From 07e2246f5700d553bfcedf1f4f245d763bd0b5e6 Mon Sep 17 00:00:00 2001
From: Abdul Fatir
Date: Fri, 19 Apr 2024 09:06:20 +0200
Subject: [PATCH] Revamp README: Add News, Coverage, Logo, Shields, Emojis,
Zero-Shot results (#56)
*Description of changes:* This PR revamps the README.
By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice.
---------
Co-authored-by: Abdul Fatir Ansari
---
README.md | 64 ++++++++++++++++++++++---
figures/chronos-logo.png | Bin 0 -> 100701 bytes
figures/zero_shot-agg_scaled_score.png | Bin 0 -> 325734 bytes
3 files changed, 57 insertions(+), 7 deletions(-)
create mode 100644 figures/chronos-logo.png
create mode 100644 figures/zero_shot-agg_scaled_score.png
diff --git a/README.md b/README.md
index 8567b9d..7ca7890 100644
--- a/README.md
+++ b/README.md
@@ -1,5 +1,21 @@
+
+

+
# Chronos: Learning the Language of Time Series
+[](https://arxiv.org/abs/2403.07815)
+[](https://huggingface.co/collections/amazon/chronos-models-65f1791d630a8d57cb718444)
+[](https://opensource.org/licenses/Apache-2.0)
+
+
+## ๐ News
+
+- **08 Apr 2024**: ๐งช Experimental [MLX inference support](https://github.com/amazon-science/chronos-forecasting/tree/mlx) added. If you have an Apple Silicon Mac, you can now obtain significantly faster forecasts from Chronos compared to CPU inference. This provides an alternative way to exploit the GPU on your Apple Silicon Macs together with the "mps" support in PyTorch.
+- **25 Mar 2024**: [v1.1.0 released](https://github.com/amazon-science/chronos-forecasting/releases/tag/v1.1.0) with inference optimizations and `pipeline.embed` to extract encoder embeddings from Chronos.
+- **13 Mar 2024**: Chronos [paper](https://arxiv.org/abs/2403.07815) and inference code released.
+
+## โจ Introduction
+
Chronos is a family of **pretrained time series forecasting models** based on language model architectures. A time series is transformed into a sequence of tokens via scaling and quantization, and a language model is trained on these tokens using the cross-entropy loss. Once trained, probabilistic forecasts are obtained by sampling multiple future trajectories given the historical context. Chronos models have been trained on a large corpus of publicly available time series data, as well as synthetic data generated using Gaussian processes.
For details on Chronos models, training data and procedures, and experimental results, please refer to the paper [Chronos: Learning the Language of Time Series](https://arxiv.org/abs/2403.07815).
@@ -12,12 +28,12 @@ For details on Chronos models, training data and procedures, and experimental re
----
-
-## Architecture
+### Architecture
The models in this repository are based on the [T5 architecture](https://arxiv.org/abs/1910.10683). The only difference is in the vocabulary size: Chronos-T5 models use 4096 different tokens, compared to 32128 of the original T5 models, resulting in fewer parameters.
+
+
| Model | Parameters | Based on |
| ---------------------------------------------------------------------- | ---------- | ---------------------------------------------------------------------- |
| [**chronos-t5-tiny**](https://huggingface.co/amazon/chronos-t5-tiny) | 8M | [t5-efficient-tiny](https://huggingface.co/google/t5-efficient-tiny) |
@@ -26,7 +42,21 @@ The models in this repository are based on the [T5 architecture](https://arxiv.o
| [**chronos-t5-base**](https://huggingface.co/amazon/chronos-t5-base) | 200M | [t5-efficient-base](https://huggingface.co/google/t5-efficient-base) |
| [**chronos-t5-large**](https://huggingface.co/amazon/chronos-t5-large) | 710M | [t5-efficient-large](https://huggingface.co/google/t5-efficient-large) |
-## Usage
+
+
+### Zero-Shot Results
+
+The following figure showcases the remarkable **zero-shot** performance of Chronos models on 27 datasets against local models, task-specific models and other pretrained models. For details on the evaluation setup and other results, please refer to [the paper](https://arxiv.org/abs/2403.07815).
+
+
+
+
+
+ Fig. 2: Performance of different models on Benchmark II, comprising 27 datasets not seen by Chronos models during training. This benchmark provides insights into the zero-shot performance of Chronos models against local statistical models, which fit parameters individually for each time series, task-specific models trained on each task, and pretrained models trained on a large corpus of time series. Pretrained Models (Other) indicates that some (or all) of the datasets in Benchmark II may have been in the training corpus of these models. The probabilistic (WQL) and point (MASE) forecasting metrics were normalized using the scores of the Seasonal Naive baseline and aggregated through a geometric mean to obtain the Agg. Relative WQL and MASE, respectively.
+
+
+
+## ๐ Usage
To perform inference with Chronos models, install this package by running:
@@ -106,7 +136,27 @@ embeddings, tokenizer_state = pipeline.embed(context)
```
-## Citation
+## ๐ฅ Coverage
+
+- [Adapting language model architectures for time series forecasting](https://www.amazon.science/blog/adapting-language-model-architectures-for-time-series-forecasting) (Amazon Science blog post)
+- [Amazon AI Researchers Introduce Chronos: A New Machine Learning Framework for Pretrained Probabilistic Time Series Models](https://www.marktechpost.com/2024/03/15/amazon-ai-researchers-introduce-chronos-a-new-machine-learning-framework-for-pretrained-probabilistic-time-series-models/) (Marktechpost blog post)
+- [Chronos: The Rise of Foundation Models for Time Series Forecasting](https://towardsdatascience.com/chronos-the-rise-of-foundation-models-for-time-series-forecasting-aaeba62d9da3) (Towards Data Science blog post by Luรญs Roque and Rafael Guedes)
+- [Moirai: Time Series Foundation Models for Universal Forecasting](https://towardsdatascience.com/moirai-time-series-foundation-models-for-universal-forecasting-dc93f74b330f) (Towards Data Science blog post by Luรญs Roque and Rafael Guedes, includes comparison of Chronos with Moirai)
+- [Chronos: The Latest Time Series Forecasting Foundation Model by Amazon](https://towardsdatascience.com/chronos-the-latest-time-series-forecasting-foundation-model-by-amazon-2687d641705a) (Towards Data Science blog post by Marco Peixeiro)
+ - The original article had a critical bug affecting the metric computation for Chronos. We opened a [pull request](https://github.com/marcopeix/time-series-analysis/pull/10) to fix it.
+- [How to Effectively Forecast Time Series with Amazon's New Time Series Forecasting Model](https://towardsdatascience.com/how-to-effectively-forecast-time-series-with-amazons-new-time-series-forecasting-model-9e04d4ccf67e) (Towards Data Science blog post by Eivind Kjosbakken)
+- [Chronos: Learning the Language of Time Series](https://minimizeregret.com/linked/2024/03/27/chronos-forecasting/) (Minimize Regret blog post by Tim Radtke)
+- [Chronos: Another Zero-Shot Time Series Forecaster LLM](https://levelup.gitconnected.com/chronos-another-zero-shot-time-series-forecaster-llm-0e80753a7ad0) (Level Up Coding blog post by Level Up Coding AI TutorMaster)
+- [Paper Review: Chronos: Learning the Language of Time Series](https://andlukyane.com/blog/paper-review-chronos) (Review by Andrey Lukyanenko)
+- [Foundation Models for Forecasting: the Future or Folly?](https://insights.radix.ai/blog/foundation-models-for-forecasting-the-future-or-folly) (Blog post by Radix)
+- [Learning the Language of Time Series with Chronos](https://medium.com/@ManueleCaddeo/learning-the-language-of-time-series-with-chronos-fea7d0fedde4) (Medium post by Manuele Caddeo)
+- [The latest advancement in Time Series Forecasting from AWS: Chronos](https://medium.com/chat-gpt-now-writes-all-my-articles/the-latest-advancement-in-time-series-forecasting-from-aws-chronos-python-code-included-0205d01248f3) (Medium post by Abish Pius)
+- [Decoding the Future: How Chronos Redefines Time Series Forecasting with the Art of Language](https://medium.com/@zamalbabar/decoding-the-future-how-chronos-redefines-time-series-forecasting-with-the-art-of-language-cecc2174e400) (Medium post by Zamal)
+- [Comparison of Chronos against the SCUM ensemble of statistical models](https://github.com/Nixtla/nixtla/tree/main/experiments/amazon-chronos) (Benchmark by Nixtla)
+ - We opened a [pull request](https://github.com/Nixtla/nixtla/pull/281) extending the analysis to 28 datasets (200K+ time series) and showing that **zero-shot** Chronos models perform comparably to this strong ensemble of 4 statistical models while being significantly faster on average. Our complete response can be [found here](https://www.linkedin.com/pulse/extended-comparison-chronos-against-statistical-ensemble-ansari-4aste/).
+- [Comparison of Chronos against a variety of forecasting models](https://www.linkedin.com/feed/update/urn:li:activity:7178398371815051267/) (Benchmark by ReadyTensor)
+
+## ๐ Citation
If you find Chronos models useful for your research, please consider citing the associated [paper](https://arxiv.org/abs/2403.07815):
@@ -119,10 +169,10 @@ If you find Chronos models useful for your research, please consider citing the
}
```
-## Security
+## ๐ก๏ธ Security
See [CONTRIBUTING](CONTRIBUTING.md#security-issue-notifications) for more information.
