From 1f1d032430a29800709929890cebf445615ab0c0 Mon Sep 17 00:00:00 2001 From: lc-arjun Date: Tue, 8 Jul 2025 06:59:02 -0700 Subject: [PATCH 1/4] docs: add tunnel flag to node server docs (#5398) --- docs/docs/troubleshooting/studio.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/docs/troubleshooting/studio.md b/docs/docs/troubleshooting/studio.md index b06f325de..3fcbdea7f 100644 --- a/docs/docs/troubleshooting/studio.md +++ b/docs/docs/troubleshooting/studio.md @@ -17,7 +17,7 @@ Safari blocks plain-HTTP traffic on localhost. When running Studio with `langgra ```shell # Requires @langchain/langgraph-cli>=0.0.26 - npx @langchain/langgraph-cli dev + npx @langchain/langgraph-cli dev --tunnel ``` The command outputs a URL in this format: @@ -55,7 +55,7 @@ Disable Brave Shields for LangSmith using the Brave icon in the URL bar. ```shell # Requires @langchain/langgraph-cli>=0.0.26 - npx @langchain/langgraph-cli dev + npx @langchain/langgraph-cli dev --tunnel ``` The command outputs a URL in this format: From f21fc056bff70d7e5c8dd7553577a1a26238713a Mon Sep 17 00:00:00 2001 From: Lauren Hirata Singh Date: Tue, 8 Jul 2025 10:36:47 -0400 Subject: [PATCH 2/4] docs: Convert example notebooks (#5381) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit * Agentic RAG * Fix formatting * Agent supervisor * fix links * SQL agent * Graph Runs in LS * fix format * Autogen + LG tutorial * fixes * update sql * remove old notebooks * docs: Add section about data region for LGP data plane (#5378) Add section about data region. * fix: remove empty notebook (#5379) * Fix docstring for _unset_config_context function (#5374) Signed-off-by: jitokim * Fix typo in StreamMode debug description: checlkpoints → checkpoints (#5371) Signed-off-by: jitokim * fix: remove unused import in generate_llms_text.py (#5380) * dcos: Fix deprecation of TavilySearch (#5375) Fix deprecation: The class `TavilySearchResults` was deprecated in LangChain 0.3.25 and will be removed in 1.0 * Fix typo: funtion → function (#5370) fix typos Signed-off-by: jitokim * docs: feedback edits (#5387) * docs: update lgp deployment metric list (#5388) * chore(deps): bump peter-evans/create-pull-request from 6 to 7 (#5365) Bumps [peter-evans/create-pull-request](https://github.com/peter-evans/create-pull-request) from 6 to 7. - [Release notes](https://github.com/peter-evans/create-pull-request/releases) - [Commits](https://github.com/peter-evans/create-pull-request/compare/v6...v7) --- updated-dependencies: - dependency-name: peter-evans/create-pull-request dependency-version: '7' dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * docs: correct link in docs/docs/how-tos/graph-api.md (#5377) Update graph-api.md * docs: Update quick_start.md Rest API Guide (#5368) Update quick_start.md Rest API Guide The curl command in the quick start needs some minor changes to work out of the box. I hope by adding these changes then new users can get started more quickly * Remove duplicate CONFIG_KEY_CHECKPOINT_MAP from RESERVED set (#5372) Signed-off-by: jitokim * docs: fix typo in persistence (#5329) * docs: fix typo in application_structure * docs: fix typo in persistence * chore[deps]: upgrade dependencies with `uv lock --upgrade` (#5358) * chore: upgrade dependencies with `uv lock --upgrade` * linting * upgrade PR title --------- Co-authored-by: sydney-runkle <54324534+sydney-runkle@users.noreply.github.com> Co-authored-by: Sydney Runkle * Updated examples for SummarizationNode to account for serde with persistence layers (#5257) * docs: move script into scripts (#5384) * docs: update sql tutorial (#5389) --------- Signed-off-by: jitokim Signed-off-by: dependabot[bot] Co-authored-by: Andrew Nguonly Co-authored-by: Michael Li Co-authored-by: jito Co-authored-by: Serhii Polishchuk Co-authored-by: hari-dhanushkodi Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: Fadel Akram Co-authored-by: David <31293924+dreadn0ught@users.noreply.github.com> Co-authored-by: Youssef Ahmed Mohamed Abdelrahman <109446360+unauthorised-401@users.noreply.github.com> Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: sydney-runkle <54324534+sydney-runkle@users.noreply.github.com> Co-authored-by: Sydney Runkle Co-authored-by: Nick Riley Co-authored-by: Eugene Yurtsev Co-authored-by: ccurme --- CONTRIBUTING.md | 2 +- docs/docs/concepts/multi_agent.md | 4 +- .../410e0089-2ab8-46bb-a61a-827187fd46b3.png | Bin 0 -> 75818 bytes docs/docs/how-tos/assets/autogen-output.png | Bin 0 -> 5919 bytes .../d38d1f2b-0f4c-4707-b531-a3c749de987f.png | Bin 0 -> 53669 bytes docs/docs/how-tos/autogen-integration.ipynb | 406 ----- docs/docs/how-tos/autogen-integration.md | 321 ++++ .../how-tos/autogen-langgraph-platform.ipynb | 171 -- docs/docs/how-tos/multi_agent.ipynb | 657 -------- docs/docs/how-tos/multi_agent.md | 2 +- docs/docs/how-tos/run-id-langsmith.ipynb | 274 ---- docs/docs/how-tos/run-id-langsmith.md | 155 ++ .../multi_agent/agent_supervisor.ipynb | 1403 ----------------- .../tutorials/multi_agent/agent_supervisor.md | 812 ++++++++++ .../tutorials/multi_agent/assets/diagram.png | Bin 0 -> 25505 bytes .../multi_agent/assets/multi-output.png | Bin 0 -> 14338 bytes .../tutorials/multi_agent/assets/output.png | Bin 0 -> 14338 bytes docs/docs/tutorials/overview.md | 8 +- .../rag/assets/agentic-rag-output.png | Bin 0 -> 81761 bytes .../screenshot_2024_02_14_3_43_58_pm.png | Bin 0 -> 104631 bytes .../tutorials/rag/langgraph_agentic_rag.ipynb | 928 ----------- .../tutorials/rag/langgraph_agentic_rag.md | 522 ++++++ docs/docs/tutorials/sql/output.png | Bin 0 -> 20460 bytes docs/docs/tutorials/sql/sql-agent.md | 549 +++++++ docs/mkdocs.yml | 11 +- 25 files changed, 2372 insertions(+), 3853 deletions(-) create mode 100644 docs/docs/how-tos/assets/410e0089-2ab8-46bb-a61a-827187fd46b3.png create mode 100644 docs/docs/how-tos/assets/autogen-output.png create mode 100644 docs/docs/how-tos/assets/d38d1f2b-0f4c-4707-b531-a3c749de987f.png delete mode 100644 docs/docs/how-tos/autogen-integration.ipynb create mode 100644 docs/docs/how-tos/autogen-integration.md delete mode 100644 docs/docs/how-tos/autogen-langgraph-platform.ipynb delete mode 100644 docs/docs/how-tos/multi_agent.ipynb delete mode 100644 docs/docs/how-tos/run-id-langsmith.ipynb create mode 100644 docs/docs/how-tos/run-id-langsmith.md delete mode 100644 docs/docs/tutorials/multi_agent/agent_supervisor.ipynb create mode 100644 docs/docs/tutorials/multi_agent/agent_supervisor.md create mode 100644 docs/docs/tutorials/multi_agent/assets/diagram.png create mode 100644 docs/docs/tutorials/multi_agent/assets/multi-output.png create mode 100644 docs/docs/tutorials/multi_agent/assets/output.png create mode 100644 docs/docs/tutorials/rag/assets/agentic-rag-output.png create mode 100644 docs/docs/tutorials/rag/assets/screenshot_2024_02_14_3_43_58_pm.png delete mode 100644 docs/docs/tutorials/rag/langgraph_agentic_rag.ipynb create mode 100644 docs/docs/tutorials/rag/langgraph_agentic_rag.md create mode 100644 docs/docs/tutorials/sql/output.png create mode 100644 docs/docs/tutorials/sql/sql-agent.md diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index a4e9a3eca..48fd5593f 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -60,7 +60,7 @@ In LangGraph, these are often higher level guides that show off end-to-end use c Some examples include: - [Build a Customer Support Bot](https://langchain-ai.github.io/langgraph/tutorials/customer-support/customer-support/) -- [Build a SQL Agent](https://langchain-ai.github.io/langgraph/tutorials/sql-agent/) +- [Build a SQL Agent](https://langchain-ai.github.io/langgraph/tutorials/sql/sql-agent/) Here are some high-level tips on writing a good tutorial: diff --git a/docs/docs/concepts/multi_agent.md b/docs/docs/concepts/multi_agent.md index 81967f610..5d3e30a29 100644 --- a/docs/docs/concepts/multi_agent.md +++ b/docs/docs/concepts/multi_agent.md @@ -26,7 +26,7 @@ The primary benefits of using multi-agent systems are: There are several ways to connect agents in a multi-agent system: - **Network**: each agent can communicate with [every other agent](https://langchain-ai.github.io/langgraph/tutorials/multi_agent/multi-agent-collaboration/). Any agent can decide which other agent to call next. -- **Supervisor**: each agent communicates with a single [supervisor](https://langchain-ai.github.io/langgraph/tutorials/multi_agent/agent_supervisor/) agent. Supervisor agent makes decisions on which agent should be called next. +- **Supervisor**: each agent communicates with a single [supervisor](../tutorials/multi_agent/agent_supervisor.md) agent. Supervisor agent makes decisions on which agent should be called next. - **Supervisor (tool-calling)**: this is a special case of supervisor architecture. Individual agents can be represented as tools. In this case, a supervisor agent uses a tool-calling LLM to decide which of the agent tools to call, as well as the arguments to pass to those agents. - **Hierarchical**: you can define a multi-agent system with [a supervisor of supervisors](https://langchain-ai.github.io/langgraph/tutorials/multi_agent/hierarchical_agent_teams/). This is a generalization of the supervisor architecture and allows for more complex control flows. - **Custom multi-agent workflow**: each agent communicates with only a subset of agents. Parts of the flow are deterministic, and only some agents can decide which other agents to call next. @@ -211,7 +211,7 @@ builder.add_edge(START, "supervisor") supervisor = builder.compile() ``` -Check out this [tutorial](https://langchain-ai.github.io/langgraph/tutorials/multi_agent/agent_supervisor/) for an example of supervisor multi-agent architecture. +Check out this [tutorial](../tutorials/multi_agent/agent_supervisor.md) for an example of supervisor multi-agent architecture. ### Supervisor (tool-calling) diff --git a/docs/docs/how-tos/assets/410e0089-2ab8-46bb-a61a-827187fd46b3.png b/docs/docs/how-tos/assets/410e0089-2ab8-46bb-a61a-827187fd46b3.png new file mode 100644 index 0000000000000000000000000000000000000000..9bd66e5654ae1f2d93b2329d3a37bcf853a93177 GIT binary patch literal 75818 zcmd?RQ*>nC_6FKXCmnPrskmd?wr$%^$F^;=V^qgh$F^6ExV)?w94sd6r%#{YB*cXkKYjYF|LGI>7&Q1l-)M-=fc|}U6qHbg zhK63*B2o8_lJL9iGmR3PkR2(9sl|=F47M=jFI5~p85A} z4u}9S1=65@r2I!DKP@;R{z%YY-}nKBMED_u(*ggz=O4Ar&|!Y$$bx?xf{zY{2rv=+ z%QSxIJwkl&d}-+a8Vge6ulRa|B|c#=P%t3cPoqbV^zVEvRV)IKCqcAP;EO7kVC$9;rUle8 zGAhk7DJi+r(57VDJgf1Fg!_qjhGNpOle1HEvxB47_9qbBFI^`QiM7WNaJIN9JE6=o zjx$nArbUgN#V%qczy|wWLPMDO3JNKlYus%zCnbd?9qqePde_MNYHDJ_K|hJislrY9;$<1T6kX9G|T zj~Rp0!!af(rJ)KK7!WX!2ZZ;kXPq=e8*?!QHbk7SLAc1XbTI2&Mv1I zrYIJIMQ6g0nI?(r&QLKm6(=7NW{%j1N4)PB+z-n(%+qozQo|)e2V+3s7dfd$hY(gx zNxqFvQT{- zbh)yOJTtVKR;6k5(pYjH@z&KcJ@3* zLV#dP1U%(8TM61K49oU<8ONcagw(C@h*Q-M3KEDQetzUDaen1%cm|#^ag>Y&zw9s1 zi|)T*kwf^!rEwoMRS4G9wIA}|J3M*f6$QH%A=37wodww$&ezSQbf{5Kj;G)}6osKt z?7?wi!YDTkOeI~|v7Vv9kRfXAY-DTLOtO zC{U+Ks|Ca_gF%JP6+9azj6`Z``h9zkUa4GeXgV zvCOY)5vc;9UBCbCP0juaMj$T_!Ho>)?WKO7NYJ5zJ-LH4Zau3L>Rc{dZQr1B4ucN6 zm`+p#6XTb0eysN^MezhetQ7-$K`D;H4i*d1N8Pikz9+=MPknX!T^vWgG0@`X_q6R} zc9IP&Y8U+sg zqHlo`)7UskB!EAeQss2sa!WLhgSz+Q<8>>Wr!@kC6r=F}8V3pksAZtlAk+4;0FGAf zPb@1jj$jhtuofn=xMmYE3(No0U-*derN(^R6bq%E1@PI~f16Eq;5N=x>I}{Rr99Qc zQPBV0hPG@WgaeNfnUGL0F%_B(#lw;Mm!P+I`jZliYG1t zN}2)N4WvJ-Xe><7B8cnN*)LgMN8$rOsSj==SeZ>)9ZCGHn266ex08I$oRyd?7L;HN z1OyxNwW`lpY&M$t!V*Z6**rSXQJ9vCm00L%8^np7A1i2hmlRzga9F9-Kc;ta`F!rC z@;;Mhq*$3kE0-z*75hgTIhg7gk&)>bp5-#xv*QuFRx$hglFHSZ{LeqZY^9_n_37-) zl^_of4?i%9|77u31WK~mtVsiHmTM^(7^XQx$2Zsfxm?;LNiH`#51+1$ST)XwmCIBM zEap#Mx`q=6A1~Ie$@DIW%M>#;9xs|mB>R|u%+yOJTP#1GmrJG6;&3>aA5P#LPGpVO zAsjGWg_USGV?HAz(b}x05Yb!}6*%3pWK3$zLH2DO9nGjYN0h7f!)lbkTO&Q5wN|?b zsg|1T>fh$XM*~DcLV+$Rpi0F4(SQ)Xt>+mI zPlb7`+F8ztivy_PIRK?n6)N=F+`1EOaqYdS{=7l6(P$l-LZg*Y ze!Dkpl>IeKr^TsEY49eZ!Q-Be5dSB!Eu+@^3tX$y#Q`x>&uk3Ns#>{r1ob{~#wKn= zB-LW)A7|N?!Dz$Fj`cFt3U-&XutX~dXmC_Y<+kgAZPVEZ$aj*h-sAoxD3#9j34IZn$otP~HCXcXwLRx9ODyg*b2R6y8`2G}xKE1ELM zwvZw#{f3T>yVWE@hd#o&TuAWVxTr`tehW)qzc?qA($y;u&#YH+wz`!J-(spVk`hxY zn(%fSI^T2)4FqUlU=$tv=>vCC)zs{H=p38FKtV%OWffCOMm2sPG5qyqcsHo)Vx6l@ zwc+ThxmpJapfRgFuPBsVRUyl{^(T~4S6A1`Nvj?QylNVqmYr#&T)BkNgF=O#hi8>l zjn_vGpN7VIcfh$N4 zvy>@ikGMTwZOZ2L<|HA(xUM(o{q;6VP#LLZ(@ey~3;{avGZ+_H$l{?lmX-)x96p;Not1?=zIqN`1tSA2qxvCQLqJIq=AT6p-e zcC=8@dFwpW`+A925QEF(v3IrInYEWB^dW@L!(+XC)35q2V0E%lp;jb|hQ-)dLaS5x zaXsk%F>9vgS?klQc{WondCltf-j&DwG)-}Lysif!tXA^E%(t?$w8Z|?e4}kYlVcV~ zBhukwrT#~iN5DL_)HdyXggk1~I;)~os8q^Zz0)zFu#k>~{M6XPIoGCLFFlK#P22Ut z`|D#4AKq1EF=R;+ERD_xFwvOJ(auR(ys#gAeqK&_92Wn?j*2dN*VFbu+#JUkgvzyX*M>mm-unZ@BTEl455--<}lD&+V`CVJm1o+ z#Af}4#ZvPd?Xz8ub4B&u0~7etApwg;dWn6(uNQ)o^=wKr#><5twXC{CY9F=ICbU|t zXpSg}kR)2X0CxAAzT;Um&x-2mD>RxFc8eN;KvDK7I}}TiRo72qK&O{3f3_o%Bn+ls zS7^pb@WUYv2J0>JdfxXXGmLS~p01H61XFyFB?4`~j~}Cgq`FfVjn{?Jy`IcH0=Pu= zpQy{wlwy~eD8|R6xg1&ZOZ(kP3f*81y(25tTHR(z_ohd?-)Q-sTE?u}O<#h)9;Q?( zvm_mFkkpr(bQ9VyUP!IFvX?X}4?b)sQ|^`84LPTm{6LXds-^j{31+k$mp~%LYy;2* zI;cxatJ(qbtxv+0Bw8Pd|5y)}wzwsROyAR>y;F$A=xTRTv>67csgBn)5;M!a9aLO* z>P)}md!gGdeL~0vytVfBs$_?>9_{@U`@;f(Pe(hYMVQe`e(YSC+Ngl&&NtU-?t5g zefM9Mvo}Oy@O9s_%zU!^D{o>p@Ok#xHH-b7(xAu?IadxOzyqjjfNbJDu~VEeX# zfrocG-s1+70l)P)Go2Yd9<2&{$XH~0> zc)yhiYMse>%RJ?0qh`y!_mju^V?8EE7oz;Kbnl;tvFFT&#Oev#F-X&z!l`TCd7d zg8GT|{%A&YO>RUIg%aA<2^&+l0B9NbdZo^*`>m-)g=8|hLYq$OL^LK# zw^Qiq5Kt-{omRiT+V*@UnM_u_tnwV5c(M-^H0O&oRMam`t;NN~dwYe6WO5(RhYKoY zsyAGdJE|D4iu}|m>F&pLp|K2VVmlTeby6u%kXk4{=@b+$qlEM&cif9rfkXvS37M3^ zP|`6C0eKHo2_z~LaaO%JWOxx;^Wn(#rBfqiq%8vPGg+%fLx%xWnc7!yBtT= zD9|z1(Zp27AJjW$BJo3N^DHg6$t7u?;ir?>k=3Cv(C2$T{Au3kp0@O}3k0>n->6Cpo|U zZz?qET$*guU>e*pp2+ zIlWJnQm#=qa`D}r`0>&;O95~AbAxIHrx)TQXhmVD-B~8zL73lJgG)ml5)!gnu?ZB` z6?Ah|j431e_5!{!qAb+LwORpq6-g!_J@A|8vSN0kOQKk598c~-2TG=Z(#?5yU_74P zdNcQTi*3ubD)5^U>I{{Qp4auXM!k8rZAN4YjFPxGJ#L32|K7wmbuBW(C_GTRtF*J; z>;NhPKV0Ne3WPUO?%|9--fVG9l#W3w^Ga13GkYY&^ZISnVwJK5q0kLMGbgyANUt*G zl8IGz+jSdZt-6zMI_<~5v{bzGoWl;>O!ikRv?i~22e$#|ZGTk7zB+_217kdRm<-?* zcy1a_F{SzYt{l)tzwnk6-MA4VJ#i8d_*c~=4rNgyPMWZB#-dG?3MRRf_E7gK@%?g7 z1TCV+;^CPXXD6~{czgtBsyg__-_Onm^oRTLc%~;lNP;XA{DVRNL&29oM|L)z{a z;!ntbadZ1sSeYwJW@B=Z^)6C!qCL7^Fz*+MScmB$@IF(nl2(MuKbj5|8 z#aOrvDiWoL1UWUXta93OKo1P_zA&z2HqHt^cu2pKHpQ>Y@WNC`=Ih1i%#xvZG8zUq zLi2u@pHIMRX`!;wc>Brh6-;SbUc+8fhsxYS6nVf z4(EOXFZKCGMjG;#kE8v`cvlN({gis>qYM|ck*CXAn1kDG?fhf4B@@_JRH%)I-z=+y<4q_Rq!34W&2w|g0igu(#i6uKquqg_ zl#W&l@f!Z~@cnD-EiUJI9^!{)<3B|;?#T*N6YR%-HmHB_+wD#Cq!v`QKg5jO-Y+8m zE^ue$UzmngF0-&Z)_MMESviX!Lah1CFX?CEAX%fp2<}{?5%p3YmCCXB!9Ro@`%{+TG8$&*zv2qP<I~>h&TBPg+8K z`!F4W2!5%Xq|w}~RJ)h2r?~Hbve(@+iYh!`*B@GMc@REOf133|B?v2W%h_1eFXKO) z4sZOiZZJHlhH@U3Fg=}8h8x9X>JtbSh)EP560$+B-LH5)U2xg)zJc9}&hU~?St)t& ziLI-wvNo;H?sOM(?iXXJi1Ne<)G|BDo*{Q!1gc{uAam)Q9S^g};bcf;j`Rx|J_37=RHsf|p_3{-ytA?Iu z$4mL1v$r~loD3@!E5!3U_YGEC#d`C?Xf;_L$MMCZPgZC~%1!peb&_20=r>L9OZWYR zW01PXb~N2VQIF1-%lXjc)UGqV2MqH%X)Kf#nvs zSeoSVuH*_BS$RAmuXMgQ>K2uWOT2~uz}Zf^Z{AXXFw2%G=bW})Rf8{shdRm;XQH9g zGDno>WML7ji1FK~&g*B)m>fx#QmfF-Q}I)$kj~)Mz{1ZLV~%-0xoA&riBV!TIUy#& zNdU{nVyigJEFD{-{9e>_d1SwR*HL~Te0gEs(Yy$j=Eby6g~OMi7UsS=X(~pdqSfxf z(L!}^wm*`h=D|ze6TQwK_;l`C_kg$bdiqn3zoHQ0@deoqIHzBOqR1omt=7Xxn0sa$ zfuW8xL-k7r2KBvVMaHTpAp}$?e&fwZv{M*SWD5N8r^GkV0-!9rny}DtKkYcp8u$IB z-D(-BARX#v=k8eP_wa4?Qrj{M*qk177|CkWv{XwYQIkybZv&;1VQy6)YpHYzfqm73 z6B|k26t5#&ASSaCea>w)G_Lz{`wMujzG{jJ6w=3Aq&QDlh=4&}d@?05PJIknz?P?d zLAxFC5xkJ&8JUnFHC_3D+{k9=^TDnDZC-OU)}kYkyc~G8JNZm~DC2`-n_CI5UHW(q z9~<7wZX3+YuU50q(d4HHxvy4+vs3= ziw;)VppHBu+aK3b^L|s+Zr8Do8(b@mf_*H;<{gcp;?vnt>?%j!hU~Zn=Q2K(jm=q6 zuhyExG~zm^UVoISRt<$qGBPsun7i@Y7Rh8-HGKw49Xk0g<=nDrZbP5jlavTIs7!dc zv#3`!WwSPvAGzNPeyy3&K9LSoci&)lV#i1;GZ`JAC<#-Y8Y&I2TzJcuA$4liYI={E zQCP-e{;@V@4UDOrdipG(7FkhVG=<@UNeo`6dp2M=(;D#c0%uuHcpez9#$u8PKu)w2 zh`R@y*D8R^mlUYYHO%PNgXlvV zci3RDZZ&5kDYdcIgAOwQO7%ZnMhlA+9CI6_RMoDw@DbN$R?LNBq$16phx`~DlIHLs ztZmtUh&Bta@JsXA&i6LkE2qYm@1as}!!y)m{t62}@F{-mgWe+N0jx-Q0e0RYk5#5X zL={V=wM{@a=c6GxYG;|m7=P|?m?oF3qDUl_JVxemB@O-4RWM0uYRV=03WF6WLn zpJpkw68F)>GmSzYpYHtAk#M--iD#&RS0Z9LdUGcC$>h78Ow6AS>l{Q&G@6g?p?E6W zJodX{tJcS}T1)7mSxq@{>cb>I^#Tu>ci!q=|KyVi9bZZ?g{rgFhrxD~cC;T&?2BK^ z_FNE}v&5x&S8L5C_rsvuJcEKmfWX5vN*oT4@8Owm=5n^*8LdE@5-Yh2jS2ZFg7N{D zRXBac%il03V}q>f$PW_WeSaJDaJQ!@lapGe zos9OlDH0dI#x^7OQU3k>r8RozseI<#mh6bV#qI6axJlKqrMsiyZ5P`36L`=RhnKyI zfK%hDb}pHwp@sH?0?QYisZ>>Z`=~8yBr6o4`_)vi!hqBlseXs1xk_Xy?QI zQD4)F%x^>SX_Pzh<5O`0VGAIH2%B7sOp=cVxTaKzg-QZ*3&YYf`18dWrvZccwDf!3 zjr{iT_&eryvXDjpW}{IsDWo&CYI7*ujKM?rXmd`^P)v;RtAGVDX_iCuqK9_H?ZG_c ze1SEEqAd1gI?S?ap}{jz$LE)$9B+`h0*OuE_6mNT3E9b zd4YwR*hsT}KIM8v2YRz+_=KDpy+J8x%Y6add|WD-5;*T;vsm)Aw(F+rGXfR3dG533 z4j#_ZV5+M_VurSsb1x_p3*Vt4_SM%s;777B1*K48ai-MV4|X8E4hH*GW&19aPevL0 z7N+Md1*Q}N%f+}jjoEnim%{5JQ$Ugu&SAR6LMOIeSB_`V7p%G*rfrr#$ym*o!nyR7 z-ljdBtvL4&qER`Nb9xwGHV05JKVssBLUKu8fD0dXANPHq=64RiAjnM`Vk9Cfsb(*pb*rzY4UCU)vAACN^h4zJ%8-5Ei;H?yw4lII*C8Px!8*1yY0vo8 z9~{6N7wD{yV`N{^T3dk(g+hW~%l4V>v*c;+Zn4qz52|>5SPr`_01kX$2JZvE*ENP1 zr*_kYTHlUfbxE#*??SE*(GCiw#~3!%aoltGVG!p}H}^p>t+^NZyZt4G=~P>GVV?T& zD#fT!dWHL)sQ!KxI)=d{@{&vubCd#bJ@L?(FGu8A{YRklk=-<4>b?Dy*z-u!sY<_H z0O{32N<{_^D&5*K^`+gWsP_=PdxH`%L`V+DPb8dJ2;YFNd%A9+9l-jhN6EENvFM(V%4IS;K93XKAAGO0PNic-lp1d+Gx^ zXV+SFl(%dTEKqn1f@>+LBhYZKP3TIRtH8t*2pc#M7TZS(FNj$NuSyyl-{H8fSwo7w zheHo(q9U=bP^VbA)#q|dm6}+oSW78z=x(QxS3?nrbPXkHnG~%4rbos+AEo2FuYCS# z&F1y9hqGh)F*Di8aJ-mlwPsf(!*|oT9(~XIv%h*f5YV%tFbGZh3;u`5&kuw8m}*KJ z2% zP%^b!t{?HLC#%C|9TPOy0I~cb+~Z$HEQYB0TX|fvp(XhXSqnHK)?u5k8Wu4hjVm#0 z?UCJ%^2n4)-iE!mK%3+W$0nPROEz~~k#rpfO{{9jFmv}SmXnU7BCTVE`|o;HDObYZ zd0uMhI^Tq$ZW)z8Fq(eBd_o#f<&i|Q91(VAXuhq*=;#>jC7mmEXp5SxY7{2E{vcyQ z4mD{&$RnXQ4!%kA5uGpn@q?T_?R$iH+=ealrlqiW%WjP4TqF*7-_!L2-d>%VoRH3R zu7CeAYCFn+kjFbiPHU8*v{b0@1DRo+)P*y-etkdl<o8UXAKXH!) zQ|Q(6eeW(;VkM^Ewop;Sqf4i^_k>=9OCliZPF?~GxvP*PLaS7)cwG0JCF8aUg-%Fv z8h=o0KfRA6k46a)pH)8mUe_v=BMiL-e);D&M;7SlkK&QBP>cC%RmV%f^T@fJr<*~0VljN$22si`ENzoq;E79S@p)TG$ zv?qAX=EW;m(XT^6FgcdV$tts9=(E3M6d%IYQbXZd=Ur;>@jRrE8!YA(;S5;Vy7Ng<4+Oh{`mx&O|gBw}K0&EqXZVZc-tQfDY7XHq@iku^S@ zlgR#=>gpx>8abbv?2gtXOZL7vJ%q_>CCn;zK@nrI#o*`h$>e;g=JMoziqrXoa|AE< zcnm|ZtaySbS6#%J*Js(ZUX|wcGnv=7Ne2~eD7BdsMeeb)l`nCf3noI_Myb(=-VlJ} zM1V-CLq70`JgtSpyNX|DIsZyqq z;<-lFNu?#xGJHXptdAn3G~gW>ZKn7 zyhRU|Nk*4K`VwK=^%ekaUu+GH-itZzIe)Lfth6{Md9AkqjziW{uKk*LPo8Z1_L0Uj zQqG-O^j5Vg26>#LYAfnKDFKM45K7>yYs+0bt5-=D>XstqDl7<6&Hu2srhBIrr!J{~ zZDf%$D>K}=S!&d6b;;`Ru4;b0fTGu}=ZxSFEQ_#4G>U*F+vbj)S15R~+w5q+sF*qq zLlytVIn`7{;4Ri@p`*m5b0_=~7L?!M zTq=S$K$(yj0{%W#3zSMyglq1j%*^ApXTK=^KA{?DGN_*1qqyZi%8;Y6J+JCluDyEG z-6j%R@xj#8Bwv%8DX*v6B)KX%R#eSg;%%~2J|?+bfM8&!QD!RZgsbB|OFEtfjePK$ z>o5bsWnJCOlv-`wP;SaTW7D!C&wpr^tZ|PA7GL$!ZV>o-Iy&c9(>KX?DBLDUF@dJ) zVnk{OqyMLfO@aAhzT{HGMQ5 zi7f`(U@cNa1wmPCP2b4EoJy;_&=`_kGhADf%&a+~@dQ@l^rN0@Qm}Zvg4Qv^ zZpXf$_Mr_B%WCx&XOhb|xInaG7}>W66|KjetJ<(326C@Y%>en9#}my6DYbBrziUjH zyAN9#!!5?ARl2rStGrd8)~_)1E1Ys$`o7YT-Pz>uzR&OAcz!;Cls6K9$db*fG>pb2 zlg!`#jzIAJhYjaeU4iIlhWDrN0!~~q4^?9Uo(LXK7eo4@4{YuuQ3x*YFyzoHF%+TX z0LRK}9szl5UyVw-m{SA>(p~^4}LRXF|FcnSK z8Fgf#k{ot+IcTHouR)?S;|BU&6KJ@NHqeZdSx`Pzjg0MzjKNheOu&Sy$=r- z&aU7bj;lA*Q(vwX&a9c}y}(~yU3NcJdO4{ShMH^0vl_jAtNDS?M0DGiguy!e93SWM z0q&4?BS1S_$K~I8jUusDr{^B3y;8$fYYMmb0||6&1X$0-E$ccfGPK7p*MN&Ejg;_w zY~3qN@OXkgk$AW+VCP~Kn=g%b*kgZ{2^~O*E!7gE8FVB!San%FT6+fT_q47i|J74) z<)zuio(h?qD5_eaIdLB%AIvnvmEa}1@7|jfhg*p zXc)e4s@(uy_4O_%97p7ZRD1Ml9TQMxK>bCZwvULt*A0ayfDwX$EQj+gU4_Ep;76!; zRLy57_%k&s7HG4*z5~orTqdmHtKU1(sw%-3@|(Ii(GwD!TyRksx9owyi42<-pwZw2 z#QhcA>xerRu1^(K5thk?EV>p1cBs(oJ{zKe zF>KnfVCXMX1fCSc^2!KMYg}g-SQxk#1Jh+P7C3&;+96bEKQl22!_}F^cIf%uCS!i} zKo=a@^jTd`X&TufxoAHmIG_Ow#}AASQmw}~tkh^7VSl61^0Az&4p$sB#2FkpPJ9J9Jjjio|f59zM*5TT5)og&MsE{ zQJKZJfA^sKepKhdA1W5@p={0HB}fwz!nWb|Rg||J-SH&4)a0qI)DoL3elj_1K*RH_ z#tgb?LJ?J!R$b7Ma&|_sBqXZt6gK1@{4hM#G7` zky?U)#6&}_&P|xGAt4Ja!_^KCHj>nDBRpf>rcTU~C_0y{=YrPh>DL)eNheD&gm-}F zD7+yHj*DIm4KrU|g+{T(u+kKC*ZMpTSOuZa!(uVbCp|q7oK03FV7)Xp>jDDtETy*% zCro*D$c-(?>FdEYue@aoGzP^6AMFujc$^3gI@B|P=hhSFF}F$cx=oGAN6xSdzYoVV z`f(^_{fi0DAjI8haPou}dq08B<`})6tj&4cA4SybEdp+3u_4+|Q#m8dXeaqd zGD|p{-_sER_BT^t1M79XbL&LlgxABi9qvIhi>1(vC*%n{3u?uv){w%wFUQr2TLv`~ zYHDGHL%#w*lde549}!v*m%JVWyG2Osx_wz23ZJr#bs^4U7DvcYf?Fqwt6V)E;fPNG zneaC`5b}2#4R`unnEGQTkJ(G^X=px+fsI+{#VCuat^&JvqbZVOXO{m=jcnG-o+ypk z+*o|i?A>7St-_I}eQz8hhH{p9cH{_Yg4Aj+Rni5GUX9bxR)%?u+GSab6Wc}XTW{>G zXiBG?MmM7<55h5V#sN!*RmV=UQk&Bpx#%V$T(sq1qJ~Dar+MH0tkqR_-xU*FQP4 zQlm^(X?k81k!VzgzWzEZu|==WtGidFTFxlX)T%vu8=3d7js@JSHmVvz7m3AW}QKVRSk&Yd3O;dwuaXrgte_L^1fZ^tg6>??}kqX z=n5YWF7=0_l-1flF4iC@{P~UPHJXATxCAPKQJk}GVNG_@g4U3b$xsY0k#xKU?K700 z2u@{o2i_;6e&i~cgnh0>Wj2!QikawIfd8gK`?CJH;kkRYO{EpJ6hk|uwa4ami$gPV86@T zW!-;L*;8Ub*E-4fCrmuZrKyPU+rs1KZhHdM!QND8K z!wCX;dE(B;=yXUQ1CFX~kz0Mp*f`M{QmT@SSQjSjF=|#F93;B~woZpVVdJc(l3{Z6 zo0$cflf19_H+mnFz90Lg85y)v)SoZE7&*>qwI1(}rjmJ4jfiAY+A(*RP1~@R3E!xd zDu)O2;p(qV*x{I@_qq%4*T|6zBAqr@);j3cNr}0e9HL?)2EFpGH+mANtAx?|_q!54 z(TgTYt7gb^{lGcSc@m=p!Nxj=S6ULsP5|Q58MSVGp9Am3Y?d2OYPhLt#^nhIkZ9h< zex!b;E{^PP?0J*hrrzR3#X%iM*_lTjrDr;f@u|%Wg+<)AS9CzaWGoBKS@);sv77$} z2X`25LnXCLv*S(}%{cg-R4uVP zbTFks4drdg!ziZc=3ob4QPE-h%n&gFap?*b#_^OG7s-rdFl;gi;o9HBMo1B(XQQR@ zw2o9)lBh(?1SCRQ-t?e^f(jtgyR0>M(8?EFSwCHFrP=qiw(w3dg%REf=?ly(43ZK? zn`~-x^1LIf zDAxb+uIuJ(&zsNBJ)%)vt4q%Ti_Lfb;=@T1Bk!}~Y_K8`YzjYJjGnCG-WVo)#FIxq zxj-Dq0=D5OPuY6nK0qvxKDgMzS@j`9wM2At-=>NP;4vd-fglDee~d#s=B#?} zam_);t7z*w3|oDyO%n;Ew~XKWm*wX`mY*Ov9u)|Y0<6aRm|EnCe*ppK0pO5Q{1ANs z`*aoP-@(fJmWpERK4stek$h=5OeRYf*Z4!gA17WQ#2m?@p2}q}cb$_hCHF5X(gq2L zAmr*UY3(oUXU2zwi@2!#hJxN|V-H6ZRw<$7iB6Sz2Fd@aKygjOsrLvTgg-vcem4ZW z)BTC^4_e$U8HD84?WahFL;*X4UPkNxO{U22tjBq30?yu!1SI4NlK?O{UCCTWAk?OM zQq04Efk5sz!Jvnp12ixZK|F>_7<_n-+TF;_SUFDee_-);{(cZ6E?_CfsQ-qezv)eqyycrWTmvVSeTb4aQ0kOAWDjAmo0xFwbq&YkP}AI0_=T zzaR+oKMB#-#s>Bk1#?F3AG4$V%)kvDYJmj=xGD*@LdhX-!G1`9`91q~Ig78zpiM(E zG7|6vzn>=_hq$Q&xsoQPnkgQ8ME`Os{2!$+BBPZB0^8fPEEMUH?Ebk|^w7n_yI&&i z>YXHmLp0~&ZV2;u{^k$+xaanvAIdG*x=F* zzc`Vze_V=I&G!omJgOG2n=3?sXlC-}&~C)|f*{@z%m-XV6f}HYIo606G&=zR1q~Mk zv7{(EF>?ZegoHGKfSQ|{S*w{4k3gOrhQ2+0@0*N_+v7l*V@*(Sc+?RtKZt}9K@BDc z*%u|1su+tywkBgfjMTk60R@VP3k*bPNKi8M1Oe%{*#R5e#BlDD1f;|v?Cy+kVGTwJ znOe}~BW0bEtPi>$HOZ~pq zjE%tn{wFY(x@pWL0_`cEX(cyz>gXsAiG=!B=-~v}`TAi+o>3amL^#`vkjUd}`J9p; zMA|SS=tGiI8UXJydEzJ%@)+PA?_07-#Ue!#{11&V_|(gT74s50><$bgo=a9-T3TLR z-Q1YH(lsT<9xH+TBReaPpL$$c9}@yOhlrAr`4{^_KvY1oN!1a2G(YtQWxqM)H;L(+ zDwF1Ahd|_*KptFi5?D|+GHxHEukIkjgcmy=VCoj)O|sUE2pB3dM)Cq`>L zNIfH7B>)=qEAmnVm!Ll~RIL;MI*9x?$BlQCoypzY^#}s|)E-1}+nGCqh{uM-2XA|* z9yknN&r9qEG29?xu#rHiFWxl@o?ptz&WMCDz0-(Krg z^yP;uiqia(cK=7C^zRYqnx$t84h)F;AHpO*@;~&;!2i%=Kxr2A{|WuFmaXshfy{HV z7#c?7%(_xB=PwJS5r=Ss_(+I|dX;NN)7!AgwI1vp!aG2&{f9gg#Lo+`bH47U_l3q!t0~Qh&yGPtwB+{^QMY&5s0R`3 zgvW-8#S44C9?}&S6>-?g#lV97rEZT8>`hrwX_kSJc>P21>4X&DayD&YW@t z0_4qP)f0#Q*Sming^l}^|5H;>)J-VJaRwPwMT(EiEe0>><#d3FvO@%a35WLg!uRV6 zSrQeJ404>y2hp)y+IsEKQm-ZL=llvm-2orsKa!si!1`E&*o4_3IK-<7zdZvgLpr2a~w)%n?3U;6^Va!$o~X{oE%)zO(`IA zo!5;NKSYx`QMk8JtAY4`Tz?_}BWXjt9#GiiMgjx|Q1>rW^(&kw(EisZkT4KAE9vSN z1V^j)tuX%>tRWV9q@Q_{>HjLb5(6n4McX@zw72;h_Pg_G*AFQ7m(c)lQW)b%M;Bqv zCP?ALb4k{}{D$`bL&2C!n6kk)><}p0Uc)O zorYH=welBpnZE;2je5R@2)G57&4m3WD~P`v0067e|1DMamr++*ARgPxc}aZCzg7wV zryy4!M08fCUjwTArSpjfj15AVYZ~R<#chV+uM7V$4jo|JY@owPf5uV&B^v+^_Rj*z zkH>k&{;&I=LO|uLW1F(!uVKK~hY1=O=uf|VOU6+)rbE4*BQE;O_O}H7<)tNjPBi0Z zve!Qxd?2P=YLOHth(!6{d9acigXe^p>Y3kc_7$Xa0{ZJ|bI+;12pR3vn^w zU#6=jVw^<(B69l(^QWY}9~q0Q{Z}Mt*bzXi?Yq!AeDA*k?1u*;mrlpZhB%4*_524v zh%WBS{&fcGZ%+|HUn~}v^t!r+i;9A%a$Zxp1aDiI+s6QaxO5`t}6Dl!o}czeHo+<9D*ze&|35 zGbxkA{x|gUNiZm-BnLEFjirvmF?6eq`)@8jE`MG(|Ilhbecpt^{k`cf%kdYp(w|VU zOO7Q>w0Z#(^xlB~doRFCncAP1!yI2uPP|lFk`p0hGFfTqwz>KriT#nORx5WI1%v~4 znhm9F7IR@qES3vDhw1*uk0TfeKfkA0xyZP;u}m10P$tu{JxkKMJw(a)|jn=EG<>~}HJnu)-L}hAK z>D=x&pslY8hmg?F@r=hT$+kZq3Po}`AmY?qG#h+{atUEL2#G#AG&CjhlhvBj>!})a zFeq;=EWG!8d)Q?CYV`5;?07K7Vp?Y*`0;nyl7m435hzK21))6!1Z0++=I7@-%Z{r7 zx6rJKV-7C&Q}sIbQ!<mWCv3C*`OccJJD<(mM912YKz(}P9=|$QMHhw-VK7 zr2YKq@N|U;uTr5YVkS?4POA-~fp75sd7A~1ql&q&udiEOFJa)|O4O=!m6cJ%E2DLF?$Gzc3wCpL$ehDo-1m)Mu)2@BX#8-4$?)N*U{gM9r zv*pCTP+UxFHGbh;cDJjkKq!RbfWa6XCBBpSvcRZfpE^MhMq1J5?RpS2a-A;coI%I` zWj7V!*HtI^8jG43#2r`1J7Nu0-O>T%gDXI1s78|n`T zkukxLhE=O*m25UsIIPLZj0=kpJLRidsr@{a*BLFw_jX0J+DQC&{dyMdN+%QRaPHy5 zI$5YFJ6WRZELW83Red+!O#xB=62#wrg2K5HV|c3i+S&|w)S^*Q+AD}1s1*lcT{KpA z5(xjQ#oz>hPng1eB_KanfXH|``fTe|;iy!qiud0z7{5pIw>qa#YbXL&ldtMg%vByh zOlS~Q=^9y%TfjqG8@pAu^JZk!^YLQ!xBV{c_NUR-R;5#g45{?5Mh4a&dTw% z-_201m`*E@{7;+L13-TQn?~h$pPXV9~2VZjHys=z&$o+cOiF0tf^=V^DsYrHL z3P+_>`C_g4HVlzSv!OPPDTPk&Bb*G^_oSqVr&u@Qucp|A1HMzZ6NlLtd!i27KgBV< z+3+V4iT zZ?%8t^Z^k$58c~;^Q}QRC^4UiR55+AW~ul>^(LUl4*1MzzwVOXao$e=f*uR(Z|bQl zp3@%-?LKg<)t?*(zGSGVsK{kj>$iCpXQ*n|S+u*KZ-U8s7|8kF*uVIBnoc%;=6}n9 z+(QJFaM3?l&reht7$|=Kbdkn6*8(2-H$J4od`XX^ao-5Cp3ilbX#9XI*Hn&z(QGJroZ4e9AHr#EjT@{uWj_$;R!td0kQpi z_(=PRrIi_lQNwx~t7)ms#LM%;*;dTURu7s0JQ|t=*3g%6^*$|&envAeXPDIJQ~!e% z{r5uaI8aQQV;%*3er3VOU&Bxc8r;t@bI}gk-`jYaHhbKh@PR(q^Jdh!CyTZz#Y>Be zm0g=lF9@lH1VP$O_8VZb-R~EPm6Uwy-=w9bje5;U{BB+Q_u`TGD6oR~|AGZECOja{ zwcOEZ-msgDMGmb>Y>({!+OaxYO7Dxp9oP@{pBacD5%9V$w-FE#5jmWzGH7+{U_|kH-soo&=(pwPQ?kMQlN*piFH@rv;7TmG|19j{LT!DF zAkC-4hMcsiR#yDyRsy6uoBcJXlT%(@l}mU0I~HtzQ%~dyRfc~;M<*pngAFr3?AQPE zt$T}uHaxe&=B+Offozhj#j+>KUGO55;V8l$>MiYD-J~jL`40%y@8chw^C^_!AmIJ zl7EZa07+8nLREmQ#-hfdQ&3P`p_@c-p^q0{YW*`FfQ<{4i?$&>^UvMDi$Zh_Aeqtk=4)yzP3NmIR$6aNDm)Hn8OcdKAuCn?@eu-;S+-H%Fnl901JWf08sovcG^L z8D|75^gjX1{qO5p&8Kzww*W!^TdJU^i{b_T^CEvpfoWhjmC*8Uk_-R0RDqYzMHc#J zj88BIr~8COs)7ezLr(I2QQw5ZKp=P7nAen}SgOd;y9h9eo^s_XD_+PgQ z5Yd><9FYf?F3&4`{+ng`$7a`K%SqCf0Tt1mY?5!shvIwhv+63 zCW{&Yz{!DF9~NFsG5+LzZ(*%P=OqHSM6KRb2r$n97C+evSymYu-w1~D3kxIj{H=~C z8#NHl9J?uz(HaNC!#~@Wzh7d@W#nd3Pog3v{F#7Iv&)eohESM1>2dWvb~xA{uudT- zJI2CJJ$u*$7l?9sfUAyMdVyIybV{(S*k!ovPir`HTyOQrCX5O<8}2A<+0$G$t{JK{ zba+J&yXT2*!ISZ;~{30W$5;C{@-GY8&h-5BJrawSoE>S=}dn`FunF% zs1*2U_O6syjIfYqZ$u<1U6|A!A*F>D?&K_gH_|`SExbXTxR{mdF;cJrl6-ZvZyMWW z^+O&;1rWZJC&MM`IyhP#ch)y-}S)%xz3LfEe!o zEy*&Hdq42aBzHhwrEP&kZy}q)Cmlolk91C&$|KoMzOT27(cj8~z{?p{MIqRi>qId&6fD2wfx>oG)~09KRYEYIm{3j3KCTQU1l#G0>3~5b6ce zk9?9MqCPIuPWsAA0WU7u&XH|w3hj8BzRz++opngRkhxo7Kt+J94hH%4ESuGPrutr8 zQCRvOWJ;*yW1|c)mcNIaPye3}r{BVa3JLRnGx<#**<;ZeQC8yJAJN_7?0lpEVGu)l zg*~)Q6@{}EG0-Wr{}iz)q{#>D!<%L95#l5`{{_n>vyZ?bPZo+)(g_x8?LJTUSHR3kht!lj_2Yj<$aQ~z3v?p? zI1@Jn5ZUCvU8;dK!3757$iUj=jbDSsWVyLQlyDp`(2=+N8H@XwTxH)#L`ws$OWJww ztgE}c|EtI}aS_Jys~T{uBbV(Yf$6=!BIxf=+RC?HYrg{a1+L?$jSr|q++|NXZl|jb z_m{M-3zFb(Y-tJ`nGLopRR;?bEiSDf)a2*)uKBHP7boPmNcKHv61`6?r~W0a=}cS<;_}Y&;bdDMr@x~@%;2K(ATGI4~7jSI=0iX<`_wX zz(~kCEn4jYjbYJdDYp=jHPusmRE^X#2SK?{>_&hF5l%%d|m%|cK zM(4$yIBYwNc>tQ^i1-13;JEFE-~Hw3CPOvqbEMKNB_#}Cn2_t@six?2_Qw^Ch&!A` zq)7gFu3T%}PT;iS#Z93w{UwWHKSu!Fkn6H$AX8p4{N;A*Mn0YMW1$SMfq0BBAtacz zRUq9pE(|}txXpu&gJ}iX9P7^DU@@Iq=>jT0J3I73+$CTho-<#EyBuu82dbxnZ&*QE zTl-+O&CB{Q4!a6#Z?Rq{8oxeWCaiQ4hu(2n1)syYnC9B^X1VMP)OMT;d*k3(zb6-z zhQv7o0ZFaejbuGw2y{c2tb8Z3-Wl)&=JD>@<0pCT|DaguoDN4LscgW3iuZCtfKIhI zg$6ypM=6h3gj-fh>N+efQ2;SqaJin|JaOUV!U0^mr@MQo;bV_>fk9nbg>Exqfpj7@ z!PG}_anHNQr<^^2j#oyXc7HuJI4FNw5R#WEv%LIcqRMdB&Y9$HyNWs)e9&kNKgn|Trf0kWI<+Ujy7o_i4>69JAKpcD?HQ6P5v z%H@Z8w~u3mjeagvW0b>)q{P1I0}QxaTA_^EJYQdtk?5r{s>~KvwUp{ay|#f52q^bB zm>vQ=x@yZA5cW(LzB}l(URz9bZdwvyt0D^=g3q^HuhkOpgBz}HPW|+L6}~ozY+Uor zm<#9kxo@Azb#_vkeONJP_3o~7!Od&GnK9Zq~W~Vq>v4mwka>vGEo^M7m zJMT^8_&%k3@LNs?0ox0`s~o=<$k{g9ML&gTf6BiQTLd>*RV#qRwg#wKlS@ zj-5dVfImYpK=quH>+NcC% z$ID+Y&A7`w4ccfbkE(*f`w>_JN<*`yY6Lzyo}Qr&>C%MTC4L*-cWqubWTe0kaY$L6 z#bNrM6^00pw7)Nr;QfJ;d|D-;lX!b?myoZ5Y@PA+e4*UK#~#}}nHG4|Ny-bc3@)#0 z_7nB~N@b4%X*DS!h5nzAsmGb`G|K9Aq@HP9k3DCAhDM0<4e_g%D?YDl-@D(*Yu;>q z9)7UMdqu3)3m2__(kqF5&br_-JrAVvT5^s0B45Gj58V72?^uu9vJ8>Gu=>alABboJwd?LH4@0Jy?yUrJ`u`(DwkTA~aBxm1z7My}*0WKaZIR_h6Vgwr^9 zWmgOXL$H@3;#C zuVWzXt_#lu?e}L9!J%+?h{0EnQ9mrGDDVq<#`y^j!JC1ouuFdK!LBDUB9R%IDVtku zv}-(hGyc^`r$4BTUc@cq0!VQ5}k(H~NbDJ%7#4nt86ne{(csg>FVl9mv-8 zpmbpdLtITo>Gvm2l=IzqjN3V0aL^HCd!(5tgvQtq!-0@V1=f2w&h&IuZ+Y1RQwJeD z#p5-ze&A-S>jIj6&}D>ACr#F)f=A0+6VBFUzn7l3@LKShAM8-O3&e+v*0O~@T+o#D zYz@S)-3vco{tAR6c5m{&i%eK-xpHgWzyZcj{n~fHL4b|b@skK&`RrrrwUIn!qEvji zn}rsmau!JNKt`+D;=BNh;+Vwa(GCUz$0K}(q^wXWAE?W$=p2@t6jUEE>|9R4py!nK z_R4%%w*O_|_gnKj=8f&n2;Q16PjVrMfGo`Yv%tO1yy#ONu4=udnuj6_4Z_6jga{g* zi!W%ui~(}xVvR9ew;=y{?^adEbEv~QVEKLiii+_KqiSnwYi9Chm0p}T8;#!`t$vTb z>c5_w z0V?0^`V*b#B-{9-)BPo0i;jT~e>^(62aFNQC1|oZR}F;36v?PmDrBvW47sg(%vtL7 z@uegfm1s+a{kUE6I`t{nIxGcf5xei>?hhkA(0IXR)_1-v&QRlRTuKc5>tQS{?!6sX zRL%0sQ`9dB74Lf%cgX}Br9l1na=G$pyix-J8mQc_kM1$N-t$IoxhPm_9Bu7CjKg}; z9s*(3Kp;30cdOE_|4hQWLhydLK1v+ei7BF!f0j=<3sK8Q*8f5yd#jsD4|)y*C$fvc zIBfDqMZSomr)I@UgZDQ{!{i984&dz)GY$y-X!z{z^VK@{*T-S*U7ID!MHuU6V6~|` zKrU;mHIp+xj<^KCle-%qyojN*@bu$hl~uiSF}FSF3I#*VCjV`z)UltsGt^rz7)=)| zl|DB+RY|I%2|T_-qnr9P_Lk^tVEShC$32S7IeS!a+o zw<_Mwcve-u6m~3l8^tmnHv?*v}% z*k8zTzxP@vph}IPG8lRsn>^iTdYzn?eWK35fQp1r=&vze>wAq#g~Upp)=xW*QyItH z{$^LBMP%k9`0kWWiFhp*hRe0%kizEL~x*6o%g3i^QW3yEddp@T zolFfYogxJlO)X-9XCxw&hsxHIyHme4Ua#wPTWk*^C9}a;0%rXmlH^nQH3H=ya)wk` z385xZSd7bOHTjNyXipV<1xvt3IzzRkaRKH&SPYw3Jpu?CISLd40cd_;=a~cgA z)-*;t_^lpQs*@X+`kIm)Mn2VO03GX95McFEy*dy~S8{H(kWhVzYd&7=kpIRW!T3C= zAip%pwcl|kQWrH&#A-C2>60-SE*S=@9!?4|Zs+`*4{H_yyCzAw+6@P=G?yTk5!d<| zJ$Ov~S}qem*)7>Kq9+E1G7)s@ZvX5aTX%8WEIKESIo2)}$s!^`W?>rYtjF+sGV?!^ zkE6aO)*6ykP}mi>EGWD9L^7gb`-J}9fR)6NZWr^lx7A#Ors=uRhp15!fzL5iODUYZ z7HSwcPD{2Lam>fkG}7O5Kth&e-f<7~9&!;-(uj3BIVWFPp z;-@Nfo28ttrorJkd$?k&QH*HR96pr&ET74X@BmzX;AjO>sAPP7_Y0C;us{!`Ycg2i zoX0hv=Ocpb?*S}f+@q<2sQWhebDR&@#^Xv>j0Uy{#9Hm%9(E|BnjQBUMY1X>`6Al_ z;-g;py9rJkQR^yVkwr7kG=c7yoePq1Q33}`4Y z@(_4R%v@8c%}zxRMb46Ik8`kbV`Ppca;aL(>fO1h=Ek3Offj^clrKStP_MRs^V~lf z;oRzWW6&>Eq@<+z@zrJi(D3j=lUP|EC-C3>!r=f``ZI^w)Mx?c;%#7cc4RtOGV4A_ zFtcbi{Na;+#;uI4+jSfh%&(+;%wvCN;|&amFdMY4C|`=8vE z>+DPahYMh`>bjJniE)>k2GqcaX%w|1)1cea?ra7y7ET!>YC9{GeGjLI^KMHD)MQy4VcIl>wz*97WtgQ-#$lQ4r5%%!Th-i^)AsQZa~# zK7Ds2QM3B)lKj2x;e@Bn^Eao7p(*I?2pc=-D&Jxy>l3Q1dLA;CRr~?1Ysfjof z9iEdWtH14}dRzWK0E7xfBYB>ZVdk*0WMS?x2xcZ)|3JPE+HZe;GMA0a6*`hbF=;zf zy)g&nKZ(!u=0fCmkD;$r>h*nnzi;1X@wl1(UTNt2vpY^sir@39B(VR=b4ht|aa?nP zBW;(~)%2S%HIXoagEt5|;m4p@7O*8s$G`~-7kN|Rdoff5T-BmAgvnlGWM(Sx(!Af3 zRPN<>`YK1%t@>e{_p1;*V3Xzduox559I-h5{NYjiaM}D?0OzQAAo}5K5W~)Lxs8<_ zj2*_6y;?t9AddbBIs$1415%A#9V>D-G9b;6z@#U~5cB2czP6$lAD8;%vQ?r&bMbk} z?QC5rp7Fj|JH8@qJX9&;Yt38X6r6J8isAO#&7Yvl7ZDL5gb&y${I@q9Sz@RnUSX;D zt0BfA53Z-_E}K2Md36>M;X3Y#pnY1VMkm&uhWOAk`V z;nF0Bu8W(|rMcDZr*;UwbF}!FT+6=D*zpf3JNr-e31lQ;E{?zT-WM)EUGW6 z7W_HfCbwfB<4Id^JH001vwye^x@Rk!xjb!^f=*kue2GdCKlQa}ayG-^y*ft2N+xtl zLg{w^(X1c3@~Ol4s8%$#6dih8D#;wZ;V)ku%Xp};d4m4h`!S?|q+r3Q2zr^b{CkbB zgbAWVj)TqNmtn>v{YVD3;@af5<+Z1(%#)}R?lEebRu8L3!prRW=W6s?(|_A zRFsw>szKoU{!!m5M#vY$5u9xL=DpcF1aK@vgYcSQy+cjQp1U1K`oe1;w`s`uecrDF z(Oy|?LRwZQt8KUi)Uvnb=&EL$MQ!Q29 zIrzn6JS|GWcrVUYgX~cH=%;*LLwg`=jTvVz$EbUasJM2*hX%*|rCPX({FaiLO1-&b zX=@}f-{L`~;iVHDM&F2=Co?eE;_p*iu7raX9R$#MoryqjC zj@yA4N_srZ2*{bJTR6z|Ds;@XMC^VQ)^nsbS{{P>NriUJ&(E~R_xX`TEjRD2ZSdl! z$y_cEz!LfR`D;oMe}7LL`L8qmhwG&(E{Bbq>If*>F?t?(4eJ6Mzyn3I`2@o zQ&mo>&2=SRlo5Bk07uUIdSUnIpbJh;r@a}AyeyjNVI#wBwPZ`<(D!>v2u0w1{Bd*o z<=Qq`gJlZ)C)U)Gy9M(XQ--S<4X)NMOvxBM(+axY6@pe-%$h!SOLSn8pKWwWTEFc3 zaI9XyiDcK7SbH2od~#F~hM5aMPyoI; zjik^}kTZM>HU0Uz8Cz>kVrDTT@Pz6>6Q|s_MMI=Lv6~rq^j(ly3+P6-MAZzI#{TY2u%q>5^Xy%L|(5 zN@{rg{Lj;u8)u_j)Z{Y)SWHUAJMrZ8P6OH6f{N7+Dd{NwGJ z1FQIW(K7H8R(rCr_b3q|G~Y-;mj~hAJQ>^NZr*jhqH$HdRr4GZIL)iNru4Rk|Ayk7 znqU4Z6H3k-%C{W1+@L3g0d;~QNY@Dmvmd+kusvkJ{S{U1_h9(Z)6XRevprCFg<5Rr z=jk*2K9#3KRy%H%XtI89T|iqW=5Me&ndet4P|m$A`K6B0g5Pwpu4=)q9XKo(JJ~jU zOtI;@I`n(G+oeCLY`{Azye-ylwP}tkc6(o;w?v-J^Mc+9CS#V^y$9eoWmHMD}PU%tQG&j~mz&ypztWDoE0 z#N%iK`JtbH>zlv|Gz^LRI%L~jq&$t|ZQISN%uAe=UDHQ+gED;#fxjzLqqGUWUcXCtI3FNc zShb(8PFoMvf;T_`v)42%;MH+CYi%#DQSY$@yoZ(LTqU0Y{gJiw`Lfe^g@K)t0+8Q; ztw~(fjZL|9xCuLKbXIfx78U`Vigs?%PaLNg($YHr}%j@>HI!ZL;Y3SuG`+# zn8&cFyJF$BJ)Z!@cX6RsgCjyi%>F|v`=`gt{k0}CZ!ns>#w>oSUP))|C%liulcB&< zuOE>%P-(6*lRGfwG>3(zYzZJ~rUMwiX>-wXw9?;w6kcn8s<@tXt-{f>e7vnnxc*{V zMJ57!=$zC;>}@~o_x$H;VnY8a9nx!tH=Jjl^}S|G_ka!zhe{|)Nkr_@pC@|fd%fVd zWofJ-5REp$HqLA8JZihr5^eMQtC5)hkp0tD(1e+mhdLR4F6_zL3%)VCdyY9J7``o{ zitveM#r04tRFyZTKr2h^{ff?&O9fDc;j3Ag zPkDe>v+K{KQR-Gv_FhnhYc<(F7s|Cq!I44l_@lZ(T9`PHCw7Z)HqbF z{H?e6z493?+2gfoY<`#s%JWSQdBB=qTn-vfJ*h&p@^K%QL(PKcotsFBlr=|#2Frai z=l(0J3{cFGRbrK;-_D#Y)Xqtn`Gn(qSk7nCG(52P5DSBBl#k>k|4H9@B@vz-T=WJ-_@W0e$dWyx}lZclWs-7Vn|2S?@k) zmoen?Ukp;HH#&Yk>yLiWJl6Mj=5*ZFTmAm$CmiRLXaLrrlkZMyPG{J<29f=jpLa?f z_>#|V;_KT2PM>NZ5Vpxk`};4Zj;3}a@mVT$xujpqq2$U8BJ$Smy0w8Wtl#}U=X-R9 z0N@?E55S)OuZn+m7^_d_yb7vS-RAV%qMS9?&3UW2V|O1%*hCDY=jPg6{$9o4=J;iS z2Z^(!l=(}obte;PjkDWZQa$n&vS`U5Nd^?AO3LI=n55_z zyPe7yht;$XwefalzNcVOJo5R_ant9qo9R^9e@H_Zt$^NYZU7iX$X;&=)7HKvUX-hzSB-Wc61R3~F*dJhfn6sBy^U#dtae?tlR{mjn z8d)x^e9b`l8d5B&P89@2!OQ8?_xQ<*o$f=1T(WNDh7sn1vdH=0EItp<&8=nh?_9Pc z@*Bb!72?y#im@hTt1t|TVxFl&VWac8n7yzu)w&+)3k3vCQ$iv!j}^aN*_l?1i6G*C z&xMWahG-IeuUM_TkMa?}tA;QGQ7w}k5{&|kSr4_U2PeCX#!j~?3i`W4RkC~d3Qd#z zF}1!m0ssE+3y)V=u|nK$OUjwC3ZwV7`%;CWGNt{Ya*M`}{QAXeovTvJZV2FpFvW|FhO@?CzK1CBk1ptL86g@i@u_sd_=cMvdzSB zY6=w8<$Fa=VQeaD@$>QValaM{YMsf76WUuM@2cl#7r%hA?PIIP?(EcJijfiaOKGX--A7LwVw+_N8A zn-}rp7FrvUdL9s@spcAL>Qb?N^}mAN+*T#CgPn77c7~>#B$*FKBg1sjLH_CFX@uxp zFp<0|qk#K7aP75_X1Iu#L8Grnh=tILneD$PmJFo|)4h#!eRR4irdi>5Ms7k_Y78$} z7(cR8uK6aBo9cD))fak)>Nb$8Ew`(8J3r8FLr8X3Lcb+l#3|aL$AS&lwCF$7;^g&= zSV-D2@7Lijei1##O>%N@DLs2FPDu0jOSJHZfDuKJ4RmTn)K@!_B@k;7jxm{CDul0xEQfIB5WtF)a>8=2A@{dP=2 z7p0d$Cz$~K_w3F|a5ne1>cYb!t(L-~unda7=ltiib$L1_sBtqXaMXYQ36}l6l8f0q z^M4Qd`@cN~Bx3mg<-^{vw9|Q}AgrEuK&SM5#^j8tG z|BO+!0C@Ck^U`ztr^iP!`fOmsOZrw)n5;k~S|J^0<=6Npw?oiPhc5_TsidXhjgWUQCH5m&kQ^U#z6(>-y0xh{p0sND-U07l{i z5lpKM3?zIMI$~a4-s97&UR)^9D0{cq2wzB{&pRA)9WuNqDR|R{gB6q}LqpadRa9F{ zj!h4*C|C#et@SkqBIPANIk|9E-THwb9${D4?!*old`g2Oy)_{n zMv50UQ}grbGekPPzS*4E2Pl&ZAnB^NTjK{(kFf)uB{Ily8hZefHaqQtO(Ud5W?B!G z5^u9g(BxAw{lfUX-zvM@s|OiUgW&qYk3kXa9Nx&oFxgbS$EZ@tE&tIPf)(X(Jnj(gzs}C_IYj00d%Fm$IY!pzZz%DR9yJ-zMFi8VAF`B@BToXAWjkYUv()$_$ z#c6&qih6>VD2US7G|S36!3bE;l#>1e24P`38n0xyq9INYYm9L4K9#Vq*R|J>C8PU) zhVsLCv$rRwG}>Avl8v2tdT)oE0a@nescJa=h9{WU+x#%S`$|ak3Yn4nvZZH!1W?~W z`U<5U&qy~UODPDVr2j*iQl5B|6dEyaF>{62tV4ws6dpEq@Y=?Uw1YUlD81OH`FAi> z(wH|xMhJ80omR*wTQl++rgO6X*V;1#q~a0tAC#2Lwh8Vz@Z_>xhy}V&6jgyf>u<*v zPaO%VG?CtKq2B6d|Cc#HpRh25J~$n(a2&;jB0xjm7Re;HE9S(|Yg7WM8kPwCD z zN|_|&ER((V)$8Qd(bsbMiTJ&}QvB#yV-~^qqEV8q?D72HrM%m&yMI&;7Vw#F0WU!> z@dE}&XF8Td^@2oGa34yTi=t#EBJ>|;&*`aQJtbBGaMi%>cqmqY!{&K7YXjI!UUwJ2 z&BrrSaTe5eqKYItc@WksyV`0vH4%K$=k;B>XGJ*#f*T*a`*tBT*J)lhs-pGJ6 zZ6*jS)n6e_7?us1e&%1mx@9eXk4QZB&}FfL7R}-Geh6J_wRPs&OP`ao`mk(4dQ`ds z|LDb+vK*vydQMZW@d>RcR4+JwT4&|0fD+Yja`MGD_Y7|PQ&(VnL`$_lgdn3sfwuSC zV2Y>fTnY0+_Y+q53nXP95A97MQkyj3iavcLY8>5HwHp6Df z1W`ILg|~w?kHvq z$x$tX8XyhX{z+to13#Jpzy8H)^$|%abtA)~=wPK4Uq!(KG5mzDYHZ)9#d)7vAyX0T z?kMw($NN?iOwQGIA1f)dok#Mjae&Q9f7V6`VVbRTf=lgUHc3PR(TOlAFD{ZS-(YDFq2gH%4Qj=b()#U9A8wX_?_-nc z&sm_7td_|*(TUhtHCUD!_t}l(bCdY*Bkz4Nc2Lav-5_d9L24BJ!{ZMJiFYiv1vy0? z=;mC4tvKL5DwlBBt)c<8O&Br|T%2PjW}Bz)8GW zaqEdba89E#D@F@(4u>=}-vA5TLL6$DGFdF86sE&9*uRTzp|=EdQ+%$}B+dy~Hvt8t znm%tqd1O5f8f@I>dAyiqWFZkld^5=6d5{f|*LItwBm}H}uLB4!CXho!2$m?D0wOAC z*Z3i$JIva!hxbM9^HOH=dOQMk7LZh}kBfxuF&=)VCv4G~{5-EPVhwf#ZF$AI2`VmO zc}*rJZFUl?(Txq)r4$x&@?7f$h2JalZC)R&SBUw&=d*=G@@ijS>*~{}Sd&29gxqmT(M)aD_g!A28g?vyDG;p)Jw&l?p#M?F%k3}kq+Y^Hac z0EoXqM}Gwcg^2w&nE2&<;I2r)1)zQ6 z;BwaxmM3H20zUP{XxfMVyYhR2p0mh}D+p*!YaRlBJnTcrZ z{Z@<k+rAV@$qv|aV5TDlrT&ih3tSn59g5@#u#1~iFQXM{Vq!bm3l9}shG$>>|} zKxUdQ1woQ~nczlkpV*N76gXPb(c!orC>nuDC7-TY&=i)@;`efrClP~5BR^GdwFWjl z{YFHz`c<#xGSK<<`Zx^;d2Km5^4OZfU~4RtG8o{gj)874H^Uy#RSZU&1>)B=OV5{<3V}?Q?t91uVp?g;C6L#o8xQacGr)h(=7| zv6)H7l74i*Ei2PtK3#4eO{FvF`|^p7#c4;@YPMWHkq*c6z2i1ot=W_HG+$5s&Ssws zgLYlES{a9{u2sVqSba*s>#P9G;Q*++4Kc{J9l%flyEV4hZ*+qRo?6QG_N@|VwSld? zQhD5Lc_B%7-J8H(ctARDEY_-!#XnQ3Ce#HwsE&+F@Y&EspMIH5P~&Iz$+S6ag@T6q z-CiCVpvQkf+Pc3w?0CLEOy~NeyXj|UdGJ2vOBvW$ju~VRaUbl6e>fO`*8TY=N!kZH zof@N%k|s!Z^A9$Q`O=BfAVUjKJO5QQZVLbbx6@}Bq}QDxGWf=M;wT;WrLUfQ8%drS zb(_!`F!~mrpWrh22p$~1VDrdlh4^5t6l>Q@1I@u1!BeyQxe?HPuD=AsP%S3{YU|yf zam_Rhlg`Ki^Q;+w9)0WzggIyG6m!|1GSxK%c3weJ?d&ET!$_*{oBdw8YK#YfIycf! z4h1sL_E(_`xbAneQN$X+9ogL-&oTfRA=pSv4kjB!0vcYUGkW1Sc zNp%9u_rb4Jhk)cfL0HG-u4M;yzHB`WCA)@cNhWAZ75yJB0Q)Cp-c zHi@BGsR7LUk#xNK`}!IiLdtG%H4tXT-wNapfqnf76teW)S-=Jj&s80(0-v!TvNJ@H z5i`0ZB5?HZ*)4!fEJM*AaKC>hRm58Qp8y)wX>ZmC+ef|1Rh}ERxSi?`eJzkf!D5~UZCx<9BAJj1Ba?Xp>m`f=iHtg~?n~><*3O7op>p56 zzuX9I^+(&&Nqo_ra?QGsu1n4J{aBn;Zz3x2#4g`StJ^fVHK34B4rCrd7UbPDcr!tG z)(hWf02h>@#6ZPE)m}_YEW4~aD~UUu)fE4bJI2Z#EQ)xn`N_B)PP?{}l7l5G{bZeI zT$fuh6z5-)a08}MsunT>2yO}^#G=9+@Q3x1>ytd1lkD)Wt z;NbdL2eL?AJ~PlCxrYBD$g70{&9c+&CWtLZzeDxJ=afZ=*=y)ws2OfB%tIb1DY zObp01s9KN_pqF?$T~ezQ1$yAUzYs3E21sx@60Qi?iv`R^-iR^o1A#LTkB}#?3R1JP zV-s!%aXW6?KU^DZw07?$3>SpZcrP^=JYUTjh>L+-1MsF`RXL4%W573)#EUQ10LF#g z=H4R|884YpC(v5lbfZhX!}nEl^%u9NiJY~+-Vduzl950-zXj<1+eHjg2uNPc=^J_2 zV%|P3^zmQ#NqwD9S8-q3z3(jVFA;#QBe4;FzgnlU9T*dM+%Q1}qWJ&(!%0rXX)3Ks zaDA3jtb6K!e(2(FV4E$(pxcyE?aSlF(qO;g9o!o)FqSR| zp->Exo^bRMP<3g7LAXQ#NKy!*k8X!AUwMQewm9PJ0TiAfh1Gg14Wtmn&ypP0^X?DV z0>$~3GsB>AN#pmyztnGVYnM(uXyUv%u62H>2fF_)mnO5(VU4QKSr-j%{gdwn@9Nyo za|yYejqX9b%pPxY+C|NrQ;bk6RwV8Ovqi7XlM&}Q&=ZaC4K-8qOqpN6EIKr)2v{@z zfaDc@&T_UKl@Z^ULF@Y)<`qz+Nrwad2D|An<=gpKs6>KIWI@v}xE>*hIHV$^f=t#S zWol)p8<=?Lg_hmHQn!<>ZVF04w^$u7ce^BUMg#j%Dr@-|5hUoO6Mlorv9>i0L8j#U zfsiB&w&6%eCo8R>I_u{{szF9nJ;IH2q1Mzh_tUmY1pl#6~ty``+38y6#4MGvN#=}LPJGm_d-$9s;yJreGDWQiv+SC42At1k(V%^ z@pF$7R#R;rm;nTfglq{wK=C_Yk`!l8Ks6slbou6N@j8J3K75lN-RM0sdY^mtqt_!R zgIHujwVW>4iHnj;Wh+=2&fI(?#rmxTIaps#!V`mBw!ywt_@QhJ8kK>us4RXH=lp(^ zDlt(Me$Rmw+AfNd@~HLHPIhURHpLtqm*iSB#1Sz$$SZM z7ZXL--`bn5NB2qtb3r2U1akKP^0mF`Y@t`ZFrhQsFHd%Fp>g9Ndw~?|cJxCKDHo9I zR|36qp}KYfV#OXMF-92blRD7#hwFmHqrb6O74*zn&X8`$WaFt(s{I}vy5ecptO$pb znVB7hPm&`|h7-=0?84R+kZXM&m{D$`dZxqC7~gU>S}%mcC8(A#=}_v+uDjELO8eKm;aOpx5zizUKGRmcxW)j4?%4$>+|EzbF7|HqWv2tQz<)ruU!@ z!^vV-&cJ^O7=@A@3KyTsDO9JfKgox(8U@l7qmz%kAK$`fa^@hFg;t?&tdu)C0BysV zpPEDbc`&Kr*VENsk~|?-)V#k~$a(jXc|s+4S4AW{ zyzivPGuu#bvN{8kFGEpwesjrG1{cWeiXbUBJGy-A%vOf>x(g1z40h%Oau-n_<)M-V z6&OTpbe}Xf^UBGY_r8zN@GQ6K)l}8ZMcutJPd#WZ<7J?t5YA*%aRKThA z9WA&3*KsrTZbpNMMxozms}2BHZze%WR`~7)TixNc{=4gM`~(lMj7Uk;+f!r-YQVUu zS2KHvfPIfZiW@r*WJa$uA#PFrzz*7*g<0#-xE;-tS-!^1nMP!QKPSMK%OPjgC6VA} z|8k4#^p{ChA_-`C>rMWkZ7Tfv`ZP2+m`6+(mW=F0t0FYuy`}%UDYghiYsBU0%P4qb zY40C47wd1-3C)-p&lhy-EaEI@ocr}IUKOM4Le1fb+Q*?h5%O?cH%fL{61T&i6DTu4 z(c#T<*cJj_@(m7TJlm>TJEs$1zRJg9A+_AgcWh#wB+H@^J^N(PW8=h)v08Oo|)iH~<;PA!a26Q6om<@)5Nv;xyK}G~zg9x%7 zVo|GfS@xAFstnR-t1~_vywTs)OzbHjw_d6^Y$)o+K0z{MwOv+-G&%b!9zKDHL)Y^~ zWf+2SQbFK}5<6JDz@EbJ7$>58Mz%&iQFN(jJgL0`C?j(Vq=8~AL=9QIV%{UXkG#!D z%(V7Xp0c_#CT3=AU&sgzU9hHNz9hyYzxVB#UB`Yo7a+y^@M>6i2;7P8{&2iL;*#1A zpLql{aWae+H?|qyUmZ3~sNj)+fWg*m%>z=Ap~Np!%Jm=7&X37t+j3ULGcL}V4BCbBjhnT6)gNeM>4ilGL~bZe zMCU8IoW$(hZf&oR_L-)MHivs*nxh^b-r*f$BnckI{#O!}X3NTLyiYvGUcesGM-$62q|T*J|` z=)hXNrP%zV_a<#gIY)2U2&Y6cu6wFyQ87p}SJZ($!0r-YbhHg2aCL3!+#`|ii7ALS zVQ3AF?wb!8T?8siGh2rj8iTEt8n~l zd+LQ5qmKG+15`XubM2RGf4;7e4(!h{I8B=A`BPHv`-rb31`q!q_TDn8s`iT>r8^Et zcXx_}i^SC4*Xw`~9-oT<*9_xI)(Z$kIj4jZblvc@W{r(5V zd1FjMcAk~oeyWN4wL`&<&Kvb_6aiL5x z@p%1a;ojzaL}*rB;e;d!it~2O?%Jqk+ZOf%s}K>tpKEs+M0kbP=H8c{31_L+FJ4Cl zV1K|?-22eRzsSdw$Z?Wc)r&`w?(NUYYP2ht9~6k0u|o}HrBVKt@riS;+?<^Xh%MKIcq zoDYztvq-e|OrrheSgwQDwO!MvV5!*}iuRlbU@#-b16aO39)hq*T7J_qGaTc#0A-Ubva2>5 zM?ywsY^y`;T0A!&vJpg51pN440}Bpy3V-eQUMC4E(5r>Sa6Lfvd03crkD_Lhm}FX^ zrYD13ZIowLIC>Y-vjVpa_G3Y|RV@+%LhO!J+9%V`i)~Ky*#4v)WYGiKfyr zl8QgHCkG{Tt08yJ`^~K}1HA4xcSrN%3XP$27RCFyd+SqPmb{QjU>z8kN}b#-XZ%R( zfYXh7?##}{qs_E9ve`~DGlGfx?A4P;VM#RN4!M2esU4jjw-v@t|OPq9fwF+n`!G4iRy7cW*(+u{TCr3ch1Ix?3k{Df`n z3Iuq#4jDg|MYq4tTn#6XTI+{)W}7#k9dm?S4GxfV<{E?crJc!ZWVnJFizqu923hI` zHQRv{e%<^GZ*GeTg4*D2Agf;%Wn0wN-|~s3LHQ{I3`AiXo`j z9l5Nj`9|5<=I1ZEcNWkqA@E15JZ}r3m}`B0T-%wW6E5u{tJg0U`GN!$I;7Nul*KubRWVgj;yIW{53ct=}Th$wS zR2?`^)bvmeC2+Dgvbiu^E7$55JFqtA#su$K?Q8}q-vY|h(jvBDl_Af_Hc0YIJ?>M2%hawdF-mi940M3JF`kB z0pUud42p)Pj_YyUyXD*$bJ-CE*z4E&D?5{_GX|`oIQ^}InwEV&FN~@d!>U-jr?(!2 z=jQ1mSrlBv!!#PsomwM&y=jXTXM>JazH*!>Yp5&h(s=oiGMq!=Fd5w(FV(mEI`(az zbEfmMp!J362z-U)Qr6&;*0=`!fHmaNfXjNDn3EkORl>~d+ctdxID&5t^RNly{@hELApnxl;CT3m+1E#H^2EZuy2OW{KuBdkFb_q9_gVHH6yzDT)}{@$zBY|p&sH<9HpRM4>Y-b)c8B#!Y$3u&8v+Kbb0(&}ABR3Sw;lOzQ`n)m_Nm?HSE-(f!D{KaDLkC4bn>e6_JzO!KfTEYC(($N(EBBH zJMBVoqA%>9JU{<#VnAj)nsZ0ZF)UKa`ICW|dhoqGOC(Y0`zKeL6K=~`H1SFMgv|2I z;2{MZ4#-iGdrz38*P~)-v#N2N{i1|RF-H(rNFPggvKVBuO3O(-RBxW{vhl@uEo>qc zM)K@~l#_A)C_8RLq?iT)X;x^2dLpJX^QP+Pvb6b&){z;(Yuq%fOG74JnTk&jo}exH zVQQd@jEouuK8KgT0~DX{TQ_Zi6$}4pns%qjjq=`31%fwu+CqYtB~^vwyf@g6yGtp# zFMSzamyDSt&{y9}n>7HY0k!n?ZhrZYYVqhfJAtv#Im}49V{|G~QTFx&<+Ya_S(1LY zf%|Z6wx*{>;}d-Wl|5^>fJ6g>w4@w?C)w!Od+DSJ9@A9Jq_7d@d1B=+WJTG?3u^Zz zoPbP1gra+FQ@u5V|noaZ`yjuRXHAt3pAN5?<)K*Oc5B)_`hrtP9Hj_7H;lQ_=9(B82&AJjxU7d|2 zx*MC~Sa2|1w8uB$y=J+bHGb_?MSkmTt>@qQ5VM-ZCJi55b#j3b;<$E)AS;b(_(mCP zd`@TbD8=1?bu#y*0Yi>wl3KTy*o|{hccz=E<+W$mF&+qpY>x>irz><5P-r1t-6Cwl z2F=bE(N3(%1YFy_Qr)G;F^?PLtQ|gh&HtQb?QKUc6f<%4nIvCB8HdZfn>N9(qb?!Z za96eZrXsO8-uB!mIo0B_9-q7aK-iPy;O>I-;(IXO8GNa0IWFV7U9!0lG}&_RxeX_3 zFNK)q_4hGhoA%(qNM#$fEU`s|(lgU0p8EX2+sfhJ-m1SM`ZR(*(h#uSJ*9#rlIaSU zOPuV!WWO-Z@^)?@F3#oiP*#|*;o7I>1%dF5;I@p2={fM8w6>in5iX~qA3$xCx|rI{ z8`SP#S!F(ocDThit)8e~{F(p-n=F;~lV4lZDE)JyJtCIN3Y)n4g&Lcn%~TO2R2l>F zJm8!lqY5;NXtn21%eQi3$t0ksF!7JURJl;+UqSCs|JH%9#g^_a_L7b%IviP&59W}d zWag}XfOSsXWxcDGk^CJr1rzMGRZp)1m;ZQ|ClfE2Teh9DKn%}e4}BQf?=w6Y1Okdo z-7?Fy^NP8kt5U{11{2iqO3L{pBr8UXYM%l@KfB`VTf^DQme|h&WKi4j)4@9AE|6mv zf*BI}BQ2{`l{@kr#+n8H`)6o9^~ru(iqh^!pEu^ZW{)VBF%NuhumTlzOG#o`yg7KmkvPP{(ex$G1T3=C>hIvH|mAC)OgdH2pn1xH+f9~eZN$L7DfGP|4k|2T#{(zsEPEk{%8inY4v=J=smi4L0{UDjw z5@6=%&4E`1P#WAyC-RrWwcCv)UObR?9up2JW$Kf67}H9G)@0M@sO#Ld0axehH+ROd zp;(2p5JDsTXhErKe}`gE8vuL5Pkcf(P*03i|Af>OqUdggIoItqVadPxsu50?k zY(cQoqrA!pA|g5Tyi(U;Uf#PHT2T)Fp8IIv9g!Z>yPR<>QgWebVIlsO*!Be8_`^1m z+`U)V9QBJMRhA-`t{>VN;G_wA`#;=CR_AAwi^do?x!?~}=K_BnA6Rb9ElBvSzF&!43E1C;vLyWr{9{tKog9^fY*m$JhP%(h z-mRw7WasS3P#oKAk)g$Lt?tIsROGUEOUvUpz`s53O-6rq@yjSWRaH*v8>-A=5{H#a zBdSrFR``*_C0VK~suwcFt=-Z^6cD9sT!rI%n*}V2t=~~*H_VD@BvDCz1Eu=1KZt9i z>b_+c-aBWa#@z)tFYgrwzm>N^`O))!#JaK1@753Zwg^vGwJ}tBqgKkS1U$h) z?PHAs&>Am5UjO!bOKFW(`k7!{2`cs~5wXc&%wZk1``CAi=~?;?o*Yr3?lx`=Wh*$m z{{Rl;P`!7X>x{UtLg0%}Y+QmOdAB+7fM<$6KvL!&Y!lXnsz72jN)VT;pqFi zf2*JA+e}cdc9w4yd5D-z2CK9XR~{I`(ktK_tkxAJ6Ck_}_Xu^HjC7>#Xk41AUN|}) zyx@~d9*HmeLVYQ+FoP#^btWj7>b*`%1m$||@T&05X05l}Cvx|`S8eHoV7O<=G4=A- zSkXt;2XfSxyyPQ?lFB~|nrk49bAdBL=?d3R>XK34(KqxrqyW&Fj^FZw5 zn9|9VN-JYGdlU!fz4endJ*S3p&a-^QhS}k zFYyASihlFW0qq%f5pZ&-jP0)tHB%edscymd!zLZ zL6=+=94n`ONtVpU#Ka(jNUTkZwI25NvUECNRwC5myRnHvLlaf{O5AdLeLbM`T=mv% zd8iG1Dj77jpU8%#-Y6gLoM$?VFXGna2=M09gIi_8kt;?aM;oeZ`k6#zD4@zj`Q%4n zKc_`rmc8x6qBcE{$jM0K&n^!&q&FJt>Eqa{46o&GiJ8f>ez{V;FZ9}I={%|ayvOa* z>$Zb8%3G}$a-cwk4B7vpOZO8|vw4G==BF7uZ|AkWeF5)+$+va|HH{6BJbOb~N(Ps| zi$g%(ikt~u(D<>Dxh1MBRDYFYBkG9JK#!HrWLL=*NT#Arabr)mR%_Z`86kiDT7C!! zp^D*1SBLy)9}VLmQ!I_8(>N2%5PcK%OKmO-3+Ln3-rq8SR45InGZr3LWnID&DEH#zdQCj*E?%gE{B( zR;|`=`Ys!$G?jSZQ9KLt0)oZEw0J@IUDU24TsWS;|^j zR`T;tYxZns^yqlx|8fB+IKtzsFl^NBe1SD4*X+?g;q|GtiDLl!+QUi^_FX=Uu7tet z2GM5{iJz=NOt$7pL#0=aI(3x3X0kNv3JNrKQ%dA#5=31iAqrtAmxy%dWP|@|ABAzP zx-f37YLy#nMs^ZUSFTI?!x<6r#~fWA2WQ%cI0($F?L{A?&r8pc^!9G7^D$!~q*zJ- zKm)l;{0zaZ$P6k3{bpqX*_=TLdrPT@{mF`5Tp146j)YqW*-c5&t4SDy=;YuzW;<`* z$y(9}Ph&5?W|1y>#)9ZHUr@rajUd2vR~_uqY;0cO`51LiIA5Mk(dpq?+ZwxJyxnCG ziXcd<^?yK2#?+h~)311bMU6y$a(0|1>|{Cb67u^|lxp02V5Ph*jP^9^y~Zbfmv3#q zWRl?Y-x_JQyZm6EioP~BZ`;Gi<~tEC=g>%Czg+$#^ZiiHewSw4?Sh9HbdZDo)J^uQ z8@)lhi&Z{?y;)iM4BWdDj;o)oqpx}$`Su!0M#QXeS=eDUE)+7j@=i8+kypfxxrQMc(CwiKyu=dTiO_A{9$;cbP~7Dt&5gY{dCPx3Q|0N z&riM}5gto+f0k37f6U}eo4HSMu>CLYf_D`-53j^Pb{Sh>-2m~f02ku+;n;E!7o|uUy-+TtA zac?6lMPg#^*DYT8ENs;3l^Ga#tir#yelwgKp%%DE)tQifIoh|9iW!O7z2!$_e1V^8 zs67O`n)Ctp#=|I>#WYwWP@ko4VJv$>EKXlXQ z>?y{=QOu06vCg}o^YTUcbAypmO3#@Y88#l@2Vu2F{vB7s=VFJ26_CW9QPui+=Jt_9 z*NP(|SSSVufD3l6F7d3Strn01t3oka^?>99{=~YbFKIe^mUh`nmFwp7w_bCM`0V{! zd65a49IYl3-_L#L1+cLqZ@H%l-+v)0RH%A*IX@vIFaG&Z$+*q~f~0A%iU-zMgX4ro zD%^(|4&pn4!YO>XrX;yp^}LBEDMxz7Qn&nhbz{En&V0}BQdgfW;97?{@@kPgiA#+H zMJxo9QGme5OvnsLrW#&DeMw)Y;7jKTdM~=@`l5ByuQp{nz+MD^>{(|&3ZyZWV!oc8 z9kK={ZbI~>KSBr>tS)U|w4l0L7ftFat^7vSe0qhr4H2qWt@Y5EA=zgc*P?zR^WMDM zP{~gG@b-Z#nPbCz91a`Z>U6)oM~VEwDi#OrLTagxOWCJ2A(pX6J(yQ0Q|MhDzQ2(& zs`R0jhh_KzUmKzu`1YLzSPQXFCo=8WN5|~9zYAwMH<~GEYU-NQc719T#cUfu?5k{6 z9>jO`C=>^OnJ0A7{81&}wUtH?R-Df0f8@>HXv35Fu|)Wm`_Q-2d-XimUR2?HzS;HJ z9u{}qm0-7!hUHh5ZDHjIX6CtyUK_HmvZk1Cx21}kI;vHH3l4u|r&=^i`{_KZ>PDub zj~JmH6w{WS=!>WNpHm4C5vJ^p46X#@-*Xi%?y|#I0zStm{7-`P`-8;@(A!YGb`1Zk zCkn=yiUPsT?nA+!zr{%ZlNn{IfCffitd`=>&!7GWR?yK++dV_;|K}PJsBg48USIel zPCC^x1%T?sBd5vgD%{$i%kBK9`lkO+4Q_T>Mhes@ zYog32^$TihYU=9Ey2}%Wf1XAaEv7UA!A^mPhpQ}GH=c2WoG^of?S+C%^&G(^DjfPv zPj8uSb^TutwNKwB`~lDRE0&dxPJVkXGB*nPmVWzqubJ~K503y(FK2OBn435z-;)4; z|09q^9UWm)av_rDzq)8dSDH>qNL~H&BidZKw=uexiI3SD_z4cUep)xm@k@3 zy)xH@ox|!pGyn4WW|uiG9v(MIw^GPvz`{%a%}&8YmyzGY_@ses!OBk{%aiGjr56nN zBy*mr_A=pY*dh=pbZ%T`6-c4DDa-Dd`M-DB?t^n~wda1S?H4k3W<|gK`)~7ph~PJH zaRv{P)3S@GntaJc`my5KgD+Fia;?(P7F#n|R{S0=S1vWyqcaHg zR@L$^9fInD321Uk6fHR|m&oNk$puPo@;|R*!|~6@)avqvYwXN@?n=v^j&9T`a?zuu z$vW!y?3}-ZbjRZOd8vbLR+eErQA~+Y=238?P!IR&mj)xHKVxf52>ywq)^-LX-?9Zk z^q2=z9c#X(FeiqCi3b8BI5xZ~G2alPtdebU1b=c3FvZ#-5b4~zn7j8U7WsZJ0Rsa2 zYe<6?iK~gh4g@Gt7JjzoNJW$1>(Yhe^xN7@a*sLwx0HV`XdRG&>i$s|{$5D!k53h_ zpC^{l|E?@#2&`RPKCS0QXN81F{1VU=;@8yvOeU^b3s~aJBw(I3ZzgFVbmHF-l!D~1PT&RWr4tx zrWAAxLW^@wlWV}s;ZOxW(qIaI+|0o)YopJPCrhMY!1!A`kPz%n;9Vb_A@5UzN6N4U z1ia60Oolot(3dLB+Mrtyr}p&AprGZUn=unYTCbD1fAl4CCj2VS35-cT?Q`}E6MMg4 zc4_ds5G?|}S!>ZT1&WSRKj=Sl-glesxLhX>3(m8>F9i>PiS!bPw28&`qa68BKzb|+ zwx6pz%lKy?GReW$Kb}@C9lwm6?5wr4MAHjCS_hZ41TyeA zF{-v&>WqkmJ~vzfmuQy{fo}zc;STaczx7P@=GWAMY0e)TYyq;mwjHki&wjb4-u_e) z6%wL6OF;3)>|8}jvDY_a!ZSQ-9T100+UvZ8;1aB90)DPNq&93SB&Z9smzI}r1@rPs z^OwEfp{q)4_Ef7+f$xqnuqgMAb_@2LGy@SV&AnVUB_;vP$ykW2h| zGa!RKJuI#dsYjBmTuu`TVQD9$ROH_H+2hQ9wzC8f%- ztdHRjB7V=<)?f_tNdq=}0F(p{w_s>b$!ONN zsWkI8U(dNtR+x^zcE*`ki{@WF02*Bz&y`11dt7VKTeg##C``vBc5zf_NlaI z0aIHK_nmX+*$wC_x#(|2S1$zPB&g!e6Q`^U# z*$&lkllvYtvc|g9f#s%6l;@YaEQ$VEbrhHq&Ud%mm_o2{p_wbkD#Ukw9vM3;9GUzvToSC!U2yfj!{G6|58jQl)9A)P~Ak2BP@a4+kXiiolcWL<3Fx zS>LHb0WxPKcHhY3{k@S!j%T&qhregPnHq#MgFK9Sx%Pewy;tThs&%&j-!0(S=N)W> zX?(g{eTiLLL)1b>I_14qH<~=ZHo(eC-_m+?&UWQ(&wP2S1t=rJJ-n9&K7ce=(R%|q zpkHD90p1I24+dtJ#Bs4;n>*9PUm{;U;J-Pq&aYmL@|`C<3XzAo+#&D-KDksM(*Jyh zOgS5*O!dwu+*(pavWXz+P(Sas*uJ$1=D<=;F-y16d{%^4augSeb9!I;?P|A4$Y1`u zKO+@l9G+pyXd=HyG_aRMSG~<=1wt;Z5`$z6u$A_s;9SN^gg#l)k#E`ca+|*bg+2e6 zL|sLtnT^~d4qT6MRW-h2}*qASWgQt~dhtLgQT=@qlqD(lmvB>=yP#Gzw$#amX`lC5;#WOd7{j-K% z8#2f)DY>VkI1wPKsHxm&*@f5jgk1=FmQf6L6U+cECwb1vTPo-ni=*skw!#6GWJF&p zNS)c$7-*`RA1!k zD=L&C38NI1r7zV%*sKi4%`cbYw$v_u^*UKl{yQZs2v5fVjM>mWQFvXwKg0B zwiZ`X1=F>be0BuR~4geS(K2?V?|t@i%Pc`u!g%-W7tBLVyaP-UO_1mz^&o zbAqXkUP`u&*DnF#!eBM3!EvlSr-%fEkcgKQ*}Vq*CA>?;ka=@c0vyiuE!8pQN_B3l zKzbpY$OHRbuviB7;HSQnIPVF)Yl2G@-L7ZBnqc03l<(Ptl#<~#c)T~- z=xVKssk_!wJe)9UaH9M2Fh#k8pZ{*Qu@H5bQFqoiVDhafMW%1u9zP;qRyl!x04-=JY7&NJ0O$Ug^D%q;%QPAGp#rM(r2{I5o8dqE4_F1B{pQ*=w zh8<6xwB?>4twH%4_UY@FE%2mamBaY&rAsYe%NG>r0D*XKZZwsKfiI z`kHH99FFNC(=;NPiG7_jQ6)~zR{HMC3^FYI52z{B^) z=QsmGFOEpI4-Mba6BBtjRjhBen6=>nUsHT4f?NKOT8f_qtmY<;W!SsG<_{>xKI>EG zODmx3fW}MM@0lszk}3~USWJ~j@S*e}5)zwwK0?TR>&%}J66Co4p<&cg&kUJY{Z=@N zKC#d08fDx$Zpu;jY-XM@bjV!>dF3@agQ|)qbiHQM%z~C?g71EgeYXy z&m`zq``pY|i)K|Il5_uDjyG;oT@@ekS=#v%Y*W}Zm6=)qD!DKAMZPJem|8@`@?nxB+*ENa- zUnG8MO{uG?gS*=C#eLNxm^vq&K?3yNG5-J)cd*Qa*`QA;#IiR)?q)5i2%a|ogm1%~ zvmdZH4yEu&_j`Zvq;mcz%Q7Jt(d?n|75evVhHMvt418^YsDJV)Bvu1NfcC@+X@6%L z{51Kg5fe6Vs_Osa*Pw_K!7}(eo&4X~f`9656Ouf@p7{H)IX!QLanA9w#RM2{&$ap_ z1UC?%2H(0As5tFBX#0!$$We@x2Y+!W#;jqWtYXAuWu3Ow#2megyg~8%#JWakuZ_Ha z|2~djP*7$3Y{iGCJs%IaMJ~HW`^pi=gS7n=!6_iLzl!t+3S`iO-l2{rZp@&dXnpIKueewyCu z>Lz+)9_;Pm;NjhnC%VOj8Fi)|H+qngmRCeKO1c;CPeRmQbb5=Syf(cu|#{B2p4dh3I|8`y+&l z3{sqv8b~>2L$ILQQ^J)%rW3cTXeR0QVt^KcU^}Y7>CLmU(+8{oO2raate1D84^- zkI(q`ngM_^20dNL|304C02qbuOs0GHzt{f%NA`d3kp-67)q8FwUmXC=4QT)ov0t6j zN2#45Z~)i?d`c}F6d6#iFj1(Tp9l^PCZZmo<=5j?$cK0W0>)Z>UAhnCNA_4hNK(E$aWw=-Cb&Fmo`i zze9aH7<#1DxmJP5`|EVe@?`x}?(-eJL?6IMDAB4vNnJ%&9+c7PBJ8USIy;l+{KlMc zx+6(FX-5EMLWV@KHkeP6V{a8& zbG_mRK(iz4<;ql3vK9Bg?@CNP_cv(P4dzPYb2Fd;=0 zah)siOZJr^c(}M9N_3g)zfIIKHk%&u8GKn_I`4*l8d#bwZhay8PPG6j_tFEU_-60T zq$)>m;ElnX1rL{B&@aeBB7_4FpD?M>^#F^Af1hM{@A4lgPRw#sR9O1y16hdGgvg2( zKEGiE7y_vDRy(Vb&Qp7gwvdGwwl!-TyNo2lk<{cK6v-8{UwQMO*7K_r%n3Ig6Oqa> z1)Of^A*Fl)M<|0I;8T5_+7`-c@H7DlbMyd&;_z^-O}J)Fgh?W&?xkt;aA7>KbxEZLAx)oAt?KTpZ@{@U}tCNh6(DvK+*fF3f>zQkoa+2 z1zhO8gh>F6=+VdsbV*zFcc{vq$X6c(*CwNzkAD(3M+NhVF#F3|M9{iD-qW!k$qfmJ zknk=-5dw3P&heHZk2AQSfSC;vPv%mW%v&)#7$zuFLr?=y;ri%rfdUTb#BjzbU_&_xM2vwy-mn>|&z`wB z-~~;h67!V9_P0K(T>?Wkwd{h6S2G5>JpP=MK#j=T^x4~(gUv^0?&r5^<*$yW)ZKo+ z=DP4p#5`9GyjF^Jxxowg?o=ZMC72t1s<9Q!E75Z#5H&?b0@kF;$13z?uKg=xH1q;E zgruo+Pmc;V1C;kk!<`SN)dE~9y}^#bjmC? z=@JD>Yyj(}7i=6pUR4r~J#I(3gTV&?P!z0s2_BGAM9q^j9$$<}Z4Mz4G%O2Ovg~&S zFf4U_lj=`MfY!)GhF~Ec>-7Kp_Aw&h1b&>&jUD%Lm((*NJX~%Sh4~#jYbAA8O}h30 zme3h?ice1g=!E!}@CeS$Th(icI;Y)}Xr_|fqJ1wjW?A{ePcQ0D^=Wo~B#eIYMQ z=hLN>744B_lzP8%ZzS0~@J3Nahimj{G*(IUdSHE8c9)6?ECAble((goFTZgU%`-{H z2fKSdrN(yJd%TwSKbAe8zTP{Fpe^Dq(pu^u+5$iXnD61vxnK?e^kZi@bN~-v6}PgP zs991o*eJhIyYu~|-ip-7;nQ>Hpp~~dbT&fwR-(5Fc=26h}vEe53?&0XMG z(m!mR9?Eu)k6>vg81qP-6y)@ZuAPf_7$Y};^O*1hTns@_qPCbkGIAT3AS*LL$ex%~ zXasjPv#dg?@VMNx^vnUD zo6RwH@7>Dq|UlV?nBuH18Cn1dX^hQR4v5<;dFL|S3pF^7ez1*;r98Zs!Fe=VUqlJc<1}40F}iG{o-ed#g`$&by+FsgJMQ_udZ1p z|DQ9+f))$95#Dxj5i1MbA91m3oe5GBVp1*IL%H4|JA310;F0n3yK#C(Q)gtGWIk93 zNYp98h9O2m^0xejzW&3_$ck$f3_~g?dFU=7&uHp6cPYf9akRZj#il*mMa-z%d5;%$ z6jBuGTc1B9PjY#~C^7x|moHRTh*2AF5$24!*vhXGa1H7SbfQzv+Fpva|16b6>82o{ znIYERS%x6jT&zE1*d$Wdc$9MU|j`vwkZ{NZ*b7`x=9b8p;ek#jSEj#f$^hIZOpk&7T4-4B?zS z5u#(rlPK!yqBI4fzw^x9=R3BeBH`oUY=T^h-W57D$gP03!yFvk#P~X0FN&@(eziU8 zh}Oybm3-9~Xv=E(b5H$Gj*oP5JJol9OKjywrukf;|6LLef8|IX|05bi(asW>@qkm) zElxzKUe6tSSA_`?woi?DKYTk9p$)sgT@#}q$o1ke}jG5Rdd)4@B}kt84l zOs+X6M$uyGnhDgc#}DFghBHTnjTYLUB%~RIsWf+n9B$~QrxjDT*hy{fINWr+>I#)5 zLxq!^Ac61S^Ua8!2{fAJ6~h+X3*JGAH}q^|WCK$+Pa$c?l*t z7V!`C-Is?TI%<4Z1rer^8)F1dk>4&0_`-BQrs|x6Hy3Bsr5RWkbhI-L_3;mKgcWg? zcsP{GSsEcvV-N@smI`ev8(|w%4!GLYgT(duc&z7fI~PAkf=Eqo9OEP(TehqxWm#JUcnU?Ez?gGqs9absNXNd_=^`J?e3qBDH+?D`Tk*YHCKUV-rt?~hCgyKKwYjkCYzqIA(#wBJXd`RjOzuD+TJl@C|N z^5HMy9+r1$yEJE)HWzpSA&zMm3d0Ks+xa@I_%v+ZDBrZ@+Qi1|t$~W6(wmsN`;+$Q zs!?zvqrQm3eV5GpKy|&;WVaUVANijX>IfBOplUWg>}?JHbWxx`YtuJwzMEXbEaT#` zwKxJ%mccVqey+kn0ApP!pnoa#S>!!}j+rWzeirfDZ*2%v?cdX=nHe!LaOdAP_1drq zy(%sur#d6j^UOGL57e#@1}xN)Ft}~AgPu*SUQNM()fhARPsM*;4LNB>E?Daw^++Ig znxl&KX109O+_K=$TFKORhARn&0tS3hq&x^O=nQcY{2gwt5{`v#?)P49wdtRCg|>kG zWri=2NG%}zaIHY;D^1j#CECTsYtCz!VaF=+BnrJpwt1qZ8D)X*E~_T+ioEfQ)m2Nz z$_&VQ3E6l@0E6Ytl}>0k*{9k74GnOdb3frNIfxHpH+6U-o}FJczpdpk#CY{$^Q<{^@le9e%{LEWOMVi|uW=M9Yw-rvYEBDipK zqEUCb!f`eDd#Q}=LWAFv)#sP@;W^XlMpbk0O3j@4(8-{ObFX_u8oj77)FhV3iq_jA~4&?#K<;-OYS?Ci!I`ojj88WRd_Mf=`@5tG&ezpUpc^ z3ah5tr{Le!cWI!N3xXq5r5dmH)C;5q)o`REE6;uN@K1MY+{;xqX8M^Lv7bv`ka+E~BlON_@yuXLJ;t~Lu|F?1TAW~d8ngZQ^ zXn6`9zI`c_&csf~%tL! zUt6oOc3Y+2x(%V49h??lShoG46I_bTjgp(;*N9_#pb(?7whr4E7w|S#*$J@0dXdJr z1Lh>(UAi`_h`X$k=qhN0O{BCj&@hTSb|KifPC5K2Bq7S;VQ~$21KoL&Ju%YUXrtvS zwNsmyIJFD$hnSBVz%egGoMXG)Dtpm$eX{;?wR1uj_(<2gEfM)EwA%*sC3NWB_(A}6 zZb?V9n2(dj4Iprc6a*f<=$-f6UB9`Rqbs$BjC1m*8Z^!T2WkN$y%@>6s6xGNZQ@vo zzcirkXpFS!?nqiAR9NVU;|(0OZU(^`OG89_-mM=k0@xn1psVeTF;C=(*u2A{6xjMj z6I{&17QrgYva;s;%j@3L)*UXZ21|W!#1z6Ksd3c}y{0I8WiJ`OXJc@zq;-hTRcxqt z_~uw2EE=}8B-oi{HmP;h1EVu%@evtrW2sXQy9^La?!nvF^H+AX!!oqblHFljEWM9A z$^@$u_c~93ywT(ZoF}lXC+*9LI(9eaIP=~7AIWqwUaeii*9b{Tj$myfWp>^P0j7#HnH` z%_4w$@-@a*uEhw;7icq8PkX2!xKj@NWvqw&;}2wuL72fE?A?Ff%|=yc8QiOt1FLMv z+tY+ipf0CyXMMc#rS5&I+BY{6*^O9-xz^@Xb;mB!W_=F`S0qARV~dO0i-iXo$v{t& z0Li(J=vfo}hd%}^9uI=&%)C_u%jS>a$LA|88~&-~Sj4xlejxDDC&9a|9-F39r+@Cc z^3>2%ViZi(`ycmA6B%6;u@3Vz-EP~F${WaTk-`@8QZl-5E zR(`EUO9^vh_{siPuTocZuhrL08G5zmk_E@KKPZTNCWe=pvZLIdl*R(|4f+gXsymiw zDnz{)G1(fn7f#ZhD5W>h{(5))gZe|D-v3S?$hckL9sN=%Q#J8u8K^E7(u^9J)_N>u z-5E|8S>VEsBtqT2cfO~h3x*13j#OyFOp`l)3r2^8(%q$pkN+a+?cz9)|5XB9#to+D z*+Zu!8S#v)fCCNDTL)Vao7r;Dndtw$W~2@!2Sc>c z|N0#KDMO{hH;1!K;_v^xkzky14k(5QI12qUcaT`wm*JbsL&6s?{CR3H#A{Gk8} zDPMG|X{ql&*973^D>{2s{(VHV|8GAwrF-;`Dl5&A$l#optYoZ5b0P|MO8+V4L?G`U zqq;Xm&VtrJKtPYX)?K?PB$yytOpK4)_jo2k>2C~I}dTh8!mLTw)F{+BlnOmm~7DMSl0ZLj~~y5#y`>G&!C+W9Gb z?8m5dXlP-f(}_GhLpLLRJV+~Q`ZnMj5lTfid>&z+>)=VqEOqJMw>!jz;fV&ssmem2 zwBMA2^UiyqR}YRbtH}EQ+@>#5wKGe1jQu-j+Lv(H4v{#aY&fn2Dv3W13e+)mzo+Kc z%7yop<6dN09wsvRr!ukOhJ^?TUq`4=sB!YX8_5qQ$g$N4o@V!q(tg2!27~$EdrYtw z0TK}^6MjrE+V_z8U1l)fJhMtMQ6-gKsD;VDZ$DQO3lYUe)lo^0l_YWSzm}v zXjN8ZFzJ6ja0@=r*B>YVztF=oNI+XU;bZmcKa=w|PPh&i4p*^c*$vFm+c@a|EJyqF zV6++fBd*%FA#%U}ETT`!NOzFSv=8b2r<&&l6aW4CRF&Po_c==m6aQvtR^0#9kKIoP ztLC0xnu#3hK$D@mO3yuM7RFs629EiEcYYZl&~|%6!2z6&ogi4^ng>g6P<%a-mE7)6 zE>6=5`zC4hmgwJO?vxI_cfOcj__^C+-&|FdSRV>-P3ReD3RbWC+bhgWOs2KBoQPYT z{77k}x*h+y=bs!pZNByG?d?#{k@ww3n|b&~~8VcY&B&+%vGpgSnagu>SvsE8**_3%DARJ96!{%0CiB?3N*i$@w0 zFlr9;m%hXTpr(K&S7O`;*^)-iLQY*fbXQdVu%hB#(DX_793xdq-)hHSWtk`GR;baZ zdcUl#Yxnh_o5Le81qTyQ)Halyo;&Pv1L;_2z`~C-?Rih;KQ|;r8T?o|f@n#ddqbAb zs0vwOEV(JVh$t6);q#-ncmJEo6vf-b5{W}|{vi$IBBlEp$pv+FO?N`(+aB8oO2V1e za0^xJxBl5%s}kEG@iBYh{}3uHUA}HB4-Xn@<#1v zFu|rV3u+j*uvC5hgc&RF7<0n?^A(EB$?;*)!w;9gdAz>jm&EP2ZczG`M>}bILSMRp z=4$3*7*mr{cSNG(#q9UD-*!bVFU2ljE7xdT;i~z_HEV_gENb#$gY93y%LP0^(Q9oF zZXDtQ(L$si)tv$mCqql_YGlYEw198C_dWY5-WPw&L~VghnAC01At8}72uNfF8Q!Ke zCkHbK)|@>kus&bZt7^l$Y;4I=L7|Z%qd9${V&v)5bdWH+AswdOTaGb!5>#ZTngg3Y zda^K+X(Ut`mAXBw{VZ0ub-~xbU~wWjO7z4P>N<4ym}n2QuvoO%*w`$7uH8W3RF6%S z>Zk8&H#QbvV*zO&8pwE-BQD`U%Z?VGAODBFuj-0pTcZU6!GeVV!QI{62_(2faCZyt z65QP_f#B}$u8q3{cZbH|R?a?q@AD7teYj)P=!fdAs@2PDe)F3%31BLTiHq}KBW5aZ za*K$oX7`~8j&fSeq#$CAOD*bE{+G_%;eInbC|&1x@Y--oaIT;||G*)lE#hmI<1*=S zrf1dp8)sm(F_}w?SJr+j!Dl$~qhPgb{qpSbzyL`3Da0q>#95ugy0N*og z>u-)J4km1@G!blPPrDCM;Kdv6;89-7iaPKQ;H}eU@v^K9^uzsJ5L>%U1bTCGlh^rR zC?Fb-r_u#9PQ!L3#XhTQy@W#W?bBnFb~HZ!acvV6`mg(ft9mP%`vKRYe4OLP!X9RH zn@%IM8DiC*m6eRbw2#S#ObTq~b1cNv)E>7{dMzqmXsCV&~Mrvy6-kK;p?(a-ny9fz@5J}Mv%Tb%~I=I`(!QCS=6eMr|@i+E&uo%K35F|+x6 zl&#z@-mQgB9?WJ;tRW7YMRyq3_$-Zo>&mGl*?O_od-_2|0)@~jVUCQ016eHqeIQ!( zVD(DRw&5^bD3K^8CT0TI$>yiUJJ_%CBj_rzA(oJ@xLrrD+AY3KD3ut&)~e1pR0I3Ci?L&66y_ONO-my_4ToO-0ny`>OE$eY$SQq!cxSqG`Q zcitM9mIx+L5Ar#m-=dqA=Iz6-_s4!nGp!L#!PYPDQ@S@-opP4M*| zB+XWvU9WN)cS)4wJt7~B!hcOfPz1>CJY%!A=XZJCuD&i=R(RdEaGu|HygpJU9t8fj zY(CbnJ6kAoE2(!n%omp!`0yd4pV{WIA$@U#)Mj0c)&Oc@-g2o`MbRE%h)`~K`*@+* zZOa{$yb>}}L4x)$*%O>ZN3|$@I~$#S+52ldR?QWeex>@`bJEyRs(wLv8xXw?5gJuz z3x3bQZaLZPaR%wBUT3w@-qw=YcC*i4Xg=w3URohQPc_JiQK}lgH^$?9E#FgXlG$=2 zT{cf7KurDWc>3bCn9fn!m9up`mcVAS^j%)59tQnAdXs-+Zxwgrx93A8$~Q>bx3e`W zSs7ZiZ4S{CGMf1_MWUnVYfix^C+qdb)2Ers85ancR3bX`}6NP*tScZe&`ND-ob`I`%H(7JNGx;$St{#*1`=S^Cj zl}w4blE7~U5B5C?*%6L|vSsJO`5y|No_wd61JMNfLF*>gTn>s;av@1y3y2TU+SM z`7RbqIV!+=y}{Nd;^h?pgIBqL%1cM<3gkRYFB!Muw=(F{cY%~%&cs}a6_DQG5VgQ8 zKQs5DyKk9y=UTgxi%Fl%9w7_HLdsedyXE7Y%*J^K{rF9>7<4hVAK)`kYb)paHB#9B z?fVg}!^4R;0mY+P{Ch#T<%Fyz?_RgW-L1Pbu;?`+yuJZ2e;!z#oaQr8l6|^2xF$OX zxd>RDeXyMd_wKK-fRu=XL^U08((0|J5#5*K1Ra_?mi3kpruNpdZ+GohQ>7SZQNivE zDXb?mLwI{1?nH1{?CAgpsJyh05dyEbm-M_I4+%U@N5q8MZLFEv_tagY>w{43Vl6kv z^JU2raPDn(=bTOkTdOCuLvcWqYV z4%oFE9mo87%FPy=+uHc@oVC};sJ`;q?v9^mwibMRT8B1zhv(k3DvtokLE?6{T4UIi zfgNayaNL&UD|xPr%VS|RmewaecNbu`O?PHn!=kA5YO))Tc=HHul$*L;b^N5lStsy* z&U9#r69o}AV&_YZ;bgQp>UY#&@Wb{Xo495}vG4CZ{^2Ku_ zluNyAdOnVrFmW2vr2gEuPTy(9XFhdXBz*k{xV;>!z*`QFj{|h!ckmwos*>kJ?QL>W zhxf>hhDBSDT11C{K-U$LPoK*)sL2jR*$s$Rs_4VDaKZYMQ9TmT zobREJX68yG)eC?OM~xBy({@q}ZQOqSiBSne%&V*DuIfm*$1$_a)7Jv|@BJTd^-Fl& zw(yFLwUYNUNxOgr9+3!U^R*=;&8p3NEWX2okJqw%9@s=7ewl}JEVeXcBLDvJ zNl=*PWQrFB&)H2`CK_?zrLBb-=47!BY1yc3$w_{DzbSuA(`L2$Vlx4uk9ku0Xs$#s zxsv!~q4}V*y^THBl7g~eLZ$5iDV25qctjW}y6r*y>WYZ*_QRY6^l5Al>_vgNIlH7~ z(6?|8$@m*fmSfUO*ND_}ZRHQijgy)URtrtmh(UV`jSIcKM%P*+94Adm*XBrAgR8Zm zvhA9^_Or6B*w8FOvS5_cJz@Sz#>dW=V?(SY!|%!nmvWr8vv0*7tCvPAJ0n>YR#k8NbzDJ0*E9 zJY_pEn6j{-R3KZf+cv6`X{_VDbMqwHlhf+42bRHA)z=g7G$ZU7yTckYa64;vZOeo-fm1+XI?C?UAFwQbtLL;%<}cE>3$eEA?C2?YwY z*{Pp=Ap;V)yd-x(g`I;dQKQ3V@qDM*>pCI`oBjW&8zcf zS*kL0m3}TXcLUYF)!iQKf~hIUl^AKbiqdxfVS_dHxb(j3Xr2zA=WrNkUDCTs;>GzW z?P6<>XdWZdo1+SWdOG#xjvKSaja>`StH=QDo2j|Z{<+oZv6oUOU8 zS)<1Opi)7N!(!#bBQ|eGNped{X3S(j%lQ3+!^mli{>df7>Z8F*UN*9}yS_v%2x;r9 zXY9p0?M9;)I>hd!M5&hR{jpQe%o+D`%Y#{$7__K$n>8T<3CmmOlj6M0X7K5(Ez_aJ z?}5Gg3~G8ZZ7+P9f*e9W&TY`3-}n;$T?%#Wxw8J$AB{6M;4yx6@*ly z@yF<**1Ex;-7kkLWy#DNjM`kdeOB5cyVW|Xk@!3gCkQq8gum^IDWFJDp&%nd0Hky; zm}<#pJR`1R1fr$eKXwVkz3c?p~RCiRDv(hIZ1{@J`9&KoP( z;zqUR>w);3mg2yGO?_^$<*XPu8b3t|ZdGSm_L6FjK#GGP{1~;% z41mb4N?bBUX2_+i_ewvg>-dt!vwb-gxL&Ho(Lon}vXq<6Y&My$jXy=XTh&yPkK^Uv zHM-<>jbpR1a-gFFX(l$>quhf80ckO7%B?9LY*A{+&rZ)B)&O5jEMv0G$ZS6({~kxKz~!e(qFnV@oMrO zvsX=PLPl~|pg0_@qa73Q)|lQMCzpRe6Th`2S`-L)&1~Mjoo)P?&N7tfHA!3NE&9^m z9D1?r?FZ|6(9K;6XerCx*U2AJ6$81N^5Rg~*dsD1o;ATM7{hUu0g1Htc^dv#4>oN# zmJ}KA5yD9vLClFo6j(Y01iAM59VZ@@uLs;W_AQw4)GuS@uf(jhHzR zk8BY^MYwAfqOiC44&hFfQ*_B(d$Wc5H=W=rOSbnrYi~Y0YhS-}#p9{MYE~7}PPJF_ zQ~&0IZoPdtTUkJbnxbZ~3A5gtBUn{lrrp##D>QiJQ*0JK@bO&Q;n7Fb!#?Brjz(W% zW#QN4v=rv(^(9Z8M%Bq_2JtuUQba;|-EZ9OZ&jM$LWJO;Sy(wYl|GApmuYrB>AuH$ zS^rh9FSW!C1fnTlo&djfk5e}RyE=eb4C1>ooucHm4Dt!BPiZ@gRl-lh0r-zz2YFaS zer#@+9b84Jtk{fAJVpj$O_#duWnNz4$DlRqUEL1hv?_Rg5!(r>NRbG)f-A|Dkz*oH z>sPlPgx$0F9W4A~cB}DkxAflV?O_I6$r(G}gZ!~d0fnufmcrg$yOEY&!f+$=K>ctI$Co9HEcuuPcx=S5L^mey-y5NInV} zs`~EpQ(Gy`;@uc5*z+5#qxImH-yk@^o$Z{9+P#hD)W)7yTC8$DB)~8|j*Bo>TIXi* ztE}b9yKIf&$gSL4oWXG1(fBQ^(~bcC;8??Q=3(2}%THWdJV$5i>5(YR6xIe2s>XEN zgw?F0D~pokz!HHWtOpQaNyQ^NJ?mE(9D%EpHT~1QK?b7~cauQUlW1p6 zJQ2aB-e5$0g4XgW`X=y31AVi7ci3BpN9d21tn1a)7f-5Uw{7QEMIWU`?h*_^oq_B| zgN#48YfknBJ=Lbwc~pC;*K-zo|1z12T!o*5tc|jLv0lGYkv3>NZT%cXkozNBEmW~p&-bEk;}a#{;$T9~ z?_rC(?ranXezw~d{o`(*$62*hbuMB+)-8>#}DTBkP zx^fK{jMZ=|e|;EK=L5}C_0EG*9u(Fgd*D`f6}DrULSqV}v)^P9xWcPxqqyjns5Zd` zP(rfIq=dp49Bw@tY=|UNiB7+Xinv|(A*AF=7hHEg0`h`IQ9N6f8_7_^hXA_?rMzO% zVgugRXK<*_VWm<*RQa43$r`f<9zRa=()p|Uf{e^v&0|G|L)R?7UsNpir3P=J!f`U%flQcH#i?Immb`BjW?K&r=obb!0J9lh$ zrG00(#tMq^9ixE$ZY;txxt_iz?Ye51#E0#>B`TLpIdUxfkwJ)IG?0l_0zW~<)9r^> zmuonI3d4#A8*)UkcNs>DWU<6b1|um#%&#r|cg5q`<?2(5r^feWsnql1d_;Jmf~X z3Di|_*|bPle$eOCr79ye72V)bs$i z96`d0_i_x095DIamX1vCgy(coZa-LtXTCEEA{efDDR4o_e!i)`&v>G)l@5sMYtctG zkgFC_jA=NYt|g}^*+axG8y`}>B9~pYB9_~OYiD(|nKPE og2xzS`l)8Gq2X$52k~mmN*==tKF(Px+s-Uk*nU|$N&}XybLbHMJT!yxVHb2|LNbA; zn+hMmQS#MzX5--om%!(#V0)#!`7mZjqE=|z@@COAdm|=8G5gIZePPdRi3|lt@*x0k zv3cvM>A3BTrs@Ji>J`vYd|h=}U#{#T9|@J(V)>A&L@p0;9NFybFF=G_`7Nn|M*VH* z+H)`uePBq?_^L+}SXDSs)z?A;J=xPqeUQtftTpH@%-%ujuW21GlE$*kos(iP6L?3W0j7+4+=8)?uN})TaOhA?FGb@Yf1nXllDt-@BGw)uu&- zSt&F1@%H#gf22XV-QDQ}x+CmHkP}GV8lu{`*`$o&H~n?HbNQ0)W2Wul%2F7Mx_0wf zr~>jEtJzd3dFm)`e$%fnH7vppw<=U;p9%%hLd2i|Uq{P@iTDIIT3XtLExDOVl>wA||{K%-E>H{M1NYtPAY+hxSd*E*yZ!{2lgN*py&z9_JHtac)=ItwhFkxkFjc6 zW6>CFIn$xvx5#BQ=X)rh9ZRHurN8U_{(MEY`TLm{7U2?LXx?VOK1n#BvIaBO{FKfC zctUN_I*85!`ig>LtY1<_1-8u-DtOqYE3-!6Y5%m|B;hO5PBC^+bYu|g=i$^4^fz&RhS$E=l6PrQTJ4-=PkU|>GEfXN- zHC{R4jNgpyj}GhlPl5*Pyb!Gw>xkqKeoJ3z5y4m?RM-z*8`f>T87uLTW z_6WQ{8AFt-C<3(si2*Cbd%boMWXN?-HXCYV<@lO*qvSL&@M~qJ>l)NfhzOUNL9;Qz zSFG^J3I*Cc=SO7i zrF@T4v_>YIZ&D4XB|eGa0JJ3ZLoRzWy9@G^@k+2*O`F}16V^0k15io_pZUC+`ox0d z>fAo%2$jM){UPRge^HNy5vCD*xrpXUt$I)J;k|xLcfQ1#-#)i<<0wV4*9vJcGF_Sk z(e@|CsTVK)o2{R@{ndV&tUOEg*zVrXX&5OVWhhK5RN=8+nB{dGMZPq(acLth6JI7k zAUo@o$*8~W#HMYcYAn^#c?6$wNZ_)fE?11>tB;rKu^8ipv&N#ed=y*UJU**5j6JpK zI%XV+lO7P@hmtYzynu|Y`NT0vW!2d#10!PhgVS;Q7tI@;F?AA4Jy(G|EI5KySZ2z2SnpcZ(gj8%?2gO^-3PcwYc;dT=!xFnLez%r$oy zK4l1r;S0Pzt(Q0LrYYKDEcfP}(rIQsZBxFE$%O$*P(%nr885kn&oMm@z}!BY$-4?_ zpmMtYL1R>|?T2u@W+~(^9(jVq{A%}fl#f)RElSO-hg!1SXf9R17g2(Vo!DTzYG1oQ zsymNmI#6e`9DBwp`?A{9R{G)qCT6nm=D#8Jd!B!OGlTaE zakM#wK3>UUpsycQIuezdC+jq4E3}Qr@8lXJ*hyORHM~CMeObv0puxoH@ZRvAfK~MM z^Iq$U(y=A(BO3~V=Em+Qt{~#IL#UA-mmksfNe0^3kt#hY)!ij+#nm)8PPbe0+XEKi zIb-s`8d)2sk_gV}$Xb_!qHuwd&Sjt_0BHz74{{n%)EI?6Hs$VG)R@R{Gur}&l}Sge z7I9oV#v?1ytVr1~L45Ir55g_4$A9&5$nCGpqX_ErIE znZx+R*QqMQ;CbEbr|apCY!>6Nrm&VHTkk7c=XL_OZ3WZkFj`^_m`Q(SdM>3Dk2t`R z$|bJNe6xq@?WlQ8zL!Nd({BIv_R{B}>Jln7WT768FPP%8lSuS%Gj)JR3K>jP`u3Xv zPGn8xK2L0k*)K9OUy9K2@gvOu__Q*cR8EWXD_EqS@gAvj?ZU&heV_IhvMy+5=55Sg zeVpgsLae`8LN&@s2`soh6g@hsDAnaKq2idYed&QJTC!}Ni9F9)c`WTt8;`rTLS*mE z5cwz8m2H_#TTrNxO*Bb|3jruJn-BaGB{vC}?vbGcpAWVG7Q?T*h7K=)d%9UE=mB^w zyUV07wQF+9eTTZ@0pz&>FbL4fFGYADa2%afl1Donyms3s+T(E3$)MAV){LY>+{}Kt z9uTq$Zfa_BS#^^0Jw(j_@=I~}-QStK^G|s-DDu9Z9I(@pktx&s7Fu1lQHs(yebN#W z404+-fkuyFgbA`|0h6^`wrS``80A*ZjEk20MtwuZvA2D}Ws&cTb|SE(wS8r!-cpH> zTcGd??Z_!jVs|a}Qu$3|-+P*;(SM(?l(Cp}tK>*9(m=T7Ql!*eixE>pyl2u;%R0YL zBA*9uCswWC4b5n3Y+Y3W!`l~v)>F}H4=^yuZD;^mB$3+~AaA54S6Dnnl@T9Jn%f;V0Z|^p63*=6KkMEbW(xa z0(Xq$u&l4|yUpsR=m1)757OAMp#!t`ggmr`V$fnj&?HGhXKQVvNz5D6yOeA`fK?>Y zqb$I-9Ah5tB!){w07P&=qbnb|OG!Qx5pdf_jPAn2_eo4NuNl&;IL){DEhB65!oPM0 znMZQ>g}WhRbK8s!eI5)!gpN@gJN zI3Hl3<93+z8E{a^u#bj8%Q>EDAM5zj7S}XM0!904Z(dlcQHM~qpNB~704QVGwx>q2 zm5z>eGBuIL?)ZP66J;@c6p7|(y&Cs`W4loK$&=`EIyVsZM3nQLU%oyaeTYh_q&{?@ zx6YvJxZ)TsdDi@X)|@_Lq}#2RqSsHilPe2zJDXw_k#|IOl}4wU#8w%$&xwsy)nGlof#4;G+)N8qRe>^nWCkC9-Ff zVZo6i#c={W0gY@rGb!hZ&MclnB*v#=LNNf-CVNziJYejii9v940ZN7Z9@U|eb@ zlOHp~#r)vJ*m8zKKc7ZFp7{_d&gwS_${x!9=Whx@7r*oKdAaJNVV#%EHVZ%ya1>Pb zbj;+%H8dG2ElkZ@w0|!x79~1{-UM{*eqv#$IH6kn@6Yw2BdA!Qin`xkoN`&g z7}VZM-!%OD1Tg}?LaJ!0Fh@Uib#=JFg@r->H%ymVo@MHPm$j1;k~Azr=<;h(D`P+= zOx1Y=T1eGSo8k@S|NQM}A*)h}57>Un<~1n1e(Lh_ots;mTb6SPgaZ*c(h=dV9b zga=O)jQVMG6z-Q~y1@c`Z;dIUFe(N{VoHj#`3JU`e+w2s47rV(17{u}R6KeRm?>m* z;nvjj#7WG|%)n62^3{y)?|OqogCm53Bi=^6VxB5wp~Q@WLgO55UE!mek^B-GrCKX# zEP#R|^-lrl;E4QCMWkUAl!wr0-r+<+#hUNzW(|Wt1Nj;SHu#HZJ)7LKWdAf3O*q6R zEPMc}2r_XPjpahzx9)iX7Z%E3^ZtQ>f|+tJo@lkxCWkD%C3hfX`QIiNav~Q}M(_?c zF~xQ=+m_#99`iA^FNKd18=8_uvj_d z@>v|$Ac+~Tn#?(gG4y0&;+>`6;0IwP`IzjKgn%-P$ibFjK^TpHp6LsBMFQc{x%H++~wFCEtc9Tkgj#pj4+e1>`9%QJD_w*D0 zjKlZvy}#yUKHCy5bJ4%Y^0`Qh?NbPaXGcdjn8qP@K6N!&f0_QOi4aZD5?};eO6-Y5 z|KIrrO_HceX&?akgw*jOj~;h&^znL^-VK*CQMDnbTq1|KAxFJf_wT;#2LVTRdqnD_ zPD6xTv;s|VgOtjQeBX!Y-@P-F{O?`mSK;hd?3^?ecghRmAP;R+ zPETgd{&^w^Di~OOdR{QizyEtqzUj}3l-wYW{_l!ZNr{RMW}Sg9A@;kwy91x_{>qnbzK#O2>c7)K8JsMs z822vMnqR4wElmKimJ$bh*@UVFI4c4czzIS^%W2l>;sXBd(jQ-gtMqKl4<7)8NILCW zy~czd5TLz@_IdIFkW>x~91dwwnM=3n{83T~Yf_`!XlLr@cn89MGKBoO#NN8(&DAhg-)M=BJR1YG9Bc(z4>0`WK>SG(UJD3t*k zC!A}pwAnz|7z4f+^)w;v&r~YJP9)Dtb2z7vOVIN2;7(1gGjd;HSuRZRasV*-zBX3k zl(I=D$s0qH?26_hLt>$$ua>BFz##4hz>CR{+Cdo(=or%gY?zB;6>KZI`BH3SNr0r{ z$Ii^bk@N~MOj%6%`Gfox!|ktpuOFX8-ERg<;T}1_m!@u!0}g%F-)_I7^lG zKq!1njShR+uSdG}ikF|L3omI%Q6Fz`2fL#P`0WA5VnCAOq5#@J-h~H`hLUo=SaaTC zW-5t3-4yh)WGM-7s2@gjz2|{M89)Q$oI3^{P?l!n=5G4&gDGaX_^%!={|Wx-;^pz= zu<}Qx-B!6a<<;esqLOAoUe%e^oKw%l&lY3?&y5;T@jgqB#gd9BQGZ{b)7cuL9WF;H zuk$D=Dv&h>B$;o$fdvO*$jI~y`t|mTDe1qyw3V}?;PZ4Y-^3yOH4sVSh0XwJU4nh) z_o?$G)A{_b2L%NTcSW2fy02IbYdqz)x(+v87hCIT>_-bj@v%e2cmOmswekabGK9{f zC#Vh(Yh*q;o3|C>{w}YU_~rNUSGiGp6}PIRNJJ}@)UsXbKSOplqH_UpGup|k$!d{s z&1+sUIN_Mv^RCM0#o=bI%w+b<#YIoEbD0S=!;pP(5KNV=ZZzg72m~5l@&sFL{1k=7 zWiwl7IO_yhPMhnqzGpXpJ3GopCO}pkGhl2N*r^OJT?1afC#&fvy#P^p8Q1o@SI%m! z^`HhMV6VzVe!KqqJTub_usnS(`mkg(Y3?KAl9D*|IxCU!c~C9z(a}YJj{vX?pfk{9 zG}PTKa>hSFThZDK2%rLqXOBm^uOIOvrHt2!MS~2{Bcr1YalHW9NkFrj=i6pu1$PXE z$@%Y}fU>PgbRy{fzM!VY1(50qs~XR^fc9)`2iTKLo_EH8xO1^?2l4mdmj-}wgX#tXg6ExNn^VXHe0L8= z6?cb`_9dd={E1UY4s{|tK~XefGIj#>@0%Hw0d zyB!~ybmx)5*xmOVbUIE3Ik^=$Ob|a8b7j&WsHp)%CbB}~lX`1i6C-Ku<{5&QQ61;x zq%6BHI#w!jn(ZsxS{>Kduc=&<31me8 z?fOwmBGZNqRYe$Rk}7JkLI?&{2mHerl6blg#t);yCoF@oImSyW>Jx}n_V-EI-WUzMXC_6rL54GwdrE)7WnTup+Nt z9x3zk-81e-5n-n5t$A#B&--y1&lajy1(NO~2HWpTM#&75*7#J!nU*-ie@ip48rD|qr7F5#Vw|95Za=p&l z*0)Z%C?R^5Z957~K>~B+WCAXcw?BkM0qU_6?aR`l>Ys=^aCE)z+?ueMs{evMze@m9 zZA8+a2)gz7dEITO5I_;;0xm@D_cEwrJ2h zJ~F;7 zlK3{u^%yd9mHN5m+AVwA+j^?oht~E}AlYtCPxI30wEDs?7x=w*zn&z-RjBiMZO@a!V!>qaqNc8k zkBgJG1*R{4x2i{&Mu0~no5BS9#04ZDd|wjqe)`BEg*hUvS8X!(NzM%pBD1kH4$%4` zE|SaOn)~&IRf)@bNm+6fw?k2h%b?fn=|k#Mcc4RAy)=r}Q|P(DQ2vUe*(oErsx2Uv zvwOkhB%)rI+IcYT7BJL*qDz6bi({y&-QQ0s z*Y*Ikt%C|i(mDkGdYup=s>-FzwT7!uH8&S+aW#UdC@<4E5pcDeK3#4 zuT?sPGkJ2|ukZ9;p1NyHi$o(v+6)J99OSOvBCTa`v%VS}!Ec@&c%15!E<4_Z&+G7d z6CbQ?ajxCv;@a3PG0Kd+t+SN-$n-m$TuN5wJnNdBTyUp=)A+s#=a{^G*DTIMxDNwc zM=GL%;cc8AY~lXH^P>kid6Byhaxx;YC_0j_9{ED1_j z4Jc*+3|E1M?X8)j?4L>!QLPyz>Im5$ry$kfo5R_b(`MZJc<8N)_6NBr`YHyQba3Bo z`z;j-izel`!B*M`F zjJZ@iw!aNAkTsme!aJgK)n&8V(J`Qg!x)IA`#*yzr1b`z=MHD7KD}m{bTn;6@qFs1(tJw_AoM~T$`()GF_)EW15y_3m&C_19w z&|G(|1k?$J+HcHhG@CxERp{6>J#3f(0vtZNwPtGotKozk5(J2N8TKnrCn0M~knQ%D zBNzE=ALd4{aVGM*v=2^WYLYGmN;PtL-|v_@Z;=>oaGFze+Cn)ulx_4!q;~exz7kN# zdG`UA`Qif-)yXn~xf0P;X#wvhoFbEdF$JhB@7-Svobk>QW2A9d>wAASF`}!U`Dvb# z+Vlc!16*%c8wBsS{iUjnkZAt1CyDFyGaPtXx2#y?hvDV0jViMlMwg>|02cep-~Rhe zcMIx!OI2WE1T@cJ*^GsDpVwZ&XLAJ;3QS>e>0Y_^6aQX81TCT8_;WhLTf9!>h_ydr7_^sYx8q>HZi}y6K z(F2EB8`F23n5KiQRz>q_yWq2UzrpKtJ(SJyY-PlQZC=S=s}eElo#)Vdqs`YfFHT}2 z@b1T}bGYruveO%E)|azz#$(=Yv${P2`g+)(L~R`5^Ji3xa#AIl1}a@lZr9sGGa3XK zepx*b=GG*+TvgO*@Ol)rtpuqU9_#r})1SgC>6R$n@+(nCc9sPnQ>pQm3s9=rq2Ur& z$V%x$LN3Rrwh)K0O0Faoi}^r6#Dp>X6&?qdW{+7U{;I z#dmdwF-N~VTtJNPKFk7HM_SA77GE(;&rVdT`=zd1WP2l9-LT;Y+KJ-sPKI5;3+ zCChPm^-fmRGT2*{LftYo=MjeCfn_02YPO!t;c_gYgYb$}Ox^iewv@I37#&PJm!ku| z!X=>zIDD=PcNg;-1&@uYsz3jOkq?Ft;hFxZ~HE3 z99~bDPusDOuA=>bT)${MZA0n(9Hv-r_96|r)uJcua+kgsuo*Uv)z$N<(AGGzTvih? zc_7B?z2T^|4PnLZ{=jR{6J!pHf+@wk0vUNlNFhESWc8(>prD3$t=62C5AMP*Jo1K1 z`pzeld+RwN{lR2;o+l_u#4!uLm^`P?CmUP>I83}M-H*H2T*kGkvQPD z;m`0SL1lrY`yPOCCEie_AhnG^jp9*nZ3%Whzzpv2dA=K%7rcNL=Y6DtE~*Tben~@K zK*8bY#hS3#15Gm{j`9 zYtEBNU&m$Dn5qb?-Qe}O%kQxtmKIP` zxjLF(19XnFSkiD-)-vzIyo7Db+=;30n^7UL>Rx)53Kgu+J5!N zrmRE=m8#G-iNeYWe5zhRMYOUFiM$FH zd@B0JEJmUyx76gotJwyNjJML_!pU;&(%~^TFC?6Z_2C2K@Vrf(rL8Kh^iQKH=Q2Z# z)5hO13gW&m5=8+eK!u!#UZ6vP)aX_Y4l{ce(uK|aB%p-W=0cnIy(X<#xmY|y-m5hLH0=s zy4qnhD-w?garO)G8_)ydk3M(j`f9zhpid<@kR97m1e`V^fF>aKb?Tt*INiokHo`Xk z^HFEZH=A`P*IHAqmFHWkN6PU7bO`+5d!>=%9`%G?xeVD_! z#o%5|TeMXJTa^YX*~R(rn!DXHO;a)As!^&>wC@9@!xo56kO0?-+7~>40eFKNqL0%T zAu~>fY*aFz%(HD`_x7UMDeHLA3UXQLEwI7vg06dq!`AC0nl#D?mi8Op^N+uXs3t4n zuqYLU1%35#CFJqfG&JCgr2{P7wrA+sTlT&^z{_4QyslDs{Qc=#ec^^vo0@xWn zA4jK4g=ipa?K%Kv5+?;$i@4y^Rao>=el)p7$e`B#A2>K~I;L+}fj+>(_N9;eM<5Y~ zsWAcfQ7y*!jI*kI=wItYB{Uaj-+Rxta1}vC_m8u67?EsB9S+j~p>27N2K#=cKVc+#nOxf?1KP^~!Z{7@I$h+jn$|b+ zmQl~VSvvL$ z3(s5qvGV7Kes8#60t}`k1gvr zp!=gl+Ol^c(K?z|SbD(^;dyJYW%@IokLjDeLu4y%TNeW!wHFz~mO8y(Ur#oIx!Y5F zGP4`6WLvbSU!fUJW|>wEPM7B$(y%KMX{FFYwtXnBKF7jT@h%O9$z-wY7=i<-i6@{x z-Hmawa`A98F#ObR0Rw`BlrIS%LIIrjN;G>~(x!34r2VF%4g9uPAFmo2s=k3_89DfX z*c8AICP9L)MpwI|Pft)}h-e(|ww>atvp1lcdQ$kh%J^ zt0kg!dV-@_LY?xZ?+1DI18o+cx8L#0>I)0M>`Mq{vzk&UaHa5lv6u#6k`YEi+o@a6 zt1-vNuk6OK9|yyR)lKKiPh5@cIXkX0)4AhumT4B5$5>I9ils6^-(eh>Xp~M)TL@DG zo}~!hD~3C)Iky0GU3bILsc_6O<Fc6B&%ZnnjYF$*n(v7Lcr1YpyHe2AHjD?HX-7 z)@jDVPc)=k7wI-E!KAvpX7btKlOcJT??*Oe8b9gog>1S5a?qc&bj5085FYDu?##4^ zW$>p6;}Vv?X4l*-L~|TeSD|^|9q9r<5?l>Awq4p>iRoA7!SP6!+#0u!NddV+y<_8e zRKY|&siT7IU{CHle z-UIiv$t*?h2C9|#OCRbpLyZ2Oq1inZgNxmfp)0jS~SGL>ld8O&kRBd};r zk5f{NT>ctbyzs@Se&4}$64vRvT$b&3n^!4~8EyLVKtEFxy!$OUJ_t{Chilt94Bzc6 zxdSu79e+u$fBAar14?Cek_M@NDI*rqbSh%a;JK**A^?GgEo~O>838Q%%wgJCfG+VT z&3^&`^3dn!Emv|r?$?B`oy#EV$_2RH<9OUG&r6iE_M;nZ@GZ5H_rox)2eMm6S1DHa zw<$!%yEu1nS_uHTHoH`)h$M`DzY&NgCzVh8@^aC33)BiXsiODvv>dNCvHzxgYT{7p z<_Wjj(mDA#F55L(g@7mX?NplX_aj#l`w%?bqF#oWUlZ-I8c{V}QM1J=xE$D^Y|528 zh%cxkRSxsza6Dh6EXB(uv@T*K2cnY$7F*^x16{@KMWj7Ft!!)vXp*cg1rx{eh9T7t z*`Ad3f~cvT5D*dV;GI_IQ1oZLgIJU8$jnw6%|%2=3eHev`YxjLI22(1#N|TNp^D~p zR2VGfBEl_K8=0Gm;?dv71O>}w5aoZo`uRHkO|kb^FfNIPH2?_7riZ-lcIj9z7AmCI zvgH`bZtt>ej=zEZe$rLwSM_@Hf*#C02;*fk^dJR9ely zL2G4PAhtcbt7<{~*T&}0r@R1kw@vu{^X>Yswm`TNof4>EuxOatEJf^`gB=YCW zP4cD@9OeIpSV_V=7ttw6EW38vmL%V5^qlz;w*u)qcPfq?q? zH-iKzMgnT#pMaqI-CwgD@JW-=ncBt^T`L_!`{(hgr%5>fg!7?Aejg9Tn}SbJe&ru> z!TTcnj$ek+#tM~k)gS&9Tl>Qo2 zyNN<1wh5TkQY$1f)RBIMlIp+p7Ay5z zxgWv27s6KXM<<5f&Se>epL#prZ2I3#BSa02%D-gE5U4OYewn8jJW8KGb6@oR@2>&6 zDkEgQ4Cr`0J7!Iw{+q@$fzPoh9UVX|Di z5qL!waPt=PnSJZ&uSy{s^% z`)nK>b9I&&|5h2m7bk`!U5kiu$ka)NHV_sT$n5VMm@lOk6jN*cpL_Y8llWSRwlY(a zZfjAhQTVw6azD-gXMJ5=UOqlWZ(~~#fvdoj=*n=jt-|?`;*%QRH9kxe{fhmce!Aas zC{`J^%jdZE}BBWzxOe zKuyg|(^7MvMp=|TtN;NGBNQkc*j6<0{zJ*-VkuSU4y~BLq7?h{N=M9LRwFw*Ju^K$ zH8yLd>T{#(@E~^{_o+DXO%6Jr48Dlfxf&YSow^%PvAultVk0ES zF~+m(X+&!rhq{7CqhHzKP4F-=(C`>dUV1xZp3>y8kVgnD{-mTMXg9 z1sqcralEY+_0+HbdRkAwmYL;ENyf8h`A&`#15%Nldw}!D{gQ3JG$->f$~AL!3OS_z z;rHTMh?EXo8Nw9O?z1XJrz5SdQ&-ZuYZJCuzzhYpK2(79 z_{5U!Tf-1y91XyVnPEZ*l3t}wppeSl+_j+y^OzJA8YLK1Lg9;p1R6b*KKy6yee*Cp TWM}4D1|aZs^>bP0l+XkKi8@46 literal 0 HcmV?d00001 diff --git a/docs/docs/how-tos/assets/autogen-output.png b/docs/docs/how-tos/assets/autogen-output.png new file mode 100644 index 0000000000000000000000000000000000000000..02106782e87441467253e6f9c4b62a7e407c562e GIT binary patch literal 5919 zcmZX2WmFtNv-aQ`G{K$410e)=3GM_I65JMdcXvw&7ThJc>q3x)0KwhELLj&Wx4Z9q z@A>|Gb7s2x)T!>OuIj3%r(!hJP@U;_Z~LQ&zp7NT!Le36*Qi1waZpdZm8xoOEu z12yARdjLSPr1)M++dKOx$H(Ac7|UKZd`($Ye5SmPn18As_MH9&{i5Zy4`9T9$bFc$C}EyJqX z$3UNc<^RkeVCBPg`$uR(Y(j{IHyMP43digydl9Tb%8D2LAzzSAEd(Ach!%?ifB7GR zhB_s;sBjjp@T-Ck`mZ6=<*mNU+4Q{@Gvs7B{qlmM11!;hY}_JyZwTa7D>}LjRx}D; z#J~w4$^A00ZZa|!VqP$gwbRUjHtJxhfbM~``Dsl3%>Y?V*%1X51&pQlSWyQ}NKER> z;beuk9`3kF;)#n!EfWpcTK~Y^;WBcGql>Xr^Ap4Dt0*l<_o+_QPtq)CUtB*WEsf(} zR?D4Gs&6orQpYRH2150$Ewb3yqT`yKL7Sx!2^9h*ZuJ$oT!%g`i-fIy%a{aR_K}S~ zm87go)BOHu&1LE1`o>ITJR!?f^Dkq@vvVuo<fA-79BmPt!9kq@ZYG5r>GN^XDbo{0Y1$cXkw zQuq4EcZfQ`)bl?Jo;O+&p3R|6NqRN$~(1IG^^3(ygqHB&jqpR z7SQL-^dZRPoB48}!)ZAKDeJXk}!{imH?Y$f;uOUaZklOn}(0nFcuBf7>=1SXR z0aX>Qy{_F}akeRq=K9KrG%`Q>hr;TXC=v&3sw`owwR z$KtoRvZ~B8KiQXe!sQGCqE*y>^Gi&&H^?QPjOlB3o34+Y-5`#l%qA{7-C3pIxN zP+t!-Gc!%iFa7&PIOFn`)yH!i1O@Md0}$M7w=f8Um6Vh$ammTb7O?Ig9f<|p z{p(J4>N-PDxMGD%Y4W~%H)GAq^G#)iuEB1hCL@E4bWWZWk9O+&Ye!puJqy~E>b(Z%@eHD)qT$-do4GA$ zNK;`-&vf*D@%8hwf6EI?HYZU*-0J4Um_zT18+8?XoDVH92|~v#r*DSt|w&EQ&JR4rm00d z?HB9lnVFdx7|bRiYndbS-!Ze*WBy2_5JWj)&g8o&V?z86Rg{!6MZ6q0e!a>XOB=Qb zutX?@1biq}*?Vr7aC^u`nv>svh|y!^4=m#;p%#lvRb^_84YX5PqyU07IXzp%Bn<>uywPc1yKd$9$Fo0*%x#U1_mQxD;IRaI3mm;}ggY2oLdbP4p9 zye#m}G21ACHOfx*%fgjtj6%l#XI=3;^%81GLooF z;1^n!OpAX`go~1IpG;G3%Dg|{b!N8s28^)Bl)XUGCIomC+wWGzagr(ct- z4{(8;T>gP~*>}Yrcm9MPDN>v+08k1#TQb;|OQf_SEFUg$5(#4zGH+S6iPo^z+uJ(6 zH1epm9FH{~{oXqoB_dM>-<(|$Q0AO4Wpu0|FrS_XTtBvvS;aG@PYRfX6aOsC=V9F! zvY$=2tD!oIHD*P;=S`fBoAdZ$mw%Sg_J~Htk2Q3u0HDtd>H{`1ERHp%!;R^KvS=`D zoo2Z6lV{TYjUDG8$|YdpBurQuSt69Qaao+Si|)LIUV1e%hk0m;rpuLWB8R^jtO9@6 zcMOw5v2FiY)HDgSvFNx(SP$?zu8JZlC3JP+42B5aX$+4YU7|PlRp}7%3Ttr435{+{ zPlpzk=jCnuIu$7Y`<%ov-9&$Ekf%7 zhVgI8kC*6ZQ2!UvMEu3`#!CNwD3K>qe;d`QS)o0Su)JFSZe5sZM(2lzEgV)n5$3PM zW(5O|46y*rG-?deX%n@DK*at;qgtLO8iH5%E9JQHJ%|6Ut{5VJ#M2jvSMlHsB^e*0vF=|_TCRfU<@H}Of*W$`UV!lc_jjQ>-}fGZtF zp)N7G>-bwViaJ5gTm*z?hT{8QFg`rJ^{4Xc8hT#9q5>cOfQh37=l&O%MYuUc0Kuxo zkdI49k`qWS*a(%{_)Gz3y?T^9FghTHF~Ful<@vii6$v3Q!pR~CM zT)A0V`t<*1Qo92G;r9IbU+$Pzq#wbvcq&~l0AD;2E$IS7gy;Ft&nD_ z5*nb_=3G#AbS~(BvncEo6F6k27?zS5l6?NT1L`_XiNqHa? zSC0G=0|4yTyS`2At~`;%!6wA6v&%ROQ8lQ3?1(ZjGCI`zxVyU>82leG$@^-_3Jn>_ zo$`@9qU=>z;$#I@&WIU*;?@vc(nZ?-hL{R3N*TYlb`o)sA(SpcidWn{CB(xwZJ-r+ z#eMFR7>VtmS|K+N(Ynld%im{0&x%9Oxur|fgQ|f(R?hefy|1{{8MYYC4H@7)jt;8v z1i@LmtGu1%?V6{B%p3-oy4QhdLsj z3>3pN&O6+1Zgdh`Ryb#F_(I6gv|HVL#WM>3}Qz%-H-HH8`YbJNFgl~(KXp&b( z+g?aTx4}_I28YXM@cGb5CsaC$qDOL&5KeGba?)X#?!uIKu*CDMP#$m0%o>UL*v{mG zO=_7PD^u3OyFdn&HYq=glO^J+=_+BaR-@I;iQST_r?58zO3R==*7IxRWvohK6QuLY3rr2h`a%NA=!Z5y&3z;afL%OPyM6<%*e8l(9vCX&%4m$~ zHEC&Cn-IKtm>+xgI_`UFNyK0%cy5>_H7^~fTr$*qy|X%sp)}&E%Dgr(d6U+%&qDGL zc=gK)0GcmhyIjXk2i`xnVj260@;;rjh)+-gbq7bVi7Ia>d_DF1y};sYN7w27*@zVY zh*Iy;0lQ!Ur>$#^0N~W^IS*bZcMng@k1iYqox2c73xb2yROw_C@(EDSZWn4r?b($T zjRbm~w5)Vxywr(9>4E`1kLmKC?MCq6mMe_(T8PVH*R2~r-g5psiYyM$I{mv+&N=Is z3phz+@ypd&8(x}$BilsIy7PPHb0q`dB9NR`C>ZDg%JRKG|kd*WWZ z;P7V4jg}Dg%k5<)yT?ROmA7h1@}sj&khfl_Dmm%7c&G2yO3FsnMQXUmFh#Aw^CxDz zK+iFn71DnPj3T<6embCk5_#8qt@2(c{+BNkJ2T&VdXgfX*U2XKG@C4ebt5j+PMh&7 zvY4C$Kx{)6qi0#%eE9?0@ZP5!p68t}>J3j_k4eTRWH^BH%|K8Ymw%wt$WBA-C_*+) zHa8!;>p0w}=Rn7EIX*x`;`SY-PgwJM(9z z`c_!1n8l%Vmc&3X>UOLAhX0}-Kbt7q{h{A*Or`Fd=d9|H?dXrscg|~%%YBRGXh86{ zcVMpEXA$5BYR~XA`~$u%P7%YdLRXkeqDJ+Xs0bBW_%jKNpnBxYR5$UJBmt<8YSm49 zVby{=BW5Q3@qx2hE^u#T_u9^B?~)w=vS5^j)p+#;I}cg%h|~RQ1TVgqL9tS)CJ za?|u3qH>U68A4nN0GXfIZ$+^Zadd3S0h5E$Ylr>4er2k~sgDcK*sl0#@r&ZP&IkO>tp1v^8>D~r{U%<&fb`& zG+pobv#dI!kY1AHXxMGPfu6S?jdH%c6H64_r%9WgedVAKWGA=!6#zb}yKU%)aS1Yn zJc`i&zPG*{=}xDprUHIIT*8n`)egSINv!D4IHV3fNc(J$nmLHCTmZf86`Z2&rgxg-Su`AFIn|p0I(>-WWJ& zfwD|O0XanBFf;OpA&$54)${)rKha>TP;CklVl-GyD=Ecu3gs@`Mc8Wr<9E7~5=A&ILf2MyUae|o(`TN~uj%06K-IiA~&-mj-xt>eXU zfh*_Wt%n?0A3N%Lu1)%~U)?{&`#JqU29yW^#lgm)ctS~!oFi~>mrB@H8M2tttqKqw%`X#w4JTh zixCnN--tRjca)u84d1k}Fs=6@aufHi;XQ+^maD?&nE+R=V|ZMDt(#ugO`CSTHF797 z87^}ctYpfXp`fSf#$D&qgXnX@*dLaOE_niGeNvx=9^7nhCLXi2>a)H=dlH-xO!pebcVWxL$Frak;qiP#>;d^@YsKJX`|e6cECSRVl19DC1J24e zZJ(3x;IeEIs31ttGt?MWF&S;z9AVI|F60?q=+vYhgl7+dg`p_szAvl)=;-c%%HC>7@oH@w+ML80sb|6O;B9HRaaR`UJ+Rv6087Dzr$7#^Y0Y3z3ZnTkPR^4?qJ`I>c@ z3HP46uigD)U*TcH$S6WBrIA3^*TUp{Oc8N$42;PiRn+DlAEwW78gh5a*)9spep#9w z5=QsK49n-lB3c_yQjA$^@9LW#;0gNjveM63Phb6+3PO-HFY%S*lqdL_weRlbK`KEb z{7#$FRDs=u!qENkwTnyqlN%-JFyul5|Ik>lhXRPvx?wA$e~+g(N@rQ0bgug=6IbRF zOp+j+vx-@G{lujpjuRm%dz0JQwqc5nXNB{0kV@=$v~15G=>7XzI^<+v()%S0VXZMp zSkFc=AunG;r=aIu-VrsQ<`*_WuKhH=7O(kVN_u>u>rru~@<}`jRwHBkc4)ogf$e*#n1|2D_1a7i0s?nS5f)8}#2hL^0dxA$_zpxS0walfU7mnPVy z9dSN)=KWAjN#c-+N;z``{7_;$9*4vz;#SfezucfFDIk94XYyK#{Ke$~b=2pQg5zl( z!hn5dV^T|WA0xw-x!QvAH4pNiZT)7<%g4h89Olk0beNovZ$c7a)kU_H1O2=T`U1z8 zG);@2`T5v7RLaEnFjcY1m{#a=w>OXG`QF^RMT8CU$}G=<&U;B}uLGT-!}b)LsN&(+ zWX+|qI4?L_Zt-Nv`&T$BKj^#m`~?x>I)cf(2uLam6~`r`Q{UXo{FPZlk~m}}i+#i4 zn7fA+duevIq(b$S#xVD`TopaFdn?53N~yLjWz)f$dUKPr^$i<4ws!F=%!hVfRH<*6 zVKLL16r$<9jh_Jm<~}pfFS$Njeh!f)vN)P|QKS;ayy43zK2^l=e{xDoZ4GrBHLL5a zbr4ZQ!YH~l^tG<7JhcvvoWkLVPwQb)I1ChGmw9%Ni2qBh>!C~;~)tx15$j?*i|g zBaYCi>9QNOm8I(v7B$h^vxM;pDFkuhe_WLHsIU9ouBZ91v0(ps5yGOL=pa*5!u`(+ z3mM^c=Rnu%;&1=+$sgZ^HL(8Yn}iWI0s;_jJ|w?(RsKgegn{O)(Zd2z{^uo3h7}7# zfgGR59P!VG0SF#t|E1U1VeoJ`Ynz|GPfPrBNZ7G36`cPZ5G94{rm@hIagcAQ{z;Laz|E13;y|BQnI!b40zXhv z!U|&0$>`(#mjMHd1pi;w;eQwDU;B*m|FuYz#adMxIse)Zx8H@EoIZPB?yF~u-ZJUd z?~K4&ef;~hB1Q;1)k@vZn>@cYSiU9~(a_Ma+OJymySIJ3yL?H=EUTapo}vHa<;^MX zrtgP>XM9GDhx^rMvw>=h$qhD>ek7sV0);dTa-X&_A*1fz+w%^Lb`LdY>NZfuz52{A zb<1u4XYak!fviX?=iA(~@3eB&gDtKHI)#cyvI%c#U%x&Cw|SrMFlG4*n)FAqU{DCA z3xyOb|7B*v={RA{F#bn%b7v&JI=nz9Njdj7=`%fKS;7iGv)`P_r0GV z6?(ccIYc?JU94sM&9`!})KGzN)9IZ0a1A0H1a zVwriS=P*NCmW%Ii8_g#1kOnYGU=hW_lWKO#%csxE67^0qbbon<+i>A&&J(JxU&@=K z%DI9ztL;hHA%Vc9tLSJQ(Ur$8-FCa?XRq~Amwjq!p-S( zqnPY{lAJ44f>KLFAjESho!P839}P2Bkg)!I9?xizjP#hw#k@S26MB9)mwu;Nu9xY; zQb6(u^3QDBh&V2g8%e70*o;~~-x*tKuoq%TVAO0std@f|ZeE3;lb)=S3zUmTJ|7rT zgnfEuo91>F$zWipx-7x#GWzxQFP}Kb6R78_-tBw0VE$~v7iI4QY$#s0!@+MR#yLv_ zKShIIJm0UroYQl*9Umwlp{z6qWwLanA4PR9_B#)mBSEQUhgK2l?NL%L<({b>&?DNanZHxV%8gZ=Fw&f%(BQ>wf~JA*#pD3$N8`|w zgNlx5^2fdTs?ADmq7M^wq`<(wtomLH&!=g$a{kcUCRD0aq1}v#wq6Sx6r?N^*2w9&|k4 zLb5h3^7elr2e$iqr~Obb(Re&CjN!?5Pm5>j=y=$E0!%+PsJqo2S`z&CgNpJRb?X-r zc=d#U#o*!LoQ3~pO~?ibF7x@MCuXN3_(lGEoB2V54fM0lKzOh^)%g<5=@oRf!}Xw8 zHP>pkRJmNwQ`Hm-77%)FUGb^VZ{ez1VErkf@VFTxw9a;*U*`S-DSs@DyF{x>bTX&U z!s7nwh*6{TUBK*4Yj1+7q3ZKUqDXYUcsS0ecbQJz3qt16sL@&^jV&CJD8A_7 zrpZR7TO4rrQmbc83Sug=?kv7W{EYS8sGj}O4me=G(J{Bcf01KhfIFzRrFsP|P5BU8OfDSs~A;S|E=;LB{?7_$5=9OY*?uR=HEt zH54?_*W^4-rc(uDjGE;R%~MMcTQWx5v{_ES?7^)(x@~muq%X{{bu#mMrVQ}}>;wpi zJo)|TP|83G3cS_<37_6VuzN{5cNjS@cSF#HO7(7|I;`5H3i1ACIE-*?RGl0*L-B>Y zh#p%4a$k#d0OwWt+_qO>6i=scwR3rQP?5^LbGTUWbH2gz>8hc_z}wdUd^JXpiUzR=Qi!;=3=pGl0q6++ex?oX&$E^F3LPP z9KXi$HP{;zOZCO`-d&^4WPruK^|W8tzU|egmai6JfRDX*ozvHB^WrquZkBSn0384K zITz^h!VkA2<|l{{MO99QOkoeu#%d7aOL)BZNwI#)DIT?|}x%#_UWJ8mF6*YV#PG}sCF^BFf9)>*kydD^w#Dv=d@I%qoS zPAt$F#3XWf)gsCt#Gm=h-?mWebv_m+IG0@$ddK|N>ES6&sj09FA2}s2O?M|5csxPUzGUi^!=x!p9pC!1p;)OJdubY_Q z(~Vf+1qMPT)}GbXq4wZgFpTYspw+D*-16pMg;8E>am20%)5Y;vF#LJ$)l~6+9K*oo zJn`6U;GD>>=*Yb9g!ey>1uI-LH|4DWUxZJ7`n}ueXaanFSmh|0Eo)-|`;#78C{rHy zD%tVTW7ozyO1XATdzb=?BVSQop6S>8y(uAXhB-rj{eV+9D0qboGaT-fZ?)=|{Tj9V ztE90kA!JOeKYCu&>sY=sE2gktQz~N`c&qCnkN&d4@6ii3BGf$`>@?SL+HXz%o3E~F zwM(*dxv55bG>}n>iU0HPYsNO@@pO941rm-UFCl!feN1@+?<4#?3!qh559y~pC2mm4 zIqel8)eQ0JD8GOQhDXc~dQiNQU|>3k?o%(RdhlN{92NsyRZbwcWSN`VKy<#O)mQRM zN{iD7%SflTG~w1~kpENESdHUlEit+Gp&_{r-}J9gObQ0ws~-_WOFE*v4YuAxKseQA zJn^<@9?=Ny+aiy26!85Ej z9onCmnaDm zN6iRi6IZsjT`gKqkas-YSp$ygj63EH`2Nz_I*&@s5pM*A>G0+gc0daA$YzGSq=A7(M>2}!4d{`HzutRZxT}4$c7JLzSE@p* zDxDz9W5ZD_?LH&lxE(k-G2LFwv#*)fG3)cD;f*%Y%O`(r5DL7mYVO!c^&|5LsT&N# zwr4Fg^kJJKaPeD@JG+=ts;<;7e}^W2QGAtTxC~bVqHfBPhE%ct)K{F|bcmqO^0})e z@^y3C8su(qhng|38+Sf_tAGW$9c#a1nIGy--`|Xpay?$2NoC@{oaX5JMrd-rJ1O-H z!t;5$CA>az781><;_|I0vZD4=V(ECGRwA;q9WC%csG?lRHd#%iC#|cRdhc^{FmL@; zr{<5i^Zy3Z!pB%)_^_FsbK|i0wX#tt*zyjOWM|z1W}}Tjc&0)hpb~9CPN+Ew0XOEts#oM)qe)+Bhnejq}`FXRBQFTinXV z&hHL-Qs~@}vAU~V8V}0(PP83n9iAVriT7SFTNXe-0ccIs^$^uRcFr3-!95TV{wbBz7G&tXf5^`aiFtOxQAp z{x;8wT6a0%0F_3k2fXmw8f?f=7#MeVZsf+5s)VWi@g^&7Yt_p-&mD6;vQ>e%p);Ay zvv9Wxxr)z)q95_Rn`T2JN~ zyAGcZDhNH?=Tq3NXOkY{e&|@IpDb{@y^p*YNB6j1@i(I%G~0~sEjz8B`XtP~-gRskb8?lYH;hLaE8X2(;~*rIqh2v|(!^n|n#qIdym8T-ID{-pBaH zRy72TI0WX?d$kW}9XA`#nI0Pq7aAcUMPHRj&cJm0ed+npvaaKa1}QBHUoE4}`}!Os z?gZ=&T`F$6Kjw@2+zwiFsy^}2Afm+PLK;Z@ipfA>>_d%iLI#;f&ie}t6FCsYP|3Sx^16_C7QaIk0F>!op!h0?{n zaeQw?o@R5BQicola-s^>Me71@QLz14+GQD5<^0Hq*v{A6Taj!(c<8N_RS6|EME z1pO|@T%vyPPF`$s8gU`&!&txNIxe03#c`|ZNl(cF-xry6VM_=s?6}SB9!VxSu4zpa z*=;m-N~Mh5(iLXlR3fwfRg6FXbgkvLhE=}-^8+%1PrsGnc{wN>uMQxy$T&Yo3a>D` zHme4f3Fg2FOJU1k3@{x>cFjkgoEqAXIqs)KGF3Vk*Q@Rf61Qsv>qkur^|0`Ip@HXA zXX&>%YpbMo4)?!J5Z3MY4Sg?NYp?_6Z3M9S_tDJOXUdP<*AaKKL2@>tDP;)t>4#52 zQi|!HlU3tCSWHlwbg$_zePe0wC69IYlFzTa?%lKoEWd1qR}wda&+rNAFVyC9`~CJD z@^oZ>J@-p}$}p>(P4(I-Hs%w*kf}GM?_&h- z8W#$vz9V+1P9BHv6aI++1hS%R{ep?13mqK2iuq(F9wKJLu~S!YT_gp-4dbkmIHuT< z5Z`|P{){DUvF+i6Lw^DAmBV*;JMLI?s0IRuzrm`A+dG45Cd2k+F?>sJMOKdUHkEv# zuL*7#Ou|?<7_WGf(>34^#}JVBQaJRlKt79g){uS5nva&}TUqZ>Ow26DDk-qu`e-RM%rEM&slgPR7e*9~u0g zZi_dme4>NO$Vd-k;VFG~sIYI{4z(KkTmj_8*X*U)`t3Kn-J0&sa4s_{<;Sa|B`6Dr z^~{GOnj`N}q4)WbCD9!u#`GCwFVeq^3w6*M?2?O6hslQvbpOu(X6UsiK~v=IlJ-SN zK{g;Q35`oUC)_-a-ee$lz_4W=Yk`StwyDEL_HVPv{~C)?@B|5#q)gYYX(mE!axLg- z*8E55s-N>KHd~=|Te}Wt-c%mlk-Rmkp^?Xj+j9VV=`P{Ho@^wED(^STfK^m&SYLQx z$RUmss?ga8T&J}QiS%>}_;^K3d>mhHp_`kORpNK+F_) z<6cI2tr`X+fcefx!C+0cmjj~U!PoxJ_waA*h3;+M>54Wk-YqGdl)o-e0er(3t_AuY z?AqrzsV5Mz2PmYXC%GF^wnZkr*&j9PuYBT#5OP7pqDb8TVuj#z5No-c&vj`hmOh&d zh&2fioPG^_MslW7WT4@Y2wUD+VV-1e!!_YTR7H7*aYN(bHxPZapkY!tdV^(T9B_@m z{76<0F_JSjN7qgGWCz}#(Z}9&dKOEE^Pd$c;IPq8nWD~dED|}6(Pa7)q1Y0FV|(41 z+bp6M=7w^OOYgH(O#+e6!i~It|Egzt)u3qan(svO`MEjHnx;WEBg{;SP=x}37n0Bi8+Ys zSlnRCh*A^$AjrBW;H4KB6TkzD!Jba&u<-d+dI+leuSV#qd6q9-a#tdQYN|pljMB=d zx2QBRQXVg%ByPtph@5MsKj{53^u7>}OU_V9KuXrNOt^oLW~2d*PQu?0t>WVQ;e zRu4zkM0Ok36hR6>pEnMWg~Li>o8nH19IrYHWp|J^PatWvK!XGvB&MYVEp{xmT)$=2 zbuJ6_O7eN(*BZYxL^uNf<gW#u zS{hDcTZQv%xlr9VyL{Jv|8xd`wl^GA#SymGr8^fLJ}D4a>y*? zD^K~Mw#%*oL{IggagGN#v|cfar@0QWc~3}g$K%sIbkfhHS8=UYA?-u69S1VP9a@Y% zD4s>XSw3K=@#`Y1aEOgul)tgJ&uJfFgY{gZ#P3>5T#U9{l#LeA8d zV zylCSS?<16!oulgaA2215h$^-Ou5}>U@!R%t1rU(-|uw@wbD?t$Dz^&;!XG)@zf+k*`Tm1(BBU$x65QpO(GP-vDCQbSnUOP(l*#^ zMw*W)?p_xv9&Kq{w!F<7iA=gn6T|NZ|`CEtVT#tF`y zdeFJ=et;i~RgO+C2O(Uh&F;volaMtr=miYJ;SJ@ep7&8F5aD;49xR-zQMSP#nrz`@ z(iG6sJyBe}>=!cyR$cL1I7}OmRb(_pOuDXZ_S#O^>%JDQYjUt$;dKWs)!UN( z*AY&yK8X~)F0A?t$`oj$iAN`LC=(j<86Tqd_!;Q4n~f}VyOnFV((I)c_WHb^v?n9H zm1d`P5>BhEbnYxo!!E^vc2j(0>3r{c8Mz0Ub zNp~yY-zWk&u64QJZT@|-uz!j({o@GB3Pjdvgmk4E3>XDr2owcwxUTNEc~5!`f5)JsS%ioLGj3MD=svh z-6x4T1sV-t3Ju3?*PskuS7ZM!U6}J&G;E!mxfCD0b8l#yxg@v61dM+- zm*gIB>fY`ZQuN1A9Pqm>^%It=N`OE8WvqQ%RM>0=1)X2uX$mZ$6SzH8M)~U9(@b}xqFT-0qH=pU^aoX^L4TBUk%d~b|ar?%p53)1W^EG>E@ zN1q?F?%hpSnxUl%f$qAp*nb78{I6rL;rQ=>#MpP7TA01TUYX0W`>TwiWaAj%qIUtn zMBE`&AnklfQ25-^(6t-2`afC#D79idN1JWN8&5i=%z#OTGbFZtp4*hE?nu?5p&l70 zK~PY%N5z?8@Eg{sF)or;fPgxjulh1Tw7~0mP=<{%FS&A58(&z84U@`gEt%&M6SUhV z740*XwGlm9(eV@#o;fH=!gs~-70VpG_I@Xgrf7k}xph}T+EydTozdl{xNv#Bz5UKA zxTx)|>m)6}a`aE^gKu$tk<@=~8Wz!+E9CV!Gg#r(nXTB)*ks4@R4}ZI4%c z1&aW8ugVJUocbf>YKyNHnKwtk+}Fi zkO#4lEv;Xq=is%}c05#1;X%-(vrJ55I`cRTl>{!oPOZF3c?O84?&!5zDWs3^h)n|v zkq-5u=;^qVzh;8zP65_yvj$W4FGf9(pMFhq;d`3E{cf%@8H`yzU3crIO{>Ku^-?H+ zl)Decl6sodGt4WeaWp6lH#%;{kSo|utSwu(+>dduiW^Q>8o{f20!TFwyx3-;uNQhk z+x){-CDK9I+fExaF%O1-kDQ39;$a66WkyZEFPOK{uR31S(sBVdHME@Bdv((D>>~N1 zcRw%|fo_nWWuwhY|FVye12t5m-bVYim2siT5Q^PJ@F+e12-C``2P4a4RJd%#<{^`;SX6ey*>4?nTrwdeVrE*;wDb;XtVKg<-0fPzuqsQ&n9Gip|8vus7!x> zEqH&?L{j&#Go~w>_-;rNtgfm5xd30`X~q4SCmM~h^_KtfqGKV8{dg6DMyv9ZVEG$> z`TB%j!y%!^jgwIT1=xqIhHoF@u$$eEoGxdWaqj@2#wc@s8HEfc@SYcB)f#))4#Q1& zBG7ya(0iDqTtCDWzZde_pNKZrFlzUWd?>hR;IUM}Qk$1=XWeOE62pST4U zd(in066ieQYFZWalcw&&M{D3AGVyfCxmy=R7}clvnipFb-4evDVFnMLTk8(U zd7SU7nnK{H(52i>b_07Ya&{ecGC2rVY#Ua(Bzpk-X_C>+?J$nGK3@KDtxF8+m2=tB zq(zu-N+5iN_vK8ILapdA<%4QE^V@pmL^f~pbb3e)X#?1cFO<(%rAs~Mdew2@2)qyr zhPFNIXAE(P7- z58f@UK7l7LErl8DsdO<*Hp*LkTtMEG{~wo26?;v0VK$F~wCd+A+bC*t?$HOay1&W* zvPWNt$mtu!Ebrr6fE=f0`{iZt(zMd7Kg0#Fs6rkf-1=prAN4j109C6E)=a>`fi)($ z+H*!Oqg z7t9Ga>ajZ|Hv7&>px^95D&&{LVOl)kQW?SAQc4reHEsx-NyP8x`R+qZ&c&SAZ|zY5 z?<>(P=xKbF>j77^hiUoIf(Pu|N!q`e0V5ph_K44^TEAKgKpV1_ z_AMl+0BRcJR5M+JhnpO@LR@KHPt7pS6ndnzGE zSkz0rJG-OUn7pOK0D&E?@(P}SIy~f}l9m^y@vZu=u=G*%LH*7#B6%UU#R}c)cTas% zNZE_#*vIS`IfdTHnB>Nf^DrKQ9Pwcph|D}bU9B6x!w_2Ey<5%H^VY4QXi)-p-Kf7pdTX@&rl zuTp?g&0|?j_OB@KPi>l00-&+e>nrR3^OL`BDER>$ThC^J%AfD~>n64YP~bt{+Hw9( zg#?o+62#tRfZg`6raIAKQKM^}c-9%9D0l zU>~B5farX)dh-k6cMO@rsz2rV&FP!Epk9MoEw;StH!~Gh5|rGilsy!33KHotD!I*u zIS3jvf3O?E?{LGeV&vY%==bT*et@u!5x>Dmz9^1m1g*%mFFOuGlEECyZThyc26ysZrm3^$fPk{_o=#?T~Bs2 za`aHCmsBN~8cP5!d?9Pg^kReg9-D+WF#^y3{RY$GeQ@g3@81@TY;peIO)GI6r%$!B zt&fgY&UJm<4BDcp1JCf^TG@)n*}MGO=^Eb@prz%ZrOp4i-mU!T7(R}oRsswm8s+ur zVp8slAV$3++3g?f`PhBGWAX|p_ z-u^g3SAMf8{rdBF99lU72A!{WQ;G$zg#5@(UroM}5!ColL_v*321~#smvls5y6d?q zE4z`v80@c{jG!sVO zT0NCI^Opt(*9y0K!#5uq8^v$mJTz`Sewcv=9FQS`lURV1mlCsf(A<2jFEZAOudA4) zqYMm%2OXZe56h~X!26$pQn8nB-z06~{M^J#p9)faABG<^NZAnLnK93ix6RqQto2sz}z zdjgEUtA05=_#ZlNeLJ-+_sUboMokk_4$?7`l2B>dOf1O+?AOz;RXM|ii4e4YzXLcv ztu5>-zv}Cky6~;d!IQN3|Bl?k+~JBgFv&jN6~G)Ujf4llW`lJW9_e-k{Nh^M0p6(Y_wK zSkE)lJLM$gjFi*L1F9YV&!^r=#{xDXZ;9>^^dQ5Zg>c4#Fk5<#?`fHsLP=qb+=UVU zeQ+9w!g$?m4(OeRJKb#J@X@F!0TH$jPFM~Mhz*)f1c_L4dH;QV%m+g(+*6VAd>1>3 z$~Syt)Y~J24WelRL$GjTu+s!{5|4t-k*~rQ8-A|6;(iCdpz!a#ViYo^wy*DTMESpU- zLmZhsT*J=37v&8WgCcU7qHsS&+=ft5**P&l7f z_>a&3US(>TPK3-C_XD)w^_BI3vu~0LO957pS3sleuaN4`Dx!}tsL}x}0rBNjf2Daq zKmK*~QUtQbP)2tI{u)4x5x(3v&vaINWF21s09{B{!_&nSCSpvAUg=xoQG^W~rn`bZ)ZxzJNh`t;B0|JeyF5jghj7)p6{d`1KuTwFev-Si%C zt<6F+P?f<`D&F@$V?IiiKcYcAnU2fX_$(TYJ1H`K-^?@Q<(6t7z2k&6b`la z2PerQ_32;gODHE&(FlBVf@vSlM^k5hWenwCT+}8oCf$zWN`(WpHNR9Lhm?G+zW4Qp zehcKLX8sF5k;Tdj3dprv-0A>j1m+jvq3`+QJT2`NLE6Ow(T>QES4$OuuKqfKO7M5% zqwf2TPIE)0bp9@wYv_{q$%IH2@sC{*nB+2DLmM1!`zOn}@{BG@tK~*Zpq-&6cc{J4 z!-+4eglrP?RHYGa8gS&AtoqbE)FKMVW_)nuZ-v;(%4R!*YzNOfY!;jllw=eXR@%Jh zVuYS5O$O?xuBdL%h}<|W*2p)%)4oJ$G9LbIS7*g}ee(I~u>d)UwO_v%SIXEH!n!L9K z#(#+RRSMqAP~uyFjDM(dDlg|=UA9K_@c$W9JYYyZKc4lh&e|gSWc9K64|mWD_Xjp& zp6$`&JmR_+t5p=cN(Re<_e*V37P^OwL8g8GNSs2vhrB|v65ryMtIwT!8&5QYe9Ewp zV;R0vZhK$rdw(sTP9Po*ml4d3l;>`+5iZeTyEMC|jmM;ew$R_lbF>A3r(alz40(Vu znHaY;LI3ly2ta?;s|lG|(xdFu<$QbBY@TtxeaM2v;gl`mub@_@a};r`nNkAvKe-x; zr+?T+A&VW`HY0zO&j{hVS zqYXcl!Jmf>vH5mqX?`@7b)Dl1@8Pp*GgW7S=lKqV$C6b97U5=jDE|otSgK z?W6mZb4`|&+#mh|4mTDC`UMl|e7o62Jb-)dL>Y*5x~19!WP(dHodI_=f4Oc>?2m`2 zbo^v{z3bSwpVgB@Q!IO=*5K?G*rxwb#pp^Zl1ZvE@dn)SZduUt=jOrWXDa(&=n0>2CWKk6Rm6`8ro!jns6l%Ay&dp zZ+Mt(TSnq*0*Xe*6q`{UpjfCxtvKSs&hB<+3}pVzfA|M#VVM;1AGjYS7iw=gjA4#r z&za`d`&f_uxLiKT*0h8~0-Q(oqeZQ~$H7!rWJ0DgzmeEjD@teMs+#i#E(Aa~wwWtzdlhC>15<5b8S(SBD8m-C93noMuKocHDo-HigF zENLK@t3o8q_61q!s&9kaa-$Oeavtgfo5jSZW)XDTC4+935%;sE6Pfet-t%#5%XVh% zGuQJSLRJQ#I7&3caNQLXsgzENaIQ=@s_&~5zIsWf=ad3>5~F6I%LCt_!&-pehI6lW zFmQANl7mBXnk2ohqtJBR_cF8*^pDKap&ympMw>?C>HX8*`$B6|wCQ^$Y0XO9^s+`T zCdewqcFkR?jpyy%igQ(cnP(!3JWqZy2Kd~b;q#B)!d<}h z*|IbjWysK)POkUxL_9;ZTb{BG5U|vC#-4}Ye``aU-I?4KB^R^`Q{wXLHaGj|lB=9s zFiG@tA1wwZ=W?;`cpd>Lx;oWFjme->SjHs`iR0p&D_Ym6q%mkeA9sLKnj;)}Zzn*) zy)5n~%5;AYVn5w!JN%9+wd-&3dvss*hdvRvtp#a#Vi-Zjj{rGP{T@<7>x#wTd35o= zii@${2F}lPQlvKwid$p4N{>Gjr`6|Cph*=wIYG)-=@jIIu$&gEGng4wa(R6#caDQg z)PC`r#_r8bx~@{_q?V<~%a(fez{EE~zb(+aA-vev-EMwDWf-a#ZJ&SPpTufu8e^rg@bi^Za8 zM5gbIKOhZ9b4vXx6D`uMc|V=9c+Q=18F$Y{T5#de>m1g`IrfiNyi@ssa!DSyQCl^# zMztU1nXa=7kPLTNI)9-2^fSjUP>#umJF ze5=Wvid(xo?P8T}>x#B$Xun@3i|tZ$k)I_&xLAO z5%07AXaRzOB8K|UY!z+mA=zePwpWKKPeJQ!mw>1MBY$FOK8arymAn!t$)&OqMk&n8 zo4^t&)2MhNV1)N~#%_e1I=POe+VjGFWjH3NcimgB=%bOHyylNH3T~hCF?Q=A`v~CJZ8M9~QG{$6iP>{h zof0yq#2N}OqKkh2r$(?_$ZO=*1m(lunRDJF98V+_V2hcQGxeg>0?mYWgWM!s!W7V z-TBVtleuh-JN(L59W#-D$}Q6jK0ehP@zX@6rS}#SVUV}Zz87ym)UAqKOa}bWIYR+k z-Bu}V`4da$_ZR}_{8^?!#T+D_$?r@(tOT8U>%5v4SlVi7 zo@sw0hz%O)&8NHH0KvSmty)#i0bO4aSVODjO56ATiu>93w9kC1?SdKbM_!W>HYy+kJrZB8Jn*n$h32NK*O?GBQ_}mAgm7adaq3 z!xnpfo4}0iaQLe24lWLkt+DY{_%^Iy@LLV}Hy5EP)t_;{E9U3&(C?G89Vw|NBqyts zzw@U}K{D8U~}BoK!w!Uu=Y5VA~T zB-1#pCrdPiz=<}z6|IAs4^%lcac3Ldv<79RyLQg0c%s{(0$6uGlM=_O1wg@V7$u>p zdiud?${;GMTXCwrVGaUdONc?Aac&AVv(h# z@{#nnEK3N2h#sw_EUhQh=is5l&@v8hr9!DRuB4Mby2Llia3bG20Ap`|e$ws2fgmFJu0qc1;1|)dbx!eAlf zfru~`Zs-NPjLX_&!N?*lkgqVaN^FXIOXYmb668IrpvPQ{CIC@?yub@_4WBGn6Me0 z6Uh9W`AbvI%DJZT5Ok*dG`!47l+CCyw8LLLlBZ#tI?qNnKc!kDrOfg5kL(B(zB#S9 zy_I4o=JzP_)RvLkl>Yii9_>%pM-Ild${GUFg(HOndZipc_a%K~RZwRj_lNxJ!(1=_ z9{<3U{ncl$UaG7FEZ4||+TVQb6+~8ci?wksAR>R(N_2~OvexbS4OEgWgjKXL z0j%xRid!RI_BA}yVS-Yl_+DZbqZN?$%%L|A$wGI3yF;I5tX4cj+jmkA_UHKOToY2g zwkD{OXd)j!$%(%3r&r93L{a*YQjpnsrdwp2z@qsSaoqM(>b;cS!PxNogi57#{{7|G zy!q~*s(;_qTwB%@FvU)Z9*MO=K%=A8fztDa=j0gPw9=JyzL)z6;o;c~ug3*j+8IWg z^;@pqL&|ITGkvd4f&ax2DFh*Qql5tNy+DBH{fuNk>HA&D^MLI* zMUVQ=BLwFazO6t3Tnk&^spO9l27dj8h4)b;@Wi~9`lLGbqmxFkaU?J4a0=c=j#I1C zU#gXARrx=Ydh9xmoSW>oZdZwPs`chF6<$PBfw)&ow=lm@#z!Hp=8xcGA7=i-0kl5m z2%`60Bj#UX`W>zm-9y=#ogEK;+v;OCDIG4h9keO)SRN$BK~Ssf=Q)%MlZJpuIi3@Qb_d^Vf6<038tev77{bobeht;+D?Nn+m;jeG-U zd_|p{G(s?sv=t@e=szp-`Cm)|F^#nLuB)m_$j`Y;@XYaS8FH%alo%dep>N7Juof$E zok$Got)TZ6;p`cO{?v@-b*?4aGYPj{wbcxFJ!v|*O|P`gj^k!+jz?caLI`>b`s6Eg z9`>v0VqE`6S4a&=P^1e`g$NzIOWJ-2LOd^298+dBFj$Rd_fh|F5+jLv{0K*3S# zlw_;~pCSnFw@i$cZmmOyzAdi*HC{wnF60Tg`)Y3die6zoFq06NtY}x z-|v?(pH1?0=Q4|+aULlQ1U-{B-mwg-^k??<^FqC~Py5y^M7VchmH84S((V|nkOY%h zQ6Zr$Hblp0WQ=e1+lXbJG293Bt|j&3!WL)~`{2o0X8rm>gq5#I{6q5oH-!0mile4Z zf<=1eR6IPWRb*vmS0qKjLU6jS=m|Si`A#I zaNL2fY+nZnL|jKX5_xO3-5Rd>%oj;d@fA3wL^Ef(-&X6N6-E6_(6USs=Xa-`C+PTK zbC$#Pe6_T#wN}G5@3J}5J*ppyk;Dhza@JcD^YOXkF$m*n`xWKQ>J&<} zggSsY_X@&Vu9^5p2kjcE81K2+9i0YxzKrXMx=LrUymV2FNrs%uo(@i1H72z6zPo3!G`ppk#Q|nAfIJkNrFfX(M9NKNo`iMZ z`iI)NKc+W$ZONN&RqNc`i0lLGtgvhowON{P?mANyOVJ!=zVrn)(6MZqk5^j z(F_T@8gJXOnd&l7})@OdF@}84C`_Kbp zzR{wH*^+UZ2I*+sITZ7mDq8+zMNwWpmCF`K$t=&^O|k4k5uFHEA(?entxwYG_N+ZB zJjik+WZ{0lM0=bFIMe}{bVIvWEanEgOAY^se|8q;o`uixV6LK0$P^(a%#8wWI_5Y7 z9#wxy~!-J3O?CA9f5GA=9??}2{1{is~Jy$@AfC5`|i8s1(bf9B6#}7x#!Bn+{$xk6nTX{2fxV-^Of!s;}psIMUdQ-4lE};!ofI9^IF9QajX2f#(!lhU_w-TyMT&IXQ@Nj&$2=j9mDFQ#n zmc6g=ZgD8Vf9$vJeCM8=8A@K$YS8@~UvWeu1mBb=J#Qx_8OIR*$>cRb5q#?y%lW6& zAGZRDuSSUJ_;b+KKJ<%(>B*W*e8nX??zWA(i&ETsW5_u<8_ulL%Ju1JcCVo$_(S;2 z@(3~-vBbgcm8I9kzCwjr=zO_LnaL7XMlLVb!=N)H zwU+vnqm(D$U@5U6X&c-yBs$K?sb6f_B8XJTSHiS~+}avOh0~>g!j0cc3{p7W`#JIf zjK|9elD^FH-A4J{(920Hh|e?G{cGGZSC#7KfPn8xMXnkntBj;r%k_QlF74paDeq?6 zML&HH__TpCVkaJXKTO@CO2BZCY`HvDcHqq1$vbo&1~IU=d}n(!Bep42-xb5_(io1H zhG<(ISK^(4?U}8NAc&bkwW(NdHWt}qFF4to!eqiem#G|U5BlZsjRKqhw{w-TbGRFZy$7&5O7yfU*P1k#HUch26T$!0>sf5rT zSd!;1q3oskJlrc#Xhtu39rV_&ChSqM)?H=#3Abv$dh$GVEIoNy%cu1n;OB|PT&fcG zXMZtVN0_J7b`q<5LH()1#LzHVe(;GX!I>+5uQy&?+a^Vw0Vd&hYyWN@9zmz=`d^@N zb^CiM#SKEDvKwbIlopzCP{Z&y0+y>b1K#U^mB3tb4BJ)Yr1>u=i1of0mq{sv_&;tiiKj4Wy=RX>`pcHGoD zAO_QzlQ>w`{TL$Ed=w`@3Uc^QVV5kAdG?@h^`7{>KX+9;zFA^(|LEBc>! zRg_Ef?;%p(P>A^rp_!8LG-mhD+g)_0%6P(IbSgz?W0G?9>1Vd2>es(k2{n&V(zHKG zjO~)bTkI)=s;-aS$LdajY8 z?eUB$Gf}Uh%qGZB9Mb@EHYc|c(UnnpS&Sg#%t>02(A^O@)E#;KgIyx_J8ohPL* zSFO+w#av>U@1~#FomY)9R`%zlDE?N_u@*kVb(O8!VfcYRYo~(JcNu0QbCT|L&T_Y+ z@nj_X7xA^)6e`4cz(+H0%-rF)Ft_s!ZClm+>(!v1`i9hz{=WR})Z!<~Jc{~fz4@01 zeUucLA&4!WUGfS_-wBa8a=3S#zxlgi`oy7U4{`N7FzCem7*I99c_MTMDtFmMx@ryf zTg_KGT|)dF1nPqy88m9k#Uml(9cGw6x?nhIR$bbkmTD8{QCCgO8SG;^Q-@}LyGq+d z3OZ##F68diLJJel_;e6S+v7kISUY02>I#?v2?DO7wm-QVGkkR|NH*u@+8g9~SQ!aG z4gTif0o@DC@G(P4?LN! z4-NDd^DHb7!o=_k0cScaghF_rJI*jKEsVftO`|UGOWRTUFYE~F^tM@j{xq(3uWL<< z>rk2`NDmO=Ng81n&6`jtzBiyO_7e^1gVpk zxlw&i-=W>wl}TxD41G2_#kw~~!iKJerC%h+;x5qwjzua1?LKvi(FGL;;~;gt&XbN{ z7$Shf;lB7KR%1R{sD1ZNCWy17+-@^o^D*yR?1hgKnKk9u3= zrHtzdv8tlY;`L_s5PT_XPf=U{vEYn$n@S^r|z%9-Np zc|{A$doPY}->t_B3Rzz?4Ow!(dr{rwAbXXc!SR&X{khX|%4mgw35T|Ws?XrZDr+Q# z?wu$6V)eL(MOMQ0<)w%NAIzo?mzn1>(YbBp{$w7hL3{N{ho$kYH~e69&CbQ(AvO#n z|5LU#7BXwQ;>MBiOnqW*`}&~_VH`Lbr|@9z_|fT9%I)A^su3u|L#0|}uO@V5c%64m z(d!zG2jFVH_1!f(SzJ+KQ(e@Ur*$E{yYx8GG2fr5x&TmK+yc*#DXz7R3(9Zr+lG)T zs_FHcY2n;Na5BWc8~wbrO%*iK%W>&{Ih>nWXd7R1)b4q1(iOHY<}3ao7)F*2UIBL# zk@lbpF|;7rs(^0cZ0OBSs&Wa_>$CEYeg<@`>t&X_dD*d4#c7yiAYo3#*UZY%xlFkq zO0SVu+%_^J6&zXL@qw6o1#Mak)8rMk3F_chgn6+asn50FUimY2%`d{LGJPrx6-R5$ z;sNW&geLV7YmVNZV`LD%95P`!Ba;_~;`flQ5SlN80<83Vm{rT~bG-WKdV3wEyxu64 z1R?&$Bu2j)Jj-<2z7WbxpP%4#Kl>PH^oaUc zZF9#DEe_g86~sw{s$v_KE!XP?y?U^5*tthFSvEa2`=C|DJ zr$Ac93lOjNVq_OMIVNshbsPo+Y&?20XZzHs0(w`v=Hq{4i~Fn*<8&vdWz-TI%4o|1 zhO$RQ@b5N2#WsO)M`h)WRUiqWYe)T|&;@ya$S`bi;oXE~N7u=@SqC;3>xqy;T`7-t zGV6SLVmK$m!Y7i?KDr{RbIF{n0CM#)B-F_K2M4(?BGfYN@T?Qjb2U7nWPq+r?fwTr z{#IhWL#FQ*FF8SRRmIyYrvx5#o;P|$FP$S|J%R@)WAx`|F>m&e5PEVUy{U1;g&)y3 z$|iFbO#l_y;Qcm=0@4Ny(bTToa_?fvQ;CbMe)%s#fvFed0;q@q(;`b559d5K(})w! z>rZGBb(q0%`xI5hVK9KRq7&1b!v^i)5C<-BdR1b!+RS9o^5RL9LhHaw&p10dj>VGn zq{{^OhErF(nc#-DHytwF9jRJRBhvEa6p$h=&d2-9i%^@$T7Plm_2#@a_zl~=ysZ-A zG1ljfleox69W-f7L=?rrA*Gq7Sc7zm8PW%{&XZhzuwZr>j;d zr=$L)wn*Vu-ucQhX*3?!w9upPLhrie9Igx`Z^J3*`@CL~k2p-qhhj!~aW^;?>YsTw z6HzD7YE_tA$U`}q6$>f+Ht*P@hej*j3TqiW4qWFIta|hm!-L~fe>gi;*bFas>10Fq z!Muhu-XLF{-v!1dqx0E^$MZcI7ZX-<93Ir4^;HDrKSS-0lNKqXx!g{JMb1Tp&b&F5 z^j39^jIV7Ng{*{W5{g?ItDLY;({XG>UZ~HhXJjuihU4ovASF&`Cm%r4sNPRG?*9tk z^nE&+&7ob3Xaubay^XGS+@kPk3;I=ThC=i~IxCz|2GipZ zaO2{;@`#tG?g}pQTiuSB;qQ6sikh3C)}hb`v0II%3Lm6pbDcK1A%>n4>a$|7uv9j2 z6io8C>_c1+U)f-$X`yHA@cJ-REs0m?b-0Kb{d|iJYCY!2Oin}*7@kUlsepNRLN%eh zH0WtV7m4MT+;hFdk|vBJ#NCo*SG+644>s=}f8R-D4BW_Q`S9GPIf?(^D}bSGrX}9c zo(&?K!d&?9mCP=GBSo$x_KU zvM5b&T5yI&g41cimk7+WCcX}&t$CeymF*M93zl8RYw-Y_=ls*Sptk(#LhVCt!2Pd| zlxJ1W@ZcP`TnWSN>vyqDQuxR+TocA+qE<|3hrNV=k|MUVVg6coQtxaO;D9M@KOVWV zGx81W_r)?PUHhY`&&d%<^O? zcu9N7vZCu}9ddY&=bp%zc1MubJ(S2dnCG}=?Y{p0JVo3UnX8D;akIVkZa<~Msfc4R z!ScNC`ivN(5R&N5WI1!~f1-g|S!xhGNJ38j9%XLzBFiOJ#4s2;e0CZJw$tAHBc+wZ!VZTej8Q@%a$$- zmB@4`6{TtZ6|zjq2E;!-2`~q14?atjr01x!S*#HJpwOLgd}#8V^V0?#&NAov%135a zeovC7q}y{$xQ^Jn^@~STj}&vIeivP01X8csNwqG#By+=hjpi#<_flUP8WC%++usdz zXT;3S+rz=RO?-P4X)~9unls$0=J4s!?Lv|cZmFBG{E>dB=to7u>ByJ`+1*QoB>o1+ zq!rX1>N^sY>-J{R>HG}?NoUiMkV(zoEQj?itnfOW zl4kdl^BvA|xoqQ{Oag$r-*tm(>7ut}@ITN~mL~cWsdPwDFMe^{DRH5Ej*&qtp<}(W zQ~pIjP2>VZ7x=oGe$09cvKX!gS2K-HbM9x)ZZB6a2%cs?|3($#QjGe%mD3L?^-Q?GT-V7vxKECTusI{&~Q)~e~kcvP7 z5svX%)y82an~WdIq&Zxz*OoL3*lk)ha4J6415X1aY@ROek4L7ClXf0=MZdCt zR*i!A6p}wvMzZ&Z=}_H=kEPG-u2XDOs!hGU3mC8a+Dwn8TFRV>W4>v5o|N-U6<--2 zw6-ubZ%+)aQL#VQkT6}KhPy0!^>qtA-~z)tb-<42UBf=Y?jC?ZWg$4ocn8P!-9nI+R}8fmXjB3`!^;oQ-2n7{YB3;t_5o+@=V%k>|uXf4%xt zsR$T)AeELg6ug)1Jqn1AUY?u)Hbl9|lK#V4d0G%tdA=nZgueuN^$VGyn~-mAl{hlH zvuW3>KlGvXzeMa_(#X}a=|q$+Mm2t0hfj3doi5AxR5+e*@|t?!#S5GgjbQ#rZxYSs z-JY*wodrWNUqrLzf@O5255FQ2<$zYmPYZEOM~_HpVbf>(xef2d%oim?(>%cXKAP;& z;1MsLQ<$W!_qrL{nJf{T*_T5)rfc0VK*W$N%$Ljl0dU~otmdBKCrTAmxt_7zxAYtu zSAB4>D9)~IE0qrr9`Bz3&4kw|^h;&+ti&FD7mGeIl(ykNQQj`hqC|zGvDz)b@w#gX z{Y*-DwU8n=KlcVRzdPctS~kO?A0=uC5@iC|nD(mB(bzK9`{v)cMc!851Gd}Jmh(3Y zP}ds!^MSP-IyBp3i07b5lU+hj-jCFVsRm!{)Q@b=6q zlek70W~uy87)*OEd{n+#*64#~cmt<*N}l;L*#3PI

