From 63f244d7373341ffeda3c34d65a16527ead222cd Mon Sep 17 00:00:00 2001 From: Leif <4603009@qq.com> Date: Wed, 24 Nov 2021 15:22:20 +0800 Subject: [PATCH 01/31] Update third-party.md and appendix Update third-party.md and appendix --- code_and_doc.md | 324 +++++++++++++++++++++++++++++++++++++ doc/doc_ch/code_and_doc.md | 324 +++++++++++++++++++++++++++++++++++++ 2 files changed, 648 insertions(+) create mode 100644 code_and_doc.md create mode 100644 doc/doc_ch/code_and_doc.md diff --git a/code_and_doc.md b/code_and_doc.md new file mode 100644 index 0000000000..1c006740c3 --- /dev/null +++ b/code_and_doc.md @@ -0,0 +1,324 @@ +# 附录 + +本附录包含了Python、文档规范以及Pull Request流程，请各位开发者遵循相关内容 + +- [附录1：Python代码规范](#附录1) + +- [附录2：文档规范](#附录2) + +- [附录3：Pull Request说明](#附录3) + + + +## 附录1：Python代码规范 + +PaddleOCR的Python代码遵循 [PEP8规范](https://www.python.org/dev/peps/pep-0008/)，其中一些关注的重点包括如下内容 + +- 空格 + + - 空格应该加在逗号、分号、冒号前，而非他们的后面 + + ```python + # 正确： + print(x, y) + + # 错误： + print(x , y) + ``` + + - 在函数中指定关键字参数或默认参数值时, 不要在其两侧使用空格 + + ```python + # 正确： + def complex(real, imag=0.0) + # 错误： + def complex(real, imag = 0.0) + ``` + +- 注释 + + - 行内注释：行内注释使用 `#` 号表示，在代码与 `#` 之间需要空两个空格， `#` 与注释之间应当空一个空格，例如 + + ```python + x = x + 1 # Compensate for border + ``` + + - 函数和方法：每个函数的定义后的描述应该包括以下内容： + + - 函数描述：函数的作用，输入输出的 + + - Args：每个参数的名字以及对该参数的描述 + - Returns：返回值的含义和类型 + + ```python + def fetch_bigtable_rows(big_table, keys, other_silly_variable=None): + """Fetches rows from a Bigtable. + + Retrieves rows pertaining to the given keys from the Table instance + represented by big_table. Silly things may happen if + other_silly_variable is not None. + + Args: + big_table: An open Bigtable Table instance. + keys: A sequence of strings representing the key of each table row + to fetch. + other_silly_variable: Another optional variable, that has a much + longer name than the other args, and which does nothing. + + Returns: + A dict mapping keys to the corresponding table row data + fetched. Each row is represented as a tuple of strings. For + example: + + {'Serak': ('Rigel VII', 'Preparer'), + 'Zim': ('Irk', 'Invader'), + 'Lrrr': ('Omicron Persei 8', 'Emperor')} + + If a key from the keys argument is missing from the dictionary, + then that row was not found in the table. + """ + pass + ``` + + + +## 附录2：文档规范 + +### 2.1 总体说明 + +- 文档位置：如果您增加的新功能可以补充在原有的Markdown文件中，请**不要重新新建**一个文件。如果您对添加的位置不清楚，可以先PR代码，然后在commit中询问官方人员。 + +- 新增Markdown文档名称：使用英文描述文档内容，一般由小写字母与下划线组合而成，例如 `add_new_algorithm.md` + +- 新增Markdown文档格式：目录 - 正文 - FAQ + + > 目录生成方法可以使用 [此网站](https://ecotrust-canada.github.io/markdown-toc/) 将md内容复制之后自动提取目录，然后在md文件的每个标题前添加 `` + +- 中英双语：任何对文档的改动或新增都需要分别在中文和英文文档上进行。 + +### 2.2 格式规范 + +- 标题格式：文档标题格式按照：阿拉伯数字小数点组合 - 空格 - 标题的格式（例如 `2.1 XXXX` ， `2. XXXX`） + +- 代码块：通过代码块格式展示需要运行的代码，在代码块前描述命令参数的含义。例如： + + > 检测+方向分类器+识别全流程：设置方向分类器参数 `--use_angle_cls true` 后可对竖排文本进行识别。 + > + > ``` + > paddleocr --image_dir ./imgs/11.jpg --use_angle_cls true + > ``` + +- 变量引用：如果在行内引用到代码变量或命令参数，需要用行内代码表示，例如上方 `--use_angle_cls true` ，并在前后各空一格 + +- 补充说明：通过引用格式 `>` 补充说明，或对注意事项进行说明 + +- 图片：如果在说明文档中增加了图片，请规范图片的命名形式（描述图片内容），并将图片添加在 `doc/` 下 + + + +## 附录3：Pull Request说明 + +### 3.1 PaddleOCR分支说明 + +PaddleOCR未来将维护2种分支，分别为： + +- release/x.x系列分支：为稳定的发行版本分支，会适时打tag发布版本，适配Paddle的release版本。当前最新的分支为release/2.0分支，是当前默认分支，适配Paddle v2.0.0。随着版本迭代，release/x.x系列分支会越来越多，默认维护最新版本的release分支，前1个版本分支会修复bug，其他的分支不再维护。 +- dygraph分支：为开发分支，适配Paddle动态图的dygraph版本，主要用于开发新功能。如果有同学需要进行二次开发，请选择dygraph分支。为了保证dygraph分支能在需要的时候拉出release/x.x分支，dygraph分支的代码只能使用Paddle最新release分支中有效的api。也就是说，如果Paddle dygraph分支中开发了新的api，但尚未出现在release分支代码中，那么请不要在PaddleOCR中使用。除此之外，对于不涉及api的性能优化、参数调整、策略更新等，都可以正常进行开发。 + +PaddleOCR的历史分支，未来将不再维护。考虑到一些同学可能仍在使用，这些分支还会继续保留： + +- develop分支：这个分支曾用于静态图的开发与测试，目前兼容>=1.7版本的Paddle。如果有特殊需求，要适配旧版本的Paddle，那还可以使用这个分支，但除了修复bug外不再更新代码。 + +PaddleOCR欢迎大家向repo中积极贡献代码，下面给出一些贡献代码的基本流程。 + +### 3.2 PaddleOCR代码提交流程与规范 + +> 如果你熟悉Git使用，可以直接跳转到 [3.2.10 提交代码的一些约定](#提交代码的一些约定) + +#### 3.2.1 创建你的 `远程仓库` + +- 在PaddleOCR的 [GitHub首页](https://github.com/PaddlePaddle/PaddleOCR)，点击左上角 `Fork` 按钮，在你的个人目录下创建 `远程仓库`，比如`https://github.com/{your_name}/PaddleOCR`。 + +![banner](/Users/zhulingfeng01/OCR/PaddleOCR/doc/banner.png) + +- 将 `远程仓库` Clone到本地 + +``` +# 拉取develop分支的代码 +git clone https://github.com/{your_name}/PaddleOCR.git -b dygraph +cd PaddleOCR +``` + +> 多数情况下clone失败是由于网络原因，请稍后重试或配置代理 + +#### 3.2.2 和 `远程仓库` 建立连接 + +首先查看当前 `远程仓库` 的信息。 + +``` +git remote -v +# origin https://github.com/{your_name}/PaddleOCR.git (fetch) +# origin https://github.com/{your_name}/PaddleOCR.git (push) +``` + +只有clone的 `远程仓库` 的信息，也就是自己用户名下的 PaddleOCR，接下来我们创建一个原始 PaddleOCR 仓库的远程主机，命名为 upstream。 + +``` +git remote add upstream https://github.com/PaddlePaddle/PaddleOCR.git +``` + +使用 `git remote -v` 查看当前 `远程仓库` 的信息，输出如下，发现包括了origin和upstream 2个 `远程仓库` 。 + +``` +origin https://github.com/{your_name}/PaddleOCR.git (fetch) +origin https://github.com/{your_name}/PaddleOCR.git (push) +upstream https://github.com/PaddlePaddle/PaddleOCR.git (fetch) +upstream https://github.com/PaddlePaddle/PaddleOCR.git (push) +``` + +这主要是为了后续在提交pull request(PR)时，始终保持本地仓库最新。 + +#### 3.2.3 创建本地分支 + +可以基于当前分支创建新的本地分支，命令如下。 + +``` +git checkout -b new_branch +``` + +也可以基于远程或者上游的分支创建新的分支，命令如下。 + +``` +# 基于用户远程仓库(origin)的develop创建new_branch分支 +git checkout -b new_branch origin/develop +# 基于上游远程仓库(upstream)的develop创建new_branch分支 +# 如果需要从upstream创建新的分支，需要首先使用git fetch upstream获取上游代码 +git checkout -b new_branch upstream/develop +``` + +最终会显示切换到新的分支，输出信息如下 + +``` +Branch new_branch set up to track remote branch develop from upstream. +Switched to a new branch 'new_branch' +``` + +#### 3.2.4 使用pre-commit勾子 + +Paddle 开发人员使用 pre-commit 工具来管理 Git 预提交钩子。 它可以帮助我们格式化源代码（C++，Python），在提交（commit）前自动检查一些基本事宜（如每个文件只有一个 EOL，Git 中不要添加大文件等）。 + +pre-commit测试是 Travis-CI 中单元测试的一部分，不满足钩子的 PR 不能被提交到 PaddleOCR，首先安装并在当前目录运行它： + +``` +pip install pre-commit +pre-commit install +``` + + > 1. Paddle 使用 clang-format 来调整 C/C++ 源代码格式，请确保 `clang-format` 版本在 3.8 以上。 + > + > 2. 通过pip install pre-commit和conda install -c conda-forge pre-commit安装的yapf稍有不同的，PaddleOCR 开发人员使用的是 `pip install pre-commit`。 + +#### 3.2.5 修改与提交代码 + + 假设对PaddleOCR的 `README.md` 做了一些修改，可以通过 `git status` 查看改动的文件，然后使用 `git add` 添加改动文件。 + +``` +git status # 查看改动文件 +git add README.md +pre-commit +``` + +重复上述步骤，直到pre-comit格式检查不报错。如下所示。 + +[![img](https://github.com/PaddlePaddle/PaddleClas/raw/release/2.3/docs/images/quick_start/community/003_precommit_pass.png)](https://github.com/PaddlePaddle/PaddleClas/blob/release/2.3/docs/images/quick_start/community/003_precommit_pass.png) + +使用下面的命令完成提交。 + +``` +git commit -m "your commit info" +``` + +#### 3.2.6 保持本地仓库最新 + +获取 upstream 的最新代码并更新当前分支。这里的upstream来自于2.2节的`和远程仓库建立连接`部分。 + +``` +git fetch upstream +# 如果是希望提交到其他分支，则需要从upstream的其他分支pull代码，这里是develop +git pull upstream develop +``` + +#### 3.2.7 push到远程仓库 + +``` +git push origin new_branch +``` + +#### 3.2.7 提交Pull Request + +点击new pull request，选择本地分支和目标分支，如下图所示。在PR的描述说明中，填写该PR所完成的功能。接下来等待review，如果有需要修改的地方，参照上述步骤更新 origin 中的对应分支即可。 + +![banner](/Users/zhulingfeng01/OCR/PaddleOCR/doc/pr.png) + +#### 3.2.8 签署CLA协议和通过单元测试 + +- 签署CLA 在首次向PaddlePaddle提交Pull Request时，您需要您签署一次CLA(Contributor License Agreement)协议，以保证您的代码可以被合入，具体签署方式如下： + + 1. 请您查看PR中的Check部分，找到license/cla，并点击右侧detail，进入CLA网站 + + 2. 点击CLA网站中的“Sign in with GitHub to agree”,点击完成后将会跳转回您的Pull Request页面 + +#### 3.2.9 删除分支 + +- 删除远程分支 + + 在 PR 被 merge 进主仓库后，我们可以在 PR 的页面删除远程仓库的分支。 + + 也可以使用 `git push origin :分支名` 删除远程分支，如： + + ``` + git push origin :new_branch + ``` + +- 删除本地分支 + + ``` + # 切换到develop分支，否则无法删除当前分支 + git checkout develop + + # 删除new_branch分支 + git branch -D new_branch + ``` + + + +#### 3.2.10 提交代码的一些约定 + +为了使官方维护人员在评审代码时更好地专注于代码本身，请您每次提交代码时，遵守以下约定： + +1）请保证Travis-CI 中单元测试能顺利通过。如果没过，说明提交的代码存在问题，官方维护人员一般不做评审。 + +2）提交Pull Request前： + +- 请注意commit的数量。 + + 原因：如果仅仅修改一个文件但提交了十几个commit，每个commit只做了少量的修改，这会给评审人带来很大困扰。评审人需要逐一查看每个commit才能知道做了哪些修改，且不排除commit之间的修改存在相互覆盖的情况。 + + 建议：每次提交时，保持尽量少的commit，可以通过git commit --amend补充上次的commit。对已经Push到远程仓库的多个commit，可以参考[squash commits after push](https://stackoverflow.com/questions/5667884/how-to-squash-commits-in-git-after-they-have-been-pushed)。 + +- 请注意每个commit的名称：应能反映当前commit的内容，不能太随意。 + + +3）如果解决了某个Issue的问题，请在该Pull Request的第一个评论框中加上：fix #issue_number，这样当该Pull Request被合并后，会自动关闭对应的Issue。关键词包括：close, closes, closed, fix, fixes, fixed, resolve, resolves, resolved，请选择合适的词汇。详细可参考[Closing issues via commit messages](https://help.github.com/articles/closing-issues-via-commit-messages)。 + +此外，在回复评审人意见时，请您遵守以下约定： + +1）官方维护人员的每一个review意见都希望得到回复，这样会更好地提升开源社区的贡献。 + +- 对评审意见同意且按其修改完的，给个简单的Done即可； +- 对评审意见不同意的，请给出您自己的反驳理由。 + +2）如果评审意见比较多: + +- 请给出总体的修改情况。 +- 请采用`start a review`进行回复，而非直接回复的方式。原因是每个回复都会发送一封邮件，会造成邮件灾难。 \ No newline at end of file diff --git a/doc/doc_ch/code_and_doc.md b/doc/doc_ch/code_and_doc.md new file mode 100644 index 0000000000..b1d8b4b36b --- /dev/null +++ b/doc/doc_ch/code_and_doc.md @@ -0,0 +1,324 @@ +# 附录 + +本附录包含了Python、文档规范以及Pull Request流程，请各位开发者遵循相关内容 + +- [附录1：Python代码规范](#附录1) + +- [附录2：文档规范](#附录2) + +- [附录3：Pull Request说明](#附录3) + + + +## 附录1：Python代码规范 + +PaddleOCR的Python代码遵循 [PEP8规范](https://www.python.org/dev/peps/pep-0008/)，其中一些关注的重点包括如下内容 + +- 空格 + + - 空格应该加在逗号、分号、冒号前，而非他们的后面 + + ```python + # 正确： + print(x, y) + + # 错误： + print(x , y) + ``` + + - 在函数中指定关键字参数或默认参数值时, 不要在其两侧使用空格 + + ```python + # 正确： + def complex(real, imag=0.0) + # 错误： + def complex(real, imag = 0.0) + ``` + +- 注释 + + - 行内注释：行内注释使用 `#` 号表示，在代码与 `#` 之间需要空两个空格， `#` 与注释之间应当空一个空格，例如 + + ```python + x = x + 1 # Compensate for border + ``` + + - 函数和方法：每个函数的定义后的描述应该包括以下内容： + + - 函数描述：函数的作用，输入输出的 + + - Args：每个参数的名字以及对该参数的描述 + - Returns：返回值的含义和类型 + + ```python + def fetch_bigtable_rows(big_table, keys, other_silly_variable=None): + """Fetches rows from a Bigtable. + + Retrieves rows pertaining to the given keys from the Table instance + represented by big_table. Silly things may happen if + other_silly_variable is not None. + + Args: + big_table: An open Bigtable Table instance. + keys: A sequence of strings representing the key of each table row + to fetch. + other_silly_variable: Another optional variable, that has a much + longer name than the other args, and which does nothing. + + Returns: + A dict mapping keys to the corresponding table row data + fetched. Each row is represented as a tuple of strings. For + example: + + {'Serak': ('Rigel VII', 'Preparer'), + 'Zim': ('Irk', 'Invader'), + 'Lrrr': ('Omicron Persei 8', 'Emperor')} + + If a key from the keys argument is missing from the dictionary, + then that row was not found in the table. + """ + pass + ``` + + + +## 附录2：文档规范 + +### 2.1 总体说明 + +- 文档位置：如果您增加的新功能可以补充在原有的Markdown文件中，请**不要重新新建**一个文件。如果您对添加的位置不清楚，可以先PR代码，然后在commit中询问官方人员。 + +- 新增Markdown文档名称：使用英文描述文档内容，一般由小写字母与下划线组合而成，例如 `add_new_algorithm.md` + +- 新增Markdown文档格式：目录 - 正文 - FAQ + + > 目录生成方法可以使用 [此网站](https://ecotrust-canada.github.io/markdown-toc/) 将md内容复制之后自动提取目录，然后在md文件的每个标题前添加 `` + +- 中英双语：任何对文档的改动或新增都需要分别在中文和英文文档上进行。 + +### 2.2 格式规范 + +- 标题格式：文档标题格式按照：阿拉伯数字小数点组合 - 空格 - 标题的格式（例如 `2.1 XXXX` ， `2. XXXX`） + +- 代码块：通过代码块格式展示需要运行的代码，在代码块前描述命令参数的含义。例如： + + > 检测+方向分类器+识别全流程：设置方向分类器参数 `--use_angle_cls true` 后可对竖排文本进行识别。 + > + > ``` + > paddleocr --image_dir ./imgs/11.jpg --use_angle_cls true + > ``` + +- 变量引用：如果在行内引用到代码变量或命令参数，需要用行内代码表示，例如上方 `--use_angle_cls true` ，并在前后各空一格 + +- 补充说明：通过引用格式 `>` 补充说明，或对注意事项进行说明 + +- 图片：如果在说明文档中增加了图片，请规范图片的命名形式（描述图片内容），并将图片添加在 `doc/` 下 + + + +## 附录3：Pull Request说明 + +### 3.1 PaddleOCR分支说明 + +PaddleOCR未来将维护2种分支，分别为： + +- release/x.x系列分支：为稳定的发行版本分支，也是默认分支。PaddleOCR会根据功能更新情况发布新的release分支，同时适配Paddle的release版本。随着版本迭代，release/x.x系列分支会越来越多，默认维护最新版本的release分支。 +- dygraph分支：为开发分支，适配Paddle动态图的dygraph版本，主要用于开发新功能。如果有同学需要进行二次开发，请选择dygraph分支。为了保证dygraph分支能在需要的时候拉出release/x.x分支，dygraph分支的代码只能使用Paddle最新release分支中有效的api。也就是说，如果Paddle dygraph分支中开发了新的api，但尚未出现在release分支代码中，那么请不要在PaddleOCR中使用。除此之外，对于不涉及api的性能优化、参数调整、策略更新等，都可以正常进行开发。 + +PaddleOCR的历史分支，未来将不再维护。考虑到一些同学可能仍在使用，这些分支还会继续保留： + +- develop分支：这个分支曾用于静态图的开发与测试，目前兼容>=1.7版本的Paddle。如果有特殊需求，要适配旧版本的Paddle，那还可以使用这个分支，但除了修复bug外不再更新代码。 + +PaddleOCR欢迎大家向repo中积极贡献代码，下面给出一些贡献代码的基本流程。 + +### 3.2 PaddleOCR代码提交流程与规范 + +> 如果你熟悉Git使用，可以直接跳转到 [3.2.10 提交代码的一些约定](#提交代码的一些约定) + +#### 3.2.1 创建你的 `远程仓库` + +- 在PaddleOCR的 [GitHub首页](https://github.com/PaddlePaddle/PaddleOCR)，点击左上角 `Fork` 按钮，在你的个人目录下创建 `远程仓库`，比如`https://github.com/{your_name}/PaddleOCR`。 + +![banner](/Users/zhulingfeng01/OCR/PaddleOCR/doc/banner.png) + +- 将 `远程仓库` Clone到本地 + +``` +# 拉取develop分支的代码 +git clone https://github.com/{your_name}/PaddleOCR.git -b dygraph +cd PaddleOCR +``` + +> 多数情况下clone失败是由于网络原因，请稍后重试或配置代理 + +#### 3.2.2 和 `远程仓库` 建立连接 + +首先查看当前 `远程仓库` 的信息。 + +``` +git remote -v +# origin https://github.com/{your_name}/PaddleOCR.git (fetch) +# origin https://github.com/{your_name}/PaddleOCR.git (push) +``` + +只有clone的 `远程仓库` 的信息，也就是自己用户名下的 PaddleOCR，接下来我们创建一个原始 PaddleOCR 仓库的远程主机，命名为 upstream。 + +``` +git remote add upstream https://github.com/PaddlePaddle/PaddleOCR.git +``` + +使用 `git remote -v` 查看当前 `远程仓库` 的信息，输出如下，发现包括了origin和upstream 2个 `远程仓库` 。 + +``` +origin https://github.com/{your_name}/PaddleOCR.git (fetch) +origin https://github.com/{your_name}/PaddleOCR.git (push) +upstream https://github.com/PaddlePaddle/PaddleOCR.git (fetch) +upstream https://github.com/PaddlePaddle/PaddleOCR.git (push) +``` + +这主要是为了后续在提交pull request(PR)时，始终保持本地仓库最新。 + +#### 3.2.3 创建本地分支 + +可以基于当前分支创建新的本地分支，命令如下。 + +``` +git checkout -b new_branch +``` + +也可以基于远程或者上游的分支创建新的分支，命令如下。 + +``` +# 基于用户远程仓库(origin)的develop创建new_branch分支 +git checkout -b new_branch origin/develop +# 基于上游远程仓库(upstream)的develop创建new_branch分支 +# 如果需要从upstream创建新的分支，需要首先使用git fetch upstream获取上游代码 +git checkout -b new_branch upstream/develop +``` + +最终会显示切换到新的分支，输出信息如下 + +``` +Branch new_branch set up to track remote branch develop from upstream. +Switched to a new branch 'new_branch' +``` + +#### 3.2.4 使用pre-commit勾子 + +Paddle 开发人员使用 pre-commit 工具来管理 Git 预提交钩子。 它可以帮助我们格式化源代码（C++，Python），在提交（commit）前自动检查一些基本事宜（如每个文件只有一个 EOL，Git 中不要添加大文件等）。 + +pre-commit测试是 Travis-CI 中单元测试的一部分，不满足钩子的 PR 不能被提交到 PaddleOCR，首先安装并在当前目录运行它： + +``` +pip install pre-commit +pre-commit install +``` + + > 1. Paddle 使用 clang-format 来调整 C/C++ 源代码格式，请确保 `clang-format` 版本在 3.8 以上。 + > + > 2. 通过pip install pre-commit和conda install -c conda-forge pre-commit安装的yapf稍有不同的，PaddleOCR 开发人员使用的是 `pip install pre-commit`。 + +#### 3.2.5 修改与提交代码 + + 假设对PaddleOCR的 `README.md` 做了一些修改，可以通过 `git status` 查看改动的文件，然后使用 `git add` 添加改动文件。 + +``` +git status # 查看改动文件 +git add README.md +pre-commit +``` + +重复上述步骤，直到pre-comit格式检查不报错。如下所示。 + +[![img](https://github.com/PaddlePaddle/PaddleClas/raw/release/2.3/docs/images/quick_start/community/003_precommit_pass.png)](https://github.com/PaddlePaddle/PaddleClas/blob/release/2.3/docs/images/quick_start/community/003_precommit_pass.png) + +使用下面的命令完成提交。 + +``` +git commit -m "your commit info" +``` + +#### 3.2.6 保持本地仓库最新 + +获取 upstream 的最新代码并更新当前分支。这里的upstream来自于2.2节的`和远程仓库建立连接`部分。 + +``` +git fetch upstream +# 如果是希望提交到其他分支，则需要从upstream的其他分支pull代码，这里是develop +git pull upstream develop +``` + +#### 3.2.7 push到远程仓库 + +``` +git push origin new_branch +``` + +#### 3.2.7 提交Pull Request + +点击new pull request，选择本地分支和目标分支，如下图所示。在PR的描述说明中，填写该PR所完成的功能。接下来等待review，如果有需要修改的地方，参照上述步骤更新 origin 中的对应分支即可。 + +![banner](/Users/zhulingfeng01/OCR/PaddleOCR/doc/pr.png) + +#### 3.2.8 签署CLA协议和通过单元测试 + +- 签署CLA 在首次向PaddlePaddle提交Pull Request时，您需要您签署一次CLA(Contributor License Agreement)协议，以保证您的代码可以被合入，具体签署方式如下： + + 1. 请您查看PR中的Check部分，找到license/cla，并点击右侧detail，进入CLA网站 + + 2. 点击CLA网站中的“Sign in with GitHub to agree”,点击完成后将会跳转回您的Pull Request页面 + +#### 3.2.9 删除分支 + +- 删除远程分支 + + 在 PR 被 merge 进主仓库后，我们可以在 PR 的页面删除远程仓库的分支。 + + 也可以使用 `git push origin :分支名` 删除远程分支，如： + + ``` + git push origin :new_branch + ``` + +- 删除本地分支 + + ``` + # 切换到develop分支，否则无法删除当前分支 + git checkout develop + + # 删除new_branch分支 + git branch -D new_branch + ``` + + + +#### 3.2.10 提交代码的一些约定 + +为了使官方维护人员在评审代码时更好地专注于代码本身，请您每次提交代码时，遵守以下约定： + +1）请保证Travis-CI 中单元测试能顺利通过。如果没过，说明提交的代码存在问题，官方维护人员一般不做评审。 + +2）提交Pull Request前： + +- 请注意commit的数量。 + + 原因：如果仅仅修改一个文件但提交了十几个commit，每个commit只做了少量的修改，这会给评审人带来很大困扰。评审人需要逐一查看每个commit才能知道做了哪些修改，且不排除commit之间的修改存在相互覆盖的情况。 + + 建议：每次提交时，保持尽量少的commit，可以通过git commit --amend补充上次的commit。对已经Push到远程仓库的多个commit，可以参考[squash commits after push](https://stackoverflow.com/questions/5667884/how-to-squash-commits-in-git-after-they-have-been-pushed)。 + +- 请注意每个commit的名称：应能反映当前commit的内容，不能太随意。 + + +3）如果解决了某个Issue的问题，请在该Pull Request的第一个评论框中加上：fix #issue_number，这样当该Pull Request被合并后，会自动关闭对应的Issue。关键词包括：close, closes, closed, fix, fixes, fixed, resolve, resolves, resolved，请选择合适的词汇。详细可参考[Closing issues via commit messages](https://help.github.com/articles/closing-issues-via-commit-messages)。 + +此外，在回复评审人意见时，请您遵守以下约定： + +1）官方维护人员的每一个review意见都希望得到回复，这样会更好地提升开源社区的贡献。 + +- 对评审意见同意且按其修改完的，给个简单的Done即可； +- 对评审意见不同意的，请给出您自己的反驳理由。 + +2）如果评审意见比较多: + +- 请给出总体的修改情况。 +- 请采用`start a review`进行回复，而非直接回复的方式。原因是每个回复都会发送一封邮件，会造成邮件灾难。 \ No newline at end of file From c659cd3a8b451b8f576888dff1ea0cc7534c0535 Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Wed, 24 Nov 2021 10:33:23 +0000 Subject: [PATCH 02/31] rename mv3 to r50vd --- test_tipc/configs/det_r50_vd_east_v2.0/train_infer_python.