From 67119344599da9a864ef07b9245851bb74be8a11 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Tue, 23 Dec 2025 04:01:03 +0000 Subject: [PATCH 01/63] Update gitignore --- .gitignore | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/.gitignore b/.gitignore index 8a3365b203..90d001609d 100644 --- a/.gitignore +++ b/.gitignore @@ -172,7 +172,7 @@ test/default test/housing.json test/nlp/default/transformer_ms/seq-classification.json -flaml/fabric/fanova/_fanova.c +flaml/fabric/fanova/*fanova.c # local config files *.config.local @@ -184,3 +184,6 @@ notebook/lightning_logs/ lightning_logs/ flaml/autogen/extensions/tmp/ test/autogen/my_tmp/ + +# Internal configs +.pypirc From a49492f0d67ee9d53a6eda0557bdb9d46fa0c003 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Tue, 23 Dec 2025 04:01:48 +0000 Subject: [PATCH 02/63] Bump version to 2.4.0 --- flaml/version.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/flaml/version.py b/flaml/version.py index a0b06b867e..3d67cd6bb4 100644 --- a/flaml/version.py +++ b/flaml/version.py @@ -1 +1 @@ -__version__ = "2.3.7" +__version__ = "2.4.0" From 8152679fa878c364fd2a2483fd741c36c39b3ce4 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Tue, 23 Dec 2025 07:47:15 +0000 Subject: [PATCH 03/63] Update readme --- README.md | 49 +++++-------------------------------------------- 1 file changed, 5 insertions(+), 44 deletions(-) diff --git a/README.md b/README.md index fcd62bd423..30a2b14577 100644 --- a/README.md +++ b/README.md @@ -14,15 +14,9 @@

-:fire: FLAML supports AutoML and Hyperparameter Tuning in [Microsoft Fabric Data Science](https://learn.microsoft.com/en-us/fabric/data-science/automated-machine-learning-fabric). In addition, we've introduced Python 3.11 support, along with a range of new estimators, and comprehensive integration with MLflow—thanks to contributions from the Microsoft Fabric product team. +:fire: FLAML supports AutoML and Hyperparameter Tuning in [Microsoft Fabric Data Science](https://learn.microsoft.com/en-us/fabric/data-science/automated-machine-learning-fabric). In addition, we've introduced Python 3.11 and 3.12 support, along with a range of new estimators, and comprehensive integration with MLflow—thanks to contributions from the Microsoft Fabric product team. -:fire: Heads-up: We have migrated [AutoGen](https://microsoft.github.io/autogen/) into a dedicated [github repository](https://github.com/microsoft/autogen). Alongside this move, we have also launched a dedicated [Discord](https://discord.gg/pAbnFJrkgZ) server and a [website](https://microsoft.github.io/autogen/) for comprehensive documentation. - -:fire: The automated multi-agent chat framework in [AutoGen](https://microsoft.github.io/autogen/) is in preview from v2.0.0. - -:fire: FLAML is highlighted in OpenAI's [cookbook](https://github.com/openai/openai-cookbook#related-resources-from-around-the-web). - -:fire: [autogen](https://microsoft.github.io/autogen/) is released with support for ChatGPT and GPT-4, based on [Cost-Effective Hyperparameter Optimization for Large Language Model Generation Inference](https://arxiv.org/abs/2303.04673). +:fire: Heads-up: [AutoGen](https://microsoft.github.io/autogen/) has moved to a dedicated [GitHub repository](https://github.com/microsoft/autogen). FLAML no longer includes the `autogen` module—please use AutoGen directly. ## What is FLAML @@ -30,7 +24,7 @@ FLAML is a lightweight Python library for efficient automation of machine learning and AI operations. It automates workflow based on large language models, machine learning models, etc. and optimizes their performance. -- FLAML enables building next-gen GPT-X applications based on multi-agent conversations with minimal effort. It simplifies the orchestration, automation and optimization of a complex GPT-X workflow. It maximizes the performance of GPT-X models and augments their weakness. +- FLAML enables economical automation and tuning for ML/AI workflows, including model selection and hyperparameter optimization under resource constraints. - For common machine learning tasks like classification and regression, it quickly finds quality models for user-provided data with low computational resources. It is easy to customize or extend. Users can find their desired customizability from a smooth range. - It supports fast and economical automatic tuning (e.g., inference hyperparameters for foundation models, configurations in MLOps/LMOps workflows, pipelines, mathematical/statistical models, algorithms, computing experiments, software configurations), capable of handling large search space with heterogeneous evaluation cost and complex constraints/guidance/early stopping. @@ -46,10 +40,10 @@ FLAML requires **Python version >= 3.9**. It can be installed from pip: pip install flaml ``` -Minimal dependencies are installed without extra options. You can install extra options based on the feature you need. For example, use the following to install the dependencies needed by the [`autogen`](https://microsoft.github.io/autogen/) package. +Minimal dependencies are installed without extra options. You can install extra options based on the feature you need. For example, use the following to install the dependencies needed by the [`automl`](https://microsoft.github.io/FLAML/docs/Use-Cases/Task-Oriented-AutoML) module. ```bash -pip install "flaml[autogen]" +pip install "flaml[automl]" ``` Find more options in [Installation](https://microsoft.github.io/FLAML/docs/Installation). @@ -57,39 +51,6 @@ Each of the [`notebook examples`](https://github.com/microsoft/FLAML/tree/main/n ## Quickstart -- (New) The [autogen](https://microsoft.github.io/autogen/) package enables the next-gen GPT-X applications with a generic multi-agent conversation framework. - It offers customizable and conversable agents which integrate LLMs, tools and human. - By automating chat among multiple capable agents, one can easily make them collectively perform tasks autonomously or with human feedback, including tasks that require using tools via code. For example, - -```python -from flaml import autogen - -assistant = autogen.AssistantAgent("assistant") -user_proxy = autogen.UserProxyAgent("user_proxy") -user_proxy.initiate_chat( - assistant, - message="Show me the YTD gain of 10 largest technology companies as of today.", -) -# This initiates an automated chat between the two agents to solve the task -``` - -Autogen also helps maximize the utility out of the expensive LLMs such as ChatGPT and GPT-4. It offers a drop-in replacement of `openai.Completion` or `openai.ChatCompletion` with powerful functionalites like tuning, caching, templating, filtering. For example, you can optimize generations by LLM with your own tuning data, success metrics and budgets. - -```python -# perform tuning -config, analysis = autogen.Completion.tune( - data=tune_data, - metric="success", - mode="max", - eval_func=eval_func, - inference_budget=0.05, - optimization_budget=3, - num_samples=-1, -) -# perform inference for a test instance -response = autogen.Completion.create(context=test_instance, **config) -``` - - With three lines of code, you can start using this economical and fast AutoML engine as a [scikit-learn style estimator](https://microsoft.github.io/FLAML/docs/Use-Cases/Task-Oriented-AutoML). From 4397fc154c666837c63f129c060723552d1cb444 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 25 Dec 2025 05:17:22 +0000 Subject: [PATCH 04/63] Pre-download california housing data --- test/cal_housing_py3.pkz | Bin 0 -> 366871 bytes 1 file changed, 0 insertions(+), 0 deletions(-) create mode 100644 test/cal_housing_py3.pkz diff --git a/test/cal_housing_py3.pkz b/test/cal_housing_py3.pkz new file mode 100644 index 0000000000000000000000000000000000000000..201c864b2fb4f89b00f875b1a44663ba9bce99a1 GIT binary patch literal 366871 zcmWJsiC+tj8%7dRLlQbBAxTZ&q-zI~keYr}(lH4kbRSzgBni z)50$nzGN$uSUQZX3JoAM=u2w3^W+|y;$+15mz`@ht&TO9PIEm$iHCy{~NA4 zVoF-8r3jG{Xlhjv{~L*e3q?{M5NEn-|MEsD1pRGe5^f@%6x~GEP`FPRIL#T2i=0A1 zkjF{*e5Zu&#|Cxqv8upddA~;h;4g1;lt10@D zlb_8sdGnXgU=z>seKN+D_*=?mJLBQW^svP~(`J70J(x)s;JT>UZmX#}%u8ckC;#x5AD3lV0SHQN}Hh1#gXh8kxXLk2mhj_TIWY~q?as-g<9 z@^~p`v5Haa+RXKdC{NZXqRencnZ`}@?Br^-Lped^TX4` zJ4W;2A;1Y-=PRDV0dY?O2eiFCNQ19qj~F5D&P%4qb=Z?hJ5aMY5MS)UO<4GjrPa(h z@4XxC4!X&#fs3M$Z$oi6;N9)+WkZRnXw!;k7uHt6J zN$T3RPTgj^QEL-x?0^fp&4)qmgxzBsmanGg7;hw7Uo&g<<)3{)P#T9q`*ptqsPh|V z-rYDVO%6a-lTlGlAV|Ip(Va-srRO}lm>bnJEASD>g0F{x8?M$2@Dc29UU_O(#i z%uCIwCqg9O7|KVV5k=9A7vj~5Qz8)bMa~au&l=_l#l88*pwHYJO0Vk0^-JUDG7YpS z$K}nAh3S#&CgunOyC1;`mV{Mv+ zMRvULwPf6$`Th^j19koqKgGVFqCwKX{z{Cq5Mw?c?em);OKn1X*fe2=n2b8DCNdRa zwv<>8HqiyFRAlM!Hc@WJR}eraYk}l9uCig*l7&@~_a} z)x{nSx!l!Mp9iMo(HTY^K{JY^LjTt2l{jt>IHDkdDH&>?bqC{aNt6G}R*irc=06~v z;J~#sqVbyy{IEg@>I1qIBSbSYHK z=OPSPsGn=f?$k#bsLg=mJHUyWs9-UTV*^hUpQK-Bkx9zKqGu6JsI#X4>^{Fp$uqC< z5e$6Mkx|ccmiN^>BEplfv(l)2FINShEuuTDZHo($de0iqzTY3oH%tkcg0G-FCGV}- z?HsP0+r*i)>v7&pUW2x;-Ps$V)MX7HSa_8E#Y26U9AMhI0n_c`cY!VT0F2%y4?UK> z=o`d@-%5IsF5Liw_qSCFn-$FYZAmUkS}rfYYl*gyVVZ~~y``*0>aO;2%p`RI{>yVr zSHag7RoZm7oyB@k`jxF~s2|6A{v!G+MU;j0tR8;`cVyYZn8&sn?!)IgkgBM-&qV9o zv@n$C8AqVY^TE}^=jB%y$W^@u9)bSiAuqoya=g+C7Z9S-BU)q%=C#4BZ^sqs8q{gN zA_MQ6ni$XK8-*%IBHSs2txEwiV~!Ahh}%3FL0Q*!X_>GhG1!qLv!Abt#~-^Fb_*f; zIND?fd8vruIl<2EB@h$li)juVQy6s{g_raFG6zCVIXZ^B(krObrLSrvGwLlf7FZ+oxs>tUbqSkM-TRgX0#Y)Iw3QPulTyOfZt|6Ug*m)D zjg)`ibYvM+2H`5UU`lSj$nDIStV{nd6W zU+`IIy#ZBdF1H#8CCnFi%x8dW1-`X{3cC!U94em#c2W#`x)MW!3-1B3%}Uta{ObK^ z{KC|Y6{3MPvbabmxGe*t&6zcj&VBhUw3*bZ8j6m4?t1}%HNsL=+psme!9?E0EktRB z^bY5^=*i6sDjx1FqNysTkYIRGG`+-mgAgR!FLk&wg4A~|fkd2$8j`F0qf~9h;06Q) zYYJ)g4bL=TNmz!_BT8_Rw-~x$!TKNL1ZP1@iPPfQgWDv*#uG+fZdB~%WHlZvLC(~e zt1BWS>L_hGb{e)YI5!3hpmwAwVnDH)WL{fwGb;Dg1sLQ9-8BYdPsXiVinDLEe<-G8 zF@rdKNj_Ddwky1tU7nulCp4{VN^cF5(AZot17yTRpbU|&cqG^e<+2{zyEU1momtgH zu@3N9=x)+6@o)p$2XNiLjn)29kXga$B0@;KC~YET{(daV_0LDf}BRfkvy zP)8WK@ha@t2Dz0m1}b|N4Y`xQHM1}U^72zU_RkTJZ@dZce)-+STP4h4%i0$zzfLgd zQhjWbU5#Cg;_T9mP)q4aW9i6^B}EDt9y9aoeRss<=xmt7c%tERLE$>csybihiJmJH z-i_QZw<+v^++DpYz6n44vV~GUjpK^5&`lER6*c~f5)sLrJtXQNKh_T|@c#_RTthed zYqprJ%r&uD_)&dqFgk9-&7Zc(tFRS1)SE5R%!?{$c>ZthAI(g-ajP`*<1FVt*I?0D zcL6*1qQ-h6dY^Dpnow!j4rIBbbmCCP=rH{GTfp#0A%s`<$$|TnGPO!N&*YR^2}I^g zp?aBR^&uwX!Chr%&`t`j0SB{FIk&C#aFDxbQ~WXdr(e*UdW0nXnvC(yUWcLin=%Et zZ*BSr32SN76Kf^mH+!w2E%anP<@NAZdmYT{R~VENh4uq_;B1f# zEtcdzEX2kqd7o&DDjd>P+Co2V5ZzWr01N%G0h6gC=NImH{f;H3D>?~>Y!eS7oVVck zZSADW-i+EK+~0k(m``14kJu+QA$;c7RS9-Vt5H2@TF9%f>)I;*rWO{u8Z{W>5=T^_ z`B!jLk;e&bbDlBZDcDLh;dsgnX)UAur|$ztyri1M*l{16np?ligj88e;aWM`qQTjo=_tP zV?{UBhQp9?S1?JwTX`@nK-1%;lmHsNUm0Ra5Kk+thPR>s9ZK;hp&QH5^?seW<6BO_VIdA>Kj#;;w zf!>TwgdJpS81jGjCUCl_AAT%$Krf8n`W8S5vpmji=e8*qR+<`KqV7O^NuJu0>=si3 zz!7vcCRBC7G8L)CjNm>tc}9toW@q)a=xv+y-!YMAdWZhpsdj2ZN1*vcY4ww-plb@7 zMcK7o*fuM$WSR1*Fp2)?*CJ$g;s^C9_Edh-Mz#Lb`CT^q{iKoW zlceF}t(ZOP8RVE|m-@ogo)cTu*QWN*0RN}_8@8ypgN$IQ*c}+gRlc!uWY)Q*=H!E1 zCbegCI``qN%lYXWVafTjK+myvAazKe7Pm4O($z~ib>k4+B_ z*>2RXU!EY`n|vnv8^yY@Ljwgp$3;QqEgK%?C3(c4ov4XTRIkEA^3D9_=GPfDEgm6E zS3}2hg2&0;{dz8A1x~{@^@QppgP3@y%!^ns+R zsr(}Rg2fbo<#8!eAJS_^Yx! z`doV4mbOendYDlUEovgsx4JOiGyIc4LRv?CtRKj1Ex3ZyYx_gEN5{){=EKiz?Ww!V zJw=^N2IZYHR~o)) zW>(gA#S5Z8d;rCrG=Ymgp!YIellkM;FYt$ZRVpZUD@1$TA!YHtW%##A3y&jgmh((+MJ36vgkau!2ke7RPm(ZPj=PxA`$~B+sfR3*V9!kq39!k_jE~fL!<{t_^iogR# z2~ja2>ODUxmkIBHq3pkR2vdyx)XrRc4YtMQRH)hKypqHnzz0fgp(XjY+2N1^foSi< zQF`z(@FiVT=~U{#WO~Ds%7^|Ixx3golpW;ke}1g&4keptuHIUvOKT%ul(d)s#J_X$ zsD&-rq0gdH`0QWxlAWye>iG}GvYoN7i#yQtsi<I>?Lhq>h%~dT3=GtB5UA;Mg284SKV2lZ7e6W7 zNx3aB9C)`N?08CSw%8ip0u4MjFk5^re0Ua0j@4F|g?_S+Ms`du=r1|SUe4x#5wfG4 zluqmRlf8tw^!a_|9{KRN%QF9Mc;z@~C;`>L*;9Uu?prvpQ|iBUz6Sir60v-Iv0a@? zIZn4z=;Jb#KhhAf6I&q08vl`%Q^zAZ&IIByb%|lb+;Ex@;>H8k<5lt<31+hWvhdHA zbS6aaf!jv7r^}?{3tdyw6`Gfs^c1H`rOxaIFxQT=8g@~gBhM_i2|2hU zg)7%*0Ux$P)uwfh1QwnI`=XD?qvWrrMnYlN3ooZ%4c7IKJ_@}AETV@YtU=ybxG-_b z($-`Z4r!uR>=Z(cG|Q2=T5rWkb*PNX(s+?b>=`>mM>6mtY>GmK+@nmNSooUuJJ=v8 zb4t~FP3b6FT9m@|+dvi+f5*a93M6n9(B-L^d=Gi`7n*dLV$(R$LNHDQ%=;oY%^$1Z zuUHM6nucPZxEeu^j0MDaBjY;5pK10(AI|eN*9$ixx=fEv%r;2Jd9l-)d%>8k^Zb>< zFPm#WZB>V;j(;BQg8u|0nu%c9%rh;{_hFm;Dxd|tKypp-sK-cZXv4|5%E-i>a8c5f zaf?nYpu&t*w-@kkAx5d2k)&C+q>F7>u$N92e3GE23f*OIcijxY2ZT0sPBd#oK`)3E zBRqU|Q~VI61!w=%0=aG1<7S^2vfN3zo!L&`4tG{RbfR99>g~?v)@{;Iby?H{&2RyG zrrWJutNj?0R)B{sIdIcIg9D`5%94svPl`GHee@O&->`{FU19A5zQ!iMO>&R>g#Mqg z8+N^T%{1vQ-IKBo`zp(sHM?NcnkUl37WUhfjX8P30d(JUkAS5{!f&xWt!eHK7;}>{ z^fygVx+{=pe9MSacj;!*t!V^?9R22%fY4)Q&+ZIJ|)$3F&sItXdC4`r_1LIIahj$0haEUp*kQON!U|2q_l3T zu|jv|2kimrmHL~Z>XFNv1*;xzBX@L!O*(<=Rlb*8z}tqA0hdBP8~uvBpPv)tH+Cxi z$PDR&2bA|8fs{w){GC20c49qs=Y~;!9nSn1GU+fu%Z}fD>*dn1!o)W0kHxNg4E7N~ zE`l~!+HTM@d?CHORqbk#-!|T7^kCjv-#FM4bJVCc!}P^yI`V9gDW#T$skM~k#wPB9REU4yx{0xdW%a$Bfx0%vyS5`NUq{!W$2vf>vy zqu0oP9~t&V(rA>Gg6r$%WLrqH`1IM%F8nZv@lS>{Z87Tj` zlSPkW{jq5oIkh;F9^$k%XP-J4)+; zy0LOn zi_7p=tkP_qzkvO3DOd0_$3H^s#ZHr)Q-&a&f8zvhyDi-StZOGpS%kBE^P#G@yPh&5 zI(Dm}*_d(X1$IN#439IFXCLr+i+j%I{EF{6W>NG5rV>x(t%fU zd8KHR^g#UA6Uv4>*bw1deoE~FrG76Y{>*u*RCj8I*B*CrE!d9;5f%WqgTHv&m&cZ$OF`Hu+wKx^` zNV*S)C;cOVg%R;2?j*30Jd8=J<&85ZJ*`K=A5u)VPVJdBC)9+UJ+~oK!b=PW;S_Bx z@!m7!hHC683U1q!LzXZL(6_7QW-GtG61IkBtXU)R=z37KZ^z;Qkt54pFEKaUrP^I8 zc?YB~wxoqHmCky;J6JlVlJ|)0$wsnVa2wq}%B0~BEM1d8a1E{U0H=Q$i zEz@gD`midH1CMT;?PV`^yP^U``}yO;pSn!PbTIB^!`0mHeBQW^7M7rIdPMV(^q%TQ zV*xIip3|Ab>~b79@rg9-E?6%A9B;0EQ`@j`Yylv8!r3BP|MTNV&{<58z^Etztx@vE zZmdTATC2fM8}1Y$p`yIM0Y~9&d_^jlNZ%$2E+VpyiQX@@XeXB^Nao2O@Hm_t8utTi zO+CuzzpZTZKaIKkHBC}NuHDQ=l7SF}6JYizne9wAs)Hl_vVBNMMd9~XV&cDNpx>y*&jr3@kp5oaR=&sZlST#W0?@2Ed=&q6Du9x*GB zPak%TpC1^BF3~(I{Fp?Ubujo`y4&7cRB`n`p`se$(U&+~E`6CGj!#ab&buH%xb(Sp z-2r}XP})RAnw0g81K@1vbi`}!klljs2y(Mdw5vkMiwLA?e?(Cs&ro2K_42nbhgdP- z0PMGjW%k_FI)xtKK-9}V81|Rw{6s4D z8DviOOq#dkpD>)en4C55RJb_*wRreZ*m}jGBoNQ;FYGT(q}!^ic?Nk4-k9AysOu4a zfp8!t(~OONOKLlqEZ8XWmvLK>u!n+>eVKF;eZdFWoL|A)l$ID-k8hkU?hC z&TlXLXJ}HZxC_da7_aE73M{KcN)lmz;(Yz{zFwr1O(P>-D{k?K5-`pVwpyLb(dVcQ zqoVJ7xEobMd6Inqy5EFP_)IW%&)o7Xi~H%{QxN;uTq$KLKw5l}v8S=B+O=LO!$OTtH_D;*nh%LaF)516N#okcT31y4)u<^qAIdz)F1)<~ehSNz_<1cWT z0hUV^vYUNG6F$t!HniGYs@fM>u?g#gKxX$sl2!`N95tSd zsBQAKMAy2)>-1|0&9z}3TSGbvl?|(e=locSZXmCJeu=Uw7@LMY9&%3zG5xaor{XnksaP6Py9~R>IphacEBDZnf8154PS90Zn>{0(ha8BNTp}Qr z(QPk}PZOV;eo}rd>#ka8t9)ZG0DKF<`FA%x#k@Qykp?tntj9St7{X8oo)s@>58MB- z{!Mys^rr1q(8b_WkvWK2ZNYN*1xO?JH1v)16OVj~aK!AodM9kBWzjcf0s7R))=uX4 zXlxDG#9i}BXC$5`bVWYgP5g>`3$*YrR#D>Uk^EIdHDAxOv5xh2v&;KkpL8|V9M3hU z9VLu!u3^6*+z_K2&1RXi0r}P%ggku-&w9kkiT{iyR`q6&%Glc$|G2dqEj+ogJcc{zjjr$(U zEW0p4$AzXhFO7G2Y@%<{KN`0(VH5VzTL`wumKV&5h%uBmotrr$*Fx2IV15=ix^6Vx zY4N1FX?|;q!n+abZ0ce}|I3Mf(hYj6N*jAJapPZ52RRPva~XxE(2rBI%w7(8P|lj| zcDg_NtKD@({8$>S^k!<)_aqzEW@`{?uq>WMFi8a(z&@}t`$)=*yky3P4%=UaSH_8r z2$s&QHs%Lx7onTGx?l^`IH5)c`HgrO_Cf7NbxdAUb3O^E*xc|#X|+|EBZWtseV^){ zUDlcb&9)6@(%l+TWNXf&AU&eZ)4K>lRv3@psC|Exko-6V zei}V4_!`F5xxaCExgtb)L~u6V>`Gf>`X$7x@l0Gd>xuF)2)(IChXKEY+lCkHjouNfZb1nP=2ZvN{I$r~1zZ#vWCDuXe2;){g%at%W6dVF;pg(}ot| zMJ~SqkDI~g3rl$tmZc@2QQ5$;zP5&c_vG*f)u^(B6+qNwz$PoV5$34qq046XE!%sB zz+K*_$4`is&u_<_1T2o;#aLkbvxM71;0A@|jH-{`cazHJzD~mqXgW$irOw@C0dfmZ zpoizKJOdmtOI0TL;ohmvQU^$)3uT^iaNMVQ@h%?sk<7ht7v4=G$bKDe#XXFD@l}hP z4S_tbc0%sF=Xu2Bobq5pY-MucLL=+iRQIHKo2pd5!6%uT9Y3>lIE8lq(@Q2H>m@%?N&rFoM z^nM7QbbEN_o#&(uP#fdVALJypn#-T6QxIc%gW9_s+UM(1z$ZeDU zVEqx4b5Jv-AKrx4+vHt+|-fyGuk~e&rcS8(eWzf1n#qZ zJg)&ZFeMl~L(IJmTcKY+zwCiZ@a=)BCX}ubiR3cRAV- z&dL!abQgvrJ`@Iv-qfFJHoOobD(E{dIMrc4SUqqU3vOzF)9;2{f~{Mt?3xqas~|KM zd_xHgyEAyeOQc_i%b(^F)p|3QoF+8ozS(hg`P?%vLftZ#txO@b|5!Z`)5Eeuk4+sP zOtywv$?-deBkYxls11bYA(^-n>agZ%L+|n^gE!$n-45s%2g>UDa5T~4=ny1h`4KL-iQ6zoJ^fV;-e#Vw}t(m_|~aldMKox9Z(`#JF&qUIvco zfM;Z^A-VCR87tah_pv+be(}O5H8T=(>?TMn_&DJXJL)l`s4g6xP6&>$ovs}-jl!wArxcD zQ=mEc!x04M8H)JHjA{-9|H+Vfc$a9-qPO7a?TWdfC`y|b1bSy~jA>=kfj{YISNM*n zF=%J%;CvLr@;!X;wB;I)gOr!ZBz=XU31X;W4H0%;Pcxc{R_y9dP7#myF=2b*&2qE+ z)eW~{dnnJN9GS6i5K-#v%Y*U##5SH`;}IF>1nhmtP8|PNKvc zt{#FgPYmu_Bau^0eo!9ai_NZW7mKSZ@wro^n~ljF{IjpxbQ}!PRyc8LJTaEHoL=Pp z)pq_!LG;kw9pI+aA3uR^ka92t;MxU5_rOfA1!Lv7i zdjj3kdGH~hj}tYQ&KO9NL;dHD8OtQI=j5xU%Ss#PU)z3j0nR-bfc!Avc$=aPIN{+l zhR|zHWrIj}yLfBk1QFeN@8BBCXs736nh8(goV2*O$LZ{K9S3Xx$lm0PJeP7P; zo%~<-nt7juHPq#}5M{GmsnSEeEj;R!){(+u7HaocRv+oO$aAVX}vNwW-1f0$OU zt%-MyKkBu`sJg$SOmDnCgPazYg9bhtvYTp-m@e;<|Yc& zG#$MIcOF^Pnf@H@SpN>OpFpL>3C+=leW?O(-^6VcZ!%C|iM@kF88@szi{DzHc8P^0 zfsf$D{zCnj>m&Bdfn1FNAg{`u>M+YDl-$OGjGp+u#|zFG;x2vZwzoP?7L$KibLP|?tnOXyVe+<}h(7-{`B^tsJLyC`O8#h9H&t46dW4T9+B^wlc~8-Ga62fym$Djn(O27P zo0g~2h2UMl|1s&5s24*-gLv>b;coSrbQw;k?!wo$W*$fQ8RXiF zPxt`$FiLc*s-yN|x*H#RAAHlc7Ot|H+RfAEhxL_`UC(!I06%`au_0>L(uFqOSSS5J zjO3l966-H;bvpdcQ9<^RtrE?Idi2t38`LLF z@pU0v^i98(%`Z|Nj{@uN%#*^wlU{AONtAr@DsDO?w7wnD_l^URn4_2e;{cV0j3--d zyK+Mc*Al%ks|0ODK-G|B{4~ylLHfWw66(6tXW9)=ogGZyop+M{wpppla%-iJ;NDY} zDdKm8({ToJ_8y-35<{63_45a_i#_YVtfwkUaR68IQfaE*0r5p8Vv~oLgB|(;m$oV5 zonx*&ZQFVhcztdu{-?OtRdxRWx8HTp@s#3RwXi8K7qeI6%_#qCswzRG@TP~t_nI|f z(gh71y~`wrcqR#K;H*W?pW+ooG&}xp+$2dT>$(FmyLk^D3i?PWi2{G^X>#}^e*rxT z*bhhOB){jedEk*)-|2P_`O2IF%YhU#rQ2DXRTYn#7oW~_8*_5K975SwF&0o+*KMOnOq7bG$I>&w&xc-Bs^8JG< zS}DAx{Leb<7+dPzi1@`_;eEs)s;8t1f(;~=zF<2){*ElbaNE?*Baa{BmgKmEeFrYU zGDi(vj&XnMk2S<9$s0oQm2SR?y~I?g*&NZp^cVpuZ+zf=0Pbt7jP-O?Co`*%zk5-6 z6i<>tX5`OR9J${@ocx+@y4UyItRpa;gisG31ZB z#b}UpU5jrmZpenz4NBIWOkm%qJ3uxfmuzFST#rg4-|L7XCMiJqe#~*z`X1^j9e{GP|o!5GPddYeJ!!*3PfvRpE*v)Z}+@JuczXixjdhHe%#| zqO5@itp6XwpavJlxuxZ(j>+Ab>Vc63;r~S1QJL1q1ZGhgr8_3K6nJSz0e$%G2^Rv^I~z}Uz4>RFbW7mYlg=n=cKA7m$)^HL>@RT z!$$K!4q=(y#j8>iy!vFdy|007#nj-c-#!p&q$NwBx7eC(uhN zojZ3kmL3r#qfa8{|2J4=g8V4Gk+28Z?%CmJUTBM}9>UxKSSu45IeREOkdN4R5YJET z_J-Y;ZjddX@_49QiX+O{4_S3n>hq4%k#opf7PD&lwP6^?c2mTj@(-0$dgyZ zZZT3lACwFzE$zYNh+BKPTDf+APKAvAa+d*1jEU6JsL);M{lmS~c(v;g>4=mkH)AZl z^JFchv;Pu^78;b8pnz~0pH({o|MXB;^n^8-cCFeMG(6WFuR(H)1zie*5E5%P_LS$;XWv{n;x_W zi_FCCpX;Bc8xpQDS})6JXbB#@RkaGdcA!c+&U!SJfvysD0cBSoD3ZwtFWfEpGUP~; zu2rpuvX?&K*~_N3P}2w=aaPWRz%SXZxvK3Dzi8v14n^s_=Aads8mt*RSY9#flM{k4 zqJJfex+#zIlr2Cu9GUTWTb*y@+OV z2euw|^gu!V`#?TE#Iz^7l#>?)QeR(4klA=?uv^9?Pi;(bdlmwz7?1qMaH0&_v%Af# ze6|&F1Aorsb@O-C*`MQ#KwO zgTF`d$88<9DI9ynwV5;Cf1~YYrvK8RKi_{Iq?v2-K9}zzgnt$?L~f>qcYl8MZ&m2n!(6|z z)E>|~_zFz$fDqF2wSk?WJSVdEQyxIAL7|?a&xis#woZTC4RDjNB9D8&09Ej}{LV?z8DH<`DDE?kNj0$v!lR>zTspIZ4(@(Y9H0nG9s@mG5)oB&fon(L==dxPz||_XU6N0+*T9 zis>tX&{rO+AIp2W;qK!9&B3Kh zgr;1WPc#SbUNB)aTMl4wU83W2^=A{;s*4!=55s0vy-(+E`&JFLF>xRCJ;e8y>Ra+1Hf; z80{8-s9>_&b7IT*A#UHYo>|9>s{b}-Hho!X=}#-Y)z_LB-iP`^S_d=?D!XCL++=z? z`BL0FOp`*FI!QZbb{uXwk`_$s7g49=$eAfG4(=Tv);+_E=5*dK(~7_ zLXQ=E0Sp0-t6uveb_a#>Vk%b4YqL}b^)!m7-Kle|8d%~MWggVN#0^hLgTT z9zkq_;EZjhA;GGH3yL{i`fujRlE`{#@!YfQGn1BsL+{3&3Db)1PK0%k*Izip|2W7H z^FBNj*W6Z7(5J+wRXpku|l42wvc zk~Y{~MOw25MScbX7cO>K0$>*OJYKjYFjlg|ukWtwImg&&>vo7u6(@_gJ) zF)tX5y&JXJi?X4R&~3m2WkP zN|I?#VvuZ-DoImT30od2Lt*lPA%8JFxqLAjC&8f#OO{@frZ)kB<={(>#5SwQwGDYs5@rQYfESX%1wo7z-mDi3-bz=1=aWHBZ(LHPgGw?^1dv@ z+0a>pnPL;Efh@->K~+`g|BXte=V}a*P!|7^@XaV?5?&vSU6e0!JJWqtw})b*9?kUQ z{nS3Pk&oQd!2XKwQc7DCuv+`P|MY;heu+AGCLP(rrfqs};D7a@CVm6*PR0kFq^Gq7 ze|e6NiYOqN+3cuP0bnBYIwWr=z6cTrKW-!*-~EL8p_;lV2!bH712*@)RFPR>(3E^Kuj+QltfW*;vi#S#|B@+v>Q0-}Iwv@k7fKeeuIBWX@b}(o!^LTe(o83u-ph3ZX>$E z8vb(QoP8Cn^w@gh{5*4$FQ%Pf259i18c13x;~Sw~g|Fv^I$`GeuMQJ3UgII{B>Dr{ zdxCe2T{OY*1x&t~7~G*a(ar>=lX~&Bf(xz5tC2@UI{p)Jmpg^IC&{5gw9y#&JDVXTul&KFZu z%Q{h&vFDh0c6;XJ(*Wy!A4=OeXRq3lzQQ0Ce9QZul3DO-avsHR4t9u`_x6?gA zD*!WVqzw9G-eAVZ#qy}67an|Z6TQ97ray6SrYKs0I|mZc<}MH7V2{kcY;IWkz^m{4 zpA;84!??u(G?epPP{ngM?)i`o02;nI%XN`op2OZk$+_ljW_JPi0q`YEVOmC*eC&4A zsoL%7&)*Q08zAr3LGW_(zr{_8N9^+gt6Q5Eb+JhCpHW0Tuzf{TI!)S&^RJh$9vJ7n&G061``tBDPJgW z@@gEnuko^bCR$M; zAKk`oE3wT6#ncMvo4IAWrjywvOz5`!+!L&_piFFnt|B2)FbuP})oTciGFI@7m&7Bi zj3-qWk*AFVazBP}KmFODI$M2Wyd9^VcmhfHl{}U7uXNEy?1Scm8;lZT=0BmVFobe{ zuv>haAN>nZv~tnwPnNjB0O@9Wj&YSU?auk8#r%60y$ob?9bcf_Ny~i#^eXUT9LyZ_ zDa9TUU+^RJ4bTwt$Xf5E>S>FzpOG=PG1v4(gAc-t$%(54nYAm2+zkJTlswIBIx-6z z>;zD!s0s9JH04D5i2bA`uOmx7$jwIhs0)-%=)~lO9dk`tv`H=2yWLyme@q1SKfHXf zDN7#(f6|26;w99JDN=v(C);^ldN`!;sF|+ndu;6e(gZzr4$&c2SgP8o_!4C%)PyfZ zY?V*wasrKHF7y4u7o@=UQTaMozxdmZ>2$rwTLx59pEgX@)ZHmi4t$4EqrqPAZL zO<6JVkY%c~2mX6)WYqmB>t1SO#`^K|6AGZrl^?{2{6I}!+N!8Vh(zhvSb?f^et+)2mh*&&*UF8LU=dCUW-_j6DWn(eZk-T2f8 zdRq`tEbDcwK52+A_V@vMOV5}XB%MZjcXpj%%fQ;x8w`)doTaQ5@vrJeUzppHW4Zp9 zdC%Bc>8}<%hgqi7nhzhU{D4~@jrpm127jW*OUM#X26Ya+?2J!+$SX0V{XG$~xB6j; z^=Yi<*h7{2NmXvq50+vNwkA_piUAMI;GqGVH9iQNX;?PQLYXLz{fs8xj}=VB6Q+H@ zjodON+W3|O#&s?3A`trv+j`4rF+d4XWGFc$7O#tYI3sxtWb2?;ouz&A;Y{3$Hro*c zoPJWmuBuhCVB|VkUX?8AsJo>ymspHDb(FB2a!ef8PtnzJj8ga~5dxfJW?k_a7D0HC z)sigp_Yemn!+a(s3UK}fWpn@QXE!wDhul`Ln($gYJ7Jv?@X{f%J7&(t>4E_1)P z;@6D_P%h71&$j8MfBIbhu&c`q?y}v;d&4b9Htj>xR(Auj(=T4 zG=!nckc4Ci-8V{-DCs6$hLH5rN~Tu3s8od3MJCmzLia|YblJME3r%#@Wvy1Lb=ke$ zw_pF9^LU(p&f{?&=kvaN-mmBDbIpk9Okjz{2*YBxi$H-de~*a^DTf4hRTcBJ*~voH z$SpE_o%~}JH88pRE=EfzTII;K@n_O1yBlfWi{0hf^v=7Wc`d!dwg$}ORS_!^% z;Fcth|Ac{K`xq*&^j+38Gs^i-4^4jHTW9;~ddLWW_$_QLYE&39Qvl(=uv~luUow(d z7VF^aZ<$~Dr$G%z1LoK8Uze6E4$f2n=bS)?Iu@F?O_zKZN9-c&S3m4Yh*HUWf+rIY zqt@#|myp(f%+gG)^{>?ED-zI8E4NN)&%T-wfSW;*2MErz+ zQp+03F@GVX&qULuEwt+Mi>J~1N6Z*ThmUTvm*jp58#VBacsJmdXZR9pt7oRq;CDdZ z$e)PJi2@&+v9f+AWbYd)`lW2w9_rl+`2A3fo?O%Bls}a|3%wQPLI^M!t9*h!KpB&g zyImxjKRoAsxB(x*>p;iF8MMh?p2x@RSn{by^zm&alwa_?urkai(Cv8)0ZX^#W$#jaN6}3 zkdvK&Z5WT6rkt$nOb}*Olu}P&Y+6W;V*$L4nFqt2+Uj%%nkd&wQx~iW8{NQhn);#~RCBB7J-A z&Cf*hUZK^4;zc z*BiO2?lYoxbjOaFCMQ*Hpwv<~ymAX2Ys( zIeV;ZBQN}c=Rp)Ug2!a`WGlcUF`v9P!Q~vvk@ctcX~}-yRSec)!hwBaSX`_Kkz|@!U@qFsyJ_v5M5u6)P4FCp4}(045Gxv_j`2jm3uSSBS7XK2F&E!}j(<-5*2 z_f2%UM=3daMm{3pSS$YVeMnbqj!B{@E6mV9#k;v@r1^1yPB9^o=G6dy1`FISdERyl zoXy}#w=auDoU3ONyk0>vqd%?l(nTd>2RZdcw7Q$aC`yfmVo63Mb$P(zv1V`+td=}A z>R=a)N=bhZgZZr8OEh?(q1R3_R%p(63ye`?w@okzVQZ4+GeMb{cWG+-9mDnxO>07m z;y1tcApJ6C$-EI%U!0q7rvsE9zTQ^d=dcisr^e+p*!}y0btJk1VOxmfD!Qh50KJuLF336r6TW zoO(LC+eZdDnYQZBi;qNz7>6l{7l?KaX?<`1gDMdsk_ba+Ca1zi&fVf zJ+xIOJTp3&e874}K~Je_YHU^qN`@r{Ob6V*JYagk4f7FGcCFa>=$6Z8^}XyULYE8g zwmP&o<4c+aow&P{>Gp7$w$z@vPz>-!oqSXg6G@Ms@!^e2+fO7>TIN;9nq zpJtXvO{lqqL6C=iL)Ia~84;n*wukYvR91Ipv|j!{SxE%M(gz zOnU}{nX_Q48c6r&Z0o+*=gw6#;jkA6}O9N%nL`H~G8w>${pW>1;D)H-u?uR|~KkWKCPVN!e+JJK?p7#~^3 zOj~XrE+J1!>4FHp+C;s(zIEPGlEfeXz&V`!hSmrxXX(~iPdOb{_bPOS&)xmg(4eTR zs;Pz1p83=B2SsvM@@w{*hvE}h$$G=8*BU{OoOOMq$^=^FO6Zcg&_K4gQ@^lgD+LqH zoVQi*O;BpbIpPMb;)xeMyM+gYK2r!o{CkHyKp`%c_r>oTTru#22O*xt*mfLeSRU*y zhq(3=Zxz?92|@I=K9hTW$H8`VddXamemx;|m}jDHGSodMXZ*eWnh zi!k!t#8|p*(t-c6gR(`nuGKgmyGLQtjEsj}gANEH!I9nx@{9fn6AKBTIEq`CQ+ihu z_1((kSpKVNfj-wM-rzv?r(4-T##1(OcNGJaYL@DJS9jd%E!4d_t2B$@zn&WQiI%_I zOYzUC0vUhS0o>8kEzs>%p4na;AibyVAebsyaI#KB?T{ij!95i@k=0|yb(f^!hE-*7 zPBVn)hJIcO%Ue1nyXo16K{if4_=hOkZo}H?B-Vk5fyZ(>yYIDEt7WE*9VTUlCWCj7mI3m`eERDNZ%=UpvTkIu z`H(7Kv6->nNBw}?B!D5KTt^(2p37d(_)7PESLZtQzc8`d9`0 zHgO7CIMr_#^`r;9bkQ$ln+`TUK`WnXvsFCj+Oe2945>@6}L zY)m^ZOCylXalbUfE=doS!sVrTbypa5Hw%i0b**JqXXu(J)u+C_N5tmoe`rtL{PvKF zLhmsN0o6Cg&p7Iz12+t$nk59R#^5>t{A`i$RJabghHDa7NqQQwdMuYQy-p)gTkceS z%CGUq3P5M*BXnL_(r1>y5|^&(!5A@51Z$-o%74Fos$Re?4p2I(pVlU)wQ?aNuMTQ z^uNi2_B!VJ>@fXPxCG-hgxB2M{D}82v>@ssVPs#7omm3$1L_WgumN~U%8`euFVX@{ zaQ^18O;sxaN3Dh?3wB|rYhMR~>m2*E*Lk^^fBm$K&2*}P-JfP`tFJW0nCeD{K1!E2 z2H)?JH#%tRzYRST?{L@LC(U4-zP_q!CVCTv#EVQG8v4CT@;rglUUvW=N({Xu9s6oH z5JM}Hc1}N3Zk*1v+{E~pRGx0>3xHjzzim>cAZpfs4mY1{zN22 zXQEvFP0{svZG*e!euv*3K*(bFF@K`DtA2|uu)A3N#DnB>%Gl6-rF7Qrc?7?4gGQcN z9=IdfWGCy^HX$_}h| zDa@EQqJVdLHPC|4GhF=sxN1x$4EV}vDSblMGqEw;BXDos&8}u_+d%X|t=kBOX#t|6fqwb4zOdjne|118IE|)sl zjKwPO{KP}ppEKDg+c}!^SS?%kC$1Ia=KqRsz!bY-Z?qbDfJ!mn_`%mvuI8J18Hc(* zVf^{nf;$9#rp7VrM?4LFhDSuOM5j1lCtJz0G_y*xzAZiUlWx^J=k|$)5Xs&PT}q&? zO8;owevW@T%wPYJUMk+Eozf>BP_7(#Gw_@O zU7x&G($VQeE1$&hH)P;Q@{byaNY~~LTU|zg9V56JDVj-Dg{rfBlL#w?F6w^^#E~H1 z0*PuL6!!_AE51NTa%rcOSicqSSS)Y}ws+ze-OU5Xx{zwtcbRS}ZzS!cMlILkMugYV zx|>8szt{$pcD(D@W!3tbTtnF#esob~KgM$`VG74bC&7xbW$dI3XrbhVEaBS7c4($( z^a<@EI9*R4e%J4gI?k!k@=D)xN0&-;#mg}863%l{5y9X4Eyuk6$P&AnT%zF*H4Y?coFjw7dMo zX=uEZZJsLgh5#ciewaS*(!W%9=Kw7ymK=J^Rlkzotk1G4lDk`L6Fbk+01JwS0pPyKQ4<`{U^m`=AS&{6nb)oqx9xZvHFI zjfOv7Ke-P($VGi7Rm{DR=tx<kG!5I|q_)znMt)drkTzF-6??BrX9J$_!&7@-RBkrHInm^7I^6UuZDeyD|DV z*#zUv>sgjJfnS>)%-$|~P$5czZnW~MUcIAWJG8mnGbc?{7X7aaHb*i4MK$cIkz2b> zb$A)g{B2*sh&)X1WFy*3Ofe759YUQdHQ!T`$*>xJqEpEGv2Dn53_n9u=zxx~oXcuY z

1I_U5&}3jf-?fjsdTKQumYad#rU%XtOJv&7xfL%keZslBVt#8Kf2 zSGr$^(HymQ2exkJdm*u6qQ{d{ro5p4M8-b6ER#y^vB4aW8h0k>7O$kZkF?V9@HZt;$1pU|!rhUt5JD+DP}B&|>TUphg#T zK^2@;iMEP9{^&Bp8+Bb#%ZFaawx#8aK)mWaFzZAQJP9%U{1!DI5DATCCz#-5H(^*~ z?Dr|1BG4(Gdx^mo{k2F@Q|mExHupih(=nyky3OmU)dBK7DZO(5l&J63JeA39l#6^OKPP)ofGH zOrY@W5gkp?U@UC6@*MJGlf`SImClYapKrh@WgN4}jpY5%P=wd~4VvcqhfbW*V(4!G zJ3JZI6hMuwW*B9wznpiH#k5_p^{SJ<7hx{d*2}7O3}LQ{g7f z%GKef$HR)FI!3yUDMyJgxf#L0+UXJ}Lg zG7oH*yMp(1id*JWM48th{0|WN=qWPy7D9IguqVYS$^1H+C+OQ%m z55)Tn?EA$9e}Ww-gZ&pZ8bt=GX5Tbf+{)%W(Ud*qz)Tdgf08kfK8>3D0`-vUdov%OIr;+WwqvHGtwiSD<0v79?90M$HhRL;}-!N zur3Zy0}$#^Z0p9QuJu2~APkve_t2V#OvfBuRV06~QoqY*T=fK7ZXPyu^tVnmjSVD0 z&L^2MYxjVM&pP}{sqNwxnAoZwCoVo~ezOd-s&`;ktG-Qi1{tbehL2}$YCzUBO==v{ z0%xz5?&<`rG>egZ<=!j#x0MCiQKJ|XH@kvfjBqmPPtnNZa^AMk;)d`Y&Y||Y{V-8` z8?~1e1elpHSZdWP6<89gc0;(^IkT;n^v+a_anGriAg)j0q7P!vL~WX}bhag~5SZ%c za|%D#Nc5y^mB!qmc`jjifF{Q05sm2i^+-`HOswSLpbldRazk(v|u$1hUO zYfVu9SeG~t_B&ON;7@?fGX5usR#7ZH=@OENE-ebU?yR>QTl=!cL7+bqu&Ctu#5DO3 zt9KI>9%Ebr`4{b4;8^FSxbT=_QveH(=%v2`S3r9yXH}c(jj#tIVaMuc6iX6TVVERC$b$UGW(gbmCjJ)tc$qMIrcj}`-(K&MNd z%WDZ)^3A$A@iU3hc3uwADxdz%kZ8TgDF<8IzaRG5d^`@X;p;f5C z5n={8tmhWVh^5kPbLDMSP20V%Y8)O@Z)d5jyYXljza?n(o7pu&ov1bFfC!pKCtLwy z2V5sw-(^R*Yh)t$%IWQ#EBY$8UJ{SiWS9NE8}d zH4G@D?A0^P{}bJpd`koK65&fobQA1~4T)gnyXHkf3O!HeQI^OJo1(a3yc~U&reY+n z9hI&*X=TmL40MuDU{!Qo_#80w1J0@t?t(ca+Zfs3PkC2!oUu|f!V8_mH2NBKynl$@ z%M66*TLwAm*Yj6CHP9UMiP}F0Sygj0#~o^@&Sl>+!OFISNWjI+3Hky2Y{a`hTVfVR zsomE+G|=?G)Ay{jOrq?CJBHmH1~2gbZqNwai8&%mTt{>ayU{%7K{>$R9!|_RN}+)% zRs4G9vd;gRBv%yU>KAR&mJcal22YK(l0KOivClQyZ1&r!zjuxhk_&ZrnSb+R%sa-AeZUj*v^JAe4`n0Lg$;&3Kujfi4oX4q!ZepRwqXQG=6*`oS zSmO5+x~zw~a%6SLY;ox523ZB|x?ZxpRFeX;8}*WZV+?{7gz}B79;dC%)x*X6immT4f<8dD$@FvdBU#`$I6D;F~g|}M%vSCSw zB~PrJP1hiQks0lHGYRs2u3D?Q_HQd1xk5IP&Lk=O1*?XDe`+_7o)8OFmGCI^B_*;X z#KTXM|BQ@JDk&jmP6V;jjg?;L@k_9kunz8Qv5mi<>`j$=3b2hFbo20BO$ySGP=MhS zM>t-OM_j=&lfE#m{sJnv$Ces`2IcL3Q<`2mcF@^HwN3q(`&)%43};&T4xZL_AZnQ; zrjhjOF!wKcMilW2I~#vl57bY0db~2R?i~jUsQn6(9=;J569Y?_H13!MapWcM5tu9)_!4ocA8V>4|_ zf|bDI`1TvC)*$nuLK!bu!|h{s&WDITVF8T8sq_PAJH(sWlfw9dOf&ZT_z1seN&I^= z!v}2Km+n{Wtbb)0I2#>g%+G%3#=7hWw~R$%1}AIQ;SSYg&D~+eG{1(vlhn!b=G(&G z`{aX#hlUc1yJ%YH^);!V)(EYMB0JpA(-oLM5&lnAuL0E?(J!NB?uGt#ON(JhJ8avI zr{pi`Ax7A%@{wN3Z^H7z*K++^`ZfRKPb!YpE$0oHRXf3R6cx2&KzyFqQtsUOxwfb4 z&)?j+?F?V!M28Q%027%DO4Qd7uVmQu5VUgn~YZrd=ruL_WtJ5PUA$_n6ugp0yu)IbY$5h08t zKNP&N=|l{0vjM;50g{o|-&7VelN}rSr2!JA?F`3*Jb89GnCZezIpfDJP=Lg5;#V74 zK}ISq^=Sy@Dd$?FxDr%puW;7;6#8}}-U-1khHE>!*-WkAp%#XfnYe$Y#*YFXuF^f$ z$H=r<5fY<1sybp>SvU24Feo^a7Azrv6^Nc^+Zo~mgAa^1FgX7l5^rFIck}~q;DwDx zmpRHqXUxVq`=jGISI4rYeckJXhkbW-saI%RL1q-2xx*?!g161pmz|fzTMB29%T}#I zZstaf^qMtOjvH_t+}OsCTl`~}GGwdQV=tITG?BE&azanN1lfZ6X-GKp_W}N^8Ml4p z5|rK4x1yt-qUDWF)9kx_vzj0eDWk>0op@m6u2kfqL9*P>*tss^qQs*U;10drXH-&d zt9&W4`SbS(C`-|PPwOEN^)B*cR6B;uasCQGUS%F)`Zt1|RDRzd8C8Lw%1zYonJ?Hz zoi0p^7tg~}ajCWIYeSANJv8K-^ARhg1m-}jRVFc!T%*UES$J-AdHMqExuUB#|Kxeh zucshPf_hE%!II$f*h_6!6Pp#9q@K_aqeK?-%?WH8OOO=G{s8j#>owb9;~n*kr#W$F zRyoKCb0tXP25p!+`?o|r&KMB?Z5!0Ic?R`J0X;I1LTGMY%Xyi<>bbGAXvu}4J^#{? zf&PdSWalzVfJ0k~HMiyj(%*}7ye2SS%2M`r{;G~m}v@0Q& z-?|Nkr+jz_TN=Kn*vKwO0G;8T=o1xDKD6z%*0AxK#Or5exe#uVxFZ!ogOemaN|7Fg zo267hVchDf2@teR-dy9Z=Y4yiFLqhBd-}&fT%>`yNt2_yGG7=d#iy4HMzC2YwQ5=@ znbH3tf3H_ve=#|1A9-((_$e?WpLGbcpOP-y(r6a%lu2fXA%Eez$j8+cveO&rsZqNb zh8ysgAGXZzMkZn>2~?8Yl!}M?D*Pmylj7DbdpYfj`T+M?yehG*F72dw($DrGo*#W$wO2uW zwBjMYIzvWn=NPJEWb}6EeZDkGSO(UvH-Lbflg*i$u)`L@I zeqx^VCrUW^_9bOuhvFi)dFD{L+&FBvY>g)_5s+erhjvl#Jk83BeCQoh{jafQOM~cA zZ8PgBPDB4rhe%RV;}%i7jj}$>zQe2MV_!_+(R&{!j_Jc~C3=_gP8dC->x2GM8X`Ug7YA-1j2=hsQW+v%CQ!`& zQ4t&AyD)G4o}j~wqO?TrO@BQ{E-DeHu*LALkXXrE;UMVTaH6@zPpY^KFT?<6-Gu#r{<0irLO_j2Q#?pRqu+!3cynU-eFKfLrx zLW>&Dy6IoHhw) zQVfft#1f2Z&&(A@1S+Rw? zV&YvtJkM0Lu{f@9Km1lXDZg`Z+^XTlj*zz^Hw|Dn7uX2c7V>PuF0Yoq(9WU^PbR+h zKTjUFUCehWVE?QjWYe}l)=B>i0-Lsgl>ngyqq9AEt%JX6t~2g4hwMPoaBz_vWEZ?4 zCF&346^2w8Bb?Y(_Vo{@mdQ_}z6gO&WIT(by@cG`s{RaCt>W%~IB-IBhbu4|hYHs? z2VRY!BqO7Nke3vD%;78-uJ1(RjEh;0WwD?=+`A$A@jBEO=JGTABq=&Bz>eRtm|M<9 zviH49H5z^u6&L&&)`)m%bQ72g%%sG!;s&-L4D~VLJMj~B#T4UZN3~y*sOo0C4;!9*uQJ67S3;NsiWnvU{1jwZ0Yzglpj07p^GqWAx^BU~X*Vqpjzdyu^WgKa; z-Hq!}vS=J8xEiLx33gEIMC}>I4V-Y66=vvnk!{W*pSNF@jTZYK$cZJMi;Np5ZGcxY zAM&AvI-aAI%%F}){sGk)e+#K@lhpWsRLoW7A=JKid!Q@iH>_s#p_d%Z<_iN-aPaUW z4Itb;s1~J!`9);7l`x2nR2u3u)qOQoVXnXL@->|)TZP=kyu9t*drFEIN#**d_Cr-+ zxqIZLV(qe-W;rB4zoF}^0wB{<-S!7d|8umO@5k*NSgW|AQAs!BTcPELKiN{SdjYF` z-=^vR6kuj>!f1E$#m6A|R#lKRPGV{TT}jE9H5TwpH_(qJK>jtrlZ{9@HAPG}PDCf1 zEB9C`njzO@4q)^AU#cGqpY!acdwWeUKvPFbFdC4*-C~%h-+SpH&|Y|~<0+XG%$_dCrXwYF*39v5Hq##2X|SFEIZn7dS9yJQWx#;LB; zPdQ+V7)$DL=JF8^htToXy{R)&+HYJo>hSD?;^q|W=MO88*Qx}g_u*(1Szn@+Y*k7<6cgv!h6fbO@xFsDwm*@xQH;UbCw5?{k$ zO6Hw{yKX)B`_b51S(_&80_OabeOCP{t*?xy>SgrOH}=oFcB}a)pCH9oDjFCurqHPG+-4vYUV7apzVG?P7bo4A`(jQ-Fz ze4!1$hVECdrD=D2w?SqZg5%f|q>Ur-6XxeT{QQ%3VC1dHCv@)tAN_y9{w{cmEn>_O z$iJtNhAmL^nGD)c233JnIo;+{b9=h)rR;_KYP83xCR7HAPbo-@s!0Jgg>tqI&9=Dp zl5RZYc4hh%Dc;$SB|GmvLy^!X@I!E>k&YI1YOpzDW^Z&FC0}lpo)`{7?qSC5|Mk4& zFzEIHqya8$)Q7>Tz~3p*G^2&KyenQz;e5o_PY{C8^E+GZzeN8{y{Vd*wwIQuS7R9k z?A4cvIx0c+F7;?Y?xCm?QgGO}f)}@tO0299GO+{^rvGuBr>R>~}P= zxR`2sGV|t*>48xec_4ZrZGa}0*I@<*)V+P%od*Ls-@GFG1tS^VUDu755WCuP&{aI| zq`7TyW4MoMeOR+}&`Y8rg4-Ew_5agAgyR2+3x^jwm-{U91x1Jg^r(1&9xOMGoWi}3 zKalfb;@KewJ=(07kOxiaF8Jri6k!x1<)0;kyTOy{A$&Z{LuM%N3SH2XpQzrYTx_>% zdQk$>LtB@F_?H@q|54IVnQXT%jh)E%fjpyR>T(jQ{!fx}k&^4|NK7@CDlBBx1Kvtx-JYUoDE>in`DI z)tmE~W?5TV4MFanJEb_uZ{_E7i4Ty_+^c!i!lSWu1N^!o{NUT?FtaW7pg9sSrexH( z)q4oATk*~5gY2cK#6qjuPdZ)CX**jK5~lo<+hIN(o+Vt<=+xKt$r9@y!9-XjOVdl^ ztdq>MPgs81l}3ivFLZa4E&3lkI3gU zSs(R_oePyIVWVy7-#*Dc)tv0#-n;7rJX#0Yx$J|@1?Gp8bQf6masmB|w@2oqi?)z9 zJM@B>=51x%CWJZ-<)1+LA$`kK@OKt_-OM>k0s1-{i;hNC^?8K)c|C3>oq|K-t&1Y_ z@hzK7p^V@+pkea(yhXna^1m>G$kbx-a<}+3?3yCM`j6CEBCb*Zc*l?pots{QRM7*^ z=PBX>`73LGk;7l*C*Coh%x*9}QzG(Xg2(n!R&rc}kQ2{-J<||B1^!3={J)b@FF`f? zZY=S*3sBnyEuI0k1fHuq zLeHg@@ivGJ0H3YH+0RA@{5M%<8|&U+;sq+#{^p{h6xlcwzFN<_qZnAN!ArE&gsyFb%ttRp5;@8wV)kgU)%hc4s zbHoM+uDpm@-?A{|Rkwcbsp1X-#I;%K7!p$sE$i9YM0$sN4|x-%#n-@`LmzGJp1Tf@-V$ ziMavCPKgxJnWQr1u~ucF*LJxHi-vo^w<8a=fM4a5TQ{Op0#7EFl3Jwm1qT5BK_L83 z4Z)DqLxI!76aDMH%C>y;Pj&*``3?}=Mu$2ui^>8@s`i0V-0#IsD`0nKf!Tp4=cGC> z3uL$F6ZJ~cS&iy-^+EEvA_I-Z7b?R6uchMyaTq6N;IH6ZJMB6-yQQ7F2~;F|#LxU% zM&PeFH7$@O9GKi*9VM&~y7gKds{d&{b8iATc2?I^wGIJer9ZYG&HdfsMt4y{6}kWO z?4fLyRrpp7>=NylEtFBbVd-ips8iZdTT;uAdBHnaI}vE5CGYCbXyr*MM~NvllM2DN z39+@M*)$HEig`lYfZo8{0Q)pofK=QX^`Wl><&vGoyLuu3dM1yu`ob%d4YgqUn+M|M zeyYu?6jXnU&jhn61dg-uz!o>6*YfU1F{`e&-ihu_GML4=vfADSOWWF!V55#a{SE_PqY(MKu$0jiY5mtk!o|3C&~ru=kW7aJ$u#aEo;>fX5jNV)#+qYh5& zD)W*kT6)@Qvf9vbc-rZmsXwt{jr{NJfa_LW<~yM0(5olbh(U8Y6B~P#>qmQF; z0V2LzE7!FEzF9pp4LLyE=d#1{z6||Q{$g%1(+aXHaPoH{={IdsCW7!YK3AC*8N+^w zvb%YA(`b5*s?v+ft=ZmO!*noyz>AzjFl`v}Blg4R=?jC&g8$J@DV1D8sk)U!2`^DPNIcFg@r~YKRGU zb=USV7x@+0liy|S<&&tH(SM{o>{V|=k-<;61!3ahPhuyl|NFHq8oM1j6aO8LVaYwB z^riQPEu|{Y@f%tfc$w3h0?&!m-6}oB7lupO1;1NRh3EWdL6to45a^R+rIgkT->1r; z_sfGRyZnPAC-Ny<)VUGIBu_kMh%4#)314eM-{Ij3#Mo_2m&T~--vBGqDNsFEV~Z$y z4bGc;OJW&hfQRdfsv|`)m53y=|3Z<7sW>3J2s4g+To{CrMu2Z|P;7;U(1g`&=U#QYO16JMli*(;bN|is zY+q}7U`@qQMDSAl|2FeZMwXTS6zpInhXeH5yN-xqIm%W`MC?*SWJTp8Wo|O*6R&ewSD#JuKf`>4W^Z zthjx)eTXuvGY^SXE*P^Ffq$tM8k8Q(?n{emYLGe?=^@_`r@@ww>rW>p2AzET_@H_D zs;f4Rf=AxK$2)Ac$+LrvkKvvZ4G3n*k@{GBn^xGTunf%opK^KuDx95Ud9}b|M zsJ!SRb{r#K4S2Xh<(X;uA?o(r6M9~E0L0F}oz0&S4yp?YLmj?E)Y88U@6x|FgT`#@ zuFm{=SU0}@L-Kow`>X{UFmt}cNgHwgk?um=I>MX4SxyYt&8C0eqb+e6nGMFdL1Q_` zWVXLyN#tmL#{_b8KnGmG8NwNt@2WlG+(S>KRBsjpoK*FVm3Glhv7U}|>$;=m&AnCk zv!9RY%J#Pv@w=w2LP7EmDparQFJO^_2Gkybl@%|6RV2T&n1U7XHP-uM;#R@>cG~%^ zBlP{7>OIt(wLiIvHdIro$m3Vs)@TseV&6#93!T1keV6xiUXyKj&!h*-)rDUHA2HU{ zK**hd8ABkaP=r0MTp2xut1Uo`|S zO?JNnobZ)LR~f>%SO4Jx{9q$z?i`rQfc%(=K8RD#+VN=+h6LB%4jCz6J;#?*O)o z?|`C(1;n_VD63xsYv_hDK_g%d^qlfSWPoZN{8Bi@NO4K3Gww#M)Mvu~U~5-kS15f2 z6GQjsf#2NWC!pJ{_JUgLXCC~)-mcq*^5rrIy|!6f$Zr$?Kk7F|4&%KK^L&{xk8?0B zk$S(>%s&2EUs_By(Dw=2=^qH0WJmMOIU8xtUU;da@Q}v+G?k20T%Ogb-#ewTo^Uks zW~|XW%zMSX$c}OEaNVgYel@>ln#Qjm8W{lUiW7cxgz4cM&PJjl&-4Zd9qAMxlX$mO zR-|uLeDOK`5btRVltJHvUBtf~=_ak_&2o2Q7ia2t6NG8R^(`{12H!EA_1Q77x0_)z z5fX`cHu8LY34!^kX5ttT z&<4>};#KM>DmHR_Jo-GTH|nMG`p|y<7G{Yc@p-Uw=T)ov4U-A#e82T*XQ^b6vp^+b z-twnA9auFbRi6}QZCjGDsx^~X@RAm}ejWm$z?~neP+Muzi{;^V?49Gcgxd2y+rfFt zm&)6Q^ZGU2jnr?30|wCAI%Z(SQRdXIYdC&X#Bq%2I6Vdm&iaLJ!0!}{*S#h-ng3S! zaW5!NF3?hGol(wHZ;pJaYNV~h{DcU9hpnY(*W?RL?*Km^ZPDN&m3j&+|Ho3$3r%fG zFr|749Q~GeZ-sbW^2eGLVQF>ly}`^w>-Fp-cjcGc+MLJAh{9JxfRd7C^*Ty!h6ixFgRsnA&*SS;7YSy-LHkbnr)18bjmlK!6%g=|y`%!j$&X z&H1q1))^>d+j0J}wf=g9k~VLX=ja~&E@&l8$MkcSJ4} zhriX}C;Vzr4LJbRkGYrH@pfn@vZ1uD6*%BxL~`-34E(pUL-D-sm?WvwhY2W>-wB5u zDhK(?{+u{e-Vy0awg@{?t{!*lg}1?f6RK(L%?t&8z)@XSvs=E|a?{ZDpEyr4PXEFk z*EUK5=tPo~cUld70t0}a)m08TmkKmmhGU>3Hlt9c-D@y(c?rabjDK0RQ;X4^P?qJ( z;CZ*5G`W|yx*k`bnWv>8(1Cg@oG^=}%&3c!=2vE4)N$`5m%azue5=$VEf0HM4rF_^ zISn0YV>wIA7}9NlH$d*_Q;OyYiAu92*cz+`5M1U{qt^Ciy_0YmtF)lP2M|iOz{XzP zHY#DTZ6ypGPUX)}CD+pWM@s=It-I5wxZ7CiS4ntJ*v* z(@eouI@_h#ZMWdomj3mRo%NDXPW;d8A6EuQi*NeRmR9zTHFf7Cjb^Ch1zlzz^#Fcr zmJe8oxmf!fZl7~?Bx2^b-z5sY;-|LkaVs>582mHKP+%1q`5fOzJ^1Ukeo@sE2zudy z!c^MASd|7k&A!WWHL}N#k(1h9j8m5OU@=$IwVq50_Dp`ZnwuYWr?*3n2)4FCr?+3s zN(HP2Y>{qel~M5;C_wTHn@8l%@c)e-g=UEy{r3tX-M7&9u$d0W)#Xe%Ctvg7Hdg)S z=tk=*qTz5_4GYQHPYg8vH1IL`{Lf&?{Ww5`pMOW?J-p!W7n*4jGQ~!IiKj{(G*W6y za~NIS<%hX<7tQ{W*#8q+HRRu+rr__}R|Tx--_8PH;@0d0vOne;L9JSs!&*7r82%3P z#!@_ev(nHT5BO;5a}(5Ow7b(17zjy1N7^rHj+wB+m_>%^>ViH~bv^=S5tz&+wt- zpIDJ*JK(%3OZDe}0p&hp;R-#R8?jp~(zf(TXU$38;>#dDBL&h5_NMVk;3xk8#Sws_ z$r`@=w@&VYin#4g1JH5I?=2hj@{Z1fXoiwX@P)ZE>_Z*Yz>y>H$4k4H&Z-U|FEt|W zYNP?;%#0-GXQ@KLKkb%X<0i}q0OlgsupA`E%)KyOluf$rjNrfNd#rK*?(uo0?ttH2 zhELwRU9pKle|!}-{AzM*l{+2_-BrbY_|s`Rju(hVOR`)0zChwwoqC9#8P-T}P_5IyO!ZQiMEg9-D|9;NSHDsHU)^`n!e`oF zM$l)_N{pBM0sjPYtIU?~@nh^2@@QZWxN@r~NMgP+F>+s6LtAggEODLg@~QP(h*vZ= z5gvs~310ta=u8}1{2w@uB)+Lh2&o|?eKn**tqq|c(WEb-V+bLd)U?$ONzyq9Nyj8v z>6nnNwUutAp<@+oYaOet+STss*Z=T5&*%BPuNQPXCzNww?3Sca224&2uQ1<0dPNGa z$UMTl^rtDw>J#}4_=_l$zD0~}>*s-s#|oIqZpe``weKK6R{kUaIC>WAKD*d$=kVGH z#Ly1K*wm@1_Q(1vJK09c;jadIKU*hRPJF`4 zLSEj3#3G^$OUC@G$7)NBP&YD1Fx=jGtawjeGIEK0xDe-&n}V<-1R z{S@w=xP97Q{W9iwPcz#In#Vod;1PxFNKAlsKuA6W-NOj)K%H{JEudS+o05>`>=3ouxy!8n@+#t`oI^=L=Apjz0QFKrfvd~d zR!Jb|A23{{(V)ANGBhT7U(jan_2+#Vy@u%zv!KIbbah*uT_dY2MFy8gb8R4HXhPCrFQG2 z2Fx7wzo_pr9mPFib>|cdUI-5Vjh{P@wVhe&cD5~Jcr0cGG?3CYIwMEyp$SeRU-5t; z?L*LJ0;}Q`ZD2ukBIhUSshb#y@G}9)+?dUh1Ald4Tlw6%^MYk2H-qzo=R zCX@b&t_X4_0`ISlMNm6+#P2q2=1(VQSP&~YVwVJS{fa!4q&-m3`#yi;!XpYVpryhK zOh0R}GM5ghrE-x7+AaTH?X^SquqH_uvg19fpwaURqPuzF^H95w$OG*Wv>rdBHwpg8 z=rg$-JQdp2dk{~o!xoepIMhb}i#jY^E`>hz|3*&JSSlLWm*J59KkG;Zv`ab&4>No% zVv!;QLUSdcUY|*hTrmR-jOq*F+o;Dd|J~;sw{lyCGTOf}>*nkMj_@XiKhAyY zg=n7txpUjdKEQW%i+vND1Xb1!v_#~TFa+%2c3xbUsLRx>i@uCVa_kEOfGsF@rQVH< zrwBj#zZ?LsxqqzWb8aeaE28UkevWcKKPmCOE*n|@#QhiAek7a2-AYS>u8}-aoCxAh z5wIp6dr|RyKc4x6>Yu8;EJ`BwpfpA7hS*!DKjmZdd3TfJcVoXPn)v=S6ZeQ4$th&G zX-|S~rSiSy+DSufE2q=jWgVY7OJ_@9JLc9xH`5kt!+p702jnQ}i7m+wy4dSV{%PE^ z^h%*@*yy`kXI3+?!Ej`)49FMxva)|bqMiBIJR!CYU*LPyW~|KUP4z%NuDp+&PQ4Y(xC7T2^_ zX!DUX(KmBW(Z17*V+%rd0d6t8<_mW(P1kHv^{|nAH$%uQ?d1OkDlw~NDM;u;CK(Hz z5}hSqCo1jL|J3MzntjF>Jkp=S6R^NEZ=rtORybu1YV6iU?msx7YU|-@Pnz`QyfSnl z+UqM{r3b+tjM=AOpdEuArbemXEsnT_TE zBTu!eATo@4B+plV6M3O=`g+PekCxH`=t&R9`D?#iOg$`QM*N*^k_=dVj`G(xonb`* z)j|r2+T5RcLzCnh;;jLy>RPurJ1jFYK)k&F^^=C(mmd6hWD8qewM{<-ph*i(4NfDO<`U@x|=H7A}_8k|m38sc{I zw?=*3c+;xvtCM7AyUdA`HM~L*+E)D+foyH+j5LOFmsNNMpA))CPt`b#t>Uf9;Od`J zdQK=3;~Y{(|8!~0^BHKN%v%lU%RQs;j1t(!qV#thSQrE8SOq>Z!9m%)96HCdp18!>0z zz6ReHD%6Ms{IGqx3wxQI%2F-|@T@w6G2)hmE9FK6KcdQwQyD*|BTWPSZ6(O#lQh?i zlwU_l ziG+v_?rZKk!qp%!JPqUIHXLVg#cD9IKlDMezQmRhP!m2$KtAMUU=Yx@n~{^5su|An zmhC22FbwtP4GYD@bU$`%%kE*LZ;@y4e<8t=e(7$=sX35lt?BpSNGJ3Lt(Nb= zN&h^EI2Q_6`@W{Fh5=Dgs@p$GhqKHj29!+oL9IhR09z}SuLp{|eGvtS!L>iq=;^Y)`h1f+7P`dmV97fj`G+Tvslkg945aS(cBZcO|){rWyGJzAZxrh z^9$hsEnC@)I}>DCf5HmFE1jR1k~`hMtkcEZ5C;|l&2rBpgM~*jKlVy0iNy3g_mwP1Up`X`8#ZF3m%i1ptM2MiMo`JPI!f3m7_JGF(U}eyH ziG^fug@RitxYwXM(C^rEV9@B8K9XwAD1>UfWFdD?&9y_g!qA^DmV=YD{a%qk-gGkI zT&M;3KjCorb8%6rKFDo(F#W8MEaZCv5+Q4p#@%CGh)de-tS@%?Z>_Dd{R`vOpI^0) zd%qv#dRZzn0`fd#JnU{FrOlv&VfE=CVC1 zEBx*@brH8DeHs2R96z2RG@ftY+Jj2sHvvC}zk!;g$twF)uQ+1{J{7qnzoPC)GC76% zu7JF)y~JU=bwlIqxGnzqrs|lr3JG4QFKE>d5z$V%O4=|6s3@Bfd}c0lcw1*2@kzZB zNjgs-glUH{&N{BkwErM5cO9iG!?vBc2WEwFYB|7Z9ZyK2Y7G3awqroam$U_|EU=N-a28>|rh#@Os7w*a$5dcx1mf_+1aTt`}FMPmxIK`VPF%vqbZ?$C%w3?Z@}% zFn!b8Zh`4cbIPe27&qEjqFObXh-Gl{6buHVHEX?3?)J@F2QM zP|VMK>(eW-M#k-3$Rk4vkJH5%#iRSjr90j2?O<=+>@%XD;jOXNZGVKmCH3`YiVf%l z*&(1I&RI0ZX^E#z1_e(X%|DrMZtXZ$*|u#)P~NwDw7Q}^Z-BT276RO-^PbxCp!uoS zLH+rl;i-W4`Ma2nR0GkIwApfl9-p4Dznjnws-3sEW85bEEqvHiP>jq zeS6#rWjEVpAE9nPZjbi+WPdkl8F9hq65bRK4Md+*63gKqN=xS81E%RKg0`t?miR*4 zJD(;mHmxM_6{r-N9)t#^WlD3dXs5I4>3?9y^)Fq2H}3t}e%;!?-b{OcGP22bT<}Sk ze>|ctF2%WLPw+OtGx{a{cAsS2Wl>=@NnQY5i+w>wu-<9~Z1D$+dz$p;(YE5w7DHbC zeuq%Zuk3W)4dEA#M|X(3zQuE*TIiwbmc~xR?wJeQ^m6+c0$9b^Lcb$teY6!0U5w5E z4XNJ7U^vSjf)3*rA-@D6S7o<=I|a85T{k>hTXG(PfS~!oo2*1hjT^b z$o{X~#S6F-JJB{Y`Db4L5b;DeS!|#D=Ue&{PZt7E)`&W3y^1H|Q2s)jSv$fbS<$S_ z(C!6IN;+_;L@>qjvd%@XN1jkRqgbeZ-B!^8yxxe#u4{>#0x$;N&zaD(=}_@jp<#!& zS)Et?JM1l$qw?^Defi>|PTJWa^Vr)ufAC!302%A0lJG03qUv}r6G|2Ka~$a@qowAG z`h84X#0%`bacv>uwwGpw()zvg1b4=KnvK-%hYtPq=Shc&JHuTyFESL(V+RC*>D5qN z`)h@slhNtUBOGBpNAIZkqaFxeYP~4fj)9%0&*k&37Zf$KQ~mH7oAY=AeJU9KY-lo2 zCO?ZSgJH8&_ANI#N5z->Ea5)<#L1Ge=-7A2t7Dh>!5Pqa!a3{LD!T;u5bnk(;{MlD z*MS>}x{9TMf5Gnua%`-dh(bs|Hx06g@mVbyv4a$Z5x#+Zq16!UyuWZxJ9BaVB%~tD zZypU`>)1hOLMy_Kqq=@2Am(um6@~I3_&R3K$Y@pGV|>FF%rBv9R`4=1W!D(Ubdxp+ zO}bV3(;vSV^F6{coR_Ud|`9gyI>z} zxWcRLn_hG2yJR=yP+oU>NFU%o*ozv($q&uF5k`ultSL{oef@ECqo&@Kb5+^xFiLzK zps70YZooYcPFEhDcPu85&S;RfjDy9vq)>pzc$Z($d!fF`_$!q45)6>O@*vcG<01OP z<6JZ&{~7I@zrko0@j2SfjopN?I-_yTm$+IhD)F%3iA$Z9fj=?CZSMIj2%O_ zxdV4D7;sI`$sTjiJfnKQhErC?sN9J^__ev6Hr>%EzgQNxw){cQK&(U`z*R@X-NXM! zhLVeL$NuceH?N`ec>c`T1}Vp^2%qvEu+sLLkyhHwRpJ4gP{=Iq+rT#Qwb#;hlni@(wG}iJT?YQpu z&+V9eNiU$x=kiouqhq)xr<{(>?cnkTg*46aG#b_efR%5GLera>-Nrg;B#Jyb8#a@e zcNFTmi7cbt-=*odeN%{WiWwW9_DuXIFO#=;+=&e>X*j;gllV2pKZAQQle>b5nW-i% z5ZSN0B*4|n=Z!xG#2KibMr`Pj{E0`)-P+GG*5+-{j{g*G@Cv@}fO{@lr+%z;8g3r0 zR_Io8?u(Ig_2Tf@P3X$mUK8uBstq3sT_LVZ(Z0K&+_jijG4BS*3zLC0<=#UJqZCgl zE0v~_g9Tj5MIN*>@;S5vJ>$ItG)~*;(Fn&V3>x& z%_Quck9i-$H7cM#oV%%0UXIt<=yyW?%MVnpYl*xJmrFfFUE*OUQt5$o4N~Gt87d8*P)FRjeariXoB*sPv5D&qVP-kjh`Tk1{Q~LKDby3CN8wq!BC%s zDe8fXJO+ME1g-V@AL|6Wqe(hpQ_Gxq;yvMb^Y`ax0I&*le+(t$CKG*GUn8t>OKQ5< z1|-D)D5(;tWrCHz%NktYZ(U!W&j~fGs!R5{wPw$p_}Cz+gmoDkNcH6(39^6V-Gy?N zY<^tvV_MN^QB>!u>}gew6+jPqd`BC$Tu?nMEpfG5T&v<`An~zfpn)Ry0El z6f0VZ7g;acztrMC?q90Edh0uABlg)qsPjqnE-aktqyCUf7jHXj`z@z49L~} zW&_y$7(}GiyKJwmDuXJ7IOG|q3b#gTr!|gR1P<;tKqZ%}{~&(en84(`*RCIhcX&6t zEnpjjaPcpveGnHyNkcW<7&7}iY2ms#5fgrLTt)t$`pYH|`r#6IQw<6#USQyD(AM@z z`^P?3l<03V%R{hYknu;tulAqJ6xJi%dF{tv(^OvvnWK92lTj}hs-VRJ9!3rz&BV$1n>7Dg#Y@~6UxjGpvfha+n7QiCKy35ZARt1xW;Up|$(7rjY*QD!yiS3I`K zQ>$OeS;y&QhnyxPxmHEZKDnB|o>mMW&~DdO^fi^yy8P!TrXHrmQjO2#bfa*)%Y*a{Yw}10Em&o zCbfA!F=Rw=DzA%r7mXBw`VBhZmxA@{@e$YSST)oMn-x|8;*$mbxRsLCnlRz93wrlQ z^69P>`QgGJvM@u)V(WE`B-|yegUGHfgNMIKnxOa5O=6D`uip7LLTGz4XfWoP8@$2~ zSvpa_k66@TLRqf;FIV~99uP>b=)fJ^B zDBvSp%k@Oq$B>-2A*yIqS>l@j?_g347ze6ZFG+im&_REbvt3Z&R;+?@@^^urX1xIg zH7MK0Ij&ZI8@P7k2l_HOkbYV({Kh5&s_INz#2zh2vR0HH(SzSb4-R@(vqC%c6M)^A zJFGkk?jJgX0nG=u7x|o$ZUGr7tvnhoL^+r{=aLvVb-pJbg5Y@>Kz;(oE7R+iT zw)?i%Jf_5Eg5e8%=u#eXd}ibh7=Aa#*TuuOKgEQphjr3U3o2NRZiig9)vr{`?hOO2 zq?=U_C+xp~&U$>}J!{Ss1aKYcwy=X!`T0%CpG9-8n!wUJ=n3>}Vem@WjTvb5C;UvC zWEK(fQa4i1FVguIRD!Mr{}LL3l!pX%mf% z{}iwnHoW7@cFD)tqxt^sZc#0;ys?j2k-vgsJTa^8v#8s|g%Q>f{-Ag2V)QqR-Hi6F zFf*`Kdh&tM^O*bWo>$xq{Nn1@2%iQs?ab{RXt|?t>`wSG_O@GvnMlUxf2B%^tA17k z&5>9?)Sf4D|7YzkEVgX+T1ah z;WH@z#*P{7B`+M2Fh6UU<{xCljD8(G1afrK9d_d^@EP5VTJW>^wGs45%!Mfdu15jv zRt??0`Uxj>DFWW7= zG7dFuNA=PZrH7@_{b5dYE4NxY`0q8Wo#20H zbKV$wj`#Ed?Hp?dBrz1CDsJoz^%?>m^ccMz)jN3^MZV<{xt9#h?3-gl)(tElc}yw{D3qfmy= zRs2DgJv6Zj)FLa%K%>w|JGm z^RnuwKQ) zx^7L(d4&mYZ<575eW`W3pe0(6$GiT-K1<-rThjbQkoBGh)On9Bl5DR+M6w8(p|$Zf zi-QhI3%%V}lW#u;#`;xS*Tb!egQVdrHqWFAz3eP_5IPzz)5a{&vVL=3JPG-EGgY@l zCU_$MO9SM}s<I0(j1j!D`9@KAFDxZ(9BP$@<1Nuj?9L${2Wmd~)Q7l~V zCpy;>v_K7R*&V}b0xi|?@oqh|SdG3ftUQ{j?}%6e?V?Iya;Lqm+1I#5W2e=AdqJ^Z zbLXqt<+TAoxiZwd&^Ss8?nG4+Cc%Dfw){8Ri}P{o90t6lD9tbV%pW6Yj0I|Y)YDzY zVhH$gTdiJ2 z(={E5=*4dz%YNhztUqZw{wqz69s(Mn=|0FqrHUnW;peE6itUkm(#MOp( zGGom~#eP1053un^N7!wg$BYI4-@9z_OaVC)wi%FY`kLzmy~V7Tub4}u9#t3F`>{Kh zLg7PMu1hKZNx{5s(rLlgp1q)-2X^PRPr^SB{OWWqQO|IdCF8M8l~vH=x;JTj8|$gL z0^KR5PlG7~J`tI25Z-WSKaQ!0=v4Lr_Ji^Sj&8N&eB&B`;;$q&$>0BrLx&sn5ewQA zxdR%2&lPQKLBrau9|n}=H_F3D6$(j@BY(Nl_Aj^G4u+9Il-U8UJ?qo!ozWZbF1-_0 zq&Tj2*u~hZQ?w6Q9j;b`#%&wVI^()9@HnL_l+QjXQ?uey+}F`^+>UZ%60D{Ulz#M* z8kKzJ>E}G*M&Onpz!(?z#HxT;q&+quS<>B}P#K^qvRI>MJkD&_1&DB5=T`k});h*z zeW(75zUt2ZfQ=XCEpK~|+_cf8g!3A<%Zh%rN@d}XM>K~bhz-zdaQG{XdUZ>Y8rJ{~ z#UHG)86ofrE63({@5ec5!|?E-*KUokN;;M0ku7!Tc%(zm+ZOPj6s68!W<7XNyf={* ze1#g0$xq?C7E&aE?U-fqEj~r}^c~}v;np^{RMxPCks_N=UWVQvAWh&)ZZ{z31q{Ayx^-N;@pOc z%v#k}VPm17RPDw`H0Nw3#&Rrb(|xhQOt`J=&VL&PXi5*w5s%Mi;s<;+EH(TlJYXU@ zN~ao*TtriE{o(1%U9pix3g2P992i?7P4CaFGYo}?;)C^Ls3DenI~0WrkMo+K2UdeUSdSP+dE%O2)Y-%BFJ1qQIyZpe zWP@`s)0-9NfByPM8z6c*9@>xnp4AT8jJE)rWIe040fF>yHR((|(wIq_`+Yt~wE^{Z zIKKrO@_K$x=xJho64$We3^fqF>!CmSxZF?M^mEG*f_Jbxdd-A^$w+&p@V02vLvj*> zx4|&KHfF=vdxuwU&ultvZs&@@=Nv1HsK=EZpJor}){Z+(u%7#`z&di{Vs4}pF{cag zZPII?eT2Tfu}Rzt>J&tT5w?>-N#_9Ipt#)Tj{R{in%m zNkzI7>;(UZ?TayyY%|+?Ru-l(CZgWh1$hVO#6z-;nkyERNpLKc9eRnCs?wR?w6YWa7QRZc@43Gv ze_2)rwg*GYlpeG$igA|)ijC4~e5%Y&dIf!S)WC!qgEIfLx5VLMpJc~p>ls$OOK|Wr z_U;jv)|67>7U1QnmlxU}x*WxPsq-iZS!Q~ga!@$*Q@;)PHU>99-?|2;ALl4(1_TZj)a~`69GD>&&4{eR_dAp}S~xP;tXkFago5 zlWmbY^;fUR&u#8iUdO)!2NpGr8Ere3ET?6A)5T2>Edf@`XF8$M#4C(In$&2B z*5Y-l37+Ba9c-%z-Q>khWZ3G?>rl!lsbmR3zf3Zp8O16oYnB{_`Y}!UJ9&0X2|wSa z+Mu+&7cHGclvN8J-UvACF-VoQ$veE__{(EcMk0YLJxuNn%e;qdGhnwk#3Gjm= zy(pi%6;fm;s1ZKy1{!JOP91i^_0)~*n@IYIf9GPWU!*JU3!or^TdBJwNtJ|6N2s#Q z%1*ar*L?6%q~_gK-PwDSmacCaLtoj}ug0@4iSGYRjB(^VNWXPz<^#q^8aV{|S-3*C zOmdCA#>% zN-2=+MsatXW+>N6R|zlk(uU|cpoiM9c1U?T9vkh)GSoecc_>qMAa3GoB~9H zVR{ec2W3a}k4w2FS;gjkj`5lmm~wGf0OE}N3@-IMsMsLM$T_@;=^_UlV=Z^BVB5WA zIiU8pOjOdwLo*nNrv&vDerAo6%CW>|DNq#)hKZvlt@tUY^>64EG3H}epraXm7CQ(& z3l{S-bZ|E?0ASK^h6w-zB~=kni3Zs1bvL=_M@jYH!&Gs=xLMW?V$}K?U1+gQd6SlK zoR702ebg;9t-frgYt#0#ZeY9WE=&LAfdG;C*bKeqiF6y{_v==*N9Y3rDgBqyJUtd8 z!ZXUAdwm(Z(Kb@^NHh)E-N5?o-Ku$7?JU0A$7r$d(*LgCOixjrbx!0L1-~fo&_l21 z2gncPpcm7tv02gsGhO3S?@&hwgXFquHV7Fz(}Hc4yU~sYq3-aOx^S@f#+uc$9)ngms+{-e-_e0$l3Wwq7 ziP|?|uV~L9M7Fv6pBjRCVJkqpLAFenhB1)&^K52W->4zBlCchVa%1GhcjxgVE*XiC zE4r(HyF;RLpAK-{!di8|r9T;CsdY@Qyj=g03BB7`H$;To>n z7&?%}j{3TbcYuRE0lj8@novRu#*ibos6NhJgAnO_dhwsKGr86GO(apA@vBT(ZWpFd z_gis*dmvpMlUT>R)&e4rc~vArD?8p@4GQ3XqV`7yUWEA}DD$?O%(ZA!A#6avk=PFc zZAprX@2!e>;nW2=4+Wt74C%JuxuhLBC4CdmPOQ(OewDmN4G_6HkOBVdbx}jQXQd@M z9Tsc}Bx$$zpDi)}jX$q<YUPHyg0V5swq}y_L7?fPVnH=4{=UVSCUD zYu$Tn^2pdL8vI?v2P|yTf?qubk01L0Z-@Mo9*qAOYTP5)+8&UhtO3zCSl8!IpnBXF z6j$47d|O>WB3E$$RZ~4EAUOgmcFhE|&*)F$KcI$~@P&UwA2mi3EOgJ!6K^WvuTkCo zf#Jwt`+}v@eO6P2PyLVorJ|~4tYmhm^(W414DMFEFw--2zBz%Z{)m7&sHT3bw{o+( z^|Ss0^M`&zmX<79qW|Ns&YnPiuwDXr979D-$eei!)^%&^2YD)U-FGK;Mj(lHiu_qU z-Pf~T-e+pWJ*@wTd8-P^<3BWh_1MD(Z=3I@8*Q{=WenFdY37LL_OQQwkct(8k!Eae zvPS`8E9#ZY9%~=zsHF``$LXQHe!0nuwrkQd=9+S`u&t7i>?xo=hTIgclFqt1V2?LI z!+DHX%LRPhVBTu==!D){aC^*1;W^K`1~5mMfs~s5=qzrFqMRz38dv2J#SyOoz3`Q> zlvM)8nf$PY1Se|{u@PG&Tr*D3z+)C~8aopR#6tVE*>w9#!79W!kmC|(PWH$eh7qf| zF0gBw3C%jjI3=0egsdWWQlX_NQx0y%{cdhcgwv!AXizJy$->L4DtC%FODCIrTH0AB zL;~fK!1kkd;eRPY=uqc->T0t4oaIQpzKwA5RKE9pGqsA99U9hbxrfyk<)OcPGKT3g z6S665!DY^jc&4cQ4to_%=Y+no|2@^DZ-c$8wpE-d&gn!jl`fj8i0;CLiKC07BaHCV zq|@}jm>Q%hH$bqSpt@M2&b1zMmsh4^c2`K3*Xc_3J54YV_{*iibI-xC^|Xs^0fIWPP~3YOlAfn_e89acBK1z*YOt z)PPEcENQ<7K=MMixFhzbRK?(SmIZtv9?O?4Sl(|Ub}8+|jYR?y<&l_>B9A6kL)KDY zKU7PFEume7pvA=SkdMo( zQ7swP@MsFiuJay40ds!R_=+dQQHK#d8MVrOD z;xo&FR?baqQct+*j-Zr;OElH7@3uC?Ebwj}+%3(`O!}6$Y`$psV^%!mm~$MLW-pCY z{cff&BH03|$P*E(d$_-~U?6Xmhp~_XzoPqtaC%JP1oy=XEz18~rN?;}vEJc!=>F8F z0^MxVOokZz?kwen>OR{TuRW}3a%VVOW7CwGNqZi>Q}~hn`?GJ0S)czz`?S`SMXI>y zwh1~}j?Nr=trr|fK!yIR8CE|7JrV@y-u))Ez=hX7fChDIX(XHj{^Il%#tvrp;v=dh3@SsL96EQ=pc@$d) z{7i-?c)VobVZ1Tnj_>CEgD+}$!2?!qdz6<0T_`3TsmfUlJ}^f(lNTRj#9d4>uWKB> zpqjiD(wNo}_IdNzIqSUCQjGA0+;%%LpuOuN%0bw34s=QmRd@5lRug_SL(-`yu(ceN zhgF%2%Bn1~)5W(T3w;zCa0fcm8`o{2xfQWYe6z!oSIU2T-$Hp@7DC9OUk%EHrC~GN z!qlX`A=5T0o6%*ki`mLML$x9WsMbVlyO>cv@ww zPQxS%uN(Q?yAk6d9(KXeB}4!;Pg4+)q|R2X5`76uyh0v6W;^BpDp9{y{3Bps;R{TY zD=h)MXg69~NWXt(w6Aqb+i0uagMP|;(z79tvP*m9PADhZ^cp;7alO`+o~^6Cm9(%8 z6Cls;bJRVT9q5;b2K!}2vr_8}&^IKe4WtZ<1BqzrGUPH%=M(h?;@8<}(j&zCeQDfe zcmOd;E4accgusWAr5(D<`N4r+vnwVy#6-!8UgINyJIf6!E*Z`U0~DhGU-&orzYRz8 zt%Dx*Xg~6eN$Nb;Pdv^vb+v1Vg772cJomP&uz&l9p5qc6r^aTkm2e-=Acj%;`=Dre zCUAoKA?PniOGO*VGrj|#goR(R>M-a;E~PnW_i=WAgFYa!l;)^c_r+JEi@@)i1W3e2 z+l?^BuI5b1FbPC2xc^aRUtz!miIx`Ci$J(+&e0i$bI&dPg#dAGT6?$SIseO6anZ%r z-ntiB58_gtVeletsJe6lnuh-5id&_-t#i9?2rZLyG5CW2SY@GWBpY(!^jmj2<+EA- zq#w9DqCGGE{I084o|uIT+rE=4L1TAUNO0TuOaEqIb#oRYwD%R7jdI$4xiO@T-fCb- z28%9yg+?e|4>}6}m1fq^UGq2{+|$xqb(3y-)4Y%`>q$Svt4 z>q6OaCXx4g22^?`D;`)B>ZxLqui^hdT3>9at_8i4ld+AE#d!bw&N5)&$Cza=*6tul zZbVejsT|8*)eY!|#g$q{O<94BiPoCa*rNZZc}#6+>;g?=TW zcR}ID++r{O04IV}aK>)=Wqe?Rko*UzQHh=+W{4HWx)9}!Tddcw%hYot?D*G5@r?<> zEweu`td-i0cb)U_Z0udDHvKHIndKSuMU-dP37(^*>G;}L?ejU&RKo>Fk8WOqEe`4y z=??*&bZ7rq4Oa6zjUk5NtC(@T$w7w0j6S|EBx@032@HL0kD!_O_p6=qf${H{Omy0J za*SRAIDqdvf4r{)5J&MT8YfUXdYiWZLI zA2|o}Ub?+U9#wk8&<=sm+d}W|k=hEc##JXWwsv*;KXutB2xER=e$%rLv>G$yoAW^# zE}b>$9w~~q^$wfdI&6><^88C8Y2 z5@`F>i-dcF|AXGc4=MaYER~BX(MsjKAPnC!g* z_7{tqQgEyJ0ID+@KK^-*`6|ZI)eC)M99^tQ!`=jmEulW`v`}Z)+UUwi!2=p+PV7D=XMFBT-UImb0$7Fa=P~+S z{(cAD4|8jwePzuo{c@~Ma4tX}J$4Dhy1z@R%^70M>GY3X!ib9mZ`ADl!WSTzZdecv z7fV%L`s(1%7fvbO6TkI(QU)X&Wnq2;HfRH79>4G-W8ABoo3wC``bIertD-qk??6L= z;C1dFhryk$QsyhM<+-*>X!-;t>x!2bNwHIlLprvUye7_FG|m+sAOAC*bvm)T&I{=) zqaC6=%PQa945>OhEFW;s)%pf`tET(*u9AmQcGCmPQ}b{Jn%}E<9~jS|HP2&;HDL=m zBZ3M(Dd|frc{v47WKh4DsBYL-0s_BL4XvH(bNDv&8<2kk2r6f7LO=FQzF-F2mH|dM z$DUMq%*Rc{WriNtG-y#nZVLPQr}D$B!dO@JK2@OXPBA&4_8FyT*G!hIv)EmCFu8fJ z{ulDFxy~ZEGT%I#&iY1ggH_{;_^DC)A1>QKkrzLNYz==SIivY*>}OqeJ3!2eWS1y^5)ewSCV|aD7_~@W#AdnDD}6 z02`7tOH|AIyHGu!i_QP_VqPy@gBXr5s0q3IaIb?Rv4Y5pW25^7>tFD&CEi;gBW_7C zES^PD{U;(+ailS5h_Q>>iStKhf9TrBoT3ayS2(k(OxQdPV8H$5=Nnc_PWAV9#%40eD<@;w%m3UTtp&TdqSM8*DSC2V zo1nTWzcA##SSkvx!Q*ajw2)DjBCP8!FcEibR4vohz0k0LM5QW-!^<3u7F@*#KX!et z``Csp@mRHhdh@}XOTTM&$?nK})$uc704zX0c509w?k%e6zf6Ha@$*AfR|pVW!bu{r zy_E8UI&VlHay^^xHE>)F@1Xy*4ZyCbKd3E@wUHfHKMtGEeBym$3}s{nG&4#z+Kbo% zxHhp^1{E&vUnfgoJ#~5G@EVvrqP9e(d6q|u+{LG5yBX)U(Cb!=;hnkH^g^n%u&>hB zzXkhM1h=Dh)OqWcYt9Mz{~;dAZ=Di8;7l}jSZ$q*S*1Bt`pw1$pAxf2c(60_ZMYwHnQN8Bgk4-Hm-CQ( za^!YIj?A8Ge3wRYbZ}{QlpO%?C*5UDOd&m_c63# z6P4U+OO@mh$mG_jd`e~Le?95q}KT$NLkf4@S1qRnC?jXm^?Bs^0QKrK&jT~PB> z^X;8-ap-z|;u-KA+;(EMq~2(Q7qa4~UHR7tO@9sb)K9b6IOS&e74Hug@SwGS8?XH} z0-qA)1}Es3dkA@(7K%S@XW32e8tr(j)cXPe5vc_ANt#V?ynq~ z#?e|POOkC?6U|ib9GmnXq96Ih;sElqZOjgC)e17W{paYR&=Q=f1Z$R0zuY2FcL6R^ zmYD|Ppo&jRV7m1bdDAnQJZK6q)96S8i)|8< zrf#Hmwr8=7aX9mt6pgRYp(JcaFc@xKPqUH7rxR-;khuQBa7v3?{`Z80NxH3uV661OTnQ#>9yfVD#`y|`nBxuXt4dTS_kJ%+Xu|Oj z0}8xM9ANmP5D0ohYsz{+CvjYw2(iLsidU*$?{1aEKIT4bAZVX!H$$TJH#ylKpr#Ij zE3!;^Be#;=2OD+ucM>`Njy{d`RCh^6+-BXC2=Eq06|g3~-H+O#*&daZUCmqfL}I)^ zJbENd5kh&kB6k_-`;-3w&8iA)6$hVcpXS(l*%Ds@J;$Ob;JMi~htc?ltfyQx%Wdml zAnpRhfX4F~IBu`qDl{p{ET>sw?d4|GCFW14QAJnr|KPOyx*gjtFxypVO?bh-wNU9l zd1Rg?gUG3gz2x<4OBLcExLCWb2f#f$Kg7=Bd7XoOe^FbC5eJ3RkA zl)DIUj(bTAoT{iIdxJHYF!3$@S>gV?BY%(ly*9Gj{drA5aFeu#xWI-IPW0z*14bCQ zru}NknHu<$6;27ReSh3&Y5m2G-y+zcQj^J;PSSAu7Qx&5j>7bDlUwk6_^Gq|htFtX zW3gz$EWaYCS@1(o%@XfNh9Al2lUw|AR>7@)NX@wYfC*e9EGcClvsn8YyJW;TqE1$s zb$<&SFy}k6F(SDC&XWZ+6MtqqIZ^9OA6A0V`xKz@o27F;R%W{6qO2EIA8iZIcpycL z3oh!4x)`DHe?FJ6X0#2!BytKRQMp;QcwAB<$e)Z5$DcQvpZ+gD1b8ITU-#v^{U+{J zq4qwe7D8jF{tw}pef^V8~%&XQA&7wWXdRc2tA0@SL_ z4o#yLR*hC)2tRXg)50^Jz+g4?Y-+7}+jSv_fHA2P4V?i4P^WlZm*5j-h4=tUh1=2L z3kwyogR`-HKGUK7)erocT&K0rbX~maT%VWh3YhI^YIRcdR~%KS-voZAkW@lTo1?7w z8QOlN?*?OsO?j{g&Tt|V{uy>NG_+EG@@jr#W`XVt=3Cmz*hPY9#c|$3Ctljj@cqC> zfC%_EM7KMreS|{bjH5V_4dZ|QSdI5qGahFBMEZ=_iM3N^0nr%(I&K5fd17~J-A<*a z^h(RAuI8J0M=3?>T5MK@WH@dxX_^-~3N_<)$S>k6g}|Axm$3TEO}I76Rmk`Kp{DNL zZpA$0uU5S*r!y82{9O?U{X|`)+lKa%Jd%2K2s9ZAj4I`?=c|NsBLOZb#nk|c*!D(@0YDyP}rQmGW_Bsr{-B(bDn z?2shNDak2pl_bYGHph*ekE7VgVHhUEX2Z_g*YEf5b-P{vT(|3WU9Z>u`F!3VoR%T( znbAaj8C7vcxg#^3|I~dju#f69o^5rq444p$R{T@f25Mq95yl_+FJN=Y6D!y8*Rxsk zvFCucb)U(sAQPJ&@H;wcw31wRax*#T`kamoySDL5+%vsp@iwP9=emEEdT^>N`_Jo^ zA^zz#o*8c8`lGz6%mg2qscd`TbD1J(igs&4a{EHb>pE~f{!{8ds+9na{)BFp?PJNaVt-Jcu@-- zwFY=I+VMLRmWX;U>ef2pIB$Hz=N)!Z_yF`Ce>)vKx~Bvn%|iKj=6g|bvmu7pLAO&} z7*|(47sAURKnI>4jsQ`yjzJ=3v)=*>P5z}cGuqJC<^P*- zAOOerAc(Vt9jv(aurYan1Zt=k{iHc0Ij~^ac_y#QXfBZF9LFEYav9gcz>w((J!BF; zneoWJNNbSju$J1>{0KUPXojqKNq^d5PEC#zVor1boA!AO;Nk^ZvY}lo^f~j>%^?PrKV8E}Y-%!y@ zWcW6weRLUVhgf^%?ycz%@3ys{hov?P^&*?)_FOSXA>G^*TdMOpL>ad)vl$f{#|VDO zAC3~5$1$ogEs=Mlv*o!rJ4ZnVWQc4#qOrbdZK7My%@4NWT^x%PMVI)k^Cb@#-(tEZ z;>0PE0M7;4Byd%E@Z4LAzObEwq=jig#b~*wqj?{rdcmyZCkmUzE4n$kyF4WwW;Su5 zq)rBy`z!`j$8x;Pndnv3A=B;X{*69>$L`9a^daSRL!YNm=6(8x_VCzWP&K^}uU%{3 zf^3pM$ZC~oi_=7JbqQgcm>c*EA5H%x{ZXP~Vw(*@Y+b8s_5?-f-HAdAf|nh#w3$Sl zPB8l1x}iiN@NSuYzunEW6V_H}NE0I(hqEAxqT`xR>!u0)>s>wY!jpP+ia3Lfhk;w> zJtRM73g5W7og7+vKJG&E@fdH0R2*E7Kp$TBTJ7g1^?(+Z|ENmG1@e6*3Js3DYZ3Al zO`HD4>osa+qg-VwXsX&6938!fmKf@>1*+nnR*M(b_gr5L5G?WQaS|XqM|lRl#golT zi{%fcXu{S_Si7umc_gEKG;rpUFUu9v9R-DElsJu6uL$-wcU22h#JOhA>|XvyY6#*c z#HZmAp6jKp_4L`uzn;*`QpJ_4Lq6%seHG`bDU1#AZ!MT7w{7c{%GFEku)e!^vBEjz zPr_@W!RUJ$4Oy}bb@F06&O*jR%%{SawcJkjGKP?SJ)^_zWAxQbfLFGAZ_s|wP1sAZ zGR<(pUP^kb%+j`@G_I%vv=bMgdnmK9I|=f@+tctp{B}&Oo=id16Yust`I9)6W-$Q~ z@XWn0ZuDnvUc@0@i4F{lqlRgX*_}Z*#*>&az{fbSfYc>XtG_UOarN_Rczr@+H+bTC zR`~DgTj&mvQVWO*nsh5Di5j;woK-F~QIx#`mm`sg$F0gEcv{f@wI+U(xwBlu<$&xg zDYrl8rM$uvy0&j&S5x;R3Hl!ktjPRQ=86qhHXSr$q>9?dYr-hA=-k}zFn^B%sYR$l z%hpdWck#jE>yzt4iPN&BiZ>e!rc8EYRrO2vzo2qd`+}ky$RE6QLYd^xrZKuh)Y}`< z!z}6_Y%qV|fn`IH>k!remNLGnFM$-unrqWqSLjO4Bc+N!T)VyS+whZwEa>@(qFAd* zYFu;8U=`$UeOeAcL545tL3s8eGMu6e9XFy!0|ss0<&{e>iLw$D$z({q;XHiAd>{S_ z+EsMt|J74Qu?eJLk&I@3lA3NZ#H`q$` zakNRTJKCh1focm^9)RV&OzN+uUHl)`b()L-pIjRAp=v;%nA?