1A>a5qX&@l*#L+jVu@4Gf-`4KbcPRZ~l9_$xzUq`f2$;*2-D;~0;9KuuwZQgIQ zZPC~y+3hrA(9Q%+u$zzbSPqx-=w>`mbu*6SV=bxb3MIw!Kpm8^H@pCyy|2otqtl2# zW$h{9jpk=W2svfMDzY8c!|ALafgaQYTQuHfR<~wOv(Zodkrc&BS(yl>)b1)Jcf20M zrb~}F*Jkni@RqLL4||kR+a9Irq_G$v-^^~rL+e(%@H1-acs*p#wkM!ZdB>1%dM5k5 ze|$3YxVXm?g!P{IYBTM?V)Lw@!9$Eh*q~#-q)xl$h}j-uI{_NeqlOT1a4fcOS*$#$ zr1_h7Qy99x*aZ^#DUF`*&Ydk#8|tQXa4!`1_pkCFTsocBKXRD&D?EM@-kx24)O(rx zG<9rs`>WDE*HCRxR2cLX>z8@1_@_0#UPYNg3(FwUe+{~;J%WF*4^+5CB3pnY9mxt? zTxLHYc6aIibk(Uwp4-E%-$_52F*A<+iJNcRB^#@g%4a0&aN;0OG{V`Sxj$OriOk~r zkATsCg0vsh%w$qHdJW3_E^aOsp7&xZ+)K_l!n@w2Sw-vk%x{$;LSo8VulX5%@S!p5 zQn)I4U9CgioX%A^NOQ8u#ZiqOf1j&mXQV=yJ!s|eoH_0yk&2KelsR|5Nn3{Sp{NF2 zev7z2%}(RtyF-i0qg~gxS#TS4>6f%LPK1~>=c*5LXfukhSi@& zvNo)0-{Uj{xq+v198VvV-@pu`xnF#M?2)&mSn{e<%aZKKAsfU1a)l4i$$QY?2ZNHY z0ifVVz|Rh~&{nK4SYpMC>hI3ZoV_upw;k6xMgvc zT>Zct`IlH&F~j}&cePW(Iguod-<34`kfi)xCDUXOPha)uT{I}@jW02L*IS~x=Uuk- zB9X((^wu~BEP4I(CHVXMK%GDsH!`(9T17CuGQu=mOMh!LH{^};sVQA;H-Z0S0@{RO zMguD8cRcWM1?lrnjn+7Y$|Y~&51U)b6tRPOZM{ohPXiKWfNZ#Z6 znV|NyZ(^G0!xWO=zbX~!Z-gW9#lGt5gm4JllQGB9Y`t;RJs=^)ESJ&v(4Zre!Z=I# zUOg+B>KB<*NQ%paO_R(GmQ7;-w|f_I95Oy_|6YU15Q&Xb%TT%j%y(%)Od;}1oHMWZ zpKmE56*h>QOda)QPPS6V2{Rc=3F}*M80BMM!BeFw1q5v&Bhl7YoA6jQdW}nsT35{ zETN7UB zQJyeIIHJd6G-b+HJiDH+Fg!QDagdL{G}#tBM-;y0vh|hFYIs)svD?g5>nwDqI<&A* zyw2+81J-4=Xe_xue4=%+lx1-ZnZ<@$!0XgDvXkJ1aOJ~bD0A})9qtR*Nf>`erjY$S z{#`_yUz=z`M26_U^A-OR7g`gBlQ{UnfsX&$L(%4NwK{&=vI80iL+WW0$NB;;WZ!C5 zQ$MU&t%{D|OkLuPzUm*~YmrDmtFT$2G~gx%Q}Q}xv*$2iy9IQ+ckGK3%mP4x{I>!u zTQ@FF%K*{G`7CK}#;CJGPL@M2Bh1JCi{|0KX%Hwx0eM4?#Mg8J3Sa#FWcEe-6&ac9 z09p3#p!J-`apA;p11TBZBTa+%7#zi9q{YNM3fbxZf)f4f6J8d4!C`SMYdRtZ5dl)F zLg9cP3%T<=?)=408B~lCLoS3*?z{)XHl5bj@3Bc7OIviIdd=V85F%g=RFE<>awcT< zvDzckK6LV-@ej>271R(=ymCJchL@6k`v3$!uF ztdVphzQk0oI-if*{f%?y^b?8xZ>0U-*D*u9eiJhIM2rjnwg7XbnZ+)q3r z0F4d3lQqdfdi%D5P5bO+GCY=MUvIw$!9XHuN@picY40odTnY+|mvF{~!FLg7O)wdl zx%-9ce{RRm^+itrmjiNWYCqc;Q!G6B&cHhLiGtE&q2n?!yCLRt z4ECo7Wdfyw5X4R-SUhg2w80=RiWb_QkI9CYTuBa23cqhPONNYv<;(l?KGkV{Cb@&G z-zro+EPga$lZo>Aq}=XeLPGi#E8P3}bRkY~04>Hc)$ck@bqWy3w+m#`ICN{HYde&1 zSBWqmQ*#IXO=f~Q-#N6sS9boq466vdUhri;@L_n2@g09b55wSi%SSIrmloGnM6s*( z+)jJT;IU9}CFzv=2VJK-c;k~LXJUnDduM*W^%5b|!P0fB>LE;4JJgFD@oLM-kqAOh z`GWihpWI5`K4}}BCF*d$`q{?Ul#C)u2T4T5cZk=U*J{yw`eg!R8qMN{XJZJK!spL@ zvHG_C?5ycAaumTtQci&~G@3c(Su^VaIX{0ez~XWArR-OXJJ7h!-MplwrKJ>J-ZXF? zoM8F=5Fu~;a&p!H4p?lh`{lNm((lfAZpz;X*4Es;`2_Dv2_b!b`-Vy+`JxTWTK#f1 zhh)uKhLb+>5IbDA8?1SH_xBXOE^fSzeZ>BhyPM;MtVfS9MvQjElZDaa1Q?yGm_2*0I zo%lPg9WG0pbuKXX9GaP#-AZ9cf7Lu&!-p$<87-U;e(Nu^die>P~XXneWnS2%;%?h%qG^@Cn+UGtHRM>c$A|r`dT)%|Bc&gR=(9w{ht`p42bxP z7|2^gn=!VT%=Y!1wWGR*A+LO2;K_C$pSv|s&c@-7+DM(^4b+S9K-aIYYO_4v`Fl*3Z`bN-ngQaaVsT$<%1k1*z~fYrFTqyI8W{isp)dQ~#pHY;zF}cYqa& ztIblXskC&EtEu@y3asO|C-$v;Frs%&9f&YK3q8^D*b@{jO z^4hv}^UoB^v^(^to888)&vxXs5n%pW)Susig_|^pXC#`b2?q$Ht!L~r0DowWd5tV! zN%uNedJ1G<+8DI1$108ZffP}$CU*?M-%lZ>93_RXfFi9|CXuOsLwY#-TeR=)**n864TaBXC?9F)_Gu#eR$Kfv}8l;m2at0o7xul%d{XfF*FOE}ipZ{$$$Y$Z>(3e?NZ4T*comm^YMcCq@>)ZC-x2~$fjYD=hOkWF#?(1ck@B$q(&Ke z`v+}PUva&9KOYD&+>r-YL$-2CLFTb;ls?F38wB%~Ger7mFaCBt{7g}DfKVj8d-waE zG=^w;IdcxkhEZU&COa6Wc^>laa) zPvUVfKi=GAJwI#P2}kU|2h9Hp-1*1R$&!?|2go>+9PPf?*`N2|k0N46>>BQ=a$1^x zf*HA>^m1J-*F56;Q{q7GJpcq-DCvabb5AvoT*&Y|j<#oS<}q^o$nCtFh)KyLkB;Bc zbh97M;{w_LfNspAKsAdOPPsXM<5;8!v>x4dE1o>!b2ptZ&nVUMNbXeUjf8=W3{tHN zW?~^uGfVp19zBQke94us2>bhpfcFcLW`Gud0xqpsxQ9zCZZsp6i9>9NAwkQ#^VyhkpOG5dI}T+GDRKfnjJh@&N1F{W#PCPjSN_ zf*WG@Q}`{gIDUpWT-Tchp#2Z|B%3+Suo%whx2H;%B!KhCwS<70VL%=KZUVYrVOn`NI=3wY;Z{J+?>`3VMsTflBd&sH8`S0=f4~d90mzxo|0m!zTN`dr9~&Vz-U2k4*`O{`C@^` zc*NDHXnAEICE=Yl{!F$Ad)?{jr-haqkuwa#h<}?30S1IDg7)spXGk8!&CS06|HR<4 zkfPSp>JIANUdd8_r(M?^r6rezB$5sLr_YY8b%bTOp-?D&P#xOH7vcQ1W0u?rhbI=X zg1dcXgD7fh3+&8yt*6TB&s22VI4ni}=a4^S#^`#xdrSK2tQJr+N5)X|^SB;z=hd~% zF5`9pg)D{Q!8rGfz>o>m7DAV~62D$+VBNeg_1A zymW&OcBiy!V>eq2N({Uu9Q4z#Jy_||G20BA{5p&Yy5X+IWEXAblv{hbzA6^Z)>_>F zZOS&ij`<=$tDvb_t=I8bTKH^2Md$W>jv0c>y%sguP2%A{a|v`ZZ?B-ofCT)#29T_? zal*WYW}JbWNYDGpiUa^bUcc<9&1!v0Zu7gD)Y}h0A0MU@8Bqyz>PrXa8*z|Tl<&_- zNojIF`w3W+y}$VnoqE0+*0vOf@e{!tuI9QPQ$_+%#j~IE%$w&+J(N6kW)7FuP3#PG zbhpq9A4hgT!6R$$LL8u4XWb(BVUXi&G_lZSF%9TkI7usH?EoSSc4yEz>%3>KMHtU; z^@H0d-${gVk=3}r1H#p)4p8K;szv~lP9%Yyow9OXt95bb-3|JibJc@i6oBc)b^q6k z#9!n2$+y^GtrZZ#!5>y|MTRZwxay7i!kS=qbo=ulMA$9)J<+HEbJMjvt{T0TDJU$B z_Js7;EJ?dXHVG<`(>TkusPrkY#mFab1-aH4rdrigg;=2}_b0&gfJ)vJeOWVsJ6bT& z=mub;E{}AUv++dDpBh>Yu6x)>DP!-|5ft=`(*s9mv4`ee>2Qlwg7sumBXoZ(zfJ#O z2$GWD@Ct~D8Lt=PxelBFOYJqGpbK_#+Dk=uEzVd#zHfb9Bgg;dXCE^^>9!jcl|h-@j96VX!sJ5TEu}jE+{GT6Y9Jn2qJ_1t7`pAZ02&))Fh08UR8{ zXO3q;I=A|VJZ)l`c1tc3m3t4D(NB2L(e+(J{B%xBO2w7lWM^fvU}S#~B0@EDyzV3N zHpudiW&DYg`N$+mB>HV;k2p*7+52DB9jqjf*Y$-7eUO`73nlgN3f#x&B?IzvR-lgl|(UbIbPN*$)C`yBzqN&(m2 z-Yzh1y7LeWVAZO;B}E!U{11CVaR2{O;~z(KC7REMDDy z+-ia`-}JwG0kqSgXH6}XMqLpERV|g3m1pIiFKJ~>uX29|L>@rC)n+#K z_{?!{`iREqKt-~p$OIJ>hsCX})2m)s3c~nx;t%^LYzJjyzCpqBYy=GdjZQnxo3@4k zfbg!@DQ=kcn~t~Ic5w&^yZP$%nGb+HKCzfAE|kl5pQ}wfpRY4as^h!(CFOPOL~g%N za2*dyeOvm|RVFq&D{Z>pJWJ%Vh(OyEkvk;kj};6lhG9I6K4YNd>^fv_VVl`*7vu3H z)NbCB6@dNH$dl{>F!wvmZ)fIkT8fzK;$%jeuFbEBS4tz2snoC-dyOZ(fDA-e+r9d0 z?H28pdA8%`clZd?&x%1Uw#_^&3{0qE_YH8`w$PPX8um5!TsnYkhEL|N9}&daXt&$H zN+Bn*5boJ$U0?T0z;n0*-^VEgXyPrmT-@i+u<(*Z_KMf_(<+pObBAgYl~u+V5h9TXX6pF22S9h^d^PTw+PQM{0Nca16!Zz^jh zo7ph4JJPh@3nzLoYta$p)887#AJa8a9aB^d4vQdm>s$xu>i9HnGcu4q2Lg}RW3rJA zd$l{b)Hg4b5v{BW)vDqbiycu)o6L6@YZPde>+ZRI?s+|!eSv~(Eg-$w(JJCBwr-kO zvZ*bK33Lc==XO|_cnwSC=OqN+CaO;fhxnqh{o$Uo145qrX1qU*r<<;P>-%cmtxXI2 zfC*J!5^KJO@S+pn2y&*kiy`=RGG76C6pZ%nG)9 zv!lt(ovOh%Q|3i==QAc5cQy0Z`Lps=($5a+_U3cs-dVgn=UI3SD_4>ZHRo6HaZ(Tx zJyX7L-6>;!zZ};N33rf@agL!H*p<34OXqF7xxcHk$p`_4T9vjWpL{Yt{JB(mr0K`A zUxL_!8t8~@2zgc2VG9!*#;<1#B3n{TW^87rLtx;LXfoFsJtd7u(+xHn9q=?dY*Xp= zX)fzzjj{q?rSeoF{0?nE^iS|8W7j0=I6u`F54Li9w=Osbx0SIdiZGs5>KH_|DR z8pmm&Gav{GrOm{AGstdKqn@Lr-iGhiQj{MvxD#4irhb+XpUv82Vb#!C%%1zl(?S#n zw$JvIuD)y|ePpd-vaV7&+*s`rSw3&g`syl|HU4U^=MLz@sJ%*tdCNMHgr`2B>OH9j zRkJBPqmmeMqgN|Y&-LostBdQ^FB?PvXP0rDGBGxrO{y0rS4NoWh6!UL(7IB&h4S;M z*FoN$NClF#i2v*~srJ`*f=1q{a&s>rH1TgU5fJ)?NHs%TFbgc<261PlQTI@k zs8%VbembWYpo3M7fdKM1G?|YF<;QyI-0R%A#?DC+!(8ByY22qejg4h41*$5|R;mGD zM%ALZh*+CnIl|myamJxFN5`w3a?dkc@WW#&KLd_+zQff@Y)0*zHwD^I81E=yeGkh#bjq;o%rj^cPF* zj`0H7)y4OoBwKetx`RjI-HD=*KaXSnjwIf|HV$|1$I`l!zFK0)x|j}CS)eRrpH`F^ z0eVU=Exz9%>|6 z9PZ1d&$;ur`_I3HtH1Hy8YPtuN|qS%#_JbjPY5Tl8}Wq|uIA7l%M-c8tfr=i|17q& z7UOXlTcgz^gSl z-lDw5b+uWU`3s*YEOmzli`F}^@6~B1>^5>@8n;zvdyp$6s9s|d5QM-;$OV53= z^Gsz%NgW6T7V~8vj0e=mi|SG~)%;p`u@ z)YTjD*2c6dXV(lkicZSfm~M8fWOJ46p-#ok>WE#Vfmp=6W$nV|W zOOv^nHW=zaw<&u<7`EW_@%DO0!9Z;hNdEvG_MtafcpfT6O(cb?Q1y$=X9Z=1s0M0u z5rM#jla(9wZE!BQ&YX^outGdeOM^>*+ViI&vP{re+j_N8l>BHy?GhxuJTK-6y%*(y zw()~BeaiI3Xaxw2nN1+T!2>TZBpmD{$FvDZ-htd$xb*5mhL^LYHVFzubc$EhUIk?j z1Zc)|8UUiO8o@(bi3^0;tS6s2nRnF52I0&+>f+)YK+KFdJulqaMHp^E4&}eptz}=1 zaZ{-SRG5&jlb5RT$`@lcEoY6&*ux+AGgi&*c)HHB#rdOlv9UWxV+_dO_ri}KnIEqq z0r#VPJC{D?#~Z$*&g{N~8*R2&mvD=bzn9AS9gop|`iUEBdlz8TdM> zx4qP)vj^JSQ#QIjh8U*ewUob;MIyrO^YupyBRwi^%0=2*Qx{6~D&}trOo@cNTAXcu zsV)l7)y6Z$Vue2n%*dY|Xms5A+U~o_=lRaQ9G>mrwm>c+#b+B`Ffxp7g^vRAY_s4D zOnl0{eZ#H7pu)=lA zl#1dh&QYJ=+m$L}nYx_49Dnb2Mn8TceYTA}u1{*}@?Xy=Nfu_XbXJl{z{8U-xje}j zO;^fCx?b!PbU%+B{yHErKkpte&-P+q$2|4JK&Vbkeaz&%?HD_{LsF{K_DvRYB@uPM zOm19s79hjl(iZ1>zWb1?ep=eNvoJohPfVf9_SuZExkMvDl+xr_1ja81Z@7xr35`zW zfvn~w2(D_IYbN)UbGO2j78+wnexDdLel*Tsm>CW^nn7xiQ4*JfJ0)9W3&ZfEfIo`q zGjh~{SApZNdQh0^Ukt;Ukt>T}mt^?d2PS`*&zW+(c*aJu()!OFV*#=Vt1mMH4ISNN z<4@!8m>y+eH#(Db9aU9~nbclXzcsAk;ipA`nrY96xJ~Nuu)nI}%J3fXtFwGd&Qq0s z1DHSyRh4(aUio)%2NPeSOfFlwla+B`c(u2uzkl+!fT?aM=lwE~;)n;xBE|`OYcpk+ ze~&g?lqAdS4Q4ylLaC*9=bSK`Fkk$DZruW?zJ{B|ERv2uG)ZFHE{NCjRf(SZ9YIm= zC><=O((mE~ogbhKW`E;^$(l?CkE|*|<{#^%V3b8R5y$h4+M);VA*KXM1T3>KiH_ zVJz%)2Lkf0Ce9Fx!I2H(O^fOBXx5`t?y8M`x2c9S#y=Bmhcmo;^qXQ5Vc7D-=e~g= zB5-;5&rgo9LPTL~w)1GTAI4t4z}4na48KZ-SE`z1+2_+H#g9n21pQNWyCB4Ei*`#2 zoVsqeY2%Yd!Hi4#4FB(U2OKC5ziEWSin8b8B4kDgZ1^oflIFUe0I!aK@Db1Jx2|D@ z$a~Ni^2f0KeiOj`)&{ZC(PwBqqKHoPkqs^9E7?uq7XFKF%W-!fFVHA z;eqHq+<)HJPaHB|=9rLw7^C0c+NlG!<~`K+e~tY<4|r8cA}{{y=KuRv1<3Ayn_>ETZMr;!YA{MeBjRfUt`D83c%=4Vg9e_2x$XD-wwgP{Lis7 z-+}pf9V+p^rsEb2z3PKT^S{P6LdIxMcq{V1relQ}n8W`szlzsC-wA@J2AxuMeR2Et z`AwQ;)+))}X&dLa8rUtXwCD9<^QHQc`915Sk!RYKlkxU{+|GZk!1ujiKIyNJ^4tq) zmYf?{=b7&8=pC88;=~SSH^L`d|xKps7#Ndn37GJa>; zy0b|`m;+y_{~;^olfQRCh%Z=G@EE;tey+%CuHwxJGxF)mLwjG9XSoh$c8@I?(6lYr zwQMFjgwD9L~f0~gjQp=@QrDX&p0VzqQj#wIy^@&#jwS!2^z z8%$me-Y)vEkKSXl(H0u*)G!MqXJ$-8XjFA42y2UfS?*5NKi)kH*%(JNfP?{x|)jSp_|@_OL8crUCLr`&(=dFef1yypdFa&NNnC< zQ8FUfnwBEbd7abrS`&5Ugm)L#UxN0}<#Eawm=cxVH&Y2rxt9>MrJ3!0=eFAY_-~6M zp!%ow{`&P<+v%y}MO%@yN@g=jWPU+GL3w6*c@$C&-BLGGoOHX^gzS)8xXKVOIqO}f z7M>3$JG1uAktNf4(BhoSW2n*?&O`o${Z04@N*f(Rs>Y@k#=;eyy<6?g-2S^>ynnol zYEb*UeA>0IHnX*bYliexaUd&e#s8MK>&OqQoY?OtbGgIhn+$V0oT2@Mz0%q~w-VBx;nL4be zv9YnUOoX}S`=~afzS`zr#LFV{?s!q4?k^T@o|L4%JsX#hqVZlp!IQ!}okn3C;dKjZ z#|=@J)>Inbt8R$2){(pHH*o~yd0__G>-(*wG}d1=Qlxcm2A92VBt9{7ZmJqgBFdkC z*Lwf!En~>}6~G^c)j51sl;h^KnQb;3`+>`dklUAW>ypl>=2v`r;^>fxuPP@~s*7NQ_6Z8JKPqZ)LL2gS{ z@+lob984?l)3>=vk2dFR6A84vwp=Ws_Jki~5rZXO_*o|KSR`z?xYW|${33EW`y8Im zt8)L`t;+{KF_0lRqZ8CrZ-Br4j7Gtn>P5XRqbgQ5V4 zD6KkzP{TfcR4LWu^SQs3g(T%K_A%IR-hjk$G2mJ^GB(8mEYF;vRfJSuAIzOVF23nx zu~~NnF~B>VYycuS3fLWw>#h8G@0$|3>2S`4+gXdU?$!VxB3z(cat%~Kj0aN*Z?aWZ zD4jQlTfyIaT>{-X(FqClbv$+t@Q)v!0-=t1+YTqMoAl^-<rv=_kC99h@@9q9XXN^^d8TgleYN?jVzr&- zk()J5r*!lxe>2pv$|pum_g10IxAmDz3O;*Rb6V+Hwvo8Tax*RZaLDr@VXzT|`uuwL z9^8|b$KKgTJVxZtRt+FcTL)-Wev!Q3ywQl{eSIBDosP`H(u*OGIf|0%JX;8a%oDx> z71e~+oBbvv)x{)MlWNmt)RTC|n_P`Lk_zVKC#3x#jsF9X)b1>_nCwPsheo;{w76ur zo7*Ur3kOqjSWZWkvxnbMzpC#Cg*=y^v`wC`59(J-q3J0^p3DoVqGX?0`;%792Hozm z><+#Vwd$L_+t?7m?IUmXsIm6ijHMk?DrwvaFgRkw7MFE}mCduMPMu^ZaU67ur1x>x zz8q;HXfD}jJz@Eu+{I_kOyoN3>QD^`NkxR-#r#W%>Ca3brL&uJdcbvJ>uw%x{#Gi|DX2G z{2l7QfBVrQ6-r46eNu!_S+diDY}rSQp;GoGJHrr)NOs+ zVHnHYudeU?y{_xH|AyO-4s*?M#mMb^ZdOM54e>Xv0PXMa7v~s zY&@?oN+7v^JN$mOk@O?ainB3ayuxF`AfF8&$VyRt4%w}d1Rrjzx%Vm(j?MEu09tN$!H{j2o*mmTM*b2=;pL90G9+x_m zT7M=$euo&K_=m-5E z(AGYJmll)ZH5<_K4}(t9ZpVl<3tvdYSI^gNgB-XV-V{AqGQ!fO!TJbZ9r{3bSMLS& zh{Fn2Q=!n?i_DS<9rU&IW(t{)m8ZvyyGHwvs~yL7#Jd(K?P7Rj86lN60@_>%JoSSZ zL<0U|8%$Ronazi(c1F*MAfz6m^uSvbM`qn}a&lS&>7*ysTcKdNRL)=o%-T&W;?d*a z40%1)Co?+FJCQdDKjK6WZ)yza%sTK9-z&ySwijZNwW4*q_w1SEw^7D&@`%NDcfLT$ z>k}C*HU?BO`uWpz;f_Jdxp)K<=yAM^WA`(wd1k3N*)!e9Y2rHA8R;cice3V(rvg1)TQu629!ajmb0iU*x#O1cmZ(lcEgXuubA^?nCp6 z@7;ZOCr<2WOeZ7j-VE({#%iFRlt}G<;g+*mOK4=ap8$opr5unzuw>!C=Lp&lgXZrn zz2do=@fQiwna<0Dp`h)(8~7yN+^pG&E`JV_kURc zu*?tLehj+T?9RqpOnO^V!6NqcGq)7+FHbYghSFUri$oBz@_@j5(OH765umxi(^qbcq8&7%RWf- zH9*xC-!Wbc?)X)77mm!$a2#uNw~=H6+3@qVKB-y>n?Xd_&v0MUFkQd>RWFc%(MZRz z5-|0i-F>NF@9+fCXFgd|4EICmbJ3};_fL{M?m~Hd9o(b&wflv<=Ms(&pnnP^SkZb; zLb$}LXL`Y~nwXye)7zbdqd>GlW1T|#psn*@xmTJ`Drb>je+%DUTGZGdu|CVMjU-Qa zX(kC)s%+?hhf?EtNO-`&4OpOw@$p)q@7vd}ZxQh8eL6}bUK`avmtt&JB3Q3*;1i%+ zXb^^Ht-CeGbmdKU9XF?>UG8RX2N(vkXT`K&CC?wMJ9J`aMVZT5g3twS_158pevpunL7g1u^r{ZUy5i$5v1S-YC=^jU55NxTV+MS2RiwaaajOeq(N;!?^_T5EL$qC{_rcDL^ zrVGh%${>c8PTd)LU|ejzH65X3$GJx#((27iHh(O6;?yFX4)-ZTi~~a(pz>%A|H+_ie;xcJ2 zpiKJ4v!&Tt%w}>e12JgrlWDn;yjmLrG=ie9ScemYvLS;(j? zS>;Tbir4(y1b9;dNg?v@I7{_f^Q-u}DDy$#EsM1UUWIit@}J`%aK3i^>zd&Kza<}i zt8kjk%?=Cuv^P%zPVzqp=bS7VgbNP)t~czvbCgysT+if4C%gURKor$(B58-D0L9yO zQsBwkvwb!hlYEv93jI_L(Aok=nN+gvbyLJHux6=+f&-^4$cRhv{-pfnUciPPy> z{hmRzFWM`POCYy4?CCU138apzjLEs%P^Ln8$1VbrKyh`<*LiunH;YS=uJ%t{^ix?q zmo!MiCKt}aXO;~>b49V~MHR@28ed8g*nO74YO`d(%56_x2<<#rpA$jHq63!HG%e+K ze?Qt02nj4I{kqXQxToX6x9NK-cC&UX1m-z{QY11XrL~J0FY=EgD2T3czp9vm=kOT$ z*KJLz;hZK33$5?{CBlXQJw-Mg^^vBV^cYH}mY6s%Zl&=$4ry85C;Am^UHB}1u~R3z z#4Iy;+ofi;Y7BG>d3|abaHgZwAHC`uKJKnFQEGN-vogLrSADxirJcr~Az-Fl+o$Kg zS;{WY7nXPR_N4;MXVHKa5oot_In_BpA*DMag&yEog~CPO+u)IUqR$Rt3!W3yg{ALQ z@vBk}xknNJcAoP3{LH7F_S0m$?n1z#}HwuP{H}qVg3K9?dYke?c9I4o7)2UsWs7ctgc$*Ee-XD zu6A?)y#RAn`BHeP)Ve~rTEnK-*ZNp-nGY-9TjQ&Wo-5Hn8j|9r3*R^bNq-)E!x<`P z5%29Y>j){{FFR~=wl{3Qc<)Xzcud3zSm@5eYDlf!H~i zLhIn8*PYfqK);-@v&p0;fls|lcl^{Ij-#`!&70iQ1BeyFN^nR7=J|!A-d8BKAut37 z;C4<&)vmKb3Y(aKOejTkZ*MkjNZeaE8tAj4x~RC@CNJr|-7H>zeP8aJc0Ac@vf7Fq z$z$}Hi`^jut1R>PdMWg#UF3e^0-_TsDTY7zq#8Ll^NQPZcnU#H%08R?(pR#G+Vs%0 zNZo6OzK6H}GlzI((KnuT@KH?v34i&dd8ej7H!R#hi@?10eOd~(R`DrobXg)k%l~d` z#k3=4B4bAfM(2dv&IXfyq#8?z8!0D(f3CbzFT6&P?f`65FK!Ob_E-xFdfe9~7oDh) zbLCOF`^`=uJVBh6b2hE06yBZJ^7v-$-dtkc5j7ErOe3S3>^17q{J8h+h6I^P}6IYbYJKu23Q<{6sD5|5pvMtMrUS(V&Ml$Ba0Gssi4?%rNzE) zE>aX$v+&QCm%AV)bEB+Xqb|RYZ)tJM5#wWAA|@(&^gFnE$tfFW_)t=E`Evm~G*(C^ z-ptkKRzNFQA38_yDJev3-R%W)+2cr45*7iy9Del`P-|oFG+k0qO zfHkcJYk8`|&hR7NV;(AZ%hi(o;Yte9C9aUaEinpgA(bjD+~j}s%+LdcfXVh>%F9{k z%LvtgSGW(sC`j1DP$PJ% zf$SWth|32t)LYvf&8y5_-`wEvdX}FJ47^!$p#?=Xqq6XIi-)u)`0YWctV^3VOM3_W zzFldF;+3V@3kS-zTp6I8^g%aD$Lgi`Hix*IMo+xrW>YNxuZpjen~?XtoH1sD3S0^) z0Yby?6>uu>!9~>8Fgwj8NeS)Vjg^6HY`%2H_;{pvRTl=Y{aF%$o;-VgWGi#c6GASU z%(wh0|Duw|Cgoqdc4*&3YpXyic@QdI_^4oN;?v|d6QYsnK9V`%Q^J7p7WnSUFnhyi z3xBoR!R|u)r^#J%B*;v0z!JYI>{&5(4Nt`mvhU^Uu}=ojj&~vRFQ{U)?f!u7W*@03 zgRAi&$@a~beR#u(6<+wlg$ZFBweqTU754C&h3lk3SL$^`F*)my*G?n$#R_PN_PE#% z*!;LX=qG2Dx%PTs(^;1OU3u}=`Oyry{J`S`31fzfjiZ$ix2S@wP(Y zf(Pw)FiXc!G`*|uv0{V5i|C0?bz@&_9ciUH!CDs7T3}@}zU441*wSENbd0PoLO0eM zBfw602#?7;wGMU^@?0X0xHva}A#v~nSW6UewtfJPLq3@lMeJjekHHoq^gC%r)ZEzB-HSUQ zJ`C1=`Pap)r};G-`V>tjZvz%}Um`xwz@pU2zTczI{0Y;Z6wR=pEAk&}XrrJ60 z)gF{iL*k^-wfpW1=kZw}ZZ%#UZmjV>e(6lsy#=agyT~Nh#yyLgW5^JSRVQ+EpDFJ> zXu#Q6xqnf|+j_7Y@deKCta?#B5NW-wvt2+tq{jiAMGeTj&z|iCNb0nyTRj!89KkG zLVXh#+V}N003_@=lzD!+eKe=fzq!qsO+CGs=W^3ZaC+SFRJjagX_vhx3_|C|!r$WM zTpeECN)!I_(i`@>6@JxP?iWKL*DyeC2!?P3Zc6n_4~kv%(ZqC;Fj zsJNsp!1m!%G_HEz9Ih?IUQ8!t^~wg;XD}Y4Brdds@OoV13t@WOuQKF5p*TFV8EpHvwYdkwqS&*Q?g~0 z`{J|O;W76@e#BNw;|l7z{^8cP*?@ILjE}L-=?<})^Ltls>eDLVqAS(&DW9VJ%8UX$ zjOIgm{oVat48I?PC#5GCBx9GKSy?uQpe(1i-|jYD)al-dP&%3cHJ3MQsHx{xtPqdS z@o4ctw=kJV&Yi^3II+7qyE&=;Quc3`k3VNgN3*;=%(Cx&@VdyE0Gc)p`dmuZ-Fos+ zC-aMRhWq1?y88p?{6PhkilAC{R_WsXpLlG@6cGajkkdcODIogVJ+02O@Yx=k?`gsx z#BbIH;{@Jw+QnDere4ybm{Ep;DL*M~U;kWn$*X(mtv#WBj+gx)OWG!qXDP+e-}HAm z)f&Ycnc|~mP6wAdl;IKAoDBo%O{XNqO;`7I?0|!3S7IRW%F~X1f4vOSd_J=ivQFW37{gy|0K8r~*xv}*eD$D+B z5lXd~Dmb_<_EJ~JF)yKoDT<@8_P!HsH~1Q*a1Zrb7n!Ur<^v^!8Y&QZdlMuW(v7|9 zTBIyJph+CB6M)~rHtA`?g&3STT zjad%9aASTrAS?J%2!b61vxvt8Yi6eJwhfVw$o{EM2wj ztb9F`vh^U-%Oda{{&_C9YJsRNlgxy8rA?1MLb@~8mH27p0OMg@_%tVQG_dkSvYUl{ zw#DIQ1A|)F>|tB@Cc@h9fms6^?fK5~cQdxh)k``)6C`d(sa#%pTu~r3IZlyZ!8e61>|s z$me2FVu>zoUjbwsK5TqPZzN&<*wRz~1S*C{$*2ypHu-w4(E*pSEmn>)8Pn zGp+$n=MY1i*N?@SA97WORV}kKxd+U$(ZmnHh%YmLv^wgMtfIdku)h;ZDlnO&jHuy; zJ&Cv|pZ+y14r-ioszP~HH6o>Yz1_laCP$s0oFQuvx7TnQky3jlAS>D9MOcLL4`M`4 zJ%`-hOf(5gGGAU+i?(LwxYti=?7vi>_UZLm?=wM?AMFFsvre5);|H6t8fHiKc9ve? z=pn8#mKxS>=X+me1T`=rH+jd!f@;uvu(}f%4>1{tb6PvMg2DhG_7rloZ}`_c=ve1D z^|@4k7Hh&kEf?QC(=UQSYhI$AUcp334}W#VUw$UVH6YM}go-915X|Wwq|T4JIcn(4 z<=e-g=arQ$LnJ*Zo9(zx(kB_V(~D}FO@db-T-(Q4ObF+_=Th&%;Og~RlD|9rM3%rSW!}7ZD|U ze)C1h=PM3(;a-+EA8Cy^Gu-vPYuA6vv4%xozQKR9^*|K%wEr2SLY40EN~^3PM|4%< zBrj)B`XWpcAIwRUOl#z1DP~+RbU^ErRmGT;ASE?ZC53T3j1iK-n}gX8>c9H z)_N3NIB8QY|6EVIK$KE%Pl3nvd97o-^W~Y5)1YX710Jzu`(&kYziX?mak|%FG`Z1Y zT+&Yc+jHbR}4utB6K0 zGrrtf1IG5nHKK&ttt+P`G6pIVa-Z1UE0Y}zDKEx(BHyNg`C=6ovPtDG=Hg=d5(IeB z+MUs5=Q^Ip)C#QYxvF;A1 zyo_j9jDVGDPWMiD;9M%PDXGWI$0{>?kCuH*a$+km1>B0*kL9ZhS+da3^+HPcWF;Qs zz_{LfNQ^Z$e{m)w)=E;YR5H~sve%Z*bF0X-e>?M}N76@Vh0WZ!P305&C{OzaPwTRG zqZx7MJ3|ZiiVTG$A>Yl8;?3`eZQSa_5aUE|U+Sv!JmGOlYHKPlTBMW2xV@2fr{{c$ zY6Qvb_c%dVyxd(mddu;oyeXmMwr%zl_LIJZ@b}RQ$>PwsQGm^3N18e33rDK_k9_jP zQ-_^^w;QdBJm+jXJXL+2odUT4}#96l^XvHHXsV|9tpDEf&4U8z3)skdaRa z0?fD5=OHsU+}xEjN`yDA-7EQWLrwnrUhw?Q`oQBAp7A`^W&wwl@aHf1omm8EXIAVH z0k#d7@ZuFl*K}4hbj|3hk(vEFb>eR+DemPr0{!*Oz@;?I{pWV(vB`H7`03Xcj|_F7 zWC6Y;U^+(KG<_)cvvLSN%& zdDR`beOCA%lU(HJ)ZKr-Bx3P@ByS!^jT=qaG9A*`*hNGja3_afQ6lM`wQWd6 z`mW@R0G67tq)Z1Fmw}8*2u2w#G5vWGVTHW=9vB9kJH=`&rT1QuW-y9klHY~}CQ2s{ zL1DOq#Q7@cOH>{9<`Q1}Mq01!w~WDNkIM!k0)u{V&m%Hlt8N&Mq>~M3Gj~4}G z<`2B$LV-QD@~z}Cfo^FeZf8@RP)igl;4r{-GMu~OAWOj5^o)1N7dL^jSMRSJN2s%W z(lxQZq@qY8!irX!6F%!35d?mqTIWYAg&9eF_gr=R`V)2ztWy#e?!#GJbiJB5B!}M0 z@|n3b#{!AXcupvbV~=;^cx*y?3i@06nWcSQt3TZrb1A>TwHi%sgSaJsdry|gOpk30 z<%I_fOr=2rbymiXN4_XGfDS z)6qT+Fk7sf1}cP5R7nP!^A*D7hPEOa^EbrI*PP*OY9nnO;+%}Gt{TQUP#_UkHhrYJ1b}8=PwhkxfAOLJoz-7I^v$|?u zB4ZFzi<$`x+inq?;nuupm;8?#(TY^^9EWqu^pdp2NeI>hdC^+wvGhLDPxXsD@iu*7 zN3!O^NJKN~v>`K{0!qKJY^fKp#`6|MM}Oetg}K5!9@0{;3Dc1{w46TGZAu4Z@NDWKaLmoi(?I3v5=rn$ z-+Y`uB@i*kgPv|M(LCi(WUA4(8X2_vgj{Ov67SLvPJGS@JPUNud{7>D_EQ0#DUO0%dH{ z8FC?XKu<{5AOBjjB0TEwiGb33g$ab1}FL69Q${`^hk%yQ&zzTISZXyE0* z$;E3Um<`0q@vMFdD7q;Y3;Jw_!3*q~7LT)xE4^y@4SE;apT>;niEAeN%q8r|<||kn z?&&C0VnSX%O}M9ynVvf9Fj?FDD&rLG8sAPZ?~}88TB%1>>TYd<3dPlPcFR*in;??H zS39%h|*vN0Xt(lo&AX_%h z>irvtu5kn)mW-yg3!BBz5>z>(9QQ+msDsVfo5D}?T+D@W_L&bemy7m`%NODU7>mfAs zPimO!uUWf$J~ZuxzyhSduIeb>awV)zY;%UX?7_QU61WxqL(WdW+Yn=wg<6yE58Es! zTn3N^NG2ka`*fUvL*CT)Aa}O1_t>!uZ0e66817_(di)Mh%M0U#*)+CVg>5kLf5SaM zIij*%6x(F-s*@cxk|og`Yc zTlAx;=u=!|MLbTpvh^{P{Ho#I3nbJD+;!5ed2XPn;&w5CRA+z#XcoBIeWL)cR(ico zc>@cV=NDBL+cOGyrs+WkM^nK^1skikBh<=0Ps%{jFxff*ny{LF-(?_hF5<;-T>9bG?OPk)U8YyXtKv)yPtKg85u=Yv38K6R@<&D1xh%OvV(haOz^Kvw3SDD`9(LN4l+#~DzzrwF;1&R_?sQdk#XaI zS5ZVf9qOmtm7a!33hN%-GrIWCgI7wuA9XV|bWhRmeoJ{gaUD=Jjf3J3wST+pj5J=q zwAM#(j5#-oGEA?Kl&pMf=A1NyVX9!Mu6i2>&+YC3oYwyQrbi zAA9fpJ&p6)5(-kau2(D$LY)(3w^bzh7Z$(Om2X?VJ^(TrC?za^64s!G?q6M>OLYDl z(RGsjGug+xI5Fe`pDLC6c=1(tUs=2UW&oB%+TJt>!$}5UWARpS{8R|^y1r8Q(vV2} z;6;d2jrRSpKVXcJt`V-O^#^|j!97y`2h!s178rYYmW@vFFD6iQA1Ci`x90rVUuylZ zR;1-rV{op_#v=0T)@AzETWRZ$wpvsDYbaaAPOWg0O5GEZj-!cP$IO?l>^WpA>xFH4 zjM{ys^nZKN4kw!{(D50!Cp1B4l4!dML-$woH+<>CyKu8{PD)gDWE!;jkpHx(`bZK=K|5%H=1D|iPhm`8|gMQ!|~&anrx3(#pvFYRymnTmL+=A2gg+%C`GXG&-8xZAvZpuqDcM(E2=ZP5VC)Qms1{3lmweyUMz z+~pSxVgV5j*k;EqW+xKloV`!*5@0UPMj(i{!rBIMWe1vfjmxMCluQ7SxAHTbYqix~ zXAfvLNiM2NeH#PL%xq&|DT~S`M1`DyK4LkTBdDGbN_~5|`TO?Z*s3sQ`IAe#2*$#t zc_8H&i!u=Irou#P3}Igo68!=TK>y>Bb^nIjo5<>cBKCKHzn4+5?@|5~tig>~cWB{1 zGS*pVCdRE2!CKMYHMsw^M>TyuP2hqR%K6t${d^z6;@&@=su*12 z?%6Mx2D0b;Ll+p@q=HVCGp9$pZZyYF_-&!PNMh)j3``Mr4K8l2k*2>lp=jM@Rt{@i zNwx^Ex+`-tGMFDlNeq_CKl@E_rt8F#%MQ54GF_E^#>=x`wT4g z!F7Moky3?*@_lpcD6{L!?gXTetl#d*t^#OBTYcbHt@_onAGQTE$N&)a6Tlw2^?lQ# zedq(_L%%SBMlLj`6*(KOv~edckF3l;?JFOMmKIruG)oq`0|mzDmx@QQS?zA(9d<5b z-j;%*O~sX=PWK3&oD+YHqP@4~v?TUU7|*Z01L;dzdoH08`Exk2 z>(gtig2{44%QYZsnogSYc!P7~c;f4`m_J`(S-$y{0R8@)A?NWeHE@-aFn0 zLItgS{np<>!ysUFpQ!%2=lp;*5##<*R}A3NTsbHdZHoZke#%l#2y4^TjV!*B*YQZe zlpYpGnHZAmISgT<`R&bSoaa7~+pQoh3?)b)Qd}WvK}(miFBo|&54IBsP^>2!-etTy zDP4B5dv}fH|h3-LS=bCPFaJ%41W5Do7`4sG|H?@Ki5M-93mm2C{4xgKZ~n)M z0z}FwCycE5iwEiSyPT5wDq?Sq0UiAs5^>_RAt#+z;Mkdbw8kt!w8beYr3!n$#LF66 z#xOsl(=JBE{aIaK6zLQ9eQnAhWD1$N52x@H)hB;gCC$ncIln9V?fRIOT5Zm>dQEr~ z6DIA_)&|i9BQ}ulBGx;4Iz=Com-`t92}Gd=>Nf2$PU7l+TvpfG=%btijROk|(Qzp$ zRX%-l+KHYVmK;$v{rN^%D>M>>GRjrO7w&=By4DkcH_o?sR+sy6O~D4nWgeAHNNQr( z{d9G|*XtzuB(QZz3-N8W^cp?;Gz8zwnel;+a*OHI{feC*M!w^!$d9ay#EuSE`XQ}c zgm2#-1~(t9yNt_1cX=nP+){*B$6bMUBV0<1KDX}uuixRV@=M0YH{Tkct(EO9$hZ(O z1;qTM0iU`b(;rxu&*^fWc*y<++nZl7F-Dm!TJBm)T`kh%Y_#vH4vIJ{xtHYJdyLjK ztfVl#pykC4Ww!scWzo8|>#LFdxS3Igs-Qd@0;V z)c%(nM)}-{Uil%Rq_2<1vLF?&?98cT>!hHkwS2?JpKn9-N=aVL8BDL;|sCfH_uzBeJ zWEK36qwU&HdOh2~`yRUe$qBLOI`W0WukbGRnfVJiwaA#luR8Ta2hZ%w2{3VS<_+;u zl?~)ef>peA=&#C&>0XOhY#8h(VRY3Pj7UheZV)*6ev5vmzF5m`Xha7Wm$W)udI%tv zLPn6eUM{h>4#W`Pt}d<~i6&iq`0+o+7@@qs@q>JS9nG3LIzywbY*L24b;uS8g_NJ! z5b2AakJEZj=LSy1uHmI@wR{$wLJVp{n`IGzvxc< z`RDxhF-4x(Ki^IpdeVuUE$+EtOUL^ObzfV@IXb0I=j|JI_uNxH z%PK7En<>ik*K-R0kE*m)~*n1E^ z${T1FKj?8*np5$d=1N^(eUK?N96KLb;Sy+lzgq(Qc9dpFo#7QS( zz1h-q{{MOhP;{Z+ZFAwtJ4tU&rxT?Xh!JqMK3A-?7CDdES}Xhj+Xx8sDwE|eoZfo( zZ$r~tWqRPH6!@*4{sfrG({|;E)A|hFl}$O=@iWtvqh^1-g;nq--(MK}*WX`aGz+l8 z?n5lN*+v6v=l}XzGem&#((%Qn+BJXrPpt#F=DGCVnQp#VD!xZn&-b8xTsC>;5^oC{|4u}1}rfubGar71Szw{IT j74iDtFVuYyNISmOqh}y@m6m!8{HUuueO&bLdC>m>66wfn literal 0 HcmV?d00001 diff --git a/docs/docs/how-tos/autogen-integration.ipynb b/docs/docs/how-tos/autogen-integration.ipynb deleted file mode 100644 index 5e58a260c..000000000 --- a/docs/docs/how-tos/autogen-integration.ipynb +++ /dev/null @@ -1,406 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "id": "100c0c81-6a9f-4ba1-b1a8-42aae82b7172", - "metadata": {}, - "source": [ - "# How to integrate LangGraph with AutoGen, CrewAI, and other frameworks\n", - "\n", - "LangGraph is a framework for building agentic and multi-agent applications. LangGraph can be easily integrated with other agent frameworks. \n", - "\n", - "The primary reasons you might want to integrate LangGraph with other agent frameworks:\n", - "\n", - "- create [multi-agent systems](../../concepts/multi_agent) where individual agents are built with different frameworks\n", - "- leverage LangGraph to add features like [persistence](../../concepts/persistence), [streaming](../../concepts/streaming), [short and long-term memory](../../concepts/memory) and more\n", - "\n", - "The simplest way to integrate agents from other frameworks is by calling those agents inside a LangGraph [node](../../concepts/low_level/#nodes):\n", - "\n", - "```python\n", - "from langgraph.graph import StateGraph, MessagesState, START\n", - "\n", - "autogen_agent = autogen.AssistantAgent(name=\"assistant\", ...)\n", - "user_proxy = autogen.UserProxyAgent(name=\"user_proxy\", ...)\n", - "\n", - "def call_autogen_agent(state: MessagesState):\n", - " response = user_proxy.initiate_chat(\n", - " autogen_agent,\n", - " message=state[\"messages\"][-1],\n", - " ...\n", - " )\n", - " ...\n", - "\n", - "graph = (\n", - " StateGraph(MessagesState)\n", - " .add_node(call_autogen_agent)\n", - " .add_edge(START, \"call_autogen_agent\")\n", - " .compile()\n", - ")\n", - "\n", - "graph.invoke({\n", - " \"messages\": [\n", - " {\n", - " \"role\": \"user\",\n", - " \"content\": \"Find numbers between 10 and 30 in fibonacci sequence\",\n", - " }\n", - " ]\n", - "})\n", - "```\n", - "\n", - "In this guide we show how to build a LangGraph chatbot that integrates with AutoGen, but you can follow the same approach with other frameworks." - ] - }, - { - "cell_type": "markdown", - "id": "b189ceb2-132b-4c7b-81b4-c7b8b062f833", - "metadata": {}, - "source": [ - "## Setup" - ] - }, - { - "cell_type": "code", - "execution_count": 1, - "id": "62417d3a-94f9-4a52-9962-12639d714966", - "metadata": {}, - "outputs": [], - "source": [ - "%pip install autogen langgraph" - ] - }, - { - "cell_type": "code", - "execution_count": 2, - "id": "d46da41d-0a71-4654-aec8-9e6ad8765236", - "metadata": {}, - "outputs": [ - { - "name": "stdin", - "output_type": "stream", - "text": [ - "OPENAI_API_KEY: ········\n" - ] - } - ], - "source": [ - "import getpass\n", - "import os\n", - "\n", - "\n", - "def _set_env(var: str):\n", - " if not os.environ.get(var):\n", - " os.environ[var] = getpass.getpass(f\"{var}: \")\n", - "\n", - "\n", - "_set_env(\"OPENAI_API_KEY\")" - ] - }, - { - "cell_type": "markdown", - "id": "1926bbc3-6b06-41e0-9604-860a2bbf8fa3", - "metadata": {}, - "source": [ - "## Define AutoGen agent\n", - "\n", - "Here we define our AutoGen agent. Adapted from official tutorial [here](https://github.com/microsoft/autogen/blob/0.2/notebook/agentchat_web_info.ipynb)." - ] - }, - { - "cell_type": "code", - "execution_count": 3, - "id": "524de117-ff09-4b26-bfe8-a9f85a46ffd5", - "metadata": {}, - "outputs": [], - "source": [ - "import autogen\n", - "import os\n", - "\n", - "config_list = [{\"model\": \"gpt-4o\", \"api_key\": os.environ[\"OPENAI_API_KEY\"]}]\n", - "\n", - "llm_config = {\n", - " \"timeout\": 600,\n", - " \"cache_seed\": 42,\n", - " \"config_list\": config_list,\n", - " \"temperature\": 0,\n", - "}\n", - "\n", - "autogen_agent = autogen.AssistantAgent(\n", - " name=\"assistant\",\n", - " llm_config=llm_config,\n", - ")\n", - "\n", - "user_proxy = autogen.UserProxyAgent(\n", - " name=\"user_proxy\",\n", - " human_input_mode=\"NEVER\",\n", - " max_consecutive_auto_reply=10,\n", - " is_termination_msg=lambda x: x.get(\"content\", \"\").rstrip().endswith(\"TERMINATE\"),\n", - " code_execution_config={\n", - " \"work_dir\": \"web\",\n", - " \"use_docker\": False,\n", - " }, # Please set use_docker=True if docker is available to run the generated code. Using docker is safer than running the generated code directly.\n", - " llm_config=llm_config,\n", - " system_message=\"Reply TERMINATE if the task has been solved at full satisfaction. Otherwise, reply CONTINUE, or the reason why the task is not solved yet.\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "8aa858e2-4acb-4f75-be20-b9ccbbcb5073", - "metadata": {}, - "source": [ - "---" - ] - }, - { - "cell_type": "markdown", - "id": "dcc478f5-4a35-43f8-bf59-9cb71289cd00", - "metadata": {}, - "source": [ - "## Create the graph\n", - "\n", - "We will now create a LangGraph chatbot graph that calls AutoGen agent." - ] - }, - { - "cell_type": "code", - "execution_count": 4, - "id": "d129e4e1-3766-429a-b806-cde3d8bc0469", - "metadata": {}, - "outputs": [], - "source": [ - "from langchain_core.messages import convert_to_openai_messages\n", - "from langgraph.graph import StateGraph, MessagesState, START\n", - "from langgraph.checkpoint.memory import MemorySaver\n", - "\n", - "\n", - "def call_autogen_agent(state: MessagesState):\n", - " # convert to openai-style messages\n", - " messages = convert_to_openai_messages(state[\"messages\"])\n", - " response = user_proxy.initiate_chat(\n", - " autogen_agent,\n", - " message=messages[-1],\n", - " # pass previous message history as context\n", - " carryover=messages[:-1],\n", - " )\n", - " # get the final response from the agent\n", - " content = response.chat_history[-1][\"content\"]\n", - " return {\"messages\": {\"role\": \"assistant\", \"content\": content}}\n", - "\n", - "\n", - "# add short-term memory for storing conversation history\n", - "checkpointer = MemorySaver()\n", - "\n", - "builder = StateGraph(MessagesState)\n", - "builder.add_node(call_autogen_agent)\n", - "builder.add_edge(START, \"call_autogen_agent\")\n", - "graph = builder.compile(checkpointer=checkpointer)" - ] - }, - { - "cell_type": "code", - "execution_count": 5, - "id": "c761fc05-e8b6-4905-a793-eb7522d20060", - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "from IPython.display import display, Image\n", - "\n", - "display(Image(graph.get_graph().draw_mermaid_png()))" - ] - }, - { - "cell_type": "markdown", - "id": "23d629c3-1d6b-40af-adf6-915e15657566", - "metadata": {}, - "source": [ - "## Run the graph\n", - "\n", - "We can now run the graph." - ] - }, - { - "cell_type": "code", - "execution_count": 6, - "id": "a279b667-0f5d-4008-8d43-c806a3f379c4", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001B[33muser_proxy\u001B[0m (to assistant):\n", - "\n", - "Find numbers between 10 and 30 in fibonacci sequence\n", - "\n", - "--------------------------------------------------------------------------------\n", - "\u001B[33massistant\u001B[0m (to user_proxy):\n", - "\n", - "To find numbers between 10 and 30 in the Fibonacci sequence, we can generate the Fibonacci sequence and check which numbers fall within this range. Here's a plan:\n", - "\n", - "1. Generate Fibonacci numbers starting from 0.\n", - "2. Continue generating until the numbers exceed 30.\n", - "3. Collect and print the numbers that are between 10 and 30.\n", - "\n", - "Let's implement this in Python:\n", - "\n", - "```python\n", - "# filename: fibonacci_range.py\n", - "\n", - "def fibonacci_sequence():\n", - " a, b = 0, 1\n", - " while a <= 30:\n", - " if 10 <= a <= 30:\n", - " print(a)\n", - " a, b = b, a + b\n", - "\n", - "fibonacci_sequence()\n", - "```\n", - "\n", - "This script will print the Fibonacci numbers between 10 and 30. Please execute the code to see the result.\n", - "\n", - "--------------------------------------------------------------------------------\n", - "\u001B[31m\n", - ">>>>>>>> EXECUTING CODE BLOCK 0 (inferred language is python)...\u001B[0m\n", - "\u001B[33muser_proxy\u001B[0m (to assistant):\n", - "\n", - "exitcode: 0 (execution succeeded)\n", - "Code output: \n", - "13\n", - "21\n", - "\n", - "\n", - "--------------------------------------------------------------------------------\n", - "\u001B[33massistant\u001B[0m (to user_proxy):\n", - "\n", - "The Fibonacci numbers between 10 and 30 are 13 and 21. \n", - "\n", - "These numbers are part of the Fibonacci sequence, which is generated by adding the two preceding numbers to get the next number, starting from 0 and 1. \n", - "\n", - "The sequence goes: 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, ...\n", - "\n", - "As you can see, 13 and 21 are the only numbers in this sequence that fall between 10 and 30.\n", - "\n", - "TERMINATE\n", - "\n", - "--------------------------------------------------------------------------------\n", - "{'call_autogen_agent': {'messages': {'role': 'assistant', 'content': 'The Fibonacci numbers between 10 and 30 are 13 and 21. \\n\\nThese numbers are part of the Fibonacci sequence, which is generated by adding the two preceding numbers to get the next number, starting from 0 and 1. \\n\\nThe sequence goes: 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, ...\\n\\nAs you can see, 13 and 21 are the only numbers in this sequence that fall between 10 and 30.\\n\\nTERMINATE'}}}\n" - ] - } - ], - "source": [ - "# pass the thread ID to persist agent outputs for future interactions\n", - "# highlight-next-line\n", - "config = {\"configurable\": {\"thread_id\": \"1\"}}\n", - "\n", - "for chunk in graph.stream(\n", - " {\n", - " \"messages\": [\n", - " {\n", - " \"role\": \"user\",\n", - " \"content\": \"Find numbers between 10 and 30 in fibonacci sequence\",\n", - " }\n", - " ]\n", - " },\n", - " # highlight-next-line\n", - " config,\n", - "):\n", - " print(chunk)" - ] - }, - { - "cell_type": "markdown", - "id": "c6cd57b4-d4ee-49f6-be12-318613849669", - "metadata": {}, - "source": [ - "Since we're leveraging LangGraph's [persistence](https://langchain-ai.github.io/langgraph/concepts/persistence/) features we can now continue the conversation using the same thread ID -- LangGraph will automatically pass previous history to the AutoGen agent:" - ] - }, - { - "cell_type": "code", - "execution_count": 7, - "id": "e68811a7-962e-4fe3-9f45-9b99ebbe04e7", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001B[33muser_proxy\u001B[0m (to assistant):\n", - "\n", - "Multiply the last number by 3\n", - "Context: \n", - "Find numbers between 10 and 30 in fibonacci sequence\n", - "The Fibonacci numbers between 10 and 30 are 13 and 21. \n", - "\n", - "These numbers are part of the Fibonacci sequence, which is generated by adding the two preceding numbers to get the next number, starting from 0 and 1. \n", - "\n", - "The sequence goes: 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, ...\n", - "\n", - "As you can see, 13 and 21 are the only numbers in this sequence that fall between 10 and 30.\n", - "\n", - "TERMINATE\n", - "\n", - "--------------------------------------------------------------------------------\n", - "\u001B[33massistant\u001B[0m (to user_proxy):\n", - "\n", - "The last number in the Fibonacci sequence between 10 and 30 is 21. Multiplying 21 by 3 gives:\n", - "\n", - "21 * 3 = 63\n", - "\n", - "TERMINATE\n", - "\n", - "--------------------------------------------------------------------------------\n", - "{'call_autogen_agent': {'messages': {'role': 'assistant', 'content': 'The last number in the Fibonacci sequence between 10 and 30 is 21. Multiplying 21 by 3 gives:\\n\\n21 * 3 = 63\\n\\nTERMINATE'}}}\n" - ] - } - ], - "source": [ - "for chunk in graph.stream(\n", - " {\n", - " \"messages\": [\n", - " {\n", - " \"role\": \"user\",\n", - " \"content\": \"Multiply the last number by 3\",\n", - " }\n", - " ]\n", - " },\n", - " # highlight-next-line\n", - " config,\n", - "):\n", - " print(chunk)" - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3 (ipykernel)", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.12.3" - } - }, - "nbformat": 4, - "nbformat_minor": 5 -} diff --git a/docs/docs/how-tos/autogen-integration.md b/docs/docs/how-tos/autogen-integration.md new file mode 100644 index 000000000..efbcf3d30 --- /dev/null +++ b/docs/docs/how-tos/autogen-integration.md @@ -0,0 +1,321 @@ +# How to integrate LangGraph with AutoGen, CrewAI, and other frameworks + +This guide shows how to integrate AutoGen agents with LangGraph to leverage features like persistence, streaming, and memory, and then deploy the integrated solution to LangGraph Platform for scalable production use. In this guide we show how to build a LangGraph chatbot that integrates with AutoGen, but you can follow the same approach with other frameworks. + +Integrating AutoGen with LangGraph provides several benefits: + +- Enhanced features: Add [persistence](../concepts/persistence.md), [streaming](../concepts/streaming.md), [short and long-term memory](../concepts/memory.md) and more to your AutoGen agents. +- Multi-agent systems: Build [multi-agent systems](../concepts/multi_agent.md) where individual agents are built with different frameworks. +- Production deployment: Deploy your integrated solution to [LangGraph Platform](../concepts/langgraph_platform.md) for scalable production use. + +## Prerequisites + +- Python 3.9+ +- Autogen: `pip install autogen` +- LangGraph: `pip install langgraph` +- OpenAI API key + +## Setup + +Set your your environment: + +```python +import getpass +import os + + +def _set_env(var: str): + if not os.environ.get(var): + os.environ[var] = getpass.getpass(f"{var}: ") + + +_set_env("OPENAI_API_KEY") +``` + +## 1. Define AutoGen agent + +Create an AutoGen agent that can execute code. This example is adapted from AutoGen's [official tutorials](https://github.com/microsoft/autogen/blob/0.2/notebook/agentchat_web_info.ipynb): + +```python +import autogen +import os + +config_list = [{"model": "gpt-4o", "api_key": os.environ["OPENAI_API_KEY"]}] + +llm_config = { + "timeout": 600, + "cache_seed": 42, + "config_list": config_list, + "temperature": 0, +} + +autogen_agent = autogen.AssistantAgent( + name="assistant", + llm_config=llm_config, +) + +user_proxy = autogen.UserProxyAgent( + name="user_proxy", + human_input_mode="NEVER", + max_consecutive_auto_reply=10, + is_termination_msg=lambda x: x.get("content", "").rstrip().endswith("TERMINATE"), + code_execution_config={ + "work_dir": "web", + "use_docker": False, + }, # Please set use_docker=True if docker is available to run the generated code. Using docker is safer than running the generated code directly. + llm_config=llm_config, + system_message="Reply TERMINATE if the task has been solved at full satisfaction. Otherwise, reply CONTINUE, or the reason why the task is not solved yet.", +) +``` + +## 2. Create the graph + +We will now create a LangGraph chatbot graph that calls AutoGen agent. + +```python +from langchain_core.messages import convert_to_openai_messages +from langgraph.graph import StateGraph, MessagesState, START +from langgraph.checkpoint.memory import MemorySaver + +def call_autogen_agent(state: MessagesState): + # Convert LangGraph messages to OpenAI format for AutoGen + messages = convert_to_openai_messages(state["messages"]) + + # Get the last user message + last_message = messages[-1] + + # Pass previous message history as context (excluding the last message) + carryover = messages[:-1] if len(messages) > 1 else [] + + # Initiate chat with AutoGen + response = user_proxy.initiate_chat( + autogen_agent, + message=last_message, + carryover=carryover + ) + + # Extract the final response from the agent + final_content = response.chat_history[-1]["content"] + + # Return the response in LangGraph format + return {"messages": {"role": "assistant", "content": final_content}} + +# Create the graph with memory for persistence +checkpointer = MemorySaver() + +# Build the graph +builder = StateGraph(MessagesState) +builder.add_node("autogen", call_autogen_agent) +builder.add_edge(START, "autogen") + +# Compile with checkpointer for persistence +graph = builder.compile(checkpointer=checkpointer) +``` + +```python +from IPython.display import display, Image + +display(Image(graph.get_graph().draw_mermaid_png())) +``` + +![Graph](./assets/autogen-output.png) + +## 3. Test the graph locally + +Before deploying to LangGraph Platform, you can test the graph locally: + +```python +# pass the thread ID to persist agent outputs for future interactions +# highlight-next-line +config = {"configurable": {"thread_id": "1"}} + +for chunk in graph.stream( + { + "messages": [ + { + "role": "user", + "content": "Find numbers between 10 and 30 in fibonacci sequence", + } + ] + }, + # highlight-next-line + config, +): + print(chunk) +``` + +**Output:** +``` +user_proxy (to assistant): + +Find numbers between 10 and 30 in fibonacci sequence + +-------------------------------------------------------------------------------- +assistant (to user_proxy): + +To find numbers between 10 and 30 in the Fibonacci sequence, we can generate the Fibonacci sequence and check which numbers fall within this range. Here's a plan: + +1. Generate Fibonacci numbers starting from 0. +2. Continue generating until the numbers exceed 30. +3. Collect and print the numbers that are between 10 and 30. + +... +``` + +Since we're leveraging LangGraph's [persistence](https://langchain-ai.github.io/langgraph/concepts/persistence/) features we can now continue the conversation using the same thread ID -- LangGraph will automatically pass previous history to the AutoGen agent: + +```python +for chunk in graph.stream( + { + "messages": [ + { + "role": "user", + "content": "Multiply the last number by 3", + } + ] + }, + # highlight-next-line + config, +): + print(chunk) +``` + +**Output:** +``` +user_proxy (to assistant): + +Multiply the last number by 3 +Context: +Find numbers between 10 and 30 in fibonacci sequence +The Fibonacci numbers between 10 and 30 are 13 and 21. + +These numbers are part of the Fibonacci sequence, which is generated by adding the two preceding numbers to get the next number, starting from 0 and 1. + +The sequence goes: 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, ... + +As you can see, 13 and 21 are the only numbers in this sequence that fall between 10 and 30. + +TERMINATE + +-------------------------------------------------------------------------------- +assistant (to user_proxy): + +The last number in the Fibonacci sequence between 10 and 30 is 21. Multiplying 21 by 3 gives: + +21 * 3 = 63 + +TERMINATE + +-------------------------------------------------------------------------------- +{'call_autogen_agent': {'messages': {'role': 'assistant', 'content': 'The last number in the Fibonacci sequence between 10 and 30 is 21. Multiplying 21 by 3 gives:\n\n21 * 3 = 63\n\nTERMINATE'}}} +``` + +## 4. Prepare for deployment + +To deploy to LangGraph Platform, create a file structure like the following: + +``` +my-autogen-agent/ +├── agent.py # Your main agent code +├── requirements.txt # Python dependencies +└── langgraph.json # LangGraph configuration +``` + +=== "agent.py" + + ```python + import os + import autogen + from langchain_core.messages import convert_to_openai_messages + from langgraph.graph import StateGraph, MessagesState, START + from langgraph.checkpoint.memory import MemorySaver + + # AutoGen configuration + config_list = [{"model": "gpt-4o", "api_key": os.environ["OPENAI_API_KEY"]}] + + llm_config = { + "timeout": 600, + "cache_seed": 42, + "config_list": config_list, + "temperature": 0, + } + + # Create AutoGen agents + autogen_agent = autogen.AssistantAgent( + name="assistant", + llm_config=llm_config, + ) + + user_proxy = autogen.UserProxyAgent( + name="user_proxy", + human_input_mode="NEVER", + max_consecutive_auto_reply=10, + is_termination_msg=lambda x: x.get("content", "").rstrip().endswith("TERMINATE"), + code_execution_config={ + "work_dir": "/tmp/autogen_work", + "use_docker": False, + }, + llm_config=llm_config, + system_message="Reply TERMINATE if the task has been solved at full satisfaction.", + ) + + def call_autogen_agent(state: MessagesState): + """Node function that calls the AutoGen agent""" + messages = convert_to_openai_messages(state["messages"]) + last_message = messages[-1] + carryover = messages[:-1] if len(messages) > 1 else [] + + response = user_proxy.initiate_chat( + autogen_agent, + message=last_message, + carryover=carryover + ) + + final_content = response.chat_history[-1]["content"] + return {"messages": {"role": "assistant", "content": final_content}} + + # Create and compile the graph + def create_graph(): + checkpointer = MemorySaver() + builder = StateGraph(MessagesState) + builder.add_node("autogen", call_autogen_agent) + builder.add_edge(START, "autogen") + return builder.compile(checkpointer=checkpointer) + + # Export the graph for LangGraph Platform + graph = create_graph() + ``` + +=== "requirements.txt" + + ``` + langgraph>=0.1.0 + pyautogen>=0.2.0 + langchain-core>=0.1.0 + langchain-openai>=0.0.5 + ``` + +=== "langgraph.json" + + ```json + { + "dependencies": ["."], + "graphs": { + "autogen_agent": "./agent.py:graph" + }, + "env": ".env" + } + ``` + + +## 5. Deploy to LangGraph Platform + +Deploy the graph with the LangGraph Platform CLI: + +``` +pip install -U langgraph-cli +``` + +``` +langgraph deploy --config langgraph.json +``` diff --git a/docs/docs/how-tos/autogen-langgraph-platform.ipynb b/docs/docs/how-tos/autogen-langgraph-platform.ipynb deleted file mode 100644 index 29bd24ba5..000000000 --- a/docs/docs/how-tos/autogen-langgraph-platform.ipynb +++ /dev/null @@ -1,171 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "id": "8381b6e0-29a6-48c5-b451-5d2549351249", - "metadata": {}, - "source": [ - "# How to use LangGraph Platform to deploy CrewAI, AutoGen, and other frameworks\n", - "\n", - "[LangGraph Platform](https://langchain-ai.github.io/langgraph/concepts/langgraph_platform/) provides infrastructure for deploying agents. This integrates seamlessly with LangGraph, but can also work with other frameworks. The way to make this work is to wrap the agent in a single LangGraph node, and have that be the entire graph.\n", - "\n", - "Doing so will allow you to deploy to LangGraph Platform, and allows you to get a lot of the [benefits](https://langchain-ai.github.io/langgraph/concepts/langgraph_platform/). You get horizontally scalable infrastructure, a task queue to handle bursty operations, a persistence layer to power short term memory, and long term memory support.\n", - "\n", - "In this guide we show how to do this with an AutoGen agent, but this method should work for agents defined in other frameworks like CrewAI, LlamaIndex, and others as well." - ] - }, - { - "cell_type": "markdown", - "id": "1113cb16-b538-448c-924c-85731ce96ebd", - "metadata": {}, - "source": [ - "## Setup" - ] - }, - { - "cell_type": "code", - "execution_count": 10, - "id": "f05993fa-9d03-4f45-bc13-0a8d87260d86", - "metadata": { - "scrolled": true - }, - "outputs": [], - "source": [ - "%pip install autogen langgraph" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "f4e0ca12-1714-4776-a30a-9527e519799b", - "metadata": {}, - "outputs": [], - "source": [ - "import getpass\n", - "import os\n", - "\n", - "\n", - "def _set_env(var: str):\n", - " if not os.environ.get(var):\n", - " os.environ[var] = getpass.getpass(f\"{var}: \")\n", - "\n", - "\n", - "_set_env(\"OPENAI_API_KEY\")" - ] - }, - { - "cell_type": "markdown", - "id": "1926bbc3-6b06-41e0-9604-860a2bbf8fa3", - "metadata": {}, - "source": [ - "## Define autogen agent\n", - "\n", - "Here we define our AutoGen agent. From https://github.com/microsoft/autogen/blob/0.2/notebook/agentchat_web_info.ipynb" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "d4a14dc7-d565-4207-8788-525f85b9fb27", - "metadata": {}, - "outputs": [], - "source": [ - "import autogen\n", - "import os\n", - "\n", - "config_list = [{\"model\": \"gpt-4o\", \"api_key\": os.environ[\"OPENAI_API_KEY\"]}]\n", - "\n", - "llm_config = {\n", - " \"timeout\": 600,\n", - " \"cache_seed\": 42,\n", - " \"config_list\": config_list,\n", - " \"temperature\": 0,\n", - "}\n", - "\n", - "autogen_agent = autogen.AssistantAgent(\n", - " name=\"assistant\",\n", - " llm_config=llm_config,\n", - ")\n", - "\n", - "user_proxy = autogen.UserProxyAgent(\n", - " name=\"user_proxy\",\n", - " human_input_mode=\"NEVER\",\n", - " max_consecutive_auto_reply=10,\n", - " is_termination_msg=lambda x: x.get(\"content\", \"\").rstrip().endswith(\"TERMINATE\"),\n", - " code_execution_config={\n", - " \"work_dir\": \"web\",\n", - " \"use_docker\": False,\n", - " }, # Please set use_docker=True if docker is available to run the generated code. Using docker is safer than running the generated code directly.\n", - " llm_config=llm_config,\n", - " system_message=\"Reply TERMINATE if the task has been solved at full satisfaction. Otherwise, reply CONTINUE, or the reason why the task is not solved yet.\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "b1170836-f23e-4e4c-ab83-ce791cd7fbd2", - "metadata": {}, - "source": [ - "## Wrap in LangGraph\n", - "\n", - "We now wrap the AutoGen agent in a single LangGraph node, and make that the entire graph.\n", - "The main thing this involves is defining an Input and Output schema for the node, which you would need to do if deploying this manually, so it's no extra work" - ] - }, - { - "cell_type": "code", - "execution_count": 11, - "id": "7b417c16-ff4e-4d5c-a9a9-0aaeeef6ede5", - "metadata": {}, - "outputs": [], - "source": [ - "from langgraph.graph import StateGraph, MessagesState\n", - "\n", - "\n", - "def call_autogen_agent(state: MessagesState):\n", - " last_message = state[\"messages\"][-1]\n", - " response = user_proxy.initiate_chat(autogen_agent, message=last_message.content)\n", - " # get the final response from the agent\n", - " content = response.chat_history[-1][\"content\"]\n", - " return {\"messages\": {\"role\": \"assistant\", \"content\": content}}\n", - "\n", - "\n", - "graph = StateGraph(MessagesState)\n", - "graph.add_node(call_autogen_agent)\n", - "graph.set_entry_point(\"call_autogen_agent\")\n", - "graph = graph.compile()" - ] - }, - { - "cell_type": "markdown", - "id": "f6a18377-ac29-478f-a76a-b213f1a3c85d", - "metadata": {}, - "source": [ - "## Deploy with LangGraph Platform\n", - "\n", - "You can now deploy this as you normally would with LangGraph Platform. See [these instructions](https://langchain-ai.github.io/langgraph/concepts/deployment_options/) for more details." - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3 (ipykernel)", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.12.3" - } - }, - "nbformat": 4, - "nbformat_minor": 5 -} diff --git a/docs/docs/how-tos/multi_agent.ipynb b/docs/docs/how-tos/multi_agent.ipynb deleted file mode 100644 index 6ee5777de..000000000 --- a/docs/docs/how-tos/multi_agent.ipynb +++ /dev/null @@ -1,657 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "id": "34d3d54e-9a2b-481e-bccd-74aca7a53f9a", - "metadata": {}, - "source": [ - "# Build multi-agent systems" - ] - }, - { - "attachments": {}, - "cell_type": "markdown", - "id": "3f0b4f70-f14e-4026-82c0-874786789ee8", - "metadata": {}, - "source": [ - "A single agent might struggle if it needs to specialize in multiple domains or manage many tools. To tackle this, you can break your agent into smaller, independent agents and composing them into a [multi-agent system](../../concepts/multi_agent).\n", - "\n", - "In multi-agent systems, agents need to communicate between each other. They do so via [handoffs](#handoffs) — a primitive that describes which agent to hand control to and the payload to send to that agent.\n", - "\n", - "This guide covers the following:\n", - "\n", - "* implementing [handoffs](#handoffs) between agents\n", - "* using handoffs and the prebuilt [agent](../../agents/agents) to [build a custom multi-agent system](#build-a-multi-agent-system)\n", - "\n", - "To get started with building multi-agent systems, check out LangGraph [prebuilt implementations](#prebuilt-implementations) of two of the most popular multi-agent architectures — [supervisor](../../agents/multi-agent#supervisor) and [swarm](../../agents/multi-agent#swarm)." - ] - }, - { - "attachments": {}, - "cell_type": "markdown", - "id": "7d43e110-16fc-4899-97f1-015d5b804b87", - "metadata": {}, - "source": [ - "## Handoffs\n", - "\n", - "To set up communication between the agents in a multi-agent system you can use [**handoffs**](../../concepts/multi_agent#handoffs) — a pattern where one agent *hands off* control to another. Handoffs allow you to specify:\n", - "\n", - "- **destination**: target agent to navigate to (e.g., name of the LangGraph node to go to)\n", - "- **payload**: information to pass to that agent (e.g., state update)\n", - "\n", - "### Create handoffs\n", - "\n", - "To implement handoffs, you can return [`Command`](../command) objects from your agent nodes or tools:\n", - "\n", - "```python\n", - "from typing import Annotated\n", - "from langchain_core.tools import tool, InjectedToolCallId\n", - "from langgraph.prebuilt import create_react_agent, InjectedState\n", - "from langgraph.graph import StateGraph, START, MessagesState\n", - "from langgraph.types import Command\n", - "\n", - "def create_handoff_tool(*, agent_name: str, description: str | None = None):\n", - " name = f\"transfer_to_{agent_name}\"\n", - " description = description or f\"Transfer to {agent_name}\"\n", - "\n", - " @tool(name, description=description)\n", - " def handoff_tool(\n", - " # highlight-next-line\n", - " state: Annotated[MessagesState, InjectedState], # (1)!\n", - " # highlight-next-line\n", - " tool_call_id: Annotated[str, InjectedToolCallId],\n", - " ) -> Command:\n", - " tool_message = {\n", - " \"role\": \"tool\",\n", - " \"content\": f\"Successfully transferred to {agent_name}\",\n", - " \"name\": name,\n", - " \"tool_call_id\": tool_call_id,\n", - " }\n", - " return Command( # (2)!\n", - " # highlight-next-line\n", - " goto=agent_name, # (3)!\n", - " # highlight-next-line\n", - " update={\"messages\": state[\"messages\"] + [tool_message]}, # (4)!\n", - " # highlight-next-line\n", - " graph=Command.PARENT, # (5)!\n", - " )\n", - " return handoff_tool\n", - "```\n", - "\n", - "1. Access the [state](../../concepts/low_level#state) of the agent that is calling the handoff tool using the [InjectedState][langgraph.prebuilt.InjectedState] annotation. See [this guide](../tool-calling/#read-state) for more information.\n", - "2. The `Command` primitive allows specifying a state update and a node transition as a single operation, making it useful for implementing handoffs.\n", - "3. Name of the agent or node to hand off to.\n", - "4. Take the agent's messages and **add** them to the parent's **state** as part of the handoff. The next agent will see the parent state.\n", - "5. Indicate to LangGraph that we need to navigate to agent node in a **parent** multi-agent graph.\n", - "\n", - "!!! tip\n", - "\n", - " If you want to use tools that return `Command`, you can either use prebuilt [`create_react_agent`][langgraph.prebuilt.chat_agent_executor.create_react_agent] / [`ToolNode`][langgraph.prebuilt.tool_node.ToolNode] components, or implement your own tool-executing node that collects `Command` objects returned by the tools and returns a list of them, e.g.:\n", - " \n", - " ```python\n", - " def call_tools(state):\n", - " ...\n", - " commands = [tools_by_name[tool_call[\"name\"]].invoke(tool_call) for tool_call in tool_calls]\n", - " return commands\n", - " ```\n", - "\n", - "!!! Important\n", - "\n", - " This handoff implementation assumes that:\n", - " \n", - " - each agent receives overall message history (across all agents) in the multi-agent system as its input. If you want more control over agent inputs, see [this section](#control-agent-inputs)\n", - " - each agent outputs its internal messages history to the overall message history of the multi-agent system. If you want more control over **how agent outputs are added**, wrap the agent in a separate node function:\n", - "\n", - " ```python\n", - " def call_hotel_assistant(state):\n", - " # return agent's final response,\n", - " # excluding inner monologue\n", - " response = hotel_assistant.invoke(state)\n", - " # highlight-next-line\n", - " return {\"messages\": response[\"messages\"][-1]}\n", - " ```" - ] - }, - { - "cell_type": "markdown", - "id": "3956f12d-285a-4799-a0a5-db13def58a15", - "metadata": {}, - "source": [ - "### Control agent inputs\n", - "\n", - "You can use the [`Send()`][langgraph.types.Send] primitive to directly send data to the worker agents during the handoff. For example, you can request that the calling agent populate a task description for the next agent:\n", - "\n", - "```python\n", - "\n", - "from typing import Annotated\n", - "from langchain_core.tools import tool, InjectedToolCallId\n", - "from langgraph.prebuilt import InjectedState\n", - "from langgraph.graph import StateGraph, START, MessagesState\n", - "# highlight-next-line\n", - "from langgraph.types import Command, Send\n", - "\n", - "def create_task_description_handoff_tool(\n", - " *, agent_name: str, description: str | None = None\n", - "):\n", - " name = f\"transfer_to_{agent_name}\"\n", - " description = description or f\"Ask {agent_name} for help.\"\n", - "\n", - " @tool(name, description=description)\n", - " def handoff_tool(\n", - " # this is populated by the calling agent\n", - " task_description: Annotated[\n", - " str,\n", - " \"Description of what the next agent should do, including all of the relevant context.\",\n", - " ],\n", - " # these parameters are ignored by the LLM\n", - " state: Annotated[MessagesState, InjectedState],\n", - " ) -> Command:\n", - " task_description_message = {\"role\": \"user\", \"content\": task_description}\n", - " agent_input = {**state, \"messages\": [task_description_message]}\n", - " return Command(\n", - " # highlight-next-line\n", - " goto=[Send(agent_name, agent_input)],\n", - " graph=Command.PARENT,\n", - " )\n", - "\n", - " return handoff_tool\n", - "```\n", - "\n", - "See the multi-agent [supervisor](../tutorials/agent_supervisor.ipynb#4-create-delegation-tasks) tutorial for a full example of using [`Send()`][langgraph.types.Send] in handoffs." - ] - }, - { - "attachments": {}, - "cell_type": "markdown", - "id": "21511f57-7bf3-4223-9a17-ce9fc84c40ab", - "metadata": {}, - "source": [ - "## Build a multi-agent system\n", - "\n", - "You can use handoffs in any agents built with LangGraph. We recommend using the prebuilt [agent](../../agents/overview) or [`ToolNode`](../tool-calling#use-prebuilt-toolnode), as they natively support handoffs tools returning `Command`. Below is an example of how you can implement a multi-agent system for booking travel using handoffs:\n", - "\n", - "```python\n", - "from langgraph.prebuilt import create_react_agent\n", - "from langgraph.graph import StateGraph, START, MessagesState\n", - "\n", - "def create_handoff_tool(*, agent_name: str, description: str | None = None):\n", - " # same implementation as above\n", - " ...\n", - " return Command(...)\n", - "\n", - "# Handoffs\n", - "transfer_to_hotel_assistant = create_handoff_tool(agent_name=\"hotel_assistant\")\n", - "transfer_to_flight_assistant = create_handoff_tool(agent_name=\"flight_assistant\")\n", - "\n", - "# Define agents\n", - "flight_assistant = create_react_agent(\n", - " model=\"anthropic:claude-3-5-sonnet-latest\",\n", - " # highlight-next-line\n", - " tools=[..., transfer_to_hotel_assistant],\n", - " # highlight-next-line\n", - " name=\"flight_assistant\"\n", - ")\n", - "hotel_assistant = create_react_agent(\n", - " model=\"anthropic:claude-3-5-sonnet-latest\",\n", - " # highlight-next-line\n", - " tools=[..., transfer_to_flight_assistant],\n", - " # highlight-next-line\n", - " name=\"hotel_assistant\"\n", - ")\n", - "\n", - "# Define multi-agent graph\n", - "multi_agent_graph = (\n", - " StateGraph(MessagesState)\n", - " # highlight-next-line\n", - " .add_node(flight_assistant)\n", - " # highlight-next-line\n", - " .add_node(hotel_assistant)\n", - " .add_edge(START, \"flight_assistant\")\n", - " .compile()\n", - ")\n", - "```\n", - "\n", - "??? example \"Full example: Multi-agent system for booking travel\"\n", - "\n", - " ```python\n", - " from typing import Annotated\n", - " from langchain_core.messages import convert_to_messages\n", - " from langchain_core.tools import tool, InjectedToolCallId\n", - " from langgraph.prebuilt import create_react_agent, InjectedState\n", - " from langgraph.graph import StateGraph, START, MessagesState\n", - " from langgraph.types import Command\n", - " \n", - " # We'll use `pretty_print_messages` helper to render the streamed agent outputs nicely\n", - " \n", - " def pretty_print_message(message, indent=False):\n", - " pretty_message = message.pretty_repr(html=True)\n", - " if not indent:\n", - " print(pretty_message)\n", - " return\n", - " \n", - " indented = \"\\n\".join(\"\\t\" + c for c in pretty_message.split(\"\\n\"))\n", - " print(indented)\n", - " \n", - " \n", - " def pretty_print_messages(update, last_message=False):\n", - " is_subgraph = False\n", - " if isinstance(update, tuple):\n", - " ns, update = update\n", - " # skip parent graph updates in the printouts\n", - " if len(ns) == 0:\n", - " return\n", - " \n", - " graph_id = ns[-1].split(\":\")[0]\n", - " print(f\"Update from subgraph {graph_id}:\")\n", - " print(\"\\n\")\n", - " is_subgraph = True\n", - " \n", - " for node_name, node_update in update.items():\n", - " update_label = f\"Update from node {node_name}:\"\n", - " if is_subgraph:\n", - " update_label = \"\\t\" + update_label\n", - " \n", - " print(update_label)\n", - " print(\"\\n\")\n", - " \n", - " messages = convert_to_messages(node_update[\"messages\"])\n", - " if last_message:\n", - " messages = messages[-1:]\n", - " \n", - " for m in messages:\n", - " pretty_print_message(m, indent=is_subgraph)\n", - " print(\"\\n\")\n", - "\n", - "\n", - " def create_handoff_tool(*, agent_name: str, description: str | None = None):\n", - " name = f\"transfer_to_{agent_name}\"\n", - " description = description or f\"Transfer to {agent_name}\"\n", - " \n", - " @tool(name, description=description)\n", - " def handoff_tool(\n", - " # highlight-next-line\n", - " state: Annotated[MessagesState, InjectedState], # (1)!\n", - " # highlight-next-line\n", - " tool_call_id: Annotated[str, InjectedToolCallId],\n", - " ) -> Command:\n", - " tool_message = {\n", - " \"role\": \"tool\",\n", - " \"content\": f\"Successfully transferred to {agent_name}\",\n", - " \"name\": name,\n", - " \"tool_call_id\": tool_call_id,\n", - " }\n", - " return Command( # (2)!\n", - " # highlight-next-line\n", - " goto=agent_name, # (3)!\n", - " # highlight-next-line\n", - " update={\"messages\": state[\"messages\"] + [tool_message]}, # (4)!\n", - " # highlight-next-line\n", - " graph=Command.PARENT, # (5)!\n", - " )\n", - " return handoff_tool\n", - " \n", - " # Handoffs\n", - " transfer_to_hotel_assistant = create_handoff_tool(\n", - " agent_name=\"hotel_assistant\",\n", - " description=\"Transfer user to the hotel-booking assistant.\",\n", - " )\n", - " transfer_to_flight_assistant = create_handoff_tool(\n", - " agent_name=\"flight_assistant\",\n", - " description=\"Transfer user to the flight-booking assistant.\",\n", - " )\n", - " \n", - " # Simple agent tools\n", - " def book_hotel(hotel_name: str):\n", - " \"\"\"Book a hotel\"\"\"\n", - " return f\"Successfully booked a stay at {hotel_name}.\"\n", - " \n", - " def book_flight(from_airport: str, to_airport: str):\n", - " \"\"\"Book a flight\"\"\"\n", - " return f\"Successfully booked a flight from {from_airport} to {to_airport}.\"\n", - " \n", - " # Define agents\n", - " flight_assistant = create_react_agent(\n", - " model=\"anthropic:claude-3-5-sonnet-latest\",\n", - " # highlight-next-line\n", - " tools=[book_flight, transfer_to_hotel_assistant],\n", - " prompt=\"You are a flight booking assistant\",\n", - " # highlight-next-line\n", - " name=\"flight_assistant\"\n", - " )\n", - " hotel_assistant = create_react_agent(\n", - " model=\"anthropic:claude-3-5-sonnet-latest\",\n", - " # highlight-next-line\n", - " tools=[book_hotel, transfer_to_flight_assistant],\n", - " prompt=\"You are a hotel booking assistant\",\n", - " # highlight-next-line\n", - " name=\"hotel_assistant\"\n", - " )\n", - " \n", - " # Define multi-agent graph\n", - " multi_agent_graph = (\n", - " StateGraph(MessagesState)\n", - " .add_node(flight_assistant)\n", - " .add_node(hotel_assistant)\n", - " .add_edge(START, \"flight_assistant\")\n", - " .compile()\n", - " )\n", - " \n", - " # Run the multi-agent graph\n", - " for chunk in multi_agent_graph.stream(\n", - " {\n", - " \"messages\": [\n", - " {\n", - " \"role\": \"user\",\n", - " \"content\": \"book a flight from BOS to JFK and a stay at McKittrick Hotel\"\n", - " }\n", - " ]\n", - " },\n", - " # highlight-next-line\n", - " subgraphs=True\n", - " ):\n", - " pretty_print_messages(chunk)\n", - " ```\n", - "\n", - " 1. Access agent's state\n", - " 2. The `Command` primitive allows specifying a state update and a node transition as a single operation, making it useful for implementing handoffs.\n", - " 3. Name of the agent or node to hand off to.\n", - " 4. Take the agent's messages and **add** them to the parent's **state** as part of the handoff. The next agent will see the parent state.\n", - " 5. Indicate to LangGraph that we need to navigate to agent node in a **parent** multi-agent graph." - ] - }, - { - "cell_type": "markdown", - "id": "e8314da4-9971-429b-9e70-58b40795de74", - "metadata": {}, - "source": [ - "## Multi-turn conversation\n", - "\n", - "Users might want to engage in a *multi-turn conversation* with one or more agents. To build a system that can handle this, you can create a node that uses an [`interrupt`][langgraph.types.interrupt] to collect user input and routes back to the **active** agent.\n", - "\n", - "The agents can then be implemented as nodes in a graph that executes agent steps and determines the next action:\n", - "\n", - "1. **Wait for user input** to continue the conversation, or \n", - "2. **Route to another agent** (or back to itself, such as in a loop) via a [handoff](#handoffs)\n", - "\n", - "```python\n", - "def human(state) -> Command[Literal[\"agent\", \"another_agent\"]]:\n", - " \"\"\"A node for collecting user input.\"\"\"\n", - " user_input = interrupt(value=\"Ready for user input.\")\n", - "\n", - " # Determine the active agent.\n", - " active_agent = ...\n", - "\n", - " ...\n", - " return Command(\n", - " update={\n", - " \"messages\": [{\n", - " \"role\": \"human\",\n", - " \"content\": user_input,\n", - " }]\n", - " },\n", - " goto=active_agent\n", - " )\n", - "\n", - "def agent(state) -> Command[Literal[\"agent\", \"another_agent\", \"human\"]]:\n", - " # The condition for routing/halting can be anything, e.g. LLM tool call / structured output, etc.\n", - " goto = get_next_agent(...) # 'agent' / 'another_agent'\n", - " if goto:\n", - " return Command(goto=goto, update={\"my_state_key\": \"my_state_value\"})\n", - " else:\n", - " return Command(goto=\"human\") # Go to human node\n", - "```\n", - "\n", - "??? example \"Full example: multi-agent system for travel recommendations\"\n", - "\n", - " In this example, we will build a team of travel assistant agents that can communicate with each other via handoffs.\n", - " \n", - " We will create 2 agents:\n", - " \n", - " * travel_advisor: can help with travel destination recommendations. Can ask hotel_advisor for help.\n", - " * hotel_advisor: can help with hotel recommendations. Can ask travel_advisor for help.\n", - "\n", - " ```python\n", - " from langchain_anthropic import ChatAnthropic\n", - " from langgraph.graph import MessagesState, StateGraph, START\n", - " from langgraph.prebuilt import create_react_agent, InjectedState\n", - " from langgraph.types import Command, interrupt\n", - " from langgraph.checkpoint.memory import MemorySaver\n", - " \n", - " \n", - " model = ChatAnthropic(model=\"claude-3-5-sonnet-latest\")\n", - "\n", - " class MultiAgentState(MessagesState):\n", - " last_active_agent: str\n", - " \n", - " \n", - " # Define travel advisor tools and ReAct agent\n", - " travel_advisor_tools = [\n", - " get_travel_recommendations,\n", - " make_handoff_tool(agent_name=\"hotel_advisor\"),\n", - " ]\n", - " travel_advisor = create_react_agent(\n", - " model,\n", - " travel_advisor_tools,\n", - " prompt=(\n", - " \"You are a general travel expert that can recommend travel destinations (e.g. countries, cities, etc). \"\n", - " \"If you need hotel recommendations, ask 'hotel_advisor' for help. \"\n", - " \"You MUST include human-readable response before transferring to another agent.\"\n", - " ),\n", - " )\n", - " \n", - " \n", - " def call_travel_advisor(\n", - " state: MultiAgentState,\n", - " ) -> Command[Literal[\"hotel_advisor\", \"human\"]]:\n", - " # You can also add additional logic like changing the input to the agent / output from the agent, etc.\n", - " # NOTE: we're invoking the ReAct agent with the full history of messages in the state\n", - " response = travel_advisor.invoke(state)\n", - " update = {**response, \"last_active_agent\": \"travel_advisor\"}\n", - " return Command(update=update, goto=\"human\")\n", - " \n", - " \n", - " # Define hotel advisor tools and ReAct agent\n", - " hotel_advisor_tools = [\n", - " get_hotel_recommendations,\n", - " make_handoff_tool(agent_name=\"travel_advisor\"),\n", - " ]\n", - " hotel_advisor = create_react_agent(\n", - " model,\n", - " hotel_advisor_tools,\n", - " prompt=(\n", - " \"You are a hotel expert that can provide hotel recommendations for a given destination. \"\n", - " \"If you need help picking travel destinations, ask 'travel_advisor' for help.\"\n", - " \"You MUST include human-readable response before transferring to another agent.\"\n", - " ),\n", - " )\n", - " \n", - " \n", - " def call_hotel_advisor(\n", - " state: MultiAgentState,\n", - " ) -> Command[Literal[\"travel_advisor\", \"human\"]]:\n", - " response = hotel_advisor.invoke(state)\n", - " update = {**response, \"last_active_agent\": \"hotel_advisor\"}\n", - " return Command(update=update, goto=\"human\")\n", - " \n", - " \n", - " def human_node(\n", - " state: MultiAgentState, config\n", - " ) -> Command[Literal[\"hotel_advisor\", \"travel_advisor\", \"human\"]]:\n", - " \"\"\"A node for collecting user input.\"\"\"\n", - " \n", - " user_input = interrupt(value=\"Ready for user input.\")\n", - " active_agent = state[\"last_active_agent\"]\n", - " \n", - " return Command(\n", - " update={\n", - " \"messages\": [\n", - " {\n", - " \"role\": \"human\",\n", - " \"content\": user_input,\n", - " }\n", - " ]\n", - " },\n", - " goto=active_agent,\n", - " )\n", - " \n", - " \n", - " builder = StateGraph(MultiAgentState)\n", - " builder.add_node(\"travel_advisor\", call_travel_advisor)\n", - " builder.add_node(\"hotel_advisor\", call_hotel_advisor)\n", - " \n", - " # This adds a node to collect human input, which will route\n", - " # back to the active agent.\n", - " builder.add_node(\"human\", human_node)\n", - " \n", - " # We'll always start with a general travel advisor.\n", - " builder.add_edge(START, \"travel_advisor\")\n", - " \n", - " \n", - " checkpointer = MemorySaver()\n", - " graph = builder.compile(checkpointer=checkpointer)\n", - " ```\n", - " \n", - " Let's test a multi turn conversation with this application.\n", - "\n", - " ```python\n", - " import uuid\n", - " \n", - " thread_config = {\"configurable\": {\"thread_id\": str(uuid.uuid4())}}\n", - " \n", - " inputs = [\n", - " # 1st round of conversation,\n", - " {\n", - " \"messages\": [\n", - " {\"role\": \"user\", \"content\": \"i wanna go somewhere warm in the caribbean\"}\n", - " ]\n", - " },\n", - " # Since we're using `interrupt`, we'll need to resume using the Command primitive.\n", - " # 2nd round of conversation,\n", - " Command(\n", - " resume=\"could you recommend a nice hotel in one of the areas and tell me which area it is.\"\n", - " ),\n", - " # 3rd round of conversation,\n", - " Command(\n", - " resume=\"i like the first one. could you recommend something to do near the hotel?\"\n", - " ),\n", - " ]\n", - " \n", - " for idx, user_input in enumerate(inputs):\n", - " print()\n", - " print(f\"--- Conversation Turn {idx + 1} ---\")\n", - " print()\n", - " print(f\"User: {user_input}\")\n", - " print()\n", - " for update in graph.stream(\n", - " user_input,\n", - " config=thread_config,\n", - " stream_mode=\"updates\",\n", - " ):\n", - " for node_id, value in update.items():\n", - " if isinstance(value, dict) and value.get(\"messages\", []):\n", - " last_message = value[\"messages\"][-1]\n", - " if isinstance(last_message, dict) or last_message.type != \"ai\":\n", - " continue\n", - " print(f\"{node_id}: {last_message.content}\")\n", - " ```\n", - " \n", - " ```\n", - " --- Conversation Turn 1 ---\n", - " \n", - " User: {'messages': [{'role': 'user', 'content': 'i wanna go somewhere warm in the caribbean'}]}\n", - " \n", - " travel_advisor: Based on the recommendations, Aruba would be an excellent choice for your Caribbean getaway! Aruba is known as \"One Happy Island\" and offers:\n", - " - Year-round warm weather with consistent temperatures around 82°F (28°C)\n", - " - Beautiful white sand beaches like Eagle Beach and Palm Beach\n", - " - Clear turquoise waters perfect for swimming and snorkeling\n", - " - Minimal rainfall and location outside the hurricane belt\n", - " - A blend of Caribbean and Dutch culture\n", - " - Great dining options and nightlife\n", - " - Various water sports and activities\n", - " \n", - " Would you like me to get some specific hotel recommendations in Aruba for your stay? I can transfer you to our hotel advisor who can help with accommodations.\n", - " \n", - " --- Conversation Turn 2 ---\n", - " \n", - " User: Command(resume='could you recommend a nice hotel in one of the areas and tell me which area it is.')\n", - " \n", - " hotel_advisor: Based on the recommendations, I can suggest two excellent options:\n", - " \n", - " 1. The Ritz-Carlton, Aruba - Located in Palm Beach\n", - " - This luxury resort is situated in the vibrant Palm Beach area\n", - " - Known for its exceptional service and amenities\n", - " - Perfect if you want to be close to dining, shopping, and entertainment\n", - " - Features multiple restaurants, a casino, and a world-class spa\n", - " - Located on a pristine stretch of Palm Beach\n", - " \n", - " 2. Bucuti & Tara Beach Resort - Located in Eagle Beach\n", - " - An adults-only boutique resort on Eagle Beach\n", - " - Known for being more intimate and peaceful\n", - " - Award-winning for its sustainability practices\n", - " - Perfect for a romantic getaway or peaceful vacation\n", - " - Located on one of the most beautiful beaches in the Caribbean\n", - " \n", - " Would you like more specific information about either of these properties or their locations?\n", - " \n", - " --- Conversation Turn 3 ---\n", - " \n", - " User: Command(resume='i like the first one. could you recommend something to do near the hotel?')\n", - " \n", - " travel_advisor: Near the Ritz-Carlton in Palm Beach, here are some highly recommended activities:\n", - " \n", - " 1. Visit the Palm Beach Plaza Mall - Just a short walk from the hotel, featuring shopping, dining, and entertainment\n", - " 2. Try your luck at the Stellaris Casino - It's right in the Ritz-Carlton\n", - " 3. Take a sunset sailing cruise - Many depart from the nearby pier\n", - " 4. Visit the California Lighthouse - A scenic landmark just north of Palm Beach\n", - " 5. Enjoy water sports at Palm Beach:\n", - " - Jet skiing\n", - " - Parasailing\n", - " - Snorkeling\n", - " - Stand-up paddleboarding\n", - " \n", - " Would you like more specific information about any of these activities or would you like to know about other options in the area?\n", - " ```" - ] - }, - { - "cell_type": "markdown", - "id": "04d18c63-a0eb-45ac-86dc-0cc5bd683973", - "metadata": {}, - "source": [ - "## Prebuilt implementations" - ] - }, - { - "cell_type": "markdown", - "id": "e0e4ce57-f8de-4f37-836e-c1e1a02dd7b7", - "metadata": {}, - "source": [ - "LangGraph comes with prebuilt implementations of two of the most popular multi-agent architectures:\n", - "\n", - "- [supervisor](../../agents/multi-agent#supervisor) — individual agents are coordinated by a central supervisor agent. The supervisor controls all communication flow and task delegation, making decisions about which agent to invoke based on the current context and task requirements. You can use [`langgraph-supervisor`](https://github.com/langchain-ai/langgraph-supervisor-py) library to create a supervisor multi-agent systems.\n", - "- [swarm](../../agents/multi-agent#supervisor) — agents dynamically hand off control to one another based on their specializations. The system remembers which agent was last active, ensuring that on subsequent interactions, the conversation resumes with that agent. You can use [`langgraph-swarm`](https://github.com/langchain-ai/langgraph-swarm-py) library to create a swarm multi-agent systems." - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3 (ipykernel)", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.12.3" - } - }, - "nbformat": 4, - "nbformat_minor": 5 -} diff --git a/docs/docs/how-tos/multi_agent.md b/docs/docs/how-tos/multi_agent.md index 5aead8b95..fdd781d4f 100644 --- a/docs/docs/how-tos/multi_agent.md +++ b/docs/docs/how-tos/multi_agent.md @@ -130,7 +130,7 @@ def create_task_description_handoff_tool( return handoff_tool ``` -See the multi-agent [supervisor](../tutorials/multi_agent/agent_supervisor.ipynb#4-create-delegation-tasks) example for a full example of using [`Send()`][langgraph.types.Send] in handoffs. +See the multi-agent [supervisor](../tutorials/multi_agent/agent_supervisor.md#4-create-delegation-tasks) example for a full example of using [`Send()`][langgraph.types.Send] in handoffs. ## Build a multi-agent system diff --git a/docs/docs/how-tos/run-id-langsmith.ipynb b/docs/docs/how-tos/run-id-langsmith.ipynb deleted file mode 100644 index 550c8ae64..000000000 --- a/docs/docs/how-tos/run-id-langsmith.ipynb +++ /dev/null @@ -1,274 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# How to pass custom run ID or set tags and metadata for graph runs in LangSmith\n", - "\n", - "