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/test_tipc/configs/det_r50_vd_east_v2.0/train_infer_python.txt b/test_tipc/configs/det_r50_vd_east_v2.0/train_infer_python.txt index e9eaa77952..dfb376237e 100644 --- a/test_tipc/configs/det_r50_vd_east_v2.0/train_infer_python.txt +++ b/test_tipc/configs/det_r50_vd_east_v2.0/train_infer_python.txt @@ -34,7 +34,7 @@ distill_export:null export1:null export2:null ## -train_model:./inference/det_mv3_east/best_accuracy +train_model:./inference/det_r50_vd_east/best_accuracy infer_export:tools/export_model.py -c test_tipc/cconfigs/det_r50_vd_east_v2.0/det_r50_vd_east.yml -o infer_quant:False inference:tools/infer/predict_det.py From 010202ff0a223a057e30485323c960f1a5ffc137 Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Wed, 24 Nov 2021 11:48:00 +0000 Subject: [PATCH 03/31] add pse to tipc --- .../configs/det_mv3_pse_v2.0/det_mv3_pse.yml | 135 ++++++++++++++++++ .../det_mv3_pse_v2.0/train_infer_python.txt | 51 +++++++ .../det_r50_vd_pse_v2.0/det_r50_vd_pse.yml | 134 +++++++++++++++++ .../train_infer_python.txt | 51 +++++++ 4 files changed, 371 insertions(+) create mode 100644 test_tipc/configs/det_mv3_pse_v2.0/det_mv3_pse.yml create mode 100644 test_tipc/configs/det_mv3_pse_v2.0/train_infer_python.txt create mode 100644 test_tipc/configs/det_r50_vd_pse_v2.0/det_r50_vd_pse.yml create mode 100644 test_tipc/configs/det_r50_vd_pse_v2.0/train_infer_python.txt diff --git a/test_tipc/configs/det_mv3_pse_v2.0/det_mv3_pse.yml b/test_tipc/configs/det_mv3_pse_v2.0/det_mv3_pse.yml new file mode 100644 index 0000000000..d37fdcfbb5 --- /dev/null +++ b/test_tipc/configs/det_mv3_pse_v2.0/det_mv3_pse.yml @@ -0,0 +1,135 @@ +Global: + use_gpu: true + epoch_num: 600 + log_smooth_window: 20 + print_batch_step: 10 + save_model_dir: ./output/det_mv3_pse/ + save_epoch_step: 600 + # evaluation is run every 63 iterations + eval_batch_step: [ 0,1000 ] + cal_metric_during_train: False + pretrained_model: ./pretrain_models/MobileNetV3_large_x0_5_pretrained + checkpoints: #./output/det_r50_vd_pse_batch8_ColorJitter/best_accuracy + save_inference_dir: + use_visualdl: False + infer_img: doc/imgs_en/img_10.jpg + save_res_path: ./output/det_pse/predicts_pse.txt + +Architecture: + model_type: det + algorithm: PSE + Transform: null + Backbone: + name: MobileNetV3 + scale: 0.5 + model_name: large + Neck: + name: FPN + out_channels: 96 + Head: + name: PSEHead + hidden_dim: 96 + out_channels: 7 + +Loss: + name: PSELoss + alpha: 0.7 + ohem_ratio: 3 + kernel_sample_mask: pred + reduction: none + +Optimizer: + name: Adam + beta1: 0.9 + beta2: 0.999 + lr: + name: Step + learning_rate: 0.001 + step_size: 200 + gamma: 0.1 + regularizer: + name: 'L2' + factor: 0.0005 + +PostProcess: + name: PSEPostProcess + thresh: 0 + box_thresh: 0.85 + min_area: 16 + box_type: box # 'box' or 'poly' + scale: 1 + +Metric: + name: DetMetric + main_indicator: hmean + +Train: + dataset: + name: SimpleDataSet + data_dir: ./train_data/icdar2015/text_localization/ + label_file_list: + - ./train_data/icdar2015/text_localization/train_icdar2015_label.txt + ratio_list: [ 1.0 ] + transforms: + - DecodeImage: # load image + img_mode: BGR + channel_first: False + - DetLabelEncode: # Class handling label + - ColorJitter: + brightness: 0.12549019607843137 + saturation: 0.5 + - IaaAugment: + augmenter_args: + - { 'type': Resize, 'args': { 'size': [ 0.5, 3 ] } } + - { 'type': Fliplr, 'args': { 'p': 0.5 } } + - { 'type': Affine, 'args': { 'rotate': [ -10, 10 ] } } + - MakePseGt: + kernel_num: 7 + min_shrink_ratio: 0.4 + size: 640 + - RandomCropImgMask: + size: [ 640,640 ] + main_key: gt_text + crop_keys: [ 'image', 'gt_text', 'gt_kernels', 'mask' ] + - NormalizeImage: + scale: 1./255. + mean: [ 0.485, 0.456, 0.406 ] + std: [ 0.229, 0.224, 0.225 ] + order: 'hwc' + - ToCHWImage: + - KeepKeys: + keep_keys: [ 'image', 'gt_text', 'gt_kernels', 'mask' ] # the order of the dataloader list + loader: + shuffle: True + drop_last: False + batch_size_per_card: 16 + num_workers: 8 + +Eval: + dataset: + name: SimpleDataSet + data_dir: ./train_data/icdar2015/text_localization/ + label_file_list: + - ./train_data/icdar2015/text_localization/test_icdar2015_label.txt + ratio_list: [ 1.0 ] + transforms: + - DecodeImage: # load image + img_mode: BGR + channel_first: False + - DetLabelEncode: # Class handling label + - DetResizeForTest: + limit_side_len: 736 + limit_type: min + - NormalizeImage: + scale: 1./255. + mean: [ 0.485, 0.456, 0.406 ] + std: [ 0.229, 0.224, 0.225 ] + order: 'hwc' + - ToCHWImage: + - KeepKeys: + keep_keys: [ 'image', 'shape', 'polys', 'ignore_tags' ] + loader: + shuffle: False + drop_last: False + batch_size_per_card: 1 # must be 1 + num_workers: 8 \ No newline at end of file diff --git a/test_tipc/configs/det_mv3_pse_v2.0/train_infer_python.txt b/test_tipc/configs/det_mv3_pse_v2.0/train_infer_python.txt new file mode 100644 index 0000000000..f9909027f1 --- /dev/null +++ b/test_tipc/configs/det_mv3_pse_v2.0/train_infer_python.txt @@ -0,0 +1,51 @@ +===========================train_params=========================== +model_name:det_mv3_pse_v2.0 +python:python3.7 +gpu_list:0 +Global.use_gpu:True|True +Global.auto_cast:fp32 +Global.epoch_num:lite_train_lite_infer=1|whole_train_whole_infer=500 +Global.save_model_dir:./output/ +Train.loader.batch_size_per_card:lite_train_lite_infer=2|whole_train_whole_infer=4 +Global.pretrained_model:null +train_model_name:latest +train_infer_img_dir:./train_data/icdar2015/text_localization/ch4_test_images/ +null:null +## +trainer:norm_train +norm_train:tools/train.py -c test_tipc/configs/det_mv3_pse_v2.0/det_mv3_pse.yml -o +pact_train:null +fpgm_train:null +distill_train:null +null:null +null:null +## +===========================eval_params=========================== +eval:null +null:null +## +===========================infer_params=========================== +Global.save_inference_dir:./output/ +Global.pretrained_model: +norm_export:tools/export_model.py -c test_tipc/configs/det_mv3_pse_v2.0/det_mv3_pse.yml -o +quant_export:null +fpgm_export:null +distill_export:null +export1:null +export2:null +## +train_model:./inference/det_mv3_pse/best_accuracy +infer_export:tools/export_model.py -c test_tipc/cconfigs/det_mv3_pse_v2.0/det_mv3_pse.yml -o +infer_quant:False +inference:tools/infer/predict_det.py +--use_gpu:True|False +--enable_mkldnn:True|False +--cpu_threads:1|6 +--rec_batch_num:1 +--use_tensorrt:False|True +--precision:fp32|fp16|int8 +--det_model_dir: +--image_dir:./inference/ch_det_data_50/all-sum-510/ +--save_log_path:null +--benchmark:True +--det_algorithm:PSE diff --git a/test_tipc/configs/det_r50_vd_pse_v2.0/det_r50_vd_pse.yml b/test_tipc/configs/det_r50_vd_pse_v2.0/det_r50_vd_pse.yml new file mode 100644 index 0000000000..5ebc425271 --- /dev/null +++ b/test_tipc/configs/det_r50_vd_pse_v2.0/det_r50_vd_pse.yml @@ -0,0 +1,134 @@ +Global: + use_gpu: true + epoch_num: 600 + log_smooth_window: 20 + print_batch_step: 10 + save_model_dir: ./output/det_r50_vd_pse/ + save_epoch_step: 600 + # evaluation is run every 125 iterations + eval_batch_step: [ 0,1000 ] + cal_metric_during_train: False + pretrained_model: + checkpoints: #./output/det_r50_vd_pse_batch8_ColorJitter/best_accuracy + save_inference_dir: + use_visualdl: False + infer_img: doc/imgs_en/img_10.jpg + save_res_path: ./output/det_pse/predicts_pse.txt + +Architecture: + model_type: det + algorithm: PSE + Transform: + Backbone: + name: ResNet + layers: 50 + Neck: + name: FPN + out_channels: 256 + Head: + name: PSEHead + hidden_dim: 256 + out_channels: 7 + +Loss: + name: PSELoss + alpha: 0.7 + ohem_ratio: 3 + kernel_sample_mask: pred + reduction: none + +Optimizer: + name: Adam + beta1: 0.9 + beta2: 0.999 + lr: + name: Step + learning_rate: 0.0001 + step_size: 200 + gamma: 0.1 + regularizer: + name: 'L2' + factor: 0.0005 + +PostProcess: + name: PSEPostProcess + thresh: 0 + box_thresh: 0.85 + min_area: 16 + box_type: box # 'box' or 'poly' + scale: 1 + +Metric: + name: DetMetric + main_indicator: hmean + +Train: + dataset: + name: SimpleDataSet + data_dir: ./train_data/icdar2015/text_localization/ + label_file_list: + - ./train_data/icdar2015/text_localization/train_icdar2015_label.txt + ratio_list: [ 1.0 ] + transforms: + - DecodeImage: # load image + img_mode: BGR + channel_first: False + - DetLabelEncode: # Class handling label + - ColorJitter: + brightness: 0.12549019607843137 + saturation: 0.5 + - IaaAugment: + augmenter_args: + - { 'type': Resize, 'args': { 'size': [ 0.5, 3 ] } } + - { 'type': Fliplr, 'args': { 'p': 0.5 } } + - { 'type': Affine, 'args': { 'rotate': [ -10, 10 ] } } + - MakePseGt: + kernel_num: 7 + min_shrink_ratio: 0.4 + size: 640 + - RandomCropImgMask: + size: [ 640,640 ] + main_key: gt_text + crop_keys: [ 'image', 'gt_text', 'gt_kernels', 'mask' ] + - NormalizeImage: + scale: 1./255. + mean: [ 0.485, 0.456, 0.406 ] + std: [ 0.229, 0.224, 0.225 ] + order: 'hwc' + - ToCHWImage: + - KeepKeys: + keep_keys: [ 'image', 'gt_text', 'gt_kernels', 'mask' ] # the order of the dataloader list + loader: + shuffle: True + drop_last: False + batch_size_per_card: 8 + num_workers: 8 + +Eval: + dataset: + name: SimpleDataSet + data_dir: ./train_data/icdar2015/text_localization/ + label_file_list: + - ./train_data/icdar2015/text_localization/test_icdar2015_label.txt + ratio_list: [ 1.0 ] + transforms: + - DecodeImage: # load image + img_mode: BGR + channel_first: False + - DetLabelEncode: # Class handling label + - DetResizeForTest: + limit_side_len: 736 + limit_type: min + - NormalizeImage: + scale: 1./255. + mean: [ 0.485, 0.456, 0.406 ] + std: [ 0.229, 0.224, 0.225 ] + order: 'hwc' + - ToCHWImage: + - KeepKeys: + keep_keys: [ 'image', 'shape', 'polys', 'ignore_tags' ] + loader: + shuffle: False + drop_last: False + batch_size_per_card: 1 # must be 1 + num_workers: 8 \ No newline at end of file diff --git a/test_tipc/configs/det_r50_vd_pse_v2.0/train_infer_python.txt b/test_tipc/configs/det_r50_vd_pse_v2.0/train_infer_python.txt new file mode 100644 index 0000000000..5ab6d45d7c --- /dev/null +++ b/test_tipc/configs/det_r50_vd_pse_v2.0/train_infer_python.txt @@ -0,0 +1,51 @@ +===========================train_params=========================== +model_name:det_r50_vd_pse_v2.0 +python:python3.7 +gpu_list:0 +Global.use_gpu:True|True +Global.auto_cast:fp32 +Global.epoch_num:lite_train_lite_infer=1|whole_train_whole_infer=500 +Global.save_model_dir:./output/ +Train.loader.batch_size_per_card:lite_train_lite_infer=2|whole_train_whole_infer=4 +Global.pretrained_model:null +train_model_name:latest +train_infer_img_dir:./train_data/icdar2015/text_localization/ch4_test_images/ +null:null +## +trainer:norm_train +norm_train:tools/train.py -c test_tipc/configs/det_r50_vd_pse_v2.0/det_r50_vd_pse.yml -o +pact_train:null +fpgm_train:null +distill_train:null +null:null +null:null +## +===========================eval_params=========================== +eval:null +null:null +## +===========================infer_params=========================== +Global.save_inference_dir:./output/ +Global.pretrained_model: +norm_export:tools/export_model.py -c test_tipc/configs/det_r50_vd_pse_v2.0/det_r50_vd_pse.yml -o +quant_export:null +fpgm_export:null +distill_export:null +export1:null +export2:null +## +train_model:./inference/det_r50_vd_pse/best_accuracy +infer_export:tools/export_model.py -c test_tipc/cconfigs/det_r50_vd_pse_v2.0/det_r50_vd_pse.yml -o +infer_quant:False +inference:tools/infer/predict_det.py +--use_gpu:True|False +--enable_mkldnn:True|False +--cpu_threads:1|6 +--rec_batch_num:1 +--use_tensorrt:False|True +--precision:fp32|fp16|int8 +--det_model_dir: +--image_dir:./inference/ch_det_data_50/all-sum-510/ +--save_log_path:null +--benchmark:True +--det_algorithm:PSE From 35e471860da76aa1bd5507a55d9129251871133c Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Thu, 25 Nov 2021 13:00:37 +0000 Subject: [PATCH 04/31] remove lanms-nova --- doc/doc_ch/detection.md | 4 ++++ ppocr/postprocess/east_postprocess.py | 7 ++++++- requirements.txt | 3 +-- 3 files changed, 11 insertions(+), 3 deletions(-) diff --git a/doc/doc_ch/detection.md b/doc/doc_ch/detection.md index cfc9d52bf2..f76ae7f842 100644 --- a/doc/doc_ch/detection.md +++ b/doc/doc_ch/detection.md @@ -247,3 +247,7 @@ Q1: 训练模型转inference 模型之后预测效果不一致？ **A**：此类问题出现较多，问题多是trained model预测时候的预处理、后处理参数和inference model预测的时候的预处理、后处理参数不一致导致的。以det_mv3_db.yml配置文件训练的模型为例，训练模型、inference模型预测结果不一致问题解决方式如下： - 检查[trained model预处理](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/configs/det/det_mv3_db.yml#L116)，和[inference model的预测预处理](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/tools/infer/predict_det.py#L42)函数是否一致。算法在评估的时候，输入图像大小会影响精度，为了和论文保持一致，训练icdar15配置文件中将图像resize到[736, 1280]，但是在inference model预测的时候只有一套默认参数，会考虑到预测速度问题，默认限制图像最长边为960做resize的。训练模型预处理和inference模型的预处理函数位于[ppocr/data/imaug/operators.py](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/ppocr/data/imaug/operators.py#L147) - 检查[trained model后处理](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/configs/det/det_mv3_db.yml#L51)，和[inference 后处理参数](https://github.com/PaddlePaddle/PaddleOCR/blob/c1ed243fb68d5d466258243092e56cbae32e2c14/tools/infer/utility.py#L50)是否一致。 + +Q1: 训练EAST模型提示找不到lanms库？ + +**A**：执行pip3 install lanms-nova 即可。 diff --git a/ppocr/postprocess/east_postprocess.py b/ppocr/postprocess/east_postprocess.py index ec6bf66385..672d16f9d0 100755 --- a/ppocr/postprocess/east_postprocess.py +++ b/ppocr/postprocess/east_postprocess.py @@ -20,7 +20,6 @@ import numpy as np from .locality_aware_nms import nms_locality import cv2 import paddle -import lanms import os import sys @@ -61,6 +60,12 @@ class EASTPostProcess(object): """ restore text boxes from score map and geo map """ + try: + import lanms + except: + raise Exception( + 'you should install lanms by pip3 install lanms-nova') + score_map = score_map[0] geo_map = np.swapaxes(geo_map, 1, 0) geo_map = np.swapaxes(geo_map, 1, 2) diff --git a/requirements.txt b/requirements.txt index 903b8eda05..0c87c5c950 100644 --- a/requirements.txt +++ b/requirements.txt @@ -12,5 +12,4 @@ cython lxml premailer openpyxl -fasttext==0.9.1 -lanms-nova \ No newline at end of file +fasttext==0.9.1 \ No newline at end of file From cb7e03738b21647578b71c40bdca44ebeedab8f5 Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 26 Nov 2021 10:05:40 +0800 Subject: [PATCH 05/31] add det_mv3_db_v2.0 det_r50_db_v2.0 ch_ppocr_server_v2.0 --- ...rmal_normal_infer_python_linux_gpu_cpu.txt | 19 +++++++ .../det_mv3_db_v2.0/train_infer_python.txt | 51 +++++++++++++++++++ .../det_r50_db_v2.0/train_infer_python.txt | 51 +++++++++++++++++++ test_tipc/prepare.sh | 25 +++++++-- 4 files changed, 141 insertions(+), 5 deletions(-) create mode 100644 test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt create mode 100644 test_tipc/configs/det_mv3_db_v2.0/train_infer_python.txt create mode 100644 test_tipc/configs/det_r50_db_v2.0/train_infer_python.txt diff --git a/test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt new file mode 100644 index 0000000000..5a93571a76 --- /dev/null +++ b/test_tipc/configs/ch_ppocr_server_v2.0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt @@ -0,0 +1,19 @@ +===========================ch_ppocr_mobile_v2.0=========================== +model_name:ch_ppocr_server_v2.0 +python:python3.7 +infer_model:./inference/ch_ppocr_server_v2.0_det_infer/ +infer_export:null +infer_quant:True +inference:tools/infer/predict_system.py +--use_gpu:False +--enable_mkldnn:False +--cpu_threads:1|6 +--rec_batch_num:1 +--use_tensorrt:False +--precision:int8 +--det_model_dir: +--image_dir:./inference/ch_det_data_50/all-sum-510/ +--rec_model_dir:./inference/ch_ppocr_server_v2.0_rec_infer/ +--benchmark:True +null:null +null:null diff --git a/test_tipc/configs/det_mv3_db_v2.0/train_infer_python.txt b/test_tipc/configs/det_mv3_db_v2.0/train_infer_python.txt new file mode 100644 index 0000000000..de9df22772 --- /dev/null +++ b/test_tipc/configs/det_mv3_db_v2.0/train_infer_python.txt @@ -0,0 +1,51 @@ +===========================train_params=========================== +model_name:det_mv3_db_v2.0 +python:python3.7 +gpu_list:0|0,1 +Global.use_gpu:True|True +Global.auto_cast:null +Global.epoch_num:lite_train_lite_infer=1|whole_train_whole_infer=300 +Global.save_model_dir:./output/ +Train.loader.batch_size_per_card:lite_train_lite_infer=2|whole_train_whole_infer=4 +Global.pretrained_model:null +train_model_name:latest +train_infer_img_dir:./train_data/icdar2015/text_localization/ch4_test_images/ +null:null +## +trainer:norm_train +norm_train:tools/train.py -c configs/det/det_mv3_db.yml -o Global.pretrained_model=./pretrain_models/MobileNetV3_large_x0_5_pretrained +pact_train:null +fpgm_train:null +distill_train:null +null:null +null:null +## +===========================eval_params=========================== +eval:null +null:null +## +===========================infer_params=========================== +Global.save_inference_dir:./output/ +Global.pretrained_model: +norm_export:tools/export_model.py -c configs/det/det_mv3_db.yml -o +quant_export:null +fpgm_export:null +distill_export:null +export1:null +export2:null +inference_dir:null +train_model:./inference/det_mv3_db_v2.0_train/best_accuracy +infer_export:tools/export_model.py -c configs/det/det_mv3_db.yml -o +infer_quant:False +inference:tools/infer/predict_det.py +--use_gpu:True|False +--enable_mkldnn:True|False +--cpu_threads:1|6 +--rec_batch_num:1 +--use_tensorrt:False|True +--precision:fp32|fp16|int8 +--det_model_dir: +--image_dir:./inference/ch_det_data_50/all-sum-510/ +null:null +--benchmark:True +null:null \ No newline at end of file diff --git a/test_tipc/configs/det_r50_db_v2.0/train_infer_python.txt b/test_tipc/configs/det_r50_db_v2.0/train_infer_python.txt new file mode 100644 index 0000000000..72b5ede15d --- /dev/null +++ b/test_tipc/configs/det_r50_db_v2.0/train_infer_python.txt @@ -0,0 +1,51 @@ +===========================train_params=========================== +model_name:det_r50_db_v2.0 +python:python3.7 +gpu_list:0|0,1 +Global.use_gpu:True|True +Global.auto_cast:null +Global.epoch_num:lite_train_lite_infer=2|whole_train_whole_infer=300 +Global.save_model_dir:./output/ +Train.loader.batch_size_per_card:lite_train_lite_infer=2|whole_train_lite_infer=4 +Global.pretrained_model:null +train_model_name:latest +train_infer_img_dir:./train_data/icdar2015/text_localization/ch4_test_images/ +null:null +## +trainer:norm_train +norm_train:tools/train.py -c configs/det/det_r50_vd_db.yml -o +quant_export:null +fpgm_export:null +distill_train:null +null:null +null:null +## +===========================eval_params=========================== +eval:tools/eval.py -c configs/det/det_r50_vd_db.yml -o +null:null +## +===========================infer_params=========================== +Global.save_inference_dir:./output/ +Global.pretrained_model: +norm_export:tools/export_model.py -c configs/det/det_r50_vd_db.yml -o +quant_export:null +fpgm_export:null +distill_export:null +export1:null +export2:null +## +train_model:./inference/ch_ppocr_server_v2.0_det_train/best_accuracy +infer_export:tools/export_model.py -c configs/det/det_r50_vd_db.yml -o +infer_quant:False +inference:tools/infer/predict_det.py +--use_gpu:True|False +--enable_mkldnn:True|False +--cpu_threads:1|6 +--rec_batch_num:1 +--use_tensorrt:False|True +--precision:fp32|fp16|int8 +--det_model_dir: +--image_dir:./inference/ch_det_data_50/all-sum-510/ +--save_log_path:null +--benchmark:True +null:null \ No newline at end of file diff --git a/test_tipc/prepare.sh b/test_tipc/prepare.sh index 458b22da30..49b0fc7c78 100644 --- a/test_tipc/prepare.sh +++ b/test_tipc/prepare.sh @@ -52,6 +52,13 @@ if [ ${MODE} = "lite_train_lite_infer" ];then wget -nc -P ./train_data/ wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/total_text_lite.tar --no-check-certificate cd ./train_data && tar xf total_text_lite.tar && ln -s total_text && cd ../ fi + if [ ${model_name} == "det_mv3_db_v2.0" ]; then + wget -nc -P https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_mv3_db_v2.0_train.tar --no-check-certificate + cd ./inference/ && tar det_mv3_db_v2.0_train.tar && cd ../ + fi + if [ ${model_name} == "det_r50_db_v2.0" ]; then + wget -nc -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/ResNet50_vd_ssld_pretrained.pdparams --no-check-certificate + fi elif [ ${MODE} = "whole_train_whole_infer" ];then wget -nc -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/MobileNetV3_large_x0_5_pretrained.pdparams --no-check-certificate @@ -101,12 +108,12 @@ elif [ ${MODE} = "whole_infer" ];then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_train.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/ch_det_data_50.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_server_v2.0_det_train.tar && tar xf ch_det_data_50.tar && cd ../ - elif [ ${model_name} = "ocr_system_mobile" ]; then + elif [ ${model_name} = "ch_ppocr_mobile_v2.0" ]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/ch_det_data_50.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_mobile_v2.0_det_infer.tar && tar xf ch_ppocr_mobile_v2.0_rec_infer.tar && tar xf ch_det_data_50.tar && cd ../ - elif [ ${model_name} = "ocr_system_server" ]; then + elif [ ${model_name} = "ch_ppocr_server_v2.0" ]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/ch_det_data_50.