+1l zB3*n^KERGX)ttFGYvu=z_IL$99-5Hf-tA)uy|=DT&M9Hl|7$~$gc=W3RVLo@h(#H) zE^r=nB-aLqWsfKhYdsIU?_L@*eR$BJ$R@$#EB5HZdfUo^UWZTVDKs&tjDLMTWD{kt zU}xaw-@n9~^xf#WjnQw~*V^AD`ePv|*XL6}7~L{vXhe9fs)TD-M$Z5VizlFPe>yNR$=fIqdazXxbeKCUB{ z*oE$#l>U`n=>qoO-cDvXllWaU)_)&JUlR9v91?f_A={&cqpL12bIM8%%0N_|u=C^X zT*(x>0_zO15x;82{rqj1(6?3u{0gxhJ{Nk~sxKE7vIl|X+{k5rlm6g7e=MJ^C~4>{ zl%&j~3uu|37M9I0bnz+Ye0QE;@4CM(?><5T;QVjBSBbRjCLnM#eueA>7jRC5d9J<8 zksXu)Nmjy`S+7dU0OXaZYZxfHvYNK$)(RWpU!|*rH<94v($FK~1s%eR!a5w|m-$8- z7%q_MvT%6-G~S4LuiWi;nD8IQSk&NniT5rc6a|L=6cT{M}Pv;K82usUXuX37Y0_on#!i)ZD}_V~_Cl!oINb zF!}Y7R6my6`;afop6IO5J>YYbD*egFp?YCEgGPLYfc9)~_(zr!aKR#j|0A)*;%3&l z`2)I1)3Uu9+Dt@^JV_Ve!4ri~S5?cquVGJ%-Ujca|EF92^bR1+sTmazXCBr(vOJSt z)NSG@izL-A=neV?`*rC+GpY9uY#1AdAnhWLhQI9j=*s159+`|xU}mBuM<`GGMe1AV z`DXQ2`ot&+CX10-aQjDEzM)7PW!seEc4U6^d04_4K%~|SKTjVM`QnR^+{VJ+(T}pN zN8f@~c`cG+3mx%A-paJjZDM<%t{4WC9)yM^uk{srUh4{Zgnl$}rO#y60TV{$ksoK{ zC^AQ;U3>OvOS;XdxFfZ$ZB7eCTJ&=<{*?;nURf;)DwCa&C>idW^Z)A^`Pih8rDbO-h zyV8IHom93+zzRPm^i74ChvA`Fg2gfJjLm7${EWRD*I5M=EC9hpUnP7v*qX($d6*W< z=rxfe;*CP&?T|>WGH_n**Gc&}iV!Taw|7QQAiBzxJuMHyO2g^nj%a>2H`%D3%>6@` zzvOM)H&fw)K{aUvb-_GkO$Itk7DTXo6jW>hlR7};m8h97A(H7|hS76OI9VX|=c8_U zu2HX0HpH=jE)bo^g5pQke1)B{RxUcineP&@xNk3i4u6LW2oAejr@GwBMSqL;JvP1; zC^S^SHq7SI7OlCejJINZNT{dfU54Uew&_WB-FLfgdLOr~hJg)XEmd^W^tksz+I#qd z@w8cEwYr7ZdWoS#$!>?+zTO1A<7O;et|Uk^R1Mb5lAF`sUVwiEdx^F8L!zJ*8d zY05ap7kG-ilA-QISg%NYE6)G1i{dloEF4kBg-@M{-XUD zR)sR#%&m5&!MGLS6UXrNSh-%+jLVJFqpO)!gsxon8vU<$Rwu!-l35ExyW+zAT zk@r;EVe7}{@MeARR1ki+f_l9HmYUe`yqj!Hq`s2~qoVbYZ6Dcb<_&=ncydFh&TOQb zZ2X#EpBy6`(IEYqt#GNwAA?GV@o?$9F+2zj{{Yqb(M&MU^7)tYw~+fRlL!hFwj2ZQ zd2JLuORE>mY10z=xGv!$!$0=3)nEZv9#c!T)_eYyCMknbuoO1#2E0OnS_z}Y#wco~B#ZV9CTa`FMX|EU z;-77RR4&O@Ix7ml+^tOg9hHlktck-z+2#snr?buRABeZ%T`WhGOAvZ7A539gmc^a5|RP(qIk#>6k}J$!|#x0Pt(TM`!-+a zwFAO}2E3y|@5t9--k;Ze0KKtmHvh79dR!7b?}mC7HJc8ns3UW+XCy@_EWHnj=~Vfv zhbi{*1#Zr1SN3l%JT*P6&8r>Imq(~)-Kk?sy@lhZSYLVXPCK4MpU^KqoNkM}=cF|2 zu#Wa`34nbNyD)FXsy>8lKiY=;^$C?7p`-s1PBrfe8ht(El}1{x=Y#xvo$eILV~oMX znqiwI=G_v%3oBxV(rj@04?TV(40od+hgJXNNXp?*!8lMNHKVe0sZ&0) z5%;X%w~&)yorl6`^ev))VH1*mV6kEgnqhgo>0t^iV-OkyJpxz+nw{s3wa-TB>EWCt z8%byKnZWkXhc=G*Dffi@Xx)>Kr@A-JeBtN(Aj~q7GOG7DiS`kX^q0RrQJ?O_+)feu zFp?85_b9JNbZW$+|IRui@0lCUQi^|vG-i*!WkUSUOayb@l40mB!89tKS}$j6l=yek z!|ULirT$L>CC>h>8$@4dV@?nKRoX{IH9nEkCZ}u7fgkBIq6(EXm9tw0{E0brjBQrZ z9zvc6OSV~?f1m?E>%X<_xHm;Ix6f9YOtX{ZsT_j7^&|d0Ay`2L6O5d!6|YZ4@aekO z#+a`_-BPF7LH|*KtY5OCUWfLgM84j&&IpNgHIL-o(>R>M#w>R~yQi#N5l|s%b(T6k zeLg`v!cJ{~W+BGk(?k9ve})^k)mT}yg=fn&(2x_LMe+Q!6otb0$SZ5?#Fi$&UvTO~ z<_EsV1A0u{k&TWI4KB%J;Ty<}d1Q`}IM*)UlLq^3S^-z_i(k0jkb@Se<# zaOVc`vThvuPKc}?OS2+{Y@?YUE`i_0{t4I>jCkNqd67XXiDxGQ!cRN}`ox~WDX~Bsy)Lg##G|<#c+5@N>=iTdH;5R9 zxE(0<>p24GDO~f>k~P1L7nVtGMGAD_M=deazBMfTt|Rj9 z=)~RKn4Y{@*~t2w_HrivLz<_cuYVu}@8yjpb!O8pu1v+!7iSr>3qKT7fR!&eOer6{q z@3GUvlvvhY=S7uh)J@N&tX`~GY-Bjo%6j(`I{fr8hS?`I2bUzl*l{#@G_JFQHpeZr6mOhDsu0?gs!M@Z5LxOU5WzG*=^&F45p zym(=X@L-sdBs5o6dWMq>xOpyDbmVYg_B^Z2btjL>B?+>K^hSi5Ps0yQR_LEC%U;3A zyV;92UqW#WUc8Xge@8E3pl<@RUrBXg!?`Jjvw7p#vv}nb+I2QLF?RYHNt97+Ce5)| zd924eZ&u+qpRY2fMeZjKz|64|P*I_A*;B|Q(Zj-&4O3ilxV*LQ8VVms_%Z9US||7a z&C3?0EfJWVVxK9?Ov=BDKhIs~-N9?saX@pu;iB{aA$s#xnm2^%p2 z;owj5o#1Vha4-Y9SMD8Z3Ul=9j$IpUV=}lAiCTMH;`nCGvB-hZxAewP`CP@bGSWjW z%v$g=y?GCi=4G_*yMux}y9^O=*&=`iQF5DXT{qNU8ln&|I6t-6Zk&a^4I#;GrT2HN z&vkRBrT5zJ8w(YriQXc?PquabDRZ?tOfD&dLp``2n~K_3KO$>*RwjcSBnd0WfKf}< ztY$7sL2dGhgiV!DWxneRS4+Z=QG8AY`#kNDPea_eKh=5;mF!97!+KrWRWs;`abcm! zDWar*gAcA|aTTO*HQGAUjHA^@o5@E2!F8d^pby!a|Pd9I!JC}u-3(a-1DYox%81zGg)s^={Vc7((Y!@K zF_OHN@mw=-r9P5ps}M2u3DroFhCPxEm7#U|KK(!uHZ@!3q_fR2W^!NbKC6&72Rvnj zR<5z=7UR10hqvhYgw}0!2+;>gFG39F4P6q3h|{K~ROZY?Y=$#<#pn(tJ951QBYaDX+5-lPqa;X9$7C1umq$IwP|4j(GLPiOK#fHpNBKb z1Q*e)ZnthRN~lmFwkYLY8#7>1OhMsL`UkALKZgp6>2W z)ASc4*~KEU6CL%SFws(nPG4~S8_JJ0J*R6Ccwb#EeSCI8>&XO)k^VPI&e2<(->Sit zJ?SPZH$%?II5E%MB~baUdBUU4tyqq399GH(RQmCZj%qhCI-oc$7=ezm2dJ;AU< z@A~3z6&5|7qWo>eESW3wt!nJgN8&3g`VP=>AK)~0f5355w$8YJNrv^2#r1Z1bO}2l z?Q6l0<*P#;LuCcEwQ#yc$WF^W|7Z>do&B%#-^2R*_9=atH~9V?m9CMv@9m~tws8PO zg>5n5v+BREZ{PQ2^UDL970iN|-@*%sY#0&y-{8AHW(e;Od))p7%ucy zvG#E)h^UrM$KaDW8@-v?Y5rL;*ns;ssNAS}fmJ^A=y$`CBc@OR4y9-MbWy{azjEZ#psJXt;YP>tdrQaP7=i9V!nWX^*e;htWYKHl>J{*y% zvMCIk`maTFBVpc*!FQjG#B6$L2cW%7My`B~H=^l%)(5n&)^CzETP^)V3HGt%OL~texH8RgdCZ0yc2IW8Icl^p4 z%{*;qJTa5IKL5l{eyWXR{k6~PYd$`aN!y_FLDj}ou`Sr-OCRCKV$z7p$glGap{_mT z#L)4w-d}IqX|3xMy`ySY3z0|Gd8iD!n@@xDed&?4Yd7gSefWQv!70gchDGywr{PTiOICRlnptBL8Uc z2>Y2f)bLHyv^4MR*JD23EpGrVA87}$0mBvLTKxZprctRP z@O??~h_#aI(8a(waG2YJG-e-$^EM3;<6~c>Y{%k8d)yItcB9bmh`49nu4%z^`h%>k zfiUR*1pG~ut-zHfcEoh30IrXmcB>11cjJkeccI71a>h!R+NymCHD9rD@p%9p9I7jj z=-6xXqlK9*NiNk}K&AW#z6AQXS+Efbq!ek9b=EZ zyg2IdWzWdC-f^3Pz~8f)$|ne;{Wr3ji`(}y{;Vb0c<<;l8QT#A>k>|CunW9A^fuCq zpKxk!7InizQff_qAr|{66{G9wW*1&zvUd2hF$)^P|9KTP-W?MOJnno*y*r(7l*?~w*y=vF&D1%OTb9k$(VE{HeH zY$~NSRCmj+r6J?TT;3vr+%pk=dd&Wat&|)q#>b+PA3TeGcDo$BH}J#s}HQFJ9R z_`ca~;i9rNn851rllNBvwg)i$#0kTr<>gbF1|8w_(X!=J{Cr3 z8Aw60l1OkEa~Ip+s;q?Y+%J=2eGaAHqC$k1yg@d!sHSSnz{nXtG_?QWq^?m@V6ZDo zPn7D=rSW02-%Y03qwN_k@fraCk?oEXQ`B^XZTG{akDW`SjXKZ|Czj2TYOg!Uu#9J9 z;vN@+mJu=yNg8?w(#o0S)dj6BTn=5d*zDzzn-A=duo~a*) zq#G|bW6x2CqT>f^NmC13$j~af&ExfNViKemGjdRl9<22i^lw?nF7#sk`AUrY1VY-; zOtzQTck^}1pf)z*H_kI__VIrTl}G+^jSffDEuDPqp*^AR<*wt6`E`zDX8CZujGo>S zE}S~JK4btk*EgE3AI4WVT~=dC4Rrjet@Y9uTT2qvx@t0Z6MS^(<}p4dY8cbPsV#S&Sy}-U9#nmk>DrtyK&Nmz2cL}>)NiLB=*sP(DF>^@z8!lgmB_xi`kvV ze1Mw|`D2rj+s+wha}E32;wVJs4gVj##OzxY^LDq2{Bm$T{HpzB(REv{i2fo56&gO^ zbw;+I-h*{{lo+UdPX{3^iP*4z08A~~2YfCsUe~TKW5ggznb!6z?jQZ{4#mW$AGBGizv%=+}KO=nGAIkZS^r+%h%V=gyH{0nh8htle zc&5j~)0T_{S-M<`OFY#=`+DC)NAmq`ZvY(6_^N;8U$u=x%?Ocm8!J9-{R1zlNknrp z(^*yr19shJ(uJE#)tE*RSlX7;JmwLm90OXfz6fW%X@8Y4Ax$C`~3X4 zj&LSk#Ma(bz9^hLc>Cn{$U=)~9jCn^jb`AZs! z-+f{KiM%$x(0jTDu>BCU`71dZH(T5>Ft=97r{d{;TTX|hvs5tZOZ?e$xZG#cfpiqcU7qk<)NUefY#bAx7n($_Ha*YsDm6j~|cR^&1>;Xu%13 zVitVU(;DFG9!3q}g}cy#lpz(j2G zmra!WLzCk0Nx5!@4>sSYQRU8Q(B84RJnV7!4v@Ybb}GMj?j)`M>iZva(C>=W4Z#F?P{-Jd5R)Irl67!)w31Q z+?|(NZmLilh!kV(mJCf^)v?8_cLo^Jlf5JD8}A4{$v?}+2ICa5b{&U!;w-dDTh7*e zc8wk0ll;o(>Kw+HBfXTRCeZI*uQ6%rIzpOSN9HAqkLBwlc za4wSurA`~O;?e^vg+5=|i_@*GFQKP~?Gx8(bs<6+geytohO_m#T{99}>Gk{i?&l{+ z_Dxp9Q%ZPd>R?neCPxC{7z-;VPyW*Ss{>|kUU`%#yq_d_2NrL=IX!@AUDHFQIu*0X z+qr%m)V1l!mIlX*uF>T{H0Ft(JHG_igh^)ka?rP~wwoL*V&18iTXU(o#@NNqp?cM@03Y`H|GqhUpL<)iZcV(#z9^s-S31>y=w1^ zY%k3Qp{eiq`oHFwr{CE9XoQJEAL|t#zT`>}*W=$K{D-tYtcB@$4#c?w&IlC!Q_C#6u`&+^}x%<<(1gZ0_ zby;Zr&te$JvO(0$BWRB z+uVnJ^^YKQA}TXL=cxWn=?KwcV+YNH8FRxoFH^uPl`Wo0Ra(ldO~Swj^pq-Q+v6fH zgUsAQEcTXG^$OCXh(W7tDil23CRr86eo4$^%xHg& zIJiOC{ih3Al0W@c6Ay2kLvu!BEIc%jvkIps-?3vB3q5Qnwh(xhFVltw;9^8q5OT|! zL_dG-uj_3xOKO+ZGVAOTzlnZ(uMadgmQq*T90Vz=1IUD_*42HEiOw!r){n=7>48^6 ztp7NCtxj@KX58RCD&m_^BF>7(mc!N^ocZ+?iu#Ksi``@9Gc~%WxKjF@C##HZt4q4$ zg*s;71o}_9B`UYu{8_k9G^? zWp4rFCqr+vWB0s+#!|AuvKhO6pDlikyL)APxHh=a^v$})q2*7w_#`c<&*OpUE6(ht ztA+tL?l(AB;k)Onuum8$$muGR?Q~5I!2XFKbS(4YX3R{5BG#|eiB>i8L$o+rTcmYu zk*24_vbKIG>w!X#BF>Jh?Oto=>kpQhB-Vwr)S`l~0E^~iFw8n?69%YvNOUL}u1n1o(FC-#{yOsj%z0S_~zNy_C%C`{TulQ!7| zD=i~B`n0tTQToN-BP{r`7W40#R|i@Cks;jq47NMMMOuE&6Q)duC9E`%fxp+HP@nwt zY!)JOs7*)338SjojhNM4;_dsNrt(9Eqa7-p;t|(3gv^E)7}^^MP*1VCqj&W0Q=?# zHte1t->EFIcQos6*Ya#k&&Q8)p0+`w&9n zPCjrbtf{pD5T^mvw`&NW+rx?K{&qD42=OOp4BUc^mX%ytZT00gQ1{`0Wv{EdEc((M z><1_|zBatqTUhgC^G}I&qX}d9{S(*#^Ljo#Zc`%a(Pr=)VDWx^bMHJ7cithy%vQJZR|GgcyeQ(2?ux$ zoEYh~T<1JvknKC}G)#d^FBR+b!_bx8Rt;?!yMc#8tCT)<{lmi1QPs75Bdj&7p@q|x z6^Omt$jPQ4-?zx0d#3FJ_l8nCM{_8lO4IvUI{<(TC6nwQZ(s?3EwJ|_ zP8j;=H=Oxd;f(sN2TRw|F)@kgZ0*t&H2xlXl5d0z;-FqzSA)KqBh&UA6i#Xh~7Vw{foc+ z-DESEV6~<+^QNbDuhy7Qnsb|?N?WhKCGs~7jNC)3vAjQV({Q3OY~FEdrE~)69T((h z7l){DI6FpF0G0g7)&G(&-Xme#ckn&#dXztTXAZhX=6wj1%(LX3SiE`To6NzclZ|5S znW<{J?(_fAG@`4G&jML_JGF##p3$Om`wV<2CQtLW{i!9-<)2$XZl)ddpq3QW+>(Bv zM<=iF58EC$nS@QXC#~m$r|0kK3sT92*IfqgdH%W~+r#NRA)AO$(VIeW~5!;vd>WeKRHg9bcE$R%6|PAG1V+y@fmp|A78ha%Cc96jV~{^@?x# zjkR0)dM`hZu-2GI&yv0#jCkpFGp6wDOoV+;CTbEczj+v%-?w)C;e_)!NS&R!ut&5R zO7tTDKNwh}47|ZRATK7>O<6gn3q|Kg4BtLA(TBXm@JG)FE<>^J5$?_UOM7(++t%XQ z#*g*lF;6J}!TZg>hX>6RYrd&)fPHVQH6w`s5m$ezCQJ^gIgL24W~vli-{fTEGj7Vs zhpafR2a*`3I89g38`6;*`llwGn$7?#?|%aQ>kHf~?0lOfc(JB)0=P{e>F$zifp3XC z8-YQ4qXz&cJmee5*KZ;nS3?1602FoWdGv;B3!k-RP^V;XJ2V}*-mVJ?{xnES;%+;* ztN;qw!G9#t32?oIj+o!}6n3q>c#jKAcjSwS?ep)dT$`>*Wq0i_%hKyxk=%Rx8EzT! zrDUkXc;mS7q%Uyn!7ES@v&nUzjy>$}&n?1VvFDx58wymx4A>h)>BQ)5pz6ZI_Ci_U zXzh=tLaT!Ce*j-!ipfS*qcxFL`Fv&EQKUKX?4&vRXj^Sj@UWHv*2sJ&XA;FN`dvJc z!h}p&j($tv=e|tVY&6LFz^{6ARvs1f%LiPTOi6dqCoP7z5M!8jn}mN%Hi<3Ji{#Na z$>fup3I6i`t=aFL%Bnv)DfWGzxTsF!eK3#X!@Tfc+k!3&Lbg5~bvgS3+w(W1=*iNj z086vG6SoZHq_?7JBSo%f(4t@8-WqI{L7|f5BWV+U>Eaw9c2H$jyPbF;?VY3Xsz$D%xg+qp9(QnL91D!j%GzqeZAA_aw&7A*2^;*JIM#D?iX8g37;z zT2dq6lps~AC1ZL?n->ZL=`%6mEia;zm}^NP<@@(R_N@eV#su6H-RPh_z~0hs51WTN zZRyGDii@-{0Ns!#uJ<6F>|KDsg6Mf(p zQnt~17uW-N!G9$0NJ+C*=yPjkMtf9$NCVhu;@Wq~Wia7w3i=<3-!SomWHZ7p5|Qtu zAM7eNo-`K>vP$9NwaC1y2vM=fXKxX?U@Ywrm(#t- zG9%$#f6S9X6`r~a-@MUVorY$kf=;a;IRxDSaisqbuSI!-7?Z}jw|Acp9Tu?PN8mtr z>G=idD>G?da*WI9e

8nFAjX&vMzh)XsJNJ=e!yM~*-KGFsf|boU*snpy7^jou+r zeggCM$UL5yB73YS?FFU1@d}-%q?mt^3kL6S&1o-T&)2JMEL1)OF(&Ku#|~pLJ5hg$ zA*To-tmo|)1t*Jq-I&Cc_a6}Q4~W%roxBL5=W&0+-m_#gAN&49b)0w38EDMAuopWQ z#-V530fj~K)bDxlVK1|=R))H8zyI`3&1qZpr>zdwJ?$z*&@-df>gJe>4}g!1JAs8K)Yfj&jmM zM+%zQox&@6j0Sx+Jx?vi=d}0JWuNZ(kkCRU>5T3&FYe=QMl;Q>b=o zVb>q(!2djx?r6zKZvV+=%WD<1KOu)NblKZ^j+js0Qvp0|eK2)jZmYMCw7;6Zjo-?tSAczOl!eDL??kZ7 zz5%w0-y6Oziai9qi$TQ-;AabHEzs-s>zyxZA5Qt2hh{@qYhc725(n28UbPYJTvH9z zja8Bok#jcZxJoKr(LecNVOW7!i65>J_#8W>kvdHDH2t2LYw}*q(&0pbhl;28duIq* z9a4UgXug)Kb48Lw^dUYckZX_Fn~!(`7j zz_bnAtZ{TMeuiD0u(1I|CTbe6eh!6@<72d#or$a|{v~}{{Cv@HyEWvl$2-Fkoq~zs zDjT@^L=0yGxCEw#qR+L(>Qau3M4y|riN$D9p!N+x>KjDk0Nx>qLIB{{cLvzAfeFz4 z;MC14WW`8vmRY}d7k?HB-NeY?oJDhA9qsY*T3EOQ28fkv2lf+G#Xl~1-nXS_H!=H} z$KJPBJ$^90+1~S%+HPwCTsMvv;0oX2?b;g zGnR7cXZ8E*kivSGDQOqJY+0oT+d{msUL)AFIvJa~4Rnn&`WELD>WMT$Ip*=go>5}} zJ4+lS$1kn>l2n-CT@D&!J1rK=16Qlfw=e9``h>rMD;E5F7Z2JyN$6HuY4}9i!a*(s zl0iD0HF2mMjyE(ZzvOv3Fp8ei7%6bYtwq?QJVN=?)SrZ@iN^ z^7t#(2z7LTZi$i)=<*Zie-%!4>zWQ~F)#&T=zi&3zu_XY3K+@sVDR)rE*-+RXkzON zm+BPM=i;_074!e$!;KhjIONZq1+5qX-fIl@&+@Zd&U_orMi(8jFcJ|hpqp_bcIZB+ z2$cS*>Yh17sR)1_AXshK=}hT*+^8vX#pA1ZT1Otp2)_%N#ECL0g1s@yW*HF_NdWI& zozE&7kb7?=xGB=}rJSmtVz*7C%8Lt`O1#p;tPe4=SBKVrAHW5jD#_6omf}P=res}r zXxs4iPdDcBx-4%z16-c=NRb^l1i@Omh?EfZPGD04YV!y!&*$g2Xzpa}W#a)@@dXb= z@7u%;G+-;UhWM6pj1ERI4#DlDm* zo;C^-cbr%55B~ibdGsmz<}hEKjO)LI-%UxAuq#AuYc_j<7$AQ6%&h4qAr5WK2Wtc_ht3oaRlyZdm!8&X3J&?N z-%wV>mXneY2SN`EB$jJK?}L=s*vw5P!v<;xQHx;S%X5sU5e;*pkeY}6B&NIC5wi&bbKTvylXjCZD zN$S+(lxhA8&?qE7I!phDUQ}EQo2mZ*)@FHUs>_v7GF^61(?8gMoq@;1-OTWh<|Q&z zqTscOFV>Z+5&F?zGzM;B>K8aFtaZc%V!_ftv{OeK&`|%B9sP%{5codib%#A^=YN#v zA10Zm((VoXed{l@wnutQ{OK#jmaEyM>`cEPt^F!g;xLPqn0Mb;e9sG15deiZ!W9dE zX<-+*PgekE^{n8Zcng1n&VoI@f&4G>jJL zP7P=vK86CEaL_&=#Kn&1qV{ZoOWR15;9`(f}>I+hwc4Bw9-_49*FYI3CX*1UC zES>DiD&GbrtNkBq; zKMd7&p5qme@eh0d9TolG*{cTC2hK<}5|ZKgGjh`4G^uEl^68rbwGBxE#!b;taHbQA zYJ=qywc`D{J&AV@;&;&>T554HyNCI#KzvHRJ-dv2!^hJi5;F-a=(m*|yPvgwn}4&> zuTq_}?(_!ktQQuI4@FH_iYTW}fkK$`hEeoAchabiX?th~*{QsS^Q4ag?WonEOQD`h zlf`N{XrDK?$M8E*9Oq8In#bgPtvWYL* z>04=V_(kNZ041u>vuo%|^AhN%pBB%QY1R)%)iuxa5VvtF`<=#8(Z$+L8V;!C&zfWZ zTPc9|E3Dex{zFawo7Vb3(P}K#~AiS8iri^j*R#X^Y2y1-v!&?!%kuIhJ=$S7mrl4frHuK)(fSLusaBip4pL(NJ+0u``UU3}<7bzp z$;==0KGYhKy=-)t&F@hk3@2ZS$dOjYhN5lndbpwKCfv$q>(wiTWlo+I#0UO@hI#l3 z?e#Q5AiGVgV|mqoK4(*9kGqmA`}SU5^bes9{1akxtTjll#g-H4@&@lRLf(<(&{uJQ z@0wy^dxfAtA>@j0^g#zGo9zExK_|J0(`Vd_I4o3_h#J31IuoMzH9l^U?x|_}fDCIFf(%JBi()ks3-;<=%3kI6yQ8v|I%_Mn#5cy9^_t z7d=_#llFy5)ld68-=!q8T}>x5=7#02Ct{Qpsm3?wiX*1GUXnM`ub%G+@L*>fVD@85 zt`x(junVS+yBE#$`jZVnXP*BkTM)lcm>o>KxZsdB=Yc=rsk-uX-fUZua*T$D>lgSf zTk&lR;^W4@CJ8l8oo?aBKRKEm!sggwv;T)4Za=7>azpE-Y_Xh+zDi`iAqZ}9967$5 z=jz^LdyO#15GOcysc2@}0Nm#E8U5g9U2%=#P|^Xe6l)XU#yhmKyeB#!ldiEX z_P1p5_&XAF%()10;RW{xj%dY3jUB*q1D`=gCND5I^sUD9M&MRwr`$}_77ZaLd@5qHQUpW`qqc@=PGngBLj z$=OAxAEo)I2Ub>gp&rK2tvuSw^!Xu?^Nm7Y_O{)b=NC%#1#&Ub@rpJZVQcAElA%02I#n#AAO!+e#PLJLxmJekTqSEMyl zzrKSU^5#1+RhBD=?ZT?`*DM`nej`9OKg$+`Co=DhcS)z%EKb$lLAPOorW9Ep4Q4o8 z5!Zzhb`rUq_;7Rtv*%@!g%b4jX0hYBl0uR#u(lcFep7l%T`S1SWfSWQ0QI-eWc9w& z4-5Fg`;vUuDb*MA&N&nPUO;=1_a^R~`ocO!dsI)?`OyvO?s4hZ=|P>(6N^{L`qx}8 zU{E0<(jrczC18;rWfSg;d>%jB9YLix$&i@F6+J z0VWJI3eBERJLdvIAF2Tfezz`ys&11zM8|yPfS}C>+3FK#|Bs?Ge`Lb{<9JAlB1tO8 zl1e2+A32(>gH*~YiE^xxRLT*Hv4bR6NRoujQ8}Yhxi=z&Hgc>3W5#B~Zo9Xy??3SV z?frVc-mlm5`FJKu)&qo*=UsWn*_P@C)fLcG(U(s9ofq+&rr)aNRegkdO-J4&QzC;T zu4&|ueChsj!3I=q74E_}JvVTOiTrKpRQ)$|sGV)FIzqG^7F8m-mY@?*9&qHil2UK> z3LViV>N)&=%GfFsw4&2YLpAkWth3Y4#xF4C*C*T%5F2E1FBAUXyRe5{G`%uyd6ZUb zd6}!h0Un;Q;HL4^z=qxFPBKM1MHh<1Tang#Dx|OG{GRQbvn=ED%9n!fXtv72-n+^L zcak0h?Sb`rdz8!|uNT0gl7VLcPQRP98BGPg zQ}zO1AezDV^_VZ4xM5{X*!Z)IYPtuirD)ab%9PGB#2lUr-l@O`=3 za(z9~TEuHXZ(bL+EeQ#Wt#Z;^=+Tx^*706!i86-X?!;5$?WR`9su!U*g3CXJX~?Fs zJ~!dZQ{5gGAIb*=;+syOpJhz(TJLN}jY}bAlc0+-`F1{e@dFCt138V%x#7`?VOIVt zIw~Kbb5U0dhi%e?ao62orrNhco+enMmK`|dR&!hYe|^!`c8dD-&HVj0;pX_B#)p=o zi*=YM*ug)>b3#-0ySLk8UeVqs>D+5klU0okiPjp&q=OQ(8Q%^9j`R0fGP(E%F4&J> zdV922$Yu`^zc4e~>kC_iId{0a)j+T+AzK%5WDowSLNllARTiiD6|e5h2>fe%z4XK> z#X8PguNHooPl6)G8v{QFP! zmaqRqOzuPG07cN%e#)7^<@M%o&d|$dJkch%L2Alw7kTGw;jzM;SO+R@8)2rMMoJ|= z90^tv{(A$yu%UTcIX_v>X54?lH3C`vR9dO_PKkUTGSRdc+mHTMUw3 z{OKbnwh!0l;62NUQCB!yIML7H*;dwu;~EmVr6^lQ8^jt$xJH@Q7X}Ye;D!~oy^HA$ z>8$7irx)7t?Mw$i*4b@6j9AMY~o$^O|4??X!gznGDGrnV@ZF1vJK;95eM<4tU zD^F{0{xnnWIHzw5@RYz^DM$*@f$w-63SmE*9^X#?64Dw#@N8a4aO&CjJQqOo8x8j&g0qx`kCe&b?feFGDq#-coby<|vLTQDqK zncQC*f@^*gZc%)uYZ;LRS8Pbiy%GGb=7hk(a`_sTH^;F3FCvpy zGUM!Ex2wSkO8-@1Qy5KJU($%<!5=EUKc`!4?HRKfY+2bjs?T3C#* z7^E`!_oUzCRz2%f2i65bM~YZqzP9d8ghxm!klMWD*5Nn=$!q+#YQ=5+93r~flfwEs zawHUMq&ozudN0nZu10MWKMVnKhJ+~S=>D9U&7Mr99gGtVPS#0{FRhM~Q zcn)2iTz-LQ)M_HfGJ87QEZ*=v0p&VdkdQ~4kUx%7b}e8*hel;*Sm(ujW)N2`dMQq&!7uygqd`PIJ0BYki`(}3AsPdeN6DB?iwU zQywq=<-lkReyX+tm!^uH-!q|O&f%*h8DE+U%Ymnr3_FB}DWgFLHbwl}v;nKAbFDZ8 z?Uu@BPD&~vk=i^59?5Ew`1|DYPv5GLDBF6Wrt-Cd6Bp^x?RH#bBKw%oENe@Y5~xz^ zQ`5fDFhfuTSX0C@$=Mx;HPA+YC$alFi6o#pfrD?M`FrHqv)?lIxGDNUxO0Y~oh~ zu_iwCGbj;zve&`*S>2W#!1IITvN-wC3obO&(^n-A7!wYp12&r!lJk`G_Oo(eeS z%&`}w(6k$_8I zaJi08QtEXDf^X^1npAd6&ZUMyF%sNPR$?r7BAu^ianw5ec&>wcuN-5J`EcC-UZ$*n zs}wgp*IXLbwF9aKog~3A?YWGr;NC(gAJ6S719sP!>iavzSUUXhqPV+x)u^i@%5pA<2^C| zGTBJv`MJ(@p_c;NM&9Afcc4qo7G2HYRHQhuImV-`^)dkW7WlauY6Z_4-U7%t1lgKE zynTVv@rHYIqy0`ooaPm%{(RVTnp~Zo5&Q460w`lQk9210LaBwbkPk{ONmF1DB(TYD z`d%8Tj<2i)-ez+hgz#t3Bf#4epSY4P9aThvI$`*gT?`Q^sTXXBqFHo*&pKkz(IcCB zq!LOZDGrO)IZ7gk+t`%YHGwHSzD_Qva^SSjJ2XKPp5I@nRp4p7hyN8hgCU~(5b5fP zitnE~{L)(`Pgb$~4lf}Ty!C>ug%*mbv4~zjoc799&9m!*2=W2 zYu4|9;2|;=m)iMYvSUbp#1#CyxN#F9X^qU8+#|jZ7H$Tgbl!~|K2HmG2ds*4h8V^D z=q+>+tdMw>AkK>B-Uqaw6cSHvgSZVs75@1=#!d?OKGiwdyi3Q7YpTb+u1e7vNiezi ztEsKsrXY<~k{P-YoF}_5ch-E+Q2s{77jL5(Y!lw?icIh7Pl}!y78XN{za%TB-h*af z+hV?LfpFH*9rL{ERQ=Ay90?jRLks7v56~v}3-)}%U9BqXj1+oVuO+)=HI`B>H%JV% zvIfdLe$vJzpJ2dFmgT&WA^p#B5sCcic-ua4P1)`ucT1_B^kxFE%~b+Vg| z9MY!R_JQD5X&ZXR#QTy`T{IjEB+h&PcZhHQqRIc8_e-K0xGQ+Vv{n6z`8f>JM1J#f z29s$S_7?Ab4E{nCb^-SBRo!mHdGVwX?~SUALmWSfEwNnVbcA87L0UR3^4Gc8j;!(d9U-^qHTcV9N?OD<>wvX=P%RdlQL z0F2Dz2NxtO+s(VON>*cqM9`ghlJ)i)o@I)A9&ri>YFg1-qzRnEj7%DNI{;D5p>UIb zbmo5$eESta#CMR-;a=>mc8%5>?=dW;0va5~O+G#QV+WYIO(cz}v421%^QUs6$tx=a zqmK=@m5%u~Tk_YrC5DPD z9YKKQGHgK^QL;a#+*}8)M{oI{6P+`a=0%A%5gx-Ro~EyfG`tMMt|J@4g_;CX7C(dC zJB0mt2%in281dFfT7h403=#(}&_3ULR;S<5>2wYebu%D5JvJkS6Sq2Y;^ChPR!>Ko zPV~I*Q7Ea*Y0wOExi@`%D&W+F->||uCRcaPGT^Ewet*92!dLk}3YynrMzHH|+M1XA zDq6re54)O)7e3w3EeJ@8*#bM8$UFu-a3gbyWd1^#09w&Jw$e8rmmiWGyD$2Hom2i3 zU%d{l8gR0|jxAA>McH#4EmR<_XSo%x^`T)LK|*FSvevDiPHuHlCp2gAd5ppQhc`oY zv026|?Q=7pZQt zpM2kryhw9o4Fi79IQkUVHXp_J*^0^xXesGqf5_~T6OxY_lby;w9MjXS>1iX%|EQ4} zME2=uK-f8Cjf=rP(318ZL}X^<@$c3TUEx1dqD^?sCP?4Qqvk5&Wup&&y9m>r};=N-LxuVzfE@@%pA1X4=52E!Nq6Y5W2LQ+? zcFsVs5np?o?*F5cK74b`cI3X(8V`0AB)gI+HiEp3xEwkm+k18-xqlREP5fwb)%M$= zO#p+|JRomE->liFRe8U#{KK}!A^Qh~homdV8c%cPE;U&3qKzTUGK!YpO)$DU9^FHh zW)c_es4C$+a*Utqx-?^DtAzJ<+O$;p!ex_lywL_ZF0Pbk2wwPPr z5b+XDuJ4{Um%gtE)92Dl@aa_EZ@l{&kn<+tP3V2Q=j3b36dloJHDF^fD0@O-c@e-L zwO4P=#W7OCF}l!+;bDY!Z2!@SbyKya7W4I)`T=QO!IJ!OpZKhQcX<30K5r&(9g@ed z)DcUw%=4^jo=Y;yE+#awaeD6az&6hOCbXF-i?|WG*T$OMCqpG*EpQi46m1R1<0!W5 zxB#2>+%BvuVj^};YnqW{R%bou+CsmpnsPBnvaYr&OA`_EU?CaKI6)PS8dzXpX@PmX3q@og`QoF-0eR>KDJV1@1*Mq%3PfSd zQCG`YZ~$r{czHddAqIuo{fN~OxIywKGXnkGtrc>%btVGSiJ`=`leT0?;vC13Kd#}v zt*rauY5{lrVWrRQ74z7GP_#ffxbRTWX)dW6<`awz)AYqXZqkZG}sz-c5pneOuNQqf*??aTWnS6?s_ zrXI1}I)puq%Ocn5Dhrb_;`H5=6Ck8_R|6tD;%iUj&4TPE^O#VTnB%@rfWZp|>WjWj zi=X?$afc^KJCXAZL#)T5iykr053Ol#!DS2LX`ik=)3E#n@kZ2ad0OBu{bM8)?u5N4 zaHN?gE5u)?W4rtc)f1aI=_e72_~5Wvzc=`gS)@aQal}D`@;b53SbA57<7jT8_sd{t z>;+*H{SgNh`P4prB-e`F_I|sDFCsqT_6Mt*AU({msvJu1kCr?lxVkz2=O#m_<3Qy7 zAF{m8D-JFvFPzBdopa{$3{voddEdQ?phVi@rXQ}(_G}Yc2LH58@~+^Ib#cIkylq(q z>tnnr29){2wOtoD9gJ;MJ02Zu@@4cW-pLIZqJ>0Z;?WGRk%oKEP_DTM0|?TGgQQ3g z5Cgm8g&hKTE1MfqmHskPRif_b4G3290Iic(eq!0q|DHikLbeU4_+1Mi1-5<2#hrp| z^Y$ixJsoizUfauyr26-v{@yS`S8z59k71by$`wTq+8PP>)6`0#vc`x9LM8l>;_+}A zu~GE!pZ1vLaDB8(fOup(J!-N|q3lrOjGNa~Ti22?xW^$=k%A|S;hBD4^$+}{jT-*6wP-H2r~QUD-(z%GX($3Jg@af60MxVAmh z>1fu&vH4O`RDu*0YDRm!bg$?D-H}NTK2$=Q2_LtvErE*HjAUS6mDa`>%kWj5G!;;c zJI|nF_XUl^2eatok6XXFh*!Gp!zXcbY-x%6_$s6Ow=^RSlyR!S0`&V?-u$4s%VbMe z(v_LNCT>=*t#5^oA7&H?vGD;G#>ts;GcCn<&KQiJS{bi_0MlHQHVOCv=MnMuoWaz# zh770QVTq8I_gWy4zAY_gll&jRZO~P|U)?HBe3DW>GxF!_`LIa~-mZefzUPF6-{ES? zKdNBnQ{A5pldt6M)r`xcL+z&#jJ)j&Etl!h=gmw2&w93nn;?rMbE|k8b!wlj2w45t zx#-$>5(Duj$LB}stexlRu`@2hjnjV@|j*qPHf%kH+nfR4aV=?J9l^CEI29f zFO`?^+?(WrUhcgku-ROMp_@)6Uhqclom=~)kd*C>J}OFnQmE!L+WG)L@a{%W6{W>C z@7!Fx3X46TZd(^?ayO7}DtoF_V z1}m+pAr-dxN8Vgh{yn<9L;PQ98M$5lMNO(=#)zB$n*?|+J{?4;*Mb#Y1AM1?dsJHe z8V!4&f;;|^tN*d?Y99-Br?2)<`-RL*x`jov-gNYsm$jzz13jQT=0)VaJ=6USP1TW$ z6H-USdR_zczf!4wV{9?*6eT)|%;W-DI}~j<3t9F3;>{GdJ)mut1b6y*2zgerLrI^Z zGNG+o}+SE_aERIiqDa!((FWCVqq#HSPkPgxckWXhqMhsu%CO?jS-{trZ z1n;>oIq=oaBhVwTxb<#V>PE!@*4aU*gXJag#vkQuaw*}+G5KDQQeTo^U)Ow0l3ohQ zbnJ<5?3w*V>yPl)Mg|w-^CDAt@mW7FWSMhX-b85AUWTbda#7X@_T9ezPVp-<^vwXK z)6@p|vR;~BR(0VF^MbwpRc{dL4CCoWUNHXZp1KvKPXpiVe6&@);_AUcG+}GrWh4IH z2*z4o6FKLGv41BGfjZt6dKPm87G>V;6ukxbXS?_`h<0RlG+C2$h2of_2If5=YI>|U zsy{Bz47`oBR(hFym>9k$_~Xp0{JL zy98=SHw;ai67o}1WX-ElrqQW{TbgijgIg-^zlDfDHXox4XcxTY|H8>OB;;|NZ<6A- z9VG-d>O4(r8M%Khd~$wsctP+UdO!B}9^BvE1nhqBulYx)-!}?R7lUP3D`tYtoQ|Og`QYp~AB)g$E!i#Fw6U??L-dVM4ArT|GjH?@i=aBidv_tZ^~UQ6%g0uN?kzn29?C!6 zBR=`=^D^H6;09;1{mG5Q$^V`RW2g=^y|)X(HG{#pckhu8lfxE2o5=D)x>NprUAVNS zRfgS+;5NbB?Ne&6{+5x9ZZdx0Bcu8Y^X zeF*J7-+3He^sq==#|{phCC-2tu}gi0s_Y)1=%TceLT?P1`$u}25~#u7#;MFRz=fW2 z_Nc&SIbI4XN=_22e38lqVtQBNI3gC!cVuY_+R}-ZBB%!t#K;S4YM9-ynLjE0SsLmx|NL4gmlMirW*p3)Q`!KaDyoX9apu~hzk z&#{VXH*7saI4Zw7t>rh3D+Y{3Kae<5ph@_;j-Kgrm69u}_}aw0$hdD!fD-QeXZC{) z)XX)Md21IE<4{(5FVlpOx5@b9Ofvb!MQ?~d!Plb3ETpC&(vM`rQG=t;(cfe4?g(=* zr^KxlgUl#8#$4NUKhlQTX-e$XzLsx_It$2|IbdI*9sPL7Wp11(ktbfTkaTW@rP<#N z!dM|9s%rvo2&6~b#&#*sh#;EsQ+UFkK8t$2BLE7_07fQH120m;HZT_V58l7mpX26v-Nf$g3b&JV ze6=5E#7d68%AROx=@*XO0)YoS{YAf{KP-LXGwJcneb@2b!(ww=(z@1`_{h?~Y0^|H z6&~4l!1D&zV3QPgc2C2^c6Nx)CT09~Vdcn>E-cA}$807XGVtFQKJW@Ly9c4q_yxZA zp>BJ7Ftz3|$xvSB!HYm;cCz6X3*aYIy|vcOYDLpYW+PZvM*Z4= z%SmGVREDd-^%>&dw1cw37GGUmmm}I4@8-1!H5z2JpTkm{gW1wv@rFzeRH9UwsEW=< zI>&M(@cf3+9hk8xkmUo9qd776A1iM8l!MQ%g%-DPA41&d`X;9!Fukd^OE3>?+5Lrp zF(Hjn;^xWlGZ?D#-9C4uLTx)^WrOt{wK#7otQ#obcd+iA$^8@Ch`djxN~mb_;CtGL zy$S7j1&%PqB>kvg|B8ItdFu_#5Bn}WUekCfc-}LrW3Mnc52xWT`T2H0%d@s_3wMx} zxyjq498cRTPWC6zUKd+WY-OhD5}bD{c^YMW-%YK7n$icvJ0o$#Kbeuf<@9=zsdg`~ z`gySx{IUj340Y=;lb+3^y``4H1oEi+pnu1d!B*<1maTV@s)cTh5YC3_!v2n?btWN4 z@lB^Vmxi4C5!VL(ofG+Ok^oPR2U~WRtQB8Kb|Sl-;JVV)xcBlce7v2vA#BgoD0foN z;nSW0j2Cje_ld_IWy9U4j)|Pms;$^GLiuW)vYTZ3>ckC{cwIg^?*$x3x)hedtd5wh z5dB&U@#bQ-^U|v%b!OUmsRo)JzSsQrR$12%*yX_YIN)7rhC6F<#V*k23}t7FuF^mj zcT{<6&O7>4;air(t$M)QAxlvEEY{&iLsz-;8|{WSpgU7-NnK^H898Z9UHl*L3of8{ ztwK%4*M`WYq0V!J{>+guZIG6Yd}RC6EZg$WXPlv>s0zB8HZt&g^!k`2YR5-OVN3IJ zZ6>$+ZDzRmjO8feGQ)fvT@#CL3?=9l)ICXeW1qM_l8 z-0e4*m?L)uWS@8fdYVa-P@T9@Jt^ke$g{=)KyZSBc*v0yBuxQ@P|UgqL|Fv(@qn~Vnut}B6@iVS^ZZTVGQ2xpB>u*beaX@dQW~SdB{_-aQ-ZB5k64Gz`CyW9o(QxkQM%=D zF0d=NBQkNp91JUcHcZ9>~8u?1*$1 z>jpT?^No{p4Jn28f2aN_g#B0BtRJ%_Vb=oCQTep%6V04SXk~~(O?LPfPY>Aqn&SD} z05)a~B6*!=da*yJK!84-V*b}BV_}WBJrX-&bv=O{1x@$Z0IYX`Ka7Fs3pDhNk1%G9 zBG3R(JY-9yo^UE8dVO7sDXQX^{R>3J%_8|Mco%x_C!%3{a)=wJ?CuBR)v56eTVDGn zsj98p=XTKbp+1@3h`y_cmDNv6aw2b*xTX@}jB~2|ZFL)+o9V$JBR{?A>5Pxc9g?1A zg7~k!drYPYA+a0Tj}CDnog!Y3*4{iK+{~$ajOz9|7ZD0qG-k!qu5OFqFKj5pHmpIm z7CDW5F<(QojymkwHA*rgH{G;keu@yWk?0qVE{_!BKm7&sYY2m?lTqu1oSAh}`97Mf zUCjXJ8`6d>i}R&T^-8+%{j#Pth{eNM#UApxm2?`N@_cmNZh~W~?7B6<+Treo@UBl& z8kQp}^~3Flh%32)@VDzk`aq{|j#FhmXx|Jf11V70O!{wFVNI63_>H0Y5vPeq^>@XCLpl=F=m#I(H0dpY)@aD%dB2jN56I z>2d^jD+3dq%F6ADz1NYlv@W(4BxQaAXb!k@FUS0w(g;n!M1BBt1oH+P+9psK=ayDR!#*amZ|Ii}wCY}2?@a6WR>^($@&E54>dM&Py1bied0#u+(*EeAwa>_h z4}v?TAv1CX=py!7e&>cFJXPq$&uHE!0wy=W3F>_NmHWuT(8o{a*Cm2f#D+)SjJ;in zvhqUc%Xe04RrP7P;cef9UCHB}dpBg=^68 z&m$}9xGo`g{bu_-g?)IdCKAPDU@gS&YEHCwT z#_il0dYB4-M1JwWr27Lv9Zl6{)~#YPemxc}Y{f<<;w}YJ3o`$B-}+SJ_pkZCS==Ss z3RJM{h=zuARpi+-g{)0`l^0@Iy`3i{fkSxK#~1SSJ#f360bj#*H_Wzr>t9n%{kW+B zFHy`Je|{T&a!o*6NDUG~SFdijzJ4`m$ zTZIh*aAz>dR4p&b_MsmB^fTOf{l)$rkoO_T9dPWxo%ZnqP~_5MS}HM&BafZ3#`Tck zYXrP*Qntauk&T%sk+VHNTbN+SuSUa-_^U^a{+B68uEEgCP4+$bFuT?Ki@lf@}ZA!UW~30#2bPVV z!zNU|#tw&wq!nN-gO6S4j6EDOTo_qh4l9v=i#fs;@o#5ZTl*}oo-2VFk&@+kXlKboe z|0Fs3_jKx z!k3Q3lFC-(59>&+fOj&M}_k4}*%_xqH(FaMwqV$1%S>1i$Y?O(Ja=%Erld~FzXV1;o!RiOT zpN%4XVYB?fsI0?cu^0E(N{8kI`bj8*kP>{R50MMxwJ!uRx%x6r37);2idxA4)-<34=PR&`Fp{wkstuI{LE zFGRj|DzZrxI{$tELbR@`%dUY=A4|tU8E6jqsPmO;PuLgG?&HDOSELcJICtt#8pe*16L9N&Hd1ZCI5guL7C6pB(HzK)`o2X-HklSzLRgCA>2>Knn7NR$nKg|Aqk=H zPGK9pBpLnVt2sf&Ah77b+!esf&rma_2?y8(_FH)f6z$EGw{AuI;^3iAFmm2=P6OLoo5AoHQL-u>?WKjcIIYJ*( z45b=?6NPlL%<5XB3f;VlIu9VwR7lWbvRYk<2hY%TG@@DT<9(|2icN(tmp_X+#I`e( zlqV8*3iF4>)K^&!eC89@(JT~VH#eyZQn|{6G8Ok2k|MZpX+;5!44U)85xiBfyCKGZ zpF!6ekv$RGqC45Jd)dVLb_BRikO{P=I>X0K!>6v7&qcLSaD6qzdkgm!o}eV}h1ZJ~b*(=>%6TjarrRU9 zHV8_g3iC9VeYX%V^q2g)*Q=M;UkN(P&#o;L2NKuT{j+2e@7oPFrG}-s;eKfni&V!( zagGwIwIK)XQuNysaH&-pBr;jzyy$55xBJaPa*bm7&b~Kr7K#x*Qg%%nx4!Q8%_(2e zE<%24Yc4Dk_pBVm2n{7|fyhf|d9&k@GCnQoce#BWwLZ<`yE(|9LSZaG=QiF-P*m@p zg_WT8)z|L(fSY)yc)+f+yo=r_fF(Gr;N$+yoOm$6+us`(tv?xVVUsdPU378cN@p3x=*LGwL6}Ob^s9P%R$d*!c&D;?Uu&D zR&}&}>_b@O1BzcFa#!P83bfc^>ts)>F5qlx&Qx4nbJi|E+}`>IKn++Vz0|Zc=Kwx2 zCVe^=^efsxd7uN%f$LL%jv!*8KG}99VN&V+!@$6QH0_u&G$p#vDc3;@yBffo4x7B& zeqtKf&3Og8mMuR&eS#Y0;W(B)kR_;CDJb|}Tw^)tS4uUHk-itMgLo!AIrfG3r*Xwo z?F|{=;heeJsn>-eF!1v8K&N8n#=Z$Fi$R1A97P6!NTn4@P76c~Z9v2ef+;7RCotu(_tZdjDIM?;i z?W$z*a}}MY;KxwY*sV2rD+rxl;laRz$yd3|P2#g&6rn3+2yTJmRhZvgvJi;y2eQrm zhIvLDv$_aPfuAeqQss&`1&(rK0hkpUSaSL+vJvZF`DkFGLz1KGzBxzGK~8A}a5$^d9n3yFcX?^NqwrHsAXkUk`!C>dHwtw|R$me^{61%FBU(K)Ta{ICuBM=l z548mx4S&2}x|^z-wYw{|_fU&PkQ|@}j&WO$qy_Hd;r4JUo-hh#am70KB9^Y?08e!s zyMYX7PY9X1-Ub!T!nmD*iy>u1J(UKh9TfOR3o9$&d1^x^yp7Ug9;?z7$u0j1VJi$m zQtI#()me=r_M$k-;quSnpti?6ZQ=c#V9yg|ww|u!dU8K6!wYK>scCGYm^c;To~&p5 zaxW^8w|~enchLFadyDc=lkc$E?v8%2o8;CyWath0pr#oc=~R(Ulh68aY=^8J+B&TD z7-whOHgIkf+OC6D-Djv4bR~ChGw-&khyJMMD5+PE94x)X;no)aM92X#Y$5BN^VJkMGEgwp(mUWu4o+C-7Q~7nEGh7 z-j1~ibC~NxDR20)xjd(7pf`>xyHjc6CDIri{#v07$+q4WQ6xz-?-W*@nJg30NKFA( zpK9zASVBzEVJjIP`%nieXzqNQwWH-Dqq7Dl|7Jah54GP(QhvG7-k=FXS#oP$!2OOZ zA6-}aN%~?uO1PIJtZcOJhaObAM}^uhdu9M%rEp9yIpYy{I+Km~!$L0(;@2V8{ef66 z=B*_g#9ms~4^9bCE%v0uu(-gi@Q>(>`^{!ZhJ0?$>d)ySC&L1;I-*ds~67!2LQp`*u zKE=bfz*feg?Es@1dvEaANxHj@=H$FhpidUSI05|QK{D7t&uu}~?RMs>V5hLvk&g`= zti`t&mQ{Mx)8)(jrNsjcADSkJcWz-z{q0{uW773y%^%u1?Gp$$vwki_u*C|YHp^2wwcUC`$~&9v24su+ao2R-=rD2b*oyB zQF=c>Q<_^`k;nS5A$P`gf$LE~6=4Dq-dWAh(P?C<$^Ar|&7TIJm{^5_&Xt@VuilT+ zQUscC0!^=(FcWF8EB-SU%R#3QJIm#_sIDf=!aag}y1ZkYI_>x)xV<|9#*3HS=ia7Z z16;5}YXl3n)*l8Ox2lL+V6(myg+tbttCx&-i5~DHTmHL9c3H?O<%ar$P&Xe-@iuR3 ztuGX*%wfMCB;lT4Bu}nk$-K=L5_JS=?DX3g%*mzTozfUu7r)_D6rqNFr_h?GwNvs6 zb}3n-D6|S!RBGQo{y|&XV8bm?OJ<(sYKTg^WzOQTH-h!tjO_$5>?2p_JUIDjSs8i+pT*?da(s?%x@ON9-?dQ;5 zqepH}2kJF2*)-YbCu4FC@8w(5hxES$^w2yG)@D8cG<@(N%*uU+-oIm{r_y{)l{FoL zq3cTfN6K!EowJsd*g%XuRf*lHJSURgA9em#?ml<8;j?moqdp^Jzf{W>TKT^D>~umg zPP-9&x9iT`adS&9I8fSIBbVsp(=O78b}xu2lHwAiWnr-X=oO0MiOi4WjY>f_R<=&! zJCC3~-#w^tKNZ*~tEsz6hkp4H=ur0)k%Ie?RnCXx4~jAyceN(I<08g;ltC5! zPn6-Mz;IeaBkryE)4xnI#16U!(y(Dt8uYEqtTx=U2IN>uj$pz_msolzAO3e>!v@rx zs}1tlkrCe0XXSY}0^~V1g2r{otiO#!jb>`oL%LbsaShHBUCH6-S1j0nwMX!HHf)t{&GXS)ecJ*f93fOM$Fty-tEU?l}+MBf3_* zeJGKFFe0?&^48NPHEFBw9m*trE#kMmj0(_ZcdOD2Q$fe)Hs$^9p&*84^uBj++H3*4 zgJiE}f%)ggt&xOaKfA&5G{TC!sRx&V5^AmCX?SN0*yM`RKaYk-e;&qSRh)orDJrW-~$ zA;IX2tFXue5cHgCRKbtSYWV;1Pb|*JoNP)lJ9a?NN=~cS8f4Kn(kzO|#nw5ZwEy&J zu9Mz7!^`4ZtGjq%1#k2!^*xSh>$+Z-lY*r2=fwU8v(<2ELdJOw|49IT1Ks`!*(>YEM_5 z{X_t&V@BHm*vWztx=^6U|kM=p=d(O*yA0mXa?HLHT%zUH^ z(p@PKYkp3#Gz2^6u32uU2`8cLh}tYT0_~`Utu>K;Ko$iuDRb3ttaa~Vb|W@St%Hxz zN+iS`0LH<9FZ4uP#1gyVhJQ@4DD=m8r>TPfpWi`qu}qWp?%`IzamEq48b`1i(G%#& zuME5I%TafglLvvF-!t6bOwI$_$N{W%`&e&DNtmcTO&r)O~HJ!>NiQt6)W1hsf5|+Q0w2 zu@nvHoatJJ7y-DgmY{a)#49}lXa+w(d954329;nFh2IXVJ|!+Gl5k?$-fI+cGJGqi zWs}mLGm+brV_UZt3q5A!U{S}bvB`UPy!E(jPIDmayYRp@B(y9Pl-$bgeFiV>%HPC$ zz)^DS@NUlYCrGGN{+~XEVKVQHI$-3ZHl@ec*D23UvewrAV)%jS&Oqngh5h~T4T&*o zlQ)u)%1p_lc5t){(VBBCT0kpCbb z>{5~}>9xdD%S5%16b!nwD|2+pDdmfYL7^(Ei6B-LpqJE!l3&&*3pPtRG|ve0LW+F9WldPXh)&HP$%pJC99QHHIe%_maB0ue%Giv z2@b}h`gFymU9Pn?KAiFgf#(8V+$~hY1ei}=Rhbl(ZIXB0{cHYBu;t52Bw|Pf>f2i# z;x*-znJzK10~$#9d9I^Ds7KE=Vb@qxD@GrIaFeD1ufkE@F8()GgBF!<#@a9+v|3pc z_bEGqxb0Qj!ClFl5Tq0os5H6mm;H*BD>rig-@!!V&DV|XPY3{=QudF6a_k(F{!e_( zQ@n4<2jmmpl1A~T`)cC7VkF;A_>S9&?%}r+kE~AHqWR`_ zJmun}&B@nFTvCkw1WPC4aCn=s4a7)gBf3cA0b$Y|=ZLeJoB>qNBO;P^Z!TDx%e~P|9`krsg&KN(pT7BlJ3Or63by* zC&}%w3aK1c36+?Pn4NT#6qS(EmXHvuWLA?L%K=QJaSIqqQWeE9kO51+^9{W?9b zi{U+QAgf}zX|`O%g#>NIO@E$uI{uE;plAkGw4E0yNj@qZh_%0rWNoLX^GaPQ*~td) z4MH(?V!!XRF-swqFisPGfDqB{JGvKETw%~=8|f@n=nIQ!_Kj~GU`BA|OSU8`F%<`FgyTGVd-T6jR#a*PnVheHPT*1Eol0zc0rOKS0^!iK6 ze&jwhtRUTJ#W&fow>@YTv5OA+Np)R)Mwv}x-oqg?9}<_`?RxT@x~D~AEbHw!=tPve zPu2TtR%wjkh0Sn}i@$vcqQ9zceZ^o@>!5c6Qd_FW#y0x{GyUxc+cZ}zsCttcCxQ`~ zi;_syX<38aDwBSZefPC05JAT@@d_;s;iIz&7?qOIr9EL$6^ z{Ip7}j*o7)Bp&lTeimrCSBWR>Q;nRisB(xFgEdzqw&$=V7ycADZWfP$#rPLf5$D^1 zKR6BS2Jw#gDJ}4gN9o>5w<$LTJR808Pz!5vc=YKx>BagbD=WlvwQ=Qos@4wyCK2pg zehH4P;{t3z{nGf?+#lY_@K*boUiWBK-%er}J}N$K+S`H&J3~i%P|b5eb5)4)L(~PI zr12%km8vhs^XHsBdlWxW>C${8@GX5JQ|mX9!hN`ixH?YO>sp{qypOO6_v%&%i3+iM zT%X9R>XW?S&h^MPm3xu*Bl6hYP1J!6qU9&T_sEfWVEsO*a*!Xiq_SJ|61A1d3}szj z`bIWfP2;}87(wZ@LvfOyO`&!F6iin0UMhaPnqnC8DWs2e9r~H~?mfHcEB#-+%Wv=p zE(J9N?iUxEX^UiwjMl}h{JuB&?EV|9glPcn>+^UEo5lVl!C!su3?)qAuT&}TVK)l) zQ$her9l-cdn5X*#*(Lsw7d@iHqzIz+DYo9+V{XECqX&+xlCl^0>N%&8o_{iEhejuI zd_=h&i&{zFOv5NmBK|1x8K^y`f;Li&`jGKd<-LJ6`wxg+{J66rE8c}5Y_|;_vXw_) z;3TI(fVmC4+_QYH^W7|R6ycW#_nDDf7*(^+tL$Lar$Y8j)oHP~Wya-ln2jwrAO+p9 z{Fi^Ei^SNLNY94gVj%%(=&&SO!t`n8w`ZiF6nxW4rWxzn(*L$C3fc$_mB&!<=`W1} zuvfx#z&65=eTQ8eFLEw>A$31~mXdv#B2#F%K}FB#Z{b{#}iQaD8$_ zv+{m#t4lb1|3=lv%mrq9JVX36h0vR2H~yD-=&5sZ(lOvaa9jgj0IMLn=dpjrAh7W$ zmndzWa;0Vcde(is+dC?q+cdkb!w){}Wy1zE$>Y3#$z(E!*Om zPsS|Fi90!CE`Rde^32@u_!#`yOh(b9f1`(s*U~Ii+;Cd{U1l^jIBtMb`OZwAFQ18? zoC~PHkbbWjadrxTl4S539Nia}B7PReJZJQ~_ZUpP<0);~EN$9;pXO%?W&D`s|zbTpwu^xxTZ$SYBRe|Bedt|ycS2Wkwdki7Y zn9|t2L7HHSnfxMqbr*0Wm4m|}onNcJ{f>P7r4;<5;wHYVC+h>h|Hw8P?Q@f^$3)m( z2<)0ld4qqEqjR6G>y@@mRXqAJj#V~83tUsfFI`xMly~91pK`Mdm;zdzQH1*xD{I*$ z5IMF1P;}8qOTS0C`e9c^iEdSPeIBIgx=Je7$LX*{{2eoI(z#lW;rA6ObF;-(GW)iy zH`BQxvMrup3ZOH8pjPJfS|gq6{}KWwf=uXpx>Z4^I7Xf(p^CDZzTo>68~J~?O7Vz$ z_zCek&z|n|$S;z##Uq3VmLO`cwc0ea`nIz`Q1Yb-Bo+8oX-U$~*WnN!FD0w5Bw7Xs z$V(3EVV_qN)iV@`&B%M>q*F5eM0edN*`r_FSKUulhL@Csd1i?2Z-jE>HyJiu^DSxU zH6W2s3P&IwQLT(y$@1|iOH?~<8#2sf*)u|-`70yZ1Xi)xMt%`mu$n28*w#W*q>FO!AKI{S6#;+Uoiy z!`4LV#6?*KSAAZKb38;5{({atdXtqZhF%l|k6Tu~U>}X&Koc-44zb2JG6S>;yytTw z(fFdjW|xNJI9HUdiuiLgm47(sM5>=nr-rO99B5NZ2y3stiw{duYb8#6OCfxlzV|=i zRs30V>`*}``F&B?7t&>#YdGOb$73PNk7}~qWHuH&LUZrk>KDAErG{;9>nr)pUYSMR zt8p1|Hn;oGvd8j*x8hdFAoxme+I`UFFyKW&Q$>-fH9(mj%|L9Xn|udUG<8!fxS_dUyw*HCQ~SsSO{a86q8b9dqUI#j`hJ(i)zndWnr z}p%OZxV2 zScib|FBJ|ufKgSG)vf<{yp6wnAPF>6Y1g~`>M4e_b`!Q?z87+88A0p$O-IkdnnHvt zqFtPB>b%(6Xlh$3U1Kc((I`&Zm>RFYEeP|-Nl+Vv$RG8X^aAIci@-;X8EynS%=T6i zi&G_2>Bm>9vL677W&Rh);`aA572s87{GkQw(hK9q41`a^r2VEifsOpxQT#YgT1u2} z0+>#Y3|y+N@F#h451?JcLM;TTxo{{^wSy>Yc}GGB?6p;&4~?-=J3s~F@qLGrp|_xc zKbn9KN&7*c7sT-qdwluv;Rq7{{R!f1xZ3JFa*ayGt@Z!%A~w1LOso92W33DO&Om>j zj@W@NisTsI!8yEe!5J^o#KBM?*KlSlgVaw(T`JNnuaLbYtwFv;9ywWB74-p55(wJVE(M~2`8Y#$K0LJ zacaE511NROll`xo6ZcnwE?<}w^L~nrjGN|c^k@gEWBoI9v%Uq`Eq&bw_T`eRQ&zT; zS0+E<^HR|vCKgfEWwaf_V#@g+2wGN%^Vlc0A(&)KFiIS6Vhw|%6SO9LVzrk3SGu>> z?Oey>oS5b9L~u6iiIM8bHL8Gc5*Bw9ooc|>uiL6Pn=MlHW@6_+dU-JP7~k9fna6R~ zlofBQBiX!HYb)!kL`Amhe#k0oH(Gj*qxtu<7^ybgVJSaNKu%iM|hE28+kwfFUJ zf9SDd>{=P;hd{AG$6ljRL#nFi}@IT>n@W=!iAKA+kXy?9ktEUp;RRHBznV>OpnZZ*PJ= z?B8BlG}hTtGC>XVpz0^;NLx4CmKroTWrw`pXH0gSTm$T5o$^)qBYQ|FlR*DicrSx4 zf3PB%Km^;L!MADe_y}NJ=~dgOd5lHgA=eoYhX>?IGQ7YHNE5Y!Bl&FMzJ(?a$Ux|g zq2kyBMDF$KWPNZjNrUN*mMk?ezN7-jd3^STv3^bm1Wz-%cR zn*1v8w_Uc=IfSL*mRZBpvLxh_rH0`m)o#{62x9-TI{uv`$fBn0prnZ5v$B7^_(GE+ z!3;8Cq`{ljnY!mL=f)b@;~SAa^lmrYiHsmMIeLchHc@~ahPeJkt+L6cHq5hF7Ki8t zdjZpky`l5jJ|fk`2?@2F5<`Kpvp-OqkY9B3_41Bs6;r6v@yi zgdeNqc#Duu0`@&O_SOLBC>VkKQ+$Nhc8?qCt)~D(@6tZ<)%$5l&yBGa3v!#bi3DOr zt+H?n!1y6;tDww6_4WZy^<@x6J3(YVb?5c|Z<4Vl1lOEy7kVXs=*T?-^!7*)dCln0 zJCfU;KKL%ucS4ldCD{9L8Z4Kr^({{W%-?Z`0Y&P-XP%!u(qPCmg$FZiqR09L71$$k zF$gD5U>+=J>|ldc4!t8cyb~bYFN89i)3`D8Oh=Lob`QXOW_JsJz-dOHV~KVPd7{H1 zXIhrNq`B`TeqIeT7arXSEfT9?%~JgWo~Et3k&AvoZf~OX)GvqKF#W+jUcHI>4VW5C zt8M$H`BBu8D6t)t=H81+?)RcxB~(>Q|_XL?FcwQ4H_EN6jKf0$YcQXg^O*dg5?e#`o?6^{=(V$=ANp{+HCaPMcyUIN{#6S`&uK*!}kcA1B!{fpcN4RNpC;>20$PJzL< zZMj4KRbw&i2Y2Jl{+XOD+t>b2`}VB{o5L3Qw{jH;1?tGYFSwrR(RS zHUvjm;9g9_`o`C99DgT1ns_kuy%%mwBXg#RiMddIT7B$V2i@n!$)L!a8rfsxG!dmPzrglgM<>K1Vt)K_%4pDV5SYT#FdihtFkF!ocg0|Zln$9 zueagt5)?kTOeQ} z8>2HuzeqS#JwF@=Xf)dTH7#(dlIoAnGiiGFfc}+6+bKXnMb?&7!7-Gn;wxcCfpPcjudtkNfy7yTAaM0RdR9E|2Za_dZ6VjDEc zvRTX|yuxrR)J9EIa@i~3B4K%+D*HTCMds!#Vz2UHvUcCJyx{K5q-4o_DHHAngGC|F zQ5RXIT_y=8jn%v3=w{q%C4f_J9MeE3uRV>4e?w*0X%LDkm1i+~ z1vfLpAEnju|F(Cf#Cw9XN0%(Qc7Y(WGVClzc3%2~ zwH@Mbx{ymM1rX>5C_8472X=A7x!osY{Z+R@_TXq;_+0Ol7Rj}BO%v_(0c+^ZwvDcb!z5NHTRHg4^ zmM&2>j)xmZ>Jx*=;dQ!H04vQ>WZuCeGwI+qcX7NgQMUDz8=UU+PbTa76@qjjle3_v z7U6tu^&v9-Dc5{G!ZS)`eTe(jn7Z73fPJEmtz3SI`2q4CwAYH*rqS*t5EmYcQZreu zNRmrjyxA|SKr7c2rz6HdGlOZOFirf)N3;|C^yt~7!^;eD{Vv7i6JFX41p2E(>~NY& zu?@TEyL5(TafcQpl?YE}!SUcOQHXr4;ol1K`HC5@HVBS>yv*?4r2(dh-g6f07Dc0^ zb~t-)>~bIljX*u9IuB$BpiZYQ+li*MKi><3kBKPjPt8?{x3e2ki{m|Z#+eh|j=qSQ z_wsMoeOMFb^V2-W4fuY&WagaEOit{`TscySx_32MAGoz?@RzbMkIvD9BpSf8Zd@jbM>B=xD+dq`>xCcew8z8BqhX_d`&KBRu{w}AE;EqY?eVP_J_bdx_|Sgulm(qgY= zWWm23q|9y5UO}oBVi#ITTE4SEQO#bh0Xow5yfYYd*Js4E%@-iPf5ZRb8}M4+Eh7_- zOY`(e=MbyBJ$ob@aDFk8y%33PZ?}lPAcIig8Auds$HX2}7lybBem&Jsef9Nv`Qk;z zLe`Lh%8+DZ`F^?6NmB2jLAsvgcU&Tk@k_K`rIkBAYC1nGqgjLF;11-aq$QJnv0 z6xBM{Paa(`ZbZ0$nmYTDRGi=!PpJ6Bm1H{PYj79eA)tS=F;A$Kuj2F-iz(=ce#mM! zJ%cnB!HvYX-l=%lIQHhSL#pe@$nF)Sh4cMV(y(!sa$g$5j5#?C~an7@UP6YL>8+q1o(-TvVz{Wcxj1HGV4n zdJ)j$_X2@1YU+2iEOP{@) zIF?tu1sh1Dxl#6W4I9HA?`1pGc*|C)YA5>fxw!@vx>KQLmeFD#Bklpl=MJT=5`C6p zNk0}*Z-Vs%VDAsmg=MvQFmN4JI%W|WTSG5~CdaLS{C|%kXO-@UQ+@Pr;oo@8U)qXu zQz!j6PC>^Az3CZqg~e3a__m0#s8Lv2Z|pcfb~(Dd<>&lNpnE9njVLBw)L;u?$6wp2 zDO%kVVXIN3(H_xRX8-J|0`I(fyG+$S^BF+L<`~MzNH}x`?!t4?E$Fe4ie#fbBI7RNK*-1lH2NjkC0udQ0y+Al zmhw+M@cty@YAEdDMDwPm$p*D!bYy{1iY=oywl@pdTYj59?CdhguH%>c&b@up?AfJJ}Eze z>xx-i3+Opqm@(QpS)hXg2vZC;tQ?rCukveH{oi%{aT!2N>6mX%6Ka%2$kwS6_lU#8 zuxc-g3|ep26|=`>CA=H}7MX)Sl}e=9d7R24E9DL2x7c;brU@+cnOXK(7OGd`^~dcI z1bmi)fY4#?#L)jCE@J0c(V$1gnj9yhaT4f_yk$_32wDK5z<`H-GUMR5g-KeG01Lcq-vky&$Y05e{suFy=v)q4UV`&)NKPLw%Q$kEb4ir`sQdS5bT3(#i*Q? zlE+wT+7GsJ#Zhz9w?!?xNLia`WKI2)mu2#WFVcZZqdvxU{kuXmQD+WKaGRx0mePBP zlpQdW5Q4!(xffqZwfxO8I1AsIM;g_)3PQ@9+e#9-_WQjYy+lot@$2F7H7buzP-4)} zO|-Ucz>SW<`FdQcOQ>$j5VWvkgA1H5!aTv=SUt=&W2p+ zi6xI=o^vaQZlrY8g4`2SpxfD<%QLGGMis1#>xKbY19(F9#|zbN?0F^Xc|E&U^lE&E zo4gEDRH+U8^aNVM#=8kIKVhb?YT9_o_jMgixc0I_D|zxZ+4L*H*st8et?t)5id82v z@I!+C{ifL@wSrcThF)!Sc<|YvhVCIlSY zA+OrT{<;Mq4te5za;JVM`#@M$3##9bc7r{(ScfRXU1ln#-}v7oz%5_-cJzadxa{zE z0{V%xRE5cEOS}e;Um|bVN4?;-W_031B!eH*^K?$RIRe$=Fo&r(alOWFCs|~n-1{U# zt$reLzfL)w7GiPZk4JShbXE$HcrJCa2DlEj#|`*wwreAco6yVDGo}fdvlhn8o zm_wpfh(PvLhB`*zSMbaxwHC+Xdt&IzyI#|8(UYg6TgoVLoockMpmNO!fiD2h*ZTvHst#>yMikChnf!6rJz2sIU`#uRE>5-#>6af!pLbX^ ztawDY4(s;gsF#(uZGjT($_FMsOFuP5=a&IZgF(NEBZ1h!xoAIJ{8GvG@V(sPzm>B? zdph8Rip{{htl6%CD+MkV;j^fM82CwTS^5_V(PkvzRUNpsu|CfkQ2{=@;um==gI!9@ z5@1rf`ixvyD&IAm_U(zoZcF#ZWI%2G+2Hccr^ccf`?uOjkFGBK2y6m<2;Uu{*E`z% z$oqXTU!P=62~5zpq7$W$HNI({3=tjbeT(YG2Xr&uyjjOb7i(ZRqDxWh(Dk z{#5q);VioyT>mi5`*99Kf?J}b&WG`yM<>x1>8u=2@lo_EU-k}vk*NEN;2)=G=`+aV zMVJl}d`Gp9LQ3S7Y04L0NndDhP$6D9AqfXO&uqT&=&ZPdd*Fl8F zF2PAx73R-FFE1V{d8y2lG@ZMpthm}1I2JgfC@YTVhdH#1v?PB)1IAX}Te%u5Sk@iK zq6bY9gH~0*R(I8P1bcP;alyl=qFav6tu{FKKT_FidW3KzurX%7(@uMQA(|KP6$GAJ zKvP_Rkn;&r{db^$`?xac>GB@*iYt*9%Th&5Hgd4&g1 z_2=abcKZS*Xr-T2bvufq!yKJFyOazvXc8Li;WLzXEg>dLGJ-D!I6yuxHrv<<*sG!_ z-J#c}cG?`-O$;sA)42(gcn=xbIJSwk%?5BDv+ycZOuqOGo^j$V;WUkETt(lH}pN>29L7ad!c zYp#lhW&Ed$6wX=Pxpl~Oo>j-w^-2FTc%3qdsY@Gs7T;5k=~UZgdZG>BrrKSr7SjSg z3|I|>%)5RgiC+7zd{x8YwT~5WBN%ub2{_6X<7U(objMX?C@0tr_&7PVPhEA>%$!IT zUY;>7chZ;MKdQIjmoK|(wkNLmg|{g9=%`69q|U*DN3V6ic|Hj4dH@zV6?OunBn+|j z=N7yPSyd%n=VWVqaF6wHbB<$Ckwn!3Ung$unwa$MwLOrc5C^ZR$R0Ab700{{RFS#=hk*YvQg>fD;f8 zV}0p)KhkZce*#u;*fs0<;hh}GB(y3J%OMV`tE&@a@dpH>8>+s=2@8`&&F|)V=o?2p z!ss2viYiIRptq}B35CiR1Dwrp!O56UvG{-MXfaBO{fl$s@YR+<)M!rfMbGe*YVO(w z>rIE8@QbikrM!ZGth0J)zdqvwi!_nsnNvR@?BV3W(ShecUmQj9%N1huJNUR@vYFSy zXn)A6gPUBabJ9*8(r6ql+ankH&?`IUXQ()y_KSIP=Z#uKxIrSVVonao$q!#F;ao+* zJbD!$S!5__y;A3~px8wF4EwtVtLg>tK-hBd4&@biMImX`24u&rY*$*mqYlw1|7mcqeF|Z)Y-=Ssc=~)Bla~} z_TwY;?jf6qz!ua7H~(X#12smz7rkER(Zb|?XNA9gR@1CCW*INUw}m`=`*fpBeDyQ( znX|+*E+}?%Z4(Ewg7!usTG&Rbs6fW`eQS=9@hHk0tQaxllAGW+I}|!O9zU&YIYaIq z!}aO!{i!-HMK=D>6HY67FY1G`rQzc{{F6NCd9s{YaYbh++RBunv? zuuk_Kqn7L{4MmDNmv(p$H)3rN3wvZrBW-cC~#c?#%P7z-FNMf_ZT4wy?sN z#Jk?;U3AY4ERQIlSrXmlyQ99(miW>|sBymC%tB8#?M+L^kI}-*5yHtAlmFXF|rc=6OA@ij4c@d%B|4hf%){N^7cQmUkkWf>fU|CT!}j z<>>iH)1;0ysJg`{{Cs(M8l5Rd+KQOTP}B3_e^rpa^xM}rwENrPu8gm-#?T@8x%AN_ zYPtR^u=rS8Wq}y`p}#np{6#7c+z`8v?u zC{XAI=vgJ(U*nI2P{4&QxBf3Bfh*8%Ca6aAF@rNJcuZ_^8K9p`3!Fta4iPQAc6$pJ zyB>x@)+(UbljNW@T6h}e@LSWl3z;mimHYlYUUlk1>znXg%E`Y}53Z?2TL3GQ)=qR> zzPf-s#OWU(^vmM2$>~&^h=bJLWVajZ8!()S1P zTQGdgw9meN=CF1MS*eR`IreFg_!_GT8!+I463X1V~%jFU@{2HZopOvurBgLH+Yh3fD8C_@lfc9RO6)Im)DKQGm26LnecOc0-AE5H>WGsNW(L*v z&Ki{UE>iF}&*QgB7sAVG_6Gu7)a^z~j7tI>wS&Vk&M zzeqFpuJjnZ<3#lYZRwXeG9C(`3HlUxS~ZzzAXYj zq3z=<+<(Ev@kJ42nQVE7_|_tHUuy5dpB#wggyar#RU;(p5qrbxaqB$!_YOoC))eeA z7n$Fg+gS2z^eoNxx4(pG<_$Up-2q+u+i0 z2-~KtMAr9r`TE@KWmTIC5GC&9W?yGktJ)C=tAwAjpDBpA(7PkzHP3>(Emdvpvmex@ zt{6f|C9d*v-7s1Wy^NCITDK0CPC^gABVvu9+*EEF?^VH6%+GK|_B7?Vk>XS?m;2->Wz=xK<%cWkd<%Q?+^=@+ zqOJ5(>Vzj!)b+H{*eY4$9Byy3VE#|R)t*s-%w>p(=1{k0A^6iMwrTfzY&BR+N<&+Gr@6?AHH& z08yudHlg1;Vl@5FfV$l*=!wl{tRdh(o+yb z6XC8Y+(kNSB`>R(A?6Z6M9|h`!}m*_kiH+`qga z%~{@nn>2YTA6U}E87Pk{q!rncs}=W+E7qw1Y#XsDXJO_zcfe9`&b7Lzx9OU$pFWb;JLPmQOTc0_5kuae z%~HOx%iOZ&97VcxnJatl5cnqmb)?^h7_BPJxKA$M#Bo0-dUPaFA7#?7{9dbXe49wN zDhjgFT74WwBP_~(>0Vj8IIV&Rnq1-n7H#|>?lY`wjZHp@x;njKA8`6*va2>FHiF5u zGGQgtmEKk|T0D|Bk+iA+wV_e60ML1hnv9_empy zNtiIzl}oD|D?CfOWRRO~)95~d5O(nds{z zG-|2J*6I0jD@d|<@*hFD4vSmg^Yoin+AH>18f_%}&PuL!(Xy!QilB)hxKy7!#`~+@ z9nEC0?2GYpcY5=#cFXqTFByS<{-zb6kZ$H++Q|HUEj>gB|H6dsrfIg~p0S(Wh<%Rx zF*Ge7%gV|V2Jcx(*hw#00-u#vw2K zm5JTHA|Hmx4^wU;v4`D)xS2jAO31Ide^a&q+RW^Va+l4IPA0!)c=H~wkC^Rtr3W^M zwCWZdcmp2eSSMkE8T6%iov-kvNt5We%z`R*PQ?Kc0chGkSkq|dsf+pOLwzVTy-N_) z2TtLBnvVND)-)_g>2ClaQs!k9S19(#(TgSE95TW7oAkT>B&%{G{N_64RYtma*iusJ z*{S=eFIc20X+}5E66yUPi0t31f<2o}B3JLnj`rQf05t8I4uUUMUCa19-Cg=Uq(hlkNqor`}9);KE#Tfr*SROFrs zeY8+8yOG;6X!#J@-_;Gt+fkMRWKj+sH6 zVduit`!YXfrFGuQe})loOA$W_G4ak-+LVqB#NYVre9d`Ola7d!1n-CV*3v1&=t5Ua z8IyWK_Gyrn5fpj5ZJKJ4NT2JEWOC<&NxBnC*n2;gpmw!y<|_Z{Repg50X-5Q)*L&s zELUGDj?1}DLLi)qt5*aLHe~}`&x}__9O8z{vg6BUxfZh^)*V2 zpGJs1m-}kNpzJXE@jb}z9o~^MqVzee{$-);I=k3m>`nrvd{PM*R-XGz3=rviSwQ% z-HH7GFj}_7Isjqi8$CC(Dm+BvD!`C+=tnc3km_ijT}v#^_V4N?ZRzZ`gVX$OxGtal_3N;^^UJDEGs`@p%V#^(nC zag#N$hjwfCMD(s!?a(q+3-2=%dVWnpPd&b%3XEe={>9;#ir$~4oC)i^;H+KZSx5f@ zcGh#!?fpdmIIDq7TH!9J=Lz*hiGJF+Z{|XKj%@y?i2HeD8y0>Kx}lMrK&MRKT$vro zuZp{fZD69vf}{w%e`%-nzirQpA;08K*29R+$HU6MR27UiD|ihm(s%aVSR+K5dU8ws z+TVlpb&@b3ta`bfEx({$oz>re82V)f8GzTcjd*atChe%a+KOF0gSo`ts>Q3>99FPH zx%VHHz`>10yJ7k)_}xz$l<8w(l-yhbstG4s!rUS-ahR{IB3C!)?}e6N0hVc_F^aKzD_m6umPT>JcY$2p{z! zc^MFT#5XrF!e7DmSaaJW-p>FH5TeW#G0p2N+xlkwze&LP{W6~LHU!98xC;`*9#Hmc zEQ2C%EXucMAQqBG`eJsG9JC?WDasWoPETa=uWF-Y=SNbFBOxDJ_&fa`OABPp6Z1bYIr| zy>LC?F)4}nZhOmtV$Yw{`57DK7=7Q-+;Pnx(pir*Qosge&NEgv|DQ^%$_%k%y1_g* zDLlaDs>JWA#Iu!J_Dkl?&YrOZ4&el&A3dWpISNu5&3lN1m^`BcJR|oCv4^?50QCP| z_Vt41rC5dX%gn^nse5v(4WzpjtQYf_^q~RI@s(3}(Xid4l-^X11DT%tPu0Zxjr#W! zA}_C?+d}%EYm6hqjV2dT;2-$d`KcEl<1fFF3%gcFm)C>G`~a%9{=N{e{-&9(iKht) zr^h0Ln)9#%O~N9X zs6#(hs($zEAqu4{)WA^sG{cozV~rLonk2dXRjFxEx^c}({qRA}TD z=%oqD(l#)T@Qncc1D3a(#&2$VGDlmgU4#5=2#-Ho8Dd*ajP=N`2eG}WzvTgELp8@kGPYrobWw0M&+IM8l4qnV zTS-$c-BbU)Lx&i2bjqv?HNmsz ztT_v$4inh@NsMOqaAl&q0^o#@7PeUHS#->E)@+ruU6F<4=vG8P<<=8lFNV6?8KRF{iO7< zhk7qNt`CrADW*Sj=|7^G_Jyng%fbJ`h;sp^bK@SDY6wT06ybr=nCbwT_Q(@})~) z^Tm|LGxoE*TW7B`>loWv@LgFKVex{lkvsMTS)AyH8wv#Ew8}i30^^S;Am7H-RWkoQEEL3M{GvRq5immpkhZsN`luWjuK3oO?>KA3nR3-B%TeAzO*xE2QSQj#wf* zC11s%h1y~?B`^4+#iY8;=U6{h@u2UVbn|j{p&e;kkY?t!XS0-{o`DwqX58J5;p0(CHYWK{7 zI01rlo0P1p`O2I4qZ#eju+Ip3PXuwExxN)%{qM5!CRD9@DbrU1NJI5aK}P1Ij+WHN z!h`HhQ0>`tI!d`s%uXEl8kkt!PcPZ7s`@QY@gMVc>etqFn$~`J%p6xNuWb{*NUZ5d zn@Z4Ta{psj&RJg1M~VfCPnyuwbF_+-{LVrYLGS+DW_dkxG>RV0;sUxRcc1n&S zt&;m}l}a}xp>kQ}ND-4uHoGOEh$TgCTe?_?Rbno)xyyC#cbog%cQd%GN()H*QW~iuG_&l2>yLYAb)si>DLbfU9 zJPtw#t(y`uY_}{3f&;CEW5jd|JJt<|aF^_Pq3d+VJo_D|ySiO&ROz|1nZHsQ{yeD22$L7;&f zA^d2EtQ-Hv2ntIILq`~oFLa-(Ph@-Z*bbod?DhZHt(%rp93rQRnx{{(-M`n%&St2Cd6pH*9;M~E<dSiLFJN{Gb4B(p74n7{oqW!l5vb-7SKZ~ z@rsmUS6redemo%!;X63_#nSz&-9yxf6bgqg#sra4h$iBg&MD@p>NK)9E#ZRB06NK@H@k<>p9Ihz+Gkd)gOnj9olNMzF|`}+^}(y zyZ0RZte7UrS2gPTG7fp3Ma5Po6?>LpTQC zIDdV4Y5wk2|9fKIj42kej?uDCi0)jp z`@0S4ue|zN~a$|w3;2CYj%IOe31KvJAcsHN&sGA;DX7!1%narGg}v(6_ph7 z8Ey1xrtF!I+Z|;Iz7;^J5msax8?kOmo;XNtW^q|GHSY!O zUUby$=d_WKoYxGPUUtb&b+@x@d=|qn@U(>(vDBx>xBp=mq=WX^s;?_axDo`Q*{b-1 zBqs^1N>l*Y#g$qsjq8iqi*)DLA45PN!ar~c;p|6wi~c4eMU6puh?zM-kFV+4c_S#Z zNQigYL^Bql^5~^!#`8}CfH0s*u~2m?+T|;i6Z>Pa-dlBE({n!h?6|7XT>o?|`j;I!&R473`Zl(=--p}l zEjxa@)w}#(&0^UXu<+XgX!AH7ceTN{>NAd7sg7*}(@(Rw?Ker&ur0E*fa0(lkUbL2 z*~;nHsHMLd4?oMFiu-4hx!FZ|88k*)a!<4FCBax<%RZ~One=MB)OY?=D95{?PtA5i zDkAKLRDDrQv=}Z@7tZY*G$#<%7Or9LDbcEgg8my2_QlMW*$bX){Tm=-;uGM^#NGV> zBO`1j>)V$mRu>DK#MBrmf#%$oJ;QCtqyB+WH7jc@X>ZrrKlq>O~U*qA>2v2*pfa2WI93uP2(IY~?B23~|a6~RnnV6$>Y zi5-=z@>9Irx4C@yzA^ZTdUmi$znvc&nIylUj;}yF#a|as0;ctcr=-J|1yPpIrwNho zD}QarVbye*oNmg})J7x!1rSpww~ywp?ZRF%llRddDXcaF1>gCiexUwlXWDgSM_BQt z{z(0SCELHg4=QMVc%5oReSY{(58~9j)z8CiqS5+xnCCKUa7daJ{Mb|2ACx%1ngnxk zi0pF5%wH+&yRsUc_1C1)yW&KT<@L#ZkA7kz3dUH^cxm`M(pF=P<^ZX0oOljd9mk>3 zUymx%5O4**oq2wtOMlm6Quw^d70xeH+78ro-DwP!kABA&s?P*QCT?ZT`bU*x0jmBO z&NEJnLy{h#U1pfZalj7UL@i_gt~FrSLcSi3WSRS|<`#se>AU5Uc8syQLL95v|3<|h z6nV@juBWB`C%Z9izk>z5=5Nf2?$V-|JF9ozF92HsePpbJwSF_bh`D-Em+AWGyZlMU zurTz(3&w|4jC>vOAEk$ESqodP2r3Mw{d3kxAFf^`t{72xBXdv6&)SPWvl$l!(wbrF zL#p0@GCW&YB(AL0b-AwICVd19im05k(HJM0(xt9Wrx6ma-A(y`IqF*BOw4Sz!)=Tn zh%rugMpf^u3-mqgvKJqUmSVFbhyhQ90aSM<2_QBZ${Osm^rx5&1NZiTMAU1p>LSY1oawA zou83omy58iYYEeaVk{N&9Jg?jK5aHZR~+x!z;~SkcM=HN8nbzcb@ra3H-yMPAWy;{ zUb!+cvS$)cv(;xHUoj;h(C=}iFFWd+*n6LLNC=}xClm??lsN;{v+*rg-^xE)wz4a$$${5<(?deFKF1CayZ&|RHo#sqR zj!|zopUVDz);qu&LSLkWyiylEI4uvZcP%Vur>!5$T?%T17Eh-#qKUl*%?F>cHbE9X z>Pw#f&j<5I@_kA4h86Q;rC{WytZ|FTR(a|Wu`C{P`+rm%gTK#r4Y!5&%1GoqRfTRm zEmAga=`^K~=&vRMU%e#>A`MW@4x(MS+I`wHLhh2`N}B?4jxg3Ol%}2&{f;6?y8ly- zghIfRo7MVBdkzwyk1#)x=h4otBo?I$y>r$V|3SDYqV-nWPOEK))nj=xLvs4j^j6Vp z8xPY}O&N|Q@d2!{9x>DGl)~+^FyYYu2t7ZgXFaLJx%^`K`F)+?R&Qv{LH9)Z=BqjT zF(Z2#B>_y5=($(a(YM_RUbrE}49IYn#n!sOt1$`?{ z^V-zO+s*eoV$Jp@K+h_(%bm_yl`U-fU0v<>KiJURLWa6qM9~%Xx3llaT$Mj7l!eP0 z()Rv1soSiS|86%GSLQ%B6RM#Zu6BR#XhvYZf)%xP@-gjp7Y&t>bj*_meAEl2Y=;0& z5FWA7j#C}U{*z{jp9Pr2_sE`&ta=g=pUq0M-Q*Nwn{V&gxrNm|6C4z zz6Hmo0Lu&qfLG*gJf`O_%9}s1_tuV+uYQKmj6)gZMTZ&5qiScq;kUYU+bP+B+4f@M zRMO1bEz)7fxee^{_JqRxkc9etRC`$QmM*#}ociPm7C+ugmAovGT^Aq?mC{OK+k8o& zDE;lQMr7@zPO{}C!0jz6mMArS!4b_$?>59qw4;Z7sKnI2tXVOJ^4QTiZLVvc{^X0A{1&%R|8mY^g+2Vr#N23S zlTwAd>c~xmHRyf0&lA`NI`P9F%6~=AHXtE$mo{RF_uxym;U3xbQ+~CG%hQz9Mes!n zNERVUjP<8;i+FvF*`ZXYdm3n`^9UzgN9~f$PKdS9c)#Snu~%Ak;|5M=xw@T z(vofY&ffE%i25d^DSh7#Us`4jACL${ogpv>-Hx>Zw}cS4QgQl%PtKi@r`2^JFw8&R zLr-=l@2-O37jN$a)ykaibqH7fgr*O%j3)b{2d{|VOmErsFK5P-c(}o%+yQdyJp0rP zXzji0Jy~*#-~N9YIKPv1qZ!~1-qWgi26@)LVt0*j=_D?0N?fh4-{qIKr#pJ^F7?rT zNHNY?Skk3JBVM9L4x&$-a&XK46fNz34|^$K7wj>GK3qH)MyLu>v8)bO&Hlh zzD6B+8N7+^5`+FR&i7_ownC^c6wPo;PtoN^pYwoe;LlU^*l~7&&|_N1ZO>|{EulsE zHI8t3ILz&oQR$>f9baOFN^;f$G0!6o8-`n}p%7=+=F?oQ+eqyXwi5rwd-wr7*S(-C z%&cU7zA_*I_;|Vo5gljH(hTNS9wL+-;Lq%arwDY=G*~I5b>l|K@k=OyaPKd+r0*vB zV%T#+-9>>U*+BHkakfW)Fphkx-u7+PR%}23f|(MYM<5}<@4~)8jb1GU!U&dYX&;dR~#Ha_tR5A&~t1EiSj`mo!a?E|xXVJN@$Xbje^ktqw^ zE(a^-A`7Mrt__Cu#1lhrZ}(BLY8JaSb4wmi9Y`bWngDc7jP-SK8)h=L_ClD5$ZI0h z3uk8y#l?%kC}-JA+c(YV@W|kClZy`i{Lgc*;-*J!V20~|nTH422@Si2UEMNSB=?kO zD)VSlbK|;$o1CGY;vY4*8_Ba^dPalaXaDrj`sy(8wf7Iew-=Q7r>Nt-nprgaosbGp zbvFb1!gt(X<2#rd!*rh#2h@50hDpvON^L?g<^6(J=kE3X3d8D$6qz$YGZEU5X5@c-dgNJ0GsQu_6ab&9+i)dNb+L63BMIz z9uGLBqP{=1&?*I}jIp14ZGjH7S8P|FtYM$%rm{X6guFT-Zs*VV{ndB{;q#AA|B|Hl z5#1`|8uqt=oV81t8a?;J0oQ;`-rKziS93nCp{)&AXA3aX?r5>#tQ*>+onZT~Gq0Ud-0Tz_ZvJkE!t@FK;^*U1N=sGZtgC+e9Ob z`L`lpY@j^@iwNl8vyy}#k-uuQAKCnggaknb(@G5FqO{Pvwr;C1&fCPzgD?Fn^ELWw zQzEY~{EW;_|4g8zEk<;}_l?+QDag5UcGe@X+QOArz>x@D%Pb9r!Cu&d><>%D3AnJn z&Yal*_X*8!#F3Jl%IghKN*&s}DgEP9#&@%hJ9(_QWmv-6xHa6$T$1`4ojS{Y&v=wJ zcnJScef*w7REz5c8)20aY+?eK&Cb5B(9OGYD(5s#x8EY{Ne3b!6RA^791!}0b)IOv z>HThg1EQ$^AH{E8p=W{psx()GcF1RrLeHQt>SSY@$Tt8%m~dOplv<>rQdqh*^rnM#~!XOuXm{qiYZH?|Mk@-Ay?U{ z+?SYB4OrXLN`LCABgbVLIK_@=dkr@FC%u)EvDn8;F$c`#rS;}hL*g7BC=+O-g`3fY zPdC8^8)n405+ml<*Pu(I%8fDpO`o$9;2lk$%5=@S>1~{VtYX-}(s5bAHf8a}&h~Z( z7r|>g_ZxBSQJ&o^_`Mn1_3lwhnMyIBmzd5?DOenU z3+`6QD^Y#YI#cBjVD_}#+6lfENt#gcyPLhUmS(`6_PLLCt5LM9qU&e3Gm79>NwS!rH zGFpWZgL5H5OLcqyrM{o*edFNF%((xgIrdlO#iY`k`tQ z>$kAv_k0QG*CYb6v9ncI;F*`6F?(JI;)Hm;9pqL;xJn-2$Wy?ByG1v*1K7rh>j;S` z=~SBrax<9Xv2E4<2qtcYUO0a^BfS$arSo?eQ@z`AJ=RkB__O?OWDUq}w<0`fhRQ66 z55mU;t-Z}%->j1UzBg(@l_SGy^VmzP8X00OpSR`uH-%1D==AXm$$RdXD}ihdi=_ytn!-xaXk2#elSv%4MJDoVGa zu*X=aa1P)Q0dDvfC?p;n1p3KJV}8hDV0&M((4SY)w9bcz^y&zETTVEOsFfqijhW+0 z))rnp2ex64X84Vx9ZHS1KbHbti0xt?B4Hmu`eVv7DboAzVa2OOh=ojj+tZfWe+rd7 zn;;$*fCBKht0g~eS7LxgmmQSlTM8g7s0K8|`Kpwe+% zz=H$ffv(c!lhR389K(4L|ESH9&C*(3sg$0bmp9E&-x4iz7L|Hs31G~N^P!MwVo zXeQhH1piDgC4g!d!Qy{~o_<W81D*H*F`)L> zau07s7%YEM)h$zDm1ti45pWoZztyCZWl~3b_`+Goz00k%_MIIA zS)RPmWIwA8?p~FD%7>F3uCF{cUK=iBndt{6hV>qRwZfw}q4m&9-u5gN6K$IV*fKUV zetp~c!#r$x;4;-Q%1(6BEeQW0q%*TlOJ`=lF29}LQtNqwA&fG%6S(A51>hwkGhVYf zkw?d9SC##nVP@HP=h%^kI{?H6`yt`s&i^m$G8+8D=nqI_GQ+AWj z7rR~uXd(gG{gYYITNquaKZ@{cHO&X@0D&p2JC~T*zcr>UPFP)8wmC7`G?y5F$?_IK zHKi36^t-WgkBJBu>dXOZn{EE$Z8u~Ct7DDHyY-!t^^^ZG*7-f|>B1qdFvEhF=Q)V> zIB0pwq1uco{>fVVQ>ogL1N+!){F10F`84CE=jezdvPoAV5289LtS3tmHzsp$i1a6c z7}kE#Vh~-`+^4+Xu%lTMal{ej>B;#vI^kiPKRdRluZChjtc0{prl^-k1Kc-sBP*I0 z+_hh<&9glcl5Fanzsc!jujduwT|`KbQo0CenOTI9$uV^U|TQY0M_!@@rOpJbq}^fBL}l zsihkiM*Jbin~bVRVmUTW)Ou*<8KBD!yP@tV!Trqys*z7?{a(R_cHRS#-94&1HNB++ z>1_9pmy)5_ z9zu;Fsb$wDPB!#LP8?FZW^nW1j|@`WxaR=S#3*+M;K+X(4_V|z>&-obnM?peGjefp z43sU75;7mc{;IRw9|A2a*YNZq9Zhdpmp3i-p|zvXdyhQ?_fKdUMdoFUM~;ONbkW64 zS*v1dE$`&#^aVt9uN|ljJxJLwuwL??LEV~#$4~4?5h(x%Syd@lOU%2;VhnxxBZ(bj zFf~|Wi+zYUh@WYRmTt|0lv0V6)7RIfCDwk{J$ADFymNqHaHp#$POL<%IhuAT{a0N9{KW&Lr9U?45la-fNyJ)( z1^EpyGMd{_4&Ctg=dH3F#JkL>RqLgWBJ25Ydn-Bdx7RF9TO7eIVKk>%a0a+`?TH_a z)Zy$vK&ql6wa2e4D{0)HUw}zB$nG`1onL#H=4eKTtTvd@gHhgF`=In$_uA8z_?_C% zIVAq(wVl3}c>`=6-_6uNC|nK=|Ex^URKEq{LOzkluO<-=d6&*w>Rsw*>pun;uD=LE zC0MvFjO-K!!l5MTlSsAOD>WqD-=(C^u7h~(lwvY20e;#Um0 zh^hv-PXw!ak#Tr(FokeGkVy{yNG~dPI&0E2puZ@_d^@HXd_UZB!Y%@c9SlU%_2vF+I%ibpNpNkp)j}1TCkM>$uxU)Vcs2^q{ z$`{!yvN`X_;_5|ne0?CoEUNm;!+(7#Qa6!y2I<1~vA|T_<;`G|c5Y-_x8ndV&CrHEJv*TuyGM%9OnK`KV)XHmU znvpWZwjWd4;1wX{f>3h@BfeUgD~Hz2C(^V@_+SIec8OjJ^V*aikc;!hEAP~buT=uLf=c@-nd5BsN0@~3RDN!stl}!w zQ62Azug(#hyxsT>PUB_=-?e#TkdGR`eQx!9wGhT_A;t5yRK1+t`Np5uJ`{#nQCbp= zR^-TeFG(^>HT_{B;r&KI9WzPv7kXnK!jsk$x8kexDZPK{W>Qj_>%-dnu|EznXK*^e z4@qpF-mrbY$fZcDv&!Ko*$x$s!ot1TtYP^e_NtN zh-YpLLG&Hfp#9ZeLN8mc&k{#nb(tAZ_YiU;bR0i7yc@i1*jceKp%#C1N~%@q#GKRS zTmO*x$~XKb{f%i@Y9_3;n52b&fE+ksQDWW$Zzw~lRC|1c*=Au-sVF_XX%GI_2h*>eNLG z@q#mu@1>PD=V1WqTyHexIY=k0(eN1WWxmJ9E!q9~$EUYiez@r)w+#?1KoWUmF zBNZ@``0dx#TRvBD=^bA9zO=fIHs`V0uV#^<)KOrNH zO219jqy^u^5nyB$el)-u6IbHLX&a1OLd0dgR|e?oKD>e%)PU7k?wnV-Utv<2zFfg* zApREpa%Aoa(1_EZZ>GNYcn^_WWf7|{PRP{y2M?AXmc4oC3?I}mAWVPO_?AH0cL79n zWW^tW#ww)8w8HneiCSlNfobPf12UrSQr#Uia6+*U;X>4SNm}=qt4BU#Zzmyf%i1CZ|#+=gSF+LjLy9(P{ z%xZtoNM8M9i<{6{LpgO*P50tt^b&bx2ee7Y66;-P&{Mdr)9fUH(yR)fL&xz*Si(x9 z3QyBI_=V!YTWB3?%R1L|O>8I3+0gOpjdpaK-x?>|ovr8InxWY#>y(bxgw@Fgi2hrmMXhQj{D_jVazo;A)o&P}$yhe2yfl z@Q;nQ^>c2qFG7@$MxEWq2rg{eDfu{KgUL6nIA>iM1RT4xwx1Gq;YSjV&TCYU72srqI$Z4(jZg1|5ozT-OSz{x@tq!?xqD^)Zis22qOnbxAl`k zIufLZJ@Lf?>45Wac22EvMFV>3xyHae{VPm+>*^jD@}dB`38AuuH`aoT3ed6pAWp0Q zB)LjwZ{_}bu+xn$?CN2XLF7UB*&ggkmaJf20F_^jmz+Xk>o&aF?9zPES@VTEoyo-3 zasA8bC9&#$Lf!k4n9=6GmH&BzQ2!fE&!$tSKH_=%G#nU_x4>ys5*}Y@VvmXoV)h+U z7A}^T@T?3_PQL>y$FD;F)%UsT<9=|A=bBa@16wAZodi~e-Mv6KqMnzbF|N9?bPY+n zDjW!NCUu6t$qf6QV{2C%|CJ@}T|PlUywJ~EWu}Y&@dL0|#{wvK`O1xZdQ|)O>XylJ za|pI)@cdes8r;KF_NI-etU95%2w}gVufIP{=?84lV$**P6(TYsfCk6xe!0-cn9G`D}%w z=o6}6#5U(1V?6?9H z)vs-km5CifzvscBgz$fu1QDZAvTt&$ozS@@q%K`;c)CHNA8f8cPs;9?{K$Cp=;7D>!EHdgPEEYr6=Iz6(XF`nee=Xu0bgdzyd;G+a`85 zr7lig@$RXp-hhW)q(^9AD!8*-7XF6nZ;$s$wqrFba%2y8^WIArKe11qRu1}W?D^!m z|ErIe?j>Tm`GC*;K&LQy{*|?9do&cfG-==M(K$K3d%pcdsMWn3=Iv#0NEvZAJ=aKgIm3co%JG0L$RWnvC^)+Y z{dWmwNjRTQz$sQw-8L(Q_RR>eci##lgq;RKfYy z+zU9N;-|E4JK&)$;k2ysG;0OK{{aY~ODJGjYp(3+u`G{q+T_iMHXDM@GF zp`=~^9QW!vf5`JsPHD|QX}gX-*Qm7!IrHz(Js?k{^=@l)B12|p*TTAQ>yqOKTrh*e?ikghwNB2b%A(51RfkvEKZ z&)3R|YP756`}_!;$L|Lvse^8;c6*j{{_s72dQrIzBWk!Fi0#{bn0o9Uh^>0X*84<6Qa=?c;O@kB$y@E2g`);7 zmm3v=vQ9-HsqQg#;=cuDE@YC4f5Z4`T0MUp2QPO5%PLMg(In zPUNV)v4^sYUinK0@z&oWc6#UywM0+44zbxQz6brTO&^BEGeketW<3C^$a}(bGfbJ- zeWF?69jS!)m@A=o8D)kmfZkZ`uz8%A(s7U86jgbc`uX87(5~^k^I_P1SJZ>i1nIG~ za^^J>O7Rmz{|ohn8=cbp8P39Hy;0VPnj7l?no%=PTFi7S6N`t`Y90JQ-^4EezViqH z&D0wOr-4=`h2SC6D5Qv={1({x6$ftf4Z(h~Iee>}v2Y#65I;>J z@ZT@)@A&Ey$C)6#_fXJ3FWz)clZ{Wrzkg-5=;`siD!ZF{(o?q0C9(~%YPQuf5cyBY zY-=IYYFM$grMoGLbQzoM|`oCeGl#(G;AWJ`bYY-}`}#S-sEdtx1MEhUr5 z?9~ER@>&!vcNJJp9fCIh9IwouO9bs(X9gnca9N}udREm~WG-DGyOyJ+Q(M9CvGyW_iV%^4rK;D|t1eXE6CKW$lGQ0N~~x z0-NUIy%<@XNCd%oWuI z-qU|MufnZk^q5M{_i5dHJw9ZDsz_0Z@)CoRiB8)ImYPw@7#%-qtjz4#s}lmKJ3vlEkN`!8J=lz z1G4grvsZP}8Ns^IK!^HKiMJMk(N4GyWEAQ6}fVWLMKf53<`7Aw?xL z|FNu&cVeb!Jtewn+FLv8H#_qh2)3bG+(>~vrgZI;NAQcQ%8k9+!+%c~B_=(VrvM+} zj}Pi(UX#zP^0O<4(egq@SZn>Jv>(X zU-A)@!3l^PK$dmP@&`ME-1PX{_5~=v)!D8c`e!!%&7lvw=X~?| zL5x>C#~r6h(&kv1Ys`{X9pta=*{%Og=nO_)uI@b$ETca;MABHwYJvkVP~>WJs$QDR zwAqTq2!W9w;K@CG^orSfeE2MZw{t}`M_yV;kn>mMtJlsf$I4}C?hV@^PES}H{1mi% zc_v~(t#GS9V7&nG>zj({LvZnO7Oe*%Z6XZOEM}sI7mlb`y71EC78|qP86lUVh6%n< zJP_nq&ortsgqh5cmuoC#xAq9qmdV+g;K?6fSwH8kSPj&rV-8^YXHGVD{C|)xL{WN6 zjZ}$WBABf9-7sP!slJd06mS+)@YN=uq{Th*SRR~tQioDc$7>VZJm9!k zPBUZ71>EyZ)NxyLaNnR#ectdf=EUoHevkpsF zA`F-qcaGQ`xu(V&rHPI^A6T@0uTAdJMluwmXn(E zA?j}$P4kIVqt%)BBcAeb6M4v(n=yg2(&P!0i$}1Z46sdQYyMI*o4Z200aaNI?nvkQ zXz`vOg_QSB4S|Z971wOGVNa@2RL`g9i4yI=R@wZSpVu-8P|}p#~aZ z6$g5~%JO;V9bu(kJ4ijI^zsldxHdxc5iw0?oV0bF=DovWp?Bu73ghX&J2V265uTLf z(G)sQr~0)g@RUZzlFi>55$Cx4*g9KZ%2Ehqc=@7J`(GI4-E33`=yWR)%Z{2*Kxkho zy%H11`A{ED+Ggc|UdyCU zA5h8AYfO*45lJC}GLn7|%HNg2J)@>PM;X|4w`Lx#$ENaQx`wzF|6?B_#%$%fa1=$g zmnp(*4Ug$P>3PB-%6{?a)G&j!I3{zueh+Q@jo0wY^7eGycj&J+hp(8`!i1&OUCt1U zx6GqxH(3Mmy#TG;1=Ylmle&P4JJmzhwed_%*a!LIQi13AF~N>@9kyU!#H4=$#DKG4 zcH5}wyfpyrjw-prOLsi4H{|T|8W7;Nv0;$hJlKZ(n2d6m?oE`X+BOh*xAa!niC@Mi zg+50SCvsW8K128WSKNdPa-a_(DM zD{JVb9MSL88&0*I;x1`neI{yDMf>(6jPVMZ-w%*!U)7= z-il2JG3hr^T9h+|O=5ec(h7*6w4AT15wpVM`fE1)$HF>&6#en+^s!42=Hc?y-=u}# z_~F-h3lfMr`;5I^Tya$YXEFS{oAU~ob+_u+7=>rzQA!VLO}_b-QMzx1=0f=frRJk4 zdgVzbeB%(e#I2;`GwGuKdI{i!hIS&vd|g^!Ia2F-!Uz@5_d`9>mffD9H4-A+xh#{V zgHt5WNq|^&cB^1;TBv3c3mvAHP?zw3gk84(WZ z+&?iGPROrV2j(^X$g26r3|Nd2x>Tl)y+xd$E_)qtSosq|K1>^ayPlLJ-_Svd*unWeJ|@)Zl_eLI*;`q)9{m0RTAHf!X-*uh z4M5Sdig)bymChx@BW&+NWRkq~FG<&({eL4RIitBujITpPUZsnQUzGRbiZWSIb&cLC ztN~LNOWy)3_(b&{AsokqZdAdyPYgc zftdL`G7EVRyQZWet0s2U(|n~@W%~gArV%A^-;kgttPQ*xi&5@tK1473to(`}S0!&|3*I%~^HE-f@D`y@4lb@!uZsq0 z-Jt64WaW~_!?)Vb?_`BF%A+shKae5@_o|;0SHT{s!r&-n!3dX1zkynA#T;)Sqs0-e zqjs7+c_Imi{)W6RhGgAkVwx{4xDn5GmaU)dq)Bc}+{4bFo2L=NK(Zas@Pq3D90knd zL%5ai$uL_j_{TxQPHRCTu@67(0+}#GH(vQ-MD-)}xx9imy`10hUKuci_VVvKK2hG^ zh&siIQ5q!5U2M4;YC0liaJ^uU3veSI@+?+bUsMD0Asg9|N!+SqviGXUsV}z1w5byd zXB-;0pg(d(_X#z5`8rSfN47Usl*I-<3{>oI{`m#u(eH=hy7&DM=Xx*rdk?GGYNlIN z6H%0v8h<6g=q(WaYdO%~Zq>|Ui-&|f-qj4be?{@Kql=C8kVyPt@mTg!nJ`{302ooX zmUdKgfOd>zE1m}XD?Y$hwgYSIRi~}*YB%pY^}~hH4E(Cjq_02k9aFX2c)->{bl?O> zv1Zv{%BA!8_A53#KzLm+cKZHrCUf<<51jR6M#bj`;UJcEZhouB+V>aQ>^M1x*!-VA z0EeEPKQZZwyz8tB(JX({@1F0fC(bRbLkkZ=_=TQRrQqF9_}q;r8aUKsX8xVrO1499 z^6U*F=S$Uy5LZ}Tl>)(j$*BiDh1}MLNbYks(eesy^h~}aUuf`6exZUen98RWQ}Q4~ zFAx$f+h_;Lb$U$M8fm;ndmMdT+~4+d%fMNhr`KW3z>wu&g5114hL-=y>eu2+P(@z( zV-u0^{^$Wxa*0)F9B~~jFO^SPD7Lr75)Z8v z`W0zU`U0#P?BaY5`|z;=)a-o=j!l6@1*-|ukK$3M=9Di)i?(T$KEUccYQhOaC&UZN32`Yq~CD`Mka|o$!`cC#gTdbpY8T2931HR#eDN9Y)k; z%EiI$@!|Hra{+dSRkQHrX$Z|oVYi|Uv-uL`XY`+T#343r8*~v!cnhx96!(yr>SN1+ zxMs>Ict56*3EBtG(O8Z}VWj!L$+8`jvD6|HNbknmHc=!_SZ*$c|BE}1@>9NFmIsNu zaRvG{?hRT_;coQOJrX7RPpsmBr(~&Ykr^(Gf4CDE0$z#}y?@T8kxa6ilNX~1*FgUn z&8)-=eX$7s{pCcql~&Zq!gPXz4-v)P(@GjsgYf+EJH6`2uokv29^Z1^KkTq2%m*jw zGi$!=6UJ!`K0>>1G{f*%-VWn_o7kj$ug9GZe=fE8XcYsu*bwygbxK#3v@C7hc>3|H z1$!J~JZ8g4dw>gSxirdLr5uYqzY{%eL{JCykilj%k}BtXg*%ioYxG5^uc4e@%zv$- zG5!y&r)0DLdK~sIagQ+jFU4hqBQvsEruD`{`AbLC=wQRM7d%(-Nr5y`)~i|pBR8n> z#|tcE3PEg!OaY?(m^yK2?0UJR3W#S@r#9OfuI|VDuxm;3p5!|JPk*s@{W)-0l_nvt zZUuG$mkTg8$-MK)4r6kDR4zn64D8?>(G~S&HKf7$NkiM17Q;s|>OAxY67C`Bs;q+z zJKy&?$+FyNI2od^&Aq`l&JWA#G*Dc|mM!YJe z5YJKQMDwvBrf`)`?