From 1edffe9f98cbf0f691dc65b4cf3fe8d981773d23 Mon Sep 17 00:00:00 2001 From: Peeeaje <74146834+Peeeaje@users.noreply.github.com> Date: Sat, 3 Jul 2021 00:26:27 +0900 Subject: [PATCH 001/368] Completed 1.1. --- .../1-intro-to-ML/translations/README.ja.md | 105 ++++++++++++++++++ .../translations/images/ai-ml-ds.png | Bin 0 -> 60304 bytes .../translations/images/hype.png | Bin 0 -> 154886 bytes 3 files changed, 105 insertions(+) create mode 100644 1-Introduction/1-intro-to-ML/translations/README.ja.md create mode 100644 1-Introduction/1-intro-to-ML/translations/images/ai-ml-ds.png create mode 100644 1-Introduction/1-intro-to-ML/translations/images/hype.png diff --git a/1-Introduction/1-intro-to-ML/translations/README.ja.md b/1-Introduction/1-intro-to-ML/translations/README.ja.md new file mode 100644 index 0000000000..568a11e34c --- /dev/null +++ b/1-Introduction/1-intro-to-ML/translations/README.ja.md @@ -0,0 +1,105 @@ +# Introduction to machine learning + +[![ML, AI, deep learning - What's the difference?](https://img.youtube.com/vi/lTd9RSxS9ZE/0.jpg)](https://youtu.be/lTd9RSxS9ZE "ML, AI, deep learning - What's the difference?") + +> 🎥 上ã®ç”»åƒã‚’クリックã™ã‚‹ã¨ã€æ©Ÿæ¢°å­¦ç¿’ã€AIã€æ·±å±¤å­¦ç¿’ã®é•ã„ã«ã¤ã„ã¦èª¬æ˜Žã—ãŸå‹•ç”»ãŒè¡¨ç¤ºã•ã‚Œã¾ã™ã€‚ + +## [Pre-lecture quiz](https://jolly-sea-0a877260f.azurestaticapps.net/quiz/1/) + +### イントロダクション + +åˆå¿ƒè€…ã®ãŸã‚ã®å¤å…¸çš„ãªæ©Ÿæ¢°å­¦ç¿’ã®ã‚³ãƒ¼ã‚¹ã¸ã‚ˆã†ã“ã! ã“ã®ãƒ†ãƒ¼ãƒžã«å…¨ã触れãŸã“ã¨ã®ãªã„方もã€ã“ã®åˆ†é‡Žã‚’ブラッシュアップã—ãŸã„経験豊富ãªæ–¹ã‚‚ã€ãœã²ã”å‚加ãã ã•ã„。ç§ãŸã¡ã¯ã€ã‚ãªãŸã®MLã®å­¦ç¿’ã«ã¤ã„ã¦ã®è¦ªã—ã¿ã‚„ã™ã„スタート地点を作りãŸã„ã¨è€ƒãˆã¦ã„ã¾ã™ã€‚ã‚ãªãŸã®[フィードãƒãƒƒã‚¯](https://github.com/microsoft/ML-For-Beginners/discussions)を評価ã—ã€å¯¾å¿œã—ã€å–り入れるã“ã¨ãŒã§ãã‚Œã°å¹¸ã„ã§ã™ã€‚ +[![Introduction to ML](https://img.youtube.com/vi/h0e2HAPTGF4/0.jpg)](https://youtu.be/h0e2HAPTGF4 "Introduction to ML") + +> 🎥 上ã®ç”»åƒã‚’クリックã™ã‚‹ã¨ã€MITã®John GuttagãŒæ©Ÿæ¢°å­¦ç¿’を紹介ã™ã‚‹å‹•ç”»ãŒè¡¨ç¤ºã•ã‚Œã¾ã™ã€‚ +### 機械学習を始ã‚ã‚‹ã«ã‚ãŸã£ã¦ + +ã“ã®ã‚«ãƒªã‚­ãƒ¥ãƒ©ãƒ ã‚’始ã‚ã‚‹å‰ã«ã€ã‚³ãƒ³ãƒ”ュータを設定ã—ã€ãƒŽãƒ¼ãƒˆãƒ–ックをローカルã§å®Ÿè¡Œã§ãるよã†ã«ã™ã‚‹å¿…è¦ãŒã‚ã‚Šã¾ã™ã€‚ + +- **ã“ã¡ã‚‰ã®ãƒ“デオã§ãƒžã‚·ãƒ³ã®è¨­å®šã‚’è¡Œã£ã¦ãã ã•ã„。** マシンã®è¨­å®šæ–¹æ³•ã«ã¤ã„ã¦ã¯ã€[ã“れらã®ãƒ“デオ](https://www.youtube.com/playlist?list=PLlrxD0HtieHhS8VzuMCfQD4uJ9yne1mE6)ã‚’ã”覧ãã ã•ã„。 +- **Pythonを学習ã™ã‚‹ã€‚** 本講座ã§ä½¿ç”¨ã™ã‚‹ã€ãƒ‡ãƒ¼ã‚¿ã‚µã‚¤ã‚¨ãƒ³ãƒ†ã‚£ã‚¹ãƒˆã«æœ‰ç”¨ãªãƒ—ログラミング言語ã§ã‚ã‚‹[Python](https://docs.microsoft.com/learn/paths/python-language/?WT.mc_id=academic-15963-cxa)ã®åŸºæœ¬çš„ãªç†è§£ãŒã‚ã‚‹ã“ã¨ãŒæœ›ã¾ã—ã„ã§ã™ã€‚ +- **Node.jsã¨JavaScriptを学習ã™ã‚‹ã€‚** ã“ã®ã‚³ãƒ¼ã‚¹ã§ã¯ã‚¦ã‚§ãƒ–アプリを構築ã™ã‚‹éš›ã«JavaScriptも何度ã‹ä½¿ç”¨ã—ã¾ã™ã®ã§ã€[node](https://nodejs.org)ã¨[npm](https://www.npmjs.com/)ãŒã‚¤ãƒ³ã‚¹ãƒˆãƒ¼ãƒ«ã•ã‚Œã¦ã„ã‚‹ã“ã¨ã€Pythonã¨JavaScriptã®ä¸¡æ–¹ã®é–‹ç™ºã«å¿…è¦ãª[Visual Studio Code](https://code.visualstudio.com/)ãŒåˆ©ç”¨å¯èƒ½ã§ã‚ã‚‹ã“ã¨ãŒå¿…è¦ã§ã™ã€‚ +- **GitHubã®ã‚¢ã‚«ã‚¦ãƒ³ãƒˆã‚’作æˆã™ã‚‹ã€‚** [GitHub](https://github.com)ã§ç§ãŸã¡ã‚’見ã¤ã‘ãŸã®ã§ã™ã‹ã‚‰ã€ã™ã§ã«ã‚¢ã‚«ã‚¦ãƒ³ãƒˆã‚’ãŠæŒã¡ã‹ã‚‚ã—ã‚Œã¾ã›ã‚“ãŒã€ã‚‚ã—ãŠæŒã¡ã§ãªã‘ã‚Œã°ã€ã‚¢ã‚«ã‚¦ãƒ³ãƒˆã‚’作æˆã—ã¦ã€ã“ã®ã‚«ãƒªã‚­ãƒ¥ãƒ©ãƒ ã‚’フォークã—ã¦ã”自分ã§ãŠä½¿ã„ãã ã•ã„。(スターをã¤ã‘ã‚‹ã“ã¨ã‚‚ãŠå¿˜ã‚Œãªã😊) +- **Scikit-learnを探索ã™ã‚‹ã€‚** ã“ã®ãƒ¬ãƒƒã‚¹ãƒ³ã§å‚ç…§ã™ã‚‹MLライブラリã®ã‚»ãƒƒãƒˆã§ã‚ã‚‹[Scikit-learn]([https://scikit-learn.org/stable/user_guide.html)ã«æ…£ã‚Œè¦ªã—ã‚“ã§ãã ã•ã„。 + +### 機械学習ã¨ã¯ä½•ã‹ï¼Ÿ + +"機械学習(Machine Learning)"ã¨ã„ã†è¨€è‘‰ã¯ã€ç¾åœ¨æœ€ã‚‚人気ãŒã‚ã‚Šã€é »ç¹ã«ä½¿ç”¨ã•ã‚Œã¦ã„る言葉ã®ä¸€ã¤ã§ã™ã€‚ã©ã‚“ãªåˆ†é‡Žã®æŠ€è¡“者ã§ã‚ã£ã¦ã‚‚ã€å¤šå°‘ãªã‚Šã¨ã‚‚技術ã«ç²¾é€šã—ã¦ã„ã‚Œã°ã€ä¸€åº¦ã¯ã“ã®è¨€è‘‰ã‚’耳ã«ã—ãŸã“ã¨ãŒã‚ã‚‹å¯èƒ½æ€§ã¯å°‘ãªãã‚ã‚Šã¾ã›ã‚“。ã—ã‹ã—ã€æ©Ÿæ¢°å­¦ç¿’ã®ä»•çµ„ã¿ã¯ã€ã»ã¨ã‚“ã©ã®äººã«ã¨ã£ã¦è¬Žã«åŒ…ã¾ã‚Œã¦ãŠã‚Šã€æ©Ÿæ¢°å­¦ç¿’ã®åˆå¿ƒè€…ã«ã¨ã£ã¦ã€ã“ã®ãƒ†ãƒ¼ãƒžã¯æ™‚ã«åœ§å€’ã•ã‚Œã‚‹ã‚ˆã†ã«æ„Ÿã˜ã‚‰ã‚Œã¾ã™ã€‚ãã®ãŸã‚ã€æ©Ÿæ¢°å­¦ç¿’ã¨ã¯ä½•ã‹ã‚’実際ã«ç†è§£ã—ã€å®Ÿè·µçš„ãªä¾‹ã‚’通ã—ã¦æ®µéšŽçš„ã«å­¦ã‚“ã§ã„ãã“ã¨ãŒé‡è¦ã§ã™ã€‚ + +![ml hype curve](images/hype.png) + +> Google Trendsã«ã‚ˆã‚‹ã€ã€Œæ©Ÿæ¢°å­¦ç¿’ã€ã¨ã„ã†è¨€è‘‰ã®æœ€è¿‘ã®ç››ã‚Šä¸ŠãŒã‚Šã‚’示ã™ã‚°ãƒ©ãƒ•ã€‚ + +ç§ãŸã¡ã¯ã€é­…力的ãªè¬Žã«æº€ã¡ãŸå®‡å®™ã«ä½ã‚“ã§ã„ã¾ã™ã€‚ホーキングåšå£«ã‚„アインシュタインåšå£«ã‚’ã¯ã˜ã‚ã¨ã™ã‚‹å‰å¤§ãªç§‘学者ãŸã¡ã¯ã€ç§ãŸã¡ã‚’å–ã‚Šå·»ã世界ã®è¬Žã‚’解ã明ã‹ã™æ„味ã®ã‚る情報を探ã™ã“ã¨ã«äººç”Ÿã‚’æ§ã’ã¦ãã¾ã—ãŸã€‚人間ã®å­ä¾›ã¯ã€å¤§äººã«ãªã‚‹ã¾ã§ã®é–“ã«ã€å¹´ã€…æ–°ã—ã„ã“ã¨ã‚’å­¦ã³ã€è‡ªåˆ†ã®ä¸–ç•Œã®æ§‹é€ ã‚’明らã‹ã«ã—ã¦ã„ãã¾ã™ã€‚ + +å­ä¾›ã®è„³ã¨æ„Ÿè¦šã¯ã€å‘¨å›²ã®äº‹å®Ÿã‚’èªè­˜ã—ã€å¾ã€…ã«äººç”Ÿã®éš ã‚ŒãŸãƒ‘ターンを学ã³ã€å­¦ç¿’ã—ãŸãƒ‘ターンを識別ã™ã‚‹ãŸã‚ã®è«–ç†çš„ãªãƒ«ãƒ¼ãƒ«ã‚’作るã®ã«å½¹ç«‹ã¡ã¾ã™ã€‚ã“ã†ã„ã£ãŸå­¦ç¿’プロセスã¯ã€äººé–“ã‚’ã“ã®ä¸–ã§æœ€ã‚‚æ´—ç·´ã•ã‚ŒãŸç”Ÿç‰©ã«ã—ã¦ã„ã¾ã™ã€‚éš ã‚ŒãŸãƒ‘ターンを発見ã™ã‚‹ã“ã¨ã§ç¶™ç¶šçš„ã«å­¦ç¿’ã—ã€ãã®ãƒ‘ターンã«åŸºã¥ã„ã¦é©æ–°ã‚’è¡Œã†ã“ã¨ã§ã€ç§ãŸã¡ã¯ç”Ÿæ¶¯ã‚’通ã˜ã¦è‡ªåˆ†è‡ªèº«ã‚’より良ãã—ã¦ã„ãã“ã¨ãŒã§ãã¾ã™ã€‚ã“ã®å­¦ç¿’能力ã¨é€²åŒ–能力ã¯ã€[「脳ã®å¯å¡‘性ã€](https://www.simplypsychology.org/brain-plasticity.html)ã¨å‘¼ã°ã‚Œã‚‹æ¦‚念ã«é–¢é€£ã—ã¦ã„ã¾ã™ã€‚表é¢çš„ã«ã¯ã€äººé–“ã®è„³ã®å­¦ç¿’プロセスã¨æ©Ÿæ¢°å­¦ç¿’ã®ã‚³ãƒ³ã‚»ãƒ—トã«ã¯ã€ãƒ¢ãƒãƒ™ãƒ¼ã‚·ãƒ§ãƒ³ã®é¢ã§ã„ãã¤ã‹ã®å…±é€šç‚¹ãŒã‚ã‚Šã¾ã™ã€‚ + +[人間ã®è„³](https://www.livescience.com/29365-human-brain.html)ã¯ã€ç¾å®Ÿä¸–ç•Œã®ç‰©äº‹ã‚’知覚ã—ã€çŸ¥è¦šã—ãŸæƒ…報を処ç†ã—ã€åˆç†çš„ãªåˆ¤æ–­ã‚’下ã—ã€çŠ¶æ³ã«å¿œã˜ã¦ã‚る行動をã—ã¾ã™ã€‚ã“ã‚Œã¯çŸ¥çš„行動ã¨å‘¼ã°ã‚Œã¾ã™ã€‚ã“ã®çŸ¥çš„行動ã®ãƒ—ロセスを機械ã«ãƒ—ログラムã™ã‚‹ã“ã¨ã‚’人工知能(AI)ã¨ã„ã„ã¾ã™ã€‚ + +ã“ã®è¨€è‘‰ã¯æ··åŒã•ã‚Œã‚‹ã“ã¨ãŒã‚ã‚Šã¾ã™ãŒã€æ©Ÿæ¢°å­¦ç¿’(ML)ã¯äººå·¥çŸ¥èƒ½ã®é‡è¦ãªã‚µãƒ–セットã§ã™ã€‚**MLã¯ã€ç‰¹æ®Šãªã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã‚’使用ã—ã¦ã€æ„味ã®ã‚る情報を発見ã—ã€çŸ¥è¦šã•ã‚ŒãŸãƒ‡ãƒ¼ã‚¿ã‹ã‚‰éš ã‚ŒãŸãƒ‘ターンを見ã¤ã‘ã¦ã€åˆç†çš„ãªæ„æ€æ±ºå®šãƒ—ロセスをè£ä»˜ã‘ã‚‹ã“ã¨ã«é–¢ä¿‚ã—ã¦ã„ã¾ã™ã€‚** + +![AI, ML, deep learning, data science](images/ai-ml-ds.png) + + +>[ã“ã®ã‚°ãƒ©ãƒ•](https://softwareengineering.stackexchange.com/questions/366996/distinction-between-ai-ml-neural-networks-deep-learning-and-data-mining)ã«è§¦ç™ºã•ã‚ŒãŸ[Jen Looper](https://twitter.com/jenlooper)æ°ã«ã‚ˆã‚‹ã‚¤ãƒ³ãƒ•ã‚©ã‚°ãƒ©ãƒ•ã‚£ãƒƒã‚¯ + +## ã“ã®ã‚³ãƒ¼ã‚¹ã§å­¦ã¶ã“㨠+ +ã“ã®ã‚«ãƒªã‚­ãƒ¥ãƒ©ãƒ ã§ã¯ã€åˆå¿ƒè€…ãŒçŸ¥ã£ã¦ãŠã‹ãªã‘ã‚Œã°ãªã‚‰ãªã„機械学習ã®ã‚³ã‚¢ãªæ¦‚念ã®ã¿ã‚’å–り上ã’ã¾ã™ã€‚ç§ãŸã¡ãŒã€Œå¤å…¸çš„ãªæ©Ÿæ¢°å­¦ç¿’ã€ã¨å‘¼ã¶ã‚‚ã®ã‚’ã€å¤šãã®å­¦ç”ŸãŒåŸºç¤Žã‚’å­¦ã¶ãŸã‚ã«ä½¿ç”¨ã™ã‚‹å„ªã‚ŒãŸãƒ©ã‚¤ãƒ–ラリã§ã‚ã‚‹Scikit-learnを主ã«ä½¿ã£ã¦ã‚«ãƒãƒ¼ã—ã¾ã™ã€‚人工知能や深層学習ãªã©ã®ã‚ˆã‚Šåºƒã„概念をç†è§£ã™ã‚‹ãŸã‚ã«ã¯ã€æ©Ÿæ¢°å­¦ç¿’ã®å¼·åŠ›ãªåŸºç¤ŽçŸ¥è­˜ãŒä¸å¯æ¬ ã§ã™ã®ã§ã€ã“ã“ã§æä¾›ã—ã¾ã™ã€‚ + +- 機械学習ã®æ ¸ã¨ãªã‚‹ã‚³ãƒ³ã‚»ãƒ—ト +- MLã®æ­´å² +- MLã¨å…¬å¹³æ€§ +- MLã«ã‚ˆã‚‹å›žå¸°ã®æ‰‹æ³• +- MLã«ã‚ˆã‚‹åˆ†é¡žæŠ€è¡“ +- MLã«ã‚ˆã‚‹ã‚¯ãƒ©ã‚¹ã‚¿ãƒªãƒ³ã‚° +- MLã«ã‚ˆã‚‹è‡ªç„¶è¨€èªžå‡¦ç†ã®æŠ€è¡“ +- MLã«ã‚ˆã‚‹æ™‚系列予測ã®æŠ€è¡“ +- 強化学習 +- MLã®ç¾å®Ÿä¸–ç•Œã¸ã®å¿œç”¨ +## ã“ã®ã‚³ãƒ¼ã‚¹ã§æ‰±ã‚ãªã„ã“㨠+ +- ディープラーニング +- ニューラルãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ +- AI + +ニューラルãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã‚„ディープラーニング(ニューラルãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã‚’用ã„ãŸå¤šå±¤çš„ãªãƒ¢ãƒ‡ãƒ«æ§‹ç¯‰ï¼‰ã€AIãªã©ã®è¤‡é›‘ãªåˆ†é‡Žã¯ã€ã‚ˆã‚Šè‰¯ã„学習環境をæä¾›ã™ã‚‹ãŸã‚ã«é¿ã‘ã¦ã„ã¾ã™ãŒã€ã“れらã¯åˆ¥ã®ã‚«ãƒªã‚­ãƒ¥ãƒ©ãƒ ã§å–り上ã’ã¾ã™ã€‚ã¾ãŸã€ãれらã®å¤§ããªåˆ†é‡Žã®ä¸­ã§ã‚‚特ã«ãƒ‡ãƒ¼ã‚¿ã‚µã‚¤ã‚¨ãƒ³ã‚¹ã«ç„¦ç‚¹ã‚’当ã¦ãŸã‚«ãƒªã‚­ãƒ¥ãƒ©ãƒ ã‚’æä¾›ã™ã‚‹äºˆå®šã§ã™ã€‚ +## ãªãœæ©Ÿæ¢°å­¦ç¿’ã‚’å­¦ã¶ã®ã‹ + +機械学習ã¨ã¯ã€ã‚·ã‚¹ãƒ†ãƒ ã®è¦³ç‚¹ã‹ã‚‰ã€ãƒ‡ãƒ¼ã‚¿ã‹ã‚‰éš ã‚ŒãŸãƒ‘ターンを学習ã—ã€çŸ¥çš„ãªæ„æ€æ±ºå®šã‚’支æ´ã™ã‚‹è‡ªå‹•åŒ–ã•ã‚ŒãŸã‚·ã‚¹ãƒ†ãƒ ã‚’構築ã™ã‚‹ã“ã¨ã¨å®šç¾©ã•ã‚Œã¾ã™ã€‚ + +ã“ã®å‹•æ©Ÿã¯ã€äººé–“ã®è„³ãŒå¤–ç•Œã‹ã‚‰èªè­˜ã—ãŸãƒ‡ãƒ¼ã‚¿ã«åŸºã¥ã„ã¦ç‰¹å®šã®äº‹æŸ„を学習ã™ã‚‹ä»•çµ„ã¿ã«ã€ã‚†ã‚‹ã‚„ã‹ã«ã‚¤ãƒ³ã‚¹ãƒ‘イアã•ã‚Œã¦ã„ã¾ã™ã€‚ + +✅ ãªãœãƒ“ジãƒã‚¹ã§ã¯ã€ãƒãƒ¼ãƒ‰ã‚³ãƒ¼ãƒ‰ã•ã‚ŒãŸãƒ«ãƒ¼ãƒ«ãƒ™ãƒ¼ã‚¹ã®ã‚¨ãƒ³ã‚¸ãƒ³ã‚’作るã®ã§ã¯ãªãã€æ©Ÿæ¢°å­¦ç¿’戦略を使ã£ã¦ã¿ã‚ˆã†ã¨æ€ã†ã®ã‹ã€ã¡ã‚‡ã£ã¨è€ƒãˆã¦ã¿ã¦ãã ã•ã„。 + + +### 機械学習ã®å¿œç”¨ + +機械学習ã®ã‚¢ãƒ—リケーションã¯ã€ä»Šã‚„ã»ã¨ã‚“ã©ã©ã“ã«ã§ã‚‚ã‚ã‚Šã€ã‚¹ãƒžãƒ¼ãƒˆãƒ•ã‚©ãƒ³ã‚„コãƒã‚¯ãƒ†ãƒƒãƒ‰ãƒ‡ãƒã‚¤ã‚¹ã€ãã®ä»–ã®ã‚·ã‚¹ãƒ†ãƒ ã‹ã‚‰ç”Ÿæˆã•ã‚Œã€ç§ãŸã¡ã®ç¤¾ä¼šã«æµã‚Œã¦ã„るデータã¨åŒæ§˜ã«ã‚ã‚Šãµã‚ŒãŸã‚‚ã®ã¨ãªã£ã¦ã„ã¾ã™ã€‚最先端ã®æ©Ÿæ¢°å­¦ç¿’アルゴリズムã®è¨ˆã‚ŠçŸ¥ã‚Œãªã„å¯èƒ½æ€§ã‚’考慮ã—ã¦ã€ç ”究者ãŸã¡ã¯ã€å¤šæ¬¡å…ƒçš„・多分野的ãªç¾å®Ÿã®å•é¡Œã‚’解決ã™ã‚‹ãŸã‚ã«ãã®èƒ½åŠ›ã‚’探求ã—ã€éžå¸¸ã«è‰¯ã„çµæžœã‚’å¾—ã¦ã„ã¾ã™ã€‚ + +**機械学習ã¯æ§˜ã€…ãªå½¢ã§åˆ©ç”¨ã§ãã¾ã™**: + +- 患者ã®ç—…歴や報告書ã‹ã‚‰ç—…æ°—ã®å¯èƒ½æ€§ã‚’予測ã™ã‚‹ã€‚ +- 気象データを活用ã—ã¦æ°—象ç¾è±¡ã‚’予測ã™ã‚‹ã€‚ +- 文章ã®æ„Ÿæƒ…ã‚’ç†è§£ã™ã‚‹ã€‚ +- プロパガンダã®æ‹¡æ•£ã‚’防ããŸã‚ã«ãƒ•ã‚§ã‚¤ã‚¯ãƒ‹ãƒ¥ãƒ¼ã‚¹ã‚’検出ã™ã‚‹ã€‚ + +金èžã€çµŒæ¸ˆã€åœ°çƒç§‘å­¦ã€å®‡å®™é–‹ç™ºã€ç”Ÿç‰©åŒ»å­¦å·¥å­¦ã€èªçŸ¥ç§‘å­¦ã€ã•ã‚‰ã«ã¯æ–‡ç§‘ç³»ã®åˆ†é‡Žã§ã‚‚ã€ãã‚Œãžã‚Œã®åˆ†é‡Žã®ãƒ‡ãƒ¼ã‚¿å‡¦ç†ã«ä¼´ã†å›°é›£ãªå•é¡Œã‚’解決ã™ã‚‹ãŸã‚ã«ã€æ©Ÿæ¢°å­¦ç¿’ãŒæŽ¡ç”¨ã•ã‚Œã¦ã„ã¾ã™ã€‚ + +機械学習ã¯ã€å®Ÿä¸–ç•Œã®ãƒ‡ãƒ¼ã‚¿ã‚„生æˆã•ã‚ŒãŸãƒ‡ãƒ¼ã‚¿ã‹ã‚‰æ„味ã®ã‚る洞察を見出ã—ã€ãƒ‘ターンを発見ã™ã‚‹ãƒ—ロセスを自動化ã—ã¾ã™ã€‚機械学習ã¯ã€ãƒ“ジãƒã‚¹ã€å¥åº·ã€é‡‘èžãªã©ã®åˆ†é‡Žã§éžå¸¸ã«æœ‰ç”¨ã§ã‚ã‚‹ã“ã¨ãŒè¨¼æ˜Žã•ã‚Œã¦ã„ã¾ã™ã€‚ + +è¿‘ã„å°†æ¥ã€æ©Ÿæ¢°å­¦ç¿’ã®åŸºç¤Žã‚’ç†è§£ã™ã‚‹ã“ã¨ã¯ã€æ©Ÿæ¢°å­¦ç¿’ã®æ™®åŠã«ä¼´ã„ã€ã‚らゆる分野ã®äººã€…ã«ã¨ã£ã¦å¿…é ˆã®ã‚‚ã®ã¨ãªã‚‹ã§ã—ょã†ã€‚ + +--- +## 🚀 Challenge +AIã€MLã€æ·±å±¤å­¦ç¿’ã€ãƒ‡ãƒ¼ã‚¿ã‚µã‚¤ã‚¨ãƒ³ã‚¹ã®é•ã„ã«ã¤ã„ã¦ç†è§£ã—ã¦ã„ã‚‹ã“ã¨ã‚’ã€ç´™ã‚„[Excalidraw](https://excalidraw.com/)ãªã©ã®ã‚ªãƒ³ãƒ©ã‚¤ãƒ³ã‚¢ãƒ—リを使ã£ã¦ã‚¹ã‚±ãƒƒãƒã—ã¦ãã ã•ã„。ã¾ãŸã€ãã‚Œãžã‚Œã®æŠ€è¡“ãŒå¾—æ„ã¨ã™ã‚‹å•é¡Œã®ã‚¢ã‚¤ãƒ‡ã‚¢ã‚’加ãˆã¦ã¿ã¦ãã ã•ã„。 + +## [Post-lecture quiz](https://jolly-sea-0a877260f.azurestaticapps.net/quiz/2/) + +## 振り返りã¨è‡ªç¿’ + +クラウド上ã§MLアルゴリズムをã©ã®ã‚ˆã†ã«æ‰±ã†ã“ã¨ãŒã§ãã‚‹ã‹ã«ã¤ã„ã¦ã¯ã€ã“ã®[ラーニングパス](https://docs.microsoft.com/learn/paths/create-no-code-predictive-models-azure-machine-learning/?WT.mc_id=academic-15963-cxa)ã«å¾“ã£ã¦ãã ã•ã„。. + +## 課題 + +[èµ·å‹•ã—ã€å®Ÿè¡Œã—ã¦ãã ã•ã„。](assignment.md) diff --git a/1-Introduction/1-intro-to-ML/translations/images/ai-ml-ds.png b/1-Introduction/1-intro-to-ML/translations/images/ai-ml-ds.png new file mode 100644 index 0000000000000000000000000000000000000000..11c9bf3767955eedb662fa6a4430064babeec5cc GIT binary patch literal 60304 zcmb@tWmsE5*Djm{PjP~~7I%ufLvi=uQrw*&h2rk+S_l*<#VJs%#T^Qi7I$yaLmzve z>;2C8_hqhR&z@Q9UNURXn#?58>ZU=lC@0KiaGkkJGHfVI!>Bvj<*7VJ7Tgy##Q zt)z-10MMBD;@%wT`JT#BK~n_)@Mi!3U=aYo?Q;`s7Xa|$1OWD-0Dw>y06^%P*RCP* z+!0}|uV|y90$_fwqXLkD$bi3yKmh3Z1NUEb;B%b>@E7^H2G9kP{-Pde|Kk}0eKVQ&X6%0H80LZic5EWqBXgmi1 z0BW_@*7wv`c`an=;>2!lx)7+QM*@O0kG~f{|H&%t;^^Y8QZ?X91O^&i~Nw*Eu>f4BXMUWDWC$o|(z z|1*sK(fS-mVqg)D|HPyi_-UG37yys}D9T7``vMPxP+%tVe#apJE8VpwiXF_g?8?q5 zTWgJbuNaVLr0k;oiX$RKP!r`*u}o1b5GnZRfN@O4*%qkC;qd{f9c;BXbM>tY9b5A^ zK_16-$MvlzY&&bhwE3bt1>Cmd;yh3MA=x2higa8f>b?IiUm&i_ZH^Uok)VH<7!Y#Z zYiUl%KMPzG30;4wJLWYI0}1_~ML+u$l<{9B089jL3c}9@k${Q*S>W7G{{g>-a6+x` zn7NU7|AqE^y9_J-S2<#W_KsfbSb^>Sjn+Hb43z#?S<&i+r61J}Lni;NDlFX|;onNO zR>FWRD|*RHMgK+vvdkd=OF<+@H6Vg+YXgmz#J|xHbSp6ar672*;Mqb)GuaB*ztNrz zCjI}(VAqxIhzc0JeCRJ$v#$r3j<;e#UmsrEw;u3_zNsgEcG0g)Py(H1a4dAq^bnG@ z+RO07@a`9hf_`N6Xz>Bf!q(IYU%<20q1Dgcad2qQ`tYI%X7lLZgN?1U(Y0zZOc&{< zjXvZ?Hsn^<&}er{E@CsiA=TB2AZ{%1Q#SwyOZ7BsVY~KbJ79Xfe7*Fh%`xTKLlT&b z&|YQ~8TJajG9Krk*Iz{OE)stFBSLxd6h;{-z?R|wM3Ua%h@r(>%{$Ix(tc6tMXYnv z!~R7qOHu?!##5=_z1~5e4$WLr`m-Mb?^3B-1k2FxxB(fIcU0H-s!`hJwF3kI|8(Je z%{}+3H=(nFl*g6@GxQK!vVo){UnOx0aeK_a!$Y9*JT$Qk=5Dr}<7sYyM^QFLoFr7K za6oaljFb>H$<+Yfbj(H+F-T%N_jeNqa3`|71!4`ng?mr+RI`1i3 zN8_#jj-458w6HVw7vJ zkYxii$`chwI@|i2zmDlG`RnE>I@SX8Vax2xVCHn<2Anj1pjdumM63Z#K!XnaXESu+mu3fU% zp7lwJc95&dw!cNe0tU7NrJOVXxUC3_*MO{ouuB;zi9EJQC&6JGZa~csMl>$*d~smt zbi(Jx<_iH4HWi(O8E3;@v<&N-ft$T}XlYqnvqxa${t#+wu$4Q^temGm{2QQ&JgMy! zLfgEbanrJ8s6KgcCPFVm2o~j5#d@@K>fb@KDSlSQ09D&Dldtrbk4gpzl_4l8OZCg9 zJ0nOvW?;(zHWLnDsX6s5si~vLdj-R@I_%_5XH#4^V-1*A4$nsBopywf(l?gcPg}7` z>^lG!mjHDtrkQnva!hq3r1WE#Wd}0hIct^cme0m3!x!2+Eiylw z3VS1txL`9T0+n;VwWkuIz0uRSg>$&L*}VMQ#ajMrx5iw@FGF~MaSkdN(EDvJ$M1lK z@LL<~9c}@4V?gx*S%cOqP^YdGxE9Ix?9-(TNoq#9)WoF=|4<}oMxz%t=b6T|agihL zE^gpYf{fsVM0G?SJ2qp<Yvp>Gqscljdd-y=hx1EeDKm+oN zcb-UEQ$Rja^?{IrA>rNkra`O#!h{(o6Yx*XaC8xelwygg=jm7nsLMr6&VO2iR#S`# zn2t>2A#H0@M-Ie-4EwWuMxt5~8@{jZR@GURGcKJ2a=@BZt{4Taai zm@t1{Bz4;gK+MFXQM_gvu;O!-<2>;}mMR-!xw)g7`sU{lf8uVun$NWe&VnyA2}RMp zHa%9+UF~DCg1DaDlm6|E2QxCHZzLQpDJnviDoiE&U6K?sy#%#NU&PC5Ml8=JY{=Et zhfI3Odr?cqrVye~Bv!%>9IL=CVL=DXfCXYez2^=(ty8Ev13v!WjMyy1r;ty=AM5GL z&8`#JuknX@TXL!^5P^rjcI2bE5e1_hTpP1@f0J&iVeVH2qU~K` z1&tm|`K=FdS&bMW7VD^zm_$v;43nbVGL=YlJ4XXW53x%p7O^45kG|#m2|BFu0?(Mv1{*~cLxLMP~zEKznZrrahr&+ zy@DU_0$=QR-$8RNTxS^o_jqfip#B&6_5PA`+U}> zf}|UR2d`6CO1$Rp`5#w)X85Q&`n~?g=ZqvZ+GhfSHWJbStWy z&PR}+Rk9ARvG+?+be-9l0pYK^pwOOAFeM3LQOP1pC2bDR(@BRKtHYQjoI8Mq`F2Go z&ls)l4S85m=w=QwgV$9nPVA#rYvzocc+XhhWb$Q=bI~Lp z*q~}QE89oE!D--4BmsYo1v!>FhF{W^(IS&`uf0$_;CMC@U?^0U3ViREJhz%h|GZcm z1)0wX^P5aQAx?C2^s#$kl`+@cN%_8%-c2&dvSLhf0mI3skMfJIcqL+|HpIh*iCn5r zM|BUgJ&pLnnPo9AUSd7RSGr_LrH`GGa-!tijf?u9AS{iEmY<09OsboF`C{P_{07z$ zAjcm6s~way+DPecfYJN`mAzZB0t4p2qPNN&J>Tn5`oCU+Eg$|ON=o3E>C^FXAtg$E zihYOYDQ^Aa*yG?+ORkaW>32At^A12xyUeod$-EXxi-XbYJG7nQ`M8u61bV1-$msOMQF*q2H$XmT z3TnG4wD9BO8-)BEPO3Z5Y*@o*0;$|huZm90@j9BU9$6(IPrL;M{upMpT=F7b9FgDV zZ2}y`Tysz4X92r`j=iZ=p&R&drTEn4GZuQZqd4@jn<+Zk+^y}$lC1sMbM*OX!j7^f z-9NF8(Aym-H-;b}+n<~eW$a+PW-okdJj#8J1Pn0Ih(m%lee42YHVj+wX-%BDFbn}H zbDwbtNlB|!0g<`T>@^`9tlhiW9Z#ALXmh7pnh>NOHGpqgYg zyiZN|dOZx`%kw12ijuVVBk7pq@51!&>Wi*cFiBlm=`g~_QuE|0-yBIhv((aml2(vo zkbud$R4E}|=}xbwEGshLhybl_otfP1q3-f=TW|AHBH6Nx#YkJPm>`U5Kx`^8zM$S1 zf5RlJx5_3NUZv`Z90mrU zg}qn~leA=+@zRb62*U=(YU-w@G|8;R2u4yyT>qfjCQa$v4RE!4_GQ>#Uot6Fnz-2? z{2`7Tb_tEKL`cXiS(xf|9MiB+sv+@ruO##RQ;sWeBSjp1t6^~QVO?=f*^J?6{dZ0C zL;s!5(!$ymJYxJiX61|5It8lj?gG}Ay<*W`K0wEfcWThwgQ(PiGgHn6D_(}jZN5!| zcI_0dpYfCh-OA})!Kd#cT91=9w%Jn1VgYbdrteQT zJh;Rn)S@4sWI(o#`6N}nR%8^3GHR_O#VoDWir?;SR7lBcc>nOLF$XrWCqB_{xZEBg zY&tnK(FxTcsxT{o;WkU`jMf@D<9th|p<z0Dx(p)x;>-qubZ zQ?umI+@+Dl&^A#Fm_J_JAYu&*0CebS)!s@^P{tg!!ZMYRI{oxXEyngrE*d>AEoNbz z?s!=!S~~2U)g^5?F$mF&diw4g$H;KE=xCP-!%8`-E<1l&%>%Q0C<=sH<@b?)>B}t2 zYtHqYZ<}w{=pHd_V?qvI$?Uc2W7}7MO#c`t8)h_-GC)*i2Lq=uJq~!6yd3!O6iBpx z;bR#yL{Wn@C9|lsDE1s?Sy)JMR!9!^SBS#c{q)}_DU($MwpW=FaGhRgGUlpcW@<&6 zt(L?1^fiiAqgsyur$v{F!RDIG4Z+o(Kij`@`(;CDb3l9|J782uD7_d_zh%ZF(RKJ@6#_2Q=A!mK;R zbHQm5Ja(v}xsZE+aQ}dGv-@4+Wiex=OV0InPB0{F z5)S9c?bjlyQni@9F^{`N)z&sr-_qlhkOdSs&n&!@?s>$R4UWe>5Mj4&=vFnJo2$$t z@hx1T-^w(uV3hvC^?MSbH;Fh00JJ$rJ{vXAWvJ(w(8^-~Wu`_NIVYdL;E;QCz7hjL z^7+#OMAmI!N>iQMPp-W~Iw)C-J&z>GW@5{HtmIjkVTE9LUF~*5aqlH0D^yG6Nxj(Y zhe9H$craF6j|+l&5JrGEol%5x4pt~W>L+^l5?Q&#BOE=gQq7*<{LY#i=GA~?YC!m! z?;}R;29s~{?RCLPuu+NV>2WY|Wxf=#zplPT zRZW|%NSXJPvoX_rPM!8GRCh~+=2*d`72)tF9Jps8D^p2UVm={~5@y{Z)1q6Os1R;? zE!St^ch=i2K`}co*ngDna`w59W?EFmgb$<-?CJ<=0A>YxMol!BtT_Tcwfp?4jRTV0WEhVOX_2-9*!iZ;{Y1{UO*Yh_~n<9V_>QQgA<&pGCC*s2iwE*-~gnurw zk$*%T%8jH1uH#itM)i@xchk73;@z-9@JEwoz$fDZkQsyMLQj?Y)kKRoTj)`cH_}n? z*H4038Ve}g(lcZGPxQ_}Op=1ql4ug`)6A2=w)P7@dL*u%C0tRgX#DOE)|zNyEik z*1dqSaPjC8sFBh6U~~4~)Jf3rtA@}yUjs2E+E74yx;7t56^jgRF1Kwc2$2;%S=;fO z@??tX3P1A`nqq7MVIE2Hn%Rd_|5R(BW)WSl5`Q3Q=Hn#OTD-<;Ie^)lpnyfYs6(*byBZyro20}P5Au~14#j>^h>+{gIO>U#izo3ieyO`K{nn>{qt~@NoXkR zl2pl2hXyv(#c#?AUrHA{37Ir;M)1y%1OP(MDBAA7Dnu!wGyOSEeZN_ zLBQP554VA2Qj8AS)ZHU~A)3;F8)iPU!vX5*_CE+&W;)?&p6>{$GiDJ@1f);-Gg~U< zp`apygTu6p#2^Kl8_|+soM4Mea4fH0QSnY;$TuOrTk@qpZ7E+@`S@HD`7M8x*qaGH zVq;RE>Fa@ty1|*(IDmQEeUN4pBWt_(PMgBE*Q*ioO!kr209~L>~Grw|xIgE*Cm?tu-DNf*qEF>SiQGUA!zYZu67QDkr{3wyp1x|zdP1N zd)}C~pNOlY0mSc)T36CB77;1yr4i(0e5}Ao@`gj?(d%iMpr4&8>G3~ewK5^+zQ4^4 zN!sluoE7b#5cOE@(kg(ixmFcLY_I!Ql=wX*23Ku+RofAz(tc9Qjuy0MvC6*Q>KX1-l3pY+cLm44ayj9bWQKvk-R;fu*5H6bdmi^7Q!m|@`~(E$_yJ;p}lPsFxoNh z^FvGS(@Jr}u>|#)pIut?`9sjUBWu=IDY@ponq$U=#`ueva z*FojFRuao-(C~mg0`YD)cP`lI54YILZpnDhAkvAH`I}A^Oq)wYV zq)aA4EBC!(PiTB@DA$Yv?|yx=cpycM#CT>&>0opk%iZFc<)%@Kp<>nk@oW%I9a$X! zy+;toKvw?-1^dckwuDEQ$y-T1Vf8n6b>Ns8GxpDRedFMWqN+{g5=XJc*_wtYo~t!M zMVP(ptEh5CAe*^{);T_6FIyDbE#;5-`V1nVVvNOrNsf+iLj$`noW_Mgyo>U>niM;K z6PMSHu}PI{#V@(E*E!OOIbl*UZ`Js0nV;!Igx#qDm3v5yx+u-eGupj*W59g;8%dNO zcIK<9W3&6$hHE#Ne4$xOxM{|=776&XL5;SBgbfY4xI@bW8m80*>Rp#l8{$KIY6^3t z=4r|e&nQfJJPyE&cZGeUJkQzkdGj*q7LEfKz^Vnm+QXVhoK1`O{5HzElr@I*Iy4NK zrK%yl@9Ga1t2mbMoS)q=I;BjyzQ%`Mh?Ew$BKq68n+!9r__onet|)M$pdC++^5-Dq zfYFs^&HbL=C+?Rg&IJ69ra#ry;9*J=XlOCKS9u}X186p!on&Fewp>`i0HOAEB7 z2GZV~FsQ6|4YV)jE1r{%At%~Z2z2_yrdgLinDA>!bgZ*NHTx+2ZnB=;ir+4r0Q!~x zjga!}s%t4?BVAXI-+Rp!`{ctMo zoJ~(gta-kq=naI$XPOT~UyN~?%1hq-8F-TWT_!z!|Ng`P7xVj&CdB`h%zK#*(+&ggKC zZj-X!HyA&eZsm9YKkvz9aUcmu(8`F$UH@@)l3)k>&c^hU_2y{jK^O(#n^pNo$uO@M zBH?1AyAafgX|-UZJTKlWJs??|h0BfF;|qsQe9ho9Yi6avJRMl72R+n{c2Wgd7(^iQ zTh$NELh}?|lgya2PU&-jW-jup^e1DTwA3-%U{Y|I1}qQ;m2DA!Fh4QHO~Wv)JBs|7 zvWf6}O;Zu+g|7J9g>@BIt;)w7|HBk=bsQ)vVAYm~>*>-b4lnS&*FRuclr_0s2p?ec zp^o7sr3kR12Uu}y{^app_0N0?_G}o#sZ*nFuq1}muDay*N4zx&eqi11AO3gisf*T) z^K-N7vQ!EP@2Xu!dkwmyimLeK^z}%`-*D!5V$oVx-TAFO?`)uJ!QML%8SyMHkWWPf zQm`N4oT75`fwC0`WkbV~>zaMn!BH`!!jW_S{ML5wJwLvxgSU~C{~?hMeNNDw#kea$ zMBjodSEJ@4iA|TvEJ^_=)U9e{^o>R6s&@^p8X=JS89wAHM>Chd$A_Pa0DcI3teKKC z=LyAJegJPrhOjKfLFv%2q&igHw1Qv&>0F`o!;rd?rx_ov&Ka-F1^qwvBX@qNV+bH$ zw&P(E){1=jVbRFjGIP6ZFlB?EAVrR3X%dbkFLWM27EK@Z->L z_>B;^V8l;yZgrsv$$G2%XfI|_ZEnd~8-Uypf+z!Htfso*c)hzK!4H#Zzhh)81 zmXDWFS9XXONhtU4#*LU;Nat|w+7sq zWCl$kO{r#5XcQo1svs{3Q_r@9Z17ej05QOMjANXXFJe@IBUUhtk5;>h$ife=Y~JM0 z^h(>urDPZ>TGImRA7vCiBk~Lf;Q)-qVU9#MaT{4`erMlI$@W|t0bJ5Z>?w{4mE4Y4 z&|l|&lH?yLaF$h=9l?vZb-ao_VA{RA#w40>U6QN93MKo)mR~Xz(l)P+G* z_)5^yq%X)R)O*T^)Le*Vcs*!~Xp;k(AV)So4VL7NhMeDQ-nX(!8I~QCuq(hOiYbZA zy~O>ieF@>Y>D4g{VR#8R3VHA4j|U&q1-LON3e4(nu8F2DAq^%zBGMlvBH~(N&vwHY zXGP+ivtaRvphxs7AWTV9)o!pyet$FH65Vt8;JvDb%rhMx9WeG;)`&oIg;sims*Iz@ z^W|t6p8-Rc(tEpVtOs>BO|RjdEa@NW5<$*#Pn`QtfOnIqUK>~k;Z2Ad%NS7Wv;o1f zue#8|1lem|!~uVflInHx_%ZO$hs+OOk`4$P>-i)!Bnf{Y!)QL3fbFX#6;(==9B3r` zs063>ErBykX3n8FdvbRRRDd@`)ANk5I(iLia)E@tYuTcO`rlvW zS){y72fKIp-*|0g7Gd7QG*Ac}(^qiof+M4Q8io*|rH?F&-h7sQE$z&5duU$?7!)~m zRF#R+Mg z7e9w{cT|`I!FEhBP#*9rNmxx}nKlZn#x+5y>jRc)K=|mf)d+f8@$WQMl8Q^04-1gV z-trKw^UW_8rAqr?#ehx{*j6eY+-K^7(?mxgd0Z%SbD77Gs`P_dbQOV_yl1TYB8fub zC4sK7nQ@?*B`JKm%=*_aEc20lWv%-C3NHIm$eI{=QCc{nemIJlOFW#P512<{EIJyL z7nknXZ_zmK`*D`Oa&Z1t@%NG*S)I{J#PVPTLz8ItDsU~pSTgVJGJXh~xb#86VE0PG z#Pez4hrvXaAa|{|6R>o07KcdgB~V?n_7zzk^SG*@4(DYLu!h)Sd^>aaULYWveL)PM z&8O|JlTli^WbkHqMG&cedrbAXiJ0GuuD;b` zNO0+XlvpWun#He;;d4F*sK6hmsP?)FC+F^##tq?;BycV7Q#>KO)V!FCgTQZ-kqkPM z`FaJ2Q!Jz_KKyI3;F%+e;Bfa?RQ z_Ze1yur+M-dhV5JZ>O@ZVU0g(GBf*I@F}&B&nqo#6hx|;)Lmy)mu_rFjoN(a zrZ02sd;k^(mPe5J2-s2^n^O1J+gYzZ!yghfAbKaCPN@Wj{a=7Q9-31IbY^~EmV50+ z7(pe9ctoOK1v{1PoeE=xl>N3}dfno0!6xF!9~Uc0H5r?~n;G#tN8TAwg*~KXlo1^y zpf>X9kLDR@!1X+>{1w%NXxwOY2{R=|UuA=Z-WNaSn`wnvkJ$z>awh z7~%Jg@Kd@nEbTqa8J{=d$=mVlW~#b%-?Q-lJB}qGH-* zRDJZxrB=|$DQQoP=`kEwGdem<*Qrm1T1>eaL3+aK>RgwyyAG&x%V)gga%oa2y@xQ4 ze`EbzHdogRPKmbK6H__lj^gS5ng^TiM(rq}$u_uZnV z7(2=xwivyL>&Io9Nc2`43%4FT5so zGZ?ixhL-F@QHdxisx@Oh8ebeLY73FIi`R_;4;>SGL!UnCa%}@7G=@+2t3-LX8lUo; zBB=!;)j88G=kT&Qwg5{OE4>mDyiOA&6vKfsF2`EeQ}Qfob{WQsq~v;Sr zS|zjEQC1z71_r59Y7df7-ocil){f;eiW(~MZ6$*bUU_Y!=242}a==z}CBwApNb?zo z*o9HcMr79h&whq(bIiVK)O77lJj{2LZ>UV(wm~Yb3m7AsG7X76*bSJf!P+*Rhh|Wj z{#PnU(&oHJ({>b>;8K&AtEOcxDqDX;IQmc#h{^xPK*9#E$bySt(KiLEi5c^Ia*MuE zFwbWGo3Ww5K_RnwN*6jXYN4YaH61_F!Ijto2U+KIRjSceqg49aI96ndI~N%nNUl9c z8E<$@R4}vda(D|#rd9xNikV&RXOVz2-h5`J5`0W~Btn1&(=u2EVK&gVP3<#&mvVKm zKyqeFvxM>R_{$1Fq(tg#Xr8zpyzgOl{Zkj}BP2WpZ%CCtBac$$)9=?;$$|3wf@>}* zvcT^&t3<(aTuDomZ&JeFNejHQME=&oYIt2v@F$_wL=<(RS2@nM$ccqzCnf!O+ef2D zO8G}`2B2bMB|}rzlykYFtxm`Zmk`SB!gGjgJefo__$`51ypr_Q2QF4p&?NQgz!w_; zsY%E$Rizx3r8)a)qA${M#{Y7Wx|Vgk0_x($|?97(!%AU94z;fU@j_1gj7IN9lGUO6VXTeDU?6 zfd!;aMoX2R3DnN!cDfe8^zDjF1@ zX6I-Ye0|L6*Md1o49Ofx2+&r?(U@Lx^RkBA9y+=2J`Qa?a(rhn>)AIszf^mqQmp}@ z@gmkKMoVm+w^hAF90zEM1k3!2G)#^TP;$OI36gp| zTF#v-z|Gf0QypEuFXNR8$awKy$?AQslf%cn(04@Ar$koquC7)?Wo3+zqnp{^R9kPk zmXkSOF>8(QN2sJ>(4Z%^ThiGu0CyU3;1v3252&wwU@nDB^lN?$ehuJ`6b(sQ=1g|F^51r}t zYACD4m;Ef;eAJiTGj?Y4@*p**4OU~;nh&*q@sOCOwxKI$1!d$b42wzXY-pfY2$p}L zyVIbug_e_UKP1(7a?=v3M6Vu>P?bQ*czw-~q<oz`L~9@8r2cZ^?Woj zJ)59R!q0RGRk5ecldAc3*YKEy-t28Sjgy6iS57CS7$+Izdzrj5_lYmljLLTXH*U4j z3LVGOuMTUgI-?9n{g*v=zxB#?^mCh>=u_tzjba_7oxW$7#Ho@nd{YYbWQxW4&2f30 zeDTA28L7H}ue4zzMNL%g9ll%G|C;NacfUuZymj{J#jw~*s5PHb^0=>*GiChP zk{xy_W7*#EMOC%z+Gq?#@4HqKlXLolAXmi`L@TbtIu$3a55WzCax|_fVkoeS(U8YC zS29O}D$l?4)1-JzH0xbMj4YKz`ZzO>e|(%79+bTP%eQ&KxeK z5D?#nvmGp}f-|V*Vq5K)DTxQZS$PCf7e(x?kqB3a*YkK9q`#KqM$|p$w74|^A-mzD zui=lo#<)SwVSOa~j)8+)qdf4vJQervJCBvkVh>@wM*_;ncb|Qp*E%%2d{z1MRimVi zQC4MNCTUnE)=L)Y%GsAcjr$+z)Oh_SOFzGFI8L1kh3U^Hl+Dn6ilg$9fnj%DVF0wC0p1DzC2Y>CM2{grlJSF#G zJZ@WE)gf>kN$PSh-lA9)Wf%<9GvDqWoq$@u~c_8EW!&dHrsXQo+&cbtndXxoA&_ zOumbVPpNNsuz)hOF7ee!S5MRHR^t%j-7h+GGD6oB8Uc?#0^zHLc3L4u) zqLJ03i__;2!TFxoY~u0_=74lU39rx{Mgmz!`KS*y?Tj$+I!d6%2u5@iR2rZ16Rv-b;TT)@%OgMIDDIU9b8T`a3Vpx9IvZe3xjn%?LEDv=lbe z*5Jze*h5pM_Tu{@lL7IoSE?yUb2CD{SBzaWNXwP2J%8|<9_cfO&+;dfzPGoWA3XgL{PdI7EMxutL9f?$Rr8}iTj!%L z6XB)XY^mp4+gg$X>%A{3MVZpx>yJh5!$p?JFOd&;XP0WRz6B>kpdDpR7&9%6!?Z{G z@(}%BcyPPAML8WRUdi`a`@L3WEV7ZEk0Z?bd*I{Hm@Q}RdtxYR^{$iO!>&b;piT&a zE1KyZQOZSEKsgyRLC2RXBV%$2ux1LzcI=hgTB@~CQG~1)6y9!ZdbM{`qZfCLsn}0U zy7i;sDHAaxEu8Diuhoy;+EOj)`=(<3W!j;r4hId4soN|Q$n>ex3rjQ)S%I%5E+Roq z7I#bgYu$JPm73jOo~joJM^v?v9rhUp$~(zsm{Ey-QT{oF(1~ayhyjeYVkc3vt+2pW zgms#BUUVr*9lh!`Q_L!Jww0TF>P|2^FA7KPH1rs}Dhcss7JOZ_D9az_u|-S?m9wF6 z4L(OxHvaWL7{vR|nb(9jN>?%X)7SUTUMGSLJ#wL`o00P~`wcOIyBEVvAD&N$j>G~% zsziXs^CE424xJs8<&A4Bbuyn=%AOT6)<{rVQ;BP*e&oo$-X|=kXco%)BP*l%Q~6&w zb`h+T#$PkDZFwX^>%w##CYd>0J*ALSqj7UZR7(Msde>4VLMX|M<552NFK-FrX$OBb z$d;Dq6~6qXr?-1VNF3=yai}CRVYSpiGaY!gbpD9ql6tZ3M@v`juymHNFp5I)r(3(h z*a4|_PRQX3^0eXi%j~1~;P32uX*O`3@w+zDM#pHOkiZ=Z7@Y;Itjpsb=mBR0W5XHE z*}SfxHG>-HpBP{4c(W4>iF$&BQI3A9+C>P(>ur?m^J-xkrOI=s8ySjOk=9JR!Fh1| z!5`#mlxWqyU@~c30jpj~l`Kb$e|tWtV$mpHhRg*61w?qwybH*JH_*z#==0iD>Pj_J zXJRl_h;k&Wut$F9Eze12W2pHS-H&sNt#U<;xgnNt3DgdqE%z5`zTH6!IK-sr%GeJ5 zw<@)B?*ooI3Hw}KbbFY*_Izq$weB^jYG7%ipFuM3gY!18Q^aL^ODdtMwbJZ^XW@-H zVEFFXsGs2{g;~H9%PCOcEja)>XW7N`ZN~j9_#$D3Dap$`sx(C$D)*a!6NLBT^b)N9 zl~_qTudoTj0Z&Q0pjo0hc4T2~!VtX>XdgbOQ&W|<`-Ur|DOD?ar>%(Qrq8gg>8D10 z2p_pRr>&5|PbDa&I;8#b0jk1@t9*Vef?mNb8 zYP}8H&gI|jF)NDzz*y$KEdc8>yejU)X5{O-pG*|r7MUI@Y_+^u=35p2luFF^D(PKn z_|6=@El&BMEnvvS!~=m270a}M@mTl+5-i9@HEff=7Ux9E{B~W=sLH4{TBtfM!x&Mj z{D9}Ar+HxfgW*B=s&|0A_a$0&yhja5D{t)^+N1EKI@S*Tt)}I(J1o)DIHU-5q>7gr z@wo`e9H_4mdD#`GZ|{LEleB~Z=ss6W_%xRpdx<$e6(MlI7z+kboJkDYktLCnK8$bg z3+ug=a>)zfehyL4NKCElXjT*zLB6crwyT)y?e4B8gt6!eev(>_Nv^9sj72Wq=ul5Hf)jQ5#mF+h= zD#kl@{V??Vkr|RC-~Go@@abq_WSVEb%+bcxzVGh4>tqpItZB^4M#xC4j~HEI@SmNk zhBR7K_wScYEWf^e43bi@d9~`|j`8~5sTxtVY1&JZy$^cWsBby$xK5zo_m(`TI8j=< zd0e^tuFh~hni3jp9S1T1vX~U(c;4>-ber|2@O2RMzbKi75jS?AOO+V@BuhF%Rd^$K zGkw!=`6aN=p`MRz+6yjQy}p{%ypn32G>Rb9`AD8J$h6?L?_}2BsMBR6vL)$r$#lRW zQOM^XKZVTe`mFM<*=JZF8aBU6fw5z8=r6mPc;;chON;~9I+{!6sWLIBDTNWmZVgs`WlTm7U|P#VT4$E+>*`mqFdp3D}W8>pV10` zukxE=#pnq6bdz_lDrFlve<|w}z1&SzUcc}M5T5sk#f~O}TuP&;@fODYcMrs6h%*#K z8Hn&^Rl9u^DDFRp6svw2IH?PJac^TE;LK0+i8A7fKEhgI?WIve4~PYiL(iiVU7bmD zd=Cp4YZ`XYcmTh#;_4iBCJz$ z&cxVWK%B?mro&ef6QOcsmIhPUByYO@Ktg`zjxtBFmwEHS3-?tP_syyAx~>KQ89Cf# zmOjPM0uaX{OVM&{UkAMPlZNT7+DUrH# z48Kx~pHw0f!O(#cIk691!#rx7%GMndKB$J)!P4s$T10pq=Hu}ge2+HwQ)t`0eIHFC z1VZ@gyEoB`dG&CMMK!`oI)w0!B-ycYxU&>1rV&XqohQGN4GFK^D#=)fyiO1~k4V$u zAGUK5AMN%1{%#brNa)0ge|UN9WB+lBY7J*)EYPExfleWL=G*SCVt1~PiV{F=sH#jb z`r>#WZ+z=8Lo{^?8j#h4PyFl~Vl6UYwvhAm_%p#dQOBGNc=jpQW&vZTWseuKiYWfx27eLko!SLdCX1||4V(f|Q zw{HZm7hml2BW%2Kbk1n!i`QIJf5Ars0Br5A#FJvW4}1h?%HAaGuE=em(s zbea7xu8S9yE)x!M(y4~`cG&D(*pljXfprjWo0D{pGMT;a+Y(WuyG0S%jTA$!r$5Hw zfe7tuy*k$U4|VC)Y+B;+jqEw&eND{1a$CV?FE zf*=OkE}l!g9rfyW1W83zQq7zAhWq`-Y1Ol9lCbmMHdm2uh~|WYKH`T`OgeFv{%gSg zZ2-RMp9>{)snZpx&asj4{NY>as@3nVaQ#GwIqDmaC%?B3a^9$sIcNPzoq}8Hl6>zJ z36L18Cvl`EibwHH(jt(e6$LGN-xl=BRwv*>ZnXbs%*mf}GYLcBbgWq0&?_kWEIdx) zqBqo}0`IQK8*>F~BxXn?^AEpu`>(-Fvo8Uu^#fwQ+M>w4%+lZXN7=>Bt)F5QI$}kJ z=r)Le?tIS}@U24|ERLmtqoK zx8XJeAZt6kLGI!CTrVayTxApsi5pY*mZu${i%_M{p}LFuCPhD}g&J zR0NSRGY*gPh1BpH$*|){&+?SLglyVVGLKP#jNMS1H#M5^+I&MfS+2PVs`!{oyUyWV z%q4P(IrVtXJmP=M6U) zJ0Y|SWW3k<3|m&o`E5#Vs1n8M$l>5 zgnsS2=(bUpLVQXX!tGd*>$Y{%9nLwvgxR*;DvKPiukKnOKmUUHHBd=U8)aT1sGIx^ z6hk_{p!$>EFH4;b+m}Gy(hV1et%84;_7-Q>5sP^?H9J)0xH{!0f6D8)f`q;IS@$a- z7W@TSzJW0}zN`g&Akv}>s`>#}bT??6K;wQi;#($kP_Pt*stKdwP*+x>KI}FkE-d}g zUGAkA;Me&^1TtsPw?nl`Sd&hux^27ARNrS9Gf(BdCv!GWBdgt8vGcK`))Fu{|- zeQ)9)*YVw00V37v*t$+zj?_=IJb$x5yo5FdGSsb_TdSBy4OF)NJ7Gj@Biwqx?`8YaOnYOBI-=5`^8_)PbbmD;FTOsIB9(}MGcWiH9s)wK! zY{geq5RD+WtuZqD?ViO3N6pGQ`dM7)yfLCyAUp)NF>ZACNlfY73STzT4^0wdOA>^f~44aZVTfiF#q#-s5!GM4&YR9=PS zdRr((mIV}q+q4!Za%wUFpEJLWthD6|k-o3dT`XSWWBq}@P4g)yxc1HsdcK-Hq*h07*na zR0#7moU#tH?SqIKjU@!Z)J!kPWyO)fD=+~lP=etxr6GdS@UTO;3VrX8L%IeW(hYzL zLt-L0xo{tW1WQ(_XIPhB91~!mYf9G!&RU~*Y&@yj#~E^BX`UJ8%b&@;Qn!KhtAB(x+MM*uo~cKQJ$z- zG*FRXXf7t@fdW|pt2}&1LA{dh>cd#dU;{f4ya3IjirKF|f|Ld*zKwPzg46MKr?i3T zk0PE<7+i?V-477M3g+V=K~N&I*9N;;CTRdp-(S~CeaJC#0JiW-Hq)CY-#cBf-q04f|) zrltChAXN!~B17VHxdNtze);6`vCoAxAffzJSw7d|kQZ8A(g=xbg4#-m!mHrx^`Lme*;X5y-qEVSm#a|@+FcED>XQ5}vL^McIa_H=V^wj2*_CAp|I7(20 zss@8*qJ3$ybfPX4$shJTI)gCz(8~jPzP??lf%ack?NV-u)sA*_D zSv-GfxAG!Z;5w)S?~gU&*u>u~wA#%MSfSyYG&KdND-Sv1kg5jNXwMqcPJORAJ2wjG zkuK*Y(TlDJQp8b!)bl5-IZoOJmCnHDmUD63Mf!j&D$&*jUkvGmx8n%_*S>a*OSzpM=-al9`-6X6o7I3j zyW!P$43gBIlTJj8usq?IBN?++h0XsB#r*_Zou&<4@=4F5-H^8WU9xGZC27$aLpx#= zI%pASI5=c9_Bu#chZ7a(eG7|dkur4_cLTBE*JTB)f@sQcqTAu~dM| zF%j6JOsP|4{@<^pDG-y+6Lg$71gYThLr%q1=?&+>rlp7Z=U$pmXnF-wp!L)-r|fPM zId2U>tI=}Ah+`1Ieq6|)MR}1caIbcD%fWcXg1ewm1LminWCg53@S|-4Ltz*iEi&il z^X&B2BxETkyphIWvl>CiH!wcm+96O8q^O=c(gKy=5vOzyLz2Qds{1G@XvRC=yv_Zm zXjhd|4?eWdDF<2|vUw#OF`D4FiY6NaXcanOvH$ti3I|FBH1Dcsv=WnYPl1Ki)->hR zbx504H8u=<6-%Dx_L4eH+C<<5*aY`kAH0%`y=v~&Pmp3_#EC9=_&xTVOYV8mDFcYl zM`?5q^Ldwa!xpHg1L!(g)QfWbFjemCcV%2V4h~ME zIObH6xuo>Fs5w$8{p4ISG50Kr0$Gu)e5lK%*!H6#lpDB8wQEd3g%dy>z$(?$7gcek z@-UeqgSnC~96jNbd%-xJ_)(bxDI-Z`uO$$)==M(U!51K59qDq(g&Pyn1n)sgSq$2% z`F8e7k=dN7(eOaWQ4R+u4jw}b=!|8ls0VD53wS)UA5=oQWmK}#bPG@o2o7@$wV9O{ z=ob~po+G*_n)OnsAIcF>(~_##)J>X^0w57b@?l6&k38#=dm&AQAYD-ya0!l3NVhbo$&&kzLv-UIbwGJm^u&^ z>9IXfLBTeqz1FmvYWf_^l}uAWNm0m@A*uGXBLe-HOSWyW;tsOD3Z;*;-%+}%fJbgM z!0Kc&0LhU7lxF8p3aoISfA(85O@5OfD3Ec%hmDr0J|!w=`~_C_1^1V9p;6bg^O>uT zj86xg-3KqfJDzl@wWA1B9OdiwHFGZJYBnf9+AApw4jA2t2y*-5F6HRr!UB?P*M%Zw zv9ts&4`ijewkLicqph#SK~T+^zz>shN`Z_@RxCLypB1~@ZNU_eOFdZ~A_@KYn4(Ro z(_^~tC$pp4Zv<+;T?bsUr_G^kQEjOB`GQLr z)t(-kg&<)OXk5w)#TGVuBBSg_?_CC<3S^CI&R&jEz9Zw3l^$)>A?)*N46)FDm9BAa zO``a!g7l#BuL_R*fpM@ZRccIKzgnUi7!tYXDYp!v(JT#uZkCZ?>W9o(TTp<2RRiaX zeOO0pXw)Saof(%3tXgdUP)+buhQrfb5_>B!SXyzg9FmncB|2(eHh9vE~jeR=6pWAc|1s2FvEb9nv)pc=;HJO*1eKsA|C z_WjMe;+(8b*s<<@(k(q$CyeS`drMe!vIwiyy62+p&^&A<_!^H1U?%030$Bqq&PCV= zyVZWwYR1L4+|p=KZ{z5TI{pyAilJWS(lHkG!UB4j%#3QIUc~l$`RjeMB%%woZb`R-6HDEQOJtKmh0jpF6Yl&(Aqw(Lt<8J^O89jfh*xEp;d=CMO zQt=4Xxc6Z$mYo5MgGLvCif-$iYc*3MlT;dM+_s#nTAGv>3S=Ct_!R^zre>}~P~wTS zpdpu|xVewtikoO!qUwP}_3-mn5Lz!rG-P%r6*dJlV1)r#4Kl)y0x1mqQDC4 z%7vhsr<#7Id5?rA+TI&stEx>6z-Pg?!mdP{mOiU!(nSAOR50f!lJ(6sD6q0= zE68E7$h0_3g-dxnj))W|;r47)e>tpx9K1>cRzD_kKZ3ysJv`|gY&XEN29a@z%`!-$ zW~dMMoCni}@^Z%SQsrkkXg<@vA}e5}E3zIU^R@7v86bn2wg|<2f&IXXu^Z!&S@x<~ zdPWhGLrGL<2oa249()>6DKP4$Vctko#aHR9(;({h1Z-DNz#hrp9HiI{c&_Q`Rxhm0D7Cz=UA!fmEuETuD{v_4~2x z{38gXREN1awg)N}T5aIEddXbwc}}(PQftlg|6bNH0qY^u`LR|odpCfAy$PU=82GL+Kz=A@o^b!|%bC6?eqLdH` z_}R2ogkfyoxyPwUpwxjlZ_mVn!JgsuOJX=IW8Ezh?dp)kC{}k)z{wfl<*i>P{>GKK z$1BMwM&<^v=2Vr88~7wvAHJqCE?x#EwgJ|=9Vo;Mi43wEP2x4Mzb{F70+wJmo`FKS zyOcGs(w|U=3ceD6^|{fQT#V64DT0-1+$G$?8I^nA5tkf=tNId*u%l#Bx`Ehg53b~} zG~>zGQ9Io(tr0aWfcBFnSDrrZl;uzb*4O4=9T-VM)jLJmd5#;wii~VaN(D&JO<~re$0zE&Q@xz; zOiPnNaW)W4d$gzyvj$eA5{XR0Zl>E8_N6FQfdl|)%fZ5G{zcV(X8ah|Jyk!?VG;=M#P6yQ|Mv5w>Z+GZXu}!e3SlV;He%?mMvlEj zPJZX>GP3t6>GU>Ou_$F=|UQcm+3 zAV9L1ThJ7r&5lF$U*CX3o&{L>ah&SUQ`~K%Qz*F8?Ga~aiNwQg;tT{O5gC&NT-TIG zV5+Ezkzoloua~MD-zg_=|F#St+$+I#n+Q7(t_*(zm*Mkl3uPS0v^izXii8AV$ISMkjRYso@ncHhN76}R zCS(YPiU10<3I(W(B%}fWQ~^*5VIPFFhutRo?4FQ;- zlB5L1XVjB?)(RCDqIwQUl%k%xs_2@!Pw%EWr?1UtuSkJ+lO`5jAOWTd6^#PK#;|%r zzsSjs5phFJSO(;;EcZ$!CbR{iuGHYkck($akNK3qp7&-QtTdt4Lwvmggzds8CWK0$ zhgI<>Ms`LN>Lg$ahFo_E%(Y_YaO0yHaeFsL&Q&m|7XnbJ9!;ZFgU>xF!~0*9cz2iL7@@Oa1H0rZaRCt3 zJT2Whbp9Y7UE`^O9DM#M8G7LaXR$CC!EEfrF^LTg zsCPObnYltC@l@9;Ahyf=Ft&+usD|lmv8*mB4NDg9Y(an`I3;lqSCSLjVjPl4#3O!e z!4a?c#-K`zxWyf_94mqe#4cpNMAjz*N4dQ>!6Z=zo;=qaObtYb;3bL;+ zBE7w1N_7~5q*YzzlPYW>0GDoJyjLGvPZ!?b?1NQ`CUpR+R|Bvpmv!nGrc&CJ)eL@=HkU@tHCGBVq!goUw<2&1b z`d#V%<#*L*>e0OGQ49cn@*lGL^WRk}#|iH8d=@}8cJc%qK5mmCcq#HZZUj@lJ8NJY6Xb&u5-b(7P16v{s9U`0aEG|f(Pv09z2v{Zroey*3pjZL1j zOW3rEF|o7IRyBm>=?`F;bFNRt5oy6BG9`S3Z|QPXhKXXzM;D&GX&CEf*+| z9ed$X8GQ7Q(vPj-4L3`}jc>Ioit5Zfwni;o9cXvJc7>qEW%Qm?MMk-%PU+GUgd)&6 zDIS1RFai}AB&QI-$pYZ2xPkq0s>JqO0rekw7i;R&b5q@& zzjG|7PV^)s81YKxOSJD^n<@G`Xt(<_BuH?-=x4abk-xFx1L7YtM zs+!hCwPzcm+d7T7!^wb@IkZ|St#uXE7h4_RWEWi3!3QBot4>i+p9s5-yQ&5&$d=3K z;RBMq;4&0nx8R9ik_d+tn{(AI8CL;z-T4dEuF2q*f9Nw(b>*OCd&C*Wg8oi^t*Nm>fJZARV2I8{oyV3lu~g(`m*tSuajbU%*jY3_U0Yn^m!YF$4{@cG2DwPVwP5 zjZfs1<_mrqI3=AdXj3Zxv)9SsAMR9YIVCP?4v@;6Bl>7xkNQS!4;nk!BK;}#UGIHE0>T7rT1p-ufmvpVBcsO;GdJL@gOJ3BK^C?l7A_rRVgM+r`DlNt83@VBwrXiUwnQ3RtEINP4C+2Ol;E*A zgXBWJrvMAT)APX335orZ6ZQ#MrxFQ@=de-gqpd0c`sP>uR7G0QAKhe~;JY3G7eVX5 zp?-OBPmc_6gpf-5KK?&*DG0Is1$g z;6|`iiC_h<#rWWWdUyGbi{W}-ra(^V)Bt1zLfYb7l~s!IS`X?TY5mIAUzP&qY#9RY z92|nm4Rqo;N>BtSX9DxF27=PmUM0;fbqb&=@tfX!ijTNqc}k(uOyD%t_r*cYiBw=E zS&8Q>Q5e#D;ytMw3TcNAu=*#RGqAoOY+3}Y!$sH7`o zOCX8%b;F}AR$tjLllcxY`;om{~u|m8_grFqFC>xv~TFC8q(r2lkXf6t{*gjg@hULY*J<>lg zrT~jK@VZ@7PQi+efMWHl;ZoTJqCmzfy2db7MN{|P56Ozr5f|RIrpnnK=h(mrR^Z3d zgXrpN2{Riv`kAQ)2Bqv)wNk$MJn6aphZ2VyxRP*GPsWe~lFiVI&&k+etN5!|inn1I zRCQ5_4fLy#c%ttFfGV!ki3Fv@=&*GD_wCa6n;)t+Uj_#YE5ICy1j1n#PuaiNKIk$} zuR^-Q6Ub=GYQN=^s;^XC^GXYt09K0VVD1E?K=V5FS@EGkFdPfyc!FTb@(w=#q>LVV zUcEoiyk4E3iczafa`13nzvZXa}F*$|96@q2o*s@n_t52Y#0oRL0ARuODL2WGOG0p>oAd!W4Ju7{%gE7yttrtqwg;xO0s6U}1QOPkf?q<9fdR{^Yel>_Y4@0$FS9e>YWYJw7D5^{g!*dT0% zF0zD@=pMfGB1qMD17P3v#ES!%C&m6F_e-b!T1n6rh6yX+fI%-xFa+b=Dj{fN3u zdXi!Kmv4}!_kRR)yUHg;u^{o|3NXu7Y?7hJ9+dd?H;EhfF-7}&rTxFYB~gTr^4F}A znwPxl)bKe++=$;m*2#HGy#Qf1-n$7AJ~n%x>T~-sfP*~^=VFksMseK_prSZ6JG(`O z3rti2T57O>zT8^5CRft*s<`k@0 z7Se`|@R|@TRuDlsM%1G%92Q;T_Y$o75yfHz?&~#zI4MPx*q~`355o((73)<6z?pRi zDA>3t7Yli-wo;PG7e=GxL8~Occ&iE(J=TknV`&JJ2f|P}V?-s&`>N6n-brSV*gAs?j9jP?4s^oTfme5-GE= zgg2;&G6bviTzE{oE4`L!q9AWdzrL`y6~oPji+7S|TXv$4woznPz8t?0$0;!0Ej4h}k`4}evPF$mp$ z3px=1Ag#D7t0i>ll@k5MtupXCsOAzeiFTY6Pg#RRhT5g!`VWbl?&Y*GLEWZoLHO|G zP*@sn{s(bFmFU5oEk?xf)YPdc7gWDF(!ho)JOzwPmWmJSW<^>LNc9bGQ&JtJH9qTG z)`({y38cLGCUB4xlXqZoKzb!oitG5&$E5>%&1?x6t3LZ339a93&8=dEWihb~Yyx4F zjs1EPCM~ab`(zcQs46EUP<%@q*EB$}e<;3yKQzTfA%gkPj#8h7tCN$ud-BP7{pKY7 z$za7Y)4ogT4{*xnv8^!gfz$c+UvzWtcom zrtev`SqAR9U7|;hzz)?e6&E3N(YfbaQWcm#Gte4=%?J_pdrX-j;M5(Ae|5EWT zX+Wa~dt^j+A44+vlv4z`F+BwpFap%7uDj8SO^5udF1%c7u6l(8Rk~+Gy*ZQjv2k$8`>wH=+y^^AJt#t*_Ee2G zlkd$}UbIZGGp}-WDF%|Sa%AJZ9H_T;I^+Z#EMSUM9X%WG$$6e=lbj$tc;5pMi=*Oh zY?fH-e$_@-eDrfr%hiFy!5T$L31CD?%8O7{ zWB|6Q;r+XXlZ6>6o^J32!Qq1kBzEvcNrrpGv*9dR^~tXYmouSMX3eMC!UoZ|~gJphaDM8qEsQhEPueVkIHkJB+{yAgRUB+ zNZN1xB7)tXBP$U@kZ$!T#H>A*I4wsw>Thtn;`s!eh@_@Nf)wd9) zPrg`%IDr{1vif|=Oq8ky(kAKiTU1@(#p^gp&?H3x)&nOEuU=Iyjmr>t84Zu;@|7ma z!Z|F3{^%WuwepT3m)wbUT2|xasURRZEKCkw#h3$q@CIy;SHnfO7jI-&FKn*P#!H4m zFYJEIuVOzh>@S+4A%GEAU3{gQ;Om9MMcEl=BVHdS;A+?N^heNssnVnB&-ay|NZ-SM zlA(w1R{>}fDT$1IoyL3LSr<#_+#OPV*)=KvF6~x~r0!e3LRNnCUu58khh^a5du0U9 z5uLTXrtp*>KNd6U)zX z5Tv?5n+B*NU*!qOT2BCu30RT>4S+7^>Gr6Z-^ms;kpr?_(YutsC_K(|ih_!#zZIGu zNE7@vN`Vtgl<5)_Mt-C{S1dl>5>^X6R$$F6p37i!ninDRCRwqR9CfjA+zaXIDo9tC zL%JHo?_z8y^avb9Ct2wjl=Hp&)!eHr^g~cDcZm&`423HK-}R(5;hA+;=-)Mwg7_NR zhPb7`i1w)&dn^b;iHxc_u4@$w52^?jN>vC)3`$UR%%BG#(}X!s001uBNEgdE!q><8 zdmuGI4G480Cm8ctdj6>Zx2g53tV}GaE(TZiL+^dND&Lx~en$b;lxt_Hhvxo=^-7@R zV3O*-r2xil0I2m)`BlIcMUcXpwkpHL_FuNE@(^rrmT6AqnW;Z4;Xon)u(jqovaAFr z#sORpXM)9+JyYehOFjf9+O2|^8@pw5jRLQHl`U_Q6+c)vj0>PUdKZF&{RYxiGpJLH z(iJ_AV1bZh$&hSLltC5Pht6I2&K7LUBk-mkMUTfy1jVBQ9Pg2;wBT2X-^=t&PlUufm2j$+-dK#&rgCEcZ- zt(C|Xa3Wn~lQ z`QhAbU0`A^nfb+bn771c%QneLc<+FS*$yI5MN+JZoh1gNdq<)YN5zw(QP?N4z+7xY zA@)pe)OpUem!_VfRFmY=^pK3WBnS7LD@9;Yc}RAP7yf^i{dX~>4Pn9M=OQ{T! zR|xB8aTMDA`H#wITZ<*-r2u8FG-cmp9<0`GJWHT`l{X|e1#4tGJp9TWSVIfvHFM|v zv!}V;rSi#A2SIESeAB6Nu>6#&=mp4cJ#}UFuPK!~*mPU~^ONnv@Dk+Ii|nXpXQO9c z!HR{`jbjI-s}K6!@*u49#TuWS{m%n^No%6KD5NUgxJDsa@!ND=>hoq#qJSz2 z+Z?E+Knu^D_WX3uVmJhejT4mvh-N`Z>;C<(P66W_Dug*x!eD)N0Q~4EaV?~%HwUZb zJbyVK_e5(&!c)UVghqE-id6BBC!cZF*oHztK|mArkU%6 zBIGymn>nJP*V9{Rz{3gMi1J&q%)gzTg)>e1l0MAn$DA2_c`oqz;f|rT8WBJ->B5atRT&P9RHdmt{hsyVhB~mPcXWI$tW4L@^l!JWs>tr=_-Wt8E&uE7w7pe9W911y6aB_6tEHZ$dEP?H4MB z%)Zt0`#U5)hFB7C5}7LT9$sN!X$Z-3KAa}r5Ui3*0u`{WS|CN|3C-lE$|IBK_?~R^ z{RreIoiF#?8dA0@eLj=lQEYipnxeX&>OiJkOII2v@=jMuS4?lL@X9Lak1>T&bs%-J zi2krsFq*}ocpYK|`ci;Z*dLhd5msSSE6b&)bOjf9PBgB?YQ=O`8ouv067B4Omm!p0 zX_ZA#AV`hk9PLtDFoE}tfhxJuU#TLXa}bhCkeWalc9aYv2%RcMQ&B7U-IS^*QE`Dq zW2>4#`DuUrfA+2eu&N?!pR|`=Ng*U5A@nL(KYnH!m+C2?@e2HEM9dh+fU9)%UvUf!}}?Gs3o+4fD}^;fCpZw{&Gdtg8y#)9G-08jg~#fkj{$ zECXSnj4O5@JJ=NHLJEM3^gDX0BprE*&RtpN@$6$(Ya>We3B3dHG67;|K-G5yfGPt% z{202*3|$lJ;XhOXav?AcE7;S>XUK2T-0P~k5EFa2q}~5{N>g=p^|Ea{y*~O@y7C*a z@+R}Kn7cX+D@@m;74=SIiw50^biFGDP}Ml$>rI0|I`4XQfXzO-r2b(b5#B#Ta_@Uq z!r_*w0UJbyxIJ!gUdn;<tOaE&#iKD7Vrk^+>ZXv)5$s77kB)X0~?ufU28iK++g(s@mMxU9n%s5hFjbA*Te zV`Y^qPRiiiHqZ?%Xep$_)h=qeT!3qbPEr6swEf-JW%EBClM46{48iK{5O5lLLZ>~~ z!2lbXv+deeGYd*c;SwQe=m2w}Ja3Be8>pa43Nx_#PziVVE;CW#v~RBB<; zUVt_?0`pSSBjaUXctKBqKR+s={cX-hL9+{hiWFm7)o^pBY2uuN;r=kHu$|SCI-Lu} z*!zmX@a)c|w(GaaX8bCc(GOAd?6`1w6Rr_T(840{z`YdGukyRQnW1zQ2}6sOPz9#K zS71aCTGjmLpwm(AO0m0*UAl z5E*4=c^Eed9qDbF1~Y^fYSQ36lV)cK(53G-NX(!i5{;t}<;(;TkHoyx5V%O9`j3*7 zzk!r@@jFr_-tT#zQ+BGsTotVV$xp~XsI}<_smhM91RX6<9*A2`q~PrQ8HRWd(3c#xp~K1y=!HfcgSjL&p5Fb(ceg?*-U z;cnLuz)HIle@QNmLQCQKq6!`_4u!XgX_%K91rHSTi03!+Qk^JcdtW73v9IwUUrEPa zO;j@XCSHQ#_%yJYQx#YZRUq9kG5DhC`<-NEsBZYvq0tVx3)0oCaPz@!M9LAtJTq|v z^_P-VY>-P9yV9^RT`7F6RYEOeIoJ%8GO>|Hz<>-(?u{g>=&v0 zb%oT!(QJ5Dwr&>^J#3V2tk}dUZF?$F%W9QZTXrXrnH`$S8Cv;`g=ODF>?xIwEjySt zL5h^@1U2z?D?!VFXUxATgSpqrE3MtuJa#&haBT)yGE$WjBpv{uvW+TQqk6~gvx;dI zScy#<YyL3t&-G6TRD(9u5o(&n)2D6!;Tt6VcBGLEHQ&hq zDLR`Chgy}6W*Z<)cFzWS?sC!nJ_7~?>WG)pE2L00mcjQs8aU|bOyKKPn5!Z!IWHFj z;&1^l>#dVXcP3lC>s=7iNLFTi(wznLTd-oG+4FL0uDwkqEIOgp^gtvsN0n*RP{hqD z-RZi@N~C(rX7!*Rme>!b<^W9+l}BqqpYges7GVOugrx|#CSwD{39#^^=Lr|?IiX5* zZrvhbsc8~~Ek@c**L=_W+xiLXJLa+Hm+^FTf__?AA+EAgaaB}+=1^&~ta#=~&#<;a zG4$}f-1-UIl}>}SL4wxbCRMDtDyjh8l-C`Ad1afbw?(;*J>%eqR)O3?gy!ts9aw*U z9SB$q2f{M&F%afTY$d{duS#*xJktmP`iH8(NI>#3IJZr74UkcxYoWft{y?2(`Mx0t z`4y*qlH#Ct+0sb^5wv&Ze-3Vi4mCq*v8wBob14JuK!-s5U|13s|M;yG0;n8ViXC_2 zX_EGvb0h@YXfUO)q!h`2^fsyd;6rgvoi2`)H0yihmJFOP!~9g)FAHQNfQp?Za<4-q zZN>$N)wh`bR1 ztJbWLyvOd7>UE2)&*5+SWw%Jup~vX8>Q$K+lONvYyG#et7&fB=Y0ww2zg7~o4ky6T z>_F_)Yt2hhl7b<{_5!IZ!WQ@alC5+o@i0BSt3MyAOiavQ?LB@e?GYnAGS5-=IWTZq78>vaw)fcG* zW_`fj+K>i~C!B@bB>(=qG+y*U$H{;Tu9h(Fzzf5L`s!-k5HV)tIGBY~r3<5mALdEi zgh>Dta9aRK)tXhZ<@IND2Vc&Qaf^9wqNwKi*Ph*^Na2G<%jU;tNqFW!i5Wd!X~7b6 zF|Pp_f}9IdOzo-vJxGu!2&4*M7LyjN}FhY}(Bp4e?@;yp( zg0%Roo@GWIQbpJvZWbC!t58qzJzs6l3HuDyf1LZJy(2-2rs~Xndz(*tzGKHRO%mJ0 zK(g{hXwrUt@xk5s1{)b08yntji4l^Ij!4@%sy}*5U?=0k#YJ^lP;&;3WJUTD8zLhb zFh8O!uX9y`#0XBTTj%Q4QqK(vJ3L&Y^Z}D)jOwet`$XGL!r{m1#)MQYl6h;U-Rw)5 z7){Yz2t{)~mYVW>aSRwH>E~P`q45a-9JeD+O$VmCaq;kUo`KW=Li4}F&cM)za$z&_ z9o8>J%d-8o7sXjz2unijEC43H{q2`^C=v{bFZH4uB=y4U#kqEg0)Ex%l}HPtRxIYN z-nh=1)8d>HB)QuBO|tzj_sH_wFP9BZ-6t3BsE~t-$|b6xpk*sRRu0)VfK`JQ77PZG zklrJ08HaOQ<)u=+ZjID#$&}o&&UV1Q zNT8~$A~){-5D=0T$v{?uL81)Dq{#b_u#SbRda47qb(?wJ5Rt73@e~-g=F5)f9@gfP`rC`ek(#cO zQF9L)1jZmpYC)Az0HjTrw<%NUi|K^pyJVe}zy1srA$IOP^jIlfHeX6U`j58T^jq$e z_(=zpa);wA9f4`Ofr=BnA(=SlTr7mxiti0?LH848z=(FUYqozq> zT%tGuj@1QQ#R2dPgDSI8f+35>0U)lgf*yjBjawqBhQ#L>kO`?N-E*&I<9Y>hmM1t4 z(yVU!SgWALItfikQrd8S%87Y4CLNZNYJF4bhK8U3g->#c07%Jsje6v-A<3V zhxm_Ak#rpn9Pm{_{YSu}-@`04#AtkD}^ zS7uo<_Y0keD_!ul#*3PGn8b|TQ&oyo^>MB%{j3Wych+A`?YWw(YUMJW7YYJ!EPCo8 zWxSVNK1)uVahdcVyO#v_PnRIfjRgaQXjacTq@8qVq$|~ykmy)02C5UOo+B#o8Ko(e zIPokjEnOl84iX2<@`G5hC`%pYrHom=li#VsJOw051xJ>hYFZQcJ$#FGK#7e&NwujZ zo$+SNfyuD#sBuNhQChuAprS-YqY7F-CLVrIjjpdI@8oI zNQjUNAZ5)%!|W}4U$JalD7*oCU9GAcfKC<|U5)uBlS%JKCf`^`R;_jLFOZLh5z4Uf zYW2po3SbtaYb!0}g51u15cFJB$03ADloWpOrj&d?SIQRtDETivBO7nM6ku2*aeE#l zaeMCvUQ1Pmo2`MN6KJX+d9iFU*c&P|HckO#=X{`+2n6|C6*L&1*90|6YQa%a60^^K znoiB~B~twLr&6_I31W#vVID0QpkWLts8ZuQR8>;SiahWbsarT#syA!TSza5Gpw0eODQD4XgzHB7O-@?z||F;XVyNILG9|Ix&8$uC!FG z5lOocl#{AFRqtWY5SH3ssZ!6mr%I?lDLGXE5RoQTqWUe;oZsXxK(iidQUa$Mf-onb zWj(9}73fKzHUeHZwa`YaogZbb>021(HlOqr@G^(9D-hxS3^ z0Je#cj0~xo^FLV+fLedsrBd|j)7pl@(}&7{OMb6&WFFO-oAK7yLj47X%iu`QbW0sg zTNs%IrIETFRN2Bc6o83=(x5UD765T#M^8}W3qC9O>zz^$)g}Wi^nz&m#&0r^Xl|w- z0zgTEG)Jei%U}D4oKRXJQ-6P}s{IUL{6IR=fwMq4Q~)9-opd)-S)|f`p@a zHeNSy&k2c2rB~2QX$9jE69}A~KcwWw3RX$XJY-f;PoBVUvX1Z$jV^STKGO(LH&l&U zG;!z0^J3EzM;Ov+J4(|R+*g$VDwu?*%5@-simKN}*52YP0dtce6^gbUfx$yCXv`xb zJW}s*POv8ftE%3B)o#>VNAm8~+B))D&&DO&B7L&3jmQ&85waKdR+)@@oJj52s_PoM zKrw@E7bL&Hlsze=1kx7PGU%PINxwT%VV`k;LvX!SKGqe4lLznChpe}_xRwm3a_)yx zdlpQ^GqNq!lAD=Zse?*jylU;H5&}R1qj*XUG)oVGWts_W7v6m-aoX=B<@nzqUUFl) zCNjRkbQ;`72_XO>%@zNcL{MOUmsW};^XKSoKH@=BZZdbO+^F8;Tvy`h7s>Wlo&xX< zHeW$x#BPB|RB4eRZOYD-aZx#P%%RgHGku`cBRW-F4yr5(4sKcykO(R+fM5eamLQ`l zJv`UBBgECbQtgNQa?Xv`d6dG`$bxAhP)!o7(dTp$g7n;K+< z#9%;xIBc&}GtvNrYCn~tKm!Jz^B(iYy3xUfQc54|A7+7L3`Ym&t4J#X6%P`#x6M-# z;KQN1F2=gRA7N>_0v=e%ABGPfEqm>~zr@3srPc|T*p#4qBA{-K8jx+jJ3~N%6$!!t zBfHq1@bmLN77QDM3rq3Soh0l2v!SJzRINk_uX62Xs5yS^6Ifrc?F14T)|WT~DDjl( ztcb#M2EODD9B_f9O9epg%%Njt^j!~1WMs5B0iJx9(xL$%^N!M<3w|jb;XPs85Nn;0 zS}QOj@W9Tgg6_lt5RBNCDzU}6jvTXVHO^7ji^d$MyeNRYcW)2qTVi|?0&nB_D z`2uuTiF2u(!2ke207*naRJof!!{h@Nth{MMgNtY3r|P`2S~&!b0DVA$zm-ZBcQ^Eo z7EtG8%NWnfkMPRwfLDuQp&QW&q2^L9=zm2qd^tU}UNSZYNntqv7p==^2-=-vBt*#| z3?s(q$p=c!e^yE1laFET(nJLxojbuuvU>e$tZDjIivRV3s^G}v$pgu#5)YrMU=xbP zzwsx`kZte0s1j7lai{9YUQMdOcsW81hQB{!bU=nmg8-GzIBy1ti0!EK+pi>Uk9{mD zjGPo8$+bu&xL@$Z-G~7OyLGFC49bPSBVbAqm^P#gswK&|X>1TVFjtbX&HT21KMpf* z_z}cAIv=N|`BrOXXDmVb$zVDsI$~uIEb^*h7F-Ju;5?p!5*;y;DS<{z8x2CaJD=+u z5$>yGh1=x%ET6_QxxJPH%#*ZHX%wJD^LJfO zW#vrE%fUb>%+_hX&iv3=hEkQ9%5x48_w^%<$$QQynRg^`P*+hOj$MF7Qs6yiwsOeY2^I`C>Cc6Z(ib4#w(ku43;k^^;ML=YyqAuDua zd+1c&*;vVN!jQq$+u&*kJdg~w!-0|w7$oPPc6|B{M*MLw&CUZxIMtg3jU@`6zF#UA zEkM$+I&+7(BTkX@-(D%nN1ddqD}oTM@k-x+Rmy*wCv`RODnSN~PECeRNr7H6ssMcD zOMg=EqZ%t5yX%HvVX}Vzqc^2tGo-h*Fms2mLK-`id^bl5{ykfYKYm>SG7gJ{2i|(0 zN@67ozLFrU>Q6ZMh(?tf^2fO;8jTdc^D+PvX;)N3=xUHSNL*n6k06w}5b;KYgvrs^ zUVZ?a(9&HJ0ks~$s6d1`G=9)Ux*pDeNNx^6cSkBA!Ag%-j|3~K2wf#5dQOnG7WqPg zBH-bg5|V-{;)Ze#jo@a|)`kYKMLs#N#rx|019cMu>59e}dc<>`SwDTB^KikCv@{@n zeusJI0%DF_XhEoDv0~_7evY^viBGpTaL`6G^$J-3#iJ-WSe!-3XM59Zw>Kd_j7Lc& z1O%zV5`p{fxlz`yU#YqL?eDIY+yD3%IeF>~j5}PieEH8(Qo2oIV-sckgh|>~d|7rj zJGS8jEG06|Y`YVzIG79Qz{=r>nq$MOFn?75;yqGb*DG8yvExDc0-LoES&GHN<3m>i z2snQQ27qV&7|3|~!O%jzu(JTW6To080YnWMDTzm)E=3=_0xN?hQi}O26IgySrwCZ{ z#qN8c(2E3tiA-7-5)+I-scF8djh8Ndr^ zu^sUkJ?2fQ$zmjO&``;|`cB#W$Zg{MOS$Bfw%^NWDb z{8dt&w^3pMEXqE0?y6SvH{iwxWgT2MMGYP!wJUxU0SGvuZmfbD?kG5)9fxh^VPFrl z8pzg|qza9R6BnKnR0v9{0%Rt<1Bk9`zW@xX1|&!@@IKX<0BBnNan!Bm=X^%sg5((t zRUYS}sOB`F)!4YKR;ETEOpYo>PEL|oNSmHUg6Q2sU!XdO_wgLGR{*uRnSwU+U`*IU zFbLuI4anjxiNrQD#7|9wH`)jUDw?U67P>lw*&dDi>UItwcd`Wj$Vf?d?(lqm_@bUhfHbJvr`#)1*HJSx!pyq`NP}?)LKz z<}u*GisdEKV)u+fCEW@O!NTp8@^463`+@0k(J-F}YOvISE?MeIfGeFLvR4=Z3zk_Z zkdK?>E=j1o34AIF1NRZ>_457tU`fPix#QkgP%?B_GS4*l)N_>-siRmjq_DJHiQHpP zi5f9R9m$4MRfi5tVo<8!+=+5McZa14E*fzMKl!E<&G|%1Kl?~(@>WXKqRmP_(ooY) z8YmHajF%)xWno^P(~}N5QdM{rE3mZ)Cb?-OKwAw}W7s zo&gI>fG@y8RdVF$eC!>NKBXNE6jMyV{ZM$(py@bip@tWHM*3oYiVK5lH*HW=pn{dY z=ao3)K{ZN=Naw7GD9o2;4{AXS4n09%!yh;8^*-~8?^s3}H0Y>O`k8n5*a7!S;Xrww z?<$a@-1?&oQWdp^g8)?F_#V~2Ub)^?GJgG>VB^9%wk3-(m_dL8DM(XiNi^DBC3hk~ zh{po$Gp1jtV8!cASmH4tBMWte`tqhfA9iyHgQ1%*f>*2d#pk;38zdL|O%in1_w|dE zA^_I6PzzEe$m;bCIlHHPv;<7Rywy^8W>3Z5S3^TL0IWEMZ%6O$o-DPRDsu$>D}x7O zz1Vh4enU&g>kj7>^+xdFdZWyle~_9}&Qc~AhE5t9OEp)xrpY+E09CK>A!Vjs*<(6T z_c^f8HEr4qNjmIUH4JcZ&I$+TRD!|WLmL4FZSU{5lQB}VBAhf}{z~6>zsC|&b<9gmmfvV2P^(Z{&*jn32P8(|2*#R{=%J(ZeFX?GZLWg?d$YJ4KiXDDRW%$`ScpbI za=@(JJCI=l39NZ6f{&(WVi4>sK+^k~C(1rZj|qCkC76Z8CP@0(7lZKvm`Dh(=R6O_ zeUhRAV09uqr0*RP0XJJz`vEM>C!74@F1a|~ZEgkFBpNST7w zJoDMw`)#G|l~jwT?n3Dt8bF?sl^TFx9*a_0SV;S&DyC1M;(DsV=;OWJle)G`H&NN1 zEQ!gw%fV4C+U5H7Yc)n4=V4o@KWiP9RS8`!aq)BXHWKP1$tsWq4tF&+{V{+w5EHJm z0az*MhAeonO~P7h4qPmD$cdIxq+V*r5@G+8EFlY3c}YwAldC) zl8Guf!5{%<<|_3V$p9U;-6=C2q~#xEVa!os;D(OwEw{Gxj=$a`sLq5MMtfe< zeO4;GhnTQAZ-auB^)3xCLSSIA2#;w>hPhiv$j!lsaUIYUgkjz?l)f6fA@G}FL@XZ! z-+aXf1@Zob0j$9=M|&|j5UP~3YMc|47U^I=5S$GdEe~{+Qo|~H`SL4Ryy~FwOHtnblyP~ARf0F_?Qw^IV}IO z0aP^Y?--xZr}}3c%La&cSk!W0k}P!515@Avd%| z5}!UJUvxpb${GNg=LN~;k|2ozW9qoEg-^5H& zoq?Y&&un)lhz4O9a6qbnWCb24G1yjynu4}kEuR__^=wR9qynhOk^_})1e=0_eEI3e zZ)FGEc_m^_c0AS)5^OjpMM;H6JtXwOLq}+h^iNNdxv+$N7du*?giS)Xz~lWRaKgN5 z)z6?qluVj*m}FqlybF>u-<3MD^|5FXZw|r+I4CfS>0^Cr{Xf!@R%>pYGPstLChl zQ%V5~HwjoT1F+65tdeyY@(%)41KCQD1{VaZQW#l$T{}b$aI6MkRrnoXu`xrNA>)VE zOVL8CwP_sqHLt2(c}s-4?q+_=fIaN#^Tefs%WYZWimbf*Q1(&Nd3`Cltg zjT*I&EM2-l#*W(yeKDlVOxp=e?B9`*Sl`wuwaHGTX!QpBQ?CvzAdN1DxK)C~(cl+@ z$~5u~)P~OrqlASOMOa80A>TQ3b@T~1KvV7J;SE3D&cFsu8-o;V9WoLURxze3IHK+g zLFcE?&gh_MDnM!?Br3|jeG;e$No)-0dju!8Q@U1atF2O^Due2;tUe6>{(=G;sHmRf z+|hFQ^&?QFr{`)HIQWngz(6oOgf z6@qnwZ7KgO)gPSG0^YOX4kgX4`m<6AtiNRFlxXO?@3mK+)y(JTuaO)6@Sxmv_uu4r zEan}`^?IDQLR=Hn+7C+&van>~PxsE2f^F+128+YveORpCP=C25v1hg&-NGgJ>|9uqt1Xz)l2C%rz z^%DTB^Whf!OE+LOHW43Dy$e4>z~TJPSZZOZIt9B+Uv%*w zU?lOg49w1!Z2+#*&$vV`z5EZt^;b^#%3O=R*PeXhewlgssWNELC|R>+u~4GA`PK*E znLRO)Y|fx9)gK(wLZZqIaPxG-Ojur)b@QeT=9cV- zovAs0RRu5fl)Bmq^rTLS>FjeZQ+hr9)O{H6#mV>czfo}NhusPC^Eb*%|9ae$%-Xd} z<)&L6mg9~;O{Ffb$Ldm#?5s|VKpmJ`3CF>}D(>#kr@)HDRKUV-JWg)!2f!+-lHXT2 zOlT}iN(UHySrEz zb>vTh{ghpHGuF%;5f&+v;CQ#s8d9`SF_amCvCd2A3JVP(1orBjkE!QMaa?mI3%)d@ z1na2as#k_W^%sl#l&E~b16#;(*WW5Crki3bL~(I8P^>M(1U@IA6&7xTw+g4E56F_= zoqv@K#p3vT@4i`9uUVvrNz?qEgspH+Kl6Mwd#81vv#yJQ3Lj#f5QJl*LVf1Fg5~RT zV8udH_la%-PY@zk!KrMCq^UV=> z2u9zb)umEdWjV}^h^6yeD-fu#AF&hxs6w#@Z73uv`uOZkK*f^M^NI;HS+^uh#cFfg z^Q)#Okf^xj$Dvsv+F<6eI8yx(R)R+%`Sg$im9To6O|n7Le4PtH_aE<)uYL@XO#rM|I=gM`%Uw&5 zZcNP#df-4vPK3qZXn++%+l)34Th;CnkwUDIV{8aD zX{Z8K3>fYpNL1JuW*79^y}b1FTlAK0!lFe#$Uz4mseonfdH)jN!y_EJ7VG`@U&5m> z%s-t6=_*rIhe>cue87Q+%jEs0sD#FTMW2EECBV^yy-U@8-kHO>&R4<3QY;rv`Y;5d z0MTXSA=@AWXE-d{Aq(N2DGl>kv=pRAZ1YSb1kfL<1gZazhSk;Abvd#KmV&MZOWo3z zxv(#a$PQ&q&Wh|i#wGnRk3|bXkUgLR)qYeNW9vdW6&8PE0ak1;#M}TY3Ub^%fi8n$ zB3x3rX|q(9LSF#_#bu^I(t@OqMnLP;?J!SI0-y~B^P|GLZJ_U-e0G1T4e*FQP(Q=< zQZGFBu&iIVTIZ9zy-t{XQ0n3W;UB)A>%Qfb?c29nfCPlEgTG7y5`U@ctAk#hh8Q%~ zV0zsNRLl@}28e{odN_>K7}%$M5&AS(k-(%gnGBDy&`G@vOTjDLmVyK6ucoans-;27}H0#?-51RgLRF0W7K`DnUgu*V6%1gD_7; zi$AuLo~3LL)F&&`n_HP{%-;o7l}gF4>!fT8_U^_2Fg`m~Vp9^d^0~!FDZq+rsCv_W zRFozO80KJ&)M2}i{_v1KdelU0Uu4Z)z5W_@xQ5eY0t??Uf#)>3D8oKybLM<1Nr?mW z5y40Uq7uyb9)GM6`?Gt~LYwAwR)vH%Z^%`tA32C8b!&<3Yw??XdT_vR} zS4-87a&g2)O7e&-Nz6)>TDaia0n_waKs`ODr<37gib(G+NbjKG@5s+Wxo z(Xti}a1Viel^v1?Cc!l<&~+=j@O{Y?upm*z$2Q2pd(_L^MZr>3X&GJkn^`;Q$j(wh zpu+sXaR90mY(ej&1ffG6-_x&Wzlcr(T5v~Y=?7_|szQDK)k9*5_d?LJ6WD4h)*hCJ* zMxKWrHdQ|R^uOv|;+m_@l-K?+9mXfj{`4mPMm2k_%zOY++V0BYL zj6{`*k@K+ts(7rS>YJ{Unh_Y>ik>N0huk}fKL5DwhcYofLv$BL9#!?B>f%lQ}0lqrWDCn+iYW%K6sa@)-_<@@hHmwri^ zGAMV5+;aP$B?EI>ROxNml7~61zsPH^J}qO%?kkHHeG5+)Ps*{!{RZ<==(*f^poaqM z)ak63^!`zjn}eOBQRiCEKFW^`R;>`4fW^w7v{i!9#IwcKaxKjlr+{cWKb6c5h0q(+ep(Kc zG29U$U(fkOu9|t8md*Q5WO$NP)$EW76AqM{ZoOYdj2I_*c^hQ*(+|t5uRNtu!-5m# z{EIJzN-$C$edJEub9#X4*kjL>8*jN!)rH0@yLbA&KBr_gd}spZ(jsiI@-^{y^*wef zGnV-jU^TO(T9#p?crZF0&D#UD(-E4u`*1a^1osTtCL_bvX*sDDB(LcSASRQegjE3( zyCuJd$O-_K1H{Lm)x!XHRbO5GoDx+LGBz42zbOE!*fxQxrzqFerEY(EW{?pR33F?p zxh)0#zj3D~%bMWhl|cuXnX=fEfHJq^yUk>tqrD zi_DDX%RRx|+sJae9$8XU5Ilni$v|u|k`=N;20}GRcy{6F2`S4kKWaOc!pp_a>)~z& z#O2P}zMjW2Ks5|NbqrL0Q822oC91}z#kvX$ZGa~O3^HIg%WDQFq;~rDt8EC@^@}v3 zAT4{7+K=U=VFc%?=6v-Lq^i?-OaAYM`(^r!i}4Jl1wA%z-XPCE`xklRKeJ`PfWeTe zR?9#C{h5pzvnSM}F4+MO6+i#<9n^x~$dbiB%aSGYWz@)hWa=qr$q`4LBs8p`C7^GC zN{iCukE#Ic8YFW-v}9*t&kAp**d%86UD#mNPTI0=cyN4FgOhj5YUH%CS_wxF-3K*E zzzWpPNP8@JDFc%cnHX9igJFCT8B`6%9EK;pR}GTs5&T2I2n%-v%NL76gv)feV4iA0 zUj#DO?Z^l>Oq>Gq^hls*pw@3!(@U!WecPJH}cRB2V1lOI4yXvZChFCfpM{F=R(P7o zk*!pZxU`3fAA&p4OmBVLLxbzPv*1^o&pz*WW2y zJRF-;W}vK;94%SGmuBwT2|!~PwiZcBY>*@Na>>a44N{K9!&(pg(~W^ZRgB)a2Y~7j z02L)F8>lRTlSy&fDJBXM73LH<&|#mbfx534Cg}vKWQfAUp}!?ik!}p#TxHfrcH=Jr zhv3751|sbVG&eW%bOaMhM<&!^?Gy`3fT1x2L5P5M*4dY+b6EmaXmG6Daog2WQc|Sx z=+Kt&=|MdsBU{bf>3o)wmp=d%`y-8HVq?R4#0t>R)&?s>iFW>yU<4!{+=2K30M=`8 z*R&qj-vG3@sbe56BqlJ?#62E>wY?!iK6MV1d9}mfVITqJsl#D;yYnd+YbSw372O~c z$GT+yVf9i5Q_LD1R1M;Pmj?pXHZU+Xrih^GuaEBR+X7_?R8;rHV{II*{y4m@g}bCO zfYc5eQ2@Gfp|9m4-#~DiL~shhB&n{6!gd|pAd$fUkzRQHp|u{>d#hG1RS@CgU4k6% zo3L@iI;n^6FmsP!LzDA*d~52o8LB3%t1pwEfBstj@%-Q5+qW9?KcSiqJyLL?FaeP7 zn>0+?J^h=U6oEWxwf9I?Y_Qr5va@OO;5aN2wRb9{tZx8X*TVw_rK~FSV}cfgPno;* z$xi(xU{U#(h=s(Put(Mh&Y`lhZlHUgEY4(E>Yh%0qysOrbI{ub8Rg)C^)hXeOLCGM zWE<3iR0(pF-xY!NWF)FFn4>xbTYN;=sy`(yY0m~94I_$#SolQ;*KpPcW6fDOBv7ci zFgL|*cZLI~`k_J?oqeN=sz&qBm46HvQAJk<)f;{F{jgxJtY5!I0Ybr}nUqCf3x$OA z-M3%K8K)f}AARtq0!%bicFD=9?(Bugk3Y;)|9u20^Bqc9bYx3lJLS}KjP5TzD`UbiV&mTU+XILjCM z(^Lj&Dq}Z@O`xJ@ia~B;ifG%ILI((J*xWFaK;^(3VjX~~5(|54Q2$O?l12l#24HR` z!QIyV8Kl@O2p|L!9`qnVYrHjU7kOID@6P+Zs_ZytREtdwjn{Gl4JEL5-hM@Hzx5K0 zKjZ9~a@LGXu#0cLoIB%i$-!3h>#&3GOW3A<*znORHSs$ps3eE>MT3bqUjK(o-1i_E zJb1WAk}~o9>6gIdgkeQ?Rt$_Q;QtbRtf#;#VAL-8XrL`7HZbHs>8K`LUmvS#K?e#Pi^arYD3MimS`K!9U?&035b#Dx zPC^Gt!)UiYNLKB@0}vRb->u<1>y2|%*;w#Pt3Lu&%X3t$MPHP(`)RqWFHmYFo#Qw+ zg@YW5j%O=zRKeOW7}x1AM;{2iAWhOqnVzB!frMXgk3?dm7lO`>H=HNs<#hgP`Q1xN z&XRAw`2ZGsn`NJg2Vxi8D3yjxx&)i-oWT<2h{OWmIkIr!eEHXlk7I3Fv~1s2h@AuK zkzSEx49Jmj$pmRgL-mDzCD^hk|MhM|W0qmo+*tSd4aVqOF;{Xli}_v5Z~xRiJmEr zhA~A#Fd^8$^se6Sea{U@MPg1W!U2^pfV&3Z#JQ;&NL9q@MEvvz12Z7i)6x`u>9I|6 z&Zn(lNk+K)S2qC@euKXM zIB)gdJFm%Icg~d5l!1_}HlXFlNK7nTPoeFUmTi^yK3D|HzjWQQ!+@R18`CkN`7?9h zAASi+l&aEFB4tQ!e2-_z2CHtU0&m=@#l_d~AM_YhgKuGt)_AmlSTq8fy7zP^-<_gc zV`#c;3dRUK2z!6k$}o7q$PU>GV}vRXXp!Apad+n{?R$rPo*jyl{@f31^Ohi4xHed} zRbV$-+{Z?y8pwcIC(Q$asv3#JK~*y~G7&o~6R0Rd;^jUTNF!?UBEq=+11>NUsDWi) z6+jBc(+UPu`4Ob3^5d|iW$Dwmgd^kM2J955aw8qQH&a?xqBtze4~ zl>YL!*L9nD)_H#3M!Dnmt7XwIKS^~>k-YZCx00JXOfCA1I*;ja4k|1xROhcQyJV_< zj$40h+OP>e{8DA(rj>H*?SD~&ikg}#jo(t90_EPw;fd;H9Yxpv)>HXro3ld~Q^2q~ zf)Q0_hhnd+=ToEQ6<7^YwZi?f$SAoT8!MnCMvB{h83;nmZzNb%KNw+f?qXhDx_kpe zj5T$c@L0gnB~%@%BD9P#x~qHyDh4O?tL#Xb=`eHO$0bK&W5*mgSKR^aPAM#jRa!t? zQ!jjnbgxB8!jQ-bu=?wVxrDw3RHPYc!w?fu4=X<`%ZP%xb&x}fT@kvlm-AAPThvw3 zP)uGWx#y;O8mK5G(Ry#ys-<$;sr$)sM~{V<@*i5=BZ?qHsc7B0UuEUW9~G$1{oOUP z@8mnjo!L5c^xPsGA-A`T!G3QN|C;0XBf!&)g=`4OaOLc<-jx%&VmDhjFu zDF)KfF9HjKFnPbDbPk9ry0%nnn>Mb~`24fS9{T{mEc0^$Sq&ENX6Fo%!;hGzaq{yw zsuf@J^mxsKi+ytt|BtgCliMIkZP~h34J-(L6;)g1^H1MHJi`S%pACo;l9je0Dalw& zhu|i2kLKO2DCyBW^y-hZQ}eL8l(k>hg$WUI1Xh8+1=Zj+RhS_GXzdFyV`b2Cu%}0@ z$U?gV%Vo^n9RMwY*m75*EXVq+jG#(*!`O~3HA-MXNEwU^NQ3p@2fA8KNYG-ZqNGIu zAU%!xc9#sw43f2*F|Sn+EO0y`P7oN{EjcAJRNX|><@uzIj%_>lsjBrc<~d-&K{zk zA9$}hxOnH6&rP7z;~hvNf*1FmVjy_6W&lCSj^E*;UoH|pY3f-rVb6VK?l+%7Jvc`# z{>I~)QkWk>yOwKLf7p-eFjtk_FGAumz-HjRo(?ux`KSg=U7Mzx1C`2ENs)3KwjOyK z=I*}-Xi+D#54bE9X@NGUMTTZ1rzk8eW(VEc5G5PyvLy-9RWjzY(nE@+e{dmGqfQ0z zM)P-8f_ajI2sRhloZcGJvKFt#C zbkp-R5Kat0u@qq;a5M{piC`>}as@+*z|g1y(o;PS1`DLHH5J+)L5ju`46I28f20Ie zdk%*K9?JJp1SOVj9+tGVt2;sh-Tc4OTv; zK9vFtWDYzZ0@Jl?0a_;lwB7?~-2>3t3I@upNVu?=Mz!R*fy6o~T``2AlpMQOfH4-= z$H~fuILQo3mJ~S6?GNBegv7pMlWk2p&|&8Py!Jt%29Ukdj<(ZxHTERd+^=5D}gz@4WSj9DK-8@FX6kR(tfI zzF__wDX-qDY4q!t3Z~83W%fW&4Bw*UMH(+(wn#TuOohdt0W)js`5;Y?2{yv z9DfV!GqeCqO@>qzZ%Lk}AN6!z4e#vfeDt~R8^V(*Fc7pTX?=^)%#$(L3LA}v-Y|;M zF(_$K7$@@xq+kIM)tlIq57Jf)#_3xzxAl`NMLuy3mT%#MaIGs7u6hzM%7?!-{No%K zjo=yhepi&4osA6GiPU0uL`m}2I4~W{2`h~X65u{44$>D*wJ}CeW<(}Vro_Wv1GC2b znY`jk7w$&GnlCygSR$ga9xBQq)uEBt&L$i{#o+@f!NtASV2M%`=cMQ&DF%dMnqIi= zuSIW?FG@cK6sUHy3@?^p&ggwz$g9#2%LOSYBQsmF2ab~1xL8@R;4}H^%MaBEA}T6c z=6^Q_KKm|{p+oo3^oNfaqv=}hzyc~7R=o1kGcx^*gXNKj?}QrA@`!FpTvk3PWwoY0 zY?5(5ykJ>5!0DgbC{_7OHf_nOhik$cw0NEa(7Fi62{6L=68mSpSm~7aAOZ9Pj~VG6 z;!xVM@H$3@1!y5eAwR}ag`I!VVu6jc2`2 zMoLD@(0FEDxBWK=sxDVBzESOv0R;&v(Gm_Jd}tt9Ifp~Z%LUF)Gk8o%s6mDkqXtS{ zO+X9rO-|bWa@Rf8tl|%U`E4E{!8FIFp#-kkc<`qloTg6YRjB)@;=>CjJ_iE?Dbha` zh7+{x7g{iuXTra12CXGlypA*_#Juf*+0X3`8%@DqyyBO1SOrs#C~$xnZK0> z?z>(F4H_-C-h44Uq*p2^{b|7Msv0yR@3%sni5nV%TG8h5L z1*YzBhDgq)Wc)(T130#2`RC@}nlAIht%>S9hm}VKFFfP-n*f=WoyNal0BtdUzORWi zzR~22dC>!RsoH}Us_{TOsE1TRkB!_ziU->d1HZ|G>FYjVxQN0*bssHL!;vVTdB?|P zzz0e=Tu=4v>__E^zs=I}a4psT2OI_|s0u1khiu8)Bqtmug~Ub4i`Ip(Ve^#L%y;c9 zb3wI-w6gP(B)S3~4gp+-rtKw-#&^F&VB*1`Kn8BlUn70hWB#ii5i}mYWa3(HSm5FP zeUr8V5-xxvV0>U2_@(4T9>{wQh(Lpsdi)ZU=m%fF&u`$c{x-H3f)GK9X8N&sCkYA0 zAt9<#nEgNnDKbRln0BN%B12^R_KkAa-A~EU$4piA9OtI~ z_x^t%Rb3>*hK`4eq%v`YhfCh3JgKcKmOtG)8|LdLLXvVScvV+b$$#L>@1X~8#4f-4 zVm;TlG7IS)bKEJq-pb^~)NvsGvi&*Xw168cW@hxP55U&wXMAnm%zn$0yUr*%6$2kOqFsF0gRhNh62E-GE4+mr9$e8 z26*wH_l$5zWXPP|savJroc234HRmGW#~!&$ zUVHUv88&PJ77G6Y^Y;-l>z=>DuixGZY6Pnd>(}U39&f+>Pk6H!qf*zbKRqo+9euLK zQLt>%2uv;l=ogKl9wQrWSwDJ262RV@+xSm8Y_Rg5x@a>e?20_Rn^7e?=<;fpdzCxTbWJic(ZHMuOTcTq8=D`gY&2M+=+T@vu7}_gN>gO#GhTNoeQx71rBppgm zIB-qawMu7FXqrn;ns}^$j&zL_l?W*+79eHKIy8cmd50+mS}+n*MMb&%^78_K#tPkG zcjV|jWWEST$*ifpI~8#|8y{ zFS06q%zbZS5jxC$10$>#@0?LO;xm4o!2prD$NT1)NrUmMw|Lexn0)a&3Q+iESveO) zDT-=7N>Pkw;^5T%b)E%gpoL#p7}Rv|-FMHd>t)`&kD3&OYP%zkI#~h4fCt~91jeo6 zue;`V^3zXqWYnm=W$DuIWeSWZ{_lpn)xFa8ZCm8uFFqx+pLw83n$0gxKjUHm-Gx|7 zHCPt{^SSv}^SFU@O;93GB_=v##Ly(a1SvyL8>|c^0`Qj%iHEneMK#bSs0M!laDA?7 z!g?um1J_kJ(x&G5B?{N4669q!An}$Y{K_ zp7E~*XtbxZ((=+eHQvKB13vtmk(NkrEAZW>SLR7;zb5B=%Scchq(tH%IK|_jMIXz} zIOZ_#wU%z6+&8K}8afbkzW@FUWZi*zl|eH7>f zL*Sys^)xB%21{VgPXfP3la?x0Vf%Rr`WZu;Wz2J`loKIki ze*JalU@cstOga20xnkx`(my>@rJFi#8}C`w-L%l={dsdglglqXMKUvo!Ehx+)^Av* z5xG&~@Zn<=*yvZVxOj)UN@CB5fE%T__ym>ExZsy*^(LSqDAA~b1{E}{@MoZ6W!YfW zv~Bl_D>5%KEpNuf`Kx>Y*D`?C_c%Vr@dXzYA*3Q=Hd-##g=Bh0Wf*7-kkUml%5lPu zY8h2rBXu1BD!kfb!PFD!t-ym5&TrB21f`}}j6h;g7OMDq13($01gg;p0Boo&4JjCp z;J^j858QXN{P&ILK#^#9=l!{om6e0_Q59NeOs6%K9Mk7BE)@Roe{aZ5H(nrIU&Y0} zn>Vl0_uqi$bgK6VNc0zI(xNJn-|lT7#X^H5C>u7`#5UWbh9|*r#k4{tiC_Nl9*Fy8 z2JK9BoC2&=Ha#*w!vM+Tvfz*KNO;G%5tFE|rA5hncvOA{i~ml-{f&6etxM<$J|9ei zjFRr2xQz)X>Inp-Wn1PzV=hM#LZcxB5ME z0MAKMj*R6X|N9hD3CLNeMnLd+hP*!LO&8|LBRSz5c&Y;ds`R39&R-)v@jb@p5y;?# z?0W(w6nZFMw)7WWU>gk=M~T?led;M^OERqUjE_Cf3n+asCRD1bDrMbVFQ(J(^ZBZC|S7f(D6gD6!Fkh&Arg=;a# zwH(JU0J3=iub&8Bd=ViR@6ZJl0gQo6*npQZWL%qX@VWQ5dEHfk>ALU-dk4xPg;;Ne zeD{?4&%2DK>|9Mil^&lF=(+wW!hjQKNKX>jsY9QUCgwfTkb&><2=qWnhijw=JkQO3 zOg{bezn&sd;z>eJ2BHnj8nvg0EWN=;Oy z(hna80}x^XR2Dcf&n6wudtM(l&%AC0{3)d>TIPK_?+f8p@DuhtP=5Mxj+}qt^>X1w zzeo8DA(+LM%G-dM&3W`wYm6aCDH8anV zg^T9N{SUsN9x1AIi+NBNpQ41vIj?O6`4STw57TXIWD|g1(!Gk_?o-le#IQt_ssaV5 zSU6j86gKBh{IR7Ca5)LZGen#dLNpA`=0kX=c!7N!O2)pHLRpo>bhW(*qaXC z-!i4~5P|!g?=!)ai|LraZ}RBP!AMoymV;sQrVaXg z(#g{SNTDkAGzT6=a6x?+AD<|P9(J;P_UXI&&b)bF$TX-9TVk|i#G%*@W!>6e6{z^a z`gN-lpp$v%4ZAIvoiG~$a5%vY3VtaLBa+Hp0ss3w;eoA0J{bPfbpH@syX3H>6 z^J^F^^Tc}uFs|{5!^Ai@k(@+#83U{ z$BsfQIDB-(Y-jZc(neDiwpCp{{pSJHdqB(jrBl&*^QfbWpCzgo^u^g zBn&rFRFr6xC#V9Xu?E$NMd&c)I7)C`001k&c`44h;pBE4rHE6C#JAvg34Uw!L+-D+ ze>@}rU|t8yzi1eRR8#%O=gf>dUQ7JPl!Lzn76lkwm~(=T0VaZG7_JFY3Ycl`?VCfo_S)o0?Vv1QJ^3(c?MI&ZiuDj68V%4W88h z@xtSB-Sx9H4y_6e_)uCZE-sR{{`<0|rev!}^&}W^(CV<3z~rSiWS;o1F9dWa)db`W zj0G*=zr}qWNM*OdDv)Z+kMe4|xcOz)+!*2cHP6_{xK1brjKKo3oeRQ+$xiej2uk`} z3E-;6c@2IEc6E40&-1*;t>|e)QL&>+BFf=K&(YMutNAW}2}0Z#7ktTmZs8eT_W)F| zGA--cyo1-81bkKle|dPT@%Vu6F_>TL%Kob~0euNR1(+W);m9LT()lP#J{ScMUqZP4cJ14X7Eq}e*W2e z*puofwe;imAHzqCRmmw38ML8SXMWlCxldJgR*b}A=}Mr|!3L|&v;zCxCKND?+WeY_ zhSBrd``pCiFQpjH`B>Z@zm0+!U)woA#kdUfRHYat1`~O_6J~5C+Wg+msWgdVqMLas z@90dt562B?yg!ZBB6S^x1~AO;((0y~hZA8NbA))tky> zJESTHj6R@A+&Q##Z|7Kgr=>i#bgz}y?c2McfEqnO>WPbuvtoiZJ|qu3^encrh?f8R{n`5b+N-C_@yDMfLx+vTdZ{@wckYJ@WUE*G zs?RUH`1cZt^<2Hpr=N#7ZJmI~-1#~pJVdew!n_IBfhtw8m^N6oRoV74pHZMCaAC2BHKYx{aQ13q=OF@LY)6$fjYp`B_ z^?7;Z;afaO-*W4t*xKU|fme{{%buPLSLF_>k;@ZR7V?I#I!j zFLTpG`tTbvbcAm6aq`sZ^8JEu0BAqT58o}o+?dFD=U*o$oOA{}SLEte_RZ(S0*Q3H z&MU>!ori0x{zG4(gWALdn5i??&e3+AydTA|!OD*+!_LaCDUiRl0vqE&{YRq;_CP!M zJq748i2#!Y@VNA{TjYtyXJI>z_vOzI#LFR5j+VU58|2|X-vfqesK^ORzmz78;77=rSFC{su{=BXK z{GL#|mBF16NP!9l6$Gh_0q%LK-AdcPUb4Z;zv{!zO$!t-QdMbLt>oobVlv$ZsusxP z_qRAxHT9Gk^2KK#%ChAP(~6dXJbFJ+DJf(aMt%SQCy+@*K~$RjAy9EWRTQ?UA3TVD z6a#yn%DZqjSb0;!KKDxsu)|T0u_><{Y94y+4h%2Reo4$8$`k>muC7*+l2YW)k3NUp zd7qRQ|M?fq&G_*LV5i@cfb*@YTcgCslMZ~0!n4;)?1Ejy;WWWKLcI_G#(86 zR$l}vmfafI1xapBEam~>%^t-)Ra;z3kT5s|3r-l^-QArKoB;-Rhv302xCEEr4uiV| zcX!v2;2zxBx!*qQPuTNR52v-NyQ)iHWuGz05It*q8GD#&TdcO-njO|FYF8ST5X)hz zodEJIFOSaAaZJ@6;F9T9dDyTeKWLg(dZMMJHX53t;!yGcG_)2Qh}8Fw)TpU`OMZ%d zoV(=Wt3i4T%j2~aC?ClPJkOIoKmoiuXHk-O*QCWTUaBlv2aX|{2wfJ_8n@OCZf}3H zx;gzHm{m@a6HlDu#{24+Vds*+{U%v#Z(5>_L{b*ek%+BMI-N-doYHZNjuB$9FszjR z$Zy$cLDy0`+332WLM4RQo-}{f30hor1^H|8WsgQB1S9Lxru*z?bPvg!9wp}9-WgtE z5(0r0Jxl7o7vG-e1SB9Bh^!bDYub4$h2Uj&OSVE?VSd*Le#bGMb@g!JH)?tCgH4(@O!7!g9nHn%B?ELK0B5Sfl49%a6c%t(oU#WqnR<^r zNH6cLP!4d(WRWU5A({J{91MF&PqOYZ^D%B2-`Lr3AhK}EDiOWg&W}yH3Q^G*dPsh^F@F0pf*y2>A;>)D=KT1ZN}3wlXj?qTmjG8!f{*Tt>dBk zHPg)4;+2v@!uvjIp#KXrqX|{r3Lqzkg3uS1^aZh4*vNNW+YMS1eMQW6vpRWBz0i$( z*r!tj>QA?B$KNwHtSr~FrmK3SCAF%a!d4!Sh<<~w*4inaJt_(f0ul1_xah=8P-N?d zeO6G?WEj^wUSn_}*!ulvyNh53MVFrM8A)k@Ye$xWiK@2c<r#CJt zHs5`EAukUTzb`B>1-bj^iPRpul+YayTtMJ>W|TyNXRX?)$y_y*Bn!PHEd9M zbt1-jRHj9*p@9N!=20=kT=<5QRtoz1!4p5P>o)wBL6@eY>IW5$zxGTGPTZjK?(H&S z<#`Pgl^RWtz}ZYMiXTRqya}}HHxJ8zd>}46m9$`mLHkBCj6Kq5J2mQJ05*2Q71TaF z>eVfxi&%simb@2Fm*cN=06G2*4tCWd3%E(u57=x=Oyc?sYd4uO=<@!ZZDdC-D^Q@4 z*hd+nXj`}^jjXAE1I?F5OBF`zjX!+tZ-jbR@ViTM-Ctj8{vT>H2(oPLdNvF`#{(Do zQ0YESzV36es2IN4w(XRNjlvLWYHB4{49I+slEM^Hu5P!rOe_e;k zNsE6BTP*>$X8{n)3JRqx2&owx{)!u&lJV-edtKM#oYEuEgE*44`7Pwscx11dTTr7Z ztKN&nM9Q}XaUT4goFbCbQUH}*Gey?Y9JV)}kf2i6TW<{ZLLSoI!(AII6X$d1djD5W zdG+2)i8PutL*GPZy(=_fdg;}`ruCFr7m<(gt($V@<`Ly@%xuT{pKI62P!9SA2bY4b^z++=rGA zdwd$w&ziDo+lsV2Nz?b}9Qx}zuFE1nC3uYmvtkRPTT`}N*Nfiz=OIf5rR5rG(@L!x)OF{b?on^)_jFP3bO=VGDbiGE&J=tKePIC2sv&oqf|GF=0`=<$(EoZ($fCgfz zGbl$K=NNaldH~aj;mTiB)CwORN;lFM>bmzsuO^X2@5*hO;ewl|{gjJEA~=vXQd&TH5_{}j zO=^@=363(0yx_V&bSUUQlDw`(6Mv>Dr^%oNsEXV3WZ&U6O%JL=BUxpJ8>uMN1_yPw zr0w(SGoa=rziPz?m7vWv`usL;us~LdWI5;k3tie<3Qk&u@)Hfsc-J#o5z()wvH87W zvV90Q9LehFND!kyM#(%Dl3KJis)ru;rVc%y?#^)qD^%go8HmLy@DerVLViCO46O7ttop|Fg^ z1kR<~>DE+gtNnaZ-2Z8kPmOXldSVE_ot6bH;=Fc03Zuc^N|c`?i|)YW;wTFnT<2s8 z$xgj*a$HUF<)SXSV35Eg>!NW=%MH`38u$$gh!fj|Z|XVvY5w_x!_5Lv(bw-Faal1T z4%i!`Q61LO8~P;Ooh=y;^~kx}$mahip074!7Wb}r{x(0)>2YfAr=_c2a6GV7B%%Vi zi){AMeHCDrlZj;H4Ren;u78$FT7tUk;JB<1bQJ_9FdHZC9r^+!tZPkVL+QJ zB6q13HEoKm{NGk;|G7NU%S|9w&k)voC(Dt|_ zw;Pf)btx|lUr|1C64?j7~hZz;AhsMvZZubI4ntd)iINx&9Ur9(x`U(Urf{BKvwp0UW%^T0?}jc6*= zcr5WZ$7$|p<+F&40qZ|mo{4pV+{GHB@?rEOI<;qWI#ei+;gZqBC3qn=iyC49dwA!^ zZP#wH-2OwHnIC+2SBkXX)PgLj$gUk92_y8~qj2imfmGu$nL3y(-d?87XYyw2{m%r0 zR=;aw*W!!oUrjVzjOX|$ZQ1x>QgI0EXdT717{RrCt~vRoBKq2VD)flOv!5qD_?5-* zrf?!%RED!5Y_uenY44V=3p5%`0{lnK({CmwtOX>eRGeGvahnxmZRMd z2e)=H(1Yuc166am)i}(+?vvK3EV}A31OiD6f^*%g|H)^I3Q9>)9UlDr-YCIn^KU<% z!fdaJZ+|q$MLAnm%FyKV~60aL4q9QlP)&RtAdOR9C@H%9#CW7=9>T=!D5a=c9 z`dzRu$Bd$TH;8|h-bzfRaCa}WfKf)I89>gRI0Sp@YhK3}n0-BeTt~+yQn5Dvp1ugS zN(=rQXxJOGb4(c+Fi$C417Lrab~Ao{^Ql4WET8We+++lLEkM-uHQ`6kG!zua|MRFH z#pC*#cErW1b?v4YYrPEe`SfvJUCvG#uOUoFna7 zgj>+FLV~shnAHb)U@JLRCTbG!m8PxthIusf?QYM7YDNqKvLM)pEZHTyvAy|Wm^bO? z!3^!nE%|v_tRPtIX zfPh;@u2xJf_xU`J-;=V=*LY2QF-x7rXgv`$?pAlQWL^DE8sIq0%AOK%R-K7M;kdb+?2drXh8n0)c`%vu4#cO?<2F-&r3LzN; z7a`h}5$eJb`-fAI=xqpPyOZ9Un41^Y@J6%UXOV79)1kz6`eqNAI2b8sT=8ki)mzIi`qF9rP=-o zByN{#>;hy+uFl@M)-bL7EE;a|axu?;?_J${+2^OHNNHL&-U4ID#71#issxz*dUMkqFMP(y ztMyIBo%&vjWP3%Bc@bHHr{ZQeHC`*4NI?>Lb?@TePrVyi{0ip;t2^AM1R3f*h!rI_ zwKcxtuv6MBy1K|3wgmcD+?*C7mr;r7@_!t)53W4qi?Om^N}I;>PTjdGwxj8aC) zQ35svCqLiTS^QE??y)`3n(-nSb?u@h6JQxX1q@1KUY|X`_x6zv4s~3I9kt<#p*w8K z>Q0WQmpJ3_C#(EQTp3T> z{|CQBP)*DHiuN9rssIO}Vgy=RZ|y_FvZ_09&td+yy0AD~U5G&ytgr~`6fr{ZW>7V8 z&{T3y>A2kvfKz>&p~Rc+zVsG&jk#V(!4{Eiv^tN-8@hoA3g0%m7IzeAzP&wca@sCz z2_oHwfAXh`lkH?gBJm_-Mqt6R|M&N9(nBNAuFQj1>FfQ8#{e(Av7(j(*0BGA2(C_d zz9#P;=!UrV&l;VgZ6nm~Fa5He3Jb@eNNz$>e)IO$^1Mg(?y30^2T@+lIfummf#GVD z?tK>Z_kP$0m+i_7%2iK)0}U6~uvzhP4zap;h}&RW)v}SFA?ee#l9J)`@4v;zinSPs z#4HfpOzCDej{=xthNW(_>ku#zM`1yy5b%A z8_-E`lD(BUTSzwsjjp1h?da}!*~X;t6W0odjlMR5FAPQNPE!%}*7G=Bp~bn|(EcHE z&SxuP&35Hbh24;&CkqV5MrV*;-pJBXP^x2j(=9Z$x>)~T()bnu*oqE<^U}itpl8#h1zyAmXoXt3+I+2x!@P~isy(8UO;h$~%_oLLjdzKV+#c+9DibhB|O)U=} zXC}ECkRq1YrtxRB*&!PGHg!kUR7+S&y`)xGo}$OB%E^AsBMF$`2X{QkwPZ}Lb*5o| zP_5K$TEGeDe#JzH&|#DoEfq{4A4bO@^S_~il(TA*e@f;e-amX#J+Kr*)a^4!c?POT z{^IAzvw>S=MPBSPYu5W*rhl3_EB0CTgJNS)If9`3Fw3XP_V48$8CB|e%}d2oc5v90 z%z3Eq?|0Qat<}eZZ0qfFi_aDc{v(`NpI!nb6|bq3IYU22O=c=c^!uVyn@)m(3keo!!DAyz(Ztu(A(vFN%5akotHm_=(US%rBQ z$%zi6K3$^QEwUN^y1BVQ?(-DBtI6Dhyg!5`;GIj*Whj02Q|7(Lvoq{ym?^P@!0DzJ z=jIzS*wgF5M*evw;SorzETF9$S>ijCH(9g$OM`8P6hV$3y-G4~#ao%gch_afqSJQe zO;dWnojk-7ywlbGB+C=_AO-$*VsI7ui%g3utcBZcc-7_lSkZwWQ}^e*HM{S!ms592 zgZ-n|ega@w9mn?)fto0dsphHHfH68vVOS4HJbAAljq&n&;Wr#34Co{F39tokOao!5 z7_)QB7ak$S2McuQ462k!YOcdPnL-L$t9gSRN#f)WY_SMQvr1gMZSTJ71dDGgUnaCG zTfaEIFNgrXqfQw0K7ExB-+OxUj*YKrz4Ff1UK9S7eYIp;D_o1rehKZ$ zj10~k&}gX-T*}x0>Iw!x;$Tc%;IkxfP8H=SqyOO>{u`s9{{)A7$7>15aykd-vVL|y z`llf+6Bb^cPNY{5v+LQI$8+51wUe8-QK6&JP_+BsMNy$yjg59Iq63P}bB*ClTUnv3 z+mQvraIaHd=-~`bQvgsh3%}h1boF&t!Z28psm9K_nWJ>BDpOuOh?! zNl4?mg#Nj~RI~7YE(vscqdUXD8l=g``swgF{fj!Cfz5D1v8BDg1p%bZjT1?+Mk~yI zaGM_Y)4A^n*u3)4!q@RzzILS~!nG<2%o;y~@Rv3>Y1W%X>>kV9Rk?FVM<+C`l|OeAXKGJ&^WQE6hZCxA3<7c?9vvi6oKDG?6YjLX+&@LGeNXyJo- zv}T_~5x~6xM(*xa8+C6v<_)}JwQ3??NEVD{NIc2aG-+1>ub+7uM)71AjA-uen+3(-IZd5Rn zj8P0T@lO+0+skC{T|6ewcVwffa-`gFXz5L*nc2TQh)rn17O>A;Z1idul;2+-$<}NJ*sf9_(BWW^M3cfpEn;HS31eSom#fG$H|p|;RY_~ z?6@X~)f7^sE!dR~k)+|b{mE(g^9{L3A7fh-S6STyUZ6!qm^23siXjH^*FOK!^ywH0&`c0K3`hyWumGSw9gg<^@^tfrNMe-I-w##G_ z-*_xvRoyp+9yeS$VY=5O>I3eqc&;#vzOoFas+!zU9;V)@r z=OYqiE<8WtX}^WGiFzOQPLBn>0?S!IYrY!g6exEl)Uz`y$HT`H|7~2r-yZwEE ztIH+2z1NFwVXw|kzh7r$Cj#W(ssmJ)0=$zJD#BxW;Q=(VWW(lC*R61r!322xIU+u4 z+3iZi20|&rwwG!Gf@auP8&zum9#WJ*DqcQ9&df@Pb_V=A2TYu zPej4(FTT+;Lh1N3?A7O~`re(d%a#ntA5ShW(!8$Mch7ygDZ8KcR9G5CVrbVE<}M_g zs0Z-ZaQBPZ^Gk~2K=_gmR8lFeEi9(0<9#_+&B<+8b=~Z!Ro*tTHYsv|L%bMXCNuSH z(0r4mi)qv{$55w&tx7CGPT08G2&Tt+=x>WyjPS(YF;S@Ds%$(x#ou-dQm!WHnn#Cz zIi?U8z2v{LSQ}8LRkg3haC+EEn^5b#H?dl4`0z0A63O(+Gwew}{j*j+&qjr%VhH8Q zUxVsc^OOYOP)n7H#r?}oJw?rt<0~a;v5YGIcA1*hfrh zi@ttdh_~{$OPNvuWe}hySH8$N*j`}~RKh2Mdh^)2kbX@SNM?Wc0AlPL`%1#rm0~et zhamJvub;Q-O@Upi?c48pfp>NwxG1gF?tonM$NWvJoF*N*N5w_VZ0{`j*%IGAY9j|& z1(Q~lQlw(TPSwUHtxT=Z1C7kFG~AMVfvkoMLtn5rBfywR36w$#(UeFUdNF;o6-Q?? z{VYw^{>El8zI0Gj!$(oman(Vj!Og>`_lCx%pTAZ-Bs-^^%D@MoC5Z}$`hlZcQOb*% z=b%Aig5f?WS?{QW=?n72jDr4~3QCf+aCP)^-cfity-@`-Y0Itl zwD4`KaDj<%e0AqJ{(MJPCM^b>9J zK41Le4i#KExCbv`7=5v%_W{uFM_A&0yLfhW%)~a9+gsV0&q3rg`xU;V8Cn7OV7jKO zT62H8O5G^3_zo*^VZjtw0xiI1k|>M4Fz8MTeEucBLU70 zdOMQWD&-Y5+9nlyGjK<3)LFNMP=BM;V3C8pXr`>Whuk@p)*gF8fWuz~gL{s`y?~EJ zQ;NNCG7ghEd!NoXVhT4=TEAEcfXVovoT+qfMk(hW#$+I*{|M4oEOHhi*cwC z?B0lPMPUdg(qGL7v4;`_E?pO7g^4~4YzFT+e4_p zph!VTDPLYkQQaQ^+a{3cLMC`F>KKg#ZS`+NP?N6!Aj_feyZ@aYzZ2^+2JiDe# z#`)N9kw68p7&=CGQk;`w%W2jYgf35B1>1HStn4eBc2XeNOIu?Z+ui+sbZQyX2S;!~ zNJ2^KJBb21gu%Owb=Sd$HEq8nN=?lFY4(q(b;-^@X3k(_)rP3N8DB{KNenyPtf~=h zw!v{h88SY~Gi7(>dcH0$;%%5_60vf`z~6=W@IG z%J129MXxP#Wd?Zqlz%}mNl|%ZEh6rQzHZ{p=}CRf>OexiBVSy3#u5tQcqh`qd}&%r zc2cNJ`EO>Yse`Kan`wVi4p^w87}T9-Z)YhDX~_Pt==3#LiBPoP5whg-NcumBru^8 zd9tw5OcJ{@og91_cn_K6Yo<5+#o>R`EA2?_JU@DS^Sf+bsV(sQTPFY0*& z$q))71#b()hO8x-P8+M)7qZkp>rHAP+^=dxLjt?;bciJA75E{UXLo(@DzIQN4=sWdp#S-@fZ1fdAy#N zksO}R@6g4f730rPS3&S{<|D+F3HPN=Sv&`8G^8RCv!naoB0l1^Iu+sbGy#ll{!)t9 z6BCdcet-k~s8vmaL-Mhh-XbL;B*DPP{7cQlk{7Q*OjFi_{&8dk{hDWCu=sQMBm2v- zdO7okfMcY%D0H9q;4DPV=8%g#5=DsB%`Y-V=Bk1|&92UMhzNNG5nX=xKyza#0wb}w zn5r8nWe;Ugqml^(qu6>*14fkbMos^MR(rV3 z>W_pXKg-3WiZ)Q(b{KX3?#Ju!-V6mzRgeh}2G%?@;gZUa#vV^y`G_PcT1{*RcYI&| zp-WQyBU2IF#?q)LojEbGYO3NtgjqXm*vMp1v|f&e@Qi&7TA0EwTJ&pG2}gdO=q9h0 zn9s-;dD@L+8jn)axcjmj;t{O1b4FDSk8n^&POEbz;p5Wvs7kXd^O=Q?zu$-w%MiM> zofLGtR0IGdBy`>Q*#G7*kjR8&_6&uFgQ^-g7VS}gIQF%+*o{-d;2+W?$`CfWvCgrC zPq})-)iEPY*;P6M&p>5G52GID(vB0wayl&_{v&>X=zXgpuNA>fe!peHE&llE@FOiMU}8#)C*zj3R9= zocf=q$IK|P1pdP{4iQrp8sKn*%l*&y_SROTQPk8BKR>BCA4-$m#Tr>+LPy0zCrX_U z&t{4EcgpRAlO}CIpO`EU`0w80ihUdrNI|r9EZXXQSxH>M5jL%{ADe!UC#y$KEmVKG zza+tq;DjQOUNO#n8U=QFHn%)!7;7g!jFxDjJ0x*6ek+hA0zKnrw)RIMh%jP5g6m~D zJp5H~Nr)>sRaTHQ(tey1t_BOr06d>8P5Yo%JPtQRV{wpYW|3&Pbo}IeCl&c21ZH%M z&ZeG4fC3AjQoagl1s#4^8a_fi5N0&$D*$^Wu!r!E{I8r?OD;nJdfv?Xx$<5qN>$DO zPWptWGG#^`-e@RH`5Om7&6Ha-+6;xo;e}-a&G#V2tIYom n5U2tj-8YeTUGo2ZuuGq!e*_Oh1;_L-&`&`|S-M)%IQV}6JAEqn literal 0 HcmV?d00001 diff --git a/1-Introduction/1-intro-to-ML/translations/images/hype.png b/1-Introduction/1-intro-to-ML/translations/images/hype.png new file mode 100644 index 0000000000000000000000000000000000000000..89469139569edcd878f420a9e8c52f463e609937 GIT binary patch literal 154886 zcmb6B2RxSV`v#6nsSqtAWmd|T%y_Jjo$+{(z4zW*l!PK7W$(TBmX%OskH;R_BYXQ_ zH@&}~%Jcude!pHX8Mo`c)_EQ0aUSP+@sWNijDvL(3k?koM^r>m77Y!P6blvEpqM?cSgoIv@3u+>GK7G_F@K_YR1G7>zl!5Xk z0iRRx)%SdlF)1jh`a*;-dOdwFQu#9YQk4+!^`O7j(!82MHRDM~(_E&Q6T9j*dev@( zYq@f5>Cx6V`z813*2qWJN7fh69){6hORVulOTf;rA9GV%?<;ZOG`phl{5Cn7@^fwL zySJYh7+m4g(c3R~DY+ce^vl0NC&T^dP+iEUCCCEp`QxikC&f5sFGlmPQwra0MMtwm z%bjOzxh83KtCof7J$Ws6Pz3DzouCMg27?Dv*NIvMKb}X6s@L-iIz!%mm5Vz(=uJI^ z<82fANvxL_{jS(euk@?Dl)4RL=jPUazjuXAbYDDTvbsScxIrgBNEcQaN?qfOeK(H9 zuJy=~_(-{EEtU;dU|a;PscJ(F_8IN(6->W9F99GMe^@IHht7Aaelj+%zR#o-!w%V5>)6+^6SaG3Vn$#^p+Ehqx%Gh;$raAJ=a}dC zE;^8ts}3uA~u_6rpEn` zXD-wRu4r)(wZ60-gE`QupexiijyaWJRAG+1*c($=xq%fkDNRGA8jR75*&yf@GcF@a z#bm^qhK6{PCfFSpCcT)#mEzWQagO<>UsBwx0d@iH02ztb&lu+R^mgSRi8JvtPByo8 zah=e(eGmO9v{{&b-XOe$ecey=3tio$wl5RnWBy0lkIo+{zFnXrNcUv@!l`wTYuSgl zfVDudz+*yEM1*`xQtv-z!ND5ws1=9MQ+p!MVyC3CWTHg81X)5cc)x;J?Si@ARzu^2 z*Y?J?);`m|(7x%u^*#rGpoWMA#mg%tFNtf}YBe8UA0yCw{otwG?VHz!UI*6WjdN;Y z=6-PS>JxUkfB$~+{b@SFd-vZd(<7J=rpOe^P2y&QaX)15Cl~t;;8u%6-$FUaWH%0vNAquX(9>X?b;G z72`d@!}NXPDc-g~9eyRA-AIO>bvaqtlGTEUQWszkQPG6j4 zoaQ&?)=B5DZ$z#Kt=rEQ&js9^xk2|p(NJ1@P(J86F7hw=!ulSsn;nbXn}_V;!jcq|Xz;Zhllj>YR3fzV~)Bb9024n4})J5Vy|% zyg#P@l>g-jgQu}il_RnqX-8B>^m0VoA1;iKzMryco|00)?v?A+&SLy|t8Mx_FU%M~o{kL?lQNA^LEB{&zPnG0==?hm9RT6IxjU8g)JI4kG zzlt`Dfrqn-#?2(Qt#%08o1mYuS>47x_t*H8UpAwR`un2ET3N5eg~a6R4=|Z98?RO6 z3G{_5IB6bBLSZcE@7=^E(MwapNG@>>-;_{8w3oLRWRW1I%_ z?lU^WZ*dfFWxTrtcVF0$Be~1z_B!)sbdy7A=?NK@zcRJ-N36km*|(DMXYeENGihxV zvXmd?_~y=y8X4G*&-JkG8n5Ie2OE#M(W7kg5A>qhiyS>P$ac)jKtg)#3GiyiK*o2?IedG#HyP~m-bI#Mm ztd+YQ)Z9C((=Vf!qp@J!P6M+xi$rzIQ~Ig;*mJSF5xec@-2~jVRkXKCpIc{2W=ZP1 z=k8B^bsSpLB@AXXDHo~uXm2?