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar --no-check-certificate @@ -124,13 +131,13 @@ elif [ ${MODE} = "whole_infer" ];then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar --no-check-certificate cd ./inference && tar xf ${eval_model_name}.tar && tar xf rec_inference.tar && cd ../ fi - elif [ ${model_name} = "ch_PPOCRv2_det" ]; then + if [ ${model_name} = "ch_PPOCRv2_det" ]; then eval_model_name="ch_PP-OCRv2_det_infer" wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/ch_det_data_50.tar --no-check-certificate wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_infer.tar --no-check-certificate cd ./inference && tar xf ${eval_model_name}.tar && tar xf ch_det_data_50.tar && cd ../ fi - elif [ ${model_name} = "ch_PPOCRv2_det" ]; then + if [ ${model_name} = "ch_PPOCRv2_det" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/ch_det_data_50.tar --no-check-certificate wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/pgnet/e2e_server_pgnetA_infer.tar --no-check-certificate cd ./inference && tar xf e2e_server_pgnetA_infer.tar && tar xf ch_det_data_50.tar && cd ../ @@ -141,7 +148,15 @@ elif [ ${MODE} = "whole_infer" ];then fi if [ ${model_name} == "det_r50_vd_sast_icdar15_v2.0" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_sast_icdar15_v2.0_train.tar --no-check-certificate - cd ./inference/ && tar det_r50_vd_sast_icdar15_v2.0_train.tar && cd ../ + cd ./inference/ && tar xf det_r50_vd_sast_icdar15_v2.0_train.tar && cd ../ + fi + if [ ${model_name} == "det_mv3_db_v2.0" ]; then + wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_mv3_db_v2.0_train.tar --no-check-certificate + cd ./inference/ && tar xf det_mv3_db_v2.0_train.tar && cd ../ + fi + if [ ${model_name} == "det_r50_db_v2.0" ]; then + wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_db_v2.0_train.tar --no-check-certificate + cd ./inference/ && tar xf det_r50_vd_db_v2.0_train.tar && cd ../ fi if [ ${MODE} = "klquant_whole_infer" ]; then From de809245456b23b5fa3240121bd62964883d2be3 Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 26 Nov 2021 10:11:00 +0800 Subject: [PATCH 06/31] fix prepare.sh --- test_tipc/prepare.sh | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/test_tipc/prepare.sh b/test_tipc/prepare.sh index 49b0fc7c78..34239f1964 100644 --- a/test_tipc/prepare.sh +++ b/test_tipc/prepare.sh @@ -53,11 +53,13 @@ if [ ${MODE} = "lite_train_lite_infer" ];then cd ./train_data && tar xf total_text_lite.tar && ln -s total_text && cd ../ fi if [ ${model_name} == "det_mv3_db_v2.0" ]; then - wget -nc -P https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_mv3_db_v2.0_train.tar --no-check-certificate - cd ./inference/ && tar det_mv3_db_v2.0_train.tar && cd ../ + wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_mv3_db_v2.0_train.tar --no-check-certificate + cd ./inference/ && tar xf det_mv3_db_v2.0_train.tar && cd ../ fi if [ ${model_name} == "det_r50_db_v2.0" ]; then wget -nc -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/ResNet50_vd_ssld_pretrained.pdparams --no-check-certificate + wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_db_v2.0_train.tar --no-check-certificate + cd ./inference/ && tar xf det_r50_vd_db_v2.0_train.tar && cd ../ fi elif [ ${MODE} = "whole_train_whole_infer" ];then From 4467ef6b82b39be01199b734384e1c6cf1d1906c Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 26 Nov 2021 10:58:18 +0800 Subject: [PATCH 07/31] fix prepare.sh --- test_tipc/prepare.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/test_tipc/prepare.sh b/test_tipc/prepare.sh index 34239f1964..44d10e2165 100644 --- a/test_tipc/prepare.sh +++ b/test_tipc/prepare.sh @@ -160,7 +160,7 @@ elif [ ${MODE} = "whole_infer" ];then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_db_v2.0_train.tar --no-check-certificate cd ./inference/ && tar xf det_r50_vd_db_v2.0_train.tar && cd ../ fi - +fi if [ ${MODE} = "klquant_whole_infer" ]; then if [ ${model_name} = "ch_ppocr_mobile_v2.0_det" ]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar --no-check-certificate From b491c40e9094add38f114db1000fbc7f5930200f Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 26 Nov 2021 11:15:21 +0800 Subject: [PATCH 08/31] fix bug --- .../configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt index d9766d150d..a6779965d5 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./train_data/icdar2015/text_localization/ch4_test_images/ null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/ppocr_det_mobile/det_mv3_db.yml -o Global.pretrained_model=./pretrain_models/MobileNetV3_large_x0_5_pretrained +norm_train:tools/train.py -c test_tipc/configs/ch_ppocr_mobile_v2.0/det_mv3_db.yml -o Global.pretrained_model=./pretrain_models/MobileNetV3_large_x0_5_pretrained pact_train:null fpgm_train:null distill_train:null @@ -27,7 +27,7 @@ null:null ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.pretrained_model: -norm_export:tools/export_model.py -c test_tipc/configs/ppocr_det_mobile/det_mv3_db.yml -o +norm_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_mobile_v2.0/det_mv3_db.yml -o quant_export:null fpgm_export:null distill_export:null From 255115deca2d3e2b7347adb2b098324a71be7fb1 Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 26 Nov 2021 11:18:26 +0800 Subject: [PATCH 09/31] fix bug --- .../configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt index a6779965d5..4609330297 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt @@ -13,7 +13,7 @@ train_infer_img_dir:./train_data/icdar2015/text_localization/ch4_test_images/ null:null ## trainer:norm_train -norm_train:tools/train.py -c test_tipc/configs/ch_ppocr_mobile_v2.0/det_mv3_db.yml -o Global.pretrained_model=./pretrain_models/MobileNetV3_large_x0_5_pretrained +norm_train:tools/train.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o Global.pretrained_model=./pretrain_models/MobileNetV3_large_x0_5_pretrained pact_train:null fpgm_train:null distill_train:null @@ -27,7 +27,7 @@ null:null ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.pretrained_model: -norm_export:tools/export_model.py -c test_tipc/configs/ch_ppocr_mobile_v2.0/det_mv3_db.yml -o +norm_export:tools/export_model.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o quant_export:null fpgm_export:null distill_export:null From 36d7c3518ad56900f698c352ef6827761af372fd Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 26 Nov 2021 11:30:05 +0800 Subject: [PATCH 10/31] fix prepare.sh --- .../ch_ppocr_mobile_V2.0_det_FPGM/train_infer_python.txt | 6 +++--- .../configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt | 2 +- .../ch_ppocr_mobile_v2.0_det_PACT/train_infer_python.txt | 6 +++--- test_tipc/test_train_inference_python.sh | 2 +- 4 files changed, 8 insertions(+), 8 deletions(-) diff --git a/test_tipc/configs/ch_ppocr_mobile_V2.0_det_FPGM/train_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_V2.0_det_FPGM/train_infer_python.txt index 77889729e6..92ac3e9d37 100644 --- a/test_tipc/configs/ch_ppocr_mobile_V2.0_det_FPGM/train_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_mobile_V2.0_det_FPGM/train_infer_python.txt @@ -4,7 +4,7 @@ python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True Global.auto_cast:null -Global.epoch_num:lite_train_lite_infer=1|whole_train_whole_infer=300 +Global.epoch_num:lite_train_lite_infer=5|whole_train_whole_infer=300 Global.save_model_dir:./output/ Train.loader.batch_size_per_card:lite_train_lite_infer=2|whole_train_whole_infer=4 Global.pretrained_model:null @@ -15,7 +15,7 @@ null:null trainer:fpgm_train norm_train:null pact_train:null -fpgm_train:deploy/slim/prune/sensitivity_anal.py -c test_tipc/configs/ppocr_det_mobile/det_mv3_db.yml -o Global.pretrained_model=./pretrain_models/det_mv3_db_v2.0_train/best_accuracy +fpgm_train:deploy/slim/prune/sensitivity_anal.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o Global.pretrained_model=./pretrain_models/det_mv3_db_v2.0_train/best_accuracy distill_train:null null:null null:null @@ -29,7 +29,7 @@ Global.save_inference_dir:./output/ Global.pretrained_model: norm_export:null quant_export:null -fpgm_export:deploy/slim/prune/export_prune_model.py -c test_tipc/configs/ppocr_det_mobile/det_mv3_db.yml -o +fpgm_export:deploy/slim/prune/export_prune_model.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o distill_export:null export1:null export2:null diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt index 4609330297..977312f2a4 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_infer_python.txt @@ -4,7 +4,7 @@ python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True Global.auto_cast:null -Global.epoch_num:lite_train_lite_infer=1|whole_train_whole_infer=300 +Global.epoch_num:lite_train_lite_infer=5|whole_train_whole_infer=300 Global.save_model_dir:./output/ Train.loader.batch_size_per_card:lite_train_lite_infer=2|whole_train_whole_infer=4 Global.pretrained_model:null diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_PACT/train_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_v2.0_det_PACT/train_infer_python.txt index 7328be25ff..8a6c656858 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_PACT/train_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2.0_det_PACT/train_infer_python.txt @@ -4,7 +4,7 @@ python:python3.7 gpu_list:0|0,1 Global.use_gpu:True|True Global.auto_cast:null -Global.epoch_num:lite_train_lite_infer=1|whole_train_whole_infer=300 +Global.epoch_num:lite_train_lite_infer=5|whole_train_whole_infer=300 Global.save_model_dir:./output/ Train.loader.batch_size_per_card:lite_train_lite_infer=2|whole_train_whole_infer=4 Global.pretrained_model:null @@ -14,7 +14,7 @@ null:null ## trainer:pact_train norm_train:null -pact_train:deploy/slim/quantization/quant.py -c test_tipc/configs/ppocr_det_mobile/det_mv3_db.yml -o +pact_train:deploy/slim/quantization/quant.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o fpgm_train:null distill_train:null null:null @@ -28,7 +28,7 @@ null:null Global.save_inference_dir:./output/ Global.pretrained_model: norm_export:null -quant_export:deploy/slim/quantization/export_model.py -c test_tipc/configs/ppocr_det_mobile/det_mv3_db.yml -o +quant_export:deploy/slim/quantization/export_model.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o fpgm_export:null distill_export:null export1:null diff --git a/test_tipc/test_train_inference_python.sh b/test_tipc/test_train_inference_python.sh index 6773c44e9e..21877364ca 100644 --- a/test_tipc/test_train_inference_python.sh +++ b/test_tipc/test_train_inference_python.sh @@ -336,7 +336,7 @@ else set_eval_pretrain=$(func_set_params "${pretrain_model_key}" "${save_log}/${train_model_name}") # save norm trained models to set pretrain for pact training and fpgm training - if [ ${trainer} = ${trainer_norm} ] && [ ${nodes} -le 1]; then + if [ ${trainer} = ${trainer_norm} ] && [ ${nodes} -le 1 ]; then load_norm_train_model=${set_eval_pretrain} fi # run eval From d4b2e486cdea2fff151a28ac7abc18a4eb58ebbf Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 26 Nov 2021 11:45:43 +0800 Subject: [PATCH 11/31] fix yml --- ...cpu_normal_normal_infer_python_mac_cpu.txt | 14 ++-- ...al_normal_infer_python_windows_cpu_gpu.txt | 75 ++----------------- 2 files changed, 15 insertions(+), 74 deletions(-) diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_mac_cpu_normal_normal_infer_python_mac_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_mac_cpu_normal_normal_infer_python_mac_cpu.txt index 4001ca1828..74115910a2 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_mac_cpu_normal_normal_infer_python_mac_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_mac_cpu_normal_normal_infer_python_mac_cpu.txt @@ -12,10 +12,10 @@ train_model_name:latest train_infer_img_dir:./train_data/icdar2015/text_localization/ch4_test_images/ null:null ## -trainer:norm_train|pact_train|fpgm_train -norm_train:tools/train.py -c test_tipc/configs/det_mv3_db.yml -o Global.pretrained_model=./pretrain_models/MobileNetV3_large_x0_5_pretrained -pact_train:deploy/slim/quantization/quant.py -c test_tipc/configs/det_mv3_db.yml -o -fpgm_train:deploy/slim/prune/sensitivity_anal.py -c test_tipc/configs/det_mv3_db.yml -o Global.pretrained_model=./pretrain_models/det_mv3_db_v2.0_train/best_accuracy +trainer:norm_train +norm_train:tools/train.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o Global.pretrained_model=./pretrain_models/MobileNetV3_large_x0_5_pretrained +pact_train:null +fpgm_train:null distill_train:null null:null null:null @@ -27,9 +27,9 @@ null:null ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.pretrained_model: -norm_export:tools/export_model.py -c test_tipc/configs/det_mv3_db.yml -o -quant_export:deploy/slim/quantization/export_model.py -c test_tipc/configs/det_mv3_db.yml -o -fpgm_export:deploy/slim/prune/export_prune_model.py -c test_tipc/configs/det_mv3_db.yml -o +norm_export:tools/export_model.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o +quant_export:null +fpgm_export:null distill_export:null export1:null export2:null diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_windows_gpu_normal_normal_infer_python_windows_cpu_gpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_windows_gpu_normal_normal_infer_python_windows_cpu_gpu.txt index 0f4faee4b3..58f6ccad80 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_windows_gpu_normal_normal_infer_python_windows_cpu_gpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_windows_gpu_normal_normal_infer_python_windows_cpu_gpu.txt @@ -12,10 +12,10 @@ train_model_name:latest train_infer_img_dir:./train_data/icdar2015/text_localization/ch4_test_images/ null:null ## -trainer:norm_train|pact_train|fpgm_train -norm_train:tools/train.py -c test_tipc/configs/det_mv3_db.yml -o Global.pretrained_model=./pretrain_models/MobileNetV3_large_x0_5_pretrained -pact_train:deploy/slim/quantization/quant.py -c test_tipc/configs/det_mv3_db.yml -o -fpgm_train:deploy/slim/prune/sensitivity_anal.py -c test_tipc/configs/det_mv3_db.yml -o Global.pretrained_model=./pretrain_models/det_mv3_db_v2.0_train/best_accuracy +trainer:norm_train +norm_train:tools/train.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o Global.pretrained_model=./pretrain_models/MobileNetV3_large_x0_5_pretrained +pact_train:null +fpgm_train:null distill_train:null null:null null:null @@ -27,9 +27,9 @@ null:null ===========================infer_params=========================== Global.save_inference_dir:./output/ Global.pretrained_model: -norm_export:tools/export_model.py -c test_tipc/configs/det_mv3_db.yml -o -quant_export:deploy/slim/quantization/export_model.py -c test_tipc/configs/det_mv3_db.yml -o -fpgm_export:deploy/slim/prune/export_prune_model.py -c test_tipc/configs/det_mv3_db.yml -o +norm_export:tools/export_model.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o +quant_export:null +fpgm_export:null distill_export:null export1:null export2:null @@ -49,63 +49,4 @@ inference:tools/infer/predict_det.py null:null --benchmark:True null:null -===========================cpp_infer_params=========================== -use_opencv:True -infer_model:./inference/ch_ppocr_mobile_v2.0_det_infer/ -infer_quant:False -inference:./deploy/cpp_infer/build/ppocr det ---use_gpu:True|False ---enable_mkldnn:True|False ---cpu_threads:1|6 ---rec_batch_num:1 ---use_tensorrt:False|True ---precision:fp32|fp16 ---det_model_dir: ---image_dir:./inference/ch_det_data_50/all-sum-510/ -null:null ---benchmark:True -===========================serving_params=========================== -model_name:ocr_det -python:python3.7 -trans_model:-m paddle_serving_client.convert ---dirname:./inference/ch_ppocr_mobile_v2.0_det_infer/ ---model_filename:inference.pdmodel ---params_filename:inference.pdiparams ---serving_server:./deploy/pdserving/ppocr_det_mobile_2.0_serving/ ---serving_client:./deploy/pdserving/ppocr_det_mobile_2.0_client/ -serving_dir:./deploy/pdserving -web_service:web_service_det.py --config=config.yml --opt op.det.concurrency=1 -op.det.local_service_conf.devices:null|0 -op.det.local_service_conf.use_mkldnn:True|False -op.det.local_service_conf.thread_num:1|6 -op.det.local_service_conf.use_trt:False|True -op.det.local_service_conf.precision:fp32|fp16|int8 -pipline:pipeline_http_client.py|pipeline_rpc_client.py ---image_dir=../../doc/imgs -===========================kl_quant_params=========================== -infer_model:./inference/ch_ppocr_mobile_v2.0_det_infer/ -infer_export:tools/export_model.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o -infer_quant:True -inference:tools/infer/predict_det.py ---use_gpu:True|False ---enable_mkldnn:True|False ---cpu_threads:1|6 ---rec_batch_num:1 ---use_tensorrt:False|True ---precision:int8 ---det_model_dir: ---image_dir:./inference/ch_det_data_50/all-sum-510/ -null:null ---benchmark:True -null:null -null:null -===========================lite_params=========================== -inference:./ocr_db_crnn det -infer_model:./models/ch_ppocr_mobile_v2.0_det_opt.nb|./models/ch_ppocr_mobile_v2.0_det_slim_opt.nb ---cpu_threads:1|4 ---batch_size:1 ---power_mode:LITE_POWER_HIGH|LITE_POWER_LOW ---image_dir:./test_data/icdar2015_lite/text_localization/ch4_test_images/|./test_data/icdar2015_lite/text_localization/ch4_test_images/img_233.jpg ---config_dir:./config.txt ---rec_dict_dir:./ppocr_keys_v1.txt ---benchmark:True + From 75772e5dc428fcf8f0b7763e454559a176dbac7a Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 26 Nov 2021 11:47:34 +0800 Subject: [PATCH 12/31] fix epoch --- .../train_mac_cpu_normal_normal_infer_python_mac_cpu.txt | 2 +- ...n_windows_gpu_normal_normal_infer_python_windows_cpu_gpu.txt | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_mac_cpu_normal_normal_infer_python_mac_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_mac_cpu_normal_normal_infer_python_mac_cpu.txt index 74115910a2..014dad5fc9 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_mac_cpu_normal_normal_infer_python_mac_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_mac_cpu_normal_normal_infer_python_mac_cpu.txt @@ -4,7 +4,7 @@ python:python gpu_list:-1 Global.use_gpu:False Global.auto_cast:null -Global.epoch_num:lite_train_lite_infer=1|whole_train_whole_infer=300 +Global.epoch_num:lite_train_lite_infer=5|whole_train_whole_infer=300 Global.save_model_dir:./output/ Train.loader.batch_size_per_card:lite_train_lite_infer=2|whole_train_whole_infer=4 Global.pretrained_model:null diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_windows_gpu_normal_normal_infer_python_windows_cpu_gpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_windows_gpu_normal_normal_infer_python_windows_cpu_gpu.txt index 58f6ccad80..6a63b39d97 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_windows_gpu_normal_normal_infer_python_windows_cpu_gpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2.0_det/train_windows_gpu_normal_normal_infer_python_windows_cpu_gpu.txt @@ -4,7 +4,7 @@ python:python gpu_list:0 Global.use_gpu:True Global.auto_cast:fp32|amp -Global.epoch_num:lite_train_lite_infer=1|whole_train_whole_infer=300 +Global.epoch_num:lite_train_lite_infer=5|whole_train_whole_infer=300 Global.save_model_dir:./output/ Train.loader.batch_size_per_card:lite_train_lite_infer=2|whole_train_whole_infer=4 Global.pretrained_model:null From 779da2596e5d16dc37dde58f8bf0342a088053b1 Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 26 Nov 2021 12:53:17 +0800 Subject: [PATCH 13/31] fix kl --- ...model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt | 2 +- test_tipc/prepare.sh | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt index bd58e96403..ba84cecdd0 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt @@ -1,5 +1,5 @@ ===========================kl_quant_params=========================== -model_name:PPOCRv2_ocr_det +model_name:ch_ppocr_mobile_v2.0_det_KL python:python3.7 infer_model:./inference/ch_ppocr_mobile_v2.0_det_infer/ infer_export:deploy/slim/quantization/quant_kl.