-## License
+## ๐ License
This project is licensed under the Apache-2.0 License.
diff --git a/figures/chronos-logo.png b/figures/chronos-logo.png
new file mode 100644
index 0000000000000000000000000000000000000000..5c51c0f8c05f2e1f69f2e44954a963b759299565
GIT binary patch
literal 100701
zcma%j1yogA*ESqMN=gZlI&_18bRDInyAB}TU574}5T(1jyE_DtlI{kjTe|r-_g?jV
z-|_upTpYk=pS|{8d#$JO&nNNH)tS01
z#468_FLbTma3?v`Yqb%|6EK{DXg6cRrN3lVk2S2gzY9oFDlo3d%V0#ky=mM&8a`Q&
z3)GNlgkep~X~hmO$HG!TO5Qix-iL^}DHs)N!@gPNFS^Z+cItp9aY*_#z-K
zdtZ-+2!$}&tLe%!hk_V;MI^AFRD3xaO4obLW{3Q8E3D38+Y7V!s>jM|A2-g*c7SDO
z=G6o43g#kbN=*X-M_P@czu{P$@q_14-ahj0}ee{}}E8a0CzhgM)tt
zhxpeq9GoOP;qPNbc-nui0p>Bl3=ZL+YoNgI`>!zI8)*H{Z%`WO*A-|ZX%BuK!_VDc
zT1U-h3;aT{eXZdD2Zu*>{|zspNO=GU2TC(jR(DjFk>)k9v0~OUw9z+WcCos@8V-)%
zg%>!qGIG=dyI5ITJMg*)kpFcBFK~Q+nuQ$v*Cmb?0_5s4@?c>bdm}I>^K<6sKxJUlGV*;v@vn1Cyo99*p(
z^<0>&9Vq_3$*=o}7c&o7p;=*;s?`@2jV8!reYXTzU%cS|>k;%@zFm``C0Dg@IU7CZ-e>k
zEMPH$sQfIytxpiuh(&1=))b^D+6fAx
zJ$*?3mX;or;iI4Kya{ghUl88lfPV~qMgJaW1l%0E%I8B$ZWJj4+g~{JPW3$8Ti|nE
z`(_gpH!a|tGgNIcNwVaP4EBKo{i`3kAAH_X`O$s+mzM*#okV=yxa|3F@9*Q&3CfyR
zm;Q?Quex)8+b~#%I~+U$7UW<3_y}~TJnj?Pzq}j)3%u)p_V3>o4=oI4z)FDsuLl$|
zgTDW-JCDGK2j2av&&K@k-W3=VpDMzI2mg8~)fpi(!JUR$|9ZCI%($A6{`F9nsRP~l
z*i-kfChD&V9kF)7{MSRNTMp6tSh@s4h{Ao|b5~(sQCZGn3Ye8pY7W*vQ1tDBGl$f=`%4W3og$k|ku7%weUyGYE&7dXWNutXezF?clO0Zqz)L_v|OK=Y|^g=;W;?^fcC$&h=U|m!piiR1%U8?4hTtt5`e_~OScRHg6
zw`dTqQ&UQ~0|oKAnE60(Qxdsswp3J+dXAwV+oboxsmR(e28JcM;Te}-!2?>yqt@ht
zknp7!wBY^1mALb|1BRc=tai(+KeZhzqBBH?*m)G&^11{&oUs
z#=c|6(ii%?yqp(b2eTM{(%DJm+kMC{|G;JaR#sA~A;mn>sF<0X8;4$HPC~Jfs_W`<
zP|-fI(jnVEn?9OB187CRDy2rN^eRZBP;?9EO)^S%!VA`y%jN0A5$Z>Gn;a454mb6T
z(`hLKy$zUD^ieGW*z3mTvhfA03=@TKN>r(+2et?p-9=x`1UIiZjax(YC8NAeqjXHO
zbOQN5pa)9^*3r8TL-!5KAIow%*wIfEsy1J|;~@1algxxxnAc8RiC+JYTbOt}wEgKk
zC$uPmqBbY0P$S>|U4lyDSjk8pEc$Ep$`4VCO$E%~Q34`}VE80j3cDPiMF;(3L}Ky}
z;mq$H89&-Hjl6w7d5js4N}=-ccAq}rppR|7%~Q@%jkb3xRLfIq*6*yCJGE4Dp%;e)
zgLpKi)tJS!?l`AeC>2{YPSNWh6UCpYn2$|m6SDH)z>4G>B|DJ{`|x|CSJs1R*rJWk
zOvW+dnLre~U-1ffOWD~uwCydduVo)K-|`K!h(uUAF09|}r>x8=TWK}A9l3Fkc$JAA
zAr*2IYL3-U-KDPFVgI-DbY#|DA^2IG&ZN+C9E_6$Re2^wE?Zm>Y+mkp#6*=5z!XBm!G%=fl_;65ps_ZX02l*I6D
zw+e@4KCh!r?CTk*-|=K!$U-oK^d1fz!@nCHX8BShi6
z4*Y~J=yjXXIGFv!EBuy>cj{SZgBxJ1;dBngG3XYxmhzQG9gkNfHEO-gD>H#u0sErn
zG}@!~ggN;*a`D3<`;Mcos#{J$hl-5ts<}#uohMV$C3>ovLZzrD5)#t*dol3;*r@Ac
zU9lvZoGY9N&wN8Ob&1dUqd6b08FdN|JyGqS{&vetq`q-=hH{Hoa*IRkzRT-x$sa3?
zq=!e_Y)$5!zwUTeoH*uHCBdYD9Z+rSq6SWr_1#E2&<)+r=foEnsD!-}HBfGW;YxfC
z7fG5aex_=33wAD3&&sW7&zjG(64@aLy{122^rk<5Uz#g=fPByzgP!$E_SM;bo3%-e
z-6t7jz^72giEnw8KT{?90~Yaiv|#Ut3>rRDM(MyX;eUk`NF?Z6AT9~#^he1(DYKEE
z3c&dAso+>Aul+J6_(n8@_hO`cgdWG={0!FLmQrU;36KOol%E<-7wvc^C#bdWjXCMQ
zlg=yav7RyPOKB;CqSXjvI#y2eNN1epBulxC&u8p4n*GV}UAbzF;)JluQ^a_PL-Pf#
za$cC#Z00ElfLBVoudGzT+nyJ@Cx^4!bUa+s1R66t@2NAHziSnJR_Z6LwxJ^BP<1P7
zpuU0({r3dK+)se#@d~G4eL$&QfOw*#smc#*PSctzAFN;geM%lINOv-npLUbtdnNE7
zLqF1TDuS$#SvBk6NB=heY=JL-!dS8S>j%8SKIn*sdW0AB7jrP@{p7t5;pLo`9}4za
z=~;wTj%Ugu}~^PP%42zFh
zZXMj7!#F##rWsLB=#wl!EEU_+&A&0F=xH^Z$#<$YcG3XlE@i0}`Bo7!=oeJcNA@@_
z=bdfJ1PoXUBs@zk8lep)jxiW+$8>USIw9OO!X@fzw;VGWa_-0z7m^7wMp|_L)WRro
zSGj)ep0pGej$jx%r6CZISmE$#Y_sXK`lyYo(PNUKG@C;)MCZdZ0^0WLP{^{rsG)X8
z#$g!xdg&b_>}FL+$cCuiwZNsEZJhq~rBgjLI$0|ATn@JhaOQ-_&nar)3yKS(II5^F
za*S$gDo{iCYgO3C3U%Ob{hJo2YFsOSBtE%X+-K}UXJ`|-ni$%)hk6z|^DC%>*%>rX
z!)8|P7WzB(?WNzg7mv;{`Ra?J)mkaE9{mVz`BLjT8)%Db(m;!N7Gvm%9~#KL-dG_>A};Z71ay*!2tu7JnhaG
z^rWjv-z*8JUGH(txW%C4t;|WlRzAORyD;Fp0go#4nrvN6!gl5roUaFR_dxOC>q^T0
z`FhKq!)ki9JS7!HTl%pQ@_Oz=TFp#-y_)R7E;w5h&>gO>*tHI(Ra39zU=t4qv+Br2
zI(~}jc(K(x6R5345a+{C+HodZ{gV_<8<@*MPrw*4^)MY1Fc2}^&OwVAOA;-Jl(TC~
z`Dw)g13ibFALuqON&?V|slvkJo3H9p`ThInu3aj5wNtHc+m-qYqF9WIFY}QA)bd5A
z<$Lp}@I+0$%eUDmgAtAW;~CIiQNuJWkjHwvXUD&DQH8x`)J
zY7pt2*P}&PIk=53pWDJBD%1yC@F{g9xVfDx`3;SpOSFXJBq>Js&Tvse#eWC2vT&c3
zjrV;&cKl|yV=-@R7z32*%_Di)OiHfQm8iefUL6qmY60teT{7zyM9^cF!hE$tb|oip
zJ#q(anW?a~38YAVwCQpO`6!w(Go4=fanKSIpHxvznIhb+Y61dW)lpqi$XC
zOxl}l1agnJwbu={v-=;7X7f1(_+Gg0u};@O^1qrP{f=pE)}RWJfH?!oP_!Q^yV=?)?o+g3&XZ6S3FZmktAe8X(y*@30n0?7lK}yb9A|}2**BM8P
z>r|nSSEYnzGLxk<3SSf$eLlfqiUR-u;*nCP@WKoqPjG5uRVs%39-9
z7JpJ-lj
zUWam@$jL`SR>r?@Sf?;I${~ttsVHfE(f5VGLdPBLZAp)(*z
zkU11G!Un=j6#znh=(Zha9_as%2v2>
zJS;w8uX=f@ypzue-Kr7sWk(Ql7F8CJ<`zBXg{?@0O^QpdLr;*UU^cQq@NdXz{y2=z
