From 413aa2fd2661a504b9a83a63eab4f731ad95ee1c Mon Sep 17 00:00:00 2001
From: Rausch <PPA714@ad.ku.de>
Date: Tue, 15 Oct 2024 21:14:44 +0200
Subject: [PATCH] updated readme

---
 README.md                                     |  63 ++++++++++--------
 README.rmd                                    |  20 ++++--
 README_files/figure-gfm/unnamed-chunk-5-1.png | Bin 9356 -> 8490 bytes
 3 files changed, 48 insertions(+), 35 deletions(-)

diff --git a/README.md b/README.md
index 524b970..6aedc96 100644
--- a/README.md
+++ b/README.md
@@ -1,6 +1,6 @@
 The `statConfR` package provides functions to fit static models of
 decision-making and confidence derived from signal detection theory for
-binary discrimination tasks, meta-d′/d′, a wide-spread measure of
+binary discrimination tasks, meta-d′/d′, the most prominent measure of
 metacognitive efficiency, meta-I, an information-theoretic measures of
 metacognitive sensitivity, as well as $`meta-I_{1}^{r}`$ and
 $`meta-I_{2}^{r}`$, two information-theoretic measures of metacognitive
@@ -210,28 +210,31 @@ quantities of information theory.
 - Meta-I is a measure of metacognitive sensitivity defined as the mutual
   information between the confidence and accuracy and is calculated as
   the transmitted information minus the minimal information given the
-  accuracy,
-  ``` math
-  meta-I = I(Y; \hat{Y}, C) - I(Y; \hat{Y}).
-  ```
-  This is equivalent to Dayan’s formulation where meta-I is the
-  information that confidences transmit about the correctness of a
-  response:
-  ``` math
-  meta-I = I(Y = \hat{Y}; C)
-  ```
-- Meta-$`I_{1}^{r}`$ is meta-I normalized by the value of meta-I
-  expected assuming a signal detection model (Green & Swets, 1966) with
-  Gaussian noise, based on calculating the sensitivity index d’:
-  ``` math
-  meta-I_{1}^{r} = meta-I / meta-I(d')
-  ```
-- Meta-$`I_{2}^{r}`$ is meta-I normalized by its theoretical upper
-  bound, which is the information entropy of accuracy,
-  $`H(Y = \hat{Y})`$:
-  ``` math
-  meta-I_{2}^{r} = meta-I / H(Y = \hat{Y})
-  ```
+  accuracy:
+
+``` math
+meta-I = I(Y; \hat{Y}, C) - I(Y; \hat{Y})
+```
+This is equivalent to Dayan’s formulation where meta-I is the
+information that confidences transmit about the correctness of a
+response:
+
+``` math
+meta-I = I(Y = \hat{Y}; C)
+```
+ - Meta-$`I_{1}^{r}`$ is meta-I normalized by the value of meta-I
+expected assuming a signal detection model (Green & Swets, 1966) with
+Gaussian noise, based on calculating the sensitivity index d’:
+
+``` math
+meta-I_{1}^{r} = meta-I / meta-I(d')
+```
+ - Meta-$`I_{2}^{r}`$ is meta-I normalized by its theoretical upper
+bound, which is the information entropy of accuracy, $`H(Y = \hat{Y})`$:
+
+``` math
+meta-I_{2}^{r} = meta-I / H(Y = \hat{Y})
+```
 
 Notably, Dayan (2023) pointed out that a liberal or conservative use of
 the confidence levels will affected the mutual information and thus all
@@ -410,7 +413,8 @@ PlotMeans <-
                 aes(ymin = ratings-se, ymax = ratings+se), color="black") + 
   geom_point(data = AggregatedData, aes(shape=correct), color="black") + 
   scale_shape_manual(values = c(15, 16),
-                     labels = c("Error", "Correct response"), name = "observed data") 
+                     labels = c("Error", "Correct response"), name = "observed data") + 
+  theme_bw()
 ```
 
 <!-- Show both the code and the output Figure!  -->
@@ -463,12 +467,13 @@ r <- factor(ifelse(OneSbj$response == 0, -1, 1) * as.numeric(OneSbj$rating))
 counts <- table(y, r)
 ```
 
-Then, the different information-theoretic measures of metacognitive
-sensitivity and accuracy can be computed:
+Then, the different information-theoretic measures of metacognition can
+be computed:
 
 ``` r
+meta_I <- estimate_meta_I(counts)
 meta_Ir1 <- estimate_meta_Ir1(counts)
-meta_Ir1 <- estimate_meta_Ir1(counts)
+meta_Ir1_acc <- estimate_meta_Ir1_acc(counts)
 meta_Ir2 <- estimate_meta_Ir2(counts)
 RMI <- estimate_RMI(counts)
 ```
@@ -476,7 +481,7 @@ RMI <- estimate_RMI(counts)
 ### Documentation
 
 The documentation of each function of the currently installed version of
-`statConfR` can be accessed by typing ?*functionname* into the console.
+`statConfR` can be accessed by typing *?functionname* into the console.
 
 ## Contributing to the package
 
@@ -531,6 +536,8 @@ issue](https://github.com/ManuelRausch/StatConfR/issues).
 
 ## References
 
+- Dayan, P. (2023). Metacognitive Information Theory. Open Mind, 7,
+  392–411. <https://doi.org/10.1162/opmi_a_00091>
 - Fleming, S. M. (2017). HMeta-d: Hierarchical Bayesian estimation of
   metacognitive efficiency from confidence ratings. Neuroscience of
   Consciousness, 1, 1–14. <https://doi.org/10.1093/nc/nix007>
diff --git a/README.rmd b/README.rmd
index 9c85c05..4012309 100644
--- a/README.rmd
+++ b/README.rmd
@@ -20,7 +20,7 @@ gitbranch <- "main/"
 
 The `statConfR` package provides functions to fit static models of
 decision-making and confidence derived from signal detection theory for
-binary discrimination tasks, meta-d′/d′, a wide-spread measure of metacognitive efficiency, 
+binary discrimination tasks, meta-d′/d′, the most prominent measure of metacognitive efficiency, 
 meta-I, an information-theoretic measures of metacognitive sensitivity, 
 as well as $meta-I_{1}^{r}$ and $meta-I_{2}^{r}$, two information-theoretic measures of metacognitive efficiency. 
 
@@ -165,14 +165,18 @@ because both are measured on the same scale. Meta-d′ can be compared against t
 
 Dayan (2023) proposed several measures of metacognition based on quantities of information theory. 
 
