2B|gOA$rD#@V-u4MF`1k!HptGlp z)7Om;SYu&DXSmAZKP4yr%3T00x@hk{T*tG-!3k)oujL2R|B} z_*;h-B4Zd!tgSSr@J9aBCA4UT$k`XS3`#@MKs!FW>ZmmVwCFY#jW&rIG(^==TPQ+~ z&^pE(Pgu?s&qOx$?@w4pI}7@67sfFCPz#GgIK}P*Kt~fPF6DhKmS!Di8;?Jnz4<|N zwm6@Lu-DT#gZ-m}_kA1zOm?lW-i>WTBp-u5F_oM>8%)@Ws(eL5%gb|OvkC&5e(r-R zB^fqf*1uq6mW^jlZzi9K@zQD{WyG7sh|w!srIix2Y6UwxWfR<*=~3_hViNJoU>BM0 zdPu9p9hI>8r-c}$OC>nns}JR$zG4QZ5l=DedL|;JcPXBHTA%`No@Y3F-O#EWVUuMU zBzo{%6Grly3MU~I8=ZV;_@716LZuR&EllMm) z#1WI|$;uh!Fwnj|DTs}f@zeX+o&|KM_nUdd6S~>{W;LKZCK@N45*Nb1^YgVBrMd5M zk~{S{^N0!TNv>7R-v-y_{8PxQM3PgQsTOGxDW9GQhP|iG2zA9z3I8feFRkG z9LdoWs`OpIX>2L%>6>^I*o2r#|KPM~LrkgPhRt3Q`SqC9BU^#=dQHg6msq+Nrd<%u z;TMnw6vL~S)Z}*ZBqF23fGWQtELUyS;u%Fnsj&r){jv?5d39S%8N%4O?&}%`uRJaj zI<5lS8x<=l@w9=|lm1gr1xEk&_t;uUWNKDD&_M1_XNn{4Y{T`rFSQ&$MO9_nIcusnZ+~s&l zNf8~Y^mt5gP~?~QJu=xO-@ibJ9yKz$cWN*y-f%1c%>Q-fsRxrZHV^Kl1_mIW*s*Iubd zfvzO}2p3kAARo!z24|oqO^9*@MEP!BcNr$)B|UrpRf&r4UX$O9y=REohRm;4OA8d! zed`E$HSZ`JQMvEH!duI%eF8Ub3sW=ALUAhvfNzh+mZhSU3Tysk&R-BEw_-$=kUyZL zk{JmM31liqEI>z9WyQVK_=4Fi*&(w=k-sp}lI!5)aO!J$~AEZw&=q{OkUksM_N zJkHu4nrE+~cw^=tY#PbAI8>(PIA_c~KeU18B!XrR|8kcbqfWE%AtPg%O3Vy=)fVRJ zU!Z=gB|2;w%YjV!6=0R0_o~p&BF^#1qBRFok~^0vM{~rSf755T(^sY1;{k&TZ2b z;0q(*{l-YpZ}utl*Y?BG@##5P3%RkXlsR@)Uwl2#;w=ridE;xt5jB{!0`^}H$-!+v`NxBZI=%;d2Pn<0GT%no zW`}Ql9GmV5UR@Qn(Vw^(qiX(!MhNSXDx+XQ=E<(SXua7}n+=B2WLs#MoKO6#nD z_zRPsHEx?XeY=k(daPv2XpesB5g{V>gU2;Bfp zI2~UkictkjMmUrpvo+yeV)LBDWUhW9b;G|K)tV_W1!0F< zCsUg8a*qB?zzgPPYXM3)Zd{IRV>~9>=2#MUhmWKU{%tYuNp~YmkOQI74YZI3DnmI( zjS}RtC%Jk{8@QX~Yi$jTKT%Fub^C|~v@gySB;Lk7&La~n$V=Kh=DH0{1TvTkL3=O$ zl;j?GML4R&sJgPR)$;~_i`!D~_2OQ7p+Thx;Fi!ZpyFTigj%QG!#pY{V&^e3rBaGHqU)V5dW-KU5_;95Q4f{3*&A>9$n=fBTX_p!j|qzU4E>A9x2QyZphrE)Oa zKi$>*FjFO|c|*s-z)S6d&HKhB1DwT4*BFJV13~`61Pt|0y^kF<%q&2>vDF+Qc)fTe z)aP$4>k>EYa=b0X$(A#nHM8FcL=XR5V8eF<^piJ4gIUSMyA`fM7qYa4zMRB=QPP~X z6);LFcTn@O=x;7v+`0w|YEkpSWy(Mgp_|5T%*S@uyTb*GGW;vOtBUo`(KB%dkK5v$ zG$sj3FSs$soa z1xf{*Bq;KPB2o<>b9V z%%t2tjY4d&`a{+ZXERIEy2wU|!MG+*3_0;B;m*v?sEqxmi1F8lzNz?*Gqmd%f_Uv zCmIX(FkSl{)b!c2ODmb0_yVz#$^w6TqB zQ>n<^NFRe3g6E;)f(;}Fx8ravbMS7kv(Kl@i9rY5dW7h)>eF91lzR6rYfF&qU8LU3 zPQrf}jOn+IWBFZc8pi*#CSwkI%kHH~5Ja!_Ynxg`bDMDa(LwMcV(k>hetqy4sEg_i zySdRW`x5RhHm4!B4sK1E9qiynCbq)_=@soF?_$9m`xBh6GdT0N<$7Z3 z{e}!*`z4A{p6gU=#u4BS-&S}SDI^rqmy9xC$Rkj>YXSU?T->ZaME&%4@N~aBp~#h< z8uF|VYNk0DjBY`+MjR=1%casHWhb_x)OiE!0f!)N-FbC1cGRgJQroG=+n%ocx-D)` zrI%l+${Z|$bd{4DIu5x)7(-xDwux_Uz&9a;caD^n8Ch#i+hhrZ>%qxvI~+d0=-E84 zm8F}WGdkj^*}m%|(B`7_Mejykz{T@{)?1iF$@l-&h-|BeyP3tPI*Cqvh~BjD(WM~C z25vvyjkfU=PB+~O%@i3!0#b`^OIc zH|=%$4B&-5<`AQ{J4nEXo&$ z{xT?|bts16a<*842yNUmHjh*7KY~r{^(Pjl9EIppJ^h&fkUcWl<-Jt{W0{qQ; z&>HB=?eHLRJ;r7g9$5@{*(qOvp!OF!+8OCV!A*)>dCp?0vQx=6|2mu^YSr4 zo-JbsV6z8|_k5v{^x^h$zew_PVG*EW@~qPWUif60rK4skyLd1)8!JN{W~(~Gu~|}? z#-?c(0%op6O;|I!n>?EZqHEJps17cV?;g%1vX}s#P}Ii3tfm0oG8oTk*7Oj>#VK4< zLULE+QZ&e_)hxI^Nze&q-#55Cv60-)Jk4ZC7^WlvZu#A8Gh9t*R=ao&gMIDY^x!Ms(5s5{ zXpi5Yn_+t-H~jDC2hSWXa0j2U+}|kQgB~f2A!9Y=x~|^UmT8(G;_s15HODH#)nM!>=QpB2awa{I4eO_$zV#Sms^^+rzO-AI7y(IXClx7HnxO zd#HhS7+q7;XZQ^)%yPa{))hZOPtl_x3!}IF_Jmsr4zSbJ%o(^r*#Drv2GE(J~V)lJ|Z#(F{O1#yiMk< z5^9OGqbMN8hHx;6f2zh=GOf)y3j zyl~_?>0gmk(c{MKOQ6~O(ZrKJ{HM}c05p&TZm=soK!DYhu@wmmNu=Lnj}xU8U*WlR zr>0b*rCiDRm9}ZK1hE z3D8VFjBChP#s^O$j6r{>u#uS{BEM;y|L&oDf^N1>L1FtA8ANLlGLmlr*W%6!&x-6n z?QCD6Ff`Qct|N9`%*l7u(=J3`S#0=RYL63-hZhJDgo}P3ennV`Vi7hku;W2}DFqYk zJ;@i!O3>Z_r`p6y0l!_1viwIq%3mElxdCB=l1GP6%dyo8EG?E<^@fC^aMrK-0R=H3Wp5Qo`WhaMCa9jaw6PIP5WU#%|OClD@S1nYrke zeEuY+0?;8TmbI`HOumH`&zw=)FN+1Vha+N)Q&PH42gU{=%{cC3Q83q~x8 z7q%v)%5#uZw>3~`->TqqC~oQBo06KZN%4y2Jr!uwNU*g+Fu2VQ5!ok^$I!NINRl;h zD1_ov%EoUt@3-jp5y1?;*^PHg4%<}gQO#YxEv?*)<*06(@x3FdAf>r*;?YuU4l^)S zR>%GCJk+iOH&T4uRY}jjrjN$4ZR@H^!C5Gs_Gn~XZD6iBO%>r|bnP!TtHI3BoTSvO zI<3J-jMydTCXM+qau{Tc?K`*W?F0{LVTB(@tvTq@S>MK03k`4Hf9>vFIUAC0JMGf6 zt9GW5ODn)ye*5lCqm~^&2C*%!SXy*i z0MB9{6fzXE4NRY8(HVd3rPtnTfrLCUSkdfTML4tPL0GAMQdMy{)u^$tc({wW?iI|u)JfWHC9a2G?;k$|HDw<5%E(SzNw z`|Q&o>I@TitpzYGoE$Zz4Z&C}op$MD*y5nJwYcr~1mCdn_`8A0LxA!zSo-Y8H#<2a z^}>r9R(ow!Na&+3>MhajT|pwb7#69Aqb)>azD+tggY5R}*oIx`A81Ci>gcvq5(wpP z$3DLuBMuUB6gjc0hP4m+ZvW|Tg@P6s(+>>0ve02N5v62=Hw)egEz(8W_ z;?-fag#z%Z^q1qluiV%(0GZYJeQw}qPq&&jpAjA*Imxuqt(AfQxpylRINV5n7NKxF z$N|{=3g+N`#7Nwo(_x4?~c3=ghN(&9}Gr93oV0q(N2 zD1M>9#xQ+@(&E|u91GAI9DK#<8d<8K(W-MCh<6&>IvPIXdx5<=RRMa=j*dn#y4H2{ z^31D3RsQ{ifeswUHC0zTMQ#AZgb?qSXhElaj9@PsH#!^BUf>hxYiZPjJKa9W7qcHt|fL;QHT zXeT*H^_`uY@*Pi_zu-sq42vy*;*sxvB6B|;%U`mpfri?(tvzJaBs z8`Vt}$SNgIOKov4^}th|0Z+>y*1tzD8xJ_ixNks3n2VaEH#f--TrmDtbHt29y1q^j z)S(-5jnIQp2 zS~qhb*cJ>D>IuH={oN~dB=bsOnuvq~mR+$cI}thQq1rM@Naa7)qL(bV;b~7(|v{zF&XU-oJDyyT|>u;0wQEF|ounIwsMN6PY8wt?t zs@!6>f(&Ymvb%viR95(7lwQNVJ0f~uB20l~w6|Yc3jT=BbY1r?MF}P<=a=Qs9L(nY3jZ0sgh;U7=|2S z$=XtLDnkZ<%aO&;-BwIxgSwIXu>kKGy!-^t zDzIne=l^xeL%`JqT8pkt`V5leq<1l|Ha`d|40Br)Vwcli-AnpLP864kY?aZPW3t2h z!E~D+k@(PYw2NG8CcB?^87}lrTYTJ=C~rZgb3JkDwy&D0Gq`z4J?PdiycFTO5Q%=Qsea&So2;wyrhgglPRM-!Fl21sIujfO)U zQoQ8E-08tDA=fP*e@UGdli1AXww8gsumzpKa+=-9K>bHQsnsK)SB6=pz`B9aly3(EMjMdVN*&CJtDV-WGqhj_%UKrgHRtPgoX#wiWHxxP}AY0Oh z2kpBnKiO;*&o?SNZ@K(p6zxVj-ruXts!Tlo=H{0O__sO(zC^+OadQJcuClp&0TP z@3ua*>)v`Pp6h2 zfrpC1k|&_xJ6Y7Mx-_7`T2u9-1T;ZbrQ5^^OHqgpKbxD1k93HZEr+;n3a>g2{ z-BeiHlh_>6wairOR&dQ-7LLQy?OQ?|({$oioHrFh$*{lPSN5USw&rWLiMMk#xtgSA zCm~B8meyDIUwT80E%7LU0qEeg3y70;>|8}CU7BoLKkx#7@8L>(DKZbrBu9Vt7n(YK{Fu4kq`dv7C2k(>rOcdiN-P{r+oDR-e<|T&6Y+ z76CAlRGOV#78R<=(ArTdmJm2D8WSS=>&du7B`O`N!(|C_v%4CsTF=xUDSHjA;0Q0Z zl8riMC{+A|4Ly$peF=a3at^4>rXmv5?3-6gvN@-~$J3tc)wZbV*^dg-Fx4Dl zt8Ho%hK|13?BA&1z7#FZG@H9AID6BInNCgJ=?apR*vbsvR+hG*M&{?Dx~PnoVd9_L zO`_o8;Rya#0a6foiaVo_eC+^ZA%{(IRekH5R@=AYZm~k4i386$@!Dt9$haP}jL#p* zjmLw?!)BcHB!7~0aJA&yQk1~<9<8qxLcDRL@wl{Fq48~VeL&aM)xV}&rRN7EtN9uO zPrfB-xz%}C;88KH5Sw`37|oz12}2Bs)V+MzSskjPgnWZm8>0DE=I-B5Oda;q4~cMQ6|e|RMlLo!;8WiOa#Wmx zy$Q1$8EmtO+PTcD6M!H^-Js;dR;8fF`BaPUu@AGfyC|g(zn33B^D++w1QmHrRDuMy!_pu zIu3HY+$V-&I{$MVs!6?M9L18BjLx;FEPR#fLr`j8-f%u5~lyJ2o#=vk|_L-s76%A7x%B)l?rvben3EYzv&c_ zq{sQwu z43=J3%4kzK(Fpe(9j*+EP&9Vuk+|Q|AfNc4Zv@4V`mZ?_Lsbxo_~H%c573{ce!-65 z|Kz;o5Odcpaa>B zU4f;WX%8UD*$(mnMgA>B`BX@0m58gZ4DO4zqZNrx`7ZGKrCG}yucFsx`g6MZb@Rp8 zRMEVJq;6;Go^~C?BF7C&x)iYgfln1-L0xgog7%VbuZ$*-=amxRM zhf9$htt9YW!itoI7KgDTXeD>>Ffq8FIUvoqi5>DN=ec_}tOz-B1AluVp0lbF_Ff|6 zrz>DYuzo_WN&)@3LfhJ@z{>Nx*e&vX`q8;(@gJIpZsYcbBVoVDp==FO@brIDd(V}| z4r>r(q5g4Gfad()+V6gn;xL1z7?i`l-r3dUIXuS4YYGVIFHMcyqM8jYdkMUXt; zKtPJu2!`lnQm<3B(s0CLptFd-COk^Ky%tya6%gJXM8EP^t`_U=v;OgGV8WNBES@)3 zULz=0kZ~<93%BEld2FH-tW2`F?noV_tn;K0Z|3{wIsfaV(3%HY9h2@%Vnrr2RV4Xf zW-FyfD=)+dhuom=Hqy+*7MaK>rGFX4V`bER!Ff>}S3l9>K4Yo{qzar4eFNcM}%5dWx?)|SutP<@ox*n-F$!b-I&(U<3gs3ZjHrABN?W~?9{ zwAo?I-vmjnWD1H8>HEA&_>FQL!gbYalT=&&%#3%-h=&k+#iab^L8$Near|k})97c{ z921w}wb;N~msao>FMO|Cn6MS`f$?%v{uv z+LH;3Il_0h0D}G0UnJW4lC$LE*G6vRaOe5UcLzjPD5vCqr?NjXhDTBXv`~RDN{!9w zhE-vSrB0(q^*9ks6~%-)ow^szK2RNvI2xHXGO*FvsxK+qeJn5olM(y@$As-DS&x70 zmPQu|)oR@gWI?CI9v10WzvTc)6n;2DUP?`xNb9I&2EiF>IHt$>{9^KEw@>0M?*tkHs7eo&$XORW z_hmw5zy{h%+o{g}`XHs#QX=v)ILD~#*PP%YX}-Wo+|DDwgJ)Vu39!0|Q^6L)5%kFH zP9t-xb3KL+QMKU9<%9)#(e>#K9}Rw)DuF2yFJb>BUZNmv-&uKPzK+AsGBjuw6~%^y zoO9gXd!G9hexzlzMQMwoV8zJMHxKO^>gbvq@s>ixzCrWPZfyc#om#ZXo*)h>pLCKK zy}La5-}9!UJ{f6XaK232s**$D#+v^8LQ-K@5M!QC3L|@yox2JwxuWz`WpBWTiyNR{=|}><2&KV97s!JvFEZ)UphX=PkBQ6``OiG zvzn@ENX*BM@X@Gi!MZvChNe-q!ar6T3=lhu9nS(Q7YX@ggO68Vfk7H?YKDiQMB zc?Y$sYQyLDD1yQ`4}e+_U5E1*o56Z#db*YJz`xu= zvVkvr9`{fC%zoo3vKkV^d45#`vw+DI$X%~8_cTSV1TODG$&#-!MO>^s>0y(w0GmbP zWL}gi%WEz?S*UBwE)V3H>i>eEE+@`?6Dv2N z>pRt7-kTS)QyQ58=ae-}wwf0NT*#ie^}MD^o5Bvkm@?F50VS#$9Fv)zLj(-BFJa5s zPQ2DBQp)D0AnwFpqe5Mn;%|~9gD7jTg>B7%^01+@)#t_3O8c9{W`oL>g(_9QF;Ik4 z-71ssBn8%&4AsW9+Bs|U@3UX`Zp!~Ia5K=IjKEJS(pr@s*f6z30aS2Fhu9 zp<_81y^)Az!Nq?hU&_WFJM0N)-bf%w%@`{NFC2^wh5h3MR?|yys-q^S%V5V{INSb| zS*g(*pEht2lqb1~+4X5~Ztw9x_{pzyw-T6obLJKM@T+uK;cnyit=xgpJwCY_aUT9r&QG_)?M`@{B)BfH}*Wulzrw9A#D1JwhbAMDH!6jf1cnhCm z#&4673!bX~8kIC1QoEi|Iq?^U$rd2dNL1aug;yu-d}3Y;8@;&!zAz9=zhAZlQHBeCDZ>Z!f%D{`EY2l((Iig*>(-@PyUKeGzs@L)Lx6F+thNtOyeo{#%SL8(td^d z)i^LGZ=}Do0moD@V@aATCyE{0X_U%R_8lOPw(2g}aY6`BaBoMiAnwa9Ipdg0`WdSu zKZliE&@n1pv(NY(|ool=L-Z>V2M@?#wI*UOBW^#LQSzH7Qo>x^*`We^qd(3QGc8u(dK z%T;{e^lqLNxw-M&P;Bw}CW32zb5K)~4 z>wFepKj$#d=^qbt}0EdZ`Cb$`%ao8K9GZ6od2^9ok*QNC;{r~ zu#G-jZ6tr;(Job{)tiW}k;N~Q*$Uk{_LBZylALLf@(v(trl0@o8dNOjYW)4l1DM8O9w@fhzkiz*vF;V2Uu1nA_#s-O>yjx25#Q@A240`Ue^u~;x&Ly-N9S}yDNmKS zzgcDi%td43%5W}3?extD>}cr5gF7co01x%+8fuwjRAkp1j-9qNUju>}?YX_GRW=K- zhK;Lx+-kKe4aQWUw~~0=_1p#5nxkT-2gHts9fS(l9s+A$w(>L%vHi5}v;X??NlyOw zy0Fn88jCJevdd*oAv3!0qM%;2_sUNvNR+P^9cL%nVCJ?$x5=xD zzq(-{ws1y73qG8yK%ZvsZ{Iarc<7Yy8L4BD<&Qg|fQVHxuF0sR%WRxLZlwkE9%(!B z%(zpk)LEJfy;~UPv4O^Wy0xq2pdrewz{1E~%qr80>8tw}{HBv@D^-cxmu?35TB6CkOwZa94g zxj?!{MNe9Z)a?sM_0{SpCmw<;V}9ni|K4El^|M<(E2MxLTSX>zBu{@68Lc9wdxDG; zE^GsqsB53S?io(koR@ec=#wMOXu*^p&3@u<^uW2xcn_iL{Q4qs5HNzbr<01e?905` zg6J9Jn$^#;G;&pI+_hu`eZA$H%g&+d+qk4Dzm+FW;ik>f?;Kc3gri3{^Ut$#?HR4p zO2sc$%d;X-Fz5~u`^n3fMsFHNPr{`AJ3FXz^h{)@=gJTABzD$as67;G+ciYU_fCS9 zE@h=q5~kKwJFK=qBVCy2*z(uy{{yK&R==SR_CBllez(g)J$*CnsNlISRAb@m^AwAz z`H2T!Z#gaGPv#p;%0@^|&&VrG8Fg%;<>EDy2uao8jhWy&(C)7lV`@*!8meBkK~cvh zA#eM7A)H4bO}Kpo%>@{SL(Q$W(N3XeJ9+(9Ashoi?WeBUAzQD{s@wu|#baVvjKXoI z^92|u!WFbv-g%=qUjS)m3o8jQTp(Sp3lR(>LE>1zZlio8DvS*n7m(HCA*ExbpsQe-(1MM+2cYVv(?t4EqqBHDie?;VrV56aWtiWU zLKROqtsaMvv%xQ9t9CN+q>zuT8>-! zo@sKBqsB|F7H$WAE@Py@RE$h2FN{tx4KmXY8tzROm>>X5K8{U}W`(d1njFiB+GvuQ znj|wtUB?qbU~)jwi7ZVFXJ~9NL!$#}icRL^`pbn05^y~^s%um*Pzy7bOkj>c2(m1J zmH}>V`7E51q|Q`>Y0!P(H(XVzRky zLdaJ!AuUiTz}%?oXWnD6b6?2MwevacumR5oNm@pnFcT(Blm|vM9zpoeCN#a=YvN*q zzkb~Xqe>BMaSQqCoKzh$IF1T1G?Eh<7|JrU!1DoSVxYFBkAeCJnT&03Kdw`xq4zQTqI`p3^kC|1tT!QV49^CEJ>=~T1gMrCeNIa zEf^w;*UJ-q^~~TMx63yC_>*$4yWg30Y-7nk0S7u5c;Y|(D-qgS(tRztXApILa-Maz~T7TDSnm9K} zy)P;|F5NY2X2sDZna0GU}?Dc$(~>kfoXbcaW3R!k4p3#i`Ps#-dXbl$?swz?AZ?IWaZUHqBk7 z?%Qm+nu1kz^h)k=zB5akEXrQdQ>DCpjZY0rT0{9h%20~7TldLvOxlq~d@3sE`=>f7 zGNo3bG2WH}<1?flK!-(1Q5&n;<*1`dG0$0k0x&+M#>z8!B{|Pbr3&v=jI|bJ>*lWt zQB74noys4iF}@dhnxdJDlf~Fz*(u)H0K8{>Dnc*i%8!|N2J+Z!QFD%0a!*RRLTUrl zaXkQ5Qj*{FRFa>ZpQo|_UO+BDG^T7Q(q|G?o6yKUOR~pGXi?`Ra~hWE>UH~Ed`D-iOW=6vjB86gxC&* zK^tTwBNs^fH3sxDPrP8jq*zV*RW9_(bS3Y(ScpqmP+mpXhZIMG-GPce-npQn_Htfv zt)`-TW#WYWIUp{?H3>6NL`HhTRFyuX`1pAS@(8IANE#B8OisC-VA8i3f}@97>40qN z8DN0&%wSF5gZkhL^*+D1K6fcWw_&KcO54_Tei=YZhBU1-P}il;P{eHj|CsqG95W)5@UrTr3LL1lX`*ELsRv~@kS-1~dK&OqZiKPJ(W&B~X49UUt zpS?g@s(`Dw$B@5L=;JC4{@(a%y|0hyv~)Va^m%>F%=481*Q+-`TGRQSex~|UIaGh; zFr5xkntq3cpzl!cq26&W=x6A>>GVsF7ZZ-^dYTIs9;flAuh*pSQTNaN43Iu=;=-aZ zADWBwJc)nL0LEn;D_p1V(f#3c+XXYV7}wfZoC3*InCA*6E)J}?LDG@PzAc4^S(!$ za4i=N*WD$tG(2r~UMR4mjcWI;rSO4nN*Xpo@opss%Sb*;hwn6StOTlZLj#{q0#zBgO#S(X|TJ7Hb;j8!?u~6p|%2Qz1 z6Dl@{bd%??X(A8KfRpZM?n;N8;yi3U>v+17Hop}Zhz!hMJH&&zS-U6&6~T@?^! z2G;^$XxvpEd6&AX7Lnj3NO|8H&X>3Ojggv1={8_u5wKXuy=IN5-lzIj)~333v(j3D zw8vMshQ@ji(ZK1a%QHs|SwywisIGlC)vZzz+3?Uw+4dg$nQZ4mO{;1@fAEt#Wc$?T z&roDIMNj_c8QC{+eqnlzGhxDn2@@tvm@r|&gozSZ|AJxso`Jvlo4=t{YVlGLdv#Ev z6~pm?c6cDsztc$?Km0spCdVlq8CT*)co}1eN)=XIc{`QGwftP(2WMUmPM3w!$tf}P zj!a}JIsxPFoEUshj%6tf!}0N~>_^6OG%}E(R3a}P=I|_dn;YuMh~x*-BREU2*)25O zo1v}~Nl6<9<|_fK!>uAJM<*#Y+E20W)3TpMw0uNizJP#~PSlzV=}YvS@-~%C!2?=^ zvZJGgI++3E`W2DZ=(I{P*YOpGsIoD#(G2DUwj7gP+J z$Dxi7P4e|wvbb!L#>{lG+L8;-0q>tD`kP2BGR&LuQ(ja)GgZdoKUCa=* zRjB=OJvHpSjjVPTS)CqAP7G4CU#)6o6Uy7!#R9DK!s{WV(TIEwd#1!_AEhRSB+sf3 z%1YaV6bLG1_1o)to2U3lT%Z@AC zHYG}vVxu!W)hqV{DSis1GbJOK6~9f2KE zuqwoRHoh|^_ajg8E>lrS3zWjMtES}rkS_9${9v2l>xv~pkOBd&Di6OyHH}UUNP5V} zd>>X_RkOSk;HYx)^(pF`Dz8O>XW-dELyCnwLUJ}T)h~I3gbevZzOm)m0RK*K-h*-< z>IZV4d=<|piX=CX0OhN4B1uBI<9|z)Rq};H=cDyE80|BbB~Q>nPOptxwghI8J78yWP<(>NuKbj! z(_uV>jhSRO(^it8RIO4H6EEVO+;kHsu zyNBx5RLgVfdE0(f4K;7_QzNI3GW6BEsJ7ip>3ELpE-Tkld7fY)+;`u7w0*m>7c7Sr zSB-=BH~;z_RNb+eoHeS_Gm#0N=*8&Nm;ayaugHX&q0MY&Nin&-qRq-VtnJzcDyn%&Kt}gI~*P zS8BD}-?K%SqS&BP+dc8a)3P;jnc+OE#(}ZGgsPY!Cr14s+#_8*|N}PtW z-LcR<+g+l*s8tD(uhaRNliFgSUg5x@p{k8KEStp1#iVhsc8XwPM(K&qo)!kf6{#Y=D%hxF1+FQR2YVJ*VR#=(M6y9g{Q^c_uc zKRmlOlD#QJp1qsNwQ&_w2f2A`Z>=ZiswPTzD*HOP@we3l$+^0jJUcf~K9&&DsB+lJ zfBRmt)d$7XBQTXwEA95?I3ZCXXnjqUO*_tv%K+Ps=lrxA~d+b#MzuIo~X zkge8F7Ppi9cU~j+Rk<9Lk0;gjOIj?M{~lpds`!1Y>;>inHYZ?QwD-7Di4deIkS0w` z4a;$)feAlsg}^jS(wKNegG8tpn7o4tfjtWtCD;yO(yxKE8!heRp=y!70H^Y=isS~8 zlk!lE9_(#!4c-Nk5=dckB1KX4^IAxR@cxk26u)CB^8DV{#=lp~KFI3j`mJxE%d5Bzsl4!W-0RnqYt33p^>p)GL-nc1m%TPb zjwb%5H*MxT^EDkRPk5Hi$KRdJ#%V|78CI`X!gN&c+(oXoHu7xQN*0HM$tlm&glFJ8 zUMkwQy`Ia;M)_QtA|vO-Qye5i3&G*^k;7Fj?-pviiYyk}EE}Qfkb{!UJfxT@7 z4{_M)iT=QGZkxSvI{x@D_3~uDlsaD?lj}!2S1p{}xczpK=GdGTk!Z~yWG>dBS;f{33~CH-84Crq{%_mel^xzuRQ!}sTR<-EkJ=LH$T_P`HBi7 z74A<`nD6xlo#YAHsixIK?K^9zw%sGl5wvT{NrtRE)?RgYi%7|~-^Fbr=unP*!1x>u zOZv!PbT~==CYMOfP;RSs)l&blNx41|UYrKaLJ}cErm}JnZ*dN|Yo7L!0yVJg`A`GH zcRyD@I(PqSq7=CtPbRowT|yFy6~TP%XEB&7os~-eB7n{5VMfR@#v2B3=)cpau( zGRu7xPpmJay1N+QV8#aI*}X}~yN9<-!lZ0b#`yp|FBn%oDBZ>I8NPtC{Q^m#r z&O(Bs7y<=I96)m+%h!T5pl$`y5|8YZQpG9;&Z*qwrI}+gkH2Dw69D1~7TM)W{6VTR zpMgKzc;gRD4G4gW9vC~Q*1eX1Qwii84yQ9136p4L~Id8LX(pj3s5n@964955^^)s1M+U$Au4Ir zyLJoNnR&u{)*$e5v#JTdcwJRv`zf)`kLjAdA>nE~TGxi_pg@}%&o0Koym;-Ll%okl5QqbZ6qVPY?IM3DxhIisC1I{oZO zSsFYM5oQBHR{=;vK*>mkM!RA>RhtskIJeI(i~?vO)UEbXsKZOppn_Ql(gDo{sJMZ_ zTF;Ep##b+VhXqtg=zoCi|_DN&|xnDWUqO{FqYR}H^$AGx=$rLiylgyIKI zQuOIpDZc+0#qq4?4^wWEClEz3z5H*uPAWQes)wo^HVWK*J-ODm)A+X^mHShtdnh*% z<*{q=dg-nzuSi+ofeq4})~dCl2G?#`FUPQLv#g_*sx=bh1`G%QNf0C(h_eHfN`!{< zy*z!-QKVqsQzg@U#4Yky)$;wLA~`9IM+A>?kea}rMM;FJW=K9@9|SLOIIfwi&Z*;N znTS$A5`{FuNI`-#J~b$JB@O22wU7uQJ`mnjd*7Q2O_o)B7oPf+~Caf%;1O7X*oC^I-n`7|?gKAv+i1A(pox*I4LkJI==4^sTd zVYv?E-+PUmPY?9VwNuHYFigIy_maJ)mWIFabxLsBiBl&eeK1?F8xBA9B&GZMnchox@1W$6yL0$V!kpfNSv*hwONMDeQLPa*u zKW9raFn6*$ycBHSM0Kn7khi8)?uD(;;OS?Ks}$nl3`Lj`n&{;^V1@>2@{yrbK~?V5 z(<2mSX6V%)^~w2Cf=9W;M0$o-WB}5k=xACbJ58JXVnYNw8rUCoy;6Rxg1r$Knm2xI zEBTw;)N^p0298bAu}20)VpF>QttCQ86ZiHq1DD|E-2c~y>G76`UVulwyyWOG^Q1gH1P5Buo3ubiV3PYhA-p>QE> zJm=IiBcdj_?an4ye}z;E%+rDU7w19n!b^ns4}%Og1)xMqOWTAA6PFBNTLzcepvwyA zs<;;f1EiTC6Dtpk*DaKrhzMmzCMYvBE=*2#Y-IxR+Cl*#Nppk==`&>p$Ao5$L&iRs zE2!K-`UFDWK)FalF+ochyOhkuCT^GEDjT*wAWWb^u;3ll;0w1uFgE7KA%mC}k&Efxixr^P^ypfPUOqocw+ z$T>p_gCON#a_}k{FSv)Y_;JxHff3K;cg5;;Aq3|NAV|ETBh2je zG4sX@pG^Cip!!xFa+Z}+*$t@@RO#$atE3YdN{T8Rs5osfgJWGWA%wa?a{|)`T4}tt zy_E?Q7aG!;amUOglY1c*_*YHzjo(7*yuNI3D<0rwU1XfR5Z(z3ztJDtb!&zOtkeZc`SFciR{MKS3> zqbmm8?kcHMNiag7B4)RnA)edlNxjYuCv{q4c~2)Lw*VQWb!E z?(3zyl{^_+J^;qc7+z;Um@*?TxDM&!p1B}t7t-hVwIE%zCshS2g%$J~hYVUMuyGfK zcHcyfU>#Xq%B#%8vHjvcTL%+0@!DW6uW_;NQ;$XqpzvpqJIPmE|l&uP2H+N?^7f_e&H{ZwQ6(FZy ziSsMPnFl^CThuMwrMv6Ys=OzpR#Jx#uaue#I)1y{jZv!>_9Lmra&ev4v9vAb8KsvbFd;1pAxlZU!mX}cf? zImHdL_N;a`uj(9*3V>3taYu76irQNLRHqz=Z4a=~vP;zMy7DJQ9jw%Y8PxFLH4ZMG z;ocg6_cjfvmhG@!$9&+lWar7;4`sU$w8y)Ks#a0YJDPt{0+VXhyQO=LyIRr%P!5an z8uv=>2~oeB&&CPZZM{Y640Q*O=6E+f>duW#g8AQ^w5x~#sRV@$M4hr`yZoZ z=Q$eu@_)(km4Yv%Y!lu723oW0UOZ5qnT}6THiJGgA!Dt~4Y+TFB*^KlHO4CSjU0Jg z1?G;7lvml*fI1iO(Jx#t+vNsSz2@gu(atxwikcauEWqF1@i3~4`{$Y8)J-2$s_gLk z2Ij~oWmL%}Z z!pb^2VP-w(z92S)NnKqmd`cBmS_m#$@LSDjXPqxdJ);At5U>E{lPqx<=jW z3-RxuqzSK&SRKPRc2R;mS7AtXFS|rw;-@7*b`O`S*Def^Fhxs*v>mQDFp+38Qa9Y@ zz_iWR?$8#>wmW7_(B?qPWk^wLc3!EQwBx}47bx?&(DFX2s6J%J3}7%Ke-qC zI*$&t7bb5bj|EIxa#WhfsdPpt-7}=9dyvvy17rxU_kgrA?=AU;{)$yU<^R8 za{W)%u<9as!7Q(djOVv5vTc-seZS|fCm8Pq5!QTXviG>06 zF-=ogu!zNk!BEtrsE>a$fvG8|x}onezMahIh)}w}k5U8u!juSOV%P>P7Fw+h%n$Me zJExg>Qu)C%h0)5Ek*?@(e}$0)v%{20L?|+Pj^bf{=lD3~vPoH-Q}avoe^%}$xK-!U ze13uwlLK-+NT+e3k&I+1Jiv9BQ!$1>P)CExSt+!G0Wd+i%#<)4U~DP{RP-ugHlVHt z!ZW}SLHdJy;kO2u84v)M;D}AMdT*_`>Yk6hH4gF=ssHJBhK=t^V9^qxn(h0jx?_{9 z7GprP89`HJ;^@-4drQk)yzeY!BIDHkqslmcOuRAR1~%=Mecx)eGG+5r3-g7EoDAan z6?vOY#weGHQ#w3KiNS74j;Pg@$m!{r3X4Ht^B$QHwA^{0ab$iBzo6VKCWm?{d}4Y< z?n2;vON1P)Eo5(2hOx01D|X!8~S!=oWO@)c(Qh;^?3?6ud zlH&t3*@e}Pg*yn1Qi+g!dPdtk___Zs`D(lI?FFW7aZu?Dy!4Q4!)ITevD&{7HoW!M zW%7&f7!&pjf#phskk8iZ-zMeiuHkmKh{3X0A_OMH>QtTc`n%U~pHW4bxEd%<{dj3B zc3@nDVcCF^{qWx@Hl(C2?Ki!X z>)%8ERjLnNEHZH=#X9{XwL+SjI)ao@{&n?BBW2ctDPn z7|QGRfwsm2sE#8v@YkQG%)lr${{DMqKl-$apCb&L;nuqntg~z{NEYB20TPc_^G}QP z2QhX^0cHRO_i#3_$)d)v*GeJ;X${W1VAI5m9Pmh1jSt04O*w40?vpLj@Q{yQ&Oa{F z8!&#rT8omr<6RM_=w$wYNXQ@^<7u*-D~Fxd8|5AVOx6`I5&E%`(4_~;w@G+jlSK}m zht=Kt4oaRo$M@W?+Pl%e>hGv9Zt-!r;11kylkAQ@@MGB)yVov&h)``@|=J1fsX)HSN^D+oJpGi_OI>HQ4Cq>89_m@sorZu(F5+MW}MF7$} zer*FatXB)k-}vKK#H-x#$7a^uZ++($*@965Q}ciC;yEZ6VSWykv{H%Cr$6wF#1R@i zIlnWccmLl#T$WYh9(wQ}j|tNT_t?M(KebO7q_x+EWPhpn=eIp3TX8DE zd0Pl89!Ch$A9q6=Gd&(2+@Le5bv#ugMjI`x1{wnoCOUEss{&f*vG@)nj{yh4RvG<_<36|on3BxLKs_nJs+=O<{BM3 zUk9%?OM#AggUM(8_8VoT8>&>8oJz;bNJqy-MJPVpL+QwvFgIzQz~MIn%%aU=5~|gM z09}4go_ymPtn7oit^~*nm>{Lf6;dt2XqQOK9cmWuY*10cx@tNyPRR+LjB!2_WBsD4 zmyV3eJ_ZqdE5&&e@b?Gl4mxc>#QPO6Ymfk?_`Oqj9u^5k`(_X_i&zp!uXsX_yufQ5 z@)jTLoOR!P6z){t`BIo50QHdPJf+45DZxx%e6UL<_OU`fdbt_dM4hXY2^l}7Vw#Ei zqxu}UoTm6X8z1A&bu=m4`0(P~{zC!;3T?WUybbLl-9x>~iYvDrlpnseS?w;&?Jlam z!>L?mcDuv<+?Tj(nwaSbOaE1AyYscImo4(8)j)xCbVBMK-$iisc2R!=GmhZbDb5et z0=xo(sZqrJGv`0Y42~=&aJ?1!dIs_cRZpac?+DjQ`)2rV+e40yCT4mXW$Wx{Cg-X~ zvNB_X7(UUb6!U?}EMSG#xMu!F_=v~nMBP;PXRQfK=_El#V(*&AymFe<#x zfw{7`@U?i>YJL_!FVjE5dF1@Ua~!9u;|N70ju84O8)bTjei%`{8$nTnN`K#%mlH z6qG}$EAdh^TRP+&biF}(H^0|7ZjW3icn>@aWrOD+4V2gUfc6>(Pi}yY z`+G1>P<^ung5+%G_Ak^zjus~VogN&JZwq4~TvJ1RO_Ja_oxP5~8y>H5J&rqDTX{U^ zevq%#YHiwU9Gn)2G>6A&`A*7#`w0u@U-~faw-Coj(l+viJmGjYQo>Xi`9n-2q;GY( z#lxMwww4@?+{X;f9V2QnP<;;Ou5bzvr(}%B+sPqHg!?Hr-Xj!cit~PAa*)rDil;bP zz{A|O0>H}~aP`|&Im6o;0*}t+>1`pP?lR0=tl1lqxH^#bL`G5~ZGp-gwveL0VF7}> z!a>0hGcs^)ARgD|8FNd$#(|-ML?|{kzxvwtyLp`Bxh|fKm_qRW23O!T?h+#jaxJ$8e)>FkSc<9Kn!GAY|}N@g$UA9TPBgEXB;n z<-`a{C)(7=oo#?SSKj*l+~K)p6$ov?TJ5Fm(6H<~Rx$I;`=~cCIL`J~vI!Ga&F7gB z3Qv;7<>c%5c&Jv?XmA}Eq@}l*5?=@B>QFcsFAA<^8e5g~>$3(UglrlnG5vpA>K?MzyW>3Q^?kKE0 z*`1{1*Z@zKyJfPBX9<(S%u$B(hj+n5*ahj0)TbgyM?iT@LK&&?A?}10UNJ#A$i$bK ziX<~#U@VHs=EjOC7{i3hyMe?=AB~@WNx9MPJu&OP`3Uc#I^=~rvU0%pg}laxx|ta} z%}h|IFhQ|_v-}?X4qU&IS03O{CdHV+2r;xVW;U7QcS&)&iV@O*RVxLahsBYmt6xvv z#x-20O-8#@E3sLwSH*zqcE?O7*Riy_7*7znd{K|(+MQCHY?gOMd#SYD&1QrEZggxWTbLo?NfUU4Z5T)R+qcyQ2mw;T^f))PlZ}}nM=O8R zd<;$(a^n$^*dXAk1U;4UCoky*?z)cCW~NIsLF-z{xu%6;&mQJ?#$=AiBhFJHpz7l< zi}VPL64Hbn6Wn)CrTJZvew4>8&eK8&@nqalnaxOHQr&7< zCzE|G`FX7vAu9rkGHc_ia!cOv4q#Nkyy!fY083=mJ;rpC*b3*HlXsITI&nV%0v2=`wMj1bBj@9gFGKoHv$zYUl=lpX@f zo)1*Y2#L2fW3k!l=5ph47q&UVbTG5wY~k-fsD+}hyuvTTEs*;FTTQKEg!r4ld0|wH z>k-2UaemT0Ju-)bTWg?VMhM>}=Lz#5%#9G7RRw!h`OtYnzmmc;Fdl#mU~B{ z&Sr#mzpYKK!Fj||0#k;-u_I@rYFi;8Hj}9O6{S5N7Gi`Jtr`br0X8xCh`3jnh3dBT zoek+ z#A8emk<8!&!^aL_(6)2_!2rQt2l)e21ZYXn*>C)-`Y4!DsRXjp%w^#SsDNNPIG27V zm=7>&n1mqTmy>E7q&wbgcJm}6N|~NMvbSNiNvX!=qiPi~^3-#}9C`OZB?o<+YHrD+ zs#C}P0Nir|&?n{!;OkW5iak5GnnF1!?`9TJg99@J_}WzPH9R>v)<==vld?Z?Ms=hJ z_$!?uI&!WfA%KqF+pDpvWBYR%7A+Wk-VzBqmIzlo34I?NXWEf42>D8aeSYn zTInCxZ>y7i#1;b1KCD#7xOJ!oGzT=Z(zx?>WiQjbt{^SPZD6!fRwCJBh7(Zj4t0x( zIsf2t7tcg?HTj*3#NxlF#^17e6CUu8phYc=2$9SI&Wn0RuBIQxXrU zjotB&J*sf%7{35 zsz>!d(tdb?zfJ|5gro|RB5}xIsPxSTOsXNt0M=Wsl1X@fo|$q?{ioDeiU~8|uTm0W zNP-}Rfz$*zmsg8Z7g{wA=^#E1RQf;wkCw7UOdc>SapOL_aU9Yt089vgJRyG3aK5j= zYyn_|0F=)v%LYD^7AaZ3NXt~+p{#@<5$6#Bes8!H=keKiKWmlkqEL;)Gr%w+{?OX0 zZConT&LvQH_zi{)SL(dSnYjWW30n#rjZLC8q1DyOyRj|;e;+n~fbZdnu}4dl^>?S* zc(S`&G%&!JNbDiL9)RNm1-_os@$J1<_7Z1K^EZ;B*bybg*AgT!MsSHO4A?xH2T<3u z@O(=xhN^y1e zDtBj|n_pWBYJY0%v{L=rYTEzJZhGMFk5Ufyw+6Fv-(TTSh{46FGJnj97|7s84=OGrxQWu@k_1b>MZFAVuDl;utg#*UcifOKpCeVyLN^&ngi z($_;x4g_Fj7t$cmVj&V1o{&1333Au93Q3Ov;!YNmPo*~1r=$l=TI~lb@>s1G)i_9D zhP#FIb!c-!>QS<~j@55WxIxQ-sPQ@6@_eYnLEQ{CK6*7CQYsCb8T!tPY-{ zGP!G9Lf%HNkc`uyg5iU~DQd=Hv9e`5pu9=&oMsI*OJ_p@lv;%Uw(J9vHrAwa68 zBokE8N|_)4!B5lYI1kLE$bDyeMHNnnf2aArd@NaDf`D8kCZwJ<^En{&_k}*6KVsM( zf%GDbcAu^9gM$^wc!|G7i6;d68?O+gH;~d`QmvVz*AkGp zz>W!$9Q_VnW=22=#tKFa!B9b98bFY~fH3*TUyAQnc0V)A3z9JqBs5?gz~CU=Y6=wM zDJiKKALHkM@H{XbVERB83-F$jCuZcZg-Ri$R$A>3q=V^#i3LH`542Q3jV;yFEez0t zOc221DPr(sMn)*v)oBoa6GCdAKU)4+nPCbBMRJsj@wdZNdVMa=$C+@A#jTip0QmtS z77+5Zkoq+wM#vMKBg}k;`mJIQVOCo*spWmVKZ3c!P6(tOKV0cRQ6b(ABr71;=D7Vf za!70;2l*RaLK@-0Buz3c7VrWGDbGT{H6WYA!c{rNa|Z4SG8nt2>lE>e5F-fyg9PuQxJKdY6p;Ykr;!xqTV(j>M( zxDK{JFyh4ZAZs-|m02h=FeEB=7Q0=HmQ&}t#6AeuF9y{u>nN~#n>ZbSEsj=w#QTjM zf1W1Jy-MS!UY6DQN}(l0kp4w_PKa8Ncv!P4*TPu+uWnjHlU>T=(_ENkBUA)<-GVyO zImC7u?pCW~6#xR1i__FyJpnv5gQ`o*|XQ>Q((#u3IOZ8srG>w1M z53u^HrGNlzviyensRsIAlzW8pU1?js)DfSGh}{baQW2=bDX(xrYP75pRWe9dpyCD< zvJ@rhNd9@6q;ZiBK?)Nm^$lHU8zGG-=VoxKTWo9~y+J;3pNqWw4z2vVQKTM_bd638 z61PlxHUBL2PMwi!kv^m*#XvgvgGwG$&+r@4l%PeG4XCcFnYRJ?7V(boo`>>C@)J-d zD7TBjMo1&YR}NDjzX{JP_-#-Qx_)8X1p6UKyHM_^3rPNSon17*cN~k6)$66|-Fw6V zfp^DM;^ED=XRp`+iHfnyMUnk4QmUtm`-aM-qb6nok`hQlu6wVFT{O{~5PO~zPtLEV zwh$m4fm4FMBVoG!eQS8UQBsmMduwUqjrG*9u9`+WVtSsM@C?Kay6soilds+-*TV@y*DDj$b7-bIC_Km^KfCT}C8)^3b}Gwb?eOVn z*|bZc?VvDLF9r+Q2wjm9AyG8}b0lgVGu2Kofvl)TOvnTg-r;fU|Yr=Vb zy!d($AIrt!{5z${I1iPVVzrk_feE6-KPjn0)VbJ1WEZE4S`0n;lfj^95(%3L&~hNE zTw-5=bZ1T^DnaqYv>31}S&XWu2?J<8i|7Ul?T(q0832RD?M~ayECq$SMVni}b_Y8n z6(AJ`))lA_mVu;Ej&b7ePx{g;6qtf%WEvmiiF{O;BoLSZ5LCr5u?N$jCekq>98)$X z3Yv)kg8_HkU}7-AF9p)Wd6n+$ww9gBh>Z`_*R<-}LQwCJHu5(84oFAR;-kttocb$e zfJ&j>H80P`eJBT{qqZttyeFQCGRaKAnn>kUpPLIPGhBx<0D;lMvlar>0g-Ouou(>N z9m1O;edJY_gHEHgu2Em*>g-|w#xgi}%1%ru3Wtkmf^v~bAz2(L>U&Hq4w$$Xlfz>b zhX&qShmcZhQ-pXqaB=|VX`!ho)|g;_14_kIa8XyaO}xaxI~yn&%?d3QP;~4e@pf+7BuX81J&-lQGVOJKG_aG_mJ52 zj2?JW_LqycyWS&#UE$OLt@^T!BZM!uTvO6ta_PCbj_b+O0y@WYp<(B3RJ-A7G38nNTi8`UPm?_cgEoph%vyv`I4{um0zs$yR+*v&&*WX9*)mpAlxL_nGe- zM-~T2Y=8~-{JLy4VngRd5{j5H%R$rLyTtPzP_u5AY=Mch2Weoxit-HGk&6jv>&@4^ zQOZ5I<|^Z;LaDO^?HpnK-D{bt;kIcs&%9h;)bA!Dp~T1m0AJzx(GZ+(M;Tac2pa6GS3})zD%o0Ve2rl5S$f0Tl$ThENt@BFZsT3xVrQys6o^ zgK>J~w}x8vPV=M>#`8;`l!3t!6{>VXJhJJ0imDP5RDnR6VOk8J!j?_V?ldX|@*{aF zfz^&YO&htIC$ZSU(?*|ofrC^CD%y+b zKZnmmMdc3nS!`82W?RTnS?Tzsc$w3xAliGEou3O@iYgksh85zjECJe{gmY<-0=&n;eYA=Bp_yFRy+I;m z_i%kOEd;bPkj>$icBiF~kbtYcMYp@9Gr79mq0LFVT2zrWT>iw>Q$7YeS+21{Jqb7618g`k4_y)j|FuqGZa^DvX%O0QV*{50#trpG^XnqK^7mq@tc z;qr@r@D}AX+$4|+vP)oDn*KpSqX7lJErNSa!<1%;4SbPN)*F`<>l70n2lFk!-k z2@@tvm@r}D%EV=1BZT0nfN#xK+1fnSWb^t}bY-lXBG{jB)>XW3li#vpDXJSYId0hS zh>g(8j~hpphfBdmDA&_Z`LPM|SgM&bw$emSc^rb2#ZrYpdmiyvl$eT9c1jtts`D1| zT6|*E8p}`0wTJ=^R9jSlHlzi~iN2@7YD8Afs=#dli}FxXZBeVDNC(#UMBmswjF!wX5p2s?HOA%A<^TT?5(uE{dHQmixRN%A;PD-OBgc zDHlsoE|H;vPgQn08j>%*a#KF6+~Y_lsCN6q@(jYIndt!LR+q%4rVpHZg&XSS`KsB2o1jcu*5L}Mvpxw+r>f$R^9sU z|1Mh%1RKV?pZ)3Qm!%ds7aXfwDe#utD0Q-vQfIp57=F)(`rmw?Y$t!B>^~Q}{(J-K zeT{$kVM?FtrLiyl&^Wp{IJ~NVtXm(TK&zKpc8195vs2?1B?X4f0penb;WWH0i8n;m z2Tb~FGxpm|Txj225&Zpiu^WB=Z)snZ) zMVszw7u!5|kE-7o$pNvMIQpYb%B3KcU%K&kF7vrt!BhEd=<0w|(pBt2J z7JXScpfB_{d*}n-P~WB4C-Niq|Cb6f{^;MoBHOuuIt6yUYdzicTRUWAy<>Nz4YaKt z+wRzA$97V&ZFka1$F@7RI<}39(XrF9ZQD5Y?Dzd}_Bg+wYE+G?weGp*ye5gY4JFOo zeb<(EWTL(yMd{(<_d0Lerqe$#x~_JR7oNF)#I1cUFHlEj*xYIwc#ZEKZM6 z^*=vvs#Z!Wu{y{=BB{ITr^uqgjixakGqB5g6Me$C-$3drdHO}a$h(9?m@03m{#V)ftngsv=G1P1fi!3YpD%7KSV|4w9P!*}o0Gh&y35&Ce;< z=;LJaNJg<3Crc6h`vz?jzcq7sCi(6D43Pb|`%d zMEy9&^N+G#eF0u2e?XGFmDqOuOfi5d62%Em(Pt9XS(%_gh7YR8FYPExC{gT2w>=qZ*VO}3gX zvd^txQ?TTXgg={MX}O6@eDp71H!CQ;TVC-SI2DKAWQjreln)yJ_Yg;k{MXd3{gn;{ zN`5}%qp*cUqfWiuUdKp*=VU_QBHbu=k_=_v*p>9xFTYUizPpV3T#yS`)FjwJB?-wS=pA zh6@SMBNgo(`;;?MolPpP84;&HD0iJb`|7EbtSF)SIS#E)LGpfRjn7Yzuhs20#=}Pt zpVN6@Uz@g7&3{&edg^6382dWJB22&5HuHB0GV|(J>BdzCZ!B1bM{3T;J;+xbI~$Sv zh$n5;PvkN?m1`u8oEg-#GsVT58cQd$*0cUjv{LZJ1e;7$w-d{?`aYzArHi|Lk7G{| z#RU;(4jDdlJWtW(@>uSN&)e4NAyB?J| zQf$DNS-DYECew}>FlEbeKL%Dzqo&>yd<3f>yF&StU|^DvbLi}8)7_#$Dc$>zrSgTa zs9UioX%0cfl%ucc_4+yQ&sy`w-@7U8uqh$dJA&2MZlVUkTPihj$|MM=R|qpQbFhh+ z=fP=9WP69sgV7ON)uqXUUyrikJjAS@Gip0Zh0pG6^UaO8|!JoMYTA)ei^_JxBjC+ZS7Z&LOZq{ z{M}DIKMItr%jpBdVp)wJQ9}<-`d8EKIdPD$I)C;JvstDvV`aRV3V8G8Qs3gWVhRMX zV+q~{o6w@o5oq@Zcu-36MU*)+MR|s%2b`lYO}4$UIQgjMr$oo0@i@=~bY%WWCE66I zGsm53%Pe9&8}2KVw*A;%0R-QXKcy{_jm52PMhmbxvs3s>zT?SkYsUXFYaE-V0==0na;8TW|fPA^RZn7PX#Df}5Sq&F^IGRnj< zgba}uY8)aGP;H~iz}}PF*=BqF{+%F(g6LxNbypAjdL1cQu1xEB7s+z;8-v}s{Ok(4D(GQGb>bS;Vp)FsXX ziw?%VH1=C)fsL)OM=tYDaf38^gyEA)wzAE|oGo?@7h@IAj36!AmPsn-nLVL8cpjlE zm5_-NJqXAwF=ZbVGh?YyiXT7ID^+%vzyey)U90fAHDytYkQf1~Ewojy(q$Xn7wJYgalV#E!tyS^f@#+3D!9y7 ze3GRB?)F)f+`3V>0L+|f$AEDy70Fz?Jh{oUWPu#i1@%prN?ou8BK-eN?+5xU>R!7ITOebd%OMl zd;NY)r^qJL2V@0uUT~M3Ko{H9oPFn~Pwkds%(J$1hA2qf|e>q8(Dc4#AADU9f30Q?_feMEqIfS+?Dy1Z1^M!v z?Me~{Dlz`f%w<%+Q?yUA7n^pbla6Tl=Z|}cd9;Ibm*7{MS!n0v0r{&~-!CXdXuqp) zV1x@3_T=nNILz@*90UKkXGyOGTbe}K4t4R3<8zcv>}(S-R7~%5AQolzDq_xN3Q}rL z$SGk`cKLPAL(vt(2uiARpZkKYhMm2?`koFiiUgbjel| z_q`D+TU_tzQi&J{QcXiOsM&CSQ>81)$lPt`oTZ1+oM6)A=QMhw2ATg0&wrY?f-G9T zF0azL9I=e3OmCZFIqh5#6-Mt_dt0BdD`NP`3R9`NlQih3p06EHu2Ttx(|916r_!Lh z>_-IJDn9AW&s$Hv=a(;~$FxXUO0knxU4?;Dw0NcRh3_lUZcQ%3Xoa|G+3KBq0+k=u zy>36)3O+&eUHjr;vGQt*NY#yM6QSdZ;;sG?i z$LcSJB}Ulev&F8P!rQgArayNAT?mzH20q#+)oHt?$#5%ro<3#Bs@FMdqxc9|MSS>m z&<(BGSpGL?7O!Cvs}OI5F?VF>x0^-hc7)Xx*)(xEB?Eod5u}uoIB8L{U62jA1mBL1 za+v@jrb_3f=Vc8#GT9lzEaNF~qOV{-_`@BzH9L~9EQwV59TfNic%R#sQ$>RG$iA+e$;!)l24!wqt zT;)KK+f5DJM{Rw|wNKouJNS16yYs}6i(hZ z?_>}&m_~8x_87nz1&iI)k;)C7UK`2XA2wqEBriyF*(x0t^;=-#+Z*> zzdmekno8bU{QWmj%@Ie_YKBca558#E$N@{I+|cE3p=9MxG4?;wToM#*wLQUowx^Ba ze~yD2!ff5t0)O92$ZZ|L4sKPe7WC}Uqg=g0WwY13G*qP zY4pN?ZkLQD`k))_aeRd)WWVYX( zoHmTw{^Ek6uu1oG{(OpN8*CADO<16eCBpfSl)dubi9=pu#0mSoLo{il*xiuxMolbY z5eYn!UewkLL_f>i3LBaQ9)dA%H#iLD$hRC~kKpRq8}PS0pZ07zH$Fd!WwWvJ59`O! zwk$LrVp;9Y^=LEKm!>8nQ za1-!IS#tHNicqY5&?Xp>@U-g1@zQ0$z8CP@^4n54&wjRjg&w7wr{AAi=Q8Q6ywDm5 zWpRN6EyZOzC%0mLriC(}lLtc<#p4P`$QBXbQ_hMeFl|gHu#VoEypLZ%7EptR8gb-F zb>o7vj6Nr@IjEraZ}AE@tq4PfWx2yeP1*-lgvznTBG-&DvmgTzeH5v|CtsYgpR(Z( zHCoJ#A@y&HX(VHK52$~e*_c_%^(d~CAA(j%0XZf{pTM{VFy0EgMCSS7UgOA04@cbT zm+f{5Z;1_rd}Q)mA0$^kUpxIx8N5zJjv9z|0tw$XpGOFd@T~nxcX*R`2(){S?DR|KbK!=pKR7a{?;VI;k+?xsz&ZVP))CFdh+CFq& zo2`7$*_R?}@ZD%k5+-Vcs>qh4B>!8!tjW`ovcvcSC{sdFIY(t4M9_Ytg)%s77f8Bd zV`IU;B>39OghncDMgeW#W$;6)k|n6yk4(GL_ty3VJ2GIDRx?K}K_wE^9bUW&e2uJQ zNomR$^6T-NG%hLtPvkclRC%mr4F_Zn;Ti52a{-f$$Eoh9!J6$k6LHR~hFCEm8dcTD zJaFMeHbBuxjZPU`azRDFx*Q7EOPLOu?;K#HPTN$TdI}6fz%2ROF_9qH#H&or2a=96 zRbB0!JvBP+oJ0KHMrw(p_*3LZ*!wD08>>fqM&0N2THrW4Z#rjn^ zx1_J<0}1)#lGk1EuWe~IsrcgL0uR3wI|B=e9`D+e*@L+ zC|5n>l;4xD7I*&-_48UM0zO3MTFh6%hhq>)O8+=?ny65GOU1A*5**bfPgX@>lo_To zHE`S7j!h(^Ji;$k5#)+4CCNPbuxAb+o1bQ6f(3YrID4A>T*1b;1=2@IFChlspSn-O zC6&zm2jBEUx2?tD&ruaI4RQ)lhg-7w&)T*m3eo&>sUe16HvAb}I15AaHVnyrTZbm4 z6kdLHfbMm=g@Qe%!1SHI{S}w%-=+I>w0#rSk3**Z3GgC0)UVpKTB=TF<#d@mbjV%PUvUgqhhkf(a5UeAnOw-LCZ z)ciC*b|>(NKE}xSJ|`($!sd8iviBNL|9Z~M`M>aVU4P{i*Y6dIVOkA^=6r&D^f9@3 z*s`Gvl^vlDGHn0*3}27YT7;K>{~WK(OiVQZOKx^h{p;#N>^D4ALwkvtE!d+?&+;+P zPlEEB5}@)O)067i*{0;?;zP1hX^1={k~giyR0|ivrW^l4_a`z|XZCn!xeTSy<15Uq z?WWl0nLV0Q^(op|SbuVqsRfv8am2{VDm_-~jugW7>%z~FPOXcgQ-hAto+~~wUqb7h zmB6#wG^;)C9D~Sb2h^DCb)}4fi4?)ADem*uP#^sxIZ4nJDwOb(uK}-xtoQjuE2Qa9 zvfuDv?^b<^&qXF+@QbbogNH;cvj&}B(q|h3sfN^C`04#B2}hJIRsm@QC!IW&_FgS+JWfs!Hhk7$#t2+MRPF~PdM?*Jq_hH z0nZ{_itclQ0n2xv*XF$t75ATX*$t|&JTl1}6^IS5r15CYd3+Z!f5I#|JcZB~rN5W! zll}^M`-_$}eM-q;&ogrV>y}sU3^b$_q-s8h z?S*NeqJlUZ;LZ|XR-9Z0&J0s$0*>6tN3RN224KReqr>MFT(kU^GcA$=VI5!`r5a7( z=bDLDM5GKGqC52hMQI<;{2P>%8k{ZL!b1xhD8Xy~=_3aHz)^!)W*zEn%Hc$>pI6^l zk)`hkT0gPr^@QSi$f{L)%`zWXHKRGXIT>Wog!&7V9FLcIh@KCvEruJPO_V4Bze&`Y ziRVuVGRyY(ENX!m0Yoczxq`Cu+R}l(je9JUq4XUor@xNcCN(aG5>~G5elZUWGK4j< zkv8d5w&jS1lbo0Kv|W(BR-XBb`8z(uoqToOpuaw^k>K*Z&w=wjOniT(ghc)9XCo{6 zuv)t+t%2O*@Nx4JDjpu2eO(+r&_|b$H9X|GOxsHV%uKV(N5ubM`X_Y_kA9+KNCiJv z0{gtggZC9_ru$<>m~Ph;tzH5BvC5HdG7Vfi`ywD0i)op>w~aXO2v?E{3bE6-6r%)(a*YjrdYOz-$5F`GBUR^N#lAOh z-DLl#ZA*%=)*IVwnkw1;kyUg!nN7|n1MJzfh0mtMhKN!&QdEjo4#{bvy<5KCQYWWw zuPaV-N4|sC5$EZ|!3%0{10s?=r|rFfF2(9td4Weboi4-M8$lbRK^nNOlgS1JR9bss zoVZ!{7M}UYr|14LOr7mRJ@ZTu#6Vq(#OzNXpom zHgSw#%j50BEQ4RDh*hccRU32%&9AOw6gXESUjRgqUt;5nB$37AS{KD;@rt@b%hS%b z>ymf{;$#GGkhiCI1y-15EeiVilHmp)h@l*dR5kDQu{N^Q5<$4Y=_C&d(&3%t@ANB0 z3}6r=F^F1s%&CraLhWW%v~Czg8}3cPHvMxqs7!Kdql=8LIP&rtE3q-vu5Rz>92)|m zzA@LoZXa4v@h&xeInJ@-OFC9%xxBFlDc%)sj1`rE%96=R6nI8M0Rzg6j*9-F1S1&Uy$^y@Dw;4 zIzBM@9syIHT0`@?E;e4aYD*p~&jufhm$&g-_ABF1=s zolrYc|EpJ$5sv-P_MGCo-)NkFS}PS!>7{U112Jg`<;DZv2w|o_a=&?OJQ>KwY&Tf|HgSdZ8=?A;Q#HY010L0uAT#2 z`Cbpy4Ltf->qS0mYmj68F;2C}++_~InWYo^WmT~higNqsDq|A;SXWnA@DW|Q)lR^T z58@R~a`eBl_QrP$w;i8<_+JDCC*CU;2LIC6)k_K`%ZZv58+NoEm@d`FUJvtY?`xDk z|CwJVeGt4yPFN(R9@X#V&zlIGSpbe|9_4R;*VXK_lpETAY2c(pHT_8nyU+8rovX&- zcdUcIb|B;%T#8=j^TqM8_uaT2Ld{8LF<)PTprmjCG-^s(aPKJF%Hd8Oz!rWh($-I=W zMMBX!LA3mZxnlI=z*{In*#?6_(!uI9IprS2BqSjyc9BStqpsS$?}TOmk8d+4qB_FPx-4N%`hS&(^aR|D_<8gdA(;&zu|g zO^m!XT)!;RAh`H-f84m6T{G+EUJM9v+b9{cD`Ju4>IgkuO^%U7VWY(x&ehub$XO~s zBb@}Vn~i2heF7JxnYXcDqegYLJ?O3+QbZ{YeVwMFtWF?as3Efs!h%;NB@}*;4Kc;6 z*eFVBN9|%28SoF!ylW&Q-9#Z11UEi|paXBwEZs?5j0YVsNHjAgFNXy+{gj9rCCu6| z^25J$9}vZdhsv)4PN#7YJ%q(SH*Z*|F!E`Vfdg0XnijqhYy!WbkUTpd)3}m|4q6OX z9F5<_d@w;)2@Y~}DZGHr29urd0Saz{^W39%L>1gD?~bYU0}L+H53YTJ_NOM#a=+j|EY_rk=<7=2i7?z_ zAQOQ$wAKs4Lt!AEdamCcs^qI~Ij{v3;cZdyJpUp_^DdboS{pGuE6yjW)aF>Up?PU( zQKcp|#9aQAmgX zc|x`V^jy%^iHGxjlUxzoZ6aK*_}KsTE6uZ*_{=hBjRm+fF@Q()Ha?u9$Go~&>7I!< zs_J68;~6?^MHMcmaXxT8TO2mtXF-P^B#XPgKrIbJ%e$`Skjr!`qO7&6qCj1F-1Pz! z%8ikJAa&h1<_ua-2F>%Ln(OzBI{qS&UtU7OAA?TviYH*M%wTnNWfix5dLEo? z38tAMZEeIsXoed#Ur3LJIALAqMj4l|Z%Yy^BUL>3KATF|jepBgwQI#4p-Td|rG7yK z__38^o5GYO!$b7G2PBAfr;bB`E2l}sYh%;mSX9LZzo2RTlb<^nJ)k)_!F2NQa0>Hb zr-1vrucMTw1-0%s&ux+P4(I5NWBJ=37buPV;ag$f-x?dilA|)5BHqk@lGZv|?)Rg8 zW9KcIEBuk8aAN__b^Kt!4#m}=W`oQK|KxZ%vbkd%KG3TJjrx^K8XNkINNh7eVZ>%w zOdNZGgEyq&r>)p$`^g(e5|{`T^ZYrMWogZ%47mO<6gA@h?i~R3iGqk<$x?2vRT+#O z@`1ip{*lv*wrQyLhg|@EI&+>x`wC3|FOq-2rxXV%kt|d5j^)aYws9we@ zsQx;P-Xpjm$FFPnPfhf*unYp^XrKcMxHNP(H7chqf+}Vj0bb`&HRtXC*c9w`BK||? z%(FiH_pmp+m=kSj_BS&5OHq8|nY`GNaqQ^K)>|(1d50bzKh}BhD6(nRT^2I!VmELh z++LBip9F@M#>zrJq%XV)Ksrpc* zD_1N)*)pVeW}8rdL);bR)aAuiF*ewN2=Lt~Su!G9OTswdzpt!8`|ytaCjIn?%?;dL_K-)=Dl1N+2rh%BJ~<^Oyf}o)4?Gy@ZVb$ZZ>-%}ClOQn*oFAr z@3Tp^oo@fqXb8EC0k#NqfZonKgk(?}71w)s_fp-!`&GkV$fW|e0@+Vx}3nz+g9y^W9G9$a6u<57M;-E@b# z9MY3yqx9?b$Wo_IE_G+XLSHVYpJ3rVDf&aO&(Y797+p49K{2w2qTmABDjc_Ux9xeepnv~4QS7X3qkhA0kc=|8lgK=q4 zYD!rK7=|Jwm1H1&@buo>_`C++?34+V)%)4=B2=!&+NMe^RiDg6GgZ33^o*Rhkx%TM z^ICHr&W}32t5EFf4VpZ^TD4ZFs0#%mb|Ns-F?)B!az*c#xx>xQ%Yeb%H5TF94?RRGs}2icmgBdP3I+ zC+cyjJVzBQ8{_=PzD8k0Rp*)F7Uh0nY2!COeaA8yLYN8f2k-amdHZhui=z0UPJ8{C znu1d;KQIyJgD)6eBSHGdCJcf6!?(4!=jj`6L z_uy?6SJA>clZ`VvFV=dEK}MWK^_Q!NollILI*I`Vynmj}n7@m*N~L;uhcE-DZ#SBy z-N$BtK~!R8g+~{Aeceo)yRU)KZ&ck7D!HF7XKhvGyOhw z#z(IoCqr1a-~P8->c%qw$ezQGd+*t{S?2xuy_6;wsS`5^fEkh;BA+B=gi!8YWZe{) z5-?Wir!?c^**_3Foia{Pnm?=4)~O(VP#wRdig&=N<%tt(RxdHVoC?O&1s<`w;S+L{ zOjjyqIJVJWL?sv6nI=;xVA;YOPtT34A(baa961g+0^WN9G)u$z3~$=Id`Km=)O{Kw zRXNR8Y#_#P_@~B?23&Q0s(#e(NggO*QmSeTT45AAE4-^lmRvC;$8F-L`;Cti zdMHq{atuNN7j;dQqs6{Cd7gexB~5#tly?y|RZ-1J5oshqQKY$Xr86=|E$i11MyGpb=DJpiq*hC-jYfj<{6ON>kwEAif5`QyDECR$;wO-vAq;-*Im^x1vorzo-ueA`f$ zPVS>(YoRm^$<22g0q;9*4lWKg+;<=FQ*&Bwz_EtKnCNwvnZ5E3tlr0+%DFLH^G;ZVV;#}2d{g;K=P_Q8cdIwq>6{~$NxH7k4*VB8z z3+N2lFow@%TYQ0nky^Yhn*e8RoCkG3#s?#BOtx8nFNFNfHiH>va}L`@)^GD`fqSI8 z*3t3}*E_~VvC|CvIsk?rec&>%OO7vxt}1?yFyJY4Y~7AzjmpM|9G>nkXid`vQ;rE% zH+fK)+%I>WUF5D!b%MQP*IF+~NHR%iC1o)#Q@tzugAM8W8~#$7#3Ra4Vy8`sq<9b2 zP0dzfW&8aQ)1G^ZmJs9!@_OAum186W$a5q?