`}&{um)`8XjH>B0%CPQv1_lSJ1e>u?dlLgOA>A{B@bjXkrs78 z%z5>5sHp@W#70g|&}bu0cjIMu9csBe+tR{6EG;R2i=H3t%OI_0U6LtszXE@EkG*9h)S+PeaXL@adz#DKwP%Tt7Ly!S~lK zj%i9{Un&#h$<=8#b=PvRJ1D<}f_W@IJHVL2SscWROsDLyUX&r=ZEd!TGiFIkT&V4) z;_I#0!SZ%IOuv)Y=Mn;L?{bz?lc$aq#tXdT;(9XS%(7N7#{gu%LQt#b>hU}jf(s46 z%#I-s$m8n1ZZ%p%K#T96cK4k^A6}QV-&^a|Ui-^`JV`DUcu)T5U{x1LuC%5rO6)-q zBCbuso^RWm58!1E8|U7;ZgrWz*1i|NIN^QrShH2T_`1U|ccU%6B`F)u-iq(0ZWpjV zOZ;58^Qw;XI!jWIeku7|=T|Drzy-EG+G#mylyO+VJFC!H(&uZ+|=R5BfZ~hk*vi z@#a+hcd#$bkTRTGP+F>u*uOCj(?c4;z4(dtEN!37k%e{GmWWn>1&5H-jVoapIXDL_CFE8!tYTP1C^Y8 z4u1iybIVtQCntAGz9p}16(b-~CR6_^BMUj#F4z(4ZTu$FZFhsNL&sbd+#`*4wuZ1{ zJe1{S#f^qQ+l1!z_4wJ3kPr}WKRD8MaSvL=t+A$rc}85+vwRo`7~AXI`D7lMZj;;W zm7-y9tpHf~#s8XB@zR9mKR&7Y1~NDPLwSn=rBxPG+N#FJA6g`Kwz2z4;6HF;j=Tej5;F)fO_wcc+Kgd5xaY z^uMhk7JCPA4|OJz{al2A?N!T5TF~r4G_CuX_;WG&uNA@#{OqUNX-;L>lwIZx{pPKc zpMe&&W+D08byK+4>s;-Q>qLdKy!sq555DyICJBD*f6@=ZP_|Phye1vCBP8ecSQXsj z-ng2NcKB=X(RL#w^)cx1O!Q(3rgx9y(71Ny#P+OX5jydT7##k(V!9x?xa8L^{wk7g zZjT)d+Eb@zf@Q7Wnk=vxaoGDDYCYr1rRs}QdZ%K2QbDxO?CP=q;XaGr`IzO~ykuq7 zK%IXBAg;V&6?=!FgyJe)nNiWZx_W()QHQj5&I3QJLFMyZoW*w9nYnmDvv%g+1f$IP z^q0?4)#P{1a%&KcF@p^dyDv7@g#}e$y{J|!#C5TtS*WK8MLKXD`^oaR+b8Y*y`hx9 z{C8%0_TR;MFLztwArf7MHmT?O#^2vY#KdQy4Ym!|U4@0P6paHel%Gr3B1a2}`R{h;;S zO2TqmejRK!Vi(HD*D!)uR@PK9Wo3)O zlw+2rluo%JOsi9tl%`fHR95CxBAN@zQdy~KQn`?&P??!aQfaA)P`NKjuH*&^A}T7f z$SAqDf)W_?)j+Ljz1_0anPdoIYlA{MfYTU?$ofN>jTtA(p-12h^$&7X{gqYS^^tJa3Y`W@^ey-HGj z)J7He&{x$I2ga=mVFj*txa!Nir{V0RI|E3Acw zBXI8MM|@jNXpd0|&D}S0{pS|hhvikC2Eu~XpZq%mP0+4Z+{Bi-1ibO`B+uembN5p} zYSSGV>q?t(Gh_cdpTbL513Es}@f4SX-_^-a)Ip_I_>l(E zB~Rkas?o!xq3z3%@RBcX*S@FPhoS7V+~BtnsGUZ^=)58zu$Tq2NeKBow)dX?`FbdD zXMXsw=4`02>8V8-@y(cF!vZhF!?Q1rY4+J`OalC9A4&`@IO)P-XN_YVoFA(=ws^VJ zFtY~wP~Cc;|6p?iX~M1bTS&DviJhhxt^S^1$+#oI1W|s(J^GrkgQT6=e5z7lk?lvC zd2jTqZ@Z#OnuHDpwqh!cyQoPL6o**(2~a|7{@YP;k)@R{&^3(7_hyy2q(o%+Iihyb zrxNcXAr%%#&?!}7^t%-H+|?ExsG4z3My@SOB+Vtu3al|CA32bot^WMbIfGGkDozwz z`9BCY1gY5?$d^9?#+?fQ;$7JsBmgd8-zX;f6t-7>wKywr#Au!<$o*@Od)OzNRn|hk zEj#1SwU_i>gdQ?*y+7oCclZO@iS#aF{(4On8KV5Lsn(NnUm4GSEv!IX(D32ims@t< zud$DD%XW~io>fK+ehh#*AKWX27XD7Y{)_ayE91&Vu92KQ)fIh_#|rb+r7O3N>Hyu7 zxDx9CG`3kF`V!dq@Gc(R19$bCJ6Ttypg}XgY?T^6M_hY%QI!K3*1T>cRd;Ys@(3LN z`f`Vxe~AWV3&@EjU_9}Y?)db3xOE)p9p%S0!lwrMPd|8Du`DNxHQ)MKX4x)?eX3fm zV@ZSG@hcz1@2DN|tLoyu+^9Y2g)2|}9_vCnTJ_PLV-hPZ&3WRv*1!ykGk-itK*g!ImYxY)SAc3bn;&q$-w>ju9bT?qu%%C$Mp=UJ#POS;oErvcDyk4s4woINm zw@dk9{dbfuTGZ>{Wwf+c9@s{O`OW!%&VNPi%#FC8TC8GjCAtV7GYP@?mEEHhmba2> ztXenU=DwC>Y>w_ddo6`ClB{U89<5*hx{`+37LI2wH|Wx)Al`>8Z-(0NZ_-f~sPnf-VTNk(^{$wf5 zO2>*{mmM{r3-OHG##!|1L3~N&1DY@@=8lb!bMBt0(_GOtzRk53)JOf`LLf!9LkI=V zd)b`&q>onArBdw7jbiXIq|`(c#+5L2gaYbEvQ*}eGYNtw>HF=8wXa=;cKuGa%A3;K|}}i>{{-{FB9m4q!D|` z3ZfWt<}s@g_pW3~(uVk_^6tS0`R}rDlX^iMuY!LygD-k|s5J)L6ZeH`hz_W&Co+yTfjY+GZMtq1OKOiw#g z{A<;PTAoPLWrpa<+aj%fYaB9^e8c<vulXi5A@6#~%iS;d5YsLT3v*QG zPr+7ov1n&{V3;g7x2>STJ#-J_1 zQ6A{9e4)aKkWkd~@&VvSQXt>l^>us2;C))~)4z}U=4=YTd^Be0KdJ9sbQ@8EDuS+= zSNe>sAzmpMBP0Q>urD2(sa^5aKdVr8H2+@A+qH~0sf=W9?r zpV++%2Tadd%SIez4X6wB`4Xm?B?Y&!5Kl4vgYT^@HxMu;2JB*(_Q+)JybqV*OXyRgHHcX zIycNf%@=NEhH)7K*Z5XgaO0sC=*sGFO@ZE^(w{KLg-6}i8Rk+?^dM_`et}oj-auim z^ah5plOe<-vH?x&jE{4d08FeoI(eh~cDZ_+-0b)fy%;5pU@3#lU2fm|ngAb+4zEY+ z$_mzau5$g$m|h)qqTV&MK5!dqhVx?`;)L3Q-G-SzW3FFBdA8NeAmGuz)RR*zE_%m* zJ**&$^iqGNg4FTuvFroSpJymGfr-_eZZeYuFdj}S2w z2$kRqe?8etlAA@{ZSQ&&&bB#C>y19|e=o82h1}w%v1PlWEK0^kXQSsYP9rhYN3^3- zJuf$#HEMEV`gv$Rv6JMxr6d%&9O-69yrHo-R*R*uK9}%~RheN8ehz!l(R)`(5cGBK zJ+O_Vv9-UR@?2#l_5gI>?bMUZ_BNXXD1xqdsT&@e&sLwSp9yZddMf5diMoGVha-Bf{+p>=5znrH^oQ!i&UDv+v77DB zm$YQm#a+m)44Q`z45)~f^F5jRJ)(LV^G}a!$Hf1gm6J}z^~rxkpm;oc;%-UPZIis6 z$}JZma^nOc?&DlwQ>;T6x60U3uzgJKFFm-iX`0ZEef<<9w>G=M+#%Wz!X{#Bhwf5i ztCzDzTpL2(^n#Foe=BhhR09Iz(T32DV~Cug*rC1rgwSX25BmR)r`#fbP~KupXpft| z{H~_|yI`G*i}cbjMC?Gc@HL%z<%G+_Fsb}z7R~*gCgStPMkuDL#|3OIso7QSVnxha zabk{0?Sx|&?yS*^^``I_+8q)P-cz*uJbfcUl5<8xn~M`*=ZKa#Nop7JG)?v^z6e;Y zZWj84Ztn{y8T)b1c$w6BA@5c*>ns$XyBdG%Vj2AW#g7b0LEYl)6%QorW2|{?u+I?DAJ37W!yxyaij5%hjJIF9i|K{+ z;t>M=sJchvgJ5G*k^|Z6xm}MCsgKFB;O>DW+SMTFzPZigUV*DI?T1DWu5!l*^ZCg`DY~WUr*yw=i|=~$-wle*%P4BzOkn5u zsCFqHF1&Ft1v0tL;@I#Wo!1>zgKlt)DGpx@<-S%7do7{wS<9g*jNA3$84sSLufL)N zcoEh$6@6!B86x7pq>(0eTCAneeyNQA)$Y?*aGjw=!=mSJ1BpdF_G!d~L(!|Ce~*H` ztW~}OWUpouUQO)$;6S^+i(N|GfY>CDfj^*vNiV)Cx#BEa#$ zx%zU`j+)qDJF&^aT*j>Ea2q!;WYlxTc0;RJJ>=+K_$T1`bnG*G{h)Nvg>tmNq+CWp zW~(;Ytd3C?Z0FOO@Zoio07Ix%vI?aam8k4DaS{9UK7)EaPm)$GpRt*oDSi7`j9pk5m3 z$6ZCzVQG7_Bu2}amh6-UR?N?h-coGdEB%7gG}?rRPBW-fwLWflnw z-46>&)s+oaxc#*W_wfT@JWB!#IHKGN52$ukA4top$!AGT6zQlm`9KFZw%H$JJWo`sxs)p9p(@%e0bt zX6jy}5#xUIux=!6ZnONJMQ?oPAC=04aXZ)gms%nnJy_T%KZjg~&z6+uU2y-$_`nV@ zrMAZ0=7k2eo^AmxBMm&6x^7Q#JNkj<;6bq!q93A6tAP& zWb4XI{&2~S7H+A`;)HNlah%Tn!J7G^pGRRtn>dd_)t_t;6S;E{Wf?AZ#y4eiQBmJY z=F(H2K+Qi!*_cp1b@^=hWZj~39cKB=EnQ8|>(l0$A*S(O>eL1Ut!7iU`>#_7S44@q z=1A;j;N~{1h_}Jw?{_#4-?@JlX!6n#iu!g~|8V)eSlbBG@^3bg?ECz~b`1`Dq4#I% zn$0EmshzXO5YIsPuFxCP5|m#(vDiizR@39Lg2kE&H*{6VZnTHy$BCwnEO>dcSPNe% zV{uYqcs7=uZRg_I+4@byj+;eRo8m`5PQ~s@1rm-ji0?*{)X=;*zTt>tR9`hyR zR+(OKtt&BoLFDH;&N0dRUgDY-%8IqobG@FAV$==8bIu)#)1U$yuHkdUrs6hr-imnk z{kTu=n<&w}C*Y;%b()HCtDwg`PkrLM{4b&)Vh9LCzL7y#C7!S@$=~7A1sbF&z z2S2is81AL!XU`o=GHe!y9X9kXm53h5SPxk!I~UNa$vN|KHIW=%4(D^;S#}}_$-S1IiO>2+7b;t7;o?PBKm>b}cXH{T! zRK2(3u0eWK{OvimN70lc#!i%kd1{hr7Dy^$+4gp$E^uC;U06D{B^{qySIP3p&cF5HC zjXVt8MYQ9Pz>R2($M)Q(GHT`eV}^lbCvCN-$|@XNogx|Z@i^C8q^_`+wm`^6?N#X! z_)p*3!ZNj6j3PMjyX3klaSG#X1^H4FFm)%1^`GwytXO?40ADIY&Q@kQcQ4nek@I(L zVC!Zy4>ImyY-D@-teEOkh4V@ABhEK0j7#K>hpAl)(g9REjk{V3zPg2mc*&UeT@A_` zTt}=el>YJw@)i(c?DWNK_>E&*Jy<7r61mKXkzj?y=uNF&6`16XW(n z=?39eDcPv=R^K2oO0@sE-$urB%ce$if3Ke7XMD8to@JvALsiMoete1E+o-{VIKby}J}Rh9P#w^vzM@2s~N=2vcP1lX9n5WO|O zLI3{7jMWzDk`y%mTD^(<^Se0S!G)04DE1$tBSH%&?mKTHccr_?NzW+y;9bg+|DoYf;F^~pX6qCj|4Y+|Rc|GKSD4;q z=cXDa$bWGF<;4+4^k6oNZm%J}9E)0FcKOrT#wDP(*Xk(h)%(%W^a<;6Gzcy_J89bX z{aQx~M6%m^bKH2_i3yXA5Od%st-{Nr`=?joo!AjmQLA<(>j%CQv<U z6%PW~Eu`4b%WJ<9r{xa~A6GZIbgz_?S+n%b9JDjg6+tteYHA1yz#MRC^E#pS?P!p4 z*59>r;oK$tgH;B=xQpTc*-m*7x*E1eRGAN8p!*)9qqp*x(k=k$*bU_6K*F1i11}}+ z-G>OfM2}BoF@7YIEctJ?L1z8{{h@9sa!h|&_~R&|f;IA}>8sd!n!Zu^5V;nnTL$UN z!mUwO{TO_|NiIt%f3Vq=CA-6?TbRT$uNh{E9-^2nXvN=T5dM+M1OHGo%-kc9##%Dw z007quKrKB;bGyp~%6=V=_f!I(Qw@;?Gx-2Tiu+=4-=nRmNHsiHcq$o5H}TvFi7W^p zJx^zMVqXZ}xxoXPbP>UQbQPbW6Rb@)0^?)2g-SmneXw^Tj+`Dl@num50;1AYUq{E_ z;Y5j8&FGZ(s$_6BM8y&re&bx0sMLtJ`8xR$H8t=j7-F!j95j?82GYBb#4;9J67Bwr zMnb1k*{Gx-vN4=(N2Ji4Hq;{WL?CO?2e5spZFQrlgy&&~Cu*oko^(4E?>>*SgIy9+ zh29Zrnm@Bk{*z-PIHj$d>2=AJbivE6W^2$-F+0K1fpjv$C3< z6{Eo?3Wn7H+|mG+bjOGH2M`!5L>{0aGx4>aSOZH{UO!vy511*lPtX1|yM=zGC>(4e z3<;Lhyup*nTxXbZ))diX7@L>k=6p{JW{62de`cyD9Z}lk=+Nh-*)nfFoX%sU%aedm zPw`o2five`dafjwCe*>$5X)NT1i2!XnkeeV#IPqct$={2s7in*SO#Od4ZprA;aUQ* zY|sF+90p4e;h?cSY)W)m$VO)pOH@LVClE-E65E7CxrOB2fIw9ZfH%}Kn{AqlA|YjV zF-4*zj+rf5skwmkr8aDvdgdZtACn6qk5|7`*g!ZQ-cBpY2RaBJI7qO$}?EQi@i zN)~_O%l>Riqa+Mw_9y+A&H&afR};c%0^zJ@sU-3e!Uv?#j%%fIeQhSRGR^3~Q^JCp z2*0Va@_B8VX0>VG1y>1gHnjB=SAv}qHyheHH6)ECDoI7*flb?%FAVIt{F^{X{Aj7D zsvZWDDpL*L@|$>)PbiT_C>bN((FBh3>|K0jtAdT-O14T@Qx#rn`ja*mqb62T;!AW5 z@@!e4%8wX{V5@3|XA!NgB#F{ZtDF6|Ou)>XXu1r)694z$sFnVo!^oSY{mj$izps(D zlaK^cui+LfUKOUyUSR0e3ojm49<3dW<`l_VQt6zvrpYgM)| z#|fCPouX=t57UR(4UCG1P-J)E2Hru}UnBTLz(C;IER6$mbf`$G+;Nh<4f!!zK&)N_ z)Z`6e=8UcEd}By%elt!@15IFJNDJ0`pn?5xwsJPcu}ruitjK)s7u2ZREg}y#8Q-8E z>sAXM)vr<;(l+}bFY5HL*}@=1oHvdL;T%L|YeZrONoN=;^cKWzE@Lg>$ej&lnL0e2 zUJiQ?d^3#V_@b}mDbZ=HyKdE0x+C*z+3$d{<`Yn-AhjV1nBmDf?|N~UxJ~>d-c1POAwB9PSeAyF zfDK--xnj0Mrilk()v|7#hOouey9cd#F5}9jOM^tjIhahqC+bDmZ+UYz((x&V} zD{Gp!PB#hTXID@I6qO0UxZAwTPg;ai2PlFnchd1@5`kCW34fsXZt%#)2}BXB0^qjs zlGbSyQlxL};cetqnFG2B9RNloV^_Pna%1|lL)A!0CmHCl-UV@)P`8W zgs6&8VevM$i**n;t2xHyN(Y*#+6kSLi*iTPQoR*d;s>&XbVJn>2oadjhC8 zoOln%aJGRgx2fAYbT6NS;H0n*5-kY7VhV7_Vf5R`vbmQ%D@^lM$e*)HB@xOm%JP_! zgZQu+$Y5+*#z}|az469dTDzHxBn0-;SExB9Xf)3$7cO10A(WqQhyMUzjx1d5!>M9T zP<>Ph!+GROo9ZlhHDp$ezYi!>t9c$jF<=6VgIlArLgk5HltztXBP^?#x7_Lcxm?i~ zy3ABtJqH9-?Vy|T;rgI51Yht`a}2g#TB`R3kPANl7seq~t2`Ssikdn}NQ9Y^Fx9V+ zuc1ZQl+~=`oC4+ZlC*HaU4a@kak3iFkr1k9DMez9*TJyOaqcM!2fStbmyX)ACgKeukAyX9sJj(m$DgI_8%J z3ur{%vaplbz_Y9T#!Vw?_p?jK&86bz=OubL)ic{mVg8FofRyFe9tqkh#NG$ZeV|(j z>W3sLn~#sLkkYfj*%KKJs0%?WG&2RXl*B8>rQ@EL$$Nl10>0`TlROw&Z8R9Tm?2CB zy$X4fSO*Eza&#+|dj?WJJ5AU`BZi&ggN30jiX6DnhS;O(hJ1C|V1V_wPoPD{XPVI> zvep9(B3r0%{Hf%CXvENMF*T7#*y3TGF$Nux$)FZ~;?F``y?iF0RpQeA zo9MQCA?Zb_`}n()s7`c?VFmHnc-!g74C&^nJUj^NN404$`La&r5#A5WuWgLm-ZU(a z+3LFBd=+D@G$%m?4pesiWhdqwUO!I!9$jxtg1l=dwWh2jKON;7Q0xqNS%|v_c-EB9 zDFQNhx77VBS>qBb<{jyU1rWg!<*?coAa0Xz;|C8gY5c?<<^k9JLj#E}$Ql3@M4RkX zP@)hacyiokR+E7*+T;CXvO5_PfD7&^aVeXH))QtV zw;ZoGee#Sbv*@VS*Gd~Gb;0-_1ft&>~ z$+$8VT#1ui_Y>dKxkm_cFd(qjA>n1ENa}RtMpB#PIlD}v{ngDH6 zVAvszhu3gz#x;)+e;dM?G#9LP*h#+|%gjOCT>v242zn&Unq#UgxSd|nwjq}PSo^1Z zgx>Txo-vpwo!_1AMcmzL%?_9nRIpo~j}rgl=2hj~PFFy`lY;x{L7~?n24W$WJM#(b z^ffxMGh{C*hxe723!-9B0gZkqL059q^ec#Ex+ZysILo-2A7JV-J4zzcI99^HG;4-v z3g-;SpYp`Z^V^lD%Sclt3u6w>gYu)1@!lP)<@!q(gw;VOG(Oz#!Z(B*L#~mK} zTR=@yFycJA) zk9Y6ptySk|7T%BdAXk*XP5KArC#yF+n*a&_^~e*95S6M(Ai<%`#*E85;SqC-*A9Jy zS~he%OZAPhVZ0khcuc=zM9WsJW=SMJ_yKb?iY_g2<4F8i{v2RCdqDRH=TJEk&rbY{ zmCvuSq2^&%Hg0qTJCL2_<9Z^3LPU(0%xT))U(=qrkm1Hq?pAqW74<3b-LNCDi9r)A zCWWu_Y*F7(*Xq3+yC*%j&hDw{FdTOu5+@5J^HsH(0Ykjx&6l{=gB0n}JC!fn6Zn}P z-k()37W6!~=3g3?WjbXdOc-S2K@=NiI;E*F|0+7GC|^`|5MR#EI<5}4%UvfesPH;N zNtQwyPk2C{xb4m&-G^;vb~L$T|Ke>>1`t`K3{X;x#mq&SIp`<6v44YUz2*VsK06@_RoQMgBmS2PxsWs!YzB9@=79#YFhuOWfrmMyWaUnE5S%uHWZEg|UBpOM7xM+0iqrO# zS7EMbj@2*kCcXwQ2rL;^{KRIHZ^{$82SzhUHRv4X_;|oNkTU9Xgoo0FZS z4W32TW1N_G3@>EP+vSz~3wMF)bDn9K$K>$c@uga4x%iT}m3P#ASWn8y9npP2`iZl} zQ45dC@>v_=4`-RS<6Q^)e@1OUzMUsbySKA={$&+1_+rgfU)^_zH~NE(9tch&9Y^7lb!Uq2G$JmhJFmY>jx` zPP>eX+)IBF(qlv0 zok(j9cBQIGRqc5NzCm!=wIzOWpvx0g3eG7LbPnXV#c?*N-D=^^6uzSmgbSl4L{?BI z4U6y>uaxGzp_X;d=4bvfQa5@@0=)~;tozsfmO5ylx7|uAKg0R66XKPi&Qk}sAo`>& z3jeWg@Ja9C+wxrYSD~NIVW#XF>% zA?G*M<#-xBq~;^xshBM)61_LxrCV`L;^@aPSKR-Q*cr62+)91zM$TiKU)pP?KnRJh zPx#6^ElMl!CuXE3aA z_v{!kHeqR;8Sb}V=ynQt#CgGUni;OL(d=shlA`Mg6Kiy2#Gn={BOt6Co*c)Sehqs!3_1hA4*QuSdbsPC`})}w2K ze~;cz0x`>j*7`bS+=sPsZ@XC19kM34?Edz4d z5-=M((?EODT#%VWdIUp^niNml4pG3DnoC{hcHXkCrBnL%&(+g;!%XN{M^&Z zZ@7-=!vy#4d{cGaFP)G$*Yp7%e2<+^oe(GU1o4-N0dESi1S?WOmu_d%5TsDL0P#eo z*t2jp?1R1jGc83>hhJ1?A*ENZh|tqwUQ_>9?awq<61qf>k(u&j zmmi=H!&$i@wI8O5+O`%7TXWZKiZeu?OFbPwT4i_`d5gGh{K6jeM%eqatkaA?h!Ej& zh4gpPo}ZF$S8G7;X?qYA{J_`7Z_zFIOZ@8_Do5fXNJ+$muS9`LfbZogk~AJ;5Fyj) z=VKT2l<1L@=^2`-ntM#r%6zU3^yj1!-_>`#(~o6d%N8VdS}Esc_(qZoFU)2)_m*LM z7CHqSpz$3>X_vS-w0mA++|_^JN>V^_5`rQ&J*W2%3pjImw~bTmoc4m2skb z_9gAHI8pr<|6Ub6LvmM8Z|HYfqHVZpdNMs!{NF7r@gBmZ3rz3cen>0!mbjc@Kd~f>*@Dyl&oTo`pK^}?zvsEoE||W7ZL6nX z30L%(dOKP5RnjMNsLo1Cs~^6E^_yChgW6x4DoDoPIYx~_t(prWB4%E$44AM|?VYe$ zmp&*7VIGCd{LsIc01-X2K;NqV{mj&Rq!7;bnoDsb+5OP0yh&b#-7lgRbJE%N2XG1I zQqnXf^6_4+`YPHQ9`r8b+eEYsaxgd`SUr@%*t54)_o3;W2p?& zG3Z6=#->DE{K!^YIa%2`0eJyFHsoOF+K6hAKg9;|VQU!2ac0VtJ=*=3ZT2`wM3Ly1 z)Wgh66Ww78jpNjuUl4w^Ni+b*+H@lZ>QZ zZ#9{fr`4%pjW1ixk8qEy6OA*raT7I9b5)AUtwx_oUlWE+syK>u3D&-l{Qmd8TpBft zp8`r;ph8Q--zeTA&{<~iy+Tiqm;VJn=DCj6?POoh-PHz-(bKfm!ZO^1BV(K4`f_-w zVg(0x(c-kuQDh=0&LKY6tV+gsJdAOqZHO%W{-JtHE-#ZE1l{Oe<)lT34`mV;(cOfB z!Y08Nk@1-8e)s!=>#)?pR29W&4bNj2k~ZM3WdZN$g2_&K79z9i(zwNWy)EuU&{kZ< zhJe4zZsW7R+WS@64DA2v>!)?rCkUqa9qyPGeJs>3069*JydiE?>wYb%2r}qB z4^UN)p08tO!{Ranh*V*P4*E@Ux@r&qSc~(hmEnCU>-f+zdH2UEcwgvVs0(WYXDfP@ z0gJ~x5iRV{R*;PM8|O^J=u-6$!=Lid-SES#9GXC6gUxg-V^Hy%#xMshOD7tpF$0S@ zQ-y_uLts0aDj|=PIqL3&SP1(r8EHZa8z@Kh`-S-RHpk{R?1Y;bCAPC! z6-matDVY9Ch@7( zAhx+2?b_@qQa$Em;-64WOEV4u>u}cQ#+J1hN_+k>pig3pfVrX@FoZl>8{T26t%t0K zu`%NzkCGDQ_2S#nH{X(K@+x$$=%z8joFR*sI)jf>XdloL@%Lw;{P)Rrw+Rcm|BIH5 zssvy=EnBqEdR|D=WieXDLafn?G*^BET-EGnE*lt53wUpwb>Sv{oA}Tr&tZAP&+&iX zT^2^&ta`V$7Sr7k=D@48dF6$kEV^B0O^-BWO~)=j<$N^ zn;|=1-blEudzw7QxNKzYM@{gjgySSXz$EGUL*Kn-bD*?AP%Gg*#g^nQxX@w@(41OB zDE_$6Y-jMf@Su$W9!@&Jt0Qhy8J?1R(mocaQquX%ZN zXqTyi`J0BrWhh6pI+rUXMM5i1oRbpVf~D)9RhSe3zN8cS6hFptmke8Dv)V*J3E}_O zYy3KIo+OOn#-jc5+Z=B37Ev|7+Z;fwkbl}id)pL>J0plI)T8)WtAU~g8xm(Sd**I#(O4lp zA3SQv#MLN(E+)9LBW3e^)LGsm-U)+Mo7Glgmc#~XYq-|4`EAl}(p`xtD!D`&)tukT zuxPw+8yLqQGjM)%%SFv^x6U2r7bB`)co*DjyJ*t+jo|FsWJaw z?%t`wGnHfej~QB$fpmH5X)zg3D|Hg-u%+5mzy=r73!4MLp6N6o6csER&LNd`Iafn8 z+IEMI^y@Zq^_ql_=PW88q5=pMwi7Qr4e}@^L1=nx3imM8`7*Z1k(vX!GS}ldPs(Nm zYF=i<6w2@DAOM?bBPL^U0Y?pkr#nL9mx|9^()sqWQ)pMG6Ea+UlzOc4%~y6nuj+Gd zn~b=VYzP`Z=!i+H3emcMWQ1rZb1(GPVChTIi}Lyj=8A z>dnifQG%cDqyE)OP>Abc%VqLI6Nn4kRF3obwJ*p}-AB4y{xB2^GdJE{gZvbP#b@Hj z>Q$$_f6`hRzEcHoAG6)?IzuitWU4YZ$s|`~^f^jvZ$SJ=ZI@xgo3!_n)V$K=1f!IOv+ zvXdpm3n&xLa>bs{l>-)PBDfrXMq3`YMR_3IMAxKSEL&Ux8~cl0)f!?2ZQZ~)yw^LI zlqRnt{Bmxb3H1Xn9ShkzON7c58kX+c{E=;@4;U}Fk0!ol!Jkea%mN=I+`*NnQ9xAf znObsWw$zm8{1te=!p{A%W@kI^HR4h9ij>4d`03x`FYYBjiQ6mkQ$C;nttmyP{LPbU zn*QQrx4GWubD13rD)T8O)Oq&RT6NIkcX5`hSl78eh!??a(RrHHPppK$Db=UR`9UiA zBZ;3zw88m2BRTp}j?E#~;SFTX+X?-_={DCa*+(YF{x9*#xRLERh!t0FJCzbd~lsDK*}9+PsYN zAK@?K==Gp(WoyVA^lFQLtLv(+Z;LI_gfHwy=?>6Gea`D}KflUG;+c_yC5Fe0;>1qC zsn+(P9oD0YFji1+4Ej-CkMSGDV7AnjsT4pTAM3UF*h7eM}1_;IgOlH1H6jRh@HULi@57Z63FK*{ zutq-(@M-E>zdQ`RedhR#rFd_T6&`&D@wTt-Mm~p9C5rNtl*}>u+LA>Ae4X37`O9O2 zzNB|^4P&lby*2ZlA)g@cxtnKvS*DD>$uHI&lo+_YDtT}TG+`8X3w3%laDyl0MXy2WUBG30rov*Ad6FxKPGaQ|^bmCIjB7@Vx{z?%uqGtA=N6o*?Iru$fq z`byzC>`>I#x+>peT{*~e@-K!-C^yP}!C#rUNB3@=RE5pbF)}Y5?PLk(|zZswXj`&Q|d-LOq zYI8KWWW*6}w6!D-BIPV3M*u z1gT7zh}6m}TCeqx(BO~PNv%`MMwr(pT{C(xbYcB$OC~;muzPny&pSKj*sI07I&}QSUPv~C_#qRUfT;@gL?UU z=mA2vVM1cv(PA_ir!aR{Q;qJqbp3!LygSmx_4*D?HtQV~7@q;ICUJsTtg#kfXIz?wvImThgp2G{=@G+R~bS zHv|cfmLu>q)vECmYFug3k%V0J=DwCILDr+uEqcc=;-AXD=STtAxps(1dIsoAPZkkthUa|)q2-9A_y^bo`yi2ov51plT9%kEv7uKt zHSmrS78^5L&CbKx{*R(_4@mla+xS{5f3203D|4xorev0+);uAsTxnTST9K)cGb2|? zXeuCwnkzL;@~ov$shL??>V^lDng$=5S zER!`P;5NU>%PlGN04tSWNK$pDEY3$~tqoZ=F&dY&UNu&b49}bLLapICq?wr9m_-8g z9Aa`+u*`p%kIYQyAyAb;L#rjYN$rb54RL>X;e$e>n6QkoYvMGf`rPD8lYC$kGpgn> z`Z$C2KEZknG3e2BNG(a!5JUOS^Kq9An+I1#Lf?=H3qra!SND-V?#-=j!BD|{Z5;g1 z;s+!76Jb{=Z)#T0pTXE*4BA$zrN;!ByB?Z4f7P7l4!Vokq1qt{(4e|`z2IL|G-i(k z#YkH@HRyY@WeiOWPlCt!PK|BwY;G;!F3~hTK|*hVM_SZ2%l-QGw!*!fqx>iV3)fC? zf~IO8)vDsP-gs~$+12n)IcMsFw(-oxAk5dL5jAl&m=c#p3jXSx5Af`vU22UVR862k6TTm?VJ(D_&I*y2(N&;k3YrlOMs3#B$lJ2iy&YRUV*msMA)P_xF+&T4G!){Y2 zD(OjSP>XklUy9&KX0>>;1TUh&4Utem93X*QZGuC7)!$u8S|JVlFC>k>4fGjj*4{j^ zx}>r+cYJ2AZ!XRyQ;Jg1)Yo;kUB)!kZtA`aKW&o<5Q7PFC%9~EJaU{gsa@>A8Myi( z>8={8ax38n^C>UP=E^=dXel(j_kOPX=xs4zw!Jcn zkP0rOkAqd}ecZogHd#L3-k)n@{pLTHvS>Nbv>OC37L1yoQ9V|?l=H6Oq)FrWX~7>* z7yUTyJ#mep?oZk+V}I7uv6EgOaUGg!4+;XVDb~6vu%nd041GjHr7yCP_6>J|mJc}z zU8h9$teew=o5xHXp#a7gCk~m!_Me5DKS`ZS%C$VwwMkRhvx7_FeENJz;$Ks5wd<94 z_A+5PDugB#dsm!kfIMM-l9U086V25(uOqdjK}GL3FB0O{nmpqGx`=m_AGGK^>hB4f zkv@a|6*nr0inux%;@d{rxbX2MX|H5Dq>&><+5V*W=+J&^u|l;Ql7BPC|7VsxJ+*UV zM2$-IiOZ4N8B_JYWlEDrW6*D7pHMQHm$e6{5Lf8^0_s5LUj2XKNZWiV{_~9bM9?Oc zU(@398yA!LaS_C}9L`^;gz(!n_f2Zd#uwW5M*BtkUi&H1@5Z$<#}C>)l5xBv)v;f@ z$-ImhHtlNCBU;lif4_0Y|P7+$QZF zx2q-=$p23KpAsv7hx0uEaA+60!azM_$F`J&*>@Hp6ZURxzJnhtjBG`zW(`ZRECSDu zxz_mm4oK{5HyPg6Nl#N$*p}7?)OMB%jIz$d_^3P-8ynH61e;m(?wWkfSw~bO%P2jqgv5-FV8CkD}lDL-HuE^4O$Tw#&SGqncqJ+q$ z_;^IL#!>%UgM}xVa2di;u8>D{OWCq#?`Nbx8UxfMm>eqG^a}M8`Seuf$E8D1BVM@;_BjlFy?lrZUG!@(UiI$-bTqEl`>*5wlUb(CH~4ihd-NiD_saYz#oA0| zX+p5)=L4A460oLLO^_!C8|@GeJpbqw#zDUw<7F~AbAxtS7pCLyh1rY~mJvNxsvUYk zZ~s$j0~Fu(2>vF_S+`;8JItK{`-jZ1kJbBNX_|ays^KD&hS|corrA)X+FYAGgZ=a0#Bk% z{h+XON_8r1q}*%^LCci3C?*a6v+PMPcd5}mYtEpIj{!94GwEj#^-wQ?`{*x{V0P$m zV2QLLIh`hi6m(9cTi8?VRlbuR?qHyjT(i2yj}2ts;IyA?Q;?S{Yv+Z{*T=5RQxp#k ztI6i-YgX{Ga68DM5qiB5CtATrd}bc8enaU%$1&;+Ldb8r1L9~T3p2kCH&`XGA2DMo zooR)yL%I@hOU5CRzwvJ^bRKD##*cd|a9nrIskHuShA_TM2~bH+(!Psrnt_T&u7uyl zxJ;I`qAyD;0VRrbfghuqwVrV6!)2)C%OirEwb_F;aLXEI}OMiGa!+3`qlCGJYd>mFJ!ACyaluJ75Y+rjBnO+oNGCqo(tS8 z`|zhx2#&Nlq-olN++O(MW_F&1HEIsf@z~XvDN%qgBg`@RqV-D&qRLApsfrYg6^TX} z@^&yh)=?qL2_21RWnd^zaNIXTwyDRJZ6v#Wp=>0X3;N2jy(FDtSlSi!^#XT&e^A8*p08Gdi`WMAXh zGg&fXRtt@*F6lnZb{=$&WZYZv`1aqq`Kz~8=ErJlmu^mW#-QuQ7HiwG8v0+8ecBdP z+N(_kvpYu38?%bgeqd+i&4JdcK^tZ+A!$bZBkO?UBC>NN-C~tXCA;!T+p(70c&Pr^ zE7WQx>@M$bn?QodGu;0l$R0}5#z@u~6Tw_VU~3 z;jL4qaZsjxJ1V9y+p?f>?YKvv?k~@E8sE^`o!x=pOVBHfRMU$Bh1U}FCMgX1L+zzQ zG$(Ss4ycjnZ2sM0Wwf&`m7N~$?0EkvsL~LC>(OT!+-%2ehfxF4G{qF-!_C#!%#(mC z)`cru=iD<3uXz5T@eP{r5JEyBJCqmZwAS2A9=N?a&gL@aAf!-VFcGG1MKCjk^<-l2 zWK&QnopW-rp%BAAT`^tV|GaXEycafs(>SyKS&#EP*9zC9 zujogS=_A}}C-u-%W*;S3Lhg90)%V8NmAoAtCmm}o*8YkS43P>th1*%8mCdqCRj9SJ9}%Jhp?41N4uh?H&E2!i_XQ6BM@Mgy-IZw_LRWA$%-{o z$$FWA)Sn^H_g?VC_4;k;0#>mLr8bRcbGcSwL6TO$=(Lb z709r#qxDLn6EkH#MRq{b?FF(b#27!LIi6s3;D_YS1PP&HA2VHcrUp6P9XFc(xY^dI z4e(7<$B+cuvSL9BjJ8bdY{~(WcqD{T>DEUWa!tgqg&2wat z7BmH({3{>47LZuXDoQe2vj4=Xw9T1{U%^l7`;6{S?Uj4ygd(eQMkO(*%};YY&AhUe zO&V_&8IJq@r8$U+E+)a$lhWNA^Yng%FW+&a2zL{Ha$Pk^)}o2qhs*Q>?L6A{OmQF0#Knkd5T`(hyRz z;EE}(Ok@(_=M&>f>!`ccwx)cT&EjHsU8j9d^EJeSuKB9O1FFfW8*lCT!4%D>Y zTVo7NO`e3|L|8eQR_S|+`TdlY-XKz6w*6JT@B z+}eAB!^SlVm)9$Q1PHwkMVZg|QMj4Q-q z=sknm+_k)xJZ%Y(Xk83rtvneyYErGuC%Z0iOG!j14dWsY+iW?WFomh$zr)_8`ST%N zV?)L~vK95d;mJ>clWm*jn?8e%C;ud0xNXu%;c4@B+}-nQ9@29XL?5Dxz9ckmU^Y>d zv}Rw?mwkrqgB9)Sm$s9h2#%}&V0Sf@_T|QI*rpYTdDkekhTNm?r70Ab=-ZM9l`)}~ z$xjroQdRFlXn+y4HI!+o)bx252sNt`?0cHKwAIaNz2YgI*n20xK-VyS_oa@~{Z2fdeN14AhNMoKJ zR!2RYOvO86HVJRWE?M-&c97s-{`m`WX4ZY6a29Gi4$;iIw~lWM7n^Dp7tWnZQs7sd zopHGzQvW1rz=|5IsmLz_lu0Hy-RE}_+JTow|BZj4iAvxhP>TTW>cM0-*70o++u&~- zD31pk7Vpy^E8Nq~a|)aLcr(UJkj=tdqH|pN6?dqEkp3a%#kA#e{f`vd%nm0PpwOX` zz}1B%ubhHxmp>Zjf4F(oKx_1v#9^uGhRo^<5P9+q-UZkZX!Lx3;wl~{!A9|;4s4u` z_Y|bnmg_1?-U!l2XXG6>8Cvfvo@N|+W?>!)!@fZ&B#uiHfn%LnDnp6W$c8vMXkvO5j)X5Lo_znc!`ppMOb%#trD(XBt34mB)^B7T>O z4UW{N!+Pczu8a9!UUy?F(^8-6b}i=3-&X$MZoOGXD4}$$c~hEcaasLhsr_cFyUTCT z3KmfKU?(RRlkyX9fcIH`*WhltFqCaiy+r8a0wE7%KnP9vr_Vdgmpi1g#mFJCZX~Wi zyOYJ$Y@gaK{5c#=1uRK}X=0+Nx{T=IvXNtm(mjxKw0mR-lCiP*w*f^xF-uXHXjWTE zKq0o_ysRvj6KqlJ&wr!KNY*gg}0V2Oob;ZMDlqUP?E z$UR)s)zOMP-S%i=uxXzU&p)kg`w2ZWbJa-j*azAF@ytO)%5z?F>DFPVg$u`mXrQ zj5UZxf}q_5G3xz%f$jCKZ7^aU0XbTk7H^tE6&vnfVn6?{##e~#>jrf;(wSlxom7p& z{hs_$P9=2bsRlqe8Z!e6mruAtf_(4FUs1F#EKhHgw{`?>MvohZVtaGlYOPQLRyNor zGo>H$SLwy>vdq*0eGrwvAx~O(!DisMY=71lF2&z8SkPhmg@Ysu)!2KZl{rZ9UwVqw z%nYe_Zu%1mMLadx=*1tUjn)oSzMyP@3Q5VEcyCEmL16OkYP0-Vr&rD#Oc~f7l3@bX zGnb$!#xaN+v_r^|mLZBBKvkHPe@>W8n~%VW^hpoQkvxHhIT;mc!!wgHRkI7a?<9lL zNwas5i&G#?ljw|6FM80vavrlT6Ltt%ob-1iK^-)(4y5QY*4u88xJ~ZZS$jW1TvqEx z3eM-YgI#-~Cx6)ldfk<-D>Z#hfLJx*HDEd%IZ2d?3102wo`H~utaF_=G~RAm2i_^t za>u>DDWrhbwqKC7aIS}{Mc|)o!{BWtzuT6mOgTa6rhka z0YaplB}*p75Z(2DhV@5_JH$f30{>xRS2sUI|0VlF9_n7xp`KH85V7QC>@H~2OYM=8 zn6BLRM&Q)7XSB0%U*!);KTU?bpu3T{o&*3l64LQK$--<=TWom*^4(&kwv6dS+gp;! zOeNV)l4mTxX)mOvOQJ^~;r_@pi)Ku|b3Ld1UeRdExG3!{oNl&*6smuC$An|=-Qk!Y zBg194I;^E$`D`&8uv%Fd_DVF@nC@E+R#i%#Iw^X9BRL2N1ooKMZDy z%9U0VapfUI@`N9>nno&VeOJ##PcE1VshJmkNW5vtA^$~k8~@V7C>xz5u~(e+=1UcQL=W27XmunvzW)vnI#eaQl32tl0t}Vvf ziRaa*b%KykUN~y0EX4TDGa0edB_O%B?r~buPQDrV>QR^<^PDM40{r~Hk$!5B-2WwR z6*-`Zl2|8z=?>`8OY{)LGity;eD@Nb-wI7VA@cg2f1wy8^4aB*uc_vuFh?dEo%|cg z7ifSdW5kAxWnbBqj+ks!hiWr)Mv>Q#09XU2z1{|`k_GkFSG4wuKX3GF0e*l7SzBS; zdo#9=t2=Ob>J!>teMzJHuyaS)1*zTQu;dBy4z7JE^&)rHqyhnjbW!d!DN=V(acjnZ zH`G!RJ?9`?309KEVjl?U3Zvg4ZGA!0s+xiW4))3#7?LN9%CRq(;34FH`gXWOa z@pTE=Q%x86?dTGXS$eV;o{hqs}gC?Ij+RZyVk+OESd^(M*+N@%6^VE1Bv$n=>|o0TMWx1Apj#2qG5F%fhA z5E$4=d6Qmfh?{F^SUsEcoHS4DmQ+9=F>p)mby;gL9WUzN<4^X|_X>IlKWiP#8na|O zG#iRRos6TLr0M;e0VQAjBOy>`F6nYteL8Qd`HBv_12r?U75YD=b=T1@X*7gDepl*FSck_+%3A)Wtjyo zu}jsLE|||Td;6ELbz`F}C)RA??${FiF1f9<^)sGnaKHd;Q__otLz9o!VnOnEFFkh% zxtXg+(~P004eFOm=(F-Ks$Aw>@SV(%$g#6rAm@SnFYI)t*waVu-Bk^yW@<6E8*VKn zWhG0$^RJ;fHn|9aoUF`T3}g$EHIo>Pc^ncQ9Z)Y3%mAdz;qRx(fIgs+nEA zhX1B7S#3S-lP&1x?s(F=02uGNnsn+DS3+1l9Nl38l6{2HzX*XZbbG_iUIBUuYg8cI zjRhud7s(>YALkdoN|n%cW{WLa{=?aX%o(z*?I;xcU-IdpkOsd-rVqk>yYG3;TG@~_ z;A?b@bp~bw2R_`ozUUzLe{|x1(Pa0r?bT}D6O4&<->DDq+grY z+&F%k)ov|)!ru&gG{vU#nWH(C>yAhvpyYx)Qboj!wi|e??K{o>3HlS-n+6LkJ&vjc zO9H2x957vbYuE&TI!oFu|8bA8s!~B(6oi{2QuD+Y;8U%6>pLp9uyVhQoxnb+_S`?& zo6o!f@Rw(uhL$072s;Pb({e8n{}`t9`T#*rdbsGu7i^xij8yamJ=q)y*&tjhHWW;a z0UHNS2%a*c2vro-6+>+%G9P?J$LXyR>VVV?O_Q{Yd{=n9^&Ysqqw0z$Xi{1?e_90| zGZt0$&b#r^8o^a=oI$4P1Vt5!@dJ?={YxeitWH?Wc#bOl%fF@E*U|M)lmzzWWohU4KZp zA4n!S!kQp@!eDFng13C)o0rJ40^Ex||AKiL`};C%xpl#VeQp|tl4q>oyBw|xf)}%a zx99IB$Or)CvBmfSzxrlHFYu{>qdw)6yZkT30t=lwcLjNR;}zK&&SS`86cwVlI|^aL z2P3cW&6NH7Kq+xlA&m&z<}slct-BJjBMU0@qdQm;CO=7}mfP8siYFg*vT-xKT11yUABsuc#jcg=H!fM{v!W*+5-8={jT* z-BA)_Ix-Ua3)HQKY=xp5Tsh2k@nx`$@e^f8qiM-OX}fXqn;1RiaZ-x;+lVySHlO|# zHS-z{vWR`Hj5j>}r+wx7xlqsvk}(3n8dn+%ZM5DwT=9F8fuPyM@cybUuani6ED^Yx zyV%(UFMud1mp+q{?~y+W;JL_`ziHnwJ?}8Arc>esFXcO9KyP9P`t_Uz$Y7-%XltAH z5f7@b9tZ3dst6@X#&69S>BpN5JnuzatoBSR^{dfPQ;uW}9@>#jLlB2SGfiD>hdiRi|P=^G;!wfQ2@RSWyH9MHH^%@B0!}`!+(j z=q&A#|151!h?<+oru@r*s%!QNz8OULX-qb}NQWEBW_0vc_sUIgq}?PwRe`=B)oY*e zqT{|-pcaU>D7jtcNK!GY$<%KNNBfwvyni(Mw5`RRUHYZB|D|sxTEmdET_?+uvJ|T^ zzE^OKc^L$dDok#~046;(GpfH})uPB`s+;XP@)bweOQPggo$Zr|68~Qzdz172v#Auq$0^Qy3(=_xY(&>xJ!G=- z8($8YYbUwLW1R^H6sQ@o1J{(-SfIX>-lO^~xn#Ruc)dfqNpRE}kMx)C*`7455^4jP zo20)peiA~gTv7=zurx86bc+NK9t>?=I~k?Zi6>`p98G3#p*_0S#i?EN}g z!VqO_BYplLKeuy@N;aYwitB1(>?&X46MA=CCRt89OLS~$jzrK##py*>a{R81ADQXS zGC~r>NlB7l#T&4YWLtY>@!o;9-#kQR3TT6=Q{gyL1du7-^=Y?f)++3 z$^thTgP`Bv4A~0%QN6PpdeYic(OCyR%eErJx>)~_*o#pl!2C`PZXkIU{e+gMPRaCZ zhBi^%lL6qZ#m47-TFU>YEak2u*$dtb3#!m==0$j?no4M{qIG8VyQAkBml@{sWtsLO zC=WQVIg8H`uo50l!TVvytaqr*^tSS<9Ki$SFQ!F32J^737Q+qUa;HzApti8-^c8Q8 zhbEHft3J{74=qtU=4n@3Ul;Z+_k7n5!{`a_3pQF;-FR%PyGVL37lDYd= zPPJ@P*EEQG8289ca+RI!Idz7lQkK*Xi{Q5tV{Z%cXua2LGYuRpeJ|w4Nr4;Lt~Sbc zlVZQysIUI^67T|f3bNYkTw6Og+SXHj+f<4yE+Kq*HaPWFviY|WfBAz0wj6v_X5fJ= zw)hj}U2}V0?P8_ra!V_eeTw(JN!=pO?u@=|4&wDouZ8c_TfS=wF*->SnU|ndbDXzO zGfzi+x6yw{n<^&ql!nLwtH9#OywZ|0C-(SiyMq0aieS_3uiWnLHiPka8ile}RIne{aZ8*>8kp253Z(&85TclH7r@Ktn=;>=v z@3z4nrAXSu*ZC@r>+ziGhB61{Z(06x`L&ti2Iq1}2kx@oE-12VT~)tD#n>M8ZaG7t z-3JmzIYPQ*r)=>mO1P0zdA(MjjlWLj#`h9EaDLa(b3KY=EteD&eq*kmz2p)jh(o;2 z5?k{c=;h#Sa8IP*)kkUOwL&dSzO$(9LdnZKzdkgS1yvX4(sxUmqZL&mYABs{4|=`S z)(!Vhj?@eHPd^8(c3@BvWS+ss+_;cAbk9;qy7Xf_F`5C;)Om+5N6V;+K3xCtN7&CD zpn`FPg&_17pzx@UGAH_w2iK43%YQ&!KzSTj|0fH?9h8-)Cm*8y+hpVo7aG39N#jRL zEEg@iSzottn`m0Hm0-QH>MP6L~m+KATjNXL3*r8vMjW(@1nDV1aCl1JwloW16%q;&g@3kXO6cmGt=XdED*8J#eaBJ2>%o*qt0`elko@J z>ex~TSPUDG9(J5@lx^Nr6D-Y6bR2n1_%eh{M+HC%8jydFfpM4TZ|d(vy_RUZTNYGz zxV7~jonDXeGv3yxLv>Fa%A>J49YZ1wM_JJLDlf#QutVht*&|Ny8=$-PiQ-Q*F*D8f zg|K)DXj~1732WjR;{`p9Wlps~Tb`<)x>XCmftVtKq@AI(QR#`v&H>qv4sAfaPv2s` z_{bclu4Ld^Ib(w1nj!2y#wKb(uN9Q6-j&c)Al7XD7!460ciJ<$^TFHrWTN(v+5 z*l*LdJ%SKWv4%NLf9<1+LsXCq>=j?$(YfP2*kAhj*%*R(Pvc#jv?n1aE@`gSE%pzMMLDBb zb-yD?tD-EeXvr!mW_^7TonwQ8!PN~BGg_noR3MTfwOv`2;v1=^@kuzU+4YK&B> zBRIj=a4KV=Y50NC3@;ZeDUe5I7vi}4b72UtdGT>1_oX5vXf%9fXkaKNFDk+# zlv_Qi9{?2>bO%Bz$*>x631|r9W!_IapvUw9_9;o}WKg;XTnx z(EO>!EH~RMaj^+Z9V><{Ml~5lO@)H|@na{Xn;ldSF1GfA4$40gpSdtO0e&=ODhlr! zesg^PpPKp1ev8BcBQwU~W=n$h$mGxQfFH@9oAFakMS7l<)M4Wx{eSi~K@zQ~MW$*1 zI!bTM#eCuA%jVaDl1|l@XqW?*l?F#ueI`0b^S7n&+y}k=NHM-NPb@d)A?qN4lS6ul zZ0+2c9NI2z#mU->_fmo&!jmDOJjQU9f14W#hKq^>7j`}_ zoOmsvbgX}luot%7D+#F=7JRGyGEt$hz2WPY6+XGD4v=rZ1v(xt9u@2e1{JPoOiuRi z`)p4qnnb@pha)VE#rUmSA9ckRc@u+cLbQ>(~Sy^jmOXy0i*2gk|fWXP*E$AY`BL_^-yly6r z?5xP`UD*QfCD!N{iThdmWtWXZp3dDFC7yjbD@*CBS6|%q6Kpxz)hO*qb>zG8YK5h` zJIkq>^b#_M)-mfX$l>mrdM4VBfi%{H1YyV@l||YnVF?DPp#JGajav2(CAi8!(}Lr_ z;9G+AHR=jO%}?;l%z{qVPc4|SR%fN_ze&rIctxNuhN_JgMnlU1K^kd+dh?+UWA}VW z*W=v8zf0g=HGu&(%-Ei&8j_y4R6fuFEF* zI-$xV1f=zCeDAg^^Yz};o`n)GaQ=DJ^T@*oOs@CVR90a3Oxe_ zbf%l_%{i8W8j=VFDQFE5P9DRwS#g0F&NnZqb>-&|&>blMV*-sQZrD1&o+;k+ataKi zvrL-7qkv|ZurjO2WF6qX7lkHLnf3;*G|tw+;43YR){-@f3jRLXJiJ31Pp$O(T1zIHnDdg5njOj`=G4Dis@pqBTf<6xW^ivsQq{gTC;yQ zk~jz70=VjxUU^cRJPBqmt)GQo@eRD0UZJYWyJol1&q{jHW}_pwHd8yl;$vR;&>Q^` zgz>}~D##bUTIHbY7sFvKAv0lnt1SN|FBA?WRg~_+@Ml$p_@+EokM2}3I!}8?+pPSb zZNK{dPts6Fi)o|BT{)Mp;oFh;>h8O zc-nI7QcpK_VUe7fTxxT0iJY*~P4ATA%#Z6IO}V6N#0NdN;eN| zI>4=e#U(F8il!>{4wysTBNpOg56mG`zeyT6l^x2zb-8#-Gxki3|4HB10-&KkVN%D@ zBB{ygAsH)C37VH2qgMM5@mEPPsSwQPJQi($8GbXWl1H-f0ZMnLLWhm}^=NM~1i1Ld zB^GJb<|SlafbojCI2c8K4Zlj2vNZ4Vxrg@vrKwCs2FP1^v)!-gdZ|tx&QAVS@*)@I z7JF=NU9gMOXXL;#H?LMs%Gcb<@$^X0mP-<>zq0-m7RhgQRju|~GLSJD^FxYcc>8u;m`x3oT{ ztA5dj7DR1FK3`B>Bm-%YlXr69PcoseGNta{t|k}+PS6&9&*ag3UlgN=LGc9G1zjy| zg-UCqu@xOMhsE8(^a`^7qFJ=UM-XUu`H6h?k5=B8eILhEj`p>$s-zL-DikHX>e56> zgV6NIrt<=t+@^*GSa(M+&}%|KMS2^Z-yttMt;dw81h!nS*u@=c5bOqwJQlYhEm6+; z9`=&%4zEhsA84d>+_@lFs_n*QCk`<{Vs#;Yo9fFzPcy|4bV;rnF;+pj@eca2!iUI3 z?PgkKP;E(E#BlzWq=I>S-l>(%({1Ow6G}0dI1-R@8B#flh%@f*oOfLG#UB4t3G+p3 zLUz~G?n6A(CeH1zgLFfSSHuml3c@3NL1svfZm1i;`1@8Q1KyRK!&nm|llR*eHnzI)2i*{%;u~^u^yue2sb9^j*C8X(#D2)* z<;RTFA=3PW?%I7-kM2Pq7qzZRUo3{RQ?mRP{i~qYop4ahA5BJwpq(79x8DD&{hY{3 zc{QE2X9D+1sQ#b_s~=b-ZCb9pNWw9D64x40H;B*m&vOq-Z*&8G(HOj2{$O|GT=Hd! zd9uSpi=x0FUt*1MsX94XM^$t$PZ4_J_u;JAE3>&tyVSOQ#sm6DZ(Jgyk59u_eG_*R z89~O!lK-jOf(*SHRCr2*Hafg?ppfMiK8G%0pm5TYEpYbTT^@M_xeQ8oh)_F=EY-rh z0H^eRR9L!4kk`&B4Flv^t3b>oZ-zfV&vfJ8A21fIWbh>1Zg+HYuptmyFn~_wzf7c7 zfb$!vqr8xsgS2-|U(`cDmqXGWP683B?e7615Vx%d>l4gB=>?KzV}f|IR>9JFh#1Q# z0{pEE-@+_Z=Ze&^!`r!_u6{A$qpprwhlq~R&?~`3rYgUC{;{enZ*>SisI7M(*^z46 zdj(Ui^}W)4n5JlOF%A?+ZCzRqBL$MtS%RMFkgjVt0m|?oT)+B}n4vd0e3x50j8nLh zWz(@XI=piLo;Riq4gv`idelY2Vj-g7Ie(whWIJ_I&`vN<({lamu9ucW5sOjLDx-)l z;}^=J-8LRJ)dJ340DesUIy{fi?|@wlhIr1G89dd8wg8dS<`1#yOYuS0L6g`?t+(~L z-b!y>oLE+$XZ*Ty?$u{#~b<>?9>#u*deNM)RkD6x{IpldA#e=jj90 zBJWegQ{$WG7I_8Y=BuS)h&lEAH+U7Qhjg3}ttoTlzsN+pjTDO`7sXbzkE$caaGrEI zO!N6UwfCR|7@b)05WSZfQI1rf$fmZn1fU=x<&U9Z$ zMfir7`=dd9p(H3r`XG;-qoaf{6-irT6t?V2h z(rTJZ9~-@GcBj8FuZ&z zdy)JifOSFo?$ilCZaI9$VAQe7pE)R<%m*1M`Ytz@O5t%%pA}RK#eiO4WlO zRr`Cnxl4jGeen%ZMG18!hVY=|wME7dhTteJWl3dRahFf264_pXMD47=aWDd48k>13 zJ1Ew<@7F=0%>bR2fc=|T{yAag+EMR$O%kAz(d|{!f@kN?o_4`(RJgtCh>F4#$oplkhuH0 z!85e{-s;1H(aubBH|F3O?uZl0Ly*khUp}@~j+R?LY4aOE0oA7j=gWmz9!BYNYgXhX zAITE{%c(X47oUggJ7Prs=CeSHX9xnldb|ob@QhowOpjn`eS`Ez1K?8WJ|Ino+}n4f zHzkGO@jE10zYZ2o+|z(Tz%8$&oj zT`l8av@>cZtE_l;rQ^vAkUqb8J+ z3+B2BvU1SwLhqK+y!cBt`nPP^C4J3Jz5-kZ?p!%&&pRn_ryEjVpx4L!mkzaqYSW3) zo4lf^3LP(F04HpVbC=jmgXcqB>U=YMHc8Mzt+ISWL)2fGuRYjbb80F9v@$*X0CM;P z*=GK#rCxnN_~wE5@k%U6^7TFA;zwg{L)1fJ;(g6Ya*jqu-Vv9oz|qA&Y99#WTn0OB zHQMjgE|n1>y@I1)>11pCK>Sl-fLfTz3f5p&sSmlvrZmJuXI(V$B|5_@-1rab^Ltd* zYH)j-a*uW*W?H>Ts5+ny>O(8AkU3Hi^wl2_tZlP2=%Ml1pV`e^=x>m~DoWgKZ8CKJ zQqiq@arl)WHDQ$!l41Q7hTGvu;T}LQ3R^aEs`vNOc9ED5RLgya*gjwt_w@kUT>DZY zO&mLb)`Ws?l;@T(Q?bTlb3W~axe-J;yR2oF8NZMO+=HgX{j6AsnbtZ@3(sq2_v!2M zHOt?U0SX393XDP8E0kb`aFsmbWq7CV`g8u+z=-13A)(4s@A1A~^}3&3xl|v=-!<&h z?kSlx9@8XOh+QOzcH@SI+Vli^;c|FDGWka}ZomA$UqvH_*)m9#b@FX|v-u$zBTag5 z2ncs)KF78_BIFq>i)ehrOvQ|#lfZl5nE1IC>2UdQaC_w`2DC>7?`P5vAlStp-y$qOEvD!o!470i(t4p+uhD$~Wp9MJlPin3#v}UJ zO4_fVYYuDB_f@JKq`MmSHb`|wdPXC$RyL)HVNdAAPF<~Wow7-F{b-=sRVfo1UXB*2 z9Ov|-9(zpbU$5Xe}ltdCZRNp+7y*R$dqf4TfBEWZqZd=5Tqk;lLJiSHxn9FJIwX z8%|;G8qmfJ(=#v;eoe#$ccRGsUihJItf_sY2r#1u!^9N%F7-?Q=G?fXZ0TcCjpjd{Mov+48|g#JnEske}#YT3DpB%Lu?Gl zm%2=~egq3T855(g!o3h+^oqhRf{WMh($1ZjYq_I26=#0V+_hsRX=i2lsTThasLjGp z|0Vm3d3cE(x2LuYQPze^{dCM>eKvtE)8F02g|bab6{;9%cDx10jekPJwWVQpoN;+- z;zW}zlovTV6QOMRg6qaEtssZZ)ZUm|Uu4SWq3x|)Qk3G&rao47C*>>?S2c}j)o0}Y zFQw7WpGH)xPKp{zmxL}|Ce`vT{`^;u=cP#v?Mlp6rSE%Vzpm|JTxvGl=U4rn*909X zLmYNBRr5?*pCut{|CxYiEd5V$`FG#nm#S5Q{k`VW&$LMD;*=`>6=m136rzk+eu^3V z`~rVTwL`hS_ZkaR65skj^MZRl60%^50UGFcGS9A6(G>&CVp*|RAB@Yu1XJYOINgVI zHHJSn#5k*zrFqHcTfbK~!%UIiN#9EG3;8YL=Vn%gA-%Pd2F2Rxk48Tzt$(ARSD+f% zy0m@jV6ID)O+**TG#`UXfZTi&ke>qolm&N#+~}n_%_&s9WsXaERC}&WYsC)jRs1?F z%|A-Y=aTCK3UvuGt@b|dj2gK7{w;H-z`mgnqdh8m-Y@bwu7`5G84sjmfQzkwPfW`` zC{tE>w0W-QX!9feiam0F-~LS12|3YAZ}8(Ij}eX8nt$AwF1u+P`0lf*LMlj98Ge9< ze_4sPBlae{3L;JzeL04&oa7C?&RXMcrqM$q-afw&_960P1Npv8a9s0?Cdyt(s}1)j z!grb**UhV+Ttcb6&>2GI4u7E%)_~s?JfP}bFkmZCC>o>u`wyw5Hww-sr74! zau5&;lb>iW?hwv44RvY8-g)BDGv$ylEUabj2jQE3?LTsNeG5F9Jz*D7&iD`Vg#%e2 z{<&UhI4CsME(LlhHCu;UG6Wt?6qz3RMEd%=OP>s>J9812DJSO`e0x>H*Sxp0j`GEO zjP6M1VC4c$sIoG_w8PoOH1>nX)kpFB^&Y=BeO%&^rd!1^>(wW>iv6QC=or(al|VHN zYdkCcVA>Rz$>s_^=C;nmQtLnUbe;Shl9N2_XM36a$+RI2D+MqpvsjJ2O?(rWylu+n zb1ckjF7+B7IhYdn<%A~Mb*cq+>kKpSf|%LWSUTN;k*Ef)ZBXbEInK&9R0i%G&OPqK z>jkJF8bL15-}3t>pcXWEJo@lk+Iff{^TA7Qq<-H&wll538H!m$tsTqy3h4#B7-n>$ z@Q)F{029H%<*S#paf6aCOMYv#-NL;7`{Df=3}T+*jQTin{KvTwf6^iy8A%&GjH|9H z)D%b+##5O+tK>H-#tt#e^khWS9aU^ORg$ULt*q`2Wf+N?iU`w%&`Fi_jd7ds{EPiA zf#eBk(yoJYwKjC3wSXwTrxgz-1+*SeD6dGqXM22znQOOHFyoBBDlIww@(>Vxqu28k zVvwHwof5&-U9Di|&2M~35qQvT+B0d1Rae9|`1ex{*VZcs40VqU;Q>U}Cm%w({uJ>; zIo3j1xJ33*#tsG@$FI`LDyTmE{ylVx>fZ7E5!FqEb?vrUQ<%Nr&&H!(k2PuN*>Ob6 z1;S5?7^=og6ZL-#oe5jgX%~m5X=Z#IrD9&@`#kQDbXS+)CUj7bq4{5Q$xuKfHVcE-pCF^PKZL_kGVYz{HiN zLHS%>M{4jkCU2)PRS_cn#`tG0;K1p<1h=gLKm>;(ndAq-FUIvnQ!G~xG##W(Tephc zRMSsg{n90vhKINg0DWD@e^1}N-?Ff-D2Tx&TuSi6IyTSG%@np40D?pdfs_{2>yIup zMyMRXuOArmbF7Oas(5(@pLcrbQSwuf>CmLG?6ZS^@SD5W67Qp_eY_P~4;bfejHjx< zpk<3F;~PG7GBy~*Zjkh?Kra!|3M@VdR&Oe`iaFaJ3~Bzg(h?=;toFx;3&b zrDv!%)Us0e#5L_U8h%pjJ1P91>@OGX=zBclmgUPHoDORlwCPtC?F3^x;|lP9x4b&o zZkNR!`!pB+21CR@e+hv+^^-C}los28znjGQK5^WPU(N@gWn84n5@RTSQJON{ZIoNg z7bK69Kj6yQqHu~i?^*+gJ@d~{oJd3dWkCr{hDJ%|$DU2B9Ga}9_26*XWkO^up`X+< zS0CZGUYtr@QJTD6`?<@a@&*R?!C?}78;tedQe`fC0oM*jH|ObhujGzJxofyP1v?+; zDNTcSITN-bWQFX4v&CVZ9t#EbZQJ`nvn=8N6TIPy)l*^_m&ie>;3Mx**e0^poD z=QA5onQ0EaISC_CP*$^wp$ib_FV&d-Je&F4yfd*fwF-&`wO%(|b&&m6s%mexI0~zI zf{v-?Q={5l;@m!KcaOUxzce&PzE5iEZ$8d!XJ}7L2|m-YwtuE6OTyV@xP+8dnd2Ws zjJmK>tqFM6$@DG!?0A#EaU=JZ$-ofor`qbtC+u7Zt^X7{e5Mlk43N$fP!urfS+wUA z|NO_w~dyrw$^{;i!2Vqc8r}Dh?dA zol3h8)7sr#tLXDTQe-t99wYo;nJN6%gFYK{tk_rJ^n4|((?ZDSr`)kSPo8;3e_>`l zJKQQ=+T>PXJ=<-))Z2e-H0dsoSm+&W6mMe7R1QTxv^ zes{o(eeIlOc@@8sj#T08O=ej^U|(!ox*iJG*-s6*oUv2ZywRtHDE&4Fo`g@m zVfZJS0W~mhT2KQ{ArCME^U--2sA=sVqEDu}QDY5-S9EG*3!=jV)5{Qh39 zp03Ev@s?=ogxD_BdP8;d$y1tu^Zv$(4OhM8nz7_b8Qys1H0?s+iT_@qayruNRdYct z`j{C*^bjR6S1$?U#63t+oVT^V-elv`8`6H9G(_uf*q#)8@QLWJ)u+-c{@;5sFBWz5Zs%S{OwVIqGvf}txrer`f1I95LEjIkqbdrSWjbO*LT!;K$i(3A z-_oPl=DFYb{P*|`i!R&CsVe$C<%9Ed-Nlrre%w&%QqADfIp3wfLT|FuMaRBr^Tn0v|} z=)(ZYs*)+vGG!b88H<{wtTDlg>lcI(OkbAR^R+7LC8>wyjU7;FIWbxq@e2IU8h8#gI)w4if3LrY zf{z8MWr4a%i5B@Dbp&_9!7%eAI#`qBsSj2DBUxB0nU>=ITeA9coOactqIf`tM<(iq}P!C8e}H!|Les z>xGo)S(tWwKTRR_Lnn|jV~q=x6+(%BLfU=4;t?)xB}5rBDr7Xrd`RKJhlmOZ&5)}O zAvNDIooB$?O%Y05k%u~p@%DE~p?V0*Unh4R<||Kh$1(QS;FeKc-i z4AdHfWR!Dwh$dt!74@j1UcDFyZnW=rReZ2kB-Laq-Z-N~St~|L@i@C;8Uv&vfV5{0 z(&(AhiZ}PnoIb^wjRY^NVt|k(amyW453iLkO9>!6w~gCr+4#A6lF3Evs7Qi??Ce9M z&!W-1;mTU#Nrd%Uo#z1KV{_h&z8TAOFzpkQUw*i(Pb)~$Mic~53Mss{)<3$i5f_Pn zO?Xlx18C%c=rI0cSJZscJe;0B-x*GH-a$SJL|$}%3T}TfZN=TOYHO#yn}L;->(Oj)``666ULA2o z?;mLMEE!ujHIw4-d9M@@6@EYMkL-pMr~?c-7`Ae5yY>}sNU=idp77PHPVxuPFb$P& zF~TEGa-{ujUd&~!X%B&uw1;V*YQIp*6+8-@TT zASJi-7*R+JwsPJMH-tdjo3>SN8Z_v!Q|S4E0{7B?;yRgFCW8!^ELeodf!yDas)LQqen z+%FYuY|bI0Nit{v;WCQPQGj+J%S$~k5%upAbhQrIPV0fzgYhSMMn9z0B8sAV@8(9P+-HH)ZM| z=e*RP`=TXu$rg1 znKqvt1tZ~4%zCP(BcI0Hg6$COUm2&|D!@VF(zm*S6*&>=x|ny{jq%;;7#|{<5rS>h zA7LYb3>9au*nObv{UX!Vrh`t7tcj0d%9))o$Rj#<2fzLiK6v!qCB196$fVvjh233;G_EN8M ziYg#0WtiG#`np#7wEWu5aKg8FnBN;a2Jq-zo(%p#Z_a&cw`n=>NF?dwBw-8h_%^a& zDGv>-64T5wgG(6lng2aY=+@Lw0Agt6ToGUm4J)$epCc_&rQcO$-Uq3cWwC}7-<^R? z$Xxwg;EYFE=Hy5awv4`2_3Lqqok#k0ZkVmM4T%In2n>3*!{26#w!7&}GqvR#Rch4Y@WZkpE(A#im3Ha8% zVb(eeiZvEmXUFyb6-ncs<4+r~p}ar37rSpj_H9x?zt>$^L(ubgPS*0*8N3&{KX90( zNx$?CL<5A8SMZBm?+$vu{|=9>5AhT4yqQEamrz}yOl0O=bicrpFCqXBU4CTVe4#1z zc9Ms3vXcT54Kpdu7y=!L{>eZQ9b0*p zd0cdC5OPzw)S)Gqg~EX67)T3v@>OcNm`)&@Rx4h_F|$w(-2d|QqO*OVM>iDr3Q&3a zF5s4~d#r{OO`8g379r0r!$cAGC>@s?dj|6zM-JKR2G4%CTlo2gIJ3Zd*nfex%w_J_ zt}OC}s40j3_ujc5gdSwct~3xvWL)WBSk@k}IUM9szlXI8IMr5p8UB?qCh|PXKTegr*_Z0`=e&QCo_E+#Gr-Pj;as<0L_FTe~w9 zHTE9MJG0>;(Q=fh&OL_k<2+({A0${dq}4U?p#8BG4t4>p$bJi&8S|oo zY*=HtP^^%2#qZlhsPmq%KZW$%sJCHfaxQYy4UaZ9XX1xl<;>S_!+m(zpVgABXao-h zb7;&cvW8;ZnmlcaQtp6hw<+=pe2}-NmGcsG5#oH+?-gX%)%?6QWS5eZl|GE0eJ}j% z8~fq*+KEH>5?Rha_i==WxB%wxL}P?A(L#vQbl9L=_^s3nk$55>)j?wXhO3eE7NifR zQ(R7wJ~<58do*8ENPHC)0w%Eq(kc8C-dVZH_8L2&q3PIo&T+JZYWv+g;0PN?V$hU9 zzy2b(%yeRYn!Wfp%moI-|AqCjWvOUzBP!o}PA+m#Cwho4a{DaTesR5>;%u`DE>|44 z{4%a(UM{pajC5rlvi|6+1GS|{9S-A~<5MbKrMKd~WCp&A#V_!i{_7R-_-q_u)nESO ztNHU}Di6%IDUSbayAgpw`gdRjgrTr_MNv$@iDmID>0JvPmui>sdkC9ZN?HGW&NhNZ z!TR0r+dL*A7ysryaM8&q`sKSb_$%WZaYYTj=Zc{9OY_zrh_+De&cg)aPE>uKrGO*7 z)cfTZb1Laico+!yQC`S5o|RHvgWzoNj$R&;Y8#Zfm}b4jX?$jGI_m&ys&v-JfDkp8 z>BYA&M{)IOF4}9VeRrDgT1m#Nn8A=2u16V{p}^e1DcXrbDb)7}D<{sk1NZHJk~;hr zBPZb)r;d-E!AjTn9;5nVIM|}p!v=;XK7z1$Hk_~!_9&}t@OskR`~5Iv%G)CQ{v-Q8J6YX1d-1|L znMdzpWmHsM2c*HZ{*gtB6B82Q6U)gH5fhYzZ3V+3Y2b*a=E z@lm|Ej$W$MZfA+shoiJF>6=uaFA^Ox_gRcAT@o^k@ak^b$3*LkX?Q0AF}|~`ZNXin zn|d~Mv=+c0j4Z?79$q_)=Wju|Ya{rRt^UC*4Cfy7Fs)ei?pG-*%kTu=3Ui@}7W1O3 zq(Z0r1i?AFT~5YgMcPAMlU>3d%SC;Ak)}D7`-xZf-Cl*KS`PWseod6_>#q3%7CVY* z3CzLh&G0wLE6}?boz(8IqVXj0tNN?Nv|MztT@dqyJuIcKviU}(tB|yX7-5&#kiI}* z6#0y}?#_2Tf7dv%(G`rL+2d-HlOU2n23T+o2B_g?}hXmh3YZspz61SuH9-fL~9I1Oe;o5u|_DNam~d&lh*1EEr5! zKm3mYB3m)_h*^^2Py=8dovs?~(E5@u`28lMzV(0RlGJI&GY^Wn)7zOhm2Rmg;QutM z9-#l6;+GZLTeh3q+rD@bu1U%7T6jljh+OEqx!ECPb;Cha5x>kZO~+TAt}^aql&X-g z^NBfO=gE;9dkc)yV2|jO^5pUN3T{u*IuhG^^^jtn;z}jn*qmXip-ZnS>zHA_9~NUy zPC0P%TDAdv(a!JX!@M>A`^}y60GH;$MMgb(vQ$DwJ0*wm9(f0j6w&|Tx^e5yrZd|Z z^avg?t1|JfZp%4*X9|>!*}}hNnT)DiDtz2&asyw08frUAk-DF1)B`P{s`s5B^2Y+A7WaG zXlzb0)L_)6-0==%Wu*JcaL*58@AHeKXAHP`+ItwvHY=FkzNwFpY~cG2<#2? z10523orVzar2Z`SG`wImZwHG@b@btU@$OP0>Lz-CxZnJ5DvVpE?d#Phj*M-`%Npvn( zp40VUKSS#GL3V)Q-=J#TE&O<}byp(%^buNes3Ikdh#tw-pF$%zon?niWZPU89%P_4 zMvcbXCMC`KkX~!^Hmk-Cw$F#vYsB5dSUOa7QftVj`fr|fIo5p!!H=Lu1& z;*5uN*);MyP+@Ztqr;{8P6@x7m`ng@D6LuV-)m*7@7fvB0~2&s0Wf*oICzxD@tN(7 zQ7mqhtsl~FlXGWT2MC)-YV{`p*zIp&Ltvy;XJ4FB_@Sw2E%&%?Cq&agC`R`sj6I=? znAN;YU41F)p?VNFgR9e}DNjq!=Fb1-ci0eV@j^3)-yYJdP3^@=ckm_;BgPG|`p^BO zq4fDvEy0)=9}L+Nrs9RM&KsXj*?ehw7%L~;O2pOrauPKrJhg+9~`%DEp$ zqI5S&1Wb?FHxbVNrs9>Mbu#R5<&`J?MmHN!G5@wMJJXR3KFFIq%UlYbu!Wd2jay=) z;t_Wt!=|29-GlXslwyC9>DxG`>3T&y2J zA?-0FIcZtfuf03~_aCs+EIm(PgjtU&B%)fJO-?6|R0V1W@}E{kjXO0q>?$I-sS_a_ zmGjnP^$66!W3ZQeFnG%rsaQh;zaGY-NO5QtoWvs zZ6UltV}bn;*@^ns8U@$v=7D;cmanvInk+L{2}(7lH)s4$SxTW-_xq=@YXTyd$Ocol zvgb|AygAS0&$%b{_yn_B2=%|g|56-R9eKl=LoxoVcukN=U;;r&Z}hNJST zk(|YGK)6wtX??y~++(un|9lLSPqBxKnje?xhlr`vx$To^vb-=QJo-2go9CGorE=*E z$RFIQ1cqo)(dRfGhLKK$DOdFX_TB!XB(5eR6Te=h2AP@HjhFPrnYIMWwcrpTQ)X~? z{o7tnFGw>x8@h>4!6lX>M&f1e^jZG&QqY>1_39(QL^Lk~919B&Ci;M)KbgM-!kGx# zMqBrv`otOInGD_<3S9eTT0bmGD)n{*?l^+=UCULmuV*P!nVzj3wiJzgoz06yK3x7n zx7P5To}$9S0qgwPChkhbP5lsbTJp{OJ#r-a+;}#knyfU7_Tq0_D>~%HqO6$n732Wc zz>mbZ#CBCF05u2UuDDPlr!xUGC0*ZlX=Q^|E<};5o}0;q5NF zF4(TiQx=8gOu}#FG)xp9t!9p zyo#p#^1G6>Rru3W6K(T4iyHnlW`nkq%F|!53A>n-w8=&I(zqDcY4zxtCdBa^jU(o5 zQ{~Iq#-i@94R}U?YpOBsahj_pYvEi}s%w`b$<2`E8;*Q^(%)<3fp_-OnKIRfDQg{A z*u0Yaso3f;>eY#WOSnm-Z0_av%#lwPR(Y}qqs5Rd$`rdHX0Kl0KjaCiJ}!c`_qd0ZMuvTR0S5MgEbWGj=!Sf~R==QY6TDob)k%CF_%h zsN1ioks=ly<2;q}+k<<+y?VyiyQlLYcW~1uv>3sqX%IjUSxGKM^XkBlz5mN9ts(`D zPy9NmfEk>|GvA(?%B&;=kj+zzAxNwsE!40uq!XOS3owu86?^7akqrrEq~UaCQ=Hmr zcdFS7pMAOT4z@T(bknk2Lsm|nHt2H~r3vhmt~V;lB1lmvyW=rdP@+M*{je8%s@PEorW`{w_=0 zVz|sty^c>kNnfj;{>hM)fgz>oVmo+;;8`s1dRz6SUB^82i=jP8|Hj_ixApaDlAU~~ zK>XP~w=_Ki{qw0(fml?tjEKL9c0jDR5tw$m<(67L+dk#1G3s*6qq*qu+jIR_KGJIM zXS%0G2NA}*V65CUr;f`*>SaC`?_xHf$cddWV97zq1$^W)_+t66;!t@RzDa+Vv!pu= zyop$)oDsXpHxQI`>{JSX%SBWJA>&WL1r8=;tZ`%8z0u$cc8{6fN- zdws|ljJ5oxjxlxpDBZqc(O%Lu;im5-@4;ka5SFMvFLsqvjUF_@WhlRv_di&!1rLlm zTr5qTys3hk*k{E-lY<+yQ9}{YSm94Hz6YuZ|EDcRzDAhyB?0(B8>>!#G@+X4lKCzq z1rnwiM#f9xesb`n4K$E;9c+ z7S8H1ZJRRm>e>R9-6lZr1J~TqxKnvRC5AI@9||8~9CT z`$|=4zrQkVG_L&xhFXw`#INUQ^UBGi>k(Q>Kly`qWNwgKvWIrhpr%N;hLdWib&#wk zJV?AHlr#1?7vlM<+>z8@+8p~)cD5VJ841}t4Y)lbg&e>{iBjxLIi>+`ok~4gDjwWG zCI^z&!{2yUZ5dvB966&9Y!s~@aUG}d9#}BTT2Hrrm`hxaNx9gp+>~%G8h6xCu6^TZ zpls2Qy5o6qVr01(C*L!u+MR={gjd*=u7vIZ_Iz*3Svvg`rd$#rjVgf*)2{MXDh|Nm zz%Gq-rO+St1otp`%Xv}`Y8#0rDMK@IoxG8_4f4B?uS=vge(&bw6qux%2z$YHY%MX1 zuy;Wr^bKgQ%k<#EQKGi;sf$V5^H+Lqd{udkHP^fH)JTV?2D*lkp>tizy?~n_dvvr+ z)DzE-ewd2sKpISoMF)irj*4(*u>7qGu`$yM5mnk|67)pj`&#ms5ZYzB5!Ta?cr%Vu zv4|ZP?z<$h8gCvGISfX75WEx~#81x=>j>y#$Vx6E{f2ItRKL>D9qYk7Z6=x@cKZl4 zTHk*7w-5{AFtE;$hWz^&+zlTNPh>izxD#pmTmGYC9p>fDv&Nx>u}+xZe)x09%o56F zQ)C=F_z`~Ajh$U&4WvtVN^M0mjz_vWo*Lu-L(d@pE%4`Qi(B`wMFQZ~vt$mwxAEFt z1`k3M8#_2#IwU-Sht6imRrt|QEj4j9hN7;xAnO#=)5=s${w~R*>$< z27BJ|-k3wUAz_ydPZ-HOhBi_zQdHc=Mu`i!1Lke-=3sIfGS~+#-rAdzIy8;6vtE#YPilTz>{fz%;uXrRR?4aXX&B( zEvkP0vjv6&J{>LFgqjcDh8lz-5d6d??K-fBMDpvN7*cN(-?3*M@e%a4+S1#fsYYea znMk*U)!*Y8Lm#{+RIj%jhs;{0aSyG;IY76z1c~yI7wD^QD5=v6hJ>AGKBVtRC^fxG zv%}1_FVC@mew9!F`H0z8?Mha^n;F( zP{Iw}NEVCMZig(C*?M{S*fEj3C z&;qHinjlQSdA_11ZQz!s0quNZRY$aSRd+uG*r+~=Nyq{|Sv2WI4&4U)J*zflB?5V| zTs6Nj2s;cDp56#m%`mI1jWA3%LSBVCi*)JRN`XIzRjBr<>Y~uOxC8!gM*UaEG#S&i ziK@|KaNq(R+7)-6K2Pkdj2>%*e`hJ*)TYbdaN*iyHo{C+8<)wV8J)i}`4ncR zvXrb?(_I4ZYFX)JS)xoBwHEMmlU7@<6{=c|-nt<1exQ1Wm85A-K8e2Wkdyg{FeIr# z$rb_iFmJYF*&~?YkIegaE;q^?+@BC$nyS*^TN zA;UUYb0eYd8D7eIC@%er6xu$I0*X(s#M=S37L3B6ZkS<@iAFE`w9jSfzDz zy3bY({&83DwHy-fLW43$sK$s$UV=D4^gDXO=q3M;-`?K`QoZahjfHP&J|X~JF_RW) zin~G53M~j9=?VtsUqU7|9rSZ&Z$$V6`ex}TPUz!%D^KX1x=;NHgOHl|6_iMfm7b|7 z&^?FluOD!wXPSBcv&gpFI*$72h0#Iux2*TvE8}~We;9MB-o)wP0d zh-tC(1$kaO{z|I4<$sNPD{lrA^!%5#9o?7!@zmbbl!;>i2(@lU>%@&p&HesGlOP7A5{BUlm6PC&$88DW*wos_(U6o$Kc4BP>_>pv&Wq ztAu-m@YePg%xH`#;$oIM2*yZHGy3(U%=){>+(W||kWR_En2GhO{%MEj;0Cm#xT6VB z9nYaCUB$a4SfOwloL(!dNJ}NRgqMR1Jv8(_(h5T;hLgIg+iq0Yj&RJpdhZ(5Z{|nf z*P-SqMW-=b@i->L32?*zeGB$!2tlLb_Y0h<4ygT%#tf|gN@6hM1&p528_(va2B_HH z@Fx52!7ZTA!Ggy2njY%xPWj!;0N`e-WxIS(Mcn4EDkF%l7OrB+oDELRuN{EFHM$^7 z!bC)gZ$(UnaX6IJa>U@-@sbly-p1o@os6|NW#eT5b?)L*;#~VLa2O@;Omneb z(bsaEP4=l`Z501uQ}e_b>cwIH-|!)XBi2OQW$ksoLC8Q_Pa8B%{3#SRUwx%jR*c@x zdjczxUPC>_jmA3s67vlG@?let6a2@>o9PePPbsKe18N&edaC|pWgx}1$ zm-Z{>t6eA3A2D`qjas-Fc)@?8oO{#eanA&a>MsPkqf3u%t7Oc*{EkRqB26KEl;#%^i^AG$M9N488EleME8z0NRxNMhkdgV z(Afsd883Xe5eZM0YbiD4G#BMWy~%6VWW2G7KTSM<9WP}#Xj6NS)dS<7=#d8*lZBVL zun??x!ietO*%4vcs>p*Pm(?iU(+W#-A)E16mSJY%Meg8Nb7jB`vRldST57kpSGt+l%4V^4X5sW24;)9cF-iyfek&7Wt0z1YQx z4-<9kB*f&qw6-fZBw;c2<;gX$qqz;Tv(3uDjR@x&-5)}Qvw#51J)F+W6I6Z96uBKu z*i}FBn2m`-nQ@hRH@5n$^0F%8dmZTnDTOOP#m1q(Pi09IpF{kb_^aX0 zl~EaWNa50K=%puMaZ;^jd5!lvpoP1WkZVhEMKjZFV_chB`Rzpc2>A&73g)Bu`~bc! z9oPo6gwtD$aq>|S+|9O}87&=lgDf*RDgvyh(Y@@2t4{D+7zblEP`B5#yxCHRkgo%E z?>s`lz_-udHvcn=P_b5l|ID;3`G`~yRjRDYZS^_c8-)|z#4)jAlE7JD*j#)@WE*O4 zP8h~c9%38zDw#&Ntbk#1u5seb7Mu3MPSku1#7PON=B(3oF+^^*Q`h?8BlXmzOe-Q3 z>zv?7-=aBZxf5Gf)}P-FyUH)&9xl&M?mVUWPqoh*thIBkF&+Mc2-dIwyd4zN<0J5)-QuIx0~__z ztldX&8bx7T2X^grD8{!Fi8KCeyi%<{XBN5ausqTKg83}!01ovsZ0@@VI8<~2h=|y> zNY*7eH%U=;)Jq=B?JbEIsK<2VXk2HZs?Hjgb&F<@;uE9$8T6!ra~t5P$k8K456vfQP9otK9JNKCQKibPI}g`6bBPB{Epr5MG4U9j`(drp{88`l zFDP-S<;z(P;4wmmL0r1SM!Rx%w!=Km?Ljo<=VU<&A(YwOZ!sJwPiQOsRTyy1? zTw=N=v)I;G--IgBT@N&#Aw_7@uN14-=yKjYv6Q3$u}br+4+rkxs#KBN37#%>84up` zc34(1`H{t3rH~K-W0}Es{sHMTFN^@*=*9lGqu-DX18WnCaY)-yo-6E_v&7kU!aCtg zRgNp0+jafO2sN#RfXU|$!eh-4R3%_pT1WAlDHS{z>RClN;zykubjvl}hgCKr>7>+4 z!X>a-+-E;b}WPo?3ofc-fiRTy>9z$XR#kgYkm$!ym(m*G2!dZue1A{d(Ysys*pV*Mqb&3QO!&%8F#9 z)#4p(+K{k~2#cbj95m`z5-0w`%vRebi#QkKpo@x-Qx*!Ytk^rAVQysajH}p0Ae}ci zHp?F|*=JW!*7Gn#g@>e&?49kdrM?ASEKFw)$$qe@c(&lny;n)kkE3Nz$rY)t!qwfV z(?XY{im)vMUR4Wura9&V+@A}zBm4o=o74dAtJDgoRHQ4`2EWbigsjg$Kr1rFpyvr{ z+k&%*cjKZ-O@qAJE>3EmD*O^93>C41JBV_-biyvD?*$499^|`W2dig@|0MPK8JkA+m5Q*xcy}7PxM= z@xLi>*Uh@TZekJ~qk!gUeiCQf=Jflm+l6tL^}myU?M+JuB_0miZYgQ%bYd^uZi)E7 zSzaTFhDwHB+W6IB{n1rMPvmM!4>9>cI!u#kKKH^{d6|1Wzl58Mez08syy9Qw3HaIc zFV-%zSN<1iAG89ob%-B#n72Zz{-^2Sm=;uN=^o^9xP8>_1MKN97s{84?NkrWGcU0S z_tG_kOCp{{MF8LG^!QOjb2(ykf&2Z;QQs5&uJR{a+K_L_^(smo0;r@^bA#^di!?Ol z?Y8++_rJ=S;$o`bfyi3fB_FN#3@pyYreG;gvo22*#k?YE384hH*?qP))LQAMrj#7L zy1bBx^AmeQJ3*DMwc7>7%~Nqdg#?d(Q>#vF^B-OK>|hh8i@OH?D1A;krd>M7IsiPG zvn>viC}6E{M@WX=PZ?s2zf5I5R2CEI7fP3bj&2!H-)rohu$kx zmOCh~kn1gYLt==N^)GB;ckI=rKutv-?(e%+MJTIoBF8ARYY24xomV6oN@emQv_S*p z`U-TY|KIOW=T|Y_A{&UETn~JR)lcZ%iK+y@P##Ia$Yc3INsCnVaTmFEa7oXE`<`MY|?a`*4`OXbZz4Fo*}h!rcz87*c|V9#CUM2!$b5E z_iCH4h}a@pu95vF53gsGVx9q^K&>%GVNbHg{7m$D!oQK6tm5J;{f-!d0aZ&XjQpmRJW@Qx{7*5{wA{>v7ZMWTA2!DBf~b50luph^?EFi6%y?rpX9}9 zSOdsBeV)Q6(4vwzeuFe0WgX%bQR5eiNncyDwSU9@t#w2RiKHj|T-mO_aVw$nlhQ4@ zDRmR$zyZQ4fc{@LVkAG#dB>;+{QU`I41SpNnp3d>tOc{1l)UvZ|C+;>tCo^qy~AbV z-g>X7eE=S{zwj>J6VXMnwFtyPclDKf@t3czkruTrxCGVk;7DcwrhsnOBZ=BF@L~pO zo5!~)>>f~bGLBnEMrDoJE>ZWp0t4~w@SnI42@x(iT2Hr*9zf_n0k+*0?(A9!e;?0j z0l?Abj z`N~;Lp@leSz;NTf$q&hCG2&zBSH1C4Pu}R9qeOIL@=MFFOJ{2ZUkV083aH97+ihep zeU3PJPaUO;!N}ieQog2J;^YiX8CoZr&wOC{&fLO(4Qo*BO*qP1`ISa|$){3ch&^O? zD&6u(BU=Gfd;yd1g8N#2Q6&`;Eo-1XMQlbeDOflxMfoA&0o|;UWGDB4@e=(avF_U!E46Fmw^}otPq$O@VBp6r&en8))I!52efSU?oLsz7EHRQyW#O(A1QV8djMf9mZ z{}K6pB4?qlY51cIDE3VF+FVW4FCMG39L7;?nvGn6M^p#<=VljyYzW%ZyrFl@n_4up zlfA}uKI$t*!AM)LH93vswZx*&8+r(DR6y%_0n zQ9F#{{1kUj>*ZnjqRvkg*+ew9d`th_CRoo>?@g96<#Gx4yv+-_SyT(ZZad9iJZFW+ zY~q*$j7=B0cNHi6H!J@W>@TjoRBC#SUMosbxwhQaroBbSM)fG-#{aan1(u#S^L{Z% z*4AfOCV4TMAdO{1Z%Ab4uk4y(?^};qWr2jt=?R!09i};+a2uc3P%#Y6@?@fB8}M!Q z!lHC{YmTx{)O-&qI;*1u*K;ACY*`QfX74B0X30}rk#*aw;DS-T36B~{sExXh5yOUh zpGKdhpRQ8jF2yZDK8-$!P0F)7=uAsJqg*1=2aC2V-*$slfBH*o>drg-2#)T5>L6M% zdRYR60bWZ0E}hocZy;}wEH*tx7G6M}eTYGYTDV5PlG^1x#RhTlnmXAj%mcm{wJrL#vO1(IX-#jDt|#c90n@1X zfbqIFT&m5mlh7j!f^ULT$7xij9{BXDs`Cp&|+9Cc>=pPnsT8KEGp`^ z1_G5T^&(+)HgN>C7yCiCb~xsYGsagxioROmv@rtvX71j0;k__opmex5Egamp#R@X* znV><3#wDH}%W1uHYWT$hOwF?C`zw72tznsnct zjkOQbE=LR~PLv-jO!=MEa6s**9(6nqcp5$mp= zh(_;bBiLJ~nbV9DK#g#N>z{q8vnp??B4SIZvNogGJ^wgnz9)@VQ5-j|hW(|+7 z?Ha!4245mRo`-a#|1G$nCD4a}65t5!KfJDi$ITt9kJ&Dt{wv07(KDUok5LMa$u@Pp zsWESsZ(x#>yB!q^dC-+ejC{z!hag^`aLzR?I{f>Y`)Bb}b{k!vMYR?ga5ld>WF@C| z0x8B+@$)ENzxnT@>sK2trQAbK-Gn_Wrn0W=)Slkos>-P$ykZ6Cxlv*Y^)80hqV~Il zy}SSk1%{q&n04~ihTTP7i7kBYJ_7&lH$xLRiVlS5KX~S9M0bM}jKsx;X+CTbi1&l~ zARzr6@c#IwuA;-1Hi+=GZE`8SFuCEar)XKpLXXMeP@aq*g7rW|2N>=+rzmdN&rQq~ zeE2@q8JoIp)^4;TdAn<-_yDy2f54xSWs2C$wr=v9BkyI;=MJ}-yu!4B>0s(h9{i&xC;hWml&l9Ze(EAU%b6|0fC14!=s$(eET3OjPOwpy98gOF<)MEIyqy&-jozb!^c-aO-tmQWw_>_rPp zN{^8eCbPW{j=8%Y#cvj>MtS#M9FXDyXh$p4{s#zKk-BXmvl9D5c>%khlQ$<9G^C>< z4++;q94wKDlK!5Et_ss|f0v(LWrU%({>S2&`~dEws4&7>CYNvozDuij@3*%lPc0EX zhk1XU%XPH!G9h5VRekYsuw$~{uyzzh9q@XYq!Lt4ZoF=sBCHg;U( z$@#F_xJL3pcB&cCX4l6%GnY`2e&l0LTpx;lPZRMLl)#RD&KJmMP4%9_>)Vhb&%0LC zu}&xGBg7nTrkM@d#jK-eAP4clTWeQpY9~ClGf0(u$7w0jjdarglA+_xw4K;NQD#JR z=WOUb8NCO%)iy1&+sJxo3>3S|eP?q7Pp*&?O)d0|(gq(yT6IUBv1{8byMcd%nNyRQ zrmQUniwFA?j@aG+sKSUt>ANJ^X;FZOIlNYYO&q> zl-Y=hx(54Gb}B?o$7mJ(sdJUf3H&8t$Trq#+k%H$z!MuTSl|5S$4&1IMyIOIrBd&4 zb-x-FY06H4;{%7WOQrb#S**Xzobh5NaiH4u!)qsMtQmNU^xFMq_O0I`TR_>#mobks z8H!G!M{l;JgmsiQj?ueKDo!aFGkl)iL|l7j7Wf|L(D+ULp9Wo;w&=WTch75LdPd=%idSAMr4QAjeO%3Z~Rak{aRnr8fK9ziSg zc1V5W+s?XExr+oc9do<1RIJ~|=ePSq*H{i)Juo-#gLx57BFJBnZ7-yfM_`iv?FX~M z@1yb$=AWMGj?|kxzP3n1S`%|e>=%H|v)dlRs@9`feenH}6lF|PK(}l&3;naRd51or zw>jR2n%qq|0Ep1x{zHWGHx*i6KU-sKcxM)&&l=l`{`i@gOXNDkzDj?7r-nvW`3IPu zs!Ect0lPf(E$I7Vd*g>z?U#>9U{gu*di1V2=pV2f!1ws_26iiMePW=UOtxwu*33U> z<$Jnj8jN-tS+uHDAth@}!&?LV9nRFF9c(NarRmdL#*yr>Z>KEJ;|#d@Ks768ZngLi zOqIrpwoLJxds9~p1(^O-U)h7(E$-GQUe`xlRvbfC<8NBFMaMZM4)lEHYiSmkqRV%k zZ8E3HK=+n75xa1&{&H23!>Q<>g3W%}BvW4=vzP(DTZQS7mtFp1mC+KoPhy_9L$s9NZ# z*z2o=o&p;g{2Ef0GJGp$A5fmW-_VlV>Mk1{C~V~`Taj9%Do%cA z5M4mk-yx?L15g)RD#xsypQ5nYjxUaM27lU0e2@H;S>*w0{GhXy4i=2+7wcEs5IZ8ci>*!O zRvsaaqrRtJYCRbp)qnkq>>}?ZPv5^X9gAcVf2u^@rlTPpdFP9qtf_40AC#xD4aWY7 z+6v%xjIX-(R*GfmSX=JpdlNF|pF-)!4CGTtf7Xn1aF^wa+W}UGPisM4G5rm0sz&{` zRgQC#<^~`hlxfk3Hk)R?78?QgnF3yyeN%*EpJ!{=k-7ubr>~5i97KoROZ&*!Yz^OY z>N@v;eJgfTbsa|W$_P6`DY-LnxIZE07Zga%`!AVt3^YtQY9~sUSrtu}CJ@g^j}5y~ zNiFuvr-eI7Ig!x)bvf>&o3@otIO;kFWMz|1>WC71a5vc=q;VeOI=4*7cUa!yL&0+H z24Q$nhq=}6rG50$&iGvz`!N{4-qAfEk$j@Vg(2;hNYUE}fV(;%(b4*fy6dpqy{55_ zQyp#g_6yQ^Ci!Z$VyW9g)>(dLl`l|w&Ukcgw<6m`0DY*J29ov>fu>4|zGtcRka0!& zaTT0BSlA@u7KxhAh+-U?HDS|C7{hXT(oPLg@17706JM_ENqJhJxU)O$HEQo2fy2U5 zOHTU{RRVrS$~65X>8r@rUXA%&w;n+)*@6Y%+0EMtSP(*1r6Etpw$>?YY-`lrp8<*# zmp>Fw@Ea*~8+T%=nx2-$JPI7Qt~032M(bIl{NFI~DlPJ%;?f-J6B;UCTbX46NX);q z9K`>ti=OfnM~@bqY-R-!U#qT~YS4j|cDbUb9v>iV^t{7t18wHt`k(O=3X+)tGiSHUJy9Ag>TB4A14TmxM{9PzhF(9K?LR& zsSDE3s+)1E(+Vf3S7!9?u5DJ`@v9L9d~B_Ci9~V3x-Cz&2hHu0|tfh!d=QawPS|qL!%@ z*bhxRQiQtKHrV&AU29d}fF;of7Czt7O znlhP$uW4KT)rz52B;Tn>Kp=5+n;ce@#o7>1)F?E+cVnh->WH{{k1X`B09I+<^E54hnpTvLRHW;=E-57CzGG6);z8$DvU|a<$FTStHr)f1)FN@1Qd;Ew` zF4A5o3Xz=fjE~=ETZC$#<`%)kc@Oh$dn|=2cK9Jx+(`Su8JM56qk|AQeozZ5Foq3mK}Zjwp%sx~jXYWTnvdS#&wiuy^XIkgN53!JbH%3F>xz$?9o# zOauR@EL6Xlzw!wPU!dW*ejW53{EjyrTq)3;Al*rK%D(9Y6{;REke!)qu_?P%u{o)(d*I=8?4;x1s z<+cadWZx-6!q`2_$$p=1K`Cf6z>wb4yrD&Mcucd>@mktm+Wj@kV6CWP=&{1jVG-T)fYYu~gi>rW_5=IX7KJmoZaVjFJ2K@Lb3eo+m%Q;n zIpP8k$zD0HD$n-s6JkH%PT}T_hYP* z({7Zzc=|o4&HkIH)V9Ixi=yC=qPgW#&Fg)zEudyvR*K$hD5Yj!jzs^2b?%J>pgK8_ zeRQy#6}D;8$t1#&0ytHci162pn)iVd_h$TLeTK+>bcAE>h@032dJ6oVw_s1HZ6AOO zXuot;WR28~hxhWEm(X5v``63LCbKJWM#+0;uchP(f+Ie2-?-XVMloIc3XtG`C}xB9 zNfE<7Q?i)Q6H=V^@Q+OwE|JY)PTF2ykPn!azdA8)kIhuDHCJEnkv_yEK1j2unAsvW z5Oj#&i8U;{eY?lNliO&fkt2nysl4_7(dhwTcs!jmy~>Rk*v$2^)J!S zMlDo7d1W`OWhp#XZlYFeut3vnS*`64YDUYW?_5~SE8RIooqFCl+6P)?=H0|o@%qhH z`#gTeGy4W;OE5YGBKLS@0df1M)ZSZ!R7qV7d3c<|BOJEop{5?clD!ZQea^r@4FGkX z`K3P5jHM<+S(I7McByLF!4<_ zBeJb4-=?C1*Q^iU5kCud&+^AGq;NF#0@Hi&nyLzKz=h&$0zzyvC8RZ!}u%>Yo`O2S?3z?+q`*&$}+0%wt%s}a)EvDxH~ zZPIV8pyr{M>+hLjTXFrxqY1&!h3%Ldcj|P>yn}#G+Q|ZYt4Sc*8}h*L)6~QGW@#4v zKDtfQU^w-_PgLf(_#5Ak+nX?;c*HR`<}UU|>-Rs1n+)75=U5Q3rPR(MHmTcwa2-($ zy(cZ4`;&F?clSE@D~bv@Q#W`8@5@P`!wCyk3_k}*gvw`+8MX!)KhMzM`W!;IjEP;F~?4@@1agL7?y{1)xiBOGQnW)CYaI7 zc$sEcGZhSD4;3^^T|a}kJezOVfgO>Ys<4029j|0j<@~4r7 zY|(ObzRuCLMe({AbHYgbph^%BrM8oe=<8J+7(A-~DQ6ca-=!LONC7(p#YU2-nkK_b zP#pVJ(DXCfd)~4FUd?1niX(aXleQ+;sM(6zB(x97h8RBt`@5Pi9MKYx=LbD-{BQh+ zw)^qNY&^k{3f52g4%mv7%1Xj32vJyb)g*91P@Wk0Mi}p@*-t$;g`8Nf4SCgEezf8{ zzrSWF3mt)Dnu<4Y>i|u=kL0Qvr6Ud=Ie9vC^1aoG|Bt0a48G$O5(mbP)e$f94^5v! zT^yt}pD$C~r5(1vaIUd{YvE+Q@!T*t3YR4+y8-jp%kK}LTcKQFu(r7i+IFIKN%#}00j zK=)hIf-5Z0XPmuki0q|7vQ+iDA8T}s$XSdYY6apdNOK}4gIFxq&J-?!+CZ)MOg^A9 zwu?~@U=iGo^6z|Ho$7zm_1%hRw+M;zx$)XKQ&<`KS4DCQ?j~L-KG+oBMtMWprv9`| zE=Wl^W%JkV*S->8odGSRkFrUvHBmaCDDZZ|RzN|rw`Q_nIX~ed%^GKk*Pe8F9Tg-u z%#r+X;3)7Fsy$z6?Idm^M0t3Jk5*GatnV}+;G zwR>W`rJ3K@e-pIZtiHBmb?0vplZq|gnyZ_pNP%XqjGT7L*wEa!oMdEr@q-9vYSyh*GhV|BBr(%oAi&1-EWwk~jD%kL2H z)^BC+c^CB?CL*yQ6TqtYj*ER${0-+~E=RSqNoijJ)?b4h8m4hSAyeI&)=m(Nh>df` z*r+qyJ66k?neBx$Fto^*WS#S%YEbc@;AJqGC#Ad;O$P|?sOM`&e`+~Mcy4&Tft!FR zV0L2qgo0*crrx17lk!lZq{kLyLC31`_oq__D~+62~F z^9OXkcAokT)~4$yd=IT%Of2q6|FX6ZH!ZtuN zQNcO%%U;J6D2E%cWk^2E38b0$?TXgg%`XrY#G}o%$6s$%NX-F>?Tw?}iXYDn2S)!x z{^ncDZPky567H$zO1?8fvm9$Jua`xQ{fkcUjuG`<+#=nFirn_y&!DeZp?B!0a5kt7 zC)*L)RV!R;U&&|Yanl->kHzM(Qs2;)GqY_$7UDpjyYu<@Lz}k zAu@v(PQ-&RWB=#(##@5Pbh!s(G~jEB6t1ffcUFjY?`YR02@^a@o0roGou9G=mB=>z z1+UwCge8QRv6@RVpF3rEeM`PqH{uTVbW!fonY9M}9ok0m^l?{jWMC4An{Q#fAR?lG zoXUMt(tS~U@C>y-=piN-WB&$xbLzQck@crG0C9)${=V9SY`K7kqX;^Cc&~4h!wH+Z zO7e=;t*lsm%oS1}%%&+;^l>3CfH{su@xx#cRg0VOMKpJRLTXv|8X z`JclGm1cda%~siJPTE4P`@ff+5mcPkTQ^ExlUV6hlpnxLUhdeOKK&qwrI;wVXyF!o zrA)(2+}qGC5qzR{wfcy?XHfm$I@)W26?$%n#445^lUco29}MZG)@R;ZopNQ7xAe^F z&d(2CVxF(xyW-OAeUU51@4fu@>w}kpXa8)!;{9U%>X-jId+!D3>{~upmAN>wDShqg zy?=iC?d+d(Da*DTKeu}I-X_zh1f7fkXH3c%=2QhiBp}4fP1P0E9C_Vi8O8!@tow_* zpA+@jR|(Wv;@YaLD2VW8Xn0XEYoQvLiE9Y|C+sRBP46?#!_=$l%PYV#zceED8|W_M zh0TLh`_kB8V?S5Tu!3K;bc?px5?_S%`D*BI(Zd7xxM}&UJ?ab}l5WJKkcN2(Kj7N4 zUnYp$fIMhTW3t|j$0a>2`07{YYX)0SFDD+xbdJ?ek3oqBEMR!d*fT-GTGvyL+6nd82Y( z4mqH%!cqS%?a0tD?2NaZkIt9T#I5O;S=J*(|5+FFU$D}Xro}5BzE`w~Sh*qqcb4^# z4FSFzDg&#|+B=YQ(75FQRE=Bk5Vg?{1o_F|ojOOjf~&H|j$CS9upEK8ocJp_l0Dp; zLz4HOv;Iw=1@jp6G)kB&6lQfmV^qWW>;U3Iw={EmRJ!7qU>DZfnVPXx)DFc$+MyjZ z7&Lf^BGI~9@nu!|TIv{R6sP_TtDw;q^S{?TBdOdYc)lr7dSwD%QM%GFR-B;RwjJ6v9^-1iY(I|_i|~V( zd$h-dK;uX6?0{t3Y2M9%HO{s<*w;O4-H3k~cOvSMv>0XRRHOUX6!G1bsev4v5$D^K z9VI#zE0cr|?;MG!^YDfC)y4o=k8}d9?l92<95y-Zja}UvXyEC6Xp)2=xFEsoir;QE zI^2|Gcf?}qyX*~7kQEMt!Q~E*)AsrVf5dB?3X$~4@J&_F9=E~OLnyG>mta5l1(xv+ zP9|gOO!oc6!7(OybGEAHEl*E1tTpAc<}*f+OyKahooLL^Lv{io{YU&a?_9YpYleBS z>x^1l4U+6`jQfbl;%9Qs5ktGpNj=Wvh@Lbc8 z#pBrYR*X!}e4)RcVQm{FM99ZxzAHB%TBj$_2a>JXRIvv`xYC$WldQH=lSgcYv(V zppCPUtL?$yXd0;;@f7elU=)p)%bAan6RfRPeHVl=ezhSF;c}VB5u_0(qN9B zW!W%kpUiM6rLel(xQK_PBMpdEC-E@%! zjWPBi=hBcdnUM{7_7T%ch%L!3dO*;AvGF9yB-t(gF-qfPppw`&wW;k4VM??P5i7l- zzqxg9@JgJElssGM4MWS&8xeAKLxX zlcnwQgEL4=1)GeN%}i|aHzl)3>huwEp7TI$cWtR&nOgBrRRQ{E{weKy`!M|;$ntE$ zhx$K87VEDt$|#?eO=Bo%bl`-&xZ5u11)8T}fl$B#bn4~gVe}b2{0qMJDZv>+INfM41+`E&7p=@=mm@(r?iW$i z$3di%_M!~?G73!-E~1A$5G71bw3KO=Bc2*^Hi>uL?ijoCIoAKh1~z0TeHJ;ki4P}2^9hO?-3Ucnm%4{I+}6&1g{lWJ-KjSbT>0yq zNUV=4RD$J2681~t6Q=IV90g)`H5kq41!DF+Fsv{np06u16ryA$rjdWAk>$34&LH9} z+eL5@Cc7_uS53-YFsxccl(RluTVsw!F&RB z#1x;IV1ihCkzUOwF^nmKD{ zTq&YCmKO!bcW8_`VVGyIA9^1x2wor9g;igPQbd}Bsnzn){!jsgTWcCQZ`jDb(xJRu z%=2}Pr?a|yQqvZL*Ac=vnI(qC(N#-$H+VnWORc}su+|oiF{AI8Qa5t_uE%bnKL%jj zS8k?VVS5L$P{1Tks{gb(scKT?8O>YzCEF7^TJM^^J2{Tj;%zuB2pg7{@*aiy+M12n zUrA@9bj7zC74xTs0W2PAEwrJ+gaB_U0s~IgTr(bWv?8KzlF#jt5@I!KFII53ELF=j z#HS6hyJ{z$x+T02WIarijdZ6_W2k>|{EFy(D}P+?d(=B1b*j#d1`W?af2SE9Hy5|X zha9U-$ceM4l@G=1?z=%3LWhpn1rf4xtnj$uBzUjkmb7Gq(%|iDPm)YUpypxj8pHRB zQ<_r;yIF?WA&1ockMJULm>1w(`js2pB?WTub>vcLCgv@b3#VQ;Sr3VBk(d3e7;3b( z(;y;*)do-B2#&Ga;c|DDB5a-QrK(ZA zzWShu?K@%m5Zf*me)5djY&+bIU@c~C&e8SP@$RbaYAjt z?2Dn_2T^}D`IEL2e`STKN$mmY4>hT_A+s`lcxsycXriO*Z_(_C9MUgtwVIxZ$qIgh zMxwp$(K6r~w)sAH9>1T_4egOX;-oB8?3SIbQIG4OM$T}wa+cc%dc51HXdfJX9ASx% zG(3(*>eh07R6C=2#Ub(c>A}B)|BCAuDAaS2ys#T#XVxg@!AIqu(4>h8vfj<@0ZS*k zH#kfl4IkFu@V;l)q~_qjI)m0D$Q70VknFdQb)no^y+7~VnH6-@f|2tAVxf1vjO6Xl z5t1!0;wz9|8W=b-^uKpi!TR^M`aA%i*qw^wRs~4z*_opm+@xozQ{TorLz*t(7;51? zd_jliROobRC`SD-!MEHf0JAs0)dAjB5}}T zBtRax>rLn+m08;loN&gTeyCKe@nC(loE=@qrKgxv#aP^QA4dYNy^E+KzC;$!3POa! zr>X2mw99W)r2o18IXJ5>O_pfhWIK$3sAj9fi;9YRefy$St3r4unZQz->X(WZ&1Os2 z=`$jG|Ab~E4x#4p;%Lk9Xe|r?@@35QS9TPVbuPhisF#9d~Fk)ydAV60LC3{ZA;D$;U79MCk< z2kc8d9_X)JF`SBsvI~Dg9GLd8d8CeZp zJ+oG)*xd%?NVmvNS$=6wy~B>f4E5lHX^2WxC4b`m>?$?YY<=X-bzyKmsLmF-4zGo5VX2|wtQ-7=pxFu# zb!LX+qWF)cLr$?%HpDyIVVb{-{{p{e)lB|2X=Yw)M(-j0BM80-NV7HZi8YEgU=})@ zTAVUjs9b9~*u7imWprk)^{ru7w!aY6=MNTARrf4@N;y*ImH=JkhC3rweVYKuh^nWP zF$oh~5rb`5kNz$K-mqVd0skByX_v@KKppY=)7Bedrv_JrUfMHd8R>FYe=UxIeq`nZ z!JARkxdm&0iFU=)sO<0zeOXMfuVFYU-S%~nWierVrXT$f^RgvKID=c^Tfh|J7lTeq zSr9Bl5FC7B2YRPSkZUe;re2d}Kp8aR9Lk?E8mRH)jJeE|Ys_{YemBv^*Zj&^gE@g( z+Km*_KBq?Vvm$GHSU*e)2>R&D{HUM^>H1+i=UMtx#bD%RS!J7bmmc&!7c?4Fo!Cgu zbT4O-i(Sfmg@+BEL@;KXY8fc13#hdp)Y4c-@-SCQ8JZwg<3@Olj$-oU^rq(^l&{*& zN$F#hwam))zG$chSATsd4{Tq^JuCahI4a~`t1t-L&GPmFQbdeD_#Re$Kzq~IA{_(T zXL@F?nuhS43t(g$`k*?v+)kMUzG!7pZ~XdhIV|^>Gls!K(q8R#R|A;8GAN=sOACaz+fJD^Q2^wq$J}uP*%cN z{+!ls#|dFr!%Flk19A?V0bCz$uTNXt%k&CEv-g1Nw7m7@DDHq%LOJQ`OeQ~opES}U zs9g%dAn_bw>@iUJSoLq`K*?kXI?MBwZKn4U_;pLr;JldU;NrHGTuYYK=27nA``$J< zgk!WIeIp`;Qyg#p!1#>0T_j|^cLT-}dh`4hW{kh|1l9V?q)1(Gd^sK#CMm$3-vNnz7ON75E;?$EP}u z<+S*?>N(-qR7;R^kk-#oc%NqJnY^E!$F0e%GOzT%Ze8T2UYQ>bt0!4k3parS*ks`8 zB_ZdvOJH-H^+%G~uTkcT(F}glHsWJzQSl6QG?