-- Meta-I is a measure of metacognitive sensitivity defined as the mutual information between the confidence and accuracy and is calculated as the transmitted information minus the minimal information given the accuracy,
-$$meta-I = I(Y; \hat{Y}, C) - I(Y; \hat{Y}).$$ 
+- Meta-I is a measure of metacognitive sensitivity defined as the mutual information between the confidence and accuracy and is calculated as the transmitted information minus the minimal information given the accuracy:
+
+$$meta-I = I(Y; \hat{Y}, C) - I(Y; \hat{Y})$$ 
 This is equivalent to Dayan's formulation where meta-I is the information that confidences transmit about the correctness of a response: 
+
 $$meta-I = I(Y = \hat{Y}; C)$$
 - Meta-$I_{1}^{r}$ is meta-I  normalized by the value of meta-I expected assuming 
 a signal detection model (Green & Swets, 1966) with Gaussian noise, based on calculating the sensitivity index d':
+
 $$meta-I_{1}^{r} = meta-I / meta-I(d')$$
 - Meta-$I_{2}^{r}$ is meta-I normalized by its theoretical upper bound, which is the information entropy of accuracy,  $H(Y = \hat{Y})$:
+
 $$meta-I_{2}^{r} = meta-I / H(Y = \hat{Y})$$
 
 Notably, Dayan (2023) pointed out that a liberal or conservative use of the confidence levels will affected the mutual information and thus all information-theoretic measures of metacognition. 
@@ -289,7 +293,8 @@ PlotMeans <-
                 aes(ymin = ratings-se, ymax = ratings+se), color="black") + 
   geom_point(data = AggregatedData, aes(shape=correct), color="black") + 
   scale_shape_manual(values = c(15, 16),
-                     labels = c("Error", "Correct response"), name = "observed data") 
+                     labels = c("Error", "Correct response"), name = "observed data") + 
+  theme_bw()
 ```
 <!-- Show both the code and the output Figure!  -->
 
@@ -320,18 +325,19 @@ r <- factor(ifelse(OneSbj$response == 0, -1, 1) * as.numeric(OneSbj$rating))
 counts <- table(y, r)
 ```
 
-Then, the different information-theoretic measures of metacognitive sensitivity and accuracy can be computed: 
+Then, the different information-theoretic measures of metacognition can be computed: 
 
 ```{r}        
+meta_I <- estimate_meta_I(counts)
 meta_Ir1 <- estimate_meta_Ir1(counts)
-meta_Ir1 <- estimate_meta_Ir1(counts)
+meta_Ir1_acc <- estimate_meta_Ir1_acc(counts)
 meta_Ir2 <- estimate_meta_Ir2(counts)
 RMI <- estimate_RMI(counts)
 ```
 
 ### Documentation
 
-The documentation of each function of the currently installed version of `statConfR` can be accessed by typing ?*functionname* into the console. 
+The documentation of each function of the currently installed version of `statConfR` can be accessed by typing *?functionname* into the console. 
 