**lPYer6U4_A{R^TxFq;qJy2FU$0Gf zim!6uwbdMj|M>E!^VYctt;M=}2kI4}hz#<)nG0vX>eQN&N8#0_K&&2S41)xu%tZGp zC`6eMoDiSuXeTwZECzQU=83dL^6jk>khVjj-k~IuaR}wEjtQGn`uW!Li%^?z^JZMO z;qxBS1pioePLKpHLj6PUD}N{w=m&Us3D1f8mq7vF{;^5Az_og^mP?p(^T)%LRg~Yp zs-(p=Oyby?!rWc`AHf!G!OkX<;GJ!S|_XT_F6sG_#RacWS%aik)~r?9G_>oF~sQKNXEOswg$5e-0nj* zTvJUz$4H@6LDYK(IfxbGVMKS;e%qOgRFCwJ_Ll+pzyNUjZG3TAGy{1u4&A$n!$PAJ zluQxz;0fl~rXmJ#4uN8kDMwBc#IyhEl1-dCb9eP)p~cP#lWD%BJpX1IP{Wi$%&~pP zf$ZqeO@Z#|p%1>ile#_+^VtQa&NJdvZ0rah=hCz)?g%O_(Se>-iArl%iN-2wZ9DWL zu`mK?#EKwc@iVkruYdh6hdx5N)-CPAV8;1Tsk}GH?$S;Q_}J!v=t=9qZ~u{36_pc_ zSPQ2r9X)$~yAwI54R)OeAI`?=A@$zuW9qo|^G<^-VcFKtEWE6r%n~Ucsg#Z@H5vQ0 zuitFm|3|bbg-j$04K3fmhdu{|)O{P4f2~fjkVd@E4NNTf;`%Zp!wG903k)SDSZDbp z>~c-7Te-DfHQwIOEFXZ3#?Bz4W`4Fb9cg~xQwNj%DQhfkw0pTWE8pXkEch&TwQ1_U zPbu{rxemcqw3G0#MaD>^GeVWePZ^{!OxN#TmWk)g1d?I&m3oN&;%&Buoqc4;WgFnv zP`D>?av#0@xsN!oC&ZMj@s}<_+>|{DT2m2O;aEQqfvA8Vg!-u0aNMIOz5Iq*Yf!eq zH80ziDek}SMy^(9U}=*^9QEjOk7C?gWU{@YD1Ccp>fuj)JtbE0cbW2uYB>7XiNZEa zQ+i(N?v9qm`TxGa=B-_{mBJRuGXz~C{#UZlR=0e1B=Ydeq+iS20Q+-vp{z;#a0rVE z)|<}jwywU1=e)H>wSPCRJ5brhIOdcl_Wp_{K_Ya@vWbPh0o||HYNGGp(jpV4NV!>PNy{=$iVVWeKV@WwB*&?2&HZA=kpX0{|O%@E?@ z;&~KFV^v`BG32E!!|Hi#1difV*abkT`?@K&U0X_>90?cyHOtEh)ar3g&6r;S&)@lX znbXrFiYM*S#{Z_?Qv`Y3TS4;vxX}Qe=l9h3Rj6)ugm*kxcJ*|%(flt^@Xbh z*I%u{WCTL*b6%25j0@6tMhPm5=b#)|`YJN2abDv389`4Hkxb!a^?(fY{1U}`b8pep|0enI|q$%)nJSGd2XqNPw%f-n2mOa^@`>ZhF$5a4TL zze0W$?n3zBXA;yYU(=#1iz-UaQJ*JK=I0IBywwm%^Y)M&VL9^6L51GOq&PAMhUI1T zUA?h872;>mzR@6(uhs!l7x16l&#l(a&d&Z$jQqJ;z($#e7&Wt7Yf#MYMy+ZG(en2; z8~j0`5raAT8j&tQsJy<;H3;dh*<&OIM+zmVy7+JxCpV+J`#;7 zhSH!~q;VgCDM$}GP_EhX z@Qqj*Bma2V1q&f{>gj@Lfc2t)D5pfl?`%{UIhSg5GKkjlu~ydUbE3M&sjpMTWk)WN z-}0LX`#k{xO&Ai9b9hACXD{Ex1-~f@i`S15Qt>i9Z|lN=rhKX`#l9O<*;4- z%Z8Y-p9m+nouoq z^>cSDhzSp?0)j%F8D$mXdXdknoCZ80zZIaD_o-Qqpu$$iN>l$6RD$SK43WqMr6!3) zE?>3Nf?p9v3RYpM%De5@=R6C(ky)5W0)fv5Pa6VX#A{qSV#s4!CO@}#UH z@uYz;o+QbNPHX!DKLT{{L5V3O%pXnyG$B4UP^ne;R*l!;Iu`kld^Ec;uGXc^;z-aJIE-%(oM{0KURnEF&3_g@A~4FIKGIZC@jfzi;p-MW zEY@BTxg++EpL6|UO1|sJqJ|zkr+z&}fnM<>-~5V4(jzmjbY!8V2-~<@SJ(kjLBKgt zirC9lbfJ~gUR&-0ScctUwxvw|mJcb{2&qHQ=ytHGUOiPVDk!MaCIOpRL`0&V)2#VQ zbf^{cmX(jMB~yiuSCCy( zdIelsxDlOdOjrWd6~n#qzc<(3LeVR!!G%pYaFk2tZ=V^m=w%uW=EWP2z7U4y?hDbS z6%JSzC+>E*<=UE%K$ib5EKUKU$Y$|T4gX8iBs)w@2p{>xrQ2ANF7GyUJVB+CeYg~C z2uTil!-;s>)wN3V!KQJH|IoC;A4&K#ecfzFfNyPWw8Jiw(7jfR z-i}W|%+2Qz!S%~ZsM_y0(XMUh7elR8*s72m`Tac&%G?e~7N&RPVA)`PuI-)rg7rs7T;~RU#>cIf4HS-C|~V4@Tge6x~EF;QxYy@NWLZ0>22EYc#~KK>ZX+^+Vo zlfx_j)!lRUBC2AaiRwxK@i)Ee3DJG5yCo_}0UfWOM%eGSm(jPK79hZS-?R6i2tORG z{5X8+V>NfA+vkr|t^iGHIsOu;?rNjB!ZGdpROV8vt&qZsej*)J^4^CmX^sDjE*f55 zR0KRk)dKTj`$e)@7%66~nhmQ!a9($6ltOLvNM4^gss^G31$_Q|Z-}xNhge2^8LPx& zo)(NeSs{fE*R=EvmhJu#TTV!)f8G&>Lr-8tM{~Qb zI?%Sc*}N8cR1<)QGAIJ~O<6u5CLdVGpaO% zp`Xxg?*#m8y@$Zx-&o?8M|)1fplaZ#x%j<_`IAoV-Vv8*N}vP1=;{5D ztE`wMF`#;XZ=OmZCEi+RrXweF;}Pr3Z~(+39pDGyq#Dy#Bzyx@QamIjF>sl>===_m z0{RZ`{*X!W@x}ZlCxfhpp3Y>?CKuVjpbsn#FoP+bCQ~GzC~+n>zSlUG9*CF~lb=BV ztN2OE^SxEI&ejV4!c}V=hasFR>Zf8v{Ge&+;;O-Z!yDbc8UsP$zs3f^m8Uve@Tfs4 zA)u~>zM6lq%f$NRNcUWcX5Z{e2&*d18*zcTC10E!8K`o&{ggLGDnSK&M;rBEhc5bC zi*x_6M0(B2MADLs%*9%6UMYeL_ zGMJQNx~GuTbWUYfFBth7mDV4*70(bX8&6;6nNgEj@BKT0?&9nZKA(<%#tuVPz+3!x z`x`Fafw0#5eZjXKc$+bkO^>KsvPqyGfe0%l97-|0t;U#h!^?LT!Nj~}eo=ay<^D)ayl zf~EyG!)D?lqE_VxS?c$E)eu_BOFC1pV2Vm+nb-{j=T1+DyW1)8HXYBPJsq48ehS=} zSSo+Y*|mr6lQ3C=V6nQc;ja5C-5o`Ovs)Ry`jz4#dX0c1#I1(gX0er-09Q=C0AzQ3 z4Jc9pN=ho$>e$bhvpN0!iH{fDmqv0Z>bVLtIefKVcgH2GfOJPHKz0=@uqsI3z=QTU4DpJ^f=`$>^mA}{}dX%OeD(1L%#{gu@jHD|=|_B#$r{*ks6 z`-4S*RBzceKkJn1YLClsa(b0)xX_ozqLS@;tRIVD`)cl3Gr{o3Tr-ArS-Wf3azHVK zEW*qah)}MFn2(UoJH8% zl;ArRZ;i5_LuJp^(aa0N9K@)s%a#xfdPm3?&-GEt{dz7r2UBB z+SO-HXVj(=D$1-nHH9^VUC$G2Lky8u3Wy&k&r1wThW_(wnM5XM*uP*E>N2X!EB?(j zEm^PMdsA+z#f^9AZTgE6gRxVOjZ_1I)J`nmLRFq_9y-0mDFA&gp!(s!fYvRIeM`aY}@{@B^Mohbx5)TEY`A95zH{pw=M=4eYeX z{VL|L!Mga`b#y!jSyVJ)5qB=X zJgaSiT^cb*_{!K4Uczr@X#?WoUPT7r7Iy?EcxT@8Twni8IAc!xYizUf6-lAKZ#H)@2FcY#>@AQ3}=I zbGO;F33R#P3-L(whlg38KhsjtB$BIuTlXqsE_t2`iuH#(3eHtV68h03iDwZ8NR0jc zvf*AtATyX5Nib=%KiJh%{orXH^odPN^x_(53InNdLPvk9Fc9^qP%s1$)If9)?u3vwbhx%Ecx`dNLgT ztS9wORia8c3}cy^kibPq5WsQURs~q2-1ujhkQxy0#&(d%uHl=|<6<7ID!+WFdY~ms zrXu}mC)E{LG9D2zx9D10lfg_gPXIWLq2mhio|5@fajD)VI>wEgrx?aJ6&;_0|#K=I-VgPiEr9)o(yV17Nl&D`e zlg}aBW2V4_F`vdc%f1RY_CJgGXpUa|iomTgMt0Rb!|!4C4S_l91YGeyFZjqdVkN1d zZG4qw=)~M?xZ1u>f`FJgaI&ga2vq8P1f*C&H&_$T^chg>!pnZtV zMbiB8Uk+*)_`;y`E<*EVP2pjsP4V>PQK`gVFng;sc2X)ZPP^LTmad0+KZRgr!8BMn z)ih(h=GMrtImiLqAY5)xzdSwkTBc<;33HPm84Y)eb&43kh}w&ERHt)HSYRlSgLPBv z6*j0zg5M%Q94P|c0>Ec!CcBGa@gk5#CbDb6VeL3sz2Qtz>jAp?s3QR4MEGh~`yeQz zeporwcAc(D(HSZCAL3UWJi^5V(RQqIl5R@#?>p1Vtq1cRw<>OE0fk5@Z34M8Y8b#L zzOkY33V@2T8&1Gf~+%#li@%(<lh1$w0mcoFv8h@`5OA&3{2OWATHjoL51roc=%&W&7t&EEwo@^=oJ8~J< zRfIp%lxBLPKK^Ds>sFNSR7hQtR?UBe>`WzZ!7`UG^;l47aK9awf4x9G^vpeB2 zc1i9$ugw%LhkYzr#*<&8?cX>EsTfOR2W7CbZV2P9a_RUYCc%yma|%(o>r6| zKIn2Z`iT8MT%A*JC1JOQW81bfF($^2Gck5Dv2EM7ZDYcTZBLwuolI=|?Em7Ns{h<| z^+i{Ab$5N=de?fNS6f`L{?1&V`_>uI0z#mPQzOa2*NXKi_l+-mWdrXFrekawrfmWE(ghpR4FR6e-lsMh2C2+q5jO+j!hw9%Qz7rL_tMLdra%_QH}XyC z60#u39F=}O`?;Yws^8$>(6}XqpaA5ll8k}3Bi9ua_0EM$8m`~!6+YJF$$EXhbci4<0%wg@iWG+cRUo|8`S@`=uSl39m zpl1xt5RPun7kNv2auBOqe(iGUv@w_^(-!GQJHH{Nl6o*K)TVIls1RcQsK5iwYh7i~ z_GS!upKb1Otzf`P*!p*VWV!jKyQb;k z5Andaz{EwXc>@K#=XW;c_=Ouk251|v9h|o;R61-ML=JE|DaBaTbBB=1-J6f`_+AE* zU8l-b<8ByG`Bd@Pja&qTcYKW@M4FMP9FIVKe~2+|QUv$597eA$!Xd#W2g>LMp; z%@hqHXb}pt{dn~~fag}F5Ey&n&0y2{xjwT~dp1Ubl*rM9cQrBToZqbmwcY|3Z27yB z9-BDz*69VHGX1i_865A29z(r`viLW8x6*Rg>4Cji`1iIB^K|yxIbxt1H$_aU-o@WB zXXB~EXUc;>Vx6ABf7p%dWDkxU&LhM+7qt*StDB&t*{rAd@ZWw1jtlR77$gKi28tim z&p!XA4I#sJYGLSB0^{azhybf|BE!vxwA2)gptz(l$5iyow3K;_MeP5i%kO94pt$1* zR^F#Jhtwa0B4gSJe&6xq;nSq=TZN>{A26Y&R3O@z+Ufrb?+{WIx>H4{Kw_{C6@c&X znmb@YHG+|nCGEskx#WeG5kW-?YboM&fW+SKO1seUz0SPMLM)K;+7E*^G9TOwItGdk za}tiJBwvrdZ)b;C!~<68hQEM$qC2=N{07DLd-Kd2ikFfObA6GKtHDDgt;U97fPqcj zMZg3&rUe=wiL`ObCL9ceM13mwuI}#VPY%GUNyr&$43)oq#tQ!Sx*Cb?;`85UEi@R& z18QRYRO{S+81kVp@W9L;+D>JAvdZsrQ;xil4>0o&VdhitU6gu>>k`y`rcNg+(Jq7=s`UL^V`xtJ=IaA5YJ z`lrn@TFH=`2;~UKmjX(RCbNO?hDq%oTU?Es&t6P=lbh!r?2nQ1v{KW>whO3J}Ys2!sZ4@t75BX3li<{daRV{{v~ z7NP{hbBV$Y{~Oi}?cn+^mU4nXbIlY~-JXkQ%rXJ3Te<)&V&!lWd-$lLf?rs#rC@n; zRiDzH7g{YWINcW@&q_VdMK6(b95e5arxW{N==(QpTrSF#Y1ZV#*a=wu_qn6;Hnx8+ zgZn`MFpbVsM4bqHavJH%ENeF`xTz$V+H@EpSd&{+$q4XD`66pII2Fu<-=$&rStk%P zVa60}^IJQ8=K-AG>|lTTz840MTOvhq!lvLKJ?+V~Dy89*`_VvT1|10HDeaStqjCV& ze4k{&rm;j`uoRs#(Mqsy!V9Lsk3A#|5`GyB&Q18+c=~T56?>8z$PREo_$0xKXXdVaDFN#POb}igfe?4PM|*qjf%h^ zC-oN%yE{)3`F*t;NlS|AMm0!jkc z^1yE#{7+GZrHB=n8C)Eth|-^d5Z{Mb`MoYM6fzMW8q62L2MWjF`{ESzQ!hhlB4B|Z z4kI!IRT(PgskpbrLJV3KMo?i8!j;lJ$st&C^YM@ zF;xE0X|}uYr==ORvRL?OCJ_UIew`L}=eWjTDhyktwcU;YY4pM>*l@Ifsk<0=Ov#F8 zem$CkwSNel5yrtQ7T?pkp2|TYh-xVMUJr-R*5nH;r5dGnD1CFnkT1sw$2Um@4v{EU zyV?>?X)lyu^hVe;q5k|>hDKg2$`Vj_C!zo%S$Tm0>LkvHgfJ8zqbwNSl$vUpgOUnK zCObwy;UAfA1p)%)&{J^aVuJaNZ%Fm) z@wQ1#XE7$h2L~G|Vtgt@iGdK4AWJ}$_LZ~<)3+f1_9{G70ab|k)MecBJQ>1!S^4!Hn$}xGtI8WoZ`?b-MzlD+SChth3J{@V|MG2O0bmf2 zaJ1CxnJ&QJX5XFQ?Mgfx_vGZa6o4qBR5T5QAJCNu4z>_uOSm2u+LI%7c-5NJS=!-|g6TkHL z>u=K8BG+j!r@oJG0IM&*C}ekKpC+A>kBL|CiWy7GiP=CYi{n%&aY0mW;aGlq%y|b8 z{JZdOxWr@UFgye_)FuW3i0&5c-9n^_q%Uw5t9^D1L&%X#zrpKL{*Wi z%(E+8##n`G6;|Szy5f^r;&+Gd*Q>y5QIZDf0xFTO_%$Hve4cb6~gi^5>#-bBZy*yKr~Vb6TpYHq(`2BH4S~{ryHdGI#j2ZLc^QD zM5X8IJ2W`g5vTiU!UA>dRfSdI&N_W_vl^pd!b6_7Nfrnp25L-v!Y#hdgE)=t#`j~} z-2l|j8QhcOcy;`On!ueKWI5uzK(}b8r6);V)ZLzplrKxOA!X8dC%$WTIFPLw7N`Ko zly}zK!xVtm7i%Ww52c4NEQ)2r!Mqi@lN1|f3OmU>h0JVJqBoO-!?c%XbIFgs!3&XT z%j>IyvJo=l1jO#Kp^$KDARJHjg{*9X6I}68rQ{Id%ufxAUn!Hv#3nojtb*Sa#dm%@ zMD<0mVqy`Vl}6?bPPH*CUHLFG9frR|YdLU|Ob3)nGZb-38hl4A!&2b|ap4hDz-*#j z@&+Jj4_3!erGLaF`3T1qUnZ27?iHDZFc7bi6GI#pjtr(u$RP}JWFDInBy1O0I7u6k zMsixBkl`d0vxzKeuaC9X#8x2(o#N503d)Y_9pY#jAo1LAhyfIOa7(W4YL7-5%S1~= z+HghHL=&LEGb1zFj>Di}z~w}jv@`Giy%jKr2lDXaW~~6o8oM%>LpqQt320zOqL3!s-8?~gDIVOk5s)-z4wMs!d5Q4fp2#abE>7f9Ph@4DUL%uj6EKQ;C8(TrY5ImKY z@2(|(O?v_U5k!?bBHdqSj1DOca{HEDx^YB$@V#gqwZW*W9G_c!mj?Pbg>tmp-g7Ue zK;Wcm9AVj8hD6|*7GcfVI2%oD{p#$E>GdCr0l&02Cb^hhJI^HA`~r~qoznp?P9SBM zpXRmg(g~(Sk8DLO_D-BQ1~r3_2MNqn)f1l*YcTRw6;m&zkXb4P8}CO7`QnaJT*0`| zupAR^LT!L-GZ_MdGd9Fk;2E1)Due=CNe$jrho^hmH2P8>cG%TUlC+g)Nk;A zSj*g7RCy~^x@+0cMCh7yxv|RZn1z9nVp#HYiY6=Jgy(;1+%ZAi$_j=6<31A|&j&ui z>kNlrVD;WP-)Y^xTJ>=sFZQ!Sakm}I60ojt>jpAqd}LUidsd}WwPooN=C(6C)9}0A zpB>f(pNrkvEBJDR8CROLFx#tbaw~`WR?C$ox+*9o{9vf^D0rTC3G4tMw3@5Z8`H!( zY7VtQpoq6qpb#4{hU7yGTMajffKvua3cB|7nt64C0b@JZpvr<$@%?2#C9)L$QA}jB z#cS<8?$1G$D>@Ou$_A^Q+C$t<6^Q4O`+VsLN+zo7yVO_h56G^wOf(;PEp|ff+65oJ z0{i9t+bL$&NjKUfkI@fy$GrDTb=8O(^>w*aQ$a*sKl_$y(dU8i+5cBt#bb4;n7Q9A z{;E|Z!nw|l9U;kB0S#^F^z6xC186MsB%sPw+dd_gR9dkkKjU-D>iaqN5%_+&{=Zuw zL+N!^S>O>!5~^Nf6KYj`s@|zBt=RO4wGqo+1iA})K_8!s>8D_m{K5d|D>xl$A57Z5 zKY+{6HkL?Pu-I$7Av_Fe#JYCuQZ*8PV1r~4q|26HaG?+dPO)kwEJQc7eFsvrC=QD?Zt9KxdJT5J*YFI}ha6`SV!(McLcTr; z)4hat5S}8LB8Y4%fs;$v6&wcphA@e^tk5YS;2aD+zw*qFn8Fun(B=H}k z-p+jTXh>cwR~Cj)DhtX3wNt%{l5DgrZrja=*LB`VzubBAF0YCc?ft|AhRI1UAI>D? zZ4cJBF%+#e=_N$eH=;Q)$`zGwf8Fv`EgZi+3rBY370WCYUaeUl3c~bWd#;Al@|}lo z-2$7p`>qh+W0hG>$gz2N+;*^|&i$2_sP*e0tMXnl5Zbh&I^O(xlt|#f5bHdz)ijU< z5G{PR-#}fIOa{fJPp?l==wWU{(7zH0?9M7gE)4vvt;kq>&B66De*dc?Ch(o1zlC&3&b@AUjOUWmok4jZAwj9@F||!o^Ml zEidU`K-TZu0~PMZWJp?ip)JKiF7JbkX=8ezKmmgufq$h_nhc;Mylzg-MNw#yl#D~li5D{un=r;vOWCnla21~!?G`4RtO5?Jb7SN` z+ss6}2b5b9(=G7XD>T9g?Aw|)pgcNNf%x>HT`^bTn=(zLJ@qtHyjr%$zN(~18;A28o?(s&U-jDa*4DE{x z7nDxT1=6lRl3=z`=M(y~7s9LU5?1(kePMY)il|*e#ciR^bx%%Nip19j`m?j49X}-{ zz@xUMmodz(Dd0B6I`PPmX1`|=9hm15cwZC?k2wkMfx0Wq$wY7$Pl*dzp8)w*08!iw z=*t#EW}e({Z<(AsJ;2)+b((%gp1@nSJboMUG919K^F z1L|wYY3VA~9!ZVM1sB-;Qks)MtTHQKEvp3VVYCnJXsEzhIQc zD)ZWU=Z=k8hxP2Sg@!ntKMp#^9C5&i%64kNaE2;-S!UhsuYQ|vk6t?ffnqpDW^{H( z$Nt@lqF27?gs^2*>L%lAv9UAt*fCY7E-bG8%#(~Z7jX9yhaXlA1jN#zeGd-a2v7Ai z%*Yn|ewetpI~7BQj>!vqtZ398x!~Y&Y`?OcV=4!SoY7>B`K2Z z{H_)P@|CVV)FY8P>W~x1S9&}xsn0$JLz57JWV}{CQzCh)6x1su!FTBK?eoH7 zR2Nkzc6sfs73NW5l85PtO*FzEPG+X*AR}g@Y?y$DRSpK>-b(N!DoPtqN}cK=>FM;p zeLjg_uv@T7@xUjS-0Al`M$mb6GnG~5UuwAn#+lQ}ziV`fU`$3v+O{1d!r95(G(A=B z?6%2$rR~_|{+?PjhMyjIR4ICP1V3?#RQLAN2Of+bXG477pjwX{B~4`_g6z}$0`w^t z-=6+#2Da3inW9qovhb$e>mvgIWD43@aF?xuqoFyL!5MkkDZl; zpL34|%@;-UR0biPm@Ydt8~R)0f2q6OuByLSbn!g;uIj120Uc59_f}9%zM4he*4Eqn z$Cd1lR7cQFXCNz4W}PqLS6k|o^~lSSMy0xQ74NPu267`!#YaN5NdhC^4*x8Ln3SACb>=u_&oeS(Zko;EEZa! zSc<=D`fNwP*cFz!jl#Bn9XPULYg}BtTrslL@?R<$?|r_h>mzgVYBZ-lcGoB`Zt$Tp z9P!*lOq}gx^29!sTU-euAJjQ3Rj6AIlj-I}$Cz3ekx2~d!bi#DDcCckvHk_0?x1nv zYF2SL5^~~h8EcRz1`rl9!k5MV7r00+WhW^C?qo@K<}|j^m78<2s=nN|S=x<=oOtrM z85YPAxME4E_v46uLUI6$-&+N!n4t?oAQ`RW@Nd@7z|-{BCnOzhSKed!z7Vn{PIn#J zHLtzamd=@is|KZ42xoEW1MsB{W;)s!{J}#}x zOFQE0a$Ebjv$j0f2~iL#0zLqVBy)kz#1NM)T#XDTx}&aE?V4wtQ?6MV>g5d!3@wV5 zxqmkt%ZL62LP0VoN$?%GFtkpDO`2cKuL4ymRVbvyg4?pg5Us)3%uJ$rFmk4!DH;Bz z+5d;m)MWI$P36l)GjrxRqqvcv%c!C>S16X2SA zK5$JDBiB`W5EIKDa!QZK2RT3IIYyL2JHl$->4Aw0x<6Lsf{f-w8%`S5)jJtIQrSno zb=U3wdkdH|d_a!SO^98uy7?nctNztsRTJfmoq3KOnb4}mYjAw!CkikCz4igkD@%P! zU;m3{$@50)q9=a+f|bnu6tVX*M5>>Qt|!}7Yk<<@k4s$oKeqh5ZB3sm%G$J#G$MbN zs_L$7mG;5i+pN6v0{zf@HM`zoVq!rhceyx54pD~oUEhRr?6SW}s#+q+K8ire#+1Id zVzJHDSGsfKrlWR*RaP#RjnhQN<9Tc9fyddOs$3F(w;ZoHNz1ny@~YYfPzcO~V~|7% zAUE7>x-bn%(4Os8N+USw(t))P2(N(}_D4Q1Vl`!qwemlnziLF+jKq zcunW-qq-UhaqstJ-oUpltdMg(Vu1{>0v1seWac~dfq?;zn0srY)JiL@qYBT^Pu-%X z^)sidRCSrTu}xB(CvbCnk5~{hSfz9K-k7Qu<{vE=D}Cs)kdDU#Dw0`6T`{SUu;7lO zgcy`hI4)UKT^DAzgUV9D$~n#i+c#zNUY?_btQ-2AZ#D~lu?ZEN$KJwrmw9%EfT_0% zwAT3uO%B;6;p36NGgJ4p+-G#KXuIS)*7c;Vv(u$d!kt*FcZtXW%C(#MzB@aPg#?i( z;y@SIscb{{pnh@}q*?cG1dq-v!`=Fvh{bl{Q(2wg{z>y`#mo2T8f;6KKMK9i$dEY< z<#hV#szg&G;NKgO1V7Q5=^( z*z(7%<5R1ADmls;fuUojH>HS+nA>Yq&ej{SWO z^7VQ2OH4O@U-}Do`=a!f&zhXLwKHl>t8H*k2oKX*fK*KzQt_L3!!y>$QF;+013JWh z1Rv^UyXd3!(-xh!Bd8={)NkAwSZI-`7mdsnWPV)`83jVA{ZXCyL7B2YP2?u{KnO{( z1fC;gipc5U5o4oZz3X9lw0YMkYkwjC$LTBajh`N0L3N#^nZ5bz2Wb438n8-rzO%Rk z(HC1$H~zbb8{oWDsaM*SQfM2;j%louK8VHdEO!oD?Xv-V$?o?SV~3JaG)p<-#!wCv z$CoF4r(Q*QKQccgczQ*)O6tSJAp<2mYA>I14N!$5!zZM{S;8wF%iE`297MB>;b4?j zny$bZg0*^f9jvz%*9Vr?6wc4)|i48%_* zz;*My3W8LYgckfP#3K;dz2Ugx#hEA_o?p?V|2zlP0N_b28xlt9En;&xP}gHYlu7z5 znfvgkh*$OZkylV`(*;E;j|c`B|CIRC&n$u4#+7ZW8UnEWxxCIZdiJbB%o#?01wmkH zQPWSmo^M`RZbK`0o@*8lmg9=pwqUjFF z@q}+v8FYxCn+i;uGBgGnp`1rGs+ltIf5`yiquW9vxh2G5n0);3-jIc z`v-P989_FN6jb-r7?$d`2oNtZagK`LtXqrHyc6o9C_!aKhvfX>+qDl`4s(pVganuU&z;9ErCdD}5#K64*XfD{ zoo}-3G9-(cjhL7zv2sy^A@qa1V025HOd;Dw4l&jb4{~iJDa543F@=(_0dxIqu~XOe zXlj?@AW9_#6M~NZ*mpH0U4g*craJ6YW8H8b7MF%XI>qB3q8Z-?;fSi5_NI2Z#=IGD zi#OELAmSu%*MjzVz_DC7;i;XKW&fG;wR&FEv&RHv1=PxE(UcW(!Q%u?r3W>egd^}i zwLi!0mOyPk3?s1OYVH6%^(cK&{`o+wv)L)^8U_c7E@+A0Ma7zjs5IoD;SiXMlxrAf z9q;$=v=OYA_=6jghP)T$H}P!(C>qkBVKCDn^i(4g%u@jWe43}k67&KHHBZ5Qb_`faP@;6;^-$Xf6FP=RAP*iQ~ z&89)^)k(mKd~TYLFZY-S#kUvN>5q&?hQhEq1y(;!K*&FIFx@fF>P!W2(xpetfmT3M z|NBg-*$NhmueHlnNw?_aQypTHYNv$3qLS6|oTL`_QPi7vMnzvoQ)_Nq<`D5?mHt!) z^elm<{Kf`4WTBFhNN4gkYkg7T>d$}QuM>-U4vfMp)c?`@c=(H|n(ZB$n|e2ZspPNu zwnn88MX%DN!KzY#VtyhK!?tf;j#(hVPwQ7hJ~ddSd;3TKZ(O_9M8_c#4M;+M3jO!< zmUJBFMzmIq3?8I;a0t@R%hP8gVuuBq^8(bj5uqy)Sx^{I~$9xj!VaJt}vIH)Hz^rrx>64}j z%QJpqGRE=recAF~tj@akK6lQIsCCY`5NQmI?Z1{N@SU_$IJvXb8Di^6tDwrnEeJHK z#No6c@kl#*?&0^P*{Ukco?I!0AMDqeOkLW+T9l79r1F)w*!8V5FbY#c>Hni`3_%Yz z?cZCPFyG2_{4eWs=w$@jJfxsJR-Hk~s!j1&7Uta0pkU*DB)|Ca={*V!=(-bk;}pUZ zX`pqN7!R*rx9ub$6Ab1N6O1h5<)R4$Og~wTp<*oPd2srv{aaB9c+{>{sH3t(1GB7M zJjtk9hz6MyAAr%yPLtieBGvvUPa!)*yH2`?ScCssmO|Z^V5W~S?W*(7)tM9DLl(q1 zQv?`8e3fc`$6Dq5t7?o4-ThUAYbj2THd%f1U|+m;W>>s(dB)6WiKgU#wkhYX45|hi zesUUuq>2B85;$h=>r5iG-T(Gl)$SFRDeS7 z{Mig+M_<7Dk8TzEYJthwz(%gY>N;;8-ibf?Jo&k&T-3e6!M^Vn2yh&Pe!8&-=~ljuhiuM_!Qxxr{+=}}%RuX@%M(AH%OLS>-qE41yv_|P^?$*O z+xD)rM$qR)2mbmTyPKTBF)7;frz?A^di%51eUJv$7HsxLVu(5VF4PP57c;ed1FV(l8C(^U%{MAE zhP}aW5Mzqm7-=iY#WR8vZ#lS4lh^_Nr&ldb(68jdxZxQq+&m!F)I_nxy)`iv}cy4aha{^ z96c{XLWX?v(0v^6`wn2CQW25BRxQNwNbyPsMFfXah|c%pdo_Yc2AHUIsu6bG$Nd_6 z!|RT$Gw1gy0eIKWjwFlI#R$Y3XvN8xW;WH@LG*>2yRw^!HLdX~UAjTR%2e`vpsvj_%m}`WM&r2WBWP-~kjN*{B{*hv? z))+H%gdc9)mrM80fWl;ig%W^S4Q>W#hG;R&Y8b|)z{gJ-DOh=kz}-<~5@Po(CtYFP zQK0=(meZ`4=#2_R`6;alAnhP)%O4mh_Pbg}&=g7ZG!+0I9|3bbbhIOPUfC7Wetc0B z0laxp<;@yACdoI&D zP5euNOl&fPQR??oYzU-2b3A;)wOfydtCzy>_^X$zujH$F=RN@x4Ss9QFCQiiE61gj z40$6jAuZTH?{l7!m$yf^zxw+f^=RwrE-dLT3JT7@J>;}MT<*KyY>278zcnGv8W?EQ z$g#~5T>ZlbDSjF$mI;$Ck&&a%!TM=f$Ttp*dm?Wp2n}t$bxO; zYKRnt-sNSAN6B^#evODn?4F(+u?K~0i4wc_>YeIUcWs!CNWJ7bxqL48)30k8Eq?oD z%YD1fU$Mz1@h%1}Q3_Nude=NPAeMrI zji&ml?*Nq~0V^7-8r-`=Y-H$id5dK|K~q$AQ8wqTTsR%;p#vSs&N+F~xUfgw3I#WN zd?8vm0eoTb_Z)B_J z>qvlOLu4LLi~lP7HD$#jS$_gn;LmdMSxyxyi==k?7xTVqbjLeYt<3Z4=UwA#tb3lv zF=0{9?|Iq7Kb^u9CwE~(M`fF&N%sdYuylD{a^KJ29TVXM{hvvx{JGffKjUQ)a@T+m z7^KpWw4^U((r^VmZCeJxTEzU!au38B**)qB~zWk0+MNvjSx~Z^2il_l@PmwA}1DqGCmt`rnIM7-?N{{A=5F-6*lYzn)qv zM|&A7)0~!~{(Nd%jQt*6VSu$+k6M((39>BIa?{(tVwj6`s{$b(T z?fd3i914Q#YGgSe=y!PXeK;w<#Aki%nm3UP>eY7FvY7u3qq&8dkL{a_Dn%3D|6BiA zl>;Bh(b;#l9dnPHwQU$KgN}3W_EQBqbRk8coDBy3UwLJQowEz8x8e564Jq$q@E8W_ z9eFM49J_Phs2CGoJQbI^Dt4gdv}#pXf5#Fv2WgsEU!l zY3beW16Ja$QO4`AYYT$8x0ZhV#@*xeYwOWu3G_21#ON-eO1Z30wn0I(_5UDSR!0x*0z};rG}SmhAV*SF+|@2e zI)|k7(Je%V_flKlkNq$bRQ?RMLJJpHfmlO3SnnStv5Wgo5zd4oH^QH`GS;cGps?+?ft2}^4R$IC&P}5Si{%cC*tDq=!wB|Ezpj+2Ic>8|K-&BUbpc_t|Xp6w!eHQ zO#Y@3hUr<8lW^&5OcO8?Au_Ihcn1{)BW)gTUHC=GU*EKk06R1gH9<^G#P`!d+t#*z zJ0l64t^N9md~Dp*$x+M+-1(G=DHR&ratn|huo3+R=Q}2^m~#W7+>z&34$-no6Mda_ z&mVsJdhdK`w5(XL8tByDGeQiP{FvqOk!p0L@tjoTF3;zSy>-b%c7*MNCow3Z_B|u(s z*G5ALmp)n*D(Qp*#%+Hw+81xZoTI|)`gnm01t$+1AEt1fnL~G_?JX1pqzdy!zF#R^ z_m|e8!-p8ze6arBZ8BvL8)HX9-Iii*L>+Ri4w;v0qlc@1woS9wH&W)a)ZVB@FYSbwl@ENZa<^`Y_Gd z@-9{9y6nQyKKt}q`^|2R;ziyG2J${~n+`We1_(O5sRCdM7D(m9ur7g`i#LwEdy{kv0nDN<_ z(V&2wU93Om=%@`NoYJk_LBa>so(2(qb@3X0QMG z$lVn4^C$g60afKx<&1PkK>~Tqi4?p<8&X=tMUFw%K*%+W4IBP;T|=V4qjAY235()u zIb41}C7bW)G^IG_?UW*fZr;a{bc7)WkG9f!j>KSCuE#cC*a!Q#i ze0_3g6~-%Ky4jvVdz$_bOmbKvMg?PwHa=7?1P+2sf!Z1(41$zcvHG0HNlGF9QhStL z_G$A|bTtTV&W*3{8{<##T*xW>n0&5iIts=rsAlcx21aMs?3hDx(iXF{P&Cqbq1`v(X6 zBIITX0I}0dhj_2TGvXZ12S;O5&7K7~f2 zG#n9$CFgWN$-)G-5;12A@}O9>mRyuPR;Tjtj~Xc8Nx|FMdY2dtn&ye63$~W=ZK1u& z?FM}miI*PGRDuA=i(U(ouXAwb!~wZiV7%Q!#w{q; z>WV1lav3?9lJ!T)p^x??lQc%z3MIEqi~6)Pj&{lvgFF-O`sI@Z^@VSfbNFfC@t^nG z_=&qxoUA z4@&~s2nEB|4@eM^p!3>(oALFTPL6nfoq3)oA`~YE<+BH z(ij`=i|O9^4P=tX(p2> z^mARzDVLj7+vfv7#jrdcFRx-qvo`x(*LLNd+%RacwJXX;+g)ag82{xgMEYNQ_aC7r zoTa5AjbAv+s7-GgeNs#w*U2 zQaWSh?={c|Lo4iux%Q>NF~tt8VgjzqVBUTq42kj~Uctt_-N1SFATs~mZeerID@(;Z>FV3#CtH3*H#4dRJz{6p z*igWT$T)@N`Q47gjT^?JuIX-r!LzU>!)K5`Bk;;8!Z%^lvQVR2cX%UO?(lfretJGE zm!#w4kKyULyHecQF$e0jGG9Z>M;C}yORS>`QVO zqGi&Aio!?+0%76uv0#LVKZe*0w^tG-zfypV)1GD~BF7OExZa>_SeJ;eT@oIcr_g)2 zQz|lL!hQd-aLBNZz*q1+Cz$^0&dibxMAt631mOa58#VCS21qVFXhrj> z&=#ljW5Tew*!LdpAh%~MLj>-9MqEttOXNl8f6eL~D2IWcs?GfcB^KO~%A`h1^?%&>S= zc3lu&>u)XWYI{2_%O8LcCv(((WPbeS@2kRRYHRFr=A9ZIclO*4j))4^HoE9piuu*u z%R_G2ecbhiyazqhY8!84aT*el*q+?^zLQQ@R^0kMVpDs2*ja*+nC}IeQmX%QsBc+P z0hdNa*1W~7g6_TB-m_JFODZPS3{BWoe9Qe_SjhIE3FdmQ<&0^N)6*!9S)f|Y$L_P? zB8kRqQ8ta0v8orB=%Xx2VB_w`@(-`)YKEj{msh#(U(araI(c8Xhg{6|(os9Pwch3# zasF=*)aLA*9?4ZlH{&VNAqLtZU2*~%e501fhl(Wm#d6=Ie&bp#c>lGiS>|D0VB)WQ zlE1Q%&&(6}2B}|c%D`Z+hx&5aBjYF4`;l?b%6o2dhE(Z(xZOUUnRKD>crG!H*uDRt zmJzFTIx90i6Du!E;J2+!6)XIjlVeY+L~>(53uW@`1ExYj=585Ohk1BL>rphv6WpFP zJGb${UuUmL0MVEIcOzOcl30UDulz5ejh^u8ATj5Au`uQO02VjcFCDQY_Lt+}qKTiV zm*+k#)VIzq4ws!+ZuhFa=HUmV@m0@1_#xcc>~#w$6QxW#3I48G=X_hc)&vVdW^r6F zPj%6XF9vPWGuXOEn83ITjx|O}-x~?OH?%bkb@mCtL3wOc=_?_YddgDK4%<}}iw*B} zcqlI^rmGG+-Fv9_^6-00a>}f&rUf6>Lh9^ri}3R|hY4_RA2u(P5kWG(s$2m^6vZWfJ-Bla8E> zaj<2P7@O6BYqyvgc*SZHXXx$!UoYlA2SpIb9Ds!XA?qB2BW=SkAKSKV+Y{S%CUz#a zZQHi3iS0~m+t&8GRl8OH-49*$eD3F}?)yI1IlnU`%1}l{lu)jBXd9*%&h#pBo!@h~ z3Dn9TU)xU_6ZSxE%JZ^0MRMQD8|=U!X<|lJaVjDrDbZLP+}qYSDFx~kn4CVoFdX#9 z`5V0V`yq0Mk5%q~(xkC7WSKE+g9dCI1Ilxrh;Mc?{ccM38h~U`>G5aubN4mj&@)|v zR&x6r*%n^$b`*S@YkTP8wk+maCG+UTjwc_5q%O_#uE?Q|48H>B3vt>!zyLn-IK7%6 z^aKwUbU8IdPTJ-|^&1xD7ZrCGk?;@)l`FSap~aeg%-JF`ak(p8+W<{w4`6WCggGd}{D3nD4jT2_E%QnGoWXyxv;t>-+ zRkigXhgK4-u>l{}zuS^a`~O@6>cz`_fuB*0WsHZ6s&GG@3K2%X)EF@KUTXAME*sxM z)M4ez)9;>g{-wkeiSbiqk-JxbIL59kQSLe-E|6CRZ6r&ksP5>%zRFDefC$J{kqN_q zH8!sW|HI)33(LNapHq&%2C>SQRTRWTcRpmlh1LU<3Uv+A7GJ4Wa|{_h+)8MD@-vTU zy{3m=S*`$K#qB|3Jmbbb%O@-teS+4oM`oaznwO<$Z?d5HnZhQcO#CK+O^%67rc8=Z zrsJRi<0VA83>8|l(S;&7+SOx7$;4RbXK#RK@6ZO4N+!UB`TR^LWQU1Xc;dPnH$9=w zp*VyZe4sZfpof6J zhK5oj8G10{{ys*s?MR7ZFSibo7{gcmN}DuK^7tCT4U{F8;=F`iD9u{(7{l#lpi)IB z8g( zM}LkDw!}7Wl#G}Ayl0C@>~6M>IaLpm-`52FKCz|P1zuO$l$;o36uzsJe_@$YlzyJs zwfhMvGwI(7s8rxUx^V2azS!q$__(w3zm|;#zeULXRR?I;kHN{DxVt%disuSHw|iw| z!#-}C8**+w4oNnGuYM`-sDT5^pg*xaeB730wJY5%M(FT-QzdoZS?##$8*`H_^P#>^ zSJq+*qx#r3x!!lPzj-%na?RZV4F`~TZ2YMOd~O(F?X2bCbF(3pfnE-%EJ6!Er=Y3 zoqM~Wm0US5-Q#1na3w5XQ#;|jpK7!!kBLOdI$~=no^#p{@GUX#s?c>TzY~HjHWv3%K z30*ed0L`Gj8$GZ`dW`M9(LN8b$y{F97Y@N~)>W8oF5N+iZ&3J=e4_X>!Z?qYsh??P zgvpP`&xEONYl0tgYqg>3RU_a$9QRmeI6gYpschKTNo+}hcW-bVK2t(W zq-c4#N3$Ji z>ky+9`ZgE>G|T3+42AU$$sfKppn2D zCCIUoW_MdwJ;>*U{>VIol`8nnCZnh0y+nVANjgF8o@OMn34|5y#aoZ^8yy-Mim)h~ z<3yQpVTLH=z=*~jHb6v(&GX_1&yg2qfIwpHW^O3s=&456YU~Q?s|?1>?J=R_cxLHG z0|R4o7g~gUCDL>W6Z7xA+--=tx9xA}@O2PY6)fD=p z7w17Vla=rE{2E&~L-(3)2?eM9lCO*J^8tMu92~$3z`G{$0O1F~W^}V5??WNRM#)B5 zJGymUNA8Gx_-5#jxM|XY|Twfi+KaMkBD6s_J ztzbY!l7UDl=l5JVCxa*pW!cQisw_Y2bA-bGchvHHa~Js>>+rjKR?Ug8 zpoA04SO4`Li~pfS$@9`cLbM4c8euH%&9w`Le*b>G+^s3CFbTVNG<_~9j}Sd@rB}94 z=%5T$q^bi7o{htpr9USCJJ1skisbQ`7%?Ox{+IKgO_D~zNnm>yPiI$w(hG@NQ|>@~$EAI)yY?zPfB z-0cfW(QFm4cVt8rF47*fdid#cjAwyN6+jxkD zDHyCApcCtnF*`??0AUc1OJRY$IQE&r2BN&8poboaw?m$Gfm$tJL?!{QZ4<M#oQ0w$HjK9`PZOH8>HTXWkaxkC@3l;*BJK@A7iM1 zvaaEoXxL(AMbig}j}W$^gFCl})m75bAUuYYxC7yL_iWk|tT#h502~5jPe+JeUiBz@ z>eQk%K*i?3n@QD6)1 z3IK$mZii^-Eu@JxM5t?O(v<>4nx?=(kdM8!#HzP zdZvwc{Wk+!_o7@@-NV9hkN)}2WnnBJc;WdE+Oy45jQ@J6smb{>!qW3jCoLpaP=LM| z`O=Yrgh~EgQ?We{n~35P^dN|ve;0NmWLy}R>wXLk|MME`>W5+_s{)QYqXA5K0hFC* zY986G+GI>FVl8{~*Y!hxI}!+|`675GffP*5bNr?=9&tOmI9&kUAA6mfbCl(_86!qH zCQQhW?20Dnh_4__xA$cJnOk^bBWDCbF7p<4t<&Yb#rDTzN$`rMvvRceBgKD(ao zhb7!bu53gufnBq`Il$}A)?up<0Ub2I>pmmf39*1ATNWDd1ii|1FRb&pWXlNh3X|gG z7Ahzy&K{FgvgOCUl8uR$fU6pek=NaHex1|cL@Y{BPbTTeI^_?LR9WyzMU#tLIvrDy ztEkW_&k0J-0fGnW+vS$58GK!!MTpMLoNWx@Qd2aF-;qhL3>OGiF zpw8dj_{pVM;?sx;xs~0&spJNJCg+!%V#( z0B`^)mx2%??s7Fi8%LU#Q6=JIhC-rR3P~mEdh;oWzVZ3QS-3j(Yg{?9wtM9j$E-UF z`AnX>s(cO+pLtme7c$ArZ-TLkYE8_VQ|0CrNZRRcGi!yi?@mB+(I&MV2n8bp|lAgIs?NPPqL?DcX$VV`K+9Q3T9OoIi-)?D(FlZ!+Z%e2n@qvPd3`*WBS4 zwsbCUMXym*0fy{}X|Mk`X0**4i52bc<#p1{nuV6J5r~UO#a%6%4sqwfv{C29w=2%( zc`1Jzk+5Xyx* zA3k-W8G&L<^b2Tky*la-a2r4Vn-8}?v2Zq?Vrjn*QWBdn1rr$&Idldl1>|A9gFupi zfQD?;kf@&R-w2@Ef;xDRrl14ndG!|{(vh2uH7KaFXRlsnUdnZG&TRg`~eOvOAP+I56ynv^Qt4#75s8x^2TKpec1j4 z9#J9E6=)=>s*(#ybU4j6&FGh0Z z>+R~te%q-M()*{OiuP3ypP#y9b5t%fpaIZa<=uY(yty9O-vdytb;p>aC&@!Nhv!y0P zopnGOvi{4g@3em)6}Dy1D^IW5SQ)w;Oo-0*>DA)D#Xo5qGa20c_gLRoPJ&4xa{Ig5 z*R-y>U1ft2AXu=^pu?BSZ-qX{ZD3m#9Ks{FUeTmQ z3GzO5oA=W?$@6oiTu@wKKBm79OQ&hyYhx7%tJlt*%Wm`dv`c&bqVnKs%`M!AJ%wwI z0>=6?z`KV)lE){a3!uD&?mwG*eL1u-nLEp zmL|_nL#=wB&l6_uk3To`W?m5=Cb0rS{))$?bBfcOvHu6tgP{K@ArCQ;IOb`W9l?Zq z>47LB;Baz&O-_9klGzwC?PXwvrp{3%j%GHgH*7uMfej(?l1*gv$qlYf`cz6*N(UNIdmv4Da?)yJFpNqVaGtwWQbUQDTbl(%7 zQHy&Lyj+K$him1A&G&2ihXbr7sE6xnCt0-sSJD={u0 z6CO>HunWeU?!6B5fh;x`esx662Cy=1!P}eX5OS(OsOIYy49vQHW%5gbTxF zWsxXGkGvV%Fbh<>h33!j#LVx}vh+GPCxDlsOw+dvyAxT6FhuyTb7%|H{$3Wi&iAx- zE|YoE7aU->zlR0SbwqLE=F(N8XKMU3$YcyawQJ?eng%L?8?p%%CJus8KqAllz4d1r zUjQRo9t|?0Y4aNN2saIQ*OO0=EYwJC*ZNd3(8()X_CR9&#mXKP;L0+4A`45NbgsXC z7VZnlhgjY(8|O*jqKg~7H<=faLFEBy7xkWf??xT3rQ78LwByb>mjze8HL zblwO)H{c?)Ok`UGo=Y{)F(+;XjMXa*iE!l4$*&Mn*3|-pm1)C6lRSCfMn}5(aCn!M zfsF*H59*MWeYW%mUhn|iRc-to_&AXcxXQQb>DCK+0)z*8_QY{zk+AR&fp6*3iF}aXz0#_h$2$HYZE!ZoXnR!a&sub>`8T{O!aKl$b+%G%&~x6S zF46_m#POw{X`1g2Nen??miviwMjUuh+;A|=PikT9grMkkoV3xnEqngW69ka=&wwCZ zdkAd<{4xWZMyhSEnPZX>AKkX|l5aIV^d#o;eZfbHN=SoF;dA~J7LRo~zj05n76ra{ zIli)!%zNzh3y%4QcqE{~KkvvNBtcT=Dl*IL-%SpHZ8?8QSTZ=p#^&sMj?`l(=4aLZ zgcr|t-=-y6Swze|Mklfu;CX2;NVn@v^t{$yRem3A7n#K4RC%_UpM3mA2ATEsJw&eE z2loiOp5b4b@70Uda*y+Hy7RhPe?~}vk&~w;U9qrHU9j>mV5&dp`|K@#I-S%ow);^@ zj+$CnI$+|f#a|{Pj5!89`l!T5G~7{sOS!e(nuctfpZoA#7_)>+>%GU0m_`@uX&kyf ze*U1=3SXEXLyXQIZg2RaH%2WvfKmGe$eKmvQ#wbDfPGhr3t**b=VQ_>tu{J&(7|qY zO<=VUkcmK*o^SQaU=c+J49kW#vu3UmlOtDhbl9heS{yd{F;eeenU1R#v8q1&8q==bWzqpzm6`cr74bP+h|2(h&#uA8OSj|G zzq7J)VXrSW_E7VFA3$Uxuy6ylI$w|T3O4zWM99Y4@YSKBl6c-szIz6C398n@XP-$5 z)Rt30el^WRY|lL%66r)>93d)TTY7jts+OKP?oQg-*r!|~Ez-gLTrqi4%7l4tI}#k> zp=nzcHW}O?gt)Zemf1sq}99Yp}S z!pkBG35#5jT}T?d65@&PdZB9<`|-iQ`A%R2fgE4?CO^*Llc4J@g9CTeq+jW#9tB{8 z)W`C9q*^B0jckk8zbt9OUK+@!3z>Mp>5=EYcjqKMHD#T%Ek^f0&#$TRIJ-C(Y8lkt zYj%eV`~{b`w3kN9G!MZV{SF2|#WR4I0dk~j__At6{#K3o@$m?q;b}%8iMS;EPd0mF7>r!&36-I^fcw``Bn2Rq4_?^8y^TxL z2(U1DZnvNvZGComBN_0bq124ey_Hp05q#s~yc__iu1JN7cSuAh=5L=zNEMM2ea%a- zbY!;vhKti*~sp#4+4c)kLolqbUPD)I&S zPJNYXr$W+CHA40H0JK}*iCIBz?L~+gsd-4HaMglZ5MLZ$l4qKt;b5V;wxaEzp_Pt1 zWSD2z#7jTx_`E6Lh1q=jBdOm$(fAgPiJ6lFniS%l4*CFHbG|JHRY#yt#~L8lMR&f_ zASl>g`5mv?1E=?Su~p7$4Tqx1eUVxG6y&B*lAn5y>oMI%K zvvQMxadae1%O6`fA>OAg4v;w4+j^*F0R_5F;rKi+Tk1jmXm|Ya9Cuo^FwQ;Rhk1-p{LdN4 zo@kF}up*-bSs0aIx|(_J#+d{^8ZMWJ-(tx3LQN)>-^4`cQx=iAZ}(LEUnNRBk8{Q! z6^pae(gIMUe_fsdYobCJNJdu)!km5cfj`W5Jwi!S!P9y(iXZtp7A<`qi;lsvbco?| z1&ts|WhK!3zU3yr%aXCWGrcwnwl~JdlWGzg$}AaN0}3OQq>Cv;^`J&J@c-JPcv%>W zS8V4OoHRb{BMR{3l!-PAH63W-^woQJ{A#v0r+{KGMfPq&rfP}u$n-wk2=x@UeTGiy zKga8paW#%sCN-z_c_Tq@;Aso(?*|UH>;ARySOPl$X9(h2bLt5BvOOdlKTD-D7$KEY z-Xcp3PmR{JwA>-#`^yC$E+Lbob$+g2&sZkt6+Pmxz1@2ja{=x41vbdx`16Gdg`bfP zlYjIKyujUCmP{y~ML09qq#!c&;6ipb_t08y?sC&I(n+V^#V9SvD;X34u0ML03y9%7 z4Fn5+6$7BEqg;mW>KfU6g8LGD$&!xziD6>psE8h@kYq3qr=~gqv}xUOA&!n17bVNj zrfwT?1G*2!&K;!i3e1>4_Ek1hl?w@8oPt(K4j+~rN3BKS+j9HMkNjy+mJ0UEVzIbk z5MV*N-z!#nQaW{UE4H|Wa%zba5{Zax(yV_)bZ%uGrITD1MN;D^FcQ}E0puEWP$Pg@ zwcYULu&D($=%nF&+!`K6Sgamh6Vox+M!i8wOqW`RN|qxoEuR5$BAT{k)5?M)Hr_Rb z|0xKv*gix&w5chqPXco7Ei2rL^<`(Q&GVCmft$xu8ZP1x7cR1?%OFOFkMpm9x6YF7 z>@I~C+~kbxSoiQTVgyWgRaqY1)mCEI#n_#9F6k1nyms;z-cC>)IJm1m<7}uG^kC>X zS8gAiLuc>_`tmXb)7Rk(x!!n6dkS`D$2cCmjmzRK0;lcfQNq%WUD+>%5sORj%o7o@ z$m!$L`wm$tk>(E$LEFAH%FzM&(b9PS35Khew4q|Leh?J2gp1mU<#GjPawltig> zrCEc`XQlCA)*&>T2@<8lDICR;pW5&y%Lh;ynu0({pqVkb=z);*Na-XtCg?cPiQ*h8 zXn{e&*k=Bv;;j4RrsSGA@(n&#zYQ6H@i9>G$k|LJf+Sz_mrgx$^YK!~BO~4e8k1R$ z8f~%IQ>iIliE2pHI|AVlZTCy2D=`puedM^=6;CWJW06>iSY=<~&lEUtF&BVX!Ae1+ zfQg2kvvrjAn^11~j%Je&tFQn0-ek+{z);^e0R- znt%>EaVj}3fYaGv0*JLoLJi3^v+Ww`j7}_rG(@7ZL}K;N9T1?YkE#B1h((o;wrEk& zB9LqM75cFoS|2=595(vDLF)-KGB)1{<_vXJ06&)N#oJ_I|sX4d;4{R(CjV4+KZ(XK#bB+vRA< zfk!c20#vkzvxMKtHorDuE%Y?8;6;AF-C|-o^Q9^8;eR*erKpj5D^x?U2SyhhyB*vA z#1*fw=p*PKAR)rg`Y7TCPKc0}x5>8Icn1}SSAn`MoK00%fA&A|0;#oDWE_xm1WYTM ze;{zz)|m^V%T^o|gtpCxL;?KeVy2p+KDHFf2`NrM5rDyPqk)|cP zF_wcdK;mPTee@D0He+YMN9~}p6R#5JR-F>4n(`*@jQp@NLGkM%Zt>80mQS!w6& z+W?%tDZgLa$kEgG$wzsoO(vi88_)+H;b+i@;CbGRMV1cE%UN8E*bkt367L$Hamq4q z>HjTPl91%0BR42Sgp1)-JDZ3t!74FzXP!O;=O0w=JNBnS!s7=V^`BBT((FeRN-y_> zXo=8F&7d@g_KU(6tHm;{fv#0s;i3v$t_Ybh*v~m+!5e&#t6E-0Hh{7#`Ow(+f+&iU zq)K>reyKq-6Igv+cfz&28Z6-*zDG8v8>IQj^v{NRma3vf^Nbhwp0Lc?UZ(NHnzO5- zJwBXYoW6HnwxohgIF*=dLc63^s3Qk`A!0$)*s0gLhVoRcSna9`SB&qiCOgIoFt(K2 zk%m_)M;-7W;ubA5L^|tKJkEuh|9$Guwtz6@-t{=lxFs-MVf1Ttg8EfFOU~LzBNxFoF=SE2W(U@q@HU3E4Kl$_<=c~KIm05K6HCT0Pn<^Hqj*J zz!AS#emf5BrI~k``<>5*)L;km@Ln_65Z3>s55(o)EUn_`i?}Z7(uy@Uwwbrk90eWP zF?EWv>aJPqlDCQjxbFZ)g=Qho4t1`pI*A|$}d>! zjrZ4(KacDV-O;EiAGC_flwy1LJ*q@nL$UL3ShINN*b}xl&N_RJC2WG7BZ|8nNXNg3 zzzn7tQPo9x)#l1Kdo$zP<77F?BqUqmTOO>z(~}JD5M%-IB=!BwR4DENy@7gn&Z>4i zPgDKwkN&YZ%5Z6?KUeGn(n}=B>ulxf_9DJ5iT>$i{-ayuBPV|v{wzr4;;J>!8?r&U zgK}g<%f^^e<>eU6%Uev2`s!id7oY}%)c_l^BedVz`_Z0B_e-mP=+&Wzit*giqJZZ?GJ%_ei}O5 zdi=dCMiI;G9ExUBkbf-xr2mnUV(ELYfbRHN?HK-A4JG{1lH$j0mDXl-V)lh(SWy8q z4#s4k->&flBTHjs!ooYS12UyIUyh8!CSvdZG*t23FL?QNF+_J>6BAhs1>oGPg&hm7 zT0%pRGbgEE9uvH+3vcUSBamKp0N6J*A+}`$4vb(ld4xrmK zz*@-%_8?(mSbrANB#ySXUbYxN{v=>aXA73_UEF;;>Dophy(b+39rvj_+rcv(s1~n1 zOti9g&olZgPp!T4Io5`XmV3?iB)Zyki5=r6JcLo<=9QeO%u0Q?_}A*W!7c{50Ugia zC$q6QB_}S@@&q|4TU zR`5_X5)n&eCvqw*J-ZahNl-}YDu1CmHr1ulz_mI3G`u&=dKrfJVdjZ7!z7Z^&2}MniXUmLmTtxKDl^Q%`=Lxb{ zT!0B3ltjmOHM0`o+X`5EDL)-*$Q*eAIWunw51wE+hs%E~K_-sKodqj}JG8h_oYj`W zcFZ`x*PUdkK~Goxi<+`=elo6}FY2(Y1HYe7l!kUf1m%nKh)4}XrHn%Q9lMq;x?C4L zy^qwWqbK4SvavAoR3;l3tfuuzj5x|wP2jZd-57JwfGAM|;k;#ml@5eMM%xYX5Xz16 z4mL&Ylu;NjXhZHN$;`M)m_-jYOd6@He=V5a|LI?BKR!(xfyl~$qiqFhiBO4i9bobV zLqjHYHTBs3rad25Fm!ue&0xDKIjB_nm(GCm3`Dwo1DJSfOYX;r2s%M(|(Fuei(U2|}cwq7mxLvFM za%3W!c+dvz`-XSmN9U;x__U+1>~k99grOr`^5W2_Q3Hak8TWV(%<#gUmYGHW(rLOB zmVL->f`hAeF7O+b7m(FL{^e%$yT0|Uk2=UZ>=69=pRhbTZ@G9o4?PR1bvFe?k%($8 zOygBO^@7`Y1uC|3s<{hf?cwV_v){&xB5SMBOhPLp7(JjWRgQA)vl#Z|@|@sxeQLVw z5OrN*zL2grs4G#s`NGdanotZ-4OX=jE*c#etxzrPv!PWsNPH^9A=-FLfrSw=`!QA& zPTH3YY?lbRHn+b^dRB;oq2VY9P)TLdcq#DnpLws7q@VrD8F{7s zC8c@J*gUQ&StM35$~YvO#=|a$u$$A%g2OkJeGwg9&a}7%2qLAx4aMTGn|C7C#f{-f z?DO$WSOpjr;;Hddk6!&Y=iP#=Ym(WK8YkQ835-N37avgCj@i(29I_)?_|6X^0|eiUDH{203D2T=$o-L#?|_EE_=6wOq!X%qPD5gcX7MKSjm? z&W9$uRg$0uZwAVsWv8HchQ(N3gDTaFfAzA~AsgqYBb2`{i2G`K4r5@Zl)CGkM`I&2 zd*chgw{^x2uJa8YH22{{eq~YPZIbnXbT7oj^2@B@DPT=3odd=34@D!f8a(?mak2pK zn;7=}wrz?!M3XFPg!a8kT-2Afx7OHIuC8?V$Q@J{Hl3*&BQt>2VR={@4U-za=j+vt$ga2V? zBI%gob9v$&n|`CzSFQT{acwq|To-43&;>m>^tD`LSZGbOhe$6{I{6!a(Fhg6ieeuiBvD}@ch?zFjC&BTW?#h&x~SQ30X z^)rJS+gvp3@98g=q}yXpsa&ZTTa8#@>ni~-_b`dTk-nESh*S2PvtqBEWns5(#Sir6 zv=1uxNU!sarpEe4yw5G&(JhYk7kvVabYv%R4~CX`1-*?xG`z=!5Oa!>Jl5Ai^ah;k zCe{Qg;__tEj0}HiE#j=wr77|?KYAk69GV_bERyb^nb{I`eK5`~^Q`~d+6s6O1^<^5 zgB1Nr^aRXf@Z>pDAI@+_i{G*ebvK&$(?f7H4(}=~1N>JFI8CGG`=*0c=-0*;8)rhw z^7d%2lWDS%5F8yn#`-;L-~5nO2*{;{|0JTf)Pgv2uiBbKyC)=d`%NplZ7?9g6cG$L z*UDe^{fU~GD&sNHut(`DtX!T6Y+~?HEg8c`V;j7i0I5t{j&ImvyIL>Te@gs7W1l$`%>Cn`@ocr}ubnCQmiOWMkN%OvVrHC^IIREm7AC?Il!YO7# ze4d5|S1;t6v)2O>TsLnazMUud_>YV~vrsWPGK&h|3hZ_VFDKu>RmsG8?VqJ~0({TB z^0LsR{?MPU;g^cEO0t?SXJemm1};D1thA_1d)YADG;)aK@TQnW^|`^W@dGCWf;2Fj zlKPy1n}%IpZ3|sH5Uy=0{%xBt`GVC32YL<*eNh%bFA|Y|8djmq#i&V&y=uGn@V$RA z@-1ox^`G!kE&Rmy*GT?_g0bIorWStDW2zLzocdy=gW|5&ThR08G|p41Slz8NIL<*sqTpVn#yJKNxOj% zWy=o2CIi6Z52f>Bq=a68+H$50v1`|}4e!C7?@;)TC|;iXNPUrn;U!+>+H=RMQP*K^ z?rfK>(Jd8xvX+wx)>bRUafk)#A`-o7YvT^LE~TNGRY|6e`AMj(_6klh@;Vw z0XV-j$1I;vHf;&>ly7hxTxHH1!p2Wd^KFQ!XUn|QBoio5%|(sDHX_uGPrnFLnw9WB zNG`{s(U(yD-n$Z0e_@(su4>&>nRuoKf(j}-1luFUuM zY|z?^^#E2@%7nRD`n@&PkP&lT0=p&d-GI%xo)j>$0h9n~vuL@t21Y$Kqi;R375Kj( zQK;6@*jxby*gnZXn6u_pV~~a*MoDdWz*izqJ)26J_fcHD!UKpmK7Rl33Dx;(Kz{vI z$SeFKk@UFp?Gkoqz?V{}@Kt+12Kl*Khqxg35_0P2cNpwmS`7}$=^VdoguapQ#P^et zG*Y+z-z8wiKhegnT|!D;JnxHwoMPsIaao|X@5r1C-zWBjDO#6B{7p8I2JkA>+2VS@O@$&H|(~H&e`v1vc?4S)j1pa<+;l#M&Kp|)r3KwTU z9=DDMVuMX%C41c5Q-Be^Z$h&S_QDgX^L3;AdbNjN)O&H8ldFTDoTw}nQ=DvmEojXG z0!vT}PN)L)b(IvzN2~$em9hZ^a79=0iQwItg5A=rc7PkO-mm=#J0L>GG2p8D%Wi)| zt*qP|S%3Jti~BH;>F4MCJSn7>%uNFdnt}*&XZIb(AJj*c!MCuq^-9C15I+JZGZtmd z&veg$a^!o3lGcz5IyM6#htx?^qM%~L5n%Mvy%J_-dxR)U3F)PR{4LBxPMtova&kQq z8ASDq=TD6~(Nu!HCBwDp;Ze$K%cb%e81x90lH+TaNQU1gC@DjTzJU4fp19x7cm9K> zUAVA0+(>^Eq}1>87yne&T+t))C706dLbk=HfIKraws3baNQ5{IH^;xXyL4}=4OJX) zXXel>_`Os2Aqfx%Utui>5$m)m)|tS#%|$vxjRhTl3OXk! zfcJ{NVDyRO$o;8C*JcH+1U(gc6h*UZ5M~iJ)n-2x4wJ%4x6#(+ALuy@#Fad9MX{1y zJChooldMi) zxA1>F1c&!iVmmA$(Aq8nzX{A*UR-2@^|1M_i#BZ5i0ZPyeuP_Q1AfV!WNwWVHam7V zi)>UuEeSwrs0a8r9I>@gU$Gqt6`4nqEGa7TT;`+^uwN<8y_!Mj8C2RgLJojs$qEQg zEKCA3vxmupxdfjE4s21i75X3AH{tn`YA0hv(llTvWCf*b4UuU$-Yin;+x33mb&l_A z#a0*q*Hq@Qd%y54kyc3BSqtsUm_|tv`}(tViYwl=V^vO`e1)eKQ_*5iM93QPsXkyi|1s%s$O5&Add7!zj|dh-Z#lxb}0T%f^1 zj4|RMaQWYR^@qGDuR!j-?@yV>)E_$r8=8X%zSm1&d9NtM`)v^14S{^^EC)U_BIf*J zqV8CujzA51r5%SY6LSLfLIyjHCfH_lR4jzdD>*GVvAphQMH7DSj(lVJ*gOEJIMm-? z%{CIRT(3`Y1ir#lu8(43-th55C3_#iwS6TOYDL=nP%a^ENc#|SGXw|H7JC}{z2uk= z4=8lZOQvbNMz{ZhU-3`kZg*RTxEN(YIiGK*-J<9Z8a1G0Tst8@IH3x5G}7Tj`m2+m;|2Pp)Uic378?{{xRT+*b-FSPU^;>(~f!( z1fBuw_I(=XJFR}%NYKI4P3QP4ArQX zwslvYajaIdEDPBXh*sV-BPynur0c3PzEu;$tae6oiNq>LCu%!>P0OF3H!Q3UtHsZ; zLGdLC8Y0{4J^^j-&yC>5$OQhmJ7O}F6TIUG81o+srPt4kevgQEkRiItte&#Nylh|n zzPTaR@d|e`iID<99tDTRskJ-OtF^leHsp+qGt7C%1Op>5V%}U&P z-3}sf$IULAUvlKG#rgLOub>yWYBj#q%}{pFcYK+`*`W_7tnp#$mFei1b?TMOfC+Wuobj!t$ckAI9AyW4EbSZ!g z-mE`1ZR7OSgH!(ZC(afLB(2S_!HXj zwNp@GqA2B6ZSweT4~CBNJrOrA0&qFiO}h)16HE8T|&U8q*eZn7sxPkitB&i6`AIN{>@n9H3hmJ zJMjO?su;ap2j2)&2bqrnR#L$2BTFPWj0o=6%&B$*FV|WPx@t^pyQB0ipr)2CS>dW$ z7WQiGmT6_jbA961lW$2xlA(&jtqjME)Y^cy?bE_Xq(%}QzT!vFdh>dXvI2A)V$ui` zX09XsIp*x3OF=7En(BE?E>6jnI~VKGKiGAYl(E~@g6H}=>7%q{SGo*E{3s%hgednrnyozW$p9k2NF5N^j@ zeMS#@dQ>mat=@wNps5zHWcsHX&!jy~${6vx$9TQOc~TS5E`SxTR^~%B0G7FHMxXEpS{vMuFi$A7}`fNQ@XyAOSO#nhn}(9F$2QZy6lGE zpc_O}K8B2FIpaz^T02-OXYAJt0hYe09-ZFpt-Vr3YWfExka=a3JL-3z9Gf+KzKiU!oCm;Cup@X@?9MAtMNJZM~efPBWyp_p} z=Icc1a>=4fsM`E>+Wo%Lu}j1f3jlV%J_vHbt!@w7Cg?oC=aXom_k;vmpeWzj6Npzn zb1M%%W1k8BKhW3z9%$+dMaR6dw0ry;D%p3_QV^=iv<+_tHrF>-RD+*V{56F^Z@!VV zw)7hoEx><+hvu2y8=+?bxAK7amZV&o{DY|?M-HSH?=Mp~x)3AdAUW$-H+8>+UmX&+ zu%i3*cF8BFUu}k4+WtT<_@YDJg*BbA0$bC^bMF!KV2|`94cO<@kn`dS-cKfq1csZO z)Eh^3nn=Qfdf6u~_5Jb?!1>z$p`=#Voa`B^Awyp99~y4+sjIence_gZ5VnFA)%Vt3 zaO!7(rj6|%iTgB{ih9FCUyzC%DkW1q5Y_?NE^TFmijs7>kjUO;`x{OXr z41d1G%f9zf#pMrf*9Sg-_{i@o0fw68WGG^Nq4AOcB>M}7dz)41Jqi4t>ST_4?!