%wqcoIucE~E+cpF&uFss9?Oz6p!% z?f&2g{2n7JH1v6k>Nsk~4UdJ;jgVK0kH6dooxBT1d{a!WQrp-&2gUa%*G78Fv~GbQ zy)PF!9rHhQPCUh?9?fL)u^)H*xYKS-2j_MM7!Rn1zWO~JesKZ%8f+MR#4&f4d#b%2 z=;Pl)L-%riRF#UtJN1lliv?kntq=gG^kE-pj#ykEvbQ-gp@_xM_Z`#a!-frpequsv z%8oR;CWM93Jhz<4urp073bG}ExD$|aOuuavzO$;@e0yvTyeKdy62B=SI-a`Em3K^3 zUY%{v2Vmc(hmKko;a&&tXKhP5=DdO_dd$IJR&P;S+i#is zriy%~pQ_SMNlf&STZuc=Xrg1`b-gN4;*v?8SVnYQwx72b{;-&vm1!a{?pz4%05w4l zGG0`@ZzCDN+}V(Y#`bsNQwNb5K_64noP^W;*|5w|{O{=i+)M=Px_DgWezYPOgGx{< z-Xcdp3kAtNnSd7JB@Wb|>lcEz;{JvHC+Bb)N%(cOl_gg$Q9q(vqi4JQ56$jL$qx}? z&hc3#ruv)}L6h$=FV-qumdaGFsKq%@Q=nCU0>hrjD45Q{{Uhj(rR{^OArY<_#a7Plq|g|lZrVX?J`dTExTfL?QdbBHQA=IBT#+;ry5(?j3o7M6S$ZW*}lM zPb`e6_mGw65#!jJN1mm8r+oW(kIG3UZ4bE>M6v%2u?!_$&JVe25x54o)~TI4fE$GP z1VtO}f|qq4zhigmH9qYb?G$r*%8E#&CHdEX%eMc>n6d+Zrd&(llD2Y2USL|~FjNx; zI~wL8(7l}W%_ErsLkirw=za|X78wpD%MT?rv+IrJ+(*H=V=EkiuQHy-8t=l9$jVUZ? zKL57kqg!=iX5mK-`mLi3crN3z2n-&C#oKoqto!9j#JBQq&;qgjQ$-TM{D+-8x9jvP znk!sTvwS8NH9TthQ*!NdO*S++DaM#He8E)!Oley(olUUK=w=z*)9B@$eAsgk?V>20 z4`J%?F1QoP7ITj|-V;q0?>J8fn1_`+A$OdYSZo4y~3e<{MScb^y%Q zSa&&IWIQ)zJiymfv0f)-H@+{0F4jR+!wk1nciec-crgQR{-nG_f`3koxVMjJ*`ku1f`uL(9Xf63nm@~@&dH)|8}FpM%9@9-tFI1U@1%y9nZ;E-!;)K(<2&$y%0Cmn zR!MowaU|_Uk?9N6$Q86b8qv?KnsOKK0$!yRh0{JV>@Qmc{GJGzFN+m9{@T3VjUn)L zCjKZ36EleK)P1pil;K}(u27hnl`5mmR7+Y4Q~-M`C2bT~wKct^r; zN}jtd#1M>HLyPB@azH(<*vvk;bGfgE)i@PCFagD$>+R^Nf9gxC+|wv;AFJPz6K3&G zilU6T$Svp;b65A2lX#_y-9f3k5AU{H!e)-BR&o!8bm#*xEaJ)24l4s$>9HL&ytV1u z@V%jj1Se|Z)*@wF*Xu_8T*NCx?-pe^la#Ybu5TL~U_1~=KB!s4zgyCb=T-m}m%!rK zAu1dRZ_U$tk&-pZER|9I7C}EIAr=>dq$bjDS-uXg8t*L^m+(26j1HosS~Rx zb(WQ4?*vMXMC92etAAbYTlZLGNCzb1Id#djgHUtD1RKI>*EZUno&98@4upM|_S9Dh z{({JwGffJry|Szwr~aUGc#$g?bkJC+e?=E{-$!gv9Pbcg8H=GC-h^3ztEjP#CLNJc zpy9bNIxTh!o}{cBpG9YqHobLg$f48WN6q#yDdI~FCdJe)I7S;{lgth^dPxjMZMmXu z1ftrg4#iIz*-UZOlk1%wpxPC_8I#b+k#|9XaAdemS)1MKru1nrYF7pa1)oZPnyMs^ zT&{~KU=CuVs5?=Y#GY%J*F-xc37$+$_lS5nXLQyQdin!Hv|)keK$6m90iz=tq_2jL z*jpIkh<`5@k23!G7Sp|Rt}lodRA;`HRs+g}R>suOIGX1Eb?zam%+9Q3&M&eS{Qz=VL+Fbdp~Fa>65T$H)=NTxNw>n z&&09PBaVZD|4gR-McElf+><;^o5ZF?v(73*LAgZ7L5ge#{H78E!U?V^hY<3KL}Bp`@eyZ}g@N?(b$l`)i$CbBYA%*=fk?Dhp$REN9Xi;|CCJ zs%9dzHh79@F)HDXU+7d+SQhW{pfWsW9X!Hz5gX}emzPn3zXsedvA9HiPc)_&{F+y( z+y@I|RI_3V2PtCuemG_YT71&KqKKfa?bo)Ad2q^arwIKHX{B7$ObaY|rBteD&5HpFZX|!>3C{O`bf)cEc^A2mK0L5(R&5 zvB{hpcGpKVO+$J(ghiZ!erwBw3wrgm6nB{)XeBZFxtk)_3MBT;Yd zBSU2XHR#G!#D~90&rJ@EH^h`nSb+R|g0FI`K++1Fqhc1Z-HF!G={Cs|>pNTj*0baj zUko0I4nQhW5Z~Kr2r^mwVjIg0BXh><4Cjk*=NXyxDneBiQfZ;tA29H3#fZ+*=5;P zRhHluv!iGsX&tpPzC|B3YZQJZ8f=9>v&v1DsK21nqk>s{mdcF|?~)owS+&3W5CLr!ydG=8CZN8M&vR=-3F0VK;&~5| zH*%Z(k*Q&UVgjpzDnD9o=eVGj$X5jEu&Ue9Ba}Q3^Z&RWF#*J1YowsY!vBb@O?B66 z{*a}pZ&qPn5&OU4l*`mD@BDu7%IbtiLV-M|o|-EAunR2_Toq&|PsXXw_aTe?R6mW_ zp&T1-4k%47{xIrBPDaSvulMig*Zrr`9gr;M&Ro%3zNu+3C#`_jL{VW(bKuZ-?V{lG zL|2u(*y#sd-n@K!BL85bdS8~?LwnMfR?M&fT9wY;Vzq9t)Qy)6^owj8cN7Ir%IZ`5 z%IO;Zr|;9CZi?s-*7v4M-yjD10;S8w{S#%39ql{KtER9^72?`OonFGcpvd#RNpeLhPt}B}=Fp&OtmS0$kSoy}+I_Vxf8}vbtZEWdrTs>Z zb{>A@zbGg&M2Tk8lbfXy#C_o9mxPU8y~BCN>)dj}J>pIpw366Jw@eXeq~Yf+=BaL( z+gF;@6XzL%)yyRcfNS-P%Q2utV!ieRsA=WMqGogTd!xVbN!Jn&<{c8>1L8 z8_W}h1bos};!5M0-hd_v>pn~L*v zb*|VzJyY1`jlZU-63r)0w(7*5UInt#0%bcTmjd%D1IJsQL@Q}W86b4=Q&PL{=bgU+ z=i9iC#-Ku7@pydXI1W@zv^G=ixRv;;@&;(C5V12RY;OF^u*`2jq*p6<4paZV)k3>E zm@UuuHBa}*n{`=2sB>;U+t1qj0SWF(bzM}@+=V1TYPIudF5GS4GB8%%@KKdm0SK#y z9&;rvBqkfvq~3_51kNU0x3EX1tE`_MMc+$(pZb$1D|ic@AaxFKW zOzkXwM@;hHQRgU)eEmM_4d^a#^QD};lnFQ8LupcTXc&zgc>{e0dUH4S1lX16fvO)5 zB$w*OPn$Ts>xr(fuk07C8$4Sc`tUe#7NFERMDZL6Qtt* z#XlHCmJ!{J-9Q+;DyKm7+N#cO&-fi-JjH@r+N)ha>Z}><=o*FR;5tYJ+sb*0ZK!Zz z^14<*W#}kjX?p<(jg*eu3#~KxAEb62Ge0JEJu>{&G5|Hu`vw#dLd<)#g#KnolwU-m zr-z38N?Ql(I--kXZBxP6zR<-MGHZ673jQ6SDwSzcdiXBsLPwvhg-)bZKs5MslgMuS z2P(;9@EW&+e}_DadlUSV`nNAGs_mXk9+Sv7*D8M;o4!3-7aZbC^-hat(E|F+&8T2* zmW$VSgvF(~a{;FkWxJlV*LQNcP~$YbN{mEYw?8rdT(E%}B_#V^76kc-oF`#ZahSg= zo{$@6{pV^sq^PWdy@zbNNP6b$?<=|o4)@IS2zg)Qt;-0^_oFR`wC^D5EGFfA^*x4Y zD@zEsHdp$8G@rnxa0XH=Nucd;>F@daA7J%$mz!o$w?-W!ak&&T9JrmRq<-YB)4y3{ zwH}hLU}7xI2eZf`U&?P*zuNKyPL@kF0@|AW4+4$THj4N(NSaMKkM}ckYa`+au6O_F z`8>hob{4NKWpZ*vE2;`(zqU@NsqW@GAU}-Xa5KBRJ!o5Pqxo71O8Fwk1@C8ATMX0% zPQxRJGk#N^@4JTgd(Ny>-4=%$Ry9R$LJr-8z65SkknnU1N|F;6yhP_Gt6kYi@0-~l zoTAw$WCnIJ+0}AUw7oo1eIW~7kM?GapVJJ~icd?Vi?F=31&l6GPYM{0DY3r@&i!32 zqwHX)vKho@tSx>MAaH|5^(7~pcGYg#2%nj73i{cZ5M}4fj+Jv5@uhFvfF24yr3{08|eW-X+tr6gFLP$siY-WQ{!O9GL$7%JVQu#EoC_l&hM zL@}7YfcQt+mZ`H)BpBIJa@eZc&`!saB$6mKte5LnkF2(LxVQ2fsLblLdSgtp|8Iyp zm==!xQJ0Im)!xjp2PN^O14Dnp++gWf@}n?$(>l6$&^BxTIi?*}H zgZm~9BWyxXm2-lwMbrk@-VPzVpUCw^--Ir#m>ebYypuTz$sb*&*a#Q^SVri%HK5!c zMXfog+riQRFNC#_e-&CQN72RFS42o%`W+RpTftCR?R{3tIp4YuB72@Dv}mKSy8k*? z=Ywfz8NygsLkB(m>?c{ljp+x7;g}n=30&Mv%jhQf9FFccK!Hxda3^Q!FA9_&c2O&A zq*K)Lv8VV&^eBFYVSgRM2l7^(Sb-SL?(=i4&`)~W>-7E!fZJsJ!DI_%{0j+ge-5Kq z$)g%rwIbwH6!94AxFr3JC8VRDiCjOe<-K#K3BeotB<#xh|-|(Y? z%7N#MK|#$ugym6Bhpf~M5M}w?Rc2T?x`+MLPq@JLszh1T)=|WI6Ghn$NSv%y%jOFY zg}9#A7cgU}8A%}-4xR9pl^8GZZ1SxH4(FVQO?#<2tv35M_#_?UtzvRGMk_jpKb0d# z%0t5F2v$k5uosO!YC}fZqlf2?(l6ujv3yZ13DJ#t06hUX!Pmr`fz@le&d$^grSSwKfcfkb`)Maoxse$v1 zLsyAZ{NNz5%Os#{F~xjqa&st;ItMvd1Rc^|GyY|w#dev-@y2aJD5kRCM=?L7Kw+Y? zxELQ5SVK`3R-C6P^qt*`i|L7qZyl<$$SISP*rJ@)dr{wo8#792toxJ8lv~sTpbkz7 zZ0a<63#>=%QtAIaqUMaunwKuLduR*8HQPpbop=?maY{)Lo zcfRDyeqy#Ky@d5JJDLoThhkv`fn6Liz97~G7%22NbV}L2u3vXv=SB@$oW!5rLEo%6 zNtR7jPM#&ihgsKAhR$i`b5RmWk?_v=T_4E#5#n#>xIUYY=K z&i}4G13N3oa7Gjp=66||Cu$kCMKlQ9-hoE{+{IFC@oypV@3JGBpg}{} z1hIFc{I^il2jV^8KD_W)Y|v6Y^ozL*r@ZXaJMzKRB-}3XbPdM7i+)p$XdC!#`iU_F z<2au%9m)-M!4YCL)0L6*btWMA7kq!!6jyzve6c<_lSnwYVp-=I&KOL2)GKwq&6m`v&uNtaGj|g|)pKn0wf(Ax?ZN zI%sriJM>3vWT^hM^y7<7UL%bAXTULdrX zqhXAJh_&Y2D?y>8!&a*g>+9s_ZoRlYpZh=l%#AS+Zp+|fOY7t}dRu&-Wi#hOT|w08 z3f)q~_FfNH<2>0O*;UytTQc6#Nk29uagG_m^>rp*!Vi5hs~IDx2DITHqCg>j*gT|i z|0*}A;L}zYDLv8ac2GJMv1vy+(t6_14s)4)gH~>KTnqO`>0@d#>mQ5Ry(JHolpqeN zMKLzZQiD1{Y7T&4<1i)6pj^Wa?kaOz*gv|`K3qeXDx_gNvyLXq=%n5Rp75HEOp=De z5S-tPL$+G)fm@U_&Y|U~Oyta|nyQuz%=25y-EFtvsr#r<`J#}0T27s))1kTZUlGyb z(2IkyRIky4C}KV@+fKEqjFYD@-J@E}BFfQp;12M%BS_<|rr*`vSm93P8rHQbECswR zf472An#NZMastv7k<}?eA{DG!eSoNIO&%o;`{FOp_PqgH4kvs>9)?QXq4hTKyXmZf zGhrb|*+ucKirki=O64u}vOxbP`E)om9Zu~RmpQl9*P_p8eucF}Vo>)b*=!19F~HK+ zU-coqVR`IqS>d4c#X^Bo*nIi|yN{?FV}}aCu@7RJoIZAxH%1q8!2f`JYu_ZmV}9X- zQ_k%`<~YmHw=*h9$O%(h&#lN61&k@Q&Brz*MvkG1)$#Sk8qLAAXq8B#TI7xKWIcCr zl(d<5SEJoy@Y43Ubqvujoo<-7e&+vkbnbCU-`^kK);3m_tXWy9u)bETELj)B5@D@1 z*DP7HqEex9Wo3z%Qjl9^&dijR6_pp(x>(+*%sUtFh$d8~cm=sADkvA_e*5wG{mb7x z-k)=MJ)ei{!~N2BX&jyn@Q$cAjr%kSDT}!AqV|eL^}hZ&*d*16Qsz)x57n88CfTq# zOKZt4xetwpAQDnBge3z(($LjlZCd{V>*pkiv}+;sCMpHNeQH0(!C^Fh(tzejwy)?6`JK9)`oVO|A8$*MKDUBhK=^9skLGCQmq{Pu)b3S#PesCr z`zL@zTXr03Mvh%mxHHmsO<^M&irbID_5Qki?H$_Au%aMV;3-HM5-^DBxPOVk7_zL8 zn63#{O|&}|8=M!|W`Jdbit975p{lGYq>Jef{XAMat;pgY#9Xv?FY!;M64ti+zryhs z{Ll4%3EtHf|Ec)pbi2#dF>Puv#^DVGTHH-uX^i?edN%-zO(-A9oG2`-B zbQ37zqWYuso?mg%81guJoSK1)_r3K@SiXPOwTkw;YVvpc!(1wUe#yMU-i{nZ*eWaW zAI3NTBl-=|P@>3v!NHaJ5?&4Lo}@!ApZ2~dat_(6a_5@yb$MKs18tW}PNuPcb3XJg54-0NZ6tJa-T^5~>?55d8%a37g2f*0Y&bX5M$={QHn$`fvb+N^#u!vraFmA@n|kMeL3e*qe>()j(iwp zUKni$B!I_ZA+^TW-tE-T8jf?|C{erWMhd-TKk5BZlUaovFc*|MOglJQ;{KH|SzIf_ zel*=^U}wQjEie2f`7h?RFdK7_U?#)*#BtME&&KJNJQ;a1(^#^g2bG$re11Srr~xu< z>f5GxtH5UEn73Le=97WT7@M8WkJ9C0`|NZ|oMbu9cu6BId*bd5T%3Ak1x{bn1qDDd zaZ^`^=5lhG-h;93I4=b4Eb_PS2OeTDXGb`7Dw}j!6?XGtT#n0=JA^haW(qXm1eWG3 zJO>*3%-Qgm@oLrPx{z&d6IITtyS{-tXHw!N~N>0(s46hJAik3w0H z`d!kb=0&LfH(7Y}-~~tfaSWo%z|cOibM0nb(;bT*ab|I+qb>QX!|a>74dE770_X$Ob-Q>?-e;11*7d}>On>~1 z%d#rPmK9m=++BE^KT={&adNkl#L=p8mCY64E~AFW*XY?3mb4&AU@3U4aZpYvH9ZyP6oKd?mGv6>-`2&l_5WO(cTp1ml&}RlsF(m^yfgg zfORQvQ8LM0h;IVG<;$Y6J~PjNj3mcZCBD=Rt2PFDD>07o#wK}_=i1q?^3d?*c3!g; zfcs1wsvDI(9PAl5DCN4nG>6rf`%FRlXcNft3LuGQ>ma*$bnJO8c?PCDA_JfQxtO}g6f7h{#xWewmS%*QfEtkjiRg2TN;XT*>N{$Uzsy{lE35=1ns~z^bxDsK z5%Y!sEF|&fz(&{L&Y}QKLs#Jpt82=yhmnf5Qe zFGDf3(Zx((Zf_3vlEQ|+z jjQ)H;^tMhW3D{rP_J5f1p?It)CEb}e*c_jwYtz{3=Lds&V8M)()hO3c{Sk_Y z87{Fcc0QF&h$W6u^TvX(w2t8)xO}{x$Z?KruAF8`mxF#q$Y3+8Fp#z?cNDEEZb@YT z=xhODZNz|e1LyNfU+5y%>2c0|&__5Jhk!>mwl8cl*pDPrqLQ}8-;|i2ThY$@xNYJ_ z?qgHJi&4!5xcv1hAL-$`z@07asLOfwnht&~jIv^_$je->D$yk-Xz}#1k~_jHqlYOE zC**8@WgA(l4J->SDD~!!-I=G?%ifJIaV5|B;xzU6`~uo|*ecs8L;54C$lesyi*FC8 zSK%jFs*H`Uny%LKBg`$eKV+_ zi@iszpddftqZgxT=&d`|FDUr^#rvHom_X-hJ(tsqC{4-ov^V%m;+k*ih4GV}!B^b!#knaz=J$wz*d_$Ga0i zDVUCRc+vm$RbCI3ol3`yc)JFBWnuZP`TV4vm^Im<{O!F( zDp()bPWQ>5J7E6FHoJkJ#nu0T|2&BrVily?yY2V2_xO^X%LPA+%SA2jhd(c-kC}9Z z3ZQNEt*E=VL`xTVRnK`6#A6w6G9>?y60G*yhQSdIr$OVF4+k+p#7Qr5S`dfcg&g2) zWuVHPSG|u7+iYXD)hN3|**}|Me}UijtDS(CY4(FqO(kQ)KCijj1#^;^+_e7@goc>( zO!F`BA8D&fW&*;)v=MY8+4zloG498a_UkDZ(|H$I(ZjoF;!M{QTSW!eoUD{5DO7lk zwqd0Z9gK6bM=Q!A4UFJF@tK<|09L{V&<90T5i4Jgtw~_?9;SE^j(XqHlB%LcS}+$d zhbzqwTh+Jtj=~Q9c~+QNT$nbNDZMskGV`KBoW3B_1=h;y%OG zf8hc-8>d?_P|N~X*Dv~z0U{owE3Ml00fs%xAfp`CZ52fO)7bGMpLp(4A<$IKC9wXH z)Fkcbq-^uc%Zd2~al1@CPR(XSP7iN5bfTzg5YKoMQ_bM$WxLb$4Lo=OzAIDTf#1C zOuf3Kx0Iva3LRkubUTYW`-%3IjY(d|vFWt5ruBMAhz3V1Vx2tJ<|l zLM){wDTjWataw>W2YVze5dSvvHF=zF-n^P=ABLiF%B#;tLzs_BMDncjoTkxE9)S82 za*SJ4Ug%I9CRfClDKe+1=2iFy&vF@u4ZN#E{tU293d?MKAS`>tY5X_2FuX8eJTk44 z@Z2athhTG*&z0k4a@xa}AiRHCe?%Rpe){ri(A#OD*tyllgn!@fJ-6Nhk1K}2@SBq* z3n4?M_L6Sudg@)%dTR{jdnL&79j-MXJMN{%7qkjXGZoHt=1t?U-f14n%d*b;n@$hh zWm_i!B-XAoBT_$7B%`6u{yltyakibnc;Cmd7m}CHeknPEPTDGn(*LY3kFHAyzNl=W zD}w%vNcVN99*ZR#1;^|e_8W$Lw^C_OrH1X%(A65WNV`7otGFB7q)O_Nke`%^!G2Cp za}I7g`ZFAH2@*)I(YNix+${OmNg~cXgJF)?+G-2eCMAIfguC_C;osMsDB^Ab3Y2l9)xQK(4?^d zK|zJL`kcPPHvw7DLBqP{z!z3v>Cp<{qb07;Bjm~i=etV+-rQzx^LMsCRDvCFpK<7; zqiFW=EWo&zQqn~8XFQsNu){d3EU(&^Z|b3}1)YtzZQy)kYk6$0si4i-xAIk6b;N#T z@W-Jw39+R~T5-o3SFU}fhqbtFy|H+KKZSQV@cWq4C|*X$o02E~ zrCuXzZ$BM^-WezVAP1~SD=eI>(C;GG>)T^FOTY=^owa{ak&U*Xim*z7tk7x(LOeslX=I54S@_IDCAdoclD6f*OD3 zl6m(<%zeczvVS+9P8w4GRWd)Cb4SEZRCOGa|AI(KtuHatqOv%z$@|3>gVQj)OWW`r zkpD+s27M02Y~91d+=J_0qh84mN#X|@vm}0-mZWL<{a(&+J0d41?vcv+{mxV}sZG4) zFjC4sjc)b{xyN-3C^D(%o|6R}>xtxM`Zn>d?JK9z8JaS2;=y^a?tnTIFu0(r^-^SV z$PLDO>pG@qaRMCWV36aqC@CO8fT@1vay8bK3)tXi=;{RaMoU+fltl)O{5t2&=C0!!BJW1V#9MV6sO*%}MPdg&l z`>dy{ZZ6t!3piIgeKqT_>orZnFa8i}d&ufiq^1f-?-xRup^??Y_47O%*&QCQKNp+6 z854$QqUP;mwk%O?E7!_)_|XQF@bx}m1z$WivXnbNFgJD3)J+VK>~S=_ygb|1MMvqT zIyQhV>MqESI z6-JWzQ!}q+o31L`C)49DB>9T)n#7NLVVwP7q*LTgu#B20CL zzSbPW^^^p$X|T!0Qr;5&%muu;IBF!=Y_TpxHIkN@k9RymdK-}NI?`VrALGW#ylb7< z;|_4gp5tWCDrWOxV-ZAWSqHT|hn|ir#fX~ahZqm%Rkb&(uB1%YhgP(TUWp8&mq(Gu z+2~q|{bzksn*LNe(Dx>(ME`b9^9Bo4WaYW`VDP#cd}`ds9M2%mgk}FUF12mjG}kag z)G|!XJ)iQttrEPAgm;EV;fgDK>wYv}wx2Uw%nr+8D$}HOb?&jBFqu!ceSWI~geih@ zHd>$wR8mUPHvt-?zTV@`9QTsrNbukNIrR!*;3CcC)!aChck!O|2X(tGwdRVz$HE|J zWld#TL_RF_528XI^PUxpS_bSw1pu>1KYFxTZfYWmX`%$uzou%e+h&&4!R2%Pxr#Vx zSKG_=bPx0A?VrCw8Ek7Q*Lt%=xR+SZU_cIW>OAskSQpCvCU_luogoWeT2RWY3$`ss zzQX4{?Dru2HW{W>H0nwY37II`T5dpY{M_SIE%sGGCoGww>g>fIJTLt3b}%CN-@W>h zPeOd2w(-JUQ7}6NDO!Kho@WyljVSxLL67BEzY#HhRu$3dD>i-#d{^pNkk*mWb{m#sirQGQN{^7#DY}ux; zeocUNw$kd6w+9XfNoQt|U}uK(Fk*|d;26AUraN{9ns~ixP;~#WV6IQ|pi44fZD??w z$=X?M?QFHgSh^8MkmOesJ1IHlI8k$+MeRg9Doi~hME)ko^5NkRs>98C?>Hf$!9w0^ z*!fa(V?i6e`h_0b!)V$k)TG5ZN03mPoY`Rc%^l&{GN0tE)^?E8|=Y@FjyWx;h z&2D>7KU6tGXVM!JYG$YIYyJ5#pWECe+$}cU_k>;C zUFfJV1-FA2bkG&ENsx<>5b4=~)AXexS{&akbGH0ozigm**?vR}SiQI%ag)pYOdIF8 z*7q>o{;|;ewBJW27Z`b6mKWxNjt{>mf6^894!(p`xcY2H4gehh{({JXf7$;yU>c@V z($HXgg*vCZ%!Gx}X3a1CyB|&!Q>SnCU&SUM4!}HlTCH)!DhL5|`OLkRoh|HE$b*j( zE3m^p!T`&Md~XUiX${_B)(rRzN^zxZ?Faq(At9=?P#Bb_Weo~-Uvy%pN6wL(Jh&Cc zgxkhG11L*;1r8;$O0{ zU{*H;3h=y!Enu${-)58UDi=AA6n=)uR$OtPXbE$kApD9Bk`&U%?9X-C1{QpwdP+?c zyq=!2=Ym%kkDW}a(a#>ycpcAb19ti+MvaqxwUbreBVGSZLj*?=Q7nM^oO+2l{8z%l z9cP4v?_ApH{;vq(($N8vC9^PYGD?3_zU{$P+5v_HfW6KL*m(nGprcCi~V2yXa@6N6`|sZ_PzF0$5I3` z2^k-s6>Kf|ReKON21|gykID_B?6F)?9KO%&HH{6LYo&<-xhY|!sjL-uOwyj)Ix3A{ z3#em8mYKT!l>_TW<*JdEgmpm1oF{x0r;;O&zK)DLEWfyrLLoWyKiyIwi&~>77)p#Ef-v*cMQ zMDVV@HJJ0n_^L`j+#(eB8b~^V{?AUy`?T>VW89XgYp8D}-Geg__wq}hkxIQntxq01 z81XH2cAB(Td~X4`8!_byw-sR(c=HK(HlIg0e_4}5&!~;_&S1*Y-q5Ndk1IY?y26D$a*qXOQ9zZQ*O30Oz?-$|2*sA+|3yer73lb3&#IENDxIB2biI z)(3U=d=EN30{-E@3EiWPRiS(AS@1AmuYAs=sbJJL<^RCdwCt_W6tj;zFUcuM28bvl zl!`WdLfI7TDc>)E-&4L6d<|UKs9r!8Ho*9Tb%7p8mNNo`H!@%P+BL{ti#f6=rA?Im zBxrUp3)+%(lr6PFdQ%F+2KP?om$ZiS!;ah1S}M9-ckF>>sUBmyw`oO=99^RX~C;hquf6%cWH1%K}JU5*Fvm`m;F8H_|j`6}9r>uJl|L4@z z_5*tbnm)H2gTM-`Z61bU5Fh7ECOe>ERz{y>=f5e)X5NGDr})CX!xh=*qUTUu`7O>Q zca+o`R3T5C*P|W?sa>4Ijh2kDmr}jb0P{ZGf4 zV!HRG)Uh5=83vCtPtlLKQ88N&-;ddKSQ=7%1D1m@2qx}b{#pWUTx6?c?X~}>9`hT|@#P>WXr`bU(>>?PViaLP-0|*SxUZ%s=Q<*Co8kKy9?nC zxDDFa=IkdM(Udx5so%UA+7Xs~etbq8#7da#N|+?!GYwg;pF*N|v$O0iW+jIfX6oWB zV7m4jU2*N_nrXMjS}b1DZ=eE3Luh>wwsD27Rm8;SfNB{-BJI|0)rj?)_n%~)xqv3y zSy*Q31l)n^Rxtu7z=M-;uXf+Bcnj?|VjagGjZ(kAh6*Jw8QNxFvTGpONZ*^XZv zy&8EX@*?Kqix|DjlT#b7YSQudaF~N)+fS*gkSHV{>&odqjSN|h+Jta+wQ`#SjlXEz zV~uJTa0dq@7gsV#(5j^dvs8vX^FbD zT7x|fhT@9lGmW(tN(DSSQBo0Q^e{)UvSmBk2yV8#CQzHzo3SIOTSGJ#^-$|nDxK;r zWv6#{Y~s?9zvuq2d2#NH_cZDeOE{b1d3F%(732_@RTrDjmK86<`NF4m)Xj<; zd(#JqvTOA13FYy!jf}~SNP+)M8nTEoFE-Lvr5_Q>vHxg{0ITDwLOU(4jcDdWU;2_X z0U>m3E`1Hb@ki><#$E6h{2}Q9|lFHTTID!%RrW7Q{P1 z5wid-dEroUP6WW*P_w)kBL@Sx>njdmJnSsrlpA0;- z?9>4+Qx9pNlhzn>yEMcn4*WYuM#sL8w0neN8_izY^L#mer7KJ?{?M?XB(c`~Eh)?2 z9OsCY4EGb&bhU0sWlKD(4=S%H)@=|?dDt|WGYUdQ;oo$2n{a= zAK=&-rcWBzM;EgdEQ05wHAf>?rN`j$<(%r{1QV<($D&fJ9!wOu(vM#Kcl1@BXTCKr_6JNmEQ~DA&bSm6=v@0 ztTNrB>~44B{Eh#0z{*YP!Yk0d&YzPua)=0pRl!YG>OipCdC_-DA5 z**_ihrzM-O5^*fwh(NPmvG-vmLY<(dB(95UH>7FcTiiLxb4BUbTm|Wv!nMW4y`{j$zggN^x#f4% z6BtebD<#Q+u6C013-~|%J&HB1&JQy=vQzYRCR_XGIkoFcI6?^!He0xPR#o58PN-fm z`?DQxM^Jty{O7$MNV@Mcw%!8X7`KJ65X6x5ns^Y;!kLpq6FQ~RpVPiJ~;DW z1w}_#Ks%$(mvB#dG+q;sTiZJraHr8MXtY-1sVcu{{v2Q4Hs77nw$y+?e5d>-)zo{( zcCgQso_bJOr(UN2pv~$EZ^c~xQz-0X7|Dd|gE`2y$5i$Uu$oq|QFtP24_c_-`l+ zY#*}fN|%m=I;IjVoObToI6Pjto#bU6)J->AE8-#NDuAPeCebSt2+kh1!zNb8C9{x!{(~W z*)pu?5RkAuE+EWTbkMvw06?*7-BP#c4e-tjwht{sYoL8!5E$vyH?DD0v5Kwhm#=Hr zNsfdJs%i|Qeu=?-X>TOs{A=#!5Pu(VMRF|KnU@%B_JibPwJ;Mes6+&QE_rc8`t$Un zL07?Qm_0}jM`c%aavZ*tTSLIwwN%HfVmI(EAj0>HP{+u`67CX&7vZ1Oyf#~_2r*z9 z=jTIz_U(wUe0K-P>F3!H;-BiA@&fGq22zWCXa3ZaUa48v1+$was4ED0V1xUw~Af<0O z-J$Cz>w;u=XTL-DFbnMO%F^VQf9K#IuQz8rVh9f7|IwDeS%54c8K0wCrRDLyElhs? z0qRPPQ*(>AxitVj-{WsH9Tir2L|?7F5<_Ei=A_kSCX)`x&^4MEuTztd1&{#TR8MSrvHFf(8y z&=U}`kRtn?lxHWJ#chyzj{Su>tZfOR-)2r`*S>am;(DZREeYoWL7>}&-NCvEaYT=6 zhPSdhqPEx7o4XoAc(ZN!&D>+iR`llcF@F;~N=}pbkz2(k&^aV|A@CB{wcoOC{0SW< zjnVTy!t$XYD_Oh39qfc(aXsV$oR8DB9u*f6LsYZ(n6Km?yZInWL$%$2Kb9R(%ZYAx zf}|;I6RudBF$E@^@a8}Hl%9SIzSxbNBS)Q^T)nMiP2asLw$7h6xf}jPuJVbBky{-b zYg@{iavGPIzjvcmt7Ve_Zi71>OrNjm_cl&^g>N|z+bM-KzR+|(u_eprTdW3?VsBT1 zPC7(L6-CZ$6EmIqvY0`l^wv$<8J z$})h|rBqJ{Qn$-YvgB7}pgo7TJ*eP>NSmdnh8YQOJgc4@7WYgU<}6PGO!w61(Q9#p zFHkebM}%YLiCL0A8!flbL_x%KlCJGIPR^4k9ut6>Gg zZ53gzM}>pubdXM5Pvut(6m$Z)l*Xp66-qAhKc7u{8SJ6V;sgjz?sY&h)N&F{>SndH(t#RbdgzcEY z%eQ65JnF-g`OFC2Qw(gCA;2}XL3@&~s+$Ia7i2ZOPD(`0E%rQFWG*yn28V3z)?7~L zJw{8zn`#3vaB;D0w=!z3)FnJm0yD^F*hUut-h>yZsXcMZyy{$$aEsnC{(W*wv#E$Z}d29BD!k79gi)dsJKEnNE{Va%va5h@YP{{YXE z<=4`*^d{3vp8nsQ#_P(#E3p4`O_Ft}+q!U$nX$rT>%lX8VD8xg?L#Z)h!dl;T2b&k zyrj}xt5~RzT$Mj`Cs{6twCU@7Q4XWu)L}bIU-yPs5sqzY`~t`R$k|Z&S}7j3&S!Q# z14wgO+@z!!$w}qWXon{s6_EF1z4WU#WY9lp|2jcWA@CTo+pGsP&nRw|n2wBB<90#p z|C0FCRB>et^*{}-s!>bDoMFp^*JAH08*SYiaV4ZnGomPqDFBPuMaoayZ5!t4JYJX0 zZb96d>wr;6q`ml}*;kCJMl{s|k**o1y+&d~zSCU5p30(^%;YJE(eke=Egu!^$Ek=x zO~Kf|Zf(xYK-|E+Ha2e3wS*194Hhe|D^G_jul7bUOgjP-KvHB@2BO{yT3a$y9}&G2 zvx!7uyJq`^`_gxfTu_ZRchy(u>4!8t#WT6&eBsEDCZ(uMH+nPb4WfxiNwM9kYnrJ> zaV&s*pOB~Ko(+VO^n_Te99 zlzCVaOGbC?qv-c$W1rNrh(hgZ&r?4RJEHL`@heEp8Kyo^5+s-F zGsq|p(f`LUTaC+2xnJj*1eFCr%INR{ee;r=+!I-NUogXdD_1Ew4{G_)NiC;V`rZ#Q zm#SNY)gZ@DDt%B-*=s?H@`%u<_epUTrYWigjdor$(sk|p#7ydSV0 zPT3MW$KIk3xi!hBw^lS|(b~E%W{Upqj(4_x*x%bdp40;FUC#KlF4349V&+3%Q$-lo zP5cn!?uH=XPP#k;Qp3Ze`K^!qE`dj<(722N*Xzdc9 zzz@R{fHBCc+}G*vGQ)MWtkm5x7>uWTFGBst{BaqVTfUFwcCnh&A+jIr!8 zCu+y6<~%6+`@k)P=m9)kN$?{F?+kId5_M~bt8%3Eylt*vVyg8&_zKx~I#SWV<>#qT zJ_njdeNSpu<^c{xXR?s8+#MaUS~T7=z_vO|+6dBxXSm_0f4RQbOI`u<)S_X<&hub} zu!9#^{3R1oT-})O!A6URy(|HeZLElOS*QI;KV)=_m)q9_5IGd?D@*-v?&%%3M8hxZ z!L0$M%|3}~y{ew`!!Km5yq6{Nrj?tYZac!${}B)VA_EkGOhQaafi&; zRL#WDNmI2W5PXX?khq8LrWuoC9W-h-_$X)}d`j(QKgb*?e5YrqG!QO%*vy?tEZF^QEy&SM{=yb9&>8#Z8{ zu(+!6oGWGF;2@l~R2MeeQqT?@A?)|ff(Hu-1(qcR66!C=2qF9%;f*x+leNf|9KjVD z;jSNDYeGiqwWa*sq?#0ku`6cKyp*dhgorP`F%_SOw*ro$wcdftg$o#6v_FZ8gGC8s zR%?AOD1StnIT@uci|jMKbnw*v>ZnKvjlY;h`fE78ls9B)VgjaD#q z1E+q-U&F+AY%4`__6FO@xjshDcrMWLe0&tc5Z^@SVW|ksR_9sW%IZDQ?;;(66h*fp z8~t)cU@WwdU;c{%_L z{lxAo2sH-9m@c(5Z_)06zd=Hx9VgA5{lmNE07h?w`B>_cO4$kRP0+HqK6qPi7_%d5 z?y)gO>1C(cfAQv{2O56REQG_ONA;CziH%5eIV?i~m3~~N`D&PA!OG^nn&+_fK+8=` zrLXE;*+-AKH$e91jNR$;K?a6n2~y*sy@#iSz+`=idE3O&MI^Lc6Uh%RK1%vZ6(#_} z*2fX{d|lD12h5zsUdA-jU)NZhxT-i?^NhchX2+XPS%Lc*TQNW&pbHn9~ z>`z@tpJ!53_xQei$ilquoGcyp7q)b%C13iWtxO7Vk8f8i%_E$TcJ9?s59oztGsy0# z{YHS?V;di1R(UtgP^qc*!*-kED)$~}pV-bB%skT_$D8GY0x`c~GF}(wm!xkcwlm{s zmny#(P0g<*y%zRklt0K_e>naIB9Mgv!VJjqg22mWMspyb>PgO1q^mpp{(d zjVYMyZH2e&?{+_-l<>B4-UX*7&W&e9A#oR7)e$Y4i0jJ#VF{mrpCn#xTLu2{!})l^ zAw@2j3D(x2M9BH}jHzDPpNf0e^Qmuisj+iY$}$tkT|S?c#G?&Nwa?3(q}8`GqJRv!PZJlvdtSKvcXH8X8S~MZZ$F7 zaW(8O!Z__TPZ|c#XS_~FWfJql0C}X<&hP!u0JETrg? z;NRgu{j#pePn^)yeiz&Gj584rMa`oRGu(9IvufL62dBSbFr6pZmLY7AjC*ijTVsZZ ziBL0r_EO+0;Cdd>y$(pxzgRp2%t-q?z+T;4dFTr?TIq7yZn`AC+wZtQ9RHOb@@e8)@DEtbXH2BQG`S zmng7BJKk&~G`Sp+`XyRhUD23{9Rgg7L zU6j8Z?Pd9#gU`+y8PXqrVij3dk{V^bk$cq084lKKE$Ws zF(~yj;BNUsZ~*9lk}D@eEb41vtNDB6Bd(w1SdbE8%?o}I)RV=H2Z18F)W{hn_pRne zhTOCh8-0WP#9DNStcY(y?$rzgK>u=1K2$fzjt%5cp6V_~8@5NZGnyXDN-}U)8=GzE z<>oxM&WJ@fuc(|2>G!tmhm74LYKZy5k3(0XinUu#JhAUEiqx5 zYB?8p2@?;8XTsv$e29mxNCqz4pI}}^sk8#yPJZGl?0?SfW&Iev)VD=y%zrIF>eu8} z2ae>JIFu`zjHqMZ2>>X`E>c)-E!lplBPls57XV(^)=sOeB&%BzWx+2$(l4%1!xE>|lSS0ASK5Vv zr_xTKm*A>%$>o&J+3n`~`QoW*8u~%ZGUzU`mwB3n2v^> zCGF1&7`zYS)+d|#srDJ^Xvf#pUf7Q4@vbsNWWA1-TOmmJ64jiytDmT!^SdxAM;{+G zor&_6{xxPhQxx*ix1feD?zhuwo@)cP__Tz1-Vy#Hk@!%)nEk7z zZlBYu>zrRe%l|Zntvea z*(O9Y*#_@MhG~6;Qt@fIC*=ro2cn!ikDU9X(0bzH%zp74>Hb8#{y_o>^)>-DA8PCQ zEbmQng-6zTu z&2UM-W*(rrNj`=g;z|>9y^}NJutkNpsJ1fF3M&JFBi)J%8xZY6>;j(g0ZfFF$^RwK zn}S1G7GxZPR*p3<8AZ+e&HY{+rCLX(8g4i*gi4uB`E|MC!OJXb#Bkg+N}l%WG(HG% zYr)(Z_mdw!*N(VIEfQrFVD;A@;Ec%hFC48b#=qzxTsd73W}vY^){C*GG6#dCJTH>X z``7(y9a48%&v2l8=c`?zspXPfU^1BLzNM% zc((9wSdDda*$96=kq~MgRgT{#6{RRAi;M;R9?FvhScf)Er|jcc2xW}Ga?ZVQ+((A& zNMV?$B#n;?jU+xVkLUNo2jURBklC=L2#5Ce%u~SrI`AM$DSRoQ-wVlS>CPQ5DVc)! ztG=_fp7?7tE#c@0B^E9%irLI5F=lHwRi>uH8_rZK7v#x5%QXwFReSM=`EX<^jni1m)M z`S3d(8=@oR>1^}V^o;Ks$~!#c8Fsr+{X~G0Wtih;kaB+|MER%fwc{#gA#gn+E(?Ec zdS)bev~KW+%8#rH1hIB8Grr-t==pJ>v}i)PhCICMm9A3)oJzCW05m&@{DII(C<;Xt zn{^WvtSUt)PLmpO+I{cT@^DO1R5H^z?>eX4uL3xP4Xxro>L(U7-qG*)s0y23D*tHt z4P&re&c?h9&D|r9CRigWllR!l=sGN-S8;B%VNYSxGq?XQZf*gFs;(e2iju-zwVa3= zNw=yScvJb&Id{lfJhy<|WQ}vZr2;=@m~solOj(23mv zEp|S@;CBr9^*tVHqEI0C8C5HM5SR6<`uAA%O7(j69l^8tf2{FU4wmwHn0>wWxiLs3 zVS(<#2O4V#5>sJgp{6Kjb(%ak-l}4O;K_lQGm;DT)j|O-R=v9CNj4avCj!R8ihy_I zVg~#TrK+dEpuSR6Mn5z=G4U0#Kif<~5EMlE1bBJYkDIYq%xGP-*24D!9pS8S=cg7@ zB9h3x^GKy>0?6G2GJZ8GU%F~7qpUrov-hfLU$Jizs-IxgiM?sT^ymuZHhyfGvMAgg z&z)T7Z*9E%59l3uN2_=`+CAOu!&o-W-ZhY^$ao4p6j&Tim@1Ll%EJx3 z-IU(`b-L#~BiV8q@~he-ewqo$_a-lwtYv39(!{gDZ!}en?-<>a>+CiCUl|!4^+}=o zuqSvY=<^ucGEV=N^zGDRj&{wCc<#lR=J?NqFysA7{^cyoS@IiP=kR-Lbp+?SeLTOY zQI`qc5$HDNymUMu7a_x<@1`sYStq;(Z$vHo!!S>KsHt2F(Gsu~_ZnQ`U424PKy@Gm(JephsvJ$FELJ@_zXfPnd0 zX&Bn8DXdj~LaM9OrFz|k3!ss!=p*oOjbBs7X3QU~r_MKmcnKiyg5-#tl~lz|XP%4h zE_+Q3H>h`wn*WcYbMb3I|Nl5bh=#C3mvO>e+aYvY8#?TZqak$CWe9N`t*o`Jb~%Kp zXc9tgL+GZaLbtVb6{StONta!^thQ>m*6!PHzrW%0_}pIa*X#Kz;W#73uA+YBZV`UOZ9`+tS4CeQGSw@pd5hg188WRr0#6qyxCD-PVETFOK77W>xPsdTOGt`s zuR6?3YR*DY;n-pMG~VD^J&*fPgYUab_Dy&D*YyqeUCC4#L_MTFa>~R1a*MvjM`U%E zkpm|G>QVg%Is9O4!#Vetw;lqr&b=4KZE62JG`dSV6Zie+!RnHdC6gy{xZHZ z2VH5^g&n1d&+AX0aGrRPHGVdS{_Aq?hR6H%?0mE~X|CV@VBe_shXb|*WB~#eUfguM zSca>sGF~aMzWX?+B`_rx%s>xmfLCJPp>Wrglj7!}c{b3Wc!%DD^R?*Wlsz|q6j%e5 zak-hYj~3JOgjHd1md@F5Ygp}PdX4vELIzKk{}(mjM7;|f=}>!|T8hI&dTAL_M303jYv`o_LV3 z2A)rTS!}DrhPUDxF^nknCH(;vRi;}wlWNXqWXp)_$j8|8bN!3thb)^Sc545&WNCb| zB*d-pZR8lT{+f5WHy3|>hU$F~Ry7-x7M_p-e_-|G&O$#R9RuGAK)uL)3I88s1*}EP z8pTAy8S1MHApEj_XO<~jQoWtom=iK8$wsQrJC>$_vdS~8eK+kA-3WuK=!*WA703OL zalSy3b4B|V-C^f?CT{p0r`-hI0nR~R3E$CEqubB%GOo7ZA8;EAL3hETmD1!& zW)_~Lr^MbiLj7P=x|h9lw%Nt4c^i=i;l$;q~#+ml}DKN`h z&G^7dfU@?j@1q1KmurgUd%mqap*_qtg^?~R%6X-7e5MZy9~I};mEHyJT-XLb1-4bs zWR@kyJlIb76NU&6=r3k9VQo^m&|bX=4dm8$=ytPPe}xd8G&9T4Eul>p#fexuHLOKm zr;STfv>7Ig;0uVvzZiH0+G&WwBkV`h*KBZ+hfWzF6VzgT12`T zm69Yz3om$$nMREAw;u>FYVM1?lr{+hkSCRbn86(#^8T1UU6 zdX>?pR5wyHj7BP&v&D+9m5hoZ6?=730W{0i$V zN|?~^3vk-6@aj-kTGdNp8_@~6-RR6p=ClWrM|&D~k0cNA^R7xX*LkjFLA;ovMt$l^ zS5?L>Rg&NiNc;`t1C1xOw{mcih1|uC^*tTJ0$h*i1xWS}Zi2bCel#-5N6mM6Bu9^DvH1S_0Lp3|t_qhy3j6X>foE4ZZ) zLkAul-0rqVdnVl3F~_lqR>!6H+@IjPTQ)Nx+<$AFn9+*AdrA1UHAk=!>O0v0J{|NK z+~j_)XZnpALzVhV%cT(}zlbZQbaIUDb2JZ_xDjGyea+gVUugWP8P-3KTpQ7l+2}0t z5-EV^wNO+A7#dyivn+dtT^#;A`h?yyl@WVo))FF_$0cpKjSo z4Cs#*h$J2DCs?Ps{b9E44lu1Vl!)(E>*1C@*n6`07&(m2F;wVMW>}b))`igzvE$^6 z+9qo^FuDzNlw3)CA~*qA2~N?n#R08}%1Bj9blnCeu~pmb+{&LA9J$B=(biaRx_%Id z^tL;KX5NVE74S4EL!yaF57|g%S&z=nn9qeeF~4|qB`Hsap`;Mqn&KqXFj{f4%g(ED zv+1?2&-ap9xF~h3r8RN_96JhG%bzv<}*k3o~WEwvO}w3;AA zNtZCP@UD!B><_e7NlKy$s;HrTCOEOmQH#=K*`o=<4Y(nu9;d13-m1sUm?&1}8 zZVOEgEYV0OzmeY~W;WJ9zEmq>^p*X5jqV(w#5CQ$Q(Fg+=J4cA{ zc(-%jI6u&w7EGJ6?XjNvWo|w7>a&6-FuaCrE>`D5LtsPDe2{2BxUN1xbbs1!?2bui z*y@T*ND+T3OKuhw4sdx#V8LvWMc5_|)?XgFtSjdHS@EXg2{$=B&y|WCg>}`*4cy~g zUn0g#?hF?>%wVxw8i5QDnT`DfQ5k2vB;@Ht5yi&*Wj>Jg)w8}xlF>g&v3F#e!d7}5yu zP912KJa1Uk?42cf=aD~Xs1r^O0dz$5X~mi@*(8d+Qgh9`eUz*KQCo?{n*X>%0VP>O ziF%T^$($biG$@buRolms&Kv2I7)7;szWt0HG?+b_EzPNKf*Buyy_Ni1 zSfeG5;Scup(Jh{iYn;p-*-%L~OvS90S0Co@?72eT8S#;zP@L3#;2_!Iu^*tU1lJIp z>2{Lv{Af{L=ods+h#ohyWNK~k%JO{p*F@zpXV!a+BE-^B*(Q-ypMy7|4E=rRDVIEy z0%FG%gY=Q!fs-|`*^+MGmgec2dz=UP&LYz2y^fJ9@U|?O_BbOaVH+&BDiyyM3ZD*B z{H@yxU94M^onCVh*V#v6nfZei$oRs9)>ac~43s?7n51g~Cl>Y~q+|LT!*%^`+qK?O z1THI&*+4u8EoGxD6{!+$^K+8W)Jkb&sbu~3*~ylLu(uUO>>-m|;OnW2l+uneKzCJ{ zm()|5*Q8Y5IOyMw9XcuoCIbqu{vPofvUchNTDQ7t3qf!ZyHRp$>Ojm2K>BQ>fI%pK3Tbj1x;r2Bl}j8%W5Q-SK6y z+D^XyGn>&WyRKQpKLuP#3k7K-K`kudc138_Dc$PH@|}=F{);HTs=jy~1igZ?@Y$F{ zkR|ft;hTEqH4g!sB;9t_PnvbgPo#TZO{nihziCGDBV#8j0TCGq&F-=G?8LFzE8wTK z=O70Kt1Zt+^?=$w5pp65dr^2_d?fY}CZF42sJ$T?SG^J1de>mLy#p!9y5)>(aJ46Y ze)KFF5Hwitg^Qs4YnChiH!~6Qb5deEIBug<*{RkP%GkwP+H`n+(lAE2X)^Vc{bQ~W zG9uTVHrJ3tD*S7fVvoc>=PlsaTbH1ZWdH6FGu4H9mnmuDeqf$*y(;ahYY^RQHNg{I z;11|%vbn{y75;;KPWfBdoI$U+4HJ>jld@Y>2ybcmSAM@2tkeLHAZVkcrA173K*!1BE`zMa>`kK6poR_5c1KAK%b^Hh5$OetQA5aNrLI*rQ`_G%uBJY)v^yh;L;fbdZ; z3Py+2)PHu3SweVa1fnib#>-d8%i%$Tk)VeBAaH&8kmS=4oHx(w7{4*H-tIxB%99j5 z(^k7$_;})>Z>Rv2h}o|3H+4oCwvPBs|4ns!ph;v%CJCP)vj^8epK%|s!W^dNNV=J` z5w$%@iD%HUZtVHoC$5EVLSgRRaw13la_|_bKFr$w1=Ex~&71ZYN@Ug`T9REE=p#g> zY80s)$VFB5;#9@`Y^G|rm8W|~`T|`a&lD_UbvxE?eOE@@$`9hQ%rD?JmTLCj%`L??TWnU93Zg7NSso&xM zA#364OqEt_Ow@DQ?GI9F`QmV2h}Fki9Lq~3q)KCR)kK9UVj)KlvpBmPNBOYllO1U- zT5}+WU)jvuZU4}5uA9L<;U67(Vtvg{EyV3TWPWj-;7gA_R0G%V7%Z<^2Ac=_q8}@` z3qUyp5@E6^@?>rt_qxB7mg=DA5^0tV#C;X_^?cL!-%Ur^>k=}%a4-8_3vQy=6&6qS z5;8(wZdh#y)15PWsyv8)LT6zk%-oVtiHg|7oVaCKCj5tErby0HGJL z4l@fFizL5(!>9wVhf7Z?T(CQ40tAmL6XmxMFUi|LIY*MMM|MR{6-*4CD!l%pi}}&Ku4sifRpLLMS$177EuF55+|Q_0<(#Gv8qK zl2gEUfppHjpYt>y#Fq)Xfh!oD7P!(7+%1Um?}}^;@M^G&5g7J(eAnb$>g71*!JOKK5N>{mc;RpJo(!1expD$nep7>a`Fd zaM)EmOSzmK|8r&4+GacXZTGv@>B48Zdx3{sUKVe(IB|s;ov=JfC|r&j#f~}p;o4X> zN3p&mbQA4go^olXv>}5wqVBclFX6rv4H07OPDcQ&G<1;G~`WKLs5c)N-&Syy3*s$l|2_<-N%-)~Cp)(nR$b{n#S_{PK@E0Zf^;eKY+m?*`q z{rK1{?-28`h!p0qj|bMT^D|-ufsRv(FNgknDrf{f1}7vFJz^ zo~|1bf1^s@@X*Cb)4H@uAa^cHLozI|<#0#T!~O&~`Nui&YuY!`C}<+fw75BmTPkl( zf}9n6P159#TfOUOIYuAd9r0Yvl1>R)zqVs({t)D#pLRW%hTAl z%5x?%p%uTN*H-C9JdDjyJ=R^Ua)~)fxGNJPFsSmeo&u*4YnPwCugq{u(Kab3Lg0AO zRhf3S{Z7E_DHaPwwzJ-d+Wrv!)c&&dheLuF|9&*hn^DsshrY;o2?}vp?YBjc3$nPQ zC77S*66|v*^X1|mmDNBVJ0a87Q8LhUzJMu&ryLns?ZcWrG3N5beoj;x3AU0b*$+e3 z{36_9*4m9NvqAs=K$u=(9;iPTx7s`|S-Mwompk8x!jf{2!ERLz3p11sils7T(^!cL zf!}O+gC5qqybVq za?2xjOwaEeAA6?OHn}RU9CVv7`DKbLrK27gI3YS_zqd3h4oxNBsYrHJ>SwND@fyB2 zA_w#*F|_~rKqI)>`553Z>`FZhMPTmy!6M4Nh;{zID}VO(!~28! zu{)GE%tvT zEIS>+tf$c-&{D+##j-U5rKA$=ySTZ5U+?-YvUm^{G0z_|vjn}!&CzSAyhC;h?W7&C z&2#lj^e(2b0h9?`{6rwkDNrZ(b)75n@|OThK_1^hJo4L~TG+VLWEhPI3twYcjT=IB zUi8ClmJD^F!mV!j)eHxJcf(WZuIc!nn@M}|dqH4c4vTMLe8uG8Uhy9mzazNQFmU|T zG)=La>Blw1>aTc*PRX6H`C;;ocZHaVgt0}f_FWtJKQfPN-@wy=`z8HVbr2OH?oJpV zqxA28#?A42$W86ogJE>mG_J(&l|Gwnc$GNfGZZu)&|-N+$!m}r9|QB@9~laa3_i5_ z+C@@h(daT+5@9pB%GWal9NV|AyTbx^vmAgeLagXXMHC>{yFCFYsV4pRYkMTZxsDEU zP(q>fwRjv{8R5Ne^0_wuNEW58SUD7eAUnY47AIIx?@)k&l7XM|*vR-SbG6OQOF6=; zhT3~z*Yh8vOlplR8Z;0=W@7&cb|yD-@8%Lw$~_puB#=$6< zC7y%-A9f5OSWQ1)aa@Az@c%~Q9zZ>tq7r}ec*|pWtD>izBKRIc`6Cpc`L?(Wzk`zN za03W|0h#sY#)rfr8swF4i{a3b62#Yg=cFBaPe zZsL3*$3(|Im6gJUDvL+Nm&&tDx;7&nr()?mroX&vP9%^GbA?bbS}oQ{vf#Pa`L^%o z1Kk!w+Cl4c-*_xwj4#1t*?5O#(YZe(NwopsKiOwEAE8@flYF!BUpb9Io!l$olSd@| zJ~^O!!FGaPP(^dG@d5CHE9^#fEu}PidVkt^Kz8BG6-BE-qVyhz%4=lYB)Bw`^eM;9{SEO`DeS_u$hPV-HeDm=mU7oE*;eL9d zea|bwa;v}+h_zE*Gput&xUEOmIXud&d5$0RO!!MRTa&~7k)G$cfSYVRh(rrlj+(_C z@s9rfJ9N0IBkTZ>gk)goS2J#T?YFOG&&4bqzCpWA?o2G8nA9@)5V~vO2zXTR5t$va z!u2mnx9O(IG|Tacq@R5wyp6L5ryf-_59V5p*QWwFnr@Ln`xL%50!aP$qBS`{S6)4O zAsL*6oZ1%5LW*W3_x0MDt}x335Fne~^sQi7;@UK-V5l;sY^u)w8|V>5(qCw=LaOq4 zU^k5DoAx&IBh!U3#_5V2m$vjYO^RDg|1x%l9SytwOZ|x}3|1v#9tz*g*kyYN_-?Ag z2~RTAvW?r3K!yH2Xz`R;A;=f325T*yHulf3cDk|c)JyOrv35K=(zHaK=Fu4aX>cQ_ zmGm9Uie6-Wgbf0JM)&`~&5l|IsB@mCw#tOZ@qH#YHm>vH@U@stdUs1^f-mWW@{RY& z?9GGTF0nm1C6)=VuV zMqBGbl^%fyUG}2?A}wLv^e8glK@3}3%0=>`!m4u?K2a;JcXV4}+-+PhoSN1b9yup4 zCuw>dBK?(f`*Jt%nzK47BZjskby33dX=*`+WB6>{uMpDwB+#z}!I#KkCxx>HBJ1Q_ z>S0Ri2Y)=iG1Tg$G*6FbN|LDJ%3Nd+a{-B&tFSgw5%_@;fC1+VfRV?%1zQJmB&9o4 zm*Yov4V*I|4Tg77Q>i>}_F*D%i=6V7zeB4>Xiu}Wjp5?g?5P4fp=;sQMtEwH$U5!b zw4^$jy9Z%Pf+OHZxKGK6*tLZJtTM|}vwuMF5&yTCAmM+6nw6Br+AVw^E;UEAfxQa) zLCw`+HS@JO5ygAdqFA9ZR@_tMaSaq2|ub}6$|=lQ_-?_Nm1@@=K+ET5+x1~HrZt_Mw3LT*atA$7ZC3n1{~MD(1F-U2^nsj z+C~iGG|KA*N>WoBK2^NNvfs@;Fg$oG3 z&drLeV<(5I8+^VUzZ~{VUG>;}**cbLwo20~A*K-J4Ja~fFYs^rVe39DXxQ0s$)42; z>%`gdJ++~(vC_ZAe%Oc=b7?DY0!TSyJk=^)b>OypS;BD_WQpxL&lB6uL?gQTE<-n}cntJ~M$@gc+a2Jr{ZDW#;mKn3Zq!zPG@Ezqs1nN^Y zk0rI9P{2%5&2o8iY604Jde2~R^9qS0i8izS$AMMB-@zHV$}EGAqT-1jDhUd2BfFUY zq~n&r8!|G%&jZ#Yj}DB~n6@$MGdhB2b3Gto*##bZBu=8)_>~}gT)lgw{X5xIo}V}! z7DRu}R&t5ijb|x#iuW27AN|937Tu-U+ zP>aK;Rb$^vR)9+(I(*D8Y5HB1FrQmtfl0n{SdKKyZ6W}U02G=p7@n&rsvo@FZwfu9 zmxwL`@6`O_C&;Mh0ix6R#p3mvkJ4|cpw4mx+gG!g4k54c2(uXThQ=f(0RL|y_t8o> zG`mBVYfB{Jg|^6T-`K_!!M6?Ns9rSxfa?=_6X0$0p6PRwiTGHJ^G~ql4720z5o6fQ z^?^v(Wuw#ZR`FlZ^E%^q5Y#QcR^BD}d0d}io&|i3&POubo|^w5{oFwmN&8313;;zN z1}&tlbStL(9*@c3Wcoh#c`rUh6bWe7oXt$9|D=|W8Y6Dx zjDruk6eEMpr%a}0&_4)W7ujIn-^70_10qntJ*}4GxJb~Xpjppd!ai#ro>{_4NpOyc zoJmc{@jhb)xV?cLL>&nG4DTpefh4aPRCt;uph(h9L#iOzd{O`e0I36^H7%_Q2jze= zURm0O=(RT`WlqaCvR=DOIhUA|&UO1}3QT`g%nnlaOs2Q+em@EuD4Nu5t3(?Q&_x~t z8DaMQ+BoqdqPe@TRIWH9_=>FohAFaPw@l{4j3kaQYxLqJc)M+cUyhn1*vkk&duuj% z|FE&~(OkcZSV@Wm<5uW2*5gw3i%yx34k?Zp(Zp=tX$S&k% z6sBV;<|b>*vP$qo9!2#>Cmhi|Ql$_r4MJBGKGCe5aSk&Pd7Fufan?HuWs;+DLi}jI z;pBXHZ7rMQLZ6KZ^Je0_NWPljQ>DWH_ zwDmJFleaP*oj~E(3uOdlkkD(`8KG+uhe1UYwng20oZow^`ab)(xkR0(>gPAJg3G^RDOWTF#s8t! z%cm<1bLnXIMcvJWQp`2ZM;Jx3UYDVXjVAFgF=)g)X@_Jh@tXorjiaGTvt--{ zR4Mfs@D`>{Hnv_G^q<`?N@%59!i+-JPvyX74ugcd*@?v8ppVB66%_ivD%>kSs;%XAhH$_kIY&l%xucqMsdk+9cWR|ONe<{(n9kKkfwQ{(m`z}E)-QT*PC z|IBf-C}yfmo&hJPhc`i76GmgsV!nA?q%;7A5811pj!3&Obvb$TWBeS-9BU_P1FLmw z%~Dd7;rD^&$BoI8#`o3%u;~DIZnZt)La=Q@c_)n+S8Em0-I!v2=+z=0{{u9J=%TH(%EC&IFtV)h45EIaTY>RwRm~rUQZJQ7)zz4+ zSfra@O^Yf59pa>Whrw-^pzD54>>gvG`u1m&AM_0mvWL1>a7244-L#1RQ}0DNEJ&45 zP^2ZuP?znnZknXODS|L1Ma3-A-!S|`C}eCkg$^vBgl5A>ai>`%2nCsq-C|Azs#D+z z$6fs5s@w9TQ7fw9(qZZ0iT@cNmGOl^Z8+gYX{DZ`k&FMBwhF_fAMqmbR?MzW8=DKy zG1zAo24jhuy|J&LYdk{yH>o#uUL5OROuRJ^&+C^@Huf0~DKf&#F^)Z}v;}tgeR+`s z9W3urVY2hPBt6)4ftl)?hM(&p)ZYa+QW7iqjc^szk}}YMA*d(A5DYG-8F15X;A`cZ z^8M&T5pU8q@Sz*5YkVukyEPSsy*;WOSSM^;Y&Nluv%I{%p~Z+}%%`}ZOQ7RnK+_M* zzb@}Ht-<9VI5~rkl4IauDm>Yh^W3x&b$Zaj{+2`CLecseCH?0q=tYvhbR&MTj$C}i zx{Ty!dc|K3=ub8e+iX4#Y|T43se~tafXnCB!#i(}r}s%B1u0I`LqBQaAm~=#ABtuk zm;+r#{!DbSk-axm z8GEVtWpUJgKsBFIcT}G*?Wj96av#5s*ZR~x2Y?_am7jzj4g)~Gk`o!^{iO3;<)Bn^ zZl*?JlL5JgtA)X&KTvH(WlO{Yfr)hhyid^6@D5<)tmfu=6aY>OZYLyrj~;_Q<2NKF zrU|Ic^<04ef!I`7fH^D9+c((PIDVKX^6>xH*IvDGAIfT<{v!K^3{QwX#3e0e zqH=>Hgoam~P^P!Y-l%`KmLhdY?>t?-s_bT-kph`74Wt}U0Q2{O^~U_F@A%934*Y@; zoN1T$BXYdF;YeQKKiyerDVA3=sPFWdMVYJdu&^R%fNlfFwfbh}Uh7flQ*4d5)Lq&) zeW>Yb%-u+SfXgiTV^D36@Fg#~;y1uLcp>X6s`mi&MV}%pHc}jEtRVWv_mYexae}v< z8SlN>bT4|~D?U*1kMWG)lBE8t9mX=w*~WPw_XODq5av#GbLw~2TEWfMva7m9iZj|P zuypU|WPr;`0OsdBlSOr=QLBezeFy}-gyA&^@SB4MneOq5j`ajtduT(pQfglB`!dB| z-`N}#3M-L6L*;n9X8AC+Pu=RqA{#IdsCRM9`_R3DM+S@6`r?S!XvI+-2TL0uk}E7` z2DpEz)o@V!jbAEPe(Fs9)^*!GPYvO?GlBjDZy$~iDGF<7O~wUE$imH1vMORmn9fPW z8-=R}_yT3>bov(-GCS0qA+`lZK>|m5}g!URR4@VsTgttT)%0 zq3%uq41Rv^6>~w3wk~lK(~1Q4iR(NRE+cQG9=z40K=XNVlo144$aLWJL$e2N<~E~( zSqEIV3&4RR&8q~@_^n#X2IB(m;fUns$iWoo0(cnt?{H%L6RxXiMM5#vFn6d~F(%wR zy*DDo6+bID$>~Q>5I?tk)Y2PlU@QKSJY(K*7_u9(t1Z_B6ZT4fnmpY$CpPd@sU#FW zl$S$J)EzK>nm&4w)az;t4nLjkE%PsOa3nK{^QT^UBqRRgE}|Zz6jN^JhJZ;6O}DW< z{P*9q?ME6AN~?)3*%h~~Aak@|NK1ie4lUsX%VJ0W8qtnkYO(Xr6#ulO*+-$GJl6P^ z6$x9q5sYn?2cVWDfPM6WC>AG6vxZa=Q`L}dr<`gMuA8~7$pu7l0M&bu`zZS*4@678 zkH56O534DN?BB^f!F=qEoz8%MhHy(PM%8Tui5~qvf#qFjO5lTKqilADCRqL%IZlj{ z)RWuolbL&TE2oDGwr&`F)pIARh&)9SZLyePrA= z`JrlWCSLm*rKZ@6P1!kDeYDdn;F-GA==h;S2_tn&LQ&3`xZxUlPg3RrKoFu<@)#&d z?A$#@yPG?a`+3QbsTS3Eq&jEn-9Bv=uAQ8^+`BFH64R&4u)Rjcv_#UQ)p+S+l!z`!c3Ol7-YJL_i=S6X5Z-GW|+n7 zeA->02(CMZUu1YWQfPXv|Cm(?Gm0a~9>^=xk{|R-pz&Pp8_v7p=gdi*dZP{yKXFin z)6J(Jv)BQc0}^!OjbaB@GiIgs6ktp2fGd=`j9HR7l9~nhog5VW2J{FE5qJ|)NT0sK z>h|g7?B@-k@Rw;><~DL|jc{f%=p&%hD4eJ3(j6&5^n9y)jo9DZV8|={u{y>-ZbA{} z1oCf@JWD!54>jMhd2`>fpTc%Rt1(r*BqFn9sw=mC>%E#y@|dwhZG?4kd#BYyQTa}x z2=YBHxd1*d@Sx-%+27EvvF15FE&rVrO_>*Q3o~1gWKG#R3D8u%bT}3EfaInRwZ)vJ zF)#bep-1S-D({GFNZRtu)Szb5A@dsPEZvqFVD@YFYRm5|Q$gjmNoIY{QrulIy&<2! z4v^%K7u5g}?heQmV1sFYf+lyuQ;#%_#AMeHVU#W89zd;iM6FmmJk)LHcB22F+m=8& z=Cf#+zz`A2B%O2YX-&Au58oaWs}C02kXFC-I#v8{A_ZT7_c5=PG*y)p zPAuzR$vt54GrUp^7iL}s)$h{*y;Hq2(hBhR1b$@+k0EOH=-0dsqDur)BB={_ha8l< zbdd7MchvK(NuWziZv;Me;bNeLYIEQ4L!oYN z3+RufZz&hea+94obBIXR&r{rEx5x`<`GQr^rI`W=q1{|a_A}L!>l1fUc563#D8mZ0 zn^2vI`}W(TgMiO9XPDJtHR1#9QJGOv!3u6Z@Qn3y;^?e#D$a2_N~A9JDjq`g2qcy5 zM@xC)o6_3a0R(i&ROcz`M=To+!?riJt0|{0a9P<}>qBIa8`_3SMp4t!m(<9;vw$yN)HwEKQogE8m8J-TpS-&Aw{f zss&JU(YuwQM1bW_^D{WbTl!}gRn$xkyyG)l6_!9U98|_=PPNJZS)Yxy@*vuOJP)?RPq{X|WZurCiO5+#S~12lN$3r$j{@c0*gSN^1INA3XZ zu{GQcGe?v7H0w339cn04W>xEZ!~W5a>{dT%^pk$k_36&z(IG%BH&x((ULCs{9x~wN z{YmnN?vV72j@;$Y$)d{!U5P2Yp;L&}Se){Tz9>oi+L2W}mv)V`D|P{Qoy80NR%!m; zlfg}ah-}xx=Q4}G;ok6-iMe4E3>0j};(byT4uyUB%)zjEu@#k$($~5yRj~GUbDcHM zfrbyV9*~`tE}kBmp4UrNe5qjr=b$7;71 zV(|CCn_L`6{r%YqFY-o|7krInAL1^2CFQk^9m>Sa24F~rU7ZK6n{6Mz9B3jOgDz@4 z-x)l%(=XVl43N$g1$~JWCIoQb$MdlbM;q&GKF2{C{{?C}aZchAP#J0f?9D|F^<}%-OUn=|((CDce7|;m%P+ePBs~6bXRJY^L$YxCUjnq)_ zxMR&;eVF9gJ>ONPZ2ueOLtShYs~gp%8MXV0`ZiR@KMl5eV*y6C?_EMei|@jG}Ra|dJ_`EoTN1foV+@;Eo$;p zSr9+6NAFZ|2Qwe{4|F?V6s?@6;4~A|{UcjV7d01^Ib9_fZr74?5MhG&Z$b_pDhE<} zB@-Wf!zX3mJElFen@w9PZWt~QmPDql;m!su&CQpKL3JTb?c$fe!m)nwU$OzDY25|g z`T&GeL?aE>n*E2~2Ubmf^|?7_y=GJFzL_HSwH=05;lHDS-($B>)=3Tu3QV=x3k>w` zh0z`iuGjP-uQEYXb6*b39Rjp|u%7I{I=Er*fiF4c9G{6Qh#szGHUTG2#;%2C_jFq+ ze&(}yXZ$Jt{Hb=!-MWCH!LFa$xXi{rh1ONl-@nH^-%vmHQ2{S@PRk!`#IF1o`(DK8QHGzu{;RZ{dZx3ea?sZMx1eLm% zcUxKba87Wxm*&qguQHax5qkmuK<(Jm7*KN?{SvZE_QNMpye#%=aVY2W)Gr*qKGhZ` zFziyIqm;1ax%I14tmCvlazDF77*gs$wpN7M6^u#&W_DvIaZlz7?J>-@ ztHLeGe1*%m>Qb&!eg?@mm*k9gaB{f~rjPw_^FH7Ccu$(WWp0m%<&g(Akcu0^e0{|e zf0(zBgTm%hp#fTV+n*|*lTmqDbwtNeJ5jz`%4vu}kH9yPkyh^>y_;!3bsU45QMNU1 z?y7l-e2dz}0+a&ApD7~a(7(|4xUb|rl|i_NlrPT0C5E~4E!||ImSt%R9M5z%O-!as zJ_??3b|zB!G9!kR*;pw4-8-DT$KI~Emg9??IF6NfCO4r~Xw3>!d3GjGc2@AyDK@`K zJmye>ggfz5Bsy(&rnqLp^u2)uWF3O0_M}Q)^HTY_9ygG?q@TM=F42^(-vFuRlX5Cf zNUwc4QrSgtv1bPqp_({naoIS z1%xe|+9o*f)kmRf_vp5Z&-c7^tFsCvwY|#TVXGK}^^wixUb7ZsrpTQ)sDm}$Bc;Ck zr&1lchD^;B;wqP)m2Y#7xq@RkRk`^4-1GJtuZcLGMle5g4%RT1y zle0ZO&>T0GQx&Z%NO-CnFtl79!(98JX-6;we%6WG39AR6RU8SowKGE>S#Kn&fLf|! zoDn}>_qmw3mc_5E$E@N$4&E)j)z&AhG`aLI1U0Ii38Ct(*xw1Vv5Xf<$INR9f3RaA za?kQBodBCaErUt<;wwt6rBEh-$nVINkrO#pk4zljDb;SJC%LZnV^su;#briRU}s2D zf5ViQJ$mADWkMIES)NQQWc@L5K}J;N^h=%%SK~Av+i$$}ThN^Bir;MPHaB1u!J?Sy zkKax$^?JsAImL!gtT3HIIhl(OD7e?4AJJcBM zH5Ow78!4d5a}zgAy@8-CHROdxR6i2*oQ?||Tb%5yc+kl|XbLy~x>$_h(6uGVncmYg z)~%HwgH-XIx9}UbU2;J?w)G3kFZU)Y74$+d%f32Gl80iVZt_9*6aF*xRKAce5d1Uv zJ6eoxM5}@|3ygd4TVWre?O~^&+n{TGpGs{yuMU`67j@YcFf|U9U8OdN29)qhjpN<{tsHQ*JHX!r)(C6LFHECH;H*i)gbVa^cD0ADK;}IE2E!H^&?s zF5Mwk@(86!MY|?7{F`<$=e6&B&Y1+BZZp2F%QFb~+Nc&x1`aC)*VMs{* zn%YTCL95XvT*LV>0lyN{oFtj!b=CTcY)1gX%6tw=rK9VWf*k&h@Vc#egX4qJQxRbD z@@DbC*TPPRF0+s;t(d$*`c0RpBsqU9ir(NpfQx#SR87vhx=pM zT|DRO(HM&&cKY~}+|Lmg=z_(8i$UI&I@34#^Q5-`dqGHl^wcWo#(4WQ?C$t*Gl;?^ zNn);=7b|H|=YxkoGFALhgDP0n#OAoTd`t)~u>*dtylP#ReSWOCw9k~Lxg5?Xi!yH( zx@<%}=xH~~*hkDOb3E<>I$@W|Zmb26S^P-h`Z#3o4&_7A7Ig1#u^ExE_oNJsZU*I-5!hdb}9wbo?=TS+@6?i2q>6yCtK zJ^s!2$w=x@SplR}t)kcEBxRk_o*J-Z_HcX1ggxT9v4dI3IY7Z9*=yVqoDDR(dy|qz7dm=YsEwo?bjcTJ_=jPVi06b&alf%aG5>y$taakpW5J|wd(>kL5 zqGSke_SGh(hIZuaV(s$AN-{g5sC~n}!^@Rid%qr3H)i5>ar$YsU_}cu-hC-QTR2v=%CsD?SZx zkK3)LcsxXxTa>7`q}Nm-i#ON6b)e&JFGROlg7P!34PK zZ$o!e;&k?S?_+)=^S=HGFTJW9c150(F#`MfrU}Q2XfxiVj)8&@=Ka$=e=lE<_!Y05 zUwIR7_iAtM%mH!N63#C6A!|xP2`Mxps5=C8%V%F#?jp!4!Cu$~euG11bFJChPdQU# znc}u-Rdz|Ve*^!~IE`=v1jm_W=D(C9igUbwKfg z_vWT&6MQ>|pfT(sr>1384EK?*Y7AV@^1ncIlXws9XX-r$S>Kf_(>2$@;>EtFioA3Ftl>)ceI_#0(J3`oxWTz=^ zn0fkMsl!m#5Y9I@uCU7iL*}eA7Ma(~q-lZw`DV&w7gi+EvI>=zTPApwXTN3@vvI2bzl*07{ zMOYlm|BxTBBy*K74alh>?RohcZYfUuegcv3Kk;QO$NJ2SZ_57AC<&MU+<{lfLIn0R zS9x9j>YW%0>I>|6Fn)UiN->)_OLv3pYgjN@_|4jzsTvoUC%d)|&TzbqCWCneB)wb` z#l9ujjoulrkKmOgtf;YIN&q_3H0!5cRWx$6k_U`r4&oCvXIuySS%h_Mem~%~aLPkj zdb1xQxC}15Xwt@atBc-IbvghAqp3<4pmdrp8ND&6X9#(4u={6Y9#!z{n~+H(L?ME`y3T<~wc;AX>| z^O+zddnM|Rij$hRWr5RHWaGJ{W`SNFG)%^SRkOcTum zEtSIqMU+eAh9AHG;r)4iUdQwKctXBFRusT<$X81n5C8X0&2$z->lQ>eEyanOO_mh&c!+y*FzUO^Tj>u;%DV~=+{7ca z3|A>W&M2r+e^iQ1xhmHIxc}WLer&HWODThbRmw%C_r&k-6DMmjhC-ziMt`)mfQhy(E>UczULjDyTNLiS7QE2;sp4Omj7hhW!OS^U47NGZ>~>yH_xAvCTnR*o^Tnp#9_uT~%KJilv=FUqcJjJyPLi++cy1vj!CQnxa{)vacSaDFpB>R97Q zvqi;!vLXGB`CUzuSxqlhq3bueVGH>;;Z$+aG~jr8+%vnEIfG_apI|L1>)zj8HTnP80pqyo4;BF}U0^2(2z0OOha?j1ylbOsHGgm-AF*PC8N8tKHvglrrKm2`vS5vhnRZQd8aeH(AJAe%GHoE>AD`Fc zZLrfc33G1c(7LK}P7?m+lP%Hl#JS{tU2Qqjg2ulIwa!N@%yTZV^?N?^W$k7>o%xX_ z#LlO!7UVE#Di!6TH5%<)bb8LaA1pmMg|^PFWNBhMoa2t!@9Ap+>NV!>YOE%T*xb!wDWG*pH#e} zo$#8>Zf5^YI$Y8>?>P6mXdVMFPzo+l^Drov8v@-&=6 zLR*ej-Hd9;UW**n%hXHNw<^$*6{yMMZ)@q0Ki~<~>GgCa@qx~v7rr2Juf>RNM9zja zr>TCEARB{pZzMDc`c=>rO^lYT4zgdg!bYo2`zQZytDQJq#-SdHAt3X%UpP^ZvaHs=(ynC1ItA*E@L0kEobx;v6JuiTq5W~7r7X##R^Ui@3z z-1fMG+=1j77dC^whJU@F5c?LRK?agd_EDF2JtQa%Y+{5_*a9qE zdl-tf9T_@68Yb?P_9m0*#nKFVWgUQmqA#TCp^8|`FFH5qYU?3eVp*tVcnHdr9wqds zGnHd@Rm3#HP>$PW{fEG+`KvDkcF%mo?|JPH*n6`7>=0oVTGgJO4GtAetbobm#G6M0 z#ZD@3^I;Iz^!;E=eYE{ObU_(>`X}oy;uU?gdnYG~-{341Ed{?vixXxKTx5bi*hyr< zRpwRx3))_$C*?;*AT_G)@bI<*6cNZdKD zE9uzqD~1zy6UaIIwje?f!Ey|s;Fs1Jg+F5rv_C*+l&?);^(;5CRj5yendc|@hlU7| zXe>-$`l|Smx2!0tz&-CIXCZkfX^VNLbV`|Vn8~(2fq+EktJ8V3e}O6p8E-mWdj4ky zFSZ3-jJyEP^=?6$7WAA~AI5p9GGrj&BFGNp5z=nNA@Vv{Or4Ycu64EjWkk1NDAR!9;m-<)nmtto?>tZ|S4XRyG&5wDvK}mez$2pN%(cR$upDLOtLmxx?8? z2^9gz?-h0&w~7+~ng50DJ)E(EJ70L|{Fjy4A>1P6^Dy&T%9l>}29G{OOdRH+i0=b( zg|4EUwrdh*IX&Q>&`*Vz^k>MEg9X`_LBR%<+(oh8V(w}{e}2^%@&eH@9BJpn^0v)s|D*&)KD7*H%5B4NUq6K z`vldS0GTtV%i01k7@e*uzinzQ_zaiHO8CXr!+Jm-s7^^z4Kj@<_|52UUf?>yaW>zQ zhA%H-Vsb=0Cl!9q;9d8kWneC+YF=KisRXI}+2~4nVp$1Wiquf$UgNc0D_c#?xx5e^ zc#g)oI?JrzBZ?e~V}?)wp<6&X9Kl3Tra6yo5~b=wg*S z%axjwaG~fSri8JTyWM?NM}b4nX89TFVgcb-8qW5)I|79Q1%{3C=J8FQ56@rLhl|4F zH`wl>HVyqXwVd5UWdp*!GAX70uF7LYs{ZPH(X#k0wn}YpZ-qE4pfZU#Rm^NLrTJv( zie)Fs?T6Q;?y-S6ncB{-XIP%?J@!}g9_}TA;h=$3(2O~gHhU{N^bhL|(I;Kw1E=3Kdx%{@N&88cp_Lkz@%ilLAfd|Nf4KQ_<>b_v&>*ajFcTrxj_n;k z*5&t{xzG8~(4M%m@Im4NxuXL&v)cLX3#v*fXw;*=FV`R`DsEY;F}rj(iVwZ8YfZ21VE2ao?g^&a_w zcAn(-Hwj|!%Y-iX>i0IP#w%TKxS(Lw2+k@vHS^-iajc$gYL5i&fL2cHVl{yG+m=)! zksmUAF%_1Zj-J{2o!_N|id^YDM+zmcxC3J^1Z)6a6E?#2Bf?UbC$ue;`sAm^CC0&6 zO<#m_8hTpkrkvo%kK1+SBgEl2swgt~74(4hgs9khF8ropJD2J==j`*5Tf%IP4g!G9 z?*xaL3kL<&eweJ_EB4NzOhJxbvUvO`L<>_NPvQ1tw45{)r<_!wbbC9oj=kw`*-L!( zIg~J+{NK6H$)k!zl>L-zxFG?hF>FjW;E_7LAr);)+h)G)3Zjk>-=j|4f5Q{-hWf$zn z0mdJ>hef&Y;B|%#aevee*uv|ckg&2BjUmQJ5i9R{MOS=t@?Z(7pTW{9q-qX~vor-0(UpbJvjsDJ*T`I1|wX3K-SiHAei3>DD%k<{F%%w%B_$d$KP$gaECh) zG9n7XhCzR%o(4(YCOK>Kdu}bZJK6Yo!G-3GSys`qqEEO!@MCrZrOXy!_Xa%HpP}A@ z{g7M&!xJ#F`7--tSQVO3tvsm=q#mJuPYQ`gp;~-efKyY&nV|(s6R)`&a;A9hN11o+ zG? z90y2pr0zmxKpG|Af4=IJ3Vr5)$& zl4zDs56p+y@-`zWc^q)nu-d)T81rpLse?C>#SN)q_?^^j0ADy;{GHskJ{;Ykg zMANKC?3seq1yS}M!7}C@zZ_Zxv&VTY`gr{pOt!!usj&4d_hoa0i0uDO{54UK*<671 z5!@HKQVz=GU0^LW!x(J861vgV!y2UODZX3~MLQV3EtSq)g((8zXgrL`Io6&tB=a+} zhT=l4>3v9sdJExJSM{~2_snIYEyBx?#^G;#%?m{PP^(=nBoL)Yv$1|OEfRe!es5em zJa`Fn9$*>ecl79x3Lbuj`v`4X-UW)Leqpny&4Y`G-+7U%GK~*6kfyRsM$ zPem}F8`(xw)lQOoM>{+RsjA>@;S+(%QDRY>Hr>0fG*q)u*Ea|vHQLTpPI~1Fd}nt} zc2}2akh6Z{7MK8c$>n2g?_Sk_%Ke0_Rf$H1-b5KATK%hxBZf$arbXQ)CH~suy1rv1=Ic-_YijY#GIZbWV8=QV8i&cg26V{tZDsx(3lhe%Z<``K z8g_xzRfUsxi+&{i>E%h8@^hfbQaQ?-q=326wEmN%;{NKiglv`81AL=Yufb_r3hsc?Ifng(N$agC zihBPa%nrhYC6bdsH9~O>llLXTw&8C^s%+lO9xg<`4Ny~PYM7e+MXPm7KSC6h2FMHx zc$HcXCV#P<~)BiEcQI;X`ENa(CYXx z5d6J4kzx$DN)VV9q_{%b;czY#s;+(m18dEx42YL+)fz?xbj07L_#vavm?ce2KQ-%I z#%I1{;~7u<8|v`TKMcS3Et+Yhz?L#D{w#D6P~hrIu-0i+s`xO~G~GKTl!DYw6KhcI zR%n|DG+s1;;M_42{Pf@s<)Fr}x#}V3u)6t4H7+2{(t_g97ni{d8AbK2_wqOHec4hO z=M^+xd6HN&gl> zNGA+8ZE@sPm}_}Jl(*$w;E0p`lI-32vG=7JUvR}izi=}P9|qmP%-0{V5Wf=J15*-v zj&YpxJc3>g3yI5;R4vQLnhhD6Hssjz7*P6Q=}q2boMRu$s3qbw?)V>)ADE98L<)}j ztrlEZ4vWLfCM}KdIn6JqNKjT`=j{*0-Qat&Nb9FArNG`xc^LT|Js-47hAIlnPW%@4 zgK%fc196#YQ%{Z&PCiF~$TP+trgqd-Zzgv1owGy6II30y*qBdRN_Dn(r{|>g6!*;E zHxxyRB>$?*_jUdUbwGW*4rIOpzvCzVEF(55CHKw69mQoc)#A(7U5ptzG%;5@4})Y) z4gRJJB2*Gq*|!SD2HrZWK6gC%@U6wYVZ_Udqq{ue{GR5_`61jKIz?3$%_FDJgnE64 zjRrcQAG2@r0}ZLp0a*F7*?&}*O7u7F!Q(}#Gq_u z`pE)yXGSU*hoRDdMnL5|8{~96`dYVi$xFVf*=P$foKP$w7gpqW@0u)@8DEY}o-gj6 zm-aVxCuai+QkT}=oyA8Mpit&F#KV@^6P8NbUNe2j-;$*h!LJp((s$D0$9JXz1tvHA zNFC=<^{902>)zzemME{Q)TXxV=JG;Q>m=yib~(&JThuV1;F>8XKM1N7kOS$>pN_;w z;3?4TuFA`Z(!8bUS*m#eY#@S-9FN)UlXc62|04Ef5P?I&6oc4j_E+^=y(XB!H2!W3 z_RK@U?QNis-mj^|T2pQX8~6eCF|vONd9lfJ^mgx(KKoMYIMV1+~W=vMt_0i&3+N4M8pZ)%htt!fa{U%9jjRG=QT+VI(af%?FIC7aK=#pS2H0dA-d zrU=O6uu+Uqis9UmJI_wvZ^#|g$5f@m+?6%t`*KP`83NzXgVz`h%}GbB zlJSfv_MxGLqJ7i@bHbg;&cWATrx8b!4nrHCK%A*5oB?oh4*6-`Mx{LNt->#TIBo25 zxl__HS;ycHwm1#|(OqzMW&tOQxP;+8G%Vlsty1xexrpS7bp|ZO#unjyg!s*}<6d#1 zU2!LZJra$9^|w{!hEQt&Xw>S*_dFw;+?TYM3@0-}*oQn(LqF9?Upv<>ph7yHfxJMz zYCq4L)CkcBrygtx#Qy#8FK8fC)Zwtx#vL5e9o0Z_&!)ltD}bxGn}@T6xq`t;L$2$x zxOw(ABD>=nDOjkDP$7*!$C)^Re%3KRA@#l>E$`rr3|K$(1fI(}iO$8TL&Q;YE9=U& zmro&kJk(bv&Q7^hFxA@z*@lJOy`Z7-{sQR&!y_BmnjzF$+8fPZGJCgpem{J|>yEE| zHQ*AtF(|uDoY0UjcBVtl#b-0RDRQ0$WL<9X&^Q@h+d_5oC}EU&`nA+HY)f*|6|1;k zy4rThnw4Gm&~V6WIafw68lk_oSN8~E?VOf09}%zafP>`v*91I;6Fc;;r4Z;Ki z#ehcphoWWVKsl+SlAJ3zmT`%n?FWo>(b@;;-33;R%83^Y!DD@CRk zNx!4~=?zEc(ARQbW{5+Opde8QD~$dhjvPvf@!G4tR~0oVC*5!v(`n$i%>zN?7R~Rb zlf>tK3)oV$y1W;|^H4lDja>`4QVk-4i~bmn4qi{5kdMgs5Ozoho_K7gmZKDkyAEOL z_k%8pB^)Q&RkvMvUn-x`5L`fpPu{q*6ddSxB!0tOo@=D+450qzIIg@o$=q@BqdqPY z&<(lI>~-#SQeTbeJLcDJ8^(hO zljDowcM%`G@9GO(_EW!j`kI}?Q;Q5rie(Z_3l<_R_*aFopui0&y%VC~kAdPW?kf0Y zzl535U~9RNi#}Ly#i_xD0L^OcJR5>=u=l`YOo4Xm@JX!N!(hCn$n}%5P@hX3jUbVk zm`VAn_*U7{EhQMEZAx21gpMa&5-q?@>~Wy% z+S}{^KvIp^WMRQR_OT_UmOT_Zrl;^7TkybsGS8=eljY`_GQX);J}Lf&>!C(2j?iY= z4Fi;^HwnH~w-j(ML-q?7+P~BtnYuyyO|E+*$nD-oeWk-%UUQFta){4EsF`+NT95uq z^&hjw>|uq?LLmVx-nJv8TL@oOkPtkx${vMSivTh1%AN?&BEKS3-4Dv!0yHQUO%CJS za(;sCGj%`QAr0-sF#z)>ZGlewdTIVwx3}$+Kz()J$g6JSuM8=Ip`}>UwpI9pRtz; zJnfw(!MN-ZQ(E0al_Ne{A7t2jTikK3mkDgGM~@N5Q74zU?YD2`@`V@buMeM{nJPAT z5$*>*u`FWLGVyD{O~DFthGCiER07b&(muEX_orbw)Jf|hneAJG6=+w=G)~g#p*Tn| zm>B}uOK}l4YlYNicC(Xeq0Uc&S&Qj7uXB|Q6L!ad3|B2B1UNmr<-Po=A=VW8IH zuq2&w9GSG9lw@r;t*c0sdq={HwcTGZ8}fwk5%ZW;v6-42Y3E(EYA5Dk_Z_U1*R&c^ zStfjiu!Ksvr`5h9F&L|$efkIBd>Qe9VL_SgFU@+xa<6Q_MV~x+q&d*+R(2h}Na+h` z({g;^joKF@k42>F^!GHQ8#En9y8#G9a`><-ka4o+7ihGus-vsn{2JN~m)Vn@xvko= zvTV+Qz@3UPm^)yN&l^Zb(el)}pXmq5i;^}Qpp+GGBiZ4~P9aDXi|SkxwO}Jm7opA~ z{|LDV>bOiHnp`?+Wle&v{Kt?fV_I46)F<8GVM;5g)#Gz$g$-<^lGjMv!UjSSD$v9u zOmKT|K;abZqkf{|Y7nzc$?nqe1p7h7jzB=%X4&Y0jwPNk!6l{#(k0{}>6<}4p9mlR zKv%(LCH*v01ne{ZlM*uJ{L%6BZIK&R>4Gntn0aJ0&+* zMiO1;lv73U_tpXux|qU0qADX+(0YcU>eh68~Fn(`cWbqEpcP1Q9+0nMLu2TEt(1lw^V2^w&Zqfqt4?u*G7ZxBV_4NEa`|04ZoCLS$(?q@H3Ac zrc~AY7Qdt;GJa9#=ay{EYT1W6Z^yM>7WdYV!S(kLj1zixJBt2U^$FAOkys$(Qo%0M zMumrU@wXhaE~whxw^<#dqFn1>a&}Po)~2i=bcgYvV<>D4)^5($7nr#Z9r3k_C#J)y@Lscfj9oZcJd!)U273!QUP{Qe_*y;Y28!HG*NZ9FFbTtMB!@y- ztFT0aV(gbqj}6h}2oV;R4H7D&hyF+{@R4@n<#FHDxK_IomvVXN0a?l^)icAHK+aq( zXZ2iI@GD%6GgF$>(ren+BNShvkMKAp-ZweXRgoj7thW^4p}nc$^?y$YwIlL+%VFL3 zvzZf+Px|*ANf}l%6lrYvaJ_W0BqwOBSKcgz*R4pdqPXE(UTx0>oM)D2_9o!M-`Nas zx}67EZcVTGbDZb^6chSGQjf>~jw1XcAg(IfJrquUp%o|!iz4yMtB#n{oPo;}4Cr3sO^(Cz^dF=s#DB_qo8MYFIqRTKNaWv=I_<#trFgT}DZ6-TCfZtV ziv?XK`004v<+P4l)=%MGhofb;1)XFh9~oM1lSY%=&Cd2iUXk`9atok!`!^a*kL4PH zP=B+h=m-BV_tQI2HuUf-*Gq`K`bCUWAmZ?(`5({?v={-7;frW_x#_)5Q> zN0H5d7TGV@LnIl=-%oW>7cobEIJFYI(PtZzU^q-HHQb6n0^f{w&3k~m_$%SK;Xe&Y z{SJCtl){uBN*PyIR65fZpf$&tJ>QHm4DXIre0g_7S~loyrRD^cUyQR&td(i$r5cKu zY%8O#M(mE;1}YG>h8W1&1XV&2t5I9dp7>il*~8Yh5%`AH1hV384QN>2$-0dsaAt=l z?lu65aHIv+$5OA6I~>Q_1-Jz4XO!G;RQ?L?p|61r4f>mLQWTK91H7nD~Vhd6)u zk(L)eQa+|sLGiyCiezD-PPUI6@lM6e80?*^>4hqAc3`#N71>MNr=SWT-TrU8xj&lF zJ`vgp|JOI^6!9Xyehd--0ch>Bn~XcBN?p5P=s-{ot|XhG#8G%i={oaJDbyY=PnEtc zB{Bd~HoS#oAnhtzsmgL2sDpmKs@(@_L{Hucx&S( z)}h=$qp%j5DZ0&V4Q9-&yWqg+RL<>|aU^l_#k^2WJ^X6r0_ttOcyIGwO#mj_J}UZ( z?(oqRS954?a8Z-2Rn(g~`hKpOaHM_}8~~~&mXrU%Wr+aXhLC63yP3lDd!qfqMP;u< zsqIU$5MRzTg`Rew#jp1JwIDoh6YU=RonRo8VduwCtJXnFaXa2oQ7-V)&gdQb-H4+_ zKiX?AQ0Rn1o=cdY^i$_qPEV;%a1EnQnIZ%-roLDgoj%)4%N3~H#HVMks`GlWC8bx` zJLRAFt+HT%6-vi_2VSi#>F)L(V{530X52*^oLNVQ{gJFJoPwj*mr;*#x5$=f~R!wSOrtK2vb69m5;!3PKme7=TN~ucaRg6bN z_R-sH|8|(fy}EvP*#bL)`j|BRd)<^I3YT%&BA5@O@G!^e5X<#8KKW9N%$B zW93&wAz@snKGW3shsr@MYKqZEGw^x4;ZQLz!;U`y-BARP1Px?5Goh6%XZ zzedXm`AHm$eaoz6)MbusdPxkN-N}pY2ptYhd#*KHo0(6&QScPl#xMi%n+-YS1oGH` z>XdOIhaIp#v>3L4c7#?FU@0^@$=6NCX1h#^fY!x^IJu7jV0l%gk5Rzs^K0JCGW)%a z%5Dkz4GW*4DV>5>fC_ZxPw459egM?f=wkoLX5AqR19{gj2jHouv37}s-z%i;1J2!p zH!w5JuJNw8V~kO4uWx?_9&P$b`ldsBPt-I$>E8hUoPQtExxE*A4Zd?0F#2%Jp*fm7 zR~n{?l;;SaQsjGPcWtJ*Z?6g- zI{j$l!LKHK{3ba$PIB<$6fR^9BYO};QLzS;rj1{*dzffu3Hd&0IqV$p8rgtR%FOB+ zsZ0V%Q3HBow86c1JuFel4=LV(SLojB&A)>=gbRu%MrV%zcg@2AJGN6^wjUm6E#E&JKTn zfVn?dsWmKUUJ>*7!4Xk-RKpx38xQtN78i9D{v_*RMSv?2MbeKF>E@+E9iSz zcB85d_p5$SaAFXt)>KVGOkE9N^+hwTvmR5P*kXdW69O5*ad|%HQ3C1(D{STqkA9f;z|N2Ce*rpxJVq@dfQAdqPr|5yY(i9-^6x%%!CeUFzQEq9iV)*XD~1FK z#!5vRV$eqa8z;8;lrai<_#nA4yn6TCeJXh@?Ej zMtNpBgW-2Dr3Fp2u`?lX+>_zL^k8!?`VIUzitIEzW&15{qC|GO>|o0J-Vn(e)sM-l z>7woGT$s0PiTy3~Dv4WDJaxxDvHfkH>ia4m`ex&DxZ^6kiD_TP>^J6EHP8vO7HwHX zcc{_IE?HLiaS)Wz)@Cq?F6Dz-^4Yb z%plrA1QY(j-8KBVkGSqS;!|lQx)-w#eM)y*wLx7W$<(Zs?VSbdh5wm1dNu;mE6&dyy&Owq@`!)|lIL+adZ(+(3(; zf~5HBAqP6J7nBq6C#-(*x=h;h#OH#hBADS{i#rS~T2ZNq9QXgy=vL{Np`v(&sUt>{&ocsBtRNf(%;BIqDSUd-6XfzzFQc|d}IB=^(r!7l*4X;jXWIJ%RdtmERHxB zP*muB_&X+H;Vee)44XE69{3IHPOGn(Ah=o{aTHKM%^LG7@3#VQv+D^=6_%+TjrfSO zpcp?U(#R*>zhXLE!S?sqv&>pkNfjOD@M^FCvS`p{XQ^8h`WE-Sa`u%vB%~mZvXQfV zFs5z^nB~0DhrA<+IZh+oRk`7hAqrr z#5diZ*e4@(ckOb|rm`8i>36FxS&L&!I$oj|DNjw{I+ea@CEA6!@Z3t3AOK-`Kp=@M ztE@i7@PC*SmPDP@GCYVUX5jzU>K(@xa*SW#FXQ`j5)5M>h&fzP1D~J%%KQJ>=X0cH zVeiQ2ofg*r=Rsbez9ro8@`das-I+=($`6R;C$T3~+c*yd&)Y7i?J!+&7ZnP|n1b|!cgb;Y_)OJSTlQ+xYFwq?2QE`nlh&9d)1vFtfeu(VzpHwtFW5ZaM`8{_ z6rn#7d8K@a|GKJ%jt0==8wW_>X21-uts5XF`7v(A-=!9YzDcTTe& zzJ=4R&$G8M{{7LchMqAhvRlDN6p&pwa*Yy zwn?%PXkDeL<5fD?h(Te9GHX)+TnbFzJXRkIHdIaRC-vH<6i(1{0zjgQ+FD~jUN6vc zlq{dARg4$El|8<=u5=~P?blx<>}7R!8Zl)PIi^UV_{`%L}z%ZTrdQ%S($vmsD;d&vdD6G8BNrb%N?JU%k}Vd#} z_*)Ik6Maf`+FB;NY1uZn0eF)XOp5lWr5~67^GS~db}FB66Gc1O#u%womI+#Hrbhbn$;zHMmym5EVapPIyb3t`(4KJ`1O{n5#UFRi`+EVT#OuD2c3IpJ!1I~ zuuJ|553l;pyiSP@YB`|R-*O6$^v@@MC2mw^*Av%5eh9iZ^owXWcR(<~1JwHyKxVdj z9UeCYCO($mnSyK2#?Ezh?p@{nZ0aik$bxiPuh{~r7awJh-_nI3oD*GX(7o<=%HA7- zEg`nfdn^IN4YJC8J3z8H1aNflU+V!zCG@E_5_Q#owrh@k-faeJ2N}Jk3kf)aMzi$> zQktCW+?3tlJ3FlUxa&WF*kh!0+0YyZ;eL=a&(j65$GS54knJXpPv}axWWMRYkXl1* zn^{MA2?g6C<`$YGfL=u@s^9+N>rBpq0~A08GyGM}G1a2E4srE2>VO+zh2>?^=DJ|R zYh^4eX{#fV>ART_YMQvm5^pOA>>>~quL+>ptrIs}WnZK0Uqf5g^*`@;RXf{0c1Cjd zQ+KBO2(j_x&v6Syt##{?R~o*zTp>H#9g!$OXVE&nHRVE~yM{G+zUU_ZhwzP{i29e^ zx@=RGrS6TO2dH*8|6TR{@XDea_5b4nP4_{u_MquMMc)^=k9aZ8QSykoCX=#xGoyw2 zH}(7UiKO?N5x)=S!!seyX{Y^eOnn4o#ap;z9>k4%~fU7lUwc?i?Q*;x|d;B z-Oq_uf}=z<<8 z*Wu04llc!TE$0af=+9;VnA#BT@2KzW>)?muo|~81j%&ZCx&RY7Hw~}r;B#vk=Vf0W zDmKmCG+Muh^+grz8;poM0yVB;!0X=YpL;aAu0crB{HX)>4lgx70)7DYKT%Mm?MvsA z)tECvtq1c}@Mm!H&vTQ}TI>gxo3+rb$(-Sl{P8Db5=MEzzIPB|$)VGzOKHy%z#S*K zYbrC{Upa7sC#{2IP!jN)0GSR8US>XFC42}74Sl9;O9WJCUAzIDFk zW?^#XIst{VhxSo&i9Xc3!+$8ff$v!&&`ktFnTnb2AFd%T5}<7}_dsVIr!W>XuXFxH z2G}T&#wv7k36v{_yXvoS^=0v6%Wl$gbE*HYwkz7Z=nyVu;%X!PXL_(>i4tqRu0AOHl#yYJrrf>5zyt^QH@>X!jR5qG3RB^PE>8hR#$$)e!R?|Dn!a zhb(tyd<%6YY$|x-z>V(i0AOI5G+*(Zxt{P#FiDLuA2CZhxxQ?kV8X3AxSvNrrBH8X z;4rut(bI%zjSdoRt8Q0xZp zG-hVExsE{EM&7~Snk7Ce{cY?7#Q;&PGj-hMxW>Z_jx*Se3Ng~&E=`}$`3QTA&E|K( zcF4WNp@_hCZ<=ogzh=8#9VfA?7W*qd-t~jOL$o_r71aGj;E=G%x1VHU?Wf(p^j0Eb z(eK^u7ghg&Jt-4TH|(Zh`kE`)_RV$lQGOPc`Q>rJ*Qq z;BxqJXde{gJb@(sA4TWlm-PLA@yeBzC6$$x7uMWzWyxAAydczE`7N8QRkQNKx>{PI z^8&~%v!XI(WsORO%F4X;;SWPDtF9g3X*6{Ylg|Z9NC)>>ugUU_xv%YLUcp>OAwW*QT+tdnLMN8oxLy1qM z#9N-!{g-wv?uV%UU{ks|(hpG>CmV^MMmdR*P4+|DAr@5AnMh9SGlASg%*SJszAir* zy(|{lX?TF)^p~c22>iHP9sh_vE8_LvvO*fNLWrXO4I!C>rPu8r>J0Ccl}?GeuWVKO zd!ShTkdi}8B(`}y1?-%z6nkI-VO;Y%r|q8kb^ZqH;UG`TYj(Al=u`=GKK2^y7CnJ> ztqdNrC>(x4`~zT=^(soI+VRFMHO1y1xP6RT68OlL?Gvut0Xa5u9oFi84xm!DnU;f& zO)GhGv8h|N&im^OD|j$k9A|uCtTFzEE?1mJdsw)T?ScAGi%|2-A@lauUlumNn6k;? znG3W-O61fNeM5_EL$kwTYg1%ui=h>St2<;!%!pW}S-#Kvq3zVDCU+ttD(i3KZ{oGq z?@aT>4T53lFN#g|U!BsH_TbDK@I#|-L(3tQtwg6xg1CQH$}lRTeMN&uyi4g%up)m| zGUkzB*b!m-F{80{2`Ae4uq``q=Bh_9YRuQ#f5-nMtE=G?c|*ojk+OfjkBdISw#EIH zBZ8PXp0=L9pSIVE9FMrw(99o4GAr#NvUEde$p0@)-k$`q$b-eqi4%YAh<2p$kZl>1# zB$&8!6SW60fkU@vn#jg@^ngx2n0`WKpF(qSM^jI$7h>XMD=H&f4-ZPiDYUT7zP*-8 z&w=~gRLt_Z-oXh)`4g9x2<)P!!E9#e~RK!%bKE zYjxixFJ-ScZ=G8<gjtZ!bt5Prisz*x2ZH^G*KHm7 z>G`{aMwIIPaDc5C*n;4AK9W76ZVr4syO>&T^QE0DN=RuJmz1ji?0qYDq`NM1Cj+EK zwmzK$MsDv_E`nt^UpDp8jCqMb_VxH~QO7k48I9nTM7TyjtG)<4jToeG;d*gpVT@x( zMMFsjoOWtzcQ8AVWOVgGp(jT5fFLU5i2n27EFG$e4*@Brmn4kSB-_Tat31bUo5PFx z823rt_SmP$#Ntc@u47`dum1w<1t}W#TTuovK}!@87L_*{%KHyr3e|eoVh@LN{^UW569e>v1eNOMJei| zCQS`Vuy7P%8T?!`1=Wu|pK$sEd;9twoOqr?Z-cCYSJtjTYQ(5HSz& zUDS_25-X7Dj#!GKb}H6Fs1&GZ;1_=BY0A5!ejo$4LjNQ*K1f-neBx|*L>p<>G^ zN9#fbyDa}j36CK5!hnzY*CP%&n(qv*G)AUhn))#2qgJ*)6E>bfCT0n#8W;i6a^f zQ*NV;NmmXxvIjof<%3d>mNGla}36>#&V8w zHj930coA}MQaC`KK!An&ipMo)RN2K{s^2p6OilUkSfaF&d|*6D%QxdPZ4Ot&72tHR z_`fjZcy&}AIpntz)_jZ5wA8;;Ie<eAmW?$9bOp`a|(=6r+G{BkKFpsz#6icjeH{MtLhz_z|iKlDD9b%(v7W)pW z9dVxqas>EJobH1)0@PF%$#E)n+$|T|w6lm*67_BT`wX!9msadl!mwk85+2PrDnymG-QM;06w#xwPSkeW&G(`bo9F(zcB7t(A_EdY^InVWiockd+p-R_+L-pI5oo zvvH`5jyK^M2}O}sPX;*X3i>C(+L2YsKg%v?a$i6K?eDbcsVNO*N+9unh1yJQn{A`5 z_HW`52(~qd*pXABz%T$82~hjB9zTy~(3V@f-;V@yl>Xdvw>k{Hl=_bJv ziZV=B%Xof>?BVDQ9>uD)tsc|&nmkn%6VDsFuTujYQE^H2gX!l~p^)g}{g!`>bB!H} z=<*o-xp1uhR^&bqdUbOzkh|P4v<`k1ipB#s6aOoRm zo_S|~r0I6byewDi=}a=Dr71^;cQP=G`a4#a-Vq*zpXmj@_Gy8=O;&I2J-rgS(^t_m zX{(f|zf8GzCY0n}fh9Vof>o&o<}sW*;+y=M3yJ&>+!(3+0wzokSN~W)P(!lqcZ@=p zd4_JHtriZ{y#q+39kr}T$9y3}SY2xJNte$!C=)x&_#0VW@l#NZc19i!dgUksKAYtR z>b7kv{?*nPIP|h&8Py%<0lii0=X+WCjp2|hDU1r-l@gqqTwkI|8rKG+x%)Am->TZG8=RG0PjK46816dPOSB1P4*qz9f{T2p(P zz0%)wT91YNVaTwHcMb|I4O%zqu5^C!;lpwHqv+lGAL_BQezP&`;>AMf4648V8EGkL zETL!mTJG4*vVzAzJt|qUvd}?!?N_Mo*m(CQgK5SnhkoEW9M87~Qvjy#0 zXA(rYuZQ0TL##uc;t-X4)-G{mUWsumEHh3{h8g~ushrKM@)#-Aj;BbKc~qnzaK_n2 zYz*>)Ew{&Ei;+eTaY+sz#QM?kPFbBGDwL`Q$I~=IEq|=osWJbC*^^#B&`lX5!I*X8 zn&oksbeO*JZ;Je`)>g)%2^4G%hBp%1{)Z-UHnX*d>gAH!Xf%7U_!V`EmJPtB~2`uAJ z)W*+#1^y=*#vWleK`!(*U5!3tmNdP1)VBvR^e=>G1GidL5^WnXxe zc(Jn%0Y(Uknk$voa17ggfn^{ zzLon`>FT~jSfsgSfD?40LU>3}3+?LZok-Kk9LECx3$cc%dWr(b2vqMxON02Wht*3h zP}VYU<=iiD)fppCl_`CZ*2y@YC}?iak0I+9T7K~PL85@Xu^FZMvP0~|uRdOk^OozT zontK}QC-gvyNr_V1ieSR;xi=?!fBZH>U6CP+(%f5{f`*VUs5q+(tM|K5y6aYnjrN9 zRl*BMmi8+lOci4M##Ym|keneJM)y-)$Y=OHz?HS{K|7NbYo+#=!dCaB5z1{C(bfGY zPQ%!H#eC@#fLD^wdvgM&{BnV}P@nIwGfVnQ>%q3!Dq9XTE7O1UW>f7XtYyz&jPXD22k}2S z2_66^00K7JG{SfX(c%rQK{r5tv-z_hRD^_0dYBH){lj@%eA;|X$OYEIw!t&UG?H9- zTxPIfxnSm*ZBrqShUkJ*_+(5ja<9qT0#2lyI;rZ9v=&YYhptm6I^(Ia@se%`a zxZ=JJQfdigCZw+6twBt}Z(7QFk{ktqHwY4Rjh!QRwh7!z zf3J`lJz8Lkc}EPVARyxUitn67&fH`Q42&i!ysi5(Q$hPlXnv7xN1x;;So=Xc7gmVt&Az`qFqa-p@ZM|?``(w*y(vW&6SvMEfb zxjnvlkT>2}jZk!{e$BLe4w$vx8aZV;!~P?`8_ok3GQLDLzpj) zu{X}^Ey1`YKA`v+6d3@B@D)_md)3h5-H7h)PpP*dODsR*mO2{kNeJIex=z9}{xzdD zZgWD{>ElH)DWgr=X5LjSHeQ>m+A-0xjy|b;ncJ$|MVrh*9xiTLN!h}S44cpBdLkb* zfn`sBQ5HqV^mYFe$TdZw7l_~X1MCO2M}7Q^uj=7u$>^(L0Gf zK%M+eb*Ol~2Ynt?GSZ9~Lnvk$M^G5m>am?F+I^iv3OGjdBfEpj<=ad`*}dlZ14*N( z@$w6$^&W?rAD@u9^rDmipn2O19JI}atSgKBH}Y(uR7eMg?747$Sm-{NePYiUW$tq8v;G48rENq)TPfqz$z zJ-2pzKgoQT6q*51$`qhu%(oVEn~JnaC2ad_+e7 z+qNE-(2H*to}njrmPNXNGaNnDz~8~Epkl^~=nnWP3V`^+=1n+FsLUDnxf$7Vl*HX0 zzuNnxIJ~)`2xzq&;wlmE(va=h+c~Cs21CZ=Wn@c zcc?0i^HrOrOIW$uleqtYLDRLl+VlQ#FQ35pmONb~=%$YwESuffEy_rFHt9Fg5j-0^ z0W6P__=7sL&aoc|Q#R@LzX4eNr5WEwZC53l6f<;<-+#pS z(xh5Uqx+Kro+H)0k~W|guS5TEm{6LwjNK0nQ|*z!DVXQ_rk+hf#Q29IjFwwYDP|JPDH4f zoo?l%m1cFP;8MLp#`_C`>L8ogUiF@OkcF91B>WX|2z#BoiptjBvYU?ZF7=MCMYN|i z=Uy*)fZ*~kK=ZP8S}Bvk4Yv^R2x0JtPoTbJYvX?U!?<>yx0XUpph=B(r-I(P23vY1yj3h@3ykgM&vW_#@T4B@4x=tzLy<4 zp|UlV?X>G`tri=I-yqrSJ|9%+Y|v>@e)y_X`wbjwP%YdPCR_>PB&`LP1Y8AdL1r{( z=SiuX^P>#r*r2C)*;4Pkz`Hp9Bp4P8xgGT#Y#-rU%lTe;G9G~Kgf39FP}YDZ;ab!t z{1G!k0o0L`%XLuWQ;n6=qSiH6tMH;n>D;0c>k9O@0w13!^GhI1-;p}&}ze*F!d z_k7reGEl~f*5=PLoZQ&GG}-k;3Xabrx&`)HKEL&5)NUsJBZ8*9<%Z!F)F0xys$B`* z5K{c0S20KyU2|s#J07$Cd z)}G~FM1d~F&3C5T?{jGK^lMUamZ3z+c?afcrs6zqTo>yY5pSX1poI1pV#L-o-;aE=PV6O>=1C5p{;Ck;)&7o?YX@m*_T5ZiW_cGeLs zEJl1{pHl(^>*B9j9fATpMDwOd>GZ#IhCuGMtU!+cD)4$MP`y@9O>-qKh~ie91GXo$ z2aqVfw*mO?i!f6SO>d_QZ@As8eTWn@WyIBNV8&4NT0ZK{wu(>-MErxgnAbB}Qyo^%WSmDPPyC#2AFVTr)?j77K~wF@e` zh`W&GxQ`i->4w-G)Iah}NT1{l7DyL3e{berBJfeCZogRoeqo2ZH-D7d_cFR`=r!*nr9n3=!7rw4^cDA4(Q0dOld%oFq z?9_D`uM11{e)40u%k1qbZni${_zxlQebl2i)zArp2lAre0ZhEEFoswHJsEAbiSKrm zvuE(c-i9d)S}jZde5gLs@QF2;HW6Ot5^Vj$Og)TWVwrnMzk!%8s_^leV3r-;2EQG4 z0z37nnKHG!xFT-2RM&+(wv#+~TCrGu%zoBz-(UC7Oh^pAd8Q_MWFhfy(S&VwDX|`3 z>;tm5k=r7#<=y~)CUXj|(Tai^>=woHI&Vf&E3%l+%^QjUK7{^MI&~@C2H82iaQQCM{Pr596JC8B>tQ>>~w86a^mkAwX0bSaD8B72y6C8RyAF1?1^wim|5v%G1)0jC5a+ibpqGTg?OMIVcIru(E3Kg3RMTf;? z@i$@8ydlg1ti=L20hImOgBTsMsV|E#UWG#XS%4-jP=^mm5?