z$3u`D6GgJcJ|oje*W1@?8vBHEzRI=Sc_ec2
z-6wfQtdB83TexA3C!E(WYR@W`x6s48?;~|Js-e+(o$Gmt{mI-uE;tb{pF{e4g*2Fo
z$-!@1!u3Uw%4YvMiD$ELU(mIc839>#H6?hFJw?4)9M3MW-d01(?uo@`gjf7GtL0{2
zKoAFevS~4>-%#ba9;!S+=Rsw;^Jvli^6_fWOMijmP%7BGvx48EvbS-R+
zmDrpC{%kgntU&bl>XxtfnagmTzRKW-$*9v^a4|JjN1tH#O_pRFmZ4z!zD0XZ!CaR<
zVG)*JzSGxauw#GGO!INEn@tzGkNuM(?UT>lt&kHz%SO$5c#^8VeUWkjY%Cnzn2GW&6QGiVGm_MrsBaG3u|Lvg*7(g
zp`TLY(3rusOvF0%nk3PL6HzOK)7}L@E{a$h^K1`bh1M{zs#}j|*Q?J;o$y243I*i9
zS7t+CViUuie1@}~c=I!`^TX-_nRAArn7OgwU+5z54Y-{~on-b-3tFfmeDN4cVpO0a
z?>UAL2rysY0-2~sPY^2{yLvd>R{Tp>iv7Rg6tzZyawAkXXnm{&SKXucWDK#)F2ZtW*nD__<@w<`m>Z=(?Yr^%DVnt
z${GO(`Xp>oJC}=8klRFQBOg-SDce-@I1mFll25(nraTLX;(X{X?-I|7&}3$Uodm2@
z5_=4JYj@*zJ~MJv&J>_;jVJgpmhM*X`S2&etVe$qk`z74sfMR~^J}@bU4qizz#oRu
zIW~7CN(P-$u-t_9OB9z%2L}Q3sG6i@fbO4+p>;St)r)RkwSU&Y_+wdpO2L2_tc8Gs
z^E%|i;^}$>NoK2jnM}efz(lGj!y;nJXZkHZ=TMM^+d+d8)r)DCNFU#fe(N{Dt3TZu
zS<6&QDblDfI@!vhM!Wo}a{{LL=Z-;90Fr_6^4HgI})&=ld2_>zJD*
zG+1cRX6SUeCbea*N@@G;ME42ul&}v$xxiTUjIpEc`IhgfR#&OzV(qL!GcUd!eZhDN
z78i&U??An#P-YNguEof}xB!T3yC_;A66$rUQ$fizK!#F{FMzG738v#P*z@l=;v#na
z>D}+0q>*U0pN}Iw8}2;@c){Lb*)+Wpt6cS}8vS!O$+iJJP0QPoqWEh}daBP@8P`
ztKgoinF=SAuV(C|ElCEzq^@*w1nr0?HVL?XCXGw`!zXBK}wdwtO0Ta)oz3J&sn}
zdN!_AW~z-kgC=}6W<~yaf$+ZWs9L<6G$>pz0|pn47?H5+a@&M@xR|dfSz0qK`9u%<
z1>z53VO5DuJan4$Ok}G=Z;J!gZ6t!_RBMc&qfI`R!Zk(v?}JVXfWEtLr|e!pMUN;X
z|8SEQ$RZa3CAYB45Y|4!)kKWq~F%T@t+dfTE7dJ)vQ2w-i3M?QPxVILvyQ6k*)rRzHx=vbEY((f8%y#-&+19ww^6!NVG*_zv-;|#lICnNK9+o|(XHzz}M_$*Jx0LvrM=|wGV5HPreD8&M>bP%p;1hj3V3~(x
z%BlUDHz0wh6zjCQGU(Z(vmo(a=eZzhusjJmW9YF%*tvf(<{e(^7XqwKXUef1Xthoo
zXKu4{Z|J{kNp~e@TOg~`msyN#iNA48xYA&9%H#*$OKd4kZD~y9ioc0vGe_J}1m5@P
zbGS~hCcE{AS)baO)`@f@J@E(|9Qj-n2M{Yhb0248E*CMITA%~+ip3jy50vC!qDG;J)r>W6k3whFHJ=x&wJfeDKy4hZax+G
z!^_e7K*S_3Ki4N>vP{inXwJe?og=j+(c}qY$#se0jlWFTWyQFoygm#r@^CO*K3#9w
z9?vmx2bend=6_tV=ua)V|0Bq|GKixEKF%R$h+6y*Gx=3@ecB`CJUJt;w#3!R|4=}q0WSJN5s2aDL
zxR0BS2L&FYfRF=yc0_EA`@+TFc%h;L9;pos5+8gM^`20*j=5G*Ne*)P6Y+d=a&*2;Z
zM95OxX4k^rar=_9lq&550GXfyUS2V=Mg+3w3_bsudO~6|<1~fj1m*gQ0FS^KtOZ7?
zlg$ni_Pt+6noE{TtE?ZjG2IY18xxOZU454V3m2tr
zLL9cHRD|0hneFh8Yn^$2<}Cs0wf0pHtpO_J4GAzS8?0+;)olyg^WJLru)?S$E3%
z4%^~a=H|#cq!>_>Fhj04VySE}lJ{4;a`03gvkUH!4HsVf#NGB{*<0WKXh)kxd!O3)
ze0TM8qW1+KJIDNKu~{ABi1p5Mm&2U7%d77agXqW8PIDOjqgmyMD@RedH#Ge&yW$s2
zW|k;!p}!3Zu+QVHvEocTnT_-6(Af`nNlS<$yF%_-;JwRa}_{9P6Zmf$5X=DMphhZdqr
zdPd%bhvQ1w>4+eVl6gH(CT(PtA6
zxN2HoJ
zpQm+cI6$oOZsBPUXF;x484lx@BL?p)7UmTXI585c2nio7s@{w*bzkgU^g>qUMEVr=S6(@Yl-lfg)a?r6!DBBnEe^_pS4p2Zz_apa$Zo
z!2{zg+ts7(R-tR{)1-(&tl%-5%b>Pa(6R{&(VMc}#Yz5qDxivS$!*7Doh3Iop_PKw
zulacue;71GfP%r@zt7i1D6!6bQsRPKtWTQ^#)ppJCpa)~RG_w#
zAq1mPihO0hbUYqm`AVZBF{Hln-U1GOI@jwa)ZuoE^To(n1bjo{By$8TQZ3&j_GtbG
z9SRrlM{;=dzGNL1d?KpIh2J+9tGFg9jnl6=1Ii)4)`Y1nM72@l=lDvrj+}&??5NfO
zLeWDYaF{MJngvo6@Z*$KLD99jx3OEAf
zpU0`!?Pq*n8GkH9C$eSD&;7(GU+WvI%UZpmWJNe3Aet*zwD2Cr3~c=zZ6K>;Q<{!%r_WH%
zAnM@o_rf1VxCs(qC
zbjdr;lBntVDBsv^`3ZkQY4
zDnH`NR?Vv^RE^{dnsG`Z1FGV&|j%Z
z0#>6qQf0v4YapdfBgpmq{M=QO1JaC~r>Og5@JNDNX|)>*2ZTJQ3EndQzV2Ogz7vWY
zYcTbGbwPwkB@|+*zs2=L%aF9HERXv{M3l1NaTu|X$?4Oz@^%>&1?)$ldenP(+qYm_
zOJc@V-=f0E3{2+QVaFql73%Oi81z#Y;wgUjYuQK{gqX9`sF*nYJJZ15xhDj|0a*Iz
zKzX0Z84+_TLOwMuNFf=B#^dNH@T+Q&80HvUz1o{{O1?;$j-!v=PS7WTn4xyWt%?>3
zPgp?6PZgHd4#|kGWasUoiphe{9}Zck8j)7X-sJSp+ENA(LaX?KYCiC9A8fpPPrXlK}iZguwoAS87w^E!#q}(rO$}g%a_&jRU*Xb6$H&q
zcaP4N?7Elm-TKW72M2rTR{=CzEO-vIEE7Y(*E9pp9wclqMnqFiM`vrh
zfH$2(vr3QVOqVaKsfyR#(;L&DFzIm@yAy4^TEL;Gtrz?y-}iat2)pwq4VdkoG5QzD
zK;)H|U}i>RBXob~+Bw2t+8$7kPJjw6s5
z0l?rWLtstGtv@&FvaJc?r-L|)w!hBQ6fNEk`;Rk3=c|@W2TK}B#@g~a@|vFS6kE<4
zjp_+K!MgqK)7s+Q>X5o8$H$FRVWz2EV(eQ~^(dCq*sWM7Z4LYZ-M0yiNT;6f`*k+v
z+O2$zTW6C_c4&ljY7u*gp2^#srUPOgu8H9+ym4eL_IQ|_2K&*8o*Ij20boCWpsc)~
zk91aU%dxN)_t|>BKKBkG0=xdze<-kQ^jC^42VSmz3COcj%lYpZ@9z(y2PVVUS}13}O7<%=liZ9!
zfnmp^FDwZY#R;wE_4J{j_-P*R>*mU?rTPjVB@&kOxB&!x;fx$-Ldk*P0srW|_sJMZ
zhwW#+0RvBi-vS2ssgALQ_hy`I*mqeq?;|~ye5#Mf3^}1}vR0*
z^Ch2vZm;BIHhF0Q`sB&*V@ePnO4b+Z0nhcxn@Ub~$LI;PNSIO@wzQmVWal_nu)1#V
z0G+(3)+e6^L^j0{HS{DPKSXdHEaX$nxR0=Z0}pv+Fe|+^V5%UD6xAw~XG#IUs(1SI
zK!KJXT=`?06aj#$+l#a_&x6U*iXd+K1s+TJmWfsVK$HP|6*cokNBBn|VQnYpI(tap
zHjJs`HS%rTND4RgpnuyK)wUN3+9c5EFt)rk92+S_p_?d(bQ@!|-JPix7@ZjKwo!(9
zR3!YM0@i?_+->|JUz{Oa=HzB$&p|nUeGhq$Qb|MJWC~%#!