 ## Contributing to the package 
 The package is under active development. We are planning to implement new models of decision confidence when they are published. Please feel free to [contact us](malto::manuel.rausch@ku.de) to suggest new models to implement in in the package, or to volunteer adding additional models. 
diff --git a/README_files/figure-gfm/unnamed-chunk-5-1.png b/README_files/figure-gfm/unnamed-chunk-5-1.png
index 8cca74a06028a22caaefb07b567846da68268223..03c60efd776e794492f54f343bac2035d234803d 100644
GIT binary patch
literal 8490
zcmc(EXIN89*XZnoB3-G{#6}H8gn)pOjRgS#gGv=F2pFkSq(}>TfFoTgDkyA4AyNeb
z(jv!$C`BYRIS>TtO{615$lc)kz2CjxkN3Iv*WFK&Co`+e?6uacS>+zl^fV8r7$*RL
z$I#&9IRG#S0MNG`SainZoC+H{uqPT{IE4-Y&<4N>05Z_l2HFY0sZ3OqAS!AvN(d9R
z&kz$6Q&UqzF;JK`aN0uAMt>8u6BsBGL6krcwI>kJPZ+^IgAhhD*?QbfOyW%3b4}vv
z?Y9}(i`s7^h1sJ|MmM^c6VP@7PEIICCvB&M1gEkxr!po==5FHdZW8D2j(&38bBDtS
zqG8*t!@}&tGQz@Wv@lwC+&1?(6tA9{pbZkx-8g{+bVmtJ31tZhjIsm<Gs9jq!=8{q
z3(M%HWpsDvx|`%~ag58Y&&?ek&K+i!Ie{`pLK%Zu#$=Q+nY1tht(!*c?yiqBsb?~~
zGwi!*Vcp#*DLU89NYG|5P_8ICgTZ7D=eiFwnNw3!%ou<&V%Eno83|i#P~dPb6J^On
zd2AsvnarTdh)Gn&95)TDy#e5@-TFcwJtz(UNVyxH)VtvSWcq1r#xOPD<MkGrblAnR
ztLIZ+SHJM+K5&Em0o*2BFFD}Y&Sv1vb!^AYn~BT3P9Eda)2qxv>85K<Ar4J@&V(Ff
zhGewFT;D?|UKS0tEOt0!^J3?FtLRruu)|pZgj;cukLR(%`@Ue1oh}gZ&r(;2rXkRb
z>v8#9E@twD-1juRpjn}(3aG0?M0X653gk2)q6tV!0~8t1XaX(>fDsvx!$I~E0G%vw
z#XUa6FMXqS>MTbvv&FCGlg3%?miq~p;ZSCKl{IB>x<Q8^_>3?&Q+U1qN8f5|#>OuQ
zP$PdFT@SlmwLbDww|sqRszs}%SiXiFk{Z4;LAdv3Z2m9r^P8h5P&^~Jrce05<&EK2
zvVALSUCgzP^DU0yIg#z7B6P*~g`pcBgVV7X;;viEoo6=zHQor0PLwv>m|xGJZ4Q6c
zAp{kx>m*%<Yu48J-mZOY30cr9Q@SH?^LFk4M3OQGN~Xk+FYmPk3JW(s2frH>A)0w@
z9;xDQ?wfa^f1lUM?W1+hqFfjG)pf3(t-hZqrh@npHYT#ICFgAQhs$o4FJ!h<Z=+0h
zA)@g`4-xCB4MMTl#>YXoK<kAx=AuO{madyqHkjdm;w{IynWO8G{?NvxltbFM2Z3UH
zrOAmV3(@!Fl*PXn>i$`LHb+QG3AtROd^<3dIc#@}M`7*c%&*utVtdGJvb~0b|GVHh
z$j%s~mj%Y7diwkIHM@l<4H>v|`J$J~jiqsI8;dTg$;q%*eK6S|M8*(HS&^%3`r;5p
z))H{Sox<b@(cNw=NV+W~2h1-5F{dlkUF6$dtO&T^29X~9KM|_B!ZT7*QZ=*TYZzE^
zcvLsIgB%i)InPSgu5S)rdjtjlYRzxd1e1F;e@<LuTWn3U7E<9d(>n5vVNC%(lQp%8
zci0vOa^EPtlrti-S5%yp^5#$|rSY;Nyk6cJd&p&0CyvIWCisc0>pz#aI49NGJbY-I
z*M-Hqx;1X~*xyum)Lbcddwa9l5v4CXTMlensx`D_!Ofg-1u8L%Z4)lRY@4>wwhiZP
z+qM(ywk&b~i6ly#9{$uGqIyt4$3vI+MW4SsfzhAbBIGsS-*S`d`U_A5!HC#ykc>fW
zu~_6BNOHLXE-3iGj+)Wkcnx^sCPeYDoLbB|Kpz(qMIyF`zcn(?%}vg6`$irVyWq`=
zYNj<RJC?&{PRSyDY!|9QX2J4Aq85~jNKu;`=Z}8opvRmAJ@zC1dw6bLN+NUWLZs;A
zcLDFQP5xK9T^;@RBr12khB}k1$}z;V1t}AWrd|HP#MO4&1Z*^AEDAtJVy@lV3kE8O
z_88z+As5X{|AVR#gU;hXXDr7JH?b|`v_-f4h<adRfV;Y7FiLH2E+y5ml8)QRUo$3~
z(hkiO80A$L)a}f-qa0U-a>|~iAk+B~6eE<^gDvwbRSQ0yyH5rdXfyh4f0cNd7oAtO
zcvZFe;awoPi_IANXD7F#WOUwP+bAiAzpSz#N){SpMOD_~TnNwh+cgKUATkHQ(r`~|
z-}8QHemE@%Ll?#+E|A*uVzI<KTk>%9+H)N%S0@YHisp~RTWlz+0;72j_+p;o=S;gA
zF5rj*Io7qT>Y@*XP)5$i9^Etix<~_(2hdkNf145V@EbHOF?;skM@2&p&evf(!A=eQ
zi?DrSCKi_^?}lNWRlj9uY>9S(t;6Xkej8+X;OiU)hA5$3)>vo!zu-x&bhX`*h%a{Y
z#@61AcAH456hY2f>8ab~K#-iHgoOiJroJA)MUa08Dxq7y?c;Xx>JL$FUgB}b@-IWL
zwkfqO;pxcSwr!d232(+=Uf>nsz!<g{v8cvelsI!)Sq%PVz;`j!mYe21JK$p#lV<-Y
z?S*YlV5nUr7@gTDSEib`Dv9jeo395_BN-m*d(pjLvNb<_<=H=w`2dD!G60;b{${<w
z!%k1ngtnF*)IJTb$89eZk88f3N6tE!0n|m#z!QO?t(a7Pp`%+)q!FTT{>C3R>U$Em
zn_g6mL6zJP7z(Hh6m*YM)xZQ$+J^W%1f}i*Y8Sua3)J?`<B*^6T9_!=$+?rk1(7ws
zlupC3A3|JM?ps8VHSi9oY`MkhAEh?r3kM51v}~XH9qd+_cd>xRcSL$+sSaM?6sNBj
zvU>9<!~{RQvM+yM5^|UASNjJ93SP<e0CJIkO=6IWA5vVjthYt%)SqH{+p9`^HTK5%
z{1YR%?f4{%{}j4``T)U?Phaedwz~h{jv@+6-=5eci)ClBqbFLBeA`@*%xmKqOZ`9m
z$ahFm+fB;NmEC7y@y66<#OK0Z8A9k!!<wNBtS}x2@=2(+q_T5%1Ie-k1#~eXh<@J|
z8mH%RlP~qzDlW5H%_s|7&Nxc(3P;x+esjIojkn+W`^J*lXnCdA#qPG~#JavJ&c?wK
zW50ua@g3K-Ycx>D)+kATr?$|hFGg=lTc0o0D!+Z!{S-&JmQ_^C#N;HcS?_K0lQ>RB