@ zx4TC-O11;%s-_6I@I>romx=sWpeR4&ma8f4LDG%eW-T(l#7SgWf6GXdSYGa3?32Ih zxyOghGt8t{mNXer3eq*Uju8;*aM}#s_pzlK>M*L!@Cr0o{y%zIz8NmD zmjY&ot6jn4cvEs0n|$ttTf*=~!K;cXfox@UwV@dPLi~R`IsoqPw|SXqir42~#_>ol zGO`PFz(`w>U&=?Fj`Qd_pvtjvR*;x#42~fFBv#9d&@7jmqh9?zy%hY6D2^%Ak5J^! zMzIbnzi@eZ@X{IMZ`Mlv+7pAAo8@{s4N5cyccfkd^o8`F*V=Co+}mx5#Uiwz2;67% zrJM|(Un%>Wi{(FTbISPK+ilWpsGFp;kKMo)k1j{*?N2c~o*?Ly8oQd_#Km*1bql;9 zaO^D8FG>er^&GafsAdOh#wctfz<)gzki%R#A-1_ zp$E{i4b}xN!a?0jAI%@0oP>G|vozs1>^e0Ua=B=-Ce7?W>b_guaVEtt7cJ|=-URFt z*j?TrFi#4PaESJh=OoU_WgIob%X=o9t?79}KiYO2w}V3i6s0c=!GVa3_~F9?e&*v5 zstO$0c;i9;M$`#x8lQjB7Bp)e?F!M+$~On5JZ_XWUu$);m76}@@$U9p60@I+dAK|= z#s+UOXgU{rTMs)QD?u@HD}=5S4tf%Vket6YY>%!io39(odtmhs4F}8i{{Sw;F__sk z%h(N*f^cll?7+Q}n|hSlZODf{>)rk`B_NPONUN}wEzA|!%?+C(F9-ZNRiJ|J|*XoCrv(M*`|)PI1l>FnPj} z8_Acu^(~!uK6$l5te5WL}Zs__@L+@1%^{hLDC<}%dYIJ7JtmDA!&76JR({9_yyAT$iAX{B4~5#Ve|M; zq~GT;$X~4Lk1@D}Uv{co8yjK(&c`S6H6;&3!*OnavHAgwItaZ-{2Iv}V^7#HWZ6*L z-jDuBJR@B8V5}5NN(S9kxMcGQh6*vip8Is zfkY_29N=J5nU|+asgUAf*ZqD&w2Zr{I`yy-5wr6R_aUnY&NL+|^O{y$5sFriF+pc&7 zEsF`@;SodP6MhzXv7Bk{?g(qBr}PZAn1^Cjft3fJz~ku{5eyf@@}dQMEXoEdqj7hZEf1cpd{WlP2TPRSYkwGTYn`;8d2`?ji7_~`r z5WCznJ%L7u-&~|}{+y0QyEb-Y69xZSV+%|xwAr7D4-<%V-*jYUJA<={c-c8eK0>fa z)j)0W(5KBzgIfu1uu8J|d)X+FhDsZ0;NY-u7zPHT{_y`CfS8`59k2_c1!GG&b=t!N zhAYyEj!vl?@^Gu!TXk5pyraukmSMWc%at)2>>Gywkh+~-Hl)91IJ{e4Okpo_T3Xbv zzO_MEZu1KeIjO3+2|=GQWw z-gHaK$uC)M0%+-LF>z=d!qA=Udt~kc4>S#Yo2&{+v{A8Oorr}d9~wRUiK2FG&rd+B zYuhL8+BV809xVLO6E6=8m=ZzH1{Jb-@V?mS0wSL54m@hRUpnBMxjPqS+ST;SV{plI zPg4-S$fgk9Y35c)5l~^|;=>**K}Un^ZOUsVxp#I4^i3?6R*TB@+7JJ8}#SHe?Ry?cdo9o-y!D z&gFZOhI5Ce!BE=K^>=P0F>z{&aHg62U1m5skyPOAoa*SrmK?Rzx}l6DK-&NvWOB&6 zKw~%y7Dxe>r!VxXNMM<}T%ViK}o_wC)Xz~Eb}G){5R?(-kQ=L+-GT)-(C0PlW{KSpnA z@${QV!S6}vIoWug?LHf4R#lM;6t*E1jSF5X zZBCOb#uM0zjg=%ycQ+=+a_rc2Q#E(*v{?nu5+b>UituJrdEc7C@VliyR5XpEz5mi+ zsEYd~NzBIoOXkQdjY8e8?(4{&N0(?4{&Vy+>B+S8`^_70mgI+8();~bj-mL@3(wCt zz+{n|QK-vG?hq`@XdKR1GW}B z=YSBrkFQf|%WrZgc2}zDXq^OIBD155^2QzoY>y4S0gwQ>yL4{ig82i>4mEyJ=lMg&F3P}+^5jnt&3=-Jc2W`$ig3R@;B(&*eR_Axu+-X0JsI0&y zR41Ce%~Y8uCY#iVYCuYrD}U3|qD*M0`utQ=PZci6m-37JS^TrGT;we#0HZj7GvINd zf{gOOTTV^+hcZRoEPzS#IcwsP{jOWabOlTbzq;h!fHHr#@uv-$OF8z%_QiKkO8HCo zFN_$H-s5J}*{F}ohdK&%x68U-YuL4%poI#L2^LaFO7uv&f;B?DuoLwx#w;U4JJtA+ z&CEs{ynhIEIM2N9sZ;^tqEfsea4E@LQnr^QnH`R#oFenLg4%^v z_O7q87=B8B!U=YmA#{-DRC3!^=t(M+=zm(z$sj-pSB;e{`~j@%2I3Z|S@ueJ#p%K1 zfbf?GcMvYMuFrkyK`F@xBUyPia@xsHg0c$m->qms7G&L`0iL|wiG#~*#fgk_wmCz&#mNY9%`_2ZaukZ zAE$LMFrWRq&I@&Z0FPEM3Ff2ogLnSVzvc%3zsI_5fZu1G9>DK8gXv>Z(M^w3%kyzl z;Hfq5vat8lrgbjwpAYrt{I2Ul-RCQd&rROJ)0(K}i`MZhyCW0xY>?BvTz(43Z3aoV z@DhTqkEr~n9z#ga4SqQ=fw94IM^~^ja~T2%KrAfX+-z1XUJ?0@7a7DrvMH{8$ywU< zL3asmk?C37rw-eu58fqmgANJG5(Dc&+fzonU1ez8;{1VLG_nGA@Q#})l~)zQ_7rKYtkP(|7!iZa<vtHMCOxa-M>sD9JBD9)l^wvJzOroVFb7Th4U8j~+w^ zv}%%ZiDB;(P7O9|QG(O$kazdk9x4Y=1N9Vf&$tq(HFI(RApNdaBIBS{qd9qdu?i{j zP9e{tg&Q_|WM|7QshDSw*VkQt!%?6!TtgnkoCYIONqyv}{+of)kDsfyHJC-&*p}2v zc&(QlvLlF|jog=|P1@7MmG~UanUw-$Jd=(ZV)J~cPeCDGwYC{XX*PUJm|E<+LN53p zp)S!3!K0;Fl5Tf3xs?27YF-JMk7TWlTZm?Lcsu6&B&=~3A-Mqfn>QW)S|ua= z1H$v)+x!dW^J}qn}7az74nzZJ7bw;uZINgJv{(t~qg5*%gaoBRo(90Q&kiBt1nvy;(~ zQlNiO>wQ+f$GRhQX$SaDqDVu3=AMo58uQs!aJ=2Nyy`;3ao;;ZTViOUu9o2jLvjqk z%lnryD1?Jz#ZkSXHGrp<nT3B%v~fEVzHPr^akD_Y>5U>4=T~+#<^O~@#lpveSM+-4)>GU@b zEA&ho6o;THj@nnQN2evpd>6|&E-(a_!n%)K4N1=CDj*((3I+sWKqLmhCo8=E-$rJg z0lZxbjTu6Z>-m2E(}ch4wqeSfTkT0l`KS_#hWXqqGkR)f*5zs6{zhOUm&^P2r+lvL zi%jOb@2NsaGx-pfRBF}-=_;>G@zW(BPIkiAXXpD>Q`z$@Aen`C+5hx?p?Boi!yYLY zw&@3p(&Le{S>us*tFoaoXy{oDlx|+pEG>p3!06O?F7VT+<7j^2(CRcIGK`waL#C99 zHY3+4^JE`7h-+d&LhrgW7#9ep43XA`J!nqet<&mr5l}<5?sqm5OV6>wktQgOia6FY z2dC^TmY`lUB5Vv#H&Ph=wI zzk`$GMf|Q|beoIP$&|-o_^>dAjih92^$vF@mM$+>V8v(;Iq%&@cRkx422}ZLsUVy! z63~$>?HZS(EAo5Fm)gU#YhY$}yKmNc=@Tio-SgkVrR$q}KtheM1@O$^1@ox{~BCacz;gneh?%L;L;31>*Kk z%}L9*Skf`wLD1m2azDGL+S>ZJc*olt!U(pfq76&X5D*`J$*$t6y{8A;3phOmk7GT- z9jeL|hct-jvuftQt4VCZb+$ZBRvLXQQSyuuy(Bq_BDc370`aLzzEl!eJ-#~lt2$y_ zyB$$+=X>J;IlxA3Ljn+Lvbx=7x4ox2G!s0@?hfP)S^YX8Zs<|VsMXC&1=lDw^h;o( zN9!83vQxUk+yaLuBE5u4Q)Nv{Swp9B)7t+H3fALWM09Rfpu#Oszy6jLNP8fix?HL8 zK3|sKJ8{Z3aCFu~zjv*gNM;sAdK@}kG+j#ZW>}gEUf875tt+&l+0v;EC-ae048-qa zkDoivUvhM{ptM)(!|`|y>yOykQ5!&d*ow#;rNspuZ?-@ai>J3sQ1lE)IS^-N5S#*M z-0I|INwCY=GG{?R%3>Vvj_vU~**KJD^iV-?0R8-D6T47FC=fz4K_tLb zySiVxWsP`mM>&139dvZUpHk-=B{q5x|M|j6{r={@rUTXnH8y#S)$$BCn3SP7@l1kV z?}#6=Dzs)%8p)b~)W-NFzS{q8(c4BnX8SA)7XcPK*kiRLu*w^Y;zq6;7lmOeae6S0 z5tFMB#VO~Fwd2&V2f~=)F^(^kR}c}Ul|jr1?KUUknwHYCv#J;%NMy~=3HRbv;Uw$- zPvCbZFDl1J??(Lw@F747vgJooNqzVyG(j_mb;5QR#&$|8%{uPZ(oSjT>I+yPnndx~ z4D{xIHTRCKW29j9Ay^0F(>c3 z)l8C!gNc{rDGieZiK20)6hUPkyD;&Gy=t&O96+yK;(~q`*(c;~|Hjawv(d*I(Pn$S`w6 zxHrid$2w7n9G{Sj;plU*ZqrG}#Ajn988>faLqEbf!#EkeI1UKUAd0WitBBI#=SMxY zbqhYe~=%-+Y5d$4KLCoe(URcqW`lIDr6xYw$GG1wA9 zb-5;qvVfIf?i`r2msrOfXa7*&IS?QTe<20?fs^Zs`8uM*)W#c~X~s#y*HpF#D&&w1IyemHIYl|MsB-usUdO0X%zysM|BQQb8bX^#JMh zfytEw6RL#ZBT(gCNI2wIulA8%=idCtFmg+tplUm0syw_OHqrY3M1o71OhZn%k^9J4 zvT3vV5{tDA+mzCYO{3P+4%XsgizbZ8lnpy?Cru;LSjHFJ}~H zpyn(#fAuF0R*#jh0<{Au#wjeDXyU}oFUg-+($Xlc$^OX<+QrnCh%&A)3KpOW~xjo+ZwE`E!g zlXBHSb~a=@;SXRMjCHmvt0k$Z$Ytjc_=L|7J)ufZT6$lg=;iX0^p=?^-rc4J#|v?I zB^E|Z4qRa%o=+ZKv3Bsg_}95e@&6i0N;~G3>PrJFZY@MJhj?@i#zQ_sR4f^fI$Z+D z#;hH626%4i0C9O`?N{mL6Y2|cFEYr^DIn-jm8SO&(kw}q&mX3s;^_rY>EU0bM3SA= z?1CC7j@%n#=elhQWyVr4=ni4S(gHIvOSfD4nFyB-Wf={lY)#T?#dVy0YPS<8*-v13ji%B#*)lSA4U!tC(VRa zM%qlbp`1c38kUq(!&Puvb6^?kgehfpeAfI}o5-w`gfw$P@93Y?13!u8rh4>y6I8fj zX>hE7dqS+wDH$QqlVS0FsxbdGn|t8kZ!OSNW(7q&Ypiw17(!~>m5oI%C?x4&=FqY6 z5n}ZC2QUWh(u@crLP@y!)e*E)A8H zA`3QF`dRCA0t-%sD9;MwsS>t)-iT>cj&Z!`F=3w0&1etT=y3&j3?VByMcHnXFP1C> zGe0%ow`CFFhOTsOB+EO!39RUvAt9!7Emu0vNRHo%t1!cvA}wq-Hlo^O#!gOwS6emY zTVXlk(-IRvTAG?LtsC_ziAofyK(9>t}L|wx@s+oe*D8 z7WWZi+x4mvKf?wDKTavZl1hjQN2Wo=`IHopQw!V5djpz^u*Z#|%9ZwbftkdQSe`@oY-krALVUzn z!_?3VuaGyNc+N<&S37OnCXKdgUo*I^AIiZWRAA9e;`AbFnNfGwkUGQ z2vf~}AapDLIl`+z+#I^gfWiWIYYtjF`AukYXo=6tL7YVwdwohCHGuPIiHii)Pj3XEk{Q6Ho!M5&6!bunAS1yJb>6pWuRehXh%MGw46>X_=hJy^wH z<(Ffl%_42Dlei%Sf8gcpljE~BkkEx2?vs_ee?}>TJ9y5uZbV$7E&{HI#CeHbU^X{m zVO(JSD&E_aJO%rclRF%1ihynH47DK-lt7AbYLgZFaE9NS>iDF$BesorVX>mo9rz$m zJExBxnLz0i`6v*W;&IVgqDUbyJ`T}eI(pNhDwu&z1qIaVIoDqYu#b(4pEoAd?K0Cj zy<=eWvMCxS+TVn({!X^0WTGC!byg%izH4H__uRE?6=%a2ws;&)HZNr!eZF}1BLq7uKH2)R3B+7qiLkY^W=0uLI%ISz}w-iyRjZTDvZOAkdQlJP(IF3>g zU@J%oXGiZ`>8WB43KTq~I%zEI>1(_UOr+&~&%B(ls#%PsA@3@jLh!31M75B6x$|H) zJdxf4ru=bU!WT^#D;p9P9H@AA59!8bjCD6Az!em-GyYyBmZE6e`3pd3oGdn2Kz z)4H;>gHFRo?@9bfcj;rcJp}d0fOgxK8%o8jJOvDcVs^?1*~mo*+t!`2cfk%_eo3dt zq(Y*!0t`g$RvE&88l^nbqVmxY#_hzkQ*tKEW{tqc>vN0C6fb8jD4L}e`2fyZHX|L3 zQhWT_m`#B9vRH)HNZ~lq<@M6^>g|!!glz)AoByGReem`09i>-Xz#`}WOG|XgDRGfQ zG_&oqn3`JR=2+UFWa0&fKc=CcQhy?PmqLM$0Y!qVV_$8E#(j9DIGeT zxr{mVFA{R`7{`e`6JkFc2`Gd~j)1kcV&Qxe>W zU`uf%%}Hx)O1l=N)rjig5hCaoOFA;v%HemNtPx^m)wP3@6cptwu#%Zs zVUoOyWkk}=aj2Bq-+(-~c!&17WcJ$Og2um?9R2}lyCA9%qs%b2p3a{htLRF|X}i?M zbJXTUGEc1f6GGP_S0qTK;eFd6it?pswHW9l#S__kXb~_qdiXs2)ebHXbm)!xsWCnD zX8?|^nANBVF2am3#H#g<%f@N!!hQMCs*RwS3n1Sl3bHG96^!t_1c1{bhhLuDBI3%qU~_ zA7O4}8YKmOhrp)DY%1;^5X*=Kl~vI5zgE)8J2UqlmykeLW=B}|(Tmpjjc&Ro@0;7V z8Mf`(d8sXz=W)6)AH{dGrlSSEmaQ#i7ODJmP_F@cWj)^9SiG-pcCpwo-LA;1t` zO@TMhyx)=GgC)L&RA$O42ShnJ@sV{riuJrv(=8o)d~v|nG%T1~ObY!`IRP*(&@NxW zc5>{l^nOo5fV?ofX4T{X-FbwK3>AbC!xYY502O9fNv3@YI>5?L!rK{PZonSk!!QutU(r;A9AKReg5R4jFN6@98-^wdeuu?J%k1?y ztDwLcRrT^iuJ5Hf8MrtLBl0B*CZW6ALL-Qyg2D!f-#d>Z2tgeEvieF1*nK%eO3`(y2RO~&3GE-GXYWIdCQ}8H!T?Q)ZuhunxOy^8HT)3V;=l*J| zmd^eYR%_Mhlq;#h+pW%SM_qc{#w~mpS8}C;FHdFe6To{vVFb+Co9_BTWCwvc^I|<+9c?( zt>FC%QRKoNa--}UkJkvn|6)~7tr<8=`_=-mU|ckPT0w!|{>v{8*yfVbW(Goqqg9XgO8 zRcjjImKoQ3qA*)ac+ZCdYCNc+a(szWd7K4nateVw>m_$?keuJ~`W1F2khNN8G}kc2aDgJMMhMLW|2|DQ1alIFOFe(=AfVX6 z-YZ>w#m@nY=DqavVJ3_^TvgZQ!p(>7Xh>$L#W@rfe5fu4L3p5fWfSZk(l`oHh7>F?gV((*1DA%BJw)zYHtMb-^cEP^5DeTU zqf9E~u`~-I!PVi$C4vT1sXS6rY4d?%4S0Jq`nnsdmXj*5(c&{VU!gWO3Ph9huv9~c zXuO3M&7fJ5OM_-q!BiU9F+W5!!TS~XfJiWDVkD@t?o}p?i$>Y_lWzE7q5vI!JCg4OAk4kg&a9(p~hpe|2D`pdb0QB<2>a z5-UCZVcB%w0EtlAtK}~OJav^24Li5JLjqzx(@LXb)6*~!kz}gf&p>FjoC726B38L~ z8#Gsqd>0kuoL8ydZvG`k0*QIq>fv{phyvdTv~^1-i|zgdw^Cr=GS=`!Bx>`tkWfE<#{X zepvdBoZyg_r&v8kr=jOBz|b`iM6Jh$DmzKfntS{ZV3T~0u$+jISNg>X2!Wws*uoLU zgoK_IE!6H?VyOQCVTEC|i(WT7Ql@_NUu?pWX`Z7e2j)ZHz)4~TY*|8UNl1|B8x1nRAq&s6b4=( zq#CXCsWSiWk&EMS@h4==?lwQZ8+hBHL3(&mhBLyUV`U8?mGrxYhlsfs1djgDyp5mp z3@Q4Pm9Z{xK#Q!8AVMX_4qq*g1@{>F@Yr-3>;K=l)r%0nWSM^C;uFrS)1$Z}H>(*g z^@@KKa6oa)Dao0jAvnEMLDJ$w?$gPj{PouiLWR;YWGkNGLov*A4x~=SPik64Zx2}SIVX)M=>1p!vCE+Yob2dB^D4N77nkv`OXgBhG;wH zQXWwF<{38Dz{BLlV^O5wyrW5fmlZpUdWQ%{xwr^WiV9Mw``28!q6QL!ozQGL;D4N2 zaz$+gnHwj~Wkh)YrL8)vrEc#v80p8OZXFf_mbfKux`9xcToHjaSPmX<(uY($uP&~| zvi}10>BE9C&4OtS3(>}qz-$QI6Klrf6-Q`CcNSSC;TWHGxF;(H$F+H*laV7I!(J}n``L_^+ z>DW^-m0;t1B28ozHX7d4;tVI&f+byva(#wVi{^^(FCYh#7f^cp#kP+B#sBJWLXNal5k0R69t zUg-_GK8+!^t_(0QJFMkQ4dJx&k~-kj{Za`k>s@c@yjVsH9@(=g-p70%YRZC!RC!sR zXZYTT$3uc0KF=k1tdgYkwBy;Q5HLKwa2REvq|brs@7(RK3%C0~XZcv*)5>A$7THn; zUYYy;S17q-BOGzb3Pgq9?HlElqcz@d)TP9;3(vjXJ*G3YVMYJY_pQwHe5N({{5^~e z11PfpBvuDqoF$fB$opKtt$@l${=Q+^{s=|+obigA9~wR`;q_Uq=qYp4OuwX4jIfny znQIO@4bgS$ede2;aP2gWqXk9X469kOZ4n!qF?+&IwUO(_ArGr&}hc}Qaj<&q2D`5(zOd%=C=Ja2g$lteXL z`L8>Qa$IC_W@r*$I4CR>-r{54lFVNSp)_tooB<8YYLr7To+(h{sUkw1`_Cwqv~%0T zUcwwDsN~LTJ~ZkX*(eC~lNHN-#nKb#uQm=D6K)_L$sd!=KTB0|tRc0c%Z8S`t$-20v-R*TFE)Z3d|7tL%4br4w#L#=d^oLq->^K-P0vLYth zZ48E8!(N6=EnR4r1>pG?@ZPE9n&8&e5)Pa^0M78`+WTBIn1>EAo0yrP{E7X(G)X7# zzY!6BbPWeQoP!;!LQYUUL>_oJlx@?PR1XJ=CtVZL4C5uY_}RA522(V!n6il%IIAe< zPF7-;YmuqJ&OUdJ9Xu=jtH6#XkXSF-)7N2R*!+H^rm$T}d0JZ)qPIl5XEdWaepLr; z#o5)s+%5Rq{QBsBVBjKf@<9QpY}EV%t^ET^2A<^DmL(`l=z~6ymsf)mrk3!rF%r7_ z^L(_I9zE!pfiUnCYYFq)mGE(tzM+}HHMj5H#PML?ZtU+8yc}4KM7|V`3pfi4)Sh5_ z%ADzj-}hhH7P1<8)nx;c7T6u*i-udN)83xRrHD{+-DH( zbP>JvxBc(xA8+kd&#=uP(IQV{z70ZvF6MmPy8P)rBA6t%M_1O$(TN{()nPMd%9!YO zPqN<(=bm82BJnqfK6_>Agf^fmX-rb@dLMC$@I@t*&DPNB*bVt89s8UWb^VuisxD0t zp4#wo5p&C<+OokNPy`rP^@~+IJETm1h(7`|MX54S|HGzjSq}zn-E4jy2=BvxyCoO0 zH8URl?bt?h=H{uvA|J&6R0%Pk;VkV&@A%TNL@vO;wdv@-Z2M*Yeg7@ffd^k&74xDW z<~?iQs!bu1UFJF00gMt3I|N%$K7&ieM`N1Y6Qx<~G&fe+0V}Yq5V=;V4VBPVg@;Ff zk)z)9y{OSdzc{>OkM|y*{!3KCxU}Bti(U6m1F9hREBZ}XOC)CD$4M3k0-zSSoO5>= zJBQ?JYtiXklN6&No-JlK~*yD%bF(_Ue7#D^~Uolgx@+!FNc9hM`qZDP+FlKK%Z(RQ4>+X zcvRci>4R8%3)s)DAa-;*?z2BK&+aLtrsv{3zfU-u<-ux=6Ci#1JgtP7&4JRbho3=A1N}oSL|rdLP;coBjEVf46%)02#}MRw7Jx(+pFM z7Hbma=_)-l=!$}N1j>X#x9|1UE0(k}MRQxeM=t~_UZY$s2}xa73`2d2&ebm>G4;MhWV1`O@e6U z*D)Dea+iH-qZ6BqOrwlSo#R;wo?Id*%LoWR(J#k1kkB)GI$YCo`T_*%o>z{@JVC3Y z6{K>LzzQ&#=y5u5xd`7~4tiB4*@Ukdg-N3cU`~Z;7zgO-BWkN(WBbafQ#!EuROwk3 z8zp>|zPLEY)~PciRpBc_UYMUG4||pE&^CpL2v>=7RXW|S!wxN1-?_SD!qJ)Wl{sIi zGxutaA^r!MF~*LrV1(Ls+ZPPqq68dxIdU#+3Y_ALX=+9}$4VoN`qcv7C}w5bwF%!(AlGZL4>;zJ z3;JJunkcP$K{mLHKmJ38&zFu*%j0{k2o3QO9PYT$WmP)^V%XvlVsdmmK-*k6|KDKe z{ST38G}5S9(H5iVWnmJZTwBiZDme5x~s3#>h`wMy803PCHi{gTQyRLJ0OiOPWn%l8e=g)kmTrE=&7wT&Q1IXLLg5>T;_p&N^zjAdF8DA6mQooYUEG1&>8 zCA$^c+ZL3?azVc?dTwhZX6Vu?aB7`06k;Z2uqM!jiHcb|E99?N{`Dy}@vIzw~ z5{rR+r>$E`TDEc1;b%7NjHG{7^eyo%pz*E)U(mKws8-n0puSPvs!5!}DRZ&j61^Xm z34*1gD`nccJV0uWtWQYluoZ~QPq5Y>g(L1s{C+YU8o(G#i>&?$-3HzU?gpZb*?qJU zEo753ri!{$&=QVL?>6`iJU&O0`qD3uS1Sl+=!p2q3-o8svSrZ~PClSmm;X`O6wESi z`+%a3Q7!|J*eI@U7UIEiVtcnE3^ao=mBfy>hr2<_6{m#l{9dJ!W`i`1+Hbf|A9V_A zzcO7P;qjlmFnv(1L@DuRl}#Xco#&7VsY?~Ju+PfUC4zspRk=k#c4Yn{0VvtmYZsh@ z=*Z-DMhrsmn276s*UBFd?7sz7BCkKE{_4@!ktSo0v#3P`h5)tcSdY^N2p}HKgegskQ}FYf5mIWNuEd*LS1U~f|f|sj66*pnPz%; zzO%{}sG)9}qWJJ}I7;X4eh;jAZGBJ9lb0F| zt%SfV1^Z_I6SZbWr9X4}*CV}uF&hs1_=2o|J5$zSMa)fis&|#1#{CN__G<0cT=;`c zTp(r@SiAd0Zql%JHf;P&_HGSX4FiF=8Jx@2l@kS0bzMa-{&Utsn{4QL#k3lQG}4!+ z5L&Zk-5PQ)maXlSO6ywh=e{>&nq>?cX;&In0M*i?y)dsDC#8+Kbu$MYQrjz^E!Y;u8AKd|Ck}Dx*pXc47v(8VAWh}gAK3^WGVDA7|cQ@=$P@gv#BZtpR` zhE;qWkod~;MugRo&Mz18jweu4(;#MXTH}kib5rWIThKLfuO8P= z@_P3*?IrMp-Rl@?HiB&(qbBNmZH>GdFJ+Gv-=!hwvM%V)8D|XKPH0 zPiids$N3I`&u)P;y)GlSjRCEEa`&S!p`D-Kw+aEZ!+lV|YKJIeLYb_i5ySpdI(VgY zwgT0f$DuQl`D-F4FB2$p##>Q0=vr|X@BPOC_h)K_dBt!2%rd?K%6LEwmV*uOSb=fO zq(?P+2LiccU*{f&zUtedQKR&y{=oHp)1WxRoG4-WZn@qRB}Z)ScQEhRbas>7Ul*fU zl%9J9xLxNo5jPp<-)k3!9I!VKBJmK-06d6W(|q{vmsUVHZQ%03(wLT(#}Tr+M2+|l z0ZIrxeWikU%nklKybj;L&|~L!iQA>l^s8ud*AiQ(C>l>pKF0>8$E8?4`slI&ozB%g zKcu^usJy(&=Cnaw@Qz{Ye$goYk+TkJ^8P}`xLRWfPI1m!6wL|xJBEKz24G@5f=7gH zM(|dZ5U2gJ8~NAn4neA>;s*g1{X-j>8^o!*R{MrGKFf`3EA;h;#YWo z>zi_EuS{_fp6JIYIblZ`xpj&a8LwN;16d4c*upT4bZFGNI>7wR#7EVH5&17IYZeC4 z-4(9CgcHUEBYE`8D074G6VN6Ia>j zy=9Rnlu8j+74yN>Ic8PYEjSp8ZqiKK$WslNCiA$jo+3Sc5u!}yuDagSBsTgso~rn7 zB{o!TYJ~5W)C#2(;>2MZ@`T%Utxx7-UUk!haSI!@14M8NEpQs6q7+zwoyMddA`+<0 zP*5DHp-7ey4ji#?oc-Ur$RmZh^G3zM5&ufj>@u~`){GzRIHpdpGR37C$^DnhU`v>+ z5j_Wd8Rdj$&dpgYB?VVXxn>zFp+=;V@47N$`*1MFf2=xxDpeACJ}8fw z2Qz3enUgYEQ_)Z8`7+O}20Nt%X~+&j@bU@JLZm%j^2FP+`Kuzv*XG&lgG(Z%_871}-1Pe$kNpTrWGo(%21~~zC?cc; zF^sim0vUyS(^EJH=t|{5bMUsGU>4u@`>3x!uac|f>&x+Es53AuDyY5w#NOlG96(U= zCZjaDD`pJRM-di^aUi-rOo5=~r3#xQ16HZwdEVWCFktmJ+ImlDdqxok(lwZ61VQLT z(6z~EU`z+VTsDJ{!)bAqm4}^3(kSN%C9H>M%8(2dJ94>}g zSlw{vblj2T@PGR|JS~g-n{p^Mc7D$?KndjP*0t%!nifOcbG1MY4&DnHU*DEJm&FF! zz{5oXAG!+V5Sfjru*kDAOV*}%3Tg|of(sYTX*qoYRPHVBp~`Tx#hm5FjxDhr|MCXw z?kF^{RZd9~jSWoAHSYD`Bu%fM=H*9(^^mf=?X4#w4oo6;Ft0qLzfe^~O^B-SY_<3+ z#kplEb$W$+zaYVJd(xa;mHeK?y12JF++DjN%?ko97O~^=Jzr(>1*h3Jb}Q+a9sg0k=@oj(Fp zf%*s_OeOfB9hY&XbAaHHluvA(3Y@C;G7*0;6`gl7q0JDBXXG@2XtUJ zb|-qO%wDS75JUIUca<(hYIlw3MWGNnMTG!ah!X1(M_xnOm73QN+JHy>LxO{Q_bDKdXVD9Po@mo~vZ{a{@%A1vt)3Pd%zDFy0CBi4{pNH(4Z>-K9Xw_ z__>{$_l_uiIXu~Kz11As{DSvdN_<unZT%4~kF`7QC>dm2k|9w+Fl%|CqI z%_iK9pM4;5bL)4?KP|sfdeiziElOtl;L>iod9L~??5JVMPxzU-up%|XZax{-j&^mg z6t?%SYUB)rQ0!PkY^AMSa?BjJgp1v-aM3Wg$0LD~Q!lc+(8}Jo^J6|NZ#Ljc{(tH4 z|6zGpICYNU-Bm54m%cMNO=L^-{w=?qazn23#M8{{&R1$ zk&w@!oGDIV?*O#A!*#3nz;7uqtxBvMEj$W*mLkWkTcPDYB8M7(s%{k;q_HV6{X%_$ zEKv1}4>|wr1Gy#%{LtIJsguSrh1)~`A*wF68O`t2>z1>1<*xl-s8u^ zgmlhB|1Yx6`Ma(!?DkP(qp@wPIdK{{wr$&JY}>Z&G-+%#_KD3C=jQv%z3+JM{tNcl zW34sznsYv%iLrU}{lV8)m=n9(M?kbGiTB5YQliq$sj0nm@c-PWhJ_HE>#&*tyT;HU zmGmej)&vS|3BHz9+vF~QA;BD;!NE9@f{y;HEogjP{`y7R)VoGK{%)+jS_wP>az;85 zhd`+4GLZBJg5=@!E9?=IQj82}Dz!M$omen|$=VB+Trw5KwrL_s{?9ODV8NKo?Po@>(pq8b4ZKr^ z@c7PVv*@j^Z%|?_u1C8_M!qkl#B}%~HNT&fTggaf)L!kJ+CO4zd7TMiV*QB*yqN3- z*jWj!O}i7pxn97a_9)tAyst)#zD)S~t%?W*>oK|UM(AH_hH;l~>uXjkPX+pki(jV0 z{C`}8T7j>de6b;{ZH2Xgu+wi#4K_VgZHx+JSG5S=t1a0KkJwoL{Fi)wkUeTvfvCQ6 z?(<<(mHx=A#Go2W{j0cBvwMQMMG|lyr22CeW=b|z26srFXji}>@(aw9^p>&{9D};A zYHyvU;(fsPMH*$w(d@Wl`Cr&U#Xu}dk$(ZV`Nm)ElFQRIUr>v1vC&hIV%C)OT-Z2( zsKv-MMAJEHib?Q7m=bqBx1qVWa|%0I)zrIPRE^el@*AXWXc+mih<18&Z*w*JbrOzUsgs8ZcGmjJ%5GrGSYJv& zzLVF(a|)&`0)1uxJi#eh6axJM*R$Vs?ZNRe*CQV#%7ciS_&8ujlV2xHvF==_ZcU|L zTIgzdtK;yloEp}rN-d>%QQkXi-nAfIS{l~?e~Ox$kMTJk4@~3WuOFPx@Um7so&%E3 z0Kz;yQm3q_ zr%3ClPUaN0y~i?sEXxy!;+8}=nwh+nxKUvI{Y`|DA6PT&_kPoms}bAs{Qa?;HpbGV z#!l^>58Q(wKRjXm5=76wx5GOUT_M>FBefdXT42K+7910Ma*mat-BKvx6uh}aS~IS> zc^yznd%suNnlXPr zJnpQ;w+ec4+lK!FxB-yt?Ml&oJU{9qw>+FIFK!vwb8{kW{JJgIFDCs}reE(ah^t>A z0MuDeYdh>!HzFkq_-A-(evPAc|FP|zma)rI^|Llu3E)@ITZeMI-7%-4MXRXs-+mW; z+wzekEjxAeNPm^+J-)}~l+*E0vy?M6s={_rd$AzWR_x2?|0czbK z$M<5!8LPEX$*g`sAh_tRSd*j?VG;pfaB7DKq%kgNTZ!@@ zK5)U$N9oF#pp=H*OvF(WO?GwY!5Zo@wtVCH>YyFGBzArVpQZI^}PfK0gGd zGJQ^hzt+|1(R2Qw_|})w^#imt63u@Grq~QUo&;Y#9z*1aNAD;fN5&%I3tmV6^$5+3q>q+Hb-8 zs)Fg)7GCS+Lq9m{>whqOPF$hiTc3ewY>idH*P+uY?)1eih!t|9URt-wD+Y%!euZUC z13=%TN6Kh%=aTv!;*@d}utX%sbtCms{7rjiHFN|6=uCS|bPea%wLaCCaWQ@w4zcML z00LjeUAAbHd#1+lQcrfg?MbtirFONs=k77ejt9(h*&Z2?qP`}>{Fd=Q_eM-#fuFt# z^T&p!^i8EKz0ZZcxDm+YUnRHs z$zxc`Fg>=#TH{67lu#*0I;QE-3f4HJy6V5}+rNf(kuzU2je;47uvhtl$Q+wO-@jNd zhQq7zABXT{q^+(Y<*24qo@`{Yn8hCfJ$=fjGKZ5m>Ou%B5{6P>aU_ai`ya5 zoY`xuB6mL>%aER@@Y6cV*jXH#L6auB%r-POJ&2sl>d^xPI#Xy1#tagnT=2gHbh+Rg z<*3dXBDjYN6Ow? z%am+8i<+_Iiu%hTgxge4`z50V!#^DOv|dPA&m=(tQV~wQPRVF1zKt7FdJjQa) zSGF5qn0$pr8EqBy6a8M1OY(xWN1fvNiSb!yWDYr`c2WX==}1aop>V&xMtPqorG9n& z?=evvyz#4#oo6MC1`oYCRP>B+h>fD8y*?HW_wt2fiCgXK;k<{IEkljXmH2kuiwzM1 zi=Ry1=3A7^v-6TeJA(*pIlR@9ON&{Z<1+d=lOXQ~xKxcA%v3BPS{E^fjPcV;o6^yB zjEmi;I!#h~!QTy44Nbq20PQB@_PFeI?nLWlvqsmdB?j-h#vn-4qAFAwg~tsV)0wKq zaKAf6sc$|pTxIC`FnaD~h?u$BDakYcifW6XGr7j)r<_JqcJBO1o+Cz5+M2BT`;D)z z$XKv7a~kjNKO(OwtsGC*Iq--YMT=c8R-!Yy_xV zwi`l9iTH(&J`6ou&^4cdKO80meKcB0=fI34)IneKi2+wJa9W@c(0&35iUnM*7F`)+!nr2Zl8Z0+J`R0u=Ko^rNR-asFJ zAWFbm|otMS)XNF~FS*fb&%Z(y$ zxz<2{OIoPe3oXHr-rTrUX@wmf^!qzQ2b{;uBc$EjxB8O|+agJ3LUKL4KB`28=7b1} z|5SY~XTjcQWu!$XW1$mzIrg;LcO2@dh`ZfqPv&)&dfe{-TV|>q1Q1c=3W)s6kK6F( zmigFNCQjqH=PH>an?xc1cmhx0`I0`fBs0mj8JhIXH{J&?{tNB=h5daovpOiPi@wxgugb_@97KF3qDQug$ia1KUTkRm!WHKPx|w_w&r3V*5AmC)g{#2uY;}5)2e{q~ zfku#flJNpqNnB3JTXAAxF*4l@ck;wUWqM2;qu9PmG!+Hk&7w@ARrokq(b=*9!znaw z@ArI4$#@*FUT-_9FE?$PaUA^5%GF0mafqaj(Qiu&Q42}imyeMdMb?Z%^Q@A|u{)<< zB}}EA@ZAzw6H?O`_%yyx97Du@;zCSzPXSDUHi{sDrZaYUlc{aay*{Ki13JX2#!D63j}YlBcGXw@}~gwhfck z2?SHAedq`fY*zAn*DI7y`+d1rXfx8w!ic}{alf)`Z%+ypA8NU`tJSDSh9f!nv@r{C z9@X{C-<%S}CGfR3Yt_k23VC&8&JnVM?!ihr>@T;S;ylEric?~JY{YR5kJ!Gx zzS3gX0aL;#MbY>)Mj}Buzxje9FL58EnrhjmpQ&}ABYGyrY+-tjPqf)coYcpT8zp7C z#~vwfojZm0TO`UD;oKJfd>j9BuJIT-C05(Xq5M^_E7Kfd z&Q=zzLaD8~)8L-vaOZCibp1FB&3N^xjT(^BN*NxHuyLBss+1!Cb*S>$-^?-Ap(y9) zfa(m$?;}u%$|~m84r4Cx-wS(9-K!i84fMlo1VNU@I^1)ZYu~Oc-9)alM5%|6mk;?_elG` z|4cSyE@4_CO^($gyFkv2i?DI}Aj-c^Y1_2+jLYEk3x8YvjQeaUJYf)JoOZSc`u=on z7=N4TB;%-;#**^Iy<1D|wv?GD*%si=g_ZKU^G@=okrxpZLL%^Q_jZi0zn>8ecYlA0 z&cJ1_*#{wvnt^EhO;_4ojh{4Op-N}%yOWhEr2xA~hs|%JB=e`7I)+OvN)ZpaV~;SH z$3(0vMz8(T^+8!SjI2@@B`Kck=&C?9>0lqG1AYJUcj#xP4v|W#h*^c#UqDy4*@y`g zMMpo=b$0DCG!*Olur>eturH`w1d@W)r{Z`_g$DjJ?x3}*qh2wYtzDXvWTNWwj6BWq zCMC|pT#tR&+rN3jn_~@;tiRy_{E8zI+vL2;{txzv14Pt(jXQ*>dZeT+fnwz#cLBNQ z3FTzq(L^qh;z-I8K7t_rDuW5q%TO!~I=+z6BLg_7H>VI`>t@$|{BfhNN(1Xe)N095 zpbn94NH21`(NO|oid09r*l7^ zW6X74Nlaabdgy5IUpX2gE_wm7GFdZKKS-Rv<2!{&5T8CZ?aQYRmz3VhfkU76DE$fj zxfj?9jBw{XZ@c_3!wMAm-QBlU>R5;NE485D$};3MJabm@Jr5nhndPAsC=bz~!~tm7 zeVu&W4srFZvqM^3RlP2je-K}FYXU%Lx__xthJPPf@l1UVA$~r2YLlU{)t+qC6drZC z;oJoO9`>;17=i$`2)(o?%9S89=oKuZldk-P$iR4uMgh>aGCp1+H$JIEv-5=aa~`qt zj_VO%Tk>|Nj(KSPu$SDSY$0OmW%(r^D0D#Qv67qc5xp5l1o3};jZl+ zf8KL(d+wlb8;D>%7^|GyqJVmWP<--i$=RFS=gm?Yeeq5FEpr|!?5wZ< z>GB5`?l{!d84l1fC$HqzyZ1S0)8!0*z-8x#X6WcPIWmQR1l?nI7q$drg1Bh+{;oSg z86hqzplom~ZdN>rI>xa_T*JoJDTd#ETP$Z?^nT^jJAiGjs%pyR-McM*%CaPBTT$=H zXy2MxX&4{)GU)GprF7xFR*#m6Gp1nTb-=qKjZS*zZ;sQ4y$kaH1bN=pKN?xP6f@bw z$Qu#tZQ%~*e8eKnuX+u>1K}PbaJH@7m3&^h%ez;z_3ZG~%(yNo)B2L)F0dN<0f$p!l<|7v5rGoxJ#;~Eb z%s+m0l8)#@6*WVJ)_Ei7I}1TW8lJbK9kDs$Hal!j?`A#mhhbri%tXEh@g}+N0&aQ- z`)>?v@^1tFK)v>Zo!-TVzPjDC;bDqu0ob8Me3t(*(A}owaHX^O==)%ls{x6BaIuEMD0B2gpjCW7yADeMClWQP?n& z4wz`H9i zcCdTO5%4GlJPU@s{LWA)*b1DAs{pi5G7|>0IEBf3Ax31OeCh0< zkY-sYwJ*IT4797&P5gRKMg#^;gG-on$MP|@SLn#hM?k~V?c@(Y7-rl+n8v7=_Y9-E z=1;~}Vv$rB8pMfq)Et)X{r>%9hvv`4S>hS>;df?*f8YF&`@Q$&;9f4hmjo2;oxeaK z(=qRqzsJAd&K(spgBK|yUom!-Y&X3u{?=p7P5em<+NuGrw#B?w#{f5ejQYCsu}k#) zHyyc$iBm`b()Xw<3GkbSAM}@ltm{dcIpQ6<@Xok7CSU$X+VTBGr~UV|(w4ar-X|2QM9Oi;m_@;>3!QrdzgV~J<-|I#TNWk*_lG-& z6gjoBY6a1Z<27QM`FII}NI3SQmnST{)skB)?=i9wIIM%8-zOE#p^sioIR_72&*()S>~l~2LeF;D^HY*%_;(E#t>I( zw8kzUj`r7>TfUJT+8=KxXp}#}2f_254?bydag*fKT5+`!gCqFU!CktCd_qf^&W|Ey zMNCDA9#e|^!{(hYSADu9^~B@+NzB&bcyrKzm+!a`u^t)=&X~ zztJcP3}&7pnfrI1@vS^3=ZhGPIn3qL(95{mSuS`MA`vYuE|HA2+^|*GANSpCmOsE_9|5tt+g!A z+b-OcxhwK>_WW8#0csf1d7Uw#A96MBVtqH^#V7kmG`<=`JYXjEz?-Fkq{v$RH%DHS zN$tb)pX0g&W%m)zY}a5bwJwjJLe|&I%SNP~n|_W#k|mHWPjJD1!@4WP#>(^2SxUQc zOu0$;;{PXQlwpi4i3G)E5KglHSX&w=N~bn(Y9v8aGEe(?OIPRoZPC)ib+zK?!FzP> zps0TfJOhXHnh2FQUu^|`6=D_el|31-t%?*YkAWFI`;@G)&_t7mM6ul1zt8iRpC1ie z`%dh$?lHi#D7Tx%70+M(!YNS}mvslg@=Q#NGpFIs%Fm#dSKh0>eM{->z8^OOow--s zx6rk;RW*)Dcorto&;E6f3;&C9)q;8LTc6o|_4%&2Dwms8$nO7u>e=Qp2D-I7C6nk% zOosFj(4qI`m1dV$mCT8`n)CleGI{tx`^l*-vS`+nCw72->Y^B-EB+$fatydihdp-BtG`M zLvPs^%*G5h*0A36kSFTBpd4%}T2n~ok%@9DjiFmaL~g}7g3HDdVX(@L;jBW(*{Zo_ zgGXz7HMtN7V#Mk~5l}~1VNf0pjjGR>o*w~pV0?e1GY2*h8?$iGUPo)BPaL!hvOX?B zX1r}z7JCMIFYzS&aFeV3He(C{Ah%zCMU5KQah5$=e< zJ;|H9+Q|hw=ho%pRytgSrOXwSJ|Q>9W@mYm`g<6Tn|s}r!htf6NAgjeIWk!`PZr~7 zr)-LHf-!EiSC`Bck*cRE`t&}YM1n)Od?-Awmzf+Ph*Rll=j;RHW0N4cIv{L#4&P)o zN$>ypmCnRL35Z4MR@?(SanE=Gs#B;&85u-B=OQJSfGt^6D94<(0DGYwuK4(lJG&s0 zqp{jICl{VPi!QDoqPle>sC!vO$DYQM&eaA^4a7$+OSpaa+}7Ss!C!z`ed0G&B!RSH zeKX|Il19?3LeG71bQ9!inJ0LsW>Fm=Hw} zOFUE2$5^zoH2wS5d~$O)=H49Q(i9j*AAoDiN!e5^ia@rA;Or3+yy9X@w92?%hqd!v;E;7 zc3(98Z6CBGq3dQ4eIH3azGw+T2s8xuOfLZEwb3(7rYpGZDcN_3oghN3h#`sxw+Y(m z?l(n&Pf~#cc0FyKZbS`S0r9eu*w`aj{j5aU*4bB^%|btj z@43LAOYTWbGhMAIU8sSUQ@%?IqJz#fkj z{<3L1&;2^bTkdXdCeb4($i6Mg0d;-(#6{oj&v|_H`zNhrTJt@zESY36;20L4K}NVm zIbr*+DSg5|ob!({U!N6mLSJNfw>25e<9c0TuRo_Lu`3#hWJMbiERZgo=_Ep(JXGZz zztQJ@D=^9`FlxA?lZGGeJ|L~QEbw*2Xw$G2sMS)>@PA>|xhOGitYAS`rLlj`eo}(D zvF&qu#4Mnb>c7TkyML6_O*~3lsF5x9-3cTN>j*rjZ(xT<=V692tn{cZPQShJi^uuu z-*P(C@PpnN{G;G9dG4BlC+{Ba`Q{TjiMjqK))QX8zv8@5DX;*P@vqE)D0Ev&`X2;WroNw6e&s-M#8StzercW-s9Db6-Jf{toH#^0^E)Ms2wrr78<=7Mo5xh+BF1zcn zpGpr;5q*a5VS=QKAj5CoeL&YuclK8om6Nvb z>6 zzc&`EOG?ripm+we9>4DTF5Xnyx?B!Uzbk3_ zrCN5_zX5qDrycYKvq_o>$f6W;5ZjFO)tRWe2|($4x8{WhD|aUp=dSU)2o#1U>L z3N+|!`_ZoDxZIFS^8BaL%;fW%Wf@mrCRo&B5*oiwPoG~2#3a@#&H7WUM?Ot1=)k`)gD1mFNUjLCsm;&m(~!38(0q z?<(nI!Mp6%&+|t8Lq+LtyCzf-3h`4OaZ)cnp8QkOk&0_zO z9$U9+w_rwo+{Bx)frwIs875tC|9|CM&f?HA!>04l&jxWtqNVHyc>~-~!KC08leA(I zD}|%LEk!z{)1Xz}^3|X@xgCG>1CdZ1*3{YThWswTYlELX$pw3k0jSfEhJEE<-G8Pw zFi3Zv4N{T6a>vx#qJptKE4 z;{vvclXP+^oLcIXA>Kf#MK-`!t3#cAn{ENQ(43wbopNQ97t?wh%F;NSi4m|B$K_$2 z$m{NOeshKQunkK3J!#pihC4P+*#}ASb5C<@AH5QO|Ghv9VmueWCQ1y3aJw#6DeS(K z2u_ZZ!;7%oOxOv6lTnh|Ks_-hHj1Js{?L8`>I59t^V#I*s~+?%Af_8%otse=JUXCu3{z<-#6+0o@;)m zCaH!$V(uw=y`*`XW_r}+f0$yDJt&>Fl+!z^TnU>~W`1usm&Dh-hI#kSOb4$TkpjK6_;q`X+u(DcX(Jz_iL(Zf zdvv!EUU(pdxmA%&uh^rQq;h)if?X}yL-<;tw5?$q!$~23Lh3^AQkdyaY0U$kttKSr0sK^gl&cbgQ78ZKYylVN`v@Z7nHd3<-vnj_-Gl(*bMF7}5s zPcm*i@QzTA@7Wu$*wm$Ku_L}wm*iD{4z-*@^T}Xqc2)VOX_m5l~!RKfFBo!+s z+*U)Gx7`#1VITG9NnHaJKmXkbX4DhWHp|(qPboucB;AlL+A2-tSEZvIif#Aq1?Kst9JAluXL9eWv@E%ebv75V z?Pae_Yt8o(y^>A2&(C!}(9US>%QSI?c8#5@%%+#18AZ(Dg{Oy^5sTm78zUA+Swtv^ z!1{hOjMU=R{BjkmDmUvk678(YbGxN~c2 zPKjQ%Y1RBQ_RrhyC|j~>BPyC>F`jvzemkwS-}b)bFT3Ig`CMJO_x6q|zOcU{*Rx&bdXmpCjbPb*Dt5aULoqY(3X2)ZBr$_e_T)1S?HkYKsqa6=2 zc?%DN+uq`7G9klQK{AE#N>~160`(JZb_x*}34Lvb@~Cx27Ek9Rzsk-0PytU!_(Yvb z@h<;D%rw3`oEabad-f%gO!IDXmKP|e_4%?qxskekjJ+NhlAx@Hj$?%kp9E;XguVJ- zYrc2S&$dh24p4A$Ta>F!@#9Zrd_JhiAqK*I(*L}jR%?0}oI3U#hig-77D0|06e?PlyIr1weYp=AE*Qz#~D5+sLI*}lr zXl%UQMshfsb)lKEAg?Jy6BnZ(F-qh*yTijoY$DV^IDiJG-8k=yg8dTp_X6kfLw#r-QROSc8sgbZ?IG{kR!gi<@Gq= zgrDCe*6)=wr*9vOwQ^OXBp7mu7c2rSNs0ONc19myLlsswnIihw#t1d?ZQZ|Ss(6j= z*DndBvk&(^`#}ZfJ{=!Hb@kEp-0ihOC}g$}igt33MeeLzF8Q~QnaT6}>M=RlEDP1Y zVo7c~PmlioX-riI?FL*PaG=xeT-K&lP2E~(d&vfc6YI$tn&5;EBfVCFPH9>41T}ia?a#J|$r#xkkP@%u$3K zJx3_%ySYu$nUS8ZHd_UWzu)!#y|X+n_#SlE!Zer2g(Cr+J#ykXBntV6<3)=5S&zU) z#lQ+6sCj1Douto;Pl);I)WT0zqopyTZR^$BcDt_L=J0K~o zSDWn|t7)3Avg(wVtX`R{zu?HQh~h43-g{Ucj3in1)d+{<;vWUBkS6hZ>LCr}Cz+rb zr)M`}IP5QfQiZ&s)SgGYOUS;2U&NEns(+0nvYK(H?8aU(54f- z#SeuFR4CjCt}6V2Y8H!N|B4MsgaaVt#+&Su(kGf1et%lkOWIhd4017l_%zBwr^LcE zkke#|pMV(vE19dG_8-VzHf1d@lk7zqGV$i@(EE#uNCVxilF>V->iN^cUUuEv!;w>b z4L@$$$bboJSSP+XlOA1H^a!@H+)g0`SY4sab5=EthQZBwJDht3{> zj_Z8df8Tq7bh!u~W#<*m^>SZBXl(l$aHY9}B-m`{wix$hDqTZWqXvM}FX}{>7>#=% zkSc-1_U}|TafW@B_dR#N@z-txw&?nL=3@59{_MW_lHT($?cMFNTt<@Mxk}d9WVtXh z=KR*%2P!InI;_`K{Nj4wd=iaR(2|rbhR=FY1|p)LM14-BWA&xVR@#q=@LbjV0KaTW zPSAC(`_Uy0ov@qTu1^F}7e|%v?w->5p6_T<%wJ1#@s`g1T=gx7?Z;tD4v1%<*uilG#?}wHxMce#kp=q#AFk=J%@P_ilxzT9ENP}Wn z&q*;6X5hFSp3B#=5hrzHe;f771Zd%jNsdV*z`SUI-xi(PILYA@!#K|Z#BcB@8DWtz zMaD!ub@&BF1C9BQKv}uYufIUnGg-Tr%9Yb27;*wj~K>+(eEs4XI}jcHNW!^ zSJ+Ss9!eW~2Pv#-np+@UBP4xUC4#H>vJjJH;5-qsss=}vUc_fASGQb0Kx5MyWGYO+ zZ%K9re+0u7D8%^IdEK}Rfh>Jlolp#GbyL~HXQz-y+SQU(#Z^7P8lXn|mBb`q4#~OI z#0g0}0^F_Mbn3jy<7ezX9{83rO%(J$rwOSVTINWwYOEAP(xR&Cpls>AE%L-1^#HPdmle|iW88PPe&B&iBA`#Za+~j zeDL+X{V?`*vaeDdN!rGp!466r0Jjk+$8Y_oK&GNqQ!rEo4#BBNDpHWBodwH_h696b z=iM_J(oho&+W}%-D}(mugF1pufK^wo-2J0DJ_=qe0;`p)BM6R0mP}8qRhu3Z+FEGR zmMV74$hVjbJEz}e?8KD4%@}Z41IO9TrO{ZRV{1fV{n9mWb|m^>HbWj$`e~$9RT8|rV8`{M#ZNW0IDWoA8M=k<5B^59|5HT=$cX3l2b|U3xTE`^vX0>VW)2v~@ zQwst!WcEEM=DkkAI9y3OTFNHNA{X(ZolUd&y@dX?aY7^wL@Y)blZhtN4>wz+J&cwP ztg^W$__j9fDz)}_l^l2K1m8GwpuU+Apc5iv+F~gc*HR?~^L(DPdB%SRStP zaxO-ZEfPO&GO^24nes@OvRcCU%WJIG>JN*Jw@k~(ZmakA`&hboR^EC@4b0T2@C=n?tDt81qh5 z3Z0_U;46^64)rXp#*x+em#Rj>wR;Lt4s&LIK$|bIXud18g<~Z?80idA>eNQ5`(b`p zGeW7j5>C`u*CeO;?nS)iEx0gPKiPggf7hPjTgwy}7|xxPn6Szu%I{;78C@k+u5s~y zsuA;@&`Is~c@cKo<<6e*!)CE43idKH-I0pLEn~sJr}9C{%TLNzUQ!^xs-<*Dd&Q|X zQYr@)=vcvjhR#Vdy%pEmEP z(7s~=3YBXUzO+GDRFQF+Z?Nv&m8-qyJ2|bWj^<-8zr5rF>hEbi^iH(fFO3@o`vgH= z9q1ss4!bOuSM&~X%^3lLqDnVWCU6rTf+f_2g8^aT32)FzKwdXdkp%NeH>#v0Yw=_2 zSUD zRx=YYZu7CA^X-@2wiD25K{Kw;E+?;@xHFnA0_-50AWD!)oCz0chJ9x0wN_q|8vQ_DE*HexF)5j>)tqtnC0k|~!? zC!YB>a4GpjIKmffF1ZFwBN+ns&Mza_ zoFJ5gAlzS1KWw!gm#jUjqJE_r{9Zm*jT+zLs^8&p(g+TMH3l=W~GA=*A_p%-o@rleOCrpdEeIs^)466ovh zE%{-BO>_~y4&Kh|ltD(YZUEvqE3jw@udSSy6%6`-E5quhjT;pgz_3yOBfVn1{2)Fb zQ(nBfl6A%DVy9c}CONxEBPAYzVVr=3ls=qHi0(UN#GTWF-DFi8;xU~mi#RT2N&5tL z?@bU6MT#0Hm?W{t!Y5y* zwI!D;T>0ZFC@h?<**8dds5zb>JO&{~W;k;x+XG^IYH}LG?h89;S6WKFsDyR4l-Tf{ zR3b)8T!7WkUwb$?k!Ilf{P9<*bnD1D$c@-r>!0>dPn_p~Q54`ULB|8#h5`_YUP_Zx zE!IXjxPB;q3s)BU4v&bMA~g&}tR$LY)97!uHii<32@S*FZ;3k$#4ZfZ-Ef%uq~8jh z3~`v=qMb+>6I?}#gB2LkBnq1wMPP7}(a_M9zIxWjP!hL0i5dqwB7_rp{QL1$PE+Jn z;8{ht5Ix&FofXh2OM#Kb3S1+wPTy0;c&p&x_IN@_gQb-q3I|%E+ETEQp2d6Vr{bB{ zg#+~Z1k8SoD~MpqR9M1RRWJMg-FV!{^BHk%`Zo(R^wzW&3K=){^en*c^OT5-J|D14 zDGUpCC2pkL6Y4`+G$lmjTu&6YGf3A@;ruJKO~bdN)!qT8cuP+P%o%HZXB{c)9D#%pg=-GLQ`oeJHCL4)q6dqGSD>b}SU8%OU&HgKVpFT3yUT6uQX(ImvexC zL|9W(kT-emJ%`$|X6JOAfa>%Is}M!lBIgy`bX{GpH%IZ&lp1=SWl1t_iR{f%vUt?oTvK+xO2LmNA$lk zKMj!+DH~XmOxi+k4&enYAbLH{yAv|TtleMsi?KX;=}QOq|7O5 z))7&K5hDYubj~-*LrU9A|4OuFNKX6&m^Wfq->N-2#Yy#fNv9EvkS1uA$%yP+_m+;l zaF{ezwhoHzbFG>#(^fe^UZ10GLBj??PO?HtcJFHv5x3g;cjtBwSZzO)`{m*^SXmJu z)LC8)1%*bxO@Ny=O1f^K_Qml8Ru20vhx&iti=F9H?_ml~*pysqKj_E*F|Z>^w)WWI zjZL9)@uHnInQ4u))XwKr>8J*GHL#q!Y)Y!-|4PKaJuG3PF%U)k+{i@e(i=~(Trr%m z@^Frwr#94@68ge>{<=5fA|*+nS2vWrYGIFR-v-htu-fQl_SH!XJ|wN2eW-z4F0MQh z>M7Unwc6PA2Qlg*6Ef7+w=20MYf>&bYgTEf;ahR3mT*GVjmfJR4IkS0KH-B0PwK1{ zABq3^DU}l15O|p(*D-&@fytzn$bwX$#?G;B9co@fr{cUqr;0RlKd#o)*w=H(opzC; zB;!0nlVtomaJ974=Ays|PD)+1I(tW$i74IYPA`80r1EsS& zqn{j7M`P7#$M5d~yY@~An9oS(+ipEQF_1Wj&BgnV)5r(6{A%xEgy2&K)2ke_uYWa1 z!RZ!?JuBbVf+6zfrY8G_DS1q~esXzSMw_oy+VnTCON202Ujf)I?`7g`+tT4O=Uvmu zeNA}cmx(YAn+s}xqY;v*iecYdiK{s`+;aj>`Og#K9Ca~^4WoX1%(|!yK7Z$?)z<{Z z&P)I8*LJ4m0qb0pJcWnUZ`_mu8XtNsvT=AySf?eVTEvOW{iWn$VT4bkXWc$+gm0E`z>o6~~bF8v4#2iLu zZ((enS700Kojzn_q0M?@r-xuyrM9{e`*n`-#}k8{Zuy5>SI+t&S*}P&vwvXW!d+61 z15F2%NL8PO(uM9eN9W<7g3}fBqDl45-}Kd+MQ&v zUxo=dr{P$k-WgChrf|`9u%Ew>c%#Ky?QE5Yr^lzIM)0>8ygzY_P`R~G2M%@|k3diXD54xAomEyVq!XXYP zs5+Y`BR21*`W1btY5ML0`s(io?$qXqIKHgDt?)DktQAq|kw3Z>>n9&NuIXmheNHx0 zp?0m)rNyGY#U&Bwgy(EkdY-96iX3Kxfe*li7@EYB{1X*|1<%gSR@0_$?-Us=r@u?R zJ%j(gn6@t3I>oqZh$8eu&rH?+eq!VNdcCzK8r7#OJ%#p`|zmqgTH?|Qx?Ms&_E<+q$K`lC(!JchHwY11vyBZ`f@%B^9C&TZfShY;m0muXB?Y%4FYcCf=1 z0N5&*RMBn`2}io~I#xUxPKNrn%sl|}7Bhk=66^q* z6}|POBqAN>6j8bKzGZ)TUODP5lA$_h{+1Y{W@kT*bi9Ba;sPBZWlqJ$7C@&A9=O=c zweWO}%)h-Ws{7*m>3nCJbXET7_X=%4c(t-&XZ|tSn?cNP$xn?{!zm&{X0~6p#WaFY zkp|=P*%x9@%NVGEAE&TT)aBZ7vT&smHyhIDCnp$3&{ybd=I0$$Z3793a_ya?6FfX4FO%-mQYf^y>L_^R{2{5HdMFP2*F2weE;R{$Nm-l%7UV5!`zIloBs*^EO# zY?A%ICw@b%NuMQS=1$4&rfPrE7 zTU`7*`m#(#z^4akJWL?P`aRZb;OXY@=!L^ufSTD~lN!xOKWJoR{N*|gT>J_{^l|9B zdp-onmqld!U_Y&Ln5IBw)%%I>U`FGnjW;K#H?ZJ6jLq5HKp=2a;k{!3x22or_Hn;V zE7^GFOH!H0+=;+&$OfzfTzHx;BtB=GbC#8#uxKR(U4Jkj9{!+8 zCfAW=j}j&slS`M*y@h*(1D?*&7|)(-BZ}Q13Z#p;(P`g4{5;*+^sew~5Wo=mG~X`! z{w=*1$=h_eK;-aI`J(N^j~;}mj~EpOh=lrTH-7B#gKs)C-SRa_c+rK)SMv5Py{}Mk zLM^Yn1AsAFz&_15B=u%ZEqYs~x5Z7miT7y(ewZ_XTgPCeQBm;S?&+44#(K?s|B5f< zaSa5Bp|>mkd~^N9Nm5qMNQR%s793?WNvc2@z(Rzh*30rg@jaEI#$45|X>0%IuaC!H zM>d?iwaE$};u!7RDrPbm_kS2rxOiih){J{>)nElqX$2QBhvU8x%Xhl?<13;>BwO&R zu-MGt44pc@P{RnxEe>-g@cT|WT+rdX;hDYtH&xYEX3S}AD%7WrjeeJD-PDr&Ld{I4 zH5c|I=~B1pfLI?Q7Q2z**UXZh-;fW_)_#U>7-2g-@@C|jkD5`;TC3u=4qi4cIxImP z`Z=?R_nmYIStnneH@Ms-0TpIJx~n_5*-$;gd1+Akx^qdk(q~nf^#~=j%}%R(fp9W> zFl7B+ENsbqKsT-W1Ia3-HY+-xD@M(^?7=IYzfB4BHyCMoPHC+pTCGmawi-4dasmra zw(h7$?LsgCTL@SH4L;P^Hn7g7@|YLR@$Xua8ql;S=V1F4J2_ujV>8gx?5?1Y;X{)n z29+yvg1M;is{jYXqhI|z?dYe&HLvFa@-Zs|MupXpt2$Ip`q7RB3Be>|$lmt|MLLF) zy@j@fhC39{zhL`8~1v7AoQ@|qR7;B@O$lD?ylQU1S0*m z+#GT(uQb2zLBB28$^>p}craY^qU>`!R5HAEo$e*8qFY66oi%cQ9vK8wd;a|O;I$#_ z2f%6xBa*zYYkAGS*18Xb;mLdo;_B|bBJF(%Z?&{Jfw41-U9QO%n@yqHlxu2}z5N*V zyGz}iPYZxOtu;{AW_AD}=U~2JKJ01`l`uqXYxJ#xniZT`031zf37nyT43m}8sz^I& zRX)%Z&Xs41^kfrGF*6Rg{6|AmWRbiQ$+mNk#pdf@Q^g`vl+9nURWW;B(9`g1g3ul$ zOetyV`zYxdHCcN%2bK}1^}Na6@iz_V_U8-yT=kKQGyJG{kt`TnZGCRgRYzov1JC-$ ze9Z)yAp&ZP&=^WoRyLcq=<&~j5@H7!`#mxN9l*8dj=3N2-CH~s0|^1p`M9;`jZmHu zADhIxj1Syl#oH~j6BNL)d2ac5cZFr5#sMo6OhRfsZWF*9EV28s$#U9oBl@CUoBL)N zWY$w0{1qLssB-xQ-K@BK@0GjI#J8esEH6gU&?Au98x2ooJ5s8&0gc+nk4-lo3(zln zy;gp(5%d#p>_VD;>94MiLI;;Fq#BHHKq!grCCI*6Q<!6cfBsNH&}WlC7+0<~n@vbK)YDUW~83Q{1{vL=QG?eq=|(348(VdXR<4(nO`1`%HSfA%Ll-W)WtSFvfFS zt4fpf2+EbbZl9i0Qgc#># z#^0HM>46Y-Ace4ihG9v$#7dMVaz~t%JKm?(%qZysUlRPFS&)lWVgc*#pi4*K zOv|VUy&O}CpwOVG0iy0%yL7C5(YIb`FUp)TWLVgA&=Jm=wZ!4!52J`O1)YpZbgSRw zy^3L?=BhFVr;5YU<>|N?X)70La*I=5xqdm?D=$agcHZf9K7ssd`#Acs2&BIAMn){N zekS{#&Y@MY0IZ3rqI_>bsfz;D=j&fO|1Y{= bAnycl0qU_~@ojvL?icgRHYQC*?s5MEw`YIx literal 0 HcmV?d00001 From 9c97d2460624bdd88b7e346bfc91dd9499c8df50 Mon Sep 17 00:00:00 2001 From: Khushi Kalra Date: Fri, 12 Jul 2024 10:47:23 +0530 Subject: [PATCH 4/5] added models --- ...suicidal-content-detection-using-gru.ipynb | 3085 +++++++++++++++ ...icidal-ideation-detection-bilstm-rnn.ipynb | 3438 +++++++++++++++++ 2 files changed, 6523 insertions(+) create mode 100644 Sucide & Depression Detection/Models/suicidal-content-detection-using-gru.ipynb create mode 100644 Sucide & Depression Detection/Models/suicidal-ideation-detection-bilstm-rnn.ipynb diff --git a/Sucide & Depression Detection/Models/suicidal-content-detection-using-gru.ipynb b/Sucide & Depression Detection/Models/suicidal-content-detection-using-gru.ipynb new file mode 100644 index 000000000..031394e96 --- /dev/null +++ b/Sucide & Depression Detection/Models/suicidal-content-detection-using-gru.ipynb @@ -0,0 +1,3085 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 1, + "id": "adce5381", + "metadata": { + "execution": { + "iopub.execute_input": "2023-09-02T18:30:49.089557Z", + "iopub.status.busy": "2023-09-02T18:30:49.088706Z", + "iopub.status.idle": "2023-09-02T18:30:53.415286Z", + "shell.execute_reply": "2023-09-02T18:30:53.414196Z" + }, + "id": "oZ7fB0JmNdsK", + "outputId": "7c1aba3c-5af4-4edc-dfcd-eb8c52a63c65", + "papermill": { + "duration": 4.36379, + "end_time": "2023-09-02T18:30:53.418469", + "exception": false, + "start_time": "2023-09-02T18:30:49.054679", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Unnamed: 0textclass
02Ex Wife Threatening SuicideRecently I left my ...suicide
13Am I weird I don't get affected by compliments...non-suicide
24Finally 2020 is almost over... So I can never ...non-suicide