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o diff --git a/test_tipc/prepare.sh b/test_tipc/prepare.sh index 44d10e2165..e021e7af35 100644 --- a/test_tipc/prepare.sh +++ b/test_tipc/prepare.sh @@ -162,7 +162,7 @@ elif [ ${MODE} = "whole_infer" ];then fi fi if [ ${MODE} = "klquant_whole_infer" ]; then - if [ ${model_name} = "ch_ppocr_mobile_v2.0_det" ]; then + if [ ${model_name} = "ch_ppocr_mobile_v2.0_det_KL" ]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/ch_det_data_50.tar --no-check-certificate cd ./inference && tar xf ch_ppocr_mobile_v2.0_det_infer.tar && tar xf ch_det_data_50.tar && cd ../ From ba2c0d395dce759c5420472babfa55bb2c542180 Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 26 Nov 2021 13:11:27 +0800 Subject: [PATCH 14/31] fix kl bug --- ...model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt | 2 ++ ...model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt | 2 ++ ...model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt | 2 ++ 3 files changed, 6 insertions(+) diff --git a/test_tipc/configs/ch_PP-OCRv2_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_PP-OCRv2_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt index 0ff24cbccf..e9524b65a2 100644 --- a/test_tipc/configs/ch_PP-OCRv2_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_PP-OCRv2_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt @@ -1,6 +1,8 @@ ===========================kl_quant_params=========================== model_name:PPOCRv2_ocr_det_kl python:python3.7 +Global.pretrained_model:null +Global.save_inference_dir:null infer_model:./inference/ch_PP-OCRv2_det_infer/ infer_export:deploy/slim/quantization/quant_kl.py -c configs/det/ch_PP-OCRv2/ch_PP-OCRv2_det_cml.yml -o infer_quant:True diff --git a/test_tipc/configs/ch_PP-OCRv2_rec_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_PP-OCRv2_rec_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt index 8826bb4f07..aff419e8a2 100644 --- a/test_tipc/configs/ch_PP-OCRv2_rec_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_PP-OCRv2_rec_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt @@ -1,6 +1,8 @@ ===========================kl_quant_params=========================== model_name:PPOCRv2_ocr_rec_kl python:python3.7 +Global.pretrained_model:null +Global.save_inference_dir:null infer_model:./inference/ch_PP-OCRv2_rec_infer/ infer_export:deploy/slim/quantization/quant_kl.py -c test_tipc/configs/ch_PP-OCRv2_rec/ch_PP-OCRv2_rec_distillation.yml -o infer_quant:True diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt index ba84cecdd0..f35c00ea2d 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt @@ -1,6 +1,8 @@ ===========================kl_quant_params=========================== model_name:ch_ppocr_mobile_v2.0_det_KL python:python3.7 +Global.pretrained_model:null +Global.save_inference_dir:null infer_model:./inference/ch_ppocr_mobile_v2.0_det_infer/ infer_export:deploy/slim/quantization/quant_kl.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o infer_quant:True From 725da5ab0d59a9d39afbe71ea933bad9d923ed96 Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 26 Nov 2021 13:12:21 +0800 Subject: [PATCH 15/31] fix kl bug --- test_tipc/test_train_inference_python.sh | 52 ++++++++++++------------ 1 file changed, 27 insertions(+), 25 deletions(-) diff --git a/test_tipc/test_train_inference_python.sh b/test_tipc/test_train_inference_python.sh index 21877364ca..210e1c2c0c 100644 --- a/test_tipc/test_train_inference_python.sh +++ b/test_tipc/test_train_inference_python.sh @@ -90,36 +90,38 @@ infer_value1=$(func_parser_value "${lines[50]}") # parser klquant_infer if [ ${MODE} = "klquant_whole_infer" ]; then - dataline=$(awk 'NR==1 NR==17{print}' $FILENAME) + dataline=$(awk 'NR==1, NR==17{print}' $FILENAME) lines=(${dataline}) model_name=$(func_parser_value "${lines[1]}") python=$(func_parser_value "${lines[2]}") + export_weight=$(func_parser_key "${lines[3]}") + save_infer_key=$(func_parser_key "${lines[4]}") # parser inference model - infer_model_dir_list=$(func_parser_value "${lines[3]}") - infer_export_list=$(func_parser_value "${lines[4]}") - infer_is_quant=$(func_parser_value "${lines[5]}") + infer_model_dir_list=$(func_parser_value "${lines[5]}") + infer_export_list=$(func_parser_value "${lines[6]}") + infer_is_quant=$(func_parser_value "${lines[7]}") # parser inference - inference_py=$(func_parser_value "${lines[6]}") - use_gpu_key=$(func_parser_key "${lines[7]}") - use_gpu_list=$(func_parser_value "${lines[7]}") - use_mkldnn_key=$(func_parser_key "${lines[8]}") - use_mkldnn_list=$(func_parser_value "${lines[8]}") - cpu_threads_key=$(func_parser_key "${lines[9]}") - cpu_threads_list=$(func_parser_value "${lines[9]}") - batch_size_key=$(func_parser_key "${lines[10]}") - batch_size_list=$(func_parser_value "${lines[10]}") - use_trt_key=$(func_parser_key "${lines[11]}") - use_trt_list=$(func_parser_value "${lines[11]}") - precision_key=$(func_parser_key "${lines[12]}") - precision_list=$(func_parser_value "${lines[12]}") - infer_model_key=$(func_parser_key "${lines[13]}") - image_dir_key=$(func_parser_key "${lines[14]}") - infer_img_dir=$(func_parser_value "${lines[14]}") - save_log_key=$(func_parser_key "${lines[15]}") - benchmark_key=$(func_parser_key "${lines[16]}") - benchmark_value=$(func_parser_value "${lines[16]}") - infer_key1=$(func_parser_key "${lines[17]}") - infer_value1=$(func_parser_value "${lines[17]}") + inference_py=$(func_parser_value "${lines[8]}") + use_gpu_key=$(func_parser_key "${lines[9]}") + use_gpu_list=$(func_parser_value "${lines[10]}") + use_mkldnn_key=$(func_parser_key "${lines[10]}") + use_mkldnn_list=$(func_parser_value "${lines[10]}") + cpu_threads_key=$(func_parser_key "${lines[11]}") + cpu_threads_list=$(func_parser_value "${lines[11]}") + batch_size_key=$(func_parser_key "${lines[12]}") + batch_size_list=$(func_parser_value "${lines[12]}") + use_trt_key=$(func_parser_key "${lines[13]}") + use_trt_list=$(func_parser_value "${lines[13]}") + precision_key=$(func_parser_key "${lines[14]}") + precision_list=$(func_parser_value "${lines[14]}") + infer_model_key=$(func_parser_key "${lines[15]}") + image_dir_key=$(func_parser_key "${lines[16]}") + infer_img_dir=$(func_parser_value "${lines[16]}") + save_log_key=$(func_parser_key "${lines[17]}") + benchmark_key=$(func_parser_key "${lines[18]}") + benchmark_value=$(func_parser_value "${lines[18]}") + infer_key1=$(func_parser_key "${lines[19]}") + infer_value1=$(func_parser_value "${lines[19]}") fi LOG_PATH="./test_tipc/output" From c244f750b7d6d0a855b082259da9de7d723a75b9 Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 26 Nov 2021 13:22:28 +0800 Subject: [PATCH 16/31] fix bug --- test_tipc/test_train_inference_python.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/test_tipc/test_train_inference_python.sh b/test_tipc/test_train_inference_python.sh index 210e1c2c0c..2b1ca65a36 100644 --- a/test_tipc/test_train_inference_python.sh +++ b/test_tipc/test_train_inference_python.sh @@ -237,7 +237,7 @@ if [ ${MODE} = "whole_infer" ] || [ ${MODE} = "klquant_whole_infer" ]; then fi #run inference is_quant=${infer_quant_flag[Count]} - if [ ${MODE} = "klquant_infer" ]; then + if [ ${MODE} = "klquant_whole_infer" ]; then is_quant="True" fi func_inference "${python}" "${inference_py}" "${save_infer_dir}" "${LOG_PATH}" "${infer_img_dir}" ${is_quant} From 2bd6b5ed5d32ff2db2ad87cc78d6875befb818f3 Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 26 Nov 2021 13:36:58 +0800 Subject: [PATCH 17/31] fix bug --- test_tipc/test_train_inference_python.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/test_tipc/test_train_inference_python.sh b/test_tipc/test_train_inference_python.sh index 2b1ca65a36..7d70de1e38 100644 --- a/test_tipc/test_train_inference_python.sh +++ b/test_tipc/test_train_inference_python.sh @@ -103,7 +103,7 @@ if [ ${MODE} = "klquant_whole_infer" ]; then # parser inference inference_py=$(func_parser_value "${lines[8]}") use_gpu_key=$(func_parser_key "${lines[9]}") - use_gpu_list=$(func_parser_value "${lines[10]}") + use_gpu_list=$(func_parser_value "${lines[9]}") use_mkldnn_key=$(func_parser_key "${lines[10]}") use_mkldnn_list=$(func_parser_value "${lines[10]}") cpu_threads_key=$(func_parser_key "${lines[11]}") From 5f4a4d2a01a3c13b10b5a3a8aa73027c0e03aa6c Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 26 Nov 2021 13:46:51 +0800 Subject: [PATCH 18/31] fix mkldnn --- ...l_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt index f35c00ea2d..1039dcad06 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2.0_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt @@ -7,11 +7,11 @@ infer_model:./inference/ch_ppocr_mobile_v2.0_det_infer/ infer_export:deploy/slim/quantization/quant_kl.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o infer_quant:True inference:tools/infer/predict_det.py ---use_gpu:False ---enable_mkldnn:False +--use_gpu:False|True +--enable_mkldnn:True --cpu_threads:1|6 --rec_batch_num:1 ---use_tensorrt:False +--use_tensorrt:False|True --precision:int8 --det_model_dir: --image_dir:./inference/ch_det_data_50/all-sum-510/ From bd2a31e781ebff16acfbac90bd252ce5f82cd851 Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 26 Nov 2021 13:51:17 +0800 Subject: [PATCH 19/31] fix mkldnn --- ...l_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt | 6 +++--- ...l_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt | 6 +++--- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/test_tipc/configs/ch_PP-OCRv2_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_PP-OCRv2_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt index e9524b65a2..1aad65b687 100644 --- a/test_tipc/configs/ch_PP-OCRv2_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_PP-OCRv2_det_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt @@ -7,11 +7,11 @@ infer_model:./inference/ch_PP-OCRv2_det_infer/ infer_export:deploy/slim/quantization/quant_kl.py -c configs/det/ch_PP-OCRv2/ch_PP-OCRv2_det_cml.yml -o infer_quant:True inference:tools/infer/predict_det.py ---use_gpu:False ---enable_mkldnn:False +--use_gpu:False|True +--enable_mkldnn:True --cpu_threads:1|6 --rec_batch_num:1 ---use_tensorrt:False +--use_tensorrt:False|True --precision:int8 --det_model_dir: --image_dir:./inference/ch_det_data_50/all-sum-510/ diff --git a/test_tipc/configs/ch_PP-OCRv2_rec_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_PP-OCRv2_rec_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt index aff419e8a2..083a3ae26e 100644 --- a/test_tipc/configs/ch_PP-OCRv2_rec_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_PP-OCRv2_rec_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt @@ -7,11 +7,11 @@ infer_model:./inference/ch_PP-OCRv2_rec_infer/ infer_export:deploy/slim/quantization/quant_kl.py -c test_tipc/configs/ch_PP-OCRv2_rec/ch_PP-OCRv2_rec_distillation.yml -o infer_quant:True inference:tools/infer/predict_rec.py ---use_gpu:False ---enable_mkldnn:False +--use_gpu:False|True +--enable_mkldnn:False|True --cpu_threads:1|6 --rec_batch_num:1|6 ---use_tensorrt:False +--use_tensorrt:True --precision:int8 --rec_model_dir: --image_dir:./inference/rec_inference From 66d1b2ed212a7d10627b48f750fc5da13312ee51 Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 26 Nov 2021 13:53:17 +0800 Subject: [PATCH 20/31] fix mkldnn --- test_tipc/prepare.sh | 3 +++ 1 file changed, 3 insertions(+) diff --git a/test_tipc/prepare.sh b/test_tipc/prepare.sh index e021e7af35..5eeecc9860 100644 --- a/test_tipc/prepare.sh +++ b/test_tipc/prepare.sh @@ -162,6 +162,9 @@ elif [ ${MODE} = "whole_infer" ];then fi fi if [ ${MODE} = "klquant_whole_infer" ]; then + wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/icdar2015_lite.tar --no-check-certificate + cd ./train_data/ && tar xf icdar2015_lite.tar + ln -s ./icdar2015_lite ./icdar2015 && cd ../ if [ ${model_name} = "ch_ppocr_mobile_v2.0_det_KL" ]; then wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar --no-check-certificate wget -nc -P ./inference https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/ch_det_data_50.tar --no-check-certificate From 76c23cf03d8b3019c07e53b315f1b7587f152b48 Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 26 Nov 2021 13:56:16 +0800 Subject: [PATCH 21/31] fix mkldnn --- ...linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt | 8 ++++---- ...linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt | 8 ++++---- 2 files changed, 8 insertions(+), 8 deletions(-) diff --git a/test_tipc/configs/ch_PP-OCRv2/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_PP-OCRv2/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt index 2cd2ba5f1e..cf4d3fde02 100644 --- a/test_tipc/configs/ch_PP-OCRv2/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_PP-OCRv2/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt @@ -5,12 +5,12 @@ infer_model:./inference/ch_PP-OCRv2_det_infer/ infer_export:null infer_quant:True inference:tools/infer/predict_system.py ---use_gpu:False ---enable_mkldnn:False +--use_gpu:False|True +--enable_mkldnn:False|True --cpu_threads:1|6 --rec_batch_num:1 ---use_tensorrt:False ---precision:int8 +--use_tensorrt:False|True +--precision:fp32|fp16 --det_model_dir: --image_dir:./inference/ch_det_data_50/all-sum-510/ --rec_model_dir:./inference/ch_PP-OCRv2_rec_infer/ diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt index eea9d789dd..6a02395171 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2.0/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt @@ -5,12 +5,12 @@ infer_model:./inference/ch_ppocr_mobile_v2.0_det_infer/ infer_export:null infer_quant:True inference:tools/infer/predict_system.py ---use_gpu:False ---enable_mkldnn:False +--use_gpu:False|True +--enable_mkldnn:False|True --cpu_threads:1|6 --rec_batch_num:1 ---use_tensorrt:False ---precision:int8 +--use_tensorrt:False|True +--precision:fp32|fp16 --det_model_dir: --image_dir:./inference/ch_det_data_50/all-sum-510/ --rec_model_dir:./inference/ch_ppocr_mobile_v2.0_rec_infer/ From 134e70704ef6292e5644333612c37b5ed2b24e6f Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Fri, 26 Nov 2021 06:49:35 +0000 Subject: [PATCH 22/31] add rec pact to tipc --- .../rec_chinese_lite_train_v2.0.yml | 101 ++++++++++++++++++ .../train_infer_python.txt | 51 +++++++++ 2 files changed, 152 insertions(+) create mode 100644 test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/rec_chinese_lite_train_v2.0.yml create mode 100644 test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/train_infer_python.txt diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/rec_chinese_lite_train_v2.0.yml b/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/rec_chinese_lite_train_v2.0.yml new file mode 100644 index 0000000000..b06dafe7fd --- /dev/null +++ b/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/rec_chinese_lite_train_v2.0.yml @@ -0,0 +1,101 @@ +Global: + use_gpu: true + epoch_num: 500 + log_smooth_window: 20 + print_batch_step: 10 + save_model_dir: ./output/rec_chinese_lite_v2.0 + save_epoch_step: 3 + # evaluation is run every 5000 iterations after the 4000th iteration + eval_batch_step: [0, 2000] + cal_metric_during_train: True + pretrained_model: + checkpoints: + save_inference_dir: + use_visualdl: False + infer_img: doc/imgs_words/ch/word_1.jpg + # for data or label process + character_dict_path: ppocr/utils/ppocr_keys_v1.txt + max_text_length: 25 + infer_mode: False + use_space_char: True + save_res_path: ./output/rec/predicts_chinese_lite_v2.0.txt + + +Optimizer: + name: Adam + beta1: 0.9 + beta2: 0.999 + lr: + name: Cosine + learning_rate: 0.001 + regularizer: + name: 'L2' + factor: 0.00001 + +Architecture: + model_type: rec + algorithm: CRNN + Transform: + Backbone: + name: MobileNetV3 + scale: 0.5 + model_name: small + small_stride: [1, 2, 2, 2] + Neck: + name: SequenceEncoder + encoder_type: rnn + hidden_size: 48 + Head: + name: CTCHead + fc_decay: 0.00001 + +Loss: + name: CTCLoss + +PostProcess: + name: CTCLabelDecode + +Metric: + name: RecMetric + main_indicator: acc + +Train: + dataset: + name: SimpleDataSet + data_dir: train_data/ic15_data + label_file_list: ["train_data/ic15_data/rec_gt_train.txt"] + transforms: + - DecodeImage: # load image + img_mode: BGR + channel_first: False + - RecAug: + - CTCLabelEncode: # Class handling label + - RecResizeImg: + image_shape: [3, 32, 320] + - KeepKeys: + keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order + loader: + shuffle: True + batch_size_per_card: 256 + drop_last: True + num_workers: 8 + +Eval: + dataset: + name: SimpleDataSet + data_dir: train_data/ic15_data + label_file_list: ["train_data/ic15_data/rec_gt_test.txt"] + transforms: + - DecodeImage: # load image + img_mode: BGR + channel_first: False + - CTCLabelEncode: # Class handling label + - RecResizeImg: + image_shape: [3, 32, 320] + - KeepKeys: + keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order + loader: + shuffle: False + drop_last: False + batch_size_per_card: 256 + num_workers: 8 diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/train_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/train_infer_python.txt new file mode 100644 index 0000000000..40d89a3af5 --- /dev/null +++ b/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/train_infer_python.txt @@ -0,0 +1,51 @@ +===========================train_params=========================== +model_name:ch_ppocr_mobile_v2.0_rec_PACT +python:python3.7 +gpu_list:0 +Global.use_gpu:True|True +Global.auto_cast:null +Global.epoch_num:lite_train_lite_infer=2|whole_train_whole_infer=300 +Global.save_model_dir:./output/ +Train.loader.batch_size_per_card:lite_train_lite_infer=128|whole_train_whole_infer=128 +Global.checkpoints:null +train_model_name:latest +train_infer_img_dir:./train_data/ic15_data/test/word_1.png +null:null +## +trainer:pact_train +norm_train:null +pact_train:deploy/slim/quantization/quant.py -c test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/rec_chinese_lite_train_v2.0.yml -o +fpgm_train:null +distill_train:null +null:null +null:null +## +===========================eval_params=========================== +eval:null +null:null +## +===========================infer_params=========================== +Global.save_inference_dir:./output/ +Global.checkpoints: +norm_export:null +quant_export:deploy/slim/quantization/export_model.py -ctest_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/rec_chinese_lite_train_v2.0.yml -o +fpgm_export:null +distill_export:null +export1:null +export2:null +inference_dir:null +train_model:null +infer_export:null +infer_quant:False +inference:tools/infer/predict_rec.py --rec_char_dict_path=./ppocr/utils/ppocr_keys_v1.txt --rec_image_shape="3,32,100" +--use_gpu:True|False +--enable_mkldnn:True|False +--cpu_threads:1|6 +--rec_batch_num:1|6 +--use_tensorrt:False|True +--precision:fp32|fp16|int8 +--rec_model_dir: +--image_dir:./inference/rec_inference +--save_log_path:./test/output/ +--benchmark:True +null:null \ No newline at end of file From 8ffa50feb7643cdaef76dbdcb2acbd61c9f338b0 Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Fri, 26 Nov 2021 06:50:03 +0000 Subject: [PATCH 23/31] opt load model --- ppocr/utils/save_load.py | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/ppocr/utils/save_load.py b/ppocr/utils/save_load.py index 4b890f6fa3..a5326edd69 100644 --- a/ppocr/utils/save_load.py +++ b/ppocr/utils/save_load.py @@ -57,7 +57,7 @@ def load_model(config, model, optimizer=None): if checkpoints.endswith('.pdparams'): checkpoints = checkpoints.replace('.pdparams', '') assert os.path.exists(checkpoints + ".pdparams"), \ - "The {}.pdparams does not exists!".format(checkpoints) + "The {}.pdparams is not exists!".format(checkpoints) # load params from trained model params = paddle.load(checkpoints + '.pdparams') @@ -67,6 +67,7 @@ def load_model(config, model, optimizer=None): if key not in params: logger.warning("{} not in loaded params {} !".format( key, params.keys())) + continue pre_value = params[key] if list(value.shape) == list(pre_value.shape): new_state_dict[key] = pre_value @@ -76,9 +77,14 @@ def load_model(config, model, optimizer=None): format(key, value.shape, pre_value.shape)) model.set_state_dict(new_state_dict) - optim_dict = paddle.load(checkpoints + '.pdopt') if optimizer is not None: - optimizer.set_state_dict(optim_dict) + if os.path.exists(checkpoints + '.pdopt'): + optim_dict = paddle.load(checkpoints + '.pdopt') + optimizer.set_state_dict(optim_dict) + else: + logger.warning( + "{}.pdopt is not exists, params of optimizer is not loaded". + format(checkpoints)) if os.path.exists(checkpoints + '.states'): with open(checkpoints + '.states', 'rb') as f: @@ -100,7 +106,7 @@ def load_pretrained_params(model, path): if path.endswith('.pdparams'): path = path.replace('.pdparams', '') assert os.path.exists(path + ".pdparams"), \ - "The {}.pdparams does not exists!".format(path) + "The {}.pdparams is not exists!".format(path) params = paddle.load(path + '.pdparams') state_dict = model.state_dict() From bba76ec45b22b77a8b6d1a3ae61698515620f418 Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Fri, 26 Nov 2021 06:59:50 +0000 Subject: [PATCH 24/31] change epoch to 1 --- .../ch_ppocr_mobile_v2.0_rec_PACT/train_infer_python.