From 4cb8387cc9131c20403bca6e4bc850f3b1bb0464 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Julian=20Zab=C5=82o=C5=84ski?= <s478831@labs.wmi.amu.edu.pl>
Date: Wed, 4 May 2022 12:29:36 +0200
Subject: [PATCH] fix

---
 README.md                    |   2 +-
 RNN.png                      | Bin 0 -> 11295 bytes
 S2S.png                      | Bin 0 -> 44678 bytes
 prezentacja-checkpoint.ipynb | 294 -----------------------------------
 prezentacja.ipynb            | 198 +++++++++++++++++------
 prezentacja.slides.html      | 211 +++++++++++++++++++------
 6 files changed, 309 insertions(+), 396 deletions(-)
 create mode 100644 RNN.png
 create mode 100644 S2S.png
 delete mode 100644 prezentacja-checkpoint.ipynb

diff --git a/README.md b/README.md
index fb21285..bf5cf9e 100644
--- a/README.md
+++ b/README.md
@@ -1,4 +1,4 @@
 # Seminarium
 jupyter notebook
 
- jupyter nbconvert C:\Users\User\VisualStudio\Seminarium\prezentacja.ipynb --to slides --post serve
\ No newline at end of file
+ jupyter nbconvert J:\Desktop\seminarium\Seminarium\prezentacja.ipynb --to slides --post serve
\ No newline at end of file
diff --git a/RNN.png b/RNN.png
new file mode 100644
index 0000000000000000000000000000000000000000..99b4215b6bf807ae7f693f44a3c76e07583447bc
GIT binary patch
literal 11295
zcmd6NWl&sQvo01O5FiBi5P~NV0)gPcA-D${g1ZG9+$Do0Kya4;gS$f#7~I`u2yPi*
zaE80{zTY{w>Q<c}x9a}5Rlu-kuhpw}uU`H1)4e~dsmKxFQRAVZp%Ey^OKYH^p`Qc)
z=iy=ie?`{89l!^=tA?B;TEz(ME^vWmC7~>VhE^Sie{1ppxPIs?ujh(}M%eZD7roc1
z*c=Uwvra);LJJH!%*6SiIeXJv^312Q>O%IN<p&H1zB;KIF)RoY_>_zEOYV$%sXM36
z6to^<Ac0=uLxe?K{Xw@-%dU4+T}FNC{{HtCmLx8vq~s^3mC;i_-?yv+7cASN7p`)_
z@^=it>hILwA%Ks>2c$&6h2H<$i_pT$MyJ&*j0IIGDNOpGBM%%XIx%uYbG%~Z9$Wv6
zAY%CQ3jJ_>xz!Vl^&qH<v=fXlWoKs}MbpGRmSAe|I<*uD&Kyj7X^;atB%tQu8%5w^
zyzJ}m4^o(3$ssmqu&1Jzka+q6W@Tli+K<;ZS%vn?N4CeO%(#<2?&EfiNmqE^ThG{8
zOdDbA@ce(*gld=O6-@|Y6HIup6y0zeHa(8Zd5Q*$0<7>GkFCSu9l{rjRCO~D-cP3H
zyBG++!ez@EK!+u8=3F+Q!%m6SM807{Sasha955l<OgUf<98%SaK=>jy>8RX0gcc!N
zmLWQ<k$^2LE)ZTqt0vNn0db;H6JY@6ra^VtuPi1j1ctVEoP-PG5(t-7H+FtIY5PjY
zlZ~!Eb@kIJn#Lzx<$tCtLd8nPmNl!DNLbbAynJp+FBg!W+9=*N@PcvA^3=aW2d4U5
z4{=NIde+Kaza3D9=@Tr$MI#7Kp>_MU>XTomjUskM%j;9}d@Iyd4Pw!8)I`kt^ylx4
zs7a1&76-#k%K|Gwzty??`;mDc@YITr0*wGcO2b=9(bAf4R+-#1>EtCvG*7Coe<${I
zxz}ooQaA>clw5k2_*{gK7|sdU9e2APQ|3d$-3I%+ONWRn(vg~|?rb}fd2GR(+vy#o
zZv1Inc;SnKgk~#vvv=L^WatowvenGrr<2qYacBKjz}a!o^O#$dxs{PC-|W!AJ@GC{
z^@UU4!V{;1M8U6PcfnDDqEk3TP&RxbqA7J9Ems2M?a@eR^~`_}OB((7nM<|D7BvTE
zgHp-wld>S386U3&HntQRY-JknU>;0LqBjQ9cSF$a$%Z4y2Y0U24JG5JxjArS*5RY_
z!s#w=>39P*A~JoovE13yL4QI%52UE{L8^9u%r%?A%MQL!hcA->8II7rO53C3x0c~P
zIhpYiALJGMwkV*!;t9>Lq|9%sBp$0OR$?4;!1l|2fKY@eFt^D!!6%a*jW;OK!bD&l
z-p|Ig1}(2@qm$ev#rcdrOvB>p6Fmpg<mJ6k^UH2&se=syhR~Z$nm5N*+|)Lrwwls=
z*84nRwe!iu^2J4$_=`)^<wlLftYX@uX8C1_A)Z6l%>w>6w)Bf(&io5BWaAi4l=xa=
zp9yDXBT<5z_}Z3>Y&X!@gF}j=QmQKcR6lfy2Y*ej(N@e#HhwDB8a$WugJ~X885Px_
z#h{l0Nq1tArL=)%WZhZ77AIITd$(P*NAtZ)C=0ZT5V@C7T{CQ=UC*t%TL>rS^$2ZO
zNAx>ML70_Owyh~2d;FU$JsMW(56|U{H%*-CQU$+bh;y1DJ}O@@SKl7SY~ByC&u@{{
z5R!6Zgyw_QRiQ1R{ftSB8;00tHA*7ASJTn4a=*y3ADdN@d!VffmwLZ(e5@>d^h62~
zOEn}kEM8yJs;>xJw3fY(j#9tsedWZwaTc+=l!au5-3!+^Nz2=G1siDR`mqpM8OuJ+
z&4POtrtHSXHU*`pmdfoej8OH?(HQ+%`JhK~l-q6L#200yz3yjg`K9H;6JcEZdIgJ#
zHdBD`k^lo1g~7-U>P~=iiCI3q>lHX1fM81Lvea-heUZP6NgEn=krnd0A#Ib)&-Ma`
z<tK#=Ht6cLZl}=1;&m>)x*ys9t%tH57!vySV39`HK~nLx<5b@s5rc2AIx6}O-+tQA
zm#|}S>%7bh=XF`MX?U$Ak;)_#6c>lP>t5#Xf0ZicH`!YYEcS?clh5<W5D0ovUmT|l
zxgYXP@C+7zq%~IL48OL{FWs1e4BnLy(9NK}#i;xw<aSp}sx?1|i3w3Uh*^GQpG+(u
zTA3VvrdGrZQoWb_HqSpW&MpFDHOC&x5Ih~vmkU5$S^4>i(!UQ31LQ}7HKt*zkTbd7
z_?mc7?y9m0V8yRzaR!{(D(`m+3(LwKO-c$(KgPS6ZXJd{5O50hwjQ)u9DDJ?`sdg0
zX}Qrt+IPHS6Ys0j<9|l)^O!kI8C_FRrPwYmK#M8UhDDNt$=9yS6tB>~zn>UbLBHzW
zMdMX2ZAq<RhB1D8;^+LuVK1gR1|OFLopFYPgTwFoOfUc?Lf<<g`GKLaLv=CUqOuhF
z(Y`V_7kZHJ=K9yomFHf(<pXQB9oxA%ySbR1Bk0e!>!vin`RWAryba!_LG_ShGdkKw
z&CUnY;aA4RC&SS_F`wM<R1ug=beo%-%1TOZZf*sIh2x|A{QMieaihb-<rNh@Oq@S)
z_cwILRiOJM?+}kd3&77#xt3K$K8PK@P7mWWF8H0yLVM47s$$f$Nud{W7yar8X4jjc
zz*(*=_94l)B(~A2KFurH&N9O%NQM0E^@A^85%i_Hl`9Pn=_MsL^78T)7R>a@>r5|T
zBqf(VJ?j&p+4q(->lai-j?A>z?kX};9=ofqdY<yBCK=9p%26FxwjMuYLg7}*#vj*8
zL6VPdV$w#l#XJ4)u4l>(XapRiV`KUG_}n}_@7E(3o?mg_KU73W7YmiVLmU$6w5fIQ
zv(E9YM9Jyi^&$DF9e#yAyuVq!Pm$u#3WZ*JU?0Z}d#bxUYK044ULrQC&3ds3X^RUB
z-vpp;TUvx}1D`W$!5VoBQnhj}&x^!1Z{9FA6NTM4=mdpsY|Q9Zf|N6bw2h4!`Y}8o
z61ZsGwvmM8>V;{3UR5L6ugkrZr56Fe^1XB>AxVJ4O_!K?A`{>Wynt$2gtJal8N!|;
zb8~Om*>~@5E}9f^|K214k&YG=6kzFzF*4@n=AwHlpur2k6Mu2I?!2&{f*TYX8v2M{
z<Pt~}niR+Wg7Ps$F2}A0b$3HV%_slKap6BeZKI5cBw`TD*}v=XyTMa<oC6k{;DRx!
z5%)$@ZES2XF=qb^{uk11q!74nG2$Kvb2&LVaSY;cuAk@tpdV_Bz)!bFndD-l<KlW9
z-v0;WiyUA~Vt&`8l$4jto~!E(YiK|!AVwMoM(%xUZoUe_#Bok<`Hv1)d2nJthpCB)
z7M}|)M|@lW{nu3yk&{KL1_lQCnEZ~Kq`+5<G4S!!?}>?t&E-jyJyZG#`Ep9ZvIo5k
z0o_b<LV))W5)i-$(5qdyRf=_eLS&;mIiY|Lw6?U6las^wW#(JlIhCO2*fDcyZ|wfy
z;$fJucjV@-$9~Yz)+Qq-52u0SEL<UQ8%FkLE48(??@>2SOXOJAt(Rnx48@>g+a^ZK
zKX=T1%IW+vGBWGfG7`z`;FNA6mn^N|yDDV>&J4IfZIQ7F7sso;4xgUT(Y1RV>M!N+
z6>I!6TbXQNaIn?$IF{z!b}%1=Mcc-P-BFYJv!+Nq6V%bxmcub2p{=bANWs>z3njiu
zA|<uE9H4*B4-Etb1#?kCG-K3&LF5u_;oYB`lx?0XBHZ2GPfkvXTd1k23s6=21K&95
z>FJv!jMQS(3f0=xri(9Vq-JJjGOxsk(UpKfTH4y<1sYmfiJ|ux`Zc7TcC%c}%*<Lw
zXUF{stS;OSuQuYut@)@O)hM*`d%SZO)<Mv<g9Aeyop73WR_a<>1!ZN?EM(W7_o%z3
z<QDlP_^z7}$4hu=Wd|NUepmbRf`$eiRaGK^Vtgnul(g2o*rO?#BPZl6fmPKRbLRzc
zX8JPJaRUMZmK2Si%Og7PQNmm=qJZPE-rI-W@}se_v6ldHa<Zvt$Vy%ue93ciirA2|
zvDpQDV~u}|T_gZStI9&Ink@#4Qzw0>;VU5#l%JoUoUEJ7W#sGYn~{+rbRWn05JJML
zqK3l4(#Dy_)rQ*5RutP7d;BZ!V4nBo%a==NLQ%X{BRQW#S;*o88yf|#n~RHa#bwo>
z@5>Uvhtmb{@bH=>^gHg(7k~dYU&;}JKp>UIojn9lHFb4#<1|vn<repZFO|GOaE1OV
zlP>@3T~=UJqLwcRv@(_}<$}56_70I4Mhy|*=kM@5rgv1Md(Yv&P-ml<%mpLJnM8+a
zlH;+V0IToq?X}(N;!Fj`Ly9JMtwyu{_ig{1^TQ`6rR7RMT62Bv&G&FztJ~qf=86H~
zUqFCD;5phJJL+U}kc*3}2|W@+O~it+6)x^~4TNlMP0f(SCum60PTOgYhgV&yrub}G
zHD8Png@uI;4Gm3<RBEb-e7U$4;8-}K<)M+P>h)Dz!lEct)A<Pmo@4H@)6*4ja&S0f
zK7Nn`9+32aMMOj#&)4Dr_aN*jVE3eWAEWj$#xWp8FU)n@LPEL#)452!&PjofW(sd_
zZidkWMxp_as=<i@pySi0PdhKy8qDAEz$y)!U-R$~3!o9B19J;>Zo49gVxps+rA&dj
zMXK2yrZBuEG-<5A`syPDoK^r3>L&OcWC~#r_Bd#8SOjz;{;QKMOjlPIh!Gp%-GEpi
zvayMaGyJz$UuET(=eO0Yglylz3=9k>s9cNzw0{;9In^~NIqQuB10n&t1OD@i;GJ|J
zoE^V}EG{xKa<18x)lnXsRMXel`4158|5@)c_GH#LU^9}rRHs~W8b*NFhzd~g0La$e
zHL{D~XHY|wanL0r6I1P{&G$=>dB;S0nfiFg$Hx~Noq7r0pTc4IB1&*MV6B-7qbM5F
z9`XCz5IMESxQd9=t>FaCX}6eYlE$X!6#Ijt(Q$T`uSuG7RVDxj)OlA#zqPdm*rmq-
zJ&1*TL8y11IDxYHCLikJ29X+jBzH<1_jJ_v4mk@@!6peCol4NMvczYzvF_)l!X@U~
zQxq@tL;qcAetuq*WP7+Ag|#l~$V4(zX&fh4<?eB~uz7UEHMbfl7ePMz6+v5xGvV=H
zax;Y72NoA;1lR~#OwTD3hIp*MG;ELvc&`5n12`x;8z}l${$H0Pee&c92?^5=@mnka
z7H-*5fH#odxaGe24?+A>490qbCG>TBgG<8hZ{!bv$c2W5IV?310}<mt^lybM4t#!)
z3LwN%v#YrOZDgwu8gT1R`K4x&is0L~YN*QN|H7kWIB~!oQr8JXtYh{k`ah$<&o7Ra
zTPa9M*Z5`r!#yWKFeV^&2z?dc<2zW(KE(lWsDmtyjh&r`m)A#X_nwdRKSY!PV)f}@
zj>r58$os^-POv=dNdVN@jf4{eo2a0m&?F)AUyo7_?gl_&a&qzwAKzFf05ZMpB7cF8
zPU|*`f;&1o`uuP)VlfdDkX-!z@(eQZqr(h`tkdc^!}Y+Y7k!X**>O)oq4MqFAq9;`
zMNqq|BXY6t@?)GuXOY1EZ+64RXI~03+C_JNE5@pTCWv9YS9ceyW@cvfb#<?Kc^O2#
zLw&pg<zQ0nbCVJ<CIDB|tN&up;#9C4tgFD32uT_+0GlhH&GkMX5o9I2P?{q%o}LPc
z_dgBH3^%^dP9F|wo_3eQ-ZL2KNS-N6C>om2mf=;V^@*zO?D7)l`kBDLw=I})XROjo
zHu`&|-KW35zmKln=K{4`l)cj77sc?j>C-0O0$%ThGgwDts+65hS0t~gDMdD4TUU33
zU(oC7gc<djAd8GB0m?q5<@=*zk*Nbj|CX1vW=(KUsntW+*7fR4IQoHxN(OymLxHX(
zXy~(O%(OCh0N-Bvw;<lx@MFKXO(iYIMMsOPPFwL}itoQC=2}WkWOe1oq67s!?T)Ed
zvA4BGY&&ck_WV=4*VLT%;sf<)eqErx>My?%z3+}}@2s^PikdqNBN@pj!5KlaZqZR2
zopG>pSKdSH%BK0BGjKhkwk}_>Td$(?eA{pd!+noX+S}j1I9wF?_ydiAAf~CQX)u|a
zx4z@wG7An)#K6G&t@ua}_N)x#b)v`lNtHwAkdUW^{LIlqffRhtVi1P9h$#@cTUuIn
z#-bLVV+|L7`SJw_f3?{{b&7~rB`LXia02(~7t3Ygqs|P!3k$?Pj(xs}*%0G($_@IS
zi&!%s{kqPF&&&Kk)_u~+5a~Lu=|Xi#obb`YsS=v6q;+u^8ZBWPj*88HLbo)m<vkR|
z?n6Umk#DIifE$gZ<keSI9Q0W6J6Eo4nD|mBB&*yz%*6q71qZxg=3Hgi|GurH0SmiN
z^yZp16bAp97kD|#Kx|U6%4!&5>BNJ0WmkV}R>Mk_UNT5|wGg}bO$+(ygvcw`j!S9t
zkWZot;sk&^xecA5k)|f^>(@{20zLXFJFee@K>jmjdPA<a9&A}<$ZI}3vofezN_}%K
zzn{);vMrdHT)Ud8)Cp=3u?fz1kW+)mk&=_=l~!Ql5U%mhzvIQrEh{UlsEB1DJIyPg
zgA`pB*vb_bgE0wOA9V5@<u7|p4I1@&!2-ZbPAIA+#omt-GFz~bvzMAv=6*N#w_G;r
zDS7S#p?L7!S^p!G0&AWd(B7s9js9BTMmL$^biUY%?0FQujI18}jZ?<JTVm@-#lv!;
zj})%?LsiflNBbGld-ow4uX~!uITi_ba{4eCB3I4PrTZ}wk{WDV?Bk3PhdW~8o{8@X
zaz>^z^*k}N)&!azRAyA)grEDBS-6?JueE5BlZQBAHT&<+l-JbM4ADIadJ#9mp!oj%
zLN55z%%I6QNu2<Pg2A}!y}kcr!)}_;-OORGrWwK6P1viE*}3F(dUOtYbsZf`b8|kg
zli%DZr+hjb<yr}h2cxXPj>|1P0hFTNbv|cWu@ZJvwyGSKs@LV7W%$9|q45H?h8RSz
zoVxB7S5<3oGz^z{1V)-uQzjZ`)|s8?8)hYO{s^d>9k@S>(lgT)aH}E~3LL$4`QB2N
z)MDZ3@XU4(Pe@&kDc?SZq=HAzV=wseW{$&<HcuCni~4S1r*N=MHoC^mQ(Gs#wjZio
z=`U^qjXNI;sfE&CAR1g&FJFf(UNb#YLb4jlTNto@?b+)9QgM92*)JfyJrX`US|Zky
zUpgZ2`C6-?WuhR%N+@*TR!~G_iTQQx2D0u^>b`dHxhz>!Do;k($5onrOG7Z<E51)(
z*Z{h4bz)&}Ushd>_!)u&fJ~@Q*pCn#B0Ma-7B+L0=TuZwKyu8=T4XVp<aatOAhKGV
zdr<Jcf3V?<ev^{eW`?&GW@;#T7-t{|_4(##v(`XQC-a%P@~}q3Ld$bV+G;o#$7<tE
z8}6=@)U>kT7v+2c2H#7M19eL%eDuDn_$ZOaxcF$8mGMrw9>-VMUe{uxemmrP@A33q
zqWA0GsgLG!>u<jE_xFvsHkFG%`?X}ha7d1`4Hy9TNchD;Pth54FEFQx4WV`mWD#Tj
zJ8<}G$EBv}Qe6rPiu?0Lr#=>6zoqvT-ggn_vTwF41;u{ax%|?mASY(#fbj~tz(dp1
z(=#)d0XS(!1?27U-2>?*ZFXK1IbR7W8CiXOJ#Yx=#eDk?egkwIstiqB-sR^h+Scl%
z;}P;+>n9qf*m&|#<95!VQZBkq2mb@bXcMgu9dB!;y?|THx&Ky@^qcavpfIN~P0P*7
zS2{^%hWX1Pho+`8W-unL!83zSEaWFQJyc5aUTm|=%yjn;_vL#v)xFe^WdgEtUfW2!
ziU4dJ1FELh^yi7MXZjuP9>G+%^DYq&vlhDWEP2g&nw-{e&LH?_IcJ|me0fSqNLLq{
zO?t_=&ym*&b2JnUOEh{&Q+}CBaQF6T_Sc<|;NUT;yW6W%08>X%tPs@IHba4yTaAc$
z-xq*lw)tETia!+^qw3n7EE2TT?IlwUQutOQ+{NDff`Hv$$?X+ehnKr}q0!s*O-HsO
zvr03QvZECPIV!pztE26w_R@wq@ijq7neC9=&d-@bLV-ov0?#)`-c0G}HXP)Z<QLBS
zz`9QwMd*GwiN<BDvE*A9J8b7Tu8gkCDvdasK1cLlMk~7voQU21nRRb2>m-8(q&4s;
zNJ;4Yv{+9Ww%I~e)93opZ6($>Wp^=W=3LG^4u!+oK+&o_O@}ec$h4uikD|ihG|m2Z
z-tXVz36<z-XzWcEjlyA3s#zj>>gwb5cDg+d#BdUo-`7=DRRDdTot@n%`iv8-CbG*q
zeVOUXtd9^W;A(C%)j(vsLRUvS5XKw`W{$*pienWKL(y`$rvBAnq5(KUsS~Mt`?Y2+
z;l5$YG5zY*sn&G&Ee@|!z<ch1*Uf7@r}^W*H@{M&1pCWg=KZNgP1yxJ&kBHToWLu=
z^pZ#3^_+nIvY<Bb*X(TI<hcR7&DGWQ`>G(292b4KJN_oZkrg37RuKOV7WH)Zn*a{$
z^VHNiu%oGAtGhm1E|^$a3%RERa~&&Gr1<UThT`Jn6n0+wQEN2{Sr1Tt@fMhsmj)C}
zR1~J$|4cDseu9FpzUQL>X%z1>XfiHY+-*fpPGAG-Y*%xM*&pBme@Fk@J&068^u{qb
zSzG@ga0y$2PynrnfBFc-xp0NQiQxa)NYE6_0e|{i)&;5?09*O=KXxu`NINYC$$PlX
ze5Yk-nUG{waVgU8IRH)pG)FYl)X<Q&o&^Pe6MNjcwY8riuSyo^*h}byJ=Taxv3_aX
z0xm2LVLo2%1bza5{*7Z475Gvb-T#3pHk1@f0i_1MFulZ=GkrkM!h@*{1*%himw(V3
zVUI*SFeLw05CEn*=2FwvM%##vz}Ci_#@8ljM9=9mZ1*Nph5{+@I)CBUw=mv&UkO7)
zT1QOy@bCTvb%5$S)1V0rHMwkz72pvO0aX|#JqS>n5<GRH2Pi@H{c%|!Si{)3vbZ>M
zFAN%P8{vbVg9#ia6@WqumI$pA*?*?(*YH(QA=IQA7#N5kr2Ew#*!1bonuHd+7Pr<L
zE#W16G`f$YS)wrswgA5UOaBiK4%BQ?&-SMAFEHTmKYn~#k(5XZr8{~MNVvAYA5{dD
zYHDkFj|#b-6kHnnUp=5mf&sh#$1L#jC(6n)H3tvgix$X@y4)ZVN&T?Y8Nl$Dg3BYU
zMlw?QY&7ic?Pn{Dij-1={kdMggvFu5aiD}iVIi@ujt@$GO{M)<92?Eo1-YlaRdBsm
zS_$9S7#{_8pcg0(^tY&ofKRt#f`c&}(ahhet0R#6vrXt|!8rzvj!#a)@YO_~9tFbD
zgD?R)m;;_bhu~lVji{PHxB?Cv{o}U)x5R+3(&d2Pt^Fm1Y*}vqUk%<SJlV()bdB#<
zl9UVv%;ESY2}qy*mPw#>g!(MP!ZhL$bvfR!-rfytK?P-HnH$0iy#A8P%F&^rZl~xv
zx7`Wmzx=yTL0@0r$cV1u{eGP)0$`@`@$qRCP7*8Ty{U0T)7v}Gjlj3Hzli8Xp8B+_
z6{~aC9j`&Z&~9ZnZ-YY$@ZYdZ17>y6)-PIyc-Ct}Md{Iq9hA<@y|mh@E9?&u?;@=?
zY>SkbVFr}_E^QL>+r9noiAh!G|7jJ9NUs59kbe&*5i$6`JAqPI^#9#4d>jPy3jgga
z{?Eg+R5Whe!y3feW9EG>hlsNi1nH+~GkY-yk{7~Sp-3vLn<|*p(&T`*5D#j$)3ejJ
z$tuc!<#hASpK@>>h~+@kcPo<HE@S>iz5M{dqGKn~rP;}n<_nT`^9TVsE1xIMcRq20
z5qAb>0$~e~2!-j_nXgKO6cC-?Q#jKm=6o&(^9cxJuNQn`uixx*pY=Pd<;m&M4<XA;
z{uPzVewJog!iO^O9PwZ2>Y~2XxvgX;mu^q<{yTH|vx{smV*^UkI1i+Q<GTCi;I2I4
z##`tOIxPQe7{@?-?O<ClY(=_V<n7J8qyxb2Y`>1AXl0gP)=ZChHD_&;3!3+&_}Bip
zQhB&B4sjQ4D0daEAES5Y8gb*`mc%IxgwW3VrSA?Ei7%~J`^ESS(_hDenKTK$3F;AA
zQICS6ZdG2?5KSiIM>m*r8F$QhjZ4=a=$%uXAYR8^fE}|eQ-xOC6v8)R%uvHKu4Ap{
zpHEYzlxA;?^>#BlCvnP8<&!q%nO3fB8U*)6#t#IOOw#Gi8ct}EjZ<l0yIzjXlO{b9
zDdT1Ha-#-&8hQt#?pHYCUS+CP^6B=~HZeC&GXBi?hmuJ1o7mn;D$&AMT^^3nyH92o
zE9W<>uMCn~TSaGB7f9(HUD#}!u^o4t;C3E9j#oXKTjaBw2}8W!7cxb<DJLNAT#GZ!
zB*@oHdtnV^A8LE;E?(>A<pkBs`Nh51aj|IJjs=^UX0(zZpXq?6612HX{nFdjMxDIQ
zVm>F?xafE-QYt6*`s}0^=@rHbm`x9s+K4uRhUy<YMxHS5ZMvOD5~*{0JLEF(ptd+W
zU6^|qI@z^WT?aFXj{lhO+nsxDzH~2tAsaGk;ic35LZBRR8N)+kM2=OwJArgfZhk#-
zgLj9=d$84$LX3R2g5nw-UTs-fK5P~nZ>T~H^M5Y^T^hN@gtF1IXj`d3Rdm|!k0cgk
zEg=<dhYC{eKE;X`S$mN}L}8Otd=@Wh->}N;JfP~QyreRIaU!KC6xSey8I<MEIj=Rn
z_**?zvmhZJk`Wz2<bbJIw)3{(#fRSaB~iKU`t~hm3c=N1D>S5wK~4F2v6&$M4rApk
zgV2h$1o4-SDIJ|zbuuYPqjRLa3w>#k2vdZOLT91lz@1g=3^`cZmL2DYeLt10*GkCu
zu)_ABSWqlcwy3`;rrM#7-+S{WeC*M+!7CrGZt8uUlohT&w#~P9?#u1d7DhhfrcySs
zwy%Fq+9#b&HmDug!gJ`;k8YcOa1P4o4NAWBjsykZ3mWN$#OI412RM{F<&)d9Ssg0f
zn24vni=koG*jctGh-f>RIN&dL^BME4j^#w5Gz>`6+Dgk8cAP_UXFgFM^&~^Q8;X<r
zVb22U3QE3#9&TKcerO$Keg|@ZQ42SCulofSQ^aUj(s%kFPbtcca7OZT$xo^o@mSQ$
zMS0IVmr}S&UOeC1oM{6sr^j{DEfP|_n;MKS2U)({gMG--<~A~{q*7s4uoRPBCOu2M
zj?gm{lXL=ccabAcEF8XmvqbJ8QW&|fzYb^U?mWo@r)Wv0X=UI)CJus6FUM5mYA$b}
zC$rGlo$*<4fL0*kD^K>y<~M|-HT`cJ9bb~g`CD(w##|^V78Zvc)!!^h@@a3~1-=gx
z@*{64KF?gg{aC({qMlUPP0m@AUeZkYjqFCCyy!<lD}1`(rQ{1v$emN6{*Ew>A)9Gi
zL+@hd9iNu&feD?_*uI{bQ)zki%T*!=u0aKhOA%eiL7$2foq8He*EsONa9G1vv+{YT
z+NCsy3diF5WG|?HpIKXwVIM7pmXW^hJWrINK|6A)Dg0UIw|2Llk*cg!1D{`}0_PT3
zd0-~yZ%&@C&9!QmQ!5Ix5{?z86qAc7G``WObH^q5Y^3>i>Qt2IA%#|4GqYQc&E-rz
zunp5p^W+2=_bjv_BO$RkHG5%uBl@c7nKvta#mU+-E@(WcaEi#}pVR?KxMTHK)Hpir
zzMs9DT@12RrUuXH5m4v$xr&ncJ+cozeXeamzNg;n4ldsTM!jYp?w;5#`B8%?*5bJw
zP;J2yEs3=Jdfz44<6|C&OL|NRV{{CMJ4DP^iLoDk?)hbwC&~Z`s9G9Pl3nPQUg?oW
zhiTPEA)a#9q4qgxw_{&XO6u-cENgc}tP_ApD@z`C;t^!rqjH5lal9TPTS)1%=^KbP
z81T4h9rH=5pNPDwDE=abtuMZL<E-cZ^-kI^YzH>%lqSGV?+|RsHx}=XaddB1Mcy*i
zbXs4Z_0<E#(~}$QLp|%<ExOc;OV>qT?oRn?DBO=Kh;3y`vaxtxY|E9+aMt6)Mua{}
zx}ElFo+fUBb+5m8f8Be2$1@?{CO6s7FN{Q;+dU5?D*&FSp+cTWkIID1=Yi}!Ib7P8
zJ|IsvYi^(6<bh}}1ZXZ~E!)dJT&QA3de1*bPB89GUnXvuEYch^{AtfHrRVqw0Gy1p
zbHZM+ZcUNp2l0g*`aNEqyxFC?Xk5}&$)!iIbOK4IxDQQzPM@RVRSiAs3MSdwej~vo
zWb~x>*6ktf->*+-XEoE8QUx0qTiqI(vx2-WJuVW2+o2jSrwp6e{Dm|TGXD4(iPTP0
z1y3%TqGb2>uV@%XVVol0_WeWLqq0+76QZN9Q?jFZYv=~p$dWfNe3baLD67%3jc0#*
z9T^@CR9rH3xx0r9rEXsqH#awi8x_2SR&<oFP{C~j{~B}5{iEt+>rS3Y?MC2ZMZin*
z!6SZ;4r7Wz(V9fEtQE;5W(taJ=ED{AD{TicVJn)$q&TBHvAwnlH@h$0blPcEkIq>3
zU=~gpBTF<c?UC(Y)w!qUYe4h@C!Z<ey2_Ga{cp|kKubxh#|l&U-@*}zX(B@TAbTIC
zQV=-I9yuZ2Max!U<NwQ@d(pf2IVU~Kw~wsT>XlfkcbIZME5`ODU)P2@qMO({(vueS
z;pf+tG%c;1>_c;Fdaoo$a~kz5J_k4i>3&a$f2N5LB#aeG6CQg*yf}~!HCPi=QDuCw
zoLY^_U%}tmBQ6>7k#au^8KTObBfWOq%P(282h*{4xQK><K7WT0>v36&UUYh%>Ff;T
zUiv6H3|rgMP*QwDCvwfaPO+|7Ts#;7566k#x@(2dCX&qNv?uSMoR$0WA$^s(c;9a}
zda@~ZCR5e&N6$vDoCxr<*bd<H6Q|#-j2mkceol9xiB(_bCmq$73hI?_S)b3#BI>B6
zd2BJ*5&egB6sr5Q6y)aCY+F>G)tarIMv@S%Wv)f#W2WM%)1IcYqE)$p?iJ!5?UQO`
zQ6|3T*7n)KX_6fIx+V6!rw6&_ox9BRMVtG)o#zW>0FpqMYIfWnX}cg~&`!xJyp(id
zTB@sR2#DyF+oeG^N((a@v}5QYPd-lX<NGinDyWpzM3gurRG|5w&*<US^2Y&YH1ylu
zf2|H01x<&sSlKbEY95e3Hj98DoI~>CAa$`{gX=pat#ew!iKGzqI7_{heFJ@*1y3Zq
zO-dcFp3x8?9lcrbbkZhRK-e-@{@RKThnuZ=^md}nfh*MEc(3zojpk=I^hEmQH9-r%
z41Z1;iq^F*1#Us)<dB(6XV&kR%U=F=*EYX~&DW|l>am{~>=rHz2kZQr;%VGuxZj~R
zE6<TlXq-(}7a0FAd|CIy$sn@9w2|v*<1m2pWRX|jSXF&tNLod+I^DR=U8edznQXNV
zJKd-Y8G^kR)@p0H`Kp4GhDik0SVkfA-BO61)3$FRy-yI!+KY`xrOB2E{CJ}*rR8&V
z=^EcOe&$FMs}LL1wqKZX5V{W?GEm*t;TC8>)|KhjwtscFC5uFOe$1q*4&E<}S|$BQ
zK>!yY9WQ}CI&iJZ^n>ny^JI(W6GVFVlOStK*a$CAR={nnr69u2Ah?OfLl2FZYWc_I
z@Wn?`2py9dqS(uf`8AIq#7`do;;pLD5&}mG_0jRHMgxDBs(R8(qdu1#uLz3Mcflu@
zPt8P=33J|DuJ*ZidW24<dj{k&j3SCf_U`{w9WUowze^W|oP3Pa?e}gRjDF)R*8K+X
zht6hLJ~%<noiQ1if7`HnvrTX~hb0H~@cLE{HumN07{;Q-o2|#I^>XgM7{b25kLG<b
zBPD~H6=+Vjzwibx-5>djg90!&!S7&1S+G@~B}+dI>8ZiYV%;AxhH-XsZ@+`h+A-+q
z&s%*+#iK-t&cVrL0GmJ~z}=QI>~gq5%L8RizZB~<4KLOe=KrZ!Tz<)qnE)X+QR3jy
z5UK1;!CM!U)D{^PrU{)5{A(Z)c+O>@UDN@^OAo)LJcfy%xC)3_a>2()_)dfyUwql$
z^a1jLjs95yt{mdAiA*pV5wXtyB~J=11WTnIT$N`%9+y?f#gBvNCLaGn7y?u*l0}gj
zR?6<V2@d%(5|`Bwy+&qiU{G4h9-_j?_-qttV>xT<Z*LAJ(~bs5KKIstZN2(an^6#W
zD}r0d^dB~d!;S)~a7`TLlVUJT;Dy-d82>jf#Hyo??@<h&ly%=TSJBo2)6o=URHQ2;
HKL-CVNC!E=