&fT0iKO)H|?{P2o&K= zHV%w?Ck=^qrl(#sJ(vT{$kA8l5T-TQiL!T5aGhN(59%xC50<>DIB6v6f9;@GjU z&!N{sZ=q(Bfx)BJp|8O>x@CD>qy9^&JVJN0b^}2E#Qs>h7bU-I)d|wVdBMK{D=dWu zaH&0kb_-a!qgOJ@&8N!XU&0?4@Tq_3R*wy85*=FMj6EGC2imImn5~4#k!I{keqVC^ zf-cRf4Pfau2PHqCA=^D5t`wj0o_oiSY1oswlzDA(i8s|QnylcfP8UXas;;XIG9ZLW zs`9&?tNYPC5jQpPR$}@v_un*J6$4^b+M%Qux_mN#KC}(gz%E0%z*vVJDwJx!Xg_yH z&@-!-6@>|Eyn{dGRVRPj>ls=7L(U}I*wSCuA+N#K8szq5WiBolsFLd_VjpYhu_=4UKdmpISVI?pR;+)Akoq3 zC9rAv^u>2d5lF5#O} z+p;*0DJXZZmQ4Gn%|8QuOACvvs_O z{m+Bwwn*x2bX!_CAd{b6p}CrTlcpPiSxC8ZOY41q7@KyL5 zz{LjpF(rt$)OWGr4DyofL#-*o7{jDM5Cq=G90e34Ier5aB2?u(lDe9ikTdd$|Fmr^ zem!i&z8i&r-oy2PcJf3`Xrv-CYn~B;_Hj{8Indh*`~1!WPoch+tVNO1LXZPmKiXSE zLIPt)!{rKnNgm%zXAlU|dJ`0;xxQo_HL^QW-C35-EzQURk{=tIkxig+N330^BN7Me z6OcMzRk+kER+GsmP5rft<_-q_HRdCORaPhQ$_3a+Tgnq5^T+%hgBivsbW_As;FlUP zietFyV7;OJl^9_|hOtbca(ZNq)N2f3?5>@{PSS58I-}d&;`)K&b6g#Z2fNB{q(i?# z*4O!mt|R>F`jR00Y8Y@b8Q6i085kE)mxt?}5YUqOdvL4eEcGF0IXd@48Zdq!?ag7` zpRIK69209L8iT#-37E(1vf=diCN0Z6agj)jkQj3GU@gz;ZQ7Iv!y?mo?XG_#a(S~S zU5iZtMtd`c%o{8;DfhX{i(c(eg&-$j}LI-9uG2gz>JI?zv;Un-gP~Tt~mCAl0 zmRCy#zgS)DU*dwkv&O{j=9K~luwi}JsxolDk~?^56b}99VI$s;-);HSr|3LvoBbB@ zA(ho<7(ALzGV}>$_^OiWSNKFB_C!3 z&q?_uM)T!5Eo>>Qetm`6m5pe4lmN@TkXqwA$b@Tf%v}1ZG>2k!2e(Cpz%Y>x?#ehBYV=6 zXNj!iMbZ?F!&#k^B&sMuh$NnI30O z-Xv4?i+a83V4gGpaV&Q1`P}8{N$JLWQIH$JD^WK61jc0mwKV86xt&{QX0vvZdAlFd1q`Pz`{c@UmWJ5Kx3p+n$u%FN6TaYk(+$VlnJ+tV#q z7((eIU^;4%uo!k6tA5j(2emZFmFznRg|{c541ww=)Nc^W(5DnzwJ`)Qo72NE>NLXG zSko8uGYU44+IUtU;#fm_X1e0)3}f|ei+);L7w3~iSm_HOl5L34>2I5k8}qE%01wAP z;y1pVed&1F))r^Zsd$OvvZoxV8yQ5USGOj#Vv&5%BrNo#$}wn}`=fVZ75t{Qd}~E} z3{kUxkgSor?}Sf(``NPz@C=#p5;LzBs0&CUGv3D4AaKMQ(?R}x!YOK1;dNj8j@$s) zuim=}S`Q}KDPq60?{e2H60NjuurDks;u`OB0j4a?->p4>>H=fK;gQowL$qrTBWN=; z5Ox9r0L8@a7UZKh3n<3WqGo6wxtHZo>kntXW&CDZiX9?}z971i>-Y|nl-`RdG*fM) zC`wWL6h_OVplUf-e zIjV`R)uOvxA4VXe$rXI9Yun^lMjn z9OAgs$WtDL+*8D+Zq7I|wccH@j!XQ>liJhE4Id7L_DR~jo=Huov`dn$ zS@K)Sw+XDxTrMGz<)iq3ymok$4-ZiOlcbDnk94+U47$d^k)rHgqW+|OIh$j1*IPrKVq=?E#kaiQ7)^|L^PrB` zu6^QqTbWCGJ-dv0)A*&NM*iuH=Rkz1M7+}obu5B03?D|V=o%Bj2ShiFBdAy1^o|C0 z4QrHSvCNt|n0ItjDh+s>)^or3Z7qhU9Qa%pgA^}IF-KHscXcFrIOh-*y!C7}$8Qj1 ztPjYQcg;`|q;*kRH{!!Ee<}yQHlZ}~cf)b^#W#p39hANc)azLZ7_0em`VZ7Ox^4G} zhhvd)3GX+^b0x3>CK|)Wc#6d(QI++z=!rR@Dq!OG;|ZexpzTlP+iJU$k6Ww=1)I0d zX4PhBv$Sr`55tUFc*0>Twh@p7Ue~DM2r*iE>h#QMU{l|K4n6^T8#(bDIc}(q_Hy8B zB3MmO$vK~4SRy;)6OuVBbbe~ldxik*mtl<;)H^usq9tC_G7V8q#_K3L+*7^dynKTF z5jm1WL^Qw-5h0b%gO!Ds6zYck;`(aD?LKm5pr~V~a{T54xN&hqwWppBh*7QcqA&>* z8;4kcT;M+j6w%9aQ^yr%!@ys}8Ch`Xiy#;5ICBPWY3-f}dkvkrK3m!)&FR)O^$}1&od8VG_IC0uYZEFVhwQ9T$~9fLaOe$n09E()%S1bkmT* zhBrRMI!IW!Ndw-*K=sBPpcfMUCESD8-h(B?zxJ=01Fwy?2BQP0PtebBUrWSEIB3c5ApP2@QDhmc2#~VTIQX+5-c~03YuI%vPSjH-<**Zg1H7rHe_Ju0U1K4$r+8Au&OQkaO>;Mvj>SWjPSTyJ3N5%L2~0g zWgf8>qF(>LA1TJA-gaF*XjyOT7(f-;bWStgX z@p1|PF`*>2hHHoyduZG^V^a&i?kNe_D?I`AXXL~Fd8@wTx|#aaeM20;TF`<%jlOL5 zQIe92Y_9B8pfb!xuCyW@CdZY2_TD0RYdw%?Rf@_r%XX$oJ&Af?c~Z4(I1B4lf`2*UxH^%DQ%amd+Y?s zlG1iybMz^o_=E53QN{ako%C=GA~x1nn;8^9c+5yKZ#G7!Z}dgP@}e+KvMZdIE<~Lp z;$IUy4UMoCODRG=aur-wnoYU@QS;QU>P#LO)cSDD;w2+RkN==!rLR()Rw*gd1tN+U zZPu_`7_&uiqx65SZKex2(XR;;(>KE<4gb;J2&$x&T~HyTmaZvOe?O(ieb1IjTO{f& z+>V}$jD7JK9VGU=$9Pw!0H;3D{}AL>d^bQ7I^GcC+Dk7|{m=B= z*o_Xiev=V2!OW#}=6(no3l66(BNiY!1s9nirPG(j=q=HGYuY0k{a2K>1Pp__!wGAk zBPTPf>ffSA#93ho)=lg7kTlI~YbnCL7GI`x-l1Suc`kq|_eTKP+T4gLH}%Xos%N8B zmlBZrKf-scY{+MuhSa-l3M#ySzw{RL@$|GA-V!%138REzzGHY3_j=!kO53ZHD!f`T zvAGfF{)-dBWKl-)!KlIImo9Sz5GiS8Nt(*^w`6Yf^*vA<6xCL5Y zz(d$?()ENyYJ3Uy0-{(Gpx!iA*gRh%d*508SGORV!031?eZKs2Y7oyNDf6 zp>#T$u$Q18-BM<8>x8c?tobk*fn4$b8m!IH2%W@+L+p;z5(Mp4+}tB~@ND$B@)vlk?VZ7G)N#VF5_=?+@-elrI28gO+3 z2gC6nU-^oZPua}f1ibHlnL(Gv0Q{>ERD4G2sF1++WdF)MXnLr>Gui%jw!tTV$l6Fd zZTQ0caMAH^X`-+_vKe^R(U_<5n$G)NxR%|_ZjvamXoT!$4?p|#$Ys*Y@Qiad?O3zL zi|41hY!7tjIMUgL;$WRy5y^KY;RfWZ>utZdf+0f_mw%Q=OP}}OM43L;yg+j<0=FvP z;3X%L$P-2NN3h^tdtRBG2y#eqwr>C&zpF6Xs)QM=vB|SRSUkD0VFlqQ+3r`cTz-tz z^hbR{x^!x)0~o@$Fax=dnX0!#EpXqUc!~K>+cxNLQa9?4NGepd#=6ML7RX=Y{>D&L zhfOb48%>8N58E8aA*lKUTYkl2>bJ0O=NTPFjn!>r3&se3Btg}Rq0n_smg@@FdHd_3y5yx5m9qXrA zP@uE0fcj2*oVkM$UiS+TKtDj(!Hyq^h_A!c@lCT^r#==&3A-TW%l>qL2)b}wyNu}Hr zwn-N~HGg@S$IO&tE@DcoMRM*B_@uYCW)DP)3&P2NTA#An2b@cpCHTG6Zy@tm@)sZXu@A%1)pBT;v>uj`4Mz%U&r2@@55y&pr>d&rN)kDkQ*Fxh^_W(~L3?bzKMs9L9h0?-6>|gP1etCuJ9J0zya2z|^ z8*i!+eyV^749)-z5(y6u-Df+X8=4S)X?Y6Y?-?_?$6P1#?{U$Grj|E>L9FNTL!Un! zW$%gIYH|E+TSz%e0(v-FyBcNvs&2c>en#5uqQoNkds9T@a4W69aVZs59B&>#`=Xg z<3*b?gpRu!SpHZC@V&cghkEyWeW(tlQmbsJmit}?COTtojfIKZT3=v&)-=4TeWt!d zsBAo31)3wG3$&|!0aShHFV)6DpkE`T=IGSvmFPO=Xx(4 z1Pjpd27ei zigW&5hG(R!V7Qbt`>#49Sp%LHiW($20`~~q8>zJ!IOnZ>uZMib6RM5VaVCEM&N!4s z`AWx6n2Nk7WPacQ<6@kjGr*;YC1nUI?r^&p;-o+jV-rC_q2db}4`{w$6dAp|e9{B)55pS`N5lFRFd#FI~ zhw;a>*~~Er=CR3goUt;iS;qrl;RM0tq-(4v&#^ybQ;qU~#7}s0bTi^a{335(&?4@Q z(s1Mw&_kcPgy+(pj5osZjRVoDt*w#oyJDV7w?t)AhSSIo$>~w1vi>*&{&q#WvOAg5 z4&p8Z4mKt$4kTP4+RPno^VtWWEqfVyffsk2C3crBZt8F)pH1R-=LWFn(j|+KCtlA2 z?CiRG_3XJn0_OxyA8`fnlC7zfo-6Ms*dN%a+)tll-mrbi0A_>18aV?OCr;8C09?iI zb53_PTAtGDRsiP!B&06aa+_wFUI@%{BXl+lIg~XpMhX6^P4zVBN!==Il+Q8D3SEoC zOX)~dvP-1RY~ZUnaR1t9{pL83xw<62{|K!Z-wrIrcEd)ukeNIg^%l&#lX{otidt*WZ#G2Xy)1x%_8vmCrb9EFjKhk*wriiJ2tiGQ_ z#y=92@-#|{?~G-Tqso9p;~Kjr_NX0WsdHnsnfa38PRYN*cfA&gwi!5{BbZ%E`j_%8 z{YG+Gy+Gq@`uglJPg=_K+o5dU7o|G02xXn88&Mgj1RcB+fsR&8i%>%T$j=76 z%DrxwX4MX%V80voP!IDwkmHUq=|^^HT05Y$!2U%ciP8I5_VOt0>{oJak`0J>rWW|q z)&LLUYNb`tW0#EJ_EP(O-X7jwNSf;4*qu>l>E{_?~Bu9%lb0_-6Emq9A%oHB>7jgU3f-r)?n)nLg{1y219Zm+CL!iILg|j_)0d zD=dSGEw~wpZ0lS@!PqgdoQ=3(CzP|mcxBnBtFPt-BU=5X z!oSc{CJB4@mj%m>WP+hAt!aW<%{JMJS@2rSR zZZnS9aBbE}cnwfl6*Z2~4yn$Ed(Q8l>*pb2~wqu0uFEyMszW+Rv4iiQH@^lvB zprv?Pn=nq{2WBLS2GEI!R}S@MK_cv)5{%v=wrE7HQnjJTWv+a}WgClaWk1ikK|JH* z{Tk>vYHcC!M2_ES_GrB@Hl+9=qllB!I`Q7gS`=!^jE2x6^4J}~90M?iN1M;d<5(AY z6yj|u8MPVSW?ol!NMhEUDg7APjJn=(Lw>QH`~ul)W5tFNpJlRDp>m-dcu#oHKO$7s zF|qTOAMH;2S=cm6H95@mx8jMk!P>dSgVX8mQ;aI@coUIf+L|0$Yc8q_WYt4ZxO-+> zPIq+bUyL+o1>#s^ho%2K!I2<8VuSjvCUoX86V04AkAZRjQ_isVGffbK(SQ}bSW znL>QNF`=zqCQWy^NsT?>DO)8@c;MPTwy?X$6AQP#g`Wd_T}Uj~H%*UsK>O36y^t(~ zu(pEs(SALrNjH|Rs7&i7Q|ej!=w+#Y8qPeo^dyu@Gz%?zc&2j}FhjjtNs7pLM=(K{ z$EAPSP;2ZzIyO)b_?JitsB(i08hS#Q{&L!0m8SMqEllpTAgPc`!XqS>MKb?Z!mZkt z!QSaBmf1C6=Nj1b5fO3BD<85-GK@FEHDkUXlXF~~st0Bi@8!n8887jF;+Nbd)@`^= zPQDlvJH{E}BoBd|$9C9Q7dprZEBhAse-%zgA2S#$(F9HHAX5{fel=dr;MiqK>)dG> zPkhtB4DYT5y(jxTjvOFF2>Th(1R>@E-lDds(nS~zTa`8zqiid@>pH4c!4U1a2?;&z zoJj zsOTA7mwlME3G};9k&Wpu=;v~)Q8kFIJ!+4HO5rZ}bYpJm&7NT5M1W)WsE;v8nod6Z09}lM^ld<&626dw3@i5B-hDG?lIGBDCV(PW^G~ zpU#<&?*H_+;W@h2^}@L5Q?2kqR{ z`5dAPwefnJn6}WEi}}9WlYfWvO+nS z+2gfysp=Qq@cNDDzns#90cyQ{6V{1nB(F*c?%+**&}fY)^i0aAM-o1nLGKk(Oec#e z{OMoQ9e1Mp(t64NQq+gb(@a0k1qsg*r!X4T*mvl^#D2!rbGrsVl#U-Y6k+w(sO&JT zdZ+HJ*%GO#D+(WNXKf_kWiC!g5WdRYDRX%b5+%F}-~dXJ@Fl5?1HeJU3`+K`$fiBK zp>$>jXsvEY2ljku_BsjQ#;2FrY-ch5FzMD9^^aL5XTJ1SsqnSoJ8FK!Qa}o{C&J4~ z>IH`#HM`ICi%wZ1)b-FV%3DN_UsG-^`?K11%>I+_i;%qj&4hAmZk{tuZ+tGEKmArO z@S~j)XM~{gTU-i>#six36Zd^NB~!5<4o3XXQb51ar0@W!7A>;3fC2LV@VRwiXv$sNWTewNZ zga&mBq6NtKtc3k(fsw%bU8^lk8_S2~aqQoTR|uD89!vhIlKQ?Bf*Z*vV3ZB?!+3kV zBbr7MeCu;GlS$GE48SY8<&D_R{Q7$L#$cKe+41rzXv(g&D z)bF>19ihXUopxnwXy8}>3IsF<9LQdR*b{2@iQAx`UduR=`Z;}LL>Pyf_Q4489x~;S zmkHa$38R*ZtrfDR(0#%a_S?oaw(kh5eRkQA(dMmO6y!&XmxCnOz+9PCIk%#3g7q>! z)^?q|g@~a32iawJwZ)m9aSc2jO6NDer#X0E9KxD_`CfBKfvU0mZ73;B{XPB!g!-f` zpZGhHX(jtoP!gh9YTlH5NF4UcAQXZH=Is7L_-MG z#y5nHR#dxbHwdF-2pwuoimB8NNvXAUYm&C^Z&X{`s;zai`*y$m_WKL=cx<1~-tX7- zx~}Ktugi!q_NhD8_?a^ZP5N$N&LF%gwTCxwgPt|)d>UFEm)EZ>6<_xImKTYo1Lxw19~># zwgG8Ht(}M5o!!CqaTzmXr_c^juyqXhCTG81HXaIX2x9#&n#^@#~> z&bHguAY`uIKeDEYGhTfXa5LnK;;=B7_;WIhTBJEeZrIFIX-poqO8OG1tJJu6Qsg_Z z70~EB9T?5w$4}U(dJ9JDsl3V8rKW%bBWa!GZE23DT^dI*%Bm;cxU9`}p#WcETH zC|ZEcp(|rax%QXw*DTQiBefhnk*osr+3mXTMeJYl^bsPY{2Qda>U~#F{R6!@Rg{7% zfoA6*um|Kzr%yyi0gK?w=1JIhlWBi4fc*06MK|c%nV`Nq&~(s;7VGm6POqZ5`Dp+u z?;j%lGRME$DtpqE)im$Cu=WOOz3};PyGE60*eqWW7@OjFRt>i-^kFLsCRh%6X}9lZ zElOOMxWM?2oF!Q=qN^w>XVIyI)y6yO=yA(Av(3y1cqNy$WMORWwXPA5W5Idb0yWo1|8BOwYQZOGnH!kcnaoKh1Yq5pp`abL%Ru zXY$?G=6l3&!adKq9`2SA0g!ssTH!4ZNokJN>X*_|4bR2T40iQLtu%i#6D4?yy=!^y zw{F#cvgR?93M%YOq-46^#+k0v2U{3=kEWyt@6NrnEWteNyr4{P7QukzW! zXG&fr=XGQ?p^-j3@!Az2SMVWnsAVgHoG((7tD_`$LhvJkM%o{I%b)}E(Uv!~sXUQJ~D7ZZktY-Y>VG4tEIwra$OBg-TI>)gIBc5*EnL-r6!NSJ+y^7^yu*YfVo-GQ@Ikj_~3Ad-7;vB z9y1zvL99~cTq9nd(BL$&JtOrC{SgD35x-S{#pTPp!Q}s>No6Nuv46-tI=fcmFSf-> z*b8P(M(xcFg7V|qBC)p!Zixl0YVRJugax^_^0$w(EV3OojQz7lG|NH)IygZi794LCL2Q=F7m7^ zoV0CchnrV>op5y0478s85d2*@J&8$0 zB~z5Qk&gjA8d+sm(Tj*r&rwN5iKg9s=FOeH_ zD(tOSZ&IgaP|T?aHMMESBpLr3w|U!WA5jR{ zN6WJ6j{|Pc3&UA9c}^#OIjMwxE=@TjD#01KH0j z`{Za-w){rAq-IHqmC2$#2?*(H%R+)LZdcM* z=GVYun4PKnIq9kX3{iUIbe2{l$B;^@HNAK7*Qm*1cc|^j-KHA!B$#ts+870hU%s+d`tWi4)1n=p`; z2E0IQ?~3r7J5%7<Rv#$AeB_Ljd}R&{ znMOj25p{Hm;tlE>vkCo?E^d2wwOdmG3}o~ln+`KgJfBd z2JJ+iy=YdLAGp}$!N1Ex~V6kj!LRU)MeZTVG!G;j0c6uj^lAs%0tS;Jb1-^kl) zdlvA(Fel-7)eoeHDdmlZkf|8+C`=-hW?G&Xu|b}lJd1C@&uU^5_5MfCw~l*SV4zT& zZ8=f}TN_T?hZ8VuvnKujKz9_JV&p&gyJkoCOnwaqpQ|)_sR^_=-m52q)ChfAYS)wV z*&h_vi;&N)@`JHddlaPI&rB#H<-2*453U{meYh_n0^+V~6h+R`?H`QOkVH;nf}v~* z;eP|)jGeCxwr#9AK-g_rI4dy?k5s=KbZ*!PMUU-OsDu`lrkn2g3+ybXTYjao1 zM+aWyZdS*$)R)H3dHLzmwgNDr!E_D&Gb!OhVK^sY5d=ExIrM{_;;|vD zfDPh|ZGbkJ_@whT@x#tDP-DHV^1|_;A z|0!G0;%>|8P4nLbGq}cOlwb>4`xP17y^O=2OJPgkjn)3$L{*$^I)Ubzl_%;eCmX>z zr(&rs^c;(krk$LqYREEPrU7zfuq1kTtsAKn@;l@g9ISO}WY!!1i3dozYucT3s(b4I zuxvLbo%n)u(*e($!+xE&l)c0=3%krtWrGMe%l@$WdK}f9R~GjiYb=;bomxcrP-?;J z7LHOoG7p2x?5Ht*C(|+Cs@^NL%7&i7nxlb&O+o=%&N`#-3o}MCGpPNF)UxViORTBh zK$*?hYMB;YOzQ-5>FOFngiRQN(~2J{Kooo8_vG{A;amgzEq(%PHF& z~-G6UR=Kw#fnqLHgihDRljJ4 zcX`2YkmAjA%*R;CLFQ>*ue--$3&^qoyFSVqWp<5DPQD`huQSrU3Yz*H1>!syM%qPw z{*?o+$`DT2}e_hWMusB2yA1Is`&cI<-QSNNBPMD_cLHDY)q#&FosQegXN+{iwS zYR)c#@)k+o`OmjE7;l7>ki^Bh8z5b$Be3E1rKSLMlX{lJu^fmne=tCX<7bOf$_Fhl zb~=Pn)8<{>Gkcjd@P$%>QsTPHUq(02;Nzmywyy(lXN?WziSJ;_3A`b>%5nk=CG%LjkYv81ISFa zW#}_fMLjKzsinA09E>Z>qlRTJ>PlBEW!^@wahm=gqhxYc{XuvFlR$zjCB#UT(m>GI zb=nt>Z=RF8{{QpAyN7|&Ww7tJK1UPsXsJk>m-TYMBEwsG;#tZYKtR$DO=O3II1_4j zC*DSQj39JuROv9!8|UDjQ@d%lD|F*LtPVS7*bV9~Xfl_0B;2bt&IPId#{1;{=`oEM z$!t9K&Cdve>TQL6Hz_;h&)(Q~O{{O_-}DtnzX_Zzv&2YyTh*uIn7DuFetDK(aj&#_ zqFaV_@=zsaLOf+f4HeN5tFYlJ`PmbHpSlF-mifXUS3jJFrY! zC3nMd3@6OZ8iUl3q(rhn5IG4mCojsZ;)jXO+-YpkHxZS|@BPN;gj#-*_ry!2%%&)z zfmcQ}pDiXNMVrG#BP~O$6rF`!U@4)>?0}}G^bLj9N#-zT6lLnz7 zIw2&HjC~8)tqbm{4{)IuMueh{K&6`j*=5#F#MUx_xpmq%A;y0(JHs-^G$eZ!B}05hiDnkRxUt;z@-RXjHS4!?y5G%Swh zm}t=PwZvi@ZjbFJ_FAhierKCvQr%ol4>*p62vCln%Q-pOuxAPn*_QZ&!?hL!`N@3E zc(oA6bh%6LvVFTwvL8}N2u3YOwBYJ{*dY1&#swf~@iF5Ss}MGuKMSf7MaDTG*(Y(xB-mKNcKC5qZNMaU zWkK1?65L?wors34y{v8{$~A!GKx-HQ!HYd5YLe znLRZ9+S`Y}kUGcFJpZ>MrfY>;B^Y`*kaMy|HoqlqdG<@Mdx$m;r(4l|_vh*;1u|+P zu9$M$E&0<`IcF*N^SpXj!yLzI7Bp)7UODh7N!%P$*2Bbh$`9r-Kw0*?z)UciBh1LA zMV3U%SB_}Pv;zT;5ZinB#4=kwcv;_Z%3fli37|a-D0c0)&fM*4rIcE+KJuu>T@VC& zO*9>T$y%4FT0M~AD3iP*U7Na$egz(dZ&)8S_AW5C8!6PW5~6EaQuSoZcRq^|< z#~~p`czBeUDF4_|)9C%m)qmoYUKc zI9&$zXTn_a8U2a~Hvq%?dPErgVk9TEC|TW0uC+NB8U3wE5Q9@vp6H&sus_x>RT!8w z=a=X>f*bB{b2EP+Ch4ll)9hp;9%Zp+JG~zJE#*Irmm{&2|?UGn@)jG3+ zg|dt6l1)86x>qu7EBT=x7@oli;Wsz}bq-))b+E_p)sYnd_yclEgl!qwjarz};zdgH zgbi_Q=RkG9wn*JA8BdelR0QZ)|IENhxcNlc#<--95C`JODXU1_Bgv2AyLt7Em*C5! zjak>r9mplCncL>g(n4uj1`tTKt${vvxMS%<#=B5j*EkgFj{8@hBhvdwrEfy8r)VIf zF=1iiVoN=&H5Kx^=!yk3T?fA*sdL?_>B74No+Ni<)#n^$tcKqu&p}-0W<=I!PYtT? z=|6fl6-^|(>AFcr=oeFU&dQX@pUdimbnQB31#IN9RweuG8CC1ruS z9S%5G;SNv*R-o2q@m3M15XqjzD!cHat#Og^rD{sA3>_1G50P=}gCt|^Ax|TILED(C zP`--(yM_vN2lKyF$Ix$-sncbgFOA2@M_|7r-&N(~A>wD>y^x!2b3%Tqd`flktM=g? z0ER-K&}AuA_)VtC0M5B$O$B$C;FG1q*rwfU&%#b7sA@91hCiFP1|X)rEFhdze(Q@{ z-~hq?H$R5^Cen*ETjod(Oul?T!tV&qnzcT8Q?DwAdL@f5(2pth{$FXC&B(aJ* zy6Qjji(Ko&b}B|I8_A+;vuAW^s#xlFq>u@0au;oy-KE-tM@pw6lOVLLOsBO8Bhg|* zDLC0#A)eI2k{*Ou_7AU~^$@_}7tbX&d*^e`z~6SScpM60BLat*_eB^^NS=4U%Pawrg4uDBU}nNr*61%tl;$9@Xn0tPboBzwlva=L=B{xuWeZXr60@rUR);2Stkik^|D z&Mr-KH!a|y@V{D|0=}yu6OUbNO9Tfs&DB!yn~*&bYxoJV3UA(i#mp6bK^Mp_Kco5l z2{I8k0vB()lhF`7S|EA|iG~*=Lf|f(1CV}L^4pYlD%-c&5&u!%P=%_xk^Hh|c>mkH zFW<$#QIA>itY<-)Y1kFSAlr?Aam!wbIGbJ1 zRW23%Ipu8-zj0#HM9ai!*soJ1C1NuzybR zQPoof>yMs%n=--`-70Fl0obpw?KM#4cO>EWjG^!fg-ApG$%Shx@a-|BU9G1o z{Y1YyRNI|NM|w6L$#z5C6MTOo{hlfb$;i%_>Ez7{r@U6r4ATjL~LCZy8_t44N= zaJv?8BO1NLPX93Uix~{cmYJoM|B!}krMhWGNg8-PQ+2^weNN4sc>FGVqO(2g3R{z< zQjcXs=VB9TM$AcP`>$(n$ix{!J=-AC?i#~%3SZH0yNRs*n%T`}Q0@`yW;8;?`Dkn^2mDRQ@3>cj9qOWqCgwi#lW&|PDJDC_*IoF6r4%I6Xq$h zKtY_4jzFX0X;Csg52YJ0DAc<~c01;SL-HMBa=k`n>EXZT8~&aa#F8(25|TP!Jz(&| zni1o0=s|Rk;Y+GaKb{q^P>mS~@oQ^r5xmTL%bRb>jK6u66lZu_rP%8~$5P_WLN~eg z2SV*00~1=6PainMvBusv3&1+DPLr!I@*-_N<<(ZBB5EsoL)#p)1^2 zh$jwnEicg5sVNuUD{^ZLALZx!`O34hqzNzqnFu!xCJlo}HunGxx3%S>bLv~7ZIoQM zaOCIq%i)+7*l&s&Qq@9c8nsM^fh-O1wMKh>mXD(AB1R`UrF!Gvq{dm2eFNd^r(t5# zyC&bfI18sQAdT{sc`Ba$FQy>=vINcIe21Ct<{?5(MB{KB>dUIufpBq}0qizNK~8?6 zs1%m_F29G51s5Y^Xo5H$5H6QVHby5E{D>xgMIUz)PiF$6O$Z51v zc(Ib0;StFfbAjXwvwsnyjaT0~i~!Qf>p1U7XW)^3S~f})qK~S6@8?JlFMx0fH?!;W z$3VOM;w!>>+FYdx?7Sk*#JP_3-AL6U!+MWPDQ(4uv-$$iSNk!h;iaY5W;be&zip)@ zga$lQ_pgy=%F7M}G*Q1&4l`c~tqJ0X>3xmP{apE32f;s)b5B3c%BC(+jKhZB` zBsCm~sQxM85+Q(eU-y%*JU3d&`Fl+g<`!=7-r;C z3#SlO2fyv5_WBm6I)i^E_Om`9)HK4Bg{DkObSUUf_M2xx~XyBIN8(Q@`93* zZ-V;1$NbW;bPJ`ILn=gE2YR6{MerG%k3%M3-~c$`ol^kE?{&-#X2SRCG~cf|CDGc5 z=GDqFUEiCM6PD2E)2!iA%U0}i(mTrto0>gtR=;h+$jX1F`({DcksV((znvW8Hk)dy zGhA>m_w4^!-qdDHmSg!-p1PZF=_Xf*Wq>Sq>kw!E0*Yr0!GRJaiVC4JV&RCstuvrRV*=bDMUgvLW3W>i%4fdedZt z^akk=dIr?6k5n`Xl4G!P`cAYeFWWuBxZ8jVomj2W1LeS@z7?HikQ;?>Tkqyl(Y0Cgin20+$#=v2p-m zSt1k%lP~&b8NZIA{banZSP#C%_Ro`EA#qqc1NhiqJ#u@I5_h`CYMQ}3Q^ zZCaX2X3eeM&GSxniunbY(@`eu9ib+k6ve__M3g7wp?n zO^k_vH#y#A4y#bC>U@7jl8^E?UVCPqWnS_DW*9D1zFTeYes&0csR`61Ob@>kNqdn! z`%U~CdVCS#Qq}Y76WE8?EcIr*3wyaen#A!Ykduq_g3TZpZo8iz%f@U^v^@~UFK9zQ8-kw4gK zpAzpEu2nFsNjQbQsw&=0zr-;tM^YS(Y2Zm@^abPJ zjTYQHL+K8$&yM;KYs_MTVQ^2|XC6MGIP?W_WWv(O7?g4ZxUFOg$I|n?U`ARK0ALO> zR4FD8LaQNfWnXcv`p)dXlxevAn1buG3SLt}DyU*zCBKr^l?)_`e>H#%3)H_i#mKC! zty2oXYx`owV>g}NL63q(MIEqMt5aIM-Gnm)%>6fguV#h|)D6FQEQEX+R73LQ&>TNu zJb}|^pD5eKko_XQTbx!0z(%I5A13(^hLi{Phug0zQcKJpCHC2!_K!`ICNJF1Of>$_ zR=!0g)|3+w!@uC(T3`wl2uk%FjBDd^+N-(n8XF(a8Ml5;N)-Xb^Yv>g;qr8f{?c?r zY_tFq0B+@c4|vc2idw0F^qZ9+)w$-h9+2sR?RnKyL0e8iI5b0QKb>d|X7!sSS2X*F zB4J)NO2#U4S0{W%AKCb16iwuC6ew-0I!))AH_`EV`f?nRUwx7DiOO{FaA}-c;Z{+8 zpNoO3FV~RE^{?yzE<}5PgSI~bfAD3`spY%^IJ`0^y~3`!tplgWmk}P0f5K{x;Z9Di zp4}$cEq!n4@XE-Yeh@iTnSz{XKb%m{CQzGxz)mR~!ewqmuX|+aDX8 z+2vgtVvLB41A8t`oyS)vak4Q>;47ul?4r~#zXw3fU!E5=dbca-alm9~B=a!jJW%L23sf@#*m*)SRN3Tv#LqPyV=h`jw^eg(E zzK!Vp!>0`UBz$xl>Ytm=2aLt?1zqF^5rr@(Wc{1)gZiU(QbOW6zk1+USI4d!U0-q@ zz&}}lTgrHQk_Tu7;-l(yyskZ8$PI(de%hKG?T+q6-#jZ&Z#2INzK|QQ?m#z{Fhz_E z;kaGAE6QeUUP+ID9Lya(G`VVKG5Mu;wQp@+*MO7!_K?D6_$}#Yzpm!95gBkXu!p2u zGI8#Xuq8Xo4l!K8URRz;6y(Lg2uFw|4ugK3q><9^tAb6aa?eY*!$N0U9HkZhq#sy6 z*@(u|iV4e0@~e0Z9WIVKtcdL_Jz9BnCH+srqj;98F%Z4BTYpx69};Z%TVDVu^U1RO zqH}N7G%fR-$&E&yj4CCwym5WfOK$8XHH%$>`1j~nKM?;helc+}GFmksXC_XLjJ0Z- zV+0HJPozk@xUXv&P{+JP#m;`Aj4+horgOR8hS>^NfP8}|P(aX=L#z}8T+0*f#NISu zODPf>L%FEBcwDpDoY=S0G-TqMF0q6Ib`cu*l|yducbHYNVH`TN5cOmHUz8U{RnNF) zr{-gdWDhH?%xCRQ!3ll6{Wwr@eOzIj2?06zy!2rjH*kVUH}G{+kAW#=(M3?PUy_lu zi4#TAkIyj5HXz5B*Z!#J&`+SirQPiB_52r%+fJ`wcS)ape@DrdN}|6+h`_*XYC3hB z3cs_1%&-T8R>yC2iKQ>+H~G26J_1p>-afKi87w;x1aP!wzV`k95)*gwY8an1+?cmY zYE4jVMy*ifM*oP)=5~F`$leUN53Oe=@H)C)2PTwimiqk3KRT42(vCexxj*nWpcatFDkl|`skB-y;}YFe?miaWM|Zyjq8-znn`uwv z236kgt>#$L#;I0kvUAux(5tS4*-q4;sB~Bn{$veaD6NJ^FEQ1Le^fD^f^SuprU+X4 z{0G^Zho29!Z?NXu>@}&(o~ds+>O*)KBz$NIbNZqsh!Ce*Y2D{`9LBLcn5DZ7+KoVY z;K)N|M)$OdqhVxg5KY%rtcl}qyuj#0v}8T|HLqV%t3Q(Rx>_FDWSxp>bF(-}A6tx+ z=65-6+CQ2ej(Yg>C+>bhoCTJ5jp15_cDu|`MGNJd*$Tu`-pKs=vE{s&1@%zPiR|zN z*@us1C$7{VJsj2BmFt_9=ub&T>>7yi*mG`2{n&~&9R9_Hw~PSBZSj@+;#}HA&-_Ka zXT_hM=i?da#mhbOJ+L?TYiO6ZGfR|V9KGi|F)@BHq|f5R4A5(2#B^W3!%-KH(bqxxuu^RM<6 zx*QW}p(ov!Z4Usj>X)hreZm=&d{`Wfo1>bNaeifXNF~~TMj%?+BY3WM;Jv^l32H4; zj!r!P>Rwhnne9|p$$mm#g!#4XFzO{e|NdjtO-#ZTcpI#k+47#t*=6KmkSFo+kbFi%K~YRU7d6uuXhnA*>i4NgITpTquy zt%^_g+^hLB?|gyr{V8HfFq@uzYY)RVA5aB^>^KeUR`gg!EePOt~pN1Cw0t%3q28;;W{d*KfTa zGU)V{`AWcHNj4=0+qv_4>`UP>xaF1gm=Ri6R*WJA-jke)ggjVH8aYd2iuZhtsV8fr z`f9+{%_i6vd%c?sqc1`Xs*@8~zP7nT*Q`IW zzu0=K8ky5O5srz~a?EPL$6Fg@*)4X-r#em}uH9Z&aINc>>xUV?h9@1_*IRGaT*q_( zN7^?6>&xDNM4R7HXP1#6*53)5f+>iFQtiL?u#G$OU3>=!4*Tqy72A9j`SHFxB}B<> zxcc@KIqBLj_#p9Yk8i5No%%DPv-$&k#sT;cs*@Vyv+dEgTg?xNl|Ex}PM$xz^rn7x ze+Zf;o&*Guf&8f9C1tqs7_%qsj3ShF2>!YxE8u6_GYc?x9m&Vmv&#A(5A5<3sCzcE zX?PJHRKQOa@Gf2>cF3W1o4R{^Dwv@PpM3iycMGu%-&oQ0Z`h*16 zc8q$Bd;s}FbqlOpRC8-9CDC+tiD^-uAA9Lq4&=5a-{MU8CCXm6qmfd95DOc(q2$EY zKGYO0?BB}F8bWu>OeE?CDK@(P;`E8W*}gLPEFk4gV)G!yo<(hE>}TgekC8vSj)5ei zmLSgyoch-P=$-E6g4$S(dL!~r#S_>{XrRAslcrWzSa#5Usn51wQI#0D5WLxiFC^?o z{+{ZmP8x<2rXE>>rN=n05xX&S*~f(ItCpzv!NkMWKDGG+sF{GVld?!oyQwh4=Gwf} zG-u?@o2watGRb$&=&&faS=4K^4s0T9SGfyb`%f&f{YLwZxGaF_PYYXTp#TU8KFxHsNlpYK@F_Cbt6S#!xiP)ESUU;6K%Ir(e8c)j7N_^wTbwMCQXcdEZj9{(43lJfE;RipY? z0{(C$lrQu9pYX-iQHyB1>~e!sV}(Rg4xWwBNU5?<&$*d&-8HvYWFuD_*y_^KNZ-m& zM+7SWPVgSKJGoP`-G$kKK;k0Y8!OG8h>Ib)LN8c;@lUhGvqJ2rSm$shmbet#A&DhT zKrzfiE^=Vmc%#b2QpKTFu?1<#9G+n=^&jp>y&V!B3dR29Mi9z_8q{O~e zAps9sp-e1YFe8hZwkd@sk0EEm&k$~`*}4aR+2k`Ql<5v6HQmlTAPh&{EXe}Bh&a!= zE&SQ|nazmzkFNB^_4GdTpEQm?GmYH0mTKa6+9H(P8pSdE6^q2t3>AV}680y(Gat;1 zZ;nr10fv4j7Bi{@rKMaq(X~_;e7N#5_HRn6&m__|(ci>>kyeaY3VXtOOqgo#dl}1Q zI_0u0r!_b3(}`KbhvBJ-R~4FpN(WbW?MFZ|i7>}sk^LRm6Rb2)M}*=tT9%*&w$5~^ z6F5RHTT$5g=A=GFYL`5CVzs+8n zT7pD9{LLqR*&-PJ#g*KeMltz5dP`-Lso;c_|6o~KOFJ{x++kZ zip1Ay*NYJbF2MgYwIR2XIAYr9auobJ zr=aZe+R=00=GjUo)!v3ap2Q+v3Grfi6XxP9*JH;hbzN_2)b7v8Q=6l%n?H3_@1#y} zJB}R&P%!-=eSg|Q9WJu{!^OB4z=hENtk|~+M+~Ws;4Y?`89wyL{8Cf7w6epfJN5!} zNBYv!i(CR&!MQ6~ZmVSPkmC{x44!dDr{0O|1~Z|Iu$dgmHvcm@*2~j3jsrQ-#P~AK z|0ZWEEIUZE+G@B0L*Qw0{>HY!(T_ z9RLxH<(WGyYpvIh|3d1#!oQ5dAM`**y+Hh#yuLcTPYack*rlHPviA@Rj1&qkQm=V; zG3tr+DaAY{0GV-7!?|S%y46+x#!lS|lJ#of}oG}CQv=6U`j&r9bE%cdPvV8bzcnC{fOyXoxVVEgb z;sIS#jpOAispY0YrJCmO$XaTtpgFL|zLywm^8hA&Vr?G!AS~*d*@;G>A^ms8_Sa| z#JmpMfMx>oO~0{zUAxL$2*`pRR%_Mertt|>6Vv`b!+I-X!oO=5CFG_XDNfDq-ZXYg zSMa=E_@$)CuiRDJ(Sx2ytQS?5vEa5q!n)k8&=ZD-q0mYdC9O`g66``-Tg7DShmLZB zJc|_d+^>{7gSOxl8J$~ji`j_U<=f)i-rRvPWULxMSd;RIUW!^2y?pXE zkNe2x91ZlmMt^H#!k1f}bzfiQNkgJTh$qOuB>!0zr-FH^Y=Sh~ZiB{7Jn5Rw@>o(PYEj!Vz>`DynwXl`ay^pnVS#0}s?Xbakm>+ORG>TG z9b&{mx;4tP0GYmBI5tryCp>5P3m2sqjzZGk>YW!|gwbHu*n^B9 zEMD8zLxv6NCrsk?D7i242NGQx&{Z{33lkDY4&E)9)=bH6vnouN^o`&KPCi$g#3#Qu zFC%+S88hvX16{b})3c3LU7MQS9TxTw{Za~kC4Vn;A9+H2o4?(LC}DQFZs(h9@M&1y z9Rx^dn6{j$a<%<$x02pgq?}D|RTupi3dpE`!r+_)qUr5aA_Q~Ei%H?bzyJpO2gUrp%k)C`r3et%5zPLA zKcfFgO=@cGf$xthMg-E>u{mz)Ld!`aUin?S8X~M33bwE0tl$h+uZrkE9|5Oj%Q5CA zjJN>S8rjm=d2viX+-H704~OPoEtlXXqdY!Jie?-sadPGMyV*|zs1RdTbcm-kZzf9^ z>?S`XjbVt4usu%2$-6Nt_&;#$bBWE}m>OU}idi<33}Y>@BbxbX!-A>gQ9jpXY+* ziDo$Mg!!0cRx+bMDk%(0`k09Q$SHA$b`x%Et`AZa5BVOfL`T`IZ?s=>8ZU@@E&iL; zbffN8O4Hy=L{ZpQ$3OISUv|H>1XSOws`;kx02V2r!Tgt+VlVI$a($(ULM)iPlZ}3Z znI$Pr>=*Wyzc?Da;+DM)Y&dP!SA34P9wNndrA}g!Qe4`zW_&GeqVa9eaLb?7@0=VM zK&tx0Y%(M``ION-8<$MPUamT?Kt%5o^x!M#JaI=cpljtadkG+lk>NzWCkzJXPyQES z@^09%3qZC#nc>@X+Kn#TJd;wDk7d73l0OZ=&LF=%bqlekKOJlP#axPS4Ic)qZETxK z9bg%BwZcBsBabe(F(>umoD4~a8ySBJ9>1JwhQ`0?zUibezvdrUQDi!~`{D$2N12k|+Osmd%jpmxuh0IES73%) zb*EibAiH-bVy8Hx^f)DIaR@tV_?!ozc(StuHEl_lZxXgo*9U#JdlF-GXA*@b-!8Wc zzyfmef5x*U;APs^*p|NBH0f)vMNZr5g}|rOE%9~E=K$TYJzSA4{y#}lg3M(kt0LYI zWbj8_2A(<$>;wMDD>0npoHJ?ZGYUgh5^W?_)S(N-RtP&(LwDfjo@S}>?sP3b-}B?K zp*nyy&KRv+E`R8S>crcG8C=6F%SDTe?lR$bRe@!-)y=vUTogXCU@~@l$8eUkfeCR2 zUg!3j4JT}(O+=q^ia11X%x`O6W;;SCAa9`g@cvW@|C?SsIDM`3N*FgI8>u$^mV!*Zp!X<&rW3C^Z68jsk% z*;t$5#fr`LPl;XnEZ!Va|333J!47jVHHuGha_rQ$8crJ*_)phI@L|>$=J#;U9%g}R z`TtRLE`CYh{}-=WnYpBLW#xr6Yp$DYtA=@j%8JU8b*1Hn^|i9HMDqg5tumu>%9@#L zDOBbyQ%Wlhxp=oS#XFfI0xCC^`~CLg_aE>8KJU-_^*ZM~54yS18k-&-mQPVfv0oeP zg0~Zq0+I5H${PE1=KaJ%hfHYfwnvt``gznW^IO4E{et&ZZD^C`x)&`YsB1HBqwa8b z_MIF86pa_sfw8}aKC+#wG(-}Q6=%=whzT(cgjb`@Pn2l!XyJs&$gBSiFsW!~3NM&1 z@e_b00V#hGPH|ieStvX`Lvw&GPHi^vl1`|bkP>@aoE9Mc?PG-fuJSKgpOR+&tM`2! z^d&;vdR#-qFJ&TCBN_on(u0Y$9m*>pVMC$U^V&{v+<-s%QQp$Fa#2Z03jnv>$5aC$ zo4}2;nxqHnNcQiPNT*C4frU!p8uWA|kTuVaAS%W;mW6EWl zE#LaPjmG+Fuv7ITmmCjrDK4OW{L`7p)VA2Gey`9n#f(J49l{i$iepa_OUijz2T3sI zI4dffTSpn?8m$!b8HK_l?cQ7ja)Vy_*hmn17#GollO87Ih#KiV;^3D1%J|3&q{ya& zx>)IA!mk$kQM;Hd2KKf3WYWwpKE(az;cX)j6;geTh6mxpJAZ2Qr`*J1pKB^%r-AjY z^?j}Fzp(&%u5sJc~x#F zCDvrENo5GmOMM3xG9DqVT~79&+z3)b`B!kjXe+^fQPF8>thdMejIl%p$R z1O24$XFg%+ek^Nij_sPgz0iy5IL3$eW31S2gO)9JB=8?Q#lZ?_%gapPurTX_3<0z+ zW1MdTN^SkwopohVgbEQ@7a^mg&IJfQ_XsS%Ax$pU!$E18@08uCFFwV)M(`Y`#HQ-UXO>2)2qw(&+By>d+GS zANEhZv%iDBMEd~CJj0r?)mmcOVv|2Z=_(aZxmF~ z7y`ZD3`6ezVRC_<=8J)8-#Gu7Unk+ z?F0RfH|azK25aA{KGUjbdO|D-MoK7^vxei%VvKs~kH8&%YPfL^ScCOb*txX<<|lIS z9qxX95k!|`-o69E|44bcy$71;GhM9#*LoR0K_8Mkpxr&JuXy}tMPrY&L;fW6J43XsJ}7=P07f8K8M^#T|KufvJ30TwXU>I zcjbCg1`5r8AzkVh!$Hn!0gdQx$L#;~$yL@ZVPA8nZ_sb2g|lzhRIx5|>s)}#3iWo$ z>1O7*JM~Lu@HgUq+rTAu6&@85@+#MB*t|@sP7EU%(vhZQ2$-2y&ybq;8Qomq_@t-)oo6b9@a&zbif8V}&(1)UUWO zLw|!%#Td-A+#}RzMMH5T=Z%}?UNxSIqmIcx6xqb>m{Z*_T(oGNgIE%D1cy0+{BLFr zq=rx+oS35I*t0X=HS8jY=M`%M^!7;BNKgy6FCM&*rO}P~D5gT^!?x3y?EcJ@8! zvpX~f$@WP@brrYoX>)vO6%XG`Kj@p&+=nj2^=Lo^UHT zoQwS~(Fuh)mO+r}jrSk*?yIKheTrw;a9k=7Zj9Ehc8l zGq8C?mS37U8Z>4sU>d4{?+tLIc?Q;XUvK^2eR+i$u0v-403py(RYUW7W7w*n6eeo+tohi8jbWD|Z^k42vYa-gPfcn4hMVGlzqo{A8HI}a@$tS)3tM*lMdZsGOL_|`YF zVIyXqd?_e9b^(-{ci!3mXwAy(SV+HZ+#u|fRm-TUY0~d1N(QP)3di-JI&nU6(Agx< z9i1ID|7qWyp(pG7fEy+1nF=vYo+>%}hgU>gEy6oFiLyMUm*oA5dK( zLixeHZ1|WPcT-|7aMGr|Hn<}!fY&o#WvGTq+F1a_)|vrQEbp&k1d`B9N3nSQw*I)1 z#*gHi15L-as89dY_Y%B6J2+ipTRJh=D*zISM;IShLk zlw7YnBwiHg%uKJuuTgz2tldl%QG4%~vipq9s3y05)lFzNhoU10_Da0i-my=_2*vUE z)CZhXq}wj`kFq=38-m?E@E!-POZ3q|D4M&d0~&`d!5pk^_eJXO;_txx$3PnOi`l^l z>X%#ES#a=8=(ET`d(c;#6#OT0EH=u=baLi3VTt-zZX?S42Hk>w#f~BM$kQ?&<{8o* zmH=IR=prjTxkaZBFnZ(vl_b1SG_A~mLB<)TpLwg~N0Eok^NI_R8`$4)YV3bI!5BT} zBq-k~57>y@)=IinKfparS(dm6`CgGj9E_J-c%1_ri}n#b(H$&#ynssv{c8P=;A=aE zsT_B;t`7^sxZoCqwr;b{JsbuR1g#^y!rInhf64=EtEV&gq;^3Jc%OooZQJE~H`a%p z-BX{2eV>t9k$fxbS;!%ALCwD$DSIGYd;e}Aq1UJUWIF#r% z9}XO{GTHJzT}S-&l1bnQO{tLJ$6f)I^19>Xw9p~Q_aF>r1)ysiD$VBVO$>gbbP2&z z{npxSn;+-+TWPq`OtsKtLIi1CjoL8OagJadpl<1^8sobe7rPLL8m*V5wd~m#g?eAFP+6`fb^FO?;8PG{% zz72gCyvpwo*EW8^z*ls`FvsZW>RRDuy!ZGO2%*xnF8~<%)u0nuj;2%aO#|1V%aMH} zg3b)8%#`G#UZ`!2Z312KQI??;KSAMu+j*;Oo2;)j-STG6j0U0NFgbJLZ8lqEx+@2j zl6*pZvOIWm%~DrGiS;(^EiH67s6}ex;niETFPryZf34ndov!H`K#r6|4= zI}YiD&;py8H^VFFab(I(ogh#iu?&CiJw9f9;vA-e$i}Z)jZ)i!F zjjK=Pr3{grAtM}rs~?1GLcho&zH$JF)$-`gc|ik~2jzV<-R4}83zJAU`lEPeB|$ge zCv%2Son_on<@cf*V z7f=%OV@j_8_g+CP3OvD$gX+ph9w?yHr9+5dGPyJV7 zk~D61a4oiv)AD#aH1h}KN#Lbo7CU*0)(xQt*$pv<@L92Zn!7iCf9zUTuE!edE!L9M z6Omdp;m6puJdo~P3Et5agNUx~A;I<*`hs%o3+>f}qp2aP&wqKcPZA|3gZIk-M}5Te^h?e|AP}PyXMaLSV;|&0#GuLjT^}_PXGb9w20Ko(p-YM^>cOd%7}t0wF@S9vwt$ zh!>&Eb%48o@cN>d%h5)@x$#9|A8=Vb(s?M&#x*8gn=SDYG^)m=^9*BJ|MXgn@SLc{4+o<(ElKvCuTgO;D z?jokJib)I&V#MPAmOSmV{T%=q$)L;;oOCYZ3~fu^DCVMe{oqz>Ue99O2FrTR!{RCM zl^s2)2bIxE&xzncWP7zaPXy~7eF#LrZamJ4C?lcMZvuFi@e&{_8@|2oDQ*diu3Etv zK=c+$hC5)z1FQl@jQ=4s(# zlKy|1ZbBa71~;w36pol0>2Tu|K(_koT-a|7I|oBzH~V2qan|tEM8S)Z^O&Bznc@4x zow{jv<`7t#*bkbq>pE)_1-3=-$MPKVcMDARQfb9KT`b`+42oA^BFAaX_-YzDO!+Dh6p z#15^qgSFe=`t|AN8Wdi0S#tFSqUm)!afmjwVd4{I79w2)844QfGicp2n=CSmHQ9Kh zD(Rysg8c}oKa2Y|*&=N}in~fJ>a1FelRG(s}Ke` z+MK^|8`YDD(qTiRHLGcYg~q0ieYU>>Zy3#fHgIFlN?VA9&AMnE9RY34GxnP|2|8cU zT12^oefpBTsG+Nq?y7!CdL?=kJmuf+WvENQAWQ+_TmW!>o-fFi@>b-n-6g*^H8|MQfxa8T#VUZUb;D7_e(p68v;+sZ^8U<=?PvhzL*?Nlg82+Z@Y8AXP!8=|P;j<48 z$a};Ka!35(lXBr$Iq_L0+0ZzOVfqtSLCy{1i^iSX8HHWZ!M+%+dROg&SrXJ2(8xFR zb02*nU1E>Z3{3IRnqZKnlci(}0>7Nsj1i2_Sn8zA=^QIK=B+*~z0*u09Y>ZV)NUfD zpH<*-)tX0W^4%;+k^wTctviIVoKRzBnqO%?G`6_)B3D} zF_Dfm(+SN|#frf5!}wSE+7ZDv?a7vMSf(4cTB7) zPL$(X+Cx)>_#Tyeup;sQv7C7s7BDL5&v9oli4|zQiZ7@GX)iddr#yS1w+P$Aej&*B zD~|Y-Qg%acvqn{IkgVHZq%TP`C=%nI5`x4N#*xcldth zTE@cIo)Jm}YY%@dK6zri44?KH%ZN?O=5=_3V4<6!KU)cyr*C`0<5Se8=Y-}kvh%t4 zRh#%_eh>(MWKj1R5tLy1Mzg0lWd>o5WPIqD-*bHICb5P%j4Btu8p3xSr<^iBf zWUBVQ%r1E5d|zF4Q&z)#*B2hj50e>R zL>*n0^Jr5X0S+3VPlyL@loaM7s}u|E{wyD&4vUyl2adx+f#VJv4rkkC6zS| zy>NK}=Dg$(0PyF?g-+g27gpLiks>Kv&mIW9Poy7Wmu9R}WCD2J@;bv0HSY%)GxjfT z=0yNWPTjkpPc&KlWoly|Ketm7Ka%osn=%Txocg+cmTdk-6(G%ODgM|gL)QBAoXt3f zx>W*s3UKk8H|A#F85^yM^}RG~?Fa8?peVBP4&*Irn^T_+6XMU4=RVg&&%Ga9!=JM; zhaFOL4VXj8r(Q)q2}z^cN-PYf23MPTU9}1jI+P|IZ*`A(QrC;AaA{gDiP%7mx%?N0N%YA<~A6nJnqN^hSUg8nZ?me*L+Rs zB@Keq-o=i9iGOCATFc*&RW4QF<9U&A^q=r;ivfN!E4L}=$A}Sv=kK| zb_WH#TI5%Vs`fKyaz4_sLj>H3<0xGtNxJeAw930OD~B|pmriSa%qg`X^H0Eh*ZVLD zz0UEFOH?tp=nS>7NpYNrd*SB)de+_Y0ngG0PgjQAe`{K zTU^e5PG|!(fv)GDBMq8>m+dmNCepM&bK|V5HgjBJpXx>{{W24+KUy8`M!C)Lv_>=C z2JNFpQq}oPklo2lD5T#ZG?`O{zyWF*{~9BCIKUVFjoCIugInSP{V^f&V;!MY#Wsx7)*0qi* zJT-Po>MJuMFmoYNE+7kZQsU}|-V;*duFZny;BOmKDm23%dU&?{wL>>gXC z%7_+;{s4l_w&srAMVv&rdoHJnQ*7M{ZDv)l7yB}Xy4#OBC>+sIG(@f%z$>m9ssLWh zSIZqA06QE#{a07QMfdxe(y$S@hI2rEHd1B=o0emG!l?8& z488>Z6pG+j4(Lh^RqS3lYRuzZATZJ9)8a3~)u%Zy*RnURK|VA4f`9j zPs>4oBf`$$tFpABqp%0%zfxq9aCRemv|PCd>OvdJP%`IMfn@Q$HryMN)h`oae!87D z^|37rDYF`1Xy;XP@FV_d52Nmls?ui=YPI$I%zw<}{`1cpm!e&P`^jHz-CyvwwT}_( zFZ99qc|AUkDd~*ml)+Lx=aTo3aEHYbu9BaL*$p|#MEK1H6$`nbBW?653S;Tqb za#$|wK++Q>K&InytDR!GNSR^1M8AoldP0}Or>at4m?735DF0!^I7d7T#@Ezzg4S4z z+-R3WLT)%{sEPf$f1}otsh2`cE0u8ei2ViNOBEAtzYiG^ON;|&rKZUp5E{}Af8K^0 zUe%eOZ(dNelwhC-?TLK|46Y=pXr|>!ONguaNfVaZaH^CgkWGBAb!3G+&$kz`Dy-;? zqK*-JRo zE*NB9)Ei$HIq5zVPiu{s80nqAiT@K|32;L!nhL1!M&I(k4(kVA(Qbag{cqxd)_WDY zsXaEvc@ua2;z@u|!0a;+TpKUx+wKtC!^zq9W(=-#Eyq!lR%5qc5b4loz*&H2(6bE@ zxNmg(LzP}<&;~s*w1#>W@&`(GiJDdZHQEOjCRNOaodG*C3^~NB<4psKH-2P8sCM^5 z^Gnp4VrYiAxqh36ka5@VmSMjmP8+hqF$&Xm$-)G6&@y0|^>>9JsA#Eg(ad({>!_AE zN7NATl=7_etA~yl32!VJuN5B2v|<8{vlB8xEikmT&V;Jh zqXO+L>jXO*2YJT-mN!46r@KA*rGMH<8&on}Me5@BxypV52uw=C#25WQiNZcU8d?lj zo-}Ww$XHqj?kJA+F6c#$EN7N5xz4W#cyse|_8maa_C#eS`EY9e917a&_=bfEvq_;~$7Z zWbcvt<-vWr$u-75hWW*IJ{J3w?+md6>&36^CTDzPr|*!bWH^M~u#f$9NPkx45#3oZ zE^2D6C~PWi)l9;1=_;FvB-|}=bzDhaZo8beM*|1uyLTr9|75=Zu~FpRrzy!wd&Ng( zfJRUapSX0ptvZDMhww_;dFv8QR!j-N9krF*LH&(_M%xVUIIn2tGL2>?K(VQ2C8@W% zy|v8$e{9@eiE&g#KkA7J^TzN2SpsmSq4>ug*;Sz<4L)6oN)lm@?|jx7mP_8Ykc|#f zyjCQn5R04ujW+UEF{GLXw|)~Gqf9lIH>2z(;U5_VUD{=m+K1?2;=@2=f&NCdeqY03 zc@i;mh1uCktOJr9anPykFECwPnwRNP=>N>W%9dGQrHquO2q*$7{|09;1z)S1kFYKb zT4-J{;@a>jcN~BVJHwZ~a^mIE5Fu3Du<615dMMeKFmOhbBc}nz3`vx9jA#jL)+P0td zEW?evO-$RtN;(!Xg@^I;g!9ru82<9bkP8o;bKq_0&vxOaWIqfPstV^u@-2gq)I4YoumPqHvdHav2_C ztRTRE@+X;4*1T9q%tB+F?UV7UjS@T0>E^rP{tvaOQC!}4_DR4m<~{cs1MrD$2g6Ce zu>uv-(Mfs;7$h$B2xm=5XPyP=p6kBaO>hiov(G% zZRi6gy@zab+^I>BpCNJ!tn{m*;7SnTRB;dC9=9(jw1P0Am~!;!@v6|V)$Emd@sPWG z>TUNdIf`;SZWg}XN$Hptet`ZOt1-4lg#oPdaj)q{51bXLaEx?~ zbb7&zNg0RIcIhHjW%fWN%^HTeNruags=c~=5xT+)&`D3hn@op#-1xCDlb>odS(H}u zIbNePhSZ5e614nM#rZKA62E=?wP7jZ9ZCq6va%toCBX&sV4wM+weU6QVM zbGTafOVNYwBS-L!%B>8Lar~tGh3R})jsiWAYP_m;Zh(u(7*Za~P}%bn`6IA}c$0Jj zOfQHj4CqBRXwJB%`f48wxUGUBlHt0jZQG4Pn=6)Vee0BgmSucnzGU=PcA0Pf!BIL7 zF5lRr)RC18UwN7qi*Q{b78HCM74A%*gqxtuB^m2QwX;_B3iHW#NR`4_ihJFdW!;bQ zq0*cWzP57$GS$A{)?xh5kOV6xO75l(Qup8=#x`epjxmElJdfmJOd>@?|P)UO{Mob+wkSe8)jctjdhbHSgDk4(k|(a(_!i8 z)VoUUNTD6zxQ6CirmSljF#0L}Ptg7K7YHc6#XZNLcOEw`%Ex`_{WuJ+G@jehs96<# z#jNaD2+Vh+KC`qPzfbeA1=JbL2mzY8SMgt7IydShtjv|$z@B2XharqP%6aOm*71(B zx8eWszBa>3p)ViMCwCk-^iI?ZSylh!)rpb2)k44 zVZ3xJ@2+qCoRT)Np7Puk+6VZWXw_GG)^gMvt*7uol3#w~^n3wp4DGYippc*;W^IjX z1O`20gk%upMR{;V8nm$)^eg;cIk}lP#cw+Lm94$n3(P=j;KUyqd~MapTih%~Vv$lO z;Y`5#Uq6p>_y1&RzZ!yCT#as{^5!LK4_E9E8Pvami--hfAl3@ zKa^Q6C%nt#EDzq*PNaI?qTm+=R_oAH{0o)!MW64a{_}LIjkQRW7 zPVdsR3=m&ixPV&}Uhja^M%^^$8@kOXVG$1zH>cGc^f@3_-mj^m-8Qyin~?;#;UMbo zIamDc`o{|8ycsbg`Lx~JW)n9mX0izj%=dB$8>Ye}zY3d0a7kA++zm^2MxCFcM=i*E z2G)a`8w}&rFq;*Tx^W7roRF-mfNz;He4d;8lJ+ux)~)C2d&Xao>*@`%RAEGDApY#_ zxtD}fo;b(Q+;OUYR%_JCiwKd2z=97u|- z6jvB(-DP3wD_W!u9J&X3hF2qRI@(nEIz(MA9w8>vaNj!$DcQzD_GOX3^SmKHHKs;z zhuFEyXAWzB8`&>I(_spx?`P4H!Ac88H+38bnJ7PaxdaP|0!Sp>w0Ycd3a8;M^eg!??z-aCz0A_z&*Xr;3BL z2wjYd_KQLw)E;RntdDg3A+jG9rIqokkc20u^+4_WVew_nP2W=RBHB6X zPDppFgdgF0>1y&MbmBTXC-t*XND76!wYMPrX5Nn}?70mB+dxOqp+;E-H zDy2CH)lHRLX;rQt6pthr!O51_2MNcmmG&H9%4-Gwd`dE-w1o9QD`9vw?9`plysGV* z@Si(H^XN<=rc)#D5}JmAK7#I^PKX4()x&l!4H&BT=YgEgxh>{2pqFdb0CW6w%t?sy$c|;R5Vs?StRo-Ez>q0lSO;1@1>5 z^ZD%~jggX)+}4SIBhMKvPi_d_>fmNLx@Hzw!FEp;AnX_X!l{DUu7DrYuZ^eCw$CBI zgZE#Z*;x%ub8BcPxQ{m0^P;qVwSRY)CH0x4JVxDtiNL-HnuB z)pW9ybluQvz3!e7qOUh4lcPwDPl#gGu=AqJOoVt{Q(+huI z^?=Ae7Mh0UHrKODY{H%NY_(dinN#FcccRlyK>01M2|kmplHc=6@Jveq=k(2bsqLO& z$W42L2#ux^uGr=}nN(tl?F@2GZd$5cj&eg+5T{V2xB~5kG)qj~1TIPcwcgVL){(|j08;YvlnwNAqi4$Rt zgs1M%9eoe1X;c}LIY)p#c`#&jW};TtqXH?HP2B(F+05*cS49*(8GQ-(*#RXA78Aln zcMOv_uH_}5Jos7gR(nvfi`AE)`h&BhubuuyLzS3?KKA^~EqJfYJq7_+9~OL0>#sdq zSdT1bkU;0yyOGc7SiF#n-+(nafAi;l}EPY5!QJ5x(4MBvpW!HLk{E@O zf!Bc4pb)-=KIb1+Pk#yVG@eUZ0h{<&2_6Ekg*POpM~(bM>YdT~`!l5lc*wgS|eppAy_@RMCL%FTYxQPI8G!FUS7a#sda zYNY-PIxn(5B|IiN;|i1sfVOiQRhtzRBOVwzYKBA&@1X4FmJ&ZINSxcQzdDARL3+oK z^_%3a#Rr2<>yF^J!f+d7rB?4^!$r>gfcXxvs1pnkwt7huBBy6hBa+Cb?xg=2U(3GE z4Evj1^dW?8#TmiRQ(u~{Vzk1GW^C$O>A@zpWSfUqBEQga47xgv+HuftRUWeVgnggm zH&s1jV+f0f2`lGUr9JP-g@3)D_c(J60mvj8A8#;iN?#4C*HmzYrimj0FJ(Qc=7QL< zTM-UDq5~TtnaS!<866=c`BwU8 z)6T0qP*VnN*h6HC;&&$wAJAJ6&uMp8uU5JUr^ns#(=N8TYn_{AyqOaQH`u~LKxZW4 zh+Vv&OA=-K-*)*2yqw%8Ef9an=Hx@4!t(hM952h3w7Qxvg*c z1M&zrsPBA!c{aASuDA2sz^oW1~TV7iD5tOF!YGUwKyOx$n+CV#~tzvux`XJk^Bm|?C1*(a-FoemCa0SvKNI_%7lW+{j#fcZzUm2SI&V9ji>hxRjfVS&fQFwh;Kh^?k#; zAi~4BcbasE(#tGsMp>`1*4)Mw7JKB?p9Vod@ozSk$yz>Kc+aKrSBD61~j+)2e-bJCpfJmiwIwGWbFrHZO(v3yxDc^Ha=eV}dP)s&D`hUiMi6j67g#&iY85DDasxaHW(jvd5Z` zf%(`sD(0Z0n_6e{g~R#T^#dWXiLp?pnOgNPJi^m*$nn@g3;c$LE0YtMqEG7;2f>OT zVz`t-Xcci$abQC@#xp}wuDn7AU@{uMH;s)--M2gQN>;kRv*rD@)qYARItnasID>Wna~)B zRh`i6i9JAUEA6m!Nk8}9y%9%BF(!+};(M)w1wCImszuU2`$2N7R-uddDs^m@nY7El zxY)nKR>92M)jKe9p*=$TGhG2_;g({1?7Z&^^?t?(86Ty?C84wplB380MF%T_y$)Jn zH(e%#LCr5rBdq=Kv=DX?((tF)mI<>3c6-{+PB!YiL!MExfn%Z9t1{_Mh`=F&^Bw*I zCwQKB2(-`s9rE|kq{iSS#oU?eX*t{+ZUD88Er#|J7RIlG9ET1S?q>g(@sr2{>fdvc ztVf$J5N`JY zIznOp#ker&$;_~BI6LzhKkSm90iXO2t5MIPIZZ6!FLybqL{u}W@BGWy2o7cLwaK`w z=piAh+7<8O7mrr0lr|Xth+a%Kd?auH_hAoF!-g}lNPD^@zZs=Y_$wz^zBj-fzSQpu zQS%q!1&(1Bgw0l^2I;0siSr8g77jZ`BGMus*$P!24HMn>_@CO3ilHD2KX`M zUU;>uaY+VUH!QGu7O@ypEBsE1!V$Y!`b)w4^Xa`3?^)Bmgeia3>o$|ormY@|(AG7J zvh)9;h3$_&6zW#E3tg3bYoNnFo>a3PC~5Xfb~yjOA?e}#?~OzZkq%NHw*H8DT$fvA zUR->v;Vd4fY&D9Vh{N5`T|uN74p^`(fUz=Tg=rHbDCh`&o1~==2EGv71N!9 zxfBY{I3sd_T?y;gJnItLnxeiSCb}mk2ImW6sk|9TK zwm&g#^jJ-IS2DGHZ$#QN_ut|7txPTeWnOc-PSGXtNw)&8a$BpOcm( zPWbRO&8SM^2Ztz4NaFmbkYQzeIn(4`OM7lvjte2Q=C z-Vy@%WJOhCjbj@jplD$zQi0aim!X^uRii(UInZO|d=9SBfX)r97jLOC{bPO-^wjhN zye0N8E0)XU=dr(m1%eh54~U})?s?7DCj)W7K~y2WTmeA+=&^#|jPBnj?EFe<%U3@X z=2!yE)6=DN-xy!QSN%D*u(JZy74j)c?qW?ew(DP`FCZnO1MFA)7sRnAqU%&w(kOqZ zV6gO*gnizid~<=_Ym#@~1KlMp1>$|@{=8-*pCjuF*v>>+9i-W~fuHDiH(au~Z#1j$ z3E*kqQm<-7m3bW-hCH1)0y;0I9D~=^v{jpbl8A;f*u!{XJEbS`6BJ*Gm!7veb3#&z zrL;qrR};bSV3i^>* zWV)N)-Tp&tP!WxuXQz=Sj(s)ERZwbZLD)}uz{^GL^Ud4F&JdI##(mYK{G^NlMjnG` zP`zOJ6a!d_My0b-0R!@t{_d+V%iKWskfR{N-)w!lxVfZ__`c74i!(*!@vPI-fv$f3 zH45Q3e8;SMgC?n}jw(wDUS5X{DI*3ab*8|!L!SkMSSxK~8x!YKFjX1OPzQJuYMWQ# zy-q3nYG58r8>6uc;_KKh<$am&H*$)2CpyEpc%O8j5+O@VD29ftLpD);@$;f!<_FPN ziUz%A>)~(g(lS*F3|VK@XJmNlm$K)f{#1qzWl!vzxc3;uH7+-g_1?b{%KixH^TJM7_z6JL(@Go>l*_I z+(*5}*hl}laHv$Z65I&4-KB$qR~j#b8=ly5V<+hL;na9lZrhO&Z`;SfD(GDYJukT} zpH%-y-uhIr6Z^r>xoapN_|fC<`rqq!{#NPwh1^z}l+raj8ld_;3T4a-8F6b6ai=N= ztG*&71H$ggdI{{Q-$zAjL@EE;YDjJJ17Z`-iwd_Z*T;BbcBXBim#A+JXA^Yl+N2`}hTZd|@@7?_~ zs=eDBIB~V-o_BiPW^G>E(=$!*rRI^CM+?417`E;*p2PnwaPe!ZFQ?r^-2^_V zff0|!JX5uR|Hn~I32sX>KXf*ooEVr$kiqwGi(BZ28r$mI=>i zgkZk6eknm@6yA zP6p1v$zF%@oMvtw1nAeYH()J4+mXxtrt=wD@xCVxNO&P1f6WXUKQnf-@SDQaV?}+z zZz7F4rF-hD5jSH^sr4zaWqGRgfMzc>gn_q}Bb|5+lHWl-^r+unv!wGwa9cY~sIwKa3blC?& z(x_PBFyOpR)TE1DQ!ao)>H`|r%PH%>wDeFZed_|r4LmGkm=XM(FlrO#PE>u!LjG-N zCXYt@n6hMd>T1Ff%AdoQGTLX};3JDCW~ZeYn0bsJ=?i2{scF!6L31zao05B9bC9I~ zi%$4($oRyWewjdzf&?gI4hBE7$;w+7)LF9`j$MWFfUKA+6%!jxFtK4@#?Pu#ax(Hm zWYyz80NSNe&8JR2&vJ_NhBAbhgegCb+(@%PV((dP#J2{Eu}TPM0B!AB*ckl)*-WN`E#Pyf2Zh3FZJaXg>PCf#9II4W4p>KNJgNqOd7 z)vet}jDhA|EmISvBJ$ju#P$rC4n0MkxT@@;epl{XkRc`>N5P@rI@E^@HriM2P_FLm zNhiIu4|K@K*xaLe!U*+o#XVEr))ZYVw+;nH?(h}4>M$=x-_1JBp8^)IrS0PXE z?y`(=%8{fcGi!)@@elHz8g|#$tAwNhQL8+TI@4$chn`LIz~@Ek$DLavmZ{!YYYuqP z*~RbjtPM#m=5z6V@%LdJabW{yW=!#B?D|N0L0+ObzrOK%P8(!mlwM<*m2^nUTiE%8=M@Zkv`4a@{ z?pgmEc%C%d?C#0F4ZUG;h9afkmFw$uIzNz|nd*r$mw8pgac=ZO{@!|+rZZyZ7$(u+ z6F#AmgYJsPGjsOH3?XXrP$>Ps7&3fxe^AP9_nYRa1i8R6y8^O=p3j@+oESNvbMgi7 z{@rakClPMIl}NRxz<*$H@+F46)*qlfU{)Ffz?z#}!k7d4m6@IwQgl(vWwcUb$+SeS zg~5WKbu2m7Sl8na@T=`{J^rOZen4`P+vzXeOi4wm+MvHv@6856>@Ri7&qQYg@g%j1 z^O2PB$PZn%FGyujVt7Y$hyj*U(nu-Na>TC+J_mIf%8;2&(;*q)t@b ziJ2%JH)tvtTwF>uHc)2ym>%~&UH1#IDCaGFA1?1D>aW&J;9?@co2{F0;h*A-NTJdM z3(yuu<+BGmJ-V2U@CV^eRl7{Ti$Xa+QzdT=BbG6rX@@ePC*%8=KF!|Yz3oSr&6*CY zYH=BsTZmv5eyUI#=+AdB)+SK^m6AA|b|=0{uwGF!^=6kd(}i#$-8tNJ%il<^y=-5~ zpNJ>}X@ZUFLfa<%#*{(BXXR?w>65$Q{K?4~0HCjN7MtV8tr94btiz@)G(P5q%5zF* ziH;+VGkGx`CMj> ze`e>2h0bt_o|BZ@#jAxq*^{C!eYG01p!^BZZ`$D#*D%%fC7~Dj2=(=tjczUYf5Ur& zloteI%qKnv&gFEF+$_A)FzNN`e}KC&TlD{fT(q#qFAIhyuJPjqmz&wjs>Iu#jWKy7 zSAaOTvc^9)u|Dr(uK2Vuwa?)LNvyi%9TM`FXMWE2-h+{Gha!h8;^BQd(_P-@*s{(E zY3~$r#C#o)@`pJj;n3FlUN3m(~JLN1VPuo6wnpJS132u(c8DjJOSa z$qi%5_txf^W3axphbDXHwc@L6!&MO|OkQHJI`|Pw0bw2bisf${Vo7)5hE~BoNP0R0 z0T~zT8^@A8sxKK|VcaXFo|cQ~3KtB^Fr$9Kx`^zDHOEWeGnN_y+Wp*(pX;fD5jFL3 zMx)CZ_0HI&@x&-wmgRME25!dr|BXL#z*f76LA5%2_$K8r6YfC6OrG3#46YKiS)7_{ zIF$Erc2S_e(oK!w$4rlXnUlnSW2sPOxEi8S4c3?VMu2G2I7mrCqLw zd88DoZ_Bk)EAOU{e@7II$7mwRz2tCS!T2rtjhsGCf%;AAP8)_<{X+Mf%$d4{GV;Q4 z3hJFQ?KdM5=P3a}FiX5SG>Obzo~pb9L24a+kJ1;&YJrq{VZ56hRy2t~H(u?--^`z` zg@5sZ>u+yXro?$0`vQo%Qur=15%Z_12-4o5=u^CO8PFr1$fi`eU6N3>cOVQ`U%PN5 z%}f^E4*w4byp?)F)CXF_T?)U>b%)+JbO64fZaCbhJ@%W$Y8)Pd3MOat4$5KKWr_mS zQWh5^cT(rvW4+tFJ50HIpzE0VCqfc)C%|rp3&N(FccGbfskac3vOexRCRCYYU&4uh zRqN<6-#R~X4&HVwE^Fwlox1t8%v1>wE{E(80G8sz{I_bplvj;m6@N8Vjs<4!bp9DM zExQWKD>_2W*qIS1XG~)XuDQ1m^^Nvn&VI9t;X>_gu5FS3{jWzxl;FtoQ~@k3ObV^v zrMWE0C;U9}$Sigf&a2MpX601pKuG9~ghJ3&9-G$XU*}x^N75qVVcvEP*go~YD0ra9 z8wq2`o&a|`uKKHVk#z?p$GC*}CnZ(gbSpnXvWYr`&td%ibq<4R<}I!#as9}DDF&u; zeYh?~Z5bVfyJiP&nV;zYi2B0@vz$f(qun(K5B$SGCjS2Dt%#nYKd8SYUGxP#f-d#^ z@m|uI=fn-xS7Y&elU3gdegVCArD|*!c(e6bQc&L@p9I-5dv{f2WM)i zwlbP4{rlp`P#v*^r)uf~fH`8hI#Y9viGP)xSZ%5c!YvCrAUnpKkNXX5R$el;kDaX_ zH2o)&g!mxV5$Ju^&A!-L(+bvUcaGYye_Rl0Ehv3~R0POZ7AYm2>Oi z^uNuYQAtGYS?l)=DMyv3^}DmF)~(*BVVtRvspO0Gk_aA_dDM0*{B#k(k5H5vZP=w+ zNL+}xQQird$|Og-+m35Mm`a-cgbFn0NZ7-Z0ylw=diof0vXuv??T&mw_o>r|Hvv6n zEyOC_N3Py{(q=28z@sL5CtdB#LM$%N_KDZ4>30lBm?mv2m-E5EuGwdXB0oYYO?I`+7KwcN>_IIyjpu zUnB2PnycdOq#^j=1i zWoJZ3-Un0BTKh7p zs%5#SQdoznCD;)ONy1^)x6YfVjqYz{H_F^IT|CZ^0djcefnDBlLLGXDH3y=8!K_7K+!Z zL)97JbRE~J9`BDA=`^y1{E1wnlk$YkHQvyti{or(B3Rwz=MxZ(P|4NCqk&hG=}QMw{VfSN)U6_Ii(} zm}D1l+cC4!_k?R=ltkRAWS7|{gr_z38%4NU&|D~AE*1+5mtb77bM>M4_zV9&0*oWZ zx;k|2@IKlN_=BRq0WTzf6g`bzIcn5hNao-4l;K=cK$tb+-?hgSPyF-hZ-MwaiCerq zu0p0UzsC;-joRG&WOJPNyLXX_Vm|#0%yu?zv3XJ+j6T%0wEh=8eXHYGU#N)Hbga7x z)^qCgV&q)iMaq^&K|ASALQ2lc>Ra4&vll~|#B!h>23~ib51~K{!tir#f>Y5*- zU(>%1StSeNow7Ji$=|THb0e66*O6>X$uT1n{ecW*j{3*;kc>ku%b8iwQu>egMt!+~ zjk_MqA)cv|eRn8iPF0#^^pp=)`zC0%jjp*C{Y6%)N|pJkiBo3GZbA`gIDe?JuEi|~ z#^tzEJ{mt_7yE^X!ie(?d+`hPY~=-IwRXPV7r(kH-@A0kMxF#hKm}_(PSg(8kr9-9 z#(F! z*@x0N%GV+2{BRCll|C4(%dl-CwmO$5MdL2`&rWoiJ*8(uBlD=ANAT<>1XRdDgVZn^aabwbz_LOUp zW#0|n##9@E`kwbj`@qSRpo(#(jV=!>AXL&T;zyq3A!*NTorz5TwQOY4BF`&poFCyo zGrvC*%Z9l~6U6=d0aHGMpFCK4dM zs*S1MV9nOYAMkddvTz#R*YEVx);Ctxtj^XbYM03gA@sg@Qz8rLsc@MXp7A0R>(JK< zzVSAMNQV9Qcxo^BgJi>2djqFu*!-rw!T>HMR7P4B!HxjqwGUzMS=$0H)>o3`W%|wB zN=pOA$sDT-46OIhQHW)#$t=ZpQMT=E*!iH&yi$~$Qx<9*GLKi#)O3tgoG$$~SY)_L zOqE#_)FdgwJUUryc@%Um`J1Elw$A@+B}+PoA-dCQJVXuWT^RcV!2++UB3X*C4U8#d zHlice+4cs#6E=kH=w#7Fa4A0LMD0c4dqyk#BdEKM#GVDtE`y#%B{7YxT9b5sUKI$AZB_EWt(R4|Upgl1H#Jz!Tvzdsb%13k;R##Ki}T5F>|W z8DTdG+w=>SdKXxKO4V#6kMNZVtsB5gY+=d$MkJt?G$DSk&bJ%4pn2aaDXJ9FEx=1; z)qYqncv?2Fh?1wd!i!(qL-B;7Vf-epOZ5W~aI)N|smlF=k#-(s`;np|z;O zd5Wnb>W3U!eYn$vD*eGOx3R78F~he~4FhK0uX+>jXB*mUh{x?$^x}khAFrCzChWj~ zs1ywpPtn_q!3sZt_oSV>q$+HoC6d4GM7!Covfd^S;t-*ovXnScnF@>G$R8?$hh1QK zhzw4P&D|{F=B8|ivCS<@&cW1+`cWrwKHPBGgl-75=QK>L)#o-5M8u^qS7<(P2>jCJ zEg|t|S(tkRQ=B5VXwlZ6i4Gu7?w`Pwh85Ow9B03NrRtrk!{x(pgtSPX(H$)++!+?m zh1`ed!-daJD0bSKHCJZoZ3Tlwz=pM#j9wR)`Sy`^6Ft93H@ z3p|_t0Q=2XOR4$>i_bu}#U>GzzHtr4p;Ul5$y%d37!o573iK31l-uI|r)0_B%9E72 z(oQJ|&8+uUHP~qh;u^t~oZ^tr%#NL^PzcpbrFS*J+jn+II)^_}wLtVDT_`OWCP_$T zBF%=EiIV0q^DsPql^xxek6u64Fk(zLb0r3?wYPjf$G^2L^{(rOEaPO0Q1y|eISCJJp+)tg zA@cayiQ$%3ZB0Tho;;O%?zw*zrlI;7M>2_0 zUE+%hc#vn(&F1^$&$Ag+j*eXuMYZiN(sjftW&J)=J?$AuK*zN}aH$u{K%J?l=v7YQg^*A9^zJ*xyZVSI({J^~-0wZ8Ywk>7PPVT>GY}B8j8kZzJd^&i_0J+A zQT0nbc9|Q6@tDh!_UhFz>=N{~vE6Zn>M{Ou4NoGAi_njmf5WXwNl6SnVE%(wq%;1) zL=e6cuE16$-$71z$usp^yKI)Lc%a?Owo=!zgnL-jpMn}0caqO%10r`U0sP&E-_J=%o#7?7Q=BoQh!x^oZI>YSEfHfhZVbgu!&oggQeQD z=u3oZe=6a36-tf}JRbm)s*4qGXjh1nhlcTJQDP;OKqWF!}CyIbetlo?DOgUsM zfDM?I=~RdbuP5VYr3uoL*ua!~@+u`lRUU%MKPuj(bNr@rzmV{DEd4;WAU36eWecXWfvgb%f!aAOhTli~}8 z;lkMOx`_$-BmRo znv1`|Yam@@O)BP!6iFGgPIMpo@{!Y$JFy(?JlW?F`q7X`N`L3n@*d{(tj}TZlg*h} zSppx>-?$c#bfq8OJ!^s4^iq3|KZx#PDhF$vu-6QZrbK(3QZ+&{bu%BL9uPm;e!vBg zd->1|tqXR!_Eszfx=)y$CEXh7GO{V?C+!J?G2r%$Q-UjPaenvA)shK20g6jV= z{#~9WyUhIn68|lqBiv$LZ>~3lp&Yq)h-o$%!SKu|_$X37!^p)m`cb7`{a-73qbABa z^5K{=)d1oGq7F+kplyA>B7ESlh+pH;t4%JE*=91Gm$9lfE&qe>D9O^G>fkk;(yWo4 z9nWHe_lp$>TSTK%Lz+0kr~6IFn{I@;mDlU%TK*6=FE508g$`p4B~Uv_ zxwc)=4$PEV$wk*`WY&_Y4?UGTP<|hqgznfl>yi;?^DBLZ#|=)trKa1P#J6#;vN~fw z4kuomo@qFEilqD@{fT&+{Jo0AjK!&S+n9qi4ml4G;lXgU@?gMG z#HZ4Ha+NKgS@j>S)3G?8j~vXCj;UWugT?9Q6NKFb4-K`^kaxBvo?FRVOhxh0&ju4x z&c?~5FCla4qb*HSdb`cBMzD7aR@Svs6&-Zm&o;3%=m;KF7GiR*uPb*%?MUqB-|KHN z;$R*_lT9Z1&u(z=1u2`3TSg=~Z*#xF2z8)cf{8>86_vwAkM`Cg5V)!Ii0C8iqobCn zAj-0=->xvj5fdN*x##f!XD(shY&Rn@4S10ALgqB`J9UOQ=~Yk<_VJK6rr+1~c&~!x zI0u_H)PHCMtl)i2;3li?a$TZ!Sm&e`6X%&;(H2;4reze_p~$-!_AK662j5YKgUDd5 z*ZDZVMLLIj)4XMeTo=uXUUm^!)XN^inYPrXtajovmiCzGku_d;R+~+owe;i;J(b#< z;={RyO{C4hZ!BQ--)Kkt$r+xQO&w0X>GmOZ!gjAxe>o&-KcPlZ2^dbE{@Am(&-)-F zQO&s>D62HqN!rt!HVd#z>BV|jU^{7Bx$c^_KsYkFb@X*($OqqvwEn7@swk00i4o5# z|A;%QM2R<xFvo~>Jr!AK)Xz5Laf(J*EL^Ak40~)*1_7XT^seslJ+sMNuNxy z#O26y>CHxe~8lt#>t>8fgMYdGJP=ZKRnratrhwMk41S9v9u({CBvIm7mwu zk)cAiu*`Feo22?Fzpd5=+a!`QtD8t?#}7sMd&N`zofxs>7rc`=R#bsLK7>t#kc@$; zUs3$Lrsbbz4{%4SdA-Q^=jd2!Mo1;_koj-S2W+hyrcJuIZf}NA`YZE7!Hl=j;~vAs zK8qj0?THqDqbuuQGTKTLAPPoi4BCO z%yiKE*%!{+4KQ~HiYMb|lcTI^G}R^KUacIdB>-|7tJ^Y8>9|!XH?V;NcFwypQVSd_ z@!e^6KRBQr;5&->?Y!H!lYLi+QnjCF!T`b9{{fFtkRM0|SwkwocO>XwITe zQekK(b=KFuhh*1tmcn8$tl)X9NQTGFakOhJ^n0O)>G$A2weG~Mq;gZEG{Pgv zNWeJI4WvCu|2-og_zNf`By7a~lBGqd7)A;~<2nu-SQUajWVds}Tj$0;E`Q5tKjzu6 z4tqs)fIGLmhXHb~Afc&3NW43Fz3PO=Iey}|tN8UFT}1DZrb0B*)(+EB4tVrgHxkbV zWree5g1&T$bD`$BoaI&ak@%z!^m^?(%wgs#@Fm^tu}U6)`S&HO>{xdseCoZE{m43w zcLP<3^8Ee_1MMQjT?$N;lrCuxZ9rjRlT1Zg#<2`f%nH@fD7)UhvP5?Z6wGYU)H)NB z<+C%qxPQ&z68m?MD&%y~_|(#?9Xm5s-yBo}mVhuZW}tNhD?Pl))P?;Y)g8M34)+s(e=On6|gR9h0X^x4(~l^cXq`w-$7J<=03 zEt5wK1P~)5H^?CJQwEEEdE$I}GWv|a@R)qCnF>31G9teccC{fWLHI3Pkydr1+zH6hZzHF z_Q=j9TvI+*ZkLWsp#}|tgX~fFu0Ykzdi97@%wOIKxZ~JdrVSC1)z@5Y4Q!?NTf~$4oI@#$P+k9d59d*7b!-Yh&<-F&$DPzpEoPf~h=v zHI)^p9Fw1@_cMLK9QD}e`3J7m`HP_>ud+1`tM@iOa30-)oo84nL4HXVCAaOor{keR zB7e1QYtI9cn}a8-z_SezXy#!Mqzjo;BoV%C7bZyO6wtAZ0c9jhQ}$ zHLuYIsk%%bNDYra3`-UNd&m&mr=hy`nzx4VE=-^lygo= zp(4o5q?ALXz5ARtbgX_7Q&LL)F|n(SvHGv*d8az)ZCQJrH$2+E@6=x(vOLZs)GY~k zU(BqqG~*d|5=&spWTtvsqz;Wp=45-`stGGZoCellk6_%pnY_*{HL|Ue|61dwZyj)C zZL#h(w{T~j%~_+Xba!$MM=T6>)#`*RWTIjX^UbVi#6C(mDP0?QFMJbkm8~*#2E=a7 z1yfVB(c*033h+idM346pwTLP=vl0E?Uu*Ldh|fg{;K_q`BRV3MCuilUY9fqBZ!=bn z7UspkrX4>KgkgmJ(RQKcv6gjH1NW>A!hhnt4R=x=o70G0rT)>}Qx>FtnQf9VcrE@p zH41Ld#En>QqM}!0(izse$^FkrZuP|KX%EfXqzE2hiSGkE{URf!&sLI(&1u1pE!MP9ntp z7e>avq=-#+Rrri2e!FSAXgN3BNoPSjSz_PBW8*9Jozf$wXNoj_ zNU-iFc(H|I#tQ)n#e+a=P5kYIw%IA*e?VoB5QqIo^L(^O^e)!Pc5CMsA60?irEr-7 zM7Q1XJSe<3dBKiqIc@%pfbPe-`Q7EHr~fy(yZRZVCo)dIi3tg_IW76k*~EVNpTnH_-D7IAi#`cyHiX%ToHd;lT?C$&i5XeV2)K5YFL7}6tXVuS26M`_CKa1j4nvNAbLnD(XQiDRG1vq1= zCca^N^ASY5MOd@6phy&?m}`#UO=uoC=I1tgf?=X$Wh#1yby@iWQU}3>*6?g_r#{Cr z2vzVir26%WK-gYV&S+Cs8mcd_oRV$Y0Yb`8*4m_5)JUc^#%S(t>VK9_{1|LLq$G($ zzmVCzf~Rqd6iA^LWKRONf8}YwGQ@<%7@R`kb!Zye$Ft(L#U~9omY;y$Go69)i`>afvpi;R@ zv3O(-O7#$n=(KO572w(;-O$+syOU4S$fUW+k4gOEiBxvx5Z-(M{KGh=(LK)Fq1;}p z8wf9W<0*lNoBUjJ$Y&kdftpjrs!zmdca|MjC9-NfvZ}y#yj&Lk6I9wbCv}f8NfX5U za;&}fX>!U&>^=Vv#>}`SqOD2H_y+`Myq{f(NFnSjctSoKGS#q@Lc!G&PE#R)TceY# zYBtIEAA#mQ`QQ`?JNPtfpW_j3cXV%XA5FK-@()!U+F2{Or5x7QV_Opa1rz3E6H008 z8e=Pz0qp!dKsNT@#Pwmyf|g$W-na!8cW&}-k$9TlSWT?Hgtp56*R{1Hr3Y}kl+qrM zMjZqSE0iB+9IauIbC9|byJedy^~3*dyr2)X|5Lmf9ks=Em9FMXXoCJAb*CD|-X2kv z9()8A!)#EQx_Q(CANKa4_gFcfRdQ<%ud2L=PBA^b=WjjLxA~NYev{ zGB9*4tws=GFoj^^nw+HyH}T*pM^)c+lhCe8k6vT_G7|%+TE-C8Tnz&I>-Y5AR<)!QryN)o+LJV1anx1bRhrWfA*13J>#+EHr#le zx!iJd(9iz@)>jO|e9{i_Z%ImJ_>Mw$XN!mN02@6KEPk3PQ2@f^ctNP}Fa(K}y~|Y< z=1U()Zf)*cenY1y8kTTMyPK)WvyXL1Sg+_3?^DUXTsO~}8B5WE+^%8c?~-yO@qNqw z*>vb3Q<`9F;*dJLEu>|GzC=H0R45AhW;($NXV^J-W}$P%H${I zFR`OyOHIyZEL5+TpV#cw^oC-lW6CoO3guHrq1;RjXj1R*;B{LuD*{JxFhWigjq(9{ zR5fq<(Dz#%;AS3m2IOBLqADztiPH;63z_?1530LrV=t055k05AW<@@A zdF2VlObQqIr)!plxOzWvEfsk)>}!q5ik+CiN!2?yJ!NZcs+0yTiG)QJtBh?4)WkD^ zc@2b4G)@)I@S}R!J@x%0sVm(_J52Z=v(PXvfMu#cXeDR6Nyaqnhe_w2$FuI$BeHs_ z%t6I2I2Qj3I?sPSZlC!Pr|RDzZ|tqW##>3PB5W9K>h8P{(=PCj$xoSA6;o0_t8>mz zwiOwOZZv<4uSg8rTRv>B5*|XnxFfN1qDV0*(Q0*bfHKB7s-;@;&nuaixO)iial=r;6>c`>31Ur>=znqZt2-H@ zXFB{^Vl2U#z6yFBc_v5`YIN6Ds_uW|y=dJ8JIE-GW&xRG(<=BFwr%*^;p?rGHaDd$ z;#7$0X$wb7GiDgK6StBsQ`d8^T5Z|QHTg9Ne8S}a_`PT=z#OASgIPvba;{@AB8T#} zy5D^d_#f>C$P;QtaKj?`7r`%>Sa#uRe=qlPONyB{RkzA@s_={IDhxfVI{^)6`ojQ7 zK2YJ)GGFhej3T_WppDJrON{>|Dott z$hPRbV+XS+r!BulWle}5Ggi1aLDf2rgMIPit>dRq(ne|)td#o6{~D!|uAF&q>um^~ zf_~r(giy;EZpQV(`06vJH)K1{@QpsVdrcug2ua~i_P&$wInA?-7fbxrSJ~l1i{h&+ z70-0U0@K&3m9{T_&_!WcwznX>pDno)$9X?oIXE#t`p;;h7rjvz0|cZm`4~dl=_=l* z1Qt+4N@6zuET#PYgYfBFBk-YnBfT;dd_L1t$fYlc?k6%{YeW1tfUlm+wqd2Pgz^3J|rLG%S89- zxus1~XJQxdWby;v&#hCc@m2*vP;xA9vFD5OX4q1X1qO`fKS$n2i$uF9(whedG!--| z+xyFu1RNMQCq5(LJg-Fd*e{*9n02CPf~ojj8A%bXwEpeEOmk?c=_EFKACB5L`koU4 z-v}EUzT+kLxSLgNQ+Emidp(*WULkJr(!)~-gDqMR5+4+%Q= zu=ZdJ#741JEH)MoUN!89{NGgMovO~xJ`oMkXt{^=H$ED4N~%a`W@ogHKurkZA@pGd zZNhC5aAo6g6_B0f)ocuzwzOv>6x^DsX%;!=7+x*$F_w-oWmOqIO8|sZGYH_ zz{DP9y*v=-$+>C`h<+?_rM?4Ia-@~kpz06!W20w8!&W4Wzo3R-G?Q7k^8m@27Na^I z>gXB592rR)0k}I6o#97E-gIA%Rg?LLkSgZ-9Kk61$L9Y_fCE5vdrH^58H$^{j>_8OKk z8#3;imX2;PC{M(HYZJ|oSjyjQ=ZLZ2cfa=Yfg*u|rb6H2L8yclzAWyn^@2Ck&-FI7 z6_H*jvhneH$V6_1qqeaJar&LBO4q$J@>cgc?my*q)eBA6UH>Xd3;SUFrLOuLI@>TQ zb|b@S6hCv6Viz?7Sr&W7x;*Gpuj{<3Uiii^wb>q`Sme3i^jh&x^E0na8^ z#{=o=;faL6syT$$B(mqv6fej_;F@?J%56QFw;YuP`sN6OX|XfUa67CwC>{l?d=X@T zqNgDZkie7w$y(mcfldVgm-99gZR?Z1*230WZi_iWxFmK|u2d|NuM2dw!}~|ipuMYQ zk~GIJnTi|y73|Nic&}N9?2^%?v;m;1(I{gN0T$Q*R3Pi4d9Kzb6N#qkL4ByZtxe@) zt&O)7_XV!fU!dy(*R#^x?y@wbp3C>)r80j+7xmvL2nxcWoR4xMrbkMiAZ8)mV(EeJr>nxgl_TyM1Y*$Qnj!)_mzwP|Tx@qC!40iQ+Mt(Xv4t zjj2;B3{euHS?r=RVCzeu4<*wOC)0e-?dmh*GtwxHdw(xqCUlEGByC-fsN|DY0% zk3TTC57*2_7!y@c#UHBO<^}=js;eMN=p_j>HntodHviKX^Z9GRR^;>QptfCFL&HTkjc0V~3`h5!!i>X!9x z6#~EA5^!2^*m!vSFV*oz$LDrr`v*iBN=U9BkME8_7v8A)QVB-x(UQx13i+@gqI@`# znzELe!;1F=p5?D1=QqOmI{sNCqL(w|p{O_POvI=o0s!m@9bT}zBx`ImAjM6p*g-`R$F1@-JQF-+;U7o6 z=KTqySrGJ8H3{}I>yc&)Z!@(B{u8LneJws!zt;Q?G7G`)G~Do?&$~f64h6vWad`oR zXcf`#->BBcRMxo2@?LTamt%#3J>q<-S4F=;55RnVkF@^tzwuhVt3D|8k;%4+zA9xu z=1GM3L6c457HjcGFI7|qykoBf8iUzI-!v*XC0DZ&AqXn2tN>sR$WhuaV8PjErQDhr zh9l$mj85nsOG9DM;_~ipyfdgZuea)%3I9Me93Zzd-5eDWn&;D;J4<3ff_>&41tG)} zW4l;Ulmbl!ttH;Au>3^S;uKlc0AgZvCUsffq=j9#f~Km|udE4EaWuTU%t25>rlHG8 zRXv2P{j)gy9Vsa%9kPu2J7FEM(7ue&1hDT#aN8?0Bs#X_M##94HWBc>_%bX^{AFMo zf@k3^W**R^+S%Pc{4u@VTn1VIyb1f_ z`jA;1(!?%c^#Zzi{J_Yt`q%U}vPM=BjR~A+`btP5d@z(@bA+i_O`s8?#Ihf=4BuPN z2aS76y$crmp7oq#@z7Yr3z0y(!^FJPbfz#gfs!+ydq=gE3&i3?*V>M;{m!A9pADR# zw88iS%LGY@7xD*>=?hGiq>hN@&1~8g8^#)7_ZtV+Fu!JFu2rN;qeReRRdMoC!oN&c z-v_3bL^xm>_oCI85~>*++OFNpxCCqCy+>f)hs)$Iw&_~eJz36P)W76l@=hYeA-%bieGjcQpfdvx==39=%bQeI2wb;X>f)SW! zd39!xD1+ZBj5AdE7bnagp;dc^y@dVl8H8z{Tz$7S>^5wC|MG^CD%1FL? z5)j}h+%4*`X2Tl%jyiHOoMDx4sUv+p0q*Go&v0qaEeOG*q(6KZ^N2Z)&n=05)?DAA zmxKx&(|ph~!3DU9kil#`b`sez7E3-&`87JpBW~8b6lK{GKTKg%j<^OzDD=A_3yiv$ zS&}9`rz9c+v_$qQw-9icd7MC_Ewl~Le}!mP#rEva?^D@Qk5rMhdFK~ zlmLIEjJba0z6;HAl?VFv2Hgb)bsJB@!$y#?WEF*4$|K7P-FgEy8s9tmmZ9VZCW1&c z3Iy+hnQf}`Atesa(6tAt3%U6zzjEi2dxH}=N+RakMgYL%$K)yayHCnd!8dabx=Mwbw*c1ofc!Kg76uP(SP!)}8LN zrn_evPaI$Ed~Mkigo@N+9QD z&K&P>_;SA_E6H%e47OC`sMccy{WkxnmQmvs=_O4LoA{L;i4DNN3^guRzj2BRG_1z| zi6$!x9m9GVf8l?mhYweSj5o?eB*Ph_*B(z=beFC-FBA3-v=SJ_43or}Q3K1Py}Kpm zwW_&F<7Mq`x?6dO_YYCiqLzt?A^wuA&5GHHVkjKAG4Ntgj1G+Yo$xuA*%T}vG$bfS z{6*%B069^P(WFG-?uI`XSr2qqF$%jKT5=agUn54hYrw5gQSM+WVf@#mPi7(~AH&an zZ*tLK(R1-2F%QcX0%~Ed2co0s&z-aNixEMRyW!r}hL{iK-eAs>zAb|y46b?B1Iw**o|N9!_M-^5q$mA zbR636r$2=T)P9Zonf@1AVGq{tghhD1gPJCcnx;kLOwg93MsSn(9iGnGoLQ0~w>5wcG&Ej2m>D$ojdXbx{{|CR4NhW;6{md$~Y?@rb z&lXEWB1NXsosAt2z|^;Dm()929!d`jp{5?g8#v!oVk_+am3qceS)-4dJUO}gbi`Y`AN&vyWdz`JBO6gl; z-oy)O-3xnNJzYZIjD23Mvc9u5ZBP&tCG2qDM%MzO`8Vr+NT+8Xyeolz5BUr9<@W-S zAxyYf_0U=IdjU{%Z?q|V59~G}YF0(G42aqEd$tFnL~77kzZ(vpq06GL*HXcY%%>PW zsB^^TRONBaeXZYD==AV!fIkU-m?3eC_JMJy#~eeHC|sFuq_8bh$qmT<)CIBwo{Jq5 zJ@LLUC(a35GyHYo_keGSJLJy&>HQaGt|lb&=Lc7n(06Op7py$vA4vAdV`>NVsHin2 z#hVohELPof+{BWa`Nip76v5UkS&Jppr8BVGf?sP(y6?X~2#()Za2pm$smHWHkL0-! z`Rq(!*C`Lp4qZB;XtpaSF$XY0Wpy!S5 z?$Ipa9}Tu&siJUBh=<{yNFBK4)u+t4HLqmNpv>SIs(495!~#}JKyw|zU%XGuC9BR6 zJ9Nq*RAVrIRe8eH>vxhnHvP5Q<7U1kEgelbY6sv#xhJrhnBsxlazmMh=SnS9X`)KQ z6P6C{mz?JXqGmuYz>7iSj^;^4!;*Ydj=%B3Y%saRbQ9WMrlAg{iR*d;$C+fW_^pv^ zfT>gJaMwt8vZy^LtBJo_YCA_?l`*LsOixSqbPg&C(0yg}Gw%SowP5-adwqYz^P^#& zb}Ml)>F01~>raGR_#X^!Od#&?XegzbGFkH_bKFa5C-L`@J|rR7c@ubq#%trXDA-M}u5JZ%n&h1R;^rc~#C!rfPrl61 zKUMk3)99QUeSo>Z=IcCss9-Ved&UH=t<=0UZ+2vh)zSC9Bpr3a#xfVaFRB%Om4^to zo5fBFVcHNzo7STE^v{31}~p16=v{8 z%}Y8CgA0F=IDp2A`ojwnYH#YFgojuotbo8KKuOWCOc-AMgCC&^EfRzE%0%o$mlrkA z9hWPHDkH=%^_U@$6oXwZoqdNY`9F%z#;@h||KkW9qDdIK8A8YDXdH)Xt?f7+Iyef+ z4&4lW={THL)@rNWAf%#6n47j8LMk;0-L0)#Nwt;kPdB^i=Cii6yLNx~+wWiaJU-WT zy|3&2d_P|)T{`m~a-DSruw$S#^wJQ;{C#Z(bph=-zrHE7ZXfq;z(&=mc6Aps7u1#g zh1?bQ7X^8XE6>N&EKNuQ=5bHD?*bMJJ*HsasGq`u!%&vvNIKw+7+mZBRU>q!gZa#Q zzJiKnn#!q@+oua&%AfF8ryQrwx3lF*4(Bl+S2cZY)NO{LJWHxV15PVu>&XPT&Pz7gvm6EWdw6_>Z z+Us#+EjyaK-sT597ty%=DbEC(bXURqQ+|hB)jy~^N10e>Mk22H|86O@Sa~T{U#Q%@ zcQDcd$G;ShPxZx?oj09Ts;|l218piN{1y19I)OtzVapn+}*eeT4`94k8xX|LyqswM?o0p6SH%5ZP_CC zQAcyWMr#(~#38uv1SO%dvmdJx6x$@_gBeMQ&WovK7UImFv>!rOqTWVow*-3mo@iYj zQ%v5>vwyKV{fzX6IMCpQ+>`K&=>+_i-<6CFWe>p+VbEKw`EgAl@EbZk=Wcb_4AuBn z<4!8-i(X+5l2fOUHW~wpxh(zyG(mqyx)JXgOij1Fi8nv$SroU*yBv^y|INvRxK&@4 zx`!KH=L}0u_)2GQ<@MOQ5u@EesNg2Aep}1~m^)<=>PjRGIaT~_ToJ-1DsNl%SV5t$ zV?M5xNk>H9?bJZZsP#;%WM7Z%74QS+S>Sq$Q7UW;)_bH@Pa~hl$?_v8Ze%|9Yr4Ds zIw_2=aIJ|>*jZz)Mv#6;SPkn-aAJngPMH6z-Z@o0-F_4L8v8$A3^3yC-ppxoxKQ_y zS8m$vH=-+1@29wMt4M%;mn*oT$yMIv)+sBo>wZuGk~f@D&DGYv4!=VVG|5g&sLVp+Yxh`KkWq|dI@jB48`L2`Ef8w1=`(N8qgFb znSBkdX7$%DmJj)izVE-&vSr3SDB2q7IUcoa6sy~d8RA@^{9x}ztXHcgry5+RyD+Vo z*>SHyq}k2F3rhv?pUi+K#V;uqCMILn`iHs&>F&xGKphV9yW5G7nxu2}0&t)hEA z!eKc>iAsO#ntUfF(z+kskASN8b`*#p@3@j8;l^p#Bh>o#^x;f?AZ>W0rrA&W6zUkW zE&d`G9JIr~p`JGArAacMA~WgNXeS*>{YIGbp~9`hMd|Lwy;UgeEeQP|^EKvk2t36Q zDSO1bVEPD;g3Y!~lmv=x5 z+UHUHDhm?tzwe0TC9Xhl`7Owi_IdKS-uJPPV| z{Feh|Ci^zW&AbQRaKBnRfUglJSbnIRqZgPt{`0LjfO8)2K+0ns0a&hnxp>Zxg0ql^ zy+J8ES(loXZVr@i$D@9&k<|!k%w%rn_*|#btGn=C*dAUh@DVeS=Qn_CC4JVdaX_EO z(1W$VoQsAE|d#Q{yhEKcS{S zM_!M0AcXVp(|l)B>JoP8XS<4S`iKsHiJTZ#hr*Jl7kMLGxafaK_G8ZQ ztRK;Ve=_P^(0PL12F^f7 zAs$&$&N)GScu)$i{b>L|FDT$NBPPi01p}zl$jn`mzod~XQ~VvPJ=OIySip_>cmIJ7 z(Yu%g1#kXyTE6UGzXX4+2X`qo&+#~BFZ!=u@@>m7KMVT=S(wvZk18IEG)(A7?3oG!?g4 zB9+}ZJe)K<O&m2pZ$C0QYP<+ z7qTDh6E8~qB)eL9JOh>A4yibl&V3$(%L4ge-g|Q2TPflX(oRWWhWu@aYWNxV3hk1h zG2Bp28)EUwfn%gqRc;8H`WAP$irSW@MCc|tA}7`LDWxjK{w_M^LwAFT^`%MqpNNNQ zc8gyZPF;FOQ?1oz$FG7@kyoHo0g)J{1M)28n`|m97#u>jAXdahI72DUoLeIOpJ z^VO*z_{t-tg;?{?+LI#MPpdLT>K`~S6W`9ZiEo!*tg!mFkIMwxl zzgWr;MD$ac?b z)a<}w(5?iX>a?bBB8__+T^xG7M)DI9hzL5#y@;DoQ}CcEuW8I_lmhj)&EU$9LkN|j8i*aa7EAy&tj-GFHKw`-Mqfq?8?qCx3f@iY+w z{BhfEGFuQHHot5>&?=yg?m#{zZ;IJ4@(Hdc!RH2pwfWIX5=3(Gv*v`~CT4ora>y>o zQKI>-C)KLsqi(!NMt)G(a;!6*U7~gL@TEX_>w4@h#f|0=aWDZ)zE4uFhkEH83U`M6>3Kl#FUhx>tj~sR zrj+qn$|nJVhM43(xCdpgRi_etM;99*isIh14)E)E_q65UX2_bt#&0cg-52T$isJ9p z$R%+D=;{_*ySlG^GFR^^cYc%niBTqRL0(R|PXLSnGk&^9n%t<_@dE9!}0m_zC{E=>YpgOfKw^`U#)pXIe>( zL3G9pd6zm(k;Llq;O|X5`em2nrHd$ZNOBY$fEYi1MSfQrl|&wERjoh#Z7`#k(E%8KWhhvq+w_spp2W zY^S&$uy}g-s~p$b8!Q4zPm-OpHE#TZT~G2AFSa`KovnY4JqK=v{bbzNyjPWGs>D?h zbBxa9E0*WZe+MZc!tMA>^TghS%kTrrT62i?4)qXiFJ)=eD{Qi(=P~#uzA3_7so#(G zXScd<(;qZUd6lh@3+<)pT=xmcGQnPto21o1v8*CadMUtB|3VKHgS9F`!Yog|jNh92 zu`EFPG--+GET@&Y(quRxE-_V)mBd=6F<(Pg4p7dJ|LAu>j8?x&$a*#&9~$VHL5?7$ zi?^%0`iApO(i_Ev?@-bz@nKC$)qbtII;s%yi$!F1WRGGmaU`Y+X6NL!6q#$0=b#Y$ zA?A6^b--}Pqffw1VZ^2q#zBq-W5X_)t zhT$JwY@JGn*{3^bc?`K*DS#;NBI5MllfsF=$)U_A90p-^F!L7%dAD}Z%B7_aJh+O* zSwHaGQe)B`DJf(~^oWmx-!+ObKmwH#Ba<8Q9f;=>gBZCmVn4+x{vPZj`)&*fuK3d< zpZhh8rg??6Kll|C%kSny&SuA(V^p^b*k50Mjq9pe#ir$u5D?ZIFcS< z<`|-9v@o5>wVDHm&_*^0%Tc!V0$^l$>L^c#i`f%>J1RxJW1xU{(c>&vc}kDA+>(x= zl>6T?8B;ZSL3>iq6@sX`_z5|tEqlf_ z5$3+EEHvdrP?pDv6cA_gs2_m=)bl(n{~2}f<@thwf$Q=s%uAum8TW{8)=S6=or8Q8 z3p&;bsN})?2J1KTp&GVPFZ`;<&Jiv{yaNg$=b5wT>vZWL()}8i{)hl+`AgsVas;^Z z2Eu*-`6wu6UG+QdF}WOndMa?rDjDdvWui4rZQzF`DBLPplxK-G+w4)M_-eTJv=}Se z`N-y%R5G+$JL_*>{dY*)eX}7xjICvRpv_oRp$kM=D?O(FmJUY4n~3yWR!oOF9JqOC z{9;eR-*pkTD{9w3p}9$9zw^)wZ(tVL2WPNj%>j0ps~MRM4U~UJ-i@zyY(jE7utJLF zXk+>s%656Be1Basw91Bh&b0((?rLUAr)C;7=ykeIR(rus-&?Q`@E7XD%b^<%}G_0k3UxoKdi-Tb39pZuv#lHkjsX#PgbA3cb>RniTz^~)#KbKMPv?*xL| zb$6iwXmL5RU6+|QR~sSUEIS)BcnJ5K=2es+Tff?JzVGm76B(-Ri|=l62`d7%p{Bjh zBcGs20ookIQV$PpHsS{Cx&K|Ez=VN`+Y;NViJ+)z<>s-x zD#JJSl+Cxe=YO{2^8C4*c9pI4;Bf>apM#$zy37+UFwFcSX zIMd#um43kXiz&q!7pyBhej+$9^n-FeQ@SkSv^Iofg<6v|=N-3j{Yz0p=Y>zvkv zEO{z%ZSLtuKCJ2?ObWi`Nq?S%>4RWDPW^Er!;8HXw@8zkyq4OQ+O$y=KMNTz4D;Cf}y&D7d8fz$4oDYo^}%p<~@xs_#4`43FH!@|t* z$P@$)-y@ajPbFXz`ycC500?GYPA?zODJmsXQ|72sQ{;)GCqZiOi6H%AW7E(fe>VG9 z_AFB$Mr#)|Lo=$?8IJ)2ygGSyMNr8rwlm;QV2(64ua5Gl*3-aB7>C56Dfi`nk4cN`ga~i%4JNyo*u%Shy zDl+??UMF>~_xeO0Tua~pedPOZnAm#g6J`eMwXKz3R$%;nNI<`+FCZ-Ue1&}Iagp-K zVy|XMT|k7qk5$MCm|FK9x*vJf=sUs)JvcKjaDw#BR!_|p-0(ZkeE@oHeV1oC64uN8 z#IJ{bg>FJ~kO^rpin#<|X)MzR=~ptCR(I=F?p(DhVjxiZl%CcicpHfqsxlgy&Y=4; zn!)tSjWN$7@9V!3`l;OC&81-!m*_Q`%Xl@xGGREGpYHg@xpA+;U_cZ1tKoEX;t%BG z%9r_(dMfdbhtJqE=itIF&ZpYXsSm==M!&NS1}0wD|2Fc~dXnAxtf|(cZdIxqeox=_ zGjWsO@RFub8TON!mGhX~P4AO+4~F!4C&f6HwUkY}{mpsC2u4PcRzg>xw=i|?)>l0E zl>70*5yR)2$0Z`Q1T5|@z6i^RlwQZL4dCLB%*1LQ_XZI|wGN0|hR0f${C&|2Lbc6N z_hyDGl&l#>HCus9jbW7FMqg3xAeVqUapRA>{BYJjWeYN?YY@;$)Q#QKGsT!G!