zFO=-uh$}I!5lu&9vO1|`X!KH|MFAl8{Csc`DMR&V`riF$&nQ;UH7_LJrA`&5NTus*
zq`Zg9Zw8H!O_d-Mjftndx~JGfYa7@czbr_X0bb+QVVp~8)tc!(7SKNduwDf`84({c
zzyxqzTLn&*HfNDcmV%
zDEf!X!Ylc+?2;I5F*Af?DsbZT-Ocj;ULS0;Q1kfVFJp$JrGWrqU2D#r(fKR>m_qR+
zJB8|YH&*OC0_C#t`qZko_p!LSN5%ZxG(7}MI$Pb-oY$>bnl+I5zY~D%(-gPO#Q61h
z6#M+2Ef5}2ENv2Aw)?cs@n1h0HMntpLBrqcnZUXFqlJWcUSpAR}Lu7Cj_QroAkdXJ`4#yRlCcim4lp$nGVy4
zE`tgfOz1NveHGO<=d+%CZ6r}7KZaSxVzRXNJ|3b~w_Q9Q=b7(0tbi=#0j8m$5oSt_
z-qM%qd`phDtIr1@lpeFx5Gb!(c;V&@M~8ta37C)FqQ~S%lXIO#3ymY3FOXvj9{B^*
zQ461J$u!@7vd&1IJ6)-66#M(OmevWDeJ+WKcrFo}P0C+g^nong6Wr^qR~Bye(1jo+
z$p--b;&}f0y{Ve-w#Cb4UdoMbo<^~kKai(HmNFVcyHL3jm1VN;$I$A$+0)U+_a*#(i#s|OHIL7@o(cLQG9qFMAIq}HUWQHheAlaVy@*F#B
z9&zNc;#%|7wnVHaEc87bR#`*asZb21j!c^;bCJRR9U4&OjlWM2{VI
zRjumDKJL^m8U020DYCj0m_yLDuN19Vh4ADa9XKOP>oniJ3;*~dQl0S#
zcs5_860J{hr@c#3G&gPkq4)u?asBSzw*R-5lL(p_Ku%Ir{F5Adf?Ov9ZPssF*^#a>
zP#uiJ`hbf7;BQQR$FHck3W$JR*d{{U#8g|$3%@eiHNtjhtO8sCLdO{uut#t9&qh1n
zwVhLzs-cK%qilE{$Kq`-BeKVMUwtuBgV#2`UnGlX%HRC@Ms)}n`&{N~5omxea>7rW
zn4czg3|jKo&$6ccfzVI2+YJ0>YlLE+YQKDUV}LgrG0GCaaJ)b7P6g-EhTu#@MSL6N^QZaC1_$5WTH
z-+^?Duy8P8mx*1!AA{JpP!>{SkaBTZw$b){TvQB#yrQPNud_{gLEWn6Kn%?5LFj$1
z)a=g`xXkc$0ULQms*Q`g;3|!0!LWE$O3wG=9CaxVQ2PL(4~?+glXSY_={(=nR|^Cj?B}a-F~cxG
zZSXT%cKyr{{+pU;h`#dNHxeQ7bA;If_$^V0hHuihofBabhFR)jG9RA>1UxIdyX(Pd
z0M@!~KfQSgdFi_`m6amAzRyRHrknU8ng&SyLU$9&*yKd%-ot^o5ZVW|)pI}6Jv8IG
za}`gWaWw5QF)#vMq-NeLCD0I8NQc|y#B_K!O_rT-{?yioirY9M&8B0Yv=1O`)&$J#
zKV{e74J8wh&x8P?A`Kf1OabD&uS
zhL>M^7+wcd$C&zV0e%B|-jjMbuI_8TtdH=Cz6>a87MHF;X8MI}-bkcFZ+0E)??&=$
zj4y+IPX6++Ek|hXUYa^F&wc~2e$+4iBK40F(x(rc2)Wty=P{Uwm4YE+_A_(NUPgSz
zMpZFeI5L$4H4T>zVdH~}n+M!Afl?wX6r71a11FR20{wot3Ex-?`VB&lxOOx?k
zVL)+k5B}8ZA)2`So=20a6f++OVbOjOj3j0Z1g??SGz49li_}|
zy9Dr76lwUrANK?`y+_Sw{D()aS4spY?HG;^RkKKu^OTo*JCQY{HBXZi%{<=w{i2b;
z0PL4Fu9<6%NQo-2tgM3N0y-={QI}w!T|`P5Jz&KE)cAA@9Td`515)rJyE3*evmzV=
zy7mnJ06oORV&iQf!-r@4NXryy=)Rsb4-C0#qpBWRy)TTVeFre@&n-*dxCskkm9@5|
zyJ|LZ5>aS{FGOT)6u)V0DMym!GcB#Eo)t*4On*{^XWJ$&j6F+Sr%u
zj+0DGPJsmFF=8LpPpH80Tp!&-Q3RI@ORa|qDNuJ(w(L(7v#{M|Q?VgnOwa;8`NE~$
zXBFJVAdR_vQ`y#{c
zk8OK#TN-#2`(jf2N6!2`^SHaDkwEwyBnD{&0s1{OAp#~TbBkIuk>5-lSXUH2K8jeQ
zu0IG@QDxy?p)(&8fjspJ%plE{Z_foy03lMem15L`-yf>ay0B!@-X^6e^fRy;1a;#qgByn
zaSQ7wxbJf2&-aC@zZdpQN>ihW#bp5gO!vO#3Xn!FK-LDD6#0*RDQ;N*n5^5zhn!0Q
z7wlg%?yydeqg->C3<+}+n{*lg?-=+8Hs*(U;0o8=GO@&J+U9$uJeh1#RIqGFb)
z>f^-^jh#o%DKvTIQ%e8gZ1NQ@mO}_9;^f8Y4*PTt&GxA(h9C&?*rgKK`2CLEyoK7eiHcv
zLDi-T!l^L#UL)y!Bn#!H9|b{1QnwRl(R>((Bc>o?D3{w&z4-oyQ+RvGV#}=AcxtMt
zg&N!Rs?0=_#B0?}Ku!d<_-Pd7y|LG0QT3FEl_CunesP{AlzFBk+LTJRBYR%t9AVJg
zTUj+h@@{Q(NHahlySuh+z>WckQ2l4CTK9VS3gn0vLC3<@dV}VOD_T-BmgA?lpAtY
z<~}BgvjpWUwt40k7Uksh4k4og^1>(Fj`SM~$a)zrtZxS@6
zNAL)TKsp}Pa1q;jh>uDMvt%#YYz0YrF}ia%GYGkG(d;XfFH52eaGxKQLx$ExrZrZp&>WuZV(y
zlIU(`>}ocp5te+e&yInM>4LG-S21NE@>}>G;Zq3R8hyBIoy2VM;7|+l-FKwb?B$u%
z#~F%mOFQF>WnMOHQ8L?%XSM5D>&rVTx~gAl$y76rT($~c4p%k+C`!ZZLPrJx3&(I?
zEcxWNOVV{i3o_>+S~ISSlUNV0^oCj#crsl}-vUc*ktz
zW-Zmfeaf#T?Bnw-<7{vt4I!b-w;aqAozDOPQ&;o_1vy5ID@y9d#axzI#N8(kW)hn&}v}{mU$iD
zV4s@P*kpGXpK4+??1;ZZ;sF!`HN7X0!2{!y+@dPTsqEV2aDn-uH*-+R<
zo-{;jhxD=bQUXpfMA46$aW>f^QcmD-3GL1z#)1{qR69*Qyj{ra@Ks4OfRVI-M4_bf
zg#v_aP<59#|AYKEUe_xhb`ktGks3tZs>
z(cuw9ma_T;r?H>hE8%|8x&ZTYwR_Z_z66jqXx6XMk*B2EY)WId>u^ualVRAjX^>Z6
zR3`2E(9}_4DSUi-5!os$gsu$gV|C(se>9o6R4fH7X6L!5G~6te$-SY4PW?QkxO7k1
zZWme}hC+L3Ngo_vY+G5VGAC~PQ5f5B#Fwl>r5O2ROokH}q;hwxxsTi>jWx@7v6m!h?R{Y=02
zc%^s1MLI7I5e8K-Edu_!+_%}^|C9W0`yoUD;&bJa47^o5Ha=EZAGZ9vwgQ_RF+@{)
zTzHNR?024sfP$bRmD@dg+8GBBjLo84*AF>ur#sXX>i8yLrnZzF<;j(xpKM=Pt{!IF
zGv&;Ke&97Gqc?dBp0ig}w5BXooMqh!3~v2c8VXHi*;5|#&>3N`yghdhxL||wGbDcT
zJlq}$WI%yFdUFw%`Je-+kQaJ1R>c)UfL(Ffg;9mdtz!Ln=+F3X4ul7V7Pd}IG-2L3
zfNE#OXjhz2#Y(YJtGE}~&qNn7D|D(go8cpg00^&1V5-?4TK!b?4dQXifk04ZE(1iO!LzlBu3-h}p#Jb_G>+aF)$X4_n>=mcE;O8E>sQFm3wZO~v;>vLn0b3q588r5=2Ruj0Q0=nEhWxVqXur_0Aj3;-nvGXjm%XT*pz
zo>daXfQr4RI^C5pqaA5q8)aFI|3<&$R)Ns3w(ihAOmdYH7YWqY*8^FRX+B8NCNRS*lDPJAVe?
zs@Oq*n9O{%((?yKXaZda-JfMW4@eldh`(>LI8$%7>dAqRcHa&DtJ
z19OH8MHwF8G=m_gQ&YLPGl2iIDO
z!qp?>L@r@qPrEL=Z{5&_cBHEF%}>w7Y6w88{3?$>!#Y^lBqPl0laYNEsKP-8K+Ar4
zEM(9GPMG)1A^fqhHc2f|#_{tZ2iZH<0fH&xtwvmx@P$!iP{ZS}RXMj%D7E|J&3rXF
zv-nK>23ex14v$AGI82(Hu7;91W#w-0H!Li0R21kkJ>Invu
z0|%=v{N01xw-~7$x?Q*%_I?d2nlSeqy8zW{=nLByjf<(E&v7>G4oHufBr=l=-6O@>)8hsx*Ef?%yL$NUEedfzWN=zp
zo9{Fs#ej}tokGHIq6|wuv8{!Ga11XZ;fJ1B(b_N>P+g=2OYw;#`Qev6W)^Q_1`z_)
zwEWWj&E59bnaEQB#X_^*$I2j_@qLHTp9!kdxIVVYZwclAJj+05&=fh`mhzjWLay0(
zo$s$-MX(Hu4@~EHSe}~ObsqCRbx?)3D$-G(yM5+ZZEIN+Bt?zM}r>%x%yP4Hmx*B4r~={M)9q{HthK5nDF=n
zGWLsaQI4+pha}i31n|>TcDqaQH=MWe+j|AZ^veK~^^a}%OaQu?-Tgz)(u#2g^JWUv
zFN2@d;SKpX?e+=m%EswiFgqIh>c)R&$k+bycbp!@C0)J>ATlzRHP(fLdfy5
z8hPsVy~JVieKgC>aeDApE)-B9K$a%!XG4?f0fY)O(_fjh+WvaVdiT;X8a4Z>b=rFF
zF;}C=>r7RgJniZ^n)#I8>UJl0x#V79LDiBI(`<;C@#(rZ<04zW%>^4()#FN#-9u6#
z(k6Gskm!G(47_L%9lD0RwHz(QcBSrz?