38i need helpjust help me im crying so hardsuicide
49Iโ€™m so lostHello, my name is Adam (16) and Iโ€™v...suicide
\n", + "
" + ], + "text/plain": [ + " Unnamed: 0 text class\n", + "0 2 Ex Wife Threatening SuicideRecently I left my ... suicide\n", + "1 3 Am I weird I don't get affected by compliments... non-suicide\n", + "2 4 Finally 2020 is almost over... So I can never ... non-suicide\n", + "3 8 i need helpjust help me im crying so hard suicide\n", + "4 9 Iโ€™m so lostHello, my name is Adam (16) and Iโ€™v... suicide" + ] + }, + "execution_count": 1, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "import pandas as pd\n", + "dataframe = pd.read_csv('/kaggle/input/suicide-watch/Suicide_Detection.csv')\n", + "dataframe.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "e1c3efc4", + "metadata": { + "execution": { + "iopub.execute_input": "2023-09-02T18:30:53.487833Z", + "iopub.status.busy": "2023-09-02T18:30:53.486964Z", + "iopub.status.idle": "2023-09-02T18:30:53.595319Z", + "shell.execute_reply": "2023-09-02T18:30:53.594257Z" + }, + "id": "RfWxfSZHBdzd", + "papermill": { + "duration": 0.145681, + "end_time": "2023-09-02T18:30:53.598330", + "exception": false, + "start_time": "2023-09-02T18:30:53.452649", + "status": "completed" + }, + "tags": [] + }, + "outputs": [], + "source": [ + "suicide = dataframe[dataframe['class']=='suicide']\n", + "non_suicide = dataframe[dataframe['class']== 'non-suicide']\n", + "suicide = suicide.head(50000)\n", + "non_suicide = non_suicide.head(50000)\n", + "dataframe = pd.concat([suicide,non_suicide])" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "eefc41b7", + "metadata": { + "execution": { + "iopub.execute_input": "2023-09-02T18:30:53.673397Z", + "iopub.status.busy": "2023-09-02T18:30:53.672993Z", + "iopub.status.idle": "2023-09-02T18:30:53.718792Z", + "shell.execute_reply": "2023-09-02T18:30:53.717019Z" + }, + "id": "N3JRK8bkOEw4", + "outputId": "b2558bbf-8b52-483a-e21a-760d4105be2c", + "papermill": { + "duration": 0.090634, + "end_time": "2023-09-02T18:30:53.722059", + "exception": false, + "start_time": "2023-09-02T18:30:53.631425", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Int64Index: 100000 entries, 0 to 99822\n", + "Data columns (total 3 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 Unnamed: 0 100000 non-null int64 \n", + " 1 text 100000 non-null object\n", + " 2 class 100000 non-null object\n", + "dtypes: int64(1), object(2)\n", + "memory usage: 3.1+ MB\n" + ] + } + ], + "source": [ + "dataframe.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "a4a932e2", + "metadata": { + "execution": { + "iopub.execute_input": "2023-09-02T18:30:53.792450Z", + "iopub.status.busy": "2023-09-02T18:30:53.791453Z", + "iopub.status.idle": "2023-09-02T18:30:53.827387Z", + "shell.execute_reply": "2023-09-02T18:30:53.826194Z" + }, + "id": "gRdpyxlKOOYs", + "outputId": "be88f206-29ba-4d01-b521-e4a3727ce3a4", + "papermill": { + "duration": 0.072544, + "end_time": "2023-09-02T18:30:53.829846", + "exception": false, + "start_time": "2023-09-02T18:30:53.757302", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "text/plain": [ + "Unnamed: 0 0\n", + "text 0\n", + "class 0\n", + "dtype: int64" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "dataframe.isnull().sum()" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "843cfd6f", + "metadata": { + "execution": { + "iopub.execute_input": "2023-09-02T18:30:54.014179Z", + "iopub.status.busy": "2023-09-02T18:30:54.013135Z", + "iopub.status.idle": "2023-09-02T18:31:35.081382Z", + "shell.execute_reply": "2023-09-02T18:31:35.079991Z" + }, + "id": "Ax20bikLPtLs", + "outputId": "69aca607-b333-4abd-ea6a-8eab7a6669cc", + "papermill": { + "duration": 41.106686, + "end_time": "2023-09-02T18:31:35.084457", + "exception": false, + "start_time": "2023-09-02T18:30:53.977771", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Collecting text_hammer\r\n", + " Downloading text_hammer-0.1.5-py3-none-any.whl (7.6 kB)\r\n", + "Requirement already satisfied: pandas in /opt/conda/lib/python3.7/site-packages (from text_hammer) (1.3.5)\r\n", + "Collecting beautifulsoup4==4.9.1\r\n", + " Downloading beautifulsoup4-4.9.1-py3-none-any.whl (115 kB)\r\n", + "\u001b[2K \u001b[90mโ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”\u001b[0m \u001b[32m115.1/115.1 kB\u001b[0m \u001b[31m4.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\r\n", + "\u001b[?25hRequirement already satisfied: TextBlob in /opt/conda/lib/python3.7/site-packages (from text_hammer) (0.17.1)\r\n", + "Requirement already satisfied: spacy in /opt/conda/lib/python3.7/site-packages (from text_hammer) (3.5.1)\r\n", + "Requirement already satisfied: numpy in /opt/conda/lib/python3.7/site-packages (from text_hammer) (1.21.6)\r\n", + "Requirement already satisfied: soupsieve>1.2 in /opt/conda/lib/python3.7/site-packages (from beautifulsoup4==4.9.1->text_hammer) (2.3.2.post1)\r\n", + "Requirement already satisfied: python-dateutil>=2.7.3 in /opt/conda/lib/python3.7/site-packages (from pandas->text_hammer) (2.8.2)\r\n", + "Requirement already satisfied: pytz>=2017.3 in /opt/conda/lib/python3.7/site-packages (from pandas->text_hammer) (2023.3)\r\n", + "Requirement already satisfied: thinc<8.2.0,>=8.1.8 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (8.1.9)\r\n", + "Requirement already satisfied: cymem<2.1.0,>=2.0.2 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (2.0.7)\r\n", + "Requirement already satisfied: srsly<3.0.0,>=2.4.3 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (2.4.6)\r\n", + "Requirement already satisfied: requests<3.0.0,>=2.13.0 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (2.28.2)\r\n", + "Requirement already satisfied: packaging>=20.0 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (23.0)\r\n", + "Requirement already satisfied: typing-extensions<4.5.0,>=3.7.4.1 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (4.4.0)\r\n", + "Requirement already satisfied: tqdm<5.0.0,>=4.38.0 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (4.64.1)\r\n", + "Requirement already satisfied: spacy-legacy<3.1.0,>=3.0.11 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (3.0.12)\r\n", + "Requirement already satisfied: setuptools in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (59.8.0)\r\n", + "Requirement already satisfied: jinja2 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (3.1.2)\r\n", + "Requirement already satisfied: smart-open<7.0.0,>=5.2.1 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (6.3.0)\r\n", + "Requirement already satisfied: spacy-loggers<2.0.0,>=1.0.0 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (1.0.4)\r\n", + "Requirement already satisfied: catalogue<2.1.0,>=2.0.6 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (2.0.8)\r\n", + "Requirement already satisfied: preshed<3.1.0,>=3.0.2 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (3.0.8)\r\n", + "Requirement already satisfied: pathy>=0.10.0 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (0.10.1)\r\n", + "Requirement already satisfied: typer<0.8.0,>=0.3.0 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (0.7.0)\r\n", + "Requirement already satisfied: langcodes<4.0.0,>=3.2.0 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (3.3.0)\r\n", + "Requirement already satisfied: murmurhash<1.1.0,>=0.28.0 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (1.0.9)\r\n", + "Requirement already satisfied: wasabi<1.2.0,>=0.9.1 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (1.1.1)\r\n", + "Requirement already satisfied: pydantic!=1.8,!=1.8.1,<1.11.0,>=1.7.4 in /opt/conda/lib/python3.7/site-packages (from spacy->text_hammer) (1.10.4)\r\n", + "Requirement already satisfied: nltk>=3.1 in /opt/conda/lib/python3.7/site-packages (from TextBlob->text_hammer) (3.2.4)\r\n", + "Requirement already satisfied: zipp>=0.5 in /opt/conda/lib/python3.7/site-packages (from catalogue<2.1.0,>=2.0.6->spacy->text_hammer) (3.11.0)\r\n", + "Requirement already satisfied: six in /opt/conda/lib/python3.7/site-packages (from nltk>=3.1->TextBlob->text_hammer) (1.16.0)\r\n", + "Requirement already satisfied: idna<4,>=2.5 in /opt/conda/lib/python3.7/site-packages (from requests<3.0.0,>=2.13.0->spacy->text_hammer) (3.4)\r\n", + "Requirement already satisfied: charset-normalizer<4,>=2 in /opt/conda/lib/python3.7/site-packages (from requests<3.0.0,>=2.13.0->spacy->text_hammer) (2.1.1)\r\n", + "Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.7/site-packages (from requests<3.0.0,>=2.13.0->spacy->text_hammer) (2022.12.7)\r\n", + "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /opt/conda/lib/python3.7/site-packages (from requests<3.0.0,>=2.13.0->spacy->text_hammer) (1.26.14)\r\n", + "Requirement already satisfied: confection<1.0.0,>=0.0.1 in /opt/conda/lib/python3.7/site-packages (from thinc<8.2.0,>=8.1.8->spacy->text_hammer) (0.0.4)\r\n", + "Requirement already satisfied: blis<0.8.0,>=0.7.8 in /opt/conda/lib/python3.7/site-packages (from thinc<8.2.0,>=8.1.8->spacy->text_hammer) (0.7.9)\r\n", + "Requirement already satisfied: click<9.0.0,>=7.1.1 in /opt/conda/lib/python3.7/site-packages (from typer<0.8.0,>=0.3.0->spacy->text_hammer) (8.1.3)\r\n", + "Requirement already satisfied: MarkupSafe>=2.0 in /opt/conda/lib/python3.7/site-packages (from jinja2->spacy->text_hammer) (2.1.1)\r\n", + "Requirement already satisfied: importlib-metadata in /opt/conda/lib/python3.7/site-packages (from click<9.0.0,>=7.1.1->typer<0.8.0,>=0.3.0->spacy->text_hammer) (4.11.4)\r\n", + "Installing collected packages: beautifulsoup4, text_hammer\r\n", + " Attempting uninstall: beautifulsoup4\r\n", + " Found existing installation: beautifulsoup4 4.11.1\r\n", + " Uninstalling beautifulsoup4-4.11.1:\r\n", + " Successfully uninstalled beautifulsoup4-4.11.1\r\n", + "Successfully installed beautifulsoup4-4.11.1 text_hammer-0.1.5\r\n", + "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\r\n", + "\u001b[0m" + ] + } + ], + "source": [ + "!pip install text_hammer \n", + "import text_hammer as th" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "c5892f74", + "metadata": { + "execution": { + "iopub.execute_input": "2023-09-02T18:31:35.156648Z", + "iopub.status.busy": "2023-09-02T18:31:35.155033Z", + "iopub.status.idle": "2023-09-02T18:31:35.308046Z", + "shell.execute_reply": "2023-09-02T18:31:35.306110Z" + }, + "id": "FRwfTks6BtzG", + "outputId": "0e20ecad-9997-4be4-b667-10226530069a", + "papermill": { + "duration": 0.191662, + "end_time": "2023-09-02T18:31:35.310683", + "exception": false, + "start_time": "2023-09-02T18:31:35.119021", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[nltk_data] Downloading package stopwords to /usr/share/nltk_data...\n", + "[nltk_data] Package stopwords is already up-to-date!\n", + "[nltk_data] Downloading package punkt to /usr/share/nltk_data...\n", + "[nltk_data] Package punkt is already up-to-date!\n" + ] + } + ], + "source": [ + "import nltk\n", + "from nltk.corpus import stopwords\n", + "from nltk.tokenize import word_tokenize \n", + "nltk.download('stopwords')\n", + "nltk.download('punkt')\n", + "lists = stopwords.words('english')" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "ef226456", + "metadata": { + "execution": { + "iopub.execute_input": "2023-09-02T18:31:35.381708Z", + "iopub.status.busy": "2023-09-02T18:31:35.381311Z", + "iopub.status.idle": "2023-09-02T18:31:35.404975Z", + "shell.execute_reply": "2023-09-02T18:31:35.403740Z" + }, + "id": "ZsoWucx6Btvv", + "papermill": { + "duration": 0.062851, + "end_time": "2023-09-02T18:31:35.408095", + "exception": false, + "start_time": "2023-09-02T18:31:35.345244", + "status": "completed" + }, + "tags": [] + }, + "outputs": [], + "source": [ + "stopwords_list = ['i','I', 'am', 'is','are','this','that','then','those','have','has','it','can','could','the','had','was','were','been','them','what',\n", + " 'yet', 'though', 'wherein', 'really', 'show', 'then', 'rather', 'same', 'con', 'call', 'is', 'twenty', 'side', 'go', 'namely',\n", + " 'every', 'top', 'for', \"'m\", 'with', 'either', 'because', 'two', 'well', 'when', 'those', 'under', 'thin', 'amount', 'upon', \n", + " 'kg', 'the', \"that'll\", 'ca', 'even', 'very', 'into', 'by', 'โ€˜m', 'keep', 'although', 'done', 'bottom', 'detail', 'whatever', \n", + " 'from', 'formerly', 'these', 'enough', 'twelve', 'of', 'too', 'already', 'move', 'something', 'few', 'nothing', 'they', 'does',\n", + " 'us', 'next', 'cry', 'but', 'eight', 'โ€˜s', 'however', 'four', \"'ll\", 'has', 'over', 'someone', 'afterwards', 'myself', 'have',\n", + " 'would', 'anywhere', 'hereupon', 'using', 'everything', 'his', 'put', 'many', 'more', 'within', 'which', 'describe', 'thereby',\n", + " 'nine', 'elsewhere', 'other', 'he', 'if', 'per', 'your', 's', 'be', 'than', 'now', 'sometime', 'herein', 'why', 'โ€˜d', 'whereas',\n", + " 'behind', 'couldnt', 'de', \"'re\", 'least', 'latter', 'whereafter', 'part', 'after', 'front', 'interest', \"you'd\", 'whose', 'fire',\n", + " 'sincere', 'down', 'to', 'whence', 'ours', \"'s\", 'all', 'noone', 'just', 'anyhow', 'ie', 'having', 'her', 'been', 'get', 'former', \n", + " 'throughout', 'above', 'โ€™d', 'cannot', 'almost', 'do', 'quite', 'seem', 'give', 'first', 'several', 'also', 'seems', 'became', 'โ€™m',\n", + " 'sixty', 'anyway', 'โ€˜re', 'on', \"you're\", 'onto', 'co', 'third', 'fifty', 'had', 'seemed', 'โ€™s', 'yourselves', 'until', 'an', 'mine',\n", + " 'across', 'sometimes', 'hereby', 'eleven', 'might', 'mostly', 'un', 'i', \"'d\", 'itself', 'most', 'beforehand', 'five', 'beyond',\n", + " 'herself', \"you've\", 'unless', 'regarding', 'there', 'doesn', 'while', 'whoever', \"should've\", 'whereupon', 'name', 'further', 'o', \n", + " 'nobody', 'whereby', 'others', 'between', 'thus', 'any', \"you'll\", 'found', 'amongst', 'hers', 'wherever', 'ltd', 'still', 'somehow',\n", + " 'often', 'km', 'becoming', 'six', 'can', \"'ve\", 'make', 'hence', 'around', 'both', 'our', 'along', 'latterly', 'please', 'via', 'whole',\n", + " 'system', 've', 'd', 'you', 'themselves', 'here', 'used', 'in', 'forty', 'each', 'find', 'during', 'ain', 'ten', 'him', 'nevertheless',\n", + " \"she's\", 'me', 'them', 'ma', 'meanwhile', 'and', 'y', 'did', 'fifteen', 'serious', 'โ€™re', \"it's\", 'โ€™ll', 'that', 'alone', 'together', \n", + " 'where', 'hereafter', 'once', 'himself', 'could', 'made', 'among', 'whom', 'cant', 'doing', 'again', 'therefore', 'beside', 'hundred',\n", + " 'three', 'etc', 'moreover', 'various', 'may', 'since', 'always', 'seeming', 'toward', 'are', 'll', 'everywhere', 'a', 'empty', 'yours',\n", + " 'theirs', 'back', 'else', 'own', 'as', 'somewhere', 'bill', 'โ€˜ll', 're', 'through', 'last', 'nowhere', 'what', 'take', 'computer',\n", + " 'yourself', 'eg', 'perhaps', 'thereupon', 'ourselves', 'against', 'inc', 'she', 'whether', 'their', 'm', 'such', 'thick', 'anything',\n", + " 'โ€™ve', 'neither', 'none', 'amoungst', 'one', 'were', 'thru', 'am', 'about', 'without', 'up', 'otherwise', 'except', 'who', 'due',\n", + " 'thereafter', 'ever', 'much', 'out', 'how', 'whenever', 'before', 'it', 'some', 'being', 'mill', 'or', 'its', 'will', 'become',\n", + " 'besides', 'say', 'therein', 'another', 'see', 'anyone', 'becomes', 'never', 'towards', 'everyone', 'below', 'full', 'my', 'only', \n", + " 'fill', 'we', 'should', 'thence', 'indeed', 'this', 'was', 'less', 'so', 't', 'at', 'must', 'whither', 'off','whose','whom','who','do','did','does'\n", + "]" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "202b807e", + "metadata": { + "execution": { + "iopub.execute_input": "2023-09-02T18:31:35.481391Z", + "iopub.status.busy": "2023-09-02T18:31:35.480316Z", + "iopub.status.idle": "2023-09-02T18:31:35.487789Z", + "shell.execute_reply": "2023-09-02T18:31:35.486694Z" + }, + "id": "9fSKPUSpBtrd", + "outputId": "523fcf5d-94b3-490f-f6ea-8eb1c55d7dc7", + "papermill": { + "duration": 0.047892, + "end_time": "2023-09-02T18:31:35.490471", + "exception": false, + "start_time": "2023-09-02T18:31:35.442579", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "CPU times: user 4 ยตs, sys: 1 ยตs, total: 5 ยตs\n", + "Wall time: 8.58 ยตs\n" + ] + } + ], + "source": [ + "%%time\n", + "def remove_stopwords(text):\n", + " token = word_tokenize(text)\n", + " token_without_stopwords = []\n", + " for words in token:\n", + " if words not in stopwords_list:\n", + " token_without_stopwords.append(words)\n", + " \n", + " text = \" \".join(token_without_stopwords)\n", + " return text" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "496e38f9", + "metadata": { + "execution": { + "iopub.execute_input": "2023-09-02T18:31:35.561717Z", + "iopub.status.busy": "2023-09-02T18:31:35.560593Z", + "iopub.status.idle": "2023-09-02T18:36:11.989204Z", + "shell.execute_reply": "2023-09-02T18:36:11.988160Z" + }, + "id": "wTsV6itKBthA", + "outputId": "482e002d-f3e5-4214-fcea-15ce05f6b1ce", + "papermill": { + "duration": 276.495416, + "end_time": "2023-09-02T18:36:12.020245", + "exception": false, + "start_time": "2023-09-02T18:31:35.524829", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Unnamed: 0textclass
02Ex Wife Threatening SuicideRecently left wife ...suicide
38need helpjust help im crying hardsuicide
49โ€™ lostHello , Adam ( 16 ) โ€™ struggling years โ€™...suicide
511Honetly idkI dont know im . feel like . All fe...suicide
612[ Trigger warning ] Excuse self inflicted burn...suicide
\n", + "
" + ], + "text/plain": [ + " Unnamed: 0 text class\n", + "0 2 Ex Wife Threatening SuicideRecently left wife ... suicide\n", + "3 8 need helpjust help im crying hard suicide\n", + "4 9 โ€™ lostHello , Adam ( 16 ) โ€™ struggling years โ€™... suicide\n", + "5 11 Honetly idkI dont know im . feel like . All fe... suicide\n", + "6 12 [ Trigger warning ] Excuse self inflicted burn... suicide" + ] + }, + "execution_count": 9, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "dataframe['text'] = dataframe['text'].apply(remove_stopwords)\n", + "dataframe.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "c884272f", + "metadata": { + "execution": { + "iopub.execute_input": "2023-09-02T18:36:12.091406Z", + "iopub.status.busy": "2023-09-02T18:36:12.090209Z", + "iopub.status.idle": "2023-09-02T18:36:12.100570Z", + "shell.execute_reply": "2023-09-02T18:36:12.098944Z" + }, + "id": "G0-F2LTlQ4zb", + "outputId": "442166c9-3815-4576-a0d1-593210341125", + "papermill": { + "duration": 0.048545, + "end_time": "2023-09-02T18:36:12.102940", + "exception": false, + "start_time": "2023-09-02T18:36:12.054395", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "CPU times: user 445 ยตs, sys: 0 ns, total: 445 ยตs\n", + "Wall time: 450 ยตs\n" + ] + } + ], + "source": [ + "%%time\n", + "from tqdm._tqdm_notebook import tqdm_notebook\n", + "tqdm_notebook.pandas()\n", + "def text_preprocessing(df,col_name):\n", + " column = col_name\n", + " df[column] = df[column].progress_apply(lambda x:str(x).lower())\n", + " df[column] = df[column].progress_apply(lambda x: th.remove_emails(x))\n", + " df[column] = df[column].progress_apply(lambda x: th.remove_html_tags(x))\n", + " df[column] = df[column].progress_apply(lambda x: th.remove_special_chars(x))\n", + " df[column] = df[column].progress_apply(lambda x: th.remove_accented_chars(x))\n", + " return(df)" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "f54a2cd9", + "metadata": { + "execution": { + "iopub.execute_input": "2023-09-02T18:36:12.174268Z", + "iopub.status.busy": "2023-09-02T18:36:12.173839Z", + "iopub.status.idle": "2023-09-02T18:36:45.284235Z", + "shell.execute_reply": "2023-09-02T18:36:45.283096Z" + }, + "id": "-Eb5RjJQRYJS", + "outputId": "38df7815-6953-449d-92cf-aa3786b9a4d7", + "papermill": { + "duration": 33.152397, + "end_time": "2023-09-02T18:36:45.289419", + "exception": false, + "start_time": "2023-09-02T18:36:12.137022", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "f8e66cdfb8dc493fa3f6ea4b93485bf1", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + " 0%| | 0/100000 [00:00.] - ETA: 0s - loss: 0.1993 - accuracy: 0.9216\n", + "Epoch 5: val_accuracy did not improve from 0.92190\n", + "547/547 [==============================] - 25s 46ms/step - loss: 0.1993 - accuracy: 0.9216 - val_loss: 0.2031 - val_accuracy: 0.9199\n", + "Epoch 6/25\n", + "547/547 [==============================] - ETA: 0s - loss: 0.1892 - accuracy: 0.9263\n", + "Epoch 6: val_accuracy improved from 0.92190 to 0.92583, saving model to ./model.h5\n", + "547/547 [==============================] - 23s 42ms/step - loss: 0.1892 - accuracy: 0.9263 - val_loss: 0.1938 - val_accuracy: 0.9258\n", + "Epoch 7/25\n", + "547/547 [==============================] - ETA: 0s - loss: 0.1813 - accuracy: 0.9296\n", + "Epoch 7: val_accuracy did not improve from 0.92583\n", + "547/547 [==============================] - 25s 46ms/step - loss: 0.1813 - accuracy: 0.9296 - val_loss: 0.1988 - val_accuracy: 0.9225\n", + "Epoch 8/25\n", + "547/547 [==============================] - ETA: 0s - loss: 0.1731 - accuracy: 0.9334\n", + "Epoch 8: val_accuracy improved from 0.92583 to 0.92670, saving model to ./model.h5\n", + "547/547 [==============================] - 25s 46ms/step - loss: 0.1731 - accuracy: 0.9334 - val_loss: 0.1881 - val_accuracy: 0.9267\n", + "Epoch 9/25\n", + "546/547 [============================>.] - ETA: 0s - loss: 0.1691 - accuracy: 0.9341\n", + "Epoch 9: val_accuracy did not improve from 0.92670\n", + "547/547 [==============================] - 25s 46ms/step - loss: 0.1691 - accuracy: 0.9341 - val_loss: 0.2034 - val_accuracy: 0.9222\n", + "Epoch 10/25\n", + "547/547 [==============================] - ETA: 0s - loss: 0.1616 - accuracy: 0.9378\n", + "Epoch 10: val_accuracy did not improve from 0.92670\n", + "547/547 [==============================] - 23s 42ms/step - loss: 0.1616 - accuracy: 0.9378 - val_loss: 0.1958 - val_accuracy: 0.9253\n", + "Epoch 11/25\n", + "547/547 [==============================] - ETA: 0s - loss: 0.1583 - accuracy: 0.9385\n", + "Epoch 11: val_accuracy did not improve from 0.92670\n", + "547/547 [==============================] - 25s 46ms/step - loss: 0.1583 - accuracy: 0.9385 - val_loss: 0.2086 - val_accuracy: 0.9209\n", + "Epoch 12/25\n", + "546/547 [============================>.] - ETA: 0s - loss: 0.1500 - accuracy: 0.9424\n", + "Epoch 12: val_accuracy improved from 0.92670 to 0.92903, saving model to ./model.h5\n", + "547/547 [==============================] - 25s 46ms/step - loss: 0.1499 - accuracy: 0.9424 - val_loss: 0.1906 - val_accuracy: 0.9290\n", + "Epoch 13/25\n", + "546/547 [============================>.] - ETA: 0s - loss: 0.1466 - accuracy: 0.9438\n", + "Epoch 13: val_accuracy did not improve from 0.92903\n", + "547/547 [==============================] - 23s 42ms/step - loss: 0.1467 - accuracy: 0.9438 - val_loss: 0.1867 - val_accuracy: 0.9275\n", + "Epoch 14/25\n", + "547/547 [==============================] - ETA: 0s - loss: 0.1405 - accuracy: 0.9455\n", + "Epoch 14: val_accuracy did not improve from 0.92903\n", + "547/547 [==============================] - 23s 42ms/step - loss: 0.1405 - accuracy: 0.9455 - val_loss: 0.1894 - val_accuracy: 0.9276\n", + "Epoch 15/25\n", + "547/547 [==============================] - ETA: 0s - loss: 0.1361 - accuracy: 0.9479\n", + "Epoch 15: val_accuracy did not improve from 0.92903\n", + "547/547 [==============================] - 23s 42ms/step - loss: 0.1361 - accuracy: 0.9479 - val_loss: 0.2147 - val_accuracy: 0.9216\n", + "Epoch 16/25\n", + "547/547 [==============================] - ETA: 0s - loss: 0.1307 - accuracy: 0.9498\n", + "Epoch 16: val_accuracy did not improve from 0.92903\n", + "547/547 [==============================] - 25s 46ms/step - loss: 0.1307 - accuracy: 0.9498 - val_loss: 0.2192 - val_accuracy: 0.9200\n", + "Epoch 17/25\n", + "547/547 [==============================] - ETA: 0s - loss: 0.1266 - accuracy: 0.9515\n", + "Epoch 17: val_accuracy did not improve from 0.92903\n", + "547/547 [==============================] - 23s 42ms/step - loss: 0.1266 - accuracy: 0.9515 - val_loss: 0.1945 - val_accuracy: 0.9276\n", + "Epoch 18/25\n", + "547/547 [==============================] - ETA: 0s - loss: 0.1218 - accuracy: 0.9530\n", + "Epoch 18: val_accuracy improved from 0.92903 to 0.92927, saving model to ./model.h5\n", + "547/547 [==============================] - 25s 46ms/step - loss: 0.1218 - accuracy: 0.9530 - val_loss: 0.2015 - val_accuracy: 0.9293\n", + "Epoch 18: early stopping\n" + ] + } + ], + "source": [ + "history_embedding = model.fit(X_train, y_train, \n", + " epochs = 25, batch_size = 128, \n", + " validation_data=(X_test, y_test),\n", + " verbose = 1, callbacks= [es, mc] )" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "id": "4061aacb", + "metadata": { + "execution": { + "iopub.execute_input": "2023-09-02T18:45:57.946292Z", + "iopub.status.busy": "2023-09-02T18:45:57.945885Z", + "iopub.status.idle": "2023-09-02T18:46:51.833306Z", + "shell.execute_reply": "2023-09-02T18:46:51.832226Z" + }, + "id": "WuKHZqHBkMok", + "outputId": "9e93abf1-e8f4-4e4a-fc78-80421a6703c6", + "papermill": { + "duration": 57.38806, + "end_time": "2023-09-02T18:46:53.613567", + "exception": false, + "start_time": "2023-09-02T18:45:56.225507", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "import matplotlib.pyplot as plt\n", + "import seaborn as sns\n", + "plt.plot(history_embedding.history['accuracy'],c='b',label='train accuracy')\n", + "plt.plot(history_embedding.history['val_accuracy'],c='r',label='validation accuracy')\n", + "plt.legend(loc='lower right')\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "id": "235d7796", + "metadata": { + "execution": { + "iopub.execute_input": "2023-09-02T18:46:57.225826Z", + "iopub.status.busy": "2023-09-02T18:46:57.225143Z", + "iopub.status.idle": "2023-09-02T18:46:57.564561Z", + "shell.execute_reply": "2023-09-02T18:46:57.563351Z" + }, + "id": "iHi9wj8sStvQ", + "outputId": "c074a850-45ed-4062-d954-7f7f3cc04a4d", + "papermill": { + "duration": 2.127078, + "end_time": "2023-09-02T18:46:57.567949", + "exception": false, + "start_time": "2023-09-02T18:46:55.440871", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "import matplotlib.pyplot as plt\n", + "import seaborn as sns\n", + "plt.plot(history_embedding.history['loss'],c='b',label='train loss')\n", + "plt.plot(history_embedding.history['val_loss'],c='r',label='validation loss')\n", + "plt.legend(loc='upper right')\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "id": "731e6880", + "metadata": { + "execution": { + "iopub.execute_input": "2023-09-02T18:47:01.537387Z", + "iopub.status.busy": "2023-09-02T18:47:01.536232Z", + "iopub.status.idle": "2023-09-02T18:47:12.158382Z", + "shell.execute_reply": "2023-09-02T18:47:12.157001Z" + }, + "id": "TaP7W8Vrk8zk", + "papermill": { + "duration": 12.404816, + "end_time": "2023-09-02T18:47:12.161336", + "exception": false, + "start_time": "2023-09-02T18:46:59.756520", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "938/938 [==============================] - 8s 9ms/step\n", + " precision recall f1-score support\n", + "\n", + " 0 0.94 0.92 0.93 14945\n", + " 1 0.92 0.94 0.93 15055\n", + "\n", + " accuracy 0.93 30000\n", + " macro avg 0.93 0.93 0.93 30000\n", + "weighted avg 0.93 0.93 0.93 30000\n", + "\n" + ] + } + ], + "source": [ + "from sklearn.metrics import classification_report, confusion_matrix\n", + "y_pred=model.predict(X_test)\n", + "y_pred.shape\n", + "import numpy as np\n", + "y_pred=np.argmax(y_pred,axis=1)\n", + "y_pred\n", + "y_test = y_test.to_numpy()\n", + "y_test=np.argmax(y_test,axis = 1)\n", + "print(classification_report(y_test, y_pred))" + ] + } + ], + "metadata": { + "accelerator": "GPU", + "colab": { + "collapsed_sections": [ + "ExcDIWLKN0b2", + "9iFYk83sQuee", + "QE41VfheTtxX", + "vn4lBP2kcSO8" + ], + "provenance": [] + }, + "gpuClass": "standard", + "kernelspec": { + "display_name": "Python 3", + "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": "3.7.12" + }, + "papermill": { + "default_parameters": {}, + "duration": 999.835608, + "end_time": "2023-09-02T18:47:17.441111", + "environment_variables": {}, + "exception": null, + "input_path": "__notebook__.ipynb", + "output_path": "__notebook__.ipynb", + "parameters": {}, + "start_time": "2023-09-02T18:30:37.605503", + "version": "2.4.0" + }, + "widgets": { + "application/vnd.jupyter.widget-state+json": { + "state": { + "04da949d7349412d960c34897b7fc5bf": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "06d8b427480a4bfcba9036e7df40b356": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "13d7db782fae4ac4acc36fcea50ad946": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_20403eae18e042b1a227d445559a9d8d", + "placeholder": "โ€‹", + "style": "IPY_MODEL_daf0a94f1c52446cb74089d0a421b933", + "value": "100%" + } + }, + "190c51d3b875404d99aea960de8a9631": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_37b04ee09cad40bc83b6f0dcbb4f1eb5", + "placeholder": "โ€‹", + "style": "IPY_MODEL_06d8b427480a4bfcba9036e7df40b356", + "value": "100%" + } + }, + "1da1f28655f74f0aba71593f100a3d65": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "20403eae18e042b1a227d445559a9d8d": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "27b922da58b0473ab0f1cf660fae2a24": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "30609b3983824c03b153bf3efaffb19b": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_364852904d5946d0809a25be65f0603b", + "placeholder": "โ€‹", + "style": "IPY_MODEL_818ef572eaf942da854e6c882d83e6b8", + "value": " 100000/100000 [00:00<00:00, 