literal 0
HcmV?d00001

diff --git a/S2S.png b/S2S.png
new file mode 100644
index 0000000000000000000000000000000000000000..1a6d618e30cacd8a0f872036800cf43ea196476b
GIT binary patch
literal 44678
zcmdRW1y>wh(=8AzgA*jU>);w(f@^Sx;0{58g)q3g1a}Ya4hgQo-QC>=zx}+w;(oJW
z4byAr)2F&>pE^~0ci0ytX*3Wa2nGfQO;$!i6$S=w3kC+(7#R*Y(yFK(0KCDvs7i~&
zRQ@AA1U|r9iYbc0z|_Q`JR2hbpWlCz(Q$!+LGStR7k0q0#0&<el1NrUOx?rqWEDXa
zXL%mxB9pmQO;ir+uegMSBzzn2!fg}yW9d%U7@AY!8kz?OhYbfO=k9o(zDn?w;XSPT
z<+E_teV2=+Y07FQpUeK53l>YVJTMpx45qNeGX(a(zXDMM8UA-f6|?BScTpL!l>aUo
zh62I(zv;jP;{2cd0);*cJ}VrIFAQP;)`t%X85tQWjGUYwUL4)rc+$uE$<)IoC>a<T
zf0vZdgTeW4eO6Xh_4doZ<O;VifdBIHXDhtSZA+BO&kga=WVlI*eoa=O9T^%L8XUw7
zn+Qb+M9)-;2@QqP!bGY34yL*d3TCw`Fk$*LI4Gl_fNF#W1Ji|IEZ38(^=rXch>7Xs
zc)lt(H<yo(58B!q7aLpM*vP}lX{#n@U_c^<>CFwJh>E5t><!%*x_^E7!PVZcr5=Py
zvy)VSqd|?6NV9_zh66G*3hy!dwPT*VV}@sLnYd$t)lBRyiMzG6)!}t3-t|E5;^Km=
z0}KpnEN76LoJ2uEVQ6SLo-J~Jch_&54ihJnAp+?vEG!HP3i9)Vm0-@9%=Npivm6%`
z6#R{**lUKXzVs2?z^*Fwc1Sd2eJI07A2Gs+DNFr~)v#H;SpS@{l`-0G^gh)IE0uk{
zc8ikJr27z|1xh8COCdj=C1h`B*B?!?Tw^NTj_2EjBt(aYi;IhclbD#86LEm`QHU0>
zGH7aQYJzfYlw9F~<VKg1jqPGGU%FT+v%AbkUPY=sA<Kl$-dd55GN_m40~SY^3@wEf
ztbv{k-445~PG6vr&Q0%4Uf@hdWQO=jSl48!TJif2w5TNf^?FSX7t1gZN`C&wzX3?-
z=;-U6KYPlcGC1stA5chKY%3}|G?u={s%n0Y8HBd>_Xs>%7MV+gyn(Ldp$Dlr#D6TN
zF28&C_xGz6DWnQ|I=8u<UNFKC5{2~0%FD|uD1^QdkR)TiFkD;@-2d&uR=b?NrKX`V
zH!<-yR6RO8JiIzQILOIG?Ih$vL{|G97Z>Mry{D+Br{~=Qy1^8#3>l7Xq&CGh`N7TZ
zIn-)st{HUpS`VkG56jT_g|bqViJbp!EaV*8%D6X{&g-z!?DO&{D<iX^0W-5VUuATE
zw)#0gAt#5tJzSJLo<Y~(M{aJI8;0%JzdbW{CvQm?<4J2-BN1Ak3yIIajKp4v@je)d
zk-MHQ;YMi{DW(?zcU%j0g8bk;i;$2I^myM_bet|GD+F=g!_{7YX=!PDAcmKlTQVT!
z?sPd}$jlp|L<CuFn1+TXQv~w*$L}4~S@POzNa>i-x{`{4W^PB3@DQOq#=5}QwN(0*
zfn9|2LKpVs>E5W;Z0KMzpFyJ)8RiQuaBHEL+cBg<jK}XIFlAfuULr#j*@@Y7^_5`D
zo{z0O`QjY2rQ5wQ$u|D&Zludm#?dOK)w`W8ovpT+a>L9ZBn&-2(_9=z{J0AlclpG9
z7l$o@OJ#k?P+~p&+h#cLElDdeHW7;|ZsMBiTrPe@Y&<~_mA{NFCM#1XF(Wn^C;R$X
zT1i&L)8mE6#KhzP>&G8oSWQjM+S*zV<`N=gHIMZ}+xg1ebqa5tKf1O<VFtU3dh%an
zAQb_~5sDYeIUS+N&*SWbRoK<vkLQ9=@Q;p;wnktso>chw`CsqXA)TF_9=NcRJjbS7
zcHxs7PVwQ49+?*m!*VgMKg0XMVcFgjVB>PY-<?-mzvoLuiNC`yNA#7OIQJG6{nH07
zc3k)H)zXmGH1s~$iO&2{?=Y*YL=q)aZ`MmgXk71YEX>SD!OO?a%OHGjsUJOle118p
z19-^J?(T><Uh}{52nfHz!^8EyKLW~HyJrAvGCzl$Lt~47c;q+k>0B~(Gs0DdXzzX@
zPKJD#s@>z+vgC@2(@-p`9P3N6fv;>VMxAz!j&Iuo80Eo?G`1YZor|NmS7~`TBY6W)
z%1z_&BCXx@{ar_okB<O$Nkgh*gAft{z;t%=X>oW-D=?aMXQ?5u{`4)Q##oxW<8Ihw
zYmD%NSrMK@aDfzjvCl>5k8`|`c}*^tz0t$z;``%yjIE?~sB}-F`}rCg@h40p`tM%t
z-8W|-XKcy|MQ;8~pF_^$u6YJ1M|Mie!tL=qopKhcko}w30mgsRB;s|53=8{fW5(?|
zW?6B>;ZvDXU&6yNjdJ}tOaIN|Sl!22P}B75*Ka$6aqOo3T#|G~<cuH`6jr0odWTgn
z@0L94k+G1-^M#r=Db4|!F27g751p0>9=$CBQjZ-`eYOarTc5Xap(a|l<3S7naL&!m
zLE*AX@h;Eo#@v4ACAa(a?+8DoK_&uTPc4NF`y5r;y>5>dYD_neU(b<IQT4lgI$~ll
z8W(;ic#c0bo+3t8K4F>|_Z!#bts1s>Idjo4VV2bZ?$^}B%gNb;_g4f?NG})Z;{Aty
zOLOyWjlxG%cJ21Q-!2%;*s=?(Q=f<(Zsuor$EjFIFS^U*;)ZtKUY{_cv~u3Ub0G8z
z?~#!Y7VGFb|G0k(Jo(+ph^-zck_nN|J@-t0gQ}5sDv$#!ntr>76ZB#ev%KCeEG$f~
z(XPe)oV1a?xOCoeHTUi_U48pFiIA#fr`L&(x4a>?kBIH#%^?y7>CD_*Kb{!MuZ)a#
z>**o^_p^<W!oBiiZqM9QH$56}oq`;r!*gTnDzXqGL2W*)`FI+6%}RZ1JG)5Mx8>I`
zw%vgkGQe@5aO-1-a?@?#`f4gq50HI?eC750l|IwU<87PI3qSFI9`#M~kB6&WAi{vP
zA~!<ans20VdCY=|PEF5QRSPfR_3(~h9R=xQP<xKPoE)g*QNQ{tY~VJthKEOUON*zI
z6MN?<uT9ZV+Rv9+E<KrqcaM7LPl>L_>b?=5=?(jMZDvZSq@(K{*GY|DwHhwig2pVA
zGx=|UMF+~^_sywx>p3f&8JdE3Q)jk4!gN1)HIrAYoNd_Z$`BcEC9A5cI^j|m5rLvB
zC9~t8mVBqv`B~hMx{84)=M&xZ{mSe3{kY2_ldIE0wXqg3|JmQTjd=B_XJ=;s91h?^
z6K4uy*3%a%VwB{g<mx!XNsV5-okNI#Oa5_p((&W|bwn$jnNPvV#f6CND@>;-`9e&b
z(X7@3*`dMF2R*JMRU2Wnw{KCAkrUb#pF_|WC^4-Mq+85};!o_>PCb(vNu)_TN*&Dn
zD;|prd{CU1xag>8QJKMvjE2qMjJJBj78uE=NJV_~5A!oxhD~PIeBxE#dT*4O@b*{h
z9GCG0Ss7TAhih$Sf3LQ=gXxjwEA#heIC^jhql(JP(trI5W_J5R=d+SSr}WU27G$k(
zCN^LH&T-@%_H?}qGV+U=4PC}{cW)0slJ<H!kO?~^VD`5V^ApSXdCdXWxmp`s<>CjZ
ztY(pNw$I)ufv2aZ9Wy51>+35p^L~6N&k|Zz=Ze3=>wdJZRrQuDFFcy&g}E>iQiyB<
z?Tc;GC%vyJTl@Q#*4AOHIqOdighH+~mq8F^D-QdqhJpjsXSK%0Myaa4q@<)8NpkVL
zygaEWLR^xukt@35Qr@czH2b=%^M@{`%Y>@wJwEOEC17po@VM-rn84ISkdRbTif(P?
z5BU2h`F-FA3_6UDrZ+M7{3a7RF_qr@&3m{&01@-MT3zprLA$$h3xgO?@R*vtZW%SR
z94UN^Yw_wys`&yr4O5S^l9iRx<Kz8Fu`UqEL53Y(+~&iH&)#c;*Z;Qn(N>)G732{0
z5Wd{x`8OfW`~37G(Gzsq>OGn*tKq=(3Be#+t2IaY@7Awovbn0#dBb8NKeW)?>8%NO
zUJ>X?uyWHWr1&PiAu!V^42_N9wmah2tCZDQxQrLkjtbNLN?E2eHkXwh>h2c*NKC=_
zFH5M+d6$tTIUsg0z&(RKvmMh2mhtgw<LY=7T8`faU~hvhz9yo0M4t%*1B3A;ug$+!
z!($9I%B$i;q@)x56X?zoez~CC0-L3!W%<fhLoA!*g;b0`JjJePde<kEn2BKq3$WuR
zcckgWA0;IvbR!t?u%1o3mFCH{HBx4I`jSYI+m8-6&xYQ<105|M0}p6KJj8s?@cgFl
zD*^-I|2_<QtMM?>%WtCO@RxR7Jsy(`3{zu=N#(zwd?=_7B)f<r;vF_g+&10e#Apio
zSf~q%r{06uN6lVh7#JD~Gctm}?<jgL>h?!d1N!LYT7z9kp`RmF!3HSllK(bA{D@YI
z|Nfz``kB{#ewVwovGD~jKQhZ4``^ktcDag~hMn$@Z5rGjyxx<nJVD|>KL1y<;VUIs
z(ZbTwzf1vqu(X|R#}Xoc$Sb;+F+r>3+m@N6UCqy@yEmngT&vA*pCa&CJAXd$=+1F8
zWPD$j&ix*Q6O=<c{CK+m37RW_#+)zkZ781Rd&U?US2_@~ikbYl;B1!+w~Hg~a#Wij
z@=)pSU=_|wV)*dza3G2XhMAMB)^BPW8}F{VizNqz%UQAO5JB&|nkmZ?prb)g_vh=4
z_C%V9-7qg2mHI7LCyRAfI&DAJhwzP5sb5zoZP0p{7<{TdBuvQ9q17gR;cCT7VwOx4
zPA3aBfF(`L&#CojG_1=}k9J>!VxF|N@Kd{dUZ!SddWV@~kV6<wmz!AIEX73Dexb^0
z9y73~mp0MCf*pbM8TipAcs4jto|>D(Yn#uvp0Yhk(o$#m6wM2tUvly4pDX+K)ylIc
z7)?5cG)QK*>gl}eZLUHuBq;#o{-saXXFj~>dX3rbZ>*j+A19}pY^nKPF5c{3h)TYc
zF?)C;TsJ=~yWD3hu<BLx);-qU`8s5)_kB6)Yh|W@y9v8o8M=iUOGn_x!oL#+A!nkl
zym{sV0s<eQ*YgIguBA*;531E9CW#_5b&zyZ_lvlp;o)KPcSX7%<Lnd{o1M06Oa~_X
z+u<pWzv-1@b17hDQnj(8gV&p#Oz%z>fr;U2RT&xpi;3<R?h}reeu4L%?f6($)$uDQ
zZFNt@(h$iJ0MS!LAi|m$X*wPr9s|)NHCOx3_FLHMz%@>cCA>=WHk+8!CM;g{zlIv7
zIA3n{(a2xSAeVoz&&9eQH^;BhXxw_2L_$I$9Yq+FB*yu4bEtL>*cs74Y~bshfg=Nr
zhBpF~Q&Q)!sYq{^+Phe_@A>7W_4{#f70g)B->;!zn|1u&gG-dmD6$oV)00tEj<5Hm
z%#kDlcR)g=Z5}~kVhr*kK40q)&<uX1kD_NtF6}@$FOH9my<i>he`qqZu*fSO@9mWc
z2ewQ|=NhPAvj+PUFNK~TsZsDMw56r@ulL89Z&^Agu<`MWrw^?*8wUi{<VgT~QQF$K
zMJB@4Z)|OGxE_2duQkUk>|{`&tz(iJ63WA1m!q0WEDm&Vh)SB={^TYKI~EKqtm!;x
zQ7RKxp)Xn#Qr4~qERya_x}y2tT(v%PB<<Bh-h57CIm~6<LsilSN2+nPje%9ueEO^r
zH4mq#-g+7l46T}t#R{nzh#GB9N_qL{ZL{e24#@;zLJ&ai+q+Y}`me%2DkiHzY6}7V
zCW7Aa59Qoeo4|$%K(&5GF*PtfJ&=tdX|6eW!E3@qDd}oH1M?{_uUgM@W(pe}dIg*l
z*w9Fzs414dT}2Kk+C)Xw#u1)_H@CN4c876+X>+?el70FH!(7i(SE~a-E`wM6Ytgeg
zxB8faaPzn}KaUR_`3*RZ_Fp=J@g%sk(9*(x%`FC#;!aMNw)0oANs^gXH;O|eBMHd^
ztT(yKb_v+T&}o09EffsI0AmvqY%DAamg4%s!?JqNsQzj_h*Om{>O^5=goB3%9R$i#
zHEx&v5FQrB&bFbb&-PW}DIw;fzAoo*Lhj?LndiJ9Jzj()WkX;mR^DXVRGp=A`Grdr
zOE0~moLnT-SSV}k+e)*OHY^tj5GY(yVkqx;7R4f`P$x21AXKcGf14cFtDm`Y4!^R4
zjuuFolNO14X-bxrF$RW<#~Wg+nvm;@dItuM4-a#L!VF5$UB1H0xg30d9teRqB<sN-
z6HZA_H%-c!?CbA$J6VX6%C~ku7kDWm+L5s*JrFxTEB6V@Ztv^s>*?+u=I`+94?^@O
z;lJPys1o}0&5wCBZBb#tPRa68NSi>YF)`{Lz5(S#n(7?ULBFdvJ;=&t9Dl2=6@U^p
z+PttFx)Jk{U%3NVltKJv_k4iIptY8TobV}At4(Bfc|SN$$_<l?16%b#T#Kr|={;hc
z<Z+>vSxSG0n~0T65#nf1jEDEfvugqdf5A>|;Q)~JL4)AxjH11zTLsvXc7t-9gn$$b
z@Exn%d4EL!zH8(r1Sd$=T>jG5ZyX(k`(SF*SV>qRe<UP~#`8k6`kX6z7iux<{>evB
zhid>29Li;kp*A||aqX0IhZ?ecH)eB9E{Je6Q>wP<TsN89<_bimm6(!R)%?~Na3-aW
zWIHi;Vcmr;h=_L&(Q3=JWp(%2HB}9q6^{fZhb4*dA4d8WB5){6F1p&(^mI-49_$qL
z<iW|Z^B0-He}6p!K&Y#Bu2qH|p!e^^P&)AJX(Pn8F1lb=PlfQW2-W@l{SCW(qBQ*b
zYkf$3NUBQ9d~rox(pskAz$Z&}xc{A3_#t`B%4XRJ0qnb1zobh5R>&460sBh%8Mc{*
zsTOE(3q+-cY54EWd!8`eP>G@(48Wo>MwOJ5AcL9KY=Uq_Io0Ol6O#4~-aZLknL44=
z$YIg61Nbx3{B09XH#P@zhh}qb&&c=e?vl2>92*F<$f&43#=qIuS65f}_tGeM*F3SM
zl<OEquDJs1CTm{x8~gkFr_5LO+uc3W;{;$-Lfc4O(QCcNX+m&dpQj1Jf9D^5x^I6V
zRT<waGe($rzQY)B|1W#2hJ(6jtAz*p%gxK8R1;n_r6}l=Q;gek_X^`o(6boq85rj+
zeGe86f+$wKLS=r;yDMpnNxXMVafVKal~1f=o>Rc483|}S>q7SbWopAd=PoQP0I5+f
z!nIBr12)@fkC9YTB1iUE`oTW1Ksq^jyqaHx7z*BrzR6Ql9*C49a*6z;YOm|PbzyDR
z$NC4RXg3a@mN%@+g-3X)G{xOG*Iq;tNus5rGex+dBPJ&HcniR6i_9LWf`RBHU%E`h
zykp)VLxXZMDY0v&G*=U8!axxG$@vl67nSlli||}NRLiMQ7f^_biNR_v@PEMr`*xY@
zx_k4`NJ~qX!w^P}?5=hEz-K<tT@bbLrl6>pl50XbXAyp!6qQ~Qxd&epX|i!l{sBRv
zGheRvypmtaGA|*v`q6gR&G#pkJWt9ZtQHbGrMGLHLridfAMOYD2jebql!OcuaB;tT
zbtKO#-yC*Z{n4f>|MKTC2v;I5Hkd&h>GRB>5@yk}xkbh0Tq002GDVp9YSiV^XK``J
z=1r*iF3^wtZ5s;Uw3QXT<~u?rVRjJM>R`|~NGdk_;qF9wziH_Gq)t0zT~$Qn!NY8-
zLH5HtpXyh7Rb#qWFE?Q(rig%GhSID!bPUK|ZeLo>hi=~ccBi$FVCLHRc;fMpAWP68
zGw#bB^k;Uj(WJH*&U@<==kq(X@(D2#Rk>$#3Ci3kj&np8p#BOBk9r)YkQdrO#)Ve~
z3nWjULUGn|cn%m}_kF0n2S63NyoM>=pJ%f9PP)r<k|!aKdqR==E<bIT^)0h)S)d#H
zB2q`yZVO2fs|`b4$I3m0$-N<F1-KhEsD1Ti9mfZgh&XmC3szR!WMca5$QZLYQ*67p
z$6DI3{6)wCVYI5+xz%6l2iD&V^jYqKilkL;)a)1b)P5oV01-+S@cNt$z3#Vf-+Z5h
zAqSSKf|_5xI82dlnHcO55Mf!b2e=@YuiU$T4)Ecyz6?`F{~`GD#lO;G->lW>hOcIK
z?4izhlFz?_q}=T7BK^nuOVhyo5N|}Z))kv`1ZUlW!gCP@-L}h2$2*5Ry;X;m*DUSO
zewvk)Y7gb{8*vu!ty<*Mu~F&))!|Kf{T3zdYo@WKWvkJhP{ZvhL7Vb<{-rSgf&9xA
z2y&P-G&t<x^6#m4&RwhW=bMiH9k(gXQ_wo!fkRi{>2Tr2U}S$gL>-TTmRfW@vHw?A
zQpl^84#+DE1%cE4eL0Nc76eX6^(#_@+q)bF$%C(Cphs95T3QqC9OI7)&%wyW!kL1E
zz6sLO<F_W*W)3RD6@68tg2>vF$Nm3EgwHl_@|3fGxd}P8i9qw;Vl3OTlxjS7K8|gO
z6fZ2_t$j)#NG356?n0{}@^M<5)b@6He{w!A#OwHy*4_5@`Kfxq*v>J8vnem&xpUp{
zyJPvohh(ywh2^aDdHDW1E1!_8WtJDrp$(3nE|QGfI|D)1#=+~i>gth)S)%9l*1qP2
z4qYoRr`0~AXSUFn+xDl<4pO~GeUWd%TM7G3Wh3WGWc&C7$J#y6`u)GnKFX7)@2?_S
zZJ#{CAM%qg71k06KkF38m~v^RPaBcXMB#Sa40*_Ud_yKZ^DHB!F@Bp9mS1~ed0C9{
zHxZwHC4Z}ouDH;1)b??oVDs_u(avqqaV&=kls<NVhN#}L`vp@ZTuy`bW1gBcyUlA4
z`y#P#j8tTuzs)?L!A?rhxDZLc@(1rvD)$EEXFo4_+vpBGJUCqL1&~6FoW^Zy<e?`+
z&SBD85IGAMly&I2(94L`fI@P0pSyQ*n-6it3zblp+>=h6@}IMwOZ&!A58Rra==pdp
zzwIA8Qdxq8^QVIuz8jk)3{%_LcF<?j_Ck+`_lbHLt`41gC22z~j<%V?Oy8sRIw6q-
z=@r)%Uwk|y$Z;O~Ikl7_B(2csH&DoSCV?SE)ucv!f%Lu){Xe4ntK)8gwGsWU$WzG}
z$TM4hO>1kSWH_Y|Y#UoJ0~aCbs_wxC{-rq0r-Fm${rIZX(E|kw36AaCkZuCHX;kI6
zZ=;O1TV*QUPbLv{BHlVsd@c0@@b;owUcO{A*_=_xchqE6&E)x+J{#YQ*!igh^(`7s
zrsFOQ!eDiNSfYsA<$oo&{qGmu8`@T2Id8I`H~m)64L}lz#H7OWkr3CdT85uqWGk~R
z9fV1nxM##voNju_b7m}#M2Whry6bVrAI@GAUN0x_6~6Zb9lM%8yf^4Yi<+zCD!^vU
z?AFOXmo6+JV5yx=*cORgn)A3X)A@m4->s<?vG?JUHjBfqNs&y^Py|;()v3TBJFCET
zUb<aOg5E}8JII9_TFjLfoK%FTni1DcOOC@}Exs2r=nmWBdc$tdpDn-UY??1+T4=B@
zo-!pdeC1Ly0<!L+t_3+8*UTLkwpJw{Bic>J80ML0$fY>Z^ko~$KZ~(`g(#Y9p{)h!
zFSf_X%F24YpCNNgt(<L0mLcQNWxV+;CPG*bQ^5cU)%I%t%0cM<jAFtD*)gAA)B+IF
zV*Nt*ipQxBQsyfE)P=dMbI5p=2Xb(5RpOyKja<xS!TwPV`Nb)`!+yP$e+}uwk4K$Z
z3QvB439~7(u}M-2Aoem*fGxET3MqA)Je)k`b)k2@p5T?liBcFD@QzK?h`eQBP+44H
zCXP?b&^<hby%~|(xe^cyvu0`IsYU|3xo0PI)GLjpZFp=Nc4x=bvP{q1ucn0no?^%I
zx{H_8X_t1|=NRoklhoaSG1SjOrtO(g+{yPNQ?;qJzz{SxCfNlh7S~p<vGigOkwr#j
z^?s>C{$ZvIx4qRiY9GzewqDes@E06}XXasB{rv1*ZW;57VdWA_wc>$n4Cup*T=DoJ
z#>V$6&suj~)reDx6}al&9uzzJz1B^hua#+l3I0kUNyw45Aq9lWs)m|1EMiz(z@^nf
zs^}C>X=MxJk(u1AL$pX)I@0>>*S^Cit8H41?o~sGgktHb14oLc;j?;;IS=%)8(fKz
zRa+*5F;vc$w{PaZpZo8TuFTH*#<~7H8+&L}k&K#&@D&?SE}G(?6m(Ak4A46jlf=A3
zXAgTs()aX529}-chR7)653x73Pnyfz$O3i~YhPKV#huHm&gkZvPe#k53v5~7w*$9-
zs$HlzaoKCT?tF30FF&)C1yB8cY-&Fxk~r}}!5|uBH9{tO`p_PcGwbhtblOPWQX}<r
zk4879;v!5y8YG&|<E2eO!9xV*im4?RU6RniXJEln+ehGfh8QGSk*ycve(8%;z|qOQ
zBvTzp^VG&V6Y+fK29MRZ^l_-J#|P0eNjNeAVtE1SkVf2ym=)b={oP1w=Hm(;Z|#nh
zgJKTYOl!RG8Za%EzUOW7Y!BNhah)M(uO5%azjlw-c;3V&>9fjUqhS!n2Ok_+n<A)C
zM;e2|0tW`|uv+=V#cs@GTONqHKyGg1KUoKE7fN58qSitx9(nmkem)I9%Y7wXEF$4h
zk7AVf@M^w|VQAf=S=n~1<FYGZ7r5!rIu3H2OBbaP!eGYS^8sTPaed_JM=C6tMhC+f
z=a-oJ;tmTlO~4@9U+TntFyW{X=gd#}7*$6tQ0d-!niZ#PY6dU8!0gm=tj0tocE(mL
zucg#IwCF&(4myxTFmh~Pb)o1Su3)G=Jc+iCFjZ!{#fW6Jz}e||J3r+X`L*l%*_A|`
zMy2`YV~mXB_>7$&G}va&I6NuwJbIswl(xVOpXM?6By}6N!i3E{?aGlq4U<M2i}xAb
z@!~d$K8N$;VRly+#t6P+Mak>(3iJ2mkLf|UOZ9(?DR3maNs2<1Y(cG5LlTDhq^*ZS
zk3*eVv%f>&S@m%IqF%|8|4>f+meQv-TlH>O<~eeH9)ZD&i0DKPV3^5GZT@7oY@3rU
z0?B)$iI5K;dD%mEpEiZORrg-hN1Z*CE+>>lh4LLzr1R%td=juhFq|gWciVZA494Fh
z?M{Y7$h&y)Nt_>5T0*akC|QbS@V0NAdI3wa*~ocdT(WjTtgU!ujgO)x>s)JJiJXgy
zM`4f8FpO$$8x2ktgCk?n+K4;tEAEB9N#YBrdA1!Mn?@E9{Z>B&OPyL+xECg?`py#v
zP#NIHur3f3clqoTm9`tP$d=vu!iK9@5+^I)qkXWuCTCD+;dWDT@9!X(M=T*BvAI4W
zsk`kDp+bDCWDoKkJgne%EXCStYOc#HodMR00f)rK#(^;|LhIrFoO>v*j-*(io*E{_
z$y?m7UqzMw($CEHzor`Wc>FvMLJqlDAKIzNh|g)@KMjLh7eO>W`v>LCd|xx|+P4^p
zVFV(T-bE^8H{tk%0Pm0szt!5%z#T;RQuL>n?)38CTn~M_Kv22hzx+QLO(Nl)=Cvly
zeEK|Mlm6H|V$#Uxl<}I^suMsR>qEa7*C^0@VKHnkDX?vZ;tFpM8T~bvhAJGM5owB(
z+oT#HlTAny8mWY5y8boZ84koorxdUr33>9ce{o7RHca^8ZC1H*;QF!ZH$l_Frqg+9
z=}(7pL2i<&M|l=h_F<(^DUouZO0=z!lNH|5RS;TH<NVR|e*Kv3-%y17Q<f-IDShKP
zy_WF|ef}pFda49j`&Eq!+@QG39z3o1#PTKIzYuLiG!<@#!eZ5dOX7^~z{nD^J_#Cx
z9;K`AijD*H4+L?k@jbu3s9D*nJE^@zChbj@yBhW<D!BP#*C1&!I0x2nOG+YS2XDx$
zTCb7hN=H~ePA`(^s+AIVucjaJj89+9hJmFSyB2@H^s5#wNr@=oSq2+)w@wK5D8+{q
zl?9#8>E%+NbYv_wImRT53K`=ca8%S+_jN_)(X(U`UQcPTA<&TtO!sZ#8z|{(evCmK
z6aDoDD`q>KjY(IctZNpX^uaW(;fI{01mBCM1&@~jxR0LxcYzNEiOYy)ML{V1@8-NT
zb#8W0_**sU&xT{-{G8A!ew)?u+txWDDKQlqU3Y7Hhl=G=6;m9R<cL0cS|L4o3V!WE
zdhGrH6&?H~_qeEEkyX~XP8gjsPZa+$*CeCA{AnT4w3;*8TJhD)UmmJYC^XyA$`&H0
z%GgsMTw_x+PY9}wYD^>)S764QYSS;DN^c4dZf~p0fxTdpB<UJ{Zo_0KyVLyLVt5bP
z_&ohh-@>D}Ow{Cw9dYSd?-#BU0YR$K&kt}Sz?MdH8FA!ajPlD#5+=5Wz6%oALdtAP
zI;M9b1v~d7FzBZ!u;+VEDNg#S8AJ2#2g(P(7j6-GBr~QJHql6OvwNDO&oWi1k$NLJ
zZ-ws<cNXB$qcAa54jwE1aF=fc>&*>rUxuh!CG-!C#wJn5W4?!fz#2gwE}jPQIb?Yr
zcka}tuIT$n6ER|K{=!`Od@Ym^A*&uZN(1!)E11^SO{Wx}7q5<4;!QWBZW0}4X#_&B
z2??WkIG?&`!4dPxin=Pede6($uE8;hjNbT{ERq+?{iyDt&q@=d{5qXe6;5|m=a#hk
z^>#n4o!8syybR8Yos8F72S2TiGJJG!k{Dd`v8+I?96NA#mkXJmTQw2_Ro-0hnt41_
z^6<@8ZI3O!J5#WMc;z&N*eWEgML<%yM0)fr_SO%d7EMqOc;Ue9KE%;L-QveJ{QASY
zN3`G@hp4KmQen+OR;+$RL`{D8*TPX_{OMpLS-t(#^r|3z?Od<AHp270Xs+7IY`9~h
zwt?tu)#WW(1Tz*w`Z8^q6gNJ;xSlx<Y%+31eiu&1R?Ajb{F?9lTUA*&)^^ZbFZxYT
zYGXr!nHRZjy?br-%NUk5bj%NEUjRFD(Qb}pT1pGyicCDAwFea+&jQh>hb`+$?d-eH
zdRl@7yZywZmnY?>VSCMndX+K4&i&4IT2dRO_}BYV)YsXgvS{l<SF<no1uPLyWIkMH
z(7v^~nS;;r0b9dF3e7=2r1Je9PgA%1nWQd{hoNI-4fNS-%h7k9F&HU0cx;Q-rW1*r
z8nYV)Ha>T<^en%fZLzo_-Wof4Ku5KCKUeUk*3FQpyv=fqGGY}|B&f>`%Ki#3;yn%U
zPxAaQBC2X>y}mUY;m^4tU?HoMA9t4oXDg<O@E%tYlw(&nHVojPF;j+>%epO9NDD0T
zSf^Ds!#mB}R5U9toq=@`PAAb*OWVFnfxD6jlSegzPSrdK`qSTI&*pwcDu<CEp|mg-
zTu33^!MuIvc)Nku8r0S&ucu%?64NwNO*!r*giJ+Pk?C=qbz81OBE{UW^k#AO=vPo*
z?327QArp9cTfbiBap}-TYa~2vdeeOOd0&szTfn%K!OsxSv2N<j?iGi|(rfVN=u%<-
zJ1O@pdy&I_gm*0|IoDcX1w#~Fco3z4%aCaK$q8mKP@Q8O>>(1MW37#qlJMZ-kdx`b
zIrgKf>29NsQ$S*%Q9iOJpOybC9xt+>tt?$068?ut!#oro4b8^rU8$1ub0y4HsHfgg
zW4t)8<R3mG?^a3<9Vl%8Jeq_9x3@nd#iwDuj}f%t?Pq-!L&q3oI!-3eZbfY_h-yWv
zk+*5>Bc_olO9kIgXG3*5qpeXji|&_+f9a>oq9lz7)XUf?=~c3Zgb3(cn9ZcK${bIU
zcn#Ocg{MA4^b0phD!2|QSl(M>7T1fS?3MR|irrfP$ID<Lz3K-E+%I9r*|esGVN#%}
zMOM%n2TiXvjPE0*5oSU<guapcc*5%<I;Rgw8T6_euFIQaB4^YeBs&re_blsO?x!;k
zjLORF#NMh$;vnSXOVpO>5;-aL$E=uWo>ft!>qsw(;u(d+9hU3PVI-5JXZD))Ww+R~
zxQy1HWM9=4zbBrHIg!9RWWYG;`7-i*oSdpYHc9g%Arz$yek@>Q*RW}TQ~F~`__LE9
zVC_IA;o*#A{-7}$Gu*#@bL%%?B$0^G|9XJF!5@BeNm3j1<o=A!R_^-AWI`hn1&ys*
zfuuF!bk?GZG+?Xlei0M*;3mF#gNDlN8sAI!DgD5~P8C%r`r7DPyjvUo_N#Kjfkr{m
zRXo#BC6OH2>9(~{`(dthy&x4S>O6rrOX^{6gaoIc9amifsAKsP@|Cb^enYoic_<To
zS!gl;8iYl4vG$Y$TTtj3HgkX=K5LptFJkq((oAl2M60xl^x34|ns>NXZcy~s^b~dN
z7H2-=GipnH?9C@a6)`=jG^VOTNwc@X$Vv}LBQ@o<r8T~8c&9%tE=3c!ZtaLo;h(PE