\n", - "

Prerequisites

\n", - "

\n", - " This guide assumes familiarity with the following:\n", - "

\n", - "

\n", - "
\n", - "\n", - "Debugging graph runs can sometimes be difficult to do in an IDE or terminal. [LangSmith](https://docs.smith.langchain.com) lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read the [LangSmith documentation](https://docs.smith.langchain.com) for more information on how to get started.\n", - "\n", - "To make it easier to identify and analyzed traces generated during graph invocation, you can set additional configuration at run time (see [RunnableConfig](https://api.python.langchain.com/en/latest/runnables/langchain_core.runnables.config.RunnableConfig.html#langchain_core.runnables.config.RunnableConfig)):\n", - "\n", - "| **Field** | **Type** | **Description** |\n", - "|-------------|---------------------|--------------------------------------------------------------------------------------------------------------------|\n", - "| run_name | `str` | Name for the tracer run for this call. Defaults to the name of the class. |\n", - "| run_id | `UUID` | Unique identifier for the tracer run for this call. If not provided, a new UUID will be generated. |\n", - "| tags | `List[str]` | Tags for this call and any sub-calls (e.g., a Chain calling an LLM). You can use these to filter calls. |\n", - "| metadata | `Dict[str, Any]` | Metadata for this call and any sub-calls (e.g., a Chain calling an LLM). Keys should be strings, values should be JSON-serializable. |\n", - "\n", - "LangGraph graphs implement the [LangChain Runnable Interface](https://python.langchain.com/api_reference/core/runnables/langchain_core.runnables.base.Runnable.html) and accept a second argument (`RunnableConfig`) in methods like `invoke`, `ainvoke`, `stream` etc.\n", - "\n", - "The LangSmith platform will allow you to search and filter traces based on `run_name`, `run_id`, `tags` and `metadata`.\n", - "\n", - "\n", - "## TLDR\n", - "\n", - "```python\n", - "import uuid\n", - "# Generate a random UUID -- it must be a UUID\n", - "config = {\"run_id\": uuid.uuid4()}, \"tags\": [\"my_tag1\"], \"metadata\": {\"a\": 5}}\n", - "# Works with all standard Runnable methods \n", - "# like invoke, batch, ainvoke, astream_events etc\n", - "graph.stream(inputs, config, stream_mode=\"values\")\n", - "```\n", - "\n", - "The rest of the how to guide will show a full agent.\n", - "\n", - "## Setup\n", - "\n", - "First, let's install the required packages and set our API keys" - ] - }, - { - "cell_type": "code", - "execution_count": 28, - "metadata": {}, - "outputs": [], - "source": [ - "%%capture --no-stderr\n", - "%pip install --quiet -U langgraph langchain_openai" - ] - }, - { - "cell_type": "code", - "execution_count": 29, - "metadata": {}, - "outputs": [], - "source": [ - "import getpass\n", - "import os\n", - "\n", - "\n", - "def _set_env(var: str):\n", - " if not os.environ.get(var):\n", - " os.environ[var] = getpass.getpass(f\"{var}: \")\n", - "\n", - "\n", - "_set_env(\"OPENAI_API_KEY\")\n", - "_set_env(\"LANGSMITH_API_KEY\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "
\n", - "

Set up LangSmith for LangGraph development

\n", - "

\n", - " Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started here. \n", - "

\n", - "
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Define the graph\n", - "\n", - "For this example we will use the [prebuilt ReAct agent](https://langchain-ai.github.io/langgraph/how-tos/create-react-agent/)." - ] - }, - { - "cell_type": "code", - "execution_count": 26, - "metadata": {}, - "outputs": [], - "source": [ - "from langchain_openai import ChatOpenAI\n", - "from typing import Literal\n", - "from langgraph.prebuilt import create_react_agent\n", - "from langchain_core.tools import tool\n", - "\n", - "# First we initialize the model we want to use.\n", - "model = ChatOpenAI(model=\"gpt-4o\", temperature=0)\n", - "\n", - "\n", - "# For this tutorial we will use custom tool that returns pre-defined values for weather in two cities (NYC & SF)\n", - "@tool\n", - "def get_weather(city: Literal[\"nyc\", \"sf\"]):\n", - " \"\"\"Use this to get weather information.\"\"\"\n", - " if city == \"nyc\":\n", - " return \"It might be cloudy in nyc\"\n", - " elif city == \"sf\":\n", - " return \"It's always sunny in sf\"\n", - " else:\n", - " raise AssertionError(\"Unknown city\")\n", - "\n", - "\n", - "tools = [get_weather]\n", - "\n", - "\n", - "# Define the graph\n", - "graph = create_react_agent(model, tools=tools)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Run your graph\n", - "\n", - "Now that we've defined our graph let's run it once and view the trace in LangSmith. In order for our trace to be easily accessible in LangSmith, we will pass in a custom `run_id` in the config.\n", - "\n", - "This assumes that you have set your `LANGSMITH_API_KEY` environment variable.\n", - "\n", - "Note that you can also configure what project to trace to by setting the `LANGCHAIN_PROJECT` environment variable, by default runs will be traced to the `default` project." - ] - }, - { - "cell_type": "code", - "execution_count": 27, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "================================\u001b[1m Human Message \u001b[0m=================================\n", - "\n", - "what is the weather in sf\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Tool Calls:\n", - " get_weather (call_9ZudXyMAdlUjptq9oMGtQo8o)\n", - " Call ID: call_9ZudXyMAdlUjptq9oMGtQo8o\n", - " Args:\n", - " city: sf\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: get_weather\n", - "\n", - "It's always sunny in sf\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "\n", - "The weather in San Francisco is currently sunny.\n" - ] - } - ], - "source": [ - "import uuid\n", - "\n", - "\n", - "def print_stream(stream):\n", - " for s in stream:\n", - " message = s[\"messages\"][-1]\n", - " if isinstance(message, tuple):\n", - " print(message)\n", - " else:\n", - " message.pretty_print()\n", - "\n", - "\n", - "inputs = {\"messages\": [(\"user\", \"what is the weather in sf\")]}\n", - "\n", - "config = {\"run_name\": \"agent_007\", \"tags\": [\"cats are awesome\"]}\n", - "\n", - "print_stream(graph.stream(inputs, config, stream_mode=\"values\"))" - ] - }, - { - "attachments": { - "d38d1f2b-0f4c-4707-b531-a3c749de987f.png": { - "image/png": "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" - } - }, - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## View the trace in LangSmith\n", - "\n", - "Now that we've ran our graph, let's head over to LangSmith and view our trace. First click into the project that you traced to (in our case the default project). You should see a run with the custom run name \"agent_007\".\n", - "\n", - "![image.png](attachment:d38d1f2b-0f4c-4707-b531-a3c749de987f.png)" - ] - }, - { - "attachments": { - "410e0089-2ab8-46bb-a61a-827187fd46b3.png": { - "image/png": "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" - } - }, - "cell_type": "markdown", - "metadata": {}, - "source": [ - "In addition, you will be able to filter traces after the fact using the tags or metadata provided. For example,\n", - "\n", - "![image.png](attachment:410e0089-2ab8-46bb-a61a-827187fd46b3.png)\n" - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3 (ipykernel)", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.11.4" - } - }, - "nbformat": 4, - "nbformat_minor": 4 -} diff --git a/docs/docs/how-tos/run-id-langsmith.md b/docs/docs/how-tos/run-id-langsmith.md new file mode 100644 index 000000000..0cd87e864 --- /dev/null +++ b/docs/docs/how-tos/run-id-langsmith.md @@ -0,0 +1,155 @@ +# How to pass custom run ID or set tags and metadata for graph runs in LangSmith + +!!! tip "Prerequisites" + This guide assumes familiarity with the following: + + - [LangSmith Documentation](https://docs.smith.langchain.com) + - [LangSmith Platform](https://smith.langchain.com) + - [RunnableConfig](https://api.python.langchain.com/en/latest/runnables/langchain_core.runnables.config.RunnableConfig.html#langchain_core.runnables.config.RunnableConfig) + - [Add metadata and tags to traces](https://docs.smith.langchain.com/how_to_guides/tracing/trace_with_langchain#add-metadata-and-tags-to-traces) + - [Customize run name](https://docs.smith.langchain.com/how_to_guides/tracing/trace_with_langchain#customize-run-name) + +Debugging graph runs can sometimes be difficult to do in an IDE or terminal. [LangSmith](https://docs.smith.langchain.com) lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read the [LangSmith documentation](https://docs.smith.langchain.com) for more information on how to get started. + +To make it easier to identify and analyzed traces generated during graph invocation, you can set additional configuration at run time (see [RunnableConfig](https://api.python.langchain.com/en/latest/runnables/langchain_core.runnables.config.RunnableConfig.html#langchain_core.runnables.config.RunnableConfig)): + +| **Field** | **Type** | **Description** | +|-------------|---------------------|--------------------------------------------------------------------------------------------------------------------| +| run_name | `str` | Name for the tracer run for this call. Defaults to the name of the class. | +| run_id | `UUID` | Unique identifier for the tracer run for this call. If not provided, a new UUID will be generated. | +| tags | `List[str]` | Tags for this call and any sub-calls (e.g., a Chain calling an LLM). You can use these to filter calls. | +| metadata | `Dict[str, Any]` | Metadata for this call and any sub-calls (e.g., a Chain calling an LLM). Keys should be strings, values should be JSON-serializable. | + +LangGraph graphs implement the [LangChain Runnable Interface](https://python.langchain.com/api_reference/core/runnables/langchain_core.runnables.base.Runnable.html) and accept a second argument (`RunnableConfig`) in methods like `invoke`, `ainvoke`, `stream` etc. + +The LangSmith platform will allow you to search and filter traces based on `run_name`, `run_id`, `tags` and `metadata`. + +## TLDR + +```python +import uuid +# Generate a random UUID -- it must be a UUID +config = {"run_id": uuid.uuid4()}, "tags": ["my_tag1"], "metadata": {"a": 5}} +# Works with all standard Runnable methods +# like invoke, batch, ainvoke, astream_events etc +graph.stream(inputs, config, stream_mode="values") +``` + +The rest of the how to guide will show a full agent. + +## Setup + +First, let's install the required packages and set our API keys + +```python +%%capture --no-stderr +%pip install --quiet -U langgraph langchain_openai +``` + +```python +import getpass +import os + + +def _set_env(var: str): + if not os.environ.get(var): + os.environ[var] = getpass.getpass(f"{var}: ") + + +_set_env("OPENAI_API_KEY") +_set_env("LANGSMITH_API_KEY") +``` + +!!! tip + Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. [LangSmith](https://docs.smith.langchain.com) lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started [here](https://docs.smith.langchain.com). + +## Define the graph + +For this example we will use the [prebuilt ReAct agent](https://langchain-ai.github.io/langgraph/how-tos/create-react-agent/). + +```python +from langchain_openai import ChatOpenAI +from typing import Literal +from langgraph.prebuilt import create_react_agent +from langchain_core.tools import tool + +# First we initialize the model we want to use. +model = ChatOpenAI(model="gpt-4o", temperature=0) + + +# For this tutorial we will use custom tool that returns pre-defined values for weather in two cities (NYC & SF) +@tool +def get_weather(city: Literal["nyc", "sf"]): + """Use this to get weather information.""" + if city == "nyc": + return "It might be cloudy in nyc" + elif city == "sf": + return "It's always sunny in sf" + else: + raise AssertionError("Unknown city") + + +tools = [get_weather] + + +# Define the graph +graph = create_react_agent(model, tools=tools) +``` + +## Run your graph + +Now that we've defined our graph let's run it once and view the trace in LangSmith. In order for our trace to be easily accessible in LangSmith, we will pass in a custom `run_id` in the config. + +This assumes that you have set your `LANGSMITH_API_KEY` environment variable. + +Note that you can also configure what project to trace to by setting the `LANGCHAIN_PROJECT` environment variable, by default runs will be traced to the `default` project. + +```python +import uuid + + +def print_stream(stream): + for s in stream: + message = s["messages"][-1] + if isinstance(message, tuple): + print(message) + else: + message.pretty_print() + + +inputs = {"messages": [("user", "what is the weather in sf")]} + +config = {"run_name": "agent_007", "tags": ["cats are awesome"]} + +print_stream(graph.stream(inputs, config, stream_mode="values")) +``` + +**Output:** +``` +================================ Human Message ================================== + +what is the weather in sf +================================== Ai Message =================================== +Tool Calls: + get_weather (call_9ZudXyMAdlUjptq9oMGtQo8o) + Call ID: call_9ZudXyMAdlUjptq9oMGtQo8o + Args: + city: sf +================================= Tool Message ================================== +Name: get_weather + +It's always sunny in sf +================================== Ai Message =================================== + +The weather in San Francisco is currently sunny. +``` + +## View the trace in LangSmith + +Now that we've ran our graph, let's head over to LangSmith and view our trace. First click into the project that you traced to (in our case the default project). You should see a run with the custom run name "agent_007". + +![LangSmith Trace View](assets/d38d1f2b-0f4c-4707-b531-a3c749de987f.png) + +In addition, you will be able to filter traces after the fact using the tags or metadata provided. For example, + +![LangSmith Filter View](assets/410e0089-2ab8-46bb-a61a-827187fd46b3.png) \ No newline at end of file diff --git a/docs/docs/tutorials/multi_agent/agent_supervisor.ipynb b/docs/docs/tutorials/multi_agent/agent_supervisor.ipynb deleted file mode 100644 index 2303872ef..000000000 --- a/docs/docs/tutorials/multi_agent/agent_supervisor.ipynb +++ /dev/null @@ -1,1403 +0,0 @@ -{ - "cells": [ - { - "attachments": { - "8ee0a8ce-f0a8-4019-b5bf-b20933e40956.png": { - "image/png": "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" - } - }, - "cell_type": "markdown", - "id": "a3e3ebc4-57af-4fe4-bdd3-36aff67bf276", - "metadata": {}, - "source": [ - "# Multi-agent supervisor\n", - "\n", - "[**Supervisor**](../../../concepts/multi_agent#supervisor) is a multi-agent architecture where **specialized** agents are coordinated by a central **supervisor agent**. The supervisor agent controls all communication flow and task delegation, making decisions about which agent to invoke based on the current context and task requirements.\n", - "\n", - "In this tutorial, you will build a supervisor system with two agents — a research and a math expert. By the end of the tutorial you will:\n", - "\n", - "1. Build specialized research and math agents\n", - "2. Build a supervisor for orchestrating them with the prebuilt [`langgraph-supervisor`](https://langchain-ai.github.io/langgraph/agents/multi-agent/#supervisor)\n", - "3. Build a supervisor from scratch\n", - "4. Implement advanced task delegation\n", - "\n", - "![diagram](attachment:8ee0a8ce-f0a8-4019-b5bf-b20933e40956.png)\n", - "\n", - "## Setup\n", - "\n", - "First, let's install required packages and set our API keys" - ] - }, - { - "cell_type": "code", - "execution_count": 1, - "id": "0d30b6f7-3bec-4d9f-af50-43dfdc81ae6c", - "metadata": {}, - "outputs": [], - "source": [ - "%%capture --no-stderr\n", - "%pip install -U langgraph langgraph-supervisor langchain-tavily \"langchain[openai]\"" - ] - }, - { - "cell_type": "code", - "execution_count": 2, - "id": "c84adef6-51b3-46c5-8490-c2a1ad967769", - "metadata": {}, - "outputs": [], - "source": [ - "import getpass\n", - "import os\n", - "\n", - "\n", - "def _set_if_undefined(var: str):\n", - " if not os.environ.get(var):\n", - " os.environ[var] = getpass.getpass(f\"Please provide your {var}\")\n", - "\n", - "\n", - "_set_if_undefined(\"OPENAI_API_KEY\")\n", - "_set_if_undefined(\"TAVILY_API_KEY\")" - ] - }, - { - "cell_type": "markdown", - "id": "be85e3ad", - "metadata": {}, - "source": [ - "
\n", - "

Set up LangSmith for LangGraph development

\n", - "

\n", - " Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started here. \n", - "