~Kkf zkfcL?Lfv!%4&^{~+QSJdGj+aSE=QItUd-^ob!S#H*M^*VQkd!~Lif6qUKae7nOSwPcjG7DemD8uEole`Z zq5MUaEBXHNsR(YN^^r+P{brL%gSCB^?c~68a75)>rii3i2BYFH^}X~=g^acqK%KZC z`6sIq&XDvh(4-e#;jswR~pLD}+1gF|%8De{GYvN+EH*w)u4Ee4m0-4?< zYHju$f7Sz?fg_0^O(Ppb{=Td+a;0O?!Tk0y5z-y7sQ$n$kd@f3A-@M5i+Y3cWULrd ziA(!r6cejtptih#xlq@Dg@tlXGlb$}Wx0dIyng1W1BC4g~m)bA<-Bo@jLX5eMREp&4)#WP>?Z;6XZRT*RCr7pFL)3XtoO z{8^m!H}NI^4Tu{33QiL*iP?r~X-n`iAGT=3dlH%O<5o$2mRGA8HgGc9NA>{g0NRdp z6GW1?LYvX`D8Aw9F5!dSvpn_H*0x<8t7qXZsb>Np=7141URaNJC@ZO2Ugv{LAwT9( zWy+`uL3ZpdzsVNy{8TU19%D}vw|=N)T7}df(VPvjK|I&f_ZqfYe~j6Qtf6uC$;5Ec zZa-EWMYuYt)ZkWw(ElcVj^926m4=O?#3apMRr|`qNBTJPteGCVpoIB$+cy+*Ebe*s zRvwO1?7A$&zQ=hFbGj0l;%($$_UV}3T=PP~o+`L|Ud%@1DU#ut-*r}^zq!UAWsUc! z6BuOf;F?Nlyt$lo0KA&{vvr8Q4d+N+%=MW46Nn)6?m_b0j5 zy}3ec`?Os-3tOPUjl@v6q%8Rs&+W5Ycn@t2dlfx{ zQ1({x@9-bWj5L{DKP|!&JxNIYHTHGp6?KE?6iXzZ$QBbwb~#d4HI6r#JL6Y-ixx5Z zlNAJs2pY)0)qwUvUz%x^8pg5pk^3#*SzAMsb6l;KuE_$dSNt`!pS7##l9ay!4~lhS!q+wU8KTvxTYd3GRlw#s%(|CA~!h;(w;8s|G5Xyc0x&cTl^S z*A7kCN`S+<{Uh1vBIX}a+d(%{L_LI@0Muk{&+0Q2xQ+-JT~6XAtWsWqzXK3*EJ745 zpJJD!)t4f|;|?(&8=oSGq+n^6-iz)QTDV#JJL9OSlm|vGK^n@bC>B1!tl7r#C!UqQ z{t?BJZdGMgVkvVg6nT?6o)5x3e?Ee+#pK8xRg1x*(wS)4@oL6l_Mrk68F%fQsTc2~(tp?WBcM>0K{ zgA~On!=-sfK+U{)&_t;yaguc<-h&WB4C%nslUu_I zJ7&b&yP8#^K3sRov@plZIh{FvLlz>`iAu&{&0IRgm+TxrpN4$p*2-g@Iq-i0WrkHEY2YLkR_M#m+WX;EJm}uyZuVY z=v8RF(`8R#USGPqA+PFJ_do0Uh3oOBYNgozY>!6MPQL>&vFKa&R2$8-7%6OeU2zBd zH>KfDc2AqvCt{!>UN)5O_ebSY%05oS$ToQ#O0gLuY30>4UTFECt;F`r#vR3=Fv-42 z;AzrnNy=C7$9QvvIZjeoO(u66yE4Qq;er{bk>eD7q8N;ewp#Za1HpjGU|umowLH+} z1S}T&1BdNkfMyrFWGD3QsEl|?k5+gqYdJjWzAo8&mMXx`^c~0Z47HZ82!ErSp%#DJ z_o{I-@+R#%O|u5M&b*4Voh4)At$~0$Ka`#{;QsFjxc0UDt8x5Hk?F(OHUY%ap*uE! z62Or>m!P}$|DI|t2Y&SZ!?xS=JODijAs^vfjM+u@Bkm=>42zS*`iH1i>a+acTSgQc zfIwX)Z6P#VKvzBZ|$LqREOmE{Iym5$@z%f+NLV&1H2(^=2o<9A-wvw{U^f{-~P@)Y} z_dkn$;k<<6!-DB5A8X$84KJlN6S}9n;+xd>3*#Wu)NXmw4spKpzZ@e&?}<4WU966U zLvhYHkK{;Vv1k44uzXxDE&wMjxJ&{rk7_k!IJXwI7C&`2sF4i-OGX4P} zzXk+UBXPz3X~`#8Vc5E)gy|fWY);cml73dX>RIsgXi3ynB2$0G_c_pG?0Q>CUITc_ z!UiC}Aj3T8bFZX{a1#h6e=DW%meWAV*fPhqE9yt@p;9 z-{^5UU;QBDlzrzrYn!MO|Cx;9(sYxsiGQ?*P5hQ~+At~A4@;g5G+qzG+U8(Ns4)2s zCcX^KC4xO0<)jU>A9>}_zO;u-LRr%8vjRl8?F_#a&f>YbS(T^RU-w;(ks#Zh;9!E% zmmN&`Dy(}*yQfRzX!8g>ULO2&bt|RGwKe^mW8Zc?LG7wsszgrL+ZBD00|oJ)W{(4y z2_*gEkC3j)BE_yZ73mgF~sgdao^|h>N6^=)=r9Q@bXA8N3pU+V8nR(5rjvhX_H(Xk-N?&mbqc4xayQN#0@u zg?;c%a)(iG+OB!E1)C{l%y`eB-7p%Xph{CQaUdyk(akI#IYQq{53{Wv zyT?H)X|Uy-7Q>lJ#aV+Jwp727FyHfDHIY9Z{-t(${#RE%;f72hNH_>PC;la|DD^~Q zXw}t#`-%JRyBXFRD6_|iQVvHWO%o)`=9&GBrVm602a{N0 zk|0pFT$(d^kNMnAa}{uDG!P6|`jM_lw~VjLLeK0af{KN>UMusJ?)Ow{ny5LI(s zI+-M##4y}E4hu{_kXy*j7$G&M!qd22IbT;HA9T>xVY50Zp8;w1iV%anFs!M=JEQMf zGWfIQt>O#11FdR7DY{Ts?y5+|IaKLz+Fzm*Ce(mqj(Meg4?i7}YirBydq#rR4k5s) zg{to?^Hqk#m>@^uChY>u32tAoW~&xGO1HT3OZ=9Pbz2g1CRmC^0~K`OQmFG-Jn-|R zP1&d~0;GKWrm12o;sEJKv!nYf%_2C+fFN&X(}OA@??TPjxa~-@?M68yMwGaf@9HQi zc?f-1FtxuJEW}KQ`{XpU0K%;}p=72KpUd1)d5Dza5EkLS4AY;%kh!!LtG_x)^AXx1 zm~+&A#A4)Sj_{+VXXYo<6DIOSp)5@m6&Ut=&C30#meyg-RvK`r^c^#Q;u-lkq)3HW*26ynNHSiBDUNhuZKUP+vU@ks^GE0 zu^!{EGq=dw4ZEU$Bqmm^WXJI+5s)k3FBgmB?y)1`QKZ_a4ERK{ITHwTT+I)!t5JSL zd?JPd>EdU_x-ESFui&4hR>!8$Rlpa)Pz5g#lIlOx-$JAs{fSS?yke4NONP*p7SB4S z`nNdiWwI}233b%)uMb00D8o-47_00sMWrxUX3Ovta@zY9_lr$zm?Ys{O9fIv52aMT zL6=?VHCEWvh&~Cn9Qt6>FPn{oH1edY$L8y*8Kg-=wp@xnh75e!y2*NkCPFJ#&aGZ% z2`BziwGpnvH~W;iF%OX&;!m*bp{V3ZgSG&_F|}TdOkQnUi1=U39z-qd1`NbrMVZgs z%Pqw&uYB!$Qhct4riwP`W2Au8B87hlG%8w0hE$JLQFF4#f_nrz!=b!9lL1^JweNB%^am7iL`h@n(zF7w!(=oHQ!sb zikWWk8%Ff#CV#o`WMjmf&wFYkp$$l4g!V_H_tdso;X5IG_vc4}1YTKBPjm2kJkg*K zzwNebm1kT%R?41{H`+BUT}*xI91Kj^&1A*U%q$>ipnI3hAt_eCpN~l+zp=S_54Bk=vM)Z{C6oZPdf-O><(62@05J#g< zvEy;!I5KWt;`OQ@<$knp=NT;WCgAa?1is$9DD{B&s1j}NWa4>4sNu})PV`GZ+#}27 zKpV#Mg+I^8vHKrnPb}*|zxWAr0K2CTZft$ZIDFlaedD=?ytY zLS$K0S4Z+F7_e43}#WyH}z0)!19yrX10;OOu3#4}9mSeRV`X;FG&d zdT3BE7cpq-J;@4+spBw7it`KrY_K%u$C5s)wqcuB7S=s5nJXs3@x$ifmKmq26wRI? zZ_Q@1S8}!az8%I8|M%3Es4-HUQLL$*4MqGzW! zb}{`t>3UMXiZeVNDINJlB+Ck=khYMHi(N<`E1SmePKl>9nx^|3=!1u_b@^&lY70tq zU|JwBwb^N;`-5i)nN_kYL6H--djsE`%?A-V+>N2vyvPllC%f)y3fg}kXd%tWQtE@- zfE3?fRQt{T`Wkq6OyTUIlrFM#WD(&93&ijjhZ;oQOWUoRgv*{@e>xs1v=l??qdTv< zhNYDKK1M|!rz5rD?EQY*fPQ!st`j&Z;GcCY0n?e&n^b;fu{|Sym=n!U2irksS#wV* zxw>wuAU&`(rWn;@Dsb>+!OAHLSu9nP@tA7>VHu(oGxIoL?r#+9wo$16UW5*^*>|uH zSXoQ*(?yd;i8=Vyw!Z|#HSBAH;8Z6j5Lfj?e*D^jZ94Hte#?AVQgilD) z1Xe+#_OT)6bx;@dcUD51`wrlIZ?tl8i|fo!X4x)7%FGq>5%Z#qoqmS}6w`ZAS5l7< zpO7xuJ3YMJ`4#6jKu*>a$Mc-CWiwRm4PkfvUdhAP=`z}9*!X1QMj=A=J$`A>pGst{ zFLV_+A_D@=if+SR6#S{p^>^iL%W)|z#(Zhj*A+B_r&@cio?HfhhL`e;xZ_}4GbJfx zW|wA>2~`K;&4(nFY3kfTBPBJ?)y{-h@Tw~HBSCrwafj*zzr`yyJ4?1_DHJyApz_wqU$Q0FUea!#T}sY%BS2eiSK zcHIALj|1&}a>4_Yxbx8=asEk+k^qd9$O8)TH3mju6StaonA}D7fRAer+Z}T(MVv;h z&27rHbJoKJ==rRVmYP|gfgxCB>S+}b|1a<&^ka30ysu3}Fsnm$8`^N>)LKhWH53AsBa!5co9+ zjr+Rr95aG)Kd6P;26w@oRLmS6jiB5GKS4o^_fUQIN7y|yY9!FyJ{k|F0w z`{3?|NX)@Zjz}`}YFH~UtRa{SqpOk8h z5_&5BB+WAx55>s}n30G=aso-LQZSe&ZA9`jV+?QaYz6VZ5vrQUC;}^-=&iQ^r^0{J z?pCdj7g(1jKJ?!|k`wY2am4xx7oqVsvdg^%I=sf{!L1_)iPB;?$&QQr+i<91r?61JULP$@Rh_^EV@3RCh>gf-?}LM@Z1*PUiR!jk>HD5D(}4``n91_NR3mEbYfq!)S+an zQxSv|Xu|tgWqT6}{8VWc&ss0;YfF0iZFXZS{IQaWi=c{AdJnwAxbbrk?zmIL1Mq|X zZ>_%@Bg6~gAfy_pcjAO_AIex`41YA-A?&5&K^;bd6}(WVJKnHeh_6$7k=r!yb1vq* zBCh5vLEVg;NW*NUTq()qYG`(@{efKM(%5;~X2P$iCgV%Y(a8BRH#s}CV9FnYIKx4O zl>3-pjuyHMc^kj6K5@Rf9<4mi{IAzrc}OPr<4a%SJ&?Y*<+vM6Qoj5~ARx!rs{+|; zG~U>s6njb>e-aDFU1&uY$I_qPjO%8vWIUF~rHI1JimjdrWy9|0Q@#2EJBWCgF|9ZjaR6}=oDb2A5ExoCBvi-Rt z(L}SQ=Sjg=&|lmPu^K-mYJ8Lk^}MHdCNIZu=oSolt0$)#%3K(Rzh9%!HG^W#Ag`rl z2L}EBys-Oi|8VzaThq}%7HXuFyNsMpdj;J;l8^mOerV)coD=O`DBV`-xx+Y+%2Qmt zT&J3y^-83}a!I+eXNQPBvRCn{-|7g}siUiO=j%ELx(1vIUuE2)Z1WUyg-+tru%oJ9 z49n8?QJicYhI6{UNQH@!Q&naWslwP$8;j{;JDPuA2Kb4dCqOkvl3%F%h5sp*YMe+Q z#qY*wy-n)Hb7VJ@uU4L?D_1F2Fq}u+J>zWReV@L8i?n9LGliZDPW-cYUZwnN@q-oV@(+Qimt%tM<{+#_ke-`cqvzSNcPn*|RQ|$XvAZfm8 z|G+cyG0XAP^~R#+Ze`e_!1Fwe)|c9dXCF`>~!#bSLBjxR1L6I8cxfUvoM_F&|9 zW9$b&@*BY)#5}X~*M1XG{HTIkLncirYRK$5nnT6}?Hkp*+Qw+w3AsZ5-x56hvq-SP^a$(DBzjs~_d#&F{Yw@0o4`FtS-oyszPR1Ft> zQ|`vyD(nrbR2**Esq)wV&8SR$On$}32GRYVLk`3&3@bzhLzi<736}8Ut;VjM_tB#1 zw#t@Kqb86N!5qB>nnYtg9XQ`v4)bY{>yA?Dk(5U0--0W+BfyH3*EARgfz(Bgy$8z1 zHgG;03M&@MrTL!fJEHW2qj+D+;>rvvRGnrDi`mFLi2g(OPyYw(M~zqax}%`etma|pXmnz=ZXdjajFCAE-dx81ab1Fcz+EY~cZ9<+zQ^Yox+IxEG z>U;b=ZZ@UwqBkWyv_Ug*;9Ib>;9uA&{Gcc@8JX2S$CUjvF@}C0+;aW@v|G5pw{`{W8Hf|;veB;cs4uB1Z*dGeu4dyTy0LRw z-Szk`sL6F`HSXVMU})$Wd5-YO5DoJw<{|%yNitG=Zq|n-zQl(x2Q*pUiX(l2`T^As z`_Xs(57ar2alhjqjrkIK92GW_1AWc-#OSW{ge97FB>u^mf}Di_j-pSIW+z;>=*`!x z+jZlF^5CRMyMB??+oN*-5?SCdul1@do7-#07uHj9Y8dL}A@F~p&1GMl3{Cdlwu8Fm z+OBfw)SqbxBLsR*&={v2`sPF0Jo6^t9R5`TS?SHbQJ3!Ybviye61lVPWNH$D^pGve zZpL~Ws!iKviv&SVl{t_-1IxIF1?!_P;3fN%na0+7n$W}ETKdEbaVG_u{Sl@9<@P&Q zkU&>nGcVd9dAvj>uwCvQmn7O{bwBmtUQtBw|9ICtx>VZRfv zAcf?nCR9iZs2|ckd(wtH7)6M%=E)ke?=$gZ_$omTsULIPUNRj?EF~V|=CFMM$tl(S z|H5jjwA)8~2fT7tK;O#s*JziSA^3j$5@TKO>fUl;)Rg|5hk`Je$iA}AcuseX<}aA~ z@Q*j^)zhm22h0Q17=-w+aNJykjs(He*e8qwQ*oE16kbiY=88aMONr?TCG(+^rm^Mn zX!3SOl0bS zv%>JGCK4#i&YnTZTip)kpJ0R=AL&8tEy_~^0Rzk9j-%FqUbjk0isE;UbxLYmgVhG( zZ4o#jmh8wGHmnpcFvLoq_w8F|x9|+VLO(Q}MrrDb_zhtiH)-4(Pd6Lfn5$f$^Nv3a z`41}HNOxT4&I@p*b=9KeJ(XJ!rDL&-G}#_0T8T=y?__?ZE{fSQwuUv7qS{0_R2Do! zZpdGuIS!jry-18-53YZyshh+aHX7bmk&yToW@{tU72Kp$CJk;hFN2-Y|6<%AOw$L7 z_aq;Vibrf>qwAW93zhTo8KmJ?Ish(i$RTW%*0olb@#rTYH2I`l{G1p4svu97?Fs{D z(7WR{Qm=)*rj$eGDlV5z|Jar>40@=iH!3r`o4Uz%HGbFYbfOdA2XWuuTflUiIJaz7bKx zklGndNZIrFUy!JvkdfEuGPUX*2k(oAXQa252afNmDlsz=;Rr|8H6G4X(Ghen<-8#O zRN*r_;`7?-0f$mhG-cF;+uZWG$ry;?Tq4rcVC4hF86wSFs5X|+A}uzu6h#x0#u)t; zTo6CPBPZs$@tS>3_h%XQ#3)R_ANvypPe^ndIZ6r_tc;dQBZd60SUySECcIGpY_9cD z#=kjd(J7&hCI@B&>tj#c`Ap!RV-J|w?4%-k~1R}uO`Bct#8_eMgRPW zB^k+|hz_ZLWxGH`i{3H6p?K9-$1`hr72~niGk}W%6Kh>HXeb=3NxOX< z^$OERJ_7txcX{rBXu2q^A>7~x4~Hk~DO|tA{?2EP>F76zEbljhrOt17M@a{j>&cDw zDI|~x*WvVIXaS>;x{?F6xyDWA*Y|V;qi^H6{uNbY)eQwJnOm(lypx^hTYko72eA~C zUVe%r>x##}B}0a=vPcqc5Cs1|aTk2P{3d!lEZ@oBoikD^{e-+Li?g*%^O~yJUkav& zCBgr;KXZQ9VFM6z7&=5*hS=tz}e^1iaY4d8#J&depC8b<9_4TBq=_f&k-^TcFQLnCM*E*+On90?Aa@(m}ID1QX z6=F^ubR=HY@+UaTzl3Ab`Td2Yjk94Hy@^zCcmeMzotd{hL3Y_y2D08)WlN$wX}!Z1jumM5P3pVy zu60L4j?wOcdoov5X6*_LS%cmriy~hK*46HpyM(<+TxIQhn=Z!7@NhFWaj?}uF@lwl z1SW4?LY-05oHhrk;9HO3!|ZD#4?i{)@{e6T@IfsjA>qGKhLNWzXw9{jLoOfE74 z$}@ABhtuu@#5LASV^D4rTK{K5*D);ciq7RH5hQ5U{ z40Sbvhp9K%`x!KAvT|L_ryIE;{g^m(!(Qdj%^U402F2$0RcmcQQ>Plj=rSI&4WV#v zo5op=k%=Uc!DhPK%PHq8WR@I-(4=HR^Ws9`JwId(f;5tbiW3 za|*-09NOHic;1bc5qex#4Kxe|qE1WR;W_$D`~|UyC+g<{=GS?W@zWTEMZsygO0f0U z#}1B4CFANE1gTK`_kfHnX(}2H9GF`mu?6srzqN$)?>3(?Of9crh^a1de?*^xtrefB zDkj@Ks#2~YAEe9>Z8TT-ANTm5fp_Sv##GZ3fPd#kQ!u2ID05}z6Nr1?k03*8bFG=2xpaUzI=wj8ybqJMEu zO7`Rx=j@g5Ck30ABX^YY(liyUvAq$bXD@=q}MV-Kv>TaZq+%L z(Nn3r6Cmn>p+g#9_=|PO&xe$;%{u93R_`LEQ?G3tJ$;_b&_6P6vHX#GRDVnq@#yKl z@-Nj7xs{L~ftC@cs$nL%Ds@sm_dpz$iRqO4BDYyqwM^dRW$CZQ>|j9L*D@AEWFz!9 zTywIj$nZd=Q4th4L&FeJ9!IsG$iZqerxdz@=@_!c@4hXbXnG*o?+pus;T%%tBGR*O}ph);UU9xYVLGo4f zlJT{@)08Fq!s_(9;V@pbY&0UV)}MoaS`&%j*1+^_sz2o66coAxt8bbR(EpF2GY?BD zd*k@8ams0v8Y?OnrmU|iVOpuFG^fSNh0G~yEOAF<8?!RCB*)Aal`$(UO*AVv zgvobltldn!TZ8o zRlA7@Zx?FeQka+Jq*J!@JkD@h=Tnmz_srN}U;n4`ZzDMk9SOWvS3|Gb$RqoDxU%(c z{oXUqTUGb&78=Z->B^Io*pBD)wxodLrKhCFN8S)OlD~1{2tg!7@E@TCj0D|QOIww^ zq9xG4mri*%f9I(5=pO&qOCPIEMXb_5^=5qq{3bR?{hK;PK<=@fN*mUW@K38Awxma& zQh&cd;%F<5$6p^iJ^Ml8i)alMZ1CZ;UnPZx{9t7oPF&JCB!WTd#8Hu{p)SpoPE&JEz+wMKTO{A_}b4-o;a#4Aumwl)p|o_@THLDM$~H%#H8 zgsPTmK}g+Gk8rd#wNG|7M|edHC~kirhcvpWELs7%S%fQwzlEQ22GdW4-p(r&Wzkq! zUI5z*pHDc0J_F3q*}IWL!OK1W^^Q!q_ZZ8pn~+4_!qtPsg|ZeGjpHQE#9mo&C}y8| zxy~;1vAWUNsd}AO%-+U!<8&sE5k5mF?RANQR1}KDq!afN0}2n3Fh1s!U=>|Bk<@5N zL9{~IaM(be&(uQS+x4?KkNBS(2bQW-4S09sMcDt)hae?hD+F)GU8fKETcvbc#kW1W zDIjyGJ}zNQQkCzKnBjES_KWL4YByGEu`zinu*yfkx)2{@bF%pO$QTpmVP2s-CO_y| z;ziKZ)(j>*&@tmUs)qS4wSNX=pkd~gWsB1}Ya`pS3PLW;pH6oIqJ>xADGk5NB@ws6oC;%XklD_gM8vx?sBxJG2} z_r8k&B4mbE!y=|7UlQaG$H^Kmvoy|&R85SqW?;gA)HnP|b@8PAC}-A9E-Q;ui&&28rjb1FUFHz= zU)TpV3*BjcF3?vw*UFQzD=Ya*FiqZBrK~G^j&=9X;R9h)+K~<~{6o`J(2UP5Zo~70 z&(LW+j6TMym6^*Uh5jv9e1=N4VO{psonJj*~ovo=t_rH-XF}v#j<5NiPoL28I zC`Fl-{JXs7XD0?(MeiD2$$4AZiAvh2l4sOz)qa)_H?!dG%gtQ@(utN3}SX z9U1{d*;1*<;;5HQCr+)PEd%3oxOdpyWDFL^2 zHmj3PDcuqpHFHz?K-xO!U-QGhIT24I{F$o|_E|Y*o4^&eGfPhaK4;UeM!-ChZ3tPa?+SW@=_L?_%1j#$#qLPF;TEGtC|8 z;Z){WdzQ~se|Fr2b!|9Z<7`w0)D&qs3X?baWdR9*KleoTPVHj1q(l>ZRdHD4MSan^B%duwE4q~V)E_MKHAx=XdqduI-l4Y|VB zE4DIk`7cGCW$v|&=>pyLs0r0PL#zN@^uRVZ))0y39e-i0kbNRAM_-vYf4R^PdCvMw zjq@F3#u4O=i)cY1qX`}2Zr?aQ5yg~v&ogK2FHD^!V23s+f>a&7)#CizYR^Kq(CMWr zx9OiAV80dM+?@AnuXl-CQ}3e1Vzl`d{C?eI?|Y2h2@g|LJ}7~|==o>~yb8WFp`Xrx z@=uc-)#RxeLt-37h?|)>r44|*2OVMh_ymlu!0$A+z@6|HEp`l%=`&b@^DI);g8?Fp zz0cHzz1w#_LKG0Zl>!Ez@&9UZruVc|lxkAzDu_a*-(;r^M1gN9V&5U0l+n~VjipNL zgXeP)<1!jHv`aBOO8z{i&4kDi zota8{%N;|EJthP4HtqN7>s`=ENWY@FK#8776 z+R$FubjjY}B!^``LK;*L)13cjaU6LAOHBT^m|UTHtuH7Th@1fFp4Mg4;d*=Lh4^kq zD%*+c`Uh~5lx_gy#u)PKSf@;k5}UNF70>v^g8@gzD#SMC4d0I1Pi#4e zILcUwE}{0I14Py`?D%8w3yUcDo((Hl(lQqDTj6m;C2@yZ->MgsQ9_x1`ajgG6xL)x z`{@@!Cj#FRB+<)>8BrOJPX*9oO<{(jE!%4?~%?CXScz5783;}a-f61&G6QhgW zRex{t(}I8aoi?#R(k=OqNvYVL!YWls;Sc^Jud0q4R-C{bDAuolM2B)0TgntLq*bem4C8 za@D3(p9^&iy~^gO(wBax$>U9|Q8kyVoXxLhBkUt~2GW$PEYA^-AaBJTw(ZaE)dd87z40f`1$Yt9-`|e>-7B^{-@%gMl;UtE;z00+99b&0iPhyZ;7YY* z%Nm1nS!?n*Ui|g+rK*{+(!y%Ob!J_t;jh_Ml4o_R%~P7{uf`r`zB$!@l0IiXL}WY( zjOk|2s^9fjc2-}8+zky7&(Ed^^b(smJ1yyo8F(8`cFiRFB&{CU9PTk6Bl?(dJD9@Q zpjcB#7T-OXAbpE3M6j{m&2G)ExB8aT*Ay060%4N8L6Xcgc6!t?DMxH7?!P{xO`jOh ze`CFf(mqhj!bC0E1OA;HlTUt%(<$0Zbc!q-=v{ET`fc@d@=nrg?Je*fD5r4z)j$^3 zkIm|#Tp!Jn$|{NtH|Zewgld?eUZp9NX2AXQFAXaS;i2t{HVSGAW+o9f#k0nam#6ur zhy~vpUB1u)NP!R!|LVAlW0gr&@(@@H9YkION#WVhR}iz%KTxlu-X|2Q^YY`&J<6Ze z#|%k^U9&WWbbf)z_zk&@abl#;b2#{ZGeWbNH@w-rGAY_CkMcFMNi>O&*DGMtWs`B% z7Ccp8e62n~YeNEyGJ}-0(D~DiG4dTq$geO|@UZIF_=^4*2h_80SK4WGD{=*O(!hJ* zR8Gu*`Eare#;WxYY>;YpC?YW)AugtM5e5r=K*OYgrdw4r{&yzoiDr}XahVO{hx~H| z$IZ72Z^Xbj$E?j~30V@9rO9goT6+9Oi2S_btg_0xGUQvg4|*lb$Z!0Pl?ZgkoQzkX|A@jC$dH z^aRPlb=WoOCuK6-Mtqcd$g~9U2Xi?om3GY{$0>?2`6$y0eEvwDrMi!$#|}|v4?cPR zg`8}6kRNEu#9XED+tt0t0{wYkBHQpbibJAbGfP$2kAb1A;7EO$Xc?##C5CwtVkG_= zq&hklqSSGlq;i**zVXUJv>-V7M~_%{A81s+co+k4Xpm(MptxT%xFukP%n+F$nP8GQ z)Lpc~8xGI@og4`p=r5eD37kNA!821@>yF_aXT`JlrV% zRpTT(Iy!w@y2z(EYI_#PNJ&qftkOYr_u19nN8qwQB@keKJ3|c-RD@lIVeBohhnrDeFPViI0O5(%OG}%P%Ani%&OVs}??}Hn7 zFPm(k?a5y{XAZmpm!_5!zeR<<0zDDQU{6n1ttFUX?c9@{)btF@nAHFU%tHD_SbYl^kI!Y?)fZt74d1r@5~JZ}GQTu!R7t>a%FZ zR0Ye^RCT%`%_|UZ6sBq89Dcv~_oR$bVV5iUlI3Bjip9yb@t7JuP6K&PhQiWH@wdXMobG*)fC}? zAXf1Y-3hu7{iB^f^~tElG{5&vTDxeA%%6T8^4Tc#O0&uHK}b^QNZ?3R=Vbi9+|g=J|moZ=lUhJc~}`pbnw4NNkQ2a?1#Zr zE_x>>h2tN3*J=rTmfEs34fPehlXM6st5du6q+XaZ%>4JDR`9@M;Lh~f&-?nl2DsxV z=JWsMc0`nl#!P=XJ>&LK?sB|rZJ^hH*0J5^fUY

am#rUH>Y*v^hb7RXfkCMng!H zq&w^#q<8M0e3a9(QOkYL8DFa$nN8SeGJz_S1Z*y(53u-Hc*Y-VfpJ=~B@ zp{HqAT|+1^3h9RLDBK>c1lIxvSAC+ilpt_HsB7nsPChUs42-MDDJi0PDypE%nNKK20|nwFLaAFk08^*yYa}H3YZsD1`AAJ z>5xfI@J;nClzn!OXz8lE?ms!6FkLLzVr8&oAR?;ET!a5;_@)Yn=xo&cbd4n#6=dK_ z1Z$ud#*ts1%uAY*tG+GPFVD|#x=peq@cvz zrnWhUK8>N74}mu+R#@KGJy8G4sdAakW0Q?*Agbf>$0mbja}9DuceBW_gtw$H$yIy{ z&2q*_FECIAYeKiVH$AtKIqdr(U(?T!90^VXwSinJxEy5l8&Pd!ynwhnpVIDt{w8i^ z&NAsIwflY3Nzqx|an8x>;$6mmiq9+aAnnXrft;-lqcsmtv8>9O5r6oNP(^pP{}*zL zZ9L$2q`?`eQcYO5yRCYn`Ak7|CTN~B$mi<6m4|or_0TRkzkoG!TEbfKS(Lc~20i%Q#a^ykX}(qB9tbUFfe?hv^Wf8a$_Ia=3uwCwgf2y9LOcPArj?*d+4a zT9)fp5WOsm8Mp99NH&a~JzNpYWHOk4V@^VZ3e#dlhvj}Uc%(gHOBDVtW9!&vwt+vKsd5Qr9H4Gdz4N9XVDg_(0ticVNXY^t zON#L;F}Zn?@wea`8I)v<;z#SZs|#)jy-R(dcb?zdBGW8zF7|?2#Vb&%`yaIP%wL%s z8IjCWre=>~aEhxn6L%buR`8O#6u2GHPLr=2qj~s7Bwm@mp1^F}7E3U=3j=lgg+ecA zrB-rEQePJ~lkV~RZESgWD6ia;no)u)z0fbTHJm$8{T|&*xWr*K0-$>F+n!&FIsLGzv2xv91pL>zk#my`)V zph{4r_iMj_Hq*{e5lqWRDCO6pS15KgtibO!ON8kQJ?J)O^c4)CBi-!naVnsLj$YHF3eP)qu>zg&FYPKN8K{ZY)(K# zlIQ#noL0+dyKKO%0AyK2x-hz)FS;w1la%AT`AGZ=W_fa+9uf1Ow*%q;Fl@#VyQ$L@ zk#{C^hS^DIC?L3G^-<4;1`_#jXR8qu4I3<3;TwU!WISsWA;-I&LSiG zpUT$`aD?H40M1F1E!f_Qt=!IcV1)4}Wu}p02~F^P_G0|!@qFb$g7F~pkAfA@P)3UD z{xpj8zF|vfbNhLy0%Qpil?mg7-MZ~Opa*Xob8L`hr91vkxms>xSrXCf&tk_j76kK% zXUWzQFo6`GQW{9%K%QgMo%?i=ep~}wqXTHdUQ*pN@Yi(FX zd3iiR$Qn)BV$tB$#=o4#Vvy16;^M}s0_*t8KuI|N zcs_l%W@)`T!_3t>!<&vJoyvS*jkzutyw8e1HIrX0TfvjYDQrwPh}m@y>Az^T*z)7P z4$uS6oA@lIfc0rQ^%)|?wv8mPHvp69fub)HpAq3);XBC|o_r5i=siHuIjDDtRzlh> zjhH}CzwBSmtGv^uGpG{cd6SnVGYg7;WY^Iqy9!wWyBXd*!w#GKUqe}CsIBQ`@^0M1 zGW6j{N^akuo z;_KQaQ?0&@oK@9idY`2wl4}1FdX!$`?&AO5q0Uy>S=zXeGkKQy3*r^nJ?oCF9i+H8 zPKjP`G%crJF#XT)lRjU0ycIj{Uq>EATl_79*uCX>PCr8@WMk$jnf!(>mp)|9P19L( zz=VC0r}%>*PK+VL*Rjqjj&OLHWmV{M<|2_mq*&x<%lj-1@hAK*Y8y0LTjDX!&Ftjd zL%o;ku34SFA9{r)J(_KE&>a!TbjLW%njaafd>)IV8E;9Mh$FVY7;~N9Cojtv!F;_F zm|=NunTyCT$q$n^j@p>zNoNGFTC`L9Oy7J}_o`@bCwykB3zj}0m4`myd+Bgk=wQgDKEBDmJi(K}N|M+SV>&mX8gRZ?r>z2**oZk0s-7wCVFV^B3RI}fV zX8o4Ad=7%fV!Uh6e2bJ>FUAJMeWuW9iD4n`wxXr>f1;w1OyJc|6RzQi49FAx*})6! zw0KDIM2w#Id{-ivzhEzL*>zU?IUgZ!_ywqHwQQfB;C_Lo@niJ}cf;;)spj|hNenJx z=ky+aYuyWg@)^jkf(RNGN*1vvu&hU_P0}|zwA&w6jo43-(&hfUD3_2>ZFUXH-Yj1t zU0mf)j=2>Qe{RUBtGbi0pK&k`KuzT@Hn+fA{lXa|nCZ5p*@;3Iez-tZ^Sv``2dD0c zAr^8^<@rrza?wxHqia!jA-)|d1QQ>T;TyeO_E7Hrb=XNyI>9(NHd^l>s9vr|trhA+ zbBv30Y3zH+s5bp(@$YS^vTQi3HgS3(;95OxBeI8j?y(%u)GIAJMTe#Vh2-=wwa8;6 zXuQByip+P?KZ($l--%fza_*|vffSv)$?wa@eGg6UCBuo zk%OqI2=?M(a1}{Fx{+u6UIy(YbIA$jjA?|atIQvDKLJeopR7<(zFRo~U#N1`&a7p1 z;*JiXM_ii=Kf4PVi;W-U{?Ub1BeY!YWj*5OuNOdhK<>EnVt&yE+j$#vtN#Z6#5j$G z_-x+5w(WJ}#WdSpLESMGHNhA8fPf!l=o!;J0(P0GWdlYki}k;S`%gC^C+!5p>yo<| z&v0o5#6o&tXL5`6_R5;HG~iM#qRY$H?3B!;+qM#)#I*Tdo(hseC%99i7->ojCr9f$ zMC&lr(xGXTJD8PM5XZ1Q{zVHZ1=yEPa1nnC5FmuGySUdn?0MUVETws;)<#Ziub*Iz zYjaedtsrfQCm`Z={0Xt_XSn46z(jqNMmz1H-BIzq;yCSIZ2sQ86s7xmQG{-F4YS?I zg^-QEP_h!pnEgG+4eNZ;jWN>;EW<+5(pn}sb_~#Gt;UUIlwkD0<0nu(L@6ZEhgi3* zNmCygq0;|SNVw%{VPoheQayd55uRZ@GLAO9*O4o#9%HheFeikZa=;56Ae+<-u0L|gu%s6)@y(+z-$@t|fS_i2(m^JSEbWAm%J2z#t7U20Qj(Zu|9>lTD+0}MrB!^%%D>pRo|X&;Ha>i-x#NZ z8P(wpQw?BgQ?n_0%iM$|o=#=?jO2$$qY4RAvSv>#TGGetBAk%swgIolDU-lWcRP}& z==V(i?#b8?FH);a|798~9WG*N$-wxCt9F}bo>zwPCOvJXx=O0Z+9+am0-JWBb|VfO zl!!&WAZr+26@j=#`3HyQk2`KPs9C*ex~kp0nunMSzN`94(@PiK4SmlK(!6m^{A*+f zAEGSqR_!n@9t3MR`Ni_`g%+$SXg6qF8^#zb+$>&^Xk$^*l%u}pd{rRz~SYBNr4MV@DHA7;z@rziV z6Be2drKU$V*cItN5V(tN}jXCDuERKx6N5JN>{GaL*htX-U3K1AsB+60xDwR$r`s+;) zSD9@LZNvOB-!R``-o7a(d&$Mg7UHi{0j96_kDKo6R63IP6oOs#lzd(m+R@<2iSXMhFQYSQqF^AN`>!l{O1+^6Ql-D@;zl2n9^t5 zt0U+O({SbY!4~c#4B!In^4hK)m<)Lwet?{Z@h&!x$i`v0z_!WJJti|f?f_2C zS1{k2_*?jkW1y*+5xGMTP_|d6t>Jgq|7{|zlWlN%r2T<^2hZ`{!?)U*cY&467>QP4 zMM2U#Z;v)=g%Xzwv|-X1WQzB{L>M0p#?2-7R1gko#qp4<}mn2(g^cQc?-D<(>&y3 zXjn_UW?J6WX*^(kNBR2lCBVU+ao0ixIzWrWE9I@4SeFQo&k`}ib%e8Ho zfV&$v^Z4{LlIzmOOK6qftg-;6J<)@f?uQCReH}~l z5nw6*#BDU4%g5~iWX%@+hu=+KhHTqh*utcUcrR|I7FvO=DFxh|zmmu-i zqfn-yp}vR~;&pwYq~GOxHQu99cbImIRvfi%rVN>HoQMdgw0JjpnyJ}j^CHVO_J#v& zdS|oNE0gfRcf~xznHZ+2?HPC4$*3O#V%;w@$-n}fGh&x!72o9&c(30D|8RJy#p52< zTiQ@T8>cTs*b!5NqX7))6~tzKPF(pX6Y9{s4gY2lDz5`8Q6dUBo_h+FJx#QVA+@g2 zr2tnt)YsWCJnbr3&0eQx+#v{K-IH3g!!$rA@OuUHi>%$st6GAC4&Ib{eP!wj^0e*Y zc&KWfV#H;sF*gp+5V!b$q$5_T3s&p%fNg>hxW?QSSVqo|Y^jOBJ3*4kh2zz}vneUn z%kInShXVftB3dwpPGILD7%^4o8)jj-Ou0(V4+cqg4aS(ZOQ5xRZ0+xzM(mo#qMtf_OkFz>2Nj<^kjAtAH+CN;*PnaGQ3wfOlCyHrAHvdSQL` zTK>p{fbe6hJ<)o)xYZz5Cuc%dl72NJO$Ua=o#vhTtjk!}Y}`Ug;AP9D5CLgb9*`!) zEs#E+)xNq9o?Y&Haq@FblVt)pT7$&g_(f4E-C=<-O6{2z0kSES51H8=^%wxJ1h&i6 zE(mG(rM#W8G^L$4@cifG1``yi+g1x|r28RuYL`4g>e!jG4J}Z1!>>~N_wGdR!Q-Zv zI%W;Di=oxEal90<&;`t&MCXSTANFe0LyaS@f5DWaERc}*TjDBFNtbbi)(Al2V2am!Ug>x9vd9DX7+{CcSX`Y5};7h+;b%Ud6o%<%tEPK$55}77uPQRg8cH zqj&nq%b~V3dsT}H?Pcj<$imSWFMyn!r^V=b8>hl0K^;3qtL8ma$uEo3=%5bYP~LMc5ON6y znzmM{;0u!02yc?Y0R1HG76M%ppIcVdJi&-3DV zEuh7hUGFf8AYDzO4)K-?#mU zIRAA05$W?W=q1=LVxIQJBoQWR>eLpRfTe&eg`a|hPWKIGlMTbHBlUT>(_P1PhxMq7 zJk=MwtG-tnA{^74KKuSB4B3e#Kf@D0a-g0LAEX%{NGNR%g9t93=pdm0AU@mpB+G z8ENr%N^a7wm|kf7lO36-wUh5vG0^upolhgMkcTmwMK?{LMnM{j16I-NhP?6kB{MG!TO1nzRIPk1+z2)ZAn-&_U!BO(5Chd;Z4ryKeqqvkBf3|tC zAq5Ee?0()tKEQJrV@P9xyLzZ4B9HKk=mWF3bvTcF_6E9IdP1`K&$=-BTMMB`YSeV zq0d3-1k|#FcR?7kiQO&dT4T|!z@c7wOG9Q))xj7bTh+a{#Ka`{;c+3iMSPL~7TTPZVugm*jeC$M`U=lAAC!8EK3rr+l zcrjmZhsMgB>(`K$Bsvj28F8(g_PV?L6!o3Bv#q<0lMk(mpkn!i0Nje==(qD*c)P5m z1maU*bg8eX;CW}Novh63nlSg0FUQh3J}xh#Yu3U%@Q`bOlMkZQMHD z(jnKKvXvEnRZ)v5;DsbzQ!hl?9m`igS&OEOnf9769_g5mwU{(sH2G(E5>saBb!M-Y z*DFJQ0Bk#bvFzP?=X{l`)L+4KP`PQ_FMumqd8%!aq}II}2csi9P2tm4|9I%ZPWg{QVP^xi}9!VXD?a`%&g_DMozT8nPd zF$3pLM=l^3Fts2K{tGUyw-Cpo*PED^@;T9vXSm1o3F4pAK&bZ5KoL2^v~fKU4J!ZneFO zCkXBF)OM}77cm6w)OrrN3?aX;l$Le;mG4R6W+801g%hi%piRjAWDG$^g17nr);5tF z`Ya>U`(I~`GM4W#sUTlwwDQKS#-q*7BK$pL2WpbxC=0t3q2l>XcjFF{N!xfEE;H{j zCv-g%$Gx=Up=(upi021~lQ`s<7N9JQ!tj<((B0W$KLN}Uw{`Z4b*#NHS2mRdNtdmu zMSe7)7x1?!q)SA5l)hWEKh??%w>^GDSyC?}`Xvi@VHsX&vYp7N(wH#VPrO4ln#ypZDyyAICi&9w%`N%K=FP}qo zs`U_pQ1*O69}z>FK1lr%$1vyE>07-+g)fv2!X_bQj{459&zl#l4`HQ6=H-wzP1l&x zkIpxR2h|jqhyOPCXH=0tOv&CDLWiVzh&XN`8Rwt7a6$Czy+^cXA!|~ z7~1+SEq^O}kaTh;-GS0ys&o=YoP)Pv*SL8XtIy7iy3Z?~S;(|iBa|(n*PRP39}{X0 zyqcKyI%7G~`w{&s{~GhGvaI57>T%lSOGR#eV|}RJNEDE;!uJ4J9K9O!o)#Sjrr|xy zls*hc_TyApj|fHz_ygIP-OY{*9)oYFc6# zCWuN1h&smH-{>=%FD1S|kb=0zcb>E+q&EB!1utbK+e7c zfxSW()7wDHq&C-VEbsr4hQ~U|zN|-N_6LsV&Jr4q|C|o_n0j??V^6#P_kecN2gHZa zmH^_%!V5&ly2L=U&o|jwuT$#Zn{s*gU)L*kb%@FeJ?e$3$|zFQgTVKSF6dpS`8)dn zJHYJ!noP!T>upD`$>o*n9hrgPjsDS*=0A|xmOtrr9oMn*9lZyc&l*(gP@&ne?y!uv zRx_E54p;s2QT=etOLN9cw-67%vxR5jfut%hVy;&hr|4#z&Rni?K+FWFu49qh-az6r zf+77Gs2e%dQmwWe4D%ptGMyp?hpYSvu*wc(&2TlOGiiCG*Vo3YB4WG~Z6@Pv@?owb zNU?Y$|0;zC-2 z)1Su1VK>~ztaIYU_`U4>kbvF*+_~@Vcs8un!kR-=$lP$H?^WBe=#S*9Lf$N5=J&k5 zBAW)Y5fqOti2hHd@3c)e%ASs&llj`s$3*jOLMXDh|HO0YF%%-Ir-54+T&mS|adq{^ z6b5{B=68w}+4i3nh}o}Q^OW{30^@hXoWwnWHpJ!GG@xWp=D(SezW;~0aG4JvjZB80 zEe^2d=WY2L7FHofUCu?#xE-L1h0hq5A+~U5Nph_pwNNT*kJeTnVut=43z$-RXb+9l zpX({RWq&BT@kye&PObWXD9j{?O6#HCur2Ecm z-u3!}|AK$XxH&?1o2kZ12}Q25lklCj~+_vyHeT%8e*A^jR&+|zS!<5&(MM25Zo=L(x8@C%1bJRh_S?{?r z!N-QS{VL9juM4YR*C)+Itf(!4uaH&z8#Lhe0AoI*em)WYDa75$E1TEKNx*w8_aWSt zU1;dnTqoV11-F@&)W!IDPps=#j44y@W76Cas*HmLqJ!(&Hr~WX`Ye6>nxGB)WmB;{zqJ$CMEv++MdL&rWxQ@5qH zNSm!I8xN@ebYQx>yt5qml{bvZLPhXP9gk8%(%gKmpR$C^mv~|LfC_Roe|8ntBU_;iXQ~(I4o=qK*Wg$+26bZ11(2 zj>d{qySu)RXGWd9*CzHf_SB~~iW<*HfLiMkcA0u?!wX<*Efc8(NBCaTEslK1AiBTr zw>p1BMNiDk0~!_U{S1|@yp&r$G=!V=UiK+L-#(7Q|Dgl>b9U*UnH4YPK5+QpJLFwJ zoqjAWI>*X_D=)OyQ*ui1A57$9*cG6Ls}!rFVF1Ls;UynXE(T8efyf%*Vh`c`yo`p_ zqtey$lQx~TF8ZxhVNt0*r~!2Hh6%QHQeILK=+|j@W-%RU_3%aP8|0kc133yiins=U zs5-6OPl=yh*iO}7rz#&fws8&P3wqI@Wh9HaDa!ky;t_fR=p6=~1f27gN3#0;nF(+W zbVyi9TFw7Xnf|(uQSCZ&tGRxDJ7pp*I1U-!JrYfptR-50= zJVVFB?q@**QJZaF$!0f`C~03%suBrQ;3g4&O@dFmeZnd|xSr4WugH(Te}v~jYPV;7 z;;Q$?|CEC=R?;}bGH-Z7;eulr^D*C1yLQW*=4h+irvyc%b8g{_iL~VUg2&UkwL&Wt zCh@|HQt+lxA#M7MiTx=&J~+@^p6v6MZ(TZH6Xt#iow+|4_)50jW58w3DtWFN(t;|p z5ZlymWzfr`2*NuI0aZEZqH}OUM2-CIgSkm7rLAWUct$6WO$AOO3Et4X{1-^MKehnv zRzEP@kTI0wY{+{#n8KhSI3G`6c0CoRih&}p-9**h&2(`)_ z8`VL!oU-&2&(^P&7~hR)7esCf$xd|c<=N_@`OM9K!(V|&lic{h$d6S;$XW(-UY$C5 zdJ`$fXoh`?szJ}IEInd-?$Knv>?(G)4}U3gO3};FE;H97ryB|Hi5>Q!jjlJr%2{oG zj{DIDvvX{ld3tPOOE&RDf1G$LW}=HbYt~Ow9PYq*yPBTK?#-ud`Rf0cb`xHsO)mW1 zxyv<*`PsyQ$t2lFv|T5-#GCUpLFkU5$+#hNsGzp`gqO58^k;3j+WAH&CWZ$&d&7hu z;vc*@r~8|icXK>b$=HflV})b42n?@l6ZQGrqjei+8u&ju74S&oe8OK-GI9SiS^(UQ zQ9@3dbt#)~EcdWINZtccT;ekKB^tw>5{JxB+aU@<;WGT+a`Q-P4B5jp&IFt2TXp!j z4@R$*r2RcU2LHVYuDyk2BVg zRsZA^7oOh3FfQHeM3@y{Y18kIl&DU$%>;78THvzVco{@p5WQ>a`+eJXzHT^gNHjVI zg!b}Ke%^#oe~I=P$XOrq<>X!lA7kh1>HQ6pI?bmng_qkaM01%IUiEDU1Z{{rLr+AB zM7hQTB+VUoS*2FX@U$J~3KeCCzOLR^5Y4SVBq*ZIT&nibwj}Yu6KK@5|7*8@?KFz~ zA#N%mjCK_dsZJ-D(t@3H$IH{wJ3wJdABpO0+r;Fd>A%f(rO9nv5i8ZclnvB)g%Q{5?;FRUo$6&TZBlVl^zN%6e>G1rkv`TGi6O12R! zL2B%gL)S463|SdUnNQ4}QHJMk#=&EfIg}_+qAc%olf)|Le|5HOCYaIS>MmZuqgl`SbD*p!QgOksOexHecgkn*S2O7>|m0E>&)+3C0;XQ7HBb!2_;d9iiysE}I7C@)yce|qIf z{KNPMuFMyYT^Pe&XzR13`s|kU+J}d9@6y*ftu`Utr+v3H5eho-1d{!@h~=%$c**6m zp4g(EjsND3${(6y1LexJPLYD!Z#vGmJ1(5;tS=*V}8p>e@xk=QyquQ>(@z% zz}ad%#`?JXeIvSMz!0YcHY0JP!mAZ z4D+UGWSd0e6S20L`JMmnd^q5Teelk=`htt)Qilh2)ZO%$-iqy{$>Dj+&^!~vwBxgoI2!ZF*Tz=9`aRt(4PcNw zD5C%M%xLuo{Yl#m#&3AYL*rrPM~f+RdJZ&1YVl=oZzD{?7@76G5@4;?y1vq=m2&PI zOPFA-E$Sh^6GQrI1c4{=kaMbktgVyO(AiYD`g+4J@kbHX4dIsju9 zIu!?2QOCp=)gn*8(~7kPZ>o(Ca?pZyuDNHM?M?7pk7g@#Byg$_s{OYFfqlQrA#jgQowW~=0vLl%fHgSlT4cCqIN5nCr9oQnS zch%^9D>;||S;31!A1eO|p1@mn4e6apQBKVNb%yjwFrD;dfEM8UwzcV6 zJyQMs0p)MvDPlTT&*|;OR=jibI3BNvqSxxzp7CD2opceT|5Y-W^496BabIwI*ssLL z~rFu#U1&6EV-IRpnsjwdcnMtGba0z zk94+V;!3;QRe8rZ_ohX50Dp(tqQwaJn=>`yjQZ zAmgy`yBM2l`Ua|s|D=ig7@Af7e1P6ux2KMAP6g6CX^f_NPI0kH8A|Kh4!Rl01RjT- zQk{)mf7lhn!sMd#i6wI*x%}Sw-o$f;zzzEA5h@D*(r6Ty?@o+lG;H^`1iTzwoi)Y( zgGuoz?-iDIq?kwEXSrGdvh+S{q5mT(KX^MfD%V(sKkMA=V=kNKeCiuOltL~tHORw( zmCqtR%@#$<%Huy}rM;Ut(COt}Kj_oRL9CK!_s&MI20*KkqE`XtMCJ3`zUidz9iq^g z@XdS+urAVZ4$_U92|3FdOnmSi`3I8!yPgA9Nl&^pMok^3NL z=M$a4#dEw!GbVZ8cF-x&;=rhQ_!}iz!`}+v2GI3_fl+=SnElPGDzRrUA!$vi;i@r` zYbu@}@pLXgB1LzY=#je6s1TC&?*^`JJ6-GzuXC<9RcSt-W4ul7k18?X&Y>fW)w6jG z@0p)8;!rX2Fex5Z7h9|j4$)k@6X=(tt$%DKd_;fXB_gCH`_9b-Rg#b&ID*jL1#Z|g z=JRUN4d5-l)biw271#XHElbmP&d~`SWSn~z6;aRCd#Fl;rP1}a%87kWg2ul?OD8?O zQE%yPdTX{rj#y=dXzu%DbVJ<7_-J3AC)|T7C#9R5XVQx<<4Obp2KVO{JU9W@-z2=2p z-|D8JzmcS~k-mn{)unf(V^3JV0s2}}29*>e)=i%wP?l4%k|t<;1_{#6WwrypWWxdy zu#OO2<)_m5hs9Qp)vZ~k8 zqK!25TQBr;SMdqs&_lBY{SfdxG%Uj$B_RfC1?LyuvZc^D{JoT^E=i>pw1}IR#ctk( zGINJKH0C1fY(=)5CPmwn`C0UAl6E$JSJyO@dp@Qy)D0EW=xp*`5CfMkl$n_gy%u|5 z1!|0nr8%*3A5awt)jGt~vVrd0t~$()TJBC*i+Vi8l`EtBJ5ur<)B8cqa>5U6NLV|W8`H5iE$fQ(v7AiCH7d50jCrV$VQS=N;xDnB#0lQwr%A2patGfrkyV{7D5 z^DJn8l1@TdcZsd$(PB)=d4oPAyjx6pM>ONl9O)JNz zl|$3%FT4rVp^8hQ>)|Gsc{y*Wt$6`KTjSJMg2Dw!QioqTp=-$kUra$r-{8dhu958Y zA-B)wbp^|_>dKgO(?9BL&v`^5j{R1WJ?9n5O^r_uVk$#)^xO1>Qx0l^xTBud>0_>0co9CXY|E?Uo#Zt2 zbE9-0VSt#7*CGKy1ouwuq+|r`1mnTjKqOECC1@O;EQ(jWbEjO`Bbov|3&y2(wlD0Q z(cB(H(*R*M+oj@GXg~ zJ>mWXBX=~rE?ey=yP5@y6>B>qWL|8%mDhmmDy{@P(i8z;t*O~#jc4Zxm0ivokQ^rg zr*|B2v0AU=9GSSdpjqdx;0Hj3TGxqBR<$C|y!s2OsI{ z-p((_s_@LBMuYWdOL%CN$c;2W7fqnQI$xQY`RBcSUfcYbFgNY62oV zp3Xkz+QDkF?5lvvPb(PoSGZ|0nxxVD{5y@lA1HBe>UJ@_pThqMnRmcKpML7>leAcU*DIP*{OP*f;? zKKrKUrAthi3^91DLIE}t~rGAEV_n9kG40$ctdP!G3gOJQNPlonQtJha? zojZU5p?0|ONIzwn^deMJR+B;O$GMux#&E-l(kMQfTSr{1jci%`0Y#Ria%CMUZj1_Z zRnXs2PbO#FT%U8E*$A&< zO?yU24mdlG#!Xw?Ze(>&=21s8}5A#6BM~HW=P?NW*1aAcP~zJ(|Xp6$CbN9 zmazHYw|P#{Erbj;!hqzk$%ds7Y=ktg@HGmb_F}YbLp%CQS3idd%833u zDy9ZYmZVP&<+#t9h8BM~&An_2VO&b)WPHLedW1_KCwc2xi8vE=phkVi1e{6bE=>4{?Pe!rZ4=NIaTbq;pRNpTm_;Q0h$L5DeKgR%f~>V9hl_JxI#3cC2(R zW*mI(g}AG@?yQd5tUu`}T_HNCv(K1Z1*|P~r~x15U$anG$9a<2e7+v;Tx;19c_4Wc z@PEXHpll#G^d{M27bq>TmXo=bW`iomJzvQ-yezdbvtDxD^YQRQ`uX9D%RVnw49(|E zEqJRY4mWlr3V>Wb6uF&SbQOr||2QdIAItnARW3DD1zA6ai->rNhpTy7R`?DvQU*Ye zlybX>!e!42=RU4+JuJ4Jk?W6N0JalmiH;h)jYB}I$C0*58&Dwuw zGqe792OQ!$t%X9$@V-Cr*5xdH5pJq`1nM^>@K%-x}BYUqf_=h&+t*t?|c+)nscS>_bg0P^5B6)%a}^JD+%gFnRfbjOYrbvRoXcdD=Z;8}|zxavCg z6L%?lR>UCMcw2=%dQH1K?_qUf4o=w20#|jPh15|w}1+|U4SliQ-py% zkQOj%yOki2$rPCxH{e{ZdTC3Dd5a;N>F39bXpX(>68uNxZ`_)&bN;t+mH$vzvWmU%eaAgDp|E3MPPp{KkU8sQ6-!Qb}w@sSA z(W?C#W*u)ow$8}jB|`h~j!u2?Xn|w6YjsTzRd`Dz^NY< zYY-{}cUM8LR{1`?a{v(ywK}znLCU_=d8?12NL7S0JnmxO!8?oYT$q5Y$pqEnjCe@E zsG%)zhsxnTokXl{2tuq1GA(LHa*QwDee5+JQaHti{^KBDa(3JmB}QPZky$GU)4I?F zIou>g3|`Y}X@%Hr>Tmau`osBejNP4YGM(ql^dD7g-+RG5T9cm>4o!~0d)9QJUx1ax8#4o39^|dS{$AQEnyZG-Saj0J^jFy_e6q|M5n+0 zDWcbi3wXtguxj^ZWNur54zpO`USme`UF99?G`zd-w`%82sA#g8@%&`gVg!MCup629 znwuTdl?A_$M5uMPJD$Dpy##+!#f>0yqriZwx5}Ju^@es^!}w_}E4;TGq*j?uwvUgk z;m&KWKZRuTHOgmkMvShm7-`t?J8RslNhOoG3 zYiLE0m;Hx^$v}bIJJa7dJFM1%^~l{n#hS)2d+}tp%@*#G@vnPkrOhrp4h)7~BiVcnML{v+^-3B%h(by<2=)EyY}YlR@*XTr33J(bq@XyDmQqH&WdfN6Au(P+3F1sdtSN| zPhI6Lk55rZ&An36d;DJz)jODZR4QdgXu(rJ&TH)#&nlRkRI;)u0QFkH_#o)%k&@P% zq!lMjQgmK&#?LC2-!qfXmrOK#m*c95{6Urx!cD&2(}His2_>g|HD=jYkR&I-sC){A zH4?vO&RF6eF(^{&r|B8=)d#Ss{8iiqU+@XtcDtn)x(@C^o=5Ez&^97P4dQqGN;nM` z6_d>#ortL!a~Z@DC=XamKjC8=?iQRFo)sXh?sUaFAb%9Rv90NB@>T5WGUzx;FXh+D z|A~Lr&N}UF5NYfD{sB#7A}Wo)YhzguUTgZU%>_108xhE?1pwuQmpK(*1c)iTeo!wt7TbgOG9^2tTfMqDI9$ zukLLIc4BFuas5wol5sWU!8IuAY^fM;64C_$K11hnkF8vH(oKPQ&|%;p-paUP1${Xs zE8&(QE9e0$=`){A- zkr?V#1K)DB$&bJmAF`v-vxkRxL;Ua0cDtOsZ^-=%L4**lxwd=BxpqoYGtb;_xV|1t7RZNyIsty z=+E^-Lq8NOw4g;Dc&o6cB#q%gNLzvgJX&FDlcl)e=JgBVyR87)la<^&_Y(N-SS`|D z9sGTC9ZT+66B`V#iq%xW>fqvP{cL{Ghzvca-R9D=QgMF;x&6)X?#h&j?7cnW?9Z3P z_|XsA#w};As&U?hYf^6HPOaJC^~ZyFi7xC`2gzebe*5@ZGB$)^u+$|8>4fiBD7Rm| z7E7$L_h`oJF5rZgW<BX<06xb#>7)0$I`t!}AWMGd_N&nD zZG5D6ibyb7XvS?c)Ani??mEx0)r$rlG&0wam3V6(yp=?*YnYuEXZE{R;42^Ui%~E9 z@ozqDU_Q4wvnr4!v-SSBdvbut>{FyP3m9DZC2-wo;~u2gIf|8iQ&_jscI!%RJd+2T zwQ?LSEGk%k&uBYJy6d^ryEdA+^hIMw)VVDNj6 z^+!q&+Oo#ZKveMqM@CYOm%Z#{`j(&d$JAq@#%DzxTU&Y#CaoB2d8FhZU*1$-!)Q#d zuzKcPKfN(~yxb*X)En!@4ThPS!<)L_%+vV z_QTYWs`!>MHc}m^L@3gYtr}tWK61*nlu6?n(sqJdV?vf%x1(>)`_xOjz*_v!YPN3y zoK?#343*POf)UImZeK25UD#Sm%;u)%gtS8Gad`j-mW2DEt~`ZaLO%OR@#1 z;)=s{XpVMnuI*(+5{$!L>1rNHADG|?2On_t1gpdfCmagTQ8XTj*LwTl#%Z;yx{=J% zsHW6OQ^ng6+_tk4K`dY-n?M2*&GOSlwkndV zwlZ}E4TX0_u!-dhVFDOg;QU1YW64mJQ<$DV>Izg$cnUzNf-4xeYT3;*_-%G{-bB;D z&H2_k<1MmyMA#mtLzF}N>%J%{zrwH0=OsBcz+x4WnGQbE!LHL-*}e1Qd`-n&*iD?)tMHfkp4U_!3nfsvlvA)^M=AFg zhHIZ&26&n;SMUiL9+8B)FU$Gl_x%a{JtNj`d{R^@&_bW!lb&Tw3B9<-;8?6@;zQM^*m8N|AEWQChFnq^pr2Tlv@s9Fc)MA z`4B~1tJqdK@a_j#YeRa<4$7mlXn#E9N4olo*Nk}0896KYjX%CT0;?-b z(~}3c5(iAf*LdPuEaxS&sFW{ClAv>otwwTRa${@9s+mkqOQ6cubo56~@7P$Q(_&fV zP@neiy%u+Y*Op7ZKLu}rpDIrp(iU$f-7VY0wp&8q&mThVK8)eAuMuF|dbPw%g80LI zI-NQ)L=R-5oUl>$JW0n0E$I&v!+O~VQM{8u;AHP#P>PbnxknadcRKM$(ng3oXd+z0u5hWk?O zmq`)SP=a8}4(-tV+BMaDhnfbVkc$0j#am0Dz-`~cBsH#Fvz0%GuASl1NG&Vo+Db3K zbX_?%3`~{H&S_&Ov{%A)bzH=XDUfM1mbsc$hmzpxSQXR;Fa-iQu_fzYIn8V9Wq&cG zguPbCirfl_2=iR|_g1>MbR89f&|CWfy1hy!4-e%R>tjx=j$5 zY=p!uRv>JJMG2jGpZUf5KH=Wo+B?0Gdk6QsUz9y?CzX){U8@k)h77TD z%5l>;T`U)N)r1fft6f{Z1ui*F$p4J|a1s!-4;9j)9Z};{Job(E!sgfsU_F|4b73Ud zPgb+Mb2;Qkr~$Yh!AYy!4oz`AUkgFTihlO(j8JA{h4g-nkb?W4T^1G z63S%mn=+1^gp1qL3`$U=h~>N-G?J4K1W81{w- z;bwc(8Ekt>5JE>4hKoCv#j?GuEPW2o_~}@@*Csf`;>HZ<_euj^rWJ$uEIDdIAUuMZ zNql=r06Emq@9E+Ax!fK@Y$rRexHK2=|nvQ6fR5YEt}5%hlBFz@s9qV-i6@l5zg$awM#b$-FX zw4B%5w~DPwb0Clv;a*`P7dr$#o^YhCBMljwhK=9sSEGthZ}CIxrOP&+^f2Bg-Q~_9 zP~u$tS&&&1V5lrrnszoW} zn`WK7-u!f1t-OvTD>lQAc^kiFo5K#Z%_Ra*M?G_QUfq1l*_c_1Q<)^d$^OhPM~IHJ zVaC5zZK&g@54Sn(G#6o~%oo(z6*?y3lyxou{oV`@@U>KWGyC+o(G69kN*<^L>}IVPv|#!n6Uc(YrxK6h~B z3PX0oR94;v?g=uX@tN{pVJ2FStn+t~1)6%3K?*wn>%CRh+f`;2D6pzykA#>4LuNlt zLR*~vYE)L1xz8IZ*miY@E{*4}?i0byQF~ajD!*fR^qmgr=; zT=owx?{1Mxm+T8HzGU`{;r$>pxS$aO^(LN?Y)6hXHO4U)S6gdGTV|*(;0*^8U9+m@ z!U%!tm@v0+z3@nC%H|5~B}o8M>taYM1|eY+ZmF8GNzeAAF)TSH)zVPK;}@yPRiX*s zrf~LL=3F{TxccQ%mB#+#2=Lp*(zdh~P_m)8pOK289$A5JnTKBKI$r<;@LR5_T}U%g zu`K_V`*#`*>8P-L-cCI)=0E|#WA!>u2Ub1g$iB~b=br1BLjRj^_JP>6=#fFVHl9iv z=J*7(&00ZQzBZ3E9Me`Gh$1aE81}_K_nSvOo&sJL%G{z6^V#`;Q^+lY-!|er2@E7J z+W!J`0wAF;ZRol~QT!UUH=_v>@FRuqnsu*WVebx^G?sm>j4l1G^ovA7_BF)O&i-5$K`A2Q!LIM2Qg7bJ#?lQglhH`9-i#5L zj)_F=C$i>aLo3r%MEoD%5kR#M{$@Uj6n#(5W$#XL`kfTIJ>@x;CqDd33yEDT{?AkY zu6Fk4w_%koh^FkO&gOxpBF(V1MMZ+go1T(Y`ZoBeKR6xtj4OhwI{8p&Kh#x7vd73j z4d*xqFvc38TPqHxR3@fiF7=Y_ibK6`5TBq=(+&c@KNZSzoHR<(CquPQ@%oO#>5P$#){OjSG)MD(hlKPN$@g?+AqR+>qHa8WI3^(06%{dy zoIvFtobhnUHA%-7M1wR;TewW0g}K0NDUzg#erKXz1r7@O)#<-?mQy(+>B1G6_=9~P ztd$#6(~W~R1rE|xwtX^1e)DV}{eDsZw8-j2yBKPLZNe;nN(C8^seY$M#do;qgq@*p zeY_`*q+{KYg56e^P<2IQ&GMwWVKE0wb9~i!p6deCE%~^3vRIQD zG8aKJ9TT=goZU1uCR?HkFSmC$!?CYecz3ske&|WU_m1Fx3UBrW+WhkcE|k{5WVZD( zWP^$u$LfeFk4c!nWrp)`CO7BTaf~C^9^T5GjT}o?!E~P)Omkya9;EeX!oS`O!@O^X z`R508MZ8z;yQJ?s2s>A6DMa30KnlB^F57Z@e0>~N%k42{UwY{e+C`e=^%ky>3Yc*Y ze|c9%w%RlE3P|TnbFE`(p6+d`I1^QfP-=w6I&nVDpDcb%OM33gBxCbw1TxmU9BCIG zMHY)MGre@xPTm+{(-5)^qj&jn~N$_D;;yGwpJV>5k!d{-L2K`dIXzM}pSN(adkpf-{Q4()8 zb^6!Qr*)uT$1_Xm1Vo#YRZl$rN5QpXI1B_S4t@*Oh0(~oufu?A7&1^L&ntnx)ny$b z>1S(F|A%%uU=J0$r{I6QMn!Qx)|xZM3DW)q^&Kw*{t(HBo?=+v)Zu`PWhFKFG(pf# zf;E~3Y!5%HWlr6+muXVnIz1ToBQ6Ts0jiSgbrWxz z**cbseh5S{trx^@RnN8oPpR-R50XCmD=OETaa7&il)P5%(X12HK4eu^m|Wkib$p85 zIGHIxgsz&YP0KlJhl7W1%axJDM%@!S*0YIFNr8E>z$woh$@tTQx!|IDvGcvm6^?1l z^sHQyt0~hYZRH4FmLIj_s+77%(xtD=GvER!h5n=;Hq!S?h&sez@f|tx57p|Ih1*$& z;hozywz>R_P#%LB0tVD({(2z$%7rmbX-{WL)O)!5UEP}Tp*HGOuvLkhD$*~(q0_fe zu}wa>iVw&K%vJ|2qVNwRZ+%wH{tv6iWQnI3BP;oXK$yj;bL9L(8MCmqcBkA(Hh~@`)p( z$xpF&N2kvhU}6O}mXEqFRB$l&J>kP%Z_PMLyHN~XnO4p61@?{D@1AjO9n@Ao&I0BRXzPJ;W{%2_+u@+8bnBlmyu+s=$+<4Do9&XIVnwgF95%- zsXT`{1vULJ()aPPl!)Y;!S->`E1d&cu)W6Yb3WH>{wo35`kvLH3UUf|qdDl`WYU(- z$t5#SrzUewim{yU-ms&169G&!e9}Hv>2p=A;%sW^SAydRZ5czo#0!mRMtdL!+Mywh zR_n9EKs6Z&CG0a^XB3WaM@*j^%mWFm5bs9yK2oDn8I>R5DQzy`#o=ee*)TP8RM};3 z)v?kzR#t~OrRQ=mzBA@VeOmh99U2q~y;?Fr;|KB&v{(K~HrSR%NwR5fm=Zs~4)-lh z-7S;iRaHnfslE*Kic3$yuZ@^8Nq&WyrBX!`hR|664zSF_9_UFBr%Du#zgFy?Kv#%n z=9h`gUv`n6jsAz4@>seOkNMORzl(Is@jiV;P%}^KYSf<*Rcp}7BJ+gz>f$=B`M@Xs zQN?w{KkE13+OcrA@a#xP9X(%3wyv$Ov*SFr{A;u*3c7@G1w$y6{rUDV)RcCi%#JQ; z&~2eN7SJu_oXVAZ2Eq6g1zK^-T6gH1Gm?q55zWRk_!7_(+&5Zc$4BT=Q$pIRAqOgp zos2?X){!*jjUl+iVc8%jxFc%2EpuJ=7Vgj$(+=LTm8X$}+#gJ=<(G2iLC-w1#7qD0 z{RM90fGSY-KRLk{S}Mu|aE(1C#v|BD=<|g-{;{?HQPXAw_srt|loYIYt*CKSfq+qN z(;c&8`!BsYr%n+E(}c|?=pNCiyU;@~=)LKKxB>u@rmpEJU1Aq$biZo4^j`4h>{s5U z_I|`v9TL)OxfQ{e6Hai?lXAliPA_B{XE&ygh1$F!M`ic%C6jw@;MgqeA^s&Z&HUPQ zRNIjdWeAem1!^mv`=sgS8E7i1X@=dBtb1{f;~c{PQaW=o&s`{ETafwgc*w#zPle4N0!zsp(3a9tswds}q8E_vdzVX=#~o@_$* zW>482_mjQLtmlTFSk=xaJdw5%J! zWzG;ALn-^#yrQ~?;r7SFbOdh}E5H3sC@9Sld>$H|&q;fMdfMD^AH)v}ezi3W?}6k$ zPjMWipMd^31G07e5tR729mbd~bd^gO{vJ{BaVkkufMUwzXd-w2@g#WdJ8Cms30k)1^vMgln}+VSQ~YtdLdBL%2*c=FNOCmf7Ey z`8JVT7b0pk&NCunTEuUh^ha+U8x)K>dRu%Df5SjE_}aCq;ie>47bF6df}UFKhpH^@ zaapO2BXG_%La)Jv`?%=5ATT~}Y3Voqmxr?rfcVYvMB#_?{x9BuS5HRqoj{d;5WGLuzKU--0$Ocq`jYB)G~_3JV|Tn(woorgt~t`oG5!F-=-FtlQ&9b zoF(&T9$h})nO<3aSz!eu`0nop=BdX=bx`oPm^vKj4hDQ*Wm^(l)*>x3bH3Ebh852%rWENU$d3nGj{CYHP#TZcTeUf;HOB@bylm~ zbGMB6og1?@CQguwnZ-~$83Sb@W1t-M6gwT~dv=01>PQMAAn>lH*L8{;ZXZMZ=qgq| zCC(w<=TM}@rCjXpY~i+!V-x2P%KRdjM^ap-81s`isgRK2l&pJYKExIJ@@Cd?gQF(& zJpI#utQ?R6ZOY$-^9w0f&<#0z40%Ly1}aoy_X_2jrbj$2+Nm1$RDBC1@q_`d zW#J{B@@zcWiQ`%RQA2ZI?(ZxHnX*1dk5o;+*ps|@IV(O%#%%_(ia zRJ7-G!8NAcL!F_E@hbvSMHd-xHU`}5Z#=!gCw6t&LJxEghh`mV+jp8YnxIF^a=N%$ z$Gh2oTk#+<&5a)B=RK2L5Pa3_T*#mC)eIMd(oRq+&&aB$@e#o}cKBOn%j+V$_hz3!<)QN- zrO^ttEN;<{3|;TdJ+&%ZxZ#97Fk|36T9{D}-eXaA0eW5!yc1* z?v)|+DGHpA1>eORjxJ^6UKwJZ!e#JP!>1lbl66~`M>h`ar0zT*WV)>XjzZ*c2mikx7-51;N3Hx57up8RfM0wF9jN&3Kk9e z&H|3*0UH&hLw9UjFn7diK=QojIl>LR@5Z&8*O6g$W!Bl0d&%oYN$*YxtQ?u%mBU|5 zfa+iap(Iy2W>WHEwIvZ;rw(GhD2P}(n;)~GLw8HFKdD__Yuq~u-zGcWZv1_$6jeL> zRQ~COWIGsmp#&=?@y|$2s7KIu6y@!qgZ3Fpe-8`!vKsR&x+2K|C3NI%>e^MQ%h|6y zc*1M+Bb?=vw(elk<<8NIjP1-J*M4}3RMC-^kUPN92E;?wxwdC*`ok}*-JT}&&s}41 z2Qv}GUM(Q>uDr%ArHgF1+W1Gb=mZiE)^Oi{w>AhGsJ>9_q^Du%yq>39WAEDZQfJ^6 z1C4MWa)W=%&yxq2cP>6cr

sZzx>cOKgnMrNhkPP{5xs(jrCwvdak*%zVX=w%k@P zGO9VlFYp`C7Tl8In0Wsb;B=0-EH+Wj5~L*ERW;g5#_lPVY70wv0GEO_ml1%^yt#yv`>cW-<`JvXjhVR7ATjwLZ4tX)%R_8r$ zI+}Z@ELazfEgh6?@Hl79sHWso4(>3mdo3t>6NSGtC1ZmO`Yp!<`>vG6s~^kyyjqznOl5_~eC*Is1Lw6&=LHIAYfsFe=8sI+E+Vr%<)W(iTMa=kye@`7-?6 zH=g*!;wg61UovrTn+$%jRi0gUgkTi?nuIm&U&3Um#Fo^Y!`acYBY1}tbG(VPM^|FlHjpegsx2;WN5~Tnr>q{9bWe_R zEE9z4ZU#ds?=5Pm7~aX&0Bhif@o#?~?1D~pOYNw5I!CQa=al3Dwb_E7zip`CZ zjOb_0TUS$t%p3{uU`nN!)iE=>z#OyT8(;x0Hc#wkXwiB~7NU`~bxr%IWBxR=@tZhr zS(HXu?oq)Q>hWQ;pqS!0$>+pku{d!Orb~n^`&8z=3gv!W&5-JW;F;GQ@0REb2RUf+ z6=+&JaR2W}z24ruGlP13@d^RO4RsqHNtu_P%e6Hyk{r*}j=pAC8#0IGH+^3LGZw3r zt_i%CBA=1%F{PLk8A-?o@n@tt^UbW7CF zNokJM$@w}>P!pu!85QkXGk>7JWoAndTw)WTKV=AQAb=lgZ?{8r|8Anc19m9T=C}pV z9ShraLxtIA=TT+ardIdkkkMmxAVwbubv}6P*dh{_m(MOzKx|KRK3o2x&{??XdR5u` z^hj>ze?YD&b|6Pa zLat^EE8%Uevll6ma~gK;(`8Otzt_02fI+3dsliPcJHk;gL-iuNujzj1Zue7YY?!(o zPa7>G3DK6fak(dtbKB=hLd8B8TmUQgBxxeH{vR8dq_-KGHNDSHi++%4qV|E5fb!`Ue_mo2C!JF_+ir?${U>RBsCxBbTo)SKo^d8)eTMbNk(3h0 z7rH~t<(-uomKc-fei2E-qNN;XhbP8)djH|OZzc1KeVkrh>IDWNHM&~zND<^PXC1H_Db$b_(iVGnJk1PpmM@k1ANw=)%uuI?n}j2s|H3U zOrx33Pzj8Mc!8cU(o#ax#M@60_S24Xk{+}9bW0#$m~t-bkl_Q0@B865-ME}?Jc6rL zol#e2SV9zj3(9b7ZlO&WoGlm z3lZ+tCuK6=5+;5*^dMM$C=2S<@3&_ZY-2fGKP+|*B$QWLW}w{TLaof}Ac_eqoKk&2 z{Q(jF)sr)B8BgN;?CEMRh#tMeS!=TNUs9}tEe1=y$Ezb(w49F_z;{2mm>5;^`XGu^qT?IMGv8zl_%1D_pSM{*as&oUv-UhD!G8Ixl@3es*u0y1;v zy!)`73Z@QFpH>H}N4N8mYPX5KG+(~Mo?wpAg?C<37xts7Oiw&- z%wyjgbMOba9qOZnI+SGnZ<5_jv4P^TG$i62D{!RlPCql=${1r;KU9r;B8qdiEa@vM zHURyQ5%qb=wR!NB_F3mY6T#K;L@-ZN)DEgKcDp%2>y)luVNe@G_Tp1A{`)%C3!4o5 zfw~q2yc*oA`%eX!tbloYub@)!?iPp+2l!_e=0g~+;(pn6nvkPUIz&>&?sdGE{K06q z(sI-c16=WKO z){l}jfopAUq91kAzlCu|aI!{_=r@|`%+J9}vBWeqqejq%DUeE8*X`cnE_4xBq?8M= z3zQ3j3y5dLsfej#Q;kDrn)vS_v@mzYOC7=%X7Ve)np!)u%BWv}k5~{@cNU0oR~czI zFh#8#P&&bj1kq>bIn|8r&4BcT%&O3$LwA6GF%%Q>DGu^&hSNVHR+z@W8}ByLzcFDcA_VZ_(3Zb zYYuyHcXzLK9XsWG7u?BYmnh=YbKVW=fdmD%N!^`%u3*35;SD*t+e%QWa#qb`wq8+n zkA%pY8ktR9!|_-^FwOlDFgVleUzoZ7uMpy&Eyn`}yW8cS@trSSPm`#nYm*D$+6MN? z_%O^Ti7+0*Cte#9$SUUen1Bmgtq^~BpRmLp)nfDo#@d-%)XY;!yp@gvOs88tIj0C_ zcak-7?IwP#aR4qAcbm5;^(~E#H%goo0R<|sJi$@_u@4t%48K*jU@{bSM1d!n#hgqE zr^Yjw8SWki4FfyG5poi-@0U|CFVFH*F^j5a%$P^|FWNIH#K-KB>NWcWa}{ZF=2Y8u^5Dsx^J zpY5^G*AN7D^o(PMiJ8stfofw#_E-b9aeJyPA`g`9k{1m`Ee| zsh05IMCZ$v#9nx^lnq|0p{1 zs3g<&kH7Oay*AUN(>6A_&^BY~q&3yJppcnSIZ38wrASRvn3lVctW#!|rL>tXBAJ;x zRBA2=Oj=o*;D)#XDIz<#u*m-7$KgK?=fM5k=Xvhyy1w7fw;9cRGNLime)c#s&{Xo0 zmVPbR17-Z#e-lAdUJn=iN#EceB(n`ZmQGiK4|@FBHk0E%C!r3M;VNuIcslpc5<=S1l@ zKv&tRPHhqVL6t4%j(+-Im8}Ppy|{C%op7yIG%$&nP_Beo@HB#n4lG{5IK zos+SLB|T+zo-wSLD}mm#-B4Eu9$PQ6mfG8J3i~#U(9kaHr=XB-qow zLkkIS^|60Nhc6`gR}cZtVD(BouXc6`EwiHB@a~}n;l9!uM#K?*w9b%E3!DNseeB=- z*Tfbb?tK91vP>;ZhLv76PY(%{Pa=P6b1A=b($sDd=?e> z*>%N&pmRlV?J`Ca;X9Nq`;-3qC>zKB!64Wn3MFW=Hp~!Hy6{?un>?!sv)*XwyIRi( zbNYZTon*MX=j+_^vv%CoBzw5!YZZO@c)0KVWHOXHA|C%CvT%`6dcJ*lP+1ysrqbHpj-KJ5&n(WXuLI|!VbY|i&0pm zP6<&HG7~-v#5p089{Fenz~HmYd{nk{q}ygsP$(!PrmQfg0M{UcM})Q`BbQPfG1Huz zmbk#z{kU@@xpv2WBOx?F# zpwVtV)YU&|_GUlU&i3lRn*AZ^V!G+*U)&|c zBSg@n5u^TjR(lwNa0ok?=k)iZ%nw#($Is_&qcb1$h!&|h_fWI6$JY3=Qrn93Gb997 zrBkP??Os7+kEk49>0ATIe?d#jVH~#$st-WN7cEtlmBP+~@*3E)!Hm>Kz=;1S8P?m#hrws=3&kOJn81#Mpk9HJrvbc0dw+*B!FRd8CmaemV~$*`e}bF4*_UJc zWjxpR=Cg2D5@BZRTfE`VO_{sh|Hkc0kq{IlybZXB&6?GZRjIT3t-c~7*W%M}EzXUn zdn$dpo3Yg1#a{)@`^~G%Cxlm-T}$CaD5jR+^muOCrPnDDYb*R+9-~n2`rC?!+BG7Y zJ6(hORM76!``nW|lgUpqji{R|L0)4v#aH>4{ZlSHf_)w`_00Ari)Pfgd$Gqi`W9F(r&{A5uOpGF^e{zL2tF!-9uCZnk3^YKf?7fQ~iIW>Ql0>u7ta;Cf|n}6#p1E2Oq zZ2Q#`vO)U9=~{U1VF%+#v;>R@w*-C}J8=8d@Jd%Mxziz4xjpn^R3(Q)xu?3`E@9gvuEMdtJ&d~ipVF5qDM zOp;K_)Q*Vr8okyYawrn~QOf_(sQdUsj-3d5npaxm=FR)Oylz_F?eFge_|!XBX;?E_ zS&b!4n9wCv^PJ1ZZ(e}v^IpwE2(O8D*({gyDpGLIiany1hpV8O;C=p*=4@%f%^|@I z&G#Ir0I(f$LI}~qM6tQiRtmTYV7Hbg9i=H}jPx1S!XXJ@) zcE>#r8_~O~urvuE8QxyN)zS#ywgWJ=wxqY7_yoGLEx!o%bukM6I%RJYrZIjoU9-#! zcC@Dz=>;pIc6Kw;jqMzMx-jbX6m4J$r@V$2o^sw4Y2rv%NGX z;jn;rxpro5^G38McL>>b?rM##}4Q=XlbK}Ll5*!`E z<<4*|2lcgf9%h$AmFb6OK=s%WkcU(1Ti)y;Orymy+FLQ+*q&6ZT3rN}Y{hXMMF6|j zN?GX%I5T{|-|#`XkU7y}Aw5Mr0AVEoer?o27e-26N(_YTBRU3++vO&u~ z4M`=ilggX}N>w+%e(@M+ND!-p)>!hAp!e0cR+;oaXIpZrGe1OUW6C{%9~cJ zKmT&<|7qBy`B-+&UQwPIt=gTBF5HDlp1zZ@CG-e_v5`;mm0DlzllVn38ev7QakMxS z$rU1!_RUq#jR|TTROa+r>zB8K(l8FO#%JwwNV#P{sEfc+9gcrO!LJk(hU~hF_))Jy zcgYN?KYMxIyg&My7ic$Z;nJn9i@;zQVIn{{Tpf`qe{jD`5im2naFW(Bx{GB^mE(%w zuSd&$4g$mbN1-f`Yg6-H84A0VZeP=0@nW^&kTYdkt$8jZcC+dg2AEQP5;!6i*176V zfu_lBdC8#kmY9X`+;ZKVA~3F3ob2{8SKc-Z|`lX0T=R4mvE=*+4#*^+JuR>ZS1|%ohBc?zg z$4w+K6WYgp3hW5(a^Ywb3Ml}2R-UA1-6%QG#dLAt;Ka2_?vxq!Xmh8>P}klrp5Ed=yX?{;h)_ z=?dG!N4bF;7S8-q3=>850-|XsSl59;RIgZZ*Aqt_7S`+T#FAeRBAcGNs@m}CwnFys zE?{^q7QV?4cxKPU?IMiG2%`f9a+_!fbU&-5l2ZD1>*%GqHP8QH2f>IaS zbfzkIPk4Y&0Sm|?#&LYT9DL*V5ca5nB|#1~W5OBosO3`vsF{ki$`ax2Tm#PP>eKdlxZLLxehgUXvRK)T?xcSpMsXF!dT%H}ienwGK25%*5r0Ro|>uf9$Dap4NcA zQuL_@*gwbNRohx|b;3yfWHkOq3sUsJ2gUBjk)P?y&l|=)^@*iVg5aYv>XV?nM?u{6 zT46yqD#}`}PefRJ%Pq>`TlK=%HE^*PUB9Qzk+oQeJZC6V>r)o$|i2xXV`3^ZEX~6#vQsYo^zsDj~GeDZ4hAaCc;S zzVZF9HfJB%tjzCK5zZ@av2Iv8qE8Rz5gw0dq|%eb>o9!ie9oBT7-ZeG zWVycb@{7Rq1FBDV7ykf*vfci-jXU$s<3l?NeSJ^7>FJIUMUDq0Z8zAkrpPIj3p%>s zAJtkF7kng!{*qb5m@i95raO3LN2l8#R5EfeRXIF&wd4-K_^17((#~tySSO6`Dl4QW zPRK8N(7V~~9qqyaJI^uw*+7;y=1k}ev>rN$xNB&sk70kY_l4Uj(n}WqT^jQ%yY_{j z!bMFyjW2CtIk~;6mNyK~9d_PA(hFhPQ+A-`bFVyCS5LQT*^h=qaFoOt@w!`Vqzy!V zh@eyErI%}D8D}nWkNX$s0B4La?bXamkD^l9#~L|I~`cV&oQ zQksOGn4ju2smO`KLU=Efd4VZ2))*v^SEHD`jdQ(xO!5eUy@IAz?78Q7%<&@62$Z9D zGrgRSQKoXt6jG7vZpQO+O_cQo(9~6?LvX0&Cz@*ad7!vH6v7&qF(JDbTtg0rzk5se zM-)T@A)+{KxeL>qT-)`z{NAnZB}3^&rBOz|nv@1*AAXBMlSrL;oXGfQ+&b^-BYvyK z#w)_u1zPjO56xLBgK)$K1>LpL)_*u(?5a~;Wi+SmEAcI0bne9mbqy@|PHn5y>~XJK zbl};pUI;kxc~I*jI>^O(|G#3>=vJsh*jYlg*h1T3cyLk3FCIrw!koD(OSzD;PZT1( z@+EA@9;fWY2@oBX*18ASjV8^}>IvWuj-PwIKnvePX=Ik*69@tRH0>v?Fab=n*3A@; z;(2oIdu=VNt9;>i$zs`rq&>RqDYZ7cLgvv?E_6p1#Ydr!Nqg(u3$8=Nm+A?b^hkR3 z0ED}J1x@E?C@36Hbgwa<)V{L|^osbBz?%8$kB`Ti4#!xkPmA$Xg03)vYiiwuw62(oQ!2b zV9KKVr99(j&+OyY!O=OW?YQtP65yXGPSLJz&B6{*+_k=b7zWy5h;H*YhD#~xlEQgk zqbl(OVr^+}%p7p6Gy;5YhRnHt80;2+qw03iRC)gwEaRc7h+?quST=qbo3kA#!`~Z3hXm2ZSH8VzNp=cUj6Y_7zzyvC_J#h zw)-a4!3XT=f|JMnJyjgrr;EMKS&Y<$D?spg01IM~lVTsp;ggdV+>3Z6cjr6f@;!) zn$+V2#CQmZBOf*L7ANHTff|(MfVkj$H@XbdYZpzXfn`ntSzd|OOJ)ycApHLS&Nuy~ zAh;WkTf+-HWs`uAweaex=5Dw$%+nT`Ra)tX0a45>qxb+@eSnSIO)hNc&nS_Bc#`an z1Z!d-yA*o6oA47tdxncSQAAFV$CjQN)k2f_uR<_hh%?wSXeR=~^)Q5BzTW ze8VMWv8=t*pxndPLPfTTHo(&+XNWqfmp{l^tQQ;B9yfg)#oYFWpCkKa`!OKOJu71L zKcYy^wF~Y2+V=SjwsB8wT5Q2!zX4N=YtZuF)tQ^eb`_Dt^F)<8$_HWDgjUH^J}Gva zs*ue?JtB2IakHV!Ojor74dwrx*KzfKSd6FsCY#TgcdCwnJ|Kih+fbZS%p+zdYuF~O zp5H?sodw_Bpk75wjb@1eM^Kt(EAj6AX4zl@ILiE`+b-!fvClm9rA*m^nm1_^M+(Ld zt6s1f3sonI066_nZ&HIDZ>Lzxt6xw`+5=3gX+@w|lSG~Ihcr(?MZ95U?V{j^S(}$? zqLUGnZ*)BM=lAPYh1>~Q7P0E;%6BkpWoV${}koVPJc6abBuU@_?i{0Mz zZQ@<`|9b9pUGnX|PhK8n`5R7Od;abA&|@!-?f?G3(DRsD5{V6DiZNip7e3p?+h&Y! z0RI0A!^f!YVG(Sb%NicxrwN)4hTjRF5ZP8V7+cb>JRswjo_Z~>Jjg`4ZX~^g1Fc`W zx#?|-jwI5WNwv9l!B^`%4Fo@_M}>9~1^&Qi;yPF~fs->!ad*!cWEk#yW^jpCIvT2waU{Yp*UgE`y zYy5UWon-~G3~&?otT=`_s9iD!t#*qea13{2Q`+!lPaTw5LCrsL3sDWiE7xh+#6oz3 zp~~8@a6h>{L+EK;wDD3F2schokjEE&o_~}RR9zomEzs zL#9m0G~1f@riW0^ABi@QN-2T)4YnonF;Tpw{V9P$a;CtS1Pi`M|#21(|GK@2Irfq4CDeicrE8E1HeXv$D6F z=HTCaT8eiIo#+?z4sS7SkH!dz^qkA!`y12zcZP`j*)P)1Ab zbsEdP5C59+9{$IS4af*-`9rCdKF~k@jqTA;YKwU&Za8U&hX+}IQ))g;-9lUAX7qL; zJh(R#-aE^z(L}I(5Euh?hA^%BO5^CNY@N*zt@eD4XC}te&x)`%|YFkBjcON zRG-_k)NyyD*&)C@h;3+yHIVyBzlEct`K<^3Ly%x26npvz%R@mB;1T$;F4QG3P$R2}n{mn*0{*Hse= zLMQo|pN%a=H=51`qW6{)rzFh-1bDph>Zgby^IeA(c`Aaka3uP1YvZ%1JtQzhvl{@H449Owr)2;lLlrzgInGwv;1lP6A` z56qus4=uV$@gCYmdgCDH0vCGD-x}CsD!PawAHfg;#{0!@O0AoEvSHnZzbZDnM&T1u zB~PWc4L!EJ(NY2da~n)P$RHwu;_quSj^u`H_UItcx(^w0)RvP(K=(WxCjHI{V@@6x z<~0Q2{FHMQ1DGM`M?zVG;`f=e+r-PO1&9W-ZL`vNvzv_F=CB)m6)027rXu&JEI!*- zJb&fd*{(-y+^Zd9?Eh_Y0^awYo&YJ0$_sW1D~w!%#~&UOs&enDD7hPL-clyY%l}N? z{Wn;TwMW`s2`^j_4Bm3gM!X>KyyV}sXPkK;9{-&;_n9Epu!V0*P@Y>rX3e!y$#*caa07h!+!>pQZ#GCMV&Q&YobU_m_-Q4~6a`+v1sNO!4#POx}G?AcRdQ1<77Cz^`&x7H2^x)2Hob>civ~6@a)fqfc?b<6SlPx`fyL5Vp z>$G--U_^Wfj6E4FkZi)~lTMN)U}U~d^ZxrhpJ{M_wEniBJCk@Ab1|m%wNs-k*X=oT zvGgKU5W7w}x$-1K4b2u;2Dv{(qEB=a63%=goHlZD0_{!ncyR>`aMCbH?WgwfL)795 zQt5UiF8n3oxWO5FSDi5!&-U`~J3XTJcsQ9aB zk4tyyk_G|zFY*vn`+;_4@66^MvQm}{*kh-R}Wb}cYoHjV#~xv>r3exduqjA zN6TNgq3pPU4fr5k((*2P@4k*M!(q0sP8w+q^K?7dGQX`}s;?2WAdX4wpBx!8RD|Xf zLPdTcPOP?XHZA_XWiqs9K9s4OQb6HRyOFNqK`GNGj9{o3K$JB6c0uDVrsoRH)tY@y z1LXiijs)UA;&6s=4Lch1O%@8KcQT*Ece;_EKZN%5^OE|^_8aEXdfT?)%k|D;unOFvBD6GEW#$3H(|La|MFVvPQ&NJy^z)_<$uSd)MZRB_9HIFvUjmX|4cHq#})n49^NCh&q|Opo4zYy zdrUENK7^=^P+@{aGw7X8Ny$LXpE7T)`*iG0C={LCMW$MK!KtlKAQSB`> z4*Ry#<2)w&FXQL4o`LoZk1)`k)H_BO)8``cOAmL;YY*Kf_x+UlZQ#ZR59!)yr|E3~ zbV4i4P;{MI>%%Q9Vg5rB1QRWMnS;cRe>Ht|pIic?4I-;0W7PwWoaFUXLo42HHRq=9 zJ?m6S_G)f$(u@#c&+pfw12WTL?X}P&;`w*f2TT{{n$OHZZqY4tI{F@SFl_whn82`6 z$>KCDEcX=*WdAtFN)CWW3>lV6X1>V}AV`ibaP7Ou-7`j6vGDmMFbgTyWEoL&-Idz< zh2UNQA=)A(KNi=*){w~6q{nt4*|&3T@v>px00L+JS@GL2UvQEDLq8Q+qals;kRw0F< zZ(}OVe~$ZD<+{!`0s%IU^hFL)(c#SECdwBV zm^GYDX%8*RnKVv*3hAAxm9J?ANyqMXFVUmJ$F)95 z>?Yk9I7@TURxOSlNm620WiEj2{o*S94->6aAXCYDiclYUkDjl*bv*kurDPFG-<0mA z(oVDkeYghtSTCtqZjKW#Br^mg3{Bft$IKE>Cs~H;UD^zVvCvHUZ56)bRHpW&)mKh- zkW(o^ZgsAe#q)L>V-h6BheB_`gi*jHN(2qFi4p{4%FX9=jN%3iS52YkE#YzQ4`%52 z`NN$G#tAN|0)(`IH7c5WWrofo} zy=>QyezPlj^dpPOzs^Ma)}lt98Nv_T)&J*C?Yh4lPcpNq$tj{PJkh(g#n0MZzm>0ep+bFTk{dCf7he_ z^lz28K+OoD=W$V!rxzDGy+CH!nb6e$r3UlHok3+R5t}lnioo5r8|Od~PyuZ6nXFfz zJV_A8_@R;J2&%Y<_w4m*K~uSpXd`{l`AOymWErTF=*xg#`$*WtYmHV*D$Pcon?Sje zNUkQ#!(by~yZE!=p0ISxM;d@_ zf#QBav1#6=YO+kdx!BLVTffp>WB?i*`y=maJRyeNU`@283$r$C{R?dV8GU_RGq_FS zZMjS+zmv-8kyavQ1J&*gMt|HkYKUS7-(}la*?e%$DfcEI`a*UbwhO|zR_oSS((et3 z6&$&Ziq8KPWUN%43tT4@ob_@KFm`rvQ0q3#^-b!bHR^z6d4>2hOUl8Rj(OmZ9g~~g zgLVT#=Xu3~>7h(i@aPf4Zg*NWZZdL<+d)6+caTEiAiB$EYhEp+`wHo` zGja(1&;KU5YUq%vn`*BP7BA;jO;w|`%T>**sRsi^F~B=clN#>X(uX4Ser?8kqT#Q( z0z-H=?s8M)CH)JRh38WqfR3_vJDqCHQQ#v^wnxRBmKWHwfl!-2*fMn{l$ib3*j#XE zxv$?$Ztkb0cWdX@b_F)vF@H2Z>*#9RWU;h=@4iFoKbIW^!a9C-cu{i7efpm5 z5$dHN@Kx%U;sGe5wxg-!EGSy%f&D^#INud$Pk8|fofp5s-?YpEy1xV4Z|dJ@Bh4$o zxxzuy-4nhzC77Y6alv~v&bAjCrD;>eH9%BFPmcJTXEd5x({bR{1Ay%ghW^sNl=k9d=(~6b2nDr9 z{?3dn)~*2Cmo?{_&|%1Beyi`?0~gxXd)_k9^hKx2^VzB=^UR>0tmga}Odj5&0^jw5 z3@Aq0XSIuOM;WRI*7V{B8S?Iwc4D#4uWb6CyDG_T+f1XpIjZe7t9RZtc2cqy?H<`U zpKDRYrI%Y7P4f`^qshY1PGb8juG_=(=T>#g7r0^Xbd~#y&8|Uvr;}Q{(1b&fAOS`8 z9a^zE>hdDzl4l8eoylR83ix32KFpX|e$U%1dD;V5miKF^5dxOXE{y;5+(+7%WcJuB zSAQB9oV-lks>EHE;Z6+qY(MP246uy6DZRs6aN^+NWn03}ARW=m-PQERNkHD370j6& z{DYMC@dvtqEuAUvO|sZlC=7dg#5Qvh;On&os|cyI#-tQvQpdYdoaTAnX#JUDIDb8w ziA?&GZ8^>$FpJb>+tb?f-WJ(T+7&^bFU*8x0DCmv5%VL+5az{MH<@M^H+nLdWa~7u+DTm*W2Ez>Jf0DF)1vc0OsTRJeJ^D zKSB}mTZGo#1jub4|Dn)Quy1;$(JO(Q z+(^1|7QU5z&i$k~<$cI%s{II-0PfqM4YgdrxX2gdOJ!=~Hn(UmfI@h}(*^WIOW>(PCmU_S7wkP11odm5pX2 zxV8&mFp!-oTrXUqHp|T?O|YqGLcPOc6$F7$+RE-i{H!zP)`bQ7TNkbm%8&qlrgYyD zB$69lrfK+n+BC~OnW$qAphGVTa=w%4RzK;@xL>J{(BK^#pY!ehY)MnPYave2acdgC z^q`;q(sN7lL~Pq1YSBNi_UkkgB&nZgD~n&oHKrkvx@en37HqJuFXF9xFajZi%9pM$ z5e*tMv+*zp5mxUOd7JvTBm$Y>k=CYzRAc1xGy_Pm_Hy{%8Ye7 z9m83IsgCoFZ_Bz;$XqN)aylgty1rSwkTNU1JXQd`wu>+>$3Dvbd12?wErs)K%(^|` z+h+1n+ogR3#fSES>}}GIK?V1yrazN|UYE6{kNw{E1F8gzTCJ8nH{P>`=C?=vu%J1i7hJ*8r*2Zo* zpN$k{B5zoQ=sTthEN`BPui~Y~I3kK!#3Fm0Azvgh+tRbtk5Rf|6Jb5o`b_&=UAj5X ztbc4OnArSnE6V?4({nAuYd(M(f!7S!W3vH6n72I35xTxsf2Wlic>T2~C_0qxu^fs0 z^b4U^{k+}b7CpFQQb5+TH1WSqqHQiE^gPhaYImSNi-88XDj2Jeg6Ngd)?hotgsxXJ zmi)t*xhJ=khA{7^qGdr?J^lKOcxo*Pe#b|ubyqUl1%5LHJ($)fj`mD~ZEYsHH@KH8 z4bJ93HSs?B-&+Mgq_bBCEz3J@sv;evro%w_>zNM;524EXIqnP&itwd#Wqor=EyH=; z8{6-fBUOtf?yvw(lEN!{+WpNY|;BpkcB z;LD*7=4fEJAMp|VNVUx28azV%0myg_IH~XCq5ez49&su#pY-SHSLL5&X4*ZXkzeD1 zyY>7-4y*WOVU>+D;(?sRlL!gYGLdfTSn_02IS(}L-z zR5;#eAGWd5#Gz}FM5O`xu~vw?QZr#xeUrUGFPvP`>f^KmVEB=_gc&;nhf=iP$v4b& z))haZ%Jt*`W-g(t{nA;tN`YHV(5uy=4T1Aud(5GTs|&-$-4`rB3fBOQyp__Sl-XFN z^E!A*U`Mpir5K*M=4p%!yUHYQcgnuK#1qa0rVC%W=y~l?gS52pp3uMrtS@vZdzX~N zhyZw0SZv57iLIi(%+?g^zzA7JFYGZp9Wg%KCce=|b~gxWru3hK^s}9-2f*PqYS%@_ zjbr2eJtp0?ypgjqMv&eXJV4*6t}ItyPI-}P2Dhjx@`9M4o($n{UGPgQjAM;G1;>b} ze33U8Tk@st^Rc!?*J!)*b_C-y84&UsEJWAl~{5apMbfstq9n}ntjhp0eZDM|280JSgknmq7b znUHIEA0HtG!o7wJf97G1ILP!*0;uV84WBME9||f4FPkGwOG8C5{^6y`8G~?_OfY(A5T)x^ zU;Ru>3*p8%p7iwtX3AfV&eO76R;YYqFlnvYu*=BlT~SCR=;VZK1YV5X6Aa~zRjy3&?XWxA=NKXBxM zD#qa?9zn@kp>cgb=if@=lklX{^#ZAX4|kM4*gP&)=}15eEIWBC4hB-W-qe;dP(s$~ zOgmb4?F8d~7~7iT)hweLTX2L+R7VnT!mv&!Q`8|9I^}odOZMM*SJktZhUev?gyk-I zZF5;v`Pt%d!DiV<-;WM4Y8as!y%E|oX}o_jQyi$8sO$`9V8pdRy^Pg5!MHOHNEce( zsQ?k+a6$~o7v68L^3`efF1242I2Hfg63zu|gNvQQS!0$xBDsz^0Xp97z5&4VmFcdW z*H=uSE^ctq%g`O51)wyoO=t9=!m*D zCeKh%Yb%5b?L|A#E;vy1k#=Pw)!k~NfWm~u+Ep-tNnPP!u2ZWeiYsWfMt1TTo|XXd zs~V?CH2Xt=lZIAOnP`3C;Wo#+d_6G8SM?)q0$BOpiy6i&61TNCf1geXH%zqtp6Ltf zr0R}kt9Vf&)QFNM$NoFlYQbq^L=e@*02F>oBN@xNPZdSdD4`rKC(rnuw$}QU-x^K0 zhE#teWcL#U4JueF^DLY%{LkGk%o{ZM+iVr?OrC_=`21~3)6*nExV!OcP4U}|?ag+B z0Zo5%4?#3&jk0tY9yc_|JvdAbd&1OT1Fpia5y~(0pdPLzxp7!5cG+hKX4HC6jMMaT zu#q;$)q;5~6w7USvAwL17l-l!4~at+OC*6vUEt5Bx(II6w(M4x2TdD|$V&&_SvCITK3(XV@9nEM8DX zkyv(4H$ejekm4!|L2d47g6nZ)llw3Um}E-T|EOt)s+lbd?#1zXFIvX>AhuMOmc@)= zPflc4mI?xx9JFHQJD&A1t23dp5z`@co-hw79qcVfrKuJlK<%Uq<+1dQ)`|9`Y_9!# zJyIQ*X1LsJ8&N*}i!l6{7Z}g=H^P55W(y``6pf)pDji`=-k1m~T6E4G;N5*jq@o74OG4*Vz8PW+TVhl#{fz z-C@`f<&cr_J$;dNJ!|W3Zxh=b%iXQQReSLD6HRErQPFo8YhHXx z5!G;AU=d^0$|-vd?&BT*ks3PP`&=fT0y4sVy2U#_78g6S9;%b-$@hgCxEH*q=`Ef% z&mQX%38eJlz6M3Xy=t49HG)pV255A20_| z%9a7`V1SR`$Q*od5 zg~snBzAm(gw2MR%d(4hnLm~|yjE0!uvvp5V@4R7k4REQjpHL6_L@%>MroH9IErK87 z8}Q}YuTxwjAL*Sm_3_t-rCQP~{N=nHJWR0Z~lDyxjO^z%yIhqGNOrl%2{ry>B)3V4G_h7z;A z=cv4}LOuWZ62Lp&M!rgq2J~6LzmA!_Z&+umZC7}Nz`Yjgg8>!+NEp^VSB4w7vhqom zqi3@h`TAzaI6cXhD4+=XkxHWv7qFGtt(^8}A~o@%1VV7YrDB;trtcf0Xu(lqrHi&N zetC7a%_~7HH#zh0iyT2y*uY&MbW2*VRtU7ZXdfBLVbcwBWEZX1(Jla&`iwm6g3J_J z6_o8gw(0*QH2m}#vE5FJ zER)PffI23*S^u@j{J2o7`ioh8FP-AB*>Y%`?S01%8>xXL@HZ5GHZ!-4_mXOOeSkjx z$Dp}s8e*Ol@5h>qB<_+CV6$EAqDF=DAE_nd&MV*}Lie7`kxO)eXpLaT>Wse?>%q`9kCyJ1OgI*4(FEpBER$X|GfuC^Mq(1T~WZBbHb7D zy9!5^zbGvMoQ51x4@CF~I+0aSB|wYzyiM!i?GSWZ!mQzeCm0{ zoY_&!+fiJ5GTiZ$(iaonM~eKyuav0d3N7P~@$N$HN#;#M2lXeUsiZs4RqM>-{tdW~ zJJReU;N$@vGLZozm^gHySfOTV9}F*uDW$M>oi8u@EzpLPw!;5=kCmIF2+8RN2>Zh_ z+Am+|-&trrR+00+BByL`@M3UL?cDl8yknP18MdX=3SS9I5TT@Q0Ar0i486V53vBMk z6Vh%zp?-ugXVvZWA#X<*zCogaFbfOu@(;N?so9_J6vi;HZ&Uvky zP=4$CJ*0=uPc%E@t^4{<->zi7+P}EgjoF=r&UpZh=&6`Xgm#Cu--ouyj1%Sf_ZN zR?gloePnMfm@9xS8c|!p2HmRTfwBi6^HmjB5zr}R_>6+{DzSWl#rpn`dyR8bz zkWey)${D^~(PeA+K147+3{any$S=GOMOZ#B?{avx{9$j;_fk%9; z&=}D$vQm_pzIUwHxEZ}4Jj)%0T36Vry~fTg3S$;9gVetX%~w9%)bg`(J;B|yE0Z?3 z$d+9}e%`$TZ0s3uI3-*=>F^kSvbvs=tS5%W84W?;=7cHW512q|R}~lF zbK7XSsqigr-al~HX)+wlfnUg*-G($zHnUH@+#DAn?1Yv!ya@K-N_}gYV@H$}?UGe| z4)x`rCcN+#%3RazMKt!FGATHen#uu63CTjjy_2N)zN!UtsI^3vwPA!4$psuh_evlvAUhc-dPf$U;_xCwFQzrIy{M_}R(C~?|ujrO*M^R}ght@zG`!L!XGH`*O#OS6P zw!w=5en6)_^q^6$7|lh^uO$VFohbM=;~=t1J2*;RC}sf=+CHqV+aV~{r$Ngpdy7g4 zRXY89f~n}qwlY1@Kb^JOGQ#}xXRlnVb-n@{V7Ss$#dv=-W`t!xt9akovc-NPT-0NZ{caWf~l zf%Uql!P{j`O{I*s+Ch@Ilc>~Y<_1<-`F=u&02@yWCT8B>`Gu*sD z^;omYUE6Tw)5ds1l%pZfjj%zf8F>>nqMj=%RTn+ZlUi@aI!W3F=2{KRrzyyV)+TvX zI5F8>Im2n=XDvvRgWe+DDe`^3PQx0sW4RsT17t^>GEGv**M~EH4T26>H_T(w3Wx!W z-wA_r59W@E^KQ%E?>(vE%T)16t+OBQAN~J|l&a#ds+O-}U9L5uzMRbj92LhNj&!=+_2m|_q%yG9Ea-Je zfX?nrz5sj~l*=nUE-W%xcW8Jb(>^_~bUw!zk9?3K1RkCj(ZVdMAvz+gh2i@r{#q^n zOf!r@&4Dsgepp~&t+2lb&H_TGYb;Sd4B^1#4uf|-R!cik@7oI(-VL+W2y9jHlTa7I z1*v|&(e0TD6wX||-#a}DFm;Fs+glH@B(Gy`!b4iU4;i^{O5w#K8rDGfNSE04_idCu z+zmJV9q&5Bx&>&*WB9&Z>>udv(dw$%oAP=NI$bduXc4aRn0$nlOiC~te3wYiWE`?l7yj%Lvub(tH6$o4t5)TV z)F01;pGNWpBDHLsx4AFtYIzp;usV1Z)dY8pc3il4Cwq_l!i@F{cwOZqu*~E)o7;ZN zg86MWB#-w8|L#V>2Nh|9C}W+{aIdka!V1gyi2JaP>??U?t9_WMxli4OKTVgGw)U0b{hiq*&`8ZoUOr@DpFJY4TnPxU_2494}uZwouR(q znxmaGoOIee5*da0>fs$1VQ{tc-I#?)YlXGmoCrNSzVSxfC8B26V%}S|gze47YkNnj z^orq}_k*l|B@jX5{>8hit!bgTIiDPWt?$qJaPm@%3$f2_So!aS?aZE-*MnioFfhc= zxF6H@`}lZtD1=6PlQ4b9`(w4&_j}y1`452RshWXXi?jDF(S%7zN|#j5VY_riX6r&JUb@lA4F3doo+WQA49GHA__4Zg?m8T36ND2i!uNtO4E}{~<#Tv;j7%TfBzR^o!8orv zd2-V69o+=AsB&F6aynFf z9J+ey>anXAMdic4K$ot|NA*>`U3p#koc*1h>F;Rg(D)brM*DYu3O{_rtKW3>1*#vY zJzV**tEPu`dTY7R*YQDXIO%w|gLgVYzmu2!e)N;~NWD*~aVGS*eZA|!xI=vS-Kj-Tt%o&FyT{WkuCKzQVLUcJZ9JN;pg zhb;g503wI?>CsF5O55R%a-3H7!G72idx6^jsozjP%pah6$DDd6Jkp(1{QBNJsP&4z zZXQComQUvu#4C-SAbzH_zH8rwb3Vm67}$ktJoGhrud0i8;d-yj>7#mZ4~_FZXuXFP z?r~wA|G~HGdEmx*A?d@ve%Q-L4*Q$zTSDJ7?;)_?2|xRtl!tqFoJU4Gd|cgABVYfjSosIq%PWK9qL8I?MO|)Z=%vUVp#VXHnh{7C-rfoc~txM>#G?c^4d@ z$9_(}Yjjnaq z^eZ~__``gM=@a=r(*(i~BFE`VxyJgpFxt`k3N=3TygK#%P`=Of(Fz~+Q%HHkk3LrR z81<3wO#P9!CrJ8H?`2*`<8it~>1POgR0F(&R^dVn5mw?H8oo zQl7D|m~wuz(Ajc_6pbZW1lzM$p%gQ*mo=>9DZ=@M^?V*3nCX9eoaTckNxKK zxA$a!^X~MI-zX#=RQbpS38(+3Ug(#wKkbkH1N-7v+JAxcUus`)$GCr+dhbzb-ffKh zZdLDr5)K{t9iH==BT5eGkQe>^#=E1T-;wqICDKPZj;Q~V5g-0@RL=hnFN}IQEaz`G zFXX-2kbC3O^P|^gu84yyckSAK*idzNgFonu7mNE;#T`roPXr z^@3dN$h(_6#+%!oQ(uFVM6DomaT=n|@i_p^oo557KcRdd$a{h(74cJQAAu4D(p}ee~CP6G%Mr zT>i*+^C+EXK|6hw&mr>NJehD1ex2XxdIV(tNxb%l-TYeXQTep~onJWAd9&(GxwTxB zM^X8y7p)(yH}wbnN5eaLPQKELx?WQ3{8IhV<>&n0VJFX(+qE0wHQd=n^&{j&Rccd zfo5KZULfP;*v~1X9H7R#aTR_YR}~qrV?2FZ>U(7&>kZ;rpNxA1ON7tqullM!=sWhk z(m$>$J)y}5L{CM>r}7*Ps@zU5?C10&TDL>G{^*My)SDu9+I60HMad<9^6gOVGwus5!yQKZdqb_Ku@mx< z=j5t=k%L_94UYZsC64=e**;#sq!oz=*;gfBmp`cS(8#BJlm|VbIalC3A?ojO^_@Ey ze)RY{A0Zw5(D3uUJn@Pu7u4_k^*d$GG4$R!