#yOfx#C$lCZZq=eO-_bMNk{h~Q^hFM{?n6}*A zviN3Gymv4wvUVBYRfcw&f}ZSO5Ev z=IbOypUFi!w2eZXcmV zBwg-0@<+!5tkzdoQB+S-5{(9YeG%>4nVV=B;Hxv>gXhex|NHvM87j21$3I6$L-RI7 zJNNr{o`TQNzjxpRYIE}W?E6<}7r=Mdz=uN&`hUNTNg8wZzpu}I1JBU-;R2$f;4@s? zN>|s++Q8g~z}VIw`~cHJ`b%=*sBOa6P0psuyHm7#@=p}843^j=LZb6Xqk+qa>P{_o$(ICUKi z|JRe5_3vqc2{J-Y7?~IzF#g}WK~os?Q%-3^2VIkAf`+DG%|IVKOb;J&z>XW7di1|8 z|7fcCzorkFnOXj9`p2Vxn7` z7`pHg4^j5zSHLAtop)4X4>Zs&y?*V9dDa)x7RT?AexNM{fpZ-k5c(ozP4eH(Oe{d zj)wl%e|%ZsxkEGl<*2j&ycK-=mGzAJ1n;>3tHtn|wNp*e(8H-ljYkcym)Fex_0Hd2 z7+zapJrr1>l4N|{a{J$Y(a`&8Kx1m<=xDb)ryrIwj2iEIkIH1Z!yS13o-H)Ym|*m( z^HE}s%DBR(XMBa~{>iYLXT2J#|#srH^;ZX2Njl5Xs^vE6}KKFixC-IP>+Nf34&%N81YvNsw z?)}<7y`FrY&$B&}U$*4=o*oqWI|}rwFYXsfb-!XK-a^qf>3EZar(?-ZKSV?Sb<((@ zJnz-%cTZMA{Z@FNJ3XB%U!_kL`r$f8C;jQQ)UXE2;V__bFkOYJ+Xn38#geOFcD{Pr zrYo+mu&xB#TfiRtZLrV?-%kIYRx?WFK0QogNgLN5{z~$#L5IKRG=q zD>Z13S_OVRNAcnhSjR&SPE)swR%Ty+I@{`7*3LLCQXA6k&0nDCkElhxXBWmLgAZM9#W&ZwW2dUm#6q z-+P>_PELyAkW-04do0td&vkkmo`3#{f4olyYgNKa{-;N#bG*lIoF8z$JpGV-^aOgw zvAs|{ymB9c%cZys6uWd$=grAV6RnBOUOYXncSO)OJ)AI{!8rZ!IbYmKW1qYX6iDm{ zka(23jmXjE^!Af$W1N6xpMpdq3akBo2edU@TaPM7v4aVcdZg+nlFc7#8QS5HCxrn)$_sGc6wnR8;aC;b{uJt9S6c$Sb=K5DI>pg8T8 zp8amx~xf7FS+mN*2LD5(n+9aztUXA>x&yv-X6YPlC z5nCZYy&%uYppM;~>JX?V>~aFMlY8GSxKV6k3kKv4ADtaSu?tr6zxzG#UC}t*FJC|8 zN<$WxB2by^JF}AscN`>@pqj8QG~v93+*t+`tHFM9gr~kMyt;s5HD!<8!@v*+*6A(e zi>WxVOPXc&#)H!<;&F)(@-1TlqFX2)R-7DRD)Z{cFjRZ20@>widOkjijy0@Lmb##H z5_PQSZB_SLrdMD8jOE>3(-8@{PqvpYSbhsUj0x2 z8nhk^r}C7e1+x&rJ8>>DA8b#*o4gG&kEus@0s>B>l1|2N+}ZP;y>WUIa~z>Dn??;= zE1Z5Yn#cPSyrU5fJ3wJtQIHdFW@aNLJiX9gZ=P6I-p`OunbWV1aXB_iDf3e7(}Q}> zXZYJXGGUD<0PoQ;;5B}7fSrCZx$$qn8^}f@L9xFb5b#iQS5Bmzo)DnRC(g7zsS2|d zMPsXDe{A9IUs)@2txJ#W2$wT;ZWPWp~!qxCfh-Og0y8Lc)chloOe9o58P}+C=U4QZ^WzE zXsn38R4Ni^0Vwb~$zMvxMyQio6>;_J{jSv!o~<7XL<`MTOTFP>lLL zL>|Q4Y>QAho{J{GA&+e!*Z>4#|OJ|~SSoAXd`mdRuDfay>EI!$Ul-uwnj zZFMFR6cmsx05Xs7oz6?Bq891*C%Z%`nooAxw7xP!KF#sVmkAUL{S4SU#7lcBtT<7; zI{3KX)meIb6kvFTD)Xe@#E@wR6lc8g81Oi*2cSADznebW!;hIby_9@T5bz`>+pD1T z4)7E780ccJ*+J3R|JXQ~2uLrWSPi0+Gi>#fw&n#23g`xPyfcfVfNH`kC+9G5nZ0h~ z^f~pQJ9dXtmuyh&!n5D~ZqnAiKylP`AwMivRBlax%4EwoUKE8`$!XJ(&oY$CC zj^g3)Z^UC)CnCi=y&xhWn)(bkvY((jy@hBVs|3d`Eu~_FcngJHK0J2fuXyhbp?J9Z zoAXp0VEr0JF)%C0E}ODj`*%s2WwQu`HxLdd!D z8d_>Rdl~uDTX^Q&!{2cE#xcZ3F`OLFP3%m(@V)mE_1^21fTU>rHcQoezdDl zzF35;%_pQ&)Q}8g!R&oTRr&dTw2PMWLawmG_?-Zl)^xPHTS+VW-@E?*#s2pd$Rlgw z??*Y*KJh0ubY8A3R&|<9l&^L>sIa*25PaARGaKZ%1#`$@HR}4xP`>aGY1v59=?-7| ziQaRs7{_?N}+qd-E~35iPi$k?`Q{H^Na zn)mmEoy*FH${v}D2n|HKTn;BItzsoTZQs*OnrHJ#_?eHHdNmXdFDB4Btarhs`k(1X z;y)&J-~Yv?RsY(8JCyJu7}sL!C+5b10>kUW36>LIiVWK0EFv-fw7akB(8li2ehorF zWj_7(zwfv9n=)`VoA>5Td#c&{zPuI1eQh$3rzeCg7SQD?QbrJp_%3KMwVmt0hwX0k zs%Hv#LpDKsL5Jm57ycTM z&9S8&Myy{xZE0T>PSE7`NX(8n`#_9BPhSSv(Q6Y)(_Eu&EW35L?hL778hMXvBB`-c zeKKIfd#@h3o8Pw z1ZIa|FJiSaiIuO|cL@ImS23`>qp4l&KhOQS0KXZwht+RNzFEd|X!l}>*~K;^<4v@E zZ3^yNKm&jBuE;@S|_9IO`^Jdg_)$OTJb(8(=Uu@>XFy%s{kU~>rPtk~* z^H#Fc#$3AqTuGjcXLsh#-tK(1Dz7b$D%^db&{!XlCMP<}>2m?2^V96kI$_3>w>NvL zTtTB}`FcLygi4Im4(kV7Bjs-M`I6ymMHBv91GroZ(pv7lnaa}(p10V*87~?HNjcX2 z{n_MblkZ`yW|i`pN#aW2&DZ;G)s|s6*VXueUin_70?jS=!yWx_ z4$HXj(YzyI{Os$Moz}vYe&rvk)cHHt^>*Wvs_o%P5|&wEs|z2s{V_+@va1h`pdb5{ znHLs2AoLNxvk_8b8OFGjA(X&-f2fUW-zNr`P z#nA-LG`7Jj?-^l5uNFtfcs(LtsuET1g6$m8^d=$LlsQ9&K?|0HZ4WtesMiw8czFv0 z8|NlV&MmW&nyiv>UyUGaJr##Bugwl|MlT8v3w;xQ%_whDr?39pn|(7sZ-=_?EwY!d zwQ?<{CP(FYVH4it&O)Ti+clEd_le$*)WN4okhc``2xugA{zmGyt+}^UzWiQIVa&Vc z#RX+hHWswh`$eZFKC@_`xs-c*^hJ?Q)BDHMG+7W>35XBN;F&-2MU+Afsg3LixD%yGLCc@ z_gSi>sn1KL@750(*8er$u;jcu6CZHzAd4pDnu`Mbz+WParD=C#e)_R6mMSHi>U5j? zVg8WY-eOOjkYD}KSXnF1W0wI3(A_|0@vu>-ij@!dulnnj?o#Uu!1;x9vF{GaN*m60 zBqpJ$9UUGFZCSO)35{`kddXEObIMZ`jh4sSv%74sj9welhqhkOqlCE)*7@ZP(RFOk zEF89SsZX=zyWg1GN}l%?AFB!IlWlNWRi%?=j}|Q`De8+l8|i^gJOSo<2MsaUVK3x}ueCUiKh&uE%gZpa5BK6)Bc zeeNo^H&-CK-zVRdTuV6}<(%bQrGs#d8@r@Bfi$l!OwXB59EQxiI z(m3^%d0b6=6t{Dk+gv{$F>O|NmMU`CbxT#@SyIV_TmT5Mxs_(Hg9Ih$u@ z*|#b^Ee4?iPZ8&T;`L5GOJx?9(m^ib#@-*Ef~x#Dg57$~eA=m>IoClAp=|aGVM6YL zW&%cPmCE{QKdz&hn|2z>3u#WlH(&rZl@vV$5hi7}^LlA?33Gg5WwaoXZH7gi`mkfL z)OzS1LVT_(H7NX=XeTGNbZi<6KRiYlE`g7b6zOEJT^{Dna(bK9pRT|NpSq54-<|Ts+`>mgvh$i+7IUcgx8L_x2X(^209*j$lakC(Q{vd55_%8ib)s)#T?rolcrh z*j?b1NBGF4&VwE8O*1ZE$n%@+pNd_6oh0ffe?fL6Su&hJ!HbGRAzDAe@)h=d45XZ+ zaCd^u5U1^2iG9-Za8G9bPOh%4d;P&?J~ej^T(Qz{FsA31UsO2@VFcI=0uUg6wYzQ~ zn=wMKnqE+IXJo5)%QCxi(+a=C#*~$asv>EzX5G#MzV04iLsxT$`o|xexdO4lCik|I zzdh**V|;%3jQIp_?$HJNPfB!rvN62<18(~}0iHbUnIe19Tsz|t zWH(b;8FpUA&rk;^){%ftQQBfMQ1}bUt%zILdU*9oPiS6y*^RgQV1!_`@*eG%SZ{&@ zQlcG4n>#8pn47Bs=cm)aUW6!Kq*~%o1;B@)=qXU!7QlPR(JIzhb+CyXJFNHO zaow#~RbDm+S!7G^hR3WJxvAAyTHP_V+EUybT_xIL#(nkF0;8frkVZvF5uF7sOyK6SPo4W$2y4TKYc`5QLQpiWU71x+p1wm z{iysp_80v6B!)L!@{Nt2!j_GHiyUs}@>~ZS1JF?gw^j;(%SjE(!<<~3oo=z&NiH0j zM-GESiQ&8(ru2+;NJ*>Vj$|ea2qqMU?tS_|Lz>DdO7{}4cn>%oyKc>KwnKZN5Opnn zTm`Qu;Whz^?;5TpKk&oQQRxmEy&0h%Qv7}zBCf0&m(P5)oVIqox zfHO~Vo4QYR20qN6NG-XM==+#!u>c-BA=?j7RNZZ7%5ubStj=4RPB3VlAonPLv+Eu0dzNs8#yrRR`_5Jn_ioZm_ zZwQYh*6#_oiTB!zZ&8Ui+)*DfMfe(xRJkg;ijM?gUcaXZ&?!EEjKO`gUq5M1rDjB6 z5ft5it^3n~sbN$cjDI)IE18HwpGKzxxG6`O4dx>x^1+_ zx%W+eM*7smmoJsIXyL2+E_N68su^=K7+fH>c#chrG2g_qT(S1l53z%3Z+J6Xu`tL& z75iv^ns+hEc?q+42t%Nx(m}7RX*?0PTS@}ep}fh zmW+WBon$od5TE(Shzk{kNC@dt0f%%D)8Y_pyB|8P#4;EsEuL{rHnt9qGa~pxcrG;Q zHJmT4gdqLvB)U51J^3@uQGiZ9_HLx%f0j`jgd9!~z0MvPmPACGMs3kqW^2rj5#|4A zv7mE2iI#S*LI(+Vw8JRA$!di6G&gCs2e5^p1*@^5rl9-fsS}el;Y)9w2gM% z+bjgst4vq2NnT}C@&dI9K2GWPk4$xv0@DmDBS&}72duYhBO)^Z#cr>|qEDVLPrMaE z=`cB8PCubp_sXkn4|vVK=)?773-10}z&y-^No?gx{6=L&YX06Sz&{)p@d#09Ay-!J zK@QOTh44YNvT3KD*~;c)-|`>y`X-u-#i??G6hwuLOb&SAG;~=HpXRml>`VsP0i*CP zKY!_lMz4ysWJ#0>zekM0SB>?VcFS(x_mM%(MMHMT?{Y*!$ILi3ZpJZ}IzvU8G9@f- zC16Qa7QPYF7pyp@sk987XGzozTp0{13|n7guzR_H#C|1-L+e_uK3aM#*={Y1IWn_t zup)w=0%G~_8Qu2`z${udyQvYwpC0Op4O(Iivw%aFA+qLvwB>HepHuSU4sF_Mf|#h0 z6PeSiyaKtX1G1xSZ)y#K0`_5tIuPREXAtZ>t9EnsldSx?sqQ!x!e%l0`F4P$uCNyQ zy=M*C^`q^M(~A?R6iDK ze;`tFCojh63Agj+rQ!tH#k;VQj~HGr@UUdzIeNS6vzEf?wco(8ATkqAj0?i;u~Eu> zgj=^(G+;yvZ-|Krp(O6eR_84T5#kH+k}FTNHp1m<*~|Ezc?u7e+ZX#>zhAajY~0%_ z`G~;oUeO95Ye*vfh^uB|tX4k23nBUrV@#t}=5nyP@BKIF5S=9)NoouEM9x* z1yFK?j>+=}AA>3hJd)Qp->0KcIW7i+vKrt505^2D3i*DR>*6V_5Uo_VX`)b?v0OIT zCQiv0*hh`H#5w-@PVM(QPu}E!*pWv?swgJ)GcMpi7QHKDb&K`ar7rdQDcxWHR;8CV}f-nNE(RLa*F-bBPrP33=5XQPpEr`d0I4>)mn2@x! zHej6X1kgRs5iSf?G|px zH;K6w^p7|AnRcbfY=UvU8mXid|Dbp!;#sLxA_&#*!?>R%KJXMi3YRte`R!9uvkL$G zNVPjhh-MCb|NWYH9U7BA&h6)Fuu%7pxF>x73@`sv%y^Dt3DXsW&nv8(O}+E36@J&U zEs>?Hrh7l}#bFF^+^-Oc6o&JObZWZpOodXz0dhU{@&~F^qa<7f!?Ytova)|vNv?$N`N8M8TMYwj;EeK+$^r{&(W+K zM>q3nq|7qLDuD=z={<+_>d|&Uk90xwX)v<}bg%|QFP8mXx+^d-G|fLmbPpCtD;9%q zmC*eqlAKKK^F1g{)r1^vk2lXKv^-KO_Q|Fw?7X1brp&A`7`Bi~u~&j$o6T z0>v{jKnuwP7zyMB?M@pte4716MNND}w2U|btPBKE5+Tl>Y}#KgL6YN(?uYI_wN_?I zCj60;)L+k~`3i>hWERkejLLujs~HxeT)@je6*p7Fr2R#4dAL&2Bh~=kecyE%fV@t3 zx`KT;qpVatFEKOUOS=eFc;lj5e8Pm)t}# z&;R*N^MUiu+lk%4ZwB@nN*Bj*I|+Tiy`?|yFM&1Z<;2T`W_&(spe-| zCl%2aCCa23!&;Q0bVzV(i{eH`rcdR!a;E>VmF>6q&e~f~l4RH?hdEz7;%1G+ zH!I&B_mgakqY?L^52`Sg9ri_F*lns-Kl4Z?l4lJUVE~EaJ>YJY9Nb-3dn|7xR&q)Zor-!Ua5;=ky1-{ z+F@z16p{@rxQDP)&jrW@8i2!^MQb`xfGHpn zK2Y2m$%U*w++egPLiB*ZA!4y?K1)cA4NU%Az#$<_sl+0d)pUSmo%n&qbIMoMwUL1s zp2Alz5&W_3BJmKyvDgKQqe3_nQ2h5Rx&nu^>99v%0lEI{7TXwWu*;kz@WbUM@YKn( z%rmd$(D8hRBo$U>UfnXw;aV!G8_E(3p(!%ye~8rbqrYZOR;(rqT%`#MLjK1RU{B?n z1IE9Nn~G^WWX?uW_lX1t6M(F41$9%fO#xi+&PBtp$}m>;%;vti?(|GxDMgxf?uT3a zBLWC#nykdttyiP%N1noA15e#Udo!L*C^D9eq65|5!m{3stS0`N567RyER0azwZi6= zABqgtMF)_EIt=Qqi8r4%FStc^G_==($xt#;6tdXUi7kaeu+Y}k2F=k`N&!qGP>x8{ zw0Ovl8AC?Qh!RAOZb0ZWsqw^ElD$Ccpy3cOo^=42R6o9g%2m?19NFZjk+mBtmq=ZY z4z_wgtiD3TSh+6iJnC_wcEg4-;Tr5Fr%g(v+n_nb7V05@EDETazJWy%v=Xc2wc8p7Y#^(ZjJ2n5zI3FJ8 z<>s#{pvvg1{HHEBQEa_Y)FyK;9G)(>>`l8tYh{D{#G0w<=R2GoxrCi`Ekqdy-T>e~l z3s1KOP0gLkH4^*5_MR-&`LcwV;HAiBSKEUj`;p2Y@ps*#$z0cE*uSgCp1pI6p=w*} z7K?#&iSFdrK)XT9Mz48*ZFpp`GC(Xc)vMiN+krZ!4M;6cZxB6ff2Gc@i8AzMx!hIg zuweH-`@e$@9e!TOPWIPdrK7TwZ78=Fv{e(2$jzu_mD)GMb|0SEAc-7KFWZjGtCjPV3k70)ioOyu5IXAabeFf9R)_}hJ3W@LXiQCz(gp`g>t)dT7uxaHp zEW$TO+}%41{ldPO9`5h-lv?X|a(Cg$$|_P1STQ*yOgnE4*-Ii~J$nj`72pq&&4$Wq zk{(Cr!GLaPy0=(d45Gy*h}}MJOvnSSKQJqP#rH8CP*X%gKGDe2ac&bvST|EA_knF8 z-C81KP%!{vI1Yn8l>Cy6;AqFI5bME{=#bNFXEW)0z_rl@j}N~{+CcMGC_L$@oCD*w zbINiPP{e)rt7e>S0g`hyj&cIfEmw6v*r=iT|12r9_R(q0pp zK2Tbj=IAyxWnT)Gko#zVElES<$6Rl=IV9m>+sLIz*O;mSZgOz2>|T6$u!jV|=qWIy z4-nM^&^=3}XULhe=j#z-vwgYohGALt@7{Q_r@h0Iz)ajT+AThd%Rycey!WkpG?4V^ znJ}O+GPtOD2^Cbe@$9`s7CZoS{T7c)iQ!F7#Fh-#LQb8@w~tI_r3FW#tPTZw-)|}m z%>ju4+ZUHCS&UnJFt2`;2gww2LsYrc+X<<&ac_Nvf7RUSAGD~$NlFvqM;fJPE_e~D>t)M6loV|bTI0iA;Q&uYw3C?g$IjH}z{b?pcz)%{k-X+P#+ za#9pOQ*G|rw2EH7=Eb@R!am(2(t6+Y2p0=mt|VhSc}^@+SdgbMmT!u0AkBiHGEl!W zO_4=_b4xlF0jskpP9wk=JBO%{p-L1Bsa+OCB4!PmgQ>?#vG7x`y-ynHSPfNo19842 z_SXCWRz5LE+m01!P=o@)YGXU!lPSHC^7DH*lrqr+8d5HZKEsEF3V@sMz-yKHm_SDl zb|;Hlv-aHlPxflX~nNEw^VO1z{rdER#GfhX29?)DdIu?PB1kC zFzcn|*t9o7XEmHIgrPq$_?P|vM~NOz8=VXI_O{aC8^|Y6E{zO;)Q0`h2ji}kKvt8! zAM2@P_Om0FET>W^Utzzqj>~q7yXd!gG%G82+BuO|lo=$A`xD|K?F?+^a~9ocY=VCd zN^r4(mCuuTz~Ai^cP2(hFYVPjhz$Draozu4$e>N9WRfp? z`vZZ=cNQ{m^UA~B`69rGUn1kMONn6})-)w+q1F2hq>ht;Guniu8Ac*lhLrN}XjWg| zzKvjiFDhYO|p*?1JUZ~nz$`4~TzQ^sdGFp?>+zRp+K}(m}XkI6!A~|hl zu9#TGvLRZ0**t~4UI?$YkFI0hkI*o$KClZ;|6#Saxwr^&Kz5Dixe)ZT5UR@W2>$7P zr65#=+1{UE2xqTMSh4$)x@!c(}i7#)50$#4ilb@l91%q@B;18H#f;DlLk*+%oYE`X5QvI>hF zLlSwoe00dDICcHcd^bc_9ag&{-;_&Koe!wmouhRIL8bU5V#=x<$fF_xBBcW|y#+SM$RkslHNW9}Z_YBROrYC?}bxo>`}orCB9WU!etg|6ML1 zas{pY8g`ne=UV?EP`9C=?%M6oIhXSPN8)(6OtW(X76vk|Q0c(IXfmC70>Ev%b|FI3 z)qvn;EA!UB6YBssJ!(&C(!e#^;~GxBS6C@w-MXQw1e<1H=siWI$p44l2w)x6l`NI6 zTAf`}X?h9Lfoam&YLx@)+Sd?I!hqi&9|I}vz5c7h#4-k4w@$%y%>GH;63vl)BgrvI zzm2}zy#SKt@bPRl#IFqHVzbSf3uGnvK=?3A6HGK*Huy8%9tE28=b#Za-PZ0^YKH>1 z1E*UY7BTdyWh_e=DV&QyKhfNn>-IyEW~A({ShsOcC-vYKA%J6Tg{?^_yyfWoa{pBD zX98-=p}SaYZF>`0$DJ{GfUvH=U6{-iNIwNY9&0!1%~DP9b5jL0ar)IB5ccJ6v6(3T zQ>C*Fl@U#53xBdrbipDuY@P>&B@LBml9Ji((ZI9On>mj+b1heBL$EUVb8``i1cZsS z`0eSsV=DKPAvGTjSqej_>+$ikCkA$gYPnsOQ6QgO6yVOw#aXK5N|{Q9kK@6Le>bc6 z^~8B5*4qRij+t}(leZ>70qp_8zzPS!pWE;`8qzV`xtlX+r{F6y{#;8rx-B0XDxZU} zlQuC;2nZ71P%<~ky%$5W+t?h)f5BoIFhFFauUT7(CqW9{OZCUca5isw~UDw_2(@`#jFg;B5&X z*T4n>Ws@<6Y~$3lt_sJs!DJdS>QRIoZ=9=GQN}z}MLG7#$n;~Rn-}Oy?D)~c$q^e2#XyToS1Gm1 zgZKS&1M8jM$r5+$$Cbb!o8W6l2L%O@I5!@UFFgn0-7{}ok_<2zKyv7|^Z`+@Y`jFq zGx!v6(h}1@*~# z*cOya`v2TUB6mFjo5W=jOWk8b#yfZZGN}pspoKuJ`j2~D2+aik7w4Y=40*q(ck9oy z=>YiL%48}Rk^LtZOwzX`Ts>Zjag3v1?*tTp;t%$+)IME|*HBG>{Wl74;&D$#%UjyC zSnE`S@iAgFk|B0@gYE04LWT7!5`dPezMlThKJq60W$=QF*K3BdncI6{{0pD-qjKxX z8tz8LlXXzJB>a%g`_LD#=D$4HkjHF+Km8;Pyr)5_oN zS9XAdN}9zt%&j?~$f0bdweLgy;arKTU4O%q08-VNZBXOU29=i;9{@&{3zB>HaO@jM zZYay#5aWO7mHQW}0jUkTSOb;a?o==?9hV_XzxfPko(UHDTjED~KNvL;R;txuFp%+I zND4Qj`PG}zF!)OKPiTAxE*p6y9e~@L{HZO}X-}D+5TC}E zCYwOr*cA$lo6T&FWk~r*Pp$17ZCz?#EYk}h<(BB203rG?2nnW>M8q>*C+Ks|CIrPLmrcd(D$jiGdj&|@Nag)S zIL9pTdURc6rL$_ghhNP%7y37WR@9*Sv@Ls7ew_IWN&P?f51tdA;8#TjE(40X^+|uo z8ceQ&t6ck%km%@@v7nFN863O zDk4;60wGpy+*^Y)_lhIvVDRLpB;`Q*Q2j{l|0kHYv6(+Yh1MuC4^yi154FxWx0s~Eu=1IpkV>KJ>lh0N& z4cp8Dx3-rm?1vp2hu!u}i;4Gz5qVERlI-yaiC3i>sK6S1u+^Wb%+9{3H2CDNsvyAI zD}Gx$HMcf{cp&)fP70=3tF(*E!Yn+cE`|n+kh)oU$o27m0)ddh=j- zig2O$7;2Rb5G^HsrFjChfD9mw&~pq<#_1=_$%xecnXm*GfC+xP#Z{*gsz2fQL@#<* z2}L@2LeEEK#@G`r+E{y4t``e*YHTcz@S24O;A7ak9c=Wj$YNcfE`h78Oop3R(5pnI5P{9{2ajl8ELRdWCa~U`PiTf^B!`INKBXD2H+FpMn_w z{Vf2W66#(Wxs*|KB`1FaT2pnKCG7xiy%A6bwf+!dx}~Z8<7^+abXO;vHo-A$jBGD9 zS4VqsOI;?#pMhQ~k|Dna-kEsX&SGE{P?Wc> z;BgNlJT4O`ifMNYv8oNY#$H}G3EkQ4j^quHqd7)7q>zs zv!If!hEDQqF{to>NFA5i!R|&85XDQZT`&To9Dfq+f$FXCtV}UEN>}}RPa_Pl$}#A> zm-#GUe`#~TlO{r9x4=ru`eua3|570TeKlDC;i8HVAZjvNLKKy4Iv!z?j%@(o3a{RI zQm;A=%2IPHw}!dQl=JW-lh@)UJL^~tevl6aFV@_VBhi75~NOwq}MC&Wc`t) z`EQ6iiO>);T( zuo9G?Vi^O*;ot%m-)UGP&7r`ko2DFO@9U=9YeAX6miysa^h6Q0%{r*FEuVX0$tA1u z3ma}0U`R5viMi7Rl`&zM4w^2}*(27COHpf5q1R2g( zZioq&-d~Hp4;HvRzm3NjNM(M`Z<8bK!k~nwCDK=$Q7JhWg<3{kCM?m+-AMNsTdR*_u>}oIaX3j_)g^%kNx+g|QgQ#y%GVmB$(O z%fr;w10dBku>EUd!av}=pBt!$E(b#F_$HalN=*_kWnsOq1U`+NMdy1&h9VQJTX+S4 z5N;kQ@qE=Qq2lnXfp8I=bRT4diC9_3rVwESw{sDMSeCs@-RYrFN#h`!<+vbJ%Uf$p z52;#q<39Jlm}1Mb;6p(00#l9#?{lRd^a!;?29FmoCgc?p1Tn z?9NsnQC$NF@xaRla+t#cXKr9<`WkaqfxlO|xwl;H?cJTJ+?!Q>xC;u35^>sQ%A+~D z2Yy~UzBmH%LQ^IB9W2ISX7mPtc0{;>m8eNWWU2Ifxvqc8NMv?k0h;M8W<7~P)3!&` z#Bhu||42N1dQch@(4A+?A;Tz=8{rV1b?UjlW3eSy_Y@u%Cm`lrDw!|=^^Y_%Taaqn z{U$EW^ekHmpHGBuKTIHWFJ)*uZ-(1duw- zP=s^Yn+-BUH*Ag*naJm7&O&Pam=nldE#|c_VE0}p>>z^e0@q|Z^CQ->co66ki-1W7 zAiw;>=|2BF!7hWfYH_ENJ@^k4j%TD33?aiCC#rzRrsPqXyysHDx);E)$_KF=VQLzU zXaH>WHB?4#*+`X>Fe+OBkj$A4R3gvZV-NY5geTsErz1;F!}QJ?@UlKoJAKPu?4AOw zgGEO_(pEmO`#GqD6?O*k&O{W5F*&oeSv9FgJ@za7o0jQQiAR?41usLsm*{kb*>%Q$6~*JbRjMYCv>3Wlk@(;xp((|(~Nb`5)-$J zGmPUP6dtNu0tT)5X=UHRDeBta-illSB4tx)Qal&|>$r%TCm|+rk_aDAg`*8o!|(yx zGIMaY&B`+S^Yw+sc%Poqv z#cXBgbb%rxLr}#*YeW36NVb1+FL-jRa&oYH9sA+IL&>vfGy$14ksz6-3{tU!*|~Am zSaOg&@ddNRH9r{ww&A=O5GAP56gO`yB!pbP)?XHibq~tFuoU zfO94gbF$|AllD8v5HMdhkQ(X^XEPlTWcO}qN}B0t1-!G54QZkc-EshU+5{?Pv_WNM z>SZhz)531~_U~qoIzjbbf(c7c5|%91Ov=r`$IA1)*#XX^Z<|zs0^#MU7IaTC4=8F4 zBo{shKUfoo>~V1s>PH+ zir-oOE!UH3#?=xaCTD;b`6TE6bDehdMmWPPRWiYDI;ETT$DnjG#(x>VD3H1#I2dtI((BZP`?PEAiJ@vY zhYCdT18ib}%P*teN|N|=O`)A?>0Thd+vJTJjjMm50@Z~R6losQ)kV@Lf%D6eLLzs{^;EriwSrE3;diqgkKvHcm9eYJUTCsda7%Hq&r) zZq+gyON2l4=foBguh;Q_>T)1Z@8UZ?1vPn|RbSUa6x1vciE(iMrd%Mo+JU0?GU{pH zg>`p(1F=Go10G@d5X?azmDL?9aDfmV)Dnb&F#Q2I>BH^N#Zf<}4?S5ahlXM78-}17 z19GZ_oD!NT0I2%`g}!co^!P&FnZFi@?{oSI$yZYsH&a;vl9!#wZ3KJSLS74WwNI1` zx(>D!Dg|Bt3VFxU=mH=qh%`L?7AEP`T=-P!pTaSk$3C?Cq)W&>o`AT3^A;iq){(tPfXH-;4 z)HbS!B8b3<36KUyiIODKMY58b93*F? z$??_!=6kQScdhTw{rCQHX&>vHs@l7DeD*F!l#~v)vWx5(ySlha_gkP)$`MmW<&oCU zJe7DYN17t;SvIqmIxMSYHiM{?O(HUp)m*>WO#LW!JY+jCp$-(`m5}b*u)RXt=UhuZT0e_f zJ59)2BBT6adJ(OilrpNE!TE*wClLdgTM*U=bZv@TxnlQ2Ekd|@GE!$S8HM0fSOh?6 zmYLUgq${MHMm(Jdybt#;(N+UrNL51Ce(2uC_0 zswFHy{0)W~%eWo(` ztHDgtCz@Prt9D4PhD23$#yBgdqyWmvMAlD$Xd^;4?}GssN)EJJ@vhT`lsR(2?o=ziAhZJM7>!`9+gH>@%s}q_?K733B*PAXaH73na;3fV z%_@Y%jB60JO>Fw(5SMu9U|%aeT^a&QH~TYoT1+dILETE_MG8Pn^;~s;pf4_S|9Mem zxuI>KuA|;i4FHM1=IRLx@2_>%kA^rs@RQEmT7xnGoHct_gyy{iNwoDmetcq4-%*yC z2hsYykw)3bHLG`>qb$f2is&}>Ocn}<4q2x1mr~!Jtp#5t?GV-wM1f+`F{|_OYw?fb zxSsrR^KX?IxwENB1*$Ipdfxz2$_${=(JU*e5*o^jfE^s(X207gJQbja5^|u=kav6e z9fFv95-ZI1BbP;g4hO8B)UbQ(ZJ<|LG7XcT|Dz4?nF|2;Z!<<)`u6Ifa`77=j7C60 z^xu1#R>`A*e@J9{@PIg(Kzz)!P}lDO8z+Q5!M^=o1c!^lOSiUor+yZZ!}LTiyD=7ZRh9Y9^bp6T!9MmIn{?{_VTt|V ziCHFZic`z|hQAa^xHG7gzg|@%`9m72)rE<=m1X`R7scUNmk8+6LCe|z71Ri}dS0kY zHcXaRS2l~YEU{G&1oY)9y+uqUh{pmD>#DHdBL6wEc4YUj6m2K|1)~oYA|MRmp4R(H`Lk9!%L@BJ z3$To=)1&~ZuaJ0c@S)FJU{$uhFjGr?pOQxMz>kRBDYg71$psW^0SLCWpAY99dJJky z>YL5_a{+LuETieFnGir9UQ#E4yb-;K-{Vt&nZf zJnV@ctBqFLyzE2(a%v_Mo2IY`s#-6nI|p6}xmqbC`uI25ZuC0{&$Th?M_ys_t;jm| zT{bVkzS(Z{=vbGl@~+4Kq501uRNe@%t{s*=Xfu-QEPr0J1VC|{&U6%1Cj`@AXQ37W zE>@)b!xx2Dba-9VFiw`3i7+{N)y4;LZxY<*GHk*Lp3FIa=e1WBJ|7pR(0N1uub(RQ zL3-5g8i4!e4i2&@5nhB%HU>`J|H-!UVLK`r|CG$;IyO0^ppcl_1g$%MN`CA$lJjIM zt1vo>d?Lv&5t)0*)1CH_X{i7w$&LIw<7Ma{Pe6G`5Q_-9cP%@XTcNH1Dp6D~APN5G zLI_9MsPFIhbb|ni)ywU2_rvkjLs+ zgt8xlMNb-oJNZRH4bQypr`zX^y54~(myy{(NzOpR7M=PT6$W|`Tw_T8X!hqn4boUS z5*WyGH=TC@afKOH6>!2)-#W)5)t{I_m{h8Fk~yGGP}>44z@BmuXVsWkqicZarR8IwHy^R{rDv}y_*3FZ|E zXQhXt6B~f618;=vzh}}z0tV#K%gXYb0RGUb(4^g|Gnrv2(8(HPG{$(JLi)pTwu7NU zShwefzIv$I0~27%r-cUfug!K^0=xof&EW}QhU>?RLlG;c3%wus0c6KHA?>nxUxCpe zA3_^I^sZjsV0KAC^{H3_Ozu^(&LStXOo^^)&)wfSb13>i_FVq${cS?_esKT}P9g}; z^#<+(Q7OwBRpp)Y(JwRk)o1SVpHhM;M={=h;ik(JUw%Iugvz+$_;sy552-ptwEHu_ z3CJS)_>-I+j(u{bm`~;%3j{kC)i^5?(g!R9$i%N_B1^k|NgyRZX-*bc`em6&E}ffy$F1h^^kvXBz7w;LP4|j{9Sm`?d=&KybT*+NMkEIc1lrdpnBEB^iGA zmHveK@N=V!FGqRyZKejwUm&^rZXYL5ohp}KEu?0G6xCNLkZ(d@#no@AOy1^GU)QQ| z$N=3G8n2QdQc*hv5L{Bk0bu(!0P+w6sE~y*69YMbVvyL=0yg~61_ZuQpxrQM`6+-?^2F)65XDRtk&2PR>_L z5b6(*zX*Gft_|T20br56{}VbA^lAC{QSxV=hnfVrG|_PVszcBh?qk2JVBu3b1|w-G zvspeBYn}}f`JaaE0HQy4?lqapjLg^_P+REUI_!PuS?W=LSwItlgV<#rZUd)k5psqC z3_zl;1Qf>J6?IhgJMv>e=}5Wxe@QfkDkWta9?q;Fk9^?1W8O8|aOc<@t^$zBG_Iij4-GJLU_Y>wkSBT{ltX~rrRaJx z;T8z1TzV_#F9%?GYJsARz$(S>Z4d$(YR^%+4OC|-5M7M0@2or^R;UN1P`Nckc7Q;D ziX?5Z!{En7P;kTUKxg^r8uJTxkZCm(bz9>S&C+Hh4Cx918gWQLD4j$VaF}C8`l~xG zcORT~cc5;_EM>6M!s>`fF$~ANZd`?J9ga(ti|Pl!@x?Sj$Yspvf27x;y#!?IsC7`swoMjNiy|baB4_PAo1tM_;oYh)S58`vp^b)!-*dMC-0slTXAG~IC+q;_FkcnF ze*j4i9{@yzFTqJ}vrJGH0b{oxLJ)l;5$WFqT|zqdiG$_#EQTw~GgiO-`{L42E{KnO zQn-KpFKtAoOG>y2aPgobLJ)9&QL21VRyF&)!rKh-{aGd3b}hh3S}dd~bB{pc)IsI> zJcJ?#)v975Z13HQKoHYr;J;X1SuQ97`WZnEg6XbD{j8R6pZ%#G;@HUMRy*b(QqQ`W zuk7cw)MiaUWI2~rxk!nmsL1r&zvnffa*iQ&Q?2eGUiIkeqSzgIg=>zydRpMjEr58~ z2cR{s!Ifxu&&RSZyyujJ3WId7F~)C#<^>qk*CI|^>Vgi!5QhT*eqDfMUqmzjGJxAO zxL*?}lF)+8PSqoZq}7Ws00-4v11W+}FG8;PNdvA7T5A*ATLBqkS8WYYZsu=UwccZe z5LVAaSP+I>d7$}o7p7GmC=lUwZ*}b_aekCWkeE8LW6GBWlikurLwtoY&!jZ{3F^4t zr-hqIU-SJ|WL%*LLMekqmKh0IX5uw_tAUujG!HpvRdL655UqL98I!u3z@TT)9e?$^ zih3H5F=ewoFJrGLV6{uF;K*&Nr}~_=rnSP%05wmfF zi9=1=TfExW$naGpvoqw=PY&B!{e85SpZ^Fw>37ItYfteD>3>x8vMT+G=@;EF3qXT! z=(`JU%WJ<-%`!(#Op2V*G^cE*qkp17cOhsn_YeCkzxXZ&TCbhP-%LL{)JN+rI9cz) z0qm>$DOV5tvljX;(=zS2FFqa55a=KGcU@hDm;I_KweZn*rT=33IY@Q%UAvR@4({l$ z@Z!I_mJw>g**_6%gHHqCXB7df07|k&V(TB^i(;RJBOP2tuyzZ7^cTxxqrq<1{ol@0 z8F6ScQ;(vqb2Rop15%FDU=roWA6DbY3yk{(CHjuq!@4I#C;XyW-S&yP^zyWJ(Z=8E zq(33nVi^AW+^?h_f8a^bZgSmL6@M_j{ZV#hiL!&|7Y}QWMP}3T@^4C~4g<8Vs}RId zZRTH#Ux*}%FBgyZi~GA{2k|f7Z4cGI8MRS*`27X*Zqchm+4*QXHhp_%@{oRWV#UhaoD%&9&6pp4-|fqcV0?9IQ!e5 z7Jq&)+&q{TShbp37#b>ccWN6JwXa^DwqNdxkKc(OayA*+jG1PTcz)(A0pUdo`s2Uw zCrW5sR*F+<`;GGCm;ZU>FQKt0P7uQXdAlU)e?}ZX_jy6=(7jTWPg`>P|Mj+$XEB5o zW{1DI@KR!8i2wW0f3MxYF_Sr#ai|g;`qMnXTWB=S%z^ z%m2IrFp?|kc&1bE%h&M+t|=vD`KKQ=3jTh2q7vfA^Xv-Q^h-Xy`il;XU)AeXO@xsT zPOtqeK|F~=_raZ&_nnSclTKotegy9o0;Dm?`yPuTt-kWwE5Cb)LIrjBwVn`Ui zCD1EAyeOhQgDSj#dgp3aFG1AblQtvABXFE*Qq~5Yk8~VvfBZe((`SIjF4UN_oTvGM z@S=r=Pj0_AA&ebf;i0>^!c&JM0^~+0mqwY_!03Ndv9^zJ*vNW#HOb8Z|sEe-m3WuVz{Yn#3C@3Xwj;gpDlV^p<|5IcAylDfTf5yy*;q<*#(L4r^XIy)(}x zu*bBR+iB6HJN?wq?})qWs20RdKT0Y+8MmnO(Fczadmxd~*Il)v}MfZpkmwQ0W{4@`4y7JP*F)PX>cPeh@V&mM0w7umFrsSzk< zUm|^)r!?)LFGlWhEy7N((XjJaN||<6zqloc_p}INn`EpN>**)YYd22TxFHr^qJ;NF z?xQa*6gZ#{GMYGS;XO*dc1(+ogr~v}Pfw>;45Ec)VpXa7X)>Mh={{C9 zMX^O#JpPW8hmO9&=P4V!M>Hogie?HupnUqRwNH;^RLgKb3I2{|A8$>p`tTm!G7=#E z?y_eLz+;Qv{NrK$`POHJ6;H2FlMW(bkI8a&#OaeLVW@w66kFXH+3;vT69X}g=rdbE zytxb?%c%DP*2F}36rKUg_fNOl_^gdbUBZfE5`w#LKJVZ?rO|~3UT!no#M{!UV;U-3 zkNi&d>hv=5`Hxmu;wXoIp0Z0%bnvdcmiXxnH}pFa!*yf>OL)?aKj=gUx9WWM!F!r^ zq}VEQVggT3M#Avh34gD}6pG?eKkP_KO$4Ae@s=NVB8cNQ*e7c7b|Cpk2Uj_%;yocO zIay=LW=xSM{uhtdhzu}t#`_{++wr!(F1>nsYo6DVK%4qFyNmMl8_)Qpo=CTjH!8w- z0&kz?vRS=PN~8>7`@1&qtDGBAb0xA^m3h?D6aIuV7#|pKhlX$H+ml9@2K&Ku*qJX z!du4L5q~G}2#K8D(pvwMVg8<%dTsHZ?m#-RNX*?N=k!TDOMp9(`@=j>Tu%>_XoqCY z993XFp0%)cBBQEY<5(!rgGXKZuf@l-s1W1n!Iwm%M_X!h0fR41PH5=1 zd)?OX^fG*kkC?P1N9QH}d8#_`FosQ==12@*cItJCIz2KUbt1k|!`x)((@(kL zA?em>*{XYbGG_?_PZZ6&_BV@oA|>*OiVWiN82sg*7>oTy*0+i9`0XY;q@reu~i zcAEUPc1HtmNk8Mnf2sd5HF(+P@m-qmBjMIaMT=*G8KO>zSd`uyjJKsFArLJFOqMNw zp`!SXRa6e+q9fi!8IQzp_y-|fyhr{gGRj|Q(g~0F`8OdMb-l2(1fRjVPP{hb2mw5W zku(kQ_xhX0F}yYKXCD)Se?g}<7>|TKM~=dJpRoC9U;ZLRw8pW?hf5%C#vAB%O*`x?^nMV8iRTBglpj)06f$4Xta;?@VQ@2 z5PwTr*ua#Hxi8^A`PDDEgoH3x@?U)re|cPm&uf!o z8rkrl{OXtIwRe{PFQ;I9ox+8eC+7SwX@&Q03YQ5{ydQr>x#zVl1Oy?*jD^2mINsL? zLMn;ye*7gZqqQVPzqKH*|MIJNUy~Spy@~hZuW8A1>7qz%Ov`DR^8Ya!Ffr`^ZVLa) z#Q*L0|1$A^XXk&J_`kFBzmoVr1@OO;_&){k|Fa~v%syYeS+%Ub>@h7I7hgSg*;;dG z^nyV2_8G6Odb-Ef?gIm{lOVBOGxt!S;#TFegLPE6LfaZ8rffR?uNN9C=bo+Z))4F(QVZdWvVKP7RPLqdheFr)5rYo!m)-N=HB|rY%>QV_ zNF|^K%cm#m=?-jgGv9C2xI~P`%1(lYjX3}$l|83)FrkzQflfNIyP3qd8|O4?cGMlK zaq)Q-t(I0iw!`Z#Kbj2=>>{4Wp&LY(Y}KhBo3MlaDt|a)F~h}L3-?1@;sTOoW&R<2uj z)?;khXTHoXzfN=_Xh5z$7mz(JJhH9m2>Rw%Yc+J|Iu)BiFnZ%_11&hvdZrq5k4nxI z-y0+Ec5Smym~$${4DsWlft>Yua49801OPH)Tc_Nt`pv_-0?Ntczi#M^?DSm2?2i(O zm+f?=i*-A=q>E|rmq=8X4j&$DYCz?|jw8Fnr4=6;I6izJ|1wM&tvp(oxNi|*UJn|L zysg+@v=+CW?MlzkNRJVyF{#-dTIQd}G%8rBKpiP5v7)Vk-Z%koRnPB~iKfP_)kyK4 z!W8{4ws(8iKtC1J;4zh`3N^<-oxyj+@_Ox-Hn7mS|9Syf|K(_ai^XPp{RlKr10d(+ z(X@=9{E-d(oX_YXFdpO0U&~80F4#B9=@-C3-_Zjl(-~#>wGR4?ak|5@*@eTwX)%Ef zvc^Qv;lA53*FHY&+&+Lfd95!w*SMQn&L2A5A%az{yx`2%|M60TSBBHIQ7#kQw&ZQh zb!2uM{&QaCuoA2{Pk+Vt1~08}J*EOQe^8&5ahSaaO0+z>xu|1Tg)=JF=)e3 z+ch8D(DWGU$=r$3P7B)E{TAX?xHr>l-^*?h`k>B-m9xkErE;aqzPwq_YT;e^aX^|L77e%MXqjt4G`-R*F0Y*pI)Z)j-q3mG^x;j@s_t0@|OG2{|ff9&nm@gn4-u3!6wwl88{JKJ9;a z&D6Q$E;AF=OBIdDwpjpFtKM0(_8z#UT|ihhr=Cac8ciJO(OVCD%)rGx33kWAaXrhR zN=usiMuelt+FSCRguMx-5ZwkKg(}jdpvH{~kKZjw_u$s^ysM~6e(wywJdt>~otPy& zI7W5{d*AFS!P_Npy!rbiS1W3hl@a@~k()BZ&W%>;1PZ&3e+8$J6Er@u%>wtE*)T)L zH46XzyhxE30G%|Iv)w z7}=;Ke%?6rDv*OlXoXoya?+}gv^erMq9=^NVD#I&5!;edeS&4zcJ~W)f!iV}Qt3lN zKZ2}1k{cb?a;!bG`d&5H!TkXsx~;Wj|? zzvXTt4*Qsm?+PWQq*_~`y=Vp1!zm#A?HwoY8MtyO`CoSHN``e4k~1)~)+;FD`V{gz z&cRlmv-&8fUKC-yvcJC@<-HxH?r$u6-LEp>8bLF2D4RX;a}EYaZQtf}s{VpJAL&K< zxSGS=8tb|J?}agkUZ6`t-W7|`v8{?PO$9ORw7Z`vzihe?3o1pHh1J<@$p9K0WK{sn zyLj}QkuTe={Fz15ocpcV@ z@6~@1sH*}3Vh+@Dnfq(VNxQ;F?VYFJ+tr>WQUgsP9Xjh0jW}HDE(OP0>ZMyw)FyKn z62ctcn$iyF&^xw*O2JOT{gnFepei|&!|rvUMWK!X(4?7N$}xw=0vN4ze}T#;pF|dS zMuBK@;8E*;Nj1j=u$tfL?LWPLbs17fS%iG4cV}ik3sl?U%JxTSMz%g1mzBuJ@$2*n z5B~^etXF}raRL!7AZ0)zVwSam`%s(gm4FqaRWGm@oM05ls0OB!Q`$rUzpx9VXZ>Bc180>RXv+foj}_ZCO8h`+dP6efyV^$aWCaOoOQov+2)*t^ii zhR6GNqQs(9#{O1eChc?12VIxf=jbJ4f!lCFR%|H(W%_ia3X&Mg%OIwTWQM z+ZPNP4>kvrUDa;5^O$C!x8YC;X7i^ zZxq8*jXEl(T-oalCv}~)!eVY~JwOLBt;8>Tq!Zt5N>HLEV9%S*{QC-Y&kCT*DA}sLf94wb=V3GkMmFXrH8IswR@Tq@8*p9axSq|O6spT1Dh@@&vVh-Z=}=Jb8xvxcQY-wb=}y@KF`UM+j5_qWrcZOMz2hJU(7>y zaeA?(A?y5Dx*`H2IajKCIQh_5)@XPnE0yd`Ni|(DC)Mv9SoiQY>5_1#7pUr}&*x!A zw38mMc3SV7jTpzL!Q#|g@RO9^K&VHXBlWDEZv7N!jrlD_ru}zNAkjO@@!8?e)WRCQ zel4kvC+oo4D&Dul30_9g?axTZDQ5F>G1@T3vuIRwM8tk>sM*ET==^r3OJB0b8Pu=w zUUu!vb69|ce~uz3tV!cZOL0u(Uty*-7Yc4keAs>oY(aN7IIxUgOhE${YAP2rX1{O* zWdk({m*#I{yn}$xI)Tb3;{@e+ts|>mSkvC4oMtUkqsq+wJ#nbZqPf9ldkquz?jXhi#F|Nu0p`jUyG%%iiqARy(cV@cwb*l6Wx3#t=M{ zHL;y=KUE|dc4c*VvwBlAl5{?lY^*XsPV|v+n_d_3!zK#L#?Sn@NYhtW*wBut;95#r z+HKvt`P?xPO|&z~J*xf=a-NZHRw^viE5TT zm(AgS7Z^Zy{6-}8-5Q?+&d^3Kf5<>X^uFAv<}gp==ZHh@JpyLy`8e7Bv>#OkLq znU&UJtFJ?N=Nk{waI6fwabeF($vsBTwT`_!Ajl@i0IWEfoeJ1ls$E0wsYZQ2<)5;Vm+twmjY8O|#0)An)m zHC>|HdnVB-x;0bO`RJ}|MlLC&zd20BnDBE>3co{+1!~N_8RJ%>jY?`q_6Hs|G9?an znJl_(fj;Gjqr&&Fe#vu_4|MIB+|3$&BVPFsXWvrqQJIchZ_ycvT6b925FLm>diR8n z7%@Ck@@g2`=_%E>wQ)V1?x&xmJN!YH1vIobDpwD-ksqK=^>3x#YB@WVxN20|0WEk> z4SHl>%fL(AWUB?a-coUeBKl0y-bTxEZ5*IxGQY5Ge`mr*k>ukE(%FO-*KmaY6FIKO zh}?#UMu>#z&AIG#&ps^ax~@NU>*goWANHHBedI_>c{wXKaIqkAId!ZydC%#K^=6*O zg`UT}Jh(Ra$a zS9fT}zPdSkvxy%eTd&LQpCV7`g1+$mma{LhzZ^X2ibQBq8rtb|vPR|iYpn-HZo%WM z8s!LBF7vDhtFtALZp#+j4|Ag5dWG#P89sn#blN`l4eytNGhAe$QtM}+wE#A?O+;;$ zG6U@sVhMyaXz5d%Oh}dbT6o$~SN6FiK_d6Z?%(C$6O{6y<8RVa&fTvgp9VhVlva?D zz;g*-QM(3*xhRLo6rK`2M9I4ro>|eu3T##GENN50?1Zd|5`xN%IfU{}c>741%F-hx z!>4m^Z5bS;i8EQHOb5aF9kTL~p7p317+UICu!0AR13O6#6yolD8B&s#^R*09$kywh zmznk~NS?syHc=*SePpeBoSI?<2Q^%AwYl6m`vN0E#7(!BdK*Rh9{Uq1)Yg`T)z+6* zujeRm^{r5n@q!LksjE<(UkI={VGQKyym%y8PO}oiR`NR-(a1OPY;&6 zP@{`8_Lh$53M39@i+P|bkP0qCHL94>H;<1Sk14ITj&s=76dg#N`fQ*<@SusW{=NAp zkHe_>hIX%wLUrzQRr7kwX1<6kg1sE+Z|o2eEio*~Ue+c%+VfGXI`Da!uS!;1H#cn=5NHuqq=_o|Z&mlbcAdI}^em(1RZ6`4+Tm zbrfP$*$cxoBeqDIjIg3;=KDdrWD}swndchoX3^?!{qBE(*#E+iq`;lN%>4MmY|%Lj zZn>p0`8%2_o>6V-NL!hQiGz;uY_y9izUa@XF-rA%jRaw&zt3(dB9TGA7Z9@yrV5{OjPZtDt6NGe> zN3^Y>2fxwP6Y3zpRNQ3De~0Y&A|j+AQ2iqaXY=U0ns4UQHy2jlv=dSz5A1^N-k?JD z9c0h+K~}8KSoVMtkyo4b0qg#aFDCqw%q-=11f5Z(a&Xg5Ew6!r;ga$!%)a)nt(b5| zy-G)2=<`&xQ-0(7b`PG;^21_3xrdH~CMBf8kx4CG?u+e*1%xmoI?BNqHkFKK4iG%x zcQlhn&@0akIz$)C4P?kG*{BfQ1lNwXPg(awbu2x{bft21mrY~nr6MAP2UL}&iU%c! zVP5QB&cMBo^^~ne!Aq>13ggs4d7pEqY$801Gy=Q9zKa&#;}NY3(h{S!CN7c>{SHdo z1_&dK#XB0qUvW0-t0Ta1@O~TgYV5>b3wN& za;vA^)zQptFYunF0LGvAkVDH9HS=z?c~zCRl=&1-bh}qIdBHLi7>v!nyB)j*?yMTp z^-HUkmHn#H==6HD=ZbtnOF}uBVmSy9Vm&_N1bxC54O?92$_UR+2>)J3#3EFUv~F>< zFSs-**zBmyBeg&BemTyld@3QmExxpxQ*0h+8H|1bjdfKy?2DuPT$D0aW;;mFO0#i9 z*C6GfX&H;;Yku@0wP-#>xjv%UAaR}AY^&6PfZbjp!aPRz#$+k7`kb8zOEUMZsQp{( z0(CnX;^Cq!4A_arNP$~7h0OZKP?N$;ohJKL6+JuE3sS(v>$FSAb@_CUANy(hKqmf; zd?pLdgEOKw+4#Y_cg8fOE_%L$5IbEeTi(Y!!h>?3-ASWf*Zj7LkI3K$F%tGtCn*&C zg&dQqG0M4sk>YNU;Vhb^e)6uUJGF>h`6Rn_pTqDgYUB63`NqQ*OjJDn!Z{tUBP0?W zsEytyI=H0bO3_ZwP~3MW*GY5k>w-8?sua^@x7mx0kEl_P*yG2HL}k+x+7@^d(u&T7 ztcpmvrkoRNi1TlBLYa6rt?|x~nS3-|9tqT6dzXgG%6_IYs_gr=pw%JYsH(4LinMuR5pVd>%tI&zfNRm?{ zM{mNcGWI=Rk2g+pHNbVR+>N2ncR+k%q7wG^Bc!~k6{%1XQIlDiiw<)e`w-%3fm@4d z_b^_sB@)Z8?AlX^3_&7gLrv%Mp`+lp`KK?6@=*kecuh7y(bXFrkx~d^3yi%u_@ zbfM24-VCPf`h5>K?xZb-z5P4v2YsIRIL(M@P1o$Rb<^}WamGgF{4&p*ZJxWrQ<|sQ zsqUzn1l^ne4V)AM-))uUbU$%yYQ3Yh*g#*N?zzhb$qe)T7{K?&QPP*sqSvST^9ZSo z^cd!qh1xu@DP*#;NoEC1m!n+g%X9}KpDczM@ltVJx!cFl_Hi!J1C5smB!9}xLE9QNRE1& z-TwsUk6m$}owVYL+aCA+KBQ!hPns=~c_cr3S7F+o;2F0lEmtG_b?_e2NG9C^LwNxF zW@=KF$b~zTR@(&;4qYkrj6u%js=~C|$q4u!B^n*}mb=487lHZUFi2q>Q%c*51Cqun zbci{xm_~|HGjuVCb@pnoTod446a{Rfsgih3c3jUlt{?|}%Uam{Au-3x2iz7#G#<_7 zT2^JYKd9PW9!9hkppoBXhz8$>*+<$(ao=)r_%FahxfW6T_m#D;`byj%?v}CkrB5k< zD+!jfL+pcb2{9#iztB4;rAevu-6CkT3cr0;wSQe%XvbmUE=CjKU(1Y(#1425fT)vW zae7`)-!Pd7qM=nk$AYHsv6%ZdXHmeyHE< zYEA@WZ|2*qbX*cy-)Amdk4;O^px^d~5f|vGBr=roH4GkD z4kj>#Oz&60tvyE$IArKeiCyT1+g)oQJE$O#k@B&>JRQfPM90_&xWTgxrz^@Il3;W? zKH3U_VrCXfg_rH8pHbJTApB-lC?2GC@@m~iJeT8k76Dln>qt3i$t5#3hIloqrF-#> zkdc~&t==e}Otn0K?lZ9tR}`hyEf?l?_KEDUVpI71t|LZGpVOP^JPgD{+i6Hk(egC) zeH=}2&{XiRy8EQ$l2H=x20z4hSV^>fMWK+258nwIyOb=L95A|Mps&0^KmQv!Xc{H! zBTaSvZbPK-N;qxipzMIYSV;IM7SxDphinx6J2beAwIRrIDVJ6(y_GcQ5b@cs8-7D4 z(BXhO^v3a zhThzU03InL(Me&4-+ZL!zWsf#lLVbq?vb{tbx+dSXnTWCEqC|x`PX^Uqsoo^<)GgH zAC*v(*yY34mAGw>^ogA;+Cfo~2?!;+@4#y@>%n68r*Iy$yhd+5;l^~nSEnL=Oi*&&NM^i%u5+)| z(RwbQkRu*6*131R@k5%zsH~Ek3Y5C@`eYusQw7hcwkCopkT}Exfl3Hl1b1oqv1#1n znqyjY3L+(c<2|q3^=$sFS^Oz!T5C)JKuQjglppO5cSmYs+`iln;MH>PXBi#n6X#Y% zy^?FNUFwoQXNKWnjTbAo&qF(HB#qH0a4YRv!zV%If6LTT6Yt|lZp&vVjE|MzxH7?W zyJWs!JCcmsws7De(u8bQ7Rm1^VCcae-weh~rv6j(=E%oREFC@LrgC zfVQWeD#R|pg8EtE9%whG@8lR)GfogvasbKLYuUS z++85)g0~5)BA@s>)Dh! zwys`ldUfabb0Gzy>!w%H-GpI{+)`C3ia#_M`xv4_!sddhKKoAbC64~G35r?455+w+ zk=x>=z3kA%7@dYDJ^Q*g0v(2^1`)cpHQv|~Ip2VUmA{yH!~OlD`*9Hsz%Aa7C3}n9 zXym$_DnFnJ2bAPJV3fICcWE4fd2OwhuQn@>8hINnOc5gZ{qHo@ZO}cNmX|x-UYJ($ zeoH=G;{;lTI%Z&(i(`fPvJV5L`a-yww=T-rt&|6r`az(9TduL%8^kHi{riAJz%i{r zAwf{K3UUqHuax1VpN*bV7r#Q$XwcIZ5+}(?{{d%Hsaa4rslBNV5*Y@U3;GAQ+njU5 zhAeeaCgqJDG2#sM9;xR9jfyeVa*+uwaTsj#BmWF@6=}wB9r&E|`9TYGZE~08>|F$k z{7%o31Wgnl3)Lb>|0f#vX&soG&^Fl-zxu+@jh4ow1S>rHlOVhxeysK#w}-5IOgAju zie!za-Ik^}QyskW2rGty}Zj^Sj@RzSK{d$sl6~UtAV2Lm5I_aP9rW&i7d=CRo^>6R3ReLO ztdS)At)vRL8>19LOu&k__&YnO%9}Yb7CfG>DZ}Df>xfWg)=?0mqcxv-J8R8g6aXr` zR2x!0h%b)}zR(rBF&O#b$CkWwsY2kV;-1o2E zuA*mOG`_b7&AR#PC5>l3bqN#qw&Xv^b03<&pyk^iNC%qDDC9r&zuIiV!7wlTFkN=i zxn@_7k{524z-HjjOtbD5!*3~}FvTf#f6`W%@*D5`7w3o7!YUo#)CdY=i75}{dfsv) zkumI!lGRyZ3LHN1v&^U{LVAX0vY3SCuNIA_Vf3HIZm0A&m(WcMgUbm16L5>mewC30 zGq0q*2j}Pt_cEjd{cGw8pq{tTJ(vRF_Ceen(VGAvBO6|PpNDb0-_puHZcLr;Y$EXj zmHIMcoHT*Om79Vhdu{yp;Ld-Djii!&nPxg0S{R9}#bB$J_aG1 z{qM<*y83&xtZNdywo_HWOcWyt5GCcLt+BUT!yGL^*KiAhkpWyAkc4_b5-J)GhrKW@ zrqS=;6>J{YO|AmzZ`13IHs6UW5VPMm=PK7dx8oD{siJ9NbC;ma}c!jc!X`8rMj_^zzc3!bB$6&ij<3e*a z0J{tuAHXS0SKQ(=OyaNsOkGkrd^l{p{5^fQ87VrrrjlpF#7k_y(wsz6orhzR>TYU| zW5w1~v0HRvj$^?7(Ic{{I{2Lb*A2PW@o+^rYQ=5Hr=)m^%`5>cNue_@;a9ikSgXvl zFG=q9h$m<15%}G?glq~W{{CF}h54KSLGzlTbvU+M5vT20diZj-UTN-g2Y4gdc3rb$ zK)<2e(0#G!YGv3*A*)G)VLwNTx`%z9yF%#_L;vI#lQE)(>p^^Jvj;xVs%D<;=M`%- z7aaZ%4(n_{E=IS`rSIoDpHx$~2A?tJ^6c_+lgTw6?n+GUs`K`nl%o}1907q#HzU6J zol?C3Y`igLsk(`|S#UaR$HLrGde3ke)$mIsjSIEyF$zSeK8Uj_PK&58=7x z9gh8LS-HYL;;<@8d8zXDue7W&WUzHH_Fdc5dwM8#%zZ5P+_&;AH;m2D+p1z%iOou1 zOi3{HrciFc1M{8;xNTV%)kG{-B|u(h&carw8Qx%NU+hQ=#TT9I_ce(A4t`TbRz!ER zON+ZCCf$=(-Q?-u)H?94L9a)n{C&8LXWV)z6MDYyO&0v#)MzS**m5`yY0nSVNnRF> zPOGNVM!r?*F1$3UpBWS};Sa##WR#{Z~D>g8FoKV$(t_q3a;@S8q77p?zlS+?^IXICiUel z=+I7xt!h?m>9bB_s%On6__;p}G2YMBGhX=DzA_0Gcdq^2X(KT0icHT{Bbg+aqUqP# z$0i3iMRGfn$XmA?(&@IpNXNQfIrC`btx)tKghb-UF&WZdcHS1?7)@hCKWug&F&Z8_ z6NjRa1Aa(%y*=s1Bmkw|iyNA=9_6f4k6Mcp0WdB7$IY)4dYSmr-;iXV<@~wXSMO$A2^HV2#`;hEpowU~q6- zWoeqd0`NB+`HOSbSFW}03dE1EbCIgjxp+D|7wURy{OEBLZ2ZuiYSmr$OkWg;&Ysxjc?k9We=Oy zW#G-;Ysqw$=hS`wsvr?*x&ZEas3?S8u#a0*;Cb8Tg=I@LjJ7ozu1(H%PJrjOT`A@} zt0mOtW;$q#>654XKh_-O*|{4`-J4nLnBlxE6ReJl5H@kXVO6*Urfq3+M=AHbbi^J% zn`9!f6B{;Q*$s<%*ea&6A9hU<_zpuHHoZ>*f(E0;{z36D?;4os(RjH|;I8}~JK5x@ zb)!aOI%heZSfuG}>uyz1i;uLB*t+J`tEU0_wMY>QtsvILak@5_wG<{%RK8L& zys&wvpZ4#?+T==oZY>6P6>5xtwk(rs0jbBrGr9p&jFmT*Ikf$eXjk`W)nQ3#m1P=V z@i}j?oo^w*y8LHlGnQ9{XxqZV&xd_r%S+{%gSd*_}4LfH4AdXrEob-ud!cuuB-J-1;NS z)*9u$Gflrkm<`bQFioH;Lw>lQEHl0+#P7U6Oy2!fGt4CgMhpKCw}&d=M@!5(^i1u4 z&_)@^bxVa?^6a`1!n&n)`4i^}r@EvJtqNPYZ^!S)FztrYzd^f2w@Joxv78aZ#%IK> zS_7-CAP2afY~D~Zq0ei>yTegibYjrZ02lv_quH)iR4HB<#3(j;VIkdx15%zsC0gYF z40-T3FVA){H4^gXqPM|cTSnR{`}D1bcT)Ft;4w>|Bf{=8?{Y5|Ym1UjC$Zngnw6sr zi&f?_2ctU=mSUzi&2Fsk7Xh8?!THI!?_5aNv#ct=UzC1UsE>Y&c9tyn6AgN`UD`kN zlHr_ejCc@w903wk9p3T0WY;Fk$aRNdc2rkq2 z(AOG7J**=>UGVPwrb_)sVf#1%x<7s(hEwX|W#p#l}U_b3WU3~Qnjt_R*)|5QHQ{FR1 z1^^qe)1*771ksq?uX~tE)>3|tq3=(`^9OVDl!}6HD~rBcFIBW}j;&D8b;8EYKzvW0 zVU~UX_CT~kv@6O!4%`{iGKvc$%S7G3FgW$#D!uctjkch+9iaw)e zfyWeX6JWl^R@nTefhVvy($NSL&~-*imG`hFr4pq&1O9g!KbHv#)L~aFS=&sdIwt_3nXN4DqZ73H)yS9WPdBxJl+3WKoCOdED zQuOK*faGrTh2>93+9nhCSnvAX-3caf*Zc$zMIQ81IOCPI*P~$_G4X~UJ+Ax9i3uzk zek7zb(rlDh3RT?YDeSkgsGsL(k78J1PTU_vCF~r!kL>D?qbuO`lo4)Ex*kb)UYMM4 z8Q18EAS*|{Yuc1(lWntZb0_fQRwB63zrGK6H=(Y8_;&=CHa9V{Z$WDAD|gDIBk&n= zz%_+Vx2+Dm=2R1@*}NmNbdAh?|7sH)e>GM7rkU_md{xrPiP$5b2ozfQ*L#3$N7=fLKc!MO`(burY+EuYS?2O3PHl121( zKD|$f_ujZ)#r?!@wvsQ6oU6A!app#gCvM9cB^-&l-AIeL?ZvnS>GE8K$KUKOM__sO zCfFDRE)RXbUV8JE5`pRnz#8!RBxj@pTpk^7jd1v!_%@Hf~rEL_knc zK&4bl8tDdu8cJY>4iy1ur5iyJMWwru9%5*uL2g7khfe7sC8fW8e4abKALU)&THkuV zwfx650&~tj`|PWK*YEm3R(V$k;K@b|D>)OE7P2rzv#SM2AxmWGxhE?wUvE-gQ4u-# z;)fCiRE1}B1xL4A)zCFk&t@F&8W9;EWQD7XrIVk_5Nwj<+6I%tkaobRk84EE=pC$v zWjGH_P_;}5-c3bv-N*_@Su|&2%9OsrhwIiigD85zq?mp=$A4K523E5m;*>|=%5K3TapCDkw|Z2(<*JW z@j{d}{0vU<11HBk2!(EI!` z3?Y&XrQ}2JWq{2xbG^GGhC*-fCbNMKjmpGpCntr@0(~Z&I}bS}+6` zAH+1P;XXjvO1b1`m!3%T(aXJ*owxIvJ(UOzT)` zV~a~#%#mT0>&a*Zp~Zp+m8D#zNFgWHa=sT^Ty68!o=74h^<1;qBwZKK1EGTM=Rb)e zv~&gx^GHmy48pfK82Exjh8dXg&B!`QbPa&JK0uu5h`2JFhnZeZS}r8va=H~y}1eo5f2x*_iw0xcRYLU^;O#W^n7Qy6vdM}T@_Q1_w<7dozsXujtGw1tSY_?14sUkzT7?a0WynXK z$cwb|6l@y#B;UsyOF-8&1IPt%`nP?U&lcZ%>FLn< zK1yL>hiH3KmYE7g>JS`pJ}pFH7h7^FI+^zI_2Vb-+;#v)miJK(6?!gayX#ZTn*TnE z%lvffZLAhdCOg6x4Dr}%%f$Csq|X84Fx2pdq6$yrQWr-`g zn=cwPP6*x>2PIQiIb3JPbb|Bzkp}fR4u&H(ebi2~3G(&og09NM%^={HbCA!OTX~}3 z{03~Gfz~#WoJcIuZ_4zps1V0Jy7px8nDGp}&rO2oS;E^63+Q^!mFQ;Vdm_@M=Eu+B zZj3sCU3bmkvVq%5zMB^0+zkkvM`@SQkGW`{b-{8}_I4dRLH#KES)MlL&4Q&T4(hQN zGaRatMLD5b&ES-_0F7=S(IEIacx+Gk#RgNS}eQn5Sx_2e?ATS(u{S)mSuJVeF!!tIKJM` zWs(Mw?XJdbny%MqIhQFm*dWi3mnY+|$OtBX8Et%$61AFa;WmgxAEYd$=hN|cJS#Dz zstj#gZrKG5L-OK>??+qvKMI5JD}C7rPne0+bX%9$0lvQX9gBgPHVD0Bt(E|~I~oI! zvQ8n*7w|sGtaNT(FJWHo7Cgm7**a)S-QzTbmm|#^_V9&9mXh-juN74%zgq%OHHmibl?!o1ID`6!CdV6pe1a1ZgsNsMnq&1GiPTlKLXS-&AGAHzx2 zv02(y-kNp|-x`G=ZJ29hC7*`o3$U=FWUfR+TI~laEIc%eETZ*i@Ziu8wQbGpRZQn2 zE{Ju+U^1y5L??~F8dYP4;u^~sAw}1!$beov zvpXJ+m@unyyp*;LT}+3Jhq%Fk2H0wuYWKVrI%%HX;UG>XKdmiIcT}s^sNZ1Rldj&K z3T_M6l<#;TPO%h&w*M(nu}xpd;Uv8dssxWb%I>K^xe-7D!nq!DEK%wDoV21{>O)i6 zlSvPY;eVv|{_(-14j=R&WEN*ZBo@`gK>KKZ4$|*Eb+!+V)o>~}VKDF>`bMzF8weKs)~0BI`8bvzjcWKl{UPJGCw)KBnZ}noRQ@v(R4_@CURqhXbsHAn{*lN58X%;OhJ50O&Z*dQSRC zD`AlX`}!fx{N?X&@Xzi2{wLz@;2}3=v;~R(*Zcq5w|{N-cgO12cK_{!d{aKZPWL~h z>6bG3Ud{eeCcl))_ZI1&Kl!hzbi_6?{+$cp{|d8N1U0a0?OO96&6@zE3#ZHk03!qN zDQ*q$9?6ogOW$j{QBMIVx}_Mp6aLMQ_`{+2{e$}R>QGSk&>iczmSq0ypL_ZHU;Xvm ziThy}MT1#uSbk16wq*#=1u_LWEHI*b0|zWwj}#YBmTFARbRs{+S{8E1NF31#)E%d^8dU|JBx9CKzA~D+g(R|+reNZ5;B9| zapC>zTcZL^{+Nyio9q8_-|x>ODIJJilsb#Pr&#=#T?DTeIzOC8JwIS2n!yE&zqb32 zegC!H-yJLUU#I(@(cryb%H*HQ%R}OqGWn-*{J%(`Df@i;NrEq}r z0_qf@MjLLj3a@{(c>MbX&8R&dia?xWU1q{~FME`GE;t0kWy`=6bhV7cbPtU7th@xy?Ny*&*_U-69-FNAoTHicaj3@WRg0e zbf5iAalD7>FJdl@wQU_#m{)B9qNM5J*ewCX?a|8lk~r%!YQ%ol=y&<5H+RMWrr8|F zPm);lvQ=H8piOXZiSXnQKmvU+hI>u49EIp?t2z>on_S_JGv_|$|m!nH(HbT+~6%$YeK9%t+S&&1{5 zkHlr#6)-~+UX!?ql&YN@$?=pzRqmL`)Fld7PY0)~q!P^tZhn`<|K~$}(>?JwvA_gC zrq?Wa(%|c0xP8lDdJkr-qlvvrTUGl>-N_rXNAjYNW7VGf3egYc&AxlxS~aGWq9I#H zQ_}%hm%|ON=+hMxDjr;mi#$jOevr0SjBg5^WPqUgZOw&VnCX}i+?|`fA}Qd zK2&3qofVE8JM(n#JO1`x*Z#F9e;lD-=j4yA_~XO>i#?fO;jmvCS_Q*oL;CZL!hn{7 zAL_boPpx7yP@xi}8c7?LV1lth$y6w_ENfvm|5o4kWF>!hmG0d8=!cW?7cp-*S8lxqhafXKAlEbj#EU0yD@hcSWz~ zjM&~|$GPM%$>DqwHSOHYq5%_w+&WV7jgsiU>*BvF7-+ZXv!9N2{6^_)ukciy>{L0J ztvm%rOtS)L|4{g*U1ySv0hr2ZfJUfMcC86s`pLrND1dr<#nrPEe9VLGe7K0S9bg@F zZX7DH-d%S|@FHcBWv{T&0DIXKD9ffns$lJ(9~vVS^nV01ckUi~?;wG((|r z!xqz>*Q05QsJ9)ef&IkbU*z(LvM5a-vT9zN1c7R;!UVly<-9pWh^PkCwo zn)tmB_`(NYC#DW|mzS*%Egfpf8^P%kMLummuPCfU<+4)KS%%d~EX$(A+js97XEcXO z9FgYF(mj+oP?gaN^w!u8siIsU{!^$&YR6oy9G}Xzj$;lGMV^-!15j*6z#J>(2a}3W zHUIeLlEGAn8rAt1&wZCbrFs<*V{Gv2rI>3O#n?J!uKmnXNisa1c*_^M)+%Hv%>eOD z!g>2223=hod>(`C7s-LY136W_IJ!=A$xP#dMy=7IQcP#*P8T$DHWl;~vX`$mGShYC z(cKAl7vM#|=5*Ecyt7Za1&JEydPF5*DLZS~q7fNL*y23P_oe7R?ck%xYN$|~$rb~c z-f9O%Td_YXj{&f0HvZkyQ_w7Ww|@IkSMNMin$2^QWRNe)a&2 zm1YI=y|w*pH==;ttBog&bfusN*GfBK%pp8I>WLEY&#rCW{U3{H(O$UPMdkQ#YowWfW3NZcGAjj^ z!5+sXm2TIa4vqAt+f9AYad)sgU8M;kjPfmm*!~b-(_AOSLqZ3(^8Z&|8q!Ypgt4*x zKHeVYk-Fw@)&Q!}BNJO=91 zh%~)mWPROZiA4I8A;jEkaSuo|8{eplbp)pYY-^<5Woa99T_XTXUN)7QD*%J{ba32H zbYO6S7kiOvre&65kOp2Ps@EY0a|3rC0B)DoyqRkfvHb-LL32)Vb0kkjAtU1eT&DAT zMf&GYxRu0L9(_#=l&b~PzSR%*+74a5;#Rr+R!63%%{kZz02)H?CKv{8Zp+(GH%Cl? z)i5(o-{{`-IO7s;R@85Jd)Reo>%8#gy0gt~V;y)5(RU$$9kLig$T@;uV=q+qxGE z1*B99M~c}&2QnALC~cVqo4wxVo+4#xiFci@d40W0u~_vCx0mJtE{#X-CKHU@viymL zO^1Y5z1Ba&-9H~1%+R<1j@bCyJGZ3ph?42}xrbJie6wEXPO6V59_*jzEZrP1ZK4a) z?o5oob11cE+%#Z{(t#igfukN}$#_(6A0%P>NqX?T@W!Wld)3feg^%1+!KB*{Ld}0zB7)JO6k_9*6GnyBRO8tN^u(P)S9X86pehdHa*x24{HC0jqU7 z#d}{MgE?R9s%r^c4l2fhx4Z*sEJb|}TsND1_P#iedEjaeT8-HDXryk8{3x>XNI7+i ziqM=^(91VPEUFtlH?= zONr9RQifk!?5#SmMTmcf4D5(voM?sF`|%f3pzlmg4IzjWJlt`|_A_~PuH2W`As9)A zcMOn|nzv{8(eDMMM{we(QuQ1l(@qa~^xA(4lS$D*ik*(}l|e_j@GIXpc+1msvBUep z=0ZQhUBws@!daULs`+VymFslv;Jc?)_UTCUKPHMmMl(G%7RfY78_pRRnk(~&F1sRi z>hi2hqT}XCLCcCoz?5e;JSaZ(S-FBCAzawKcFx^3rrg&{e|P)i1A#WRkI5D%9qm@S53Q`@Qjv}Z_q5? zYBE3y$|MwxRBQ~4%wQ7%ZZ6AUhz2k{(2;RF#BKwvwuL7O4JYO`?U%22%A&{d_L6p*LbWV3h(Lxa516zqe)F9Q&;|_G zqsMdH2XwGYxfT2Jx5Rsn7_h0UEs-nHt28iTWA(~>=AZ*FN7tL#{uM}P0 zO;qSSsUw*JE!$?ckr1eT3hMRShDcXUhBzznp6dW1+At8TB--jXb+;H(+2F z1?X(tfAI);zWg9AK(MF2ZN3+KV^ylu;!+$!RU_F=2-!>PdtMlss@Y^;f0EpcV@v>K z8xG4NdD(@ggeCxy?99{MLFpg(=HJZVN!8^Gf{m74c$8CH3jqatMcWVu|JPaD^1YK$ z?}*5BY?Nf4glvYI)$k6QsqB**-w)Va<8jph=8=Y5edRjBY6Pf$H8lr_CoiOR2LLJ6 zi19JqtK$cC>~2ap0eK?#0{~JQH^$C9O9+r%H$u|{dUjc8j|GW;B;*@v*LU5GLa6B! zMuT1<*AyF*Wowut-5t-`?KiRcvqzUg1A)mu~#j-2P1%S{kz1W@NxRwEr$Rd+?kMm_{ zLO?2v(=hINC~-NaI~$Pdnpk`^ll}0;RBu<)Z`cCZIi?)6^GN+>4Dl5qz^1L|gVw@)m$3+1*oV?sBMRBo%8)#`tSDn%j0B*@{ zv$4i)WAeHXY zeIWF(h*i2PJ6ZiUzVM_jm#i|r6s=gm=&VZl)qDvZhjA~uh=38+UAhzgAIQ)D$5wiu z7lw_|(-jdT`%E&j8bL?B8iKDBSUlueNu7o%KHJmHZpnO|0TEYvk2`4m zGeumLk-mC0yp+z&zvTFQybR2a?sO1`wCDvf&`P0|;4h~Zg5h5b@f$cnyCJX^8Q%$J zD8wPC-zEBqmLd?dMq;}fh<%1#+UT6agWa$ezR~j1HtdVab!pl6f|+h^_l_90L?T>g zF(xp)^K{!=zV_vbNcdVJ^?n+9H2Sc7tu^C=GM;2N!r7{A!;woKO;)+$)kCLj%t*bL zPz0Zl5l33!EkSgy>5a{FzY$k87l()XNP!K0BIBQGaVb^sR1f0?)#WhJSkTpl9)$BW ziuAJH42D}HtI|{?7?;-!La!h;6(v#%Tym4!7PCYg;3&Tua3Q=Ocr!c7ub1E#xH8|}*)?Hf|vhCH$!srX|-cNDSJKbXeI7>`S zYMiQyD;CY^tjkUQo*DSN28e{hdPi1@Fk0Ey)yn`^5sd zG;&2ErZ{am!>#4C$~L7?h%ZLN2xAT<5P)Q8@jO$iuy_YTKHyP-Q_^TaoFdIZcShfN z7w{4zi0NTk_q|A9oz@kDkZUP#5V`?5mLzPL3dNB_h|w0B+EtP4 z3M8HdV$auOoe^6+9$2mz2l<&{;92=BoeQ47fVyj#P6wus5@s@&NEs^+EzOwq(X=R8*#HvkH5d!gQUCE7CV2+2i5Wus)^bH(yqUGLR} zLNUH)IDX0=Oe)9dpZ1P$vZ^1PEldRsln`_6%q5F6FX?VXavM976Vz|6wKPqQfYfgT z?E2#*)-F2mPEG>u4BCcK8_wpTB&UF=#pn!;&5nQ~4Q(JwT`eeZdHHk`ZCLeoB z;pNe02Es&RG_hE~U<~)dx#=iNHjs>h`whP!K9KICn^EM6hl~~WUBQrLm{+_u`&5;( zBR$DO%zPoPH1XMOFl<9>WFy;tV^(%QMdYjvfIEjNx?)r|Qm9Z7Y-}p_$%Se*jl1q& zz1^?R#JRhXX3P}|%EyK7%Ai9)NkX;jjtdfW~m(bE4~~N30eLqPb?4 zy1~yYnP@|H!zl zAUQ-8lW)4@4Dt|cZ2;*)@<=wKY?GJNcX1{vtVv%0ttm`QVWQZZ!r7}evKlHno^`O) z05{!lQ2%Z;=%4WoM-`4Bj`7FRGs6_Y#>KM_Lny?VD5vZv96OZFXhQr*pRXw4|MqY? zCZVWVF!&uUqebB3-a8gh481>&?I-AUV}Lz|u@SAK4DJ}^9AOza^Gp0rkW#X@S#1)L7?|r=H(5>rr-6O+=5d|`U6=fYxqZTp7 zm7K$B!`UFwV&YU*zD!xrSh^53X*7UCrrn+hgQqxVq$)I#*#rjc7s}Cut*Cs^kGPy= zu~?HhQj_8oH(pLO)la@qA`_AZH-wSTRx0n$LDPK-^-}QU9wDnjeRT2qlXTqoH$*-l z^vC-~ZR?#}nat5DOw&WAnYkZd%2I3+5z?)Mw<|FP5*ALKG4a$f-k@$upfVNvQ94bZ zwVm-Umujzi`)w)EH+dH^q@P(X7{ic%Wr}ZXFJ*M3v}F^x&uma4YnOGsEP8Y{=*z|H zD(%ZP1Y;+GoSep}pY(&g%RmlUO{Qb$n=z%%xAEK`Hs;Ccg!0~fPct}}ifrsQgIsIX zGXf65>0Q8y%EyF1Pa+PzoR)0TsqZJE+5>zNPV4FBo^1PJTQ!8pb!OtP{XpiMfEw

B5F2QN8rOAiuw<=KyIl)^2Up|F7yR&+roHRc?uSijh z8yT==q0nN8swZ$%ARf!mn#OM7cC9s66x+A)%X7ygq_g3UwgEkD5q_yNbA z(|L1Yb7`wAdJYY2HT!y*>pSzaAVKkxkM$cxV=zSE(`O$%*SeY<#H&6PxigublRs!t zhPL$Hi##)CUOe2Re<&v#n%)N`0(*9zvljs$8>7d>ype6yJ@942EB6!?U3K6U*#Qt! zA-);nrW5-NT~&025ys8RPf{xi6$eA(5m;)*i#3th-m9pJPF%C!4Ng31-|0EY74&bT(?>By zAdM0RH~nGQ_oi4p^=XUrZnwNgc+FC7(Ce{|dnT^M1XLS5moS$>vH)h=hcd271U>o{ zL%?~YbP`-aVuup%$&j0@%=$2AUx&sfRWmFZncgC06zLhU0FnGEu*QW+R`Mw#rj{U2 z2qhXK2RVC(!4Yf(V`ru|*=)44naK?FQ}$qD^PyW2_$*l>_iQ?4 z7(#*+MN2f}#Ex+pHm8NI#bKj*^xXG*cE?$w9j7DE+;#$n?{NJ-Ha_7$NF3%vyjZ2^ zW$kJvOjTC$Edw2LO_>_`7!^1oh!kX+FKIh*<6>4uE-o@qQ&wHP>*u0t1$b<~S)=B{Sw z?ITMQX?*$Lo#wIe_n}WO2;%Jq#J9xX=9202jyO!@flvU^bbGr^?_f$#wF_xGz(?1Y zUAcp?0(A())ueKi7Sn0a+G=btP++`)#MrWUN~M1&0Ev3lnL!YwHbK18RK2t7^j+!| zD<239RwoJx2`l3P?QlUoPNGu0pG_uo>tY=WCEwmHG5=-0W#v|`JmEGtW-R@mbf<%(0j8^rsU zDBzKgc&dNP-QA)qFMKi1%aegC9Uiz;eLJirBlelPP=eRQr*I^{!zR(%P-X{k^zc87 zoIC*ryVwU0nMN~MLFvF*AZ@{&dLS>aYAfX^FA+OllyK6Iv-ti8E@M1wgVF-W4k%W# z+WS&pobqk}#QKelsnMMi%0NRo9c<6YBAb)w3RWRrym}s|Do0S(9F^Wd1hoh zeA$&aP;mVDc_cba+e#*a+gQ2ZZFeza3!2o`1O`!cjCjKlfy%6mddlfo=rT@t+zG+V z@Y$L~|D=KLN0dVCt@#K=kqtPBZU(#mm$n&Xg^y}QR#h){_CSW$lq0#qyv_|qcZxNQ ze~#SwTUCi$K3?UMVUDPs{zbzaXG}#!kgTDm%-8ExrhWzxB;{@;h=N0LuG+xMXpI6i z9=zyUjaq`jA4sFP5#*enaT$J(8tb?&nFT}=OiS3Eg2wJF%`P7-EjNxHERGWGF1@%c z-1rL0imXDFI8dDlhN?yCS6@-YvB&Q&SBbs;R^Q7&RXP<8?TSV07=wZi8=ep=0HDyx z)PhXe`@%tWY(J|!)}p?5v;h&jGIh!8=~cYIR(#7x3_yWkG_xKXgmq|eV9EP*4NOnk zhM~8>Da{7>34B+DIOBdydza%(jM@m`!vm$79FT)QbayK25G0xw?iu95xxY(2B}{D! za)|!yK}an;4m4{*LIS;M9Egwuk$Q{f!0Dx2;cBl$8T`nYu4#cSO{o%pVpZJD(jYQN@;Tf61g>k%)Sg+ngA8+d)|w|jLYYMF)_n!vUX%FzFo z2e$NuW4Wi4-abaHNpBymw=w6KvM~VGpAj#6iwnwmL7B?k1vQkzY=ZaBbJrK=g$1Fk zf&`fkJcjLR>T}Z_(}+M8JwcUJ1@lEKFmx{jFwnW8cj4qapz1I%(BnLgI^01uz2zHG z&#Ih`G==R72Lo3GD!*uhYge41?%dV)6-bE&Q4qSaq`1vxnrok`7^v4UsQx$L2F|n^ ziJjo$UVxuVQc@M(lZTB+_zf5xvXm3CtZ%iyPQ#jz_}uTL2~;xGFfT+qEpW0wliC+? zekQ_~zOoL@_`w(*979ixBu=2JN&sjHhSW@^GqZ!H4K~FbtbvWn!A~$43{lh7_-2^R zzZqsl0=%B7gF&qnt)>Fg+r~76R4ox)srZx?z_MmFvE(;%8eBWLX?9%wPZ#q~Rr+7E zJc|!$)Dg<%O%A!`-Aj$T3TzSPGfz(t&3-Av+-*fW$APj-lFPi-sOH!$sl}~`2iLdF z#kL}E(WH|Wj=vyRt<&Iimd_(4({;EnW0+(QmGBxtwuEl=PLE4Sdd0Zv zk*Hky=Gy1i?4fm}TA4dmGG1{btX+YN#iMQro&*%&d@=}<=S{Z*CP1wbBGWK&6^-7u za$0sC9o$`61->p~{=7B}wgOUZ%lV2a$uM*%{1$8@8Jpq7U}qa}6pE0U!R*89Mr$p6 z%hr+DMwNS0>}vBypv3cS0O8iCQGKu49vFC zIhD8_IM2oawSBs7p5DhDPXe|yPnh#!p_=KMI&&v2I~)Q75}he1wuM6Gfbu2Z~u`lNpexxS%$8R zsF$tVSv++r%9+Vq%UMD6@<>B%*iGN&_XY8MP%yW$cbaN{HLNNVOKQ%7DV_rK zgVWuQk{c}q`QXB(Yh8hKkGV~IGEH=tQfS`?uIn%@?SWZ|*~r@NuDi(>`V`$(gH#QY zYOrXt2=<>`z4;DaBO@#Q0H}xb>_doCDN>U4YQl4UVr3ZAMXe?ZJSDZJZN(y2Fsdu7 z(cfUHmLhJfrNFqW5i*4c0vKJAAY-_`weJY)s~R~(NDN1wC11PEBfQ%N&L1?4x39HbJiR^@ zxIPKF%uFqzjbbHn&uY;DpE^yB0W50)f=SoPWlBl%cFBa%HmzUv6_x8S~R5HX<>};Z#LvF)f^yX9xw20 z@cZR{144CwZZdEbsj{86)ht-yx418w|0T)6yIn~!qM0SAKaEpX5&vO9@!rR%2W(O1 zMH=&4dz*^{jgab-v~oM?xBAoq6c)3NDGV>dOVgRJqUVXl`N3ePCUE|BQJAlh#to}= zIiS#0TyzFV%i`C1uXX0}L}=J*bBs{&)^yMA^FSiO*?3X3dtPmG1XsqvX6#|E?Fuy+ z2-q^L!1y0k#9JrJdAP>)U2PC@pebHgY*EkdgSx)T!FJqX2`zS~p6t_E0H|BAZ_L@8 zfFc}a+zL(!UCM5g(bO#~IpQ1)CH`E+C`4~>vedkuYN1ale7P~7ug{u`swUCl&Bxr$ z=uuQI$ioR`jA^?rJ|6-F!7M5jlAb}y)kq^DL?Ne?v2#=R8^-U+1S+@7Yg7NF+{m}?{w{W=!tFXzsBKV6O>%#j%Q z0Ha%dtRAo7RE@YdRMKrt@ak#XgYqoWCzXo>fE72ZJWV{?xL@GS#hI*)WQK?f=@4K} z9b47jstK}WA?mJ5N*Fmz7LA1Kfg*8+CP^k{LixB6$Z@k>I(%L_FbK4Y)lyAtMNMqf z7ksMZWg6S&+zIq*U9`!aib1HBPUolc)-;Lrei0Z9l%tr#HeLuN1V-EXq){{cKQ!L5 zSFT1C0L5tMrjy5<-Nooq0O48lFA+|iWCIEd3OCpL5brZ^rRL>*g40BXZGS!->U@<^70 zW~<}_KJ)4AJ_Vs^H)4*aD!IeZ-Aj)m3lwH9zP8;~Zgi#l8vI1@nw48|fUxsgncp)V z6rS(1jogs~$Rflgj<`Q2u59``TGy579IiJ;~wM7wY0in8io;p_eP*##_I(!1uP`@8wQM+*9DWF zige!xQ~Urn37Mql`@^X5tomad@a?mP?-LR=X~z@c3rb2z(o0`?B(AcR&c3s%CFEOA zf80G9uszJ1uagpH}?dE&Rzd?jNMufC&yCaX+{x52~%mc%BpnJI)`S z6_8whg!ix~4H|lisCn7bc$`#HIl$y;SIe3_06<0e;ee2Y!^y)%W9E#qd&uh-)?N@^ z4ED(oBl`A7n+vq-bXGMeylR3gj}p#Zo6pQw=m-9TcY)=3J7Qsz57%c;fIk9-YzVsZ z4Pf=sh=lk+Vd z*+YRtWkSaBq~~ui?kXvA=~z+_XpaMR&;CanIl*wnf!l2}TX++T%czwF__qE23gyFg zgCUe4<0QSyd!L~uk^YUR=-(~PzZoG`Q0*c0^1=}x0}ln_F)J8C`a%346kZ7cm|4^M zTo>*@meL@$-(?;O!JqehouYC)Ny_b!02jd&nMhp%uw|S@8C$U*)fJ#PG`#2m9#mUB=?`) z01`!l=hdJFt?srKH5x2e6&h-chQfJHP=m`Z7pqCkrr?eF5%S^>hnS-h%TsD%ei?I= zn_`zLi?5_Hd&{RD3A_CU1ptp-;%dyYD(6Ts%_k~f}%AiItL+5|zdnxX})ea1TO zb}#(g%t9tJ0&9X)SQFriq8Q(){h{Q4_~A4YhlNTO=(-Z*wfjRG|L8-zI?&V~iQfAD z$RqmW0*?S%U&*>&vhDkW`!7G?Rt1JJ_jaQG510JAl8gk#t9vN7gX=pz@o%rWaA@D*Ap z$whtyfByX_U-`A&-@UJ2+x`8?`gOYhP&vPU_@zv~(?-9P$uDK{KWx}v?c6VA^1BZC z-PJhtOPTyqCO>yGe|aXqJd^KK#~(W6j}QOP*sH1>CG-H#SD;KQ1W>iuYQ)~ zLhw6J?wd}_S9rK09&*YoIiF;(ZH)+wkm1~poq1w3U9jaU#z4E}IBkD;NLxG@!+zCl zdwV-DEQ~=HrDkPim81|OcvZWO@zj;4$FPp{&)su40O5V+Muc2oWF%8VLqlM2@P!`k zpi@*{Q*{kTx&(A9+ULh=5}!M~VZl0%eOCO~5&rS`ihowv2gR3o^{7krKZ}L!NBsFl z-l-$r;yCtaX7QEll_PIn9C0_Eir+eQRv7lO{?oOiJq38smniV)U!AxvA8o_I)8btP ztGY@>^ZG~|1?}aPO7Q5P=UX2AdLchQTfbh&F9|t<{`;k}eyOaZ7?S@FZ3r1TD+7a+ z9KA}es}ps(E{Z=!;YL);gF;O@Mn;+CZqI#S>_6L;)lGbPSv&6E-QDe}Uw-rF3h_T? zCY}Y@kj8gpHHDfNrEiax*`p`~ZKT=Wk73Ek$aGYwNJ_o{NODO|E-r%Bj#kT|_qowl zjf}@nNDo{j=d)OU8xR$x^*CnOBCN=K5a#6MR3kca1S$K^pPYRKY@~GewfB-4s!3jN zHx~yPG?Qr#dlIG243BjL&`3R;Y7Bwg7&Hq z<&OEw%B){iDx7RPZb=LnUJ%wSGK+ZRkpKt*q%%~rTxLuU6TQMi+4YtPIBb`OO4MIU zm!a(BOvmj2-pi-Okomm@pvlWKYNJZm$WO7sbf@1F`#)HW8d%Kx2j`-zZ0A+=Lf)Mu zyez%Hvmue8THx^I!D!67r=+{0(S%u1jCDy6SZi@qj%>3(s z;h>@B8E!J5B&t+3!@=z)&+6>-#38t2T4x71WnT6;y(K(?TlVcwJfyKtQAxZbo#xTZ z;Lhp{aJMcSzWGc_)8JaxEu}2YLYJ(Lz1|(Y{gAf>K9Zp8Hru_JMf-hwukRw5FS%N1 z#LT}kL+(yO$C{bs72kK{ck=QM*alh?gL3-du?}#`9s(L{Tvq;*Azg6 zK>n#-8Sf%?R$xcDZfrK-t5YpjSfX-3|pkkWqJ3qtaT zz1cc|_T{`W8X4lwhdT0YpVNWcvSiOjQ@~a}=|!YDu1$BolX$IsVrF`}ac_Igb$1{A zb`RlS5CfpV3AA){A*B{zAM*6;iH2dYWVU;_NA@?5gSUwz{%Rw;@uKiRrK@wM#(S}& ze7O!gR`;$vFQo!w5o*@EtnEp|r411Y`!7@;RwR#lGBrwksh4Z;sOq$|-;ynkDQztc z)3LJ3e@J)ePE(>W2Jhul27|)|P7;P|KX>u(+`Nb4lAt=sjvSB4wwMVOO?`MNw)Q6W zXGT!}vax5QIiw|d*T5%O!d^SXi^ltzu!MLJLsd z&>HvW88FhctkhGxuP-ilBzh+$CB3Yl2*4pAy)dY@`hk#~_fxh`na#4+s*5Pi&zA?+ z^&ly;HfZqGE=kWQ>16%f;?>V-u2D2zzFWZ6bb?Idbldn9tW&=!JoS;v%|};QU$03t zA1rDJWz|Z$`RrsTIL*piRwdf2Yi(k(CMHPj$2B-O^)oF|GXxH{ivx&ERnr=`!zgry zT9d~K+)0SIEJZ$Q$9~P{=aGy5J*FIjM2}nVDqjy+6kSGl#Cv7Hmc0*pjk*?~8X-OCv_5;4k?aE5L=LbVc;MuCYE(GOzcg zTtf&owzPz>>68M4y|UPQz(7NDJCsG^bD0tNZL-q_rcxi+^h$i#?$&ZxsqH*$7l4(s z#|eMB=BMJHPf=LUG;N%Z4h%q)%4C;O8fvwSXzUf!}Ua``G-w6rHn^*+%Xv16Y`!8!?RjcGdt zw>$1BHMN3$n~=&_rK=U~d8giwddur=fvcu}8sHD@%+%mo)UP%N8<%H2qo8+JM(8HN zRYZnr3mH65GwV3&KEn| zuKDb4M1iP?Ixkb3AGYlcoOd>e(l_)vjy zbfqJgCE&Qe-)oao7dAJivgRJ}ge>pX#&VNg@jnJ-a>lOaVgK9#_dqh|%&5n{iJlEy z7&=8P3eu93l0Q4L+B6sz@!{>eC!y5Y78Q*|XyRb$06uk7AF5 z^Q%Ww3>!=LM>yNw*=yws_cbX`@7{5GW?C;~aQLPCC0i9LOn#?G{_U(L+pS|BpKyS8 zWKkb-q-$qX@~Dxr`a!qVn;UNdJVWr?(;C|_XK zHPP{dzFx_cx4gi)(LC8TxZ@JO)^EMmSFujrp$CS5lP~+$&tI~5yf)4Y#m5c4G$k*r z`Ah74%7uPg5;AJ!^eeC6D8NX6G1IYYj$z-gH@ujgDE|zuQKncGqg`v#!1?8*_qYf@ z5QGjga9&+imX073PsQbMMl@#w(wpAgzQfpX%~Ef|pR^DiC04BF_9;|R&2U!?Q%O(v zN_d+r%0@^4u@(J@GQ4z%#*&pow@ItFZ>q;n`^9ZHdnumU35q+%JZ2$Ltz@*nbA$)I zVu)qBIn|)k6&*PYTc%p)s*Zrjowpr+L?YU6W0yyIFMPM}?8`!+lhp~V-G43*$Fbkh`iBIu*D{n$RzKN4=F1Dua7tgDz8Cc< z)Iln1@^tKNrE#(SrwYzjZ+!GDXK2WN><(D zW^qxGXjs6L+lxx=*6W7{$~@Uq!BR)(;=7&N5cX%IU6$;j??U%CW=#SqMpkX>QE9lV-TD|l`rF$5 zHF3QQRmrb4p54EFMG{zv2iQ&KM_Az5moKUp3U#;JqMPH3oiJ9ATHu*=^iWW@zoNFZ z&EJ)?7;#v-lo~#XhwfxaKkXQ*BegcUb0+3M>!b@|H|Nr7F&AxDBK7-AeBD=DOz3dS z_1-OL=L)GU#U$u$r_T!36FrfBJ@9H#$)=**uL_97*C?)QKbHt`&r=kQ&0#a}!HlTV zsARL*`xLP&!9M;;WgjRt4*va5c9&U<~j=eduDU%GcBLXYC5h7;(O8 zW`LLI_1IbAruZqUF90QZgjHf=y)Je}%zk3Mgp|<8*JR|JzU&Ow^e2gFtcZJWP{%g$ zRKeUi?+vvfnbs?Rg^u0lI|ubGx{fZYDU)80YjW6(IX_1ioNZ-s^HDJ{JX!9nU0zbR zMYeBaOr9@n^JMN`*V?E-gPN(Zgd*$a!N+XUk|h@FT-R5d_gDB+_r_08;k_Dis+2fs zFK6a=$GI}ZTm;*XY4E&$$L;3&_?^(9qIxkj?5@-<+U5`JHws`?ng5^iwbd>9Q1y;f3AeLR zUJlXi>Yk!!ZN1Alyr&OLwQF7QSk4!z5sCEm4_4k~kcPEEw@ zbY_I+9&tm)QZh^O`EkIO&fd5Ou(LmYft)o8d(yJWJshhqaBR0 zba-~0vTJe^@NThOs``y)!oBm%*9$Shy3%R00urNJ)KORKhj!*Wky3BTP1v1eYIfrg zA?a1_p&BbClTo3zDpN70W_0!p%`>m6oF<1Q5=6+%JB;EU2D)$M(>L#ENY_Db-j`TB zt|NE=X-~1Z5IFCO-I@c<5MQJ%Gt^i$MyMZ!}q+o$LASl;h_V(Xo;uaN+EBAuf5 z2<1aK-tkln{^k%xG0Q;U`A#R)qd5Al^*81C9p17coEA@3ki3>U=3%NEuwohl+5_Poc0wAv;BVSxiFNdXuW@4Fo=cJ=xQ76+^cjDVMqa-*sl*1ma?NliaO`?&prXvg_`&GsPLu}|YtEKKrSOeG)_KBPoC(J^zKOUA-V}UPz>@7# zsBzQVX9~vYGMt;5W!%~4M)7&U`D|$aqEK7N&mzMqQz&4ac1)?k9387&;#`_Bx%40o z7t%b#&v0MRA7BFP$>xgRm1V|gELB~%*?JI^7EfKUqE=NL6PTr2JG@lV&ZHKdO?`@B z%PrbZTnR+D5j#iN zcTp1Q*SaC^p_|V_fY>mj0w|A3y)cdBC&8N0`vol-nj>?lk|d>$Hnlk80IkvxH+qS- zaS^NlsD=;so&Ra%e;k|1tD0HtVN)veTV}jEldGB}@n4Wrn(|R@@gK=7lxoswA{<|I z@$b?G?Ulu$SEALaY?wZnjIr)CrSmF<4sAagQCR6+6)8OyfG+xo}6%S2B;H5A@9rA{r`aNB=>(Uy{+ z&`CL7hVn5|)fvQTTdpJG-nApWWmy_aO9AwnQRULb2l`x-;e-4u$@a^nw#qJx+v~ph z)E;>Ns=)M`s`m)lRk*sxf5bhYFqKmi4s5w_M=!nkW!k3D+m2REQVI=(ffS|Rf?U>2 zm-80y>AFMBCD(jE-tdF%p5CpdbLZb6i%weQ0XY}PtcQB))5tB()2{2D8j@=jPu|HE zQMaN+rv4+17ltRCa7mH87KaBttp5*XUl|tVyY2lKK@k)TLZwus1(a@3LSCXc&eLk?w{Wy1TpPeGuJe?`t37ynGb7pF38s-&(8v5h!judjL5G7}h*H zgwT%HYs3{NMtgHj^BMK{$=lg|yJ;y;n-tvWoG_@9UvS9VoLQwKdL7M|K8t%G`mD*e zD4c*sQ~ru)b3Q#{5$GRr=cH2_PC(grx0Tkv@YPq`aQo?#1YhVs1}5;r=MpE|PrGIT zMSySGgq8JCJf~mRVLG@I5s2STYr9sdn)cbs*W^Sa=}f3S1$8y)J6U{C;bOyKe7Zeo z3_s3&A-q>)x1B4B!l4|Qa>ofxiBn8G!axumS&emf43y z6w5l6%GkPhL3eHuPY5Z2mI})FY=vMT`FnkoOMtAI4&^!oIoM{bH;dEyYRTD922-5C zt5mpHyU5sso!(4EtJ%#qpCptvJ44?pHM*ohNO)6Ti8~@6_=#_J{{tPJ_;{k0*%U-r z)l(%|^HthamXTNSv~TWp8DFvzSz}`aAWWWf38|7z2a2@h>rCjLU`gSs7>0;R(i!jQ z*$d)4x$Oy<_0MxT@Gm`3x+vN`x1Gi^A1tLWeISjm7J{^@J0vLPp|n>IlhaD3ztDA5 zL0fYOp=^44+wbl{Y3%F1oyW+M?rCPI(JPWxwbEuSg33FC$aYSnOb=|lTnFpi;zx~Y z7lDY0qt~cL?Frb1CaXDgRK6$tSI7&0z|5fv3fv2_apTNvSvJ~CU>4KnIrSsLB! z8MB-Yb2#LeWd#XkE|3#%_P+{vry{>^2AjT>!%rpdX_rBa-*OgB#a99ztx$`>82WC z^U?_OaK544v<9(>Z+zA}OO#M};17v!tu*p;=-4Q9qP_Nb`I7t!nn8mr0Zx`%DIG-h zVITO)N}@T?v`vcAA@5PgPh@vVaTycYWIliLeE8{8;qxBsF}1o0fwTt zBu)$FO1b6=V?Aw_W1k74O1Oki_bOZkM6f$~oki-7?1}T z+|NLGn(*Rr{M_2cXIns&PZpXq%edpxuNO>Ie+@T(^9qbCRLIE}u`yZ*^y;$tMvBaV zFpO~2&?t@`OEJ=LFw(P%td@4tx_}jkksq5D$`9a>hub@Aq;biIWl$PaMlFe^ypg{( z(e5{~-TUZmswK^j)7alJ*f=F8tjJJ7NnmoIZw*I??O?7H5f!Usd-BakYQHa3gO?qexRK+@<; zPHA3h1U=e7uV>h8{GM`X5@VRoCQo+L3O$qZo|&IAsBqT3VmOysH{X*%QCcPChATz!hm>Lj#@)CH`&6l*PzRkv6$!|L z1oe(uwc?qyteUx$HZyf9d^XM;};K!+zLt?m7a3d ze!lu>#!=~xLI6a7W&3IbvE}D}&6QsnIp4{elwhM9y9YB4Lz_cpLp2g~w&mZj5>CWE zcc$jF2AK^kDui(5q=>{z%^kpfRC!n3h}LcTfvEY5IyP%Ub9GzFLO?uyaxH&$Iqzbs z2w-peMu*ar-9#whS0K;hdpg*IaWwEC?Wrg{Ab#SNL;|;JSK)0=L>G# zuxUMFENap*!BQ%?CHoD++pvw|Ew}RHH8jGq@sEwzKg$oG9u&N~jEbqRyVJDrSlZ}W zarzP;1m>JX(lg6{QjHVBZuccx3`7*|%S%TUJQh|xk6q|0BaaY>sFAQ$nR$ALO_m;?$jea7 z?R_v5>Pve|2MKPN+8D~nXJChf8+j=^$guM&lCEfl*xm}BdSN}>5DD+1lWgrbb4eAS zr|(;Zft;GuieLajg>KcT&o% z5P#55FWlSl`a6=I_O)Do5<<9zce5SsL`wEt`Gmjqn&v`B1H2%t#vc-w5v#x>rYmWX zA57-&Jg1MNXzXnQlYd+nKiyIZZ&j5orj5R;FyjQ*cJM=L zrKno|ja8pN1|)U2mLP{`H^p^b z%xG;-6E%N$J%*~`-dXHol3z>y(u9o#USuAVuL4n!4a zf;zTqvA#*XAozAS>u!(FR=!Sq>G~__+S2b$U#z-c#c|CENvr?zqaDY?T{@ICLabth z*@w3d1BRGfKD%PGuFP5XnND>|i-Fu1NI+jRc#1mY=NemR!7IWo7M;l#}$ z(jnV)+Dr&e z+4F?mjyWPafuMKkNd8U6iIAPMoiHqk%ppJ^3h;=+q5_F7&jwPV_pf$@bjgY-5F5zx z3Q;PPopcH*j*uy6cpVV=L2Cx(jB@upxu!iRhM0_C7Bk-pL!UlK{y`yb_%y=6a4d7L z*$Q$C#1SKpG22urZ=eb_*79PeoV!LaXz^6i6Dpw8wQ{AeIaBf`IO+lDJuO$|=n=Cd zQ`6Q7?VF=KIH7}fbB>~;Ed3y=8HB@Gp)sDY1s%0&)`)NoG(Wb%F#y-QC1BhGunnRg z#_#Z7yDZl6K_C=eb)Qd+J@#Ikp+HyUsxm%@BN!v#qY18ZCK|bkj1V-HNf~6tWvYdmRmeHW!otNwv25q68`#q!jBs%a5m(poDfiu7W1ZmZPKrUzXtxV%{DY%l!NJ1<^dCJEASqJf|5B|_LH zchMzUFDYNODKlT#PMc!b6*RywoYway^Z0o&#y%myT4~H zX6K_k!g-wFb=zwbBcL#0!5l7zjGX!ALqDNPu(hQ^o3ryBvcJ<()O{t1&!lbh1J=-j zNk@BA@R?mgS*CL0^q|vS{*$BYGVYmJ*wXOhGl$XqeO(a3UY{vD#ko zl?mw{+UT5f7rZ4TIY0~5?^8E~3?w0yuz8tm6xVH~w7fXJ3$3h+4j-Gj7PTtdB#BCN zy|`MzpCaMSlMPs#5Yo~6K(McJ5Yty}GRUAMe}){i(s-w~dgPby0k>$W7NCo@)GF<@ z!9yM3SgSg$968(gmv6glK(r-K>nKN*<`tfGw>F@@qt>j{8(1E=xPl$%zgpar0?viT zCl~(dvV(097Q28`k1KPmKL-ahpU2c%7?*F03W&yR^+39%6+j!*oougQ%1GPBVp?eu z;a`~(C`}=ZVJL_Wx7`}!VRmmN9oP3JUS3Se(2p7F+=aC230X}+H6&U@MNWz^o1^89 zBGYeom7%Xbu!BC_R?j^+OAAOk$59aB#_o?WE%jUH-9L^wG!^vc*|FKkmFap>TJ*fB z$1RUwq|OPw8OTJWzjIdhz3eQQUzUG`)Hno_=Ci;Qy!xHAG~Z)h+jn2ltGuZ|-=AMz zZ<29XD7zBN&}}wXxS*ard@Q{lDAhno>5%8P9_S&9&sl1bB#Tqx8&hlmvRW~hrOQ#0RSm_6CP-)|{y+#x@*UZXNq%xfv8 z0F7*GJE&pY`v{_q&g~^%qwhCcm`<2DpcmiV*QuXKZU7|k>6s6E{tG?B(ey@JRR#3Swr2PNUE5Im{MIgJ(=bA8i3H4OMx~^9b+v9>lu4; z9rJS#K`5&ym8z$g*mmBa*@VerG5h<&!`U z)aE(TrEblIS{CLiOTVZ*R6n5BUH{Cj>V4HtjN46BXtqFZuAka$7atoCrq#SYXnIO5 z%SN(&7@@qlBLw2i)SB8qUZr8>;JzyXu;ve58G_4M6R-FeWKUz5@>!=dxL-WTE<5i! z7v7EwT%sxT*7a=dL*be6y4CAZ+Ou4oL?OB^}Se)}*%rb7v(cyb~YddMZEOIXA z*}spdAS)iEbU}P?_5}gFbUf~!LU@Qwu?lIN#cu)^r&?ki>zRhPo{ng79GD|8AC9Jf zBqJ^uBoB$uU1+zQA2MLt9G>iJu$IhDx}INg5F9Q#_#(7Ha-ABFqW$vH0W{3=v{$_M zUSaN_7x%)U{^W|3%4URCqQ;fuqvp>QOk*a4ou?#pEwvPUlNLkEm$>5)E*pW?ULR$A2S!zGhBl1xTeSpZ zBHHoar8dUAVU?M#cw7|uIp6zc9TNTa`@%jt)czCZanVBMv0@Dz0~vD8@r~EW5@Utz z;k~b7c8Z9q19-N?aky3KKi977o`#d7V&=kuh`shQYyuw7Qf%{s^of0FZb10wOOk6+ zdFZi_djsrCmfF=1zDf&eZrJAVY$}c?dZ@X6f;`Uo#QuYpXk%5Xj0Hj(C*Vn7Tk2G*)W5P9Zg2;Tx!49JH#?C}ttly%!CK}E zry6h>zsa#jhe!i-5maI(Q=R@1Y>YEuES9trLc10VZ=#cQ38OisURd%V=}vXEc4eo` z2gYf>b}h}kq@2JV>*;!jqUliDk1JYtzkB~oRx8n|NhMa9SnPj#pe4t_W0+>72%+Y*JyX;|Z!K%rxZrye&`eCkV!t9u{ zGKy*3X=U0=kEcY%PC3yXNi1!neYx0kg%h`iygtZVKs#a@>XarM+6^USR&J8W$euW5 ziXtg;NC7fcM2kf!&)OA$+;@3QO@S1lf*l7A2|*_3>q zD;BN_(^!_j6UROFyoQ1sBqeQ@94+~+we%`&Iw!AeetDf+2SQGNW$F%2UL@__81zS2 z{VYjM@?f#0!evJ6zN`TDsBh#np2F@|Mw_+849Id}fN5BqzPeGEN4!Qk3?0$mSNzP5KAkw^$ls=2~pBinQ6ja(Rt^=pJ^2JIUR0r>e4O z*;~Sf2A)B8r-_z>!ylF3*LfzNZgg$J-su;_g5S2u0zdp_A+LF! z_+g>QE$c5XZh#{{DSYKJ6|PBZ{>#9CKEHXpMU*?}+3Tayg9rHi2POdur+p1TUJV_$ zLYaHi;Qr0U!JkhVcNsG&zk?n>+;o@SYgu_as}gQRTG2|$t5kp5Vx4u^ssaiAQgkX+ z^ebU+)d!Wh=Y4Aqo$&rI$OrAs{gGtw4b?j{@*ivTq+hh#?|2y>Xs3wEE*i7j!WwP# zKr3!W;F(fnYRUA6myHDg1lM7QHpc%+87tt991~yWHwv_(>z;?a#70m)O5H+6TC#VF z*a@^-8d|Wc5^;AwA47OcEcdJT`1iIZ@cl^~G~630_Njfih6H|@otVz{=lL+-QteZ4i;v}LpK)MU5t@7P6 zIk8MvP$7jpwQcOZXo2`cy`iI>sqWG%D|toG+x>oSp1EiX29JLlkQ2M(P;6H^?=Wwk zLClLwa!>SfF^t@>Q~lmX$a#}ue1Bq(7x+2uG812lzGg5q6>it zG|kA3&o=NYF=aT*b(*NBv6Q4JMX!m?orAs>yuu1&D!6|>rW)u+c^>xD9J*&FhS1dB zIJZGCiNDa+n$(tBcLPR<%KP#ka_8<$UuOAT0#Yikr{*K*5yu5^A#fjk!UvVNA-=UD zV2=C_q6ePjZKq15k+RJT5xEnsAIV~qTqB-Ih4nHRm<0Psuzn@~sb<3HoIp_^zod}a zLsXI^x6l)#=CBI9V*H`6-^F@F01eBSBajX*VS%Bd=l~i!&~eJ<*;ZUXz*nix*pZo0 zb9uHosGyj?pgvA3fjBUcO9>meKIE+mUOb6*aj~wBP9>1n;)0OyzPVlcdo&$|C_>z&hmk?QfkHAXG{i=LYDT|{!-#|AeyK*aLpc_|2 zUamHYe&^U~x&OygLO7F1hwKh=$xl{e5En$v7QABI=ejj^g8BAZKDj;K+s`hSz2fbl z5Ar%wvrSR?Q*TW>m#$}szJkXQYDRe{+F`GJ>T}wfK9y7nUgGgA@{y``?nK>hmFsygkAkKQ*yV>$cmI z9V~0T05lYX6u#17(F?H%-Fl73CiMKq&0#~dKMNcs$3@?<#~K9@wX=EUe=3e=Hqr6F z$>q_^Vj-lLOG~kSY^S){_wb}h(~WzKR>3Ay8Fj7USCX#|$-2nf`=(D;?jM@IJ#L40 zUkL&n2kx&_oyM{}7^SGxP)t6M3)KQ}m(3ySq?epc&eDAbNJ^miZsdF(;^HVDuhVI-q8EKASPoNvHscy z&w;LrzjN6ros)lj`ZddQU&~?%9=tcf4D=Q8tp2ISySe)0B(VJ)gB?co_TCa(78iQVDx+ zI0FG);jv6YlPccA-Z@TXmDN6e?v$QpMG8PpZdp@Sp-}DUZ^;65Xv(eU%k_x1E< zCcCV<8Mh;)x=fAqmGhcR4eZw7J+=&fu&i+XteJdM9&L->N8zv;kr$GeShTc>)f#>E z#`~|+I0|0jE`%!A|B^o@hD{?nr_R4jK02y-mu{Jm=@fT10~N-U@nK#_T*tkF0`1x> zW;J}t>!mZ-1$(R%_6{f9e`3ieFJ1jQKp45#e^_5e81K+qq9u-gk;sYACzyEd|6(YpS~&O6a+AFQbFp34_}0$+b=MYi4pPpZ(=x zI->i3bOxauvt&l2D-LzV<$OwUX+?avP|=zPqG!bqI>qn-C`wRyvj}J{lyovTwz;)M z!9Gx|tleaEJg|95b*n%++Bu}h%FD5I)gfb%drc88cW;IDA#&LU3@seb8Vl~-Ym-zn z1#(723QFL!lwE8b%>yn_oGP(;guh=h%Ypts=D5Jak4ye`5_SY5?bP(cs^V2&AYZ!z zf(NLVI<`BIUEH&Whse&n4{sd4#mS_QWw*k2(kdjblwY3uO4u`t|BK-D2o3Lf%t3zv-V8J&D4^*-^uc!6%)G%yYR z2H;;@B39AyIRb+ie^ml%4Qu|gYVv#(4=X|*mt0Fbtm44oG7r3WYM$YcQN)vN>``-C z-f%Ty@VlhZRA-P|u^=R6_#R}l-pt9{ElNcVMjf3A#?LzoG~i2*k0j*lOx`_#n0Rn2 zP4}%xA&Ve1)w5fbC;Fv zMcUJ*6Xuc4C7crok&7gIiwj}+iP$hocIp}X{f4S~rBbEo#r6r8q3p`&kQ)Hx-6J6% z>y=z`uZ!m=2Qn0~{h9NiNxvQj8Qm&>xf8zyj-LM7l7w8$&osi?F?+|0f9$*lF4Na| zl5A4zt{hcRLx;^i*H~L&(w*(@I^XD)5HYLMGFP>5y{NFth%CVJ3wh(<=@G>y5D*bP zv<@KQrmQ+WgbuFQ0<(Usyt6csMJD+=Ed*UZgwo<24fC{|FUdqEW}<~eIB2)N>g%aj z7n2@as&w%a1o!f7V63H8%{0<&Z6otRaLOiM=>^T>oxGY4i=A^&tLc=rorZO_ERe=S z7u9H#)?kj&-b~XJWC!t5FFx9eHA8i`TyD^691Y-tM|kdGPL?zA3k)N&O1&aMd9>xS zK}7~Y{bn1PuUmwXo8l>Z;~kkV)WNVUS&ft`@h-cbRfh;e*@2voMHxQYZ9%Gn>(nwO zmX)i(xUdk9>$svz{aGo}$#3d(3tUeehbo-lg~mfNKn?zP&4&F;icbU-(Rm;kZU6Lr zq-1{U1DluwPahUw(_xWV&$jeUG2QfXb;%V(^x~?1ebOnuYm_xc$wAR#1e8Zfau@KJ zOAm#-VgO-IE!l9MRlFxIKC1uSn08!r{bI{dIJ>Kq-)MSfE zQG%NW$0{GA_BXwPC4hmg845Y76v~ffFRqVGPeGOLxZZ1JAhbW=4^5ScO5P1+p;xZB*rvck+5 znff5wy&1r+1E9}v)41iK0$~Zh1MM+4iilk61sP7)tNClu9gTX*{+?q=|D%r2;;;wg6Fbskfrj|$@Ta{0EVT6t;9nyJG z+a4l&&Psy*r8^!%f{3!2;E)~Y6MPcoBhlq1j6R*Epzgjf@Av^~Udi}nhj+m{Ui6M! zYk*Pc{^@RHc|F6JF&oZOFMO^kP9?URTBV9fCR55%E_5PcOd;ZgVQOrq&1kwQd5%n$ z^UM3T@7pc|Buq_Jgj}Lz)cJaHdJ)!UGWR76fzBd9$zM_7jl)i_6AL}?Qk`2HA?C&doHtwk-)Y~R0% zpul1@M$~d(*wIx5=5G0sZ}AGNob0u?$7}m+4@`Yd3CFXR>}P6B>-W}i*ZXv#?^Lul z+yVw(TSf6h^qrX zguzEJpoNl94K=7^tMf^YTr#vGE{2`2`z@s}6<9 z-Sd=Rx#yNa4}gy)(`G&^7k>0XAPIXJXpdnuhMN@}JC>vY(>@r`kv9M_)>Y9h=1A<= zjo05`;GCAJ=)YNX4sUq{9n?0D`GQL#yJLSCz`9$u+bcK-{er zp9!i;+m`ZMhkK>#EFs5RGAB-cC<>)|RHo?j2gU{v>+UkamF{%MzDxlPf!fD_?)!*F zgfrWmtX4WB;AzpAM(rnZa@Y&0zqN`NBhc0}rrR`Z6Q3j)?r8R>Qia|=9oAbXJ-h*A zYJeEs{un(Jf@z~c7ot4c{xnPLey8^&lcpuzN8uM5ey8GYrw4i#CCc5Y;zS06x!+d* zL0s+oAq4bY3SI1jMsTe)29f7GA6U|oh6QU}+@9z|eDy@DbIGF0lp!uk3gpdzC6m(V zhT^2P-uA^JTQC8~z!*(MaP8k8Z65are1y-GRCs}|SwGK?#xwuN_6*W}gEkR#-_g}H z{J1#7{sSerg%_p@;XyP!@e{;!ZbwdsQB^W!C#AJ5twkWFgL{0H?B5ujzS&bOAE66y zw$-ep8juL0>v-N$mn9E)J?i+omSLY`3*IY3HCXcSu(H8eWkqk6(!jbZ z&_LlLeiRV6cMgZ?fckXF!lE4t?d)GBsAnOsS-%TGc1UjdMM1f)R^%X{;^1idgZeR9 zpWT$bbC4q#Q`AnH_{=E2`(7O=C3koS!$n#=Wv7Img&p0h25EQiRn`0W0e~Uefn0Hv znYXRwEv0wG1xaVC(*JGyuC}TrabFkla->UDs0TybEF&ZhiP!8p1qzY0w z%Q@#92htbFJ88a>Zi&wX=P0HKCjVHp`6jP)Xvbd`>moB@^CIdIyPr3JBT0W=ze-tk zvKvFpV=wiP-SpYneAULQ2Dr9>|C9t0*fay{pEYCZ2ByA#qA8b}yBpOJiu^d?X2{ZA&Y$iV8EMsVsohhL+F14EZ z6|@#GaCe9)-~6F$*H~pFHulm@GMJXnt}Tk-frY=~uHJN0<5z|zd3UPhtn>aUn$}ni zw1L1;u(qFTzQE5JpfiOlb$55ahPwW|*ev-Vz=6i@7D^L?v#3B+3b>lEp0CT;lR+WM zK=%l1#PN5fU}qBth>}4bLHpK>GVpu?_d0(h2DoONR0gyIQ`sT(OR4)Gg&!!^e@jL{ zM7#GjY`b1%8Hb(KE6vzTv|duLAXarB`F3fa1LngCeid^ESwZhVq@^)~$*?c25V_tJq+ifo2-EEMb%tX1+d zD~nWY=;@&EZxc%65fYI+`IL`-#o&erCWXKwbhM`J!%^?Uwkt1U<13jU zcjriAj|}2e&zLLTvtyaG%nPAsS=TpU9p2;6Euk;Tv!v`Z5 zGgaWjzYk9M&+g}mw=vpF3mM`aGUiO|MvD~;j(fVUB6Uqpwkl61Z8j_gcl)`Qj!aLt z0^>F}n(omiA>7hNW9#*RKJzzy1Z0hbowZzLmEq|QKCA35G7PTPZ*)Rujq{3x0?Ezu z#=9^R%P9V9oW=j_#{Ctcy|sOE78>gm1RHQM>lI<@k8=$u-%}mFIJ;t=E+31>mN-pIuYY+wn>6n92Ol zE>LZDx(v(hlOAZ%yek8#B5=Z>9lHzXZDLTZcNVNzoZ`cxwhlXFp;~JO9qX@jJ%A7ev{1B-3lmc10f6_}wUn2r|y+Y2K zYZ=UedwQyeD!gI3xuz8PNF4F}?A8Klq+*T%J22bm3peQ+-}}P>GwOhzUB)(IL9GdF zOj7*Qhf$CJ^Hcw^3wMos^KVzJ7Xu8n#NUK@+pVY(xNUjXAUBt1>c7~Xs+$-q6Ow0r z+W02eGNrO5dYyip#rTWq>$7tZp=UehQZSUJyog40qdq|fIjoR@k-u<%@fjCQiM(&@?v!BY<8~~rGMkIC0fSnjqMv$P+t@!+8)&%NQE2PW0q_T(ou$Jv zmyY?`i!VK_cxQ*%798IoL{5!m8+>U{)%VF)mR7d3wRx8%-IPuC2?GgQPgjqq|9R0s zl$%x6zxid0%}CbBdunqJgnRgtp$g7fS(W%gVzhvT&+@fmXxKh(w}-#60Ps=YL;*3G zjrl49mIetW!UmTzVHwc)g+RI^=Yn32@YjG9qIlm z!G`!az*f&_^22yg)ddd^{E7>s(ppeUxhlOx*pGyAW-?7rVP`P=7YkwDzOM!d!YQK) zUiu3baW}aw{mQ{?3oNq(@_y#a6w2F1_=SBBZieJ2! zQM6xMOK;JVKdW!DgqZ2xlW*9)XHw7)s(MCcEdxF&fc zdh1(m@hWg5ObkVi_YA^_vNy1Qp95%aGWNgz>F*7?TQHb!#EDxB{YZWHxyn({plALZs%FP8dqaA(-|_V!!BlD)f=ijQNX(`m-`MXP=@ z#FPrKP&o?2-u3V2akrpf?;|3biuQ_OzQt$$)*;p=(NH+*?ZaR5Wve-Wk+2WFd{%=l z8aMkLEudfNn;hvoN&t0{q}jxpcA;6MWE$C%3IAi-zO z`4n63sD;61`zcb2C>NIBGHps0IE-_4S=t5KUCf@?2l&v!SP_3krZ);*NR{ri7AB?FW0BF>7#yRTFfZX(3+u7T`fMjc* znTy{&jADjy*dieIk!#1zD`%XrEVeXxV6zgnS2e#4iL4N3`2Sp$Iyd{pq;EpDPrtTT z!B#-#7xvW`$H>T&yGZm(D>`H*x;|(49UypPjhiv4MdY+#_Pvdvdt8Wy-{9uFQnYH-r}>fsM2bl1a@V1|jJS zHFNG?`@K!~RX|vj+F68O!0FYAH=a3E(zjQXXAf-P4AGD!@Z|A8#IC!=6m(G5q$xRK z@?|9qs*!+6*EkccPlWmZa9I>Ur%^oTJouOOjZ3NzsvrSg@pv@I65i$k zGfq(boz4Dne|($cNYBOv>(ttA$-|D}s~c}66QrfoXvNWV2I2oyzxVD|NOe~-)Crd; zi%cVerlfeE7qrBGDDr#L^C5trk`QVA1@tE*wQib>dw`h6{dG(dyNrsSZ%-4bxz>gr$W zz&D}4PHbo8Dv^3*b~Id&ukv))Tcb1v`TLHwL;&ISEGt#-@!vN~eczqzL{0X1IXm@Z zl%(Xnl3)DoXDG~jz#(~l=Iv*|)Y1Y!pdriOb@tV}?|TB}xDPE%SW=u4@z*VuS5r1J z?loiEd#3)LF^ZQD+}z_F|B?&Fu=1s-;NhBFLs_z@m%x`SFGVZYbEF9=F(Our2;`i$Cv_sO3A90 zK-S6~!m%#ms?xLe?a+H0cI7n8qhUM^ptGd%q6o%IxEB_&h{}Vx%RjZ+r_`0T0Z?CIc%kWNG^&!zwh5q3HDScp_&qYsWfo^$!yN~CK=-n$_?Zw~O zSnHq5|3fE# zge>q`_1EZzF;8Y0jlM1gS9f`I6yoUP=n=~ikU6>xeaN=C^XI+;0bPaVVAJaN*H;rz z|A{VN-b*^CB25T$85GS*6Ol-#_HG$fA_+@@knCtr|7L*aj=)-P)(}ygeO@B+Guvb=iQ9QP-P;~HdO-iMw(gJqd<6_urtJ-T!T5KYbiZ>aU93n!jE;o2 zB-O-cmEYu*g>tkUH?91#z7Mg=yYr{VJDU?dk?j|CfjGNaAYi2=?7ZE479&{tuvR`- z7|dZR!RH^Ra<{MOJyC8l_POqL{{g5j+GAEG!3(yfaa(`w%YMYzK!Zxa9=|n1&W|Z@1rzX)tDi((Afl=pv7Hw@>x*5btrDn$<(auOZTTby zdC8n4J3bdnCoEY1ruq6WfZcgt@om7%zmxoPbN988emsTYp5CQ@N}j4%cS_`Z@*B++ zW7u4{asBs;smB5ns{MLC?+sp{M{+QaH%DiS^AJag6_{-Y>Yex7%1+M9 z12$c~UcFrl*Fzh;aOs7F|J-J76mY+~kN7SS1->Tk_>4gGPoY=HL~AX@-#f1X^Pjh) zt4eIX(NoQ5Sd1xvzmIvrNjSy@qoqV>n8coZwpTviQDPPF;3xP!34dKkKwS>Yc_}<~ zf2@i{r`^R8WZ}G~0(t>^lal<7+VA&+xW7JA*jN5eG}V!W_J=dO97k!es18LHFy5Hx zD`9REQ0&lonDI>5R?`7{IlcRLBM7~|yjm~Y^ovpiOq*|Rj8yebT6>G$Zr2~#`e;2G zcxxj9*9$bzeGfETUL@(9T9C?$KH}44@$~W{we+(7%~K5fr{{N~o{o+l;3?W){J&7! zpUq)BHeY32)M|qCx0Wc|Tj~(ALZAN2Wdowc;sCj@?W}*j;976q7%ncPsI(UsoW0i; z557wnCm%H=eD2Z$w;^!$#@UqZf7H;QFaO)7e6P6o?|xxs$nfifH8DPDfRBMloJ{os zU^kzL|9KcwU?ZhE7Qq*6q!3%#)y&G-?WNYN7IB^uAD@$qkS^Jur@ewU0`RiUn#^-O zBm$9?oJ?-TM?o>N#vZFW(Ajyrt*uR3MTLZSr-fMgUwi+rkE!EfWr_jRcxVq;~j3NR^>nvIm*SKYcZXg!8Pkx?;!`c z)Y>3ZxrC{Ey!$X20-@a<)Cx$?$WU}^mi)cMe|`HybtK;Y>c`amoeJ@&j@|Z?=E?Hf z=6^FfZqF+Mu9IHONBxqMU1V}z-V0z-2_C*pLJ|M*dM&#_*;?h%WF4W}a;7{MN!iluxU!LdQu<)Po z(g%SeK;ad2e1W$o@be>($f+;n{_{QZ21fmvU!B$};LOI>)mN^+*3p^q)O%lUZ^5(G z%KXA%b9&IM>Vz$J?_cZu*9E9QyWzYW?OCj2AT~=L=k(s8+^MSj->fC{e1JysUEUPJ zQc@^PD|Y(Z4TXhC9kdf>H>NQD*n}8$ZxB3te|tgRILB#Z>B)#= z+uzmJeX-xHCLMF|wuch0fkIC~Sj0k?{eMWS0??y7Xe_~PFc{2YK*i4H+NbVBDK&qj zX&HTVbo3C*%Y~ff)dYtE}tG4ggIcoMd!65Hp7L$-gigEpK;VVP_XMFxb=K?-6~o&~LoZ zA&AS%(%e#JDtg>0h+AjX!8@RMizdtaUw-$0e*26BeDU26TEJ6V3Ilq4?aLVy7F`Bm z|M7WofZWU^$h()K^y0tP`49?=i;L^^B?CYLYkB788Os+2-QOd5oowag zWyAuK{kdqpeaiZ;tK@5o@a*);=GEeGTdP)0Xjc9A7XJJT5ID1jHq#Z*;Lw~v+tqeSP zC(1uCFwhWZHyU;NP#CDH*BficB|1saX$h|jk>q;-WM+MU0Xn9=lL~)l#CiPrx2dix zN&Ut9{`K~G*+=VL69t<<%t!4=W^djF%uMy$8piVkctImm5MJ;W4pz$36q*Y+04997 z$SEk;d44^ceS^!lug=24^7GS;duLR#T&}vFuJnfLZ)(;31w-!f=iU9K7Ng{>qjKrF zo;mL)47_QSMNkn}i$B+edD9zEAc@a$P>&4u=U28vUK?{0NwLt`4DOv6>b_AZdy~MUh}b?ifglxoIIk%#-Wr#Ec~aeBriVqYOz*p zG%HlhNs9d|amUtwj_N4`5TRQi zg-`HG<>59;P>gP?5DphMGBPSGEc_WtBkO5N(oi>u%d;-PgF4>IYhft+DZ#zcd(gTy z*Rf{JHytRLl$5kYTDL%80!)bW9$XO|V2I^;`ZqcFoDGB2v55Ed)Em(_sFAf* z!|!M<6q~Q$YW&2W_Kya@#7ehk`M9qLm$(Wx&;h=Y)V}z7p`$GTn8rk<>bQ_{3qPQH z7q|%W;OFm`0Tk|;3~Xau7>+jqCc)(a_9lK1?dQFOff^XL_q_L_Ij%6|MC?RB>@x3Me|{lgbx9|cNm?*8{iAe zz^A0aU<7B7bxSEAPVBDvLZHy8L=T+BOy5wpY$Z4Y0fX}W5@m+jwix*1z|d<^-BR6W zlI($|XzjZ{d7uBLq-uS<_xsvPK>S`MN49EwznQ~A;@{(ex9>*+>{31KIRmjSgR#60 zJ)>P18z8;X)6?gNRkJE9`MiC6xZ!e2YAgCD-}|qJckcrKGe1B7*1+`ELA!v=LtV>1 z6#3r<&KsG;$`W#h(kxE@NYLHu?TtGIv@r2IbH|x%{k?i-dH;{GtB$HFjrJpAAW|aT zBA}!o-GX$taJi&3mym7|1O)_XX^`&j&MV#B-CVlk9cBh(obkQ47JnRAi*vs7eY<{p z|MouCOSDIabrHt70~GR6SO0$dvbI4yW(E#1NiA|b+U{r(GDwWyeM7_tn9=l%j5TXY zDbB6ez&F00gh!}9rw{k|N-q8l)nc(@{UM!|gMZ-f{fg#ez-}#+tZg)PaF{44;0twj zKy&lOv^x5t`GsT7wfV>P^~DybDda2i8b{{llHIPhRAcEfqU^Ai7LqI$H8M8#VRQ^q z<&A<1g&){AAtq9Eb$qKvs2D)|FfvHr0Punr)+wLqt$UZyx zVuS@0d$Y1(JuW-wSkdx0LhHKTh3(+$n*2i`w&>RI#`U{e(G@!SDUh*BuGwQl*n`}s zEgeY8QHv%@kD%nY5-O80phDRhejaZfiZW(%l#dQin~UDGGW7hNbwtIc2IAD(07m5B zX+03i#=u|Y1fSw@KU|?N&0}?br)$+0gtC%z_c~Z#dIgxId)UdpH>>a-f>YYn2A@u% z4RLmBt4~AHSZ#Gpx}Z$NR}bes02E}roK@`KFL?Er7w~X4+4|mo7pCK}evUJQiRQJr zq6A3n8Mm$;j04;Qhc*04qW(i5?n)7ltE}_$WP1Z;WW0)3f&NMI$)F4c0Ud+xSYM+T zSEk3+7}y+RqRjue#ns*2o|>)C3>>!X?YS|I7}YzjLln$w9)dC|$!A!uY~m&mCL#Id zm;YW>nhGy7A(4B%WpBWv+$qSW!{yZH($fJeS{@lTWU6!kS#`_v! z;I+TFmpS|%x;OdP*qA`Ps9;7djx+fvwB85RStm5Tm|UY1IJ{+c1{M`#o1(VMeL&1_ zin3bevhd|+;PdbjEXK;C_czx4fFI?$Ka^{YTSP!<#+VsKzY|q1;Pw9n0*JAm1IrBd1v)6#^@u^~1 z{JxX|0LUE|1tOG8<95e@&gxVKW{HlSk>ftf zURY-ikbgVePyI)<`ioX|0h)0tPaw(efN= z&pX$Wt^)&&$CtYDL4ZNvw>ZzdHaS5ewqkKykPaHRn>VT6x2^!Kc! zrt}&Gl%fWgiLnz*BO;zE2FY-3@4~HPo0-3peXhBH;4)ed4g~%_rR@QHCNy+(*6l(L zO9BF{hoC}E#PKMog(Sk$_4-T%C4q{>+}ik69q3>3Li~dA4tqdG_vC3FcSwd|d!&nm z9T1Wz_tWRrVH$-=|16Zq19u9djWqcEYYB>8rz0!RuqIS-JU5KJ3)SMTtHM6AfSr*k zGfR^A)d&2PUj$c*!Ka`Mup%C0eb|P_*~GwcyJUM!Qsw766<#Wm8h7CD73pCNFD^6t z0hZO)(n?WVgb7#yKQvMSBRCO=N8GQ?9`-z-PVtVnBz~XLe4z@5H97n^l*V{v&syC@ zNTF?%d(eo7;HOp>1B+QwNBwPNrLiv!L@u)K{}Ld+ zah>D>D&i%jBo@v;DVoN)Aqa_AYWczg2tu2>wCo@MRuyVA5E+Xs8^S^d8pF%>+u}Oe zOQ^Xw*+D;9Oh9&n9yo6LqF6%DYaj6I#fxunp=14zLr>M^7QU8CHa7~06vnImBk`ks z$%gWwNrC>pbp9};Cj8i!bv9r`-*A1&{0G21>7W^~9uZwh5qgSrrI+-SfKCZw4QT&f zr;6mKB6cEXic#JwaR6EnLRo%=n*SeFFO47Y(e|la6Bqd-`m3cwGUAuzv(8lCZO87f z?e#U+_VMHY?Uc~a4|j?1W{QMB46q9uA;kuc{pOq5>lfBQ9Amk7F} z`aAQ^hyg$k3CK^6PhicD_?v-?A)9cLm?J|#h#m7MhTEvxqah<#te|nW4bTuai7J`g9O#1r>@c7AQ#0h9-)Wz_}n*Cn|mvrl(&k5I3 zFL^2T&3q2xzms~6xA2$6TNlR?HbUDZF7^N~K7Mn@rLH`hEMED^F*H5$eqS|{AJKDu z9T~n+MY8Mq&jz}`UCCcJIk3bL#Wu%lOLlEV{3kRr`Hrrg-NVcHYy6@Sxf4yOX)03O-v|EdL3uGw_9fG zJwtSF1UWCNE{|c*-r=`g1r1bqwG7y&h>)k6@d;(!0Sf%p!746EoEGV-jwgL-DYUDj z3V#w34`W8!I{N1Y%#~%=Q`m=lZe*Y}DgnD~k8n-6`c>v`!QsGtp&>9|-7ZSVlMx%s z|4!y%M^We#W(5h)hln`N5@5(E_Bco{p;u@7l@qoiOZ}BD=jrz?6Xm!f0ANo7_ITFe zjEn0!;QZy{D@j*3n)!iOQ3-ddvE}G?h`zq0q)&kMCE;NNLHC)lL`+YEBdU&)lm0A3CmnxAr<`>k7d!4#qPS8B*DW$YhN}ifdw;8D zIB-U|>@y$IakYkbhCzNh54*K{5iG=Je<+j5_)k8&)29tv^U&!mhPsT8d>vozA7I%~ zz`d}vv|CIT=q!R_Wo7yW8=YV#jaZ4&I&Vg&agPh#c>E<`Iv*HA|7{R5e!q`Nz;SNS zYHJ7)GG4HMAz$`#n7i3pkD{%a!Zm&@7T%XRx^e!18G3-lXyTX#os{Cq;cN-~F2GZo zZ+bR}vfB@IJGj1lcY|mK&@9jaPbR_YcpawvdeIcmYXS6J0KwtbWM19(PmUO@D9F6h@F2|K5z z`m~B6^vvr#60$NXJh)JXJbVxg}G>~I_Di~0EXg;$l+5OIOp=)aTANE*C|V5rq;W zk*q8T_++3|32LI<=SccB3q`(%OTkFF`iMnd9Qsg4$C<nqsqxDEjx4>7_17hi~ zz?1HjQj`D9Hv2b{dRSPKXu|^agXypJLdX>`p$HbDvk9eSxxI?@<-sa3*L(WHp}mUX7E^8 zb`47W2<6a{*Vx$k*ysjq8UUg|qlMAv$*yrP#llP-z*ya9E>c`Y|0a2T{|Ep9LF%S& z|6&-fBlQ;pcTAiLn79w1$4g?2i`kRXgvsmb>}kT$cBPds9e3aH3oTSg0mUC8t~BZv zQRB=TfgEqF5Fk{923J(u>xW)nd?fyALZYa^%=%LK@BD4dT@I`2yu;D@yUR^ZWwTRO zD%*~i%(mkc(+yniG-?_z7tu6Ufn;KfD2fgGS$8a9a*(a9r*HdBCOgo*pTMmK6Hs3WT8#I#(f>7ksD1Qi z3m`QfA0Tbbg%`Dv6lYQTA_9b)%47_^eIvv5Pp0Ro&$FCL*zzZjn=TpD<2DL=RDA{K zDmUDq&!xm{m8?YneDXil`SG_J7?I)UYzk2N-XiPb-n@Cot8r6;ecQakk>raZRH=hU zS}HEpnE@|)4<^7Eqw;=HqC%@N?Ki6K{Fh!tKCkH-|NLrB?J+I`A-eG{Y{Zbl-I3yQ zWCtm_&z;e#Xn|Dfh!4Sg8e*!JV%m&kqeXl3DYSH6rj_}W)+3bVA7s^=dVa!<8rw@+ zkW~7*@^3bc_Z9+YSs4xODL;RP>S&cfe)t=i$1&`=3=75}eD$&-`VHF(7)cV){o}}m zK+rCcWpzXOyLf=DI=5TCGJwCn#r=mLYA{?3KEuX)sx$lVP8Wea4riVRoBxm`19x!T zfM~ohz@Am&E#LUlKll7$Kn;)pZNdQDR^B4$ow0?~IXSt$!9k$nZ2%OTUz?kEI!M@9TD}H>KntfmtNBeoD&G8#)v+X?0oMLA zh1-w90*z14pHx0z?DG{-`0nChAs`=$*%+E_>hn>mXt%FWYSF00duA4@*nnr9Tc{FK zbA#XA3B)UPupI0HpKgJP%pS8-%$jFGtaC2D?>x2T73vklmj5l4A)4}-Q z{Am+~Zbu|e`9|4lQ}XHW`ZbD#iI3EbA0Zohr(-ceA+4bARvLEk9c+<#Qp%AwLrDss zvkt|?PB)91^i5V>`V%UXLmNI)C^ZZOr{P1opOk;QKCK!-;3V-hACF{TRo!qI!x?l$ z@<)k zo*i%;Z>B;In4sAk5a0g7_1lD`l{uFT5!S3H$g6RbGAdllJA;f(UQA9>Oioi@w^>(f zdT7E!Z<>rH00sB;ClGfeAX;3U9oU$77G0d5K!M^ob6a{=7VQSW_3K#gvmS*M3-iCm zBl6=px20CEs@QQH+4z@$|1D+w&o{jR9$Iyv=xdPuTJAWN!sjGKJtZe)-6h z-WJHT-ZN`$-Q)+Q5W9CUJUlJ1_F(3qZ+MiS3wIJH<@Oo2Szx!*jhBpN0Tg_~T`