\n", - "
" - ] - }, - { - "cell_type": "markdown", - "id": "349a5673-962e-4de9-ac19-70ee426ef433", - "metadata": {}, - "source": [ - "## 1. Create worker agents" - ] - }, - { - "attachments": {}, - "cell_type": "markdown", - "id": "ad822cf5-56d4-47f7-ad64-d9fd3f57551e", - "metadata": {}, - "source": [ - "First, let's create our specialized worker agents — research agent and math agent:\n", - "\n", - "* Research agent will have access to a web search tool using [Tavily API](https://tavily.com/)\n", - "* Math agent will have access to simple math tools (`add`, `multiply`, `divide`)" - ] - }, - { - "attachments": {}, - "cell_type": "markdown", - "id": "5e34989a-8989-432e-b6aa-8de91ab83372", - "metadata": {}, - "source": [ - "### Research agent\n", - "\n", - "For web search, we will use `TavilySearch` tool from `langchain-tavily`:" - ] - }, - { - "cell_type": "code", - "execution_count": 3, - "id": "0fdc7345-2967-484b-881a-114dfac3f534", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Find events, attractions, deals, and more at nyctourism.com Skip Main Navigation Menu The Official Website of the City of New York Text Size Powered by Translate SearchSearch Primary Navigation The official website of NYC Home NYC Resources NYC311 Office of the Mayor Events Connect Jobs Search Office of the Mayor | Mayor's Bio | City of New York Secondary Navigation MayorBiographyNewsOfficials Eric L. Adams 110th Mayor of New York City Mayor Eric Adams has served the people of New York City as an NYPD officer, State Senator, Brooklyn Borough President, and now as the 110th Mayor of the City of New York. Mayor Eric Adams has served the people of New York City as an NYPD officer, State Senator, Brooklyn Borough President, and now as the 110th Mayor of the City of New York. He gave voice to a diverse coalition of working families in all five boroughs and is leading the fight to bring back New York City’s economy, reduce inequality, improve public safety, and build a stronger, healthier city that delivers for all New Yorkers. As the representative of one of the nation’s largest counties, Eric fought tirelessly to grow the local economy, invest in schools, reduce inequality, improve public safety, and advocate for smart policies and better government that delivers for all New Yorkers.\n" - ] - } - ], - "source": [ - "from langchain_tavily import TavilySearch\n", - "\n", - "web_search = TavilySearch(max_results=3)\n", - "web_search_results = web_search.invoke(\"who is the mayor of NYC?\")\n", - "\n", - "print(web_search_results[\"results\"][0][\"content\"])" - ] - }, - { - "cell_type": "markdown", - "id": "cc899eb6-b08e-41b8-b92b-75908c44e8b0", - "metadata": {}, - "source": [ - "To create individual worker agents, we will use LangGraph's prebuilt [agent](../../../agents/agents#basic-configuration)." - ] - }, - { - "cell_type": "code", - "execution_count": 4, - "id": "15e179e4-1a49-4c76-9235-0ca989eb0d2d", - "metadata": {}, - "outputs": [], - "source": [ - "from langgraph.prebuilt import create_react_agent\n", - "\n", - "research_agent = create_react_agent(\n", - " model=\"openai:gpt-4.1\",\n", - " tools=[web_search],\n", - " prompt=(\n", - " \"You are a research agent.\\n\\n\"\n", - " \"INSTRUCTIONS:\\n\"\n", - " \"- Assist ONLY with research-related tasks, DO NOT do any math\\n\"\n", - " \"- After you're done with your tasks, respond to the supervisor directly\\n\"\n", - " \"- Respond ONLY with the results of your work, do NOT include ANY other text.\"\n", - " ),\n", - " name=\"research_agent\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "023ae2f2-517b-4611-a0ca-506b8c0b734e", - "metadata": {}, - "source": [ - "Let's [run the agent](../../../agents/run_agents) to verify that it behaves as expected. \n", - "\n", - "\n", - "??? note \"We'll use `pretty_print_messages` helper to render the streamed agent outputs nicely\"\n", - "\n", - " ```python\n", - " from langchain_core.messages import convert_to_messages\n", - " \n", - " \n", - " def pretty_print_message(message, indent=False):\n", - " pretty_message = message.pretty_repr(html=True)\n", - " if not indent:\n", - " print(pretty_message)\n", - " return\n", - " \n", - " indented = \"\\n\".join(\"\\t\" + c for c in pretty_message.split(\"\\n\"))\n", - " print(indented)\n", - " \n", - " \n", - " def pretty_print_messages(update, last_message=False):\n", - " is_subgraph = False\n", - " if isinstance(update, tuple):\n", - " ns, update = update\n", - " # skip parent graph updates in the printouts\n", - " if len(ns) == 0:\n", - " return\n", - " \n", - " graph_id = ns[-1].split(\":\")[0]\n", - " print(f\"Update from subgraph {graph_id}:\")\n", - " print(\"\\n\")\n", - " is_subgraph = True\n", - " \n", - " for node_name, node_update in update.items():\n", - " update_label = f\"Update from node {node_name}:\"\n", - " if is_subgraph:\n", - " update_label = \"\\t\" + update_label\n", - " \n", - " print(update_label)\n", - " print(\"\\n\")\n", - " \n", - " messages = convert_to_messages(node_update[\"messages\"])\n", - " if last_message:\n", - " messages = messages[-1:]\n", - " \n", - " for m in messages:\n", - " pretty_print_message(m, indent=is_subgraph)\n", - " print(\"\\n\")\n", - " ```" - ] - }, - { - "cell_type": "code", - "execution_count": 5, - "id": "bfaa81e6", - "metadata": {}, - "outputs": [], - "source": [ - "# hide-cell\n", - "from langchain_core.messages import convert_to_messages\n", - "\n", - "\n", - "def pretty_print_message(message, indent=False):\n", - " pretty_message = message.pretty_repr(html=True)\n", - " if not indent:\n", - " print(pretty_message)\n", - " return\n", - "\n", - " indented = \"\\n\".join(\"\\t\" + c for c in pretty_message.split(\"\\n\"))\n", - " print(indented)\n", - "\n", - "\n", - "def pretty_print_messages(update, last_message=False):\n", - " is_subgraph = False\n", - " if isinstance(update, tuple):\n", - " ns, update = update\n", - " # skip parent graph updates in the printouts\n", - " if len(ns) == 0:\n", - " return\n", - "\n", - " graph_id = ns[-1].split(\":\")[0]\n", - " print(f\"Update from subgraph {graph_id}:\")\n", - " print(\"\\n\")\n", - " is_subgraph = True\n", - "\n", - " for node_name, node_update in update.items():\n", - " update_label = f\"Update from node {node_name}:\"\n", - " if is_subgraph:\n", - " update_label = \"\\t\" + update_label\n", - "\n", - " print(update_label)\n", - " print(\"\\n\")\n", - "\n", - " messages = convert_to_messages(node_update[\"messages\"])\n", - " if last_message:\n", - " messages = messages[-1:]\n", - "\n", - " for m in messages:\n", - " pretty_print_message(m, indent=is_subgraph)\n", - " print(\"\\n\")" - ] - }, - { - "cell_type": "code", - "execution_count": 6, - "id": "71474c5b-6a9b-460e-bf30-77831976aaf9", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Update from node agent:\n", - "\n", - "\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: research_agent\n", - "Tool Calls:\n", - " tavily_search (call_U748rQhQXT36sjhbkYLSXQtJ)\n", - " Call ID: call_U748rQhQXT36sjhbkYLSXQtJ\n", - " Args:\n", - " query: current mayor of New York City\n", - " search_depth: basic\n", - "\n", - "\n", - "Update from node tools:\n", - "\n", - "\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: tavily_search\n", - "\n", - "{\"query\": \"current mayor of New York City\", \"follow_up_questions\": null, \"answer\": null, \"images\": [], \"results\": [{\"title\": \"List of mayors of New York City - Wikipedia\", \"url\": \"https://en.wikipedia.org/wiki/List_of_mayors_of_New_York_City\", \"content\": \"The mayor of New York City is the chief executive of the Government of New York City, as stipulated by New York City's charter.The current officeholder, the 110th in the sequence of regular mayors, is Eric Adams, a member of the Democratic Party.. During the Dutch colonial period from 1624 to 1664, New Amsterdam was governed by the Director of New Netherland.\", \"score\": 0.9039154, \"raw_content\": null}, {\"title\": \"Office of the Mayor | Mayor's Bio | City of New York - NYC.gov\", \"url\": \"https://www.nyc.gov/office-of-the-mayor/bio.page\", \"content\": \"Mayor Eric Adams has served the people of New York City as an NYPD officer, State Senator, Brooklyn Borough President, and now as the 110th Mayor of the City of New York. He gave voice to a diverse coalition of working families in all five boroughs and is leading the fight to bring back New York City's economy, reduce inequality, improve\", \"score\": 0.8405867, \"raw_content\": null}, {\"title\": \"Eric Adams - Wikipedia\", \"url\": \"https://en.wikipedia.org/wiki/Eric_Adams\", \"content\": \"Eric Leroy Adams (born September 1, 1960) is an American politician and former police officer who has served as the 110th mayor of New York City since 2022. Adams was an officer in the New York City Transit Police and then the New York City Police Department (NYPD) for more than 20 years, retiring at the rank of captain.He served in the New York State Senate from 2006 to 2013, representing the\", \"score\": 0.77731717, \"raw_content\": null}], \"response_time\": 1.81}\n", - "\n", - "\n", - "Update from node agent:\n", - "\n", - "\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: research_agent\n", - "\n", - "The current mayor of New York City is Eric Adams.\n", - "\n", - "\n" - ] - } - ], - "source": [ - "for chunk in research_agent.stream(\n", - " {\"messages\": [{\"role\": \"user\", \"content\": \"who is the mayor of NYC?\"}]}\n", - "):\n", - " pretty_print_messages(chunk)" - ] - }, - { - "attachments": {}, - "cell_type": "markdown", - "id": "95e76c28-e9c9-42b7-8381-5e2a07f900cb", - "metadata": {}, - "source": [ - "### Math agent\n", - "\n", - "For math agent tools we will use [vanilla Python functions](../../../agents/tools#define-simple-tools):" - ] - }, - { - "cell_type": "code", - "execution_count": 7, - "id": "95910bb8-7144-4015-9ea4-d97d0c22db12", - "metadata": {}, - "outputs": [], - "source": [ - "def add(a: float, b: float):\n", - " \"\"\"Add two numbers.\"\"\"\n", - " return a + b\n", - "\n", - "\n", - "def multiply(a: float, b: float):\n", - " \"\"\"Multiply two numbers.\"\"\"\n", - " return a * b\n", - "\n", - "\n", - "def divide(a: float, b: float):\n", - " \"\"\"Divide two numbers.\"\"\"\n", - " return a / b\n", - "\n", - "\n", - "math_agent = create_react_agent(\n", - " model=\"openai:gpt-4.1\",\n", - " tools=[add, multiply, divide],\n", - " prompt=(\n", - " \"You are a math agent.\\n\\n\"\n", - " \"INSTRUCTIONS:\\n\"\n", - " \"- Assist ONLY with math-related tasks\\n\"\n", - " \"- After you're done with your tasks, respond to the supervisor directly\\n\"\n", - " \"- Respond ONLY with the results of your work, do NOT include ANY other text.\"\n", - " ),\n", - " name=\"math_agent\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "0df97bc0-f38b-4893-a135-c16cecbfa759", - "metadata": {}, - "source": [ - "Let's run the math agent:" - ] - }, - { - "cell_type": "code", - "execution_count": 8, - "id": "f803943c-6dd2-4b42-93ba-0210ec276132", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Update from node agent:\n", - "\n", - "\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: math_agent\n", - "Tool Calls:\n", - " add (call_p6OVLDHB4LyCNCxPOZzWR15v)\n", - " Call ID: call_p6OVLDHB4LyCNCxPOZzWR15v\n", - " Args:\n", - " a: 3\n", - " b: 5\n", - "\n", - "\n", - "Update from node tools:\n", - "\n", - "\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: add\n", - "\n", - "8.0\n", - "\n", - "\n", - "Update from node agent:\n", - "\n", - "\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: math_agent\n", - "Tool Calls:\n", - " multiply (call_EoaWHMLFZAX4AkajQCtZvbli)\n", - " Call ID: call_EoaWHMLFZAX4AkajQCtZvbli\n", - " Args:\n", - " a: 8\n", - " b: 7\n", - "\n", - "\n", - "Update from node tools:\n", - "\n", - "\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: multiply\n", - "\n", - "56.0\n", - "\n", - "\n", - "Update from node agent:\n", - "\n", - "\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: math_agent\n", - "\n", - "56\n", - "\n", - "\n" - ] - } - ], - "source": [ - "for chunk in math_agent.stream(\n", - " {\"messages\": [{\"role\": \"user\", \"content\": \"what's (3 + 5) x 7\"}]}\n", - "):\n", - " pretty_print_messages(chunk)" - ] - }, - { - "cell_type": "markdown", - "id": "dceefc4c-95e1-4565-8946-4c113a11416a", - "metadata": {}, - "source": [ - "## 2. Create supervisor with `langgraph-supervisor`" - ] - }, - { - "cell_type": "markdown", - "id": "eb5637a4-8ecb-4c2c-82f5-3674205e37b5", - "metadata": {}, - "source": [ - "To implement out multi-agent system, we will use [`create_supervisor`][langgraph_supervisor.supervisor.create_supervisor] from the prebuilt `langgraph-supervisor` library:" - ] - }, - { - "cell_type": "code", - "execution_count": 9, - "id": "6f32a820-1f3d-4698-831b-46252e2316bc", - "metadata": {}, - "outputs": [], - "source": [ - "from langgraph_supervisor import create_supervisor\n", - "from langchain.chat_models import init_chat_model\n", - "\n", - "supervisor = create_supervisor(\n", - " model=init_chat_model(\"openai:gpt-4.1\"),\n", - " agents=[research_agent, math_agent],\n", - " prompt=(\n", - " \"You are a supervisor managing two agents:\\n\"\n", - " \"- a research agent. Assign research-related tasks to this agent\\n\"\n", - " \"- a math agent. Assign math-related tasks to this agent\\n\"\n", - " \"Assign work to one agent at a time, do not call agents in parallel.\\n\"\n", - " \"Do not do any work yourself.\"\n", - " ),\n", - " add_handoff_back_messages=True,\n", - " output_mode=\"full_history\",\n", - ").compile()" - ] - }, - { - "cell_type": "code", - "execution_count": 10, - "id": "4c45a3c2-4d9e-4760-87bc-3cc2c007a167", - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "from IPython.display import display, Image\n", - "\n", - "display(Image(supervisor.get_graph().draw_mermaid_png()))" - ] - }, - { - "cell_type": "markdown", - "id": "3a94c3d1-3617-4d68-ac9a-b9bb2c05a604", - "metadata": {}, - "source": [ - "Let's now run it with a query that requires both agents:\n", - "\n", - "* research agent will look up the necessary GDP information\n", - "* math agent will perform division to find the percentage of NY state GDP, as requested" - ] - }, - { - "cell_type": "code", - "execution_count": 11, - "id": "89b77171-58d0-4b36-8cb3-b69d69e4bdf9", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Update from node supervisor:\n", - "\n", - "\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: transfer_to_research_agent\n", - "\n", - "Successfully transferred to research_agent\n", - "\n", - "\n", - "Update from node research_agent:\n", - "\n", - "\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: transfer_back_to_supervisor\n", - "\n", - "Successfully transferred back to supervisor\n", - "\n", - "\n", - "Update from node supervisor:\n", - "\n", - "\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: transfer_to_math_agent\n", - "\n", - "Successfully transferred to math_agent\n", - "\n", - "\n", - "Update from node math_agent:\n", - "\n", - "\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: transfer_back_to_supervisor\n", - "\n", - "Successfully transferred back to supervisor\n", - "\n", - "\n", - "Update from node supervisor:\n", - "\n", - "\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: supervisor\n", - "\n", - "In 2024, the US GDP was $29.18 trillion and New York State's GDP was $2.297 trillion. New York State accounted for approximately 7.87% of the total US GDP in 2024.\n", - "\n", - "\n" - ] - } - ], - "source": [ - "for chunk in supervisor.stream(\n", - " {\n", - " \"messages\": [\n", - " {\n", - " \"role\": \"user\",\n", - " \"content\": \"find US and New York state GDP in 2024. what % of US GDP was New York state?\",\n", - " }\n", - " ]\n", - " },\n", - "):\n", - " pretty_print_messages(chunk, last_message=True)\n", - "\n", - "final_message_history = chunk[\"supervisor\"][\"messages\"]" - ] - }, - { - "cell_type": "markdown", - "id": "599b8d13-70a9-40f3-99b9-5e3d35cc6f56", - "metadata": {}, - "source": [ - "## 3. Create supervisor from scratch" - ] - }, - { - "cell_type": "markdown", - "id": "e23764f8-f818-45f3-a5ee-4cb7b7002336", - "metadata": {}, - "source": [ - "Let's now implement this same multi-agent system from scratch. We will need to:\n", - "\n", - "1. Set up how the supervisor [communicates](#set-up-agent-communication) with individual agents\n", - "2. Create the [supervisor agent](#create-supervisor-agent)\n", - "3. [Combine]() supervisor and worker agents into a single multi-agent graph." - ] - }, - { - "cell_type": "markdown", - "id": "78d32089-0259-4aa3-a473-077ecce342dd", - "metadata": {}, - "source": [ - "### Set up agent communication" - ] - }, - { - "cell_type": "markdown", - "id": "a6c6b436-e95b-476b-b427-7b02d8384cc5", - "metadata": {}, - "source": [ - "We will need to define a way for the supervisor agent to communicate with the worker agents. A common way to implement this in multi-agent architectures is using **handoffs**, where one agent *hands off* control to another. Handoffs allow you to specify:\n", - "\n", - "- **destination**: target agent to transfer to\n", - "- **payload**: information to pass to that agent\n", - "\n", - "We will implement handoffs via **handoff tools** and give these tools to the supervisor agent: when the supervisor calls these tools, it will hand off control to a worker agent, passing the full message history to that agent." - ] - }, - { - "cell_type": "code", - "execution_count": 12, - "id": "7873a6c3-7de0-46d1-96ce-2dca7c62245a", - "metadata": {}, - "outputs": [], - "source": [ - "from typing import Annotated\n", - "from langchain_core.tools import tool, InjectedToolCallId\n", - "from langgraph.prebuilt import InjectedState\n", - "from langgraph.graph import StateGraph, START, MessagesState\n", - "from langgraph.types import Command\n", - "\n", - "\n", - "def create_handoff_tool(*, agent_name: str, description: str | None = None):\n", - " name = f\"transfer_to_{agent_name}\"\n", - " description = description or f\"Ask {agent_name} for help.\"\n", - "\n", - " @tool(name, description=description)\n", - " def handoff_tool(\n", - " state: Annotated[MessagesState, InjectedState],\n", - " tool_call_id: Annotated[str, InjectedToolCallId],\n", - " ) -> Command:\n", - " tool_message = {\n", - " \"role\": \"tool\",\n", - " \"content\": f\"Successfully transferred to {agent_name}\",\n", - " \"name\": name,\n", - " \"tool_call_id\": tool_call_id,\n", - " }\n", - " # highlight-next-line\n", - " return Command(\n", - " # highlight-next-line\n", - " goto=agent_name, # (1)!\n", - " # highlight-next-line\n", - " update={**state, \"messages\": state[\"messages\"] + [tool_message]}, # (2)!\n", - " # highlight-next-line\n", - " graph=Command.PARENT, # (3)!\n", - " )\n", - "\n", - " return handoff_tool\n", - "\n", - "\n", - "# Handoffs\n", - "assign_to_research_agent = create_handoff_tool(\n", - " agent_name=\"research_agent\",\n", - " description=\"Assign task to a researcher agent.\",\n", - ")\n", - "\n", - "assign_to_math_agent = create_handoff_tool(\n", - " agent_name=\"math_agent\",\n", - " description=\"Assign task to a math agent.\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "3bc810a4-1866-4915-abd9-12655954e673", - "metadata": {}, - "source": [ - "1. Name of the agent or node to hand off to.\n", - "2. Take the agent's messages and add them to the parent's state as part of the handoff. The next agent will see the parent state.\n", - "3. Indicate to LangGraph that we need to navigate to agent node in a **parent** multi-agent graph." - ] - }, - { - "cell_type": "markdown", - "id": "6a7bd301-c954-4189-87e2-b3bf42d187c1", - "metadata": {}, - "source": [ - "### Create supervisor agent" - ] - }, - { - "cell_type": "markdown", - "id": "ca787ce6-46f9-49c3-8f9f-ddb7ff52cdeb", - "metadata": {}, - "source": [ - "Then, let's create the supervisor agent with the handoff tools we just defined. We will use the prebuilt [`create_react_agent`][langgraph.prebuilt.chat_agent_executor.create_react_agent]:" - ] - }, - { - "cell_type": "code", - "execution_count": 13, - "id": "de5d1ff8-e7a0-4efe-93fe-d304eaef29b6", - "metadata": {}, - "outputs": [], - "source": [ - "supervisor_agent = create_react_agent(\n", - " model=\"openai:gpt-4.1\",\n", - " tools=[assign_to_research_agent, assign_to_math_agent],\n", - " prompt=(\n", - " \"You are a supervisor managing two agents:\\n\"\n", - " \"- a research agent. Assign research-related tasks to this agent\\n\"\n", - " \"- a math agent. Assign math-related tasks to this agent\\n\"\n", - " \"Assign work to one agent at a time, do not call agents in parallel.\\n\"\n", - " \"Do not do any work yourself.\"\n", - " ),\n", - " name=\"supervisor\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "dd4bc4e2-e133-42fd-a47f-6fb63fa51697", - "metadata": {}, - "source": [ - "### Create multi-agent graph" - ] - }, - { - "cell_type": "markdown", - "id": "8e6dcae1-eb91-4a07-aef8-53cbed14a002", - "metadata": {}, - "source": [ - "Putting this all together, let's create a graph for our overall multi-agent system. We will add the supervisor and the individual agents as [subgraph](../../../concepts/low_level#subgraphs) nodes." - ] - }, - { - "cell_type": "code", - "execution_count": 14, - "id": "584443af-587a-44f3-a320-f942868191e4", - "metadata": {}, - "outputs": [], - "source": [ - "from langgraph.graph import END\n", - "\n", - "# Define the multi-agent supervisor graph\n", - "supervisor = (\n", - " StateGraph(MessagesState)\n", - " # NOTE: `destinations` is only needed for visualization and doesn't affect runtime behavior\n", - " .add_node(supervisor_agent, destinations=(\"research_agent\", \"math_agent\", END))\n", - " .add_node(research_agent)\n", - " .add_node(math_agent)\n", - " .add_edge(START, \"supervisor\")\n", - " # always return back to the supervisor\n", - " .add_edge(\"research_agent\", \"supervisor\")\n", - " .add_edge(\"math_agent\", \"supervisor\")\n", - " .compile()\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "12576211-a946-46ec-bf81-24215597a1fe", - "metadata": {}, - "source": [ - "Notice that we've added explicit [edges](../../../concepts/low_level#edges) from worker agents back to the supervisor — this means that they are guaranteed to return control back to the supervisor. If you want the agents to respond directly to the user (i.e., turn the system into a router, you can remove these edges)." - ] - }, - { - "cell_type": "code", - "execution_count": 15, - "id": "0175fe14-5854-4197-b7e8-559335d0f81b", - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "from IPython.display import display, Image\n", - "\n", - "display(Image(supervisor.get_graph().draw_mermaid_png()))" - ] - }, - { - "cell_type": "markdown", - "id": "d36496de-7121-4c49-8cb6-58c943c66628", - "metadata": {}, - "source": [ - "With the multi-agent graph created, let's now run it!" - ] - }, - { - "cell_type": "code", - "execution_count": 16, - "id": "56ba78e9-d9c1-457c-a073-d606d5d3e013", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Update from node supervisor:\n", - "\n", - "\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: transfer_to_research_agent\n", - "\n", - "Successfully transferred to research_agent\n", - "\n", - "\n", - "Update from node research_agent:\n", - "\n", - "\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: research_agent\n", - "\n", - "- US GDP in 2024 is projected to be about $28.18 trillion USD (Statista; CBO projection).\n", - "- New York State's nominal GDP for 2024 is estimated at approximately $2.16 trillion USD (various economic reports).\n", - "- New York State's share of US GDP in 2024 is roughly 7.7%.\n", - "\n", - "Sources:\n", - "- https://www.statista.com/statistics/216985/forecast-of-us-gross-domestic-product/\n", - "- https://nyassembly.gov/Reports/WAM/2025economic_revenue/2025_report.pdf?v=1740533306\n", - "\n", - "\n", - "Update from node supervisor:\n", - "\n", - "\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: transfer_to_math_agent\n", - "\n", - "Successfully transferred to math_agent\n", - "\n", - "\n", - "Update from node math_agent:\n", - "\n", - "\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: math_agent\n", - "\n", - "US GDP in 2024: $28.18 trillion\n", - "New York State GDP in 2024: $2.16 trillion\n", - "Percentage of US GDP from New York State: 7.67%\n", - "\n", - "\n", - "Update from node supervisor:\n", - "\n", - "\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: supervisor\n", - "\n", - "Here are your results:\n", - "\n", - "- 2024 US GDP (projected): $28.18 trillion USD\n", - "- 2024 New York State GDP (estimated): $2.16 trillion USD\n", - "- New York State's share of US GDP: approximately 7.7%\n", - "\n", - "If you need the calculation steps or sources, let me know!\n", - "\n", - "\n" - ] - } - ], - "source": [ - "for chunk in supervisor.stream(\n", - " {\n", - " \"messages\": [\n", - " {\n", - " \"role\": \"user\",\n", - " \"content\": \"find US and New York state GDP in 2024. what % of US GDP was New York state?\",\n", - " }\n", - " ]\n", - " },\n", - "):\n", - " pretty_print_messages(chunk, last_message=True)\n", - "\n", - "final_message_history = chunk[\"supervisor\"][\"messages\"]" - ] - }, - { - "cell_type": "markdown", - "id": "dac583a8-af2c-402f-801e-15a6bc318e44", - "metadata": {}, - "source": [ - "Let's examine the full resulting message history:" - ] - }, - { - "cell_type": "code", - "execution_count": 17, - "id": "1eff504c-7437-477e-ab28-936c1238eaca", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "================================\u001b[1m Human Message \u001b[0m=================================\n", - "\n", - "find US and New York state GDP in 2024. what % of US GDP was New York state?\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: supervisor\n", - "Tool Calls:\n", - " transfer_to_research_agent (call_KlGgvF5ahlAbjX8d2kHFjsC3)\n", - " Call ID: call_KlGgvF5ahlAbjX8d2kHFjsC3\n", - " Args:\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: transfer_to_research_agent\n", - "\n", - "Successfully transferred to research_agent\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: research_agent\n", - "Tool Calls:\n", - " tavily_search (call_ZOaTVUA6DKrOjWQldLhtrsO2)\n", - " Call ID: call_ZOaTVUA6DKrOjWQldLhtrsO2\n", - " Args:\n", - " query: US GDP 2024 estimate or actual\n", - " search_depth: advanced\n", - " tavily_search (call_QsRAasxW9K03lTlqjuhNLFbZ)\n", - " Call ID: call_QsRAasxW9K03lTlqjuhNLFbZ\n", - " Args:\n", - " query: New York state GDP 2024 estimate or actual\n", - " search_depth: advanced\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: tavily_search\n", - "\n", - "{\"query\": \"US GDP 2024 estimate or actual\", \"follow_up_questions\": null, \"answer\": null, \"images\": [], \"results\": [{\"url\": \"https://www.advisorperspectives.com/dshort/updates/2025/05/29/gdp-gross-domestic-product-q1-2025-second-estimate\", \"title\": \"Q1 GDP Second Estimate: Real GDP at -0.2%, Higher Than Expected\", \"content\": \"> Real gross domestic product (GDP) decreased at an annual rate of 0.2 percent in the first quarter of 2025 (January, February, and March), according to the second estimate released by the U.S. Bureau of Economic Analysis. In the fourth quarter of 2024, real GDP increased 2.4 percent. The decrease in real GDP in the first quarter primarily reflected an increase in imports, which are a subtraction in the calculation of GDP, and a decrease in government spending. These movements were partly [...] by [Harry Mamaysky](https://www.advisorperspectives.com/search?author=Harry%20Mamaysky) of [QuantStreet Capital](https://www.advisorperspectives.com/firm/quantstreet-capital), 10/14/2024\\n\\n#### [The Worst of Allan Roth](https://www.advisorperspectives.com/articles/2024/10/21/worst-of-allan-roth/)\\n\\n by [Allan Roth](https://www.advisorperspectives.com/search?author=Allan%20Roth), 10/21/2024 [...] by [Juan Pablo Spinetto](https://www.advisorperspectives.com/search?author=Juan%20Pablo%20Spinetto) of [Bloomberg News](https://www.advisorperspectives.com/firm/bloomberg-news), 04/05/2025\\n\\n#### [Your Clients Are Clueless About Their Risk Tolerance](https://www.advisorperspectives.com/articles/2024/10/23/clients-clueless-about-risk-tolerance/)\\n\\n by [Dan Solin](https://www.advisorperspectives.com/search?author=Dan%20Solin), 10/23/2024\", \"score\": 0.78650844, \"raw_content\": null}, {\"url\": \"https://www.statista.com/statistics/216985/forecast-of-us-gross-domestic-product/\", \"title\": \"GDP forecast U.S. 2034 - Statista\", \"content\": \"[](https://www.statista.com/statistics/216985/forecast-of-us-gross-domestic-product/#statisticContainer) This graph shows a forecast of the gross domestic product of the United States of America for fiscal years 2024 to 2034. GDP refers to the market value of all final goods and services produced within a country in a given period. According to the CBO, the United States GDP will increase steadily over the next decade from 28.18 trillion U.S. dollars in 2023 to 41.65 trillion U.S. dollars in [...] US Congressional Budget Office. (February 7, 2024). Forecast of the gross domestic product of the United States from fiscal year 2024 to fiscal year 2034 (in billion U.S. dollars) [Graph]. In Statista. Retrieved May 20, 2025, from https://www.statista.com/statistics/216985/forecast-of-us-gross-domestic-product/ [...] US Congressional Budget Office. (2024). Forecast of the gross domestic product of the United States from fiscal year 2024 to fiscal year 2034 (in billion U.S. dollars). Statista. Statista Inc.. Accessed: May 20, 2025. https://www.statista.com/statistics/216985/forecast-of-us-gross-domestic-product/\", \"score\": 0.7576693, \"raw_content\": null}, {\"url\": \"https://www.bea.gov/data/gdp/gross-domestic-product\", \"title\": \"Gross Domestic Product | U.S. Bureau of Economic Analysis (BEA)\", \"content\": \"Real gross domestic product (GDP) decreased at an annual rate of 0.2 percent in the first quarter of 2025 (January, February, and March), according to the second estimate released by the U.S. Bureau of Economic Analysis. In the fourth quarter of 2024, real GDP increased 2.4 percent. The decrease in real GDP in the first quarter primarily reflected an increase in imports, which are a subtraction in the calculation of GDP, and a decrease in government spending. These movements were partly offset\", \"score\": 0.75723875, \"raw_content\": null}], \"response_time\": 2.48}\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: tavily_search\n", - "\n", - "{\"query\": \"New York state GDP 2024 estimate or actual\", \"follow_up_questions\": null, \"answer\": null, \"images\": [], \"results\": [{\"url\": \"https://comptroller.nyc.gov/reports/annual-state-of-the-citys-economy-and-finances-2024/\", \"title\": \"Annual State of the City's Economy and Finances 2024\", \"content\": \"[[1]](https://comptroller.nyc.gov/reports/annual-state-of-the-citys-economy-and-finances-2024/#_ftnref1) At that time, the Comptroller’s Office expected five-year cumulative real GDP growth, 2020 to 2024, of 8.7 percent while the mayor expected 9.5 percent.\\n\\n[[2]](https://comptroller.nyc.gov/reports/annual-state-of-the-citys-economy-and-finances-2024/#_ftnref2) U.S. Bureau of Labor Statistics Local Area Unemployment Statistics (LAUS). [...] [[19]](https://comptroller.nyc.gov/reports/annual-state-of-the-citys-economy-and-finances-2024/#_ftnref19) FY 2024 actuals include $742 million in transfers to the capital fund through Interfund Agreements, as detailed in Schedule G6 of the FY 2024 Annual Comprehensive Financial Report. [...] (constant 2019 dollars)**22,550 15,322 13,596 19,451 21,211 21,096\\n**_% change_**(32.1%)(11.3%)43.1%9.0%(0.5%)\\n\\nSource: NY State Department of Taxation and Finance and Office of the New York City Comptroller\", \"score\": 0.82795376, \"raw_content\": null}], \"response_time\": 2.76}\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: research_agent\n", - "Tool Calls:\n", - " tavily_search (call_DxuGWv9Ii19fPys582KcSLOg)\n", - " Call ID: call_DxuGWv9Ii19fPys582KcSLOg\n", - " Args:\n", - " query: New York state GDP 2024 (in dollars, not percentage growth or projections)\n", - " search_depth: advanced\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: tavily_search\n", - "\n", - "{\"query\": \"New York state GDP 2024 (in dollars, not percentage growth or projections)\", \"follow_up_questions\": null, \"answer\": null, \"images\": [], \"results\": [{\"url\": \"https://comptroller.nyc.gov/reports/annual-state-of-the-citys-economy-and-finances-2024/\", \"title\": \"Annual State of the City's Economy and Finances 2024\", \"content\": \"(constant 2019 dollars)**22,550 15,322 13,596 19,451 21,211 21,096\\n**_% change_**(32.1%)(11.3%)43.1%9.0%(0.5%)\\n\\nSource: NY State Department of Taxation and Finance and Office of the New York City Comptroller [...] | Retail Trade | 301,565 | $56,208 | 14.4 | (0.6) | (5.2) | 2.7 | 2.9 | 2.2 | 2.1 |\\n| Leisure & Hospitality | 431,227 | $56,086 | 96.2 | 32.4 | 15.4 | 8.9 | 3.8 | 4.5 | 7.3 |\\n| **Total: Low Wage Sectors** | **1,997,491** | **$62,185** | **184.6** | **102.2** | **91.6** | **62.7** | **32.0** | **26.8** | **29.3** |\\n| **Total NYC Employment** | **4,485,695** | **$114,294** | **307.7** | **119.9** | **78.1** | **88.0** | **65.1** | **53.2** | **51.4** | [...] [[1]](https://comptroller.nyc.gov/reports/annual-state-of-the-citys-economy-and-finances-2024/#_ftnref1) At that time, the Comptroller’s Office expected five-year cumulative real GDP growth, 2020 to 2024, of 8.7 percent while the mayor expected 9.5 percent.\\n\\n[[2]](https://comptroller.nyc.gov/reports/annual-state-of-the-citys-economy-and-finances-2024/#_ftnref2) U.S. Bureau of Labor Statistics Local Area Unemployment Statistics (LAUS).\", \"score\": 0.8098936, \"raw_content\": null}, {\"url\": \"https://nyassembly.gov/Reports/WAM/2025economic_revenue/2025_report.pdf?v=1740533306\", \"title\": \"[PDF] New York State Economic and Revenue Report\", \"content\": \"Sources: Federal Reserve; NYS Assembly Ways and Means Committee staff. NYS ASSEMBLY| U.S. ECONOMIC FORECAST AT A GLANCE| 15 Actual Actual Actual Forecast Forecast 2022 2023 2024 2025 2026 Real GDP 2.5 2.9 2.8 2.4 2.1 Consumption 3.0 2.5 2.8 2.7 2.1 Investment 6.0 0.1 4.0 3.1 3.8 Exports 7.5 2.8 3.2 2.4 1.1 Imports 8.6 (1.2) 5.4 3.5 1.3 Government (1.1) 3.9 3.4 1.5 0.1 Federal (3.2) 2.9 2.5 1.7 0.2 State and Local 0.2 4.4 3.9 1.4 0.1 Personal Income 3.1 5.9 5.5 5.0 4.9 Wages & Salaries 7.8 5.4 [...] in 2024 on an annualized basis. National output, measured with inflation-adjusted Gross Domestic Product (or real GDP), continues to grow at solid rates, surpassing its pre-pandemic peak by 12.1 percent, as of the fourth quarter of 2024 (see Figure 2). However, the pace of average quarterly growth has decelerated from 3.2 percent in 2023 to 2.5 percent in 2024 on an annualized basis. 121.7 114.5 115.3 112.1 109.2 106.3 70 75 80 85 90 95 100 105 110 115 120 125 130 2019Q4 2020Q2 2020Q4 2021Q2 [...] three-month moving average of 2.5 million in December 2023 to 872,000 in December 2024. The ratio of job openings to unemployed persons, another key metric for assessing the balance between labor demand and supply, has also steadily declined in recent months. By December 2024, this ratio had dropped to 1.10, down from its peak of 2.03 in March 2022. 24 | U.S. ECONOMIC FORECAST ANALYSIS | NYS ASSEMBLY With diminishing supply shortages in the labor market, growth of labor and other incomes will\", \"score\": 0.8051581, \"raw_content\": null}], \"response_time\": 2.86}\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: research_agent\n", - "Tool Calls:\n", - " tavily_search (call_wjYTJR9EnmCqzKrkrONpxYO0)\n", - " Call ID: call_wjYTJR9EnmCqzKrkrONpxYO0\n", - " Args:\n", - " query: New York State GDP (nominal) in US dollars for 2024 (actual or forecast)\n", - " search_depth: advanced\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: tavily_search\n", - "\n", - "{\"query\": \"New York State GDP (nominal) in US dollars for 2024 (actual or forecast)\", \"follow_up_questions\": null, \"answer\": null, \"images\": [], \"results\": [{\"url\": \"https://nyassembly.gov/Reports/WAM/2025economic_revenue/2025_report.pdf?v=1740533306\", \"title\": \"[PDF] New York State Economic and Revenue Report\", \"content\": \"Sources: Federal Reserve; NYS Assembly Ways and Means Committee staff. NYS ASSEMBLY| U.S. ECONOMIC FORECAST AT A GLANCE| 15 Actual Actual Actual Forecast Forecast 2022 2023 2024 2025 2026 Real GDP 2.5 2.9 2.8 2.4 2.1 Consumption 3.0 2.5 2.8 2.7 2.1 Investment 6.0 0.1 4.0 3.1 3.8 Exports 7.5 2.8 3.2 2.4 1.1 Imports 8.6 (1.2) 5.4 3.5 1.3 Government (1.1) 3.9 3.4 1.5 0.1 Federal (3.2) 2.9 2.5 1.7 0.2 State and Local 0.2 4.4 3.9 1.4 0.1 Personal Income 3.1 5.9 5.5 5.0 4.9 Wages & Salaries 7.8 5.4 [...] New York State Economic Outlook (Percent Change) Actual Estimate Forecast Forecast 2023-24 2024-25 2025-26 2026-27 Employment Percent Change 1.8 1.5 1.0 0.8 Level (Thousands) 9,441.6 9,589.7 9,694.3 9,773.6 Personal Income Percent Change 5.2 5.8 4.6 4.4 Level (Billions) 1,581.5 1,671.1 1,754.9 1,835.0 Total Wages Percent Change 4.2 6.7 4.4 4.0 Level (Billions) 864.0 915.9 964.3 1,005.1 Base Wages Percent Change 5.1 5.4 4.4 4.3 Level (Billions) 764.0 803.1 840.7 876.6 Variable Compensation [...] is 0.3 percentage point above the forecast of the Division of the Budget. The Committee’s forecast is 0.1 percentage point above S&P Global forecast and Moody’s Analytics’ projection. The forecast is 0.2 percentage point above the Blue Chip Consensus forecast. Actual Estimate Forecast Forecast 2023 2024 2025 2026 Ways and Means 2.9 2.8 2.4 2.1 Division of the Budget 2.9 2.8 2.1 2.0 Blue Chip Consensus 2.9 2.8 2.2 2.0 Moody's Analytics 2.9 2.8 2.3 1.6 S&P Global 2.9 2.8 2.3 2.0 Sources: NYS\", \"score\": 0.846159, \"raw_content\": null}, {\"url\": \"https://comptroller.nyc.gov/reports/annual-state-of-the-citys-economy-and-finances-2024/\", \"title\": \"Annual State of the City's Economy and Finances 2024\", \"content\": \"(constant 2019 dollars)**22,550 15,322 13,596 19,451 21,211 21,096\\n**_% change_**(32.1%)(11.3%)43.1%9.0%(0.5%)\\n\\nSource: NY State Department of Taxation and Finance and Office of the New York City Comptroller [...] [[1]](https://comptroller.nyc.gov/reports/annual-state-of-the-citys-economy-and-finances-2024/#_ftnref1) At that time, the Comptroller’s Office expected five-year cumulative real GDP growth, 2020 to 2024, of 8.7 percent while the mayor expected 9.5 percent.\\n\\n[[2]](https://comptroller.nyc.gov/reports/annual-state-of-the-citys-economy-and-finances-2024/#_ftnref2) U.S. Bureau of Labor Statistics Local Area Unemployment Statistics (LAUS). [...] | Retail Trade | 301,565 | $56,208 | 14.4 | (0.6) | (5.2) | 2.7 | 2.9 | 2.2 | 2.1 |\\n| Leisure & Hospitality | 431,227 | $56,086 | 96.2 | 32.4 | 15.4 | 8.9 | 3.8 | 4.5 | 7.3 |\\n| **Total: Low Wage Sectors** | **1,997,491** | **$62,185** | **184.6** | **102.2** | **91.6** | **62.7** | **32.0** | **26.8** | **29.3** |\\n| **Total NYC Employment** | **4,485,695** | **$114,294** | **307.7** | **119.9** | **78.1** | **88.0** | **65.1** | **53.2** | **51.4** |\", \"score\": 0.7992077, \"raw_content\": null}], \"response_time\": 2.65}\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: research_agent\n", - "\n", - "- US GDP in 2024 is projected to be about $28.18 trillion USD (Statista; CBO projection).\n", - "- New York State's nominal GDP for 2024 is estimated at approximately $2.16 trillion USD (various economic reports).\n", - "- New York State's share of US GDP in 2024 is roughly 7.7%.\n", - "\n", - "Sources:\n", - "- https://www.statista.com/statistics/216985/forecast-of-us-gross-domestic-product/\n", - "- https://nyassembly.gov/Reports/WAM/2025economic_revenue/2025_report.pdf?v=1740533306\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: supervisor\n", - "Tool Calls:\n", - " transfer_to_math_agent (call_WEBMqUupSKqT5slb6ilX2VEO)\n", - " Call ID: call_WEBMqUupSKqT5slb6ilX2VEO\n", - " Args:\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: transfer_to_math_agent\n", - "\n", - "Successfully transferred to math_agent\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: math_agent\n", - "Tool Calls:\n", - " divide (call_fsn6la6HTqfNl3sLp9v1WVoz)\n", - " Call ID: call_fsn6la6HTqfNl3sLp9v1WVoz\n", - " Args:\n", - " a: 2.16\n", - " b: 28.18\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: divide\n", - "\n", - "0.0766501064584812\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: math_agent\n", - "\n", - "US GDP in 2024: $28.18 trillion\n", - "New York State GDP in 2024: $2.16 trillion\n", - "Percentage of US GDP from New York State: 7.67%\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Name: supervisor\n", - "\n", - "Here are your results:\n", - "\n", - "- 2024 US GDP (projected): $28.18 trillion USD\n", - "- 2024 New York State GDP (estimated): $2.16 trillion USD\n", - "- New York State's share of US GDP: approximately 7.7%\n", - "\n", - "If you need the calculation steps or sources, let me know!\n" - ] - } - ], - "source": [ - "for message in final_message_history:\n", - " message.pretty_print()" - ] - }, - { - "cell_type": "markdown", - "id": "239700cd-d2d3-44be-8cf5-021d15615de2", - "metadata": {}, - "source": [ - "!!! Important\n", - "\n", - " You can see that the supervisor system appends **all** of the individual agent messages (i.e., their internal tool-calling loop) to the full message history. This means that on every supervisor turn, supervisor agent sees this full history. If you want more control over:\n", - "\n", - " * **how inputs are passed to agents**: you can use LangGraph [`Send()`][langgraph.types.Send] primitive to directly send data to the worker agents during the handoff. See the [task delegation](#create-delegation-tasks) example below\n", - " * **how agent outputs are added**: you can control how much of the agent's internal message history is added to the overall supervisor message history by wrapping the agent in a separate node function:\n", - "\n", - " ```python\n", - " def call_research_agent(state):\n", - " # return agent's final response,\n", - " # excluding inner monologue\n", - " response = research_agent.invoke(state)\n", - " # highlight-next-line\n", - " return {\"messages\": response[\"messages\"][-1]}\n", - " ```" - ] - }, - { - "cell_type": "markdown", - "id": "1c2a0aff-935a-46b8-bdcf-7125b80f1f41", - "metadata": {}, - "source": [ - "## 4. Create delegation tasks" - ] - }, - { - "cell_type": "markdown", - "id": "1a13d919-49bf-4cc3-838f-714cd3fb8b26", - "metadata": {}, - "source": [ - "So far the individual agents relied on **interpreting full message history** to determine their tasks. An alternative approach is to ask the supervisor to **formulate a task explicitly**. We can do so by adding a `task_description` parameter to the `handoff_tool` function." - ] - }, - { - "cell_type": "code", - "execution_count": 18, - "id": "dc1e4d24-8738-4548-a5e7-8556e476b805", - "metadata": {}, - "outputs": [], - "source": [ - "from langgraph.types import Send\n", - "\n", - "\n", - "def create_task_description_handoff_tool(\n", - " *, agent_name: str, description: str | None = None\n", - "):\n", - " name = f\"transfer_to_{agent_name}\"\n", - " description = description or f\"Ask {agent_name} for help.\"\n", - "\n", - " @tool(name, description=description)\n", - " def handoff_tool(\n", - " # this is populated by the supervisor LLM\n", - " task_description: Annotated[\n", - " str,\n", - " \"Description of what the next agent should do, including all of the relevant context.\",\n", - " ],\n", - " # these parameters are ignored by the LLM\n", - " state: Annotated[MessagesState, InjectedState],\n", - " ) -> Command:\n", - " task_description_message = {\"role\": \"user\", \"content\": task_description}\n", - " agent_input = {**state, \"messages\": [task_description_message]}\n", - " return Command(\n", - " # highlight-next-line\n", - " goto=[Send(agent_name, agent_input)],\n", - " graph=Command.PARENT,\n", - " )\n", - "\n", - " return handoff_tool\n", - "\n", - "\n", - "assign_to_research_agent_with_description = create_task_description_handoff_tool(\n", - " agent_name=\"research_agent\",\n", - " description=\"Assign task to a researcher agent.\",\n", - ")\n", - "\n", - "assign_to_math_agent_with_description = create_task_description_handoff_tool(\n", - " agent_name=\"math_agent\",\n", - " description=\"Assign task to a math agent.\",\n", - ")\n", - "\n", - "supervisor_agent_with_description = create_react_agent(\n", - " model=\"openai:gpt-4.1\",\n", - " tools=[\n", - " assign_to_research_agent_with_description,\n", - " assign_to_math_agent_with_description,\n", - " ],\n", - " prompt=(\n", - " \"You are a supervisor managing two agents:\\n\"\n", - " \"- a research agent. Assign research-related tasks to this assistant\\n\"\n", - " \"- a math agent. Assign math-related tasks to this assistant\\n\"\n", - " \"Assign work to one agent at a time, do not call agents in parallel.\\n\"\n", - " \"Do not do any work yourself.\"\n", - " ),\n", - " name=\"supervisor\",\n", - ")\n", - "\n", - "supervisor_with_description = (\n", - " StateGraph(MessagesState)\n", - " .add_node(\n", - " supervisor_agent_with_description, destinations=(\"research_agent\", \"math_agent\")\n", - " )\n", - " .add_node(research_agent)\n", - " .add_node(math_agent)\n", - " .add_edge(START, \"supervisor\")\n", - " .add_edge(\"research_agent\", \"supervisor\")\n", - " .add_edge(\"math_agent\", \"supervisor\")\n", - " .compile()\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "622bc89a-0787-41ea-bf18-c591be85509b", - "metadata": {}, - "source": [ - "!!! note\n", - "\n", - " We're using [`Send()`][langgraph.types.Send] primitive in the `handoff_tool`. This means that instead of receiving the full `supervisor` graph state as input, each worker agent only sees the contents of the `Send` payload. In this example, we're sending the task description as a single \"human\" message." - ] - }, - { - "cell_type": "markdown", - "id": "4d1dba40-c948-4117-9df6-5bdb1ab162c7", - "metadata": {}, - "source": [ - "Let's now running it with the same input query:" - ] - }, - { - "cell_type": "code", - "execution_count": 19, - "id": "882f9e00-5fdf-4278-ab26-a59fc949fa91", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Update from subgraph supervisor:\n", - "\n", - "\n", - "\tUpdate from node agent:\n", - "\n", - "\n", - "\t==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "\tName: supervisor\n", - "\tTool Calls:\n", - "\t transfer_to_research_agent (call_tk8q8py8qK6MQz6Kj6mijKua)\n", - "\t Call ID: call_tk8q8py8qK6MQz6Kj6mijKua\n", - "\t Args:\n", - "\t task_description: Find the 2024 GDP (Gross Domestic Product) for both the United States and New York state, using the most up-to-date and reputable sources available. Provide both GDP values and cite the data sources.\n", - "\n", - "\n", - "Update from subgraph research_agent:\n", - "\n", - "\n", - "\tUpdate from node agent:\n", - "\n", - "\n", - "\t==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "\tName: research_agent\n", - "\tTool Calls:\n", - "\t tavily_search (call_KqvhSvOIhAvXNsT6BOwbPlRB)\n", - "\t Call ID: call_KqvhSvOIhAvXNsT6BOwbPlRB\n", - "\t Args:\n", - "\t query: 2024 United States GDP value from a reputable source\n", - "\t search_depth: advanced\n", - "\t tavily_search (call_kbbAWBc9KwCWKHmM5v04H88t)\n", - "\t Call ID: call_kbbAWBc9KwCWKHmM5v04H88t\n", - "\t Args:\n", - "\t query: 2024 New York state GDP value from a reputable source\n", - "\t search_depth: advanced\n", - "\n", - "\n", - "Update from subgraph research_agent:\n", - "\n", - "\n", - "\tUpdate from node tools:\n", - "\n", - "\n", - "\t=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "\tName: tavily_search\n", - "\t\n", - "\t{\"query\": \"2024 United States GDP value from a reputable source\", \"follow_up_questions\": null, \"answer\": null, \"images\": [], \"results\": [{\"url\": \"https://www.focus-economics.com/countries/united-states/\", \"title\": \"United States Economy Overview - Focus Economics\", \"content\": \"The United States' Macroeconomic Analysis:\\n------------------------------------------\\n\\n**Nominal GDP of USD 29,185 billion in 2024.**\\n\\n**Nominal GDP of USD 29,179 billion in 2024.**\\n\\n**GDP per capita of USD 86,635 compared to the global average of USD 10,589.**\\n\\n**GDP per capita of USD 86,652 compared to the global average of USD 10,589.**\\n\\n**Average real GDP growth of 2.5% over the last decade.**\\n\\n**Average real GDP growth of 2.5% over the last decade.**\\n\\nShare of the region's population [...] | | 2020 | 2021 | 2022 | 2023 | 2024 |\\n| --- | --- | --- | --- | --- | --- |\\n| [Population (million)](https://www.focus-economics.com/country-indicator/united-states/population/) | 331 | 332 | 334 | 337 | 340 |\\n| [GDP (USD bn)](https://www.focus-economics.com/country-indicator/united-states/gdp/) | 21,354 | 23,681 | 26,007 | 27,721 | 29,185 |\", \"score\": 0.73981786, \"raw_content\": null}, {\"url\": \"https://tradingeconomics.com/united-states/gdp\", \"title\": \"United States GDP - Trading Economics\", \"content\": \"| Related | Last | Previous | Unit | Reference |\\n| --- | --- | --- | --- | --- |\\n| [Full Year GDP Growth](/united-states/full-year-gdp-growth) | 2.80 | 2.90 | percent | Dec 2024 |\\n| [GDP](/united-states/gdp) | 27720.71 | 26006.89 | USD Billion | Dec 2023 |\\n| [GDP Annual Growth Rate](/united-states/gdp-growth-annual) | 2.10 | 2.50 | percent | Mar 2025 |\\n| [GDP Constant Prices](/united-states/gdp-constant-prices) | 23528.00 | 23542.30 | USD Billion | Mar 2025 |\", \"score\": 0.65359193, \"raw_content\": null}, {\"url\": \"https://fred.stlouisfed.org/series/GDP\", \"title\": \"Gross Domestic Product (GDP) | FRED | St. Louis Fed\", \"content\": \"Q1 2025: 29,976.638 |\\nBillions of Dollars, Seasonally Adjusted Annual Rate |\\nQuarterly\\n\\nUpdated:\\nMay 29, 2025\\n7:56 AM CDT\\n\\n[Next Release Date:\\nJun 26, 2025](/releases/calendar?rid=53&y=2025)\\n\\nObservations\\n\\n| | | |\\n| --- | --- | --- |\\n| Q1 2025: | 29,976.638 | |\\n| Q4 2024: | 29,723.864 | |\\n| Q3 2024: | 29,374.914 | |\\n| Q2 2024: | 29,016.714 | |\\n| Q1 2024: | 28,624.069 | |\\n| [View All](/data/GDP.txt) | |\\n\\nUnits:\\n\\nFrequency:\", \"score\": 0.6152965, \"raw_content\": null}], \"response_time\": 2.53}\n", - "\n", - "\n", - "Update from subgraph research_agent:\n", - "\n", - "\n", - "\tUpdate from node tools:\n", - "\n", - "\n", - "\t=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "\tName: tavily_search\n", - "\t\n", - "\t{\"query\": \"2024 New York state GDP value from a reputable source\", \"follow_up_questions\": null, \"answer\": null, \"images\": [], \"results\": [{\"url\": \"https://www.ibisworld.com/united-states/economic-profiles/new-york/\", \"title\": \"New York Economic Trends, Stats & Rankings | IBISWorld\", \"content\": \"#### What is New York's Gross Domestic Product (GDP)?\\n\\nIn 2024, New York's GDP reached $1.8tn, representing an increase of 1.3% from 2023. New York's GDP has grown at an annualized rate of 2.9% over the five years to 2024. Moreover, New York's trailing five-year GDP growth ranks it 37th out of all 50 US states. [...] In 2024, the state of New York has a population of 19,482,372, having grown an annualized 0.0% over the five years to 2024, which ranks it 44th out of all 50 US states by growth rate. New York's gross state product (GSP) in 2024 reached $1.8tn, with growth of 1.3% over the 5 years to 2024. Businesses in New York employed a total of 11,671,845 people in 2024, with average annual employment growth over the past five years of 0.0%. The top three sectors by total employment are Finance and [...] The Finance and Insurance, Real Estate and Rental and Leasing and Information sectors contributed the most to New York's GDP in 2024, representing a combined 46.3% of state GDP.\\n\\nGDP trends by sector are an important indicator of which sectors are contributing the most value-add to the state's economy, in addition to how the state economy is evolving over time.\\n\\nSector\\n\\nGDP\\n\\nGrowth 2024 (%)\\n\\nAnnualized Growth 2019-24\", \"score\": 0.8051581, \"raw_content\": null}, {\"url\": \"https://comptroller.nyc.gov/reports/annual-state-of-the-citys-economy-and-finances-2024/\", \"title\": \"Annual State of the City's Economy and Finances 2024\", \"content\": \"(constant 2019 dollars)**22,550 15,322 13,596 19,451 21,211 21,096\\n**_% change_**(32.1%)(11.3%)43.1%9.0%(0.5%)\\n\\nSource: NY State Department of Taxation and Finance and Office of the New York City Comptroller [...] [[1]](https://comptroller.nyc.gov/reports/annual-state-of-the-citys-economy-and-finances-2024/#_ftnref1) At that time, the Comptroller’s Office expected five-year cumulative real GDP growth, 2020 to 2024, of 8.7 percent while the mayor expected 9.5 percent.\\n\\n[[2]](https://comptroller.nyc.gov/reports/annual-state-of-the-citys-economy-and-finances-2024/#_ftnref2) U.S. Bureau of Labor Statistics Local Area Unemployment Statistics (LAUS). [...] **($ in millions)****FY 2025****FY 2026****FY 2027****FY 2028****Total****Change from June 2024 (FY2025-FY2028)****Source as Share of Total**\\n**General Obligation Bonds**$6,900$6,230$6,920$7,110$27,160$1,960 42.8%\\n**TFA FTS Bonds**6,900 6,230 6,920 7,110 27,160 2,460 42.8%\\n**NYC Water Finance Authority**1,726 2,474 2,452 2,505 9,157 435 14.4%\\n**Total****$15,526****$14,934****$16,292****$16,725****$63,477****$4,855****100.0%**\", \"score\": 0.74541736, \"raw_content\": null}], \"response_time\": 3.45}\n", - "\n", - "\n", - "Update from subgraph research_agent:\n", - "\n", - "\n", - "\tUpdate from node agent:\n", - "\n", - "\n", - "\t==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "\tName: research_agent\n", - "\t\n", - "\t- United States 2024 GDP: $29.2 trillion (USD) \n", - "\t Source: [Focus Economics](https://www.focus-economics.com/countries/united-states/)\n", - "\t\n", - "\t- New York State 2024 GDP: $1.8 trillion (USD) \n", - "\t Source: [IBISWorld](https://www.ibisworld.com/united-states/economic-profiles/new-york/)\n", - "\n", - "\n", - "Update from subgraph supervisor:\n", - "\n", - "\n", - "\tUpdate from node agent:\n", - "\n", - "\n", - "\t==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "\tName: supervisor\n", - "\tTool Calls:\n", - "\t transfer_to_math_agent (call_EqT01HDOQDd5Timk89cccdXP)\n", - "\t Call ID: call_EqT01HDOQDd5Timk89cccdXP\n", - "\t Args:\n", - "\t task_description: Calculate what percentage of the US GDP ($29.2 trillion) was New York State's GDP ($1.8 trillion) in 2024. Show your working and round the final answer to two decimal points.\n", - "\n", - "\n", - "Update from subgraph math_agent:\n", - "\n", - "\n", - "\tUpdate from node agent:\n", - "\n", - "\n", - "\t==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "\tName: math_agent\n", - "\tTool Calls:\n", - "\t divide (call_OpdUlYEFqk4RZ2rF46M9eTrS)\n", - "\t Call ID: call_OpdUlYEFqk4RZ2rF46M9eTrS\n", - "\t Args:\n", - "\t a: 1.8\n", - "\t b: 29.2\n", - "\n", - "\n", - "Update from subgraph math_agent:\n", - "\n", - "\n", - "\tUpdate from node tools:\n", - "\n", - "\n", - "\t=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "\tName: divide\n", - "\t\n", - "\t0.06164383561643836\n", - "\n", - "\n", - "Update from subgraph math_agent:\n", - "\n", - "\n", - "\tUpdate from node agent:\n", - "\n", - "\n", - "\t==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "\tName: math_agent\n", - "\tTool Calls:\n", - "\t multiply (call_vdKJTLtIzmG6JbPcGQn93MR6)\n", - "\t Call ID: call_vdKJTLtIzmG6JbPcGQn93MR6\n", - "\t Args:\n", - "\t a: 0.06164383561643836\n", - "\t b: 100\n", - "\n", - "\n", - "Update from subgraph math_agent:\n", - "\n", - "\n", - "\tUpdate from node tools:\n", - "\n", - "\n", - "\t=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "\tName: multiply\n", - "\t\n", - "\t6.164383561643836\n", - "\n", - "\n", - "Update from subgraph math_agent:\n", - "\n", - "\n", - "\tUpdate from node agent:\n", - "\n", - "\n", - "\t==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "\tName: math_agent\n", - "\t\n", - "\tNew York State's GDP was approximately 6.16% of the US GDP in 2024.\n", - "\n", - "\n", - "Update from subgraph supervisor:\n", - "\n", - "\n", - "\tUpdate from node agent:\n", - "\n", - "\n", - "\t==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "\tName: supervisor\n", - "\t\n", - "\tNew York State’s GDP was approximately 6.16% of the US GDP in 2024.\n", - "\t\n", - "\t**Working:**\n", - "\t- Percentage = (NY GDP / US GDP) × 100\n", - "\t- = ($1.8 trillion / $29.2 trillion) × 100\n", - "\t- = 0.06164 × 100\n", - "\t- = 6.16% (rounded to two decimal points)\n", - "\n", - "\n" - ] - } - ], - "source": [ - "for chunk in supervisor_with_description.stream(\n", - " {\n", - " \"messages\": [\n", - " {\n", - " \"role\": \"user\",\n", - " \"content\": \"find US and New York state GDP in 2024. what % of US GDP was New York state?\",\n", - " }\n", - " ]\n", - " },\n", - " subgraphs=True,\n", - "):\n", - " pretty_print_messages(chunk, last_message=True)" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "6d3916d1-7de9-47c1-ae6f-80865b804e77", - "metadata": {}, - "outputs": [], - "source": [] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3 (ipykernel)", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.12.10" - } - }, - "nbformat": 4, - "nbformat_minor": 5 -} diff --git a/docs/docs/tutorials/multi_agent/agent_supervisor.md b/docs/docs/tutorials/multi_agent/agent_supervisor.md new file mode 100644 index 000000000..ba8f84ff6 --- /dev/null +++ b/docs/docs/tutorials/multi_agent/agent_supervisor.md @@ -0,0 +1,812 @@ +# Multi-agent supervisor + +[**Supervisor**](../../concepts/multi_agent.md#supervisor) is a multi-agent architecture where **specialized** agents are coordinated by a central **supervisor agent**. The supervisor agent controls all communication flow and task delegation, making decisions about which agent to invoke based on the current context and task requirements. + +In this tutorial, you will build a supervisor system with two agents — a research and a math expert. By the end of the tutorial you will: + +1. Build specialized research and math agents +2. Build a supervisor for orchestrating them with the prebuilt [`langgraph-supervisor`](https://langchain-ai.github.io/langgraph/agents/multi-agent/#supervisor) +3. Build a supervisor from scratch +4. Implement advanced task delegation + +![diagram](assets/diagram.png) + +## Setup + +First, let's install required packages and set our API keys + +```python +%%capture --no-stderr +%pip install -U langgraph langgraph-supervisor langchain-tavily "langchain[openai]" +``` + +```python +import getpass +import os + + +def _set_if_undefined(var: str): + if not os.environ.get(var): + os.environ[var] = getpass.getpass(f"Please provide your {var}") + + +_set_if_undefined("OPENAI_API_KEY") +_set_if_undefined("TAVILY_API_KEY") +``` + +!!! tip + Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. [LangSmith](https://docs.smith.langchain.com) lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph. + +## 1. Create worker agents + +First, let's create our specialized worker agents — research agent and math agent: + +* Research agent will have access to a web search tool using [Tavily API](https://tavily.com/) +* Math agent will have access to simple math tools (`add`, `multiply`, `divide`) + +### Research agent + +For web search, we will use `TavilySearch` tool from `langchain-tavily`: + +```python +from langchain_tavily import TavilySearch + +web_search = TavilySearch(max_results=3) +web_search_results = web_search.invoke("who is the mayor of NYC?") + +print(web_search_results["results"][0]["content"]) +``` + +**Output:** +``` +Find events, attractions, deals, and more at nyctourism.com Skip Main Navigation Menu The Official Website of the City of New York Text Size Powered by Translate SearchSearch Primary Navigation The official website of NYC Home NYC Resources NYC311 Office of the Mayor Events Connect Jobs Search Office of the Mayor | Mayor's Bio | City of New York Secondary Navigation MayorBiographyNewsOfficials Eric L. Adams 110th Mayor of New York City Mayor Eric Adams has served the people of New York City as an NYPD officer, State Senator, Brooklyn Borough President, and now as the 110th Mayor of the City of New York. Mayor Eric Adams has served the people of New York City as an NYPD officer, State Senator, Brooklyn Borough President, and now as the 110th Mayor of the City of New York. He gave voice to a diverse coalition of working families in all five boroughs and is leading the fight to bring back New York City's economy, reduce inequality, improve public safety, and build a stronger, healthier city that delivers for all New Yorkers. As the representative of one of the nation's largest counties, Eric fought tirelessly to grow the local economy, invest in schools, reduce inequality, improve public safety, and advocate for smart policies and better government that delivers for all New Yorkers. +``` + +To create individual worker agents, we will use LangGraph's prebuilt [agent](../../agents/agents.md). + +```python +from langgraph.prebuilt import create_react_agent + +research_agent = create_react_agent( + model="openai:gpt-4.1", + tools=[web_search], + prompt=( + "You are a research agent.\n\n" + "INSTRUCTIONS:\n" + "- Assist ONLY with research-related tasks, DO NOT do any math\n" + "- After you're done with your tasks, respond to the supervisor directly\n" + "- Respond ONLY with the results of your work, do NOT include ANY other text." + ), + name="research_agent", +) +``` + +Let's [run the agent](../../agents/run_agents.md) to verify that it behaves as expected. + +!!! note "We'll use `pretty_print_messages` helper to render the streamed agent outputs nicely" + + ```python + from langchain_core.messages import convert_to_messages + + + def pretty_print_message(message, indent=False): + pretty_message = message.pretty_repr(html=True) + if not indent: + print(pretty_message) + return + + indented = "\n".join("\t" + c for c in pretty_message.split("\n")) + print(indented) + + + def pretty_print_messages(update, last_message=False): + is_subgraph = False + if isinstance(update, tuple): + ns, update = update + # skip parent graph updates in the printouts + if len(ns) == 0: + return + + graph_id = ns[-1].split(":")[0] + print(f"Update from subgraph {graph_id}:") + print("\n") + is_subgraph = True + + for node_name, node_update in update.items(): + update_label = f"Update from node {node_name}:" + if is_subgraph: + update_label = "\t" + update_label + + print(update_label) + print("\n") + + messages = convert_to_messages(node_update["messages"]) + if last_message: + messages = messages[-1:] + + for m in messages: + pretty_print_message(m, indent=is_subgraph) + print("\n") + ``` + +```python +from langchain_core.messages import convert_to_messages + + +def pretty_print_message(message, indent=False): + pretty_message = message.pretty_repr(html=True) + if not indent: + print(pretty_message) + return + + indented = "\n".join("\t" + c for c in pretty_message.split("\n")) + print(indented) + + +def pretty_print_messages(update, last_message=False): + is_subgraph = False + if isinstance(update, tuple): + ns, update = update + # skip parent graph updates in the printouts + if len(ns) == 0: + return + + graph_id = ns[-1].split(":")[0] + print(f"Update from subgraph {graph_id}:") + print("\n") + is_subgraph = True + + for node_name, node_update in update.items(): + update_label = f"Update from node {node_name}:" + if is_subgraph: + update_label = "\t" + update_label + + print(update_label) + print("\n") + + messages = convert_to_messages(node_update["messages"]) + if last_message: + messages = messages[-1:] + + for m in messages: + pretty_print_message(m, indent=is_subgraph) + print("\n") +``` + +```python +for chunk in research_agent.stream( + {"messages": [{"role": "user", "content": "who is the mayor of NYC?"}]} +): + pretty_print_messages(chunk) +``` + +**Output:** +``` +Update from node agent: + + +================================== Ai Message ================================== +Name: research_agent +Tool Calls: + tavily_search (call_U748rQhQXT36sjhbkYLSXQtJ) + Call ID: call_U748rQhQXT36sjhbkYLSXQtJ + Args: + query: current mayor of New York City + search_depth: basic + + +Update from node tools: + + +================================= Tool Message ================================== +Name: tavily_search + +{"query": "current mayor of New York City", "follow_up_questions": null, "answer": null, "images": [], "results": [{"title": "List of mayors of New York City - Wikipedia", "url": "https://en.wikipedia.org/wiki/List_of_mayors_of_New_York_City", "content": "The mayor of New York City is the chief executive of the Government of New York City, as stipulated by New York City's charter.The current officeholder, the 110th in the sequence of regular mayors, is Eric Adams, a member of the Democratic Party.. During the Dutch colonial period from 1624 to 1664, New Amsterdam was governed by the Director of Netherland.", "score": 0.9039154, "raw_content": null}, {"title": "Office of the Mayor | Mayor's Bio | City of New York - NYC.gov", "url": "https://www.nyc.gov/office-of-the-mayor/bio.page", "content": "Mayor Eric Adams has served the people of New York City as an NYPD officer, State Senator, Brooklyn Borough President, and now as the 110th Mayor of the City of New York. He gave voice to a diverse coalition of working families in all five boroughs and is leading the fight to bring back New York City's economy, reduce inequality, improve", "score": 0.8405867, "raw_content": null}, {"title": "Eric Adams - Wikipedia", "url": "https://en.wikipedia.org/wiki/Eric_Adams", "content": "Eric Leroy Adams (born September 1, 1960) is an American politician and former police officer who has served as the 110th mayor of New York City since 2022. Adams was an officer in the New York City Transit Police and then the New York City Police Department (``` +``` + +### Math agent + +For math agent tools we will use [vanilla Python functions](../../agents/tools.md#define-simple-tools): + +```python +def add(a: float, b: float): + """Add two numbers.""" + return a + b + + +def multiply(a: float, b: float): + """Multiply two numbers.""" + return a * b + + +def divide(a: float, b: float): + """Divide two numbers.""" + return a / b + + +math_agent = create_react_agent( + model="openai:gpt-4.1", + tools=[add, multiply, divide], + prompt=( + "You are a math agent.\n\n" + "INSTRUCTIONS:\n" + "- Assist ONLY with math-related tasks\n" + "- After you're done with your tasks, respond to the supervisor directly\n" + "- Respond ONLY with the results of your work, do NOT include ANY other text." + ), + name="math_agent", +) +``` + +Let's run the math agent: + +```python +for chunk in math_agent.stream( + {"messages": [{"role": "user", "content": "what's (3 + 5) x 7"}]} +): + pretty_print_messages(chunk) +``` + +**Output:** +``` +Update from node agent: + + +================================== Ai Message ================================== +Name: math_agent +Tool Calls: + add (call_p6OVLDHB4LyCNCxPOZzWR15v) + Call ID: call_p6OVLDHB4LyCNCxPOZzWR15v + Args: + a: 3 + b: 5 + + +Update from node tools: + + +================================= Tool Message ================================== +Name: add + +8.0 + + +Update from node agent: + + +================================== Ai Message ================================== +Name: math_agent +Tool Calls: + multiply (call_EoaWHMLFZAX4AkajQCtZvbli) + Call ID: call_EoaWHMLFZAX4AkajQCtZvbli + Args: + a: 8 + b: 7 + + +Update from node tools: + + +================================= Tool Message ================================== +Name: multiply + +56.0 + + +Update from node agent: + + +================================== Ai Message ================================== +Name: math_agent + +56 + + +``` + +## 2. Create supervisor with `langgraph-supervisor` + +To implement out multi-agent system, we will use [`create_supervisor`][langgraph_supervisor.supervisor.create_supervisor] from the prebuilt `langgraph-supervisor` library: + +```python +from langgraph_supervisor import create_supervisor +from langchain.chat_models import init_chat_model + +supervisor = create_supervisor( + model=init_chat_model("openai:gpt-4.1"), + agents=[research_agent, math_agent], + prompt=( + "You are a supervisor managing two agents:\n" + "- a research agent. Assign research-related tasks to this agent\n" + "- a math agent. Assign math-related tasks to this agent\n" + "Assign work to one agent at a time, do not call agents in parallel.\n" + "Do not do any work yourself." + ), + add_handoff_back_messages=True, + output_mode="full_history", +).compile() +``` + +```python +from IPython.display import display, Image + +display(Image(supervisor.get_graph().draw_mermaid_png())) +``` + +![Graph](assets/output.png) + +**Note:** When you run this code, it will generate and display a visual representation of the supervisor graph showing the flow between the supervisor and worker agents. + +Let's now run it with a query that requires both agents: + +* research agent will look up the necessary GDP information +* math agent will perform division to find the percentage of NY state GDP, as requested + +```python +for chunk in supervisor.stream( + { + "messages": [ + { + "role": "user", + "content": "find US and New York state GDP in 2024. what % of US GDP was New York state?", + } + ] + }, +): + pretty_print_messages(chunk, last_message=True) + +final_message_history = chunk["supervisor"]["messages"] +``` + +**Output:** +``` +Update from node supervisor: + + +================================= Tool Message ================================== +Name: transfer_to_research_agent + +Successfully transferred to research_agent + + +Update from node research_agent: + + +================================= Tool Message ================================== +Name: transfer_back_to_supervisor + +Successfully transferred back to supervisor + + +Update from node supervisor: + + +================================= Tool Message ================================== +Name: transfer_to_math_agent + +Successfully transferred to math_agent + + +Update from node math_agent: + + +================================= Tool Message ================================== +Name: transfer_back_to_supervisor + +Successfully transferred back to supervisor + + +Update from node supervisor: + + +================================== Ai Message ================================== +Name: supervisor + +In 2024, the US GDP was $29.18 trillion and New York State's GDP was $2.297 trillion. New York State accounted for approximately 7.87% of the total US GDP in 2024. + + +``` + +## 3. Create supervisor from scratch + +Let's now implement this same multi-agent system from scratch. We will need to: + +1. [Set up how the supervisor communicates](#set-up-agent-communication) with individual agents +2. [Create the supervisor agent](#create-supervisor-agent) +3. Combine supervisor and worker agents into a [single multi-agent graph](#create-multi-agent-graph). + +### Set up agent communication + +We will need to define a way for the supervisor agent to communicate with the worker agents. A common way to implement this in multi-agent architectures is using **handoffs**, where one agent *hands off* control to another. Handoffs allow you to specify: + +- **destination**: target agent to transfer to +- **payload**: information to pass to that agent + +We will implement handoffs via **handoff tools** and give these tools to the supervisor agent: when the supervisor calls these tools, it will hand off control to a worker agent, passing the full message history to that agent. + +```python +from typing import Annotated +from langchain_core.tools import tool, InjectedToolCallId +from langgraph.prebuilt import InjectedState +from langgraph.graph import StateGraph, START, MessagesState +from langgraph.types import Command + + +def create_handoff_tool(*, agent_name: str, description: str | None = None): + name = f"transfer_to_{agent_name}" + description = description or f"Ask {agent_name} for help." + + @tool(name, description=description) + def handoff_tool( + state: Annotated[MessagesState, InjectedState], + tool_call_id: Annotated[str, InjectedToolCallId], + ) -> Command: + tool_message = { + "role": "tool", + "content": f"Successfully transferred to {agent_name}", + "name": name, + "tool_call_id": tool_call_id, + } + # highlight-next-line + return Command( + # highlight-next-line + goto=agent_name, # (1)! + # highlight-next-line + update={**state, "messages": state["messages"] + [tool_message]}, # (2)! + # highlight-next-line + graph=Command.PARENT, # (3)! + ) + + return handoff_tool + + +# Handoffs +assign_to_research_agent = create_handoff_tool( + agent_name="research_agent", + description="Assign task to a researcher agent.", +) + +assign_to_math_agent = create_handoff_tool( + agent_name="math_agent", + description="Assign task to a math agent.", +) +``` + +1. Name of the agent or node to hand off to. +2. Take the agent's messages and add them to the parent's state as part of the handoff. The next agent will see the parent state. +3. Indicate to LangGraph that we need to navigate to agent node in a **parent** multi-agent graph. + +### Create supervisor agent + +Then, let's create the supervisor agent with the handoff tools we just defined. We will use the prebuilt [`create_react_agent`][langgraph.prebuilt.chat_agent_executor.create_react_agent]: + +```python +supervisor_agent = create_react_agent( + model="openai:gpt-4.1", + tools=[assign_to_research_agent, assign_to_math_agent], + prompt=( + "You are a supervisor managing two agents:\n" + "- a research agent. Assign research-related tasks to this agent\n" + "- a math agent. Assign math-related tasks to this agent\n" + "Assign work to one agent at a time, do not call agents in parallel.\n" + "Do not do any work yourself." + ), + name="supervisor", +) +``` + +### Create multi-agent graph + +Putting this all together, let's create a graph for our overall multi-agent system. We will add the supervisor and the individual agents as subgraph [nodes](../../concepts/low_level.md#nodes). + +```python +from langgraph.graph import END + +# Define the multi-agent supervisor graph +supervisor = ( + StateGraph(MessagesState) + # NOTE: `destinations` is only needed for visualization and doesn't affect runtime behavior + .add_node(supervisor_agent, destinations=("research_agent", "math_agent", END)) + .add_node(research_agent) + .add_node(math_agent) + .add_edge(START, "supervisor") + # always return back to the supervisor + .add_edge("research_agent", "supervisor") + .add_edge("math_agent", "supervisor") + .compile() +) +``` + +Notice that we've added explicit [edges](../../concepts/low_level.md#edges) from worker agents back to the supervisor — this means that they are guaranteed to return control back to the supervisor. If you want the agents to respond directly to the user (i.e., turn the system into a router, you can remove these edges). + +```python +from IPython.display import display, Image + +display(Image(supervisor.get_graph().draw_mermaid_png())) +``` + +![Graph](assets/multi-output.png) + +**Note:** When you run this code, it will generate and display a visual representation of the multi-agent supervisor graph showing the flow between the supervisor and worker agents. + +With the multi-agent graph created, let's now run it! + +```python +for chunk in supervisor.stream( + { + "messages": [ + { + "role": "user", + "content": "find US and New York state GDP in 2024. what % of US GDP was New York state?", + } + ] + }, +): + pretty_print_messages(chunk, last_message=True) + +final_message_history = chunk["supervisor"]["messages"] +``` + +**Output:** +``` +Update from node supervisor: + + +================================= Tool Message ================================== +Name: transfer_to_research_agent + +Successfully transferred to research_agent + + +Update from node research_agent: + + +================================== Ai Message ================================== +Name: research_agent + +- US GDP in 2024 is projected to be about $28.18 trillion USD (Statista; CBO projection). +- New York State's nominal GDP for 2024 is estimated at approximately $2.16 trillion USD (various economic reports). +- New York State's share of US GDP in 2024 is roughly 7.7%. + +Sources: +- https://www.statista.com/statistics/216985/forecast-of-us-gross-domestic-product/ +- https://nyassembly.gov/Reports/WAM/2025economic_revenue/2025_report.pdf?v=1740533306 + + +Update from node supervisor: + + +================================= Tool Message ================================== +Name: transfer_to_math_agent + +Successfully transferred to math_agent + + +Update from node math_agent: + + +================================== Ai Message ================================== +Name: math_agent + +US GDP in 2024: $28.18 trillion +New York State GDP in 2024: $2.16 trillion +Percentage of US GDP from New York State: 7.67% + + +Update from node supervisor: + + +================================== Ai Message ================================== +Name: supervisor + +Here are your results: + +- 2024 US GDP (projected): $28.18 trillion USD +- 2024 New York State GDP (estimated): $2.16 trillion USD +- New York State's share of US GDP: approximately 7.7% + +If you need the calculation steps or sources, let me know! + + +``` + +Let's examine the full resulting message history: + +```python +for message in final_message_history: + message.pretty_print() +``` + +**Output:** +``` +================================ Human Message ================================== + +find US and New York state GDP in 2024. what % of US GDP was New York state? +================================== Ai Message =================================== +Name: supervisor +Tool Calls: + transfer_to_research_agent (call_KlGgvF5ahlAbjX8d2kHFjsC3) + Call ID: call_KlGgvF5ahlAbjX8d2kHFjsC3 + Args: +================================= Tool Message ================================== +Name: transfer_to_research_agent + +Successfully transferred to research_agent +================================== Ai Message =================================== +Name: research_agent +Tool Calls: + tavily_search (call_ZOaTVUA6DKrOjWQldLhtrsO2) + Call ID: call_ZOaTVUA6DKrOjWQldLhtrsO2 + Args: + query: US GDP 2024 estimate or actual + search_depth: advanced + tavily_search (call_QsRAasxW9K03lTlqjuhNLFbZ) + Call ID: call_QsRAasxW9K03lTlqjuhNLFbZ + Args: + query: New York state GDP 2024 estimate or actual + search_depth: advanced +================================= Tool Message ================================== +Name: tavily_search + +{"query": "US GDP 2024 estimate or actual", "follow_up_questions": null, "answer": null, "images": [], "results": [{"url": "https://www.advisorperspectives.com/dshort/updates/2025/05/29/gdp-gross-domestic-product-q1-2025-second-estimate", "title": "Q1 GDP Second Estimate: Real GDP at -0.2%, Higher Than Expected", "content": "> Real gross domestic product (GDP) decreased at an annual rate of 0.2 percent in the first quarter of 2025 (January, February, and March), according to the second estimate released by the U.S. Bureau of Economic Analysis. In the fourth quarter of 2024, real GDP increased 2.4 percent. The decrease in real GDP in the first quarter primarily reflected an increase in imports, which are a subtraction in the calculation of GDP, and a decrease in government spending. These movements were partly [...] by [Harry Mamaysky](https://www.advisor``` +``` + +!!! important + You can see that the supervisor system appends **all** of the individual agent messages (i.e., their internal tool-calling loop) to the full message history. This means that on every supervisor turn, supervisor agent sees this full history. If you want more control over: + + * **how inputs are passed to agents**: you can use LangGraph [`Send()`][langgraph.types.Send] primitive to directly send data to the worker agents during the handoff. See the [task delegation](#4-create-delegation-tasks) example below + * **how agent outputs are added**: you can control how much of the agent's internal message history is added to the overall supervisor message history by wrapping the agent in a separate node function: + + ```python + def call_research_agent(state): + # return agent's final response, + # excluding inner monologue + response = research_agent.invoke(state) + # highlight-next-line + return {"messages": response["messages"][-1]} + ``` + +## 4. Create delegation tasks + +So far the individual agents relied on **interpreting full message history** to determine their tasks. An alternative approach is to ask the supervisor to **formulate a task explicitly**. We can do so by adding a `task_description` parameter to the `handoff_tool` function. + +```python +from langgraph.types import Send + + +def create_task_description_handoff_tool( + *, agent_name: str, description: str | None = None +): + name = f"transfer_to_{agent_name}" + description = description or f"Ask {agent_name} for help." + + @tool(name, description=description) + def handoff_tool( + # this is populated by the supervisor LLM + task_description: Annotated[ + str, + "Description of what the next agent should do, including all of the relevant context.", + ], + # these parameters are ignored by the LLM + state: Annotated[MessagesState, InjectedState], + ) -> Command: + task_description_message = {"role": "user", "content": task_description} + agent_input = {**state, "messages": [task_description_message]} + return Command( + # highlight-next-line + goto=[Send(agent_name, agent_input)], + graph=Command.PARENT, + ) + + return handoff_tool + + +assign_to_research_agent_with_description = create_task_description_handoff_tool( + agent_name="research_agent", + description="Assign task to a researcher agent.", +) + +assign_to_math_agent_with_description = create_task_description_handoff_tool( + agent_name="math_agent", + description="Assign task to a math agent.", +) + +supervisor_agent_with_description = create_react_agent( + model="openai:gpt-4.1", + tools=[ + assign_to_research_agent_with_description, + assign_to_math_agent_with_description, + ], + prompt=( + "You are a supervisor managing two agents:\n" + "- a research agent. Assign research-related tasks to this assistant\n" + "- a math agent. Assign math-related tasks to this assistant\n" + "Assign work to one agent at a time, do not call agents in parallel.\n" + "Do not do any work yourself." + ), + name="supervisor", +) + +supervisor_with_description = ( + StateGraph(MessagesState) + .add_node( + supervisor_agent_with_description, destinations=("research_agent", "math_agent") + ) + .add_node(research_agent) + .add_node(math_agent) + .add_edge(START, "supervisor") + .add_edge("research_agent", "supervisor") + .add_edge("math_agent", "supervisor") + .compile() +) +``` + +!!! note + We're using [`Send()`][langgraph.types.Send] primitive in the `handoff_tool`. This means that instead of receiving the full `supervisor` graph state as input, each worker agent only sees the contents of the `Send` payload. In this example, we're sending the task description as a single "human" message. + +Let's now running it with the same input query: + +```python +for chunk in supervisor_with_description.stream( + { + "messages": [ + { + "role": "user", + "content": "find US and New York state GDP in 2024. what % of US GDP was New York state?", + } + ] + }, + subgraphs=True, +): + pretty_print_messages(chunk, last_message=True) +``` + +**Output:** +``` +Update from subgraph supervisor: + + + Update from node agent: + + + ================================== Ai Message ================================== + Name: supervisor + Tool Calls: + transfer_to_research_agent (call_tk8q8py8qK6MQz6Kj6mijKua) + Call ID: call_tk8q8py8qK6MQz6Kj6mijKua + Args: + task_description: Find the 2024 GDP (Gross Domestic Product) for both the United States and New York state, using the most up-to-date and reputable sources available. Provide both GDP values and cite the data sources. + + +Update from subgraph research_agent: + + + Update from node agent: + + + ================================== Ai Message ================================== + Name: research_agent + Tool Calls: + tavily_search (call_KqvhSvOIhAvXNsT6BOwbPlRB) + Call ID: call_KqvhSvOIhAvXNsT6BOwbPlRB + Args: + query: 2024 United States GDP value from a reputable source + search_depth: advanced + tavily_search (call_kbbAWBc9KwCWKHmM5v04H88t) + Call ID: call_kbbAWBc9KwCWKHmM5v04H88t + Args: + query: 2024 New York state GDP value from a reputable source + search_depth: advanced + + +Update from subgraph research_agent: + + + Update from node tools: + + + ================================= Tool Message ================================== + Name: tavily_search + + {"query": "2024 United States GDP value from a reputable source", "follow_up_questions": null, "answer": null, "images": [], "results": [{"url": "https://www.focus-economics.com/countries/united-states/", "title": "United States Economy Overview - Focus Economics", "content": "The United States' Macroeconomic Analysis:\n------------------------------------------\n\n**Nominal GDP of USD 29,185 billion in 2024.**\n\n**Nominal GDP of USD 29,179 billion in 2024.**\n\n**GDP per capita of USD 86,635 compared to the global average of USD 10,589.**\n\n**GDP per capita of USD 86,652 compared to the global average of USD 10,589.**\n\n**Average real GDP growth of 2.5% over the last decade.**\n\n**Average real GDP growth of ``` +``` \ No newline at end of file diff --git a/docs/docs/tutorials/multi_agent/assets/diagram.png b/docs/docs/tutorials/multi_agent/assets/diagram.png new file mode 100644 index 0000000000000000000000000000000000000000..8aaad47591014bf58887f4697663a977a8141ae9 GIT binary patch literal 25505 zcmdSAbzhuK(>90%xPlD<0t62U?rwu5xCaQXK@)6n*O0*A!QI{6HDQq8?t=t(8{GEH z_1y11_wH}M!2UJeXLnasS66q}QC$=AL0JX|lN1vP2?<9|R#FWK2@MYk2^I4n-~-Z} z6@Mo1gJPl}BZ>4B95vJl)G(aol%z0rFi2mZk|#%HK!8uk5J%;Y$fG%KDM&~rd2*8P zKDsX)!adT+UDBUUwFJHgAmaqf(g#$1QbKwm)q*9d8tsn6MCa;7jr9)`vvq`YkCTSv z=06%cm=!3PG$Tn_U-h3uso!g?znG_byI&(6qZpp$FudCT>AkPt#|ypoD2LcbA0X=q-fqC&V_enw=-V+S(5U#YbJht{Wfi>ZGKC22 zXt)-^zdS8fGTn~{#WSN#e@v-UTO_Jqv4E-=qEg=xnEiT&)51ppQ(UmsTSVqR#Z6fB zY>|PZR=o_B$`c=A>Z>1c8LcsGzO=q_^zx422%}Rp6);@smP*!^#Dv*c!LmJnDLUBy zY;gV>CKZq*$2;H>7ZfDOHfQygC=`V=(7Q3b)XfT;oVZkMH~j|*BVKJ^<-*i7I6Qcr zD@)yg91}KlT1r$jk55c%lFfj}U8p&;>6mt3f?q8_2-R7i8ce~#XI{|=Y~ui`+}&rI zwCx}%b>{syMbsacUYn8+lF52fVfMMUrY5r`XA_w;Tx$%6D~Kx*52O99xes)|oy$mp z_GF0<5%KR*FcgGt(zRpZW1*SM6a^%GtNEWMLJZ+u0|q+gmeAN33rv_|acMwOcxb$= zfqh6S3a2_)cvpmhr5+FKT|g4;UY~q;d92Yt!0hEpzIGOcXahx-Q|HR2yd4;EAkckE zTrphP=QS`JbH^*Rc!tkO&Hn(>=UnAMRWT)_U5gwH{~aZ81PCQ1`5sg9Mf498Rmrz?5);DIG0$6kBwiPNPh#AgC+3v)JWFV8zL^QZk0{vN?jNPDU~JFeX7Y4veO+xc!F)(y zf{u*G)4`WfWpnmjr^kw(GPxh+JTYKkG?Rk7ykxF6LU*ZsQ$Gw(SG2k7+C{rp=-M8$ z@5*1-+*FH+6xI#yT!BDKcXKDcleogW(7Y(wXCMBG%g^CfIMX+SI{E4ROt?}<>}<7V z%hIOp^y@`e^|2tExAA-`@=UqcYRNjgyYEEb+po8H=3JV{!BHLucE)~ukBek%6vGmR z3-{k0LZ_?TptA+ZXJfk9Mtx z4`0XW_0N-|*~J>S8M}HXgHwGU+AV$GS4zpH8Gf5+enP#3_!QyVKLb5mNf+}= zv-5QQ97)t?Z$#x`$8_~+)8m9S%hKU_(p~akClQaZR7}2j+=wj-Un98YRMU9NS@|Ln z3Dj`~JqKAoo(4rBi)iVbq$Q%>UwD9h#ZH=6e|a@VWi;3f55|a=*nZfpry0wVFUZ?a zYQ|b?>we)@hM^ol!TyAVG^h-A(wU5^BtbXizJE&w5q@PXREB)%rLg||uH&eTnBM0j zjn(Ucw+C4FEgna!u}06_397nfl!YL#y3*Y`&QN8(iy6A0bt&7q%MYw9bF!^9>+7% z!}3=zpI5UVqaplHcc6fbj+4bhBd0@^(OqQsRSJkNiThS?7i}MJhUj_lDD6KmidfMf zxH}C)-?ISSh-@o$p+zYF66?IuMJoJ4Gcpu4Ew+GKb$}xlwAuY3*>4!MJcml|a>~k0 zT@+rM9V?bn7srxHm`eyvj!Cc>@CsPtHxk0WKg!W!LB9U8v?@b(Oo=}CVB~iiLg^2> zPpcPyW$1ZxW;8`*fM_rXFy0$6p6DA7nEFVHANwy2x6Rj<$ohQS8yOIUe+l4y!F@~FDMkRXD5DSIohu=Tbb_3v?<5T{N6MR=dj zlm@EC4=^uJ>F=z$te|T94ez{}ynMV)G``q!kF%;i$!xz2y+fx~-Dp zFn79I{Lv(EB%ZRI5a{P~J8ytXs@J>b$JW}sc{eFK%3Z00V{dra+_iGw3NZlDXSgiy zSxpLUg62A7Z+zyAVB}>@xM?U|{+uD6%yuV!m1Q zwA9t;N-uh9a6~~Z>+9mmgJyWD&rMK#=M#Lzb;WeFI2he6M`+`+HW9guj5YgVM6vyP zVp8k1bTM$e^0E^s!5qL35zHxjKPhiR_mJg|Rb{L)Dw42XOHfV3mctpsbjsURvQ488?!pBQhFC^q7YqZ@+730^h;!fShpr6 z5mXoY>f`y;=a7)x{Y?{9Y75oj+5ykZlX6(&gPOx2u$>XroS;Z#-}p4 zI4PJ|@B7u`=sjO5w{vYnWrEx#8;Z=8dD4c;1|*qx%%%|?21lY16zYfLaH`)sj7Ybq zPjWjRY(jNcVmL8j-egyf?m3^6iWq0?CX|L@c-&_)-sQauU@)+< zq+Q$I79n41@~BB1v0Wn6siuRg2WaoA3BQ|a{bZYeNRPU@lcIs270(nS??PVQ7d^*_ zN1MOIR(K9|Rt{ir*a0u;PYcFg17Fc&llphEg)H%f(Q@))3Yn&3kbpvTU_?bStt+Nv zh(zHz2ww0+91?+|YBJW_wnv1kP4N^XJ4*T_kSJa^5PZ3G%b z+4zj(r$ZPx)v4vNh4c~+=Gc%D@~tgG5<`V$_nOAEP?_Ly@h_8~0jwfSa z85|~PYhl^Br_nWw81BYiR&V6X0AhSc{Uz<8)0#Ir(M1e+wC1D|Odq>u1ZrXO0SsT( zxg1+t<+U3333lU9RR4^`4;yKIKrT30%!xg zYlj_gof@+HzM222?y5nsi=VyHtzOEzlEst^VA#bs}^O_T#OJtOW&f2mHm`xQl1%?h8t zk$Cq%UIyWd=YY~FQ-Jwr)~4&C9GeQiMl(-?kQrW`t0F5h_KGPa> z5lV_@jrj+V&T{&B0o>WXrwnB=23TAYe#hZ5W(zsNzr`Be9wko|$p*jxx8{GMRpoO2 z+mNL76BTrz>?_v}#TP@%fps|+EB>bPoC*Vf?erzqiE#0K{!HFmU=KJ{ ziIkfrV1J;N2a}i9s^Y(li3g;lrDRKSepB4|*G=p({z4jI=C2E^Tn&7lMRID5;rRzF z)juDKofjBC@ABp0cQU5NE~7>}43Pk2?*Da8FJAwzL;D~1bfPYGX%|6stg4a}na3g# zSBtlpd7K+?_-m5EiSBxJx;4Cl&*?RYAZKxSV?^b%JN}?VoB*!qFO3HK%*tx+-33V$$1ygqef`~%^is#SEVi($KL!DBv3J(i4c{QB!n_< z6lIF4TLCb|BU@aNs?Q)EIJ@cZOa9i+jCR!<0j=)J_M8c^a=vBE7gwa{n?!T!ookXi z{wgUEkR()*j-ZqcX40IJqNRO}39B22%A?OyByPq#t5Fw3rXz4=`?K`$A5OIy0mpyjeNu%`2S!D)go&e6UzU>C!8A2)3h;vkq-~g zP#P+LoER8zod8?Kf6m%6g!&)6V`}4hu=WzT!iaWL1VOH{o)~#Lf6!_(%b!*gJ{9SU z^6sif3(7rB$`JwpVy~J?zZMtiDZ!1tot@@6V5U32TRZ%&`uU%r>+E(@r7CLDK2Vgd z^-RAtK{pH|N*4icBgiAWsOQ-`$iqq=NLU_FB(je*R@GXnk*)Vl#-swcLkpWKF4NJA)41##+6?wK>7J7rX>;VOa>iejryrIWLIqGL zMBsX~JecI3j7-c&WF=`4(Y5O1)v;CI_muMp6N>`ZT@pXtEk425SDgxJectDF71$_y zg7QMtkE0$Sc0x8J#(OU}j~Z$n4vnssI(JcBcO#zaQaw@_g(F0-G)iwK zgj0MyP8z7hNW3*(&D~~xS++?TGS--ArG~3h7sy9kqei!?U_I&yUL`Reg8fczgWWvW zPJ`<&8^8|vSR7C7wZAYbZWPyLw!KdTu{u$foVxmJ@5E-(yvckY1BGRL>x|Tu`eh(S z=htj^avKHUPd0#Y_}j4R;S+O*4AO?-F>fC zXZvc#Non)6R*=%@%y0)sKrHVIG!r&pr_v^iZ-hXMUTn=z`IR3Ld0%W`1yp7%0pSCx zr6;UO|Fx^-MfaV#L9gygKHrnA@d&sJUe%cftKaGyV$>1F2uS6Vfc-IIHC3sv61zj- z&rYD1+sa|q$uG~j8;ePF3*UXDhNZH*gtsFz*M!jKtamR1;G-g@r{?>}Gk(5ng2#)< ztMB~jWilMs-b8s{GOhlK25&6884-WmE{Iyf=6FC8?P7gG_a*Z0~;NUgsT5!zL!zsdQ+k22<@ z3l|~uUklqVN4^#jFFrl*TtRh1v95UDk-31%BJA@9Df0oDsuCH#aYoy&shF4L~5Xv5B@=TCR=Rk7N{M+?!DuJ`^%`dC9PH||MR z$Wf?QNKNnjHV~EUE@Brg>#G6OqB`eaR%=^-ax8&_{jiX>cPb+qWNy4pGEM!!S7_{_ z>%G<%sN13~n-)=uw`IKJYFogOG60nrHG1^$zku8cmmx3L_^iu>`8RiQAr;^r)K7k= zM@g&`(-`fcz>jH(;=D-e#gu(NQ{pLCI6O)kDk}t`dl1{KhHI&xztg;P4a3 zA9RSkXvFifiQ?_NRo#oz#MY2bU^HRgYNg zcMBg*5EPeaxSdVoJ?;CmUaOYtHhdye;zMX?IduXyznxc)R#PpzG5CG)jhXIWfBrR$ zv2#OxvSQPbcUtKS zG1$quzw%13%5E|t^<_vLO7)JN$Mk+jXYuZx2oaJ>7Z%BHDK@`8y?dMjIZ@L+=WgOj9Z38s~*Q!3r8FM&Zl#cgV=pe%|98v z!H9fp=%mDj`=$+tO3DRj+oy`vIQGaq9;)Pmz?T}v1d13(Iw(~)Dc^spNK3m^B9Vq= z?RVWpJ?;9DqaoTJ^%AedIq0~iZiAIk*be-zMY^ku1g<=(R<$BBkDbr1>;#DX6;>V_ zSF^>U?$5e?x)k~r%VgttK-H@m6zFyxH*o_$TaQkL)X9AB2JeCboK~-+P%T?!jxJWs zuP{{F*99K&yXcurTMF#H)uDVu1V)tvQv=y zR)eaTPd0>SNAK0pR-x8@E{|MIPZ$udtHw|8=f|apvZrnX1tiaoj>RTm0e`S1haH98 z*`AJdwh^zQ7Dk9VZaho|P@S@zeDV+OxM{x--z;xKx*kzCZK!(Q_{U<<>-!221Ju=nFw)Uzmr_sI|iTg5LbB!Qd$M*a+;W2{Aow^ff4Mz$39K5QH= z2ks|czJ3%Z#kJCu0)%tKNhtQiynR&nZ5ftN`khqQZ@hQMF=YH)wss2-_{J@ya&pC4 z+TbfskMUwR!$yyY#Pqppw>^bHnaDP}<;w}ka%0OjZ>1-^;zRQq<}r>9x~tIb5j!ig z`}!_Bs^PhxEfY3ky>yt^U>up=#nq> zn{G4vuECEZARNH7iM^C$ijlP~mAf+PY-YB6k*kyrk;~2Fh4S;^Bl|AoO6A+Jyc0s`M8G!6 zeq&q2Lx<7h*u~X@orvPYcDK>vgn`hC9q#8DCE%9KN4*2=mg5qmVR$8DsXR%C;U-%& z`J)_Q=sx~pN{DXwr6#3(FC);=Ym(udDMkk1+qK+vq2kj1Dmo4Tj2Nz9{Gi0Q?Zh+% z0678~R3Q9I@IPgGI2kXJF5O5(;nZM)Lw;3Zc?#G`1D>@Bl+L0s zJo8L3P3I}|4T6!$<8K7=;Bf+QRbeDPb7B9!_e2-b12HA!{|bl~C@c)Zp&KcHN6KZd z{xW*$Kkq<{xz+N~8~^QdnuEo5MSlDqx4rt;XZ_bL&Kp_L|3F<3k&3|W&vQ?CpMnAS zxN;t+?8(qdzm~sE;rYriMKx4TD!`dnpV5uCnMjgnQ=Cr&{PNOQW%8cHc~0C19eHl= zXsRA80dM|Qx7PRI0|%yoZhatrvw&DDA?9>i%5xLX0o?C&%t&>0I*ZUiQs_;VD6Gyw zU;F_&pF;_28#B04>v0pQ0<&uWChAk8+ycCK5%1aBE~Zvpj0Dz5;U55SkaYf?M%1y7 zoS^h_i?!i4cA4r+;yUi8!OjoN@327-)62gmNxn z`dG`hsu+=2syyLEduU%m`r-Er11|beHxZUp`SPXhCe(QQ;?Uz}&%*qSW6xWy0Ro(o zx2(`^%x@k$-z0|OF)OUCef3gqDlAKD`L(9rD<&L!YL213-1;=AYte^RYV)!(j0uY% z-6yfPGkqgS7aCIHp~>-nxAhaH(6AHU-L3szMVLm=`yLaxf(y;9y9*dAf^9nz=w}u8 z@}kr=Q~lU-$%G$w>XiB{2eezR?b1!kv>v2%M!*xeXG7$=o;jl6W?DbAUf@vky~oZc+<6Z7e%ZVP`gYn$;(FfEGw01TDoohMjYeyw zG>2|-A`^0lf>w7WchznB+@fJ%exzTIgDPVr`8ezVCJ1!W#QpOIE4S3vL8P^GrIBUy zn9SZZBkc&+wIf%XLn3zP{Cz+YPanljX?!=i^I@JFJ!!}6MOKqWeocvCm`fuu>(R9q zQylWF>f0^gmFZKYaF;BlkYKZM?_q;~^5ML|m{DZ?DX%R)R)6^|Qo?30Jn1DjA@sN7 z5^_ldcu48>?LVYY<7QFByaSz5@wtMNTQg#Ghn0o-L0qkE3mIPBrKJUtk@d~TaG>w-G&$lS1v z8n1C0|Gsq0gIbd%Q52l20RZefLg;hHz0$TlK-dO&pg__YI?Z_5$stxN-mMAqcfMrt zXQPFk%FQ#412a4FGmLn#>r;1np{%X8kIQ2O?d`UryszU&w$CHXy+lzBf3Q`%yiK6v zE-^u!tN1Rl!lakWJYOKLI5Y>gX$=tIiMtdJm+irXs8SfwDAV9ECt$?~lo+^WLE_8z zw^-Dv>C9~?)gW#IO0Q9>zb=w!A>X1eX_-}l1QgEK^`4#T zM$=2PGnZ-})dC=#j(YsG|xT;(1E7 z*~Un$5`Zc7j#kFY3<@fKb?HsZ1AY2q8phJe^lnpH7;FF4VOFG>c~%K>d^pf=q%xu0 z?(D}z8p4pPHUFrPx}j*!@b*oC$<5kG3+k)S;{z~4sKa?_gT^5pXmaRkAo>DX#on=XQg~M!`+{O4yk)lPWnB^ZJ+tVu}hH} ztY_n)$@${pvFa3E6#cf8w}w3CcD5a&i!(|X<$R1 z=Dj&0VaCUNwnEwJO^g7Vi4D11SE*ZOu*Bm;-Ar(KlCD0#?#9oiiMOcH1)GCD#(iF` z;vvA3He00d>hv4!sMK4>c{-TeICxjwhEYE+d#enP3RVnqk7e5)v-!-h*F&v?x7Z}l z7?UGJYBd6^IylXF`i?r~%a&X$Lky`DgK!o5o1?JRX`2_4}U4;vQaz;oH<*=2qSA#CMo5lBvDR z@cscoo1OFo#A&$c7^lb#0l-~Sq|PTiE-e1w19gfP^WqD(Sp`e1ahHgvkn4hFNePKz zfHO~pW@KHOL~+`$cNTup=H8Wm$h1PBK`Gp_0iSit$!!HOtbH+2$8qdt?KFwxd-x_l zIH4>3z1%kR%N3suUZ{;!>b_87TwwYexr%me>UL>Ox%0dCR>c};@RNRV#*u+egXqy^ z;tQpi`QS9efTwg;>iXhrEUIB>820CtR)~F@J|IZ0I8T4&g5-yh*B47)T)}<}3B~+@ zcHcsZvbVxjsZ-XqY;zd4f4BxNw>9B%CD8rJVR#BU+#AW$$r0zF>+jCHqQQ0L2@_tx z>zhjbJRlr?a4ODQ!OMq_9Kn_Pr^z*a=bR(`yvB6BkHhNjOfB!pM!8+z-*=CHm0T&!z%OI3->^=&6rKJd8`(mRV@w}D$gM*0_D1oZP z@O-v7&o76=k`OS9Mf(BKmY!a0d-=dOo~bknOi2RrSyv^vPjlm#4|L2jc_cw@MauYn zL#c4A#W$33q(VCuP;gMFHhb&$Q#z9nx#dvj_wmSD0MRV*@C=+aL2r^|Hq2)}4K9`wHXo>^M~V~S$Pym?=)eQoweH*TBhF>XgD~G1#Y}_Qbh%pb021EX z$q}`g_lWj~^{3q4R45%HJ0P9~7+Pk*2g;P4O$xP^@^zUHQdHjGuT0cn<9i^4j+dzN z>5hCFv7Y=$Ide52$MUn^jNrwhl*DjxVZK*%Njwg_k@7696KhVYtLs3eMVXpS(`>`c zNq)L6RFx|ucsJ#H(g#$|QYA@1Z9tg#MM=C;oRe&29w~ISH5oh#XgmB?W5R})1Ipx=HM=`Sjg)w|M`F|*)R+KeHn@2 z-2(A&s8M}*T%%mIni}yCpcY)u0;|t+b~H}&`39J<&(|7*-1YkG^)YG4X6|e`#o}ey z?yedkNrsp(Sy+6qL=RigT%O6<5!3jxRF+GpEq|o2`R})kc)6<|Luf%5#4lS7UAbec z)lw`f4$Lyk9jh4umj3IyB+S5$K8f6vqq4RlX-77KxxM&2t_9e#CnH|#c2ffJDCRQ_ z3382UuoGF-EgK5!Enoux(~6$NU`D;BM2^^Qrsx%^EUC_*6m^AwIo7^`vH~b>G--MdUQE5W(R7WlZ80NDR|8I$DK%=>zT&>*V3oH% zO~oe3WSWex7*gPh62o8e#ScQ&>3cT^0|SehTx075>~tZa04Mo?Ih;b}w84v@Ao(v$ zpj3gyRe{%&tvZ&2qm?0eliaLyNW=YPiNv)ts^&;Bg$)z-`^3_nutV%VgjP(&leoy( zYHOIsRH`=hVwM$$w=fjN0so$9Kw>NNVPu0~yo79_YHfN5r@``zhEoayp76F-^&h6~ zA10-1`*O#q=IX4HfxuV|^fv6u#^0~eO#1Gb5-b#u>>(4v{^m@L@{@nQG<_)&G~gLn zV%+@n8U17@4tDixO04%d^Jd#UQXEiX4v|T48dJkr)!MP#F;cG`j*qdT2XOR!RK=?1 zjx5~`&VJP;V<94R9h|+K(Z1a2_TFo)Nyo4ZzvzY5Zq_^Pnq*Z$dlvy zyEU!aOxl$CAC#CdWd;!Pp2F;t$Ib0b-?ERdnwM!A0jIR?bQ-BJ@cgZl1Af-DkKhC$ zhJ4qsV#yeIXF^DFLk|~tX-!BO@D>w0!QSqx%Y8vtsjhQ!GR`LoVX2+TI-+|gsD0r> zw!Q}{ql}06i|@>uM$0?6uFJ^5kjb*E_N<%PdnS%6jl1+l+#><^ z1L@<(UvJj5o_wWQENMWwOj+4maUZA~7!5r7(>_JESF|eGc;B>aK}PRklOz189ho=f zs7T8$>}~sBdjnO{MDD9iHJ4oaqg#J^tON>MM$GT0R z`H>Xpp82Gk8dT`l-FmtL!U$UNBzX5b@ztA7Ct~a;$o0ihSJH2OF;UR!l1H5uA-u)&IuqJ2m@wC=HXdcyeLMKl`W0#>mYsFWq72pb=t;}@9M!1L&#=*^ z)0FwH=BP7)u2j;_+8bnP3=LQGj4$&We-mpSj^~rN+sIcdh0`ss#f$!R}Sx5);l?|(;QE6k<=rF_lEA)&cbF;YAyIOE=;p5m9K{jSou79 zI8IX5=-uYs`{TcMbUO_XA0%a{Is|Be$&pjzho4p){ zql|6*tZ(<4H0^GYxES3k=lgH$O3v)tT%3fJEhHqu&nx}5cHXp)zOsN9B%B#0mSad; zFoClS3WU*9DaQ$=lu!vtP9@Os0|s3*h%6<2h40idzCs-<2JrfqR? zk_eB@!-(|VtqhiET)}99G-3GPbv}PL`x?n>a)*|@>4%u%Ah*~vw&t6olRoawhb4Dq zxwgQeAnQL?MS`C&VddOO!OtsGy)frEE`N$e9T`xas1+c0C;F1ECPFZb&CdRG@SvPh zHen=yB+MW0jSO6_ILLXQUQa9z$x4)@9w`!Drt}8I+ zFJ$Lel)0(LxrSOabTR==yy)llJ;E{_ZIjjp{8u66WvRlZ&)@8CHIVDsm8G^`E!$S6 zEWb)JG-6Oa(zKK$1RVgE{RT6g(Tjq(>4vZm$5^y(r3*NXmd7&JP-}fg{`q(2< zlM6Fz$dJx0(_b9vUc`E(d2I{L1dYzEonWjDFe6{j5_i2uvV8hnlFiNcnmwq8DLUe? zF}E|QG$8&P10KB@;Ne!;RXiJd7}_)nq(TR&JWYLydb*VX{W{U>^gL_Y-@<*~k>L(s zDi#*I$c+kkO5C>f0hJ_zxcB|HWd0K#v_V813wWAUZS{yXUpBmKJDZEW`tCz)^|FLX!$Yvcs%&sJsO?LB1?+pj(r8 zR@C_aG)?h?&~yqyE_m4&|BU||Vr=l!%ZyE3sloP7e!ARZL>g_g;O+9L-$wB}Kkjf5 zNxSA%FSLh+M~1=_t!$ps^in!Pvgzywwme?~$vp)2`muu-6IFIcJ&h3ZZ6m*@hec5+ zHr`=-uZI?lpGz32igaDGIjb$azNvn2y_o(jM|%Q5fXF@v9JjR%3-}Vr|0`3< zgw&p`@nLAA?{cvqd&`UgukRBWfD{g(*(M%CWOh7EaZv?XoRILESD+N}x6}ZMR=ql)eO-&AMtW&Q(8v5J+kN^{A9QH7EKv<&ma5M6X&E z`EYJG#-mJo;HMQ_|D(_=Gv&TH-^E*CmmEyJCx4T1Hs|$g4&c4pKXeWeM)JSm=R>C} z5linYrcSw9iOh8ORxL8+AB&Irq3pW$^BN?vX=Mv|+_kM`LBHKTPd}TN=M}a=DWjZQlNtcQw|g-23rY^+N&1|NF@U zh1}`)4tK$_J<|!i*RAUW4TR|Q@h_f&6}5WJXT}C<{L$ai%rZ~#R}_#| zY07%8C-ik^N-U;xbw#RzzDy6^6Jp?#1$_Q&tag0{w)rG&{fEoF$M(3PyDu?(2KF#Y z;a}PsoC8kF8%XJqJ4Tt4I~Gq1EF zJn&^UVEicl5#1ikrR(?HW3AlryBsU)*v_NDl5AS9xA!`Z^Q+<$^sl99y;mkmr>jgF z@vMvVEGD*;6mC=IYhy(*s`qMDk!>e2DWC)0c5z$fa+fL1_QXGY5->lfYHo;^#m5K3 zlVoa?V4iE55QLs#`CWOxCZRi}NqJiT-R_l`N743=Zi`Fj&iuUiGBfs7OxGT4V=tyG zTg^{;(8qAIGy9!F(2&3`tlRNNura+5Z>zo5(C4E0wjAXlWt;KYF~=1SNRp={DRC(BjN?2_nb~jiTTn^_lNlCgz!{&!VDQ#w zHQTsQVN3ba6dWcKEY3$T`B|kSfzeI^d&lh0!^=O_0tN5f%B}c{{3q@Huy1oTZA4}Z zn)Bl2ZsbGobM$IVJw9ommjL3*pU!agWduuHd0$xyXI`^0XPOQ^8gsofs#Tx;-t?)w z(?XuyQld0vm{A{#^Ab$1?}$-Z?QQ#n(Kk5eeWQl-@E?K?UChy7m%nDoh z(^Q$5xWavy?gUy8+(qSgbyrZ4rP+UAbe)jt$X6>SRDU!6tCHTR3LJ$=1mz{_UCv6r z91v`dkYErDAw(`6RxrYPA+Pz{sI39NoBo@|mD-&2NPvk_E922e94A9%6k6KJR{5T* znT6`Ks8=mowa=weT-TXFI7y9kA^0HIiC~dvc_Xj~RFjuX707cpif|CMcAzr>%`hdx z9%U@^>H|pLbe8aXk25P{an1mf`j`|lm+#?wAhS0oN9BU~(WV5*_<1Cx!&@Kfe*LsZ z#xNv#;BmU*JF0Y>wD{@XDQ8rA(sB)=~WHWY4TNE{3~;jWSIcEeykAHwyz)==sQ(R_`c%8182p(*WvfbZofYY;x!ZKyrwdD&By*|(Fnc93GX9Xwjhnb*afJ1< z{|%U2cW=2%Z!{?uF;4e+%(hHZ$HskLzLyl5J7lhD?hizLX_}8D!Y!45ozc1s+^~kg z6MmA4fjT@reji}|wz;5=B-Q`W5Z+Eip>?$uCYzxLJ>6IvixJ+Mv+PwRIwn_&m`5iI@ktoyk)~-3 z*lE)2oZ)%J&s9Wj$bF~CeTTvA)r0+Vv6q*3YYMnSWH)&6hTXV9tn9LAbLFHqZ8+Q2 zB*qcrA#T16+e;sBZJd-o5MQR+_ObshFd!y&<{=32VXHF4~)@6s#BegnGdt^(FNOZb}PtvFBX9*YBs298KuJ-@x z>1p2zJ)b_f*T=(UrAW^=APwEiNuK!AlOS5s8r?{g1LJiPJ+eh@K@I%J?4}<~} zhniE|^{q1UMMgvS=$7M=ad_S6EmVh1Q2BxR4kX ze_=F&eAzXYbaP~_)<&yqjjj|AKch2~xVpDHPS_#y)RhzHSe}^J{N#c)t4%GGY|snt zh|xGBuEg20iw~+uAyT~%_{&EtV21_D0ZKIFbe5dVKOL({ebiT@iL9LRnkJw}3T)p( zsg_>eT8fj;ulKcYk!Mbjc5$*+xUxEu&bY;+;2!|DrEJ9*4purI&;39k7tFTZgb@Wc z$G5B+)BF602J9-rA$~_Xg?DNxp1d9}cchzJx=%UisQK11zmc*qd5$2Z)?_ygKa+}_ zXZrn+5$cPqDUysR`#c?qVR*j^wGbQXdG5koO=%|OaCf#jKl~nHz2cJAn?OIKr+btr zFY)*zTip4ce7N}w*I9}cY0{qNZ{0RcROZ;YRB`8}e+GZzvOU7}RP5R`F(56zJui-@ z2I9(@hN`YQ_TfBl^DW++k#5iCFJG2NytJmg=7J^&!f;=ar#S|N z*lt$_?37VhwQQY;bv8KvT98s5ZXf?U=nbBASo!zy_&!6&i`SNX?b>c^Us{xeJ-5rW zg?M<(OfH>q$)Jg>1|7YbTqGSMo?f%JA7`Zbu7W&VaLN?r*-)ytbRNbZ_Cd=W0_+e- zr#2!XLCmJ0*YA-O8t0l>lhd(eRO|Y=>RCH*y$JDmNtwZ+L#6@v4)|wyP+sjaY(f{aoc)$GG@AORY zoMwNNG3MtEs)Kah6YL9N9U(H9S=NuLG7>{OLjzZ0oU}25M4u!vTRBvH9tR(zhG;Ph3odi*cLc;Q(HuWeNP56EWcX2MuH}N7$mOQZ_SIijKcp(muW)fh+g$BG7d1|`K2f&k z8+!8A#pI#z;Q7qfG64T-t?l0J*k&SZq-_l#@u6gMqY_=mgB-3J7uP-2smaMDp>rRQ z!@ZpqL@O-k4Jm*>wvrHBiVz!!LX<2}Id8$_ZJw&fudy#j=AZDsf6^6jVebUQ>8Z8R zpH%9e5!-A2sZaU+k9!R%V2WqeeNsnbl=)|M(KNj=C^xp$=)?mFWGa(Y7b;#AoAYc9 zTn<>Ta9}CK4^|cV3RhaB1>u3IgXlS8p4n-Nbb?K;?PZEf@A**2B`LXh@~}4Rew(9? zyRCPApbS>OviU&QFknJ`tKrzXV3e|16+qH}L(uer($WY_ym&@)95Gshjh=JHv8l^k ztF72EA-+she9Z>@lbaNsT%XMTd=5V|?f@;v$b6rI{pZ=pz>IHFLZT^~AX_*~K83mJ zQQd;SL`4GUU4>=*euxX1GOXCEVY68(>1jD=2rXwlQt!w*Pv9R8T_UqQ%8Ab*dC`T> z_BSUl7KSRu^>V$~@Rl@|zq2aTX~!+5lYs-tl?Gh*cCE-74pPIEIz)Zwn5}_}n8@_J zO1kPcc1GUO{bF1jd)6t^8IoegMFLYoF_*LzR8mHoRw1 zCz6A@`iq?Gl&|y5S@0VBC00t;b(Z@rQ&n>F3eYwa9+{~?|AY;_do^2=%~gZck$&|# zAR~~G> zM0_tTu*Kbn$(=Q9*B^E#a^L3tT%586CC-clk`fo1m6@3tnDqlLN%mZ zPE&>}_>l&+?xFP9(OR(J?SC4D;02kNP9W@9RnE#u3rwqCj$)SJFcZEO7+$}SQFL`O zElu@&$G`rp)cS$^p3E%QL15q~tLs|}X1oCD;5-7w*O(0?(IBi9ah#TDWK4Qd#@(LF(@M?Ry}CLkZ2qeM25VjK@p#q5 zquGl}s6vVe@?&##>o7CN_7&z|%LNbG6E-}`vF~WGj97PHeLc;;7Q;xhPf)fisS>bD zsEBV|x(VmCc@q;Yj=fx{ZHPT91}>$EHCHxE^V9J`J7CpYWn_w7kH0^hz$n=GJm+K|$T7S^j;2_&-rIJuPm zlFWCoYT*@5OYHE^_i)%pjGSZi^JkOe<%DJ?K`Vs>6p5*!1)svQlrqNoO@{|nE~0-) znRudBJ7lL7xLI9`DoKgTyovZi(!Uoj(3vv-mMT%btOmr01LG<9)aCG+z4AJ1g~MmN z>8C^OtioE;7t!2UaE^D|(qhbjxZ)M!@nk~qyfT^v;fyv6J4mxyQ8I9;{D+Vcf07uw ztK2PKwZ%OMH0J@dYMn~XQ^-9;Djf1y|DGw-4~cR*)-81d%H;2x>}gO7g))b`#Xr$- z8U*aLCNpBktjP{(lB3MKi&=c4Y87e;Z?~YMmkJvod{mUS0(2-ewD4+F|E&cN?(5Cy9WC`Xrx$V;V&Wego@^$(ko-HllKG$_#x9-3 z+B!bh9SRn%i#i;(*#`2S>~wv5-90MX#jZr!SljgN1O+n%+IvQEVF4UqN4_!|&Wx5F>RFtTTWepNXi^E9Y9nt8YOrX{ zb3?L}hUc^~4OSM3F1lI@$Q}3a32^VB3c9mimDAW3wrHk-^Hwe`uHg{6_Lw6HXDjus za@zd|cFnz3t$Q?U?nMuQpzo++W|^KOi@>N>hvPQ@K| zBlA+}>jec}K}N$$=Ba#wb{SCuI-b|eDa=+R3v0CrHuoV9a z2;LB1>1O4*T)=v7MS4}Dz$iG&ifTOp3Q-=wP(J1GQPyvX75L!WZBl1)rS$>T{{qM zU(zn2WAP_Wz99Gqwy^N{6CCLk$#B8yI&pHktA%kcYTm@dUM-!ozEuxt5Fx@kOM+Ho zXScSt>Q>!-xsUhhy!>6Kt52WvcXjvo^_`pc7w~Aj zw}lMm2-r;b)r;hbxJT`#8RY)caQl#wi|lk!Eb6tT~D29j;HNC-r@cRR!_$ z)=9NvCTI?zgG@Gu!%4BHWhY1j{=V-%4EGKwKa#&BibXDZr`I(?m5M9E4cTNx^-<{M z<9I07zi^FkMs`5?W=FQ4oN!E1Wth#Y zi0KCzwr((zSp6z1|2;Jlt3jW;vJ9u@g*Vb)FkVRNP##va;Pj#-gRpn_-Bv}kd-YsK z()!$C4(F!xp?W~*4u5qK6*pP(#~$o0i}ESgd)%|95Uo_|->s0U7V#6@zk;9@!3h6+ z_Qa@&k$2U*;eu>`^FC6*j~;*2fMrJ2T)dbW3T)YPwNu%Dh#G{4sC^{ij|WyzY*A~- z=u!UoOK;8Rz%*&AXZhgYJNoYS7kzg#OgY`?Jh9e4$cv+pI`5sr*^hL&w?x^IT>*;g z&WfVs{R2p=o^)*Yz7Q3Qm14oMaXwF`FicMZ#8JJ4j$5w`^p{W({}eGbo-s?`ACw9Z zHyh!|qmMEg(4iDe`)t!+(R{@gSU||d8)g4?Agn9YnB>N*yvJ~cRt6cQq~+2#fSAtG zH;fW4*Z+o|gf5LGm3KEs$V2hXce0|R#nl>mt^e%)>nLjh>nuIYuvu;bvZKp#=$o&u<09X_hfbxmM3w z3@D6+YYS(hSL2pt>L&oItoQlaCFs_@)e4$cLdJ~jQLf59OL82^iPN7o1^kv{wUH)& z?!|(4I}#Mrx_jVSJe%uiGpy`V=yP?`y(f3Jeb{t1p1>>&oOT|78y7quK@+xr#(5BC z1Qz|7=v^$9rO_&ym*>8x!FuQK1gKciR>PY7~=kwiwll(T4 z(13YMX{8@nq$b5RcR;a(Uxj?|R6jaz;+KJ}0~U)Q?T$hV(IsV?owqi=+wlgvi z-xGrh;OMerQ5Ne7xVj+@-ilia&XhU7%fA%$?Of2VQ4^s+$S)=x_$GEJkcfoA+L(4+ z)HX}Fkus>r$Uf;e)_=F|6$hDy{4KMnJdatOTha|3MMsKWgTJ-sQY5^ z?ay_Lc4p1UMF~N{@CFwZShB=xae1}};q^n*6Ssd%o$*Hg^Q^WoXS&WI&7fmutl;Lq zr5Bw;Nc0^W1Os?8=*O8tADm(ejm}bJ5oP56Seq~MduXuP z(BvZ~NE1`@P1M!BEgHmsSHxM~Uf;eo3fd0z+t|CCC&_B$b+jF=I;#W6Evlk}vZk+c%Qo&bCu0|&VuIRnc)vP==2L~x2M zf6LzFg1vE3<1@@bV-zZ# zg`(VtE%vlCNWNgpN!yp|YJ5v|`EzQ|v>H(B$+qn$KlPSoPgk3@8q$`hX7Herye zU?;5WD~nYvnuA^tKj))h!eG>mD^49&+P4HJmFf~P=itu0j{DUJtx(`gBN!Ne%P`7C zk$=Fa$f11);XN8~5KJ${iIw`EfAfB4Vc8URw0xuU;_?p3ky2^C9f6?`p5<+@qN)tzeIkH@7}~j%}}q; zTXCu0ia-{7*hK%-HgWepe=TYa{|L*a<@dIfrYYfZ&6P90Z5Lbj@tL=)G_!e*I$1ZC zu&waqN9lX1b`?#t(OtLiSs}KPOL4hn)b~G9iR7049c_YJr~!%*E>F`-FA8daPcfVe z!RN^?fal%VxzhL{-Z?(wmrcQ5&)vKfLuqfCZQB{=PU|&gAD8?5vm5rg26JQa1sNS5 zlp2LBwRl;t_jd*J%_shhAkgY)#~*oMC&oXs%f-!ol6z6cynhucc4Zk!g!aeP3t|Zi zKFFSaCqf-X!3(kTE}a1%JvQIuknvVoB@;h9-zus0SoS;S&+@v@5qKEozh6;Z$Sg(7 zSew#JE_y^m7qO~6Xjk{D@Rxm0JC<-0h|^w5rrCj!h8F{<%}Qe3Z}D4e(`Q@H2^pET z*6Pe~d<4a@ZBIszKgfVPh<&)OD_o_?d?>0f&9OlHbT&Rymz{DN&-?uBv;1q%e2cnp zfKa)DfRSObxky^2%__x0hll4obC*-Ch9)OK)zl@RE_**OSbCOp5$25p*@SRGg6Zx-m=+GT4^Rs06(^oeHi-fm3Dd)DV(Hjs73( zA7)Do6R=df<(+_~Gs&HY%49sPsI+2eEk?;|WTZ+*5SIB-6dBjcRJ+ojt)%0?ZaV@J z95Uun4X_?kRL<2@b3E%f-@6Hgf^{-5oGm2;MtPyB7Rk-|S5F^6=P;c6;oNof0ix5y z+kKiM@(>AKKvx)vH~Yv4aL14#vKM)Kv{{r-d=sa&D>w-0lkt?0(Xs%n&s&)bZ!&PyOr#Ay*b>PqCnn?XzHoWVN%N;Rb#X|!Y`@VPlug+Epyhc z|1ybFe0(FJ39DbxPmQAXoy)cGzteyxrlT)hl?mZ6)Zg?THWJ@r_o9 zcaXOT1zMmP6>@`rgaL0~-37UHmFI`^n+f_CI4y~@8=|i-x%>VlT_3)f$j*7+m{kAF zOo&5FXOHu2n=NN^CYeiIzyWKupa9<;vaqb;msGf`CDjas(l6x?~$u>!oVEMp=JdQ%wJv^}6#)ok^M@um^!>P%3! z?bS&c{w{dw7>Q<>zrnp@zW!sU5f@5TFeq@|SLGR~di1!+L8e!%b=-sNmtZ+N_))-{ z+QXRQ1s*E?u*Qc$j)Xv~$g_UfjNU{u_R^U@1WIatjY?$Bm0By@hSY)*KwAliOufY( z+*Nzx0Urw**isg?D}H9xbk~GPzpr2H+$uuBP9!>-@zN#a{!t3nyqBa{sHP{7x&*cU z+%IQ*7!0bCu;-3$dpIu`Q`2ByDi9;5#Kbn5mOh`@QmY`3he@Q}mf$L`yBDaOBw0$M zH6_z;{>B}V-73o&9FXferH;VI=`zE$q-wXsvG4?Qa?$6LbWtBP5Mk)vb2cFTozYW< z?)iW|yiz?0N9QfqgB=9-LZ>*?+lmXA!u;WpzN+4)a34+taR?g2}d}mKZ7wa@B>>nh3MjQV*wD z>eE0jgxP7kCCg383j1JWX9KHE!v%-tReW}sObJ>DJ)h? zeaP)ihZ?>OZ}Z+8T3cf7#lFza6&A+s%vFB$KcSFZJ;jh$2wUqgZO#@tQN#tv^1Sq~wI~F%@Gm|%2?ogXIIauKB#cLxz(3Q8Zh7U2( ztouGmF^V%rL6?