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/train_infer_python.txt b/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/train_infer_python.txt index 40d89a3af5..7bbdd58ae1 100644 --- a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/train_infer_python.txt +++ b/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_PACT/train_infer_python.txt @@ -4,7 +4,7 @@ python:python3.7 gpu_list:0 Global.use_gpu:True|True Global.auto_cast:null -Global.epoch_num:lite_train_lite_infer=2|whole_train_whole_infer=300 +Global.epoch_num:lite_train_lite_infer=1|whole_train_whole_infer=300 Global.save_model_dir:./output/ Train.loader.batch_size_per_card:lite_train_lite_infer=128|whole_train_whole_infer=128 Global.checkpoints:null From 17a4642103db8acdeece2860b18497a6457ad282 Mon Sep 17 00:00:00 2001 From: zhiqiu Date: Fri, 26 Nov 2021 06:36:51 +0000 Subject: [PATCH 25/31] add no_grad --- tools/infer_det.py | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/infer_det.py b/tools/infer_det.py index bb2cca7362..1c679e0faf 100755 --- a/tools/infer_det.py +++ b/tools/infer_det.py @@ -53,6 +53,7 @@ def draw_det_res(dt_boxes, config, img, img_name, save_path): logger.info("The detected Image saved in {}".format(save_path)) +@paddle.no_grad() def main(): global_config = config['Global'] From 2c003d56457644a6dc2d3aeaf9b90fb23abdc6d5 Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Fri, 26 Nov 2021 09:26:29 +0000 Subject: [PATCH 26/31] add rec kl to tipc --- ...rmal_normal_infer_python_linux_gpu_cpu.txt | 21 ++++ .../rec_chinese_lite_train_v2.0.yml | 101 ++++++++++++++++++ test_tipc/prepare.sh | 7 ++ 3 files changed, 129 insertions(+) create mode 100644 test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt create mode 100644 test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/rec_chinese_lite_train_v2.0.yml diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt b/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt new file mode 100644 index 0000000000..92f33c58c9 --- /dev/null +++ b/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/model_linux_gpu_normal_normal_infer_python_linux_gpu_cpu.txt @@ -0,0 +1,21 @@ +===========================kl_quant_params=========================== +model_name:ch_ppocr_mobile_v2.0_rec_KL +python:python3.7 +Global.pretrained_model:null +Global.save_inference_dir:null +infer_model:./inference/ch_ppocr_mobile_v2.0_rec_infer/ +infer_export:deploy/slim/quantization/quant_kl.py -c test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/rec_chinese_lite_train_v2.0.yml -o +infer_quant:True +inference:tools/infer/predict_rec.py +--use_gpu:False|True +--enable_mkldnn:True +--cpu_threads:1|6 +--rec_batch_num:1 +--use_tensorrt:False|True +--precision:int8 +--det_model_dir: +--image_dir:./inference/rec_inference +null:null +--benchmark:True +null:null +null:null diff --git a/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/rec_chinese_lite_train_v2.0.yml b/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/rec_chinese_lite_train_v2.0.yml new file mode 100644 index 0000000000..b06dafe7fd --- /dev/null +++ b/test_tipc/configs/ch_ppocr_mobile_v2.0_rec_KL/rec_chinese_lite_train_v2.0.yml @@ -0,0 +1,101 @@ +Global: + use_gpu: true + epoch_num: 500 + log_smooth_window: 20 + print_batch_step: 10 + save_model_dir: ./output/rec_chinese_lite_v2.0 + save_epoch_step: 3 + # evaluation is run every 5000 iterations after the 4000th iteration + eval_batch_step: [0, 2000] + cal_metric_during_train: True + pretrained_model: + checkpoints: + save_inference_dir: + use_visualdl: False + infer_img: doc/imgs_words/ch/word_1.jpg + # for data or label process + character_dict_path: ppocr/utils/ppocr_keys_v1.txt + max_text_length: 25 + infer_mode: False + use_space_char: True + save_res_path: ./output/rec/predicts_chinese_lite_v2.0.txt + + +Optimizer: + name: Adam + beta1: 0.9 + beta2: 0.999 + lr: + name: Cosine + learning_rate: 0.001 + regularizer: + name: 'L2' + factor: 0.00001 + +Architecture: + model_type: rec + algorithm: CRNN + Transform: + Backbone: + name: MobileNetV3 + scale: 0.5 + model_name: small + small_stride: [1, 2, 2, 2] + Neck: + name: SequenceEncoder + encoder_type: rnn + hidden_size: 48 + Head: + name: CTCHead + fc_decay: 0.00001 + +Loss: + name: CTCLoss + +PostProcess: + name: CTCLabelDecode + +Metric: + name: RecMetric + main_indicator: acc + +Train: + dataset: + name: SimpleDataSet + data_dir: train_data/ic15_data + label_file_list: ["train_data/ic15_data/rec_gt_train.txt"] + transforms: + - DecodeImage: # load image + img_mode: BGR + channel_first: False + - RecAug: + - CTCLabelEncode: # Class handling label + - RecResizeImg: + image_shape: [3, 32, 320] + - KeepKeys: + keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order + loader: + shuffle: True + batch_size_per_card: 256 + drop_last: True + num_workers: 8 + +Eval: + dataset: + name: SimpleDataSet + data_dir: train_data/ic15_data + label_file_list: ["train_data/ic15_data/rec_gt_test.txt"] + transforms: + - DecodeImage: # load image + img_mode: BGR + channel_first: False + - CTCLabelEncode: # Class handling label + - RecResizeImg: + image_shape: [3, 32, 320] + - KeepKeys: + keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order + loader: + shuffle: False + drop_last: False + batch_size_per_card: 256 + num_workers: 8 diff --git a/test_tipc/prepare.sh b/test_tipc/prepare.sh index bc9b29032b..d3676415e8 100644 --- a/test_tipc/prepare.sh +++ b/test_tipc/prepare.sh @@ -174,6 +174,13 @@ if [ ${MODE} = "klquant_whole_infer" ]; then wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_infer.tar --no-check-certificate cd ./inference && tar xf ${eval_model_name}.tar && tar xf ch_det_data_50.tar && cd ../ fi + if [ ${model_name} = "ch_ppocr_mobile_v2.0_rec_KL" ]; then + wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar --no-check-certificate + wget -nc -P ./inference/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/rec_inference.tar --no-check-certificate + wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/ic15_data.tar --no-check-certificate + cd ./train_data/ && tar xf ic15_data.tar && cd ../ + cd ./inference && tar xf ch_ppocr_mobile_v2.0_rec_infer.tar && tar xf rec_inference.tar && cd ../ + fi fi if [ ${MODE} = "cpp_infer" ];then From 389e315ae0e5f482e979c354b5f18bd6d7bed34c Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Mon, 29 Nov 2021 02:48:09 +0000 Subject: [PATCH 27/31] opt east post --- ppocr/postprocess/east_postprocess.py | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/ppocr/postprocess/east_postprocess.py b/ppocr/postprocess/east_postprocess.py index 672d16f9d0..c194c81c69 100755 --- a/ppocr/postprocess/east_postprocess.py +++ b/ppocr/postprocess/east_postprocess.py @@ -60,11 +60,6 @@ class EASTPostProcess(object): """ restore text boxes from score map and geo map """ - try: - import lanms - except: - raise Exception( - 'you should install lanms by pip3 install lanms-nova') score_map = score_map[0] geo_map = np.swapaxes(geo_map, 1, 0) @@ -81,8 +76,15 @@ class EASTPostProcess(object): boxes = np.zeros((text_box_restored.shape[0], 9), dtype=np.float32) boxes[:, :8] = text_box_restored.reshape((-1, 8)) boxes[:, 8] = score_map[xy_text[:, 0], xy_text[:, 1]] - boxes = lanms.merge_quadrangle_n9(boxes, nms_thresh) - # boxes = nms_locality(boxes.astype(np.float64), nms_thresh) + + try: + import lanms + boxes = lanms.merge_quadrangle_n9(boxes, nms_thresh) + except: + print( + 'you should install lanms by pip3 install lanms-nova to speed up nms_locality' + ) + boxes = nms_locality(boxes.astype(np.float64), nms_thresh) if boxes.shape[0] == 0: return [] # Here we filter some low score boxes by the average score map, From e97bd9d3aac5478dcf340d15e4dd70693c203818 Mon Sep 17 00:00:00 2001 From: Leif <4603009@qq.com> Date: Mon, 29 Nov 2021 11:15:52 +0800 Subject: [PATCH 28/31] Update joinus.png Update joinus.png --- doc/joinus.PNG | Bin 197726 -> 197265 bytes 1 file changed, 0 insertions(+), 0 deletions(-) diff --git a/doc/joinus.PNG b/doc/joinus.PNG index cd9de9c14beaf0be346a1f7f1d09450a0905a880..99964b62d0e8a5867d5eb7a29640f0414c7af3b2 100644 GIT binary patch literal 197265 zcmeFZc{r5+-#2XEx1zF*e4~uAWjFQ|5kewsRJIuV(%6?!VpL?!o_)=}uSMBLVJu^e zC1eJJ3EA(D@AbQm`~Ks)kLP&)c>a5iW2j@C<~%?9YkPZbps&Tuz|BBLM#ijt_qGuk z86^eyBBP@L?<9MV)PR5J-S3)tk&!X7pM8;&rG4ZiBjYF2zO8Qj!g?(O@izY<;B;6N6I`L4#HtZgWUGkn^1)%r7grD z+@1@H^l?RDrF8{bUKhnr7yj?7Pw6=L^181Dd8p~J5gV2I+3V%g1PP{f+z4m-`$T;wV(7MWTrPAD!B>km(8+-^kcF;j4^|sG zmXnj|2p_gU8KC)VetFXx&T!1zj#3=u^3*udG(9umu?#BlYJOr{!-4LeTW(zx@lf`( zwIc1bJj#MEyG)4e8;Q4_h&k;G!FqJytJE)i`F*aQ9OFDq2zl;VP^Xz>M;RrK^7`Yi%n?aX^Kz(1Mnr4C zeA=ztiF{{uEA@`yr1PX%(}Zyd@d99oM!}WqIjNZk^_P!MpFDWwWI2e)?9; zhx~cA!7AMFf%Y&BVEei8&xLwn_ zdkr}v*AMG&9pBgECrf;nc&NzK^E#HFVdq7rL*7J54mFb3r||Q<+a9c3&)9PPS!$Hg z(ko?&6uzv7$<{u1;g#;lbz?1e%Og<7B!po?YiL7}ixyk@6~7IG>G* zJW(>w+BLn&Q$p-Ff6H$8_Rp~m9%e$YogK9YuZA!+Zcht&@yAxquz_Yg^_1Yp*EbkO zUtj$COzQ_{U{jdWxISDqku{&ATBN3WeY(NvAH4^H+E08kt6}s-Qi-S1dhq7a+i3-l ztgY)kzsmAleQL|eVwqmB8ah;wha*0(tgEy{Yy>yKhBg~`$&1dVodGXSv@}W%4 zg6B>KoKa{AT6LQb$Fl*%uO596(>BW#Ht8zqYDA_$%#|0J(I%yK8BHZW^7Hxvlu zWDW|4uX;M$;C5LFrNMEBOt)|Tu>Cm1vI-loxFHYCchR~?)%9)IMU+ZCR3n5e1Fm&Pb*A(a|)-VmM?L-T+xZH0i%A!ja ztlqAdE?z8K%uT#iI$mO_z}GEto3~HieU5B-W#wE{jt{hu8WxtJPm6V(2(ITYbQiEy zEH5jwIXyXctTRWSAh}W^gZKXa9R^^xa zg_O8GI1hT9kyf@^r02U(mJ2bbt*FOGh4r#ZtZxf5_-PP*LLcPQ>s&f7FE10?v}so< zu33T$g)6lO6NGVq;qxg)#Kjl;Z&ZG9P(3s&r49Aq z?8W+~-(jPA(JPf1Hqp8)laiiZqd)WshLT5aMT!TK69aIE?sJ{w9Fb+lh@(Z-6M>|pBpUnAz2&6Z(-uEsuLbVF z_Iv)AD<8dv6gSLna^v=?ByC+IgzPV#Kr+wf5JQ%$bCeZz**0K+E z_4w6(-@b0ADG`M0%kZ0h?H0JmGk!0UUM)I3>N*|8Gqb40cvZf`-SrxzYDi0C))rAv zdlO3xz69nq?ZL51w?wVxYxodqa`}#-&B}0wy!Y>sbEwi2-yhS1bb+8pdf*LM`-zwb z!q|XSn_mbdKi$Zx_vVNOt2)7xrt3JM^Z^HT@ODDh9OL7^6z>JqIio1aI~x^WLlk}I zx*5e^u6(#>c5QWxypuKLuf`g-Wr7A)5uvoPUH(D8Fe17APyWa>Rk?Yx)RS7j9kG+Wd^ z!}5zvU;#s8%4V0FQj0TaCRJ*0j;Z6ZjS&y0&%;i@r?dLNYPk%wl9Q6$`yhUEVYGZz zl<>R*Fqf;qz~cMjI3iOoOm&2)x^^7<^2)n+?$+h~Wbch|Zdj1lteogMI%?es>6R#) z`MiJD<#&7)?;!cTILsy=)rw6Vk9IO0W@eC8`phdciQt*umnsFaL{RQ=m&bcE%&gqZ za@E>EZZz4sk5cKbbWhRxO6zb5?LG?i>0_COcRa^bfK5@C6(fQg8ooV7Xuc}p90{L% zUiY>@I2U@0w~vg*Axt)1L`s}tr3tYbFVMo2f(N?>UASkHpo9FQjy?)o&S^8H`s9@l zHNK0Q!jaGz8oW4a=^d&aSD9JJ$r;sa5Ej9j{)wE+w z_n>uSy{~b`h<**LNX+(exZ%>`-jSY#fD3Q!%29b@>Xh^IhM{LiXrH($FRK20b#R{6 zEp_**_fbKaMVRTSo2GA${%(ZGU+DBPk$>J3K$Z|iNNGlSjV^l4)XubS`!_t8BUzR7 z|6WGvMSDIHC79e1X8%oQU(h?=pxt>@;f}3+5R@^&z4lSPw4Nb{eeD>0!08IzV8D6H z_FC>fGUx=e(CewV-0OW75M*ynTS@)XnMPEG38c_c%H~)OH}Lx<*K3vz*-g zSeKWWr@1#uK@g9*fjtr4~>+dM=|V<%g}Pm8^tQg!k4XnjK>$EWG*C!-~mlU8(4LHdM} zTtKK-^o5NpIXiP3=Uoq4^^3c(E;C0QcbY7I*!LW@oVF->ORrZg;KMPp8^;g9WI3=N z(C65w!Q1RVT4{AKhPk#m!hN5+FL{{TOhHY@n=d~cE+?~lU-dE5G_%N@)-S*lgH;yx z=X^M|qf~PWTe9IaPxsQ9!CIzt6v(Cq{qQjI2 z9Ykm7CQKL#&Fp&ETK`SQY)W4C*%wrb%YaMjoX+_AF4@=BUqyLg&@*0Hs-W{fu9L@B z^uFGmi80@H__`TfVI2y7Kj`tHWDwDH>HjOWM*#gYQYpvL&$ZG}(I|NK=| z?D7~JYMignst_=MdtJ)cUHPW*dZyfZzYG@Gp#&c*d8LENBWmNe@GfeOOm*sdL z{9zAmmSrxKyQr_(zo?r%$tu9$IE4H6-rjIM*HWP%a5Ti} zXGgha&BMtnBa>COfjBqmZ#86btjuo-g7NsmZ}IboqnJ&(Ie%N_*y8zpnq*^Xcz2nFn&FVi~*Y207j?e zWjh1TJZZgk!?i@5;nJJjZ+K<9A2M^J6s**g{R!TqV;{K{OCa?OPI8$lZKp}oI{3;UrFL#V}^iYUM{o>oQ#-v zTTMEt`X4i>nVk(ez_`5lJ0DHE-RWE_R@er16^H1i+S+ILUF{4ymGmjiodI$B=>1ch zj7Dp165Ruc9H8<-S%H!O*tkT0>2bw>d zB`u010Ul$dLX4J$SFHux3(jj6fJUJ)0HL>FHA}}~3!$Hfo&fwr>8}m;@go1#q!hB9NUw1tmGl%F^Slc@s zzejubf!FqqaN)=BLt+PRI@A#UyMp)TTvr6>(aqLz>bOu3>pW=p{orkOsGWTbp})?|u^bSO ztxihT?D6Madqabqa0jb47wNGN&=`OSg&y&Ar1{F57?ylw22cmIoDz0Z6fVbte61Er z6Vv~YgW7jv=Jkz=Mv9J2#*R5_*n=0D3br2v&I9CEsCq0sd%WSdSdi{|%a92PE!SGz z7g34GcnG>&adwME>1eaZmns_aCLqW#<(!gl$&DK~8u2Bin?09LMqtGo%1-8)&5t

Op+KZ7PqaE9U`bRX1c3OSfH7jLpx#YUwp3)3}`GOk;m8YJ5>1(81nu)5meFLBBPA z@A#Yeevm92L{p1^kwpdX)@-?*?z(|-(cp{%y;!edF?5%x`PPn0UvXWUOw(kQZuc%01oA`{L7yfhfU@832cIVf5d+Tk zeI9o}a!h~+9#$gh9_>6nm4Rg->)vIq$iNc8gw8q`E8GlY4J4yw<*SALcxlv|6Ee^E z_|#duS;h+mC;g>WLy`Q!^beGVfmi6K>O?( zG+_tK_MF7j^Cv&zgI}&=8Z08D^$U?v7@0Xkj``2=L2Kt|(lV4(SD+tOQh&4t9q!Ke zsh-^2-M|Li<~QKzedj|-Uf)50vfM~Wm&@-rtjmRN9uq@wCqqQUP?ZCmrrd;4^?3FE zJi`7ZGSBgjEpJNm@F+>?Y&{EQc%P5SwcP$fC|%b8eBWbKPi|^BRf?kXm*$;~7NW`w zVkHA~B;VU}avcvZUc9)uaH)fHCBrWhvcIAH5 z5?_vvlxk(WURYty;Z-u!k=e+SQ#>o4-quC@S|r;`Ju>w0f&8o z9VfySy%Vza<}o1uWU-miwvJf}Ov?@k1Xlx~p5y;a*E@wB(?+yA`IJqmU>e-|Hn1&l z9vXGh`P!#k^%+L?vT^qce~arSd@N6%pKglG_ELLVQpd+BTr8#52jhKB*E52SUqR4? zA7orZQ)0q0z>&8m{qxI1{_7V*NPNMX0{w>oK2YrSMec38q17R}$)3g>c52@P^(b94 zo?8uR{Pyi*_W=baJTs{W=Yy2ViEV7wv&!aWjyPCHXbrnc zQ*(%f0J8F)Dnq}oKuRYX+A*d&y;+bdmefNU)V=*aB^hb9VOV9YT**uIWU7!HO};K= zhe+F|OY~XBrxAHVo*TDZE;V2c?-WreOxx=OECi&7ey1TUr;ivduPhmsG$Ene-e3pPtxVLjU)GITs&Zbtp zJdBP!Situd?Xqf+u3Viku5r~;Fn5mZt%nYdiP+T&izW^1>An31EAIbh-_yJpPfYid zC=iZc>?HMP$F0jT;46u&Vpq-MEFTn@RhE7}PUOn2y_zX^5X4;26<~t#ZKaj>)RIi9 zqetf%Evc!odq4ZTHa%3>{ZU+>mj#PN$=gD(!c++II=}H5Gn3Q6ec3TT<4@;{U*UKQNyrqsg{@ zDANDeEb%Ze*{sPJQ%R4$!oJKv|&?TLB`m!$P%V1>wPPTh@zdU-e?${KYhcBQABZcDR^`o z)klpcWvGrPSs3thfq<2wgr&P)hF?eHj;RVVQD` zS|JIsh|H7DNV%LU8}zTAxwvTa(a&WvNwbdaCrlJ#DX^)SGiAN+8z|*PFS#Y zw4x~)Yk8g?n~*Ev;8ew8R_Kal(Gae=DzuY$_@|}oIYfjuOZoX*TjfUBFIeKn$|!}; zE6)Y|SlI&2QAHr~XJnj6ZVBbAvd^k?z1${2uMbQ)ggiJH}Ee`bc$ zY3>&Pq|ZN3k(UwS6hV`Etm-gusZHwEIKNd^vi@ z)>`zx5cB7LSyJ3skO~7(b#2xT+kG}Ct(}8hA&FKn(W9(FB-%&9yL)AMnT(f9FitVc z`v_=NE!cX2{6+of+lJ2|Y$&P1ug~H7$GY!7nal!hj*W`0$6ZacGz={P=$rXUW{8Uf`Sh~D44q)Hf>GPNa`lC+!`E#*Eq zSQk)NoHpz`_3>+WB|yHxzOs2dxnVv z=J@R<6HfQ42xf|!U(ri0I`^H~CEkytsRzFCca!=Ji3NW^Px`h1X_XCNZ1T=;_CEH- zF~oF!CAz^P8G`rugHLs4=xpOATRYXJ=8qR(7#If_V7o(Id6WiJIduo@}_R&WC&79#w{> zil!F?x(b@7M_hAq2xyz_487|^)@64dW<_h4tompD+7EG;hxqb+CN+Tx;vS$#-4b86 zw_27 zJh;D_z!pLInu}ziBf3~l?X^zM;5P1PS#G`NuX2XY9$zr=P{n$$d1BrpKHwt)iY)W% z4j7fblUDFII^-`c(Cb-06&g1ccz*;G-npWQ=JDb5gbVkEpU=9APA?vS1Ka_+VeR)N zNT!mUl#^a>uqdEzmVsMeC+nOqB_$`rwJbR9JawpkAX#h>vgz_I9h*Z{t-x9Zgc zklicm(DGxzMw2hsd$c`=?|A?c=c@$!%+P_2sLZ z$d(?HetGeRS%2^#3k-*W(|mqa`7uTeXFc+{vC+}zqq@FT1~^n=HG%<1o|nqn9}H#h zw{hoeCjz`FA|f(Z?iO?du$YXY8!^lIp0=D%P?^eShEcGLRRKBOy(d%$G^sULg+YQoUjfM1zh>rx&}MXRK;3l zqS0hqfSmQg&*Uezwzki*B<~Hn@OL89qmr^9FmJl7!i3vAWGq>`4%9v@O0p`JWNZBO{$5m4WY(7T2{2g~%6M)f7rcO^md8+boT>*Te{752Q_v-psFmN|R{Nk?FB{s{ z;b*M3s~?Z3w7Lze_6+C+fxk5_cPrxd^?^}f zGdF!L6-O^0RKH z9IVltQTzQISTr;zfT~_#Rrxy?>ZLN6`7Sb&`e>GT%y;I1*#OOCq&cu7XV@?=_<>1@ z{{ugqxow8>$yR)yQ(;DwZwq{2Lhvb1oY6J+*4=66EkNxCheygZqmd6C{eVG;^~1%L zR`~^7@IG|rVTEqzt3&|h)nZHo&C(EZ`)D8wc!;&p zrTx7|+v67f_vkz=Kk@`qv;dsGs(J5`5=unl38JQKrmcuRp945miPtH$@x~CByb_i2 zf>`G+=AX)L?i1P9te&*B)Q7`ojoa%wcb{Nxx(z-~uv|ej6;rwYs)_0%l2&wh|LPTVhCM@1A{>cIkDw`h7r3A!NNOPBQ0v~zA?z;my z`Vvp-GW`NMZd2k+bLyu+eZ>lR{_M|T-CUvttBcSsO09`JLZT;(%z&G_Lc^E&M=i7g zW7dSEC248sT%CDtOj<)97ReUXvnvkbrAjjD443L@!657=lwz$Or32R z=Gu#OK|-s}gUb@467gjLmnKw(xyK&RP*jxZF;)!Tcu0fh5ndb}Mw57{|Koi;lKX!b zJCbU1ek&2w*plAZ65WWS+)=dQh@8Tux7fL#dT(&d%}T8WZ;0|m0Z@#|Lu4wbIZp1( zsD~@ENd!r*WtP(rhLo3*cg(Qr5svWsINebW5R=L9iRoSoerPK0n(n1&y4Qkd5%>Zx z&6Th~crvy$*P{A zW5uy#b~Rjr9$y~D?;bm|tAQ0C85rpvjvaG4S zh&Sq0Sa#5j*RQ~sx%rd?OpcNlqubsXEgwU7y?>?BS*~;J>W&^^d+>dMk}Hj#9VfM2_js1}ckP`S+U(jwJ-azI8lpG@JKObq zwx}JUGRAK6w`yvW#!*}vzTRtpN^s}6-S@b>JFLBd!f_ zMP#?n+Z3eH((g?u?H|WgpqE;Q8nw$e-_@#QAm&JQ4rfSF`3z^Qq@LRi?uZA6!BTU@ zNuPR>9dH(aCH!aRXgoioKzKo3rQ4QkoW_s?L@+FrsdONf_ZTH6IPQc2NK9%7I`mz` zb#!di{jw2Q+wHwv-3h!eZu^Pig6?Ug`;<23DPYy`9pg zF9*{wWwAIomHE;OEd!P>2q^~ z`N~!8^I#;;>;jlx*+jt~K#e+ZeHB(M;CiY~Dhj7GWE*xO2F(osDWFzB&Z}2g*T{&d zsH4AVQGT)B4d+^E+Rs~W;_$c@eFo-B$HURRU2#`xew9Gel4+dEJ${e;Od6C*A7*JSCfP3#r=Q0DI@pkpzODa zfZ7!dRkm!vEQ5B&cNMu;MWhOQx2==;>D&RBumw(w)+f1s+i3>ScpleA@~o3Th3{7E zr^p~bu0;nsInhs8`9a!GkBJtAAK;8*exX2wet%D_t0wE%e9h5_o294TaiIP%mr(2g#!_Gh~tRmAP zsavM*P}8VTC@)({Wy2%B$Jv_14W8do`2__@-mw3VTVbaEb!$@E#9ZSdGQ!AS*fzn% z$eM5eo7Qw_q-?IXHI=RC*j`z9Ro95&m^s)lXWV<38~2yfbLu>1-aL_Dn{!Z8@WihB zZslHTqeBI4c$G`t)j&yaAAOU-4-5=69^gy2mc&C4Ht5cE_nAN6fTC`Fj5l3et3lOw zF;r65pZv`J`t{DypF-8I%yKg}8JEy@30`PmX#szT`7yY(eDs!9ZZSI%u-T==2ku?e zmnQ9$T}anfgqSqeX4`liGJKLg{4*={nu{E^^>H0ca+CT;sy%l{IT~=miWSZH>pU}6 z*3iKV)DxU+xDKO@jpn(#fEc=%#jD(kgr`}POG z`6)+41_)t3yJv=wIO6-gP4J)R+4qa?B~Q42=U>;s?u1wI{d*K%aZ0;F*I&0+Didn! zm~`4W{4mMR_x8_776HYlIJp5cvV_bP=F4!qLb{Y6hGOd9ylRZ|#M;UkGU!oK1Gf?- zS=`yz$gF6)*U1QTJ7ua9^t>|M-|KfEuj%4zDw zBdE*uI-ATP&w^yX=QBDqlgi>>0qBj^=cHhJy3|>JACf8e`)$oo5U$LF&rh++ zB~XQHNkmwN^R|;!yM(1_vfAoz|Jipfy|Fr8<&E1L%cIAxoZ|;`$Cfv&q60(;MZ%Cv zrbJ=dgcN%78{;rrOu2nAOUR5`sHBx&B}U&onS8(Kz`ZgD{*+HXCdkX38Rx#iicTKWV&1H4Y1W~tkXU$v$HMZG<8*5=Q{Kkjv8S0y zX8ICbuj_}uZE(qf>2UzZc&IzaMqSU(3ZvuUy&e7lb>dJ)Mt4orr1KYR6|{^>zlG8w+cR>0V4FEiYp! zlhP$B0S1xciRx;8%7umb5sDdzwy@CRQW2@>_e}_ooD&biEu0+2Yx^K>a{j4|0Zn81 zbATogS-1J#OEc>=qxav{1yVVE!G!$Lc+xK3IIOsv)y%Qq!~O zQ1PSG_r+SLRgaFqeXZ>0-unTQVKAjeYUuqRRQ;lhN+`tzj{Et8zJY57DUYoy6ITtQ z^OFmfinA+7{hl&hNAgJ8U09_TvM-W&2!N7-&Ob3%(yi-MN*v39>Q*hnQ#4v2$~|?m z?#mO}l$6C>sDV20z`2M-aCTL`tOfyRh z+iqb-2=p zK&qchum;&jKR}yk;q{ZRY5PIsZa!w+98*!Jun1y%h^sv|<~qVaSSaC`H5o2EYTmCW%n45IBIVM9G64Q+5-9#+% zwdMyFc10D~rK8IdEYZYd0gQQiDjjm8W!38s61C#jIY)O*@m`WBYREAlVRS%$ru;JM zh8FQ!n4)KS-RiH@J=-d&_(e`&l8jocb_!}6HT87*gHS?!(A2Ce*3=fVf$+R3}?= zH*EClMFZ#B+6doy6-}6LcBRkkwmKu*`}F zD=-X5!T|f(mUH0?`~5Hd{R|)iw2tV&yrV40%O?eauron=zdc@18_CJMc)S+dBk=_Z zQKTt|+gD18U;UiG7)}OcDv)`pud(Bj=eE}3i~>Lyro3j*i*=$XCU z11GDYxpo$$w- z&UNM=O8Y1WzElJ6{CoEu+m8ehsT#VY!M!0e?Tn`^oNcS#uuY}KzlZHyjgHOH2YQ&^ zLoT&3YQ9jXXnEwj0{h~Be4d;~x)~FandGn*_DP z36&judoy0F+pWA;3rC5*VmcoL4s?W%7Q|1kWCC<-QTP_ZMfA&{49;3nXg^+eTclO} zOCHl5ds2&m6R3Wqe#!