zW#;5chkPmtV-RyTZ?Kw_)K1Jv_tzO;1ns|l6gNM#2{jSNzW(z));8{**wq#*01Bua
zX5E72-)0m`EVpehm7X^4S2q@3=yaR*Xm$qWu_^<JC3dl-eyTgKVsv?4`I&vn{TcH6
zJ>n3fW&qBhC%gN2!7mI^3v<4#NQJ$@d*ZopIn6@>?YPAIm$iqRD+-dx8#;UwmrJ)V
zfis#_$ymI*ee>=^yHU6&WMq4#qz!4z>K!}RYMx}6cgB7q5)`r-ZH;8$bDc|Uuy4`c
zB?mT`n6F9IpT7H0br{((C2!U<M^bMM!AgQ~zwOT@23PUNHHm+MKpUhVlw(}YX(aZ$
zdbJ?gmeV=VOy!FWTpqXUZ`JRogfsuZ%ppy=5Fx15nfM#scxT%x+NV#$o4y|ueSG`!
zHm&qOAJ(_uvk7@kbV{RW!nd|Z-~V_#99)nd=+NFx6&fO(Zv0qk8FX^+_S{=Y4;d#T
zZECk4c=Rr?jx_qMw#zeZl$hQ9<V&t#%@ko_uQ|V?81!K$`QT|!K=XYXdh&<?J6`s-
z$?8_Vw%80{baBzxhKOR5zGGMj`}UQHeTWOAW(u1zoF)ufKvU5L>s(NHbD#P<G!W(D
zvQ>%LZOFU4v`fR{P0lAw%Bmj@*%wxQO!v!Y!~r^=Syzwq2FQ|vg}=^UQ*5_!__MYf
zh&-?PhdG{k|H`gya!J2xB*;5zroad2Rp7|h=9{7C*ZbtL+wUV&8J<69!q>#PJy3Pk
z2=T)tTAfe;s;R&>djc#8TKXLK50UZ>=pHC_LEIu7x4#r=Wa4_dZ})-+ymZNfqZ;HF
zpHo`10AKUHU4I_n(gExPpfQp%lFZ7gTaV0a2hC7U!YM)w<Tu?PC=8ayn`DUBHJDlc
zt%=5(a%emhcZr(k?;n?1iyfAg-bX1TJB0GjVUY3ll8;+Bil!*fY(-Jlu(@BBBOG)(
zx-$XyXw0nbdtU|v`RDc3y~;g%qC4Yzhg}RR&?w9w8Dgrn3u?;w_K7&|pCDK!R8I{V
zo?T~F!}RjTlNT;0M9yXTJpA6v#XV#c2_mOQlr1B~dx)fo$w#g8b1bW8Ne5%*Hw%#2
zsS6xwN3m6(RE(xeBrwrdX|hX+#nIe)9V|!M%$F5c_tBTs#vlULnoqjEukMc8eq|d`
zo@R(p@u7LMVrK`*mGt1u{ALcR?v4*x<f4tlwG(FIG0p}HyY&o)CcnWvht+x_#s`R)
zAduU|i(kYP$tSY_?Y;0udo+iYcK;-Nw6@Gt&Cnz<s--5-dBOAg;pemp*Ba*$@)KjZ
z^s}Bl4C|OA*2gPa$9H4f7kcdsG)$<@2E6{76}kXtH8;rUug4M%xt-6?&DD9}rKPn`
zV-FW)byxn7P!#!h*tS^kD@SgEoGA$HdH0(zCbNv3T)9pSL75$Uppb&tsjy!RyUx8}
zQ?>Um%!$AS;_gZ<4>i!}_Ov!Yj0<|@Jwyp)5I@qwJkLK|ZR2f3CY;}m;AOxu%syuA
z4Aa{e+D2Yizo2AfW`M|5jcD_M%Mj3cvHaL1M<(oUsZq+kp^E&PFp4&V4<0}vnIQ^U
z;LW^G8h78!gj>`D$I^0#`a){-zv`Enxw(ip)Hlh1SRA9m1#1tMI4`d`bWM4l6dV#5
z7#h(OJ1OKZL9QTXFA#nIhySMXVr2Y^S}ljcTZzdjW^ZiEbTS`_tB@<N_IkaF^}GfS
zlhm9HG<SfICO%(jF!H#)F7R~u?891)N*Fx0=b9`Mnv|KDIl}*^fh`ZKsKn|hb2&m8
zgMG_{BLIQjLv8aL^4^`S_TusY(trK6es7=wW<}9ILI5-(sQQ-?URWPt#jMr}?Ck8a
zgnk4kji23qYFH*&QYQ=^GIPgU4@gqZ78U{+IOQu>p;9USy6vY57*vMnN%e4Nep7z6
zIg^T=e&ad-^D1~%1~Do~f-;Ik0FPOs)3e(m@t^y<K$u4-iDqFbl<oc~Ds0KU68gOU
zNCKA^{;ccME!ZlwX|hRcHhH%foV{rU>U)!2TyO0bN$gZIYOj;#^8>%6`z{7l1z$ap
zsBxjz3v1+a4`Va4nJOa@pg*2hs^%G%WbL`=<l|UFNRGPYy*}}ffNnyDX=MZOJX<`k
zX*GRb=)HgM?q(<$r!C7i7e0?3=LQxZE!3Ki0IY=?XL%aAc{DFXDa#nUo?*O!1l4QL
z7dlZ>*<o{Z!HD9140Ln=oShaI7D`Gl2@!fSFFkpE+5QNl-{gylkOLK`2Aw(!PcJW%
z-ICaPpVqrLirAx8Hwsu0qyw^nVSuv1=Ye^iKII>#eZ3ct_T%Z6t$#S+AMF&yrXH1X
zb@Ni&#j=o-lau=I33Ft7f<d#ZouPzD@%nVC{EsYTL2&scY)dZ)x+xsypfHJUMlsHV
z<t9fnGio=NR58mK-nMUJYxszaZ~ZljmD7(KVORPF20))VCP|M63q>tdVk=?Z$9bxq
zU2C&6LlV6!=UyW}R+}sW*Raz&%$(w#qS;ct^#MeueoJPaP8$u{Mya2jLx!aLN0@LT
z0~HM|At7OGY^+A|Y0NQ%Ap7xA?7-sS`IuuGB=E&;^;5K^0SE(wjLQmB)8SI0sxGCy
z-Fssnv)p_C!%OoIL`_7rxoKf^upPh#1~{ME1YB4*p>}qowpggmUXdeyu0arjN30w&
zcwJGE#Q#XvNjZEBU~m`Y?e###TR1JdW>Jl~$}a7Nnt^oBX?ycqvovq#Sy|uHFB2;+
zLO0Y0X@KS8&tWxx7Ikb}-hL^$hWmz>;$F3={d&J`N_lR!2+;9Ahg>-5<Q$SC`v(VD
zQE=Yr-3UmRV?Q&iIA&Y+N#+DG>owLFEG|i)#L3h*Gypu%veHrp&FibRc`-yfHzu#F
zjM#|T=YVGYv+L!*nZ0i7&lt)_rjCv=ngA=pIN}Xi!gZiLoA0~H?oosGLpOC_PzP=e
zM@ieikgaYtPxkPF0y7mX*CZ=ifU089=5~2`c_dZjlI<N>x4rI9`cfHno`r{1P+W{p
zLZV2kC2+}UC44}4budX){-&9Fd99f2aJnp)p6qh4j&*=5VhJN`M?}NIl9iUmNJkgK
z`X(<?%Z*THGMm;@n3<6*ING^`^x?x?%)d%)CMgvyt&op}d@A&bis?MLK{KJOPQs4W
zaud&6?FN${5txA+-AmsMgSJSgm?lv?&1%>ls)dEDT<JqFu_v;e8uMA?6?^sgs1@Zs
zbkVO?o=xiEV)HY=?CT=P&yi^6MsT?qj8lB3b)9Jm14S?=RdYbmZa9;dKdSXAvsqYK
zeIy_lszMwKk;*?jJ}xdOu=;)~bX|$hRl+55GvRKVrlDHh6&x(3ufNhY7t8WI@pbp{
z_E<i7>tTRh&St9Y{+5iVB|p$io+xV2#3@VvsoniS-aGYtD1i=ukskb4jRJrqx-B_5
zvpl6LCwFwJXIRU$@5qH7Vr);2Ht@VopXWk7H(+vI0`$>^sNv6JP}4qmohngd8yFZ|
zt3ep8h)M|Dy0v`wu<y3kiUo=%x&iQGVrH(9BNv=l4^2&7{Kc#{@oB0rh&KuLcR}dp
zCeHW6YGogXu)Nv=IdCc-=GMMB{<+Sp^JxK+A<Lu(uHUzl1@;^0h4S~Lim<_^qOT_>
zCwBJsHCON-j>ZTo`WP^Ot8eRpI;M@itJg0!dyG2EF<S!m0u7nm&Q<^tzpkpv_ZpYx
zKtb0^8)2H3Mf_j_Kw1MZKJw!DxeAf_ZUlEXHvt}9+<X;yojTKU`Nu+U^r@n5n0~ve
zx~NP02j{Q2q<%^N1skArFrEtLV5U|OTkZV$W3F6Bdr>_CYS9wyPi<!Ndqd<t5-r+O
zdvR^e2M_}Q>@((rIQ;a)oRL6;p8KTrNF>EE&!@atI;}~#M=WhUJ<}vnC=eXSQ-$(R
zcc(qc2uuqz1I}NUS@3k#*>le<)b%AGPfRAq`rfa79BFB3m6erj*X9}kM^w!7$B+Bd
zWqfd%3>DeDtu45F$5Rj$ujzEHxj^u_&gbp5BPa+#6A>~GFbN%3TZ8Z{)^4rsmw+vj
z;?cfk*D_~0kSg)H4DbDUM;X{>k}Lhxc7LYdraLw;zb5f>VMnRs@tDZXT9O(h`2i(i
z2>m7H=B7W{lGnqa5xfYH?$Wqzad2~m8oZbp6~rVZB^&<ZUXx>HjcJk0>fjkcwK`D8
z$bQn00LOrJ(aikq_9vwv2M5Q5qn(>uqw_NQbryVl>9_k_>za=@w%el|rd>d)C@Cey
z2nKJ1eUZhtaamcb1l%20oBu`Jmf4t=PdTbTp%)SM8|Yad4Dv2VQQY*VrRsjf$~Za4
zbAK{}G1Qy<tx*PeMnir5h546&UwB110JEEki3$6|hnha~U)TfvR?q1M8lXes(@ZB%
zE^j9Qrrpn0{KROw5j%3Sc^yf04|u*WZe%ih*G_kx-Dd3?5wqkVxHgTaaw3(%=S7cM
zKmZo|=`wHA`ATFYG<sgsPl{3VGny-EqV6%ceok#-M$IY_>$3u_fDnTS4g0y4c;$H6
z!F){jCGPBEi_3nwP8~320BgWf{U0_24SoG~i?K8yH#Uu^FpFy%X^5P8t2mx<S<?Y$
zCt9YXqoJKH)KGQ81ipVHA{O$hPe}NnOX_rBvCqy~?KN@GDcpAJe5_y3=DZ^VFh-CU
z^$OzSu_2H*AU0Pz-0v+-unnK|Y|?6%>~R{F8Gl<14l>9LGRlO8#f3`1Q&1j2YT(lQ
z7)Fcp->zJ&4jCSReiuZ-@+GoD4gIDs0zaR<CN-6yVBgyJ7eUcifME_WnkBsbL_Y~j
z`J^Uc2hChzD8Pz@|Hzi<$43rK0&8UndSXWpb~Ewv;tK@={IU3wtvZtDnJ9$|+?-2*
zy57n+8$XuQV86`6#`eYVb$!_8(NSFan~2Xd^lEpcSeYzaKsX{i@~;+GjS~PzBm#3Y
zGxY$zW2Qal{vs~Q!N$g(7%7^V_+qD`i|WZH{XMh>wQ=DpkY>7-_#ue(0?;y9TUl+O
zh=%@rd7SO;_D{A|m501;$80%+prDD5a?W*CifrTZkAP<Ga<eByDU)1fOa$~2);BOP
za49J*?fK;K6YYJ^{4CkycR<2%1mv@X(FYf$Wo13w#@jZV#-Q*%lmBS(fDR5&7ECmC
z+$`3Bm`_N*=+oxEyh=3w3GoCje6h}Q8$}dnrNzZ^Z#30PtZO`0cAq%x8`hl6C#Jcc
z!u&07gGr9ogkt6F=-GGV$U;PbhR)p55+4UA#~hF?Vvz$g8KI-1`b;U-7_m2f)q{gd
z_^C>_VR&(oz=#&}7wXKyL_&En^%6nyq%x4deVb@e2T&Z~MIO&p1e}QGS?vxdS^moq
zL+S4+j^=kxK$h+i={K%3vdg#~>w{&^Z@ZsiFx=CD@hyRBb>l=97w-eqOdwh=%;Ahv
zg4QhL9ax!b2OFkvC;6n;ipV_lLRqiv7i-M{zS}pjx-0v~NI+WwFpLSA-AQ@fp`G%`
zf$0$6k<Ip%M)I2n#d<-rK}^6U85;`65Z;N7+&Km@(DS7KJY8y#jKJF*@#ZHS4k<c_
znjXe`MQ=P9xTWaym<*`Dz<DhfENpaGZFSjK8hluV14M<WWWp;wJ@1&^(`#KP{!M=?
zYcrk*NIzb~UAWi=^m85F531tgaC!(Zn7D8HDXNH3T3I5H=&aTJ{1M9^;sgW-%w*Hn
zcIkHCy@CW7ASxCVPmduhW30o{_@-Rk+>=yQVecbB)_!LIz1Fl};YJ-?@tbeEF(i-2
zOs6FKq<2Ap(;hUPFp<9ktc!rk4j}EqgeF3G6&S7o2kHsI$frQ*)E^$4__}s*_OpJG
zmV+U&T^%ktpvnr5ug`9stUkhTR7)(?Bgxl(xp8r6X?SR8b0mkR71%oglCBB=g1n;w
z6jm<&?*m6-74e_nk+BeHWQ@soMxIGA{UC)A#Su&K2L!23uCDE_hf@Gi{E`tC(SBKh
z$a;2JbX1>DFZeDZHd%5*y?Os?-%7zjorjJ|`Xt`H($V7<jSt|=8n$}?bPN7YRNYn-
zd=@5y)=1Hf_&%C}g;m>m2z4ji4Aaus;lpDolh=DMZvn{b3eae}z=eUqf;W~kUtC<&
zZpX&_@*;ss&#f_=`|inYN{#0W9nm1C`jBVs4mKrKwkRw4MEHAD#MfpUayv%u2-&fm
zu)?HR$S&4uw+q=1iHVO2-am1HaI$jcFzrVL6hQ5uXL%qk;Q4;U#h)4M<7qcX)|k>o
z+WLD2UMLMgzvlW=CjzsajLb@#yW?mIJCq;B*9uVTN&!*f=jSIZEDW$zpC50*hI&Zh
zAkdU*yaHarU(q-A6sxcyGnXT$aTzr=7a$Hl2cxpHvH7}Si&E_YbaZnwvz%RYs$z`g
zA90lE24Y~-DLH*NCJ@9EgE^nQBOyTzAdOaQz`#rbn%>N6?Q#HV05q+xuI6=I>+S2?
z1QUj}wD5T?&R6I)p_B0Eu!9UyL`$JiC=d*$o3%D=|EOZt)HzQgE8t4%2V?PxEG@nF
zchk?N3M2CK^P{8D8$mGSQ$Rh4oG>aF5v-!30<bdxv9PtZb^FSQlT%Gijg>-xND1Y;
zcLCAfz#y6$8X9V90i(k?XMm6Zz>;g|rD?qawOYC6X+ugA=~b80!ijI0+gh=^lXkm6
z@dsK3fD<2_n26Je2Sa{Dt^g})meVx{s2M7ax;mSiKN`#&E&b@-Jbt^<-UXaHIeq^X
zboF^8K0f~L?hg1`a5I_jFZga87>Q9pl3+QO7WDnHNJi#%-D>{(ls|-7hdH_Dy_ZMd
z=>ejJELQ6RAS(tmxd5W$>7l?_vHv&V4Pdx@>6o*XmWZ*vRb#u^zvXYV0;gv`s~eaN
zO>$zE<Ios3vnM!G6&BmvV=5|`N3mc<Bj5oQdLAP^y%msD85={nZcZO#8W+Dw$j-y)
z*~Qgv<qMtgNVoBsM34x$wFAlwY7@j9ga9@K7_5FX4nX|Z4v0KQ6$~3q0wvCPzFTMS
z3Q+GPRU_4Zq-oI+w!Ht@3CS996jjyKTmvq-I|zjV2XqIh;Qu!+B`YiI`9$t#I;D)9
z>B9p`4b^_LZtprB@GkfF8l=$rl>(@8lM<3h=P*5sD1KjW?_8xpo9DGf3KN_!1Qw9Z
z*JA=Z5Wo%{kT>DZNX%snLQ7%GieF&}hEz5_2uR+|K7vQ5b+rTy0D!n%Y={kp!Tsra
z2UuA*5zuWfFE0aijo7jN;>@W7J7Zzn2C=uey!f_Xvfziay_?Iqp`9vdATBkw^;Cf@
z5sw`kH+P>649pucvKow)pU>;FC!pp7N=Cc8yT;t=vZ-$Gq=+yw^XAJ$!sPlGgFxwj
zp_}J-r%9iIP1?yTx}1`d10Z4mYnEHBwsLE}gMk){rof3tXngtd4TzZ!bV|MN70$_7
zimd?A=jFuE(&|%eV9pyu`j7(Z&BNXn=It2)Gb>RQ4d%`5?R>cor~8@ye+~?Sxj6S@
zL-56m?%9MsJ_>tWYye@g@Er6pJTUQ>wId+U&LsUg8`Z{bIp`luTe>bx=Y1guJ$N{d
zAp4bvmY<i$0SKlT)QWD-R@0#|yXEg-)c<=n!5auK0?5!eH#RPA4yRXE^xJ=7py=qP
z>;C@es?NivIx$7ZW7Eim9+7knX1V?$(SKi^ZR8COqNVrre7M{mP|gxu?C(d!Zh(au
z1}6GPB;@Z`k6$2E)N+0u+Bk8IDQ<NK+k%yTZKcoBs$x2-DmvmSl5%1}lnffv1$+m8
z19kM(6BAV8GgUG&butoClQUuoo|rjlDZ#>kP=6x^9286itPM>plDNa9oG`v!D8P+<
zwFcDpfCkR0eQR?Qi<np;fARvCn_Q(Akokx047-PZVzQ7`qD>!?U%4Uzr2VwCv^+dK
zR{B$ATC4}|a4=9I;O4@7;z=hVBYRU6m5`87Q3-0}yrl%x=w@bSOrly^$r6;CWw0<g
zJpVn=!6uND3<LskaB^}&tHlBRvySR74y%jYf`S4-F;`w5`HD})%xo}CLJ0#?g7V+F
zVPMD$zP6M7AAa=1|CbN2=-zd{1Z)g5VYe`JZUm*B;Db_<iN}ALnczNI5Yvl})OruA
zITeE6c}d+NOf|;6vDKOSzlm%(=5%C~{y)Cnf-TOb2^J>7A-E;DySuvwcY+0X2<{LF
z?(XjHZX4V^xGe6lxXZVBUOC^<Kd>`<Up+nDRb4eaZwj1{2J|7QMeGw&3~Sv$*Vb+*
z{SafH<#hmH-MsvErZHvgqJ41i6YtXMz7kZ~)FpFCP?fZW{AGg2w*<S<NgqNF(?agM
zT*?`IzpP>ZR#XUx-$1chy(@l7v=mf~r>%+gg0IigAv~vq>l2PybBdgA-1r$XU^OW%
zqv7_Ly#EBf2(X|uaQRg8a^i9^4ELX4c@coqmX&GXQ0S1Cy7r&r%=EaW**wnVY}~|9
zm9)7Y?;#oy<4iTZUmH2v@>h13sM0DUA+*cxk)ii)bH)5sW=>$x>F)WX6}^V02);I6
zjDMX!K!w&`_4U2o58@x=k-#XmkhT>37-Y@dplsdx^{(7bZLy8osaiC$;BSXpV@+Gw
zt#*X`7Jtm7dcK!nJDyce-+x|EmHJd;VG}f&L4}Opeq^e!x1av>=JNjWpH2S>&$uD&
zUtpxHC9FdC3tjU)igPi$x;b2xvL94yzlD4w_{R&S6pKrUX7OI+p*{?>e-`c>QN}G;
z5*CYeFn&TOL4DKqT+6SJw^v+LY>Elz+G%Po^~~S>!(=g94>l$TaJy}c5M?kyJ1k6;
zKIuiBqMaHys`;j-AQ=c~nL6kv^-X{gAiL_yFxSHNQ|ACFM<)}Wzh3o-r6>v^UZ%6Y
z92D&t*~b{;m|q)wLIgdmCZ_8T3*Q@Xtg6lNvCRT__4pjV1MBD%yGUp0It5Xv$krBD
zpRZ#`8^=e(E?hk~FCONfmjd=5V$%t6z+qPk{k+Pfh}dvowge!YOjz{uAyg9|TI7@t
zoBJd|>%ZpOr3kOD3e7gU5fF)51f!>5YRCcYRxC?W*gU3NC&9?K?9fz%{)L2CoHZzr
z_|Ot5kM>^ZHcI4r#p36Jrs%Bxk;b^6*~m^=RT$>nYu<8?rsmB;yeJ&z+!>L(bvR*U
z60xc?PRAVgi<Dyo;gmWty3Q2uyUG17Bdyx01X_ILWnu$~#DlTYi>;U%6_BZyPO0*c
z?3D(}F~lXilcF7X+^q6qj!66NMEmYS?V7w8Clk!N*JXmrGVzv?-OqM;%m0FueAs@w
zzS>uSv>Wzt7dP)|>zNV_;+lc3`s;9W95xYo8@>vdw(XtG?dX0SaJStdvu>G5D*SO6
z<^VsB;!EQ!P<u7*<&dFQUDxIeQ^3qIJGQmdnLUVGHN5L;e4i-*7Q+8=H4*RJk?Xkj
zHb+rlE!<Sx&kM;R1o~|~{wT|CSg<h=5GAUqhT%FoFE3IsQa#OS@8Hv0Wt-4y=icsF
zvCIJOZ=sYzI%b4<z4Q5<Pm4pse=O?5?{He;JWDf*6w0k9eHSp6vh|3cw_6xV@voDC
zCpqU@yf1#h&*F4GN@xs^qPPjvaVplOnBc!(SUnk-uY%+lIRc$6u5{IO%S#7tZiR;$
z7ejBXG&omAc^urGt+;VE5OR23YcDSKUJ>KG&}e8}jGf%^EixEbb{rHsRSv@*f*o(l
zmeN(Wa&~gR5HYrzn8ND=7bU<`R;}zU66xDe@q?_ImGFfsUXW}iEfmmkIhWiy04*f5
z)36Pb{PE3(BX@}}zSany*y7~v^qv=DwR)vb3z*`)&euZ$S97x6?;CdL;`wSoRfkWN
z#diyspXkml>^;P7Hj8dXsmf%lXc~X}P^#reE-yg%-%I59Is|gY6*f)NAl7WzGk?fS
zwYy_jO<G1dC`W^j??M>}eb{P;ZY=&0pV@MIHf?=Zji`|0i%ll3;rb$#q7#yW!eTY|
z4yYj|l0tw~hz4bUpj?-lY!GlhKZbr5&|&g35!MiffC$bYNOLqT0j^kUM*+K%vvLf6
zQxZE~U8&mvFHo`6F=+xn#?X~?#h2eKm12mOuRd9qP#&vX?g3+`Po_oMz1d7Wt5?`1
zGnDW*Z?16k=Z7uMQ|v<OcSC9nR>d}b;K{qRGm4ue;&XJ}GZ~k1MbDW%t?uhjomdH?
zTteaa_}CH{Rzj@fP*e{|sEQSYp1NBnZdA{OSmObH5f5){WOQ2vec2;0_>2;DVUk_%
zs#lp@^lKo+v>$_9Wx0ig<W!?BI!rJ7^bnglQsmY;hb!4lZLQK}GuNvPzT!ie4Wa0s
zFrEQ!R->pK#g*EBlsO-7Iq|wQJoRh`nburK$npqNO5g^(k53tmkoa;(z~z<yxamNk
zZyy~l4&BPT7;T8npFUCK8XAg?CNCU#qc<vp9y&?VqNeWZ-}?FSBT#|A4a1&@X_Qe^
zuk`VG&Qh3W{N58z=&G7zhgu`7Q{^{2-us3sFdjUjdvZ4S+c(_t=?t6pAL8|o4+$wp
zn$K*Dk)E+2go~Srrd>fzZzu|GX(o*8P};l8@b>(@%TP3`8QmPZPUu*+(OsUNCJF|&
zm-&I;r%E)D8NV#eJ0870@MdSCX|Ho42`ZwR?wJh3m9*xL#^y-&OpXp4qN2jnl+l`7
z#Rn#mhh?Q?F)aB?U}50od&J^Kg@(}F)aZleS`!)jhH%whoeQ0B=eKO0hO~r?cOsV6
zpedHIe}teyOjGv{$V=f70E9O0@{q;&*q2A@4*S}^cC(8+48N4|auAJ=q%Slt3hc1W
z84I@Ru5tn8$`0p>k}~jBF6NtvRjHBe@b*m#%^O<9zfeeY<~zhCs$rJuDy-CuV!)s-
zdC-h_(8p)y8-0!ZN^7~#rdU7)m(P}}%Sa`_f#vwl{)Mt(d6wq`L?FJ*w-qbNk}<_6
zgSCz7m`#RY$)OS74%^$dQqrJo>t==WzRrF%51;EirxpVbH=jJt_#Mz4TFuapX*Otm
z8}`h`7LaIKXrfFEqq{su>%ahqOM01Y#_-wG%^%{r{0(lu3U}1~!j4XOlOMZYH)oe9
zYkv0U%tdh+P|EE$oT{YtMs{Zc&+5b(juYGxBP90c*@CtAOG!8)U*@xYc_aV<8v;(w
z&+)M_cKR}jt7mvE`cS6Z)+h0Rk~RB{=mf&^gR@s2{}k=eB&?(BZT~)FoOdGV`5~Rt
z%+eOON&Mvo>qXz^F-7mkDRi|EKm}!F<Yvm^J74Y8`4vWvy`^&}cHg^>Q~1o{4(rup
zSm<nh`0?#A#Lwgpaf_OSb*-ye2b^~mxAJNkceL3Fc%Dwn7cwP6P8?E19nCX^``u(g
zdHTLGSkXG<tnlmKSXeqSAV^7~vq!bl%4i-QgGKBQhr@fB7@06ek$*-Kp)?curbsS)
z<}eOK(3ipzgXF?aNa;Wt3ir6>%)b&zXpt9Zgk7h6`DxI{8MaEF0n`g1a5t)2AF8Zh
zqz`Lvah;0Ljel#L$54L@lC<;gS7ft%ff5-2-KWXs6Je6Bn{8f+EgJYdY7Y-;st-Or
z_BQ)f)ZYmNRvLAW(#ZTM#ir}AG=E>1)w076%s1iP0i=Y4sK6&yuYSA;`4Mp*sV;@u
zW{Cnl2&glPxrwKWMdD11fv&hcd#DjRs=r=eeJzMGm7Tj(N@Qs6`damXv+KjpSCYUl
zvuPf#lq8?Rn;6Yn@d_%3=j$8uwc&fb0`ZU*+cxxE!V0!V>p1&GY8CxL8}ME{c#7tQ
zg7|6m9Wp+Q7LqHIJBrWcWqQ)ZDcDs7OKe(9yCvwJj4npKn%gAhyd#fJLbl_+AFCtA
zpw$C(z9%+T`7MmZ<W=3y;{`SsbY<2Z%q_a`x&3i>*?9Fp>q|86HKa}tUaXv&!7e?s
z_h;3<pC*Cg$y|gDM-NCLtLARw%$lSHtED`9uDZ)J$=J1hp@Ff9&u0!#F9MvT7ZbM=
zVn}-)hAIOeM%K^|9%0A>D+Szl${L26C!09&ebIH4DZ20kPRF^C5CWu+QS|F~9+nfH
z7Zi^3g(X9;lKBLjE{8Vp<Pvi(q=^^t+4ceYCi$sbRzmH+znh`mX!c>g03+hQcCTHI
zR9wlIk(rIZ77-MFRcwM&R_WT=^yoPE8L4!yXS!IiAdx9-1Y;6IOuRHK<9oYn3XUdw
zzOHCR26f&w1e6y2zhc!orj{J877DzmQf_Q0=*T2&q*GN+Rxb@gN=u81Ot{Tjer3d@
z%r(hF-cH*Wq|QA3yt2m<IQ>J7g-%*6Fnu>W_H%kT^t(`r^|P7wArjy1uXosmH@A6m
zpgo|CKCRNCeKT<GV5u;$hH#Pd(?UKv#ma<FcCR1nNype8qPc6dr95dUo;=^^S25>H
zhwqZ|4#W@;e;~1o)dEq=>dV_HA<b+7$5|@t6v_h_JbcT~FX3C1<=YQMgijwa2N)YB
z{ZP4uaXjpgXw*F^N@oq1_RZ^Fo;*Wn5DHK>R@w8Tz(a8ip4Bt8THb7hy$RYtrf5qY
zPa>E++AlG>UIW+tpX){q5n86ba6rz#1mu?8cZ-L#e3F=<B_e$5olK<ES>AF&i7|tk
z7wBsI7-N)`z*>uJN@>)7P1HSan`JF0y`U>j=<>VPxg3OxTV2UvLpGN9%H`zyJszMI
z3(g>o>dk`tRRc^NDe{eyFyKRQ!+HPg!-;a6KHZ88*)58yl^@@!3tXtU`RdR3*`Zzj
zjB;6tQh_t0i@Q%6FMB<XCh?=Oh%;CjnM*;?L7neaeR%Uf-%LZo@RfU=HHD8jU{UDy
zx1rQYW&GZ*@jHXZ9y&|dzQi5*u|oG^K>SoP!l4Z6zIWf-y;(W1mfgF<Q8PUC(oAU)
z>=AOb>hP{0pyPT=IMAtyEj(tbT86&#KcAem@9gg1d%K&s+WDwou!^ZR);F41DDmX=
z5J-XZVl8E~1+QA`$7H5A{_@KUVf+;28#85>Gb^9P)d$)1>WM<XUHEK&DoIHz*PStF
zK(Bsdl^v{hua`4T>dIPdA?KA&@!6dfaC#_F=9a4RyqXkDg9XZ~%|}ihLmLjx02<fy
zA*`>41jO>KcrN}hsARNP>)O_&Ym^qh4Rfo%3|C=)Z-P+~Ps>}Jy~siSrxSa_&O7p*
z<Ax6yMzHmYGH3>M=PWMAwTnysDayj>Zh(hUP?$b&=S(?otEh>~wWG^&vIH3I%$V={
zAcvWe$u1PY(k62bU!cy<)6p-=5Nsfa#5UcyK)>m;rY)vxB}Y@e_k~bn=YKae{pJ(U
z-wseVt%L(8TH2_1MG)4z%hS|N3drrYN#wi;veOrlflK;ni$E}D8V)65bhe+fVnZQ9
z|FpZ8F+}kL@hV~os@|-{ACeZA)nTLl$VcZ|a&~vOu1#EmW1jYdNLN83Z>Qdu@>cN)
zvZtlC<+e|Z%KG|w!I2+s-3w%w(!R9e`>y2D`Vq>g$l`D9nwzQ%&HQbnrfuG=QfCw4
z!3kw>&`OPwPlzi<g1D@-BAhJr_;=LM<@xTO0)fu=XhB_l{p7$O^>ed$hUmE~z-l|`
z@N!v;{@AQ+jXp_^mni8`+AHyoT(65$qsF~Fk1`{Vy7`syDynAU;AJ3=H$j0kQ7Ocs
z8mim)u6OsTpI~G}YLM)`0ga|WMx*h~l&KvnSx74TS(%*vi7VH;@ru~^z`inX!Wx*C
z!c@wK+P_N{dTG00@9aNy-&K=fIsF2L+L)Mbg!hNLpP_3zhzs$(PMpNFh03AOMNkfV
zgDoo|{z$Qf&_rbsQ-dCL=S28&q;Ph|inz-ZR^OeNC=a93tok~tL}oME+$(v<t}l++
zdORMCA1oZIW4hdVl4y#J2VG(fDFoK)=F8Z3ko^<(+|DP5WYauKJJm?#dR_#^sTW&-
z;TO=^|9p=pHk>GBviH^opBky)@A!%D6PGUqHI+i{Wbhdh5gl~N-aK7*OV640Da8^%
z{RSXK%y(|5J#MU23t_n01fY=DA@O0)uc7Fgu3tOW(ok>^5DbVLNgC{zuz9HPaOzEC
z!+TznYi^gGM4QipTm9i5p0?pgH|!>_ckAQyPj?UKq9{ViwRbM@qIG?CA9nZ4s+?dO
zlZsrIHlqZH`y90K5T2~IZDDKp9rw<w5>_c)G~yFcE4Yc%-h7R@Sp7>mzYmQu&D5dg
z=+@ZcmDdbU0m)LiiWXWb=6;)Pobc-ylFNvb-#$$>33%VV_<3<Lq)n@eMV9hkGWtn7
zR@tkOeR~+#s+xW!w`8`sRM20^b=i?mlz~$&JK)S&yV_D<o;|QHXRL!F5>-EDcAOZ9
zN%IrzMu&h1OtA=|IV|JccpcyVaJEM`K@b#by;|w^qMZB=)u(V*n`2b2|D+X2Ut!Mn