zU4z*NM_$t0>Y#(XsPIH|i$jE46rhLUHXc}j)SMozz@|q@OADj3zBgJ7aNh^b{~5M`
ztb#Uc@@A7K)Vc=M9$!U60Mt!B{_I#<@R#7J{>OQ9?dH0_CJek%lGC3LE4tzy%g9#z
z3QC{u(P^&uL`W1Z+w!=hWZn$zR1F)9LZL!oZ_2lqr}X?omO{PtU&dk<2OVapa(dgF
zL1PUevXiz5>(_vg>GE)+3qv{JlZi-If0MtWYn`LsQVGk(QI+4goKnQ0d?C;?;oN2L
z;4jA{yS`6yIx0Ee^9K7x-d0QxI@GioG!nS>XtLY)A6C*ucOvf&2l74R6Q}o`7+0!%
z{J!qG=c?RyN0VUX+v%L*-OWjMRX=Ddg;#&R$yir#hdElINNAkgxvxid-#eQB&w|l(
z7m$jfYv2GYQpH7fE+~R3*$^ijkQAjuQ&WI0oGa3mauEmA?J}c0iZDb_Sp_JgVj_r(
z`h%W4;A0C177|nDwmgxpEyFz@&;`k5*3dZhx93LhX!=(n4J7|q*L|qRZ`ZkoEf4l@
zHhhRyh9`8}^W4$H<^8i*MDn8}8)Z;U8D2085RqNAI%lT{xy@9>anaM8j-2Ko>FOHA
z$=u++!iHjwu#(=5nUaP1zD+$)JH#!Tp=AuVZWtj^aO^u_{cqm=xhvCG+RjNp)8o5i
zI8nEq8go%s5>vVQuB9O83C+D}U&DMZm)Fv-x^YU~VLLq6P0-_MYf4TDs$^NbB7b?J
z`~L0=iUe(9{a0hqam9y=Qq9)<=h14Vt2BSDKs&-M^93xi8P5shg0x9>9%qQG6o0$7
z)9mzL8Q1je&K$C3`u&qr6a9tO^+=zA`*Lff#Jrq#fxY7zhmlXHd);f6>B)+?S1jlS
z<d!Fl>@lJbigjQS5hVcvpzBqj9_2$e7rfx?TfusvbdZ*W?1Or`Rd>Mc^*;{2Y~EbW
z-w!3ZKFs<Kp$u+x*ceE*MzE3QUhn8OICl}fd1(s0?lQngbV-;WxRq4in*Ony1u=tC
z@dI~r2FF-P!TGi&F8Set0*GW33;oc=xi91b@h=pcxsf4X6M{S0X0hsDx}<oZ7t#lo
zac7--TDNKi$h?hig78$(XF>XF_gffx*8Oj&!Wi$liD@AZD<GpeQBlM#p7BZvHK|K4
z35s4m0p5m|i@=BIehu_|OZ1gqjo~6zxR0B?b+)ss2BpqlF!T-{WGp&JbX7R^=d7J1
zy)$G7Ds$>H{L@Z%@&nE1!~gV`_#XcXQqv8!)2_vx2ik|;>;i7{h7|=o*3&F<0lABV
zmAmJ0>5|GH#VsSGnr`jvZy~rVjCyyMRrs$ywI4WnfAMkuJ-zqN8HdYQ_kIC{Oxaj~
z-1UmyNuTF@E#YCyO}I3D#;c`p^Hd5VRI<C`poXmF%ssJ60|iazjCm>tc_$8@PAx@!
zt6O6ugs`4SZuRn-wY4aA%BuEK=}*>D@f$H89${qPZnMAqqnjT-bV3)9?A(ELp`K<7
zg`ax3&+38y{%qGqKTFWu{G3ZI@S$S5l8a&ITLI_K*?VUWHggc~v}8^7JnCG@Uzx^}
zXe`ZKHToPjDfgahwD4;h(d{p1F1oV78u5!nAa?@p8qJgNQFFUEH{w6kID5^E$Ag!k
ztlHc>*k}&>7pth)m%eb8`dTMC_~>9$9C{~KPO~e06c?$DH^&CgEY4@S)>(eK9)R0i
zYst;`PhNCoM0se?Pq-e0h#OOt(l{r-2Nwb<kr)=AtJz0@iDE4);@kv?&y8<n7ewMt
zz#J2*d5mIvHc}|+t$)XywpbRbzX1!`2qG5NKn>Z#wafdmtU;xXVv5goIVt#+Hyb<#
zjdyo4wXEAp{=m@VhD=#VtAgX)AUQEwv4X(I2}5$%VWrg4{E535^@-;N^Sd{=fr;hw
zD3Pgycp({tw!VnPQnfz~F{q&#AtdtX)IqZ#Hcrw^-rsCgKZ|#ev+#A;yyB3)1q3SP
zYCAtPK*U$CAbfhUXYn8!-HDEIQ+6jhSVgmud-m9$?Ba!&!ajI|MdQ~ERZ@avcPO9d
zj|>QbjB7}`jzbQtaK$TDI{U|uDnVx7u}E3u?iT#MV_fi*+PzPJ?nU!|7Fn~hpT;~_
z6s?97`Kq;&vlM!@Fg@e!LJ6aoZTwOp`iBd~C4gE`%|eq=cZlt4b+lQ06KvAIM*&Dq
z`B&9f|6R1h&#jknA`H7Wn>^+rlo61GWF3}n<X2bCk}ZJ8%<gE&ybfNp(31nD$KI0)
z+{paiN}o|(urDr%Phz>?*lcd{XACiTGL2m&q&A=BsSt8cYvWF8o-Z5uiws{?Xwawc
zVuIw+R2=fXE1Dgt=*c+#<G%S7vR6i0Lr^0aR|IYXTf7Yl?sa?fz-jg@q!~|%W5xD6
z#!Xv4u*uDM&W-Z>Q1f<UdJ?W`uW5nCK--`*FWKSikle!6k0ls&)!V6A|BfS$3=KAC
zyIPRNy3RbpB6%O{u*6pZSZ85yD<Q4>5m8?r7;$Z7QLgG3=S%bhzHs#V6I+j4sZB|$
zB^V?>hnk^z)=x1qVbvn|?fMHH4ryn)L}{PIHaRJ(%&|^1f^6}Y=gxV#N1k)&;?H7^
z1FEOP%ioGFvF9Z*?(7+YDg)MNck2cXxPeVn_8%f#o*rg-*q`nZ5i80T_nPi8j&2m#
z`nr3sm^zuSb6y@ul<lv_y4U0=emYZ|ayUM^ixo}#oYYOZ$My7kjjp`m|1{lH1Y@G3
z6Mt1%b(Dg|y_#;k%bky&C0pHB``H{c|KdzT$y#dF*lF2?u9J&*AJ}dsaF1g+9F27>
zo`?$29p{!8U*(-G(X6n0<rdK^Oz$;uj}8tme4U$P2iZvE8|aS~$|s?2J9Q0vfqcHY
zVJWgCKd9c#+dnY*=c0sL?n?WEF8j*^7k55RA?IW!Tl>iGgiW$tzvo$Dd)^MlsKA}=
zj}9Wv83OP~;(ph4J&2?&gxH6FM~@GKR{YsCT4D3h%)H{V8*bi7s4y;q-v_~Q!2*qs
z&|}zb&|?|<B5&uXBe|Ln-fu0FD}C8H_C=U9rQ$0=3CTLIaogevG;PJ-unTd}Z0~#Z
zrI~v9udHQ!)#J8B*P_B(Y7i0hOjoYcR^Ca@mMP?}2n_|!0cv1#!8nBf8s4%cPJMBK
z#vQPZx;zqbesnMHE!nrA8_$D$?$q6|m)}aI(NI2j@==H<ZhK&$&!9PVaf6ZDJgG!|
zO>0$<6emp1ik%Qa>jUf+_w_-J{2vn{5T!XMQVk?c(frY?PY1sb;c)ax6vDy{BsYs=
zEdPpbudSU5QHut|TywrLAlIa7xQzrxWH&Rl$5`cr^1l{80T(#Yq_pt3P@y$M%ARM}
zSe78XrzYPza3iMfR%CtkXd#!~wcXQlQ)2C6EaT7WNXv8Y-=@hwi*YYD%2v^PMN$&c
zShl#M0BaIfx|8Eex~8KAwA%k%Ir8nM9U2NI??cL_KE!d1S3dLV7pC8J)XiG>E_bZv
z_Sq1hy*y~Bdphg-kF8`sRbPf&CSQXF9G_sJ!sv%3T}$UJ@ilL`W3ME39adjb6V^md