=Ovt{aP9(fk6e-SoUqH&N>12qCrP_<2z)&1BiS#zNh26+Q{em^&JIv$8PAuc~0~X zFI;{vt>2Sy-ly*~fSh+JhI$+CGC(Vz#%ntDbCnOG59vVlTWFQ5{;%o6{@Zol7k1vN zcO=4|kE(VS?fcR7|98?yKmEuGkNW%m`ks;e$(Qr_GxJ Date: Thu, 25 Dec 2025 05:23:57 +0000 Subject: [PATCH 05/63] Use pre-downloaded california housing data --- test/automl/test_python_log.py | 2 +- test/automl/test_regression.py | 12 ++++++------ test/automl/test_score.py | 2 +- test/automl/test_training_log.py | 2 +- test/default/test_defaults.py | 12 ++++++------ test/object_store.py | 2 +- test/reg.py | 2 +- test/spark/test_exceptions.py | 2 +- test/tune_example.py | 4 ++-- 9 files changed, 20 insertions(+), 20 deletions(-) diff --git a/test/automl/test_python_log.py b/test/automl/test_python_log.py index e18e33634b..6913064948 100644 --- a/test/automl/test_python_log.py +++ b/test/automl/test_python_log.py @@ -38,7 +38,7 @@ def test_logging_level(self): "keep_search_state": True, "learner_selector": "roundrobin", } - X_train, y_train = fetch_california_housing(return_X_y=True) + X_train, y_train = fetch_california_housing(return_X_y=False, data_home="test") n = len(y_train) >> 1 print(automl.model, automl.classes_, automl.predict(X_train)) automl.fit( diff --git a/test/automl/test_regression.py b/test/automl/test_regression.py index 892ad1eceb..9f5aa3a04a 100644 --- a/test/automl/test_regression.py +++ b/test/automl/test_regression.py @@ -47,7 +47,7 @@ def test_regression(self): "n_jobs": 1, "model_history": True, } - X_train, y_train = fetch_california_housing(return_X_y=True) + X_train, y_train = fetch_california_housing(return_X_y=False, data_home="test") n = int(len(y_train) * 9 // 10) automl.fit(X_train=X_train[:n], y_train=y_train[:n], X_val=X_train[n:], y_val=y_train[n:], **automl_settings) assert automl._state.eval_method == "holdout" @@ -141,7 +141,7 @@ def test_parallel(self, hpo_method=None): "n_concurrent_trials": 10, "hpo_method": hpo_method, } - X_train, y_train = fetch_california_housing(return_X_y=True) + X_train, y_train = fetch_california_housing(return_X_y=False, data_home="test") try: automl_experiment.fit(X_train=X_train, y_train=y_train, **automl_settings) print(automl_experiment.predict(X_train)) @@ -268,7 +268,7 @@ def test_reproducibility_of_regression_models(estimator: str): "skip_transform": True, "retrain_full": True, } - X, y = fetch_california_housing(return_X_y=True, as_frame=True) + X, y = fetch_california_housing(return_X_y=True, as_frame=True, data_home="test") automl.fit(X_train=X, y_train=y, **automl_settings) best_model = automl.model assert best_model is not None @@ -314,7 +314,7 @@ def test_reproducibility_of_catboost_regression_model(): "skip_transform": True, "retrain_full": True, } - X, y = fetch_california_housing(return_X_y=True, as_frame=True) + X, y = fetch_california_housing(return_X_y=True, as_frame=True, data_home="test") automl.fit(X_train=X, y_train=y, **automl_settings) best_model = automl.model assert best_model is not None @@ -360,7 +360,7 @@ def test_reproducibility_of_lgbm_regression_model(): "skip_transform": True, "retrain_full": True, } - X, y = fetch_california_housing(return_X_y=True, as_frame=True) + X, y = fetch_california_housing(return_X_y=True, as_frame=True, data_home="test") automl.fit(X_train=X, y_train=y, **automl_settings) best_model = automl.model assert best_model is not None @@ -424,7 +424,7 @@ def test_reproducibility_of_underlying_regression_models(estimator: str): "skip_transform": True, "retrain_full": False, } - X, y = fetch_california_housing(return_X_y=True, as_frame=True) + X, y = fetch_california_housing(return_X_y=True, as_frame=True, data_home="test") automl.fit(X_train=X, y_train=y, **automl_settings) best_model = automl.model assert best_model is not None diff --git a/test/automl/test_score.py b/test/automl/test_score.py index 2976daadee..0df480ae7a 100644 --- a/test/automl/test_score.py +++ b/test/automl/test_score.py @@ -142,7 +142,7 @@ def test_classification(self): def test_regression(self): automl_experiment = AutoML() - X_train, y_train = fetch_california_housing(return_X_y=True) + X_train, y_train = fetch_california_housing(return_X_y=False, data_home="test") n = int(len(y_train) * 9 // 10) for each_estimator in [ diff --git a/test/automl/test_training_log.py b/test/automl/test_training_log.py index cd8db81182..6960fafc69 100644 --- a/test/automl/test_training_log.py +++ b/test/automl/test_training_log.py @@ -30,7 +30,7 @@ def test_training_log(self, path="test_training_log.log", estimator_list="auto", "keep_search_state": True, "estimator_list": estimator_list, } - X_train, y_train = fetch_california_housing(return_X_y=True) + X_train, y_train = fetch_california_housing(return_X_y=False, data_home="test") automl.fit(X_train=X_train, y_train=y_train, **automl_settings) # Check if the training log file is populated. self.assertTrue(os.path.exists(filename)) diff --git a/test/default/test_defaults.py b/test/default/test_defaults.py index d8be7b61b5..acf50e4ea9 100644 --- a/test/default/test_defaults.py +++ b/test/default/test_defaults.py @@ -60,7 +60,7 @@ def test_housing(as_frame=True): "starting_points": "data", "max_iter": 0, } - X_train, y_train = fetch_california_housing(return_X_y=True, as_frame=as_frame) + X_train, y_train = fetch_california_housing(return_X_y=True, as_frame=as_frame, data_home="test") automl.fit(X_train, y_train, **automl_settings) @@ -115,7 +115,7 @@ def test_suggest_classification(): def test_suggest_regression(): location = "test/default" - X_train, y_train = fetch_california_housing(return_X_y=True, as_frame=True) + X_train, y_train = fetch_california_housing(return_X_y=True, as_frame=True, data_home="test") suggested = suggest_hyperparams("regression", X_train, y_train, "lgbm", location=location) print(suggested) suggested = preprocess_and_suggest_hyperparams("regression", X_train, y_train, "xgboost", location=location) @@ -137,7 +137,7 @@ def test_rf(): print(rf) location = "test/default" - X_train, y_train = fetch_california_housing(return_X_y=True, as_frame=True) + X_train, y_train = fetch_california_housing(return_X_y=True, as_frame=True, data_home="test") rf = RandomForestRegressor(default_location=location) rf.fit(X_train[:100], y_train[:100]) rf.predict(X_train) @@ -155,7 +155,7 @@ def test_extratrees(): print(classifier) location = "test/default" - X_train, y_train = fetch_california_housing(return_X_y=True, as_frame=True) + X_train, y_train = fetch_california_housing(return_X_y=True, as_frame=True, data_home="test") regressor = ExtraTreesRegressor(default_location=location) regressor.fit(X_train[:100], y_train[:100]) regressor.predict(X_train) @@ -175,7 +175,7 @@ def test_lgbm(): print(classifier.classes_) location = "test/default" - X_train, y_train = fetch_california_housing(return_X_y=True, as_frame=True) + X_train, y_train = fetch_california_housing(return_X_y=True, as_frame=True, data_home="test") regressor = LGBMRegressor(default_location=location) regressor.fit(X_train, y_train) regressor.predict(X_train) @@ -194,7 +194,7 @@ def test_xgboost(): print(classifier.classes_) location = "test/default" - X_train, y_train = fetch_california_housing(return_X_y=True, as_frame=True) + X_train, y_train = fetch_california_housing(return_X_y=True, as_frame=True, data_home="test") regressor = XGBRegressor(default_location=location) regressor.fit(X_train[:100], y_train[:100]) regressor.predict(X_train) diff --git a/test/object_store.py b/test/object_store.py index 6d32237b48..1b3f98502d 100644 --- a/test/object_store.py +++ b/test/object_store.py @@ -7,7 +7,7 @@ from flaml import tune from flaml.automl.model import LGBMEstimator -data = fetch_california_housing(return_X_y=False, as_frame=True) +data = fetch_california_housing(return_X_y=False, as_frame=True, data_home="test") X, y = data.data, data.target X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=42) X_train_ref = ray.put(X_train) diff --git a/test/reg.py b/test/reg.py index 795ab1c5ce..4dfbcab2a2 100644 --- a/test/reg.py +++ b/test/reg.py @@ -11,7 +11,7 @@ "task": "regression", "log_file_name": "test/california.log", } -X_train, y_train = fetch_california_housing(return_X_y=True) +X_train, y_train = fetch_california_housing(return_X_y=False, data_home="test") # Train with labeled input data automl.fit(X_train=X_train, y_train=y_train, **automl_settings) print(automl.model) diff --git a/test/spark/test_exceptions.py b/test/spark/test_exceptions.py index 13c265d37c..26075fa005 100644 --- a/test/spark/test_exceptions.py +++ b/test/spark/test_exceptions.py @@ -22,7 +22,7 @@ def base_automl(n_concurrent_trials=1, use_ray=False, use_spark=False, verbose=0 except (ServerError, Exception): from sklearn.datasets import fetch_california_housing - X_train, y_train = fetch_california_housing(return_X_y=True) + X_train, y_train = fetch_california_housing(return_X_y=False, data_home="test") automl = AutoML() settings = { "time_budget": 3, # total running time in seconds diff --git a/test/tune_example.py b/test/tune_example.py index 9b4ba68fe4..592f94762a 100644 --- a/test/tune_example.py +++ b/test/tune_example.py @@ -6,12 +6,12 @@ from flaml import tune from flaml.automl.model import LGBMEstimator -data = fetch_california_housing(return_X_y=False, as_frame=True) +data = fetch_california_housing(return_X_y=False, as_frame=True, data_home="test") df, X, y = data.frame, data.data, data.target df_train, _, X_train, X_test, _, y_test = train_test_split(df, X, y, test_size=0.33, random_state=42) csv_file_name = "test/housing.csv" df_train.to_csv(csv_file_name, index=False) -# X, y = fetch_california_housing(return_X_y=True, as_frame=True) +# X, y = fetch_california_housing(return_X_y=True, as_frame=True, data_home="test") # X_train, X_test, y_train, y_test = train_test_split( # X, y, test_size=0.33, random_state=42 # ) From 519cc5e2c00b4759ea039c99e3eced2a33981ae5 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 25 Dec 2025 05:31:49 +0000 Subject: [PATCH 06/63] Pin lightning<=2.5.6 --- setup.py | 1 + 1 file changed, 1 insertion(+) diff --git a/setup.py b/setup.py index 31cc563726..0d919e077d 100644 --- a/setup.py +++ b/setup.py @@ -94,6 +94,7 @@ "sympy", "wolframalpha", "dill", # a drop in replacement of pickle + "lightning<=2.5.6", ], "catboost": [ "catboost>=0.26,<1.2; python_version<'3.11'", From 229826b351cec613328880a531941ec03bba1375 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 25 Dec 2025 05:42:12 +0000 Subject: [PATCH 07/63] Fix typo in find and replace --- test/automl/test_python_log.py | 2 +- test/automl/test_regression.py | 4 ++-- test/automl/test_score.py | 2 +- test/automl/test_training_log.py | 2 +- test/reg.py | 2 +- test/spark/test_exceptions.py | 2 +- 6 files changed, 7 insertions(+), 7 deletions(-) diff --git a/test/automl/test_python_log.py b/test/automl/test_python_log.py index 6913064948..b3f141780d 100644 --- a/test/automl/test_python_log.py +++ b/test/automl/test_python_log.py @@ -38,7 +38,7 @@ def test_logging_level(self): "keep_search_state": True, "learner_selector": "roundrobin", } - X_train, y_train = fetch_california_housing(return_X_y=False, data_home="test") + X_train, y_train = fetch_california_housing(return_X_y=True, data_home="test") n = len(y_train) >> 1 print(automl.model, automl.classes_, automl.predict(X_train)) automl.fit( diff --git a/test/automl/test_regression.py b/test/automl/test_regression.py index 9f5aa3a04a..daa5f3830d 100644 --- a/test/automl/test_regression.py +++ b/test/automl/test_regression.py @@ -47,7 +47,7 @@ def test_regression(self): "n_jobs": 1, "model_history": True, } - X_train, y_train = fetch_california_housing(return_X_y=False, data_home="test") + X_train, y_train = fetch_california_housing(return_X_y=True, data_home="test") n = int(len(y_train) * 9 // 10) automl.fit(X_train=X_train[:n], y_train=y_train[:n], X_val=X_train[n:], y_val=y_train[n:], **automl_settings) assert automl._state.eval_method == "holdout" @@ -141,7 +141,7 @@ def test_parallel(self, hpo_method=None): "n_concurrent_trials": 10, "hpo_method": hpo_method, } - X_train, y_train = fetch_california_housing(return_X_y=False, data_home="test") + X_train, y_train = fetch_california_housing(return_X_y=True, data_home="test") try: automl_experiment.fit(X_train=X_train, y_train=y_train, **automl_settings) print(automl_experiment.predict(X_train)) diff --git a/test/automl/test_score.py b/test/automl/test_score.py index 0df480ae7a..7e9abc864a 100644 --- a/test/automl/test_score.py +++ b/test/automl/test_score.py @@ -142,7 +142,7 @@ def test_classification(self): def test_regression(self): automl_experiment = AutoML() - X_train, y_train = fetch_california_housing(return_X_y=False, data_home="test") + X_train, y_train = fetch_california_housing(return_X_y=True, data_home="test") n = int(len(y_train) * 9 // 10) for each_estimator in [ diff --git a/test/automl/test_training_log.py b/test/automl/test_training_log.py index 6960fafc69..0d96284732 100644 --- a/test/automl/test_training_log.py +++ b/test/automl/test_training_log.py @@ -30,7 +30,7 @@ def test_training_log(self, path="test_training_log.log", estimator_list="auto", "keep_search_state": True, "estimator_list": estimator_list, } - X_train, y_train = fetch_california_housing(return_X_y=False, data_home="test") + X_train, y_train = fetch_california_housing(return_X_y=True, data_home="test") automl.fit(X_train=X_train, y_train=y_train, **automl_settings) # Check if the training log file is populated. self.assertTrue(os.path.exists(filename)) diff --git a/test/reg.py b/test/reg.py index 4dfbcab2a2..1b758e2ca9 100644 --- a/test/reg.py +++ b/test/reg.py @@ -11,7 +11,7 @@ "task": "regression", "log_file_name": "test/california.log", } -X_train, y_train = fetch_california_housing(return_X_y=False, data_home="test") +X_train, y_train = fetch_california_housing(return_X_y=True, data_home="test") # Train with labeled input data automl.fit(X_train=X_train, y_train=y_train, **automl_settings) print(automl.model) diff --git a/test/spark/test_exceptions.py b/test/spark/test_exceptions.py index 26075fa005..63a22e625a 100644 --- a/test/spark/test_exceptions.py +++ b/test/spark/test_exceptions.py @@ -22,7 +22,7 @@ def base_automl(n_concurrent_trials=1, use_ray=False, use_spark=False, verbose=0 except (ServerError, Exception): from sklearn.datasets import fetch_california_housing - X_train, y_train = fetch_california_housing(return_X_y=False, data_home="test") + X_train, y_train = fetch_california_housing(return_X_y=True, data_home="test") automl = AutoML() settings = { "time_budget": 3, # total running time in seconds From d4d1e2ef8ccdb6e3ea04c2922a33edd2b5de695b Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 25 Dec 2025 11:47:01 +0000 Subject: [PATCH 08/63] Fix estimators has no attribute __sklearn_tags__ --- flaml/automl/model.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/flaml/automl/model.py b/flaml/automl/model.py index 298f3cab80..53a92ece24 100644 --- a/flaml/automl/model.py +++ b/flaml/automl/model.py @@ -111,7 +111,7 @@ def limit_resource(memory_limit, time_limit): pass -class BaseEstimator: +class BaseEstimator(sklearn.base.ClassifierMixin, sklearn.base.BaseEstimator): """The abstract class for all learners. Typical examples: From 6c731ecf47de5c7b02d1cc0ce36d2f97e9c6f0d3 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 25 Dec 2025 11:47:26 +0000 Subject: [PATCH 09/63] Pin torch to 2.2.2 in tests --- setup.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/setup.py b/setup.py index 0d919e077d..6bd73d55ba 100644 --- a/setup.py +++ b/setup.py @@ -62,7 +62,7 @@ "pytest>=6.1.1", "coverage>=5.3", "pre-commit", - "torch", + "torch==2.2.2", "torchvision", "catboost>=0.26,<1.2; python_version<'3.11'", "catboost>=0.26; python_version>='3.11'", @@ -86,7 +86,7 @@ "nbconvert", "nbformat", "ipykernel", - "pytorch-lightning<1.9.1", # test_forecast_panel + "pytorch-lightning==2.2.2", # test_forecast_panel "tensorboardX==2.6", # test_forecast_panel "requests<2.29.0", # https://github.com/docker/docker-py/issues/3113 "packaging", From 5f1fe8e4f72e0f252640ee6c5358eb68c5fb1a78 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 25 Dec 2025 11:50:28 +0000 Subject: [PATCH 10/63] Fix conflict --- setup.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/setup.py b/setup.py index 6bd73d55ba..64ef08c9a6 100644 --- a/setup.py +++ b/setup.py @@ -86,7 +86,7 @@ "nbconvert", "nbformat", "ipykernel", - "pytorch-lightning==2.2.2", # test_forecast_panel + "pytorch-lightning<1.9.1", # test_forecast_panel "tensorboardX==2.6", # test_forecast_panel "requests<2.29.0", # https://github.com/docker/docker-py/issues/3113 "packaging", From f59b66771e33624a5efc649b010db4ed2e7c39b2 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 25 Dec 2025 12:17:42 +0000 Subject: [PATCH 11/63] Update pytorch-forecasting --- setup.py | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/setup.py b/setup.py index 64ef08c9a6..8d12d1caed 100644 --- a/setup.py +++ b/setup.py @@ -78,8 +78,7 @@ "rouge_score", "hcrystalball==0.1.10", "seqeval", - "pytorch-forecasting>=0.9.0,<=0.10.1; python_version<'3.11'", - # "pytorch-forecasting==0.10.1; python_version=='3.11'", + "pytorch-forecasting", "mlflow==2.15.1", "joblibspark>=0.5.0", "joblib<=1.3.2", From 4173a9b0cecea774e24d1f8eab30ee063be06ea7 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Fri, 26 Dec 2025 01:02:38 +0000 Subject: [PATCH 12/63] Update pytorch-forecasting --- setup.py | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/setup.py b/setup.py index 8d12d1caed..e0d09ed39c 100644 --- a/setup.py +++ b/setup.py @@ -78,18 +78,18 @@ "rouge_score", "hcrystalball==0.1.10", "seqeval", - "pytorch-forecasting", + "pytorch-forecasting==0.10.2", "mlflow==2.15.1", "joblibspark>=0.5.0", "joblib<=1.3.2", "nbconvert", "nbformat", "ipykernel", - "pytorch-lightning<1.9.1", # test_forecast_panel + "pytorch-lightning", # test_forecast_panel "tensorboardX==2.6", # test_forecast_panel "requests<2.29.0", # https://github.com/docker/docker-py/issues/3113 "packaging", - "pydantic==1.10.9", + # "pydantic==1.10.9", "sympy", "wolframalpha", "dill", # a drop in replacement of pickle From 12106d0b6e9404084a479c73876165a80593e89c Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Fri, 26 Dec 2025 01:04:44 +0000 Subject: [PATCH 13/63] Update pytorch-forecasting --- setup.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/setup.py b/setup.py index e0d09ed39c..e737649984 100644 --- a/setup.py +++ b/setup.py @@ -78,7 +78,8 @@ "rouge_score", "hcrystalball==0.1.10", "seqeval", - "pytorch-forecasting==0.10.2", + "pytorch-forecasting<1.0.0; python_version>='3.11'", + "pytorch-forecasting>=0.9.0; python_version<'3.11'", "mlflow==2.15.1", "joblibspark>=0.5.0", "joblib<=1.3.2", From 022d64a6b6293898d9df1c13e5c153aa13fae3dd Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Fri, 26 Dec 2025 01:14:23 +0000 Subject: [PATCH 14/63] Use numpy<2 for testing --- setup.py | 1 + 1 file changed, 1 insertion(+) diff --git a/setup.py b/setup.py index e737649984..65f5c71a02 100644 --- a/setup.py +++ b/setup.py @@ -51,6 +51,7 @@ "joblib<=1.3.2", ], "test": [ + "numpy<2.0.0", "jupyter", "lightgbm>=2.3.1", "xgboost>=0.90,<2.0.0", From 28afa46281e3593381b4b3dff62a4d7de667ed9e Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Fri, 26 Dec 2025 01:19:12 +0000 Subject: [PATCH 15/63] Update scikit-learn --- setup.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/setup.py b/setup.py index 65f5c71a02..8d71c6ef49 100644 --- a/setup.py +++ b/setup.py @@ -58,7 +58,7 @@ "scipy>=1.4.1", "pandas>=1.1.4,<2.0.0; python_version<'3.10'", "pandas>=1.1.4; python_version>='3.10'", - "scikit-learn>=1.0.0", + "scikit-learn>=1.2.0", "thop", "pytest>=6.1.1", "coverage>=5.3", From 5915ae0acecd1a336a90c5c61945093633451ed1 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Fri, 26 Dec 2025 01:29:16 +0000 Subject: [PATCH 16/63] Run Build and UT every other day --- .github/workflows/python-package.yml | 3 +++ 1 file changed, 3 insertions(+) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index 38b98ff3dd..d8137ac8ac 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -22,6 +22,9 @@ on: - 'setup.py' merge_group: types: [checks_requested] + schedule: + # Every other day at 02:00 UTC + - cron: '0 2 */2 * *' permissions: {} concurrency: From 21d346aac80fd1b311fbef1847f30c4382de19da Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Fri, 26 Dec 2025 02:01:58 +0000 Subject: [PATCH 17/63] Pin pip<24.1 --- setup.py | 1 + 1 file changed, 1 insertion(+) diff --git a/setup.py b/setup.py index 8d71c6ef49..6334e57d4d 100644 --- a/setup.py +++ b/setup.py @@ -51,6 +51,7 @@ "joblib<=1.3.2", ], "test": [ + "pip<24.1", "numpy<2.0.0", "jupyter", "lightgbm>=2.3.1", From 768a8d8b12b0b292507ab4c999130a53a9cd409e Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Fri, 26 Dec 2025 02:06:29 +0000 Subject: [PATCH 18/63] Pin pip<24.1 in pipeline --- .github/workflows/python-package.yml | 2 +- setup.py | 1 - 2 files changed, 1 insertion(+), 2 deletions(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index d8137ac8ac..fe153e1c36 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -59,7 +59,7 @@ jobs: export LDFLAGS="$LDFLAGS -Wl,-rpath,/usr/local/opt/libomp/lib -L/usr/local/opt/libomp/lib -lomp" - name: Install packages and dependencies run: | - python -m pip install --upgrade pip wheel setuptools + python -m pip install --upgrade "pip<24.1" wheel setuptools pip install -e . python -c "import flaml" pip install -e .[test] diff --git a/setup.py b/setup.py index 6334e57d4d..8d71c6ef49 100644 --- a/setup.py +++ b/setup.py @@ -51,7 +51,6 @@ "joblib<=1.3.2", ], "test": [ - "pip<24.1", "numpy<2.0.0", "jupyter", "lightgbm>=2.3.1", From 843e860fdeff6388b791656b88dbf2e51263951d Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Fri, 26 Dec 2025 02:17:19 +0000 Subject: [PATCH 19/63] Loosen pip, install pytorch_forecasting only in py311 --- .github/workflows/python-package.yml | 2 +- setup.py | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index fe153e1c36..d8137ac8ac 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -59,7 +59,7 @@ jobs: export LDFLAGS="$LDFLAGS -Wl,-rpath,/usr/local/opt/libomp/lib -L/usr/local/opt/libomp/lib -lomp" - name: Install packages and dependencies run: | - python -m pip install --upgrade "pip<24.1" wheel setuptools + python -m pip install --upgrade pip wheel setuptools pip install -e . python -c "import flaml" pip install -e .[test] diff --git a/setup.py b/setup.py index 8d71c6ef49..912a1620d2 100644 --- a/setup.py +++ b/setup.py @@ -80,7 +80,7 @@ "hcrystalball==0.1.10", "seqeval", "pytorch-forecasting<1.0.0; python_version>='3.11'", - "pytorch-forecasting>=0.9.0; python_version<'3.11'", + # "pytorch-forecasting>=0.9.0; python_version<'3.11'", "mlflow==2.15.1", "joblibspark>=0.5.0", "joblib<=1.3.2", From dd5b99257aaa18a95ee0cf83b5e480859109cd58 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Wed, 7 Jan 2026 13:32:25 +0800 Subject: [PATCH 20/63] Add support to new versions of nlp dependecies --- flaml/automl/ml.py | 16 +++++++- flaml/automl/nlp/huggingface/training_args.py | 10 +++++ flaml/automl/time_series/tft.py | 37 ++++++++++++++++--- flaml/automl/time_series/ts_data.py | 12 +++++- setup.py | 25 ++++++------- 5 files changed, 79 insertions(+), 21 deletions(-) diff --git a/flaml/automl/ml.py b/flaml/automl/ml.py index bd13d8259e..3000e0d637 100644 --- a/flaml/automl/ml.py +++ b/flaml/automl/ml.py @@ -127,9 +127,23 @@ def metric_loss_score( import datasets datasets_metric_name = huggingface_submetric_to_metric.get(metric_name, metric_name.split(":")[0]) - metric = datasets.load_metric(datasets_metric_name, trust_remote_code=True) metric_mode = huggingface_metric_to_mode[datasets_metric_name] + # datasets>=3 removed load_metric; prefer evaluate if available + try: + import evaluate + + metric = evaluate.load(datasets_metric_name, trust_remote_code=True) + except Exception: + import datasets + + if hasattr(datasets, "load_metric"): + metric = datasets.load_metric(datasets_metric_name, trust_remote_code=True) + else: + from datasets import load_metric as _load_metric # older datasets + + metric = _load_metric(datasets_metric_name, trust_remote_code=True) + if metric_name.startswith("seqeval"): y_processed_true = [[labels[tr] for tr in each_list] for each_list in y_processed_true] elif metric in ("pearsonr", "spearmanr"): diff --git a/flaml/automl/nlp/huggingface/training_args.py b/flaml/automl/nlp/huggingface/training_args.py index 6a408b1a2d..8a71f3950e 100644 --- a/flaml/automl/nlp/huggingface/training_args.py +++ b/flaml/automl/nlp/huggingface/training_args.py @@ -77,6 +77,16 @@ class TrainingArgumentsForAuto(TrainingArguments): logging_steps: int = field(default=500, metadata={"help": "Log every X updates steps."}) + # Newer versions of HuggingFace Transformers may access `TrainingArguments.generation_config` + # (e.g., in generation-aware trainers/callbacks). Keep this attribute to remain compatible + # while defaulting to None for non-generation tasks. + generation_config: Optional[object] = field( + default=None, + metadata={ + "help": "Optional generation config (or path) used by generation-aware trainers." + }, + ) + @staticmethod def load_args_from_console(): from dataclasses import fields diff --git a/flaml/automl/time_series/tft.py b/flaml/automl/time_series/tft.py index c9ab30be1a..48520a8ba7 100644 --- a/flaml/automl/time_series/tft.py +++ b/flaml/automl/time_series/tft.py @@ -1,4 +1,5 @@ import time +import inspect try: import pandas as pd @@ -106,12 +107,17 @@ def transform_ds(self, X_train: TimeSeriesDataset, y_train, **kwargs): def fit(self, X_train, y_train, budget=None, **kwargs): import warnings - import pytorch_lightning as pl + try: + import lightning.pytorch as pl + from lightning.pytorch.callbacks import EarlyStopping, LearningRateMonitor + from lightning.pytorch.loggers import TensorBoardLogger + except ImportError: + import pytorch_lightning as pl + from pytorch_lightning.callbacks import EarlyStopping, LearningRateMonitor + from pytorch_lightning.loggers import TensorBoardLogger import torch from pytorch_forecasting import TemporalFusionTransformer from pytorch_forecasting.metrics import QuantileLoss - from pytorch_lightning.callbacks import EarlyStopping, LearningRateMonitor - from pytorch_lightning.loggers import TensorBoardLogger # a bit of monkey patching to fix the MacOS test # all the log_prediction method appears to do is plot stuff, which ?breaks github tests @@ -132,12 +138,26 @@ def log_prediction(*args, **kwargs): lr_logger = LearningRateMonitor() # log the learning rate logger = TensorBoardLogger(kwargs.get("log_dir", "lightning_logs")) # logging results to a tensorboard default_trainer_kwargs = dict( - gpus=self._kwargs.get("gpu_per_trial", [0]) if torch.cuda.is_available() else None, max_epochs=max_epochs, gradient_clip_val=gradient_clip_val, callbacks=[lr_logger, early_stop_callback], logger=logger, ) + + # PyTorch Lightning >=2.0 replaced `gpus` with `accelerator`/`devices`. + # Also, passing `gpus=None` is not accepted on newer versions. + trainer_sig_params = inspect.signature(pl.Trainer.__init__).parameters + if torch.cuda.is_available() and "gpus" in trainer_sig_params: + gpus = self._kwargs.get("gpu_per_trial", None) + if gpus is not None: + default_trainer_kwargs["gpus"] = gpus + elif torch.cuda.is_available() and "devices" in trainer_sig_params: + devices = self._kwargs.get("gpu_per_trial", None) + if devices == -1: + devices = "auto" + if devices is not None: + default_trainer_kwargs["accelerator"] = "gpu" + default_trainer_kwargs["devices"] = devices trainer = pl.Trainer( **default_trainer_kwargs, ) @@ -157,7 +177,14 @@ def log_prediction(*args, **kwargs): val_dataloaders=val_dataloader, ) best_model_path = trainer.checkpoint_callback.best_model_path - best_tft = TemporalFusionTransformer.load_from_checkpoint(best_model_path) + # PyTorch 2.6 changed `torch.load` default `weights_only` from False -> True. + # Some Lightning checkpoints (including those produced here) can require full unpickling. + # This path is generated locally during training, so it's trusted. + load_sig_params = inspect.signature(TemporalFusionTransformer.load_from_checkpoint).parameters + if "weights_only" in load_sig_params: + best_tft = TemporalFusionTransformer.load_from_checkpoint(best_model_path, weights_only=False) + else: + best_tft = TemporalFusionTransformer.load_from_checkpoint(best_model_path) train_time = time.time() - current_time self._model = best_tft return train_time diff --git a/flaml/automl/time_series/ts_data.py b/flaml/automl/time_series/ts_data.py index 2587a70e76..0709c63d5f 100644 --- a/flaml/automl/time_series/ts_data.py +++ b/flaml/automl/time_series/ts_data.py @@ -9,6 +9,7 @@ try: import pandas as pd from pandas import DataFrame, Series, to_datetime + from pandas.api.types import is_datetime64_any_dtype from scipy.sparse import issparse from sklearn.compose import ColumnTransformer from sklearn.impute import SimpleImputer @@ -390,8 +391,17 @@ def fit(self, X: Union[DataFrame, np.array], y): n = X.shape[0] assert len(self.num_columns) == 0, "Trying to call fit() twice, something is wrong" - + for column in X.columns: + # Never treat the time column as a feature for sklearn preprocessing + if column == self.time_col: + continue + + # Robust datetime detection (covers datetime64[ms/us/ns], tz-aware, etc.) + if is_datetime64_any_dtype(X[column]): + self.datetime_columns.append(column) + continue + # sklearn/utils/validation.py needs int/float values if X[column].dtype.name in ("object", "category", "string"): if ( diff --git a/setup.py b/setup.py index 912a1620d2..213d14caac 100644 --- a/setup.py +++ b/setup.py @@ -63,7 +63,7 @@ "pytest>=6.1.1", "coverage>=5.3", "pre-commit", - "torch==2.2.2", + "torch", "torchvision", "catboost>=0.26,<1.2; python_version<'3.11'", "catboost>=0.26; python_version>='3.11'", @@ -71,31 +71,28 @@ "optuna>=2.8.0,<=3.6.1", "openml", "statsmodels>=0.12.2", - "psutil==5.8.0", + "psutil", "dataclasses", - "transformers[torch]==4.26", - "datasets<=3.5.0", - "nltk<=3.8.1", # 3.8.2 doesn't work with mlflow + "transformers[torch]", + "datasets", + "evaluate", + "nltk", # 3.8.2 doesn't work with mlflow "rouge_score", - "hcrystalball==0.1.10", + "hcrystalball", "seqeval", - "pytorch-forecasting<1.0.0; python_version>='3.11'", + "pytorch-forecasting", # "pytorch-forecasting>=0.9.0; python_version<'3.11'", - "mlflow==2.15.1", + "mlflow<3.0.0", "joblibspark>=0.5.0", "joblib<=1.3.2", "nbconvert", "nbformat", "ipykernel", "pytorch-lightning", # test_forecast_panel - "tensorboardX==2.6", # test_forecast_panel - "requests<2.29.0", # https://github.com/docker/docker-py/issues/3113 + "tensorboardX", # test_forecast_panel + "requests", # https://github.com/docker/docker-py/issues/3113 "packaging", - # "pydantic==1.10.9", - "sympy", - "wolframalpha", "dill", # a drop in replacement of pickle - "lightning<=2.5.6", ], "catboost": [ "catboost>=0.26,<1.2; python_version<'3.11'", From 54e310185684240b42e27ccb91076ddc6f6e0b77 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Wed, 7 Jan 2026 13:52:37 +0800 Subject: [PATCH 21/63] Fix formats --- flaml/automl/nlp/huggingface/training_args.py | 4 +--- flaml/automl/time_series/tft.py | 2 +- flaml/automl/time_series/ts_data.py | 2 +- 3 files changed, 3 insertions(+), 5 deletions(-) diff --git a/flaml/automl/nlp/huggingface/training_args.py b/flaml/automl/nlp/huggingface/training_args.py index 8a71f3950e..383fc9f390 100644 --- a/flaml/automl/nlp/huggingface/training_args.py +++ b/flaml/automl/nlp/huggingface/training_args.py @@ -82,9 +82,7 @@ class TrainingArgumentsForAuto(TrainingArguments): # while defaulting to None for non-generation tasks. generation_config: Optional[object] = field( default=None, - metadata={ - "help": "Optional generation config (or path) used by generation-aware trainers." - }, + metadata={"help": "Optional generation config (or path) used by generation-aware trainers."}, ) @staticmethod diff --git a/flaml/automl/time_series/tft.py b/flaml/automl/time_series/tft.py index 48520a8ba7..fb660ce433 100644 --- a/flaml/automl/time_series/tft.py +++ b/flaml/automl/time_series/tft.py @@ -1,5 +1,5 @@ -import time import inspect +import time try: import pandas as pd diff --git a/flaml/automl/time_series/ts_data.py b/flaml/automl/time_series/ts_data.py index 0709c63d5f..0c7d255587 100644 --- a/flaml/automl/time_series/ts_data.py +++ b/flaml/automl/time_series/ts_data.py @@ -391,7 +391,7 @@ def fit(self, X: Union[DataFrame, np.array], y): n = X.shape[0] assert len(self.num_columns) == 0, "Trying to call fit() twice, something is wrong" - + for column in X.columns: # Never treat the time column as a feature for sklearn preprocessing if column == self.time_col: From ffb87f41a09d75616f6af6144c2bde5b5287c2e4 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Wed, 7 Jan 2026 13:55:54 +0800 Subject: [PATCH 22/63] Remove redefinition --- flaml/automl/ml.py | 2 -- 1 file changed, 2 deletions(-) diff --git a/flaml/automl/ml.py b/flaml/automl/ml.py index 3000e0d637..8ba05120e1 100644 --- a/flaml/automl/ml.py +++ b/flaml/automl/ml.py @@ -135,8 +135,6 @@ def metric_loss_score( metric = evaluate.load(datasets_metric_name, trust_remote_code=True) except Exception: - import datasets - if hasattr(datasets, "load_metric"): metric = datasets.load_metric(datasets_metric_name, trust_remote_code=True) else: From bc30b13cf142f6b04434a3077bd3a7427df53eee Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Wed, 7 Jan 2026 14:28:19 +0800 Subject: [PATCH 23/63] Update mlflow versions --- setup.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/setup.py b/setup.py index 213d14caac..e81831d7bd 100644 --- a/setup.py +++ b/setup.py @@ -82,7 +82,9 @@ "seqeval", "pytorch-forecasting", # "pytorch-forecasting>=0.9.0; python_version<'3.11'", - "mlflow<3.0.0", + "mlflow-skinny<=2.12.2; python_version<'3.12'", + "mlflow-skinny<=2.18.0; python_version>='3.12',<'3.13'", + "mlflow-skinny<3.0.0; python_version>='3.13'", "joblibspark>=0.5.0", "joblib<=1.3.2", "nbconvert", From a688c5dedab551e62ac4f0c2d4f81af50e56aade Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Wed, 7 Jan 2026 14:43:45 +0800 Subject: [PATCH 24/63] Fix mlflow version syntax --- setup.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/setup.py b/setup.py index e81831d7bd..9efede83f6 100644 --- a/setup.py +++ b/setup.py @@ -83,8 +83,8 @@ "pytorch-forecasting", # "pytorch-forecasting>=0.9.0; python_version<'3.11'", "mlflow-skinny<=2.12.2; python_version<'3.12'", - "mlflow-skinny<=2.18.0; python_version>='3.12',<'3.13'", - "mlflow-skinny<3.0.0; python_version>='3.13'", + "mlflow-skinny<=2.18.0; python_version>='3.12' and python_version<'3.13'", + "mlflow-skinny<=2.22.1; python_version>='3.13'", "joblibspark>=0.5.0", "joblib<=1.3.2", "nbconvert", From 71ccd5abfc3ecc2f44a2484094ffbe1f589cf939 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Wed, 7 Jan 2026 14:48:25 +0800 Subject: [PATCH 25/63] Update gitignore --- .gitignore | 1 + 1 file changed, 1 insertion(+) diff --git a/.gitignore b/.gitignore index 90d001609d..18c858dad2 100644 --- a/.gitignore +++ b/.gitignore @@ -184,6 +184,7 @@ notebook/lightning_logs/ lightning_logs/ flaml/autogen/extensions/tmp/ test/autogen/my_tmp/ +catboost_* # Internal configs .pypirc From 1902c9196f8156346bacfbe6d352c97e6e10f6fa Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Wed, 7 Jan 2026 17:55:42 +0800 Subject: [PATCH 26/63] Clean up cache to free space --- .github/workflows/python-package.yml | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index d8137ac8ac..d87b29f338 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -91,6 +91,14 @@ jobs: if: matrix.python-version == '3.8' || matrix.python-version == '3.9' run: | pip install -e .[vw] + - name: Clean up cache to free space + run: | + echo "Run `sudo rm -rf /home/runner/actions-runner/cached/*` to clear GitHub Actions cache" + sudo rm -rf /home/runner/actions-runner/cached/* + - name: Clear pip cache + run: | + echo "Run `pip cache purge` to clear pip cache" + pip cache purge - name: Test with pytest if: matrix.python-version != '3.10' run: | From 61fe7ef32defcebf64b206caadac876ca71caa84 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Wed, 7 Jan 2026 18:05:15 +0800 Subject: [PATCH 27/63] Remove clean up action cache --- .github/workflows/python-package.yml | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index d87b29f338..a40a189fee 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -91,10 +91,11 @@ jobs: if: matrix.python-version == '3.8' || matrix.python-version == '3.9' run: | pip install -e .[vw] - - name: Clean up cache to free space - run: | - echo "Run `sudo rm -rf /home/runner/actions-runner/cached/*` to clear GitHub Actions cache" - sudo rm -rf /home/runner/actions-runner/cached/* + # - name: Clean up cache to free space + # if: matrix.os != 'windows-latest' + # run: | + # echo "Run `sudo rm -rf /home/runner/actions-runner/cached/*` to clear GitHub Actions cache" + # sudo rm -rf /home/runner/actions-runner/cached/* - name: Clear pip cache run: | echo "Run `pip cache purge` to clear pip cache" From e551629f58b90d99433e326897ed6b4e9f557c49 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Wed, 7 Jan 2026 13:10:29 +0000 Subject: [PATCH 28/63] Fix blendsearch --- flaml/tune/searcher/blendsearch.py | 33 +++++++++++++++++++++++------- setup.py | 5 +++-- 2 files changed, 29 insertions(+), 9 deletions(-) diff --git a/flaml/tune/searcher/blendsearch.py b/flaml/tune/searcher/blendsearch.py index 8cd8f52e3b..c76a9a162b 100644 --- a/flaml/tune/searcher/blendsearch.py +++ b/flaml/tune/searcher/blendsearch.py @@ -244,13 +244,32 @@ def __init__( evaluated_rewards=evaluated_rewards, ) except (AssertionError, ValueError): - self._gs = GlobalSearch( - space=gs_space, - metric=metric, - mode=mode, - seed=gs_seed, - sampler=sampler, - ) + try: + self._gs = GlobalSearch( + space=gs_space, + metric=metric, + mode=mode, + seed=gs_seed, + sampler=sampler, + ) + except ValueError: + # Ray Tune's OptunaSearch converts Tune domains into Optuna + # distributions. Optuna disallows integer log distributions + # with step != 1 (e.g., qlograndint with q>1), which can + # raise here. Fall back to FLAML's OptunaSearch wrapper, + # which handles these spaces more permissively. + if getattr(GlobalSearch, "__module__", "").startswith("ray.tune"): + from .suggestion import OptunaSearch as _FallbackOptunaSearch + + self._gs = _FallbackOptunaSearch( + space=gs_space, + metric=metric, + mode=mode, + seed=gs_seed, + sampler=sampler, + ) + else: + raise self._gs.space = space else: self._gs = None diff --git a/setup.py b/setup.py index 9efede83f6..622d04d91e 100644 --- a/setup.py +++ b/setup.py @@ -51,7 +51,7 @@ "joblib<=1.3.2", ], "test": [ - "numpy<2.0.0", + "numpy>=1.17", "jupyter", "lightgbm>=2.3.1", "xgboost>=0.90,<2.0.0", @@ -67,9 +67,10 @@ "torchvision", "catboost>=0.26,<1.2; python_version<'3.11'", "catboost>=0.26; python_version>='3.11'", + "catboost>=0.26; python_version>='3.11' and python_version<'3.14'", "rgf-python", "optuna>=2.8.0,<=3.6.1", - "openml", + "openml; python_version<'3.14'", "statsmodels>=0.12.2", "psutil", "dataclasses", From 7ea4bc699ee7adb333fe6f69e256496bf7f6ab3d Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Wed, 7 Jan 2026 13:23:40 +0000 Subject: [PATCH 29/63] Update test workflow --- .github/workflows/python-package.yml | 21 ++++++++++----------- 1 file changed, 10 insertions(+), 11 deletions(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index a40a189fee..94a6152f16 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -39,7 +39,7 @@ jobs: fail-fast: false matrix: os: [ubuntu-latest, macos-latest, windows-latest] - python-version: ["3.9", "3.10", "3.11"] + python-version: ["3.10", "3.11", "3.12"] steps: - uses: actions/checkout@v4 - name: Set up Python ${{ matrix.python-version }} @@ -73,6 +73,11 @@ jobs: run: | pip install pyspark==3.5.1 pip list | grep "pyspark" + - name: On Ubuntu python 3.12, install pyspark 4.0.1 + if: matrix.python-version == '3.12' && matrix.os == 'ubuntu-latest' + run: | + pip install pyspark==4.0.1 + pip list | grep "pyspark" - name: If linux and python<3.11, install ray 2 if: matrix.os == 'ubuntu-latest' && matrix.python-version != '3.11' run: | @@ -83,22 +88,16 @@ jobs: pip install -e .[ray] # use macOS to test xgboost 1, but macOS also supports xgboost 2 pip install "xgboost<2" - - name: If linux, install prophet on python < 3.9 - if: matrix.os == 'ubuntu-latest' && matrix.python-version == '3.8' + - name: If linux, install prophet + if: matrix.os == 'ubuntu-latest' run: | pip install -e .[forecast] - - name: Install vw on python < 3.10 - if: matrix.python-version == '3.8' || matrix.python-version == '3.9' + - name: If linux, install vw + if: matrix.os == 'ubuntu-latest' run: | pip install -e .[vw] - # - name: Clean up cache to free space - # if: matrix.os != 'windows-latest' - # run: | - # echo "Run `sudo rm -rf /home/runner/actions-runner/cached/*` to clear GitHub Actions cache" - # sudo rm -rf /home/runner/actions-runner/cached/* - name: Clear pip cache run: | - echo "Run `pip cache purge` to clear pip cache" pip cache purge - name: Test with pytest if: matrix.python-version != '3.10' From 86afb2a3d9a9956e1a3179c88686cbfcb0395a6e Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Wed, 7 Jan 2026 21:47:02 +0800 Subject: [PATCH 30/63] Update setup.py --- setup.py | 23 +++++------------------ 1 file changed, 5 insertions(+), 18 deletions(-) diff --git a/setup.py b/setup.py index 622d04d91e..4bdd839b88 100644 --- a/setup.py +++ b/setup.py @@ -66,7 +66,6 @@ "torch", "torchvision", "catboost>=0.26,<1.2; python_version<'3.11'", - "catboost>=0.26; python_version>='3.11'", "catboost>=0.26; python_version>='3.11' and python_version<'3.14'", "rgf-python", "optuna>=2.8.0,<=3.6.1", @@ -77,7 +76,7 @@ "transformers[torch]", "datasets", "evaluate", - "nltk", # 3.8.2 doesn't work with mlflow + "nltk!=3.8.2", # 3.8.2 doesn't work with mlflow "rouge_score", "hcrystalball", "seqeval", @@ -106,7 +105,7 @@ "packaging", ], "ray": [ - "ray[tune]~=1.13", + "ray[tune]>=1.13,<2.5.0", ], "azureml": [ "azureml-mlflow", @@ -143,23 +142,11 @@ "prophet>=1.0.1", "statsmodels>=0.12.2", "hcrystalball==0.1.10", - "pytorch-forecasting>=0.9.0; python_version<'3.11'", - # "pytorch-forecasting==0.10.1; python_version=='3.11'", + "pytorch-forecasting; python_version<'3.14'", "pytorch-lightning==1.9.0", "tensorboardX==2.6", ], "benchmark": ["catboost>=0.26", "psutil==5.8.0", "xgboost==1.3.3", "pandas==1.1.4"], - "openai": ["openai==0.27.8", "diskcache"], - "autogen": ["openai==0.27.8", "diskcache", "termcolor"], - "mathchat": ["openai==0.27.8", "diskcache", "termcolor", "sympy", "pydantic==1.10.9", "wolframalpha"], - "retrievechat": [ - "openai==0.27.8", - "diskcache", - "termcolor", - "chromadb", - "tiktoken", - "sentence_transformers", - ], "synapse": [ "joblibspark>=0.5.0", "optuna>=2.8.0,<=3.6.1", @@ -172,9 +159,9 @@ "Operating System :: OS Independent", # Specify the Python versions you support here. "Programming Language :: Python :: 3", - "Programming Language :: Python :: 3.9", "Programming Language :: Python :: 3.10", "Programming Language :: Python :: 3.11", + "Programming Language :: Python :: 3.12", ], - python_requires=">=3.9", + python_requires=">=3.10", ) From e90836f22a317a6cbb08d74a5eaad33372f53052 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Wed, 7 Jan 2026 21:54:49 +0800 Subject: [PATCH 31/63] Fix catboost version --- setup.py | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/setup.py b/setup.py index 4bdd839b88..0346f7f6a6 100644 --- a/setup.py +++ b/setup.py @@ -65,8 +65,7 @@ "pre-commit", "torch", "torchvision", - "catboost>=0.26,<1.2; python_version<'3.11'", - "catboost>=0.26; python_version>='3.11' and python_version<'3.14'", + "catboost>=0.26", "rgf-python", "optuna>=2.8.0,<=3.6.1", "openml; python_version<'3.14'", @@ -97,8 +96,7 @@ "dill", # a drop in replacement of pickle ], "catboost": [ - "catboost>=0.26,<1.2; python_version<'3.11'", - "catboost>=0.26,<=1.2.5; python_version>='3.11'", + "catboost>=0.26", ], "blendsearch": [ "optuna>=2.8.0,<=3.6.1", From a051966cbd064b24e060bfa626da4a373cd5e22c Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Wed, 7 Jan 2026 21:57:21 +0800 Subject: [PATCH 32/63] Update workflow --- .github/workflows/python-package.yml | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index 94a6152f16..ca8bf68e81 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -78,8 +78,9 @@ jobs: run: | pip install pyspark==4.0.1 pip list | grep "pyspark" + #TODO: support ray for python 3.11+ - name: If linux and python<3.11, install ray 2 - if: matrix.os == 'ubuntu-latest' && matrix.python-version != '3.11' + if: matrix.os == 'ubuntu-latest' && matrix.python-version < '3.11' run: | pip install "ray[tune]<2.5.0" - name: If mac and python 3.10, install ray and xgboost 1 @@ -88,14 +89,16 @@ jobs: pip install -e .[ray] # use macOS to test xgboost 1, but macOS also supports xgboost 2 pip install "xgboost<2" - - name: If linux, install prophet - if: matrix.os == 'ubuntu-latest' + # TODO: support prophet for python 3.11+ + - name: Install prophet when on linux and python 3.10 + if: matrix.os == 'ubuntu-latest' && matrix.python-version == '3.10' run: | pip install -e .[forecast] - - name: If linux, install vw - if: matrix.os == 'ubuntu-latest' - run: | - pip install -e .[vw] + # TODO: support vw for python 3.10+ + # - name: If linux, install vw + # if: matrix.os == 'ubuntu-latest' + # run: | + # pip install -e .[vw] - name: Clear pip cache run: | pip cache purge From b5732adaa565d5028fcfb4b8286f3bd7d8f04812 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Wed, 7 Jan 2026 14:02:27 +0000 Subject: [PATCH 33/63] Prepare for python 3.14 --- setup.py | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/setup.py b/setup.py index 0346f7f6a6..1b24f44318 100644 --- a/setup.py +++ b/setup.py @@ -65,7 +65,7 @@ "pre-commit", "torch", "torchvision", - "catboost>=0.26", + "catboost>=0.26; python_version<'3.14'", "rgf-python", "optuna>=2.8.0,<=3.6.1", "openml; python_version<'3.14'", @@ -79,8 +79,7 @@ "rouge_score", "hcrystalball", "seqeval", - "pytorch-forecasting", - # "pytorch-forecasting>=0.9.0; python_version<'3.11'", + "pytorch-forecasting; python_version<'3.14'", "mlflow-skinny<=2.12.2; python_version<'3.12'", "mlflow-skinny<=2.18.0; python_version>='3.12' and python_version<'3.13'", "mlflow-skinny<=2.22.1; python_version>='3.13'", From 336287551a1e0f24586dae72390ba767553fe29f Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Wed, 7 Jan 2026 14:10:47 +0000 Subject: [PATCH 34/63] Support no catboost --- test/conftest.py | 25 ++++++++++++++++++++----- 1 file changed, 20 insertions(+), 5 deletions(-) diff --git a/test/conftest.py b/test/conftest.py index 47a74b2896..4b46207752 100644 --- a/test/conftest.py +++ b/test/conftest.py @@ -2,11 +2,24 @@ import numpy as np import pandas as pd -from catboost import CatBoostClassifier, CatBoostRegressor, Pool +import pytest from sklearn.metrics import f1_score, r2_score +try: + from catboost import CatBoostClassifier, CatBoostRegressor, Pool +except ImportError: # pragma: no cover + CatBoostClassifier = None + CatBoostRegressor = None + Pool = None -def evaluate_cv_folds_with_underlying_model(X_train_all, y_train_all, kf, model: Any, task: str) -> pd.DataFrame: + +def _is_catboost_model_type(model_type: type) -> bool: + if CatBoostClassifier is not None and CatBoostRegressor is not None: + return model_type is CatBoostClassifier or model_type is CatBoostRegressor + return getattr(model_type, "__module__", "").startswith("catboost") + + +def evaluate_cv_folds_with_underlying_model(X_train_all, y_train_all, kf, model: Any, task: str) -> List[float]: """Mimic the FLAML CV process to calculate the metrics across each fold. :param X_train_all: X training data @@ -17,7 +30,7 @@ def evaluate_cv_folds_with_underlying_model(X_train_all, y_train_all, kf, model: :return: An array containing the metrics """ rng = np.random.RandomState(2020) - all_fold_metrics: List[Dict[str, Union[int, float]]] = [] + all_fold_metrics: List[float] = [] for train_index, val_index in kf.split(X_train_all, y_train_all): X_train_split, y_train_split = X_train_all, y_train_all train_index = rng.permutation(train_index) @@ -25,9 +38,11 @@ def evaluate_cv_folds_with_underlying_model(X_train_all, y_train_all, kf, model: X_val = X_train_split.iloc[val_index] y_train, y_val = y_train_split[train_index], y_train_split[val_index] model_type = type(model) - if model_type is not CatBoostClassifier and model_type is not CatBoostRegressor: + if not _is_catboost_model_type(model_type): model.fit(X_train, y_train) else: + if Pool is None: + pytest.skip("catboost is not installed") use_best_model = True n = max(int(len(y_train) * 0.9), len(y_train) - 1000) if use_best_model else len(y_train) X_tr, y_tr = (X_train)[:n], y_train[:n] @@ -38,5 +53,5 @@ def evaluate_cv_folds_with_underlying_model(X_train_all, y_train_all, kf, model: reproduced_metric = 1 - f1_score(y_val, y_pred_classes) else: reproduced_metric = 1 - r2_score(y_val, y_pred_classes) - all_fold_metrics.append(reproduced_metric) + all_fold_metrics.append(float(reproduced_metric)) return all_fold_metrics From d7c77f1097b614bdd6e27b2725ab9f6ee9d27c7d Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Wed, 7 Jan 2026 22:29:50 +0800 Subject: [PATCH 35/63] Fix tests --- .github/workflows/python-package.yml | 8 ++++---- setup.py | 2 +- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index ca8bf68e81..abca8127e9 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -39,7 +39,7 @@ jobs: fail-fast: false matrix: os: [ubuntu-latest, macos-latest, windows-latest] - python-version: ["3.10", "3.11", "3.12"] + python-version: ["3.9", "3.10", "3.11"] steps: - uses: actions/checkout@v4 - name: Set up Python ${{ matrix.python-version }} @@ -47,7 +47,7 @@ jobs: with: python-version: ${{ matrix.python-version }} - name: On mac, install libomp to facilitate lgbm and xgboost install - if: matrix.os == 'macOS-latest' + if: matrix.os == 'macos-latest' run: | brew update brew install libomp @@ -84,10 +84,10 @@ jobs: run: | pip install "ray[tune]<2.5.0" - name: If mac and python 3.10, install ray and xgboost 1 - if: matrix.os == 'macOS-latest' && matrix.python-version == '3.10' + if: matrix.os == 'macos-latest' && matrix.python-version == '3.10' run: | pip install -e .