zzSI#|AA%L6tNp!vTyZRt*eG1GpoN#)u7aI7Z7i~#Q}S#54F~loL2wmd-@_=?ap2<P
z?Q=_oacY1OMcb<UQMpDox}<0W=*5^3IF_w-0#q-E2Tip^pXTc(b4qv1zwc=VKpSC#
zJ!1VZu#01zQ}|<K`d=Ri?=H_?F{~v&kD;SEbUA4D;--H=18<!$;7!QY5yP^vkclw+
zLyt0iS`a?GDFMa_nwZy->=qqeUWpV`(BJ_&R%4WW<b1j7@9zf#J_s@iwu=Kwdhx@^
z>(=YGiy_gg-rc7lf1#hko0ycokDrq#lNXo|*JL!`THK_pygJ(ROQc|e25`bH4_oX@
zD#-P3?beAp(HO{BsrkPaXNA>C&cpZh0L8s{RGnV3yw7e4^IVgY1pnj{Kac$4Y@e2I
zZhr)w=wBC7F(A|N^mCEwOtva6YDBGG5<(Drf0Is~H7Q2{gY1+5MAhkQmuc6zZrJV-
z20sEHv!77<%U65)gjTgn1t$YJ)rjhES(~4J`eEwwK~8J6$}oxAoEd#gNVi6}tgntP
z{_LNS7O(4IW(@&hK?7z<a_fFvCph{L1u|tk*lkoxw|w>_7_5IsvujS#l>Q*fdAOvE
zUbDS}Sp?P9<#zmAzhf+#<Y>P5$k<$}9mwP3JXNjte8wbr`%J>D6qTHjoSvMlV^}EE
z!HrinPE8#F;QkWlyCG<RUK@9Hb2I$YjWF+pm5QtvxO(7Zga$}U&PIFOm|&xi0EEY=
zCStyvpO^jeiqUgKP^dtic3|!G>FN@?<_b>2*gCuUy>(e}UkAPS9H$-|mz)m9r|`^&
z=*Ok^#9AuQ>IuM23nm~0p%k9(>BfZY*ImKsuS*07D>aAF^8Wx(YJJ{vF27vzC|dMD
zWJ?w4Az6gqxkPgUVjHQ}pR_kR(vmKWtch;xqiNwJMr>XY{clUxrp!y+l9)1-u!EG;
zIvBiBhjR4F?NqC-s3MEBPLW2D+uB-NqM<}%4?BYZ!9pkW1I~UHt{3xIT}_3k9|xSD
zVc%#O*=7cs6;G}vH%%xanx^8%IHTVe(Dhony}aBTMUG%dO`7yGHj1uc*=JrgTB=s|
zDrcSRAsS7_4L&I1<7fLxKfT7nn03CB!NQjO#`M-s1Z%}Wai27}y(6N3=q^InK;%X*
zjce7kz9Ozu(j&E>WEk)?DR%vE{6&=A*he7HXnuWF_8PUSz06Bxi6r!JvHhuP7U<A@
ze?F5C<4t_K=Dd3ErLw)KSyF8A!p&p7pEf+pi6yyvBvYWd{!A(KecklnJ}(Zfr+ZS<
zZEAWda*7&}=(v)5d+l?^qh@da)3|Pz%er)DatNWXr}ul%L0^&Q<hKT&TR0zmJ%3=c
z^4uBZl14MvE8t!Blg}$qEh|*j$vg)ENA81|Vk?LFsQT=<lYLj0zV14&QbOk2hHurn
zm;NvKwu%0(_beedi+q15|5?Lmu#FAB`Mn&=U^b#KV&5lT3q~fg0`^l!OI{|)-viKd
z#Z}!mkN5JS?7%svVg&=QSGsQc*^3|05EmuBY#&<x12xUNJ%Q4%iix+YycAYl?wyi1
zDX`dVYSKQ+%sL-DUn!rDFIDs<_9JO_?J(8(W#Ta?l9!|KVdy!01=ngX?EJS@@>{kc
z&LxkE+fX6h*E*{1*8hM|1)VfuUpiZ2xRvXNTn%2Llqy>h`x^UK>F@I|4U*p%paPBB
z+q}hn@A}%T!~;w~Vl1ZcMp_P1USniNkW57>_9PGt=wv_Kq*wz&wR7Zq{UBa)2$jv>
zB1X?Du&~zBXoDJKgC#BGKjeP-(LknJeR`wg#ZW=AZcVo<AJ)oI#Sr#ey-B~_A-ydB
zH&g1{QO|3x_nx7(hR*P&I@CvPztP#5D!(W9E9|A2mKu+JwAINCL16B-l74=oGZN&3
zA1vC5mDkP<w*ZTG8$Rj0?Q4@NhRc{Jz2tP)vxwXyNyXwFtshzZMr09LlJApeGRIx+
zFoC4vnLcN!OmRUSG8)*`!>ry=j`ZO`=~HpGi}&UBQO}m_yWO`RU=-ZKQ@p-8sQOGh
zBO)RZH~t8c<SUJ3^I%hs>yKw83598zRW`$bS6v=$gMe-BOwpXKt|%xMos-OQ=|~Bp
zo{8rNlFHCRlbI6h`QP~;P|FBa&0Nath}lJ1a(Fu;glTbHU55{r(2*zX&Fzi7J08Xt
z1jh{DBcGKCi&1H+ZlMnz{|zAZtU+uVkS|VJErp`>c^<QmDr&HUc@JS#Qq@%yi;nwm
z@318ua;zzK?w&#)$g_?@xW0YG@@riEq5XYQ#?IFlEkxMPmNz$^1ml1)%l4OGKvqV=
z*mmqL*2BH^8$M~ps$|!SHUCDsMu7DL-cXU13qh#8S{Vt(K0}6a5ygd8xfa<}GBO0J
zDH!sfX8&K%p7XSnHN*M0y(|+};5pFq-Tdf>UVM2wLXxl&QSDp*(Wt^Q)2;xR77P<L
zF)b=m=aX)GqfSgO=Au%ibseTCD=+(N1K^8SvD)fk7KIM@Fw;K_=j-tbfP6k>6uNv&
z)*yjKW#B?mcW~wpR?SQRr)xbm!jgA8d_gL<-YIZW=I6Y~&L+Hl%aU@xv)4I1Y&){F
zJo0#%*upH5gzEHtTG}%fmG{~Da%7NIoyA=XT1qO>M>|sIO3flTyME}6)8lg`;P-u7
zh=`#HKruX!^bl(M;m+l^CtvcL+&}3B>H|kqTS-xf7r^^sdT(wDmvW{iA9)!e5M@K^
z?>W6-Y!s*jR`H?XC29cA>*vcngdLl&M>$PM)!mv^k4}G$FU@IelY3ZqTx}F5)f$~W
zQIn8YQ80SA%d5}3<e3KIGgFYMw@+)V#>4=Fl1saX+qUJzD!IMrDDoTUM?S;otE#Vi
zf|6bH&iq)aiUAm*2(G~?uS>31dnsd~sm&8GLqrcLfsYd#0kW8E--wj%j&>Z06y_=7
z8&d!bn@kIw$bf!_a(wk(8%>MJ#fpfkCh|}y*`cf)zZ5!%=iruXm$%Dv8cJHVmn1ZY
zS7lfKM=Y-3tUVW#_&hm}2x8LWWxu(Ehk8he|B2wZ&`10izj1F~AD^k|rdiV(pD#H5
zhIe<i|4vQa9<g3$xBn6E?k|5)gclrZ5?PXog!Ztdwiwj>_v@iIf`ds5)x&2GEuv%H
z!_FA+4Jh7ZX|G59X}R4z5}9vnW8pO$`USWCJP~Wh8L=~_TJZHtMbC_1r)NJ%^y_R^
zop&HF$<bu99-($XN_^1P6eB-!ALCM~5oX|T6P7Wr%KCiWHL^FbEryKwV7T%zMYohj
zHUcB6H;w0M#n49C?U6@-bf{{Wu7a7>GO9dpp&Nz_^jk=9%-U{ZnGoo|V8Mcp(@V7C
zm(HcQ_atIGh0JBmG_=><Yy<c^%ujq1N6}x=i`M(oG&+%&n_~J7c4qSxqP{d<$i_DA
z-8Atnpe&X(snT*iZj?XXJ9;s`gBH#j_zu36#~nH-=oH(#r~MLwP~QcZpO!Kx)~`(P
zKC+^|l8LU_SA*m)d3bFPXQ?b-iYG-EA?c<%aaxx5a5KJ|Gb5;yb>T-(oLE-&WW<*b
zyDUDj8MMCMxA*VOxYLn&pA%_Y*cEDT6<eCF6?Rl4^M=XS`5m91JKIjN8y{_l!djQu
z`euX(<>I=meeH!q(fr1#5f4m&MV|WjJ=%cS$jVM$gawcTVv>^V?>I;9&tTM|VZy;1
zArW6z*jXUzc$&T(slr&@Uu{gn;nIx|_&KKF(>yGWE>d`7yKl?y{7)^<NB|W38cEGI
z;1BE<MOm{TmnWu0S~+{z^Rf}&T&>WNY>rD2u3332<$xYg@ab-?AQw#pF)!R5CP_sr
z+GVc^mKgA9hxbVOy^Z~;;|S8(<_G=^jBSFo=XbeVT4-Ac?o-^h{t1`ow`@o%mB_q~
zMZDBJ6yC-yk|aHw^UfdAGOW<Sq3$m|n{|(G&ADYfl!eO}q{MIsbB#-HMHCX#Qk;9h
zC!XIIKg8lP(mMPeOtwv+>1me2v63XNo3@f3h)Tf_BeioT@L%5LXTTK`LkgDLf0>u{
z<k<COOmW{qWB-whG3GwEj8SwfN&&QoNlhgmYXnlm_Q!(XeEZL!GfEM1*;@gu5PPk?
zg~aw#3d39bK$;g2wYG8eR~aM&=$XNn&M{eiSy`6{?aq&lp8-g!(`2s)>lzm0c}w@h
zEwk8W8sUp^H0$cJ9l^vV%bW|>(STR~8p%49HRP$&a^00^d13M<dYl!WflVm^jopEf
zsxo&zI#K0gZtn?~nz=q|_3Pq4Y;q!Xs_QZ1r{yXur;)LHH_wcXPV)VecrHk)tw9?d
za^!Uk;-#a5WLyG+nh-O}FIGX=VWonX3Y_}N;roCE;c#lU8(ddaCZMGT9ZURZf5>u?
z0ffe%9zq1u5|fo@0!i}tl|UT}apK%z%WQv>#j^)=>|$ag0aqVXxvOhDE%sdQH`gE#
z3*Y6bFb5PG-Vf4bfbMpEEkRf*_nt3|U!88))V`}p{`N(LrOn+`LjgHj{C+HCHlX}P
zcWJ~Nww#!}$uML+zI?qNw>NPVFL)hVtRE&jyf2w9h{k2*YcE1?BKFTmhYFq`;n;Jx
zk2MhQ{NKI>zVoGMo`c9#A}+xI!x7rsdCkUCaKCAynw_Z$PQRa{ldJE}EstOei|3M>
z=-2ms(BNvTJ<U9fflC|_&c(`Jm7<a^qq7p4%=pwQ12t55)9XDg(f7PQarwYf)_3S+
zxZ2(It7vj@-?dwVrTav*))I`gQr2TfTKlvtXV<{b0n%knwg$>N95nT_YHw^8su0TK
zSAbc4rlxWd-0wrnD8=Z~?TDG~U&mw5vP{veKvw$A7bD_dSha17RpCRj_PUD`KH{d6
zVS4CcoH!sh>pRCV+0Gc6cP}r6;W!o0xwNlDYPX(sN(Bi2R#{{Y%5pa2Za4HqG%y;&
z>Pu`tqcDW_Q5A;+_i1wzWYk5D0nGJ1^c@EZ3MeR%y>!2Y6y=(HrryLtmb4l*D;R6w
zW#x)|6f;O5J>3>mv1`0P6U&}BpK#eM3Dc<(v{4=!k`){MX&|!SiTs1d-zz0qGqjlp
zNiTo1YjyS=A`p#_jVwh>n7pu1R_As!mb|DRO~F?Kd%;|E&FapbOr?9Mt3;-=n(*e{
zEl2W4V(>VA#o+VCf*>;Bld^`C!7B=2O59W(nzfgPW;f=Gk-6zE^Mjvw5*v+-sBC=y
z<k+McT4*@ukiLKbYe`MByRvBQ!5kfI%BNGSwi{pmnr9?s%S{m+s*dGk&+VWv5yB}Z
zA}!A8!h@@jy4$1bAkxv%kz70tU8S?w98d5`3=E-T&_T46A8BJ3|Kb=32_a6jkv?nt
z95Q<5a@u(POAYFqs{1_&%8e4S>`i@)*^h`d9P6)2&iftM+aW!bCkchIRyw9of8H^b
z#0sG4j$@I6gsnPWknF>N>7u#L<-8W5NuDmT-0;@&JrznfnM){)i+2;CYFUX;*@ScQ
zom=ml(^iy8uDi*zGFeG!T4PZ0)q5^)OL6})N`A7fR#JxGKY{)2@8A(TTYX%T2OT=y
zcGFjrfqg^Qr;J4+C$uy|4eTOYTk>CJY@T0dKj!d94!d|ACib&ua<8^GVo5HPD{`?Y
z)FzMt9P;bbae;%;KKnOGLDFeAn~IWO?G{T)V8XC&2Oo6?gaQ)IuP9@C;y^PkQrbk!
znCkjK@58k-%?&++(x}~s4xj#$$5I5({XR2L>)rknzL;oojnsFJWh1M88Lwx?MD1on
zb@nrTD<cL9xkvec+))e5EkOko!fph)ew9}^qXX}IPwz|S<bZ?XoswOIf=%}_hj_7#
zH36URpgKCH09INb7B<q?PM>b_@3waTZoqGc75Zsj&TkL%U;Wp;qV;`u3DX!Jjw=Hq
z8~^;d5rqj6{*G-{cm-ehQbq$g4RS;L_pArXhVS=1ad$p}AMP*5J~kqj-L<SuXV~Cb
z52F7CKr=>Wc)Ck}W%PjROQRbqOI)8Kvr<gw6pluH*`BE?v7ZagqMABZ=G0L;+quAt
z*1<gA%hLCZ+~4)#ZXaYMj;LgD9E8L2O=_K}e_W?n@~2u@@foLmCX>xSuUfw*O(_(z
zy%onI&13#E##GijlOitp&Mqz|7b#4QaC;+P{bQFhUK-BcG|8(j00Ge2DM!sG#q7(!
z<olq9t8J`y6s*3G=v8*KRYgl$nG7YuuN7fXafFj|jt7(=AcFLd-Tg$R_lXg2fADS3
zVpR_f$p^aOlp2KxExaYZ&bPbv3|LOzLMf`jVz8&iYfTXj#FnaC!b>D@baJpKfDa^|
z@lbsr!D9(rvn){D<Lt>aS5+^?Tn_?0{KCEJI+Jqy4P#C$PKve;Q*H&Ss!6!0j&cUf
z+g2!i(%i_^)T7MdJ;nBQ$`@a(0&Nfy+7icf)4d!13n55()%)jzTcG{}<@KY9&wiiY
z&KjD^UTd-I3}b~Sp}k$(PtZ9!2YUgg*-B#~RcT;qTX!slOXaK#ZE@Wcqll8yIj1{r
zKJ~$B##9lQ(nk@(j^L`=4Z(K`{foMKk~#!aSHJG1DzpAb_(!7dF}Qi|UlDEoAL+Kd
z66n8lo37_?7%@xBtn37h^Ii=?-B}bLj=JWr>veH%9BVfsTyh)k<Hb8%_;xw#EPc9R
zU>g_t9;6CSl~NxjcslSr!qw|t##sDFcroHAI+X3_N(l<t#*AnFq8qXuKPfcQuyPyT
zgH3sJlzW`TdwEt3aUpB?K2y6TP~OEvz{0l4?c1}#5N2)LY5tA<21+#jm#%#@&Yu;(
ztqX$4<7QXP`Un7rh5gJa2~K<?aY@h15{cdL*u~D-?bEVxa`Dz&77NEdTrb-2DKN5e
zrB7c4XkYl3MiwbYFvj{)I#=h|qa<zHu4tKQi)zW)9wzG*;>F{SQ+2dH_bc3XxgEW<
ziD4j-!zCg+ola&QwEb9org~qV_a-}O)Oiqs$j?tsj_+yhx?l6|{0HAa6tIHpS8qB6
zpBszTe?yz}g8*4k)Nt>Nx3L;1&1hx!aA4K^4hP$CRJV%!_MdNqTU_jN%Ab7>-;(2O
zO^075$YdLj>mJOc2NX!CzuhEck@1x$zoH@7R%z<LkAvB)QxniNU2sNe_>N{kC!)YV
zgxV$rvtfy6mOfi$L;G*p`udR@ME|0xYLdS{e6dlCW-H4jr_z@{8?5ej_ASH)T2!XL
zFOio5juuIknvzctfxmLI@6ImV;uHU-IJ@VDnYVz2EM>X8kJ299a336x4W(o}$Ldvg
zBM=*;#~7(vd{}ka59@se5a^82Z=ZXiawsAjUHZ~8XoV(&i<kwqk)HLVJO@FPW1T1%
z7V<P8Gkk{@(SU)Kl;ltMQ2qr1W;R)8I%8uJAkJlt;e`m|DS>7ERELw1{dr55gH)vc
zo`inu37nk3X*?kQ2F?!!I9=FV{z2O35BTnzthInRDI1+seAs?mU^`p{an+DB%mVwU
zz_F2A0n@=ubJI0MM^&;YXi)K~-yG)+p-<x}m~zo4QSiUSnI}OGo&ysH$_%pVaXN7*
zQ88-K5HF@CcTYiyi8?9i90(xJnUdt}kqZ@8DR>wvWt2qE=lmYEJ}e$I;x>1IAU9Lt
zfvd9a=F^+<V2D6!)_&j=G$jaH7DYl=HXiM&k23LXI1!;3U0y=uB+!WWvCZce{gVv}
zWhywvFaG1w^4$<&z(i5{k%4Bv#TY!&VS^q8JpUUtX;6Kb7j9wBINvk^fhFvmWnB^|
z+MUITS_g!bS8Mr-zC^x@+TWTLr=rONu?e1{MkvZ+TJ&salkdz_0i0FN@=q~3wLAIG
zik0RVR8)sf{25>G*X9?cm0FS}+Zqk(E?-~`{|&IjmWjF!N7k(vl~vyx6dLN)X^FTI
z`52DgWbm{F^H3}`c2=&Tb3QFITOg%Y7W#GsNn2Lc&@oRjB*{?{Lm?xsE&%^t5D>Xx
zVa};4_fR79WMJR?_|Vh*0MVOFIknFg_iLBKFB>cI&|xI!Q`P4NyrtGt5;4BDaBBfC
zq@4bh%eKVX=p}z*!>*I%L94iaJcQyJEm59Dn_``dl{8t>ZRf~}=#R@`A3*}2cFK{{
z1SQ3OfN;bCJRPPR7uk~OS1Kw9qfcWw*jzuVTRhLP%TpihSqWdwTEM%4ke4g$!;G&F
z=a2g>|9;zWh5AjwSH)2<g}S1Ud|E*XnvaYuVuEA@{BDImv`hiu3=8;hbCVjwR6~?%
zT%3;M?9z1B<t(nPr;ab825ZOf+g`6agq?bw6Iq~C-c7dbJSeUz8wT{zH|rs_4>5OO
z&zflnzNn`%4)4y)TxuJecF}}H+d&(<%kn!s-E#PzDxO4LVE5+SqNXutSN3uYrvF4y
zi_~scaSt2g(T<!<#)xN$M10Vp)PE&xAHO9HB%Q1jI31~3c&kCbp7$Ntbe+2zD_C0v
z*XR(7t<>A&u$Wzp9XOb3D;x){`OZcPjs_7C3Pq2)06<rWmTbNXfO@pS%B7Q&X8y6$
z+B;kEi?NEK=Bi9S-_oYEXk8>_gpjdUpPv^Y9YvJtZeuwUrzj8*y|UiJTZ+oc1`08D
z7v9LZb)l|>S~*Z6P>g30*AZ8k-Zu|a87OW081qsSxMiV?I6+=dZ{^cR0Vx5nTc*n*
z(w=NMG3w?Xg!(>|+$aWPQV~u*r%3@!FEGm1+m*gor?EIFb^<Q15*%r)1zwLSIa%?E
zz2wqIu1BFQ$GM+@1?Z_sy%6G}nX<c%tg5OPLIEN*-3iiNu3s5^TXZfEAl#S2x?_m=
ze;@>!aY6_i_jb&NSAx@6FAO;ud41(fYt$VVfnCl9&zKr~70HQ0d)#lLThZ&)g>2HI
zRqNjJ^}82(z7QN8OI1|IXRPWMs2~4N>TW_O=KfQStCd%`?oisXk(zv!hDq4lyM+4Z
z?SyOZc3dnH8lCY~zuTDz9OQ$Rsro;gNVZhvihhAY!6p43-5=mdtWh@Hk5B8WoT`vQ
za{4vY2uKY|)&utBAI~};|K9ep>M|GLSoN!hHp#0J3n2Cuz8WUGZ$Y*rQ`{Dcrp}x3
znrW(?Wn-rsxW5^V%TlL+sz0SppX9%$0i3!ljS$|c%6&~(x>DIWTpky?t9G6}`__wT
zH?nF<AZ>9-B6RvqJAeb9?r!p9-z*O&T6m^!&N3`#j0&X&!oRv$cU9kOLJ4(F!nPw8
z4Fg-soE5jP1igviDh1*JOdvt9e}H=RdwC%{w)he!vZc*>CKt(3K3T38okc{1kFQF-
znCF$E5$~?Yqo6$7QVRmGBA|J6F|PBLYGy&@WD}xc;^ur+o95(X(Xj4(PQLk5dvCD)
zm3)51<#<g}=2J~o{^k(u`|!sy4O?E9@>R?`@O1#te75`W-nkvJlUM>!%K+heY!r){
zGyT0$u`$6Sgg4T^W7qP+jUCf}pO*ZWK4)t{=6(O<?3JP$m7$@&pwocg+5Hq9HO1iU
zJFbGV?DVQv3G!)E$LZ&IeU>^B1}+8;28I?<6*E@{&TnbTN`f|731~+LXV(<|_pY7K
z8F3##{GyMhm{dZ8&szgK&u$<d*P}dhag9Eh14szr*9)&wF#m`KaAF^@8Oq>#rfN*s
za(mmqHPUEaJ5E<N^kviNji~bZGj`rYj7iaO%QQ+AzXw3h=Y8z{4Mp)kW|7^YH;^)o
z>bR)+cK5e8c+ufR@~^Qs!aoA~3OcB-ykf|}rKAW@xCw-Ppw?sild&<EC9&UKp!0om
zeo-jvSJoHM`*2SW4KuU>3@IYG?Ll=sZ9TazUN@QS+&w<Ky&By&{3e|YoKe3c8q1Kp
z>T-+8_TGgoJ1Ac51}2~bx_LPyc7{f@*`~zb?CL|)F{Lu`jM?60Q?+u^Dq5=XMrxiP
zFUbl(PhYG3<7gfwEPL*88qM8_(z>n`lH`UUUeFYxE7l{+*qQ|&qBT*q3|#NkW0N)T
z9FS{D@zOy@JgOx%U2j-_HwRX$1maUs$@|Wnd99Y*v^3%hzE2B0!?(lHO#-Pt7)&7u
zwY}L;&zfHE;Dngn*j3E5&&6UbW)l&JBXGNI;Q7f3FatIKR}xNleYrK&T#7b1>j|1v
zJTF$_hjU_xFlk@PWy0v(vW@(li;?5c93S;Pdc`5qVg9W-u#_H0`UU|G98}lstU<YR
z4Ngz}b}=bibpMQmG=GLUA%Rw+#2S8^2`j`el`s+uIrYxYCpm)I4~{;>MtZjPnS>=&
z%(U6m3<eGl9de4-(d488oUHYZ^(GLwILqIY*PW<JLSQAguzn`tZg1^>$^7#h{`>pI
z(FF28f1GsVn}O7KY-@zHmfwbkXP+}GnM54$74oJwHY5Yut6UV!$Oo^8RGC86VrUxL
zUl=)%Gm=x1RC`Gr+rX~C*BlVj?z>3OpYT|WN`nZyU#52cqf!{c@67jo1FwVfcO3!=
z-TZx;s<y{R3H*<3O-DavgXJ<ioSTu}N<D>5TvkgpC-Ai%Ugz6vor3SJYRZUdigGJk
zhJ#@4y}&2x?^eZeWz_Y8WQj{rwEVW6O-oL3wqwds5Q-d?)IK}gIS*C$E$SkZHWD@2
zObvug5@t(1MJ#%J{0iawD-HuZiMfTbs0cqF(>N47XO5|6)3$@@d++Hy^Y;5(K#EDo
zF=o8_n&6Y9BzR){onXn_?J;bsg`HPl2VQ{exa)cjN53a5=*2eQ-244-|EB*@6^wG(
zLB6nL=I_RGxhWDYd7X|%<VLpJQS$Mv99#yGb-14Ly1%?s;Sj0T-Wy?=<l4tiuHxvL
z8A{R`4EXfXgx-3)kN66;8#4;tJtQ-@b0~+DEY|m&u2F-iaF9X!_KRTVq7dOZH*Ay4
z7kvAbFhYi<)s<BB%UmtEvLqFUqbseSF#mG9t+o?(O!6qxs=M<^Lc6Ta=$ajaCB_l>
zOs7Cequ`;T!u#?WhWwW&9Xy~dl?##n-6Sw}cO_=Y+vBSrt^cfx4#XSk%c_pyWWBfj
z^9+F+6zs67x|h9A7CE<Y_i;n3?tf?s&4Yi_6ra2Wh}dRKjSaG9ZSzZ=30NyTqS=M;
zAe4RAAUB-NB-{qe$wJo}LmnTTlq#3Vx_04Bej#;#1Q0Gv;WWSMBuzp}2@limS7gRp
zz2zA#jYL;4Bn=<tNRX>a!@W-x{-5L)dG}lJ1elMj`sqBvT~m^av)a2>UY-ni<FbzW
z>v~<=uF8L4SNtl@P@cr5(?9;SAUjdeW>IMQ{d{;lmxFl0EBF+n86(><tujD;ScI9N
z?AtW@&xET&itFi(!3c%8%=#oj(pWUk&EcVYr_(10q{O{?kJH{U5gsE+;}(6-tHJAu
zq@Z&BjF@&!8_qHquB~e-g(;*I187N#1Dqn*b!{uHhqvL%J+}nE`z-nkA$IRw$`ZL(
z52{STAxaAG+t%id^c^?v(anMAvLZ#3Ac73mq4R>>#JTD`WkWlh1M9&~m)B*FDLP<`
zLASSi*ARM)!Eyey*GWIk-G<57t=mg}0(zgW!nj%tFPGbX*f8+T89@)FlK)zN%`C;*
zhB2#$S}y9wk(;xoq)w8KwQYS#z<er;kGqNB3mS-f?wlXE{43$pCH8b+i~*d_g1*Nq
z3oS(rwZQIV03&p_t5XV4cK`j#+Up6#)%rhE5vy2RPVd8_PGl1r^6S;%A#|h_hCJS>
zfREYS-2CrKBu{p58UZLzcT9rCNnJ|w>A`sep8TEfzbVL{6}c{_>U<vrCEp7b%jEId
ze$gJMCPyV*Io8GDizc42+r8FRpb+G&@z&XP%j<YJ^ffI(T72M(l!lXO0}_@@0gcH%
z0(F7>f0Xp`1-2LRID#^s-P;AxNXpWu1l(`8z_kr{dbVzQvnc-80_OCO@mK=A8K$bv
ziM*H~cP!c=%Wkx?<^@bw5e2)T0_Cj=i9?*sm3u8redSy28zw%v-B9%6j*nLtxP%(x
z1MubI@oz5&JTH5q2d3%djv-CwMF8%n0|EgW@uDLX)=PwJ-lk$^QRIxUIFpIyPA2+i
ziCUw7cbxRLU8OMpk&I|<4$v75W*^E^j(!r%h}KLjt2aj2V4RI34!^CBVX$W@KRRl%
zVP^AgDWg|5oW=0}a+?tXHg;jB4?!W@+zzmu(=d3_wO-Ok^!$CP&1P2Jj;qf}-(V&G
ztJY~91^k|k`P^&kF=*lK%fjONzSZn(cmdN}zqN13xjDe-_y&us@7>wnpHftX$amp&
zU`r;rOkv3OKoM8KcVUGr2EA#JfWHC3>u2h6B?sARZ2g*B)T!C@VC(%K<E=B>93n!=
z0gXjbyPNJ`Y@Z6*1oeTuezT=BY%S|yEZ@3aOI0RbJwKEpU5_PD<vspsX1lLuN$5RM
zV&z?tZOTqA{}h{YY&a7q`u#ODvY)P;+5zAy&?xNE-##y3a!E_5lCgYJ+=3S4=YFZp
zizjeS86r{S)E_x9iVmR^)$XQDx+@j*Ug~s!^J!;pU8+ZJInR1gOFD_;ckIfzF1IYj
z`qU1dzA{pNemH56(x$^+qh=SMui7%p>^^pu$@Fh>+WdtAH(NY8H6>;AoqaDofwMSW
zv6yVik6b=)@-}8NGGNU)Sg?X>H#-Nya^$u|RvoP`&V@z(6$SdNYnJL`&>0X@XxDW4
zHk1SCegrsq?7h9M*8KS#Q(+Bzm|ShwaVVp7Bjsj(z@=|sG&k34fS&_&yJh<}uo#E$
z$B;1*4PaHlulXH!O9_0z#oFH5-rYv~oRx`v;ADG3av%pock=f6a+WN8x3FYDZZHOs
zS%9@PVr{d=!A88TB+7DGv((=ma@XnW(&0#?@(zT@e_QP!;vlI?Clm}1yBR6~iiMnT
zyn1XAYN`j1hiOX4z%q}7uDGN^x3A8~2>cWN)c&l1+wFiV)c5u(=gFei@Tego4W#_R
zM2-mFQA)suS;3-gO~u4R&?=(`Ir}ay+7$lJWUCDpQ>*pw`70)?Z3F|Xx|}G}>Wds_
zT&^>Bh}mHiXtlzrSBj)9X&!g}33ClTWNMeViKTllAJoIjnnBBchp2cFcUJ1?IjBW0
z<2J!acoy^OHQUgd>!D>X^@O}erA^v8d+kj<5Epo-Yil1lQ>aTyiDS`A2@&#vnd_|f
z?uumc9yB!2rmuU75q@qH1C!&!dthx#*;R0W3b_o8LtsKq(!iKw1;0sQ6a6CS${`2;
zZEVkn+u_c(&g34g&*k{OU<-Q}rw-m^!VdBh>O&E@#Avbc1TH-Em5Zr0^4uLC#qszw
z4Md;^2;}JVw3>X^94op%Nxp2OKE4f_aFV>SsiBw}!_Vzlqv%q=DQL21{=c&jE-TeS
z5Qd`wndYwfX|P1|cA5cTgWZ-X1OX8z5w96bMv~cy&+b`Tlg~s&<$BKQygo17@7`bU
zb3nxF<&OWE^0H$e`9Bzdx+ym8e;|N<K$wimiBUe8KE9))^pM>z$xo~&+?(G~O)vaa
z$7G*N7qDa`{?qzLm%viH9WQ-!dNJf6w*I!Ne;+v|7xTF3XbbdaE6P_)(Xexr?$KN<
zq~6^<tfuVK(0lJaVzwITjzGy2;77U9ELY`I%Hm8doVPzi%2^t}S^9^>Xi}nw>u&A6
z6Ztj=L;H_L)WYtHGHWHiH4P*Hn1-&oYr}VCz<K|_Ht}orii0L(bk`c>8vWKO!{HMH
zr#D<RLi?nw0+saci#F^vI&R%=D~<<$M*Vp+MFTcA_eu<4`SonS9dTN4v{DSP_n=Du
z4=4diC`5Jdb|PeQ-FI*OHfTjU8n7zWN!YjLNY}4>3c}&CcX?V~Ubxk%9ZM=$)tpRY
z{?pzOZG-edW3Ob<_*dEgMs>_lDq7|l%6757PtW#s83?)`ULBc}=&Qn)P}4lDudBk|
z6L#!kW-k2pB5eY|k<pDVUT!6D`Cmf20J?L*QrM!R1NBakVaDyfFXgFTaIY>nW=V@0
z(y3}CnL?jkT@!(0QYeHS5AW0#$HjtRmj#I<USC~v?ZJDU4n=6JEtk%_pHwwgU#O-O
zWMmD|3!OWh)o43V`W^_AE}o-Z(K#cq-8AVR9X-$L?s-gMwZ7e)aXjl?!3DEvMyqah
z;gV5s{F;bk({a8TY;eoAC)DHVtOQj5KGonZ|C}cL9Rc~mT(YhERpZSK6^^_sEn=v=
zfghzt5F;~W@p5Kcm&oh2TCeb5ECxxIxMT|w8b%)%RNN-gtg&fv@g`_0_d0t1^p63}
zCvnl1uOU|o!;wp+52J2<R_>lvc)FO!M;W?Xg~H@<s2i?fTn@zkF%M|kNMkR*kUDJy
zPF@n$6FN#aA-Dc5cg2#@VZqR|d2YlWb}cZeyVG`DwpJ$y)LyS0|E3=}#_;CqmR*MX