$e40_gZvVh&Svv7wkWLqA2?#?@(pV&W2l^?Hntk+}QL1@C9 zR*-3^Ie6HX7_MYM%mUB*?y zXY3_y^sEoYZUp6;EVTm~2N zTCgV4qEYf-NaSW;2=Kj_Bfk-!`1u+rP7rOFd4g$#4#Jk41oa+(0L=Z)tI3uWoxjN; zBL$uZ?eWS1yK2Hxl7lEtvhke6Zn^d?<{9r%y))CO%hgaiex#CmI+8T3$oemWya229 z*=$zaXrw%gsr$EO)^Gq%sQURs#CwdRQ7#?p<5FarAf$^=vSicxw<<}4?7m4cwDP8=pl#&Skzi) zgBuRH*KvCM0{s<% z2WzFgSUigxSnl-95E;0(uvXv{ z(-yD?CeVU-*@NNKiw&YBRe8BJgnH@$`B>M6H=xR4>czVA<2TS4P-p|nPT&dzSSLX9 zt%q1*Z@?^x56nw2hp6m@zx@1Zl&8`74lGJwGS~nCz5XQe`eWD2ek#ey$y*>Z#Burb zZyy|Mn3XpD58a?-vXY$oi-#cvuO*l~&}I(vdT##gz$VMCqgQHTDqxvrg_Y=VIf!e& zH@A9X0elmy7ZWYBk;jL782_af#l@Gh%W-aj0>EMs${D|&+$xj*)wg*Oskqq?a@YW} zjS`{T$3+GHV9f&IqA)AC8Io3pW8nodx~ntih;NZ1p3@_(5XRR_1IepZy0};;kp47<29+D(seay>TEWk@=4u5 z>;n7CA%KW%U5pj!%ZKhgGP6hTdyb_GoP@5X#B}TWvie(85foKXtM! zqRAX0Vq()jj(}aBP7}cfb7@T;?ghETG7VHMNCYr>Dsts>oGA*3OsoFHm8Kv>?CN+| z>+a!lSuV{N_*C;fYgQOcNx+`p3x)xJ;tkua$(0xgzJk2PC{_x+J~{)N~fr!rW+AW_{vSHH@qK6y9Fd51&o>msg( ziTN!CQ6rFkaq`ErR0JX>s`B|;EWetw)DwLnY-CoGr&`G=yY)cx_iE2VZd565ivpVb z2?!iYBb}y%sbk!_U@9MJcx4=LjDertfto(o36Ti940pjYbInV$dB#n{mV~bLyU|_E zW6eyTep5YM>b@hqs5UZhJ6l9MN;aUjLMuVeZXljT$brVzV2FG0Lql}vfpnd0mptLE z!CC>g zEgsg@fvsK9b}JYi@<4Iq^U@sSh?2!kmYD9z8GZ?(YU6(CfRK%S=8JOYth5aejBw1L zzM2bdwt{l!OwXUt)oQyMGR$k=(PN8g+ZamnoJ=Q~D;4Jo-jxkF??^RJuK0OhEIpN* zpY37UbIf692G7m-S3~jlT{28!x<4*_m!f)PjT66LF+R;J@&(mMh-gt3!o)g| zv28HSTN7Mmw7BNlf2uXrFq6ocT*Y;O%r3W4f9}+JNM(J)XGU9yZqSTEPtm)Kc6oGf z$Tw5XY!)^H(TeGI3He5%#rt-ROuLBd?CXw>ZPfp$b$)9>OhnlHK`NQ2uPdc$?X?&D zNujB9YryRCzG?=AF8BQoHC=slnz-vkf0^5=q_X(r7Q4!FFGnJH#(lqx2cfUpRcfnD z01@w`rQS;hVaZwnw`odYov?p6AGBBz0rUq1lAh2v^z;ElIDbuneq)~|&)Z44>aoqq zH8ZV>ih#^1Q|E%D3RiR_IIH za3MMOV+QdLdW0@7D$D%zi`15)WdRM4T3OoVDv^q=CAU2`KI=Y>idxwfZnAvC+$~>L zF@cR76+|Yhw@Bifk}lK!wRAm=--vvH{!^|&rieqsOOYPNh*QC5_njA0NMw)+1h2_u zFz3S((;v&F6fGR8Ot04;<3xEd2+ljUEHmFFy+%O&H!$7XtSkLj&o%B0AKfGZbG#qP zj~eKi3;me`i85N&jg&{uBbJrOZT`&eI7e=;9WL7lOo%bM<1N=M?fu`WY`t06|WWb0bQ6GrRX?^SdOLUdnxN$D~384 znP&PTF=XrRUIHkH*9}Xan;QqeNH5qvp?ll6jDI%=R(t?3@C@W6F(b(m;tf56M0z|& z-L{Xn*@iqC_Wf;C2isBWsFz$URKF8BBDsPd_Qx*#a7r8w-i zU%4tD3W^3@5n4!>v>t$H<&?fL-qk_4SibE(_PX(#?@#p`bfUpvPWocxNf-at`_4)5CfXWBSgKlIvRaojjCHKUc zG{8v@UI9AJqdSC1AI<1DS$a>k3hF#uv1AGBrsWQoaC_^6A_QqBjau_^bA%Zy)0fp> zbr|@wR0TGVnf0Q&`n+Gj*!FjaYqQ&O=8mpTvUep2raofptk2V905-m>x`;4mAqBFF*hC>2iu^1N*a; z1VMU`;Nc7YYa3vmPcUfiO^!;1fFj>BY5(jDOJAaw)~~op8ym53XfF*)YZpN&K{Zm9 zWLWMnd%v#{>-rSbJMQj-KreEJ-75MAuv~&o32bE-P$c&mkG}a&3HqS=?##r#riY*! z^en}nhk}M8Gtk^1Rx;TU(gBzAi^|q_4QS*0H!8ImbQu^0fF*!U=dbd)l}e-ABy?4Z@DCluS&Sd79*~I> zsxG#cKL`OFtyHTPvB>hGR(9As$xeRj3p*$&X1--2Amt1q8|0wmwu|;Ms8s=~WzDLz zf&h5Bv(jU5^Xh^`1@aeTF0(&wnc)HPJ#h-CtH^50K2{F=yQ%gT7(Qh)^0q1w{8jZ~ z!y$n9H+#%auYiit;7-?x@$Vv?skxwX0d&Sc$i>&Sd;U4cfQ?qc70TZ$7lfE$f8#Z1vMb~y{uh5>`mLYAQcjDLyN6OCrK09Q1Dcs*9Rmp4-}#9ZEgQFzQn|G7Uj&~eppLp& zgG2q=Nf~Fkay=`)iF=tV^2dKhLGq-v8$bK$=k#@VwcoGr2VOY~ZP_|GeXl>P1rCa~ zMOwkxbzrea=cWvX4w={;DWfl5LwSKsMu}jCIxB-PKK6ew(#)Iji>M@b?}>a}`Pi={ zL0#_aPwHZUqL6p_xac9er2Dy5yYXF6g1X$g*GBwWler_JDETFoyur(pxp4*Y2rwg% z?KcWsy75@O!wt<$8;0;mM){6DZ2_CY;?v8AoM*)zic48`_U4JmI}mBjKv_le6IK0i z{!YMdH-8DKPNBh8kjj?i!O1&+ct#Bs&8%^PUVsb(7pV*IR*Gz<&Y5bKr@_H|ZZCRx zHu0t%+(5Zp;syX&Oo6i(xJFfOIF*^;!*#OQh`t}A1=;zalwARZU*aVBGc zgZ$d&4HIRatbW)C3h>WL$G5;OCzGzyQ4osaEUeKQnJbMTsN=C-HvzcPpkM#x0~@D<$~Q5%av!)iVhPWZ3=pn@6Sx-yV;u9;x0<^ZNbAN0NSPqX*LMa~zjS|A@=K4;DYw%K|zP=`!9ntp1V#nz=fk~2HCC6>g2Y;=^=ioTxyC8C^=UGD`+b|MCq(N z=P(1Ez{O_9IL6M>NF8SO^DDWgZ7(Hx{@`7KNKj6x&+lDTNDkxl&X&h_>HCy2dRj$m z7B)KJ)AfZFKjuobOgX}Ny?rYzjex0ktemCyuWbs#SLYN}A|RE-^h?<@f5@KNOsaqb zL;H#EyBzi|WG)L0v-JPs?5*RPe8c}y5ClX@Md?sl5fG406_hSPX%I$8$A&bBg3`hO z>5}g5R$>Dt;fPU^!su?!z0Wz{-#Nd3&g*r~Um|X9d-mM-eO>RmUV+^J80Fqou1JKt zm+wvSTF|430_aW-rSAMbZgrX^2hDT@|Mq)_coXNvaP@|>aC7Gt7}fV2fq*_y(|aXF z47HCi!E8Dye{4TO!n17JPn-U>b=VQdpAa(SR@B}_;50LL* zl`gM-*YZ!&MDX@x8=Eynl0HCRTWb9CKJ(A}iFQX|sQ;rf7J;5b0DC_gyipRc>Ha}J zPP1nG4YY}}#dqo`fyo+=vF8bZ?9!40@1XYH6Wo({ zv;$#BTKH}%ArV0=iFVyJgXB3FxEqxdCroBV3C#Y;z=Se+KV>Gi{Ao;isvKi!)(85o ztsKUM-)1kg2y+kzKhOOQGQJhi*B0)$>`AGeuqf0R@j1R6{&YqrL9U#5%eVINm50k# z_8pKAVrDBqD>n+h1GphimDQM9*`%Vb&n&O;J8y8_dvCF4d?N}E|1Ge4WzMa;ncsQ3 zBZVqmDzlUI+vJl(2=48&YNn?p%U%vh3^)4S2nXRK^1IEho;|y}t_!~HX6=fJJ5oa4 zuZhc6p1Q+1Z{gwPGK)X&^fG5o-ff^OXR7A&csl=TBOkcJ-}rIrDk7@272iw*$G!@K zWbom0`bwaow(q5nhb1)~1@T5_>-6#^|D2*KmlNJid)G7C$NavBzwArYK6)40!k2ER zfEA?0x=ZLheOsRhww_00)xM*7wjFUHg+?OehyAyO7)bD5lCIZB^oY6aiSVXpx2D4i zJxec5a|(ZVWdy!k`AyjNX012jYZT|>Bk}Bf6CtGq{e-f$D_yumrH*c)5x>zJ(M{>D zO{Y$^*aL~epcUwOa&hiPU81+)zu6cTzmGqCRnYM~vED=cDqZ32C9e=ygDZ>2=4Jb|TD|X# z)rL&x#{T!#mAE|bMnjKA!&iE;Z1&!M-PTQ*d%KrwzJH#Fy_KC3(XgwET33}Hp~<0}Kd55h&~%t5#q zDIiQ-R3ZHq%{hLreJ(!3Hc8zwd5M(-2=U5b%dx^aInaje(fsLR^i6Dl;ImX{-8>V~ zR%R))THA}ncN0e9l1VG7G3F4DgSOI(;AvubZN;iLN=VYC!J1(-XQ9(tn`;WC=XNy| z16lUDo_5jEELvk6pSbs|S_Bq>xjW()DVe@bS?J55>HlS_f&vMQNzCKsNPQR0l&>_Im-F~b=4+B*jq%UwvkSZ3_N zkuNw5Y#a_riPuK5ja*pXBX-_T`m(v~Z6TgA#We2TGI@UDf|PO`9%ZauLgcL^Rv9Nv z4nfQF&)^kWOOghK2L*7K46s!tyIsh8zPOuGOb4hA4Efz z0xdz53pkP>keowSL6a3v&JPxt5x>zmkihhaO|@)ol#(qpEAa4|Dm{G1CA)^s-9mDN zXm}YN$FDYmpO@|+?I@}C7FIGKE}?Jraq7?b7mK?Hc1T1eLPRS5d*sY}rr*a^u@6bz zI_Cfrfd?P_Q!KyGrQUOd9Lna*9^0R0cOJ>=Fqg6YG<6l&F~l%to{?1<`$I>LO_dEX z7>fGS$O!#gqneP)kiG?L6<)M`uUKX)hR5cSl33UUeNe+kAaXhVNjr#5%+n-Nj6BG; zoSC+{UWj|)6AIW5Yd7H!DIMQUQwA|{zD%Q2U2JeU3`hH1$< z1<8!8Yaf!5y2|opTW5V}=L|BKcS7W?X@h|?N>Y;;@+i{6#zCEJ>*x%vPTmDTnG|VD zjQUNodAt!v$2pxbKi*_4e9vobT;NzLK3ycKQZp||DP}yj2;V|9Y2T-%T-~T8o7^!5 zWUCP5sIvTI8Jme>Snk1TLeh0X=+b=K99(nXr ztmlu*^W@;F9Uu(8p2W0bC4ar>l8F!>z{wjiH|gz6gj9KaXJ1l%A4*N_D^$Dtvo_diNdf?-pce=vAYT8P z35qj>eCqRx%1R&-t8l=GV1Ix*9MnROM7wZ~jsMb@0-*ossBDb+GK*qX#KDV#7VlQ@ z-vGcb8Q3w(_#A&3Qkd-J1ZWSnBy3%-4x`5PolRaU-B<&a2#R&f~EJqnBZ@RO~)zWxg!^pF#a)I{-q`Z z{ft)n_N0JzjtVD|l7tT!ri&W-*pUrJ|YUFYq^>n zYYoPAQTp-sk%GX4VfVHOcwuO{q)4tA0aXyrI$p99wlGd_3>eu8(C=n#odFS$=#(>o z%t3+1gjNZ+AToT)JJ~0sIgoIlx}14O+%SKthyEAYeNlDBZ?N4QUnGjR7S4 z@?>QJsK(@}6+pB0CClbra4RUoamHX=BgieNA_-E0?72+ebPKEDEckP%t8)}SoNWt; z06}q9qi#b*i1Ufe?8euPUxjrxpABum?3liReh@G_S6!IdOBgpKp5{YZB4p9~X%~ni z_%HkE?`Il=ydLPkk7g38MQD29&@VR(X(7 zgX{Blb`#nd<;_0Ah1E)Y%h=R(K=u(0&U^u+?A3g;FB1tp(-x4XmxBLwX&{P#K;1Gn zcKJ*uQugvg^T~jdVp7rsUj`#cQ{yJ}0m7t?6`3P(;E^%(7FPo~cOOx}gwuTdi@#Q4 zW%AG*XwC5)>;Dq@fGDOHKoXGtXZm?H`ae0FXv?Y^+JyspG~DX9B01YJ3%+Mc!j#y2 zoyo>6pB2duvu86Qz{!ts3XieF6I%!uat3);a<;(6U0Z9?*l*iPA5$(mB)-{9`tg0R~X8t0c3aE)p^q7ay)hV@c85H(`hWYM1}zk1;sdcR(&%CytRsJt82=m2?D+I?lv|d+EwRu=1IB>;MscoZNQ8 z-~)UpifJn_+i#MMvf%*3_)B@UO^J&D1O5<@StTW!9h9n`SL=v_p3Xa1&hcf=|5Ac9 z3*3??>9=7@jWGu-?)f?DN{X+3?%E?3xs%QIYH(M)X*<<5g8tD(pJZ%?BZ`3D42T`0 zz}MPYr+;i8!j@N}tR#@QR_xw6wkvAgG*i(aFilzsx6Q%S%m1E;7XH#+hdwx-d zwI!T$B%$qmX}rQB1ez7#<(nrh3*aK$y2KZNm^hr^6o>ObCnydURIzVq)h)Tz=IIX$ z5jbub|9kToIdw73T0buL!UxDDe}O!LbyfmRQ~uEkpoyIuR^hJ+(%&pT6dRf(`+ffJ zrEF)E-+p!=Ecx}c{dH~?Wgbk(FBhqABIBxSCFpyBOgc&NBPT{gF%T=`b>7Zz1Uc}h z67Aox>v@5N+B~}I6(O1W-dIl%saH)jyccPNDS+R=J4vQ6i(%&`6=Fil^+G?6@7Opf zrZtR)t9XCdD&K#_*n>8ANho|O&gSC?7K8sI$j3m=kSuOetkiFOEku9#pPZF5t9>5S zKv*ytVBvt3&Sz=uV4>C=>=7V-4#0T_UJ-%n)C$|aLY;^qjdn3_xr^Hb=*<6wECl{` z##g(gm~Ju&5ERLuhhDvbm5H=?lx$?BFr7@~GrP{%?c;j*vGO+|ga^d9Nahd`ynA!m zgP(c}^=OPoMjwmn<7EId6Y2IYS9jjpnorOa@L0wxZ(pJ9Ys z%#5h7gQE|eUx3%!;!)7DbQrLDe)jK720UQ>3TO@o8ek(pd;H?tL|`Z^{B)e%)F7eB zEvLZqWtc?4>pe$P#@B1Grd~lIJJkldx@Kp>#M2n|Wv1(si4yRK*7XHW>!`QgE|owD z4rI?Hw(n#vRQK1&A0@-!pjRjQAJZtzj0AxR06Cc-GYH<$re1FymB-v6_C z8|+(m>$kwjN%l_PUdPk^^|#|%uU#Z*v zt2wq82Dd>d1ef0?{0^tKRfdlzl1^_|8T zUnjV6cgM3>;|(ln4L&<04`XV`-Tx;A95Pze+#atAo2-R2uhLtDC~Md2v1h=FICeC{ zYw!#~uW;{HOmV=c4hDJJU$-mCV_dEyiF!1nKM^u;advd*7e0%-5};-mqk1S0N&fkh z)VvD+mVRY8(`&@r+MLQKr1!=R!kfdTkcOLH@q;X~t<%1!6t~Ufqs?TGSylhS+~sL+ zlz7X9UmFM^Jac~pVN@I2v3bZaWI&W?fT@|W?`a7*9T*(@^DDPGIMaCTXS2)ptoy8b zX4>$rKVC4JQVAOi?rdJu5oy&&NOsr3s|gXcqcn~2qhX@E3g-Kn1sB}O?ye|>)^Sr% zFU#R#f1dq0@)GoQRl+ zuOmLAYyB+)@@=U+sKmDk@aOCy5B*!m3koTZnhIx8H0z^#YZ zlN={?;bo8f+n5jzPqt=ux0Xo8V`U{nKu0A!^d3loN)^g>e zRT}12pyS$#Sgj2?GuBC3nnk43D4k3!p)ehfHBPP`erWODP@mlP(QyRq_jF~$SR6L* z#!=dyg18fgCmVx47e&Qy#P7Nx)A>aZ4zT-GE%&Q?7Ir>JF-P?*G-QA`4*vY_uPYCJ zQWF06Yw*{!Fe+K9?f>&pJ0JL-oy)cVuP2e{7a9j0D@q+TFJE>}%_H6Xoa-jtHU7ER zEiCAh8zApl$g_HHxXxN$lJ{@cFUU}qV^SIoy&pm-$rhZZ3Q?a?5N3EKrgBHrF;2gx zO%$me|5(()M=hO1>;%Kot<4y(O|H{KS#RB_k9JKYTmgt zpY0sr`QPwdR zJGJFRlGt}b+zN+$9Zm8 zFyTO|k6LF%91Ox>LCP+)o)>G$`gY?R8&~@I?wRL%e&{8sPE8`FYK~2_3O?x)h=6j4zc4+1 zIewhP$*4gxY|H>;V#(XxAQA@01m(Z^S>=hYlC||D!_q#>k2N(<*(BA}r5$895-~Kt zW!`j1{c;IEIDi6d%%aV9tH^mzUXpS=s4dCN%h1peC^Egh+Q-y0{T}6%gW7H9LwHtn z&6@$(XdUim4@&+Hm$#!x1w z6H@@TC&oUjsr}aBNv-h2O*>!GIXQtD+pvCI478gb&HjmYxU|$qWO#GC%7%@mG&WeuB z*XM#@bc0uI9UQ)S`x;Xxj|oNDhY7$5ZekbjloN=TK=1~A5;nfybfAHJm*|C+#nQdW z6L0@J_iMm|Zu0t^eUns_R6>9zRWIJ%Xu*(7GOr?%YvSLFrX7eFR>>4DzT{he)6 z(*E?m5eH~JK_pw!R|+nrk3%-2+#u%p9oV)49=}P-eWjClx9FW<;`e;BG?o4E)MAFe zg%6N|#bsp-#0lRzK<pMdV~Pl%;7jBtep@=#L@_1*sS>2qqsPV~dH9+bC78} zJ~=74JK&e8N;N*$xjK?JwCb)ZZKba#`5T{IExv%;K0~SCG^oy|tOng}g(s5nI?o>{ zZ>5+m?CfrW9sbJ$@xT;VmU?dD3<81ZKWINWU|arsXa`zT0A%bds0r4p@~Zfv`(|Jn z;ot)&>C>(C3Tiu$AKrjX35x-xERe1_d5t7*OV92}FCnG910J2*gK%z87vj3U5znFD zjUSC6KA_L?`xoNJs+HaSKUx5i*^|^++XtHgEgC$M(R1#itf*65pqcDxwrrBi{DeMS z(r!%tq}J(71lLXmqbGU)g7&&weo;1Oz4|DKQszsP&m!AzgWLt45Rh7|W}WwU&nW8% zuy}E0J{&lqW0JM)W38tYK5JsoHR_p@L62+Efp}wN8l15P*+5E)>u_078>jBDS8BLG z;-07l<1McQ7b4@AKO$G%LtnD^j*9WE^PZIPZTxlq0{nASTDU5sb z4?-;`h?!a?s^fbvfqBk+T{*tbHJtP9tp9nBe|SU$Ibk5-&CYT^aTcJAD=I2F&b!)A z&jO+hq~M;TrVay8YJvyhWS1hyGg{01ksBo(2Ca>ap*UX|xZ$224R`Cw8wxT{hEAkVIQQRqmH#2C z-|QjTM!3seplK9LoCB?(d`z$Xt&P8e9jDDZ-3&WAvIjasRX@`9g0x|Y+o13L2o{os zTS8n;4Aq-k1*71;1H#&uB^HbNyuhlHNq#NS56XmpiLxt7`2_O6aUQ5{CD+W6AEOg0 z0bnuhbjhA~iTRx$>#N@!F zM6rmrHLdQ_Ema-?qhZBC%a7`3!=**@y6d8y?99J;1#JKjHxGb6m?%{;c_JCL}z`fT-La(0TdH>shh${VJ-FjyQ=JPZ5r4izdOUb^j==IR=5L52e0ww=ff7+X$v z0T_{d{LA!Q0X#7~Obueu)|ulOFv>;D44@ol-xH*1HwY{@i4~nJl&HS<637>v{?qy5 za0s=fiv)pUQ|+HCM1l5F?ZRuNvl<}A1W0p3ib@+1KDiG2>(Y+1pacT(B7kQ^i@Yri z4A2%?>lswGi%`$2a{i2oBGMP($|OarG_=}&2-IJxT0gH>{2q!oPUCSRP5vlbZqt=E z2mUpwL1$Z)i^MYe9|hyFD*YPt#W6uFX#f^W%}K=kDsWD3m|1xYMD1JU%rWZ-7a z_pu#TUd#YO(j-!q`zUw<#OTykKbwCMi^Q#pDq(A3`b^A5e5n)zf&kr+w{{_PkjGK@h}TiW#V=%pDxa0Wb^rX63&2WY4l7kZ45*DVr{x z30!5(f(xpcg80WG4*o&WzFO@nf_VV`{B?dSNy4fMwsqvcHP|p4I4b{jIE&d&9fWP< z)R4Wkd6_sSadxe@1*oLJUd17U8Oq6uUh-BAJ9wfo82XG+*W;xi!kVcx&5iY!$WWu7 zVjx7^Wu6d#fubaF(q2I3;4fV$hrStE#%Js zv+rDN-+Gi&vwpR)8Hap4MuC`Vpds~i)Of6`T{4!hvjGZmDIIr*H) zIKJS>(Nje<<5S2Jo|!43 z^c7Fx4hPuGTh5fbf+S8fDvQP83O z{7vDgT^NmgU_Og_{NA=IE~dozs8er?aLk`|)Uaf!IO1Wo%o(e4$p)S1(rkPnfh-l? z`u!K4X$*PVb`fCNVZ}b&yGb5=Mp=Hgx?5Gm4pj{9{uZi3ZJjQl@5jgF3-eR%yBuqN z!{hB$VZD_Z+h28iobK8?RgeQm(c>%YOUgZ-ZU+e+FvP`YU#TzS=*OoTVt?1v3bHG&pNpNV90>+ zm07JS#>FV^_DUQD4byMW!8|nD#T9BJm^jK@-Q)BAF>&Zl^Bsi!binoURgSMI4@+vu zSgYd+7_h&+4mJ5EXC)*l--aZ(c2r|axte9vmV{w-CIu|I^`2zU8JyqEwy*G!q7~vH zR#b{7wg|2*9c56*zQfIKu_s7{9&NyRkydL@NG!`M-}vTgyM6#9qD{@>{)ZAsdjqc~ zpD`2U^82hrwTw5FJPRZd@#-GNaN~}>fCxP?0p>|%+Ap~v%6^Fp(3RHkka`DDnnf^a zAz<;sgU#{EOSLuLEdC2nA0o)Rw0SRZ*-a*H6$wW3fg56^K>FMF?$}hDRA3Ui{lp?q zJN~PJhm|9-23_{R#LY_8to3j!?F8D)$s%`;kJ^VByl-ZYl&`95^M(k}9m9>6pC+c| zPO^L|jpTnz{dU>Vq@D;OU0E86JIRV}@0Rk^Q(ouZyAymnB6nRlM~O}c!HoXJZNpTm zgTz*+$;v@48pe1yE^3njnxqy#LB|^3^!2~KRi+OL(|;NEDUZ@EJ{pytps^fv1;3s8 zPKHKJtJnSxHcdnUTSd7r_1K3qNXGGuU!p`|ZcU81?!1X~(kfn!Cy~cUL~=uWx`b^g(!!1$WN&lpE@)qJUIb&^KjLKO98`79Xq9ZO~8?2J#YIv$y_B|I&|-G?8NC zz)Q?^u{T^t{<)S2vuYg>3XQDxzxeQ7F?HVSFc=gjDHam@3il(;jk(5G-NV`D6(85# zDPf4}QEaW`eGZg{;2zcD^<#~OBz_&PaLbt1cQ**=J$4rRIo5Uw8k|UWXl7plh7#m; zO-;Q36-IY=@Q`fnRhGFgqqyr&GLo!U>L+P+`^!d|WLAkU7FU0`WQb^hJU~8hu+Rqg z3O12_bTteP!Z_C1rKPXjW=%}cOH+B>Hio5J;X_)Al&V{Ck4Y)Lv6GB;RF0AW*)h)h z|FlMk+lF(qyQsD#E5cJ%d6^p7?7PK zkxBsy5nG`{4D#cn5R~*$ivE1g!MQvi=IGy0Hr|gZ@B0>Eb029*esfzT_7eiCGerVQ z3Nju<(}0C*cPULe!+vUf=)g<*d_3xc7RPsF6E3M2BYp$dQ(g1YV_dX=jK$)#b-nG* z5eNY|Uy)@o=NX$9cca%@<-hr3wcRXMpyzxMue*}MHc zRh#-cP=?B%|7E$%bV+(U$hlHFvie_}Q&Kk*3Y@18%H^u|=zE@&wj3G&V0htseQgP8 z34{(+WP0%))U*A)BxBp6ISK4Ha0M=avDU403<4sOTGt``U2RuKSInU7i3(bfmNX0`0Ywwx1gW!<=6(+v0GfwsbG~Jb|Ca(_NW0c$1 zA6wVKnV+Yg+R1V;#6stt4S%75HDkiPo8vS8t1W;@(kGhl*HEDEmyH+@)vFx|wBRyV z#*Ew|h91p~xbu&YJolVcwri?*&UC)9R#etKd(7ZZpxn2UyJ%VD;O|JD zJ-0G)`E?f)xxUiSyw}_6UPe?KFt7p|#=AHf^Uo0tS%-C)vGFDm1tAWc0CzSw2hw`{ z9`f3f2YBJ58E^iy5A!#K^)>pBvwI6nKz)c?p&t2`$@Rfv$ zC-ze40Ys=*(9?sc{IEHcJG6g&_Vl!|fD8RHb&xo)nPl+%nxNPUURJ4302$*L8r^?$ z6ri;=Cz!_yS{Y;5`zSXYO9y@iz;(fLjeS$nJ*67q(j z4~@A5QxTlEka$aac013Bg77O<;f&Cb`RKa!~RJUJ^wZPui~99;UhzYwI~-+Ja$Qk=Y%*gtEf^(!l8=c@r-OfH z4!xxJM6m^>>H!l8;ju4fo?FO{d@|%BqrV1gAA&Ed1JOoM2c8zYN+7WIC>0f`uaYIO zEv^`pVSd$EyI+cHPCmkP$~!Y6Ei}=H{f;dpQGjmXz82mCAkt+^_Q=&O%_I#Bi&$OhX1f zf}SxqVPagu1{&?~n`)N*GT_r=lhbSc0~)$BZ~2b4vbdvHT(1?PD{I*5sK~EP@HejM z?>TQ3V0P8)njkEDTaJ)Ehn-5*+Qg;H!Og9yyvB#ph-gDmeNJ!mSYOditUgcEXZj@@ zE~xAMb_dx#qeBUns2+gH>V9JgUg^F zW+84?94S!EW&Vi6+G0DJKHi_sc@OyX3s6!C&ol0+QTI*o%d_SUxJo`>(`Qo9qHXHD zWJ3zFf0f>8Af4o#baK3#Ut_P-U;}8jTSS2=1wf>=vi$9;Q$QFNw`-C|d4_+cG1;0G zEhIHxe`!yib|S!|Y9>x~N>{YwwC|#tKQEi+6?iGCK$$4PH_!bn{io48I;Mv3XA* z8U^-WEo!>d@uCu4exkLfRxFyziO)Nxf^&p4L(qa|fb0Qiz6~MX>4o1aqAHWM=}b=N zRZ+%njARQCPxmg2B?Jt=f`N|7MAh}`I&ViEF zZLhr=Sx!Eozs(W>KQn5Va1l0v6Z@g|0RgT+x9Da}*E*hIZ<&~qzlt^;k-2c6`$t&b zk~>mhi+WfcPazU&neMd4lUmX^_{PpbU10OTi#9?LU{SG9Pr@Gq#vv5ZfdofFrkvbJ zdDgS4s%(#?=E&UHHM3+}2PPMZduoy5(9aGpjo(-H1k@Y^OdGcGx`hD-;T=D#_fJ&S zo%7Tq>d*VSC3<$Nj4ZXDahK>-LJTG@k2P8zudVE3V_LAjb*DBY}=(CKNO#e&-%=Z$(Qq&w<2(=Oxy##A6{PwDsOI+IDlMlY?IHTel8Rg0w|TE0n+7~a^L zDvY`-BTBOVwU3NQxQ}+@Ev^ML5OUa@{H5?M*Yyi6ae_e@%dbHHJ7hMgMG?+eSu%(6 zl2Q|+)W3Vlr-Z0O{YHBC2oA|@>vUh(@}fKqX{z?{w|k%a$x>u|8@1%Pt5u^Ei?~U_ zag*;}^ZDbN1nIgM_{xz9C^>>Yiga68lkoz(kjA<15UFewD;06*)k=eyIH)kBn3H?C zB!wRSC`t*SZD*y*g8{q~RP|H+8IeHfGjpSK$lQ16I|#I9Mg8iq^1b%p&C>o9tbV-` z8J#Pr&{qqEbUo%zZ3ETY+9-(md;d773uGK*<-_Fg=lTjZ$ z?J410>c?-fpRqlDo?p3GwlI^0vKKD&t$!HzNW%1j23{5!fI`{tAJBMWeuspUkIHL=k);7hSkz5B7OYi}ve2=D&t7Olr zb|FOcKJ?5bkjcxhTxr~moJ5Y?AhsI=>c>zb>Hc0i6ylvAAm8}V@sR1nPzvGfc6XYZdUDuuP@e( zK)VK5neC{{Az8sg)=2Bwy?me}lREvEEd(L$S|(Zc192v}j;r~wrtE3Z6NFtH^9GnJ z=?tbR0VKYkH#ymwU{9*&rR=HJdEM3FY*iUx0>&~jAVS`s+3$u#?29l1*3ah z08Sy@-4JF$2OnG2(&{~XVtU+318R*KmU69SiMQ7Ou@m6>b&Xz!wrrYEbmEk~x4;)J z8`c#}`IAwAOC5s{zArM}>jSL%nRKn#s?pz+Vv4H#RnN~kh*1a3) z-XIfv&f0I;+A3amtS=XRJ5ytX8YeaUFOrW-BTX1qeuL=Ve{qeY1v+{)UweQ zA+5k}4|BKd=kWgmQ%?08yR`KbOP;3r>?%$9UtMfM+;~V=$#88h@TxL>vG^)r=a4#Z z(H?P4b4NX0CIU#%+bT@Lo~4*Aw{eq#%Ph6q%Vu43%ZzDVArt-rYI=1#h8wJvexsoHoeZ` z6~CnnxbYUAvz$hI-U2R|J2&iH*TSB~W6L*d;se^HaF_%bI$H)L4&ayu7x%k*UUy`K}X8mOcQ^bR^ zsCr5Z$luNySs@`po5QK#DVIB7A3mNVmL+K?3eFGG)ofMFzID0Lc2Q>)fNZtkpG>y`xyVGLeufck#XsbIf(Mm_FxpInd^bc7jybs$fC= zE3=X^eJT!Xe=l9-zN5_QvTD{vos;|=W#P-xzOUia~cxK z^?Fd^*N{Q3uB|5BXB)xNPqx~tpj6T~`OW?_AR+*4n*gWMQ!#m6n2o3)eDQZxO1piU zqQd9mB%nL-puM>&+dcIO`p(|l%i@|BF1Dl7Ua4<-)P84LOKMVNy)r?}SA}&B-uo>U zht=VE!nt{yP!tiJTlqSl+lu&_PrSC^ox43d;p-6e1v5Nw z=nUtbi8$hNU&vkiG7v(%0jp?eu$^SiPhfjPWXr!hRZ3H)!jIx8|CN(sfLxD4R|#E& zee@+}-9Y(4m~9_!F=_Kn97S(PEf|@3YpuX5J3=|fopzmWD-1Ld**+Fx4yYM{TxF0> zx|VQ^cjq9yGs|f197s>g$Z;`86Gc6k(6cwG=0&o56~a{$MQ)fSK4j}xZ>3n~AX#s8 z%t<1q6qYN!pYbuzHk3QYgD~`2R?_BEHI%WiYPw>ar;DUs34}EDBM8_PiY4|`(`Vzk0UM)u3<`;F6eT_t8|GMf zRMn67x_vFx>E_nZw_@9b)=Lod!lg8~sPTsV)-AJNT_aVV;MUPsBsqvt>BQ5&(#ZR9 zqpl0N*JAj-iiEZR;)!3h1%+1W7`+UswU+#pNf^p4I-27d`C0q+_1rtR33n@ZO|%AM zwfCULc64Hu>3GG`h2Bn%vSt{l1QRz0#Zznkq)EK;+T_5bjdl?t`>Si?3ENmrWgo&; zo%XD|m$`(dfL>FcXbJG75+ zbl%fHIC=K2Z0&arFaPE&reMUC8JV6CSXh}-G*v(LM%t;(1+!(8-4U}KtiuqD9y?%871HGBghiUeP zspuvdjA40QFz>k>OfG%yyUY<^vw!XAkyB$2?Q2O$p3@5i!PtNd1}dSgq}m@@9~T=^ zpzV)~mv4}!C$hV)i?N<~)G_kPTF8dun_+7nIxLUr7Ec3+xFCug>^SD=ZdkH~RT5xFdwHEbtuN(blK4 zP0_!#Pauhg+vqs*a#%Dhd_%UZ#JC}Jfp%es)?DlgkGED!M^ZNvzHu|(ZJj15ZRzkS zobHG*s?ledll^N7;Xq@eD(9=1ABQp~DierG``kSiPN zqNd=?OMJJ_B|Q9Ligs8VogH}PAjRGSOkH_m?>g+7$m{bX#V4z41ecublD z7cVZJc0`TZ>f|M_%<~RcIH!KCRNVl1UMQa4%6_q+(Fsn#z1UhMYYtBQxV~~ncu#Q>Af+kaG%1Sx{pp^YRmTR$FZP zs-2KsYl`yG>OHxx6!U$n&^c8=rU~H~@3J>faX8U~miPRjHfFHsopw1S>g-S>DF-_v zqA|&9er2DdOE=eL6Sy2+wtoDYJ(rX)1lD%Hm0-s$Q#+D>!g~w*_1ISgE(3A8p z>4PPzYi~hv*3B(}FGa81ranKTG{C|7-ZLILmyr*@DpC}0j^~T~t-|hB(v%yyze{9! zc#}ymK6Q+sisA%jJ{^aRdCgOhY!&zHA%}y~LJ^>!2hL|(#vT(aeI)7{V8aZwl=0s5 z%gDF`&gyu4IjwP><^rAbW-|B&ACwJH*2@xNzcY2-Qh#=d$4-JRm-jah-cQa( zVgkO~XivZj2MvOYHkkqHrY>rH$*>IQ#Tv|$6jW8`usDVm`)UB-sm=)ZMv&Bk#rH4DDq<|q*45WND^H2sVg zsV5z;%}&jKX1C*%|590LmSViU%PC+%N>Y89#7DZX@b=_VpD zzc+bn3I>K$Osy=lk+6G#kZ^#Y>|~l}G4dq|z-7q28@aO3t21bH{8zlH8J?GN2Y4`9aPqF#7?f1G+nO)~TBRC|qjlES z3cuz?DzdRFUATVP8vt!v+6tf=yAp-0`|Z`KCm^38`CRg>Tys|X!9YIRe6Y;{ z*%QDA=6b)eML^{%`_o`E*aKD5xZK;+MJiDuKe$Ql9UKz`% zU_&>(6ORPj%Dq0&KyLL3?-AP*}M@R9MXo1r8vGx_EWFJ zjwq-j+-VFxBmZKRgXh-gsE>J9lmCD~at@dJkq4|?!%QF1w`;O~Ynev@vey4R>govy ze`&=1dfu{8@gdo!t-@~FXDnVD7>C*}>eD64s>YG^m;P#lcI4qyV_x9Rk9Yuj z_;A!T8_{--Y!i0qAN(@5=TwW&dUA==nBry#aQZKK{rEfR#_&rmh7d>iQ=<*ZJ&?b^YVn~*awZU$ z-!DAGUuUu^kN_q%_@CzahvGQE-;;_{o)=k2F`!2T?|&YbgX1y}uM`T0o~XtUM#-FV z1LGa)J{iY@{z$hWhgOCd_KyE1`hsx+a(N?98eb%>xdCc7;%peURL6erouzDdS8Gy@+pun=RMZ#~=Wx;l})A_f>4BOtwd+} z6SLYDaN3r~oF6YlEqr^)9Ecv~a?EoA;_?5eMV!F5!%X1&P$oF6(YP+O$8h#k57=^b zNI#h9$!oIJ)B59y0$V6aZ4R0 z^2bWenL0PA{%vnC@^~J!r4={Y@edk%P7!;IQBu1kexR#n$kRv{Ik7`<`8Mn1ss9Zm zW)?W@y38car6I(}HOW%I`V3~}MW3{&>m3ptj3pM-J|GUyD%UY^7X}d5HcnmYzp{0n zy1Rt;qXH0Gt>sm}4Hj1O(z|gUUK|a{J8iD@Uc{gn1kL)6i2hvxoHZ`jrxBo-*Lvtk zN}j5xhy{AIV2zK#01RWx#9|K%E^39UDuR`pXbvWk5{|W9@DI};2JB2MBv@di?g6j| zbH#UR4Yj_+`T|7KXO4iSou=(rA=}x}_Yo8Z{~Ek{ytD9Of9R#Pp9e8vX;*HIJ2$uo zz?$o<88f+Q1~V|ku6+s878LscgqxAP$;OSHwMA|(V2BS=-8&0CK+)=REkF&$RL^Kx zjXv&nI#)FEn1*tmF4P`mAw|JW%zw4i_VenP<*JL!nNFIYU)&|L_*RbnVw9%|5oqvn zLob?lBAkS)j1}&u@@znu^yB}|UNrfK;38_l*OddtSdAhp!GzfNf4ThTxwdmfTW*+h z2UR+P*^MG{Ys;d33!*ry?@a~<2HfKvhS^FW4gFD9$59L9P9F$B$ST+$S=Lf^r(;%j z!NiP*{6kgKN%l9P%ix~uD7!i>vv6}8884FXCVF)O9&G|Js3eI+){IT&_Q7fs@Nw!l z8?jdFx#$7pw@ThFprhXbBk8bs#e3I<^AaI+d~N5hKsS3%0v60SY%cz7=phJ-;-1;h z6*UJirzfA?K(b}hQ42uYercTl)Sx}vq=+msV~{>uuR0!>Jy-h=h-F0H&gPWVqMqNM z1+@PMtl?}ZZZI1-o|Ox__i-0I?dylV6SY-t2-7@J!;*p-c~9{qEoyD5YYDhZ{twQ+ zI;!da|Ca`71tkZlC>_!rDk+UfmoP#)MyDt!N+=Tqq@`grNP{TI1`O#2B_@ne(vkNy z@9+Elk%GK4LO z_RBVLeZdpY8M}0fd*;5~yPsNAnk~7_Z$G#Tf2-#qmzCin0IQ><<=|lmj*RxbOa2kz zH%(_DWbX71S~~~OZLn(@^kvZQccSo5CXXR}+BNh-m?}6slfst04{w$Lfy%+V+=(2D z79H$SFi}@50F*4YjmAk166M}=UyAL?tFe9c%oonRTe}?qsSr#tUloND&IKu`XCNOW z=>pYMAebZYOY-+DXS7Sou2p{bCQ~a&HT+=c%v)f{Ewbe0;UPJ3D~%9p?&P!_E8DI zC7^LPcM6n=zY)AssoTy232(Bs9Vg^7QZNVk&m!sbj+bN4upjTKNcrLy0d7_>4rADyH zK4V)ktRpkR+GdEy0<8{n5uH2T=FX`g4xo6br}PSjm0NzeQY7cAoK^*0z(Hu@8A3SUTZ~w4#e*b0I=j6u zi9XXyQc)&yhNengGe#TDMvv=89!`*oP-*uAjbVr-BdKuvzR^)_9q_VfJzj(vXg?)9 zZA}zJ4aMIFyCH(XqpfFI8&X->ylmanv>YIV`K&DAAZ>$Y=di9q{we6B5e2~h6tGcOj z^}k?5>XrYyhx7lAOeC*OenyL%$TR$|+}irw)QV{z-|CRHIqx`~f3DK3qw9ri^_rNx zQx$QP;d^ez5hxwID+%3*5J9OJTL-67Z4-6R>qp_XOg+8q&K1dD^3R(+LIw*)F3`4V zV2?duV3K3AHwvV)gMMvyT9A76>a!4&Xry!M?g5l#k#<{g)wdOnEp(5>ZaXN~rhr72 z{!r8+=Y6xtQ_S||NNQH9xTj#gU6SnV?=3P$*(0a85-+*ej^t>Zx^u{IlU?fhMg-hF)>P zQ7i59i5VGD&vPcNX+y2IA4d{RF01(}Q9@;@dwN^^A$1bEgFhm`(!o=_dHBj%HkyGz^WbKs6Za`?i|?`@wyX8 zp;oMJAGH;z^#EVFW-G~uSr$V1>(tQdOw1m!;(#q~hO2~57;-<=iK|OH;fCz@#7K=d z9~NdEbqKI+AoC3%#}4cp&yZUZ@VBZIY)5yQ4{AEB)SPg0i&5qdSk}i8`f9LFIwCA{ z-MtT!>aV`&mRBTe_vhJ4NLaiwHtbO6W4q-LJe6LjaEdYa1hw4Ak%xNw)I>f3*1<6Q zI5;%D27(xr`Q>5DC?=E7UYBC4-t5%gdd%KeBir9V8!)jc0$$*P-kW2;ZyGCh7e}xn zYf_g3HH)l|Ys=P~&Cfd{T~R3jUK<58TbBGxLw+*x67MYhIWTCQi%@VANOL3&_J6jR z)NsQMlywyGDXMH2TJ3UnR^6oD>P~S#sufcH28RLD?CX)@((=qQC6Rc?*z)lw?+}e} zQ^IOR85J{bK`N9c3-MtDJ%Q#i93qD^sou7>eR1#w%|!Xirt8Z+K?RR`kmb$%^cvS8AlnXSyiPdR_k zmKv!4ng6uVIsxe%=z#%FOfbKZ`0xQy^KN_$aAn_#$tp@8hjO8z>TAFus_wjOmU>%i zaAbFEC51=-Cm2V2(?zi1EONxasm4rA=0x8Ngqc83-!TWk19}odyxvGgO1wYRo@a^^ z4B3DJ1iu@25O3%jJEmu?mxAS%byi&?{jeQj0|MnxfXjH@W>B0l6Wv%hy8izB-TL?e z6EOH5DuRiDEGD`JYjT+NEN{RaTt#@@D{)ZXlw08@b^pY~HLxiJV&~39>2i)T?