zQol@nxLr6XgbdzVEfayYd{GUYwl<7~C?EL+$@AEx!9)?H4o`_!NWV-V4aGooRc6jW
z7ZN=pOze#M3CYzsl15ocYBmpAZa_<lqg!!~nMDN{FF=#%?LTn^&1xDda93QYJE+V=
z6CG(zA}V1#P6g({)nbXq6M@kz2mJ9@E})c7{cTzVD$i3v5?xSw+}=W(q}GE+qX0Zk
z7LMie*v6hz!GT8e7${lfYDlp?K-UHe??s56#<TZPb)~hYT1K!Ve{4xW0fVp=v8q4T
zLX>c|XL-geB<5Qu5BJsj8AB!y9^2iOaex~6iK`v6);iCMu&J7)v5u&-=c3NWkuTY{
zbR}KP`h6`{H1@8qel)2IQXYFs+X@BbwYNfnhR*(l;6^KfGq8DNQw%VfY&GQ~wGk_3
zf-d@|48=wde9R3t{IS|uQ_oMe7e4Ut{pj%IG1V88-{C;sJualYQ?xPv^TM5HjU+8C
ze`TDt+WAWr4`&*mJgoBihw)#5fOS(G7a}D6qSMfVXG&d~)OM(VC;b4^{O*T?$1;uR
zjsDjRpFe^_X4L!bAa%{?XU5%C4s)*7XL6S1&)WS0lEw2H;ORFp>ZLoQ;g>U6l$uVO
zrCq-#E=?*bG`r1Ng!Q2pKe>ucRgam*)q3BkdpHz}tBtS2kB`m{@wfLX7TB`>A-g`m
z-(rj5H6=|avPI#CF$}3qy_n61!DkbvsT?6C;!>A!sV^yiC2CdCqwX>@pj4>vK{^^(
zG+RpsK-xaqYYC7^>W8@GfowJ(_E*b9z9NjztZgZTHNta1BzaRb6{=TOhN~<h_;%Ab
z?ODco+Yg0v$#q_bVB8>fT-jTlWoy)0arh{Z9+sd#)!!Y-Ng8XL1xZ{;6%T<^kp3h2
z4J5}03q-X%#`Ms(@dW7(U6~Oaa$l#oV7w<#?_Q<oVT^vR+2nEa!5msF>fIZZeHiEQ
zC03*|i(OrGfHDW1OHk70rP<`6A@Mqq>tqcny!g!y0O5O60D^xE1uuTd-zv#<Xj*Q=
zygFt93VUrMx`%9Z%wIlmy;|<LBxH2dS!0-j1{0M^5Ip5v!zT2SBpi_tUFR!Mb4K(c
zn)G{WWR*)QyBeb<FsaZLo4M9g96hS#2L|~&eEL7RJNhRfM)7^U^Ai|)!8^%0jB~XI
z8!YbZR@X0A6T?Ru`xZxnUV1zUcG-H><hB(FDdL*my)d+)L9C5c14fP9;~UrZS?2pk
zx1KJ+cKAB_10#|H?4~buijy+&Ss6&<Igr@wDJK@n+-K#A*Q($thB&F*Xglh3&d(Q{
zp4Lzz13G+fo{lMDl=gO)#tW1A#rTL#?<8|}bj#;%7MzDF5B<RfAH$|v@%{LdpE#7s
z4mu8ckcAhJp3QPqjL|2rOcvN=9-?TRnu&q%n`~2ffNxMmN+&WIwo{@11OdBD7JQQ^
zvh=F?qV4szD9vNlFV_gIQL-Z#dReYXyNYlqfwDX+yr$=dKa6zRx6j4vINjzWi|y%~
zd~{<MIgX7~V#JPgGHu%`f27v)lH;(HjZPVmZ&f9`3qmrJ5tmn~_kBy%Kd{JkCKi=F
z%7n;qK|D4u(I_ppqdnXLdjz6HTU0+6Bn#_LP8_cqI1M_i38a!n4=uQt%%Aw@&KhnY
zJ+fkYF7X^lUG~=yC9jo>6q;!hVDX@HLR<h}&uBNED#Nuug52h{PF<mT{viMGWgb2P
zPd}etI*aB|1hX20du8V($ow`PT=ZKtnHXX}xe=o#S#>r8heWR(@R>?KjTSW_bMs1K
za;pK9e-dYYk{h1JOCmSITUS7G{O1lJNl%1eSypu=h4pgqRrkxwQw;N~fOxLm_U`2V
znIlWu3gWO!i-g6h#9|~wM{A==Q*nN$w15$>r~LC$k^qKy&KQ`aDXo)1vTK`={@cg8
zhkB0q!GV4qQ!x>MWbNL=r;b66sRR1k4r>Q1Pe~s}-#&!KY)~J0`nHyN2xyyfIO{l+
z$sO;pi*8J3Mg9;TX3dtj-6PUP^;9PJ5G{|`RC4M+#-xUSbHk_uZ8jug;q{d0!pdLX
zQm~0?UK$3eYMQ~53e{xts??`xm+lsH$>}}))04x-YAJiZI|Tt9_w)f(>J~P8ugy;n
zEfYS<#272G?8QlOTq11@aT<tXoyC>F^u8JC@dARfmQ@lvLi6^{2+=2}oEh#B5;V1(
zNKUFeGFuwfTFKM(*dBAiO;$zkOlGv9G4YO9V!w`|%+!^xB3acUTB5qC`;T@kJtcDm
zT!{M7%8q<3@WePXtd`g)hio*Tu3?-5csOCn8_(+;PWv{;9{5x|<QhT`D*);;aJ!8g
z9}JPwZsjbLhYO<R35}_wG0$;zB3dus&#I0un}+1T(u}iF|9XKO7|AM!Ox8`YXdWPc
z8xGf&rL!WgcrG~HCT$rc<vU?XIh^Ov3vr{WCp%lppA7|S!e)1}Ku671NyV89uH}yT
z?|LZUQ9HC<y+kV@Ib}fo(_f^m+9{g#P<k-*L)gr%lIQ<d!U*2++RH&^K`h%n3SD<t
zN}>k9cB8+$ADh2IpRwFZO<P|y!&c^_WlIy<p+YGkvgwAO|G9%?P867c0vF`;!vAB(
z!ta6le{4x?7tOXhasM}6j4$NRC($m^ewfjDfGAtwKL0a~F<fwz6<&9lBtaw(lZ_z^
zeV1*3vjn`a{yva<ZvzDFl>ldvFxoeC5rlD>1RaUXTDotqj;(zwKYYtNa!)8JfWH3m
zSe$cPN9%1~Kl@l{9Lu;o(0cJrTV!Ehv8Rw=Z)n3jU%5xqyRnoIx74FHH3O#kyFjPY
zJC{zZC-3)AJmdM~fJ&%~-=~jK)&p%h4Vg>gKZfTo)@27&*vmH^9+^8GTAa7AhZf*!
zH<5~WqxRK4wR(%Tlm=YN*n>v`suBa4ct-SLvw<I%@{Gmvf8LZ$xeiOwZUlZf#~2(N
zD12Mwj^8mIbXDU9b#sgK;v6kRIakV3-8?>i+HjVg5>48O)G4aQ^bExu^Pdx6=Q_gJ
ze@X)FG6mmKtBCsIT`tgsLsmm;z80pf_FPI{n%qs*rI?*RJ*CxqFI>m)cN_J8NDJo$
zw-omF9Su$`5Dv`lo}zufW@10%m$7*4Qi^^1QQ={D4x385{6ujin)tz5s$+f#G7U{-
zMKE&M3m~atKO0&ZE&5qNdC;$8rMRgOIL-B1*~3Xn@Y(~@qMzf6YA<txADadwrT=}+
zTBq;6ui^+4Tj&qbj%YPrnoli~(<|v3X%4n#JR~>Lo;y@m%ZuWUzPG5>V7HuknTvaq
z_AximSkf~oN`z=FUN}h>F7@UiHhdqLo=)FSBMC|>zAOr;uKegzq@U*x**Vswl1<qc
zZH0XB@TsK&=as*QY_lw@O5TjlP3}ny7b^jj!KvAr1++;yf3BMBj=Pekpj(ylYpj;D
za{or(L|Om&T>2T<WgB;p6@9aQx<O-_-sx`Nv)>y74J@OpBl<Ps%1ri#Umo$#NjCra
zNWIreNCk0xLlmOB$+nGNDO!0r<<<50J5F0}?Vi|IW6qj+|NX43j`*v2a|-tGx?$<S
zJ|_%fl$-gAiTRTsBtHnA-g#oHz5PG*@?YKWwNRnx0J*t`fu8~T7@jgcS#bQyo&N=x
CH(MwG