zoj#}|6Z9(cy5oC?!66QTpE5NT@za*AZ939#gqFANFQY|exPjt9*j<9qrCTGg@3Gk33Qxhp}an?jqem9JQI3UB2F
zJQD%gy9i7*I$)0ENA2XO!wvjC@&9A$s-vRn-mVPYCEXp;-3;B`ARyh{At|5&(j_e=
z-JOEc-QC>`E%jaB`h9E7It%_8*5RJ>#D4bPmlVK*-rp{Ev?ss<Gj=Hs`?Y5Ai{uU*<(=U&3oewGA;y-6FDuXas0|yQf
z1hJ`+w{;s0ZEL^%8P9PAK3}LO4e3=Ym;8}B{9VP4CkH*$7-qt?skOaja0n}Rt^=gE
zaICU!NW&is5bx>Cc8#OU(B6>c0&FmOT;;mJ)W;~)PrFlHAtVC6
z>VL5(&uPI|;2PY(OnMA4pL!db(eWGXZm(YP>q-ZVTTj^DueG}O`$%8Jm)Z-WFe)j+
zvK?;#LJK3w_HI;YWo=2S)ViM)qY`2OguM5WDuYxkO!|dRq;HmWvAU(V0w%WF%6v46
zJj{q@oNVfSbx(GE2j>@`l5a8~pAq||-=k&)G|6u*@?P3AP8XXVOmb_Z)GO8_p%3L9
zC8!rM(9uouNUB*TBQl3WTiTu5j%}&rddPH40_04a{>}iVUH0ic
zvW$ZY0My)!U&aufB-qtQgD$ouz#z;Z8bpj@3A^Wxgg4`^V0{?
zt1Jyc%sXz}J2|BI#5Ze}QwT!LSoLr67(-JM5Yf-kZ~8|^Pl7n@eJ4h8Ege_!){k89
zji3P!<5>S-GI|u}PSJ5^e1Yr|mnjLakI-%pCkrii-Ui4-?@!eq@p8h*+#}dj2Z7Go?{v2Wv|P)#
z0q#dstgU75#**AogVKGCHjc!ZIQawL(37ZTBA?Au8#L04_os?k
z#|~IN^=XYK$$mH!^>7_AYL@x+g?*e0bdMpcAluwgauMZp>B5`>jFcdt1f664b)4`1
ztv9x6j#GEfS;6Ld2%v+BpU>Bgv?QFoVO7%0y4vqPv25vMuz;cNDU>Uk
zjd!5p=as6}`?)=xEv{~M4%
zn8^d{2lXoBNaVRpHsG#ujwJlFYn^2u0rEy$K_nDfzQp$ah1-R
z0Qy?-r*rW=F8>m}SU+4(4GTy#IzSQ1uu6vYYaQ(LYrh+)R{KVc`~lQvH$C|{S_yts
z-ZtdkG(#(?tMSZ}$)Kd^KGxM$k!sJL)3}%13UK6je*a_CSR`Usl8aXTg!?1ztUD}UR4PI7jfc_Q$Ugb!|%Otu>VO=D0J)V>#DiRT0XM+_amcLg%+{_Jz;sPcq{D^
zZJ_x)`TN@scs9IVxtvrTk&*|{E!miBP>hj0Ksp!n0g8MWzYNjVll==VKl7a`!Os`!
z1W@vmd#BI6#c4x%!oy(V
zIEoEWF2Qw7ZKO0WV>rUe85R-FRxk-F?hsBvGU*;!syh!wm602X{R%CfMpd{5W|RV9
z0lBrsh~d-qP@!;q5?|}ftU-rnz0I4J;x@p;qcZC5Zs0b1O~5v8j3rs+&3>0rGd`yK
zoqcFS%2do{hCY;(CPU_ojhz+FBo4j4N_P1ad^Uk4g2QnX^WHxSk
zyeS=s=i6hbegfQJcV4zq+Ce$dJvgbAO{0x@zDWhmNduN}_Z9YGqWd9XEE@
zf}?@A(9i1ybcEAoi2y+NUGklIc10Ko$q0NxYz9vN19HYA{J}%<5s=)T-TLuah2{tx
z%PT`r%C^-zIk%P?Y2QnYMi)duess$OqMz{RW}`L4Ro9(qvj*B1m(Cmw%Y^AGwSKy4
zue18y;_sfSfJCoajy~
zEUc5gW-V^1g|vO88Yd*`&e2WprSNi2IM6-&>34RvPMOZhDTuhbz6&UgrV{4Qhh~ST
zo!$onbW-X22>nhYTL6%=Ll}1Erx#Sz%cCuyWq15!P(dUsht^l6l09dOj*9Em#}f@u
zjkE2F=$O~A0b;owyTo4=f#;-t9C~KHFOf+Yayr^oGs$TUYlBjHzrxPfVWQYIUpu&wcRjUjC{;K^fEhqF>)gw#(^hs(
zX(AzyY^~`WQ+WwseIVR-OXU8}7ZBXV`f9T
z%@j`fnqq6vRnDi|os%qaRTzEZesT4v2@Jk%5y`T{fXbbSZ1mmM9}a@0MsynWH3O&;
zOObSvZ@JfX_wt5ZkEC9r17~dqh}p=zUz#O7-yU?@xG$};z5rxlfIX0p>7?`ULO5r>
zII?kXAkhwDnkck?h9!=Cot$LYvBs;F+*0zOJN2lO-}>gwbn)-qpU5iTx1_6UpS*0;zwpE%wa}MKS1S>oTe%#l%33e)WnvCgo1W5CSIhH%i5eGmA
z>9e=ylyCk-x3=E@kU6vF!s?kixRf&+P#4EErJCzF(}q)|$pV7tZI)7r-g{55?bXn@
zF$^9&Abj5-GCwI;0hlN!_^pp3`NR!{;Oh9%lTLq!{c63M8igG=?sC2QFNg6j-`lu-
zg7Rqwo;9e~=6&C-6qaom(Hl@{i2lg^mhQDw1b3Mu;Ax0A1t;DLRk?pb=k1D%TjAY4W1KDEC
z-#lz4&P#Gjilh8t9KM@=B>7Qx;oL(h*mrf*cCx#M4`IKiX&NRf-a7s@)Liyy0YOj}rFJWycx8*mHv_+;CPWseoR4)cMSdRP(6LhO$&7h2B^IV2}cS4O7bH4=quQ
za>_-pmM^%deBQLo>^p07H{ONqb?JcI%%Z?zzGqO^p}~9SfDG+<=DP1zL5IPqENic{
z5rU&o$zU9K#B=b2E`6sI#4-q2Lfk8?3Z8Vz>t>l20E5`SZWF#%j`^>eF?8RD2C@@YE`U@zCaXNu7Z%=RP3?Y(U$FAr#V-$r<>`>Y7lgh$0eGomCWGmcj_VCcEQV5
z_3bG=K5+mV_}#t3dv6~{tb)A5M1RU-Qx99f@SlElUM@#-$F-dn{m%92Z074sy#|7S
zy|%~;FyK>;AQ^H?0q#f|;|pvW=HA+A#hV@M9#-DRiKkl;rdOUR(MjmUsr45%w^DC(
z*GyhQ7Mp)=owKaNWmjvAMKA
z1Kj>_G=ZbnD`H?bP>+>U(lF6Tn8Q&J>j&esD)(EG^8QtIg$AECA!l5Lm1T;&-4p
zqg(Yqy6NpVP*j4t?Pq|n8E`(xV7hK|TXt=}96r1zYUr+-3I^tU>F)m1RUmibe+p+V
zI@NNUOsJ*4%6rF_l{vu~;uR98h)l#m`Jq^bVUj+Nmw1Belkikvu|NBM59I2~m#ZE?
zODNdl+`7yV(5GjF`?kwtvjjLwa>-UdXo|&5g4?I&8jfd^S#e%l7V*P4BFRF2R1ugv
zeDxEwUk;QeO&Uxt)32^`a-M!_$mGz0-J7Ps$WFYOY2wCCK4nn<$U*EIJL2AQ8x7d0+_SNd(y*fW=yY6hDF@DsYv^rm^mIlj2l_bss+SOMPDr9aJ*d?C
z*O#}8&-B{z+I(X8GhsWG_EUrIhygs(7jCV7O3J{t@7Zu$
zrWb(iNjB5ERye_g>Zg(oA{|y9bH6u%Tcf~DZyHm&_2^k{d3B=VZT=2Lla=={vA8$Y0ee1bym}_+sj5uHA4mL
z_rCS{klYN~1FyIiV08={qZQkCmmzOC6
zqkUO&fUq)F|KKrA(Gn8G0|O{Ecu3y)F*u;$s!`@(1=otD>^bt4#lI`V4<9g(@02dn
z&CM!dG^~F#NPv&MH>qTISS-FqmRPOVIr?hH%fcl9=unMJ?8r}Od8iNP)ph)!4(e9`
zj~4%W3;Rf&tXVtwnOW974jPdR-lnDRt6!wJYirXNQO-X0Ibb2LE{k<5{0m@)2OYnH
ztPY>`w+-A{8w%}ap4ruVQ*fFZL=ghZcYVzwCe$UsIi@`YmDQ*rUkXp1PX}rau@uPg
zuFTSac?fNl#dP=doq4^Kln}zcE;)_x<(^);$Ff)pKojP{O06?>(sgf(W^B}{pMWws
z_;kGG=Jafo0G~*2{CjneujC^*BA~LdYOPo9v-`x?Xc&@e1NmyWNk@Rs&2F)$JGFG~
z!F{q^2n;Ih)5;!+HV$H1Z`QcJzMn0pB=cY9J+E2`@X#+n*8G=r6^=uGSx7Dsc#xlM
ze_pSmyyy%u8eN{!AfvTf5;L$l@%VN_dJh&j_2&9sfvpVs=_jMMg1!@o%g@jR|1x
zrA11;L6JijjHs}}(nq~tEOP}5(6TUP$NJ-+4iaq?s5B+R%17Z9@?eJn9PAyB`
zdpGgU5~ZX48awT?M3PQHZNEh&Ku7?%Y3La)_v3gHp7#Phmj~c9+mGOR>=(U@jy5mS
z0gReFqr`ZXT9qt+5UrrllKhQ`04#o9PAA=bUNY$Hq$~nYT6!`(+Y2nqpW0;Hz(U9T
zE^Nn62XLkQ>h-=R=-T=IagfYzV4?%rMjq6+COY-X25)!e$vH6I+eKv5t
zo~|QF0k%5FAGMfg^UL&>6mzz=1rzZaj~=jR+`9|lp#0oe)(#i?