z1`$XOw0l^f)2X5s4^zkxV!0eri?}p3z+qKxpNiA+SGIUm+^;}~N7rli#dc?;x<3E>
z#qkL73ieNu*=-ra0m|E^DT*#iG-tBpJvEA`a`Vz1<1Gj!1qLFtm_CIg*hu`Na`Rk2
zLy5hRV{4<i=c2|HYkWTtMS9XhP%kPxHj!cf4DfU~fEaSUYOZZ{5uGi5dlYjqq~>Oo
z8)SgLA&W0Ujs(N7;qh`h;&U>XL7jk_uQFvT*!HlM{`eZ%gsCkzP+KA|r1t&>2@{jr
z`1@{i4_qNSicJIm>K9PwVNTs2DR?=z6Hc0J#cr`4Azz^9I_eT~YS@#B<4vN5?71Cp
z<8~4<N}4ALrZtjR41aOG#ad1Ka5)!ATi$248r@$aa)H*nye-ny=~x0Q0i#vU)Gi|#
zkGR`#+Qx_b3Fa}jImV{k$KxoCyw8{XE~KIw=YsFQu{NT4TAa)j-#kJ}nNP}o=Ynvv
z^~Yf5AUQwFem%L#Y;?qF)`LyC5p(O%S}i(<#!^U+s)i1x@n|Wr#En{U+*ALqCTl2T
zc9CV8MIJ^z&(Z?r_<jj?mX^*hb<J=&T=<IOR$}c{>;g`Da7}%(i@Gn$hwwnRg0PQq
z`7?ey^6wx3x|l&uswJ&dJkf)}=eKVz@Tyi`XbQ<E=#Z9BB%@744X@ifE&6sS(M+tW
z+ddsu;8BK~7`&*$u=v->SNi9ZaQ#WPT<S8~a2GyQeL)SugB}|S(zYEKd+_5GqN%;I
z9)6(h$59?X#<k1Gpf0ZW9&s9V4%0rC4gxR%l38twTPh`j%2IIq;imXC;3LF0#+8=U
zF|X3E-{Iv)uFi(ey|1jKm}D6|RJH2dus_RgX<oodC;+BTAv2_CFs$KKl_-EY_5{lk
z+L3;BLByDkoUbVc@QmPRK(C#P7kj!#=Ut8^sMBBXAIx?_w9WoBv3!f~#M!D*IcK(c
z<(R_gOQbW;lTdv5i;B&Etx-m+ugso!F$SSqJUm&|CUi~Je74J{jW$5L+7PZAtGp(H
zx+=v4%aFp{U?iz&34(2XvBxURUR&8JR4u6torey;O)LW}Q(@{<QdQ>WJghdpyOu8=
zarE(4All$)FPl6C{l3C5!?5%>l+|N>|85u_c(fn$(eV5KPoTWCob}(SQQ(zgDwdmJ
zYO;IAp1u~QiWgY46~Qb7I+TzI9o2XiN}-zHGUH*GqabMtq0wz$OlLF^_q<7(-6u^y
zU?s3D@Q8qwmc9|d)^4Xl$eFSeBR{FRET0xu$tL@<9*wICb4es8OwPLslel*^>cOd=
z6t3j8GN3lh2G(NS_A!P&Z{Cg4ahH`6Mm01hDxsKD;l#wSqyc#L7IVo3dVzgh>}9pH
zc>CN_^XOD9xQ$bnPBtnCijFGY;+q*>u1+0g=xF83(Ld{VSUQMFrlYui3mr-+1Ysy}
zSXcr#t}&nkOEktaAMj3l3#ac1YcezSpS|VF_O*lRKze!`Pf(h^Zr?^V72l%6CU*_u
z7jlbBjC~t=PJ0aums}sg?95}$BN+p;Tp8{2inv2ysQ?hofDuU*a|nnneI7RI!>yhY
z8j8e^HKeo6r51}hv_3`~L#Py_@c&i!l|gYtYuY4eLXe;Vf&>Z9;0_bqb&%lhmf%hT
z!QI_qU;<=tx8UwNSa5guo!s4f_kLBo`~UmXU0rp$s=JPLKlaLTLG-SWOv-R-0_kg7
z(atB$O^9AL9P%H(Ug&ruG!?0v%nBDo)oMhGtC9LKrngC-&o4&hvEQHewwOCsK0e8H
z`3BS2hgWye!_Ze&ze>XMa>8y|zNbr<79(ysB2}(#T@Rm2AR~MOs~=QX<Hs(VRT)g5
z3;4Rf|H4p6mWyZkp4qoR3Euq)vG81C#MHsCSDS@srWV28QOv7j53X0-J-ACJsbxM%
zoliv90X%)J8t}*D0dUfRQ%Wt3mda%F!W-Y#y_+z|OglUuD2b7yC>PaTT%+;4sM<4~
z)Ql$BxeH3xj{Wp}0~1Ot<oohNB7lSbgDJC*-uE75xYsE&5b?Yb0YMw3(sf^h+I~Xn
zxGzt(MNuo%Hyfuuq%y0uy*V%^wR%WTe{=|F7E<ZRZu$eH+fH)UIXntj`vTF{D=}?K
zWp$W;&sfG!a}T+caF>bV)s1V;GwzIgpG)l({Vu1!|IB0GK-5s8%`q~Lm3&S~rg_1T
z@iOlE*Ky@I8gpMaRaU;C(4s5_y^D`ZHw^_JKSEA-p0mDn+kDu5&2S}Xg{U@oM6d07
zY;~)9y(u|@lPlegXV$PM)-J$?JoOE+&LBF7IOewK4K0;WdN<7g9eo};wK|vj>)6s6
zhxJ#2y$TJWtMj3~3zxEgJRjWf&Qs(aq1z%e|E5lL1AJGHocQ4AhGmlShHNoRI%bEf
zPcroAcu%W6?5hvSu=<X`lq9Af3If8nfsL~3MsH_f;icla^NwHm=U}BQehbK9Y|#q7
zq;1Rkv7bwxVgU4`NM}toYoSQ0pX}F(<yLevO4W}-jX825^3c|$O<8~a))46AW2W`=
z9c73vZM<t-=>Y<C6n`S)3*SLwFEcx0(C2>23gIjRA#eYN9>n0i(=^Wlrr^&IQ}>3u
z?#6~-e8x*s;T)DK)h`V2uX}`Ve!=xmf2OxHfsX^jGIAlIcTZIG0sx_j{h2JQOuW4*
ze@tRMIdsro;%(7M_~77=N_o-$*I5kpFW@lgrn|&hp<+}*CPWj?CcEFOaI;?GH=DYn
z+!(>_bpcnJI2;KQsrrgl)9qB)sM*cG8>Dc2l)O*@2jA&_y&vLy&x$s>&t$^UB`P@m
zyvcQKm?S=G6tHny!v*=UjfD^ylzDgC9e@e|7K&28`9UfpE77FTRNGj#VTf7MRtm3c
zZ%oSE$QV134JCnZi8Z}&*%Zq`%93B#>h(0Y87nS1Ga2{Sb3Y#P#lL%H_EFWZUMHPM
zV6)qgmos~x;4^e*(VwLO$^*N2-*!4T-3!QC=C3iae;xtDeMYlo=ys&~KNekve<7o&
zrXDSnddb6lx31PvHC;!mi9GmIl#4!62<NTktZw1Ze2;?q?qR%#A&*W=s*wqn3-Dcy
z+qD^ol8FWXe0=KsNWZgRqHL8rTYJFkGGgDm*qTYJ_cY9zg)l|Jv2$0sJ2|WW3_H1!
zL1=G7AwQ+6Bu=_TC5b6zz5COmKn3eN8Qp_hcRP(whhxIT5*l<}f8NKY>Z!}p7D@Fv
z-uV#18a^6}%~+47cU<2St~$z59s%}p={3?eiaU1>NA-`{$`-zvzF}My1%Ac(UO>g$
zlQI}4U=9hv#?~Uazj?jB^OGnhLSrW8JowpA+#V+O{X+B~IA6&I`^!IWvbsq&aqew1
zcNXxe|B#P;@qD{oND!DseDx&f9a^B>q?oxtr=A&;`_ffk^vlULC9+A?2r-<e!M}(I
z0py-~g!o6rhQ@j#3V6+)QD*qxjFMBysR_94=co`{aMk72BeWHZ%0$+Z3hWJZX3i1(
zNPNFW63c9{SY%VuHj5r}Wi7D~OGr)7^dpm0ZEzIsBfHTyPkD(@Om}^z9MDas2Y7PU
zEB5Ev;{yIAYc8iJ?%>P~oUG~o1`kvF7g>{)>=2Vky7{=4+i8;gt^UDUYT!$ls{it`
zfQun?$*ec#qEi3oKc!2{1?NK@5o)c~5ie2T1P(K^5VRxW6v<V^RohzIK5+bW#|F<_
zz0|D0&!y3It-aMmGs{z=K+~J^U*h)?At9l?M}g>*17fM3ituqad4)j+CQ|3#5qQlR
z@72Cvq)nJ)K)?O&-UL~azlml4q<MmQ|3&kBSba?IA4t;C<6yW5HK%EwgGseiM{~c(
zrAsaIJB`1_{u?f%9*-3dWAVE5kWzXrT2mX2lj2o+((f!p`F+B!7xnkAg<T$S--l)y
z8;`^TCMZ?>8?yl~mJh~hk32%5058$dbv3@9KjwdpWk59vb!=pEptE1ME%e<^f9hEh
z0m+8BV?>HQIqaoQ$$l%k>e@lXCDG7g<v$qt3_u1Rm3#t-SVYVMfe)nQ{DRIGp|~m4
zm(h6L_k6;l2IGKQ1gH@hg`W?e8oK=y-afUrwNpYW0Dv9NRAiFOGU-&>9$rx;uXJ@3
z2YCAbqpz?)Az|9u11R~|OMcty|KOC$=+RWaPV@>w4?p5i{{&xd{gX}-s*1<aTD~(y
z3K;y~<THnlm{d3__U$shyq%gCo>@T!aX<6au9sFfQ8$}W<7+}UTuC0s-Iy*GtZ1aD
z>D|)3&ddyLRYgPYhcO<VAt5Q_{=Ja7{gY0-(<l(Zl{#(Q1`I2s!S(nOE44Yf1P{!L
zn>FeH2v@GAHNfb#ehr@~@zQCCZcRO#Rd5kewKM>=iH1s3c*LsU9NUuZ#3-UUrFr%V
zih=%t!Pp1L9YelL6}RQpdTLN{b4#gykc6o6b)@P{=4e0Ut&4-Al*Gf?%YY|GL<&U+
z`CmRs|A7@*2-9tmSeaW@_<B`MKOlH|T)NqxsCHKVJGE8gxzBCLSM!5uBiXf}lI7LG
z()#?C&9@cblP8czuAQH`{@S9A;I`-h@BXHmMzsJBITX3S2qwK<JDoS5P9(#OYXeWD
z82X}xD9u2M9R?k?JE_KE_gk2F+>B@u(=sJcUIeQzL_C-#+60L<9sdm|MFx=Unjl4&
zw$9?sEM&_e=BlHLeM=OL9Mp=a%xr#AJ5o89+IB8!@EfWn#s;*jYG*eIBZ6Y(oVTtb
z*A+>xi3H8fr8ms2?}g!eRb~?d58&W-=hJw}tP;Tfea@`QJ}Me$4;4P6DGs{eg}xWd
zY|ERSgQH;7FL`J=5Ou(Qo?n1;1|D;1(**3FF~UlLX1hBxKp(-=O=}eesW(#+VGjo9
zTSek*FEf|oH=D3@#%5DUFJPO5Gl<}>(P^RhGA=`LH@3$6!i8L>+tvNqs~CcMQR>}c
z7-t!C+`~r4I(%K=2uZo^v7zm$gK!7Z@=Uutp3D<c(MXcHHMZ1)cK>+)8+eK1Ed9cd
zys+jML7-cbcrncGVvmPEGV|bWh$lW=IN6xr;dtPc7)rFBa~Ib}X4x*eyg^@-%N(oS
zMaVd4W-)|rFh}1Y@{DpA?O$ZxmjRCd<nl6A|0h$^CJ_A5n4q&v=fkZpwE4!zxl_L-
zG`?cTVtO1572Z9!(exeqeY10rDyn=4%&X)5S3pkQ+54n5y(YI<iY>ph$G&#l_t=o7
zqM=>w!>^f%>rEbA5gfm@3TDKT<aypoYlu&!RGO<u&2P*u*j{Mwy3`Q0eOaK1MoOuA
z@xW~CgB55FKEKPb2qN7FwVn81GV<Z4r@~Q)<aVVRaUzLZptNq6qFa;q-f#46Lab2(
z-^96g(~-X~`s%;f#FsYVzc6`#Z6RV5b=;)N{V}0V$s@JD$<+}dtz!3#@slljE0gs@
zXqvNhwevYovc?|+Y^z^C=q;VsBlXN`9{XY^*ad0J;Ld3g@x6IT`9YoQp5mX{WR@0Q
zs)4<o5=}3`Eox!jFAP7XI>8)Vt0RrffT{sv!!Zkb*n6+bi@uCj*qX$v-&sn~_qy!m
zC7ToXr!MIt=-(uz5>-?|AW$sic}SS)qg1)3Y`$TlqRgOK70E-sWdfm<Bpg+vn2`Rt
z$FTEO&8Phiw}?@|tPR^wO`+fp^LREa^I)A!_3XS%legb=3|oOxq<L1Eu~4*)hs<NB
zxXvM8E}<+~XOIFZE-vh4;-JhuTt!q7nVB4y(M#D+&yXbX5l<E)7e6_i4wLs}@p5{J
zv;{983a@~e!h~In-S@vo`5>qTr{cIj&w;x+Xrt0uow|R8lL2K<@L%$JtYnd{@Ol+Q
zddx<VHl|z>ad%!7IN9F*fKIQP`gwXk0q>tjfHyH8Zn!o#f`xz;MptzE6SHYGt6e4U
z^z*5MZqFa7l4Eo+kGmTV49Ezh$bn8g6g4wCOx5*u+uWlhX7khaK3?lATaP`%S{B!=
zteMw>ik^S+oiv5wZiuuR7CE0up5b}Cn%-L46TP&(Jb?GWdZ<L5qU51~bIp71*{iwr
zoJyPTjm>Cb@gz?jzMUDdGeYG+hnu0xBHg?SJVd`2f8Aab-Zy%dZ>RgOk-=%k66Cu|
zA7TVb&%2WTn_g`rI{fAacWPBs*V(2H2~6UJ;0=l$LnG~JjA|L*kZ5L{yXX0*EaxS^
z?L`fLyGgq&yFQy#&tk76T1?;-z%atut=P(~^*w|xx-o8~r)eyXYpXs%jVvp`EjRO7
zFukqs#uc0`n3~7czp5FTp0(Yq8!SGRcK_Z01p%}im^3x8*ck5}c*;rbab9+;iqvyZ
zRrL^!HICOy1`>v4eoElcx_kM2r#=~&MD%s&eR_YQaV0I%anaxiCNH~qwCZM!G<&8d
z-^a6+grlZqxPYIsaJw~T<w0`XV7Jc6WZJ_IRkT31Us^TfMkfTTDW=Q3`~8vDA@y~0
z@+L5o$h)0*|0G=5bn0?}WS8EA{<S;e)3Nt<i_0Ov|H*^TS4z@K{qa169n6^Z@zuek
zIU+oEE`<Bv9bOFx(DnRh_simHbY{vAiZqUV@bn>fF|>dvx7!$U&MDO}>@AngrDVa*
z?*SMU0VTF8ACki8eMe>m{Xb8%1Yv5d&4&GZ+R&L77mdpG%#lZ|+(dS*XP)e#6keWx
z62Z)DvT^lB$M3g}{Yu=Wjj8P`zdDowZ3VAyx1>#R8ZV8FFZWx}o@g>6>esRa$nSpq
zK=xRiw_GBQWSnatj~-q*h>NUCD4)!nQC!|i6JN(kygzm)OVsP2lfB~S)`;&K#%Qhg
zUeA~RorLpZam~0Hz>}<@+Tn*yyFQ<$wV><A+92+dUZxnU#VI)sIDQVYQRYK77-g>T
zN&D`b2qCD|BtIZ4j4~snw2EGE+3KhXCQNd<SrWH$ZmW_qD5?yF;j&0<;cXC(pRGOE
z+mlX9s!Bj3U&a>HopfXl3iMi=07UFLtXAgFw*iKLL8z#-Eq8peVe5wKCu2`p+pfA4
z%jCqRG53JMn^VK{@3Y9dqKAME8L!%}d)!3^@+>li$0LL*mE%A_|MlD#A)LBDgAIl0
z*#*<(C22Thzp9yN*z~cEf$o=yK*7q@&p>Z-0p8}DsrG7Nga=4tH8|)0bc_N0|7Sbb
z96cdQNvykxE$r7dJqbg<q9BK8Tq#U8kO{a`-A;q|e?=9Yo~G4QgOwYPaZ;|*@%>z?
zF=LAb_eAl~TTqt95-AIpyw<{&dI5{gZr5kIW*TK(*p*KVb%GD42!;bXpb-9Zg9M+i
zTKzmhgJ&jo;;utA)TJy3IT&ppkX7EfGvpo&4&A{&elW}6%xqIu6xl|Jm4cD?@+6Us
z!iT!vvrN{`L#8hG-@Gc|=i48ZAjJLsW#MKNG&t5ZZCV<()dC@1#+h_+usoy!YguWc
zoJr!V$aCMJnOYiI{O}H=@9>k20Ngj87~tl@%hT!cZ5w?bP0UxyM+z30Y_?rkTNpF)
zqG-iv^LneXA%f0C5_u6*Rbkc?M&vmc5n)e~$Ge6D?XLEy>6^j(clA<v>k+qRe3Hg{
zpHo_W5Un%t9v`k=swct}!}z951;GODKdqfmcd8Yw*Phcxk^yZFW2Hj1*D<rS0)hS%
z%WhZIiZ~hKv-u@S4%JSw4znt_24ZTYf{PJp@hUl7hVIvulIX@Ixrhp(TJ))=RYqdd
zgVDHCn1jb2{0HxH4R$=No*Oqy+_K!0icK5Vel83O`7lO5L(t<!4ko2~N(gx!v^-hj
z@`2Xl#x~u@irZJEI_zGsi$I-`wxYUhkGrQQ-?>+>s<E9{CeJOV&6gF6HaxFSs3F={
zyA;EWU&H+$+fH6*yaBjz%72pEh(biMO7cmSx!VFaL%EA?bWZB{$u{_`vh;s9d(fh-
zW!_;;G1<KTKy<;n+CorAO?gA2uc<^$FCihr*=`&(7rkT6VbzU!zW{BeCd5q?aI*oW
zHE&)QFmEqJ+0n9+g?7G8PG}G3OHHjgmUnH8_EIKmEWvi{eG1S?K4hpAaRa$9+->9g
zqDL97*{Kbq!i|-9u4nVHYMlN{1m84S+iVZ%0?3R|vlQEH_6l}JKD+!VXqhN`2SxZ)
zKR!YEu+q@%=tmx_J)F!9WcgDwsFR&jzu03*&Lk6laeeYij>YmKoYB=tb;m1OQ&AD>
z(+d8?P?}Wy=SdKw8&-UI5BRwygJW%}u<BUFf~m1Ji)$Keu%<o+gw<}NvLMB2T&tKM
zU)Vs~1#oev30j#-b!=WQt<2yx0c2-5Ng~_c82Fj+J=s`|I&HLAJ?enQTGnb|)wUhz
z!s4P9e`aXkn)xjq>%j5swwFibdO#ER<0#KPM`P+A!OybPku=Z~lMV;{A6LF!UZwF^
zGM&vG**rNd?*SjF>e3E%3lpm$i4t<m#0|qkRahArS%)OKVY3hCDf9+Hi-MwYB)pbS
zSlO6qAt~1PS&zY(@Hls9kb(HCxbb~9_Z<W|T%o5I+e)D&#8=g|2V=60<5-m9^pQ~n
zGh&hv?`-E~9<fx7UgXjUZCa~p*r_hfRF`;A`$kF^Oe|OH61aXdW}AONLAZ#Byq4|m
zC(?>0<MF-gh@2eWzY{WHuiy@4fYR8^VfmC<q0bH<e4m|4cW%p%K5`(Q2Ya^z>!?=`
zbFEBx+8~hg2n5An&!<*;kzPs*?hmVk_M?Hp<TpM_7;}>iqAP1K)>Dxw^(R51eBb;r
zpu_h=CMRnj{c@E*tFqy$g<OT-wH<9_w+btak@m{bZ|tV-?mCMQ5!%Ti5U#*X$Hr{R
zHZqkg_hik^FyS%8rVlct&L||IpfdQ1THZ=p-a|$;`D}XTK!UTslIG8Nd%4@qb$>Wm
z1bfF|Tt@a<Y{dBZGCe4!IeD=9_U$VfqxRGZOa`-PsEN&Mj<wHNwuvdj4W>rB3AYY#
z_ca{eAS_>Nr8Nxb?z0mhK7UFsSjtZ)XnPThe|hV~2BipzPtMR)o?T_677G8R{a~3x
z{nDpwpoNpygWkojY+&vcfJN{|V+eotfZOq?(Z}MbA^3*8Tul;faqUOFxJKah8qVj^
z*D{~q87w`HE2A(!t}JItak-S)A#7(*V}D9TMnWp^&tngDQrSab(8e#FS^g<OkI6f&
z-kTRvlk?4U+Luo8b4eK!5-PTM+3OL1iI~@U674S#(L~U2C_0lfkfgsM*&gbxE^pT*
z`K5eObmeB-K0RnzDcfbAZ;HOyiuu-Qb?&QE%@<1Ca;>b1ep->&9_P`fkclU9>E%2f
ze7=GMBA$zOpzi81Xg1JYvZr+{k<}{P<T{%yK;lA%RG_*&VK>t}d|Iqo98escv8MA%
zdDkLfYSXnkY2{(|nwuQr7w;!<>)cUuKp3}T<8vDy!T9R5zb#o+NiI!Io5qdvS&wE~
z@isatE$2#o#-Z&1<576c&{)H&gcg5YQ@G(RA%3mL<kpzpy`Q4ApF#@(6x^K9&9EI(
zl)&Y<Tuyi9Lsk5A?s@QCFE1*T_g#9L=7$BgOy1^vl=G4^Xs?tV78vHj#D=&~htO{D
zv6R_E<>O=A8ew)=7y!q>;1x!%l236;9(Z+&XF)~JxdOL2WTACG$`qCHbsaB_X;R60
z9tS)dH<(hs4!Xs=7vlK(_%gMewC-mCR`?fmK#&KPqc#D+>FN4IX{*PEj=?BoCDUh7
z?en)0)3Dhm>^CMLTn3W2{IH2(8YF0Vc$Za#P702Ppk6|Dgy)VNICttfHIY6G?um>@
zmXV5oO@pY<j+ZSyrQS(6<nnej`fWmwLwG%C+R;{Td=vXP_m`aUiY)mJR@}1sZpR)O
zsSanKGQY0mnDf1YmY-!O@2yC7Z6{C0{sNH1@Dr&-3^o07;EeMWEF5a);8er2(+QPE
zPoJ*y8k`f(H4?Vh^N!VRVATIAn$x5OqW=&yo+G0#=D8~^Y%XM;mE#-4&zE}-{hT^%
z=z(lRbvd<DTjEf=*kS9P)oiW?uqq=RKXvY0Eg7L}o2Tmm@YPKW$FO)B3~zd%x#WNK
za^!<Bk}gb<s+xQTG!7Jgl$>6K2@m&w^K1b}vAM3qP%q^hd|vszO3IaT&0w{O4Z3w@
zlNWN8tyMrp*h%Oqjhql6KXpr=$98xwe9;O*O_ydtMw+ymHcKUq<hvFX!k{tuown6k
zbdOT1dj$0IkB5MFKKQc<iZ+wGuRlf(2MCv>qxiHskX_7wVwNU=G!U+`JcdmjaM{AV
zPa9fKoZ0iu+t?)oW$IYAmp`uv?o`#_1_)zZ``^i-=vt=rkT=wOc}DFiwUDT~bwo!O
z8pJGC7tX|4Ss+b%kIgCR4%Jw`+3(J?AP$dG<jzS`ia|jvx<S5Ca62?(a}S9uz~HRv
zrdJ?j@1}jbzKG?*6Yrdu)UE8WcQ}0I%w4#gHPfeTW=8!?q$dn+meIB<Qp7R6ep+53
zjc7}7-JBXhrYep$L1o*lj~@PJwI5Ybuw7-nw=5{Q3jCZ-G3=dZCM#a;G*y-}-bg*%
z=kk<wPJ|2%tAdj02+e3`j+H<2|JoI0vF(GgRSM`eJtxN|lW#qLJ#3#9bCI6#*1joR
z*o9YLa&yvG22MAmsQNgSRSEaAlGI(z9Towyw+jxk1f-L}+o#9C1V1(FxMb}K%Pq3s
zgxTunfe)~WUMn^M0>4JVsA0wid?&-_)|k}CUF$c!9;^5CDzp>&G9#aH*C!{Jy&iMJ
ziK4YK8qW^mpD;a<tT@k8Oy!gR_Bv-u{DljUc-XXKW}|~Rt6tB8M==OhUhi!uI0^+}
z2hUe)=)J(lNQzW2?#pZMef#-p$sog>c`pY=cNPPBt&7BTH|whC5+abfHC?W~;qfZ@
z*CUn48`PW9cR;z1nG{^5=H&4S<(r8;!@JW87nErPRlc#QCBHqq`M>ligKnLOu1MPG
z9uC%tcm3wvN2Bjktm2UrnyxjAR;0<*f_fX@)_A7b0PGm3l+hZGq;O|5tstY68$!bD
zy+QP(C~sTT7gP+8dbgC7P7P=rp1S1UH+LyP(jzVwvQO_O&RW+RC5d}*uWRJRR2*?+
z?Vcd*TED>BXL{4KPDY<avxboOfra6{2|Wi2ER9GI9&?>Wd7<~NdD%i`0Vy3&qe19E
zX+!gvZ2Z)_hN_%fxA{9d`yLGom&*G;$4A^g*xl}<nDQ0WRs>I>an_DCYy8^I3l30F
zws-8^fx-wsXMSph?Yxq;g|V#3%yoWB!AyB1kIpjbu~tGd@_Bxtm4X`pHkm?t&_i*t
z)kS|NeVP>%ZBZ?NTC32jX7JfwfvRSymay*ur1HgG3d7zTPf$8|DjR=Atqcv}k65wK
z?H|}}R{^7&qg=y8Fsu{G+~4ALIC1;qJlVk%U$bU}{W#Fz{%(9pt-ql%eBN;--A{~k
z|8$8_W_u({PI1uQK|~lH6{lf+b}Pq2hF7`y5l+}?VNI&Wu`G>1m1odNGsH!w)T57N
zni!gP0@T=UY#LqVV;5;A%XNIk>}hz8R|$9|QP~R1>c&A4uO>`$-Cwkbm*nroJsw&X
z{62`^oHEdvg4ZLCMTy5s2z$wk&(&u?sZ^^Qx{u=xq$G;^jb(a#??MTlDuUJ=OLCmt
z)`%YEjmYR4|2nFsk?YqJe+ut|Xd?}K8%ysEx(aWFIZ1l`j8al5n)JFGU`p2f;vS5#
zl??xNp|Jd@zW{AkeTs%h2j-MOFCYV+HxWwjBHwXH^J8s5*PK=~$|0mA#Pwl6-=1YH
zw)WzT9R_1K=-I~avUkbEm$v$Gw1^Z-PL9v2Yub1Z`Ys42C+qb47D?{y?6nwB-6qbr
z5l+3Tuk#l^asZf|{N%8G3bzQ9wN8pLuWHlcRW72Y`67Ru)uPukbb=-5UREFPLSQB%
z5QHaz9Y{J^x{eb6F&ml+RL&RlcyzY`?TeVS5ZXt``modm!)=zIs#r)tH|pwK-X9<~
zIV-NJ7PM1vgtqKDlRYmxFe|CHHTH61gWasE<raVQ0EJ5w)Kqw-h?@qkC5U{ktyFCW
zcN=y%^JMl21iOJNAhad+&+j}MO7F+$<2_ppAe2TfE@hvJ?N`^>^q0Fv36O50l1j@#
z-#FjleqD7=fi}@p)$%f4UaXFFld;&#DH6DtbK!392Ksz&jhDLOm?d<RKMw3)RAlzw
z6AzFPiyqtPA~k5ZG<VDb*1KLQOMeSK4=Job&eS(UhL^07D{xb91Xl(K%lmU)3sC{f
z$3sQP6=#khwetQ(2dw=L!!s>XN;C#?7(^D*F?Z*RjPEY2h6t3HEO?O~mp!Wkd->P;
z3&s_;KB{(7E?yW)@zd+S<g|P+js8ub%^3vQ&SFX(Uhm>3yLel!BtKnX6a5<SAQ0ex
zdEoEWKd$o_^sWTAV6_xC&WJcLK$t5v7F1UKN)k&7dTZl5=ftTv#b&gg!kPB_fP)m{
zu30l+pD}w$?r!(~mMUTwsVw*!Ft-AaH-zoL0)#zBFnugLc-&MuAU=g@XvlUNVT0>H
zq4KAuF2X1qmMwD{a?F+b#p^1W>^Bt?DxP%#_B_@<h$55y=;<|Eu|sm0ZdD}%J~^b^
zv!3o%+*R+SbgbGa9z2?R1Gu8nHsSfoprb6ERcbxRuZZ#^(z?O}^l_hpd+en~)NS0b
zCrJfQ7oqgr_$>#)hXoZsIe(w~*EPJYp>%G2iK(kVfqH8vW5`nD3Ym~}7$%nPs({!O
z+YG<;A(mzh|3eFL>8};9;#tk~JOJtCVTs1qb8VEcBKTml(Gvi(>YCs1?r{n}^0^-e
zB>8*P%%2jTCCq$UuhMUYM}L)NqP*jF`y?jftCF7P@0eop@*&9bjKdWQi`5_Z_V_jU
z>@uD7gCT9WO=INXgD^?7dIteoBbI*iV^LKy2+&OqtF<jv7a)-=wh7_sv@PIXNU$7g
zbLZ0ppO6n*k5odyE<Bw{$HPBk?X-P<2iEU1nXOZ2WK`=wH@3XTb~sgH#>Qf7zsp&&
zOT@~4{_V-^a~tG&I)&>yKN!caItcrkqvF&%1@utuvbofLjd@)EyM_<KBe(IfA!ZGO
zSNzagn-ul0#nV%>aPX(}6BvpDk=)}QT0B`l1+k&ldGAhupiKl2=dmJo`|tEJW065O
zf$xPAW3a0LmPcqmDO2M_$SdxEPoupNBG}s8xC9O`{058tpIVA*61$%EA`pebAMUqJ
zA6==YBkwWDIW$Zc>rZjPL)zu83*QSe!jjUMI_?B)^meTa`fn7rIkNFKmfGHt^7i+9
z^1gJ?iebV#)3P9Jung5V{C1so(pfapO}<by?pB(K+39+C=$d>YJ+-2H*fCc7;J#J)
zM^B~K_i=Nq_S?7+7h2<R7^4FX@RWtGJwuXpNQmbkG6GGP73Mn8N+_6H&Tu405|=}S
zl}YYNtue(Go9-P9QV-ui^}!=uRteZY{&loamOdS#HOM0e!*6siEe3MkF*PtW^^m~z
z855f(Vak2OLaojz5O6^by&g2%P5B{Efl44Hy@OGZZB#^uKS6}mxlHQ*seEKG;(!7i
z@+G?q)rWuVa{z<481|b-;cO=0otIpLn6DX(Kxwa5)&{AwYdyGn09-tgLN2r<TGH@J
z60x~8QgP|fjG7Q6ep6av<fP}xIm?o_539dFuD{XMQWx{>zkD40C_qxDMo#$-Me+P<
z#U>RE<Yg03^?gHY94_r`-m=ZD{{bJcDITtk!(|iuhulEZmgEol>hdbi^(k>E=UP@k
z-}Oy85WY+Ter8+5Kny+MEBK)Jf&Cb(&)UfOqUgleQ+RglYk5Gb25m6bo)(`QNBD|W
z>c!VDm0W7c!Z4Y|$_LbjO@5!u!%dz_3tFbHXMEHbbM*k{Up7joc@()hPlqKy4v>kf
z-7|gsByh)7vb^M?8!2Nub=NQ9v?+C-pa1hdy-p*@AS*c)bdfYxSLs%Je6V%p4~tvO
zparaaK@(yu`fYdBwT-vblGr<bWVO#aCaY;)q2^;Xk~y@-uK-&39_15(=NggMNt3|!
zS^1s;u)T|}(sWTqs)u@Yu(1GFuRL=-GqXvke;K{!lfO@`c+xQM8`B^(c*=f@dk<|9
z=B}imX&1eF2mg_#3-SmjCum77e5H57PsI$qB%5m{bYl-g;>=&>MwyU;H~B=rsBk+r
z$Kwz7CiAT8891CIiWjb}JII~yWARAx;d+|tUUYLk$1>o(2`sD}Cxzb<BPxH3`XyZT
z>Tu4Ba#eCf4pV8W1*xC2nuL4xT%ttN_oy_Q<TJcsW<bgr*4k{VNB2+L9G7FG8PK&*
z(av8Vsqt(0p_ne>6a4foOcd^YML@t0glifQ5Ex#w!M7CzHN5|7(}f4WQdSB1TL>m6
s5BqQ0%qNKte_?|E4{uB?0AD>X<^-)YWcHTBUjaczLQ(vesG<LV0J(t}rT_o{