>7mMhItFK$%Ocdi1`MzJa+yAfoU#?NjnRbV@a+4Dqa1f$ zy6jTXp@e$ZvZwIY9iDiL78@}~Xk*vl93ybOq9drHWmILk?g3EjVo}s!M}81_G5nNr;Ua>bMC~2zZR))d{>3aXj~CW z(<#%WOdTsrIPNUffEMBhTLHl8y2O>AEeUoBRcEKUP)yFBp?w8Rq`g_J-@ z1M52JE%nPxA0dk_w=~zYEL=dwqm`t^(FD^&H=KLq0+%n$Y6K=O7X1hRR&%+f6*@FT zwbq9_TU|dHxTH;Ub(6_BAf?U9$zju=6F2jbQmI`ZL<>d8YOmH`A?&#{VE|Gec1sJ^BuiEX281M zsRvqYX~yM(2k6=Wcn+I>61L9ygV|r(Oa!vB(@kn(D{N)A`Zn>c?H%3<G7zIX-R}dyhVM`_U2U6elsxTMe>n?N%mlZl_xEsC|jj0#J9_5p77 zRWpjm<64Wx=3a^bjTc5YYZy!dbsm*9Lw>Ehsf}7T#EC`MihtYe^AJUnApn<2+hSxS zD`126DBb`|@AVRWcJ43M@oN&|towQFd5)Xz^0;dF09;MsTNU3euJCl{mo~AKnm80( zI5p?Gg&L!mr#kC4DQjVDjgI!lzq>Z>>H)rz@cjLBl!i4{l@H`vqa_z9Eh1Hn{l<$5 z&pzFv-ZOiy!Zv!n&xhtsM@;HeS%35!uT!Yggy#pTsm%!C9>FYqoaTuAvH36MSL#*7 zt5jvof=FLGX>zZL#Z{+HIPMN3$| zraM!W^d~%D;SW@!V%Iv{zhZEL5bv#uLcqi@G?SX0v*<;rf1NJ5eFKAk^ zRY!zf47GI)bY+l5%O!w{A%CX6ske`dO!?dxCZbPfZS^B6I-edzGaXfJwl1}PbIx^3E;#!E z(-FI%^WA6gC^db=2YvyQFX%8GSB&*vtP}b)9@NuwM!r+m&KSgaBT55hL3+EZK?zf+ zwLgNXTqOac_h0t~Kk_t^@KdVgi)_5m54^DdZ3r8K)tL)i)j2Eo&_>OK4b6+T`2BXu z>R<`A7oe`x=@icC{JMWKVzmBc@bjzN{_9yBUkE_rWu*cY7Uud^A{AV^s@SIn5A_bh za{hf~x6-T^ZlNTxGi{K#66EO2g7q@>|`tIV+i`4~A6bIwg{IeZ34( z=c&I~jKftMklHC3c+4dE7_r`S>U<>y1(yaV877vsw(Rg#wMjbb*rar^+)ZKZqtqDeC{XJks+cgWotWBsN zC{Y1A;z`%@*#2$rHfESS^|?GVJay+y&;rjey|KV=kyWHz_Pl~^aH#NT<>G32!vP~q zcOLwSC^10nX&}wQ^?@=+jh@8*v(qIWV?J?rH)kGrZ{wHCTRUR)-vJ?9OoO<9OVh~( z?K8aM!jm5!S8K_mY>gUAb-ZL?5fiDq*Cbi5Xrp}nA(E8{GgZe2)MAbUpw(HgJ^1%wKK*h|Rq&$-dAP5ky^n<_+`Gno>Wzb}N_6eN)-iuW;m z4Ir~tu1|5?_=)MQ2I|S1Uz^k_{BkI2$AL=LF&hIgtrQA=2O=(}f!)TUk?VmwgAF~r8v%5Ut zt(!53m!I$*-`rP!^Yx@C`U_RmTy#%t|BD}~A5LJZb6j5-;~9Nuzo$hijhjCvnvW5M zC{qJMC5acnwelm%Qy7X4C>#!Ad|TZphKL-$UgR`ojL&{Vwkr}gQAZmmF=Ux*M21?A zkmI*(0OwA=@(QI@2E;%m18AUsxARqoKez|_uaxEI;lw3}s>}i&4_&MVtJ5QAsV6*T zsCAy;B#Km-4w_~4e%S8UL}_@61cV)zxt=7~(to9RZ_5(T6RPPaZWxh#x0D@+_+1Hp zDGhJl$fM(>I6Yk+xoc3gjVQOj^CPip2mD(BHi~bva z)+hE`)1|sr=Iw`ueLR4<2z7=rD|_?TI_WaTHZ4SkC~V?o;7rejZb%=O zQPrkInjSuxDz^*UQY}+d=)}tpaaLAkuF$PC)%S5}pf?JGsXsYQD)DJL1lZ{-`h1;T z7OjbJN!bE&4Ksxau16=FvBa@ zJeBl4tt>pPB`rOyF$L<60Z9RYZT_T+|3kpp)ym$+ t_kR~i^WdVu2)zEE9X#!wt%05v&Tjv^OqVUu7)A!`gOY|~D_Cuwe~)VS{jO&XryQ`Fff?PN^&|dFtAg=&krbYz{qU>M+^oA z2S!;=8XWNXWCPiNV2KE38y&r8@OME(Nof=sGhQKmSbU-RAA z+%hiV31%=XqC5fu!C_&!%U_CI8(9LakC7`YSK(pDC)0X<`_o+qKi9Qc*xBh(Xku6k z(4i&F6H!-!=)$Rw&77QV6y2#`O?%%<)Qr;$&YiR{IO`_!KpVjq5}Jak7BbuDM9QL#L^FzZQqjk2S%oefqbo2n46>1f#lyx36$^F#DD5@3xv>7hhx_AyopLljhzTCeM6k<6f z^wm2d#so|5V0kYt8b-;T(0;2iJqybpnwof_Avvlz!@y#_&R=5*nEYi+2S2Z4o|!8# z0N3}@J`j-=)w&PW8=6CAY-_Yte~jpyhB{zLt3~Y)PIgWZ3&#bZ;nR!}^(GL*Mdcz4 zvt_uS%OdKAJ7pIX7@L@oKWKdX_)$FQn)9sXu+K*zPkMkQF6xp{&3(kvmhF{(@nYP-g4}`sZ{``5{vQne4w1^THcLxd$IN$7R zFzvyj6bm@`S@=DhLsdyBB5ypEf{tHHSNEq95$*`d`%ShC>nL`f(CXuVf5-36x6Tg_ zQFm?@>y6hsT>oA@lqsfdwmITYi-o_2{KyyBu{>UCe7ioJ8;V8G%FHY+E2Eju&(9wM zfpv8)4>$XQp{lG#t^N5y(vOTNl6)OOjixJ>=^2?!HcBbTx!tJ(jHC_aKrVn zvACM-+p~?Hz^nbPu5$jqxBI`?kx~|Rc1UySd5@2en*-rgLu_)!xpOtTwGi3(4AyT; zy2_~>)CoMmHU(D5fgQ-m);2f8ju-3A%*?Q`uzvphxxc?(E`i~Ml74`MgfuoXqUBvx zm;Jj+n@&d;49?1gt}c}%{m+?*6n`qz&~L}BY0$#_8u}zSuLfMOzP?_eUSMZ)U97EF zu5$=16|x{{J4K*m+TPYSmP%oK2`gr{K*68=iF3c&I>Z(SqfseUx}8T{SqTcMt6T3e z0E6fMj;HybZx(*~)avB`bK47Co^YZfUfQRb&7YS@_8CEFbyA6a&Ee#C~P9Vh1 zV|U_XU0od%3YC0&#(!-2^Y^c6=r^Nwr}8gf*1LUG^~&o_xX$es-9|c+w$ zUT($z_Tq(>$|q9?{roIEx3J**aJk3H$%#(P8S^kZzIS%k?|q;Fz31iOF>JOPt<`T7 z@jH2MJ@#Sp6Vu)C67~@1de}phD-ctwZHFf-t?+PgO18;CS=Ns6%~^&Xz54W9v{5bi z*66F(U?^_uV03czBV!F_z9aq!Np2tQgWf>3-Y>!$$K&k`^p(TXPR&-H4qi;VMx`_9 zp8FqvPtL5>RZY5mtxlGs#q||D&Dp-n>}KZBzF(GXKc4t~$5};c86eNz9I7`G(|+@wyjvl_ zl0Ek2&vbk{0A_aFS%X9L?aGYM(epWR(u z(*9K#bqBD*%39tgEz8Opm#;Z|a&4spNi>ISS6zRJI&GyhK_;_yvE>Co58YHagF1`I zR8rTt*!33^lWfQqmpm(r51G@D_P2bPGGIn0*Jtv#FX+{nYmjwAYG!|=P9ar!Y~^ZlVuXlQ7R>iKdwtg!KO3Me#2z}^1;FhUjo1&A>) zFhtFU1J4DjNJv13@R29F*;bjt)V!;gQi~ zVpJL60Dvuth=|fC#WPGRshOA*S+hEv7LusJYI)Q#tTxlRJPHa53O@)P8*U@o1@sSQyhGF|7ywYoe{O>k^3B={%Fz9Tw z>DG~x`fvb846;DaDknuE1u7`BAJW~kx&;5Y_^MBDUu2>j{KQ@`ti34kCn-iJ>NUBn zCbN(ck4UMWg5UVXMM~=iIDWgzCGJU0C9}<~xYFiVf9z`H=nrSNa!Z{4OSvhEIiM?q z*dOZ{sE>l*I_DC=tBde&Rtxa8d`c3#cFB@42bL}`oVT2hlb(o%)wEGH@IV%&`i&PO zSN$m3vYVRGg69K|2d21~fTgEunHxo_5R2;nRD{}ikAn`z{4e8cF*6K>h-qy1Bwm*5 zph_>m0gn*n*Z-?gaayJ-iQQ1dfAPc|Rp&n#1IvFdG?k|%C)0`qLcJ>KtKW%CI_4FXE=ENxvB~!fP4K^sO0$Tzv#QOH6(1D z#MeG4nBE-Y9KEitE{0@yY%0W`xx%m#le0UN;dO=?`S~(G;^XTL+EL7ERE3!C!3RH! zzw8%Fc=x)$ao)kXcXhWfce_$ZU&0KtSuvq_6!DFFzyB;y+u88u9gn7HyHon^;o)IV zGI>~AIPl0IjWX2DB)G+|T?heMzZtHR+~HyxeAK4to)#%F&iVCy52vBv-EHA9x4Et^ zp&K^UhN#Ejg&`lYufx_mnO%U5*fs{^f-`e9%VJO&*8io6FU^bnNv`}c5x#v5MHEzZU&E%fowo7 zbwWTzl~Yu-*g0e2TpQ=%Klx|j8&|O`(&GZtVx#@WdNZNg@-qL~LMxR`nYY~wXO;?l z!#(CZv6{+YbAxN|PBR*sZ}Dj-k?(84A&H~`DsN%3)%x6geDUe46)EUyvPO@OH*5iS z8Sc23tV-(YuFJDLQ&Up{PPzOp7Wg+3Q3RE3=~m}k!9%}WDAy@OY_7}3Bjd5o&H9}9 zdeKkFkT6LfKFNAV?IPGUCKRfvP5b!X=rx)KT`z~B#Ntel=ZiQwIfdOM-W&_1t1t1l zMzE9op1imal7TdJSv#+kR^0w5VNcJ?Qr>9qaH2QfU94W)2o|ExC{0FK?+TVRHSKdP zxMkQHYVZUlWN8IPs6AM(BU4bA-n($Wy*#0y^nhKm`cBq58*Tl`1e+L?!YqI+glaTb zAavFha7f+;y*u5)dbd}F`*&kK?DYsgP)uN@Ex%i29SuQPc$9FtdTu=IcKf;wOqf-g zlJNcMPQ;?nq=!PJ_uwI!DgF0mkAJIaPwKq24K>870{;c{Q>9j{> z(7*K&5h8$Ae8S8q*JUO^ESyb0^Shet)RQx?1qB6_RaGBO5P$q|UZ}MvBnkMw+O4g1 z!I$u57oEfS{Wr;bmzg4Qkq)h9-#_uIOm!jQ85)WS8xCLDO(Ic-Ih=Mib6P8I2CE=!r}*V28u2q$^Mj zA$CcVv4H`~imj0m%Q+T>Nm!V@(^5lief^bfJ3lkjFtQ0|s1^dz&fydgA<}szr`tGGd};p((7Tn^#LQZn+%NjzBkNc`-#h`vS%@& z&!cQKRYe;4MovZs_%d;lMOh0kHh`3j?9~6CT0B-EO_K~s?5*gvt@IK;f#uF&>w(5< zTcEGu`DT9~S&V9UxOsp`{P2T|qmskLZ(DD!`@+cyrSKQV-RYITIQHsJxg8EW2=X5= z$@Jp-LJPCA(F8D28sVR?nUo~{BP!LAzGZoQMC4*^$Bmy9cN(l)v&k0E>p+#9YMOas z!7q=5XML-(b|M~w9!sgFkRa+{tF*H|Sy^)#Wr7Ntf1gEcu>`_r2XSwX44=p)QLy)~ z1YBM7r}NvZ>iioW?Xak}UGEDVK?h-7K2mR%I0k*4^=nO@gqa{wKWK~Ruv?8p&ye%W zPtyWMbxPE$+r1&B?N2jZuGuybs_wf>(u{?+&O#5YQd0mx4{*-vUe{4g-noh2yq=bTsO0$ff zyo(m6KceNn*}DReOF#ly1OfDTXJ-e2rVGelJQIM5%H}kEJTd$g2!e%$rTwU+sfk_D z>AJR9sYz8Thm%2P5G=o-K#gT~64>FYD0OiBXk+w9k`U%*L;Pb?5} z@9yrF2)KwKNU1h@TW+>~dAx~sR8&+%Mn-nha*A07q(qGkULrj=_w3RVaRp88&wK$k zR#vnM$O85iG6r*Zy7K92J0>P3s!Uzn4xuhwvq8w8k;ESbLc=~bHuP$sVL1W!&1XO> zfwODs>(yCVnVF%}d3*@t<5FITJsnzz%H_Xrj+e$qMgoF^?d|P{<&GESmsFZNIyPJO z&SAeul9j&&iw9oHsjAjm4##!6ZM1Z_NjF7A?4sEFY)f z^6Q3n9xW{`FE6i!)oPpLMvwo@4-V2DDv>)#AGT}wgxY_atD2!&4A{F11wcea9|9cd zbiHeCdisNrjm6Z4z;;A^Y>%n?C>h_925&Tl|Hs;R60 z{qx6Zu}&EZ>$@f%d>it7@3}XX1?&ib-i(Zl7w*!#1%&ApE0LW-*e z{tbR`irmrtC2bNtWml$L-Ll-p( zfKvBAnr9Ie6$MBq6;<)bh@8W|Y_FXur)3N07m&XR3V zyDHa-h2Od?>?2=JywfZ9ZIW7aVl?4&9yR!c`2R|aBeM$Q=t8Wp)R}z_vNQC)-N>cbar%Byw2bqbTm7Nv1$tNeKZrSu5OxS3tI5g9tI$>S-$ALO z;TmddodKVOj*jkl!FtWM{Oa;jA&nA~TTVHgRNw>JcZsE}*t&O%o4({zAi^&al=jmJknney=9=&*x;hdH(Qu_AD+D5uIqT zt+|;e2CNfH{blGB2utceQWY3;fa->52w5;d*}FY<;1)nWhrbz)-gj^to zS1}`Pyi85$_ASL=!{_maum<=~%ui2GQ=0+@jMl~f@%oTWzd$yLHbT=5yR|mqw0bh$IG`gGex_G)+vX~ho z0o32$@1eR8OZ9zbMspG7O$|8DG&2kfHQ{fsTIDRlo7R9O*4Njk-(Z6J1Ixf?cLFzx z72D612Ypw1_7}GDY$}DQg>kF~%>?hyB5`L+%U^Ydd~xLDSiWXdR8$V8Lu07&FaU}= zxZsV#XXVu!cTNqlv24z$e}0b})SW;cb@^{+IG!diHXFw$wz);iY?QlKZy=3IQf56S z3~5=9CNv^s_UBKmqZ%#2m!prjx-0-R2?!9%9DR#M*45ReAuPS=h2K90aLvc(+Me%t zz8b{;Ro~Id%If<1I?QIbsRovfX2h^cKsvqNd_Y2)Fm|V%6eV`(V5V4>46}Q_(`5w? zl^Q$r>v|TiMDaN*4gR4^}6OhAYGkP^0o>joZTjKad}#9Zk(h2mYoq2 zteR&W8%Tsi$i2sMCEoQ5FmYTcsNe52zoNo!slk*+G~oVXhil*&Zi8B@9Sj1|eaOtq z6ArxG?e;wm3JlbK5k$z^CmUun=yaoaKJx;r23~B#)-^QU%A}tz){{wZ0hBJKgb$6? zFhnB!`Tie(_)!Aw_A^Bt9UU<%KZ_-?`*whkQ`OXr=XC~v5QTNAGhSc#=JwPWrA4zy zEYj=SEJPMig8{oz<&Zep*mM!V3k7V%Z$3IYcy09}{%06-z&hj2e8&Xh=gfhgs2AEBHirZ=%>Z=l{1>gq& zoQGU&e?uS~UUkMjS_f4ns2596bSqju?tRYzEDao8%OD_o13=X9&;9#n$VW@GL_djocizdfHwq}F>=7fZs-Gm#`4j|c9Z{Qr@O#>c%QBl#X zuQUN(!7v;BzO=NoR#5!m_Z%)+T@+S1ZcK_jE#wFdR-JW&?7O~NDeCt=P#ghLd^sIx zS1FcWIV)L09yL-5x$xz88uaCuZo6P0ko-?q5gkw3otF}#ER6$V1@nuGsRSnYcPouM z-C`vWbV9%b0|Sx*ZUfcTEY>r!FMf`Wj({Z{n+W_}+L!RsGdFJE0C@7K8OT~{z+tAb z>;W>@>1N+k=T=xx&<*SSi{|I|(ga9kPK~n@QTZY<1+nPFaq;m~qJ99Y*SILd6O1we zw36&q`1)5*P=y-A%LCj;@Gw|+o{^poBd`MOsSF73(y&Xk;a~8ib3m37eKIlE>%|WI zvk{KL4B#S{%A+q9JM)J=`T$iN+iGf696s>--u{4pLlYBKNnFdI(-;|=g}c4|eI#k3 zM@babNkva;35g=oHj+gPl(H{h8_*sO0cTtJAL1o-mbJ4>D8Ph23wv*K!OTS?V^@sf zjV@1-q6G*ra^{Hooh$>)dTn)8W{lf=Ukyq}<1(^-LKsBG!NEb}Bf~`x2qFFJZR02re-DV#p`%{lMo*kEv(ALrt_LsD&X<^BE zo7E^9bF0)CtGR5W&hYEmW*-@^+k^NPs?L(+>-}HADvt4Ox}C!FJzk8JuoV&#@-y$R z{{8Uu1i&$x$R329+JD#xH2f-qTGUNc`Ps^OC(GhAOFc6)6Ikip^MiL(zclawGg&~u ztSy34rwSbg9D+(NJ%jDdiZwiOMy*akoth`4UAJ(Wdr zWwgET&DP^n8y1L`mX>wwy?A10S@lR*NLL&eX{Qc+8xDVKu5<|3cad80A z%~YcQ^0hK`0=^ndN=ccz)2?E9dV{>Ox~hYGw%#9VEZC}EAPA3y@k{KZ$N%;MsL@>1 zn}M+M20}lk92jJi;#~&L)er*z-eZl&jr}Cp=vgJB%HI9R76}Hk~Rs4aR>cq4%}L^wW5m)`-wwV zRu<#{pOk<4T`W4_7@eM?s#>0Ay;$&l!W;##X~Zx*JiL@7BqSsVh=v}2cV-l3zz6(y zaX}&x5&#?mU*deO52Wu&lHnI%oEx(yR#x&@3A6x~pGqaEu?*~*zib>vVpx!*PZe45hSpes z$l~GS1Kob4o{4hA$N!|7!-)$XnyYn$V7Gt+r%D@Rmom_jvH1|1n!nW}kts4Ux08)2F# zS*6=c;Y3rMY%Ru;lc(#q!1#KUMIx1-6?vVJ&vthaKDO;{6L6oNooQ&HzQ*J6-{^{_ z)m2v~?;R$6|BlNvvP)q`$>$Tsk%}hUc`+XVIX}~+K+5GmxBF5$Rg(<{-JdG@f#90rHh@No>yvDm+&+_l)nMN)aaUQd`n z$l@5D!juxx!&%btf#LX|8m2W{nMdFhWBWopXO;g1n-MK%E6iZs*7V}y;?JKNe#H3! zKwtw&l%dMt+BvdDWV|AaQaWp7R_#Z3p2qB{&yZLeXQdyxivmuxFNsaixrvIQ>bp+W z7?obg&ek5ck}2Uc{m&WtQ09mrHt8%CkpAA^o{%9i+;76&3MNq`7-1!Mkf@eyUN__R z=H`#4XI}w!6BLwN{ZdP6w(vAx!*RZ9KcV7A;zHW!ER6a4O%PKO@r@@Z9pM1$D z|Bw-+eve7f?liAQ2ebsGontlrJHI;ntggGT&02kN)V>JP-opu`o5ruKKyfQp)nB4M z6~kz{gQ1a&KSys|dd+!eVciz9RBVGCAY=3H`nt*6O%Bh#NntQE0A*r_dX# zi_tdYQQAK6Cl1XD%ys#5#%fz{)5h$A19xtIc7bJ8LrX2x>Tb@8R2hK(u%5zbMEwnvc6)o!E3{2WX;BST65E- zSm0wi^9SBxXSpV+#Ih=QK{gh}_;RC`NE~jPxJGQ0}!aoJ!bkUFw|U$^U%g@p2FS6KO>>Qi_km(fw1#%B+s_@~q!c zrZ6Vi@*thJmzPG7$m9KFG!&`?A1cell*3aNt_uQPeHNB@e=jVyJ(ZQ*<40L;CMY0a z`!IzFfp7tJ5=_$frB5@Id*L5i%svyC8+l%33jpOmPu;GD2GdiQ(*v$+u~1$q^DP(^ zoTpp&@O!A4j!sme>D*=?Qk&C4c93+MM4>`jer{f#C-lP2*Y{>WwZ2lV$koEa;Bt;~ zJ-o%wRkc(po4xS+j)&7cw|Ll}i?jb5ReT50B@rwzE@E7=Pn}n~9abXdN;aU}py_>Non<`q3$VIJkO~$b(L=-~)|F zp;wGC-~X-%Jvg2$PYV2)wtPM08xMW;w}?9kxrd>ogs;#jD$FVbRMfs|&df0sDkA9o z@cnBwP_3JgIC#WlgG?a|5Qhx?S9iDx&bsLzyc;?kcol(jM?M}~{`Cn7^RU@rM+GQo zSbPj0L^ODO4UH+ceMzMdxjW^{B^C6@TN2BK?ykDttd;}$Nl-9_DYbaB>Z9po*1`J9 zvB;{(ZhTfz-@Z-4)fKC)!#ABTU&5AsJ!+J|+_uk~A7vLunEKb6-}Do|RDB-T*8cJR zyM({z222`ac#yb@)zZI4Z?@ljcox52#w*VyB^D~5J|E&^kwEiI`?}I*kzOB#4{4>d z*Y!zsb(5z=H`f-wDR})|#Wxmx5)bZlTkU=9>{5m2+&b_KH1239rwrU@ccDQ>#vC$tft z32S)Rg73Z0_H}J~Np!ql=&d6mM@PmVN!9r$l>1Jvhd9O*3VOWB4~FMPls_qsg+8LBiFxne|FV91x}oq|AD`LiJ0Dd&pr##uPV=O)n<>mRPe1Rw8N9ST zE-l^skt1(K)#!T_W}IDA)Zw*2dlcGb;^x*cxO6xF9B9@=95X&s zAj!cX@NN<3-u6=l{sBiWW9<^v`>(CapNcJhMZ6A|@Pz%{&$=7{3!gtRvitn4kTfw7 zp%C_|=<{Z3B)poGWSeO}esx~tj22its5<}J{K-=Oa(bBTC&W;-20ck2ieya5VD`J`r9}t_$)rL35)(Qs1^me(rAE+cX z{9yAs=5FYax6bu*Sry-YJo|d@A#VC982i=9>KF}{cwbh6?1_h8IBSe;)^p)~l$xjT zQbZ*U^Y7Fq;B(|>Un2dPO*!?Kg| z27op1`aSLr-1@c+|ErTPK`6CaD@L0j)=3ro?bN}T*x|ASl+UTCsKtJ6k}~ZWQBg&s zV+xJtozags;f2t50dKd5a}w#?NCP38C6VemobsnblpDdK@A+rGa~!!un1giX3E|jm z3O)R9>$Z@wed9Y2pQY0GD3p?5H>UUY_n&!v2C`*U_lK*^Y+srI zCOW?{rK=F|gUI`c6`JC~h5tF;uj*8A*&0?Xn^ zWupsc;kKAQtE^l2iS2ELZFHyGo_tBU=BOTwviS2}b-H3wG}kjGP18_YVd2Y} zxG_+6M&zmTb+$J&7UkzBF`@^%ATlym*xBuKnfFCppp{1h6$!Vse#+YlxNlo@JoaBI zC-M5?Fkd%a=;*ejt4gidA_%;GMG=0vpMd@9>aqmBb@XlXwUjhEBc|Q!<=Q<=ch|AZ zA!rl>njvUL37M+mKajj8pU7Lh!Sd}SYxUSC8~Fki3J?h0t+lnahO-xrVU<%dGpG29 z=IjPY^Vy&CvR=+173OA!Gy_u9StrsDUR_|j1wA-&DJ3G?$w8B&*l}q^3 z3|HeaWb7{+o3|W;gid7oCVVBm45}zA{1%+mI^w`w)1A^%**=GGRO6_V zZ-cIA3dz86LPyGr9U&{O4~~ zkcLzQ6Jl&o$J$YVr%GB-H>UFP-p!)I zAo5lzXqR&uBtk0x04+By9l8wtUqR0H%SFXvp(v26-D+z9tchZNY11)DnG^Yt%gXhK z9*A5!xC4Zs)}868?ZQm1S>UN!KAA}4c##9JxFZI}x~`a*UPacowup&L2X+Pm4nG4` zJ1LGQpwkRsp-rvr@eePLn6a+-A}cf!VfQ$#&Ue{7f01Mo@rf21?J*_il1nPJ#N*AA zf|6rKbQE&EeOwR+hR!yO&o+BdH6zVAgjt%0ki9am`dvSpSXpg8J)cw8iCvHw=-SLY z+MW<3bZA9KHNs!@WzQ4~2h-Lm?G5C(>^(Ti-;-rgn)zY?%3nE5kYfpUadx&yd^Enc zrr+^kqTLutPG0i-y!oQmRJYJXq0?pYc)k2at1WiE-CjsjXZ?bne?o--`i^b!o$~;f zN%L`0s^9M zsqa*%*s|U4ogUeZ16ga|SAg|mf;@O`tbHKbrsAam60wW|{tInhv$Pn0!>W#QO~94llO5Ouaua}y{RTayfmyb zoo{qk-w4Gg5^=no5bp`G@ve!RshQw0^VJP3=fC@2C%)8qK+JkHL7}38r)o@Ob{iK- zU6fk<{`xxa?`%+?=hc)sQ`+sxN*eVwrdce)3$Gsw@JCVeNJN*29Bh@vuUu;Ud0b3VCr}Uf5pJ zR#BgE$_Jys3sk z1%IZ{8jbQF`IsnJd~vT0J-f=O?knm{CST-jL*q)*Xjr#vvUOawwOPE`5v%3$pCww~ zc7qT?(^mM1QRU0^kE+WWr+7kF-4R%ts0;(EhQFhLvWQQ7Fu-K(^J?lp7kOFB(NGg1 z{YCGp-LjtEwfx~*yAz^^ioPS+roz<4N!BlF9{qBA=^mH^#%<^Lu@%|F7bile7_-ypzYZHT05Hh}_n7;mw zM5&ZAY+3~mM^7p3J6aE+^@>#a8Z$XF0u6Yw|D3y8@C-?iqaCn4)Dx#UcB`Do{so3S zhz9a=)l!QJ>l#n1S z`PzQPt)_}&S3zK!t&+}?oluU)gFnosudbX|HoXL3%|6NEuG10o5#kuGj4o7h2J%~I zXGk&%M+Q1@F&z{*w4}Xbv2VJPnU#61{w{}>fCx`^F6o10N=>dO$4yI8DT&!$qKLUL zLxUI6qdaf*zlW*m9*>-ck4`7Xkp}b(3Ex`gp!ME^am+ zes72SDj0QG1m7odC|>9)hN<~X=^Em)OHH0r9=83JHJQIY=nrwR9eXWIsp;_aPL-!4 z=RxjTFTEKM0}TplQQ-02ilnSwKItWMKE|xXJp6tNW6^Fz2pU_rsq5vFN-=UeOdJZj zefGN41G?yhkt8Wxf)PR`ZBy|=mt-oG`=y4K*^QeCh%Wj255b!M>&3Xbed-t-4xn%Q t|6zQ^hJkr|d)W2h0N&(7`Y%1h!w4P)&hUE1JORIfQI^+`Yml*s`X7qd#8vD_Cuwe~)VS{jO&XryQ`Fff?PN^&|dFtAg=&krbYz{qU>M+^oA z2S!;=8XWNXWCPiNV2KE38y&r8@OME(Nof=sGhQKmSbU-RAA z+%hiV31%=XqC5fu!C_&!%U_CI8(9LakC7`YSK(pDC)0X<`_o+qKi9Qc*xBh(Xku6k z(4i&F6H!-!=)$Rw&77QV6y2#`O?%%<)Qr;$&YiR{IO`_!KpVjq5}Jak7BbuDM9QL#L^FzZQqjk2S%oefqbo2n46>1f#lyx36$^F#DD5@3xv>7hhx_AyopLljhzTCeM6k<6f z^wm2d#so|5V0kYt8b-;T(0;2iJqybpnwof_Avvlz!@y#_&R=5*nEYi+2S2Z4o|!8# z0N3}@J`j-=)w&PW8=6CAY-_Yte~jpyhB{zLt3~Y)PIgWZ3&#bZ;nR!}^(GL*Mdcz4 zvt_uS%OdKAJ7pIX7@L@oKWKdX_)$FQn)9sXu+K*zPkMkQF6xp{&3(kvmhF{(@nYP-g4}`sZ{``5{vQne4w1^THcLxd$IN$7R zFzvyj6bm@`S@=DhLsdyBB5ypEf{tHHSNEq95$*`d`%ShC>nL`f(CXuVf5-36x6Tg_ zQFm?@>y6hsT>oA@lqsfdwmITYi-o_2{KyyBu{>UCe7ioJ8;V8G%FHY+E2Eju&(9wM zfpv8)4>$XQp{lG#t^N5y(vOTNl6)OOjixJ>=^2?!HcBbTx!tJ(jHC_aKrVn zvACM-+p~?Hz^nbPu5$jqxBI`?kx~|Rc1UySd5@2en*-rgLu_)!xpOtTwGi3(4AyT; zy2_~>)CoMmHU(D5fgQ-m);2f8ju-3A%*?Q`uzvphxxc?(E`i~Ml74`MgfuoXqUBvx zm;Jj+n@&d;49?1gt}c}%{m+?*6n`qz&~L}BY0$#_8u}zSuLfMOzP?_eUSMZ)U97EF zu5$=16|x{{J4K*m+TPYSmP%oK2`gr{K*68=iF3c&I>Z(SqfseUx}8T{SqTcMt6T3e z0E6fMj;HybZx(*~)avB`bK47Co^YZfUfQRb&7YS@_8CEFbyA6a&Ee#C~P9Vh1 zV|U_XU0od%3YC0&#(!-2^Y^c6=r^Nwr}8gf*1LUG^~&o_xX$es-9|c+w$ zUT($z_Tq(>$|q9?{roIEx3J**aJk3H$%#(P8S^kZzIS%k?|q;Fz31iOF>JOPt<`T7 z@jH2MJ@#Sp6Vu)C67~@1de}phD-ctwZHFf-t?+PgO18;CS=Ns6%~^&Xz54W9v{5bi z*66F(U?^_uV03czBV!F_z9aq!Np2tQgWf>3-Y>!$$K&k`^p(TXPR&-H4qi;VMx`_9 zp8FqvPtL5>RZY5mtxlGs#q||D&Dp-n>}KZBzF(GXKc4t~$5};c86eNz9I7`G(|+@wyjvl_ zl0Ek2&vbk{0A_aFS%X9L?aGYM(epWR(u z(*9K#bqBD*%39tgEz8Opm#;Z|a&4spNi>ISS6zRJI&GyhK_;_yvE>Co58YHagF1`I zR8rTt*!33^lWfQqmpm(r51G@D_P2bPGGIn0*Jtv#FX+{nYmjwAYG!|=P9ar!Y~^ZlVuXlQ7R>iKdwtg!KO3Me#2z}^1;FhUjo1&A>) zFhtFU1J4DjNJv13@R29F*;bjt)V!;gQi~ zVpJL60Dvuth=|fC#WPGRshOA*S+hEv7LusJYI)Q#tTxlRJPHa53O@)P8*U@o1@sSQyhGF|7ywYoe{O>k^3B={%Fz9Tw z>DG~x`fvb846;DaDknuE1u7`BAJW~kx&;5Y_^MBDUu2>j{KQ@`ti34kCn-iJ>NUBn zCbN(ck4UMWg5UVXMM~=iIDWgzCGJU0C9}<~xYFiVf9z`H=nrSNa!Z{4OSvhEIiM?q z*dOZ{sE>l*I_DC=tBde&Rtxa8d`c3#cFB@42bL}`oVT2hlb(o%)wEGH@IV%&`i&PO zSN$m3vYVRGg69K|2d21~fTgEunHxo_5R2;nRD{}ikAn`z{4e8cF*6K>h-qy1Bwm*5 zph_>m0gn*n*Z-?gaayJ-iQQ1dfAPc|Rp&n#1IvFdG?k|%C)0`qLcJ>KtKW%CI_4FXE=ENxvB~!fP4K^sO0$Tzv#QOH6(1D z#MeG4nBE-Y9KEitE{0@yY%0W`xx%m#le0UN;dO=?`S~(G;^XTL+EL7ERE3!C!3RH! zzw8%Fc=x)$ao)kXcXhWfce_$ZU&0KtSuvq_6!DFFzyB;y+u88u9gn7HyHon^;o)IV zGI>~AIPl0IjWX2DB)G+|T?heMzZtHR+~HyxeAK4to)#%F&iVCy52vBv-EHA9x4Et^ zp&K^UhN#Ejg&`lYufx_mnO%U5*fs{^f-`e9%VJO&*8io6FU^bnNv`}c5x#v5MHEzZU&E%fowo7 zbwWTzl~Yu-*g0e2TpQ=%Klx|j8&|O`(&GZtVx#@WdNZNg@-qL~LMxR`nYY~wXO;?l z!#(CZv6{+YbAxN|PBR*sZ}Dj-k?(84A&H~`DsN%3)%x6geDUe46)EUyvPO@OH*5iS z8Sc23tV-(YuFJDLQ&Up{PPzOp7Wg+3Q3RE3=~m}k!9%}WDAy@OY_7}3Bjd5o&H9}9 zdeKkFkT6LfKFNAV?IPGUCKRfvP5b!X=rx)KT`z~B#Ntel=ZiQwIfdOM-W&_1t1t1l zMzE9op1imal7TdJSv#+kR^0w5VNcJ?Qr>9qaH2QfU94W)2o|ExC{0FK?+TVRHSKdP zxMkQHYVZUlWN8IPs6AM(BU4bA-n($Wy*#0y^nhKm`cBq58*Tl`1e+L?!YqI+glaTb zAavFha7f+;y*u5)dbd}F`*&kK?DYsgP)uN@Ex%i29SuQPc$9FtdTu=IcKf;wOqf-g zlJNcMPQ;?nq=!PJ_uwI!DgF0mkAJIaPwKq24K>870{;c{Q>9j{> z(7*K&5h8$Ae8S8q*JUO^ESyb0^Shet)RQx?1qB6_RaGBO5P$q|UZ}MvBnkMw+O4g1 z!I$u57oEfS{Wr;bmzg4Qkq)h9-#_uIOm!jQ85)WS8xCLDO(Ic-Ih=Mib6P8I2CE=!r}*V28u2q$^Mj zA$CcVv4H`~imj0m%Q+T>Nm!V@(^5lief^bfJ3lkjFtQ0|s1^dz&fydgA<}szr`tGGd};p((7Tn^#LQZn+%NjzBkNc`-#h`vS%@& z&!cQKRYe;4MovZs_%d;lMOh0kHh`3j?9~6CT0B-EO_K~s?5*gvt@IK;f#uF&>w(5< zTcEGu`DT9~S&V9UxOsp`{P2T|qmskLZ(DD!`@+cyrSKQV-RYITIQHsJxg8EW2=X5= z$@Jp-LJPCA(F8D28sVR?nUo~{BP!LAzGZoQMC4*^$Bmy9cN(l)v&k0E>p+#9YMOas z!7q=5XML-(b|M~w9!sgFkRa+{tF*H|Sy^)#Wr7Ntf1gEcu>`_r2XSwX44=p)QLy)~ z1YBM7r}NvZ>iioW?Xak}UGEDVK?h-7K2mR%I0k*4^=nO@gqa{wKWK~Ruv?8p&ye%W zPtyWMbxPE$+r1&B?N2jZuGuybs_wf>(u{?+&O#5YQd0mx4{*-vUe{4g-noh2yq=bTsO0$ff zyo(m6KceNn*}DReOF#ly1OfDTXJ-e2rVGelJQIM5%H}kEJTd$g2!e%$rTwU+sfk_D z>AJR9sYz8Thm%2P5G=o-K#gT~64>FYD0OiBXk+w9k`U%*L;Pb?5} z@9yrF2)KwKNU1h@TW+>~dAx~sR8&+%Mn-nha*A07q(qGkULrj=_w3RVaRp88&wK$k zR#vnM$O85iG6r*Zy7K92J0>P3s!Uzn4xuhwvq8w8k;ESbLc=~bHuP$sVL1W!&1XO> zfwODs>(yCVnVF%}d3*@t<5FITJsnzz%H_Xrj+e$qMgoF^?d|P{<&GESmsFZNIyPJO z&SAeul9j&&iw9oHsjAjm4##!6ZM1Z_NjF7A?4sEFY)f z^6Q3n9xW{`FE6i!)oPpLMvwo@4-V2DDv>)#AGT}wgxY_atD2!&4A{F11wcea9|9cd zbiHeCdisNrjm6Z4z;;A^Y>%n?C>h_925&Tl|Hs;R60 z{qx6Zu}&EZ>$@f%d>it7@3}XX1?&ib-i(Zl7w*!#1%&ApE0LW-*e z{tbR`irmrtC2bNtWml$L-Ll-p( zfKvBAnr9Ie6$MBq6;<)bh@8W|Y_FXur)3N07m&XR3V zyDHa-h2Od?>?2=JywfZ9ZIW7aVl?4&9yR!c`2R|aBeM$Q=t8Wp)R}z_vNQC)-N>cbar%Byw2bqbTm7Nv1$tNeKZrSu5OxS3tI5g9tI$>S-$ALO z;TmddodKVOj*jkl!FtWM{Oa;jA&nA~TTVHgRNw>JcZsE}*t&O%o4({zAi^&al=jmJknney=9=&*x;hdH(Qu_AD+D5uIqT zt+|;e2CNfH{blGB2utceQWY3;fa->52w5;d*}FY<;1)nWhrbz)-gj^to zS1}`Pyi85$_ASL=!{_maum<=~%ui2GQ=0+@jMl~f@%oTWzd$yLHbT=5yR|mqw0bh$IG`gGex_G)+vX~ho z0o32$@1eR8OZ9zbMspG7O$|8DG&2kfHQ{fsTIDRlo7R9O*4Njk-(Z6J1Ixf?cLFzx z72D612Ypw1_7}GDY$}DQg>kF~%>?hyB5`L+%U^Ydd~xLDSiWXdR8$V8Lu07&FaU}= zxZsV#XXVu!cTNqlv24z$e}0b})SW;cb@^{+IG!diHXFw$wz);iY?QlKZy=3IQf56S z3~5=9CNv^s_UBKmqZ%#2m!prjx-0-R2?!9%9DR#M*45ReAuPS=h2K90aLvc(+Me%t zz8b{;Ro~Id%If<1I?QIbsRovfX2h^cKsvqNd_Y2)Fm|V%6eV`(V5V4>46}Q_(`5w? zl^Q$r>v|TiMDaN*4gR4^}6OhAYGkP^0o>joZTjKad}#9Zk(h2mYoq2 zteR&W8%Tsi$i2sMCEoQ5FmYTcsNe52zoNo!slk*+G~oVXhil*&Zi8B@9Sj1|eaOtq z6ArxG?e;wm3JlbK5k$z^CmUun=yaoaKJx;r23~B#)-^QU%A}tz){{wZ0hBJKgb$6? zFhnB!`Tie(_)!Aw_A^Bt9UU<%KZ_-?`*whkQ`OXr=XC~v5QTNAGhSc#=JwPWrA4zy zEYj=SEJPMig8{oz<&Zep*mM!V3k7V%Z$3IYcy09}{%06-z&hj2e8&Xh=gfhgs2AEBHirZ=%>Z=l{1>gq& zoQGU&e?uS~UUkMjS_f4ns2596bSqju?tRYzEDao8%OD_o13=X9&;9#n$VW@GL_djocizdfHwq}F>=7fZs-Gm#`4j|c9Z{Qr@O#>c%QBl#X zuQUN(!7v;BzO=NoR#5!m_Z%)+T@+S1ZcK_jE#wFdR-JW&?7O~NDeCt=P#ghLd^sIx zS1FcWIV)L09yL-5x$xz88uaCuZo6P0ko-?q5gkw3otF}#ER6$V1@nuGsRSnYcPouM z-C`vWbV9%b0|Sx*ZUfcTEY>r!FMf`Wj({Z{n+W_}+L!RsGdFJE0C@7K8OT~{z+tAb z>;W>@>1N+k=T=xx&<*SSi{|I|(ga9kPK~n@QTZY<1+nPFaq;m~qJ99Y*SILd6O1we zw36&q`1)5*P=y-A%LCj;@Gw|+o{^poBd`MOsSF73(y&Xk;a~8ib3m37eKIlE>%|WI zvk{KL4B#S{%A+q9JM)J=`T$iN+iGf696s>--u{4pLlYBKNnFdI(-;|=g}c4|eI#k3 zM@babNkva;35g=oHj+gPl(H{h8_*sO0cTtJAL1o-mbJ4>D8Ph23wv*K!OTS?V^@sf zjV@1-q6G*ra^{Hooh$>)dTn)8W{lf=Ukyq}<1(^-LKsBG!NEb}Bf~`x2qFFJZR02re-DV#p`%{lMo*kEv(ALrt_LsD&X<^BE zo7E^9bF0)CtGR5W&hYEmW*-@^+k^NPs?L(+>-}HADvt4Ox}C!FJzk8JuoV&#@-y$R z{{8Uu1i&$x$R329+JD#xH2f-qTGUNc`Ps^OC(GhAOFc6)6Ikip^MiL(zclawGg&~u ztSy34rwSbg9D+(NJ%jDdiZwiOMy*akoth`4UAJ(Wdr zWwgET&DP^n8y1L`mX>wwy?A10S@lR*NLL&eX{Qc+8xDVKu5<|3cad80A z%~YcQ^0hK`0=^ndN=ccz)2?E9dV{>Ox~hYGw%#9VEZC}EAPA3y@k{KZ$N%;MsL@>1 zn}M+M20}lk92jJi;#~&L)er*z-eZl&jr}Cp=vgJB%HI9R76}Hk~Rs4aR>cq4%}L^wW5m)`-wwV zRu<#{pOk<4T`W4_7@eM?s#>0Ay;$&l!W;##X~Zx*JiL@7BqSsVh=v}2cV-l3zz6(y zaX}&x5&#?mU*deO52Wu&lHnI%oEx(yR#x&@3A6x~pGqaEu?*~*zib>vVpx!*PZe45hSpes z$l~GS1Kob4o{4hA$N!|7!-)$XnyYn$V7Gt+r%D@Rmom_jvH1|1n!nW}kts4Ux08)2F# zS*6=c;Y3rMY%Ru;lc(#q!1#KUMIx1-6?vVJ&vthaKDO;{6L6oNooQ&HzQ*J6-{^{_ z)m2v~?;R$6|BlNvvP)q`$>$Tsk%}hUc`+XVIX}~+K+5GmxBF5$Rg(<{-JdG@f#90rHh@No>yvDm+&+_l)nMN)aaUQd`n z$l@5D!juxx!&%btf#LX|8m2W{nMdFhWBWopXO;g1n-MK%E6iZs*7V}y;?JKNe#H3! zKwtw&l%dMt+BvdDWV|AaQaWp7R_#Z3p2qB{&yZLeXQdyxivmuxFNsaixrvIQ>bp+W z7?obg&ek5ck}2Uc{m&WtQ09mrHt8%CkpAA^o{%9i+;76&3MNq`7-1!Mkf@eyUN__R z=H`#4XI}w!6BLwN{ZdP6w(vAx!*RZ9KcV7A;zHW!ER6a4O%PKO@r@@Z9pM1$D z|Bw-+eve7f?liAQ2ebsGontlrJHI;ntggGT&02kN)V>JP-opu`o5ruKKyfQp)nB4M z6~kz{gQ1a&KSys|dd+!eVciz9RBVGCAY=3H`nt*6O%Bh#NntQE0A*r_dX# zi_tdYQQAK6Cl1XD%ys#5#%fz{)5h$A19xtIc7bJ8LrX2x>Tb@8R2hK(u%5zbMEwnvc6)o!E3{2WX;BST65E- zSm0wi^9SBxXSpV+#Ih=QK{gh}_;RC`NE~jPxJGQ0}!aoJ!bkUFw|U$^U%g@p2FS6KO>>Qi_km(fw1#%B+s_@~q!c zrZ6Vi@*thJmzPG7$m9KFG!&`?A1cell*3aNt_uQPeHNB@e=jVyJ(ZQ*<40L;CMY0a z`!IzFfp7tJ5=_$frB5@Id*L5i%svyC8+l%33jpOmPu;GD2GdiQ(*v$+u~1$q^DP(^ zoTpp&@O!A4j!sme>D*=?Qk&C4c93+MM4>`jer{f#C-lP2*Y{>WwZ2lV$koEa;Bt;~ zJ-o%wRkc(po4xS+j)&7cw|Ll}i?jb5ReT50B@rwzE@E7=Pn}n~9abXdN;aU}py_>Non<`q3$VIJkO~$b(L=-~)|F zp;wGC-~X-%Jvg2$PYV2)wtPM08xMW;w}?9kxrd>ogs;#jD$FVbRMfs|&df0sDkA9o z@cnBwP_3JgIC#WlgG?a|5Qhx?S9iDx&bsLzyc;?kcol(jM?M}~{`Cn7^RU@rM+GQo zSbPj0L^ODO4UH+ceMzMdxjW^{B^C6@TN2BK?ykDttd;}$Nl-9_DYbaB>Z9po*1`J9 zvB;{(ZhTfz-@Z-4)fKC)!#ABTU&5AsJ!+J|+_uk~A7vLunEKb6-}Do|RDB-T*8cJR zyM({z222`ac#yb@)zZI4Z?@ljcox52#w*VyB^D~5J|E&^kwEiI`?}I*kzOB#4{4>d z*Y!zsb(5z=H`f-wDR})|#Wxmx5)bZlTkU=9>{5m2+&b_KH1239rwrU@ccDQ>#vC$tft z32S)Rg73Z0_H}J~Np!ql=&d6mM@PmVN!9r$l>1Jvhd9O*3VOWB4~FMPls_qsg+8LBiFxne|FV91x}oq|AD`LiJ0Dd&pr##uPV=O)n<>mRPe1Rw8N9ST zE-l^skt1(K)#!T_W}IDA)Zw*2dlcGb;^x*cxO6xF9B9@=95X&s zAj!cX@NN<3-u6=l{sBiWW9<^v`>(CapNcJhMZ6A|@Pz%{&$=7{3!gtRvitn4kTfw7 zp%C_|=<{Z3B)poGWSeO}esx~tj22its5<}J{K-=Oa(bBTC&W;-20ck2ieya5VD`J`r9}t_$)rL35)(Qs1^me(rAE+cX z{9yAs=5FYax6bu*Sry-YJo|d@A#VC982i=9>KF}{cwbh6?1_h8IBSe;)^p)~l$xjT zQbZ*U^Y7Fq;B(|>Un2dPO*!?Kg| z27op1`aSLr-1@c+|ErTPK`6CaD@L0j)=3ro?bN}T*x|ASl+UTCsKtJ6k}~ZWQBg&s zV+xJtozags;f2t50dKd5a}w#?NCP38C6VemobsnblpDdK@A+rGa~!!un1giX3E|jm z3O)R9>$Z@wed9Y2pQY0GD3p?5H>UUY_n&!v2C`*U_lK*^Y+srI zCOW?{rK=F|gUI`c6`JC~h5tF;uj*8A*&0?Xn^ zWupsc;kKAQtE^l2iS2ELZFHyGo_tBU=BOTwviS2}b-H3wG}kjGP18_YVd2Y} zxG_+6M&zmTb+$J&7UkzBF`@^%ATlym*xBuKnfFCppp{1h6$!Vse#+YlxNlo@JoaBI zC-M5?Fkd%a=;*ejt4gidA_%;GMG=0vpMd@9>aqmBb@XlXwUjhEBc|Q!<=Q<=ch|AZ zA!rl>njvUL37M+mKajj8pU7Lh!Sd}SYxUSC8~Fki3J?h0t+lnahO-xrVU<%dGpG29 z=IjPY^Vy&CvR=+173OA!Gy_u9StrsDUR_|j1wA-&DJ3G?$w8B&*l}q^3 z3|HeaWb7{+o3|W;gid7oCVVBm45}zA{1%+mI^w`w)1A^%**=GGRO6_V zZ-cIA3dz86LPyGr9U&{O4~~ zkcLzQ6Jl&o$J$YVr%GB-H>UFP-p!)I zAo5lzXqR&uBtk0x04+By9l8wtUqR0H%SFXvp(v26-D+z9tchZNY11)DnG^Yt%gXhK z9*A5!xC4Zs)}868?ZQm1S>UN!KAA}4c##9JxFZI}x~`a*UPacowup&L2X+Pm4nG4` zJ1LGQpwkRsp-rvr@eePLn6a+-A}cf!VfQ$#&Ue{7f01Mo@rf21?J*_il1nPJ#N*AA zf|6rKbQE&EeOwR+hR!yO&o+BdH6zVAgjt%0ki9am`dvSpSXpg8J)cw8iCvHw=-SLY z+MW<3bZA9KHNs!@WzQ4~2h-Lm?G5C(>^(Ti-;-rgn)zY?%3nE5kYfpUadx&yd^Enc zrr+^kqTLutPG0i-y!oQmRJYJXq0?pYc)k2at1WiE-CjsjXZ?bne?o--`i^b!o$~;f zN%L`0s^9M zsqa*%*s|U4ogUeZ16ga|SAg|mf;@O`tbHKbrsAam60wW|{tInhv$Pn0!>W#QO~94llO5Ouaua}y{RTayfmyb zoo{qk-w4Gg5^=no5bp`G@ve!RshQw0^VJP3=fC@2C%)8qK+JkHL7}38r)o@Ob{iK- zU6fk<{`xxa?`%+?=hc)sQ`+sxN*eVwrdce)3$Gsw@JCVeNJN*29Bh@vuUu;Ud0b3VCr}Uf5pJ zR#BgE$_Jys3sk z1%IZ{8jbQF`IsnJd~vT0J-f=O?knm{CST-jL*q)*Xjr#vvUOawwOPE`5v%3$pCww~ zc7qT?(^mM1QRU0^kE+WWr+7kF-4R%ts0;(EhQFhLvWQQ7Fu-K(^J?lp7kOFB(NGg1 z{YCGp-LjtEwfx~*yAz^^ioPS+roz<4N!BlF9{qBA=^mH^#%<^Lu@%|F7bile7_-ypzYZHT05Hh}_n7;mw zM5&ZAY+3~mM^7p3J6aE+^@>#a8Z$XF0u6Yw|D3y8@C-?iqaCn4)Dx#UcB`Do{so3S zhz9a=)l!QJ>l#n1S z`PzQPt)_}&S3zK!t&+}?oluU)gFnosudbX|HoXL3%|6NEuG10o5#kuGj4o7h2J%~I zXGk&%M+Q1@F&z{*w4}Xbv2VJPnU#61{w{}>fCx`^F6o10N=>dO$4yI8DT&!$qKLUL zLxUI6qdaf*zlW*m9*>-ck4`7Xkp}b(3Ex`gp!ME^am+ zes72SDj0QG1m7odC|>9)hN<~X=^Em)OHH0r9=83JHJQIY=nrwR9eXWIsp;_aPL-!4 z=RxjTFTEKM0}TplQQ-02ilnSwKItWMKE|xXJp6tNW6^Fz2pU_rsq5vFN-=UeOdJZj zefGN41G?yhkt8Wxf)PR`ZBy|=mt-oG`=y4K*^QeCh%Wj255b!M>&3Xbed-t-4xn%Q t|6zQ^hJkr|d)W2h0N&(7`Y%1h!w4P)&hUE1JORIfQI^+`Yml*s`X7qd#8vITzGRg+OjYUdcXJcT3oqcK0$cO_biTjWe^1t6XjTd9${7m|fwc<^`Rf zJb%fNw*`J`{vgK1*EFE|{O^lsR&z#Yb>mylp3&=+Z=fz*e4mxkshQ$Ak*O3>X?eJC ze*U0A>7PIMH(wzp>M=zO{9Eq5hYhP8Jq^kDgyoLc8ErFl6U{3t362^@A- zZOr7e%41G1tQ1(YQd2b=z3D8h6w60j$;HXf?ci`Kiuy@Ya z)^F%(cJCulv^|%<9wLCfR-0o5%nP1&(<(P^9);FIPc;?7MU5F%6QHp6qaao6DJt9k zl7O2VLR9NYY>A78LB75&-MwphCUfhNsx^hFJ)6Fv-63SskA)5=eBLY^4B{yvklOHm z%9R0;LR)`^^V0i<1U+{lk#b}Wvvkd)Ln%9B#kj{092`R4U812`%mbU#i!5<_Zpvfu zX4tCb+titqeZRbK+jA(>eJ?V|2R?QuGLL+7C~ruUWO=3rzW;lT{p@!c*fxPe4$;^Q zsa%BiMWgq_3IYO!4|pf`PY(h(l$n_DtWU*Wn1Qtf+^D@Esoc?$>$WqpSb~IieK1}a zJBx9d+KD;-c!4SRx{FDUe-*LuPEvB&Q%&AH%Mz=hYNY)JXM5X@fIAZx6$IiU`9d@W zojpFrsB&g0SARHvO265PUO@S&+@hhqe{J&i6{1_J#QEv+pX`~j7(`cRpxCVc9s98> z92}e~RiY5cm@2q|`~w|rx+6m0pX{888-Fb@-~u?LrMT{^@OB!T7F*{Q!2`|NxI(m& zbXjuDWE$zH)>yI(sy(l>6pR8a*tnd^^39%K(vaRKv4JbDiFs4p+4vS=SKzyrakt3I zYiq4xR6^Zrlao4))k>*xo!x;!dDIWf&w|=a;Nqz-h-d9XqZ#ub4Y5iCw+m-a>5Xpaj0Dd{LzD;ub9F{1mpLN^)_qL&~adV%$rRLe>Pw$6rIw988Q zrbWCI4HF!;su_iZI6p@|-_|giyy;4GvH!u7t6UYrJj$=!WZ^1!i45|L^UCxGgqFeh z-&F>6k47Tlv58JZYUV85C>9sx|iQO&c_Xt9Y~G z$G~Vcm%;TFsr6D@$BQl6MvaT~l`v%}9p1<7AM(7vbpMg-t|_~_B(3!w!7Ns7n{a~K z6)-MN%7#GtX@I#WZS+U~t~y_?Q>{TL%(_f*%A+Z$gtf(SvVBsz%s!qLN**=D4;wKD zi&q5f{AYouRH7if&#JNu5P{;z3Oj|_C_4K;Pf&k}uKddEW28YNO`JU3vA*rq`%MgT zJ?-t+H35c;Q>3!=FEX#a-9=+~0t1Eh+U>+wH|EEziO-qxudetLGO0r&q{T%V>XC5o zvt7}k^hRZQcd$8Ta9u`R9Q#%c(mr2THv%D!7_s=svLSnTHrp&VmoKoo_-9er9zz+F zp7Zdt+y%%pYECDz+9B*wzQ5_VdPcOgQ%mu2^Bu&)uZca4YC-e<5%NEyqZK7WiR4JOrqssfb_?CACHNV_U{~0b<_&_3XXHQc}_Q#shnOC!?|+pia5w1ENWphf6@<%j*0@oN0{<2gA2$>7nm>Sht|er`ujh)H6$wKbmo6u z>!MTiDmv7^<+(Qpds?FiosZ{FCl*f2_edxVlK4_(AflG5eGdz zfvr-N_bif$`f@{>!r)B^0}Nw&Ql}>+%b>DS70LT?Cfm*A`Kf(J9ZOZ| zz5lVmL!Ih-PAr0{KIGgNJksXPimr8G3VMyf|GzN-pd17!q{%bBcgJJH4S>?0vE55F zI`M2px|*yZ2RYsJ7XcyZzODEyE3zOzWXKb?!vIp#$)!Df;afvPwxY&Us~s;Uqi;Jk)U!P%n3mqlfjrjJJfpqZ$FK#X_ z6I37_??ihTG#MEhZ*J_34G|^1MY1;;`E`E*SbJNN=N8jorc#hx5WSqUvva;dUFGm@ zS7+z^!Kjh#{uGCNXz(2wU#DEdiRg$?PIAhQ0B6&K?^b~i^fXvF+AF;a{sLk54}i~! z+R|Da_E)Q;pms-LUVXg;qTyiVczPRI)dN?NI$4MjfW5rsv6A{dA|e9Ov=nD;(wp#P zbL%LB-#e%0&>q{4ns&uZFV*_``nsL%qOqCrh`>=pPk4L>lmbQ$TCVZ8fmLDkFsGV+e^vP_-5DL$ zGGa}}Pc5hY{Q17F&wEfw6xDRT-c-p_Ehn2`-_ArsW5&JtJg!M9-a8WB%N)9dT&2k~rfvmXuK4AAc{^#uL-^JjCyb!)1I6p!*;!zcC)Y~Hv5 zQv6ia_W09q%?Li|pLPi0JkT<}cl_(T))VY$7r>p@@<(Z+w(#Ep|2wD%h}u8dq~uP z4OOpZ#?MFTAuOGPtqW`JQ%RrgxN!Q6MFsIUib?Iq_Q|@qxX9YXb-yF|KuMT4PeV_b z&l%Q#%*lE^m`PA~L)!AsPbY#?!#P=HuzGN4V|xh{A}3qitM`s7iMK6xb^^zpQr+SF*xZs3GnvW7?DOwoBKepW8}dJ)hck&zX3-bf_UYe_J9wKH7z0T)*wXR=qR z%w!S>rCmSivs8`khVbxkWj}XHbDlSLYOAXL3h2a`jqFi3GSI%j zNB$h4{@*#|?`zzjejv`tT{`2Lm6}CLl_8S~ACt;S~i&O;Wue1gdd= z6soLkDmMq3e29oA)E#+wa3pta)h$wB&0HW$9PxVDQv39Cy6aaE-)JcD7=1j3mZ}`y zBq6mJxNEIxebw69+VR9_Qnc}Gksl4PkO{b?O7y5)->}4w zzMg|964an#fPV=mfA61NF)%;gncHpH_T~r3n@|m69s7#{YU5b9x3?!oJln_mvLO5= zkEY7H)@t@TxnVQ)?gbM;;5pZfj$d@g2ymR@ESd82wM*EQpDh01%L_;6K)St(BFVPWlBZuN4K=lmC1Wedmn+}zyb@jm2;sW6X0 zt|n0^|pq`r@!lMZit^S&qM4bec#>tyx9W<*h%~ezjdX ziKs)7P`n9jdGglwCW^r*_+)?C_1H~GiM2`ic*b)ZJ}iQmsy_Zjjr{w`lP9cF%9+9< z%%6+JHS)*zPTvUexN_@}@}fNHEcdDVdZg6{B{P-c>Hoc@ewFdBPgKyXw6wI=Nhtzu zuwa6(diE3JBjaa&VFR@Jx)of5_i$W`dDeqrLB3Z>iOGNz!NS|CBWE(g)8+tP?JsKOVBqNYjEpbSpG8G~~P+^wR1S_TL~=`hi~=0{sl^^NvTY zb4FlbV0OW@OU$??e;J@*B2}#7E;f0Uy;Fe`BQ>RERgy19!I6*V7I5_OULm8PvMFn5 z7}t7bxIfu<;6w&VsRC>&QN+Rmv9JJNlT&ZZlv=YEml9!|bOFxaW|@^pPc3z9TnGE6 z0VvsipEM@dduR3+93gE#-=+z#nsQJW8g}LpA~aJ-_2+Kd`r!5ICbQn=RG*)&W7qPE z5EWY?qBH)02#6fM)=ZZ2HUY$@V8XU;vyx9*lm<{OZM@-1)U;?<3@K?mB_)m0Qug+Z z-I-_SIsT^C@*eB;^SwV<8{QpKQ<1Z@wA6PQH{D%T@p%fof!2B?|CMc-&3I|`LbyU& zd3pKp`Cyic?S3`E6A2n}rH*^_=@+1vv{FPuH}YklwWgwY5*#kjHb! zgl!)Tuk!(U7lVe}T7yKRn^%(s7GVTx+>C=Brwh%!Ic_hW^B^5zva&or=CEK))}vOs zO=eQ6r|+_DBDi?1>tSB=Qv*KK+0`{U<=W;?13&yH$Y1PqB|W{mS}cm+XeP}Y#1!Be z=0DA43UViIq?`{*6Vx_J&}=j`_JAa{JC7QSnt02$X*Hb=GfE@9HUOVd1-ug^QRTV= zVOKyB+g%&?6GhMl3V*Vs4st}y4I7STJh8yd)#dr!4_!QSMoLjgM+*SoGjx!}$Cce6Lo(G6p85EJ)KE>Tt|SuPceFt|mG4 z^u&u6si7wf8T9R*-}2ljVV!AoANzfnQ(zTH7nH}r&Yq{z7Bl^CjQm@1WPplz`eO7kcYa9D}s+0VPXVLlQdmgh(GY(sL?vJhpBXJ==caF9VK zr)>4mXLHm(jM4fI9U@h{e0&I3a_y&EJ??bhU&B4>Cx4!QPlP=KwUGxF#9J$qg_|~8`Y)!Q_Em8bh zUAJJL1TCpTjAoUNd|lyL_dcAIaj#8#d;owBB7ve7cWc^N$-trG-ZXP<#@l#93Xw^3 zt4SEyU!aCC-eETOQyg=BhJzMc>X%S1Qr=hpV7O7>zHVZm<1H(D^ZM`50O(3>k{2-0 zAPCTnZ&KdunLUhbJ~f`Eva%BRowc!}<}`oA#-@l_cT9HnN^NlwVDP17Wp(8Yx(Ebf zVQI-_e0R0FrKN?08bTr>F5^#X*W=bgXPhi^{Fm8u5)CDA>8&mGHvuS{-G2*l(N=?T zAstbAs%jMve~O(+k=Ye}$I8u+bIQBarm3z_Fe3k#&Sq zoEyk)|9G5qNEi5L7OvmDhoxq z&SUecN+^o++)e_7RZEr{N~W%s!PdS{O9u&nB?$rjjmF!tY<^|%Wkz(er<0Pp=8!o_ z8v?IrL2BD8<6)=A4;sL|Z-ZNZBB}JjAQ1l3<4WqZfC7|~C+Tt&XpxT*lGs?8#-nB^ zsQz<0K_K49l0+{_*jybkJ6LqNpv&g@vuJn6nOe6aXe15I_eeVXsr-4MUO^y|e*KjH z41w;1{j7YYV(!jw_&wE%#}T z%{^U%T^z1mVQPIW3_3I5LC@;y4ajTtU6eJ?2IV|u2Z|G%CJCavVJCf2=EYno97#Sl z&Qm_vsDL-*R2>0>{1Y6dj*1DOiKZ732^x^Ge1EO5a&fO=E3Fw`F4p=wP2#Y=D8EqX zcks+Zi9Nr?-7tnq+-qLOZ>-7ZYhNDlbdGl#g2#RV9UU{4QPH-BAJs{ zua;qdI<~iu1rVxAIr}wU(`%v{M0o-x)-s(YNs2Y;+!OQ-R4+hM1c8a$f5x#Vc^GTd zAN!cDR_5pXucE1gEP>L>lf0X%Cj@lRRX6**K+d&|wOjG5lllDPkibA1L?pW9y&E z{%_wUtQd$f=Q!ygSWjmyZ-`F(eI2ng7)QvxOIPxAF@HO5@ddapX;oT4K|e+6ybo*g z`*;0y4Y2h(AQbnfpKX5s<90YF%sb)WP&N-xq8+sE=o+8@Wr!^0NpjCFm)rgW_5>Pm zcd5kf-{X4hZ9)3mcmS@&g3f1?#wCnWBmlw={>M9{d^~57aM>#bDsD3kU8SI9q zU2##8uVI6mQ)q7o#q0bo;^6wbpkBji7d2?ps7zHo2E9Su@}rJW#wOc$8-f)N?kltX zJ??$b`SEaQSO|4y-i6KJ>@>li`qHnTE}5a;aR!stGt_qPzfN|4Iw<;3+$xbMg|MQK zxZHNlbw`IXlRro@6U^rzlcR>Kf^Y8NBn~Qa+>QDs_uSf#UBZ3AAZ7 zi+mN-*>WfL1U*u|R#2t`5P(s<ld9Xjh?ny!xs8>eg@mMzfs)$LDEyqCj?y_q)nNLj_3PYp8&miX zAdpsMI1h~VpKrGS5Dg}yVz~ipXK;co0#~X72$84>^X8gy_m-(Ye<8=??Fk#Qz4zhs z%=_Q|35?=7j}FJO9-2i3LF`0sdv+c?xOPH) zXjE9vKxwd&`XbYik1^)KgYHn|SMPKq;>LpqQ%-ugZ>sJBUg}97Ku632xDg~k>};ns zJu~w@JG-IRF6z1#88}uQa(B|e!iO&;g%KF>-3#C~5Xe6VN$8@!R&$^Gz9chexS0P8HCIKe4C>wxeVTO5pH)PGtMKYy2VJ-ai~sG zeF2MZ3k99G{=v)Bkg;5yGP4Zmw#Ckf7g5D~O-`Ct$Q#8exNCM040Y`-tezmijZC3^{<=yQJZ7r+_g=u9An zPO9tau>j4&3TU2Yb)I#?+lz<&GSqUQ2xm|S9Ijt*WtF^H1c)Cfb7}ne=1kMck@#|g zV`wvgsIf%=Tr1xOmBo3A!$9^4ivOt7f_V!v(V1pHigFv;nckNgED_cPBg%?Ck7(Dp5JPv`H#o`+UwFaqFW+ zEL@jJ9ac&3xMI)(5d$mu(=~ZIlj;6Q)Ov*HBT25a2IgA%@$X`;Gxp~&Demv@ll0J) zmRw+H4?vS4iElw8BKaJbHQ2Rz;NnRP&XALnGcW{KKT~<9C<=L60O}f&x~N@+FsOso z!4Xg)i#Od#l9L4);4?*Oz+#D-;2xwBT~_wJscFU9=7@wpet}c^A*kUX9;{%AV|+kk zvv0Y=DGOfnoa1GxK<(|e^NWJis$*X%7Z)|BZl_Cy~po0t0WvJ9US;Z$#cJG-inIJAtUQ~$;0ij%)u>D8-9H9|( zxY4MiDv|0FJXK%>3fBPS*YWR(o0G2~kcpA8n83Bp_Uwr_(2TBJ*tj@LdP(Zyb+EbtYgbQ>W1FmokJXXva{xAL^t0^*&<_u-!nz zll3*XRpQ9c=rX(+Dy#AF9p5m4eL={^)h~$c9MG^9CEyZqqM_s7WhU*MXu!X}G8JA2 z!8%;($_49X?fPSMN)IY>$404vf=o#9wMXyQJ(s35{-gyFJC?WjNJ4=^NTN1lRzj8h z`=e^YC+<@e zjIz}L!xfO*R~n?86olm@9PnUHB?(U*EP2QpLz-b(fxBR=pKUldN4=_JtFxB-8TVNG zZ%A>ghxbb5N>AqlBLgT+e7n6kZtOc#;boxnuY3y&^>3_+pGhj>&I@xj!W4%{g2>Jv z^+VV;)k+bLKXCZUHJy_eTmrqXgLRB3)!FZU+}>nSVVAKmq&akLI#rL$g2*_ z3iDz{5KDC(YJNYV{4qwW^lFf?fP=OEg0y@vekj@y@sK%+)}2r)s1K%R8g!HmlNlu> zL_zPwIvxxw!oiRRq*AMhYMN_wI19=^+1E$v(-*%QD6Yq0I}a$C@-IThXd0e6P4@FR zaX%R90wQ=fH)FMF@LcD?f!_EH$V!KY^NMCessKR@a}mHG4sz+M-b}P62rP(rfgvkU zS>V#-kPnKM=auCp1M%vRF&(hJe`#Cor~l2@0AT(vtMwKH+Arkp+*gL=klKj}oWr}h z4`V*W&!pL>{&*7i9a|L!l3tVd{yaTgPVO@{Nm?pAH4wN}?| z+91YRfE$CcqI?^rDfiI=``#M!wh&Q+Ys3unw~F1N!LY1JE2~^V z(ySF^;|%BY*$^?1=Bj&ha;2c1E>Z9uDv3Yx--eX~5)10#Qy9l7UPVYkh;D<&= z+RvK$LPm{<@;{=ZQNEPxK$4v+nM7bsrVyTOVBiA${f<>?rq-FR#9=s)9^R9zMjGrn z>%0c!*mxl624?3L2446|W-^p5An>l`rxv^CLGxh)>%@g=tA;Jfv&rKQ3pnKsTrl>xMe!HwSHho$vyyDix{1+Vyzy0#gx?VAI0%YzYx_<{5w zSeyTM-a}S>p`RL)mFk+B?R()Fyc53H*_d1X!d8atYg_ihP5(WUTn+B@dN26U@TlBC z$}a(nH>!$pAkJCO$)t)f&)aR{arK5HL|Y!6;ZO)*i~r09sd` zH^atxuU@+*Zs2NQV9@P(u&7;AdRH_5LBO!}YuIe_a{)6a`uy_84g&ME!ZYw8_Dm25DD}LKb#68NxyYwh5j$aol5DSTcyyrPN8j zg_0(TTcc-Z7%;eDl{eWVv50d!iDUpxN(pxMC|9t1wro9*X zuG9SUA)-c;W|C;AtAp2h8b4`41&E^sBWMtmaO}CdScF`Rw_ocPOiN9@gE%Kp+7I@O z+By>nhvK*S2!|SpS1CC}zG}Np?c%E!M=)^5$+=efwrKwIm$#5(s*_WPhvGqqWOMGlfJT4+moo=7w3GRo}yOZnZd2*Ax z=)_x?(hA2?K93 z9?p&2g^r#-Nb)|+p@I{$q=_XOlWW5nCACf(b*Xw_0_kicZX2C|W?Qo}pq73tettS+ zG(XSpEg34kwLs+q^FgyVg~6mfOuVr1Mu7n2@FG>oz?l<#esN+;6`G)fSK*oZ>MG2wrjyMIy=k+hhgntGV7R50U9*U9 zbLR;cS0^@08jqAX`%T#0v^8K#mk(KWN)*o(FN!QY+;CJuYiJq$NnQU5N(LkjyN&7* z#q=Ype^^Q$)FvNvIL>2y+}wjIn~2_qJUXS>Y7Iwh$oDRJm95m`rLB0*9H0e(%ecjF zUWOkwe?CHyQO__$8#-v*!Vl0s_3i}-QN6aNQ03l>4HM59L@Q5(j(DIGtJ-{sr_xWo z8cFxn->E5^&sRxgG51+Yp4=>$1bB)$J=B_ytl@WAsqT{|DoPN~1nq~uquX8iiO>BG zYi3(ioL7WNkz06jLOmbmO1y)_eYO_I;YSm}6vSBpAGaQ@mf^uB0EaD+Do_ut4GDuD z7;ErqWkeoGenrGd2Ju_f6{Z_w_Pcc~h%hp$n2DuR=XI`mYvOjA^4ACPa4Pn~>OH;h zuQQ&OlhU32UU&La-2KKA_rlZjjFMD`R*@?2om6QQC{3@iQnAH(p00A5jOb*wSr_$e z43j-8Sy$Vr54DBFO>y^`iO6!iFS9UAN>+#cD z6H`S<0u+qu#EI`ThaS8MSvj+A0}sLO8;Nm7Me|IqxlDUcJ0D=yKra@pMa0Oll6q1e zT&a`FTeU1mrmiSnOUG#ec_@Q<0#Y;KTqQZM1ct8g4DM>&Cyc?uuD}be!Nd`F)6u4j zEe#JB^DW@cZ1Sw`HXCI}3od=T3r55nVG<5R&7_Wz%|`sO*m-pXHk%_1epI8bp@9M| z-I2qdRRSM>!hyXsbB1Rz*{YA8&=&<6kA%-Lq)3|08JuWvL7qdU zgEjIrUDoQiiSE=;DfZAOr`2K$v#YC>VacZf%B*F1Hns(#(_Y?rlcl2E0HetclbK0| zyXDsWYxCSjL6axUh_JR?FcBTi69hzvhUqwJiP>4Pjbe_;dP?L(s$WGNW`tzs7y&E{ zw42ZMyjgRyAB3$J5f?SXB6HRYYkr<<+Md5CX*d$4iEn%~HG-i9Wq_mSs`3Wt9ER0x zpUdPOk}S?ww}FA#>>^ex0SKF%Eme)SOCHS(Sm8al8wAfcS)wc4L)UGOHR3U+VXWeO zjVB>pu+AR@y>d~Ityo=WZj2fbgEaB__@Lyrp`1aJvEsQ&FGRc&FzWe%b#&W82fo6> z=Wsl0K6l$_B6qYGif9s0#vjMRsp1+O76XsB=!hdr?oz_*W%j+}4CS?p7IQ7G^8nz| zhTUX=69&Zl_{iwH{bCS5(-*-{ePuw$CLcR_JX<|r89{S|O+f`iB}-5)pULS^Pt1+J zv2mJrugq8PwDAAS%}6}8qa(oIzrwofUl?_raaPE4B#?6NU>57lnmZ*pAtbV=h{I&? z7#?kK9zwsa^5g_}d?shKHMHM}s|f!=ijbqiQhi|FJcqqku|CIygG&<$JE> zpt0uC#3Oj_8lGxR?Dqy3r(qJk?ZaA)Ou85(Dw)PFh)O zi!Y>R4npf1B-ubK-gLPvD-=qoBKb*(p_9NyCNJay4l zDPVOG(_6QwaIrEdaDMj5pGx3?*%bB-lGro3Vc*KbmBSVo3RyU~Vo zAKhkWcf=}+a?{h6DfBNbE&0j?MX6t#f;5khGnE-4DZl4l_+*VF!mjP@?_b$d^K1kTWE!6I3V@fY8h#=pGg z6u*LD_n#R7wzmkIu6Y;v&!(rQioM34DtaO_SJC8Ha+=Md_2p26^{#fjWhoL3`5Fo#&x097oJ7j zpNSFh7MY2`ht8g|h?ZV&{4R7a2Oap3a6ITSa4#lN+!eK^Am6N z3@iCNUaaB$h`Exvuxo|WHEPu=lMQ_OV*D{QgXKB=Y)G-6%Fb~u4&Uv$qdn(>`ZTY6 zqMq&D{*_m@Jda{9q?xMR?}Jy{&3PrV8Sx#pg!5}^T`Pgyew1>lU1HZMajR=;?QNcB z9x6N|`SMn}4BptQ?ojJG(^<@l;kS41Ugw`I$vfS=!XmAGECMFF4vdne`MnA^rp2!V z9zrJ}f9X<1D;mm+A~qZt1w*B4mTYlsM1x|+qTVF#Q~kN^(Odf%!tdF0yDm+y`xAVN z)q)@+MR0}%A)mRK*L!g=kJQxE%th)DThw5b(7$ED&)rMD2+a+}Qafm*j-U~(Mv>ieZy-M*C87vZcE6tI!5q5xkiSL?}2NwY?r%5(hrMWv8Yjnzr5kC(Q6tajAw zAeXPKhH4f}T%IUH=rfcXC^fOy-d!~+YWV8+)G5D&o|BVP(evC?=!p97DVO=5yPiGQ zSs(w!W$KfTCv!EcgJW{re_G7h{Pd(`e5CFatMqcI;;G~k;hRH#%v&TfX(jl1&D6CZ zdZfVFlvQM6p2FcStHRC1RH6M0 zpAiq;l`&3Ee?GKcg@;a}Rm^z{H+j{yY}>WB){5Q9X`ZQI+|_5`XMk(SsE0@A?)K=+ zjq>HiMa^s0+}zxkyb&`(^+lPA_ho)sJxfd;-KJzh$>x&J#s;We+T>JWL(jF54Nv^M z<(he}M+srsdUC0z6F6`Xxu?ea$4p7_Pt&6U0b=^|1f|AD0)bb?DtAY1e#VtIl4dYJ zHd*zTqNQcdO8kKjah+&IshWG8O-O&|MO}9GxN1-G~ck9W5D;0_bUwdydsh! z?|I!R(!6#glGhb~kAZ=q{&wlqUfRb+ev|tuab`{!{GEu79WefPag&K@iE{cU8+Yap zw}ljox>Tlj2Z8o2Xussn*t>*It?05Pgy}s2cj> zSugRgm|Nt>1Xcvo!Ojk}!c%x372hMLVl*3gs)%An6Z%tI)D)}WnD@m#d*g$jmWyWn zo({XY1=r{|@?7UYOVSx^e)tiG+va3@vAFP_#hK85ha;DFuAmS2(yLbi7Pr|(v~;r2 zHz)hH=24*RYioZB>5e1ZnmLae|a_94TDeMJE#`)E%C_oYzE$gPgA&CkEiseC!Qb@R+oJd{n+1_ul9Uh|o2d6HfT zpI-YF?P*z|yK&~o;&NWY3XDWM?W!1DBO_|f#N96D{`$Bf`v)Nexv!h7y8WeeUTV9W!mQ&N?!h&dh@5sn1zrdT`J^i0qfQm0c(QhxoSwBNXPCZ0LMM>e# zlU|RMB%7IO>7O@zdokoFkFO$0*p9t)EcL4J6Q90cw;^`B^B6pBrR5-ft8eJd?kQd_ zt`9;bBWmV=e;*8B!H8=p0~S{f;b9+x(5ii@!5q`m_~xzt`zjW#N1b@bRB z8i?IW|2YNn3}mGhr&`^kP^*FYbIf%ub@kht4J;V1GDRnq^obwVUDn$;NWN)4qYO{V z^~;YCC%WFS0_;pj#}y8FiGYfAO-*a!ejn?n{({odw@#Lp21jF$zlMk3_!TS;**Pqn zyi(uTi2Q`vw`N)u-ZIr#oFQ{sE2|xfgSRioq|L8E4JAaK$wL8Zc^8IgY`@I+5Uq7x z54b{`nS$nMfFAE!V<*+c)Nq}VO)^!dkG6u}bz`g`Vy;`kDFt%Z=1Pq~B(YonTFlVw z92ih_%4}Z3Aai$XvFNbC##oE-w9{}k$E}Rv@|>cMt*&4Gi?4iJDAX*rpvFUHn+Qb6 z*82H11fB6xSD(o_-gZW7*6hm2V@<2{!Btsv>eCb9Ds9Tl_dQtTY;)?TN(}Ls5{z&T zPbeo#-4T0#M?(Kz$CXkOzX3?La|F>-m`*3ub2HAq313*)LTFmcs~xiHE!ptt+pdb% zqUAYps92fF(g+W|XVPoUKX>3>H2;?MP_!N}$G(s5?lq^>ch_1$YzX+|UAWC}h9+4AZjbhB4OalsqSc(v`!dWM~*#)TF~@VLLITRs2QP=1OkP6#hPX z3W4Dhn)I%EjD7eKg+j#~ElGc~F}BojBtiiCJvv|QaC2#7v&fK#D%8_B_xtt6#ipyl;3Xg8S7;&~N%?rv) z=}BmUjm(bl?t2qWpoSfbhR3)Fx|qi&DCQzbJ7eg5X>oc_*cSz?5(=er28q_Ix(XQ^ zsUEbdNP~I_9h6M@@sqqavmKnA&B{0QPS2X@r$Z)wrFIra%yZ(o%B>Z#e|wVM1%gh6ZW+722D^;-#0gJzq>ekNnC6Gf)rv-5n9Hng-B(IJx|QU-&6 zo0Hl4E6%mbZ3mwF2p>^5m-Q*jt1mnw5N#v7yRS<~KCh-}cmnnhzWnqtK0ba{$q12T zhq+^IT!3xtOb-jaVUq+ZH$9phX2@(W3vSvIS>#d&NY^QjO4Ji>a-C4$)zxJ!IlYT8 zQFs_1d;B-OHHV9X-J3V!UVGnnwv7y!CZNM8)TMzwMHH-69vf{WYC`ZGCf~+3ob10z z9X6SLHVCaYv9N>>+bDZ7x-OkTK7w@ldkUPj05mx>Mu4EfV{n)=TX_k32Br7f@mZOw z;ojSGum`DR0DqY>C5GNa3Qrf1RTorGvR|#KXrqT0!kE)ag zt-d$)(!_|T4Y7|!^_aPN^~&kHd4!Ih-&`^Y!JZ>hz1c6`SxJ?_e~)x#ZEWnZn_m$4 z_24@cF}qO-&Ibd^!iq2Afa3*42EM<>2S(b6xRULr@-{Y1**4EyDUg}WXpGG?AomvRKY!E%&f1-HU^_5bU;pL#Yb)JO0^LIanJzL)g8tfP zR&gY4`6cqoHlOpdX7-PIa&lh>yy2Gl7Xxj^;GNzfR%&3V6`pTDh3Nt22he(>e0ye} z0`jb2hNHRw&?d8@u5Y=`vxZM<#+_@SCEu!a1c~^(%ZZ^HH^hO3otr&au z*N9IQ)AQ$>vgr}8P;^BV$dsAQCx^RUM{4&+#l@M7DDC&4rl$?;o^nHO5xHM|vI=%2C2?MldG#RP zs_wkj@YmpYF7$lmDnr>8g^4Fw)a$h4>Okw$ z-eeV#6YS2DNPs+*c0CmdPSxJ6n30v17Gn1z69}S$QTHn2CHv{yy?s@#vZEob1 zEsYusmtgiZoU0U+rqHnvt;cxCJpCC*X`rUaQG!>L{6^&00 zKnd39P50?=+Vi~mmkeI%+&d=z^{`PnAzfugbTwQ`&7L84U}<(md)-++lWign!>y9I}{wHrJ7(Rb<#c@UykTFdn&?yCN0JAs;O8 z3O=H|@$C~R`gTBpsdy>|68=j6v zZ~c3k@h+OHH zVhjxped)7D35noTqJ=vcA>8#=u(PzqEy4~Rlah!vbuWM zerCB7H(6y1n~5hjgM$zDxNgmIWU@uc*ydyb<*EHK@`ps`DE*}LaflG4GBRZyo%<(F_Vw?Sc_W+4e3|Q3XxL3)$z| zN1I$zjoy>SO`j>gLLc~V+CIFW9@VNta%Yg{143h84Ghk@_uY9|DYT_qVfC3NO!AIJ zlhZ@Wq9%d5kz@lC_UN=5;EV2rmj0eV;F72e1{ z!ATa)a1#SL<+x z#s|TUtYW8e6Gfy84;AY~m_eNjBkPTrK}R)n|H|#D8cHl@aC2|sq#eN{Qz~Z2s^Xs} z7dopRHpSS}n^Hs*8jyB&7sJDP46W-bSuetar1c(LE(RqOpHZX#*Pt2Hns688jFA$j zU2R=mP?A~AaoJaJzGy3a>~y_5S<_i94gSbF_n;!9MN zI@j`b{{c7mASNn4{xh$h&#yr&lr|yT@}DmH0y*e1;S!OievThFfvq00y=`ScOdl;8 zIU=-$5^}K({>AQDHx2*&4b7vjwAuF83h@6&z*4Cr8##yKZ{EKB)W$)v`=|TiCt8*Z zvtKRCX89hcce2$~R~xqnmC4!aReA{+#++Zj5q2y85U&M3KQwgHX5x&aU4`+ZkeoHEvB%tjZW?LNkIe>DE*X_?)F~P;g{$6@uu-la`WGu z{s`+7S)Z|aKL~cf)JnXqtxZpsH+ zD#xPET-qB;N9Y8Cl_#eat1It+dpy5zUv3uvK*h+s_xd*XiRth)#&_=$i{EAirhl() zqNuJEx4Nd17pU)g6aTCt%~gG^no-*VfBx&UrNsV9*Qw_{nBqSa?e5#mm}b1FSq^R$ zv$qs<$XBbbyj5S9{KlOW#v-^%YOz4?T5Idr*R|LBmP_kDZ2Npgg}ETSq~vMT@z)Yy z?itniUVQSCa^=e96&7BbwrJI$aq0&nNy9aBD+ML>HkLg$(`~_~^BF+7-4pyvdS7^h zoa<*@nCaPHoa2s#aw&hx|1qQIzVD#^We7_5_wO7|A-uj~U=fsE?bhKw>XXy_L2gmY z_7=;H1e8@HNW|Wek%^L!AqLzvKxO&9gtPs2csIT_h$>8Y4}$Ot;OvZe$ zH|>ryKQy7p6Q?y9_|}X5&cU8?s?&tiSl5z<^GY{6#%WlESb$mvU6#F0-w-O*`dG-` zo;Q6^)7iviO+lgS0i81yVWGb-`4&?K@x|0Lx|H?kIV>0Y?Ol8nRphkvCTM&zS5bKS zP_MnU`z`sZ*Jh1l!g^3`rxIV7 zk3li;YNNp$DjVkHTNo!g-D|t__FKPUSejpob1)ngo0-zY(8gg_s$W~oQ}WjCnVVsL zke%c%GPt(z85gTN6Hb5S-A&p&*lBf;Q%*`h&a8Lg2CK+&-pR@3fuH-ksk69gX=xpP zOxgQ(Fz^Tq%h}NF)!xBF!z@x-Ale}~UnCDg)?p@8uD>QGq`HxsrTEb0-y5^1vw?K6 z;yImjB(Rq$wYJ7$Jt3Hxeo4jU>zopg03e!PxovDn)Axm!+GRR-dF!@q$N6TY?{813 zO0}TOGOb@%;WBn0`b}8iU8$ue?zyWoZ*Tu%6#lJv+fVpf*Fo{$C#;e$R(lX-a}`BAA;@Pht`>*=Hh{b*ER4MlRLXE_G3hu$s%G3Qd{zT(2!!@7(CC zted^-2TEDK+3aAoRKQf+>AY5muJpT#Mpq86)Q_QaC#fI4_ZdGS zzBGHbzaxc<)SZ#&CI8@%*^Ze@Ew4vBUK-xLHC-i{U77qu`t~}SgJ!)f)NCSDoy)Ii z=Vy}47B1M2HGpZ7*dQnSoy*+oVNz zgdlRq5AKBoUK$@OhX+z!yu4|+d4kfdeNrZ+_!D3iY!T;%rwI+czHi11YOSQYzXk`a zn+L(;p(IQ)41ZxT?x^wL$fwYlWeu?-M=pXfzE~BZfJ|1?;ro8VO*vQ3>WfQF?Uj|I zc~@4wnB-as7B|{-4=|_zWpJvgy@8m`9uel}DUGX0?dK%>BTc!l%RSA6{hYNZudHuj z-6=u^9b4HyzmJdaHT7q1F7i~!>vir8;99Rs_8hR> z_H*~aS25L@Bqb%KN6(+$N={aO@?`7#*8Kf=#JjizemL2wH%49`)Ii~)>)_Lj`r2f# zpP@IRr5zt1mlf2P60&Ha_GJ=YO%fLz*zh)FDy8@Xql6d2Ti0NvjM{S_e)*#0;5;93 zu3W_Ug7j4yc-dp_=5M|4cO*54nD?J35a~NbalVdUD`k7%H?G~HT&Ni2c+Jl4W_0v! zafu$2*4Gq3`6qOv&VfV7b(#|3(^G=CwiYM^zaF1_vOE4%{=AleBaE@7k*Y)^!@YjZaOCY^$`g9%o%zFLcCwJnYZPgBY};tSfuD4V{-bqP3N7qR`c=(nsBtuBC%t&B% zUf8=>|3<>eL-7!i>a+4@-$zDn=wI}`e}Cgq+u;YNIkgfEHyVCG5GwSj?V*9q0ni`T8K1)9gUY(N2qNZX-ZS(ZDEGUm;PI~aZ2nHzJ* zzTd_q%7d23Y?m_7{_(A%BIWeSUj#EhNa1BE0{ycs_i%oH&~_#na~Ttc-A^xXzR=QO zQT+W*+1>cqt&CNh!C(jv8CEUw@bfdUOmhS|v%fq!qm|Y(o507##aw>g&2sva<6MFX zxW|TJKO7jAl~oTvWRq#VHiv5bcX9$!{61h8btj~6F>5d0Q@o1}wsm_}h_x5VKY``w z6V)GjlChh%)_pUT-r{@~2+Y2(n7wJ)w5iJfnvG5Jl+(lTL2K)bbhAdzYBgWI1TXPxpB3~lE- zi>U6`pORG%7#&ITX1lrsl?l4@F|(~N;bt>ug!f+PR2H?w>^pw>1{oGcd(bh-6c0&g{OlL zx>4cx7Ill)7t|Ze+IIqQ9S2cyfz6vs9%!GQ9`j?G{#+t^nm(l3E-fO$?%B*o=j`d- zkKavH7T)O#cUBw}D3g91pi=MPoE8@OO#JHe>X}c!&W1v0h;Ug#IYB*eWNf5WV>mW5 z#+Ir7>1BT(TfZd7Ig0TCE}qe$J103fPCnl%L9?-AdF$!Z3k6tmT!T_YH_|Wh@$uo7 zOo(8xT)MPW`fz7)ciSJM=+vF??^xLux2~qB$REDv(|_ETb#bc~9kXY|uAvJb70HuS zT2-;mL1?<9jNb5|bt&V|zP@hcb)Lw7cCW3Cw-xm?70>^?Z_L^8{LftEF> zpT1jhT(1l*5h3r$lE!#^tHFmYhdk}X)CV)lAbtApwZ*Vsp8QtCy)RL>uS^pDt^8m` z%!qIJ-#uhQ5V7P-;E)%NL;SrsKUNW_hkTWlGUgGVCSU*WA^+=i88=3TL}7R#X4MAKvTQL^KIGb(Mz3sbFw#UNE0({qwp)GSU;#^(J2s9 zoqS){+tRl!baI(5PnZ4TUh-PlxA85A*_hQeC3uD52kPO?w*3uRZ<3Zen+HDp*KHaC zaho|2d;eFPF+QhFc02Qqrq|izi7i`pW*>SkWVX-ZIPryJ_yUE;EB1=X%yzXU?qgp*FFfZb>jHtuCjoyB z+b@hYMqIodqNquOX9lm95p4?|iDg-u*%;EKpVmEGt+V0p{jZN|@?CN-T(@BLRD1SS z?28_31S1N@yhRni6IOTbI1Us)Y<+rHCO>meGBG1Q=Iz~u7plxZ*SWVR#8EoUSaDyJ z!-$Ny1wvCz`_`}a80pk|`^GS5QlD~-uym8El!MbmtvvBA0=~Ci2n;VvY&p;0){v7u z^!?!{&YUN&xQSow^|Ivn?q2*5&q>ObnDpUl@pO1se0lfXMDi?{1zrCm!uMj|vETj$ zrXvy@A;5AP@7;)pzTHh8JeQr zqpQ0&{$6rD%9^2$kpK|F0sSJKL2I9~6T!j3v#4u#eSUhRMXWP!8;{1bk)k0nO3EEN zZ7;fdN7o0*KU51M?C$oTdu&%Vl*OE>zt0(FqSV0q%3`<8;HclTXN4!wW<^+$Om4uLeBbHkYeIftQ?99b)?+E^N#6qZ{D#cG#o zJJf)be_oU=G<-ZSSZ{uWSPrufJu$K(gI@x9G_y^&p)3uZNwya07+$9(+l=e_Wuv4# z`>}_ZxxzXtrD^7pbk`8%g@grP)}GhBbwZT zCuQqgQF;c?Oo?y>@rLR$FV>(Cr@!#$ii!_7D|oH<>kiTkLqWS(?MSybo0nR`9i^D> zlw@nNTH9l{EuB61?B9w!jatf!)ShsPd{_VDo;J$d!&em%#YZF@L5 zS09vgNE)2MGRmZA?B?U6MTMIvS7u?;JE*@R9iw(UQnjSLz5Pz-Dq^N>FtQ5?3kxpE zdWB2@Tq@fo*0A&Q`+4u)L|;uouu^SrZlqgvN|HXVwK{k5Q3>S2r-C|~9Ey^$U%Nt$Bk;QSm^CkR_AZSs=hB%FxP>v5U zYh1W+ckRaQ1}NG{GL;O@8jR_I2vB$E%ge8u9&O*~D}Nj3Kq8}}q_gyg8d4(F5+X1k zF((;V>j-_X;Tg&7AGi-=L?PK*NP~=wh&a|c`u+PY5SMT3>(9L#^zlFeJau*TW1Inz z3Vy7F{2Py>B;_R}#gV8h9rEp;YFpr#0X+i)Www^@ojYs5(rRag>+nG>tieHzx#^LO z3=Bb(v!{v8`xT(lKQPyA@u>U*#>jA*T>+mDI*c1AhfD5UwQAMo^9pP3czZ`@Wo;D* zruL>|dkSSBKmoQIRg%8`IH5#i%EaA#1JY=cnm#O&rb?ZoD z#Bqwr!TODuP84DspPo*g7bP~Wkugw5avAF5`LMY!(Bs#rGbpR7`lxO%{Qz)B23ykk zw2vqOL|T8PV8AAiDd=g#2h?GU)JTe~Ho1mEakm$5-M&2)7rBm-GUR7E0qt40%7{ie zaq{!ioln&{9>k}+Cg<{%D?jF@hWhIg0@ei5*8|=7yZ*uYjpHcGs##~d*4tmLlTKM$ z@?A1C#I3K-j$ed!xHpY5VLP@`t$m`j*J);y2Kh_@jl>ceQA1#-N>j9RUfGRknYZQA zp9&HAfCDaCKR#cRouUE6n$NucL*4~7OsfWL>YoNb-))nxt*E$xmgO)B>+d>0&CbQ8 zkQENs2y2lPzP7Ks`(i^Lu^OX5RVOZC=QO>#NP+5p-tXQ?yUtUxx!@S>x)Yi^&a@xI zyr|^ck9`cWD%y~ukz^%k-)|?11%aRs21Ulk#@?GIQfE$_@R9T|YtPI6K=!%kONkrb z=>2-!+}s?wgBGpdQQ=O`#3UXjrM^A!Tv9~6X?ItbM_;(R9-(~SCEl_>xz6&EfPlcJ zEn6sj<)fv}T~LOMO|8<&v25?^>N4n7!i<+SV5J^64b;Zb@7zg)tgsDlGb<%gh;#kj zBy5NbHymkg%CTY-VxpNI>(sZ9|I;omE)M*&Dn{)t&b0#bUdUA=J}C!j7`%D8{Ar|0 zxTM2Re}$KPp140J5?5FxoVGIguEU?kRNXBjh^O8OlW>+!Q0EX3*j!Xpv_16+CeM5= z=3X!k)XYyw8>QLU5zI;tbd+T^sg=`0?nrlez)MfR476dN1czO7X8 z>fDo~?aZLyvaS)1l+h{TKF5`msA5&ZAA)~29l3e)W>bM8gF=`typf@^}f7h;K zQWUA$IhNPpk^a3aWYLKPS(Lwi{Sd43@S(RBkX+K}0V(zjN-Sns*w*daIV2@n5&k1Y z07^-2{!M=VZafg1y}P^nn!SAv`CfarVN=a?qlU3&p3VN#p+`OCeVH!8?z_;RvLoEy z<|J?wd-m=fpffz7qQXEH0n7Hh<1uRN+}u=dZf+|gaNuqv^O>D54qsob@A7Lc@}RU* zTv4(W6>g&lIj1AsoKXZDR-^1~1ok?Ffzjzr`)eHcxf+iYQ7>iA%$1cqO-Q2w%5^ps zI_E46m6RM)4iTY|@uI$Y`}XZSf%J;RsQgv+^upZMwY|MfIn}DtmV`Q}s{6L$&w3h~ zp6RWS2oVorKriFs3@7`mq53|e6rMLMg7pofsAWYZC20^j$QpE->}LmpkexaW8Kh+u zUk^Qzn|rve!0~KWcxI-c)70SFKqb$zvJ(!Djw|~`0C}9U99Sj#9Xs|rwJt5r({JBy z+)ZHkvZwtvM_)*~GGW7*W;5;Izkl7l-C&)fJCiAC{V!F=kDE`^3iWGTIdG3dDN#qo}i9R7j>mF?8S8zxEg+ttJT+r zn+_e&D`4bO39HMIn5>dP&BMYwWS}Ymx?bAFMZC^b`1_9^*M}N-tKC4ANp}{xPO-D7 zsKM(=)Mw1|Z$GJ|1ovNx--Hc_u=iUi&v61^%i6=vbv30M&Rvd2;CI_d9~2a^>e%>L zb=aqsRUiFOLJhhRG7w#s7;6SkLz%x5|9sb68}6aO9+e1b^3)zWYBM@ z?}kc`jT<&}?;pa0HXHQIuj!iUbpHgsa_~m4a+YPg7Sv#J4d5J-@?c;n|H4uqb3He= zA}sn%3Q9!TI9gQrf&iyiL!vKUym;;U^>nfTO+yl!Ju#RrDC(PWaa_igVZ2ykk^TQ; zyEO+Q%E%(_uxUlbiCT61H=F72??3+Q*WbI|#$B_1{xl+>jq}W?=jpiHpa}`BWnA%G zQJ2ZT0uV)3-TlRp8}h}}Q$RZF=GRhE4h0c5Qu~!O*0%Lim9Ke2t5vZt6vT#>PWNS3 zYF2U0Tlaj-nV~?V=JxaOY)ns2uQCxjp{#sIBMI|9G?YZw_16q;Jg)>zr;TN+&?%cUEl8Nz4xX=x_;4meSQ7AJi9Ya5nFAr`1|)urJp{@L|5;E9)JWzEduY;Gci#? z*?CH@Qc+P^x!odhJ4nkrsO)=HRjQ~2OF=+MNMi`WG?e9mp!`mV!h6P4oi~_up!Yp{ z)?n6gVA%UKBzgyC{6cn!sMUAWvlMhVERB-uKds+*CUIn_C@Lz($H#jhr`L~F;w7Iq zhSraYlXJHjAF@f&%2;jD$})-6_i9Pe7Cn6Qs215E4V6^+CKGkrHBz)ZVNrHK8_%Lt z#cgZ|CEe`=(mYjGS*f6-!vgt+areN>R=csXvaYA2^Hz$H*3@K#2{@p$p9pe8M3a>+ zUSuy?oYppPdKKAnUkVqBkgi2gXR`+3vKcvNY4OtTJo++xZBSs~-j+(F8cH}#dZ;DH zs;h6Esou-Mk&m@Y{<30hbyHHz8dC{vjz!&*NZ$Bvd$_m&kul$0s1v+yJ{!~_i~(i| zT-{vO&1kZltin;+`JSG$3P!;wrnpev5uyB+)Cax@?CvQ$q))TC=Dt@tC*)V);Q ze4qdlytem6%guWDL+1 zp2B7gU4WR!kgZ$fTAz5Xza={f`rAJprkdWLePCdqDc_zS@C1@TQ1#VrDDvq|_H?9Y zM%%lw*~lhY**lWCeq*Kh1T`k^Y)T4H0^!4L+Psj@7B&pU!#8&pcF z|32RRUNxy6_k2G!)F?N90V0IwK2V; z&9^PgnK?x zcEUOzJ{*DF4T;7&6Bw4f&f3r-qpcl<>gO0xx4?*!S&Qk5cs%er8-5hW}H?#XrZU z7@)|-3o{Ql;@CS7EIT{9P6GI}8D9wC1qdnh_bu!@(-D*`msVGQ0-WSu_wN0BR0O69 zR`!BgKdDQLopKz7U9o1c3Wsg_Hvz~3ZT48hIh4UmLcmi~Q^_xxyaPKg6PWb-?5`hl zY0RRQM73lp9N88Umyp<-{J`fWd=XHJXa3_70qUT-C|s_ph*eJ0!-QzY$M2h+oz)~X zafS$Wd;B+~9{cw9o8X5A@9u0Mm?NpF-B?~ItSN-TPQN(3%2%EO9xv?tZC5s@es)-U z)V0ncRGtwp1pGMG2z4DNUqiRa%FAnkH&3i2Yk`!{Wo5^dm8pf8eBH*>4#A+(mS20biaj5rCHs7@Sqg9e$T~o7^d;J7k9~A&S3*QhD>f5 zi4g9zA8S3_>9Vp-BU}NWp>$TmdIAxOShSGm=<#M5(bc4dKYih-LNA(_P3)*2AAR40 zO<6GwKJiN@6&0()y_^Q?_TOGyO+i77Np)XUp|d?<@`%lnA>?JbhO`e^IM^+Z!o%s{ z^rFrK721Qypu+zOkn$-r#ac|@$fEESGB$t~ zCH5y272gur02VKyoE=*NWH(%e!w%!Spm?!^9|Z*LCfhqd@!#v=alY(Zw1%kicf_d^! zz_%$da~}rlll(zLkxc61>0euQ(%ReG1EY|~p#qYT0J>T!A8|gH40Iixon<0bk3p=3 zgtq?U+u0N|<~0*V=4n4Y+?mwSi$n*A^i~MT(|B$;um5xRm6^i*xdiiyEz?x&E=6{K zj);jx1IiNn`oV)cDAPpj>!>_~e&Zk8`E-23-R3rss=$|^sl+;TyztcF3(6JX$Pk9J zgo*Z4=6v7%-aCc>FlF5Taf|`-ckI})0`iNgv!SH7CqMS_dGNNX;v@3VwxE!7$*YU+ z3Dq?_H>ae#nZ9`+A+fTtL609tbyAMjC7fMH0DLG5sJM5S@muFGfs90)6wD($SV29v zKbNapCs4&d)3{YLL<2>92LfH~ z2BGM0&Ed!5!ft>9An&VH)a}VysG1CWa}$spy+t+yX&pgvw3mGM%Pf0-{$y-h)68mg z<;swZ0(|bfN&Xs;#KZ4F5PnUBA)%yY@z6?<2L>Y>5~$5xK!q!i2cbu@r#g*;Slckl z_l^4QzWQ!|)&j`QO+R|H$psBGHfzWmbu#81vPmM_+ye<10(^w3F$#OKNzs074s`_I z+!KK0-P^Zs3CU_^Km>N5h{$#rr-X*>OicGs_-s35*N4!ah2EBk&(FE%Bb=vscap4O zp48OV(t}5VenRd`G)|0Ug~JFy&w>vn9Jd+w;=4Y|zBm}=)Ej)|(Tf)kP!sGTNZj?b zv`>M8^RJ9WGB097(bLy6{2OIRXVZNi?fuJ$M5^vWZEea|Fxn)U9H(JmTtt#(q1xBv zz`dmf@IqQd40teNZPmdx;&uwqFQE1s!o8jc23~$;1s@Td8XU$Ae>YO|Ixov9Rd51? zs|b~9fMWir98d&a#h;!r?+oF*gHV1P5HPi>4LvkyrYC;=f{h`wjgtfK&*t6rc~DxF z!V0;13vo(!oG@Nwwug1$)qY*l*{T11&?lZy@&Ijwx`zi;H%1Zk?3u~9Fp(Aj$0J4R z!CM?VAW-K8I*g_ang=a|d`&IzS|8n)CfKqpS~a2)eabZd>psRf2IAxA=MDLQ4~nzL z)YBtvZ!+C>m?FRI0KDKci=Qy4KXXHA=`Me)NyoftcOEqSNR54QLT4NOj`9u}(J;L^ zQS#gI;3OYGhOF2a5R**9Lt|rU@QNn;-xhny8qF}~)$b;d2I|{`3iAb&Q-tBHC&U4! zGkz`rF7WQ_!ZnwOxa(_fUfxs1bZx*!2K^&O(el1kIad1MlfWwF=I3h@b`KnAW(1(s z0t)*zWt%Y+{pn>cV&9!{h`OCI?;PRh$xiR$=T~7X*$L+n`ZQ{hW#`W7Y_s$G0+nRt zHAs#b$eFp=;ez3$v8`MUuw(8`DkwiY^zb~p|30ymyB8|y5Vf(F|MENn#kuK_B@ z)pkR>KnD{3q@v=Xrt>TQG6*WdU$(W)HUM?j`=P4hE4apkmrBfk8iC8?*Ue*LNjz=SWAzwF`XlT0C#L#cY^%uQl* z^?V5|0~{IaC``8*P$R+$cMa!&s{c@L6{@Y#IL< zpz^zPCx_VO)T)%ht6H1^!3#wxz`8K7^BPENK?jjtP?nfv8SrIqi$L5EG+Bare@CI% zGV2!qmyCJsaqao`M0*^Qj)Egm(b0z>2=Fsi2Qz$5NXOE)qvSr89449;(CrcsAZNan z8ChlfBofpBT*h1S(cr7-lg0Bw^7F-DL$HMKxmLm2Z`4qdg#yx;pBYoo)O^-&0Ut^Z zg-H(RK>wYOrE16fcp=;ZbbP#Xx^#<*!C+kiol9r0x~ghKc;J;U?+pnxzTz+bg$uua zd_vL1{VkP9t;E>jG7@h3!r~$?JpSh0MY}ge)|;l=4r#cceTDzeF`iXqe zuiZ5nw`0oVNk#e~2{&)q0{6g2 zMe@%YoS46<6mv^Dk|6opEq@#*cFzdk(NJ13`%$919MYrDxz2;rA0KmH)_(VeaC0Ch zobW?HCm;5rW1~&cxDS*tAd{P%oNVe^TT}Ds=~FMb%E$PERa~ zgk^_1?5|H^SiyF3=f%O-+>dP4#aJE^ls^JYuuTX+vT&Yyeh0-{PUD9v-Gf{1hHY#7P9+tD`o+X;1Ik63ury9p z$KYbBojvRC?|<{_S7rF=LS_xrOG``Zsi|f02_l%1?u5BKaa2-Ll2Od=07{14hvp}! zGXy+Aox$SQ&v9~b?LKm32S~*`X=!Pa9z&Q=WW4UGwm&A@@NV=pKZ+`SK)kF9u~@jM z${;Yyyu`}}V$cD{VIU@Gbg0E~q8A9g;G-mN*pz#HH^Ra?LyJIu7i(TBVpG~uMS+BJ zTVAvWNwo?;7oJ8zx(^`Uh^8J|6z?gI6iz~PXAutG#LTR1CW6p2#@_?wJnu34-wByLwX}2XW2E=%C#^t6rH@M)# zJr}il2;b{1&a)2I0Q}cAR`KrNzhR=UGR^K*&>?AQ>4R1s;?Yl=t{5ACM+EO9$flu5 z#;~w3%~uAih_Lyq%0(|rIN=Uxls2eNM>?RJw-wTtI$?_hHz<3fMJxkA3i@Bsgw^o< z%ac0yV~B|(gWvAg!>+}d&QA!o;2+Ktt zKw~7oaEq1H9u2Xc3~GEaMStWVrozG?A~> z3Gj63y%!A;;$#9lFg|__cm-_9FNuBrq3QRlM#7gDxX^>-Cfd<$u9TcV(Ac`=OyQ3o zKW?L|6v<;YqwL~)uZvL7wt1CACA=)%NYySxGq(-LJOysH?{HMu6k*Yxw+-9{L+GmQ z%*^)@`aA9nZtEBNq(z_xa(xR9;QvC@{5xpU`WAluq6bJ7LCSnp{8$PnY#byqE}{?| z-!n0LJ|r|ISfMn;d&u)E>mDT%VVrS7`>s#_a&!K~EAF9%++MEY9c}`6& z3XAbUg31z{(&f3{zCJTBrdsKjJW=?j4mgl#NjKP>8TVZ1>ZKYnlB%xhR}Bk9cASua zQy4Z>6LTMkJ{IBSMC#Y`9D1D z>~9Sx3(9-cBQQLor3Y6M6T3Xa6v@v`nou12-Y;wlKSK^d(Bhs)oUw^gk3Et-@^* zCns&7zfPgs+^?f4HdvJBV)m)%zM6(K6|^o4tAS0X{Yzx}r-`EP|uD<2w*ZtHA{K zS1%!Sra#`YF+sn$sBunedEtiJeQFXZ@sxu@E;1kDApP)#m5^Og2SDJA*YsmK5``X! z{r(E9)ik1Q5%3K3(cb7#q$x>0!@PgGY`5Q&4Xt|;5b47feWAE*m52M`Yh4}++Q=Aa zd{iESf}J$jEzzzZQb!Gg%40x`IZh5hLwXJkH>QOvgydAP@OkB2jna3W3S3&6DIS6i z@e)V%>!f7oDbSxuU0i)hxv=kauq1E6-i#=MlIbot^J=qPeKW#t3E-9%8%C{Ep zT$iRV)sD_(lJ?US9xXp8|C0ffhe&T1$xkhT=E`g1C8qER{j*_B+K)RP?_ zY!F96+r>2Tmd4}@(a`#4$TG@9r>bpCOiqfGs9m1RcwXnaXKLjoPxsnLDr7d>CkOMKR*XNF7S&@ zGFRQcb?Z;t5n-!ZL;jE8?j}np=w`+BH8tbw>eOKzP6Gf~ztQm$XCRTfE_vR$^JGKI zGd(KJkRBOPQi*4%q-&nqqU*@_3#zKSKF?d}QmkH$d5Wg8czNw<&Oj*mkofT_2J3R( z920CV)6Sy8N=f8(JQp^5O2qw%-Tmqc(ZwRZ9p%|0ybfIo%|M_hZ3Ld#qnAo4bmv*E zYSnqRw49TL6Cey9TjI~p07(X(r!SL}4-lf@<>iL2lT7*193%c>frU(XK)~4qXBJ79 zrsxy^^HF-sjDpb089jTZQRjv)p%EbAuvp*B!iR^=d3E!$l)~t6 zw+Op*^L`vbPsHXwl~y?Y9JBkCkCK!Khcu46Ee9=FSuHxGvwk~l96$iDTXK5)14rg; zjWDd9LzbtPm&Tvib%YNDHcsf`yESQjeH9^}ig}2U@&#YEC-9c-vWKYk&lBN8bgPb? zdK40(-CfinA9>l>I59U@6ZRveN(fj$_$zJzA3ra(s7b*qUNG0qr1fr&X);7c{}LAzf=Q?psk3w4C&d{Pa z$M*Q+kDX~m^=oY*5-HxSPVsG5bWa!=y}WhnK>m4D;y!o=5%;#@eP7GU9#t3%zTqG- zlxOU2NAF%C$KOV$!o-6CJ$!FKwZtPaPO_3=`}U_BzJR13tp=V{Lg;~Kh82*dXPqU$ z0^wJ>hS9BGuZk?B#tP zUfyl2b|$7#^}(^I76BLea8axH&qw$O#P(wDgkH8BXWI)DgR)l9CdtFMb28{@E5C?{#_~d+l8%3n+w?lsxNs_yf9_h;r6&?x4-!F&l@&N^sp(dDq`?iCp$a0;x>*?+N(qj#( z30W;V1PTlhu}}v+b~4W$@pe%1Hpr&jdDP|vD4#s^Pc59beWP>K`lk;}^-K6YUyL*N zA3m&EU0r?e;lt^s!20!fP_4b|68g*jGfWv#&5al^DOVE>$|^24hp&f25{P(_;&U{V z_iRA>L-Lh)`?5?9yz3(@w2GIEHmv+`;ofCfI1l$DVl>i^!3Zg%CjhRm=*vqwD0}f@ z$xs@NV7DsWghHJ>6myD>D$HKYl;7&6uklQj>P5h_JCNS(2*k- z06QMP+#zTjuUF{AN%RLKt``@Zt9S(k&(eC!JVqXiaw;)QT!%lcJPCb8a{ZdsiRT}) z_^%hjF~&WZJ>rAR`-y+!T4(qYD~`ObcM>&8BF_(AyalucdxJDrW4j} zS$p#DvfRF4_fp%%r2tTcQ;~`)W4LwY3CC!RnzOj~?)@;ldF2PBwEv9pD-;xowA6#l ztwSplGtT_rza#>ElvlpO_%);Uq@bW+kIm7&iXa_0iFRMa^}-M1bt=0Xd(ca|*6!_x>{hKvXLi-o@#vn%ZMfQF|3ZaQ*9f z1OLqGxWfOq|8JD6e)f+@%CCq9WMcP!y$~Vd=y)*Nev$}nA}$ty&Kl(F+bf&?-fM^V zC;|}e`TKc~&R6D`fpLD!dLwX@QD*aGNaw=GVWKGmsYIO+ZW-lZ0lCw0XGKCzAueeg zM4p$}OaG@Guz&yOLg%a3uD$wPzw%sCpXSkjhPU9qvV{kbw*#@=wyER`c><3STeuXb zM$G>aGRB90zy1FgGA=G2BwRlh0`9C&I$!zo-pc1Z1P?Ih0jZHC5P{ZzJzWG(pA9;4 zqqhvI4XLkNxK0cGH<7a{_~rHDbey>eGohPE9$J3BWYNRlZ~xtXnBz=VDAfFOBT1M=r8 zr0;lZocS3V_u#>*NAtu6@sNVE_Y_$tI3=)m&mKhnE3u3bADJXwa_5_Tx<-a2=4Qs+ zKhKcm!Ts+j;eE!y+zFgj;BRQo?p(ckH35z(63MH)EQ@v7cizS(%ZqmBpqp!Ybi0zi zD1hyU(o!933u3HYX@p2K*dQ?=2{tWAyrlwz+PCinIv3>}M{gbs%$R!cFAv7zPY?XAm8JDAfehekW`b9F@UU+ z<661bB46GJ;ZXk=as0L>TEReRrj!6vAM*qTxd5mdNY(ny>ju)bhcyXZ716MZm{85j zjdnZ|OqN|GH%-Bi1~!mqdx>cNqPia^Ia2EI{M&Ef#b$DslV_<|ni$&?3>KA5OBP!i zO>#0vu%37C-bLxmbJ=w(ckm$Ya238`VpWSJevJ|2A-f zr=ANDx@uIo&0HjeIIZC=wea@m7xE*;Uq7D6c|qY}+QrRRT@<>OyG$w}>Q*iHfoqB` z!Wqd3NA&lF3$~|;+2*Msc3*gRTleFpmyfEBc{XfR*IYbm+9IHAJLuGT5JJJ2ghUT? zYT6DH8i^^l!(1+2(G`pwIuw_;z^|i%eOS}CC)wOl&6Gz?zeTot0p$j=rOWnxH zr*pkgw*Oud(=%OS;SDxGgC%}=t^a$GN35z<5T^t=03Evu?=Y}=Nr z{%A~1?{xZa>u;Vt3bbK$kB{VUbC!E%xwBoGeQNcoWI}_C>l2Uv`c^_tOXy16X%?{_ z%cLzV5#QRQ>JnurZ}xtIGDE#~*CNTq!{bG`biVofUV+UKIKNnUO3YQBcj z%Bg5SMjDqQC=?(FXyV6D9z8Muc=?tim49rk=!tGD@96XOdnZkk^VlCeSe4xWaw5xZ z_WL7|j~fz(F>gqJ+aUGJ-eckib@cn%s5C_c2-IxXSSbJ zA!HG9da!A58|9N5U0ZO;#hIC*QHqfGs|SX}R|(f&K6G*LX;bOq(Hj+KLJ#)}q;uf&CGNyS zlzV3sY`@qV`^J@DlBrKUDFeHxT7T5WBA%1?{T{_Q6-BA=oX><1sA{hb)-Cw}y^Pw$ zMSJ!iewJY7A=UJ|_R!4rK1*Mq&SSQDiP}$C#3~QW7EHw%=8Zo6xB~Q^Rn}ldKs*F5j6O^ok)LMwL9c2YH+Nxo4ypV)3C>ag+#KJBfi zbNvikJx>ms)7EhxvZzey%;le#JtBc=@@Tn zh)2sk;)k%esj<3*I%c^eXvqn`CDmy_2&w(U?MJ~rhK%mcHyQ7)q%;jFDY@Z3ru$)U&m;jW zO({=$TM}4E2t)~>A^Do6@rqvg+rOd6f3VfRaiXZ&2_+@#F+C<07W01raN2)>k$dT+ z+RDBDjjjKKoBnN(<*DEZfBF2`e^14;Lo)EQ2$z(6by34ABOA>6Fw5gTLFJeK^Q8Yk zs{d_#jAR6WTNoTHw|x2N@nep-+Wm?dtt0=!XTXD!k{$mW*8I0;#!^zmuBB#T@2iSwPRE7A&cHS(Kb~c$B`p;ryOn2GNn0QV$a6J>7hXy@+YPh| z>nF_96e=eYDxH)Vbh$PWKr9n|s*6FJh@>nS1wTFDs_*V@nwL82E@&#esIul?kz zl6OV(mbSqq6>NxSU0ukPdU6%LZd7as7MKIL2~FWK_z9rHquYxvUnZx`I!QpF=-mAD z^?O%4-VHw{o*SV>NNb|(?Kv8oiBql>d}P3*{MQ%!mnJ7ZO#0tfpk#1*q%&z@!Em(k z=Bk^ft*=ApZQ<)`xX;JfjdkqxHViikV5-YWa-3lNU_TfI#_Fl_a)F`aT(|dFmf88& z*4DJ;7lzd{V=J~{{?Py|(J^y7a`1}{LO=PjHFmh>@BO^ET!XrniDmKftRsEJ;R223 ztsIT}gXV1=99)I1w*PQ*ogHUSS(-k1gk|2gfBblT%GvYu^fdKRa>=QRxu@CLBU>`< zgDRg7)o)Cj+yHsLFclgmAOBcp7K)06 z|4CN5d(Zr3LkfSIIxl)&a_>8ke@HftM~r9Bo|kPy59YoSa-vCIy@M?QDw44k+9H8h zo!#GB3!5`0*ms1+)!Fwt4VO7Klx{fd;$`VPJ=*Z%NM_dHeHDI98HAsvN*3K4dV=)3 zGb`f;%BQ{;hfO)UePhlP>hryzSLr(5K_w?IA9z*FZ6TLaS&6%LJLV{-W8oMvHRXcg zjh}dGT_ne181x&z>52(0{)uy)zKVJb9bv^~eSc~m_OR97T01diH`umu+V`Pl^8E7Q zbD}mw15*p$Y(+_)_qRLfOjr)8tzDyT&L4?f{{AAVj-KkJmL#i$_H8>0^b9Aj5H4bg zQxmq*m+k|0gJZQz=hG^K2LvJeMLBJQ*{*M$t;{B$MQ)?g-*ff7SGdqcwa3QAl4lWT z(3^kqP}2JGO87OT#sLNEhf3eL5Vb+%5qU%zw<^q6y9>Z3)g|@uK7QGi_jX%e8o!zQ?CDC$8T;_QR^9635_z33E!!>C zZEOs~7uCZ4gnTkdkIZQwt+v8oj$Z#>x_0PGOYU(b?Ch)Ej*+ppTmA0W4*WZp>BG-0 zkx=}lvNCATT1x#EI%m=~c4W)Vwd#lO-|~8-IbYkX!6hYkLS8LJZ!xChg=p)L#M0p& z`l-UbXGB6Oemw6B3y*DUi@w|jb05IiQ@*``txOtWd-p6U4U z#hniq=Y~32r<_I~`Wp%xPp*N2WtMg)Zmd|NG|HhX!mXGVQPu{W|zJ4S4wTA}@M3RH6dyk{p5{uul0|yQ$ zrxgjP1aO}IH78tTJ5(+_(eozHqHIj!TTIae+elM-#o|#AmG8YBPg9=K)>!&T3)Wt% z;HRtKv^FI*-$Y)lYcH`49zapkkP^!!0+wH*JBE7vSYLQrUD2YmL`%2>#|Oj4s)^nj>8A-F{zDBac(h^bI^`h zzA3fXZ20mSKe579k-H^Y>_@((UZauUAu>q|&s<|buxOT}Jvyb&w@^a9`VWYRsLugo zGbWXiRy$vp^O=3O-T8WQ($BTSc|ZYSCCiBnH~mv9x|Y!!Z~UdYxO3Oqb?Y?Zt-x_N zuuc>n|0fqsg_t68a{n%y)lzeJKoxa4Be!j4YYWnTqtgp;^!{ia90 zi?_VCT&$0_81L$4wP54bjz>a(W7b%6=E>mT%L?VTCTwxJxw8Tm!Xtu&A8o(_$VDR6 zM+)pNbd#{ksmv_keWf49+h;+$OR>4RwO}$?t0E*jzGfbUHl5dKKc4*c#V^ZJSKf0B z=cT`+5@V~e-|JM}&0Du@(Kc@BCA1!@I@{PbnthpE{8`(>+DIp6wJ4d;`*(Q>W=~kS zzR-C9;p}GRRmTE-uIl~j+TT}pF>QoR*Wm5ay?y%*VkdqZscW`hBl-$kF-)J^J z)0tNE(-}vvh}hWIb?e|^zxQsB1k#CB?%c)}7r3Dv6x7oLixJAA^}h~-4VlnAd*R)lgZ6uxl8vcb+H zkE8iduD$lSF0PU!Cnx;diC^*tv6qVAgn1n27^Y{np#3D-T$m+eT2(TdT^P}ICn}ox zeDjUgtgA_cN@sJQd)xl{;(L{rY#!~-D=tIT?qyx{l4BoMR0{)w9v;V;95!en2RzZH zo!<&{^|_SFe9h@~9k&x|^6zT|M#@;tm}vj`EWoSx+b#Gt?Z}US8d>mX6b*L)BKZ<1ekpe(-L-N?OpCc-g zu&&XIAA-WpAuNNtBOGo~nWx^U-}pXyq;RIQVR}S6Y4Qy~D3i$KL+g22h@0Q2sO~F2 zDHpOPS|CH9aRt{o>Twg~^hR_%mx@ygYs~$1nbm?>WI=OpWg?YmheKxL^uY3rO`uA^ zNT-#-n?COj5jDVpH(joY^DBD{ZNEtd5vAHzXJut&{HE*dm!j8>74im^pBtTW{PcUW z=4+p@)9@=Q-zWEc|3A*YIMO z!OHHWanW~+F(TV_jpF(-+mhq$PLrme>?y%o9Ln=r*NB)~bZy4U0W_*35p5%}`)6C= zQI|jmb}tnIo_CA+kL1=yKlqpc2&Si-FYy)9a02}B_$G(RXh=Lq+2LYN^H-RS6~`|Q z_pBUgS6nJ+W3bXe-{w1QCpA?cY(;oaD+n5Fx`X0f+k4nPmfv8!T{~x_ZRb&ODQE6B zX9l!B>@!xFwo=w|pP4PBLF8`j>4(qu&eD1Hp5*F3*94vvx(`rWgo>rV>16HY+~zY5 z^b|sc!WNzc@&=I`)xo)t49uV%{(jqew4(B}&*U}d!)G?GZRMKtz017pkW@9<5_dYE zv~V2iX~ajee~=l#xCKF2oHt=c+Lmp~=AWZ~@~41~pc&+#>xb3{zq0^XlsJCu#CN^T zCXhbJ&o7da;OR7#UDN*JGw%}*2fP#8p1(*%#jMwx@VIDn=qJ;5PngM=bb$erQulU^ z%V_jPV!@0pGj_uz@!7VO-`6u@{`GrEkLE<#=#?9vya<|kCbe*bclg#ph4}Uugw#Ko zQoma@Cm0_pTPRwXm6W?0lqW}Wm3B`x=T(#`8I~R^vu`f>%biTgMIOPUVH?#_QHj#3 zX!k&~9EE7$m?KdLK!e5kqOtGZ*yt2@avsWPYoq5`HP;*Sxx$fUB`4vDV6OweMl8HB`{N{jhoys!uU#Uc;Z;?QA;5Gcqhx1SG%p#M_HYp*u-ZcpgI26ujcLt#o2f-U&_;TLu=A^s?u(`Wi$gD?fx{Ym6*BBC%mUw4;l0G zfLw%O)rz>ZZKl+l_(QKNiVg}~hBwm|_F9gre9sBl?hTsri**G><5^YgzTtq9_&%^% zCXVCP0B8s0+8?*h6v{A6J+nn+xdGrSPp5|$aAe6jTK&sd#l?#Z5U?yj`=|Rm{}TrA zL9ZjR2}fiKF!pT&hf7%h*zn_q9Nm9fPRC8qR|5pWH1x~dlj9K+4@Js{|8pbN1KQ=M zFNdoCRKof9zLnQasFD9ueOOXC&GjL}uYwf&^!0x#2mSkZjR*h=zU5!--2d2s`>#UM zzkkx}CZG`if17~+=ZX@r<}(p-2S0#R95Vj7F_Ck!jOnj|P4lNc3F+WNaB_eC_HUiN z|McY{8cbr1EiD1iFeQVYPl&?lIC~Vz$Qvs7Kz2v^=VI2@#nA7E=l|y`wNB9k$bUcn zkM$vN5T`F_XqWhM{XeuQ;Q}kKe|*8X`=0-YHul3aDlT1#0>cgQ6-3DQ@S+^+zCbAPC&E0%Tdg7LV<)q(9cG|VnATig%k!l{?NGWwoV^( zE%C9?Yhgluz7gJUx7_wnLV{-xvT%pNpGYtMXCg%rCfc4IR76Tw8w}&qB{KNLIZMma_joc63m{#HZW8cJBbNw+uiT%>!ngpDBQxD4#yHUv57RclL{O=`dsXmC>7mEm(i8n=E6pOO8#;)tiX* ziQ{rk$aQk|{`%OlJIWuT$BtGku3%#BFz?n}<4gh!X$qig0{hTW4w9gk!wn<_zc^3N zs{lZN+PhIU(%)KBc5YC#-WFsdncy!AkQ_WZdWk&|Ei_DMZlR{T2nWeO*cmP zJ-nyT%FSh%P<|fEKqFE-hPHf<3v>c{0LMYXZw-iK)4E&I3Svi4MZ@h9`1sYS9pic6 zGh4YPIU9*FTjN&^-h}|7tRznwd;F?z#~Xx2^^;KN8k)7#)uCNy-#cP4w!Hm8NXQ## zcb+zPQ7Aj;H6*dXvm^gs8}SmgnVd7)n+I&m{ia-rLUJ3;*F+Z2Ul@R97kl2Sj`ShT zj*h(@xTBSd6T1=X#dJJ?(g9CIX&HbxB8rzr9KKFX&V34jjrbQRLuY&;%VQ^eS!Zm3 zM*6+`&N8*gi(~#zK8;5#;YVH5fky20W)jMd`p;4vT z5i;oFCPWp_+SZ(O3ee6&EdrpSzPD+W(m;1%5#WhZfHi!RiU3;*uS3TGmG&qp<4v~o zW3HY^W&%vxlZs@&IKb2;uezNu6p@U3g17Xx^I=awz=)L&XI3Z~K^jB^;sU`8Ky;(lydc-U%fI5w7baJ!E2nax#`u8-i{IRjv*|J*Pki5cNUr4O7&t3ILv-+aE0|=e_G7Ssuu3SQ0~V z?_U`PhH4VP)ZPJC2tiv*WhhYET|*uvDS{LFpQANY&fFiXdwdR)GV`&K_C1Rq7Jhz^LbvCamGeQ9&rdyBYI-@FmFr zbfjWt?yp@yp%}1EcS=PBJnFT3P)jQqvnxT46#|!py%|IX0Q`@YNHj8|3Xc!{4++u$I7(5`(i3Q=Rp}T*JxG9t3%}fq3v~mFFQ4g+MM3R2?Bx?1ddu z>Bj+F3dF7>2w2!G$fC?(;L`H|cG$SW95PwtenCJBI~823gX|wT-$2=CnSTXP4teSt z8b2nC8YhgBK+pw33K^wTsr0R3lQw{#zd<-Gh$Ix0&K7{1^t9wDTyjVy5>iOx%5LhQ1RlhheaQ>-hbb>+5#%yJH;OGEAwNL~J z1#V{sJlL-?xbRz^>H`c2V6rDo_6fh3$QO;V3#gD7;8DsI8OH;!D#6k5ETpMAfgt|H zzFv!s*m)2`Nd}P(iO$fxaX4(c;4&cs_4u1z!GG(dLOt^xugA1d=Zni5gb%E31}7| zpjW73%V8CsVQ2)FSQNmI=4tKsKl6GMBm*(8<+Np~!+_lF%bc8?;RevPL7)Whh(D}- zJhTL^hxJwmA7YktV6UHAc@Y~}=;x2bSWSTt6bokWF+jLC5y}@OJ$>g0427lO{SJpI zgsgE1{^?1yKRE_$a4#&w!1FUnZy~7#_LU`|K!@Lguq7ERJsyj3+)CjT zW>2Q_*isG(1yQ=r4Fq`yw+Q8yIC!g&AB_5=J31UYk%E|V8`K%X6kv-#|A`>KP$HjU zB)J$c6{CQ~KBA8(fxy-|eaum_r?}#zm%ylC(CK~vOce<@&4pk^?(rjL6~V^9sZ4{u zd>#)x#$wa?7~PB_1arj*@IyOwxKV`9*JlD8%3cKY4l_s)gQ#;KE>Xw(ZZ3ME%o5Z{ zZWVyU24r#wR|D9c5b}$-WL7=`6eJ>20zN$!WMn`JnIDRaJqmQ<3)Tub7l4X$ADDs5-eZ7EaFA00s@jSg?1)A%cPGt@rn z8XAHRhosGD>I9z%HS`}f6R~+PO4GMHYU4Wy{Oi=YFUJ_{!Tm2wSyPPYoE9(Mrs zGX;?7^M8TX?Lf!R)BK62KU^qk|9gs<0m2`E&3;t`euKlrpST|A6{kBGSR-!%_ctEQ zXfO=p8Qt6-L)y%&2CNbdC4?aZ;x~ODG^#K55`I#W6wWw}l%{3^6v^VA40WGFu|b3N zhb_UKJ6fvbD5H8{L7CNHFg<|w2Qa!LK%$R^AEBR}9uypeNCMHqGw3seF{b9v8PEgI z^qUvR#byYH6z-!lfKt30N?;@AL;9M*$XY)@tW`9~5~M(DiQkW=;OXEJ z6NCfur#>b5E8JqV15_?k0Wm8LTIhap&^FDP2dkHL{deq!s<7-f0{A~I=qR^nPfpS)} z%hKq=B5RX3u+Y$WXcWhRncS zK*V?8x{@ZO}r+lP)z$^guT_{GnZ z3Q7zp6bNd!CyA1V!X49>LIO+#{-y!8Oh?vRZd4^ut~9(6sSK(o`Ji8u^BB~>UcPv7 z#^HYYa9aA$o&76VqM=Y+z6@DJH9(Wv%WZ8sG{Dm!3fD1Ef8h0oQ{+a4J?aj6n<0}@ z2B@{A+_om1%UHf2Sj`s5Rzsi`H7T0#ZfTk!os zpXoLQJc8}q*w|Q4Uyyi?0B+(L6huFt01_VyPIeDicTWwB-QA{PyT(}Kj4v%wn>^)_ z5M5Gw&7=MnA_y0-18_pw*wnpU+#0`o;?SLiajA%iBkJy>M;M4uBM4=TT25hv#*)S~ z(LniP`-u(p^-#{-H^tCzPcop;*Z;YrdmW1(hB0A)Aegc13m$bp@O+NopKrjitpWRg z9dyKH`772(c*5RsEffSY`mq)d< zygY;H?sh4dR$1z37#5J;0t7O{OrK~5&c zPIFg`WU~Rb7Q93+Un63BZfOZ4cyix^Soc+&$0%@WQW>Qy$9O>v6II#Ma|`O> zelVgT2xKNDf=Lu?s7a)~0c74gC`AHG%0`}i` z&II`TJAs2sQPR@VQUiq~L`n{7{~Dg2ZY}NyE%D!YJ~kRZp67*~3U6}(`FZ$$qe?>( zl5?PjUkP8Fr9Z@I3L@B&N{K>*mNqucIJj7$*l;Fu1i>s4VA2+9G93~HeCwzj()%tA%t zhhGwOLzc!HZU$cJ=fNY39(@4I2`bGvksmfL!}(PA^l+Q5=fCMjedq5A@vgxds)A|& z-V_$VB`$(0PB0zvWnfulX(&M7iMjutNLXp^t5g4%(;8tUZCkR%-HAp@Q zXN5_AK(eOi|tgNgBxep*Zwl+7T!_%(? zHV`5s1T=>quuwn9?%hKaWKbxOhyjVoCIYbbKhpdSAi4tTo72T7f9_B-9&E4N*bu0R zfx53JFbk{u`uf~p_tRzTQu=B<8TxZBeaWFYg%w1j=m7y80gLhZ%+if35J^w0tk77z z;~}Gio215X_%pAH{vN0Tb0H@+3PzeD8uf^tmP*xG|S=P50i;TZFRtOWsBUgvwW)osZuuJ^G!IQ2GcSH#RqB1`wqVd;Dc|K zR4YdG*I@W2J)+_aiel+-ykLcI0425#Qd<$2{8+>dGF~9(qXq>nlmn!y6Hxx}M-Ic+ z0q=Q6C|dL^#haU()AJ!xP&{PpAX?S(l>r@gloMC2PjECw&CX!~Y zUc{Ur+wQr1^&^l&N8Xy5NlPL0%?&&k4huaA740_vP38!EUyaF-KPOIe0V*jMQ6TI# z0nF7JkO@YFU_#-@+8QBi^4A3Vh62Z|%aRae?BH}x0qTSvgftEwP*ZomDrP@i%BAP! z=5eV$T|m$?KxU`IM2C;iJu0dJ`NwkruB%zCH~~Peby5>l&fFlH!i^<~>4X&^MXcEP z)YLf}8yn<2aI1iQZmeD!t00DO7Zl&>KWLA+l4j$Y)T9b|(0~X4qoz^8ElRT-lcEs_Cn%{n_V02?rVefdzL7hMz7a zG87KIf@YX0D6jIstOgWFeIgbV`Q-d-kWfM*kG*xZ@9#R!B3i+@nm?y8KPPpH8P3;( zJEgu;5_~#RO|}qB` zF@m_W4(ciw!IVi3fXiQlpaHg*=*&%g{AM`QRF_YWt0#zVQ@^{PsGQ@8uM! zbUZ{?3UupXD1(wNdKic$2ubj&-z8}CO4@il6bI*fQ{s(r__uF}hevGS2aqE~(%G#t zkclS*g^((UXXrpgpdZFFvapggCyaKcgAOBN3|%xOL9U<7;ySW{dB`fDptH3+bjM<4 z_zViBd?0G=5Tbna^73lwzzVH^YSDKP8G;ZyD<=nue3%pPebJ`=Zq)Or!uL}bzynKW z!jx_&*kr13a_D(qsoq<_WpAYNY0{wm5&_tdU?aLc;Bm*prZ+P zve7unXRh~eG@$lpv%!=fEGWWNLy;Qx?LFkwt80RGA|3G9mIm|CLH3s`kV^8e_q_eW zmW>H9E|IY1K&*KZH1ym+wFvbbMr4B0jye*cwfMrA29Dor+@8doL|2@-OhXZ!PiQ=X z|7`q_e1C-5gwl5gY?(aB!mo^!Ttvwf8r@5L3L_98Zba!{E&u@{WW)rb9}kZfVw$(m zHkkh9*1`v}@t}D4&rXuK9HPJ52~PJLyUlHg0G!2=1D+uxx=3PP&wvWeSrn*zt3go= z2PT++`Jph8;7M47(~-h|6Ut5~dHG!ZbKE@E^ndLt)(@?v{_M5QP4TjF7y|_ScW)qs z|G>HBJ5OaWb0JHYQdFbrj|s*Kl!YqnI*=|^Km>hl3+Nr-B#|l}cntNg-C8snonA@C zDmGFO=yU-!|I?>WC=_%X$dQmm{RfE6wS%ZO5-n8x-av=p49pq$zPA_n3m~*?hR^~R z9dT#36F{d2O3g*c#uX*stj9NeXKM&*7?!{H5EU@yGkB7(Ao!A{XvBd(bOBW2<_W1ZRh~`9Ay3Wie_P|h;E)&!s;}tp#a?#()pe4l3;JB z3v}vZV`1IO&Cllr9h`MA=}7tov07GcZVe>7R}YGyg30h}$GwTGEFNErhCGEW&7;2n z4zwO}m0$$@^irAUXIECd0jX9!__ggV{A~@68z`5i=P}5EUfs9w@aDW_h!rGtji7dd zDAqwFGO;I1_7LQ`p%@(%6@_ACWMqCndESRV=CvclnLL_2WqI;1BSMo3jL|VLG^!!; z_Dle8Cx>kSb{bJ!yiP=f491~1lxLr~fC4uPS=Y;bCvv|^FqX*L>-B%R%oOvwR&O;2 z(S60sL>O#Htl~KdHRG0!>|}+ zP_~Z-rLdGf2!r=_h72K)a}XoWxl{1%p_M*0%r$|-&#^y^4Z^Ad#7}*Rz`UoFS|+{( zO9F%0T(u&sc(;p7(?>w0tphZ+ldECVq_16ef#_{ zqHYK>%KVdLR1dd91$rQN@|{sb+~ZYfY2ZB{kS|FtFvbVzbNM3hRX?<0LRD{u;%9KR zMBs|*k&1wcm45oLN3#d9kp5^X?p#}BFoy~WA;>O*L>ne5D?9rIY&*nadAeK3!vllz zF%FE+nNb?6sW}JK(7?xVog~_3zuM1$mID_wI(WuTK^%{d^thHyOJ==OIhIqhxso z)PySGK2cHKVcx1xb)S|2W1-#)!THtaC-2|%WFn~#6PO!ks0rYcF_M@q_&c7$(SvDf?wVR}n(R!(nQyO%L14azw5q=T z0wR3`@f|%QvdLOP^70sva=Uf<@@(`T?^JrhDAlHCB3wfPZ=Yv;aZ;P6HHc5@frQyc zznh0*`3G6EQU>kLPlz4j`ki{8gx&&{bfF>#; zmX(s__B2U2=5%}e!}bQPGPqVqmoF7@-y`u;To+6rkK9Z3nE}n}ru%8IdOJ-+P3dtl zF(jbDhs^JTvX$5;-nftOyS@XXw(vjifeJcAUI=CgaWk@c+Fci*Dcych4nsjsFCtg^ zn^EcmFcOlTDoLh0*8Te0vTE0 zc*(FJBi61?mx1#L_}(={1{C0So%-T406&aV7|tkre!F*g@c7Ju@b@UYGC=`>8o;K!{Zg2`A-XEpghXVtsWBdcN2IYx4G8jX z6e5t!%);`604tD>caXI9Tl4eDKs|LWO5b2bhf)Phbpd#=H5p$SmI(Dk%3L0m;!E)r=5c{aIV!)RNT!a%9{MJqS?hVhjzZPqXZCDb8hM|e@3sJy2=$r*dNEcFLr zP}M*a;d3jfo>ap5L_z$QuN>Q1Y<&;=gs{}Lq+nwj5>2;PiNA-rrsr;xY>ujtHOx(R zmp|<`Pxa^>QE1;|aUPz)n&_;U-`iEeqwgeYCcFZ+BZ<$-*)l-d?cy%+lsOkoD?zyr zUTeN(c`CZx_e^pN%L{7xk475VMv6i|b63QVn{#ChM?fD};$z6LOY;x(7 zl%mdrGsg=3%Xc_%EusYdUp*VUaPc%AvG&%R5Gtz32c9d%hRTTu)n}u1bvzn^3fHqU zT{;r}QVi=GBDA=qu}x9q@3GAFOWg~buSsf1E~Tvxqtm<_^>$Z=qfJbwwK7#YWix<2AO?M7Bz|P1#0%AF9(-D zK&mGYVEs^7?J84+mTpWlN?0aTUq1pVH^`pRFD}w18;udtx=S+~?H$U5`(@p);*?SnPL0#S)1HMA!X8Zy!AQ(k5%G@sDkbsc}9 zl$BoX3HgyMg}nU3(x+KOYbU|QpB@erTCvNIMm4G`l8oQx z3HOz%l+^RSKDM`;Cq+)4-y7W6%qjx%1fjaBeqGCZSo~ zT5_z(N_+&f{RU)SCVY{-sbBLmoEI3FT~2?!M2GR+K*!7NgJGn0@$}}cv+L3k93zIk zE?s@tSfVpgc0UD5zdD^Izwe&tVFjmkx=^8KadufY6I^t2Z<^Voo_2}dRgUT!&xVZ) z5@%rFO&io{Oo@*Ad-wIE3`KcsF(4^62|eZFd+%`v6cA8gr4VTZzmx(55e`k&Io!U_d@_?zC{0X!t( z<5B-`{PUa9M6LgdgN48GcwcRH0E63i@l5Icr^%EM`SS48i&to4kkxTmSz0ks76in< z36{M-(g^K1N=eR=ZVjkokfbQ+^x3j{YYRzz^z{8g-YA+XolFdo2~SX3{$ZDbqqTR{ zaFz`vKh4}k7w4sCHbUuehyp(=<|X!Vm}HRoX*Ws<3=c}HmBACDx>%wk+$A~t`?^0r zlb?d!Xcwm&l{s52RDX?=Q_Po}+vvt#lZ3kP73plJq@)3Y2o`zN;XZA{2V&KYrZ=79 zVM*z3{`v6#$E(F3xc9yA6BCUd#sbTS%jfGned})ykAz5mA^$2sM6_H*ePWXSy{gxp^L6o*-yU~<`?SLjuwdGYt<_uoE+#HW8>uT+st=kM#k-&W}IQaq&Q zLC=tB%70#>P!;*t9-1J#sA;j^nbhj|5yk{F01iu z=zj&He=nNs|8t2UCi8{aIw(+Q_G7{OXEkCH|F@-i2qhjdG2MZ5#xdu)`<&BYRb8Fr z$`?0vX%P_zwupel zIGa~WkUXn3V7%nw`j(Pr7}@vOc90L8Q;?$?JU`qQfRGx#2YX(We7A4x8N&@|Id#pN z!K~v>8~n7cFO(^b@Ash|)b*iv7BES>EeTl2dltr3U9 zNaE&}bmQ55u{->)+_*J0d6O4>!x4c}&sMFF3y2Kiaanb=m6@3ONwCJhh=+FpZi; zHA6uvaAt1i$7!Wp(medx#i@ln9IqS)haiy$-|{kQSTRCLD1DlDmRa96x8a|kAu=m# zA|w{2gH96$A?r;L3`C)xJ$r^6l1?B2@q4{S!M^kzwl=j#yLd9-)Zf2|LOECClDhcx z6k#qtRriQvqX}cs{YVq;^nS1LjO0M!mme@XB|!a9qA1ktpFO@MUXqXA>tgsqq8p9P z@{9Xiq@gR7WJ)99ZQmu1FqUg-25%sb4aU0JeOrhjd>Yf8`qOEWiR_`IC7>GLv$(`V z(a@YrRLZvagYbeZ4pfxLvK31#RM!0<3<)%LMP>{mFZ#ETvaqu?T07K!Vy?YkOTxHv zA%{qvD>@17+BMV}3p+qS>QXJ?gM*vSI+?_$H`Kp?iTfnFH*aqI`n?roT3wCle* zC7LNXMMgGO?wpo(F{O{ZLL7RiLdUBQ3`PZ^%u26<)}1zSURap+{8Qb_oSgPmjGsL` zpkLtesg{|HtQOLj+gAHlS*bb~EiA4*H5f)lFZGsK?Sl^D`-l4!sH9_Q3(_4{ogQ&|Z%QBoH8nm)v`D?d(>OwAu%=Y6U8T{nWgTmv*hcS_Zar1c@Blr6Bf!Ws?ToNb;<;Oi@$r+>C*CCF zGmK(gD=|8ODX_;yGb5Ow8|aAV&9HS$O+t_1Qs|QEAJh6aWlocFM-kTFc+)tq{;oO3 zQ>kzYsHyvk&S4SSUY?)d{Svs1%F`{0QDQf=>`T3N?HX^t+;L6%w0vV@ zq0iEiE8Ocvco2L04BhWu1ez{VG7S1hDVpb>=Z(>R9AMWEjMMFyeyZX2ICQ~-Jyx7+g_A@&v`_IENg2k z5j09&K7W2Fm;VmUaP`32P6-lwU?|R#6vbx><(Dg;9}cx{yrCNaSk3R@d_xK?=1P8 z&(+r-J79t*j98Lvt7cXZuyxD8r#JWGb!K0DdmFeQ&OhaQ7xUVVe*dO3+qn2{^fya* zGa>UHQ@QeQuio9h<17MQdl{O8(afQp06Bi9?rc?Rm%&w((LbJ@b+dKvLod^%OYdYCv^3P?)DAP{5i1){GZbrMXgYtxQ|e>7WChGw zpSyhF*w*CNJ5RzV(cQ~uDt86StCGbVIx<&bB=JGC(&BK1{WbqY9n64q5oWc9oifO9 zZz#?!6`fosDwoK7;bCLC>eaq3ZrW;hku90phcJVQp)f(R8Q_Z4>lI47^IUmDqZglE%OO%(?0sXc zD4?jgsM0jO+mX#3EEXJCQ?!Y-_jUxP4=J`6DKuazK7L>QEGs|1ehn)#rOEXwk^B35 z{!fG66q(4!DxG5s)i7Tt8QCSP+gTa$(rCW(F5per!M%fRLYakPm%`(7w=y};bP@%p zO3~JTbh7zwkownG#n*oYI~2xv}Kw~EBQyc*U0nZkjV!Gtztq8dqk2L#kwdC8( z|9ckmE$8diyHpEj>%H;ibgD5>;1r+UFsPfvR+SXat{oQQuiqFHn}IR?Ze#wI#y0Arw#asi}UcJnt!f zq#f7oX_bcKcyG5Pq=pakE|`HJ8hHo!ihzSq5El=je{>!n)EG>Py1PsMT!beo%VNyn zlPw#D{n3$YN}52hf#VD&;b3CljzJ>XRCn?eD&n^nhluV`RCD%}>v;{ulAY^s3PDD0 z;xjF0UCHX3b{nazvP1K!pA$#fTB{kxM}(l*Tt9@(w;)&u5oYlGWjd_T{0)Vg9)88S z$J-{k#U3lqY;kggOm>Ftap=GMjP`ZH+j(F`|5hY_%V$$-8j^aiN~6^1bcBUtmf?k9T_7OTA3 zyPf{iPc1{CzW>X;6w0Ayl6&_s{S&?GyvOW1Y{J7!?CL)8$8HH(-n4JEKrg2@D}wG* z%ZWLSRGLEEZhnvcXbzEDJtoa&-F>zX+R1mq4MwcBe`5leO)LZF^7F= zp^+t*JS1vAc^P%lj4fL7{tnr0U2BzNPD{&GL3^#z*83HMU~}g8JW`Cit7A3a=cl%W4pJTpvsbrYwK@t;Iab&4CSZ5l+dar^pOKG@l}$g zDQ&sFM?_Aw87YL5lZJQi-4ik`A{VH5DU!-)U2yzh-|CV4cSl#(J1dD}+=4CS9sYa! z=vqeWlPuwizZs7UdDqBNLY$n=;W6l)g&-_66t~cLJLlJ~5V+6gU|JZ5X_wGvaprE) zufpQq7teEp8k?*gPDq5#pbmENE|GtsItyy%+D&wq&ToLCVVLNWZJGTm}|nbaxTKvEB(=pNj0<) zGY|cK9B;I-^y>k{?A9WS{>m`&n{9V^v$3tYX6Dx`ud!f0Md7^zMv=0N#g&hmKjkh( zco1{G=h2d0gvfl#Cwzzjk*)T$v%R>m6l3&^Ty{-1Bg5?cR1qaqR*%^P7cC?neCsG& z&&?{x<$L`)r+3Di=6?GOA;}S-D9>DZjKBo&_5*7F15=#5yOO%F4 zNj60ybeUBOPmOY2fZ3R2=8DX0t;4;Ar&wA-Q(^li-|aMGUAzDc@zJsRtU?*yqe8}; zVY}PZpE3{l^9@WO;?L!{V}2+(2Q$=Xf1aE8C37x2yRT_`yj$j~EwVw*zp_yPu=mdT zFOT^6c-3CLWkumgU0fK(6j=u?E~EL4mmL>BgCXqCq9YKesL0%%!&^}DzWSM;DdvXV z@OP=L>i%?jTEWn*@_JO9T7zKaoCznd`m|NmhAZYWw}r3f5fsN4yCTdXXkVDVcK)2Q zCP!m!r^eVs<;M!N#g-ETn%j?LBokh2mY8DlzG@_c&{850lqARD_+2m8Ojg+^W=gO7 zZTM*hWwbFZ+CftYLwG~j=GnJ?jmo#bnVkL+8*&n!6*dj-Y<=?cNPB{rO<9G&y4KHw zC0#b+ZYjwi2y#ID!5fGd+SK^nDPFws2m4X^FeYVUZDY$NRhLjkUOpBE!T}3grg-U@ zuc!Iqxf7hgXU1UhVfA! zk@ExQ!g9hn`23<7`U+*?0ISiOp(g>IRF+=$NjNJ7^KIDc)dMCc_mU*>(-#TZ4}7B1 z=9+EMn4ENbX&S)yI%<@amab16K0!N#vAn8_`H?(rxSxYt`BJ6y$0sLK7-;A?ILh}_ zSI&C5-)ar_sdTy+O5p#gkt=Wp&)^rUD`iBOS%<~gQI1AI3R)+esr4y}5pGIUXCDX; z8%*WO61$Z%sjt2A#Nm)Kitzbd@-eeKg8Y>M!0{&=-sob{o5zes(#CsnW?-cso_oF~ zK&z(Es1Ap?<3U?Xi=`OnLp}j|?k8K@%1sCU0Sz-YtB9rla23UDvZMTAZ9=`Q_nZT9 zZ~DzgtDP)3#CB_;<1*z0gwD8;rKHT#zjqF$wpN*vTCUZ_mj;v8JQ-Sv=W!unv;#@| z<@I>(l?0g@uKaVVOafXwpz?&5Y|SMzHs=Z3j1xZ6e2;i_m$rzP?YH|AhNCa%w88 z4Q=)1OU25FvXbw1k$x@^od|R8b4`(9Abm|7R&_F7pDVZ0u;0%<4R{?NrI0Lf<$^X^ z1Y7d&_cfTM!?8ZU3dxp-Djc>s^8Qu zuElp>Qm!Zw=BX~>@=7LU+83#ny+$S`(vqg>kRmKA5-Z#1;gl)k-jz9t?<{(z#z6)ylVh3A zTa974@x3tjRb7}{i?VFSM17J;$Y7CdaL&>*XavD2su>(C?p~j~Is^LxE!`A?`z+%+ zUQhH;vICX6zJfDBj}%?c94X%^EwE~~uB{t+EK@o-ToRnXpv|Jc;2N^NPSA09_Wpe| z%@1rB@2TB)PopGOQ#grF6rGGj@rV*09QE zu{V=Sd{}I>+4ntk|Kcb}Nfp0;9-4Mv)lqDKgc$hcl}^O;qjH*rE~lD=^5zr-(_y+|MQ2nsHZ0`=z0RGtux)CW!^_w8MO+cmey zl(CR!uz5m!?geN%Nq^Vtcr9YL8u54_+wer$F-KdR^X7Mnj{1fM>8BIEmQR#2w;^p? z_xl*>*psnLauB!NO1SRvT>{S6G)0}=?#5Pcs?8KSq4Lduk=sSto7>xW0+xqP?!%0< zZEybY{@ug5cBKxDlLPrZ9?4DH^>r@@qaHk1>`*jG3;g~2L&rAf@9|u*hr@CDZk8D=!@Qf9vd8{cQe-h*dpKsiwemsU7W#b8ggcC|X$CmUlzU zMgUHQcUR&4XzZ-|0t+a%bPP&b)hDa19c;wM}d( zzr1X~w0xfZ=xA(<{cEInyZbQLt&u(eoKp)#KHtL3=y;AAz6hPzaEbIL`>%ZrrfGYs zL-nA*d3kj( z@i5s~49br(^!%j=Sg8rFJ)xr`YRkJyWu_?XW36!jr6hZeA{YGJ*0)r1nu;yS8f(C+ z;*r0wJhFSgHR4Ekb@`_7`RElFU-{L~gq+U@{3{ZwW%}%s7ln&T9X9ZoStCr|S-Q$3;f+AeHlR`|_>L507NTAACde z?*_!p!)AuzemJzuOp9W^2R_-_B^@WbWB#{B&4q+80Tu1Z$il;$uy&?|Yl1pNuu=hM z@xajSz+d5Yz9kyUqM%q~kat;*#xZB_{IY*ufGY>8n?w-n8jUUbGG>sner#aJM z$(H9R)ObPa?`)DwGn0POSoNji8JtqyaIb9upgQJQZ@(sXsh7arJfJl@;-_(UuYUWa zc7@ms3%{3)VLPKeJNDnisQXc8sB zE!?Fr*rF)b+(cD2q;G|rGsFY6b-{1HgL>nHA))Djl)@8_g$P1`Xg;4&YjVsub?t^L zDny5h%3hJ$9vF-HpIb3g4GuHI=ybO3hED}TeHX$d1ILZY(2Cj{2eyah7oLAGPYHi! zkA_kxw3KR}fw_)5F%>qp#qGht<(ExKNEAK)+-ZDctKGXS{X-n99jWYyv#RMqEza4Q zCt|&kg8Z^JZ~KR&u^+Yl_z^@xK>tEB>?{VroXc#4HC+?{%9$daNn^sJPI=*~CE?s_ z&&0s5)_3krxv1`>8Meo*=cWKnVl#wg@mvQ_btBf-H%WQy@Avd(*~;qKI_rA#JvuIS z-@RmEXwh z+idq#;n3-M{@OUm$;s!%$X6vLx!fo^_vLU{k1|CEcTs1t&0flK+dH)FEb1}HZ6)13MP8+4>O2ce7 znZ;`bsjy%O^A&LX(6Olc8Fo3w|30d6D7tHWmO@V_O@tY}*J|%^Gxg!fLmpKjetyRJ z)f|()OvkFvXH>cy*Kt@>)2?D^Yww6vVF z!fKpauI94!f@}KmcLBbW-vG+aoBWq0jAkz>O5D84D7eWwzrg8udBl$iF-X89KDD0Tx%)@HR4b{YiE5kB|3-l}Je2bl(($jW6)ILZ}Z%oi>$cZ7+Sw?N_Y zmnHdJD<4u5>Odm6j#MrC7yD4IV&JT4#!o*huN%PLm2?^WMzrWoP4O*9{A!3x@`R-B zGUmV1`(|Z2e>bM-g*U6~@Ei=~Gz_m#POv=7w=jU`RrRL1zwH_^UIuzPYc{UJ!|IzE zC%2M0-RZV?w1|kRYjt(LK6X#|+`lopEoiZEDTNtYyF3uMVxZqit8Tv++w^_(OBE@d z*3O6Yk#k{zICB^%8UYV6n%TaUl|Lu~fkks&7a zj0~Ogk%1sYhbqT?0x!o(qNlVi`|dcaT?~A>mUl14pQChVLB%(op)4-;(9b$}UlmFm z7_61_Sfa9xl5<}+hxHb=(z}3+0$nX>5xa>=mSk>vX8m)sP|10<(Uy~a8xu{egGb|s zbugvkNwd5?RWh~N^2T-DWy^_bTsG^iSF8F2oOUuInyuKzBJ9|`yg3;y0z(;SYP~PyWe>#ys!ah|VB^Yhw;@Vi( zCF0Q#-R&Ar@Gv23%nP_~DAQl|Zt-?Yty0|LDvvy8*v zo6FzJ$4XyDz1we+9Me?<5?RP}ibXfu!xnp)R2z0@2?;nCZPMk>bdyX0_VI?}&33cip`k)SkVhXm}2*72!QB$hi z{i`{Uu10+{)KT26qZxUdsBu>pl_VQWF_^bV_JGsK=o-X}VvE**Mj_)1=WaF0rA&So zo?tQ;%+p+K|WyeZ7?}epAoS^$!kuw5A=d+Q%yg3!m8B;PW{WUofsO zJ;skYmgsB|wP$m;`7L130lEGb_-Qgl#^m3=F?U_$ zzdzsm=ELF09T-BXG%AB%j>c)QcWbrm**Q`^&zGyx7orqPgw}R4loOg9%gWg6OBr5> zL_mZf%wc;6^7eRxsqT4f9F7f!-}`;5@ie;ckSf7Jt>I?Sc7O45qS$?wA0NKBlS2<@ z%{LRb)J$P{kj9Zr3JmCQVRYFKYY161TMwQ;#-SIn7Pl}ooany_4dc$kxlkSrmWP$d&@Z?kI_i--bWTc z2WMF2OTwAopC@*H({uR5)9i?e=O{U*LBq<6kTfk4;^enxdt%|c&S7?P_XpK9YS*p{ zDQQd~P>So<9RR!z!?u4AUXcw^Yj_%VajYWJ1@IEt_9x1ajhLrlDtq{FT3s~6T=co4 zY)wSFyWl2$U|PS|J`wW7!lm`|WfT+ygttiWOlIhLeXPzcGJPVYqx&E2y?0Pm%l0mc zvc)ZCB}p`pL_m?8F_KY`oMjOtErJ9kE0QGVoRhE!5*I;0L_mV%90f!&NX~G_T)59U zx9Ze;_gD4m{dK#nU2bG{&pEn#jL~C^{yt6Xc2+3=R#jz!0Gddj&&=PuugKwscb-G- zeMr||T?oH@1Gb#h-?8*NBkx1K?;I?QdGfqeMRj-qCn_9?>dmUmYTLMXb?W`%14BYQ zA&15G>XXhy%J*#@_aEa~@=O-s@ugL&&DW zN*I)@o^xKhJi!@O6oRUa*c}e&=#aCR?XrL{a&7pkzK;6z9ZDffk>trSi1IW;@6#sP zttX4xOC2HY+dgVrZWD`yr>4r_WckAQ420spmtmysg%VInPprd^EC|Y zqqjOY+3MCupRP09^lMuAy}bNrR4DgUBX;l8u>yVPcWMpd<6{RUSG)GJ6_k#}fKblfsr<&%ovxv1*t z6RHj8_uIc%KbI_<5$fk1cR5R;gsQ3Q#!E1|Mq8{7+l~vZ?kawHw0*by%6$kYneW$e znhY0<|JDT6`suKm7Vk-=ci$#sF}%by$nGqjQ82!GJHS#B z-eN)DEgiWhbjhsF`<#>$l*0Hf<<2d=ZXeIhGphBsb7s95SovqFPry_=I5@Cvu9`gx zz7M8jYc{$+v$xF28Sy;&mLE)aE!Vzx%piQV&|STyH}|-m46dL5uJ#C4$d%+Lv7YJN zRWKPm2)=re7*e8UPsmip7DypD{Lk!>3z?j;S=>>0>a_RuIi;`#8K~OdtHz!X!pI&y ziw8vjpfVlZCa!hrL$$K!nbgE8^MsLn2{F%_LfbmySiNtZHE=OC>k<{_o(4EiEvIKj z?WIsp_XF}Sv*~_(^>SBdNB(apQTZT$&C5>cL-us=O*|gwu`}l=hn}S=@~u|Jxq0*H zj}lUGKeAFU)}eE*T3)tztH(U1%(mgg9Wv~Ekhbcm5tGz4_m??Wck%pe$S%L28cdX} zsry@N{W2(Eg&b4c>9#9PVObclS3!ATQA+~El`E|@B3V~zi3km(qJI1JeDw*bbCV*W z;!vbe>g7`|{&|vl51l%_3(?pY8bE$U+Pr6_nGxEgV_`9ldaC>-Pp5ew&3JY{{^lwl zlB!|17eXA)-W6(VLC!EL4)YJ14Seg9-PrDj{yo;SFii8V&NZQf7sVs>*?gd*ns+TP z!ksncU)ElpYu7+Vi(tJ2GG$ifda3(9qRXe5Lw8_tU5gMB%vw8fHcPKbQMuBjqS^qp z)V2IbDsr!Kq={6kzG*7f7A)p{yz(I2a#3`1ks{M|`fZgTri1e<>G``o7!R;+Y>R z(rDY;g=!%vVlu2aR)oche(Up^?wWaIvNLm1z068>>heW!mkLcgFGAQyV^%s_Lt)HE zSz$ef4g+_kzbwq;p@Zqae}riJUcU8Z$mXgW>7;;LFx_@=0&MG~T(sh#*C#VfJ#z#d z(9TjJ&KAJ`KX2$TD-z0s*zu#*@ZBWPVVQlEQJaEcth+MCrysMKs$%QU8#`6^jyBgp zZb&s8Zo*&K_p3bv)dsGKoA$(NV|m(I(mXFSx#-Fz<%zda^-XdxN8UlQ%%y=B#vtaF%24!an&55f0w>B9WRAm0C1$d~dnNsGCVb_g) z;G;gJpRoz(Am;WvElErYv2}2?eJOlM0%d#N!{yF3-Nx;Fxfu|J5}BJCOjc^G{Gc)T zC|B7xQGCK6Z_y*WuV_wberFXcr}oyyG`~v25SXY+H>8#!7w`>l# zDZrj(uL`6*<+-ty8b`KQXI|ll;$58lb|>q7m*_V8jSBnC#}bXf{aWX!1-@O8lK$3W z?NxGd{Q_Rw%Mxw@$J%r`EC)DQ5nPTjKXi93`lhByvYaQGA#!T5%pDLJ`34HJ%q`^- z^v^N?eE?GGIs5UfM27rzB$0C}kCBlPxpoRJDE!jr)?TfRWhuD!0)PwBU+XlKq@837 z(k_3R>+a=kk%a6c)ZU5o)SP^kGr4N#e<~Qdr^xF??8x!dOF21cbBk>J@PV93U_0wY z)%NWUUx4^lK%5he?X#sHEwO`{B-v=|p|M|zq)@{}B7JA(20J_cA?1Ka*&?-_^*ZCs z(DvVOr#tDvZp(*f_1WrdAtBMx-dpdK;RRMrW7RTP*fAT-ZIk#*$*qPq*Dbra80u+!Tcmb6is1?S{cnKtWeE z1bapOsJ}v0g<%#hZFxE}rPn~fX^*U-LA)d~#03KQ;O!@&HJ-S3V^S**DDoiOCwC~_A;CGG7sJ-$~Ot|a)tO~@yA}>)$d7Ks;c{V5&xs{ z<~hW5g{yO%VNaxd>Z6L9*o3S~ray0hi*i?D2v@TTR|5F@;0EwJ>p7)U`m^HF72YRK zPi1KCZf_?iB#3ZgGPuCWyPs&zrJVI}lK4W~B`8-syK`l6(fVweBa}jC4rTJ=&gPtK z!HGP)H_yONaaZ*L9RfRaE#L$b+^VnKcMdC4H<@3T3qN&k(F=^6*hP}SpCq>l!*S*1 zNrVtQeFIt9Q2sO5uV?pFgoydV^*hDvt&19bCzW3C!^!JGpGU#dR0(3Q>;IIsztlPM zA!Mky9k^an*U>-uTIuLKz9h&W7-fg;S_k8F2GWgoo765*oJQo=k$}5Q2N>LTOpFD%BxwK+gam zSxDpi_=)J~Z~xKL9f|+)6+s(U-2m7pv$RwN$HOfHSa3+Ql9>JHWyBu-kKPr43-dZ9 zb4!1bmky(YF9UcU(x`&UtXAzRa75rw|H$j=^Z(In0!%{V(xzk(fV4=-xT2Xgs8!E7 zftWJBeHnGVBZuVi{=*=3tOD-0G~w=mn%qQ zKQD!nJ?7_!X|n&eG_lKs_xZ;UwUZwubgG{(EGsj$%`!?{ew7>SJO4;x>M@ zRPEk%=aG_nD<2ToKDnEz2;%(wqpfFPdq0GM-EhiVwygS9+_hM(3)Z1RmxBSz^S9Bh z{k(d^2SJ-!`hjE}BY_wr_3H@$CU#Bg{#BKPd&fyT>f6j zzM|vLot>p5iE8@_K(Jr6)Fj1>g{PSxO0+i2O`sDF(;SOdE7ls`t8VjBQR4+9eX0H+ zWC!eMj^io~`v8<0h88oaMRh9NEkcP)(_XJw9~XLbo!;X(-c^|-*_mw8jhmMix0L#2 z>tAsseSPU^aBQFb`Rs+WR3mv?E%%J>faVKe24b6^yzoq>Byy9 zc2NJDmih5?L|HZGEDiV9oR@?#d<$Q$6Z`8`0r2!nMqg*|PoQ{$KB|m2G+4 zkd|_*zDndXYvw$JY2_~ZT1P0FGE?aNK18_LROc=p-nl0ae6mBaXoX?F`%zk_S-0Y7ca4gAEJ>oh|ECFOnQ3{6C(H$OroR*(NVZ+MZV@nUdzAV46 zYB*Q+cww|Xt>;&OgX#0_?Mgzt0K6D9W@v`%wR<0@iBuGQTtG%r`;315dwx(ix6!|FzFYk|6W>$9(^X#BsjF|3CH_|L8TWBnCQ60g+c<^yvTb>}vlfa|KIL z9&Ry<@8llt+Ca2AD>TJB030GTG*l0gq z`P?X~KW@`B@&9Qj%!ZO!*f_=jqFlCIM*%n%^aT@mB47IQ^(5!>YIWGy5IB3hGn(S| zc7{MLi%4z;=;{*}D|gWv4$EK%j^nL%_4dvc528B(Q0coL&Y5)8>}bLHne`K18l%>| z8VD3U(K!!ksS|j=lq>9t^qV8vv(5QuQi81jgFQL|LHx!exGl*+?i)olE(W6c9?32m z@3^62GM|ab=>WVn#`ZU<(l!rdziVaJZAQ3*&PbrH~|V7RItd9DIoO%oTsd{wYAAS0E71CI`139y~?hyy2z{J z-E;$mNyd!eaR}~ACl7CY{PZiz*lP7VcawfWx&+9L5Lrc z)X=4ms`eMmP;P2~{OhzkI69jyynoXTx3p*jLgHt*3|c5F_BRHDQjBve6uF13E(=E- z7vnDk@hfq6J>(Q#<^=7Z!|6fyi-*nMgDX3(JaKMc!O`oSg)CYjV=kzFrx4 zKYlX<%yzG)y(#vKDWGBj*!IOgV|{?J=*3s3T6=@&Bz1b>cwo2QtrQv-rmw?;*lu>%WPK+?@K6F&*@W!%60_b7(lF}|4bBZ z#0_yVwal zIWhqM1Wnv060NnhzvP??P`%a^<-3dJ*06}9L;eLBI zQMDBIc1CV)FO}_3BF*7eeH3yCvcO9T(7y9wdv$6KfXF?4OVwG2Ur)uB{r+-VS_=27 zcgVrIpTW!R#UK<>O<^o~c?gO(;06I=f7YK>vHW?Zqvprz_S$r^f8*3@B-II#OWQwO z&szZ>gUziW){-B`fj8(K+&_aRM+7UV2Bu~W0H%z$evd^7J68hYbhF=ISKqkZzlLls z)laW1Oqyq`i~w=145Tl0^fI2@5?H5!d_C{V!~G>&3OMF=0%tW^?mC`VR%WII?A~_3 zR70TvdmZT%YDCSBR?}yZ1S|eCK*`v3+iCfu>>T|3BY{?tlL z2QdM`-=FajB>X?4=_f!$Ex8b2boc9>_dLkS$#qZ$vDRUW6`Pv+SY-WuzN*K~FT!L8 ziq*QAt^UQz&j!#m7ew1%$a|BpbO1bX4*sFjbL^mX6AKohil9%&nU@H!NHl>O9KXRO-;cz<;)KQi2t`vZSOpnC&%C zs<*flm6bo#Q$tp&&IcLc#vZV&bRi5t<~%q+^Rn$<(bj;79}#n;uEnzSPl|B@CfUyL zl2iIzPB|-=enG#yvhoYiS)!QK)JH%o@9nf%K3YqNg#vekOM?g(k#V7pm?05GVARo7 z`!>kW5kbScr%Z%TL@3WPv$A}AV=WtRMgbTX5<@szYGU#sXp5eXjxW?tL@wDMY)>Kj z?iLZDcurxE)z5~vmQ+-vfbiNVW+`8!0C3~Z zeIUb~=}iBle*(_*b$+Ks*kI5lV+4W#ME(BZb^ZEdZMA?&Yu&A&WPNK$hHr*e@NjSS z&zCqQTCrDyZAb%EcoU2yp!6aZUE$d?lnpapkfz-X-TrS6VR;} zKxEBm+{&Tk(l3krJYzMBcmZZks}yis3iyF2C?F2^I5_y2YbOAha>52gyA3Ya)3jRR zJnzpJQEAZ)fZES#zhUA_#&#cJi0HR)xi|H{sn?AGW9v?7?uLJv(ST;g;Y$zI)S^M* zfA>3fhWw3)2!K0xY8|YE_bGz`g_s%U*%QI;Bh&065BldXi;rJS!OOvq`Af_ZAvyt= z<>72|#AYJa0W1_Wmut{*AEe_DvpsiGcU^M}q zP|G03!u`*V&hYGDc>~tO9(XK6|QQ@sK*-)y1>wH$!$ z{`2_z$<$Nx7_I$GggOX}SbZHVt3>TPu>7xIRMV8^dE-a^;oLfC4d6- zMOI+}3hz|ze=1nSz~i46XfyrAh=>($`Tk7U@yyIjqQ4>B`t@({r_y#|KzT*QG-!Q& zwpF_t-{S<;&Uw`QnX>A49V`uxl4U^$N&-7nH{xu?+6=(?`nc4^a~?&2TD(7Y-Wg2N zFnZW=bC_@G)9RnCN#?^NcC7(HSZgT=Bn%B_ zWGhA$b2YQCu=c;HIgFljl1cm}|1 z0Gu0!uk?|g@pimBSWf5~7|aEVQn#~Uhpq#I4SNgE1B5s5{H`?|e~En`TWd#G0anC| z+V{kd>rc>%SAj)DEpdC-S}ND7IXYH2tpvwU!%pZga>}+3O4fE;~WveaxlQ!iOL8M5}E;Y4MSR!`4#t>H$G+5J1aK z2%R{9c6TBC6q2_qI1UnVOMpiz1nnxKp%IICS=$amh^YgbaYax!DG-O#veEWu5r-J% z*@9v12!^;<*NejktM!#_SVn;4Edlt=&mF(+M;u4oL3qcfT*&tME{{t zId~&$J9D{`e4jSsXo0ZC82qz8nw$27e%aaUqq`%Uh2XrLr4bS+F3ipCm~g?bm?T5q`-sTSQ@>Zv6TonRYM7mPNm@}S?TkS*nU!gkXDgwso0Hjx+ zv-@m|x@1Yr8)upDnjg+^d0GkbBv!%PE@>1Qg#x#U$qUw$O1IObOkj24!!&z$cVGTK ziCe#3Z#zsc@Wwa}o!f8_NBGDB^8&^*F_`rM<6QBMi>=E1cP4ztaqeaSGrWfvzFuu0 z30fn56Y)905p(}}e^%|&tGxS(0na5L-WrB`6(KG2BMO$;oLq z=0-|u{mI$ym*BU9Bey~q6yCVs+K+b+5I?u@c@0607HNha*kr=Lj=P8I%`f5RPK+c3 zeRHwtk;lc_`Zv9=w#GdCcZveaK$5xPTGv{0fjEo$6M_ri#B1YN$j);9(-Pyq-i4)> zO1Z-};lp&EcHXmibUm z_+x-~G}dkwQg%iP0yv^OJUmL`^8RBNoPjYjY0s$&>?QmGqMHu@N;(VuDWIU*%1um* z-KUKD!*6h?je!RO_!#dyIA0WFf&<86QOW*_b?@d#_^5|Xxdz;dLRbh9Tu>kvK+F-a zRc#cd$%3z8w$YOVtSWm*JeP6*SHTCs_2C{{7i2Fq;OsQk4Vp%TheA07jv7{@ljbcj+H}QA`304$ejhOZZu*nMDyN;Qd4NM|mLUGu~{=7S$dQ2ewrTHcZUAM@+LcWN+A}e_m0=L{Vjd zq#yp99YV!pH}>>lpgH_0oUmZLuB9e6_9M|iWq^{zH}YruyNpPt$e5ey(FNMLkMQ#9 zV0o0iYY^L$^+CbW`gg)+A0^l8yBaG$OD1J(zfD>+#*UGF4iQOzdgr>c@zcs4mjSgs zBjewg?2M(P5?u**8Nn(Mw;P=VjC8pE_8-iI-G^LD&jCi}R(S7P8F%z|w1Q2d=oMZ= ziB_55TfPWtSgaQ+M%Z`A{AGBBwQfiV;K!Up`f(YNElc(MZcJV{ z9x749Wqo@v%LbG0k#@^KInDiN37w+nbuYKK7daJ+Ge-uq!tAvC{rxH9b8@w0v$6!z zPtD-oW%?nQZ6hWJJCtGRC;M{J3+KW$+)xFf+yc)?iYx9D5^mjJS^V(K@sW4g!A<=l z@y^zkPNhmAc^d{6cBF&Plp$zwrXSwZ`&9}?!&S58#OFdx$xP0Lm{}9E$c5*gc#5Z81 zXf#4iaQ)C8h}zrPyPSLfO-pKDy%a@~x4vIMciml_Pav`f&z5953hghw)mSMKM_12goeWLKS)kH!^wN` z&MQ!Jz6+Kgzw|tbE^>TO;N^w;^#-4g@3iq|Mm>ux*?Dt$3`C7aMS#66@)#XOST?`3 zu^~xJ=#fNEk6Saex1jHL4`f-K^RC-y++mVLpWiJx=B3@YMcH4)!&{QdmU4dd!=GH$ z)hqDwKPIr5-5q2>wEeMMXDQ-Q5Q{&lY^vzXQ+(B?Vm4cqvxh)R3Ri+5_BaHM-#kj) z(}*=f^(NDkelwcD9uuSOf6&QE%Pk#5uhoGpd54X#c+G442FC15csXxvVxC@*pNItgiAA?mYq$oKFoUdnKY2Lls$$#Xo`#XqDv)_-`tRy#Y7`P5f_296wX|V z(c2W@pCK80Dt~KqHs&U~i9?*1VZi%hlU?O%?Ws%OIIyC06EYlZvMZ8J=1iEY8ou48 z?;odxn)S}cNP0K>;WBt)$KjK82X`KC5Af-imUd7zGKsPCUX&B(q3j5LGCE+&ToasJ z-l)lTQ+Ae-GCa9%mP@?D-G6SqTar+(Kut>Ks23~d+f-N8dN^vSNqt)GvYm3N35=VO zv(vAa`<)=18LOr$C37TU0vPV<5APSfzglRzP|*eWVubyVV$;uUM_aY2^V5+hW*mo3U6s|d;oZ|6wUIcozI@=!m~Qw~Mt}HRu$H z1u<*KVO+;MYohzyxp_Vnc>euIB5W(yG+c$l(%5Z zH|eNql6TM*g0hBxxEmU!N=5JK$1KSn|>! zJx6sTy>obSVqxIZ`UCip+{>5Q$KTjYPWgAWz0G4H=X?D6^$pzus_p!7bDpf&@?N8* zN@*jLEJZc@;JX66jZIWJpN7WBIahc})#r(MF5I;oGRj!mipm)p%xSMTzk|!uB-ldq z3(_y$$`m2A{gjHl8hcBASW|w*xguxgwvqjKs+ExrL(-Y?(`Jf~AIrpVW)J@MJ6K!U z&zozipNhm- zve!>mv`$L(ZfotfKd<2O^RX8*=pT4`nd2xmb+x_yc&+S9$EWQt>QU)~&(RS?q>nA z>@y9OImHTiz+cE7d{1*R=K&OGt+i+xol6S2bSOX)O!Ps?CG&9-PZxYwUY^m#Q|s7$#S zgVc5u;mYF;JAdJ}wDUp^8Lhpi{&G;=>S~qODBn!jB>L&wwH&;Auzpo2iwk4o=O@fH z7*#{)v7^*=G&)V1R;{e$v#GczHVpgv0-rYg846KTQPdMBUf)=+XlmrJXPbD<{k1&9__Z zTW5W8)pH69N4S3+?&gnT^6ivzY06>#=28qvt?Y9~N7Jfq?^YaaFE{PC_#~rj`-sok z5Y4jq-qV}pmHwkN9&+5XAX>9+d7Ib2rDa+eHL!2D=}IS9_ayd!Wq)rpeo?kWJm|ZA zZ>fP+cKqSbo24Dxm!0F+U**=09D@)HsUxcA+NBysE>VLSk~K)E z;$}%q7m>j93Vvxf)S`*co+8-ad9{uatjjH#5BFdmI>#uuoJ5@(0c#*FZQy~BZP&L6 zUF>GXllWSl_EXPVm)J(yrWe>0H7b;%i`SUGqw@V!AUeP-*LAxtSx5^8ICv&GV(g>Zsf7 z!^Mu3ne`D(kvRpepBg_h-*GQ8o~b)ZxqrTA9&n%1TCl&(DBYU9qh)x`&ZoO4e6BX! zelfWx{7LWaB{oHc0<{TA8NP87jkC)?H^_|98FOqKuWx(Y3yFIb*wM*;w&B=>^1Mb^ zIDraq;zFCIg1781pWi1swOmsgsH$qGb$q9NR>^Ot2>;DSvXd{qKwMmEWNpW4rQCl1Y+mO^TjfzzZt-%^aR z->h}EBBrCmowD1Qtai$njsLh^yJH%KJtuDV?L;Aqp%EVE)@MHa^h#U|H7eWmY~#0Y zT-Uz0=k?DwHr-S-b$pFoF0;$6^W&{fk>`ofPFOYWADh>66++~N_jRSCr16{8SL%}k zjKe-0w8>K$2_ucI(qcM3UlLLFKHc2%5dnXBTzYZ7j{1Cq&6))R@|5nsAnf+D03`O_F6XEvvdCE%rBvSPOb}NSWPV!?U*!GNr}0-o+`8so9aKo ziqtzoyM#=!-Y-7gZJp~|4Fbw5_V5W0aB`Z?BO%A*rdjGG>GSI!DaQ1f1{+Ra`-E4hQZh0UVM5xpqJEC=hlV2r=}yP7Q03qqbE19wL(2T zJt>0r+%ut7!kME2EBK}(8;Od!8fmfRJ?C`SG-D5oXx8~Et?MPe#~un+Y_lTWHFk`v9_DJUDbZb9LF9YdonwWPiua;?V=$)J#InL@BW^tp8 z&0EeX4R7R{Q^WVv6}s&Uja$p8Y1FXGrrCpAHuD3MQsKgFJrhFWcH5!bJ5}7bL3xgZ zKh)QcV-+|gs5tobMN{~ox9X5}qa;`B^!SZSe&2e>!)aBr<4UaD99fHafm=u(p7k) zxB{5k?668_7qfQ(nNX;M!eqP zQ5b%G^TrqSQt9)`nz6;n7R@;0xyjOv%6Q{~e(M>mQ_c*7<_GME<6{aTHH$X;g1a8G zNx7pR4C~|5%bnK0quGB5%zSF~>5}i5>wAbYuR+ciUwf}hs$%=T+n2-- z15pkuMJFI46hU=svQ*s=@dSOzU&r*z`n_)o&B;T_+~KcFP=wREy}Kt>ZF(I7S{3a#jC1Oga&JnVOy_>8#u4; zk}Gns=EUFV-kY1Kda=0}$MA_fqs%5Ayp^x(FV)Qtm?wXJ-Q4evE?fE4H-VBYIJviq zo426zse^dD#T@*KW!;O$itT-rDp~mDpDcP8i&QIYRE9K#$$f2@;ZTfy0vz&pw~a&4 z9)@$p{yReyEsXXXrap6X@vnD?@iogh6^m{X{Vf>;#gaS^aXY2$c2*$yRg~ZI3o#kd z`YVL*6qMQ>JPq)~{vzg|)$vb{jg027KZet~&5+ADrjw^?)%bc_^&m+pNWQe-UTN(E zg0+)dF)`1XzENDk_MZ1}C3HV-Sw3!`pUfc0o|7B@9(;GjdkT9_OLQ686)H1^55{+R z)mz@g>h-8D1?SSiA)S|*Z@Kx?vHd+{MY#N|ok_)Kq4+Pv)6$m(Nu z@5GO93(C(De+}4(cZeHI&CZ*V{<(e-z^tsv=VD^xEyGv7is#lCyH1~WtSx*aOeD*^ zKsmLeAaW8Hfu0tIlMmN|j{vVam88UAG=Ao4F&tJz%r;Cj{RD+_IMm{pJ-iH&_rPm# z?=W@Y9afCBb8?z5bCAsblIRHsbO#lUM+K@B`$2pgmi+r~kN>H*3p$$`eM*LpU#owc zszCq#n~#my9;)#xcmxuRWbFqmMR{|bm4UfFCip9k;%M3@q-B(FjRaZbr=pS$_(;Va1;{?Z+!wS$fO z^UDNex=U$Bp=NW;L-a9s9!@!%Mf?fsWPCf)bRtirt&i=`A=xnHWqOy-B0ysf^>ltQ2pkx2>!2xu>yd$iv8%)W57prO9!QL+0wCeuC@BIiyzMr`+X1WH05Tax^pKgb5NnET$5$J5lZQf_?#}JOc|K@{`LxBWlG(&-QXdk8GMosM7q@e0v6JstC^DtS?@lgSjE!VeE zlKG9iW_6Fs4YL}K#lADa9k|^AS&+M`BzY#@JZ zznPd!yJwerZ_ZAmKBofqCZ&(n?koKq4YZ^8%9MW;sh4i$ zx^@le39lmAN>cWo^TYUT!!`p9xk4poKVehvXGv&aFulhhYx*E-c@wKg6McCNpNk<| z-wB6nZhOy9fM2qAtq8l-tv;9|V*jCccR|LRUb?9xWk!XZcZB`3A?|Z|I3SyykjM2w z8Z%e4JVC(OyQ!(OR%e85)Guh%Dfp&e@XaXMvqtv)W8!n7#HSuX>;?Wpw4k>!Q0)Wh z1q%iP!!IGs{glleIG>4@75YodVM=g>OCbhom<_a9XfBK8G{Z6ahTE1#Utjq@rVdV& zrK9E@CaS{8>8HpH*EfamU4>`uIqp#|vrPL&A5LXf4&-XPU{4VEF?6QI91pi7jA?AK zWy<}Q{XDtj=Z5)Ft;1Db*|J2XF>U!{&dSG}<*P*=-g8zSt5GKSo6@JoFto-nxKl5% zMh`MV{YP`s_2A!k*KX3|gbN$x-xmS?T`B$lT=)F%>u&$YrPu$rLqy#Fwaowja~Xa- ZJpWr-)bP~I#WZA4@kcTbbM9-u{a-{_0=57E literal 0 HcmV?d00001 diff --git a/docs/docs/tutorials/rag/assets/screenshot_2024_02_14_3_43_58_pm.png b/docs/docs/tutorials/rag/assets/screenshot_2024_02_14_3_43_58_pm.png new file mode 100644 index 0000000000000000000000000000000000000000..65402f1e047a8b2767e3162bb7e9b9f2497ab761 GIT binary patch literal 104631 zcmeFZby$?&x(190qO>3&B?!_f-5~9l|j1 zz3j8kIs51N_V3^C`(t}u1F!S0wcc23J?nn%`(A{+RhGrUB*R2OLBWxG^I8oB<$($c z3aZk>2f#N8Q@=J*P_V@Le_sAQVGpFyB)4CD2FOo<_Q= zD?#*M3||a2-y$~$eTP$oYq3#w^6EY~;uhYMC}A`rASY;hnw-)>?Rp<=+3&$azqa-D z5S`J{$0)PQmj~zj^ryjYLz1~?BR98JNnVoA@K9cm{o?U?#zFE+3S~J~r2GXHO6C{- z^dyeB_tSngX!H+{UXc#JU6v3R_hSepKi{!F3LC}3_M7>WL$2=+pm-k` z9vd;DN~GQ^x3}9j75h{mPc}lF>&95T*VI@OBva?Fh8(;7Jqo@FdBIi&%*2=?g@6a;6q4=CZEd+1Py!#XLP9X0| z-SPE`ei@(t=UC2s+Fo+xrXs)fg2&;o8Su zF!yJJ6U<*RCPzQK>ZpmXzzqW8LWYqHhA~|@VVY}!9QPml;h^5T|K`2I{b)&I{F?Kr ziw+iP70zYUN4+Sh%im2O$iC=mnf!9k+*47rmftpy{{Qbw~b#1UwvnI z`T6M^Y`O=8wvPAbNiMo+mq)4iNK!%|xWlWB;wwxUsfP&;fj=edxaG82FPR2je!sti z!D1bGZo=lklD{2{u`KqMl6keE?w&)N!FvkgeGJ-0$vEhso+eSZ%B*j4JD$7LwyYaX z*4leh--@9TT1z*c+UG<%P~YMR8dz0^XXPhPpYV~Jj4LlN?(Gbr-e<}m!zOv* zpK+E&Yu}#=pp0M1m^^#k((oXwtN)}gSg6DI2YFLvh5Ux#q`;Ygx9aIS7AEMxqAiUz zjsCHj=?1su* zQT8tUB5orOQg8@;gf#BeIecV@eZnZ9??&BL`%%bi?9~4yCkCH`7<(Hf_`WYbT8G4w z&lr8iG=pM=u=9gBoVoau ze8+3LpZ*x_`E!kr;ajxypXokC)KM$`AbIqP=9~5lQaNcZ{KquXX4EI&xJ=P2p3MYk z&>|vfdo9;!#=_?%pFJ1twO%6(&9KVnv-+r;L90sN|EhBo?1Gx)uPhn%^Ky&8g-thT zUuO5`_!j8&!K{Ca1mdM%!H0$frVYQBFZqara*VeV(2|2E3AiQvP+dRu7 z1Rm$DV$a_zS$->WZ*xWcQ0MmtFVPml($3@~>)*6Zc(*d&39uiXqj8{q;*9$ynIO;j zJch7_5SMVc!@MBuF#OGn9Rg^`vqK$4db7uQgd~Lg_!k5!kLN;CfA|oDWUEv#Tz<=Z zqpv1b@UbeqD%mBU(D4hs zH=^?*1Qp18`v@U0~R1W4c z^<`{uQop!f;Uv?nmRN2;p5;wgDX)=r(G)3ESxAOIuX0=DO5}>~O2;#vPjyZ=IV&~C zAh%hz@m-^etU}K_2GzW`d9OC>ls~cOTIURm4viWVLL@G7_w(M%bZU2^brRzgQM$QK zLz-7SZ#+?()tZ|<`A@~RxlaR6E6?J0VY}j-Z8&?_AFy-DYMk4>564z8 zf|qz-VCercYRmt4wa&vj7AS^m`I zWoKshuM)bMD9c34`tIYVbqI@f(TgH<%`?q~DbFd2U7B68T?C&7xp*=Y-#b2-ZAgvc zoXs3gjjdaLYO+DDLC^vbWMXl_(4w`y#T)JY$IXYMp9>$?QcO^2(aq@oRs&SY{RVeRA=YX43}{yU#)3__Q8u9KQ;bno@(yt?4X^@8e1h+PbMAGA8cBdJH$RrfqM*| zWW_gXwZQI`wl+~VRW`eOxJT4HV|&_c>yj6dhzBzdQ~AWv$v^2{Y5ys`fC^UHe^B09 zFTUj4c*`Bf#u8SYe8$zrBthUHuq-+kCK~O{jE{x=h!w{MXAD#Rp)S=a=QOJTZ5@3b z#}77B&)$N!uPne@$3OKbXmKOdhqkd!9ww7cQIrrrBWV%4cKm4Y@kOR1wh+C|VE@Yt zS~gng$TL;yP{L}i9my)!)0nNSmp>A=5<(O5%7@FjKaBK@uF`kGLm?zo_^muQh5w|C{1L;Qw7hRDkLj7>z@b3L?u*+=J@LWteaeI=x+SVLF(%q zaQ3m|5eWmC;$R}!E9&aSes*BN$BFC-g94*E)p=$+Uor+L7c`=01G>_@j5yIcf}n{7 z5CxcTi*2)vHFA4#X_}pyUV-Aw&oO5@&N{SoGIWwF7R_Zr2ffdF<@#rmi=vW~Ag{CO z`iXf}_|o(R>J4l-*dMzq9wk+H%BR23uMoxSOa2D?R*K`hrkBk(W|SR)>4n9NDW<5D z;#u~#@~y@+*O1+8rsYUzK$uT*xXPxIebRd&ll?EU+&bK~gRjhThOClo#UM9(C4$eL z-G*}phtd#`a`mIoU$rMu8@$8(N#f={n;g$uCt`lezo4rn26IeT;TtjPPd8XhWV(j6 z#2zbf#0tj7#(V1w>UnKzZ5qLDgvwrc%k8RNMJAsWh^G)HrvJwr`p5n!iw?|Xa_iIIdGoHl^h2u{?Hpo;-l zHaoUYzU~O5LlMM!fhd9n$2QIOD?DpXMgOeP^GbKQ+8HJY<5>FKH0r~>vH{mywRfH0 zSlVxicHW#^2xv?{7^+vPGrE#kmAsVd{PfOu`lh_Lvw8J*R+}ta#Okx9-8I)1aMZ<` z0_-|(B5!>}G3}juqYKsUqkT(=*uDo0>nLobnamvZ@SGZ)I!2o^} z0H4?2QUCkw1C{Uh|N9tK>8_xpx|Ez8@LS!)2?Vlpwy<}ZtuToPnu1wsXuD`DDGHg` z+p@nmwf_KO2irQ_RY4H}3jv3=AeZ;lU|SnIXCbgC%|A*A0mpY=bI?%#qlk;OD2=w# zTWTqLClECsJ109QjTk01HMNM7shN=4Yngvm2mU8YW8vcBAjHAp?(WX+&dqM`WX|zY zP*9MAlZ%6kiw!8j=ImkT@*d1)=S=%wjr?~zuR+cxPL>WXmiBhkckRCaVDIW8N<(vZ zq5t{!U-t8dTR*=+338sv;|%da1AkD z0YQ;}l>a|x{`ZRiQd9eXYrf=r$@{NW|K-%5Ro^*-oTTh+for;m{cqR&v+}>5{AWcG zj=M+ymsjy$@A)6!0xwz&Q-tGxI!z3dg}}`h1w{fy?zN-__}<^?2Tn)2CDRaT=~A2{U88dTx5b}7kZ8PaKxa}zN84`3c!*zkaLpiAS(&Uf z{Q@8LI&H*!!Tf5_ograx<|IYx<~(T>UjpSG+Q0qb=05-N01}5EA%FL@_zxf9i#&S&Q~K`;Jxp`J`}gkyxX$x?0SwQv8~*NQ zQQ8Uj|EBNoC8%FW9KQG{N%D7vsKFhg|F&Q5PJBdrnD&u#i{|eNp{dki{M#`EobdmK z(jJZfz~J8xG<*ra9-U6kfG?y7$!a<_N?$^I*EXbH4>)H&<;P&ucw3 z*x&|FQV^&=y+XthqIna)U>=<2i8Z&JAB%7NIH*Ylv0|AU_x#n3Th(Opg~KY*bS(BNV{&7m{p-1KIg6ziz!{~ zQ~k?rN0kWb-RYEg;Fr_VcAn}xH^QxnKr(|c^^nY6-y~7|MW_KhiRbk&1@*7^xb24R zaKG_jRtit_iT=k5-f+=#Z5F+n;;*FK-9DG=B)&T7V&28ho5MHP#vma_bTS?b<99`x zIB`s2=D&V~w4Y{0h*{x~^HuTL&Z3=E%ra>fC!B4K(oQQT@}jAr;tD-9W&UY@N&{J` z)Y+cWSwq)2B@^InDKz@$} zK+*{LwADWlECoJlUtiz(gt`hk%xN>#eRha{$>#ES{bX~&ZO)d@xD5sJD}-ThTt%GD zf#3~jQ58yr<4SBtMhwr1H>~hMUR_oiLEzOC=qeqE+z1rZ5UeJFEe?c3Z-8;28eLSe zT&kL8Hzb|YI+aRfv|~h;twuq3q70J`>%s1rBrqh;N$&1zfT7Fdj~iQ)y9Z*a%V;QR z*kM*Ldr@QHGAix1=&=dXe&T~Thx1uYFp5iou1>c#YwZk#FV~`>>UqlLjhMNvgCpX% z^BNq|)vuVfN^Cw+`P3OWw4-YlsO1V%KX-$s5Bor->u-Wot-$f8qExPSFHVV zk=}9BcBY~$&G*K`SLbYJ5+}=QZ9h+_?o7e5D`Z?p-XpQo}tN?B`;_{RMT z#UBm*VcLs_T~>c4Pd^A9@H(sau69c!YD2|2d~d$oqG`#524^@a@v4bkZP7NKP3gf) z8uy(l11~Jcb1$paiP`m^4mzi58G3Z#Sup-4n2Tf6GrL&wZ3Tuat5sw$vAQ}AKBzCh z*PBL(_m%$R%K?s7kw_({rf^iuuVlPjq!x^&slx6Lbi{<+_R8*}T1+H?9VW3wFHhp> z&g1}(J}k{bM>H*bx2ZdVii#hcNu8u8+g|5@*v?DeD=KZl9KpPb5Is-mLSR2(bwsE; z?Clhl;?}6jA`rlHAQ5cm&xZDqQ40v}ogWFkm8i`vS1bax;}OsgwW%K3oN%Fd_u}^k zv+rtAFUKNa)w}kU9NnztkDR06rF2Yj8q056W%?}*zsAIUigjyzxF~+(QLdy{*4`hl zSV67~wO+bunfJLesU9+{7`4)1wVK_p%HqO(nwPy_6(Pe>wH?Mc-2_T5epT@5=BS~n z$n0J770pVG!w~VVDA*7j=W3@ngj=dRzevirBy(Hdt;T0xs~g+wDc1w5%B_}vuxq#2 zFmAUiI)f`XalOl~#@RYqsyi57qBk3UT;cR#t`J8LF3r)F9T(Lgsx1N<_Jouzof^-& z+6h3z-!U2VDm^$j(X^@H=w`1@4C9;acP(gr8T0+G`hiyUj<>#y!v_H7ZhG)J2&NnfY%zQ_e%&|%sV3$pVXRw z)V7}2qheUiD$ih$jmhId&%WMTm(+L1;HZ}9XKPmmV8aaD4K$SS(4S)xF`IgRE6ufy zr8|MwF6lybtFJv0k()3hUmRj8=fRhfzV>cBib*BvP>+)h$>rDMWWy{J0;&s%+1W41 z`R!7CZN;w-1DmgQirhv-FD%X|YKxzhDyIth1V`)?_Id$9YP;fN3u3=tI*4+kl zc82YJTGE>4;rOXWeVd_~`Q{#k?MahPja3+OBXtC9e;G(Z`M5fjo_<)VMtCX3CRyDb zZJ=ssZDx1xBr>cRL5H(fOdLx)FlARx)h**wQuW;|sN;OU44uzvtlzkqj)i}2BwA~x zsFq?KL$=CzdsZX)7}j7+=~|^kbiqIWKLw)#yeYxwE<)0ye`Wxjv(Q7xT*-%nR z^13^^cnnS^bto4e&zC6c6_lWj=8Hv0yOcdN&*tM|GX&8o$%JU*1?hfF=91Tcc}m7> zHND&(h_t-}6NuFa@qjCu{^6ev%8k~T6iiQ7bb)nW-@UYXKrfND#Bz%TrlzL0*T<0V z5`FXbUH_uT#e%!HIy(DAOyVl4N_+pzmgxl;Q@Nh_e7p&&SNkB54m0!)@<4Z5Z!cPJ zt2tV)$6EP7SeDb(B!=MU+If(ReMxacSs3(r(`h10-9`YDuKQYVq-wJ1g-_o;lGN3K z{0;SyK$hZTgBl9xr6tG$J+dDIm^oo&2SQ2;lFvy!QRYC;-NYETUB801 zn}I^|h0gBvQ2Ao# z@)m1>^XWjijQ1sn(BlVE}?Bp}) z#`cZllIYN3w`)#mV!=PYK^$N-Y6I(UgEeM)(M~#}vy4);7JzH;ZHk=LfBq`SOFP~i zO62b8k7FZe9Z_az>=8zujw?yxSKGH-(p%+29lqZIB14>z-p&>9W_~dlh@R>5@$mo$ zLh?B%+f?Irt)Hb7TffSFvd3d9TA4*pEDQ3b?ThB75Ygr#hne5$|NY8|W5 z?$tr-kfAkqlf=iQXQ!>3lTwcD@k`Cvk>rLc%N)CB>o&smI?}CFh;*8ulQ{@pwrzZa zZT!M4Br07x^Gs$$dmX2g=+Td+qfUMGHfgA9meaYf7T=@b#}MyB{au8C?(*uo{ib#V zI}Y+}$Kzhc#eqd)%@AA)dGUyQO1FSOZt5UyO0z~g3VIO?7rMMry{VX4e5+H!Td3O6 za&=H{$7Opt_5@OOi>+=pq{}gG>KAWta&SJ01o8G0l=uXbaf`@~`MRX7FWpxLZ~a$rgPK7! zSjuKw(o5j!M(HNUJcaP zMVM+@G>;iR=$>TAMMC&&tk#x#OBi6ds0_L)ypl2A<;FUpRHkoHsQU)mKHIvdX>m-K zydBFpFFl_OSndmm9lg~IPazOoT-cn%LuHuF+lj4P?sF+klQ&Z$Wp)Smyjy+^&E|JW z2w>_XuR+&gBeg9x9xPh?sl(t(wOz`!fR(uylG_$ZepUP?cmnqL(shS6?lh#;+LAQ{6^BSvrt8$C%RJQJW`-jk-#kYnfY=>4kdm%LzuBPHR*70RbCs}OK$5aFqX8u}l(}Gb@X9%gm zQKI}cul`7N+b~Jk_wa7BC4w;hScIjPHkS0D00Pe~#^M)=y&|nXEYs*`c4X5xogW(L z19FQw=L(PDxkc}b1xZR`Xq6Ar(|RmBr1d7*zQq!z@BGz%#qDjbA}Q=_iMvvcN1Ld+ zGa)2}qST}(ynXQjF2xpxiq9?bV%d4pLAM;?GlkK4qpTD6q&88&F*9A1pi#*! zuks(*V{qT2!eAdugC(A?>3ew7AUOf^yhc$cPBc;4!PUSkDW8R#+1LES`;ZEL9?d4r z{(-AE*-YG>sbN;W`iNJv-;9rHTz9+lay>9ny3E>R4ukQy!+eUb)~9pMU2YR|6=T25 zmMF%7s+5{86`DHEOTJzN6OFTE)%c8ue<6NJbq~#r79XkJk{qk^<*%7HAasnP7G7lR z-9zWOI)_`=&O0fbVnV?3q>ntcOO`(edt<^x`nKK{+OGtksC}2!xu-KkJbGSOp1J%j z*kfl3n(SMbCJLEl%1en_C8HrtZA>mLegraX0`Oc*HD>gZLnn@%#Sk z4fDxokp$O!g51UzjV8#KWCczH4WA`P2lVqB)b;SFst^$gX5pp{|RILcAPk zE}Hp=f^S=B5;XH%jP*+0Ew*(b6WpU?!)uwgaLGC(R~&saOzDtB27ye>AW7D>F(05x!J&nUCk zXY2VNa*@MZ%>3y_mMC~|S*Z8#W`_3~n^NtB=#r;DecYuq>P`5qr&t;5N@!M^As6S| z=JUr3&DNhQ2vtNzji5@&e7Rp9O^T_0TM1!E<1|1l+A)JmmP2M0IGRo44cv6TZL60W zHZLFEo&O^JXR?tb#%~*DcaDms_yk~*Fv2jfpME=ZYnPeXVyt^E%ZqO7(o#~ebw%kuzoKH57~I`(eLR4q zjttfqH$MUb?%e0cwDa`ai|<4hctI1rT4J*$vJuaGy2JI-##gH?MrnGdRG0g3^4wN} zNE)Qj4(?c>E6lnrG7SqunCr<9KLWO)`4->oaJjw4Zb68jMem<$q1o<>8H@CX^PW@h z+Z~r~+)9vK&PtrAI(c%?H_cIyr9`XbrYmYlC2yfY|59`Cy65k#ZOF;n8q;^C4=baS zWdh2Cwmw_3aO0#{q>f2P2V5~#HlZGn*O-b;XsVXRg${g9+#?ExeR~gp%;^(U)3iMg z_47X|xo?jc$DbRucuBU~i#AQM)D$FK09+86u&}~7SxJ#rsiBXb47t@qHegs|)=ok7 z5KV|V94m$e_Z^>2dB62F#HgDk$8+a-9~{u(07Ny*@X=Ve-`~ruXPMxTanhZU3|*&VcUvSdb08+m^3~CD%u~CrNX@^!*1u*LF`mB6V#a?|wdV zKGz&yF-#|((rDtCN9o4Zn2{eYBJh5Xvs)SsVG{S6o0vhXNyO8MiW@q}7k817SWv_D z4>zt4u$q*sLk(p{DyqJu4Xb=DR}h}8DgU45TasL_=eViOjps-jq)R6X;#p~(SL+Wqn_qbFl{JIMNoRB{V8Q&Gg zh~8eGOy#Li`PMPUlm40lKuLuWNjAXK+Qm`XI1(1Vd~CO}Erq9ao7Wg@r$*m(he}K{ zKGtgauJ>1c@?N;%+OZ2ukt@4l$}ajH#_t7mC2_|u11zy6^EPn)dBpEZ{~j`ni22y z?;wALr!FAIZAT!BB;b{#T+KNwa2=4J%R`2-$mV~K9k{{BqXWpO8gIg%s9eZ@o95Qy zJKFKfM(SmL53Zm$ymYErIPO1~LP86sAj=fwmp#n2a}>?-9I8L};bYX(JsT60xhERm zSEjV+XK4?R?+@_O?0r4%VZqWaKh*d>6C#1}W!vCwh$cxs7zpZlUC|l*MbAv@W=WuN zt?#AJ^6RrV%674QU*bbyt$UQKqdwc*?;DI<>V&hQN{c)85c`9^^QD&Y>zmqE+2Dim zP)hPxfDkJnu&JCaf^q>F?M?k6fGr?}x7d)?WI^JqhXJnJ)P52lP(NM6tAA*M*#68e z@!x1jbkYMnPX8&650UUvveG>~oYm}42+!V9#h~o$Y(>0+o_j*#@n_#@HiWdAsTh^c zB#1c-ZS-nvj&?sXvScbIay>Fw48|e9&Zv4KF}9iaytAV{8!X({ck~O+u_$5?R-tlw zsad4SrY)VGg-VxHs%0h_X`IFC;^fq7x_L9$QBb}kM(~rctM=1SEk`rPc_ACYoCZ&` zZS?W^G#ySBvF&;AuBXqk5g&MuOW`ZG3ycZMq8Gs)XTECMc+4@jfzbkXB+<_8BtyB)P>b;(5(oh_xgbS zW$IwiA`vbb74wc-aC%RhHpra(V8qhATZ+-#uBImm<%$yOY zfwiAHx>eoe0oI8mTdJI82_*BQW!fPr=9w6(5^|kB&~rmxmM6TE^s3{ocE=^EJ!1+| zo=k)cS-C4U#HDh=Dxi`WrZiI`1>$4WI#S#BKEZtzYZ`j1lzRS#-ng;yM3`0urcHDT z0KG^H@c5*CoB*Qyw-y*&dH>VMi?KbD}rK+jc7{D4EoQ|1ProdMv-djPqxx5v!iK6P6ETrlT01F&41 z{pH8~k~Q#YkSE}4qds`FHUi1}LHZ{a{U(gc#KZ)MJhd|s;x|uS+nm<>rIn~267IrE zYcZ(2ezral{n)aV8OGIAtePdG6Y)&67$9BGTW^sgRe0#9cr@?nCf!2@F+M=_up8dkg?o#* zWW+MP`$;y%{N8paz8Wi{!0Ho4g>QpvvVZyevO|8(O~r?rI=vFsV$Ud0ck z3d4-_t1MK7*JIR+rNfBbHWJNs`=aPpZOs75<&68C*1x&>|PLgX87qAk?d^25R3HC01M$RpKT?84mZ-35i;w0W4MBN$6f+fkB-nV7O@h92v&)Dqu6byE5K&Ugnr+;}B9 zkHt6F-Kil!wR}>Y7J%V-2A_lNH(wYwt^|`e+2b6v8j+X0&3I+3-r%yGZ~ZkDAPiGo zx5v8Dy|;7XX92qSS!2IlUZqph!oCev>357PfcxdM9D9*0;!%-2&vJl8%vu&nD@$K> z0Q4Sxm7$?7lkd&7+s)-Bq7*>&OMp@(L@w%Cb?f{a0PBeM!6uiMl!1NpIH*FkLOP{2 zK#e0w-(C&IVq?muIznYS(C}}KkdvprbN_H+0E1hejQbMZD_{NLm3P+bOVQJTsi;YFG zv21ZMCxK-cKm?BiI3Jtx^2dL9e|~K@tYub{%&m3-C6ZDE1sigST%&kvQ&_&n7cd62M~X5k zHj+PwP^2H~^AgL4SqtdOQPI(GfFX8Do=$aNwB%p%ssv^gUv}tDF7{K-wkK%6Psdd* zcH2VO)e;4OoL1YR8PXU%*E2ThQ3j*R*lxYiq`;=)}2*^u3r5wbi#%o>qWKGJoX8tGQ9FHs%~y-)R2W;#Y@dl zcxGDb6)EF=U&9al>4q@*CmWCVZ_ag`S}=l9#62*&H+$dx!LOQ1E(7BaZ>r_611r!+ zh7yDa=+!YH8OCxp>i5o1hJ{CXe!R6&Ka}QW%2i3XBl{qSJ{@}x(4t8aTI8tafQioH zYd}=edfd-~6vFGb89fK)QFXM6*o=_t&XdMO@n&lnd#l*TmdMfuR)DXoH$lq1Hg&W{${ zmo8!X8J%Q&HWoukyeFk+fWPGH>2I^4TzAMnH#T^J>A&Jl%zHSoN0Ye{Bl6JlRY}eSI|hSe7D^9bDmgwo|t| zk}Axo-~Ti=@x@uTI;LKw6C%8g5PH z`TOS16$DHXlRK*5Ut_U~#x~nfpzQtv1O1w%R!0kz!yC|PQ==~iR@@}E zRk4P)@nl3{x+Vu;W_BY1rgG<&U%!)l4qs!C3xbH+qZbM*QBfhXe#?D}j-b+iU5Efy z%|qkP^V6P4cg*JQowXpkWeY%(f;x{(rKPevm5|aD$~ce>xEF5tud5Xi&F8b$NY>9_ zSLV~7t#=fKn><*U@)jUxtDHwfVI!{{H&3SY?4{9_V|X(=&Mjf;pK1X(JQ$8`a_;XKCFac$nqX`;Xx&*kA+F8-1P6e6s zvva>h0r%kjtud5d@jIClh zK7izjs;c|+Y-36MGM@XjwE)n!!hT+m`H!Vhj6fT3#5k)yun_wk9!!lx0M8klK88#p2Ih|G#9D4KX*DpJF5A)1RD|I?5Mbs{ z4pjm1awS#tWRSZ(duO;;N=Yil@+psC?+HLroqS@&(!9x|wMUm*%g9|03FcKFckQWh z9v195b%GzX-kxK{6zzXK&yffSzV2s{0*fB^()vjBq?SUo;$pGD^&b5~j!GpYMqcIO z^|o_-;zirQg6$R-c=39_(s(@~yaJ;CAa36syJD|UelOa;R5VHm@ul#wOX zoxEKL)?rZ4vOC|jRbF^s3mL$;urZgrq?JN8U&B-K+vK7<dRCQnavrR*sc??){EYnuE^GsqgJ#M4Vj2+hM%qu7UJxmc8G;{_85sH!h~ z!FUFy*5>p2efXVfPBktFmBFe%W^2CQxvW1^IxOG@;CddTS@GycVWn80dxzZum&M`D)A+Z?XkXjRMZB|i2t<@B+>{T!aAlG0-aL9b@ZVXtQBJ$L&=+^B^XHN&y;qa zCmjB?=wt|bMtIY9ybu7td)XP-g4jLMa#DcU+!$hBfRV>o6ZtC)Zn z6!gOyGwlZ-x}gTSjM9u~xzZVrOgP0h-bD^XqO+8#ZFx2Toz$0n0UE7$(C_yMgKpl@ zuX}k+~(n}Ea=t%W4zhgfpC)=M*=M2$_(i_ zzs)p0E7onw1MclLLH|i|8gU~~7)kH(CFQ$34bGHFh~c>$76;3VEWZ9CqF=L824vc! z0hiZ(TwPInEn#HfPLCY)C$iv9M_18|xK^97rZwv2u@;@6{O8Tg;6!Rznjr&o;}5vF z>l}M>t!b9HPWD@$#=#={Ur80ZB_ic<3u3}-)7+r;2g21JbhQGV7hH5Sx800jHNhu} zeGat5)P8Nxh=8ajD>y^?$7?qvd8>p0BOu*j5UJ7<81IDBX1R-gbYBWC1yaCw)~2=x+gu~)e&`Ag(dhsym|%o!Q3TF?ee&HnBedCkVtzzPW?Vc%*q|eKFk(VsROl|K*s&dA z_A`u4-Kvwz?8dP_b5txibjm3{dZJ<~N>u31cAn9) z@o5uuif`0uM>oq;e5Yk$6XBfX z-|ukg?r#u^I1(U6Ad?l52)oOMH{WSkv>ber3(se@d@o69(!uDntGu89o25wN306{u zPU&?iOpipyLWWPv(qIA^#ohPk-c3!r4BWBrZ z=c8tJZn!A(44XZDG(*C3Y&-G3oGrsd_eu>K&+RU18;LrWnv9yj<<&rXX%$^8LLD*d zz)!eBt(r-Czi|c>9M-8LQP})=?gns68fyYtYM_ssaEV`i4HdW9mvpAJ^BmN(yfn|k z~{HY*oU|lIkBHd4;TVIdNz-dP>0Dbv^WCp-_T_y^K*dx2=r7j7)XY>8V<_=+A}o*(#sn>&&KytN%)8eGIL1R5- z+Zw=)QDy>CaI`mH`{9LlUIQ4)AbJT%+P@0cK)5X`Y3LXGy?p<+3GiB*n5VjG<}L-; z1Ad59@AH}SGnf%#ZXUe))vcc?Px})tr6T$~Kj7XG?*$wXOYrPy&PY>jque1l5#eSAcy*yH+-RCCu1f&5Gxepsn=`qc*{HxD>3+>W|Pk}EJyZEs6Yw^d~+LH($2ISTQrLZP&c1a)U*>c8Bp@%UqsR_7xbcFB3m$qZCs~a#zWt68Dgx-c#yu3y zUo0Q;p)22IN5&VrC4(3X48VK>5HwWV z(ZgP?Fbfh@D9+G=WIJ7U)YweM8G}ovSxOCZ5|q+nHM5ZmkFoYP#&;eRoj&6d-+~p9 z_30NFJm*T8Hy39+A5rpu3TvUL{Nb9l7>ko-{52RQ`A+e5O`R*0pAgbX{B3cXX?&Ko z6OVg!XBu`#$360~?4veIYGmm6Kf9!RmHYt+yQrbB2d^4euDo+&j@y?qFfh!~Mu-rg zNpT6m0G=A%pA>omuv*gS@*ip&cS{=gfK2Oa67*C1eGJ^w&7l-Q?w$8_J5^P73$sRd z+cgpbxJrO+EyL_hhp+fa&q|hq2D~2H*8-T?7hIkn%boaC%&U$nR?y002os4;C6lAvVni?m*ERwY+oF-8n4^9xEzQ92Ixn&1%9*K zBOOOl~vWt(WH7-2;Wcui96+DB}v@3!O-mz7RI%nbr7b zr1~>{Msd@X>4Vt0L`_e}xR0I1J#IW3&o7VKkGaF4J0#&$o9O&JYw^~|1MzMvEyiE; z@cAC!qd=}C$J584Vg&?Qjct!I`2ncTi`UcfTRDJv&>Sa{129RmOt~pMm{b^^F2!lrYMqDv!4UVL&v$yI<_0jv>pO*5{>~J?y&iYlR}_!=@+^jBPJ)z{th*1#V~t&1 zdMWlRfsaKgs6XmwAbdLW&~ppXO>UnJVEg35z~yPw%@s?K>_tU(MFYL%#kmcQm4s zj%Yb7YnI1TfC@qSQPgC}W$>`c!3giN(RG*Dm4Z^;p9W`%nBAD_V1A(q*yTWHfJ4SM z-qJxTZig&x`849;J{Qc!>jL}sTx#zJp~{z8#dZQv8IK)>2PcK$^GbvwdrAHCsrjnK zGww{qR6ws0@8@kA29%G^Xdfd|iT-OfzkYP(DhdD6!_8znDKj9hUcC#`37n-vp6Dhw zO_q3g1b{T+UsOhm7fFK#ifi7C>(W8TM`HsFu#W@SRL;^L6A1{GdROPSg=n^H26P76 z(hut2rR;tCzK}w1F{enPk?+P$d|T7`O{qSLFXmo%+TmP#gt+=z7v4F@Zl~?NxO)a# zu-wv1vR1m-4|v%hrT`m+DEy;c6;O()ok74hFdw(C z!LI%N_bOZu94`9@H#-NfV{j)W*zR#sv`8{It%#V89?GiW;?jT+l<)i^zNe#7L zWTKL>?t&@(Ua*xEGb`*7uYFd*T({_cE75}d!e*~}mGc;`!@yVmgW}APh3a5vv0NXH zBlJo;FIt?Fwdj7|RHm4%TdojyiuUqT-9&RKusWGqhS~4Zp0&#a9Sri?d}Nl_{QW|P z&fi`#;;ERdE$G!BC+EdB5XnwxY|rfPW!vwLH{Z@ulqFCcwdghu#8SWxRFYATPm7O! z_Qp3c*$fogi#NSS$P?GBZ0&li8u}z34=HNQ==cm6;~H5#N(K*zXHZzHy+Ds96L@g{ zRa4R-P^P7&C&*v7L(5zo-UiXuQPQQ*Zu{`0U`ltzdXeRqWnRr1rY4l9SkHahNbuv5 zlVcGB!qmgvM{~-Wq@&cs>yQ?t60L_2x@UYnF8qdh zl&-zRMuPheHRs40M4a49#s-DG?b>(ES@QMKbC#b6!AfmXD4O{(Jmc?1p z4@Qybkc^LhAHET{mu%FcZ~CD2yn;lCeVL^BE-^18`;Fspa}{;@_*vcqmi$`SFN(1` zTQb?$|Gj;DYCTGoJ8vJ6M#^a@AAzNd9;i*{L0tT8zLD|4V~UWN#4zrIZ}+yNG(5B+se z*w3i{c54En^{j~p(Ax)P-S`!ARVLtrY$BaU|JeIcX@;5y9hoC(PM0G7z9jpn#*0C~ z1*CwSEg81`FV**DbgGaBP*wNVdoO52m2^Jehd3`XP!3-NHEqwLZ+HAdzVGn@E4)5L z(6|4(`2Q~Mue4yXU9>a{6r=vCxJUb zY3`8pjc(D^;NNdOaMsoFEV=MGqy^cy!4X=vAX4^>0Cg#mUfT<3aQ!b1IZ)?T10HHl zKTh2M*a3br-hiFTbfv}fsevNJkyr{w$`1)!2lO@}pDiRl0n@8cYgA_t*1V#o+vf2# zu@6*O7I-M@{~E}da0JP z$5HUTNwS&fqP8#MR>jn1-@$NPTP;a=UJLm6dM~AU=G1JZQB=q~hBl8zAgN{hFZ9jK zq|P~QCmR+h=D3kK0fOYJ9ffF8sKY~$QBI5`1#8K5)`nFEo;oGrk>YdG&0O4=iJWsof6wKJ-Bh1 zjr22e2Au!3GPJl;(>snKBvx>AzAcfg!Z@y-oTQj6ca8~CUd?j7KV-IAgCi{6hPoQm zG;}$7K*{BNq<0dWAeg3Z;I2lAUg>pn%TWQ%y8;*~;y&s5zuL1|*Fm*}1UI0}q@;05 zLIq&Q2QQ5v=LkNSVgtx~5JPx(G}GU1XQUO7@$U;0r@sJ5#>6rZ$C$bVM#?Md;xj0z zD!}4>R}l@7)e<=Oi`nGG4tvENwa|xpiqe4fG+t+C!u^ibe1VOd5DQm)7SUfX4lta8 zp1(GOlp1l!GOAT7b)tToFYXQ(s22TUHC-oaJ+s!>>TUXtg(o+8VO?~2$gdvf@UInX z<;R;F-BB1^2D6xu#Ht|!$Jv_It9kDBcor~_mgKCWe$t9yW~msJT1It-$3Oar(&5m@@OI-dhhv4 zV&&5@ig{z(p*!8U9Fa}L=4z;dki~_2ur_s0;UMMK+W~ElUMUSrg!rTApVQO zc_mp!4&nrimSzF`qcff(RRe^pIpBp%GE$G&*wDLwI3Iqz#QLbDRb?`LWfGwVXaOq| zmj=+@W~bK#V$qoc5!hxTFW0;q2uZ8M<0Co~?|6Bj_ffV~@v=*n`lKZ`x zYMGAU1cwBrv~4>wYgQsbO1iE)uytnJOdA#r;JI~ywM?hZnQ7VeJS3Io_hG&5?pmsp zqvqalJVSfa!#6{Jp~$Sk zqYz#BE+Xqp)M}G{!DGrXc(TqEvjm5rn%I7=E~cGVB#Ba zzl*S*gWIU%x$_{he#Br)gZlbd_TdLxy}mx~^R|lW(W6zN$BM=0BJ5Yq$~w#mWn#_Q(tWd9Dwj}5mnSL!-MOui z;$VI(6`sqx`J2D=mMX$ns?o~9|+7m4$x}lwAfwM?p`WT>rLWizR_2A0D>OVc|}$>Ue>M1(gA=w zoB{kV_JwZ1xRoBRi{0FZf<=- z-}wZ?6eEDyRp44QFvQc4ETX*lSqL0$5akIqc!OrLTLcVR(UBGHL0H8d#IKWT?x&l~ zXXOPjGP4KK3s6BAY}V5`ghACIW{>yRtkCNdp`Z6CpBEp661YASu&Vt^EV+KkF?fl2 zSpLM3$^{9Rh4qffq-gPROK-l3V{krRQIGbHzhC^VcvFS&GXyW*vy}Y z{;P^TD1ns2Dy-q+-RP0-^hX2w`4Z$L%%`4D z&-7TGAGQ-An$B!|vws#8Qlhp{q61PX2tWCI+2F7hRgoETUEKIM(A1XgbWaohEKMzw^`r4>QnOXZtbl>2+ohl2;t6(bj+L9H z4DgrsQLXnVx^d8M(|PiBgstfQt@rNKBcK{uS%5{x(d#3AG2#}2JJ|ZLXm@dp` zM*wK1Pe;Z+&?J9-{TC!%U!`ozSBX>?07#hu{DAi`_3z|^P4w5F>kgfOYv~h&+w3M# z-l3bNcCR}=XiW9W9$#}jhW=jOG{25TIhq}cY*-#u>p!wLcUk(^zU2A%ZDo+?{NRyb zYrguhzECCm7h>3SvEF_s=4zBxPH*8t14FP z`Vjpw-W=<0N_Y8qTls+CSVe|EMBUYt`AdWGTur$hP2v;$$iP{-mRi9vG85W7m`CD# z2{MAZ`U1f@q7d8ilE#o)k6*oS_X2KWLjw(J{*ZYW!e+V2?UDH0`A_ixl733=BM_*b z=rznK0N8Af$|l3Zr%e7q64V9G9**?TG8jBES6+VVb664y47y#xrzk~fUW__wGUp;u zGi1P0&Ag|4vOsZO5KCPTl+u?> z{T8e}V{@U?;SYtjbF&|tL6`Bw{M~tPA}@OYtI9CNiTn?_nxkhMS~srxtwQ1%n2XA3VV=xZDFEnShxZtD*^@ z0U1lj<|d8W4Yc>oW!27$3fw==T7~J_k9|~Hr)C#vXCU7Ht4h=2#(bAB?V4@Fnq=DxxZNQDd-Xkj9 znsCtkA?fMjN3auF}6Rb#d#T<{Qj>e6PGJ3pd}|RMpTtyijfJp7VK;u zAHe_~Av7<&8vq$NKb*(#V48|2cbc;q669*BNq;Z>Gxk={VTN>V`7)u- zS^@%Z8e;>#=EdqBwR$hl6!UH47negEbEea(GY>+`2vAdWJ!RyGn-w`HetQzIjn3%@ zlFhnd2^!0W#N#|#2-w&3--s)D{cD5SFoOsgVzhm*wus;Knz|hzw;yA$yHZ*qb={QM zQ>>)TVJ-WhLEdLEt}q*7Aqh|SlD-5R59TNwC!Vhpac|SPMZ_If>p}zZK5+*?{ok3J|fq+ZN zlg{&_Kt1D?-zzyGJS!i~6Ln^Zg@PC8$;o#b+2F`i(p1|KIqLTTUZbrfyAjUYfvWs( z1u)Z_#dNNzzT6eSv<-V*N{w`UweEK?_BajtAJ%yf2Z(Gy($0VqJ0mnl-@K<1>=I1I zUtSP*6|Ims%);#8wXe(>Ob%ftsPJC2s*UW%?FZ&qRp{U+{xQS?e8NidDx%k@vZRUT z@}7KC(U6bRvULG!J?Y`g zO1nLYKl=HDLaNM{@~^fsW9^MZ?P_9+)#G+fLMv%+DP=NzH0k)DE55DFV|q@H2;tz1 zR!OWeo5cywgRf5iG1AzggGHZ~@?IbG%k}TafX{=I{k=OX!O&-`?ia(}{?Vxew8?_B ztaY649Eb+cY}OFD`LTLFaWpV|`0R$b)glicriPhYPO+wi;S|A9nA zWeb3otvN&}_Z539rCB4|-e7z)DR1#70GTQxxa`2(0aeas_n%#NDFLGE{ry=Qt%W~} z-9agizV)Ut!p5*6R?s6^VFZR8m*3?CL<+JJMaUh!8<$BPqxu>KOkN`{3UwhX`|y3RESB-lQJr;bwuRo z2v^pFzjn=H5&Cpx{ZRELRkc)0CH7afwAERXn$4P!gsR-7gqr17-`Pb}Z4i@J_a3Ak z$<$NJk>6n*SawNI1TG0=Korj$+Or-G**Wa%8#=MskXM+DH|fTu0@N6wu>CM7_y@33 zZT$Ahc2mkK7@yp>=KOJCXzV>&12rf(T$LQTfZv>+T8S`{uy}cOt?F=oe2Y3@P31p| zRS@vBXR>FJF@uJDR_F88z-hZ1`pJkTS8+7dOo4M+>UKkSU!`x$jKz6O^nL?t60-Q$ z{&p!Hic*Dv#B)&MU?MAP9M|QYv zwHG+4zid`Bp`6EIUO|!+A0mWpZyFe!6cbgqVN>lHPGGpc5TGgOBPW9SE^EfTg(_7a z@@YjC!m^wxKk$_Bt=`!;^O9UsM;%8l^i9R z^iTbUQSRCl!vdvj@Wz2D)&XI5g^?I(0hs|>)IT3^%O7wJz%!8jSD<_I@kCyf(Yu{$ zvw0ZJul)sax1&3PFP;kz=OUaOf(W9Tq5F3wY>!E|1x3dw=Y`3veBzrU#|E`_ukZKR z02@(cB)kn!;VF;W^O5DBIE<>E7$6wXL0`iN2l5e^<6lb(G4qtvZvE2w^%_#!vqy2R z&1QG|S4%0C5x@Q6(T_e!}i=zS{ zilSz4#-`dzx<7t7AR2}oro$Uq7j2_nrgr9_I2m6yoqhCD8XEq%&RMwJu*h|9RMyP5 z4ii%(>f3WHfo)Oc5GzlsIiSK^pl#B)oal@qQXo<15E^}bb=UF=C4MhtPq_{B+ZIPb zKjpH;%U*&O+YM5|MEB`r6pW6_B|w&Bs%FQ@YP%VTx@z4os#%)b-tK_kQ1|Sao<2Kb z$1gRk1~GRandoj>F;7kVey@2qh{}V@;3wy#M|Sg2fKkC=!_>Dz(=sn!nBG6TCLd=J z^P|tYlx)o&H1{7)OBQW!8f6Y;SxgVDJ&+Flxqq2I*I)FUtTj%RtW9cmMB0 zP312w(JWO>)B3z83IEvw_y}Eqb$oi%NfU?Vlvd2oFJ}k%kayZufUGhiM$heu!`|BL8klxXb^AD!Ir&+V)(m`iooqcMqbGU$>LlrQT&AqrN5%MW>C(3k%p2MKsUJ0hgJ;lpy z1eXRCnGQNOa2w`3N4yyX><|atv{y4fCIVsM`)Tzf6g#7+iW@MUd6=r86B?f33 zoy5RNe{V~bCVR8FXQ5Bx^I)uvVVXghz{oQp`FWSrWHLV{6`BzU5+)#ZG8O)}MHH1~ zmdTVmGq*;5li!xLo%2&_5<-)f zFC`lEN`piQB-RlUK>bYx2S?0&@y7Z4-SV-|?1B1Gq*(1nWW)8XY4zaHfP2Y?z5Dsn zI=~*d_2M{(9?#t~mQL5Z6D>7;ibzn93q^>-c5SYJ16$b1CLm6dmD#TMX0uGcJGS(2P|slHPwleRbyk@ zB!?Pe5|bN*ho^k@!yg7$P2@Yxk*2X&^f$&$u>*7~gG$ ziED&V)$c)4=*@iYzTXIfp5Mr*Bg=|e_l%4QHWZ2F^@d4+cL>I$KOOp7qm|_4YlWWT z;KW@^;@A7Kxa*vM9JH6{HHgieg zSfH{G@1Kubl$5(M)qTKy}7)aiJ+_7^(tm&8KLH~k)W zHqinE!I!c`3=a;&RrUp0O`%fH2Yyak3ffUPf*W2cW#+#3uam$ac7z=e(zo}+qwUHUT`z( zYl7=nsJVS4sEy%d30e!KvA5GzZ+gO^iW|;EMF!BlaY-x$fWolgqNjk8f%uj92;lVf z+wW{Gz^r5Tc(x*kFw!vu7WV$fDyMnP3k7FjKI=C1K+n?xr$;JAz^E56Oa+DZ%s~J$ zRU2MXNpq4Gc?+>Q97`8tz?;JxmGP4jnKX3_>yCzeRloIot6Q-4|Q673rAJowNz8;%9)Xfuc4DJ#VK8 zZ}I!WNi0t}Rcm92&1@X4ic#iesdv>~;L=v%xzEV%42fMU1=pz`G9Buluhn&rhjru= zyx*w703S1HBTC4G=4gACdbYTxbgH}B6GY}IoX_}uER~DzC%6AlQGZ_x?c^g0#rZQ;nKZa=F}KF6D8sP%kCwYcr3R&ReGVBH1t z0*%S(^9{FAnaPz5&p-|J&MXekG*ZAfR83U{&Pt@(okFtz4nQ|XEJ1FNKT%kPsGt6g zsdv?8*5$1?{Iq_yGf(YWj9AU@7G_&Q#lIY87xe|l2AnS_Ql!*i&wH6MRw_$$oQ`33 zU6+;N%4kurOV_(Y;n6sn(UnbjN+y7)9PD@VARVtM7npasB{<}V!52p<#i=cjOd;| zXG6Ek+og#n{r8)QJ%8w??j7GUL2NbAe)}y~(ZWO%<&tdnJa)I;*Y#Rx<8bzrUuwp( zu)44~=I?cl2R;i1A_e<}kad@0YDJK`0rgznw{<%m3fNVf|S7;oXoH_HDN?{Z-hZ_>qXOSD~mC^R?K12`eO5QOD7*^|~51(2YRt>V5m= zNe5uEIH#p;cDz+62;XZ|lY7_i?S_p&7y8Phy^v+OdLmdrR1?NbZox>)V+(VK@tv4x zxW#5`fFTaGE3{Nspq#LVVp&m52x zW=4iM2KxQuSEzs}iucDhLekX{1aKiCbWpltPs5v*$sj8{As&DzDp4kJJ8z$ zJ`e9V;A^zsAApzIV2m<-`kkWIz|SOlfwi8%>=cUKn|ufCI_Zp|jgJY>V}g;KiJ5^= zrLE%%{ifZAC@uvaa_GgLW(~sl@hAR;B(81Py5uWAXeaebnZ(feqi-(CZdO~WKQ9sU zXBS6sCzYc3jGfiwSmec)T(^F^QEP;>p^s${>3S2Yv{HMD zr3#0SU%zAWY8#5++-f5vE;Csg6dy(D@d^Cw>ci28bsv73c#4aYu-?Ydl+!O(QRgm~ zxluKt%N}IOP`OT=_&w>@8e$(Hz;A)wRXFLdSA{-227iS^2wH@840GKNh_7Y5zsXSfxy&3OFR!S3#ha z+Sky+1O!&Ug~A{|CR+Jj&CnYylEx85Jl1sDg=-$L2c|D_AX$7c`5esn`(M(|>pZm_ zfsWI)|Awgir|~NHTfT|F4)UC)$E$h10IP%Qd2=S>*G{@C?Mfjb6FkCJcY~x;b#}Vl z>J}iikcoZ5KHy$kwlJC0B&=e~9WAwS6*WPL-F8RXf*}D|eoqWz z2jw|>p-OJxyQ6dutJ{a#O(m&fxpNKzLK)I)|5TAP2le1S2YsXWUfk2z=5@VqdY7rP zWn;i_#*@;H#?+>N;m|&mX}vaK#8j$AILt1aZ+VoqCiCYm{qkS8;2*okKktV$WQ5=f z;K6NRnMBw^;@h$cz>F#pF9d0gK{ZMFP($}CD^cUS85F&N+^+kj<;c%1yB=-|OG0eg z;nx6IXHHLukM||Yq8CfSiwC}VDLVlInt({GoT3PhTALm>&S9y`J2~0aK)ZUqDTjjJ zQRpgm(mzJ>f%%U~qu)ocfeTfpAyu8vpCpYi%dtHr#|L5w2f!vQO9bk5whZi2ryq#~ znElxJBne8KI?>5JiN#Sh!b@kCdT<0@r8iHV?CpREzeVOXLBQ1xW{DCjppFIa*M@Rr zz0;nckocun-76CKLp$>)y-4im-c?a986+8hwLkAtIh&}>QYpiF_%4ueKO|dkB$qjv z!dE?>b`vE2Om$q=a%>}n8l~V+ZAhT5VrbcgAgL`k+hiQ4Ncc6&7a2QDD zhADULW0?892~YCPU1j2x06&s|YKrv4ID_c$oD_X(f^^uEPg91b6o`eGy6yU%^QUJ1_tqJ_ z>;l$qKg=LaVk03fF)pbFoB$;4O#T_OCKe%n8XnVc10@+&g z#j{^OmLk4-G_A-EX68zTvvy1?t6p6NrV@_zORat~hY;3Zm{=zpErnRH^9>e{o#>&e z>nePXtIEEp{P8r2*U=}s8mB&`8dGpCp{0pRpWZ;Bn3SG6hodq>8aVbUpCd$FbS~jS z!Y!LzRh@0A`|g4TO#;(r&VW~~Y<1%FganCTwFb>eOQ-SSWS}=Ndrbt!V={R-AJ3+Z z67RFU?3BR!*j>a-z0p3-BlXMgHYxQ*xc|7!f3GzE@t6nbLHfMv;`eFg?g#;tKy9rg zD#L^BgALsPK7|_A*LYkQ#8c^3kWWFRsBRZCYYZOQi$% z49lxY&!Jgo)M&UU4Aa3{=^+Yvm=Xw>PNS$16P5>@ie0((DGP4nd@6Nfb7Cvwi9{T` z4WMLCh=#64<7i+bjh(BMIr{oVPbEhb*QBq?&KXi?$x4B^V{tn;Bse%n?UVOy*P$Ly z((|<1Ri?t9`%cvLI-|h*FsEa1cv$gRNcOh+SGfKZYIvdJ3mUO*=FhV`h`Wa=g-TNd zq8ElSCE5IP`J2>`h-iuUE4yBGZEXpTlz;tGRrviwir$FgGV**z-vN}*DVH`mGsDASu(8hR{OrUMZREfLj<#u`WgCM* zEK`4yr<{QjQ`)e&%SZNCNn34tPfh@pk@!PEtyfwdgo%5=wI*P|)ItB{dvc_UE+j)N zwc!S`$(F5khqnLl#pnU)IZFEPHU_Q17PeQnOTG_0OK#O>=Cam$Z^*-#gD&aImdF|$ z8oOt@#M)d2RQ(R6mzUviNIyy;w|gU#cL6f>kY5mfPK>B2MckTT^$r3$1 znpx(R(&%@hWzl#UJNq7hOX5jvqi{71&xx!S=6v(B<@SdXJCd)zQL_*qq2{A9bnZK? z4Z>Hvfa1v4ekW~f%p7~Dk|%ItJ*aWLvnC-6MQZ*5-7dt`DjRP7dYy_y*w=U;YdO9} ze0utcYa%i~*BL2Op^@>=HuAsUf6oh&mcMA&MdNAAe9>W`i(r3x3^!CX^+!{YJgw@3 zbd(*xW1Xw&2+8N_sPbpv!6N(Zm`yFVQ%NP4opsX``UOp?G>M(qLW#+6Na>Vx z>!z8iNs8;$%Zq01XfO-iN)o<`={eKtbgUBw{f6+>BzmZ>YszWl4G=5`=jLJfI)Gc% zg74@Xm~)W^U;CAx#Y*g5f#`^ieWS1Cwk8UWS_Bcy@s3{MsQG9xK_^NZmTigNz7Iq! z{%|5&AWlcyrlfl4C5*tl4+w)k$IQ5*;phPrQ|BE`SZty1^qVQ*EoRdBVPC_3Ql(uH z+15%!_07XHr&CIdaOXH{(kD(<^0_GangB_DD?fB#C{}zhOIi}O)2Q>4LpD>~ol1uO zB+k&sR=qy`|9lnov>?y7hgo|A6R<;A;k0#R=Y^{Wg3WXubvHi4v+S6(&r65S-oekd zdY9cYQbr{~4EgLy9^f?q%Wr~GS!oKzF(l%v1N$%OZVeca$=mPJJHbILDCjINT6y*| z(jKvBgnT_{$+^tv=~Hmt+Hfdv^xAdxlQ8dz7kwVG|D)3_OJq6?CN1$Qi=xxcaH_u3 z{_3cV4PDzKv`g!GwISPn%us${@;9fR%ItP9Sr9i|wsEw=+-(vd?L+DP-6z0{Xd_$| ziWidz`aF=rF$GMQML|&2Ap(x8hb7)Q0tcL)>yhcADJ)ZoqDC6y&tw!^2?Oqz)z8hB zI&Og4yH0*M4IC;%{lDMVmm(kg=U*A^7rWz|fRSNYEiqm0Wq9*imnR(J+Z#OH!}%JN zIq}oA&P5w|WIWlPXIEKRwne|lT3)WWVDsk3t3-GkffP7uH8h-%6Fx;XD_;!iw@bp5 z5eH7zCnxadXR&n2beYWqR7cO?XbnB!FJrXX5@&^qDXLVdGA`DVtup@R!e+V;o)e<1 zIy7cfhsZVkNa@ZA8!*65^c_+38RWz8gTDfUus>_?%Mu?02us_=Hoyrc%-V45p%i_q z?f!#yG1Q7ICik!@8}57$EBIdUgVnC3cGe|`3YrFN#cLu4j+mGD|4%zCDh#Ap>_>)- zUT}H`IWYR28LftXISL29f}U+9<0D6O*7aS~fp@M|@awyVAUnnj6WNi2>Qxd1bq^#g znh2_x5~>tE-vg>jJ~#}etDm%p9)l)Y&i;s<{2rw%>su62jm^r5m0a~NbZqj1Ie@E% zd}p-VlRtHKZf$ptFO{Tj=LcY`ayC0h2CC;{h^5L4ILICxhd``ljJs^eY&?N_#;APS z;kR0DcAB8l=v;M=O`=BR#|Eir;=_E7v!oF~0Y}^JkHRXl+_-J`3$7?q8hJU$2yw!#j^?4~&hQlc|FD+UIJbVj7M}x|zIpWpH-Z?k6$72sh{U?Q6vqdiD#@#dl0uG{U+x!ol^RfMmS{l#RjpC2y z6&O|qo>TeKJSy_kAEuKvyMF#o(MR10`MV5fM26t&a|MWrfkGv6cgG+g6~#oODTx5~ zff)%aBqY=i_-fkO3d67fm*$N~G4UPTw#gHY?qk6=S77!OJrULBiExF$yMDby?(l<& z*!**FGT}6&sgfzd0I(HYAGLijViY9e+5H?5Yl0ale8+9QA>$LUM*S zGO-y@1ZSuJ9un_4{rP=An+JUUjBab{}#*Qlv%>PPbiOgEo6mr@z!3 z4Do8Es(Yz7EdsVaDuE!OO>i8_IbG!D)i(~2oH#XtO2G@>FUd=;bEZ6y11W2%knjx+ zZ;zCUma4~$6K|5GI8J?xLjP1E@xU^<>v4eC6FwyEH~BB8At{Z@v!_fQ{ua%7CC|s^ z4bbY3E!Yh3%n7X#GNwJ@L|)Vk+y`Se2I&%N-C;a$-A5QI{FJ{uanmy)s2&=2QCObL zP}Ob^>X)0_sI4g0OG%m~j`g({|3{&}21kE>0}`x4SVs(Cxxt7+Zz5M_3(^Pu0A#;0?hD_v z6K*5YH4rY=JFvoUp^3o->GH47aB%^Cm1|AJ8kWt>Ex~B8(2*~dC_elXu5Jr%nGg2C z22(FMlSp8}+B?u!ocrpK8wc(2)Lv&Xo7131?X*X0^H7zo8oRV+8F?bC$Io&W?7d9V z1feWZDmjWAd8#$$BmJkMyd10dx*)`OtKRE7=@jHos#f9{`lB}kJ(^zIZ?#bnLGrv@ zq3!M%*S!MiDM1kY23ngThQ*sEMEH_!x0^&(@%X`#HO;Og*g8!YTI!-<>&VH6xEOP` zH%H7f-d8IrRhwsiWF+Z9r+28Ku!q;%W)X^p9i97`qj%pB_r2{o?wko1pzMzpjMT7- z3W%S1ZgScSvw&y&!}p9*@ct(#B{=4;q==gKf)zEfbI2OOVL3|f8AMYq5HN7MTzy&- zE_J?%Fs)s1Js)i2ai(vliA&H>0zrtK9tHxX`wiVuV=32l!t-04;ApvuAB%1Ec|f3n z82-IMcsYkn{ByFM{{+OhB`7i2;CmvMc}+7-kh&0kOX}cjXokb(9XJGB(+~%S9CWH; zB&st%GijMIO)*rYpw|72D5gWJais4A-q2qY+A?w52-@h%F*dX*bw)gc` z_uR78#*KY9(o_V}DktlB;jyhuPGE&o zCCN)G_FJrO^<<(yDr3=5PYWSN`IPGT+Vr`+`Xw^zYW49=?pGD{VM;r4Lfohi#qV34H5Jzk+4eg%Rt|6ogdY{n=qn6y>ilyIh>$Xi5z z1;78I00KGJo-{75LN-o-?q!PtPHbyO1_!)zRII*Vxj>0n*JQ~&KF9F@`9kG(`wBMH zFIpnJYKInM%GZeRi<&AQpH0~blWB{{n5i+t0%(t{I+9N1JfdPR1ABYs_k{#f+ZN&5 z1c~?OV5SBerE?;l1H(8h)448^P}pM-gLo zlQIxF7xht=D;@T5YclXgIv@-7Ftb~#|9Jl@@dp_**74_+wy?s=L7 z?lg*KF3MalaXuDVTniv@R}r9)P@A8h?$=7L?xzRCy#4W=DN`cD99zynzY14oe`|Fl zVZEa%m36%S%c2wg85`}%jlz$dsgWL^@8!IZXX2kn6u%na!p-!M!4T49a>k(PSj> zj592@@r`mFvdDC}V~w#Ed^Jq}V6v}0TS>!0&bnVun@8x&v4+r$a(KI*BPr(ItHeDj zhA}|(TwfX4^PAWtGwI&NuB?Nt`Tg|)H?3~-9ATJ7;#FA@)AAR|W8s&zCcF-+hm6(v z;(IFgx;Rut#!ol=GX#nagykUQ-KpN}ulLzO?{VVJ?3v{( z&N(yQ#`Cm#L3>MfA-f9h;cwa`p>g%+xlHG2s&+Xuiap*-o#R~cetBKX^}V}nr09aA z)aBZInmM^t0IB6I)f%vLa*2OeN4iQ}h)NrY__kwcCmGm`ei9kABiCZ- zr6C(Q6G5Fyans<3kJU9=He!&DkuD@K8}$3Ca+H~snU7~nc%9}j)y2V}Gyh<^CE5)3dy@0mW5&sC>o!@5_cLmNo5R@FHee`a4F!&!;6p%#!l&lO+sZ`jqF2sFM7*1}KN#VuEk63FM=Mds2@D5OQY> zGw_H`%G5jHx2~P>`u-hL|NdlwVKvDuikArG(#bAo3pPAb36Qj+hDqPty)h;*|C3fI z{_K?a@HlM#(|k`{oGRR%o!K-s?irg*e`t_e{H=l|dsqkxCfuVgpjeflN&OiHc8u{2?