216781.80it/s]" + } + }, + "35926388a2c4490794ceaf7024dd6433": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "364852904d5946d0809a25be65f0603b": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "36c68825df1745a590481ba854e95ff2": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "ProgressStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "ProgressStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "bar_color": null, + "description_width": "" + } + }, + "37b04ee09cad40bc83b6f0dcbb4f1eb5": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "387fd6ca806142b6b711954a95cca3a7": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "395829dc498c40dd87b300f8ae117a35": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "39997e14a6544ccca86872d260bb041b": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "4dd4179ea129449b88f317815706469a": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_5897dea8600c48d5870e4027bceab39d", + "placeholder": "โ€‹", + "style": "IPY_MODEL_395829dc498c40dd87b300f8ae117a35", + "value": "100%" + } + }, + "558f17dd5da64c069d29fa073f39efd6": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_1da1f28655f74f0aba71593f100a3d65", + "placeholder": "โ€‹", + "style": "IPY_MODEL_a37332d6d541476780033dfc22aa028d", + "value": " 100000/100000 [00:22<00:00, 4592.57it/s]" + } + }, + "580e73b0af3e4662bf24780233cc6277": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "5897dea8600c48d5870e4027bceab39d": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "5c1de130702a48f29a41a0c775b35f65": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "ProgressStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "ProgressStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "bar_color": null, + "description_width": "" + } + }, + "652087bf023546b699fb6f6d9b5a2709": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "FloatProgressModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "FloatProgressModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "ProgressView", + "bar_style": "success", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_580e73b0af3e4662bf24780233cc6277", + "max": 100000, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_e23d7b0bab9b46b4a8c5018a8500c3a1", + "value": 100000 + } + }, + "675dad1a55994b05a0f0997871fb2e13": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "758fb8e96fff47afb22953cfbb79cad6": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "7d5ee6685ac8441e8526422da531e70b": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "ProgressStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "ProgressStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "bar_color": null, + "description_width": "" + } + }, + "818ef572eaf942da854e6c882d83e6b8": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "87471ae735844ed6b91923045c56d1c2": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "8e211abd90914dbb90866e173297916f": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_04da949d7349412d960c34897b7fc5bf", + "placeholder": "โ€‹", + "style": "IPY_MODEL_a5279ccaca974a7fa6e8bec0ce725fa9", + "value": " 100000/100000 [00:06<00:00, 20342.07it/s]" + } + }, + "9218ca1151554645ad3673ea99c02ce2": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "971f66621b2742ae8a207ac4c6060aae": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "FloatProgressModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "FloatProgressModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "ProgressView", + "bar_style": "success", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_cf8631ea54c744da8c2790da053a992c", + "max": 100000, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_36c68825df1745a590481ba854e95ff2", + "value": 100000 + } + }, + "9a72a2041ab945ee8f7b50b86d098db6": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "ProgressStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "ProgressStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "bar_color": null, + "description_width": "" + } + }, + "9db9f6b1559f432787ec66700259f9cc": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "a0d782f04d444afb977a5f45cfc6a241": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HBoxModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HBoxModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HBoxView", + "box_style": "", + "children": [ + "IPY_MODEL_4dd4179ea129449b88f317815706469a", + "IPY_MODEL_971f66621b2742ae8a207ac4c6060aae", + "IPY_MODEL_8e211abd90914dbb90866e173297916f" + ], + "layout": "IPY_MODEL_9db9f6b1559f432787ec66700259f9cc" + } + }, + "a1f407aaef194eaa980f5d98b7aa45ff": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HBoxModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HBoxModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HBoxView", + "box_style": "", + "children": [ + "IPY_MODEL_190c51d3b875404d99aea960de8a9631", + "IPY_MODEL_bfc99de7ea2145ebb4084a2eca1e16ca", + "IPY_MODEL_558f17dd5da64c069d29fa073f39efd6" + ], + "layout": "IPY_MODEL_d56e47eb0c854e559868dbc1682fe765" + } + }, + "a37332d6d541476780033dfc22aa028d": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "a4fc3ba9eb0f4d56b18c6b4c08d5e4e8": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "a5279ccaca974a7fa6e8bec0ce725fa9": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "ae05ce36086c40d89b521d637091dd36": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_87471ae735844ed6b91923045c56d1c2", + "placeholder": "โ€‹", + "style": "IPY_MODEL_a4fc3ba9eb0f4d56b18c6b4c08d5e4e8", + "value": "100%" + } + }, + "be3a4cb0e3e5412e9708b33953163bb6": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "bfc99de7ea2145ebb4084a2eca1e16ca": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "FloatProgressModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "FloatProgressModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "ProgressView", + "bar_style": "success", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_ea7f307b8cd34f629dc5c5b0c2fcc69f", + "max": 100000, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_5c1de130702a48f29a41a0c775b35f65", + "value": 100000 + } + }, + "c55c1f6b7d10488f93c118f9640f55c4": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "cf8631ea54c744da8c2790da053a992c": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "d56e47eb0c854e559868dbc1682fe765": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "d6d478726bf84b6d8594b33e02667bc7": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "d7832886ccf147528037e9831e042525": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "FloatProgressModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "FloatProgressModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "ProgressView", + "bar_style": "success", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_c55c1f6b7d10488f93c118f9640f55c4", + "max": 100000, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_7d5ee6685ac8441e8526422da531e70b", + "value": 100000 + } + }, + "dae6535051a54d6b8a0c93e986383887": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HBoxModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HBoxModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HBoxView", + "box_style": "", + "children": [ + "IPY_MODEL_e32b7b7c5e3e456088a1748d196963cc", + "IPY_MODEL_eecef2455f5d407d85235b185eceff66", + "IPY_MODEL_30609b3983824c03b153bf3efaffb19b" + ], + "layout": "IPY_MODEL_d6d478726bf84b6d8594b33e02667bc7" + } + }, + "daf0a94f1c52446cb74089d0a421b933": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "e23d7b0bab9b46b4a8c5018a8500c3a1": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "ProgressStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "ProgressStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "bar_color": null, + "description_width": "" + } + }, + "e32b7b7c5e3e456088a1748d196963cc": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_e96202178eea4b86933920ad73be7bc4", + "placeholder": "โ€‹", + "style": "IPY_MODEL_39997e14a6544ccca86872d260bb041b", + "value": "100%" + } + }, + "e77554c023fc4eb6b6d54a93f57b54ea": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_be3a4cb0e3e5412e9708b33953163bb6", + "placeholder": "โ€‹", + "style": "IPY_MODEL_35926388a2c4490794ceaf7024dd6433", + "value": " 100000/100000 [00:03<00:00, 45600.54it/s]" + } + }, + "e908bbc197c2478786fde69292842a5c": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HBoxModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HBoxModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HBoxView", + "box_style": "", + "children": [ + "IPY_MODEL_ae05ce36086c40d89b521d637091dd36", + "IPY_MODEL_d7832886ccf147528037e9831e042525", + "IPY_MODEL_e77554c023fc4eb6b6d54a93f57b54ea" + ], + "layout": "IPY_MODEL_27b922da58b0473ab0f1cf660fae2a24" + } + }, + "e96202178eea4b86933920ad73be7bc4": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "ea7f307b8cd34f629dc5c5b0c2fcc69f": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "ed00f96044b74670bdab3e9dbb0a46cf": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_675dad1a55994b05a0f0997871fb2e13", + "placeholder": "โ€‹", + "style": "IPY_MODEL_758fb8e96fff47afb22953cfbb79cad6", + "value": " 100000/100000 [00:00<00:00, 258128.81it/s]" + } + }, + "eecef2455f5d407d85235b185eceff66": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "FloatProgressModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "FloatProgressModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "ProgressView", + "bar_style": "success", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_9218ca1151554645ad3673ea99c02ce2", + "max": 100000, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_9a72a2041ab945ee8f7b50b86d098db6", + "value": 100000 + } + }, + "f8e66cdfb8dc493fa3f6ea4b93485bf1": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HBoxModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HBoxModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HBoxView", + "box_style": "", + "children": [ + "IPY_MODEL_13d7db782fae4ac4acc36fcea50ad946", + "IPY_MODEL_652087bf023546b699fb6f6d9b5a2709", + "IPY_MODEL_ed00f96044b74670bdab3e9dbb0a46cf" + ], + "layout": "IPY_MODEL_387fd6ca806142b6b711954a95cca3a7" + } + } + }, + "version_major": 2, + "version_minor": 0 + } + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/Sucide & Depression Detection/Models/suicidal-ideation-detection-bilstm-rnn.ipynb b/Sucide & Depression Detection/Models/suicidal-ideation-detection-bilstm-rnn.ipynb new file mode 100644 index 000000000..9ed3de0c6 --- /dev/null +++ b/Sucide & Depression Detection/Models/suicidal-ideation-detection-bilstm-rnn.ipynb @@ -0,0 +1,3438 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 1, + "id": "dd124687", + "metadata": { + "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19", + "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5", + "execution": { + "iopub.execute_input": "2024-04-18T19:23:15.357413Z", + "iopub.status.busy": "2024-04-18T19:23:15.357039Z", + "iopub.status.idle": "2024-04-18T19:23:21.283431Z", + "shell.execute_reply": "2024-04-18T19:23:21.282040Z" + }, + "papermill": { + "duration": 5.947662, + "end_time": "2024-04-18T19:23:21.287126", + "exception": false, + "start_time": "2024-04-18T19:23:15.339464", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Unnamed: 0textclass
02Ex Wife Threatening SuicideRecently I left my ...suicide
13Am I weird I don't get affected by compliments...non-suicide
24Finally 2020 is almost over... So I can never ...non-suicide
38i need helpjust help me im crying so hardsuicide
49Iรขย€ย™m so lostHello, my name is Adam (16) and I...suicide
............
232069348103If you don't like rock then your not going to ...non-suicide
232070348106You how you can tell i have so many friends an...non-suicide
232071348107pee probably tastes like salty teaรฐยŸย˜ยรฐยŸย’ยฆรขย€ยผรฏ...non-suicide
232072348108The usual stuff you find hereI'm not posting t...suicide
232073348110I still haven't beaten the first boss in Hollo...non-suicide
\n", + "

232074 rows ร— 3 columns

\n", + "
" + ], + "text/plain": [ + " Unnamed: 0 text \\\n", + "0 2 Ex Wife Threatening SuicideRecently I left my ... \n", + "1 3 Am I weird I don't get affected by compliments... \n", + "2 4 Finally 2020 is almost over... So I can never ... \n", + "3 8 i need helpjust help me im crying so hard \n", + "4 9 Iรขย€ย™m so lostHello, my name is Adam (16) and I... \n", + "... ... ... \n", + "232069 348103 If you don't like rock then your not going to ... \n", + "232070 348106 You how you can tell i have so many friends an... \n", + "232071 348107 pee probably tastes like salty teaรฐยŸย˜ยรฐยŸย’ยฆรขย€ยผรฏ... \n", + "232072 348108 The usual stuff you find hereI'm not posting t... \n", + "232073 348110 I still haven't beaten the first boss in Hollo... \n", + "\n", + " class \n", + "0 suicide \n", + "1 non-suicide \n", + "2 non-suicide \n", + "3 suicide \n", + "4 suicide \n", + "... ... \n", + "232069 non-suicide \n", + "232070 non-suicide \n", + "232071 non-suicide \n", + "232072 suicide \n", + "232073 non-suicide \n", + "\n", + "[232074 rows x 3 columns]" + ] + }, + "execution_count": 1, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "import pandas as pd\n", + "import numpy as np\n", + "from matplotlib import pyplot as plt\n", + "import nltk\n", + "from nltk.corpus import stopwords\n", + "from nltk.stem import WordNetLemmatizer\n", + "from nltk.tokenize import word_tokenize \n", + "df = pd.read_csv('/kaggle/input/suicide-watch/Suicide_Detection.csv', encoding=\"ISO-8859-1\")\n", + "df" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "78e9008b", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:23:21.322853Z", + "iopub.status.busy": "2024-04-18T19:23:21.322537Z", + "iopub.status.idle": "2024-04-18T19:23:21.447127Z", + "shell.execute_reply": "2024-04-18T19:23:21.446193Z" + }, + "papermill": { + "duration": 0.143796, + "end_time": "2024-04-18T19:23:21.449310", + "exception": false, + "start_time": "2024-04-18T19:23:21.305514", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Unnamed: 0textclass
02Ex Wife Threatening SuicideRecently I left my ...suicide
38i need helpjust help me im crying so hardsuicide
49Iรขย€ย™m so lostHello, my name is Adam (16) and I...suicide
511Honetly idkI dont know what im even doing here...suicide
612[Trigger warning] Excuse for self inflicted bu...suicide
............
5982089731She was only seventeen oh why are girls in son...non-suicide
5982189732i wanna talk to something until i go to sleep ...non-suicide
5982289733I know this is like super random but Fuck labe...non-suicide
5982589737Ok, I think I have officially decided That I a...non-suicide
5982689738What are reasonable/average grades for HS fres...non-suicide
\n", + "

60000 rows ร— 3 columns

\n", + "
" + ], + "text/plain": [ + " Unnamed: 0 text \\\n", + "0 2 Ex Wife Threatening SuicideRecently I left my ... \n", + "3 8 i need helpjust help me im crying so hard \n", + "4 9 Iรขย€ย™m so lostHello, my name is Adam (16) and I... \n", + "5 11 Honetly idkI dont know what im even doing here... \n", + "6 12 [Trigger warning] Excuse for self inflicted bu... \n", + "... ... ... \n", + "59820 89731 She was only seventeen oh why are girls in son... \n", + "59821 89732 i wanna talk to something until i go to sleep ... \n", + "59822 89733 I know this is like super random but Fuck labe... \n", + "59825 89737 Ok, I think I have officially decided That I a... \n", + "59826 89738 What are reasonable/average grades for HS fres... \n", + "\n", + " class \n", + "0 suicide \n", + "3 suicide \n", + "4 suicide \n", + "5 suicide \n", + "6 suicide \n", + "... ... \n", + "59820 non-suicide \n", + "59821 non-suicide \n", + "59822 non-suicide \n", + "59825 non-suicide \n", + "59826 non-suicide \n", + "\n", + "[60000 rows x 3 columns]" + ] + }, + "execution_count": 2, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "suicide = df[df['class']=='suicide']\n", + "non_suicide = df[df['class']== 'non-suicide']\n", + "suicide = suicide.head(30000)\n", + "non_suicide = non_suicide.head(30000)\n", + "df = pd.concat([suicide,non_suicide])\n", + "df" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "eb5c8c7a", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:23:21.480388Z", + "iopub.status.busy": "2024-04-18T19:23:21.480112Z", + "iopub.status.idle": "2024-04-18T19:23:21.517602Z", + "shell.execute_reply": "2024-04-18T19:23:21.516345Z" + }, + "papermill": { + "duration": 0.056088, + "end_time": "2024-04-18T19:23:21.519904", + "exception": false, + "start_time": "2024-04-18T19:23:21.463816", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Index: 60000 entries, 0 to 59826\n", + "Data columns (total 3 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 Unnamed: 0 60000 non-null int64 \n", + " 1 text 60000 non-null object\n", + " 2 class 60000 non-null object\n", + "dtypes: int64(1), object(2)\n", + "memory usage: 1.8+ MB\n" + ] + } + ], + "source": [ + "df.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "e70745d4", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:23:21.553038Z", + "iopub.status.busy": "2024-04-18T19:23:21.552680Z", + "iopub.status.idle": "2024-04-18T19:23:21.575874Z", + "shell.execute_reply": "2024-04-18T19:23:21.574838Z" + }, + "papermill": { + "duration": 0.042677, + "end_time": "2024-04-18T19:23:21.578147", + "exception": false, + "start_time": "2024-04-18T19:23:21.535470", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "text/plain": [ + "Unnamed: 0 0\n", + "text 0\n", + "class 0\n", + "dtype: int64" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df.isnull().sum()" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "cd77ba0a", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:23:21.641171Z", + "iopub.status.busy": "2024-04-18T19:23:21.640823Z", + "iopub.status.idle": "2024-04-18T19:24:18.372989Z", + "shell.execute_reply": "2024-04-18T19:24:18.371980Z" + }, + "papermill": { + "duration": 56.750262, + "end_time": "2024-04-18T19:24:18.374884", + "exception": false, + "start_time": "2024-04-18T19:23:21.624622", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Requirement already satisfied: tqdm in /opt/conda/lib/python3.10/site-packages (4.66.1)\r\n", + "Collecting text_hammer\r\n", + " Downloading text_hammer-0.1.5-py3-none-any.whl.metadata (2.2 kB)\r\n", + "Collecting beautifulsoup4==4.9.1 (from text_hammer)\r\n", + " Downloading beautifulsoup4-4.9.1-py3-none-any.whl.metadata (4.1 kB)\r\n", + "Requirement already satisfied: pandas in /opt/conda/lib/python3.10/site-packages (from text_hammer) (2.1.4)\r\n", + "Requirement already satisfied: numpy in /opt/conda/lib/python3.10/site-packages (from text_hammer) (1.24.4)\r\n", + "Requirement already satisfied: spacy in /opt/conda/lib/python3.10/site-packages (from text_hammer) (3.7.2)\r\n", + "Requirement already satisfied: TextBlob in /opt/conda/lib/python3.10/site-packages (from text_hammer) (0.17.1)\r\n", + "Requirement already satisfied: soupsieve>1.2 in /opt/conda/lib/python3.10/site-packages (from beautifulsoup4==4.9.1->text_hammer) (2.5)\r\n", + "Requirement already satisfied: python-dateutil>=2.8.2 in /opt/conda/lib/python3.10/site-packages (from pandas->text_hammer) (2.8.2)\r\n", + "Requirement already satisfied: pytz>=2020.1 in /opt/conda/lib/python3.10/site-packages (from pandas->text_hammer) (2023.3.post1)\r\n", + "Requirement already satisfied: tzdata>=2022.1 in /opt/conda/lib/python3.10/site-packages (from pandas->text_hammer) (2023.4)\r\n", + "Requirement already satisfied: spacy-legacy<3.1.0,>=3.0.11 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (3.0.12)\r\n", + "Requirement already satisfied: spacy-loggers<2.0.0,>=1.0.0 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (1.0.5)\r\n", + "Requirement already satisfied: murmurhash<1.1.0,>=0.28.0 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (1.0.10)\r\n", + "Requirement already satisfied: cymem<2.1.0,>=2.0.2 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (2.0.8)\r\n", + "Requirement already satisfied: preshed<3.1.0,>=3.0.2 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (3.0.9)\r\n", + "Requirement already satisfied: thinc<8.3.0,>=8.1.8 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (8.2.2)\r\n", + "Requirement already satisfied: wasabi<1.2.0,>=0.9.1 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (1.1.2)\r\n", + "Requirement already satisfied: srsly<3.0.0,>=2.4.3 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (2.4.8)\r\n", + "Requirement already satisfied: catalogue<2.1.0,>=2.0.6 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (2.0.10)\r\n", + "Requirement already satisfied: weasel<0.4.0,>=0.1.0 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (0.3.4)\r\n", + "Requirement already satisfied: typer<0.10.0,>=0.3.0 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (0.9.0)\r\n", + "Requirement already satisfied: smart-open<7.0.0,>=5.2.1 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (6.4.0)\r\n", + "Requirement already satisfied: tqdm<5.0.0,>=4.38.0 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (4.66.1)\r\n", + "Requirement already satisfied: requests<3.0.0,>=2.13.0 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (2.31.0)\r\n", + "Requirement already satisfied: pydantic!=1.8,!=1.8.1,<3.0.0,>=1.7.4 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (2.5.3)\r\n", + "Requirement already satisfied: jinja2 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (3.1.2)\r\n", + "Requirement already satisfied: setuptools in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (69.0.3)\r\n", + "Requirement already satisfied: packaging>=20.0 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (21.3)\r\n", + "Requirement already satisfied: langcodes<4.0.0,>=3.2.0 in /opt/conda/lib/python3.10/site-packages (from spacy->text_hammer) (3.3.0)\r\n", + "Requirement already satisfied: nltk>=3.1 in /opt/conda/lib/python3.10/site-packages (from TextBlob->text_hammer) (3.2.4)\r\n", + "Requirement already satisfied: six in /opt/conda/lib/python3.10/site-packages (from nltk>=3.1->TextBlob->text_hammer) (1.16.0)\r\n", + "Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in /opt/conda/lib/python3.10/site-packages (from packaging>=20.0->spacy->text_hammer) (3.1.1)\r\n", + "Requirement already satisfied: annotated-types>=0.4.0 in /opt/conda/lib/python3.10/site-packages (from pydantic!=1.8,!=1.8.1,<3.0.0,>=1.7.4->spacy->text_hammer) (0.6.0)\r\n", + "Requirement already satisfied: pydantic-core==2.14.6 in /opt/conda/lib/python3.10/site-packages (from pydantic!=1.8,!=1.8.1,<3.0.0,>=1.7.4->spacy->text_hammer) (2.14.6)\r\n", + "Requirement already satisfied: typing-extensions>=4.6.1 in /opt/conda/lib/python3.10/site-packages (from pydantic!=1.8,!=1.8.1,<3.0.0,>=1.7.4->spacy->text_hammer) (4.9.0)\r\n", + "Requirement already satisfied: charset-normalizer<4,>=2 in /opt/conda/lib/python3.10/site-packages (from requests<3.0.0,>=2.13.0->spacy->text_hammer) (3.3.2)\r\n", + "Requirement already satisfied: idna<4,>=2.5 in /opt/conda/lib/python3.10/site-packages (from requests<3.0.0,>=2.13.0->spacy->text_hammer) (3.6)\r\n", + "Requirement already satisfied: urllib3<3,>=1.21.1 in /opt/conda/lib/python3.10/site-packages (from requests<3.0.0,>=2.13.0->spacy->text_hammer) (1.26.18)\r\n", + "Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.10/site-packages (from requests<3.0.0,>=2.13.0->spacy->text_hammer) (2023.11.17)\r\n", + "Requirement already satisfied: blis<0.8.0,>=0.7.8 in /opt/conda/lib/python3.10/site-packages (from thinc<8.3.0,>=8.1.8->spacy->text_hammer) (0.7.10)\r\n", + "Requirement already satisfied: confection<1.0.0,>=0.0.1 in /opt/conda/lib/python3.10/site-packages (from thinc<8.3.0,>=8.1.8->spacy->text_hammer) (0.1.4)\r\n", + "Requirement already satisfied: click<9.0.0,>=7.1.1 in /opt/conda/lib/python3.10/site-packages (from typer<0.10.0,>=0.3.0->spacy->text_hammer) (8.1.7)\r\n", + "Requirement already satisfied: cloudpathlib<0.17.0,>=0.7.0 in /opt/conda/lib/python3.10/site-packages (from weasel<0.4.0,>=0.1.0->spacy->text_hammer) (0.16.0)\r\n", + "Requirement already satisfied: MarkupSafe>=2.0 in /opt/conda/lib/python3.10/site-packages (from jinja2->spacy->text_hammer) (2.1.3)\r\n", + "Downloading text_hammer-0.1.5-py3-none-any.whl (7.6 kB)\r\n", + "Downloading beautifulsoup4-4.9.1-py3-none-any.whl (115 kB)\r\n", + "\u001b[2K \u001b[90mโ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”\u001b[0m \u001b[32m115.1/115.1 kB\u001b[0m \u001b[31m5.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\r\n", + "\u001b[?25hInstalling collected packages: beautifulsoup4, text_hammer\r\n", + " Attempting uninstall: beautifulsoup4\r\n", + " Found existing installation: beautifulsoup4 4.12.2\r\n", + " Uninstalling beautifulsoup4-4.12.2:\r\n", + " Successfully uninstalled beautifulsoup4-4.12.2\r\n", + "\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\r\n", + "libpysal 4.9.2 requires beautifulsoup4>=4.10, but you have beautifulsoup4 4.9.1 which is incompatible.\r\n", + "libpysal 4.9.2 requires packaging>=22, but you have packaging 21.3 which is incompatible.\r\n", + "libpysal 4.9.2 requires shapely>=2.0.1, but you have shapely 1.8.5.post1 which is incompatible.\r\n", + "momepy 0.7.0 requires shapely>=2, but you have shapely 1.8.5.post1 which is incompatible.\r\n", + "spopt 0.6.0 requires shapely>=2.0.1, but you have shapely 1.8.5.post1 which is incompatible.\u001b[0m\u001b[31m\r\n", + "\u001b[0mSuccessfully installed beautifulsoup4-4.9.1 text_hammer-0.1.5\r\n", + "Collecting beautifulsoup4==4.12.2\r\n", + " Downloading beautifulsoup4-4.12.2-py3-none-any.whl.metadata (3.6 kB)\r\n", + "Downloading beautifulsoup4-4.12.2-py3-none-any.whl (142 kB)\r\n", + "\u001b[2K \u001b[90mโ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”โ”\u001b[0m \u001b[32m143.0/143.0 kB\u001b[0m \u001b[31m5.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\r\n", + "\u001b[?25hInstalling collected packages: beautifulsoup4\r\n", + " Attempting uninstall: beautifulsoup4\r\n", + " Found existing installation: beautifulsoup4 4.9.1\r\n", + " Uninstalling beautifulsoup4-4.9.1:\r\n", + " Successfully uninstalled beautifulsoup4-4.9.1\r\n", + "Successfully installed beautifulsoup4-4.12.2\r\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "100%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆ| 60000/60000 [00:00<00:00, 203319.28it/s]\n", + "100%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆ| 60000/60000 [00:04<00:00, 14221.40it/s]\n", + "100%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆ| 60000/60000 [00:11<00:00, 5201.42it/s]\n", + "100%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆ| 60000/60000 [00:03<00:00, 18516.80it/s]\n", + "100%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆ| 60000/60000 [00:00<00:00, 148440.99it/s]\n" + ] + } + ], + "source": [ + "!pip install tqdm\n", + "!pip install text_hammer\n", + "import text_hammer as th\n", + "!pip install --force-reinstall --no-deps beautifulsoup4==4.12.2\n", + "\n", + "\n", + "from tqdm import tqdm\n", + "import pandas as pd\n", + "import text_hammer as th\n", + "\n", + "tqdm.pandas()\n", + "\n", + "def text_preprocessing(df, col_name):\n", + " column = col_name\n", + " df[column] = df[column].progress_apply(lambda x: str(x).lower())\n", + " df[column] = df[column].progress_apply(lambda x: th.remove_emails(x))\n", + " df[column] = df[column].progress_apply(lambda x: th.remove_html_tags(x))\n", + " df[column] = df[column].progress_apply(lambda x: th.remove_special_chars(x))\n", + " df[column] = df[column].progress_apply(lambda x: th.remove_accented_chars(x))\n", + " return df\n", + "\n", + "df = text_preprocessing(df, 'text')" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "b4514909", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:18.436843Z", + "iopub.status.busy": "2024-04-18T19:24:18.435717Z", + "iopub.status.idle": "2024-04-18T19:24:18.448893Z", + "shell.execute_reply": "2024-04-18T19:24:18.447984Z" + }, + "papermill": { + "duration": 0.045875, + "end_time": "2024-04-18T19:24:18.450787", + "exception": false, + "start_time": "2024-04-18T19:24:18.404912", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Unnamed: 0textclass
02ex wife threatening suiciderecently i left my ...suicide
38i need helpjust help me im crying so hardsuicide
49iam so losthello my name is adam 16 and iave b...suicide
511honetly idki dont know what im even doing here...suicide
612trigger warning excuse for self inflicted burn...suicide
713it ends tonighti canat do it anymore i quitsuicide
918my life is over at 20 years oldhello all i am ...suicide
1019i took the rest of my sleeping pills and my pa...suicide
1120can you imagine getting old me neitherwrinkles...suicide
1221do you think getting hit by a train would be p...suicide
1322death continuedi posted here before and saw so...suicide
1423been arrested feeling suicidaleditsuicide
1625iam scared everything just seems to be getting...suicide
1929yeaputting a knife to my wrist didnt give me a...suicide
2030i am ending my life today goodbye everyonei am...suicide
2232trapped inside a voiddear whoever cares enough...suicide
2536the graveyard of redditanyone find it eery to ...suicide
2738i think today may be my lasteverythings becomi...suicide
2839iam trashlol i normally cringe at the self loa...suicide
3041what is the best way to do itiam not looking t...suicide
\n", + "