  • 1(Hj66+G!_g5E~qMD66caMUAyjzrG(+jZSf zsEv`j&}`)sx~$9*1-|d%?zOlbJ_Gb}9o7|oF~8`giTX3PI|5MuXA9B0jM3gVZnjho zXuxFVgS5r(0^VBlEy46o$GhJE&JDSRSg|Ik+!}&!R&$Hqe5Rp;X1E8rbgbq)5fWH? z^QHdbt3KB_NfkFjv(vl>iq0*>Q)zwUZCat-e^x4KTx`buA9I#*4sSqS;kvV@J}F9< znNPV$pLuYF2X4Wg0hRBS>ful7mvU`0W-i3Y#00sI1>sBuh&h@0kdyUdzmbuV4pdp8 zZBsA?1_r%$lh2T*Cjs-u#_yVByXahnl!_aVl=_6UjRt8t>~T|)DxM+71p8?N4Ivfl z7XBvL3pb8Q!}2nPP`7s_D@7Cbt3x)EPAqmG7&RDO&QFD1l7wWoO-+Exdeo}~#+)ER zU~!N|Op_?0(-;}DO&6fUaaETRg)JtTU24soLQXas@^#M8E zGWnu2NaeQ0Y)$8nIe%6)T>+n?;{xBwePYZB(86S7Wgp!max~LR(VrC(t`{(PVK821 z6&3uY$4r(x#&7pj+m8$Lve2w|04%L&Z|{u@Sp8&`a9Brn5fb8sYbts1Emb_YrOn6Lznqu=R3wvOev&d1EP92~cijN&U^FWpGu z-GbA;WTUKB>%FD1l;P(mGE5~WO#rzhD%hxiK=A-8ElE5qfGMk}9p2E$X!eo{wj1$M zPZDqsPWgmRC*R21dp2dt&PL)q2}g>(cX7>K3oN%B=rY$Mp?@W9R?gJdgdBVRGm7PX zb-!W^+x%=Mq{Ldr0tg&3r7T1zc{d83xFx>@`T?pVV7@(x+;O4cMD0*@u$nB@)@U;c z#z1d9v81}D9RMLSb|V}mP=jd{s{?N>@?5ZmM?`1?oVG7CBGANJNr1Pa^78VUS&H`~ zyd!1mm0IIMqwl3YKHB%$d;$$)tHWtWoyWYFQv*bx}3XX^d`3`csMF?*d&k`qFIoTyhH7<#E~5va|gNY{uwy z+rxwH0xe(lnsv`L2WkV<74UJ=XyG`(F{C84@_)XDZdz^&&TuL$8LH(7Z3vxIRUSfT z=D2a=>t)(wq(C+N4Y3Uv3ij?BRM5)G3g2KUgiULY2{dPS0EeZ%E2 zgO%#k;mzTnu5auUdnjzou2l81?yIEaLRug#NMKVME2TRb|6 z5vdoqYu zOYPHhxugbwH8ftzrDEDEK}DJfN-Zodwz(S&aFoiC1_8M!7?Hu%7ln=jydrx%Nrzka zFY^s}RqF-L?CK661^gGJ#pm%cHx__2CMJEGBiPPEOqi7@I#2rO#;6Rms1HQ++Uz6C5EjThER zL*%fQD|{j12kNF6fOXjP-&FkdcN70f#d;t#6{j~ko0O2(eVUMV6n^ZFbPp-9 z1qi{KWb*UU8Ib1BzT61HYkpiNIwGAGqI##Kafw%2uW42ygO-}1G$2faI|+M2fb9)? zsn_I#^O9NqmCEFTp5_D9zjCapt#i7NKyO%#TK{~-f3in?Dr#E4psAN}r%GPfuKtip z3dIqR3GyCnr~W)jX!KdZ=!D4kc!?gvye0ZbZR7h5EajSeS%y$l+V5X9gh0M{cz8Uc zqVg^{_SfFM|Fki6hJZ(Pa5j3a`dnHxvI7(C&HMR^*HkNw0)yd*#)2*ym> zH+0w68nxR406NcB$QZ$&N2ic)LMW$FuPF8I{11T=Z)InS+YpL1Q&DYBMnA87G~;-75knDR0bLc6TR13S{a zwQ^I?BMO+mpSE?Y(B2>K_}ViG%#F6Aw~rfX@QQpnOe76XTel`8eEY)7e&!5Z`2e6u znb7|N3oiS?W$Tj7g+4e5n$R*+vhCJ?lahO_Q>er*3gC7CgEnYlxKu$v^gNN520!_V39vfiB*#`I5g{y|Y5q!v7;YK?P=9BqcoRa+&CPJjmI{85@@@`OJ4V z5Yj0b9^to6VSN3H4{E{dg_2TqmK(oe5(fuR+hx3s=cMBR9Ql1E1u<-E0OK zTx51*MebDWAiuK!Gk~qlL1g-cht}}(omQ2yb#7Bn$H;F!Hgj2}&pO5Mti0!L@{H#Fr0u_q+S9kJ+Z-(gBqm?edAg(rnhW?3Wx*FC5Bklht}8jCAq(`Vq*XnNGaq`6>p! zszhA=hpWsbXM-}N`6&zL8--D{Vx9rqd+Xo4*D1)!T<%Vn7ywP(UxS`k?c|0Ww&76) z20%&C=Z=oJF6RQ^p6tDbMJiz~5~vMi8(IEWcodt?TNXURe92%10`GjAtj`#uY;U7mY+2X;Fno&OA7M|FBfT-5>^Qpyn_g-mY1+) zsgFIPQr|ej6PvWvQ&{8R!m$LVVFL!+53I}*%b|9D9{x>*>m8^8Nxz5oKg3x254dHA z(!Z_yX;ltgZ^m;vFR4grG|#6n3WA$ricJUDi&>pSp=OIDjzpeKL$dEjt>N)Sxfc$} zQb49&+lCIXzhsxhsD%Ci6SWcCK{B!cB9-A0iqHJkEl@0})Frve+LhhLRouX@3OnyT zPN3RUGZGF%RL;D%|FJmiP&kN4#}d?rsj(Ag;^li@?J?G1?3ciBvN$1+22whHH%0z} z`-WD$Sh%cJpFZ;Qz32E&!a+4uo>`O%jjP(EwMv=sjnrviewsG4nJY*VHzT5gT@oVK z0II6Cu8Ais1^SCDdN%JS-sM!rx=y`b-`R)bQeU=4@7>!^ELY3*^3uv)e`j9fQ>FAm zy)cWE+eLz(^VE2H**~e}yBN5$I}XQ*Kv!JP$Ooqub>~>}BzYt!ySal4DKO`jRwtdBu1RpK zvDB^VlPsWB6WmEB{{VBi;X3}lzG%w0q<@~yn&QsH1%nioQ>uAK`yNBh%R|XuC8(~k z&j*`HvxfqMqVTI7h(-92riN@z9Pwk|eSIUMynCKH{%NvxMEWk<{RLV(UnoDqg*?`R{x9p^BDs^dfRVDowPaG)y4td_T;xC{Wc0zcB2AB}T)7VL+sJ({%F z(~7Yje#)MF;`5TiJ88u^eTCv_eEMCQ(SrAeS zJ24wGeF4k8B-+kxVdQpNf>M!^(sQqm7QmGbfHlxIl@Mx2Po$INcG;t%(|Tanvq-2& z2%9j~|3D3tw6}_g`UF)LN|R^X9fBL7zO>#v0LR(oe6lRMOyHQHY%&IfKF^OGu(Gw_ z*0nhWIG0PmP*K39X0kLKUkplom3eWN9zF+pU%A)!uZ$`-r|YGUmm=$TVrO>9lnHw4 zyB`AtV?dObY*3R&xddY!W zU-QH=MLW$Gh*6m{hq1&XSrPRnRp2v4j#jY5fL;yJ<85LBi^+8h^ASHIZ7WEpLTHja z7L;yMo$D#$cDbhtP|0I@>w7kCqXY%!sl<~ZP(*D{4wEEyRsjTgW#cTR5xU!wX5kS;H9Z@yqPo(b%l!)m3IbCRm=RBW>+}Al)`I`DIpt zhxAlc^61R0Qon!^U3O9nFa|(aHgle8a282N;M$+5;o~lsqM4X6wNIq$F%hT48xP9pj(586Wp(hdGKzy6A=QcD zNHIL#6mEI4K`FqLWa4Ou>QznagcACw75et-sJ{I*1?5kj0;mW%Zxuj}yDWR%U1LsG z-=i6OzP3ib8DrwJ&n{mH&vmBZ7F_%#gR+lic<9@v-$nLw+DfVukQ9%D zsg<}?8UfO-a4TrCB!S5~Og7Tp!~VQUi9c_A*nlVX%Lt*UQu4v1-Ky&8rq3W8@9n7G zra>B)I4*8W<@KiIveR?_c*+@I%n(Z?Y`syU?BKMsvpK6{7e$G=;Y{j4&A_l;g5_98 z435h?Fw_i-MxSuAk|kU)yE;iBGwW2^Du{gYiRD|w1kmAG^(vm0#JG`Wf%alhQ?f7C zSW8InFOv4;&(RJSX|w}C$JiGC^fyIC_9v-nK_-pXH9ktMlzUr(%jJ2yQ7S6n(^@^X z6Sy(%(Z$(@IWZ@Ll+-7aXB?qdeIvb5$YPWMu(G|H$+OIA5p5%xv15HP&K+60`J@5$ zuJ+D@sx0aW8?*h;qq5Sr=QkqV`>4VIv*#4Y=Tl9&vpgQ1@ng#&$4AS>`IobD+NDs2 z7rg_wJ45$m&yQM7Q@z~u_F)1N1Mby($v{W>Nq)Uj_1UI-%EVF>6kR0oJK*@N-z@j! zr)!^xj8_*N^*hmQsTb=i*|daPjO-u1)-dH()fuv?>uh2H3yLmp5zYzD!d<-8!*MhViyXuTqN2tR6PA0&aOHPmy zQIL)1u6NaQ#9KzMRyWwplD1b$p13EB?B1%9&QPI;7-Hov_2b;+q&hmqa=gFtK8PbG zkz{Py#uCZEQJCon+NR+>F4;w2P~ z+g8+@@n|(CPp2db@3sDdwOYZb#`E)i1-5x^R2&G0mJ>I1XUw-Ks#5OnEOHBH$$BE8 zq)WIuwwwOHE>$?z5#jM;SSa2^b1+$iS3agDS5P7! zDfJkh86%W)rZ&PZ;17haH-6-~;2skyR2r#e(M(j?RQos%Oe!y}NiJ`@j6=2>AIZ-m zt=zXXzb_QaEtK3$aN7;Rl~RU`gp|d9iHCP&pC#9DOtB#$K=?n3E(Tx3FK^NIMwIPP({kc z;->LbS3_4VjZW6x`q;#FG#=G@bwTA-SxDc5%clFkn|N_LxSlxa`%aVo;|2bDnWtXw zN41LT4Hb*=9q$@o=aLtG@NBI+Z0XaX6jTP;JCbu$hLaiMrfBo3O&6Qk(2i!U=%h<}*W4H4$eZJKxIb7mtsc!r7YP-| zU_FC)GmsxK)dslJCiM?GZ&ryFI!v?5GnLk}( zf(qn$XTWG!KeTJn#(95cDw>Xo;NhPs_otUnB z5-Jyp)v_CWCwjVgEz=}v_VI1`E=K8#l|_8)3v4bCs4t_VQi}UWO~066GNzYltq-}J zL&_*Wu6}hl|8mYQ*i$Ir)hE4$ChO|uiD3zoL)|;>DM=ff~7Z$nd=4i0`_Y>&s zpB6$;dsb^xCpyFz@H52MlJ}P?K}o`^^<(2-+HyOf2x8uI{>$e&z^BIepHghM&Qmii zt5xsXxpE%(R}Xb&%UI>5w^D`biXkC^u^2V)Os~Pb5*KAQZdbnAnKu8`AOa@x4>(|` z{Eqj^0=^_}3S7di6_viVt`!E6M!~>P%pXbjG3UN9a4<^*W)IxMI_6}50%B)=&ef5t zxyX&dmD=0m#n*|dM}Z%$Tgr0KHQC2TskSXvdVJ#Gdww2$1aWmN8*qsHLFe~+jpJ#4 zxbJ}uXz&TzeD1?X#JB$W&@ zW#~+D_Tfbnef4qx30BmbW89M{qvA-|WRgcAd;Shu;yut=$09K5=a)CQ_2>nB6CG`k zOIsTgUEs`C2Z?f7O^Z`TBcnz?vlWd>qRUE^JWp~N8S_dnCC&FO^109S0bwie86^ZH zyjc|S(H`OFUzqEL6l8Up-?CUg68_bF1%LVDqfqd=SBRVgG2`=rXOhla-LSC8^rD^# zs`|X%^yb6K0qSH-eCmB|Fp2f5tqWF^m{jvuF{uZ%4ZIuaE+fes{O2a3tQ)|GC=KJ* zc~}eewR2CWXG306EMu&)8L`Kk=y3Et5)`@h2zmW>{v0)R^ z1=%7l_9-RDFVeMz_%L4dJFpJWogKMVZZG4)ZWW@3CpOpX9U_*VZc8OOfvnd<4UP#T z;@o_UoV}u26+A{t{VJ)23|AjCN;j`MslAL3*C;elCwL7RotmnAM9H&3@vkjSDT80g zpbh9I@*o^d5q1~cYYv)*miXmAO^e1R`WA~U71+Q0JRoN^DH>ajEosYja`R;0+cD1j zL8$RIp&R|;FwuBq%-YFOZr zY-Oc;UoX7ZS0a>WKMk8cmzJnpsA_V9W-KKU&c`O}q14m82&MfWBdA(m1#w}{Yl&<+ zx5|uh_G7?##k@jJ--QVn%tv}41#%>4e^Ct zIyhHI&)cu(Oer`jZI30n?5O{3uI7n1+moCaR@%D|ODJWH=s2H)vse?gd>}ZqZz`Di z>&qoTGN}6a2-U-$1!93DklqAI#o@`d2Y@yCJ>RB!H*FJF558$aW zNF-DgpQ1?YVCA36DCI<9n%x=gMSL(<_%RI?!9N;1j1a4`Fo{*G~>c1+fWAfn5sQ$k*x5|c#&b$(iB13zOFdX2-n zMz)?c>4W3yEKT?e^d~;CpomNJ?ktq7JU#SI%R;OKqaN~$`4kbKsx*C|qc`D<6I)Gi z8W|E;@)=FZY2QX5PivKXd5}`TH%*DYN&bn6&w2yX3a(QnI~T0k&!je(rnQ8JV64hx zY-~`6No`CFS*_+FxP(yv{FPgw*wmp#CUoW%E@h?Z>ZYm}R@&_zdGLKEfseu+mU0)= z7<4k0b8~h7if^TH0D+dr$u|5j;*KGAUu3~@e0hvhfsmCZU7m)I?S5fa4CMuP@h)2; z!!u8;Mb2W7^VoXjyY$qoKODAZZqxEO_am0^!79)h=8C~?o~8~i=5d_NFAMU&Dvj6S z{BaC}9&NYi#qnn`rA9r$v$o|Bh@)j)7~aWC@OTiT6)Y?oVKyAbRP2_X0TBJ5bg z}F#B;}Xv4|elUN?w?6EurIA_m(~yY)z#tMrsL~#U;Z27M)Wd zOc=I7SG3yAAhAIx8={Rmm@2(jdtSS4_ygvUB&HiQeefZuQTvy8A;7I>ZeCz}s?V)j zp>5_PH*V8#yEz-13Pywq->!LY`Mvramh1>Enlk)&gELq)qEv3ayxZS)R6kKKE#<8uLp8x$|f3lFTwM@A^y1K*WCI*XRGN+X&%T! zHP$=wKfqdIEbpz6F^kV)dQ?4l7os8!fh#qG|tPjr&z zAbs+q5u4sLR`;+e^v*WS?wnTOV(_J02X{yL1h#FdEPn>%xDiz2UPhFq6}d?FfJQ5? z*kTr#j@IoRH~w*L@!MjQ74NdE<}^a(2Y0WRM<-mqY#c&SE@NPe(~g}{79$1!v6ov@ zyal7V+3IR?$gQ8FhMF@U8oLpjdLd7=%dn)4mpvJzu~NF#*>2zIIlJlh)gzu_t|c6G z`gp9x!eq6FQ{tSHSk90gQ_JJg{U2+VJfA12#m$JWCTIM2x{O}gfk zC3JsZo3VhMPxWA6GG_X8pBlDL8hL1Cx`9xdg0RQ=UDE4Rus&R3@T0J^xBuyg`$x0e zL2?%Zsj^i0;TdBp)I@2B`)xo#SiKx|RA^*dbH)*Zm=US-w0E_g65r@sLtXJ(t@g5C zFveG20TKX-36wq(k@v&Fomj4Fj*GJPaqGO@yRAwSx-M}(wGtkXXz~K%x#kDLETK5# z%F94#^qx*hQ>DvW-DvSQsskn?!yS4ZoxZ~Q=F^Df{9fUaI@CW^Kd)hD@c zvh4S!L`C(gUa6^@$(E_6x_!u%ysu-ukYCPqPgQ0DX8*S3hwsFbT>TYy*Jhl_l^A64 z0Joh!s(XIE8hQIWllye} z^g0)PYG{6&Q6J7qq^~N$^4z2KRi26{DhVT{K)e%HdA)AkwjGh1V3fECJ||^1#OA=B z);13-G+I_7rzSqeOGr|NURFnnXN?urV=Yv@e!|m`4l`z#gI%GFD43EdNI^itjvd*e zzPvQg*kbvGhRL1)_6|j72Q7~Ga3)+vnAw=J$}=(kNEdxYt5$QHSOx<%sDoyoDv~yf zNPU1)!N{m4&P*~8H&XeE$6I#oJ1XK8n~ z12^efSG#VMaw66k3GcEH4`;V$;W9{Cke}pbXR;FdeB94nR~Gwf762z}&5=KQ?HFba z!h%;8&bw&@e+24tdJgp_x(9S?15x~WmW{Ku_TSH6^|sp6TpY;g6Pg?qO1B>9(cSOt zZ>Ay2h8##W2QE`p2&h@t`X~`j8&Ad}zm`*QXK~Vzi4FdyW~XkZ<)J#(cL(*WB zUl6G;*IF+-hiOH^kv8GlWvF)a>RX<^n~R+dQhPv%ReEb-*#o8}-eM&?=TRwkP6?BK zGcyG)#}7pDX8YrDzREk*9A*%@jY!(i)y0?SKmEEq1{{Y%Y_4v3HF{>CfUk9Lx1X*k z=PhbSoIS}hZ5X^tKSRxQT;^RTpL|hC}uTsi=17+8Y0f*j%vf+L!O-XCMv0#ZX z$EW`a=7?RrgaKUYWvdVnD@m97$9(DiJI^{6xKiVaN1&{`A^ASS>K6(0uuf-drKw)e zlus+i(aW-LR3&H$%0bp^TSP-9>DFV~9msTZVU-L!idfB2M8_O`Lw>MSDqbO!<^t0P zCr1~Rxk`%02wbR3ZLG3N6nflh+5F0zvlY3a13`kGudB+?Y2V0M=AgNSiZOretV!QM z6&R;y+{Df-;s<~4&RgIcOm!79l`nN{J-1_6r}s#%cf!>c zz2^JZd&vuePs4dC`O8k#4{!sbagTtn))!~z&HD)u=Bms|tGj3;L2P)9WqEwMM$I0F zdwe8Eo_-MNW-^e|KVQfSr}-mm@DA~+HU8u3hLht1!(eijWvyJ>QhC#3fhaFV`J(`& z#_etu;DgfoR4ZGD%E}~I{n)hX235}ho?#HL7L`TxASvJ2D|bdIEUPgWm5^-h042gP z$zK5nV8yR62YC+HUw{B3dek@He#^SnJOrzG^{+%!BE{O-wt=I**8{)()*|rvchb47 z9x}F`*GdwC;ZvVpuJP2m%e?t~P~q{W)wgHvZ4@$``f<-jPTFY%EzY~{&l_3XnmUhX zrh0?#to2vQCTz&YoJj-SNV^vD+mtVttRF@{!5{0gPTgLqgR|3p^|nLyS$2As20obgjA+VvA!jD`DWOuyPY{ZviW6nisfn9wV9|ufHSZVDN*^ft$=_t9`NyG>7vrj*m zbPjfn#!s{a2ZOYW@005f?-?nEg6`?IK-=dWM+&;f+ZNJ==(KXsmpJ7H*NCgb}uebrv4<BH?29b9pD??*n>vA?X)nMd7Gq_m%b|uWeDjol z%G@mU)H=Zf=LB|CTUVK>7XwiELPs4aa2+JpTdpKZO8Ty!+AI(Yw`qrQ2KKKkimfT&$A>peNpzdO~|^t&J!W z(~cv`igS6B(ay})lB82S`c>To5(hc>ypN(N_F~gIi+W_rWxUra?;;E_FL61}znneH z`mWunBRF^LG|S~%&LUllf4z~z;5{wB=dE?>RyYaHTda5Eh&+l0r48xToVWxD+9$la zx~Vw`1`dYby#dci`r)GJvs@ZP9l$v$tpVxM%4waS zJb8S(I^Pm@Eg{7{TFbZI0uD9%1KSS;>;zs{7WV783#VEJW_}|H8OdTOq6Xs|5iSqFS@_k z%lMDsFxX$%v|`*qOMN~EE=jrC{z5K8cl_Amm_X{z>No1t zELC!fyjYSy>h2hM&F!P`dHX&@zh6d36h=ei*r2sA^w#$Ek_S%;-k71#r-$y5Yw>VT z-3Rr-Jl|T;;lBjR`><*Y%xySp>_e-<-l?24I1`fB7itmw9YPX^z)x8pDp}2g5BVMM z;4-VVKw07oRhMUlNG*YEs&)0g+ zaZgVKrkIQ29hs#SF-}Nzk4SD2r?|HD&5L8+aaY2qY1)Q5%5kV*LsS3r$g*>7>nb1h z1eOFL!?Y){Sah@mCXli$cxUx}uZZ_^!>gegm5Bnr`2)yyZnB7-<74%4AXG%(^s_ddDdPk^f>HWdi-o->te z^|H|di7t;#aobB2q7J?GxNUTpZ#( z{Do|1Kg^sl*c28iM?rYJ;|^hUfj#qHa$2iX_$C~7oIiuiX>3au@U57CKv6j7ZxXd^e!QsyBPqv&^X zD0WVJSQ!V(AH4(1G&sGJ!`j7TIeTR9cn2Eez~`YlUg*JAIA*zzn`pA*N@OH{dUjli_p-a?F)_Q|MCL;mr1mK*n8|^BDdthR1mb&;p`r?O|$|V zD`seJTT+8D2!(cU}8xiByDoJTJP3_fH%b9`&RoFgjKC*pq*8H#ZrEuSL|bU=-N z=n^C!2da$qw}J>znNMcVY&$HEx+;L*A*OpHGwGYx|qq3f> zt-G_XI3v*&K|(J8qsx6a%2cVCTKp3k!(Yi$Dksy@V3**mEG~7WLBR;-CA}Y&XyX4v z-dnd-)o$&>3J8l{ba#hzcM3>1g3{gHA&r0_A>9JfB`LiqX(Xk)Q@VL4_jBvM-Oql0 z`vbiHaB#qiIp?^pG0t(0^BSY=bOQbZ!F8?|LJN7})a3k*N?uM&7eQdKH;ajW)iO2~ zZ1CzW%<@Lf>iM^Xw-|`DyCqQ|Znq=oXOO%_EB8vJ%+~x2-RS42vN3=ogol(<>3HgE zXDIM>y^zK|tkoN%5Fgscz4M}8I5WNno|}5alg@I#F?&VoIW)0Gu|51DH|;7karfy~ zc|qJ2ctbUwl`~~B4>n&MYJvyz-Np@?FgM5boBidp?Pb!Psj=zmVy)Vc@66)-L&B`$ zAvKCMuPw!Dth&piikCw;Q9jz4EcnR-Efrjj5Bc{e&xpp<^#ebc0vX#>sG_NB;OxW* z%iO)eUgehQWRYQVbCJj~;@!bLG-a&??cRI-4lli~aIzM!>)N7h@;Z)fFonNP$xgTT ze}vU@fW_B_f^|NT9?eJN(@WE|-oYv_eyuW9;kW$pQ@`rL7VBln%!Y8q)e+vbqu7_r zpoANTVPD=&zs9|Z*v}Iaa8Lmfxp;i_i2krDX2@U8-iFN54|AC6wltmP)fFqfu;lDL z0hwUsE`&~$*p{B|`>2IiM~&3LEO!x49vM1m1iPAUPvwW)7P?tWa*EDaL^Z5@6s+nt z!#ev06ucV`;GVH;G?^1W+&Hu}9VcIKI`=omg!GGT)l9P1AL11BcLehN{Mt_ul&y0$ zWMf8{$TW3uscMY{vj06be<^hsr4nU?QmvVO{+E41%VB!1FoTs=lk5jrS68`|1yr`d~mtQT;13Jf0&(3_5| z*ozi*IyNBFdHv>P6uo=JOu6b69N{!?yA7_9kwqgR$qxcU`w0t$d#p!rSn$ zhM1w%nnLW$6mZmd!iDmO8e-P0W|Xivm91KL+_>~mr17Y zJj`Am^CbRAAHj7H$K3+LK4y%qA{T^PoOH;d$Z*Zfh0hShuHe5$pN(e?c) z--*C~0PF=@0K6;O(xjEytusCmx>}uW%=>2IU2fLTfmgcnwl5yR<;m6J3F2CM`Ft?? zXQEUlzdbDNw_xPBe%%EMfDC(bI~q5(K05UC`S7bLe(JTp9AwGk6-i#G;xj9jrn0Yh*Ty=<#scOV3d!hflDnKBV1(fNgYpa#y zechqc0F;qi?OCN?RSIq7c8smHFZcd@CFpA4FVL8^A`kPdfx71Gi^=H+#)CF!X}vXA)KXaYy=@Rk(sB;V_2GrI%njgBo;cVHFtt;e#t()&Q3G%PjxVjAweWU5O{RsAi;>MZ&6(XO3oxs(08!ekeRkJ^*5 zJdc-jWcQo+wR;OPUZ1qnmeg%2XO~CLquIZ}wEC}nJkmL`yG|WAuqQ08B7NY0G2a8b zWoA{L;G7(RrMV&gA2^G3QGjGR&eE0ayEQKj$LdbRVi4dFn8kN5jl><CiRrnIeYJ zaG4!r$%|K!0&eE@>L8T>>M-(|5c^&D6wNB)P;gDF=67Wth3KiXzIe2r7p{vrhtA~- z?B79M%V}{|i1O(rdl+vHGp=8;YtIWBB`?KuuFe`JU)hKjj!fC-?0s5p_ADwc9V_h9 z^{vlLJ$Ev?YEt(Ji(-40+5b}kB>MhBqy#cW#wgI4240^iCvA0G9sF<=_rD66A4E(Z zm02dOkoCD1nf{2i>_WIMsWa3>k4FOE5NOZnw z_s0zn7ec-!*W3Px{IYjE7s=5GYZJp1ZX)fu=nz+`SgO}w^cPW|Le_V*)ni8cY`Po= zt-M|q!5EBx;u|5d6HTXML3faO;}XkTDvsbFX>2l;dtWcg`Vx?tZutH@URy+8Phr|{Xmw?k3ZxVdGday+PM77SK zC$~TIy7S@-!QnhWO<@`sLF75r<@iO!-B8qUnx#8}VClwe9FMhJ*xS4s6d4|R=SZ(; zc5thW`pPdlt}{`?X)w@WVV8xL2UvB)VUow?7B4{a99qwq!Dij$K2SJ_SOX8S1h0+F z`huZ*NjQ+KtI4I~^m2&jvEe0Dxq=WTH_c!U;z8;HlI16#uUk;K$@Z8JLdiUZIg||? z)L(*ugEIi3!Kc3k-13r8G8%@%mfzV$7lZO`3vKzXbHG=v752*t!iV7~vAco}FXN(f z#f!DjhX(NOITgeQ1EqNP^rrPQlXd}~>DKJ#1l@ro)^q}p;YFlo5Ed?bV(f9;mV|T= z6PWT?cfdt>!deLDmY*MDunHOWt^X%s*as0q5FkIm9a8@q@1Td%r^=Td?d*|=)Nmj7t< z37bu*!u^E8(Gy1{E+EZ{!(%|mZ3wT*;9;0mMMCFjwU5>*w#9I>a1VDOgHY73N5~xd z^3nZhb+IhO8gO6yE4g8kjeJr74?_{iZd?`4ws_0Nih-r|?CwWc2^)s3f&rLRI|#Q}-=YMU5p(AH+BWOa zDm#z);R5jHP8v^|Xi9FxZ*vbnW2APmGDYfqg0>FKGRXg`FQ*EO>0Tf9xX#A+M9h@E zw6!QT;QSRkP)nO`$Wl+B)xFyn$V>D-vJkMQC9m>cAvukOMZ zQ@!5I*|ydD#nPM6o>q3fP|-?fI{>#N3*On(#h^#s^gR_mNNwUS0ox4;h&8v=Zu(cW zgJ`&4wSJ;5Bz2uFI}&>MzS-%lqWFCl-WEYK!Ob3TdrKPAZ`S7=#B1-Cym3uE8k4D2ROAvKIvz4SLQT79{clXxCP z#qxHy2bnKsxI+9O@tFa&t%qH?KoNO@G&Red6SAF^e~8faW*XF$zQbA$5&O1JSea#u znPGBK3vpuX!YtRG*y0sgNT?+wgm`Lw&O}}=hfB`3b#8F z3y%~7)Vr*>Z|C0g869WvOzxH7n)>xp8ik{6L6(Qcz$muLlBq%HDr2C=L-Pad2_MA`?(=CCY{^u%kH`JNr0%&D zq|b4{X|%D?ucPyX(>z2#4xS)i(;x}URI^9tptR;taJJp6*bnhdb}XuG&B;fmU#K@z*%L4Htidsy z$ld`3-(-I$n=p>>MO`|9%^9{A3-nk&tRsqgO0|M0-XuOcayA|r6ykc(+J?K?FvirSg9%3A zZk}MddB&%~i^9uqyMk}I#Px%RKPlNn|fPsaz@;c(B9!AN^3wKm% zbZJa~oN9(gEe~Z4~&#rf!jQP)$0oeZA6bx$K8M za23c6s;?GQzr1XKX{g!Aq{nvWN2+|A=|rSqy;vF-@DuxGeNBfAfdua-b;k8XduFog z$$$wsYE(2XI1DPEvpKE&&zu7NRt^ZmP&oPgrzgeoK5qWg{1FK2z0+`Ma1<68oydYN zY>f_)!X&#>axn|fF>pNl(bKj&w>ez!e>$zb3yPWL^_j3(g+*@0jvDJ`bU{eXjpB|8 z(se8r=b2?wZ<>@UXAcs|h6y?M<~9#YX;dGtv`oDTD!$Vy-?Xo7yFr0UF(=Y#QI<*u z#N@sAcMD~*`NUpo!@RN`$&%>W!)wdl64C%^6zuv$yU8Jw@ih+rcd_QQoyoKQf+GzGG+C_eL*t2j+qyHnV`B zxNn^wZ3Tvhg=psi3)j{Oii9M3_yW+~!eM6!JDS!KE{@;vv)WTY5ndi}2OkejJwt17 z;QNAlo&MDS;`M%`oVKXkDBJ*;o7>bTJ`?Y@5da3Z(Z@1#Ldk@iMGPdwHqp%=@+<9BaxFx7HdOv-o2 zL(F{NbiO?#V?26$^tI_(*Sq=!#z26D@`V|R9v&1ty5W5OT>c1$5@^iR+v0OW$*GQ` zHeP^ldm~zZg!@PSCyfaZeCU!X>p&VrnbyXaj?{dbH&50t*oYxkOw}7Z^P#km56(I` zF0!opO%>g7??F~In0#q?XKShM_*yXC+tXWPZ3V^a4U-_1IFn%Ep2#a6o0dww1U@F1 z)VFK~r+Y3(UMMtafG#9H1rxs&Ca$s4qE&BN^uplyVR^aT>@{n9TzQ z;p^q9D>xbJDF=We^>g6GAhFFD^qAA?#?I>0r2RP=CQ91J6*Nz)*@HWcux5*$NhRU{ z9^ngw4v-jL{eCCgX>2e5sMI*Y0a7*u(Gcny9;?Qq!A?TOR})!0(6KQTE%q6c5GWZA z%|Oz%G9pNdEgkwh8lfBPcBfP5lNE+Bu2vSOOGc6$jEi|rMr9?bvi+SGffwV(%_Nhz zm>EX6EsJVac&f@t7G~<+d!H|$UhC%EdHX5gpf}|rA_P#@OP<@}l{{cjwvzTRPB$Hj z{1^g>Zv~H;eVKwQv%urSA?Q0B@kPn(No;dOPQ8k74do|6<6lmN?}9aJj6;cj(^5(9 ze?yY}@{>Md40iii)jZ$qYJUqP7Y72mVTL1Y6iKj#bpei*gCUycDDD}j)kL3q36ZnSTo*Qsp z5(R$Yk=}u&k*VER6-ShN(R}RD@>Qp^Wi)B6DPJli?K{~I1b#WB$czsR!V+M%o^jub z5VM8`z3mZwQxf99>MQOyY->3fs0d1Xn_hyg%gwK^*djlW)vy&8%4#K43zU2n;1Fa6 zy84z^88tVr8_<2#7LpxxF>#la1r4)lzsYWy#wBSdgqY6r(W-XHL4r-ot>^QwX=Gwc zJ|5GH1?ZP;XJ-=`HKnL4RqL7hX}-r4}8InYWBFT)QZ! z-9=*bDPhF~-&_faruz^lw{R<|OIf(rEI+-m_Gd#n$r5WueNH1;db0Vn#Sg4veL5m% zf!*b%o20(B{I@}smf?Q{OBL(^qZ_L2jdGx2(C=OyL>1C_{j zA<%#DdRg>9ai@S@*QGWYrU>k>iBoiU69AoXac){?=S*6y-E6%%{_&1hND&)KuZ%h2 z=JVQ1#ttD4K-B1Fby-BQ8$$E8ZEZ8j;~AT;a>d6h*KYgpdR0jZ(NRr;@O2> zi5};MGO0@x-K5UepYyL@)$$4jB$3*Lcd^#nK8%BjYMN?vDru>A1a$A4MD8z?r-$%~NGBVjPA*3QAsW>F zn8flvhD9dd33BC~SyLo^kc=zm3{yF`P7}KL@D7K6E)_hreE)irTv~Ylhx3(DdARDL zyMunq-54G9JX+S6E^oP}50+}t^ymMl_^gQS zQ7L&mVKyt1g#Z@0{D_#bs9uzZWn-A5wth)&2uPheSaoSSSlcn)@dq|9iSF<@`wX9= zyP@8+WDPGj9eoICm`TTxdxBd<7=;7=H0aMb{0-7WsZw9>t)9$vWE;kQaf)n z5>dIkl!?6WWcOs`8h2WhUdMr~PFzo;5FP~hb#^u>L$mBFLwbrlj8Ez>! zw4y0D_CW4QkGH&^avo#uVsz|HoRf4Fg3ugwYTLaAI@tTos=Qk}gxa7sUolDr!D z_yNGaq$?h5dKyV{5_3+M22Q3_dJl6t09NF=l0M7X3 zFbhPc`-|;B>fdmR9{(x#GMZSKd>W53ir}fz@Q#Z6+Pe_`KlTJE8NhzN1es!sK~XAY zz?2cyO*uLgrmMfP)#OfAkiVme?f+87$j=4SqAc1|jRi;T4sj3>F)tH78;pM{^|?Ej z9Q4`wF!%MVs@_FOQ1Qb9D>=Ji_LvTkx#H|b0P5Mm2C87oIfzTwv8Kz1ne2dItvs@Q zt8a%B@8~E7{Zz2lMXNbibzBlb6p>?O_zBxGE__7;%P%uhGMQIfK#MZ{yZGeJ-|ZSr zzlEgZtC+uP*&8T*9ROACdSf$OqAY)HRbS^@swzvUcN?{sRVI94Nn7+$=WcP;*CVJA zm8=A|P7v%!U+@hj>N)%A9fya9A(76b=(eZO-a7W{I%AZod%OKru$Q+Vmh<@)=fGU- zU{^xDOJIbTO#M{|AMt^W_-vlcEG!*VC2E>p{`Ogf=GT0$khHO0VF@#jxx=59%*V|@ z$j6#tjb(E!6mt9Ot0=ERlWV^40F+GbCI%VQEq5=pKEU*JSX}#NWQ!$)ZyjLgB3BCJ z&njCumWMFNCY;8F;FY9e?P=4 zVh*#PON0eE4_*+oLN<{Ot}`XLaQq+O7xbgDKU)>Y>Gn)3ZW~deLI`rR@!u7T=;g^H1tyswhL*{6N~V0Ds;U3!N!CG zFOdseWaN+$+s9L32x5tL&nGX!5hyjWJJl*QWy_s}v09L=yFdbS2+)|&9K*iWUWXib@r)DMXBVQyOt@go|gq*@6C|A^6w%G>5*<6 z+|21+IMaw=Y{X~s`<81c{Dg7LrF*9rv2-rP3xpKVUnhhW{SXa)59HypD*Y{Q$oryc zeoI>8g#q(!l{Fs3T8!oQ-d>%QhF=B0MkE)wzS_)OubI+~!(!0*mcnV3YdimSG+&;8 z_i8OJ&ulm&55H9p?gOdwOs!37Bmwu(j9WLJne=M+`#i}AY&OHrFqNLwuFsji^uTD~ zxb^FC;!~g7rm+Sl|Sb;ud(= zlkX%Xbu@{$VQA1_*GTM$%LR{pB|(J7gYa}U<1WFdp9A^g%gYX(>9J1~Re45nMeMsE zvm9WgUbCXQ2)fKyK*-xr^5x>LMQ2{!)VFvA)8$y}D{mDWd7);`FfSnXM!KEgrNutL z_$xjyVVZhHOU>p3XI)cTq?Lg7m zwbt97j1n&&d))T6%+0;Nnq9o7rtNsnX+9!C?0adjJz1JMYimpKTWbpZ;{Woo6!Ovm zZyNNMzwx&Rkzx{1hXaQblwO?mfvl>y(;U-!409gZw;Lv5Tm~a8k2_yYJ4&Q~n@35z z>e)Se)Z7VTIIV}WKoIZpWkX&&=lK1|klwxNKtay?9QSq1#qq)t8FgIsjb)J`f+i-c zK^IYARjyCR(#0IC%W+aEBwWm8GE7}eL9cX?CW1?XX*{f&Db(_tJV+M&vef17hvGfO z?Wj0AN>dL-DMb+fostw*@c2L&N8-^C`(Ys-J#6@kYWr{957@SpkV{pe>KcGx#7qeo(pFd-WMVv@4vzWYS} zyd@g!=u@~{_uA%6U?^x-QO^WZzE9QQqC_1IGW@aUhSz+4cU?`Glx_S%odZQ&w7

    7H>K>3|GI?T(>OurUy zpd;cWYr*1Dd*JgJOh{)l(7(FM^USg^zm=aB_DE9w-k)U$-tLWb5I@+1nP=clya~OLOcwGGaF-R2(4E8=RS^g z;Sk^y90@yO6usY0oN{&=85baA6NkYn04a4b@!#!|><4d^twd$-*gC(2A$uUF^SWID zH2ecg3I-qy5E|$n|B~mKq)MXwu)n!McU>@>Y?qJlGiSSif_)~eni^#B@r{xxMXw7T z1=cY&%&*ry-wxpBHtnB>hklCu(&GIl<*kRX6#h$8!?}f0t8az)zDPMDAQ!X34^~xt ze9xyblPPcxLr14*W6JG?qA0+@s$8Vxh8*(d;pZvp**4dKO}J zBMoPI?$F`!*pp@?P%J={4icx-;0Y4y6WZMfbk zUSEwCt|g2IhQX53(N*ffs1_-;{Be-f-;ors53h(>%-`txLkY}AF@LZ4&P+8!c}bj@ zB9`0SsBmC5GKRMU83Ms02g*#(fFKUadsG51abeA3Aut&y$$n-$0C;35XFBz=(=}?LIFG7&MFTaKn(BmtdmMeu8Z^pv1(O9l z9O<6MBpz+G!;!=T!EnZr&=Y7(o(I6GviENhN9P)xs3<7Jg|5~xdV6~-qfRi0v2bv7 zP+sWNTBF53?0p&jy3uSph>ypj{}w;~AQ6t(U1SJoy*#yA|7RzJ*knNM+v|O%<@Psh zqlJUArxx9+vEw4P&wg8PqIM{UrKD!W#4a0bv^1dn=ED z4Vr4#c7(t%%Z%{ZrIaeA*-y+JFYOmZX=~c)1lh6BYE%Gh}ZP?#F{5oP1lJ}w6t=jc?0NugGhhZo&Wgh z2Jzh z-lf2a6(U>ch~pDMZn9VA;o&TUE||p_R_+LUH~E7sP?3id6y)jniVCKZ&dC2K!1!x{ zkI;$94ySaz0@QcLT@h-+~*l8(D=ath}Oizc+{$B$Bdz0YehjUjZ7CW-iY=Tx(Z zpt;HYBiBZUS{lI`?O?1E?}ipF(j4CGSI3G;n^69-te|3i0Y?VufD5c`KXP(w}t0!xcY+Wmk^U}vH@u|%(R!4WTD_Jd{sEog-_ zHqR84f#3xOaXT6Ondr`FUc`8*&XniDIV#i3*;<>pvNw%=vdqnE|NXK2L5lsB zIEz5_Dw4(@fo|YXi5VN+4r_yifSm+%K-OGV78Ui!(ncm5dhs5gYD&+qnI`M@_BNKq zng^wa;tD>RloVgkv)bkLh@X5aR}EoO0P*mjhZIO(J=QZBHN(&T!?~Y7G0DrvYZ&ys zJDXYu8vNFPl0I1JMoAIzIe%#Oy3jTlAKRBcf?xENvy-Uo%$pCxlrZ41EFd|ML?`E12(M9XSog#Sh{qI=w*QvWA`bWxR=c8HUKZ%e(()7Px zC!G}l2GI*Z9H`f{)A?MC7h8OePlkmP(TLx;&AYGx=%G-rwPhfUSCx&8?X?a<A!y>lYa-z@kfJR`Z#iudNc~oZ7K~vACX3qO$u*}0~6&Pz__R(j~p{MulQ)q$c zRg^zzat)xp1=a3Hqvrp1d9ja|C+4*1zKYDin9>ZW>!yjKO$U<5G~c*WPI0v4UbbrEgljF5ssz%MS*L30`NR_(6z!-+na=q4AA9usFL{jqr{lZsY zO!LlHyT%Ju*Hf&@Dr>zavIk`5{;_xgqshDg@GNDcehJfms#5+#(V@l#wG3F+mEx9_ zFbzUs-3qLs=XIar<1rhy^HWQvfN>gM6s`^$57%a_%0Jq*-f37)0-Z8jE^`Qh&eYWW zl~bFiV~UB35BC?bz$%kZ;i!xPF7Q2-96cjrwS<(tJm38#0M@)N4zNI(w#~p1JOIp( zm%W+w^!Y{>8oxG7^T*m#dfXmrugWNAMSrCNf8N>eSIfL?V>)fvEo)qhm1X*A)4MU4 zT6$#J=(=AO<%B_O^P@r(j9@ceWirXi%gZ}jYojv_q=?=?*sJn;@a-KMO2E-|7v*!= zQOj=va!U9<*52)K4%;~%JYfH>t*^fd6^S3OcQCpFpk-cGR@+R-ZBG;t1F-v8rvI(~ z4?PyJVCg5;jsN|;^}nxK(koC9gMb*~qW6h>!+L^FN%vRy=!#DT1=2uBfcLmsPkfd= zAs29>oXGTMt((9uel?1J$eQV_-uucCkJB=~jR>ARuezG!Rh99(HqqYP8mp;3pou(L zrq6kD+ugTZz+LuN7<4dx&ZUSJ7`5K`Q&6GC_t+}vt85?aA3*mXi{W=B%mZ&dT_vdj zEW+gE4NLkE=2w$~$H=OQGtIkwgsIt;`Y5#tVf8`PX zBozMNw-4q8))Bc`5BndErA^^2gn}Pg=zsV8__v!VIDNe3)5Ij?e?}H*Fz})1FezPs zT-ARvx*F2s`iR_Z`{M>EA7%fs5BOEj{x2i_{<{HCqNYXC5{UnmI{f9Py#e#DL5DW} z`)2qbAO4S@^dyhlXpz$-;-4?=3k71-dPjQRe>|V^&zp+J0N&pRFAzcX4?`C?KRy)a z#CYVt@z_89_ixYd2uS#1NQe&pWCoUi_ah7yy(GDsa~f9PKL7G(A?<%z^kb-qgG1%s zY@MQtib}5M+0Mt$2G;QZSVHtt8;n zRd_A~{_%K8LmqDr8+*{Qp`qa>Fs%~sZB}O@%E~dA*w`GZhvV!ZY$#ZJm$rm-_uleX>~< z6&1PVnAOZ%l12dTta9Dg-X1TEbKd*$eYfJn>?$i)=smc5gM|rih)1W|sbIM}*vN^@ z4w)E0K#IU_fzgF{oNV;gE&4J@5xK+zPMY1HDpYeEK&>VKvZ(@c+Tsjg4i6$Tbo8f3 z%!P=nCRb8!WL!$fIt19r*|T!aGzwpgQ8bBx!@{!x;KEws##>A zh3h^lL-TIx)|SJmV9PICAo8$OuT__+=}2r*(w8um!Nme_f%Ke@Va?rxq3X!|aG`lJ z5vGMUE1a{?@8Lc}Pi`ZLG&+AjX|ElJNTWb}9>Ikrz8(xlCg3U=c+wqDr)mr&E0cX5 zX`wQO&vj}+wy#Tg2zczXIIMqEs9~{?aRR82VX*^+Uk;zcoYb78<;l$v%Tn6Gk|(v<0w{=8Fk`&meV|@~Fq&dT6)7xD!o_ zOHY3evtp5rFJU1)k@r_?JKOVp}up# z1S-4dErBCjM;)kjgmiUwi?(N(Bc`Be1>W^|mBHxhOzh%X`@OR9^(Dk#JM}4?|1l=x-CXQ1skiNu8=GZN^ICi2*#pj@J6VhJ?vrM(EytpwGV{DRW zuzhFbWUE&H9)_OzFN4~=zs%3!B?cDIV9G#yA}Km`G|N*MM2HqBI?Fqq%O-^i6Q9qy zKp)aiSr7l7M0nxau(R*-e1A6e&1B+gu1s8s3}qTzaG4AuHsc4NW~9Y12_*5>TsqMh z)N&P_GDA<%sx5#iOHjMQYjC2mnqtjMuErYUs_tu&;j_Q22{4`r(O}p;fcRpf_)uF% zD15Bvwa;1n5VU{U{w5{NhCY+zL(~-K`!7M!QFGD z7y6JQfT%G|?Akb+(yJrBIo+lhOyy>tzLSgU>C@NRV`J(42P_RK|=Hro7A_@ z`%(lvswSA~&BRDe(L49G-oaJ^%V(>fEZZ-_mu@bFUORJcJpEsrsgg_|hnTE~>CIOy zg&)2R9c1vE$!f|$Qa_0s0}0jvZNYP8p6=^+tMlV z_&0yo79rw+5WT|76?*lWaa6n&;b7Swnmc=~!tVa|DnS~wY6xCT^;yE~CEVd&^rp!4 zyKe}_^Du}+F|v}@g-}EdzH+~ehAH|>L+0MVP;>;^(HJiU2!Tl$BibO zbkg$FoMkXWa8{)al7bJ-HFOGGR39+OsHB%hPAu8Dq%>OaT#I)(2xGu>7j1=1?!dtC z2q2aj;nxks(%O*W_41RNs%TU-fIX7ycDQilfn9xo6B7(-~;>v{rHTwOulFGLq+;yh#`vo&s{ms2^%9;2fIsttxj|_TE%5$fu0SP zT7YG@dliCe>y)ik!tJ`JwfnN$5CzN7i6`194NZ(@4Yl?^*hU;an#CdD<90o*)_dfq zV2uEAural=oC{hXGzLr9_je@MN#Fddur@_0p({}!iO*yzR6~?{G7RYi0zRR^hyw15 z-e^0NZIBX^Flk; zH#(XpyB?<~gccUXq8!RBg%CX_?LhE1L{w9Na$QEdRs&51SLi2&{d`H{edIrZv-_j+ z)B4aVl|TR%E2J{JHB@~Dgc>5l62FQh8{&(2O$Ab(EqlYy0K-@-dKCCk&bN2El^sIM zT=_vycgP3q_zBr&rW%b$gAZfonG|I%zzJ3xtAE6^f15?P??U~>umz|DkCy6iTh-x7 zKOK%>`dT2Di95BEm#ABH-LaYBRfdHQLnEdI({CZi6j08d>`Td$Q$F}FmwE-l1+JPA zY+w}7X%z`y+GE@YM3)kwS3rCz{a~sK`9x~uzI7p5@*c+R#p!lhZ{%SOKt16HIv#MB zr~B|*M^F49!oDy-LV*{vHM`Wd4vnMKD4i7tOU)=oNb2yoh89D9e;YXbZR!!o(26lq zJOD4i=^nWD(gY03&Rv>JZ$}xld|`kQqG`X>D_Gk*>2$P4@%auAF|Y1qc~rjU0rb`c z!aG>*et@=o6)M3Ox?|lBu%u!L(L7w6uIi8em-{dO>jTizmu=mNghEF#6?{Z-v_GQO zxV}H5V(?uu8imCW28Itb6Ag(&oPs{Kg&#|+l)y3v#M{*Wph>Q9+y-z(^-Q2Bs&_j- zMvL0=3u6)(w#s={{jp{*w@tc97|fQvCTi%)6l4sOC2qq`&~&{6MX;0!tt0fWWnlZb z3^bk@1an_gQ%FTIQlLP=m)v)=+G|TZAP&!8;{z3mO4w&CZT>+rL>L%EO>hzc4D zgd!+dQV+optNbT92U67gq)vZn&>d6Q$2}tj9_l$(fmKXnoIO$!%1F`ofF{8+s>?x( z5=Y_X+K=?)>MzXq5xJQg6D{_SE|@v|U+sjJDD|&dw3b46r70QJ*cE;NCC^uQ%V--! ztPfiplpsy_wGvtKo{go2sC%qfpKM(K@aD^B2T2j8vgck(KOQv>mV$3uc%cu?z$aIq z6r_Io2xnJ58}U=89c4$hqof7)GT$VU53xfi1bXOgN&8qH?z_&S9)Lw#cXl2F@gB|F37o;Dly){7~&dMvy>3@lu}EO!XOVfNK4mXz9M}0 zqo!_URP#~*BFs{+n=qg!%h(;jIzhnIyFp`ye052q8DJh8|Z#d&<~K~FGrGC2c$QZUJdy6QW^#ZS)j_d7dv8L?gKCiRJ39wtC8$O6`iXf z_T9bE_H8%}Eim#jT&v)~a|w3vFloC38j6&3J_!`fi?Cj_1g_C|+GVt$6ELwq)?CgK zlu0a}y~qvg(Xt#^52%im#y7?|h#ot6Jk}!DfwE*3WH`B{8cmp&22q*1jJb?PW1Alb zhG)0)UoAX)W!pf}0~gKkH8Xj#I1I)m%_e$&cQsV7b_-1nSdP#>=%?~&KJV;%j!4ZR zxCC?T=>g`dnQ;nyGEux6xLz($f{L9#h_2;Hd@r-HsDJd5kNT$QtOnEu8@bt@0$%WM z0t5g<>X^jsMCfIyxWz%s(AX!*n1h5xz0lF3kK64F#;Ox~0X{&zyj*fQM<>2gZx0It z$XA-r>8{$n*)QD3%k_l)FTZ8%VSDN>Q|>eMvi0fVq8=cdGA$SDeM#eG z?Q!$e7eWzM7J@=#DcrQyfJQtmUf=!{qqu$vGx*5O9##@k6MRH;69cZj48l;T$T<1+ z{U1cu$@PPzUdRQFSm1IOseX>4f?_KTrUk{RB;~R~S*#YVV$LQx%@NDDj_`(<9ak~O z22xxA`I04kY#QR~+_CHI#-yJ!KU0d0DI*4kMpAthvo{Cill_f_?HT+icGP@B%qrLK zj6xVnUEV#2S0QrHAci7nA6!4Hn(I_>E^s6eb`)mhq+sbS5A`wgFNfE?(ew3{#X2Lt zAwfNXC{GZ{SQ2i#Q^}GmiuNUZ^zCb(G!xHel;gm~-8s_P${jx-L{DZz!y?gs*%2gM zMq~Vf7LGzU$}EqOI>DP|Onpl$P<3?t>$d;ru|sr4E};`8>z|WL;EDw&2a%irWyUX% zt(@pkA(T01pxzEZenF%sS(%|DR??uvi1w$A`aP%Ov7+ul6Sw;QTrNA~GJLLk=HQP4 z_&uK_K+#GC+lyrJ2tn6Lw1E^D*L~n7BggN9KeKWn-oI(`592VAK)^RYtsFrW_oHNH z9!6NU)bInULAt=KZu)b3^T(WK4lR9E0#=&G$h|wSC!=+GgJy=+#6z&#Yj3Dtq`enC zg6yZc;yY|WTb_QBk|f05bFilw`d6otzlcAPEPD9pk*Q#-O+Y_W!mByZ43V7AAWs*s z+4SE8Ry)to=^_so9KIZ-(qE0B+pPAe1n8NV5*nN~wJX0X69p}_?N|#K8p4oMZU(8c zAx47E%NrnSa!14y^llz*ag>FM7u)YUw`Qn;GRbt zoXj@q0SFl0EdquYou zBBkqX!2|{qScC~pJBG(g|MHs0=#P-VJva_+6c8IRJ(hRqF03(NA;iZ6Vg2dkiPGIi zEcegcDAKIvIKBfBjs_piSs+-dyBU7f!9Kb(DVbezOwnlrs|djQPUx#vzqWxtDNIr9 z5$I$9TqhohJY_GD>X0wJp_J~iKZPJN$IBfgkC>esy;s48q5!=I8sRXO%PHO><`Z<$3p}NgA$JL#hl^>M8e*Hj9^9-o_^(}N$YVB))x0# zGHmcTF=eR67yHhushaGW!a_0M=vGI4GvIn+&X}~U5M&RLrgkPG3!pED1CB;XU%2WB=I8xRfNrp*Za<@98R&i3X42b z%w1O2C?@z0f*UGie2#qTDW1Xnp_W5VLMcL=wFGf+o2=(kKr|U!0-{M15KY3r*L3Yr zA(4|B)Wjx_S8&~450_+BH3Gd=HUNsP7M9>K9J6hzQv-}c5Gh{(QH@W)(9pO0IcHE4 z#34yE3xEo;Kwey;ItY`VsLsg<%gPHI^F0iWg3-<2VYPcY0{O}t$URj)E2Mwu`T6Ac zGFziuq7q<>W4Qc@(vCnaJpw1NTk@r3gwVV+@#veJH;}WuByDn_9y^o)z8V&Jw*TBKPlLxQW!zmxL<((Gqqc9cQ3>~a3!_xvYPde@JsL6qQO z(+MRz`>;# z7HN1+Rs6(~d?BH7@+4@h%Jv#!xJW#lVjP$j#~8a3DKyuOAcDwy7MOK#kXN48`Pl6Y zl$lc0yvLhowtrVv{EK9t++l+(^aJ%BSU&@YG zJ!K`FC<2c~a}5n%^#aomlt9V9g71-W3LqG=f(%x$MR%)b0q7?oso%PXg2zw^3O?X@ zNq`g^5iD~?_lpss>|lS2@@hbv*iMNTKNw~fy_R-#%$N9!oO%lr8}~Nx2vo{y(EJ$K zgLeR-V~hx;O>DB7V~93;an9T|C)muW4H}xPmr{T&N)c>K0TY@U{AD>n3Zn`dMI3C) z8wcMNA|ot!PbNACql`_RHh2BQl;vAkU%CgMULY%El}s3v$AZ@TJM{#(!z5PD&qv>a z+`6<-uVKL>`mx;yA_vq0;m$~?&fi-le(0XwEEP1{%yMl5DHc{fm&&!TVY>8j=Z+sE zZs(8uMDaYqMZhLvH2~CdJ^oit@C3>pr-z3qcGi$%%&o)sCP(y!oU5Ln%VHEJG@g=T zO$N4yf}JUZl{a!Qm`y&+7^Ma#a$knyk#U+9Ii`bGV*|Wj*HOSwhM*3qntuu=hSCe9 z!1iE8;XAd2=Q^fzcC6nG&!;tI+u`9E3iVAQomhAS;fW}DXhrpnR6IhpI}$Nwp!vAwwlsPu64)s2)m=k5KRW9t#A0a0Uu;56Cyz%({$^%7AsP zccPf8*kUV&l0vYtX;;evUSi1r7aiE94nNFV5Y$dZ{o4`r$+n%g0y>b>vw;IHes z{l82+41S>p=R-`%XvzO@ds6Bk_CV1tfZVdf1I6h%i%28QIjx zF+@5%gKhino#AsYK)6IAFWd0xW*mjlSA%U|2B(mE!q80(-yG3S!A2zJ55aWuGt_Y_ zSh(D=J{L1C>2&ua&reQhl)F}*1!oT|oVI1YgO2wR!&9Ww7Vl5deJx6q)Z8~e1P(Ls zJldO)_MW%SZ?7{_oRdo&;BG6%tBH&yBuj1q-1Hu^Km98rnZkZN8T* z4-fSgB1w`8231*f*r{bS0sOs(!X52#j5ci_MG z)Nd1XUrv!*N2|X(wG@=8$X!*?jF+rtt3ipJdE%NAMG}Rfc7#JehAH|)cqj>?%8zCq zls#OWM72_yQwr$NFM1Y~^{I4F5ro#ot7yOAfDf5S<@_^}q90uX2e%|0=I3~tD6rpX zK(Z~~gH)~3pseEB*8C^$rBa(qH{=~Hg`bK=2d9suqZJe`LR3vOPi8g9;kLFzBD@QE zr1(I>4*~*MPVjymXeRXQ20G?YeH+TUjabcOkaaMTYp}BvE0LKSdg63`G(2B$Ev3EtTLcqql%PxrGaJpbH zBT9B=esaWKIZ}Ha6ItBQoSVC_5L;AOP{1v{f9Q!OhX99w%rlKpK&DRLv?dSSiXQ#OCyn=k^x@1b2#u6t zoJIY0dGAIAlFdu_sF-8L)M$^~(}D9c@iMs_*cV3{pOfaAB8lcXjPbA|e75`-x})B6 z&R*H5q772#crz4rl6Yr;I-&JSs#2`~eQau!5d^KwP>E4CQJQxqi5qHD;#H4UljdS~ z-%u@`oSNGFcvdFGzth9?U49r6nx{Urp~U-jlVtlRluI|B5Owi*aiXL5?W_Z5JMWEI z?!6vJEp3HpOVNU5|RBv(J1DAO|68BJtzx0nu++3k80Op$UtW%eu(R-^J%+` z!Te3xyGVZtGerV>)wA(p!8HOnFq)Fq(&OmIpVZw_c)VQ~m>r2~NL(x7^c4hNra#)^ zyokamNwy;BzKxK^6V1RGFYZ)-5OOmf-7V6ePxz}_-`1c`n;DX|OVqFMLBbq$bFEI0 zoE^>z0sZh#RYcSsj``qbAsS^Q(udLd@?cfO#;AInjQwPB7f*{kZDjf^1vh;y0vjA0 zu)QR8)&7yz`ybmDiG-097Eg&3)id4e!N1^0PlZk}gONBO8@~yNs9V<~_%^0FPz(CJ zEHnt-T}zCMfh^;f^ZJ|8kZqvT8kSTY;lQ zr<&Pl=Rlj~6=VI8HrWm6)gEj5ZpGsOll`ilJ^*&n4@1XSqo)5)5BiXonUHSfC7{xQ zK?lZ)_Z98cK?epM7<6F#Z11I6hQ5{1w~|%8GdeKnz@P);H@{=(J_g;#pzWk9ph5=* z9T;?A(49|ufQ#ls(8Gw!0;u%J=hCZz4vhZ~Fm6fvdCkpCz@41#jU+uhObe4OO~fjn z;4d==%~Q1yGoM}5f~m84B_*muBO|7bjKc`ZIB9Cv>tjQ2a_y^7B03jGdnz`9Gml$j z?Qo9>x$P0Ob-#+#rLT}1pwlxGAM+|Ig3Br~X*$82;MxrV=>9TLUetH{B`d`xe5vs= zk79+d1rkNA?BY}L^;x99oi@wb@PiDDuljCb3BQC*D5$pa*h9-)PH+bTRJXmXA~cnj zlKR!abS06@u~Zu&w`p*B;K`QsuvpLgMaS5M@A%0UUC(7kL3IgpHlZ>oQoZ4?#C#t+ z-jV$jG958IGQt@U5Dc-H%N-k&vK^g0sF0n&xAlU7_1I20M!_rL3+3(fbVBc> z#4l%tCK@)!_*dS}hXOHhOqS1LU(+3ID5$(~oAa#kzyY%Jy3#;eQtSu2(S$7MfX>15 zOB>fZ??n{i+RZ1E3bF+?33@fhV?14oJV=<%vRCcAo?k$M?Ds0G*&-GE5)Ng@RAM^c ze;LcaFJ!V#m_%e3iP6JcNXKY8imgEi$i!iptIwLd*N|KMbV?-On(}HIr;}czt}_|c zPno@l>)Ez8DcVR9ti=DB_C54_Re-u+_N3XOJ%cG<;=TNP zLHCArMwqY@%a-k`uHLu7CCia>k(jY9)N^EK&A7GoVJw0OuAi^HB{9fhZZ6ICRO^7dzl0g>Nsc^R znm1!!Ab)PVWnx@WA;gx+hIYj@tqS>M(AqVDbN^bV7Zw20ki|}|sZwiF@-1lXi189# z)9iPMfxNjEcJlC=7GT2;Q9mVtJ4}ly<7Mj8Y_`l+ZNi}PU^uDB$#_jlsvzh8(5hp3 VHTqHN?jPXqpn>WBoPDs+e*wdtYAFB! literal 0 HcmV?d00001 From c00f6552413c7f9f38c69fc5fa82225519450971 Mon Sep 17 00:00:00 2001 From: feiyun0112 Date: Sat, 3 Jul 2021 10:48:41 +0800 Subject: [PATCH 002/368] translate 1.2-history-of-ML to Simplified Chinese --- .../translations/README.zh-cn.md | 116 ++++++++++++++++++ 1 file changed, 116 insertions(+) create mode 100644 1-Introduction/2-history-of-ML/translations/README.zh-cn.md diff --git a/1-Introduction/2-history-of-ML/translations/README.zh-cn.md b/1-Introduction/2-history-of-ML/translations/README.zh-cn.md new file mode 100644 index 0000000000..67c93dbf5f --- /dev/null +++ b/1-Introduction/2-history-of-ML/translations/README.zh-cn.md @@ -0,0 +1,116 @@ +# History of machine learning + +![Summary of History of machine learning in a sketchnote](../../sketchnotes/ml-history.png) +> Sketchnote by [Tomomi Imura](https://www.twitter.com/girlie_mac) + +## [Pre-lecture quiz](https://jolly-sea-0a877260f.azurestaticapps.net/quiz/3/) + +In this lesson, we will walk through the major milestones in the history of machine learning and artificial intelligence. + +The history of artificial intelligence, AI, as a field is intertwined with the history of machine learning, as the algorithms and computational advances that underpin ML fed into the development of AI. It is useful to remember that, while these fields as distinct areas of inquiry began to crystallize in the 1950s, important [algorithmical, statistical, mathematical, computational and technical discoveries](https://wikipedia.org/wiki/Timeline_of_machine_learning) predated and overlapped this era. In fact, people have been thinking about these questions for [hundreds of years](https://wikipedia.org/wiki/History_of_artificial_intelligence): this article discusses the historical intellectual underpinnings of the idea of a 'thinking machine.' + +## Notable discoveries + +- 1763, 1812 [Bayes Theorem](https://wikipedia.org/wiki/Bayes%27_theorem) and its predecessors. This theorem and its applications underlie inference, describing the probability of an event occurring based on prior knowledge. +- 1805 [Least Square Theory](https://wikipedia.org/wiki/Least_squares) by French mathematician Adrien-Marie Legendre. This theory, which you will learn about in our Regression unit, helps in data fitting. +- 1913 [Markov Chains](https://wikipedia.org/wiki/Markov_chain) named after Russian mathematician Andrey Markov is used to describe a sequence of possible events based on a previous state. +- 1957 [Perceptron](https://wikipedia.org/wiki/Perceptron) is a type of linear classifier invented by American psychologist Frank Rosenblatt that underlies advances in deep learning. +- 1967 [Nearest Neighbor](https://wikipedia.org/wiki/Nearest_neighbor) is an algorithm originally designed to map routes. In an ML context it is used to detect patterns. +- 1970 [Backpropagation](https://wikipedia.org/wiki/Backpropagation) is used to train [feedforward neural networks](https://wikipedia.org/wiki/Feedforward_neural_network). +- 1982 [Recurrent Neural Networks](https://wikipedia.org/wiki/Recurrent_neural_network) are artificial neural networks derived from feedforward neural networks that create temporal graphs. + +✅ Do a little research. What other dates stand out as pivotal in the history of ML and AI? +## 1950: Machines that think + +Alan Turing, a truly remarkable person who was voted [by the public in 2019](https://wikipedia.org/wiki/Icons:_The_Greatest_Person_of_the_20th_Century) as the greatest scientist of the 20th century, is credited as helping to lay the foundation for the concept of a 'machine that can think.' He grappled with naysayers and his own need for empirical evidence of this concept in part by creating the [Turing Test](https://www.bbc.com/news/technology-18475646), which you will explore in our NLP lessons. + +## 1956: Dartmouth Summer Research Project + +"The Dartmouth Summer Research Project on artificial intelligence was a seminal event for artificial intelligence as a field," and it was here that the term 'artificial intelligence' was coined ([source](https://250.dartmouth.edu/highlights/artificial-intelligence-ai-coined-dartmouth)). + +> Every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it. + +The lead researcher, mathematics professor John McCarthy, hoped "to proceed on the basis of the conjecture that every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it." The participants included another luminary in the field, Marvin Minsky. + +The workshop is credited with having initiated and encouraged several discussions including "the rise of symbolic methods, systems focussed on limited domains (early expert systems), and deductive systems versus inductive systems." ([source](https://wikipedia.org/wiki/Dartmouth_workshop)). + +## 1956 - 1974: "The golden years" + +From the 1950s through the mid '70s, optimism ran high in the hope that AI could solve many problems. In 1967, Marvin Minsky stated confidently that "Within a generation ... the problem of creating 'artificial intelligence' will substantially be solved." (Minsky, Marvin (1967), Computation: Finite and Infinite Machines, Englewood Cliffs, N.J.: Prentice-Hall) + +natural language processing research flourished, search was refined and made more powerful, and the concept of 'micro-worlds' was created, where simple tasks were completed using plain language instructions. + +Research was well funded by government agencies, advances were made in computation and algorithms, and prototypes of intelligent machines were built. Some of these machines include: + +* [Shakey the robot](https://wikipedia.org/wiki/Shakey_the_robot), who could maneuver and decide how to perform tasks 'intelligently'. + + ![Shakey, an intelligent robot](images/shakey.jpg) + > Shakey in 1972 + +* Eliza, an early 'chatterbot', could converse with people and act as a primitive 'therapist'. You'll learn more about Eliza in the NLP lessons. + + ![Eliza, a bot](images/eliza.png) + > A version of Eliza, a chatbot + +* "Blocks world" was an example of a micro-world where blocks could be stacked and sorted, and experiments in teaching machines to make decisions could be tested. Advances built with libraries such as [SHRDLU](https://wikipedia.org/wiki/SHRDLU) helped propel language processing forward. + + [![blocks world with SHRDLU](https://img.youtube.com/vi/QAJz4YKUwqw/0.jpg)](https://www.youtube.com/watch?v=QAJz4YKUwqw "blocks world with SHRDLU") + + > 🎥 Click the image above for a video: Blocks world with SHRDLU + +## 1974 - 1980: "AI Winter" + +By the mid 1970s, it had become apparent that the complexity of making 'intelligent machines' had been understated and that its promise, given the available compute power, had been overblown. Funding dried up and confidence in the field slowed. Some issues that impacted confidence included: + +- **Limitations**. Compute power was too limited. +- **Combinatorial explosion**. The amount of parameters needed to be trained grew exponentially as more was asked of computers, without a parallel evolution of compute power and capability. +- **Paucity of data**. There was a paucity of data that hindered the process of testing, developing, and refining algorithms. +- **Are we asking the right questions?**. The very questions that were being asked began to be questioned. Researchers began to field criticism about their approaches: + - Turing tests came into question by means, among other ideas, of the 'chinese room theory' which posited that, "programming a digital computer may make it appear to understand language but could not produce real understanding." ([source](https://plato.stanford.edu/entries/chinese-room/)) + - The ethics of introducing artificial intelligences such as the "therapist" ELIZA into society was challenged. + +At the same time, various AI schools of thought began to form. A dichotomy was established between ["scruffy" vs. "neat AI"](https://wikipedia.org/wiki/Neats_and_scruffies) practices. _Scruffy_ labs tweaked programs for hours until they had the desired results. _Neat_ labs "focused on logic and formal problem solving". ELIZA and SHRDLU were well-known _scruffy_ systems. In the 1980s, as demand emerged to make ML systems reproducible, the _neat_ approach gradually took the forefront as its results are more explainable. + +## 1980s Expert systems + +As the field grew, its benefit to business became clearer, and in the 1980s so did the proliferation of 'expert systems'. "Expert systems were among the first truly successful forms of artificial intelligence (AI) software." ([source](https://wikipedia.org/wiki/Expert_system)). + +This type of system is actually _hybrid_, consisting partially of a rules engine defining business requirements, and an inference engine that leveraged the rules system to deduce new facts. + +This era also saw increasing attention paid to neural networks. + +## 1987 - 1993: AI 'Chill' + +The proliferation of specialized expert systems hardware had the unfortunate effect of becoming too specialized. The rise of personal computers also competed with these large, specialized, centralized systems. The democratization of computing had begun, and it eventually paved the way for the modern explosion of big data. + +## 1993 - 2011 + +This epoch saw a new era for ML and AI to be able to solve some of the problems that had been caused earlier by the lack of data and compute power. The amount of data began to rapidly increase and become more widely available, for better and for worse, especially with the advent of the smartphone around 2007. Compute power expanded exponentially, and algorithms evolved alongside. The field began to gain maturity as the freewheeling days of the past began to crystallize into a true discipline. + +## Now + +Today, machine learning and AI touch almost every part of our lives. This era calls for careful understanding of the risks and potentials effects of these algorithms on human lives. As Microsoft's Brad Smith has stated, "Information technology raises issues that go to the heart of fundamental human-rights protections like privacy and freedom of expression. These issues heighten responsibility for tech companies that create these products. In our view, they also call for thoughtful government regulation and for the development of norms around acceptable uses" ([source](https://www.technologyreview.com/2019/12/18/102365/the-future-of-ais-impact-on-society/)). + +It remains to be seen what the future holds, but it is important to understand these computer systems and the software and algorithms that they run. We hope that this curriculum will help you to gain a better understanding so that you can decide for yourself. + +[![The history of deep learning](https://img.youtube.com/vi/mTtDfKgLm54/0.jpg)](https://www.youtube.com/watch?v=mTtDfKgLm54 "The history of deep learning") +> 🎥 Click the image above for a video: Yann LeCun discusses the history of deep learning in this lecture + +--- +## 🚀Challenge + +Dig into one of these historical moments and learn more about the people behind them. There are fascinating characters, and no scientific discovery was ever created in a cultural vacuum. What do you discover? + +## [Post-lecture quiz](https://jolly-sea-0a877260f.azurestaticapps.net/quiz/4/) + +## Review & Self Study + +Here are items to watch and listen to: + +[This podcast where Amy Boyd discusses the evolution of AI](http://runasradio.com/Shows/Show/739) + +[![The history of AI by Amy Boyd](https://img.youtube.com/vi/EJt3_bFYKss/0.jpg)](https://www.youtube.com/watch?v=EJt3_bFYKss "The history of AI by Amy Boyd") + +## Assignment + +[Create a timeline](assignment.md) From 136658e970c40ff8bc16dffa018186b6e3293104 Mon Sep 17 00:00:00 2001 From: feiyun0112 Date: Sat, 3 Jul 2021 10:53:06 +0800 Subject: [PATCH 003/368] translate 1.3-fairness to Simplified Chinese --- .../3-fairness/translations/README.zh-cn.md | 212 ++++++++++++++++++ 1 file changed, 212 insertions(+) create mode 100644 1-Introduction/3-fairness/translations/README.zh-cn.md diff --git a/1-Introduction/3-fairness/translations/README.zh-cn.md b/1-Introduction/3-fairness/translations/README.zh-cn.md new file mode 100644 index 0000000000..063c189813 --- /dev/null +++ b/1-Introduction/3-fairness/translations/README.zh-cn.md @@ -0,0 +1,212 @@ +# Fairness in Machine Learning + +![Summary of Fairness in Machine Learning in a sketchnote](../../sketchnotes/ml-fairness.png) +> Sketchnote by [Tomomi Imura](https://www.twitter.com/girlie_mac) + +## [Pre-lecture quiz](https://jolly-sea-0a877260f.azurestaticapps.net/quiz/5/) + +## Introduction + +In this curriculum, you will start to discover how machine learning can and is impacting our everyday lives. Even now, systems and models are involved in daily decision-making tasks, such as health care diagnoses or detecting fraud. So it is important that these models work well in order to provide fair outcomes for everyone. + +Imagine what can happen when the data you are using to build these models lacks certain demographics, such as race, gender, political view, religion, or disproportionally represents such demographics. What about when the model's output is interpreted to favor some demographic? What is the consequence for the application? + +In this lesson, you will: + +- Raise your awareness of the importance of fairness in machine learning. +- Learn about fairness-related harms. +- Learn about unfairness assessment and mitigation. + +## Prerequisite + +As a prerequisite, please take the "Responsible AI Principles" Learn Path and watch the video below on the topic: + +Learn more about Responsible AI by following this [Learning Path](https://docs.microsoft.com/learn/modules/responsible-ai-principles/?WT.mc_id=academic-15963-cxa) + +[![Microsoft's Approach to Responsible AI](https://img.youtube.com/vi/dnC8-uUZXSc/0.jpg)](https://youtu.be/dnC8-uUZXSc "Microsoft's Approach to Responsible AI") + +> 🎥 Click the image above for a video: Microsoft's Approach to Responsible AI + +## Unfairness in data and algorithms + +> "If you torture the data long enough, it will confess to anything - Ronald Coase + +This statement sounds extreme, but it is true that data can be manipulated to support any conclusion. Such manipulation can sometimes happen unintentionally. As humans, we all have bias, and it's often difficult to consciously know when you are introducing bias in data. + +Guaranteeing fairness in AI and machine learning remains a complex sociotechnical challenge. Meaning that it cannot be addressed from either purely social or technical perspectives. + +### Fairness-related harms + +What do you mean by unfairness? "Unfairness" encompasses negative impacts, or "harms", for a group of people, such as those defined in terms of race, gender, age, or disability status. + +The main fairness-related harms can be classified as: + +- **Allocation**, if a gender or ethnicity for example is favored over another. +- **Quality of service**. If you train the data for one specific scenario but reality is much more complex, it leads to a poor performing service. +- **Stereotyping**. Associating a given group with pre-assigned attributes. +- **Denigration**. To unfairly criticize and label something or someone. +- **Over- or under- representation**. The idea is that a certain group is not seen in a certain profession, and any service or function that keeps promoting that is contributing to harm. + +Let’s take a look at the examples. + +### Allocation + +Consider a hypothetical system for screening loan applications. The system tends to pick white men as better candidates over other groups. As a result, loans are withheld from certain applicants. + +Another example would be an experimental hiring tool developed by a large corporation to screen candidates. The tool systemically discriminated against one gender by using the models were trained to prefer words associated with another. It resulted in penalizing candidates whose resumes contain words such as "women’s rugby team". + +✅ Do a little research to find a real-world example of something like this + +### Quality of Service + +Researchers found that several commercial gender classifiers had higher error rates around images of women with darker skin tones as opposed to images of men with lighter skin tones. [Reference](https://www.media.mit.edu/publications/gender-shades-intersectional-accuracy-disparities-in-commercial-gender-classification/) + +Another infamous example is a hand soap dispenser that could not seem to be able to sense people with dark skin. [Reference](https://gizmodo.com/why-cant-this-soap-dispenser-identify-dark-skin-1797931773) + +### Stereotyping + +Stereotypical gender view was found in machine translation. When translating “he is a nurse and she is a doctor†into Turkish, problems were encountered. Turkish is a genderless language which has one pronoun, “o†to convey a singular third person, but translating the sentence back from Turkish to English yields the stereotypical and incorrect as “she is a nurse and he is a doctorâ€. + +![translation to Turkish](images/gender-bias-translate-en-tr.png) + +![translation back to English](images/gender-bias-translate-tr-en.png) + +### Denigration + +An image labeling technology infamously mislabeled images of dark-skinned people as gorillas. Mislabeling is harmful not just because the system made a mistake because it specifically applied a label that has a long history of being purposefully used to denigrate Black people. + +[![AI: Ain't I a Woman?](https://img.youtube.com/vi/QxuyfWoVV98/0.jpg)](https://www.youtube.com/watch?v=QxuyfWoVV98 "AI, Ain't I a Woman?") +> 🎥 Click the image above for a video: AI, Ain't I a Woman - a performance showing the harm caused by racist denigration by AI + +### Over- or under- representation + +Skewed image search results can be a good example of this harm. When searching images of professions with an equal or higher percentage of men than women, such as engineering, or CEO, watch for results that are more heavily skewed towards a given gender. + +![Bing CEO search](images/ceos.png) +> This search on Bing for 'CEO' produces pretty inclusive results + +These five main types of harms are not mutually exclusive, and a single system can exhibit more than one type of harm. In addition, each case varies in its severity. For instance, unfairly labeling someone as a criminal is a much more severe harm than mislabeling an image. It's important, however, to remember that even relatively non-severe harms can make people feel alienated or singled out and the cumulative impact can be extremely oppressive. + +✅ **Discussion**: Revisit some of the examples and see if they show different harms. + +| | Allocation | Quality of service | Stereotyping | Denigration | Over- or under- representation | +| ----------------------- | :--------: | :----------------: | :----------: | :---------: | :----------------------------: | +| Automated hiring system | x | x | x | | x | +| Machine translation | | | | | | +| Photo labeling | | | | | | + + +## Detecting unfairness + +There are many reasons why a given system behaves unfairly. Social biases, for example, might be reflected in the datasets used to train them. For example, hiring unfairness might have been exacerbated by over reliance on historical data. By using the patterns in resumes submitted to the company over a 10-year period, the model determined that men were more qualified because the majority of resumes came from men, a reflection of past male dominance across the tech industry. + +Inadequate data about a certain group of people can be the reason for unfairness. For example, image classifiers a have higher rate of error for images of dark-skinned people because darker skin tones were underrepresented in the data. + +Wrong assumptions made during development cause unfairness too. For example, a facial analysis system intended to predict who is going to commit a crime based on images of people’s faces can lead to damaging assumptions. This could lead to substantial harms for people who are misclassified. + +## Understand your models and build in fairness + +Although many aspects of fairness are not captured in quantitative fairness metrics, and it is not possible to fully remove bias from a system to guarantee fairness, you are still responsible to detect and to mitigate fairness issues as much as possible. + +When you are working with machine learning models, it is important to understand your models by means of assuring their interpretability and by assessing and mitigating unfairness. + +Let’s use the loan selection example to isolate the case to figure out each factor's level of impact on the prediction. + +## Assessment methods + +1. **Identify harms (and benefits)**. The first step is to identify harms and benefits. Think about how actions and decisions can affect both potential customers and a business itself. + +1. **Identify the affected groups**. Once you understand what kind of harms or benefits that can occur, identify the groups that may be affected. Are these groups defined by gender, ethnicity, or social group? + +1. **Define fairness metrics**. Finally, define a metric so you have something to measure against in your work to improve the situation. + +### Identify harms (and benefits) + +What are the harms and benefits associated with lending? Think about false negatives and false positive scenarios: + +**False negatives** (reject, but Y=1) - in this case, an applicant who will be capable of repaying a loan is rejected. This is an adverse event because the resources of the loans are withheld from qualified applicants. + +**False positives** (accept, but Y=0) - in this case, the applicant does get a loan but eventually defaults. As a result, the applicant's case will be sent to a debt collection agency which can affect their future loan applications. + +### Identify affected groups + +The next step is to determine which groups are likely to be affected. For example, in case of a credit card application, a model might determine that women should receive much lower credit limits compared with their spouses who share household assets. An entire demographic, defined by gender, is thereby affected. + +### Define fairness metrics + +You have identified harms and an affected group, in this case, delineated by gender. Now, use the quantified factors to disaggregate their metrics. For example, using the data below, you can see that women have the largest false positive rate and men have the smallest, and that the opposite is true for false negatives. + +✅ In a future lesson on Clustering, you will see how to build this 'confusion matrix' in code + +| | False positive rate | False negative rate | count | +| ---------- | ------------------- | ------------------- | ----- | +| Women | 0.37 | 0.27 | 54032 | +| Men | 0.31 | 0.35 | 28620 | +| Non-binary | 0.33 | 0.31 | 1266 | + + +This table tells us several things. First, we note that there are comparatively few non-binary people in the data. The data is skewed, so you need to be careful how you interpret these numbers. + +In this case, we have 3 groups and 2 metrics. When we are thinking about how our system affects the group of customers with their loan applicants, this may be sufficient, but when you want to define larger number of groups, you may want to distill this to smaller sets of summaries. To do that, you can add more metrics, such as the largest difference or smallest ratio of each false negative and false positive. + +✅ Stop and Think: What other groups are likely to be affected for loan application? + +## Mitigating unfairness + +To mitigate unfairness, explore the model to generate various mitigated models and compare the tradeoffs it makes between accuracy and fairness to select the most fair model. + +This introductory lesson does not dive deeply into the details of algorithmic unfairness mitigation, such as post-processing and reductions approach, but here is a tool that you may want to try. + +### Fairlearn + +[Fairlearn](https://fairlearn.github.io/) is an open-source Python package that allows you to assess your systems' fairness and mitigate unfairness. + +The tool helps you to assesses how a model's predictions affect different groups, enabling you to compare multiple models by using fairness and performance metrics, and supplying a set of algorithms to mitigate unfairness in binary classification and regression. + +- Learn how to use the different components by checking out the Fairlearn's [GitHub](https://github.com/fairlearn/fairlearn/) + +- Explore the [user guide](https://fairlearn.github.io/main/user_guide/index.html), [examples](https://fairlearn.github.io/main/auto_examples/index.html) + +- Try some [sample notebooks](https://github.com/fairlearn/fairlearn/tree/master/notebooks). + +- Learn [how to enable fairness assessments](https://docs.microsoft.com/azure/machine-learning/how-to-machine-learning-fairness-aml?WT.mc_id=academic-15963-cxa) of machine learning models in Azure Machine Learning. + +- Check out these [sample notebooks](https://github.com/Azure/MachineLearningNotebooks/tree/master/contrib/fairness) for more fairness assessment scenarios in Azure Machine Learning. + +--- +## 🚀 Challenge + +To prevent biases from being introduced in the first place, we should: + +- have a diversity of backgrounds and perspectives among the people working on systems +- invest in datasets that reflect the diversity of our society +- develop better methods for detecting and correcting bias when it occurs + +Think about real-life scenarios where unfairness is evident in model-building and usage. What else should we consider? + +## [Post-lecture quiz](https://jolly-sea-0a877260f.azurestaticapps.net/quiz/6/) +## Review & Self Study + +In this lesson, you have learned some basics of the concepts of fairness and unfairness in machine learning. + +Watch this workshop to dive deeper into the topics: + +- YouTube: Fairness-related harms in AI systems: Examples, assessment, and mitigation by Hanna Wallach and Miro Dudik [Fairness-related harms in AI systems: Examples, assessment, and mitigation - YouTube](https://www.youtube.com/watch?v=1RptHwfkx_k) + +Also, read: + +- Microsoft’s RAI resource center: [Responsible AI Resources – Microsoft AI](https://www.microsoft.com/ai/responsible-ai-resources?activetab=pivot1%3aprimaryr4) + +- Microsoft’s FATE research group: [FATE: Fairness, Accountability, Transparency, and Ethics in AI - Microsoft Research](https://www.microsoft.com/research/theme/fate/) + +Explore the Fairlearn toolkit + +[Fairlearn](https://fairlearn.org/) + +Read about Azure Machine Learning's tools to ensure fairness + +- [Azure Machine Learning](https://docs.microsoft.com/azure/machine-learning/concept-fairness-ml?WT.mc_id=academic-15963-cxa) + +## Assignment + +[Explore Fairlearn](assignment.md) From 2b69c1e7268a20a087c6eb5f3a2cb650681e51df Mon Sep 17 00:00:00 2001 From: feiyun0112 Date: Sat, 3 Jul 2021 10:55:28 +0800 Subject: [PATCH 004/368] 1.4-techniques-of-ML to Simplified Chinese --- .../translations/README.zh-cn.md | 105 ++++++++++++++++++ 1 file changed, 105 insertions(+) create mode 100644 1-Introduction/4-techniques-of-ML/translations/README.zh-cn.md diff --git a/1-Introduction/4-techniques-of-ML/translations/README.zh-cn.md b/1-Introduction/4-techniques-of-ML/translations/README.zh-cn.md new file mode 100644 index 0000000000..ec42fe705e --- /dev/null +++ b/1-Introduction/4-techniques-of-ML/translations/README.zh-cn.md @@ -0,0 +1,105 @@ +# Techniques of Machine Learning + +The process of building, using, and maintaining machine learning models and the data they use is a very different process from many other development workflows. In this lesson, we will demystify the process, and outline the main techniques you need to know. You will: + +- Understand the processes underpinning machine learning at a high level. +- Explore base concepts such as 'models', 'predictions', and 'training data'. + +## [Pre-lecture quiz](https://jolly-sea-0a877260f.azurestaticapps.net/quiz/7/) +## Introduction + +On a high level, the craft of creating machine learning (ML) processes is comprised of a number of steps: + +1. **Decide on the question**. Most ML processes start by asking a question that cannot be answered by a simple conditional program or rules-based engine. These questions often revolve around predictions based on a collection of data. +2. **Collect and prepare data**. To be able to answer your question, you need data. The quality and, sometimes, quantity of your data will determine how well you can answer your initial question. Visualizing data is an important aspect of this phase. This phase also includes splitting the data into a training and testing group to build a model. +3. **Choose a training method**. Depending on your question and the nature of your data, you need to choose how you want to train a model to best reflect your data and make accurate predictions against it. This is the part of your ML process that requires specific expertise and, often, a considerable amount of experimentation. +4. **Train the model**. Using your training data, you'll use various algorithms to train a model to recognize patterns in the data. The model might leverage internal weights that can be adjusted to privilege certain parts of the data over others to build a better model. +5. **Evaluate the model**. You use never before seen data (your testing data) from your collected set to see how the model is performing. +6. **Parameter tuning**. Based on the performance of your model, you can redo the process using different parameters, or variables, that control the behavior of the algorithms used to train the model. +7. **Predict**. Use new inputs to test the accuracy of your model. + +## What question to ask + +Computers are particularly skilled at discovering hidden patterns in data. This utility is very helpful for researchers who have questions about a given domain that cannot be easily answered by creating a conditionally-based rules engine. Given an actuarial task, for example, a data scientist might be able to construct handcrafted rules around the mortality of smokers vs non-smokers. + +When many other variables are brought into the equation, however, a ML model might prove more efficient to predict future mortality rates based on past health history. A more cheerful example might be making weather predictions for the month of April in a given location based on data that includes latitude, longitude, climate change, proximity to the ocean, patterns of the jet stream, and more. + +✅ This [slide deck](https://www2.cisl.ucar.edu/sites/default/files/0900%20June%2024%20Haupt_0.pdf) on weather models offers a historical perspective for using ML in weather analysis. + +## Pre-building tasks + +Before starting to build your model, there are several tasks you need to complete. To test your question and form a hypothesis based on a model's predictions, you need to identify and configure several elements. + +### Data + +To be able to answer your question with any kind of certainty, you need a good amount of data of the right type. There are two things you need to do at this point: + +- **Collect data**. Keeping in mind the previous lesson on fairness in data analysis, collect your data with care. Be aware of the sources of this data, any inherent biases it might have, and document its origin. +- **Prepare data**. There are several steps in the data preparation process. You might need to collate data and normalize it if it comes from diverse sources. You can improve the data's quality and quantity through various methods such as converting strings to numbers (as we do in [Clustering](../../5-Clustering/1-Visualize/README.md)). You might also generate new data, based on the original (as we do in [Classification](../../4-Classification/1-Introduction/README.md)). You can clean and edit the data (as we did prior to the [Web App](../3-Web-App/README.md) lesson). Finally, you might also need to randomize it and shuffle it, depending on your training techniques. + +✅ After collecting and processing your data, take a moment to see if its shape will allow you to address your intended question. It may be that the data will not perform well in your given task, as we discover in our [Clustering](../../5-Clustering/1-Visualize/README.md) lessons! + +### Selecting your feature variable + +A [feature](https://www.datasciencecentral.com/profiles/blogs/an-introduction-to-variable-and-feature-selection) is a measurable property of your data. In many datasets it is expressed as a column heading like 'date' 'size' or 'color'. Your feature variable, usually represented as `y` in code, represents the answer to the question you are trying to ask of your data: in December, what **color** pumpkins will be cheapest? in San Francisco, what neighborhoods will have the best real estate **price**? + +🎓 **Feature Selection and Feature Extraction** How do you know which variable to choose when building a model? You'll probably go through a process of feature selection or feature extraction to choose the right variables for the most performant model. They're not the same thing, however: "Feature extraction creates new features from functions of the original features, whereas feature selection returns a subset of the features." ([source](https://wikipedia.org/wiki/Feature_selection)) +### Visualize your data + +An important aspect of the data scientist's toolkit is the power to visualize data using several excellent libraries such as Seaborn or MatPlotLib. Representing your data visually might allow you to uncover hidden correlations that you can leverage. Your visualizations might also help you to uncover bias or unbalanced data (as we discover in [Classification](../../4-Classification/2-Classifiers-1/README.md)). +### Split your dataset + +Prior to training, you need to split your dataset into two or more parts of unequal size that still represent the data well. + +- **Training**. This part of the dataset is fit to your model to train it. This set constitutes the majority of the original dataset. +- **Testing**. A test dataset is an independent group of data, often gathered from the original data, that you use to confirm the performance of the built model. +- **Validating**. A validation set is a smaller independent group of examples that you use to tune the model's hyperparameters, or architecture, to improve the model. Depending on your data's size and the question you are asking, you might not need to build this third set (as we note in [Time Series Forecasting](../7-TimeSeries/1-Introduction/README.md)). + +## Building a model + +Using your training data, your goal is to build a model, or a statistical representation of your data, using various algorithms to **train** it. Training a model exposes it to data and allows it to make assumptions about perceived patterns it discovers, validates, and accepts or rejects. + +### Decide on a training method + +Depending on your question and the nature of your data, your will choose a method to train it. Stepping through [Scikit-learn's documentation](https://scikit-learn.org/stable/user_guide.html) - which we use in this course - you can explore many ways to train a model. Depending on your experience, you might have to try several different methods to build the best model. You are likely to go through a process whereby data scientists evaluate the performance of a model by feeding it unseen data, checking for accuracy, bias, and other quality-degrading issues, and selecting the most appropriate training method for the task at hand. +### Train a model + +Armed with your training data, you are ready to 'fit' it to create a model. You will notice that in many ML libraries you will find the code 'model.fit' - it is at this time that you send in your data as an array of values (usually 'X') and a feature variable (usually 'y'). +### Evaluate the model + +Once the training process is complete (it can take many iterations, or 'epochs', to train a large model), you will be able to evaluate the model's quality by using test data to gauge its performance. This data is a subset of the original data that the model has not previously analyzed. You can print out a table of metrics about your model's quality. + +🎓 **Model fitting** + +In the context of machine learning, model fitting refers to the accuracy of the model's underlying function as it attempts to analyze data with which it is not familiar. + +🎓 **Underfitting** and **overfitting** are common problems that degrade the quality of the model, as the model fits either not well enough or too well. This causes the model to make predictions either too closely aligned or too loosely aligned with its training data. An overfit model predicts training data too well because it has learned the data's details and noise too well. An underfit model is not accurate as it can neither accurately analyze its training data nor data it has not yet 'seen'. + +![overfitting model](images/overfitting.png) +> Infographic by [Jen Looper](https://twitter.com/jenlooper) + +## Parameter tuning + +Once your initial training is complete, observe the quality of the model and consider improving it by tweaking its 'hyperparameters'. Read more about the process [in the documentation](https://docs.microsoft.com/en-us/azure/machine-learning/how-to-tune-hyperparameters?WT.mc_id=academic-15963-cxa). + +## Prediction + +This is the moment where you can use completely new data to test your model's accuracy. In an 'applied' ML setting, where you are building web assets to use the model in production, this process might involve gathering user input (a button press, for example) to set a variable and send it to the model for inference, or evaluation. + +In these lessons, you will discover how to use these steps to prepare, build, test, evaluate, and predict - all the gestures of a data scientist and more, as you progress in your journey to become a 'full stack' ML engineer. + +--- + +## 🚀Challenge + +Draw a flow chart reflecting the steps of a ML practitioner. Where do you see yourself right now in the process? Where do you predict you will find difficulty? What seems easy to you? + +## [Post-lecture quiz](https://jolly-sea-0a877260f.azurestaticapps.net/quiz/8/) + +## Review & Self Study + +Search online for interviews with data scientists who discuss their daily work. Here is [one](https://www.youtube.com/watch?v=Z3IjgbbCEfs). + +## Assignment + +[Interview a data scientist](assignment.md) From 8f7dd4181b2ae896daafecc29edf8679c6818da6 Mon Sep 17 00:00:00 2001 From: Peeeaje <74146834+Peeeaje@users.noreply.github.com> Date: Sat, 3 Jul 2021 15:07:36 +0900 Subject: [PATCH 005/368] deleted some files that are unnecessary --- .../translations/images/ai-ml-ds.png | Bin 60304 -> 0 bytes .../1-intro-to-ML/translations/images/hype.png | Bin 154886 -> 0 bytes 2 files changed, 0 insertions(+), 0 deletions(-) delete mode 100644 1-Introduction/1-intro-to-ML/translations/images/ai-ml-ds.png delete mode 100644 1-Introduction/1-intro-to-ML/translations/images/hype.png diff --git a/1-Introduction/1-intro-to-ML/translations/images/ai-ml-ds.png b/1-Introduction/1-intro-to-ML/translations/images/ai-ml-ds.png deleted file mode 100644 index 11c9bf3767955eedb662fa6a4430064babeec5cc..