zC8Jhn+?FiM5L?@bGwrw_uhw&CZQHN5Bk(!G>&|(tgX#!?=<+%P{U0+&0caJze5mjm
zZ~49ZIA+VgbH3i!03Uy^!*3|{J?pzlR^RD@W%y?$W}?v5REcZy)$UQoRH(anKGJj>-WI2j~Gzb
zNKvWn+|0rn?H79ZdVo<9awVJOMW3%n+2r?Rik6Zv-6*ko}HPKH*W}U|z`wq--l_jY$kL^lY
z7Z@o21P1bVao|2ECobE$$!`kuOdz}%AEF<@1l2e)0Lc%B-9asBnFMeZ;1zw#mamIQ
z2ZYitfm>zcl(=}yt}jy-<$sj3#HsCaNA2VRe|QJ}d!O9IZtex-cR2nKGyxJ@j9hAq
zZQ$xqX>)!VHi3(X?g-{0O#`K}RI7hqIi_cIr&FFGT)Rf0e%M3S?#~zG{u0!e7tM(m
z9goH6QYH_#lds=M1suu%*ulaq4;alH)&=UHXzx|}b#NcNiRk{Ke-ebWd9&al#LO-Q
z8;k+^cVn?+Q1I{(jYaQ!G8WWZ*AoJgjlTVY41X-=wgna%nt^ti_F@1jr(>Oug!BJk;(48nIwavlncWmDcH7(hito?{aXAn6g3y|yjL
zyrJl~9oD8$RjqK4YNJef9frU;6SR2^u&2fWPf4RN)wt}^hSfIu`|ZrAbX|(n*sc$m~`iwezydfw|kJy7&LH&
zNYsN+mqZ{tEo%yz^m&U2TBNB(Ln28svflp`ut7mS!7UD7g{Jm!0M3|SMAu9trWf(V
zf&vf=2vmW^M6(I&Pqfz)I=*NWe6w8f{8d3gmY*+T#P^D~DttcW&3v
z0jvBFQSK`Ik3jYo!}~DKw`S;2SlRky{H59pcU|r6%1hvuV-t+m?dIv}j}`HYWcA5S
z)w0;lieK`WSA_!ue)ZP}OcXqu@W8h#LDIm(CD3asE^?F_*TcqPYV_@6R4QH+t(vK$$%I{+
z)ye9JQcY_HPkZ%A{;^-^0E5vvjbJeLrDu+zh9|ii&Qw$Ly=#>>n
z1HMk^e5s#5PtbK1Jkxz|pi)W`Ag-Y~PFE5;`A#wVRLPnRn#Y=EU7y-mI7VB)i0zIQ
zYEp#rFcUGLf`Pe}ztLZ`0)-0E{mw_Vwt6K>Gi2u*CjHyupQ%r74w=VNOKQqQ_hikq
zbYAPoz#shU69htx?Sx($!~YKM<8LJ!?69?{W6#HFsV|!e6Q~j2vx@KFPd=)W_qWJo0mRoMj>Q{3+dn6D9a0MyQZ+U2Yd<>BxC#L|x
zE)%1(lB_h#%~bM?k<{g6g#VV$WFXfv~uC(uAcbny#G9SYPAH(#YK+=~~EXaP?0pElMV~VBDL)`Jhgd!O@
zbEFsFn
zA&?z8pK3yjI9kpslt9d3wm0LiJysk?p8~+-I!QSj6>}~=exW?0pEyxB#hOf*qP7(ZUFiQvu)d!uE?tHf
zCkb12Y%go9t4ZUEmd!|YGO2WdzVf@1AEk1|2kN(=
zo6sOh#Hh0=_do7M26l!wdmb+H+)f`O^DL}
zwJbx~t8QWm=q`sMpEXlKRR07)AQW~XfPlm7&PaX_zY_K8UDsslyXF-V32A%qo!ci$
zs?jh>ZRpvXhP<0$*Ge^QZKF^7QESF2rlOLcn$g@t6Vu=+qM0cs#g@sqenC`tRIg%?Hj1H2%nx)C6z-G
zqh*})S5%-G8_VI*juG$3digbuZN#wt4e&wgAYAQaRJg{xX}-bX_=OdA^PjzMje%2`
zOP5sNpnQ{eJHikZzLx3gJQm+5gw9e_P`^8|T3D8X5)ey)$<6?##je5n|Jl;NPa^Z@k`2T3v-?BODNQYXH?}mo+#JP=-tCs>O4|*iNlXm_KNjt~b#Bd8
zBlepq6l3dF-^NQ{N*x+?1z^PD2G3qI5%_wf{}%`0B<)idtlmWBE0j&50A
z($d2O^c<$6`}wU^_6dg$0sC}@4@%be*lPwZCULY}uM9$i*kK$`_?|9&V%Z6+aA6Xq
z*Y(zmF1?HyN9^v9L4`DIEi!oywQvH9OI2K3x59Wjr>#HLb+Jr~97O`}pKGZVpXT}#
z@e^14u~}Rj&Unu*!W}`|pAs3H!oE{U4+@(fp02tHq`pi21iriwW~?tBq7LRiz@}d-
zvwX;glU+k+7rHy=V}p<&-Teed3;!1|U}#i}JxIi^ha`L~OZ*2NV=>)YY2z^6wrE4O
z%8RZT=IX*xh%pz!SdBaGPCDr!h4W}Ica&h@^(w>5`4iT6>IlN*vm2o
zabF0e`#r^s1n>Wg>0(_Bop#kJ@Ye8(sOgL`3QOVc9*D$S3{-$Fx`AT;qUo*44^
zgZ-dM?JLl^oc>S$t7Y8vg2Mj2rd*V=
zu&C#W&G*-X=2M#qSKi;+Y=~qGcKDfnVie{oDSgO3xR7%zPiMPKrL4qao5WLYeF$bS
zC@-hGWmsz~v|kA|6F}f1o~;IMvH&gn;|pSJL{?803<&;zM@=>p9LHF2ypoZB8kTco
zHTZ1_mBSPb4-r_#w_84B^uGn7Ub8P^L|kwwNBQD|#uWWQ}Oa%!QQtMXSqcOX`9kUnU4*YdK@!-%V+
zv>*&tfojHfBrz%CeT9$-1FJ`MPYS8jFB*;0Fh*<(UVe$hxQr|}gtKBg)h1I!f4P4PSCMWx`Wvs=oFr^x5wQt(V9wzO{fQg?6Kimn
z!PxTy)(iYU1uBqG5wXDwsBY(gn^Y?7ezp{_ov6RIg9t%$R1W1;;%#Jma;OVHgr1@5@9U%i-)
z5<~U|y_1^b_XES~J}mb~R}R1Ga~ItfS($|sgy}D6
zwDcG+cnkddQ1poRl=#glhTrIuG-A{{YpE_o3n=W|&e1TVu;qt>7kX!G-5tuzuzHXv
z$1QHyV14YGN&!MP(2EgvpuPDon}CB@fWrv{@WPMQnznLy22@m*pDP;p&3`U_e|P+X
zZOu8-)2vFm{!V2bNLR;U6o{bLSmc3lcis-P5rt~jM2a9NY^-^oO|%0xitkW24YR9L
zzQT8xdzVMFXlLqcSg3n2zfgA)F~1wv=by=sx&A_`YJ4NiX?OMQJ>HX1wQ+Zf`4(Zp
zEi>_H=U#U9EW0paLTL5f>O<-wi?+Xm&l|pwoUkwgS1zrx2hBg$shMyEwPoC2D1#T
zSB2x~ATEskxdxtMQx`WLaydHDBr3e)_8T@;^9MPFEapy>`A%!egs&Em=I6^P0Xt3j
zGn;lPo>2S7r+J5HX{zTACNE8kZ%azbng?xkpy@dh(+Yr6oSP5D`7uj0)gg@dO`lrDBQWX
zT5vO{rD-PQRF=C5X1D|{Xyo4M>91_i74<21Nc%prfNpQ{jKuJWq7=fr(rEnR&ozt8
zyJ00@>72I*%_a@R#~psxRm>Cy{5m#B1ey!G)!0lo_!O3}tPgxfp+Ol4sp)Axs7{Uk
zt}1U7kZ%GC3a!wxJNIT2$Zr&)QmCK4TcS9)x%6z`puX7~pRJWbg-~UfsqQ!CV!8l_
z1Gyd9>Zv4(Y1D3~g-u{T;jLjpqj4DH>%PX`W{C{K>qn)=!A{k2k&88^j4|
zmMPsnOUjqAX&TOb_#hvJ*S1`3dF+JS?L6BX>3|)f&jZ5Qe`h}t+*PhCM$t@!O&zdK8!V0Y;vrt>9g`;S%@(CJ1h`j
z$9~dxhS1^xbi{6b?vpK2rGR6>Qo0_MAP|nMak|KXl<|ted_ll2*6OSft~oa2E4%sV
zY3o=hs=h_O^O-hxrDK*ru9F)P$(Vft*+z&@M%41J@tszcRc`lB){r|hKHkHRye7@?