[ray] - # use macOS to test xgboost 1, but macOS also supports xgboost 2 + # use macos to test xgboost 1, but macos also supports xgboost 2 pip install "xgboost<2" # TODO: support prophet for python 3.11+ - name: Install prophet when on linux and python 3.10 diff --git a/setup.py b/setup.py index 1b24f44318..08f36ed311 100644 --- a/setup.py +++ b/setup.py @@ -51,7 +51,7 @@ "joblib<=1.3.2", ], "test": [ - "numpy>=1.17", + "numpy>=1.17,<2.0.0", "jupyter", "lightgbm>=2.3.1", "xgboost>=0.90,<2.0.0", From fe1b1445f171faabd9adb104392dfb98ad8b9e4d Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Wed, 7 Jan 2026 22:33:11 +0800 Subject: [PATCH 36/63] Fix python_requires --- setup.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/setup.py b/setup.py index 08f36ed311..f600d89c68 100644 --- a/setup.py +++ b/setup.py @@ -156,9 +156,10 @@ "Operating System :: OS Independent", # Specify the Python versions you support here. "Programming Language :: Python :: 3", + "Programming Language :: Python :: 3.9", "Programming Language :: Python :: 3.10", "Programming Language :: Python :: 3.11", "Programming Language :: Python :: 3.12", ], - python_requires=">=3.10", + python_requires=">=3.9", ) From c28d9db403a936c28d8bb871865923a049c71212 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 02:28:18 +0000 Subject: [PATCH 37/63] Update test workflow --- .github/workflows/python-package.yml | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index abca8127e9..657a819f4f 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -83,15 +83,14 @@ jobs: if: matrix.os == 'ubuntu-latest' && matrix.python-version < '3.11' run: | pip install "ray[tune]<2.5.0" - - name: If mac and python 3.10, install ray and xgboost 1 + - name: If mac and python 3.10, install xgboost 1 if: matrix.os == 'macos-latest' && matrix.python-version == '3.10' run: | - pip install -e .[ray] # use macos to test xgboost 1, but macos also supports xgboost 2 pip install "xgboost<2" # TODO: support prophet for python 3.11+ - - name: Install prophet when on linux and python 3.10 - if: matrix.os == 'ubuntu-latest' && matrix.python-version == '3.10' + - name: Install prophet when on linux + if: matrix.os == 'ubuntu-latest' run: | pip install -e .[forecast] # TODO: support vw for python 3.10+ From ffcfacc8ac82a8e38970d91dc73cf5faff5560d7 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 02:35:01 +0000 Subject: [PATCH 38/63] Fix vw tests --- .github/workflows/python-package.yml | 9 ++++----- test/test_autovw.py | 14 ++++++++------ 2 files changed, 12 insertions(+), 11 deletions(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index 657a819f4f..465d6aaa07 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -88,16 +88,15 @@ jobs: run: | # use macos to test xgboost 1, but macos also supports xgboost 2 pip install "xgboost<2" - # TODO: support prophet for python 3.11+ - name: Install prophet when on linux if: matrix.os == 'ubuntu-latest' run: | pip install -e .[forecast] # TODO: support vw for python 3.10+ - # - name: If linux, install vw - # if: matrix.os == 'ubuntu-latest' - # run: | - # pip install -e .[vw] + - name: If linux and python<3.10, install vw + if: matrix.os == 'ubuntu-latest' && matrix.python-version < '3.10' + run: | + pip install -e .[vw] - name: Clear pip cache run: | pip cache purge diff --git a/test/test_autovw.py b/test/test_autovw.py index 06140f435e..1c617ef910 100644 --- a/test/test_autovw.py +++ b/test/test_autovw.py @@ -16,6 +16,13 @@ from flaml import AutoVW from flaml.tune import loguniform, polynomial_expansion_set +try: + from vowpalwabbit import pyvw +except ImportError: + skip_vw_test = True +else: + skip_vw_test = False + VW_DS_DIR = "test/data/" NS_LIST = list(string.ascii_lowercase) + list(string.ascii_uppercase) logger = logging.getLogger(__name__) @@ -351,14 +358,9 @@ def get_vw_tuning_problem(tuning_hp="NamesapceInteraction"): return vw_oml_problem_args, vw_online_aml_problem -@pytest.mark.skipif( - "3.10" in sys.version or "3.11" in sys.version, - reason="do not run on py >= 3.10", -) +@pytest.mark.skipif(skip_vw_test, reason="vowpalwabbit not installed") class TestAutoVW(unittest.TestCase): def test_vw_oml_problem_and_vanilla_vw(self): - from vowpalwabbit import pyvw - try: vw_oml_problem_args, vw_online_aml_problem = get_vw_tuning_problem() except (SSLError, ServerError, Exception) as e: From ca1e0d185e9f5afd61cb5c96ef33ffa03cfc4228 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 02:43:38 +0000 Subject: [PATCH 39/63] Remove python 3.9 --- .github/workflows/python-package.yml | 27 +-------------------------- setup.py | 3 +-- 2 files changed, 2 insertions(+), 28 deletions(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index 465d6aaa07..0fc0fb6e7d 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -39,7 +39,7 @@ jobs: fail-fast: false matrix: os: [ubuntu-latest, macos-latest, windows-latest] - python-version: ["3.9", "3.10", "3.11"] + python-version: ["3.10", "3.11"] steps: - uses: actions/checkout@v4 - name: Set up Python ${{ matrix.python-version }} @@ -116,28 +116,3 @@ jobs: with: file: ./coverage.xml flags: unittests - - # docs: - - # runs-on: ubuntu-latest - - # steps: - # - uses: actions/checkout@v3 - # - name: Setup Python - # uses: actions/setup-python@v4 - # with: - # python-version: '3.8' - # - name: Compile documentation - # run: | - # pip install -e . - # python -m pip install sphinx sphinx_rtd_theme - # cd docs - # make html - # - name: Deploy to GitHub pages - # if: ${{ github.ref == 'refs/heads/main' }} - # uses: JamesIves/github-pages-deploy-action@3.6.2 - # with: - # GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }} - # BRANCH: gh-pages - # FOLDER: docs/_build/html - # CLEAN: true diff --git a/setup.py b/setup.py index f600d89c68..08f36ed311 100644 --- a/setup.py +++ b/setup.py @@ -156,10 +156,9 @@ "Operating System :: OS Independent", # Specify the Python versions you support here. "Programming Language :: Python :: 3", - "Programming Language :: Python :: 3.9", "Programming Language :: Python :: 3.10", "Programming Language :: Python :: 3.11", "Programming Language :: Python :: 3.12", ], - python_requires=">=3.9", + python_requires=">=3.10", ) From e962e8392b4b3d1abf88f501b8308a5e574ca113 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 03:31:49 +0000 Subject: [PATCH 40/63] Fix nlp tests --- test/nlp/test_default.py | 118 +++++++++++++++++++++++++++------------ 1 file changed, 82 insertions(+), 36 deletions(-) diff --git a/test/nlp/test_default.py b/test/nlp/test_default.py index 239fce2276..5f7622a1c9 100644 --- a/test/nlp/test_default.py +++ b/test/nlp/test_default.py @@ -30,21 +30,33 @@ def test_build_portfolio(path="./test/nlp/default", strategy="greedy"): @pytest.mark.skipif(sys.platform == "win32", reason="do not run on windows") def test_starting_point_not_in_search_space(): - from flaml import AutoML + """Regression test for invalid starting points and custom_hp. + + This test must not require network access to Hugging Face. + """ """ test starting_points located outside of the search space, and custom_hp is not set """ - this_estimator_name = "transformer" - X_train, y_train, X_val, y_val, _ = get_toy_data_seqclassification() + from flaml.automl.state import SearchState + from flaml.automl.task.factory import task_factory - automl = AutoML() - automl_settings = get_automl_settings(estimator_name=this_estimator_name) - - automl_settings["starting_points"] = {this_estimator_name: [{"learning_rate": 2e-3}]} - - automl.fit(X_train, y_train, **automl_settings) - assert automl._search_states[this_estimator_name].init_config[0]["learning_rate"] != 2e-3 + this_estimator_name = "transformer" + X_train, y_train, _, _, _ = get_toy_data_seqclassification() + task = task_factory("seq-classification", X_train, y_train) + estimator_class = task.estimator_class_from_str(this_estimator_name) + estimator_class.init() + + # SearchState is where invalid starting points are filtered out when max_iter > 1. + search_state = SearchState( + learner_class=estimator_class, + data=X_train, + task=task, + starting_point={"learning_rate": 2e-3}, + max_iter=3, + budget=10, + ) + assert search_state.init_config and search_state.init_config[0].get("learning_rate") != 2e-3 """ test starting_points located outside of the search space, and custom_hp is set @@ -52,39 +64,60 @@ def test_starting_point_not_in_search_space(): from flaml import tune - X_train, y_train, X_val, y_val, _ = get_toy_data_seqclassification() + X_train, y_train, _, _, _ = get_toy_data_seqclassification() this_estimator_name = "transformer_ms" - automl = AutoML() - automl_settings = get_automl_settings(estimator_name=this_estimator_name) - - automl_settings["custom_hp"] = { - this_estimator_name: { - "model_path": { - "domain": "albert-base-v2", - }, - "learning_rate": { - "domain": tune.choice([1e-4, 1e-5]), - }, - "per_device_train_batch_size": { - "domain": 2, - }, - } + task = task_factory("seq-classification", X_train, y_train) + estimator_class = task.estimator_class_from_str(this_estimator_name) + estimator_class.init() + + custom_hp = { + "model_path": { + "domain": "albert-base-v2", + }, + "learning_rate": { + "domain": tune.choice([1e-4, 1e-5]), + }, + "per_device_train_batch_size": { + "domain": 2, + }, } - automl_settings["starting_points"] = "data:test/nlp/default/" - automl.fit(X_train, y_train, **automl_settings) - assert len(automl._search_states[this_estimator_name].init_config[0]) == len( - automl._search_states[this_estimator_name]._search_space_domain - ) - len(automl_settings["custom_hp"][this_estimator_name]), ( + # Simulate a suggested starting point (e.g. from portfolio) which becomes invalid + # after custom_hp constrains the space. + invalid_starting_points = [ + { + "learning_rate": 1e-5, + "num_train_epochs": 1.0, + "per_device_train_batch_size": 8, + "seed": 43, + "global_max_steps": 100, + "model_path": "google/electra-base-discriminator", + } + ] + + search_state = SearchState( + learner_class=estimator_class, + data=X_train, + task=task, + starting_point=invalid_starting_points, + custom_hp=custom_hp, + max_iter=3, + budget=10, + ) + + assert search_state.init_config, "Expected a non-empty init_config list" + init_config0 = search_state.init_config[0] + assert init_config0 is not None + assert len(init_config0) == len(search_state._search_space_domain) - len(custom_hp), ( "The search space is updated with the custom_hp on {} hyperparameters of " "the specified estimator without an initial value. Thus a valid init config " "should only contain the cardinality of the search space minus {}".format( - len(automl_settings["custom_hp"][this_estimator_name]), - len(automl_settings["custom_hp"][this_estimator_name]), + len(custom_hp), + len(custom_hp), ) ) - assert automl._search_states[this_estimator_name].search_space["model_path"] == "albert-base-v2" + assert search_state.search_space["model_path"] == "albert-base-v2" if os.path.exists("test/data/output/"): try: @@ -106,7 +139,13 @@ def test_points_to_evaluate(): automl_settings["custom_hp"] = {"transformer_ms": {"model_path": {"domain": "google/electra-small-discriminator"}}} - automl.fit(X_train, y_train, **automl_settings) + try: + automl.fit(X_train, y_train, **automl_settings) + except OSError as e: + message = str(e) + if "Too Many Requests" in message or "rate limit" in message.lower(): + pytest.skip(f"Skipping HF model load/training: {message}") + raise if os.path.exists("test/data/output/"): try: @@ -141,7 +180,14 @@ def test_zero_shot_nomodel(): fit_kwargs = automl_settings.pop("fit_kwargs_by_estimator", {}).get(estimator_name) fit_kwargs.update(automl_settings) pop_args(fit_kwargs) - model.fit(X_train, y_train, **fit_kwargs) + + try: + model.fit(X_train, y_train, **fit_kwargs) + except OSError as e: + message = str(e) + if "Too Many Requests" in message or "rate limit" in message.lower(): + pytest.skip(f"Skipping HF model load/training: {message}") + raise if os.path.exists("test/data/output/"): try: From 0157b4dab68f8b3dda336f6b1c4cb4f087b3632f Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 03:52:50 +0000 Subject: [PATCH 41/63] Fix prophet --- setup.py | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) diff --git a/setup.py b/setup.py index 08f36ed311..fb14a541d7 100644 --- a/setup.py +++ b/setup.py @@ -129,19 +129,19 @@ "seqeval", ], "ts_forecast": [ - "holidays<0.14", # to prevent installation error for prophet - "prophet>=1.0.1", + "holidays", + "prophet>=1.1.5", "statsmodels>=0.12.2", - "hcrystalball==0.1.10", + "hcrystalball>=0.1.10", ], "forecast": [ - "holidays<0.14", # to prevent installation error for prophet - "prophet>=1.0.1", + "holidays", + "prophet>=1.1.5", "statsmodels>=0.12.2", - "hcrystalball==0.1.10", - "pytorch-forecasting; python_version<'3.14'", - "pytorch-lightning==1.9.0", - "tensorboardX==2.6", + "hcrystalball>=0.1.10", + "pytorch-forecasting>=0.10.4; python_version<'3.14'", + "pytorch-lightning>=1.9.0", + "tensorboardX>=2.6", ], "benchmark": ["catboost>=0.26", "psutil==5.8.0", "xgboost==1.3.3", "pandas==1.1.4"], "synapse": [ From 12e3502905a3929c12bb8eabe896f02e205f6d00 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 04:09:03 +0000 Subject: [PATCH 42/63] Print pip freeze for better debugging --- .github/workflows/python-package.yml | 3 +++ 1 file changed, 3 insertions(+) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index 0fc0fb6e7d..dd1ae78c90 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -97,6 +97,9 @@ jobs: if: matrix.os == 'ubuntu-latest' && matrix.python-version < '3.10' run: | pip install -e .[vw] + - name: Pip freeze + run: | + pip freeze - name: Clear pip cache run: | pip cache purge From 560b7c7ec068fb64b7f2077c88175466bd6b5aa5 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 06:55:39 +0000 Subject: [PATCH 43/63] Fix Optuna search does not support parameters of type Float with samplers of type Quantized --- flaml/tune/searcher/suggestion.py | 98 ++++++++++++++++++++++++++----- 1 file changed, 84 insertions(+), 14 deletions(-) diff --git a/flaml/tune/searcher/suggestion.py b/flaml/tune/searcher/suggestion.py index 6142668929..552f65a661 100644 --- a/flaml/tune/searcher/suggestion.py +++ b/flaml/tune/searcher/suggestion.py @@ -35,6 +35,73 @@ Quantized, Uniform, ) + +# If Ray is installed, flaml.tune may re-export Ray Tune sampling functions. +# In that case, the search space contains Ray Tune Domain/Sampler objects, +# which should be accepted by our Optuna search-space conversion. +try: + from ray import __version__ as _ray_version # type: ignore + + if str(_ray_version).startswith("1."): + from ray.tune.sample import ( # type: ignore + Categorical as _RayCategorical, + ) + from ray.tune.sample import ( + Domain as _RayDomain, + ) + from ray.tune.sample import ( + Float as _RayFloat, + ) + from ray.tune.sample import ( + Integer as _RayInteger, + ) + from ray.tune.sample import ( + LogUniform as _RayLogUniform, + ) + from ray.tune.sample import ( + Quantized as _RayQuantized, + ) + from ray.tune.sample import ( + Uniform as _RayUniform, + ) + else: + from ray.tune.search.sample import ( # type: ignore + Categorical as _RayCategorical, + ) + from ray.tune.search.sample import ( + Domain as _RayDomain, + ) + from ray.tune.search.sample import ( + Float as _RayFloat, + ) + from ray.tune.search.sample import ( + Integer as _RayInteger, + ) + from ray.tune.search.sample import ( + LogUniform as _RayLogUniform, + ) + from ray.tune.search.sample import ( + Quantized as _RayQuantized, + ) + from ray.tune.search.sample import ( + Uniform as _RayUniform, + ) + + _FLOAT_TYPES = (Float, _RayFloat) + _INTEGER_TYPES = (Integer, _RayInteger) + _CATEGORICAL_TYPES = (Categorical, _RayCategorical) + _DOMAIN_TYPES = (Domain, _RayDomain) + _QUANTIZED_TYPES = (Quantized, _RayQuantized) + _UNIFORM_TYPES = (Uniform, _RayUniform) + _LOGUNIFORM_TYPES = (LogUniform, _RayLogUniform) +except Exception: # pragma: no cover + _FLOAT_TYPES = (Float,) + _INTEGER_TYPES = (Integer,) + _CATEGORICAL_TYPES = (Categorical,) + _DOMAIN_TYPES = (Domain,) + _QUANTIZED_TYPES = (Quantized,) + _UNIFORM_TYPES = (Uniform,) + _LOGUNIFORM_TYPES = (LogUniform,) from ..trial import flatten_dict, unflatten_dict from .variant_generator import parse_spec_vars @@ -850,19 +917,22 @@ def convert_search_space(spec: Dict) -> Dict[str, Any]: def resolve_value(domain: Domain) -> ot.distributions.BaseDistribution: quantize = None - sampler = domain.get_sampler() - if isinstance(sampler, Quantized): + # Ray Tune Domains and FLAML Domains both provide get_sampler(), but + # fall back to the .sampler attribute for robustness. + sampler = domain.get_sampler() if hasattr(domain, "get_sampler") else getattr(domain, "sampler", None) + + if isinstance(sampler, _QUANTIZED_TYPES) or type(sampler).__name__ == "Quantized": quantize = sampler.q - sampler = sampler.sampler - if isinstance(sampler, LogUniform): + sampler = getattr(sampler, "sampler", None) or sampler.get_sampler() + if isinstance(sampler, _LOGUNIFORM_TYPES) or type(sampler).__name__ == "LogUniform": logger.warning( "Optuna does not handle quantization in loguniform " "sampling. The parameter will be passed but it will " "probably be ignored." ) - if isinstance(domain, Float): - if isinstance(sampler, LogUniform): + if isinstance(domain, _FLOAT_TYPES) or type(domain).__name__ == "Float": + if isinstance(sampler, _LOGUNIFORM_TYPES) or type(sampler).__name__ == "LogUniform": if quantize: logger.warning( "Optuna does not support both quantization and " @@ -870,17 +940,17 @@ def resolve_value(domain: Domain) -> ot.distributions.BaseDistribution: ) return ot.distributions.LogUniformDistribution(domain.lower, domain.upper) - elif isinstance(sampler, Uniform): + elif isinstance(sampler, _UNIFORM_TYPES) or type(sampler).__name__ == "Uniform": if quantize: return ot.distributions.DiscreteUniformDistribution(domain.lower, domain.upper, quantize) return ot.distributions.UniformDistribution(domain.lower, domain.upper) - elif isinstance(domain, Integer): - if isinstance(sampler, LogUniform): + elif isinstance(domain, _INTEGER_TYPES) or type(domain).__name__ == "Integer": + if isinstance(sampler, _LOGUNIFORM_TYPES) or type(sampler).__name__ == "LogUniform": # ``step`` argument Deprecated in v2.0.0. ``step`` argument should be 1 in Log Distribution # The removal of this feature is currently scheduled for v4.0.0, return ot.distributions.IntLogUniformDistribution(domain.lower, domain.upper - 1, step=1) - elif isinstance(sampler, Uniform): + elif isinstance(sampler, _UNIFORM_TYPES) or type(sampler).__name__ == "Uniform": # Upper bound should be inclusive for quantization and # exclusive otherwise return ot.distributions.IntUniformDistribution( @@ -888,16 +958,16 @@ def resolve_value(domain: Domain) -> ot.distributions.BaseDistribution: domain.upper - int(bool(not quantize)), step=quantize or 1, ) - elif isinstance(domain, Categorical): - if isinstance(sampler, Uniform): + elif isinstance(domain, _CATEGORICAL_TYPES) or type(domain).__name__ == "Categorical": + if isinstance(sampler, _UNIFORM_TYPES) or type(sampler).__name__ == "Uniform": return ot.distributions.CategoricalDistribution(domain.categories) raise ValueError( "Optuna search does not support parameters of type " - "`{}` with samplers of type `{}`".format(type(domain).__name__, type(domain.sampler).__name__) + "`{}` with samplers of type `{}`".format(type(domain).__name__, type(sampler).__name__) ) # Parameter name is e.g. "a/b/c" for nested dicts values = {"/".join(path): resolve_value(domain) for path, domain in domain_vars} - return values + return values From 557c2505693aed6ce5f4b7bae8f9503eed928025 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 07:48:42 +0000 Subject: [PATCH 44/63] Save dependencies for later inspection --- .github/workflows/python-package.yml | 17 ++++++++ test/check_dependency.py | 58 ++++++++++++++++++++++++++++ 2 files changed, 75 insertions(+) create mode 100644 test/check_dependency.py diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index dd1ae78c90..7a324e12fe 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -100,6 +100,7 @@ jobs: - name: Pip freeze run: | pip freeze + pip freeze > installed_all_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt - name: Clear pip cache run: | pip cache purge @@ -119,3 +120,19 @@ jobs: with: file: ./coverage.xml flags: unittests + - name: Check dependencies + run: | + python test/check_dependency.py > installed_first_tier_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt + cat installed_first_tier_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt + # commit the generated file to a repository branch for later inspection + git config --global user.name 'github-actions[bot]' + git config --global user.email 'github-actions[bot]@users.noreply.github.com' + git fetch origin + git checkout -b installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} origin/installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} || git checkout -b installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} + git add installed_first_tier_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt + git add installed_all_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt + git add ./coverage.xml + git commit -m "Update installed first tier dependencies for Python ${{ matrix.python-version }} on ${{ matrix.os }}" + git push origin installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} + # checkout back to the current branch + git checkout ${{ github.ref_name }} diff --git a/test/check_dependency.py b/test/check_dependency.py new file mode 100644 index 0000000000..b71b443725 --- /dev/null +++ b/test/check_dependency.py @@ -0,0 +1,58 @@ +import subprocess +from importlib.metadata import distributions + +installed_libs = sorted(f"{dist.metadata['Name']}=={dist.version}" for dist in distributions()) + +first_tier_dependencies = [ + "numpy", + "jupyter", + "lightgbm", + "xgboost", + "scipy", + "pandas", + "scikit-learn", + "thop", + "pytest", + "coverage", + "pre-commit", + "torch", + "torchvision", + "catboost", + "rgf-python", + "optuna", + "openml", + "statsmodels", + "psutil", + "dataclasses", + "transformers[torch]", + "datasets", + "evaluate", + "nltk", + "rouge_score", + "hcrystalball", + "seqeval", + "pytorch-forecasting", + "mlflow-skinny", + "joblibspark", + "joblib", + "nbconvert", + "nbformat", + "ipykernel", + "pytorch-lightning", + "tensorboardX", + "requests", + "packaging", + "dill", + "ray", + "prophet", +] + + +for lib in installed_libs: + lib_name = lib.split("==")[0] + if lib_name in first_tier_dependencies: + print(lib) + +# print current commit hash +commit_hash = subprocess.check_output(["git", "rev-parse", "HEAD"]).decode("utf-8").strip() +print(f"Current commit hash: {commit_hash}") From 7b2aec9c9bee076b0d3b6b4972884effec875055 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 08:49:17 +0000 Subject: [PATCH 45/63] Fix coverage.xml not exists --- .github/workflows/python-package.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index 7a324e12fe..c007a3b301 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -131,7 +131,7 @@ jobs: git checkout -b installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} origin/installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} || git checkout -b installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} git add installed_first_tier_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt git add installed_all_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt - git add ./coverage.xml + git add ./coverage.xml || true git commit -m "Update installed first tier dependencies for Python ${{ matrix.python-version }} on ${{ matrix.os }}" git push origin installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} # checkout back to the current branch From 3196551394ae5f58d20fdb85b610849650c8be33 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 11:16:58 +0000 Subject: [PATCH 46/63] Fix github action permission --- .github/workflows/python-package.yml | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index c007a3b301..ceafb9c3a9 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -26,7 +26,8 @@ on: # Every other day at 02:00 UTC - cron: '0 2 */2 * *' -permissions: {} +permissions: + contents: write concurrency: group: ${{ github.workflow }}-${{ github.ref }}-${{ github.head_ref }} cancel-in-progress: ${{ github.ref != 'refs/heads/main' }} @@ -127,12 +128,16 @@ jobs: # commit the generated file to a repository branch for later inspection git config --global user.name 'github-actions[bot]' git config --global user.email 'github-actions[bot]@users.noreply.github.com' + git config advice.addIgnoredFile false git fetch origin - git checkout -b installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} origin/installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} || git checkout -b installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} + if git show-ref --verify --quiet refs/remotes/origin/installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }}; then + git checkout -B installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} \ + origin/installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} + else + git checkout -b installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} + fi git add installed_first_tier_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt git add installed_all_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt git add ./coverage.xml || true git commit -m "Update installed first tier dependencies for Python ${{ matrix.python-version }} on ${{ matrix.os }}" git push origin installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} - # checkout back to the current branch - git checkout ${{ github.ref_name }} From 2eb598a6dfdab02b4f008008f67ace1f1d01aeff Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 11:17:42 +0000 Subject: [PATCH 47/63] Handle python 3.13 --- setup.py | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/setup.py b/setup.py index fb14a541d7..6ccc296ec3 100644 --- a/setup.py +++ b/setup.py @@ -51,8 +51,9 @@ "joblib<=1.3.2", ], "test": [ - "numpy>=1.17,<2.0.0", - "jupyter", + "numpy>=1.17,<2.0.0; python_version<'3.13'", + "numpy>2.0.0; python_version>='3.13'", + "jupyter; python_version<'3.13'", "lightgbm>=2.3.1", "xgboost>=0.90,<2.0.0", "scipy>=1.4.1", @@ -65,10 +66,10 @@ "pre-commit", "torch", "torchvision", - "catboost>=0.26; python_version<'3.14'", + "catboost>=0.26; python_version<'3.13'", "rgf-python", "optuna>=2.8.0,<=3.6.1", - "openml; python_version<'3.14'", + "openml; python_version<'3.13'", "statsmodels>=0.12.2", "psutil", "dataclasses", @@ -79,7 +80,7 @@ "rouge_score", "hcrystalball", "seqeval", - "pytorch-forecasting; python_version<'3.14'", + "pytorch-forecasting; python_version<'3.13'", "mlflow-skinny<=2.12.2; python_version<'3.12'", "mlflow-skinny<=2.18.0; python_version>='3.12' and python_version<'3.13'", "mlflow-skinny<=2.22.1; python_version>='3.13'", @@ -139,7 +140,7 @@ "prophet>=1.1.5", "statsmodels>=0.12.2", "hcrystalball>=0.1.10", - "pytorch-forecasting>=0.10.4; python_version<'3.14'", + "pytorch-forecasting>=0.10.4; python_version<'3.13'", "pytorch-lightning>=1.9.0", "tensorboardX>=2.6", ], From fbe4192c20970807ce8f798ee2d12a9497456cc5 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 11:18:36 +0000 Subject: [PATCH 48/63] Address openml is not installed --- flaml/automl/data.py | 11 +++++++---- test/automl/test_notebook_example.py | 19 +++++++++++++++++-- test/automl/test_warmstart.py | 9 ++++++++- test/spark/test_performance.py | 19 +++++++++++++++++-- test/test_autovw.py | 18 ++++++++++++++++-- 5 files changed, 65 insertions(+), 11 deletions(-) diff --git a/flaml/automl/data.py b/flaml/automl/data.py index 4c473963f0..096ba46d90 100644 --- a/flaml/automl/data.py +++ b/flaml/automl/data.py @@ -50,7 +50,10 @@ def load_openml_dataset(dataset_id, data_dir=None, random_state=0, dataset_forma """ import pickle - import openml + try: + import openml + except ImportError: + openml = None from sklearn.model_selection import train_test_split filename = "openml_ds" + str(dataset_id) + ".pkl" @@ -61,15 +64,15 @@ def load_openml_dataset(dataset_id, data_dir=None, random_state=0, dataset_forma dataset = pickle.load(f) else: print("download dataset from openml") - dataset = openml.datasets.get_dataset(dataset_id) + dataset = openml.datasets.get_dataset(dataset_id) if openml else None if not os.path.exists(data_dir): os.makedirs(data_dir) with open(filepath, "wb") as f: pickle.dump(dataset, f, pickle.HIGHEST_PROTOCOL) - print("Dataset name:", dataset.name) + print("Dataset name:", dataset.name) if dataset else None try: X, y, *__ = dataset.get_data(target=dataset.default_target_attribute, dataset_format=dataset_format) - except ValueError: + except (ValueError, AttributeError, TypeError): from sklearn.datasets import fetch_openml X, y = fetch_openml(data_id=dataset_id, return_X_y=True) diff --git a/test/automl/test_notebook_example.py b/test/automl/test_notebook_example.py index 536ef0484c..b4558f1092 100644 --- a/test/automl/test_notebook_example.py +++ b/test/automl/test_notebook_example.py @@ -1,8 +1,23 @@ import sys import pytest -from minio.error import ServerError -from openml.exceptions import OpenMLServerException + +try: + from minio.error import ServerError +except ImportError: + + class ServerError(Exception): + pass + + +try: + from openml.exceptions import OpenMLServerException +except ImportError: + + class OpenMLServerException(Exception): + pass + + from requests.exceptions import ChunkedEncodingError, SSLError diff --git a/test/automl/test_warmstart.py b/test/automl/test_warmstart.py index 677883741a..ba390bf7d3 100644 --- a/test/automl/test_warmstart.py +++ b/test/automl/test_warmstart.py @@ -108,7 +108,14 @@ def test_nobudget(self): def test_FLAML_sample_size_in_starting_points(self): from minio.error import ServerError - from openml.exceptions import OpenMLServerException + + try: + from openml.exceptions import OpenMLServerException + except ImportError: + + class OpenMLServerException(Exception): + pass + from requests.exceptions import ChunkedEncodingError, SSLError from flaml import AutoML diff --git a/test/spark/test_performance.py b/test/spark/test_performance.py index febd1a4b88..030fcae493 100644 --- a/test/spark/test_performance.py +++ b/test/spark/test_performance.py @@ -2,8 +2,23 @@ import sys import pytest -from minio.error import ServerError -from openml.exceptions import OpenMLServerException + +try: + from minio.error import ServerError +except ImportError: + + class ServerError(Exception): + pass + + +try: + from openml.exceptions import OpenMLServerException +except ImportError: + + class OpenMLServerException(Exception): + pass + + from requests.exceptions import ChunkedEncodingError, SSLError from flaml.tune.spark.utils import check_spark diff --git a/test/test_autovw.py b/test/test_autovw.py index 1c617ef910..100021d565 100644 --- a/test/test_autovw.py +++ b/test/test_autovw.py @@ -5,11 +5,23 @@ import unittest import numpy as np -import openml + +try: + import openml +except ImportError: + openml = None import pandas as pd import pytest import scipy.sparse -from minio.error import ServerError + +try: + from minio.error import ServerError +except ImportError: + + class ServerError(Exception): + pass + + from requests.exceptions import SSLError from sklearn.metrics import mean_absolute_error, mean_squared_error @@ -23,6 +35,8 @@ else: skip_vw_test = False +pytest.skip("skipping if no openml", allow_module_level=True) if openml is None else None + VW_DS_DIR = "test/data/" NS_LIST = list(string.ascii_lowercase) + list(string.ascii_uppercase) logger = logging.getLogger(__name__) From ae0e68782e672dfa5506fdb308cc818f45303db1 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 11:37:46 +0000 Subject: [PATCH 49/63] Check dependencies before run tests --- .github/workflows/python-package.yml | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index ceafb9c3a9..f2f5107ee2 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -100,8 +100,12 @@ jobs: pip install -e .[vw] - name: Pip freeze run: | - pip freeze pip freeze > installed_all_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt + cat installed_all_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt + - name: Check dependencies + run: | + python test/check_dependency.py > installed_first_tier_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt + cat installed_first_tier_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt - name: Clear pip cache run: | pip cache purge @@ -121,11 +125,8 @@ jobs: with: file: ./coverage.xml flags: unittests - - name: Check dependencies + - name: Save dependencies run: | - python test/check_dependency.py > installed_first_tier_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt - cat installed_first_tier_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt - # commit the generated file to a repository branch for later inspection git config --global user.name 'github-actions[bot]' git config --global user.email 'github-actions[bot]@users.noreply.github.com' git config advice.addIgnoredFile false From 3c49b6c49e8542a3a5d1c2820456d6f8bdbbe258 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 11:38:23 +0000 Subject: [PATCH 50/63] Update dependencies --- setup.py | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/setup.py b/setup.py index 6ccc296ec3..1b5597b35a 100644 --- a/setup.py +++ b/setup.py @@ -81,9 +81,9 @@ "hcrystalball", "seqeval", "pytorch-forecasting; python_version<'3.13'", - "mlflow-skinny<=2.12.2; python_version<'3.12'", - "mlflow-skinny<=2.18.0; python_version>='3.12' and python_version<'3.13'", - "mlflow-skinny<=2.22.1; python_version>='3.13'", + "mlflow<=2.12.2; python_version<'3.12'", + "mlflow<=2.18.0; python_version>='3.12' and python_version<'3.13'", + "mlflow<=2.22.1; python_version>='3.13'", "joblibspark>=0.5.0", "joblib<=1.3.2", "nbconvert", From 6a7cddf8e1c4f09d44bac66230d0ae019c8b1092 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 12:02:09 +0000 Subject: [PATCH 51/63] Fix syntax error --- .github/workflows/python-package.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index f2f5107ee2..1356e8b813 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -131,7 +131,7 @@ jobs: git config --global user.email 'github-actions[bot]@users.noreply.github.com' git config advice.addIgnoredFile false git fetch origin - if git show-ref --verify --quiet refs/remotes/origin/installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }}; then + if (git show-ref --verify --quiet refs/remotes/origin/installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }}); then git checkout -B installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} \ origin/installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} else From c5d693718e534aba926a4cffe5eb2d39d8d8c257 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 12:16:25 +0000 Subject: [PATCH 52/63] Use bash --- .github/workflows/python-package.yml | 1 + 1 file changed, 1 insertion(+) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index 1356e8b813..3fb16d4ad6 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -126,6 +126,7 @@ jobs: file: ./coverage.xml flags: unittests - name: Save dependencies + shell: bash run: | git config --global user.name 'github-actions[bot]' git config --global user.email 'github-actions[bot]@users.noreply.github.com' From 2176cfcff3cc0554dfc08a1b979d5e7a11a8940f Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 12:34:08 +0000 Subject: [PATCH 53/63] Update dependencies --- test/check_dependency.py | 1 + 1 file changed, 1 insertion(+) diff --git a/test/check_dependency.py b/test/check_dependency.py index b71b443725..0ffe7ad788 100644 --- a/test/check_dependency.py +++ b/test/check_dependency.py @@ -25,6 +25,7 @@ "psutil", "dataclasses", "transformers[torch]", + "transformers", "datasets", "evaluate", "nltk", From 8ac1583659d713e8029541f1877eeddbe7133839 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 12:51:58 +0000 Subject: [PATCH 54/63] Fix git error --- .github/workflows/python-package.yml | 2 ++ 1 file changed, 2 insertions(+) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index 3fb16d4ad6..3b5179ae15 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -132,12 +132,14 @@ jobs: git config --global user.email 'github-actions[bot]@users.noreply.github.com' git config advice.addIgnoredFile false git fetch origin + git stash push -u if (git show-ref --verify --quiet refs/remotes/origin/installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }}); then git checkout -B installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} \ origin/installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} else git checkout -b installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} fi + git stash pop || true git add installed_first_tier_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt git add installed_all_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt git add ./coverage.xml || true From caffde47776cd270e664e214359f454af5770106 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 12:58:54 +0000 Subject: [PATCH 55/63] Loose mlflow constraints --- setup.py | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/setup.py b/setup.py index 1b5597b35a..8bd7bec604 100644 --- a/setup.py +++ b/setup.py @@ -81,9 +81,7 @@ "hcrystalball", "seqeval", "pytorch-forecasting; python_version<'3.13'", - "mlflow<=2.12.2; python_version<'3.12'", - "mlflow<=2.18.0; python_version>='3.12' and python_version<'3.13'", - "mlflow<=2.22.1; python_version>='3.13'", + "mlflow<=2.22.1", # Refer to https://mvnrepository.com/artifact/org.mlflow/mlflow-spark "joblibspark>=0.5.0", "joblib<=1.3.2", "nbconvert", From e609bf26a475a20514d2045b55663021189a90d0 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Thu, 8 Jan 2026 13:13:29 +0000 Subject: [PATCH 56/63] Add rerun, use mlflow-skinny --- .github/workflows/python-package.yml | 4 ++-- setup.py | 3 ++- test/check_dependency.py | 1 + 3 files changed, 5 insertions(+), 3 deletions(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index 3b5179ae15..eac8a3ecb0 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -112,12 +112,12 @@ jobs: - name: Test with pytest if: matrix.python-version != '3.10' run: | - pytest test/ --ignore=test/autogen + pytest test/ --ignore=test/autogen --reruns 2 --reruns-delay 10 - name: Coverage if: matrix.python-version == '3.10' run: | pip install coverage - coverage run -a -m pytest test --ignore=test/autogen + coverage run -a -m pytest test --ignore=test/autogen --reruns 2 --reruns-delay 10 coverage xml - name: Upload coverage to Codecov if: matrix.python-version == '3.10' diff --git a/setup.py b/setup.py index 8bd7bec604..898e19b5cb 100644 --- a/setup.py +++ b/setup.py @@ -62,6 +62,7 @@ "scikit-learn>=1.2.0", "thop", "pytest>=6.1.1", + "pytest-rerunfailures>=13.0", "coverage>=5.3", "pre-commit", "torch", @@ -81,7 +82,7 @@ "hcrystalball", "seqeval", "pytorch-forecasting; python_version<'3.13'", - "mlflow<=2.22.1", # Refer to https://mvnrepository.com/artifact/org.mlflow/mlflow-spark + "mlflow-skinny<=2.22.1", # Refer to https://mvnrepository.com/artifact/org.mlflow/mlflow-spark "joblibspark>=0.5.0", "joblib<=1.3.2", "nbconvert", diff --git a/test/check_dependency.py b/test/check_dependency.py index 0ffe7ad788..526f25b189 100644 --- a/test/check_dependency.py +++ b/test/check_dependency.py @@ -13,6 +13,7 @@ "scikit-learn", "thop", "pytest", + "pytest-rerunfailures", "coverage", "pre-commit", "torch", From 545f9ea78133ac429ed9cee303a56b17531e4fce Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Fri, 9 Jan 2026 01:14:34 +0000 Subject: [PATCH 57/63] Fix git error --- .github/workflows/python-package.yml | 24 ++++++++---------------- 1 file changed, 8 insertions(+), 16 deletions(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index eac8a3ecb0..e7460a8ef2 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -100,12 +100,10 @@ jobs: pip install -e .[vw] - name: Pip freeze run: | - pip freeze > installed_all_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt - cat installed_all_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt + pip freeze - name: Check dependencies run: | - python test/check_dependency.py > installed_first_tier_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt - cat installed_first_tier_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt + python test/check_dependency.py - name: Clear pip cache run: | pip cache purge @@ -132,16 +130,10 @@ jobs: git config --global user.email 'github-actions[bot]@users.noreply.github.com' git config advice.addIgnoredFile false git fetch origin - git stash push -u - if (git show-ref --verify --quiet refs/remotes/origin/installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }}); then - git checkout -B installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} \ - origin/installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} - else - git checkout -b installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} - fi - git stash pop || true - git add installed_first_tier_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt - git add installed_all_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt - git add ./coverage.xml || true - git commit -m "Update installed first tier dependencies for Python ${{ matrix.python-version }} on ${{ matrix.os }}" + git checkout -B installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} + pip freeze > installed_all_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt + python test/check_dependency.py > installed_first_tier_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt + git add installed_*dependencies*.txt + git add -f ./coverage.xml || true + git commit -m "Update installed dependencies for Python ${{ matrix.python-version }} on ${{ matrix.os }}" || exit 0 git push origin installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} From 803acb319baa9bf1d7f6fde046aefeb6b7fe2864 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Fri, 9 Jan 2026 01:39:13 +0000 Subject: [PATCH 58/63] Remove ray tests --- .github/workflows/python-package.yml | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index e7460a8ef2..2e49f98e44 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -79,16 +79,11 @@ jobs: run: | pip install pyspark==4.0.1 pip list | grep "pyspark" - #TODO: support ray for python 3.11+ - - name: If linux and python<3.11, install ray 2 - if: matrix.os == 'ubuntu-latest' && matrix.python-version < '3.11' - run: | - pip install "ray[tune]<2.5.0" - - name: If mac and python 3.10, install xgboost 1 - if: matrix.os == 'macos-latest' && matrix.python-version == '3.10' - run: | - # use macos to test xgboost 1, but macos also supports xgboost 2 - pip install "xgboost<2" + # # TODO: support ray + # - name: If linux and python<3.11, install ray 2 + # if: matrix.os == 'ubuntu-latest' && matrix.python-version < '3.11' + # run: | + # pip install "ray[tune]<2.5.0" - name: Install prophet when on linux if: matrix.os == 'ubuntu-latest' run: | From 308965c7d3cf4f76571cf9cd94bb70f3468022f8 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Fri, 9 Jan 2026 01:39:29 +0000 Subject: [PATCH 59/63] Update xgboost versions --- setup.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/setup.py b/setup.py index 898e19b5cb..9b44b78142 100644 --- a/setup.py +++ b/setup.py @@ -55,7 +55,8 @@ "numpy>2.0.0; python_version>='3.13'", "jupyter; python_version<'3.13'", "lightgbm>=2.3.1", - "xgboost>=0.90,<2.0.0", + "xgboost>=0.90,<2.0.0; python_version<'3.11'", + "xgboost>=2.0.0; python_version>='3.11'", "scipy>=1.4.1", "pandas>=1.1.4,<2.0.0; python_version<'3.10'", "pandas>=1.1.4; python_version>='3.10'", From 028701119d3cc43346b35798d7fe0e232ae18a74 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Fri, 9 Jan 2026 01:39:59 +0000 Subject: [PATCH 60/63] Fix automl pickle error --- flaml/automl/automl.py | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/flaml/automl/automl.py b/flaml/automl/automl.py index 1f6fdadc55..c4c6e2dbb5 100644 --- a/flaml/automl/automl.py +++ b/flaml/automl/automl.py @@ -401,6 +401,24 @@ def custom_metric( self._estimator_type = "classifier" if settings["task"] in CLASSIFICATION else "regressor" self.best_run_id = None + def __getstate__(self): + """Customize pickling to avoid serializing runtime-only objects. + + MLflow's sklearn flavor serializes estimators via (cloud)pickle. During + AutoML fitting we may attach an internal mlflow integration instance + which holds `concurrent.futures.Future` objects and executors containing + thread locks, which are not picklable. + """ + + state = self.__dict__.copy() + state.pop("mlflow_integration", None) + return state + + def __setstate__(self, state): + self.__dict__.update(state) + # Ensure attribute exists post-unpickle. + self.mlflow_integration = None + def get_params(self, deep: bool = False) -> dict: return self._settings.copy() From ed4b1ac08ac20f9e125218f808c321c34879debb Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Fri, 9 Jan 2026 03:23:08 +0000 Subject: [PATCH 61/63] Don't test python 3.10 on macos as it's stuck --- .github/workflows/python-package.yml | 3 +++ 1 file changed, 3 insertions(+) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index 2e49f98e44..036f292003 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -41,6 +41,9 @@ jobs: matrix: os: [ubuntu-latest, macos-latest, windows-latest] python-version: ["3.10", "3.11"] + exclude: + - os: macos-latest + python-version: "3.10" steps: - uses: actions/checkout@v4 - name: Set up Python ${{ matrix.python-version }} From 36c07451333097c18a89c61309f3a3d3c3db963a Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Fri, 9 Jan 2026 03:38:40 +0000 Subject: [PATCH 62/63] Rebase before push --- .github/workflows/python-package.yml | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index 036f292003..8d5e90e750 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -127,11 +127,17 @@ jobs: git config --global user.name 'github-actions[bot]' git config --global user.email 'github-actions[bot]@users.noreply.github.com' git config advice.addIgnoredFile false + + BRANCH=installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} git fetch origin - git checkout -B installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} + git checkout -B "$BRANCH" + if git show-ref --verify --quiet "refs/remotes/origin/$BRANCH"; then + git rebase "origin/$BRANCH" + fi + pip freeze > installed_all_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt python test/check_dependency.py > installed_first_tier_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt git add installed_*dependencies*.txt git add -f ./coverage.xml || true git commit -m "Update installed dependencies for Python ${{ matrix.python-version }} on ${{ matrix.os }}" || exit 0 - git push origin installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} + git push origin "$BRANCH" From cc0089e7ba4e8574c85d9b716693f6f58c26b950 Mon Sep 17 00:00:00 2001 From: Li Jiang Date: Fri, 9 Jan 2026 03:48:31 +0000 Subject: [PATCH 63/63] Reduce number of branches --- .github/workflows/python-package.yml | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml index 8d5e90e750..9e88a4e8da 100644 --- a/.github/workflows/python-package.yml +++ b/.github/workflows/python-package.yml @@ -128,7 +128,7 @@ jobs: git config --global user.email 'github-actions[bot]@users.noreply.github.com' git config advice.addIgnoredFile false - BRANCH=installed-dependencies-${{ matrix.python-version }}-${{ matrix.os }} + BRANCH=unit-tests-installed-dependencies git fetch origin git checkout -B "$BRANCH" if git show-ref --verify --quiet "refs/remotes/origin/$BRANCH"; then @@ -138,6 +138,7 @@ jobs: pip freeze > installed_all_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt python test/check_dependency.py > installed_first_tier_dependencies_${{ matrix.python-version }}_${{ matrix.os }}.txt git add installed_*dependencies*.txt - git add -f ./coverage.xml || true + mv coverage.xml ./coverage_${{ matrix.python-version }}_${{ matrix.os }}.xml || true + git add -f ./coverage_${{ matrix.python-version }}_${{ matrix.os }}.xml || true git commit -m "Update installed dependencies for Python ${{ matrix.python-version }} on ${{ matrix.os }}" || exit 0 git push origin "$BRANCH"