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 60304 zcmb@tWmsE5*Djm{PjP~~7I%ufLvi=uQrw*&h2rk+S_l*<#VJs%#T^Qi7I$yaLmzve z>;2C8_hqhR&z@Q9UNURXn#?58>ZU=lC@0KiaGkkJGHfVI!>Bvj<*7VJ7Tgy##Q zt)z-10MMBD;@%wT`JT#BK~n_)@Mi!3U=aYo?Q;`s7Xa|$1OWD-0Dw>y06^%P*RCP* z+!0}|uV|y90$_fwqXLkD$bi3yKmh3Z1NUEb;B%b>@E7^H2G9kP{-Pde|Kk}0eKVQ&X6%0H80LZic5EWqBXgmi1 z0BW_@*7wv`c`an=;>2!lx)7+QM*@O0kG~f{|H&%t;^^Y8QZ?X91O^&i~Nw*Eu>f4BXMUWDWC$o|(z z|1*sK(fS-mVqg)D|HPyi_-UG37yys}D9T7``vMPxP+%tVe#apJE8VpwiXF_g?8?q5 zTWgJbuNaVLr0k;oiX$RKP!r`*u}o1b5GnZRfN@O4*%qkC;qd{f9c;BXbM>tY9b5A^ zK_16-$MvlzY&&bhwE3bt1>Cmd;yh3MA=x2higa8f>b?IiUm&i_ZH^Uok)VH<7!Y#Z zYiUl%KMPzG30;4wJLWYI0}1_~ML+u$l<{9B089jL3c}9@k${Q*S>W7G{{g>-a6+x` zn7NU7|AqE^y9_J-S2<#W_KsfbSb^>Sjn+Hb43z#?S<&i+r61J}Lni;NDlFX|;onNO zR>FWRD|*RHMgK+vvdkd=OF<+@H6Vg+YXgmz#J|xHbSp6ar672*;Mqb)GuaB*ztNrz zCjI}(VAqxIhzc0JeCRJ$v#$r3j<;e#UmsrEw;u3_zNsgEcG0g)Py(H1a4dAq^bnG@ z+RO07@a`9hf_`N6Xz>Bf!q(IYU%<20q1Dgcad2qQ`tYI%X7lLZgN?1U(Y0zZOc&{< zjXvZ?Hsn^<&}er{E@CsiA=TB2AZ{%1Q#SwyOZ7BsVY~KbJ79Xfe7*Fh%`xTKLlT&b z&|YQ~8TJajG9Krk*Iz{OE)stFBSLxd6h;{-z?R|wM3Ua%h@r(>%{$Ix(tc6tMXYnv z!~R7qOHu?!##5=_z1~5e4$WLr`m-Mb?^3B-1k2FxxB(fIcU0H-s!`hJwF3kI|8(Je z%{}+3H=(nFl*g6@GxQK!vVo){UnOx0aeK_a!$Y9*JT$Qk=5Dr}<7sYyM^QFLoFr7K za6oaljFb>H$<+Yfbj(H+F-T%N_jeNqa3`|71!4`ng?mr+RI`1i3 zN8_#jj-458w6HVw7vJ zkYxii$`chwI@|i2zmDlG`RnE>I@SX8Vax2xVCHn<2Anj1pjdumM63Z#K!XnaXESu+mu3fU% zp7lwJc95&dw!cNe0tU7NrJOVXxUC3_*MO{ouuB;zi9EJQC&6JGZa~csMl>$*d~smt zbi(Jx<_iH4HWi(O8E3;@v<&N-ft$T}XlYqnvqxa${t#+wu$4Q^temGm{2QQ&JgMy! zLfgEbanrJ8s6KgcCPFVm2o~j5#d@@K>fb@KDSlSQ09D&Dldtrbk4gpzl_4l8OZCg9 zJ0nOvW?;(zHWLnDsX6s5si~vLdj-R@I_%_5XH#4^V-1*A4$nsBopywf(l?gcPg}7` z>^lG!mjHDtrkQnva!hq3r1WE#Wd}0hIct^cme0m3!x!2+Eiylw z3VS1txL`9T0+n;VwWkuIz0uRSg>$&L*}VMQ#ajMrx5iw@FGF~MaSkdN(EDvJ$M1lK z@LL<~9c}@4V?gx*S%cOqP^YdGxE9Ix?9-(TNoq#9)WoF=|4<}oMxz%t=b6T|agihL zE^gpYf{fsVM0G?SJ2qp<Yvp>Gqscljdd-y=hx1EeDKm+oN zcb-UEQ$Rja^?{IrA>rNkra`O#!h{(o6Yx*XaC8xelwygg=jm7nsLMr6&VO2iR#S`# zn2t>2A#H0@M-Ie-4EwWuMxt5~8@{jZR@GURGcKJ2a=@BZt{4Taai zm@t1{Bz4;gK+MFXQM_gvu;O!-<2>;}mMR-!xw)g7`sU{lf8uVun$NWe&VnyA2}RMp zHa%9+UF~DCg1DaDlm6|E2QxCHZzLQpDJnviDoiE&U6K?sy#%#NU&PC5Ml8=JY{=Et zhfI3Odr?cqrVye~Bv!%>9IL=CVL=DXfCXYez2^=(ty8Ev13v!WjMyy1r;ty=AM5GL z&8`#JuknX@TXL!^5P^rjcI2bE5e1_hTpP1@f0J&iVeVH2qU~K` z1&tm|`K=FdS&bMW7VD^zm_$v;43nbVGL=YlJ4XXW53x%p7O^45kG|#m2|BFu0?(Mv1{*~cLxLMP~zEKznZrrahr&+ zy@DU_0$=QR-$8RNTxS^o_jqfip#B&6_5PA`+U}> zf}|UR2d`6CO1$Rp`5#w)X85Q&`n~?g=ZqvZ+GhfSHWJbStWy z&PR}+Rk9ARvG+?+be-9l0pYK^pwOOAFeM3LQOP1pC2bDR(@BRKtHYQjoI8Mq`F2Go z&ls)l4S85m=w=QwgV$9nPVA#rYvzocc+XhhWb$Q=bI~Lp z*q~}QE89oE!D--4BmsYo1v!>FhF{W^(IS&`uf0$_;CMC@U?^0U3ViREJhz%h|GZcm z1)0wX^P5aQAx?C2^s#$kl`+@cN%_8%-c2&dvSLhf0mI3skMfJIcqL+|HpIh*iCn5r zM|BUgJ&pLnnPo9AUSd7RSGr_LrH`GGa-!tijf?u9AS{iEmY<09OsboF`C{P_{07z$ zAjcm6s~way+DPecfYJN`mAzZB0t4p2qPNN&J>Tn5`oCU+Eg$|ON=o3E>C^FXAtg$E zihYOYDQ^Aa*yG?+ORkaW>32At^A12xyUeod$-EXxi-XbYJG7nQ`M8u61bV1-$msOMQF*q2H$XmT z3TnG4wD9BO8-)BEPO3Z5Y*@o*0;$|huZm90@j9BU9$6(IPrL;M{upMpT=F7b9FgDV zZ2}y`Tysz4X92r`j=iZ=p&R&drTEn4GZuQZqd4@jn<+Zk+^y}$lC1sMbM*OX!j7^f z-9NF8(Aym-H-;b}+n<~eW$a+PW-okdJj#8J1Pn0Ih(m%lee42YHVj+wX-%BDFbn}H zbDwbtNlB|!0g<`T>@^`9tlhiW9Z#ALXmh7pnh>NOHGpqgYg zyiZN|dOZx`%kw12ijuVVBk7pq@51!&>Wi*cFiBlm=`g~_QuE|0-yBIhv((aml2(vo zkbud$R4E}|=}xbwEGshLhybl_otfP1q3-f=TW|AHBH6Nx#YkJPm>`U5Kx`^8zM$S1 zf5RlJx5_3NUZv`Z90mrU zg}qn~leA=+@zRb62*U=(YU-w@G|8;R2u4yyT>qfjCQa$v4RE!4_GQ>#Uot6Fnz-2? z{2`7Tb_tEKL`cXiS(xf|9MiB+sv+@ruO##RQ;sWeBSjp1t6^~QVO?=f*^J?6{dZ0C zL;s!5(!$ymJYxJiX61|5It8lj?gG}Ay<*W`K0wEfcWThwgQ(PiGgHn6D_(}jZN5!| zcI_0dpYfCh-OA})!Kd#cT91=9w%Jn1VgYbdrteQT zJh;Rn)S@4sWI(o#`6N}nR%8^3GHR_O#VoDWir?;SR7lBcc>nOLF$XrWCqB_{xZEBg zY&tnK(FxTcsxT{o;WkU`jMf@D<9th|p<z0Dx(p)x;>-qubZ zQ?umI+@+Dl&^A#Fm_J_JAYu&*0CebS)!s@^P{tg!!ZMYRI{oxXEyngrE*d>AEoNbz z?s!=!S~~2U)g^5?F$mF&diw4g$H;KE=xCP-!%8`-E<1l&%>%Q0C<=sH<@b?)>B}t2 zYtHqYZ<}w{=pHd_V?qvI$?Uc2W7}7MO#c`t8)h_-GC)*i2Lq=uJq~!6yd3!O6iBpx z;bR#yL{Wn@C9|lsDE1s?Sy)JMR!9!^SBS#c{q)}_DU($MwpW=FaGhRgGUlpcW@<&6 zt(L?1^fiiAqgsyur$v{F!RDIG4Z+o(Kij`@`(;CDb3l9|J782uD7_d_zh%ZF(RKJ@6#_2Q=A!mK;R zbHQm5Ja(v}xsZE+aQ}dGv-@4+Wiex=OV0InPB0{F z5)S9c?bjlyQni@9F^{`N)z&sr-_qlhkOdSs&n&!@?s>$R4UWe>5Mj4&=vFnJo2$$t z@hx1T-^w(uV3hvC^?MSbH;Fh00JJ$rJ{vXAWvJ(w(8^-~Wu`_NIVYdL;E;QCz7hjL z^7+#OMAmI!N>iQMPp-W~Iw)C-J&z>GW@5{HtmIjkVTE9LUF~*5aqlH0D^yG6Nxj(Y zhe9H$craF6j|+l&5JrGEol%5x4pt~W>L+^l5?Q&#BOE=gQq7*<{LY#i=GA~?YC!m! z?;}R;29s~{?RCLPuu+NV>2WY|Wxf=#zplPT zRZW|%NSXJPvoX_rPM!8GRCh~+=2*d`72)tF9Jps8D^p2UVm={~5@y{Z)1q6Os1R;? zE!St^ch=i2K`}co*ngDna`w59W?EFmgb$<-?CJ<=0A>YxMol!BtT_Tcwfp?4jRTV0WEhVOX_2-9*!iZ;{Y1{UO*Yh_~n<9V_>QQgA<&pGCC*s2iwE*-~gnurw zk$*%T%8jH1uH#itM)i@xchk73;@z-9@JEwoz$fDZkQsyMLQj?Y)kKRoTj)`cH_}n? z*H4038Ve}g(lcZGPxQ_}Op=1ql4ug`)6A2=w)P7@dL*u%C0tRgX#DOE)|zNyEik z*1dqSaPjC8sFBh6U~~4~)Jf3rtA@}yUjs2E+E74yx;7t56^jgRF1Kwc2$2;%S=;fO z@??tX3P1A`nqq7MVIE2Hn%Rd_|5R(BW)WSl5`Q3Q=Hn#OTD-<;Ie^)lpnyfYs6(*byBZyro20}P5Au~14#j>^h>+{gIO>U#izo3ieyO`K{nn>{qt~@NoXkR zl2pl2hXyv(#c#?AUrHA{37Ir;M)1y%1OP(MDBAA7Dnu!wGyOSEeZN_ zLBQP554VA2Qj8AS)ZHU~A)3;F8)iPU!vX5*_CE+&W;)?&p6>{$GiDJ@1f);-Gg~U< zp`apygTu6p#2^Kl8_|+soM4Mea4fH0QSnY;$TuOrTk@qpZ7E+@`S@HD`7M8x*qaGH zVq;RE>Fa@ty1|*(IDmQEeUN4pBWt_(PMgBE*Q*ioO!kr209~L>~Grw|xIgE*Cm?tu-DNf*qEF>SiQGUA!zYZu67QDkr{3wyp1x|zdP1N zd)}C~pNOlY0mSc)T36CB77;1yr4i(0e5}Ao@`gj?(d%iMpr4&8>G3~ewK5^+zQ4^4 zN!sluoE7b#5cOE@(kg(ixmFcLY_I!Ql=wX*23Ku+RofAz(tc9Qjuy0MvC6*Q>KX1-l3pY+cLm44ayj9bWQKvk-R;fu*5H6bdmi^7Q!m|@`~(E$_yJ;p}lPsFxoNh z^FvGS(@Jr}u>|#)pIut?`9sjUBWu=IDY@ponq$U=#`ueva z*FojFRuao-(C~mg0`YD)cP`lI54YILZpnDhAkvAH`I}A^Oq)wYV zq)aA4EBC!(PiTB@DA$Yv?|yx=cpycM#CT>&>0opk%iZFc<)%@Kp<>nk@oW%I9a$X! zy+;toKvw?-1^dckwuDEQ$y-T1Vf8n6b>Ns8GxpDRedFMWqN+{g5=XJc*_wtYo~t!M zMVP(ptEh5CAe*^{);T_6FIyDbE#;5-`V1nVVvNOrNsf+iLj$`noW_Mgyo>U>niM;K z6PMSHu}PI{#V@(E*E!OOIbl*UZ`Js0nV;!Igx#qDm3v5yx+u-eGupj*W59g;8%dNO zcIK<9W3&6$hHE#Ne4$xOxM{|=776&XL5;SBgbfY4xI@bW8m80*>Rp#l8{$KIY6^3t z=4r|e&nQfJJPyE&cZGeUJkQzkdGj*q7LEfKz^Vnm+QXVhoK1`O{5HzElr@I*Iy4NK zrK%yl@9Ga1t2mbMoS)q=I;BjyzQ%`Mh?Ew$BKq68n+!9r__onet|)M$pdC++^5-Dq zfYFs^&HbL=C+?Rg&IJ69ra#ry;9*J=XlOCKS9u}X186p!on&Fewp>`i0HOAEB7 z2GZV~FsQ6|4YV)jE1r{%At%~Z2z2_yrdgLinDA>!bgZ*NHTx+2ZnB=;ir+4r0Q!~x zjga!}s%t4?BVAXI-+Rp!`{ctMo zoJ~(gta-kq=naI$XPOT~UyN~?%1hq-8F-TWT_!z!|Ng`P7xVj&CdB`h%zK#*(+&ggKC zZj-X!HyA&eZsm9YKkvz9aUcmu(8`F$UH@@)l3)k>&c^hU_2y{jK^O(#n^pNo$uO@M zBH?1AyAafgX|-UZJTKlWJs??|h0BfF;|qsQe9ho9Yi6avJRMl72R+n{c2Wgd7(^iQ zTh$NELh}?|lgya2PU&-jW-jup^e1DTwA3-%U{Y|I1}qQ;m2DA!Fh4QHO~Wv)JBs|7 zvWf6}O;Zu+g|7J9g>@BIt;)w7|HBk=bsQ)vVAYm~>*>-b4lnS&*FRuclr_0s2p?ec zp^o7sr3kR12Uu}y{^app_0N0?_G}o#sZ*nFuq1}muDay*N4zx&eqi11AO3gisf*T) z^K-N7vQ!EP@2Xu!dkwmyimLeK^z}%`-*D!5V$oVx-TAFO?`)uJ!QML%8SyMHkWWPf zQm`N4oT75`fwC0`WkbV~>zaMn!BH`!!jW_S{ML5wJwLvxgSU~C{~?hMeNNDw#kea$ zMBjodSEJ@4iA|TvEJ^_=)U9e{^o>R6s&@^p8X=JS89wAHM>Chd$A_Pa0DcI3teKKC z=LyAJegJPrhOjKfLFv%2q&igHw1Qv&>0F`o!;rd?rx_ov&Ka-F1^qwvBX@qNV+bH$ zw&P(E){1=jVbRFjGIP6ZFlB?EAVrR3X%dbkFLWM27EK@Z->L z_>B;^V8l;yZgrsv$$G2%XfI|_ZEnd~8-Uypf+z!Htfso*c)hzK!4H#Zzhh)81 zmXDWFS9XXONhtU4#*LU;Nat|w+7sq zWCl$kO{r#5XcQo1svs{3Q_r@9Z17ej05QOMjANXXFJe@IBUUhtk5;>h$ife=Y~JM0 z^h(>urDPZ>TGImRA7vCiBk~Lf;Q)-qVU9#MaT{4`erMlI$@W|t0bJ5Z>?w{4mE4Y4 z&|l|&lH?yLaF$h=9l?vZb-ao_VA{RA#w40>U6QN93MKo)mR~Xz(l)P+G* z_)5^yq%X)R)O*T^)Le*Vcs*!~Xp;k(AV)So4VL7NhMeDQ-nX(!8I~QCuq(hOiYbZA zy~O>ieF@>Y>D4g{VR#8R3VHA4j|U&q1-LON3e4(nu8F2DAq^%zBGMlvBH~(N&vwHY zXGP+ivtaRvphxs7AWTV9)o!pyet$FH65Vt8;JvDb%rhMx9WeG;)`&oIg;sims*Iz@ z^W|t6p8-Rc(tEpVtOs>BO|RjdEa@NW5<$*#Pn`QtfOnIqUK>~k;Z2Ad%NS7Wv;o1f zue#8|1lem|!~uVflInHx_%ZO$hs+OOk`4$P>-i)!Bnf{Y!)QL3fbFX#6;(==9B3r` zs063>ErBykX3n8FdvbRRRDd@`)ANk5I(iLia)E@tYuTcO`rlvW zS){y72fKIp-*|0g7Gd7QG*Ac}(^qiof+M4Q8io*|rH?F&-h7sQE$z&5duU$?7!)~m zRF#R+Mg z7e9w{cT|`I!FEhBP#*9rNmxx}nKlZn#x+5y>jRc)K=|mf)d+f8@$WQMl8Q^04-1gV z-trKw^UW_8rAqr?#ehx{*j6eY+-K^7(?mxgd0Z%SbD77Gs`P_dbQOV_yl1TYB8fub zC4sK7nQ@?*B`JKm%=*_aEc20lWv%-C3NHIm$eI{=QCc{nemIJlOFW#P512<{EIJyL z7nknXZ_zmK`*D`Oa&Z1t@%NG*S)I{J#PVPTLz8ItDsU~pSTgVJGJXh~xb#86VE0PG z#Pez4hrvXaAa|{|6R>o07KcdgB~V?n_7zzk^SG*@4(DYLu!h)Sd^>aaULYWveL)PM z&8O|JlTli^WbkHqMG&cedrbAXiJ0GuuD;b` zNO0+XlvpWun#He;;d4F*sK6hmsP?)FC+F^##tq?;BycV7Q#>KO)V!FCgTQZ-kqkPM z`FaJ2Q!Jz_KKyI3;F%+e;Bfa?RQ z_Ze1yur+M-dhV5JZ>O@ZVU0g(GBf*I@F}&B&nqo#6hx|;)Lmy)mu_rFjoN(a zrZ02sd;k^(mPe5J2-s2^n^O1J+gYzZ!yghfAbKaCPN@Wj{a=7Q9-31IbY^~EmV50+ z7(pe9ctoOK1v{1PoeE=xl>N3}dfno0!6xF!9~Uc0H5r?~n;G#tN8TAwg*~KXlo1^y zpf>X9kLDR@!1X+>{1w%NXxwOY2{R=|UuA=Z-WNaSn`wnvkJ$z>awh z7~%Jg@Kd@nEbTqa8J{=d$=mVlW~#b%-?Q-lJB}qGH-* zRDJZxrB=|$DQQoP=`kEwGdem<*Qrm1T1>eaL3+aK>RgwyyAG&x%V)gga%oa2y@xQ4 ze`EbzHdogRPKmbK6H__lj^gS5ng^TiM(rq}$u_uZnV z7(2=xwivyL>&Io9Nc2`43%4FT5so zGZ?ixhL-F@QHdxisx@Oh8ebeLY73FIi`R_;4;>SGL!UnCa%}@7G=@+2t3-LX8lUo; zBB=!;)j88G=kT&Qwg5{OE4>mDyiOA&6vKfsF2`EeQ}Qfob{WQsq~v;Sr zS|zjEQC1z71_r59Y7df7-ocil){f;eiW(~MZ6$*bUU_Y!=242}a==z}CBwApNb?zo z*o9HcMr79h&whq(bIiVK)O77lJj{2LZ>UV(wm~Yb3m7AsG7X76*bSJf!P+*Rhh|Wj z{#PnU(&oHJ({>b>;8K&AtEOcxDqDX;IQmc#h{^xPK*9#E$bySt(KiLEi5c^Ia*MuE zFwbWGo3Ww5K_RnwN*6jXYN4YaH61_F!Ijto2U+KIRjSceqg49aI96ndI~N%nNUl9c z8E<$@R4}vda(D|#rd9xNikV&RXOVz2-h5`J5`0W~Btn1&(=u2EVK&gVP3<#&mvVKm zKyqeFvxM>R_{$1Fq(tg#Xr8zpyzgOl{Zkj}BP2WpZ%CCtBac$$)9=?;$$|3wf@>}* zvcT^&t3<(aTuDomZ&JeFNejHQME=&oYIt2v@F$_wL=<(RS2@nM$ccqzCnf!O+ef2D zO8G}`2B2bMB|}rzlykYFtxm`Zmk`SB!gGjgJefo__$`51ypr_Q2QF4p&?NQgz!w_; zsY%E$Rizx3r8)a)qA${M#{Y7Wx|Vgk0_x($|?97(!%AU94z;fU@j_1gj7IN9lGUO6VXTeDU?6 zfd!;aMoX2R3DnN!cDfe8^zDjF1@ zX6I-Ye0|L6*Md1o49Ofx2+&r?(U@Lx^RkBA9y+=2J`Qa?a(rhn>)AIszf^mqQmp}@ z@gmkKMoVm+w^hAF90zEM1k3!2G)#^TP;$OI36gp| zTF#v-z|Gf0QypEuFXNR8$awKy$?AQslf%cn(04@Ar$koquC7)?Wo3+zqnp{^R9kPk zmXkSOF>8(QN2sJ>(4Z%^ThiGu0CyU3;1v3252&wwU@nDB^lN?$ehuJ`6b(sQ=1g|F^51r}t zYACD4m;Ef;eAJiTGj?Y4@*p**4OU~;nh&*q@sOCOwxKI$1!d$b42wzXY-pfY2$p}L zyVIbug_e_UKP1(7a?=v3M6Vu>P?bQ*czw-~q<oz`L~9@8r2cZ^?Woj zJ)59R!q0RGRk5ecldAc3*YKEy-t28Sjgy6iS57CS7$+Izdzrj5_lYmljLLTXH*U4j z3LVGOuMTUgI-?9n{g*v=zxB#?^mCh>=u_tzjba_7oxW$7#Ho@nd{YYbWQxW4&2f30 zeDTA28L7H}ue4zzMNL%g9ll%G|C;NacfUuZymj{J#jw~*s5PHb^0=>*GiChP zk{xy_W7*#EMOC%z+Gq?#@4HqKlXLolAXmi`L@TbtIu$3a55WzCax|_fVkoeS(U8YC zS29O}D$l?4)1-JzH0xbMj4YKz`ZzO>e|(%79+bTP%eQ&KxeK z5D?#nvmGp}f-|V*Vq5K)DTxQZS$PCf7e(x?kqB3a*YkK9q`#KqM$|p$w74|^A-mzD zui=lo#<)SwVSOa~j)8+)qdf4vJQervJCBvkVh>@wM*_;ncb|Qp*E%%2d{z1MRimVi zQC4MNCTUnE)=L)Y%GsAcjr$+z)Oh_SOFzGFI8L1kh3U^Hl+Dn6ilg$9fnj%DVF0wC0p1DzC2Y>CM2{grlJSF#G zJZ@WE)gf>kN$PSh-lA9)Wf%<9GvDqWoq$@u~c_8EW!&dHrsXQo+&cbtndXxoA&_ zOumbVPpNNsuz)hOF7ee!S5MRHR^t%j-7h+GGD6oB8Uc?#0^zHLc3L4u) zqLJ03i__;2!TFxoY~u0_=74lU39rx{Mgmz!`KS*y?Tj$+I!d6%2u5@iR2rZ16Rv-b;TT)@%OgMIDDIU9b8T`a3Vpx9IvZe3xjn%?LEDv=lbe z*5Jze*h5pM_Tu{@lL7IoSE?yUb2CD{SBzaWNXwP2J%8|<9_cfO&+;dfzPGoWA3XgL{PdI7EMxutL9f?$Rr8}iTj!%L z6XB)XY^mp4+gg$X>%A{3MVZpx>yJh5!$p?JFOd&;XP0WRz6B>kpdDpR7&9%6!?Z{G z@(}%BcyPPAML8WRUdi`a`@L3WEV7ZEk0Z?bd*I{Hm@Q}RdtxYR^{$iO!>&b;piT&a zE1KyZQOZSEKsgyRLC2RXBV%$2ux1LzcI=hgTB@~CQG~1)6y9!ZdbM{`qZfCLsn}0U zy7i;sDHAaxEu8Diuhoy;+EOj)`=(<3W!j;r4hId4soN|Q$n>ex3rjQ)S%I%5E+Roq z7I#bgYu$JPm73jOo~joJM^v?v9rhUp$~(zsm{Ey-QT{oF(1~ayhyjeYVkc3vt+2pW zgms#BUUVr*9lh!`Q_L!Jww0TF>P|2^FA7KPH1rs}Dhcss7JOZ_D9az_u|-S?m9wF6 z4L(OxHvaWL7{vR|nb(9jN>?%X)7SUTUMGSLJ#wL`o00P~`wcOIyBEVvAD&N$j>G~% zsziXs^CE424xJs8<&A4Bbuyn=%AOT6)<{rVQ;BP*e&oo$-X|=kXco%)BP*l%Q~6&w zb`h+T#$PkDZFwX^>%w##CYd>0J*ALSqj7UZR7(Msde>4VLMX|M<552NFK-FrX$OBb z$d;Dq6~6qXr?-1VNF3=yai}CRVYSpiGaY!gbpD9ql6tZ3M@v`juymHNFp5I)r(3(h z*a4|_PRQX3^0eXi%j~1~;P32uX*O`3@w+zDM#pHOkiZ=Z7@Y;Itjpsb=mBR0W5XHE z*}SfxHG>-HpBP{4c(W4>iF$&BQI3A9+C>P(>ur?m^J-xkrOI=s8ySjOk=9JR!Fh1| z!5`#mlxWqyU@~c30jpj~l`Kb$e|tWtV$mpHhRg*61w?qwybH*JH_*z#==0iD>Pj_J zXJRl_h;k&Wut$F9Eze12W2pHS-H&sNt#U<;xgnNt3DgdqE%z5`zTH6!IK-sr%GeJ5 zw<@)B?*ooI3Hw}KbbFY*_Izq$weB^jYG7%ipFuM3gY!18Q^aL^ODdtMwbJZ^XW@-H zVEFFXsGs2{g;~H9%PCOcEja)>XW7N`ZN~j9_#$D3Dap$`sx(C$D)*a!6NLBT^b)N9 zl~_qTudoTj0Z&Q0pjo0hc4T2~!VtX>XdgbOQ&W|<`-Ur|DOD?ar>%(Qrq8gg>8D10 z2p_pRr>&5|PbDa&I;8#b0jk1@t9*Vef?mNb8 zYP}8H&gI|jF)NDzz*y$KEdc8>yejU)X5{O-pG*|r7MUI@Y_+^u=35p2luFF^D(PKn z_|6=@El&BMEnvvS!~=m270a}M@mTl+5-i9@HEff=7Ux9E{B~W=sLH4{TBtfM!x&Mj z{D9}Ar+HxfgW*B=s&|0A_a$0&yhja5D{t)^+N1EKI@S*Tt)}I(J1o)DIHU-5q>7gr z@wo`e9H_4mdD#`GZ|{LEleB~Z=ss6W_%xRpdx<$e6(MlI7z+kboJkDYktLCnK8$bg z3+ug=a>)zfehyL4NKCElXjT*zLB6crwyT)y?e4B8gt6!eev(>_Nv^9sj72Wq=ul5Hf)jQ5#mF+h= zD#kl@{V??Vkr|RC-~Go@@abq_WSVEb%+bcxzVGh4>tqpItZB^4M#xC4j~HEI@SmNk zhBR7K_wScYEWf^e43bi@d9~`|j`8~5sTxtVY1&JZy$^cWsBby$xK5zo_m(`TI8j=< zd0e^tuFh~hni3jp9S1T1vX~U(c;4>-ber|2@O2RMzbKi75jS?AOO+V@BuhF%Rd^$K zGkw!=`6aN=p`MRz+6yjQy}p{%ypn32G>Rb9`AD8J$h6?L?_}2BsMBR6vL)$r$#lRW zQOM^XKZVTe`mFM<*=JZF8aBU6fw5z8=r6mPc;;chON;~9I+{!6sWLIBDTNWmZVgs`WlTm7U|P#VT4$E+>*`mqFdp3D}W8>pV10` zukxE=#pnq6bdz_lDrFlve<|w}z1&SzUcc}M5T5sk#f~O}TuP&;@fODYcMrs6h%*#K z8Hn&^Rl9u^DDFRp6svw2IH?PJac^TE;LK0+i8A7fKEhgI?WIve4~PYiL(iiVU7bmD zd=Cp4YZ`XYcmTh#;_4iBCJz$ z&cxVWK%B?mro&ef6QOcsmIhPUByYO@Ktg`zjxtBFmwEHS3-?tP_syyAx~>KQ89Cf# zmOjPM0uaX{OVM&{UkAMPlZNT7+DUrH# z48Kx~pHw0f!O(#cIk691!#rx7%GMndKB$J)!P4s$T10pq=Hu}ge2+HwQ)t`0eIHFC z1VZ@gyEoB`dG&CMMK!`oI)w0!B-ycYxU&>1rV&XqohQGN4GFK^D#=)fyiO1~k4V$u zAGUK5AMN%1{%#brNa)0ge|UN9WB+lBY7J*)EYPExfleWL=G*SCVt1~PiV{F=sH#jb z`r>#WZ+z=8Lo{^?8j#h4PyFl~Vl6UYwvhAm_%p#dQOBGNc=jpQW&vZTWseuKiYWfx27eLko!SLdCX1||4V(f|Q zw{HZm7hml2BW%2Kbk1n!i`QIJf5Ars0Br5A#FJvW4}1h?%HAaGuE=em(s zbea7xu8S9yE)x!M(y4~`cG&D(*pljXfprjWo0D{pGMT;a+Y(WuyG0S%jTA$!r$5Hw zfe7tuy*k$U4|VC)Y+B;+jqEw&eND{1a$CV?FE zf*=OkE}l!g9rfyW1W83zQq7zAhWq`-Y1Ol9lCbmMHdm2uh~|WYKH`T`OgeFv{%gSg zZ2-RMp9>{)snZpx&asj4{NY>as@3nVaQ#GwIqDmaC%?B3a^9$sIcNPzoq}8Hl6>zJ z36L18Cvl`EibwHH(jt(e6$LGN-xl=BRwv*>ZnXbs%*mf}GYLcBbgWq0&?_kWEIdx) zqBqo}0`IQK8*>F~BxXn?^AEpu`>(-Fvo8Uu^#fwQ+M>w4%+lZXN7=>Bt)F5QI$}kJ z=r)Le?tIS}@U24|ERLmtqoK zx8XJeAZt6kLGI!CTrVayTxApsi5pY*mZu${i%_M{p}LFuCPhD}g&J zR0NSRGY*gPh1BpH$*|){&+?SLglyVVGLKP#jNMS1H#M5^+I&MfS+2PVs`!{oyUyWV z%q4P(IrVtXJmP=M6U) zJ0Y|SWW3k<3|m&o`E5#Vs1n8M$l>5 zgnsS2=(bUpLVQXX!tGd*>$Y{%9nLwvgxR*;DvKPiukKnOKmUUHHBd=U8)aT1sGIx^ z6hk_{p!$>EFH4;b+m}Gy(hV1et%84;_7-Q>5sP^?H9J)0xH{!0f6D8)f`q;IS@$a- z7W@TSzJW0}zN`g&Akv}>s`>#}bT??6K;wQi;#($kP_Pt*stKdwP*+x>KI}FkE-d}g zUGAkA;Me&^1TtsPw?nl`Sd&hux^27ARNrS9Gf(BdCv!GWBdgt8vGcK`))Fu{|- zeQ)9)*YVw00V37v*t$+zj?_=IJb$x5yo5FdGSsb_TdSBy4OF)NJ7Gj@Biwqx?`8YaOnYOBI-=5`^8_)PbbmD;FTOsIB9(}MGcWiH9s)wK! zY{geq5RD+WtuZqD?ViO3N6pGQ`dM7)yfLCyAUp)NF>ZACNlfY73STzT4^0wdOA>^f~44aZVTfiF#q#-s5!GM4&YR9=PS zdRr((mIV}q+q4!Za%wUFpEJLWthD6|k-o3dT`XSWWBq}@P4g)yxc1HsdcK-Hq*h07*na zR0#7moU#tH?SqIKjU@!Z)J!kPWyO)fD=+~lP=etxr6GdS@UTO;3VrX8L%IeW(hYzL zLt-L0xo{tW1WQ(_XIPhB91~!mYf9G!&RU~*Y&@yj#~E^BX`UJ8%b&@;Qn!KhtAB(x+MM*uo~cKQJ$z- zG*FRXXf7t@fdW|pt2}&1LA{dh>cd#dU;{f4ya3IjirKF|f|Ld*zKwPzg46MKr?i3T zk0PE<7+i?V-477M3g+V=K~N&I*9N;;CTRdp-(S~CeaJC#0JiW-Hq)CY-#cBf-q04f|) zrltChAXN!~B17VHxdNtze);6`vCoAxAffzJSw7d|kQZ8A(g=xbg4#-m!mHrx^`Lme*;X5y-qEVSm#a|@+FcED>XQ5}vL^McIa_H=V^wj2*_CAp|I7(20 zss@8*qJ3$ybfPX4$shJTI)gCz(8~jPzP??lf%ack?NV-u)sA*_D zSv-GfxAG!Z;5w)S?~gU&*u>u~wA#%MSfSyYG&KdND-Sv1kg5jNXwMqcPJORAJ2wjG zkuK*Y(TlDJQp8b!)bl5-IZoOJmCnHDmUD63Mf!j&D$&*jUkvGmx8n%_*S>a*OSzpM=-al9`-6X6o7I3j zyW!P$43gBIlTJj8usq?IBN?++h0XsB#r*_Zou&<4@=4F5-H^8WU9xGZC27$aLpx#= zI%pASI5=c9_Bu#chZ7a(eG7|dkur4_cLTBE*JTB)f@sQcqTAu~dM| zF%j6JOsP|4{@<^pDG-y+6Lg$71gYThLr%q1=?&+>rlp7Z=U$pmXnF-wp!L)-r|fPM zId2U>tI=}Ah+`1Ieq6|)MR}1caIbcD%fWcXg1ewm1LminWCg53@S|-4Ltz*iEi&il z^X&B2BxETkyphIWvl>CiH!wcm+96O8q^O=c(gKy=5vOzyLz2Qds{1G@XvRC=yv_Zm zXjhd|4?eWdDF<2|vUw#OF`D4FiY6NaXcanOvH$ti3I|FBH1Dcsv=WnYPl1Ki)->hR zbx504H8u=<6-%Dx_L4eH+C<<5*aY`kAH0%`y=v~&Pmp3_#EC9=_&xTVOYV8mDFcYl zM`?5q^Ldwa!xpHg1L!(g)QfWbFjemCcV%2V4h~ME zIObH6xuo>Fs5w$8{p4ISG50Kr0$Gu)e5lK%*!H6#lpDB8wQEd3g%dy>z$(?$7gcek z@-UeqgSnC~96jNbd%-xJ_)(bxDI-Z`uO$$)==M(U!51K59qDq(g&Pyn1n)sgSq$2% z`F8e7k=dN7(eOaWQ4R+u4jw}b=!|8ls0VD53wS)UA5=oQWmK}#bPG@o2o7@$wV9O{ z=ob~po+G*_n)OnsAIcF>(~_##)J>X^0w57b@?l6&k38#=dm&AQAYD-ya0!l3NVhbo$&&kzLv-UIbwGJm^u&^ z>9IXfLBTeqz1FmvYWf_^l}uAWNm0m@A*uGXBLe-HOSWyW;tsOD3Z;*;-%+}%fJbgM z!0Kc&0LhU7lxF8p3aoISfA(85O@5OfD3Ec%hmDr0J|!w=`~_C_1^1V9p;6bg^O>uT zj86xg-3KqfJDzl@wWA1B9OdiwHFGZJYBnf9+AApw4jA2t2y*-5F6HRr!UB?P*M%Zw zv9ts&4`ijewkLicqph#SK~T+^zz>shN`Z_@RxCLypB1~@ZNU_eOFdZ~A_@KYn4(Ro z(_^~tC$pp4Zv<+;T?bsUr_G^kQEjOB`GQLr z)t(-kg&<)OXk5w)#TGVuBBSg_?_CC<3S^CI&R&jEz9Zw3l^$)>A?)*N46)FDm9BAa zO``a!g7l#BuL_R*fpM@ZRccIKzgnUi7!tYXDYp!v(JT#uZkCZ?>W9o(TTp<2RRiaX zeOO0pXw)Saof(%3tXgdUP)+buhQrfb5_>B!SXyzg9FmncB|2(eHh9vE~jeR=6pWAc|1s2FvEb9nv)pc=;HJO*1eKsA|C z_WjMe;+(8b*s<<@(k(q$CyeS`drMe!vIwiyy62+p&^&A<_!^H1U?%030$Bqq&PCV= zyVZWwYR1L4+|p=KZ{z5TI{pyAilJWS(lHkG!UB4j%#3QIUc~l$`RjeMB%%woZb`R-6HDEQOJtKmh0jpF6Yl&(Aqw(Lt<8J^O89jfh*xEp;d=CMO zQt=4Xxc6Z$mYo5MgGLvCif-$iYc*3MlT;dM+_s#nTAGv>3S=Ct_!R^zre>}~P~wTS zpdpu|xVewtikoO!qUwP}_3-mn5Lz!rG-P%r6*dJlV1)r#4Kl)y0x1mqQDC4 z%7vhsr<#7Id5?rA+TI&stEx>6z-Pg?!mdP{mOiU!(nSAOR50f!lJ(6sD6q0= zE68E7$h0_3g-dxnj))W|;r47)e>tpx9K1>cRzD_kKZ3ysJv`|gY&XEN29a@z%`!-$ zW~dMMoCni}@^Z%SQsrkkXg<@vA}e5}E3zIU^R@7v86bn2wg|<2f&IXXu^Z!&S@x<~ zdPWhGLrGL<2oa249()>6DKP4$Vctko#aHR9(;({h1Z-DNz#hrp9HiI{c&_Q`Rxhm0D7Cz=UA!fmEuETuD{v_4~2x z{38gXREN1awg)N}T5aIEddXbwc}}(PQftlg|6bNH0qY^u`LR|odpCfAy$PU=82GL+Kz=A@o^b!|%bC6?eqLdH` z_}R2ogkfyoxyPwUpwxjlZ_mVn!JgsuOJX=IW8Ezh?dp)kC{}k)z{wfl<*i>P{>GKK z$1BMwM&<^v=2Vr88~7wvAHJqCE?x#EwgJ|=9Vo;Mi43wEP2x4Mzb{F70+wJmo`FKS zyOcGs(w|U=3ceD6^|{fQT#V64DT0-1+$G$?8I^nA5tkf=tNId*u%l#Bx`Ehg53b~} zG~>zGQ9Io(tr0aWfcBFnSDrrZl;uzb*4O4=9T-VM)jLJmd5#;wii~VaN(D&JO<~re$0zE&Q@xz; zOiPnNaW)W4d$gzyvj$eA5{XR0Zl>E8_N6FQfdl|)%fZ5G{zcV(X8ah|Jyk!?VG;=M#P6yQ|Mv5w>Z+GZXu}!e3SlV;He%?mMvlEj zPJZX>GP3t6>GU>Ou_$F=|UQcm+3 zAV9L1ThJ7r&5lF$U*CX3o&{L>ah&SUQ`~K%Qz*F8?Ga~aiNwQg;tT{O5gC&NT-TIG zV5+Ezkzoloua~MD-zg_=|F#St+$+I#n+Q7(t_*(zm*Mkl3uPS0v^izXii8AV$ISMkjRYso@ncHhN76}R zCS(YPiU10<3I(W(B%}fWQ~^*5VIPFFhutRo?4FQ;- zlB5L1XVjB?)(RCDqIwQUl%k%xs_2@!Pw%EWr?1UtuSkJ+lO`5jAOWTd6^#PK#;|%r zzsSjs5phFJSO(;;EcZ$!CbR{iuGHYkck($akNK3qp7&-QtTdt4Lwvmggzds8CWK0$ zhgI<>Ms`LN>Lg$ahFo_E%(Y_YaO0yHaeFsL&Q&m|7XnbJ9!;ZFgU>xF!~0*9cz2iL7@@Oa1H0rZaRCt3 zJT2Whbp9Y7UE`^O9DM#M8G7LaXR$CC!EEfrF^LTg zsCPObnYltC@l@9;Ahyf=Ft&+usD|lmv8*mB4NDg9Y(an`I3;lqSCSLjVjPl4#3O!e z!4a?c#-K`zxWyf_94mqe#4cpNMAjz*N4dQ>!6Z=zo;=qaObtYb;3bL;+ zBE7w1N_7~5q*YzzlPYW>0GDoJyjLGvPZ!?b?1NQ`CUpR+R|Bvpmv!nGrc&CJ)eL@=HkU@tHCGBVq!goUw<2&1b z`d#V%<#*L*>e0OGQ49cn@*lGL^WRk}#|iH8d=@}8cJc%qK5mmCcq#HZZUj@lJ8NJY6Xb&u5-b(7P16v{s9U`0aEG|f(Pv09z2v{Zroey*3pjZL1j zOW3rEF|o7IRyBm>=?`F;bFNRt5oy6BG9`S3Z|QPXhKXXzM;D&GX&CEf*+| z9ed$X8GQ7Q(vPj-4L3`}jc>Ioit5Zfwni;o9cXvJc7>qEW%Qm?MMk-%PU+GUgd)&6 zDIS1RFai}AB&QI-$pYZ2xPkq0s>JqO0rekw7i;R&b5q@& zzjG|7PV^)s81YKxOSJD^n<@G`Xt(<_BuH?-=x4abk-xFx1L7YtM zs+!hCwPzcm+d7T7!^wb@IkZ|St#uXE7h4_RWEWi3!3QBot4>i+p9s5-yQ&5&$d=3K z;RBMq;4&0nx8R9ik_d+tn{(AI8CL;z-T4dEuF2q*f9Nw(b>*OCd&C*Wg8oi^t*Nm>fJZARV2I8{oyV3lu~g(`m*tSuajbU%*jY3_U0Yn^m!YF$4{@cG2DwPVwP5 zjZfs1<_mrqI3=AdXj3Zxv)9SsAMR9YIVCP?4v@;6Bl>7xkNQS!4;nk!BK;}#UGIHE0>T7rT1p-ufmvpVBcsO;GdJL@gOJ3BK^C?l7A_rRVgM+r`DlNt83@VBwrXiUwnQ3RtEINP4C+2Ol;E*A zgXBWJrvMAT)APX335orZ6ZQ#MrxFQ@=de-gqpd0c`sP>uR7G0QAKhe~;JY3G7eVX5 zp?-OBPmc_6gpf-5KK?&*DG0Is1$g z;6|`iiC_h<#rWWWdUyGbi{W}-ra(^V)Bt1zLfYb7l~s!IS`X?TY5mIAUzP&qY#9RY z92|nm4Rqo;N>BtSX9DxF27=PmUM0;fbqb&=@tfX!ijTNqc}k(uOyD%t_r*cYiBw=E zS&8Q>Q5e#D;ytMw3TcNAu=*#RGqAoOY+3}Y!$sH7`o zOCX8%b;F}AR$tjLllcxY`;om{~u|m8_grFqFC>xv~TFC8q(r2lkXf6t{*gjg@hULY*J<>lg zrT~jK@VZ@7PQi+efMWHl;ZoTJqCmzfy2db7MN{|P56Ozr5f|RIrpnnK=h(mrR^Z3d zgXrpN2{Riv`kAQ)2Bqv)wNk$MJn6aphZ2VyxRP*GPsWe~lFiVI&&k+etN5!|inn1I zRCQ5_4fLy#c%ttFfGV!ki3Fv@=&*GD_wCa6n;)t+Uj_#YE5ICy1j1n#PuaiNKIk$} zuR^-Q6Ub=GYQN=^s;^XC^GXYt09K0VVD1E?K=V5FS@EGkFdPfyc!FTb@(w=#q>LVV zUcEoiyk4E3iczafa`13nzvZXa}F*$|96@q2o*s@n_t52Y#0oRL0ARuODL2WGOG0p>oAd!W4Ju7{%gE7yttrtqwg;xO0s6U}1QOPkf?q<9fdR{^Yel>_Y4@0$FS9e>YWYJw7D5^{g!*dT0% zF0zD@=pMfGB1qMD17P3v#ES!%C&m6F_e-b!T1n6rh6yX+fI%-xFa+b=Dj{fN3u zdXi!Kmv4}!_kRR)yUHg;u^{o|3NXu7Y?7hJ9+dd?H;EhfF-7}&rTxFYB~gTr^4F}A znwPxl)bKe++=$;m*2#HGy#Qf1-n$7AJ~n%x>T~-sfP*~^=VFksMseK_prSZ6JG(`O z3rti2T57O>zT8^5CRft*s<`k@0 z7Se`|@R|@TRuDlsM%1G%92Q;T_Y$o75yfHz?&~#zI4MPx*q~`355o((73)<6z?pRi zDA>3t7Yli-wo;PG7e=GxL8~Occ&iE(J=TknV`&JJ2f|P}V?-s&`>N6n-brSV*gAs?j9jP?4s^oTfme5-GE= zgg2;&G6bviTzE{oE4`L!q9AWdzrL`y6~oPji+7S|TXv$4woznPz8t?0$0;!0Ej4h}k`4}evPF$mp$ z3px=1Ag#D7t0i>ll@k5MtupXCsOAzeiFTY6Pg#RRhT5g!`VWbl?&Y*GLEWZoLHO|G zP*@sn{s(bFmFU5oEk?xf)YPdc7gWDF(!ho)JOzwPmWmJSW<^>LNc9bGQ&JtJH9qTG z)`({y38cLGCUB4xlXqZoKzb!oitG5&$E5>%&1?x6t3LZ339a93&8=dEWihb~Yyx4F zjs1EPCM~ab`(zcQs46EUP<%@q*EB$}e<;3yKQzTfA%gkPj#8h7tCN$ud-BP7{pKY7 z$za7Y)4ogT4{*xnv8^!gfz$c+UvzWtcom zrtev`SqAR9U7|;hzz)?e6&E3N(YfbaQWcm#Gte4=%?J_pdrX-j;M5(Ae|5EWT zX+Wa~dt^j+A44+vlv4z`F+BwpFap%7uDj8SO^5udF1%c7u6l(8Rk~+Gy*ZQjv2k$8`>wH=+y^^AJt#t*_Ee2G zlkd$}UbIZGGp}-WDF%|Sa%AJZ9H_T;I^+Z#EMSUM9X%WG$$6e=lbj$tc;5pMi=*Oh zY?fH-e$_@-eDrfr%hiFy!5T$L31CD?%8O7{ zWB|6Q;r+XXlZ6>6o^J32!Qq1kBzEvcNrrpGv*9dR^~tXYmouSMX3eMC!UoZ|~gJphaDM8qEsQhEPueVkIHkJB+{yAgRUB+ zNZN1xB7)tXBP$U@kZ$!T#H>A*I4wsw>Thtn;`s!eh@_@Nf)wd9) zPrg`%IDr{1vif|=Oq8ky(kAKiTU1@(#p^gp&?H3x)&nOEuU=Iyjmr>t84Zu;@|7ma z!Z|F3{^%WuwepT3m)wbUT2|xasURRZEKCkw#h3$q@CIy;SHnfO7jI-&FKn*P#!H4m zFYJEIuVOzh>@S+4A%GEAU3{gQ;Om9MMcEl=BVHdS;A+?N^heNssnVnB&-ay|NZ-SM zlA(w1R{>}fDT$1IoyL3LSr<#_+#OPV*)=KvF6~x~r0!e3LRNnCUu58khh^a5du0U9 z5uLTXrtp*>KNd6U)zX z5Tv?5n+B*NU*!qOT2BCu30RT>4S+7^>Gr6Z-^ms;kpr?_(YutsC_K(|ih_!#zZIGu zNE7@vN`Vtgl<5)_Mt-C{S1dl>5>^X6R$$F6p37i!ninDRCRwqR9CfjA+zaXIDo9tC zL%JHo?_z8y^avb9Ct2wjl=Hp&)!eHr^g~cDcZm&`423HK-}R(5;hA+;=-)Mwg7_NR zhPb7`i1w)&dn^b;iHxc_u4@$w52^?jN>vC)3`$UR%%BG#(}X!s001uBNEgdE!q><8 zdmuGI4G480Cm8ctdj6>Zx2g53tV}GaE(TZiL+^dND&Lx~en$b;lxt_Hhvxo=^-7@R zV3O*-r2xil0I2m)`BlIcMUcXpwkpHL_FuNE@(^rrmT6AqnW;Z4;Xon)u(jqovaAFr z#sORpXM)9+JyYehOFjf9+O2|^8@pw5jRLQHl`U_Q6+c)vj0>PUdKZF&{RYxiGpJLH z(iJ_AV1bZh$&hSLltC5Pht6I2&K7LUBk-mkMUTfy1jVBQ9Pg2;wBT2X-^=t&PlUufm2j$+-dK#&rgCEcZ- zt(C|Xa3Wn~lQ z`QhAbU0`A^nfb+bn771c%QneLc<+FS*$yI5MN+JZoh1gNdq<)YN5zw(QP?N4z+7xY zA@)pe)OpUem!_VfRFmY=^pK3WBnS7LD@9;Yc}RAP7yf^i{dX~>4Pn9M=OQ{T! zR|xB8aTMDA`H#wITZ<*-r2u8FG-cmp9<0`GJWHT`l{X|e1#4tGJp9TWSVIfvHFM|v zv!}V;rSi#A2SIESeAB6Nu>6#&=mp4cJ#}UFuPK!~*mPU~^ONnv@Dk+Ii|nXpXQO9c z!HR{`jbjI-s}K6!@*u49#TuWS{m%n^No%6KD5NUgxJDsa@!ND=>hoq#qJSz2 z+Z?E+Knu^D_WX3uVmJhejT4mvh-N`Z>;C<(P66W_Dug*x!eD)N0Q~4EaV?~%HwUZb zJbyVK_e5(&!c)UVghqE-id6BBC!cZF*oHztK|mArkU%6 zBIGymn>nJP*V9{Rz{3gMi1J&q%)gzTg)>e1l0MAn$DA2_c`oqz;f|rT8WBJ->B5atRT&P9RHdmt{hsyVhB~mPcXWI$tW4L@^l!JWs>tr=_-Wt8E&uE7w7pe9W911y6aB_6tEHZ$dEP?H4MB z%)Zt0`#U5)hFB7C5}7LT9$sN!X$Z-3KAa}r5Ui3*0u`{WS|CN|3C-lE$|IBK_?~R^ z{RreIoiF#?8dA0@eLj=lQEYipnxeX&>OiJkOII2v@=jMuS4?lL@X9Lak1>T&bs%-J zi2krsFq*}ocpYK|`ci;Z*dLhd5msSSE6b&)bOjf9PBgB?YQ=O`8ouv067B4Omm!p0 zX_ZA#AV`hk9PLtDFoE}tfhxJuU#TLXa}bhCkeWalc9aYv2%RcMQ&B7U-IS^*QE`Dq zW2>4#`DuUrfA+2eu&N?!pR|`=Ng*U5A@nL(KYnH!m+C2?@e2HEM9dh+fU9)%UvUf!}}?Gs3o+4fD}^;fCpZw{&Gdtg8y#)9G-08jg~#fkj{$ zECXSnj4O5@JJ=NHLJEM3^gDX0BprE*&RtpN@$6$(Ya>We3B3dHG67;|K-G5yfGPt% z{202*3|$lJ;XhOXav?AcE7;S>XUK2T-0P~k5EFa2q}~5{N>g=p^|Ea{y*~O@y7C*a z@+R}Kn7cX+D@@m;74=SIiw50^biFGDP}Ml$>rI0|I`4XQfXzO-r2b(b5#B#Ta_@Uq z!r_*w0UJbyxIJ!gUdn;<tOaE&#iKD7Vrk^+>ZXv)5$s77kB)X0~?ufU28iK++g(s@mMxU9n%s5hFjbA*Te zV`Y^qPRiiiHqZ?%Xep$_)h=qeT!3qbPEr6swEf-JW%EBClM46{48iK{5O5lLLZ>~~ z!2lbXv+deeGYd*c;SwQe=m2w}Ja3Be8>pa43Nx_#PziVVE;CW#v~RBB<; zUVt_?0`pSSBjaUXctKBqKR+s={cX-hL9+{hiWFm7)o^pBY2uuN;r=kHu$|SCI-Lu} z*!zmX@a)c|w(GaaX8bCc(GOAd?6`1w6Rr_T(840{z`YdGukyRQnW1zQ2}6sOPz9#K zS71aCTGjmLpwm(AO0m0*UAl z5E*4=c^Eed9qDbF1~Y^fYSQ36lV)cK(53G-NX(!i5{;t}<;(;TkHoyx5V%O9`j3*7 zzk!r@@jFr_-tT#zQ+BGsTotVV$xp~XsI}<_smhM91RX6<9*A2`q~PrQ8HRWd(3c#xp~K1y=!HfcgSjL&p5Fb(ceg?*-U z;cnLuz)HIle@QNmLQCQKq6!`_4u!XgX_%K91rHSTi03!+Qk^JcdtW73v9IwUUrEPa zO;j@XCSHQ#_%yJYQx#YZRUq9kG5DhC`<-NEsBZYvq0tVx3)0oCaPz@!M9LAtJTq|v z^_P-VY>-P9yV9^RT`7F6RYEOeIoJ%8GO>|Hz<>-(?u{g>=&v0 zb%oT!(QJ5Dwr&>^J#3V2tk}dUZF?$F%W9QZTXrXrnH`$S8Cv;`g=ODF>?xIwEjySt zL5h^@1U2z?D?!VFXUxATgSpqrE3MtuJa#&haBT)yGE$WjBpv{uvW+TQqk6~gvx;dI zScy#<YyL3t&-G6TRD(9u5o(&n)2D6!;Tt6VcBGLEHQ&hq zDLR`Chgy}6W*Z<)cFzWS?sC!nJ_7~?>WG)pE2L00mcjQs8aU|bOyKKPn5!Z!IWHFj z;&1^l>#dVXcP3lC>s=7iNLFTi(wznLTd-oG+4FL0uDwkqEIOgp^gtvsN0n*RP{hqD z-RZi@N~C(rX7!*Rme>!b<^W9+l}BqqpYges7GVOugrx|#CSwD{39#^^=Lr|?IiX5* zZrvhbsc8~~Ek@c**L=_W+xiLXJLa+Hm+^FTf__?AA+EAgaaB}+=1^&~ta#=~&#<;a zG4$}f-1-UIl}>}SL4wxbCRMDtDyjh8l-C`Ad1afbw?(;*J>%eqR)O3?gy!ts9aw*U z9SB$q2f{M&F%afTY$d{duS#*xJktmP`iH8(NI>#3IJZr74UkcxYoWft{y?2(`Mx0t z`4y*qlH#Ct+0sb^5wv&Ze-3Vi4mCq*v8wBob14JuK!-s5U|13s|M;yG0;n8ViXC_2 zX_EGvb0h@YXfUO)q!h`2^fsyd;6rgvoi2`)H0yihmJFOP!~9g)FAHQNfQp?Za<4-q zZN>$N)wh`bR1 ztJbWLyvOd7>UE2)&*5+SWw%Jup~vX8>Q$K+lONvYyG#et7&fB=Y0ww2zg7~o4ky6T z>_F_)Yt2hhl7b<{_5!IZ!WQ@alC5+o@i0BSt3MyAOiavQ?LB@e?GYnAGS5-=IWTZq78>vaw)fcG* zW_`fj+K>i~C!B@bB>(=qG+y*U$H{;Tu9h(Fzzf5L`s!-k5HV)tIGBY~r3<5mALdEi zgh>Dta9aRK)tXhZ<@IND2Vc&Qaf^9wqNwKi*Ph*^Na2G<%jU;tNqFW!i5Wd!X~7b6 zF|Pp_f}9IdOzo-vJxGu!2&4*M7LyjN}FhY}(Bp4e?@;yp( zg0%Roo@GWIQbpJvZWbC!t58qzJzs6l3HuDyf1LZJy(2-2rs~Xndz(*tzGKHRO%mJ0 zK(g{hXwrUt@xk5s1{)b08yntji4l^Ij!4@%sy}*5U?=0k#YJ^lP;&;3WJUTD8zLhb zFh8O!uX9y`#0XBTTj%Q4QqK(vJ3L&Y^Z}D)jOwet`$XGL!r{m1#)MQYl6h;U-Rw)5 z7){Yz2t{)~mYVW>aSRwH>E~P`q45a-9JeD+O$VmCaq;kUo`KW=Li4}F&cM)za$z&_ z9o8>J%d-8o7sXjz2unijEC43H{q2`^C=v{bFZH4uB=y4U#kqEg0)Ex%l}HPtRxIYN z-nh=1)8d>HB)QuBO|tzj_sH_wFP9BZ-6t3BsE~t-$|b6xpk*sRRu0)VfK`JQ77PZG zklrJ08HaOQ<)u=+ZjID#$&}o&&UV1Q zNT8~$A~){-5D=0T$v{?uL81)Dq{#b_u#SbRda47qb(?wJ5Rt73@e~-g=F5)f9@gfP`rC`ek(#cO zQF9L)1jZmpYC)Az0HjTrw<%NUi|K^pyJVe}zy1srA$IOP^jIlfHeX6U`j58T^jq$e z_(=zpa);wA9f4`Ofr=BnA(=SlTr7mxiti0?LH848z=(FUYqozq> zT%tGuj@1QQ#R2dPgDSI8f+35>0U)lgf*yjBjawqBhQ#L>kO`?N-E*&I<9Y>hmM1t4 z(yVU!SgWALItfikQrd8S%87Y4CLNZNYJF4bhK8U3g->#c07%Jsje6v-A<3V zhxm_Ak#rpn9Pm{_{YSu}-@`04#AtkD}^ zS7uo<_Y0keD_!ul#*3PGn8b|TQ&oyo^>MB%{j3Wych+A`?YWw(YUMJW7YYJ!EPCo8 zWxSVNK1)uVahdcVyO#v_PnRIfjRgaQXjacTq@8qVq$|~ykmy)02C5UOo+B#o8Ko(e zIPokjEnOl84iX2<@`G5hC`%pYrHom=li#VsJOw051xJ>hYFZQcJ$#FGK#7e&NwujZ zo$+SNfyuD#sBuNhQChuAprS-YqY7F-CLVrIjjpdI@8oI zNQjUNAZ5)%!|W}4U$JalD7*oCU9GAcfKC<|U5)uBlS%JKCf`^`R;_jLFOZLh5z4Uf zYW2po3SbtaYb!0}g51u15cFJB$03ADloWpOrj&d?SIQRtDETivBO7nM6ku2*aeE#l zaeMCvUQ1Pmo2`MN6KJX+d9iFU*c&P|HckO#=X{`+2n6|C6*L&1*90|6YQa%a60^^K znoiB~B~twLr&6_I31W#vVID0QpkWLts8ZuQR8>;SiahWbsarT#syA!TSza5Gpw0eODQD4XgzHB7O-@?z||F;XVyNILG9|Ix&8$uC!FG z5lOocl#{AFRqtWY5SH3ssZ!6mr%I?lDLGXE5RoQTqWUe;oZsXxK(iidQUa$Mf-onb zWj(9}73fKzHUeHZwa`YaogZbb>021(HlOqr@G^(9D-hxS3^ z0Je#cj0~xo^FLV+fLedsrBd|j)7pl@(}&7{OMb6&WFFO-oAK7yLj47X%iu`QbW0sg zTNs%IrIETFRN2Bc6o83=(x5UD765T#M^8}W3qC9O>zz^$)g}Wi^nz&m#&0r^Xl|w- z0zgTEG)Jei%U}D4oKRXJQ-6P}s{IUL{6IR=fwMq4Q~)9-opd)-S)|f`p@a zHeNSy&k2c2rB~2QX$9jE69}A~KcwWw3RX$XJY-f;PoBVUvX1Z$jV^STKGO(LH&l&U zG;!z0^J3EzM;Ov+J4(|R+*g$VDwu?*%5@-simKN}*52YP0dtce6^gbUfx$yCXv`xb zJW}s*POv8ftE%3B)o#>VNAm8~+B))D&&DO&B7L&3jmQ&85waKdR+)@@oJj52s_PoM zKrw@E7bL&Hlsze=1kx7PGU%PINxwT%VV`k;LvX!SKGqe4lLznChpe}_xRwm3a_)yx zdlpQ^GqNq!lAD=Zse?*jylU;H5&}R1qj*XUG)oVGWts_W7v6m-aoX=B<@nzqUUFl) zCNjRkbQ;`72_XO>%@zNcL{MOUmsW};^XKSoKH@=BZZdbO+^F8;Tvy`h7s>Wlo&xX< zHeW$x#BPB|RB4eRZOYD-aZx#P%%RgHGku`cBRW-F4yr5(4sKcykO(R+fM5eamLQ`l zJv`UBBgECbQtgNQa?Xv`d6dG`$bxAhP)!o7(dTp$g7n;K+< z#9%;xIBc&}GtvNrYCn~tKm!Jz^B(iYy3xUfQc54|A7+7L3`Ym&t4J#X6%P`#x6M-# z;KQN1F2=gRA7N>_0v=e%ABGPfEqm>~zr@3srPc|T*p#4qBA{-K8jx+jJ3~N%6$!!t zBfHq1@bmLN77QDM3rq3Soh0l2v!SJzRINk_uX62Xs5yS^6Ifrc?F14T)|WT~DDjl( ztcb#M2EODD9B_f9O9epg%%Njt^j!~1WMs5B0iJx9(xL$%^N!M<3w|jb;XPs85Nn;0 zS}QOj@W9Tgg6_lt5RBNCDzU}6jvTXVHO^7ji^d$MyeNRYcW)2qTVi|?0&nB_D z`2uuTiF2u(!2ke207*naRJof!!{h@Nth{MMgNtY3r|P`2S~&!b0DVA$zm-ZBcQ^Eo z7EtG8%NWnfkMPRwfLDuQp&QW&q2^L9=zm2qd^tU}UNSZYNntqv7p==^2-=-vBt*#| z3?s(q$p=c!e^yE1laFET(nJLxojbuuvU>e$tZDjIivRV3s^G}v$pgu#5)YrMU=xbP zzwsx`kZte0s1j7lai{9YUQMdOcsW81hQB{!bU=nmg8-GzIBy1ti0!EK+pi>Uk9{mD zjGPo8$+bu&xL@$Z-G~7OyLGFC49bPSBVbAqm^P#gswK&|X>1TVFjtbX&HT21KMpf* z_z}cAIv=N|`BrOXXDmVb$zVDsI$~uIEb^*h7F-Ju;5?p!5*;y;DS<{z8x2CaJD=+u z5$>yGh1=x%ET6_QxxJPH%#*ZHX%wJD^LJfO zW#vrE%fUb>%+_hX&iv3=hEkQ9%5x48_w^%<$$QQynRg^`P*+hOj$MF7Qs6yiwsOeY2^I`C>Cc6Z(ib4#w(ku43;k^^;ML=YyqAuDua zd+1c&*;vVN!jQq$+u&*kJdg~w!-0|w7$oPPc6|B{M*MLw&CUZxIMtg3jU@`6zF#UA zEkM$+I&+7(BTkX@-(D%nN1ddqD}oTM@k-x+Rmy*wCv`RODnSN~PECeRNr7H6ssMcD zOMg=EqZ%t5yX%HvVX}Vzqc^2tGo-h*Fms2mLK-`id^bl5{ykfYKYm>SG7gJ{2i|(0 zN@67ozLFrU>Q6ZMh(?tf^2fO;8jTdc^D+PvX;)N3=xUHSNL*n6k06w}5b;KYgvrs^ zUVZ?a(9&HJ0ks~$s6d1`G=9)Ux*pDeNNx^6cSkBA!Ag%-j|3~K2wf#5dQOnG7WqPg zBH-bg5|V-{;)Ze#jo@a|)`kYKMLs#N#rx|019cMu>59e}dc<>`SwDTB^KikCv@{@n zeusJI0%DF_XhEoDv0~_7evY^viBGpTaL`6G^$J-3#iJ-WSe!-3XM59Zw>Kd_j7Lc& z1O%zV5`p{fxlz`yU#YqL?eDIY+yD3%IeF>~j5}PieEH8(Qo2oIV-sckgh|>~d|7rj zJGS8jEG06|Y`YVzIG79Qz{=r>nq$MOFn?75;yqGb*DG8yvExDc0-LoES&GHN<3m>i z2snQQ27qV&7|3|~!O%jzu(JTW6To080YnWMDTzm)E=3=_0xN?hQi}O26IgySrwCZ{ z#qN8c(2E3tiA-7-5)+I-scF8djh8Ndr^ zu^sUkJ?2fQ$zmjO&``;|`cB#W$Zg{MOS$Bfw%^NWDb z{8dt&w^3pMEXqE0?y6SvH{iwxWgT2MMGYP!wJUxU0SGvuZmfbD?kG5)9fxh^VPFrl z8pzg|qza9R6BnKnR0v9{0%Rt<1Bk9`zW@xX1|&!@@IKX<0BBnNan!Bm=X^%sg5((t zRUYS}sOB`F)!4YKR;ETEOpYo>PEL|oNSmHUg6Q2sU!XdO_wgLGR{*uRnSwU+U`*IU zFbLuI4anjxiNrQD#7|9wH`)jUDw?U67P>lw*&dDi>UItwcd`Wj$Vf?d?(lqm_@bUhfHbJvr`#)1*HJSx!pyq`NP}?)LKz z<}u*GisdEKV)u+fCEW@O!NTp8@^463`+@0k(J-F}YOvISE?MeIfGeFLvR4=Z3zk_Z zkdK?>E=j1o34AIF1NRZ>_457tU`fPix#QkgP%?B_GS4*l)N_>-siRmjq_DJHiQHpP zi5f9R9m$4MRfi5tVo<8!+=+5McZa14E*fzMKl!E<&G|%1Kl?~(@>WXKqRmP_(ooY) z8YmHajF%)xWno^P(~}N5QdM{rE3mZ)Cb?-OKwAw}W7s zo&gI>fG@y8RdVF$eC!>NKBXNE6jMyV{ZM$(py@bip@tWHM*3oYiVK5lH*HW=pn{dY z=ao3)K{ZN=Naw7GD9o2;4{AXS4n09%!yh;8^*-~8?^s3}H0Y>O`k8n5*a7!S;Xrww z?<$a@-1?&oQWdp^g8)?F_#V~2Ub)^?GJgG>VB^9%wk3-(m_dL8DM(XiNi^DBC3hk~ zh{po$Gp1jtV8!cASmH4tBMWte`tqhfA9iyHgQ1%*f>*2d#pk;38zdL|O%in1_w|dE zA^_I6PzzEe$m;bCIlHHPv;<7Rywy^8W>3Z5S3^TL0IWEMZ%6O$o-DPRDsu$>D}x7O zz1Vh4enU&g>kj7>^+xdFdZWyle~_9}&Qc~AhE5t9OEp)xrpY+E09CK>A!Vjs*<(6T z_c^f8HEr4qNjmIUH4JcZ&I$+TRD!|WLmL4FZSU{5lQB}VBAhf}{z~6>zsC|&b<9gmmfvV2P^(Z{&*jn32P8(|2*#R{=%J(ZeFX?GZLWg?d$YJ4KiXDDRW%$`ScpbI za=@(JJCI=l39NZ6f{&(WVi4>sK+^k~C(1rZj|qCkC76Z8CP@0(7lZKvm`Dh(=R6O_ zeUhRAV09uqr0*RP0XJJz`vEM>C!74@F1a|~ZEgkFBpNST7w zJoDMw`)#G|l~jwT?n3Dt8bF?sl^TFx9*a_0SV;S&DyC1M;(DsV=;OWJle)G`H&NN1 zEQ!gw%fV4C+U5H7Yc)n4=V4o@KWiP9RS8`!aq)BXHWKP1$tsWq4tF&+{V{+w5EHJm z0az*MhAeonO~P7h4qPmD$cdIxq+V*r5@G+8EFlY3c}YwAldC) zl8Guf!5{%<<|_3V$p9U;-6=C2q~#xEVa!os;D(OwEw{Gxj=$a`sLq5MMtfe< zeO4;GhnTQAZ-auB^)3xCLSSIA2#;w>hPhiv$j!lsaUIYUgkjz?l)f6fA@G}FL@XZ! z-+aXf1@Zob0j$9=M|&|j5UP~3YMc|47U^I=5S$GdEe~{+Qo|~H`SL4Ryy~FwOHtnblyP~ARf0F_?Qw^IV}IO z0aP^Y?--xZr}}3c%La&cSk!W0k}P!515@Avd%| z5}!UJUvxpb${GNg=LN~;k|2ozW9qoEg-^5H& zoq?Y&&un)lhz4O9a6qbnWCb24G1yjynu4}kEuR__^=wR9qynhOk^_})1e=0_eEI3e zZ)FGEc_m^_c0AS)5^OjpMM;H6JtXwOLq}+h^iNNdxv+$N7du*?giS)Xz~lWRaKgN5 z)z6?qluVj*m}FqlybF>u-<3MD^|5FXZw|r+I4CfS>0^Cr{Xf!@R%>pYGPstLChl zQ%V5~HwjoT1F+65tdeyY@(%)41KCQD1{VaZQW#l$T{}b$aI6MkRrnoXu`xrNA>)VE zOVL8CwP_sqHLt2(c}s-4?q+_=fIaN#^Tefs%WYZWimbf*Q1(&Nd3`Cltg zjT*I&EM2-l#*W(yeKDlVOxp=e?B9`*Sl`wuwaHGTX!QpBQ?CvzAdN1DxK)C~(cl+@ z$~5u~)P~OrqlASOMOa80A>TQ3b@T~1KvV7J;SE3D&cFsu8-o;V9WoLURxze3IHK+g zLFcE?&gh_MDnM!?Br3|jeG;e$No)-0dju!8Q@U1atF2O^Due2;tUe6>{(=G;sHmRf z+|hFQ^&?QFr{`)HIQWngz(6oOgf z6@qnwZ7KgO)gPSG0^YOX4kgX4`m<6AtiNRFlxXO?@3mK+)y(JTuaO)6@Sxmv_uu4r zEan}`^?IDQLR=Hn+7C+&van>~PxsE2f^F+128+YveORpCP=C25v1hg&-NGgJ>|9uqt1Xz)l2C%rz z^%DTB^Whf!OE+LOHW43Dy$e4>z~TJPSZZOZIt9B+Uv%*w zU?lOg49w1!Z2+#*&$vV`z5EZt^;b^#%3O=R*PeXhewlgssWNELC|R>+u~4GA`PK*E znLRO)Y|fx9)gK(wLZZqIaPxG-Ojur)b@QeT=9cV- zovAs0RRu5fl)Bmq^rTLS>FjeZQ+hr9)O{H6#mV>czfo}NhusPC^Eb*%|9ae$%-Xd} z<)&L6mg9~;O{Ffb$Ldm#?5s|VKpmJ`3CF>}D(>#kr@)HDRKUV-JWg)!2f!+-lHXT2 zOlT}iN(UHySrEz zb>vTh{ghpHGuF%;5f&+v;CQ#s8d9`SF_amCvCd2A3JVP(1orBjkE!QMaa?mI3%)d@ z1na2as#k_W^%sl#l&E~b16#;(*WW5Crki3bL~(I8P^>M(1U@IA6&7xTw+g4E56F_= zoqv@K#p3vT@4i`9uUVvrNz?qEgspH+Kl6Mwd#81vv#yJQ3Lj#f5QJl*LVf1Fg5~RT zV8udH_la%-PY@zk!KrMCq^UV=> z2u9zb)umEdWjV}^h^6yeD-fu#AF&hxs6w#@Z73uv`uOZkK*f^M^NI;HS+^uh#cFfg z^Q)#Okf^xj$Dvsv+F<6eI8yx(R)R+%`Sg$im9To6O|n7Le4PtH_aE<)uYL@XO#rM|I=gM`%Uw&5 zZcNP#df-4vPK3qZXn++%+l)34Th;CnkwUDIV{8aD zX{Z8K3>fYpNL1JuW*79^y}b1FTlAK0!lFe#$Uz4mseonfdH)jN!y_EJ7VG`@U&5m> z%s-t6=_*rIhe>cue87Q+%jEs0sD#FTMW2EECBV^yy-U@8-kHO>&R4<3QY;rv`Y;5d z0MTXSA=@AWXE-d{Aq(N2DGl>kv=pRAZ1YSb1kfL<1gZazhSk;Abvd#KmV&MZOWo3z zxv(#a$PQ&q&Wh|i#wGnRk3|bXkUgLR)qYeNW9vdW6&8PE0ak1;#M}TY3Ub^%fi8n$ zB3x3rX|q(9LSF#_#bu^I(t@OqMnLP;?J!SI0-y~B^P|GLZJ_U-e0G1T4e*FQP(Q=< zQZGFBu&iIVTIZ9zy-t{XQ0n3W;UB)A>%Qfb?c29nfCPlEgTG7y5`U@ctAk#hh8Q%~ zV0zsNRLl@}28e{odN_>K7}%$M5&AS(k-(%gnGBDy&`G@vOTjDLmVyK6ucoans-;27}H0#?-51RgLRF0W7K`DnUgu*V6%1gD_7; zi$AuLo~3LL)F&&`n_HP{%-;o7l}gF4>!fT8_U^_2Fg`m~Vp9^d^0~!FDZq+rsCv_W zRFozO80KJ&)M2}i{_v1KdelU0Uu4Z)z5W_@xQ5eY0t??Uf#)>3D8oKybLM<1Nr?mW z5y40Uq7uyb9)GM6`?Gt~LYwAwR)vH%Z^%`tA32C8b!&<3Yw??XdT_vR} zS4-87a&g2)O7e&-Nz6)>TDaia0n_waKs`ODr<37gib(G+NbjKG@5s+Wxo z(Xti}a1Viel^v1?Cc!l<&~+=j@O{Y?upm*z$2Q2pd(_L^MZr>3X&GJkn^`;Q$j(wh zpu+sXaR90mY(ej&1ffG6-_x&Wzlcr(T5v~Y=?7_|szQDK)k9*5_d?LJ6WD4h)*hCJ* zMxKWrHdQ|R^uOv|;+m_@l-K?+9mXfj{`4mPMm2k_%zOY++V0BYL zj6{`*k@K+ts(7rS>YJ{Unh_Y>ik>N0huk}fKL5DwhcYofLv$BL9#!?B>f%lQ}0lqrWDCn+iYW%K6sa@)-_<@@hHmwri^ zGAMV5+;aP$B?EI>ROxNml7~61zsPH^J}qO%?kkHHeG5+)Ps*{!{RZ<==(*f^poaqM z)ak63^!`zjn}eOBQRiCEKFW^`R;>`4fW^w7v{i!9#IwcKaxKjlr+{cWKb6c5h0q(+ep(Kc zG29U$U(fkOu9|t8md*Q5WO$NP)$EW76AqM{ZoOYdj2I_*c^hQ*(+|t5uRNtu!-5m# z{EIJzN-$C$edJEub9#X4*kjL>8*jN!)rH0@yLbA&KBr_gd}spZ(jsiI@-^{y^*wef zGnV-jU^TO(T9#p?crZF0&D#UD(-E4u`*1a^1osTtCL_bvX*sDDB(LcSASRQegjE3( zyCuJd$O-_K1H{Lm)x!XHRbO5GoDx+LGBz42zbOE!*fxQxrzqFerEY(EW{?pR33F?p zxh)0#zj3D~%bMWhl|cuXnX=fEfHJq^yUk>tqrD zi_DDX%RRx|+sJae9$8XU5Ilni$v|u|k`=N;20}GRcy{6F2`S4kKWaOc!pp_a>)~z& z#O2P}zMjW2Ks5|NbqrL0Q822oC91}z#kvX$ZGa~O3^HIg%WDQFq;~rDt8EC@^@}v3 zAT4{7+K=U=VFc%?=6v-Lq^i?-OaAYM`(^r!i}4Jl1wA%z-XPCE`xklRKeJ`PfWeTe zR?9#C{h5pzvnSM}F4+MO6+i#<9n^x~$dbiB%aSGYWz@)hWa=qr$q`4LBs8p`C7^GC zN{iCukE#Ic8YFW-v}9*t&kAp**d%86UD#mNPTI0=cyN4FgOhj5YUH%CS_wxF-3K*E zzzWpPNP8@JDFc%cnHX9igJFCT8B`6%9EK;pR}GTs5&T2I2n%-v%NL76gv)feV4iA0 zUj#DO?Z^l>Oq>Gq^hls*pw@3!(@U!WecPJH}cRB2V1lOI4yXvZChFCfpM{F=R(P7o zk*!pZxU`3fAA&p4OmBVLLxbzPv*1^o&pz*WW2y zJRF-;W}vK;94%SGmuBwT2|!~PwiZcBY>*@Na>>a44N{K9!&(pg(~W^ZRgB)a2Y~7j z02L)F8>lRTlSy&fDJBXM73LH<&|#mbfx534Cg}vKWQfAUp}!?ik!}p#TxHfrcH=Jr zhv3751|sbVG&eW%bOaMhM<&!^?Gy`3fT1x2L5P5M*4dY+b6EmaXmG6Daog2WQc|Sx z=+Kt&=|MdsBU{bf>3o)wmp=d%`y-8HVq?R4#0t>R)&?s>iFW>yU<4!{+=2K30M=`8 z*R&qj-vG3@sbe56BqlJ?#62E>wY?!iK6MV1d9}mfVITqJsl#D;yYnd+YbSw372O~c z$GT+yVf9i5Q_LD1R1M;Pmj?pXHZU+Xrih^GuaEBR+X7_?R8;rHV{II*{y4m@g}bCO zfYc5eQ2@Gfp|9m4-#~DiL~shhB&n{6!gd|pAd$fUkzRQHp|u{>d#hG1RS@CgU4k6% zo3L@iI;n^6FmsP!LzDA*d~52o8LB3%t1pwEfBstj@%-Q5+qW9?KcSiqJyLL?FaeP7 zn>0+?J^h=U6oEWxwf9I?Y_Qr5va@OO;5aN2wRb9{tZx8X*TVw_rK~FSV}cfgPno;* z$xi(xU{U#(h=s(Put(Mh&Y`lhZlHUgEY4(E>Yh%0qysOrbI{ub8Rg)C^)hXeOLCGM zWE<3iR0(pF-xY!NWF)FFn4>xbTYN;=sy`(yY0m~94I_$#SolQ;*KpPcW6fDOBv7ci zFgL|*cZLI~`k_J?oqeN=sz&qBm46HvQAJk<)f;{F{jgxJtY5!I0Ybr}nUqCf3x$OA z-M3%K8K)f}AARtq0!%bicFD=9?(Bugk3Y;)|9u20^Bqc9bYx3lJLS}KjP5TzD`UbiV&mTU+XILjCM z(^Lj&Dq}Z@O`xJ@ia~B;ifG%ILI((J*xWFaK;^(3VjX~~5(|54Q2$O?l12l#24HR` z!QIyV8Kl@O2p|L!9`qnVYrHjU7kOID@6P+Zs_ZytREtdwjn{Gl4JEL5-hM@Hzx5K0 zKjZ9~a@LGXu#0cLoIB%i$-!3h>#&3GOW3A<*znORHSs$ps3eE>MT3bqUjK(o-1i_E zJb1WAk}~o9>6gIdgkeQ?Rt$_Q;QtbRtf#;#VAL-8XrL`7HZbHs>8K`LUmvS#K?e#Pi^arYD3MimS`K!9U?&035b#Dx zPC^Gt!)UiYNLKB@0}vRb->u<1>y2|%*;w#Pt3Lu&%X3t$MPHP(`)RqWFHmYFo#Qw+ zg@YW5j%O=zRKeOW7}x1AM;{2iAWhOqnVzB!frMXgk3?dm7lO`>H=HNs<#hgP`Q1xN z&XRAw`2ZGsn`NJg2Vxi8D3yjxx&)i-oWT<2h{OWmIkIr!eEHXlk7I3Fv~1s2h@AuK zkzSEx49Jmj$pmRgL-mDzCD^hk|MhM|W0qmo+*tSd4aVqOF;{Xli}_v5Z~xRiJmEr zhA~A#Fd^8$^se6Sea{U@MPg1W!U2^pfV&3Z#JQ;&NL9q@MEvvz12Z7i)6x`u>9I|6 z&Zn(lNk+K)S2qC@euKXM zIB)gdJFm%Icg~d5l!1_}HlXFlNK7nTPoeFUmTi^yK3D|HzjWQQ!+@R18`CkN`7?9h zAASi+l&aEFB4tQ!e2-_z2CHtU0&m=@#l_d~AM_YhgKuGt)_AmlSTq8fy7zP^-<_gc zV`#c;3dRUK2z!6k$}o7q$PU>GV}vRXXp!Apad+n{?R$rPo*jyl{@f31^Ohi4xHed} zRbV$-+{Z?y8pwcIC(Q$asv3#JK~*y~G7&o~6R0Rd;^jUTNF!?UBEq=+11>NUsDWi) z6+jBc(+UPu`4Ob3^5d|iW$Dwmgd^kM2J955aw8qQH&a?xqBtze4~ zl>YL!*L9nD)_H#3M!Dnmt7XwIKS^~>k-YZCx00JXOfCA1I*;ja4k|1xROhcQyJV_< zj$40h+OP>e{8DA(rj>H*?SD~&ikg}#jo(t90_EPw;fd;H9Yxpv)>HXro3ld~Q^2q~ zf)Q0_hhnd+=ToEQ6<7^YwZi?f$SAoT8!MnCMvB{h83;nmZzNb%KNw+f?qXhDx_kpe zj5T$c@L0gnB~%@%BD9P#x~qHyDh4O?tL#Xb=`eHO$0bK&W5*mgSKR^aPAM#jRa!t? zQ!jjnbgxB8!jQ-bu=?wVxrDw3RHPYc!w?fu4=X<`%ZP%xb&x}fT@kvlm-AAPThvw3 zP)uGWx#y;O8mK5G(Ry#ys-<$;sr$)sM~{V<@*i5=BZ?qHsc7B0UuEUW9~G$1{oOUP z@8mnjo!L5c^xPsGA-A`T!G3QN|C;0XBf!&)g=`4OaOLc<-jx%&VmDhjFu zDF)KfF9HjKFnPbDbPk9ry0%nnn>Mb~`24fS9{T{mEc0^$Sq&ENX6Fo%!;hGzaq{yw zsuf@J^mxsKi+ytt|BtgCliMIkZP~h34J-(L6;)g1^H1MHJi`S%pACo;l9je0Dalw& zhu|i2kLKO2DCyBW^y-hZQ}eL8l(k>hg$WUI1Xh8+1=Zj+RhS_GXzdFyV`b2Cu%}0@ z$U?gV%Vo^n9RMwY*m75*EXVq+jG#(*!`O~3HA-MXNEwU^NQ3p@2fA8KNYG-ZqNGIu zAU%!xc9#sw43f2*F|Sn+EO0y`P7oN{EjcAJRNX|><@uzIj%_>lsjBrc<~d-&K{zk zA9$}hxOnH6&rP7z;~hvNf*1FmVjy_6W&lCSj^E*;UoH|pY3f-rVb6VK?l+%7Jvc`# z{>I~)QkWk>yOwKLf7p-eFjtk_FGAumz-HjRo(?ux`KSg=U7Mzx1C`2ENs)3KwjOyK z=I*}-Xi+D#54bE9X@NGUMTTZ1rzk8eW(VEc5G5PyvLy-9RWjzY(nE@+e{dmGqfQ0z zM)P-8f_ajI2sRhloZcGJvKFt#C zbkp-R5Kat0u@qq;a5M{piC`>}as@+*z|g1y(o;PS1`DLHH5J+)L5ju`46I28f20Ie zdk%*K9?JJp1SOVj9+tGVt2;sh-Tc4OTv; zK9vFtWDYzZ0@Jl?0a_;lwB7?~-2>3t3I@upNVu?=Mz!R*fy6o~T``2AlpMQOfH4-= z$H~fuILQo3mJ~S6?GNBegv7pMlWk2p&|&8Py!Jt%29Ukdj<(ZxHTERd+^=5D}gz@4WSj9DK-8@FX6kR(tfI zzF__wDX-qDY4q!t3Z~83W%fW&4Bw*UMH(+(wn#TuOohdt0W)js`5;Y?2{yv z9DfV!GqeCqO@>qzZ%Lk}AN6!z4e#vfeDt~R8^V(*Fc7pTX?=^)%#$(L3LA}v-Y|;M zF(_$K7$@@xq+kIM)tlIq57Jf)#_3xzxAl`NMLuy3mT%#MaIGs7u6hzM%7?!-{No%K zjo=yhepi&4osA6GiPU0uL`m}2I4~W{2`h~X65u{44$>D*wJ}CeW<(}Vro_Wv1GC2b znY`jk7w$&GnlCygSR$ga9xBQq)uEBt&L$i{#o+@f!NtASV2M%`=cMQ&DF%dMnqIi= zuSIW?FG@cK6sUHy3@?^p&ggwz$g9#2%LOSYBQsmF2ab~1xL8@R;4}H^%MaBEA}T6c z=6^Q_KKm|{p+oo3^oNfaqv=}hzyc~7R=o1kGcx^*gXNKj?}QrA@`!FpTvk3PWwoY0 zY?5(5ykJ>5!0DgbC{_7OHf_nOhik$cw0NEa(7Fi62{6L=68mSpSm~7aAOZ9Pj~VG6 z;!xVM@H$3@1!y5eAwR}ag`I!VVu6jc2`2 zMoLD@(0FEDxBWK=sxDVBzESOv0R;&v(Gm_Jd}tt9Ifp~Z%LUF)Gk8o%s6mDkqXtS{ zO+X9rO-|bWa@Rf8tl|%U`E4E{!8FIFp#-kkc<`qloTg6YRjB)@;=>CjJ_iE?Dbha` zh7+{x7g{iuXTra12CXGlypA*_#Juf*+0X3`8%@DqyyBO1SOrs#C~$xnZK0> z?z>(F4H_-C-h44Uq*p2^{b|7Msv0yR@3%sni5nV%TG8h5L z1*YzBhDgq)Wc)(T130#2`RC@}nlAIht%>S9hm}VKFFfP-n*f=WoyNal0BtdUzORWi zzR~22dC>!RsoH}Us_{TOsE1TRkB!_ziU->d1HZ|G>FYjVxQN0*bssHL!;vVTdB?|P zzz0e=Tu=4v>__E^zs=I}a4psT2OI_|s0u1khiu8)Bqtmug~Ub4i`Ip(Ve^#L%y;c9 zb3wI-w6gP(B)S3~4gp+-rtKw-#&^F&VB*1`Kn8BlUn70hWB#ii5i}mYWa3(HSm5FP zeUr8V5-xxvV0>U2_@(4T9>{wQh(Lpsdi)ZU=m%fF&u`$c{x-H3f)GK9X8N&sCkYA0 zAt9<#nEgNnDKbRln0BN%B12^R_KkAa-A~EU$4piA9OtI~ z_x^t%Rb3>*hK`4eq%v`YhfCh3JgKcKmOtG)8|LdLLXvVScvV+b$$#L>@1X~8#4f-4 zVm;TlG7IS)bKEJq-pb^~)NvsGvi&*Xw168cW@hxP55U&wXMAnm%zn$0yUr*%6$2kOqFsF0gRhNh62E-GE4+mr9$e8 z26*wH_l$5zWXPP|savJroc234HRmGW#~!&$ zUVHUv88&PJ77G6Y^Y;-l>z=>DuixGZY6Pnd>(}U39&f+>Pk6H!qf*zbKRqo+9euLK zQLt>%2uv;l=ogKl9wQrWSwDJ262RV@+xSm8Y_Rg5x@a>e?20_Rn^7e?=<;fpdzCxTbWJic(ZHMuOTcTq8=D`gY&2M+=+T@vu7}_gN>gO#GhTNoeQx71rBppgm zIB-qawMu7FXqrn;ns}^$j&zL_l?W*+79eHKIy8cmd50+mS}+n*MMb&%^78_K#tPkG zcjV|jWWEST$*ifpI~8#|8y{ zFS06q%zbZS5jxC$10$>#@0?LO;xm4o!2prD$NT1)NrUmMw|Lexn0)a&3Q+iESveO) zDT-=7N>Pkw;^5T%b)E%gpoL#p7}Rv|-FMHd>t)`&kD3&OYP%zkI#~h4fCt~91jeo6 zue;`V^3zXqWYnm=W$DuIWeSWZ{_lpn)xFa8ZCm8uFFqx+pLw83n$0gxKjUHm-Gx|7 zHCPt{^SSv}^SFU@O;93GB_=v##Ly(a1SvyL8>|c^0`Qj%iHEneMK#bSs0M!laDA?7 z!g?um1J_kJ(x&G5B?{N4669q!An}$Y{K_ zp7E~*XtbxZ((=+eHQvKB13vtmk(NkrEAZW>SLR7;zb5B=%Scchq(tH%IK|_jMIXz} zIOZ_#wU%z6+&8K}8afbkzW@FUWZi*zl|eH7>f zL*Sys^)xB%21{VgPXfP3la?x0Vf%Rr`WZu;Wz2J`loKIki ze*JalU@cstOga20xnkx`(my>@rJFi#8}C`w-L%l={dsdglglqXMKUvo!Ehx+)^Av* z5xG&~@Zn<=*yvZVxOj)UN@CB5fE%T__ym>ExZsy*^(LSqDAA~b1{E}{@MoZ6W!YfW zv~Bl_D>5%KEpNuf`Kx>Y*D`?C_c%Vr@dXzYA*3Q=Hd-##g=Bh0Wf*7-kkUml%5lPu zY8h2rBXu1BD!kfb!PFD!t-ym5&TrB21f`}}j6h;g7OMDq13($01gg;p0Boo&4JjCp z;J^j858QXN{P&ILK#^#9=l!{om6e0_Q59NeOs6%K9Mk7BE)@Roe{aZ5H(nrIU&Y0} zn>Vl0_uqi$bgK6VNc0zI(xNJn-|lT7#X^H5C>u7`#5UWbh9|*r#k4{tiC_Nl9*Fy8 z2JK9BoC2&=Ha#*w!vM+Tvfz*KNO;G%5tFE|rA5hncvOA{i~ml-{f&6etxM<$J|9ei zjFRr2xQz)X>Inp-Wn1PzV=hM#LZcxB5ME z0MAKMj*R6X|N9hD3CLNeMnLd+hP*!LO&8|LBRSz5c&Y;ds`R39&R-)v@jb@p5y;?# z?0W(w6nZFMw)7WWU>gk=M~T?led;M^OERqUjE_Cf3n+asCRD1bDrMbVFQ(J(^ZBZC|S7f(D6gD6!Fkh&Arg=;a# zwH(JU0J3=iub&8Bd=ViR@6ZJl0gQo6*npQZWL%qX@VWQ5dEHfk>ALU-dk4xPg;;Ne zeD{?4&%2DK>|9Mil^&lF=(+wW!hjQKNKX>jsY9QUCgwfTkb&><2=qWnhijw=JkQO3 zOg{bezn&sd;z>eJ2BHnj8nvg0EWN=;Oy z(hna80}x^XR2Dcf&n6wudtM(l&%AC0{3)d>TIPK_?+f8p@DuhtP=5Mxj+}qt^>X1w zzeo8DA(+LM%G-dM&3W`wYm6aCDH8anV zg^T9N{SUsN9x1AIi+NBNpQ41vIj?O6`4STw57TXIWD|g1(!Gk_?o-le#IQt_ssaV5 zSU6j86gKBh{IR7Ca5)LZGen#dLNpA`=0kX=c!7N!O2)pHLRpo>bhW(*qaXC z-!i4~5P|!g?=!)ai|LraZ}RBP!AMoymV;sQrVaXg z(#g{SNTDkAGzT6=a6x?+AD<|P9(J;P_UXI&&b)bF$TX-9TVk|i#G%*@W!>6e6{z^a z`gN-lpp$v%4ZAIvoiG~$a5%vY3VtaLBa+Hp0ss3w;eoA0J{bPfbpH@syX3H>6 z^J^F^^Tc}uFs|{5!^Ai@k(@+#83U{ z$BsfQIDB-(Y-jZc(neDiwpCp{{pSJHdqB(jrBl&*^QfbWpCzgo^u^g zBn&rFRFr6xC#V9Xu?E$NMd&c)I7)C`001k&c`44h;pBE4rHE6C#JAvg34Uw!L+-D+ ze>@}rU|t8yzi1eRR8#%O=gf>dUQ7JPl!Lzn76lkwm~(=T0VaZG7_JFY3Ycl`?VCfo_S)o0?Vv1QJ^3(c?MI&ZiuDj68V%4W88h z@xtSB-Sx9H4y_6e_)uCZE-sR{{`<0|rev!}^&}W^(CV<3z~rSiWS;o1F9dWa)db`W zj0G*=zr}qWNM*OdDv)Z+kMe4|xcOz)+!*2cHP6_{xK1brjKKo3oeRQ+$xiej2uk`} z3E-;6c@2IEc6E40&-1*;t>|e)QL&>+BFf=K&(YMutNAW}2}0Z#7ktTmZs8eT_W)F| zGA--cyo1-81bkKle|dPT@%Vu6F_>TL%Kob~0euNR1(+W);m9LT()lP#J{ScMUqZP4cJ14X7Eq}e*W2e z*puofwe;imAHzqCRmmw38ML8SXMWlCxldJgR*b}A=}Mr|!3L|&v;zCxCKND?+WeY_ zhSBrd``pCiFQpjH`B>Z@zm0+!U)woA#kdUfRHYat1`~O_6J~5C+Wg+msWgdVqMLas z@90dt562B?yg!ZBB6S^x1~AO;((0y~hZA8NbA))tky> zJESTHj6R@A+&Q##Z|7Kgr=>i#bgz}y?c2McfEqnO>WPbuvtoiZJ|qu3^encrh?f8R{n`5b+N-C_@yDMfLx+vTdZ{@wckYJ@WUE*G zs?RUH`1cZt^<2Hpr=N#7ZJmI~-1#~pJVdew!n_IBfhtw8m^N6oRoV74pHZMCaAC2BHKYx{aQ13q=OF@LY)6$fjYp`B_ z^?7;Z;afaO-*W4t*xKU|fme{{%buPLSLF_>k;@ZR7V?I#I!j zFLTpG`tTbvbcAm6aq`sZ^8JEu0BAqT58o}o+?dFD=U*o$oOA{}SLEte_RZ(S0*Q3H z&MU>!ori0x{zG4(gWALdn5i??&e3+AydTA|!OD*+!_LaCDUiRl0vqE&{YRq;_CP!M zJq748i2#!Y@VNA{TjYtyXJI>z_vOzI#LFR5j+VU58|2|X-vfqesK^ORzmz78;77=rSFC{su{=BXK z{GL#|mBF16NP!9l6$Gh_0q%LK-AdcPUb4Z;zv{!zO$!t-QdMbLt>oobVlv$ZsusxP z_qRAxHT9Gk^2KK#%ChAP(~6dXJbFJ+DJf(aMt%SQCy+@*K~$RjAy9EWRTQ?UA3TVD z6a#yn%DZqjSb0;!KKDxsu)|T0u_><{Y94y+4h%2Reo4$8$`k>muC7*+l2YW)k3NUp zd7qRQ|M?fq&G_*LV5i@cfb*@YTcgCslMZ~0!n4;)?1Ejy;WWWKLcI_G#(86 zR$l}vmfafI1xapBEam~>%^t-)Ra;z3kT5s|3r-l^-QArKoB;-Rhv302xCEEr4uiV| zcX!v2;2zxBx!*qQPuTNR52v-NyQ)iHWuGz05It*q8GD#&TdcO-njO|FYF8ST5X)hz zodEJIFOSaAaZJ@6;F9T9dDyTeKWLg(dZMMJHX53t;!yGcG_)2Qh}8Fw)TpU`OMZ%d zoV(=Wt3i4T%j2~aC?ClPJkOIoKmoiuXHk-O*QCWTUaBlv2aX|{2wfJ_8n@OCZf}3H zx;gzHm{m@a6HlDu#{24+Vds*+{U%v#Z(5>_L{b*ek%+BMI-N-doYHZNjuB$9FszjR z$Zy$cLDy0`+332WLM4RQo-}{f30hor1^H|8WsgQB1S9Lxru*z?bPvg!9wp}9-WgtE z5(0r0Jxl7o7vG-e1SB9Bh^!bDYub4$h2Uj&OSVE?VSd*Le#bGMb@g!JH)?tCgH4(@O!7!g9nHn%B?ELK0B5Sfl49%a6c%t(oU#WqnR<^r zNH6cLP!4d(WRWU5A({J{91MF&PqOYZ^D%B2-`Lr3AhK}EDiOWg&W}yH3Q^G*dPsh^F@F0pf*y2>A;>)D=KT1ZN}3wlXj?qTmjG8!f{*Tt>dBk zHPg)4;+2v@!uvjIp#KXrqX|{r3Lqzkg3uS1^aZh4*vNNW+YMS1eMQW6vpRWBz0i$( z*r!tj>QA?B$KNwHtSr~FrmK3SCAF%a!d4!Sh<<~w*4inaJt_(f0ul1_xah=8P-N?d zeO6G?WEj^wUSn_}*!ulvyNh53MVFrM8A)k@Ye$xWiK@2c<r#CJt zHs5`EAukUTzb`B>1-bj^iPRpul+YayTtMJ>W|TyNXRX?)$y_y*Bn!PHEd9M zbt1-jRHj9*p@9N!=20=kT=<5QRtoz1!4p5P>o)wBL6@eY>IW5$zxGTGPTZjK?(H&S z<#`Pgl^RWtz}ZYMiXTRqya}}HHxJ8zd>}46m9$`mLHkBCj6Kq5J2mQJ05*2Q71TaF z>eVfxi&%simb@2Fm*cN=06G2*4tCWd3%E(u57=x=Oyc?sYd4uO=<@!ZZDdC-D^Q@4 z*hd+nXj`}^jjXAE1I?F5OBF`zjX!+tZ-jbR@ViTM-Ctj8{vT>H2(oPLdNvF`#{(Do zQ0YESzV36es2IN4w(XRNjlvLWYHB4{49I+slEM^Hu5P!rOe_e;k zNsE6BTP*>$X8{n)3JRqx2&owx{)!u&lJV-edtKM#oYEuEgE*44`7Pwscx11dTTr7Z ztKN&nM9Q}XaUT4goFbCbQUH}*Gey?Y9JV)}kf2i6TW<{ZLLSoI!(AII6X$d1djD5W zdG+2)i8PutL*GPZy(=_fdg;}`ruCFr7m<(gt($V@<`Ly@%xuT{pKI62P!9SA2bY4b^z++=rGA zdwd$w&ziDo+lsV2Nz?b}9Qx}zuFE1nC3uYmvtkRPTT`}N*Nfiz=OIf5rR5rG(@L!x)OF{b?on^)_jFP3bO=VGDbiGE&J=tKePIC2sv&oqf|GF=0`=<$(EoZ($fCgfz zGbl$K=NNaldH~aj;mTiB)CwORN;lFM>bmzsuO^X2@5*hO;ewl|{gjJEA~=vXQd&TH5_{}j zO=^@=363(0yx_V&bSUUQlDw`(6Mv>Dr^%oNsEXV3WZ&U6O%JL=BUxpJ8>uMN1_yPw zr0w(SGoa=rziPz?m7vWv`usL;us~LdWI5;k3tie<3Qk&u@)Hfsc-J#o5z()wvH87W zvV90Q9LehFND!kyM#(%Dl3KJis)ru;rVc%y?#^)qD^%go8HmLy@DerVLViCO46O7ttop|Fg^ z1kR<~>DE+gtNnaZ-2Z8kPmOXldSVE_ot6bH;=Fc03Zuc^N|c`?i|)YW;wTFnT<2s8 z$xgj*a$HUF<)SXSV35Eg>!NW=%MH`38u$$gh!fj|Z|XVvY5w_x!_5Lv(bw-Faal1T z4%i!`Q61LO8~P;Ooh=y;^~kx}$mahip074!7Wb}r{x(0)>2YfAr=_c2a6GV7B%%Vi zi){AMeHCDrlZj;H4Ren;u78$FT7tUk;JB<1bQJ_9FdHZC9r^+!tZPkVL+QJ zB6q13HEoKm{NGk;|G7NU%S|9w&k)voC(Dt|_ zw;Pf)btx|lUr|1C64?j7~hZz;AhsMvZZubI4ntd)iINx&9Ur9(x`U(Urf{BKvwp0UW%^T0?}jc6*= zcr5WZ$7$|p<+F&40qZ|mo{4pV+{GHB@?rEOI<;qWI#ei+;gZqBC3qn=iyC49dwA!^ zZP#wH-2OwHnIC+2SBkXX)PgLj$gUk92_y8~qj2imfmGu$nL3y(-d?87XYyw2{m%r0 zR=;aw*W!!oUrjVzjOX|$ZQ1x>QgI0EXdT717{RrCt~vRoBKq2VD)flOv!5qD_?5-* zrf?!%RED!5Y_uenY44V=3p5%`0{lnK({CmwtOX>eRGeGvahnxmZRMd z2e)=H(1Yuc166am)i}(+?vvK3EV}A31OiD6f^*%g|H)^I3Q9>)9UlDr-YCIn^KU<% z!fdaJZ+|q$MLAnm%FyKV~60aL4q9QlP)&RtAdOR9C@H%9#CW7=9>T=!D5a=c9 z`dzRu$Bd$TH;8|h-bzfRaCa}WfKf)I89>gRI0Sp@YhK3}n0-BeTt~+yQn5Dvp1ugS zN(=rQXxJOGb4(c+Fi$C417Lrab~Ao{^Ql4WET8We+++lLEkM-uHQ`6kG!zua|MRFH z#pC*#cErW1b?v4YYrPEe`SfvJUCvG#uOUoFna7 zgj>+FLV~shnAHb)U@JLRCTbG!m8PxthIusf?QYM7YDNqKvLM)pEZHTyvAy|Wm^bO? z!3^!nE%|v_tRPtIX zfPh;@u2xJf_xU`J-;=V=*LY2QF-x7rXgv`$?pAlQWL^DE8sIq0%AOK%R-K7M;kdb+?2drXh8n0)c`%vu4#cO?<2F-&r3LzN; z7a`h}5$eJb`-fAI=xqpPyOZ9Un41^Y@J6%UXOV79)1kz6`eqNAI2b8sT=8ki)mzIi`qF9rP=-o zByN{#>;hy+uFl@M)-bL7EE;a|axu?;?_J${+2^OHNNHL&-U4ID#71#issxz*dUMkqFMP(y ztMyIBo%&vjWP3%Bc@bHHr{ZQeHC`*4NI?>Lb?@TePrVyi{0ip;t2^AM1R3f*h!rI_ zwKcxtuv6MBy1K|3wgmcD+?*C7mr;r7@_!t)53W4qi?Om^N}I;>PTjdGwxj8aC) zQ35svCqLiTS^QE??y)`3n(-nSb?u@h6JQxX1q@1KUY|X`_x6zv4s~3I9kt<#p*w8K z>Q0WQmpJ3_C#(EQTp3T> z{|CQBP)*DHiuN9rssIO}Vgy=RZ|y_FvZ_09&td+yy0AD~U5G&ytgr~`6fr{ZW>7V8 z&{T3y>A2kvfKz>&p~Rc+zVsG&jk#V(!4{Eiv^tN-8@hoA3g0%m7IzeAzP&wca@sCz z2_oHwfAXh`lkH?gBJm_-Mqt6R|M&N9(nBNAuFQj1>FfQ8#{e(Av7(j(*0BGA2(C_d zz9#P;=!UrV&l;VgZ6nm~Fa5He3Jb@eNNz$>e)IO$^1Mg(?y30^2T@+lIfummf#GVD z?tK>Z_kP$0m+i_7%2iK)0}U6~uvzhP4zap;h}&RW)v}SFA?ee#l9J)`@4v;zinSPs z#4HfpOzCDej{=xthNW(_>ku#zM`1yy5b%A z8_-E`lD(BUTSzwsjjp1h?da}!*~X;t6W0odjlMR5FAPQNPE!%}*7G=Bp~bn|(EcHE z&SxuP&35Hbh24;&CkqV5MrV*;-pJBXP^x2j(=9Z$x>)~T()bnu*oqE<^U}itpl8#h1zyAmXoXt3+I+2x!@P~isy(8UO;h$~%_oLLjdzKV+#c+9DibhB|O)U=} zXC}ECkRq1YrtxRB*&!PGHg!kUR7+S&y`)xGo}$OB%E^AsBMF$`2X{QkwPZ}Lb*5o| zP_5K$TEGeDe#JzH&|#DoEfq{4A4bO@^S_~il(TA*e@f;e-amX#J+Kr*)a^4!c?POT z{^IAzvw>S=MPBSPYu5W*rhl3_EB0CTgJNS)If9`3Fw3XP_V48$8CB|e%}d2oc5v90 z%z3Eq?|0Qat<}eZZ0qfFi_aDc{v(`NpI!nb6|bq3IYU22O=c=c^!uVyn@)m(3keo!!DAyz(Ztu(A(vFN%5akotHm_=(US%rBQ z$%zi6K3$^QEwUN^y1BVQ?(-DBtI6Dhyg!5`;GIj*Whj02Q|7(Lvoq{ym?^P@!0DzJ z=jIzS*wgF5M*evw;SorzETF9$S>ijCH(9g$OM`8P6hV$3y-G4~#ao%gch_afqSJQe zO;dWnojk-7ywlbGB+C=_AO-$*VsI7ui%g3utcBZcc-7_lSkZwWQ}^e*HM{S!ms592 zgZ-n|ega@w9mn?)fto0dsphHHfH68vVOS4HJbAAljq&n&;Wr#34Co{F39tokOao!5 z7_)QB7ak$S2McuQ462k!YOcdPnL-L$t9gSRN#f)WY_SMQvr1gMZSTJ71dDGgUnaCG zTfaEIFNgrXqfQw0K7ExB-+OxUj*YKrz4Ff1UK9S7eYIp;D_o1rehKZ$ zj10~k&}gX-T*}x0>Iw!x;$Tc%;IkxfP8H=SqyOO>{u`s9{{)A7$7>15aykd-vVL|y z`llf+6Bb^cPNY{5v+LQI$8+51wUe8-QK6&JP_+BsMNy$yjg59Iq63P}bB*ClTUnv3 z+mQvraIaHd=-~`bQvgsh3%}h1boF&t!Z28psm9K_nWJ>BDpOuOh?! zNl4?mg#Nj~RI~7YE(vscqdUXD8l=g``swgF{fj!Cfz5D1v8BDg1p%bZjT1?+Mk~yI zaGM_Y)4A^n*u3)4!q@RzzILS~!nG<2%o;y~@Rv3>Y1W%X>>kV9Rk?FVM<+C`l|OeAXKGJ&^WQE6hZCxA3<7c?9vvi6oKDG?6YjLX+&@LGeNXyJo- zv}T_~5x~6xM(*xa8+C6v<_)}JwQ3??NEVD{NIc2aG-+1>ub+7uM)71AjA-uen+3(-IZd5Rn zj8P0T@lO+0+skC{T|6ewcVwffa-`gFXz5L*nc2TQh)rn17O>A;Z1idul;2+-$<}NJ*sf9_(BWW^M3cfpEn;HS31eSom#fG$H|p|;RY_~ z?6@X~)f7^sE!dR~k)+|b{mE(g^9{L3A7fh-S6STyUZ6!qm^23siXjH^*FOK!^ywH0&`c0K3`hyWumGSw9gg<^@^tfrNMe-I-w##G_ z-*_xvRoyp+9yeS$VY=5O>I3eqc&;#vzOoFas+!zU9;V)@r z=OYqiE<8WtX}^WGiFzOQPLBn>0?S!IYrY!g6exEl)Uz`y$HT`H|7~2r-yZwEE ztIH+2z1NFwVXw|kzh7r$Cj#W(ssmJ)0=$zJD#BxW;Q=(VWW(lC*R61r!322xIU+u4 z+3iZi20|&rwwG!Gf@auP8&zum9#WJ*DqcQ9&df@Pb_V=A2TYu zPej4(FTT+;Lh1N3?A7O~`re(d%a#ntA5ShW(!8$Mch7ygDZ8KcR9G5CVrbVE<}M_g zs0Z-ZaQBPZ^Gk~2K=_gmR8lFeEi9(0<9#_+&B<+8b=~Z!Ro*tTHYsv|L%bMXCNuSH z(0r4mi)qv{$55w&tx7CGPT08G2&Tt+=x>WyjPS(YF;S@Ds%$(x#ou-dQm!WHnn#Cz zIi?U8z2v{LSQ}8LRkg3haC+EEn^5b#H?dl4`0z0A63O(+Gwew}{j*j+&qjr%VhH8Q zUxVsc^OOYOP)n7H#r?}oJw?rt<0~a;v5YGIcA1*hfrh zi@ttdh_~{$OPNvuWe}hySH8$N*j`}~RKh2Mdh^)2kbX@SNM?Wc0AlPL`%1#rm0~et zhamJvub;Q-O@Upi?c48pfp>NwxG1gF?tonM$NWvJoF*N*N5w_VZ0{`j*%IGAY9j|& z1(Q~lQlw(TPSwUHtxT=Z1C7kFG~AMVfvkoMLtn5rBfywR36w$#(UeFUdNF;o6-Q?? z{VYw^{>El8zI0Gj!$(oman(Vj!Og>`_lCx%pTAZ-Bs-^^%D@MoC5Z}$`hlZcQOb*% z=b%Aig5f?WS?{QW=?n72jDr4~3QCf+aCP)^-cfity-@`-Y0Itl zwD4`KaDj<%e0AqJ{(MJPCM^b>9J zK41Le4i#KExCbv`7=5v%_W{uFM_A&0yLfhW%)~a9+gsV0&q3rg`xU;V8Cn7OV7jKO zT62H8O5G^3_zo*^VZjtw0xiI1k|>M4Fz8MTeEucBLU70 zdOMQWD&-Y5+9nlyGjK<3)LFNMP=BM;V3C8pXr`>Whuk@p)*gF8fWuz~gL{s`y?~EJ zQ;NNCG7ghEd!NoXVhT4=TEAEcfXVovoT+qfMk(hW#$+I*{|M4oEOHhi*cwC z?B0lPMPUdg(qGL7v4;`_E?pO7g^4~4YzFT+e4_p zph!VTDPLYkQQaQ^+a{3cLMC`F>KKg#ZS`+NP?N6!Aj_feyZ@aYzZ2^+2JiDe# z#`)N9kw68p7&=CGQk;`w%W2jYgf35B1>1HStn4eBc2XeNOIu?Z+ui+sbZQyX2S;!~ zNJ2^KJBb21gu%Owb=Sd$HEq8nN=?lFY4(q(b;-^@X3k(_)rP3N8DB{KNenyPtf~=h zw!v{h88SY~Gi7(>dcH0$;%%5_60vf`z~6=W@IG z%J129MXxP#Wd?Zqlz%}mNl|%ZEh6rQzHZ{p=}CRf>OexiBVSy3#u5tQcqh`qd}&%r zc2cNJ`EO>Yse`Kan`wVi4p^w87}T9-Z)YhDX~_Pt==3#LiBPoP5whg-NcumBru^8 zd9tw5OcJ{@og91_cn_K6Yo<5+#o>R`EA2?_JU@DS^Sf+bsV(sQTPFY0*& z$q))71#b()hO8x-P8+M)7qZkp>rHAP+^=dxLjt?;bciJA75E{UXLo(@DzIQN4=sWdp#S-@fZ1fdAy#N zksO}R@6g4f730rPS3&S{<|D+F3HPN=Sv&`8G^8RCv!naoB0l1^Iu+sbGy#ll{!)t9 z6BCdcet-k~s8vmaL-Mhh-XbL;B*DPP{7cQlk{7Q*OjFi_{&8dk{hDWCu=sQMBm2v- zdO7okfMcY%D0H9q;4DPV=8%g#5=DsB%`Y-V=Bk1|&92UMhzNNG5nX=xKyza#0wb}w zn5r8nWe;Ugqml^(qu6>*14fkbMos^MR(rV3 z>W_pXKg-3WiZ)Q(b{KX3?#Ju!-V6mzRgeh}2G%?@;gZUa#vV^y`G_PcT1{*RcYI&| zp-WQyBU2IF#?q)LojEbGYO3NtgjqXm*vMp1v|f&e@Qi&7TA0EwTJ&pG2}gdO=q9h0 zn9s-;dD@L+8jn)axcjmj;t{O1b4FDSk8n^&POEbz;p5Wvs7kXd^O=Q?zu$-w%MiM> zofLGtR0IGdBy`>Q*#G7*kjR8&_6&uFgQ^-g7VS}gIQF%+*o{-d;2+W?$`CfWvCgrC zPq})-)iEPY*;P6M&p>5G52GID(vB0wayl&_{v&>X=zXgpuNA>fe!peHE&llE@FOiMU}8#)C*zj3R9= zocf=q$IK|P1pdP{4iQrp8sKn*%l*&y_SROTQPk8BKR>BCA4-$m#Tr>+LPy0zCrX_U z&t{4EcgpRAlO}CIpO`EU`0w80ihUdrNI|r9EZXXQSxH>M5jL%{ADe!UC#y$KEmVKG zza+tq;DjQOUNO#n8U=QFHn%)!7;7g!jFxDjJ0x*6ek+hA0zKnrw)RIMh%jP5g6m~D zJp5H~Nr)>sRaTHQ(tey1t_BOr06d>8P5Yo%JPtQRV{wpYW|3&Pbo}IeCl&c21ZH%M z&ZeG4fC3AjQoagl1s#4^8a_fi5N0&$D*$^Wu!r!E{I8r?OD;nJdfv?Xx$<5qN>$DO zPWptWGG#^`-e@RH`5Om7&6Ha-+6;xo;e}-a&G#V2tIYom n5U2tj-8YeTUGo2ZuuGq!e*_Oh1;_L-&`&`|S-M)%IQV}6JAEqn diff --git a/1-Introduction/1-intro-to-ML/translations/images/hype.png b/1-Introduction/1-intro-to-ML/translations/images/hype.png deleted file mode 100644 index 89469139569edcd878f420a9e8c52f463e609937..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 154886 zcmb6B2RxSV`v#6nsSqtAWmd|T%y_Jjo$+{(z4zW*l!PK7W$(TBmX%OskH;R_BYXQ_ zH@&}~%Jcude!pHX8Mo`c)_EQ0aUSP+@sWNijDvL(3k?koM^r>m77Y!P6blvEpqM?cSgoIv@3u+>GK7G_F@K_YR1G7>zl!5Xk z0iRRx)%SdlF)1jh`a*;-dOdwFQu#9YQk4+!^`O7j(!82MHRDM~(_E&Q6T9j*dev@( zYq@f5>Cx6V`z813*2qWJN7fh69){6hORVulOTf;rA9GV%?<;ZOG`phl{5Cn7@^fwL zySJYh7+m4g(c3R~DY+ce^vl0NC&T^dP+iEUCCCEp`QxikC&f5sFGlmPQwra0MMtwm z%bjOzxh83KtCof7J$Ws6Pz3DzouCMg27?Dv*NIvMKb}X6s@L-iIz!%mm5Vz(=uJI^ z<82fANvxL_{jS(euk@?Dl)4RL=jPUazjuXAbYDDTvbsScxIrgBNEcQaN?qfOeK(H9 zuJy=~_(-{EEtU;dU|a;PscJ(F_8IN(6->W9F99GMe^@IHht7Aaelj+%zR#o-!w%V5>)6+^6SaG3Vn$#^p+Ehqx%Gh;$raAJ=a}dC zE;^8ts}3uA~u_6rpEn` zXD-wRu4r)(wZ60-gE`QupexiijyaWJRAG+1*c($=xq%fkDNRGA8jR75*&yf@GcF@a z#bm^qhK6{PCfFSpCcT)#mEzWQagO<>UsBwx0d@iH02ztb&lu+R^mgSRi8JvtPByo8 zah=e(eGmO9v{{&b-XOe$ecey=3tio$wl5RnWBy0lkIo+{zFnXrNcUv@!l`wTYuSgl zfVDudz+*yEM1*`xQtv-z!ND5ws1=9MQ+p!MVyC3CWTHg81X)5cc)x;J?Si@ARzu^2 z*Y?J?);`m|(7x%u^*#rGpoWMA#mg%tFNtf}YBe8UA0yCw{otwG?VHz!UI*6WjdN;Y z=6-PS>JxUkfB$~+{b@SFd-vZd(<7J=rpOe^P2y&QaX)15Cl~t;;8u%6-$FUaWH%0vNAquX(9>X?b;G z72`d@!}NXPDc-g~9eyRA-AIO>bvaqtlGTEUQWszkQPG6j4 zoaQ&?)=B5DZ$z#Kt=rEQ&js9^xk2|p(NJ1@P(J86F7hw=!ulSsn;nbXn}_V;!jcq|Xz;Zhllj>YR3fzV~)Bb9024n4})J5Vy|% zyg#P@l>g-jgQu}il_RnqX-8B>^m0VoA1;iKzMryco|00)?v?A+&SLy|t8Mx_FU%M~o{kL?lQNA^LEB{&zPnG0==?hm9RT6IxjU8g)JI4kG zzlt`Dfrqn-#?2(Qt#%08o1mYuS>47x_t*H8UpAwR`un2ET3N5eg~a6R4=|Z98?RO6 z3G{_5IB6bBLSZcE@7=^E(MwapNG@>>-;_{8w3oLRWRW1I%_ z?lU^WZ*dfFWxTrtcVF0$Be~1z_B!)sbdy7A=?NK@zcRJ-N36km*|(DMXYeENGihxV zvXmd?_~y=y8X4G*&-JkG8n5Ie2OE#M(W7kg5A>qhiyS>P$ac)jKtg)#3GiyiK*o2?IedG#HyP~m-bI#Mm ztd+YQ)Z9C((=Vf!qp@J!P6M+xi$rzIQ~Ig;*mJSF5xec@-2~jVRkXKCpIc{2W=ZP1 z=k8B^bsSpLB@AXXDHo~uXm2?#yOfx#C$lCZZq=eO-_bMNk{h~Q^hFM{?n6}*A zviN3Gymv4wvUVBYRfcw&f}ZSO5Ev z=IbOypUFi!w2eZXcmV zBwg-0@<+!5tkzdoQB+S-5{(9YeG%>4nVV=B;Hxv>gXhex|NHvM87j21$3I6$L-RI7 zJNNr{o`TQNzjxpRYIE}W?E6<}7r=Mdz=uN&`hUNTNg8wZzpu}I1JBU-;R2$f;4@s? zN>|s++Q8g~z}VIw`~cHJ`b%=*sBOa6P0psuyHm7#@=p}843^j=LZb6Xqk+qa>P{_o$(ICUKi z|JRe5_3vqc2{J-Y7?~IzF#g}WK~os?Q%-3^2VIkAf`+DG%|IVKOb;J&z>XW7di1|8 z|7fcCzorkFnOXj9`p2Vxn7` z7`pHg4^j5zSHLAtop)4X4>Zs&y?*V9dDa)x7RT?AexNM{fpZ-k5c(ozP4eH(Oe{d zj)wl%e|%ZsxkEGl<*2j&ycK-=mGzAJ1n;>3tHtn|wNp*e(8H-ljYkcym)Fex_0Hd2 z7+zapJrr1>l4N|{a{J$Y(a`&8Kx1m<=xDb)ryrIwj2iEIkIH1Z!yS13o-H)Ym|*m( z^HE}s%DBR(XMBa~{>iYLXT2J#|#srH^;ZX2Njl5Xs^vE6}KKFixC-IP>+Nf34&%N81YvNsw z?)}<7y`FrY&$B&}U$*4=o*oqWI|}rwFYXsfb-!XK-a^qf>3EZar(?-ZKSV?Sb<((@ zJnz-%cTZMA{Z@FNJ3XB%U!_kL`r$f8C;jQQ)UXE2;V__bFkOYJ+Xn38#geOFcD{Pr zrYo+mu&xB#TfiRtZLrV?-%kIYRx?WFK0QogNgLN5{z~$#L5IKRG=q zD>Z13S_OVRNAcnhSjR&SPE)swR%Ty+I@{`7*3LLCQXA6k&0nDCkElhxXBWmLgAZM9#W&ZwW2dUm#6q z-+P>_PELyAkW-04do0td&vkkmo`3#{f4olyYgNKa{-;N#bG*lIoF8z$JpGV-^aOgw zvAs|{ymB9c%cZys6uWd$=grAV6RnBOUOYXncSO)OJ)AI{!8rZ!IbYmKW1qYX6iDm{ zka(23jmXjE^!Af$W1N6xpMpdq3akBo2edU@TaPM7v4aVcdZg+nlFc7#8QS5HCxrn)$_sGc6wnR8;aC;b{uJt9S6c$Sb=K5DI>pg8T8 zp8amx~xf7FS+mN*2LD5(n+9aztUXA>x&yv-X6YPlC z5nCZYy&%uYppM;~>JX?V>~aFMlY8GSxKV6k3kKv4ADtaSu?tr6zxzG#UC}t*FJC|8 zN<$WxB2by^JF}AscN`>@pqj8QG~v93+*t+`tHFM9gr~kMyt;s5HD!<8!@v*+*6A(e zi>WxVOPXc&#)H!<;&F)(@-1TlqFX2)R-7DRD)Z{cFjRZ20@>widOkjijy0@Lmb##H z5_PQSZB_SLrdMD8jOE>3(-8@{PqvpYSbhsUj0x2 z8nhk^r}C7e1+x&rJ8>>DA8b#*o4gG&kEus@0s>B>l1|2N+}ZP;y>WUIa~z>Dn??;= zE1Z5Yn#cPSyrU5fJ3wJtQIHdFW@aNLJiX9gZ=P6I-p`OunbWV1aXB_iDf3e7(}Q}> zXZYJXGGUD<0PoQ;;5B}7fSrCZx$$qn8^}f@L9xFb5b#iQS5Bmzo)DnRC(g7zsS2|d zMPsXDe{A9IUs)@2txJ#W2$wT;ZWPWp~!qxCfh-Og0y8Lc)chloOe9o58P}+C=U4QZ^WzE zXsn38R4Ni^0Vwb~$zMvxMyQio6>;_J{jSv!o~<7XL<`MTOTFP>lLL zL>|Q4Y>QAho{J{GA&+e!*Z>4#|OJ|~SSoAXd`mdRuDfay>EI!$Ul-uwnj zZFMFR6cmsx05Xs7oz6?Bq891*C%Z%`nooAxw7xP!KF#sVmkAUL{S4SU#7lcBtT<7; zI{3KX)meIb6kvFTD)Xe@#E@wR6lc8g81Oi*2cSADznebW!;hIby_9@T5bz`>+pD1T z4)7E780ccJ*+J3R|JXQ~2uLrWSPi0+Gi>#fw&n#23g`xPyfcfVfNH`kC+9G5nZ0h~ z^f~pQJ9dXtmuyh&!n5D~ZqnAiKylP`AwMivRBlax%4EwoUKE8`$!XJ(&oY$CC zj^g3)Z^UC)CnCi=y&xhWn)(bkvY((jy@hBVs|3d`Eu~_FcngJHK0J2fuXyhbp?J9Z zoAXp0VEr0JF)%C0E}ODj`*%s2WwQu`HxLdd!D z8d_>Rdl~uDTX^Q&!{2cE#xcZ3F`OLFP3%m(@V)mE_1^21fTU>rHcQoezdDl zzF35;%_pQ&)Q}8g!R&oTRr&dTw2PMWLawmG_?-Zl)^xPHTS+VW-@E?*#s2pd$Rlgw z??*Y*KJh0ubY8A3R&|<9l&^L>sIa*25PaARGaKZ%1#`$@HR}4xP`>aGY1v59=?-7| ziQaRs7{_?N}+qd-E~35iPi$k?`Q{H^Na zn)mmEoy*FH${v}D2n|HKTn;BItzsoTZQs*OnrHJ#_?eHHdNmXdFDB4Btarhs`k(1X z;y)&J-~Yv?RsY(8JCyJu7}sL!C+5b10>kUW36>LIiVWK0EFv-fw7akB(8li2ehorF zWj_7(zwfv9n=)`VoA>5Td#c&{zPuI1eQh$3rzeCg7SQD?QbrJp_%3KMwVmt0hwX0k zs%Hv#LpDKsL5Jm57ycTM z&9S8&Myy{xZE0T>PSE7`NX(8n`#_9BPhSSv(Q6Y)(_Eu&EW35L?hL778hMXvBB`-c zeKKIfd#@h3o8Pw z1ZIa|FJiSaiIuO|cL@ImS23`>qp4l&KhOQS0KXZwht+RNzFEd|X!l}>*~K;^<4v@E zZ3^yNKm&jBuE;@S|_9IO`^Jdg_)$OTJb(8(=Uu@>XFy%s{kU~>rPtk~* z^H#Fc#$3AqTuGjcXLsh#-tK(1Dz7b$D%^db&{!XlCMP<}>2m?2^V96kI$_3>w>NvL zTtTB}`FcLygi4Im4(kV7Bjs-M`I6ymMHBv91GroZ(pv7lnaa}(p10V*87~?HNjcX2 z{n_MblkZ`yW|i`pN#aW2&DZ;G)s|s6*VXueUin_70?jS=!yWx_ z4$HXj(YzyI{Os$Moz}vYe&rvk)cHHt^>*Wvs_o%P5|&wEs|z2s{V_+@va1h`pdb5{ znHLs2AoLNxvk_8b8OFGjA(X&-f2fUW-zNr`P z#nA-LG`7Jj?