" + ], + "text/plain": [ + " Unnamed: 0 text class\n", + "0 2 ex wife threatening suiciderecently i left my ... suicide\n", + "3 8 i need helpjust help me im crying so hard suicide\n", + "4 9 iam so losthello my name is adam 16 and iave b... suicide\n", + "5 11 honetly idki dont know what im even doing here... suicide\n", + "6 12 trigger warning excuse for self inflicted burn... suicide\n", + "7 13 it ends tonighti canat do it anymore i quit suicide\n", + "9 18 my life is over at 20 years oldhello all i am ... suicide\n", + "10 19 i took the rest of my sleeping pills and my pa... suicide\n", + "11 20 can you imagine getting old me neitherwrinkles... suicide\n", + "12 21 do you think getting hit by a train would be p... suicide\n", + "13 22 death continuedi posted here before and saw so... suicide\n", + "14 23 been arrested feeling suicidaledit suicide\n", + "16 25 iam scared everything just seems to be getting... suicide\n", + "19 29 yeaputting a knife to my wrist didnt give me a... suicide\n", + "20 30 i am ending my life today goodbye everyonei am... suicide\n", + "22 32 trapped inside a voiddear whoever cares enough... suicide\n", + "25 36 the graveyard of redditanyone find it eery to ... suicide\n", + "27 38 i think today may be my lasteverythings becomi... suicide\n", + "28 39 iam trashlol i normally cringe at the self loa... suicide\n", + "30 41 what is the best way to do itiam not looking t... suicide" + ] + }, + "execution_count": 6, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df.head(20)" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "76a402b0", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:18.512730Z", + "iopub.status.busy": "2024-04-18T19:24:18.512264Z", + "iopub.status.idle": "2024-04-18T19:24:18.522117Z", + "shell.execute_reply": "2024-04-18T19:24:18.521411Z" + }, + "papermill": { + "duration": 0.042921, + "end_time": "2024-04-18T19:24:18.524044", + "exception": false, + "start_time": "2024-04-18T19:24:18.481123", + "status": "completed" + }, + "tags": [] + }, + "outputs": [], + "source": [ + "df_preprocess = df.copy()\n", + "posts = df_preprocess.text.copy()" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "0a285d7b", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:18.584373Z", + "iopub.status.busy": "2024-04-18T19:24:18.584094Z", + "iopub.status.idle": "2024-04-18T19:24:27.107429Z", + "shell.execute_reply": "2024-04-18T19:24:27.106606Z" + }, + "papermill": { + "duration": 8.556036, + "end_time": "2024-04-18T19:24:27.109667", + "exception": false, + "start_time": "2024-04-18T19:24:18.553631", + "status": "completed" + }, + "tags": [] + }, + "outputs": [], + "source": [ + "def removeWordWithChar(text, char_list):\n", + " #Remove words in a text that contains a char from the list.\n", + " text = text.split()\n", + " res = [ele for ele in text if all(ch not in ele for ch in char_list)]\n", + " res = ' '.join(res)\n", + " return res\n", + "\n", + "char_list = ['@', '#', 'http', 'www', '/', '[]']\n", + "\n", + "removeWordWithChar(posts[1], char_list)\n", + "\n", + "posts_cleaned = []\n", + "\n", + "for p in posts:\n", + " posts_cleaned.append(removeWordWithChar(p, char_list))" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "cd2d0bbc", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:27.172190Z", + "iopub.status.busy": "2024-04-18T19:24:27.171815Z", + "iopub.status.idle": "2024-04-18T19:24:27.177390Z", + "shell.execute_reply": "2024-04-18T19:24:27.176532Z" + }, + "papermill": { + "duration": 0.038909, + "end_time": "2024-04-18T19:24:27.179198", + "exception": false, + "start_time": "2024-04-18T19:24:27.140289", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "text/plain": [ + "'ex wife threatening suiciderecently i left my wife for good because she has cheated on me twice and lied to me so much that i have decided to refuse to go back to her as of a few days ago she began threatening suicide i have tirelessly spent these paat few days talking her out of it and she keeps hesitating because she wants to believe ill come back i know a lot of people will threaten this in order to get their way but what happens if she really does what do i do and how am i supposed to handle her death on my hands i still love my wife but i cannot deal with getting cheated on again and constantly feeling insecure im worried today may be the day she does it and i hope so much it doesnt happen'" + ] + }, + "execution_count": 9, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "posts_cleaned[0]" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "d52888f1", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:27.240256Z", + "iopub.status.busy": "2024-04-18T19:24:27.239716Z", + "iopub.status.idle": "2024-04-18T19:24:27.245028Z", + "shell.execute_reply": "2024-04-18T19:24:27.244217Z" + }, + "papermill": { + "duration": 0.037563, + "end_time": "2024-04-18T19:24:27.246764", + "exception": false, + "start_time": "2024-04-18T19:24:27.209201", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "text/plain": [ + "60000" + ] + }, + "execution_count": 10, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "len(posts_cleaned)" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "6183a5ab", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:27.369610Z", + "iopub.status.busy": "2024-04-18T19:24:27.368835Z", + "iopub.status.idle": "2024-04-18T19:24:27.374412Z", + "shell.execute_reply": "2024-04-18T19:24:27.373749Z" + }, + "papermill": { + "duration": 0.039526, + "end_time": "2024-04-18T19:24:27.376156", + "exception": false, + "start_time": "2024-04-18T19:24:27.336630", + "status": "completed" + }, + "tags": [] + }, + "outputs": [], + "source": [ + "\n", + "def tokenize(texts):\n", + " tokenizer = nltk.RegexpTokenizer(r'\\w+')\n", + "\n", + " texts_tokens = []\n", + " for i, val in enumerate(texts):\n", + " text_tokens = tokenizer.tokenize(val.lower())\n", + "\n", + " for i in range(len(text_tokens) - 1, -1, -1):\n", + " if len(text_tokens[i]) < 4:\n", + " del (text_tokens[i])\n", + "\n", + " texts_tokens.append(text_tokens)\n", + "\n", + " return texts_tokens" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "60d05498", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:27.478571Z", + "iopub.status.busy": "2024-04-18T19:24:27.477820Z", + "iopub.status.idle": "2024-04-18T19:24:31.564469Z", + "shell.execute_reply": "2024-04-18T19:24:31.563602Z" + }, + "papermill": { + "duration": 4.160516, + "end_time": "2024-04-18T19:24:31.566555", + "exception": false, + "start_time": "2024-04-18T19:24:27.406039", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "text/plain": [ + "[['wife',\n", + " 'threatening',\n", + " 'suiciderecently',\n", + " 'left',\n", + " 'wife',\n", + " 'good',\n", + " 'because',\n", + " 'cheated',\n", + " 'twice',\n", + " 'lied',\n", + " 'much',\n", + " 'that',\n", + " 'have',\n", + " 'decided',\n", + " 'refuse',\n", + " 'back',\n", + " 'days',\n", + " 'began',\n", + " 'threatening',\n", + " 'suicide',\n", + " 'have',\n", + " 'tirelessly',\n", + " 'spent',\n", + " 'these',\n", + " 'paat',\n", + " 'days',\n", + " 'talking',\n", + " 'keeps',\n", + " 'hesitating',\n", + " 'because',\n", + " 'wants',\n", + " 'believe',\n", + " 'come',\n", + " 'back',\n", + " 'know',\n", + " 'people',\n", + " 'will',\n", + " 'threaten',\n", + " 'this',\n", + " 'order',\n", + " 'their',\n", + " 'what',\n", + " 'happens',\n", + " 'really',\n", + " 'does',\n", + " 'what',\n", + " 'supposed',\n", + " 'handle',\n", + " 'death',\n", + " 'hands',\n", + " 'still',\n", + " 'love',\n", + " 'wife',\n", + " 'cannot',\n", + " 'deal',\n", + " 'with',\n", + " 'getting',\n", + " 'cheated',\n", + " 'again',\n", + " 'constantly',\n", + " 'feeling',\n", + " 'insecure',\n", + " 'worried',\n", + " 'today',\n", + " 'does',\n", + " 'hope',\n", + " 'much',\n", + " 'doesnt',\n", + " 'happen']]" + ] + }, + "execution_count": 12, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "posts_tokens = tokenize(posts_cleaned)\n", + "posts_tokens[:1]" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "7a42e1ec", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:31.689718Z", + "iopub.status.busy": "2024-04-18T19:24:31.689427Z", + "iopub.status.idle": "2024-04-18T19:24:31.812520Z", + "shell.execute_reply": "2024-04-18T19:24:31.811601Z" + }, + "papermill": { + "duration": 0.156381, + "end_time": "2024-04-18T19:24:31.814409", + "exception": false, + "start_time": "2024-04-18T19:24:31.658028", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[nltk_data] Downloading package stopwords to /usr/share/nltk_data...\n", + "[nltk_data] Package stopwords is already up-to-date!\n", + "[nltk_data] Downloading package punkt to /usr/share/nltk_data...\n", + "[nltk_data] Package punkt is already up-to-date!\n" + ] + } + ], + "source": [ + "\n", + "nltk.download('stopwords')\n", + "nltk.download('punkt')\n", + "\n", + "def removeSW(texts_tokens):\n", + " stopWords = set(stopwords.words('english'))\n", + " texts_filtered = []\n", + "\n", + " for i, val in enumerate(texts_tokens):\n", + " text_filtered = []\n", + " for w in val:\n", + " if w not in stopWords:\n", + " text_filtered.append(w)\n", + " texts_filtered.append(text_filtered)\n", + "\n", + " return texts_filtered" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "5b1b1aec", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:31.876060Z", + "iopub.status.busy": "2024-04-18T19:24:31.875761Z", + "iopub.status.idle": "2024-04-18T19:24:32.978125Z", + "shell.execute_reply": "2024-04-18T19:24:32.977133Z" + }, + "papermill": { + "duration": 1.135721, + "end_time": "2024-04-18T19:24:32.980331", + "exception": false, + "start_time": "2024-04-18T19:24:31.844610", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "text/plain": [ + "[['wife',\n", + " 'threatening',\n", + " 'suiciderecently',\n", + " 'left',\n", + " 'wife',\n", + " 'good',\n", + " 'cheated',\n", + " 'twice',\n", + " 'lied',\n", + " 'much',\n", + " 'decided',\n", + " 'refuse',\n", + " 'back',\n", + " 'days',\n", + " 'began',\n", + " 'threatening',\n", + " 'suicide',\n", + " 'tirelessly',\n", + " 'spent',\n", + " 'paat',\n", + " 'days',\n", + " 'talking',\n", + " 'keeps',\n", + " 'hesitating',\n", + " 'wants',\n", + " 'believe',\n", + " 'come',\n", + " 'back',\n", + " 'know',\n", + " 'people',\n", + " 'threaten',\n", + " 'order',\n", + " 'happens',\n", + " 'really',\n", + " 'supposed',\n", + " 'handle',\n", + " 'death',\n", + " 'hands',\n", + " 'still',\n", + " 'love',\n", + " 'wife',\n", + " 'cannot',\n", + " 'deal',\n", + " 'getting',\n", + " 'cheated',\n", + " 'constantly',\n", + " 'feeling',\n", + " 'insecure',\n", + " 'worried',\n", + " 'today',\n", + " 'hope',\n", + " 'much',\n", + " 'doesnt',\n", + " 'happen']]" + ] + }, + "execution_count": 14, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "\n", + "posts_filtered = removeSW(posts_tokens)\n", + "posts_filtered[:1]" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "7f9338b0", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:33.043835Z", + "iopub.status.busy": "2024-04-18T19:24:33.043536Z", + "iopub.status.idle": "2024-04-18T19:24:34.503453Z", + "shell.execute_reply": "2024-04-18T19:24:34.502453Z" + }, + "papermill": { + "duration": 1.493799, + "end_time": "2024-04-18T19:24:34.505669", + "exception": false, + "start_time": "2024-04-18T19:24:33.011870", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from collections import Counter\n", + "from collections import Counter\n", + "import matplotlib.pyplot as plt\n", + "import seaborn as sns\n", + "\n", + "# Count word frequencies\n", + "word_counts = Counter(word for post in posts_filtered for word in post)\n", + "\n", + "# Convert to DataFrame\n", + "word_freq = pd.DataFrame(word_counts.items(), columns=['word', 'count'])\n", + "\n", + "# Sort by count in descending order\n", + "word_freq = word_freq.sort_values(by='count', ascending=False)\n", + "\n", + "# Plot the top 50 words\n", + "plt.figure(figsize=(15, 20))\n", + "sns.barplot(x='count', y='word', data=word_freq.iloc[:50])\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "id": "567e6445", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:34.570934Z", + "iopub.status.busy": "2024-04-18T19:24:34.570396Z", + "iopub.status.idle": "2024-04-18T19:24:35.219063Z", + "shell.execute_reply": "2024-04-18T19:24:35.218121Z" + }, + "papermill": { + "duration": 0.683629, + "end_time": "2024-04-18T19:24:35.221586", + "exception": false, + "start_time": "2024-04-18T19:24:34.537957", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from wordcloud import WordCloud # Import WordCloud\n", + "\n", + "# Assuming you have defined feature_names\n", + "feature_names = word_freq['word'].values\n", + "\n", + "# Create WordCloud object\n", + "wc = WordCloud(max_words=300)\n", + "\n", + "# Generate WordCloud\n", + "wc.generate(' '.join(word for word in feature_names[500:3500]))\n", + "\n", + "# Plot the WordCloud\n", + "plt.figure(figsize=(17, 12))\n", + "plt.axis('off')\n", + "plt.imshow(wc)\n", + "plt.show()\n" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "id": "8ed86d6f", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:35.290367Z", + "iopub.status.busy": "2024-04-18T19:24:35.289751Z", + "iopub.status.idle": "2024-04-18T19:24:35.299392Z", + "shell.execute_reply": "2024-04-18T19:24:35.298533Z" + }, + "papermill": { + "duration": 0.045747, + "end_time": "2024-04-18T19:24:35.301347", + "exception": false, + "start_time": "2024-04-18T19:24:35.255600", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Unnamed: 0textclass
02ex wife threatening suiciderecently i left my ...suicide
38i need helpjust help me im crying so hardsuicide
49iam so losthello my name is adam 16 and iave b...suicide
511honetly idki dont know what im even doing here...suicide
612trigger warning excuse for self inflicted burn...suicide
\n", + "
" + ], + "text/plain": [ + " Unnamed: 0 text class\n", + "0 2 ex wife threatening suiciderecently i left my ... suicide\n", + "3 8 i need helpjust help me im crying so hard suicide\n", + "4 9 iam so losthello my name is adam 16 and iave b... suicide\n", + "5 11 honetly idki dont know what im even doing here... suicide\n", + "6 12 trigger warning excuse for self inflicted burn... suicide" + ] + }, + "execution_count": 17, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "id": "3e3f9ed4", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:35.438986Z", + "iopub.status.busy": "2024-04-18T19:24:35.438212Z", + "iopub.status.idle": "2024-04-18T19:24:35.443468Z", + "shell.execute_reply": "2024-04-18T19:24:35.442624Z" + }, + "papermill": { + "duration": 0.041227, + "end_time": "2024-04-18T19:24:35.445300", + "exception": false, + "start_time": "2024-04-18T19:24:35.404073", + "status": "completed" + }, + "tags": [] + }, + "outputs": [], + "source": [ + "def lemma(texts_filtered):\n", + " wordnet_lemmatizer = WordNetLemmatizer()\n", + " texts_lem = []\n", + "\n", + " for i, val in enumerate(texts_filtered):\n", + " text_lem = []\n", + " for word in val:\n", + " text_lem.append(wordnet_lemmatizer.lemmatize(word, pos=\"v\"))\n", + " texts_lem.append(text_lem)\n", + "\n", + " return texts_lem\n" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "id": "a7039f8d", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:35.513556Z", + "iopub.status.busy": "2024-04-18T19:24:35.512994Z", + "iopub.status.idle": "2024-04-18T19:24:58.066791Z", + "shell.execute_reply": "2024-04-18T19:24:58.065899Z" + }, + "papermill": { + "duration": 22.590471, + "end_time": "2024-04-18T19:24:58.068962", + "exception": false, + "start_time": "2024-04-18T19:24:35.478491", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[nltk_data] Downloading package wordnet to /kaggle/working/...\n", + "Archive: /kaggle/working/corpora/wordnet.zip\n", + " creating: /kaggle/working/corpora/wordnet/\n", + " inflating: /kaggle/working/corpora/wordnet/lexnames \n", + " inflating: /kaggle/working/corpora/wordnet/data.verb \n", + " inflating: /kaggle/working/corpora/wordnet/index.adv \n", + " inflating: /kaggle/working/corpora/wordnet/adv.exc \n", + " inflating: /kaggle/working/corpora/wordnet/index.verb \n", + " inflating: /kaggle/working/corpora/wordnet/cntlist.rev \n", + " inflating: /kaggle/working/corpora/wordnet/data.adj \n", + " inflating: /kaggle/working/corpora/wordnet/index.adj \n", + " inflating: /kaggle/working/corpora/wordnet/LICENSE \n", + " inflating: /kaggle/working/corpora/wordnet/citation.bib \n", + " inflating: /kaggle/working/corpora/wordnet/noun.exc \n", + " inflating: /kaggle/working/corpora/wordnet/verb.exc \n", + " inflating: /kaggle/working/corpora/wordnet/README \n", + " inflating: /kaggle/working/corpora/wordnet/index.sense \n", + " inflating: /kaggle/working/corpora/wordnet/data.noun \n", + " inflating: /kaggle/working/corpora/wordnet/data.adv \n", + " inflating: /kaggle/working/corpora/wordnet/index.noun \n", + " inflating: /kaggle/working/corpora/wordnet/adj.exc \n" + ] + }, + { + "data": { + "text/plain": [ + "['wife',\n", + " 'threaten',\n", + " 'suiciderecently',\n", + " 'leave',\n", + " 'wife',\n", + " 'good',\n", + " 'cheat',\n", + " 'twice',\n", + " 'lie',\n", + " 'much',\n", + " 'decide',\n", + " 'refuse',\n", + " 'back',\n", + " 'days',\n", + " 'begin',\n", + " 'threaten',\n", + " 'suicide',\n", + " 'tirelessly',\n", + " 'spend',\n", + " 'paat',\n", + " 'days',\n", + " 'talk',\n", + " 'keep',\n", + " 'hesitate',\n", + " 'want',\n", + " 'believe',\n", + " 'come',\n", + " 'back',\n", + " 'know',\n", + " 'people',\n", + " 'threaten',\n", + " 'order',\n", + " 'happen',\n", + " 'really',\n", + " 'suppose',\n", + " 'handle',\n", + " 'death',\n", + " 'hand',\n", + " 'still',\n", + " 'love',\n", + " 'wife',\n", + " 'cannot',\n", + " 'deal',\n", + " 'get',\n", + " 'cheat',\n", + " 'constantly',\n", + " 'feel',\n", + " 'insecure',\n", + " 'worry',\n", + " 'today',\n", + " 'hope',\n", + " 'much',\n", + " 'doesnt',\n", + " 'happen']" + ] + }, + "execution_count": 19, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "import nltk\n", + "import subprocess\n", + "\n", + "try:\n", + " nltk.data.find('wordnet.zip')\n", + "except:\n", + " nltk.download('wordnet', download_dir='/kaggle/working/')\n", + " command = \"unzip /kaggle/working/corpora/wordnet.zip -d /kaggle/working/corpora\"\n", + " subprocess.run(command.split())\n", + " nltk.data.path.append('/kaggle/working/')\n", + "\n", + "from nltk.corpus import wordnet\n", + "\n", + "posts_lem = lemma(posts_filtered)\n", + "\n", + "posts_lem[0]" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "id": "e8b02628", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:58.137834Z", + "iopub.status.busy": "2024-04-18T19:24:58.137523Z", + "iopub.status.idle": "2024-04-18T19:24:58.304298Z", + "shell.execute_reply": "2024-04-18T19:24:58.303462Z" + }, + "papermill": { + "duration": 0.203188, + "end_time": "2024-04-18T19:24:58.306475", + "exception": false, + "start_time": "2024-04-18T19:24:58.103287", + "status": "completed" + }, + "tags": [] + }, + "outputs": [], + "source": [ + "posts_ready = []\n", + "for posts in posts_lem:\n", + " string = ' '\n", + " string = string.join(posts)\n", + " posts_ready.append(string)" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "id": "725d486c", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:58.376314Z", + "iopub.status.busy": "2024-04-18T19:24:58.375999Z", + "iopub.status.idle": "2024-04-18T19:24:58.381774Z", + "shell.execute_reply": "2024-04-18T19:24:58.380935Z" + }, + "papermill": { + "duration": 0.042542, + "end_time": "2024-04-18T19:24:58.383732", + "exception": false, + "start_time": "2024-04-18T19:24:58.341190", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "text/plain": [ + "60000" + ] + }, + "execution_count": 21, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "len(posts_ready)" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "id": "0b73e2fe", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:58.453841Z", + "iopub.status.busy": "2024-04-18T19:24:58.453541Z", + "iopub.status.idle": "2024-04-18T19:24:58.478180Z", + "shell.execute_reply": "2024-04-18T19:24:58.477327Z" + }, + "papermill": { + "duration": 0.06263, + "end_time": "2024-04-18T19:24:58.479986", + "exception": false, + "start_time": "2024-04-18T19:24:58.417356", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
original_textpreprocessed_textclass
0ex wife threatening suiciderecently i left my ...wife threaten suiciderecently leave wife good ...suicide
3i need helpjust help me im crying so hardneed helpjust help cry hardsuicide
4iam so losthello my name is adam 16 and iave b...losthello name adam iave struggle years afraid...suicide
5honetly idki dont know what im even doing here...honetly idki dont know even feel like nothing ...suicide
6trigger warning excuse for self inflicted burn...trigger warn excuse self inflict burnsi know c...suicide
\n", + "
" + ], + "text/plain": [ + " original_text \\\n", + "0 ex wife threatening suiciderecently i left my ... \n", + "3 i need helpjust help me im crying so hard \n", + "4 iam so losthello my name is adam 16 and iave b... \n", + "5 honetly idki dont know what im even doing here... \n", + "6 trigger warning excuse for self inflicted burn... \n", + "\n", + " preprocessed_text class \n", + "0 wife threaten suiciderecently leave wife good ... suicide \n", + "3 need helpjust help cry hard suicide \n", + "4 losthello name adam iave struggle years afraid... suicide \n", + "5 honetly idki dont know even feel like nothing ... suicide \n", + "6 trigger warn excuse self inflict burnsi know c... suicide " + ] + }, + "execution_count": 22, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df_preprocess['original_text'] = df['text']\n", + "df_preprocess['preprocessed_text'] = posts_ready\n", + "df_preprocess[['original_text', 'preprocessed_text', 'class']].head()" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "id": "f1e56d9f", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:58.549760Z", + "iopub.status.busy": "2024-04-18T19:24:58.549495Z", + "iopub.status.idle": "2024-04-18T19:24:58.634342Z", + "shell.execute_reply": "2024-04-18T19:24:58.633384Z" + }, + "papermill": { + "duration": 0.122139, + "end_time": "2024-04-18T19:24:58.636726", + "exception": false, + "start_time": "2024-04-18T19:24:58.514587", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
textclass
0wife threaten suiciderecently leave wife good ...suicide
1need helpjust help cry hardsuicide
2losthello name adam iave struggle years afraid...suicide
3honetly idki dont know even feel like nothing ...suicide
4trigger warn excuse self inflict burnsi know c...suicide
.........
59995seventeen girls songs always seventeen movie s...non-suicide
59996wanna talk something sleep check recent post h...non-suicide
59997know like super random fuck label wantas long ...non-suicide
59998think officially decide satanistthe satanic te...non-suicide
59999reasonableaverage grade freshmans many student...non-suicide
\n", + "

60000 rows ร— 2 columns

\n", + "
" + ], + "text/plain": [ + " text class\n", + "0 wife threaten suiciderecently leave wife good ... suicide\n", + "1 need helpjust help cry hard suicide\n", + "2 losthello name adam iave struggle years afraid... suicide\n", + "3 honetly idki dont know even feel like nothing ... suicide\n", + "4 trigger warn excuse self inflict burnsi know c... suicide\n", + "... ... ...\n", + "59995 seventeen girls songs always seventeen movie s... non-suicide\n", + "59996 wanna talk something sleep check recent post h... non-suicide\n", + "59997 know like super random fuck label wantas long ... non-suicide\n", + "59998 think officially decide satanistthe satanic te... non-suicide\n", + "59999 reasonableaverage grade freshmans many student... non-suicide\n", + "\n", + "[60000 rows x 2 columns]" + ] + }, + "execution_count": 23, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data = [{'text': text, 'class': label} for text, label in zip(posts_ready, df['class'])]\n", + "\n", + "# Convert the list of dictionaries to a DataFrame\n", + "df_ready = pd.DataFrame(data)\n", + "\n", + "X = df_ready['text']\n", + "y = df_ready['class']\n", + "\n", + "df_ready" + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "id": "09318cc7", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:24:58.783603Z", + "iopub.status.busy": "2024-04-18T19:24:58.783231Z", + "iopub.status.idle": "2024-04-18T19:25:13.948813Z", + "shell.execute_reply": "2024-04-18T19:25:13.947968Z" + }, + "papermill": { + "duration": 15.206639, + "end_time": "2024-04-18T19:25:13.951095", + "exception": false, + "start_time": "2024-04-18T19:24:58.744456", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2024-04-18 19:25:00.606537: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n", + "2024-04-18 19:25:00.606631: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n", + "2024-04-18 19:25:00.744374: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n" + ] + } + ], + "source": [ + "from keras.preprocessing.text import Tokenizer\n", + "tokenizer=Tokenizer(num_words= 30000,lower=True)\n", + "tokenizer.fit_on_texts(X)\n" + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "id": "a4e78c24", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:25:14.028790Z", + "iopub.status.busy": "2024-04-18T19:25:14.027773Z", + "iopub.status.idle": "2024-04-18T19:25:16.977166Z", + "shell.execute_reply": "2024-04-18T19:25:16.976070Z" + }, + "papermill": { + "duration": 2.986879, + "end_time": "2024-04-18T19:25:16.979311", + "exception": false, + "start_time": "2024-04-18T19:25:13.992432", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "text/plain": [ + "array([ 30, 12973, 20, 217, 96, 0, 0, 0, 0,\n", + " 0, 0, 0, 0, 0, 0, 0, 0, 0,\n", + " 0, 0, 0, 0, 0, 0, 0, 0, 0,\n", + " 0, 0, 0, 0, 0, 0, 0, 0, 0,\n", + " 0, 0, 0, 0, 0, 0, 0, 0, 0,\n", + " 0, 0, 0, 0, 0, 0, 0, 0, 0,\n", + " 0, 0, 0, 0, 0, 0, 0, 0, 0,\n", + " 0, 0, 0, 0, 0, 0, 0, 0, 0,\n", + " 0, 0, 0, 0, 0, 0, 0, 0, 0,\n", + " 0, 0, 0, 0, 0, 0, 0, 0, 0,\n", + " 0, 0, 0, 0, 0, 0, 0, 0, 0,\n", + " 0, 0, 0, 0, 0, 0, 0, 0, 0,\n", + " 0, 0, 0, 0, 0, 0, 0, 0, 0,\n", + " 0, 0, 0, 0, 0, 0, 0, 0, 0,\n", + " 0, 0, 0, 0, 0, 0, 0, 0, 0,\n", + " 0, 0, 0, 0, 0, 0, 0, 0, 0,\n", + " 0, 0, 0, 0, 0, 0], dtype=int32)" + ] + }, + "execution_count": 25, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "from tensorflow.keras.preprocessing.sequence import pad_sequences\n", + "X = tokenizer.texts_to_sequences(X) # this converts texts into some numeric sequences \n", + "X = pad_sequences(X,maxlen=150,padding='post') # this makes the length of all numeric sequences equal \n", + "X[1]" + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "id": "15ddc6a4", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:25:17.050728Z", + "iopub.status.busy": "2024-04-18T19:25:17.050369Z", + "iopub.status.idle": "2024-04-18T19:25:17.056139Z", + "shell.execute_reply": "2024-04-18T19:25:17.055291Z" + }, + "papermill": { + "duration": 0.042855, + "end_time": "2024-04-18T19:25:17.057952", + "exception": false, + "start_time": "2024-04-18T19:25:17.015097", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "text/plain": [ + "(60000, 150)" + ] + }, + "execution_count": 26, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "X.shape" + ] + }, + { + "cell_type": "code", + "execution_count": 27, + "id": "19dd24e2", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:25:17.196471Z", + "iopub.status.busy": "2024-04-18T19:25:17.196150Z", + "iopub.status.idle": "2024-04-18T19:26:39.839999Z", + "shell.execute_reply": "2024-04-18T19:26:39.839206Z" + }, + "papermill": { + "duration": 82.681197, + "end_time": "2024-04-18T19:26:39.842199", + "exception": false, + "start_time": "2024-04-18T19:25:17.161002", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[==================================================] 100.0% 128.1/128.1MB downloaded\n" + ] + } + ], + "source": [ + "import gensim.downloader as api\n", + "glove_gensim = api.load('glove-wiki-gigaword-100') #100 dimension" + ] + }, + { + "cell_type": "code", + "execution_count": 28, + "id": "2b1c7963", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:26:42.614378Z", + "iopub.status.busy": "2024-04-18T19:26:42.614023Z", + "iopub.status.idle": "2024-04-18T19:27:56.114464Z", + "shell.execute_reply": "2024-04-18T19:27:56.113667Z" + }, + "papermill": { + "duration": 74.891533, + "end_time": "2024-04-18T19:27:56.116702", + "exception": false, + "start_time": "2024-04-18T19:26:41.225169", + "status": "completed" + }, + "tags": [] + }, + "outputs": [], + "source": [ + "from gensim.models import KeyedVectors\n", + "\n", + "vector_size = 100\n", + "num_words = 30000\n", + "gensim_weight_matrix = np.zeros((num_words ,vector_size))\n", + "gensim_weight_matrix.shape\n", + "for word, index in tokenizer.word_index.items():\n", + " if index < num_words: # since index starts with zero \n", + " if word in glove_gensim.index_to_key:\n", + " gensim_weight_matrix[index] = glove_gensim[word]\n", + " else:\n", + " gensim_weight_matrix[index] = np.zeros(100)" + ] + }, + { + "cell_type": "code", + "execution_count": 29, + "id": "939746c8", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:28:01.538398Z", + "iopub.status.busy": "2024-04-18T19:28:01.538050Z", + "iopub.status.idle": "2024-04-18T19:28:01.545076Z", + "shell.execute_reply": "2024-04-18T19:28:01.544271Z" + }, + "papermill": { + "duration": 1.374891, + "end_time": "2024-04-18T19:28:01.546986", + "exception": false, + "start_time": "2024-04-18T19:28:00.172095", + "status": "completed" + }, + "tags": [] + }, + "outputs": [], + "source": [ + "import tensorflow\n", + "from tensorflow.keras.models import Sequential \n", + "from tensorflow.keras.layers import Dense, LSTM, Embedding,Bidirectional,SimpleRNN \n", + "\n", + "# from tensorflow.compat.v1.keras.layers import CuDNNRNN\n", + "from tensorflow.keras.layers import Dropout, Flatten\n", + "from tensorflow.keras.optimizers import Adam" + ] + }, + { + "cell_type": "code", + "execution_count": 30, + "id": "5b568d58", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:28:04.213565Z", + "iopub.status.busy": "2024-04-18T19:28:04.213216Z", + "iopub.status.idle": "2024-04-18T19:28:04.243641Z", + "shell.execute_reply": "2024-04-18T19:28:04.242873Z" + }, + "papermill": { + "duration": 1.392627, + "end_time": "2024-04-18T19:28:04.245845", + "exception": false, + "start_time": "2024-04-18T19:28:02.853218", + "status": "completed" + }, + "tags": [] + }, + "outputs": [], + "source": [ + "#Splitting the data into training and testing\n", + "from sklearn.model_selection import train_test_split\n", + "y=pd.get_dummies(df['class'])\n", + "X_train, X_test, y_train, y_test = train_test_split(X,y, test_size = 0.2, random_state = 123)" + ] + }, + { + "cell_type": "code", + "execution_count": 31, + "id": "aa1f0937", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-18T19:28:09.632226Z", + "iopub.status.busy": "2024-04-18T19:28:09.631877Z", + "iopub.status.idle": "2024-04-19T00:31:05.393353Z", + "shell.execute_reply": "2024-04-19T00:31:05.392432Z" + }, + "papermill": { + "duration": 18177.120726, + "end_time": "2024-04-19T00:31:05.395685", + "exception": false, + "start_time": "2024-04-18T19:28:08.274959", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Bi-LSTM-RNN Model Summary:\n", + "Model: \"sequential\"\n", + "_________________________________________________________________\n", + " Layer (type) Output Shape Param # \n", + "=================================================================\n", + " embedding (Embedding) (None, 150, 100) 3000000 \n", + " \n", + " dropout (Dropout) (None, 150, 100) 0 \n", + " \n", + " bidirectional (Bidirection (None, 150, 700) 1262800 \n", + " al) \n", + " \n", + " dropout_1 (Dropout) (None, 150, 700) 0 \n", + " \n", + " bidirectional_1 (Bidirecti (None, 150, 900) 4143600 \n", + " onal) \n", + " \n", + " dropout_2 (Dropout) (None, 150, 900) 0 \n", + " \n", + " bidirectional_2 (Bidirecti (None, 150, 1000) 5604000 \n", + " onal) \n", + " \n", + " dropout_3 (Dropout) (None, 150, 1000) 0 \n", + " \n", + " simple_rnn (SimpleRNN) (None, 150, 125) 140750 \n", + " \n", + " dropout_4 (Dropout) (None, 150, 125) 0 \n", + " \n", + " simple_rnn_1 (SimpleRNN) (None, 150) 41400 \n", + " \n", + " dense (Dense) (None, 2) 302 \n", + " \n", + "=================================================================\n", + "Total params: 14192852 (54.14 MB)\n", + "Trainable params: 11192852 (42.70 MB)\n", + "Non-trainable params: 3000000 (11.44 MB)\n", + "_________________________________________________________________\n", + "\n", + "Bi-LSTM Model Summary:\n", + "Model: \"sequential_1\"\n", + "_________________________________________________________________\n", + " Layer (type) Output Shape Param # \n", + "=================================================================\n", + " embedding_1 (Embedding) (None, 150, 100) 3000000 \n", + " \n", + " dropout_5 (Dropout) (None, 150, 100) 0 \n", + " \n", + " bidirectional_3 (Bidirecti (None, 150, 700) 1262800 \n", + " onal) \n", + " \n", + " dropout_6 (Dropout) (None, 150, 700) 0 \n", + " \n", + " bidirectional_4 (Bidirecti (None, 150, 900) 4143600 \n", + " onal) \n", + " \n", + " dropout_7 (Dropout) (None, 150, 900) 0 \n", + " \n", + " bidirectional_5 (Bidirecti (None, 1000) 5604000 \n", + " onal) \n", + " \n", + " dropout_8 (Dropout) (None, 1000) 0 \n", + " \n", + " flatten (Flatten) (None, 1000) 0 \n", + " \n", + " dense_1 (Dense) (None, 2) 2002 \n", + " \n", + "=================================================================\n", + "Total params: 14012402 (53.45 MB)\n", + "Trainable params: 11012402 (42.01 MB)\n", + "Non-trainable params: 3000000 (11.44 MB)\n", + "_________________________________________________________________\n", + "\n", + "LSTM Model Summary:\n", + "Model: \"sequential_2\"\n", + "_________________________________________________________________\n", + " Layer (type) Output Shape Param # \n", + "=================================================================\n", + " embedding_2 (Embedding) (None, 150, 100) 3000000 \n", + " \n", + " dropout_9 (Dropout) (None, 150, 100) 0 \n", + " \n", + " lstm_6 (LSTM) (None, 150, 350) 631400 \n", + " \n", + " dropout_10 (Dropout) (None, 150, 350) 0 \n", + " \n", + " lstm_7 (LSTM) (None, 150, 450) 1441800 \n", + " \n", + " dropout_11 (Dropout) (None, 150, 450) 0 \n", + " \n", + " lstm_8 (LSTM) (None, 500) 1902000 \n", + " \n", + " dropout_12 (Dropout) (None, 500) 0 \n", + " \n", + " flatten_1 (Flatten) (None, 500) 0 \n", + " \n", + " dense_2 (Dense) (None, 2) 1002 \n", + " \n", + "=================================================================\n", + "Total params: 6976202 (26.61 MB)\n", + "Trainable params: 3976202 (15.17 MB)\n", + "Non-trainable params: 3000000 (11.44 MB)\n", + "_________________________________________________________________\n", + "Training Bi-LSTM-RNN Model...