literal 0
HcmV?d00001

diff --git a/prezentacja-checkpoint.ipynb b/prezentacja-checkpoint.ipynb
deleted file mode 100644
index 6622c4f..0000000
--- a/prezentacja-checkpoint.ipynb
+++ /dev/null
@@ -1,294 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "slideshow": {
-     "slide_type": "slide"
-    }
-   },
-   "source": [
-    "**DST** -  Dialog State Tracking   \n",
-    "Monitorowanie stanu dialogu"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "slideshow": {
-     "slide_type": "subslide"
-    }
-   },
-   "source": [
-    "**Czym jest DST?**\n",
-    "\n",
-    "Monitorowanie stanu dialogu polega na obserwowaniu jakie informacje dostarczył i czego oczekuję od systemu uzytkownik, na podstawię całego dialogu między systemem a użytkownikiem. \n",
-    "\n",
-    "Najczęściej składa sie on z **aktów dialogowych** i  wartości **slotów**. Akty dialogowe mówia systemowi jakie ma zadanie wykonać, a sloty jakie informacje potrzebuję od użytkownika do wykonania zadania.\n",
-    "\n",
-    "Źródła:[3]\n",
-    "\n",
-    "![](DST.png)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "slideshow": {
-     "slide_type": "slide"
-    }
-   },
-   "source": [
-    "**Dlaczego jest to potrzebne?**\n",
-    "\n",
-    "Prowadząc dialog z człowiekiem z natury wychodzi nam zapamietywanie oraz łaczenie poprzednich wypowiedzi w całość, nie potrzebujemy się zastanawiać co oznacza odpowiedź \"tak\" bo zrozumieć że jest to odpowidź na zadane przez nas pytanie. System dialogowy potrzebuję właśnie takiego rozumienia kontekstu danej wypowiedzi aby prawidłowo działać."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "slideshow": {
-     "slide_type": "slide"
-    }
-   },
-   "source": [
-    "![Dialog](Dialog_1.png)\n",
-    "\n",
-    "W podanym przykładowym dialogu system nie potrzebował pytać ponownie o miejsce oraz rodzaj jedzenia które szuka użytkownik gdyż zostało to już podane wcześniej systemowi. "
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "slideshow": {
-     "slide_type": "slide"
-    }
-   },
-   "source": [
-    "**Dobry system DST powinnien zawierać właściwości:**\n",
-    "\n",
-    "Incrementality: System powinnien aktualizować swój stan po każdej wypowiedzi \n",
-    "\n",
-    "Context-awereness: System potrafi używać informacji z poprzednich tur dialogu \n",
-    "\n",
-    "Trainability: System ulepsza się po dostarczeniu nowych danych \n",
-    "\n",
-    "Portability: System powinnien działać w róznych aplikcajca w różnych dziedzinach, nawet bez danych treningowych z domeny"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "slideshow": {
-     "slide_type": "slide"
-    }
-   },
-   "source": [
-    "**Problematyka tworzenia DST**\n",
-    "\n",
-    "Jedną z przeszkód stojących na drodzę do idealnego DST jest właściwe inerpretowanie tego co było już powiedziane a to co właśnie powiedział użytkownik. Trzeba rozpoznać każdą zmianę jaką chce dokonać użytkownik w istniejących już wartościach slotów. Jest to zwłaszcza utrudnione przez nie zawsze precyzyjne informcaję podane do systemu. Trzeba inertpretować do czego się odnosi oraz co oznacza zdanie np.\"w takim razie jakieś tanie miejsce\" a bez kontekstu jest to nie możliwę.\n",
-    "\n",
-    "Gdy użytkownik zapyta się o rzecz, która nie istnieje w bazie system musi odpowiednio zaragować. Proste odpowiedzi typu \"Taka restauracja nie isteniej\" mogą tylko zniechęcić użytkownika do korzystania z systemu. Trzeba dopasować odpowiedź do brakujących danych można zapytać czy użytkownich chciałby zmienić lokalizacje lub rodzaj restuaracji o którą pytał.  \n",
-    "\n",
-    "Problemy wynikają także z błednego odczytania co użytkownik powiedział(ASR) lub nie precyzyjnej odpowiedzi systemu(SLU). Takie błedy mimo dobrze działającego DST mogą spowodować minięcie się z oczekiwaniami użytkownika."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "slideshow": {
-     "slide_type": "slide"
-    }
-   },
-   "source": [
-    "**Tradycyjne rozwiązania w DST**\n",
-    "\n",
-    "Początkowo problemy monitorowania stanu dialogu rozwiązywano na kilka sposobów: \n",
-    "- Słowniki semantyczne, które umożliwiały\n",
-    "- Wyszukiwanie najlepszych cech z pośród zbiorów własnoręcznie wytworzonych\n",
-    "- Tworzenie bez znajomości kompleksowej ontologii domeny slotów na podstawie historii dialogów  \n",
-    "\n",
-    "Źródła: [3]"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "slideshow": {
-     "slide_type": "slide"
-    }
-   },
-   "source": [
-    "**Modele wykorzystywane w DST**\n",
-    "\n",
-    "W przeciągu ostatnich lat do opartych na uczeniu maszynowym rozwiązań nakierowanych na DST wykorzystywane były różne architektury modeli, przeważnie coraz efektywniejsze. Przyjrzymy się następującym konceptom:\n",
-    "- RNN\n",
-    "- ReDST\n",
-    "- BERT\n"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "slideshow": {
-     "slide_type": "slide"
-    }
-   },
-   "source": [
-    "Rekurencyjna sieć neuronowa (RNN) - łańcuch kopii tej samej sieci korzystających z parametrów poprzedniczek"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "slideshow": {
-     "slide_type": "slide"
-    }
-   },
-   "source": [
-    "BERT"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "slideshow": {
-     "slide_type": "slide"
-    }
-   },
-   "outputs": [],
-   "source": []
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "slideshow": {
-     "slide_type": "slide"
-    }
-   },
-   "source": [
-    "**Dialog State Tracking Challenge 2 & 3 Data**\n",
-    "\n",
-    "Z powodu tego iż zbieranie danych jest bardzo kosztowne, mało kto mógł korzystać z dobrze przygotowanych danych aby rozwijać własne DST. Dlatego aby ułatwić rozwój powstał pierwszy Dialog State Tracking Challenge i jego następca Dialog State Tracking Challenge 2 & 3.\n",
-    "DSTC 2 zawierał ponad 15 tysięcy dobrze opisanych dialogów udostępnionych aby ludzi mogli trenować swój system.\n",
-    "\n",
-    "Źródła: [2] [1]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 53,
-   "metadata": {
-    "slideshow": {
-     "slide_type": "slide"
-    }
-   },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "system: Hello , welcome to the Cambridge restaurant system? You can ask for restaurants by area , price range or food type . How may I help you?   dialog-acts-[{'slots': [], 'act': 'welcomemsg'}]\n",
-      "user:   expensive restaurant in the south part of town      dialog-acts-[{'slots': [], 'act': 'welcomemsg'}]\n",
-      "system: What kind of food would you like?   dialog-acts-[{'slots': [['slot', 'food']], 'act': 'request'}]\n",
-      "user:   any      dialog-acts-[{'slots': [['slot', 'food']], 'act': 'request'}]\n",
-      "system: the good luck chinese food takeaway is a nice restaurant in the south of town serving chinese food   dialog-acts-[{'slots': [['name', 'the good luck chinese food takeaway']], 'act': 'offer'}, {'slots': [['food', 'chinese']], 'act': 'inform'}, {'slots': [['pricerange', 'expensive']], 'act': 'inform'}, {'slots': [['area', 'south']], 'act': 'inform'}]\n",
-      "user:   address      dialog-acts-[{'slots': [['name', 'the good luck chinese food takeaway']], 'act': 'offer'}, {'slots': [['food', 'chinese']], 'act': 'inform'}, {'slots': [['pricerange', 'expensive']], 'act': 'inform'}, {'slots': [['area', 'south']], 'act': 'inform'}]\n",
-      "system: Sure , the good luck chinese food takeaway is on 82 Cherry Hinton Road Cherry Hinton   dialog-acts-[{'slots': [['name', 'the good luck chinese food takeaway']], 'act': 'offer'}, {'slots': [['addr', '82 cherry hinton road cherry hinton']], 'act': 'inform'}]\n",
-      "user:   type of food      dialog-acts-[{'slots': [['name', 'the good luck chinese food takeaway']], 'act': 'offer'}, {'slots': [['addr', '82 cherry hinton road cherry hinton']], 'act': 'inform'}]\n",
-      "system: the good luck chinese food takeaway is a nice restaurant in the south of town in the expensive price range   dialog-acts-[{'slots': [['name', 'the good luck chinese food takeaway']], 'act': 'offer'}, {'slots': [['pricerange', 'expensive']], 'act': 'inform'}, {'slots': [['area', 'south']], 'act': 'inform'}]\n",
-      "user:   thank you good bye      dialog-acts-[{'slots': [['name', 'the good luck chinese food takeaway']], 'act': 'offer'}, {'slots': [['pricerange', 'expensive']], 'act': 'inform'}, {'slots': [['area', 'south']], 'act': 'inform'}]\n"
-     ]
-    }
-   ],
-   "source": [
-    "import json\n",
-    "f = open('log.json')\n",
-    "f2 = open('label.json')\n",
-    "data = json.load(f)\n",
-    "user = json.load(f2)\n",
-    "for i in range(len(data['turns'])):\n",
-    "    print(f'system: {data[\"turns\"][i][\"output\"][\"transcript\"]}   dialog-acts-{data[\"turns\"][i][\"output\"][\"dialog-acts\"]}')\n",
-    "    print(f'user:   {user[\"turns\"][i][\"transcription\"]}      dialog-acts-{data[\"turns\"][i][\"output\"][\"dialog-acts\"]}')"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "slideshow": {
-     "slide_type": "slide"
-    }
-   },
-   "source": [
-    "Co to jest\n",
-    "\n",
-    "DST jest to rozpozananie obecnego stanu dialogu na podstawię całego jego przebiegu.\n",
-    "\n",
-    "Po co to jest?\n",
-    "Dst umożliwa nam rozpozanie czego użytkownik jeszcze nam nie przekazał, jakie sloty nie zostały jeszcze uzupełnione. Umożlwia on szybkie zrozumienie gdy użytkoniwk odwoła się do poprzedniej swojej wypowiedzi.\n",
-    "\n",
-    "(Ludzie prowadząc dialog naturalnie zapamiętują na jaki temat toczy się dyskusja, System jednak potrzebuje mieć zaimplementowanę aby patrzeć na każdą nową ture dialogu jako nową element pewnej całość, aby mógł zrozumieć pewne odwołania lub skróty myślowe.)\n",
-    "\n",
-    "Problematyka\n",
-    "(Największym problem w monitorowaniu stanu dialogu jest nauczenie systemu do jakiego slota użytkownik się odwołuje, Oraz samo tworzenie slotów) \n",
-    "(do rozwiązań - Problemem jest że najczęstrze rozwiązania działają najlepiej na rozpoznawaniu co się dzieje w aktualnej turze, a nie koniecznie patrzą na ogół dialogu)\n",
-    "\n",
-    "\n",
-    "(jeżeli nie ma czegoś w bazie np. restauracji tajskiej, system musi i tak zrozumieć żę użytkownich chce się dowiedzieć o resturacji )\n",
-    "\n",
-    "Rozwiązania istniejące\n",
-    "Problem klasyfikacji \n",
-    "- Znalezienie istotnych cech z pośród własnorecznie utworzonych \n",
-    "- Słowniki semantyczne \n",
-    "Tworzenie bez znajomości kompleksowej ontologii domeny slotów na podstawie historii dialogów  \n",
-    "\n",
-    "\n",
-    "\n",
-    "RNN- sieć neuronowa która wpierdala w siebie ciągle wszystkie tury dialogu -co powoduję że dużo ma niepotrzebnych informacji\n",
-    "\n",
-    "\n",
-    "tera - turn-by-turn z pre-trained model -czyli biorą to co wywnioskował model poprzednio i łączy to z tym co zostało powiedziane w tej turze i zwraca nowy wniosek. "
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "slideshow": {
-     "slide_type": "slide"
-    }
-   },
-   "source": [
-    "**Bibliografia**\n",
-    "\n",
-    "**[1]**  DSTC https://github.com/matthen/dstc\n",
-    "\n",
-    "**[2]**  The Second Dialog State Tracking Challenge https://aclanthology.org/W14-4337.pdf\n",
-    "\n",
-    "**[3]**  Dialogue State Tracking with Incremental Reasoning https://direct.mit.edu/tacl/article/doi/10.1162/tacl_a_00384/101875/Dialogue-State-Tracking-with-Incremental-Reasoning"
-   ]
-  }
- ],
- "metadata": {
-  "celltoolbar": "Slideshow",
-  "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
-   "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": "3.9.7"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 2
-}
diff --git a/prezentacja.ipynb b/prezentacja.ipynb
index 6622c4f..7348099 100644
--- a/prezentacja.ipynb
+++ b/prezentacja.ipynb
@@ -16,7 +16,7 @@
    "cell_type": "markdown",
    "metadata": {
     "slideshow": {
-     "slide_type": "subslide"
+     "slide_type": "slide"
     }
    },
    "source": [
@@ -101,14 +101,17 @@
     }
    },
    "source": [
-    "**Tradycyjne rozwiązania w DST**\n",
+    "**Rozwiązania w DST**\n",
     "\n",
-    "Początkowo problemy monitorowania stanu dialogu rozwiązywano na kilka sposobów: \n",
-    "- Słowniki semantyczne, które umożliwiały\n",
-    "- Wyszukiwanie najlepszych cech z pośród zbiorów własnoręcznie wytworzonych\n",
-    "- Tworzenie bez znajomości kompleksowej ontologii domeny slotów na podstawie historii dialogów  \n",
+    "Slot-filling\n",
     "\n",
-    "Źródła: [3]"
+    "Hand-crafted\n",
+    "\n",
+    "Generative approach\n",
+    "\n",
+    "Discriminative approach\n",
+    "\n",
+    "Źródła: [4]"
    ]
   },
   {
@@ -119,12 +122,9 @@
     }
    },
    "source": [
-    "**Modele wykorzystywane w DST**\n",
+    "**Slot-filling**\n",
     "\n",
-    "W przeciągu ostatnich lat do opartych na uczeniu maszynowym rozwiązań nakierowanych na DST wykorzystywane były różne architektury modeli, przeważnie coraz efektywniejsze. Przyjrzymy się następującym konceptom:\n",
-    "- RNN\n",
-    "- ReDST\n",
-    "- BERT\n"
+    "Proste podejście przypisywania w każdej turze określonego aktu dialogwego wraz z odpowiadającym mu określonym znaczeniem. Działo to tylko dla prostych dialogów z wcześniej zdefiniowaną ontologią.  "
    ]
   },
   {
@@ -135,7 +135,9 @@
     }
    },
    "source": [
-    "Rekurencyjna sieć neuronowa (RNN) - łańcuch kopii tej samej sieci korzystających z parametrów poprzedniczek"
+    "**Hand-crafted**\n",
+    "\n",
+    "Na podstawie gotowych danych ontologicznych i własnoręcznie zrobionych reguł przypisywania wartości slotów "
    ]
   },
   {
@@ -146,19 +148,50 @@
     }
    },
    "source": [
-    "BERT"
+    "**Genartaive approach**\n",
+    "\n",
+    "Podejście polegające na tym że stany generują obserwacje, zakładają to że obserwacje na każdym etapie tury są niezależne od siebie. Umożliwia też tworzenie własnych ontlogi na podstawie historii dialogów oraz wyników jakie generuje NLU."
    ]
   },
   {
-   "cell_type": "code",
-   "execution_count": null,
+   "cell_type": "markdown",
    "metadata": {
     "slideshow": {
      "slide_type": "slide"
     }
    },
-   "outputs": [],
-   "source": []
+   "source": [
+    "**Discriminative approach**\n",
+    "\n",
+    "the dialogue state given\n",
+    "arbitrary and possibly correlated input features.\n",
+    "\n",
+    "Podejśćie polegające na tym że stan dialogu wynika z obserwacji, zakłada żę stany dialogu podane arbitralnie są możliwe połączone z wyjściem.\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "slideshow": {
+     "slide_type": "slide"
+    }
+   },
+   "source": [
+    "![s2s](S2S.png)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "slideshow": {
+     "slide_type": "slide"
+    }
+   },
+   "source": [
+    "![RNN](RNN.png)\n",
+    "\n",
+    "Recurrent neural network- Biorąc jako argument ostatni akt dialogu oraz to co napisał właśnie użytkownik, tworzy nowy act dialogu."
+   ]
   },
   {
    "cell_type": "markdown",
@@ -178,7 +211,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 53,
+   "execution_count": 1,
    "metadata": {
     "slideshow": {
      "slide_type": "slide"
@@ -214,41 +247,102 @@
    ]
   },
   {
-   "cell_type": "markdown",
+   "cell_type": "code",
+   "execution_count": 2,
    "metadata": {
     "slideshow": {
      "slide_type": "slide"
     }
    },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[{'asr-hyp': 'expensive restaurant in south part of town', 'score': -0.512136}, {'asr-hyp': 'a restaurant in south part of town', 'score': -2.429358}, {'asr-hyp': 'expensive expensive restaurant in south part of town', 'score': -2.429358}, {'asr-hyp': 'expensive restaurant restaurant in south part of town', 'score': -3.210609}, {'asr-hyp': 'expensive in south part of town', 'score': -3.211676}, {'asr-hyp': 'expensive restaurant south part of town', 'score': -3.347932}, {'asr-hyp': 'the expensive restaurant in south part of town', 'score': -4.268281}, {'asr-hyp': 'a expensive restaurant in south part of town', 'score': -4.34658}, {'asr-hyp': 'a restaurant restaurant in south part of town', 'score': -5.127831}, {'asr-hyp': 'expensive expensive restaurant restaurant in south part of town', 'score': -5.127831}]\n",
+      "{'asr-hyp': 'expensive restaurant in south part of town', 'score': -0.512136}\n",
+      "{'asr-hyp': 'a restaurant in south part of town', 'score': -2.429358}\n",
+      "{'asr-hyp': 'expensive expensive restaurant in south part of town', 'score': -2.429358}\n",
+      "{'asr-hyp': 'expensive restaurant restaurant in south part of town', 'score': -3.210609}\n",
+      "{'asr-hyp': 'expensive in south part of town', 'score': -3.211676}\n",
+      "{'asr-hyp': 'expensive restaurant south part of town', 'score': -3.347932}\n",
+      "{'asr-hyp': 'the expensive restaurant in south part of town', 'score': -4.268281}\n",
+      "{'asr-hyp': 'a expensive restaurant in south part of town', 'score': -4.34658}\n",
+      "{'asr-hyp': 'a restaurant restaurant in south part of town', 'score': -5.127831}\n",
+      "{'asr-hyp': 'expensive expensive restaurant restaurant in south part of town', 'score': -5.127831}\n",
+      "[{'asr-hyp': 'any', 'score': -0.02858}, {'asr-hyp': '', 'score': -4.777333}, {'asr-hyp': 'any me', 'score': -4.791492}, {'asr-hyp': 'any give', 'score': -4.885694}, {'asr-hyp': 'any the', 'score': -6.819719}, {'asr-hyp': 'in', 'score': -6.9403}, {'asr-hyp': 'any get', 'score': -7.371329}, {'asr-hyp': 'any a', 'score': -8.3173}, {'asr-hyp': 'and', 'score': -8.77938}, {'asr-hyp': 'any any', 'score': -8.946321}]\n",
+      "{'asr-hyp': 'any', 'score': -0.02858}\n",
+      "{'asr-hyp': '', 'score': -4.777333}\n",
+      "{'asr-hyp': 'any me', 'score': -4.791492}\n",
+      "{'asr-hyp': 'any give', 'score': -4.885694}\n",
+      "{'asr-hyp': 'any the', 'score': -6.819719}\n",
+      "{'asr-hyp': 'in', 'score': -6.9403}\n",
+      "{'asr-hyp': 'any get', 'score': -7.371329}\n",
+      "{'asr-hyp': 'any a', 'score': -8.3173}\n",
+      "{'asr-hyp': 'and', 'score': -8.77938}\n",
+      "{'asr-hyp': 'any any', 'score': -8.946321}\n",
+      "[{'asr-hyp': 'address', 'score': -0.455029}, {'asr-hyp': '', 'score': -2.265617}, {'asr-hyp': 'address address', 'score': -2.268945}, {'asr-hyp': 'the address', 'score': -2.274268}, {'asr-hyp': 'the', 'score': -4.084856}, {'asr-hyp': 'the address address', 'score': -4.088183}, {'asr-hyp': 'address their', 'score': -7.867221}, {'asr-hyp': 'address uh', 'score': -8.652037}, {'asr-hyp': 'and', 'score': -8.784164}, {'asr-hyp': 'i', 'score': -9.450777}]\n",
+      "{'asr-hyp': 'address', 'score': -0.455029}\n",
+      "{'asr-hyp': '', 'score': -2.265617}\n",
+      "{'asr-hyp': 'address address', 'score': -2.268945}\n",
+      "{'asr-hyp': 'the address', 'score': -2.274268}\n",
+      "{'asr-hyp': 'the', 'score': -4.084856}\n",
+      "{'asr-hyp': 'the address address', 'score': -4.088183}\n",
+      "{'asr-hyp': 'address their', 'score': -7.867221}\n",
+      "{'asr-hyp': 'address uh', 'score': -8.652037}\n",
+      "{'asr-hyp': 'and', 'score': -8.784164}\n",
+      "{'asr-hyp': 'i', 'score': -9.450777}\n",