-^l5uNFtfcs(LtsuET1g6$m8^d=$LlsQ9&K?|0HZ4WtesMiw8czFv0 z8|NlV&MmW&nyiv>UyUGaJr##Bugwl|MlT8v3w;xQ%_whDr?39pn|(7sZ-=_?EwY!d zwQ?<{CP(FYVH4it&O)Ti+clEd_le$*)WN4okhc``2xugA{zmGyt+}^UzWiQIVa&Vc z#RX+hHWswh`$eZFKC@_`xs-c*^hJ?Q)BDHMG+7W>35XBN;F&-2MU+Afsg3LixD%yGLCc@ z_gSi>sn1KL@750(*8er$u;jcu6CZHzAd4pDnu`Mbz+WParD=C#e)_R6mMSHi>U5j? zVg8WY-eOOjkYD}KSXnF1W0wI3(A_|0@vu>-ij@!dulnnj?o#Uu!1;x9vF{GaN*m60 zBqpJ$9UUGFZCSO)35{`kddXEObIMZ`jh4sSv%74sj9welhqhkOqlCE)*7@ZP(RFOk zEF89SsZX=zyWg1GN}l%?AFB!IlWlNWRi%?=j}|Q`De8+l8|i^gJOSo<2MsaUVK3x}ueCUiKh&uE%gZpa5BK6)Bc zeeNo^H&-CK-zVRdTuV6}<(%bQrGs#d8@r@Bfi$l!OwXB59EQxiI z(m3^%d0b6=6t{Dk+gv{$F>O|NmMU`CbxT#@SyIV_TmT5Mxs_(Hg9Ih$u@ z*|#b^Ee4?iPZ8&T;`L5GOJx?9(m^ib#@-*Ef~x#Dg57$~eA=m>IoClAp=|aGVM6YL zW&%cPmCE{QKdz&hn|2z>3u#WlH(&rZl@vV$5hi7}^LlA?33Gg5WwaoXZH7gi`mkfL z)OzS1LVT_(H7NX=XeTGNbZi<6KRiYlE`g7b6zOEJT^{Dna(bK9pRT|NpSq54-<|Ts+`>mgvh$i+7IUcgx8L_x2X(^209*j$lakC(Q{vd55_%8ib)s)#T?rolcrh z*j?b1NBGF4&VwE8O*1ZE$n%@+pNd_6oh0ffe?fL6Su&hJ!HbGRAzDAe@)h=d45XZ+ zaCd^u5U1^2iG9-Za8G9bPOh%4d;P&?J~ej^T(Qz{FsA31UsO2@VFcI=0uUg6wYzQ~ zn=wMKnqE+IXJo5)%QCxi(+a=C#*~$asv>EzX5G#MzV04iLsxT$`o|xexdO4lCik|I zzdh**V|;%3jQIp_?$HJNPfB!rvN62<18(~}0iHbUnIe19Tsz|t zWH(b;8FpUA&rk;^){%ftQQBfMQ1}bUt%zILdU*9oPiS6y*^RgQV1!_`@*eG%SZ{&@ zQlcG4n>#8pn47Bs=cm)aUW6!Kq*~%o1;B@)=qXU!7QlPR(JIzhb+CyXJFNHO zaow#~RbDm+S!7G^hR3WJxvAAyTHP_V+EUybT_xIL#(nkF0;8frkVZvF5uF7sOyK6SPo4W$2y4TKYc`5QLQpiWU71x+p1wm z{iysp_80v6B!)L!@{Nt2!j_GHiyUs}@>~ZS1JF?gw^j;(%SjE(!<<~3oo=z&NiH0j zM-GESiQ&8(ru2+;NJ*>Vj$|ea2qqMU?tS_|Lz>DdO7{}4cn>%oyKc>KwnKZN5Opnn zTm`Qu;Whz^?;5TpKk&oQQRxmEy&0h%Qv7}zBCf0&m(P5)oVIqox zfHO~Vo4QYR20qN6NG-XM==+#!u>c-BA=?j7RNZZ7%5ubStj=4RPB3VlAonPLv+Eu0dzNs8#yrRR`_5Jn_ioZm_ zZwQYh*6#_oiTB!zZ&8Ui+)*DfMfe(xRJkg;ijM?gUcaXZ&?!EEjKO`gUq5M1rDjB6 z5ft5it^3n~sbN$cjDI)IE18HwpGKzxxG6`O4dx>x^1+_ zx%W+eM*7smmoJsIXyL2+E_N68su^=K7+fH>c#chrG2g_qT(S1l53z%3Z+J6Xu`tL& z75iv^ns+hEc?q+42t%Nx(m}7RX*?0PTS@}ep}fh zmW+WBon$od5TE(Shzk{kNC@dt0f%%D)8Y_pyB|8P#4;EsEuL{rHnt9qGa~pxcrG;Q zHJmT4gdqLvB)U51J^3@uQGiZ9_HLx%f0j`jgd9!~z0MvPmPACGMs3kqW^2rj5#|4A zv7mE2iI#S*LI(+Vw8JRA$!di6G&gCs2e5^p1*@^5rl9-fsS}el;Y)9w2gM% z+bjgst4vq2NnT}C@&dI9K2GWPk4$xv0@DmDBS&}72duYhBO)^Z#cr>|qEDVLPrMaE z=`cB8PCubp_sXkn4|vVK=)?773-10}z&y-^No?gx{6=L&YX06Sz&{)p@d#09Ay-!J zK@QOTh44YNvT3KD*~;c)-|`>y`X-u-#i??G6hwuLOb&SAG;~=HpXRml>`VsP0i*CP zKY!_lMz4ysWJ#0>zekM0SB>?VcFS(x_mM%(MMHMT?{Y*!$ILi3ZpJZ}IzvU8G9@f- zC16Qa7QPYF7pyp@sk987XGzozTp0{13|n7guzR_H#C|1-L+e_uK3aM#*={Y1IWn_t zup)w=0%G~_8Qu2`z${udyQvYwpC0Op4O(Iivw%aFA+qLvwB>HepHuSU4sF_Mf|#h0 z6PeSiyaKtX1G1xSZ)y#K0`_5tIuPREXAtZ>t9EnsldSx?sqQ!x!e%l0`F4P$uCNyQ zy=M*C^`q^M(~A?R6iDK ze;`tFCojh63Agj+rQ!tH#k;VQj~HGr@UUdzIeNS6vzEf?wco(8ATkqAj0?i;u~Eu> zgj=^(G+;yvZ-|Krp(O6eR_84T5#kH+k}FTNHp1m<*~|Ezc?u7e+ZX#>zhAajY~0%_ z`G~;oUeO95Ye*vfh^uB|tX4k23nBUrV@#t}=5nyP@BKIF5S=9)NoouEM9x* z1yFK?j>+=}AA>3hJd)Qp->0KcIW7i+vKrt505^2D3i*DR>*6V_5Uo_VX`)b?v0OIT zCQiv0*hh`H#5w-@PVM(QPu}E!*pWv?swgJ)GcMpi7QHKDb&K`ar7rdQDcxWHR;8CV}f-nNE(RLa*F-bBPrP33=5XQPpEr`d0I4>)mn2@x! zHej6X1kgRs5iSf?G|px zH;K6w^p7|AnRcbfY=UvU8mXid|Dbp!;#sLxA_&#*!?>R%KJXMi3YRte`R!9uvkL$G zNVPjhh-MCb|NWYH9U7BA&h6)Fuu%7pxF>x73@`sv%y^Dt3DXsW&nv8(O}+E36@J&U zEs>?Hrh7l}#bFF^+^-Oc6o&JObZWZpOodXz0dhU{@&~F^qa<7f!?Ytova)|vNv?$N`N8M8TMYwj;EeK+$^r{&(W+K zM>q3nq|7qLDuD=z={<+_>d|&Uk90xwX)v<}bg%|QFP8mXx+^d-G|fLmbPpCtD;9%q zmC*eqlAKKK^F1g{)r1^vk2lXKv^-KO_Q|Fw?7X1brp&A`7`Bi~u~&j$o6T z0>v{jKnuwP7zyMB?M@pte4716MNND}w2U|btPBKE5+Tl>Y}#KgL6YN(?uYI_wN_?I zCj60;)L+k~`3i>hWERkejLLujs~HxeT)@je6*p7Fr2R#4dAL&2Bh~=kecyE%fV@t3 zx`KT;qpVatFEKOUOS=eFc;lj5e8Pm)t}# z&;R*N^MUiu+lk%4ZwB@nN*Bj*I|+Tiy`?|yFM&1Z<;2T`W_&(spe-| zCl%2aCCa23!&;Q0bVzV(i{eH`rcdR!a;E>VmF>6q&e~f~l4RH?hdEz7;%1G+ zH!I&B_mgakqY?L^52`Sg9ri_F*lns-Kl4Z?l4lJUVE~EaJ>YJY9Nb-3dn|7xR&q)Zor-!Ua5;=ky1-{ z+F@z16p{@rxQDP)&jrW@8i2!^MQb`xfGHpn zK2Y2m$%U*w++egPLiB*ZA!4y?K1)cA4NU%Az#$<_sl+0d)pUSmo%n&qbIMoMwUL1s zp2Alz5&W_3BJmKyvDgKQqe3_nQ2h5Rx&nu^>99v%0lEI{7TXwWu*;kz@WbUM@YKn( z%rmd$(D8hRBo$U>UfnXw;aV!G8_E(3p(!%ye~8rbqrYZOR;(rqT%`#MLjK1RU{B?n z1IE9Nn~G^WWX?uW_lX1t6M(F41$9%fO#xi+&PBtp$}m>;%;vti?(|GxDMgxf?uT3a zBLWC#nykdttyiP%N1noA15e#Udo!L*C^D9eq65|5!m{3stS0`N567RyER0azwZi6= zABqgtMF)_EIt=Qqi8r4%FStc^G_==($xt#;6tdXUi7kaeu+Y}k2F=k`N&!qGP>x8{ zw0Ovl8AC?Qh!RAOZb0ZWsqw^ElD$Ccpy3cOo^=42R6o9g%2m?19NFZjk+mBtmq=ZY z4z_wgtiD3TSh+6iJnC_wcEg4-;Tr5Fr%g(v+n_nb7V05@EDETazJWy%v=Xc2wc8p7Y#^(ZjJ2n5zI3FJ8 z<>s#{pvvg1{HHEBQEa_Y)FyK;9G)(>>`l8tYh{D{#G0w<=R2GoxrCi`Ekqdy-T>e~l z3s1KOP0gLkH4^*5_MR-&`LcwV;HAiBSKEUj`;p2Y@ps*#$z0cE*uSgCp1pI6p=w*} z7K?#&iSFdrK)XT9Mz48*ZFpp`GC(Xc)vMiN+krZ!4M;6cZxB6ff2Gc@i8AzMx!hIg zuweH-`@e$@9e!TOPWIPdrK7TwZ78=Fv{e(2$jzu_mD)GMb|0SEAc-7KFWZjGtCjPV3k70)ioOyu5IXAabeFf9R)_}hJ3W@LXiQCz(gp`g>t)dT7uxaHp zEW$TO+}%41{ldPO9`5h-lv?X|a(Cg$$|_P1STQ*yOgnE4*-Ii~J$nj`72pq&&4$Wq zk{(Cr!GLaPy0=(d45Gy*h}}MJOvnSSKQJqP#rH8CP*X%gKGDe2ac&bvST|EA_knF8 z-C81KP%!{vI1Yn8l>Cy6;AqFI5bME{=#bNFXEW)0z_rl@j}N~{+CcMGC_L$@oCD*w zbINiPP{e)rt7e>S0g`hyj&cIfEmw6v*r=iT|12r9_R(q0pp zK2Tbj=IAyxWnT)Gko#zVElES<$6Rl=IV9m>+sLIz*O;mSZgOz2>|T6$u!jV|=qWIy z4-nM^&^=3}XULhe=j#z-vwgYohGALt@7{Q_r@h0Iz)ajT+AThd%Rycey!WkpG?4V^ znJ}O+GPtOD2^Cbe@$9`s7CZoS{T7c)iQ!F7#Fh-#LQb8@w~tI_r3FW#tPTZw-)|}m z%>ju4+ZUHCS&UnJFt2`;2gww2LsYrc+X<<&ac_Nvf7RUSAGD~$NlFvqM;fJPE_e~D>t)M6loV|bTI0iA;Q&uYw3C?g$IjH}z{b?pcz)%{k-X+P#+ za#9pOQ*G|rw2EH7=Eb@R!am(2(t6+Y2p0=mt|VhSc}^@+SdgbMmT!u0AkBiHGEl!W zO_4=_b4xlF0jskpP9wk=JBO%{p-L1Bsa+OCB4!PmgQ>?#vG7x`y-ynHSPfNo19842 z_SXCWRz5LE+m01!P=o@)YGXU!lPSHC^7DH*lrqr+8d5HZKEsEF3V@sMz-yKHm_SDl zb|;Hlv-aHlPxflX~nNEw^VO1z{rdER#GfhX29?)DdIu?PB1kC zFzcn|*t9o7XEmHIgrPq$_?P|vM~NOz8=VXI_O{aC8^|Y6E{zO;)Q0`h2ji}kKvt8! zAM2@P_Om0FET>W^Utzzqj>~q7yXd!gG%G82+BuO|lo=$A`xD|K?F?+^a~9ocY=VCd zN^r4(mCuuTz~Ai^cP2(hFYVPjhz$Draozu4$e>N9WRfp? z`vZZ=cNQ{m^UA~B`69rGUn1kMONn6})-)w+q1F2hq>ht;Guniu8Ac*lhLrN}XjWg| zzKvjiFDhYO|p*?1JUZ~nz$`4~TzQ^sdGFp?>+zRp+K}(m}XkI6!A~|hl zu9#TGvLRZ0**t~4UI?$YkFI0hkI*o$KClZ;|6#Saxwr^&Kz5Dixe)ZT5UR@W2>$7P zr65#=+1{UE2xqTMSh4$)x@!c(}i7#)50$#4ilb@l91%q@B;18H#f;DlLk*+%oYE`X5QvI>hF zLlSwoe00dDICcHcd^bc_9ag&{-;_&Koe!wmouhRIL8bU5V#=x<$fF_xBBcW|y#+SM$RkslHNW9}Z_YBROrYC?}bxo>`}orCB9WU!etg|6ML1 zas{pY8g`ne=UV?EP`9C=?%M6oIhXSPN8)(6OtW(X76vk|Q0c(IXfmC70>Ev%b|FI3 z)qvn;EA!UB6YBssJ!(&C(!e#^;~GxBS6C@w-MXQw1e<1H=siWI$p44l2w)x6l`NI6 zTAf`}X?h9Lfoam&YLx@)+Sd?I!hqi&9|I}vz5c7h#4-k4w@$%y%>GH;63vl)BgrvI zzm2}zy#SKt@bPRl#IFqHVzbSf3uGnvK=?3A6HGK*Huy8%9tE28=b#Za-PZ0^YKH>1 z1E*UY7BTdyWh_e=DV&QyKhfNn>-IyEW~A({ShsOcC-vYKA%J6Tg{?^_yyfWoa{pBD zX98-=p}SaYZF>`0$DJ{GfUvH=U6{-iNIwNY9&0!1%~DP9b5jL0ar)IB5ccJ6v6(3T zQ>C*Fl@U#53xBdrbipDuY@P>&B@LBml9Ji((ZI9On>mj+b1heBL$EUVb8``i1cZsS z`0eSsV=DKPAvGTjSqej_>+$ikCkA$gYPnsOQ6QgO6yVOw#aXK5N|{Q9kK@6Le>bc6 z^~8B5*4qRij+t}(leZ>70qp_8zzPS!pWE;`8qzV`xtlX+r{F6y{#;8rx-B0XDxZU} zlQuC;2nZ71P%<~ky%$5W+t?h)f5BoIFhFFauUT7(CqW9{OZCUca5isw~UDw_2(@`#jFg;B5&X z*T4n>Ws@<6Y~$3lt_sJs!DJdS>QRIoZ=9=GQN}z}MLG7#$n;~Rn-}Oy?D)~c$q^e2#XyToS1Gm1 zgZKS&1M8jM$r5+$$Cbb!o8W6l2L%O@I5!@UFFgn0-7{}ok_<2zKyv7|^Z`+@Y`jFq zGx!v6(h}1@*~# z*cOya`v2TUB6mFjo5W=jOWk8b#yfZZGN}pspoKuJ`j2~D2+aik7w4Y=40*q(ck9oy z=>YiL%48}Rk^LtZOwzX`Ts>Zjag3v1?*tTp;t%$+)IME|*HBG>{Wl74;&D$#%UjyC zSnE`S@iAgFk|B0@gYE04LWT7!5`dPezMlThKJq60W$=QF*K3BdncI6{{0pD-qjKxX z8tz8LlXXzJB>a%g`_LD#=D$4HkjHF+Km8;Pyr)5_oN zS9XAdN}9zt%&j?~$f0bdweLgy;arKTU4O%q08-VNZBXOU29=i;9{@&{3zB>HaO@jM zZYay#5aWO7mHQW}0jUkTSOb;a?o==?9hV_XzxfPko(UHDTjED~KNvL;R;txuFp%+I zND4Qj`PG}zF!)OKPiTAxE*p6y9e~@L{HZO}X-}D+5TC}E zCYwOr*cA$lo6T&FWk~r*Pp$17ZCz?#EYk}h<(BB203rG?2nnW>M8q>*C+Ks|CIrPLmrcd(D$jiGdj&|@Nag)S zIL9pTdURc6rL$_ghhNP%7y37WR@9*Sv@Ls7ew_IWN&P?f51tdA;8#TjE(40X^+|uo z8ceQ&t6ck%km%@@v7nFN863O zDk4;60wGpy+*^Y)_lhIvVDRLpB;`Q*Q2j{l|0kHYv6(+Yh1MuC4^yi154FxWx0s~Eu=1IpkV>KJ>lh0N& z4cp8Dx3-rm?1vp2hu!u}i;4Gz5qVERlI-yaiC3i>sK6S1u+^Wb%+9{3H2CDNsvyAI zD}Gx$HMcf{cp&)fP70=3tF(*E!Yn+cE`|n+kh)oU$o27m0)ddh=j- zig2O$7;2Rb5G^HsrFjChfD9mw&~pq<#_1=_$%xecnXm*GfC+xP#Z{*gsz2fQL@#<* z2}L@2LeEEK#@G`r+E{y4t``e*YHTcz@S24O;A7ak9c=Wj$YNcfE`h78Oop3R(5pnI5P{9{2ajl8ELRdWCa~U`PiTf^B!`INKBXD2H+FpMn_w z{Vf2W66#(Wxs*|KB`1FaT2pnKCG7xiy%A6bwf+!dx}~Z8<7^+abXO;vHo-A$jBGD9 zS4VqsOI;?#pMhQ~k|Dna-kEsX&SGE{P?Wc> z;BgNlJT4O`ifMNYv8oNY#$H}G3EkQ4j^quHqd7)7q>zs zv!If!hEDQqF{to>NFA5i!R|&85XDQZT`&To9Dfq+f$FXCtV}UEN>}}RPa_Pl$}#A> zm-#GUe`#~TlO{r9x4=ru`eua3|570TeKlDC;i8HVAZjvNLKKy4Iv!z?j%@(o3a{RI zQm;A=%2IPHw}!dQl=JW-lh@)UJL^~tevl6aFV@_VBhi75~NOwq}MC&Wc`t) z`EQ6iiO>);T( zuo9G?Vi^O*;ot%m-)UGP&7r`ko2DFO@9U=9YeAX6miysa^h6Q0%{r*FEuVX0$tA1u z3ma}0U`R5viMi7Rl`&zM4w^2}*(27COHpf5q1R2g( zZioq&-d~Hp4;HvRzm3NjNM(M`Z<8bK!k~nwCDK=$Q7JhWg<3{kCM?m+-AMNsTdR*_u>}oIaX3j_)g^%kNx+g|QgQ#y%GVmB$(O z%fr;w10dBku>EUd!av}=pBt!$E(b#F_$HalN=*_kWnsOq1U`+NMdy1&h9VQJTX+S4 z5N;kQ@qE=Qq2lnXfp8I=bRT4diC9_3rVwESw{sDMSeCs@-RYrFN#h`!<+vbJ%Uf$p z52;#q<39Jlm}1Mb;6p(00#l9#?{lRd^a!;?29FmoCgc?p1Tn z?9NsnQC$NF@xaRla+t#cXKr9<`WkaqfxlO|xwl;H?cJTJ+?!Q>xC;u35^>sQ%A+~D z2Yy~UzBmH%LQ^IB9W2ISX7mPtc0{;>m8eNWWU2Ifxvqc8NMv?k0h;M8W<7~P)3!&` z#Bhu||42N1dQch@(4A+?A;Tz=8{rV1b?UjlW3eSy_Y@u%Cm`lrDw!|=^^Y_%Taaqn z{U$EW^ekHmpHGBuKTIHWFJ)*uZ-(1duw- zP=s^Yn+-BUH*Ag*naJm7&O&Pam=nldE#|c_VE0}p>>z^e0@q|Z^CQ->co66ki-1W7 zAiw;>=|2BF!7hWfYH_ENJ@^k4j%TD33?aiCC#rzRrsPqXyysHDx);E)$_KF=VQLzU zXaH>WHB?4#*+`X>Fe+OBkj$A4R3gvZV-NY5geTsErz1;F!}QJ?@UlKoJAKPu?4AOw zgGEO_(pEmO`#GqD6?O*k&O{W5F*&oeSv9FgJ@za7o0jQQiAR?41usLsm*{kb*>%Q$6~*JbRjMYCv>3Wlk@(;xp((|(~Nb`5)-$J zGmPUP6dtNu0tT)5X=UHRDeBta-illSB4tx)Qal&|>$r%TCm|+rk_aDAg`*8o!|(yx zGIMaY&B`+S^Yw+sc%Poqv z#cXBgbb%rxLr}#*YeW36NVb1+FL-jRa&oYH9sA+IL&>vfGy$14ksz6-3{tU!*|~Am zSaOg&@ddNRH9r{ww&A=O5GAP56gO`yB!pbP)?XHibq~tFuoU zfO94gbF$|AllD8v5HMdhkQ(X^XEPlTWcO}qN}B0t1-!G54QZkc-EshU+5{?Pv_WNM z>SZhz)531~_U~qoIzjbbf(c7c5|%91Ov=r`$IA1)*#XX^Z<|zs0^#MU7IaTC4=8F4 zBo{shKUfoo>~V1s>PH+ zir-oOE!UH3#?=xaCTD;b`6TE6bDehdMmWPPRWiYDI;ETT$DnjG#(x>VD3H1#I2dtI((BZP`?PEAiJ@vY zhYCdT18ib}%P*teN|N|=O`)A?>0Thd+vJTJjjMm50@Z~R6losQ)kV@Lf%D6eLLzs{^;EriwSrE3;diqgkKvHcm9eYJUTCsda7%Hq&r) zZq+gyON2l4=foBguh;Q_>T)1Z@8UZ?1vPn|RbSUa6x1vciE(iMrd%Mo+JU0?GU{pH zg>`p(1F=Go10G@d5X?azmDL?9aDfmV)Dnb&F#Q2I>BH^N#Zf<}4?S5ahlXM78-}17 z19GZ_oD!NT0I2%`g}!co^!P&FnZFi@?{oSI$yZYsH&a;vl9!#wZ3KJSLS74WwNI1` zx(>D!Dg|Bt3VFxU=mH=qh%`L?7AEP`T=-P!pTaSk$3C?Cq)W&>o`AT3^A;iq){(tPfXH-;4 z)HbS!B8b3<36KUyiIODKMY58b93*F? z$??_!=6kQScdhTw{rCQHX&>vHs@l7DeD*F!l#~v)vWx5(ySlha_gkP)$`MmW<&oCU zJe7DYN17t;SvIqmIxMSYHiM{?O(HUp)m*>WO#LW!JY+jCp$-(`m5}b*u)RXt=UhuZT0e_f zJ59)2BBT6adJ(OilrpNE!TE*wClLdgTM*U=bZv@TxnlQ2Ekd|@GE!$S8HM0fSOh?6 zmYLUgq${MHMm(Jdybt#;(N+UrNL51Ce(2uC_0 zswFHy{0)W~%eWo(` ztHDgtCz@Prt9D4PhD23$#yBgdqyWmvMAlD$Xd^;4?}GssN)EJJ@vhT`lsR(2?o=ziAhZJM7>!`9+gH>@%s}q_?K733B*PAXaH73na;3fV z%_@Y%jB60JO>Fw(5SMu9U|%aeT^a&QH~TYoT1+dILETE_MG8Pn^;~s;pf4_S|9Mem zxuI>KuA|;i4FHM1=IRLx@2_>%kA^rs@RQEmT7xnGoHct_gyy{iNwoDmetcq4-%*yC z2hsYykw)3bHLG`>qb$f2is&}>Ocn}<4q2x1mr~!Jtp#5t?GV-wM1f+`F{|_OYw?fb zxSsrR^KX?IxwENB1*$Ipdfxz2$_${=(JU*e5*o^jfE^s(X207gJQbja5^|u=kav6e z9fFv95-ZI1BbP;g4hO8B)UbQ(ZJ<|LG7XcT|Dz4?nF|2;Z!<<)`u6Ifa`77=j7C60 z^xu1#R>`A*e@J9{@PIg(Kzz)!P}lDO8z+Q5!M^=o1c!^lOSiUor+yZZ!}LTiyD=7ZRh9Y9^bp6T!9MmIn{?{_VTt|V ziCHFZic`z|hQAa^xHG7gzg|@%`9m72)rE<=m1X`R7scUNmk8+6LCe|z71Ri}dS0kY zHcXaRS2l~YEU{G&1oY)9y+uqUh{pmD>#DHdBL6wEc4YUj6m2K|1)~oYA|MRmp4R(H`Lk9!%L@BJ z3$To=)1&~ZuaJ0c@S)FJU{$uhFjGr?pOQxMz>kRBDYg71$psW^0SLCWpAY99dJJky z>YL5_a{+LuETieFnGir9UQ#E4yb-;K-{Vt&nZf zJnV@ctBqFLyzE2(a%v_Mo2IY`s#-6nI|p6}xmqbC`uI25ZuC0{&$Th?M_ys_t;jm| zT{bVkzS(Z{=vbGl@~+4Kq501uRNe@%t{s*=Xfu-QEPr0J1VC|{&U6%1Cj`@AXQ37W zE>@)b!xx2Dba-9VFiw`3i7+{N)y4;LZxY<*GHk*Lp3FIa=e1WBJ|7pR(0N1uub(RQ zL3-5g8i4!e4i2&@5nhB%HU>`J|H-!UVLK`r|CG$;IyO0^ppcl_1g$%MN`CA$lJjIM zt1vo>d?Lv&5t)0*)1CH_X{i7w$&LIw<7Ma{Pe6G`5Q_-9cP%@XTcNH1Dp6D~APN5G zLI_9MsPFIhbb|ni)ywU2_rvkjLs+ zgt8xlMNb-oJNZRH4bQypr`zX^y54~(myy{(NzOpR7M=PT6$W|`Tw_T8X!hqn4boUS z5*WyGH=TC@afKOH6>!2)-#W)5)t{I_m{h8Fk~yGGP}>44z@BmuXVsWkqicZarR8IwHy^R{rDv}y_*3FZ|E zXQhXt6B~f618;=vzh}}z0tV#K%gXYb0RGUb(4^g|Gnrv2(8(HPG{$(JLi)pTwu7NU zShwefzIv$I0~27%r-cUfug!K^0=xof&EW}QhU>?RLlG;c3%wus0c6KHA?>nxUxCpe zA3_^I^sZjsV0KAC^{H3_Ozu^(&LStXOo^^)&)wfSb13>i_FVq${cS?_esKT}P9g}; z^#<+(Q7OwBRpp)Y(JwRk)o1SVpHhM;M={=h;ik(JUw%Iugvz+$_;sy552-ptwEHu_ z3CJS)_>-I+j(u{bm`~;%3j{kC)i^5?(g!R9$i%N_B1^k|NgyRZX-*bc`em6&E}ffy$F1h^^kvXBz7w;LP4|j{9Sm`?d=&KybT*+NMkEIc1lrdpnBEB^iGA zmHveK@N=V!FGqRyZKejwUm&^rZXYL5ohp}KEu?0G6xCNLkZ(d@#no@AOy1^GU)QQ| z$N=3G8n2QdQc*hv5L{Bk0bu(!0P+w6sE~y*69YMbVvyL=0yg~61_ZuQpxrQM`6+-?^2F)65XDRtk&2PR>_L z5b6(*zX*Gft_|T20br56{}VbA^lAC{QSxV=hnfVrG|_PVszcBh?qk2JVBu3b1|w-G zvspeBYn}}f`JaaE0HQy4?lqapjLg^_P+REUI_!PuS?W=LSwItlgV<#rZUd)k5psqC z3_zl;1Qf>J6?IhgJMv>e=}5Wxe@QfkDkWta9?q;Fk9^?1W8O8|aOc<@t^$zBG_Iij4-GJLU_Y>wkSBT{ltX~rrRaJx z;T8z1TzV_#F9%?GYJsARz$(S>Z4d$(YR^%+4OC|-5M7M0@2or^R;UN1P`Nckc7Q;D ziX?5Z!{En7P;kTUKxg^r8uJTxkZCm(bz9>S&C+Hh4Cx918gWQLD4j$VaF}C8`l~xG zcORT~cc5;_EM>6M!s>`fF$~ANZd`?J9ga(ti|Pl!@x?Sj$Yspvf27x;y#!?IsC7`swoMjNiy|baB4_PAo1tM_;oYh)S58`vp^b)!-*dMC-0slTXAG~IC+q;_FkcnF ze*j4i9{@yzFTqJ}vrJGH0b{oxLJ)l;5$WFqT|zqdiG$_#EQTw~GgiO-`{L42E{KnO zQn-KpFKtAoOG>y2aPgobLJ)9&QL21VRyF&)!rKh-{aGd3b}hh3S}dd~bB{pc)IsI> zJcJ?#)v975Z13HQKoHYr;J;X1SuQ97`WZnEg6XbD{j8R6pZ%#G;@HUMRy*b(QqQ`W zuk7cw)MiaUWI2~rxk!nmsL1r&zvnffa*iQ&Q?2eGUiIkeqSzgIg=>zydRpMjEr58~ z2cR{s!Ifxu&&RSZyyujJ3WId7F~)C#<^>qk*CI|^>Vgi!5QhT*eqDfMUqmzjGJxAO zxL*?}lF)+8PSqoZq}7Ws00-4v11W+}FG8;PNdvA7T5A*ATLBqkS8WYYZsu=UwccZe z5LVAaSP+I>d7$}o7p7GmC=lUwZ*}b_aekCWkeE8LW6GBWlikurLwtoY&!jZ{3F^4t zr-hqIU-SJ|WL%*LLMekqmKh0IX5uw_tAUujG!HpvRdL655UqL98I!u3z@TT)9e?$^ zih3H5F=ewoFJrGLV6{uF;K*&Nr}~_=rnSP%05wmfF zi9=1=TfExW$naGpvoqw=PY&B!{e85SpZ^Fw>37ItYfteD>3>x8vMT+G=@;EF3qXT! z=(`JU%WJ<-%`!(#Op2V*G^cE*qkp17cOhsn_YeCkzxXZ&TCbhP-%LL{)JN+rI9cz) z0qm>$DOV5tvljX;(=zS2FFqa55a=KGcU@hDm;I_KweZn*rT=33IY@Q%UAvR@4({l$ z@Z!I_mJw>g**_6%gHHqCXB7df07|k&V(TB^i(;RJBOP2tuyzZ7^cTxxqrq<1{ol@0 z8F6ScQ;(vqb2Rop15%FDU=roWA6DbY3yk{(CHjuq!@4I#C;XyW-S&yP^zyWJ(Z=8E zq(33nVi^AW+^?h_f8a^bZgSmL6@M_j{ZV#hiL!&|7Y}QWMP}3T@^4C~4g<8Vs}RId zZRTH#Ux*}%FBgyZi~GA{2k|f7Z4cGI8MRS*`27X*Zqchm+4*QXHhp_%@{oRWV#UhaoD%&9&6pp4-|fqcV0?9IQ!e5 z7Jq&)+&q{TShbp37#b>ccWN6JwXa^DwqNdxkKc(OayA*+jG1PTcz)(A0pUdo`s2Uw zCrW5sR*F+<`;GGCm;ZU>FQKt0P7uQXdAlU)e?}ZX_jy6=(7jTWPg`>P|Mj+$XEB5o zW{1DI@KR!8i2wW0f3MxYF_Sr#ai|g;`qMnXTWB=S%z^ z%m2IrFp?|kc&1bE%h&M+t|=vD`KKQ=3jTh2q7vfA^Xv-Q^h-Xy`il;XU)AeXO@xsT zPOtqeK|F~=_raZ&_nnSclTKotegy9o0;Dm?`yPuTt-kWwE5Cb)LIrjBwVn`Ui zCD1EAyeOhQgDSj#dgp3aFG1AblQtvABXFE*Qq~5Yk8~VvfBZe((`SIjF4UN_oTvGM z@S=r=Pj0_AA&ebf;i0>^!c&JM0^~+0mqwY_!03Ndv9^zJ*vNW#HOb8Z|sEe-m3WuVz{Yn#3C@3Xwj;gpDlV^p<|5IcAylDfTf5yy*;q<*#(L4r^XIy)(}x zu*bBR+iB6HJN?wq?})qWs20RdKT0Y+8MmnO(Fczadmxd~*Il)v}MfZpkmwQ0W{4@`4y7JP*F)PX>cPeh@V&mM0w7umFrsSzk< zUm|^)r!?)LFGlWhEy7N((XjJaN||<6zqloc_p}INn`EpN>**)YYd22TxFHr^qJ;NF z?xQa*6gZ#{GMYGS;XO*dc1(+ogr~v}Pfw>;45Ec)VpXa7X)>Mh={{C9 zMX^O#JpPW8hmO9&=P4V!M>Hogie?HupnUqRwNH;^RLgKb3I2{|A8$>p`tTm!G7=#E z?y_eLz+;Qv{NrK$`POHJ6;H2FlMW(bkI8a&#OaeLVW@w66kFXH+3;vT69X}g=rdbE zytxb?%c%DP*2F}36rKUg_fNOl_^gdbUBZfE5`w#LKJVZ?rO|~3UT!no#M{!UV;U-3 zkNi&d>hv=5`Hxmu;wXoIp0Z0%bnvdcmiXxnH}pFa!*yf>OL)?aKj=gUx9WWM!F!r^ zq}VEQVggT3M#Avh34gD}6pG?eKkP_KO$4Ae@s=NVB8cNQ*e7c7b|Cpk2Uj_%;yocO zIay=LW=xSM{uhtdhzu}t#`_{++wr!(F1>nsYo6DVK%4qFyNmMl8_)Qpo=CTjH!8w- z0&kz?vRS=PN~8>7`@1&qtDGBAb0xA^m3h?D6aIuV7#|pKhlX$H+ml9@2K&Ku*qJX z!du4L5q~G}2#K8D(pvwMVg8<%dTsHZ?m#-RNX*?N=k!TDOMp9(`@=j>Tu%>_XoqCY z993XFp0%)cBBQEY<5(!rgGXKZuf@l-s1W1n!Iwm%M_X!h0fR41PH5=1 zd)?OX^fG*kkC?P1N9QH}d8#_`FosQ==12@*cItJCIz2KUbt1k|!`x)((@(kL zA?em>*{XYbGG_?_PZZ6&_BV@oA|>*OiVWiN82sg*7>oTy*0+i9`0XY;q@reu~i zcAEUPc1HtmNk8Mnf2sd5HF(+P@m-qmBjMIaMT=*G8KO>zSd`uyjJKsFArLJFOqMNw zp`!SXRa6e+q9fi!8IQzp_y-|fyhr{gGRj|Q(g~0F`8OdMb-l2(1fRjVPP{hb2mw5W zku(kQ_xhX0F}yYKXCD)Se?g}<7>|TKM~=dJpRoC9U;ZLRw8pW?hf5%C#vAB%O*`x?^nMV8iRTBglpj)06f$4Xta;?@VQ@2 z5PwTr*ua#Hxi8^A`PDDEgoH3x@?U)re|cPm&uf!o z8rkrl{OXtIwRe{PFQ;I9ox+8eC+7SwX@&Q03YQ5{ydQr>x#zVl1Oy?*jD^2mINsL? zLMn;ye*7gZqqQVPzqKH*|MIJNUy~Spy@~hZuW8A1>7qz%Ov`DR^8Ya!Ffr`^ZVLa) z#Q*L0|1$A^XXk&J_`kFBzmoVr1@OO;_&){k|Fa~v%syYeS+%Ub>@h7I7hgSg*;;dG z^nyV2_8G6Odb-Ef?gIm{lOVBOGxt!S;#TFegLPE6LfaZ8rffR?uNN9C=bo+Z))4F(QVZdWvVKP7RPLqdheFr)5rYo!m)-N=HB|rY%>QV_ zNF|^K%cm#m=?-jgGv9C2xI~P`%1(lYjX3}$l|83)FrkzQflfNIyP3qd8|O4?cGMlK zaq)Q-t(I0iw!`Z#Kbj2=>>{4Wp&LY(Y}KhBo3MlaDt|a)F~h}L3-?1@;sTOoW&R<2uj z)?;khXTHoXzfN=_Xh5z$7mz(JJhH9m2>Rw%Yc+J|Iu)BiFnZ%_11&hvdZrq5k4nxI z-y0+Ec5Smym~$${4DsWlft>Yua49801OPH)Tc_Nt`pv_-0?Ntczi#M^?DSm2?2i(O zm+f?=i*-A=q>E|rmq=8X4j&$DYCz?|jw8Fnr4=6;I6izJ|1wM&tvp(oxNi|*UJn|L zysg+@v=+CW?MlzkNRJVyF{#-dTIQd}G%8rBKpiP5v7)Vk-Z%koRnPB~iKfP_)kyK4 z!W8{4ws(8iKtC1J;4zh`3N^<-oxyj+@_Ox-Hn7mS|9Syf|K(_ai^XPp{RlKr10d(+ z(X@=9{E-d(oX_YXFdpO0U&~80F4#B9=@-C3-_Zjl(-~#>wGR4?ak|5@*@eTwX)%Ef zvc^Qv;lA53*FHY&+&+Lfd95!w*SMQn&L2A5A%az{yx`2%|M60TSBBHIQ7#kQw&ZQh zb!2uM{&QaCuoA2{Pk+Vt1~08}J*EOQe^8&5ahSaaO0+z>xu|1Tg)=JF=)e3 z+ch8D(DWGU$=r$3P7B)E{TAX?xHr>l-^*?h`k>B-m9xkErE;aqzPwq_YT;e^aX^|L77e%MXqjt4G`-R*F0Y*pI)Z)j-q3mG^x;j@s_t0@|OG2{|ff9&nm@gn4-u3!6wwl88{JKJ9;a z&D6Q$E;AF=OBIdDwpjpFtKM0(_8z#UT|ihhr=Cac8ciJO(OVCD%)rGx33kWAaXrhR zN=usiMuelt+FSCRguMx-5ZwkKg(}jdpvH{~kKZjw_u$s^ysM~6e(wywJdt>~otPy& zI7W5{d*AFS!P_Npy!rbiS1W3hl@a@~k()BZ&W%>;1PZ&3e+8$J6Er@u%>wtE*)T)L zH46XzyhxE30G%|Iv)w z7}=;Ke%?6rDv*OlXoXoya?+}gv^erMq9=^NVD#I&5!;edeS&4zcJ~W)f!iV}Qt3lN zKZ2}1k{cb?a;!bG`d&5H!TkXsx~;Wj|? zzvXTt4*Qsm?+PWQq*_~`y=Vp1!zm#A?HwoY8MtyO`CoSHN``e4k~1)~)+;FD`V{gz z&cRlmv-&8fUKC-yvcJC@<-HxH?r$u6-LEp>8bLF2D4RX;a}EYaZQtf}s{VpJAL&K< zxSGS=8tb|J?}agkUZ6`t-W7|`v8{?PO$9ORw7Z`vzihe?3o1pHh1J<@$p9K0WK{sn zyLj}QkuTe={Fz15ocpcV@ z@6~@1sH*}3Vh+@Dnfq(VNxQ;F?VYFJ+tr>WQUgsP9Xjh0jW}HDE(OP0>ZMyw)FyKn z62ctcn$iyF&^xw*O2JOT{gnFepei|&!|rvUMWK!X(4?7N$}xw=0vN4ze}T#;pF|dS zMuBK@;8E*;Nj1j=u$tfL?LWPLbs17fS%iG4cV}ik3sl?U%JxTSMz%g1mzBuJ@$2*n z5B~^etXF}raRL!7AZ0)zVwSam`%s(gm4FqaRWGm@oM05ls0OB!Q`$rUzpx9VXZ>Bc180>RXv+foj}_ZCO8h`+dP6efyV^$aWCaOoOQov+2)*t^ii zhR6GNqQs(9#{O1eChc?12VIxf=jbJ4f!lCFR%|H(W%_ia3X&Mg%OIwTWQM z+ZPNP4>kvrUDa;5^O$C!x8YC;X7i^ zZxq8*jXEl(T-oalCv}~)!eVY~JwOLBt;8>Tq!Zt5N>HLEV9%S*{QC-Y&kCT*DA}sLf94wb=V3GkMmFXrH8IswR@Tq@8*p9axSq|O6spT1Dh@@&vVh-Z=}=Jb8xvxcQY-wb=}y@KF`UM+j5_qWrcZOMz2hJU(7>y zaeA?(A?y5Dx*`H2IajKCIQh_5)@XPnE0yd`Ni|(DC)Mv9SoiQY>5_1#7pUr}&*x!A zw38mMc3SV7jTpzL!Q#|g@RO9^K&VHXBlWDEZv7N!jrlD_ru}zNAkjO@@!8?e)WRCQ zel4kvC+oo4D&Dul30_9g?axTZDQ5F>G1@T3vuIRwM8tk>sM*ET==^r3OJB0b8Pu=w zUUu!vb69|ce~uz3tV!cZOL0u(Uty*-7Yc4keAs>oY(aN7IIxUgOhE${YAP2rX1{O* zWdk({m*#I{yn}$xI)Tb3;{@e+ts|>mSkvC4oMtUkqsq+wJ#nbZqPf9ldkquz?jXhi#F|Nu0p`jUyG%%iiqARy(cV@cwb*l6Wx3#t=M{ zHL;y=KUE|dc4c*VvwBlAl5{?lY^*XsPV|v+n_d_3!zK#L#?Sn@NYhtW*wBut;95#r z+HKvt`P?xPO|&z~J*xf=a-NZHRw^viE5TT zm(AgS7Z^Zy{6-}8-5Q?+&d^3Kf5<>X^uFAv<}gp==ZHh@JpyLy`8e7Bv>#OkLq znU&UJtFJ?N=Nk{waI6fwabeF($vsBTwT`_!Ajl@i0IWEfoeJ1ls$E0wsYZQ2<)5;Vm+twmjY8O|#0)An)m zHC>|HdnVB-x;0bO`RJ}|MlLC&zd20BnDBE>3co{+1!~N_8RJ%>jY?`q_6Hs|G9?an znJl_(fj;Gjqr&&Fe#vu_4|MIB+|3$&BVPFsXWvrqQJIchZ_ycvT6b925FLm>diR8n z7%@Ck@@g2`=_%E>wQ)V1?x&xmJN!YH1vIobDpwD-ksqK=^>3x#YB@WVxN20|0WEk> z4SHl>%fL(AWUB?a-coUeBKl0y-bTxEZ5*IxGQY5Ge`mr*k>ukE(%FO-*KmaY6FIKO zh}?#UMu>#z&AIG#&ps^ax~@NU>*goWANHHBedI_>c{wXKaIqkAId!ZydC%#K^=6*O zg`UT}Jh(Ra$a zS9fT}zPdSkvxy%eTd&LQpCV7`g1+$mma{LhzZ^X2ibQBq8rtb|vPR|iYpn-HZo%WM z8s!LBF7vDhtFtALZp#+j4|Ag5dWG#P89sn#blN`l4eytNGhAe$QtM}+wE#A?O+;;$ zG6U@sVhMyaXz5d%Oh}dbT6o$~SN6FiK_d6Z?%(C$6O{6y<8RVa&fTvgp9VhVlva?D zz;g*-QM(3*xhRLo6rK`2M9I4ro>|eu3T##GENN50?1Zd|5`xN%IfU{}c>741%F-hx z!>4m^Z5bS;i8EQHOb5aF9kTL~p7p317+UICu!0AR13O6#6yolD8B&s#^R*09$kywh zmznk~NS?syHc=*SePpeBoSI?<2Q^%AwYl6m`vN0E#7(!BdK*Rh9{Uq1)Yg`T)z+6* zujeRm^{r5n@q!LksjE<(UkI={VGQKyym%y8PO}oiR`NR-(a1OPY;&6 zP@{`8_Lh$53M39@i+P|bkP0qCHL94>H;<1Sk14ITj&s=76dg#N`fQ*<@SusW{=NAp zkHe_>hIX%wLUrzQRr7kwX1<6kg1sE+Z|o2eEio*~Ue+c%+VfGXI`Da!uS!;1H#cn=5NHuqq=_o|Z&mlbcAdI}^em(1RZ6`4+Tm zbrfP$*$cxoBeqDIjIg3;=KDdrWD}swndchoX3^?!{qBE(*#E+iq`;lN%>4MmY|%Lj zZn>p0`8%2_o>6V-NL!hQiGz;uY_y9izUa@XF-rA%jRaw&zt3(dB9TGA7Z9@yrV5{OjPZtDt6NGe> zN3^Y>2fxwP6Y3zpRNQ3De~0Y&A|j+AQ2iqaXY=U0ns4UQHy2jlv=dSz5A1^N-k?JD z9c0h+K~}8KSoVMtkyo4b0qg#aFDCqw%q-=11f5Z(a&Xg5Ew6!r;ga$!%)a)nt(b5| zy-G)2=<`&xQ-0(7b`PG;^21_3xrdH~CMBf8kx4CG?u+e*1%xmoI?BNqHkFKK4iG%x zcQlhn&@0akIz$)C4P?kG*{BfQ1lNwXPg(awbu2x{bft21mrY~nr6MAP2UL}&iU%c! zVP5QB&cMBo^^~ne!Aq>13ggs4d7pEqY$801Gy=Q9zKa&#;}NY3(h{S!CN7c>{SHdo z1_&dK#XB0qUvW0-t0Ta1@O~TgYV5>b3wN& za;vA^)zQptFYunF0LGvAkVDH9HS=z?c~zCRl=&1-bh}qIdBHLi7>v!nyB)j*?yMTp z^-HUkmHn#H==6HD=ZbtnOF}uBVmSy9Vm&_N1bxC54O?92$_UR+2>)J3#3EFUv~F>< zFSs-**zBmyBeg&BemTyld@3QmExxpxQ*0h+8H|1bjdfKy?2DuPT$D0aW;;mFO0#i9 z*C6GfX&H;;Yku@0wP-#>xjv%UAaR}AY^&6PfZbjp!aPRz#$+k7`kb8zOEUMZsQp{( z0(CnX;^Cq!4A_arNP$~7h0OZKP?N$;ohJKL6+JuE3sS(v>$FSAb@_CUANy(hKqmf; zd?pLdgEOKw+4#Y_cg8fOE_%L$5IbEeTi(Y!!h>?3-ASWf*Zj7LkI3K$F%tGtCn*&C zg&dQqG0M4sk>YNU;Vhb^e)6uUJGF>h`6Rn_pTqDgYUB63`NqQ*OjJDn!Z{tUBP0?W zsEytyI=H0bO3_ZwP~3MW*GY5k>w-8?sua^@x7mx0kEl_P*yG2HL}k+x+7@^d(u&T7 ztcpmvrkoRNi1TlBLYa6rt?|x~nS3-|9tqT6dzXgG%6_IYs_gr=pw%JYsH(4LinMuR5pVd>%tI&zfNRm?{ zM{mNcGWI=Rk2g+pHNbVR+>N2ncR+k%q7wG^Bc!~k6{%1XQIlDiiw<)e`w-%3fm@4d z_b^_sB@)Z8?AlX^3_&7gLrv%Mp`+lp`KK?6@=*kecuh7y(bXFrkx~d^3yi%u_@ zbfM24-VCPf`h5>K?xZb-z5P4v2YsIRIL(M@P1o$Rb<^}WamGgF{4&p*ZJxWrQ<|sQ zsqUzn1l^ne4V)AM-))uUbU$%yYQ3Yh*g#*N?zzhb$qe)T7{K?&QPP*sqSvST^9ZSo z^cd!qh1xu@DP*#;NoEC1m!n+g%X9}KpDczM@ltVJx!cFl_Hi!J1C5smB!9}xLE9QNRE1& z-TwsUk6m$}owVYL+aCA+KBQ!hPns=~c_cr3S7F+o;2F0lEmtG_b?_e2NG9C^LwNxF zW@=KF$b~zTR@(&;4qYkrj6u%js=~C|$q4u!B^n*}mb=487lHZUFi2q>Q%c*51Cqun zbci{xm_~|HGjuVCb@pnoTod446a{Rfsgih3c3jUlt{?|}%Uam{Au-3x2iz7#G#<_7 zT2^JYKd9PW9!9hkppoBXhz8$>*+<$(ao=)r_%FahxfW6T_m#D;`byj%?v}CkrB5k< zD+!jfL+pcb2{9#iztB4;rAevu-6CkT3cr0;wSQe%XvbmUE=CjKU(1Y(#1425fT)vW zae7`)-!Pd7qM=nk$AYHsv6%ZdXHmeyHE< zYEA@WZ|2*qbX*cy-)Amdk4;O^px^d~5f|vGBr=roH4GkD z4kj>#Oz&60tvyE$IArKeiCyT1+g)oQJE$O#k@B&>JRQfPM90_&xWTgxrz^@Il3;W? zKH3U_VrCXfg_rH8pHbJTApB-lC?2GC@@m~iJeT8k76Dln>qt3i$t5#3hIloqrF-#> zkdc~&t==e}Otn0K?lZ9tR}`hyEf?l?_KEDUVpI71t|LZGpVOP^JPgD{+i6Hk(egC) zeH=}2&{XiRy8EQ$l2H=x20z4hSV^>fMWK+258nwIyOb=L95A|Mps&0^KmQv!Xc{H! zBTaSvZbPK-N;qxipzMIYSV;IM7SxDphinx6J2beAwIRrIDVJ6(y_GcQ5b@cs8-7D4 z(BXhO^v3a zhThzU03InL(Me&4-+ZL!zWsf#lLVbq?vb{tbx+dSXnTWCEqC|x`PX^Uqsoo^<)GgH zAC*v(*yY34mAGw>^ogA;+Cfo~2?!;+@4#y@>%n68r*Iy$yhd+5;l^~nSEnL=Oi*&&NM^i%u5+)| z(RwbQkRu*6*131R@k5%zsH~Ek3Y5C@`eYusQw7hcwkCopkT}Exfl3Hl1b1oqv1#1n znqyjY3L+(c<2|q3^=$sFS^Oz!T5C)JKuQjglppO5cSmYs+`iln;MH>PXBi#n6X#Y% zy^?FNUFwoQXNKWnjTbAo&qF(HB#qH0a4YRv!zV%If6LTT6Yt|lZp&vVjE|MzxH7?W zyJWs!JCcmsws7De(u8bQ7Rm1^VCcae-weh~rv6j(=E%oREFC@LrgC zfVQWeD#R|pg8EtE9%whG@8lR)GfogvasbKLYuUS z++85)g0~5)BA@s>)Dh! zwys`ldUfabb0Gzy>!w%H-GpI{+)`C3ia#_M`xv4_!sddhKKoAbC64~G35r?455+w+ zk=x>=z3kA%7@dYDJ^Q*g0v(2^1`)cpHQv|~Ip2VUmA{yH!~OlD`*9Hsz%Aa7C3}n9 zXym$_DnFnJ2bAPJV3fICcWE4fd2OwhuQn@>8hINnOc5gZ{qHo@ZO}cNmX|x-UYJ($ zeoH=G;{;lTI%Z&(i(`fPvJV5L`a-yww=T-rt&|6r`az(9TduL%8^kHi{riAJz%i{r zAwf{K3UUqHuax1VpN*bV7r#Q$XwcIZ5+}(?{{d%Hsaa4rslBNV5*Y@U3;GAQ+njU5 zhAeeaCgqJDG2#sM9;xR9jfyeVa*+uwaTsj#BmWF@6=}wB9r&E|`9TYGZE~08>|F$k z{7%o31Wgnl3)Lb>|0f#vX&soG&^Fl-zxu+@jh4ow1S>rHlOVhxeysK#w}-5IOgAju zie!za-Ik^}QyskW2rGty}Zj^Sj@RzSK{d$sl6~UtAV2Lm5I_aP9rW&i7d=CRo^>6R3ReLO ztdS)At)vRL8>19LOu&k__&YnO%9}Yb7CfG>DZ}Df>xfWg)=?0mqcxv-J8R8g6aXr` zR2x!0h%b)}zR(rBF&O#b$CkWwsY2kV;-1o2E zuA*mOG`_b7&AR#PC5>l3bqN#qw&Xv^b03<&pyk^iNC%qDDC9r&zuIiV!7wlTFkN=i zxn@_7k{524z-HjjOtbD5!*3~}FvTf#f6`W%@*D5`7w3o7!YUo#)CdY=i75}{dfsv) zkumI!lGRyZ3LHN1v&^U{LVAX0vY3SCuNIA_Vf3HIZm0A&m(WcMgUbm16L5>mewC30 zGq0q*2j}Pt_cEjd{cGw8pq{tTJ(vRF_Ceen(VGAvBO6|PpNDb0-_puHZcLr;Y$EXj zmHIMcoHT*Om79Vhdu{yp;Ld-Djii!&nPxg0S{R9}#bB$J_aG1 z{qM<*y83&xtZNdywo_HWOcWyt5GCcLt+BUT!yGL^*KiAhkpWyAkc4_b5-J)GhrKW@ zrqS=;6>J{YO|AmzZ`13IHs6UW5VPMm=PK7dx8oD{siJ9NbC;ma}c!jc!X`8rMj_^zzc3!bB$6&ij<3e*a z0J{tuAHXS0SKQ(=OyaNsOkGkrd^l{p{5^fQ87VrrrjlpF#7k_y(wsz6orhzR>TYU| zW5w1~v0HRvj$^?7(Ic{{I{2Lb*A2PW@o+^rYQ=5Hr=)m^%`5>cNue_@;a9ikSgXvl zFG=q9h$m<15%}G?glq~W{{CF}h54KSLGzlTbvU+M5vT20diZj-UTN-g2Y4gdc3rb$ zK)<2e(0#G!YGv3*A*)G)VLwNTx`%z9yF%#_L;vI#lQE)(>p^^Jvj;xVs%D<;=M`%- z7aaZ%4(n_{E=IS`rSIoDpHx$~2A?tJ^6c_+lgTw6?n+GUs`K`nl%o}1907q#HzU6J zol?C3Y`igLsk(`|S#UaR$HLrGde3ke)$mIsjSIEyF$zSeK8Uj_PK&58=7x z9gh8LS-HYL;;<@8d8zXDue7W&WUzHH_Fdc5dwM8#%zZ5P+_&;AH;m2D+p1z%iOou1 zOi3{HrciFc1M{8;xNTV%)kG{-B|u(h&carw8Qx%NU+hQ=#TT9I_ce(A4t`TbRz!ER zON+ZCCf$=(-Q?-u)H?94L9a)n{C&8LXWV)z6MDYyO&0v#)MzS**m5`yY0nSVNnRF> zPOGNVM!r?*F1$3UpBWS};Sa##WR#{Z~D>g8FoKV$(t_q3a;@S8q77p?zlS+?^IXICiUel z=+I7xt!h?m>9bB_s%On6__;p}G2YMBGhX=DzA_0Gcdq^2X(KT0icHT{Bbg+aqUqP# z$0i3iMRGfn$XmA?(&@IpNXNQfIrC`btx)tKghb-UF&WZdcHS1?7)@hCKWug&F&Z8_ z6NjRa1Aa(%y*=s1Bmkw|iyNA=9_6f4k6Mcp0WdB7$IY)4dYSmr-;iXV<@~wXSMO$A2^HV2#`;hEpowU~q6- zWoeqd0`NB+`HOSbSFW}03dE1EbCIgjxp+D|7wURy{OEBLZ2ZuiYSmr$OkWg;&Ysxjc?k9We=Oy zW#G-;Ysqw$=hS`wsvr?*x&ZEas3?S8u#a0*;Cb8Tg=I@LjJ7ozu1(H%PJrjOT`A@} zt0mOtW;$q#>654XKh_-O*|{4`-J4nLnBlxE6ReJl5H@kXVO6*Urfq3+M=AHbbi^J% zn`9!f6B{;Q*$s<%*ea&6A9hU<_zpuHHoZ>*f(E0;{z36D?;4os(RjH|;I8}~JK5x@ zb)!aOI%heZSfuG}>uyz1i;uLB*t+J`tEU0_wMY>QtsvILak@5_wG<{%RK8L& zys&wvpZ4#?+T==oZY>6P6>5xtwk(rs0jbBrGr9p&jFmT*Ikf$eXjk`W)nQ3#m1P=V z@i}j?oo^w*y8LHlGnQ9{XxqZV&xd_r%S+{%gSd*_}4LfH4AdXrEob-ud!cuuB-J-1;NS z)*9u$Gflrkm<`bQFioH;Lw>lQEHl0+#P7U6Oy2!fGt4CgMhpKCw}&d=M@!5(^i1u4 z&_)@^bxVa?^6a`1!n&n)`4i^}r@EvJtqNPYZ^!S)FztrYzd^f2w@Joxv78aZ#%IK> zS_7-CAP2afY~D~Zq0ei>yTegibYjrZ02lv_quH)iR4HB<#3(j;VIkdx15%zsC0gYF z40-T3FVA){H4^gXqPM|cTSnR{`}D1bcT)Ft;4w>|Bf{=8?{Y5|Ym1UjC$Zngnw6sr zi&f?_2ctU=mSUzi&2Fsk7Xh8?!THI!?_5aNv#ct=UzC1UsE>Y&c9tyn6AgN`UD`kN zlHr_ejCc@w903wk9p3T0WY;Fk$aRNdc2rkq2 z(AOG7J**=>UGVPwrb_)sVf#1%x<7s(hEwX|W#p#l}U_b3WU3~Qnjt_R*)|5QHQ{FR1 z1^^qe)1*771ksq?uX~tE)>3|tq3=(`^9OVDl!}6HD~rBcFIBW}j;&D8b;8EYKzvW0 zVU~UX_CT~kv@6O!4%`{iGKvc$%S7G3FgW$#D!uctjkch+9iaw)e zfyWeX6JWl^R@nTefhVvy($NSL&~-*imG`hFr4pq&1O9g!KbHv#)L~aFS=&sdIwt_3nXN4DqZ73H)yS9WPdBxJl+3WKoCOdED zQuOK*faGrTh2>93+9nhCSnvAX-3caf*Zc$zMIQ81IOCPI*P~$_G4X~UJ+Ax9i3uzk zek7zb(rlDh3RT?YDeSkgsGsL(k78J1PTU_vCF~r!kL>D?qbuO`lo4)Ex*kb)UYMM4 z8Q18EAS*|{Yuc1(lWntZb0_fQRwB63zrGK6H=(Y8_;&=CHa9V{Z$WDAD|gDIBk&n= zz%_+Vx2+Dm=2R1@*}NmNbdAh?|7sH)e>GM7rkU_md{xrPiP$5b2ozfQ*L#3$N7=fLKc!MO`(burY+EuYS?2O3PHl121( zKD|$f_ujZ)#r?!@wvsQ6oU6A!app#gCvM9cB^-&l-AIeL?ZvnS>GE8K$KUKOM__sO zCfFDRE)RXbUV8JE5`pRnz#8!RBxj@pTpk^7jd1v!_%@Hf~rEL_knc zK&4bl8tDdu8cJY>4iy1ur5iyJMWwru9%5*uL2g7khfe7sC8fW8e4abKALU)&THkuV zwfx650&~tj`|PWK*YEm3R(V$k;K@b|D>)OE7P2rzv#SM2AxmWGxhE?wUvE-gQ4u-# z;)fCiRE1}B1xL4A)zCFk&t@F&8W9;EWQD7XrIVk_5Nwj<+6I%tkaobRk84EE=pC$v zWjGH_P_;}5-c3bv-N*_@Su|&2%9OsrhwIiigD85zq?mp=$A4K523E5m;*>|=%5K3TapCDkw|Z2(<*JW z@j{d}{0vU<11HBk2!(EI!` z3?Y&XrQ}2JWq{2xbG^GGhC*-fCbNMKjmpGpCntr@0(~Z&I}bS}+6` zAH+1P;XXjvO1b1`m!3%T(aXJ*owxIvJ(UOzT)` zV~a~#%#mT0>&a*Zp~Zp+m8D#zNFgWHa=sT^Ty68!o=74h^<1;qBwZKK1EGTM=Rb)e zv~&gx^GHmy48pfK82Exjh8dXg&B!`QbPa&JK0uu5h`2JFhnZeZS}r8va=H~y}1eo5f2x*_iw0xcRYLU^;O#W^n7Qy6vdM}T@_Q1_w<7dozsXujtGw1tSY_?14sUkzT7?a0WynXK z$cwb|6l@y#B;UsyOF-8&1IPt%`nP?U&lcZ%>FLn< zK1yL>hiH3KmYE7g>JS`pJ}pFH7h7^FI+^zI_2Vb-+;#v)miJK(6?!gayX#ZTn*TnE z%lvffZLAhdCOg6x4Dr}%%f$Csq|X84Fx2pdq6$yrQWr-`g zn=cwPP6*x>2PIQiIb3JPbb|Bzkp}fR4u&H(ebi2~3G(&og09NM%^={HbCA!OTX~}3 z{03~Gfz~#WoJcIuZ_4zps1V0Jy7px8nDGp}&rO2oS;E^63+Q^!mFQ;Vdm_@M=Eu+B zZj3sCU3bmkvVq%5zMB^0+zkkvM`@SQkGW`{b-{8}_I4dRLH#KES)MlL&4Q&T4(hQN zGaRatMLD5b&ES-_0F7=S(IEIacx+Gk#RgNS}eQn5Sx_2e?ATS(u{S)mSuJVeF!!tIKJM` zWs(Mw?XJdbny%MqIhQFm*dWi3mnY+|$OtBX8Et%$61AFa;WmgxAEYd$=hN|cJS#Dz zstj#gZrKG5L-OK>??+qvKMI5JD}C7rPne0+bX%9$0lvQX9gBgPHVD0Bt(E|~I~oI! zvQ8n*7w|sGtaNT(FJWHo7Cgm7**a)S-QzTbmm|#^_V9&9mXh-juN74%zgq%OHHmibl?!o1ID`6!CdV6pe1a1ZgsNsMnq&1GiPTlKLXS-&AGAHzx2 zv02(y-kNp|-x`G=ZJ29hC7*`o3$U=FWUfR+TI~laEIc%eETZ*i@Ziu8wQbGpRZQn2 zE{Ju+U^1y5L??~F8dYP4;u^~sAw}1!$beov zvpXJ+m@unyyp*;LT}+3Jhq%Fk2H0wuYWKVrI%%HX;UG>XKdmiIcT}s^sNZ1Rldj&K z3T_M6l<#;TPO%h&w*M(nu}xpd;Uv8dssxWb%I>K^xe-7D!nq!DEK%wDoV21{>O)i6 zlSvPY;eVv|{_(-14j=R&WEN*ZBo@`gK>KKZ4$|*Eb+!+V)o>~}VKDF>`bMzF8weKs)~0BI`8bvzjcWKl{UPJGCw)KBnZ}noRQ@v(R4_@CURqhXbsHAn{*lN58X%;OhJ50O&Z*dQSRC zD`AlX`}!fx{N?X&@Xzi2{wLz@;2}3=v;~R(*Zcq5w|{N-cgO12cK_{!d{aKZPWL~h z>6bG3Ud{eeCcl))_ZI1&Kl!hzbi_6?{+$cp{|d8N1U0a0?OO96&6@zE3#ZHk03!qN zDQ*q$9?6ogOW$j{QBMIVx}_Mp6aLMQ_`{+2{e$}R>QGSk&>iczmSq0ypL_ZHU;Xvm ziThy}MT1#uSbk16wq*#=1u_LWEHI*b0|zWwj}#YBmTFARbRs{+S{8E1NF31#)E%d^8dU|JBx9CKzA~D+g(R|+reNZ5;B9| zapC>zTcZL^{+Nyio9q8_-|x>ODIJJilsb#Pr&#=#T?DTeIzOC8JwIS2n!yE&zqb32 zegC!H-yJLUU#I(@(cryb%H*HQ%R}OqGWn-*{J%(`Df@i;NrEq}r z0_qf@MjLLj3a@{(c>MbX&8R&dia?xWU1q{~FME`GE;t0kWy`=6bhV7cbPtU7th@xy?Ny*&*_U-69-FNAoTHicaj3@WRg0e zbf5iAalD7>FJdl@wQU_#m{)B9qNM5J*ewCX?a|8lk~r%!YQ%ol=y&<5H+RMWrr8|F zPm);lvQ=H8piOXZiSXnQKmvU+hI>u49EIp?t2z>on_S_JGv_|$|m!nH(HbT+~6%$YeK9%t+S&&1{5 zkHlr#6)-~+UX!?ql&YN@$?=pzRqmL`)Fld7PY0)~q!P^tZhn`<|K~$}(>?JwvA_gC zrq?Wa(%|c0xP8lDdJkr-qlvvrTUGl>-N_rXNAjYNW7VGf3egYc&AxlxS~aGWq9I#H zQ_}%hm%|ON=+hMxDjr;mi#$jOevr0SjBg5^WPqUgZOw&VnCX}i+?|`fA}Qd zK2&3qofVE8JM(n#JO1`x*Z#F9e;lD-=j4yA_~XO>i#?fO;jmvCS_Q*oL;CZL!hn{7 zAL_boPpx7yP@xi}8c7?LV1lth$y6w_ENfvm|5o4kWF>!hmG0d8=!cW?7cp-*S8lxqhafXKAlEbj#EU0yD@hcSWz~ zjM&~|$GPM%$>DqwHSOHYq5%_w+&WV7jgsiU>*BvF7-+ZXv!9N2{6^_)ukciy>{L0J ztvm%rOtS)L|4{g*U1ySv0hr2ZfJUfMcC86s`pLrND1dr<#nrPEe9VLGe7K0S9bg@F zZX7DH-d%S|@FHcBWv{T&0DIXKD9ffns$lJ(9~vVS^nV01ckUi~?;wG((|r z!xqz>*Q05QsJ9)ef&IkbU*z(LvM5a-vT9zN1c7R;!UVly<-9pWh^PkCwo zn)tmB_`(NYC#DW|mzS*%Egfpf8^P%kMLummuPCfU<+4)KS%%d~EX$(A+js97XEcXO z9FgYF(mj+oP?gaN^w!u8siIsU{!^$&YR6oy9G}Xzj$;lGMV^-!15j*6z#J>(2a}3W zHUIeLlEGAn8rAt1&wZCbrFs<*V{Gv2rI>3O#n?J!uKmnXNisa1c*_^M)+%Hv%>eOD z!g>2223=hod>(`C7s-LY136W_IJ!=A$xP#dMy=7IQcP#*P8T$DHWl;~vX`$mGShYC z(cKAl7vM#|=5*Ecyt7Za1&JEydPF5*DLZS~q7fNL*y23P_oe7R?ck%xYN$|~$rb~c z-f9O%Td_YXj{&f0HvZkyQ_w7Ww|@IkSMNMin$2^QWRNe)a&2 zm1YI=y|w*pH==;ttBog&bfusN*GfBK%pp8I>WLEY&#rCW{U3{H(O$UPMdkQ#YowWfW3NZcGAjj^ z!5+sXm2TIa4vqAt+f9AYad)sgU8M;kjPfmm*!~b-(_AOSLqZ3(^8Z&|8q!Ypgt4*x zKHeVYk-Fw@)&Q!}BNJO=91 zh%~)mWPROZiA4I8A;jEkaSuo|8{eplbp)pYY-^<5Woa99T_XTXUN)7QD*%J{ba32H zbYO6S7kiOvre&65kOp2Ps@EY0a|3rC0B)DoyqRkfvHb-LL32)Vb0kkjAtU1eT&DAT zMf&GYxRu0L9(_#=l&b~PzSR%*+74a5;#Rr+R!63%%{kZz02)H?CKv{8Zp+(GH%Cl? z)i5(o-{{`-IO7s;R@85Jd)Reo>%8#gy0gt~V;y)5(RU$$9kLig$T@;uV=q+qxGE z1*B99M~c}&2QnALC~cVqo4wxVo+4#xiFci@d40W0u~_vCx0mJtE{#X-CKHU@viymL zO^1Y5z1Ba&-9H~1%+R<1j@bCyJGZ3ph?42}xrbJie6wEXPO6V59_*jzEZrP1ZK4a) z?o5oob11cE+%#Z{(t#igfukN}$#_(6A0%P>NqX?T@W!Wld)3feg^%1+!KB*{Ld}0zB7)JO6k_9*6GnyBRO8tN^u(P)S9X86pehdHa*x24{HC0jqU7 z#d}{MgE?R9s%r^c4l2fhx4Z*sEJb|}TsND1_P#iedEjaeT8-HDXryk8{3x>XNI7+i ziqM=^(91VPEUFtlH?= zONr9RQifk!?5#SmMTmcf4D5(voM?sF`|%f3pzlmg4IzjWJlt`|_A_~PuH2W`As9)A zcMOn|nzv{8(eDMMM{we(QuQ1l(@qa~^xA(4lS$D*ik*(}l|e_j@GIXpc+1msvBUep z=0ZQhUBws@!daULs`+VymFslv;Jc?)_UTCUKPHMmMl(G%7RfY78_pRRnk(~&F1sRi z>hi2hqT}XCLCcCoz?5e;JSaZ(S-FBCAzawKcFx^3rrg&{e|P)i1A#WRkI5D%9qm@S53Q`@Qjv}Z_q5? zYBE3y$|MwxRBQ~4%wQ7%ZZ6AUhz2k{(2;RF#BKwvwuL7O4JYO`?U%22%A&{d_L6p*LbWV3h(Lxa516zqe)F9Q&;|_G zqsMdH2XwGYxfT2Jx5Rsn7_h0UEs-nHt28iTWA(~>=AZ*FN7tL#{uM}P0 zO;qSSsUw*JE!$?ckr1eT3hMRShDcXUhBzznp6dW1+At8TB--jXb+;H(+2F z1?X(tfAI);zWg9AK(MF2ZN3+KV^ylu;!+$!RU_F=2-!>PdtMlss@Y^;f0EpcV@v>K z8xG4NdD(@ggeCxy?99{MLFpg(=HJZVN!8^Gf{m74c$8CH3jqatMcWVu|JPaD^1YK$ z?}*5BY?Nf4glvYI)$k6QsqB**-w)Va<8jph=8=Y5edRjBY6Pf$H8lr_CoiOR2LLJ6 zi19JqtK$cC>~2ap0eK?#0{~JQH^$C9O9+r%H$u|{dUjc8j|GW;B;*@v*LU5GLa6B! zMuT1<*AyF*Wowut-5t-`?KiRcvqzUg1A)mu~#j-2P1%S{kz1W@NxRwEr$Rd+?kMm_{ zLO?2v(=hINC~-NaI~$Pdnpk`^ll}0;RBu<)Z`cCZIi?)6^GN+>4Dl5qz^1L|gVw@)m$3+1*oV?sBMRBo%8)#`tSDn%j0B*@{ zv$4i)WAeHXY zeIWF(h*i2PJ6ZiUzVM_jm#i|r6s=gm=&VZl)qDvZhjA~uh=38+UAhzgAIQ)D$5wiu z7lw_|(-jdT`%E&j8bL?B8iKDBSUlueNu7o%KHJmHZpnO|0TEYvk2`4m zGeumLk-mC0yp+z&zvTFQybR2a?sO1`wCDvf&`P0|;4h~Zg5h5b@f$cnyCJX^8Q%$J zD8wPC-zEBqmLd?dMq;}fh<%1#+UT6agWa$ezR~j1HtdVab!pl6f|+h^_l_90L?T>g zF(xp)^K{!=zV_vbNcdVJ^?n+9H2Sc7tu^C=GM;2N!r7{A!;woKO;)+$)kCLj%t*bL zPz0Zl5l33!EkSgy>5a{FzY$k87l()XNP!K0BIBQGaVb^sR1f0?)#WhJSkTpl9)$BW ziuAJH42D}HtI|{?7?;-!La!h;6(v#%Tym4!7PCYg;3&Tua3Q=Ocr!c7ub1E#xH8|}*)?Hf|vhCH$!srX|-cNDSJKbXeI7>`S zYMiQyD;CY^tjkUQo*DSN28e{hdPi1@Fk0Ey)yn`^5sd zG;&2ErZ{am!>#4C$~L7?h%ZLN2xAT<5P)Q8@jO$iuy_YTKHyP-Q_^TaoFdIZcShfN z7w{4zi0NTk_q|A9oz@kDkZUP#5V`?5mLzPL3dNB_h|w0B+EtP4 z3M8HdV$auOoe^6+9$2mz2l<&{;92=BoeQ47fVyj#P6wus5@s@&NEs^+EzOwq(X=R8*#HvkH5d!gQUCE7CV2+2i5Wus)^bH(yqUGLR} zLNUH)IDX0=Oe)9dpZ1P$vZ^1PEldRsln`_6%q5F6FX?VXavM976Vz|6wKPqQfYfgT z?E2#*)-F2mPEG>u4BCcK8_wpTB&UF=#pn!;&5nQ~4Q(JwT`eeZdHHk`ZCLeoB z;pNe02Es&RG_hE~U<~)dx#=iNHjs>h`whP!K9KICn^EM6hl~~WUBQrLm{+_u`&5;( zBR$DO%zPoPH1XMOFl<9>WFy;tV^(%QMdYjvfIEjNx?)r|Qm9Z7Y-}p_$%Se*jl1q& zz1^?R#JRhXX3P}|%EyK7%Ai9)NkX;jjtdfW~m(bE4~~N30eLqPb?4 zy1~yYnP@|H!zl zAUQ-8lW)4@4Dt|cZ2;*)@<=wKY?GJNcX1{vtVv%0ttm`QVWQZZ!r7}evKlHno^`O) z05{!lQ2%Z;=%4WoM-`4Bj`7FRGs6_Y#>KM_Lny?VD5vZv96OZFXhQr*pRXw4|MqY? zCZVWVF!&uUqebB3-a8gh481>&?I-AUV}Lz|u@SAK4DJ}^9AOza^Gp0rkW#X@S#1)L7?|r=H(5>rr-6O+=5d|`U6=fYxqZTp7 zm7K$B!`UFwV&YU*zD!xrSh^53X*7UCrrn+hgQqxVq$)I#*#rjc7s}Cut*Cs^kGPy= zu~?HhQj_8oH(pLO)la@qA`_AZH-wSTRx0n$LDPK-^-}QU9wDnjeRT2qlXTqoH$*-l z^vC-~ZR?#}nat5DOw&WAnYkZd%2I3+5z?)Mw<|FP5*ALKG4a$f-k@$upfVNvQ94bZ zwVm-Umujzi`)w)EH+dH^q@P(X7{ic%Wr}ZXFJ*M3v}F^x&uma4YnOGsEP8Y{=*z|H zD(%ZP1Y;+GoSep}pY(&g%RmlUO{Qb$n=z%%xAEK`Hs;Ccg!0~fPct}}ifrsQgIsIX zGXf65>0Q8y%EyF1Pa+PzoR)0TsqZJE+5>zNPV4FBo^1PJTQ!8pb!OtP{XpiMfEw

    B5F2QN8rOAiuw<=KyIl)^2Up|F7yR&+roHRc?uSijh z8yT==q0nN8swZ$%ARf!mn#OM7cC9s66x+A)%X7ygq_g3UwgEkD5q_yNbA z(|L1Yb7`wAdJYY2HT!y*>pSzaAVKkxkM$cxV=zSE(`O$%*SeY<#H&6PxigublRs!t zhPL$Hi##)CUOe2Re<&v#n%)N`0(*9zvljs$8>7d>ype6yJ@942EB6!?U3K6U*#Qt! zA-);nrW5-NT~&025ys8RPf{xi6$eA(5m;)*i#3th-m9pJPF%C!4Ng31-|0EY74&bT(?>By zAdM0RH~nGQ_oi4p^=XUrZnwNgc+FC7(Ce{|dnT^M1XLS5moS$>vH)h=hcd271U>o{ zL%?~YbP`-aVuup%$&j0@%=$2AUx&sfRWmFZncgC06zLhU0FnGEu*QW+R`Mw#rj{U2 z2qhXK2RVC(!4Yf(V`ru|*=)44naK?FQ}$qD^PyW2_$*l>_iQ?4 z7(#*+MN2f}#Ex+pHm8NI#bKj*^xXG*cE?$w9j7DE+;#$n?{NJ-Ha_7$NF3%vyjZ2^ zW$kJvOjTC$Edw2LO_>_`7!^1oh!kX+FKIh*<6>4uE-o@qQ&wHP>*u0t1$b<~S)=B{Sw z?ITMQX?*$Lo#wIe_n}WO2;%Jq#J9xX=9202jyO!@flvU^bbGr^?_f$#wF_xGz(?1Y zUAcp?0(A())ueKi7Sn0a+G=btP++`)#MrWUN~M1&0Ev3lnL!YwHbK18RK2t7^j+!| zD<239RwoJx2`l3P?QlUoPNGu0pG_uo>tY=WCEwmHG5=-0W#v|`JmEGtW-R@mbf<%(0j8^rsU zDBzKgc&dNP-QA)qFMKi1%aegC9Uiz;eLJirBlelPP=eRQr*I^{!zR(%P-X{k^zc87 zoIC*ryVwU0nMN~MLFvF*AZ@{&dLS>aYAfX^FA+OllyK6Iv-ti8E@M1wgVF-W4k%W# z+WS&pobqk}#QKelsnMMi%0NRo9c<6YBAb)w3RWRrym}s|Do0S(9F^Wd1hoh zeA$&aP;mVDc_cba+e#*a+gQ2ZZFeza3!2o`1O`!cjCjKlfy%6mddlfo=rT@t+zG+V z@Y$L~|D=KLN0dVCt@#K=kqtPBZU(#mm$n&Xg^y}QR#h){_CSW$lq0#qyv_|qcZxNQ ze~#SwTUCi$K3?UMVUDPs{zbzaXG}#!kgTDm%-8ExrhWzxB;{@;h=N0LuG+xMXpI6i z9=zyUjaq`jA4sFP5#*enaT$J(8tb?&nFT}=OiS3Eg2wJF%`P7-EjNxHERGWGF1@%c z-1rL0imXDFI8dDlhN?yCS6@-YvB&Q&SBbs;R^Q7&RXP<8?TSV07=wZi8=ep=0HDyx z)PhXe`@%tWY(J|!)}p?5v;h&jGIh!8=~cYIR(#7x3_yWkG_xKXgmq|eV9EP*4NOnk zhM~8>Da{7>34B+DIOBdydza%(jM@m`!vm$79FT)QbayK25G0xw?iu95xxY(2B}{D! za)|!yK}an;4m4{*LIS;M9Egwuk$Q{f!0Dx2;cBl$8T`nYu4#cSO{o%pVpZJD(jYQN@;Tf61g>k%)Sg+ngA8+d)|w|jLYYMF)_n!vUX%FzFo z2e$NuW4Wi4-abaHNpBymw=w6KvM~VGpAj#6iwnwmL7B?k1vQkzY=ZaBbJrK=g$1Fk zf&`fkJcjLR>T}Z_(}+M8JwcUJ1@lEKFmx{jFwnW8cj4qapz1I%(BnLgI^01uz2zHG z&#Ih`G==R72Lo3GD!*uhYge41?%dV)6-bE&Q4qSaq`1vxnrok`7^v4UsQx$L2F|n^ ziJjo$UVxuVQc@M(lZTB+_zf5xvXm3CtZ%iyPQ#jz_}uTL2~;xGFfT+qEpW0wliC+? zekQ_~zOoL@_`w(*979ixBu=2JN&sjHhSW@^GqZ!H4K~FbtbvWn!A~$43{lh7_-2^R zzZqsl0=%B7gF&qnt)>Fg+r~76R4ox)srZx?z_MmFvE(;%8eBWLX?9%wPZ#q~Rr+7E zJc|!$)Dg<%O%A!`-Aj$T3TzSPGfz(t&3-Av+-*fW$APj-lFPi-sOH!$sl}~`2iLdF z#kL}E(WH|Wj=vyRt<&Iimd_(4({;EnW0+(QmGBxtwuEl=PLE4Sdd0Zv zk*Hky=Gy1i?4fm}TA4dmGG1{btX+YN#iMQro&*%&d@=}<=S{Z*CP1wbBGWK&6^-7u za$0sC9o$`61->p~{=7B}wgOUZ%lV2a$uM*%{1$8@8Jpq7U}qa}6pE0U!R*89Mr$p6 z%hr+DMwNS0>}vBypv3cS0O8iCQGKu49vFC zIhD8_IM2oawSBs7p5DhDPXe|yPnh#!p_=KMI&&v2I~)Q75}he1wuM6Gfbu2Z~u`lNpexxS%$8R zsF$tVSv++r%9+Vq%UMD6@<>B%*iGN&_XY8MP%yW$cbaN{HLNNVOKQ%7DV_rK zgVWuQk{c}q`QXB(Yh8hKkGV~IGEH=tQfS`?uIn%@?SWZ|*~r@NuDi(>`V`$(gH#QY zYOrXt2=<>`z4;DaBO@#Q0H}xb>_doCDN>U4YQl4UVr3ZAMXe?ZJSDZJZN(y2Fsdu7 z(cfUHmLhJfrNFqW5i*4c0vKJAAY-_`weJY)s~R~(NDN1wC11PEBfQ%N&L1?4x39HbJiR^@ zxIPKF%uFqzjbbHn&uY;DpE^yB0W50)f=SoPWlBl%cFBa%HmzUv6_x8S~R5HX<>};Z#LvF)f^yX9xw20 z@cZR{144CwZZdEbsj{86)ht-yx418w|0T)6yIn~!qM0SAKaEpX5&vO9@!rR%2W(O1 zMH=&4dz*^{jgab-v~oM?xBAoq6c)3NDGV>dOVgRJqUVXl`N3ePCUE|BQJAlh#to}= zIiS#0TyzFV%i`C1uXX0}L}=J*bBs{&)^yMA^FSiO*?3X3dtPmG1XsqvX6#|E?Fuy+ z2-q^L!1y0k#9JrJdAP>)U2PC@pebHgY*EkdgSx)T!FJqX2`zS~p6t_E0H|BAZ_L@8 zfFc}a+zL(!UCM5g(bO#~IpQ1)CH`E+C`4~>vedkuYN1ale7P~7ug{u`swUCl&Bxr$ z=uuQI$ioR`jA^?rJ|6-F!7M5jlAb}y)kq^DL?Ne?v2#=R8^-U+1S+@7Yg7NF+{m}?{w{W=!tFXzsBKV6O>%#j%Q z0Ha%dtRAo7RE@YdRMKrt@ak#XgYqoWCzXo>fE72ZJWV{?xL@GS#hI*)WQK?f=@4K} z9b47jstK}WA?mJ5N*Fmz7LA1Kfg*8+CP^k{LixB6$Z@k>I(%L_FbK4Y)lyAtMNMqf z7ksMZWg6S&+zIq*U9`!aib1HBPUolc)-;Lrei0Z9l%tr#HeLuN1V-EXq){{cKQ!L5 zSFT1C0L5tMrjy5<-Nooq0O48lFA+|iWCIEd3OCpL5brZ^rRL>*g40BXZGS!->U@<^70 zW~<}_KJ)4AJ_Vs^H)4*aD!IeZ-Aj)m3lwH9zP8;~Zgi#l8vI1@nw48|fUxsgncp)V z6rS(1jogs~$Rflgj<`Q2u59``TGy579IiJ;~wM7wY0in8io;p_eP*##_I(!1uP`@8wQM+*9DWF zige!xQ~Urn37Mql`@^X5tomad@a?mP?-LR=X~z@c3rb2z(o0`?B(AcR&c3s%CFEOA zf80G9uszJ1uagpH}?dE&Rzd?jNMufC&yCaX+{x52~%mc%BpnJI)`S z6_8whg!ix~4H|lisCn7bc$`#HIl$y;SIe3_06<0e;ee2Y!^y)%W9E#qd&uh-)?N@^ z4ED(oBl`A7n+vq-bXGMeylR3gj}p#Zo6pQw=m-9TcY)=3J7Qsz57%c;fIk9-YzVsZ z4Pf=sh=lk+Vd z*+YRtWkSaBq~~ui?kXvA=~z+_XpaMR&;CanIl*wnf!l2}TX++T%czwF__qE23gyFg zgCUe4<0QSyd!L~uk^YUR=-(~PzZoG`Q0*c0^1=}x0}ln_F)J8C`a%346kZ7cm|4^M zTo>*@meL@$-(?;O!JqehouYC)Ny_b!02jd&nMhp%uw|S@8C$U*)fJ#PG`#2m9#mUB=?`) z01`!l=hdJFt?srKH5x2e6&h-chQfJHP=m`Z7pqCkrr?eF5%S^>hnS-h%TsD%ei?I= zn_`zLi?5_Hd&{RD3A_CU1ptp-;%dyYD(6Ts%_k~f}%AiItL+5|zdnxX})ea1TO zb}#(g%t9tJ0&9X)SQFriq8Q(){h{Q4_~A4YhlNTO=(-Z*wfjRG|L8-zI?&V~iQfAD z$RqmW0*?S%U&*>&vhDkW`!7G?Rt1JJ_jaQG510JAl8gk#t9vN7gX=pz@o%rWaA@D*Ap z$whtyfByX_U-`A&-@UJ2+x`8?`gOYhP&vPU_@zv~(?-9P$uDK{KWx}v?c6VA^1BZC z-PJhtOPTyqCO>yGe|aXqJd^KK#~(W6j}QOP*sH1>CG-H#SD;KQ1W>iuYQ)~ zLhw6J?wd}_S9rK09&*YoIiF;(ZH)+wkm1~poq1w3U9jaU#z4E}IBkD;NLxG@!+zCl zdwV-DEQ~=HrDkPim81|OcvZWO@zj;4$FPp{&)su40O5V+Muc2oWF%8VLqlM2@P!`k zpi@*{Q*{kTx&(A9+ULh=5}!M~VZl0%eOCO~5&rS`ihowv2gR3o^{7krKZ}L!NBsFl z-l-$r;yCtaX7QEll_PIn9C0_Eir+eQRv7lO{?oOiJq38smniV)U!AxvA8o_I)8btP ztGY@>^ZG~|1?}aPO7Q5P=UX2AdLchQTfbh&F9|t<{`;k}eyOaZ7?S@FZ3r1TD+7a+ z9KA}es}ps(E{Z=!;YL);gF;O@Mn;+CZqI#S>_6L;)lGbPSv&6E-QDe}Uw-rF3h_T? zCY}Y@kj8gpHHDfNrEiax*`p`~ZKT=Wk73Ek$aGYwNJ_o{NODO|E-r%Bj#kT|_qowl zjf}@nNDo{j=d)OU8xR$x^*CnOBCN=K5a#6MR3kca1S$K^pPYRKY@~GewfB-4s!3jN zHx~yPG?Qr#dlIG243BjL&`3R;Y7Bwg7&Hq z<&OEw%B){iDx7RPZb=LnUJ%wSGK+ZRkpKt*q%%~rTxLuU6TQMi+4YtPIBb`OO4MIU zm!a(BOvmj2-pi-Okomm@pvlWKYNJZm$WO7sbf@1F`#)HW8d%Kx2j`-zZ0A+=Lf)Mu zyez%Hvmue8THx^I!D!67r=+{0(S%u1jCDy6SZi@qj%>3(s z;h>@B8E!J5B&t+3!@=z)&+6>-#38t2T4x71WnT6;y(K(?TlVcwJfyKtQAxZbo#xTZ z;Lhp{aJMcSzWGc_)8JaxEu}2YLYJ(Lz1|(Y{gAf>K9Zp8Hru_JMf-hwukRw5FS%N1 z#LT}kL+(yO$C{bs72kK{ck=QM*alh?gL3-du?}#`9s(L{Tvq;*Azg6 zK>n#-8Sf%?R$xcDZfrK-t5YpjSfX-3|pkkWqJ3qtaT zz1cc|_T{`W8X4lwhdT0YpVNWcvSiOjQ@~a}=|!YDu1$BolX$IsVrF`}ac_Igb$1{A zb`RlS5CfpV3AA){A*B{zAM*6;iH2dYWVU;_NA@?5gSUwz{%Rw;@uKiRrK@wM#(S}& ze7O!gR`;$vFQo!w5o*@EtnEp|r411Y`!7@;RwR#lGBrwksh4Z;sOq$|-;ynkDQztc z)3LJ3e@J)ePE(>W2Jhul27|)|P7;P|KX>u(+`Nb4lAt=sjvSB4wwMVOO?`MNw)Q6W zXGT!}vax5QIiw|d*T5%O!d^SXi^ltzu!MLJLsd z&>HvW88FhctkhGxuP-ilBzh+$CB3Yl2*4pAy)dY@`hk#~_fxh`na#4+s*5Pi&zA?+ z^&ly;HfZqGE=kWQ>16%f;?>V-u2D2zzFWZ6bb?Idbldn9tW&=!JoS;v%|};QU$03t zA1rDJWz|Z$`RrsTIL*piRwdf2Yi(k(CMHPj$2B-O^)oF|GXxH{ivx&ERnr=`!zgry zT9d~K+)0SIEJZ$Q$9~P{=aGy5J*FIjM2}nVDqjy+6kSGl#Cv7Hmc0*pjk*?~8X-OCv_5;4k?aE5L=LbVc;MuCYE(GOzcg zTtf&owzPz>>68M4y|UPQz(7NDJCsG^bD0tNZL-q_rcxi+^h$i#?$&ZxsqH*$7l4(s z#|eMB=BMJHPf=LUG;N%Z4h%q)%4C;O8fvwSXzUf!}Ua``G-w6rHn^*+%Xv16Y`!8!?RjcGdt zw>$1BHMN3$n~=&_rK=U~d8giwddur=fvcu}8sHD@%+%mo)UP%N8<%H2qo8+JM(8HN zRYZnr3mH65GwV3&KEn| zuKDb4M1iP?Ixkb3AGYlcoOd>e(l_)vjy zbfqJgCE&Qe-)oao7dAJivgRJ}ge>pX#&VNg@jnJ-a>lOaVgK9#_dqh|%&5n{iJlEy z7&=8P3eu93l0Q4L+B6sz@!{>eC!y5Y78Q*|XyRb$06uk7AF5 z^Q%Ww3>!=LM>yNw*=yws_cbX`@7{5GW?C;~aQLPCC0i9LOn#?G{_U(L+pS|BpKyS8 zWKkb-q-$qX@~Dxr`a!qVn;UNdJVWr?(;C|_XK zHPP{dzFx_cx4gi)(LC8TxZ@JO)^EMmSFujrp$CS5lP~+$&tI~5yf)4Y#m5c4G$k*r z`Ah74%7uPg5;AJ!^eeC6D8NX6G1IYYj$z-gH@ujgDE|zuQKncGqg`v#!1?8*_qYf@ z5QGjga9&+imX073PsQbMMl@#w(wpAgzQfpX%~Ef|pR^DiC04BF_9;|R&2U!?Q%O(v zN_d+r%0@^4u@(J@GQ4z%#*&pow@ItFZ>q;n`^9ZHdnumU35q+%JZ2$Ltz@*nbA$)I zVu)qBIn|)k6&*PYTc%p)s*Zrjowpr+L?YU6W0yyIFMPM}?8`!+lhp~V-G43*$Fbkh`iBIu*D{n$RzKN4=F1Dua7tgDz8Cc< z)Iln1@^tKNrE#(SrwYzjZ+!GDXK2WN><(D zW^qxGXjs6L+lxx=*6W7{$~@Uq!BR)(;=7&N5cX%IU6$;j??U%CW=#SqMpkX>QE9lV-TD|l`rF$5 zHF3QQRmrb4p54EFMG{zv2iQ&KM_Az5moKUp3U#;JqMPH3oiJ9ATHu*=^iWW@zoNFZ z&EJ)?7;#v-lo~#XhwfxaKkXQ*BegcUb0+3M>!b@|H|Nr7F&AxDBK7-AeBD=DOz3dS z_1-OL=L)GU#U$u$r_T!36FrfBJ@9H#$)=**uL_97*C?)QKbHt`&r=kQ&0#a}!HlTV zsARL*`xLP&!9M;;WgjRt4*va5c9&U<~j=eduDU%GcBLXYC5h7;(O8 zW`LLI_1IbAruZqUF90QZgjHf=y)Je}%zk3Mgp|<8*JR|JzU&Ow^e2gFtcZJWP{%g$ zRKeUi?+vvfnbs?Rg^u0lI|ubGx{fZYDU)80YjW6(IX_1ioNZ-s^HDJ{JX!9nU0zbR zMYeBaOr9@n^JMN`*V?E-gPN(Zgd*$a!N+XUk|h@FT-R5d_gDB+_r_08;k_Dis+2fs zFK6a=$GI}ZTm;*XY4E&$$L;3&_?^(9qIxkj?5@-<+U5`JHws`?ng5^iwbd>9Q1y;f3AeLR zUJlXi>Yk!!ZN1Alyr&OLwQF7QSk4!z5sCEm4_4k~kcPEEw@ zbY_I+9&tm)QZh^O`EkIO&fd5Ou(LmYft)o8d(yJWJshhqaBR0 zba-~0vTJe^@NThOs``y)!oBm%*9$Shy3%R00urNJ)KORKhj!*Wky3BTP1v1eYIfrg zA?a1_p&BbClTo3zDpN70W_0!p%`>m6oF<1Q5=6+%JB;EU2D)$M(>L#ENY_Db-j`TB zt|NE=X-~1Z5IFCO-I@c<5MQJ%Gt^i$MyMZ!}q+o$LASl;h_V(Xo;uaN+EBAuf5 z2<1aK-tkln{^k%xG0Q;U`A#R)qd5Al^*81C9p17coEA@3ki3>U=3%NEuwohl+5_Poc0wAv;BVSxiFNdXuW@4Fo=cJ=xQ76+^cjDVMqa-*sl*1ma?NliaO`?&prXvg_`&GsPLu}|YtEKKrSOeG)_KBPoC(J^zKOUA-V}UPz>@7# zsBzQVX9~vYGMt;5W!%~4M)7&U`D|$aqEK7N&mzMqQz&4ac1)?k9387&;#`_Bx%40o z7t%b#&v0MRA7BFP$>xgRm1V|gELB~%*?JI^7EfKUqE=NL6PTr2JG@lV&ZHKdO?`@B z%PrbZTnR+D5j#iN zcTp1Q*SaC^p_|V_fY>mj0w|A3y)cdBC&8N0`vol-nj>?lk|d>$Hnlk80IkvxH+qS- zaS^NlsD=;so&Ra%e;k|1tD0HtVN)veTV}jEldGB}@n4Wrn(|R@@gK=7lxoswA{<|I z@$b?G?Ulu$SEALaY?wZnjIr)CrSmF<4sAagQCR6+6)8OyfG+xo}6%S2B;H5A@9rA{r`aNB=>(Uy{+ z&`CL7hVn5|)fvQTTdpJG-nApWWmy_aO9AwnQRULb2l`x-;e-4u$@a^nw#qJx+v~ph z)E;>Ns=)M`s`m)lRk*sxf5bhYFqKmi4s5w_M=!nkW!k3D+m2REQVI=(ffS|Rf?U>2 zm-80y>AFMBCD(jE-tdF%p5CpdbLZb6i%weQ0XY}PtcQB))5tB()2{2D8j@=jPu|HE zQMaN+rv4+17ltRCa7mH87KaBttp5*XUl|tVyY2lKK@k)TLZwus1(a@3LSCXc&eLk?w{Wy1TpPeGuJe?`t37ynGb7pF38s-&(8v5h!judjL5G7}h*H zgwT%HYs3{NMtgHj^BMK{$=lg|yJ;y;n-tvWoG_@9UvS9VoLQwKdL7M|K8t%G`mD*e zD4c*sQ~ru)b3Q#{5$GRr=cH2_PC(grx0Tkv@YPq`aQo?#1YhVs1}5;r=MpE|PrGIT zMSySGgq8JCJf~mRVLG@I5s2STYr9sdn)cbs*W^Sa=}f3S1$8y)J6U{C;bOyKe7Zeo z3_s3&A-q>)x1B4B!l4|Qa>ofxiBn8G!axumS&emf43y z6w5l6%GkPhL3eHuPY5Z2mI})FY=vMT`FnkoOMtAI4&^!oIoM{bH;dEyYRTD922-5C zt5mpHyU5sso!(4EtJ%#qpCptvJ44?pHM*ohNO)6Ti8~@6_=#_J{{tPJ_;{k0*%U-r z)l(%|^HthamXTNSv~TWp8DFvzSz}`aAWWWf38|7z2a2@h>rCjLU`gSs7>0;R(i!jQ z*$d)4x$Oy<_0MxT@Gm`3x+vN`x1Gi^A1tLWeISjm7J{^@J0vLPp|n>IlhaD3ztDA5 zL0fYOp=^44+wbl{Y3%F1oyW+M?rCPI(JPWxwbEuSg33FC$aYSnOb=|lTnFpi;zx~Y z7lDY0qt~cL?Frb1CaXDgRK6$tSI7&0z|5fv3fv2_apTNvSvJ~CU>4KnIrSsLB! z8MB-Yb2#LeWd#XkE|3#%_P+{vry{>^2AjT>!%rpdX_rBa-*OgB#a99ztx$`>82WC z^U?_OaK544v<9(>Z+zA}OO#M};17v!tu*p;=-4Q9qP_Nb`I7t!nn8mr0Zx`%DIG-h zVITO)N}@T?v`vcAA@5PgPh@vVaTycYWIliLeE8{8;qxBsF}1o0fwTt zBu)$FO1b6=V?Aw_W1k74O1Oki_bOZkM6f$~oki-7?1}T z+|NLGn(*Rr{M_2cXIns&PZpXq%edpxuNO>Ie+@T(^9qbCRLIE}u`yZ*^y;$tMvBaV zFpO~2&?t@`OEJ=LFw(P%td@4tx_}jkksq5D$`9a>hub@Aq;biIWl$PaMlFe^ypg{( z(e5{~-TUZmswK^j)7alJ*f=F8tjJJ7NnmoIZw*I??O?7H5f!Usd-BakYQHa3gO?qexRK+@<; zPHA3h1U=e7uV>h8{GM`X5@VRoCQo+L3O$qZo|&IAsBqT3VmOysH{X*%QCcPChATz!hm>Lj#@)CH`&6l*PzRkv6$!|L z1oe(uwc?qyteUx$HZyf9d^XM;};K!+zLt?m7a3d ze!lu>#!=~xLI6a7W&3IbvE}D}&6QsnIp4{elwhM9y9YB4Lz_cpLp2g~w&mZj5>CWE zcc$jF2AK^kDui(5q=>{z%^kpfRC!n3h}LcTfvEY5IyP%Ub9GzFLO?uyaxH&$Iqzbs z2w-peMu*ar-9#whS0K;hdpg*IaWwEC?Wrg{Ab#SNL;|;JSK)0=L>G# zuxUMFENap*!BQ%?CHoD++pvw|Ew}RHH8jGq@sEwzKg$oG9u&N~jEbqRyVJDrSlZ}W zarzP;1m>JX(lg6{QjHVBZuccx3`7*|%S%TUJQh|xk6q|0BaaY>sFAQ$nR$ALO_m;?$jea7 z?R_v5>Pve|2MKPN+8D~nXJChf8+j=^$guM&lCEfl*xm}BdSN}>5DD+1lWgrbb4eAS zr|(;Zft;GuieLajg>KcT&o% z5P#55FWlSl`a6=I_O)Do5<<9zce5SsL`wEt`Gmjqn&v`B1H2%t#vc-w5v#x>rYmWX zA57-&Jg1MNXzXnQlYd+nKiyIZZ&j5orj5R;FyjQ*cJM=L zrKno|ja8pN1|)U2mLP{`H^p^b z%xG;-6E%N$J%*~`-dXHol3z>y(u9o#USuAVuL4n!4a zf;zTqvA#*XAozAS>u!(FR=!Sq>G~__+S2b$U#z-c#c|CENvr?zqaDY?T{@ICLabth z*@w3d1BRGfKD%PGuFP5XnND>|i-Fu1NI+jRc#1mY=NemR!7IWo7M;l#}$ z(jnV)+Dr&e z+4F?mjyWPafuMKkNd8U6iIAPMoiHqk%ppJ^3h;=+q5_F7&jwPV_pf$@bjgY-5F5zx z3Q;PPopcH*j*uy6cpVV=L2Cx(jB@upxu!iRhM0_C7Bk-pL!UlK{y`yb_%y=6a4d7L z*$Q$C#1SKpG22urZ=eb_*79PeoV!LaXz^6i6Dpw8wQ{AeIaBf`IO+lDJuO$|=n=Cd zQ`6Q7?VF=KIH7}fbB>~;Ed3y=8HB@Gp)sDY1s%0&)`)NoG(Wb%F#y-QC1BhGunnRg z#_#Z7yDZl6K_C=eb)Qd+J@#Ikp+HyUsxm%@BN!v#qY18ZCK|bkj1V-HNf~6tWvYdmRmeHW!otNwv25q68`#q!jBs%a5m(poDfiu7W1ZmZPKrUzXtxV%{DY%l!NJ1<^dCJEASqJf|5B|_LH zchMzUFDYNODKlT#PMc!b6*RywoYway^Z0o&#y%myT4~H zX6K_k!g-wFb=zwbBcL#0!5l7zjGX!ALqDNPu(hQ^o3ryBvcJ<()O{t1&!lbh1J=-j zNk@BA@R?mgS*CL0^q|vS{*$BYGVYmJ*wXOhGl$XqeO(a3UY{vD#ko zl?mw{+UT5f7rZ4TIY0~5?^8E~3?w0yuz8tm6xVH~w7fXJ3$3h+4j-Gj7PTtdB#BCN zy|`MzpCaMSlMPs#5Yo~6K(McJ5Yty}GRUAMe}){i(s-w~dgPby0k>$W7NCo@)GF<@ z!9yM3SgSg$968(gmv6glK(r-K>nKN*<`tfGw>F@@qt>j{8(1E=xPl$%zgpar0?viT zCl~(dvV(097Q28`k1KPmKL-ahpU2c%7?*F03W&yR^+39%6+j!*oougQ%1GPBVp?eu z;a`~(C`}=ZVJL_Wx7`}!VRmmN9oP3JUS3Se(2p7F+=aC230X}+H6&U@MNWz^o1^89 zBGYeom7%Xbu!BC_R?j^+OAAOk$59aB#_o?WE%jUH-9L^wG!^vc*|FKkmFap>TJ*fB z$1RUwq|OPw8OTJWzjIdhz3eQQUzUG`)Hno_=Ci;Qy!xHAG~Z)h+jn2ltGuZ|-=AMz zZ<29XD7zBN&}}wXxS*ard@Q{lDAhno>5%8P9_S&9&sl1bB#Tqx8&hlmvRW~hrOQ#0RSm_6CP-)|{y+#x@*UZXNq%xfv8 z0F7*GJE&pY`v{_q&g~^%qwhCcm`<2DpcmiV*QuXKZU7|k>6s6E{tG?B(ey@JRR#3Swr2PNUE5Im{MIgJ(=bA8i3H4OMx~^9b+v9>lu4; z9rJS#K`5&ym8z$g*mmBa*@VerG5h<&!`U z)aE(TrEblIS{CLiOTVZ*R6n5BUH{Cj>V4HtjN46BXtqFZuAka$7atoCrq#SYXnIO5 z%SN(&7@@qlBLw2i)SB8qUZr8>;JzyXu;ve58G_4M6R-FeWKUz5@>!=dxL-WTE<5i! z7v7EwT%sxT*7a=dL*be6y4CAZ+Ou4oL?OB^}Se)}*%rb7v(cyb~YddMZEOIXA z*}spdAS)iEbU}P?_5}gFbUf~!LU@Qwu?lIN#cu)^r&?ki>zRhPo{ng79GD|8AC9Jf zBqJ^uBoB$uU1+zQA2MLt9G>iJu$IhDx}INg5F9Q#_#(7Ha-ABFqW$vH0W{3=v{$_M zUSaN_7x%)U{^W|3%4URCqQ;fuqvp>QOk*a4ou?#pEwvPUlNLkEm$>5)E*pW?ULR$A2S!zGhBl1xTeSpZ zBHHoar8dUAVU?M#cw7|uIp6zc9TNTa`@%jt)czCZanVBMv0@Dz0~vD8@r~EW5@Utz z;k~b7c8Z9q19-N?aky3KKi977o`#d7V&=kuh`shQYyuw7Qf%{s^of0FZb10wOOk6+ zdFZi_djsrCmfF=1zDf&eZrJAVY$}c?dZ@X6f;`Uo#QuYpXk%5Xj0Hj(C*Vn7Tk2G*)W5P9Zg2;Tx!49JH#?C}ttly%!CK}E zry6h>zsa#jhe!i-5maI(Q=R@1Y>YEuES9trLc10VZ=#cQ38OisURd%V=}vXEc4eo` z2gYf>b}h}kq@2JV>*;!jqUliDk1JYtzkB~oRx8n|NhMa9SnPj#pe4t_W0+>72%+Y*JyX;|Z!K%rxZrye&`eCkV!t9u{ zGKy*3X=U0=kEcY%PC3yXNi1!neYx0kg%h`iygtZVKs#a@>XarM+6^USR&J8W$euW5 ziXtg;NC7fcM2kf!&)OA$+;@3QO@S1lf*l7A2|*_3>q zD;BN_(^!_j6UROFyoQ1sBqeQ@94+~+we%`&Iw!AeetDf+2SQGNW$F%2UL@__81zS2 z{VYjM@?f#0!evJ6zN`TDsBh#np2F@|Mw_+849Id}fN5BqzPeGEN4!Qk3?0$mSNzP5KAkw^$ls=2~pBinQ6ja(Rt^=pJ^2JIUR0r>e4O z*;~Sf2A)B8r-_z>!ylF3*LfzNZgg$J-su;_g5S2u0zdp_A+LF! z_+g>QE$c5XZh#{{DSYKJ6|PBZ{>#9CKEHXpMU*?}+3Tayg9rHi2POdur+p1TUJV_$ zLYaHi;Qr0U!JkhVcNsG&zk?n>+;o@SYgu_as}gQRTG2|$t5kp5Vx4u^ssaiAQgkX+ z^ebU+)d!Wh=Y4Aqo$&rI$OrAs{gGtw4b?j{@*ivTq+hh#?|2y>Xs3wEE*i7j!WwP# zKr3!W;F(fnYRUA6myHDg1lM7QHpc%+87tt991~yWHwv_(>z;?a#70m)O5H+6TC#VF z*a@^-8d|Wc5^;AwA47OcEcdJT`1iIZ@cl^~G~630_Njfih6H|@otVz{=lL+-QteZ4i;v}LpK)MU5t@7P6 zIk8MvP$7jpwQcOZXo2`cy`iI>sqWG%D|toG+x>oSp1EiX29JLlkQ2M(P;6H^?=Wwk zLClLwa!>SfF^t@>Q~lmX$a#}ue1Bq(7x+2uG812lzGg5q6>it zG|kA3&o=NYF=aT*b(*NBv6Q4JMX!m?orAs>yuu1&D!6|>rW)u+c^>xD9J*&FhS1dB zIJZGCiNDa+n$(tBcLPR<%KP#ka_8<$UuOAT0#Yikr{*K*5yu5^A#fjk!UvVNA-=UD zV2=C_q6ePjZKq15k+RJT5xEnsAIV~qTqB-Ih4nHRm<0Psuzn@~sb<3HoIp_^zod}a zLsXI^x6l)#=CBI9V*H`6-^F@F01eBSBajX*VS%Bd=l~i!&~eJ<*;ZUXz*nix*pZo0 zb9uHosGyj?pgvA3fjBUcO9>meKIE+mUOb6*aj~wBP9>1n;)0OyzPVlcdo&$|C_>z&hmk?QfkHAXG{i=LYDT|{!-#|AeyK*aLpc_|2 zUamHYe&^U~x&OygLO7F1hwKh=$xl{e5En$v7QABI=ejj^g8BAZKDj;K+s`hSz2fbl z5Ar%wvrSR?Q*TW>m#$}szJkXQYDRe{+F`GJ>T}wfK9y7nUgGgA@{y``?nK>hmFsygkAkKQ*yV>$cmI z9V~0T05lYX6u#17(F?H%-Fl73CiMKq&0#~dKMNcs$3@?<#~K9@wX=EUe=3e=Hqr6F z$>q_^Vj-lLOG~kSY^S){_wb}h(~WzKR>3Ay8Fj7USCX#|$-2nf`=(D;?jM@IJ#L40 zUkL&n2kx&_oyM{}7^SGxP)t6M3)KQ}m(3ySq?epc&eDAbNJ^miZsdF(;^HVDuhVI-q8EKASPoNvHscy z&w;LrzjN6ros)lj`ZddQU&~?%9=tcf4D=Q8tp2ISySe)0B(VJ)gB?co_TCa(78iQVDx+ zI0FG);jv6YlPccA-Z@TXmDN6e?v$QpMG8PpZdp@Sp-}DUZ^;65Xv(eU%k_x1E< zCcCV<8Mh;)x=fAqmGhcR4eZw7J+=&fu&i+XteJdM9&L->N8zv;kr$GeShTc>)f#>E z#`~|+I0|0jE`%!A|B^o@hD{?nr_R4jK02y-mu{Jm=@fT10~N-U@nK#_T*tkF0`1x> zW;J}t>!mZ-1$(R%_6{f9e`3ieFJ1jQKp45#e^_5e81K+qq9u-gk;sYACzyEd|6(YpS~&O6a+AFQbFp34_}0$+b=MYi4pPpZ(=x zI->i3bOxauvt&l2D-LzV<$OwUX+?avP|=zPqG!bqI>qn-C`wRyvj}J{lyovTwz;)M z!9Gx|tleaEJg|95b*n%++Bu}h%FD5I)gfb%drc88cW;IDA#&LU3@seb8Vl~-Ym-zn z1#(723QFL!lwE8b%>yn_oGP(;guh=h%Ypts=D5Jak4ye`5_SY5?bP(cs^V2&AYZ!z zf(NLVI<`BIUEH&Whse&n4{sd4#mS_QWw*k2(kdjblwY3uO4u`t|BK-D2o3Lf%t3zv-V8J&D4^*-^uc!6%)G%yYR z2H;;@B39AyIRb+ie^ml%4Qu|gYVv#(4=X|*mt0Fbtm44oG7r3WYM$YcQN)vN>``-C z-f%Ty@VlhZRA-P|u^=R6_#R}l-pt9{ElNcVMjf3A#?LzoG~i2*k0j*lOx`_#n0Rn2 zP4}%xA&Ve1)w5fbC;Fv zMcUJ*6Xuc4C7crok&7gIiwj}+iP$hocIp}X{f4S~rBbEo#r6r8q3p`&kQ)Hx-6J6% z>y=z`uZ!m=2Qn0~{h9NiNxvQj8Qm&>xf8zyj-LM7l7w8$&osi?F?+|0f9$*lF4Na| zl5A4zt{hcRLx;^i*H~L&(w*(@I^XD)5HYLMGFP>5y{NFth%CVJ3wh(<=@G>y5D*bP zv<@KQrmQ+WgbuFQ0<(Usyt6csMJD+=Ed*UZgwo<24fC{|FUdqEW}<~eIB2)N>g%aj z7n2@as&w%a1o!f7V63H8%{0<&Z6otRaLOiM=>^T>oxGY4i=A^&tLc=rorZO_ERe=S z7u9H#)?kj&-b~XJWC!t5FFx9eHA8i`TyD^691Y-tM|kdGPL?zA3k)N&O1&aMd9>xS zK}7~Y{bn1PuUmwXo8l>Z;~kkV)WNVUS&ft`@h-cbRfh;e*@2voMHxQYZ9%Gn>(nwO zmX)i(xUdk9>$svz{aGo}$#3d(3tUeehbo-lg~mfNKn?zP&4&F;icbU-(Rm;kZU6Lr zq-1{U1DluwPahUw(_xWV&$jeUG2QfXb;%V(^x~?1ebOnuYm_xc$wAR#1e8Zfau@KJ zOAm#-VgO-IE!l9MRlFxIKC1uSn08!r{bI{dIJ>Kq-)MSfE zQG%NW$0{GA_BXwPC4hmg845Y76v~ffFRqVGPeGOLxZZ1JAhbW=4^5ScO5P1+p;xZB*rvck+5 znff5wy&1r+1E9}v)41iK0$~Zh1MM+4iilk61sP7)tNClu9gTX*{+?q=|D%r2;;;wg6Fbskfrj|$@Ta{0EVT6t;9nyJG z+a4l&&Psy*r8^!%f{3!2;E)~Y6MPcoBhlq1j6R*Epzgjf@Av^~Udi}nhj+m{Ui6M! zYk*Pc{^@RHc|F6JF&oZOFMO^kP9?URTBV9fCR55%E_5PcOd;ZgVQOrq&1kwQd5%n$ z^UM3T@7pc|Buq_Jgj}Lz)cJaHdJ)!UGWR76fzBd9$zM_7jl)i_6AL}?Qk`2HA?C&doHtwk-)Y~R0% zpul1@M$~d(*wIx5=5G0sZ}AGNob0u?$7}m+4@`Yd3CFXR>}P6B>-W}i*ZXv#?^Lul z+yVw(TSf6h^qrX zguzEJpoNl94K=7^tMf^YTr#vGE{2`2`z@s}6<9 z-Sd=Rx#yNa4}gy)(`G&^7k>0XAPIXJXpdnuhMN@}JC>vY(>@r`kv9M_)>Y9h=1A<= zjo05`;GCAJ=)YNX4sUq{9n?0D`GQL#yJLSCz`9$u+bcK-{er zp9!i;+m`ZMhkK>#EFs5RGAB-cC<>)|RHo?j2gU{v>+UkamF{%MzDxlPf!fD_?)!*F zgfrWmtX4WB;AzpAM(rnZa@Y&0zqN`NBhc0}rrR`Z6Q3j)?r8R>Qia|=9oAbXJ-h*A zYJeEs{un(Jf@z~c7ot4c{xnPLey8^&lcpuzN8uM5ey8GYrw4i#CCc5Y;zS06x!+d* zL0s+oAq4bY3SI1jMsTe)29f7GA6U|oh6QU}+@9z|eDy@DbIGF0lp!uk3gpdzC6m(V zhT^2P-uA^JTQC8~z!*(MaP8k8Z65are1y-GRCs}|SwGK?#xwuN_6*W}gEkR#-_g}H z{J1#7{sSerg%_p@;XyP!@e{;!ZbwdsQB^W!C#AJ5twkWFgL{0H?B5ujzS&bOAE66y zw$-ep8juL0>v-N$mn9E)J?i+omSLY`3*IY3HCXcSu(H8eWkqk6(!jbZ z&_LlLeiRV6cMgZ?fckXF!lE4t?d)GBsAnOsS-%TGc1UjdMM1f)R^%X{;^1idgZeR9 zpWT$bbC4q#Q`AnH_{=E2`(7O=C3koS!$n#=Wv7Img&p0h25EQiRn`0W0e~Uefn0Hv znYXRwEv0wG1xaVC(*JGyuC}TrabFkla->UDs0TybEF&ZhiP!8p1qzY0w z%Q@#92htbFJ88a>Zi&wX=P0HKCjVHp`6jP)Xvbd`>moB@^CIdIyPr3JBT0W=ze-tk zvKvFpV=wiP-SpYneAULQ2Dr9>|C9t0*fay{pEYCZ2ByA#qA8b}yBpOJiu^d?X2{ZA&Y$iV8EMsVsohhL+F14EZ z6|@#GaCe9)-~6F$*H~pFHulm@GMJXnt}Tk-frY=~uHJN0<5z|zd3UPhtn>aUn$}ni zw1L1;u(qFTzQE5JpfiOlb$55ahPwW|*ev-Vz=6i@7D^L?v#3B+3b>lEp0CT;lR+WM zK=%l1#PN5fU}qBth>}4bLHpK>GVpu?_d0(h2DoONR0gyIQ`sT(OR4)Gg&!!^e@jL{ zM7#GjY`b1%8Hb(KE6vzTv|duLAXarB`F3fa1LngCeid^ESwZhVq@^)~$*?c25V_tJq+ifo2-EEMb%tX1+d zD~nWY=;@&EZxc%65fYI+`IL`-#o&erCWXKwbhM`J!%^?Uwkt1U<13jU zcjriAj|}2e&zLLTvtyaG%nPAsS=TpU9p2;6Euk;Tv!v`Z5 zGgaWjzYk9M&+g}mw=vpF3mM`aGUiO|MvD~;j(fVUB6Uqpwkl61Z8j_gcl)`Qj!aLt z0^>F}n(omiA>7hNW9#*RKJzzy1Z0hbowZzLmEq|QKCA35G7PTPZ*)Rujq{3x0?Ezu z#=9^R%P9V9oW=j_#{Ctcy|sOE78>gm1RHQM>lI<@k8=$u-%}mFIJ;t=E+31>mN-pIuYY+wn>6n92Ol zE>LZDx(v(hlOAZ%yek8#B5=Z>9lHzXZDLTZcNVNzoZ`cxwhlXFp;~JO9qX@jJ%A7ev{1B-3lmc10f6_}wUn2r|y+Y2K zYZ=UedwQyeD!gI3xuz8PNF4F}?A8Klq+*T%J22bm3peQ+-}}P>GwOhzUB)(IL9GdF zOj7*Qhf$CJ^Hcw^3wMos^KVzJ7Xu8n#NUK@+pVY(xNUjXAUBt1>c7~Xs+$-q6Ow0r z+W02eGNrO5dYyip#rTWq>$7tZp=UehQZSUJyog40qdq|fIjoR@k-u<%@fjCQiM(&@?v!BY<8~~rGMkIC0fSnjqMv$P+t@!+8)&%NQE2PW0q_T(ou$Jv zmyY?`i!VK_cxQ*%798IoL{5!m8+>U{)%VF)mR7d3wRx8%-IPuC2?GgQPgjqq|9R0s zl$%x6zxid0%}CbBdunqJgnRgtp$g7fS(W%gVzhvT&+@fmXxKh(w}-#60Ps=YL;*3G zjrl49mIetW!UmTzVHwc)g+RI^=Yn32@YjG9qIlm z!G`!az*f&_^22yg)ddd^{E7>s(ppeUxhlOx*pGyAW-?7rVP`P=7YkwDzOM!d!YQK) zUiu3baW}aw{mQ{?3oNq(@_y#a6w2F1_=SBBZieJ2! zQM6xMOK;JVKdW!DgqZ2xlW*9)XHw7)s(MCcEdxF&fc zdh1(m@hWg5ObkVi_YA^_vNy1Qp95%aGWNgz>F*7?TQHb!#EDxB{YZWHxyn({plALZs%FP8dqaA(-|_V!!BlD)f=ijQNX(`m-`MXP=@ z#FPrKP&o?2-u3V2akrpf?;|3biuQ_OzQt$$)*;p=(NH+*?ZaR5Wve-Wk+2WFd{%=l z8aMkLEudfNn;hvoN&t0{q}jxpcA;6MWE$C%3IAi-zO z`4n63sD;61`zcb2C>NIBGHps0IE-_4S=t5KUCf@?2l&v!SP_3krZ);*NR{ri7AB?FW0BF>7#yRTFfZX(3+u7T`fMjc* znTy{&jADjy*dieIk!#1zD`%XrEVeXxV6zgnS2e#4iL4N3`2Sp$Iyd{pq;EpDPrtTT z!B#-#7xvW`$H>T&yGZm(D>`H*x;|(49UypPjhiv4MdY+#_Pvdvdt8Wy-{9uFQnYH-r}>fsM2bl1a@V1|jJS zHFNG?`@K!~RX|vj+F68O!0FYAH=a3E(zjQXXAf-P4AGD!@Z|A8#IC!=6m(G5q$xRK z@?|9qs*!+6*EkccPlWmZa9I>Ur%^oTJouOOjZ3NzsvrSg@pv@I65i$k zGfq(boz4Dne|($cNYBOv>(ttA$-|D}s~c}66QrfoXvNWV2I2oyzxVD|NOe~-)Crd; zi%cVerlfeE7qrBGDDr#L^C5trk`QVA1@tE*wQib>dw`h6{dG(dyNrsSZ%-4bxz>gr$W zz&D}4PHbo8Dv^3*b~Id&ukv))Tcb1v`TLHwL;&ISEGt#-@!vN~eczqzL{0X1IXm@Z zl%(Xnl3)DoXDG~jz#(~l=Iv*|)Y1Y!pdriOb@tV}?|TB}xDPE%SW=u4@z*VuS5r1J z?loiEd#3)LF^ZQD+}z_F|B?&Fu=1s-;NhBFLs_z@m%x`SFGVZYbEF9=F(Our2;`i$Cv_sO3A90 zK-S6~!m%#ms?xLe?a+H0cI7n8qhUM^ptGd%q6o%IxEB_&h{}Vx%RjZ+r_`0T0Z?CIc%kWNG^&!zwh5q3HDScp_&qYsWfo^$!yN~CK=-n$_?Zw~O zSnHq5|3fE# zge>q`_1EZzF;8Y0jlM1gS9f`I6yoUP=n=~ikU6>xeaN=C^XI+;0bPaVVAJaN*H;rz z|A{VN-b*^CB25T$85GS*6Ol-#_HG$fA_+@@knCtr|7L*aj=)-P)(}ygeO@B+Guvb=iQ9QP-P;~HdO-iMw(gJqd<6_urtJ-T!T5KYbiZ>aU93n!jE;o2 zB-O-cmEYu*g>tkUH?91#z7Mg=yYr{VJDU?dk?j|CfjGNaAYi2=?7ZE479&{tuvR`- z7|dZR!RH^Ra<{MOJyC8l_POqL{{g5j+GAEG!3(yfaa(`w%YMYzK!Zxa9=|n1&W|Z@1rzX)tDi((Afl=pv7Hw@>x*5btrDn$<(auOZTTby zdC8n4J3bdnCoEY1ruq6WfZcgt@om7%zmxoPbN988emsTYp5CQ@N}j4%cS_`Z@*B++ zW7u4{asBs;smB5ns{MLC?+sp{M{+QaH%DiS^AJag6_{-Y>Yex7%1+M9 z12$c~UcFrl*Fzh;aOs7F|J-J76mY+~kN7SS1->Tk_>4gGPoY=HL~AX@-#f1X^Pjh) zt4eIX(NoQ5Sd1xvzmIvrNjSy@qoqV>n8coZwpTviQDPPF;3xP!34dKkKwS>Yc_}<~ zf2@i{r`^R8WZ}G~0(t>^lal<7+VA&+xW7JA*jN5eG}V!W_J=dO97k!es18LHFy5Hx zD`9REQ0&lonDI>5R?`7{IlcRLBM7~|yjm~Y^ovpiOq*|Rj8yebT6>G$Zr2~#`e;2G zcxxj9*9$bzeGfETUL@(9T9C?$KH}44@$~W{we+(7%~K5fr{{N~o{o+l;3?W){J&7! zpUq)BHeY32)M|qCx0Wc|Tj~(ALZAN2Wdowc;sCj@?W}*j;976q7%ncPsI(UsoW0i; z557wnCm%H=eD2Z$w;^!$#@UqZf7H;QFaO)7e6P6o?|xxs$nfifH8DPDfRBMloJ{os zU^kzL|9KcwU?ZhE7Qq*6q!3%#)y&G-?WNYN7IB^uAD@$qkS^Jur@ewU0`RiUn#^-O zBm$9?oJ?-TM?o>N#vZFW(Ajyrt*uR3MTLZSr-fMgUwi+rkE!EfWr_jRcxVq;~j3NR^>nvIm*SKYcZXg!8Pkx?;!`c z)Y>3ZxrC{Ey!$X20-@a<)Cx$?$WU}^mi)cMe|`HybtK;Y>c`amoeJ@&j@|Z?=E?Hf z=6^FfZqF+Mu9IHONBxqMU1V}z-V0z-2_C*pLJ|M*dM&#_*;?h%WF4W}a;7{MN!iluxU!LdQu<)Po z(g%SeK;ad2e1W$o@be>($f+;n{_{QZ21fmvU!B$};LOI>)mN^+*3p^q)O%lUZ^5(G z%KXA%b9&IM>Vz$J?_cZu*9E9QyWzYW?OCj2AT~=L=k(s8+^MSj->fC{e1JysUEUPJ zQc@^PD|Y(Z4TXhC9kdf>H>NQD*n}8$ZxB3te|tgRILB#Z>B)#= z+uzmJeX-xHCLMF|wuch0fkIC~Sj0k?{eMWS0??y7Xe_~PFc{2YK*i4H+NbVBDK&qj zX&HTVbo3C*%Y~ff)dYtE}tG4ggIcoMd!65Hp7L$-gigEpK;VVP_XMFxb=K?-6~o&~LoZ zA&AS%(%e#JDtg>0h+AjX!8@RMizdtaUw-$0e*26BeDU26TEJ6V3Ilq4?aLVy7F`Bm z|M7WofZWU^$h()K^y0tP`49?=i;L^^B?CYLYkB788Os+2-QOd5oowag zWyAuK{kdqpeaiZ;tK@5o@a*);=GEeGTdP)0Xjc9A7XJJT5ID1jHq#Z*;Lw~v+tqeSP zC(1uCFwhWZHyU;NP#CDH*BficB|1saX$h|jk>q;-WM+MU0Xn9=lL~)l#CiPrx2dix zN&Ut9{`K~G*+=VL69t<<%t!4=W^djF%uMy$8piVkctImm5MJ;W4pz$36q*Y+04997 z$SEk;d44^ceS^!lug=24^7GS;duLR#T&}vFuJnfLZ)(;31w-!f=iU9K7Ng{>qjKrF zo;mL)47_QSMNkn}i$B+edD9zEAc@a$P>&4u=U28vUK?{0NwLt`4DOv6>b_AZdy~MUh}b?ifglxoIIk%#-Wr#Ec~aeBriVqYOz*p zG%HlhNs9d|amUtwj_N4`5TRQi zg-`HG<>59;P>gP?5DphMGBPSGEc_WtBkO5N(oi>u%d;-PgF4>IYhft+DZ#zcd(gTy z*Rf{JHytRLl$5kYTDL%80!)bW9$XO|V2I^;`ZqcFoDGB2v55Ed)Em(_sFAf* z!|!M<6q~Q$YW&2W_Kya@#7ehk`M9qLm$(Wx&;h=Y)V}z7p`$GTn8rk<>bQ_{3qPQH z7q|%W;OFm`0Tk|;3~Xau7>+jqCc)(a_9lK1?dQFOff^XL_q_L_Ij%6|MC?RB>@x3Me|{lgbx9|cNm?*8{iAe zz^A0aU<7B7bxSEAPVBDvLZHy8L=T+BOy5wpY$Z4Y0fX}W5@m+jwix*1z|d<^-BR6W zlI($|XzjZ{d7uBLq-uS<_xsvPK>S`MN49EwznQ~A;@{(ex9>*+>{31KIRmjSgR#60 zJ)>P18z8;X)6?gNRkJE9`MiC6xZ!e2YAgCD-}|qJckcrKGe1B7*1+`ELA!v=LtV>1 z6#3r<&KsG;$`W#h(kxE@NYLHu?TtGIv@r2IbH|x%{k?i-dH;{GtB$HFjrJpAAW|aT zBA}!o-GX$taJi&3mym7|1O)_XX^`&j&MV#B-CVlk9cBh(obkQ47JnRAi*vs7eY<{p z|MouCOSDIabrHt70~GR6SO0$dvbI4yW(E#1NiA|b+U{r(GDwWyeM7_tn9=l%j5TXY zDbB6ez&F00gh!}9rw{k|N-q8l)nc(@{UM!|gMZ-f{fg#ez-}#+tZg)PaF{44;0twj zKy&lOv^x5t`GsT7wfV>P^~DybDda2i8b{{llHIPhRAcEfqU^Ai7LqI$H8M8#VRQ^q z<&A<1g&){AAtq9Eb$qKvs2D)|FfvHr0Punr)+wLqt$UZyx zVuS@0d$Y1(JuW-wSkdx0LhHKTh3(+$n*2i`w&>RI#`U{e(G@!SDUh*BuGwQl*n`}s zEgeY8QHv%@kD%nY5-O80phDRhejaZfiZW(%l#dQin~UDGGW7hNbwtIc2IAD(07m5B zX+03i#=u|Y1fSw@KU|?N&0}?br)$+0gtC%z_c~Z#dIgxId)UdpH>>a-f>YYn2A@u% z4RLmBt4~AHSZ#Gpx}Z$NR}bes02E}roK@`KFL?Er7w~X4+4|mo7pCK}evUJQiRQJr zq6A3n8Mm$;j04;Qhc*04qW(i5?n)7ltE}_$WP1Z;WW0)3f&NMI$)F4c0Ud+xSYM+T zSEk3+7}y+RqRjue#ns*2o|>)C3>>!X?YS|I7}YzjLln$w9)dC|$!A!uY~m&mCL#Id zm;YW>nhGy7A(4B%WpBWv+$qSW!{yZH($fJeS{@lTWU6!kS#`_v! z;I+TFmpS|%x;OdP*qA`Ps9;7djx+fvwB85RStm5Tm|UY1IJ{+c1{M`#o1(VMeL&1_ zin3bevhd|+;PdbjEXK;C_czx4fFI?$Ka^{YTSP!<#+VsKzY|q1;Pw9n0*JAm1IrBd1v)6#^@u^~1 z{JxX|0LUE|1tOG8<95e@&gxVKW{HlSk>ftf zURY-ikbgVePyI)<`ioX|0h)0tPaw(efN= z&pX$Wt^)&&$CtYDL4ZNvw>ZzdHaS5ewqkKykPaHRn>VT6x2^!Kc! zrt}&Gl%fWgiLnz*BO;zE2FY-3@4~HPo0-3peXhBH;4)ed4g~%_rR@QHCNy+(*6l(L zO9BF{hoC}E#PKMog(Sk$_4-T%C4q{>+}ik69q3>3Li~dA4tqdG_vC3FcSwd|d!&nm z9T1Wz_tWRrVH$-=|16Zq19u9djWqcEYYB>8rz0!RuqIS-JU5KJ3)SMTtHM6AfSr*k zGfR^A)d&2PUj$c*!Ka`Mup%C0eb|P_*~GwcyJUM!Qsw766<#Wm8h7CD73pCNFD^6t z0hZO)(n?WVgb7#yKQvMSBRCO=N8GQ?9`-z-PVtVnBz~XLe4z@5H97n^l*V{v&syC@ zNTF?%d(eo7;HOp>1B+QwNBwPNrLiv!L@u)K{}Ld+ zah>D>D&i%jBo@v;DVoN)Aqa_AYWczg2tu2>wCo@MRuyVA5E+Xs8^S^d8pF%>+u}Oe zOQ^Xw*+D;9Oh9&n9yo6LqF6%DYaj6I#fxunp=14zLr>M^7QU8CHa7~06vnImBk`ks z$%gWwNrC>pbp9};Cj8i!bv9r`-*A1&{0G21>7W^~9uZwh5qgSrrI+-SfKCZw4QT&f zr;6mKB6cEXic#JwaR6EnLRo%=n*SeFFO47Y(e|la6Bqd-`m3cwGUAuzv(8lCZO87f z?e#U+_VMHY?Uc~a4|j?1W{QMB46q9uA;kuc{pOq5>lfBQ9Amk7F} z`aAQ^hyg$k3CK^6PhicD_?v-?A)9cLm?J|#h#m7MhTEvxqah<#te|nW4bTuai7J`g9O#1r>@c7AQ#0h9-)Wz_}n*Cn|mvrl(&k5I3 zFL^2T&3q2xzms~6xA2$6TNlR?HbUDZF7^N~K7Mn@rLH`hEMED^F*H5$eqS|{AJKDu z9T~n+MY8Mq&jz}`UCCcJIk3bL#Wu%lOLlEV{3kRr`Hrrg-NVcHYy6@Sxf4yOX)03O-v|EdL3uGw_9fG zJwtSF1UWCNE{|c*-r=`g1r1bqwG7y&h>)k6@d;(!0Sf%p!746EoEGV-jwgL-DYUDj z3V#w34`W8!I{N1Y%#~%=Q`m=lZe*Y}DgnD~k8n-6`c>v`!QsGtp&>9|-7ZSVlMx%s z|4!y%M^We#W(5h)hln`N5@5(E_Bco{p;u@7l@qoiOZ}BD=jrz?6Xm!f0ANo7_ITFe zjEn0!;QZy{D@j*3n)!iOQ3-ddvE}G?h`zq0q)&kMCE;NNLHC)lL`+YEBdU&)lm0A3CmnxAr<`>k7d!4#qPS8B*DW$YhN}ifdw;8D zIB-U|>@y$IakYkbhCzNh54*K{5iG=Je<+j5_)k8&)29tv^U&!mhPsT8d>vozA7I%~ zz`d}vv|CIT=q!R_Wo7yW8=YV#jaZ4&I&Vg&agPh#c>E<`Iv*HA|7{R5e!q`Nz;SNS zYHJ7)GG4HMAz$`#n7i3pkD{%a!Zm&@7T%XRx^e!18G3-lXyTX#os{Cq;cN-~F2GZo zZ+bR}vfB@IJGj1lcY|mK&@9jaPbR_YcpawvdeIcmYXS6J0KwtbWM19(PmUO@D9F6h@F2|K5z z`m~B6^vvr#60$NXJh)JXJbVxg}G>~I_Di~0EXg;$l+5OIOp=)aTANE*C|V5rq;W zk*q8T_++3|32LI<=SccB3q`(%OTkFF`iMnd9Qsg4$C<nqsqxDEjx4>7_17hi~ zz?1HjQj`D9Hv2b{dRSPKXu|^agXypJLdX>`p$HbDvk9eSxxI?@<-sa3*L(WHp}mUX7E^8 zb`47W2<6a{*Vx$k*ysjq8UUg|qlMAv$*yrP#llP-z*ya9E>c`Y|0a2T{|Ep9LF%S& z|6&-fBlQ;pcTAiLn79w1$4g?2i`kRXgvsmb>}kT$cBPds9e3aH3oTSg0mUC8t~BZv zQRB=TfgEqF5Fk{923J(u>xW)nd?fyALZYa^%=%LK@BD4dT@I`2yu;D@yUR^ZWwTRO zD%*~i%(mkc(+yniG-?_z7tu6Ufn;KfD2fgGS$8a9a*(a9r*HdBCOgo*pTMmK6Hs3WT8#I#(f>7ksD1Qi z3m`QfA0Tbbg%`Dv6lYQTA_9b)%47_^eIvv5Pp0Ro&$FCL*zzZjn=TpD<2DL=RDA{K zDmUDq&!xm{m8?YneDXil`SG_J7?I)UYzk2N-XiPb-n@Cot8r6;ecQakk>raZRH=hU zS}HEpnE@|)4<^7Eqw;=HqC%@N?Ki6K{Fh!tKCkH-|NLrB?J+I`A-eG{Y{Zbl-I3yQ zWCtm_&z;e#Xn|Dfh!4Sg8e*!JV%m&kqeXl3DYSH6rj_}W)+3bVA7s^=dVa!<8rw@+ zkW~7*@^3bc_Z9+YSs4xODL;RP>S&cfe)t=i$1&`=3=75}eD$&-`VHF(7)cV){o}}m zK+rCcWpzXOyLf=DI=5TCGJwCn#r=mLYA{?3KEuX)sx$lVP8Wea4riVRoBxm`19x!T zfM~ohz@Am&E#LUlKll7$Kn;)pZNdQDR^B4$ow0?~IXSt$!9k$nZ2%OTUz?kEI!M@9TD}H>KntfmtNBeoD&G8#)v+X?0oMLA zh1-w90*z14pHx0z?DG{-`0nChAs`=$*%+E_>hn>mXt%FWYSF00duA4@*nnr9Tc{FK zbA#XA3B)UPupI0HpKgJP%pS8-%$jFGtaC2D?>x2T73vklmj5l4A)4}-Q z{Am+~Zbu|e`9|4lQ}XHW`ZbD#iI3EbA0Zohr(-ceA+4bARvLEk9c+<#Qp%AwLrDss zvkt|?PB)91^i5V>`V%UXLmNI)C^ZZOr{P1opOk;QKCK!-;3V-hACF{TRo!qI!x?l$ z@<)k zo*i%;Z>B;In4sAk5a0g7_1lD`l{uFT5!S3H$g6RbGAdllJA;f(UQA9>Oioi@w^>(f zdT7E!Z<>rH00sB;ClGfeAX;3U9oU$77G0d5K!M^ob6a{=7VQSW_3K#gvmS*M3-iCm zBl6=px20CEs@QQH+4z@$|1D+w&o{jR9$Iyv=xdPuTJAWN!sjGKJtZe)-6h z-WJHT-ZN`$-Q)+Q5W9CUJUlJ1_F(3qZ+MiS3wIJH<@Oo2Szx!*jhBpN0Tg_~T`