^!n} z^y^qVmd&I)Kzo#NbbXAPgRm9(*w96&SYz**6VPQh={*V{-ij7*R^>twWa`xhKm&o0 zAPVw(l7@@+#sf+EC0_UtkNj#=$3bNf%5uS|BaDnSFF%>$8~9~-QW@`_yNfd-m{?y` zZW@jJ`y=jwDW}3zw80BpBvAY&o&QSusU{Ng=FSFURNu-08Qpa-Y3qgXN!X8Iwwe6Z ziUWc&ugH_xe~oJ9P6a2bc9RuG(R?D>T8t|oKt;VDXNCNdUvMM)tMA$X3!aAk3o>w> zX6ktzlhdGLCN^r|*Wg=_@%gXnO|z0mcT|Q~Tp%QXjz%=Eu`U}JBN|d~8J^^|!a#i< z-l%r5#Cb|wQCUe?{ouw1yQxa5;VCZXLU!S>vB@598`FZn_cZ|3>UfpyZ#-!a@{m9` zHW25l84kcQ@IxEBVu=45&NV|y4k@q7XHM&j5B$UrnfM4AAf5(vxx)pZ!53(n@wLJD zQLqbW`F5>RpFz_r`VheO4xVy~7YY9}*MaJa+d+&ZVh%>nC&Z>YfYt0zjPlq37~5`xiTB3p+3o`BxIUcUTj2R&G)d;m3!bfBbGuABTXgr#Iw& z#2Vvrlj&?wl3J@%t)1OSsp`Ayey3SqE1P}+ifG2NZ-61!Ctqt`{%uCRX z{@aykyeJoR-kl7<1TEQ7s1@K>Q?pN#yYrNEYS&Nnhq)&h3Gmcr=uSHm5T|^*1BsUl z2cy{Ep`BB`xnNAziyP^RKN(NkhW2t*g3)5>S96V<+*g7V1*+Pv{4+ZOoJ#?%JJ#;X zG5|)7w2u8-_%xzupPz9(_FGGI^qBy8z-tSBf}TAP38j&Z3T4n8YMT&q`=Ufn&$2v# zvH_t)AnZAD@wjsFGeFav+E*hvV%XT+bIc-1rz6geH{zf?d2!TR$)j?jUzV)^x`!l8TyivGI3FH~{qEi^+e3QV=?|>)nNN?eZl9X^DH!i~<6wDz_d)0Ast3 z0nyz6-dn{sW4rL+WX^Xu~`C`#jwKxMm= zKVJN<>?kVCXayWc^Ed2iDtH*!bA_E`LYkrYMiRl#A#+va-25K2+AZeX~vq2sh z!@L5hJ^|_k>y{aS?>D5TMQ9Yf(Q2Zs{Cd6MUL{;?9csQelAfi~3W>eKSL`lL<4fRU z^VicFsJxnHc`2t&%K4I1xnfukBf)5^%JqQPOy8usQ_I*~S(m(haVHJ&0b~NLGz(&= zh|E9d!|zvH&)USUdXv8xur9I>F@f6J?WQc+Vh4bohcgRT;mcCh6;0A4%WH} ziAd^x@Uf^EGc^*g-g9dm*s$7GW~y%#*(AE5Z1}KKVBFKrdXsVD63nj3N$W5r z1iSIy!A98Z|BGc!-IfeJs`T-+mE586INI{tP9_-hsv#XXzeZAL={WcR{WV-~G<>z~ zxA>x_NRRYUU`%TP?I^Eu9luADQbVJq?}TDpu#j?S>p(eGS8DY7n6*aZ-oVEC0K2!g1In}E2hG@>4W)0qDIBpx2}_g=E5g7V?DLi62yd+UkUz1 zZdQEnwyjiIxV#kZFJtzI zvLrLW_#E>iH7S%N<@OIJh%c3x=RKs}vz+LrPQ;Rwd!+7)Qoc zHQy;#Si8Hl)l(G^cmO9(wvwHq`fG)JZzSb1e4qYXW8Q#>2j}n{ZM{&1ep49Yby}X) z-i1U1&*jO&zWVq2sUGrr7PzA=xwWW<=F`j&lem*M!IV+2(nwv`tIwyxDh~z@-{>A~ zl5I3wUTbtPwr-jHEoU0Hv5V?Rn5^dD^8=2J*TA0#vd0>y!;rL(C;$=OTog#szaVqef~BmLqA*{TuWnCO4r?bv zI{Z2q=EDBe%?6HjpT~+7HPGAzi<8I71|PJ_wOZTUf3HVXc&F+q4piomeA=}lGj)9u zBM&v>mnOpOuv^=Mjk3uFFAS zpMUUELdaC8b;`%fU@KMy74Qs0BGwzM>-ou-o#_Yk__C?C6+Kr{^$6utd|6L(;y-r= zL*DgQ<`l_g$7SjtO}l%^UDb)fEwy|0T_zD@URfUjM0U%^J#!w76VlZi6kp!tDgeAn z3u7bo4B=<++{icIl~E0k2Cn}I)D6jRV767Sas7lTEN-};_Uj7}>ZR+bYJkycggG~rbPTTWs2l23BIA)ZfRL$$w;cL!GTd5eK>sxDa5 z5#m2W&>Saq1C1q*u9qFNL>AA+n*0tabn;+7_&6l&DkLS1zBT-1+O5$U5{3w6^;4XO?uU=vPkOYh$&N z?W8YhqBv2>e&wLclMWn>8x=cNE{;G$3(#>FiO=}mI^bX4r`#yFBOd*pYmhDLSM;cc zaQ6DHlFDub7&w6DD-{7~Qm1E-zANc~UczJdM6zUak^q?IJ<%i4!)I0!SALbvUkZ?e zfG-rhB4f7|srQ5YV+4Th`2CU0?Q4>rN=$?~;Fgo&R5dA1e76#9WAk$@hD=)P$_H6L zG9ZpEQsPMgJl5gey%tLl!V4(beg|vk=A7yXQ&Uqs8tCW_W|J(v;(%cKn^-)zH( z@+)m}zrKME7l-<$Q7zDH52kof#jfIbGoD+{oKelAFBG`*z(4|YPxO#0c3w^{U|YPl zR@Be1rwOt4=aFGesBUR_awBi-`b^$!a63IdQ=;WmQegOzgJHk_=DxN@!R7i?bJB=T zCStZdk3o?O5i*#3C=Tm?aI^?6U z)V>FktYp+J4^hZksB!K;Cy}zVC4ZDWV|W%Xhc|2)tsH0Oq})> ztftg+hzCeare;Ct5%>>S4jY1Zn;E8TMI}`+j0XE*b->o#T#ye z*#Zm{g+chijH`Nww%Yt@uF}jfiN;j0@SBf-?grlpUR%A>$%)^H)013zj)>1_(KF1( zfLO7h+1Z7E5u-)M6h!6`^=xJro-@JO4uDO?_ZL>z)cBk>a$C>D+H$Cma7s)d zw~rRMrGShb49!5s#p+y+KVJAEoxgJ!rMT}dj{ZW4w~Q=JJI2RN+J4Y@bwL8>?5V3P zhgwbm*9KB)qIhz`u0&DuPW>S|Ff0NgXKsjQGE2%H+3urJhs=W_ytfH)x1OkN) zkj(TG^cQ{ZPvHU6Dkz5m25nJhYH$FPPrxPfkGRHvGH?ax~g$ zIpUHOH2uCBmdIMW-?zaFm+IC1%4Yut5PMIJVS*G8km_VS6NUyok-6~wH?Oz-ydTBV zZh-isXi!Q|30wmM_XkqYf+%l1uM1qIu|^|a)g2rA`fisVp6y(mb^cj6>R;4Fz%wtH=#)>ZFD;Ev-%R6;$6q%3;Ty&T?!)}3 zvKWR|Rlb@x_`|wzwjG#B_`rCYU|pU7Ut=T35qat}8c>-kSYyd;fwGf}zk#=+e%dH2 z*Xbx8b*&WMWJJ1WWAFe>EdC6a@%j0T&o6rt?f*RQZyx^Z0DUm7r)3}*XXfkvQ?4#x zFL*}I5kvrBHL*_Vh8~>y8O*7yvRg^j9!lya6!R4Sk4cJ$X6sKL{F}Of&PF{-1h6cp zcnqWCz_PO;tT{R>WukHP0(6W%T|8-D6tICt_i`a976{G@uB6KBJkLWRQqNat& zJ;OUiNS(53fP*OQadtKR}@WcUk$|0Uv!VbudK1 zYpYTT0=)JY{Y?A7iuTUb<|X(p+;7)rLjw|E5d4MZ zA|WPK0d`-g+Fpu7yhH+j7XEI-D59rs4po0LY!d_Xtu2Td^S(Gc`o}Pf%RM)75qfjV z=GAr?P^{DnR-w~)=Wzl)It=Ka{#BH5p~S>_1q8c?)1Q>Gwr&j|w9#WMKF;)|=zB$$l-8z@Huab{|8Z*$C;6mbz!O~PLa zc>Otn3cGWyOE^Kr&_#LVK^0K)NR4{Jf>#y6>if6Q_V^ai`dmjtN$<@oMzvP~khFsr zCB{Hub!}=Je^Umj-@Z4M_h!6!4Em;qEM^J170#wF)t9T%{}Y>NDD7VUAmc+66@ICW zvJrAdt1|OOJcQ}W!8Lo%r7Ou?)o!2UAi;T!h@X5V;G`zlJYJA#6nQ-q6WccaB%L;x z1EKmcMHcPC2c4Ygs|n9bgwW?1wdA0$@~0XFG(KDW?h~+wUCUWtZMHuCg2l1rG^Dgi z{$S0LySEY{Tnte92P$8kvvwMlj-KWxt8!!81<6rFu=q%(=hhQO7-qnXC_u8n%^w(> zcwvX~%8e^;QC7f7m_S&Szg%TPt`L#jjyno? z8eYR1t+3{7RNn-JZsx8oTGgjcKa+T8{mzP~ing-Sfplus7|3rtE&M@%=EN!X0=D<7ZXcTM~J8&@mgt4|-P=?o~}W-h2f=70Gs z{wVni=@nfADhCPrZW_|pZ2a`_Tf!w~-4r(4yV)K?1N%EkabAHr&7QM2< zSm=VR_~^X8GI~`1Xy4zB#jtWzH#HMxohF?ole%826VvCy+0iz#qyC2a3ar(-zzb$@ zyWS(f>~@OQ`oO_0;O7qcm8Y)n8mH?UpHBg*2t<#Uk>!*~2dNY_&HBa|rw+G6axm!_ zjIm9i)K|ajQdM^cWF^@ORTyvZT#~WZT#~zyd(K+4Y-1_?^!_gb0z(x!bN*|nUq!l@R*Ux>S&)QG&Mr>I zy1bt|rw?TBRC}Wuu36z)efgK=+@!?dnMQj_R?{+n`Vyv9^W+VrK!5?dEeA3 zz8uS<52g`e0=zCBX~KS}tyI68cJqWrMf{Ke`vnT@TxAR~1o^S;SPf$3+Fw~!TM_nY zw@QBPvxy<{*syVtzg9=e*4KmOck$}oX5I$OZl{u!xzTN%z{|Kznl@={va9s^Z8QwF zSZSQ2GrvzDa8a~RD`92c(s!78JFB$Cc~tU**t80H(0&;QgS~xlv$ZdLIEQ#7z_|F; zV~5mzDkT;r5tEefN6(!V>CVS=!jufXSVRBm2Rb*@{IfakW?=+be!EB_J=9EC2eg zbRu70M}%TC*4Km+Nm`9nm-cuJ%k6?1prB+Ku-WlDO&nP71H;m<@);EcWo=U7Rwt_( zZZ!@?J0PVj=Na!?-3pM>^4Z{!z>XtGSJn%AV5ptS9%CsHKF3L;kx{~QQfHl8B zKp?+Xnm3a`-tc!iQ|B;F8%vv~p_kL)Shve(Go0|WH4-V$Qk@M6X!nqc`L=G3+HC%r zs<{^do0~*<6fR_nBFhjd?2KsGVjSg=y1B&=EosCU;5cZ~C?=uTbh+^Dr+D*XUR5qd ze;d%>bJ@>DYM4Q6Ixg8nTNJ+JiyWCSi#>$T}|$V{LaDNg%q5< zQG4n)!GJCZmn@VADqLvo%MWbSBayhJ1x9NA0!^vUKO3+wCs+X~2>*J(06Q*)Z|<|j ziY4D$BzvI~jvLsU+hkK+tkw)6XGI|3NhUwCnMlOG^$s>6J6pYPir6iXORi6y+eu{t zI^QvUvvzA{L_N2`gMlwJ;z%{r80m>I z$4qi%ku*9SS`C?f3uD=>M%8A?ES=Y z|5G;$dbt7NbEn!dxo7zl#l!aSa5ubUU#HJQG~Kz-oxY(-bWpHvl*tSN^2RqrBF0zC;t>PY{;n_V7KBuLM-1YXOn|Gl|NA0ky}M0UQ#wR!__L zOT09*9yZa0E(Y;_61pHk&>M`Lk3mOakxi6S`$8?0Dqw2{>Yp;{d9C^IB%C_#9#9U* zDi%Nx(ja}B0c@fF+7-=nGaPg4Eeggf<`C?7{n~S?V*ySgKAvS(PJ0*fn_$QU8}-i0 zK*eer{n4>6&n%lf=j+8tXA-;0@~(^jmfs(~=QAXWSqhB61R{%xs%aAjLoM3zC1 z?#xY<;*sI^)=-+K1g(KPyo2){&6XgCZgWMH-a2{+bLHPj?V z_dN*~zG)ns4I+;+H0rhaz>NRF`|~9~+IiRr;%zGqMBC$MKW{LCgG?OIShZXKC}!A# zXi*d%{1xbU;%|Ha*Hvb$haO{y@AAnm7{xxjzlp;A06cfII@4 z(CpvOc$V-lCMIy7vI-Xx&(ib3PuGd^+fhb`hRUlu59kO$5N{PhwLYg+jzOR`JqhU8 zwTW^Dfyzb841GS!4u3)X>CcxuKS8vWmuTjojz(OTDJLkJ)-}fGi4{3~A#jpmPt~_& z*oP<|e@{AOZsR+d6$6n1$Ji?Atj81Au^p7)^p2SIXkv*RTt?{2B?E31MCFR(Q^kM~ zq(EF3*opvD40O7kQV{4ySrS0kiufYVu(r!RAV%r~&BXeChm%x_YJNeI5-WZ@!uMaf z2xYzZY--&OF08z&7X3iD`3rmCY$bobX0|qm-@u(95UKkDY)rrWeF;=fO5vS*K z7vctN&%ONDK@hOBV68MrPcotfdf9^ZG8f&$c1A{wD)lXB_Jg(g6gCW##E3zctVH9^ z2~}rd!^67{`8~yTKu1UQPv2fB5?&hG5?#;~74A+skkr+>sD+aj143%NSsIV+toMr( z*blR*?ebD5_nPxP2NL!xJ0X*;zkyo7A5kMjDSv>!RpDAZK|>tV*HcR3fwWBJY~?Jy zKhx-lYqmfIF!=2mTiaiv{@ctF&gpHAyxiO?)`b~tl*$`P5b3?B~?c6@*j>Nb2s_%+<1WM9Zowh4~U~B z5}yY2PM$p`+#H@4y>p5UM-_^PYtnPRJG@%Aq)z{p zjcravbHCwJQaMZEDrCgB4FAa52`jz?x9T(`JvhgMK8UtmZH2!}F%2yZopu;pNkQW0 z=S(&)z>xgL&*Mu7CcQCy+{E+;(4BN6tXp;D%}vQwxkl^2orqll)%V4(z*& z^BWY7UD)l`on8%m6vKY8vQUoECFmI_J0?oA^1YNO^OJ%B^LN9pAz*z%q;oWePVC)@mv^D<8`~9+=pirY7A=j7~|G;f}dec?$ct<-v$I$ zFOwXq@sokY2@P)$3^W?@V@9HKR7I6D(DfU3$sZfCP%LV@SBxRrd?}_#p8|I3otz?& zWs;2nnl8YrN&U{-oUS3Mpn*e!xbDsxXv<}|@)u0o^X`QHMV7!n+lfXQQ36G@+V#C) z*Ayiy%fgPe=B&`ojTWjT=X8Z@IXbt&opX;r^2gozkCoNS=|y5$gX9Nq{eHVTo8Q*o z3?l(mW*Rm<^;mz823}}RQS(9J*lF$qmbC;ajT6gf`R^m2P)LAHLKhF5I;5lkW^fIJ zv&(B=N#{2wUu9Oi43?4_sxCdY`+Akmqj33>exSfyP+*M5|sdJwldMnP}kBRGApRmE)PSugAGKm za%DsCW*>mgg%!55Q!u+9qkd;A^zNlcrn1ndf6{}#5lh{>U$*i9o-ujJ&c=Cyt*CF; zMM&hqlX->0?r{yf4&vb4<<#fV0tY?gF=B1B)l#$AgV(}BzFD)?bv1vw_ugx8p!U|7 z9tPVcX8buVBIqqn(w)3a_Dp+P&(qk(lnSfMw_YMTC}-9e1qm~kNb(G+NCA@+2uPGr zN|5W_Q9?qUlOGk2cC{s~oXy)+3lD~0!lrK#&)O88Hg-JvX_noxcTd?kZ2@pk<;p4> z`K26(x3s4T&1MoN_U@QnHGvvDiCC(mTs+GtX+;RPpY^W|n1JoW|2$BWc27wg}y6ktaz^xp^h^>I`xyWMY(-QC$AFo7qZ1eAZ%M4a0m zIZ2~eGk;heL>=uFHd*|QZOb%0uawg>19Sqmi8!=F)%LtI?DhIygw^w|vjm)X=U&_H z8E!(^^>|+BrfI29m3?!1$3#Gr3CE)0QTkMFr9)&dM zno(mzi{B&_V(D`TSFrF&!dSOo$!tBg2zw}5?{_4!R26uxx2S$$v5GC5n%#%LHEHEl zo(MN-|MUs-ovubNs12xe0>!O`FQ+@)}mq9 zZ)rYfRHBPgxLg{8^N;H-4Y>^!dS6U@Ha0Rpb&BV>J_&qr_+#Q-xh+g75P(q~nm8n8 zuct6`lCAP<;+#qK`?v)7;FIYvSqoBsIFKI5(%qZ(2)GP{dIB z(at;RMvcZl`I(YjJ*0Q@>R|8QahM32r0qf5t5M?UJL z9ei0dPm7XdZF;@B)25FTu|s`P_!D`=F5TJIBp?tGxPzuc3sGL${L6=Sw3y(oO5yQ%zc_y)nb%-V`+oBGNk|H`Hd`6o(gri15?+Ci zrdlgEy)meIY00B&5CFU+VqQ}4w_Np4mibc~PWdiq2Tn5O=Jt`3@X^xaPsh6Qe}IYI zNqm~6V=e&5R)T1^Jm0uIqJnlyWFX1R7W@4gu7MCNqg05pXY1#kVByeR7i+JbwvKfL zrX#E4_X)l4s8CwU8z%&fTg%RJS*hDry3IOl)_Mz>cd{oAlwNMpq5H+A{VUz~u+Jv~ z_qTH#L{S4o6e*GU@m>%}~Wke$ZoC>+;^mFCxp%|me*7Y})fu590l88XMV zUd1@rfWjvnSDOYd$i!xD=X6ns#|KI^rBL%)&7N4^rmA87)0enRQz<(zGxz^eiG>!w zCh+-;7z9hSR6c)_k`!#?-CHgdx9{OIlGys8b-zAr?ZSk0odu8&v-eTlWAT7#aHlWw z$KNHAuiQGP8EZ)31{0$502_1#B@meFWNkVU+U_L__B>Ls@Ivk`FBXMN*}A`GJiKo4z*}oo zL|#Gquh6}|Zpvj`2N!8|eDT34?i7WxAYET-(_%04*0E8hZ|Je=clGECtM)j#E z<4g1uKCi;y(#3V3aCXyE%XIf z<(XhOiz!kf$`{wA&1_UT%U38RHfg;3v`gwkMV(+uDSby6B6ZpwNWO`>$gB|_e<_Hj zDdQO@} zG~|0(MSP8*Em*pxwZvf>w81vIPd`QuMLiqZSBWKz1fzBhh#|&oO@y*kuC8BWy8-^T zshTWr5JDzMb)86z0|OjL&f$$G0-6Z0^-7qmcplM7r-E#*uP3S8Qc<_YIn||K;wT<1 z{UdK8*OKkxL40o`+&HSsjyvA*S#s8|7J%i4Jz?+LU7)lJzn`VE9$JbrHjezjFw|0~Gb>a>w-WhHwKT5$2gbWq)cmrUn{&14v86Fne%6h*S{!@gXT z6fSsv=yh=|io|AH^FydGBqSC3e$eUDYN{n(Iy^3OTkevaX)0g5V@E_P6ePF(lPnY~ zpRM24Ty2KVIe<;CS5OZdo=@}=P9C<9^F$asvh;|40fBi+U?l#@ql|kgn{+fD62VGm z-*ExRa@vdYoeR$q`*wb)d>BZT4EVVF!EdOW^ZW-V2w3_>xA(%{)x-SjP=#teuK9Pro7Q#lyn24hM)MvG0)Dk0I;@D zGd(A_0KguyR-#UGyj^Ob*rpXVdC>jZo77KQSv4%WK#*Db9|MU=-4bwhGg=KPKtA%fPCc~CEa=+)BEJau7Q%Ri zoJiKKyNi*4#CAB^o+oe*HODN!GXlvte|Ijs%h@;lQE+h87Ua1MH1yg_sB+C5n@NR$v3D${|OF=?X3>cpe)#_k^ zJ(bi9e?jB}et~QNbAdvBWr6_>zQSdh2?BrSSUKvR=EZWQHVyVZIqG|g%+v#a%Jk_; zfMTH%J_0)I++w(fwcc*}Tpav;W?usSC+Z_G{zi-ya1`Rz1&Zvn2kug%DgE~`!3gKN zbWnUIhZ0lF(o10PTdo1QUdQ;8RtW$Iy4mJrg>WrUoxZ1x3i)y4)Z=3ulN1s#$7Wu@_AoVk4i|21?=HK(e^K(5-|iF zPY|^?{Y;}JCwOzbsINWFxz2oA;!pq_-LV}I)!+}zN@7kde3NBp`#4C^B4|PS^~2j` zAJk?U_vZ8RD=T=3rT<#@F9UDyd%4`S`(lUn1(3p|Zy%kgoBqwhA&%L7zg<7BMG23B z&sXePwx=)&ZgRKLO6mn~B0!+v$I$j{K5;u(%{uXmb5N7(UhzTtsOJRVVmP<(u_;x* zYI0PS{uhuvxR-oTK4R#rQQB$S+dtfz)eYbE0)F{*;AsdOB5w`oTmUR)0WiK_m?;D7 zp;v!|>D|G|I2vql&MwXuFGfLkd(i;=fh^4tPzdbLE*wgMy7KMkqXj!3J@sxgSTjys z9Dl3T!N2$zvJ(`iv{~J=wxPk-iQp^TpQyAv01^j|c85M%W2cCWzaV+xKJNn;e`$2cnFTg(OZp(z*P~W3n^=WMt2XbTE0kO{S6^6=M{w<@!ck)dh*oXR-clrE&E1Z*Veq$cB z@d~Ai0eSW7X*`ueTwnTal*Zg*Oq_b* z++Og9MM8PX%kl+X18z8ZpHZ#>9yH89NSO2MJw?wZt7iwj4xSJqvDL?-%mdavyPw3I z(h2(Y+56U3>z4AqxBI4-@lwhQJ-=vm<^1L%bW@^&3i<39bB|A+PLASfGasYjpeD!+ zo`3o)kB6`xx=;t&6P~;<+TG`&h6+nZT6)22cLqdtyEa-xbdjPW30A3Fj}`50YU9*v zi|BOI{SG?gr`?^@EeHiJ6c(-a9Ia1^!Mk zHUN>~;$4V?;$MI2WCZ|wl5}T}`i8kyUPXH2J3a-jj(0CQ4>4XwmeL*<$fWYUiqg4U za5moEvaD#v6J;y)8l*Tbs1?i;M_AfY2`YzE!z%bH1yhQDe!r*B=1QeIImi+6mL+!A z@vU6l3&jeu35L~ga=-g?z|0!iGFioA@ya%Lb2(^vA>B;mR0vD>)a@ZcZ1`6OuYd+l zNhv7NNKC&bEUZ13=<0J48HBNO^}|?gy#mLAo)gFP@SbeFsrFQ|!TJ}oGM^^zC`u2C ztu7BpLmp+%O8*V6a93z6LgHE$gTooNWi3z-qIA++o4cKH*%iduJz>o_VO!%^U*fgV z5Sh!dhnM9Ja`l7}2=6CaDefhqI^^%YT*B9z4T^I^-C}Cm!BrhY?D*!GxWek&jdjAkm>Vj4xGx7# z&zTPFyqtq#+Fo1s!&T?Yt)!b>w6op5Sr79X{7+-LgawHt0!ys3M4Hlcd2c}?vMo*4 z4({@OcWxYK56TM*`#pb4g>38Eb7Pp5xmfB9MVTD)t0EIbXF?fyrN;@nh2rU8JH@wb zL&yK(Bo9zM+5b=G|NjZ9l-kBSD0fREnn!e*XFR&fm&(Yj?nHS#nXL3Q?8q^yC89QE zP zY6vr#nnVig#Cj`M-8uc6b5%5RU(uHAQ!@b zY2Md+Jb(9zWhQBp-us=~S=T*f=f_bSyp*jyvakl%gW~Y#mYpU7Wq_n>D8HJ#v^sG$XUpsnxCf7+xkHoQx5014XWplXa;$XA*672C z`)6x><sk<~52 z+@;*C;**bbN5{tGvEG-JGeC}sW`Gk|8rRHHY(|}~9XKfH^Zj`M=d0RV9aT(|+CWU( zlEd!AJFa8YX1Bt%K$-8&rI8iOxh&*-i`$6X1hGdeZ$=uswNQmy*RieXd1#80d7#B3 z4cJ&UNZH@2IXiT}J)1DPR1Ws(RHxQ%=B87yMiYbRxsxHMIz@Foq=8mLs(K=&bs$Z( zvkLXpQu~JV;++VfmUovD52ht*74~hm0zw2os-q*{;%=D2>f&6l2(?FC^ie?!Y%~DE z1fhyF)n-*&i0gZQ?Skl0N1Le$?vSj>UF~|P0G#bw*M_}uPb(`>iGT**@ASyHp(*m! zudx|omLCbnEIx!Y_GjQci-F3Cj>vT`T{?h2tUAKXmip~`jm3t6n`{V$G}r^NvKFeydK zeexD?YtR6PUM_l05I!BUudUT5t96AWn!1+pIfUj&_2TZVF*1NM3=%Y9ZusrCnJSmk z2T0H&Y@ZxtSWaTWoRHCd?R*>bD*)0CY*yT78)@p&-C0ThZly-N@U>&HmM?uO_9M_# zzX9RNMrf%tku&kQ&ZNcWHU$xw`&XLX2QEpFPo1!uO7q=BPVq3PAPP7gAO=<8Ekv4_ ztQO6)z_Wvp~UBbq5zRZ?mPuJcmmXpg;Qa}iZou~vuScA3>H1OSv#M(=& z%2m*qC2@kHdl+@{FLpyTqNq=_iAT;6);wzqF{;}C!jS!N{Ap;|1mG>f+bA{`6cQSg zoff4YROhmI&_gmn-WP_vd()x76#7wnP`P#`5Q!v@^kcJbq6Qt^qw98ZS**C4*utkub0 z^OiQAQD#FYAh8Q0Kn{@Euh-d?-DqJNhw!2%lK-hb`=`4l$)UBN$mpPZ`W%F5XfbhR zF>S!Iw%+`co+d<)Mb9_1Qx>lD8-lGdra9UIqGlfhse~fG-8dYwOOD72m|XX7+k#I> z{Vk#q)BMPZxXb5i8dNzUX12OvaCN^Gkyt%nT6YsEg7({7XlJu<+Yf11PY02}{(h)b zp|$t_i?p{2t8x#!ekamOi9yE_5ReY(6iI1R1SzFNx}`xB6bWe%DJhZeZlyaW-6h>! z`<{OLeGm4*cd|KH*NSW5g!#yclv?n>{{d*l1`B%52hE%AQiEDq6GOA2qdU%_3j@Ss#` z(d_0JxMqQ1?m*Y;ex_b7{E zMJfzxcXfo+EQnmb<9j3fKfxtoPhdPh<@PW?7eS%AHwCBfR(oN)#(A-p;^$B(1#rb( z!_kpm&EygRDIvG|)jFgn<8`@t`AH3hbbrJT8ZlX+cx70*L(LeNCJ<%AY{c z3*4y>D;v7SX(g#Qx@617KdK+bV~;6KK4gdb15Q%Ji}PKP6U7x5ZDvbY6 zff)wtH*5n6jjHck)6y8`y|QLbs_%Cg@m7H|!f6QmAPbsUt{z|PM}~+MO>ep*XioLcuyy#o{r~T2}P!*N#{<*=C+W`yy7)Z<4`7=Us?^4+V=q{R@IctcnPobOcd zZA9RyWrUUJIPCZ86I9W4J7gD^4)pQeh-gyrI@NH8?LCUTaY!xAJ=8A?i{l^}0T_Qf zAL5m+!Yx~1qj(_Dv-fB)xun=y{J4S8X0MB%PRaN&1<}>wsRhYtICL5SQUhG^h)?lr z=?>0d)IbF{WUhp%T)5YKPtK8LwZ_+;ImN7=->_U_R&lwWu=jnLw!=tUom4?$tAp`k z0Yq)s+lN4nH+Y*RMf6@`8ec(Bqi2B!=lRtKmy{$O>Wja!DBlBI9jfih8V8ZkcmjNj z^4)e8Ale7vdDP;4-G>ku^zC6eS+;7SYe%jO>JB&|g?VvC0Y&zFbYcrX0}ysc_y|FA+q=a9k7;f7V?8j2v3)8+&UVTHQEx zsR`dBpjGZ4r4eO}T2gZ&YL&a&lWFhgm>*8&=d=$Lpah%ST|HG9czo|s`5qK}5;6WN zg=b%I17u$m8Sq`ZM#Aghv`i8C6so78>CFE)g78wCKp%yry?lsq?me9K39}qy!?-AZ+gy&d2g@Y{Xa*w~JB%XAS@SNsP;5^W&jUWE&3+z>c8~%wO>nu7s)aLj zaUnxhG8rvkCoi0Kff{a?6&l)bBAx~CaS?=lh6t^v=YeE zU0*eIz&?=PUj_>@l;eT9_N2>tt|ZTc&1rq))!3~>MZ`E7hS68LfOV$gyv;*w3&LQ* zCDMOcx0|w*XlX{NwYgGQt+bsGu=)6fqn32M%vzR6u6izt@pjc+7l2pvk9x>CB)I~4+j~9S|CEy+^kPw75KZo@j6yGLT(VW~yAd`8h z4$Ga_N7UB5+j|8cr1fqc6%-eLP?lXXufp)bn8Mrt8wkcpFo)d_Z_|;G5iDSp$QXPdk|3Xbq(JIu=iN`?=6 zXQTI$2NODq*uwnWf>^Qu?#!t9AwE|K%yghcF1`1W-feYt18J8lzWqA*825IxWW^%F zuq2sNc!eUB}bpFKsftO~R3 z=>)5DuuW2=Z6o*v)=x`exZo_3KSPJB2q1e471KKgM_FbsR-`hg2R}b3E~PqvC&Q>> zy2Gw@JU+;Td-0amlPC0$Gx}s{b-S$!-&;RHyCXaf9`dNiDAdaRr&TEyWl3S2=~$$+ z&hj%(k|^egeorSM5P1?U(gRLU)SF;F%#LcBDajFgPs@q^c;8HEIQx#169KgHI8<@t zbK$HPr$6{v6(6h^|6}ubp|6m1Xv#ob68o`x$fx%5VKv*7pO-uAZT7KwwQqW&ugkXd z?7vUC#1yK>Qt@^a(&D@N8xcJ6P+}@hdFY`k|3%>d{`(VU?>MnNZRMVa4=&lFJ01#o zI`7FaQe*5GMX~0m_z2Kt(gt614D43_ z_!_EOzhuvI6T`{)LC(f*POV*qstW152#c5zSrjxa4w?v~IwHU8RO95{Hn$m>45t^(c~ z>VobLp{Mt84&C*MM~ZToHEt85t0icUcdyEPUEaOjHM9tO*ArbqrlINp-6x{hel1r2 zDC-yq7s|V(fcIQ#hv91;(2;!O4f3x*PU#}M{W*#5Z-sff1CSaKDU112?PPx z9d#lzbheC-t6p|)uE6(o@w%E-g2x+6&1(C3%%h_R5n)c= z2Qp-*-HeOVD^f0ryIaQZE-m<)EbhArn(0m}Z*lM`$s%I(k)4*LBsX+n!S!WstYet+h08uo}%k7KmHE zY#32UJMB@X{mwn!*{Dd4z;A_S7X9v|AKkOjzT=fx-#Dq3toh!9KsA!e4ZRVX{ik3?;N|Vk9YL22z7o zx)}`CZ->pWEoL5jN%E=d{~dTCcB;|-3(qo0G6UH!|HC>foOvT{J>;7^-(urd=SvOO zEFX%jf`OJwXJPwiy~BM@FPaa!X>g^?>a@Hfa#T2QQ3zxuYS?IvMs#C$q(vMnwx?tU z7Gpmm>kIZ^GLGiV@cKawZlTF8`xP6%$L2q`2&$@QDw%Y5>oUn>`v@pt*F?mj^@0BbM_dkgbwBAE)oOhu{69GM)JkCiyaOx zsFzwMvq5IEzN|~g4^5vcgovC>dX~)^Kc2C!OPbNeVxeMGH#z}va-al~<6_l2I#rN^As;5xU7ZA+0{4h4jzka(T4=v^*~gSs zuB|eSOOdT&SY9WJcRX<9A(FlsM$hG>LY3?_&C2Puj?;gLXfwYx*a2>WPo-O*`DqX0 z=yBn|z*j~i}v!3o0io~MtOKSX&yQ%e}_RrY_#$8*FXLK z;4j@f52Ds|^WPc6`)&Jow4aQZ$YrZ4qvxxP#pO=EY@RPz`E8rzZ~0;}Ns_45Qv{*~ ze}5pk6}I*&vi=+l$=?$;8F)V0e4M0`VSWAD=cD7A>#O$;{-Fl(e%+U!qw2Iv&b=*d zkIjs;jwo0gjx3}wKTX1y4#gwqq|!gDt}{E-J|n`y&CQ2Q2XwFus}(ljysuaMZ-&7C zWvRuEkjZl<19YnNEse(pVu(jY9w~eg0J}me#IS`y1RQ5p>^xNn6jdgE7;O3<1 zt@Qe=ei&LvI2F=89?3u1==zaYLdX8(+6%qhNVzL1Xk`92MG9B%huioqDInw{-Y?ri-mwxIQeex*j4!2O1dJ z$za+Yzg>*`J@&9|9~cQA`m3Uw@;kmIGchi9RMDv-T`#-~`3@>^9jYcp$G+U37LOY> zyRBs+TRh2^3KdE5PG1_8Rh}v&k8ilHO%0s^2qeu2qlC$b-*)xwok6uJUp~BJVWO;$ zwV#em4P+^HA|d@&-wo%?2z9ul-M_m%YCzcne`n%+P~)?^c68P}Z4`@9v&Kde1r`8U z23U+1VoDka2aEl^dQoRJUN)6)**n&J6Y?Uq?kA}srh34+(21MLvaffdOkn~Sa{cZu zHyn)9!A?eR^)Ea)8zPTVbeSBED{<|#_}$dN)`_3o)TSm z=Ojh_S}pbU{O&Zh6EyW&2yd=Q5T!r6UFdjVz{wf;0y7@3Cz;!fEd@_$1vHJ%kP$9^h^OKq4w>*i-x|9h?>APn~6#Tx+wbHRIGfTD)SGv z7R$8{7~%`XJ0s3DcspPaFlVP(H0OH{GV1`M zz!PCGoU5D0^AYB8PCPi6agqli?9=={u zni8zgCPx+T6lFZ6M&m^OgXs|r=({g|Q)>b{9ApDY+LTi*lSGrKgZeb4RH^p>K8?&g^{rK~omu1lXh&pDf(WeB7!AKpU4k4ubv|4(4GN2BGv@8ywjqEd7 zW9flW!(k9fae4wj>Mf$eqPZKJh?U3XmsT*DY4M5wRrR|x;3S>luRFmHqu)6dBS}kD znj0|5c?jP$;88k8F2DPu!s241bLxwk&B>~c42n5{W;ny6^89hGskK%>a~MFuP-wsU zh?H_X%h!;0oAs8~O;QxrP=Kcx7YG+b>+q7#f-i^C^B_wTn4rnmZlIVKCZ9Ng*@x zDz0Jt<*)ThN+nTJ+&=;bd_$aT!auqZ@s4*|xcZ+iy@MW77ywYioc-|E-K)puWXb;+ zB~ahd5lBn&(L-UT%1Nz#molW~_#n^Avb}L$T8)M@I||xToACK#n&VSA6o;dy;J{K= zcVT%856j0xGE(LCFR+6#CcL|+EOuf9IugwPz~`!W3dn-@`uMP{4iyX;S_tj_cBt_Y zNsgjbpvp*jEd3lcmmg~lljzu>?M`%eVJsw==Nxf^ISYba+?RfZrnHZQIl_&}YEOsq z{<*2leumR?(Rn!fgoIQUf{*Nc*z1A>k@@6Q5H&nO-Y-01Kea+A)w7S$5(`>4pg$rI zOd03VJp`Xnkv*<=!|S9&?RLLdaEl?_8zWT`rOa0XuR7|^ zzw8{K%3mL-?773zom^ejmi)mYhrf{7yfOVwo7*rI=Y^(hCf4y2)p;);Fq z+WnE>%WZ%0@8QPripclKel%1pYzNvU$iZv(*9qA98o#_QhP@uT42?~j6n4XX&q3%9 zxR4u`C6fyGd5grM7y8nVm5 zIn8BTEPT(p->7E8tH_5@4<+RB1Zmxoq}8x{F$&6vird(L%}QGI9JGUQBwFEITS0Ih zV>}l07-0E?+{FpJ4m}2Ncd!NVZVs+Yvl9pOJCDuJV&u)qENvsf(>!xs&c(m!KS-7J z%mCVj&rX@x$E*Dd3J_bkn zb*a5VG9N^_Xj{WYVJd#6g{r*h)qX!uA9@lRp>ju|?bF>g(tIZk{(Hz3hi#BDC`k0) z{Rih35c2<#;|NT6t~vrxjiG>_Ihuv&9H(PSNj&wep-}uBN-smnS@x{KhOk2z`0wXy zSWlri!S8Rdm)HNLqczU1vSN!6CI@H4b{odm;CjblU4Iq#xmzY5BI%T~x>C$3p7pYW zszsJ&Z(OYPX`U3M)B)R0t`^e7^;7>Y#F}8hdIa_9`^w;7>fHUF^FjP`pzSE=$x&}LT|}y7nf0yoWq8hGn63` zWqT6Yv<>k^?&`%@dxvp@G4r_7RLpnWs;aAcda>Eln8`5b;XkPse&h9hvNS=gMO*1l zlMl{?Q!hNdlL_bw&d9BVs)O;x2H1p+r?M2NBdGF>$tuXv+%C^8b5b+2r)Js% zwl67WZFF}<65T{Py$f*xHLyf)LvQqzX?!IOC=T2i&vbY%+xTD#!sw71c*R|FE#M^kz94p-8RTGYt}~>IL84IEp|-+31xcOG5o`(xz5y)VJlYB|8ayD4d8srJrnwqy~Py9NL}T$ z@dyA`AhOk5#;alh&&r=Tq#Nb9^fh35vwIRYog7*5 zQ4HlAdp>544OeU-_v?C+ZDyH4vCYAfhs*y}YcRRj()SU-aJauUPOlL8j=}PSTU*;0z{IRZD-izT7 zjV`pZis;_Hw3;cAyM+4bE3K%|kO%Q|qO1LE!$Ut-a-ukkRdy%2E2llS`JL{scI&bz zoalek#TfHZFGZhk8q#%Yk9ey$f|2joOc+ujRuCAcr`D>`Cf&ZairF(c&S$S*WczL6 zf2kq;g=YU(k?8->FaVv%&z0x8y`AZS-D5@;gioeZ{%_w%# z_k&~WY4&^9+0!zS#A&q}g4I%ULe?!4AIn=CK^-Spk{920U%hdb*>Eu8H)=1HXE&z# z&A{5xlBN2#=U{WK-~J*SS!g8|GAucsF8(s&*S=1j#vfSOy5;q9s7Af|tN*@mJ^YE){zSdvZ_n#v{X|}f;It8Vb!Jm?= z$)rDyKeCj*T<>lkca$-P>AB|IIi+umZ&Q*MAGY5s9DGyfIf^GSYZ4K=94;$-VG3Ow zUN5oZT}MNA-rZ3cT@;?QQ- z@F9sfjMTn3#rtqCMqx+!#9~y2bKgzKCOrB2>^AW^bj%ul!BZf7{^F1|JSdMTCFnIT z`F}V@(_~`?-_lzPvz_eaa*U;FMWQ7lE7eWQw8w+#JnmV>OZJ*1F{RQe$EB#)c^KU} zpDa;<0Fhk%bgkNEF}fdoEOq`{hyY zw>mXY{uzlHFGjW7WOd5Z%_6GZ&&oIW29TAW{LT^(3U7D~eir%V$hzvZOnv^fJ# z>V}ht&_-_tODa<;RmhLlzdYF1s|Ieeqlhm;)F{60NL!M|l5LuW!FQ0tC_)OpJ4i&t#xWjU4Yj7>3OT1Y+4UBj z+UzM$a96bGj8+|`@jUrn-`OX`Dl77@9KVhWn_cP zuydsC_`_imyB5tv(caR}>Q2+4Ki*6UC^4JKW(rew9j3LM7X;d}3ilO-BX6e?`b&WY zq~Gx6YWNdD(b5C6Z{3(D#*L~8*&mYEk&13_`jsTs-@xn!TAiR@a$xD%F778!JM`yB zsf4rP4r~jGsPvVIrDW+@nv-fxX65n2`8S^syRcR#Kza#y1$ue>B*%8V zUCY}gQ4*g|9t2#2buZY()0L68oQsqXjCS%St|mK6IF#_)yH@ZdC8AH=yp1+ zmuGsdKRt=F$GF~qIbS2SLo|m7iR;+ls^xH8IBJkgB-Cs%k&3*?w^gnjNwmPH;9dEskGzf z0GAf{3V^KRfa7k7Zuw~iHc_c&*K-ag*xL!Qe=+pVeY`IF(WoBk_@K^f_{}%UAVTtp z@h6F0Ce38{Sm?hCI<@929I*=cdu`QrAUgsJ?!J}Q(z)EdW&Iq>`4_i|y$@V#DyYxi zE~rY6lJz%WNurPtMH>Qp>WhPa>x=JZ@BuxAbp^4y(6$d}DnHinleP%5tqr;!dsN>b zBj!lUcgJCAwGUv#B6X=+)ZV(e0x_!`21HVmk-^vH%V3DJ1O5|?KqkMXnLon3d@DPL z5z10{-&czt6zVLqORL2wOAvi3%20&vZ14jZaL2}&NeQP%jhDK~_qQaj2cy>An zZfhx-CC8?TdX?~#rA!qVVj~J_cHZfn@glxfs)Rw<$~w1}|CdFUB(qr@U{F8`fap#b zz1hCJw~jDCZucLY_Ehfnd7**^U>W0-f4P5Cn)^MZIMC#DgJn|2_Dy(%A37NuW5N;? zF2Nw7^av4`@~g`;u2@CI41#>$C(pKVrHSN(dx8VD+{5T`VWI-}QQy{4as)=+AjR|n z}JbAT|qajp18yeEThx)yQP(BRTv3=TDC6M+CFSvDVlQjCWv@?0|oYDph zwI^r>ZmbK69l7`fi4pWwQVDV~XrkcW_#Fp@e1gkFeusUpG&nUd2U2{@EoG&_#G`lK zB!2E|dmMrOiu#9~Og0)y7J^QcKXCG|@t++OY`p*Ld${_X5h@cj|KXN0f_AyDjY||h z#bJ+wfsf&h+nx=dhGWt2?-6R@9ILh>A2gu)^K5csY@0D zKbL!nArL?53+&t*3j){EcZYuY^i30<*lWP&m8OA@e2?d05$ta}D=bEyr$j4(@BytG zj_kdk4P7L2mHAL;NdS4+LU#!*&NZI5?RKZv+QI39)H0ndpAY6ajZELQGuCy*MhD{$ z3LXK;o3;vC`har{Cgh`pWr4*A4I5x}fZ%8pQ#-2Z<&EJiem+aMLvyu5bDy?zk1{^E z+DPvsa!xbxazA7Kq{pi^_D7q6VNv~gqWf+eV%?3$`#7W06@r;4#^pAcBRYT#n6RL8rWFZ!r8G<8VO`C31^w>21{%AsuoZ3K-8mlD~ofX`zBR<6XuUKSo0t3qbuv9LmQ+inO5_l&Kql zg7Gh?uEtBFNABp#)}ctGD3}k}A~=0cWR6{;L_4_8k6Xl>7o+uq!4nY6$SCGgD4KqC zk|KUehma^6KZwGQ*&?O3{e5$!)H1GbFWpt3ciXKkO1uB5p6ZXnEwEk;>)7L?T3aK# zZxd+wX|i99Ko z76L9}dH7N*1^&r_1i%2ceK#>r?Jp3;$jIAyza)CS z2C@U+WVkeM>^0jJi73(CgQ>2^6y7)uWGNHIQ9 zd4V|jr$zkI_Kjttd;Whc!