+      "[{'asr-hyp': 'type of food', 'score': -0.016592}, {'asr-hyp': 'type type of food', 'score': -5.183987}, {'asr-hyp': 'the of food', 'score': -5.404256}, {'asr-hyp': 'of food', 'score': -6.505753}, {'asr-hyp': 'type food', 'score': -6.617192}, {'asr-hyp': 'type global of food', 'score': -7.340683}, {'asr-hyp': 'type budget of food', 'score': -7.360768}, {'asr-hyp': 'type the food', 'score': -7.5319}, {'asr-hyp': 'and of food', 'score': -7.617202}, {'asr-hyp': 'i of food', 'score': -7.727244}]\n",
+      "{'asr-hyp': 'type of food', 'score': -0.016592}\n",
+      "{'asr-hyp': 'type type of food', 'score': -5.183987}\n",
+      "{'asr-hyp': 'the of food', 'score': -5.404256}\n",
+      "{'asr-hyp': 'of food', 'score': -6.505753}\n",
+      "{'asr-hyp': 'type food', 'score': -6.617192}\n",
+      "{'asr-hyp': 'type global of food', 'score': -7.340683}\n",
+      "{'asr-hyp': 'type budget of food', 'score': -7.360768}\n",
+      "{'asr-hyp': 'type the food', 'score': -7.5319}\n",
+      "{'asr-hyp': 'and of food', 'score': -7.617202}\n",
+      "{'asr-hyp': 'i of food', 'score': -7.727244}\n",
+      "[{'asr-hyp': 'thank you goodbye', 'score': -0.30683}, {'asr-hyp': 'thank you goodbye bye', 'score': -2.107196}, {'asr-hyp': 'thank you good', 'score': -2.109648}, {'asr-hyp': 'thank you good bye', 'score': -3.910014}, {'asr-hyp': 'thank you you', 'score': -7.879262}, {'asr-hyp': 'thank goodbye', 'score': -7.957914}, {'asr-hyp': 'thank you goodbye bar', 'score': -8.884375}, {'asr-hyp': 'thank you you bye', 'score': -9.679628}, {'asr-hyp': 'thank goodbye bye', 'score': -9.758281}, {'asr-hyp': 'thank good', 'score': -9.760733}]\n",
+      "{'asr-hyp': 'thank you goodbye', 'score': -0.30683}\n",
+      "{'asr-hyp': 'thank you goodbye bye', 'score': -2.107196}\n",
+      "{'asr-hyp': 'thank you good', 'score': -2.109648}\n",
+      "{'asr-hyp': 'thank you good bye', 'score': -3.910014}\n",
+      "{'asr-hyp': 'thank you you', 'score': -7.879262}\n",
+      "{'asr-hyp': 'thank goodbye', 'score': -7.957914}\n",
+      "{'asr-hyp': 'thank you goodbye bar', 'score': -8.884375}\n",
+      "{'asr-hyp': 'thank you you bye', 'score': -9.679628}\n",
+      "{'asr-hyp': 'thank goodbye bye', 'score': -9.758281}\n",
+      "{'asr-hyp': 'thank good', 'score': -9.760733}\n"
+     ]
+    }
+   ],
    "source": [
-    "Co to jest\n",
-    "\n",
-    "DST jest to rozpozananie obecnego stanu dialogu na podstawię całego jego przebiegu.\n",
-    "\n",
-    "Po co to jest?\n",
-    "Dst umożliwa nam rozpozanie czego użytkownik jeszcze nam nie przekazał, jakie sloty nie zostały jeszcze uzupełnione. Umożlwia on szybkie zrozumienie gdy użytkoniwk odwoła się do poprzedniej swojej wypowiedzi.\n",
-    "\n",
-    "(Ludzie prowadząc dialog naturalnie zapamiętują na jaki temat toczy się dyskusja, System jednak potrzebuje mieć zaimplementowanę aby patrzeć na każdą nową ture dialogu jako nową element pewnej całość, aby mógł zrozumieć pewne odwołania lub skróty myślowe.)\n",
-    "\n",
-    "Problematyka\n",
-    "(Największym problem w monitorowaniu stanu dialogu jest nauczenie systemu do jakiego slota użytkownik się odwołuje, Oraz samo tworzenie slotów) \n",
-    "(do rozwiązań - Problemem jest że najczęstrze rozwiązania działają najlepiej na rozpoznawaniu co się dzieje w aktualnej turze, a nie koniecznie patrzą na ogół dialogu)\n",
-    "\n",
-    "\n",
-    "(jeżeli nie ma czegoś w bazie np. restauracji tajskiej, system musi i tak zrozumieć żę użytkownich chce się dowiedzieć o resturacji )\n",
-    "\n",
-    "Rozwiązania istniejące\n",
-    "Problem klasyfikacji \n",
-    "- Znalezienie istotnych cech z pośród własnorecznie utworzonych \n",
-    "- Słowniki semantyczne \n",
-    "Tworzenie bez znajomości kompleksowej ontologii domeny slotów na podstawie historii dialogów  \n",
-    "\n",
-    "\n",
-    "\n",
-    "RNN- sieć neuronowa która wpierdala w siebie ciągle wszystkie tury dialogu -co powoduję że dużo ma niepotrzebnych informacji\n",
-    "\n",
-    "\n",
-    "tera - turn-by-turn z pre-trained model -czyli biorą to co wywnioskował model poprzednio i łączy to z tym co zostało powiedziane w tej turze i zwraca nowy wniosek. "
+    "for i in range(len(data['turns'])):\n",
+    "    print(data[\"turns\"][i][\"input\"][\"live\"]['asr-hyps'])\n",
+    "    for l in data[\"turns\"][i][\"input\"][\"live\"]['asr-hyps']:\n",
+    "        print(l)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "slideshow": {
+     "slide_type": "slide"
+    }
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "{'output': {'transcript': 'Hello , welcome to the Cambridge restaurant system? You can ask for restaurants by area , price range or food type . How may I help you?', 'end-time': 9.93, 'start-time': 0.001283, 'dialog-acts': [{'slots': [], 'act': 'welcomemsg'}], 'aborted': False}, 'turn-index': 0, 'input': {'live': {'asr-hyps': [{'asr-hyp': 'expensive restaurant in south part of town', 'score': -0.512136}, {'asr-hyp': 'a restaurant in south part of town', 'score': -2.429358}, {'asr-hyp': 'expensive expensive restaurant in south part of town', 'score': -2.429358}, {'asr-hyp': 'expensive restaurant restaurant in south part of town', 'score': -3.210609}, {'asr-hyp': 'expensive in south part of town', 'score': -3.211676}, {'asr-hyp': 'expensive restaurant south part of town', 'score': -3.347932}, {'asr-hyp': 'the expensive restaurant in south part of town', 'score': -4.268281}, {'asr-hyp': 'a expensive restaurant in south part of town', 'score': -4.34658}, {'asr-hyp': 'a restaurant restaurant in south part of town', 'score': -5.127831}, {'asr-hyp': 'expensive expensive restaurant restaurant in south part of town', 'score': -5.127831}], 'slu-hyps': [{'slu-hyp': [{'slots': [['pricerange', 'expensive']], 'act': 'inform'}, {'slots': [['area', 'south']], 'act': 'inform'}], 'score': 0.9282257165548344}, {'slu-hyp': [{'slots': [['pricerange', 'expensive']], 'act': 'confirm'}, {'slots': [['area', 'south']], 'act': 'confirm'}], 'score': 0.02923086440308149}, {'slu-hyp': [{'slots': [['pricerange', 'expensive']], 'act': 'inform'}], 'score': 0.02236294307342127}, {'slu-hyp': [{'slots': [['area', 'south']], 'act': 'inform'}], 'score': 0.020180475968662844}]}, 'end-time': 12.19, 'start-time': 9.93, 'batch': {'cnet': [{'start': 0.0, 'end': 0.525, 'arcs': [{'score': -0.0001, 'word': '<s>'}, {'score': -5000000000.0, 'word': '!null'}]}, {'start': 0.525, 'end': 0.7875, 'arcs': [{'score': -19.41379, 'word': 'i'}, {'score': -16.88836, 'word': 'a'}, {'score': -11.6955, 'word': 'ah'}, {'score': -13.1059, 'word': 'uh'}, {'score': -0.0001, 'word': '!null'}]}, {'start': 0.7875, 'end': 1.05, 'arcs': [{'score': -11.16217, 'word': 'i'}, {'score': -9.916637, 'word': 'uh'}, {'score': -13.39644, 'word': 'ok'}, {'score': -0.0001, 'word': '!null'}]}, {'start': 1.05, 'end': 1.12375, 'arcs': [{'score': -0.115834, 'word': 'expensive'}, {'score': -9.822551, 'word': 'ok'}, {'score': -5.432811, 'word': 'the'}, {'score': -15.02797, 'word': 'extensive'}, {'score': -18.94896, 'word': 'are'}, {'score': -2.254255, 'word': 'a'}, {'score': -5000000000.0, 'word': '!null'}]}, {'start': 1.12375, 'end': 1.1975, 'arcs': [{'score': -17.54265, 'word': 'ok'}, {'score': -0.0001, 'word': '!null'}]}, {'start': 1.1975, 'end': 1.345, 'arcs': [{'score': -20.00645, 'word': 'expensive'}, {'score': -23.83931, 'word': 'extensive'}, {'score': -0.0001, 'word': '!null'}]}, {'start': 1.345, 'end': 1.4925, 'arcs': [{'score': -2.212958, 'word': 'expensive'}, {'score': -21.30504, 'word': 'extensive'}, {'score': -0.1158337, 'word': '!null'}]}, {'start': 1.4925, 'end': 1.64, 'arcs': [{'score': -19.56206, 'word': 'a'}, {'score': -15.88049, 'word': 'or'}, {'score': -19.37142, 'word': 'uh'}, {'score': -0.0001, 'word': '!null'}]}, {'start': 1.64, 'end': 1.7875, 'arcs': [{'score': -0.0001, 'word': 'restaurant'}, {'score': -14.9832, 'word': 'the'}, {'score': -11.33279, 'word': 'a'}, {'score': -9.494602, 'word': 'uh'}, {'score': -11.74803, 'word': 'or'}, {'score': -17.88491, 'word': '!null'}]}, {'start': 1.7875, 'end': 1.935, 'arcs': [{'score': -13.69902, 'word': 'restaurant'}, {'score': -0.0001, 'word': '!null'}]}, {'start': 1.935, 'end': 2.23, 'arcs': [{'score': -9.268782, 'word': 'restaurant'}, {'score': -2.381682, 'word': 'in'}, {'score': -0.09705002, 'word': '!null'}]}, {'start': 2.23, 'end': 2.27, 'arcs': [{'score': -0.1433645, 'word': 'in'}, {'score': -12.21065, 'word': 'or'}, {'score': -13.78668, 'word': 'i'}, {'score': -12.6393, 'word': 'are'}, {'score': -13.47915, 'word': 'a'}, {'score': -11.99133, 'word': \"that's\"}, {'score': -7.282886, 'word': 'that'}, {'score': -2.348785, 'word': 'the'}, {'score': -13.8318, 'word': 'this'}, {'score': -3.286841, 'word': '!null'}]}, {'start': 2.27, 'end': 2.31, 'arcs': [{'score': -46.60755, 'word': 'it'}, {'score': -12.64047, 'word': 'which'}, {'score': -22.6327, 'word': 'this'}, {'score': -12.27715, 'word': 'the'}, {'score': -34.75492, 'word': 'as'}, {'score': -6.923453, 'word': 'is'}, {'score': -7.139083, 'word': 'of'}, {'score': -10.67487, 'word': 'a'}, {'score': -7.923023, 'word': 'tea'}, {'score': -0.002173584, 'word': '!null'}]}, {'start': 2.31, 'end': 2.48, 'arcs': [{'score': -0.005246917, 'word': 'south'}, {'score': -37.23841, 'word': 'well'}, {'score': -10.74857, 'word': 'ok'}, {'score': -46.34189, 'word': 'gastro'}, {'score': -46.3515, 'word': 'aswell'}, {'score': -9.687854, 'word': 'child'}, {'score': -5.276869, 'word': 'shop'}, {'score': -41.20222, 'word': 'style'}, {'score': -10.09526, 'word': 'stop'}, {'score': -44.53774, 'word': 'sell'}, {'score': -5000000000.0, 'word': '!null'}]}, {'start': 2.48, 'end': 2.65, 'arcs': [{'score': -22.39381, 'word': 'is'}, {'score': -36.75235, 'word': 'of'}, {'score': -32.42216, 'word': 'how'}, {'score': -38.02597, 'word': 'in'}, {'score': -0.0001, 'word': '!null'}]}, {'start': 2.65, 'end': 2.74, 'arcs': [{'score': -0.0008488684, 'word': 'part'}, {'score': -25.69577, 'word': 'partner'}, {'score': -13.48137, 'word': 'what'}, {'score': -25.05888, 'word': 'post'}, {'score': -27.47913, 'word': 'are'}, {'score': -11.4708, 'word': 'close'}, {'score': -23.77218, 'word': 'park'}, {'score': -14.65481, 'word': 'code'}, {'score': -15.24271, 'word': 'cote'}, {'score': -15.58045, 'word': 'or'}, {'score': -12.75343, 'word': 'i'}, {'score': -15.39197, 'word': 'in'}, {'score': -18.09848, 'word': 'it'}, {'score': -8.927246, 'word': 'a'}, {'score': -20.65172, 'word': 'for'}, {'score': -31.9557, 'word': 'the'}, {'score': -7.264429, 'word': '!null'}]}, {'start': 2.74, 'end': 2.83, 'arcs': [{'score': -18.37638, 'word': 'what'}, {'score': -18.29665, 'word': 'for'}, {'score': -20.94735, 'word': 'or'}, {'score': -11.55013, 'word': 'on'}, {'score': -0.0001, 'word': '!null'}]}, {'start': 2.83, 'end': 2.94, 'arcs': [{'score': -0.000271893, 'word': 'of'}, {'score': -30.09363, 'word': 'post'}, {'score': -24.60016, 'word': \"don't\"}, {'score': -11.64039, 'word': 'the'}, {'score': -24.75157, 'word': 'okay'}, {'score': -8.243156, 'word': '!null'}]}, {'start': 2.94, 'end': 3.1, 'arcs': [{'score': -0.0001754378, 'word': 'town'}, {'score': -14.22256, 'word': 'cocum'}, {'score': -8.735506, 'word': 'hotel'}, {'score': -23.20746, 'word': 'care'}, {'score': -11.17799, 'word': 'tell'}, {'score': -24.11379, 'word': 'okay'}, {'score': -19.73659, 'word': 'down'}, {'score': -21.45679, 'word': 'it'}, {'score': -22.00763, 'word': '!null'}]}, {'start': 3.1, 'end': 3.18, 'arcs': [{'score': -20.51506, 'word': 'on'}, {'score': -21.36264, 'word': 'down'}, {'score': -0.0001, 'word': '!null'}]}, {'start': 3.18, 'end': 3.26, 'arcs': [{'score': -11.9611, 'word': 'a'}, {'score': -0.0001, 'word': '!null'}]}, {'start': 3.26, 'end': 4.14, 'arcs': [{'score': -0.0001, 'word': '</s>'}, {'score': -23.32922, 'word': '!null'}]}], 'asr-hyps': [{'asr-hyp': 'expensive restaurant in south part of town', 'score': -0.480987}, {'asr-hyp': 'expensive expensive restaurant in south part of town', 'score': -2.578111}, {'asr-hyp': 'a restaurant in south part of town', 'score': -2.619409}, {'asr-hyp': 'expensive restaurant the south part of town', 'score': -2.686408}, {'asr-hyp': 'expensive restaurant in in south part of town', 'score': -2.765619}, {'asr-hyp': 'expensive restaurant south part of town', 'score': -3.624464}, {'asr-hyp': 'a expensive restaurant in south part of town', 'score': -4.716533}, {'asr-hyp': 'expensive expensive restaurant the south part of town', 'score': -4.783532}, {'asr-hyp': 'a restaurant the south part of town', 'score': -4.82483}, {'asr-hyp': 'expensive expensive restaurant in in south part of town', 'score': -4.862743}]}, 'audio-file': 'pt344x_0000993_0001219.wav'}}\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(data['turns'][0])"
    ]
   },
   {
@@ -265,7 +359,11 @@
     "\n",
     "**[2]**  The Second Dialog State Tracking Challenge https://aclanthology.org/W14-4337.pdf\n",
     "\n",
-    "**[3]**  Dialogue State Tracking with Incremental Reasoning https://direct.mit.edu/tacl/article/doi/10.1162/tacl_a_00384/101875/Dialogue-State-Tracking-with-Incremental-Reasoning"
+    "**[3]**  Dialogue State Tracking with Incremental Reasoning https://direct.mit.edu/tacl/article/doi/10.1162/tacl_a_00384/101875/Dialogue-State-Tracking-with-Incremental-Reasoning\n",
+    "\n",
+    "**[4]**  https://arxiv.org/pdf/2101.08333.pdf\n",
+    "\n",
+    "**[5]** https://www.cs.hhu.de/fileadmin/redaktion/Fakultaeten/Mathematisch-Naturwissenschaftliche_Fakultaet/Informatik/Dialog_Systems_and_Machine_Learning/L6.pdf"
    ]
   }
  ],
@@ -286,7 +384,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.9.7"
+   "version": "3.8.10"
   }
  },
  "nbformat": 4,
diff --git a/prezentacja.slides.html b/prezentacja.slides.html
index 32b6568..6650dcf 100644
--- a/prezentacja.slides.html
+++ b/prezentacja.slides.html
@@ -14103,7 +14103,7 @@ div.jp-OutputPrompt {
 Monitorowanie stanu dialogu</p>
 
 </div>
-</div></section><section>
+</div></section></section><section><section>
 <div class="jp-Cell-inputWrapper"><div class="jp-InputPrompt jp-InputArea-prompt">
 </div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput " data-mime-type="text/markdown">
 <p><strong>Czym jest DST?</strong></p>
@@ -14149,54 +14149,57 @@ Monitorowanie stanu dialogu</p>
 </div></section></section><section><section>
 <div class="jp-Cell-inputWrapper"><div class="jp-InputPrompt jp-InputArea-prompt">
 </div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput " data-mime-type="text/markdown">
-<p><strong>Tradycyjne rozwiązania w DST</strong></p>
-<p>Początkowo problemy monitorowania stanu dialogu rozwiązywano na kilka sposobów:</p>
-<ul>
-<li>Słowniki semantyczne, które umożliwiały</li>
-<li>Wyszukiwanie najlepszych cech z pośród zbiorów własnoręcznie wytworzonych</li>
-<li>Tworzenie bez znajomości kompleksowej ontologii domeny slotów na podstawie historii dialogów  </li>
-</ul>
-<p>Źródła: [3]</p>
+<p><strong>Rozwiązania w DST</strong></p>
+<p>Slot-filling</p>
+<p>Hand-crafted</p>
+<p>Generative approach</p>
+<p>Discriminative approach</p>
+<p>Źródła: [4]</p>
 
 </div>
 </div></section></section><section><section>
 <div class="jp-Cell-inputWrapper"><div class="jp-InputPrompt jp-InputArea-prompt">
 </div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput " data-mime-type="text/markdown">
-<p><strong>Modele wykorzystywane w DST</strong></p>
-<p>W przeciągu ostatnich lat do opartych na uczeniu maszynowym rozwiązań nakierowanych na DST wykorzystywane były różne architektury modeli, przeważnie coraz efektywniejsze. Przyjrzymy się następującym konceptom:</p>
-<ul>
-<li>RNN</li>
-<li>ReDST</li>
-<li>BERT</li>
-</ul>
+<p><strong>Slot-filling</strong></p>
+<p>Proste podejście przypisywania w każdej turze określonego aktu dialogwego wraz z odpowiadającym mu określonym znaczeniem. Działo to tylko dla prostych dialogów z wcześniej zdefiniowaną ontologią.</p>
 
 </div>
 </div></section></section><section><section>
 <div class="jp-Cell-inputWrapper"><div class="jp-InputPrompt jp-InputArea-prompt">
 </div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput " data-mime-type="text/markdown">
-<p>Rekurencyjna sieć neuronowa (RNN) - łańcuch kopii tej samej sieci korzystających z parametrów poprzedniczek</p>
+<p><strong>Hand-crafted</strong></p>
+<p>Na podstawie gotowych danych ontologicznych i własnoręcznie zrobionych reguł przypisywania wartości slotów</p>
 
 </div>
 </div></section></section><section><section>
 <div class="jp-Cell-inputWrapper"><div class="jp-InputPrompt jp-InputArea-prompt">
 </div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput " data-mime-type="text/markdown">
-<p>BERT</p>
+<p><strong>Genartaive approach</strong></p>
+<p>Podejście polegające na tym że stany generują obserwacje, zakładają to że obserwacje na każdym etapie tury są niezależne od siebie. Umożliwia też tworzenie własnych ontlogi na podstawie historii dialogów oraz wyników jakie generuje NLU.</p>
 
 </div>
-</div></section></section><section><section><div class="jp-Cell jp-CodeCell jp-Notebook-cell jp-mod-noOutputs  ">
-<div class="jp-Cell-inputWrapper">
-<div class="jp-InputArea jp-Cell-inputArea">
-<div class="jp-InputPrompt jp-InputArea-prompt">In&nbsp;[&nbsp;]:</div>
-<div class="jp-CodeMirrorEditor jp-Editor jp-InputArea-editor" data-type="inline">
-     <div class="CodeMirror cm-s-jupyter">
-<div class=" highlight hl-ipython3"><pre><span></span> 
-</pre></div>
+</div></section></section><section><section>
+<div class="jp-Cell-inputWrapper"><div class="jp-InputPrompt jp-InputArea-prompt">
+</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput " data-mime-type="text/markdown">
+<p><strong>Discriminative approach</strong></p>
+<p>the dialogue state given
+arbitrary and possibly correlated input features.</p>
+<p>Podejśćie polegające na tym że stan dialogu wynika z obserwacji, zakłada żę stany dialogu podane arbitralnie są możliwe połączone z wyjściem.</p>
 
-     </div>
-</div>
-</div>
 </div>
+</div></section></section><section><section>
+<div class="jp-Cell-inputWrapper"><div class="jp-InputPrompt jp-InputArea-prompt">
+</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput " data-mime-type="text/markdown">
+<p><img src="S2S.png" alt="s2s"></p>
 
+</div>
+</div></section></section><section><section>
+<div class="jp-Cell-inputWrapper"><div class="jp-InputPrompt jp-InputArea-prompt">
+</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput " data-mime-type="text/markdown">
+<p><img src="RNN.png" alt="RNN"></p>
+<p>Recurrent neural network- Biorąc jako argument ostatni akt dialogu oraz to co napisał właśnie użytkownik, tworzy nowy act dialogu.</p>
+
+</div>
 </div></section></section><section><section>
 <div class="jp-Cell-inputWrapper"><div class="jp-InputPrompt jp-InputArea-prompt">
 </div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput " data-mime-type="text/markdown">
@@ -14209,7 +14212,7 @@ DSTC 2 zawierał ponad 15 tysięcy dobrze opisanych dialogów udostępnionych ab
 </div></section></section><section><section><div class="jp-Cell jp-CodeCell jp-Notebook-cell   ">
 <div class="jp-Cell-inputWrapper">
 <div class="jp-InputArea jp-Cell-inputArea">
-<div class="jp-InputPrompt jp-InputArea-prompt">In&nbsp;[53]:</div>
+<div class="jp-InputPrompt jp-InputArea-prompt">In&nbsp;[1]:</div>
 <div class="jp-CodeMirrorEditor jp-Editor jp-InputArea-editor" data-type="inline">
      <div class="CodeMirror cm-s-jupyter">
 <div class=" highlight hl-ipython3"><pre><span></span><span class="kn">import</span> <span class="nn">json</span>
@@ -14257,29 +14260,133 @@ user:   thank you good bye      dialog-acts-[{&#39;slots&#39;: [[&#39;name&#39;,
 