%sjx!t1x}P+jhgqlW`_Wnh%0aS>F0CviY2pvsQ5SYl7%$V_m$@xBq>njKLIo!O-gJot zc}Dnm&XEBuVC$ zdlSvpA|#lSW0$@oT`@c#(7~6E@g$c|=K4+{AVg4!o}FpfE~qt>#BM$bd>zU4?@eE! zKqlHEN+ET;@gSKp+#X6MlU}3mz%p=t3HRpk3BN$3`zzA-wmw{r4*oTj*xKE^dHvvq zb4!JWAkswXEqKp_v~o#ug~>r>>gl!`wx>X>Y^OaYfcM3zA}Ur3QE!WhHq! zqoSFRYGK?)B=KUv<1-{ImPP9cQb`jK;B7Fzcg1*^YNg0U zzbaKG86wn!W<0Cs0O#9AQY+ro|AgJsr0hNcU8cO%Ao#;ETDjFeOS+csAemTvoFnNz zh$28{;}ze<2`ess8_*c74@kma7#}xvhr_v;?@bvzmpcq(;qs6H1WLDKw~^6EjnbZY zk<>ygv#W4~$B(7#q^BEkRredMR?Yn)hUA%5hxK8-B_2mHRLd3-G7hZacbEusat3Z4 zy{uz~2iD4$Qk5=AELg=eB&?bWt(n7Mpc;qx@GUP0bR|;v=VTX1Pq~V(MO-j58 zTUSqa#HW&_%sBw{bA56Wt?@s`wBcM(H;+vx{^!6xx5tIEx#EQ5^V+)`gp*MhAT=@Evs2zEYUJ7I+)EFlBS2+cx6 z8@vu&36a+uKcQX_(&4p4?R^hNu~E?a%D2@OX$T|_WoVndPP*j}$R(HfB`hu;jFye+ zUl^em`T5P8-=?#|YIG`AV&(*S#QtOGimAK#hOk z2Pp~=)^5`vZzUn>l?*8rQxnj2hG3GH~jeZxnUyst_9?f#nz9Z78PoOj#MJSu|S5DZN{E6IK+wUMR*B`WCI!sbnTS4}~di+`QbTvZ9y6N$?eenb#X zTyz0nyjEm(@_5(F7RCU9DJ2_BUx{$R3$|H)mz?i;m87MB39PA=hEZ#Lk*`!ZzF8M2 ztwOE(qDqq*3bYXIGVA;WiQpvrlC*f$U++2cV;gaQy*sQQ9G|@5%uhce@s2LozUNsh zWJGT&6Iu#Q?V{29xf*XDxTOPGQIM8k13tGRnL7?@)jJ%sK)#wMO>_m}xijJ;k#i|U zrX2xj_HsS!uMmTkt$cXAAwIr@!X{Yio3AqAmVPBITSa(RNgB|o^ z(bw_M%=QGw`1@|1F7G4yosF$t)8W6wurh*kvt5)&?Ean#E)D0pNV7(R-8ogZMmthU(n=o>3thN%4svhXVYBb9&X_f1D)7T~Fe2ITW&>P2|LMsKyR>XcR%k5XcXGy7DX4B~++g3cILCxkoOZuo@ z_U&woUFfJGi!WYf)EJ-W0qzN+D$@5zGw)fXpK6FFa; zyA!+3`vqg2cOD02&RCRC4eH=8+DnR9$`r;9xazIUPMTajgbSq&D@TsFDFO~5fWPnWxcb?KMgq5jHS~& z8W(KHi`Nc_KV?vfw0D6{-1SStwv^Mx*ks{& zkS=%@XeNT^-PahE+g?RBU(!kESVIQ)&wHh0_{<}w*xe$AsqAvB~^nQ3WY+jw!9rRPKxVH~+T(i)KfFH*9AW4PEfYKgAZSg}P zFmI4DWq*d1aN*#ba%>^=yP*p()L2g`R0sZR^2)?T!w^HX^tA^Xvg5MLeO+JMyDo1w zzF>7WL(N1`rSvz?{EZ%BktH zLj&_3OX}@M&836#Wwf8}4nFVd3qw9{NVXAqh*H|Gh?~FbdpktXO%rYMf6?_8P*pYU z`>>Q8N>W0)LmKH0l?DL?DQQr;4_$|nZWJUG1O+MS?nV&l?vMuQ`sO@}@B6&p|F;&) zsDMYZq3BQCTUO^}SZ#kUY>pb+Nx!uvG)4SBMbj=n8RVe(qtohEEsvO@Y}(UnV9DuUQe(d6-d#iE0w%3}(t zEZX)aCdw%#}P0@@Js$GzquiZllYd%Hv=z&hs=4%#eCZ4!rO(9A0xj^ySl2`~j z*ktWR^MAL$UMe=;Pn|>9Ib6%W4;PPiNq6vN+7HJ|Jfe~yNT8ISlMdbMrs;gZO3JW>*J1<;3miR73DpS59fdD&)ViL&?}4RDDeT% z#1g(^NY;V##z$*Bo9p_udOTTlzmCE}Y7}yniLMv6@w17xTnfvt+p?Z5FR10nU&!4@ z8}_F92d zKQP-Jy~CiHPn^Wa^!;@J!S1*E*=py!2Ox7g$3*RRmlwODNymM>z@NO)vIz((9-_XG z<0V^AFjRP~0cQrsintm>{A9;VxTQla%u0_g7jE~kWiwF5Uz+UoC4f-51z&$B&w%=y za$0!0mJnElUJZXnqo&4r?0tEtnV!nnRpp#=9b{Q|9G6=C>mEH0Izp*tv-bSEz5Wqt zhoQZT&P7}fsZRmihdPkE32=jR%6fNM!DNe)Z%m!t!pTZ1C9X0U<)d!l!y1sEL%qI7se+Dp z*6@aiV^1y6rSAi;|H^YTYb_O&Zg3iG%UR2ki<`Ox-PNkeu4$^-z_L&3Io&?jr;+i2 z6XrUPb2os)O7;5zV0;j6uADo{kWCh_J#rrbI|KCHDwE=FU4n*$^D#-TN+|V%=T9;L z`9-W5?J#IREmJoX&;Dx)jxJ5xZheW=4Dg3-b(a{wnt!^8_1!coXGwIV(%x7d3z5{# zSk0*RXyhPRml;}~%j{Q5rCR79`2%yIEvKQKNK9%SDUDdOoYnfZdy9cvL&<%i`MP|i z)S51vw}&UTeAVYqds^KJG+1$yaxBg+zI?mBRgpY?@+*mH7AdVx=q|<^(X)f0;5-rY z*!t6X6`iTDdr!l9xK=2Ju)vtZ>vkeU{dp5mDFcao67nWM5)|3h3*&sDzYK0uOB7V( zR1Zg2pgu^r^KGY~8O2cupGbumcRA+;f9KG1pH2naz9b^X-lS0mws?DWpM+4J#O`oG zpYFmQK82xjq(=hG8*~9X)5WjD9rsgm?P4YMJ-0sM$azlKvEgx`pDi$@LZ4 zZJ>UyIjcXU9D5@D+vHFD+dbz7y#Yr+C#;9&Hn8pZ5F{#Ridp>NN2at46}9kbunk_v z#-5P&7%ESqcl&}pA@}GbQC4OPq*F{<#Dac=ws$4X^822bF{y7bnvAO_Bc)CJs% z<4)kS!~pDDaq%j7q{9%V0bTJeY7B7RNp^sHLbX=ciGwZ}!xKI!&;n6_xqN}Eeq4CQ zH$co1u`RyY;Ue=MNNJD1BkFjJB90Mdg7xe4WzQXY!|D#=|6X)+HL=)Y8uLkm|l>ASChN`PGhF0mvuSK8+BdiU9)`!tXbe^pvk* z-NJm(HD&26(KK#3+deK%AZ^yEX;a_4-K4NAtZ)6lgd<-jIc7P==k-Va5sri|E8Oj8 zn{=g&%hh|lJX740<+0%aCssbM71K7mBB-sq`j*ic)#LJZ_VX;;4E5Ht?HC5$*PSS3 zujQFLYD{OT&5I0Iao5MoYomz5Cr{pBHHRx*a}~VPPij5vZ^+;NP2iB2z}-b>yzpKj zp-$ebAp#TI0QL*|h{iXh+5gLwr>gfXR3h;jpY8s;!+ML^+PZjZSKB)Iz-+^1sli+$ z$xjQQwu6JnmUse;mNPDAQ)OZnpyPH{dr&$IgK!#n0jF9BS~*GU_K%#XT9Dy`_RYWt zSSr$khIs*T^BhN5!#gpr3C!1yWKDCRS>hBJ3*>rz4V56>NJ5Ram*6hXE$?aP(Kmcz zS0~1X#gASfx`|WSIql3$6NXj5U0RB?Xg_6L`(6Y3UIbK%1bPGxr-b~LM_-_^u}8#O zt7*GJ1~^xsgL3>b_+3)~DkBrDt%j--fr9u;FtSwqC{A=|o)i;*&9Cigow+&QS{kkV zB-mY5t2ZB&QreAQ_S2h3Xlh8c2XHT-jeIp%aoedoMpnT&a|OA$r@Gyiij{&s3TvWf zbIpiXP7l4Z{PPT z!p(igkUr!@k=3QyjC8D5$PvR;M~dbq9c#3lilZ~!O%z?0a_4q!zIUL%!;{6>0@eLC zf2NLuu~Ek?$Bh!TVi}6`4a)lRY7tJeD#3D-J~6}MXSou0LdA#$G!=mOVL4`-SBagl zn=(R>9^RPz{L!>z?A^gba>xyMF>q^PQ^q6QSpIc)hFjc#s}s@9wXjhIz@XT%6T*^cniB{ zOE-#HAhn2Dpeo#Yi^F-IL7pG(HBi3l*6Hot{pz?WWwD8(bcL66{=mTmuG&7y}k4^Srkg(fUmT(LExYyeIZ|cvkBU$M?~`ASE!0j04=&0Ol_- z8XM52ow2c1C1GjnCPIs&vb(m|K(s$dWp;XST=Hx6p~sBzjP>nPO7}JDbQg<*?)!xV zGU&F5=MtCY9l&fQ0(xpL@K{gZmQqLTlcjAse$7Zv?0_DydQuYEL->x2ooLV{iSHie z)<#Nd)Hlf}MhX218scric@@1tVqIHfMw*FI$btO%`-jIhtGNBOg@qnc04X6x6B#J} ztRqY#UqR#@6Y9vHG9J!H6v{TYWXSJQK3&9^df2zs?RIp+E~I|a``wU#G|^k{b0|Nv zNY4I}(Bs?FTI#o5K5aXS3Imuvf$Q=fhE(XNJYf%;?aU9RCX-$Al*HI|mGbzY;V@*j zibX{lsLrt*WBq*|0Y=Z7C)HxtVOBy1DN$JmUP2nhZBBnAZ0j=M*f0e$4p%CyrSytt zI^GF6%jYJi{hnK3rmUl-F89v-3WKGt?+EJIMx*xVe=us{7{H}4swri;AMtM7gSThp z*iE+1yqB9ty(lEkg$yQF9~m-8A6idHuq5bSpDaV zlIOqGC}YHvN0oXUj#GrF^)=1rk7&He+|0e!_nK+&+cTX5@VFO7s*mto>zrq~NEYbC zE?dy(=fy1~GXIfG2ng!r?`71z>XAI%+PRccwAob#kwy07KIz~TL#bcq&7oQjKUT8&=oLeYk zQKRESqoM9T`4r8lLUEjS3D^ncragrC1>wg-o8w<28YKmhTj2J3*LdldYikw8OoKEta&rhSgeK@ZIpH^9XWs zkyNtIW}=P_ZtWUAkM{=mllG0ev@u?H{oj>&yo4ED#)6zs?dP)$atu9yugv%5C^5+1 zlJwhE?CmX~LFZWm4g+V%g+JoJBG-!BsDn`qyAT=!hcJ|=ocMdxZJqqE`M$yY=REjn z-^87CCQORJshnnml(kILKH#u2((B=Ihn+QKTp0mTe2(}V`htjb{VKG;bKoUv|MPc za$&fbMc7(^i%{KZ>3B3`PR^enJqgT_eK+c-y}GQ(jfj+phVy> zE4~oSe%hCcI}lg8B}LDegBJGZH<+*^f(sdU<@*YxHfJKBRSBnRUAAH*(`OD z|1(eEH~1#`Ru!pj<@{~OU?N>RKe&0|n-iPEk((=qrTYeLp^F<^Oa9k?`malZLr1%P z{Ii|#T*jZaL%A`Y`(2ENmJ0Wu{j-CD8T%BH`tGTX#*X^`^&8$qH1@vX(VNeD6mgFw z)ufzAYv=Lre>)iA`)FZPqLdcb{*v``|&p0<4Cd?G<=4Wp&I+eM{k6jKgY8?l?N2hSLu#_ zFP=&i0=|!8s7&eq_`V-Fz`-W(*Uw>vpCe!H!?Ab__DTscV82r|H4A!X^dBBYIBbAA zzpUmRE*}uqQ-iZ8??Tu1bj>t-6iA`9!Uh$m!&2uwMFlCy73O^(HjP~_=Kd>p?^ULa zh~8gO4P6p=GF|jJqc>CTe{TJGS|d;Zf2oV1Hwo+588EB5(|xYi z6Aft@7#^j8U_A8#VEI1y{vpOokTIf>_e=@SB2O>*&v0oVnhRcsTw#;hru9SBXLOJ@+0~0X3^vB)xgC$u6`)OQ+iD)81Us zImye2$FeDW*5SdOpsgLw*av8q%Rs>u2d6UBC18tRsNTLqW^sfId^0BKfzTllU;}S= zEl|)2I9Dk=Ho0g^ihh6`kR$~sr&3H(zTRp6z{99NlhzLh%qlz@Cbo}%R>%Kb>z`4E z%vp2qFDEU}WlB5PfBnc^t3t^4lG8{rMsP*Zy1LRU)#nX}Z8_#t(i} zsJj3zR`I||xbmFDKA^}9x=O`btOv)o>f(S<*HO6l&>Ao!a0)QIp>e5VUfQ>Thu;x! zml32!CTe@yUjZD+Y0LoZA$yl{fxQHSTY^p(i>&RZ8?&U$&uRV zx2p?VFR3!5uupzI4?6(X$~?_zXiLvQcre{k;>&>ygd9`AGdAl>596pGvpT320Tdl> z*X^n3{STFQ*Me?F^A_^Z0o^SE(X9V#G&vAp`1k!EVvGKY{$-EM{-4q1rSeqGko;*Et*gsp z@lQ|qd|-!)HUF$zohQh(&2p(lgR|^B*2H}k@oChc@bL$eiR&V`*uN>PP7l@AgJ<MB;nWXR7NC-7!&q{PTL1GPV9KBx)^Obq7jd7o&V1qd(S+!+T@#KfgzX`O zV}sXF`Q-8xT>LWNGrL5Y@kQYux8sd5X%%IlZ9MdciYV+-0nN3DtRtFXXKBUY&%+(3cm!KoNpWN zow=~C#7Egl`JF%F;91t@WnvfAPKpakKyRy2#{RRT8-R_M?nRo&{))ux1%qE9y64dS zj4=^PKje52H{G54OJmKvoO=d(WCKBqQc zonGlZb1!9PT)o>zUMP06k=Y|(1jcpsi^W+)##QDa*-~c=jFK|D?LX7mu8L zRaC&R8Tlo!=$z~3{{FTBS-0t?cO#dP4fkCUkqeJb)BIB;{8^RPCFVu}`Dv|`cTYsoVf7n( zLJv{g=jAZjL-dDUu^YF4G8lEw>Kt`<110Oh$mK`b`r7g3nY{ZzNNkFz_rBwws^xi_ z1Qm$h#xJ>FW_8#e;SfffK?dBfYD4=!T^Bj#gVaU`0HvJ5w&lM7A6`*L<{Xk`P7WA((!n8+k+6u|XyrhrH@8X+0P; zE&gz357iOkkL6xnaXnUdG@kxgw|{zUxpcCp4q7$1Zh0y#L;3m}YVe0{B8ymHw?Idq z{@`T22%V8|AoJ^2xehTeIu;gr6atKgqc*X>@RO5?pRwqCd*1mB9 zypCz$S;(%s*s$P6zyMqZ_3Pdz5wLr3gEaMYZP37T*2ea_E+yrui4wz&M>kYciL47#|vn%jjW* z>7aQ-&-;X$1Sk}*OUUw9oR4COypPum$)`JBW=k!vv)`LripMU3p=qfhp%X8A4fhA#0DZ7%C>o&%B}KA6L@9k0O&Og zou+_C$_?1gf}3>B-#(Z7Tziu|u!siMFhXsR&Vc#l^ab=X#3Wy(JGOyw$B)Bh9p_tJ zKE5i@-EWn~elu+|UEKwY8#kH%E*q+d>k@!C;N)AN^e-JfOxa+*Ik1^(5ajhxfxMCg z;0F}!ZGW4o%5e3G+i?%p#~cV_+5Ma?EAxBoPaZ_+4wSxMZ`nw{rmeZtn!J^NVGcAR zW=F2Ts-oAUun=F_8~~Rk-~`1Qx(p^&44I%qr9vVsx;5pIz!`w_k*S%pT>yMrv*|`Y zq!#RZR~d(y-FjAE40`yq%R|Jg;aZ3$1DZO<3_F!aFMt=gI5{6^MfN|GZuZrGeE?RX= zmOkz3AGK=Eui_Pn=3DEPwE5W??{?pfnLdx~3ga1`^helvOySlF+Z_i70#9(U`ygJh z8@SriL6|oW(lo+oi(Gf~1p$41%8yD1(CYA-Ro{t`r_3_Saycd=m`DX@2YnZ}JU?o4 zcOUoKIa}(A`v}^Y$LYs*jn-$bz(~w@HP#)6{TGH`q&qyoFY8(6a-77i_K}(oJ*G)r zmdA&2BB<)Cb%`*WR*s*w*9D)R=%u-esGe+RM*VvFr;7+?pR5PGIVHJo78n=UR+bJhT40nb5A%c@h(`uvi5ZyQK2ESfXmOwF#( z(j4-_h@gj)4-ywDxLf7sCW{QE|0yi|P${B;)k?SYw_Er~)0vp9AN8W7Et%Wp?p#mr z>j93Zi(rFy)c`xX@u%vi#?oAul}in0lZlih_;v94GoY25aRleYDy!N0X($AzI}C+78*KW6K*lJ$=J58=5H#fA6~LP# z1K*HSKzWs&sdZaCM97T0=F$QV7AQ{;TtXOWQ$_Gc09}Y3@C^BDAgsWGp6A*p%~6VE z*}*|-j@P1&yLJJU=;V55fjIN(?Sq#zVMH8!ZdTPUTd%HMcirK=mO;?F74*KS#Fd8X ztqU*Xj=gYOzWakg>YmvoJ>wnI(}H;)oP}*L2DCt2mZ+93PvVDci?DB@`T~e%dk-)6 zUJFAdX8`?nMHg`1BP^Ol+gn?SsoUyg`PCwP4+RYMs-gBI+Yo}7S9%j=yn3X+&Nkp= zY1)aqOsZ6BHF=l+Y_(v&j9!PAlE_J)$zWQS-wiEc{q3uD~~|F|RN* z+k$XEKdaYjbt*1j^x_`HRav<%)=_pP>iSs!bK4ROKDpugB4+u5m%rxg!*=NbV`@PG zM06>6AorfOvn@i~qIF(C9VaxcPaKPZg#=8+}+`kB%K?4)iz#0e%ce4TIS ze~gxH2y?%>-W@-BjZ?m@Iud}2ohahApGoC7#(3FQDjn{5hOuMA)g-DAuSuTNwKR~3*iT99I2#=6R5b@ft zb!JqS5}#mD!coEdA5g_;P!J(r^Baw0!-#iPV-wSgOVd&i?aQOQ@rktz5ZW<2G+S59c-# zF3Jay%M9Bknmu<`nuW0)e5q(dXgS^RrU#OkXB6)1=rbI3?l>-l0Ke-YZkdmOTY*lK zr1y1ex<5^{rpT_>=D@pbzVNo7zoru71D!{l$nnFZe4QSVtAzS0ArXIQ>EJ*haRP@y zyxS-IZ;qsxk_KS&+m;|Ip2StSh!EK7{fza1hmhlrnGsJ|q9jvu{!oG5WNZ9!5L2-; zcTW-mW;d7dam`LW4^Nn2yYC5HJ|u!+W86Dzl&RSw_ zPfGIbGc5Ha^?OWMIlh4N=UGEru<@b$_c(WR2-77%W^14+Fauq=G-+95)vZ+3N(TBp z2au(KR_NwF>Cfw-NF+}XaebC!KD)ot#ZFHyX|AAA>RE14^O`ZVme{lXd1rQ$#~oZ0 zE)+?3kU?(-JR22yU30EoC25|Thx{0EY!M#ozYYOhq0^--np&dvBroDO*&Die#-Jbl~HhEyqnlTvL@SJKMnL2=zZ%Jlv0!V6l=dgK-E>W>a=oUmW8|eOh?9fBSx{O@~4PlR;~o zC~9PmJ0-Df9{?@)6g$(!8=fDK%n>(JS?(Cs&8pSW8s_f8;_TjfG)ntNt3Wx&j;Y== zA3uLDgH{DVY91vU0@C^zoTm`#rttIjtBV7bbea*2jLJ9-l5 zf#2`QssW<6@|n9`A7>00lwW6#JQlqFnW~n~1n$YSaBqgubM(OHw%FfTmNdp{AR)W zr09#lVs>Rj*P!&1>N}tjc){?c%A=7-gT$zxT>wDSpgH}-)eD@A%|_U$^{o&Mh2scR z5Ts`2oc*|RolkZRzeb!GVR78`IY?iZ@4VBTsUVWQ#NWTq$@6CHtW@QM*sPML(oA|Ft5nHDw3j&7%&?I=llSG_=9wBP zle$Pl72VlDXqNt43%6z(F1v1X2zEifiKkET+>=IE;?QB9>hT@(*gRUP1Gfc_cb*LT zU;Q~|Lg941y~uNgV4t7V=+?;=l*tK&x3^1Euamh~Jb z1sxYr9>-fQRE9yuwL6^=!{cavPglaVR%e4{XI?BpqAbatB+?z{xNJu|eq$r)njWsd zIa_uT06&T@0*((eu$z@J9=B*-KDX_KlWhJ9y+UHLNmabGRx)sy;LL6FJpvZArcxtz zk+mVY7Uo&}#b-k-+V=OlYk66}m2i!0QR$^3QfqUS@ zI^Uc@XB0m=*;dDdLJIBN722KPf+XC<{0=*BgQ{d@lSc@ezPnqbJa1Zec*Be=VJua|0W7d}MBkM&u6(&?q{ z4bpCGv^s{aBC)F$*EMir!s%7gb|KxKt8*`_-CNq7s$1fDg@0|iDx!G3^ElyPIIKfd z9}%w8zBvi4z#F+Mo_dD`H^+3kP~Sev=7)Sv^0{|8nM@*(HzAC~`pB4`xCfNdm}m?D zqU@x7B#=oK@EFrZYY(Q%8DlgR^$wvggE8|NP9W!PDRq`K&Rq~>qpMDd6oAR34ctL+ zUL;~<+NyaNMC@`xG-_RX7t9Qz%JkGMXGbV=u{;`@MldrGj3_X$-ln-oQcxWneVR`j5@ z#qgsWEzX>vnJ826wexb6u+IjN7B+m2=7+bV3JXYOPZbWt<@NCzrrI$0ZP7 zW*sDg`G3}3tV!I}(CWR*l`D=J9xkb*{|*PTg%<7L0Hfv6d6-xE zmvQU*q#5UC=GfJrAWeqz^6SDKQ{ax~LB6`l7ut0hq04ZwLrai->79c)KXoEr@_mnv zN+SuY&!Ui#U)>DqGj<*da|SI_x(1#Y#GQy$GU|omN!u`(P2f=QD4Bh%x2CJNvEJl7 zdY20WroJqZ<=?tmo@yL}^2EpOhcxaNIX7nR&}t7g=S_+BKJ>RUBqyV-lxSyt?~ zJbFn9dwZtSR3eTx&Y7rGN71LxOr{oB6C4K(za?%SMDW35~ z5>x8~l<}+ms^&1nmJf)9_*XvB9k&TRU1`}pP*AQIsX}nd!%IRApGB7+maL4y-LVq^ z-2*_JsDV4;%?55g6(FY__JHn!bQMS9L>$Jc;MP6JVL4;63wKK>f=1L&E{1STA)UJJ zzAg<4S>VW0Wf*hirr?DL!By(J;y9VvKowHp`Vj0p*+2l=>_v05;M{b5Z^vu!2#{sN zH9u|z7sM&w`o2Z~*r4V9x@bpFiCNq0!EC+BEY0|K&3qS`xY3+8^blGyLoSc(Zin)N zG&oi7PdlSo1yn1$JR^iR@0W}S0x5f7B*XBJmKFJVZPM9y)|UDzgl3qicq*H$MjlFZ z1$6=D^FJx)8}hk+ObAa-P*hl&vE`vKNaR)k9dh!URsE0g470ZbMp^x4!u%{T`@b)m zi>MeW%cMaan$d<$zsO|jx~PvLKseVT9wYdN_C=V4Q5Wr}8Ant4QPbY@m!jgJE@U#= z={fg`%E0Nxxou^1Vqm&y4@35upGY%~Ffe*Zc7!&Wa4&IU_B4d1srgA$GClR;b>8&` zPJTV?XV=eFoZhY`sqG?3g=c{;(-!h0z+&Ou%Wrj@zSb!CDFi}z+r<)BWF;JDm90?NYuPm-uaG@x z8UaT%k6QIzu1l)c831URi=^z@EY$qC*H!<;zyaW0lPx4JdO#Np6#OkS5)A7C`l+AD zKXb)p$n9H?MJJO;N)*pmv)jvzp{7%$E_n=N+&nzs#tgq3(~DoPZpxZBLASc>fv1)Z z{)@CJY_Ibpto0cdjoi6Jod)l^eoO;psI*F;(2ub%oPJosizC7oZzT4dgq?xwJJdRk zl0G_ti$sP+E23bYFKFBn^jGCDBP{?*^udm|5N-z=Ee}jO zaD)K%c(0g(sZcm`0y6rv{nY1ZpM>L$Mzzo=gXfF&{*>FnN|jtJNKLrM5me)0$jygi4aZ1YDPqrP zu4G=jwG>8kk|4Cb7R9V&*D9hi&}2qp#AA;mrnrI;KRKZHr8)+Ak6oZ#_iXN+k1%e- zgW=;jJTBqQ*X;~p%)1_^SK5e=%q^fNs(>W!%rJz54nSI>)dv-tt z4TomQryDokh|vgWLwTAi1yw8ecqC+xcdAw?c85q#1rVJ&3QrN;H2a+YhO2u{@R5<>Sy;3*lW`jc*MgsDIWN@YK5wSo$(J*)oI>(S}5xZ%n)8ef9MH$I;N~3)~{NN}|1_-q& zuB(PX*Vqo%R^9Ka(ZmYT{Bl*Ry&mP}6F5+$jf_K-2Iy3V65`ax<%aN$wmMx8%VE4bzC;4y^W)B;H_kR>@UZ-x| z*RPrL#~A#K(Pe|WlMHVT)m$ZL>f_o*eukp_l1QZC1eH>g&X6VK7d%6bqMLgX`e47| z6a+l)4+1rW^R&=@I-|Lcwjc4*1z+;smh1awIc?lgmzy<-PEkTs1#jhp`3*zDgc%Vt zpUXhDG8aub86u)o+tt_g+&>r{X;i=E2dsz>jsb62H@V@ZUGMl@Kh<) z(;Z>7(NFl1eq!R$Xq=FqYSt3wQ`nN!Y+27zV3Zz>tWSSrMLLHr&B+;&a1+Cgan4|G z(KU8J1hUobS3pC?4zxxJ88juJ5yPxTXNmbL39v9BjF|e1Xv3v1)jt9FLxP=T%iuhS z-egdd)U;2$CfoG#03_3xXP>iePI6*t4C~ zVRx=+MVLF&rK`~SppQR=1}cR&!o@?977F|LYXW{8cZhR~3e=9Y1Z0sHg3dqVJp|`j zia`^ps-t9~{V}B%9k$p>M%Cd)YEf4IVJ*E9r)RZimB{`k&Y^%D{=`Jc&CziHE$cc& zYRzjn_GVWBhO;jE@s&MW=>?KO3VomGZ$)F{tshKo(dUJub+c<{F_+M-XkIE1_4BQ_X$n$!XzLR z>Uo#+G~TnSB->oGbn5{XFp!4Fyek0sG~zKSHLki54E5mH1WjWWUi#t~e=-dD4%2o2 zCU>-jg|_~JzZqwSFO&H4D>accbGjDEME0db5I;@WD1K`LnNhIbMZepzHOa`UE?QsT z{e>`&w_!QLX^Nuh6s)K%@z*9F@hR>`#Q}#u3x|`XI$igKn@3lKH}FEkuq`j)X1n}- zOMWmk4dw(*cH1Wt1lm7=nH2xF(uqIkJH29?oB{85-GR&6j|ZOgOd&(iPC!;ia!RXX z%PfRVtKQSC3iPeKlySvgFD{KX0t-#!qjb-GxkY+4P*J=Dh2H*Pg{_>wm63ygaXISNh3LJ+B-2A;qDemM}_uMl6{9ga?)!L_66o8 zol2DLjaF5l%cTvBLFoG4^70%~_rmE@s}sO8wFjPOB%%(=Fq3YzJ5ji? zBls?1l9r@Gp<%)cC&0XH4U4?t&kj8Xwg` zpq`KZ5;~duI;k=Q{~zQI6GKG7Z-P2OzjEoxQ8_s0YwAzzBc`03y}A-;FT_WWA5z8@ zs~e#vrdO3kKFGW09{GXwBGz-%i?E;YN{4yg#6N%`3!*!N@f9+X=hn{wpmn}AUF zZtmolY2^D^_&=i*ZV6~>sv<^sUG5&m z)OrD%ERg81+I^@If3cEcy95$B%G(@t#Q_6YhsQv(!ba~DsP#;H64?;_LI_=v!=r=A zm~z;4Rk7pUT7u<9FiGBv=2oTf$@TY(Vbi8&NG^S8>PMBS0k5kr@Tex<@`h5QMc?}8 zN1_*P;DBxga~$#JlYc&MO_r#Rmori8&%wxE<(P^;ZP<$I8NEV#La=xz^?}{)$+Bv* zTwm;))FGo7aa^#>zm|~T3TsX#jqJmi|M1?psi}!2bq=@-|Bwf)ucy5NUT39z73TdE z)6I)Z03mobkBkDmd)1!Md!7AqpP9CW+`pZtUssKcg7UHf@(x89u&Q%84VVH#jGGyo z!fq4l`@4ved?Q6BCT_MPnMdv*7j}t<6J_xjmBXdW&X0F6THgc9M2HL(^i!i^EL3z( zCl!~Ig30fbS{Igr)uss=<@Y{6Zja#|mQ(|7_)(7jedZ?6JUv5L2K|F_gw8o{rt;s7|>Kz0sH$#h1Qs1$DA-&vyrHj3l<4CUHf97mXYnz!mV!b9_Hl$ zgP)f#`WDXtJ8@KO^I;~C4X@YEfAW{v27SM;5wsvgsr)jTPLKaRu0xKD3Ao$QhYm{~ z_76f>9_&!vLdcelp&1;Spn++cZmM z^Zh%A!?S5(SBGglrNl47th4=7plXe9c=I$r(+nXK;6+ zYV1*K43_)>pab_|dwa5?pP;MCan*{tUSxR>)JlX~wxCB{?)`3yii`jj2l`GufL3J+ z*pGvN=WPB|I^oR}PbT_1vP$j#u54BZD%DCm!#6f)ky z-IsgRL%Q5qtgzp=gvH$<|FeFl^Pwht#c-9~X}Q&evvy*EDBY~`I8(~C5cMZ@hS+N@ z!DqK?^|h!*24ov!BwiCRBm1Kg9^uZQ18_RjBCx2rto;CKhGG{Q%1TnbciS%+2miim zhOd6)1$`%@KWrZwOPMI@;UbqNQng)k-Umg`CB#5 z)#QxQ++@|yv1HU8bO^Nk1Jq>S`@-S>DcO0oOct(fu$H_g}KniWG7sN(H#3!BecH{5J73bsjxK zpVXb3Nc1@=Z2*v{y~+vVS|q>SUDEV3FxoG4 z+L{=4sjG2~J|SI{9y|e*fjlNeTO_(Hl~}saAQLGIo>XbIa{w(rmgcb@D{4yIFac!> zV$>UOWF8Mf^+Q2h^8f+}V7p5I;k1AU{gn%i*_Y|(hC%1>5EexThq}$bcA(chD>G&! znLa>bwhcTislcjnkz4H#wn!t07w)k)dNDih9dH|q4xDHasZkKo#qgizJ#&6GY}<|V z3VW8H^8yoi;i@0+Ifm`fK$jcU!yettUP&-}tyIk@|22C-+8HD-aC9%s)nxtXV817b z$KmHMUhBQF(EGUMskC%oYBjh&46(hFl=6>#sFlPh*O=A_RzP+j=gNbVy#w(>umHU! z4p0d5P`kl#)~c(Ix?F(g+#v+a1QO7w3F`Alea!_MEFn_CsYY*Z1hlHG8IpGC55QxS zC5=k6E1p@sce$KRil!+#o?-z_L;+|l=pj*otvg!257UUkQLF{uUPAlCTJIR+@XXcW zrBSD&BWKHm;&lNnJ90Bz3GsB(Zw^WNx?RY1J0Dm<7vj;&=T=C5HS7Cz&#stAMZ463 zI#08Z5QfA<@u0H(yY81&D0!3~t#AKNErm{t+0O}^#ol+c57+PS@L96xz2zIejq?!$ z%tS!@zTcUGM;@T%-)~2&knb15-Hm?z6wHma2g~@QBHkBceBXeEW~>z-@&$OGUb#^< zK8q3lGs1J!h-mrZ7tPl_{nGDBp8t7qe!o6ce#y9%ZpSgHbqmW=jj1O zMDu%lJ>J!F&=``-95Dd_U1Zmq{T(CRm5?)jn<-VkIZTy%kCDWO&@eH>qvfE7RNPG- zZQp@8^NLpMA_OwWGc~k7;O~prbLOg5rcT5@R^Ts-+Qj*>99Y19dYYHWcL1Q4JVtrN z$zSaoV9UA>miG!QYk-{mtU}>nhdMcLjy-8p(FtsDW!LK6JBWMmQmsQ>gyEmnh04zY z-|mX&RcXrNfA*Z;!&)p!^|a(xpQe8*JqJ-g4yj%VHYJsBo0EB0?p7V+d!dNAIfF_1 z?QUE|+}69)$nVkdoN8L;_6&CpafkV+ji_{=%a}D|jgR`N8aqUfB~9)GMj%Mi$b17rsG3|OLK7>y+A~I0cM3%-laug(pMs--LGQ>dcqZf$x`)qq2zGJ-5^k= z-$ndh-=JOZc{;R#tEX!v~LKksFH(-@yL zw{`#PRlF%8=y2XPg0%%Db1k|BwV6PqF?yq8`nF0DU_x1bg*-k6%&6*`Fahg+Xxam- zF;(SsF*Z={M}u0n8?5gw&@`(2S)jYgUu4tBjb=;7bs%jrrWc7S7kEb{%=mllW=2Hw zdR0os`mZA8Pjz<-4N8IlVSx0bsG!k3IELBzGq>PlR=FAg=?gyW^?|0P>2kbeZe3d# zQScj@=E(>$LA4%KY>IqmxK0K~Ig(PG9cU!b9kDQe->odwXJn={nFQjExZm~T|NJRI zhAW%M`X>3}0lY2cv@yaK`DkI>2E4EZnMctAGvDX@0*LYfIk&nGR00~G)E*um@5~nJ z)W`fOM-?HeBB5xa`ZxRZ|BR|V$ctnGxB4|2qjYpm^3!wb6eb+@Blz=OKc>FGTv{s)Du}z4OnZ&>BBz z?kbR*z3Kr5p%}wJqd@ZO1Qat42P;aGg7|~1;4E3ry?s+dJE4J#W*5|d{IBiv@1@%- z-U0_m310@yLU6}qQSEipw83(q;IZzSQ`+df1pfQ-qPjr7WP{s|5xQUq%AOx=vo{0I z00z7jMp5|ZK7QZL-#`5X80-LjLN0+NAr%P371ursP{dAF+k(O|xg3v9mI{zW2ls*R zNGA|CvH%y75tBU}UBL9YyK;~4=K3|@*U!Q=@_R=8S=9c11p)`HZ!HzX+!=kJu?M zHml`3aUXrP9A&nyIL70RDhG$(zp_{Xyr;!IDRqBl+W$;k-wwzH2#Zt8Ri(DR28$gL z?7(4I$oVH7pr|{QRcj7O|9^D72RPM#`v;!m2q6bq5eLU8n+CG?C_)+8t3_GSvG7!x65i2kcIR2qshdNTCBzwar}3w{(CQkPoV*3$|e64kOe9}7*ljHIUE>CN|kG) z6vhSCAzHtTBX)P<#=U(Fbbgg(X)L>6(+EvKbHs&y_}?u$N*w?l!e;Q6$)>vjGkxXH zMYIDX^5gAN-cxNZ=H<4AKhRmm10BRv?t9Clb5x3i)FYe=56@un+{d;3rMXV6JFj|QJT?tDmd z3tV(GfpEK-wlDjC&mD#)0LpTmOPy)YFCca}pw6K#m0Eg?eeVtORV0v)d>$j;eNTkT zD54Cb$u7F_-L+D$&LDT_FF6ZhVs3n(6P!`l+w|lI4t!wh0V_^Kh}oK{5VXNeG_{wD zA*()&3V?P-D}+%yP#T`3o`8AJQWZ-O!V6s{G?5`7h#uEvvVFRkjY4B5A652*tl+nH>97;Q}QAF#RAE5|#B z&WhcAbB7*o{GfH0G2vL|5;#xfD!@f?keS82FQ?{az%NrBP?f;AwlicM$SutMxmL4mrT+l>@wZO_m5* zMj_KH>rIT29*BlO*2;Hl!(%ZtyWJ=8RMQbR6u}7(YX&HV0$((Y^7 zt-Q87E3;p8qFE(rr%!;=@{1<`$L?>hJHEujol_%?K-Ql*6peg9VAHdXdlf==hEhj# zY;g3#(=*Ja9d2-;w=RGs#SMG$hoPpMQL3NzcOj8Z-jo&(U^9s@6953NQis{p@ssy1 z7q7uQA58Gl!@XkDIIHk%ww(O z&YSetB1y7gMYI~Dn7_?z5~UN~)JUf>FR}T+_v+LRIp*55R?T2jvnl_kgfJg@^i=;j z;)o~2_^AM6TdRrMoT-`1$CIAc-PnHP7tw=0t7e-%GC*|CX5%Al1p5by5TXxjn1+PV}*$tG4Dt(@DP_f?>n-+Aj=VzF^}q3J-0HpwY(O zMa*51>!ZVO>7$O^G)gdpu4*eB|84y483wB?y{Bi60^PiXoR{ee>`f(d4}e8t;bdq* zT)x?#t;4=a>E)?M#@GY1s9UabmkS!?reGC>Xon|T;??rMv(pEiOW;yF=#$j}($(j; z)WA0E37{XI4t=Mgf8!aTP|s)hDs4s1x}L26W0_X6P>(x8Ag@3xKD)|dvYs&k-Teml zgL1Rsaxo`GGZ2+8)OoL|yu*4+12$n6SSm+IpX$j<$lJ{N3)#c2R!5PVJBr}TM+O!7 z+F&N|#;nEPIJij28^?}{+lY9a$E3FV&Lcgzm$NX!sSQvxwzB@ ziGl5m!h*XSP#i?00ZdPM(e+ER7@a72>Vq>=$yK2+q(1aTnMUmB@ z5vIfeJ{mW4z!Qk%pangG2$>G-!w?aU^D8Ur^n!l&e6{=s6KNHS5wynlf@c|UvG92X zRufZN*MjlJ^3*|{@zE*Fb-B$gsBP zhGLtNS5eeiUqpDkzgV&{3D3c_M$~&s*U&5Ipw$D7LGH1Bh;a4~sJI?|OgrJJOF2~i z7~tZOb!m?m^?NExp|WjHU~ZC?x_3(qjpaR@v80s%~BjU$svq(0MaT)ebdWY0f# zki33>0JI`h<-CUA@Flz}s*mKQ<2lo9|B~0F&eUG8c2=@Y3D+*U9tbSbYclrln<{%t z96b!bdu(_cONu&7h<+25%QM)bxPIWLs2#vTKGq&e-wP(F7SXN&#)|Pt#*wR@4ub-D zG=h0>jU07(_gY%t!N=&~^UnprhXjj)BSpR#J$#W~6#P-G4aWU4$pCj>+bljOTK9iGXwKRsEz>Ep*<|-9vy)6nDaHlWxQ&ck!g( zmbW16?&~MpSzFtne%qe zfhXR_?6->vl8Wv%kWW6*xnPrF&YLM8c^37Vy{+#-&d9yt=cv437&6(b7?4sfo6dA9 z?%PEQg}&ppgT~e|*iQ*LV|N_R5Do@$s8LMXMyRK1*+6&uqWN1FJD-cIaC&Fxtp(+; zvN4UzN4%;TtMgXFq7VN82S4gZpVD&1l+3w%p@HCF&u%U(qBU2LXhrH8m`x#p07o*? z=c4nkI$wp6I<)1Mm)e`<^q*+b3yKf=zWV$za&3Fx$jT3gFq+@>{=IaSsR01q*;R7@ z5+-2&EtR-uu%fb&I+OI_mth_^*{u$a_@IK4N8mPd%W&LFY8TKfe?s`bpgs`0SG9cq>NIIUy{*yNm0`rADBjboHqO{ii3lr?;G4QTt8rz zPUR9%Lqng3Jzh!|!||wH5gb4EwcMi!n@2`zbp~83aAcc@=)`L&`{aB1Y?6{A=fMuo zg5Yu-_lo+!0T%)HX(mY(?C0vc1h$dFhu2<3Q(L_iQ45|gEI(1qLRN|iz=*XmG4$Q_ zYPk2U%5TWnIX96*fRl2Sm7_n3Lyn@oy{*S{((?wAZB_S^%Zt>^LRMZ-@l^Z4G5m<* zxs4y;j4u8)*froCnb9=}TU45sN~mfd-JrVWG0*~fnB{d@@Rb~t^@)zzO~tI0Axn-T zEL!gsS&~Hf`@{72e1JPql1H}+<~&YAnREx%@ZB6FG3d5KEtL`I1i^r3#BzvkRm zDUM?ga|-mPGit6;q*9J2$d;1R;=0>ld)N=1N%7Gaen{ep3saNjsJ9;~z8Wv<9NFN@ zKM;Q#g~2|whZMlg^Epi-)}ePwR5Wqqw8)CN2nP$UVL**X&Kpc3|)ZBqm5@Jrt z`Bhx_>ynZ33iF9yZ;>)M;0>qc)?od@H*j*BrSMaVW)f9&bykp9II_`s42*UVk&NFh zfNd3A69vuQoo8QB&_0A`Ry;Qi*M;fXu+7~oM?M#W|6_46_hy6k?r8)(M{8eXEZGTk z9o-OYn9>s=e&#g3WIADLQ`4>{N&_4LS7`ZwyCI)e9Y|SEr$$sA+|*a;Q?}x7JG+)U z*>T@6J~ix%VBA|SW68b~=|r6~hdHc(02LcL8aPE%e>gn$3X4^0JL+9giH#miD;BL_ zZD}8g4I~(@2(M?@#P_ml{PA>-Q~tiASs7u1M+Uicr71bCTf!M^{d_+s zLsvh;1lb|s_NXtQ2SKahiuL$PxBQV#U(N&pugh=mrb0pkLV>H&&Xd@2J+?YU?tYe=c% z00X#cVRZb|Sgp}8jg2&o!yVOpame+4P7NeR9S@`&Oz3T5r zmjvU8aXXkS;>oHTEsN~k-Mp5a2SbRkla=;8_+U4JY^3^EI0ZITfL1*?RhZ#I0U?s) z6M*~Yk$QYani{DKEo%-3HiJ$WJxFKduRx0@Bg0T(_sg#-;RWypmI3T2Qik-{(>B0r znR&90`j|jFi_$@M>EE~AM~qQ57>XgdJj$Oxj2vtx3dGN2n9otCAHy(qpZufLNw*v( zTEl|Mx}V=03g7DpA;BP&Go(3GA@C34e_LgJjAij=$ zFEbgw9^@6Yj&fw>Qq79a3GHH-=VGd(*#G0M2$5cgY*xo-$(9#KO7liB;BCDa-5dH zfFj6O2fo{USiIRgtPo&3t?n7!z$nnps}WbfvRvh3HaFH{S5V(cm2+$X7(j!RBK%+BHPpe=+1TAw^!=WBg;aMBRNy{1>&-fu2Q`&$7X6 z;{i2vt#5cOW4rmhE}`%uq#o_VcdQ-P9^47zoTIWnPCS}WT1<>|2`*yRfY-8wj-wb5LLSJijOaoeE#Tym99GvE0I|B9#=XAo&Z9W75aALRz7}O8shVc8 z(90i+@>zQkU5-|LJSKiLZEvT2A**h)g)WEG95hn3Af$4p1(M(E28I^d>>_UWadvQH zeBeGsj~aU^Kwi?Xq+j2=Q`6`yJL4s)NLY8^LV0gq5itJ&#Bb{n6OvtlQaTFL0u7`m z0EW}j(sDwBAr6|jYDO8nyU-Y{^}qjnsERaM#v1zevOxnBkbv^ZdBYix+;a--1gu0z z<@TZX`R0eT2Ru*rGX3+QyD;}Ypu&5m6N2`7GtSDETN}TZpapYn0cjosJSiEiTDzT@ z{rl&4o@S0*#Hw6|YJRu#Aum8<1cq*BC8qmc^M}M6jdpHB4FB)ntfogD(lO1dm)d)H zRTKbmap2Q!Vp6{XiM^}Mti$bpzVgQXY9Nh0F$MH`XE9jbaO#+tOC8U5-Z<1gw|NWBrfu$g zbiEd;ztk}qh?z43_B}7Sxm2UVlUXIM?riqmpW#P|xYFy3kC^Wr9~v6@xf6Af0 zEJ!ENF*>%jXx&;)6tO~)pGi{JGtR3XkHC3_9*Pr`c>DKknm7W{1W4du^6bgyzygcT z1f!@WK$IItuJoSx=jE@F!b@Lk@);LEzG}4z5>q~(9t&`O+3TU;dFLD3)wX+q@froD z(~r-mAJ-GIkZMVnReAg+f141sMjHV2eH&ndb^t#GZXz?ipks`~;p%^UMBEaOU+I?b z7lfP>VeRb<3of}J^?jOAqduxl>ONy*|lNfdnd1_dNyP0neP0vs^ua6i_%&6 zQNdfmW7`MzFYa#jpOBx&&pr0*6dvrlHWt8NIG5*I_;>v5SFh!Z2_9&p^rI7z``aQ_ zKD22yh-vm6ig+}zUwHeM9Xa#Cnau;Bu?xqYSq*`4d|nyJtj3;SIpR3*MmL1n)D-S? zW8lve_bq->0bY76AjFygTO;%4OWa(Szw16XeOq|o7))a=!@;fL9=+ED-#Iy$FKFmF z90N2x4r$y2PhiwjVlHF*OI7P2>-&+V?AjNV-|p{eB)ym3T@MAUD(U9ccrB;CLvcJJ z_i$%fu^Myc5ijMWEJ<}t<7~2DHUFkKb^VJquWRV|lGVq+q*I#bp1H7Xw^{Nq--oKI zyx|*34Y+V?Hk+KiuU<0XG2TAWwiNQK>kC-YKL(i>N|)oQIUi&n@sNbuTOd;%flT3X zWcK?)<;Fwfk2!D4aI4> z*OUoAgDD8YDXin1i}ZK3PP(seEGx19Z}4`L&eZ+sP+?tc;fNoX_bN6Pj(4^yCr zuLp{lJ=c4DH_z)=!cr1%c*Qq$CTx)3n!=^cjUvk|c8(nEW9263mvA|jFNT_@%KCT$ zFs^koH6qcT=S&I~X|Oy7FQX4RhQbKF`}~XZ8ikj;g!m0|&y5fh2RxbqNMjse%0!dW z^et_^i^b0+Fys#7&WTr! z-PaJ$hx$1=b2{2AUbl)h?)rM(cacZ@sS=%w?;kZ4I!|5w`ffIXesc0v`YT3p8E!W;8(oHW8sFGd`voF9z`t}?2~2UbyBep|n@Cj%AC8KhePB9bv#$0%W6RktP_4S&OY6c3YY&>biy( z#*d!|PakOY;>Z?4#(<}!0SwXCLY)fXzu|)r8?TFvdwtm5%z!rl;e=Mg{LGGyTf@~v zw@!AH?A-{e*sSX@r)9M)J3~Kym}+`(EDa`mo*lfQ*Z8BBeE4y`%hU5h@k%ewx4+Ca za4f@$e@*?jz+I0>tOx}0sihq2z^fFxCrY3yk%$^Do4 ze0~(XRQ3!J^dt4{%U^5naT~|&yZ3jN4MME$b0cg8PDfFVr!EY#>VP*~kanKuH~c(t z&a6|_K{&JhV!nKZAlzPDSH-rVU?Hwc4xtlY*!cZ(_68WA)osunh@rDl@4Sc8pmlxL z4@=$~o#8Rb{2SW7Ia%dMS^x+$jYypv#tzJ8uH>{G`S&hA2odjnR-^sS#(Ag-VAS~t z4~h|B})wS)K*^SHFSCAgFLGLg=GZPDq&+z9nANvWTs$5@MAS1UMt`0aB z3bdD|lU>k*)&gGCI8+4)m%-%i*B0DwjnP8^1xPK*JB@~7W=Wcznen(!fn#L>HHQL+9|8oy?=G|`FW=t6sTd{;`>cxm(7 zZB{amIi8=GS<>aP5aER5pedxLk9D-lT#n4b_o$Z%c!SYy&pkqtumVO_|7o)AMap{M zs0Pq#eJfzuKFa}kRsYe5m~VuT`&~2PN@6nf{N;9Q#^SO^x+Zc~ZD&Vv2E}sKn{>q268?S;J@q7P7|&*Qg2zz2&1M41)w4o&)y?p z?G@<+ED={Kupff%em%au$Nfd7BB*`H=1tgmO;%*7A`|&>^R5qiA5m!EH$~}lC$gMc zATPJkoYJ=U>$6||In^bM@XG`M9OuzV7eENC2z%#GZ6@3m$AEs23{ZxBc&JgK&SDlC8WfIKW73$}2FRT7{MQ6}P0?CN=mTR+kd z?&nl_I-D>lo?oRFcU0+`mh@XJW8To}y8@Mynb0gGH^=A#q``$0!N8hnb*$QaDr`zk z4Uz0cGgwnU%c|tG|M+bx4;qF>Ot*@?R?~fjSlb|#=K!zy__@If6Cv|*KG4G^6A|xM zKX>k~&4oM}_|heak2}D39Ej>wE281ArJiSZTwOIuzG8nK)zNVU_`X1oXa?j{J>+eisX!<4gA4=YFd z`@K|Fr3hFsy;-9zz1XOGSu;cLAlqB))XQ$O%4@31U>EsJz_?IZB>~1D>Sq(@9V)j( z968GW%;fy3t@U-d$iiQWQjB6&8oi_zWTXcLhKQsolo~TeCqBo(@3_A8{PiMK!sVp< zI??qKMGna3bs`5LwwfiB$SusSBWApSqB2| z5yZb%`T0qDEIsCN_}l%*SpC6=)4{f6d;v%GT6L;6I2}z^o%*7f29KnH#Rgq9ZkIALS8tK=Z8S~a(;jNp;i_@?8O+C0}3-cI3 zf_La@f5}U1eR2T=tY%PQ<#AMCG=WjEC6+B+8L~>EO#0L1^9X{%*J?mqwT0QfP+nO5 zl8?Ba)%3M@_-hd6Dm9z5le^S;W61ZP0B7L?r2R)hp8Cvn%V{6cJQS*bjsZFGjlRok zw!`WvB3VC*XzoeS-v~2zM-B2Ek-FKP3}giZe58}Llh3D;JCr4jue}z`JziO>e6{mn z*yFnFp%?=?0KCj!Ze`fn^0l;?QyNA*<|Z!leIK2SRu|-!T19!zvSyCG8zmi$kQs(D zoZMUluLv%17oj|Pv~JYAIrMicJMYc1Y5}|O3dejkkg}dyHRflCh3b6glJ13<$Sonk z)DdED^u3Y25X|_=OQUmm@8;M*VHImo<+b`MnkuV>Q#?u-i^VodpOGUhA#6vSR#N++ zuO-k7c)tm(@h}z_N^8<~hhNn#_az-mv|p^*(fBqF!aOnhzalmJg)h+1@wBn<4s;)8 z&jne`hg{J6R9wk=+!m|fw$q^e?-e%81{c;qUF^%98&?^Tiqw0T&p12`Cci1Y%8oCc zY6;Tm!xQa!0g&n_I_H3m``o}=ea0|=M>0(W_Ki*8=W^Q%v~8OT-m4-P6%amBEk+R* z>7V7R^a9w6%aGD8O?L2;B`>#&cYT+p(L7q>f*p=g$EQ{gza6>K%oVtJxhvf)1nXJ} zw3wwxdKW~zpETMK*=9aF?SZQVBJFD+t{cf1ojV09u>?8i^hW#ZW@>TJPch4Q77Lzd znt6TYyEGae~3G0IMP|5piVHZofH4k}&!5Vq1c! z4X(MDHJ_X^s+oK$)<;U8w zejPCCL2-igJq4#n&^fT$%*(-szrd(qpJH#a1V8=#bBZGPP%k_}j^bgLq-uNeJjhmB zVY4W881nxVq$I+pF^xRF__omehx4ceo{W#>){g{9euLIko62}UlZWQ=@32kR_89kj z_w0Yt3xXbO>GI`92^NErss_5W)4r4MY`f;nTEg=N@u~6LYK_^P)QwjeVhh+aIs}Hj z;t;`wE!^EDLc=uB-aFX8=(SLR6g;nwl)WP!u>}&_F|<5oos1b^?AR|9XzJh#Vzn*c z7Ap*08U4H%cm&%QLly~uMt*LDo<$=ii^)B-DB&q}+u z3wTNHi^v3-M1I5i!AF%JU7u#rz4#D-FwxrKI}#O?Of8nj-(NG-O;tL zIp)0SX=TgDy>+9{{KcnR*04rqI$wohBVXT%+oq7y0C!H*ETw6*ocvyhG$WZG2@!E^ zNH*2T!*0gl!(-ow6dH_f(!JtpBA4c~y!!EJ;zO0l7kK}3c4V8cK_)3f8bcRcD|RV^ zzcJyhYov%pK2UnJ&UJs}{WIH>lRZkVXjdcN5KN>bv+iU}cnGZ++>1~z%AiwsXtbRM z3T)II6u*Kjq=%T)nr7=wSn`3!P~766KzbT*qmhgWaI%jZBF!{G8SD-nZCeuW*54d7 zje2usrJ$kB=-1K&&OVfKvOBwk%ED7)zX|7POwlDRxmnj_k?t~O$Be+1nq9PiC!eey zLbv~=Sj!09t6tMNp|sMsmFd1Q3M#R*lg~}xdFphUznKnTv;I6jkyH9Q;BKa$GTS(w zU>L+pR4ZGvUX8obWXfTU`2U9lgJDc#-*U&OOs%GZbin0cpDQoFA4FS6R45!OECq+m z87qJke?-bu#*p}?RPX`KoaezQ=?iz8Wnd;^i8Ge^j%c{~`xD`+`PL1$ReI}zKlbEp zq6q7tX!L&FhWj1HuhM$i9EJF^F)nG?xeJ!J6AHdn^so1OEr;XO<39cPQK;~5jT?iG zx?-)Iu-5GFx+7{KNJBJAzdZMut?;o<)@aLbtwH|uaSxtElc~A0cN9brun=hzbxN4m@Fd_{7XqwnzEe%qLEnr)2zi1($t*-g&( zyQK_e%`O_3SUqYl@cle^{quP}lfcZpuW9uBH=8+(m2e7rw|Ohw|4Nwu@0u*&ES2fF z9^$9?B#-U5K_l-0&|Yl{O!~ zekj%-<%n6X095^KIoktS=d-|_T-F;G23kiD0tvVlWMW(kslPs<|F(QdTKn!Q33N{xZjIa8z`Kuzd6^?T}qF z#-m8b;$C-Eh7LKOoY0!e%9c;p{o)%xj`@E6w|Pe!9t_~8`o6{8GR6dINoR`2Ui0lfWZi z$%2d3dGLo_fof3le^iktYNX1|k;a1pP@JNEL!ZmR?7raTmpvSKsLpN{r!!cbbmkI<@yn9zxBMz6E%a+yQbWr2RkjK;! zszn50>2M2Nj4E=mthTuX{ z3-ZWcMwnCS49*8+LLWfUI6HIJ6;MxlZR|StrNEW8_j675D;(}jK`B;S`%c-dn2hTOWTJ)R0I7KHBdQoI4fyB zmD7Yd_vauJpVy$bK32WD+zfy-`(*EHj{=mf+r+K zvmawM91?Ky?k!w7#=O}XIw&K#*jLy&gPT;Vl)}N4|DVdnN?DE-|05Fn!5!cRW?f1m zOvF5eYi8(;v(i_8@ud)n!)sXwWgTfm$@AEhdGL7>ePGwl8tFa5l<}PBbJvKuR&Nmg zL$@Z+!kzsA*^vHA6*t~F+(Mw?DtAu#KPcPUaUx~06lsq_EaN07z`Q+$SfzI$id86n zB31|QJE<8wQTIF=EB9dI_21jK85))@Czu+WQ4}1y@#wv@^V;}FQQMXj#l8z|G#v(` z^J2k~*P!e}=;0*_@**D7Pm<3|EIu{mF*>7$=Tj?MO?Q%J{OauFl>hQ}nM<%`jod>O zzmFWsiEs&~)-hcl*^xUoVz=dvp@d7(`?DIyk2y+_R_HZd9ZV>B=>&w*e#W3>eK%jC4FA7R&($M0MEjy0q%cU3j^ zaCwm|%(zPYy1w+Sf%g6MAi0kP0$@Yw`n!I!S_1XIcMUB;2L%MDPpqkk3pKV1x3`_8 z6|nAx{W79K_`XYvtXCh`<-B!UNzNNcl5AUTlpM2L<5!`1vmASj#2KSN}J!z zqGR?ca{)8*y~5Mv0eOD{l+`vW{;rPuK7ShT`N#2TG2F9nFr0s%lYHYWX(uTvTVv zF3h3NU2%PLxq~xKf<{g?-MYgael?Bm%7=#`-s`_FpClsrt!n)do7Vo@<|Q%^u}^Sm zm6Ban`eayjG71FlYj0H7M0s-C%6CD%-_zhoI>PZP700}vB{j5SD^jf zOM8wMy}xxU=gVWizgtr6^0Q1?fA4-u-^(g@7XssjP7#q^F)HHSUdtE+D6KK;IzaTq z$JwLA@^0s3+0O4l{B({^>q?N} z#OM=~!YdQ9UwyupkGI8|i|v1s5LI)O+v5Ylar1Xu=|p?TtV=bm*4=Qn(E8<$4C+`t zgFyUYBxc}Q0@$BwCq9LXC7nynVWiBn^eWeL{iiqk&*{kFv4T?NHa{g4>Y&dIm-NY6 zhwy)>4yIhNS5Ii?s_gUy#6ePTA)#aa8J)LcfO0*Rp2VCK?n0=)Bcen}d4}gXaeU}e zIlqyo(s)9%f%z--{NsKc_oWR*dJP)4TTld56xpK-&lO2J0s7Y(M01| z(}x!eyoDfMf8d!R>thQI4nW~)5ppSLI~rMHozDDosD_i#TcCnygNTcR0s@g%RHbq8 z?mp%!Y_R%G8NwRbN%nV~CI`(D1xfRoU)6o#Y!3nljEF-8aW#xx_Y7Jm&hnBdp=4xb z-M0_wJm)^=N6+b1XB#{2_dPzn-FnfO)gf)0ukd<(cKw_gWc<^eS8g6+oI;(Y4< zUGV=uWqY_kWS3;aghM+wP(PAgvS2Zy$Fpx1^C2nY~!pEG)H=~3Ei0>Z!^iV3)nk|Fo_nlUQ-&uwtm zpOg#+KAK(XN*9FoS~8&NUYVpPHMZZGJc{!S6FapW$;1*#L=j_1Jbuk;I6s3ft#YsO!)#6EJ%C@{(Wy z9SQvr0IG2VvzI&5jsneCmpTD}-(Zs%-hX14KiXOw@Vs{BLXz!66wr*o^=UNrJ7C`$ zHUK(bT2OpVfQQ_q>G@bI zcdpXz3jMRSU?&VeL9m1AJ0tnKQA24T8yDHjkTh-r-}LdD&#>g>;z3SJPbZvR>ZFEN z<8{z%0=fl2U{OafU}*g#ey`6=Ti^Yk4;Di5k;qJ=opr>jBHNUVU3j-XB|#AsMNzFw zzgQl-Rc)cN>IZ@42+68eJkQxVow9N{u%{wu>2nCcH1YT*nf{IQp~v-xwqqDf`zjMDe8D&=oFR@85FcFI~feNx?44K0#=DkK>>VQx7(1RS7 z{@k_Cxy+ABMrTH$aiawi$K}^>8v8{6LwaO~C{|xa{U3(WeRK++?8&P|!rKwP3i)P3 zUS0PPBwoIpYafvvG)Pm4N#g%r>$`ch`&q$FAg?I7B?_CIY^;q4kOGCyQIOBgeJe9W z4ApM?Tv}gSwCOkEJ(FVD@3;Itb{-UbUAs}_zlZHCs_`f8VYAH~f!2Uv;^NED>AYIdG@yiD~%i0)>G-e_X9p!{#BM!jXvYf9*j+a4ffq?hP ze=2|*aL&k?oSuJ|e76mnD98KL=zI^LJ?me-V{^I5m`-lY=0KL84itItG&%N5yZEX0D@a?WK1 zVsCx{DfNScK452rj0t+{Hk%}fJ6wJNq3*y~_M4tmd&=gW^a%SDNXxYPf#o|=b1WzP zQO<{TuDg#i%!4AMH^{eCxZd#vodD#N=OTfU5{J|Out3Sf09_6vrH7yZ*rqLkypWMh z?f5^Jo)a(8n8#@6$*&B%(MtkylTr2HroGSSKmp+X{;t85!DMs-f}Q~tsAs*NrUOuL zw6GBb47dBDHmb7#9nJf%`gvHlVXAZr5sy z=f z0=AmaCs(pDWGec%Q|2c(&i=B>z={Qu3aAzCX|hw}oR zfLvwQG6KU4%N9#B>an{W2$%rdmbjKc5(C(0bjZeurqmhoEOA~E+FU}uD0?+}Gk)Fs7mkkYP5Ro7( zOdQPaK3R5YH^A)d)WvXl&>bJAM94S@({1s}WZJRqC6#xeU7K?L&^wpt^PkAkOD6A- z0%dn_aPSr8Epp|-yWs+SeCmj~5g5l1QdTvTH&34?b)?6IXFH8X{mWn+T@R!M=w-f#2Q zm-M+t$Nuya_DM_qRen`Now0QAfP{mnW?mEr(qfa!1n-4bpsCayVXbu^L0~F?V$Qz# z^`CvYbKEu}5Yvr!_tPklwZ;n*uxeZ1YTcB3Ps-29DC^J1nPevR6sd9HWSE zr{QPo2X1Z8G+i0L;j07?M*?%UOBfN3{XuB@@)9EZb?E4X+8rK{Gr<&fu#{KIz<5A| z(mIvwb%pCxG~9V=VB5isbod~Es>8KhizAT*=Mcn!1Y9_d0zaEB__~4-!>5>5b~4O8 zk$LE=nl}*@w6qiV1Cb{9@DKtMQ+k69gz*Gvy6*JHN(F&Yut?E_hHcKz25;3UCJv0g z?HqEHxR@H9Ef zEN0t%5u*WOJI2Waxcg1F4zn@&H~pVndgFn=aW*7}RYO;IgnFY%@(#R_Niqo zt5tR7`uu4bL$LbQ$dIj=@46?hRyxU8%x2{HotUG-68P2^gu`IrDEi!rftG!3{!oI= zT=NseF5=lOa^(X|$$j;}*L+4)W{FX>{d>cl6xB{hsvZ?*p)`&$h z1?2~YNaD{{4}V1v+KGFdDhFgP&%7%PQKM$~33T5_o1?zN=7<|c9EjfN?}!M8AB6*v z1eb3s1c38K(}NXd?u)=ZFq|FT_gfV*q!$wyB`3l9OgKzeZtM$}5m^pfx6B2`LZMC5 z!5=IFP7BU2X`JR3c%=g^8}*++r&_Q0iY~o-_YF|lFHhh2Sn}IpxVNV({cX}VcMH%9 z$YAaMEKmP2I~Y?9wDZ|jD&JGRER~Ln=ZNa#LEzE9^6jcc*!oUH-2FK$49&77U1P4r z=A=`fR5yj?Zs{ff+= zECIvWZ%+5`iP_!$T!1D%!j8Yl8J7ew%7{7RAh}lf2m6{&BC;;y;Yk4j6~BJ;8@83X z&92nbzP>r6dBf3Sg1j>zg1hp2MWe=Sgo9vF`7H-GFYqn0`~Ky?k0%q~*;kt`o-ghp zrYI&o*$DMXwp~hT#l5b(+*`ej{VR`_z80C)Z&&-_C?c))lFjJ4TVMKK+t-;QzHBqM zQZ^ULL=uhmg3Zoy`pQ>(lIj0ld1;uJ&fXrXzjKWf3{bMnC2d!u9sSxVuOEWWvsJJt%))mDoSA#P?B}X}{jLvWa zQP3&k+6z|Rgo^=RC@7mmaAYM6jW^kKsJXH!Mh8w)Fg{HjSiTf-s!tE90B*RSWlVGB z6pPzwnb1*>te?s_1$@~*8O95fKm)tQRwT8CV}o3zMem%mArL^Xsq}1qSOx|2Yp2j- z3SJ}>44(=?_&Uwth)~?;PeED(_Ho(kZ<+|!S3x?)26myaE6djHyx{9EhV83mzsOnY zD`)eWAn!~$(gQwsBw|pktr3Tc@8vGr6-_caSEU3m>8PE$ov~rv$=4iug)GH>9|Nw! zC9OZvA%8&DDoWV^^~P0Z+VDCIZ5wX1C-6$8@gRpMcdaS27C7Ng>WNbP<_kNAx{q`N zmaFGG#;%<2P8^fOF$JXNI9r!5J30bj^y8=LB}wbgf{ z%JGY%BPU6^7?{d6!2r5K*RI*TE$LdS$fZWd*0agxU3}b!=@oM8%32&*U>rw{wn>TD2J|XUOOy{@;n!?(y9=q@Mxhi#4=n_+4+U+P+& z**GhbAXTrED_^G{)S4s~M#rCLO(#Pxo>;39)^t*Wr#z=~(f8|ebI^%?!AKfR&fgSb zS^X)`L6#sMRhL`v^U78=A;1ra;1mXJR4u3)^~5=& zffPo_4=|)we5hcgWK;+Yp8FrON>o{m%kc#t8`Se%CW;te=5&Ql?2pAZ;0x4G7RU0ehi@wU^*}&XaS4LPp9hmi9aX}ze*Wt#Xq^b*LEE~mNn=@|#qqaXe> zAF-A`1fD#XC}=|7;iVX1YJL58O77@i-BgUTXGfdq06<)2Fl$d!&Mmsln#`EAq zv`enMh`$7JQCw5MOG-7r70fBQ7+o;-4&2qA4==$Wo~x3Iu~w91u_U^rff0%I=KVp* z6qDhSDm@oV9G=_MxEaKGQniGXz)I%4RmZYnGss%2;mwf|!Q2D{dRAUF^QkMZz6PFJ zp5)3@_{r)rVnubR+KO4(IE~io2EfoDD_sZ^IlL$avWJO?+hqKq&MZ19Fo}JJsfPKj z)tPxF*0C+-I$7xpBEoa`)AnH^z$E-p&)sB_v3e%L*V=*iPno$oPh9(?KC7ti0Rw=> zAq$z&wMn0I|5bFkv0+<@6s%?mx9$>bqdrCbhvuMCJ%1)F#hTJu41MTlg@oJ8`CkSI z0b|Rz_SQ5kuj32Xe-#(LO8X=o7X9$RUq;5v`-%%|@2GmJ%gE5)*7t5v9bLYHF(7ZL zw8Ag|MXR2kVREFva_Q2rv@}`1e6;NZ(pMG?CdDzTF&(0_rm#K}LCcNSXpWU<8)O3Z z=?71v;3oYFzS)o2F-TrzafLn9lTS(gs5BL29x)j)Mnj|$;1r6uSC6QR-&!<_GC`_mv(_Zp*zEw47+ zoV#kbZSN_MMh^!B4jHfAXk0yBPvbqcvGKF(nAq4leN{H$$a%;r6@2Nuzrg#@_v?}$ z{+4eQv?WgdzAq6IN<$fzy;HXB{;ZxQYfqOk`K#9ScNV%dKi@4n%kSL%TME^)piYbY z4V<4;WwyvI%vsRz%@iLM7*OK#X99}ZJHD{&LdQcMM)~**BzHu2Ujq1ME~j^;fGJdY zN+M?Q>y+qQ;GN!)US3|IYBoe*VOdP?5VTH=7AzwYh zLq5vcpb75nF{2%urD*gpO$2TcKeFDQvN#+R4#TB9J!W5)E}v$Y8VLxLHf@eqS)fyj zf;b;QAY+yF((>Fa)_-MM#k+8#RX8ufGnHMW{Ext%ho>Xf@}d3kg7x_NqdmZimu&4N zew5tEA$}L384dKO(Z#&@5FAo3f!U(r*w2CCI>dptWKulq3!~I#vQ}aG&lQxXObtLk zz-wQAP^h+&_J-}Ld8=&%?()+^dYp!bpBtWx#GS!By{*IGE#e!N`{(C#iTkywW4F7~ zJ$t0?w+E@|jK|0@{yLwOIUXmc^+~Bk%0Ag{=VjqA7f@stk$0!V*--xvZQmVFbsPR4 zQKIZoWE8SO$;hT`*nM+%%-ecSIv142=pFX+f`3IU zm5!xT3|z)NXpk4s@RJ}4*jV#->$YET^zbO!QGVm#5^@&4lC5d!*><&?w~RhDkI11! zf#l%AS*iBZ2j_lb-uNsg;cap$4*nVx0qa36-$C50*q7vBE-jt`MCsN5yxSZ5B1a|1`>z*&0(HKp-qCB-ml7+wi!g8<|J#T(K$tgl`bg5u~! zvSW!!+Gap^HopqG<)t-E3*TqpRlxL=rn8!7;94D5!j^l4e+R~>YiLjn8lhuuFc_H(hV@EDAHC>WbY?n0{`1Qgr4_*CZJpgli zR&M}dvfN>96>cABJ<~qA`eR`NX;Gvo&W{w%^^N9QD7*X7F5ICuFKW29jcZ~0#Yd}2 z0xC=NmT-T$?3IdS7#q63{4v9F!07%$wTivIOK>Q1VZ&5&xLqLAEVl>k<2yy`CFNGxW^* zvE|-T7)q(1-d-H0FdO0-(EF9T;um9h5d#ZfX5!5^N(mn(H1A@w##_uQ148${k>?DC zpbrjGq8DH1X$~fcv@Q1A;|J(bi9U-3oqy`PEz})qLlwkp9`1(CToAv2tyEaNSbTRp z*KW3WVepp^Qx=!gAX8{jM;Tsi;NvSj3*j&1Z(8;2l3A1tYl>4V8T5P$FpZ!t9iqWf za(6j0Kx*i_yk@ch$q&tlHJj05rup3>H@;C$=d~Hlmc`&;(9*!Vf)$6-L(rni|0=RI zt~9X%@A7nW4aVs-v2k`LWHGc$-<*R!ixptAy01J35sf9r9w3BP0M`5%>FH4c9Abs_ zW+wS4;JA!}5^=}L4@@Xf!5MSAEubtGaEZO-#Q4T%`t!MPmaQKTY=>rxNbH>V+O32x z#_dt3c!2qC^$^*eA{hp?$a5XI#F#jwu!-n0Dtg~H(l38c#l!YQi0^k%vam-d4|mim zhcRXqm%EL{mX5eyND}1sk<55;XHue;1p8*#=iK3?;=O`fI6=%gqwry?%_NO)^VUe~ zJAIEE413>pS29MAI{rvh;N`#iLKPv$D0sh=m+5~0##6Zt_sYh&Dpo<~5S^4l#3iZp z9k^SE2*vZ&2A2|^*EIoofRQeq-a|#$Dk^2R%^LCg;3xMpfIjjLEk~&3c;=|`k)Sg?7Y09q1WPe9-+)r zj2suP7h3}R8>0O5BCfvug1e5Msd!rHA8`pQDYGNG6O1#_t4?@D3w{l0<%eFM;%G0$ zapy$Ow9hqc(FNt-xymvbiisEXRGwbc-JIl+LrU*gv+_gq-omC<0Hr{GY_tGV4_vWj zB|4zcb&e%f<@TY?VV!<7?Wd=!ELA#2RI9V}hhNR$t$E8kTy`-|X+~9W8GRt8sB2#P z%|>&5DY=KgoJQA-$Cr1iJKu+9wJ#V_(2jRc4^cl-aNNAtWA)>}znIrP%|Eglr8qa{ zjNIqpHf$`gq_z-E29G6fo!yz~wQJQBlK_(lvHXxy?FN6LCKH9L+T z!wzVr75yQ$X3$2j`DiTyNlGcikIpU{kuLx=4%H*riVR$~6Jok>W;cW7^>)DyY%aTG zyRD~f5%$uSZsVT&IaIrL^BHNxLLdA;>EX}?8DfIofm%L@_cL@sJDA$hPQx}dF$Gc~@xM(=JnHMQMCd=_bAwYIg0%C~T=H{8hdv0~Q`kHIQD~Oe9~7eWEm@`PBcVfSGMzlia`i zrkVje4EYUp0V8CkG4x999?M6sYc6(mKd&r?8)oK6qKmQVmF#?&a=apo*)(gvz9uQ> zy-XeHUpf7oPPwRq+gD3F*$8d_s2rYQc8_9~S2iQ0Czr9-XN}>{H9@vg8>+P)N&s#N zpw-S3$@on)0pjO{X12VShdnI^Kp_rrqaM&p;d9Rspnz3VKJ2#H9JfWM*AHUyX2`=u zo5jSB9EBhP!e^HOYHEssy`c-@;DTvp%Ak>pVT7KY?47jDjxDHeKGp_cAcglAiuyr3 zys-)wvkJ6++4@eRB$0}%Rt>z9VcMgWi%z+~oe4fbsd)TM-s|q}9y!@G1UROxt-}JPnwWF+EBXV3xJ>48{Fm;}#@GWE9nZ%pazjV1adyot zye93}2NVxG!%Tm2yYJZY>B!D*EC8H>q}x=m^i7Yxy0$e-;I;1XLiYi*d~~u{D4q^& zDhZ3wosEg(mx%d{@WZdexu~EjsSo&XMR?c;5S16TYr}Np6wp)dEHXVN?68V&*aEH> zK2{)~ezcU$zu*^NERR9xf2)bp?D;z>`!{M^jb94)_j~7}e2v2Pu-EY)qr=yo{Ug=Zuj3x1g4_9^j_pIQNzAJ7^R8 zoGHfk0g+sDQ@PE!h*Lu0W>fSAm!4Ti+=AE09$ExvBJhxQP$x*U$C8B*(0+=R*0cg4 zm^FyouLMpsaxNys=N2VK;;Xkl_b64861-DWR4GBSei)yUn|T1PFOraeq>BNhlhzx4 zKBRz{+&;;EMBytfmzoGN_oENL4!0J7J1C`E$WE?8`Vho4Ne^#B<;#70qBz(VGqh7I65~#sI5%Q&z;P@`#r?`jRIXNr$jXMUij+% zF75P_yZh|}Q=I8L8rO7|pi~pL*1HW_poHlr&s*sg=AD;d6)G%rvbDu2cJ*y*u~Uf@ zkIE#ije)&$Vlxrz)a20GJ!_{BI&$jO>+xBZ>+mXW%oP=$#p4ptWTLNse34!*43}yy zI471y7>zVB1LjzH|xa}T; zl0Z^^=fsF1h&^Cxm*r|Vh@v#E#vg4kRcxUw=o<=_eoya#x@9Zr>o4Bu{?@0zJ?3@n z0V?ngVcU~{XLUNb=0oaW_eYn*8&F8Hg09mnEq2D|K7sH?P*yiGAGut;pLoHEJLNMb zwH&2*(u6F00~86A*L(omtO7bHP`d#qa52tp>BaW`#0R8ij{`)xZH;0H&yegU>+FhkdqE4kV*!W-~9^i*K(Z6&b!WC+IunWgmkds^Tdb#huI%p0L z=vLqxo$41-5o=IdCeMXoz`<7b#&)l(cFNG3lU~*ZtXk`WhZ{ZLO7a8$m?5&aYCJ+N<}HHD1Bt6SB8;PL1mO%HA0O|J&@(x z{TN01M&(?6;7aGAz*anL0gWYBiNXdgW#2F9tsOcjjZKe% z1XTEpt`w%Ue=Z@?n+dOed=RFp(8F~U+MY|)bGqIyU_xCH4LceuUTmhBj$u;^d;JL} z_32S`d=-sMg&{j>iYr3RN;3aK^AAG_KUFkcx@v!hqJLOpdX!5o=5n+XsZifze?#+o zqBo7D^6wn{ZW{99>OJSK*@#n;=_B?|sl)eZcB>pxe=K-Uq3#$|V3czM% zo-ImvWhJIvH@orKMV8fGkn3-YCxJZR-ejUR57#YZut;ql^aoW8EyD91_tEF_ZXXcxJf)uu;!6g&1<50qGc z73;P_&ROz_imH{MJig5i+S9Vb$Moli3u5PWbmYi%5iin=uorhnKji4sc%JM59pOjl z2H{NoVMpL{D(uXeVs#glpe`j8_q$Brfbhi_T&CP$Z@-H~a?w_0eIFtA5;mqnDaj(w z*&hy6^Dj)kf?V$c5FkwNaje%VkrHDz6DEq^;2OUeCui~fBaFcV)Ui7PRkELHZ$(yG zHvL(5gkM2?Mmh#qVbsC(i}s_00*w{q3cpozN43_c%UL?S32s^qg%l}U-j@eVbR%C= zt$Wtxlse)P>!R7TBp;kyy63#7C`_+mT?$yz8NcO$9K$AvyX3>j7dzDDaQll<3r*$5 zo&CdSYL+H<<|S&e{fj%9VjQmwMzA?|GibF)B;9vk6Avj?=w8QCDjX13slGTKX=6`e zqC_;aJbWEX+}h=ao4RA@1vB2g;%5gbdc%U_UluWi$`p_65)$5XK1|EK+BeMHvMmg*z7aX-#MbM(S!qe-T(2rcsafio}Eo&p$u>p}%#c$u`n3 zp0t?41jm<>dDrnr$BF&K#8SfLyLk~ho;BWh3YgRWzSd%C=a$#w-X z&t<9&OST6bJuO=tIkcZDeVD)0B<-|WAnp;Ri!f}?G&NHow8dfaN8zOLA|G}1@pf?Q z=TOdIDZy#v#GAHn=*0QtPPfsr=SjgwoU}EwU`#wVj!ufTdK&x^QA2g~4mYYiaKvv(G({FBHo0{u+UH70)^6 zCq)cJ(Q{h9+U$YuQ{$f772`u)(zN=?=46$f%;Z%Y0KDMX>vj?U`fqY!%^-35`x}K` zpeyT=FA|gAY1AP?_uwE*O|bx!1)xA%CCZIP#fdX60G3&BaRZw(bqHV^^n)Hyz-21> z`BYS^bYRb+lFs@m#QRE^g~UL-^W=xdmrr%vtAXvDiV%l8mto4_X1edqr8Up{+|(hE zGTI~JXVv`}Z2@mBXrMv&PjxzM*5@G;>*o$poaXz0?r=%U=U_F^>*X}))G~_A{rngZ z0F>Dm90e_d`t_Ljr7j&#z;wsUE|VuY!P70_0?+lsyh5S>G8oj6w&YkO@uG(SgrO(= z1?Wp6of#f~y)Od5p325ADg@ukJBfM39c&?HKK5IysM4hXM)F%!*U(N9P5WjrwNemw zm#=8o_0Fp7DJ}bWaB2#l^R{Ep7ryFU=>83VxWFPWIx*9Xkh$yT>NKNSdCoMIV>#z6 zbAlQleLZDg=Hm$*;=`6_8vTmHgpUMMDGVHZcwm$Rx#ZgNDV2kr1Sg(JH`&Uv-dgoX zmQfuRYDiHhBwm6J$6o&t=wV=LQ?>b77#IsnDZ42V2J=2wA)a9m(Dai-M!6T<*#vu_ zg)T2PF+9KAU4>|?Mm>|wf*9UI|rabeWYa@G({Q2Hsr>vXHrNZ z!;d>S+D1BgV^oF3-+lIWpLw}#HV*B>gu7$vy=mLcjPIq0L5{7&UHBa2+1i!G&jB;Y zSy@;8@ffLf-neFdQo3wUNEIkD+yYeFpxtHFOzTu?@?Ln@EI`o|LGg|0E`IE0`0qW3 z0eutmywpnEi%J&=XcjP4qwre;6I5(jY(Ug$Y(J5j?frTqePC%##KA^IdM~J|msydh zrNvgK0o^xbCM0Ta3R_?@b*b}1J^(34K5%#_1~nD2b3JO1D}qhlVo!0@38dm*UY1G< z8&DV2(v^%)WVo_mMwy+INO0-o?(?3+-PtCuXMeYUdXV1V3RjO``jr5aYfLN(W`0S&)r{6lvyXgS@!6ohR{xDNtL@*jCAF&r$QlJ$) z;0JJgB*`q3XXs_dcqqD^o)lnL53_4xwEk{l)+br^M@<34qlrEtl7m~a)HvB0poc=! ze!@wiC7{c|BKDk;)#}M}`f9)GdN=2ltaGi6sQJy2-19>Nc*j2A3gjYF?-HP64DQ_u zG51nIC5b8Q3`th-S$-ASmmEl%?km$e1OXP9uuF#pTvlAxG|HQ}fcoY; zcRAFRC;_Y1f{cE?rGi2%f^hw~2p;}B8TZuS6l>FvN98TF2i~Aj1 z>fB5O{0GVDhMn%3Gh|U!qEkdpPKqC+*h=zKVH?+vCswmqE2fgj?(?to7|%F9efnns zaKiG>(JRSa4Vyf<@$D`V3N_W&{t2DpTz_8!wATU%A~}{A0A0rmV2#oDUnDBT2@NTX z4gx|+Dj{dvUY=@r?f`K8O+aD>s#VZ}((iyTtRF-H&YR(GpPCoY=)>$WMAzO-$ntx2 zzJ!7e%+G-5V{~vXo#u27Yjg>QO@N_ku^e<@>^LM3@dD18{;Mf)xEsmDh5&2PsPixX zt=fyU5?4tjAR4t%$ZW7*_~#XL6WBO$&XXo(&D6}Z1{}XixpPua zONGsfK%l!}V=@~6>m7R(_!zB3P4Ayt7MDI<=jWRD-?RUkFwq25ra5j5^BgT4L_ri8 zoC^UQ)X6~9bLZzpwR4hlA@Up_Mv{I3D9tcPK`#O=)_3EQHy7F9_PUqV_lTB7(e-;) z?ICTS5iTYAF2)xtrglCXp>mxcA9Vzh4@N(W&>2N#^rtu|phpYhV%=UAO(0jDi$RL` zWNRA9fsVn6TcJ7$f+|O0lJPoofv9lFog9M|(uw&^1}o>+q@wJTYjTK@xky#ASR4(Dn~r)p z@^y1DvAUjMjXP7LEj%I;(NY~#-R>b0Tl|CO(1R~4PKQ6ZTuJw2KCG`mt_61`XznIh z=%91ZY#>Tv6u@+v0d;TwwKmkC%5@0bB7hfJqX@W%j5W`aF0GO>H!sTx4NK}ndcq9P z9k7$8)<~;_N~L}e0Tv_xyDdA(!|nHtCb2M6nDfnlDOUXc+EnZ?Ort43Eq8I~2_VSo zn6OuGC@(WVYCz5oC-X-x>E#7DEOeO69ND!GMp`UJ22zxvIrZ${7$t{~!M>F&qfGA2 z*t~S;%T(dPiJLN?$DocG$@ywETGo)--chKHSNJIDg?fX_wo;+y@yp~sd8HA3sj~?#XiH$j+S3aFnEVlaCr)r{>mv(rZmMZisEDVNk0uL10Un>eTv$t) zA{6lkeEyVb25tIz6t=Fn>SaM$jDuR`@A{bQ03B|*TcxKq?6^E;<$#{?EF;SCL?{^n zq31zO-;C`k$=uw}Fv%M4GDTyPW{7b?CwE$1tmnyI@^M-yp$xaT^jIwIrI_T4;_oA@ ze;cu*QzWfsMy7SGnwoQ#qKYHpM^NSijhC3234i4+qbfSMvUj{LenJJQ3Jm2x&OG!w z9FE%Li!QnOd;AyhGzgQmkuM^I?->V=y5iH<%YBG@dUBoO5*ng%F>O!tQOcIyZ(ysd zp-LhJV4P)vsTdGcMgL`&q~qX|cV#Rh1_0%a<704O5ydhMQ8En!QlZYajHP?AAU2tM z+h0JXn_I8i-3qxk&lca;M^c^YQB_I6l;b_QuIJFF!E4sdAUKyCx&V|}2_8pj4fXy3 zC=IZ+uP!m``4J$rCIjGnvHC94;i+PQN>LOdHi9n~f&O)v)i zWa-X>i4H`7gh$1=uh;SZ_)+2TymrQ;L(fgns&MmR00=f2GQdi zn$+X~k}PZW9zd2vZ&dgiy8D9YA;$Vf@tqC~uaH-M5r(?tm{6XEqGzByGYh*%wTgcs z-TgdM7m9BUP1dCt@zI_-I%RkzI*`l%yNc^^7Fp_C+aK5FyL=7Oi_V#t^Tj-*XIN7q z|9Io6BfSP>2=io9-fjZBaXo{6yuAQ=~v) ze84TsRRRkeCqqbMP;Y9~-otL+fS67I&0=eK!>Rc{g1J}pgUKV~#({}<`z*!x?d^QY z1a9oO-RCcockK6z^Y^Z01pm@<9E%D%-2R}ocI#Q3u|tog*zy8W$pSY|Q>5c)-H zQWti;8^^ah(F`d0tN^AdH`et=Uis#l0p}7A=0ja}fjqPC)dy?xeBoi3L}{^(Q3{~% z~pWA;G~Qh zms0?j;5g(uYys!|3B_f*i^v1zs`-zYHz4mQI_lB3Iaa&q(JBFjmdfL*71as3!@g>% zVP@gYUNw%LaHqRZoC<4#N^`wCTJ!{MO?CH0WE(ztDy+cs9l}&=6yHjthhw3KYkX71 z=ZkOEZlB$L$H60A!bd56sj^oDt29Z_owt)Z1xl7iOl%(MnHNP^6-#CnYkx!aWj|ZL zOf`~L#&U;Pz8s)6_E*s19OyTZ=!K{rRJ|S5S!@i-w~$uQ;hy@s3Fo}u5X4PcRY*}W zD<>pVOzJf1b50^=-WkkWNPH0SO*X?sj$@RZ^c!5`zEyAFIhUXd@TdC}wDKx#l$o@$m(S~@q5^mp_$<(2QAOXCKP~y3##nMjnfo=|QWodL z1SOTwxU5tTwF&v2mu9^oaaOX~Woo^}LA2o9FtcEIxO`1W77Yy)Tq^Y`Bz0AQ8F$7{ z%N=if9nXp!-BNg+&Uq_{kQomHTQp3EM}f2Z>2HKZsum)wxmW<_Rps7p_ZJ_g@zxeN zRebCdx2>7%!y@xh{OE&k7pU%9M+)yXe9O^A>y%wB#zy<=9k#xVuDZi!XW*c?=UH{- zTBm>=2dW)D5l7{-SG?RBX)eEMhOtxfn&@7M-r$hpa-?u31M2Wd4}GC8$9hp63d^B= zc=XRSy6&rBrd9GZ$ZJY?dWk`QMFQ=*lN`ai$GjV{hsz-07M#>YyDIaOMW)XCJ;I_0 zS(Q$7XKzYD{=29mucxf zQ)N=ZhtGf2Pf`HVcd7o5+vSffzr`R6sK4SIQjgVZ6!_O)*Ybam^2_p&swdI4K7|Uj2xrvV54l+w&>C zXkNaaeycH~E9Nt>w^FZ0RyJPt^lHlKNGF@!nSVZwT3m|3y#DYt`q|RlGZqcIqujL) z0*WcznCt!30odb71;9eI+)3M2`ic<7Dc4({g3BaGq08Wwdf)nOP}4qi<_*GmZN*6f zy*%v1Fs)45DX?H48_)(we7PsvFHYh(=e#Fzc7mCHA_K|OgqBXy`C@cYV%^wETwIAt z4<`ohs4($(M9{`g3R2_a{*Lh~Ai>w!sjMGLO_e&9HU=yP`USYTrg9mTEJCqi{jJK~ ze6|SuCZG$K0}wdnN>B=_f|5Eh1G$z8+L6o%hcCYvGzU>Dj;kRsW0AR&>&f7Q9 z6_^~$ih4_1o4vZ{` zIGUnPgFR+9c(Kntrz=ZqE&97Qlmi*Y>z|+Jv8R_uV>SE%4#?>Ox6=$y4N~l{USZi) zSAU+O0gxR#bX+OWNy`OUVvjbF6*7h;3UP<4u=9#`~#e83RWrS-q=Kpo5SBK zxi;lV4wD6SQc^|*JW`9=$;b@r2>jhC&**QvI=y410^<8d5&%+?oT|B{A3eN$(U0hZd z0xO=rLGR?mgIwm)G_wH+DN~v=-vhTenT>ZHc6U$DXq3)T2l41vm9Rbgu8W0uoW9oq zlFQKJj?VWq*J~Pw)~&;EV8DmE@;L5}N0!!kUQ;LDJmU00Ja8|6W@`fSBrircNU0nR zws0({0|`=pt(om5X3Nw}mwuy?!elDX?Yil8KPRW7eEcy02)z8IO-&CVFFlmEMB0uR z=_Ui$hx@OjvaHM2k^l%96m(c61lSDcUPbA|88#z+xQ5 zb7Dv3Iy{U6NqU)8&*$|yC1d7MQ4+O;?5v(3>LM+r&%6$WLWOnG4KS$Vxrrb4@R{9p z%K4Zm_ASqIldUAb(g4S-V9mC>#gqNP@@zCHVeI)9p)W|r`~^U{bafxre*K0o*ow{l z@@VoSXL~p@|8)yTMQd)=mJXj)GxGS*W;V6nZT%KPK8c09{Gm9NX!pb2mq;b9u=sw0 zYuu(mx*sJi`~Y*rkUwtsb8y*jZ7y=&Nx^DhxeJts-?T{kT^4j2dJv;? z!J5bN#)~g3p3cUb?5HoK?k-93-*pV9;#u3>=Kt6+Tjlid&UCnHpP^jtL*4#Xhuy*U zAup!xCJl6nbZ$=L2+@#ewZ&m$^*DdBBp107;}*MTvOPuWd2};@h)62DlT<`DBW$ zfsrOR~ zf=#V7@?`pB_r%0e@5~qJZ9f*N{46r8+gJ zv`EGu+|fSY`CwJSe71BtG`8pD6yk)ve?{vOvOT`csCwV`H1@di@g%=Al%@9h9x=NQ zpko%;R@E1Of6#wZrG%#qYE3i}rWQr>9Vx-7K#Z*ie^u5`rLw=KE5p~%%iNgAl1DnC z>~b-R6i0bNCdrL7L0Qm^9Dtw6r3*&Jv!AV3^{XStk^Q}b z+^^I=U37V)L5I^HeRPf*(tJ1zvGBeR;HMT)+G}o6h>jh0u2pjmk+xA)w>!aIcb}<aajxBon&59-0aC)*j}Kgnfl06yHVQWOEBeV`SM)b-P05y z2)P|#`5?VdKsr9S@N@QXHEiWM3RQ?LIK5{O~fxc z-!g5DB|jyRe{V0E?B_SV{ddI~ivfLc$m(EW)?1oyZ(gsw`$C(+DKNkRR8pQwv@#0Q zlQ`VRxR!a5SNY_I7u%(1+%?zTU1k%8g3f2^^OiT;1$a@ONv@O2F^}ViSf@wp=P6TG zhYf!^i|T8_{qh8IVh^8;g)O-LxLw@J?lw}ldK`)D`k|mtT{cZ{9wo7b$ae?n^lQdx=x?@iQ@QKO4_7(GD`*E9Ei1Oh(cHFsXgd2=e9@{7&SNdnGRTsZGoHQKU!wGLXvb~D&m<@M8Gy>d6muFL6! zt(}ZN5u`jmq}K7D&E}jpE_-KzcDaMKTFcCB)bmK~iiI^VMMPTO?Hl#%@ zX$Z{Z?{Tb5;_v(c2Vcj+iTjQb$$B#OY}4bt^Oyf_1QHF)?83usdl}^CuStjNBYabY6V zs#Vz(cWhY@d%SI9*_4^HR8Ag?+=FkGt=Rz^=8sfE{|dY4{o2RW14CZ!^E;02i+@)g z6dOD;RLnO#2tPjH;A*amup2khUwo5$e3Dg%ieCE(M2icDRDAJjtj#qz@b5nnbk-C`q z8;P+2xx1$ak`CtG-geJ&%BpIyoygv@-$x;suDO|RP?<8r+SPAPElmAhj$hwp)HTJ; zd#c~wVJ{8f&G)1wz`5+Dr|ew)Gy9VsQdb=Mw5OKKp03V!-5M`~FWA3+u`79i--y){ zPrwAp&@Mf?XRkcxeG`|NaP^o&bYwJzh+33m+?;(qTU_<~%}|d0yg<{Z!l?dB~sVID`JI=^be0#kzrn%a%(^y$11T?_^Rs=a8^Kf2q^6aIz)7uP|AV^Vj#&0}4%_ zne`48zqr!#?zCRwZ12=by+=Z?X@(i}bi5Y>MV{H92LV9;=Qr{@{#lTJ^ualA2{KT2 zk|=TwRBVxPc;QiI{HoapW0I9Ncz>*g8 z6kr-1_d#~G2;@L34nct)^SnpQzix}OrD)axGuIZyn1>*+tZzZO%x^V@z6x~6G=Bbd z2ehHo0T^r~CT`;8t9NRq_W^Fg67;1Eb{jmYW%_$7o=6|DBzRV$&?OY4B*SbVdE4Jz z8W0qnPX=Iaz>iYW*hqq{qA0k?Z2_R?pgM1aRP4q{di~!b@K-F9p;cDTPdfr=9XoDO=3UqGTH)TOU>&iB8^_207+lO$=3E6PaREce;k25A>t;Q`bwk==<8R$zs{X;;r3DQuhWlDfAi+*?LpN(bK^$v#RifP2%Yr}CW(C@3f^lsfLyx^H9T;;Ydvy5LfSjnc8y-~;PC(#7uNy^ zgBF9L=23r<8_vY!&;BZNW-9OPqoU@c4{|RQVkCX>;rK8;C7bDWl_ju)@NC zGEgve`*j($|nH5u4S3DiipY`yin1IQXO=uc| zL&$5O5K3v_v3d#+w9q6}Ib({>1TEu&%m(QI$8szsWcqU5M2No6vXck@1G3l`2qfq;+JRh3S@ge4~^7EFJz{r4nY9e23mmU=1m#o@sm1HJPCtf)BF@=*Xqd38Y3 zAxlG~Vm~8JOxW{z)Yo9V@9w*!T+u=v8KVaymn)pbEGp)E9T~N+7@evaWpl{r^#AO0 zk_3K$8UA2@;jx%^Z?F0Q`Zc})v-$F+7zv}F3}?2I@+4+-@p8eiLF_apQ(uqj!UD#- zFz1Qyd^+f_F@qKUu`lc>oaX!Peq8(zNWRlzc?#9~qc^;9!6GO79OHrx8VDb}PLH#D zfbK39C?*}AVIQ^BId?RohL3P$x9ewT$swm9{$yS@9q6IIikl#`dlw_DyD(J``kHqL z3AoIgv{g?$6tN>PjkS^tLQS9`-#0vB`tx48T5JQ#lELcO@TxlpuV4=O#{OKBsL78V z^FQa^rk`Olzwk}tI(giAfT9UKUmkgi9w2w+7|5eH19)7h%tqendAccn&+ z^c`=uZf>o@`;1WJtbax%iXm!6L9et%GAVRom&pR@W4Tn%!0|25W%773_`Cksxif(W zMw|N`MnN%^Dpx;;<=)cP;c@tra)-r{?}%qRJ4QxhG?agiJ2%EL{_6=@Zt9beGfr3- zn0H9Xqi#D;<_hRyWKrj%{Ld~nhP0ZvD$!v+d*0F?PNw?fnU8Z~2&w0~$}%h(Mi`?P z-x)=dBr0%Yo?Hap`QSyDyk&F~{4-O?69* zyu7FL1B^uHzh}Lbj&C(6u8%i9#h;yxK%E&}3}o{!mdA$x8*6DOmeN+q&&z*>J_?h> zQ)|}GMp<+;im#oyEz?iWvE>$GMED+s4eMIf$e%fz;oVtFhWoOdXmCgk?+DvMKPCU( zo0DvJJY5}ElGTPUtmfQ*i;CjQp0o4d6TSC#Pg_;lMZRx?pO z2$vQvv#F6J?=Y)#5wmk`U=K;*5P+=Jn`?(c8N~; z26#I--+InWm*PmRCGgJ_NJ27JzfIG%HkGn&aYhZxj4-t%%Gyv>?-b>e!~GQR`b*`f z$JcZg@KsM9CKibchOc(?JtGGD7s;%4C-I*rF&|@QXxL*C3py>zH{$2%m-HdBq|C$L zSkn5RXn^!mOixs;Vqs_8lcN+Nu>UTE(5w)jqgLV@supF@a-tSKNFYb?5yM$J{7Y;)BrbQ%8sR0a()qLKi4cOieei&(lXl4gKO$ao=Yo zIQ}jZ1C~y&e33cw_V>6l7in2i?L?#Oq5QV+0ffRLF?&SaAEA-;2261g&tHT=juJ4a zIUwiw(91sdC+HjoH}7NZhFx*md}y$4?U8CayB^2(TMKEutFKADtKZ@#g5271FZisX z?DmwM?K9U?maKe{U9JOOqzHM6Fm|{`@79!Nzc-voNBbfwuF{AsP55V_-!KjAu`FEL zDC0UKY{}c2Bu99xtkeT=sp6LJ{`E0} z63y9l1Ga@2Br@OKrKQiqrK4W;t=H#*F*I7m)5~TjW_)|tbb3BY6>fnY+fvQ8Uv0X3 z-~gFkVv-c#_UQ6D{~M}~QWgNJNv&ECAqIctCpt^@D-dyI8SM(UKXM`@?zmaT-*pA8LL|Tm<2yD&PIqcd0v~nJg$g>}7h+iK*SldT~ zFRtw#cv9uGCx-oOSskw2-dp!X=b=xY<4iTy5#2GWRRf4y_X`hpD|NkuE3Yh5JV&?S z&Jx70Z@=x>^+5VN<{>#z2%q+hM~u@f%rLmckkjW8FQ(l-BI{@$rn7{iPJ)~)g^bIe z4LebVgwZu0Mk&iDnnH{sd0qN(1=l(YsXe%!Em01v;Imtxf~wh{p|Q%EABk~Fh94BS zQa-5v*Vn@yddp4cE4}a8i?z}}-BYvh=ZuPSbgaj~#m;QKhal#tL?yjYQ= z`%YkM<@h@J_Fe|z*|mG3*<`qZ3i_}mVfrr+xi3lS41BA<#O>k0a#K$P>K(~!L)%0V zomNtULdDFrtpyF?$;X^z|Mh=A+y)vpI3C~X;n^4o)9+vs=ymsT8-4EnMEe-^`-SLd z*YWXv_uNA!oPN{M@Cwi2lwNca=9$(z`62)%;P2T4g`^a~{~8F8W5%&kH(0;5fTSq3 z=CuxcQdQ1d@^IKBP>v;|+Qe$iY}xZDDMt^{w1?@`hUVqfT zh5c`*7(`wafFgHsOWgS9H3@n+?spu*g+=z5T2O_8-SHgMR715nXx+5pP5RlVL)rz# z9}i2NtJ+}jFg@#oWb^)9vfj8%KoQjm+X;L+({6M@w=uJL zm*7M&j*_lgli6I4VB@qD4Mczyl~zRYl^jK>%)U%*b~^>Z=#qs(1Xalps@1C4Ry&nl zH+VXC;IhxHv@LT>Bl`U(sd(OPLU*Y>A> zV*TsO;E8{nh@ZIs{q^54W*$AQD|83Zu{`%j4V*Edc|9`!X4hBx+ZnrJQ|1oOT zL?Fq2JH2!F{*UQTv4PVFqvZIn%>H{k;LZS{TveiY0gqGH_V2O%&-C3vBx_3k8}@(B z=zsq-{XGyI)=9Orj{iLWKifeUlm|xbZdUrAW&0n0PX_sNlJ4^x|JSCi0aH7K^9sfP zTDyM+m2eYmN5*@lSdqU*@Berc_$UV1?`tVP|1FsR^RLt;z;=+WX7rf+AJhN;DMVpj z-VgyVV!*%x)Lj>W`n%7^_&pE*Ga9-J-mG`0Za{oDe-{@E6Oa6MK8fP}8xfLKuA2}n z+ObbVQyVyR_WJT~U$dyC4uZa^=qQ2bjf8?zK@YbqwgA1cQ%iGmA;9bQsH&=NRkUCF z&wCOoFJ0dlrwcDp?4teGf5<&F1ffYsP$Z}8Psx)Fh)#0-0JuaVlM=MRa#$V{wu`we z2r%RM$b)H42uxpr&aKD>l-m0|6@Xr}7#OJ#!9IH5^V9MzfNgxv#a6IK&~u%&+{ z7FJ#J3gi4j)%KWuJjhB9xy;7SvU&hGa^mA^z|IaR4|RJjR_~XAaJ_q$th{^?RED<% z{yg0~)ZM0awhpAxEdhezhZB4aW0L#dYsTOUBsAIi(P<&V>0P9P4aNOH<;sh;o<`Dm zQX%I>#6UKvvFuNhCMco4Z<}xkf|4Tgt-9e|0B;@v8eT1zfJqKD(0VQOGj~->y=8{j zpq8TA%_s|?4mq++vCqMnN+ITmr9q?1o_YYRA6(g-r{vs5g6)3b0wDRSf(QIoJm&Rj zpwHMKKrJmuOz_a#L(D5F9cBceQRyxw zqL&)9DjA&#K#(_s3Xrxy)m?!7&Ngm~3a<2BG}RA>2w5z3oFvUwC+mv=Rba_DK#G_2 zxiV5>-?N*NDG>qiY*Fj-)P0RKOxAIa?bsH#UKMF=z{L|8<`+PCvtn{z6dr5-^LatJ=oM?DBw%IB75PwJD;wi~+Mu#`fk zWxEvrikJ5rqR#YBPysAd=rQhfTm|u>=URDUz4svm2>t}YZ7`Ga;2R*cUj$*PUuLr) zx!0g}-&+7PG&7tB)dyBmPXE7cEO{AZ{`I=Q{of~_v@13N7gub*4?N3wyDjjTBOiHI z>t|VBX?d?=|;HYhPC_7?Q09<%{1l9;b{<*K9^|M&A>}1!=J>K-%SNBg5|metT$I5LTM!fXW|Y{ zcB!-4!V9kif$b3B8rQ7ran)@4P0U`zbm)y zU|-BkC-!|dYr_{YoK0+(edf09M`^$z$4M7*Rqal>EKaFu(1`H|_UrCG1QjG(F8cvb z1o{m;e(lbC8FiF8rO^dAYB%YGfBQB>Lsw-_gIk|Q#iExxwrp!(cHz=2C9aH&xkA0C zcKmv^dTr2BFZMgrf#;6zuHO1$w^-}XZOIve8R$}y{yp*p6sfXjA$i;N>&dBV8zP$K_CNB`#KbHWC)fl-Lu$3Z}%G`aU0NW zrNC5Dx@@9U)N7YUw%#KF7ajx8hU+W=)*Vxql?EsI);;O`l%3Lj3AlymO_Ksw$&V$j z4{H@%o|mjKDS2@r=HO)D{R8gncJl21QvvkKoqNEtcZzn!C*Y|z_kcUqlYtv5S7|0J z?w=&KF8qTF<8)xi-UJm-(!gz@zO&8F*2XM*zN#GEN%HiNey^y3}a$~k30|VzWPZ!4!L>-945-@LMj^ApO=etA+ zUS`4*4A@ABxplyj^-;PnXwny6+@i52uz*Tmq@gk7W+2SWj`_f3`6Dt|7&)z?nW@qU zO1c#uCW0_sSWJ2Z?A@r;UAuyN+7Z}IPynjoloRsBGW!e5sf|BC%WvkJnj-Qjnsb0H zJt5#gxo1ArnXfze!%(9~yZL};JEv|894@Q>J8f`>Z`Un2=ze`m+ U@L>6#8U`TnboFyt=akR{0ODduAOHXW literal 0 HcmV?d00001 diff --git a/docs/docs/tutorials/rag/langgraph_agentic_rag.ipynb b/docs/docs/tutorials/rag/langgraph_agentic_rag.ipynb deleted file mode 100644 index e9fe98d47..000000000 --- a/docs/docs/tutorials/rag/langgraph_agentic_rag.ipynb +++ /dev/null @@ -1,928 +0,0 @@ -{ - "cells": [ - { - "attachments": { - "7ad1a116-28d7-473f-8cff-5f2efd0bf118.png": { - "image/png": "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" - } - }, - "cell_type": "markdown", - "id": "46397e48-aba2-4bd3-bb31-d840f2fcbda4", - "metadata": { - "editable": true, - "slideshow": { - "slide_type": "" - }, - "tags": [] - }, - "source": [ - "# Agentic RAG\n", - "\n", - "In this tutorial we will build a [retrieval agent](https://python.langchain.com/docs/tutorials/qa_chat_history). Retrieval agents are useful when you want an LLM to make a decision about whether to retrieve context from a vectorstore or respond to the user directly.\n", - "\n", - "By the end of the tutorial we will have done the following:\n", - "\n", - "1. Fetch and preprocess documents that will be used for retrieval.\n", - "2. Index those documents for semantic search and create a retriever tool for the agent.\n", - "3. Build an agentic RAG system that can decide when to use the retriever tool.\n", - "\n", - "![Screenshot 2024-02-14 at 3.43.58 PM.png](attachment:7ad1a116-28d7-473f-8cff-5f2efd0bf118.png)\n", - "\n", - "## Setup\n", - "\n", - "Let's download the required packages and set our API keys:" - ] - }, - { - "cell_type": "code", - "execution_count": 1, - "id": "969fb438", - "metadata": {}, - "outputs": [], - "source": [ - "%%capture --no-stderr\n", - "%pip install -U --quiet langgraph \"langchain[openai]\" langchain-community langchain-text-splitters" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "e4958a8c", - "metadata": {}, - "outputs": [], - "source": [ - "import getpass\n", - "import os\n", - "\n", - "\n", - "def _set_env(key: str):\n", - " if key not in os.environ:\n", - " os.environ[key] = getpass.getpass(f\"{key}:\")\n", - "\n", - "\n", - "_set_env(\"OPENAI_API_KEY\")" - ] - }, - { - "cell_type": "markdown", - "id": "3d07e8d4", - "metadata": {}, - "source": [ - "" - ] - }, - { - "cell_type": "markdown", - "id": "c74e4532", - "metadata": {}, - "source": [ - "## 1. Preprocess documents\n", - "\n", - "1\\. Fetch documents to use in our RAG system. We will use three of the most recent pages from [Lilian Weng's excellent blog](https://lilianweng.github.io/). We'll start by fetching the content of the pages using `WebBaseLoader` utility:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "53ff068e-053f-41da-a8b5-b23efc2442fd", - "metadata": { - "scrolled": true - }, - "outputs": [], - "source": [ - "from langchain_community.document_loaders import WebBaseLoader\n", - "\n", - "urls = [\n", - " \"https://lilianweng.github.io/posts/2024-11-28-reward-hacking/\",\n", - " \"https://lilianweng.github.io/posts/2024-07-07-hallucination/\",\n", - " \"https://lilianweng.github.io/posts/2024-04-12-diffusion-video/\",\n", - "]\n", - "\n", - "docs = [WebBaseLoader(url).load() for url in urls]" - ] - }, - { - "cell_type": "code", - "execution_count": 4, - "id": "0e027561-c144-469c-913d-c3d8295500ed", - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "\"Reward Hacking in Reinforcement Learning | Lil'Log\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\nLil'Log\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n|\\n\\n\\n\\n\\n\\n\\nPosts\\n\\n\\n\\n\\nArchive\\n\\n\\n\\n\\nSearch\\n\\n\\n\\n\\nTags\\n\\n\\n\\n\\nFAQ\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n Reward Hacking in Reinforcement Learning\\n \\nDate: November 28, 2024 | Estimated Reading Time: 37 min | Author: Lilian Weng\\n\\n\\n \\n\\n\\nTable of Contents\\n\\n\\n\\nBackground\\n\\nReward Function in RL\\n\\nSpurious Correlation\\n\\n\\nLet’s Define Reward Hacking\\n\\nList of Examples\\n\\nReward hacking examples in RL tasks\\n\\nReward hacking examples in LLM tasks\\n\\nReward hacking examples in real life\\n\\n\\nWhy does Reward Hacking Exist?\\n\\n\\nHacking RL Environment\\n\\nHacking RLHF of LLMs\\n\\nHacking the Training Process\\n\\nHacking the Evaluator\\n\\nIn-Context Reward Hacking\\n\\n\\nGeneralization of Hacking Skills\\n\\nPeek into Mitigations\\n\\nRL Algorithm Improvement\\n\\nDetecting Reward Hacking\\n\\nData Analysis of RLHF\\n\\n\\nCitation\\n\\nReferences\\n\\n\\n\\n\\n\\nReward hacking occurs when a reinforcement learning (RL) agent exploits flaws or ambiguities in the reward function to ac\"" - ] - }, - "execution_count": 4, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "docs[0][0].page_content.strip()[:1000]" - ] - }, - { - "cell_type": "markdown", - "id": "ca258767-ce4f-4eaa-8ae5-25e9361675e1", - "metadata": {}, - "source": [ - "2\\. Split the fetched documents into smaller chunks for indexing into our vectorstore:" - ] - }, - { - "cell_type": "code", - "execution_count": 5, - "id": "24d2cefd-d0f0-44ae-aa4b-0436a675dc56", - "metadata": {}, - "outputs": [], - "source": [ - "from langchain_text_splitters import RecursiveCharacterTextSplitter\n", - "\n", - "docs_list = [item for sublist in docs for item in sublist]\n", - "\n", - "text_splitter = RecursiveCharacterTextSplitter.from_tiktoken_encoder(\n", - " chunk_size=100, chunk_overlap=50\n", - ")\n", - "doc_splits = text_splitter.split_documents(docs_list)" - ] - }, - { - "cell_type": "code", - "execution_count": 6, - "id": "a82479d9-e7d3-481f-893d-0a95c404f0bf", - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "\"Reward Hacking in Reinforcement Learning | Lil'Log\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\nLil'Log\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n|\\n\\n\\n\\n\\n\\n\\nPosts\\n\\n\\n\\n\\nArchive\\n\\n\\n\\n\\nSearch\\n\\n\\n\\n\\nTags\\n\\n\\n\\n\\nFAQ\"" - ] - }, - "execution_count": 6, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "doc_splits[0].page_content.strip()" - ] - }, - { - "cell_type": "markdown", - "id": "c5e7b434-c4c3-4a42-aa4e-2a0bb1e7a2a0", - "metadata": {}, - "source": [ - "## 2. Create a retriever tool" - ] - }, - { - "cell_type": "markdown", - "id": "732d6512-51af-426d-995f-98ae72eff4af", - "metadata": {}, - "source": [ - "Now that we have our split documents, we can index them into a vector store that we'll use for semantic search. \n", - "\n", - "1\\. Use an in-memory vector store and OpenAI embeddings:" - ] - }, - { - "cell_type": "code", - "execution_count": 7, - "id": "e50c9efe-4abe-42fa-b35a-05eeeede9ec6", - "metadata": {}, - "outputs": [], - "source": [ - "from langchain_core.vectorstores import InMemoryVectorStore\n", - "from langchain_openai import OpenAIEmbeddings\n", - "\n", - "vectorstore = InMemoryVectorStore.from_documents(\n", - " documents=doc_splits, embedding=OpenAIEmbeddings()\n", - ")\n", - "retriever = vectorstore.as_retriever()" - ] - }, - { - "cell_type": "markdown", - "id": "225d2277-45b2-4ae8-a7d6-62b07fb4a002", - "metadata": {}, - "source": [ - "2\\. Create a retriever tool using LangChain's prebuilt `create_retriever_tool`:" - ] - }, - { - "cell_type": "code", - "execution_count": 8, - "id": "0b97bdd8-d7e3-444d-ac96-5ef4725f9048", - "metadata": {}, - "outputs": [], - "source": [ - "from langchain.tools.retriever import create_retriever_tool\n", - "\n", - "retriever_tool = create_retriever_tool(\n", - " retriever,\n", - " \"retrieve_blog_posts\",\n", - " \"Search and return information about Lilian Weng blog posts.\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "9cfdc027-5ea8-42ba-b0f4-0590505d76ff", - "metadata": {}, - "source": [ - "3\\. Test the tool:" - ] - }, - { - "cell_type": "code", - "execution_count": 9, - "id": "58c17a90-9a85-46c9-96a9-1e7b224330c4", - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "'(Note: Some work defines reward tampering as a distinct category of misalignment behavior from reward hacking. But I consider reward hacking as a broader concept here.)\\nAt a high level, reward hacking can be categorized into two types: environment or goal misspecification, and reward tampering.\\n\\nWhy does Reward Hacking Exist?#\\n\\nPan et al. (2022) investigated reward hacking as a function of agent capabilities, including (1) model size, (2) action space resolution, (3) observation space noise, and (4) training time. They also proposed a taxonomy of three types of misspecified proxy rewards:\\n\\nLet’s Define Reward Hacking#\\nReward shaping in RL is challenging. Reward hacking occurs when an RL agent exploits flaws or ambiguities in the reward function to obtain high rewards without genuinely learning the intended behaviors or completing the task as designed. In recent years, several related concepts have been proposed, all referring to some form of reward hacking:'" - ] - }, - "execution_count": 9, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "retriever_tool.invoke({\"query\": \"types of reward hacking\"})" - ] - }, - { - "cell_type": "markdown", - "id": "7c4e9e70-3de1-4ab4-ab42-c298cef00eb6", - "metadata": {}, - "source": [ - "## 3. Generate query" - ] - }, - { - "cell_type": "markdown", - "id": "2ba89a91-3dc5-4663-a6a5-43ccb2575849", - "metadata": {}, - "source": [ - "Now we will start building components ([nodes](../../../concepts/low_level#nodes) and [edges](../../../concepts/low_level#edges)) for our agentic RAG graph. Note that the components will operate on the [`MessagesState`](../../../concepts/low_level#messagesstate) — graph state that contains a `messages` key with a list of [chat messages](https://python.langchain.com/docs/concepts/messages/).\n", - "\n", - "1\\. Build a `generate_query_or_respond` node. It will call an LLM to generate a response based on the current graph state (list of messages). Given the input messages, it will decide to retrieve using the retriever tool, or respond directly to the user. Note that we're giving the chat model access to the `retriever_tool` we created earlier via `.bind_tools`:" - ] - }, - { - "cell_type": "code", - "execution_count": 10, - "id": "32f21d6f-9d0e-44e1-8edd-1fd903164d6d", - "metadata": {}, - "outputs": [], - "source": [ - "from langgraph.graph import MessagesState\n", - "from langchain.chat_models import init_chat_model\n", - "\n", - "response_model = init_chat_model(\"openai:gpt-4.1\", temperature=0)\n", - "\n", - "\n", - "def generate_query_or_respond(state: MessagesState):\n", - " \"\"\"Call the model to generate a response based on the current state. Given\n", - " the question, it will decide to retrieve using the retriever tool, or simply respond to the user.\n", - " \"\"\"\n", - " response = (\n", - " response_model\n", - " # highlight-next-line\n", - " .bind_tools([retriever_tool]).invoke(state[\"messages\"])\n", - " )\n", - " return {\"messages\": [response]}" - ] - }, - { - "cell_type": "markdown", - "id": "552e0f6c-5af5-4c13-91a6-437f320a4ed6", - "metadata": {}, - "source": [ - "2\\. Try it on a random input:" - ] - }, - { - "cell_type": "code", - "execution_count": 11, - "id": "5cdc7e76-8680-42e6-9cdb-44aa4104f0f7", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "\n", - "Hello! How can I help you today?\n" - ] - } - ], - "source": [ - "input = {\"messages\": [{\"role\": \"user\", \"content\": \"hello!\"}]}\n", - "generate_query_or_respond(input)[\"messages\"][-1].pretty_print()" - ] - }, - { - "cell_type": "markdown", - "id": "df45f7c8-df6c-4687-9f25-0357e29532cb", - "metadata": {}, - "source": [ - "3\\. Ask a question that requires semantic search:" - ] - }, - { - "cell_type": "code", - "execution_count": 12, - "id": "845e1bca-3097-423f-a9c8-967fe1216375", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Tool Calls:\n", - " retrieve_blog_posts (call_tYQxgfIlnQUDMdtAhdbXNwIM)\n", - " Call ID: call_tYQxgfIlnQUDMdtAhdbXNwIM\n", - " Args:\n", - " query: types of reward hacking\n" - ] - } - ], - "source": [ - "input = {\n", - " \"messages\": [\n", - " {\n", - " \"role\": \"user\",\n", - " \"content\": \"What does Lilian Weng say about types of reward hacking?\",\n", - " }\n", - " ]\n", - "}\n", - "generate_query_or_respond(input)[\"messages\"][-1].pretty_print()" - ] - }, - { - "cell_type": "markdown", - "id": "3e4756cb-b952-47a7-be4a-4f8a0866b67f", - "metadata": {}, - "source": [ - "## 4. Grade documents" - ] - }, - { - "cell_type": "markdown", - "id": "0b278967-18ab-409e-9ca4-6ee44a23e708", - "metadata": {}, - "source": [ - "1\\. Add a [conditional edge](../../../concepts/low_level#conditional-edges) — `grade_documents` — to determine whether the retrieved documents are relevant to the question. We will use a model with a structured output schema `GradeDocuments` for document grading. The `grade_documents` function will return the name of the node to go to based on the grading decision (`generate_answer` or `rewrite_question`):" - ] - }, - { - "cell_type": "code", - "execution_count": 13, - "id": "98f0e1a5-5ca8-4d8c-a94c-fc0976ae1937", - "metadata": {}, - "outputs": [], - "source": [ - "from pydantic import BaseModel, Field\n", - "from typing import Literal\n", - "\n", - "GRADE_PROMPT = (\n", - " \"You are a grader assessing relevance of a retrieved document to a user question. \\n \"\n", - " \"Here is the retrieved document: \\n\\n {context} \\n\\n\"\n", - " \"Here is the user question: {question} \\n\"\n", - " \"If the document contains keyword(s) or semantic meaning related to the user question, grade it as relevant. \\n\"\n", - " \"Give a binary score 'yes' or 'no' score to indicate whether the document is relevant to the question.\"\n", - ")\n", - "\n", - "\n", - "# highlight-next-line\n", - "class GradeDocuments(BaseModel):\n", - " \"\"\"Grade documents using a binary score for relevance check.\"\"\"\n", - "\n", - " binary_score: str = Field(\n", - " description=\"Relevance score: 'yes' if relevant, or 'no' if not relevant\"\n", - " )\n", - "\n", - "\n", - "grader_model = init_chat_model(\"openai:gpt-4.1\", temperature=0)\n", - "\n", - "\n", - "def grade_documents(\n", - " state: MessagesState,\n", - ") -> Literal[\"generate_answer\", \"rewrite_question\"]:\n", - " \"\"\"Determine whether the retrieved documents are relevant to the question.\"\"\"\n", - " question = state[\"messages\"][0].content\n", - " context = state[\"messages\"][-1].content\n", - "\n", - " prompt = GRADE_PROMPT.format(question=question, context=context)\n", - " response = (\n", - " grader_model\n", - " # highlight-next-line\n", - " .with_structured_output(GradeDocuments).invoke(\n", - " [{\"role\": \"user\", \"content\": prompt}]\n", - " )\n", - " )\n", - " score = response.binary_score\n", - "\n", - " if score == \"yes\":\n", - " return \"generate_answer\"\n", - " else:\n", - " return \"rewrite_question\"" - ] - }, - { - "cell_type": "markdown", - "id": "ddbdc0c2-6f0c-41ec-b4d1-ad20d4498006", - "metadata": {}, - "source": [ - "2\\. Run this with irrelevant documents in the tool response:" - ] - }, - { - "cell_type": "code", - "execution_count": 14, - "id": "0a62e5f7-4c4e-44cd-b4bd-77b6879ce970", - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "'rewrite_question'" - ] - }, - "execution_count": 14, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "from langchain_core.messages import convert_to_messages\n", - "\n", - "input = {\n", - " \"messages\": convert_to_messages(\n", - " [\n", - " {\n", - " \"role\": \"user\",\n", - " \"content\": \"What does Lilian Weng say about types of reward hacking?\",\n", - " },\n", - " {\n", - " \"role\": \"assistant\",\n", - " \"content\": \"\",\n", - " \"tool_calls\": [\n", - " {\n", - " \"id\": \"1\",\n", - " \"name\": \"retrieve_blog_posts\",\n", - " \"args\": {\"query\": \"types of reward hacking\"},\n", - " }\n", - " ],\n", - " },\n", - " {\"role\": \"tool\", \"content\": \"meow\", \"tool_call_id\": \"1\"},\n", - " ]\n", - " )\n", - "}\n", - "grade_documents(input)" - ] - }, - { - "cell_type": "markdown", - "id": "ba25a787-004e-4e6c-8f22-62087af7da1d", - "metadata": {}, - "source": [ - "3\\. Confirm that the relevant documents are classified as such:" - ] - }, - { - "cell_type": "code", - "execution_count": 15, - "id": "8bd5d24b-f05e-47c4-b954-267ab514a090", - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "'generate_answer'" - ] - }, - "execution_count": 15, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "input = {\n", - " \"messages\": convert_to_messages(\n", - " [\n", - " {\n", - " \"role\": \"user\",\n", - " \"content\": \"What does Lilian Weng say about types of reward hacking?\",\n", - " },\n", - " {\n", - " \"role\": \"assistant\",\n", - " \"content\": \"\",\n", - " \"tool_calls\": [\n", - " {\n", - " \"id\": \"1\",\n", - " \"name\": \"retrieve_blog_posts\",\n", - " \"args\": {\"query\": \"types of reward hacking\"},\n", - " }\n", - " ],\n", - " },\n", - " {\n", - " \"role\": \"tool\",\n", - " \"content\": \"reward hacking can be categorized into two types: environment or goal misspecification, and reward tampering\",\n", - " \"tool_call_id\": \"1\",\n", - " },\n", - " ]\n", - " )\n", - "}\n", - "grade_documents(input)" - ] - }, - { - "cell_type": "markdown", - "id": "4c337d45-bc35-4e88-8b27-ebc6631f9fed", - "metadata": {}, - "source": [ - "## 5. Rewrite question" - ] - }, - { - "cell_type": "markdown", - "id": "e8d67fdf-0b77-4b90-97f2-63a571c3e496", - "metadata": {}, - "source": [ - "1\\. Build the `rewrite_question` node. The retriever tool can return potentially irrelevant documents, which indicates a need to improve the original user question. To do so, we will call the `rewrite_question` node:" - ] - }, - { - "cell_type": "code", - "execution_count": 16, - "id": "278d1d83-dda6-4de4-bf8b-be9965c227fa", - "metadata": {}, - "outputs": [], - "source": [ - "REWRITE_PROMPT = (\n", - " \"Look at the input and try to reason about the underlying semantic intent / meaning.\\n\"\n", - " \"Here is the initial question:\"\n", - " \"\\n ------- \\n\"\n", - " \"{question}\"\n", - " \"\\n ------- \\n\"\n", - " \"Formulate an improved question:\"\n", - ")\n", - "\n", - "\n", - "def rewrite_question(state: MessagesState):\n", - " \"\"\"Rewrite the original user question.\"\"\"\n", - " messages = state[\"messages\"]\n", - " question = messages[0].content\n", - " prompt = REWRITE_PROMPT.format(question=question)\n", - " response = response_model.invoke([{\"role\": \"user\", \"content\": prompt}])\n", - " return {\"messages\": [{\"role\": \"user\", \"content\": response.content}]}" - ] - }, - { - "cell_type": "markdown", - "id": "1ec05f29-48a2-4325-8c56-da60490f2a60", - "metadata": {}, - "source": [ - "2\\. Try it out:" - ] - }, - { - "cell_type": "code", - "execution_count": 17, - "id": "15404075-b7c8-4248-84d3-ca2b361aa3d2", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "What are the different types of reward hacking described by Lilian Weng, and how does she explain them?\n" - ] - } - ], - "source": [ - "input = {\n", - " \"messages\": convert_to_messages(\n", - " [\n", - " {\n", - " \"role\": \"user\",\n", - " \"content\": \"What does Lilian Weng say about types of reward hacking?\",\n", - " },\n", - " {\n", - " \"role\": \"assistant\",\n", - " \"content\": \"\",\n", - " \"tool_calls\": [\n", - " {\n", - " \"id\": \"1\",\n", - " \"name\": \"retrieve_blog_posts\",\n", - " \"args\": {\"query\": \"types of reward hacking\"},\n", - " }\n", - " ],\n", - " },\n", - " {\"role\": \"tool\", \"content\": \"meow\", \"tool_call_id\": \"1\"},\n", - " ]\n", - " )\n", - "}\n", - "\n", - "response = rewrite_question(input)\n", - "print(response[\"messages\"][-1][\"content\"])" - ] - }, - { - "cell_type": "markdown", - "id": "55eab9fe-f1d6-4ad3-93d6-4638a92c9395", - "metadata": {}, - "source": [ - "## 6. Generate an answer" - ] - }, - { - "cell_type": "markdown", - "id": "d9362002-e66e-4e58-90af-d225c184b5d1", - "metadata": {}, - "source": [ - "1\\. Build `generate_answer` node: if we pass the grader checks, we can generate the final answer based on the original question and the retrieved context:" - ] - }, - { - "cell_type": "code", - "execution_count": 18, - "id": "20b0e60a-a5aa-4e95-a15a-2c91382e508f", - "metadata": {}, - "outputs": [], - "source": [ - "GENERATE_PROMPT = (\n", - " \"You are an assistant for question-answering tasks. \"\n", - " \"Use the following pieces of retrieved context to answer the question. \"\n", - " \"If you don't know the answer, just say that you don't know. \"\n", - " \"Use three sentences maximum and keep the answer concise.\\n\"\n", - " \"Question: {question} \\n\"\n", - " \"Context: {context}\"\n", - ")\n", - "\n", - "\n", - "def generate_answer(state: MessagesState):\n", - " \"\"\"Generate an answer.\"\"\"\n", - " question = state[\"messages\"][0].content\n", - " context = state[\"messages\"][-1].content\n", - " prompt = GENERATE_PROMPT.format(question=question, context=context)\n", - " response = response_model.invoke([{\"role\": \"user\", \"content\": prompt}])\n", - " return {\"messages\": [response]}" - ] - }, - { - "cell_type": "markdown", - "id": "8f59a1fc-fe75-4907-ae86-0dd33c4d9a7f", - "metadata": {}, - "source": [ - "2\\. Try it:" - ] - }, - { - "cell_type": "code", - "execution_count": 19, - "id": "15d84cb0-b529-49d0-b18e-1d03ff3f7c0d", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "\n", - "Lilian Weng says that reward hacking can be categorized into two types: environment or goal misspecification, and reward tampering. These categories describe different ways in which an agent might exploit flaws in the reward system. Environment or goal misspecification involves unintended behaviors due to poorly specified objectives, while reward tampering involves directly manipulating the reward signal.\n" - ] - } - ], - "source": [ - "input = {\n", - " \"messages\": convert_to_messages(\n", - " [\n", - " {\n", - " \"role\": \"user\",\n", - " \"content\": \"What does Lilian Weng say about types of reward hacking?\",\n", - " },\n", - " {\n", - " \"role\": \"assistant\",\n", - " \"content\": \"\",\n", - " \"tool_calls\": [\n", - " {\n", - " \"id\": \"1\",\n", - " \"name\": \"retrieve_blog_posts\",\n", - " \"args\": {\"query\": \"types of reward hacking\"},\n", - " }\n", - " ],\n", - " },\n", - " {\n", - " \"role\": \"tool\",\n", - " \"content\": \"reward hacking can be categorized into two types: environment or goal misspecification, and reward tampering\",\n", - " \"tool_call_id\": \"1\",\n", - " },\n", - " ]\n", - " )\n", - "}\n", - "\n", - "response = generate_answer(input)\n", - "response[\"messages\"][-1].pretty_print()" - ] - }, - { - "cell_type": "markdown", - "id": "955882ef-7467-48db-ae51-de441f2fc3a7", - "metadata": {}, - "source": [ - "## 7. Assemble the graph\n", - "\n", - "* Start with a `generate_query_or_respond` and determine if we need to call `retriever_tool`\n", - "* Route to next step using `tools_condition`:\n", - " * If `generate_query_or_respond` returned `tool_calls`, call `retriever_tool` to retrieve context \n", - " * Otherwise, respond directly to the user\n", - "* Grade retrieved document content for relevance to the question (`grade_documents`) and route to next step:\n", - " * If not relevant, rewrite the question using `rewrite_question` and then call `generate_query_or_respond` again\n", - " * If relevant, proceed to `generate_answer` and generate final response using the `ToolMessage` with the retrieved document context" - ] - }, - { - "cell_type": "code", - "execution_count": 20, - "id": "8718a37f-83c2-4f16-9850-e61e0f49c3d4", - "metadata": {}, - "outputs": [], - "source": [ - "from langgraph.graph import StateGraph, START, END\n", - "from langgraph.prebuilt import ToolNode\n", - "from langgraph.prebuilt import tools_condition\n", - "\n", - "workflow = StateGraph(MessagesState)\n", - "\n", - "# Define the nodes we will cycle between\n", - "workflow.add_node(generate_query_or_respond)\n", - "workflow.add_node(\"retrieve\", ToolNode([retriever_tool]))\n", - "workflow.add_node(rewrite_question)\n", - "workflow.add_node(generate_answer)\n", - "\n", - "workflow.add_edge(START, \"generate_query_or_respond\")\n", - "\n", - "# Decide whether to retrieve\n", - "workflow.add_conditional_edges(\n", - " \"generate_query_or_respond\",\n", - " # Assess LLM decision (call `retriever_tool` tool or respond to the user)\n", - " tools_condition,\n", - " {\n", - " # Translate the condition outputs to nodes in our graph\n", - " \"tools\": \"retrieve\",\n", - " END: END,\n", - " },\n", - ")\n", - "\n", - "# Edges taken after the `action` node is called.\n", - "workflow.add_conditional_edges(\n", - " \"retrieve\",\n", - " # Assess agent decision\n", - " grade_documents,\n", - ")\n", - "workflow.add_edge(\"generate_answer\", END)\n", - "workflow.add_edge(\"rewrite_question\", \"generate_query_or_respond\")\n", - "\n", - "# Compile\n", - "graph = workflow.compile()" - ] - }, - { - "cell_type": "markdown", - "id": "fb7c8138-8b01-4858-b9d1-b4bc5f7ebb33", - "metadata": {}, - "source": [ - "Visualize the graph:" - ] - }, - { - "cell_type": "code", - "execution_count": 25, - "id": "7b5a1d35", - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "from IPython.display import Image, display\n", - "\n", - "display(Image(graph.get_graph().draw_mermaid_png()))" - ] - }, - { - "cell_type": "markdown", - "id": "eb655cfb-d4e4-4f6a-8a0d-01e8611b18ab", - "metadata": {}, - "source": [ - "## 8. Run the agentic RAG" - ] - }, - { - "cell_type": "code", - "execution_count": 23, - "id": "7649f05a-cb67-490d-b24a-74d41895139a", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Update from node generate_query_or_respond\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Tool Calls:\n", - " retrieve_blog_posts (call_NYu2vq4km9nNNEFqJwefWKu1)\n", - " Call ID: call_NYu2vq4km9nNNEFqJwefWKu1\n", - " Args:\n", - " query: types of reward hacking\n", - "\n", - "\n", - "\n", - "Update from node retrieve\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: retrieve_blog_posts\n", - "\n", - "(Note: Some work defines reward tampering as a distinct category of misalignment behavior from reward hacking. But I consider reward hacking as a broader concept here.)\n", - "At a high level, reward hacking can be categorized into two types: environment or goal misspecification, and reward tampering.\n", - "\n", - "Why does Reward Hacking Exist?#\n", - "\n", - "Pan et al. (2022) investigated reward hacking as a function of agent capabilities, including (1) model size, (2) action space resolution, (3) observation space noise, and (4) training time. They also proposed a taxonomy of three types of misspecified proxy rewards:\n", - "\n", - "Let’s Define Reward Hacking#\n", - "Reward shaping in RL is challenging. Reward hacking occurs when an RL agent exploits flaws or ambiguities in the reward function to obtain high rewards without genuinely learning the intended behaviors or completing the task as designed. In recent years, several related concepts have been proposed, all referring to some form of reward hacking:\n", - "\n", - "\n", - "\n", - "Update from node generate_answer\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "\n", - "Lilian Weng categorizes reward hacking into two types: environment or goal misspecification, and reward tampering. She considers reward hacking as a broad concept that includes both of these categories. Reward hacking occurs when an agent exploits flaws or ambiguities in the reward function to achieve high rewards without performing the intended behaviors.\n", - "\n", - "\n", - "\n" - ] - } - ], - "source": [ - "for chunk in graph.stream(\n", - " {\n", - " \"messages\": [\n", - " {\n", - " \"role\": \"user\",\n", - " \"content\": \"What does Lilian Weng say about types of reward hacking?\",\n", - " }\n", - " ]\n", - " }\n", - "):\n", - " for node, update in chunk.items():\n", - " print(\"Update from node\", node)\n", - " update[\"messages\"][-1].pretty_print()\n", - " print(\"\\n\\n\")" - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3 (ipykernel)", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.12.3" - } - }, - "nbformat": 4, - "nbformat_minor": 5 -} diff --git a/docs/docs/tutorials/rag/langgraph_agentic_rag.md b/docs/docs/tutorials/rag/langgraph_agentic_rag.md new file mode 100644 index 000000000..2aae24820 --- /dev/null +++ b/docs/docs/tutorials/rag/langgraph_agentic_rag.md @@ -0,0 +1,522 @@ +# Agentic RAG + +In this tutorial we will build a [retrieval agent](https://python.langchain.com/docs/tutorials/qa_chat_history). Retrieval agents are useful when you want an LLM to make a decision about whether to retrieve context from a vectorstore or respond to the user directly. + +By the end of the tutorial we will have done the following: + +1. Fetch and preprocess documents that will be used for retrieval. +2. Index those documents for semantic search and create a retriever tool for the agent. +3. Build an agentic RAG system that can decide when to use the retriever tool. + +![Screenshot 2024-02-14 at 3.43.58 PM.png](assets/screenshot_2024_02_14_3_43_58_pm.png) + +## Setup + +Let's download the required packages and set our API keys: + +```python +%%capture --no-stderr +%pip install -U --quiet langgraph "langchain[openai]" langchain-community langchain-text-splitters +``` + +```python +import getpass +import os + + +def _set_env(key: str): + if key not in os.environ: + os.environ[key] = getpass.getpass(f"{key}:") + + +_set_env("OPENAI_API_KEY") +``` + +!!! tip + Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. [LangSmith](https://docs.smith.langchain.com) lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph. + + +## 1. Preprocess documents + +1. Fetch documents to use in our RAG system. We will use three of the most recent pages from [Lilian Weng's excellent blog](https://lilianweng.github.io/). We'll start by fetching the content of the pages using `WebBaseLoader` utility: + + ```python + from langchain_community.document_loaders import WebBaseLoader + + urls = [ + "https://lilianweng.github.io/posts/2024-11-28-reward-hacking/", + "https://lilianweng.github.io/posts/2024-07-07-hallucination/", + "https://lilianweng.github.io/posts/2024-04-12-diffusion-video/", + ] + + docs = [WebBaseLoader(url).load() for url in urls] + ``` + + ```python + docs[0][0].page_content.strip()[:1000] + ``` + +2. Split the fetched documents into smaller chunks for indexing into our vectorstore: + + ```python + from langchain_text_splitters import RecursiveCharacterTextSplitter + + docs_list = [item for sublist in docs for item in sublist] + + text_splitter = RecursiveCharacterTextSplitter.from_tiktoken_encoder( + chunk_size=100, chunk_overlap=50 + ) + doc_splits = text_splitter.split_documents(docs_list) + ``` + + ```python + doc_splits[0].page_content.strip() + ``` + +## 2. Create a retriever tool + +Now that we have our split documents, we can index them into a vector store that we'll use for semantic search. + +1. Use an in-memory vector store and OpenAI embeddings: + + ```python + from langchain_core.vectorstores import InMemoryVectorStore + from langchain_openai import OpenAIEmbeddings + + vectorstore = InMemoryVectorStore.from_documents( + documents=doc_splits, embedding=OpenAIEmbeddings() + ) + retriever = vectorstore.as_retriever() + ``` + +2. Create a retriever tool using LangChain's prebuilt `create_retriever_tool`: + + ```python + from langchain.tools.retriever import create_retriever_tool + + retriever_tool = create_retriever_tool( + retriever, + "retrieve_blog_posts", + "Search and return information about Lilian Weng blog posts.", + ) + ``` + +3. Test the tool: + + ```python + retriever_tool.invoke({"query": "types of reward hacking"}) + ``` + +## 3. Generate query + +Now we will start building components ([nodes](../../concepts/low_level.md#nodes) and [edges](../../concepts/low_level.md#edges)) for our agentic RAG graph. Note that the components will operate on the [`MessagesState`](../../concepts/low_level.md#messagesstate) — graph state that contains a `messages` key with a list of [chat messages](https://python.langchain.com/docs/concepts/messages/). + +1. Build a `generate_query_or_respond` node. It will call an LLM to generate a response based on the current graph state (list of messages). Given the input messages, it will decide to retrieve using the retriever tool, or respond directly to the user. Note that we're giving the chat model access to the `retriever_tool` we created earlier via `.bind_tools`: + + ```python + from langgraph.graph import MessagesState + from langchain.chat_models import init_chat_model + + response_model = init_chat_model("openai:gpt-4.1", temperature=0) + + + def generate_query_or_respond(state: MessagesState): + """Call the model to generate a response based on the current state. Given + the question, it will decide to retrieve using the retriever tool, or simply respond to the user. + """ + response = ( + response_model + # highlight-next-line + .bind_tools([retriever_tool]).invoke(state["messages"]) + ) + return {"messages": [response]} + ``` + +2. Try it on a random input: + + ```python + input = {"messages": [{"role": "user", "content": "hello!"}]} + generate_query_or_respond(input)["messages"][-1].pretty_print() + ``` + + **Output:** + ``` + ================================== Ai Message ================================== + + Hello! How can I help you today? + ``` + +3. Ask a question that requires semantic search: + + ```python + input = { + "messages": [ + { + "role": "user", + "content": "What does Lilian Weng say about types of reward hacking?", + } + ] + } + generate_query_or_respond(input)["messages"][-1].pretty_print() + ``` + + **Output:** + ``` + ================================== Ai Message ================================== + Tool Calls: + retrieve_blog_posts (call_tYQxgfIlnQUDMdtAhdbXNwIM) + Call ID: call_tYQxgfIlnQUDMdtAhdbXNwIM + Args: + query: types of reward hacking + ``` + +## 4. Grade documents + +1. Add a [conditional edge](../../concepts/low_level.md#conditional-edges) — `grade_documents` — to determine whether the retrieved documents are relevant to the question. We will use a model with a structured output schema `GradeDocuments` for document grading. The `grade_documents` function will return the name of the node to go to based on the grading decision (`generate_answer` or `rewrite_question`): + + ```python + from pydantic import BaseModel, Field + from typing import Literal + + GRADE_PROMPT = ( + "You are a grader assessing relevance of a retrieved document to a user question. \n " + "Here is the retrieved document: \n\n {context} \n\n" + "Here is the user question: {question} \n" + "If the document contains keyword(s) or semantic meaning related to the user question, grade it as relevant. \n" + "Give a binary score 'yes' or 'no' score to indicate whether the document is relevant to the question." + ) + + + # highlight-next-line + class GradeDocuments(BaseModel): + """Grade documents using a binary score for relevance check.""" + + binary_score: str = Field( + description="Relevance score: 'yes' if relevant, or 'no' if not relevant" + ) + + + grader_model = init_chat_model("openai:gpt-4.1", temperature=0) + + + def grade_documents( + state: MessagesState, + ) -> Literal["generate_answer", "rewrite_question"]: + """Determine whether the retrieved documents are relevant to the question.""" + question = state["messages"][0].content + context = state["messages"][-1].content + + prompt = GRADE_PROMPT.format(question=question, context=context) + response = ( + grader_model + # highlight-next-line + .with_structured_output(GradeDocuments).invoke( + [{"role": "user", "content": prompt}] + ) + ) + score = response.binary_score + + if score == "yes": + return "generate_answer" + else: + return "rewrite_question" + ``` + +2. Run this with irrelevant documents in the tool response: + + ```python + from langchain_core.messages import convert_to_messages + + input = { + "messages": convert_to_messages( + [ + { + "role": "user", + "content": "What does Lilian Weng say about types of reward hacking?", + }, + { + "role": "assistant", + "content": "", + "tool_calls": [ + { + "id": "1", + "name": "retrieve_blog_posts", + "args": {"query": "types of reward hacking"}, + } + ], + }, + {"role": "tool", "content": "meow", "tool_call_id": "1"}, + ] + ) + } + grade_documents(input) + ``` + +3. Confirm that the relevant documents are classified as such: + + ```python + input = { + "messages": convert_to_messages( + [ + { + "role": "user", + "content": "What does Lilian Weng say about types of reward hacking?", + }, + { + "role": "assistant", + "content": "", + "tool_calls": [ + { + "id": "1", + "name": "retrieve_blog_posts", + "args": {"query": "types of reward hacking"}, + } + ], + }, + { + "role": "tool", + "content": "reward hacking can be categorized into two types: environment or goal misspecification, and reward tampering", + "tool_call_id": "1", + }, + ] + ) + } + grade_documents(input) + ``` + +## 5. Rewrite question + +1. Build the `rewrite_question` node. The retriever tool can return potentially irrelevant documents, which indicates a need to improve the original user question. To do so, we will call the `rewrite_question` node: + + ```python + REWRITE_PROMPT = ( + "Look at the input and try to reason about the underlying semantic intent / meaning.\n" + "Here is the initial question:" + "\n ------- \n" + "{question}" + "\n ------- \n" + "Formulate an improved question:" + ) + + + def rewrite_question(state: MessagesState): + """Rewrite the original user question.""" + messages = state["messages"] + question = messages[0].content + prompt = REWRITE_PROMPT.format(question=question) + response = response_model.invoke([{"role": "user", "content": prompt}]) + return {"messages": [{"role": "user", "content": response.content}]} + ``` + +2. Try it out: + + ```python + input = { + "messages": convert_to_messages( + [ + { + "role": "user", + "content": "What does Lilian Weng say about types of reward hacking?", + }, + { + "role": "assistant", + "content": "", + "tool_calls": [ + { + "id": "1", + "name": "retrieve_blog_posts", + "args": {"query": "types of reward hacking"}, + } + ], + }, + {"role": "tool", "content": "meow", "tool_call_id": "1"}, + ] + ) + } + + response = rewrite_question(input) + print(response["messages"][-1]["content"]) + ``` + + **Output:** + ``` + What are the different types of reward hacking described by Lilian Weng, and how does she explain them? + ``` + +## 6. Generate an answer + +1. Build `generate_answer` node: if we pass the grader checks, we can generate the final answer based on the original question and the retrieved context: + + ```python + GENERATE_PROMPT = ( + "You are an assistant for question-answering tasks. " + "Use the following pieces of retrieved context to answer the question. " + "If you don't know the answer, just say that you don't know. " + "Use three sentences maximum and keep the answer concise.\n" + "Question: {question} \n" + "Context: {context}" + ) + + + def generate_answer(state: MessagesState): + """Generate an answer.""" + question = state["messages"][0].content + context = state["messages"][-1].content + prompt = GENERATE_PROMPT.format(question=question, context=context) + response = response_model.invoke([{"role": "user", "content": prompt}]) + return {"messages": [response]} + ``` + +2. Try it: + + ```python + input = { + "messages": convert_to_messages( + [ + { + "role": "user", + "content": "What does Lilian Weng say about types of reward hacking?", + }, + { + "role": "assistant", + "content": "", + "tool_calls": [ + { + "id": "1", + "name": "retrieve_blog_posts", + "args": {"query": "types of reward hacking"}, + } + ], + }, + { + "role": "tool", + "content": "reward hacking can be categorized into two types: environment or goal misspecification, and reward tampering", + "tool_call_id": "1", + }, + ] + ) + } + + response = generate_answer(input) + response["messages"][-1].pretty_print() + ``` + + **Output:** + ``` + ================================== Ai Message ================================== + + Lilian Weng categorizes reward hacking into two types: environment or goal misspecification, and reward tampering. She considers reward hacking as a broad concept that includes both of these categories. Reward hacking occurs when an agent exploits flaws or ambiguities in the reward function to achieve high rewards without performing the intended behaviors. + ``` + +## 7. Assemble the graph + +* Start with a `generate_query_or_respond` and determine if we need to call `retriever_tool` +* Route to next step using `tools_condition`: + * If `generate_query_or_respond` returned `tool_calls`, call `retriever_tool` to retrieve context + * Otherwise, respond directly to the user +* Grade retrieved document content for relevance to the question (`grade_documents`) and route to next step: + * If not relevant, rewrite the question using `rewrite_question` and then call `generate_query_or_respond` again + * If relevant, proceed to `generate_answer` and generate final response using the `ToolMessage` with the retrieved document context + +```python +from langgraph.graph import StateGraph, START, END +from langgraph.prebuilt import ToolNode +from langgraph.prebuilt import tools_condition + +workflow = StateGraph(MessagesState) + +# Define the nodes we will cycle between +workflow.add_node(generate_query_or_respond) +workflow.add_node("retrieve", ToolNode([retriever_tool])) +workflow.add_node(rewrite_question) +workflow.add_node(generate_answer) + +workflow.add_edge(START, "generate_query_or_respond") + +# Decide whether to retrieve +workflow.add_conditional_edges( + "generate_query_or_respond", + # Assess LLM decision (call `retriever_tool` tool or respond to the user) + tools_condition, + { + # Translate the condition outputs to nodes in our graph + "tools": "retrieve", + END: END, + }, +) + +# Edges taken after the `action` node is called. +workflow.add_conditional_edges( + "retrieve", + # Assess agent decision + grade_documents, +) +workflow.add_edge("generate_answer", END) +workflow.add_edge("rewrite_question", "generate_query_or_respond") + +# Compile +graph = workflow.compile() +``` + +Visualize the graph: + +```python +from IPython.display import Image, display + +display(Image(graph.get_graph().draw_mermaid_png())) +``` + +![Graph](assets/agentic-rag-output.png) + +## 8. Run the agentic RAG + +```python +for chunk in graph.stream( + { + "messages": [ + { + "role": "user", + "content": "What does Lilian Weng say about types of reward hacking?", + } + ] + } +): + for node, update in chunk.items(): + print("Update from node", node) + update["messages"][-1].pretty_print() + print("\n\n") +``` + +**Output:** +``` +Update from node generate_query_or_respond +================================== Ai Message ================================== +Tool Calls: + retrieve_blog_posts (call_NYu2vq4km9nNNEFqJwefWKu1) + Call ID: call_NYu2vq4km9nNNEFqJwefWKu1 + Args: + query: types of reward hacking + + + +Update from node retrieve +================================= Tool Message ================================== +Name: retrieve_blog_posts + +(Note: Some work defines reward tampering as a distinct category of misalignment behavior from reward hacking. But I consider reward hacking as a broader concept here.) +At a high level, reward hacking can be categorized into two types: environment or goal misspecification, and reward tampering. + +Why does Reward Hacking Exist?# + +Pan et al. (2022) investigated reward hacking as a function of agent capabilities, including (1) model size, (2) action space resolution, (3) observation space noise, and (4) training time. They also proposed a taxonomy of three types of misspecified proxy rewards: + +Let's Define Reward Hacking# +Reward shaping in RL is challenging. Reward hacking occurs when an RL agent exploits flaws or ambiguities in the reward function to obtain high rewards without genuinely learning the intended behaviors or completing the task as designed. In recent years, several related concepts have been proposed, all referring to some form of reward hacking: + + + +Update from node generate_answer +================================== Ai Message ================================== + +Lilian Weng categorizes reward hacking into two types: environment or goal misspecification, and reward tampering. She considers reward hacking as a broad concept that includes both of these categories. Reward hacking occurs when an agent exploits flaws or ambiguities in the reward function to achieve high rewards without performing the intended behaviors. +``` \ No newline at end of file diff --git a/docs/docs/tutorials/sql/output.png b/docs/docs/tutorials/sql/output.png new file mode 100644 index 0000000000000000000000000000000000000000..006d100aad6a03be8929fe5af5806217e83f79e7 GIT binary patch literal 20460 zcmcG$byOT}m-ZWZ5(q&O91| z0l(>o{qTG8GWf)Ce2=M)kQv4mGYb>3yaS^EQ{B38m+>}FC|AuM)02;iS<1B zE`?0_B=?Q%6IoK|6R04^Qw~4tXVxgozTwzsOjLXj96UIva3l&7L~x1is#_+k|BZ~5 zF>Z>3hTu{df76%8<<710u#Mwp6SVL>Fu}WQYz&eDO*5QR+>%)E|CHODoR!t+`u#JIa%j(~Ya-uN>=3AhaU$q(Ds(qC+Qm(-+Z!T8^phh1@O=bG= zz23-Pl`g3SHe0^VP<2IAM~J)78S_mbR+lkK#1MqvcI8FGkN;Zs#c;O&m^3h7<+Fbl zN?w#886?Ns$o)G#q{&D~u@lV~%l72Xl9qKa>8w(hYw0AC(N5jPZ>BQ@1DR5PcX0Hr zLVAIrhdo*1k41DpN&O60B02)3$_53D7aJY-HA9mUT(Qz4rDlQc9OtY283aggWPd-y zNmyXdoJFlNsLNHe+GQWa6kPmNmf>$Lh)YlM+M%RdYP9f3DNoOKTl_PPDKcdP%@BKJ zfg?*ow|y+E5LEo8*%f;*2)p&wVMbZ>4XkGCEtTLoSDdqZ#ZUDVWXd}+U5$c6A62tl z!TYUk@8?JxgQ9hLi9clB{R4 z_n{R`-SUO#i6E^)k4W&)eQUr>_Vl@fD8Ke+t0&>=o2#pt%7wi#RLrhtTLnsm=DmSXdgIkI8zi#x6XqSqogW&6I`ZZCkyD2u$=LAGAM7i;Ls)do~mn?qcxYm3rJ> zjAToN6LGU~ak-8cC>BD(dl z_bBC3hVCR|sqasE8Cas@h`Aj@aGrc*-hP+*m$LnlhwN4Om2s@MjW-lWKtRC8#>Qf% z68N4T5KSp(xzNBF8!DL%qK+N$%{RH(gutNC&vDqANfH=tmj}EyEA2mjBHP*7xn6Db zCm%N54z#!X=0r!01dyGdpZf(|hk0V0&3+R>TK4tIE>N@8Ef`7rx}z8t7KY73jZ^Qq z52<&wC}>hHRNlWI&kNf*JZyH{pP4Mtrlg?A$;xu-mVZj)8+{@FRlz+~ybt>(%2ySU zqBHEiLNX4yQYqk)8O%#tC?Ut}o`9<-vzj@s{f=z%xUI5Ud}t5EIxKO#IosYI%~dH< zjoc}AyVy%)HB;Wm#@1B+d4}4aFLKFh>x=b39vcjb0!6>xOZsSNXgIc)ML`w{8!J+? zTWD}ObgBl~7pj!-*{+H2T$NQ;X0=A+z^X~9UQ&o$h;Rab?K99(^TWhg!9Zy=S8`sE z*=PVexMMJtA8*506fG(`+OISByQ?KLgNJqt8hg7)Vekb_5wJo7V1=IU(%xfhc|yiJ zWBDv7CnxK@DkWM9IPt9#u;R+!z!$fWzWygAW2?EEe{q)fe*^$8W?zqchwo0L& zjgK#7CXOXUA~-KRU?w#HgQOUgjS`yA<91I2vrB?EhgK>$#q_M$qB?sDp8R+WuAQ8uXOn2m`4h=%%qb&Lr*cppN&$#yHiA> zAKok0ns9Kvs_&c9RP#H!8`#3a#DqrIvc{5)$P(yloyfUp8S@Qk>^h%<%}6P|Qaoue z%Xa!*8e+wpGQInM55?TTTxx}1x4t5=KF=+!F?I;fxE-)ruCeE!Rf0|AslWQO&GPl# zD))8xq&v(&z{T3)+LP4M7*@jC+{7E>6Xiefv{jyqaFD8isZ7AWO|I`%n&JYbM0_%c zV1sy8K8o%noi}r#6_wF$h-&bT-Q?pBYeC|6k%B|bQ1@#_dd@H7*%|4=NuIS&^j;@I zvR^n=#r3ogFH?aH48QC9UnZ6TZV*_l9C=^970%$i0IaKQ0JAu^|SDXh6Z|#nutSfmSOrUbGm<4 z7ynuUw(Vb3zF6NC|9Rp3?;qf-Z(%`)8JJ)x#NjR0($Z2h$N9f{X-QwJ{-^viRTebS z-CNMJ%ggGD3asD`Q$(G9choy}OZ_?|a!PCAPneA^r=khy$jE7_qN1YDkJAz zEN^co)>4a&8uZ4&l?ri3C6Fntv&6!QtE#G!0y;Z8xzx>>e6eakR6aRJwAxJ#1Gvek z%YQAL9n8T?$AeQ-9h}}^Zj1JS}==# zMCPx}SAbHmC)SL@!L>);uKc>#J_ddpmH4hI`|J$Wz;ofasPqGri;m}mM0$Xzp(WK> zht7kmF7!Tcu7BZ!K7In&aS9o&DDVJCnvFgRT zd)#YS9m{J}ly4EQ3J^z;8XA(r&Cp4go3k&yeoISfd3oY5AN{Ob#WGCy>&|HgVb?x6 zhx7G~=eG_`O?<$m2j?>hy$roL%H#4}DXn*U%*f7_y>D9~< z?@5o_tp1e8>0zwJr(Wy&dS`o^-DJ-8_tLOrHdFnoeuE$amrqGhu(fY`l}f~H@~+f# z{{mE;ojdsc9T8Te@p1uE_IxsnrOnxqnhIX^9$N``*zN$D%~EEvT1+sCOF>R9ll)LC zO|N}CeuK8_7*iteGB78Ro=%!{UhP?3EPBqh=-!A>pX9s6*t^@l3P0RC>WK{w#vLqB zhEK%xyXDFzI&GiS#unlkIYGaaiFlp&@1?Ez%Hd#QV&maWHmJR$q{MUOFg#lmMN95A zu%#~g1fuHbYj0->=+6e_H#I$#YHqJD^Y-?Z%!V}DM*I0?myV}B{fI%baJjTR-x-?p zdUx{NbSm2pBPnf7yV>S?uU4DeVK2@l!0o~!^Tc{9;tPv~_K40_#9%xvjM@Gu!Pocd zGF*{j{ZjA^aJi}7+{BQOayz7qoQxL{uIDpgkMpylPCi@%R*~59jaded6#mfA(6`0t z1VlvHYU#Gw)JfYY>olo64m_tD`T6-{Lqtc-%SoIm#M?-xD;*VttUGI~Hn`7#-8fd5 zOzrf1qVcvWmBU8AcLFgn!3Lf~7!SvJ#IRf4QOFQ0bX0Wq5ooJIV$w<-4yP$jQC_}$ z)S{7>lVkYuh4gf!y7cm441mp<{%&!7JNc6EEPH=Y6g{T#yn znupS8vfQAjtbyBWnI+roqNoGK?}vf?T2qCfU=B-XSC?YE+(tGyKBv4|YBR-R*8XPj zZBn9<3zWXta{3&M!)cG)l8b_Tmdl5)Gc!Hyu;QCI=5aG}S<6UysX~W0AuS=H{5eLD zsXMWvxuIgl2kFssaEdFP{p0uG;ri}miQ~bHB&EmoNvv^XWO`(HI6GR<8c&EscX}$jrAGQQIMhu9 zay=Q%?-vk2#Ortzcqv4Ekdj-?ujd!bpk4bl+e}`5sVo(JzS%3+AlUJ64{j9x5Ko`F zSmE@J2$h%h{ADN+dqUlhavoWyOOz0myouJ_49i#(#SJ2k%H05qclnx|Yil_vd*oD7->3Me@wX=h6z+LNc@vbO~HbF8a}D~9=K-72@6nSAprgv#Z?tZ0`I+E@?{ zZ{BVEjdul)?9PBeLu(_*e&#&2Z8V_<e`@SEzh~rsH-n#1nGHh+ z=OUxkX4zn3s3cuWyvUT;{@qwcl)0abYQ?Xd-!xz>`?h}5(dqsW{{L+r|H-Dfd<*50 zG5PRT3V$t1m3D%iy5R(MdCbIc;v_Bp6{;=en!}I47p<@STFZ(&b&I%^LJ!Y2B<)qp z?Oq5etG;5^ID@Oq)!rhh#Lz>KDUW{7 z&c2K17W=nO{9j&&Pl@i$zp{Y;nfeGkf^#m36)M9&VZM_23kgYGO)XG(dovdf-*;N{ z=(5Lad+C@aX}{SiwQA3W+rahzw7%c(@=4!XEv*^f+(1V(G8|5_eXK33S_rMZKg z(U=(08KseDw1EcpP7awm7p7L)VwHUrY4Vj!l5?83*mJR;(^Ve!;&~WxY!gV@YN0KU zNydl!t4-~2*1TgJ67*)kT;K^c8EK>593N6IjMd@~WM>*ws(-kPs`7TpM_?>>;x&DASnxRAV9+3KoIJgj}TfOMu|^1zqH@1{s*-3UoDJOxL*+w zUwJcd)3@4p4ahj{&Gl4v1!{+}Swb|kL0>EY&Y*B|?y*qe#QjoD0D&7cGMXDsp#OY% z3EKxC_ZIn649M>NNDe=;&Z(|p6%5}BcboPuO3oHchi9LcwEP_q8u;wGT%R5 zA06;6Hha8NXwt}jwQ6){^=-1&E{@GI##3p_Zc~nkr)D=cwIn5lkM}mBB{NoXy=p(X z>TpU>Dj$W*b*#NO)xgchsUXdAy6$4d%E)*#C&v)&9m_kmGrL6TxVBv~tA#7LDneXU zcyOw~z)k02)JU$%QgiqItV}W%gpibR?Pg=EV)1He-9D?(6&dsia1;@30*WA?-pmhP zw`v%-7ZWFEt=gz^YIq{qT+e#o<;R(}|ENxsgV))?W)6biTLTa^95~q!Gp<2ptub$;{cHf$eor$%K_^b*8&tkKKkWrejL8yzP!J7G>D)^&qO~O=R{dsh{Dw z6%iEg_h4+6Q))ekN^p`Vh5R(D*?=%Ltgo+6jTpvv&uqc!x+d(3A||$(w)8c1Q_fEW zP%o!|WVHW{pH#)L&1nC7Qz~wNfXf-R14ODK|6N8}+HRq?q$IpBSUq$HVUmR67?knm zc25gWp>TY@{<2WBkX{A@pYaDZGT1lnv(EYGu~6myo?37yYN+vW-?sMI&gcUm1d`az z`k8{Z9`5}|tlGxMLsPzp#EztU+|<9s39*j=CVG%&p zfqmUHNE|t6C6I&D+Xk~^_up!iO+w+ucY@#*7TK4wXy;jJ%vS|%!YSRt;k9U5f;K#j5C&uwtXsD=% z{52md>-nC+V7PJQEExHTl+vFlC_tp5h#?~GYJDw!9i$mxc~**RL+y}k8zd2MBN zofaCM@Tj{qZ$o*~nvW$d-M_NBx{xY2E@&FAKpAwzNOinw<@I&bW}}~q zt@ST*-=HLVm|K;_DAnmrjtebxcfvya5fc;B^3Ky>{w7u%ey7`=F+SZ`{C>zyK3g&+ znxe0VC%P`*wi(G)ER!9%0G|dMC2shq!Hefaf?Vm9>eF3v#PV|fM@MJdSbAL;>D6~@ zF>8JMygPCFk1-f723Ch#_h*rYv%M9px2qiQbIz>Bc7daiM z6c#b07J~`gv67A*;YVyCO9|t{sV>>eo3kn8Cz0)BZ z(N}ct$KxX{Q|pDsTfF_d3s^Fq7x(V68ePC-Y+G@jItMAK=XLz)ws_u;I=#&aZ=ZU8 zmy?Kf!uyeo3^Fg?Yb>JcVLc?EEz24v+M&U$b3c7sfZ+13FEQ~rg=_++ zxiiU5NVW*t4tx5>^n%Ci=Q?moYI2!qOEr9`1(u2;_Ut$DLwzn*?9pXChI zFIp@$UHFkdrke#Sc7(eMF@^EhPsTnc9MM47j&AguZK3^eje^3Y=x%hC0FAAxrDYO+ zaL;C;B@_TIG@Rd7#Cv~Za|yiJ?xYpj8RaJ4T$lijMZ!tLB#VlK)&Zr;(3Wt}Pvprl zjC{uT>^fVQ>X-F3OaAm~Q;4nWQgV{3*BOx4*FHN?&HCyL2-!GJew0LZP`%p=BO{fU zB(KLIzu^QGo&3($l)WNApF#Y^eJ;f0^##dXtoKxySF5<- z_e7c#9=Fd&0g1adu@+d7@5J6KGNx~E=<)J{+)VXjk*BLae`SX_AKhHey zDI`z-e`68@C!PhXxd`RC?ta@)NpA_8vP0ej5WSMRrHs~MDZSuygb_J%W& z{zd58tahTKqbCJiTwDNT*RT#{`LCw4y=g!;f2Y7k5h>AXOh5;K9s`|vRhB1&P`I?B zVsu?MI=x!isG}>7prp=W``yt^a?1Qe*Qo$C+NYIg^fQbD`RXb6IBuKOze~S=O2mxK zkrVvL=lU{=I#jxMoXPOdensl!73{hnEnnMIn}~q6P`bT?-fg*NCB{}No7pKpc_XW7 zb9y#=w1i!Pu8evz-mj>`^rCQbIBK1J%)(v}oJW1-fElIsrl7hqBoq|TAgg-~Xx>*X zp0CKtehRciNOrW8V5)fvaX1K?XZ7?5gbTEkR;qN07OHx-$p^`bONxt33brU@XHj%y zm)`;E9`FeN)AjzZCgML10N7XM*J@&T^Q|ZE=ojD~%DsM_1LJXk3V43yAm!)vxJeM; z6G+A?$Hm31vprv#t$t)-VcEG+JFKahl8PU+dR%@(7Zl=N@432D?_?z+B2rfuN6e)! z4ah2%W}DHo)ZS)|Qp@|E)h_yt(o?rfORIZN7$8p|aC)VJz6gO#UY>%HyO@|5i&=IU zpB)(+TW3Ff?q{4_kEG!5ccfR5L7U%;gln3Gjr3^u@|NG$LzL=P(x7h&E>n98iiAUILRb276Kd)kbHh{R8-V1 zPrld<_PoU`$Nd|i7fMSkD=P!XU2H9s81y<`kQ3$gWsNC}undZHSi!i?MaUF@CTUv- zs8iJ7JX#)UNlDk!8=QXgiAvVSUKW$N)`f*gEvr>gI>1P5;rMP!r|(S`85gHq*1D)9 z`TV($>9}d*iBvNsNwy$Gw2Lbk3s=DgQVYT%Avv5ogfcBHEmb;i@1gJ4E0<{1SE{SY z4W2bBm_VUWje*7=rG&h_0QE7gDgd;p;Y{Dyox+_*lJmnIm6BPZN+DcrZCN1{lbD1= z?(pr|P|Jcuf|apx!oyHYQ`Ow5(@ zv$MbLYC*6lXMXPEA*bn`oDMz-i5Oj59K4YwIIrK((z3-?hAgt^+I1Hw$#PJRa)Vun znOU;HbI~f}VK$1)sC@a)6?4>fALY*mWq7u~ZEgBWw6Qv!W7D4#9DDqNwvSnJLpc`3o$2Z=bkrK z+@0)}d3pq}SS^SqV8tsyPo7u+j=V$X&>-akhF`!KVS~f2lg`Y1<1;0`ZK$KYJ?=yI zI*L@USIN)YmfjdelEGLSJUcxfO_Mp_bO4Y;0by2> zVA^c8C~A!EFX>#6)1>%q<9mvqR&b8i*OIfDb1A5X2@$ZyV=5~c6lXjx4-WH*wl*(q z0o-i(2-YZ6f=j^L$)z^uYVBwkKQ96OcborZjut_jg2|_tVk5qVG3Q>(&3}-|XVDjT zH|K@0`YI4dexZMoA6(sGs&VsnG*`}<-=hI&QVCh%s@=>1t64HLh+RS6>aj%W#Ac7p z4WQqR^WaQfCp1Xb8pt2zzeQ?rdAK8R_&8ah6kqVl>S6mQwP4JnQqt0;Wo3XWLk`tJRzKh2{I>Y84_Mg?h||2t zbcHy#oP*T=2V$C7$;uXM)2a$+3RnJoi4U~1o&AgQB_YAV@6K07W`zg3I8EFCa?R*h zpYRO`P#B#ol=W%x5v^!QU~)N7p}|EboP%lgrg?wttyOWqS~^&x`gpm2orpsBF)?+z* zmj8NZ!s7RmkUV_+6^R1;UU*C;W}b{$(QYT!paZ!8ktSnqv4`(Y_}}yUa%CfS0~JS) z=I4KqMN7rJynfv6Ad#lTPm~b4$Hd1DiI{x#lHyu>fE3Er(j$POZT*?URH-0V=)o(7 z1SrNi+Y?q=R_v?94c^zFY~v73h}p z^fI&jSTdaPwJA8jdMD1PS`W~K1cVWvVr^iu2nczk^}ITjA(@9_ppnq1Eo-P+%U2Eu zupl(arbs544Y2n5?{Vl~r=~PzrJ0f7OAafzzQpV-%ZkCzZc%{)}oEsaW-C3_BDSE~8eSC!e2)4hmCZIi)KVv5aU=E}Odx=K@$LVvvtBEyOO zTkXx)Iq3bZ_U!j(l<4CtLUb$iy22#0qqeDiv91J{?d|QeUQr90ByBSSHt9#~ls|Es z$w?zFR3WO4F1VeQHeX1z_#c^-5T}?A#tdZ1>{Sj{n z-IK~!t@-C!wq<^3t=-&MQieF0yupFLbOJaB+77)c+gP z+ba7Bp1Vit<=%Xl00khwU(2QP86IhW=5!c7PQ_6CGG30D^bKsgztaNv=6!1ib8AK} zz@K4kf9z5~uq7}%*~-hOG5(DR4*_51{8l@jZGQ)YWcYG8^EQbnh1<)t{kJE4c4Bt~ z(J|vdgN23VcMaI4+-IvPCd{2x#!3KyiQ9nkI@wMg@l_(Z1|?QdULiH4C(c$`K`z$s zmXy zDKk-i{pwYY2(UvATrX`hB#1Ix1Lls&@POvxFa-wj-1uI5*Cti}E95hy7cZB`IKxwU zG!U+V@hWX}$rqKD3A>1*QsTofZ@@lD8V6cy5vh)Ky;0)+tH{n-~xgTbJ_(Vn5W1aoBiGhK~+0pdHKA+ToDmYW2+LQD3 zULN=TaDF@AA#Z)U)}Vxr#1P%Bb#){EJDyH$v`G_>dp62w;2~o=VxqE= zR>w@Q3k9_`-%d95VY;u+{p@#OSl9p%_kwFT2O18`bzvw-8Vy4Bc6LM~6FBdKe|>#i zq=#X}QBY8X5OONW$%X4q6q%S&+xiI_4mMd36DtLv6XFZN2nTxB7B zG&ns#JFlE4pJ==C?Xc3OZQ${oL{v_5piyvg(3RioVQ+sHc_v-3I4zdy6vc041HMRw z8BN(=49WJ7j_yquj)uc604NfFx|^{!v`fgT%d_7g3%h}FIU3bQw6)}Z!F_XNcvJbJ z@<)EYTA?x#y#@y?>g3W-bZc@X!$n;|KnsEgTwLq|z?GHygt@G9?;6S)ocGP1LA?cl z1kT157J6ptBVAqU&Z;8=0(P62A2ZK_?>pLs6{*DfJ9frR%^X*@6+%6moR6iQ#mmNM zflysm!A~MiX6FeQdN{tS0 z@95w=J@wsVgM#?ni{xRqXS@DRRAJ9%WUwx z?`dsi3TRX?l#F-8Ku6DtOJNK^X^a^9rk6=eZwrG)nVTG2rF$H>yQ=Km#pZ_i^WKD9GE+VUw`E;by`mf7w8 zYMD>vaY;Buce~i~+CI4kd=Z5xlQuwJh5%M`Q2ga2?WWf8M*q99tgKS_gKK`Ri|bxv zBZ-17Dq3i3Tid~Wy-aX{5?pNtPP@N308{rCcuJ7T&=`Ji&$!uUJX49Yud{y>JJv0H zzPZ}vHAYLVnDhSrZhcaZHnZDfZ>lAn!#QsFa<*Kg-s9G3tZ*`6X0YfxpkliIKt);; zU%lT$DPTt#Xarx3AT0ZAw6oK3yhrBk+hzM7Vff>R;ru24z=5_4IzBfth67hW99fOU z!>JU-a`QmOOYOgYE#0%;o)N4qtA+X-b%qcb8ybFb=$QlV55PUz>KPA~Nv_PmOvb%; zJ;8TSrqjNF2N%vZyii0!-K(>kobbH&KigVh*lf#FXuQaa#G=>aVz;$_B*8c&;xz?8 zE#MOfy$sLsH>gKPYld8R@YK|Mwo#j|a&p%*{eC+R1Q(~hmX?aseY#2oy@C63 zbz9A%v+~6)#=%_0C6ej(@b85qomJr(>r{v-0mw-D-ykK8ICwDEa!uhQC8+*84hk4> zP_+sXGc;6Gfj4dyKoJSFGyZH#Rx%EWfu^xy6C@0L{jXsfLW7V-0hEuH+W6d1Cs*I) znunP0s(4+PmQ#X=D6a3j)$wq_)=&LJCXB;OCP0%1|Mz#RW+K#r(Fq>86NHRup2EFpgF=17?n6QE#k zj52CVjew9vV%zeo=z^leKVd3fCW(K%?={}jON@zVtRdM24X4Sfzs8ax=>6!c$ZbAb zwGYH=7d>z3UW_a*E{={e_l|?cX<0r{$-vLOMs?c!;p)}06uqaf?x{8gV7EKtG_Hz6 zqodKi>p+MlXWW!khE}!o`#RB!`4MIMNRPE-RDaObIS?4eqLc;1;CO56kDW5T&Ma-B zOkw{$wzpCV2qBxR&P;U?DRsO?jD_t_UZKz-Ga$@My1cyCJek()!mLuQ(BoumXS+F^ zDF%=vKsj`k*+n3}pCEwzLZaiff)j##IvvYC?2?-y7 zwD#DZoVa*e96u|&>arlBN*>1VmBOW&E1h_9b8be>s9tHDUYk#Rv^oIe2HTI4NHGVv zKHca`TwPukSXC`5OsY4ILoJVk7``j_EsZSRp_R;c#nHkQgy)s->lcN3XGT zB$zNfJk0W5GIWHY=_(`bHd?`UgLzqYZ?@XXYO#s;p47y(D~u=@mrjuBGz)na%E`>e z_HeUXl7E*a5TELK6NvLE270{#LsugUMHfy*l-gI$39V8pm|yqB zljXZl5)u+X+#H9mNvIK-HnX;MDH>;-!$@ns!6lk22#Dc-`S$Bc+@d`E9-oho&+Q^> z5qo}eGCqnbH*y~e%QaeJ{A06BHZ5AmT9*+z@nKcil$#KZmg1g>H&u!ZdSwj zfu+x?+x~NCNd7{>z78UF*x&s4;MVARz7<~sTqPhZ527Xb`fv>_Zj+*HAh$yDWg;V( zoYLVa4@|w|lwB8aM1E;HUf^*)t|CsspVzPEn+oq217CBwoah63`to4@lxreGh@A2T zFe3ocy9Xu|0KpQc|LqsKCBORZa0?SeJa~9v^UK#t0wXEu8QX&7L&*Uv;LJ)A$jpaXVmSwCc+umGVAuqOb17UUz4GrM)8`lA;ckvuYDmIIO%VE{t z>P1hf`3QU4hr823&LXvqW{ z)XvjacYH&m()g{9u7J}Fl=Evcma{J#rBE`~)&Kqn&C6vdi^&LZ%1s16Nj4Kl9Yqr_ z)HwuuPh_xBj=Dg!;Y2LP5wWqc)+U(!avI*sv+4caK>{2!`oG=L1QZoV`8;kIhYXGF z%2$R-&zc|a2S-Lgr$Y-1O(vr`sFegW&?0Bw=o@cp<)Y8h@9^={rf6R>4s-|g%lG{9 ze(pExU0YYjX0-tIj;pud#+-q6e1d{)0ZKp}+q2N<#ysQ+z&f$y)92NC;u4a#yLYp1 zn%k$0Sy$C1B~O4Y1zuD{uqaJupc|N{Pt_RY5K3^TlFVkI&qM%=e*5WP=cFak=p`Pj)4wyM(G;`2l`9AK2`)eB+BjS5O2@b! zs(A8JaLLOP-1&@b9Ud}BBfWSrEQf^^?RkIm1r}q$ai2=L!0~p$$dKygb$`30D8JMc z%|(g(yjj6VA20dSPmGL=r|;R>tNqd5F~MRSl9`qIi{GPJ%yfkbc$Ybj6fPNfUK`{4 zcmb0eKGbD@r@?^<3!^&fUF#-KLWap`@vJ->9NK!&jod?s;SmN3-!&kbk~kc#OPRG?!EYdM_6=@Hn43wm6&#Cv2)3 zKz?@xk4iudbaHaqDW%h_vj+f90{ZAVhR&~CNxkDQJ2l2if<=(nD1Ijit4U|;b~CkG z3SO-zJ-G>oG_ze6o^&SEUhM`&QF;e zYMp9`As&4hSL`L%?oY{6i{*7G84;MBC14~~5u!uK2jmTrn(y%+&KyIOsO;S2p69?& z>CU!DWP}6SNwn~mG-R7i1~EgVIvxiWA{fRBDlJ}XH-^T<74Lr2&4hRN^-W1OgDq|o zoYyC&o4}k#cLE&CZJ4Vdb{RJvm4(i+fe0LioT{6F@%<_~cLKbQ`R};Ea0W1)!gq!6 zx;7lJ^;wxlj-t*wK!c%|V1%Ho?z%AodE;MU9Sl;gQ{^iIEu)s3D=HzBzz18)?H8aQ z-?v{YWFCX^HL!iRdA{sx&Rh4Eh=dYE*qvx8k1wt`Hp1O3!o>Fh8UY7#c#<_q%$mCE zhMxP*9IMY;U#ZnQarTUK3zJR4nT0>$PRFn{yTk>zR{ZihpLeZ%IJiBh!f}{v>R|(9qVy1d{Zb z>~kUxD*$tkPINSmukeSuzEJCI-F`^HLAbwDmX(m0uJRb&W~Kx*=-gZnItf>mZb!3~ zDj9BY_2HE5Uan7No@6ZHXzlzfxVE(693+2ZLG=>wi;njTG>#WeyQ<3!ckl0fDK z$<+4F3Y*CaZBp9vhx2wE9YjRL5^>`Uy;@}*KoZW5wks&rtBcP)Z&b4-^C8(AKxWt6 ze5C7UgCT(eGw?kT4CDPAtHVhHg9`1|sGIyEpN}srZ+vI44zV{@Hv9b%&1i=9H-kh0 z?JwQai^kDr&3UZ%JB9LFD*0m>4i%BMwNV>`+7ksPZQhXzEtP_{whh72M%7`=p`oMA zZbUdlu18zCB#Y{j**Y*nO;v%Y-QH)~I366Q@U^*8n{yyn7)Ho>ygRNg8E2osrCN5IFxlP6Or8Xo7d|xXmv9!qOW(H>;LQty68b&=Dq+)*{03-K*PX+&Dsf{IGj>>V7NvW{Lfoa}s{fUrf3}27f{>kScQ{ z?Gf*uzBsy>{ptEp-1#QE(aq~oAd9CA2W8r>Q%6(uJC-`3yna2-L$t3med2=EIhd!p z`y4N!)^pWX+8mX)K3P4{_5deG21Y`3zg&X+V1M@pC;Uu#WoKtYtLhHO&JDe5w~8Mq zJ1fCucAZRvx;cYczkIQn&52H3?4?`?_eEpSYW7R6+_q=V9!^nj6~2 zYfY0kGHL1F_rxBX&g!0NJIR4esAv&Yw#90QMqx5m)-rnax&gDtE*54c>h{f!{?7B; zct^PG@cx`-IY{9_lMtr=Q1e{!|M$%wM(gAYpmK?_dd2UyOhVY@@$oSLe&c%Au~BtAO{&XJH|Zw9WGj&(2ql6%t{`V-<_o%2RZ;59h@e{Nolgwdq;>}MW z;#*QT+s4{j33662^hasDzXGL;jO6hbPqXlVGo3U)a_8a{5-9PAMvdJ=8xIV&81NA; zM}IpecnkZA=<+IlN2=k*w_?GuFO%PQLDR5Uq$c__N0h49@5=vwHwE(KKCu zs&!IAQkmCDoK!0npVG6Kq!}s1US2J{NEvv6fRD?`vfVm_sH~xJ3E8>_=eG;Su*ppJ{yR2+1wCLIe zOpK8QQ7CJBG2Jm>1Gq}P7~GVIX^@@e_jabZt;sSQToBGtFa$DU7OQRA=*yreYk))UTNnKUd@EY5`&!LeJ(DOlr*>u4ou(+5~#jJwUO;gRowBp;S zsspQ!&*73BuOIvRB|-cVO`ft&VN^Pi3&5PXD+Ld@P-)QPQTK%?T(W{T|C$$bWcrc7YaEd#9r@p<vF83U2XJ_|cXz!2=q?rx#MbQ0wohDQ z$aNGFLo64xJ$XAvl5-efsk8YQI@&v=?&+`KQElJ2& zZv+TjELQG(8!2AZBGpnG<#ylB@*l%_TG~5iyj~`5;#)M2PY!tlHsG^dv|qmDn#R{A z!lQ-AXWC~soo3dShRH!DPpe*SBW7!{|CEhAVm|Ztp}gjEcWAH)dU{JKJTJwOgXh&a zgsi&H9O}#$%KcHk+S$D>;X-?fLDweCBHPGOPCy~+7@shUfZg*a0#(0haXz4?(aGOCq_5Vy_hlRUxJwt~gnlnL$Q+|~UKv^M(UfG;Jv zc**I#`eniXH+5#S1HR8rt}5c7f+mGm`!zLMd_lqD`hjH@NO+x%pX{1meBf)kln@jW zD%Pw&8xQi+>cs_vsU$#Z6Bbn|nj1>A@mB0=Y06vI zz1-y*Szvq$3WIs`gLwy?>vAp;vTvzntcf7;GBSgwr2^!c5fPkoiDUO%2)%$y-=KNPhND5~kL2#pSd~IfTEa7k^E;gR2wW@HiOsGou zvz;tokXljFMb`AD`d1V5XHDMHU+i+3hP=S&cWj6rpu-bmeKOP-Xl^cVOax1{ zbydHDG-@QtvBdna?{Rj^^-bwi$a4u3#Od}7l+XL1J+b;TK*%b56L<*5!!ZQ9w8;Y`883gL{ZIEU8!VuCs{(YjG3ySL)wAS8Q;~x|j z>`_?mFdY5ce&wZGgo;|g=!^nsbg=sY_L2eQJCMU@uXuY%(jSl!WzkkX{%F47+~xYH z2Ia3*svJ0cRkUTWqCuQG6Hv51ratSLuK_jEP=R$>~t+8OfP=?R_9<^Yl6sQFHq^v9+Ui^DydjDnY z2A5NR&lhLGTQdlndt4wThW~Z=8h~J1=I%FVCWEM;?SOi>APWvw^!FTJo3XjMxyKyR z#qX|G!|h)}AdnG)sZyP`IV;t>HmbCZ&-Xv_>cyQGF$1leA)i%&M`PSF*>4Sb%W|ry zsKklYVah%ocko*TrUd7=FW0xWem^;xP4#+^*i5Ldb~-RCRU3g@nIpUgU9nu)!8e)Tc7sv1N0G-OmaDE(H4kS5PZGJdnq+e&&*um zp5J2lYS|rc+<_{?)6+9=xApdikB=YwG4N;&1*P8p{!kozUuH8Pa*xZ`?OjKY>`S