\n", + "Epoch 1/50\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n", + "I0000 00:00:1713468507.130218 88 device_compiler.h:186] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "375/375 [==============================] - ETA: 0s - loss: 0.5687 - accuracy: 0.7175\n", + "Epoch 1: val_accuracy improved from -inf to 0.84058, saving model to ./best_model.h5\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/conda/lib/python3.10/site-packages/keras/src/engine/training.py:3103: UserWarning: You are saving your model as an HDF5 file via `model.save()`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')`.\n", + " saving_api.save_model(\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "375/375 [==============================] - 199s 498ms/step - loss: 0.5687 - accuracy: 0.7175 - val_loss: 0.4006 - val_accuracy: 0.8406 - lr: 0.0010\n", + "Epoch 2/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.3795 - accuracy: 0.8470\n", + "Epoch 2: val_accuracy improved from 0.84058 to 0.86292, saving model to ./best_model.h5\n", + "375/375 [==============================] - 184s 492ms/step - loss: 0.3795 - accuracy: 0.8470 - val_loss: 0.3460 - val_accuracy: 0.8629 - lr: 0.0010\n", + "Epoch 3/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.3161 - accuracy: 0.8731\n", + "Epoch 3: val_accuracy improved from 0.86292 to 0.87408, saving model to ./best_model.h5\n", + "375/375 [==============================] - 185s 492ms/step - loss: 0.3161 - accuracy: 0.8731 - val_loss: 0.2860 - val_accuracy: 0.8741 - lr: 0.0010\n", + "Epoch 4/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2740 - accuracy: 0.8921\n", + "Epoch 4: val_accuracy improved from 0.87408 to 0.89617, saving model to ./best_model.h5\n", + "375/375 [==============================] - 185s 493ms/step - loss: 0.2740 - accuracy: 0.8921 - val_loss: 0.2601 - val_accuracy: 0.8962 - lr: 0.0010\n", + "Epoch 5/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2486 - accuracy: 0.9030\n", + "Epoch 5: val_accuracy improved from 0.89617 to 0.90958, saving model to ./best_model.h5\n", + "375/375 [==============================] - 184s 492ms/step - loss: 0.2486 - accuracy: 0.9030 - val_loss: 0.2304 - val_accuracy: 0.9096 - lr: 0.0010\n", + "Epoch 6/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2324 - accuracy: 0.9099\n", + "Epoch 6: val_accuracy did not improve from 0.90958\n", + "375/375 [==============================] - 184s 490ms/step - loss: 0.2324 - accuracy: 0.9099 - val_loss: 0.2274 - val_accuracy: 0.9095 - lr: 0.0010\n", + "Epoch 7/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2259 - accuracy: 0.9122\n", + "Epoch 7: val_accuracy improved from 0.90958 to 0.91383, saving model to ./best_model.h5\n", + "375/375 [==============================] - 185s 492ms/step - loss: 0.2259 - accuracy: 0.9122 - val_loss: 0.2252 - val_accuracy: 0.9138 - lr: 0.0010\n", + "Epoch 8/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2163 - accuracy: 0.9168\n", + "Epoch 8: val_accuracy improved from 0.91383 to 0.91458, saving model to ./best_model.h5\n", + "375/375 [==============================] - 185s 493ms/step - loss: 0.2163 - accuracy: 0.9168 - val_loss: 0.2213 - val_accuracy: 0.9146 - lr: 0.0010\n", + "Epoch 9/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2037 - accuracy: 0.9213\n", + "Epoch 9: val_accuracy improved from 0.91458 to 0.91758, saving model to ./best_model.h5\n", + "375/375 [==============================] - 186s 496ms/step - loss: 0.2037 - accuracy: 0.9213 - val_loss: 0.2136 - val_accuracy: 0.9176 - lr: 0.0010\n", + "Epoch 10/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1971 - accuracy: 0.9222\n", + "Epoch 10: val_accuracy improved from 0.91758 to 0.91817, saving model to ./best_model.h5\n", + "375/375 [==============================] - 186s 495ms/step - loss: 0.1971 - accuracy: 0.9222 - val_loss: 0.2152 - val_accuracy: 0.9182 - lr: 0.0010\n", + "Epoch 11/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1876 - accuracy: 0.9269\n", + "Epoch 11: val_accuracy improved from 0.91817 to 0.91858, saving model to ./best_model.h5\n", + "375/375 [==============================] - 186s 496ms/step - loss: 0.1876 - accuracy: 0.9269 - val_loss: 0.2152 - val_accuracy: 0.9186 - lr: 0.0010\n", + "Epoch 12/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1812 - accuracy: 0.9305\n", + "Epoch 12: val_accuracy improved from 0.91858 to 0.91867, saving model to ./best_model.h5\n", + "375/375 [==============================] - 186s 497ms/step - loss: 0.1812 - accuracy: 0.9305 - val_loss: 0.2129 - val_accuracy: 0.9187 - lr: 0.0010\n", + "Epoch 13/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1732 - accuracy: 0.9342\n", + "Epoch 13: val_accuracy did not improve from 0.91867\n", + "375/375 [==============================] - 186s 496ms/step - loss: 0.1732 - accuracy: 0.9342 - val_loss: 0.2132 - val_accuracy: 0.9186 - lr: 0.0010\n", + "Epoch 14/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1718 - accuracy: 0.9347\n", + "Epoch 14: val_accuracy did not improve from 0.91867\n", + "375/375 [==============================] - 186s 496ms/step - loss: 0.1718 - accuracy: 0.9347 - val_loss: 0.2313 - val_accuracy: 0.9107 - lr: 0.0010\n", + "Epoch 15/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1558 - accuracy: 0.9411\n", + "Epoch 15: val_accuracy improved from 0.91867 to 0.92167, saving model to ./best_model.h5\n", + "375/375 [==============================] - 186s 496ms/step - loss: 0.1558 - accuracy: 0.9411 - val_loss: 0.2142 - val_accuracy: 0.9217 - lr: 0.0010\n", + "Epoch 16/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1541 - accuracy: 0.9402\n", + "Epoch 16: val_accuracy did not improve from 0.92167\n", + "375/375 [==============================] - 186s 497ms/step - loss: 0.1541 - accuracy: 0.9402 - val_loss: 0.2154 - val_accuracy: 0.9192 - lr: 0.0010\n", + "Epoch 17/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1457 - accuracy: 0.9451\n", + "Epoch 17: val_accuracy did not improve from 0.92167\n", + "375/375 [==============================] - 186s 496ms/step - loss: 0.1457 - accuracy: 0.9451 - val_loss: 0.2240 - val_accuracy: 0.9197 - lr: 0.0010\n", + "Epoch 18/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1203 - accuracy: 0.9536\n", + "Epoch 18: val_accuracy did not improve from 0.92167\n", + "375/375 [==============================] - 186s 496ms/step - loss: 0.1203 - accuracy: 0.9536 - val_loss: 0.2299 - val_accuracy: 0.9215 - lr: 2.0000e-04\n", + "Epoch 19/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1133 - accuracy: 0.9574\n", + "Epoch 19: val_accuracy improved from 0.92167 to 0.92267, saving model to ./best_model.h5\n", + "375/375 [==============================] - 186s 497ms/step - loss: 0.1133 - accuracy: 0.9574 - val_loss: 0.2349 - val_accuracy: 0.9227 - lr: 2.0000e-04\n", + "Epoch 20/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1061 - accuracy: 0.9595\n", + "Epoch 20: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 494ms/step - loss: 0.1061 - accuracy: 0.9595 - val_loss: 0.2536 - val_accuracy: 0.9218 - lr: 2.0000e-04\n", + "Epoch 21/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1049 - accuracy: 0.9601\n", + "Epoch 21: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 186s 497ms/step - loss: 0.1049 - accuracy: 0.9601 - val_loss: 0.2494 - val_accuracy: 0.9191 - lr: 2.0000e-04\n", + "Epoch 22/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1022 - accuracy: 0.9612\n", + "Epoch 22: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 194s 517ms/step - loss: 0.1022 - accuracy: 0.9612 - val_loss: 0.2450 - val_accuracy: 0.9212 - lr: 2.0000e-04\n", + "Epoch 23/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0974 - accuracy: 0.9633\n", + "Epoch 23: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 495ms/step - loss: 0.0974 - accuracy: 0.9633 - val_loss: 0.2483 - val_accuracy: 0.9215 - lr: 1.0000e-04\n", + "Epoch 24/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0941 - accuracy: 0.9642\n", + "Epoch 24: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 495ms/step - loss: 0.0941 - accuracy: 0.9642 - val_loss: 0.2524 - val_accuracy: 0.9213 - lr: 1.0000e-04\n", + "Epoch 25/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0926 - accuracy: 0.9644\n", + "Epoch 25: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 495ms/step - loss: 0.0926 - accuracy: 0.9644 - val_loss: 0.2558 - val_accuracy: 0.9190 - lr: 1.0000e-04\n", + "Epoch 26/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0892 - accuracy: 0.9662\n", + "Epoch 26: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 186s 495ms/step - loss: 0.0892 - accuracy: 0.9662 - val_loss: 0.2616 - val_accuracy: 0.9196 - lr: 1.0000e-04\n", + "Epoch 27/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0863 - accuracy: 0.9675\n", + "Epoch 27: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 493ms/step - loss: 0.0863 - accuracy: 0.9675 - val_loss: 0.2708 - val_accuracy: 0.9193 - lr: 1.0000e-04\n", + "Epoch 28/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0869 - accuracy: 0.9674\n", + "Epoch 28: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 494ms/step - loss: 0.0869 - accuracy: 0.9674 - val_loss: 0.2605 - val_accuracy: 0.9200 - lr: 1.0000e-04\n", + "Epoch 29/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0863 - accuracy: 0.9672\n", + "Epoch 29: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 492ms/step - loss: 0.0863 - accuracy: 0.9672 - val_loss: 0.2644 - val_accuracy: 0.9200 - lr: 1.0000e-04\n", + "Epoch 30/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0845 - accuracy: 0.9680\n", + "Epoch 30: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 493ms/step - loss: 0.0845 - accuracy: 0.9680 - val_loss: 0.2823 - val_accuracy: 0.9187 - lr: 1.0000e-04\n", + "Epoch 31/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0841 - accuracy: 0.9685\n", + "Epoch 31: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 494ms/step - loss: 0.0841 - accuracy: 0.9685 - val_loss: 0.2698 - val_accuracy: 0.9195 - lr: 1.0000e-04\n", + "Epoch 32/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0809 - accuracy: 0.9688\n", + "Epoch 32: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 493ms/step - loss: 0.0809 - accuracy: 0.9688 - val_loss: 0.2662 - val_accuracy: 0.9192 - lr: 1.0000e-04\n", + "Epoch 33/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0800 - accuracy: 0.9694\n", + "Epoch 33: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 493ms/step - loss: 0.0800 - accuracy: 0.9694 - val_loss: 0.2762 - val_accuracy: 0.9192 - lr: 1.0000e-04\n", + "Epoch 34/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0783 - accuracy: 0.9704\n", + "Epoch 34: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 493ms/step - loss: 0.0783 - accuracy: 0.9704 - val_loss: 0.2756 - val_accuracy: 0.9168 - lr: 1.0000e-04\n", + "Epoch 35/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0788 - accuracy: 0.9714\n", + "Epoch 35: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 493ms/step - loss: 0.0788 - accuracy: 0.9714 - val_loss: 0.2768 - val_accuracy: 0.9193 - lr: 1.0000e-04\n", + "Epoch 36/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0750 - accuracy: 0.9721\n", + "Epoch 36: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 493ms/step - loss: 0.0750 - accuracy: 0.9721 - val_loss: 0.2848 - val_accuracy: 0.9190 - lr: 1.0000e-04\n", + "Epoch 37/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0744 - accuracy: 0.9722\n", + "Epoch 37: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 493ms/step - loss: 0.0744 - accuracy: 0.9722 - val_loss: 0.2767 - val_accuracy: 0.9187 - lr: 1.0000e-04\n", + "Epoch 38/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0716 - accuracy: 0.9732\n", + "Epoch 38: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 493ms/step - loss: 0.0716 - accuracy: 0.9732 - val_loss: 0.2942 - val_accuracy: 0.9164 - lr: 1.0000e-04\n", + "Epoch 39/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0723 - accuracy: 0.9716\n", + "Epoch 39: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 494ms/step - loss: 0.0723 - accuracy: 0.9716 - val_loss: 0.2952 - val_accuracy: 0.9180 - lr: 1.0000e-04\n", + "Epoch 40/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0731 - accuracy: 0.9718\n", + "Epoch 40: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 493ms/step - loss: 0.0731 - accuracy: 0.9718 - val_loss: 0.2887 - val_accuracy: 0.9176 - lr: 1.0000e-04\n", + "Epoch 41/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0727 - accuracy: 0.9727\n", + "Epoch 41: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 493ms/step - loss: 0.0727 - accuracy: 0.9727 - val_loss: 0.2930 - val_accuracy: 0.9177 - lr: 1.0000e-04\n", + "Epoch 42/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0691 - accuracy: 0.9733\n", + "Epoch 42: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 493ms/step - loss: 0.0691 - accuracy: 0.9733 - val_loss: 0.2954 - val_accuracy: 0.9168 - lr: 1.0000e-04\n", + "Epoch 43/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0676 - accuracy: 0.9745\n", + "Epoch 43: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 493ms/step - loss: 0.0676 - accuracy: 0.9745 - val_loss: 0.2997 - val_accuracy: 0.9172 - lr: 1.0000e-04\n", + "Epoch 44/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0678 - accuracy: 0.9736\n", + "Epoch 44: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 186s 496ms/step - loss: 0.0678 - accuracy: 0.9736 - val_loss: 0.3102 - val_accuracy: 0.9162 - lr: 1.0000e-04\n", + "Epoch 45/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0679 - accuracy: 0.9743\n", + "Epoch 45: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 495ms/step - loss: 0.0679 - accuracy: 0.9743 - val_loss: 0.3179 - val_accuracy: 0.9133 - lr: 1.0000e-04\n", + "Epoch 46/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0653 - accuracy: 0.9755\n", + "Epoch 46: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 493ms/step - loss: 0.0653 - accuracy: 0.9755 - val_loss: 0.3089 - val_accuracy: 0.9154 - lr: 1.0000e-04\n", + "Epoch 47/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0636 - accuracy: 0.9758\n", + "Epoch 47: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 185s 493ms/step - loss: 0.0636 - accuracy: 0.9758 - val_loss: 0.2976 - val_accuracy: 0.9141 - lr: 1.0000e-04\n", + "Epoch 48/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0626 - accuracy: 0.9756\n", + "Epoch 48: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 186s 496ms/step - loss: 0.0626 - accuracy: 0.9756 - val_loss: 0.2932 - val_accuracy: 0.9146 - lr: 1.0000e-04\n", + "Epoch 49/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0610 - accuracy: 0.9763\n", + "Epoch 49: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 186s 496ms/step - loss: 0.0610 - accuracy: 0.9763 - val_loss: 0.3147 - val_accuracy: 0.9161 - lr: 1.0000e-04\n", + "Epoch 50/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0596 - accuracy: 0.9772\n", + "Epoch 50: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 186s 495ms/step - loss: 0.0596 - accuracy: 0.9772 - val_loss: 0.3149 - val_accuracy: 0.9153 - lr: 1.0000e-04\n", + "375/375 [==============================] - 25s 66ms/step - loss: 0.3149 - accuracy: 0.9153\n", + "Training Bi-LSTM Model...\n", + "Epoch 1/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.3592 - accuracy: 0.8496\n", + "Epoch 1: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 133s 334ms/step - loss: 0.3592 - accuracy: 0.8496 - val_loss: 0.3094 - val_accuracy: 0.8757 - lr: 0.0010\n", + "Epoch 2/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.3059 - accuracy: 0.8784\n", + "Epoch 2: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.3059 - accuracy: 0.8784 - val_loss: 0.2666 - val_accuracy: 0.8937 - lr: 0.0010\n", + "Epoch 3/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2542 - accuracy: 0.8999\n", + "Epoch 3: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.2542 - accuracy: 0.8999 - val_loss: 0.2492 - val_accuracy: 0.9001 - lr: 0.0010\n", + "Epoch 4/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2350 - accuracy: 0.9088\n", + "Epoch 4: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.2350 - accuracy: 0.9088 - val_loss: 0.2200 - val_accuracy: 0.9151 - lr: 0.0010\n", + "Epoch 5/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2219 - accuracy: 0.9134\n", + "Epoch 5: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.2219 - accuracy: 0.9134 - val_loss: 0.2186 - val_accuracy: 0.9153 - lr: 0.0010\n", + "Epoch 6/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2134 - accuracy: 0.9155\n", + "Epoch 6: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.2134 - accuracy: 0.9155 - val_loss: 0.2295 - val_accuracy: 0.9151 - lr: 0.0010\n", + "Epoch 7/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1998 - accuracy: 0.9224\n", + "Epoch 7: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.1998 - accuracy: 0.9224 - val_loss: 0.2191 - val_accuracy: 0.9171 - lr: 0.0010\n", + "Epoch 8/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1878 - accuracy: 0.9257\n", + "Epoch 8: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.1878 - accuracy: 0.9257 - val_loss: 0.2081 - val_accuracy: 0.9195 - lr: 0.0010\n", + "Epoch 9/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1756 - accuracy: 0.9304\n", + "Epoch 9: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.1756 - accuracy: 0.9304 - val_loss: 0.2133 - val_accuracy: 0.9197 - lr: 0.0010\n", + "Epoch 10/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1683 - accuracy: 0.9340\n", + "Epoch 10: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.1683 - accuracy: 0.9340 - val_loss: 0.2094 - val_accuracy: 0.9180 - lr: 0.0010\n", + "Epoch 11/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1580 - accuracy: 0.9379\n", + "Epoch 11: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.1580 - accuracy: 0.9379 - val_loss: 0.2161 - val_accuracy: 0.9212 - lr: 0.0010\n", + "Epoch 12/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1489 - accuracy: 0.9416\n", + "Epoch 12: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.1489 - accuracy: 0.9416 - val_loss: 0.2399 - val_accuracy: 0.9156 - lr: 0.0010\n", + "Epoch 13/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1374 - accuracy: 0.9460\n", + "Epoch 13: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.1374 - accuracy: 0.9460 - val_loss: 0.2463 - val_accuracy: 0.9221 - lr: 0.0010\n", + "Epoch 14/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1071 - accuracy: 0.9576\n", + "Epoch 14: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.1071 - accuracy: 0.9576 - val_loss: 0.2564 - val_accuracy: 0.9195 - lr: 2.0000e-04\n", + "Epoch 15/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1008 - accuracy: 0.9601\n", + "Epoch 15: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.1008 - accuracy: 0.9601 - val_loss: 0.2714 - val_accuracy: 0.9160 - lr: 2.0000e-04\n", + "Epoch 16/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0938 - accuracy: 0.9622\n", + "Epoch 16: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0938 - accuracy: 0.9622 - val_loss: 0.2645 - val_accuracy: 0.9203 - lr: 2.0000e-04\n", + "Epoch 17/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0872 - accuracy: 0.9653\n", + "Epoch 17: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0872 - accuracy: 0.9653 - val_loss: 0.2685 - val_accuracy: 0.9185 - lr: 2.0000e-04\n", + "Epoch 18/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0810 - accuracy: 0.9669\n", + "Epoch 18: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0810 - accuracy: 0.9669 - val_loss: 0.2875 - val_accuracy: 0.9183 - lr: 2.0000e-04\n", + "Epoch 19/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0750 - accuracy: 0.9695\n", + "Epoch 19: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0750 - accuracy: 0.9695 - val_loss: 0.3081 - val_accuracy: 0.9153 - lr: 1.0000e-04\n", + "Epoch 20/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0722 - accuracy: 0.9712\n", + "Epoch 20: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0722 - accuracy: 0.9712 - val_loss: 0.3160 - val_accuracy: 0.9202 - lr: 1.0000e-04\n", + "Epoch 21/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0713 - accuracy: 0.9711\n", + "Epoch 21: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0713 - accuracy: 0.9711 - val_loss: 0.3289 - val_accuracy: 0.9152 - lr: 1.0000e-04\n", + "Epoch 22/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0666 - accuracy: 0.9726\n", + "Epoch 22: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0666 - accuracy: 0.9726 - val_loss: 0.3335 - val_accuracy: 0.9158 - lr: 1.0000e-04\n", + "Epoch 23/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0652 - accuracy: 0.9737\n", + "Epoch 23: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0652 - accuracy: 0.9737 - val_loss: 0.3318 - val_accuracy: 0.9164 - lr: 1.0000e-04\n", + "Epoch 24/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0626 - accuracy: 0.9746\n", + "Epoch 24: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0626 - accuracy: 0.9746 - val_loss: 0.3714 - val_accuracy: 0.9132 - lr: 1.0000e-04\n", + "Epoch 25/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0607 - accuracy: 0.9750\n", + "Epoch 25: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0607 - accuracy: 0.9750 - val_loss: 0.3439 - val_accuracy: 0.9173 - lr: 1.0000e-04\n", + "Epoch 26/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0592 - accuracy: 0.9763\n", + "Epoch 26: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0592 - accuracy: 0.9763 - val_loss: 0.3428 - val_accuracy: 0.9143 - lr: 1.0000e-04\n", + "Epoch 27/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0553 - accuracy: 0.9774\n", + "Epoch 27: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0553 - accuracy: 0.9774 - val_loss: 0.3633 - val_accuracy: 0.9185 - lr: 1.0000e-04\n", + "Epoch 28/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0560 - accuracy: 0.9773\n", + "Epoch 28: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0560 - accuracy: 0.9773 - val_loss: 0.3575 - val_accuracy: 0.9177 - lr: 1.0000e-04\n", + "Epoch 29/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0550 - accuracy: 0.9771\n", + "Epoch 29: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0550 - accuracy: 0.9771 - val_loss: 0.3566 - val_accuracy: 0.9183 - lr: 1.0000e-04\n", + "Epoch 30/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0524 - accuracy: 0.9787\n", + "Epoch 30: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0524 - accuracy: 0.9787 - val_loss: 0.3736 - val_accuracy: 0.9182 - lr: 1.0000e-04\n", + "Epoch 31/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0511 - accuracy: 0.9798\n", + "Epoch 31: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0511 - accuracy: 0.9798 - val_loss: 0.3750 - val_accuracy: 0.9171 - lr: 1.0000e-04\n", + "Epoch 32/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0507 - accuracy: 0.9798\n", + "Epoch 32: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0507 - accuracy: 0.9798 - val_loss: 0.3921 - val_accuracy: 0.9175 - lr: 1.0000e-04\n", + "Epoch 33/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0495 - accuracy: 0.9800\n", + "Epoch 33: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0495 - accuracy: 0.9800 - val_loss: 0.3817 - val_accuracy: 0.9156 - lr: 1.0000e-04\n", + "Epoch 34/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0469 - accuracy: 0.9815\n", + "Epoch 34: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0469 - accuracy: 0.9815 - val_loss: 0.3919 - val_accuracy: 0.9171 - lr: 1.0000e-04\n", + "Epoch 35/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0462 - accuracy: 0.9809\n", + "Epoch 35: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0462 - accuracy: 0.9809 - val_loss: 0.3956 - val_accuracy: 0.9158 - lr: 1.0000e-04\n", + "Epoch 36/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0458 - accuracy: 0.9819\n", + "Epoch 36: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0458 - accuracy: 0.9819 - val_loss: 0.3919 - val_accuracy: 0.9145 - lr: 1.0000e-04\n", + "Epoch 37/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0442 - accuracy: 0.9820\n", + "Epoch 37: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0442 - accuracy: 0.9820 - val_loss: 0.4040 - val_accuracy: 0.9164 - lr: 1.0000e-04\n", + "Epoch 38/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0429 - accuracy: 0.9832\n", + "Epoch 38: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0429 - accuracy: 0.9832 - val_loss: 0.4113 - val_accuracy: 0.9128 - lr: 1.0000e-04\n", + "Epoch 39/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0400 - accuracy: 0.9840\n", + "Epoch 39: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0400 - accuracy: 0.9840 - val_loss: 0.4239 - val_accuracy: 0.9112 - lr: 1.0000e-04\n", + "Epoch 40/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0390 - accuracy: 0.9845\n", + "Epoch 40: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0390 - accuracy: 0.9845 - val_loss: 0.4333 - val_accuracy: 0.9175 - lr: 1.0000e-04\n", + "Epoch 41/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0385 - accuracy: 0.9850\n", + "Epoch 41: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0385 - accuracy: 0.9850 - val_loss: 0.4345 - val_accuracy: 0.9124 - lr: 1.0000e-04\n", + "Epoch 42/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0384 - accuracy: 0.9844\n", + "Epoch 42: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0384 - accuracy: 0.9844 - val_loss: 0.4329 - val_accuracy: 0.9132 - lr: 1.0000e-04\n", + "Epoch 43/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0370 - accuracy: 0.9852\n", + "Epoch 43: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0370 - accuracy: 0.9852 - val_loss: 0.4459 - val_accuracy: 0.9143 - lr: 1.0000e-04\n", + "Epoch 44/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0366 - accuracy: 0.9848\n", + "Epoch 44: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0366 - accuracy: 0.9848 - val_loss: 0.4312 - val_accuracy: 0.9142 - lr: 1.0000e-04\n", + "Epoch 45/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0357 - accuracy: 0.9859\n", + "Epoch 45: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0357 - accuracy: 0.9859 - val_loss: 0.4393 - val_accuracy: 0.9159 - lr: 1.0000e-04\n", + "Epoch 46/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0356 - accuracy: 0.9859\n", + "Epoch 46: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0356 - accuracy: 0.9859 - val_loss: 0.4298 - val_accuracy: 0.9156 - lr: 1.0000e-04\n", + "Epoch 47/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0337 - accuracy: 0.9867\n", + "Epoch 47: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0337 - accuracy: 0.9867 - val_loss: 0.4391 - val_accuracy: 0.9153 - lr: 1.0000e-04\n", + "Epoch 48/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0321 - accuracy: 0.9878\n", + "Epoch 48: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0321 - accuracy: 0.9878 - val_loss: 0.4486 - val_accuracy: 0.9143 - lr: 1.0000e-04\n", + "Epoch 49/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0331 - accuracy: 0.9874\n", + "Epoch 49: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0331 - accuracy: 0.9874 - val_loss: 0.4201 - val_accuracy: 0.9160 - lr: 1.0000e-04\n", + "Epoch 50/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0307 - accuracy: 0.9878\n", + "Epoch 50: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 123s 329ms/step - loss: 0.0307 - accuracy: 0.9878 - val_loss: 0.4477 - val_accuracy: 0.9171 - lr: 1.0000e-04\n", + "375/375 [==============================] - 19s 51ms/step - loss: 0.4477 - accuracy: 0.9171\n", + "Training LSTM Model...