z?Bb$OK;!OTfHdV))Q|i5QK@=h(D7V0S`1G+g}z!**BpRN0`6r?NOd61?=
zkLJ{kwveHf;yM%^%_!OW!%yB;An)gE)!aRp$_ygTxAND2olbY9QnFrcH5gND5^R9`
zU)nK3Ah)vYwJzQw5ZTqD{gZc85rDk&0l7P0i%o=o!5EJ3JfXB^bZKfS2zvXl1FEZb>B6VS!3H+&=lyd`RZ-`pGs)1ktANh(9@l(Fh@386{v@y
z)Nrz}d#smb#{9A+5I?lnewvS%hMS>38|_twguX3T#N>$=8ax*9iIH&-Dyea)+fqxn
z*Ua6BmXtVn%N8Taj5lf@2NVj0$KgVX5~RhLXx}op5^nm1aifzHWX{86K&TO(RUS2a
zYPlJju-}fQl{Y@H{*#$%)3w3WkJ{
zOFmOzcwgEmo?BJDf-1)T7(f=_Gh$Ng2lXJEx6S=Q1)F
zr4+fp|I!TL?aA1*G`i7}VG&i9R~i!$AFv}@A!(zp9;ddHY{RTnOOPW@JSt=fQ}s;X
zY}mF=%#aZ8;8j{cB`q`|o3b=(jiVhn-k~U_2W(!B+26nGOWMaXDstK#7
zR!J3?9)4Z~OAlw^P`hD}JA|fxIgS5o%d(&3yaR5sW*JaIg6a_3TAv<&tfdrKaF)SB
z86jk}+zW}wooB5X#H_a5-;wZ!BJa8Bakf{ELD&JDcBc@k6bvQ?!Cvx_7zvbUY%zvA
z-m-Iaj14=SiT=WG87IFN8og1Vbr)>+^$)czK+(U`N}YyNGTynoE>C3z+MMi;C;Htl
z_BS;yl8r>n8Gr_Kb;Y+3c&6^y&J2kuj?hZ)G6f{2K^)cSX_VJ4=xCHLCLSg@OvYu>
zbG>Df6#4d`Xc>2Po>!Tx5yCsENQcuO5&!qxF*1WtkuE&gQcN3xxq?7Qu?0RsH7-3q
z#|LoanV&wvs_R-T$u^ar)0xjbAGAuM()xDy3ZNbyw0w1zB>V0~kKI?+ykuHp_cW_<
zM^l`m4=YGMMcaigjH*ki>P`6z7v!Ypt-hm^9npY|I&XBXZ5JwEixlkLv#W@{T0
zks5=Z()1#&lXfB*lK8yk6_>2OpLc
z{fYKk#G7mKHSD|h
zba#HEk6B0pY`}x<5JL`Ego>b`0MJ?YxBfWx-#d4)qbr#Y6!M>u;aD0%?{
zlhn1DgrT)nBV?p*1=@FQH;!>lL&&`Fh~KykU&C1tFoo2m-pF1S~>oNY@%3>E|})
z`^A8p!UR9%hd9b0TJ9m|`IU)P%iVKN=*2n1rl;XBze+yqE-Z}olrG7b-6{CS-E(~Z
z^@#6KwVu+wNHf;unsIGzz3%o}jUcN)IrOYwf@4|DPa(B%q)VOC~dev(=)5ta*zL3-3}^YoERMWXue{hmRP0#ctrYMz%#a7MD3T@%JUD#DGE
zlZ*0)UjWc@Tmf95P1gLVTbYUhauTP4A6cF05G&Bj6L`&<7}0}EgS5B6V>dq1ced*w
z{KUf*9}33=1-11Db!W<+tOfETv1U8kCwx9TD|dgShUuJsm#gNgcyoSpZMga%<@aiDTu=``Y6#-|Ie}r41Ik8a_+&?HNU>NSm
z+tPh(?G5d0Sn`uVMtV4&lfyJQhrBl&Jap#wZH$LVRm{tnx{tal?gI+KxC%MMVSWZx
z=AaQ9Krk*~Dnz`AEGAjsj;_34@O}JR&^ks}2q)EvKu(I&?D-o|mYvFAd~6T(o-fTvZkucj=0-gDg^
ze}U)FtqGae0VOTHzrE%*@~6Qdz`p^0_-7*f2he^Eq#bFUY2y6Z6QIMZP~F9skI2P>gcmmb%5OEDx$EDC3zleqoc4J>wC%aB)Ipa>S>
z&5qQfq%_9s-X324CLd1j{J2rnDw_e0YrnMdjA=@;gB(?LO`ARGnUE5if5
zR#SFi*Tq60JIt@y`PrM6G|1)kc9g1bPM2H&C1n56W|z`L-W-{);u^=u!$)TsV;9cj
zBSv+st7~plR(oQX(Zo@{*l=FLXz4vJTA+LBpdmL}K)oM8;qyCNTDee0hH^b?dhTk_
zr)7fyMZuKR$t*|1RT;kT#GsO>vb;`5wJ)9LnrSdt@lz~ofl8mw@()I@W@K8Qs}6n)
zsXEd&3oNjnAG=x~Si3f$@uKR&f{qdE?km0lIaLa|sm2(hHThye&u_p;LGyHkV9W|M
zd=ddLC5bGQ)S238MIE}=0)cstQ8&1SPb)ZypccjRnKdFIoqdPJ7yuBsaaX+(h;Dh
z5XvoP>+gkHajb(od_gzT^*O)Rn)1>%a
zOm%cb{(xurRoj9UMedK|XhGOLWI#B8WZ?%;26*K)X2YSTJy^+qB`5eDt4yrI$>+Bx
zb#zgh^5VnJZHUo>LOHB{_cMBriAD+|VKy2TGiZB0LtVz^>rl|4}a_BsCX6N&w?jEe8o52G!?6c
z7HDulO)Br6UcR$Mj15T#291vV0oMaWV<_AlOdPA@G=RA_56s)l;~Q4OiAo0&p6+Ig
z3Q-hdkwElD)0*W)*C@Wl1+EG;fxev8#>x=U+ZT(zreE+PlCsA|yW1mHS9RCSsA39a@8`
zYe#l5z;xVvAOEd7Td>~)3yi3D!Mj(^h7B;v_(r3YB9SU;h}g+5D|>qmV{7kCJ0A+E
zc;}HNuwZnt=#R2qJ~RTL3yt339!Bf_t?hJoJ3*Rw{S0g3+``}_WnUqR6F>+LTv=ao
z84MR3Tjl(yCWu3tdADvsPqT|dA5lZ}lzVr_WIYsgU+g0HTZ#e-ghWfOzM#~_?_N9;
zm^5=ZGyG#ju9KJ=c%>xNBw@Ap2}FQ4NsyS}POW*g$L+sodw|_Vq0wTxug_Om!LZmH
zhgtgx;qP;t1|Dd^St(7~XnRY6Jd;+!t3$@{Q?IF*7$jnpKEqh=`WRM(MB)5fJ4jQg
z{e?1rs7^_=QW$q+bDS5rnp7!d3ctAL8E`;Kc1ip#ruG-&K1~Gvr=!S|YP1^t#x#JS
zb$ipwEK9z8b;GadeqQt@tGRN*-=y|~avJs=s$&>UEY-ee50-s@sLoNV->-l!8x5vl
zF~v_Xl8miB(fOIbX0#fe6mZosprB|0Y;}XjR9PhM3}|c<5Lf>jwaR`3C%dCr<0X`6
zK=DozOAV7mF96jEkuOlDsD?zb!yNjqS40K`+_g=PAz-Ofo!?vZUyt+^SaKi30$-aS
zIh;3VqKK6^5;tZ3EO55we_X;W<78=haEuq9oOQp&NH%C6!5>e@R_6A!;uaww=&VRO
zAZD|bLaJHJEgnjaCwh3fFnkzQ
zY`n(%3l_mvc*w|K);^H7|2CieEMXhH&dvK%(DfpL49d{4&7Y&mlHbpTFmpg
zNeNOO&u8EX?L8SXbi`w>QZsacxX^1q@QccY7{Uh+}Eham2YLAzvQTOJWuv=>6rNR7R+GQZRN=wYVSN0k1S!gs
z`hdHoi*6WrHKxZ8_$s12buKk*OdxJ^AI)IEKncUhx)%LShq?>HB}zFt1&KbbRKE*TNwt<|($>6Q1`J4A{B$;g
z!%QnG=+SpxpbG`6p&|EnGlaBmoLtk@?DQp=?R=f1en0H34#zDIoM2yqWvU?;Oi8%#
z=GTJ>0HW~UhQUB5`3dto=`1Ntv3skSbeV9#+7|#+ONL)Xl+6(z(*JUAC}-ymbsO7p
zI5$ViOERVPJAi%;dUC$FXDrm;(ovlSM3VI|x}?_OqhvtmM{{R2nUlf4J^^#wkU*fPz(TwNV4vci7ICF1K1CXcSX9}GSFoPE}sYp%KGraJc&bJ|V{|9D7=n5eU7SxqN5elr8ppf5~H
zNu@hjGjD2VJ`0*6#*bPM@qCwM`1UP07a2c9WnwOe4yoL2x>T?!!;bs3cp}uirM;Pt
zJHNZ5M#KJyGgb5CpFK_?PKqvlW!EnK+z>Q@?V2=VGFqq~0Z+Fa0@
zn+d!`OkJJ~@tW%zPU%?H_n2E}sSR*Ny#6}CFcW8swa35k`i^YgRuaBVc#4S4Cy@X_
z)8jaaCtWk85h;K`(YAbLU0QK$)}V{_Dly43C3`tJ@52fRoj04T+b2<&)v=r@R)#64
zGGd&_gjp+M_bVwLN>HA+*WWVJ&Yp9;%wj@?n1QG=#74^-MOfVJ;0;;Um_wpzk*tCL
z2N()M5YqGw*0D(#R_G}~Zz&voD};kZukj;k3nPres=fP>wTFF=6hUqF{?9$(+8lO+
z)wto=jV9ycnIzfrVV3AsuJUF_hqS_xFT$mXEAbsnNRUwsG0DLOKr2hqhOT)N;832Sv7`-Se_e
z|Evo*c$0>MGH*I#TZ1O7G9{!6Z?hAu4Fxp0f3Gc)rGA2j;FKashCnc$F6bCF
zo>GSaInNXQqs`_br@eZ!V7LPI=oXNZz`=(W15~f>m@PjEtj6Nt$4tvzk)@?P;J#cv
zPu`i}xT;E6m9Mr$x-N_%daOn3>s1mU72!e9dZzshN@>Ko7AKqU?kbx9n
zb>E{{3_Z^Vr&V$QeXz99?-1*Ou%N-(_)>d`jHfX;i}Z(qQrx!~Nr_ekdGqUYy7kpl
z#SDv2{d``pA-FgD=187wpT!xVvq&|>K!_B<&e`jr&DS4t{ucTypz(CS@ce3ZGJ^1j
z_;X7eHotq57NDUUl$#FxzA4cgT{ob8~o{IYajND+gIg;64zw$8i`S`KEK
zar3~7e1c9JY+;NY=H-_yyqBH2`~&5nD!N++SNRtE?wQ)t&Tmp^=l(5~0qf(ZC}<5Z