literal 9356
zcmeHtdpwj|`}ewM3?V6sknG*bp|ItYQD&q_g<;5<y(y+Cqa4d|uy<|8Y9k@d9mOyX
z9h@0bIVV#Ln{lX|CK|Rvjnlk0?fpE@=l4E;zn|Y9?|kMq?t5L=cdd1;Ypv^A-*v|l
zEX^du<i!9066Pn4*#Llo0f63ZKqF5qY&3<D8xO*1+vCVB09XLL0H6Xa7GQ4zuW~s#
zoSdA894<=EgC;L8udS_(NFZfc;I$@#MgHEz-sB)cI5`|n&I5--zEL<28ZL@yvG$=@
zSR`3c7#2w_9_v6JavtkKQ69*L)`wK)1+ZSg%L_sD!g}4j=~Z6tRW3l(C>9ioMG}RA
zd>ItRXcSH^Y8^T%$|H&v6~$ymG5eC%(UTCl7Qsy{xQVpJ3*1Cnbkpl*`OTZ0@|zq1
z%|njnfuk{_XnjmtUmt^F!B|5~Vze+Aqoa&bLAe(w=iDsk2+9SVa)E#ug=6+HnSFgN
zNfs>vK_AVdj~Uh1ho~aY`ZzbS91enth;ujs!6<_=DiBOfO$ibJ0wicj5^!#=!61dB
z3;_ZuKrq$>1p+}>^|eXF#-f)_IQj#y>BZUwHT%-f0`LPcKW1!u<t{JlM&v;G)h4U)
z!cz0CCDy%V*8KtXYI%bxt0($>%oXy?3n&(E*(WDT&9fW^KXFH=4sgB-Zh%(94`Zix
zp7?;l|8><!Da&$(xo5f|9JBdQv`gWSDjw=b>!A&kQsB32ng4O!R)@ci-@jVL?B(BT
z{t=K#ZWM?+s-sx}0{k-4oB`ILqg@?j-5np3p;}sg@e{`@=UuJBz{y>;11OeJ^-|(+
z?d;vR_JS)jR{Rq0vG|K^na)M-kuz=-yCa7~zfQF+Y716PS9_T?!^=4|{U45}nluV=
zMaEt*ap7Yzq3*S_e}tt=T=i$nn^R_!BW)x#w>8BRX5X3E*c#X-P$M3nH#HJW`*+sz
z8?6NMRTFM)3U-o8nfet@RgIbvi@uZZJH!ZMSV32p!p-cq`yN82i;r~jBNr!?hOR6O
z4VK_)B*fR6)G1hv=@iUL;+B38xTX^?g-pGNNJ?x_WCVX85^3p+{N<TZ1sPvTnQc42
zX!(Rrq)L+Umlrd&p<jDmpjk!-omtE1(D9Dw?f{7Swe_n}>(cLOe!KJsUkv)LHaJU4
z;+KBk88d!3cPM&1=c}M4bj~;bddDvvovU5zg><91$Z?Am_yiL%*#mLzgaB?n7P(~1
z1J06DN6h2YuE$9c!jCPaE;|~;S;??8hX_xZ^$`nhS2q@nADdZCtk}8pzl=cZ`F{pX
z6$-H?e|foCVgAl!p~?q6Mb9t-f{KDoQz|W*w=1O})&cY_dnsL7O(GBJeky=`ES3|H
zFl($@utSV0Z~ebgr2x$mF*7q8UM*aHK}?i5je-4z=DI2_L|!R)9F-3Ou2GFvxIsKk
zD_y;xwozYiW;Oecxz<vS(K^R?62=5q9@S{^J<n<X<VW|ApVgdLa^<j7<_O`tIEsM#
zNA3DB-*08E;*OWp0gH&G6v*xitm!NK`5yV=&uSoRH+=0d`_E^#Kn%aq{r~Q_yPPK*
zbP~g=qd@eM?wuS>EHRd&DmOxOtO4D@Y5)TllMl>IY`{Zw!CKo~*$&*!EmRaikPo&}
zs|y}KqvhQ5I&Za?qI|Y+Z1`50I6gej$=BUds61DyqD!l1ai*mY1J@BD6%a`DypJ;j
z<Oe6M#|YZ!t<A7$!B)FXy83u;y|~Vw;?fu{02{KjvFo&7>i#bVpDn~d;u+9=VhuY%
z-0KcKjWOVs-@~cUQ9X~!_#%pbs8-=l<}4igMNauZO~HV5emTdx#wM22y4^E6{rEce
ze*Dv{{Fgl2gv-W2&7mrs$Sdy*%9E;))@s6NXWD-EOGMyUf>y&TH9N$cGq%8r;ck9Z
zuZyKy)*N~qO|qw@$-{CAKEnU<Y4hzDgthhBgpddHKG1)aAdJsRRfmc8GK&&dY;D{o
z-bM%AkpF{hj0ml1(bGY-XZKAh1Z6A+S$4V)gTX+;?agbbqHE>=-M>p!yivFy4AFlc
z+hjTTIJ^ndXZt}f$g|_;CR0xIn$8X|c!YdHn9!yS0v)P$+d!l(7ylB)586%N>3+>q
zzg89t`bVXySO#JVR}LpZ9*|1c=VoHR!|oPD($#LdvO9@abgj%Bw9Sg*744>5x)*r7
zYh|jS?K6U8s{s_5;d&2L5WA>d#;Z@b=1P+E=<ByWc%SQKK*tFp>^?bQdLe;45gloy
zf~>WBCFt|D?cLBI^BqlctuLx*GH&J|qTWON^6!~%<GF|Y^j^lggebbrsumEyBrdRr
z>Qt0BzGf5KYHIoE+C3ul!4UFyU~v-X<tZ}NNyC2KXj_49_Di3HrdM>HpWV8|Low3S
zs~-?2eV)%OnHtPN)+AxP84^3fFSzV0yUd_2<WIFiZ6wd$h$^-9<>{?Z_xR`Diz$^7
zltKm8P6ot>5f@C|1D{E&_sV(cbICJ{y5K0$^6%yAT~jsSmHip_y=y3PaCng3$pK+~
zVZ5|Z_vgh=-?7lFh<~tzVDXV+ZY#Qgx8QQIaSe#R6{sttPW27psfkh*)+nUJ_3x5V
zP*!JTsIZqZg6X!fMy#;jNp&DmycICt@lS7KiQl#Y*jo8fqN7nVCZO5kUFfGZH(eiC
z3Cc!lE`~%iba{lVt(97OcOX`7lWk0x+R~T~7_74Y*ScG&1^Pv~U6RAa_mc93`L+w~
z=Rvi9Nj9m<S{as1@(|kU-93`9c=SfZa{PA4M|`0cBfi$EpG<W~M|71^&b7y4LHOdE
znK+GAzDwOu=cT=YH-<Gjn&@XlU`aRHD=luwGewO*RW2*UJif+?mI;|l)`UnxP4sK(
z-zH+-1{e?wk$O_xGzIcRiO=~al!oV~v78MFx<dGh3`=qBJDCG(mh>%@L&44v>s?dE
zkD!fwTDK){Q<d3fq$x4T<PdS|y1qUR>(iwn5n}BDiUHF5kpa1ej66h_+e<M~K*OLr
zL8{_@qHc^yN!))#B32V1Nl`b|Nc@1{O-OJjf9s0)$<?jqg^{3rW{H=lu@<kKLH{Xr
z>cWtf>-RYcVZS}k3wyMrEdnRmKwMU~lhG&T3YqwDOiyTh<NOjoIi}I$vdv5X=`;F6
zuAf%sf4O%hf@dk`?5?P}Ftq=E;k_x|CiciL{IJ{9O|@N`3K%$}GkoPKwlcz9a7O*z
z<hx@>+-dIh^;ngN3G?fApC-TkH1PzKhC~F0Wy_u6*mpwQ%51=`{*?9Q`d#xuTnS~M
zugU*yLFHn;$6~PP&-2TI&rSBew7fIxlDk9#zCFz0|11tyJ8L%BS`4R8&-P9<e$F%L
za2yILxcxLPYsyFLkBUysFKG*w;u`T2+Cl+b=G#6MFb3FWslC&rzk2LUu3TM-6gTbN
zm%^E_Zw}3~6L&fy0*}O@V6h`?ma9y2k^<dU-%XEtoI5$bWN^c-5PN`*>xJEA)JgL@
zs7d?9sKuEvdn8;e`;wgv&ZXI<Wd$!Usa5>$syNdtY~JYKI5LT620bahhbHVU`zp0q