dB z@eWr8KSpdes{L1S@t;Q%b2&2aZtC%j(f&^<=N`}W{>Sm;v}}~~iygUNDrpyLGbais z(F#XdLrt#bwvjr~v9KLcxhBP&SB!9|;;_GX(kYo63TA9I(OSET5w~{VhjGkT=9s(gD zO-+dr&qKg!AM|Hb&i+`2PPAO zYPg3CiaP^r7Sc#u$S`AWtEfF7t=hPD8wl?+up zURzeyw%AQ%+A~qDSp{ngZ@j+XMLbCQo;cnn5Yod!drC@5!p! z?tY`=AotFxzYcoVYw)F_jh}&1vhnzNV3mR z0*Jxuw0Tei^dW5#p$>11HB`nwiJl9ZZ(D>L4-}^}7`arV6#KW8l%GL(nzsE~N?34gy@TCRB;K-6Rn1^Uxtyq_&DPQ_ zhFyaT!;5OqEic;L!Y(!vf(pZNgD5cgv)g55miV5$Fbhi9+X`;>vjGji#(`+aIW9{@aU|mg9hPFd%z-|Dw|82#eOIFC5Ii zxR(NXdgW3|=puCc&HN_b<^*y`_vE{$^D9`+r~*H%>->Gn6Jx1J?nDzVqpDRIzvE#) zYZ$A zL?2lFB#PEFqm2~fXRc~%^pHg&AKj?LuCDtVyiM?Ie`VxjfbN72RX-7~S9h&~cWGef zm(x6dAuTg-LJ)Rg?CqXcAHGiZKXFN~&UvX+6+Mx^f4!w!K5VC3sqC0_(0KVj@s2DD zi}~IT)*Ykz-)1HG_-B(sd*_X#ePvDl!(REzc0>x5YcNmXG-t`y2Zg zp7R-Dpf`lu=i%x)fTKiE0}zYqvLV-<;^_5}Mp z{*F8o^ssl`7^$g)3p{Y}`zv|qp)wyVAd5lu5~k@{f7u+3E_Usx{gxDr+lgl~q_vlH%60l9rGZfK?v=|0-Oo#N@(6uPOD*WHbHC9F-jUacM=@P0jdB{+vB zE~92fZ`Ra&3xgv*4zFp?CK^-+FCXqOKq56&R8EtP!9|cdn2Hn9PK>k566~!JO2FU% zrc}iZS%2^*oVII@t_+^s%8O+BcaY-a<2|2VEBAZ#RS9#X^2wR0t~z91;A&j9@2y<_ zI~GT8-acD)$>L5(**zTP!$UZ-4lr#yTwD)TthaY{O}-v+H{~EodLN|~ z_}8=0115!G^QQjKr2ZYh!6uZ2+TP21F`q*|)@kQEvECH;V3~NgyFb*p_L;bkB*5`T z=GiT_cK9tz^`14fyyRpGfHQWkstF2vCBk~_m9I-;rev7fJ_ZJ2M-GNhFENWLmI${@ z4Ar>$`e-~Ejfd5f9ORR_yPb#7d5x`$@s3s3$!Jw~469urNRs>Oe^LYJXJ~a}lXk!0 zs*jOuitL3|gY4%^6$kJr2ZXNyVdoeL6jB7Pjfwl>l6U!WH6YTLz0hVoHTGzt@l^h! zP-1YWPh7Ie$V>BcmKs{>N$q-gEB^V->}K3-P*b8wl;7<`bKTP%J|w}`fU?)JJZ>g5 z7;aTgRMlo@WTbNYRQQtt9f`QOUn literal 0 HcmV?d00001 diff --git a/docs/docs/tutorials/sql/sql-agent.md b/docs/docs/tutorials/sql/sql-agent.md new file mode 100644 index 000000000..84782d371 --- /dev/null +++ b/docs/docs/tutorials/sql/sql-agent.md @@ -0,0 +1,549 @@ +# Build a SQL agent + +In this tutorial, we will walk through how to build an agent that can answer questions about a SQL database. + +At a high level, the agent will: + +1. Fetch the available tables from the database +2. Decide which tables are relevant to the question +3. Fetch the schemas for the relevant tables +4. Generate a query based on the question and information from the schemas +5. Double-check the query for common mistakes using an LLM +6. Execute the query and return the results +7. Correct mistakes surfaced by the database engine until the query is successful +8. Formulate a response based on the results + +!!! warning "Security note" + Building Q&A systems of SQL databases requires executing model-generated SQL queries. There are inherent risks in doing this. Make sure that your database connection permissions are always scoped as narrowly as possible for your agent's needs. This will mitigate though not eliminate the risks of building a model-driven system. + +## 1. Setup + +Let's first install some dependencies. This tutorial uses SQL database and tool abstractions from [langchain-community](https://python.langchain.com/docs/concepts/architecture/#langchain-community). We will also require a LangChain [chat model](https://python.langchain.com/docs/concepts/chat_models/). + +```python +%%capture --no-stderr +%pip install -U langgraph langchain_community "langchain[openai]" +``` + +!!! tip + Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. [LangSmith](https://docs.smith.langchain.com) lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph. + +### Select a LLM + +First we [initialize our LLM](https://python.langchain.com/docs/how_to/chat_models_universal_init/). Any model supporting [tool-calling](https://python.langchain.com/docs/integrations/chat/#featured-providers) should work. We use OpenAI below. + +```python +from langchain.chat_models import init_chat_model + +llm = init_chat_model("openai:gpt-4.1") +``` + +### Configure the database + +We will be creating a SQLite database for this tutorial. SQLite is a lightweight database that is easy to set up and use. We will be loading the `chinook` database, which is a sample database that represents a digital media store. +Find more information about the database [here](https://www.sqlitetutorial.net/sqlite-sample-database/). + +For convenience, we have hosted the database (`Chinook.db`) on a public GCS bucket. + +```python +import requests + +url = "https://storage.googleapis.com/benchmarks-artifacts/chinook/Chinook.db" + +response = requests.get(url) + +if response.status_code == 200: + # Open a local file in binary write mode + with open("Chinook.db", "wb") as file: + # Write the content of the response (the file) to the local file + file.write(response.content) + print("File downloaded and saved as Chinook.db") +else: + print(f"Failed to download the file. Status code: {response.status_code}") +``` + +We will use a handy SQL database wrapper available in the `langchain_community` package to interact with the database. The wrapper provides a simple interface to execute SQL queries and fetch results: + +```python +from langchain_community.utilities import SQLDatabase + +db = SQLDatabase.from_uri("sqlite:///Chinook.db") + +print(f"Dialect: {db.dialect}") +print(f"Available tables: {db.get_usable_table_names()}") +print(f'Sample output: {db.run("SELECT * FROM Artist LIMIT 5;")}') +``` + +**Output:** +``` +Dialect: sqlite +Available tables: ['Album', 'Artist', 'Customer', 'Employee', 'Genre', 'Invoice', 'InvoiceLine', 'MediaType', 'Playlist', 'PlaylistTrack', 'Track'] +Sample output: [(1, 'AC/DC'), (2, 'Accept'), (3, 'Aerosmith'), (4, 'Alanis Morissette'), (5, 'Alice In Chains')] +``` + +### Tools for database interactions + +`langchain-community` implements some built-in tools for interacting with our `SQLDatabase`, including tools for listing tables, reading table schemas, and checking and running queries: + +```python +from langchain_community.agent_toolkits import SQLDatabaseToolkit + +toolkit = SQLDatabaseToolkit(db=db, llm=llm) + +tools = toolkit.get_tools() + +for tool in tools: + print(f"{tool.name}: {tool.description}\n") +``` + +**Output:** +``` +sql_db_query: Input to this tool is a detailed and correct SQL query, output is a result from the database. If the query is not correct, an error message will be returned. If an error is returned, rewrite the query, check the query, and try again. If you encounter an issue with Unknown column 'xxxx' in 'field list', use sql_db_schema to query the correct table fields. + +sql_db_schema: Input to this tool is a comma-separated list of tables, output is the schema and sample rows for those tables. Be sure that the tables actually exist by calling sql_db_list_tables first! Example Input: table1, table2, table3 + +sql_db_list_tables: Input is an empty string, output is a comma-separated list of tables in the database. + +sql_db_query_checker: Use this tool to double check if your query is correct before executing it. Always use this tool before executing a query with sql_db_query! + +``` + +## 2. Using a prebuilt agent + +Given these tools, we can initialize a pre-built agent in a single line. To customize our agents behavior, we write a descriptive system prompt. + +```python +from langgraph.prebuilt import create_react_agent + +system_prompt = """ +You are an agent designed to interact with a SQL database. +Given an input question, create a syntactically correct {dialect} query to run, +then look at the results of the query and return the answer. Unless the user +specifies a specific number of examples they wish to obtain, always limit your +query to at most {top_k} results. + +You can order the results by a relevant column to return the most interesting +examples in the database. Never query for all the columns from a specific table, +only ask for the relevant columns given the question. + +You MUST double check your query before executing it. If you get an error while +executing a query, rewrite the query and try again. + +DO NOT make any DML statements (INSERT, UPDATE, DELETE, DROP etc.) to the +database. + +To start you should ALWAYS look at the tables in the database to see what you +can query. Do NOT skip this step. + +Then you should query the schema of the most relevant tables. +""".format( + dialect=db.dialect, + top_k=5, +) + +agent = create_react_agent( + llm, + tools, + prompt=system_prompt, +) +``` + +!!! note + This system prompt includes a number of instructions, such as always running specific tools before or after others. In the [next section](#3-customizing-the-agent), we will enforce these behaviors through the graph's structure, providing us a greater degree of control and allowing us to simplify the prompt. + +Let's run this agent on a sample query and observe its behavior: + +```python +question = "Which genre on average has the longest tracks?" + +for step in agent.stream( + {"messages": [{"role": "user", "content": question}]}, + stream_mode="values", +): + step["messages"][-1].pretty_print() +``` + +**Output:** +``` +================================ Human Message ================================= + +Which genre on average has the longest tracks? +================================== Ai Message ================================== +Tool Calls: + sql_db_list_tables (call_d8lCgywSroCgpVl558nmXKwA) + Call ID: call_d8lCgywSroCgpVl558nmXKwA + Args: +================================= Tool Message ================================= +Name: sql_db_list_tables + +Album, Artist, Customer, Employee, Genre, Invoice, InvoiceLine, MediaType, Playlist, PlaylistTrack, Track +================================== Ai Message ================================== +Tool Calls: + sql_db_schema (call_nNf6IIUcwMYLIkE0l6uWkZHe) + Call ID: call_nNf6IIUcwMYLIkE0l6uWkZHe + Args: + table_names: Genre, Track +================================= Tool Message ================================= +Name: sql_db_schema + + +CREATE TABLE "Genre" ( + "GenreId" INTEGER NOT NULL, + "Name" NVARCHAR(120), + PRIMARY KEY ("GenreId") +) + +/* +3 rows from Genre table: +GenreId Name +1 Rock +2 Jazz +3 Metal +*/ + + +CREATE TABLE "Track" ( + "TrackId" INTEGER NOT NULL, + "Name" NVARCHAR(200) NOT NULL, + "AlbumId" INTEGER, + "MediaTypeId" INTEGER NOT NULL, + "GenreId" INTEGER, + "Composer" NVARCHAR(220), + "Milliseconds" INTEGER NOT NULL, + "Bytes" INTEGER, + "UnitPrice" NUMERIC(10, 2) NOT NULL, + PRIMARY KEY ("TrackId"), + FOREIGN KEY("MediaTypeId") REFERENCES "MediaType" ("MediaTypeId"), + FOREIGN KEY("GenreId") REFERENCES "Genre" ("GenreId"), + FOREIGN KEY("AlbumId") REFERENCES "Album" ("AlbumId") +) + +/* +3 rows from Track table: +TrackId Name AlbumId MediaTypeId GenreId Composer Milliseconds Bytes UnitPrice +1 For Those About To Rock (We Salute You) 1 1 1 Angus Young, Malcolm Young, Brian Johnson 343719 11170334 0.99 +2 Balls to the Wall 2 2 1 None 342562 5510424 0.99 +3 Fast As a Shark 3 2 1 F. Baltes, S. Kaufman, U. Dirkscneider & W. Hoffman 230619 3990994 0.99 +*/ +================================== Ai Message ================================== +Tool Calls: + sql_db_query_checker (call_urTRmtiGtTxkwHtscec7Fd2K) + Call ID: call_urTRmtiGtTxkwHtscec7Fd2K + Args: + query: SELECT Genre.Name, AVG(Track.Milliseconds) AS AvgMilliseconds +FROM Track +JOIN Genre ON Track.GenreId = Genre.GenreId +GROUP BY Genre.Name +ORDER BY AvgMilliseconds DESC +LIMIT 1; +================================= Tool Message ================================= +Name: sql_db_query_checker + +\`\`\`sql +SELECT Genre.Name, AVG(Track.Milliseconds) AS AvgMilliseconds +FROM Track +JOIN Genre ON Track.GenreId = Genre.GenreId +GROUP BY Genre.Name +ORDER BY AvgMilliseconds DESC +LIMIT 1; +\`\`\` +================================== Ai Message ================================== +Tool Calls: + sql_db_query (call_RNMqyUEMv0rvy0UxSwrXY2AV) + Call ID: call_RNMqyUEMv0rvy0UxSwrXY2AV + Args: + query: SELECT Genre.Name, AVG(Track.Milliseconds) AS AvgMilliseconds +FROM Track +JOIN Genre ON Track.GenreId = Genre.GenreId +GROUP BY Genre.Name +ORDER BY AvgMilliseconds DESC +LIMIT 1; +================================= Tool Message ================================= +Name: sql_db_query + +[('Sci Fi & Fantasy', 2911783.0384615385)] +================================== Ai Message ================================== + +The genre with the longest average track length is "Sci Fi & Fantasy," with an average duration of about 2,911,783 milliseconds (approximately 48.5 minutes) per track. +``` + +This worked well enough: the agent correctly listed the tables, obtained the schemas, wrote a query, checked the query, and ran it to inform its final response. + +!!! tip + You can inspect all aspects of the above run, including steps taken, tools invoked, what prompts were seen by the LLM, and more in the [LangSmith trace](https://smith.langchain.com/public/bd594960-73e3-474b-b6f2-db039d7c713a/r). + +## 3. Customizing the agent + +The prebuilt agent lets us get started quickly, but at each step the agent has access to the full set of tools. Above, we relied on the system prompt to constrain its behavior— for example, we instructed the agent to always start with the "list tables" tool, and to always run a query-checker tool before executing the query. + +We can enforce a higher degree of control in LangGraph by customizing the agent. Below, we implement a simple ReAct-agent setup, with dedicated nodes for specific tool-calls. We will use the same [state](../../concepts/low_level.md#state) as the pre-built agent. + +We construct dedicated nodes for the following steps: + +- Listing DB tables +- Calling the "get schema" tool +- Generating a query +- Checking the query + +Putting these steps in dedicated nodes lets us (1) force tool-calls when needed, and (2) customize the prompts associated with each step. + +```python +from typing import Literal +from langchain_core.messages import AIMessage +from langchain_core.runnables import RunnableConfig +from langgraph.graph import END, START, MessagesState, StateGraph +from langgraph.prebuilt import ToolNode + + +get_schema_tool = next(tool for tool in tools if tool.name == "sql_db_schema") +get_schema_node = ToolNode([get_schema_tool], name="get_schema") + +run_query_tool = next(tool for tool in tools if tool.name == "sql_db_query") +run_query_node = ToolNode([run_query_tool], name="run_query") + + +# Example: create a predetermined tool call +def list_tables(state: MessagesState): + tool_call = { + "name": "sql_db_list_tables", + "args": {}, + "id": "abc123", + "type": "tool_call", + } + tool_call_message = AIMessage(content="", tool_calls=[tool_call]) + + list_tables_tool = next(tool for tool in tools if tool.name == "sql_db_list_tables") + tool_message = list_tables_tool.invoke(tool_call) + response = AIMessage(f"Available tables: {tool_message.content}") + + return {"messages": [tool_call_message, tool_message, response]} + + +# Example: force a model to create a tool call +def call_get_schema(state: MessagesState): + # Note that LangChain enforces that all models accept `tool_choice="any"` + # as well as `tool_choice=`. + llm_with_tools = llm.bind_tools([get_schema_tool], tool_choice="any") + response = llm_with_tools.invoke(state["messages"]) + + return {"messages": [response]} + + +generate_query_system_prompt = """ +You are an agent designed to interact with a SQL database. +Given an input question, create a syntactically correct {dialect} query to run, +then look at the results of the query and return the answer. Unless the user +specifies a specific number of examples they wish to obtain, always limit your +query to at most {top_k} results. + +You can order the results by a relevant column to return the most interesting +examples in the database. Never query for all the columns from a specific table, +only ask for the relevant columns given the question. + +DO NOT make any DML statements (INSERT, UPDATE, DELETE, DROP etc.) to the database. +""".format( + dialect=db.dialect, + top_k=5, +) + + +def generate_query(state: MessagesState): + system_message = { + "role": "system", + "content": generate_query_system_prompt, + } + # We do not force a tool call here, to allow the model to + # respond naturally when it obtains the solution. + llm_with_tools = llm.bind_tools([run_query_tool]) + response = llm_with_tools.invoke([system_message] + state["messages"]) + + return {"messages": [response]} + + +check_query_system_prompt = """ +You are a SQL expert with a strong attention to detail. +Double check the {dialect} query for common mistakes, including: +- Using NOT IN with NULL values +- Using UNION when UNION ALL should have been used +- Using BETWEEN for exclusive ranges +- Data type mismatch in predicates +- Properly quoting identifiers +- Using the correct number of arguments for functions +- Casting to the correct data type +- Using the proper columns for joins + +If there are any of the above mistakes, rewrite the query. If there are no mistakes, +just reproduce the original query. + +You will call the appropriate tool to execute the query after running this check. +""".format(dialect=db.dialect) + + +def check_query(state: MessagesState): + system_message = { + "role": "system", + "content": check_query_system_prompt, + } + + # Generate an artificial user message to check + tool_call = state["messages"][-1].tool_calls[0] + user_message = {"role": "user", "content": tool_call["args"]["query"]} + llm_with_tools = llm.bind_tools([run_query_tool], tool_choice="any") + response = llm_with_tools.invoke([system_message, user_message]) + response.id = state["messages"][-1].id + + return {"messages": [response]} +``` + +Finally, we assemble these steps into a workflow using the Graph API. We define a [conditional edge](../../concepts/low_level.md#conditional-edges) at the query generation step that will route to the query checker if a query is generated, or end if there are no tool calls present, such that the LLM has delivered a response to the query. + +```python +def should_continue(state: MessagesState) -> Literal[END, "check_query"]: + messages = state["messages"] + last_message = messages[-1] + if not last_message.tool_calls: + return END + else: + return "check_query" + + +builder = StateGraph(MessagesState) +builder.add_node(list_tables) +builder.add_node(call_get_schema) +builder.add_node(get_schema_node, "get_schema") +builder.add_node(generate_query) +builder.add_node(check_query) +builder.add_node(run_query_node, "run_query") + +builder.add_edge(START, "list_tables") +builder.add_edge("list_tables", "call_get_schema") +builder.add_edge("call_get_schema", "get_schema") +builder.add_edge("get_schema", "generate_query") +builder.add_conditional_edges( + "generate_query", + should_continue, +) +builder.add_edge("check_query", "run_query") +builder.add_edge("run_query", "generate_query") + +agent = builder.compile() +``` + +We visualize the application below: + +```python +from IPython.display import Image, display +from langchain_core.runnables.graph import CurveStyle, MermaidDrawMethod, NodeStyles + +display(Image(agent.get_graph().draw_mermaid_png())) +``` + +![Graph](./output.png) + +**Note:** When you run this code, it will generate and display a visual representation of the SQL agent graph showing the flow between the different nodes (list_tables → call_get_schema → get_schema → generate_query → check_query → run_query). + +We can now invoke the graph exactly as before: + +```python +question = "Which genre on average has the longest tracks?" + +for step in agent.stream( + {"messages": [{"role": "user", "content": question}]}, + stream_mode="values", +): + step["messages"][-1].pretty_print() +``` + +**Output:** +``` +================================ Human Message ================================= + +Which genre on average has the longest tracks? +================================== Ai Message ================================== + +Available tables: Album, Artist, Customer, Employee, Genre, Invoice, InvoiceLine, MediaType, Playlist, PlaylistTrack, Track +================================== Ai Message ================================== +Tool Calls: + sql_db_schema (call_qxKtYiHgf93AiTDin9ez5wFp) + Call ID: call_qxKtYiHgf93AiTDin9ez5wFp + Args: + table_names: Genre,Track +================================= Tool Message ================================= +Name: sql_db_schema + + +CREATE TABLE "Genre" ( + "GenreId" INTEGER NOT NULL, + "Name" NVARCHAR(120), + PRIMARY KEY ("GenreId") +) + +/* +3 rows from Genre table: +GenreId Name +1 Rock +2 Jazz +3 Metal +*/ + + +CREATE TABLE "Track" ( + "TrackId" INTEGER NOT NULL, + "Name" NVARCHAR(200) NOT NULL, + "AlbumId" INTEGER, + "MediaTypeId" INTEGER NOT NULL, + "GenreId" INTEGER, + "Composer" NVARCHAR(220), + "Milliseconds" INTEGER NOT NULL, + "Bytes" INTEGER, + "UnitPrice" NUMERIC(10, 2) NOT NULL, + PRIMARY KEY ("TrackId"), + FOREIGN KEY("MediaTypeId") REFERENCES "MediaType" ("MediaTypeId"), + FOREIGN KEY("GenreId") REFERENCES "Genre" ("GenreId"), + FOREIGN KEY("AlbumId") REFERENCES "Album" ("AlbumId") +) + +/* +3 rows from Track table: +TrackId Name AlbumId MediaTypeId GenreId Composer Milliseconds Bytes UnitPrice +1 For Those About To Rock (We Salute You) 1 1 1 Angus Young, Malcolm Young, Brian Johnson 343719 11170334 0.99 +2 Balls to the Wall 2 2 1 None 342562 5510424 0.99 +3 Fast As a Shark 3 2 1 F. Baltes, S. Kaufman, U. Dirkscneider & W. Hoffman 230619 3990994 0.99 +*/ +================================== Ai Message ================================== +Tool Calls: + sql_db_query (call_RPN3GABMfb6DTaFTLlwnZxVN) + Call ID: call_RPN3GABMfb6DTaFTLlwnZxVN + Args: + query: SELECT Genre.Name, AVG(Track.Milliseconds) AS AvgTrackLength +FROM Track +JOIN Genre ON Track.GenreId = Genre.GenreId +GROUP BY Genre.GenreId +ORDER BY AvgTrackLength DESC +LIMIT 1; +================================== Ai Message ================================== +Tool Calls: + sql_db_query (call_PR4s8ymiF3ZQLaoZADXtdqcl) + Call ID: call_PR4s8ymiF3ZQLaoZADXtdqcl + Args: + query: SELECT Genre.Name, AVG(Track.Milliseconds) AS AvgTrackLength +FROM Track +JOIN Genre ON Track.GenreId = Genre.GenreId +GROUP BY Genre.GenreId +ORDER BY AvgTrackLength DESC +LIMIT 1; +================================= Tool Message ================================= +Name: sql_db_query + +[('Sci Fi & Fantasy', 2911783.0384615385)] +================================== Ai Message ================================== + +The genre with the longest tracks on average is "Sci Fi & Fantasy," with an average track length of approximately 2,911,783 milliseconds. +``` + +!!! tip + See [LangSmith trace](https://smith.langchain.com/public/94b8c9ac-12f7-4692-8706-836a1f30f1ea/r) for the above run. + +## Next steps + +Check out [this guide](https://docs.smith.langchain.com/evaluation/how_to_guides/langgraph) for evaluating LangGraph applications, including SQL agents like this one, using LangSmith. \ No newline at end of file diff --git a/docs/mkdocs.yml b/docs/mkdocs.yml index 7cfbb246f..020c6e28a 100644 --- a/docs/mkdocs.yml +++ b/docs/mkdocs.yml @@ -269,19 +269,18 @@ nav: - Examples: - Template applications: concepts/template_applications.md # TODO: make tutorial - - Agentic RAG: tutorials/rag/langgraph_agentic_rag.ipynb - - Agent Supervisor: tutorials/multi_agent/agent_supervisor.ipynb - - SQL agent: tutorials/sql-agent.ipynb + - Agentic RAG: tutorials/rag/langgraph_agentic_rag.md + - Agent Supervisor: tutorials/multi_agent/agent_supervisor.md + - SQL agent: tutorials/sql/sql-agent.md - Prebuilt chat UI: agents/ui.md - - Graph runs in LangSmith: how-tos/run-id-langsmith.ipynb + - Graph runs in LangSmith: how-tos/run-id-langsmith.md - LangGraph Platform: - Authentication: - tutorials/auth/getting_started.md - tutorials/auth/resource_auth.md - tutorials/auth/add_auth_server.md - Use RemoteGraph: how-tos/use-remote-graph.md - - Deploy CrewAI, AutoGen, and other frameworks: how-tos/autogen-langgraph-platform.ipynb - # combine with how-tos/autogen-integration.ipynb + - Deploy CrewAI, AutoGen, and other frameworks: how-tos/autogen-integration.md - Front-end and generative UI: - Integrate LangGraph into a React app: cloud/how-tos/use_stream_react.md - Implement generative UI with LangGraph: cloud/how-tos/generative_ui_react.md From 4de1bd6e666ba9305d259f8f043608c4f9c6aac9 Mon Sep 17 00:00:00 2001 From: Lauren Hirata Singh Date: Tue, 8 Jul 2025 12:33:14 -0400 Subject: [PATCH 3/4] docs: cleanup (#5401) * docs: cleanup * fix nav * fix * fix nits --- docs/_scripts/extract_images.py | 70 -- docs/docs/additional-resources/index.md | 11 + docs/docs/agents/agents.md | 2 +- docs/docs/cloud/concepts/threads.md | 12 - docs/docs/examples/index.md | 23 + docs/docs/guides/index.md | 41 + .../tutorials/multi_agent/agent_supervisor.md | 2 +- docs/docs/tutorials/overview.md | 21 - docs/docs/tutorials/sql-agent.ipynb | 759 ------------------ docs/mkdocs.yml | 3 + 10 files changed, 80 insertions(+), 864 deletions(-) delete mode 100644 docs/_scripts/extract_images.py create mode 100644 docs/docs/additional-resources/index.md delete mode 100644 docs/docs/cloud/concepts/threads.md create mode 100644 docs/docs/examples/index.md create mode 100644 docs/docs/guides/index.md delete mode 100644 docs/docs/tutorials/overview.md delete mode 100644 docs/docs/tutorials/sql-agent.ipynb diff --git a/docs/_scripts/extract_images.py b/docs/_scripts/extract_images.py deleted file mode 100644 index 781ba0a22..000000000 --- a/docs/_scripts/extract_images.py +++ /dev/null @@ -1,70 +0,0 @@ -#!/usr/bin/env python3 -""" -Script to extract images from the graph-api.ipynb notebook and save them to assets folder. -""" - -import json -import base64 -import os -from pathlib import Path - -def extract_images_from_notebook(notebook_path, assets_dir): - """Extract images from notebook and save them to assets directory.""" - - # Read the notebook - with open(notebook_path, 'r') as f: - notebook = json.load(f) - - # Create assets directory if it doesn't exist - os.makedirs(assets_dir, exist_ok=True) - - image_count = 0 - - # Process each cell - for cell_idx, cell in enumerate(notebook['cells']): - if cell['cell_type'] == 'code': - # Check if this cell contains draw_mermaid_png - source = ''.join(cell.get('source', [])) - if 'draw_mermaid_png' in source: - print(f"Found draw_mermaid_png in cell {cell_idx}") - - # Check for outputs with images - if 'outputs' in cell: - for output_idx, output in enumerate(cell['outputs']): - if output.get('output_type') == 'display_data': - data = output.get('data', {}) - - # Check for PNG data - if 'image/png' in data: - png_data = data['image/png'] - - # Decode base64 data - try: - image_bytes = base64.b64decode(png_data) - - # Generate filename - image_count += 1 - filename = f"graph_api_image_{image_count}.png" - filepath = os.path.join(assets_dir, filename) - - # Save the image - with open(filepath, 'wb') as img_file: - img_file.write(image_bytes) - - print(f"Saved image: {filepath}") - - except Exception as e: - print(f"Error decoding image {image_count}: {e}") - - print(f"Extracted {image_count} images to {assets_dir}") - return image_count - -if __name__ == "__main__": - notebook_path = "docs/docs/how-tos/graph-api.ipynb" - assets_dir = "docs/docs/how-tos/assets" - - if os.path.exists(notebook_path): - count = extract_images_from_notebook(notebook_path, assets_dir) - print(f"Successfully extracted {count} images") - else: - print(f"Notebook not found: {notebook_path}") \ No newline at end of file diff --git a/docs/docs/additional-resources/index.md b/docs/docs/additional-resources/index.md new file mode 100644 index 000000000..6e4e7a93b --- /dev/null +++ b/docs/docs/additional-resources/index.md @@ -0,0 +1,11 @@ +# Additional resources + +This section contains additional resources for LangGraph. + +- [Community agents](../agents/prebuilt.md): A collection of prebuilt libraries that you can use in your LangGraph applications. +- [LangGraph Academy](https://academy.langchain.com/courses/intro-to-langgraph): A collection of courses that teach you how to use LangGraph. +- [Case studies](../adopters.md): A collection of case studies that show how LangGraph is used in production. +- [FAQ](../concepts/faq.md): A collection of frequently asked questions about LangGraph. +- [llms.txt](../llms-txt-overview.md): A list of documentation files in the `llms.txt` format that allow LLMs and agents to access our documentation. +- [LangChain Forum](https://forum.langchain.com/): A place to ask questions and get help from other LangGraph users. +- [Troubleshooting](../troubleshooting/errors/index.md.md): A collection of troubleshooting guides for common issues. \ No newline at end of file diff --git a/docs/docs/agents/agents.md b/docs/docs/agents/agents.md index b00184266..04ee419ce 100644 --- a/docs/docs/agents/agents.md +++ b/docs/docs/agents/agents.md @@ -52,7 +52,7 @@ agent.invoke( ) ``` -1. Define a tool for the agent to use. Tools can be defined as vanilla Python functions. For more advanced tool usage and customization, check the [tools](./tools.md) page. +1. Define a tool for the agent to use. Tools can be defined as vanilla Python functions. For more advanced tool usage and customization, check the [tools](../how-tos/tool-calling.md) page. 2. Provide a language model for the agent to use. To learn more about configuring language models for the agents, check the [models](./models.md) page. 3. Provide a list of tools for the model to use. 4. Provide a system prompt (instructions) to the language model used by the agent. diff --git a/docs/docs/cloud/concepts/threads.md b/docs/docs/cloud/concepts/threads.md deleted file mode 100644 index ffd48faa8..000000000 --- a/docs/docs/cloud/concepts/threads.md +++ /dev/null @@ -1,12 +0,0 @@ -# Threads - -A thread contains the accumulated state of a sequence of [runs](../../concepts/assistants.md#execution). When a run is executed, the [state](../../concepts/low_level.md#state) of the underlying graph of the assistant will be persisted to the thread. - -A thread's current and historical state can be retrieved. To persist state, a thread must be created prior to executing a run. - -The state of a thread at a particular point in time is called a [checkpoint](../../concepts/persistence.md#checkpoints). Checkpoints are persisted and can be used to restore the state of a thread at a later time. - -## Learn more - -* For more on threads and checkpoints, see this section of the [LangGraph conceptual guide](../../concepts/persistence.md). -* The LangGraph Platform API provides several endpoints for creating and managing threads and thread state. See the [API reference](../../cloud/reference/api/api_ref.html#tag/threads) for more details. diff --git a/docs/docs/examples/index.md b/docs/docs/examples/index.md new file mode 100644 index 000000000..84c721ac3 --- /dev/null +++ b/docs/docs/examples/index.md @@ -0,0 +1,23 @@ +# Examples + +The pages in this section provide end-to-end examples for the following topics: + +## General + +- [Template Applications](../concepts/template_applications.md): Create a LangGraph application from a template. +- [Agentic RAG](./rag/langgraph_agentic_rag.md): Build a retrieval agent that can decide when to use a retriever tool. +- [Agent Supervisor](./multi_agent/agent_supervisor.md): Build a supervisor agent that can manage a team of agents. +- [SQL agent](./sql/sql-agent.md): Build a SQL agent that can execute SQL queries and return the results. +- [Prebuilt chat UI](../agents/ui.md): Use a prebuilt chat UI to interact with any LangGraph agent. +- [Graph runs in LangSmith](../how-tos/run-id-langsmith.md): Use LangSmith to track and analyze graph runs. + +## LangGraph Platform + +- [Set up custom authentication](./auth/getting_started.md): Set up custom authentication for your LangGraph application. +- [Make conversations private](./auth/resource_auth.md): Make conversations private by using resource-based authentication. +- [Connect an authentication provider](./auth/add_auth_server.md): Connect an authentication provider to your LangGraph application. +- [Rebuild graph at runtime](../cloud/deployment/graph_rebuild.md): Rebuild a graph at runtime. +- [Use RemoteGraph](../how-tos/use-remote-graph.md): Use RemoteGraph to deploy your LangGraph application to a remote server. +- [Deploy CrewAI, AutoGen, and other frameworks](../how-tos/autogen-integration.md): Deploy CrewAI, AutoGen, and other frameworks with LangGraph. +- [Integrate LangGraph into a React app](../cloud/how-tos/use_stream_react.md) +- [Implement Generative User Interfaces with LangGraph](../cloud/how-tos/generative_ui_react.md) \ No newline at end of file diff --git a/docs/docs/guides/index.md b/docs/docs/guides/index.md new file mode 100644 index 000000000..1d57f63ea --- /dev/null +++ b/docs/docs/guides/index.md @@ -0,0 +1,41 @@ +# Guides + +The pages in this section provide a conceptual overview and how-tos for the following topics: + +## LangGraph APIs + +- [Graph API](../concepts/low_level.md): Use the Graph API to define workflows using a graph paradigm. +- [Functional API](../concepts/functional_api.md): Use Functional API to build workflows using a functional paradigm without thinking about the graph structure. +- [Runtime](../concepts/pregel.md): Pregel implements LangGraph's runtime, managing the execution of LangGraph applications. + +## Core capabilities + +These capabilities are available in both LangGraph OSS and the LangGraph Platform. + +- [Streaming](../concepts/streaming.md): Stream outputs from a LangGraph graph. +- [Persistence](../concepts/persistence.md): Persist the state of a LangGraph graph. +- [Durable execution](../concepts/durable_execution.md): Save progress at key points in the graph execution. +- [Memory](../concepts/memory.md): Remember information about previous interactions. +- [Context](../agents/context.md): Pass outside data to a LangGraph graph to provide context for the graph execution. +- [Models](../agents/models.md): Integrate various LLMs into your LangGraph application. +- [Tools](../concepts/tools.md): Interface directly with external systems. +- [Human-in-the-loop](../concepts/human_in_the_loop.md): Enable human intervention at any point in a workflow. +- [Breakpoints](../concepts/breakpoints.md): Pause the execution of a LangGraph graph at a specific point. +- [Time travel](../concepts/time-travel.md): Travel back in time to a specific point in the execution of a LangGraph graph. +- [Subgraphs](../concepts/subgraphs.md): Build modular graphs. +- [Multi-agent](../concepts/multi_agent.md): Break down a complex workflow into multiple agents. +- [MCP](../concepts/mcp.md): Use MCP servers in a LangGraph graph. +- [Evaluation](../agents/evals.md): Use LangSmith to evaluate your graph's performance. + +## Platform-only capabilities + +These capabilities are only available in [LangGraph Platform](../concepts/langgraph_platform.md). + +- [Authentication and access control](../concepts/auth.md): Authenticate and authorize users to access a Langraph graph. +- [Assistants](../concepts/assistants.md): Build assistants that can be used to interact with a LangGraph graph. +- [Double-texting](../concepts/double_texting.md): Handle double-texting (consecutive messages before a first response is returned) in a LangGraph graph. +- [Webhooks](../cloud/concepts/webhooks.md): Send webhooks to a LangGraph graph. +- [Cron jobs](../cloud/concepts/cron_jobs.md): Schedule jobs to run at a specific time. +- [Server customization](../how-tos/http/custom_lifespan.md): Customize the server that runs a LangGraph graph. +- [Data management](../cloud/concepts/data_storage_and_privacy.md): Manage data in a LangGraph graph. +- [Deployment](../concepts/deployment_options.md): Deploy a LangGraph graph to a server. \ No newline at end of file diff --git a/docs/docs/tutorials/multi_agent/agent_supervisor.md b/docs/docs/tutorials/multi_agent/agent_supervisor.md index ba8f84ff6..71335c8db 100644 --- a/docs/docs/tutorials/multi_agent/agent_supervisor.md +++ b/docs/docs/tutorials/multi_agent/agent_supervisor.md @@ -206,7 +206,7 @@ Name: tavily_search ### Math agent -For math agent tools we will use [vanilla Python functions](../../agents/tools.md#define-simple-tools): +For math agent tools we will use [vanilla Python functions](../../how-tos/tool-calling.md#define-a-tool): ```python def add(a: float, b: float): diff --git a/docs/docs/tutorials/overview.md b/docs/docs/tutorials/overview.md deleted file mode 100644 index 6eacbd4e4..000000000 --- a/docs/docs/tutorials/overview.md +++ /dev/null @@ -1,21 +0,0 @@ -# Examples - -The pages in this section provide end-to-end examples for the following topics: - -## General - -- [Agentic RAG](./rag/langgraph_adaptive_rag.ipynb) -- [Agent Supervisor](./multi_agent/agent_supervisor.md) -- [SQL agent](./sql/sql-agent.md) -- [Graph runs in LangSmith](../how-tos/run-id-langsmith.md) - -## LangGraph Platform - -- [Set up custom authentication](./auth/getting_started.md) -- [Make conversations private](./auth/resource_auth.md) -- [Connect an authentication provider](./auth/add_auth_server.md) -- [Rebuild graph at runtime](../cloud/deployment/graph_rebuild.md) -- [Use RemoteGraph](../how-tos/use-remote-graph.md) -- [Deploy CrewAI, AutoGen, and other frameworks](../how-tos/autogen-integration.md) -- [Integrate LangGraph into a React app](../cloud/how-tos/use_stream_react.md) -- [Implement Generative User Interfaces with LangGraph](../cloud/how-tos/generative_ui_react.md) \ No newline at end of file diff --git a/docs/docs/tutorials/sql-agent.ipynb b/docs/docs/tutorials/sql-agent.ipynb deleted file mode 100644 index a06626f40..000000000 --- a/docs/docs/tutorials/sql-agent.ipynb +++ /dev/null @@ -1,759 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "id": "cb71c4a7-9f99-4f08-be34-b3ba81ac075f", - "metadata": {}, - "source": [ - "# Build a SQL agent\n", - "\n", - "In this tutorial, we will walk through how to build an agent that can answer questions about a SQL database.\n", - "\n", - "At a high level, the agent will:\n", - "\n", - "1. Fetch the available tables from the database\n", - "2. Decide which tables are relevant to the question\n", - "3. Fetch the schemas for the relevant tables\n", - "4. Generate a query based on the question and information from the schemas\n", - "5. Double-check the query for common mistakes using an LLM\n", - "6. Execute the query and return the results\n", - "7. Correct mistakes surfaced by the database engine until the query is successful\n", - "8. Formulate a response based on the results\n", - "\n", - "!!! warning \"Security note\"\n", - "\n", - " Building Q&A systems of SQL databases requires executing model-generated SQL queries. There are inherent risks in doing this. Make sure that your database connection permissions are always scoped as narrowly as possible for your agent's needs. This will mitigate though not eliminate the risks of building a model-driven system.\n", - "\n", - "## 1. Setup\n", - "\n", - "Let's first install some dependencies. This tutorial uses SQL database and tool abstractions from [langchain-community](https://python.langchain.com/docs/concepts/architecture/#langchain-community). We will also require a LangChain [chat model](https://python.langchain.com/docs/concepts/chat_models/)." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "1b613286-41c3-4ca5-9514-36aac0121a5a", - "metadata": {}, - "outputs": [], - "source": [ - "%%capture --no-stderr\n", - "%pip install -U langgraph langchain_community \"langchain[openai]\"" - ] - }, - { - "cell_type": "markdown", - "id": "f5258ad0-39e1-418e-bddb-56e9aab1b43a", - "metadata": {}, - "source": [ - "