 </div>
 
-</div></section></section><section><section>
-<div class="jp-Cell-inputWrapper"><div class="jp-InputPrompt jp-InputArea-prompt">
-</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput " data-mime-type="text/markdown">
-<p>Co to jest</p>
-<p>DST jest to rozpozananie obecnego stanu dialogu na podstawię całego jego przebiegu.</p>
-<p>Po co to jest?
-Dst umożliwa nam rozpozanie czego użytkownik jeszcze nam nie przekazał, jakie sloty nie zostały jeszcze uzupełnione. Umożlwia on szybkie zrozumienie gdy użytkoniwk odwoła się do poprzedniej swojej wypowiedzi.</p>
-<p>(Ludzie prowadząc dialog naturalnie zapamiętują na jaki temat toczy się dyskusja, System jednak potrzebuje mieć zaimplementowanę aby patrzeć na każdą nową ture dialogu jako nową element pewnej całość, aby mógł zrozumieć pewne odwołania lub skróty myślowe.)</p>
-<p>Problematyka
-(Największym problem w monitorowaniu stanu dialogu jest nauczenie systemu do jakiego slota użytkownik się odwołuje, Oraz samo tworzenie slotów) 
-(do rozwiązań - Problemem jest że najczęstrze rozwiązania działają najlepiej na rozpoznawaniu co się dzieje w aktualnej turze, a nie koniecznie patrzą na ogół dialogu)</p>
-<p>(jeżeli nie ma czegoś w bazie np. restauracji tajskiej, system musi i tak zrozumieć żę użytkownich chce się dowiedzieć o resturacji )</p>
-<p>Rozwiązania istniejące
-Problem klasyfikacji</p>
-<ul>
-<li>Znalezienie istotnych cech z pośród własnorecznie utworzonych </li>
-<li>Słowniki semantyczne 
-Tworzenie bez znajomości kompleksowej ontologii domeny slotów na podstawie historii dialogów  </li>
-</ul>
-<p>RNN- sieć neuronowa która wpierdala w siebie ciągle wszystkie tury dialogu -co powoduję że dużo ma niepotrzebnych informacji</p>
-<p>tera - turn-by-turn z pre-trained model -czyli biorą to co wywnioskował model poprzednio i łączy to z tym co zostało powiedziane w tej turze i zwraca nowy wniosek.</p>
+</div></section></section><section><section><div class="jp-Cell jp-CodeCell jp-Notebook-cell   ">
+<div class="jp-Cell-inputWrapper">
+<div class="jp-InputArea jp-Cell-inputArea">
+<div class="jp-InputPrompt jp-InputArea-prompt">In&nbsp;[2]:</div>
+<div class="jp-CodeMirrorEditor jp-Editor jp-InputArea-editor" data-type="inline">
+     <div class="CodeMirror cm-s-jupyter">
+<div class=" highlight hl-ipython3"><pre><span></span><span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">data</span><span class="p">[</span><span class="s1">&#39;turns&#39;</span><span class="p">])):</span>
+    <span class="nb">print</span><span class="p">(</span><span class="n">data</span><span class="p">[</span><span class="s2">&quot;turns&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">][</span><span class="s2">&quot;input&quot;</span><span class="p">][</span><span class="s2">&quot;live&quot;</span><span class="p">][</span><span class="s1">&#39;asr-hyps&#39;</span><span class="p">])</span>
+    <span class="k">for</span> <span class="n">l</span> <span class="ow">in</span> <span class="n">data</span><span class="p">[</span><span class="s2">&quot;turns&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">][</span><span class="s2">&quot;input&quot;</span><span class="p">][</span><span class="s2">&quot;live&quot;</span><span class="p">][</span><span class="s1">&#39;asr-hyps&#39;</span><span class="p">]:</span>
+        <span class="nb">print</span><span class="p">(</span><span class="n">l</span><span class="p">)</span>
+</pre></div>
+
+     </div>
+</div>
+</div>
+</div>
+
+<div class="jp-Cell-outputWrapper">
+
+
+<div class="jp-OutputArea jp-Cell-outputArea">
+
+<div class="jp-OutputArea-child">
+
+    
+    <div class="jp-OutputPrompt jp-OutputArea-prompt"></div>
+
+
+<div class="jp-RenderedText jp-OutputArea-output" data-mime-type="text/plain">
+<pre>[{&#39;asr-hyp&#39;: &#39;expensive restaurant in south part of town&#39;, &#39;score&#39;: -0.512136}, {&#39;asr-hyp&#39;: &#39;a restaurant in south part of town&#39;, &#39;score&#39;: -2.429358}, {&#39;asr-hyp&#39;: &#39;expensive expensive restaurant in south part of town&#39;, &#39;score&#39;: -2.429358}, {&#39;asr-hyp&#39;: &#39;expensive restaurant restaurant in south part of town&#39;, &#39;score&#39;: -3.210609}, {&#39;asr-hyp&#39;: &#39;expensive in south part of town&#39;, &#39;score&#39;: -3.211676}, {&#39;asr-hyp&#39;: &#39;expensive restaurant south part of town&#39;, &#39;score&#39;: -3.347932}, {&#39;asr-hyp&#39;: &#39;the expensive restaurant in south part of town&#39;, &#39;score&#39;: -4.268281}, {&#39;asr-hyp&#39;: &#39;a expensive restaurant in south part of town&#39;, &#39;score&#39;: -4.34658}, {&#39;asr-hyp&#39;: &#39;a restaurant restaurant in south part of town&#39;, &#39;score&#39;: -5.127831}, {&#39;asr-hyp&#39;: &#39;expensive expensive restaurant restaurant in south part of town&#39;, &#39;score&#39;: -5.127831}]
+{&#39;asr-hyp&#39;: &#39;expensive restaurant in south part of town&#39;, &#39;score&#39;: -0.512136}
+{&#39;asr-hyp&#39;: &#39;a restaurant in south part of town&#39;, &#39;score&#39;: -2.429358}
+{&#39;asr-hyp&#39;: &#39;expensive expensive restaurant in south part of town&#39;, &#39;score&#39;: -2.429358}
+{&#39;asr-hyp&#39;: &#39;expensive restaurant restaurant in south part of town&#39;, &#39;score&#39;: -3.210609}
+{&#39;asr-hyp&#39;: &#39;expensive in south part of town&#39;, &#39;score&#39;: -3.211676}
+{&#39;asr-hyp&#39;: &#39;expensive restaurant south part of town&#39;, &#39;score&#39;: -3.347932}
+{&#39;asr-hyp&#39;: &#39;the expensive restaurant in south part of town&#39;, &#39;score&#39;: -4.268281}
+{&#39;asr-hyp&#39;: &#39;a expensive restaurant in south part of town&#39;, &#39;score&#39;: -4.34658}
+{&#39;asr-hyp&#39;: &#39;a restaurant restaurant in south part of town&#39;, &#39;score&#39;: -5.127831}
+{&#39;asr-hyp&#39;: &#39;expensive expensive restaurant restaurant in south part of town&#39;, &#39;score&#39;: -5.127831}
+[{&#39;asr-hyp&#39;: &#39;any&#39;, &#39;score&#39;: -0.02858}, {&#39;asr-hyp&#39;: &#39;&#39;, &#39;score&#39;: -4.777333}, {&#39;asr-hyp&#39;: &#39;any me&#39;, &#39;score&#39;: -4.791492}, {&#39;asr-hyp&#39;: &#39;any give&#39;, &#39;score&#39;: -4.885694}, {&#39;asr-hyp&#39;: &#39;any the&#39;, &#39;score&#39;: -6.819719}, {&#39;asr-hyp&#39;: &#39;in&#39;, &#39;score&#39;: -6.9403}, {&#39;asr-hyp&#39;: &#39;any get&#39;, &#39;score&#39;: -7.371329}, {&#39;asr-hyp&#39;: &#39;any a&#39;, &#39;score&#39;: -8.3173}, {&#39;asr-hyp&#39;: &#39;and&#39;, &#39;score&#39;: -8.77938}, {&#39;asr-hyp&#39;: &#39;any any&#39;, &#39;score&#39;: -8.946321}]
+{&#39;asr-hyp&#39;: &#39;any&#39;, &#39;score&#39;: -0.02858}
+{&#39;asr-hyp&#39;: &#39;&#39;, &#39;score&#39;: -4.777333}
+{&#39;asr-hyp&#39;: &#39;any me&#39;, &#39;score&#39;: -4.791492}
+{&#39;asr-hyp&#39;: &#39;any give&#39;, &#39;score&#39;: -4.885694}
+{&#39;asr-hyp&#39;: &#39;any the&#39;, &#39;score&#39;: -6.819719}
+{&#39;asr-hyp&#39;: &#39;in&#39;, &#39;score&#39;: -6.9403}
+{&#39;asr-hyp&#39;: &#39;any get&#39;, &#39;score&#39;: -7.371329}
+{&#39;asr-hyp&#39;: &#39;any a&#39;, &#39;score&#39;: -8.3173}
+{&#39;asr-hyp&#39;: &#39;and&#39;, &#39;score&#39;: -8.77938}
+{&#39;asr-hyp&#39;: &#39;any any&#39;, &#39;score&#39;: -8.946321}
+[{&#39;asr-hyp&#39;: &#39;address&#39;, &#39;score&#39;: -0.455029}, {&#39;asr-hyp&#39;: &#39;&#39;, &#39;score&#39;: -2.265617}, {&#39;asr-hyp&#39;: &#39;address address&#39;, &#39;score&#39;: -2.268945}, {&#39;asr-hyp&#39;: &#39;the address&#39;, &#39;score&#39;: -2.274268}, {&#39;asr-hyp&#39;: &#39;the&#39;, &#39;score&#39;: -4.084856}, {&#39;asr-hyp&#39;: &#39;the address address&#39;, &#39;score&#39;: -4.088183}, {&#39;asr-hyp&#39;: &#39;address their&#39;, &#39;score&#39;: -7.867221}, {&#39;asr-hyp&#39;: &#39;address uh&#39;, &#39;score&#39;: -8.652037}, {&#39;asr-hyp&#39;: &#39;and&#39;, &#39;score&#39;: -8.784164}, {&#39;asr-hyp&#39;: &#39;i&#39;, &#39;score&#39;: -9.450777}]
+{&#39;asr-hyp&#39;: &#39;address&#39;, &#39;score&#39;: -0.455029}
+{&#39;asr-hyp&#39;: &#39;&#39;, &#39;score&#39;: -2.265617}
+{&#39;asr-hyp&#39;: &#39;address address&#39;, &#39;score&#39;: -2.268945}
+{&#39;asr-hyp&#39;: &#39;the address&#39;, &#39;score&#39;: -2.274268}
+{&#39;asr-hyp&#39;: &#39;the&#39;, &#39;score&#39;: -4.084856}
+{&#39;asr-hyp&#39;: &#39;the address address&#39;, &#39;score&#39;: -4.088183}
+{&#39;asr-hyp&#39;: &#39;address their&#39;, &#39;score&#39;: -7.867221}
+{&#39;asr-hyp&#39;: &#39;address uh&#39;, &#39;score&#39;: -8.652037}
+{&#39;asr-hyp&#39;: &#39;and&#39;, &#39;score&#39;: -8.784164}
+{&#39;asr-hyp&#39;: &#39;i&#39;, &#39;score&#39;: -9.450777}
+[{&#39;asr-hyp&#39;: &#39;type of food&#39;, &#39;score&#39;: -0.016592}, {&#39;asr-hyp&#39;: &#39;type type of food&#39;, &#39;score&#39;: -5.183987}, {&#39;asr-hyp&#39;: &#39;the of food&#39;, &#39;score&#39;: -5.404256}, {&#39;asr-hyp&#39;: &#39;of food&#39;, &#39;score&#39;: -6.505753}, {&#39;asr-hyp&#39;: &#39;type food&#39;, &#39;score&#39;: -6.617192}, {&#39;asr-hyp&#39;: &#39;type global of food&#39;, &#39;score&#39;: -7.340683}, {&#39;asr-hyp&#39;: &#39;type budget of food&#39;, &#39;score&#39;: -7.360768}, {&#39;asr-hyp&#39;: &#39;type the food&#39;, &#39;score&#39;: -7.5319}, {&#39;asr-hyp&#39;: &#39;and of food&#39;, &#39;score&#39;: -7.617202}, {&#39;asr-hyp&#39;: &#39;i of food&#39;, &#39;score&#39;: -7.727244}]
+{&#39;asr-hyp&#39;: &#39;type of food&#39;, &#39;score&#39;: -0.016592}
+{&#39;asr-hyp&#39;: &#39;type type of food&#39;, &#39;score&#39;: -5.183987}
+{&#39;asr-hyp&#39;: &#39;the of food&#39;, &#39;score&#39;: -5.404256}
+{&#39;asr-hyp&#39;: &#39;of food&#39;, &#39;score&#39;: -6.505753}
+{&#39;asr-hyp&#39;: &#39;type food&#39;, &#39;score&#39;: -6.617192}
+{&#39;asr-hyp&#39;: &#39;type global of food&#39;, &#39;score&#39;: -7.340683}
+{&#39;asr-hyp&#39;: &#39;type budget of food&#39;, &#39;score&#39;: -7.360768}
+{&#39;asr-hyp&#39;: &#39;type the food&#39;, &#39;score&#39;: -7.5319}
+{&#39;asr-hyp&#39;: &#39;and of food&#39;, &#39;score&#39;: -7.617202}
+{&#39;asr-hyp&#39;: &#39;i of food&#39;, &#39;score&#39;: -7.727244}
+[{&#39;asr-hyp&#39;: &#39;thank you goodbye&#39;, &#39;score&#39;: -0.30683}, {&#39;asr-hyp&#39;: &#39;thank you goodbye bye&#39;, &#39;score&#39;: -2.107196}, {&#39;asr-hyp&#39;: &#39;thank you good&#39;, &#39;score&#39;: -2.109648}, {&#39;asr-hyp&#39;: &#39;thank you good bye&#39;, &#39;score&#39;: -3.910014}, {&#39;asr-hyp&#39;: &#39;thank you you&#39;, &#39;score&#39;: -7.879262}, {&#39;asr-hyp&#39;: &#39;thank goodbye&#39;, &#39;score&#39;: -7.957914}, {&#39;asr-hyp&#39;: &#39;thank you goodbye bar&#39;, &#39;score&#39;: -8.884375}, {&#39;asr-hyp&#39;: &#39;thank you you bye&#39;, &#39;score&#39;: -9.679628}, {&#39;asr-hyp&#39;: &#39;thank goodbye bye&#39;, &#39;score&#39;: -9.758281}, {&#39;asr-hyp&#39;: &#39;thank good&#39;, &#39;score&#39;: -9.760733}]
+{&#39;asr-hyp&#39;: &#39;thank you goodbye&#39;, &#39;score&#39;: -0.30683}
+{&#39;asr-hyp&#39;: &#39;thank you goodbye bye&#39;, &#39;score&#39;: -2.107196}
+{&#39;asr-hyp&#39;: &#39;thank you good&#39;, &#39;score&#39;: -2.109648}
+{&#39;asr-hyp&#39;: &#39;thank you good bye&#39;, &#39;score&#39;: -3.910014}
+{&#39;asr-hyp&#39;: &#39;thank you you&#39;, &#39;score&#39;: -7.879262}
+{&#39;asr-hyp&#39;: &#39;thank goodbye&#39;, &#39;score&#39;: -7.957914}
+{&#39;asr-hyp&#39;: &#39;thank you goodbye bar&#39;, &#39;score&#39;: -8.884375}
+{&#39;asr-hyp&#39;: &#39;thank you you bye&#39;, &#39;score&#39;: -9.679628}
+{&#39;asr-hyp&#39;: &#39;thank goodbye bye&#39;, &#39;score&#39;: -9.758281}
+{&#39;asr-hyp&#39;: &#39;thank good&#39;, &#39;score&#39;: -9.760733}
+</pre>
+</div>
+</div>
 
 </div>
+
+</div>
+
+</div></section></section><section><section><div class="jp-Cell jp-CodeCell jp-Notebook-cell   ">
+<div class="jp-Cell-inputWrapper">
+<div class="jp-InputArea jp-Cell-inputArea">
+<div class="jp-InputPrompt jp-InputArea-prompt">In&nbsp;[3]:</div>
+<div class="jp-CodeMirrorEditor jp-Editor jp-InputArea-editor" data-type="inline">
+     <div class="CodeMirror cm-s-jupyter">
+<div class=" highlight hl-ipython3"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="n">data</span><span class="p">[</span><span class="s1">&#39;turns&#39;</span><span class="p">][</span><span class="mi">0</span><span class="p">])</span>
+</pre></div>
+
+     </div>
+</div>
+</div>
+</div>
+
+<div class="jp-Cell-outputWrapper">
+
+
+<div class="jp-OutputArea jp-Cell-outputArea">
+
+<div class="jp-OutputArea-child">
+
+    
+    <div class="jp-OutputPrompt jp-OutputArea-prompt"></div>
+
+
+<div class="jp-RenderedText jp-OutputArea-output" data-mime-type="text/plain">
+<pre>{&#39;output&#39;: {&#39;transcript&#39;: &#39;Hello , welcome to the Cambridge restaurant system? You can ask for restaurants by area , price range or food type . How may I help you?&#39;, &#39;end-time&#39;: 9.93, &#39;start-time&#39;: 0.001283, &#39;dialog-acts&#39;: [{&#39;slots&#39;: [], &#39;act&#39;: &#39;welcomemsg&#39;}], &#39;aborted&#39;: False}, &#39;turn-index&#39;: 0, &#39;input&#39;: {&#39;live&#39;: {&#39;asr-hyps&#39;: [{&#39;asr-hyp&#39;: &#39;expensive restaurant in south part of town&#39;, &#39;score&#39;: -0.512136}, {&#39;asr-hyp&#39;: &#39;a restaurant in south part of town&#39;, &#39;score&#39;: -2.429358}, {&#39;asr-hyp&#39;: &#39;expensive expensive restaurant in south part of town&#39;, &#39;score&#39;: -2.429358}, {&#39;asr-hyp&#39;: &#39;expensive restaurant restaurant in south part of town&#39;, &#39;score&#39;: -3.210609}, {&#39;asr-hyp&#39;: &#39;expensive in south part of town&#39;, &#39;score&#39;: -3.211676}, {&#39;asr-hyp&#39;: &#39;expensive restaurant south part of town&#39;, &#39;score&#39;: -3.347932}, {&#39;asr-hyp&#39;: &#39;the expensive restaurant in south part of town&#39;, &#39;score&#39;: -4.268281}, {&#39;asr-hyp&#39;: &#39;a expensive restaurant in south part of town&#39;, &#39;score&#39;: -4.34658}, {&#39;asr-hyp&#39;: &#39;a restaurant restaurant in south part of town&#39;, &#39;score&#39;: -5.127831}, {&#39;asr-hyp&#39;: &#39;expensive expensive restaurant restaurant in south part of town&#39;, &#39;score&#39;: -5.127831}], &#39;slu-hyps&#39;: [{&#39;slu-hyp&#39;: [{&#39;slots&#39;: [[&#39;pricerange&#39;, &#39;expensive&#39;]], &#39;act&#39;: &#39;inform&#39;}, {&#39;slots&#39;: [[&#39;area&#39;, &#39;south&#39;]], &#39;act&#39;: &#39;inform&#39;}], &#39;score&#39;: 0.9282257165548344}, {&#39;slu-hyp&#39;: [{&#39;slots&#39;: [[&#39;pricerange&#39;, &#39;expensive&#39;]], &#39;act&#39;: &#39;confirm&#39;}, {&#39;slots&#39;: [[&#39;area&#39;, &#39;south&#39;]], &#39;act&#39;: &#39;confirm&#39;}], &#39;score&#39;: 0.02923086440308149}, {&#39;slu-hyp&#39;: [{&#39;slots&#39;: [[&#39;pricerange&#39;, &#39;expensive&#39;]], &#39;act&#39;: &#39;inform&#39;}], &#39;score&#39;: 0.02236294307342127}, {&#39;slu-hyp&#39;: [{&#39;slots&#39;: [[&#39;area&#39;, &#39;south&#39;]], &#39;act&#39;: &#39;inform&#39;}], &#39;score&#39;: 0.020180475968662844}]}, &#39;end-time&#39;: 12.19, &#39;start-time&#39;: 9.93, &#39;batch&#39;: {&#39;cnet&#39;: [{&#39;start&#39;: 0.0, &#39;end&#39;: 0.525, &#39;arcs&#39;: [{&#39;score&#39;: -0.0001, &#39;word&#39;: &#39;&lt;s&gt;&#39;}, {&#39;score&#39;: -5000000000.0, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 0.525, &#39;end&#39;: 0.7875, &#39;arcs&#39;: [{&#39;score&#39;: -19.41379, &#39;word&#39;: &#39;i&#39;}, {&#39;score&#39;: -16.88836, &#39;word&#39;: &#39;a&#39;}, {&#39;score&#39;: -11.6955, &#39;word&#39;: &#39;ah&#39;}, {&#39;score&#39;: -13.1059, &#39;word&#39;: &#39;uh&#39;}, {&#39;score&#39;: -0.0001, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 0.7875, &#39;end&#39;: 1.05, &#39;arcs&#39;: [{&#39;score&#39;: -11.16217, &#39;word&#39;: &#39;i&#39;}, {&#39;score&#39;: -9.916637, &#39;word&#39;: &#39;uh&#39;}, {&#39;score&#39;: -13.39644, &#39;word&#39;: &#39;ok&#39;}, {&#39;score&#39;: -0.0001, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 1.05, &#39;end&#39;: 1.12375, &#39;arcs&#39;: [{&#39;score&#39;: -0.115834, &#39;word&#39;: &#39;expensive&#39;}, {&#39;score&#39;: -9.822551, &#39;word&#39;: &#39;ok&#39;}, {&#39;score&#39;: -5.432811, &#39;word&#39;: &#39;the&#39;}, {&#39;score&#39;: -15.02797, &#39;word&#39;: &#39;extensive&#39;}, {&#39;score&#39;: -18.94896, &#39;word&#39;: &#39;are&#39;}, {&#39;score&#39;: -2.254255, &#39;word&#39;: &#39;a&#39;}, {&#39;score&#39;: -5000000000.0, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 1.12375, &#39;end&#39;: 1.1975, &#39;arcs&#39;: [{&#39;score&#39;: -17.54265, &#39;word&#39;: &#39;ok&#39;}, {&#39;score&#39;: -0.0001, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 1.1975, &#39;end&#39;: 1.345, &#39;arcs&#39;: [{&#39;score&#39;: -20.00645, &#39;word&#39;: &#39;expensive&#39;}, {&#39;score&#39;: -23.83931, &#39;word&#39;: &#39;extensive&#39;}, {&#39;score&#39;: -0.0001, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 1.345, &#39;end&#39;: 1.4925, &#39;arcs&#39;: [{&#39;score&#39;: -2.212958, &#39;word&#39;: &#39;expensive&#39;}, {&#39;score&#39;: -21.30504, &#39;word&#39;: &#39;extensive&#39;}, {&#39;score&#39;: -0.1158337, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 1.4925, &#39;end&#39;: 1.64, &#39;arcs&#39;: [{&#39;score&#39;: -19.56206, &#39;word&#39;: &#39;a&#39;}, {&#39;score&#39;: -15.88049, &#39;word&#39;: &#39;or&#39;}, {&#39;score&#39;: -19.37142, &#39;word&#39;: &#39;uh&#39;}, {&#39;score&#39;: -0.0001, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 1.64, &#39;end&#39;: 1.7875, &#39;arcs&#39;: [{&#39;score&#39;: -0.0001, &#39;word&#39;: &#39;restaurant&#39;}, {&#39;score&#39;: -14.9832, &#39;word&#39;: &#39;the&#39;}, {&#39;score&#39;: -11.33279, &#39;word&#39;: &#39;a&#39;}, {&#39;score&#39;: -9.494602, &#39;word&#39;: &#39;uh&#39;}, {&#39;score&#39;: -11.74803, &#39;word&#39;: &#39;or&#39;}, {&#39;score&#39;: -17.88491, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 1.7875, &#39;end&#39;: 1.935, &#39;arcs&#39;: [{&#39;score&#39;: -13.69902, &#39;word&#39;: &#39;restaurant&#39;}, {&#39;score&#39;: -0.0001, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 1.935, &#39;end&#39;: 2.23, &#39;arcs&#39;: [{&#39;score&#39;: -9.268782, &#39;word&#39;: &#39;restaurant&#39;}, {&#39;score&#39;: -2.381682, &#39;word&#39;: &#39;in&#39;}, {&#39;score&#39;: -0.09705002, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 2.23, &#39;end&#39;: 2.27, &#39;arcs&#39;: [{&#39;score&#39;: -0.1433645, &#39;word&#39;: &#39;in&#39;}, {&#39;score&#39;: -12.21065, &#39;word&#39;: &#39;or&#39;}, {&#39;score&#39;: -13.78668, &#39;word&#39;: &#39;i&#39;}, {&#39;score&#39;: -12.6393, &#39;word&#39;: &#39;are&#39;}, {&#39;score&#39;: -13.47915, &#39;word&#39;: &#39;a&#39;}, {&#39;score&#39;: -11.99133, &#39;word&#39;: &#34;that&#39;s&#34;}, {&#39;score&#39;: -7.282886, &#39;word&#39;: &#39;that&#39;}, {&#39;score&#39;: -2.348785, &#39;word&#39;: &#39;the&#39;}, {&#39;score&#39;: -13.8318, &#39;word&#39;: &#39;this&#39;}, {&#39;score&#39;: -3.286841, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 2.27, &#39;end&#39;: 2.31, &#39;arcs&#39;: [{&#39;score&#39;: -46.60755, &#39;word&#39;: &#39;it&#39;}, {&#39;score&#39;: -12.64047, &#39;word&#39;: &#39;which&#39;}, {&#39;score&#39;: -22.6327, &#39;word&#39;: &#39;this&#39;}, {&#39;score&#39;: -12.27715, &#39;word&#39;: &#39;the&#39;}, {&#39;score&#39;: -34.75492, &#39;word&#39;: &#39;as&#39;}, {&#39;score&#39;: -6.923453, &#39;word&#39;: &#39;is&#39;}, {&#39;score&#39;: -7.139083, &#39;word&#39;: &#39;of&#39;}, {&#39;score&#39;: -10.67487, &#39;word&#39;: &#39;a&#39;}, {&#39;score&#39;: -7.923023, &#39;word&#39;: &#39;tea&#39;}, {&#39;score&#39;: -0.002173584, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 2.31, &#39;end&#39;: 2.48, &#39;arcs&#39;: [{&#39;score&#39;: -0.005246917, &#39;word&#39;: &#39;south&#39;}, {&#39;score&#39;: -37.23841, &#39;word&#39;: &#39;well&#39;}, {&#39;score&#39;: -10.74857, &#39;word&#39;: &#39;ok&#39;}, {&#39;score&#39;: -46.34189, &#39;word&#39;: &#39;gastro&#39;}, {&#39;score&#39;: -46.3515, &#39;word&#39;: &#39;aswell&#39;}, {&#39;score&#39;: -9.687854, &#39;word&#39;: &#39;child&#39;}, {&#39;score&#39;: -5.276869, &#39;word&#39;: &#39;shop&#39;}, {&#39;score&#39;: -41.20222, &#39;word&#39;: &#39;style&#39;}, {&#39;score&#39;: -10.09526, &#39;word&#39;: &#39;stop&#39;}, {&#39;score&#39;: -44.53774, &#39;word&#39;: &#39;sell&#39;}, {&#39;score&#39;: -5000000000.0, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 2.48, &#39;end&#39;: 2.65, &#39;arcs&#39;: [{&#39;score&#39;: -22.39381, &#39;word&#39;: &#39;is&#39;}, {&#39;score&#39;: -36.75235, &#39;word&#39;: &#39;of&#39;}, {&#39;score&#39;: -32.42216, &#39;word&#39;: &#39;how&#39;}, {&#39;score&#39;: -38.02597, &#39;word&#39;: &#39;in&#39;}, {&#39;score&#39;: -0.0001, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 2.65, &#39;end&#39;: 2.74, &#39;arcs&#39;: [{&#39;score&#39;: -0.0008488684, &#39;word&#39;: &#39;part&#39;}, {&#39;score&#39;: -25.69577, &#39;word&#39;: &#39;partner&#39;}, {&#39;score&#39;: -13.48137, &#39;word&#39;: &#39;what&#39;}, {&#39;score&#39;: -25.05888, &#39;word&#39;: &#39;post&#39;}, {&#39;score&#39;: -27.47913, &#39;word&#39;: &#39;are&#39;}, {&#39;score&#39;: -11.4708, &#39;word&#39;: &#39;close&#39;}, {&#39;score&#39;: -23.77218, &#39;word&#39;: &#39;park&#39;}, {&#39;score&#39;: -14.65481, &#39;word&#39;: &#39;code&#39;}, {&#39;score&#39;: -15.24271, &#39;word&#39;: &#39;cote&#39;}, {&#39;score&#39;: -15.58045, &#39;word&#39;: &#39;or&#39;}, {&#39;score&#39;: -12.75343, &#39;word&#39;: &#39;i&#39;}, {&#39;score&#39;: -15.39197, &#39;word&#39;: &#39;in&#39;}, {&#39;score&#39;: -18.09848, &#39;word&#39;: &#39;it&#39;}, {&#39;score&#39;: -8.927246, &#39;word&#39;: &#39;a&#39;}, {&#39;score&#39;: -20.65172, &#39;word&#39;: &#39;for&#39;}, {&#39;score&#39;: -31.9557, &#39;word&#39;: &#39;the&#39;}, {&#39;score&#39;: -7.264429, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 2.74, &#39;end&#39;: 2.83, &#39;arcs&#39;: [{&#39;score&#39;: -18.37638, &#39;word&#39;: &#39;what&#39;}, {&#39;score&#39;: -18.29665, &#39;word&#39;: &#39;for&#39;}, {&#39;score&#39;: -20.94735, &#39;word&#39;: &#39;or&#39;}, {&#39;score&#39;: -11.55013, &#39;word&#39;: &#39;on&#39;}, {&#39;score&#39;: -0.0001, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 2.83, &#39;end&#39;: 2.94, &#39;arcs&#39;: [{&#39;score&#39;: -0.000271893, &#39;word&#39;: &#39;of&#39;}, {&#39;score&#39;: -30.09363, &#39;word&#39;: &#39;post&#39;}, {&#39;score&#39;: -24.60016, &#39;word&#39;: &#34;don&#39;t&#34;}, {&#39;score&#39;: -11.64039, &#39;word&#39;: &#39;the&#39;}, {&#39;score&#39;: -24.75157, &#39;word&#39;: &#39;okay&#39;}, {&#39;score&#39;: -8.243156, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 2.94, &#39;end&#39;: 3.1, &#39;arcs&#39;: [{&#39;score&#39;: -0.0001754378, &#39;word&#39;: &#39;town&#39;}, {&#39;score&#39;: -14.22256, &#39;word&#39;: &#39;cocum&#39;}, {&#39;score&#39;: -8.735506, &#39;word&#39;: &#39;hotel&#39;}, {&#39;score&#39;: -23.20746, &#39;word&#39;: &#39;care&#39;}, {&#39;score&#39;: -11.17799, &#39;word&#39;: &#39;tell&#39;}, {&#39;score&#39;: -24.11379, &#39;word&#39;: &#39;okay&#39;}, {&#39;score&#39;: -19.73659, &#39;word&#39;: &#39;down&#39;}, {&#39;score&#39;: -21.45679, &#39;word&#39;: &#39;it&#39;}, {&#39;score&#39;: -22.00763, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 3.1, &#39;end&#39;: 3.18, &#39;arcs&#39;: [{&#39;score&#39;: -20.51506, &#39;word&#39;: &#39;on&#39;}, {&#39;score&#39;: -21.36264, &#39;word&#39;: &#39;down&#39;}, {&#39;score&#39;: -0.0001, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 3.18, &#39;end&#39;: 3.26, &#39;arcs&#39;: [{&#39;score&#39;: -11.9611, &#39;word&#39;: &#39;a&#39;}, {&#39;score&#39;: -0.0001, &#39;word&#39;: &#39;!null&#39;}]}, {&#39;start&#39;: 3.26, &#39;end&#39;: 4.14, &#39;arcs&#39;: [{&#39;score&#39;: -0.0001, &#39;word&#39;: &#39;&lt;/s&gt;&#39;}, {&#39;score&#39;: -23.32922, &#39;word&#39;: &#39;!null&#39;}]}], &#39;asr-hyps&#39;: [{&#39;asr-hyp&#39;: &#39;expensive restaurant in south part of town&#39;, &#39;score&#39;: -0.480987}, {&#39;asr-hyp&#39;: &#39;expensive expensive restaurant in south part of town&#39;, &#39;score&#39;: -2.578111}, {&#39;asr-hyp&#39;: &#39;a restaurant in south part of town&#39;, &#39;score&#39;: -2.619409}, {&#39;asr-hyp&#39;: &#39;expensive restaurant the south part of town&#39;, &#39;score&#39;: -2.686408}, {&#39;asr-hyp&#39;: &#39;expensive restaurant in in south part of town&#39;, &#39;score&#39;: -2.765619}, {&#39;asr-hyp&#39;: &#39;expensive restaurant south part of town&#39;, &#39;score&#39;: -3.624464}, {&#39;asr-hyp&#39;: &#39;a expensive restaurant in south part of town&#39;, &#39;score&#39;: -4.716533}, {&#39;asr-hyp&#39;: &#39;expensive expensive restaurant the south part of town&#39;, &#39;score&#39;: -4.783532}, {&#39;asr-hyp&#39;: &#39;a restaurant the south part of town&#39;, &#39;score&#39;: -4.82483}, {&#39;asr-hyp&#39;: &#39;expensive expensive restaurant in in south part of town&#39;, &#39;score&#39;: -4.862743}]}, &#39;audio-file&#39;: &#39;pt344x_0000993_0001219.wav&#39;}}
+</pre>
+</div>
+</div>
+
+</div>
+
+</div>
+
 </div></section></section><section><section>
 <div class="jp-Cell-inputWrapper"><div class="jp-InputPrompt jp-InputArea-prompt">
 </div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput " data-mime-type="text/markdown">
@@ -14287,6 +14394,8 @@ Tworzenie bez znajomości kompleksowej ontologii domeny slotów na podstawie his
 <p><strong>[1]</strong>  DSTC <a href="https://github.com/matthen/dstc">https://github.com/matthen/dstc</a></p>
 <p><strong>[2]</strong>  The Second Dialog State Tracking Challenge <a href="https://aclanthology.org/W14-4337.pdf">https://aclanthology.org/W14-4337.pdf</a></p>
 <p><strong>[3]</strong>  Dialogue State Tracking with Incremental Reasoning <a href="https://direct.mit.edu/tacl/article/doi/10.1162/tacl_a_00384/101875/Dialogue-State-Tracking-with-Incremental-Reasoning">https://direct.mit.edu/tacl/article/doi/10.1162/tacl_a_00384/101875/Dialogue-State-Tracking-with-Incremental-Reasoning</a></p>
+<p><strong>[4]</strong>  <a href="https://arxiv.org/pdf/2101.08333.pdf">https://arxiv.org/pdf/2101.08333.pdf</a></p>
+<p><strong>[5]</strong> <a href="https://www.cs.hhu.de/fileadmin/redaktion/Fakultaeten/Mathematisch-Naturwissenschaftliche_Fakultaet/Informatik/Dialog_Systems_and_Machine_Learning/L6.pdf">https://www.cs.hhu.de/fileadmin/redaktion/Fakultaeten/Mathematisch-Naturwissenschaftliche_Fakultaet/Informatik/Dialog_Systems_and_Machine_Learning/L6.pdf</a></p>
 
 </div>
 </div></section></section>