z2@M7Y|2}kz>umICQ70Mut#Ff)T^NVxUXY%r)?}$Y)?6G~|F-$jQs%@{bm<CXn^Buv
z>h3Lgxj^?{ALwhWV(^K!A7rS)vOw1c8AExO^}sQL3TRmg1f*ek|GWLOY%#zD@XCda
zR3xuNCXm`?g>7)mG_rUv^=t?8qFgEJLZMc3zN>o?V8xHYGAWz~x{glYL(nF!pMz|t
z&)bXx>PQHas?dV4?{?k*rM`((mb|-F_m02@G*D`vXkdkb6^$E}0gvc8O@v6H1SC?0
zcHhBICIDWj!bEguDPUrYE#(G$M+18-=$((37^#kEr_bp!aAQF&W5h!}KmT<#1l#1B
zKt<n9i?ck!Q`{`xIy@V2)Caa|x+*YR^qbd{I@F7s^XV^y*7~it6R)k;FEj9$D#T4(
zwRhkCV|o1WfqjDoPbsMIB*QwVbWPaV2m`r>^OkyzUPn%I_qHxtiTZU|*E%Vd9de}U
zEBy7n${0{It18NEh+TE0DHa@WT+FBD9_|pr_x+q@_1Kejw?!dXr_jD+9H(#fFl1E}
zU43C3yBDdvO<2MRV>+8DQz~S;a`<YqB;1H=K04~~oOx|<zIOq4P@^x@H6wD<V~!-@
zhPsOOO5c?(lJRa|;|}3&6LBZSm=G{mF!Ee7D-ex;XcLM)`7H&Ma=ondsJC;jXsUBl
zz@UMD)>^zg7oE~sT4(O0KQW;{S(y0|kdsbjA`!eBJX~4YR6#_SVjX{dQTmhu=(c1W
zUSTEFgBq0o{Z<p7xnc--#agZ1gWA0x!AaloS9yACHV)f>%h%XRmD1%SH;U`H*beJu
z0$x)}q0!f>;=*`fj|50PStb(DDvQB$a<cPfw0m2@N&cu!n)>m*LR|j%uJHlcds*xa
zNC$i%*Ya(to8hj-0%>Yb*+q3VReVjiW8x~#EHk#j7Wn+tcfz$?HdO{z%;N-}*}aC0
z3h1-8wV_UC_TKwTG2B}hMaZ`WCvO<XCog4O{Tmjz25s95amTVN6T_|#WQCbQM}}$7
zb&}by0(VjM_&UTimrFU9sq*f}>@=s!>b(blZ3ERtT>*N%J4zt#v$vy=t6xKrDb%Ks
zQB$3Nd%3{>dc{=Yh4vfs;hQ7v4!g3Sr-fYfpP#H@CrYz(HCJuacWJoqk8!0Igi>j<
zvWeVcVIVKY+ACYwdIaX|e#Q$KA@gVXb^E@4UZe$zJBc$=PfhmD&My^QFV<t%xAnHY
z(!d*byGj6^&G0D=hcp0hGbv;Hh0MAPJsH2B?LTaE`+fEx+w7Lrkb$m#k=y=*h9K+J
zACj3ftNqzTbjo}G%<9qTYXeF##cl_jS^XiN79Sqhoh8otNw9Vm=Kl1>tm8wUXkTd<
z?@rBl*PeQF6#Vc3q+6k;nxABRUffA1B2gxQv#((2@Nngu=rbj{PF7+n{^t7txn_7X
zRig@4jXIy(p%PIuR9U|45cq<jpyaEi{(Q2<&iG1giKIuOMSJfz)z_-U=KU*!?q)ha
z;?rDip77flfG9XcxNKs5wWJNa&H3i^t#Ete@Un6ER(DEBxm=^Z2$#5HPlqTWsq<i9
zD<*%}(!Ts1lJhJf$n{UvZU5vbxfot<u^F-qq2Yv->rG-e%t61qv``BF)wl@BA0q{L
zbYY;DvX-041=^#*{6>|hzji6Y)~!NwVUG{JS2n3(K-5J{inGUI5qxK{pmSC<GL<z=
z{)=zmN9!8^WtxcN%{Lym^uQ?vSW>-A=u@{QV;<f5zW6D_?Nz91BYW4{YAYM!AbazE
zm^m>Q%ZY{hLoMC!mAvtoPJTHnCkYp6_gcp7q9_ZKX0jiA^n+V1v@d&p%XYnFz4fxf
zc2<J8hSMh$tAYJa2iYKPCzS2WA9K_3END2V*834+-9H{XqU#o5=gk@X14&NWZ2{I3
zpLdDhXaHqAP6wr^x%Y40RNxqH3PcGFqPHa1uZc>Tku90yn$e1sYHgT0x$--5pJcfe
zbe%p7%QEZCPl~ge4%4GcS$D{n?ug+hBvkC)L>tlC+cW!$Zr7|@evxT1d>O*l*S361
ziLb6T3gL}$*k7~+d7Fq6-O*ky1R6=7BU5@A;xyw*HYgG$rfv9_c?m$>QYdRb{AwdN
z&o|8qX=+V3(!|RRSJ7m*#qsySb;rcT@Ft9r55GtzCG9d4K6j%nWgJDS>h>(Annu=N
z0Zz*q`{6h4ndcExyIdNl2Aw~xd+1b_A<Xu$1pVDeYg3(qLYTdm=Tzb<k?h5qLPYCK
z2hM(~?D4UF>9dhaVTafMY5?T<)!&eswHo@+@=H%V!DO|H(yxi#zk+kkWl8b7C^@Kr
zW<_r=KpLI;0MWl>kMiY<M2q5S(y*Mg5bTO%SV=bCo4CcHLD1T$>`PV9JER#`iGdr+
zpwh*A+WjS1r1*7c5FeE4MYu5^jRbADN;xv2SwGAaRi?bz5nNxTh3Ew${pfhN6)IZM
z-C+5!wJ?6-n#43xKSWmr{X4S^;F_5b?j{ynd`9Vj%ylxvip8F~^7v18*{@Ndnr0>|
zMblFDn=3-z9|`AFu@i&8koZoi48OPz`;hCIxb|KBz7p+da&f!}{NZ7lFd7|lQ#!c#
zk)i!7-c%W0EFLkmZy>Q(j1{x16#J)U@5U^Kp`T3ERm;a8W5=tOpPoSSD`}R;&Xlnb
z#LVVd(y*g8SC*AL`TjTJAZ=W7@#|H-p&12Dx|JJgUt(9K<`%@;3$up`Iu=p=#Yq2l
zr5dOutEip9t4}LTJ&DH+8gZ8-C+ZQJh@lYtIzIj03k%uE>antrqPC*n3n8azA7>%J
zOBKThqnz#~wxXQs_MzxT#63e-S%{1HdwQzk*@{mX8g45?oQ1xrg7|vt(^t=w_+USK
z$+}NVsB@ontAGyq1}c>zV!kx?gQ~BP|7-3K<|~fOGh_0ZwCynuqSQ%Cm61qyPMo_Y
zvFkWQ5K}SiC=Di`*5=tiN-fdNOYfO_+E}NBgk)({$`GHhon<6{81iNK$VTq`5<ClR
zAJAu)uLza$oWC9QEchji{IpA2Ev`}mMe5mrVhtj$F+!Y>AdL()ZGqH^r`t6ty1DXA
zc_>19_F<rR#!#ca^35Icb{bY0uPuLZXh#I%+Myf!53403D2M~GFY}L^Aix1?AXSx2
zMhf|jj_CeTqC6qU>^|w<%Q4vrA(Ckf=st-^T_oppCGTHToG*&ru+9LX@N!1p8@X3g
z>}4SM;BgB{)`?zZp=2cp85ntslO8az3TEg0ke<Id(x>+OI-LZO!Wlng;AL^%ct@ib
zD5txR9xI1Qb(xmOJh()zYPg({?UnZ9(92DDUxz}k;Jj~n>7+bm2itElDR&Z|VATJe
z_P8ic?N6lVd7ikqWqmJ6!lylF4=y-rDpZqDA(w0lKOQ!8ad)dPp^3GCxz?f=<^I!6
zwl6Aeff_L-;(6TQ76ajq*jL7OxPMZhh%b6NaUCrb6B9$`1ITCNErcB;8=u0nJClQJ
zf&7?`G-9TLD~PJ~et4%d0ijSNzMf?q-O8HW-fbqUG;)}m0J+B5>vsy(?DpGOeW(vO
zC4C47^6z~}Dl~I+|J{c-HWNDI!ZSpvZCmY(J}wu2JY$R`eaJ>@E9>D|>#^k5of(vE