\n", - "

Set up LangSmith for LangGraph development

\n", - "

\n", - " Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started here. \n", - "

\n", - "
" - ] - }, - { - "cell_type": "markdown", - "id": "62886241-caa7-40bd-8867-096a97acaf40", - "metadata": {}, - "source": [ - "### Select a LLM\n", - "\n", - "First we [initialize our LLM](https://python.langchain.com/docs/how_to/chat_models_universal_init/). Any model supporting [tool-calling](https://python.langchain.com/docs/integrations/chat/#featured-providers) should work. We use OpenAI below." - ] - }, - { - "cell_type": "code", - "execution_count": 1, - "id": "2b981efc-bcc9-49f4-aa66-4c15a76210a4", - "metadata": {}, - "outputs": [], - "source": [ - "from langchain.chat_models import init_chat_model\n", - "\n", - "llm = init_chat_model(\"openai:gpt-4.1\")" - ] - }, - { - "cell_type": "markdown", - "id": "804ca604-7ac4-4229-a3e9-3a9825eeefa4", - "metadata": {}, - "source": [ - "### Configure the database\n", - "\n", - "We will be creating a SQLite database for this tutorial. SQLite is a lightweight database that is easy to set up and use. We will be loading the `chinook` database, which is a sample database that represents a digital media store.\n", - "Find more information about the database [here](https://www.sqlitetutorial.net/sqlite-sample-database/).\n", - "\n", - "For convenience, we have hosted the database (`Chinook.db`) on a public GCS bucket." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "b58ba6b7-5f80-4970-bc29-961de0e24fbd", - "metadata": {}, - "outputs": [], - "source": [ - "import requests\n", - "\n", - "url = \"https://storage.googleapis.com/benchmarks-artifacts/chinook/Chinook.db\"\n", - "\n", - "response = requests.get(url)\n", - "\n", - "if response.status_code == 200:\n", - " # Open a local file in binary write mode\n", - " with open(\"Chinook.db\", \"wb\") as file:\n", - " # Write the content of the response (the file) to the local file\n", - " file.write(response.content)\n", - " print(\"File downloaded and saved as Chinook.db\")\n", - "else:\n", - " print(f\"Failed to download the file. Status code: {response.status_code}\")" - ] - }, - { - "cell_type": "markdown", - "id": "5ae005fb-e7b0-43f1-95b2-565d19c82610", - "metadata": {}, - "source": [ - "We will use a handy SQL database wrapper available in the `langchain_community` package to interact with the database. The wrapper provides a simple interface to execute SQL queries and fetch results:" - ] - }, - { - "cell_type": "code", - "execution_count": 2, - "id": "18822c9b-6df2-4a5c-a12d-046d2b2131c9", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Dialect: sqlite\n", - "Available tables: ['Album', 'Artist', 'Customer', 'Employee', 'Genre', 'Invoice', 'InvoiceLine', 'MediaType', 'Playlist', 'PlaylistTrack', 'Track']\n", - "Sample output: [(1, 'AC/DC'), (2, 'Accept'), (3, 'Aerosmith'), (4, 'Alanis Morissette'), (5, 'Alice In Chains')]\n" - ] - } - ], - "source": [ - "from langchain_community.utilities import SQLDatabase\n", - "\n", - "db = SQLDatabase.from_uri(\"sqlite:///Chinook.db\")\n", - "\n", - "print(f\"Dialect: {db.dialect}\")\n", - "print(f\"Available tables: {db.get_usable_table_names()}\")\n", - "print(f'Sample output: {db.run(\"SELECT * FROM Artist LIMIT 5;\")}')" - ] - }, - { - "cell_type": "markdown", - "id": "3bf47b6c-30dd-405f-8a03-21327af3c764", - "metadata": {}, - "source": [ - "### Tools for database interactions\n", - "\n", - "`langchain-community` implements some built-in tools for interacting with our `SQLDatabase`, including tools for listing tables, reading table schemas, and checking and running queries:" - ] - }, - { - "cell_type": "code", - "execution_count": 3, - "id": "aab9e00a-aa28-494a-a95a-a6638ea03bd1", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "sql_db_query: Input to this tool is a detailed and correct SQL query, output is a result from the database. If the query is not correct, an error message will be returned. If an error is returned, rewrite the query, check the query, and try again. If you encounter an issue with Unknown column 'xxxx' in 'field list', use sql_db_schema to query the correct table fields.\n", - "\n", - "sql_db_schema: Input to this tool is a comma-separated list of tables, output is the schema and sample rows for those tables. Be sure that the tables actually exist by calling sql_db_list_tables first! Example Input: table1, table2, table3\n", - "\n", - "sql_db_list_tables: Input is an empty string, output is a comma-separated list of tables in the database.\n", - "\n", - "sql_db_query_checker: Use this tool to double check if your query is correct before executing it. Always use this tool before executing a query with sql_db_query!\n", - "\n" - ] - } - ], - "source": [ - "from langchain_community.agent_toolkits import SQLDatabaseToolkit\n", - "\n", - "toolkit = SQLDatabaseToolkit(db=db, llm=llm)\n", - "\n", - "tools = toolkit.get_tools()\n", - "\n", - "for tool in tools:\n", - " print(f\"{tool.name}: {tool.description}\\n\")" - ] - }, - { - "cell_type": "markdown", - "id": "6baf93d3-afd5-484e-9bf7-3ff20401e06f", - "metadata": {}, - "source": [ - "## 2. Using a prebuilt agent\n", - "\n", - "Given these tools, we can initialize a pre-built agent in a single line. To customize our agents behavior, we write a descriptive system prompt." - ] - }, - { - "cell_type": "code", - "execution_count": 4, - "id": "f6b9f10f-c6cb-43bc-b671-79aadc71280b", - "metadata": {}, - "outputs": [], - "source": [ - "from langgraph.prebuilt import create_react_agent\n", - "\n", - "system_prompt = \"\"\"\n", - "You are an agent designed to interact with a SQL database.\n", - "Given an input question, create a syntactically correct {dialect} query to run,\n", - "then look at the results of the query and return the answer. Unless the user\n", - "specifies a specific number of examples they wish to obtain, always limit your\n", - "query to at most {top_k} results.\n", - "\n", - "You can order the results by a relevant column to return the most interesting\n", - "examples in the database. Never query for all the columns from a specific table,\n", - "only ask for the relevant columns given the question.\n", - "\n", - "You MUST double check your query before executing it. If you get an error while\n", - "executing a query, rewrite the query and try again.\n", - "\n", - "DO NOT make any DML statements (INSERT, UPDATE, DELETE, DROP etc.) to the\n", - "database.\n", - "\n", - "To start you should ALWAYS look at the tables in the database to see what you\n", - "can query. Do NOT skip this step.\n", - "\n", - "Then you should query the schema of the most relevant tables.\n", - "\"\"\".format(\n", - " dialect=db.dialect,\n", - " top_k=5,\n", - ")\n", - "\n", - "agent = create_react_agent(\n", - " llm,\n", - " tools,\n", - " prompt=system_prompt,\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "78b1c7cf-9a0f-48c3-baec-be8101bc3c77", - "metadata": {}, - "source": [ - "!!! note\n", - "\n", - " This system prompt includes a number of instructions, such as always running specific tools before or after others. In the [next section](#customizing-the-agent), we will enforce these behaviors through the graph's structure, providing us a greater degree of control and allowing us to simplify the prompt.\n", - "\n", - "\n", - "Let's run this agent on a sample query and observe its behavior:" - ] - }, - { - "cell_type": "code", - "execution_count": 5, - "id": "9317c234-6251-4723-a335-7fee6d11ac2b", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "================================\u001b[1m Human Message \u001b[0m=================================\n", - "\n", - "Which genre on average has the longest tracks?\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Tool Calls:\n", - " sql_db_list_tables (call_d8lCgywSroCgpVl558nmXKwA)\n", - " Call ID: call_d8lCgywSroCgpVl558nmXKwA\n", - " Args:\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: sql_db_list_tables\n", - "\n", - "Album, Artist, Customer, Employee, Genre, Invoice, InvoiceLine, MediaType, Playlist, PlaylistTrack, Track\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Tool Calls:\n", - " sql_db_schema (call_nNf6IIUcwMYLIkE0l6uWkZHe)\n", - " Call ID: call_nNf6IIUcwMYLIkE0l6uWkZHe\n", - " Args:\n", - " table_names: Genre, Track\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: sql_db_schema\n", - "\n", - "\n", - "CREATE TABLE \"Genre\" (\n", - "\t\"GenreId\" INTEGER NOT NULL, \n", - "\t\"Name\" NVARCHAR(120), \n", - "\tPRIMARY KEY (\"GenreId\")\n", - ")\n", - "\n", - "/*\n", - "3 rows from Genre table:\n", - "GenreId\tName\n", - "1\tRock\n", - "2\tJazz\n", - "3\tMetal\n", - "*/\n", - "\n", - "\n", - "CREATE TABLE \"Track\" (\n", - "\t\"TrackId\" INTEGER NOT NULL, \n", - "\t\"Name\" NVARCHAR(200) NOT NULL, \n", - "\t\"AlbumId\" INTEGER, \n", - "\t\"MediaTypeId\" INTEGER NOT NULL, \n", - "\t\"GenreId\" INTEGER, \n", - "\t\"Composer\" NVARCHAR(220), \n", - "\t\"Milliseconds\" INTEGER NOT NULL, \n", - "\t\"Bytes\" INTEGER, \n", - "\t\"UnitPrice\" NUMERIC(10, 2) NOT NULL, \n", - "\tPRIMARY KEY (\"TrackId\"), \n", - "\tFOREIGN KEY(\"MediaTypeId\") REFERENCES \"MediaType\" (\"MediaTypeId\"), \n", - "\tFOREIGN KEY(\"GenreId\") REFERENCES \"Genre\" (\"GenreId\"), \n", - "\tFOREIGN KEY(\"AlbumId\") REFERENCES \"Album\" (\"AlbumId\")\n", - ")\n", - "\n", - "/*\n", - "3 rows from Track table:\n", - "TrackId\tName\tAlbumId\tMediaTypeId\tGenreId\tComposer\tMilliseconds\tBytes\tUnitPrice\n", - "1\tFor Those About To Rock (We Salute You)\t1\t1\t1\tAngus Young, Malcolm Young, Brian Johnson\t343719\t11170334\t0.99\n", - "2\tBalls to the Wall\t2\t2\t1\tNone\t342562\t5510424\t0.99\n", - "3\tFast As a Shark\t3\t2\t1\tF. Baltes, S. Kaufman, U. Dirkscneider & W. Hoffman\t230619\t3990994\t0.99\n", - "*/\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Tool Calls:\n", - " sql_db_query_checker (call_urTRmtiGtTxkwHtscec7Fd2K)\n", - " Call ID: call_urTRmtiGtTxkwHtscec7Fd2K\n", - " Args:\n", - " query: SELECT Genre.Name, AVG(Track.Milliseconds) AS AvgMilliseconds\n", - "FROM Track\n", - "JOIN Genre ON Track.GenreId = Genre.GenreId\n", - "GROUP BY Genre.Name\n", - "ORDER BY AvgMilliseconds DESC\n", - "LIMIT 1;\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: sql_db_query_checker\n", - "\n", - "```sql\n", - "SELECT Genre.Name, AVG(Track.Milliseconds) AS AvgMilliseconds\n", - "FROM Track\n", - "JOIN Genre ON Track.GenreId = Genre.GenreId\n", - "GROUP BY Genre.Name\n", - "ORDER BY AvgMilliseconds DESC\n", - "LIMIT 1;\n", - "```\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Tool Calls:\n", - " sql_db_query (call_RNMqyUEMv0rvy0UxSwrXY2AV)\n", - " Call ID: call_RNMqyUEMv0rvy0UxSwrXY2AV\n", - " Args:\n", - " query: SELECT Genre.Name, AVG(Track.Milliseconds) AS AvgMilliseconds\n", - "FROM Track\n", - "JOIN Genre ON Track.GenreId = Genre.GenreId\n", - "GROUP BY Genre.Name\n", - "ORDER BY AvgMilliseconds DESC\n", - "LIMIT 1;\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: sql_db_query\n", - "\n", - "[('Sci Fi & Fantasy', 2911783.0384615385)]\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "\n", - "The genre with the longest average track length is \"Sci Fi & Fantasy,\" with an average duration of about 2,911,783 milliseconds (approximately 48.5 minutes) per track.\n" - ] - } - ], - "source": [ - "question = \"Which genre on average has the longest tracks?\"\n", - "\n", - "for step in agent.stream(\n", - " {\"messages\": [{\"role\": \"user\", \"content\": question}]},\n", - " stream_mode=\"values\",\n", - "):\n", - " step[\"messages\"][-1].pretty_print()" - ] - }, - { - "cell_type": "markdown", - "id": "6b2afc54-0a26-40f8-bca2-01cbb2fc3443", - "metadata": {}, - "source": [ - "This worked well enough: the agent correctly listed the tables, obtained the schemas, wrote a query, checked the query, and ran it to inform its final response.\n", - "\n", - "!!! tip\n", - "\n", - " You can inspect all aspects of the above run, including steps taken, tools invoked, what prompts were seen by the LLM, and more in the [LangSmith trace](https://smith.langchain.com/public/bd594960-73e3-474b-b6f2-db039d7c713a/r)." - ] - }, - { - "cell_type": "markdown", - "id": "2ed79712-ca6b-47dc-9bc9-7cb6fc50a8aa", - "metadata": {}, - "source": [ - "## 3. Customizing the agent\n", - "\n", - "The prebuilt agent lets us get started quickly, but at each step the agent has access to the full set of tools. Above, we relied on the system prompt to constrain its behavior— for example, we instructed the agent to always start with the \"list tables\" tool, and to always run a query-checker tool before executing the query.\n", - "\n", - "We can enforce a higher degree of control in LangGraph by customizing the agent. Below, we implement a simple ReAct-agent setup, with dedicated nodes for specific tool-calls. We will use the same [state](../../concepts/low_level/#state) as the pre-built agent.\n", - "\n", - "We construct dedicated nodes for the following steps:\n", - "\n", - "- Listing DB tables\n", - "- Calling the \"get schema\" tool\n", - "- Generating a query\n", - "- Checking the query\n", - "\n", - "Putting these steps in dedicated nodes lets us (1) force tool-calls when needed, and (2) customize the prompts associated with each step." - ] - }, - { - "cell_type": "code", - "execution_count": 6, - "id": "992f6ce6-f932-4f44-994c-c0ba5032dbd1", - "metadata": {}, - "outputs": [], - "source": [ - "from typing import Literal\n", - "from langchain_core.messages import AIMessage\n", - "from langchain_core.runnables import RunnableConfig\n", - "from langgraph.graph import END, START, MessagesState, StateGraph\n", - "from langgraph.prebuilt import ToolNode\n", - "\n", - "\n", - "get_schema_tool = next(tool for tool in tools if tool.name == \"sql_db_schema\")\n", - "get_schema_node = ToolNode([get_schema_tool], name=\"get_schema\")\n", - "\n", - "run_query_tool = next(tool for tool in tools if tool.name == \"sql_db_query\")\n", - "run_query_node = ToolNode([run_query_tool], name=\"run_query\")\n", - "\n", - "\n", - "# Example: create a predetermined tool call\n", - "def list_tables(state: MessagesState):\n", - " tool_call = {\n", - " \"name\": \"sql_db_list_tables\",\n", - " \"args\": {},\n", - " \"id\": \"abc123\",\n", - " \"type\": \"tool_call\",\n", - " }\n", - " tool_call_message = AIMessage(content=\"\", tool_calls=[tool_call])\n", - "\n", - " list_tables_tool = next(tool for tool in tools if tool.name == \"sql_db_list_tables\")\n", - " tool_message = list_tables_tool.invoke(tool_call)\n", - " response = AIMessage(f\"Available tables: {tool_message.content}\")\n", - "\n", - " return {\"messages\": [tool_call_message, tool_message, response]}\n", - "\n", - "\n", - "# Example: force a model to create a tool call\n", - "def call_get_schema(state: MessagesState):\n", - " # Note that LangChain enforces that all models accept `tool_choice=\"any\"`\n", - " # as well as `tool_choice=`.\n", - " llm_with_tools = llm.bind_tools([get_schema_tool], tool_choice=\"any\")\n", - " response = llm_with_tools.invoke(state[\"messages\"])\n", - "\n", - " return {\"messages\": [response]}\n", - "\n", - "\n", - "generate_query_system_prompt = \"\"\"\n", - "You are an agent designed to interact with a SQL database.\n", - "Given an input question, create a syntactically correct {dialect} query to run,\n", - "then look at the results of the query and return the answer. Unless the user\n", - "specifies a specific number of examples they wish to obtain, always limit your\n", - "query to at most {top_k} results.\n", - "\n", - "You can order the results by a relevant column to return the most interesting\n", - "examples in the database. Never query for all the columns from a specific table,\n", - "only ask for the relevant columns given the question.\n", - "\n", - "DO NOT make any DML statements (INSERT, UPDATE, DELETE, DROP etc.) to the database.\n", - "\"\"\".format(\n", - " dialect=db.dialect,\n", - " top_k=5,\n", - ")\n", - "\n", - "\n", - "def generate_query(state: MessagesState):\n", - " system_message = {\n", - " \"role\": \"system\",\n", - " \"content\": generate_query_system_prompt,\n", - " }\n", - " # We do not force a tool call here, to allow the model to\n", - " # respond naturally when it obtains the solution.\n", - " llm_with_tools = llm.bind_tools([run_query_tool])\n", - " response = llm_with_tools.invoke([system_message] + state[\"messages\"])\n", - "\n", - " return {\"messages\": [response]}\n", - "\n", - "\n", - "check_query_system_prompt = \"\"\"\n", - "You are a SQL expert with a strong attention to detail.\n", - "Double check the {dialect} query for common mistakes, including:\n", - "- Using NOT IN with NULL values\n", - "- Using UNION when UNION ALL should have been used\n", - "- Using BETWEEN for exclusive ranges\n", - "- Data type mismatch in predicates\n", - "- Properly quoting identifiers\n", - "- Using the correct number of arguments for functions\n", - "- Casting to the correct data type\n", - "- Using the proper columns for joins\n", - "\n", - "If there are any of the above mistakes, rewrite the query. If there are no mistakes,\n", - "just reproduce the original query.\n", - "\n", - "You will call the appropriate tool to execute the query after running this check.\n", - "\"\"\".format(dialect=db.dialect)\n", - "\n", - "\n", - "def check_query(state: MessagesState):\n", - " system_message = {\n", - " \"role\": \"system\",\n", - " \"content\": check_query_system_prompt,\n", - " }\n", - "\n", - " # Generate an artificial user message to check\n", - " tool_call = state[\"messages\"][-1].tool_calls[0]\n", - " user_message = {\"role\": \"user\", \"content\": tool_call[\"args\"][\"query\"]}\n", - " llm_with_tools = llm.bind_tools([run_query_tool], tool_choice=\"any\")\n", - " response = llm_with_tools.invoke([system_message, user_message])\n", - " response.id = state[\"messages\"][-1].id\n", - "\n", - " return {\"messages\": [response]}" - ] - }, - { - "cell_type": "markdown", - "id": "cd23a432-09c1-428c-88a7-6b4d1fe55577", - "metadata": {}, - "source": [ - "Finally, we assemble these steps into a workflow using the Graph API. We define a [conditional edge](../../concepts/low_level/#conditional-edges) at the query generation step that will route to the query checker if a query is generated, or end if there are no tool calls present, such that the LLM has delivered a response to the query." - ] - }, - { - "cell_type": "code", - "execution_count": 7, - "id": "8b3ce09d-0eec-456d-8447-d04f79ccfc23", - "metadata": {}, - "outputs": [], - "source": [ - "def should_continue(state: MessagesState) -> Literal[END, \"check_query\"]:\n", - " messages = state[\"messages\"]\n", - " last_message = messages[-1]\n", - " if not last_message.tool_calls:\n", - " return END\n", - " else:\n", - " return \"check_query\"\n", - "\n", - "\n", - "builder = StateGraph(MessagesState)\n", - "builder.add_node(list_tables)\n", - "builder.add_node(call_get_schema)\n", - "builder.add_node(get_schema_node, \"get_schema\")\n", - "builder.add_node(generate_query)\n", - "builder.add_node(check_query)\n", - "builder.add_node(run_query_node, \"run_query\")\n", - "\n", - "builder.add_edge(START, \"list_tables\")\n", - "builder.add_edge(\"list_tables\", \"call_get_schema\")\n", - "builder.add_edge(\"call_get_schema\", \"get_schema\")\n", - "builder.add_edge(\"get_schema\", \"generate_query\")\n", - "builder.add_conditional_edges(\n", - " \"generate_query\",\n", - " should_continue,\n", - ")\n", - "builder.add_edge(\"check_query\", \"run_query\")\n", - "builder.add_edge(\"run_query\", \"generate_query\")\n", - "\n", - "agent = builder.compile()" - ] - }, - { - "cell_type": "markdown", - "id": "c377e9dd-92a7-45d4-91bf-408a6f2af8ca", - "metadata": {}, - "source": [ - "We visualize the application below:" - ] - }, - { - "cell_type": "code", - "execution_count": 8, - "id": "f728ed5f-7bcd-489e-9174-4e4725962335", - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "from IPython.display import Image, display\n", - "from langchain_core.runnables.graph import CurveStyle, MermaidDrawMethod, NodeStyles\n", - "\n", - "display(Image(agent.get_graph().draw_mermaid_png()))" - ] - }, - { - "cell_type": "markdown", - "id": "977173cf-8f8d-416e-a122-aebff115f29a", - "metadata": {}, - "source": [ - "We can now invoke the graph exactly as before:" - ] - }, - { - "cell_type": "code", - "execution_count": 9, - "id": "6c800a77-2359-4011-ab0c-dad22ed333b1", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "================================\u001b[1m Human Message \u001b[0m=================================\n", - "\n", - "Which genre on average has the longest tracks?\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "\n", - "Available tables: Album, Artist, Customer, Employee, Genre, Invoice, InvoiceLine, MediaType, Playlist, PlaylistTrack, Track\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Tool Calls:\n", - " sql_db_schema (call_qxKtYiHgf93AiTDin9ez5wFp)\n", - " Call ID: call_qxKtYiHgf93AiTDin9ez5wFp\n", - " Args:\n", - " table_names: Genre,Track\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: sql_db_schema\n", - "\n", - "\n", - "CREATE TABLE \"Genre\" (\n", - "\t\"GenreId\" INTEGER NOT NULL, \n", - "\t\"Name\" NVARCHAR(120), \n", - "\tPRIMARY KEY (\"GenreId\")\n", - ")\n", - "\n", - "/*\n", - "3 rows from Genre table:\n", - "GenreId\tName\n", - "1\tRock\n", - "2\tJazz\n", - "3\tMetal\n", - "*/\n", - "\n", - "\n", - "CREATE TABLE \"Track\" (\n", - "\t\"TrackId\" INTEGER NOT NULL, \n", - "\t\"Name\" NVARCHAR(200) NOT NULL, \n", - "\t\"AlbumId\" INTEGER, \n", - "\t\"MediaTypeId\" INTEGER NOT NULL, \n", - "\t\"GenreId\" INTEGER, \n", - "\t\"Composer\" NVARCHAR(220), \n", - "\t\"Milliseconds\" INTEGER NOT NULL, \n", - "\t\"Bytes\" INTEGER, \n", - "\t\"UnitPrice\" NUMERIC(10, 2) NOT NULL, \n", - "\tPRIMARY KEY (\"TrackId\"), \n", - "\tFOREIGN KEY(\"MediaTypeId\") REFERENCES \"MediaType\" (\"MediaTypeId\"), \n", - "\tFOREIGN KEY(\"GenreId\") REFERENCES \"Genre\" (\"GenreId\"), \n", - "\tFOREIGN KEY(\"AlbumId\") REFERENCES \"Album\" (\"AlbumId\")\n", - ")\n", - "\n", - "/*\n", - "3 rows from Track table:\n", - "TrackId\tName\tAlbumId\tMediaTypeId\tGenreId\tComposer\tMilliseconds\tBytes\tUnitPrice\n", - "1\tFor Those About To Rock (We Salute You)\t1\t1\t1\tAngus Young, Malcolm Young, Brian Johnson\t343719\t11170334\t0.99\n", - "2\tBalls to the Wall\t2\t2\t1\tNone\t342562\t5510424\t0.99\n", - "3\tFast As a Shark\t3\t2\t1\tF. Baltes, S. Kaufman, U. Dirkscneider & W. Hoffman\t230619\t3990994\t0.99\n", - "*/\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Tool Calls:\n", - " sql_db_query (call_RPN3GABMfb6DTaFTLlwnZxVN)\n", - " Call ID: call_RPN3GABMfb6DTaFTLlwnZxVN\n", - " Args:\n", - " query: SELECT Genre.Name, AVG(Track.Milliseconds) AS AvgTrackLength\n", - "FROM Track\n", - "JOIN Genre ON Track.GenreId = Genre.GenreId\n", - "GROUP BY Genre.GenreId\n", - "ORDER BY AvgTrackLength DESC\n", - "LIMIT 1;\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "Tool Calls:\n", - " sql_db_query (call_PR4s8ymiF3ZQLaoZADXtdqcl)\n", - " Call ID: call_PR4s8ymiF3ZQLaoZADXtdqcl\n", - " Args:\n", - " query: SELECT Genre.Name, AVG(Track.Milliseconds) AS AvgTrackLength\n", - "FROM Track\n", - "JOIN Genre ON Track.GenreId = Genre.GenreId\n", - "GROUP BY Genre.GenreId\n", - "ORDER BY AvgTrackLength DESC\n", - "LIMIT 1;\n", - "=================================\u001b[1m Tool Message \u001b[0m=================================\n", - "Name: sql_db_query\n", - "\n", - "[('Sci Fi & Fantasy', 2911783.0384615385)]\n", - "==================================\u001b[1m Ai Message \u001b[0m==================================\n", - "\n", - "The genre with the longest tracks on average is \"Sci Fi & Fantasy,\" with an average track length of approximately 2,911,783 milliseconds.\n" - ] - } - ], - "source": [ - "question = \"Which genre on average has the longest tracks?\"\n", - "\n", - "for step in agent.stream(\n", - " {\"messages\": [{\"role\": \"user\", \"content\": question}]},\n", - " stream_mode=\"values\",\n", - "):\n", - " step[\"messages\"][-1].pretty_print()" - ] - }, - { - "cell_type": "markdown", - "id": "fc3d8130-15c3-4c22-bb17-eb82ad41586e", - "metadata": {}, - "source": [ - "!!! tip\n", - "\n", - " See [LangSmith trace](https://smith.langchain.com/public/94b8c9ac-12f7-4692-8706-836a1f30f1ea/r) for the above run." - ] - }, - { - "cell_type": "markdown", - "id": "838b3acf-3186-4aa1-b211-cf136e95c39d", - "metadata": {}, - "source": [ - "## Next steps\n", - "\n", - "Check out [this guide](https://docs.smith.langchain.com/evaluation/how_to_guides/langgraph) for evaluating LangGraph applications, including SQL agents like this one, using LangSmith." - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3 (ipykernel)", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.10.4" - } - }, - "nbformat": 4, - "nbformat_minor": 5 -} diff --git a/docs/mkdocs.yml b/docs/mkdocs.yml index 020c6e28a..a66966905 100644 --- a/docs/mkdocs.yml +++ b/docs/mkdocs.yml @@ -109,6 +109,7 @@ nav: - Agent architectures: concepts/agentic_concepts.md - Guides: + - guides/index.md - LangGraph APIs: - Graph API: - Overview: concepts/low_level.md @@ -268,6 +269,7 @@ nav: - Environment variables: cloud/reference/env_var.md - Examples: + - examples/index.md - Template applications: concepts/template_applications.md # TODO: make tutorial - Agentic RAG: tutorials/rag/langgraph_agentic_rag.md - Agent Supervisor: tutorials/multi_agent/agent_supervisor.md @@ -286,6 +288,7 @@ nav: - Implement generative UI with LangGraph: cloud/how-tos/generative_ui_react.md - Additional resources: + - additional-resources/index.md - agents/prebuilt.md # NOTE: prebuilt.md is auto-generated by `make build-prebuilt` - LangGraph Academy course: https://academy.langchain.com/courses/intro-to-langgraph - Case studies: adopters.md From fb1c0ae9f9072363e1c934d259549ed131f1bcd1 Mon Sep 17 00:00:00 2001 From: Jake Broekhuizen Date: Tue, 8 Jul 2025 09:55:15 -0700 Subject: [PATCH 4/4] docs: Updating mcp_tools_node fn & referencing runtime graph rebuild docs (#5328) Fix: Updating mcp_tools_node fn & referencing runtime graph rebuild docs --- docs/docs/concepts/server-mcp.md | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/docs/docs/concepts/server-mcp.md b/docs/docs/concepts/server-mcp.md index 0d40bfb2e..b66eaada3 100644 --- a/docs/docs/concepts/server-mcp.md +++ b/docs/docs/concepts/server-mcp.md @@ -209,7 +209,7 @@ To make user-scoped tools available to your LangGraph Platform deployment, start ```python from langchain_mcp_adapters.client import MultiServerMCPClient -def get_mcp_tools_node(state, config): +def mcp_tools_node(state, config): user = config["configurable"].get("langgraph_auth_user") # e.g., user["github_token"], user["email"], etc. @@ -223,14 +223,19 @@ def get_mcp_tools_node(state, config): } }) tools = await client.get_tools() # (3) - return {"tools": tools} - + + # Your tool-calling logic here + + tool_messages = ... + return {"messages": tool_messages} ``` 1. MCP only supports adding headers to requests made to `streamable_http` and `sse` `transport` servers. 2. Your MCP server URL. 3. Get available tools from your MCP server. +_This can also be done by [rebuilding your graph at runtime](https://langchain-ai.github.io/langgraph/cloud/deployment/graph_rebuild/) to have a different configuration for a new run_ + ## Session behavior The current LangGraph MCP implementation does not support sessions. Each `/mcp` request is stateless and independent.
\n", - "

Set up LangSmith for LangGraph development

\n", - "

\n", - " Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started here. \n", - "

\n", - "