\n", + "Epoch 1/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.5045 - accuracy: 0.7572\n", + "Epoch 1: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 57s 140ms/step - loss: 0.5045 - accuracy: 0.7572 - val_loss: 0.3502 - val_accuracy: 0.8615 - lr: 0.0010\n", + "Epoch 2/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.3539 - accuracy: 0.8521\n", + "Epoch 2: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.3539 - accuracy: 0.8521 - val_loss: 0.3051 - val_accuracy: 0.8782 - lr: 0.0010\n", + "Epoch 3/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2993 - accuracy: 0.8781\n", + "Epoch 3: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.2993 - accuracy: 0.8781 - val_loss: 0.2849 - val_accuracy: 0.8737 - lr: 0.0010\n", + "Epoch 4/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2726 - accuracy: 0.8911\n", + "Epoch 4: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.2726 - accuracy: 0.8911 - val_loss: 0.2494 - val_accuracy: 0.9020 - lr: 0.0010\n", + "Epoch 5/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2575 - accuracy: 0.8993\n", + "Epoch 5: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.2575 - accuracy: 0.8993 - val_loss: 0.2449 - val_accuracy: 0.9024 - lr: 0.0010\n", + "Epoch 6/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2436 - accuracy: 0.9040\n", + "Epoch 6: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.2436 - accuracy: 0.9040 - val_loss: 0.2372 - val_accuracy: 0.9081 - lr: 0.0010\n", + "Epoch 7/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2322 - accuracy: 0.9085\n", + "Epoch 7: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.2322 - accuracy: 0.9085 - val_loss: 0.2270 - val_accuracy: 0.9135 - lr: 0.0010\n", + "Epoch 8/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2237 - accuracy: 0.9114\n", + "Epoch 8: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.2237 - accuracy: 0.9114 - val_loss: 0.2265 - val_accuracy: 0.9129 - lr: 0.0010\n", + "Epoch 9/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2167 - accuracy: 0.9149\n", + "Epoch 9: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.2167 - accuracy: 0.9149 - val_loss: 0.2313 - val_accuracy: 0.9107 - lr: 0.0010\n", + "Epoch 10/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2108 - accuracy: 0.9175\n", + "Epoch 10: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.2108 - accuracy: 0.9175 - val_loss: 0.2160 - val_accuracy: 0.9177 - lr: 0.0010\n", + "Epoch 11/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.2025 - accuracy: 0.9218\n", + "Epoch 11: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.2025 - accuracy: 0.9218 - val_loss: 0.2094 - val_accuracy: 0.9207 - lr: 0.0010\n", + "Epoch 12/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1956 - accuracy: 0.9241\n", + "Epoch 12: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.1956 - accuracy: 0.9241 - val_loss: 0.2277 - val_accuracy: 0.9154 - lr: 0.0010\n", + "Epoch 13/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1939 - accuracy: 0.9248\n", + "Epoch 13: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.1939 - accuracy: 0.9248 - val_loss: 0.2089 - val_accuracy: 0.9218 - lr: 0.0010\n", + "Epoch 14/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1835 - accuracy: 0.9287\n", + "Epoch 14: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.1835 - accuracy: 0.9287 - val_loss: 0.2106 - val_accuracy: 0.9222 - lr: 0.0010\n", + "Epoch 15/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1742 - accuracy: 0.9311\n", + "Epoch 15: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 52s 138ms/step - loss: 0.1742 - accuracy: 0.9311 - val_loss: 0.2114 - val_accuracy: 0.9224 - lr: 0.0010\n", + "Epoch 16/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1694 - accuracy: 0.9345\n", + "Epoch 16: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.1694 - accuracy: 0.9345 - val_loss: 0.2317 - val_accuracy: 0.9172 - lr: 0.0010\n", + "Epoch 17/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1615 - accuracy: 0.9365\n", + "Epoch 17: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.1615 - accuracy: 0.9365 - val_loss: 0.2093 - val_accuracy: 0.9179 - lr: 0.0010\n", + "Epoch 18/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1550 - accuracy: 0.9399\n", + "Epoch 18: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.1550 - accuracy: 0.9399 - val_loss: 0.2232 - val_accuracy: 0.9155 - lr: 0.0010\n", + "Epoch 19/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1313 - accuracy: 0.9490\n", + "Epoch 19: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.1313 - accuracy: 0.9490 - val_loss: 0.2244 - val_accuracy: 0.9193 - lr: 2.0000e-04\n", + "Epoch 20/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1269 - accuracy: 0.9507\n", + "Epoch 20: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.1269 - accuracy: 0.9507 - val_loss: 0.2293 - val_accuracy: 0.9214 - lr: 2.0000e-04\n", + "Epoch 21/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1237 - accuracy: 0.9512\n", + "Epoch 21: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.1237 - accuracy: 0.9512 - val_loss: 0.2304 - val_accuracy: 0.9186 - lr: 2.0000e-04\n", + "Epoch 22/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1194 - accuracy: 0.9536\n", + "Epoch 22: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.1194 - accuracy: 0.9536 - val_loss: 0.2287 - val_accuracy: 0.9211 - lr: 2.0000e-04\n", + "Epoch 23/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1147 - accuracy: 0.9556\n", + "Epoch 23: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.1147 - accuracy: 0.9556 - val_loss: 0.2499 - val_accuracy: 0.9166 - lr: 2.0000e-04\n", + "Epoch 24/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1085 - accuracy: 0.9586\n", + "Epoch 24: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.1085 - accuracy: 0.9586 - val_loss: 0.2538 - val_accuracy: 0.9170 - lr: 1.0000e-04\n", + "Epoch 25/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1100 - accuracy: 0.9577\n", + "Epoch 25: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.1100 - accuracy: 0.9577 - val_loss: 0.2610 - val_accuracy: 0.9136 - lr: 1.0000e-04\n", + "Epoch 26/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1079 - accuracy: 0.9585\n", + "Epoch 26: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.1079 - accuracy: 0.9585 - val_loss: 0.2455 - val_accuracy: 0.9188 - lr: 1.0000e-04\n", + "Epoch 27/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1048 - accuracy: 0.9590\n", + "Epoch 27: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.1048 - accuracy: 0.9590 - val_loss: 0.2438 - val_accuracy: 0.9175 - lr: 1.0000e-04\n", + "Epoch 28/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1042 - accuracy: 0.9598\n", + "Epoch 28: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.1042 - accuracy: 0.9598 - val_loss: 0.2525 - val_accuracy: 0.9189 - lr: 1.0000e-04\n", + "Epoch 29/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.1019 - accuracy: 0.9606\n", + "Epoch 29: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.1019 - accuracy: 0.9606 - val_loss: 0.2612 - val_accuracy: 0.9171 - lr: 1.0000e-04\n", + "Epoch 30/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0978 - accuracy: 0.9632\n", + "Epoch 30: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0978 - accuracy: 0.9632 - val_loss: 0.2711 - val_accuracy: 0.9133 - lr: 1.0000e-04\n", + "Epoch 31/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0980 - accuracy: 0.9630\n", + "Epoch 31: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0980 - accuracy: 0.9630 - val_loss: 0.2767 - val_accuracy: 0.9147 - lr: 1.0000e-04\n", + "Epoch 32/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0967 - accuracy: 0.9634\n", + "Epoch 32: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0967 - accuracy: 0.9634 - val_loss: 0.2499 - val_accuracy: 0.9178 - lr: 1.0000e-04\n", + "Epoch 33/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0942 - accuracy: 0.9633\n", + "Epoch 33: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0942 - accuracy: 0.9633 - val_loss: 0.2586 - val_accuracy: 0.9203 - lr: 1.0000e-04\n", + "Epoch 34/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0923 - accuracy: 0.9645\n", + "Epoch 34: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0923 - accuracy: 0.9645 - val_loss: 0.2706 - val_accuracy: 0.9164 - lr: 1.0000e-04\n", + "Epoch 35/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0923 - accuracy: 0.9647\n", + "Epoch 35: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0923 - accuracy: 0.9647 - val_loss: 0.2815 - val_accuracy: 0.9154 - lr: 1.0000e-04\n", + "Epoch 36/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0920 - accuracy: 0.9654\n", + "Epoch 36: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0920 - accuracy: 0.9654 - val_loss: 0.2606 - val_accuracy: 0.9182 - lr: 1.0000e-04\n", + "Epoch 37/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0886 - accuracy: 0.9660\n", + "Epoch 37: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0886 - accuracy: 0.9660 - val_loss: 0.2832 - val_accuracy: 0.9181 - lr: 1.0000e-04\n", + "Epoch 38/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0866 - accuracy: 0.9671\n", + "Epoch 38: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0866 - accuracy: 0.9671 - val_loss: 0.2707 - val_accuracy: 0.9193 - lr: 1.0000e-04\n", + "Epoch 39/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0881 - accuracy: 0.9653\n", + "Epoch 39: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0881 - accuracy: 0.9653 - val_loss: 0.2676 - val_accuracy: 0.9152 - lr: 1.0000e-04\n", + "Epoch 40/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0854 - accuracy: 0.9673\n", + "Epoch 40: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0854 - accuracy: 0.9673 - val_loss: 0.2905 - val_accuracy: 0.9170 - lr: 1.0000e-04\n", + "Epoch 41/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0825 - accuracy: 0.9672\n", + "Epoch 41: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0825 - accuracy: 0.9672 - val_loss: 0.2840 - val_accuracy: 0.9153 - lr: 1.0000e-04\n", + "Epoch 42/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0803 - accuracy: 0.9686\n", + "Epoch 42: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0803 - accuracy: 0.9686 - val_loss: 0.2869 - val_accuracy: 0.9184 - lr: 1.0000e-04\n", + "Epoch 43/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0800 - accuracy: 0.9699\n", + "Epoch 43: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0800 - accuracy: 0.9699 - val_loss: 0.2823 - val_accuracy: 0.9153 - lr: 1.0000e-04\n", + "Epoch 44/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0814 - accuracy: 0.9687\n", + "Epoch 44: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0814 - accuracy: 0.9687 - val_loss: 0.2901 - val_accuracy: 0.9162 - lr: 1.0000e-04\n", + "Epoch 45/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0768 - accuracy: 0.9696\n", + "Epoch 45: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0768 - accuracy: 0.9696 - val_loss: 0.3085 - val_accuracy: 0.9141 - lr: 1.0000e-04\n", + "Epoch 46/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0747 - accuracy: 0.9710\n", + "Epoch 46: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0747 - accuracy: 0.9710 - val_loss: 0.3119 - val_accuracy: 0.9153 - lr: 1.0000e-04\n", + "Epoch 47/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0748 - accuracy: 0.9720\n", + "Epoch 47: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0748 - accuracy: 0.9720 - val_loss: 0.3084 - val_accuracy: 0.9151 - lr: 1.0000e-04\n", + "Epoch 48/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0725 - accuracy: 0.9718\n", + "Epoch 48: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0725 - accuracy: 0.9718 - val_loss: 0.3139 - val_accuracy: 0.9150 - lr: 1.0000e-04\n", + "Epoch 49/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0757 - accuracy: 0.9704\n", + "Epoch 49: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0757 - accuracy: 0.9704 - val_loss: 0.3141 - val_accuracy: 0.9161 - lr: 1.0000e-04\n", + "Epoch 50/50\n", + "375/375 [==============================] - ETA: 0s - loss: 0.0709 - accuracy: 0.9722\n", + "Epoch 50: val_accuracy did not improve from 0.92267\n", + "375/375 [==============================] - 51s 137ms/step - loss: 0.0709 - accuracy: 0.9722 - val_loss: 0.2981 - val_accuracy: 0.9158 - lr: 1.0000e-04\n", + "375/375 [==============================] - 9s 23ms/step - loss: 0.2981 - accuracy: 0.9158\n", + "Bi-LSTM-RNN Model Evaluation Loss: 0.31487181782722473\n", + "Bi-LSTM-RNN Model Evaluation Accuracy: 0.9152500033378601\n", + "Bi-LSTM Model Evaluation Loss: 0.4476720690727234\n", + "Bi-LSTM Model Evaluation Accuracy: 0.9170833230018616\n", + "LSTM Model Evaluation Loss: 0.29813352227211\n", + "LSTM Model Evaluation Accuracy: 0.9157500267028809\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + " Model Loss Accuracy\n", + "0 Bi-LSTM-RNN 0.314872 0.915250\n", + "1 Bi-LSTM 0.447672 0.917083\n", + "2 LSTM 0.298134 0.915750\n" + ] + } + ], + "source": [ + "from keras.callbacks import ReduceLROnPlateau, ModelCheckpoint\n", + "import matplotlib.pyplot as plt\n", + "\n", + "X_reshaped = X.reshape(X.shape[0], X.shape[1], 1)\n", + "\n", + "EMBEDDING_DIM = 100\n", + "\n", + "# Splitting the data into training and testing\n", + "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=123)\n", + "\n", + "# Define model architectures\n", + "def create_bilstm_rnn_model(input_shape, num_classes):\n", + " model = Sequential([\n", + " Embedding(input_dim=num_words, \n", + " output_dim=EMBEDDING_DIM, \n", + " input_length=X.shape[1], weights=[gensim_weight_matrix], \n", + " trainable=False),\n", + " Dropout(0.275),\n", + " Bidirectional(LSTM(units=350, return_sequences=True)),\n", + " Dropout(0.25),\n", + " Bidirectional(LSTM(units=450, return_sequences=True)),\n", + " Dropout(0.225),\n", + " Bidirectional(LSTM(units=500, return_sequences=True)),\n", + " Dropout(0.45),\n", + " SimpleRNN(125, return_sequences=True),\n", + " Dropout(0.425),\n", + " SimpleRNN(150, return_sequences=False),\n", + " Dense(2, activation='softmax')\n", + " ])\n", + " return model\n", + "\n", + "def create_bilstm_model(input_shape, num_classes):\n", + " model = Sequential([\n", + " Embedding(input_dim=num_words, \n", + " output_dim=EMBEDDING_DIM, \n", + " input_length=X.shape[1], weights=[gensim_weight_matrix], \n", + " trainable=False),\n", + " Dropout(0.275),\n", + " Bidirectional(LSTM(units=350, return_sequences=True)),\n", + " Dropout(0.25),\n", + " Bidirectional(LSTM(units=450, return_sequences=True)),\n", + " Dropout(0.225),\n", + " Bidirectional(LSTM(units=500, return_sequences=False)),\n", + " Dropout(0.45),\n", + " Flatten(),\n", + " Dense(2, activation='softmax')\n", + " ])\n", + " return model\n", + "\n", + "def create_lstm_model(input_shape, num_classes):\n", + " model = Sequential([\n", + " Embedding(input_dim=num_words, \n", + " output_dim=EMBEDDING_DIM, \n", + " input_length=X.shape[1], weights=[gensim_weight_matrix], \n", + " trainable=False),\n", + " Dropout(0.275),\n", + " LSTM(units=350, return_sequences=True),\n", + " Dropout(0.25),\n", + " LSTM(units=450, return_sequences=True),\n", + " Dropout(0.225),\n", + " LSTM(units=500, return_sequences=False),\n", + " Dropout(0.45),\n", + " Flatten(),\n", + " Dense(2, activation='softmax')\n", + " ])\n", + " return model\n", + "\n", + "# Compile models\n", + "num_classes = 2\n", + "bilstm_rnn_model = create_bilstm_rnn_model(X.shape[1], num_classes)\n", + "bilstm_model = create_bilstm_model(X.shape[1], num_classes)\n", + "lstm_model = create_lstm_model(X.shape[1], num_classes)\n", + "\n", + "# Print model summaries\n", + "print(\"Bi-LSTM-RNN Model Summary:\")\n", + "bilstm_rnn_model.summary()\n", + "\n", + "print(\"\\nBi-LSTM Model Summary:\")\n", + "bilstm_model.summary()\n", + "\n", + "print(\"\\nLSTM Model Summary:\")\n", + "lstm_model.summary()\n", + "\n", + "histories = {}\n", + "\n", + "for model in [bilstm_rnn_model, bilstm_model, lstm_model]:\n", + " model.compile(loss='categorical_crossentropy', optimizer=Adam(learning_rate=0.001), metrics=['accuracy'])\n", + "\n", + "# Define callbacks\n", + "reduce_lr = ReduceLROnPlateau(monitor='val_loss', factor=0.2, patience=5, min_lr=0.0001)\n", + "mc = ModelCheckpoint('./best_model.h5', monitor='val_accuracy', mode='max', verbose=1, save_best_only=True)\n", + "\n", + "# Train and evaluate models\n", + "models = {'Bi-LSTM-RNN': bilstm_rnn_model, 'Bi-LSTM': bilstm_model, 'LSTM': lstm_model}\n", + "evaluations = {}\n", + "\n", + "for name, model in models.items():\n", + " print(f\"Training {name} Model...\")\n", + " history = model.fit(X_train, y_train, \n", + " epochs = 50, batch_size = 128, \n", + " validation_data=(X_test, y_test),\n", + " verbose = 1, callbacks= [mc, reduce_lr])\n", + " histories[name] = history.history\n", + " evaluations[name] = model.evaluate(X_test, y_test)\n", + "\n", + "accuracy = {}\n", + "# Print evaluation results\n", + "for name, evaluation in evaluations.items():\n", + " print(f\"{name} Model Evaluation Loss:\", evaluation[0])\n", + " print(f\"{name} Model Evaluation Accuracy:\", evaluation[1])\n", + " accuracy[name] = evaluation[1]\n", + "\n", + "\n", + "# Plot training curves\n", + "for name, history in histories.items():\n", + " plt.plot(history['accuracy'], label=f'{name} Train Accuracy')\n", + " plt.plot(history['val_accuracy'], label=f'{name} Validation Accuracy')\n", + "\n", + "plt.xlabel('Epochs')\n", + "plt.ylabel('Accuracy')\n", + "plt.legend()\n", + "plt.show()\n", + "\n", + "# Model Comparison Table\n", + "results = {\n", + " 'Model': list(evaluations.keys()),\n", + " 'Loss': [evaluation[0] for evaluation in evaluations.values()],\n", + " 'Accuracy': [evaluation[1] for evaluation in evaluations.values()]\n", + "}\n", + "\n", + "results_df = pd.DataFrame(results)\n", + "print(results_df)\n" + ] + }, + { + "cell_type": "code", + "execution_count": 33, + "id": "ba7b4865", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-19T00:32:26.810696Z", + "iopub.status.busy": "2024-04-19T00:32:26.810048Z", + "iopub.status.idle": "2024-04-19T00:33:17.010231Z", + "shell.execute_reply": "2024-04-19T00:33:17.009398Z" + }, + "papermill": { + "duration": 56.335587, + "end_time": "2024-04-19T00:33:17.012037", + "exception": false, + "start_time": "2024-04-19T00:32:20.676450", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "375/375 [==============================] - 24s 63ms/step\n", + "375/375 [==============================] - 18s 48ms/step\n", + "375/375 [==============================] - 8s 21ms/step\n", + "Bi-LSTM-RNN Precision: 0.9131800893004796\n", + "Bi-LSTM-RNN Recall: 0.9181908879281676\n", + "Bi-LSTM-RNN F1 Score: 0.91567863361247\n", + "Bi-LSTM Precision: 0.9078498293515358\n", + "Bi-LSTM Recall: 0.9288327236448287\n", + "Bi-LSTM F1 Score: 0.9182214185912715\n", + "LSTM Precision: 0.9032726100274061\n", + "LSTM Recall: 0.9316594612570669\n", + "LSTM F1 Score: 0.9172464598510273\n" + ] + }, + { + "data": { + "text/html": [ + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
 ModelPrecisionRecallF1 Score
0Bi-LSTM-RNN0.9131800.9181910.915679
1Bi-LSTM0.9078500.9288330.918221
2LSTM0.9032730.9316590.917246
\n" + ], + "text/plain": [ + "" + ] + }, + "execution_count": 33, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "from sklearn.metrics import precision_score, recall_score, f1_score\n", + "\n", + "# Assuming y_true and y_pred are your true labels and predicted labels respectively\n", + "\n", + "# For Bi-LSTM-RNN model\n", + "y_pred_bilstm_rnn = models['Bi-LSTM-RNN'].predict(X_test)\n", + "y_pred_bilstm_rnn = np.argmax(y_pred_bilstm_rnn, axis=1)\n", + "precision_bilstm_rnn = precision_score(np.argmax(y_test.to_numpy(), axis=1), y_pred_bilstm_rnn)\n", + "recall_bilstm_rnn = recall_score(np.argmax(y_test.to_numpy(), axis=1), y_pred_bilstm_rnn)\n", + "f1_bilstm_rnn = f1_score(np.argmax(y_test.to_numpy(), axis=1), y_pred_bilstm_rnn)\n", + "\n", + "# For Bi-LSTM model\n", + "y_pred_bilstm = models['Bi-LSTM'].predict(X_test)\n", + "y_pred_bilstm = np.argmax(y_pred_bilstm, axis=1)\n", + "precision_bilstm = precision_score(np.argmax(y_test.to_numpy(), axis=1), y_pred_bilstm)\n", + "recall_bilstm = recall_score(np.argmax(y_test.to_numpy(), axis=1), y_pred_bilstm)\n", + "f1_bilstm = f1_score(np.argmax(y_test.to_numpy(), axis=1), y_pred_bilstm)\n", + "\n", + "# For LSTM model\n", + "y_pred_lstm = models['LSTM'].predict(X_test)\n", + "y_pred_lstm = np.argmax(y_pred_lstm, axis=1)\n", + "precision_lstm = precision_score(np.argmax(y_test.to_numpy(), axis=1), y_pred_lstm)\n", + "recall_lstm = recall_score(np.argmax(y_test.to_numpy(), axis=1), y_pred_lstm)\n", + "f1_lstm = f1_score(np.argmax(y_test.to_numpy(), axis=1), y_pred_lstm)\n", + "\n", + "print(\"Bi-LSTM-RNN Precision:\", precision_bilstm_rnn)\n", + "print(\"Bi-LSTM-RNN Recall:\", recall_bilstm_rnn)\n", + "print(\"Bi-LSTM-RNN F1 Score:\", f1_bilstm_rnn)\n", + "\n", + "print(\"Bi-LSTM Precision:\", precision_bilstm)\n", + "print(\"Bi-LSTM Recall:\", recall_bilstm)\n", + "print(\"Bi-LSTM F1 Score:\", f1_bilstm)\n", + "\n", + "print(\"LSTM Precision:\", precision_lstm)\n", + "print(\"LSTM Recall:\", recall_lstm)\n", + "print(\"LSTM F1 Score:\", f1_lstm)\n", + "\n", + "results = {\n", + " 'Model': ['Bi-LSTM-RNN', 'Bi-LSTM', 'LSTM'],\n", + " 'Precision': [precision_bilstm_rnn, precision_bilstm, precision_lstm],\n", + " 'Recall': [recall_bilstm_rnn, recall_bilstm, recall_lstm],\n", + " 'F1 Score': [f1_bilstm_rnn, f1_bilstm, f1_lstm]\n", + "}\n", + "\n", + "pd.set_option('display.max_rows', 300) # Adjust the number of maximum rows displayed\n", + "pd.set_option('display.max_columns', 300)\n", + "\n", + "results_df = pd.DataFrame(results)\n", + "\n", + "# Style the DataFrame for better visualization\n", + "styled_results_df = results_df.style.background_gradient(cmap='viridis', subset=['Precision', 'Recall', 'F1 Score'])\n", + "\n", + "# Increase font size\n", + "styled_results_df = styled_results_df.set_table_styles([{\n", + " 'selector': 'th',\n", + " 'props': [('font-size', '16px')]\n", + "}, {\n", + " 'selector': 'td',\n", + " 'props': [('font-size', '16px')]\n", + "}])\n", + "\n", + "# Increase cell width\n", + "styled_results_df = styled_results_df.set_properties(**{'width': '300px', 'text-align': 'center'})\n", + "\n", + "# Display the styled DataFrame\n", + "styled_results_df" + ] + }, + { + "cell_type": "code", + "execution_count": 34, + "id": "2d9da526", + "metadata": { + "execution": { + "iopub.execute_input": "2024-04-19T00:33:42.188409Z", + "iopub.status.busy": "2024-04-19T00:33:42.187551Z", + "iopub.status.idle": "2024-04-19T00:34:33.411775Z", + "shell.execute_reply": "2024-04-19T00:34:33.410888Z" + }, + "papermill": { + "duration": 57.539646, + "end_time": "2024-04-19T00:34:33.413754", + "exception": false, + "start_time": "2024-04-19T00:33:35.874108", + "status": "completed" + }, + "tags": [] + }, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "375/375 [==============================] - 24s 63ms/step\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "375/375 [==============================] - 18s 48ms/step\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "375/375 [==============================] - 8s 21ms/step\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from sklearn.metrics import confusion_matrix\n", + "import seaborn as sns\n", + "\n", + "# Plot accuracy vs epoch plot for each model separately\n", + "for name, history in histories.items():\n", + " plt.plot(history['accuracy'], label=f'{name} Train Accuracy')\n", + " plt.plot(history['val_accuracy'], label=f'{name} Validation Accuracy')\n", + "\n", + " plt.xlabel('Epochs')\n", + " plt.ylabel('Accuracy')\n", + " plt.title(f'Accuracy vs Epoch for {name}')\n", + " plt.legend()\n", + " plt.show()\n", + "\n", + "# Plot confusion matrix for each model\n", + "for name, model in models.items():\n", + " y_pred = model.predict(X_test)\n", + " y_pred = np.argmax(y_pred, axis=1)\n", + " cm = confusion_matrix(np.argmax(y_test.to_numpy(), axis=1), y_pred)\n", + " plt.figure(figsize=(8, 6))\n", + " sns.heatmap(cm, annot=True, cmap='Blues', fmt='g', xticklabels=['Negative', 'Positive'], yticklabels=['Negative', 'Positive'])\n", + " plt.xlabel('Predicted')\n", + " plt.ylabel('True')\n", + " plt.title(f'Confusion Matrix for {name} Model')\n", + " plt.show()\n" + ] + } + ], + "metadata": { + "kaggle": { + "accelerator": "gpu", + "dataSources": [ + { + "datasetId": 1835, + "sourceId": 3176, + "sourceType": "datasetVersion" + }, + { + "datasetId": 2568, + "sourceId": 4304, + "sourceType": "datasetVersion" + }, + { + "datasetId": 213609, + "sourceId": 464671, + "sourceType": "datasetVersion" + }, + { + "datasetId": 1075326, + "sourceId": 2250642, + "sourceType": "datasetVersion" + } + ], + "dockerImageVersionId": 30648, + "isGpuEnabled": true, + "isInternetEnabled": true, + "language": "python", + "sourceType": "notebook" + }, + "kernelspec": { + "display_name": "Python 3", + "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": "3.10.13" + }, + "papermill": { + "default_parameters": {}, + "duration": 18779.382989, + "end_time": "2024-04-19T00:36:11.771940", + "environment_variables": {}, + "exception": null, + "input_path": "__notebook__.ipynb", + "output_path": "__notebook__.ipynb", + "parameters": {}, + "start_time": "2024-04-18T19:23:12.388951", + "version": "2.5.0" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} From c00be22caefb52425242958629ecda5ca4d9d249 Mon Sep 17 00:00:00 2001 From: Khushi Kalra Date: Fri, 12 Jul 2024 10:56:40 +0530 Subject: [PATCH 5/5] Created README.md --- Sucide & Depression Detection/README.md | 56 +++++++++++++++++++++++++ 1 file changed, 56 insertions(+) create mode 100644 Sucide & Depression Detection/README.md diff --git a/Sucide & Depression Detection/README.md b/Sucide & Depression Detection/README.md new file mode 100644 index 000000000..fcb673efd --- /dev/null +++ b/Sucide & Depression Detection/README.md @@ -0,0 +1,56 @@ +# **Suicide and Depression Detection** + +### ๐ŸŽฏ Goal +The goal of this project is to detect suicide ideation and depression from text data using various machine learning models. + +### Purpose +This project aims to build a reliable text classifier that can accurately identify posts indicating suicide ideation or depression, thereby potentially aiding in early intervention and support. + +### ๐Ÿงต Dataset +The dataset used in this project is: https://www.kaggle.com/datasets/nikhileswarkomati/suicide-watch/data + +### ๐Ÿงพ Description +The dataset is a collection of posts from the "SuicideWatch" and "depression" subreddits on the Reddit platform. The posts were collected using the Pushshift API, covering all posts made to "SuicideWatch" from December 16, 2008 (creation) to January 2, 2021, and "depression" posts from January 1, 2009, to January 2, 2021. All posts from "SuicideWatch" are labeled as suicide, while posts from the depression subreddit are labeled as depression. Non-suicide posts were collected from the "teenagers" subreddit. + +### ๐Ÿš€ Models Implemented +The following models were implemented in this project: +- LSTM (Long Short-Term Memory) +- BiLSTM (Bidirectional Long Short-Term Memory) +- GRU (Gated Recurrent Unit) +- BiLSTM-RNN (Bidirectional Long Short-Term Memory Recurrent Neural Network) + +### ๐Ÿ“š Libraries Needed +To run this project, you will need the following libraries: +- TensorFlow +- Keras +- NumPy +- Pandas +- Scikit-learn +- Matplotlib +- Seaborn + +### ๐Ÿ“Š Exploratory Data Analysis Results +Exploratory Data Analysis (EDA) was conducted to understand the distribution of the data, the frequency of words, and the sentiment of the posts. Key findings from the EDA include: +1. Word frequency analysis revealed common terms used in suicide and depression posts. +2. Sentiment analysis indicated a predominance of negative sentiment in posts labeled as suicide or depression. + +### ๐Ÿ“ˆ Performance of the Models based on the Accuracy Scores +| Model | Accuracy | +| ----------------- | ------------------------------------------------------------------ | +| LSTM | 90.3% | +| Bi-LSTM | 90.7% | +| Bi-LSTM-RNN| 92.3% | +| GRU | 94% | + +### ๐Ÿ“ข Conclusion +The GRU model achieved the highest accuracy in detecting suicide ideation and depression from text data. This model can be further optimized and deployed in real-world applications to provide timely support to individuals in need. + +### Accuracy Results +The models demonstrated good accuracy in detecting suicide ideation and depression, with the GRU model being the best performing model. + +### Best Fitted Model +The GRU model was the best-fitted model for this dataset, achieving the highest accuracy score. + +## โœ’๏ธ Contributor +- Name: Khushi Kalra +- Github: https://www.github.com/abckhush