z$8Mqf-$_%e!|VrXx`Th)Z3VT)2cQVrClMlb@ozP8{FF1s`T$kqQ;HazGk_wPBI87w
zj6RQ-hEm=jKL}H!oAXi~6!~sDw!Q^=4_r^mVDMQFCS!iW9X}3KE{o$&HHwoAPXA?X
zD@2v5XkR#;HdC<lG}jrkZrA_N1*LU%4^J^ibEIX1m4_R}xwT2g+P6DZAhx#zEl?UO
zj@LYd>}$I(DQLs?`thM{S%rTmB(CET-xHF>>ZkT4b<TrvmOhIgH0KA0<9plQodq3F
z<e&`KV;E8oOONl@6-4P(uy4Q7`Bk?ntyfU!%yM!la<QpTDjBvv6<1&}U-G%^ptcQu
z)^5VKZnCy;dqL6duK0=1w{<1AWi2f{-YCl}c3W5&%_91DO_uFKB=euRg=X!5a}P?n
zlzU+#>`j0x)1e$Ln{UcE76|c~T_48e;o}bK=h~EX<Gq#7b|Dq`i&4Gib|>#|OAv;1
zb~l{6kUJHuE5z#1Qt>g4Zz_2cP2n74XHF>SJFzCqOsC}3VaH*nr9A5z2ig84efZvU
zwib7?soD8-{3EQN-R7M$Q!x&nYv6|<B+pi}*_wz<V=86b7?Ea-HM5x?tp0ljQZl(+
z33hBcemstBiU$4Pf?dkbRY%&t0E+kh4tJZpfw0{2yVAm?{Fx2-sW=JNFJC`pXdp}I
zorCboHrel*A97w69fD6sJ|5LJlWoM*;4W0#Z9b0SzB?6Vxs6a<58!JasDN*Q&!tcS
zt=>r*aS}n1Jpqg#;zU@5kLz|{#aE#-Hh{#FZvd9ef1?APn<+u#EgDfOFT5wIPs|2R
zX&kDim#<MU+eF}wY38t7-UgM(mlF8cY@nl1(<_SV_glO3K<=2mXTd}LX%wtTMFKjP
z%bS#}k|z!>Ed!q_HF@xmEHXFS>06?-@>dW)%^|Uj#!o*8tNY6&8Y-b!L3pLQKX74A
zy7z5#O|5zq%X{;oem!ppuP|vOY*)k$SqalffBV#Nu<c6DnO<WGWk&@T3SWFb#U|d0
z{5Y+ROZGA5o5$-0`+Anbu2iFl3or6Ez_0SX&|i`KV*@v*cW7w!eO=QN*SvjsVx;y$
zd2o)siQbSGiVsE_q5}IoC11E^1^OXw%<jGU)`vs0U%q2nZeS&01R%xwj_)7b;F&K_
zsb>T3g))#p(+AUO2*G*{N?FSA-Ph|?P?z@mV*hFDRvbK#9~edsv9Fk_J{!k$25l|o
ziFUG|pZ3*VJbFz=_ovj#Ko_T_y+j%HAc!uDzaX1lT09^G@xFc7NG&&)VlA?*LhUIb
z)S$4-<wCHJ6i69+5<K==l(pB1b;dv_#m}JFFBx4rpE)=t<r-Eq0NV8K<m}haw+`bR
z&%4c@Q#PF~^l(Ui5S`x{^T)`gS26u+GWh0)rjn%ge35q7-U19lzaV60@yQXZBM#Gb
zmfI124mlv$hUxYaJ^cOuKwJ6XEw>)B#gNoSnDy}XaN9Q3xXM7HymtK?`T@|tD@~p2
zw_h@K$}iL!pPpP%*jb2%kGj^z7MvKX9&nRl?W3Lk@-de_l#eiGDc`hI&Y$rXL40b%
z_XCN~1}DhpiDuL6#)X}b-6}+9q7pv%EUoMTd&7cuWT6RXvZ+Z+tA2je{%Q<-WR!^+
zyen<(d!h?;6D}_%To!I=`U#Y!`Ql~gliTzX_6c*x2a2mXuE;4^N~$-WVM)pH=uX)L
zH^p8LtT6Pwv;pt<o3L%HQlFJ1tI7PabA?H#I0S2fjJziamZF59*i&P-GtlsYSja~5
z1HHTxc_<R(-7XmUjTe{<sW2|(T+v{k4M0(QHW}~*W()BafLu}um_Kb#+(CHD_(JWE
zyey1&Y&>K=d=+cEgL?4#-8*#IHvoU#+P>9qHrqo-48J5uW_Kq3GKgaB^y`Z5*GMnQ
z*vue*y6hG^ER&L#WYF*^%@wq@iNZZfCM8y%p0CzQl~6Ne^jwk<#@?|Cl#cmd0^G(s
z2FHHHpOQ>mkfO%5%xH0uIy#U1h7}$ZqbAwUsndJlD&;<zZm(t#P`d;4->ANy_Jfq`
z!CgWsrn4h^#khB-Z>cg?UgjMFWr5=O(3VhfY_VTogWmzFSgjTyi<82o_g6(Se(U}<
z$;xp7<u=@PSLd;etq4Zb<LE#(Mna>0c#i>N;9;ZOv!m1`B|c#7%us=4DYAnQi!`=!
zc1s$~<m*9meX7&W7!=`0i{;Pp#U(`3fot01d21{+`2*E0;O-p}d~og>6Wt{;7w!C7
zrxf9ZOOfFahSgeJ?yRvt)YNuiiIYgAlW?fwHN@E<VHtg9I_YOl*$L#~JgvW)ITodg
z96!XO3(A(BZKawr3_yWhARn2Ni&kYrbz@;RK)%RJSjCC3dyNENXT@_p6aLLke%wB?
z#%g~jt+7qt))qO0q*A$@nWVK&UjKiUlLP<FVz$cgjfgqN`R?zBaZiPi?O_E(KYRW&
zM_GcKrzwziBD{G!T45#6vXB*oY2aprqkq4&@t>nc>(YVmiRRkr<W_b4b)tOjbaLzQ
z_QZdkP9A^bxBFkGlWQu&d*uFi9#6K(uqg9~-n;Fw%MGi4h_Ol}g3|h<H5iz=t=bQw
z$8;aU;FtE7Is-A|#~Y|et;PT2Y8Qpe*fc{acSCFlM`QlIHBYGX^ikyfjF)Tg0=zwE
zI5%g~o@I1s@^P=;Z$VRil%K1>+<hPQ!>fGWe2&skxw+le!&e$VOB{YQ_Kw(ItAPog
z>ZOnjKzF5>Ti(us&S!N!7c&dYaI3!!8qp%o{&p%ax@P`@e&uq|>{Pbv<CJrmpXYLh
z9T%?N7{)%C4&ZF-DGIKM8dG59Db-PGx_aHpmo6#QQbGy2*n!!OlMWR_JmK5kc%}Mt
zOR+^hnKUyT|7}P&th4y+0Nq&-gl_+;QfD9NXjd#>uv4)w`JE&|v|<%evB!soc}UiO
zdAI`I^-0bSTAZt>%T0eU5K^w-;0CF$F7e3von=SVrgXJP%&s%v`}#g}j|v$g{NLB!
zP290ELx!wn??gPaYk6X2S@l-Pyk`AOO?_Croxl4{>RS<=$Xcw?8(f0l>qYp?C&^2T
zT?c?2z6_7S@3wzVbz9^9iLRfQMB*FW$8bwxg~}D5mG^pPSk6zrciibjo0;OZCHUmF
z9r|&}P-nr;3qPHz0oylU=ie~$Ds*BkPqY?=4|04Vc1gkQ*;e%qht&HK3>v#(erf8<
zh2V^(9J%|7%J~5hn>lNMDV@v`g){mTg7r#lISk`B4Y8Z_zM5%2ziQ#|5fySz|AH7+
zn$`Z~{%&f)H1pGYM#3p(INxuTHK5e7!AtiX#QLK|LEt*~7tqmoD9+7Amw){5#j`cC
z&f3J@r#`qk{i_akP95H}`wly0y_HMqF%LYp2iy@15jD4IKbVNmzCOG$NzVk9s_$QI
zvzF=DK-RZIkb<1(o48(5=K~)(EVo;%yiLEiA4?yPmG3&eH-fX0uYO0BL3Qi0drn;-
z8eF@+OB$#LMr>bH#NP$<+xIS9-?jCByh-w(Z<h23(8tc?iM9rA*$$A8`Eko*#YcaR
F{y#CYXL0}l