z-aMd(;^L}IBmd(GR!~*ioYi?95Guv8defr*w9@Kb5;)s5bQ*|t4_=K)iZKrJ2%lADczkvn!=qjVp7_1op3o6K50B{g>_t%Y$FWM7tvv^Yg{tFQ64c(SEQsH-2+z$T!KXiY{y$rvLUuusuVmRBt;PFg2NF*WH$r;6
zQ4G-GK!kxF4;|`DE?4y4-WCaF?-PPit#P6taRY`qqZtJa3}6i{jzkgb{DfkFK9rZ0
zT{EMz&CC#^1wpYV_(j8jbjx_BNz~iBPsIHIPmquiDSN2d+TBe(+K%H>Acb&}B;9^B
zu!-V-9-xYZ1&coF;eox$YCG~ljr`1g@8&SFSuGn!SJ4axqKARDkMl6ER^BMT-A=^D
z^|NWhvH+aV9}D4QgQP)z235uRazyqZ0PXmRmJ;~!^n;fb)GTFNax{-L*#pn`u_t?iZIN4
z!I7^QoCiqh#Ip!NToSxYsOiO1tF2E!`1FwxTYX}&mwa89G+koQ?U$?VTFE@l7@qXw
z8)R1r;bT1HPAvEmYEqpkl)9l!6hWXwKJNJ^^Zn<7F99m&kjcKQ;0#5)
zQnnvnE|mGG+?iaiQA1waaxmX8@JXh8wz906_4;Jxmrj~iHDZKWPjrdrNX|;#(%8gE
zi=81y`O)#dK$=~+aXb=;6U9)>i*ffX5ga57g^_@VS8?#MsOV#+sF6Iy
z;Qk_A1#(U-X1YFS8a>7%GegRvw|Y;5too>;!!LB!X?3>(F&`uO2K+L;eD2BetKrhO
z6q4f331JiOYz97xUFi~_F@0#6Gcxvl4}C=!1M58?olmUBlCMLYDML7*kk_p3vOn6t
zR>t=flUhugERT+@mR0y`tV7?Pi47~3F`vOgS9$L#{jJ3|s3(doxbU%$#l6uM%3j{z
zIp+03<;y1s%fbo%ZwmIKKk-?829&_x{6Dy$kntkiK|km`=1eut6fvM?#Zgb?WJEa&@I6CeA?zV4}!&f+u^q^+$l^?ya-(cj!XB68x+O
z4h6QNE;T$nzmCKm&X$qWbw3?gIV-jo7qQF)k$NBbR
z-9X&y`i+CjLF#;xw8oT4PwmRp_wRFbxB|#f8ZuI)De2=Yt6}0m4C)QGhc^`)!AJ%F
z;@v<=ix1z>c%hW3FrtI-h9kO2TNOi=5p*+Lr@)UOc(A2OImfLBuvnMT?k2RG+;n4G
z3w+^KR8&vZHlLCC2(zASm)(3qFU>Fy2;uX3=zSbqKyCTTM&%3Z2H_L0rpoEVjXfj;;0x)f#KDz|*w@tV1jx@3M0`N?85f9mgj{G6
zp)%l26{^~kvwz7p3RNjt>m+9zWWvaj>lHWDeHSZ1vr}*qYpUEfE0DX_uc{8ydWZh;
z?uotaRbIcl5n}=S=0Zhfn7d_8C*_SLelZ`It`u;0OGl-*O`i+I?$)k@|2=5)?`?-A
zxxn6IQPH9Gyb>R@oP_s0^}9dbbwP@yB)pUk@h&DS#QSmeZIk1u;S8zffCd{!)clE0lbSKK7OqfF5$q$P)dCblO}sdu?Rb1-fXb=8CYdG$V*+cksz^e4*E@4C$10bzx1Z<%xn&gG;8Y;S8F(^1KGk~ql~MSxz94>h>sy`p
z4uR4o&qv%}XvXTz8H0QJV(`JU49t}7`^j@>+P+Wn9`*WBS-tiOeDcQZji|Bg=lLg)
zW#n1Zu%BAjOM54CjbaD`T7%d?wJLEQJR%U{9a!0bj+V9UpG6blox>=(j1u`E(26hp
zf!t@Hq5oINFc4T6ixFix@O{hKQyt#Vd-0)W9=|jfEjR;FsIA4*U%2i4&dYGH8X(?2x+|oBcKGx(e
zvJM%pHDZ1Wg&EY6uDWJv`8hrP6#y+<~7g+faV$>UwB#4V6kAcF&klld^ZLXZ7)km
zKF|@?Ap9B9b!fbC*MM>~1A2kjZz}#x+MvBp7xcyZ9+f}E4DT7Tf2w~`N90@}6O(Xq
zLa){B&sPi}<%kN$?*0lOlA4y|xU{F!9|2|f{&R|_lqlLy5B;LV2OF4E`upuY43#q*
zu5N|NPw5G@3eC`|=;K*Ky<)OE0-y0Vic}bVgzY8zg`s$LSGJ_OKplIC#!BxkYMGlG
z+M}K!7xQ>N0SDG$jc?)T>kreUP)i~vPZ@9~`&EBq!>QPMW;K^z;6rgbW`D1NX0l-U
z%MIZY=3c|*s%6I*Gs10yJu1Tg!O;tRq`%Ccgwpi%_xH^cWAB}=)UK!Em&!jSni9ibClO-w3=*bTDtI0BwG5XJnk|U-Diw9
zxD9gi*Ut?=Kb-NWCx;!`va~Sit7r5S$$^<45<_<0J%^v1aDr?~jRDOOD0Zuhwg!*#
zQD%5eHVRX|e0S$qVh1+M>y)HoQ_|;>&t(=ywi@!F5);K{`Guv7|JASj)w)`N?(TAn
z*7{K~7fZE0cNlm~H8uT18t7Rt1^@GgnW{@@RI4+(lO#-EL}N(MYHb&qv*$SihHa{s
zR_L&CUb6WH7yB38Y&0TFau~TlmYI10pzKp`L{gHJ8f!d;hd<3Bt=PY7MvUx(_F(t!
z)ej%lKc`ISnCug2r;12S;`|(8=1V4s>MdHFm&-z@T8#0m|H~sJx)Urivgxy!!q+z$^Q^(Pilf&dBgZ9p;2
zm@8GY>4`}voAihV7mHq}6@Vn|KH`{YEt>Z4*3YH%SIOs8Y1OwwWRUhyJDkH9_&begM+gQn$?X}hWQiw4n;Ctvn`}fYQEsW
zb62~L$zOi66@7SISi~s{%(hqcOm5{Cq+E#YB+Z{!mWrO?|CF=h-8U7`6PhJY+wzoE
zuRpIZ?M;%|PpIA4_&x*pvg=)lIU=>rGv7P0?u5Kw{*M9n_yB4RJ_pFgr)Q!bKjYn@
z#n@t`A41#QvBiiP0Mq+XcH!V4Y1x|a;4IrsRFdGUNeaG##q!&ua4--0`2?0luTX`_
zSOC(oZ-qtfZdb3{PQLTqM%r|Ab835NXlvNP$8rnE;brM2*>(Drny>{);H}8Q6I1a%
zZo)7uBdaK~HTTwNNh7O-OOK9E_mn%1@DpvB^Vry49n1ia`YEb|NP2H{O@cn+u?QuW
zQq6RA209jx7RDC>tg!(%
zlOOk9qtMQ$hvRYXXSUFSUvH1J9WZfm5>kC<+IfoUCNeIFj_{Q^3l(wiJCl5TFi>#l
z<=#$x`UBC$r~ld3*MbYXX+|(lIl)lG<}}aL8u1q^D9AN(zjKW^e9Z)%QKWlz*?XrJ
z*SvVd7u;y;JdWpIdvD((JR=
zXtrbmW?4vOJ--jCI#IB}5a)~9#(xaZv3cyr57h+}$0
z(gSCl*Jq?DC!oHQ7HDbh|6Zs96J$J5ebCwg{eQ!wiEZ%C)z}5F-s!-j)>K{i^Tg`qy!4q
ze)@{I^vx2ZC$+o4LZ4Bhp<&58|{1YPrvNI!=@JU6GQb(a`P_|Mv#JOpnTSGIVI?1tPLxDF3*);zRk
zGJi!mgDsjv$v2R|QV28UXvvZvl)!z<*ApuAshBl#L`&=Nb*;`s{^{!hVEqF)-o0yQ
z4i+Y9P(l#|$HUDRM_mkHk2Cv2L#f7)(C8~WjvSqGjwP6J7
zd+_)AY9?rm@(U?T8DC9Npi}7w(`n=hjetq`A=2JTkmQE6tGb8TxhSzAs?XWEIYpn3
zOAHN^J8DTCy&{YXe;R&}wd$|*GyF11f)e)GE_%1w=ds6T+Kz)CbQwY&8q9=v1WxYk
zMa}e{Gft2ywq(8W@YD=M
z2}$n~P|;I^#9mtz4=8l_iyVGcl$kz^du9SAXQ;#eJCFY#9tbE`%z09upc^$1U%aPL
zX;J}YX4WrUuua9t{wjenF|3NU64|c#2CX#7_}Wl44VKw{o(6cx%FYHN1w}?Fc3vIs
zj+Z|QC>q_F8FqT};{k*az5yvnA&>pO_2mkA^)^+b?rwD=VuyEnl*$viM7vunGMQs6
z-{L}lBGqND9_vzdNSuUw%=9$(F?l_^o@|)v;XzUfnXLnLP8MsBuCf4BtKm!T49eoT
zL!xNQlqTOw&>sw)j7hem2JWHP3O63rV(1ahK)rm}(eQdIUtEOz+ZgiSKh>3RqrV_0gIJ>z)tRpu@s*_`cE3j(d7)JDxYbD>3O2qu`_dX}?5*
z4Gl;EU5KniP}atT#pggu{IKo59y|DF@elGeJonF5nsW+n3xCG1NK>Ep%^oke7cR#`
zVaerlw_e802{og>r4e(mzy*9{3Tf8A5kI)b;wPo0*PlO+$v>77b}~(mI<=6gbY*9?
zig?xmw)F^*JOOl^L+P~@4Qr(*u?7-;arAq`xl9S7j1u?b&
zHeiJIpF<5fx8LK@Zs4K~>mFXlxH1|Vo)>0T*7c-X_ms_Ze?opNS#RGkdQ(7
z$I1GhzS=nbV4SOCS|hc)MdS#}{lMihniIiuB^W|Zl9}7d`8+Rz>PnStKFD&@4={fJ
zJ){YiU8KN__hfS0d~on5`4`1!XnqjQ)!J>iux2K0!CjhqT{z!=O(BCIR$oF@#0$-K%(ZdKf!>y;TyTV1Ld&UjvpI_gfkQKX=SN
zDjQHyJ+3?Z5JF+V=cRowY`b^^Z{7zjQ`%&IR9ux*vb(zE*caK^H$9~$yE*XzYu@S9
zL{H!Reem_w;(Lp?vn02;(rGU(|W3Ff~P
zKmRJ^`1s_Z#6!igKSBg+E`@PkK+Y-m=+@}Y37xs_ya
zH-&j2D8-