  • 1(Hj66+G!_g5E~qMD66caMUAyjzrG(+jZSf zsEv`j&}`)sx~$9*1-|d%?zOlbJ_Gb}9o7|oF~8`giTX3PI|5MuXA9B0jM3gVZnjho zXuxFVgS5r(0^VBlEy46o$GhJE&JDSRSg|Ik+!}&!R&$Hqe5Rp;X1E8rbgbq)5fWH? z^QHdbt3KB_NfkFjv(vl>iq0*>Q)zwUZCat-e^x4KTx`buA9I#*4sSqS;kvV@J}F9< znNPV$pLuYF2X4Wg0hRBS>ful7mvU`0W-i3Y#00sI1>sBuh&h@0kdyUdzmbuV4pdp8 zZBsA?1_r%$lh2T*Cjs-u#_yVByXahnl!_aVl=_6UjRt8t>~T|)DxM+71p8?N4Ivfl z7XBvL3pb8Q!}2nPP`7s_D@7Cbt3x)EPAqmG7&RDO&QFD1l7wWoO-+Exdeo}~#+)ER zU~!N|Op_?0(-;}DO&6fUaaETRg)JtTU24soLQXas@^#M8E zGWnu2NaeQ0Y)$8nIe%6)T>+n?;{xBwePYZB(86S7Wgp!max~LR(VrC(t`{(PVK821 z6&3uY$4r(x#&7pj+m8$Lve2w|04%L&Z|{u@Sp8&`a9Brn5fb8sYbts1Emb_YrOn6Lznqu=R3wvOev&d1EP92~cijN&U^FWpGu z-GbA;WTUKB>%FD1l;P(mGE5~WO#rzhD%hxiK=A-8ElE5qfGMk}9p2E$X!eo{wj1$M zPZDqsPWgmRC*R21dp2dt&PL)q2}g>(cX7>K3oN%B=rY$Mp?@W9R?gJdgdBVRGm7PX zb-!W^+x%=Mq{Ldr0tg&3r7T1zc{d83xFx>@`T?pVV7@(x+;O4cMD0*@u$nB@)@U;c z#z1d9v81}D9RMLSb|V}mP=jd{s{?N>@?5ZmM?`1?oVG7CBGANJNr1Pa^78VUS&H`~ zyd!1mm0IIMqwl3YKHB%$d;$$)tHWtWoyWYFQv*bx}3XX^d`3`csMF?*d&k`qFIoTyhH7<#E~5va|gNY{uwy z+rxwH0xe(lnsv`L2WkV<74UJ=XyG`(F{C84@_)XDZdz^&&TuL$8LH(7Z3vxIRUSfT z=D2a=>t)(wq(C+N4Y3Uv3ij?BRM5)G3g2KUgiULY2{dPS0EeZ%E2 zgO%#k;mzTnu5auUdnjzou2l81?yIEaLRug#NMKVME2TRb|6 z5vdoqYu zOYPHhxugbwH8ftzrDEDEK}DJfN-Zodwz(S&aFoiC1_8M!7?Hu%7ln=jydrx%Nrzka zFY^s}RqF-L?CK661^gGJ#pm%cHx__2CMJEGBiPPEOqi7@I#2rO#;6Rms1HQ++Uz6C5EjThER zL*%fQD|{j12kNF6fOXjP-&FkdcN70f#d;t#6{j~ko0O2(eVUMV6n^ZFbPp-9 z1qi{KWb*UU8Ib1BzT61HYkpiNIwGAGqI##Kafw%2uW42ygO-}1G$2faI|+M2fb9)? zsn_I#^O9NqmCEFTp5_D9zjCapt#i7NKyO%#TK{~-f3in?Dr#E4psAN}r%GPfuKtip z3dIqR3GyCnr~W)jX!KdZ=!D4kc!?gvye0ZbZR7h5EajSeS%y$l+V5X9gh0M{cz8Uc zqVg^{_SfFM|Fki6hJZ(Pa5j3a`dnHxvI7(C&HMR^*HkNw0)yd*#)2*ym> zH+0w68nxR406NcB$QZ$&N2ic)LMW$FuPF8I{11T=Z)InS+YpL1Q&DYBMnA87G~;-75knDR0bLc6TR13S{a zwQ^I?BMO+mpSE?Y(B2>K_}ViG%#F6Aw~rfX@QQpnOe76XTel`8eEY)7e&!5Z`2e6u znb7|N3oiS?W$Tj7g+4e5n$R*+vhCJ?lahO_Q>er*3gC7CgEnYlxKu$v^gNN520!_V39vfiB*#`I5g{y|Y5q!v7;YK?P=9BqcoRa+&CPJjmI{85@@@`OJ4V z5Yj0b9^to6VSN3H4{E{dg_2TqmK(oe5(fuR+hx3s=cMBR9Ql1E1u<-E0OK zTx51*MebDWAiuK!Gk~qlL1g-cht}}(omQ2yb#7Bn$H;F!Hgj2}&pO5Mti0!L@{H#Fr0u_q+S9kJ+Z-(gBqm?edAg(rnhW?3Wx*FC5Bklht}8jCAq(`Vq*XnNGaq`6>p! zszhA=hpWsbXM-}N`6&zL8--D{Vx9rqd+Xo4*D1)!T<%Vn7ywP(UxS`k?c|0Ww&76) z20%&C=Z=oJF6RQ^p6tDbMJiz~5~vMi8(IEWcodt?TNXURe92%10`GjAtj`#uY;U7mY+2X;Fno&OA7M|FBfT-5>^Qpyn_g-mY1+) zsgFIPQr|ej6PvWvQ&{8R!m$LVVFL!+53I}*%b|9D9{x>*>m8^8Nxz5oKg3x254dHA z(!Z_yX;ltgZ^m;vFR4grG|#6n3WA$ricJUDi&>pSp=OIDjzpeKL$dEjt>N)Sxfc$} zQb49&+lCIXzhsxhsD%Ci6SWcCK{B!cB9-A0iqHJkEl@0})Frve+LhhLRouX@3OnyT zPN3RUGZGF%RL;D%|FJmiP&kN4#}d?rsj(Ag;^li@?J?G1?3ciBvN$1+22whHH%0z} z`-WD$Sh%cJpFZ;Qz32E&!a+4uo>`O%jjP(EwMv=sjnrviewsG4nJY*VHzT5gT@oVK z0II6Cu8Ais1^SCDdN%JS-sM!rx=y`b-`R)bQeU=4@7>!^ELY3*^3uv)e`j9fQ>FAm zy)cWE+eLz(^VE2H**~e}yBN5$I}XQ*Kv!JP$Ooqub>~>}BzYt!ySal4DKO`jRwtdBu1RpK zvDB^VlPsWB6WmEB{{VBi;X3}lzG%w0q<@~yn&QsH1%nioQ>uAK`yNBh%R|XuC8(~k z&j*`HvxfqMqVTI7h(-92riN@z9Pwk|eSIUMynCKH{%NvxMEWk<{RLV(UnoDqg*?`R{x9p^BDs^dfRVDowPaG)y4td_T;xC{Wc0zcB2AB}T)7VL+sJ({%F z(~7Yje#)MF;`5TiJ88u^eTCv_eEMCQ(SrAeS zJ24wGeF4k8B-+kxVdQpNf>M!^(sQqm7QmGbfHlxIl@Mx2Po$INcG;t%(|Tanvq-2& z2%9j~|3D3tw6}_g`UF)LN|R^X9fBL7zO>#v0LR(oe6lRMOyHQHY%&IfKF^OGu(Gw_ z*0nhWIG0PmP*K39X0kLKUkplom3eWN9zF+pU%A)!uZ$`-r|YGUmm=$TVrO>9lnHw4 zyB`AtV?dObY*3R&xddY!W zU-QH=MLW$Gh*6m{hq1&XSrPRnRp2v4j#jY5fL;yJ<85LBi^+8h^ASHIZ7WEpLTHja z7L;yMo$D#$cDbhtP|0I@>w7kCqXY%!sl<~ZP(*D{4wEEyRsjTgW#cTR5xU!wX5kS;H9Z@yqPo(b%l!)m3IbCRm=RBW>+}Al)`I`DIpt zhxAlc^61R0Qon!^U3O9nFa|(aHgle8a282N;M$+5;o~lsqM4X6wNIq$F%hT48xP9pj(586Wp(hdGKzy6A=QcD zNHIL#6mEI4K`FqLWa4Ou>QznagcACw75et-sJ{I*1?5kj0;mW%Zxuj}yDWR%U1LsG z-=i6OzP3ib8DrwJ&n{mH&vmBZ7F_%#gR+lic<9@v-$nLw+DfVukQ9%D zsg<}?8UfO-a4TrCB!S5~Og7Tp!~VQUi9c_A*nlVX%Lt*UQu4v1-Ky&8rq3W8@9n7G zra>B)I4*8W<@KiIveR?_c*+@I%n(Z?Y`syU?BKMsvpK6{7e$G=;Y{j4&A_l;g5_98 z435h?Fw_i-MxSuAk|kU)yE;iBGwW2^Du{gYiRD|w1kmAG^(vm0#JG`Wf%alhQ?f7C zSW8InFOv4;&(RJSX|w}C$JiGC^fyIC_9v-nK_-pXH9ktMlzUr(%jJ2yQ7S6n(^@^X z6Sy(%(Z$(@IWZ@Ll+-7aXB?qdeIvb5$YPWMu(G|H$+OIA5p5%xv15HP&K+60`J@5$ zuJ+D@sx0aW8?*h;qq5Sr=QkqV`>4VIv*#4Y=Tl9&vpgQ1@ng#&$4AS>`IobD+NDs2 z7rg_wJ45$m&yQM7Q@z~u_F)1N1Mby($v{W>Nq)Uj_1UI-%EVF>6kR0oJK*@N-z@j! zr)!^xj8_*N^*hmQsTb=i*|daPjO-u1)-dH()fuv?>uh2H3yLmp5zYzD!d<-8!*MhViyXuTqN2tR6PA0&aOHPmy zQIL)1u6NaQ#9KzMRyWwplD1b$p13EB?B1%9&QPI;7-Hov_2b;+q&hmqa=gFtK8PbG zkz{Py#uCZEQJCon+NR+>F4;w2P~ z+g8+@@n|(CPp2db@3sDdwOYZb#`E)i1-5x^R2&G0mJ>I1XUw-Ks#5OnEOHBH$$BE8 zq)WIuwwwOHE>$?z5#jM;SSa2^b1+$iS3agDS5P7! zDfJkh86%W)rZ&PZ;17haH-6-~;2skyR2r#e(M(j?RQos%Oe!y}NiJ`@j6=2>AIZ-m zt=zXXzb_QaEtK3$aN7;Rl~RU`gp|d9iHCP&pC#9DOtB#$K=?n3E(Tx3FK^NIMwIPP({kc z;->LbS3_4VjZW6x`q;#FG#=G@bwTA-SxDc5%clFkn|N_LxSlxa`%aVo;|2bDnWtXw zN41LT4Hb*=9q$@o=aLtG@NBI+Z0XaX6jTP;JCbu$hLaiMrfBo3O&6Qk(2i!U=%h<}*W4H4$eZJKxIb7mtsc!r7YP-| zU_FC)GmsxK)dslJCiM?GZ&ryFI!v?5GnLk}( zf(qn$XTWG!KeTJn#(95cDw>Xo;NhPs_otUnB z5-Jyp)v_CWCwjVgEz=}v_VI1`E=K8#l|_8)3v4bCs4t_VQi}UWO~066GNzYltq-}J zL&_*Wu6}hl|8mYQ*i$Ir)hE4$ChO|uiD3zoL)|;>DM=ff~7Z$nd=4i0`_Y>&s zpB6$;dsb^xCpyFz@H52MlJ}P?K}o`^^<(2-+HyOf2x8uI{>$e&z^BIepHghM&Qmii zt5xsXxpE%(R}Xb&%UI>5w^D`biXkC^u^2V)Os~Pb5*KAQZdbnAnKu8`AOa@x4>(|` z{Eqj^0=^_}3S7di6_viVt`!E6M!~>P%pXbjG3UN9a4<^*W)IxMI_6}50%B)=&ef5t zxyX&dmD=0m#n*|dM}Z%$Tgr0KHQC2TskSXvdVJ#Gdww2$1aWmN8*qsHLFe~+jpJ#4 zxbJ}uXz&TzeD1?X#JB$W&@ zW#~+D_Tfbnef4qx30BmbW89M{qvA-|WRgcAd;Shu;yut=$09K5=a)CQ_2>nB6CG`k zOIsTgUEs`C2Z?f7O^Z`TBcnz?vlWd>qRUE^JWp~N8S_dnCC&FO^109S0bwie86^ZH zyjc|S(H`OFUzqEL6l8Up-?CUg68_bF1%LVDqfqd=SBRVgG2`=rXOhla-LSC8^rD^# zs`|X%^yb6K0qSH-eCmB|Fp2f5tqWF^m{jvuF{uZ%4ZIuaE+fes{O2a3tQ)|GC=KJ* zc~}eewR2CWXG306EMu&)8L`Kk=y3Et5)`@h2zmW>{v0)R^ z1=%7l_9-RDFVeMz_%L4dJFpJWogKMVZZG4)ZWW@3CpOpX9U_*VZc8OOfvnd<4UP#T z;@o_UoV}u26+A{t{VJ)23|AjCN;j`MslAL3*C;elCwL7RotmnAM9H&3@vkjSDT80g zpbh9I@*o^d5q1~cYYv)*miXmAO^e1R`WA~U71+Q0JRoN^DH>ajEosYja`R;0+cD1j zL8$RIp&R|;FwuBq%-YFOZr zY-Oc;UoX7ZS0a>WKMk8cmzJnpsA_V9W-KKU&c`O}q14m82&MfWBdA(m1#w}{Yl&<+ zx5|uh_G7?##k@jJ--QVn%tv}41#%>4e^Ct zIyhHI&)cu(Oer`jZI30n?5O{3uI7n1+moCaR@%D|ODJWH=s2H)vse?gd>}ZqZz`Di z>&qoTGN}6a2-U-$1!93DklqAI#o@`d2Y@yCJ>RB!H*FJF558$aW zNF-DgpQ1?YVCA36DCI<9n%x=gMSL(<_%RI?!9N;1j1a4`Fo{*G~>c1+fWAfn5sQ$k*x5|c#&b$(iB13zOFdX2-n zMz)?c>4W3yEKT?e^d~;CpomNJ?ktq7JU#SI%R;OKqaN~$`4kbKsx*C|qc`D<6I)Gi z8W|E;@)=FZY2QX5PivKXd5}`TH%*DYN&bn6&w2yX3a(QnI~T0k&!je(rnQ8JV64hx zY-~`6No`CFS*_+FxP(yv{FPgw*wmp#CUoW%E@h?Z>ZYm}R@&_zdGLKEfseu+mU0)= z7<4k0b8~h7if^TH0D+dr$u|5j;*KGAUu3~@e0hvhfsmCZU7m)I?S5fa4CMuP@h)2; z!!u8;Mb2W7^VoXjyY$qoKODAZZqxEO_am0^!79)h=8C~?o~8~i=5d_NFAMU&Dvj6S z{BaC}9&NYi#qnn`rA9r$v$o|Bh@)j)7~aWC@OTiT6)Y?oVKyAbRP2_X0TBJ5bg z}F#B;}Xv4|elUN?w?6EurIA_m(~yY)z#tMrsL~#U;Z27M)Wd zOc=I7SG3yAAhAIx8={Rmm@2(jdtSS4_ygvUB&HiQeefZuQTvy8A;7I>ZeCz}s?V)j zp>5_PH*V8#yEz-13Pywq->!LY`Mvramh1>Enlk)&gELq)qEv3ayxZS)R6kKKE#<8uLp8x$|f3lFTwM@A^y1K*WCI*XRGN+X&%T! zHP$=wKfqdIEbpz6F^kV)dQ?4l7os8!fh#qG|tPjr&z zAbs+q5u4sLR`;+e^v*WS?wnTOV(_J02X{yL1h#FdEPn>%xDiz2UPhFq6}d?FfJQ5? z*kTr#j@IoRH~w*L@!MjQ74NdE<}^a(2Y0WRM<-mqY#c&SE@NPe(~g}{79$1!v6ov@ zyal7V+3IR?$gQ8FhMF@U8oLpjdLd7=%dn)4mpvJzu~NF#*>2zIIlJlh)gzu_t|c6G z`gp9x!eq6FQ{tSHSk90gQ_JJg{U2+VJfA12#m$JWCTIM2x{O}gfk zC3JsZo3VhMPxWA6GG_X8pBlDL8hL1Cx`9xdg0RQ=UDE4Rus&R3@T0J^xBuyg`$x0e zL2?%Zsj^i0;TdBp)I@2B`)xo#SiKx|RA^*dbH)*Zm=US-w0E_g65r@sLtXJ(t@g5C zFveG20TKX-36wq(k@v&Fomj4Fj*GJPaqGO@yRAwSx-M}(wGtkXXz~K%x#kDLETK5# z%F94#^qx*hQ>DvW-DvSQsskn?!yS4ZoxZ~Q=F^Df{9fUaI@CW^Kd)hD@c zvh4S!L`C(gUa6^@$(E_6x_!u%ysu-ukYCPqPgQ0DX8*S3hwsFbT>TYy*Jhl_l^A64 z0Joh!s(XIE8hQIWllye} z^g0)PYG{6&Q6J7qq^~N$^4z2KRi26{DhVT{K)e%HdA)AkwjGh1V3fECJ||^1#OA=B z);13-G+I_7rzSqeOGr|NURFnnXN?urV=Yv@e!|m`4l`z#gI%GFD43EdNI^itjvd*e zzPvQg*kbvGhRL1)_6|j72Q7~Ga3)+vnAw=J$}=(kNEdxYt5$QHSOx<%sDoyoDv~yf zNPU1)!N{m4&P*~8H&XeE$6I#oJ1XK8n~ z12^efSG#VMaw66k3GcEH4`;V$;W9{Cke}pbXR;FdeB94nR~Gwf762z}&5=KQ?HFba z!h%;8&bw&@e+24tdJgp_x(9S?15x~WmW{Ku_TSH6^|sp6TpY;g6Pg?qO1B>9(cSOt zZ>Ay2h8##W2QE`p2&h@t`X~`j8&Ad}zm`*QXK~Vzi4FdyW~XkZ<)J#(cL(*WB zUl6G;*IF+-hiOH^kv8GlWvF)a>RX<^n~R+dQhPv%ReEb-*#o8}-eM&?=TRwkP6?BK zGcyG)#}7pDX8YrDzREk*9A*%@jY!(i)y0?SKmEEq1{{Y%Y_4v3HF{>CfUk9Lx1X*k z=PhbSoIS}hZ5X^tKSRxQT;^RTpL|hC}uTsi=17+8Y0f*j%vf+L!O-XCMv0#ZX z$EW`a=7?RrgaKUYWvdVnD@m97$9(DiJI^{6xKiVaN1&{`A^ASS>K6(0uuf-drKw)e zlus+i(aW-LR3&H$%0bp^TSP-9>DFV~9msTZVU-L!idfB2M8_O`Lw>MSDqbO!<^t0P zCr1~Rxk`%02wbR3ZLG3N6nflh+5F0zvlY3a13`kGudB+?Y2V0M=AgNSiZOretV!QM z6&R;y+{Df-;s<~4&RgIcOm!79l`nN{J-1_6r}s#%cf!>c zz2^JZd&vuePs4dC`O8k#4{!sbagTtn))!~z&HD)u=Bms|tGj3;L2P)9WqEwMM$I0F zdwe8Eo_-MNW-^e|KVQfSr}-mm@DA~+HU8u3hLht1!(eijWvyJ>QhC#3fhaFV`J(`& z#_etu;DgfoR4ZGD%E}~I{n)hX235}ho?#HL7L`TxASvJ2D|bdIEUPgWm5^-h042gP z$zK5nV8yR62YC+HUw{B3dek@He#^SnJOrzG^{+%!BE{O-wt=I**8{)()*|rvchb47 z9x}F`*GdwC;ZvVpuJP2m%e?t~P~q{W)wgHvZ4@$``f<-jPTFY%EzY~{&l_3XnmUhX zrh0?#to2vQCTz&YoJj-SNV^vD+mtVttRF@{!5{0gPTgLqgR|3p^|nLyS$2As20obgjA+VvA!jD`DWOuyPY{ZviW6nisfn9wV9|ufHSZVDN*^ft$=_t9`NyG>7vrj*m zbPjfn#!s{a2ZOYW@005f?-?nEg6`?IK-=dWM+&;f+ZNJ==(KXsmpJ7H*NCgb}uebrv4<BH?29b9pD??*n>vA?X)nMd7Gq_m%b|uWeDjol z%G@mU)H=Zf=LB|CTUVK>7XwiELPs4aa2+JpTdpKZO8Ty!+AI(Yw`qrQ2KKKkimfT&$A>peNpzdO~|^t&J!W z(~cv`igS6B(ay})lB82S`c>To5(hc>ypN(N_F~gIi+W_rWxUra?;;E_FL61}znneH z`mWunBRF^LG|S~%&LUllf4z~z;5{wB=dE?>RyYaHTda5Eh&+l0r48xToVWxD+9$la zx~Vw`1`dYby#dci`r)GJvs@ZP9l$v$tpVxM%4waS zJb8S(I^Pm@Eg{7{TFbZI0uD9%1KSS;>;zs{7WV783#VEJW_}|H8OdTOq6Xs|5iSqFS@_k z%lMDsFxX$%v|`*qOMN~EE=jrC{z5K8cl_Amm_X{z>No1t zELC!fyjYSy>h2hM&F!P`dHX&@zh6d36h=ei*r2sA^w#$Ek_S%;-k71#r-$y5Yw>VT z-3Rr-Jl|T;;lBjR`><*Y%xySp>_e-<-l?24I1`fB7itmw9YPX^z)x8pDp}2g5BVMM z;4-VVKw07oRhMUlNG*YEs&)0g+ zaZgVKrkIQ29hs#SF-}Nzk4SD2r?|HD&5L8+aaY2qY1)Q5%5kV*LsS3r$g*>7>nb1h z1eOFL!?Y){Sah@mCXli$cxUx}uZZ_^!>gegm5Bnr`2)yyZnB7-<74%4AXG%(^s_ddDdPk^f>HWdi-o->te z^|H|di7t;#aobB2q7J?GxNUTpZ#( z{Do|1Kg^sl*c28iM?rYJ;|^hUfj#qHa$2iX_$C~7oIiuiX>3au@U57CKv6j7ZxXd^e!QsyBPqv&^X zD0WVJSQ!V(AH4(1G&sGJ!`j7TIeTR9cn2Eez~`YlUg*JAIA*zzn`pA*N@OH{dUjli_p-a?F)_Q|MCL;mr1mK*n8|^BDdthR1mb&;p`r?O|$|V zD`seJTT+8D2!(cU}8xiByDoJTJP3_fH%b9`&RoFgjKC*pq*8H#ZrEuSL|bU=-N z=n^C!2da$qw}J>znNMcVY&$HEx+;L*A*OpHGwGYx|qq3f> zt-G_XI3v*&K|(J8qsx6a%2cVCTKp3k!(Yi$Dksy@V3**mEG~7WLBR;-CA}Y&XyX4v z-dnd-)o$&>3J8l{ba#hzcM3>1g3{gHA&r0_A>9JfB`LiqX(Xk)Q@VL4_jBvM-Oql0 z`vbiHaB#qiIp?^pG0t(0^BSY=bOQbZ!F8?|LJN7})a3k*N?uM&7eQdKH;ajW)iO2~ zZ1CzW%<@Lf>iM^Xw-|`DyCqQ|Znq=oXOO%_EB8vJ%+~x2-RS42vN3=ogol(<>3HgE zXDIM>y^zK|tkoN%5Fgscz4M}8I5WNno|}5alg@I#F?&VoIW)0Gu|51DH|;7karfy~ zc|qJ2ctbUwl`~~B4>n&MYJvyz-Np@?FgM5boBidp?Pb!Psj=zmVy)Vc@66)-L&B`$ zAvKCMuPw!Dth&piikCw;Q9jz4EcnR-Efrjj5Bc{e&xpp<^#ebc0vX#>sG_NB;OxW* z%iO)eUgehQWRYQVbCJj~;@!bLG-a&??cRI-4lli~aIzM!>)N7h@;Z)fFonNP$xgTT ze}vU@fW_B_f^|NT9?eJN(@WE|-oYv_eyuW9;kW$pQ@`rL7VBln%!Y8q)e+vbqu7_r zpoANTVPD=&zs9|Z*v}Iaa8Lmfxp;i_i2krDX2@U8-iFN54|AC6wltmP)fFqfu;lDL z0hwUsE`&~$*p{B|`>2IiM~&3LEO!x49vM1m1iPAUPvwW)7P?tWa*EDaL^Z5@6s+nt z!#ev06ucV`;GVH;G?^1W+&Hu}9VcIKI`=omg!GGT)l9P1AL11BcLehN{Mt_ul&y0$ zWMf8{$TW3uscMY{vj06be<^hsr4nU?QmvVO{+E41%VB!1FoTs=lk5jrS68`|1yr`d~mtQT;13Jf0&(3_5| z*ozi*IyNBFdHv>P6uo=JOu6b69N{!?yA7_9kwqgR$qxcU`w0t$d#p!rSn$ zhM1w%nnLW$6mZmd!iDmO8e-P0W|Xivm91KL+_>~mr17Y zJj`Am^CbRAAHj7H$K3+LK4y%qA{T^PoOH;d$Z*Zfh0hShuHe5$pN(e?c) z--*C~0PF=@0K6;O(xjEytusCmx>}uW%=>2IU2fLTfmgcnwl5yR<;m6J3F2CM`Ft?? zXQEUlzdbDNw_xPBe%%EMfDC(bI~q5(K05UC`S7bLe(JTp9AwGk6-i#G;xj9jrn0Yh*Ty=<#scOV3d!hflDnKBV1(fNgYpa#y zechqc0F;qi?OCN?RSIq7c8smHFZcd@CFpA4FVL8^A`kPdfx71Gi^=H+#)CF!X}vXA)KXaYy=@Rk(sB;V_2GrI%njgBo;cVHFtt;e#t()&Q3G%PjxVjAweWU5O{RsAi;>MZ&6(XO3oxs(08!ekeRkJ^*5 zJdc-jWcQo+wR;OPUZ1qnmeg%2XO~CLquIZ}wEC}nJkmL`yG|WAuqQ08B7NY0G2a8b zWoA{L;G7(RrMV&gA2^G3QGjGR&eE0ayEQKj$LdbRVi4dFn8kN5jl><CiRrnIeYJ zaG4!r$%|K!0&eE@>L8T>>M-(|5c^&D6wNB)P;gDF=67Wth3KiXzIe2r7p{vrhtA~- z?B79M%V}{|i1O(rdl+vHGp=8;YtIWBB`?KuuFe`JU)hKjj!fC-?0s5p_ADwc9V_h9 z^{vlLJ$Ev?YEt(Ji(-40+5b}kB>MhBqy#cW#wgI4240^iCvA0G9sF<=_rD66A4E(Z zm02dOkoCD1nf{2i>_WIMsWa3>k4FOE5NOZnw z_s0zn7ec-!*W3Px{IYjE7s=5GYZJp1ZX)fu=nz+`SgO}w^cPW|Le_V*)ni8cY`Po= zt-M|q!5EBx;u|5d6HTXML3faO;}XkTDvsbFX>2l;dtWcg`Vx?tZutH@URy+8Phr|{Xmw?k3ZxVdGday+PM77SK zC$~TIy7S@-!QnhWO<@`sLF75r<@iO!-B8qUnx#8}VClwe9FMhJ*xS4s6d4|R=SZ(; zc5thW`pPdlt}{`?X)w@WVV8xL2UvB)VUow?7B4{a99qwq!Dij$K2SJ_SOX8S1h0+F z`huZ*NjQ+KtI4I~^m2&jvEe0Dxq=WTH_c!U;z8;HlI16#uUk;K$@Z8JLdiUZIg||? z)L(*ugEIi3!Kc3k-13r8G8%@%mfzV$7lZO`3vKzXbHG=v752*t!iV7~vAco}FXN(f z#f!DjhX(NOITgeQ1EqNP^rrPQlXd}~>DKJ#1l@ro)^q}p;YFlo5Ed?bV(f9;mV|T= z6PWT?cfdt>!deLDmY*MDunHOWt^X%s*as0q5FkIm9a8@q@1Td%r^=Td?d*|=)Nmj7t< z37bu*!u^E8(Gy1{E+EZ{!(%|mZ3wT*;9;0mMMCFjwU5>*w#9I>a1VDOgHY73N5~xd z^3nZhb+IhO8gO6yE4g8kjeJr74?_{iZd?`4ws_0Nih-r|?CwWc2^)s3f&rLRI|#Q}-=YMU5p(AH+BWOa zDm#z);R5jHP8v^|Xi9FxZ*vbnW2APmGDYfqg0>FKGRXg`FQ*EO>0Tf9xX#A+M9h@E zw6!QT;QSRkP)nO`$Wl+B)xFyn$V>D-vJkMQC9m>cAvukOMZ zQ@!5I*|ydD#nPM6o>q3fP|-?fI{>#N3*On(#h^#s^gR_mNNwUS0ox4;h&8v=Zu(cW zgJ`&4wSJ;5Bz2uFI}&>MzS-%lqWFCl-WEYK!Ob3TdrKPAZ`S7=#B1-Cym3uE8k4D2ROAvKIvz4SLQT79{clXxCP z#qxHy2bnKsxI+9O@tFa&t%qH?KoNO@G&Red6SAF^e~8faW*XF$zQbA$5&O1JSea#u znPGBK3vpuX!YtRG*y0sgNT?+wgm`Lw&O}}=hfB`3b#8F z3y%~7)Vr*>Z|C0g869WvOzxH7n)>xp8ik{6L6(Qcz$muLlBq%HDr2C=L-Pad2_MA`?(=CCY{^u%kH`JNr0%&D zq|b4{X|%D?ucPyX(>z2#4xS)i(;x}URI^9tptR;taJJp6*bnhdb}XuG&B;fmU#K@z*%L4Htidsy z$ld`3-(-I$n=p>>MO`|9%^9{A3-nk&tRsqgO0|M0-XuOcayA|r6ykc(+J?K?FvirSg9%3A zZk}MddB&%~i^9uqyMk}I#Px%RKPlNn|fPsaz@;c(B9!AN^3wKm% zbZJa~oN9(gEe~Z4~&#rf!jQP)$0oeZA6bx$K8M za23c6s;?GQzr1XKX{g!Aq{nvWN2+|A=|rSqy;vF-@DuxGeNBfAfdua-b;k8XduFog z$$$wsYE(2XI1DPEvpKE&&zu7NRt^ZmP&oPgrzgeoK5qWg{1FK2z0+`Ma1<68oydYN zY>f_)!X&#>axn|fF>pNl(bKj&w>ez!e>$zb3yPWL^_j3(g+*@0jvDJ`bU{eXjpB|8 z(se8r=b2?wZ<>@UXAcs|h6y?M<~9#YX;dGtv`oDTD!$Vy-?Xo7yFr0UF(=Y#QI<*u z#N@sAcMD~*`NUpo!@RN`$&%>W!)wdl64C%^6zuv$yU8Jw@ih+rcd_QQoyoKQf+GzGG+C_eL*t2j+qyHnV`B zxNn^wZ3Tvhg=psi3)j{Oii9M3_yW+~!eM6!JDS!KE{@;vv)WTY5ndi}2OkejJwt17 z;QNAlo&MDS;`M%`oVKXkDBJ*;o7>bTJ`?Y@5da3Z(Z@1#Ldk@iMGPdwHqp%=@+<9BaxFx7HdOv-o2 zL(F{NbiO?#V?26$^tI_(*Sq=!#z26D@`V|R9v&1ty5W5OT>c1$5@^iR+v0OW$*GQ` zHeP^ldm~zZg!@PSCyfaZeCU!X>p&VrnbyXaj?{dbH&50t*oYxkOw}7Z^P#km56(I` zF0!opO%>g7??F~In0#q?XKShM_*yXC+tXWPZ3V^a4U-_1IFn%Ep2#a6o0dww1U@F1 z)VFK~r+Y3(UMMtafG#9H1rxs&Ca$s4qE&BN^uplyVR^aT>@{n9TzQ z;p^q9D>xbJDF=We^>g6GAhFFD^qAA?#?I>0r2RP=CQ91J6*Nz)*@HWcux5*$NhRU{ z9^ngw4v-jL{eCCgX>2e5sMI*Y0a7*u(Gcny9;?Qq!A?TOR})!0(6KQTE%q6c5GWZA z%|Oz%G9pNdEgkwh8lfBPcBfP5lNE+Bu2vSOOGc6$jEi|rMr9?bvi+SGffwV(%_Nhz zm>EX6EsJVac&f@t7G~<+d!H|$UhC%EdHX5gpf}|rA_P#@OP<@}l{{cjwvzTRPB$Hj z{1^g>Zv~H;eVKwQv%urSA?Q0B@kPn(No;dOPQ8k74do|6<6lmN?}9aJj6;cj(^5(9 ze?yY}@{>Md40iii)jZ$qYJUqP7Y72mVTL1Y6iKj#bpei*gCUycDDD}j)kL3q36ZnSTo*Qsp z5(R$Yk=}u&k*VER6-ShN(R}RD@>Qp^Wi)B6DPJli?K{~I1b#WB$czsR!V+M%o^jub z5VM8`z3mZwQxf99>MQOyY->3fs0d1Xn_hyg%gwK^*djlW)vy&8%4#K43zU2n;1Fa6 zy84z^88tVr8_<2#7LpxxF>#la1r4)lzsYWy#wBSdgqY6r(W-XHL4r-ot>^QwX=Gwc zJ|5GH1?ZP;XJ-=`HKnL4RqL7hX}-r4}8InYWBFT)QZ! z-9=*bDPhF~-&_faruz^lw{R<|OIf(rEI+-m_Gd#n$r5WueNH1;db0Vn#Sg4veL5m% zf!*b%o20(B{I@}smf?Q{OBL(^qZ_L2jdGx2(C=OyL>1C_{j zA<%#DdRg>9ai@S@*QGWYrU>k>iBoiU69AoXac){?=S*6y-E6%%{_&1hND&)KuZ%h2 z=JVQ1#ttD4K-B1Fby-BQ8$$E8ZEZ8j;~AT;a>d6h*KYgpdR0jZ(NRr;@O2> zi5};MGO0@x-K5UepYyL@)$$4jB$3*Lcd^#nK8%BjYMN?vDru>A1a$A4MD8z?r-$%~NGBVjPA*3QAsW>F zn8flvhD9dd33BC~SyLo^kc=zm3{yF`P7}KL@D7K6E)_hreE)irTv~Ylhx3(DdARDL zyMunq-54G9JX+S6E^oP}50+}t^ymMl_^gQS zQ7L&mVKyt1g#Z@0{D_#bs9uzZWn-A5wth)&2uPheSaoSSSlcn)@dq|9iSF<@`wX9= zyP@8+WDPGj9eoICm`TTxdxBd<7=;7=H0aMb{0-7WsZw9>t)9$vWE;kQaf)n z5>dIkl!?6WWcOs`8h2WhUdMr~PFzo;5FP~hb#^u>L$mBFLwbrlj8Ez>! zw4y0D_CW4QkGH&^avo#uVsz|HoRf4Fg3ugwYTLaAI@tTos=Qk}gxa7sUolDr!D z_yNGaq$?h5dKyV{5_3+M22Q3_dJl6t09NF=l0M7X3 zFbhPc`-|;B>fdmR9{(x#GMZSKd>W53ir}fz@Q#Z6+Pe_`KlTJE8NhzN1es!sK~XAY zz?2cyO*uLgrmMfP)#OfAkiVme?f+87$j=4SqAc1|jRi;T4sj3>F)tH78;pM{^|?Ej z9Q4`wF!%MVs@_FOQ1Qb9D>=Ji_LvTkx#H|b0P5Mm2C87oIfzTwv8Kz1ne2dItvs@Q zt8a%B@8~E7{Zz2lMXNbibzBlb6p>?O_zBxGE__7;%P%uhGMQIfK#MZ{yZGeJ-|ZSr zzlEgZtC+uP*&8T*9ROACdSf$OqAY)HRbS^@swzvUcN?{sRVI94Nn7+$=WcP;*CVJA zm8=A|P7v%!U+@hj>N)%A9fya9A(76b=(eZO-a7W{I%AZod%OKru$Q+Vmh<@)=fGU- zU{^xDOJIbTO#M{|AMt^W_-vlcEG!*VC2E>p{`Ogf=GT0$khHO0VF@#jxx=59%*V|@ z$j6#tjb(E!6mt9Ot0=ERlWV^40F+GbCI%VQEq5=pKEU*JSX}#NWQ!$)ZyjLgB3BCJ z&njCumWMFNCY;8F;FY9e?P=4 zVh*#PON0eE4_*+oLN<{Ot}`XLaQq+O7xbgDKU)>Y>Gn)3ZW~deLI`rR@!u7T=;g^H1tyswhL*{6N~V0Ds;U3!N!CG zFOdseWaN+$+s9L32x5tL&nGX!5hyjWJJl*QWy_s}v09L=yFdbS2+)|&9K*iWUWXib@r)DMXBVQyOt@go|gq*@6C|A^6w%G>5*<6 z+|21+IMaw=Y{X~s`<81c{Dg7LrF*9rv2-rP3xpKVUnhhW{SXa)59HypD*Y{Q$oryc zeoI>8g#q(!l{Fs3T8!oQ-d>%QhF=B0MkE)wzS_)OubI+~!(!0*mcnV3YdimSG+&;8 z_i8OJ&ulm&55H9p?gOdwOs!37Bmwu(j9WLJne=M+`#i}AY&OHrFqNLwuFsji^uTD~ zxb^FC;!~g7rm+Sl|Sb;ud(= zlkX%Xbu@{$VQA1_*GTM$%LR{pB|(J7gYa}U<1WFdp9A^g%gYX(>9J1~Re45nMeMsE zvm9WgUbCXQ2)fKyK*-xr^5x>LMQ2{!)VFvA)8$y}D{mDWd7);`FfSnXM!KEgrNutL z_$xjyVVZhHOU>p3XI)cTq?Lg7m zwbt97j1n&&d))T6%+0;Nnq9o7rtNsnX+9!C?0adjJz1JMYimpKTWbpZ;{Woo6!Ovm zZyNNMzwx&Rkzx{1hXaQblwO?mfvl>y(;U-!409gZw;Lv5Tm~a8k2_yYJ4&Q~n@35z z>e)Se)Z7VTIIV}WKoIZpWkX&&=lK1|klwxNKtay?9QSq1#qq)t8FgIsjb)J`f+i-c zK^IYARjyCR(#0IC%W+aEBwWm8GE7}eL9cX?CW1?XX*{f&Db(_tJV+M&vef17hvGfO z?Wj0AN>dL-DMb+fostw*@c2L&N8-^C`(Ys-J#6@kYWr{957@SpkV{pe>KcGx#7qeo(pFd-WMVv@4vzWYS} zyd@g!=u@~{_uA%6U?^x-QO^WZzE9QQqC_1IGW@aUhSz+4cU?`Glx_S%odZQ&w7

    7H>K>3|GI?T(>OurUy zpd;cWYr*1Dd*JgJOh{)l(7(FM^USg^zm=aB_DE9w-k)U$-tLWb5I@+1nP=clya~OLOcwGGaF-R2(4E8=RS^g z;Sk^y90@yO6usY0oN{&=85baA6NkYn04a4b@!#!|><4d^twd$-*gC(2A$uUF^SWID zH2ecg3I-qy5E|$n|B~mKq)MXwu)n!McU>@>Y?qJlGiSSif_)~eni^#B@r{xxMXw7T z1=cY&%&*ry-wxpBHtnB>hklCu(&GIl<*kRX6#h$8!?}f0t8az)zDPMDAQ!X34^~xt ze9xyblPPcxLr14*W6JG?qA0+@s$8Vxh8*(d;pZvp**4dKO}J zBMoPI?$F`!*pp@?P%J={4icx-;0Y4y6WZMfbk zUSEwCt|g2IhQX53(N*ffs1_-;{Be-f-;ors53h(>%-`txLkY}AF@LZ4&P+8!c}bj@ zB9`0SsBmC5GKRMU83Ms02g*#(fFKUadsG51abeA3Aut&y$$n-$0C;35XFBz=(=}?LIFG7&MFTaKn(BmtdmMeu8Z^pv1(O9l z9O<6MBpz+G!;!=T!EnZr&=Y7(o(I6GviENhN9P)xs3<7Jg|5~xdV6~-qfRi0v2bv7 zP+sWNTBF53?0p&jy3uSph>ypj{}w;~AQ6t(U1SJoy*#yA|7RzJ*knNM+v|O%<@Psh zqlJUArxx9+vEw4P&wg8PqIM{UrKD!W#4a0bv^1dn=ED z4Vr4#c7(t%%Z%{ZrIaeA*-y+JFYOmZX=~c)1lh6BYE%Gh}ZP?#F{5oP1lJ}w6t=jc?0NugGhhZo&Wgh z2Jzh z-lf2a6(U>ch~pDMZn9VA;o&TUE||p_R_+LUH~E7sP?3id6y)jniVCKZ&dC2K!1!x{ zkI;$94ySaz0@QcLT@h-+~*l8(D=ath}Oizc+{$B$Bdz0YehjUjZ7CW-iY=Tx(Z zpt;HYBiBZUS{lI`?O?1E?}ipF(j4CGSI3G;n^69-te|3i0Y?VufD5c`KXP(w}t0!xcY+Wmk^U}vH@u|%(R!4WTD_Jd{sEog-_ zHqR84f#3xOaXT6Ondr`FUc`8*&XniDIV#i3*;<>pvNw%=vdqnE|NXK2L5lsB zIEz5_Dw4(@fo|YXi5VN+4r_yifSm+%K-OGV78Ui!(ncm5dhs5gYD&+qnI`M@_BNKq zng^wa;tD>RloVgkv)bkLh@X5aR}EoO0P*mjhZIO(J=QZBHN(&T!?~Y7G0DrvYZ&ys zJDXYu8vNFPl0I1JMoAIzIe%#Oy3jTlAKRBcf?xENvy-Uo%$pCxlrZ41EFd|ML?`E12(M9XSog#Sh{qI=w*QvWA`bWxR=c8HUKZ%e(()7Px zC!G}l2GI*Z9H`f{)A?MC7h8OePlkmP(TLx;&AYGx=%G-rwPhfUSCx&8?X?a<A!y>lYa-z@kfJR`Z#iudNc~oZ7K~vACX3qO$u*}0~6&Pz__R(j~p{MulQ)q$c zRg^zzat)xp1=a3Hqvrp1d9ja|C+4*1zKYDin9>ZW>!yjKO$U<5G~c*WPI0v4UbbrEgljF5ssz%MS*L30`NR_(6z!-+na=q4AA9usFL{jqr{lZsY zO!LlHyT%Ju*Hf&@Dr>zavIk`5{;_xgqshDg@GNDcehJfms#5+#(V@l#wG3F+mEx9_ zFbzUs-3qLs=XIar<1rhy^HWQvfN>gM6s`^$57%a_%0Jq*-f37)0-Z8jE^`Qh&eYWW zl~bFiV~UB35BC?bz$%kZ;i!xPF7Q2-96cjrwS<(tJm38#0M@)N4zNI(w#~p1JOIp( zm%W+w^!Y{>8oxG7^T*m#dfXmrugWNAMSrCNf8N>eSIfL?V>)fvEo)qhm1X*A)4MU4 zT6$#J=(=AO<%B_O^P@r(j9@ceWirXi%gZ}jYojv_q=?=?*sJn;@a-KMO2E-|7v*!= zQOj=va!U9<*52)K4%;~%JYfH>t*^fd6^S3OcQCpFpk-cGR@+R-ZBG;t1F-v8rvI(~ z4?PyJVCg5;jsN|;^}nxK(koC9gMb*~qW6h>!+L^FN%vRy=!#DT1=2uBfcLmsPkfd= zAs29>oXGTMt((9uel?1J$eQV_-uucCkJB=~jR>ARuezG!Rh99(HqqYP8mp;3pou(L zrq6kD+ugTZz+LuN7<4dx&ZUSJ7`5K`Q&6GC_t+}vt85?aA3*mXi{W=B%mZ&dT_vdj zEW+gE4NLkE=2w$~$H=OQGtIkwgsIt;`Y5#tVf8`PX zBozMNw-4q8))Bc`5BndErA^^2gn}Pg=zsV8__v!VIDNe3)5Ij?e?}H*Fz})1FezPs zT-ARvx*F2s`iR_Z`{M>EA7%fs5BOEj{x2i_{<{HCqNYXC5{UnmI{f9Py#e#DL5DW} z`)2qbAO4S@^dyhlXpz$-;-4?=3k71-dPjQRe>|V^&zp+J0N&pRFAzcX4?`C?KRy)a z#CYVt@z_89_ixYd2uS#1NQe&pWCoUi_ah7yy(GDsa~f9PKL7G(A?<%z^kb-qgG1%s zY@MQtib}5M+0Mt$2G;QZSVHtt8;n zRd_A~{_%K8LmqDr8+*{Qp`qa>Fs%~sZB}O@%E~dA*w`GZhvV!ZY$#ZJm$rm-_uleX>~< z6&1PVnAOZ%l12dTta9Dg-X1TEbKd*$eYfJn>?$i)=smc5gM|rih)1W|sbIM}*vN^@ z4w)E0K#IU_fzgF{oNV;gE&4J@5xK+zPMY1HDpYeEK&>VKvZ(@c+Tsjg4i6$Tbo8f3 z%!P=nCRb8!WL!$fIt19r*|T!aGzwpgQ8bBx!@{!x;KEws##>A zh3h^lL-TIx)|SJmV9PICAo8$OuT__+=}2r*(w8um!Nme_f%Ke@Va?rxq3X!|aG`lJ z5vGMUE1a{?@8Lc}Pi`ZLG&+AjX|ElJNTWb}9>Ikrz8(xlCg3U=c+wqDr)mr&E0cX5 zX`wQO&vj}+wy#Tg2zczXIIMqEs9~{?aRR82VX*^+Uk;zcoYb78<;l$v%Tn6Gk|(v<0w{=8Fk`&meV|@~Fq&dT6)7xD!o_ zOHY3evtp5rFJU1)k@r_?JKOVp}up# z1S-4dErBCjM;)kjgmiUwi?(N(Bc`Be1>W^|mBHxhOzh%X`@OR9^(Dk#JM}4?|1l=x-CXQ1skiNu8=GZN^ICi2*#pj@J6VhJ?vrM(EytpwGV{DRW zuzhFbWUE&H9)_OzFN4~=zs%3!B?cDIV9G#yA}Km`G|N*MM2HqBI?Fqq%O-^i6Q9qy zKp)aiSr7l7M0nxau(R*-e1A6e&1B+gu1s8s3}qTzaG4AuHsc4NW~9Y12_*5>TsqMh z)N&P_GDA<%sx5#iOHjMQYjC2mnqtjMuErYUs_tu&;j_Q22{4`r(O}p;fcRpf_)uF% zD15Bvwa;1n5VU{U{w5{NhCY+zL(~-K`!7M!QFGD z7y6JQfT%G|?Akb+(yJrBIo+lhOyy>tzLSgU>C@NRV`J(42P_RK|=Hro7A_@ z`%(lvswSA~&BRDe(L49G-oaJ^%V(>fEZZ-_mu@bFUORJcJpEsrsgg_|hnTE~>CIOy zg&)2R9c1vE$!f|$Qa_0s0}0jvZNYP8p6=^+tMlV z_&0yo79rw+5WT|76?*lWaa6n&;b7Swnmc=~!tVa|DnS~wY6xCT^;yE~CEVd&^rp!4 zyKe}_^Du}+F|v}@g-}EdzH+~ehAH|>L+0MVP;>;^(HJiU2!Tl$BibO zbkg$FoMkXWa8{)al7bJ-HFOGGR39+OsHB%hPAu8Dq%>OaT#I)(2xGu>7j1=1?!dtC z2q2aj;nxks(%O*W_41RNs%TU-fIX7ycDQilfn9xo6B7(-~;>v{rHTwOulFGLq+;yh#`vo&s{ms2^%9;2fIsttxj|_TE%5$fu0SP zT7YG@dliCe>y)ik!tJ`JwfnN$5CzN7i6`194NZ(@4Yl?^*hU;an#CdD<90o*)_dfq zV2uEAural=oC{hXGzLr9_je@MN#Fddur@_0p({}!iO*yzR6~?{G7RYi0zRR^hyw15 z-e^0NZIBX^Flk; zH#(XpyB?<~gccUXq8!RBg%CX_?LhE1L{w9Na$QEdRs&51SLi2&{d`H{edIrZv-_j+ z)B4aVl|TR%E2J{JHB@~Dgc>5l62FQh8{&(2O$Ab(EqlYy0K-@-dKCCk&bN2El^sIM zT=_vycgP3q_zBr&rW%b$gAZfonG|I%zzJ3xtAE6^f15?P??U~>umz|DkCy6iTh-x7 zKOK%>`dT2Di95BEm#ABH-LaYBRfdHQLnEdI({CZi6j08d>`Td$Q$F}FmwE-l1+JPA zY+w}7X%z`y+GE@YM3)kwS3rCz{a~sK`9x~uzI7p5@*c+R#p!lhZ{%SOKt16HIv#MB zr~B|*M^F49!oDy-LV*{vHM`Wd4vnMKD4i7tOU)=oNb2yoh89D9e;YXbZR!!o(26lq zJOD4i=^nWD(gY03&Rv>JZ$}xld|`kQqG`X>D_Gk*>2$P4@%auAF|Y1qc~rjU0rb`c z!aG>*et@=o6)M3Ox?|lBu%u!L(L7w6uIi8em-{dO>jTizmu=mNghEF#6?{Z-v_GQO zxV}H5V(?uu8imCW28Itb6Ag(&oPs{Kg&#|+l)y3v#M{*Wph>Q9+y-z(^-Q2Bs&_j- zMvL0=3u6)(w#s={{jp{*w@tc97|fQvCTi%)6l4sOC2qq`&~&{6MX;0!tt0fWWnlZb z3^bk@1an_gQ%FTIQlLP=m)v)=+G|TZAP&!8;{z3mO4w&CZT>+rL>L%EO>hzc4D zgd!+dQV+optNbT92U67gq)vZn&>d6Q$2}tj9_l$(fmKXnoIO$!%1F`ofF{8+s>?x( z5=Y_X+K=?)>MzXq5xJQg6D{_SE|@v|U+sjJDD|&dw3b46r70QJ*cE;NCC^uQ%V--! ztPfiplpsy_wGvtKo{go2sC%qfpKM(K@aD^B2T2j8vgck(KOQv>mV$3uc%cu?z$aIq z6r_Io2xnJ58}U=89c4$hqof7)GT$VU53xfi1bXOgN&8qH?z_&S9)Lw#cXl2F@gB|F37o;Dly){7~&dMvy>3@lu}EO!XOVfNK4mXz9M}0 zqo!_URP#~*BFs{+n=qg!%h(;jIzhnIyFp`ye052q8DJh8|Z#d&<~K~FGrGC2c$QZUJdy6QW^#ZS)j_d7dv8L?gKCiRJ39wtC8$O6`iXf z_T9bE_H8%}Eim#jT&v)~a|w3vFloC38j6&3J_!`fi?Cj_1g_C|+GVt$6ELwq)?CgK zlu0a}y~qvg(Xt#^52%im#y7?|h#ot6Jk}!DfwE*3WH`B{8cmp&22q*1jJb?PW1Alb zhG)0)UoAX)W!pf}0~gKkH8Xj#I1I)m%_e$&cQsV7b_-1nSdP#>=%?~&KJV;%j!4ZR zxCC?T=>g`dnQ;nyGEux6xLz($f{L9#h_2;Hd@r-HsDJd5kNT$QtOnEu8@bt@0$%WM z0t5g<>X^jsMCfIyxWz%s(AX!*n1h5xz0lF3kK64F#;Ox~0X{&zyj*fQM<>2gZx0It z$XA-r>8{$n*)QD3%k_l)FTZ8%VSDN>Q|>eMvi0fVq8=cdGA$SDeM#eG z?Q!$e7eWzM7J@=#DcrQyfJQtmUf=!{qqu$vGx*5O9##@k6MRH;69cZj48l;T$T<1+ z{U1cu$@PPzUdRQFSm1IOseX>4f?_KTrUk{RB;~R~S*#YVV$LQx%@NDDj_`(<9ak~O z22xxA`I04kY#QR~+_CHI#-yJ!KU0d0DI*4kMpAthvo{Cill_f_?HT+icGP@B%qrLK zj6xVnUEV#2S0QrHAci7nA6!4Hn(I_>E^s6eb`)mhq+sbS5A`wgFNfE?(ew3{#X2Lt zAwfNXC{GZ{SQ2i#Q^}GmiuNUZ^zCb(G!xHel;gm~-8s_P${jx-L{DZz!y?gs*%2gM zMq~Vf7LGzU$}EqOI>DP|Onpl$P<3?t>$d;ru|sr4E};`8>z|WL;EDw&2a%irWyUX% zt(@pkA(T01pxzEZenF%sS(%|DR??uvi1w$A`aP%Ov7+ul6Sw;QTrNA~GJLLk=HQP4 z_&uK_K+#GC+lyrJ2tn6Lw1E^D*L~n7BggN9KeKWn-oI(`592VAK)^RYtsFrW_oHNH z9!6NU)bInULAt=KZu)b3^T(WK4lR9E0#=&G$h|wSC!=+GgJy=+#6z&#Yj3Dtq`enC zg6yZc;yY|WTb_QBk|f05bFilw`d6otzlcAPEPD9pk*Q#-O+Y_W!mByZ43V7AAWs*s z+4SE8Ry)to=^_so9KIZ-(qE0B+pPAe1n8NV5*nN~wJX0X69p}_?N|#K8p4oMZU(8c zAx47E%NrnSa!14y^llz*ag>FM7u)YUw`Qn;GRbt zoXj@q0SFl0EdquYou zBBkqX!2|{qScC~pJBG(g|MHs0=#P-VJva_+6c8IRJ(hRqF03(NA;iZ6Vg2dkiPGIi zEcegcDAKIvIKBfBjs_piSs+-dyBU7f!9Kb(DVbezOwnlrs|djQPUx#vzqWxtDNIr9 z5$I$9TqhohJY_GD>X0wJp_J~iKZPJN$IBfgkC>esy;s48q5!=I8sRXO%PHO><`Z<$3p}NgA$JL#hl^>M8e*Hj9^9-o_^(}N$YVB))x0# zGHmcTF=eR67yHhushaGW!a_0M=vGI4GvIn+&X}~U5M&RLrgkPG3!pED1CB;XU%2WB=I8xRfNrp*Za<@98R&i3X42b z%w1O2C?@z0f*UGie2#qTDW1Xnp_W5VLMcL=wFGf+o2=(kKr|U!0-{M15KY3r*L3Yr zA(4|B)Wjx_S8&~450_+BH3Gd=HUNsP7M9>K9J6hzQv-}c5Gh{(QH@W)(9pO0IcHE4 z#34yE3xEo;Kwey;ItY`VsLsg<%gPHI^F0iWg3-<2VYPcY0{O}t$URj)E2Mwu`T6Ac zGFziuq7q<>W4Qc@(vCnaJpw1NTk@r3gwVV+@#veJH;}WuByDn_9y^o)z8V&Jw*TBKPlLxQW!zmxL<((Gqqc9cQ3>~a3!_xvYPde@JsL6qQO z(+MRz`>;# z7HN1+Rs6(~d?BH7@+4@h%Jv#!xJW#lVjP$j#~8a3DKyuOAcDwy7MOK#kXN48`Pl6Y zl$lc0yvLhowtrVv{EK9t++l+(^aJ%BSU&@YG zJ!K`FC<2c~a}5n%^#aomlt9V9g71-W3LqG=f(%x$MR%)b0q7?oso%PXg2zw^3O?X@ zNq`g^5iD~?_lpss>|lS2@@hbv*iMNTKNw~fy_R-#%$N9!oO%lr8}~Nx2vo{y(EJ$K zgLeR-V~hx;O>DB7V~93;an9T|C)muW4H}xPmr{T&N)c>K0TY@U{AD>n3Zn`dMI3C) z8wcMNA|ot!PbNACql`_RHh2BQl;vAkU%CgMULY%El}s3v$AZ@TJM{#(!z5PD&qv>a z+`6<-uVKL>`mx;yA_vq0;m$~?&fi-le(0XwEEP1{%yMl5DHc{fm&&!TVY>8j=Z+sE zZs(8uMDaYqMZhLvH2~CdJ^oit@C3>pr-z3qcGi$%%&o)sCP(y!oU5Ln%VHEJG@g=T zO$N4yf}JUZl{a!Qm`y&+7^Ma#a$knyk#U+9Ii`bGV*|Wj*HOSwhM*3qntuu=hSCe9 z!1iE8;XAd2=Q^fzcC6nG&!;tI+u`9E3iVAQomhAS;fW}DXhrpnR6IhpI}$Nwp!vAwwlsPu64)s2)m=k5KRW9t#A0a0Uu;56Cyz%({$^%7AsP zccPf8*kUV&l0vYtX;;evUSi1r7aiE94nNFV5Y$dZ{o4`r$+n%g0y>b>vw;IHes z{l82+41S>p=R-`%XvzO@ds6Bk_CV1tfZVdf1I6h%i%28QIjx zF+@5%gKhino#AsYK)6IAFWd0xW*mjlSA%U|2B(mE!q80(-yG3S!A2zJ55aWuGt_Y_ zSh(D=J{L1C>2&ua&reQhl)F}*1!oT|oVI1YgO2wR!&9Ww7Vl5deJx6q)Z8~e1P(Ls zJldO)_MW%SZ?7{_oRdo&;BG6%tBH&yBuj1q-1Hu^Km98rnZkZN8T* z4-fSgB1w`8231*f*r{bS0sOs(!X52#j5ci_MG z)Nd1XUrv!*N2|X(wG@=8$X!*?jF+rtt3ipJdE%NAMG}Rfc7#JehAH|)cqj>?%8zCq zls#OWM72_yQwr$NFM1Y~^{I4F5ro#ot7yOAfDf5S<@_^}q90uX2e%|0=I3~tD6rpX zK(Z~~gH)~3pseEB*8C^$rBa(qH{=~Hg`bK=2d9suqZJe`LR3vOPi8g9;kLFzBD@QE zr1(I>4*~*MPVjymXeRXQ20G?YeH+TUjabcOkaaMTYp}BvE0LKSdg63`G(2B$Ev3EtTLcqql%PxrGaJpbH zBT9B=esaWKIZ}Ha6ItBQoSVC_5L;AOP{1v{f9Q!OhX99w%rlKpK&DRLv?dSSiXQ#OCyn=k^x@1b2#u6t zoJIY0dGAIAlFdu_sF-8L)M$^~(}D9c@iMs_*cV3{pOfaAB8lcXjPbA|e75`-x})B6 z&R*H5q772#crz4rl6Yr;I-&JSs#2`~eQau!5d^KwP>E4CQJQxqi5qHD;#H4UljdS~ z-%u@`oSNGFcvdFGzth9?U49r6nx{Urp~U-jlVtlRluI|B5Owi*aiXL5?W_Z5JMWEI z?!6vJEp3HpOVNU5|RBv(J1DAO|68BJtzx0nu++3k80Op$UtW%eu(R-^J%+` z!Te3xyGVZtGerV>)wA(p!8HOnFq)Fq(&OmIpVZw_c)VQ~m>r2~NL(x7^c4hNra#)^ zyokamNwy;BzKxK^6V1RGFYZ)-5OOmf-7V6ePxz}_-`1c`n;DX|OVqFMLBbq$bFEI0 zoE^>z0sZh#RYcSsj``qbAsS^Q(udLd@?cfO#;AInjQwPB7f*{kZDjf^1vh;y0vjA0 zu)QR8)&7yz`ybmDiG-097Eg&3)id4e!N1^0PlZk}gONBO8@~yNs9V<~_%^0FPz(CJ zEHnt-T}zCMfh^;f^ZJ|8kZqvT8kSTY;lQ zr<&Pl=Rlj~6=VI8HrWm6)gEj5ZpGsOll`ilJ^*&n4@1XSqo)5)5BiXonUHSfC7{xQ zK?lZ)_Z98cK?epM7<6F#Z11I6hQ5{1w~|%8GdeKnz@P);H@{=(J_g;#pzWk9ph5=* z9T;?A(49|ufQ#ls(8Gw!0;u%J=hCZz4vhZ~Fm6fvdCkpCz@41#jU+uhObe4OO~fjn z;4d==%~Q1yGoM}5f~m84B_*muBO|7bjKc`ZIB9Cv>tjQ2a_y^7B03jGdnz`9Gml$j z?Qo9>x$P0Ob-#+#rLT}1pwlxGAM+|Ig3Br~X*$82;MxrV=>9TLUetH{B`d`xe5vs= zk79+d1rkNA?BY}L^;x99oi@wb@PiDDuljCb3BQC*D5$pa*h9-)PH+bTRJXmXA~cnj zlKR!abS06@u~Zu&w`p*B;K`QsuvpLgMaS5M@A%0UUC(7kL3IgpHlZ>oQoZ4?#C#t+ z-jV$jG958IGQt@U5Dc-H%N-k&vK^g0sF0n&xAlU7_1I20M!_rL3+3(fbVBc> z#4l%tCK@)!_*dS}hXOHhOqS1LU(+3ID5$(~oAa#kzyY%Jy3#;eQtSu2(S$7MfX>15 zOB>fZ??n{i+RZ1E3bF+?33@fhV?14oJV=<%vRCcAo?k$M?Ds0G*&-GE5)Ng@RAM^c ze;LcaFJ!V#m_%e3iP6JcNXKY8imgEi$i!iptIwLd*N|KMbV?-On(}HIr;}czt}_|c zPno@l>)Ez8DcVR9ti=DB_C54_Re-u+_N3XOJ%cG<;=TNP zLHCArMwqY@%a-k`uHLu7CCia>k(jY9)N^EK&A7GoVJw0OuAi^HB{9fhZZ6ICRO^7dzl0g>Nsc^R znm1!!Ab)PVWnx@WA;gx+hIYj@tqS>M(AqVDbN^bV7Zw20ki|}|sZwiF@-1lXi189# z)9iPMfxNjEcJlC=7GT2;Q9mVtJ4}ly<7Mj8Y_`l+ZNi}PU^uDB$#_jlsvzh8(5hp3 VHTqHN?jPXqpn>WBoPDs+e*wdtYAFB! From dc2874ad537e71280a4a21828126c148c7b588a7 Mon Sep 17 00:00:00 2001 From: Peeeaje <74146834+Peeeaje@users.noreply.github.com> Date: Sat, 3 Jul 2021 15:26:19 +0900 Subject: [PATCH 006/368] translated 1-introduction README.md --- 1-Introduction/README.ja.md | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) create mode 100644 1-Introduction/README.ja.md diff --git a/1-Introduction/README.ja.md b/1-Introduction/README.ja.md new file mode 100644 index 0000000000..1113d19f2a --- /dev/null +++ b/1-Introduction/README.ja.md @@ -0,0 +1,22 @@ +# 機械学習ã¸ã®å°Žå…¥ + +ã“ã®ã‚»ã‚¯ã‚·ãƒ§ãƒ³ã§ã¯ã€æ©Ÿæ¢°å­¦ç¿’ã®åˆ†é‡Žã®åŸºç¤Žã¨ãªã‚‹æ¦‚念ã€æ©Ÿæ¢°å­¦ç¿’ã¨ã¯ä½•ã‹ã‚’紹介ã—ã€ãã®æ­´å²ã‚„研究者ãŒæ©Ÿæ¢°å­¦ç¿’を扱ã†éš›ã«ä½¿ç”¨ã™ã‚‹æŠ€è¡“ã«ã¤ã„ã¦å­¦ã³ã¾ã™ã€‚ æ–°ã—ã„MLã®ä¸–界を一緒ã«æŽ¢æ±‚ã—ã¦ã„ãã¾ã—ょã†ï¼ + +![globe](images/globe.jpg) +> Photo by Bill Oxford on Unsplash + +### Lessons + +1. [Introduction to machine learning](1-intro-to-ML/README.md) +1. [The History of machine learning and AI](2-history-of-ML/README.md) +1. [Fairness and machine learning](3-fairness/README.md) +1. [Techniques of machine learning](4-techniques-of-ML/README.md) +### Credits + +"Introduction to Machine Learning "ã¯ã€[Muhammad Sakib Khan Inan](https://twitter.com/Sakibinan)ã€[Ornella Altunyan](https://twitter.com/ornelladotcom)ã€[Jen Looper](https://twitter.com/jenlooper)ãªã©ã®ãƒãƒ¼ãƒ ã«ã‚ˆã£ã¦åˆ¶ä½œã•ã‚Œã¾ã—ãŸã€‚ + +"The History of Machine Learning" ã¯[Jen Looper](https://twitter.com/jenlooper)ã€[Amy Boyd](https://twitter.com/AmyKateNicho)ã«ã‚ˆã£ã¦åˆ¶ä½œã•ã‚Œã¾ã—ãŸã€‚ + +"Fairness and Machine Learning"ã¯[Tomomi Imura](https://twitter.com/girliemac) ã«ã‚ˆã£ã¦åˆ¶ä½œã•ã‚Œã¾ã—ãŸã€‚ + +"Techniques of Machine Learning"ã¯[Jen Looper](https://twitter.com/jenlooper)ã¨[Chris Noring](https://twitter.com/softchris) ã«ã‚ˆã£ã¦åˆ¶ä½œã•ã‚Œã¾ã—ãŸã€‚ \ No newline at end of file From 99fda249e52fa0ae5e8a1ca07f3baad9f4c8936d Mon Sep 17 00:00:00 2001 From: Peeeaje <74146834+Peeeaje@users.noreply.github.com> Date: Sat, 3 Jul 2021 18:07:11 +0900 Subject: [PATCH 007/368] translated 1.2 history of ML README.md --- .../2-history-of-ML/translations/README.ja.md | 114 ++++++++++++++++++ 1 file changed, 114 insertions(+) create mode 100644 1-Introduction/2-history-of-ML/translations/README.ja.md diff --git a/1-Introduction/2-history-of-ML/translations/README.ja.md b/1-Introduction/2-history-of-ML/translations/README.ja.md new file mode 100644 index 0000000000..437f156fa1 --- /dev/null +++ b/1-Introduction/2-history-of-ML/translations/README.ja.md @@ -0,0 +1,114 @@ +# History of machine learning + +![Summary of History of machine learning in a sketchnote](../../../sketchnotes/ml-history.png) +> Sketchnote by [Tomomi Imura](https://www.twitter.com/girlie_mac) + +## [Pre-lecture quiz](https://jolly-sea-0a877260f.azurestaticapps.net/quiz/3/) + +ã“ã®æŽˆæ¥­ã§ã¯ã€æ©Ÿæ¢°å­¦ç¿’ã¨äººå·¥çŸ¥èƒ½ã®æ­´å²ã«ãŠã‘る主è¦ãªå‡ºæ¥äº‹ã‚’紹介ã—ã¾ã™ã€‚ + +人工知能(AI)ã®æ­´å²ã¯ã€æ©Ÿæ¢°å­¦ç¿’ã®æ­´å²ã¨å¯†æŽ¥ã«é–¢ä¿‚ã—ã¦ã„ã¾ã™ã€‚ãªãœãªã‚‰ã°ã€æ©Ÿæ¢°å­¦ç¿’を支ãˆã‚‹ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã¨è¨ˆç®—ã®é€²æ­©ãŒã€AIã®ç™ºå±•ã«ã¤ãªãŒã£ãŸã‹ã‚‰ã§ã™ã€‚ã“れらã®åˆ†é‡Žã¯ã€1950年代ã«æ˜Žç¢ºã«ãªã‚Šå§‹ã‚ã¾ã—ãŸãŒã€é‡è¦ãª[アルゴリズムã€çµ±è¨ˆã€æ•°å­¦ã€è¨ˆç®—ã€æŠ€è¡“çš„ãªç™ºè¦‹](https://wikipedia.org/wiki/Timeline_of_machine_learning)ã¯ã€ã“ã®æ™‚代よりもå‰ã«ã€ãã—ã¦åŒæ™‚ã«è¡Œã‚ã‚Œã¦ã„ãŸã“ã¨ã‚’覚ãˆã¦ãŠãã¨ã‚ˆã„ã§ã—ょã†ã€‚実際ã€äººã€…ã¯[何百年も](https://wikipedia.org/wiki/History_of_artificial_intelligence)ã“ã®å•é¡Œã«ã¤ã„ã¦è€ƒãˆã¦ãã¾ã—ãŸã€‚(ã“ã®è¨˜äº‹ã§ã¯ã€ã€Œè€ƒãˆã‚‹æ©Ÿæ¢°ã€ã¨ã„ã†ã‚¢ã‚¤ãƒ‡ã‚¢ã®æ­´å²çš„ãªçŸ¥çš„基盤ã«ã¤ã„ã¦èª¬æ˜Žã•ã‚Œã¦ã„ã¾ã™ã€‚) + + +## 注目ã™ã¹ã発見 +- 1763å¹´ã€1812å¹´ [ベイズã®å®šç†](https://wikipedia.org/wiki/Bayes%27_theorem)ã¨ãã®å‰èº«ã®ç™ºè¦‹ã€‚ã‚る事象ãŒèµ·ã“る確率をã€äº‹å‰ã®çŸ¥è­˜ã«åŸºã¥ã„ã¦è¨˜è¿°ã™ã‚‹æŽ¨è«–ã®åŸºç¤Žã¨ãªã‚‹å®šç†ã¨ãã®å¿œç”¨ã€‚ +- 1805å¹´ フランスã®æ•°å­¦è€…アドリアンï¼ãƒžãƒªãƒ¼ãƒ»ãƒ¬ã‚¸ã‚§ãƒ³ãƒ‰ãƒ«ã«ã‚ˆã‚‹[最å°äºŒä¹—ç†è«–](https://wikipedia.org/wiki/Least_squares)。ã“ã®ç†è«–ã¯ã€ãƒ‡ãƒ¼ã‚¿ã®ãƒ•ã‚£ãƒƒãƒ†ã‚£ãƒ³ã‚°ã«å½¹ç«‹ã¤ã€‚ +- 1913å¹´ ロシアã®æ•°å­¦è€…アンドレイ・マルコフã«ã¡ãªã‚“ã§å付ã‘られãŸ[マルコフ連鎖](https://wikipedia.org/wiki/Markov_chain)ã¯ã€ä»¥å‰ã®çŠ¶æ…‹ã«åŸºã¥ã„ã¦èµ·ã“ã‚Šã†ã‚‹ä¸€é€£ã®äº‹è±¡ã‚’記述ã™ã‚‹ãŸã‚ã«ä½¿ç”¨ã•ã‚Œã‚‹ã€‚ +- 1957å¹´ [パーセプトロン](https://wikipedia.org/wiki/Perceptron)ã¯ã€ã‚¢ãƒ¡ãƒªã‚«ã®å¿ƒç†å­¦è€…フランク・ローゼンブラットãŒç™ºæ˜Žã—ãŸç·šå½¢åˆ†é¡žå™¨ã®ä¸€ç¨®ã§ã‚ã‚Šã€æ·±å±¤å­¦ç¿’ã®åŸºç›¤ã¨ãªã£ã¦ã„る。 +- 1967 [最å°è¿‘å‚法](https://wikipedia.org/wiki/Nearest_neighbor)ã¯ã€å…ƒã€…ã¯çµŒè·¯æŽ¢ç´¢ã®ãŸã‚ã«è€ƒæ¡ˆã•ã‚ŒãŸã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã€‚MLã§ã¯ãƒ‘ターンã®æ¤œå‡ºã«ç”¨ã„られる。 +- 1970å¹´ [ãƒãƒƒã‚¯ãƒ—ロパゲーション](https://wikipedia.org/wiki/Backpropagation)を用ã„ã¦[フィードフォワード・ニューラルãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ï¼ˆé †ä¼æ’­åž‹ãƒ‹ãƒ¥ãƒ¼ãƒ©ãƒ«ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ï¼‰](https://wikipedia.org/wiki/Feedforward_neural_network)を学習ã™ã‚‹ã€‚ +- 1982å¹´ [回帰型ニューラルãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯](https://wikipedia.org/wiki/Recurrent_neural_network) ã¯ã€ãƒ•ã‚£ãƒ¼ãƒ‰ãƒ•ã‚©ãƒ¯ãƒ¼ãƒ‰ãƒ»ãƒ‹ãƒ¥ãƒ¼ãƒ©ãƒ«ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã‹ã‚‰æ´¾ç”Ÿã—ãŸäººå·¥çš„ãªãƒ‹ãƒ¥ãƒ¼ãƒ©ãƒ«ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã§ã€æ™‚é–“çš„ãªã‚°ãƒ©ãƒ•ã‚’作æˆã—ã¾ã™ã€‚ + +✅ å°‘ã—調ã¹ã¦ã¿ã¦ãã ã•ã„。MLã¨AIã®æ­´å²ã®ä¸­ã§é‡è¦ãªæ—¥ä»˜ã¯ä»–ã«ã‚ã‚Šã¾ã™ã‹ï¼Ÿ + +## 1950: æ€è€ƒã™ã‚‹æ©Ÿæ¢° +アラン・ãƒãƒ¥ãƒ¼ãƒªãƒ³ã‚°ã¯ã€[2019å¹´ã«ä¸–é–“ã‹ã‚‰](https://wikipedia.org/wiki/Icons:_The_Greatest_Person_of_the_20th_Century)20世紀最大ã®ç§‘学者ã¨ã—ã¦æŠ•ç¥¨ã•ã‚ŒãŸã€çœŸã«å„ªã‚ŒãŸäººç‰©ã§ã€ã€Œè€ƒãˆã‚‹ã“ã¨ãŒã§ãる機械ã€ã¨ã„ã†æ¦‚念ã®åŸºç¤Žã‚’築ãã®ã«è²¢çŒ®ã—ãŸã¨ã•ã‚Œã¦ã„ã¾ã™ã€‚å½¼ã¯ã€å¦å®šçš„ãªæ„見やã€ã“ã®æ¦‚念ã®å®Ÿè¨¼çš„ãªè¨¼æ‹ ã‚’å¿…è¦ã¨ã™ã‚‹è‡ªåˆ†è‡ªèº«ã¨ã€ã“ã®å…ˆè‡ªç„¶è¨€èªžå‡¦ç†ã®æŽˆæ¥­ã§è§¦ã‚Œã‚‹ã“ã¨ã¨ãªã‚‹[ãƒãƒ¥ãƒ¼ãƒªãƒ³ã‚°ãƒ»ãƒ†ã‚¹ãƒˆ](https://www.bbc.com/news/technology-18475646)を作æˆã™ã‚‹ã“ã¨ã§æˆ¦ã„ã¾ã—ãŸã€‚ + +## 1956: ダートマス・サマー・リサーãƒãƒ»ãƒ—ロジェクト +ダートマス・サマー・リサーãƒãƒ»ãƒ—ロジェクトã¯ã€åˆ†é‡Žã¨ã—ã¦ã®äººå·¥çŸ¥èƒ½ã«ã¨ã£ã¦é‡è¦ãªå‡ºæ¥äº‹ã§ã‚ã‚Šã€ã“ã“ã§ã€Œäººå·¥çŸ¥èƒ½ã€ã¨ã„ã†è¨€è‘‰ãŒä½œã‚‰ã‚Œã¾ã—ãŸï¼ˆ[出典](https://250.dartmouth.edu/highlights/artificial-intelligence-ai-coined-dartmouth)) + +> 学習やãã®ä»–ã®çŸ¥èƒ½ã®ã‚らゆるå´é¢ã¯ã€åŽŸç†çš„ã«éžå¸¸ã«æ­£ç¢ºã«è¨˜è¿°ã™ã‚‹ã“ã¨ãŒã§ãã‚‹ã®ã§ã€ãれをシミュレートã™ã‚‹æ©Ÿæ¢°ã‚’作るã“ã¨ãŒã§ãる。 + +主任研究者ã§ã‚ã‚‹æ•°å­¦ã®ã‚¸ãƒ§ãƒ³ãƒ»ãƒžãƒƒã‚«ãƒ¼ã‚·ãƒ¼æ•™æŽˆã¯ã€ã€Œå­¦ç¿’ã®ã‚らゆるå´é¢ã‚„知能ã®ãã®ä»–ã®ç‰¹å¾´ã¯ã€åŽŸç†çš„ã«éžå¸¸ã«æ­£ç¢ºã«è¨˜è¿°ã™ã‚‹ã“ã¨ãŒã§ãã‚‹ã®ã§ã€ãれをシミュレートã™ã‚‹æ©Ÿæ¢°ã‚’作るã“ã¨ãŒã§ãã‚‹ã¨ã„ã†æŽ¨æ¸¬ã«åŸºã¥ã„ã¦é€²ã‚ã¦ã„ããŸã„ã€ã¨è€ƒãˆã¦ã„ã¾ã—ãŸã€‚å‚加者ã®ä¸­ã«ã¯ã€ã“ã®åˆ†é‡Žã®è‘—å人ã§ã‚るマービン・ミンスキーもã„ã¾ã—ãŸã€‚ + +ã“ã®ãƒ¯ãƒ¼ã‚¯ã‚·ãƒ§ãƒƒãƒ—ã§ã¯ã€ã€Œè¨˜å·çš„手法ã®å°é ­ã€é™å®šã•ã‚ŒãŸé ˜åŸŸã«ç„¦ç‚¹ã‚’当ã¦ãŸã‚·ã‚¹ãƒ†ãƒ ï¼ˆåˆæœŸã®ã‚¨ã‚­ã‚¹ãƒ‘ートシステム)ã€æ¼”繹的システムã¨å¸°ç´çš„システムã®æ¯”較ã€ãªã©ã®è­°è«–ãŒé–‹å§‹ã•ã‚Œã€ä¿ƒé€²ã•ã‚ŒãŸã¨è©•ä¾¡ã•ã‚Œã¦ã„ã¾ã™ã€‚([出典](https://wikipedia.org/wiki/Dartmouth_workshop)) + +## 1956 - 1974: 黄金期 + +1950年代ã‹ã‚‰70年代åŠã°ã¾ã§ã¯ã€AIãŒã•ã¾ã–ã¾ãªå•é¡Œã‚’解決ã—ã¦ãれるã®ã§ã¯ãªã„ã‹ã¨ã„ã†æ¥½è¦³çš„ãªè¦‹æ–¹ãŒåºƒãŒã£ã¦ã„ã¾ã—ãŸã€‚1967å¹´ã€ãƒžãƒ¼ãƒ“ン・ミンスキーã¯ã€Œä¸€ä¸–代ã®ã†ã¡ã«...『人工知能ã€ã‚’作るã¨ã„ã†å•é¡Œã¯å®Ÿè³ªçš„ã«è§£æ±ºã•ã‚Œã‚‹ã ã‚ã†ã€ã¨è‡ªä¿¡ã‚’æŒã£ã¦è¿°ã¹ã¦ã„る。(Minsky, Marvin (1967), Computation: Finite and Infinite Machines, Englewood Cliffs, N.J.: Prentice-Hall) + +自然言語処ç†ã®ç ”究ãŒç››ã‚“ã«ãªã‚Šã€æ¤œç´¢ãŒæ´—ç·´ã•ã‚Œã¦ã‚ˆã‚Šå¼·åŠ›ã«ãªã‚Šã€å¹³æ˜“ãªè¨€èªžã«ã‚ˆã‚‹æŒ‡ç¤ºã§ç°¡å˜ãªä½œæ¥­ã‚’ã“ãªã™ã€Œãƒžã‚¤ã‚¯ãƒ­ãƒ¯ãƒ¼ãƒ«ãƒ‰ã€ã¨ã„ã†æ¦‚念ãŒç”Ÿã¾ã‚ŒãŸã€‚ + +研究ã¯æ”¿åºœæ©Ÿé–¢ã‹ã‚‰æ½¤æ²¢ãªè³‡é‡‘ãŒæä¾›ã•ã‚Œã€è¨ˆç®—ã¨ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ãŒé€²æ­©ã—ã€çŸ¥çš„機械ã®ãƒ—ロトタイプãŒä½œã‚‰ã‚ŒãŸã€‚ãã®ä¸­ã«ã¯æ¬¡ã®ã‚ˆã†ãªã‚‚ã®ãŒã‚る。 + +* 移動ã—ãŸã‚Šã€ã‚¿ã‚¹ã‚¯ã‚’実行ã™ã‚‹æ–¹æ³•ã‚’「知的ã«ã€æ±ºå®šã™ã‚‹ã“ã¨ãŒã§ãるロボット[「Shakeyã€](https://wikipedia.org/wiki/Shakey_the_robot) + + ![Shakey, an intelligent robot](../images/shakey.jpg) + > Shakey in 1972 + +* åˆæœŸã®ã€ŒãŠã—ゃã¹ã‚Šãƒ­ãƒœãƒƒãƒˆã€ã§ã‚ã‚‹Elizaã¯ã€äººã¨ä¼šè©±ã™ã‚‹ã“ã¨ãŒã§ãã€åŽŸå§‹çš„ãªã€Œã‚»ãƒ©ãƒ”ストã€ã®å½¹å‰²ã‚’æžœãŸã—ãŸã€‚エリザã«ã¤ã„ã¦ã¯ã€NLPã®ãƒ¬ãƒƒã‚¹ãƒ³ã§è©³ã—ã説明ã—ã¾ã™ã€‚ + + ![Eliza, a bot](../images/eliza.png) + > A version of Eliza, a chatbot + +* 「Blocks worldã€ã¯ã€ãƒ–ロックをç©ã¿ä¸Šã’ãŸã‚Šä¸¦ã¹æ›¿ãˆãŸã‚Šã™ã‚‹ãƒžã‚¤ã‚¯ãƒ­ãƒ¯ãƒ¼ãƒ«ãƒ‰ã®ä¸€ä¾‹ã§ã€æ©Ÿæ¢°ã«åˆ¤æ–­åŠ›ã‚’身ã«ã¤ã‘ã•ã›ã‚‹å®Ÿé¨“ã‚’è¡Œã£ãŸã€‚[SHRDLU](https://wikipedia.org/wiki/SHRDLU)ã‚’ã¯ã˜ã‚ã¨ã™ã‚‹ãƒ©ã‚¤ãƒ–ラリã®é€²æ­©ã¯ã€è¨€èªžå‡¦ç†ã®ç™ºå±•ã«å¤§ãã貢献ã—ãŸã€‚ + + [![blocks world with SHRDLU](https://img.youtube.com/vi/QAJz4YKUwqw/0.jpg)](https://www.youtube.com/watch?v=QAJz4YKUwqw "blocks world with SHRDLU") + + > 🎥 上ã®ç”»åƒã‚’クリックã™ã‚‹ã¨å‹•ç”»ãŒè¦‹ã‚‰ã‚Œã¾ã™ï¼š"Blocks world with SHRDLU" + +## 1974 - 1980: AIã®å†¬ + +1970年代åŠã°ã«ãªã‚‹ã¨ã€ã€ŒçŸ¥çš„ãªæ©Ÿæ¢°ã€ã‚’作るã“ã¨ã®è¤‡é›‘ã•ãŒéŽå°è©•ä¾¡ã•ã‚Œã¦ã„ãŸã“ã¨ã‚„ã€åˆ©ç”¨å¯èƒ½ãªè¨ˆç®—能力を考慮ã™ã‚‹ã¨ã€ãã®å°†æ¥æ€§ãŒéŽå¤§è©•ä¾¡ã•ã‚Œã¦ã„ãŸã“ã¨ãŒæ˜Žã‚‰ã‹ã«ãªã‚Šã¾ã—ãŸã€‚資金ãŒæž¯æ¸‡ã—ã€ã“ã®åˆ†é‡Žã¸ã®ä¿¡é ¼ãŒä½Žä¸‹ã—ãŸã€‚信頼性ã«å½±éŸ¿ã‚’与ãˆãŸå•é¡Œã«ã¯ä»¥ä¸‹ã®ã‚ˆã†ãªã‚‚ã®ãŒã‚る。: + +- **é™ç•Œ**. 計算能力ã®é™ç•Œ +- **組ã¿åˆã‚ã›ã®çˆ†ç™º**. 学習ã«å¿…è¦ãªãƒ‘ラメータã®é‡ã¯ã€ã‚³ãƒ³ãƒ”ュータã«è¦æ±‚ã•ã‚Œã‚‹ã“ã¨ãŒå¤šããªã‚‹ã«ã¤ã‚Œã¦æŒ‡æ•°é–¢æ•°çš„ã«å¢—加ã—ã¾ã—ãŸãŒã€ã‚³ãƒ³ãƒ”ュータã®æ€§èƒ½ã‚„能力ã¯ä¸¦è¡Œã—ã¦é€²åŒ–ã—ã¾ã›ã‚“ã§ã—ãŸã€‚ +- **データã®å°‘ãªã•**. データãŒä¸è¶³ã—ã¦ã„ãŸãŸã‚ã€ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã®ãƒ†ã‚¹ãƒˆã€é–‹ç™ºã€æ”¹è‰¯ã®ãƒ—ロセスãŒå¦¨ã’られãŸã€‚ +- **æ­£ã—ã„質å•ã‚’ã—ã¦ã„ã‚‹ã®ã‹ã©ã†ã‹**. å•ã„ã‹ã‘ã¦ã„ãŸè³ªå•ãã®ã‚‚ã®ãŒç–‘å•è¦–ã•ã‚Œå§‹ã‚ãŸã€‚研究者ãŸã¡ã¯ã€è‡ªåˆ†ãŸã¡ã®ã‚¢ãƒ—ローãƒã«æ‰¹åˆ¤çš„ãªæ„見をæŒã¤ã‚ˆã†ã«ãªã£ãŸã€‚ + - ãƒãƒ¥ãƒ¼ãƒªãƒ³ã‚°ãƒ†ã‚¹ãƒˆã¯ã€ã€Œã‚³ãƒ³ãƒ”ュータをプログラミングã™ã‚‹ã“ã¨ã§ã€è¨€èªžã‚’ç†è§£ã—ã¦ã„るよã†ã«è¦‹ã›ã‹ã‘ã‚‹ã“ã¨ã¯ã§ãã‚‹ãŒã€æœ¬å½“ã®æ„味ã§ã®ç†è§£ã¯ã§ããªã„ã€ã¨ã™ã‚‹ã€Œãƒãƒ£ã‚¤ãƒ‹ãƒ¼ã‚ºãƒ«ãƒ¼ãƒ ç†è«–ã€ãªã©ã«ã‚ˆã£ã¦ã€ç–‘å•è¦–ã•ã‚Œã‚‹ã‚ˆã†ã«ãªã£ãŸã€‚([出典](https://plato.stanford.edu/entries/chinese-room/)) + - セラピストã¨ã—ã¦ELIZAã®ã‚ˆã†ãªäººå·¥çŸ¥èƒ½ã‚’社会ã«å°Žå…¥ã™ã‚‹ã“ã¨ã®å€«ç†æ€§ãŒå•ã‚ã‚ŒãŸã€‚ +ãã‚Œã¨åŒæ™‚ã«ã€ã•ã¾ã–ã¾ãªAIã®æµæ´¾ãŒå½¢æˆã•ã‚Œå§‹ã‚ã¾ã—ãŸã€‚一ã¤ã¯ã€["Scruffy"㨠"Neat AI"](https://wikipedia.org/wiki/Neats_and_scruffies)ã¨ã„ã†äºŒåˆ†æ³•ã§ã‚る。Scruffyãªç ”究室ã§ã¯ã€ç›®çš„ã®çµæžœãŒå¾—られるã¾ã§ä½•æ™‚間もプログラムをã„ã˜ã£ã¦ã„ãŸä¸€æ–¹ã€neatãªç ”究室ã§ã¯ã€è«–ç†ã¨å½¢å¼çš„ãªå•é¡Œè§£æ±ºã‚’é‡è¦–ã™ã‚‹ã€‚ELIZAã‚„SHRDLUãªã©ãŒæœ‰åãªScruffyã§ã‚るシステムã§ã‚る。1980年代ã«å…¥ã£ã¦ã€MLシステムã®å†ç¾æ€§ãŒæ±‚ã‚られるよã†ã«ãªã‚‹ã¨ã€çµæžœãŒèª¬æ˜Žå¯èƒ½ã§ã‚ã‚‹ã“ã¨ã‹ã‚‰ã€æ¬¡ç¬¬ã«neatãªã‚¢ãƒ—ローãƒãŒä¸»æµã«ãªã£ã¦ã„ãã¾ã—ãŸã€‚ + +## 1980s エキスパートシステム + +分野ãŒç™ºå±•ã™ã‚‹ã«ã¤ã‚Œã€ãƒ“ジãƒã‚¹ã¸ã®è²¢çŒ®ãŒæ˜Žç¢ºã«ãªã‚Šã€1980年代ã«ã¯ã€Œã‚¨ã‚­ã‚¹ãƒ‘ートシステムã€ãŒæ™®åŠã—ã¾ã—ãŸã€‚「エキスパートシステムã¯ã€äººå·¥çŸ¥èƒ½ï¼ˆAI)ソフトウェアã®ä¸­ã§æœ€åˆã«çœŸã«æˆåŠŸã—ãŸå½¢æ…‹ã®ä¸€ã¤ã§ã‚る。ã€ã¨è¨€ã‚ã‚Œã¦ã„ã¾ã™ã€‚([出典](https://wikipedia.org/wiki/Expert_system)) + +ã“ã®ã‚¿ã‚¤ãƒ—ã®ã‚·ã‚¹ãƒ†ãƒ ã¯ã€ãƒ“ジãƒã‚¹è¦ä»¶ã‚’定義ã™ã‚‹ãƒ«ãƒ¼ãƒ«ã‚¨ãƒ³ã‚¸ãƒ³ã¨ã€ãƒ«ãƒ¼ãƒ«ã‚·ã‚¹ãƒ†ãƒ ã‚’活用ã—ã¦æ–°ãŸãªäº‹å®Ÿã‚’推論ã™ã‚‹æŽ¨è«–エンジンã§æ§‹æˆã•ã‚Œã‚‹ãƒã‚¤ãƒ–リッド型ã§ã™ã€‚ + +ã¾ãŸã€ã“ã®æ™‚代ã¯ãƒ‹ãƒ¥ãƒ¼ãƒ©ãƒ«ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã«ã‚‚注目ãŒé›†ã¾ã£ãŸã€‚ + +## 1987 - 1993: AIã®å†·ãˆè¾¼ã¿ + +専門分野ã«ç‰¹åŒ–ã—ãŸã‚¨ã‚­ã‚¹ãƒ‘ートシステムã®ãƒãƒ¼ãƒ‰ã‚¦ã‚§ã‚¢ãŒæ™®åŠã—ãŸã“ã¨ã§ã€å°‚門性ãŒé«˜ããªã‚Šã™ãŽã¦ã—ã¾ã†ã¨ã„ã†æ®‹å¿µãªçµæžœã«ãªã‚Šã¾ã—ãŸã€‚ã¾ãŸã€ãƒ‘ーソナルコンピュータã®å°é ­ã¯ã€ã“れらã®å¤§è¦æ¨¡ã§å°‚門的ãªä¸­å¤®é›†æ¨©çš„システムã¨ç«¶åˆã—ãŸã€‚コンピューティングã®æ°‘主化ãŒå§‹ã¾ã‚Šã€æœ€çµ‚çš„ã«ã¯ç¾ä»£ã®çˆ†ç™ºçš„ãªãƒ“ッグデータã¸ã®é“ãŒé–‹ã‹ã‚Œã¾ã—ãŸã€‚ + +## 1993 - 2011 + +ã“ã®æœŸé–“ã§ã¯ã€ãれ以å‰ã«ãƒ‡ãƒ¼ã‚¿ã¨è¨ˆç®—能力ã®ä¸è¶³ã«ã‚ˆã£ã¦å¼•ãèµ·ã“ã•ã‚Œã¦ã„ãŸå•é¡Œã‚’ã€MLã‚„AIãŒè§£æ±ºã§ãるよã†ã«ãªã£ã¦ã„ãŸã€‚特ã«2007å¹´é ƒã«ã‚¹ãƒžãƒ¼ãƒˆãƒ•ã‚©ãƒ³ãŒç™»å ´ã—ãŸã“ã¨ã§ã€è‰¯ãも悪ãもデータé‡ãŒæ€¥é€Ÿã«å¢—加ã—ã€åºƒã利用ã•ã‚Œã‚‹ã‚ˆã†ã«ãªã‚Šã¾ã—ãŸã€‚計算機ã®æ€§èƒ½ã‚‚飛èºçš„ã«å‘上ã—ã€ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã‚‚ãã‚Œã«åˆã‚ã›ã¦é€²åŒ–ã—ã¦ã„ãã¾ã—ãŸã€‚éŽåŽ»ã®è‡ªç”±å¥”放ãªæ™‚代ã‹ã‚‰ã€çœŸã®å­¦å•ã¨ã—ã¦ã®çµæ™¶åŒ–ãŒå§‹ã¾ã‚Šã€ã“ã®åˆ†é‡Žã¯æˆç†Ÿã—ã¦ã„ãã¾ã—ãŸã€‚ + +## ç¾åœ¨ + +ç¾åœ¨ã€æ©Ÿæ¢°å­¦ç¿’ã‚„AIã¯ã€ç§ãŸã¡ã®ç”Ÿæ´»ã®ã»ã¼ã™ã¹ã¦ã®éƒ¨åˆ†ã«é–¢ã‚ã£ã¦ã„ã¾ã™ã€‚ã“ã®ã‚ˆã†ãªæ™‚代ã«ã¯ã€ã“れらã®ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ãŒäººé–“ã®ç”Ÿæ´»ã«åŠã¼ã™ãƒªã‚¹ã‚¯ã‚„潜在的ãªå½±éŸ¿ã‚’注æ„æ·±ãç†è§£ã™ã‚‹ã“ã¨ãŒæ±‚ã‚られã¾ã™ã€‚マイクロソフトã®ãƒ–ラッド・スミスã¯ã€ã€Œæƒ…報技術ã¯ã€ãƒ—ライãƒã‚·ãƒ¼ã‚„表ç¾ã®è‡ªç”±ã¨ã„ã£ãŸåŸºæœ¬çš„ãªäººæ¨©ä¿è­·ã®æ ¸å¿ƒã«è¿«ã‚‹å•é¡Œã‚’æèµ·ã—ã¾ã™ã€‚情報技術ã¯ã€ãƒ—ライãƒã‚·ãƒ¼ã‚„表ç¾ã®è‡ªç”±ã¨ã„ã£ãŸåŸºæœ¬çš„ãªäººæ¨©ä¿è­·ã®æ ¹å¹¹ã«é–¢ã‚ã‚‹å•é¡Œã‚’æèµ·ã—ã¾ã™ã€‚我々ã®è¦‹è§£ã§ã¯ã€ã“れらã®å•é¡Œã¯ã€æ”¿åºœã«ã‚ˆã‚‹æ€æ…®æ·±ã„è¦åˆ¶ã¨ã€è¨±å®¹ã•ã‚Œã‚‹ä½¿ç”¨æ–¹æ³•ã«é–¢ã™ã‚‹è¦ç¯„ã®ç­–定を必è¦ã¨ã—ã¦ã„ã¾ã™ã€‚ã€ã¨è¿°ã¹ã¦ã„ã¾ã™ã€‚([出典](https://www.technologyreview.com/2019/12/18/102365/the-future-of-ais-impact-on-society/)) + +未æ¥ãŒã©ã†ãªã‚‹ã‹ã¯ã¾ã ã‚ã‹ã‚Šã¾ã›ã‚“ãŒã€ã“れらã®ã‚³ãƒ³ãƒ”ュータシステムã¨ã€ãれを動ã‹ã™ã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢ã‚„アルゴリズムをç†è§£ã™ã‚‹ã“ã¨ã¯é‡è¦ã§ã™ã€‚ã“ã®ã‚«ãƒªã‚­ãƒ¥ãƒ©ãƒ ãŒè‡ªèº«ã§åˆ¤æ–­ã™ã‚‹ã«ã‚ãŸã‚Šã€ã‚ˆã‚Šè‰¯ã„ç†è§£ã‚’助ã‘ã‚‹ã‚‚ã®ã«ãªã‚‹ã¨å¹¸ã„ã§ã™ã€‚ + +[![The history of deep learning](https://img.youtube.com/vi/mTtDfKgLm54/0.jpg)](https://www.youtube.com/watch?v=mTtDfKgLm54 "The history of deep learning") +> 🎥 上ã®ç”»åƒã‚’クリックã™ã‚‹ã¨å‹•ç”»ãŒè¦‹ã‚‰ã‚Œã¾ã™ï¼šYann LeCun discusses the history of deep learning in this lecture + +--- +## 🚀Challenge + +ã“れらã®æ­´å²çš„瞬間ã®1ã¤ã‚’掘り下ã’ã¦ã€ãã®èƒŒå¾Œã«ã„る人々ã«ã¤ã„ã¦å­¦ã³ã¾ã—ょã†ã€‚魅力的ãªäººã€…ãŒã„ã¾ã™ã—ã€æ–‡åŒ–çš„ã«ç©ºç™½ã®çŠ¶æ…‹ã§ç§‘学的発見ãŒãªã•ã‚ŒãŸã“ã¨ã¯ã‚ã‚Šã¾ã›ã‚“。ã©ã†ã„ã£ãŸã“ã¨ãŒè¦‹ã¤ã‹ã‚‹ã§ã—ょã†ã‹ï¼Ÿ + +## [Post-lecture quiz](https://jolly-sea-0a877260f.azurestaticapps.net/quiz/4/) + +## 振り返りã¨è‡ªç¿’ + +視è´ã™ã‚‹ã¹ãæ•™æã¯ä»¥ä¸‹ã«ãªã‚Šã¾ã™: + +[Amy BoydãŒAIã®é€²åŒ–ã«ã¤ã„ã¦è¿°ã¹ã¦ã„ã‚‹ãƒãƒƒãƒ‰ã‚­ãƒ£ã‚¹ãƒˆ](http://runasradio.com/Shows/Show/739) + +[![Amy Boydã«ã‚ˆã‚‹AIã®æ­´å²](https://img.youtube.com/vi/EJt3_bFYKss/0.jpg)](https://www.youtube.com/watch?v=EJt3_bFYKss "The history of AI by Amy Boyd") + +## 課題 + +[時系列を制作ã—ã¦ãã ã•ã„](../assignment.md) From b0e06d6927cc49865a2dff68aba147bb5a194417 Mon Sep 17 00:00:00 2001 From: Peeeaje <74146834+Peeeaje@users.noreply.github.com> Date: Sat, 3 Jul 2021 23:20:24 +0900 Subject: [PATCH 008/368] translated titles, alt tags also fixed some paths of images --- 1-Introduction/1-intro-to-ML/translations/README.ja.md | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/1-Introduction/1-intro-to-ML/translations/README.ja.md b/1-Introduction/1-intro-to-ML/translations/README.ja.md index 568a11e34c..aded0f7e65 100644 --- a/1-Introduction/1-intro-to-ML/translations/README.ja.md +++ b/1-Introduction/1-intro-to-ML/translations/README.ja.md @@ -1,6 +1,6 @@ -# Introduction to machine learning +# 機械学習ã¸ã®å°Žå…¥ -[![ML, AI, deep learning - What's the difference?](https://img.youtube.com/vi/lTd9RSxS9ZE/0.jpg)](https://youtu.be/lTd9RSxS9ZE "ML, AI, deep learning - What's the difference?") +[![ML, AI, deep learning - é•ã„ã¯ä½•ã‹ï¼Ÿ](https://img.youtube.com/vi/lTd9RSxS9ZE/0.jpg)](https://youtu.be/lTd9RSxS9ZE "ML, AI, deep learning - é•ã„ã¯ä½•ã‹ï¼Ÿ") > 🎥 上ã®ç”»åƒã‚’クリックã™ã‚‹ã¨ã€æ©Ÿæ¢°å­¦ç¿’ã€AIã€æ·±å±¤å­¦ç¿’ã®é•ã„ã«ã¤ã„ã¦èª¬æ˜Žã—ãŸå‹•ç”»ãŒè¡¨ç¤ºã•ã‚Œã¾ã™ã€‚ @@ -9,7 +9,7 @@ ### イントロダクション åˆå¿ƒè€…ã®ãŸã‚ã®å¤å…¸çš„ãªæ©Ÿæ¢°å­¦ç¿’ã®ã‚³ãƒ¼ã‚¹ã¸ã‚ˆã†ã“ã! ã“ã®ãƒ†ãƒ¼ãƒžã«å…¨ã触れãŸã“ã¨ã®ãªã„方もã€ã“ã®åˆ†é‡Žã‚’ブラッシュアップã—ãŸã„経験豊富ãªæ–¹ã‚‚ã€ãœã²ã”å‚加ãã ã•ã„。ç§ãŸã¡ã¯ã€ã‚ãªãŸã®MLã®å­¦ç¿’ã«ã¤ã„ã¦ã®è¦ªã—ã¿ã‚„ã™ã„スタート地点を作りãŸã„ã¨è€ƒãˆã¦ã„ã¾ã™ã€‚ã‚ãªãŸã®[フィードãƒãƒƒã‚¯](https://github.com/microsoft/ML-For-Beginners/discussions)を評価ã—ã€å¯¾å¿œã—ã€å–り入れるã“ã¨ãŒã§ãã‚Œã°å¹¸ã„ã§ã™ã€‚ -[![Introduction to ML](https://img.youtube.com/vi/h0e2HAPTGF4/0.jpg)](https://youtu.be/h0e2HAPTGF4 "Introduction to ML") +[![機械学習ã¸ã®å°Žå…¥](https://img.youtube.com/vi/h0e2HAPTGF4/0.jpg)](https://youtu.be/h0e2HAPTGF4 "機械学習ã¸ã®å°Žå…¥") > 🎥 上ã®ç”»åƒã‚’クリックã™ã‚‹ã¨ã€MITã®John GuttagãŒæ©Ÿæ¢°å­¦ç¿’を紹介ã™ã‚‹å‹•ç”»ãŒè¡¨ç¤ºã•ã‚Œã¾ã™ã€‚ ### 機械学習を始ã‚ã‚‹ã«ã‚ãŸã£ã¦ @@ -26,7 +26,7 @@ "機械学習(Machine Learning)"ã¨ã„ã†è¨€è‘‰ã¯ã€ç¾åœ¨æœ€ã‚‚人気ãŒã‚ã‚Šã€é »ç¹ã«ä½¿ç”¨ã•ã‚Œã¦ã„る言葉ã®ä¸€ã¤ã§ã™ã€‚ã©ã‚“ãªåˆ†é‡Žã®æŠ€è¡“者ã§ã‚ã£ã¦ã‚‚ã€å¤šå°‘ãªã‚Šã¨ã‚‚技術ã«ç²¾é€šã—ã¦ã„ã‚Œã°ã€ä¸€åº¦ã¯ã“ã®è¨€è‘‰ã‚’耳ã«ã—ãŸã“ã¨ãŒã‚ã‚‹å¯èƒ½æ€§ã¯å°‘ãªãã‚ã‚Šã¾ã›ã‚“。ã—ã‹ã—ã€æ©Ÿæ¢°å­¦ç¿’ã®ä»•çµ„ã¿ã¯ã€ã»ã¨ã‚“ã©ã®äººã«ã¨ã£ã¦è¬Žã«åŒ…ã¾ã‚Œã¦ãŠã‚Šã€æ©Ÿæ¢°å­¦ç¿’ã®åˆå¿ƒè€…ã«ã¨ã£ã¦ã€ã“ã®ãƒ†ãƒ¼ãƒžã¯æ™‚ã«åœ§å€’ã•ã‚Œã‚‹ã‚ˆã†ã«æ„Ÿã˜ã‚‰ã‚Œã¾ã™ã€‚ãã®ãŸã‚ã€æ©Ÿæ¢°å­¦ç¿’ã¨ã¯ä½•ã‹ã‚’実際ã«ç†è§£ã—ã€å®Ÿè·µçš„ãªä¾‹ã‚’通ã—ã¦æ®µéšŽçš„ã«å­¦ã‚“ã§ã„ãã“ã¨ãŒé‡è¦ã§ã™ã€‚ -![ml hype curve](images/hype.png) +![機械学習ã®äººæ°—を示ã™ã‚°ãƒ©ãƒ•](../images/hype.png) > Google Trendsã«ã‚ˆã‚‹ã€ã€Œæ©Ÿæ¢°å­¦ç¿’ã€ã¨ã„ã†è¨€è‘‰ã®æœ€è¿‘ã®ç››ã‚Šä¸ŠãŒã‚Šã‚’示ã™ã‚°ãƒ©ãƒ•ã€‚ @@ -38,7 +38,7 @@ ã“ã®è¨€è‘‰ã¯æ··åŒã•ã‚Œã‚‹ã“ã¨ãŒã‚ã‚Šã¾ã™ãŒã€æ©Ÿæ¢°å­¦ç¿’(ML)ã¯äººå·¥çŸ¥èƒ½ã®é‡è¦ãªã‚µãƒ–セットã§ã™ã€‚**MLã¯ã€ç‰¹æ®Šãªã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã‚’使用ã—ã¦ã€æ„味ã®ã‚る情報を発見ã—ã€çŸ¥è¦šã•ã‚ŒãŸãƒ‡ãƒ¼ã‚¿ã‹ã‚‰éš ã‚ŒãŸãƒ‘ターンを見ã¤ã‘ã¦ã€åˆç†çš„ãªæ„æ€æ±ºå®šãƒ—ロセスをè£ä»˜ã‘ã‚‹ã“ã¨ã«é–¢ä¿‚ã—ã¦ã„ã¾ã™ã€‚** -![AI, ML, deep learning, data science](images/ai-ml-ds.png) +![AI, ML, ディープラーニングã€ãƒ‡ãƒ¼ã‚¿ã‚µã‚¤ã‚¨ãƒ³ã‚¹](../images/ai-ml-ds.png) >[ã“ã®ã‚°ãƒ©ãƒ•](https://softwareengineering.stackexchange.com/questions/366996/distinction-between-ai-ml-neural-networks-deep-learning-and-data-mining)ã«è§¦ç™ºã•ã‚ŒãŸ[Jen Looper](https://twitter.com/jenlooper)æ°ã«ã‚ˆã‚‹ã‚¤ãƒ³ãƒ•ã‚©ã‚°ãƒ©ãƒ•ã‚£ãƒƒã‚¯ From cb1470b007cbc415c5bbe735f86f122bc9939d93 Mon Sep 17 00:00:00 2001 From: Peeeaje <74146834+Peeeaje@users.noreply.github.com> Date: Sat, 3 Jul 2021 23:29:40 +0900 Subject: [PATCH 009/368] translated title and alt tags for 1-introduction readme.md --- 1-Introduction/README.ja.md | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git a/1-Introduction/README.ja.md b/1-Introduction/README.ja.md index 1113d19f2a..7752ae312a 100644 --- a/1-Introduction/README.ja.md +++ b/1-Introduction/README.ja.md @@ -2,21 +2,21 @@ ã“ã®ã‚»ã‚¯ã‚·ãƒ§ãƒ³ã§ã¯ã€æ©Ÿæ¢°å­¦ç¿’ã®åˆ†é‡Žã®åŸºç¤Žã¨ãªã‚‹æ¦‚念ã€æ©Ÿæ¢°å­¦ç¿’ã¨ã¯ä½•ã‹ã‚’紹介ã—ã€ãã®æ­´å²ã‚„研究者ãŒæ©Ÿæ¢°å­¦ç¿’を扱ã†éš›ã«ä½¿ç”¨ã™ã‚‹æŠ€è¡“ã«ã¤ã„ã¦å­¦ã³ã¾ã™ã€‚ æ–°ã—ã„MLã®ä¸–界を一緒ã«æŽ¢æ±‚ã—ã¦ã„ãã¾ã—ょã†ï¼ -![globe](images/globe.jpg) -> Photo by Bill Oxford on Unsplash +![地çƒ](images/globe.jpg) +> Unsplashã®Bill Oxfordã«ã‚ˆã‚‹å†™çœŸ ### Lessons -1. [Introduction to machine learning](1-intro-to-ML/README.md) -1. [The History of machine learning and AI](2-history-of-ML/README.md) -1. [Fairness and machine learning](3-fairness/README.md) -1. [Techniques of machine learning](4-techniques-of-ML/README.md) +1. [機械学習ã¸ã®å°Žå…¥](1-intro-to-ML/README.md) +1. [機械学習ã¨AIã®æ­´å²](2-history-of-ML/README.md) +1. [公平性ã¨æ©Ÿæ¢°å­¦ç¿’](3-fairness/README.md) +1. [機械学習ã®æŠ€è¡“](4-techniques-of-ML/README.md) ### Credits -"Introduction to Machine Learning "ã¯ã€[Muhammad Sakib Khan Inan](https://twitter.com/Sakibinan)ã€[Ornella Altunyan](https://twitter.com/ornelladotcom)ã€[Jen Looper](https://twitter.com/jenlooper)ãªã©ã®ãƒãƒ¼ãƒ ã«ã‚ˆã£ã¦åˆ¶ä½œã•ã‚Œã¾ã—ãŸã€‚ +"機械学習ã¸ã®å°Žå…¥ "ã¯ã€[Muhammad Sakib Khan Inan](https://twitter.com/Sakibinan)ã€[Ornella Altunyan](https://twitter.com/ornelladotcom)ã€[Jen Looper](https://twitter.com/jenlooper)ãªã©ã®ãƒãƒ¼ãƒ ã«ã‚ˆã£ã¦åˆ¶ä½œã•ã‚Œã¾ã—ãŸã€‚ -"The History of Machine Learning" ã¯[Jen Looper](https://twitter.com/jenlooper)ã€[Amy Boyd](https://twitter.com/AmyKateNicho)ã«ã‚ˆã£ã¦åˆ¶ä½œã•ã‚Œã¾ã—ãŸã€‚ +"機械学習ã¨AIã®æ­´å²" ã¯[Jen Looper](https://twitter.com/jenlooper)ã€[Amy Boyd](https://twitter.com/AmyKateNicho)ã«ã‚ˆã£ã¦åˆ¶ä½œã•ã‚Œã¾ã—ãŸã€‚ -"Fairness and Machine Learning"ã¯[Tomomi Imura](https://twitter.com/girliemac) ã«ã‚ˆã£ã¦åˆ¶ä½œã•ã‚Œã¾ã—ãŸã€‚ +"公平性ã¨æ©Ÿæ¢°å­¦ç¿’"ã¯[Tomomi Imura](https://twitter.com/girliemac) ã«ã‚ˆã£ã¦åˆ¶ä½œã•ã‚Œã¾ã—ãŸã€‚ -"Techniques of Machine Learning"ã¯[Jen Looper](https://twitter.com/jenlooper)ã¨[Chris Noring](https://twitter.com/softchris) ã«ã‚ˆã£ã¦åˆ¶ä½œã•ã‚Œã¾ã—ãŸã€‚ \ No newline at end of file +"機械学習ã®æŠ€è¡“"ã¯[Jen Looper](https://twitter.com/jenlooper)ã¨[Chris Noring](https://twitter.com/softchris) ã«ã‚ˆã£ã¦åˆ¶ä½œã•ã‚Œã¾ã—ãŸã€‚ \ No newline at end of file From 5619c3079922a2adb22f0cfec6226c030698b3cb Mon Sep 17 00:00:00 2001 From: Peeeaje <74146834+Peeeaje@users.noreply.github.com> Date: Sat, 3 Jul 2021 23:40:50 +0900 Subject: [PATCH 010/368] translated titles and alt tags --- .../2-history-of-ML/translations/README.ja.md | 24 +++++++++---------- 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/1-Introduction/2-history-of-ML/translations/README.ja.md b/1-Introduction/2-history-of-ML/translations/README.ja.md index 437f156fa1..f9b4c0457f 100644 --- a/1-Introduction/2-history-of-ML/translations/README.ja.md +++ b/1-Introduction/2-history-of-ML/translations/README.ja.md @@ -1,7 +1,7 @@ -# History of machine learning +# 機械学習ã®æ­´å² -![Summary of History of machine learning in a sketchnote](../../../sketchnotes/ml-history.png) -> Sketchnote by [Tomomi Imura](https://www.twitter.com/girlie_mac) +![機械学習ã®æ­´å²ã‚’ã¾ã¨ã‚ãŸã‚¹ã‚±ãƒƒãƒ](../../../sketchnotes/ml-history.png) +> [Tomomi Imura](https://www.twitter.com/girlie_mac)ã«ã‚ˆã‚‹ã‚¹ã‚±ãƒƒãƒ ## [Pre-lecture quiz](https://jolly-sea-0a877260f.azurestaticapps.net/quiz/3/) @@ -43,19 +43,19 @@ * 移動ã—ãŸã‚Šã€ã‚¿ã‚¹ã‚¯ã‚’実行ã™ã‚‹æ–¹æ³•ã‚’「知的ã«ã€æ±ºå®šã™ã‚‹ã“ã¨ãŒã§ãるロボット[「Shakeyã€](https://wikipedia.org/wiki/Shakey_the_robot) - ![Shakey, an intelligent robot](../images/shakey.jpg) - > Shakey in 1972 + ![知的ãªãƒ­ãƒœãƒƒãƒˆã§ã‚ã‚‹Shakey](../images/shakey.jpg) + > 1972ã®Shakey * åˆæœŸã®ã€ŒãŠã—ゃã¹ã‚Šãƒ­ãƒœãƒƒãƒˆã€ã§ã‚ã‚‹Elizaã¯ã€äººã¨ä¼šè©±ã™ã‚‹ã“ã¨ãŒã§ãã€åŽŸå§‹çš„ãªã€Œã‚»ãƒ©ãƒ”ストã€ã®å½¹å‰²ã‚’æžœãŸã—ãŸã€‚エリザã«ã¤ã„ã¦ã¯ã€NLPã®ãƒ¬ãƒƒã‚¹ãƒ³ã§è©³ã—ã説明ã—ã¾ã™ã€‚ - ![Eliza, a bot](../images/eliza.png) - > A version of Eliza, a chatbot + ![Botã§ã‚ã‚‹Eliza](../images/eliza.png) + > ãƒãƒ£ãƒƒãƒˆãƒœãƒƒãƒˆEliza * 「Blocks worldã€ã¯ã€ãƒ–ロックをç©ã¿ä¸Šã’ãŸã‚Šä¸¦ã¹æ›¿ãˆãŸã‚Šã™ã‚‹ãƒžã‚¤ã‚¯ãƒ­ãƒ¯ãƒ¼ãƒ«ãƒ‰ã®ä¸€ä¾‹ã§ã€æ©Ÿæ¢°ã«åˆ¤æ–­åŠ›ã‚’身ã«ã¤ã‘ã•ã›ã‚‹å®Ÿé¨“ã‚’è¡Œã£ãŸã€‚[SHRDLU](https://wikipedia.org/wiki/SHRDLU)ã‚’ã¯ã˜ã‚ã¨ã™ã‚‹ãƒ©ã‚¤ãƒ–ラリã®é€²æ­©ã¯ã€è¨€èªžå‡¦ç†ã®ç™ºå±•ã«å¤§ãã貢献ã—ãŸã€‚ - [![blocks world with SHRDLU](https://img.youtube.com/vi/QAJz4YKUwqw/0.jpg)](https://www.youtube.com/watch?v=QAJz4YKUwqw "blocks world with SHRDLU") + [![SHRDLUを用ã„ãŸblocks world](https://img.youtube.com/vi/QAJz4YKUwqw/0.jpg)](https://www.youtube.com/watch?v=QAJz4YKUwqw "SHRDLUを用ã„ãŸblocks world") - > 🎥 上ã®ç”»åƒã‚’クリックã™ã‚‹ã¨å‹•ç”»ãŒè¦‹ã‚‰ã‚Œã¾ã™ï¼š"Blocks world with SHRDLU" + > 🎥 上ã®ç”»åƒã‚’クリックã™ã‚‹ã¨å‹•ç”»ãŒè¦‹ã‚‰ã‚Œã¾ã™ï¼š"SHRDLUを用ã„ãŸblocks world" ## 1974 - 1980: AIã®å†¬ @@ -91,8 +91,8 @@ 未æ¥ãŒã©ã†ãªã‚‹ã‹ã¯ã¾ã ã‚ã‹ã‚Šã¾ã›ã‚“ãŒã€ã“れらã®ã‚³ãƒ³ãƒ”ュータシステムã¨ã€ãれを動ã‹ã™ã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢ã‚„アルゴリズムをç†è§£ã™ã‚‹ã“ã¨ã¯é‡è¦ã§ã™ã€‚ã“ã®ã‚«ãƒªã‚­ãƒ¥ãƒ©ãƒ ãŒè‡ªèº«ã§åˆ¤æ–­ã™ã‚‹ã«ã‚ãŸã‚Šã€ã‚ˆã‚Šè‰¯ã„ç†è§£ã‚’助ã‘ã‚‹ã‚‚ã®ã«ãªã‚‹ã¨å¹¸ã„ã§ã™ã€‚ -[![The history of deep learning](https://img.youtube.com/vi/mTtDfKgLm54/0.jpg)](https://www.youtube.com/watch?v=mTtDfKgLm54 "The history of deep learning") -> 🎥 上ã®ç”»åƒã‚’クリックã™ã‚‹ã¨å‹•ç”»ãŒè¦‹ã‚‰ã‚Œã¾ã™ï¼šYann LeCun discusses the history of deep learning in this lecture +[![ディープラーニングã®æ­´å²](https://img.youtube.com/vi/mTtDfKgLm54/0.jpg)](https://www.youtube.com/watch?v=mTtDfKgLm54 "ディープラーニングã®æ­´å²") +> 🎥 上ã®ç”»åƒã‚’クリックã™ã‚‹ã¨å‹•ç”»ãŒè¦‹ã‚‰ã‚Œã¾ã™ï¼šã“ã®ãƒ¬ã‚¯ãƒãƒ£ãƒ¼ã§ã¯Yann LeCunãŒãƒ‡ã‚£ãƒ¼ãƒ—ラーニングã®æ­´å²ã«ã¤ã„ã¦è­°è«–ã—ã¦ã„ã¾ã™ã€‚ --- ## 🚀Challenge @@ -107,7 +107,7 @@ [Amy BoydãŒAIã®é€²åŒ–ã«ã¤ã„ã¦è¿°ã¹ã¦ã„ã‚‹ãƒãƒƒãƒ‰ã‚­ãƒ£ã‚¹ãƒˆ](http://runasradio.com/Shows/Show/739) -[![Amy Boydã«ã‚ˆã‚‹AIã®æ­´å²](https://img.youtube.com/vi/EJt3_bFYKss/0.jpg)](https://www.youtube.com/watch?v=EJt3_bFYKss "The history of AI by Amy Boyd") +[![Amy Boydã«ã‚ˆã‚‹AIã®æ­´å²](https://img.youtube.com/vi/EJt3_bFYKss/0.jpg)](https://www.youtube.com/watch?v=EJt3_bFYKss "Amy Boydã«ã‚ˆã‚‹AIã®æ­´å²") ## 課題 From 2020f144afc20780ccded726d7dbc231657d999c Mon Sep 17 00:00:00 2001 From: Seryozni Date: Sun, 4 Jul 2021 16:52:27 +0400 Subject: [PATCH 011/368] Translated README files to all the topics to Russian --- 1-Introduction/translations/README.ru.md | 22 +++++++++ 2-Regression/translations/README.ru.md | 33 ++++++++++++++ 3-Web-App/translations/README.ru.md | 22 +++++++++ 4-Classification/translations/README.ru.md | 25 +++++++++++ 5-Clustering/translations/README.ru.md | 26 +++++++++++ 6-NLP/translations/README.ru.md | 24 ++++++++++ 7-TimeSeries/translations/README.ru.md | 22 +++++++++ 8-Reinforcement/translations/README.ru.md | 52 ++++++++++++++++++++++ 9-Real-World/translations/README.ru.md | 14 ++++++ 9 files changed, 240 insertions(+) create mode 100644 1-Introduction/translations/README.ru.md create mode 100644 2-Regression/translations/README.ru.md create mode 100644 3-Web-App/translations/README.ru.md create mode 100644 4-Classification/translations/README.ru.md create mode 100644 5-Clustering/translations/README.ru.md create mode 100644 6-NLP/translations/README.ru.md create mode 100644 7-TimeSeries/translations/README.ru.md create mode 100644 8-Reinforcement/translations/README.ru.md create mode 100644 9-Real-World/translations/README.ru.md diff --git a/1-Introduction/translations/README.ru.md b/1-Introduction/translations/README.ru.md new file mode 100644 index 0000000000..887fdacdbf --- /dev/null +++ b/1-Introduction/translations/README.ru.md @@ -0,0 +1,22 @@ +# Введение в машинное обучение + +Ð’ Ñтом разделе учебной программы вы познакомитеÑÑŒ Ñ Ð±Ð°Ð·Ð¾Ð²Ñ‹Ð¼Ð¸ концепциÑми, лежащими в оÑнове облаÑти машинного обучениÑ, что Ñто такое, и узнаете о его иÑтории и методах, которые иÑÑледователи иÑпользуют Ð´Ð»Ñ Ñ€Ð°Ð±Ð¾Ñ‚Ñ‹ Ñ Ð½Ð¸Ð¼. Давайте вмеÑте иÑÑледуем Ñтот новый мир машинного обучениÑ! + +! [глобуÑ](images/global.jpg) +> Фото Билла ОкÑфорда на Unsplash + +### Уроки + +1. [Введение в машинное обучение](1-intro-to-ML/README.md) +1. [ИÑÑ‚Ð¾Ñ€Ð¸Ñ Ð¼Ð°ÑˆÐ¸Ð½Ð½Ð¾Ð³Ð¾ Ð¾Ð±ÑƒÑ‡ÐµÐ½Ð¸Ñ Ð¸ иÑкуÑÑтвенного интеллекта](2-history-of-ML/README.md) +1. [СправедливоÑÑ‚ÑŒ и машинное обучение](3-fairness/README.md) +1. [Приемы машинного обучениÑ](4-techniques-of-ML/README.md) +### БлагодарноÑти + +«Введение в машинное обучение» было напиÑано Ñ â™¥ ï¸Ð³Ñ€ÑƒÐ¿Ð¿Ð¾Ð¹ людей, Ð²ÐºÐ»ÑŽÑ‡Ð°Ñ [Мухаммад Сакиб Хан Инан](https://twitter.com/Sakibinan), [Орнелла ÐлтунÑн](https://twitter.com/ornelladotcom) и [Джен Лупер](https://twitter.com/jenlooper) + +«ИÑÑ‚Ð¾Ñ€Ð¸Ñ Ð¼Ð°ÑˆÐ¸Ð½Ð½Ð¾Ð³Ð¾ обучениÑ» была напиÑана Ñ â™¥ ï¸[Джен Лупер](https://twitter.com/jenlooper) и [Эми Бойд](https://twitter.com/AmyKateNicho) + +«СправедливоÑÑ‚ÑŒ и машинное обучение» напиÑано Ñ â™¥ ï¸[Томоми Имура](https://twitter.com/girliemac) + +«Методы машинного обучениÑ» были напиÑаны Ñ â™¥ ï¸[Джен Лупер](https://twitter.com/jenlooper) и [ÐšÑ€Ð¸Ñ Ðоринг](https://twitter.com/softchris) \ No newline at end of file diff --git a/2-Regression/translations/README.ru.md b/2-Regression/translations/README.ru.md new file mode 100644 index 0000000000..58270106de --- /dev/null +++ b/2-Regression/translations/README.ru.md @@ -0,0 +1,33 @@ +# Модели регреÑÑии Ð´Ð»Ñ Ð¼Ð°ÑˆÐ¸Ð½Ð½Ð¾Ð³Ð¾ Ð¾Ð±ÑƒÑ‡ÐµÐ½Ð¸Ñ +## Ð ÐµÐ³Ð¸Ð¾Ð½Ð°Ð»ÑŒÐ½Ð°Ñ Ñ‚ÐµÐ¼Ð°: модели регреÑÑии Ð´Ð»Ñ Ñ†ÐµÐ½ на тыкву в Северной Ðмерике 🎃 + +Ð’ Северной Ðмерике на Ð¥Ñллоуин из тыкв чаÑто вырезают Ñтрашные лица. Давайте узнаем больше об Ñтих воÑхитительных овощах! + +! [jack-o-lanterns](./images/jack-o-lanterns.jpg) +> Фото Бет Тойчманн на Unsplash + +## Что вы узнаете + +Уроки в Ñтом разделе охватывают типы регреÑÑии в контекÑте машинного обучениÑ. Модели регреÑÑии могут помочь определить Ð¾Ñ‚Ð½Ð¾ÑˆÐµÐ½Ð¸Ñ Ð¼ÐµÐ¶Ð´Ñƒ переменными. Этот тип модели может предÑказывать такие значениÑ, как длина, температура или возраÑÑ‚, тем Ñамым выÑвлÑÑ Ð²Ð·Ð°Ð¸Ð¼Ð¾ÑвÑзи между переменными при анализе точек данных. + +Ð’ Ñтой Ñерии уроков вы обнаружите разницу между линейной регреÑÑией и логиÑтичеÑкой регреÑÑией, а также когда вам Ñледует иÑпользовать ту или иную регреÑÑию. + +Ð’ Ñтой группе уроков вы будете наÑтроены, чтобы приÑтупить к задачам машинного обучениÑ, Ð²ÐºÐ»ÑŽÑ‡Ð°Ñ Ð½Ð°Ñтройку кода Visual Studio Ð´Ð»Ñ ÑƒÐ¿Ñ€Ð°Ð²Ð»ÐµÐ½Ð¸Ñ Ð·Ð°Ð¿Ð¸Ñными книжками, общей Ñредой Ð´Ð»Ñ ÑпециалиÑтов по данным. Ð’Ñ‹ откроете Ð´Ð»Ñ ÑÐµÐ±Ñ Scikit-learn, библиотеку Ð´Ð»Ñ Ð¼Ð°ÑˆÐ¸Ð½Ð½Ð¾Ð³Ð¾ обучениÑ, и Ñоздадите Ñвои первые модели, уделÑÑ Ð¾Ñобое внимание моделÑм регреÑÑии в Ñтой главе. + +> СущеÑтвуют полезные инÑтрументы Ñ Ð½ÐµÐ±Ð¾Ð»ÑŒÑˆÐ¸Ð¼ количеÑтвом кода, которые могут помочь вам узнать о работе Ñ Ð¼Ð¾Ð´ÐµÐ»Ñми регреÑÑии. Попробуйте [Azure ML Ð´Ð»Ñ Ñтой задачи](https://docs.microsoft.com/learn/modules/create-regression-model-azure-machine-learning-designer/?WT.mc_id=academic-15963-cxa) + +### Уроки + +1. [ИнÑтрументы торговли](1-Tools/README.md) +2. [Управление данными](2-Data/README.md) +3. [Ð›Ð¸Ð½ÐµÐ¹Ð½Ð°Ñ Ð¸ Ð¿Ð¾Ð»Ð¸Ð½Ð¾Ð¼Ð¸Ð°Ð»ÑŒÐ½Ð°Ñ Ñ€ÐµÐ³Ñ€ÐµÑÑиÑ](3-Linear/README.md) +4. [ЛогиÑтичеÑÐºÐ°Ñ Ñ€ÐµÐ³Ñ€ÐµÑÑиÑ](4-Logistic/README.md) + +--- +### БлагодарноÑти + +«ML Ñ Ñ€ÐµÐ³Ñ€ÐµÑÑией» был напиÑан Ñ Ð¿Ð¾Ð¼Ð¾Ñ‰ÑŒÑŽ ♥ ï¸[Джен Лупер](https://twitter.com/jenlooper) + +♥ ï¸ Ð¡Ñ€ÐµÐ´Ð¸ учаÑтников викторины: [Мухаммад Сакиб Хан Инан](https://twitter.com/Sakibinan) и [Орнелла ÐлтунÑн](https://twitter.com/ornelladotcom) + +Ðабор данных по тыкве предлагаетÑÑ [Ñтот проект на Kaggle](https://www.kaggle.com/usda/a-year-of-pumpkin-prices), а его данные взÑÑ‚Ñ‹ из [Стандартных отчетов по рынкам Ñпециальных культур на терминалах](https://www.marketnews.usda.gov/mnp/fv-report-config-step1?type=termPrice) раÑпроÑтранÑетÑÑ ÐœÐ¸Ð½Ð¸ÑтерÑтвом ÑельÑкого хозÑйÑтва СШÐ. Мы добавили неÑколько точек вокруг цвета на оÑнове разнообразиÑ, чтобы нормализовать раÑпределение. Эти данные находÑÑ‚ÑÑ Ð² открытом доÑтупе. \ No newline at end of file diff --git a/3-Web-App/translations/README.ru.md b/3-Web-App/translations/README.ru.md new file mode 100644 index 0000000000..c252074677 --- /dev/null +++ b/3-Web-App/translations/README.ru.md @@ -0,0 +1,22 @@ +# Создайте веб-приложение Ð´Ð»Ñ Ð¸ÑÐ¿Ð¾Ð»ÑŒÐ·Ð¾Ð²Ð°Ð½Ð¸Ñ Ð²Ð°ÑˆÐµÐ¹ модели машинного Ð¾Ð±ÑƒÑ‡ÐµÐ½Ð¸Ñ + +Ð’ Ñтом разделе учебной программы вы познакомитеÑÑŒ Ñ Ð¿Ñ€Ð¸ÐºÐ»Ð°Ð´Ð½Ð¾Ð¹ темой машинного обучениÑ: как Ñохранить модель Scikit-learn в виде файла, который можно иÑпользовать Ð´Ð»Ñ Ð¿Ñ€Ð¾Ð³Ð½Ð¾Ð·Ð¸Ñ€Ð¾Ð²Ð°Ð½Ð¸Ñ Ð² веб-приложении. ПоÑле ÑÐ¾Ñ…Ñ€Ð°Ð½ÐµÐ½Ð¸Ñ Ð¼Ð¾Ð´ÐµÐ»Ð¸ вы узнаете, как иÑпользовать ее в веб-приложении, Ñозданном во Flask. Сначала вы Ñоздадите модель, иÑÐ¿Ð¾Ð»ÑŒÐ·ÑƒÑ Ð½ÐµÐºÐ¾Ñ‚Ð¾Ñ€Ñ‹Ðµ данные о наблюдениÑÑ… ÐЛО! Затем вы Ñоздадите веб-приложение, которое позволит вам ввеÑти количеÑтво Ñекунд Ñ ÑˆÐ¸Ñ€Ð¾Ñ‚Ð¾Ð¹ и долготой, чтобы предÑказать, ÐºÐ°ÐºÐ°Ñ Ñтрана Ñообщила о видении ÐЛО. + +! [Парковка ÐЛО](images/ufo.jpg) + +Фото Майкла Херрена на Unsplash + + +## Уроки + +1. [Создайте веб-приложение](1-Web-App/README.md) + +## БлагодарноÑти + +«Создайте веб-приложение» было напиÑано Ñ Ð¿Ð¾Ð¼Ð¾Ñ‰ÑŒÑŽ ♥ ï¸[Джен Лупер](https://twitter.com/jenlooper). + +♥ ï¸ Ð¢ÐµÑÑ‚Ñ‹ были напиÑаны Роханом Раджем. + +Ðабор данных взÑÑ‚ из [Kaggle](https://www.kaggle.com/NUFORC/ufo-sightings). + +Ðрхитектура веб-Ð¿Ñ€Ð¸Ð»Ð¾Ð¶ÐµÐ½Ð¸Ñ Ð±Ñ‹Ð»Ð° чаÑтично предложена в [Ñтой Ñтатье](https://towardsdatascience.com/how-to-easily-deploy-machine-learning-models-using-flask-b95af8fe34d4) и [Ñтой репозитории](https://github.com/abhinavsagar/machine-learning-deployment) Ðбхинава Сагара. \ No newline at end of file diff --git a/4-Classification/translations/README.ru.md b/4-Classification/translations/README.ru.md new file mode 100644 index 0000000000..e9359d59a7 --- /dev/null +++ b/4-Classification/translations/README.ru.md @@ -0,0 +1,25 @@ +# Ðачало работы Ñ ÐºÐ»Ð°ÑÑификацией +## Ð ÐµÐ³Ð¸Ð¾Ð½Ð°Ð»ÑŒÐ½Ð°Ñ Ñ‚ÐµÐ¼Ð°: ВкуÑные блюда азиатÑкой и индийÑкой кухни 🜠+ +Ð’ Ðзии и Индии традиции кухни чрезвычайно разнообразны и очень вкуÑны! Давайте поÑмотрим на данные о региональных кухнÑÑ…, чтобы попытатьÑÑ Ð¿Ð¾Ð½ÑÑ‚ÑŒ их ÑоÑтав. + +! [Продавец тайÑкой еды](./images/thai-food.jpg) +> Фото Лишенг Чанг на Unsplash + +## Что вы узнаете + +Ð’ Ñтом разделе вы воÑпользуетеÑÑŒ навыками, полученными в первой чаÑти учебной программы, поÑвÑщенными регреÑÑии, и узнаете о других клаÑÑификаторах, которые вы можете иÑпользовать и которые помогут вам изучить Ñвои данные. + +> СущеÑтвуют полезные инÑтрументы Ñ Ð½ÐµÐ±Ð¾Ð»ÑŒÑˆÐ¸Ð¹ количеÑтвом кода, которые могут помочь вам узнать о работе Ñ Ð¼Ð¾Ð´ÐµÐ»Ñми клаÑÑификации. Попробуйте [Azure ML Ð´Ð»Ñ Ñтой задачи](https://docs.microsoft.com/learn/modules/create-classification-model-azure-machine-learning-designer/?WT.mc_id=academic-15963-cxa) + +## Уроки + +1. [Введение в клаÑÑификацию](1-Introduction/README.md) +2. [Другие клаÑÑификаторы](2-Classifiers-1/README.md) +3. [Еще клаÑÑификаторы](3-Classifiers-2/README.md) +4. [Прикладное машинное обучение: Ñоздание веб-приложениÑ](4-Applied/README.md) +## БлагодарноÑти + +«Ðачало работы Ñ ÐºÐ»Ð°ÑÑификацией» было напиÑано Ñ â™¥ ï¸[КÑÑÑи Бревиу](https://www.twitter.com/cassieview) и [Джен Лупер](https://www.twitter.com/jenlooper) + +Ðабор данных о вкуÑных блюдах взÑÑ‚ из [Kaggle](https://www.kaggle.com/hoandan/asian-and-indian-cuisines) \ No newline at end of file diff --git a/5-Clustering/translations/README.ru.md b/5-Clustering/translations/README.ru.md new file mode 100644 index 0000000000..eb0241b477 --- /dev/null +++ b/5-Clustering/translations/README.ru.md @@ -0,0 +1,26 @@ +# Модели клаÑтеризации Ð´Ð»Ñ Ð¼Ð°ÑˆÐ¸Ð½Ð½Ð¾Ð³Ð¾ Ð¾Ð±ÑƒÑ‡ÐµÐ½Ð¸Ñ + +КлаÑÑ‚ÐµÑ€Ð¸Ð·Ð°Ñ†Ð¸Ñ - Ñто задача машинного обучениÑ, при которой она ищет объекты, которые похожи друг на друга, и группирует их в группы, называемые клаÑтерами. Что отличает клаÑтеризацию от других подходов в машинном обучении, так Ñто то, что вÑе проиÑходит автоматичеÑки, и Ñправедливо Ñказать, что Ñто противоположноÑÑ‚ÑŒ supervised learning. + +## Ð ÐµÐ³Ð¸Ð¾Ð½Ð°Ð»ÑŒÐ½Ð°Ñ Ñ‚ÐµÐ¼Ð°: модели клаÑтеризации Ð´Ð»Ñ Ð¼ÑƒÐ·Ñ‹ÐºÐ°Ð»ÑŒÐ½Ñ‹Ñ… вкуÑов нигерийÑкой публики 🎧 + +У разнообразной публики Ðигерии Ñамые разные музыкальные вкуÑÑ‹. ИÑпользование данных, извлеченных из Spotify (на оÑнове [Ñтой Ñтатьи](https://towardsdatascience.com/country-wise-visual-analysis-of-music-taste-using-spotify-api-seaborn-in-python-77f5b749b421), давайте поÑмотрим на музыку, популÑрную в Ðигерии. Этот набор данных включает данные о различных пеÑнÑÑ… "танцевальноÑÑ‚ÑŒ", "акуÑтичноÑÑ‚ÑŒ", "громкоÑÑ‚ÑŒ", "речевоÑÑ‚ÑŒ", "популÑрноÑÑ‚ÑŒ" и "ÑнергиÑ". Будет интереÑно обнаружить закономерноÑти в Ñтих данных! + +![Поворотный Ñтол](./images/turntable.jpg) + +Фото МарÑела ЛаÑкоÑки на Unsplash + +Ð’ Ñтой Ñерии уроков вы откроете Ð´Ð»Ñ ÑÐµÐ±Ñ Ð½Ð¾Ð²Ñ‹Ðµ ÑпоÑобы анализа данных Ñ Ð¿Ð¾Ð¼Ð¾Ñ‰ÑŒÑŽ методов клаÑтеризации. КлаÑÑ‚ÐµÑ€Ð¸Ð·Ð°Ñ†Ð¸Ñ Ð¾Ñобенно полезна, когда в наборе данных отÑутÑтвуют метки. ЕÑли на нем еÑÑ‚ÑŒ Ñрлыки, тогда могут быть более полезными методы клаÑÑификации, подобные тем, которые вы изучили на предыдущих уроках. Ðо в ÑлучаÑÑ…, когда вы хотите Ñгруппировать немаркированные данные, клаÑÑ‚ÐµÑ€Ð¸Ð·Ð°Ñ†Ð¸Ñ - отличный ÑпоÑоб обнаружить закономерноÑти. + +> СущеÑтвуют полезные инÑтрументы Ñ Ð½ÐµÐ±Ð¾Ð»ÑŒÑˆÐ¸Ð¼ количеÑтвом кода, которые могут помочь вам узнать о работе Ñ Ð¼Ð¾Ð´ÐµÐ»Ñми клаÑтеризации. Попробуйте [Azure ML Ð´Ð»Ñ Ñтой задачи](https://docs.microsoft.com/learn/modules/create-clustering-model-azure-machine-learning-designer/?WT.mc_id=academic-15963-cxa) +## Уроки + +1. [Введение в клаÑтеризацию](1-Visualize/README.md) +2. [КлаÑÑ‚ÐµÑ€Ð¸Ð·Ð°Ñ†Ð¸Ñ K-Means](2-K-Means/README.md) +## БлагодарноÑти + +Эти уроки были напиÑаны Ñ Ð¿Ð¾Ð¼Ð¾Ñ‰ÑŒÑŽ 🎶 [Джен Лупер](https://www.twitter.com/jenlooper) Ñ Ð¿Ð¾Ð»ÐµÐ·Ð½Ñ‹Ð¼Ð¸ отзывами [Ришит Дагли](https://rishit_dagli) и [Мухаммад Сакиб Хан Инан](https://twitter.com/Sakibinan). + +Ðабор данных [ÐигерийÑкие пеÑни](https://www.kaggle.com/sootersaalu/nigerian-songs-spotify) был получен из Kaggle, как и из Spotify. + +Полезные примеры K-Means, которые помогли в Ñоздании Ñтого урока, включают [иÑÑледование радужной оболочки глаза](https://www.kaggle.com/bburns/iris-exploration-pca-k-means-and-gmm-clustering), [вводный блокнот](https://www.kaggle.com/prashant111/k-means-clustering-with-python) и [пример гипотетичеÑкой ÐПО](https://www.kaggle.com/ankandash/pca-k-means-clustering-hierarchical-clustering). \ No newline at end of file diff --git a/6-NLP/translations/README.ru.md b/6-NLP/translations/README.ru.md new file mode 100644 index 0000000000..6743a968a3 --- /dev/null +++ b/6-NLP/translations/README.ru.md @@ -0,0 +1,24 @@ +# Ðачало работы Ñ Ð¾Ð±Ñ€Ð°Ð±Ð¾Ñ‚ÐºÐ¾Ð¹ еÑтеÑтвенного Ñзыка + +Обработка еÑтеÑтвенного Ñзыка, NLP, - Ñто облаÑÑ‚ÑŒ иÑкуÑÑтвенного интеллекта. Ð’ÑÑ Ñта облаÑÑ‚ÑŒ направлена ​​на то, чтобы помочь машинам понимать и обрабатывать человечеÑкий Ñзык. Затем Ñто можно иÑпользовать Ð´Ð»Ñ Ð²Ñ‹Ð¿Ð¾Ð»Ð½ÐµÐ½Ð¸Ñ Ñ‚Ð°ÐºÐ¸Ñ… задач, как проверка орфографии или машинный перевод. + +## Ð ÐµÐ³Ð¸Ð¾Ð½Ð°Ð»ÑŒÐ½Ð°Ñ Ñ‚ÐµÐ¼Ð°: европейÑкие Ñзыки и литература и романтичеÑкие отели Европы â¤ï¸ + +Ð’ Ñтом разделе учебной программы вы познакомитеÑÑŒ Ñ Ð¾Ð´Ð½Ð¸Ð¼ из наиболее раÑпроÑтраненных ÑпоÑобов иÑÐ¿Ð¾Ð»ÑŒÐ·Ð¾Ð²Ð°Ð½Ð¸Ñ Ð¼Ð°ÑˆÐ¸Ð½Ð½Ð¾Ð³Ð¾ обучениÑ: обработкой еÑтеÑтвенного Ñзыка (NLP). Эта ÐºÐ°Ñ‚ÐµÐ³Ð¾Ñ€Ð¸Ñ Ð¸ÑкуÑÑтвенного интеллекта, Ð²Ñ‹Ð²ÐµÐ´ÐµÐ½Ð½Ð°Ñ Ð¸Ð· компьютерной лингвиÑтики, ÑвлÑетÑÑ Ð¼Ð¾Ñтом между людьми и машинами поÑредÑтвом голоÑовой или текÑтовой коммуникации. + +Ðа Ñтих уроках мы изучим оÑновы NLP, Ñоздав небольших диалоговых ботов, чтобы узнать, как машинное обучение помогает Ñделать Ñти разговоры вÑе более и более «умными». Ð’Ñ‹ отправитеÑÑŒ в прошлое, Ð±Ð¾Ð»Ñ‚Ð°Ñ Ñ Ð­Ð»Ð¸Ð·Ð°Ð±ÐµÑ‚ Беннетт и миÑтером ДарÑи из клаÑÑичеÑкого романа Джейн ОÑтин **ГордоÑÑ‚ÑŒ и предубеждение**, опубликованного в 1813 году. Затем вы раÑширите Ñвои знаниÑ, узнав об анализе наÑтроений из отзывов об отелÑÑ… в Европе. + +![Книга о гордоÑти и предубеждениÑÑ… и чай](images/p&p.jpg) +> Фото Элейн Хоулин на Unsplash + +## Уроки + +1. [Введение в обработку еÑтеÑтвенного Ñзыка](1-Introduction-to-NLP/README.md) +2. [Общие задачи и техники NLP](2-Tasks/README.md) +3. [Перевод и анализ тональноÑти Ñ Ð¿Ð¾Ð¼Ð¾Ñ‰ÑŒÑŽ машинного обучениÑ](3-Translation-Sentiment/README.md) +4. [Подготовка данных](4-Hotel-Reviews-1/README.md) +5. [NLTK Ð´Ð»Ñ Ð°Ð½Ð°Ð»Ð¸Ð·Ð° наÑтроений](5-Hotel-Reviews-2/README.md) + +## БлагодарноÑти + +Эти уроки обработки еÑтеÑтвенного Ñзыка были напиÑаны Ñ Ð¿Ð¾Ð¼Ð¾Ñ‰ÑŒÑŽ ☕ [Стивен ХауÑлл](https://twitter.com/Howell_MSFT) \ No newline at end of file diff --git a/7-TimeSeries/translations/README.ru.md b/7-TimeSeries/translations/README.ru.md new file mode 100644 index 0000000000..9cf9be25fa --- /dev/null +++ b/7-TimeSeries/translations/README.ru.md @@ -0,0 +1,22 @@ +# Введение в прогнозирование временных Ñерий + +Что такое прогнозирование временных Ñерий? Речь идет о предÑказании будущих Ñобытий, Ð°Ð½Ð°Ð»Ð¸Ð·Ð¸Ñ€ÑƒÑ Ñ‚ÐµÐ½Ð´ÐµÐ½Ñ†Ð¸Ð¸ прошлого. + +## Ð ÐµÐ³Ð¸Ð¾Ð½Ð°Ð»ÑŒÐ½Ð°Ñ Ñ‚ÐµÐ¼Ð°: потребление ÑлектроÑнергии во вÑем мире ✨ + +Ð’ Ñтих двух уроках вы познакомитеÑÑŒ Ñ Ð¿Ñ€Ð¾Ð³Ð½Ð¾Ð·Ð¸Ñ€Ð¾Ð²Ð°Ð½Ð¸ÐµÐ¼ временных Ñерий, неÑколько менее извеÑтной облаÑтью машинного обучениÑ, котораÑ, тем не менее, чрезвычайно ценна Ð´Ð»Ñ Ð¿Ñ€Ð¾Ð¼Ñ‹ÑˆÐ»ÐµÐ½Ð½Ð¾Ñти и бизнеÑ-приложений, Ñреди других облаÑтей. Ð¥Ð¾Ñ‚Ñ Ð½ÐµÐ¹Ñ€Ð¾Ð½Ð½Ñ‹Ðµ Ñети можно иÑпользовать Ð´Ð»Ñ Ð¿Ð¾Ð²Ñ‹ÑˆÐµÐ½Ð¸Ñ Ð¿Ð¾Ð»ÐµÐ·Ð½Ð¾Ñти Ñтих моделей, мы будем изучать их в контекÑте клаÑÑичеÑкого машинного обучениÑ, поÑкольку модели помогают прогнозировать будущую производительноÑÑ‚ÑŒ на оÑнове прошлого. + +Ðаш региональный Ñ„Ð¾ÐºÑƒÑ - иÑпользование ÑлектроÑнергии в мире, интереÑный набор данных, позволÑющий узнать о прогнозировании будущего иÑÐ¿Ð¾Ð»ÑŒÐ·Ð¾Ð²Ð°Ð½Ð¸Ñ Ñнергии на оÑнове моделей прошлой нагрузки. Ð’Ñ‹ можете увидеть, наÑколько такое прогнозирование может быть чрезвычайно полезным в деловой Ñреде. + +![ÑлектричеÑÐºÐ°Ñ Ñеть](images / electric-grid.jpg) + +Ðвтор фотографии Педди Саи Хритика ÑлектричеÑких башен на дороге в РаджаÑтане на Unsplash + +## Уроки + +1. [Введение в прогнозирование временных Ñ€Ñдов](1-Introduction/README.md) +2. [ПоÑтроение моделей временных Ñ€Ñдов ARIMA] (2-ARIMA/README.md) + +## БлагодарноÑти + +«Введение в прогнозирование временных Ñ€Ñдов» было напиÑано Ñ âš¡ï¸ [ФранчеÑка Лазерри](https://twitter.com/frlazzeri) и [Джен Лупер](https://twitter.com/jenlooper) \ No newline at end of file diff --git a/8-Reinforcement/translations/README.ru.md b/8-Reinforcement/translations/README.ru.md new file mode 100644 index 0000000000..6d6217500b --- /dev/null +++ b/8-Reinforcement/translations/README.ru.md @@ -0,0 +1,52 @@ +# Введение в reinforcement learning +Reinforcement learning (обучение Ñ Ð¿Ð¾Ð´ÐºÑ€ÐµÐ¿Ð»ÐµÐ½Ð¸ÐµÐ¼), RL, раÑÑматриваетÑÑ ÐºÐ°Ðº одна из оÑновных парадигм машинного обучениÑ, нарÑду Ñ supervised learning и unsupervised learning. RL - Ñто вÑе о решениÑÑ…: принÑтие правильных решений или, по крайней мере, извлечение уроков из них. + +ПредÑтавьте, что у Ð²Ð°Ñ ÐµÑÑ‚ÑŒ ÑÐ¼Ð¾Ð´ÐµÐ»Ð¸Ñ€Ð¾Ð²Ð°Ð½Ð½Ð°Ñ Ñреда, Ñ‚Ð°ÐºÐ°Ñ ÐºÐ°Ðº фондовый рынок. Что произойдет, еÑли вы введете определенное правило. Имеет ли Ñто положительный или отрицательный Ñффект? ЕÑли проиÑходит что-то негативное, вам нужно принÑÑ‚ÑŒ Ñто _негативное подкрепление_, извлечь из него урок и изменить курÑ. ЕÑли Ñто положительный результат, вам нужно иÑпользовать Ñто _положительное подкрепление_. + +![peter and the wolf](images/peter.png) + +> Петьке и его друзьÑм нужно ÑпаÑтиÑÑŒ от голодного волка! Ðвтор Ð¸Ð·Ð¾Ð±Ñ€Ð°Ð¶ÐµÐ½Ð¸Ñ [Jen Looper](https://twitter.com/jenlooper) + +## Ð ÐµÐ³Ð¸Ð¾Ð½Ð°Ð»ÑŒÐ½Ð°Ñ Ñ‚ÐµÐ¼Ð°: ÐŸÐµÑ‚Ñ Ð¸ Волк (РоÑÑиÑ) + +[ÐŸÐµÑ‚Ñ Ð¸ Волк](https://en.wikipedia.org/wiki/Peter_and_the_Wolf) - Ð¼ÑƒÐ·Ñ‹ÐºÐ°Ð»ÑŒÐ½Ð°Ñ Ñказка руÑÑкого композитора [Ð¡ÐµÑ€Ð³ÐµÑ ÐŸÑ€Ð¾ÐºÐ¾Ñ„ÑŒÐµÐ²Ð°] (https://en.wikipedia.org/wiki/Sergei_Prokofiev). Это иÑÑ‚Ð¾Ñ€Ð¸Ñ Ð¾ юном пионере Пете, который Ñмело выходит из Ñвоего дома на леÑную полÑну, чтобы преÑледовать волка. Ð’ Ñтом разделе мы обучим алгоритмы машинного обучениÑ, которые помогут Пете: + +- **ИÑÑледуйте** окреÑтноÑти и Ñоздайте оптимальную навигационную карту. +- **УчитеÑÑŒ** пользоватьÑÑ Ñкейтбордом и баланÑировать на нем, чтобы двигатьÑÑ Ð±Ñ‹Ñтрее. + +[![ÐŸÐµÑ‚Ñ Ð¸ Волк](https://img.youtube.com/vi/Fmi5zHg4QSM/0.jpg)] (https://www.youtube.com/watch?v=Fmi5zHg4QSM) + +> 🎥 Ðажмите на изображение выше, чтобы поÑлушать Петю и Волка Прокофьева + +## Обучение Ñ Ð¿Ð¾Ð´ÐºÑ€ÐµÐ¿Ð»ÐµÐ½Ð¸ÐµÐ¼ + +Ð’ предыдущих разделах вы видели два примера проблем машинного обучениÑ: + +- **Supervised**, где у Ð½Ð°Ñ ÐµÑÑ‚ÑŒ наборы данных, которые предлагают примеры решений проблемы, которую мы хотим решить. [КлаÑÑификациÑ](../4-Classification/README.md) и [регреÑÑиÑ] (../ 2-РегреÑÑÐ¸Ñ / README.md) ÑвлÑÑŽÑ‚ÑÑ ÐºÐ¾Ð½Ñ‚Ñ€Ð¾Ð»Ð¸Ñ€ÑƒÐµÐ¼Ñ‹Ð¼Ð¸ учебными задачами. +- **Unsupervised**, в котором у Ð½Ð°Ñ Ð½ÐµÑ‚ помеченных данных обучениÑ. ОÑновным примером unsupervised learning ÑвлÑетÑÑ [КлаÑтеризациÑ](../5-Clustering/README.md). + +Ð’ Ñтом разделе мы познакомим Ð²Ð°Ñ Ñ Ð½Ð¾Ð²Ñ‹Ð¼ типом задач обучениÑ, которые не требуют маркированных данных обучениÑ. ЕÑÑ‚ÑŒ неÑколько типов таких проблем: + +- **[Semi-supervised learning](https://wikipedia.org/wiki/Semi-supervised_learning)**, где у Ð½Ð°Ñ ÐµÑÑ‚ÑŒ много немаркированных данных, которые можно иÑпользовать Ð´Ð»Ñ Ð¿Ñ€ÐµÐ´Ð²Ð°Ñ€Ð¸Ñ‚ÐµÐ»ÑŒÐ½Ð¾Ð³Ð¾ Ð¾Ð±ÑƒÑ‡ÐµÐ½Ð¸Ñ Ð¼Ð¾Ð´ÐµÐ»Ð¸. +- **[Reinforcement learning](https://wikipedia.org/wiki/Reinforcement_learning)**, в котором агент учитÑÑ Ð²ÐµÑти ÑебÑ, Ð¿Ñ€Ð¾Ð²Ð¾Ð´Ñ ÑкÑперименты в некоторой моделируемой Ñреде. + +### Пример - ÐºÐ¾Ð¼Ð¿ÑŒÑŽÑ‚ÐµÑ€Ð½Ð°Ñ Ð¸Ð³Ñ€Ð° + +Предположим, вы хотите научить компьютер играть в игру, например, в шахматы или [Супер Марио](https://wikipedia.org/wiki/Super_Mario). Чтобы компьютер мог играть в игру, нам нужно, чтобы он предÑказывал, какой ход Ñделать в каждом из игровых ÑоÑтоÑний. Ð¥Ð¾Ñ‚Ñ Ñто может показатьÑÑ Ð¿Ñ€Ð¾Ð±Ð»ÐµÐ¼Ð¾Ð¹ клаÑÑификации, Ñто не так - потому что у Ð½Ð°Ñ Ð½ÐµÑ‚ набора данных Ñ ÑоÑтоÑниÑми и ÑоответÑтвующими дейÑтвиÑми. Ð¥Ð¾Ñ‚Ñ Ñƒ Ð½Ð°Ñ Ð¼Ð¾Ð³ÑƒÑ‚ быть некоторые данные, такие как ÑущеÑтвующие шахматные матчи или запиÑи игроков, играющих в Super Mario, вполне вероÑтно, что Ñти данные не будут в доÑтаточной Ñтепени охватывать доÑтаточно большое количеÑтво возможных ÑоÑтоÑний. + +ВмеÑто поиÑка ÑущеÑтвующих игровых данных **Обучение Ñ Ð¿Ð¾Ð´ÐºÑ€ÐµÐ¿Ð»ÐµÐ½Ð¸ÐµÐ¼** (RL) оÑновано на идее *заÑтавить компьютер играть* много раз и наблюдать за результатом. Таким образом, чтобы применить обучение Ñ Ð¿Ð¾Ð´ÐºÑ€ÐµÐ¿Ð»ÐµÐ½Ð¸ÐµÐ¼, нам нужны две вещи: + +- **Среда** и **ÑимулÑтор**, которые позволÑÑŽÑ‚ нам играть в игру много раз. Этот ÑимулÑтор будет определÑÑ‚ÑŒ вÑе правила игры, а также возможные ÑоÑтоÑÐ½Ð¸Ñ Ð¸ дейÑтвиÑ. + +- **Ð¤ÑƒÐ½ÐºÑ†Ð¸Ñ Ð²Ð¾Ð·Ð½Ð°Ð³Ñ€Ð°Ð¶Ð´ÐµÐ½Ð¸Ñ**, ÐºÐ¾Ñ‚Ð¾Ñ€Ð°Ñ Ñообщит нам, наÑколько хорошо мы Ñделали каждый ход или игру. + +ОÑновное различие между другими типами машинного Ð¾Ð±ÑƒÑ‡ÐµÐ½Ð¸Ñ Ð¸ RL заключаетÑÑ Ð² том, что в RL мы обычно не знаем, выиграем мы или проиграем, пока не закончим игру. Таким образом, мы не можем Ñказать, ÑвлÑетÑÑ Ð»Ð¸ конкретный ход хорошим или нет - мы получаем награду только в конце игры. И наша цель - разработать алгоритмы, которые позволÑÑ‚ нам обучать модель в неопределенных уÑловиÑÑ…. Мы узнаем об одном алгоритме RL под названием **Q-Learning**. + +## Уроки + +1. [Введение в обучение Ñ Ð¿Ð¾Ð´ÐºÑ€ÐµÐ¿Ð»ÐµÐ½Ð¸ÐµÐ¼ и Q-Learning](1-QLearning/README.md) +2. [ИÑпользование тренажерного зала](2-Gym/README.md) + +## БлагодарноÑти + +«Введение в обучение Ñ Ð¿Ð¾Ð´ÐºÑ€ÐµÐ¿Ð»ÐµÐ½Ð¸ÐµÐ¼Â» напиÑано Ñ â™¥ ï¸[Дмитрием Сошниковым](http://soshnikov.com) \ No newline at end of file diff --git a/9-Real-World/translations/README.ru.md b/9-Real-World/translations/README.ru.md new file mode 100644 index 0000000000..73691de97c --- /dev/null +++ b/9-Real-World/translations/README.ru.md @@ -0,0 +1,14 @@ +# ПоÑÑ‚Ñкриптум: реальные Ð¿Ñ€Ð¸Ð»Ð¾Ð¶ÐµÐ½Ð¸Ñ ÐºÐ»Ð°ÑÑичеÑкого машинного Ð¾Ð±ÑƒÑ‡ÐµÐ½Ð¸Ñ + +Ð’ Ñтом разделе учебной программы вы познакомитеÑÑŒ Ñ Ð½ÐµÐºÐ¾Ñ‚Ð¾Ñ€Ñ‹Ð¼Ð¸ реальными приложениÑми клаÑÑичеÑкого машинного обучениÑ. Мы обыÑкали Интернет в поиÑках техничеÑких документов и Ñтатей о приложениÑÑ…, которые иÑпользовали Ñти Ñтратегии, избегаÑ, наÑколько Ñто возможно, нейронных Ñетей, глубокого Ð¾Ð±ÑƒÑ‡ÐµÐ½Ð¸Ñ Ð¸ иÑкуÑÑтвенного интеллекта. Узнайте о том, как машинное обучение иÑпользуетÑÑ Ð² бизнеÑ-ÑиÑтемах, ÑкологичеÑких приложениÑÑ…, финанÑах, иÑкуÑÑтве и культуре и многом другом. + +![chess](images/chess.jpg) + +> Фото Ñделано Alexis Fauvet на Unsplash + +## Урок + +1. [Реальные Ð¿Ñ€Ð¸Ð»Ð¾Ð¶ÐµÐ½Ð¸Ñ Ð´Ð»Ñ Ð¼Ð°ÑˆÐ¸Ð½Ð½Ð¾Ð³Ð¾ обучениÑ](1-Applications/README.md) +## БлагодарноÑти + +"Реальные Ð¿Ñ€Ð¸Ð»Ð¾Ð¶ÐµÐ½Ð¸Ñ Ð´Ð»Ñ Ð¼Ð°ÑˆÐ¸Ð½Ð½Ð¾Ð³Ð¾ обучениÑ" была напиÑана группой людей, Ð²ÐºÐ»ÑŽÑ‡Ð°Ñ [Джен Лупер](https://twitter.com/jenlooper) и [Орнелла ÐлтунÑн](https://twitter.com/ornelladotcom). \ No newline at end of file From 74adefb0ff1a4eee7585d9cc269c4ab5c51a08ed Mon Sep 17 00:00:00 2001 From: Seryozni Date: Sun, 4 Jul 2021 16:54:08 +0400 Subject: [PATCH 012/368] Link fix --- 7-TimeSeries/translations/README.ru.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/7-TimeSeries/translations/README.ru.md b/7-TimeSeries/translations/README.ru.md index 9cf9be25fa..6f69ed58dd 100644 --- a/7-TimeSeries/translations/README.ru.md +++ b/7-TimeSeries/translations/README.ru.md @@ -15,7 +15,7 @@ ## Уроки 1. [Введение в прогнозирование временных Ñ€Ñдов](1-Introduction/README.md) -2. [ПоÑтроение моделей временных Ñ€Ñдов ARIMA] (2-ARIMA/README.md) +2. [ПоÑтроение моделей временных Ñ€Ñдов ARIMA](2-ARIMA/README.md) ## БлагодарноÑти From 129f1b05e06401561755b90e434be56c6c2cb1f6 Mon Sep 17 00:00:00 2001 From: seryozni Date: Sun, 4 Jul 2021 18:33:19 +0400 Subject: [PATCH 013/368] Moved all the translations to their proper folders --- 1-Introduction/{ => translations}/README.es.md | 0 1-Introduction/{ => translations}/README.ja.md | 0 2 files changed, 0 insertions(+), 0 deletions(-) rename 1-Introduction/{ => translations}/README.es.md (100%) rename 1-Introduction/{ => translations}/README.ja.md (100%) diff --git a/1-Introduction/README.es.md b/1-Introduction/translations/README.es.md similarity index 100% rename from 1-Introduction/README.es.md rename to 1-Introduction/translations/README.es.md diff --git a/1-Introduction/README.ja.md b/1-Introduction/translations/README.ja.md similarity index 100% rename from 1-Introduction/README.ja.md rename to 1-Introduction/translations/README.ja.md From ee1590f3020b8ba6e53b01ba78dc48030300bc43 Mon Sep 17 00:00:00 2001 From: Peeeaje <74146834+Peeeaje@users.noreply.github.com> Date: Mon, 5 Jul 2021 03:39:57 +0900 Subject: [PATCH 014/368] draft of 1.3 fairness readme.ja I also fixed a typo and an expression --- 1-Introduction/3-fairness/README.md | 2 +- .../3-fairness/translations/README.ja.md | 211 ++++++++++++++++++ 1-Introduction/README.ja.md | 2 +- 3 files changed, 213 insertions(+), 2 deletions(-) create mode 100644 1-Introduction/3-fairness/translations/README.ja.md diff --git a/1-Introduction/3-fairness/README.md b/1-Introduction/3-fairness/README.md index 063c189813..79bf481929 100644 --- a/1-Introduction/3-fairness/README.md +++ b/1-Introduction/3-fairness/README.md @@ -29,7 +29,7 @@ Learn more about Responsible AI by following this [Learning Path](https://docs.m ## Unfairness in data and algorithms -> "If you torture the data long enough, it will confess to anything - Ronald Coase +> "If you torture the data long enough, it will confess to anything" - Ronald Coase This statement sounds extreme, but it is true that data can be manipulated to support any conclusion. Such manipulation can sometimes happen unintentionally. As humans, we all have bias, and it's often difficult to consciously know when you are introducing bias in data. diff --git a/1-Introduction/3-fairness/translations/README.ja.md b/1-Introduction/3-fairness/translations/README.ja.md new file mode 100644 index 0000000000..5794b23fca --- /dev/null +++ b/1-Introduction/3-fairness/translations/README.ja.md @@ -0,0 +1,211 @@ +# 機械学習ã«ãŠã‘る公平㕠+ +![機械学習ã«ãŠã‘る公平性をã¾ã¨ã‚ãŸã‚¹ã‚±ãƒƒãƒ](../../../sketchnotes/ml-fairness.png) +> [Tomomi Imura](https://www.twitter.com/girlie_mac)ã«ã‚ˆã‚‹ã‚¹ã‚±ãƒƒãƒ + +## [Pre-lecture quiz](https://jolly-sea-0a877260f.azurestaticapps.net/quiz/5/) + +## イントロダクション + +ã“ã®ã‚«ãƒªã‚­ãƒ¥ãƒ©ãƒ ã§ã¯ã€æ©Ÿæ¢°å­¦ç¿’ãŒç§ãŸã¡ã®æ—¥å¸¸ç”Ÿæ´»ã«ã©ã®ã‚ˆã†ãªå½±éŸ¿ã‚’与ãˆã¦ã„ã‚‹ã‹ã‚’知るã“ã¨ãŒã§ãã¾ã™ã€‚ãŸã£ãŸä»Šã€åŒ»ç™‚ã®è¨ºæ–­ã‚„ä¸æ­£ã®æ¤œå‡ºãªã©ã€æ—¥å¸¸ã®æ„æ€æ±ºå®šã«ã‚·ã‚¹ãƒ†ãƒ ã‚„モデルãŒé–¢ã‚ã£ã¦ã„ã¾ã™ã€‚ãã®ãŸã‚ã€èª°ã‚‚ãŒå…¬å¹³ãªçµæžœã‚’得られるよã†ã«ã™ã‚‹ãŸã‚ã«ã¯ã€ã“れらã®ãƒ¢ãƒ‡ãƒ«ãŒã†ã¾ã機能ã™ã‚‹ã“ã¨ãŒé‡è¦ã§ã™ã€‚ + +ã—ã‹ã—ã€ã“れらã®ãƒ¢ãƒ‡ãƒ«ã‚’構築ã™ã‚‹ãŸã‚ã«ä½¿ç”¨ã—ã¦ã„るデータã«ã€äººç¨®ã€æ€§åˆ¥ã€æ”¿æ²»çš„見解ã€å®—æ•™ãªã©ã®ç‰¹å®šã®å±žæ€§ãŒæ¬ ã‘ã¦ã„ãŸã‚Šã€ãã®ã‚ˆã†ãªå±žæ€§ãŒåã£ã¦ã„ãŸã‚Šã™ã‚‹ã¨ã€ä½•ãŒèµ·ã“ã‚‹ã‹æƒ³åƒã—ã¦ã¿ã¦ãã ã•ã„。ã¾ãŸã€ãƒ¢ãƒ‡ãƒ«ã®å‡ºåŠ›ãŒç‰¹å®šã®å±¤ã«æœ‰åˆ©ã«ãªã‚‹ã‚ˆã†ã«è§£é‡ˆã•ã‚ŒãŸå ´åˆã¯ã©ã†ã§ã—ょã†ã‹ã€‚ãã®çµæžœã€ã‚¢ãƒ—リケーションã¯ã©ã®ã‚ˆã†ãªå½±éŸ¿ã‚’å—ã‘ã‚‹ã®ã§ã—ょã†ã‹ï¼Ÿ + +ã“ã®ãƒ¬ãƒƒã‚¹ãƒ³ã§ã¯ã€ä»¥ä¸‹ã®ã“ã¨ã‚’è¡Œã„ã¾ã™: + +- 機械学習ã«ãŠã‘る公平性ã®é‡è¦æ€§ã«å¯¾ã™ã‚‹æ„識を高ã‚る。 +- 公平性ã«é–¢é€£ã™ã‚‹å•é¡Œã«ã¤ã„ã¦å­¦ã¶ã€‚ +- 公平性ã®è©•ä¾¡ã¨ç·©å’Œã«ã¤ã„ã¦å­¦ã¶ã€‚ + +## å‰ææ¡ä»¶ +å‰ææ¡ä»¶ã¨ã—ã¦ã€"Responsible AI Principles"ã®Learn Pathã‚’å—講ã—ã€ã“ã®ãƒˆãƒ”ックã«é–¢ã™ã‚‹ä»¥ä¸‹ã®ãƒ“デオを視è´ã—ã¦ãã ã•ã„。 + +ã“ã¡ã‚‰ã®[Learning Path](https://docs.microsoft.com/learn/modules/responsible-ai-principles/?WT.mc_id=academic-15963-cxa)よりã€è²¬ä»»ã®ã‚ã‚‹AIã«ã¤ã„ã¦å­¦ã¶ã€‚ + +[![Microsoftã®è²¬ä»»ã‚ã‚‹AIã«å¯¾ã™ã‚‹å–り組ã¿](https://img.youtube.com/vi/dnC8-uUZXSc/0.jpg)](https://youtu.be/dnC8-uUZXSc "Microsoftã®è²¬ä»»ã‚ã‚‹AIã«å¯¾ã™ã‚‹å–り組ã¿") + +> 🎥 上ã®ç”»åƒã‚’クリックã™ã‚‹ã¨å‹•ç”»ãŒè¡¨ç¤ºã•ã‚Œã¾ã™ï¼šMicrosoftã®è²¬ä»»ã‚ã‚‹AIã«å¯¾ã™ã‚‹å–り組㿠+ +## データやアルゴリズムã®ä¸å…¬å¹³ã• + +> 「データを長ãæ‹·å•ã™ã‚Œã°ã€ä½•ã§ã‚‚自白ã™ã‚‹ã‚ˆã†ã«ãªã‚‹ã€ - Ronald Coase + +ã“ã®è¨€è‘‰ã¯æ¥µç«¯ã«èžã“ãˆã¾ã™ãŒã€ãƒ‡ãƒ¼ã‚¿ãŒã©ã‚“ãªçµè«–ã‚’ã‚‚è£ä»˜ã‘るよã†ã«æ“作ã§ãã‚‹ã“ã¨ã¯äº‹å®Ÿã§ã™ã€‚ã—ã‹ã—ã€ãã®ã‚ˆã†ãªæ“作ã¯ã€æ™‚ã«æ„図ã›ãšã«è¡Œã‚れるã“ã¨ãŒã‚ã‚Šã¾ã™ã€‚人間ã¯èª°ã§ã‚‚ãƒã‚¤ã‚¢ã‚¹ã‚’æŒã£ã¦ãŠã‚Šã€è‡ªåˆ†ãŒã„ã¤ãƒ‡ãƒ¼ã‚¿ã«ãƒã‚¤ã‚¢ã‚¹ã‚’å°Žå…¥ã—ã¦ã„ã‚‹ã‹ã‚’æ„識的ã«çŸ¥ã‚‹ã“ã¨ã¯é›£ã—ã„ã“ã¨ãŒå¤šã„ã®ã§ã™ã€‚ + +AIや機械学習ã«ãŠã‘る公平性ã®ä¿è¨¼ã¯ã€ä¾ç„¶ã¨ã—ã¦è¤‡é›‘ãªç¤¾ä¼šæŠ€è¡“的課題ã§ã™ã€‚ã¤ã¾ã‚Šã€ç´”粋ã«ç¤¾ä¼šçš„ãªè¦–点や技術的ãªè¦–点ã®ã©ã¡ã‚‰ã‹ã‚‰ã‚‚対処ã§ããªã„ã¨ã„ã†ã“ã¨ã§ã™ã€‚ + +### 公平性ã«é–¢é€£ã—ãŸå•é¡Œ + +ä¸å…¬å¹³ã¨ã¯ã©ã†ã„ã†æ„味ã§ã™ã‹ï¼Ÿä¸å…¬å¹³ã¨ã¯ã€äººç¨®ã€æ€§åˆ¥ã€å¹´é½¢ã€éšœå®³ã®æœ‰ç„¡ãªã©ã§å®šç¾©ã•ã‚ŒãŸäººã€…ã®ã‚°ãƒ«ãƒ¼ãƒ—ã«æ‚ªå½±éŸ¿ã‚’与ãˆã‚‹ã“ã¨ã€ã‚ã‚‹ã„ã¯ã€è¢«å®³ã‚’与ãˆã‚‹ã“ã¨ã§ã™ã€‚ + +主ãªä¸å…¬å¹³ã«é–¢é€£ã™ã‚‹å•é¡Œã¯ä»¥ä¸‹ã®ã‚ˆã†ã«åˆ†é¡žã•ã‚Œã¾ã™ã€‚: + +- **アロケーション**。ã‚る性別や民æ—ãŒä»–ã®æ€§åˆ¥ã‚„æ°‘æ—よりも優é‡ã•ã‚Œã¦ã„ã‚‹å ´åˆã€‚ +- **サービスã®è³ª**。ã‚る特定ã®ã‚·ãƒŠãƒªã‚ªã®ãŸã‚ã«ãƒ‡ãƒ¼ã‚¿ã‚’訓練ã—ã¦ã‚‚ã€ç¾å®ŸãŒã‚ˆã‚Šè¤‡é›‘ãªå ´åˆã«ã¯ã‚µãƒ¼ãƒ“スã®è³ªã®ä½Žä¸‹ã«ã¤ãªãŒã‚Šã¾ã™ã€‚ +- **固定観念**。特定ã®ã‚°ãƒ«ãƒ¼ãƒ—ã«ã‚らã‹ã˜ã‚割り当ã¦ã‚‰ã‚ŒãŸå±žæ€§ã‚’関連ã•ã›ã‚‹ã“ã¨ã€‚ +- **誹謗中傷**。何ã‹ã‚„誰ã‹ã‚’ä¸å½“ã«æ‰¹åˆ¤ã—ãŸã‚Šã€ãƒ¬ãƒƒãƒ†ãƒ«ã‚’貼るã“ã¨ã€‚ +- **éŽå‰°è¡¨ç¾ã¾ãŸã¯éŽå°è¡¨ç¾**。特定ã®ã‚°ãƒ«ãƒ¼ãƒ—ãŒç‰¹å®šã®è·æ¥­ã«å°±ã„ã¦ã„る姿ãŒè¦‹ã‚‰ã‚Œãšã€ãれを宣ä¼ã—続ã‘るサービスや機能ã¯è¢«å®³ã‚’助長ã—ã¦ã„ã‚‹ã¨ã„ã†è€ƒãˆã€‚ + +ãã‚Œã§ã¯ã€ã„ãã¤ã‹ä¾‹ã‚’見ã¦ã„ãã¾ã—ょã†ã€‚ + +### アロケーション + +ローン申請を審査ã™ã‚‹ä»®æƒ³çš„ãªã‚·ã‚¹ãƒ†ãƒ ã‚’考ãˆã¦ã¿ã¾ã—ょã†ã€‚ã“ã®ã‚·ã‚¹ãƒ†ãƒ ã§ã¯ã€ä»–ã®ã‚°ãƒ«ãƒ¼ãƒ—よりも白人男性を優秀ãªå€™è£œè€…ã¨ã—ã¦é¸ã¶å‚¾å‘ãŒã‚ã‚Šã¾ã™ã€‚ãã®çµæžœã€ç‰¹å®šã®ç”³è«‹è€…ã«ã¯ãƒ­ãƒ¼ãƒ³ãŒæä¾›ã•ã‚Œã¾ã›ã‚“ã§ã—ãŸã€‚ + +ã‚‚ã†ä¸€ã¤ã®ä¾‹ã¯ã€å¤§ä¼æ¥­ãŒå€™è£œè€…を審査ã™ã‚‹ãŸã‚ã«é–‹ç™ºã—ãŸå®Ÿé¨“çš„ãªæŽ¡ç”¨ãƒ„ールã§ã™ã€‚ã“ã®ãƒ„ールã¯ã€ã‚る性別ã«é–¢é€£ã™ã‚‹è¨€è‘‰ã‚’好むよã†ã«è¨“ç·´ã•ã‚ŒãŸãƒ¢ãƒ‡ãƒ«ã‚’使ã£ã¦ã€ã‚る性別をシステム的ã«å·®åˆ¥ã—ã¦ã„ã¾ã—ãŸã€‚ãã®çµæžœã€å±¥æ­´æ›¸ã«ã€Œå¥³å­ãƒ©ã‚°ãƒ“ーãƒãƒ¼ãƒ ã€ãªã©ã®å˜èªžãŒå«ã¾ã‚Œã¦ã„る候補者ã«ãƒšãƒŠãƒ«ãƒ†ã‚£ã‚’課ã™ã‚‚ã®ã¨ãªã£ã¦ã„ã¾ã—ãŸã€‚ + +✅ ã“ã“ã§ã€ä¸Šè¨˜ã®ã‚ˆã†ãªå®Ÿä¾‹ã‚’å°‘ã—調ã¹ã¦ã¿ã¦ãã ã•ã„。 + +### サービスã®è³ª + +研究者ã¯ã€ã„ãã¤ã‹ã®å¸‚販ã®ã‚¸ã‚§ãƒ³ãƒ€ãƒ¼åˆ†é¡žæ³•ã¯ã€æ˜Žã‚‹ã„肌色ã®ç”·æ€§ã®ç”»åƒã¨æ¯”較ã—ã¦ã€æš—ã„肌色ã®å¥³æ€§ã®ç”»åƒã§ã¯é«˜ã„ä¸æ­£è§£çŽ‡ã‚’示ã—ãŸã“ã¨ã‚’発見ã—ãŸã€‚[å‚ç…§](https://www.media.mit.edu/publications/gender-shades-intersectional-accuracy-disparities-in-commercial-gender-classification/) + +ã¾ãŸã€è‚Œã®è‰²ãŒæš—ã„人を感知ã§ããªã‹ã£ãŸãƒãƒ³ãƒ‰ã‚½ãƒ¼ãƒ—ディスペンサーã®ä¾‹ã‚‚悪ã„æ„味ã§æœ‰åã§ã™ã€‚[å‚ç…§](https://gizmodo.com/why-cant-this-soap-dispenser-identify-dark-skin-1797931773) + +### 固定観念 + +機械翻訳ã«ã¯ã€ã‚¹ãƒ†ãƒ¬ã‚ªã‚¿ã‚¤ãƒ—ãªæ€§åˆ¥è¦³ãŒè¦‹ã‚‰ã‚Œã¾ã™ã€‚「彼ã¯ãƒŠãƒ¼ã‚¹ã§ã€å½¼å¥³ã¯åŒ»è€…ã§ã™ã€‚(“he is a nurse and she is a doctorâ€)ã€ã¨ã„ã†æ–‡ã‚’トルコ語ã«ç¿»è¨³ã™ã‚‹éš›ã€å•é¡ŒãŒç™ºç”Ÿã—ã¾ã—ãŸã€‚トルコ語ã¯å˜æ•°ã®ä¸‰äººç§°ã‚’表ã™ä»£å詞「oã€ãŒ1ã¤ã‚ã‚‹ã®ã¿ã§ã€æ€§åˆ¥ã®åŒºåˆ¥ã®ãªã„言語ã§ã€ã“ã®æ–‡ç« ã‚’トルコ語ã‹ã‚‰è‹±èªžã«ç¿»è¨³ã—ç›´ã™ã¨ã€ã€Œå½¼å¥³ã¯ãƒŠãƒ¼ã‚¹ã§ã€å½¼ã¯åŒ»è€…ã§ã™ã€‚(“she is a nurse and he is a doctorâ€)ã€ã¨ã„ã†ã‚¹ãƒ†ãƒ¬ã‚ªã‚¿ã‚¤ãƒ—ã«ã‚ˆã‚‹æ­£ã—ããªã„文章ã«ãªã£ã¦ã—ã¾ã„ã¾ã™ã€‚ + +![トルコ語ã«å¯¾ã™ã‚‹ç¿»è¨³](../images/gender-bias-translate-en-tr.png) + +![英語ã«å¾©å…ƒã™ã‚‹ç¿»è¨³](../images/gender-bias-translate-tr-en.png) + +### 誹謗中傷 + +ç”»åƒãƒ©ãƒ™ãƒªãƒ³ã‚°æŠ€è¡“ã«ã‚ˆã‚Šã€è‚Œã®è‰²ãŒé»’ã„人ã®ç”»åƒã‚’ゴリラã¨èª¤è¡¨ç¤ºã—ãŸã“ã¨ãŒæœ‰åã§ã™ã€‚誤表示ã¯ã€ã‚·ã‚¹ãƒ†ãƒ ãŒå˜ã«é–“é•ã„ã‚’ã—ãŸã¨ã„ã†ã ã‘ã§ãªãã€é»’人を誹謗中傷ã™ã‚‹ãŸã‚ã«ã“ã®è¡¨ç¾ãŒæ„図的ã«ä½¿ã‚ã‚Œã¦ããŸé•·ã„æ­´å²ã‚’æŒã£ã¦ã„ãŸãŸã‚ã€æœ‰å®³ã§ã‚る。 + +[![AI: 自分ã¯å¥³æ€§ã§ã¯ãªã„ã®ï¼Ÿ](https://img.youtube.com/vi/QxuyfWoVV98/0.jpg)](https://www.youtube.com/watch?v=QxuyfWoVV98 "AI: 自分ã¯å¥³æ€§ã§ã¯ãªã„ã®ï¼Ÿ") +> 🎥 上ã®ç”»åƒã‚’クリックã™ã‚‹ã¨å‹•ç”»ãŒè¡¨ç¤ºã•ã‚Œã¾ã™: AI: 自分ã¯å¥³æ€§ã§ã¯ãªã„ã®ï¼Ÿ - AIã«ã‚ˆã‚‹äººç¨®å·®åˆ¥çš„ãªèª¹è¬—中傷ã«ã‚ˆã‚‹è¢«å®³ã‚’示ã™ãƒ‘フォーマンス + +### éŽå‰°è¡¨ç¾ã¾ãŸã¯éŽå°è¡¨ç¾ + +異常ãªç”»åƒæ¤œç´¢ã®çµæžœã¯ã“ã®å•é¡Œã®è‰¯ã„例ã§ã™ã€‚エンジニアやCEOãªã©ã€ç”·æ€§ã¨å¥³æ€§ã®å‰²åˆãŒåŒã˜ã‹ãれ以上ã®è·æ¥­ã®ç”»åƒã‚’検索ã™ã‚‹ã¨ã€ã©ã¡ã‚‰ã‹ã®æ€§åˆ¥ã«å¤§ããåã£ãŸçµæžœãŒè¡¨ç¤ºã•ã‚Œã‚‹ã®ã§æ³¨æ„ãŒå¿…è¦ã§ã™ã€‚ + +![Bingã§CEOã¨æ¤œç´¢](../images/ceos.png) +> This search on Bing for 'CEO' produces pretty inclusive results + +ã“れらã®5ã¤ã®ä¸»è¦ãªã‚¿ã‚¤ãƒ—ã®å•é¡Œã¯ã€ç›¸äº’ã«æŽ’ä»–çš„ãªã‚‚ã®ã§ã¯ãªãã€1ã¤ã®ã‚·ã‚¹ãƒ†ãƒ ãŒè¤‡æ•°ã®ã‚¿ã‚¤ãƒ—ã®å®³ã‚’示ã™ã“ã¨ã‚‚ã‚ã‚Šã¾ã™ã€‚ã•ã‚‰ã«ã€ãã‚Œãžã‚Œã®ã‚±ãƒ¼ã‚¹ã§ã¯ã€ãã®é‡å¤§æ€§ãŒç•°ãªã‚Šã¾ã™ã€‚例ãˆã°ã€ã‚る人ã«ä¸å½“ã«çŠ¯ç½ªè€…ã®ãƒ¬ãƒƒãƒ†ãƒ«ã‚’貼るã“ã¨ã¯ã€ç”»åƒã‚’誤ã£ã¦è¡¨ç¤ºã™ã‚‹ã“ã¨ã‚ˆã‚Šã‚‚ã¯ã‚‹ã‹ã«æ·±åˆ»ãªå•é¡Œã§ã™ã€‚ã—ã‹ã—ã€æ¯”較的深刻ã§ã¯ãªã„被害ã§ã‚ã£ã¦ã‚‚ã€äººã€…ãŒç–Žå¤–æ„Ÿã‚’æ„Ÿã˜ãŸã‚Šã€ç‰¹åˆ¥è¦–ã•ã‚Œã¦ã„ã‚‹ã¨æ„Ÿã˜ãŸã‚Šã™ã‚‹ã“ã¨ãŒã‚ã‚Šã€ãã®ç´¯ç©çš„ãªå½±éŸ¿ã¯éžå¸¸ã«æŠ‘圧的ãªã‚‚ã®ã«ãªã‚Šã†ã‚‹ã“ã¨ã‚’覚ãˆã¦ãŠãã“ã¨ã¯é‡è¦ã§ã—ょã†ã€‚ + +✅ **ディスカッション**: ã„ãã¤ã‹ã®ä¾‹ã‚’å†æ¤œè¨Žã—ã€ç•°ãªã‚‹å®³ã‚’示ã—ã¦ã„ã‚‹ã‹ã©ã†ã‹ã‚’確èªã—ã¦ãã ã•ã„。 + +| | アロケーション | サービスã®è³ª | 固定観念 | 誹謗中傷 | éŽå‰°è¡¨ç¾/éŽå°è¡¨ç¾ | +| ----------------------- | :--------: | :----------------: | :----------: | :---------: | :----------------------------: | +| 採用システムã®è‡ªå‹•åŒ– | x | x | x | | x | +| 機械翻訳 | | | | | | +| 写真ã®ãƒ©ãƒ™ãƒªãƒ³ã‚° | | | | | | + + +## ä¸å…¬å¹³ã®æ¤œå‡º + +ã‚るシステムãŒä¸å…¬å¹³ãªå‹•ä½œã‚’ã™ã‚‹ç†ç”±ã¯ã•ã¾ã–ã¾ã§ã™ã€‚例ãˆã°ã€ç¤¾ä¼šçš„ãªãƒã‚¤ã‚¢ã‚¹ãŒã€å­¦ç¿’ã«ä½¿ã‚ã‚ŒãŸãƒ‡ãƒ¼ã‚¿ã‚»ãƒƒãƒˆã«å映ã•ã‚Œã¦ã„ã‚‹ã‹ã‚‚ã—ã‚Œãªã„ã§ã™ã—ã€éŽåŽ»ã®ãƒ‡ãƒ¼ã‚¿ã«é ¼ã‚Šã™ãŽãŸãŸã‚ã«ã€æŽ¡ç”¨ã®ä¸å…¬å¹³ãŒæ‚ªåŒ–ã—ãŸã‹ã‚‚ã—ã‚Œã¾ã›ã‚“。ã‚るモデルã¯ã€10å¹´é–“ã«ä¼šç¤¾ã«æ出ã•ã‚ŒãŸå±¥æ­´æ›¸ã®ãƒ‘ターンを利用ã—ã¦ã€ç”·æ€§ã‹ã‚‰ã®å±¥æ­´æ›¸ãŒå¤§åŠã‚’å ã‚ã¦ã„ãŸã“ã¨ã‹ã‚‰ã€ç”·æ€§ã®æ–¹ãŒé©æ ¼ã§ã‚ã‚‹ã¨åˆ¤æ–­ã—ã¾ã—ãŸã€‚ + +特定ã®ã‚°ãƒ«ãƒ¼ãƒ—ã«é–¢ã™ã‚‹ãƒ‡ãƒ¼ã‚¿ãŒä¸å分ã§ã‚ã‚‹ã“ã¨ã‚‚ã€ä¸å…¬å¹³ã®åŽŸå› ã¨ãªã‚Šã¾ã™ã€‚例ãˆã°ã€è‚Œã®è‰²ãŒæ¿ƒã„人ã®ãƒ‡ãƒ¼ã‚¿ãŒå°‘ãªã„ãŸã‚ã«ã€ç”»åƒåˆ†é¡žã«ãŠã„ã¦è‚Œã®è‰²ãŒæ¿ƒã„人ã®ç”»åƒã®ã‚¨ãƒ©ãƒ¼çŽ‡ãŒé«˜ããªã‚Šã¾ã™ã€‚ + +ã¾ãŸã€é–‹ç™ºæ™‚ã®èª¤ã£ãŸä»®å®šã‚‚ä¸å…¬å¹³ã®åŽŸå› ã¨ãªã‚Šã¾ã™ã€‚例ãˆã°ã€äººã®é¡”ã®ç”»åƒã‹ã‚‰çŠ¯ç½ªã‚’犯ã™äººã‚’予測ã™ã‚‹ã“ã¨ã‚’目的ã¨ã—ãŸé¡”分æžã‚·ã‚¹ãƒ†ãƒ ã§ã¯ã€æœ‰å®³ãªæŽ¨æ¸¬ã‚’ã—ã¦ã—ã¾ã†ã“ã¨ãŒã‚ã‚Šã¾ã™ã€‚ãã®çµæžœã€èª¤ã£ãŸåˆ†é¡žã‚’ã•ã‚ŒãŸäººãŒå¤§ããªè¢«å®³ã‚’å—ã‘ã‚‹ã“ã¨ã«ãªã‚Šã‹ã­ã¾ã›ã‚“。 + +## モデルをç†è§£ã—ã€å…¬å¹³æ€§ã‚’構築ã™ã‚‹ + +Although many aspects of fairness are not captured in quantitative fairness metrics, and it is not possible to fully remove bias from a system to guarantee fairness, you are still responsible to detect and to mitigate fairness issues as much as possible. + +When you are working with machine learning models, it is important to understand your models by means of assuring their interpretability and by assessing and mitigating unfairness. + +Let’s use the loan selection example to isolate the case to figure out each factor's level of impact on the prediction. + +## Assessment methods + +1. **Identify harms (and benefits)**. The first step is to identify harms and benefits. Think about how actions and decisions can affect both potential customers and a business itself. + +1. **Identify the affected groups**. Once you understand what kind of harms or benefits that can occur, identify the groups that may be affected. Are these groups defined by gender, ethnicity, or social group? + +1. **Define fairness metrics**. Finally, define a metric so you have something to measure against in your work to improve the situation. + +### Identify harms (and benefits) + +What are the harms and benefits associated with lending? Think about false negatives and false positive scenarios: + +**False negatives** (reject, but Y=1) - in this case, an applicant who will be capable of repaying a loan is rejected. This is an adverse event because the resources of the loans are withheld from qualified applicants. + +**False positives** (accept, but Y=0) - in this case, the applicant does get a loan but eventually defaults. As a result, the applicant's case will be sent to a debt collection agency which can affect their future loan applications. + +### Identify affected groups + +The next step is to determine which groups are likely to be affected. For example, in case of a credit card application, a model might determine that women should receive much lower credit limits compared with their spouses who share household assets. An entire demographic, defined by gender, is thereby affected. + +### Define fairness metrics + +You have identified harms and an affected group, in this case, delineated by gender. Now, use the quantified factors to disaggregate their metrics. For example, using the data below, you can see that women have the largest false positive rate and men have the smallest, and that the opposite is true for false negatives. + +✅ In a future lesson on Clustering, you will see how to build this 'confusion matrix' in code + +| | False positive rate | False negative rate | count | +| ---------- | ------------------- | ------------------- | ----- | +| Women | 0.37 | 0.27 | 54032 | +| Men | 0.31 | 0.35 | 28620 | +| Non-binary | 0.33 | 0.31 | 1266 | + + +This table tells us several things. First, we note that there are comparatively few non-binary people in the data. The data is skewed, so you need to be careful how you interpret these numbers. + +In this case, we have 3 groups and 2 metrics. When we are thinking about how our system affects the group of customers with their loan applicants, this may be sufficient, but when you want to define larger number of groups, you may want to distill this to smaller sets of summaries. To do that, you can add more metrics, such as the largest difference or smallest ratio of each false negative and false positive. + +✅ Stop and Think: What other groups are likely to be affected for loan application? + +## Mitigating unfairness + +To mitigate unfairness, explore the model to generate various mitigated models and compare the tradeoffs it makes between accuracy and fairness to select the most fair model. + +This introductory lesson does not dive deeply into the details of algorithmic unfairness mitigation, such as post-processing and reductions approach, but here is a tool that you may want to try. + +### Fairlearn + +[Fairlearn](https://fairlearn.github.io/) is an open-source Python package that allows you to assess your systems' fairness and mitigate unfairness. + +The tool helps you to assesses how a model's predictions affect different groups, enabling you to compare multiple models by using fairness and performance metrics, and supplying a set of algorithms to mitigate unfairness in binary classification and regression. + +- Learn how to use the different components by checking out the Fairlearn's [GitHub](https://github.com/fairlearn/fairlearn/) + +- Explore the [user guide](https://fairlearn.github.io/main/user_guide/index.html), [examples](https://fairlearn.github.io/main/auto_examples/index.html) + +- Try some [sample notebooks](https://github.com/fairlearn/fairlearn/tree/master/notebooks). + +- Learn [how to enable fairness assessments](https://docs.microsoft.com/azure/machine-learning/how-to-machine-learning-fairness-aml?WT.mc_id=academic-15963-cxa) of machine learning models in Azure Machine Learning. + +- Check out these [sample notebooks](https://github.com/Azure/MachineLearningNotebooks/tree/master/contrib/fairness) for more fairness assessment scenarios in Azure Machine Learning. + +--- +## 🚀 Challenge + +To prevent biases from being introduced in the first place, we should: + +- have a diversity of backgrounds and perspectives among the people working on systems +- invest in datasets that reflect the diversity of our society +- develop better methods for detecting and correcting bias when it occurs + +Think about real-life scenarios where unfairness is evident in model-building and usage. What else should we consider? + +## [Post-lecture quiz](https://jolly-sea-0a877260f.azurestaticapps.net/quiz/6/) +## Review & Self Study + +In this lesson, you have learned some basics of the concepts of fairness and unfairness in machine learning. + +Watch this workshop to dive deeper into the topics: + +- YouTube: Fairness-related harms in AI systems: Examples, assessment, and mitigation by Hanna Wallach and Miro Dudik [Fairness-related harms in AI systems: Examples, assessment, and mitigation - YouTube](https://www.youtube.com/watch?v=1RptHwfkx_k) + +Also, read: + +- Microsoft’s RAI resource center: [Responsible AI Resources – Microsoft AI](https://www.microsoft.com/ai/responsible-ai-resources?activetab=pivot1%3aprimaryr4) + +- Microsoft’s FATE research group: [FATE: Fairness, Accountability, Transparency, and Ethics in AI - Microsoft Research](https://www.microsoft.com/research/theme/fate/) + +Explore the Fairlearn toolkit + +[Fairlearn](https://fairlearn.org/) + +Read about Azure Machine Learning's tools to ensure fairness + +- [Azure Machine Learning](https://docs.microsoft.com/azure/machine-learning/concept-fairness-ml?WT.mc_id=academic-15963-cxa) + +## Assignment + +[Explore Fairlearn](assignment.md) diff --git a/1-Introduction/README.ja.md b/1-Introduction/README.ja.md index 7752ae312a..83f0ba6cb4 100644 --- a/1-Introduction/README.ja.md +++ b/1-Introduction/README.ja.md @@ -9,7 +9,7 @@ 1. [機械学習ã¸ã®å°Žå…¥](1-intro-to-ML/README.md) 1. [機械学習ã¨AIã®æ­´å²](2-history-of-ML/README.md) -1. [公平性ã¨æ©Ÿæ¢°å­¦ç¿’](3-fairness/README.md) +1. [機械学習ã«ãŠã‘る公平ã•](3-fairness/README.md) 1. [機械学習ã®æŠ€è¡“](4-techniques-of-ML/README.md) ### Credits From c6f90b6c8370053b9e6f8892e77c9e3ef6b92346 Mon Sep 17 00:00:00 2001 From: Vitor Mouro <49460655+VitorMouro@users.noreply.github.com> Date: Sun, 4 Jul 2021 16:34:26 -0300 Subject: [PATCH 015/368] fix typo --- 2-Regression/1-Tools/assignment.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/2-Regression/1-Tools/assignment.md b/2-Regression/1-Tools/assignment.md index dd58a16970..de37856c51 100644 --- a/2-Regression/1-Tools/assignment.md +++ b/2-Regression/1-Tools/assignment.md @@ -2,7 +2,7 @@ ## Instructions -Take a look at the [Linnerud dataset](https://scikit-learn.org/stable/modules/generated/sklearn.datasets.load_linnerud.html#sklearn.datasets.load_linnerud) in Scikit-learn. This dataset has multiple [targets](https://scikit-learn.org/stable/datasets/toy_dataset.html#linnerrud-dataset): 'It consists of three excercise (data) and three physiological (target) variables collected from twenty middle-aged men in a fitness club'. +Take a look at the [Linnerud dataset](https://scikit-learn.org/stable/modules/generated/sklearn.datasets.load_linnerud.html#sklearn.datasets.load_linnerud) in Scikit-learn. This dataset has multiple [targets](https://scikit-learn.org/stable/datasets/toy_dataset.html#linnerrud-dataset): 'It consists of three exercise (data) and three physiological (target) variables collected from twenty middle-aged men in a fitness club'. In your own words, describe how to create a Regression model that would plot the relationship between the waistline and how many situps are accomplished. Do the same for the other datapoints in this dataset. From 6bb8b8eee824175f35259c40285faed8b027ede8 Mon Sep 17 00:00:00 2001 From: Peeeaje <74146834+Peeeaje@users.noreply.github.com> Date: Mon, 5 Jul 2021 05:31:15 +0900 Subject: [PATCH 016/368] translate 1.3 fairness readme.md to Japanese --- .../3-fairness/translations/README.ja.md | 103 ++++++++---------- 1 file changed, 48 insertions(+), 55 deletions(-) diff --git a/1-Introduction/3-fairness/translations/README.ja.md b/1-Introduction/3-fairness/translations/README.ja.md index 5794b23fca..3ac9418c81 100644 --- a/1-Introduction/3-fairness/translations/README.ja.md +++ b/1-Introduction/3-fairness/translations/README.ja.md @@ -105,107 +105,100 @@ AIや機械学習ã«ãŠã‘る公平性ã®ä¿è¨¼ã¯ã€ä¾ç„¶ã¨ã—ã¦è¤‡é›‘ãªç¤¾ ## モデルをç†è§£ã—ã€å…¬å¹³æ€§ã‚’構築ã™ã‚‹ -Although many aspects of fairness are not captured in quantitative fairness metrics, and it is not possible to fully remove bias from a system to guarantee fairness, you are still responsible to detect and to mitigate fairness issues as much as possible. +公平性ã®å¤šãã®å´é¢ã¯å®šé‡çš„ãªæŒ‡æ¨™ã§ã¯æ‰ãˆã‚‰ã‚Œãšã€å…¬å¹³æ€§ã‚’ä¿è¨¼ã™ã‚‹ãŸã‚ã«ã‚·ã‚¹ãƒ†ãƒ ã‹ã‚‰ãƒã‚¤ã‚¢ã‚¹ã‚’完全ã«å–り除ãã“ã¨ã¯ä¸å¯èƒ½ã§ã™ãŒã€å…¬å¹³æ€§ã®å•é¡Œã‚’å¯èƒ½ãªé™ã‚Šæ¤œå‡ºã—ã€è»½æ¸›ã™ã‚‹è²¬ä»»ãŒã‚ã‚Šã¾ã™ã€‚ -When you are working with machine learning models, it is important to understand your models by means of assuring their interpretability and by assessing and mitigating unfairness. +機械学習モデルを扱ã†éš›ã«ã¯ã€ãƒ¢ãƒ‡ãƒ«ã®è§£é‡ˆå¯èƒ½æ€§ã‚’ä¿è¨¼ã—ã€ä¸å…¬å¹³ã•ã‚’評価・軽減ã™ã‚‹ã“ã¨ã§ã€ãƒ¢ãƒ‡ãƒ«ã‚’ç†è§£ã™ã‚‹ã“ã¨ãŒé‡è¦ã§ã™ã€‚ -Let’s use the loan selection example to isolate the case to figure out each factor's level of impact on the prediction. +ã“ã“ã§ã¯ã€ãƒ­ãƒ¼ãƒ³é¸æŠžã®ä¾‹ã‚’使ã£ã¦ã‚±ãƒ¼ã‚¹ã‚’切り分ã‘ã€å„è¦ç´ ãŒäºˆæ¸¬ã«ä¸Žãˆã‚‹å½±éŸ¿ã®åº¦åˆã„を把æ¡ã—ã¦ã¿ã¾ã—ょã†ã€‚ -## Assessment methods +## 評価方法 -1. **Identify harms (and benefits)**. The first step is to identify harms and benefits. Think about how actions and decisions can affect both potential customers and a business itself. +1. **å±å®³ï¼ˆã¨åˆ©ç›Šï¼‰ã‚’特定ã™ã‚‹**。最åˆã®ã‚¹ãƒ†ãƒƒãƒ—ã¯ã€å±å®³ã¨åˆ©ç›Šã‚’特定ã™ã‚‹ã“ã¨ã§ã™ã€‚行動や決定ãŒã€æ½œåœ¨çš„ãªé¡§å®¢ã¨ãƒ“ジãƒã‚¹ãã®ã‚‚ã®ã®ä¸¡æ–¹ã«ã©ã®ã‚ˆã†ãªå½±éŸ¿ã‚’与ãˆã‚‹ã‹ã‚’考ãˆã¦ã¿ã¾ã—ょã†ã€‚ -1. **Identify the affected groups**. Once you understand what kind of harms or benefits that can occur, identify the groups that may be affected. Are these groups defined by gender, ethnicity, or social group? +1. **影響をå—ã‘るグループを特定ã™ã‚‹**。ã©ã®ã‚ˆã†ãªå®³ã‚„利益ãŒç™ºç”Ÿã—ã†ã‚‹ã‹ã‚’ç†è§£ã—ãŸã‚‰ã€å½±éŸ¿ã‚’å—ã‘ã‚‹å¯èƒ½æ€§ã®ã‚るグループを特定ã—ã¾ã™ã€‚ã“れらã®ã‚°ãƒ«ãƒ¼ãƒ—ã¯ã€æ€§åˆ¥ã€æ°‘æ—ã€ã¾ãŸã¯ç¤¾ä¼šçš„グループã«ã‚ˆã£ã¦å®šç¾©ã•ã‚Œã‚‹ã§ã—ょã†ã‹ã€‚ -1. **Define fairness metrics**. Finally, define a metric so you have something to measure against in your work to improve the situation. +1. **公正ã•ã®æ¸¬å®šåŸºæº–を定義ã™ã‚‹**。最後ã«ã€çŠ¶æ³ã‚’改善ã™ã‚‹éš›ã«ä½•ã‚’基準ã«ã™ã‚‹ã‹ã®æŒ‡æ¨™ã‚’定義ã—ã¾ã™ã€‚ -### Identify harms (and benefits) +### 有害性(ãŠã‚ˆã³åˆ©ç›Šï¼‰ã‚’特定ã™ã‚‹ +貸与ã«é–¢é€£ã™ã‚‹æœ‰å®³æ€§ã¨åˆ©ç›Šã¯ä½•ã‹ï¼Ÿå½é™°æ€§ã¨å½é™½æ€§ã®ã‚·ãƒŠãƒªã‚ªã«ã¤ã„ã¦è€ƒãˆã¦ã¿ã¾ã—ょã†ã€‚ -What are the harms and benefits associated with lending? Think about false negatives and false positive scenarios: +**å½é™°æ€§ï¼ˆèªå¯ã—ãªã„ãŒã€Y=1)** - ã“ã®å ´åˆã€ãƒ­ãƒ¼ãƒ³ã‚’返済ã§ãã‚‹ã§ã‚ã‚ã†ç”³è«‹è€…ãŒæ‹’å¦ã•ã‚Œã¾ã™ã€‚ã“ã‚Œã¯ã€èžè³‡ãŒè³‡æ ¼ã®ã‚る申請者ã«ãªã•ã‚Œãªããªã‚‹ãŸã‚ã€ä¸åˆ©ãªäº‹è±¡ã¨ãªã‚Šã¾ã™ã€‚ -**False negatives** (reject, but Y=1) - in this case, an applicant who will be capable of repaying a loan is rejected. This is an adverse event because the resources of the loans are withheld from qualified applicants. +**å½é™½æ€§ï¼ˆå—ã‘入れるãŒã€Y=0)** - ã“ã®å ´åˆã€ç”³è«‹è€…ã¯èžè³‡ã‚’å—ã‘ãŸãŒã€æœ€çµ‚çš„ã«ã¯è¿”済ä¸èƒ½ï¼ˆãƒ‡ãƒ•ã‚©ãƒ«ãƒˆï¼‰ã«ãªã‚‹ã€‚ãã®çµæžœã€ç”³è«‹è€…ã®äº‹ä¾‹ã¯å‚µæ¨©å›žåŽä¼šç¤¾ã«é€ã‚‰ã‚Œã€å°†æ¥ã®ãƒ­ãƒ¼ãƒ³ç”³è«‹ã«å½±éŸ¿ã‚’与ãˆã‚‹å¯èƒ½æ€§ãŒã‚ã‚Šã¾ã™ã€‚ -**False positives** (accept, but Y=0) - in this case, the applicant does get a loan but eventually defaults. As a result, the applicant's case will be sent to a debt collection agency which can affect their future loan applications. +### 影響をå—ã‘るグループã®ç‰¹å®š +次ã®ã‚¹ãƒ†ãƒƒãƒ—ã§ã¯ã€ã©ã®ã‚°ãƒ«ãƒ¼ãƒ—ãŒå½±éŸ¿ã‚’å—ã‘ã‚‹å¯èƒ½æ€§ãŒã‚ã‚‹ã‹ã‚’判断ã—ã¾ã™ã€‚例ãˆã°ã€ã‚¯ãƒ¬ã‚¸ãƒƒãƒˆã‚«ãƒ¼ãƒ‰ã®ç”³è«‹ã®å ´åˆã€å®¶è¨ˆã®è³‡ç”£ã‚’共有ã—ã¦ã„ã‚‹é…å¶è€…ã¨æ¯”較ã—ã¦ã€å¥³æ€§ã®ä¸Žä¿¡é™åº¦é¡ã¯å¤§å¹…ã«ä½Žãã™ã¹ãã ã¨ãƒ¢ãƒ‡ãƒ«ãŒåˆ¤æ–­ã™ã‚‹ã‹ã‚‚ã—ã‚Œã¾ã›ã‚“。ã“ã‚Œã«ã‚ˆã‚Šã€ã‚¸ã‚§ãƒ³ãƒ€ãƒ¼ã§å®šç¾©ã•ã‚Œã‚‹å±¤å…¨ä½“ãŒå½±éŸ¿ã‚’å—ã‘ã‚‹ã“ã¨ã«ãªã‚Šã¾ã™ã€‚ -### Identify affected groups +### 公正ã•ã®æ¸¬å®šåŸºæº–を定義ã™ã‚‹ +ã‚ãªãŸã¯æœ‰å®³æ€§ã¨å½±éŸ¿ã‚’å—ã‘るグループ(ã“ã®å ´åˆã¯ã€æ€§åˆ¥ã§å®šç¾©ã•ã‚Œã¦ã„る)をã“ã“ã¾ã§ã«ç‰¹å®šã—ã¾ã—ãŸã€‚次ã«ã€å®šé‡åŒ–ã•ã‚ŒãŸè¦ç´ ã‚’使ã£ã¦ã€ãã®è©•ä¾¡åŸºæº–を分解ã—ã¾ã™ã€‚例ãˆã°ã€ä»¥ä¸‹ã®ãƒ‡ãƒ¼ã‚¿ã‚’使用ã™ã‚‹ã¨ã€å¥³æ€§ã®å½é™½æ€§çŽ‡ãŒæœ€ã‚‚大ããã€ç”·æ€§ãŒæœ€ã‚‚å°ã•ã„ã“ã¨ã€ãã—ã¦ãã®é€†ãŒå½é™°æ€§ã®å ´åˆã«å½“ã¦ã¯ã¾ã‚‹ã“ã¨ãŒã‚ã‹ã‚Šã¾ã™ã€‚ -The next step is to determine which groups are likely to be affected. For example, in case of a credit card application, a model might determine that women should receive much lower credit limits compared with their spouses who share household assets. An entire demographic, defined by gender, is thereby affected. +✅ 今後ã®"クラスタリング"ã®ãƒ¬ãƒƒã‚¹ãƒ³ã§ã¯ã€ã“ã®"æ··åŒè¡Œåˆ—"をコードã§æ§‹ç¯‰ã™ã‚‹æ–¹æ³•ã‚’ã”紹介ã—ã¾ã™ã€‚ -### Define fairness metrics - -You have identified harms and an affected group, in this case, delineated by gender. Now, use the quantified factors to disaggregate their metrics. For example, using the data below, you can see that women have the largest false positive rate and men have the smallest, and that the opposite is true for false negatives. - -✅ In a future lesson on Clustering, you will see how to build this 'confusion matrix' in code - -| | False positive rate | False negative rate | count | +| | å½é™½æ€§çŽ‡ | å½é™°æ€§çŽ‡ | サンプル数 | | ---------- | ------------------- | ------------------- | ----- | -| Women | 0.37 | 0.27 | 54032 | -| Men | 0.31 | 0.35 | 28620 | -| Non-binary | 0.33 | 0.31 | 1266 | +| 女性 | 0.37 | 0.27 | 54032 | +| 男性 | 0.31 | 0.35 | 28620 | +| ã©ã¡ã‚‰ã«ã‚‚属ã•ãªã„ | 0.33 | 0.31 | 1266 | - -This table tells us several things. First, we note that there are comparatively few non-binary people in the data. The data is skewed, so you need to be careful how you interpret these numbers. +ã“ã®è¡¨ã‹ã‚‰ã€ã„ãã¤ã‹ã®ã“ã¨ãŒã‚ã‹ã‚Šã¾ã™ã€‚ã¾ãšã€ãƒ‡ãƒ¼ã‚¿ã«å«ã¾ã‚Œã‚‹ç”·æ€§ã¨å¥³æ€§ã©ã¡ã‚‰ã§ã‚‚ãªã„人ãŒæ¯”較的少ãªã„ã“ã¨ãŒã‚ã‹ã‚Šã¾ã™ã€‚従ã£ã¦ã“ã®ãƒ‡ãƒ¼ã‚¿ã¯æ­ªã‚“ã§ãŠã‚Šã€ã“ã®æ•°å­—ã‚’ã©ã†è§£é‡ˆã™ã‚‹ã‹ã«æ³¨æ„ãŒå¿…è¦ã§ã™ã€‚ -In this case, we have 3 groups and 2 metrics. When we are thinking about how our system affects the group of customers with their loan applicants, this may be sufficient, but when you want to define larger number of groups, you may want to distill this to smaller sets of summaries. To do that, you can add more metrics, such as the largest difference or smallest ratio of each false negative and false positive. +今回ã®å ´åˆã€3ã¤ã®ã‚°ãƒ«ãƒ¼ãƒ—ã¨2ã¤ã®æŒ‡æ¨™ãŒã‚ã‚Šã¾ã™ã€‚ã“ã®ã‚·ã‚¹ãƒ†ãƒ ãŒãƒ­ãƒ¼ãƒ³ç”³è«‹è€…ã§ã‚ã‚‹ãŠå®¢æ§˜ã®ã‚°ãƒ«ãƒ¼ãƒ—ã«ã©ã®ã‚ˆã†ãªå½±éŸ¿ã‚’与ãˆã‚‹ã‹ã‚’考ãˆã‚‹ã¨ãã«ã¯ã“ã‚Œã§å分ã‹ã‚‚ã—ã‚Œã¾ã›ã‚“。ã—ã‹ã—ã€ã‚ˆã‚Šå¤šãã®ã‚°ãƒ«ãƒ¼ãƒ—を定義ã—ãŸã„å ´åˆã¯ã€ã“れをよりå°ã•ãªè¦ç´„ã®ã¾ã¨ã¾ã‚Šã«æŠ½å‡ºã—ãŸã„ã¨æ€ã†ã‹ã‚‚ã—ã‚Œã¾ã›ã‚“。ãã®ãŸã‚ã«ã¯ã€å½é™°æ€§ã¨å½é™½æ€§ãã‚Œãžã‚Œã®æœ€å¤§å€¤ã®å·®ã‚„最å°ã®æ¯”率ãªã©ã€ã‚ˆã‚Šå¤šãã®è¦ç´ ã‚’追加ã™ã‚‹ã“ã¨ãŒã§ãã¾ã™ã€‚ -✅ Stop and Think: What other groups are likely to be affected for loan application? +✅ 一旦ã“ã“ã§è€ƒãˆã¦ã¿ã¦ãã ã•ã„:ローン申請ã®éš›ã«å½±éŸ¿ã‚’å—ã‘ãã†ãªä»–ã®ã‚°ãƒ«ãƒ¼ãƒ—ã¯ï¼Ÿ -## Mitigating unfairness +## ä¸å…¬å¹³ã®ç·©å’Œ -To mitigate unfairness, explore the model to generate various mitigated models and compare the tradeoffs it makes between accuracy and fairness to select the most fair model. +ä¸å…¬å¹³ã‚’ç·©å’Œã™ã‚‹ãŸã‚ã«ã¯ã€ãƒ¢ãƒ‡ãƒ«ã‚’探索ã—ã¦æ§˜ã€…ãªç·©å’Œãƒ¢ãƒ‡ãƒ«ã‚’生æˆã—ã€ç²¾åº¦ã¨å…¬å¹³æ€§ã®é–“ã§è¡Œã†ãƒˆãƒ¬ãƒ¼ãƒ‰ã‚ªãƒ•ã‚’比較ã—ã¦ã€æœ€ã‚‚公平ãªãƒ¢ãƒ‡ãƒ«ã‚’é¸æŠžã—ã¾ã™ã€‚ -This introductory lesson does not dive deeply into the details of algorithmic unfairness mitigation, such as post-processing and reductions approach, but here is a tool that you may want to try. +ã“ã®å…¥é–€ç·¨ã§ã¯ã€å¾Œå‡¦ç†ã‚„リダクションã®ã‚¢ãƒ—ローãƒã¨ã„ã£ãŸã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã«ã‚ˆã‚‹ä¸å…¬å¹³ã®ç·©å’Œã®è©³ç´°ã«ã¤ã„ã¦ã¯æ·±ã触れã¦ã„ã¾ã›ã‚“ãŒã€è©¦ã—ã¦ã„ããŸã„ツールをã“ã“ã§ç´¹ä»‹ã—ã¾ã™ã€‚ ### Fairlearn - -[Fairlearn](https://fairlearn.github.io/) is an open-source Python package that allows you to assess your systems' fairness and mitigate unfairness. +[Fairlearn](https://fairlearn.github.io/)ã¯ã‚ªãƒ¼ãƒ—ンソースã®Pythonパッケージã§ã€ã‚·ã‚¹ãƒ†ãƒ ã®å…¬å¹³æ€§ã‚’評価ã—ã€ä¸å…¬å¹³ã‚’ç·©å’Œã™ã‚‹ã“ã¨ãŒã§ãã¾ã™ã€‚ -The tool helps you to assesses how a model's predictions affect different groups, enabling you to compare multiple models by using fairness and performance metrics, and supplying a set of algorithms to mitigate unfairness in binary classification and regression. +ã“ã®ãƒ„ールã¯ã€ãƒ¢ãƒ‡ãƒ«ã®äºˆæ¸¬ãŒç•°ãªã‚‹ã‚°ãƒ«ãƒ¼ãƒ—ã«ã©ã®ã‚ˆã†ãªå½±éŸ¿ã‚’与ãˆã‚‹ã‹ã‚’評価ã—ã€å…¬å¹³æ€§ã¨ãƒ‘フォーマンスã®æŒ‡æ¨™ã‚’用ã„ã¦è¤‡æ•°ã®ãƒ¢ãƒ‡ãƒ«ã‚’比較ã™ã‚‹ã“ã¨ã‚’å¯èƒ½ã«ã—ã€äºŒé …分類(binary classification)ã¨å›žå¸°(regression)ã«ãŠã‘ã‚‹ä¸å…¬å¹³ã•ã‚’ç·©å’Œã™ã‚‹ãŸã‚ã®ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã‚’æä¾›ã—ã¾ã™ã€‚ -- Learn how to use the different components by checking out the Fairlearn's [GitHub](https://github.com/fairlearn/fairlearn/) +- Fairlearnã®[GitHub](https://github.com/fairlearn/fairlearn/)ã§ã¯ã€å„è¦ç´ ã®ä½¿ç”¨æ–¹æ³•ã‚’紹介ã—ã¦ã„ã¾ã™ã€‚ -- Explore the [user guide](https://fairlearn.github.io/main/user_guide/index.html), [examples](https://fairlearn.github.io/main/auto_examples/index.html) +- [ユーザーガイド](https://fairlearn.github.io/main/user_guide/index.html)ã€[サンプル](https://fairlearn.github.io/main/auto_examples/index.html)を見る。 -- Try some [sample notebooks](https://github.com/fairlearn/fairlearn/tree/master/notebooks). +- [サンプルノートブック](https://github.com/fairlearn/fairlearn/tree/master/notebooks)を試ã™ã€‚ -- Learn [how to enable fairness assessments](https://docs.microsoft.com/azure/machine-learning/how-to-machine-learning-fairness-aml?WT.mc_id=academic-15963-cxa) of machine learning models in Azure Machine Learning. +- Azure Machine Learningã§æ©Ÿæ¢°å­¦ç¿’モデルã®[公平性評価をå¯èƒ½ã«ã™ã‚‹æ–¹æ³•](https://docs.microsoft.com/azure/machine-learning/how-to-machine-learning-fairness-aml?WT.mc_id=academic-15963-cxa)ã‚’å­¦ã¶ã€‚ -- Check out these [sample notebooks](https://github.com/Azure/MachineLearningNotebooks/tree/master/contrib/fairness) for more fairness assessment scenarios in Azure Machine Learning. +- Azure Machine Learningã§[サンプルノートブック](https://github.com/Azure/MachineLearningNotebooks/tree/master/contrib/fairness)ã‚’ãƒã‚§ãƒƒã‚¯ã—ã¦ã€å…¬å¹³æ€§è©•ä¾¡ã®æµã‚Œã‚’確èªã™ã‚‹ã€‚ --- ## 🚀 Challenge -To prevent biases from being introduced in the first place, we should: +ãã‚‚ãã‚‚åã‚ŠãŒç”Ÿã˜ãªã„よã†ã«ã™ã‚‹ãŸã‚ã«ã¯ã€æ¬¡ã®ã‚ˆã†ãªã“ã¨ãŒå¿…è¦ã§ã™ã€‚ -- have a diversity of backgrounds and perspectives among the people working on systems -- invest in datasets that reflect the diversity of our society -- develop better methods for detecting and correcting bias when it occurs +- システムã«æºã‚る人ãŸã¡ã®èƒŒæ™¯ã‚„考ãˆæ–¹ã‚’多様化ã™ã‚‹ã€‚ +- 社会ã®å¤šæ§˜æ€§ã‚’å映ã—ãŸãƒ‡ãƒ¼ã‚¿ã‚»ãƒƒãƒˆã«æŠ•è³‡ã™ã‚‹ã€‚ +- ãƒã‚¤ã‚¢ã‚¹ãŒç™ºç”Ÿã—ãŸã¨ãã«ã€ãれを検知ã—ã¦ä¿®æ­£ã™ã‚‹ãŸã‚ã®ã‚ˆã‚Šè‰¯ã„方法を開発ã™ã‚‹ã€‚ -Think about real-life scenarios where unfairness is evident in model-building and usage. What else should we consider? +モデルã®æ§‹ç¯‰ã‚„使用ã«ãŠã„ã¦ã€ä¸å…¬å¹³ãŒæ˜Žã‚‰ã‹ã«ãªã‚‹ã‚ˆã†ãªç¾å®Ÿã®ã‚·ãƒŠãƒªã‚ªã‚’考ãˆã¦ã¿ã¦ãã ã•ã„。他ã«ã©ã®ã‚ˆã†ãªã“ã¨ã‚’考ãˆã‚‹ã¹ãã§ã—ょã†ã‹ï¼Ÿ ## [Post-lecture quiz](https://jolly-sea-0a877260f.azurestaticapps.net/quiz/6/) ## Review & Self Study -In this lesson, you have learned some basics of the concepts of fairness and unfairness in machine learning. +ã“ã®ãƒ¬ãƒƒã‚¹ãƒ³ã§ã¯ã€æ©Ÿæ¢°å­¦ç¿’ã«ãŠã‘る公平ã€ä¸å…¬å¹³ã®æ¦‚念ã®åŸºç¤Žã‚’å­¦ã³ã¾ã—ãŸã€‚ -Watch this workshop to dive deeper into the topics: - -- YouTube: Fairness-related harms in AI systems: Examples, assessment, and mitigation by Hanna Wallach and Miro Dudik [Fairness-related harms in AI systems: Examples, assessment, and mitigation - YouTube](https://www.youtube.com/watch?v=1RptHwfkx_k) +ã“ã®ãƒ¯ãƒ¼ã‚¯ã‚·ãƒ§ãƒƒãƒ—を見ã¦ã€ãƒˆãƒ”ックをより深ãç†è§£ã—ã¦ãã ã•ã„: -Also, read: +- YouTube: AIシステムã«ãŠã‘る公平性ã«é–¢é€£ã—ãŸè¢«å®³: Hanna Wallachã€Miro Dudikã«ã‚ˆã‚‹ã€äº‹ä¾‹ã€è©•ä¾¡ã€ç·©å’Œç­–ã«ã¤ã„ã¦[AIシステムã«ãŠã‘る公平性ã«é–¢é€£ã—ãŸè¢«å®³: Hanna Wallachã€Miro Dudikã«ã‚ˆã‚‹ã€äº‹ä¾‹ã€è©•ä¾¡ã€ç·©å’Œç­–ã«ã¤ã„㦠- YouTube](https://www.youtube.com/watch?v=1RptHwfkx_k) -- Microsoft’s RAI resource center: [Responsible AI Resources – Microsoft AI](https://www.microsoft.com/ai/responsible-ai-resources?activetab=pivot1%3aprimaryr4) +- Microsoftã®RAIリソースセンター: [責任ã‚ã‚‹AIリソース – Microsoft AI](https://www.microsoft.com/ai/responsible-ai-resources?activetab=pivot1%3aprimaryr4) -- Microsoft’s FATE research group: [FATE: Fairness, Accountability, Transparency, and Ethics in AI - Microsoft Research](https://www.microsoft.com/research/theme/fate/) +- Microsoftã®FATE研究グループ: [AIã«ãŠã‘ã‚‹FATE: Fairness(公平性), Accountability(説明責任), Transparency(é€æ˜Žæ€§ï¼‰, and Ethics(倫ç†ï¼‰- Microsoft Research](https://www.microsoft.com/research/theme/fate/) -Explore the Fairlearn toolkit +Fairlearnã®ãƒ„ールキットを調ã¹ã¦ã¿ã¾ã—ょㆠ-[Fairlearn](https://fairlearn.org/) +- [Fairlearn](https://fairlearn.org/) -Read about Azure Machine Learning's tools to ensure fairness +Azure Machine Learningã«ã‚ˆã‚‹ã€å…¬å¹³æ€§ã‚’確ä¿ã™ã‚‹ãŸã‚ã®ãƒ„ールã«ã¤ã„ã¦èª­ã‚€ - [Azure Machine Learning](https://docs.microsoft.com/azure/machine-learning/concept-fairness-ml?WT.mc_id=academic-15963-cxa) -## Assignment +## 課題 -[Explore Fairlearn](assignment.md) +[Fairlearnを調査ã™ã‚‹](../assignment.md) From a742220e80b676d05c1a38771f714c19d661d9a8 Mon Sep 17 00:00:00 2001 From: Jun Hyuk Han Date: Mon, 5 Jul 2021 14:27:39 +0900 Subject: [PATCH 017/368] fix broken link and change wording --- 1-Introduction/4-techniques-of-ML/README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/1-Introduction/4-techniques-of-ML/README.md b/1-Introduction/4-techniques-of-ML/README.md index ec42fe705e..7061821f27 100644 --- a/1-Introduction/4-techniques-of-ML/README.md +++ b/1-Introduction/4-techniques-of-ML/README.md @@ -35,7 +35,7 @@ Before starting to build your model, there are several tasks you need to complet To be able to answer your question with any kind of certainty, you need a good amount of data of the right type. There are two things you need to do at this point: - **Collect data**. Keeping in mind the previous lesson on fairness in data analysis, collect your data with care. Be aware of the sources of this data, any inherent biases it might have, and document its origin. -- **Prepare data**. There are several steps in the data preparation process. You might need to collate data and normalize it if it comes from diverse sources. You can improve the data's quality and quantity through various methods such as converting strings to numbers (as we do in [Clustering](../../5-Clustering/1-Visualize/README.md)). You might also generate new data, based on the original (as we do in [Classification](../../4-Classification/1-Introduction/README.md)). You can clean and edit the data (as we did prior to the [Web App](../3-Web-App/README.md) lesson). Finally, you might also need to randomize it and shuffle it, depending on your training techniques. +- **Prepare data**. There are several steps in the data preparation process. You might need to collate data and normalize it if it comes from diverse sources. You can improve the data's quality and quantity through various methods such as converting strings to numbers (as we do in [Clustering](../../5-Clustering/1-Visualize/README.md)). You might also generate new data, based on the original (as we do in [Classification](../../4-Classification/1-Introduction/README.md)). You can clean and edit the data (as we will prior to the [Web App](../../3-Web-App/README.md) lesson). Finally, you might also need to randomize it and shuffle it, depending on your training techniques. ✅ After collecting and processing your data, take a moment to see if its shape will allow you to address your intended question. It may be that the data will not perform well in your given task, as we discover in our [Clustering](../../5-Clustering/1-Visualize/README.md) lessons! From 20c50a2c3f887c812908a6c527528f2c204842b8 Mon Sep 17 00:00:00 2001 From: Jun Hyuk Han Date: Mon, 5 Jul 2021 14:50:24 +0900 Subject: [PATCH 018/368] fix broken link --- 1-Introduction/4-techniques-of-ML/README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/1-Introduction/4-techniques-of-ML/README.md b/1-Introduction/4-techniques-of-ML/README.md index 7061821f27..ae9d2c44be 100644 --- a/1-Introduction/4-techniques-of-ML/README.md +++ b/1-Introduction/4-techniques-of-ML/README.md @@ -53,7 +53,7 @@ Prior to training, you need to split your dataset into two or more parts of uneq - **Training**. This part of the dataset is fit to your model to train it. This set constitutes the majority of the original dataset. - **Testing**. A test dataset is an independent group of data, often gathered from the original data, that you use to confirm the performance of the built model. -- **Validating**. A validation set is a smaller independent group of examples that you use to tune the model's hyperparameters, or architecture, to improve the model. Depending on your data's size and the question you are asking, you might not need to build this third set (as we note in [Time Series Forecasting](../7-TimeSeries/1-Introduction/README.md)). +- **Validating**. A validation set is a smaller independent group of examples that you use to tune the model's hyperparameters, or architecture, to improve the model. Depending on your data's size and the question you are asking, you might not need to build this third set (as we note in [Time Series Forecasting](../../7-TimeSeries/1-Introduction/README.md)). ## Building a model From 473566c61a197354d5c40371de961fe6e6edbd6b Mon Sep 17 00:00:00 2001 From: Rohan Hasabe Date: Mon, 5 Jul 2021 14:44:30 +0530 Subject: [PATCH 019/368] Fix typo in 2-Regression/1-Tools * Also add image according to the change. Signed-off-by: Rohan Hasabe --- 2-Regression/1-Tools/README.md | 4 ++-- 2-Regression/1-Tools/images/notebook.png | Bin 151537 -> 28476 bytes 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/2-Regression/1-Tools/README.md b/2-Regression/1-Tools/README.md index d18f529587..3d1fd89e9e 100644 --- a/2-Regression/1-Tools/README.md +++ b/2-Regression/1-Tools/README.md @@ -52,13 +52,13 @@ In this folder, you will find the file _notebook.ipynb_. Next, add some Python code. -1. Type **print("hello notebook'")** in the code block. +1. Type **print("'hello notebook'")** in the code block. 1. Select the arrow to run the code. You should see the printed statement: ```output - hello notebook + 'hello notebook' ``` ![VS Code with a notebook open](images/notebook.png) diff --git a/2-Regression/1-Tools/images/notebook.png b/2-Regression/1-Tools/images/notebook.png index 27a187118d9254d2e589fc6f13359ee7c084e04b..bea9e0e9ff26d6e98bca4ca677a1c8ccc46829e7 100644 GIT binary patch literal 28476 zcmeFYWmFtn+bs%B;|?J>G!h^Q?(QB45(w_W-5aNI3lbo>O9&*mOYlH&4elOXg3B%5 zy?6F|?z#W&IOo^L=pJ2E)skBB$egpnm0!zXqLHA%!NFn5$x5oi!GUDq;NX=}kbpCi z&af^xI5c5P2?=F62?>a@lf9XxjVTcB z{@`#v<;w43Xn^k&Ih%ah^j&@qJKRnOqe@X7sh6)BXBsDnp~zpb7ZlzzJm7$dg$2iW zKj0Tr#{(a?Zv7qa>x1Bffy8T^tRKjt#;i}#*speoE{yiR5964^%km@an`#l`)pjI5 z>mccxmgxxRU~EVt;OQhbC`Cng?WxD{SJ*n142PRg)Xd1}RO)gm3T5TSNBa;=6v^hc zN`DN7nvIAOz+hq!0{g1bHNxzk$*IS`Z zKsX2>l3pf`>eNXV#|a8z#i7d$vVb&(;TppnaOu17QANe$7-*l$Ne0PsSAw)c6CfM1 zT$RX{qQ276)Su!`d*Pf&iN&SkG_tp>&z=dA04ZpG_p>PX_!@zxv^m@UsrJ&o^; zdWsqzWEw;*j%Xy3ljsoQ5XdSHeMO5$5Jr~EG#W}Ir9icb|0-M{oT=+g7yK%#V}K;} z^SCJ4=zIuXOoG%4YV|m**w-IQKS;|S{@|=cnhDsGil#&Vfw+$5L~{kcdU3_(5kW06 zS}^@4fC1F!Bu{KgSrG4(!Y-bY+o#H>CSyg!72NnyTh6vHGiNdHTj8jMHm86#3R6mL z$Xhu9NdXzsT%YXzQOuo#Ga^rJADWhE-w5Bd`_y*BZ8lw^(J036($WZdX~+J@`Vr4f zsW)Cei`8aQrC5Gu{f z!7r*3|0K5+%`BBvl9u13SEq52K_SDUYTu&Yw1^)~-IEmGW4Ss%ZoVzB%Q9Iu*=i@s zxlEEzvd39(97du|N=N*YNS|B$y-n1I_iiMZ19StmX-w&Q0$wA@c6(OoCiCwE-{l(O z8LAqN*`0K#^>EI*R9Tkom~&WHaaQ@5xesm(+zp@)2n-r%5osSWSuvLCurszWYUsFW zeye7$jHsJ6b?r~JB)4c;Z0x^|fB*G;Z{dvD#M>V3pKhW1kfd(R&5usQGq`*6dvE8+ zt8NVKO|ne=rtjU6O+wW|r9W?l?o;KGi zIq9tG(^Zl_at9w^7U#_hjaizGmI&ohwJbd965f zV~X1!z23Kp=6v&!lzFNA)-h9KQR(W^+t+srZIHlP1LLcUUV}M7|RD>?V{SkbfaB{PE-!%dO$P#0K^_DL(gk(oDkstmHiHyo+E^Z5%t`af3KL9c zwzjhFsl88??Q!iX!CQi+x;J&^_JkTpYHn&fYPQAKni*<7dEdVlnuh+0#uQlGf3nxy z*l((FYFD%Htcl)F;r{q1#i5XGjb{ykSGbq+x$rDuMcD=YffD9T;ziQlgBtx8{Hf5N z7o5}FoSennAGvZAXB6|0jj?q3wjIm!X6ikZBE?_*pyEyN+1?qh5Lpu3^Q+%QxsW|& zX>8rIFs*%BM`hbT_kAvYwwqjq96{zT+Rc^UrS{(H#Cm~k(1fdJBPJ;$JCi|>&4>MP zz3X$-8J7mcm4y(ndljBo@r5P;;UU#eS8+efHLYA7jh&UL& zJ*)q{?l!xhO8DU%TxwE=AhY?FJTgo2V?nVTN`L3W=Gcsd6eM(4K{317g-*_^F zg)-XlkH2Zclhuv{f&DdH;i4GeHVFnan39j}{h(x$olZ#fv#rOW#br^l7zJHxKB%Pr zErD=GS8(u2aK*)J)(tj5R{u83MaB*?qXJK)7cV~9zV77NMXJO;x<6~iEXJKb&0`6dP z_ONp?f->1T)BIJ*KlMnOI-59II=EQc+d&@dH8Qq$brGVber)K^-(USSgMP{r}sQe_Q;olA8aP zimyr=G~@X}cX%QABa_|SZ0K+x z^YnmcX&_uqUY`FsCmsa;&nHRrOH&u(pMQY^DSuEnCJZd=|M!Ox^hEY4PU#0Y)Y!j1 zqASX{UEc}fRuiq?@shzShSu7};`chktp?Nyq&*71R6e z3-@g*0_}{Hl``G&^GcH;TKc(Wf`_|llNg{|5qr}WG$o(XEw2qvh4-t+*U#tdlRQp0 z<+Qc63$H>_A=B4EsUk5nfAt)8s{o33J;|5=zrf)b%vWt2IDGF=s#c^dhMMoN zNxai+Hk`?HnHTrZAg$|xX7FTj#9hl8-@NV!bQ#-u9T^^%OHf=~rB3}r3!64Jg5V|H z!_~0gyPdIuu_6_k=W0cRwL z2%GnNxKpjM^3)m@x=25v{;E|aJzHnbBII?Ru-YBj-yVp(5lZPBo5EqAY0CKfk`YGA7q`p9%!;ZM&zC$ zo?huKakb^R^d!)ve(Wpe2FVv1rC(~3SPb;B$oXko@2^V~GWaK%7)Kf!!|VU7l~|Z) zt#n2AlNXGpB()za-_Od=eBYliSFoW{?`^mJQE5J7K#a!8)5Tw*fFlKs_4_@kA+In| zq$1Jix?A``tQTTtelXu~GAcv)e01WI5+y8_Sje+_Z?Q!uoDKyIr6iTh`o(dD?HqTa zFVJ15ZUPP2-@hgojH$XH)WSD=9O`bs@FI4A{Vo>0%DTF{Kl_M0+*VC|ej`&(`Lu}H zZgWUH6pLJq+-x8@PK=C7&)d8C1!4d7EtR+OBB6nIj-avof@>0p=BHQn}S3&0RT3!OTmH|hY-RGG(c+Rmyyvs-Aav6-oq`gVV9tyS-k(|pqR(n>%QzgB{g zOvqF1?N{vrVF9OfhPwGF33fEb;EPJx+GhEY~dTzec$;hagoUVB@bO*$H}QuKjA$o zPq8U96C(PWe_NOidTNEY=|?H)nRc-g>5R4#E2I zga=1pJ6D`AHAxJdz@Q?|+O?3C>Q8)Ni9N@r|D5G>nM8i?aOSWOox6i5t;7rmk zb|V4s1h6W~g8GQf)2UFFf3hhSVKnwF6FBLQy{ubsOT!a=`Sy!4Z1qSv2O}ZeWjl2| zGiBgcK=Ww9Z@`qiP{@ai%7q%0q0saWfeuol}sd{=k6buhF?Zmc%Ee#5fs3byOa zbH2aYDk-Ids(p{4ECW@>JzY>l$qy+K%jTBEsCC_+RoAYzfYAmpzZ62l*u|kBdXHp} zdvdcLuQZQJ8rSolXOZDz-Z`cipDxjR?LCh}G8#3u8ezUz%ZjDvR)+h#^f?#r_~-iNZrCNPa;c|bu)NFqolpld$?m6>^;(Mn6{Gi|@Uh$b} z^GF(-o+ohk7W{Y{?+Qvdj|SW!d3{ll&tSpIl_cv?}I8XOD1Qp!I?A*}Fd z`EB0yN&lS4hQrGiec?}-@ks$RDi6moG=4$|uRnh4Ds(zNqYT;jRPw`_>Yo?{1sy0| zZupMw&rDr~!y&5rHl_*^{u7tL%5kzq(a@#M#Q%veVzW^G4^{*dq={Yh*dlM(OttQC zQq6hiG%7_XS7kOR&X>?58QZo!nn#l?P(}CmGuVTITX?o^e*!*T{?Qs0B`lD`ug|o#&Y|a3$dFM!ha@9bOQefR~7*YBQ}mk#&hu!VIQ3^cq`NE^=pwk7h?8z zVcs|UbthL_S)ng?vrkgEZBuFK==x^NvzP!o5-Sz?6wy_;9xua9PbX2GMuv!s{}YTR z;=I4oAgRQbU{_5URJ>8Z^lp!4?V0J|kORSP{4nNP;q{jr9byklLcgY5H-WCvVIP)Q z8Tvt$E!HBFS9}-asH!DznPC1uf9A{KUjbI>-H#8?Jn=>T+8HOr@RYELf>(q3T$&#oWI#doS;& zY@JW=Dy7#)cM?O$=ZI)N4qBCywJM>YZ$j{lw)D`^J%-2 z)vQVit5Lx4l^zi02=D6twJ>}A!0cTR#SDJ+3@8&@JRpVyncWp3)Vk`siLg*h3F%ayQ4G7a7XT3f#|0mpu6kek73fy z?Mf4jTmG*owzn0`jzW8@EloDk`fz98upEHyJtR0HoE2oG4ddAX0$_KOc$~b1_`()B zes0vPxax$OodvTnfyUd`Bw55yI)mHA( z4`d4IJY~_}JMFp5J90FC{kXqk2~jUyDt6GQYmG;;#pG$JM1SYS2#+53q~If7EEBvjIMS22?Nf#DJD-}sh` z=1t!2Hyt$Yc6k2UHr~De0Vc@D93E}Rih3(Hi_*)z|D|dO59397Q_di#?q7Q*%np zWchjQGMD%~qV%t6cwj*VlcEOob`u&f;z;ArAFZ}Ug2%Jsp2#VtaVLggk``BvE_{3w zQd%V{afSN@^K{Fo@3ZITmeb}y3QhgJ{>Kj=#1G>LiR>Py5#}uk2^1B-fNvzk34Moc zf>q6uuAoqcYaF}W@>%Wkhq>;M#+(w&1v`}d#cw`dypbBGRF-A9g`}c9IM!6)) zK5%jT?5n=GKVxLiQ4PTNn;0R!Z>7!-I}uDK4Wxs^R)r|xu70ri%RF8si6X&*+HAUr zJi0Z&vxLl&wNtWN=(zdac(q2Q&l?L&C)nBOc^=%pXXFf zVI0r%Ev=FtS{5c4xV_NeILMPaCE!5w{j38^>4ZOCV%Gg-`6YRPofu?vdrudEQO%-@ zvP>MndP?uTmos_p>9z~=h>87_G=+jKxi_Wf8K(H_=hw8`vvDOO@)OKq74LB9j<=4C zrH7?6L3~2(Ll}Oup8jC5PZzGr5-+}#F{e&GW9yJ}0)akC$G6ppmx9=q^YxBMhsEaN zcC`O7sA1Y5VbIRky2TSOYGCsVF`x_3DW=i_=GO@!r^>-X^Fygl-D_^Ud4_xA_mK3> zp^TE%Y-LE-XtT#@$BXE*@Z0NOv|NGQU{&K&Nx335)xqNTKWGS}ZUl1mtgjqHYzDbr zX2E7}#tL2`?gvZ%$EC#jp?2|dnBQQHh6gy41HtXtugRAhRuNOLA;PF)Yu>V(ea{SY z5R*mwKQfsDEFW-mFW?IJGS^wbp*9+nUUXT^`j{DDDSD}Cxyl?m@i2uh6vf=_>5MS0z;*S^f zLwWK-ukj53YRd&mN2^?pYDfodt}556G9$icy)xA&fe}+umeCdJH8`1$l%W@^7DU+0 z)iRAo>Q=s!Yq&k>Hw8A@hmRjWT3$CNz{QW{9wZl4tK>;ndR^H4PD=VkG}H)gEYxdq zOMJ{1TddG*8!bk2D}lH_Hqvi1#XKL#ORQcu#d4sYv`tTI?u_Pz>U!=JsOCvya`h~( zbcPDWeCKLi|L|(PxTF-X`S6MJS0M7A=U~!@ECGi8w|u$h?_WOv2L`G=HXL8$Kiivm zBcJ%9U=w`Cc$@V;ITapXAcj&zAIMYVsCELeg&BC}9Rj^C7Tk%?ujZ8VWpOEKnW-y|BZS3YI$SQ8JI zT3bDN=qO?vg(Qx!h*%+Sa=;%YLb22j_jSK{H@v_)yy|kN9t)XL_Bm=tF829USXdk$ zd@R&(`0XYko2|dc9d4>lv%>gSs!66$^-GkfyPJzNxychC%`XD4ee=1=2YUX33A%i< zZ`xk?u=H^M>04wnqsR5xj=omI+);*93l*pZ}<`A7O@E+K%Egwp3AFiib zX{o5h06+2NCsow35D}MELa|2n$*o!HAo-$s(x8F|FKVpl0SI)l zu$%qHM%S{JDIBP zs-LJP*+PQKvZ#M<-N=esbApUbp|*%oZ2;xT^g7^!eOjsLq3GxA2eMXl!SKs|%ci3a z^ht?37&1kyby>?>JrNYaX6p%VUF)zs0DCadJB1-2zfhV1(pzi##5B7`x$wMD3QyS$ zbX>%^0KsO)>9ZmlNo+|C-aWw9pEWEko1P}V)WAx1-5SbxOc~wBMGeXMU3&F~-y?d) zItB9g)53JAkB)mNOPGuKU4IoJvr+J8`Q0}tB(sW1M-k;%O#<*j!(L^7f;$S&5`C7p zDv&`J__pUYZH3ck#vUv*l^{R$ecw5Fe|NF;@`AS!IhLYhr8kZSNPVAp0Jw$vH>h$u zlR=B8>&C~2jU)r7AW$REB6JP&>Z0|5X?rNJ1$g9N?E%Qs%k6>cguOt3p?-lRXm?cI z0wnyu1c4P{2cws8-@jdYxXW_7QcB|<@rqg*HwdK7-Qj+SHj;c1KlS4}BF9xLfLjn7qr+@-`B+I^v$GICl~ z9ABQTWMo6NI7^>j`kwrtFHSq(Eo*(dk?lpSPjai=M!)y@Q>LmusjdvsV%`-HFMeJ- zhVuooze(%T^26KH;UrS~=Fx>vciY@KLwLt4*gsWKm?hDdJ+f#_N*U ze!e%8_YGU0r9qnKMKkwoe~s^*=S+=t(g{VXhcWS1oz>(Et!j%nQ8;SyZ)Nw4$!E(b zY{DkrFUXsgN;#i9>WU1rA(4Fd>2-#hhhb!?Aj%Eb*{h()&0W3dhzeGYT!OLgeQ! ztCK&1bJOAMilH+H!y*ri!@@UT_(^X)x%^B`h)ZpWjEx@idhqvbvA_xL`Gjw`C$Ba_ zdInLk9R#x-OoEF>cwP$X;siP_xK4dugG3jrm(X>FVrwkO>SbJhNN1f-KMwV@?392% z@Fo(bQZO_OKt=>Zj3JPOJ%f8-NyY-W0+bzY&g-B-Mu-vTFd!6(Vjzs5T>|{uVzIfw zvq&O;hTCyfQhgX$AteUOn55hZ>+A2HmB(^7UCnI-n~<-f0G`v}2M>0PyKt+^B4G;M z=Q-+A@qILc7YxIoW^x=9LR|-#7YP7hRd^NwvzK)OM4To2mnv{EA+_2=ySXFWRJkL* zR8XTZ$BvrWO4G>Y*_DOt?Oz@4a3O91=jg|Qp==b3I34;SQ!Cl8o?!bz%|~n_-17Z` zl6#f~M<^5nd1U;M2lrnOltZFX%jT|ucz+^tQpMaaGs|%7VN2veMf?_se9N@dl>0BS z*5*etP(~x{_=z+v*--2$S3e(h!cmX+k0eG}!=Nv0w>JZcI5F8iUM?<&68dslRQx{a zXDK-(x?F8sZ8q?`b3Gl2{kS8BhDwS}4cmU^U!N|@vc@&6uvEW79oo}%TT?9b;(Ea% zEL0R{KbA^jLIR08*Q6)2A)Dhv*@ZMu=A3o!h`Gcc+N-f(_Sdg6lYIlwdIOgPoA2XPosdi zrvG*b&*MaB*rH>r?!$|_9r60}NeY2q1tMW4^neDcFOc*lpr+Mg)vGWpiL@Y`5c((L z==*TCc-Yp}2q>yy-uutQZ^y@#M7~A`e*)W*FNzA@wqSVBV5j%oUY)8(#bq4K*|)0O zV(Y(#_Ky6lUg7mtsunYp=v_g^8X~7)sb+#&EqWaaWdKY*qfj$j5hXZBD|!fyO~>N2 z(DLat5?cTzE-Q*gwZ`1?K1j*ipnT7i_a~O&l*sbHnIX0k={NKvX&B-GeG&b)lIz7v z=dIVSw#zEwBko1guqQA0C}cn?@^Gz3ejX-7$s*pHpOAR^DfI3nY9wC4iB#!)7kVBcYm3bxz zGPQ}iK%i+Vmu9eP_M;g?foF3+&{aoTmjv+4vS0`NZWLVG#$D9ida<>T z=uj@-nU0K@G-2Zen{giM`E{y%g>mdg?bjDbzA2qY)*`wgaLV%oqhDd9Oulg)4)Iew z`uU?Y?;Iq!{VH;IM>(NPXMB(N^9xErhGiyJN}NqL%@BtGFG_5IxoV!FMY~cOJ0v~c zusKl-q#nDDgmTa7wO9fK zX`p{Z_}EbL3#PPpj=hN9gqIG~IS)O_d@jHlc1&AKsQc(I-b^C_#L)nfZC(o3pF>s*Kp;#cv<5wg$I9F{!VsIWwhMbBs8E{3Gfqx+OXAgVxC1KY z1Uj#3WUYWijCi+;dLw~KAASBPJG6YFP)MFg#q1t#_SiUF>vgImppcI7r&BdM6(swi);t|c-qjGIg<3csDogR`aB%?(j zlcM*gj6ZqqXqOV?0ZW;zDBE5DY}#K#_?ow$N1h@=);vaoDfy-Kn))ZUUR;du zbot_E@+&HB2;-S` zy)S)ymqHG>ppmQzeuHNKFI@QHv^ zJ3B+{I-8xyXS$ya%&qp7bRe3IRF* z_Ghu=_7a6D9Abswk@q2d(H(pe++~)9(-x887sHucj27So(J}iQ%#C%nofHakoWA2fqy

    2?_kOSk_7ei9_*Y@XE8q70v;I zI4z+8VvbI&+SX=e`hg4ooEReZusA*&Z2urRqe^WAvYVgVUM6Uf#ij`3^>6l{E`M#O zwY@h>crST1@DEfZo4t(J3yEbVeZF`RC~BeL746yu?6DHe!6VKY9O^dl; zx!Ejz|*s!^M4P?jJvXs8*RN zYLFSx7rMyIIy4Uqe=6727dxt1s6i{jbQZWS(&po*hRsxjAw0()UurN1#yVj~tRkY&&Oty_v5>G*&vl`in|^cq zs`&q+swi;*7Gqmn2gn3DttMoxrb?MmQBm2=28j4}zf|P|FG+F}+-9Jkb0-qylO!Hs5 zo=7;v4~Rh}fBFfKc3s?m6c?hPr$7bhC>npJN)}N4*#0B=0b2auR6lTtcys@#7674) zCYT|X5u*F-Zw3so`N{eD`SD%<)da0W1e}oUCryX&e|SNE@Do@ElTIxy+CPxyWAQ-X zE+h=*e+5QxD6B}~`~OR4&LCQ%_45DhdQnht_1#kf&afYAiKO~Xi`u- z8+yEY(n}K~tDq2N{{GQ@%+%RuZ3#beSO9ta8o(VNf#P}mo4u;xcPkyi2cG&m^#FKY z;C*#ctli{(_z{5K*vi!7epA8x0c#3c#r>>Vk=uN=nPyi^B@xoQ&I)@GrH1RLg zi;D*RZjVEo0T9jYL-x;Ci7%en&Q>3%)#^4lJ(2a@d1ouL(CAv=uyR(`vDo5e3W)l& z`u#tnbHV_Yt7*x@U$e>WD;b{??VbVvT)zCOQ|ERX%-h!w+5yPf5B2rD>Pr9^rUGfu zh(;e3_Xj@UBUJaUzgeS{oH4#fK6;X5s_}&9Y-QFfJ#Hwb0rb^}!@z=PL z1Et{A1K{*Jz$>_qKZ4lo9w+bIGtdF!Gb;C3-7Y{f(b2z7kDZ=Ab70yhKXFe!mo$9X z989}=*6$*qJ0@~_6qtt9`duMs^I#eDj|~=`MpcF&!p;CVTos_K67lrx?lGw4(Vyqb zy?%Y3#5ul?V^-FB-z@gL2(oxO0wCaiu(aNOpt@qJ zo2N!$TmoQid3PY2fAy$C1=6#`!`27=gXYuMVi<^Bn?b1f6SWvAyre@S58eb`>P>F8 z1ccR&YtmY^HYwi6-GokO+bU%emCauPDQhuV8Ny0I)6GGXk6X?j5Jg^A?)%oPtDp&wky~0n!;jG4+Mv(hSh%&QZ+T z7#rt81_6kAqDC^qR$wiPW1?ni4i_hi8$p7?Jj0uY6;N{4&pqPZ zzt`Yx(JEY*F$#f|qrODWV-$w|$j(9RutPk30gBFPryxnbCz^b+A$IO$=1P_TnB9abk^s6GSY!;f zaHoA2(EKzRq<<)#%cjS1(UTawJXw3Qj`^Zaz?4cYCTVOV?|^MvU@q-g)>3ASzurJHlJl zUZFcQ#eq$?dS5w}EojaD{QQPKF-aJu0}vdQj+Xqwx3zEchd+-YB3Kv`9?vXi9Lfl$qXIY|qt| z4VNk(AYU>YIQ@7A=pvKO5pwYL5#IN$#`~e1NBNM6Xkv=aF4@NuS1~6=1kb|JZ zvdXbJ4ib*oIYO7gwPCHuQ5(Xq5UU7+8{VeZ3=e=}`}a#V2f?Ac4EM!~DyOOa6C1~% zpv%OsnAo)dgC}B`(j7tY2HV3AB)^r2pM_(!*}24wpIM!x@A>1C(6nWMUTEF+(MqSC z6Zk$&oLFGlm18hUuSCD4QFM7`YDNLu9*06Rt{R9RgKKP&i(fX=*7Ayb+JyG9GLh{s z46r0`I4od8IQI@o`g$%-+mR{HKFhn!it)~Q#nUM!K)H7@%LsjyM`R&okKFufL~@1X zwOHQ5?zO}G7X{cm;R)8PILjl}>{t7MmBPdDn?yQx9arSpnY8g)4GBA-9EVH;(($j- z)6410I~AB4Yw%^5mPfmG_qSKVJ8ac6=kK;nMkcFWc2rx6g_9VysMEM@o$7KTRY)7F zkB-KW%0R!uOPdjpv96L}xsk?z>o=Y>)6#7wD&>OJCfk z@-)F6FV~bc4bhZj6<(nAq%N`m(4*)ClF>at&TUU(I9s+yf|PJVa?_BMz>UToO3ba99GB7;^?>KXWV|^RWE!?M?0q#cGP#la3-+<*`~q@%lo2& zA3RRQo^Fr#bgHL)vs@*N=R3i~R_UU$?8HP_Af5F*+s?UO{q0$lWu}03(iVv7D8Bjr z3HIAlqez$SQZ~%RP{3vC*@{Pn5NA10s%!k%%JQ?L)9Bx}gIV~~%ad*fT8(7G+DZ^P zt}*~(jX&kO8IZ%xBlFr7I!x{!;CIo9DXbt-7K=evl+tz_+g-i(DtgXJF|xMx*)dlU zqYUeZj)HcO;Yyi;@)rtBC@2R`dMN!EEeL0E2n+W+&{@uoslRAebo|y8r=45f`Ke={ zm?7X^+&O9+QqOA3%j1$E-PeYsXFEHBw5$~oM3spsj%$l(Y84aw{9eVbXn!GnX<*Mz zCNoba=BgEXt2J0e!@-`T8}#BqkMO2fiUOnU17o@8zC4{}pwZnI zdifi3>kdwzcYon0g496cwwEZd>FQ?%z{NYyzF!2b_BLzl4n=~u#n4uZMKwjcwvik7 zicvE0x}JzKCiCD6k#Y!96jBipXOj7auy?GuGQ25_Qo?mGH8&$#Fij=)DyXAb*7kbl zFy5(1q-Rwqwm!Y!K5d33b{F~0aqatN&M!i`W-=HC=y{zP&d)qz6%j#iDkMb-!`drR zaXnXusubpuej>yWRMZ+V-Ml$G8jPA89YVeM*}dKHvqGw5(kZD(Bq6i`1^iM|m(k4x zF=&eM{N8u&UBgqC>OGr9dYSwIvc?5 z3yZaETa7I~E6ZK+VDr2C#aY`s+P>=9dWTw;^o_0KC4s^`*5*6*!QJmRk@0{|BQ^A$ z$ILxJR9bV(;LwEiB87lr{?Fssk*TF!lIeh9v(yaFcJwf#2iC9@KyaYKwCfk94)93B zoK1Gq=4h5Wx2Zi=htF$w2xC)kICQWQ+|7EVty+C-opM`3cYWdL%#wtE+Lo~u(!xwX zkq1?^(EKd(x;U^)696DL_4MQgrmrss{b*y6futJI1J|L7E%2)&QRABf_|*NmjG0koj0s;a6B3tBs3{Y{b@gPebFic&@hPQ9NDQ=x~l zS~()cnXM{Jd@7z|zMc%vb+AJchZz(ZT*d5bqLZx3bh2f}%GStI0Y1V4U zOXf_q(<~EBwu44%Gw!IXDea!Z)NFek5l)KR`~~w3L2?IjS>0&ir=pSFLWaIUB38XD z2k(%Qmfge9()gs2v6~K85mDb{$q7!cU#Lt`mE%|)lZz`k9h`LA(U00B9b0hy+-Z4( zKn7y9z`*AM%nN!9+VU1tPZ1G0NIH1B`zfvB%}QXW%S9^jWpUNGo(=!|#l;3^^9mVT zBTBQo?rj#oRk;U}i7P{DQH+h>=0vIXB+|Rd&e-jwA*Anx+OvptG1&r8k!p-O6i#8q z_HCYi3|aM~DeY^Q>QS1cRb=bvD=>eHfo(E7j;42Qh*Lex21%AZ_J*_Y`&vQtDv|X0 zGT&C7$U^I<_eH1r$-+O?8RchQ9ZBPG&k|!`J8~4F&JD!*;7k|h>@N$lkv8o&u#QX3 zIq}Xm`6wRgou2%JDb^n#_o7dU#elj7HUM|kWj0m)iA!J~GQUrbZ#tB}{h$J;V+Zth|(Y5{DPZEBUm6lPXVEdm*7_<8izl)~%ndnvd&TQd?b z$|*U>82GYsRu=!fm+}}D<)B;eS$%suBDTE(#Bn)J@gsM;^~e`3j>)8$ob4m&l_Xs= z9>{=m&?RMlCjuNb^H@ln({do62_&Ko88nHd7ULom@Pks%c{@as_4(#CYLdpF!3;`* z%v8pCv5`x1w8XJ{)MB&=PCsl}zgj?n%+ib7Rr@ad#Nlof3sUK{dc9leyvD&Z>*aQc z4DH6#U}C=e)FkgC2GL3H+&P%@bKZsf>+SQEpU2H7ZcaF&e(@ z(QVtQ^j}zeuQ2=@2%D_HjbI7_s$@c2xlt!LJ~q`+f&nKJ)FmDACo}AnM601+46JSRHz!o-wG7|WNgoi>3#3wDPv>pz^4T=;s1aDJ64gX7z zj6)2H5{tml*!%-}s$q)Jg!jauTKWFJO`YBq9-&=-+?+0n{Zk z0VVM|aTWL<;2Os82#lB#Mos>~?*UE=Ex(t;>|eNCMH;BNkC~|C)jzO1jt2m!$C(e{ z{ejGLqkx(d^jP2idG!Z!0zg^R!^zM7z~+*~K+OzIhF_TfU^?pnrX!Wf_UaFG9>xyT ztm4+LsrRRgz4*Xm2%3ESgWUqp&k&ot7f{jo2hfhA2Oh)YJL8|Xl_7nAvzD4c{GS2( z8vy(N7@#N7l0NY_(WuF`WArVcgW;C~Yip64iN)mNCipwW4@1x}da@kiv=u6AY{5qra6-x`PrKP1+!{4tg%U2k8Ph0~S z8ug>@f4Bxnd&YnlD8k>rPXvUNZy*it{;hvGMl=&ow($dSA&FeTZCm{_kl+75Wu0|Y zRPFcn6+r<-5Tqm=kdPKpy1R4e77%Ia5TvA)h5(g|@HIqqj?|;l7$LDaG7+p z&>s3fq+dqqpES|2=WyB2`caAys$6Yt0FnaJODtjUdzg60;`5jt&LR{(>z|rp<>GQR zeOS1bM6MsMP!S}cOb=aGNd0u{7P$$61pxB2km%>*-xpGnlJGZ-gb}sO;u-8N>-m*X z`0YD)pNH)aD9N>h)(Z=%_K(W4>YDTcb6uf~`oDe~T9tHfUcK+NsNMM$2Nn7~UMzdH zCQmQHWQA1U?(ARUii#K#5)z1xQLfVVM0LdeO0RZ?%$ZO=H8r(gntmkzF1A=3u3!MU z#LmH4p=-X4%}Ee`#{|<7m;3$mjQej}IWX;_8bo^{c;XIEDUPT=o?9Goj2IXJW6^(J zbYy)OX^QBJ+W@n08Z=vRi_`afzDIg~>+=u6G0$f&FPhX#-t^IOtQW}k z?X~Y2_Bu)#Zo%#^_?I$0_8!UU&#pK>80A*ya{Df)`nBP_WKL$R_?gISDpW`nrM~=J z!=7&(*;Ejb9Ir^u)y+OOQphs??A5sxP_RLyL`&;e-l;lRIOw7bwwyZ%kc z@08(9XUvanH}9-<(4Y1_i8y~tF!#{u%P;Of`LEVTn|G+lSZQc9r_!);AMNX}aIORY zM!zwaB+8r+y}o9u=ohbUZ?vmi18Km1?_IX{6K!sgyfx&hs&ZrA@azleJMU+H+&-u@ z?n!)axXOm<^m$*`wVmASI8DB0e1owf&&lsxrF-l)^DDzPB}rm|=gdQGNX`F7sGdkN z5IHo8O=k3ET=n8NJ~K>x(el}8cGaa3Pzw!_p`IxSOT1;}z~)&gbYqNgoyWVZG*SW& zJ5?$lIJIaGS9V(O)lNM_e)sGy+TD269iA|~`oK4KPHQQ@-r<9~ks zlR}-3uFRuIS2unm;0N&%rd4=#0VZlo2XbLA2tUtpsQ;#wDkNcvyPfW<;Ji;+|AN$h z4^#DDwnM)K(wMo9A?h=NI#DH^Bs?2Ehh)={xmF!ayS!mulHL%1cmLF+kCOP}YGS`D zwk*AmR)7iM2<|n0l`bK=Af>^V7JY`_1+pj@X6dAw<2~)0MeQ8&JqaxnU^m%84omY1 zpX!mjiXLrd$l(!eSrX)%`hQn(Bz$!=(bIm9iVT<;s?cd^o4!rFGSfjk-%)!6Kz6=9 zVrr7wAWIZ1C!n69*syYu*&=dUaRXsqFyAVe6#uF2E3pm({hKf!Mj0Bt5Raae$Uw)YOQgFxpzs|}l>2zd$}+g#X(Hj960~h| z&!4+U3*Q7)CbTLR6SJ^19tmb<<{Zs@h)eK(lum_twax4f%sS)&=^;$-+OLEaHEMU*8)L-=DuncI6gF+1haad2>lAW(0sn;0`+5 z%^5v5M_k5*(^lB$2-+|3h1yaE1x6j(0h*}o?jY)aD-Ep$rHu`&LDR)GfSse%9mo1E zrsQ6bspO(>`lQ>e+H>vnm{;rSEX3-I+1#`P3=9~x${kAfZ;bx!tdt~Bg^^bKE}BM1 z&bP{DjX^EdL=`Kac0YS<@cL@=CC@x+dl8sFL)JFQ!l;>=_G(o@`!@=nCc6sfbo7>p z>11vN{iQ`|THYiz?s{~%=rRF~Y)%7v`BLXjD@_z2e}L2#jJR_d6nq@|iA#@x=A5`&fLb zRfXBS#(@^c;cr2gSC#L5&@}(1W@7JbF*sOZrkz8dSS8SRnAvkmnkBu^zZxwX&dnxb zT2=sADY(<*Y%nDGtxC#TH!*mz4A_P#&zWdgG%!j3eALhenhpeSp0cE2vgpn^EwX7MsqtL2Z$^q3e-e zfMi;Az&ZWECRF4%3F;$jyvp`Gn${ zUEW5#s(vi*n`+J;xABM7XD>B9-b;_W3op5S7nM|{V^oeyYt=APq1t6oq0&JkLrvvK zDs)&Y*Twa-tc7DpcFa9xYnz@Q=IAimHF}`>*1S3i5h>AWL0HqO-xYgE8UYo1@Z&?| z`S3%^j}?sL3|zw*@UT0r!0M#)#K{-(q@r+)nh1XbT3#?wl(!+rLBIZX6N|Xix!01~ zfkvA+vM?&Sj*BkCJj{abkp0Y|+W4z{cGm9PhE@`-h*WDwZ&T}_?wcSR%O~#laT7dN z%1bNghCdIa0oC;S^r6BqXFf z`ZN2|4ZAwAan((GVa*zR+R~Ty4u5{-$JD7iaM|CHQY&9tG@2grUaNT=%#>E3aG-*_ zPj>##Rob=;I$6;j_ymc8Z-T`I*wnsvD??$!VM47?{DH%$4^g24szVYKMS1T8?*rCXO;z)x$TGXg^k*GMU+oF;RCThKNvpYwnL2$vE??1TFV?h%IYsiW0v|?K0PNu z#ETQo*5+S1mf!9KQ)e~vVIG!Uv=UoM*1YGZ|8j^vs@!s?HVua@6@&Q|7DEx(FEI;`5S^n}Jd_mATTWhcV$xGaC(y03vn38km4jX0v)4&2&m()V18SVF}g@hV>D zZXrdU*bnPSV9zC2Y>-kmS!xz*+(QXb%!*cz<apM?yz=p*W8HQr zUfgZbIb0XznDzG~TFGqM7Uf%2u`lG^D!AUSq4s|4IZk=En7N;|QH5OaRrgBz{Ay|80Kc@LY3d9m#fd`-_f>hEx z4Clf0sm(HXp;dB3JAAs*u!?s}mhoTOU7$l^29^Hkc|u~Cn9VdZA=?wpaNbp;yv50u z<71^s#cFM-=4OZkDl@AKK|=wKFdp|yEb)qL#uv*osj^xTu*PS&>kk6C>b7Hyhhf9a zBfh82oJNUma<(T1$1eqlLo4_X8c{hPtcp1%Y^~Xb*osc!I*1ODjC~NcLhmq6e#t_2 zYuW3OCkX5ciS?a*wtWAkUj8X6ME#N22rR(WQJMGJMZzlaXPz^Q_^WI3pm$AVz>1kd zS5Lf<(wUExO|(?7<6yQmjSo3)ccw!VHe670Fl!1QZgEg%bk>p01t$hlvN;-=amdHU z8WFE(z2u5M`&YN5z(97~*QwRY=AeqF-})>L0#Ps0BR(K|pj z|Ja3BEEA&^{;vn+lBF>**2Let2zd(?6U4GyZn%ikS!@-(zhobDoneKB;KG!eP29^yM4iZvQ5rFWqR7-`~A?b zW;274p#?Act&X)1xc4?VWi2cX2=aJxp)+ z#;o5)jkQm$N7lMu-bUC^=f7{nJzFfsLKMGk5=im)XsAuj zXx8Jb?shD7aiy!|)5-=$0A?4l^TolA<(hueUGUb-v+qoBtS2L4-H`8K&?p09Dz#W{=MoLf;fA={itug>)3zyZlifor_xAIH zRa;tR1|>n#wNZtfNE({fkfcy+TU9u$G}^FsgtW_yYbtCBNv^-(j^BdM_S?2Gl3JE7 z$KNd^gSo0}D)iK7EkSMZuj3Y@0!kN}10o*62nd~Ixj0fkrmVfxDs>AvcRYVJn&bdY zY&bL@w%u=D)Ly7!u{|aho546Ush};@-pqlo`xZ1__=UiWl8mb#gwCAf9V|R-rSmIj z+p$$`TUiYnv`!Cf>1e`t4A;Pw>OiCJnTA<}I#OMvPif<4@viC3>-MM|q^s77?sgte zqa1V>bj5Q>**D_Z)XBGo+)FIysWsYhy1FC#kVOg$KBI3zc^NpvEH(>r@cRR$#vf>m zno}SZSweoIprxbWJpW|}OzndPg`SKGh<`udB?t%!iI{7dpTrKtwS8O5sHnC!T*xTo z7p$DJ=1WwQd*OawEKKU|nuaa(;gA^Gp*CG}5c&SyNYOWJR4#|K`&0=#F{2%ex#}GQ zYt25ni+izzBApAA>kE5?gq8EjqO$mX9ekaUWuRs8zLAxCCaJz8?y)s#3Dd<3p?cjS z$rfB?T4khKA=c)}2^$qN^=f-q-RZnE4Mp+pNP2=dywJ6Mt}jDH`NU+$^yJ-`pouP< z!kQ{$W-w2ovo=SmKE}z;_TcX90E%SU1MbR%F*dt7bLZK!RnwC@#?ukg(owxEeIgIL zmm3p}+rE-~SXj00kCZKU^*Qoxy z?0vqSnW9p+r50EJX-w?1Zq&*6rp(U0M`B_jBrvNb`iN1pikNoF7_;v-i<0&a#_s@acr#xbD@tzHWjUVKbvdT^xJ%8$8`4C+uUqB}wp7~HhB37g6t6wiAk$9fHH zisv^8cca9c)SgC-oH*)t)F^(}qM>}8$#oKQH)Oz=)QdI7EFnHVuLMZ+GU(Ok7y`RW{zW~Y z)Fe&gwn>NB>8wg09nsw4!{__< zWGEr$cv998M%o*_;v}j!NNoLyU;o@BNP^d7t|y%k_tO%Rjr*Rc=OCLA-i;zo0Vr)ytl8R1~Nyt8ld9 zec{5bsy7{HWYe`1VwK6GAUIYXS=|!m7=tv!BcJD}cxTI%m!Rcrhz`=O(Er?xjf_m& zRj+l>R@*iBj|CR03Wg++Vg<-GBW2$@ECzGBT$J6mrjxd!{)`ed!vQ+zJmaV4mcjXh zTHPAE?9xIQ&K^{v!bbxdJo?G>NtGLG1p9(fUW*I-ebdnTGo2V z57$Qci;2Id7th}VL&{HGZZ3p6q!wtw5N{8`k6BV*c2@&izv{*bMwY>CyWw3ll6v4f z6j9ilNm*B6E1r0QbGKumQ_|BFJq(ZV+Hy`LTDb-~K zu6l-!m7PT%wRS!mSwvZ{ltO9cve(!&HQh>fkj+yp!4WRp48@_lTR2$VWtuzlK3^b{ zUPeWy^kc%;+#10;?X6})T-zFs^2XZ86{im`sIlurSya+L;IVO%=JtcSS6Gqe^%0cA zczqxhNK2X`x72J1u;ru``FPF^XdBU8{?HBx5q|3+YCU>woCC}05S4p zGhX@anZw!<{|LY<-#$4UAIgw%b6$!in&^__8I)rPA=oLM_Z&-$#%T48lV$q+luxO~ z&V;P6u20uF{O&_nz+aopaWMFs3RtLcKmei8{dkhjOR*rg$qskIBll-JO~M&*ZNMoy zA3NZI68?#S=#7^JxEF@uy{6@|eUOlf+=j9J+0n77e>c;%8`aDAN6lX9(F>Cccxni?0+H! z^OYzWc;q4XIW}hwhBfd}`RrH9eAxRfJ|>NM(hn2WdQ3;F1w)T=%bNo5y)o?9coyIV zyD2OU*yaiSMw|{?Hs!?9ZcxGHY4?Ox?5Ubnq?02b5@>SO6e-*o2#L~xwz=CdtVxI? z{y&s0eWT{cUsXPaG&(3~*_r z!`-$!v9MkHirbX+U6^`mEqUH#teWrpco#S6$gKaeBXKItZ(BEucmbuLr4kx1~Rzn6S{+w30}>_Hkk*c6k9 zD9BU1JgjCUpiquW<=q}WSQ|WQRr}4!_J)(_RdtT#2815)JHzH}p$g8)czHK{P!B48 z%y6p4|F7+L47g0#z$JLo()>SGsQ=^2c_BWI^rqnMSGq>&)_((F#0>ac1O#uuM0EZi z^GP3q6&71os#$pc635Yj5zH;htsBzV${z%RTb zrM3fN)6IKkPNP*-v@upLMe4lp2)GiuRoee>3TOfDqadjIX_H)c18Dh5rLC$L@c`>N z2$(R05jYsl3iGhcL>G8IW*x(@mhBe+9+ogQ&EdY-?=N{jE9i;;4k*Cah~577%pZnGc;DYse+9Z*q8BS?&IQV|cV2`gyjM}x^4Jdx+a z0=VgsV1*rns=E2XLUl^h^=?t`!Hdx_$|5(=vYSqW`$dCb3?h)%XvFmMR^t^GrhLr` z2&jTQ5r<7$AVv)!3QnBzLjXq#weF37bq!GVurRQU?0sT)1MY&TObq*%o4JMc;miiV zV&eW4MBveg1?*AywJ&_e7Fy$^L$bo?_KwM#o%UrKx*d8!iOYd0f}g~h z2PR%?I8!9ZJ|Uz$8Z9D&`I@SOm%!DbqC<%)LO+i=MMqFG3GtZel@u2b1<%oauVERX z4Q!|OcQ@(HNebNhA9Cg^}gjOAecqcD$_PKO~{0{Bh@LtU>NvqWWO=O46wnHRAL$1 zyiR|T#}9lA=#+EP9DbA{Gy7c>o$M{=_5k)WW!gYsD<83+dDtPY?_AVv~h zt_6DY!9`S7Px$AM5LLf>oIOCZkJdMf*-18aGbz;ufn}8PxMzoRs;&Fs8?>_*ryWVF zew=f#GXEF-TOMWGF{j3XStY112;BPu#N`=O>o6(w~frZTCwh?Jt(!u1%hv zSZGB-48rdlNmxX{B<=CI!+9>bXOb$>l zvfb?S56Ur^)nM8Vu^K~{TL8m_fRuJ8aOD<=nqxxWzF0={ME-Wr$iAl^ZkKcNo zGEfn1pSBfQ64b)LhHuBHDLR1phRx^iBT)AM*~i=8x0NKeVK?^v`812OBPHbfK;hZgdt!Jq2e~e?Nh+Orv;QO74t^u@2E87f6Y1EO zB3*dBsd+A&FZq=6Wt^#Pta$BCNuHL z{g>7eNBe;=-Qwf^2jl!tb^(s`JQRI@ak>3Q-r)o!QM)NW|L4d@`3mldXZ<<1^L0|Z z%>;I{4z%j}I*2FpgJ~`8Z&(tw+;wuDLHK`z(61Po;Q4&~4txmZba!vzeASu$#2r8&_gMdi4(hLm}2BD~QcQ-02(jna~-H3GW z>mHtZzxUaDeSh~_pKHMkGxt4n#TmzWoX6!WdD#a<_>}lqSXe~TQj&^TSa5GFEZ8Lc z0(ix%$mIYF3%}M>LPB0zLINpoV{K$=VTgq#_3C37u5xHIndkgzkHlSR>~6eD=`ePN zmt^AT&zIhd-^HV&V;cN;4`;ypHJs@+`)j5zWa9nU0nhXovpzyPJ&j5r=Hi5SX!R? zw)EHUo0uTP4UCS;UB6t6Z{7&1H^}xlJ<+}Qsx|aEmgn6|vNI3)7vXWJErtj5?bulM zSOv=*tt2uwRNuKd-y^?^ghmVhxDgu7UvF}I_A*7gWcUTF*gB&(q34iYmxM&3Lf_WW zIbXM6o56nxe}izC|26dFrQCI4J`oZ9_lF2x>7z%{Gu8EvKhzr(g&GQ1hTW=hA)rs> zc4$9!rasl`?n&k{ze@LT>dkTOyLaBv8ggd$Ewa*tAY}oF< z_#rn$7k@MXK zi^tO0tbFVpBgXxV59c^qe!tIQcu4SZf3y3@z(LyKrG*ul4W044UrYOKUmAp9ttTE# z9PWgq$v;Q4citMS(5bcPCg1lq#B2ag^AW4y`Hb7wVi;!B+fzm~NBGa**2wD__{rvx zNle%BEt?HDV%_vS)(V&T8*LnMBE>Xjydqv7x@3zGm%idxH%N`;MpDi2&IgE}>-41T zym0Uhrj9&?;ZEWUdtHQM(IH9dycj&s?@5h6%O}i+EsKjX!A5GVIA&rIZqkG&*6T5 z_v;xyMf*$JpTbToPq5X#H~vI_!KuQV@Hzac_UkHs{ER#^llBLk7QA{%-}os7X(moH zo=hy2x0#ZCiIMWF=|bu5z3?S2$~UQrizWocEW@-kzH{+hU0GdPohb`R3urs417b9m z$m^3JI(=@=xvS)a1ef1PH?Y>u=)dMv2`>s~33mynYr|zF%kt)F5PWuA@M7@F3jPY& zil_xmDaA-2O=lnd;0_{ZP-ulyp&-cGnewqIA=*!3#x>3doY8_nbH@o$X)4qG# zd(VzIkM12=9@!r8quxE0dQSHe@#Q7;ci!)McQ5}W(+jvQt9+gE@>syTx+_zH&+rO9 zIr$DgaAjmql5Zp%bxU*$-HEV} z+@Ws_Eo#zg?we_yVVsfst?c~B`IE4kd!TECYvqyU{s2AJr{_Ds0N~g@~@6_AAzIdhaL;6GBhjb4YA8LIFx}~j#M%5_1WfHx@&Jb)9U=wji z>Yl=O#$#sxGB0#E+E-Arw)?Fqt0sO$=?;F zm4;Gn9?n*^gb8TcUbS_yuX#Rcsqp~|OD7mDt&XI+rvdFTNGQ}~UvY4`IbOTxe zy}Z4&b#3|bcFb1jmgC~*rC`d1tE{)xP384}HHAD~Flgc%)gQFl>uMTgYA{Js@aZDZ zSJD6UBKJk$i#WWYx?1|47Vu}A^_zI=f6MeN z6<2&J`P6i>2U&!CA>V`Tg!<+kLvY|tq4!76~2S#|Zq%hB#Vu?K}!Lre@#daVY4%FP-_W=$o>Vl6Ltv9w%NSNmZf5 z;K!8$N!4f0Q%wHNL$%KJ+cRrZ#G9i|_bTS4t)%ZsbG#pW?|Ad#rpR9X(WP@r^2|=5 zQ*ny#esl8BZ%ZvnYChJ~78DhsJB|*y{+P7X(8QRvV}UH<&F_215$qA=To3BKrkL^Q zHCRpC66pdJ{Cg!2eC&`TBIT@yJ%tN~X<;mESP^c=6IASjf#km#;V_O&@uYB?_{7ThB#*XGqp93gMcHCeaCv0)5x?# zE6uu)+Ir#Icm=s@-l2Q%aV+keYg@a|U37#WyX?&=>?CcBm$-#jc{jceeiLURnsM}F z>vNgA{|UkQ@$)otH)3~)Z7%2@v+o}-%a3#YE}EJb?8wp`tl}425gOgL`PrJFmZ&y; zW8Z`6Sm9)NW;;Sm6L$KDm(}8%ux~%ZwbW&bQ}y z>w|cYwdXDD(^qLOYqwrtP0Dj&aRyvKYF3e9Nl0cirOC@-U5`3vVa(X$@tVH5-9eBH zZ}Drv!mC$~^;nbWb9<>moSfr#I5%=grAtXXTSa@^j*_3g8<2U&MrfhR{h%}l=k)q1 zR&?sc-qRrLD*)BLHdL23l99n;2G8MGuyd4HIN;ei@FRMT>aXYb&oN=0$Gnb>h2?LG z1^eR~S?~z`3jsgSJ!g;S-}_k5fGMpI5`Z?8=DS{mBvA)PEOtZW6HM6P455CqSmmpQH@F_+kxi(FTikw;2c+ZZBw*tywn zUl+wkB9X#221bI4lK20(9sDJ7-NerBxgZCJqoX6cBNw~1jWGwOfPetU?K>QI?y!L? z*leAx?DU-2tZZ+beaN4mBWY->Z)5u0&eYlp34N~KGi!T0k?Yr?FZ%1>**FcIO#l8S zE89P&1t!P=y~DxDew*X3&jz;&L$3ra8(3n@j zN{iwPbNsbyqWFBq^g&=FsZAy2l))n~Gw2^Sw7pDckI?fK?q(f$G!~XPmb4^F+3DQM z7~aw4zRKN=I87AqJ@M>^1b0>BClxd`%t%YpRbG5u(;CB3TdLOyD;*}Yz|oYy6`+|z zWj2=Lt+}019-u0DMX~}pn)n4tPKrQn?0fVWPZ`WT-jTl9JYS&`*=F#GYrR?QvQ3q3 z`-O?)HOFdRd~7(<6YIZz^(KEi-4{D$Od8Lt+|g{{zI=8ScOuq6<9*>}YA)ywPsaRn zC`7RL0~kUa>)ikRWhUq8mpd$E+CO1B_w~kN5own=wx@GqE1(M?qAjy8KMMMCt8vf^IoSX7%J7TWZNdpSWVA) zBTeyR|7~B8@-Eoic75XlNe!4wHwd1)y`2ab!CXtP1{TMrgNR4szjp)dKpI#bfmiC7 z_u|iiq4G{j&xc%s=G=`9Lv>#fM*Z7;@jEb+h0IFZ`+=BCr^qJ{u2yrpV3tZM7c9<9 zER~txf1424ELpHRHY5d@_r%ZRKXjVl{?8%*&msTMA^-p7kfu$M ztChRs{W)5EQNqrJZYTTmM5fWb4x>ch$ABR6JGWyZ-m7SnBPCQLECfjj+ygyYYs<$RbyS!h=ZCgdF z*hJx&hV{Wq%tt$Uv6T*fZwRK1Q|QnS7U9*8#QQTS#Yci5t)dj?hKcU!>P(RrS_81K08<$!~qQV8^RBxwimpddRJt?!xsNCv) zv_(wvUhe`XY(|3#O!XR$U@{{dDNhWny(OWE7c=`M8Sp)oF$zJrn8_kQ8u2wLu&}^Z zKp|G;D}5H&M~8}x8%sTSqglPukpX7X>&Rw7wcZQhauXQ1I`H`?f%|83Fw=sdji^WO zHR!XD*C1%~rBDTW(`y)kgey|bXRMh=u)(Kjge!fVNVZSIq30sdbKy^q;$<3nOBX-o zSVoO&rSJ}A4D?e(ovhS$tb|?^Eixn!Ee$4P=aW%-f8%*A>U^kD1c{l=frMOGK5te7 z0|;U48l`XkNSm`HAZB*%n30mPLN{Dv*Iq-?p48Wf)+_IIZcE{WNTO&CBo$DE7)N#V9W0pzW2R z9zP-~BN{QY^>f} zVxmEB0ziHCPKpBbn2Edx)B5zd=>lfPX6i^zRLGda$dBGZt>(hUYc7La9l}b@+J$*= zugT8aVcHH6l^YFI87WbG)>diJHm_I9*Q?q~FzepK<%0?#*Kw*j+0OGw2{NrE;w#pv zL=O=Qtvsc&$VV;kFVfIUDt_j|$Mfd4?rS5iu+v@cb;)||WYQGg*eGI&?3`+aPAR2QP zwcP9`js2du69WQr_TpEU86(k`*>Rvt*ws_wu^Wzhn+#$Puvsys0NAf*S+{giiX9 zbfi@9HIc_e2l@?fGo*GK@x4u&E)fZEqPI-6N-Zd?Uo+#)T!tz1X%?HjTcR=Oj7POc zitf*Z?u1nz)vM-c&br1#l+cC9AtWCZcHO5+LE{d&AFeLaB=gEzYM4jJA(9*8S1sdy zo+lVyFZ87yc|Z_0lIAroEVw-^SVA-KoO{S2J{C3Z`pmubo>LAX2(BS6bL>pD{jJrn zZaA(O!B8#hlmldV-gOLG`>ShZePj-`C>K8mYSIfYe8nq1=EVpM+&s1o;xk>MnDIPM zEF~BA+hFpW*J8pYJns+Jo68`0JPq&c?Bk!0ym1do8BGvaKT&0U1ucQR2lguh?*ON< zGJ9UrZu%%uH+1%t4nk5$qY_;bNX6Ga*S0^`II8a0-oOR>P2}ywvb{Do0=#%`g~p)% zH7T#&cNiLXBj4;7=L9mq(PWEwN}9W0KNl%gMHf;cBjT7o5aio4(VVqQyv^EeFCLbv zXH;5hbzV@bNEl5S@uKl@6zBfO20;|5sHQiP^n!K%*dcFYTK4e7$5$?4tCj2q7+jA z$(LilBJ_5Oz9@E-KL+-u)cqK3t*SwLj6ztdAAa3z7qVRH%`hChNY%CpZd+7t8 zMgRG{D8CoCQkkxFufJR$Y4_i1`zk2nGege|m8c9G1vbRsQNYve*X$F$kX=zNQcB_m zp|&p)$l;49gj^Jc81Qn1DkQbnvzVAeDGhe(MvP?;DHL{*Od$)%0A?ZUabYSUhwPju zU+<{zzEAT(A*_8kD!1OtMDus`=d$JMb&e1R?-N|P4BIG`T}Q@B9t)F!=RrEPBHQZ| z6PhJvvp*h+EmKljx0HuR5ukrL4NwD$Kic>yY*K(W3%NOTp3t zFJt2T++!i*vAW`S!LxnKQvHFy5|uoW0inJTh9n?y-fVhlc~k4^%-F{)3;!KU37%X@ zxTFvRoFvis3~WH)!Uk*OjFpgpYTKrsb`%EE9C>)v;i8l~W?REX!a4OVHm2$gMP5KS z!=U*+1MitHYG~{>H;iJZz);2awux;H7Sl#_TvlgCYn0?f4Ee^Ll5#XlTJ{CGLK);A z0&lzvp^>;3eK(om@6A0ro`{VZiR|6at$&TyuJqXw%9WcbTXLQWvR&iQ!eX#_K(vRA z3IVJMsjvYs!@AsGx$H+j+Z|jxR(}thr`$zm5`x?(i${R5uRUOGzx94y!6ZVjQ&4Wd z`;3jrf;R|yB@vR0CjW4!p?0-WWPn2zm!c1HT=j5kQGY(x zCAY+qs>;Xq4~L2puuc8UAB;!J!~;+e83qyUt3O60lhhV%T`jI8q89uX$wNdn&HTuS zqvVjo2!1C_=GG?M`$J-5*;g3cdxj))b>QMq!kfUWU(EY`(hcB}q;&B2+jhUML&Nlx z@bqKH#VITk;w^&G$#KXF;r7rX-Q9^dY3?@BGSz#(hypZ)Wt>8!5=NO`OPaG;uh-YvDu0 z3#XayPI)9#i^|a`y17Nk^&axGWD+`r++)2EH@|W*pW??1%Ts+B_b+dI~U;7N>#J--O#Nri2X#3At}hXZdX`^ks!@ua5B53ID^NwlwLu z01UT2(3@aU({WB&tv)^JuXJ&=Ry|N4s=v%A%k6uQX_lPH7$iS{w46vvcGcTK=-WEb zTA!${F6z4PF_2evkjbW5+@uxVNQq#ghm*`dzpbD=x8v27rEHi6YL!m?L*4 zL^6tOQuARp7e+z_>$-irLwNHo(KX@0AdwyQk>|BUTf zfF5Z#YkFnV{$X``-hZ4cwz_9DjYe(T2MaTYy?>ZYiI`zN5*`u#aGg@FLi%c*e|-^b)?7=J6< z=TSw)J(2dZ0!s`Gn*o{nW3(+w2IHFVZvY}sVBR|t!&WNm#NYHrK3nWe@X@yGy4Rnf zlorx`BUt!ZD5G5I_VR$gPAl&XErjH9M(aEQM@8pwiFsH-L|*mD@LI*Dtjj{e(x_gn zLWE#MfDrt8DyWc0Iuj%;KaV-sX5MnL18{UOQza`pkE_TA`voq_jvl}r0BEJRiKhe$ zQyzsy`n}XFG74NaoMRA2P9UX9GgT8mhluzQ=f_-JT+D7bCYLK>D4MXXYH0foLeQ1D z`fgu51ngE@7TKqEGVJ}4qkB6YDYQcYUgd%p8LnPH8lk2C4E*F#`ilwCprjPFJpq31~xHmz(qy4CLeiT4N~ zzxEjU!(*AiW;b6-|A>^@RQ(Z&#u4a`F%>3hX954r7On`u0Iqp#9$<0p^)ck zYt@9nBNfH%r=BtP{Tij9N^&Vu3_j3v82$2s*SJ0U?f&B#4*5l7fEpLkW!nY{&hg(anKWp4C;h}yza9OH> zK<0ABuO^g>rujtmIaS`{sOet+*uu}Ba-=v!pHGSlZMJds^vt5~*vv>p2{~GQX!%TX-7w#O9S{<#vBU&##4~;9=l* zM#}?v7c1v}Fmb~Zx9!3v>dAF2EJsSWc|*}h2g`Y@x#eqxx0F&1+KjJ@d70s$l!bi7 z+Xd#S(YVd)$|?pK;3d1~OEk%HaiXQGr6YpDA2ESBO2h1GvLZ<07rC}#L!)JT?u!}` zq;fa_zdQm}`HXe_dtbop-x_6YHiy+Ba}fbmn^}y+d3&|UkAz+(r@%jo{ugmLKGg>) z=sEWzY6S3RibS6ycfy=E=Y|Fg^etywKAsFM&esE^R_J|R>~z1wvV8UX*ikiLpLkzP zNotilSfa&*o7v=lfbA})JzjW_F^(E8Hk}6B&8)oj?fk{$G-r}LaUOt>-eow3oE#sH z+>ElA0&iDs_iOP2_HK(?E|W$6A!ep{_lOL+rh_7Q^?$SU-OKi=ybR0m304122dnP+ zn4eLXeJ~egJsfc928mJ}I=iV!#IeMK{kSQ$qJ+uL398vm)5VYHNqtedkNrO$kv$M}|`%3`8k zJhyAeUAmLIh`PL*O;?-?x}|ycs}lmnd*8{)M|v*>`_6GC-G(xdbKgjJ&AQ$8b5|HD z(}j!_wg?y@ZI{&&3cZkreJvx^J8DhU3d|~EJ3dg93K0e=_`RcS`X!7}Zvz%Y^^{uv zJjSY*@Ootsk}oS=c1t%q#3rE9n650>lyF@-^ybjkVh`6GE)~)1{de{NE_|kw@Jsd! z{blXCz0}8RHOdbWK_0vvH=y};g&-f$x)sx)7ZOtAgR47l1nAy4nU`L;B$-eM0{I9? z>{rlY2-vGr5&2-31;_8ZITJ73?)s}X>qy@f%1fevp_Ud5$X+mknccIv{vrzj%UHX$ zfy>!jI|-THYt^Re(`P4vid`**nAVV&B$QtRdc+wpTS;;gu{crF^&yC=V~gP!Bh=wo7~_7R`>Gls&wWeIjp8kPU9TW5&DVF4p&0 zW+prePVZ-!_zq!uY~p>OV$3|a6%TC|7B(Je3qoh3V5 z%F0vx1`P67!9$fhQ)21qsYu6`Bc=umF&tZBi-37dt>kf@ser6%b_6B1P2X7<;@01=%irZHzf;lBZ`OrFuJWn-y1{94xO6 z8OQzL*PVrXOA-+!>736kTVjq9adBu6jd)I=d;?y1yfZS=;l9H)VwRP2_oF7D#kJkG z`*=N0jylaFFA;h<5nQ^kLI2Pt`#wUl(sK)x%ibb~3sPIV!ukz_hm$@RuU~wNl%gl% zIPw6VwNXV1sLth)1uaH+x6?Wn;JqE&1)hPC#rH$IvdtxReAWj>fIhH4^N1wG!K9OV zmEn1(Kkx@DGHmyty4n@gv%F@#>Af*WC;Mm}P{CF{h<2KKRXX8*h{G3U##%M?nx&AJ z(qGl=7X_60$wQ`t#m0#R7veh)8PqtCf-(wpi1jNZVCb(svJDWD9<58~Bt^VnF1}rt z43=@~JYK?3N0h>jc!X6Jy351WwhJk~9PFhTH?9%=Q@QvE#8CgD zCQitn0vQ0B(T}kXBEz5=F;G zOP!v#&wP+MTAxb^*uQ)$b6meRyk?$P`Dsb|t_WQIH+i};LUMkkVzZqklN>=K19FjL z-AiB@Q`cBXVG8TC5OkN}8{<~)az&rp$diP^oNd^*pGg-BEzxd5TitJkjE3Jf=e_(Z z%oW(3%_P%5yj;_-!0Bc(s8@{H;A1F&qEU#^z~Gp%c0k;LA1 z3!OFWWeeArdoz@zj6ogY@PlU{+|TzFGstFwj%yZBXecosU@sm5sd-S_am;W@#al64 zd=!Pi@pC`ui5otLOD(Sq_9N0&0%WhtU_bCZ-U`7Lb9Kaz9^A(S>UeR0E-e;B7;%i# zQquxm+sGd4uDAw9?1Q}AJn@==02D`xUo>*^P>AII4&?R&s4|DtAq_(F)Mr`^E3gS) z1J$uH({jf#QANX18uhsgAjjO#~P1ns3Dhoh(<<_qjSBg}H!*s!yRQ2Qn$NvI=JWYn%<8vCyF~4>G83TiSuMIK1{DJT2{^L*2f)0R zjYAF~j2sB{^QU8MgU;kGpTTB~-e>i~VB~6S9txPi-lWfJxPB1c_Da6Z>_dp}`Vqy) zk;nDP8h8h8I^V0fqZ!78Rd$d;?+0JAeDsMGud?Co5zZ7G5a{EhvVx2BBi7Z&ch^a` zhwoSpbC&~@WC0SZ1`8!Awg$*r_KV#Nm1|3VSrRq$l~PyfWj^eN@_Qo>gw{Z3%S{QDT?zM9{rv+8pQ@cSm3kWY(?v5FHR3zRf|ZDbr`Ul=kU~}HYXTm- zc()H;qt99NWvW~!mYWdR770C5ZHWJW)wS`d-bl*v>1f;bf-1K|zxG(~t7=bP__FC% z32$lkzIFBf(~XdQPSm48*o+=~wX}PByl|yDyRSY-vtP8l7U8CQ@lTsSlOuA-NwFh9 z`I@1y0KlfKwL7SGcV4jSc@-4AIL2JsP`mMZr#gc*n5eT-0rD*4z?G$ebu!l! zX<;I=BoNs=-daqs!K5Qjg5JjnNumRCZM%2&OTD)w6X`U#x!?Gi+)AFo_g;|w_D8Ww zKD+{|+4q#*Mxa*!_1eCO11<^JJG zTT3eGrX4PXG*5+SyL&YJ4>@9u9vM~716MpNUEib-Tm~i^PW}i(J^(6x7K-T=0xp;+ zE>#7Lb|&tI8K$yB#EHwmP`%#-UH5)<=2MvsRs&MOqUFP_dw7Nf-h@|ge`^<9W*u>w z4&*9526=IqXW)r`Pnz0iJk|D0dwC84VZHL(Ds8L(E9_-Lm`d1}A&YrKw?MkJdEs`iDqt9yz*?DGk?E6%B zU3OL=dg%)==(X9s>EO@jiTRPm86Y7H0(Mui68=)h;J!4(LqBOKYEPockYyS{PJq2;Q(#dwQ8r#&5|$1*PIrs`g1b;Nz|NOM|*bWE^lgc zaOS~*3Ezt^csLF*GpU(TI2`3GcR-``!Sx{HnWlHVj%%Zmy}l36RzeM|#9z8Cc@eM@ zeMQC{n3d=wmjd=JVYc#RI>_`Lm32iIdV9&x?Qln?V>7k_0JaSA`x%a#!Jyl@Ra>Q& zTk-qhh{x%1j?3;E=$K7|5M}9cG)=qOBPS*U3fz6zYa(xeq8r;q4=v_@i98BpFUiWQ z8lLOms7SCMvHZ+)ys+kr@}@)3I0NG2(}q9l3U=4wliGLkg#=fqU<%f^j}Ld{e4V{r z0+PrPr=oE0J)X?Xqc3>=8vePfF+y=?O&8L6^A|r>{=n4HQE7Y##z)dzplMXs*dhRN z$XJ=RVStbQDLoqkfTWioIcS}?%iZ?($Ast)!1NOb)$uS1NESwN8HRT)gB2_5KXisF-Er*zXDKYt8B{!3|upl0!(;Fd;;Yzo9E!5qpz>;0P|`Po&&h2m?^HXZ$qgrYB@@_xdB- z0E-;k8PpG!_igDTkTR*&Fq#>;A@xb-m*PutQLu&Bg`j(;1xQy6%Wo+y@oRx=8mJ-& zE6ExOL31vhtV+h0BPhx#9E7EtHnZf`9`!n9l@$Y*t&QD+*Y%HsPpkMS&`-+Ea>m@- zk5d!$m}VgnhK^k6r&~a{(UWv3^#}z%De!p%MoJRD1BKsX+onqxulxXYGWTVM&iMUR zIK3#m5JV14S*=l+ypjpJsr7Hj)9xZ$!#Fq&MTg?GkRVS0p-hQwyFerFm|aijg_bce zHQY;p3X}ZyLC*oW6y*%SH`bsqCtQ`Oz5gSevsM&9kxuiJ_NM-#M|7wVi;-I;LD|-y z%mc1cutijI_X$aT#iqcI1vX%HKGEeym-`=}E<5eR9f)h;Vv1@7GMD!?U5Lvt6y+4R zZ3qY*;@XgGQ!Opj=S<|Xy70x^UVn;Qqc{(%%xh(MC0J051 zaufh#3RXe`%b=d!5{|(_X98$Umiww^-{(M{JVd}Y%s=cpDF zl^$^36A9TeUpi+nFX1xxq>`kzdnQ|wh6893g5<2Qi35>R*s^zhv%ufqCs&9u5y1fP z2M=!=oX12+dr0L;EXVABCUyEO4vK<*v||q{E~a+kfPU=ie7n$kG2pey$|yc@`1~Po zad>RK2&P@~gLkCOkC5Vhx{i1^sRbJ1JQT*~YmiNVgowj81`K!y$S0SEHKHp$O^kKm zP*sxQ#`2tm&W}cFhq?7mm5`i?VY0wX7CZ{PTCxHU_Kpt)kWfvr2ASlcQrkI9 zW9AASm@1rG6|EfNSOTG#I&b3l%t* zd&jWlO8In?$N>V&w!NqTCB|&O!$?SBL!h#@=uKCE8cpAW5}Q&;VH)KQ zIodDkNlPt;3a@|uZZ?Ym!k{N3kSOhs>f3dol{D@Ff?o7DSdmY= zzTg7`<@MPV+Tq2EA~Br7uM7aoE#2UpZoqZ~By(#6B^=&JcG5XaEeHdfi2wRFM&;+} zNipzX90y~$fvh3n$R>pcG={;V2kCC0NGBDdWsG%790O!bsnL(f*Q=p4UvUa?P*o=f z=25(skpPfb7Hl6(=Bn&`g`o-$Km0ZWAqD#?ps$NLs*egkj9n?VQVy>r@R7tpCHCKd zDIz?Y3C#CC)r74Yk8B+{G>^0xpNAs82-JaJscqA__gf>(@b%^AIH=bnirf9y;ff-M zPh4)o-#g0f_;Zv_Zs2i`7!oVoGyPkj~yH=+j{K1$_yClw=)ofuJcE z2pQ54Az%k(o>{kRq_7MF4rK0)*ZB5Kgo!`3(K{;l+VXmwp~Z6`f)nk!q}!X|lRwG&7JYhJ6(N`X2)Kk~>1RG&T>IHv!gW%V`X znaItXKb@uGxR6#!4)<~* z>2q7*G;tZ3?TqbDFh49*mOK|i&e&1R7bn%wsbVOQPB{Wcg-{CV@u+HhBY7OYr?a2c zG(=1Q9Q-X5PXodE&*KO3jQ}z(?alNLCzM!?4F~a0!AA_kv@JkQ&FkS_NTm#TtwZ7w z!1rbYkk=3T)-JOj)mD5iiUv6P<5-9=6TUaU{8HQoPvvBrt*nI4_b!kFWlqWme5VEF zhjmeEg%vH^-k0AWje{w`H{Ib6XDWj7zE2@WT?YazQ#nwn88RCBoW)lp6klbp`vXh^ zBt1}fDsfT0pbOCb+&eM4q6tjLm!fn`MUN-B z4vTLc6UQ;`q-D^`Hx&OmX(h z3mZ5fGvaMM276K71$kW}+KBj!nSnaxFv6lsO^}!~OkTetmWA*xJehMGmzv>^`V#{O zkn%Dh>ttqhRAD~lJu_f;GmQPv7!Oy5K84}K&;xC7lqRFKN3$1o&ld&vx>Dc#5VMGX z|H6SZzsRd^Ff||dOR({N{6DDaF-`^5zwoO@H$onPk7G}0K#N?xKnLqqpN~tzh^zjZ z)cod{81))vv;Xl+oCnH8 zsiKExr#B!`Kyoa_k}t+4@W2jylsBUNdzSxrRtG6LALmy#CFuyWQ}%>Xej6m67#mY5ho z*`jWkC!46bTHR`g!IyQhC3Lh?0?MBQ~WWAul2B2qq7=D6Znx#oG5cJ#s(6B?UJ0+ zp!w$z|J>34mC*gqi~r*T0wC(=Pw}$244{6p7keJeVg~T=3K;8kWY!<-2{ekoNBqC> zR^0!%I9xO7thKzQ5^(uFY&`U(WhhcE;Cf8CJ(bISj7%Dvs81kmC|JU6C`jC0hOUR zPyyI1?mN=bja5JWs5A$1wJm{`rx-{v8pmDME8mx@Wqjltvr}3guPiUI80!ASuHE7o z_ppZi3#~U2$Gd?4zwG)7$8&%Y$f|8FnY(Y)lRH3?{+5p{Z`7kymNX{CPNu~yf<<mE&!@9fXV`Ok5k_oBzQsMu&IU9UPE zjz{@qgDQO|$J(t`;|C-N0_==Fq<9$vJQU!$p3K{{Nn&~J!I<6LsD+*KAv%tI&vVOp z127zEl%A(Ru|5Q%5S|-cNt6)05*I znJC`UnhR9?EGZt2i<)WX<1BdTB#Ldw9T_q1+X3QPV&ZzFhaD=hluK9jyuB{2P#>uwgU z%Su6p&_fB1+vmMtcuZI};HXg=rPY%_+OL_J*X}j|;XFsx-m~0$&LH5n(916_PeX*S zcA4F+uRCUsV}OcDCRSL0^d^h558miEIgbOSp6v-;pvyA$r?v_3AnV9tioYgy|Ls+U zUKh=CSiU?)?a>-LH{|86cpNY5wpNWzUYOo|p$q8rGQscyuLr6~0T7CI>!5nlGyZrNz8QuaPS%JP!Zrf9}MU=;)6TtqkrS0?&tCBfc zo-LrZju!c_H$VrU8kiLLFTX+}d_dZ}DnDeP05ZetXDp@+XG-jkupL0XD#Meb%TFgx zk9CD+KPXv%fG|7CI&wFD3Hyf@PZ5VI!@Qd|kY>jy6TNV%!66U#3(N?+3yvupD^{vT zqn9!av*P`zlhf3AVf#`-V4I$N=&*1ybBph(v4Oa>ucD0_&dz_@Lf+J%0^44_~ zBxYmWc7}FN4%ZKj!)CvtWz9hJXt%rsYyR_4>_3yFr?@GU&^F)buZGHr4tWzXEH&%{ zs^^Wq@k2W_C-<@HmxrtpeUT0#rrOo9s$`+Kz=; zVHd}?v)CtW$L<}2C^ZBrIzg*NTG(#srFadtmYq2ek&iWgYSS?b4t6P<+@4ma+dG~E zl+hb~zD3JV<|g4=7u@8yd;R=ANImp?Oq92jL9gx(jBz9_&Yd!vCh{hnigfydh(|9@acJCIRrvl(JJpT<^RhebT20uvB*|ubfErt>~`(YVp`mqOe(GMk4Tly ztIb?SZMS*8D2!rXzWF#mSG(M(Vy|DzI${kwYyK`G`L}TxWb{cp>Eld6!oQo|hE?xP z5%dTo9%+dOYVA&VRD;0PCY|UP_28Ow-N}B3N2H&X{EZwnyJN+G8Nj2Kw8sd!6iEbs zG-<{acDZp{-y6>lGgI0>{b|CL?Q|XWq)iO{} zFTDUcrfoP}5U7=V4BB#8+-EzOxTyt4@w>~8hrWEkAxJQk0RQ9v1lss4SP%08(SR8zS~pGW=PSaU$w1h$82-4HVLGx*HgIU zq2>s;aGM*sTuCsQEA@P?tyOvVzymcv$5}3du2qMI8e9B@r$pyAlW32dhQf0VPe586 z7CYIvD3W7F2Uk=%QFq9_W-QD9^JQl#>A1X{SIxJ(CN-%PmFVq|9H#5GAS_W-rX6~O$rV`6p zYK{CYki5WhxERk;)9dHyLEdTM>B;&jF=)vrJdg6jHk+ub^l&)X*D#JTd_&G5D@VR5 zd9%wczNbO3C@P7{~Eum!90+P*wIh7v<+^)dzifD}X!KdtS=+dPnz$aH~_8;Rn~^T@PK(eM$AyX;g7updnHlw#j=dj^eW{XRcrcku%Cy(RKSfY zjeJi`hA!IJ@c^NHrH!kD_30zl*W!0z@2_FMQ_KM}SVy9QOFV>cDc^~md?~BU0a7=O z%(y}zdL8+2OX)7|)^`GqLC^xe78z%GAB4o!AfYulo1)Y@f5J!o$OOHIKvX~9+26N$ zkM3vKyRZkn6QK4<-8?P+$<|k_fl~ku1OeT2Ciy?zJb^`94hJxb1!geV`bL z^Q2r?en9!4`e>Fz!bJsZEf#?Mxk>|x)Jatnd9H< z;5DK7#4+A9hxb~7Za9A7mg5ix3G1PvvHomXfA@ATFapqJDF@&TFY1P-lI*6m$>pl3^@lxZQ3HS~FL z0k!oxw_4(Ob(u>Qg?l`RoK&Ds1lKbVlofWe{ z63{EV#r4{&1~wmLTZ0I;guVj^-CmxQ(Z=lOo+sMz>+jK3pt-Dc`&+r+qAf>mqCNk z2hDlXOds@0I1q80xYKa~sOh&Qu$8x46nESL6OzF6xlLxcW4#$TwBLnj_9PS%$dUN642Cpx}n*SfVzB``kzkmNz5s_@l zJVF$bO3BPtDtm)_aXoXU){IoUpBAVi{D}8S43)R$T9LkHu=rrb0a7T8!>A& zl-B--*PeC5#jNYbP6*dFCAHhRdLZxNqUn>edS=^M{mdUVB8b8YP$|{yL>VGQddlbb z^JP!xj-VUJw*+Ac{sMFSp;Ps4*<7UCwl;1xLmGPN+F_=aev9nQ>rJb_K{ya+aT!A| z8$l79ZO1=!gG$$6X;g^cSHl7=l!REQHaq71!~qjs?!p-DFG(yN3_8!LKl-kG-~{4f zr41#G@E|=bUNU#CRCg@(CDVVDGaVbN4Nf%4^EDiT#)Ct;7D&a$YJv@i8g?cN(QH%j zruA2&ym_Pnd6zvjnSfpcffQEF~BaNfoI zDxr1sDW$FfGy^(a{!Zzi&hJ6x$ADTtL8xrhA5@+q8E}Ns7AWR!5boAikKq0GhwNsq zZ3n*$zkNezacxMmg*wTcBtO*~An(?rC%`zRZ&slefjHc`OP-837KjBD=x-EtzQM+@ zkyl1`(C!6*lRe)JL#U|cwJc}R)H3Q<1zCeR>-vO-8n|I_X z9b&ne1hN(W!&rP>o~_YYwN~#AUb@ZQhrM|ugJP>MD?e$oZaCzwcz42IO5&N=x<##f4dKrOM4%^;*Z5<@Thy zd0wwu1M&2)bLVMat$h5$@%)!_Qb_IHD5#JN>2xeP0rh9vCxj^nI0|ydUA2{09iQ{WjPO zlYk3jgV!qprsv=#y7Z!tstgvLNftg`d>$Rl#V2c=KT2YPl|Zm)*zg@>Tpq0%Sth7~ zt(}wGzPaHa0Qdr_dH=~9@G8BK-8q9XQUj4w6Zlz^yizVlcEc#T^DI`^h>YSE+KN$Nibqy1yTiiUtP2+a5|5 z@+HV(i$*=kWHz-?Cc3bsqVJ#6-P+zvriejO2X^77xz2xrfpVkQS7Ud%Z_8^irTsvP z+|5$oYewR|=+A<88UlpjUG}8c5v$GK?qofTRM_3I$a1E~ZR&H$?2{w9Z|+FwqBRjtUnsCKnY1EJ5(7J9=*PB|4+amCjUD~^Q#ZHUY|NRJ(>rz z$wx~?W0!8g=K6O0e$hFu%z2R@XO$DbjDW0oB=q8c#1LAMw|SJA;A<^o#^&m?4~tZF z0Cmg9vLt3o1(CMrPX}SD3I1QKv-K@mSlQFY-bY<-N zJ2#n67cYC>UT+U)h!LjEJ^32s-i3*dPsB@_W!OjcQVDvUW6zAg*!hj_=KU)1VaI=9 z76)0N`z{BI1uA1C**zW{*k_)&MIOgUF;jUI3%OH4(X z**34QJtC&11A8D5l5I~rST`r=>Grep?6|eVc@1QsIffWr9uhn+OKe}?nNdP-N*r@k zWaV4&vCV=Y<|W6q+s)o1YJ%fJ{Z)lMdvxF5WeY6-jCbSy#@tE|<2@}#!m z6%AGFZ;;p(BXo*2E0e`c3c`<#4rX)pYmF?5(~XN-v5M0@TMk;3q|}4~CF!a{+8=+J ze=deRe?bexpr4lpLzIS%@-&%dTFWoj#4rFkW3h%Y9KOa7k$bQRLcC_z88{cbzQH6l zkZ=3kuIdE}6Bz`p^4a$_F6VWHSY>LjRNZ>2$?lso%mD4=(+A@?%6iA0HdTc;cfH>@ z_28o`E-xGvHEB0mab&I8BpcKz<5AWQM@|_eG*f!(G3!AyVFa#TtT}( z>~7RMd`k@*;e5QaoU>TJQk%RsJn9*FTnMF4FTgYg^1*H=RAX0@*UBqp<9j;BvnSbn z$C$W3<|dYh%k11WWU3haVQrysX1V!2YObmUQ1n}%96M=l2@uO?c}Y0*!A_Gd3ipaH zBt+&gnAgYxBZdFT)UvBt#xD&M&|x5_B%L>r{?`0S;?7}w*~Z~s+g>Y^x#6-vjaB?P z5GOyj@44GrKUO_|h+p+AJDT!0SFE}MibiKhf;nK>de^6zGeA*xqEiQZ7?e{Sn6l*L!LUOSdnD0gCPqyxn7 z*~}U4#0rkR!7E(jozg#}umoUqGzs2hPYGJ$h_is%KgzhnrP7XXZw!u<1N`z~JEIw$ z)dw==JnkQk<>1+PhJugORnIpB{~Tuw0f=b?Vra*c_^ikpHtJFrypdAb<&T?b_OZ{N zz1!FXQ3N8pH4x?Qaxf}jE!vK@`U-TTTD(M>xk|RrS&H0I(vLVJ-WPlKrq|=*yW^>j z(zGmaT`o;aoXZTn%<2^v`_4l3Do%H5ke~`?m@=iE64 z2%#NE#{*2e0gg#9TrhYQA=>k6ind%m>~>aTaMM@+jkpXdAjWh%kscoEFn+x>7Cr4Z zOFZ|3ne24O-*RQu7Vx?WYxOWk+Ol`W;oKuW>g7dt7o=@yv5c_ z-24{Mlk`!uw`-sb`7BqwE9&D$%u4?8qU}AIdG^r>(;D6d9Pz(vLFD2{wyo9e!jwy*r->555 z7O>4Jh~oAB_Hm~@r(~JN2yVpFqOIMB?Y1FmVXn3pxvKx^0BKpUw}k=FXu#@yqW&in31Q+U>era+(z66A>OaSge{cEB&PJ? zYr6MEn;(pFRV2-feQH)wn^vF4o~7XI{M?%l?!lgZ7UqCoCa$h-Yx+Dc`LO|XW}^CK{d35RP>w;+`dnCTv?*CEQ+AQ>@@+bOEa8Z`hDl}P~OPYrqa;TbbhKO z*Mxo9V7|NsVbmuSaN+vVjXT?5+k}>nHAnP4R3+QnXYncczpr0Hz7a`AGyh#!fcWJK#A~m@#YoN9or=U##ebGO66{iwMVK5e4OXPJ?x9+lD;&SuI26r z;Jgz}{_H)*fS8(n_TnV8!-?=czltIOZH7lps<+HNX5h%hy!Di?L;GoD%7{jVmZ?>ndo(JjfPifs)Gl z(f)*-#;u~gQmY}m6n{w%_MkmOH|~#HFDw);G_;jF5dmrRt$%J*^R!1dZ%$Un(=`JHt>Qg)Vl(8F@TCuJ z?ms?_HcFcCvJcbkTIEy}k;?QP^{j5}PsbkD1V8F8Gz5Fi57yv=K3F)iz)~SMcwKZ} zj2IRwkE$e>CuKlX>J`0iX?(&!@$v6(mHl$Q#F^g#Q&C)7qS!6X=Yrl4GifJ+8WorWWl#)TH0BA3o%pHfbBd9;h*t(oM3P z@675A2s;XW>DD3?*LVU&9$X2~^{dZR>z2~5b_(jQGew&8WiqZg1e`v25?TCr(cre_ zgP*nl>b#zm>i?Yn#y;Icg+~_dAE^c0@Mm*YA5@5b&#}y?N-n~A0Z7sfDRKkT@;_E=q_!3=7RUNqwDdRMfR`1GR3rd!)bL{@T z7+I{s9vF)ot4k-8nb7_td+P;^2}i;B9*czkOl@vMjGqIM3pT}AiRza*uqBuP7=?k4 z^2yaDM+n2^yp@lkH0hs&NV)tYcyNFKZH9-mkP|^Q;*9i+(Hpm+v=O>rR59z29mdeS zW(#O5H=AcfX35ZPf(R{(JpE!O3J>W4*={Db^8$=rTE^kau9a$*zY*=3zTr9-^AX0` zkGXP6W5Pal^M~B=SZMoH{3`C|rDuy}d75ydB5?B}myUYQ+(DAJB)G7*Rr#o56y4zR zBo>hHEZa3A(mUlbcB>ABnZI~g3KQ3gyV)4D3#>zO?@DLn0TG-$W=S^x2Gu;>b6-tF z@7HeA^!T}*M9-O*_1`$}w|^C?9W?_Y(+Ij}UaR90uM9sv2V&@%gC>*#*5fN|>l|Mf z{Jw5p*(YYbAKgV(A$3N*gIB^%a(gL^DTXcrz~gRE`St+D?oG(p$6RZ%+%^%KRD1!n zeYfYH!b_?Eo#>|ciyfs`Jli^J2kr*>#5pDVP&JQ z6G+Bja-z?z&B5DWAj~OQ4+trw|4Y)-L2-x~nlTOY8iB1qv?varK10+xFV=%GXQv!X zCG0$;&o7M#MwU3aNV_*RE%zJ4@F%x`hrtW3KEM%>YQ`c zRJm)Ir<@;rPx3wLK-?TuumZ6^%?eUNA%`E$l%H8BSRv&w{VB2qa06*?o>~s zar#++0J2^j-Ud#{aK-wq8Pu$1lPb*TUi05Z89^q@tpQ26x6y0AQ>^y0$Tyi!w=2g> zSB78(e{`t>ie7IK2hIZ&&q!~@pXsw6y^U+m1c}|+yD)xfHBBHOxQZ+pFqS^u+^fOu zlEh`>mhMs~6-Fa%VR=P<8767Y=w|jn;TkVU2S`Tx*+ZzDM@RIH;bD3>b!^!sBB3IVpDu$xYn1Y$+dFr z$^m-dCqj5xPAbd6;m5>LI8$VTzxQX3FX9FLz@sX0#cc{k$3Wc2;FU0gs_D*EQHYeU z*-CT$O5~$hzpw`lpTtNm%KTryCyuLG22Ye|ln5PBQ zN9j>1B@yvfoV}>^j2PmkWxJXEw=gRKS{Am8I4_U~N4HL0MdmCgAdd2il?Py6sZ}LQ zVa`4#nl3P(x%0iYxnukXF&+^8${sajgqw&j?7j#9MiS3#U{j{M)-SI#I=}6%o z|1|T#w$}3Az@`h_d@?OH^5kB>n|<9YAIpxDz_{UJ?qt8wXkU2A5}p$#n5cR zfWn>yl6e|thk@TK=kzZ{z0;|j3nSwDmKxjIsUy4$LR-VB^ok5Hx2gZiz1tYVq&GEi z0}VV+C=96x4!yrRq&sx%KOVOe{f^p;K&vsV{DfV|UPA4;-*RS$7&piuMVXLR_5$kd zKd9%fnUMh-;U2>+&RU7E28;I1Vv$#%RxjjOy927B4$-5mV3hgp{UHqh6I1coZ51jY zDrVcZb&w8$p!ofXh3awd4w>&mFDrqRco#5dICt_I5$33q-GTm89y7Nt%mK%w!<*R{ zPomb@mQ}HD8z{MUK!DheU6TNYIzR|o8tZrPpkdt^B;@6m9(`j^nbfTFi)Oj29Et>P zGaUbT)o<(PpzamN+e1|!(bU8;bq2{zJ`vquk5AaWamPaYt zvOgHLGc=46q}>AcffLF!yll07JqIYyu1Uw5|Lm%6`F?nL#c{(2+8uq% z7~}owse&Qo7f?xRGXxd7$CoV2#jMU5p&s|$8jyY;UssNBbFfc)%4zbUH3?C@zq;hl zxP-mfgAN|`#3Wt=lEBm5^Af8FLtQNd>|;ezik@UDu5~X9u{r7)wufq%j~+YC!#|mZ1jl!0o7m-Tj6_gMkIQP~Z(f@{R;yzf)1URUEL5 ztuPcULA14rXn?7~g_X?4nRo#G@pk5M(J&4S{06Y>J6uO*{oXpa{>exL<8w#awdCa# ztQXEJhVedvi|u(52`)hBT5R>meTuH^Xs1xAR!8wb2mp;ESij#l?<*uyYga0Kq^e(d z+Xk$7^;H!QU}D{9shuDl9NuckCmd*9rDU4ltvHj{y%_A$a9z@BUx0#($4nmS!$_S! zR{avayuAug$7}f8sXD-buvqPXj!GWD0FCO`BkAI$vj1Cr<{$@|7z92#5?6j6~0xYYlCjXx zpN>vil{H4K?XPlld$#V4#=H%WJ(a=<%$@;VkU4{QswQ0hX_yqAxB^%aQIF9mggHC+ z(=8KSWozp2&_%GYtE=}K|KNuTryLGItxN__6rO6Duw0#Ku8#w0XHBHocMfT_-0=59 zgVzr@z_zIWKOvXv{(u3(0QS*G$i8(zpY^NW3Lb?<0m|M@q9{N;m5hI3nT459IA;-eZ9QJ$RUjT(jOIqb5w_R-(*O#MKaijJ(5G~ZTv{DRI zfT=v3bR{x8_EzAI)#{*hwX(57gngq*A1K0n1J11X%5S**w5_u`oZs;p&N7r0UBNP| zF?^s<5skUR_OI7ug)8YUcuBRp9Qb^v6BP$|7*HS1!@gLwe=4I#XEGNTWn z;OhMw6{t5&-W&!NU`CK6iZ{YOX?}FdQYkw;qaRfTq?r>y+T1sY`ZbX4Mya;j2RA;k z2tHUGFVx8%{58rQ9m3wpfxn)|hX_$WHor1bNMpx&y)4UvZHJKqqpSJxan*gXw1Pp+ ziWXp^bxxtSD76{zjBm1K7>hozHp5?d?FN?DMfLx+yZ{CA3u}&sT3l#^n6VVPJQ*79IiwYkKY`U!D2HE%txMtGQ zHv2%Fh}JmB#(-54r$EQYZeRf&FX8qFs`N{LyuQGu-0yn_VNR#!E9_``R@!$Nci!Zp z9g`Z=n-EalQ;V}GFVX8Khay_3l)a{#PCQg-q|&ZX22l-X{WYMket2iys}AIxs|}wZ z6hB3ZrNLAj)aR$fF0hDRS$VH;2S73V2hdZ3827WNAJ(I~bKszhy`;_nO{vr!Q7Z#S z+(?#xA3aBS_^Ie5w?j}mt*6uQoi+=%=(lvrV*w^nRG9(qh^} z9p%I#rpyqxrV)(Tn0vfGE*tO^c9qE6N|s~3rB%1*76_)~1$4Gv%wu{biXVSHUua!l z8f^VaPk+R2jY9ZoEu8I-m>3sqWcTpMoxtHDK&quTdG{NzN~!=OLD@~0Oga9CtW2>P^+emb$PNzx%QEL-{;=#mdu_R+lFPRlOGIM1b=*0 zf@%N0)p<{o=t8ZU9p*YUW^2(H@}%j96~Ie!E+tO}+#TKf4%#>yyw^OJzgG`zl_mgf zxq8PdKkI1C4RLLTM|S|AdkWdm_&?sUf77M?p>sUv;?SZ@_L{@z3Lr_@H*sbXYM67W zo9K#HJhUf(>lS#hArJe4cPH$MF`=stP2V7q!7>KrRHUszRmL};ARb8o1MF|x&cr=( zHfue}6S(p5-1gQR`-!T@NWzcrRTQ4yVy1Cu)oS3ymUm356&3)ytpn^`yDyi8et$13 z%6bgElK@YZ4r3Iv`m*H27Y?(}^bvVF^#$2qR3lP8Kg883dGZ7AV=k+)AsRu^KzJyW6MgVXGiN4Em z4TZKd+qZ&#Jv^r4({iAlu@DRc>HRnwfgrEz58wsypbkl)_Jwz0cISd{65321uW6MW zwQ^Y>$OtTi0sp8cvAE35Tw4_%lEbtsLN9fIhSw_*VSIRiAYu>`fb5l=HU8Qew-`$u z#ER;AR>WVI$JK766J}aNR*C*}LLp%HblIyCflp)K7Njp|FwlE~l0!jL>zTDLQhDDX zJ!D&4@H@%k)u!hN?PDh#*nRva^*QdBLc`kgn4_bNNioZbA)RI}2Zk(#EY+u%dyEHf z6(1B1$0zAr%6gFAXm0q3?N__L=8s65XU$x>rJJHnF@-4B+%B=DMUG7&P8(#4!^7p=%-q z$n?vXGnOSi7vgRf`0~HaAI-${);F5bFE!aBMk{4wA|7{h7B%sHncbUBLHV`Uie385 ze6pDCd5!h;#Aho{nS?%7GhY%*@kJ{xWqA7k=tE9@QtNgOY~wDhk)qXymn1q2h!_ca zhM6b-Xz91zMix%DY4XBHt=^~2J9!)(eVseH?=h^F0L)&hc9IH9c|M9>qHVsTq;tP& zeo1k!ab*!4)JXNWXBpn3hiR46Xmc&*ahFxjzfM6WJ{)DvF0=KD^Gj+N6wt^QON9}~ zV3~&;&pezu_xZtt+na#?BbO#ld-Yu!;e>Kx<7;H%U&oG9(K119Kl|H1(jsm{BPJ@$ zFR@|d$|bcI=m&vcQK=~$*H-}o?`am*wXVeG#jOa$eniQ-_K|@S-e5%zy)9?*g$rhZ zm}kv4l`31mGZx(4GPG$9Pgryag?R8U(KaM$zo2g02$M__VGQCk$xp%Qkx@RaPMyT& zg_xiFDvcJ%@b8xg41z7kn3*U};rV=leD&XNfr2Z4#zvf`$5iri2VbFK(8r@}Kv7%i zlnnM*oHsHN>yDK|h~IorygHqAot9;lYp=Bs4P{P>7Fg;tp-;Gm{9GT{SHGiCc2SaB zbvOHI@5K@gkti|mg;v|?t1e7JDiCIU(NqW7m@MLNyK<~hl=Vmq?}^GR^RpD2aSEIFFq0<&H?XONZ!j@NUD;Q2JQZiO_A#9paFR;jXP0Rl~7YU}UEC6>`-<&uK(x0Tm{ zu$tjwh~(lxmDCjAxsCwbm`0o8{vu$!(UCH+ml?8J!-onls1GYL9nd?utr-h&Xa)$;lUuZ>q2 z&~qWi3P$5NblIuSEg3T=1pT%h;U^C9mzfQ`hjC?l4c}P-PNaCGfA02qji;h1$o>gm zBBg2w{B%4$#wWWyxL;)xC$h@|p-k^z<=?Da@5OjTyhp8G;kqFdFymr0pK{5=DcR%h z*d}k^SP(M|bB?c1jS14$z_-`7>tWHo3BAm}|AFR`>ixw7q$`=F9u_So)@NOLNA;4C za-wx!HCufd7I#I~`}{itcue+u6|sA0q6SpijPf{*pk?XR3ngfU#W^G1K&gSX`Z-I# zlI*Q5qXB->M#P@%<33&&DYr59*86f8Uc!Cv45P+3DgO13{C4he)3A~EQksWR^kmnK z7Oe6Z?RgqkwA@GjuG|_KKO}NBt8SCj-324YCTI1K|YdrW&;$}Is^=$nsM`w^yX5)l`|Comcf6l1Tf=^KkvyGk%Pu2%gR#|oqI1D7nHDGi+nUkb zXl815x2|`}5%qSHn{}SNf0lO%nCU%~)}A3pGxJ12c1ne3_ew~d@#gYfb_X_5F`Og< zn%peou);?lA`X2$Dey1*7NmmbI4DRJV`CY9c*uGRiZfd$M9BXXNRHD76YJhk4xvX= z@oDD~+~h1OZG~GWH}t3(O=vc0Ve9MW%rnYIRv&@@W0ir|QX1hg*$m`pZ0&bawNLx; zm^_QSw3+{8y0-w^_jN1TM$crdu4_j!Xhr*?MrovEYD98_#GTV)7H1}RQkkh%v`a}u z=o}up66Y4UfaJ{G7HTt%i_f>tchy?8&WUe&P#}|)ckR(=5d=ui91gQ^2U)#eDas`F z&LUaB0iW#=E659#ko}JAZK%KIKVF@c91ITKAa(Z20pMeA?r}Oo6H!EFzsBgtVv^Vu zoTTmbBJNN{eq9~5ZV~4{A3X%TEUij^%q}1Y8V;*1kf`4NiA^I*1o^--YdeE2^No@4 zC&;E*eSyGLr{5*5(UK4F-k~US0$y`0G$M;r!Vos(o zhg;2q*_8G=ZVAdN39{*lZ=PEP5)ZKlT!%8)_}j2X!`6h%ANiKaMq6N{I* z?Tr@e0J@tkqRE?`TG$@M5od8_bXPTk;F*itvI$>HvbNf~x10V_=aFQ#7fpZ|7-=FV z0r_o~-eE7B6VYw=xuNq3^v%w+9;$F1@p}KuHEgwGBZ{=`P=B%C0A8OILzWcmkwT6V zaje}AfxFy3RM~o~D`-CJ1KO{t+B`Is6zwAXAm3;}aDyw)Iz_dQc+U1wgw9JxV^DoeAK+u8lVr@%ZdZW`U-^5{jUts0sV zYVpcFYO~3=Rb-|*$J*=s&*wDQSZ`NTW_;9?wOY?RP^f?o{N~W^DzOLc0|8JwBUQD%uh!3J}U#PnNc2car2y{O*3erPYVZcy$9^q(*tlPc^~G%wH?9#3OqpUo6CYr|lXoBY+#8 zyxi=!Y=YYPOlIC9<;mCc4AlbE2H#l#^-R~~gBVhaHCras2%&AwYBJq5r*9O+r2V22 zDNAFc_G)=l6!tO`?8%PYgb^(2XB@_sJMYAG4T#Z{36Gky?n>+0nRI z_Hulm{E4&7rm>KpMQ1RzbRnae;9nxqg$yy@nz? z%cV-cF@st+sWL~l4pNI$ulBdo^edfq?;7BbUi?*ZcgC)q)bct*?y!#yL0(QE+54|( z!^|>91LQyVL@$Qf%@pjnu#wS*$+QY>yS3$OL`H*Pg_J0A9#X?0%5P!Y`UUD2be@bm z`&{UFDcCJinST}2GicHZx<;`oS0aZy$Yvl!>!3~Fe3QRl&sRShO-|?&p1h<1xfK6| zg>c}hwC9=Z!X3;kx|-@+9!nPM;+?5NPdA{cmAjI!^8ahV2^lQ<0O(S=f|6~{V!_OY zFCrz?@=nO3NKs74<0x3%6#C|RD@ z?k`kWb0}G7oGXW@@JPvaRb0I^$jsa6Q5uLf$i6#tj2ZSV8>n?_SAUn6pswYjuC<^o zHhoG7lsg7^r=)i96qx_Pt;M9?k{7m(SEA9icQhm;#qKOb4_hz7%LAbP;4+(=&M8La zorHTcW33b#bb!m*J=cWzXS6VO+yY;+5nO}agr!@>;>X#Ksrw;vvy8SK5x(Wf0LeIk zj*_}!N2s2i#l)NHjyaA zPwu77N^W~m*2t+4wT49?AYF6&_1)+6E&lqMiz!We$uegIznO{;O|WD-F0!mK8UJW> zUTpCpdtnoLnXO~{U|yXa%`%rxv#_U}5a~>=k)~oH42~XD;%e;$Lpr>P9~h5rmwUFM zE^Rquk#n9eH<5>rv-{P5MH$&WmI7M|^oR zY9F@2eDcCo2L+x%a#KrTi<=nCDoJ*m6ISE9X&;S;1swK`6SLsy^VROL z1KV`LmUG$L^V^PNq6tKN=O%0Yg;?^~)0BrG-Y0h?n^-<6;-qgt@23=`OU1|>|HG)e3EV3h!I;_(607NqAgW(e{32hHilfkrQrN#%V|>> z@j^dZo#}m1<)H@;+G8;Ys@a1T#ulm-U6j2(A2VPrJbv!SGbACI(HymJHtnC+@)Zti z7>9>z(s<5{z3hZD1n>xI8NzKC%`a&5M}07b789CF)6p`m>xvtle)Aqw)kuvek?4^o zF{3_stUn(j0WK)HxW3O-=Q8x1OVD5x-gVmCd(>&jk0tY19_i%Ur+=GiN5naDQ2xUg zbv7LtHpSd!kpA>V^S5KrHD0{Brunp^_7M-U!QG<5fY;WJ{E-LzwmSGS_z)*N z7N;u26!|lnNqRHOj7Hk>ix+iP7!siP$s&i%Fh8Dj5A%GRHy{}|`KxQ0sna?;1|R1y zzg`J7(xI}(lIrG-wb1b8ak-JC2AaSb``ev|SYT7n$BW|y7AF_`6*m`j0_RyE=FvgK2m@j_=i~Mr{(D>}9Rg^Dz%>(X!ATwV{8DC6qhdLJ zfqv2^ejR1dtQ@xx(Yf0t?v+otdVh@PeNn1sfe+n2hSszw5jan{d7252XM5Z(WL z5A&2MU!{=cYN|E30aiqVDk{EVULk?d7k@Z|E6sk-DjA=!|H{?yS$@O1eFD%T6x{rz z5_rzEEujITrB_1onBF=sn_O^8y~Be|`4zAUZVIMmFIPJ&gc9QMR`>L)Jo`$~wTz`+ zAbPG-AG0Z4dQCpOE#ALZ@H5-0gi^fiWmHIpC)8UU->y z7(oOKcpS7-W`$t5NvwpKjNuAo+mPck*&KNv)nwdW?WL2dI_D|}5QKWfl-z~+O(Eg> z?Fw5Z*JD@i6mxe9xdw0uYO!k@5xQ25ctt%>{@poK{YeS671K@d)t(`v=v9~Y1wR^P zo9u3eP-ow1sg<`k*B=nv2IO8K*1egi4w~EpZv-rG&zXMaT`HuN(=peW`6Pjbc0jZc zh&)e9zDgkIbdz|qPtALJPm|mYi~X~?@r_N$eh|MI`S_XfTm$#+e7+~Rfme!%3pn%U z=d|AR?4?v+1y`v4nd6+P+)JA^Q-(%9)3U;@0@uINdNPMcnXle^8t@V6lICR=dyg|4 z&5tkAyx(N+Mrsh&iS2n6K=pdQ>nj(yltg-7q2$tzrL@2PjEnxLgiKG3(MODnDqNu_w<+UNc7!kZ2F`Pi&PbdWUK6~3s zQ|`kFfPL$f5Y9EER@RsilS_#SaO+MD_rMPp#qaIw-)a+$?Vm9v3khS}Ha^h@D5oOD zuwBGYvZ&J7lI$#?crM45in-}b3m#Iiynq|8y`y(mqa{TbQdm5c)|y!0-HA)gKM$st z6fx-%ZLty&CxD$Hm5ENm_W_Ds$O8rZ=*zWf80FfWb@Kw-sNs*J|vhVLP4yjqJ}xty&6^s&6BmeDEnf3XOD-W~;cQ3f!JX=2%NjwWLY2PnBo5GCc-m6{(u#eI62)`q@ z>Lzvm3>HDDPP-TTfpdttppCJygik*WvJVY$3J3h;O@5PsE022LFU1EN!E?qRYMzKX zESxDI)fo{LLH^|5ZNAC#P#ShGX4`5yYX%KXG`JET%astI;bcSXvr1)GQdbZa z9WObx(1%^NiWOTwzwAD8|K@#t=vvE=!2Y*x#X3wBC(Nod^^1q}oKWWTAw7P0htBi^ zA=z&W*w3^!9b?X$o!@0LKh$zR^cQHm7YAk>$B9jKi@=MVIcc=~9k|Dr@B4VMODvp? z&52!boYYj@dbt9u4Y0haL*q1Oc(g#*kzWGigE}uzUFa|4%@v#nmj|rGePPafs5UhR zDrGC+-Wp$A%+|yB5|0oFk(j{jW%DBPtblO*rm5QzZx^Xc$^^gYhM2e$jDb%0jv1NY z+kCU$qpY{Pt{jMDxT{D8a%^`7n8o&9Ss5#|W=8A{In;P8bg|{MEz89C&*P-xqaKTH z2J_V8qW2b-&qnQiy%M|E=L?Jf&~2wgnAialFE^l?$-8vMd+jv_r)>f~)CmRJqsYY1Z&Kw& zc!NbZh9sGf_Df<=jzm$=qNq71zhH6KZZoBLbT_SbvfS3f^g}*gmlT({!v(i%ilLpi zuFA?Rv|f=_k09P3sOtB-%ss`AQwLFYaCd-zAnH`%w}|}=a+qYGbykV0yLy zcl(8Yzs>E92qk%ScgcJ6h?o0hk$yks@XWe<{_APp3a`aeozmw$^{KSmuk#`BC9t)L zFx$YCSn|b%>Yek+_S>{4)&Ei=^)DRpW#bRC{Q#oMbHc--rnQ62xp_{BR5(%Z2E)Rp z^=e_8v&`FFWOXT6&5sHeGJ1!cnqZ#NzjgDPv%=X}0J$->I>rm#cFMRD z0@*fS-FR#B^Bo0ri@hiHTYF{eD2o*`sQ4V1d_#opQc08c)lI%U7P+KJeeiR}0TX5- z(WwOq5?bmm*H*K-rlBL2N=YcRZ@qosr=%bPpdb@!D&?lht=;s0TNRMoRu`$fyuOy7 z;E*(0!}2{uYdVw& zI8dgwo)pKq5n$xd>ezG&G_>%;o2l${zP-`=p_y{-gdFn*i|Ypx7EK(-PgS6@`4hk% zu>-1yw-k1A71B`$VRnzQ{^gxPb zAEXCZ-XymHf0qK>emqV~CwS}ih%MiCS}!!_LF-=2l0tI%8?9#_wodVQjQ&^=#(jHx zMro%RcK-C!uQ=QT0wa;HP=1MhK;A3njYw%4{(}f3_1()E^14;Iuc#5jwc85-2;1@c zBGvt#g5It~=>*!yf|Y7AV5Mz*h^=w456=Oo2(Ar{38dr{d6lcZrHc(2r+x zqunHKLux$*wkN``B^9k@ayPX<0sII?n8XYN@B48tV5;byf3PM3CX}osgm(Bq^{@X} z!q0A}Q#Xwcjj74Ge5(-P_530wW()_)%fhoS1x+4_LZuyZMi#$BK@PpP29lo0-*JFo!Pw5ZCxGbBLJ=d|NH;?5(;Vp!TSx z<3Id*7{t%RfSc3lwf*&GA%zImjU2u6>_Z4WWz5``Sn3ge(7VVSH#jYI%OsS-&~5bu;4-_|q{j)|fyjIa-%?G!F%2G5QdJFkkA&!wXLh_x%E z0CbB@lgR5+GD;7AJ$9wE-t;mUk3~nF0M-BagM(7woZ)qHZ`OYHVVL&M4)Yt6@n&3U zT`x3fk}x|pEbZ3dy)hlSLY={1SYItf*~)5Qz+jl@V$b-GBmlhQ?_0X8_`AH4_Bc1p zDJJ{Y+ft{`Q(c|=|a+g4$qOW1j`D38I9GoQS5KrK?h zP5P}`HnpohYqiv^?#es$XG%4Kef24QN%nsX$FtiSpgHF}jP6tZd>OlYu(Qj^ggd1V z(n6m>dU%+Ol>f_%%fOjemfV780gd=@o?b@j^3&CCHz>dAkpTBxn41#+uYv!+rB~^e zmwO{i94U|4mp$9IbxQ9Kr%aa*JYc%~dR9b7`Jcac%Rm0%B>XPVsqAN2D&^BT9)39R zcxm2U#|)mj?Fz=AsiA2lFYf8yS?RenB9#S83FY{{1!J z#inAQC!B=cS}7k<(F104%(QG>9OYL}>4CExhp`f6DF5}B&ts3vHeH=ZzNI{7Ciea4 z+hjrY56V+lXu#*)f*?cw{o_D;%}y;Zn~C|Y3R8ARbq@Dni=X4*259iBM04PAUSDwt zCPU_j6bbWTaKmchRiEsG+1XB-NEgA?}7bG05%8>RP`gkEvnFRHsoJhW5ohG;I;jmKV98)9<(Jnr7oND zjIJ;s>*m{V9Xp=Mkp@KfzMmkOiZ6c4|8swEu@4T7iPk6o85(UW(C5RtH~u6MRkT1^ zU~RP0LQNs9*SO9!`I*QW1c87(&B2de67@kI075>WATd*2b+P~4TYW-c?BAG;v{Cx} zLkswuOTxu`f2L0!*qyk|%WS831QIC_5IrZ=E#D0*+%7-xsJq>5_UOQ;o`!uk_HJ z(V>jW{~s@_elS;lokq%1T3||dY^ginUPJEB@-m>K4Y_y9uJRX^&Gke;k-Ay`=@drj z^i=^<`X|}Kv|CM>*30TL?`7`xP~_K_TX%+Q+~(Riez)Ei`oF&q8u0(v`|7AHx2A6c zl@Pf>LQp9QX-Sn*5S2zjx(qr*QW_}%QMyyQyBkrYyCei91*97UzPWfj=Q*Bp-se1P zz26_-`WFAVZh`x{_MScaH@}(Ld!`T+OHbLeU+q9hg4J6&7@D)l5FBq4DAbb+FJbQ7 zb{T})V*@3oQnMND)pQo!GkEdxNbx}Lv zR_)ag=F@C%6<(d{Iw?T;4O)4~$0LX0BOUg(rV)?Z@`LgDM<~}{gp1y+eDiK!`R$&& zh!C2{zx*gZouNZ^psPxV#T!QxNW2;T$oKJ;C-*@KcxuSmIcA(#xD8PnpPphko$gu! zo%@59HdDfC5%M)1;7%?k;EsON01NUL=kDG_X1!lz_{JHs8a;(5Wt^qpvt@bi~G> zF&oP6MW5#~NFEv;&0#>77SW*mRW<@3>)-HNj22{VCZXqcAnRiwYu21s7vDRPKZ$M8 zEqtuK4RtG|&60zgAcP^)d{xB5BAvc=t@ax!?)*79G=Z1X`6~tR*6;7K$?LQE7`+H7 z(b>cs_5ZLGfN-8*eiT<34=hs?J-jfTNx$mgg%=!Mah$%7RNbg;QwlB~t#at8A{Wlf zxPMv;T2knb^b}w?nH5aTUH#((2I#;LM$T!aAr_rH3+#iMdclEtc#Oqpi_*Y_+tXo& z1Oh%<6&Xy_0;s!3|M+e6d=bF>aHXa~L^p2A0DMb`DB~OmJQg-Geg@3o#K37!bf@MQ z+8rriaS|Wrp?-Npx-U$K)2T=aQ78o+*r2XYNzWhXZs-X@Xf6dYCH$h}#B!rSOi=OG z+9(Lh1H?o55}fdgjREk3_(G=OAjNsBolxi=6%XAiH-g|~K|AL%w4I{z%1F%x7&068 zy}PdV>It1!$|Gto&ujd#uWf!{r?u(p13k@MH++coD`?mPD32+J9w2GyB>?5f!onnE z9kP`kesv}vb777>y9b4_rsq`6=GCpI6Vf*FIW!@S>r4lw4EyV|MRJ~4Mi`Pjn19!N z3~o1#89+yyYobk8rIr@>)}O+W7n!;7^|`@QM{|uw!V{HzU4ME1+b&RD?$19gqL7WF z2N+A?GUtV1Ay{o;~nIxnR@WvUeiSyC_r=Oo-ONl)je>R!8vzxYvIWk%9&+(Ch^?{q}~S)sD$3zAFh=@%S+Ea`$Js!#-B2P5~@9_4MQ4MJ_0hf&YrZ1^;U(9ZE@ zS)(4gYcd(Bh%)b0MV=SHP) zwZyG;;lPLdx7oV#1~!FmH{$^lZeoUA{+-ES9}&RA5ZG5&BE$O71oR%VNq~sZ(gu_q zq3PQG+CW<8zOuvPf&INAsOUFRC8O+3F3j|X=nsN}2Y4X{(LII!V84+d22od(M-Z_U zGN4@Ijx@h+*_S5-wkdq8T%-=2>g%c(a+{dNpkF012>#ypbEMkgsWVp$xnbp}hZs2>98J*U ztKh_`JGroW{efHbyIkr4tr(IrOAp+S3_u%0!|Q_w`niHap0Ih3qhCe6-NIn1C$k zXlPrgM=JFYN|vjOT|bQ+1X}tv9A$zK?3A?&2f{M%Q84 z?5{uX!j@O#52-u;%RVCakpTdgR5d>!?aIJp2M0_F4<^Gj>Ly*u4c$8X zQ%lsooXGm&pZ|1aMXq_Vz3qp7>B8_%=T}-7<%k(W4 z)QL#>GylQrAl@ts>pLR>AB^nDD`|+EsH|2CxYh;k%2X-Q*xtn(3v~_@yJKa{Wa(wh z=y>cx=f;<4qx|^I=473>{xUO&2~7O-uk!E1%-&>Zg&)jJ4FNF#rv-(gm)Cb!yKHE{ zZ8&EY=+JBl*KKf4Khbz${1m%u!JV||xK;Ks94$N)TBwH+9U+7Xg^8ajpyGM6rVyR| zH%)$v0e~%WS^3m~93|jwhD=&E2XBi)-bPGomILg|tbsasRfY1J$#g=n7smjg#lxs3 z<}w3~Pvg*}nb_s-$Noft3DRapa46Xy%`%|SbnVU#AkX~w%jz!!NL*s+35e~OE*^{5n);jog_MMc{C$pV3rcKx(D>Po*9z!;gk!K>wD7>}nIJRvBS32~w`@roevD4eJs)@6!M72Vt_yr+ z>u|F#)sFela_#m&dBUflxxu2OXHnYyKi!?TjnP(vmG8zxNWrwkY96>qzssqf0#?c` zmHS|&<`6hH5*R#)=;om#?E0n?VMirD^Px`RWzy_Y=#Y-8R|&Vf*Z&jbPPI)fJ^TsyE6Ji45pn)o%nv0%4jFV+fd|Id z3Gq{P-hX05grXw^*omvfKymXSA<&H9+^65a-^2_aX}|KPJw*MhIbeahk`-Sf;t0A8 zI-sc3X^&EP4Bw=Fs#5lAxR@REFD&%l3FKMXF7QpUPstV@$X7ptxQN3WU!$R?2UfF3 zZ#CzGK`Y6$hm*r}_V~XpV!H(>;qTgi9rmB;&t&ibCB)Bzf^s8&AMC$B(1UG~eH(HR zoQed^{r-dE2%>e7$U>Q|3|jsA5)hSRAq-c~25Z0``Bgo|mmeXwWDg$QjtHhOKVa^c z*aotzAQA=pH0@X|74n~NdEPA;I9l`u}nL;CGlH1%k^o2Ou;BWQ4~7{RlkzW)k4z@9xkqi~e>F z*zCZtQ{h3B)dc<8jmh7L1*}{=6O$&-uY2)em=DkaG;XpUEE6{ixOYDj6xy!+M03g>1^LLS7`M{=N{SCO7fPNY3(hdGy7+esfaN?B+@||`o5JD^2u_MT5 zT@be9)vCz9hEa9OfKe9~y`4jT<#+)U>iUD8gGU?Ug0&M9>zDk?svbQ?08*g8Cw>sz zXobkI#UAPi@_k8&p_-L$&HWNIJeb?fLD($?6(3>~SR5Ah$j*kyAg=T7iwkvK&j=*@#lt9>BEk;v+74`t^t+d|=9=RK6&ly?sgK&W| z@Z*6_>B2uBdO%hj39@72`y)ery8=Rpyn%KE`P>Du?JJ+8=JA z!!2|GT!&WZ06-6|5ZeFSR_MSY|J&3$;)0dC^rBq+&*t~92Khf3>p%bX9bjIsk1VtO zi`4zaw*TSJ-~IxuyZ5PBnHwxqobKW2s`9IJ5&p%1VhYpw;tiML$|FpdSdPX}xkn8n=P$nc`^AC@G z04d-Hxgo;;_=hN+Xkf)V1I>j0g+|hYyF#xh{x_b~pGTB01v+DAsO9-r-u#!j1p+7p zGSd7FEC0jJJ($P(>H(7(qH7)9uM#LALzaP zkJCrhUxthuu6)lg-~U^&Inu!U!@3KZ|Hcpg)3*;*^-xt0RrL^64^j0$<22yxA*vn{ zfI|Xs$W;%y>LFMCCu03SxK;aOgKyj9A>VNL+}63X08_Z`^%#5rKPf* z`LZJ`{E{N?!OE z^k{;(5+qbH=aM~MVkm8FV4co?-@j2=Fl;qtKm2W}RZTx~xBvOG3R}K|tJeQYr6Khe z$TNF?L5aUlI*Ehoh#ThpV;uG9F-OR`I2J4Y7kOefB#~Co@?T^J(I7a9HmdhuEL=U3 zN9^YMFOs7u$hhITr~MZhS8^Fhqsj~(9HssPocv9o2qti?M&; zZ-*FqXq5iJ)E*k8Lo5FeXPVFtY11JO{m+E;h~pu#KP2}5;Qt+ZN{6oJ-#EL!Sf4{* z{xCT5_vf{LSChlw$YHeU4?gbS8l7K)BWResPytzUx--G??7X4f*vB^}P~5|YWfSYq zk9mmZNG%o=Q0qw~|Gd~@PH77Y=cdl$qudECp-FAykFAxoI+YTWgb1GJnWd5kHQZho z!3`_R=&H{LH}t*HEarz4qoA{O_eCfPmIi&OY0MKJ{b`Iht4<(g9!#LWGaD=G{fdrv z4oaaB5$TySH_4?RcEsM+gywIz-V|G}=~+Py;Fqbg2ek>t7~qOnrMA&QO_+9eh@?hc zOrHC7w>YvAoI?}vBD=xgJ%2*rMgnxDKW|W=!bLhLVr$GiCqH_)UbJtJoreOP5$KJvNU#2GOilS-8fLMXRq4 z;@DLFGUk1QnL#c+Id^ldPrV7c7pCb^5*iKUrx*;Em!z&0J4kqt3->hDLeb=3eAu^f z5E;J6cdm#NsheO^>WW1aw7auW#ImV95z4OJc#6#So@x?eXzU1}fWvIV0CL4+kI-^9 zqo@%&hpKBX?JQIUL%%Ay9wElzoF{j!zPb{nD(!DgMNdowt7oQ8cRd#H;g3wL_>*jx z;CaL&c?+|xCjS6I83MKb*ou>R{)P)TdJQ4}P^0e9gzri-Dfh9fa^w#h^z zQid&6b*RKNDP66iG?nJ@5)@8tTApZ1Q`(2#D4I~nS(ejIi4nassVkodo)Ey+=r(fI zXqXYcSo`}`u0<9{fqU!F5`MQ&ePDn4<9VY1Pa>8=mTVI$$NR%rBgMvX;*yeH9O|r& zDNrv%mxcasM|9=+{Lpfbyt?edMq=xY`5u#0Z=?);^ zgCKSiQHq)AXgq zw51K2?KU|cKcy#{Xgl|1niPj`^`7j=^AWou+x^`QWOd682Md=+Jy{Z;PQwP%YzgB` zVaJ9MR~MtdUjAA{>-4%Q?F0Y6Zp}H8z$-~Ird4?wZ62v(vsY+qZ|{P#%`cc#Rbe^} zN!5*ak`QZ^a|i#)&ny$yveHNFMoNC?l>rxC?082n~K^IQOB_&9v0}7LMl;7!li25i(gk+V-l)J;nw+q>{@wzoBTHWZ)9 zg$i4#kx+`5WuoQkNP22$W34A46FqFGG`+1lGKT;tf-B#nU@LzUHJ<5;2(6vK@jm|# zSocP{Qk;Lm*;_F{F;(t2lQzT?6;K{EFC(q%5@_@=G$j5TjkcT&)oyenvsWnaXOPbf zkO=3J?Gc|C4P=lHQZ2KPEkKagYIm(KO^>k>QQN*6BScqTRbM8v$8s{8!2f(uG7N@7 z4Xex?d1s~K^|Bi5p8npRX`~{`0)Y2lkc1SSQ6c;7g=^2=>e2p0h74r7ZG0OlR#0R< zO$ls=x8@?}HT#!H#pYHC0$5ba6L+>2l~Gj78mm7qH)^GViDuDI-DE;% zYt6b~gA|#8iSLQQidSirZfqP()NFzG^L;v)qxy+|PxPj`Gr7ajBS-f4UC*+g{jKNM z_sCO27oQC5FO)aV%=H9SmDi-574jyOQ{=;=#HDm6ch9v(bzCwK@^Qgdqg0YpCQNfl z_VppC7`)6Nx?0HSP?h69#XEIfbDmAR_4-N$fArK!1Uvh4Xl@@`(zH;SX#w@b_WK{A zUK~LqN8xTl@2MMkAMnH!IZijy5A*5S*P}E_)DbcLaE-6&}+uAvLt7o zQY`jW?G;RT)+UkJZ@&9pIQW#1?i&6vHLjtd>>r8B8=DQ!FpyzIK2RB0SBXM&QSyy5K4Zk*44 ze}>(tVl7CPq|cf29L1v+u(gIk&~NM8O4a_Z(Fz#6+YT__N2~i_flZiVC#O+P0e36w zDLtDjdp%hyEzMep6doR3`TTq&l%;aB#uiHYb&T*Vl*{DVWZN3fZG2O)hemyO9A zDrUv!=-L3z`)pyaiR314v6{AU26veEaQ5!I5~*EFodvWpSWVyl;*1l!Q#45sS1p8Q zd$KD0K6>t_$DW~@L5eiiG_zJ9$QyV4Sxde5x7r*?u)Q9mbkL4%8>V3skFzE`Gh?ear`<^SNcRVhV+njCCsH$QbDx&6_2w!ZbN^CI2j@236 zdUuM9Lpq^v)`eTpM?EV!Go!r2pkl4p_ndBwE1{@KiZ)rCh={5w#^t@ejjC|y?h=jg);(iSl%Q=s6j%vk*ZUz3;_FI=&Ra44G^YNTSc$20kp${~H zQRcLR2mfmN{^tdIA`Ce|_9&q~`M_>YLQkl%916CquY7NNmNUbK6w|*xwUh6`t9KVn zn;+4LhuqVX?=jpq1TnQwl%1xB7GhiiKao^;e+EA#?b_8Ya}Yu#{Og zulMcU4%L)$)05TLA_FJsGvx<1d}k82OG4&D^J)9m)H-+5xf0}JXuqCg_doK*@f4~1 zp5Csq_%<{1BxzrTrH$Xhc1D`r=;MR|H0CY*q@|J#|88ARrotw@3mK? zTD+vuHY%AQT!C8Z`0~LEn&`!9b$V^?V4BZcdH6h3SlAX#w69@r<2}2Es|FpCMcuS= zvwXKPkBwVPWL2yyU2^pFyl7HSA-jVkvb?*uv*3S$aP=HU&GS!>PUZD$h2O4OkGB($ z6>MYkphUcqy5>E3G_t`;^m64Qch#R4-@HSbxu{i)|QE!VV$m z6>-)+b>Z2Iev>Aq410N3D}0Bbs{zpdz2P-6I}Hhnad=@zwaS6-l;0-+B0cMk`HL4* z-?hZta-$*in+{!tg8kuLQa38zga}pbV6l{6Q6dP6CtYGC!S}5mk8vY?bv0pIy-&XF zxrdG5a1}R>1pf?4-d{>$9+*^+3+He&WoC>V%)#y_M(|*u!I4@Amemr5s+B7Zqz*4(KL@rpP1}illNifjG~5cvBCVaZ1qK0M;=H z?vxjAY00F&n;G+0>Dt?dl&-;-2_MfvwOC(IdsQ)_uJ~J|~&$N=jlrBBZm-<`TEDAZ?D@PL~7b^CP{+ojr+S5M6yUJ^Q`SM6=% zp0%1YX7yOwTMn+Gn$Ic?@a2#g9%y|tT-a&eB8{F4y`G7!WY#$=aC%x9)O^Omjf|~b z$ZLf*&wz4Ozd0g^y81N!K)LRrQGL1w9SPEB(i`q$e~@ESe)n>j8~l+LuqO(1%BBFj z$`VIXE|EU&$)hir4AZ}3-tD3nmgqqzk=?78TRc!QE{mSq!ZjFTzq?JW+(y=888n!M zv35W1PAD`j${3;$C=&Y;D{{GyphAQ`uHg;f{*7d!)8{FK8PQ}&BvwnsNt@=fwC zDA9cF`RYdYwD7aVVD|?$YdPWO*lS%&Mtm77y1lj&IX9g!hvD;Yc4ngceOoU=q4t1s}P;d)~@fT*f`!X48W<${mej+4AgwAQm)89Cfc}5Oue1 z43CW}Sw5+x-=sWPqBtN|DwT`pWKg*^d4H$7Xf9@?eA~6)T$2#G^eipbguszwEuu## zg!9xCTd{!t+&p?T_+uW;UMgK}0cMRwyDBA)M9l*>*RE`1j0~?LniV_oNLJVOlr_)$ z!<5|~Vf=QbQfn1mlM?8$i!5Pgr&(Xj9^0?j8203qk0-EyGFQ;YRd?}u#C`EEN+C1~ z1Wp2JwYS8x2QUS;*XPb-zuiHnG?O^P{p~G!uxEV+zBYzN=Q;BJhtAl|z*P+MkEI8# zH!@^}0LXSVhjFF_)pAKFq6>(sHv&W93qwztl&mt-+sRORN8h1PVKte1!+#*Dom2WU zW*-K%2hsYFY1W7jA{oeCBu;zo=5_{9R-=mMfe0Q3A!4+vuGp z1hzEcSgruLDW9_lT69BPCh-^_CSbUroa*D82Ffs(@H_P=$eGiP;x-6IO_%<)E;l47 z++D)030)TN1O|9-zreZi;~5nxP6yfRCTwc7`~1LwG8OeR3P#hNfAp28Iqa@Y!dAxe zQ}+{J?zVVZSQwaTL@0>jTWPyKbzxyWMTi_&r&>!*XUS_#&zy3BGo-u097R z6FGvdqZgQFw0T0A!W|70X=J-Weet@9#`w1CqZq*uG&Y;Lg5ay(HKi^pwQ?awoo!0j zv!*gao>-s(b(&0gp7}wV!oDoSRB~M+2EUbiHeKVCX}7^Kyo;OSvBDn+x3^-Rsqmkj z#k{+^K4{d&e>WHhPfg9DUz2m?ik!mTf=kp-{ciHaQ=sX_Otbf2z!K z$=)^wg)pmc?}(i%d6>_GXo;w?&MQ`2(tEnf?5hNi@9Xz%Dk|{T_B231piP56YF^&|^YVgf=QIw~16WD%`U+ z)#PkezUQ@h)pQUnh|Tf{`sCO9v7nLrTs?*^uxwF#T^;= z3+%~3rGlQfB<1boA#b*dN%|VQcQ-~x5{tXNWSci|FWrd`a!7c+LXOr^?ZKU?gnb_; z=p||#iUzZRrBY2y+JIOciA1W1_@o{e7U1XtPY|j2yejhVrdK^hdU7Kid7?Og2sBo5*Frtq*oI;nZjyd3Vxk zd)h?0P)-jLXBp5s7HmDJgOXL;r-mUw6PC|=GF7joQ{J6*=LgpqZ|*jk5_;^tBus&@ zz`7WC@951Dl;1iGkrWhT;A(1 zoq3$_#*anke#4^AdT$y^5pD$tCT>dLW0(MVI&+(f{LF~Mz8%LVu=~@>j2%mg7y=RW z%GCGWMToJ7gsBR_qm!r0C@eE7{(6$O?Os8K=57G2W}p6jsj~#BzBPh$_-Z87;NVSp(Et-#ID;QJ*j2DfHBEC}x_<>3K$e(Ce(e+IK-;BYbB|B!nP!5a z$i&b&;XilF0RW4+OjM4+ipT8Rmk$q|@0m2HHF8^gKZl(hJRtuxe6)0szN_Gvg8!*C zomUwQeX|AmT7*tIi#kKa#&<4$wa)bsL`euv7s*H^zN-QX5DMXLCHb^a7jkYZ3AU=W zUgh2kqiA?#OwcD{sjqZ9%f_fPUc_8B$vWe-AnI(!%f8gj6NCzHGnxHqo{V2i@1<-m zb$WQ8m~XJmd{)ahb}i;r|3y8&544Q0X$;1RTR9^_RSd=AlPvD=DJEM`3Hk)3&9L{U zw|f_0inycVC&B`^$1kDX5Ajs0D8tuk>^Ux&veH|qHOuf}YI&SGCs2`p&T+tX6~~i_+LQ# ztW$(G6k>Bjsb3&g8f-q5B4W@-Kyl`BE8uMK zDtRjhF+x_*4$F@P}~aUdZJFDHFCMCrMS{{Uv)>s z)#{rmAJ#|N8zU>-J6!o!MAt;S=BxJYFU9DgckS@97*6r#XP8OFGokq|icq3P?rGsU zMp)0JXhGivil>@MyeOw@k1pXNFv5kHDuJ4lh-bzLDCqm;bIa=Bc2Y*_rr~Y>? zdR9UIY{Np@&u^dMPS9j-4Cu4ujo9x{x6pI9(v&xWS8RY!lSC@tj31RxBVYTdH(k*+ zTxMq|UC?J_e{X}+^n|{Ru^_rYH)}8El|GRxgWP2ovTVfW^9|GcUC15wcSc$oXrm3B z$+<0@vANubvXyAljh#3?hxBbekY-|D5cC0X>b+pM+L6Mx-l9TrAyG1NCR~EKQ{AlN zwieP!lUu&P7+U3Jz}F_hJ!CK^ksAyv$U@!H2^i5*8L8Q?Jccnr%{m5M;_ihQs@kvP z-r`gKRY#2^;k7}X!TeRL96z!t|CS=dDOuyE2OgF13i>ve_3)OIIC(21f zBs!Pkh@%$e3sih)Ce73_gtZ)VWGehrW4$&(Ag}TWT&G^HlFWi$CAAna`dT-#D-U1X z7Y>W3=?f1N7uJK&3M&#g>U_`6H}8zYe+4|tn-VkauLV)HqLqNbvvO3pg6N%>j}JgU zXeX=rk}15ArWP7B-?w6CO}G*XHpc3_uR^o%B8l8)uZ;0&x;)Vy#dlVOPT#I(5Tq5D zwal4pRqgNDbV19Vse2U6np}Ee5?uXMO4Q$!_P@S&3x|69ZJbsCt#@BP9<*mnPZkR= zy<(Ta4++GThPT&cEiYUsvI=(5RtG6Q4>-mvXqK?Hf>NpTbF!4ZeaZTsA~K*f3*s0Vr)6W z@CQ@jhJ}htler^?CwVzb;;K+IuH}z!@2^Dfb1Xlj#JPkQQokEJde<%NI}o+JWWn#-c0l7;1<&v zN%&>vHqvv|2Jp!_UssY%HHGZf^X6DjX1yR)3&l&Vco8neliYQreAZVS)z;}l>N5(W zh~DT_K9V63ahc|MHsapbSI}zhUGfrR6{9J3d!{7QL%MGyU_egqNq^V}tFtuX$u?~! zOIEsm4;F-okk!h79bmY9n@Kx3T1wZ0vQy`V8#+!`GkDXPhCfxGUl3KXv&tN%9c6`? zXSSE_%8i$q7X8+m2}sHvPkKT{4c~{F^*kTJh1XZT8>t`X`qD!xcPK=@ysC4bw;*z* zyRehzT{;G1#szN)|LHIV zbqKt+K&bj@cXcYEeLdK&krlujxq{cl_JMn1W`MO&eWLcYL8dqS|9Qhs0->(Pw#Fz8-5mgTWGTQo&N zq%wK|BSbJn#aiBGA&}v2+k0?tiie)2&4s;f|4{eDn~zJ`EWce4D>S2UpwVn@fHP!M z%Bj!$#@?sYr&he15V+$&;;3CIvwPo>RQyz$#2ph?mf+1y8YyDETLVhKH)5>^6B+o` zrdT>UAY%^+;p#H+ zwkCq5B4koLMx;G@=%lCW>yQH@!sjk6oD`9WRFaU?J@URsg_gmfQ6DstCJxx*L+7{O zbMWjUZ01XRue#=*3jCBalbRL=SaAc-EdWot_{`~)%T|5YpxLunK$xvR%?12Dkv3P+ z#MTW+6zC8t8{~+($$r*8OIizNtS4kUL-$_Rrw98za>^N&racQvvMbH(ZTZ(Fom-W@ zYCZu{CO#@akRiR$o#HRCp5aA+g)zcm9N0~z)wz|K*UBy84}a@KEb)D2xTXfRNc^LB zt=))HA4wZ2LWZS0bqS&lw>Pv4dUTfu*mc_9zB-#ALca`IyD;cIMcWWl!hdJM6=|cw z-p=0GqwI=S%aBE{KRxkgnkHu9zq3y*6p)l4ZRlPz zmTJepP`=U}FXEqiu~+K`#BZr|F$6fKkp$swr=e4%mgF5fdC+*)w~pN-X52Nqy-@;1 zKDH54Sx-rQTA?rMRcmjc_uz4PHB=%ZaZ%;PDo*(b}Q?6;?D zT6&{r8e*EX(xFhQ?BTtY|E(o!-$ANn3`x{*^)va z@Qp_YbJ=%kHV?7gR)`D>krGhV7f&aKN7Oz^-GE55-nc}4E&fyC%nzz_uuHtldx{fj zRrQbYJCCe^6K|^bV{<7Z$}XcVd;6+Yp>JJSJf(Oboo{#>T7>;wJkCJ8AlWSAO0Pv@ z)%lsV=BtdaqjKc;NBbLk^GTGP$N3X`>cGHgjWUkwiVL3M!%~r`@pnGMUC(_TmR~~` za!BbPj8X)qrBqAJJY3%C1=mBuZW7<9U{3qvH0J*7XL{ct1$j2t7U_MNId1u`SPZ79 zXE|C648>RI#K$hh*DImt!ffuu_SLN9Z2g!{ioO%Wj8&mL5FdQ6_wKc)O>Id`Dz<8P z_TMh@!jXbeRl~{!$c0Wq?(J#bXA5UT%aC%gWxa)+yJPHXvOl8wEM2$v_-PiN=#7Xb z%No6es@9q4YmP_INCLq_zf)n-eIBBSFOEBkF_(Nu&#uArkO?xOt=Js(4A;*rp)n?% zIKz0g!5Rq)m$Z^`>IL{-N$}G}srqMU()77k9^Y5uMB=d5Wn=u3jwhWKqtc_H5O#0B zO)}oYDMlY~*)hL`#(dM9j6UDJSr$Fkhl$Hy*pPm?>7*-`@70UvyT~K#cle_O{4$T3 z;wsgfy74Lnum!&h(lbssJYgJW+y2&+6Z#XqZ-$OX59YFScg){BE&lp4cgO#GF=X@N zK7A4KV&YFEDHo8@0MUiPE3)M4v8>J@MSvU&}rjuPPDqMujGD^?avML!C-_ z6NI%WNTtzZ8{{+A?SCxQDPu1Fh!-K3I7uLdveFtfYq*6ja1w{qAW=Nj&;88;TI6=E z_53F#(rwH_@mJnP8WiqAeKUi_A;XSJrx(G3mr{9m3J+1)XBskECHqygrflWAdNv8O zSv&~&sC}BYwXU!&8a(}B2nMsLe54uONm=A$$>OZU`bc%G!)|^>+GR)`Q#|drRJ}QKvt=h0d(^%jZQk9}Ffr;$cs=KnSBu zb37c^Q=*Ia;hz}dE1OI+OAUT}DXohchi^Nt0EjnFZH-rx$joh8{>sHFA@ZnyhsS!*9s;+^R+>mzrv8|wh3WrgJ0f|UAVI0bb%*+#xO86)$C5Vo75&k; zfV7Ftb9=P3U)VQQaRgDWmpe6b)2OZLf}3n0G5Ts$C3sNX(B|t+{emYel_Q07a`jEM zJZ@5(SFz2?TrgLO^ZT~+-m1nD5gmQiIav5LFtMG(I0M7rJi;XzAiTffSe_S;EB6@O z<2N<9jaVn-!d_&d#vy9@s)jSLwPH@L;{8JL*qJGd6)*()cV3_C;2NV>UT&99c6vKA ze}~{8_I1a}P#%cK&g*#`bs5V7FrRgF=aoB(rId@aotA5;76J+L8$SBYMSahZgfvTCIR9skY&QVS&ha3D@{2?I1VR^52nIizua*?4_3?g>iT@>#x zY`W{vNJgB?AU*xA;sGghB*Vm|qAinfC6>wA-P=f znVIXZ8Yu7@--|FxSb$*LH;D6piMyI@!#Scc z36FiF+yFTiYc^Ov6ubMsT!+x@s!pU=5ke!6_V8E%?;Z!{9yh)|kTT6unwV zlHHmXo_X{2sR+3q>5zLBSU0JErhG$dqko$^aAx5bFv1}90p_Z;+1Wp84svi;FB zyAQtnd2IKOp~;tIn3~0G)A!PDsWZ7aHMy9m+@bTnyu=^8&+mLLMs7Pe?Rg#qqz3LV zkEu%de}e<3&n2R)xubU!1H0ZPoqT84+;=fsB_`UuSCJs?i%ArmxXmd6FQ9Q!e^xhc z;MMg($KBB>)s$DF$+YjHJY6fF!%4Ljxi;-^ZrLZ={_?}#YgR@YOlZ>`ac=qc?}CaCdsvXb!Jlxu8hjZ0Br$~)$g+h> zT>acV9grvyt`#u+QDvarb?d6AdQ5gJbM$#xbdz@{h(h4#=I4okIAcQBI~baHg(+2E z9+zgBJJ7NM-HE75yXtdocXMe>+wJ+AqW&L!n?inEHz7%CWuQD=xrFPKERizel5r(= z(O#Bx*^5Z;sK}#XUE83IYKTTw)9b&v>U#Q}w(PZ2w_v9w-vU>CW-<02HL`NLxIQm# z37$DGb4vYeNRIC9vGSGH^A1c~>c=%jJsg*Wn1k zT?#>!&LA;=7xk3!o#y?E?EXD>f9G?z?{^G1A7QZ4x)FqAUVIbTzSvZ=F{FfR@cDC92W8LqBefk zVZ$1-Py9AmSzid9wHAo_hKDxpzt$>d~DH!VgqC)9L zTDBLz;n!GYET%v-~kr^NUW(-toP^pxa5!&xK&8j1I8T1cOggAjvi?Ndz8#aA+_ zEgY;B&}p=p6u$s_)>lJN@%VzsjC5+IZn|F48s$BOENH~oS-w7??;7r8=arRX_>w^y4q}+3`{+fD8Qvr4)eo{t{L-fw|`0NJS=KD(vmH0eF6z(e8O|BN#^HfEonKIL( z-Y}$R3zm!y7)Y^&7%A7)2#)7&0%c%*Rc@RA7zv{1dG+Pdk>uB*aq7< zA--HneC^r8T1udGC-WdSAj_LL)C%d0kOqI+;K8~4&)R~i_X5TqSh@4$CF&wEVhQ+{}rH?t>bUZ*#rl+8N!wFPm8+N<2_wO zHKQIJNzLase3T$X`LWNqr}=KY?}tl@2=o~$PxCj4qCYu=P(u*x(L_ zz^Gm&^Apvk0+Fei7y(Te{u_hZB44fa$kmQ#PoxvOlCx_yyVDdkMVcl_M!x>Qn3>q{ zQCe==Dt?S%Zo^8AbLOTD-Al!%E5tuqrO`+8L^<;B5=VPrs_r0k3(goYG!4Fe)W|>f zn!OpWE-XA4xl3lpp~oX9%9xWR6wPb_p3C`+dUIBf2NpxdcZ1{K-O}93H}`XSz{o2C z$i@L!Q#hn7Iw7M2n+LTf-(yrz0Fj@;di_Yx6382O%0K(67?OzVN{&aRw>;D{lafW= zhKBc?28>SjGKOCU`~%qJ-vjm?m{O z{Obg)Hih-L+JfV)OfDeg?hY-)*gLYw*W5tn4@IiE3bz4xui2RJd8GAQPCV)K`V5qm z+a@qU?#N*ATsl|s0*}r^53aTM)V)6PJ{;-9FnGaOK`&ja2bUQyebDZkjnBPhqq9hS zI*rgG0fYpcKC9Q(|-Ae9Fjzta2X}CgmE_DlhV#_NAbRJT=c#F7Br%8D~iHEY)mRj@s zmoW?al&=z`vj_n>!*|-WBxK^=ZLS(9_XaHDzW?Tt?or>Y6u#uWFi$$h$eJ$MT*S8> zDD33(0ij}j&evH0@uZqD^nT`l^Mo#5LRu`2B{bi_3rWn~%K|X!o(+sqL%>Ly@*&TQ zMdXYtc#H?a0 z{Pb80vY)QuB-^HbJ-L~*-?(+oVL7NYVTZO_B;0(||E67W)gvzaV=8({>hL4Ci&uc*+$dHm)HC)z)BN+0}f?FmUQ)}1Z=u=yL9RB3pYVAcHF$!Tgu*`rN zO@AL_qNXhSbhOBj-lHVUrmuM(yunPR-(n>%pTs870CFV9mND=OVoMJj?0UFpLSLi= zxJ7AS zZm%&E@eD+PZ>k1{`V%DBAFM+h6lL|QXX;K4%5Aj4F$4~bTe<;OB@G|x9p zrlucb>c3q)Ns$3fjb%^eHwGqo@Jnn4=4s(O*=~MyOEem(oRGYZdH%^5Zi>Mj^V}yu z$tK}kOP2ot$5T!?yqt&c^i6F6Sn%0R4k7QUXopHg2);+oT$0iYv#j&OZ5O@Ow;_@6 z{5>51AjOWLO}jlC*Uc=u)7>-v#?X~O?wM$&4+?QYWClz;#wYFJ1`tzG*{DTHFEbJ8 z+iMgMEuus+7FcVNAK(c3B{btm@yrAim6Y9yk+Kw63u`aNq9K#YB(lg`|N|ZTwH12`3k$aBno?@ZFIXW6~rT&cUr^E@eK4pbQ7Z zhDH3Dt$y;_!mp_uv4`=XEX_kBTVbi)@x&@^WeNO z&jfOT*8y^h;A4%t@raE zr>YJ@oktMnh$E8EQ&L#OKxim4`nPuukR_t!@3SPh-6^7Oh2u+vt0;{_hPgl~pTy;C zD#E?SKhJaac@i7`(8?1DxMpL{6I5n-Gwo;x(diq{vY=KJE*GcTN_4O zMcpJp5D-+@B*~x>1PPKPDIht6ppv72WRR$!Ac9B+l_VJ?BS8rQA~{J`a?Uw^W6|yT zb)Rm0L9dvD$9KeVV~v-Vtbjy#_6O#P3Kj>s6*(=MG>SF(N{DNgz1d0=uM=a&Z` z_=Nq*I2r31Ic{|4AJg%si%!75xA6Uwew;rSW17rRQ^ci?DkF*I)9U@ud(t1NeLn<(3ynCOMzM{UZa6V<~Mv)`iNna+AKX^_KMfGC!C^qvQrxaVK zlC3CZ3O-FuyymyWyVXkRaTXO0oSPV4wQP8~5ybrELcJqVfcK+SaWnvtg9gZo{`%F3uph?n_=e--;@p7XC@B62zOYEUN8(nJm zl6M*b7fY0U*X!sT9+N}0np%~g_9)!B(v!GIy2I*9SJ1#;$|6wl^!I2c{sJT_V&pBF zrhsL6k<~$KrrjoiL3R1b(^4oyigl$nI}glLdYy$oK;UCvqj+)_Y)Yq&A5e#qr^HP2WxeO^1p|YKwJ|S{ku%AkEQd@qD9;$au?5NBh*_Ae1~( zt~cqGx+(^RO~vQ&_@M~0Q;3}{W7Yp*>k4Z~>2-BSV^-@AE(ikqJCex6cI^^C3x@+xjlc?wt_+PVzN z03nq8<`tvrH3;-nH(y_I0w}-Lcox(B_c&$UTxklr$6FUrP#}p6Q;_A)%L=QCT|~7K zhbVtyTC>SHD;#)9<-MFYTEqwE@;TEicS0`?WvdHrJBx`oLs?~3i=dg(@Z>agRO=Ye zi)or(K(*aFvn3ldc#mwQ6aZb(6&22aiV#cPaA`2;m;zU-Kj1jcGRT-KHP0UI0;t`H z$+o0|FPVKL5WL-k;LQs4^;O@-OQIvOgEG|i7D%Y0+w8YL#ivo^?H;|V8~J%`mu3E%LgKA#7LQ>h z(d6Jf;)RxWap2)W$^Z$N-#!y}Nv(?D@#ju`6gfYXkkualHW*axDi&^BfGdF$R6Y)p z^TlT9aHdRMCSre=6O9sL)Cq)!Q%xH5tU~Y@b5w$za0G`v=Wr4-O+PNW0`eq~&kgf6qFtO_TnzX~dUrCz(II3NSRc zmri?3Y_dZG;nmlTm!wA`edC9uxI7S{^d6eNyIl%miC{3EGQWQ^o7_3On1}B{gGm>4 zCa`H_H&n;f=*(C~&?r>HkmT@4tW$C;|HymU81@mNxPgX?IJv~fjCZjAZX(T5hL2pp zCERo_Bco65v{KSf7%HZl{a!a)a3U$V2cRN`tT3PD_t*|*(v$m&s4N7UAV!r# z1Thtp@C!}qq`t#Q(aih$s4r2RCa=1|@<&01Wn|cDeBN#U1JCA2aQ%I=C3*}V=;bzm zPuF=R88<74A{g`IOu--dcnpuD|AU>K#{zVl!vJK`#b8z>ily+@l9rXrxSc+3n>2?y z>*G=Evw0M~c_s#-3!!n?{9lV?J|gqV`Y=VsfkiuE0*LYC$>ukn3S7mZwz7{oVwHZt z1SBWcxsZeXeNc)QQVGGAKHz_Pf=FYpP&dyA2lA~FIa|C%B3-`6&JGDL$^c31K-vrN z2~S104kZ;%b@Zs7KNmC#%woBwB{>(p_J2a3Ox{C8M$(Y z-L>D#jE0nm$2%T-U8zsUpeEp@2@V zK^SH3n}HH^e%;$)I_U^Sx%d!qN{LtjmuJdlSD#X1U|Q2c$&N}EpG?sv8{~xtOL9CY zUz5yNnsZ`G4^AxK0g$DdL1O~`4O6AUs@YFO*7qSsP%LCxH1NJFr}_$ijGCskD(B*P zk%*W~>#vs|E4<>GpM^4nLA33wN$3Z+ZF(;qRnRPK!eAB-)QjkLv3N)+5wnF=W+z9D zC8Rl+T$a`XQ!a%WddUk8T8Ne zS}*lhuUiIf=Wt7>m(Rm^o|xX4@x>#BCDXr8EjC98204^=GLQ{xe1Jaae&1Q8gjYzA zYb-0Mi=d%HtB&tO5m~>HXFScmyolNYpT6D!FK;}ml zyMp#RP8UP3N@^}OUk>X*QBo#dFnSyqA7m1w+;*pRqL*+d%{!8h9`_x|(}z5T>08iF z+{NgJFqJZbdF5K`>b2zJH!5P9_CLJ!^E#BRspO<-0wH3^XnJ(ySd_gh9Q7xaC|;#_ zbPJfRdr({|@?decp=^E!b-_m zwYMU<(v5jKPH}I1_iQl#FsnS96|^1vL}B?4KPZ8MxXGx$H`|E0Ui>$fPpMlskti2z zH67UM1_v!3Cd5_~VH$G2R$hM0fy??bqH0AgTrt^Ia#&yhyuOTio!QBHrjESRYef&R zV0S~?z~4?M>{)o#m4Sn7GraV*S~wQzsTdGz*~o>aD6UQf!=5&I8AS|uM6q7-^*9Y^ z1Lcj!G)zezO8MwG&Mr-MEE1J0hQ*rZc^wL$l-!b4XXq39PB*!yV;1^cMn;9HC~F}$ z@3>ZpF-UXg;VQl|qZ0+5>YLam0m2)jqP+=?YQ0D?;yzL54+O3uTt#GlbBmNxLSBB8 zj}`AARd$~L4(21uvI0{d5Oy}#Y%#r9i$(~Na@FXT%xWYpghp4J@HAzR!YvM$N4~_udQbvUwB8P&G1!#F!xxjJ`dHB+TopZyp{b%#_~= z2Xi&OK@tVnoi`SOK4{S$k${`x{Rv%*eULhK;v;zh&AfOET{dv-V!k_o^e(9!#C?D5pe<6 zzL*yDKb#}wj*sGVcu#ue-V>28t&+~lYCd}xwib}=TN(&VZpPm8O@~a2;&QS?v?{1QgMG-I;Fe)Yp`M+HfyDJRFHo(BvkC$B2qEJ!maDnQ)7-g|@ zgLu5E`A}Jw%pb5v9JrEV&WRsK)Wt z+`HMzpP{GR%`J{7bM=mZ$19+YXr(W?#UJU4Wc)TMLr6F`(#qv!mR(IeRz)al^M$dl z%9z|6?J?=N@0YAxZbFOKsfkSEipUD~21P^q)bcw2siYy7b+7sI_!UQ z)+UcSC<1DiHFu+)iy!ksOJWKFB;u7C0C%k6G3UST5rNb*QR-wynRY zck)2j+xbD)ZQXp|p^?PF2307-{VLU8ISCn2rnO^wcC)~|Uuw>}TQz-IAuQ>6eH{(c zRjhjBktP9cjS);Sa^0y<&R*A3%ByhOl4LU zRC@8CK_JpT;e`57hN_shH(Ev0%3BR=J=1%Jf@KH_t|BQ_QNO9h?StYmwh)Kf8%b;u zY#!Yo?CWW%E?gnOjZl28}iXwiNS3j2y{m4g^>RG4J(w1`Ih~Dd7U1 zu`rPvYLUX*MplZ{0(cc9Wk0z9n9^ilVH}iiM4@~Wz{-_`^?Ui|dGEIh=ao%}IaBFZ z7JTj0HD8tW+H);fVY2ByP2KL0U(FadQ=d#xB4y?J9d>>Rimq*jL<@jQ{vI_C#=-0a z&e`u#0<&H>S@K@W&8Vof{Z@G5g!8SKyBvnY*9sS)^BN92UaCj5#E7wIwTVPtk$`b6 zqf&Xub@52Gt=(nZEtBupn0jGKIZm}LWAyO?*J$tFJGwVjF3;+_i>UsP)G z>%YD`Df_8@YnWo+@Qiv(bxK}v@Yp1^xiV=4M*DB&{r~~cYGitS^3-`l827S`9C}-K zP5RykG6dnjDG_=SP@YIsF;i$Y{ESbzNcEMrNM^~BRL)B4%@2yaj#sBHbQ=;2qvRQ$ zSrL%`5f+LBiQFU)<`}X3JZ5V^l2X}SjScet22tm z$l{1Z)VfccGrgnjFiNysO`;R8emu6L*eJcsmgk2|^fgl4#?3I*K4=#Rs3}VKCfp>6 zJDUdVNJ1TVpW@x_N8+BL2H!H1D^eL&qwo7WoU6G)H6QpUb7e=XP}V}de|*&CJ4_s0 zhA`B!G*^3azPuivM zBqsC$L%=5Cf`c6`i`Ecbx}M_BN^@tDEJa(6f$X@cOKjkKdIf|jt_ zHgL9#vILE{I4%-lO{?yx9mP-I(bV&$eQW>Tk0~;df#wTRR@baTB1>)O#T(-*ZwR)q z+!XO}&s?zlcw`JV4@)V%d(-f!_`oHOAM@AL-VF9&F{NGNmF>k_0$8utQx?cU#5}Pw z2;~?R2#q?tq#XdZySI9}q1*oEgQC6BR>ReNXZX(DJ!_BI`}Z1-VkKUS;ob&8eK{?`Cq_WY`sw zWLQF2dW~;=Op^VUs=^kB$*rLQxXUPhy0XR3=fsYhYSaVt;+5p$BDHt`(n`=<91|U+ zy|-YbL&F&VRfvrGEI>%YOFOQn#$GmVe>ZjV1fL#MO8q}`=J!3Qe-~TYYyLrPlw26~ z_GlirA@B_--=DtYAkZ=Yt%IgjWuHEIv&2qGjQXG{km8Mh5AvO>2n4&eH>1IdEK_cO zz~G2nflft1^(~h|YSecb@DJxl%N~t(b6FNqFDGUNheum~Hi> zvis≻*7PNoShG+1pRK!^tpI?4MK+eZ?9M&c_!$b_E8%EL+0q_8D8e^yWK7#H}Wp z-{`03bxLV(5{Q4X-FhKTPTKy#Lzyrd;L$n5avx-rO^YWJeffu8a&yKJwIL(6!Im`! zRHC+$*Caz#D4(1)dMoFtm%-(*bT9LWUgQI!hoa?9)l>nvloGe7`rP=LIG)Vu0r^&P zsL7qsE7)xtemHr44-ha(zBbRuoR)5`W+~m|d1R_Rj;K=g>`~*c_|xwbvK~IT>%$ z5za(2<%^272&lY6kx+8tEG~xEC0w8D04Vm&i9qVeL>>a#I5akaPF2v2xyX8T(tWBA zfQ6@8UO&0!U9Ak6;=D0VhnlCZdUnK#IK7J@WskW@Ml&&S9Of5h1zBO<^2Iw*T@$QI z>djlq6ET5yr&QYJm5wr;%cN6xNmMcfXsESh-;r@O)cd*O=?H-I^q#b~X_9gtFJ3B< z|3Tnls5AotWB3zHDFpNB_oIEqdS$dBc&SR0>`@W>XDnahzyxwou05DcE{@|zU zAzQ31&oIstcos%nbSSo)kFh`6n-^7*5X^0mobXC&G#AQhHHNRbL%!|@PeMRxiuhHf zsy`5GFSwtdx^OKFL2|6os1JD^Kk52M>*KgaklP9_Ef3JqScAGLu~>pUK2v_C!k_i! zh=5tDlp0iCM3b!6MggYH)H;LoBxH2;X~V4Y;)fn~?Jy+-hY8JEssy8j&FUA2Yd+}9 zU5@_D4txd_)Ti%e-=T$81S5-kHlN$Y#ksf}=OZxINk!ztmr@p>S~Gc<3T8-oo|~nr z16cZ0!twiL7B-jpnQ~Qcxn!r^+C8yKVajr|Njl$l)E?2=JT#m%@k@qdNRj;lL0}C= zy4PY}{AF39p(@VvH>@sbX)eb=LL@dP;c|;mgzS#;tvkLL&SsGKK36tIDoI1XiLk(M zA_>DDwl@~*I324W=04zMm2vlf1i7_x{AXN=iF_TfnPS0b^Do@-wrCLyzPNVNYBfCV zb|Ka7W>`ju@*qN`LX@Kh$u@OpVS=a>5h3yHEwU!%I?WRwQyebJ^I23U${2TCY~s9Z z&=!~<{$yE;G{IW&yk$pr&#Lqn5lT#s%dBsdlDBypA+SIcTl(!3|Bt(fLq1c{ASODg zpsm{jY&f-PB}-qtBh#+e^JiCuJ;>CmThZ>wxfA0smTaTi;etc1d}G8$P(seq>Uazc z%Sn`|mGH0FIrzz)Wu^!=IF-JUKE@!Dnn>GYbK&Fj=?OxrsgYn=E@Y}GDx_`SxJxWp zkibdFv4Do@GzC4Obr_D)@{PoDL$Zp|-S9$oNtF0cWQct3?v?xiM_`A+G(CYNmO%L;t zMhc)<^9Vd9$`tG`vs|<026&wZyb18W;8 zNsAIX0`A9j$TtOqOW2Rjrg12~&Q1D}%JNx4b_AF9?BXYvhO4d*l9!06B!b`J@hv=R zVa%!s5iKr9h>{|dl+~&c>V+4Af-#1S;Z66vPb7>Gf3vfcT6F?5>Mx21phDTyW&vZG zwaGR*g7;qZ;;C{$0@h2c7%wJ-2W(-|o)SPbg+@yij)} zQD^;Os}n@jJSlU%BDcn>5~hpNE-^W}PRb6ZhKmM1U;&k;P~puTaV_0GLtNjrve)&x z^r99tez|ygrM)KldmwV6KBE0(-5D8%F&Lf^-oW>Ilo8V7AsI`@<$dZrcQeAlj0=eyofg`{UYTbD8rMSBxsNH2x zc9bQ=Nfc@s_(ep*rkE|12%f7D_I(~ro@dIgf>?sY*Jae3MRGe8l9xrEcMY$t9zzw7 zt5~1l+q-#cJn}MZ^w5>|v6iz>(~+`rLB0|T0M0C=!!A)Se!H9O)S-VqNzpOZEThR- zbXTBV5-4*xsF8Ayy%G-{VkIy!`=E?Ez$E#y=6zskFvoIgXWroyk#on`wSG;dbi-DPaB}H-9te}qH z9@Td5jjZ7#8r$ZNvDdm-$9UDOR$9c*RkNHjia68BQ|F&Ky{n!g01P~$FPCKW97sH# zyl`C?_~9mH)hW7#&@5s8wgQDnF|)QDsV3%KqZ_d?Dj7cbRUw+2HUwctH{4!9l2ld8 z9%yo*6@~e>4ue0*uRq{J=Vh5B-y^ljE@{F76RxMabrgWehT)m7C<|F{puOrW^Qz=ntlM7qFj)sN-`5Po zmdB0+9%q!L=YKRahm@DN8%EVuBZ<03Z^|o}PiOdhoxkRkUT~taFg~uAQHNh@)~B&3#tY{lcEBfw>_l>GTwzYrg^tN4`plpj)BCA@ z9~@G>ydhl5C%$~=k>TN0SEJr`JW7dkG(ucn_~u)RqN$=TC63HFU$2b8R7hs;xYbuN zVHsQ-Vn=7c^~LKrM{ot{O_d~4iN7%2Q~Pq0HGkCj!+^9l_nEhiT~%FI`+_Vk1SPJU zX@oF=-9Fe?3*a|6*wu`}AuMXPXti)?H`^iuWJuE?R%tc*OFG1LZFP#0H1jF~!ShJr zDA`QOgz|~G(rTqMU;jPYm*Ob-QK3rn-z6Bj^eDf|gHk57J5y#!E5fg!oph0nV5*re z8)&Q|VJeZf?#=z@*_byJ7zvBm_#wQxIh)dJ2KE*8sBa${;wj z$<3Ecd{y_nt5^|^8%CzEFCD!Px1oM->T~LXaR?4&ZD!#qC|~4&*;LK0-@4<4p42>2 zZ3!b+3fIEvYCBX=Z>gp>9nMp8|1s+{l({48`$+T|LfIDBm*-~(XllQb3&-m&>0T^W z0Um*{g^3=D#EwW9e6rQ{)l?l0mbu1&<)l)Wu6_Nnp+?BHt1vG9j$7`;G<#2@*-USUWPgJ#v_0RGdO{{`{^J0W@5#!>U0 z1AMG++F`^Kj2Ij|A3D`dlz38h%!{!4p*j48;&!gMRnw?dMa>!6ewA7)ARI%+VYe0I z5{*otE+TFMF++Gp&nd{-y;Hi5c;5~t^1C}TMN2n(PAzrHX?-r5o931#)Dkaa5@Nbs zvzcgq|EBReC5GhqQSY-Q%gJg`q5hJ2GFzAD6pW>FEoplt$O_IM-}TOdP`MWeNWw&w zl^tqX_F#T-o=B=bMX=tJt$2cJxuATv;fm5^*$>C=^{ne`k`gS+Kmvno_tWnIB4$>j^rO1hi>IV-YWEQXkItgnWza zPlU<09+9mVpW&%>?hoBi3YD6IZB8?cLtrBcyvfqP(Eum|IuVuU=M|R^Qu#9Fc)hH{S?-EBFQU z%EZHPSRO!YOYQ|SeoG27Tj3Bau^0yqyvfmTxKeCUzBQG)f<-ybg#jaU&T*eyDcU|H z0^f3{c9QcKW|?(8Fj@2RvI)UT^F({j<_PQvj(=;Zfb@-^;8s06N`}mw^?Cf~msQfA)gzF8+xS+90 zziB<3Jf?xKGn8hsBdjGvP;Y~qN0LwI=NhMwGyB*1{p_oT$HZK;IsxUt0he{HaYs;; zrx|ipCMcmE?GJ$(96TwKujWDF*ml2eh~g7aVh#$G#~=>qis7SnYQV5q}y+{7eoBY z9TF_oD@7!P8nWS;VwH>WMlV(yzA6)T7NFc?dtXZ>w(e6R(b+1DC8TH}&05c1>w(Rx zTHlt%$OQSTFsTl(#&N3s79VJ}^g6@`BB{QwUJVrr~o042?DGZtB5AYQ@~j_Q!ow1-56{+4xXGUxtI5^#DA? z;wJdn3Ev~$?Z`@F5WRfUgt{6SkEwJ_lA@IT8y|I2+YFT&q3Uo)hWm$C1dd)lV;EZw z<~x_+;Yo?Nm0db+E)QMRTvheM@4wwvGcU-wggZ6OCf%0T`eie>WD|Op>~cGfm<{k| z4z$aWR}r#J7vPo>9;`NgiU1NhS8)i`QB)U=&wUwIF17b%E|US>HqNdc1==!Bf?J73 z;GM6U^fH=I)~cJN6M1v36sfckQ`Fdspa_D$Bk-gn*8%Lcv+;rqk-JL5Lcj$@4b@2< z9{|3eq~-m3oxP!v{1q@tSHRHaD1>(qBnLBuzFFlkT!IFtxHgdw9tC+jwiG9E_4eq) zQUEVCF7`ZKPF>?S19#CSnu38%AaY+9xIJQ!!KH>si?VR3MP!UrxhRQKO+nMBKNSHK zwv=uG!iZ!(e-B_qSFSjCH=|uU2hL!bFvxueeK&+qh=BRxL)|@Rb7{fs#*=^=MYsTx z?ChueQrvEW*vpRwY0der#w07%o|6`ac7uT53$5L+JXF$0p{|vM@GC|61ukT%WNOP8 zwMoVMUd*F~BiPOLGYro33dOpR7>VIuF>qo4sMS*d#a{t_u7OS;1@9Qa+L4j;IE z86XwlDP4%-LNdp&MIAaLWNQ>VHH$~(=#Nm0!jI>#hXblv?3lhPHcTSoTyi-WT zR8Ag2U>?C7-?EVgSP`)0`)BxLTwT3G1~=7yn|EFR&`uX?LW$uUp293G+$nWw5h^XB zOg<1FZx&)D?`2lsNsKuz^7s@dL@78ASpg>bru|;rMR}9iJi3hN6OUlP((SD8HcXUj z^D$N3G2YObw~&sX?q(Lct3IEed$MqJZVSfCISF_YY)=*IP{5rGwM^aRd#jmb+O&R* z;&;H`b1t=x>LEa}q9H#f&7i~>F{XN1zC}?4*OjFFYVHTj>VAfjDNAyQ9P8cEIL;R& zjLvr!mybiRx@5Xe3ndhmPhzu7x%|PF8twD_@X5G^Y^L;FdE>W%9Ny^5{^tF~Hz?0@ zgvWSG((eHcUpUBLH}v8glCj^uynx2zn`?|K*L{%#O3YUR za2n0b1*jI#e6cH~0Gck5wgL;Ct`Xl>h>N^@58!Ev$TAqIP9e?a<~|I~8>D3R$|a^G zya3s000g9dv?5_cL~f?Hk);NfU3HeM1`QeyOTN{FGf9)#`1ZRPX~O5Q7t!5-R;6Aq z!e+@0Y!R%2)Q&X12iVA}h&s{X!WSf8pK&CB7{8nY3BF+7^b#^)@HDO4O^eMplIz=@ zV{9Zl-b9_R2aq-kYx&NDV4ru@H&71%b`^%q+h;|j3+TX&7y5waH_uCN4K;4!?p19w zI`tA^azLf_iGHey-y-!tzTnp}Q(1$aKM$EiL;qnHV7n*`O3+k$`$N+>;#VG z+HdxNKjFzd&&F3U)HJm$VLGnob<6tAD;YBHk{JZ6H0C%qv2I}ENQZZh>noR6jKuqh>+$qBT-wAER-ZEK7h zJd9FF*mu0%6v0_fzcT;~2w$NBJjrr~mvbm;A{>L+(^}&Oz>Oc;uDY zwE$!|qB#v|lud_>mK=0z&}WaNzm}svUCq=r5^=Fp^VTuyh`?6ScCcUg-cBu3BY`nz zG^cJv5Cgw;4^V4Khw2EY8h-CwQ=cX?d8_$G2dFVR44bEbAHs|QJuIs(NBOD#pzuLu zQzki!aDqZ(_$t`p=iyiGOSRrRuX6V&!3Tu9S3+GKykUiXYp~(#b?g zdE$ud?A_)T_x4Dp3&ytxuj$_-9q}yXAwN>)mALf|b2GPMxPy>l z>ZrRIwV=+MmW(hLmyWU^d_4#wd@;;3bS+o-LeUC(b}SEXe_--jsN8$jr|`tP$`NV3 zD7ioD2_?8R0MHFNH&xxI(@3k`d@Nt7v|pwDBsevo1c(Lv=;v~us3VBKL){Y;JT|B< z^p;3SKDowiD$v-wr0TZ_(GE8UuVzt3kLNh7ijSyhFC)`Aaj`YayLQ|2 z&i>B$b!s-!BYtUW#O0d}7H^LTrk`rtoQ&yY+=1d4H(7fde=6oX$4L7W($^Om452U= zFsIdDzLfuVJVaC`+fuTLp}M)#t-mQEs?7lB=&GpTdYbac&6)V|kh%N)l@X7cV`Dz4 z(`bgk4A*;RG{#o(5>M%E!ck;41#~<{R{tuhv zfN&DYuVf)|m|8BreE>{CHu^w@;Wv*26$lCu)T{8k(eVc`0do2b$EemPNhB-#^T*&X z@PpuN0y*js&euOZ0Qu4VJx1{kf$J>b&e%Sbw*~-NaRl%Ir?*lb^q7Bde*O8cHUKA# zZ+`4xaesZ7vG8}8-F{M`P;}-Y5YR=yzBbS@82#6J|Mbm*8PG7T>lhuohX7@f_x#VJ zy74FaKNPgYK6L?ehH$j8N=hQ+r+@hn^7#kiPzs%6<)Op*+i%~oxFhWgB%|El%G`gz zs2U@RG`xYcAK@|n+jlxpKfEVb%1wX$ZF}HH$XV_o12hPmdXl``)+Ak9y2jPz zOt;;727cfF_W6T#3JxIe*Gn|+{P*``AP56;pIgA|+5p>S2w*(AobMjkjv-vOQ;OCA z&2tzP())+kMF-M>C@k6J@u>Xmw~y%DWB8rOc8N4TJe;M?b&%!vxGlHK3rpTuu_m(+g&FEt6^}j{TExpid zW#K@$RwDw6HX-khME=7n;0cV-rIq(RwfzsjHN5c}F!IPf2guOn7{uDfN?bbs@IiP? z3bK&s)8c=ZIrIX>Q6p@jGX!H#r~@v+IbQtw537YooCg80dL%~hx1$o&#ss=T1cUq) z;?{7LlBlcyw51EsqpUhdN&5FCEQz3{DQkggydf4$<<7^ETmQ5IK`iij@xO&I|NWN@ zQfOStTI7xhYlu>i{$`R|w$;I;+kfAV*ZT03b{x**e>W;*v4Vz_wQvoU1)>`IjZxhj zmfZhj7SOj(A`kg=<2BabB~6SHUl_Fn*>gCI(}W=`;0S-EoJ_PMLRkU(_?H#A`yl;1 zN>;)U7(V^WZvY;5(w+8|D%H|=gno-QqZwek7Dy(^ZqmsNVD)>;Kr?^NSLo8;4g`6U z`vNS{%RnmNZ(sRWY5voi(1f5rS3b$S`Ct4AAEX*T!h@*)``Z70$Uk4M)(b56tt