{YFu5J1tD&UaeSV`yjvM*6hpQI{gT6_O}n!1V?1%-G9g zvDQ5>=Lnve|M~(-Ad0@?OLI;OcLw}=tW(Ef*b`KPLf0H?nqX$(eOjues*unsx~A3B z2pzzv24kqKSUz%x_Z4bZ(5KI)jyeabgWbxzd!{1#>t3+o1XS6S@Ap4}lH&c!V#FB+X}&MHH$L|o$R^Iy9{iv`Q6rpeatTfo zHIxN00l&mvYBl*5i*U_Akos(-Wezj~&RPa}qX+qfPWAuPD3pS8Km7Afg(DbD%8n-N z2~q7vw?BFW%!dadiS5B8$#Yx^S6#hww667tUe(EhhKRlx+RIv0+|e3U-1M^U;3Z;# z3SD2gS;*kq`uU#%k;S*an$SN7W#;%Ig1_cRXO=YE2=Iz-AJ9X(xPPOD+h7HTre6em zl9;(XcrmGWcvfWls&ez`*U`>exwbu*tD%LE^5M$q74N9%Xcqiu z%w#z>d9$rf5QWAM?ZK_Yf1S|rH(8wS79L?13OBzt13-zj3GgnaFVHQ$fe0qak%ZqX%B2h<60|U5>?@6`hQG&ue?8=9Y5g zsh^d**yO#Mk{6`XWM|D&-wQH1sg%yNH)2-|KXG=nt(g_=XmB{W;s)x9)X79pDBaoGCv&Dx|w6&E2 zoiC@~J>;KVk`!Q)L=~;-88h(@GZv-{Ve~BHxQ$|KCP&oSt6xs1t=IXS?%@t(36-tB zAj3KwBuP83{iy19u6JlD7V*u?wd$uxR;}j6)+4e?zWP!dx}WV&bSDvePD(T?WJ9ya zn^)}p1)@Zw?$p`;V+l!1Gj96-AqM=>lbV(-x2~|q#9_yOe&pv&)Y`1Psc=)Ftv}Ln zILGftzWGd#Nn!G>=meHLp+Y^IiYK)z-FMI?_@`qz*e5k_A5cUc&k!3cw6#-ccMzK> zcIdj|=Vx)+m_0;N(7M>|bzesLV+2e1kFk(FM7QHDkOAR%^07Qt41?*)a(ouQ|W~-7oMHmv;6&5RbftQ zjNSP0biwEdub>?j)fKkUXimC88k3P`k3ewO;h1;!jk+n-n|mVs=39k@{MV)u*Cy3N ze=6oy|D73T;mGxg2>PRx0wxc)(HB{pFE4MbyG-AOc0$U4M?!k2W~c!f#?{R&#QKHZ z`#36oqGVBCN#S(Gl2VGVY<36!T5 z9VvrSt*R~UXR%u$Jk#0L56f1kHftvbQZjoUw>V6{d_Uxevp7Z(~oYOKO z6RsWT`2WbRDPkSmGQT!J$Ua=ny@b7jev-la|uJLwCQ82O2ofEsAH4fkpHN6Qf4;@aC3DRcnX(n$ZXg!aCR7r>srKN@&B(A9OgPsv zoQ+kZ&r7SO7S`T=I~CT6cw&v7{5mx=Q0NUCwgw;)l{3QY+%eK_lLukJ$~9xaR>ocw z55h{hregg3*a*O~mWCTa_*C6*x3_LI^#(DWm?2Vl=sHQh zl2Mb>TzT^8Erkcg;7|1M<3A=c;=yO7@De9iQ@BQe#0;8E1zzQXgZ@;@%{!|-p~{pk z-Fi%sU8+jJS5Ka_stS+I=`yKrU(n)m>whYWS{l=9HQHFD=z~p+zPxQmKr$mH%mpHvX3{UtYX;!Nhx= zmk&l6=GMl>YlzGR?)hr+b3e1dz`+m6-!GRbinWpl|GpyQrLVm;zkink1Ha*IpSH5p zzVG2VPEg6{y05+|knrSNK8osrMe>n zXddkGLzn{IQ{}T-k23>Hmu)GR-ckt0Gg6e!+Ny^5bSCs$_buI(^a@jXnb{ZqXD@$p z$Jt=eFT@v!b+0S4zZC8VWBi0JY!ch=LEh1hzYv1)v3zmEZ*$c0h6#7_#4MyJN%60d z5iw_Nm4wmvyU-B$TWP5hwW?R#D~?L~UigABZ&O|gW|<16a2kZXPh5Q!cVP=)GXxrNX^syTN1L6#DSX-BSMZf8jIn*|65Jw z(B0F6{vwf@W&l@8YaY%?KIqswGh9!Kqw2NI$r#Xnx1Kh5_eiQvCA)Ug@kMyV{YmS(ir6 z4<(0R$v(9_88^(Y%+H^3=q4q@<<~DW&m?OW%CyG#f8mE>N{PC55?cLW8 zdqalb7EZ7QL&t%S_04~g9##&Hkl$D(->Sp830AAsiH+>Z1bolrg{~X*fR7(O4<-Iu zQsX?71O7KhBE5M{0k!Z6-tULH}Av5w*6yjx8!h zqADvK9h(sqOflGnGiNBj#+u%d5P!^%r=_^R1>C*tVe?ukVJ$%UXl;G*LPU`e;kn1} zbt_1w=_HHWt$A7WVnxL{-iKcP`krh=Sa;*a(R9e!W>rtg-jDMW=mG^d%p7YEFcUF7OXCY}PI}5H1A5IL40eQo-SlVz(ygl68Lz!qr z7;5>`w9fq=P!`7Tap|_t2dRdMVMifUF_t9F=yR{X`1z}7Yo~Kl+>KA_*HK=4)PhmT0*?e4Xe*4x0+F;S0VHs^EW!0H}D062;w;h zaIeK$8Ijj29>*q5qdIUdxelrjxxmohKFXYJc;1Is$#yy3 zc@DG!yXL@y-?UDYK{DvW`d*^!b?0zlH5r)B$z@D${hMU{#rFS$Sq=}&H#BtWUPfw( zgB|WcxMkAoc~u204OH(BJT5C-<%nGp({RQ-a+T7>nX0hAdLK33;P?&B9ZcGn!5Fs* z`CgdAt3qixGm~4M{6g{Fa`hd>doNgu#{MyYqROf<;btH=VBuW$W8XX59q-Z%n~XC= ziWzqXMcAlD@~DN-bYu11S2-9$|0iwh={dDn z^3#2vYB|wMOj-QG0mFwi#f&fRql-xYCqpgIW-a2?CuIAd|E0fBRi)rGv3FHP-ZZ!% zrR!*oiQ2L@#a%msOd2;TDI(;-n|}ULXKYj;>JEB@Y^+XHK%%A(JV34g~_ zLTh{PLuo2%C-1c{>-7PpDanPcOCM`Bp1&}+MTsEb;E*XA{h%J_oH_oE#V#f(>Ba2# zdkfr&w#N#}v*tM{_d?RZN;$*rqW1yq#ZYJ(OD$l7$a&sTBO$`#PS-AjUC*`yaVEP4 z<;$@l-g8_0KCk;w?uaBJ`R)|Gz4YP8ji^SMTD%dMv5 zY^C&dxmcaR(3Lk%5k3|dR%)EK^vNX{``84O^*|xz!U9f8ytY15#{sEeRt$l;FHIYrch5sS7%bq@2+tu|Y{kcmeEa5($+T9oW zUcc_#gFK==>L%>Lz(yp=yf|h2;8A8aPvz8Z5#P%kiK|9l#+3k$8>ffz+tjDECnqiz zfWd8UT@C!P?;f8w^j5%T1_4K;S2@+znhw+22Ce{r_&H=0PH?r?)tvqr5h3Id9c+dk zVT)xuVOsP7f85$(zCHgrbfd=wbuv91LpSk6DrjAH)S?qyh$SAKABjm%7X+^TH%@m4 zDXz*|cCUG5MIo_3tF6O`VdKBNM3#!O*LnB;L>J-dt*u3qw+ z%%kkl!la};lb-DorH1m~+LH~;9xUTVY)n>@XLpr;8qhv;NOaO9Y>!(EoS*ak<{m@6 z_#$sOBdi6LP&HzF%@RpPR7!2LCb$$?n({ncqpl-KPc_%Owek&xXwISzy8e}SpA$dZ zrD*s*&y{Q{rqUU4ugh#t&%ir&0n{J9K zQyGp1tP-|bM3uZAzu3DOa4!2TT~1?Eocim2*yC>P*}Z_7sIFSkI4rR^Crr{w(Kk`5 zR_uxnxT7@ZPVKg-qgrb7u8G2+klyH}<*qY37r@=iH%bE{(YoBfP*} zBO;1LzO`#zHSu;`TO;WZWDdC$y{s`Nr86a0)ps8~z3L&?Co$t;EE#S3bNt-t_6!Re zOEclSfTUT|`P8BT*HgA1?kVjwm+}5s!IHp~mMGOUHjr+KcCM$u9LU(+^KY*UP0gen zGz@HjUmB%pO9FaI10kqN_0jsbrE?YMXYAkf?)AifNQ_c{O=|v7@SqH@J%TEn`yv)I zTQ+HgYgvBaHe|IPvrF4`&O=`GBANBw=>z60;9?XPM@iXcNBUqSojWKB-P31Rkhjsi z`?!DQIQ~wW8O&pJzo@+1)m0I5u@V7?&mk zwtzgbGsl5xje#2+PP&oXSKLfIdWo=IU=_cRh_zb$d{5Ab4NnSg`7no2a5IaV7iqk5 zSv4OOHTS7}G>X}G}-c}6RO*K>|7Ca72JUO~0o>$$YP1d)&Wq1)G*|OO|n@1>(;S>1aB*kmdFR}N>y%~|;dT4NXyl1gN@EHKxHd$WK!gTvLH#r>EbIJ36rdIgDo44ykXshV{y)tY>X&f=8p{q2(TtUjO$X5 zi=rQ8)gW#EyrZbA#3 z5ov^-=y|b&0wzH;ocTVBS0>aaXRLd&qgpJ={n};@wIYJrIDuG&4v$3ec@jCune}tC zNv=GF-irAS-{xW$H8(bj?G{kcBXk%2e!u)Bvw}RQMC|>~fjT_pERdQYIp1p`B^$zP z%Fj=^jWN28K9C}L&=JolpQxGk#~*yly`dtw==e$16KUeJ z2FAKb23kqML@<)*>QaQq{}Ln6$13K^4$;fgWKy9ur)zG-NmnNhRPKLhp0=?lEZSvq z%eR5ns{&;w=Ol#~x0HG8tA(~YW7Pz25xM}Q`yuhIS}76aWUutKwcm4bqpPP|D|`-m zt&3c9OkG(}&eiuikF)&o)4kcKh(Oz%mmD9@ztHGcYbhdD?`${KO5@7pv-&6##|9JB zf0ANbrmAz}XlAu3zAYN1WImhqyKE>+_8##m>;~Yi+(2hW{zU;nM?T-nLxq!FvN+6A zv@y0){5raD|H3`XyOtWUtS+L92qPW^BuR)VO9BlT-dWdJtTQj@82F7xb2!x6U{YS4gA zVUadbIehPJ3Hj2TWBs`#CHe> zZKO8*2NL(w?i#rkNN7VF@Axb>7M}K(O?2-+f20|u^wzK~rEQjQTZIBC<9o_5X-cR6 z)i#uJ2OC06hF@Lhw~YX0rkfMSmsP^A`%6np207BGbLBu6DLEZgr*g*pf}ESiUE1uZ ze*sw{479sBuO#3&VKs-t^m&^am7mA@ih=0>X??tW(SbKp%xD@ca-!?Z!BG z3tRcC%L|xa?D^cubzd@f5^l(SYq!vdO+aX=V#>w6Ezy=6kis4y;MX=S()eBjH||>p zVH;1(gWx8?|7qwAuYxJYU-azAQ0uh9zU{)p%D(*g{=0l}O8hoYm(G6r!qD8@YB8MZ zWR>$9bMRqZS0Pmwv3@tq0I0bB#Kc5;m{-Cs@=%~T@AjotV4!x*3?@fCL$uCZ?GY%9 zPUNh7rKr&m+^%iDO$m%PxZ@{vBYudy24$lphjdCvw_!{O|8GZvKW1X>(!Dkia6om@y>a_f_64gv#p7#H5ef zp|v|v_P8(c>_BJfFqtmaBkd>K3iFbNsS$x+QA^-Wo}8j+A;z{VOpbSG;2PfCjiv7& zHRyd9ED*c{oK!q;L3^Kd_L%D3H#dgQC&W?8*K(_fT6S{-rdJlPT)uVPM{Ung-6vMX z-J~m|wiwpnDL^D%tee^b*RLW?6y7)GqI6p0rNF1Vd@bN8kWOUndX#B$LS;Is1GZL{Ljk9Ffq&hO{J za2PtCBdPm@G832V(6>cVbP}&lqh1$`gOysrg+d7KSbgZX==K}vRyAz~tAriO)r_ix zP0e($yAOk!I4Xq^_YwE~x^%}qrYh90ccjQPB>X>1w@|&jyJ<#Pb7Vd5uyQL`A4JQx zr^vB;yC0#b0yM|<*g6%(ch7{^fZPni^&8&pR7O$UJoiPGN%e=NpYQ!_fudKTsmsGM zz^#;*l3@IKQXHX<>b^am^|ysOm5SItv+*+-fX=|l4At6!`n}WnQcA~5tm=o+qiShT zWyso3|VP0R_f>zKf$vHdqCWDb}ajh|j|$wUPz-lEU`0M{}g zMQ6}yjuCLj1=@vnPE;H|lY%uEzDFt@Sx}J_#wSRR*zn+aQq2aR(XmpGd)D+kL>>svyhKR zX{s2BY7#=Bq7kDrh_5Sq0%4Dl{?o9~MyXg$v%M=$wfHB4Zy%!g4=B3L_a;SRvOohe zxau17&fD{5LO=^E01?FWR{J$l2@7)~466kP^~+BscL~MMIZ?m~e(lw@iR{dwx9^6s z^gg0Qm=`)aA4*1NT2lG8oGo6^gzv_rGgG^7Oa4EUy=7FCec1I)gEWeCNeBo?!_Xko z9U>qS(gM;ALrFK%jr0Xbhjd9dI+S!XAkq!be|(<%UhDn%vexy@ONaBEGmd@y_TD}_ z(zH&)OIFMgQ?)uM0HW<8EizGV@E(NB_5p3?9tU9Afv`DM6NQ?Z#n2)t)}tP98FPZu zhHNHsK%9g(Fm!>a(|Pzy@Zf)Ub0PB1NV}mYpq{7a=4Y|^Qvgv!zq3=<5s++egsp)L zl|CE86N(FRcx4FD)bFn; zLuzg-cf4qU#_!!nZn**ii+>n%f#pEv-cnLREb7+wcGl=(IaNJ1vP~_i+K$VmCt1iR zY{6K2vqVkRwA%>IzzE|C7y48kptVR5WaZ@K&%|+YN%}^hA_-9 zTBdh%XW(6AK_JSt`7mOo7JhaupAJXg6eAmW%nDmoJWCfSDYx}^iQ7B79vB{Ik4c`i z6?p%?n)P3{jtLXuHyNFh6O$ALI!)b?>d0?cV>-R7Jt6!%wM(0 zPk~o#lW4Bzl}MW=fOX_*2Ofgzir)*xWp<;73E=JJOV$iRk)1bl=aEzIzz4_6kN(6* zq5jb+BR5OTVa5L>)NHDa5*;3aN(SXJ!Z6pURD;U(h1L|@nPyGO*^XX@$NwjDvn-9Z4?) z4es52+UrAOVopxzZ(BD{^iFSNxyG<=9#WF5ITDk;v%H69r{o)vLYKfu*IPHUjc0WK z;Fy?XEf&4-abQNpaHfBCIHMz$&7NGjzr($4V{O~D@XEk0yRT61oZaY8D|EhfD<|Wj zCCcah^+P4*eeGK30}o%yfhGu@E8GfcS%-PqQ0X3vFf zxGK?=cM(%%hn9#ql!N)2~W6Y);YWjYqZ?2 zNJJXDjvIT9m$PStMRJGyZgLj5j_pl1)W32i-rYpAO8jyIuiMmTpDOmxhHr0L<}S@m zuku+$RN`i;4a-N*@nOOu5Vjc86rFSn%n76#21&w!>}?;KO&YYhJmL1+DZYiCoOdny z8ChH-E|)$<{dIRPx__+QCBc9+sE3jG=mb6S5JP(L+&ry_HGR}#)wge;OHN#WIFIx9 z2L~$H(9+&-0Um`KRHNJG^rt_WV3cR511{FdWDE;bJYt14cDu*NL+fyAhaFgh3>m|= z*$LMmSC>@mo)>!Zd5RfmNwR;tW-Q= zdvhzhhWGV;m6a0f#^SSvo^F`q{M;6|=WTMJ8)4hz@||(5@|V-Pg@c%VuDWcVysWcv z)uRPFqVl)d+epfmvEbfL5goDsrHngm;kfr=uyigK`bLjD}UWxs4+iVR6J5t z-*0ys6BbPUD=DUqZ2(n=rc@&%PwiZ`!1OxX45L3hW~Ys##_#1ha>uyauR_WXWteO_ z_A~53${No%z{=|c^DZ9qS|I8ir693x%MvigEyYC+q~QQmjt+0T6M|(fl|0!d$B&Z%{nhAZPSRY zs97geQRei>6KxFbPskYVL+nHqHf*El^}eAm6KqPb_1NdzUQE}a&MV#H57Y`!!iM|_ z;_zyo&WY1jFg-VZ2kuD$8T7OGEEzt&YEvn{jO_%-u;jfB2EXIWVb`YoCPGV!ldyI8 zf_wcDykRlwN`#^ImRv?hfaKm`nH^5TL{dktrRF;Yy z^@TtKUlMBBt!0&g2C)>Y;|g>x>GAf+c7HWh?yD8V8rBuedKqqnvKUk5ItZ`+vq7zdg84|n#ge1&G`LWB1-}) z&Rj^sYm7ZG*-`3}gQO>BV@P7ptE-OXQd&$AZ?N}7LHfa~8j4!T=mEwkn4<0DkU3m6 zJ&ElYq#eu%D3?4cxGc!C z@C%Sl)#ZY^FiZ}tIwda!vp7VsB)60ty`;{eMSFb9>uacGHUSINtU(Lzcmm{# zW|k&Xa-ml?6%b8~lPjJCzNDC^mgTz4KWAE_Keu|pryq_{V0Zi{>*vyy*z_4{?pH6cGfX}k5EwK* zmG7S)nJFy2jM7(0#(rAQFqCo5g;g_V7q6m^gWDG!pN2ip?baKy*QH-Q48#sab5s={ zLk+{ZZ`SV7gUKot+beyFc@Id8 z-R0;rEtRKQOR@kseId>ScrJ?hnnarZqZ& znHII-kQhM8WFDE4-!2PcCy%F?CoqE83gF;(bR;_Sx{LFfYC5vFVe;Q9IazA+UbJHC zP@9%6etSjrQen2nd^pTmRZB1Vwce4Qa;%c@`JY9&8JKGg2tsR~&vb-ZzhKIzA7Q{d z_o{7b!YVZTOs`-TQK$2T$#UJ8u)*&E$`~HC6>1W;>Y#fagUD*AgIP5$AoX5?k@_dx z6)+UidjWL|c7UP=c6~5}Gc%i-HCTvc&H*w25tIfb4It9dU*Fv~L)4Oe!G^5<2TOvv zy=~(&4V;gSHYJ9s0a+y zWjxEdt^(ejv;IMvbvCsj_u?_8pt*qB|7Wo#TrkcFqJECm$XL$uU=NJ2-uv9GfNt+g zX+LE;+&e;(5~vZEkcb_A*@6pVC^heE`17-`i!X>xT0e8}h4 zKh^y-J>bsLFC4^U*+4Tv_ED@kWKKU45&sB23sRA@0bi97XjTDVpj_8;r?My&a;D;ky#Q>K!u|#Shl`Z!^Mvxe~&YA3W6kr{kGkqn4s-3f=w{!go~Z# zNRI>9c7TR}m}gCtVHzgiBce*UF+j13+7Sr+1e^&#T^M3aFEsVziGJy?16aOCfjR5* zzwiry9xWz%Xm|nW2SFnI%Z?ov#R(gUk+}CjK+WLUpg?b9gUl+Oru=r;0n&?T zd9?%#^w?N+kM2)L103Kf5&Z3dijY1)Qekt$O5>cUZ85_BYfj)vL zd;vb{0oQvgoQS*=uHCLt)3j8-w7s^XF`VVA?d-dKR}W z`!x}*_1|#>e0FF3Ba99iw?~BU=i9sm#ew_!`<#KMwqMx~{oT_nprth3U&9x3Nl<6L zdKJH)1%HOd_f%u%QLHKX;L9OkAOqJYBF58U(RJqgMeUldg;K#HkTm?2aMQILk&57Gd#DmiM7{5Jq!asc;4&bDJqEO*Y|-u*}wBM?Iis)-t0)bNYuM_1V}%n3duN+@{w!^7Q5(i z&6yKh4%@vqPW0jk%*f^Kz_ri;!ZKrRD%(f^{SvMkY=Mfw6ACOD6F^h}M*zw1GSn{& zVY&x>N@LaB(WCw_ij$L*%Z!dIN>RY4zbgAhag~j%PZ-tym#m)~9MMoj$w^-jfw(S2 zUolNLqYW|#{w`9#BT^4wQpYzn8;Z}zvQAfl>t&1%fqNCWr_8i&@&6+<){CV~>4t?1 zdLH!DhB3NgII14N0-;F|^Pqy#~q!LLnUYyFXD^42;xQ zDf(5)P+Kr*wL);eLhu)qj&{!xCD+P7M?z$c+#Gh$f}!5YOW}?KxVwO5My%;#$y=jr zu-QPkv!;f%@NbNJD{7_JuQKqwXkO$RNvb{im;SnO6~z45N$64CjnVO@dBEOMfE(>7odcR24SfHC0J`yY6)lF zS5Ss(*qZ&rx&2WdfiIS1gLr%S;;~4DxdQVhEn3Zf^-ep;!5*1E02foG#}BsO<@vB5 zU7~cTl;8%0P#gcI_+6!30z6@VPqp+YNK+DagotiEs!3_X1%IUKd?Izz;RzZ~nBx#@ z@k&yq4-?|r6qMJ{5x76Rkt>W4=@mC2TLXTf9~F&D5r`tD5bn*OAdpZUjtY_=ueNg# z-z2wQH}ULQ1O56g#=#wV3h#{0>nN59T|G9fZvQ9?rf_BIDM!G&yk@ay0ZdHX2ZH7q z_;)=Nzx1>Vdq&$?=y>a|k5V=V$>PX*mSLM_rup$@zdrsG`jb^BA(!Us%lC)u5_09y z@f%c9Z zk!N${ILe#EWNzU=0BHDKY?~sbnVw<>v8qVn>yw-0@L$Quo)%<72^bxy^1d&-36|29 zTci3+qFhRVmc8lY`*hL1!o~ml>(KmgTU42g(Fld>YVuse+NI}%y^SK0N7sKHb=evu z6wGu&i3@j&waNx@dRiN7%`WA^66URFMSZQ%m!YlU5Vthy2R+k@%L6W4gxF|-_Pk5@ zzh580g%;}K5RnbUE3h)S$-W?d@7J{AuTC~3BB?qZ}Aq|fBHZ*3>%vf|XhaB$AiLn~F}O1+QaeVXTU zUkNsW7t}dE>G#W5hXauBM~BvXot3po>WN#5IVT=@CHvDu?o$_@KAx4QhO|qEja~Uk zJsNICTWku;9m!{ccvpM^Hd(Q3%XL>H%kR)SL=2icCU?R1l)(IdaRqH{(N87+-&z0v z58R+KO;n}R@I7*U)-E<>d8itzUFw^*Tem4yf2b`p1P4F`MC;3nl{mL508n~8nCl9wuoj{wVM4-wsd4?9{EP|2i--pv7l z?t@1sXMMDE3pM--8?CtFmTkG&AaER{_Coy@r>Jl@_#6v23<&*1X6RE%D>kX{kUrV_ z`Mn$0>)NgvB{#(vdRvHfO;Nbv6J>ou+J7`aG6&NKH<{}8HTj?gOJ}%cLwM``_?{)8 z*S?K{m9TpZrfXXmc+Cad$xn^oIgnXd%rRWtLtDWQBmt2XCkPe2$|hG$%_+EtmOs_w zQe|n$*3FoEcdybCKc4ZBpJ*#HLR)C2bi(t_uA-M;^ek(9&qcVfdGw~FgPpXIeRSKd zLLxCQhr};uqd$gE;MaMct(692zn0cYn_=aI6$L}6>{S5&L-9!_6Q6(rJ!YUT8@|0} z(0aX#=m_)hKoGQi6u*CnjcNFinfFm`*OlyqAfLaBWEilufAf<_${N?j<3nL5MIG^B zrc>A;%;ZawmYXfnTORSIvSj8wHtdKB4RYLlkwpG#OUp)xG{7l~33^37f*C;dVG}M* zV~UI5hwNaJ1$o;vOY2~%1zvLfVe`ApXIwGs-L*I#Vq9J%LEKZ^D`}5CLEO9uQ{D1m z%XsnPyW1nh;%Ok{?sOW*9Sqv{c!KpyJ{g+R3ZV_y6Z^su+c@{jdDaCqq#SAI#Gr6+U zh9>S>v>ArNyn3_MoZEx#Ba7@XCEYB5OB1q{Y6_@NFdzI0KQ`ciSp?Phk-5Wucon!p z=0Nz%3kh#`?QaMWSkl)Dmn^FKNj#&I)s$B!#-PhIpX)9HxpT0c;9J9`n{xp7Q@kv^ z;32YpClAc&i^J*b35JIH$9yPx@Rvi7 z<~kpVIGek6=+FMKrk%6+!kG544q6CCtH(T81++(~m7?r|K}<#s3JD`VgyC)i;5Hs& zmm=_Opr|^=_vVp&KcLM$B#jc}E#uGWaSKs9g;_uP$zdm=TW0w@AVtyE>9GN!=?Ze~ zbzSq-jSnn?Wq6(}*a~FfWta_v<0A}`(!qGM@&VV)r;i9NfD&@%8wmRmDvP2=I0hsY zRN(;WJL+oK*BSV;2s66DRCbaz-U#`}mQs0Ro%d~YU<9T|7 z_jEYz?F(_|%phZF{^yjz=hSF}g0CfVzN8jWt*_Lk;b*1PPo>_jaNDQe`fwkvi<=7J zGegC7G5JVtRc+F)(p=>vS*@9Ip~u(SpEY?F1im>_p*d~bK`sp_+@Tm@+~e?^D_E=Ihl=bz8j+w)8DKwsElBY5yVlGs2KmLitoTpFGzrJ<(3XOHqNj*08eW=FW>Ha0V|_ znlD3ToAU(9OOP|{;wxcbwvIelTxqFj4vi8P#7~wfozx1@<20Y>u*v!%WPZVoSJ<2V zm$#xmLnfO^U&*%b+&T*P(lE1_pIcs zEG?ECmw2z?>{vM&8EkfZe1=f_j!7$H2vtn^LH;W?rf5A@N1C93R+0E39z~`8Y)~VQ z7dZY^+JgM{)c@#06_t`V8ikTA8Z||L(56;3T$6_XdEnEzS?djDw*cyzETzy<1Ssf0sD!)`D2k7rRT*!KVUjVfk35 zW_s9h(Y8*EwJh>*x0Y{rGw?H_Hsb|x3h+H#{3McP>V1)AMbtOoR1RKlgO#UJ^s4Na z#T(5+yRI0s^zaaXd_8#oW+uQel9|A86Q&0HPhB>fojW&CC9rJ)tj4QQP*+g|DBTuu ze3*LzBk=J65cW0I@QuBU9m#;~2>A#ZgVabl=5u_~Q%i*eM1U5^SNO(p+3uh-(Kj|a zY68@VF#5E@pUtA`-)7*5X$#DkU~djqxKV@2_aq=;5rOCblS0I1$Wx0d2ry8=!xEfh zz)T?J4+r(H_G5q>Fl;En@;U|RlNcSqaDhIhjp6xx&RQISmKaTdrO|yPSS<2cr-NS| zSbTg;^u%WZVaM%#$H&LbyECh@b7u)l*wt<>+F2-vp#1<*T>+M*F-s2M)Zn89sTC3f zJ?SrE-#XjkZhxPg)+tMl##1N&)C5F}CjgS6_nJt0S5~Q^(rGmKfk@!qE5@jPeu!Go z1-_Frj^GzZOc+O__RwkaE1%fqho}%6a6mjJBMmq>dA;Q*$6 zkWO#nb3L6ybJYqiCeoV3_~=ri?h967+y5bRCqHBM$m&V$@X{& z+ZPls8z%8N0PTQP|A78=%}rv0>hEkZ@82s}n;Oqa-Bb~<2x1CkXlP>x5SYR_fZT(# zMBJM%x}Wb_5)RlNgJ6bGOVM@mPYRk;q#Y6#6}_)Oum=MA0KNu*ne6{zn^xKL0XSdF zqFXI5xbTaGl**sd<`l82iXOd2paUQ_({*`tzL6IHrwD7lJA5$|2K0(1ykQ8y%biG%0H%#4UX5Wt) zurCURdxRc%yPI>*ifB&}&>_P>xq_!h7mlGn zON4>?u(u9QkY)}b5y0i&3b6P-&|cs?utNop?#q#~BOdgA%qPhgA){)ae`F5YoIy(G zJwgiv!3N)y6CcV?49b2X0vXn=)x6WoIQJ9`F?|3Ug*aS5DDy4ILJweXVcnBNN;`U^ zD7y+06sA)kF^(XLbUlRflLNIsSEBVVIu<`x_L$oDAsz(g`Z|3KG$O&K8b6%aHsZGR zCBp-}jalP~`gX^0QsoM;uBYR)Lf+{h?m-ng zJs^YI=v5Bo3kQcS}Z0SQVB=(KIgDeKt8?BddI0@x}PsN1V>n zxD9Yov1sRtPcdDg)HSxZS;UJihEZjxTccz_zBlKa9fqV)gD{fm)k)o(d=<~cgxv*4 z}x`y+6WllQTzWbqp{~Y!{$_Zhs*-3Uow!8c}v&x zl^}N3DF1f~{UBX6@XEwn}_&o*I07@FRWiKNU~C+DJbt( zg_(N=N-dVp z6Ge}7@hU=gWJoAF*E(}a$qN*lndV=xaxvt0?-tbG;?;Ev)Tx-55M_|{n1xCG z?r%J3tWI?M+jeidQ}o`&mNr~3VJFb`7u$mV}_*ukxzATd?TghKZgfUjT5o;L`nClk)X7@*ld!bj$ z3BNIqGzY+O!r z`~wHQq+kFN*QBUwAAY9gIC8pV>-%+rY{=_n7XVF|UK(}1{kt=jf4LDW5#2xHx86Ug zW`c~?fN#!M&Li)FU4n*9a@K@V_OnayNeuh*`bVDz2hN3ek(ea22O9uunm;L7Ig>HF zSw{Pv@~}@y+>Y=LS`YkInbRLlH`O&AREs7H8+0 zX$hec{*J^}n^UiEx1r8S^Wu%g7qD?lR}mWhK4}FJp+K91ZOW)`jquuO>WPDp+OF#t z2@EEBEZ+C+Kx??NF61w7dPI0K(t4i+9 zG-tJ*?F%R}?fsy{*XfmsrQ>izM`U^NXv4#m3@W^3eLnbliWaa1p6!dp^yu%} zA%CFqK^0Ko_Q}snBpXy<;A<6MHnw6Ec;X_!&P->1P8|m0iRh8lwWu5Zdw2qmA3wzt zIqAKvcq$%C_C?4Yh7(PZ6%5Ol;x&&8m7Pslg?MDyix#s|ZqkJCXz_52uw~(L&4*aX z5Ds8bo@Jc@G#o72h|u8EQ)y)~DBQwu^BFj)^Hep7~@g zF@%5pbrDdGpNO3`z1s0qw8&?_wz1SI)H!=nCNhR*BOUg%{)KkeMgB3@(|sx4Pqr(% zaU$h`Cjz0$JU*nwM1^zVYLxvdqQSEeM*w))fN>Z1*JY!w?lR_blGjaB`JfMKKXuB! z&wYfc#2WxR3U8(ODQRs8s&P4{q(!3toV%zl=1;pqMv@g}6^$#OyAm+Ufp=|yr{^YK zGSHm6Hko`6lw8w*kuSi@eNWjzkVlQcI}{|dE`v5hcjN(xRi+hR;*hMEE|E;t2Om>u zbdI3q6R3_`e7q=?`jxK0&=)D{IQ{{;xRoS{vVgu1nXBhHyTdVl`m2jL1u$rlVSQ#L8P;Sjppy!MJ*R<9W^|NuHyq#Z2Zyb&5`nvS<=vT zy}z(9`iK0IEawM}){eMi*fBrSCehWy{+81EK6Vd%2+yPGd=c#RxKp?3eYnMhfnXGm z@6iH>U6AHu+eGx!I&Ob7rBiNmR2}~v?zlp^_nfty0K?ZN8)hBBd0&?L~Bb+#C zSezxa9wY!Px^O-KPr}&cXo8P);R#otF$f?K(l_mWbPZS7mDk+@OaS;FhQ@_4UVmE0 zCB{pwCjkqb4@eq^Gw<}UcMB*TMc_p zHQd5n+pDh!?IjYoDb&IYQ;yv4^opOUwpMV(&gV~jG$Bt3As6}Um-TQ1VHU+)_+m-%ISgVX@2ULGS4!-$ZOwsTXjzs}$tqMzg@tkeuu*bexf!VHDtCd>zlFUOfqt#K0c1MLS4= zQ2@Q7C$Cj$VDN)fuC@eUuXOz_4Q(G#MsHE`Lix4?bneG|_nBZB&L;SaU%N zp&uc-=i~T;_1F3O3m3Ji2xB6;7N`I&XMQdwb7$84#7iH5n&L)A77-m?Z;bVEsWAxl3>HpwhZ3f z;guaU2z8;2O>ULtQM4@~8o)sa5KO$!&BF4S(Vtu??H6!DKzSH|WWlEWn&IOV)PWai zef2YNDVykng~s(=H9s#ve1+OK?dF!(Biz4;arSoG@18j^!4QIaQICmA&lg?p09Z%4AcbFI4yxXP=rW>Y?h@dTn{AezU7Cz<#mwq*6REmD#`P* z(gv8)*ZwQ})(z4G3|Nvl#6DbE;!S&R@#ciI{?bZ($ML(%e<1BKCdXzi$2QJeoWo?G0+%W85jZ^C zSFu!mWdFPbh*d7o?za89t8STogHMLYc@%N~L-oiQBS^}6K_vkV1pkwM>bP?Q3P}o$ z-A@R}Zs~pc)*t|A9>DtlEP|d?GN}+Gk|G)!f%0d$JdDCIb$D@9s^*Ig%rTfz@#Z3< z7PVZlk``K(z~$0fQ4z2foH9+st;?R%H-NN0@ z;YJU_OAq%Krd(uEiDfKpd@3Mr53${$wXWb|OIu#{nYu#?D#NEeA&7 zeZJqde7Crah&lPCq9;}!Qa`Xjw(F1$qANWX7M8Xw*dz9C9C#Nf3>k1U*L2_PxhUV}I*Q!4D{UeV1Q38iOaf5FkVkm{lIqHmUKrIr zfE-D3@MW#!g|sV2M8$DqAn+U-j}_$C@$5?KA6?px*;j{k5h)kIlMxo8$&=8pk&kc| z%~|F|zXi+*=y7V(*5RvZS-Snx^_zWe24U!-U#j%As%!{`$~bhTe8 zx$&~7R7cOg)|T+4LT@G5qo{ z&^7uDQSQ-lo>^3C(;@?a2}H;z0ssLbj16X;qYUtotb=^dM~KD+OA&utxo?ht)3!66 z@ePk6-Wme{PlJ(n^2@wod)rhI*8~lw-7;WM!a!}xQw4fn66IQ!97&S>jvX1!SK!ne zbEP(!ft`dpWIc-40pJcGUsMp=Fq!0lFG!N^-AnBdZXV;~9@=|CM29IjF)RC95D~S+ zA&cZ)wjKCT?#KneZve*gV>eB66NnGO@`*yed-^^u$)OEj=fwHYZOIwspF~O``Z*qiz9vm?v{xSHz6+a7mnT8axOn~G3y2gxC^x{dn)-$=AfeRRJ7i>xEhyG2-jyc;*6}b-C7`c#87YWmy_`iR0NX=}k^FQZA z24TJ_noiHFXOA?k&>#)YRN_8!%WHL~;a2k)b^mp7I6wkJ-qpk?qOqh!RL)W4;*F`P z_LDDUhuJiCV9og6&m4`;KlE|&srS8_Q~A#GPZs-tA{Q5l_;+S?6NOKi@g9G+M(ga} zL0i}-5w%&UJ=)kH{u#Jy=q55%6%c|)&$mw%Lzx*x`o}|DRnT$r+Hk7 ze;;vEoA;m113k)nPNN@>?8&olu9Gyfmn&bELyO09Vneb+rLgn0cd?WvhwUhT1q^iH zLR6Zuu~EfdtC!^qS*@rrw>=scx(~iQd2oFqp5}vs@u<9l{tyv=!4N%2gGk9(WBD;J z?N|3G9!(lepgVl;QB%o1v8@7WoRwrA29+q8ZDJ5LHqM-dHhF3gHFBB1Hg#(^N)-l) zMIYyeJ`xG5zxKNV#q3W{jSajoI=kv0Fdeg=vgOML@JJ%Dr^{rLX9i>yzYmtB>SFQ! zTaS!x^OXG|I(}L9{vYC!$UA0A{f6vImNw>qo7wMk>#1=-IvR61EDWl&-2GNG@sgq| zUONt5NWJoNcHzvKIlDhPL6JvVeK^?0AzKvIvYD$*ojEBS`nr8|;kZfd&H;DU*_d8C zw-g6PI6G&Z(W5NBGt+{@d#*QLfah(0Bczc;jZx$SKogdNaqi8I*;X^b1VPi z^R+&WwqC#nZeFwDiinxh2hWU_Wltj_*id*I;BUmPt%I)a*!?_oEy*lGmGm?0c;A1n zhwaKO`E5=^G9|=e7Mmk}f;c6Vii5Bgn8mv|l-d*=E(#@GHt1mGa6gXR-4q=KJ4F!4 z73;4iMWe#Tdrvs9O5bP3Ogv8UkY=5Jz*+gzig4ek?C0uSS!S20$vv5NwFYv27vCnP z8D8wn&F0_Aqs^=i3j<*vm_CKWWLuwE0LO55N8}&mnz2C5{E&+MY=&Syl|>Ui0Kfd@ zN)7v3!X;@emnnq>&&;Zz2UT@r9=Xoj&*Xnkx0Z+zggOm@jnn;~pR}emE6i$CFIO}F zpkGSloa!62U5cVLha3Ae@I>`VQtr^nh`q6LVQ;=>ZeoLCNx%;R4hS5#VAQiDSy5Vn zwd%DL0S&wq|0ys+1SNT9BXbHRy8Q^;*WrtDli05exE_dYBV&G$u;+e*SLOYfZI6

Z(E9;^VRcts4!%em| z4XBQe_EbjxApO{qE~^zyx6Lpl=8LLfI$HA=sNMvh!id-B2~`@-4I)Y`v~^1q2%{v@ zy437!UoD5J=s-hA*>1K8!J9i!T5;$bGbVJgw`Z`cI4@Dr_GePayJWS@zU066zYflQ z1V1HeYM1>|epzrEl4MFW*CGX9sz2LK7?%z%zvDc6R>S`v^$2P8s`Z}W&=($}cRb2E zTamN&70%u-4=kUVZ8|uOV4{4bsyJ&)T0#Gqy7u_>cl3beBQ>gUM}|PG?>+1hnu#S7 zm2~)pRrp%A2^fQ+w;j4VV|Z@2n~iI2KNKiK_%wxjC8KMh7=_ika&h(^yoxa!;(eF? znp2Q+_E^SVWL8)J>?&rsC(b$Ym|mn)A1=i;D|=+WEO5<|47&O^?Y&ObJS zZ;$jkf}EZ}NG3*u3@p z@@?%)ar9tQT}C?B{KW46h|P(&X9WHhJ*ixaIdIQERknuU;b)2;n|}}c8892g62U`> z(j(aOnA*~=oqUc(?!q{}LF;Gf6iL!?S3j58=T?X!Saa7laAUTAb&+v+A#_&mx zIzmpUPi8z4DR>9BTP$%=ykXxv`vAp!HXgB_3W8JCZDWVUi9vq%XA^mfO@oud*-NGhXAOOvNiNxEin>$JcmJq8j;NcG& zw5v7=qq}8=MY1mhS3^dgV_oXXq4i6SPy02XPx_q_MF;FEM|A_ zp|s1Xn9H^9oNZk-FRKmx6`#oracVXN8e|BBf{&# zA7P|44hZU)CVkB>*2{Xio&tM6jK_IPtz{t+zW3p2)3YxHF)ptg^4|87m^aW)Dol8{ zX+nGVCE;!Y5#xcPogsI{>BN)sB7b>!@NeAbqrN^E6v|>!uZ%zO&DCCVkA-#kmjv1vxr^OHcE=3^Jls0t|k;68zF=240+>Q zuaD}`rC25SNYIdz6q$Hm+(nx_qv&JFJW(axh{@1w7ko?Xn|^K5^FjYcLTNqO`^}4_ za%h1`#D_$$xDMEsmLv#@-O7wDW4|7rnfIXq(VgD4d^jI6L#j>tW7x;@$3z~-pYsU` zg*!tHUtZE^*@SXd&)Ev3rKZxU_{KuXLcmQ$FY8euafJ>W{1++e?_C!jCh;(q2rE8{ zpC9W-q#{fu_|JS9KlJ6s1_HMHrr|F(h1L;KipG_cBQov-!sPL*#b8xFJ)Z~6_q=~4wcHzKc ztfnqV>XCdFni2UR_Hx5hig4m%eT@sc z6qlyXhTt|;nNG#f29v_$kM;Aj(?UjdVmU&8tIu}__Yizh8+7+=xXDFCJ;2$yw)`cr!tL7XH@ECUdv1V*6# z9J6&Cvmv4-?oy)AHC7U~ENTbqXAFk{ZXio`j^9NhLHiL|kzWfnm70bgdAfQ$u3^@ktW&?8d6 zc$EwRlc*d4q@|S&S4-qbEpE0oCIg3VjPAM^o{*nS@lyV|>QlU(&Is)XE^5QQF^W9yrBRU{SA|gJ&15g8mT=hCC*UpUyYhvWp7ipF{0AI$*DuExw zTTgd?kqh&k%RXpaPz0{E`7-V+gFnEM@%~fGfdG&ti`X6HwU)zqad*jtf$ne_Byowl zcjk4TsbS74d=B~%R)8&M2YPP#8$3`N7dAn2>c%?l0|eByj4dgVC8_%BSM*(tL{l5O%Hl#ioG&KAx?nGFp{vNpfz7ol#dK%E_6Re6>T!IVC;>*A$6|$t(o|3(* zUm@GFxCO!_whrw5yqD)r2T4{gwzirfAa2>>|n7RQ*1I5iIc9fp_j0=yDcq zEZtp@ImOM8uO8!kDp#QTQ?AXhczV~&obF6us0q^epv}Wg6`3ZrIT6bpat5ENh2D96 zWf{^LS0BO@?V<8|7IR|##w4|Jj~~~DBSX-B3#6C4W|b+P&I1)r&iwVF- z^L61Y5j7V$jJ{(+btf@C6fRe3*W=CyLyh^Rfx+s`wWos(m>TD&E_& zboR}*nfmeu&!%x|@Ch39MfM0$iWBu&D;<`{R+=S(GpgkSY^AYB3%cJSqPq0<&*A+` z{hQKM@SjQ#p;(>MrXolJP9KRO6+!M(NWY$proCOkH1NASo5L#P&fiXrjG_X9Z0kwZ zq1on8Qd-dBf`a3v!&vMoD4JMH4&9-tsIC1l^X{)hVgx>=p87z(S-M9I+OfI!jXFXQ>&*s|f7etH7nVLvp9oclFwtL#_j-y06TuO)OjPxMN)y}Y<1%-OBq4vB;vx9|yAZ`?^` zWWK!xORe&}2K|bCQC8B@maAly2UkT;^NK`w(Y%=|(rMwa4@-qU1(-h7xj(~ppE0l) z&PbTCxe8#%KP!Ehn_4kw|D0DhI2VZYcbo__SAXGe$73ls|{k8Apnt3g(+E(N=#f@Ww?y(9MHtSR*;d{fT@loTw^ z3Dxlv17{09!yXr<7yc?VW@8b%@N(yUNS=LU=-vIp$G6`-P1UHN4DX}GsYx}@j41Zq z_V}A(V%Jw1a-;`?5RIYnxnvp##{^5b8DYj+ru(rM${Nu@!*cv5fee55R=H>AQmnM1 zRtLo=UHj(PZC!%~)UUo*8VJ6lAi$`&+i@elN*O|*yy!VDNi@1rb}81p;wkR=CPYOH zS4-t!$>vnNV%3q^(%(LLhINZ`@^~ZSy~1?9{jc0rj+O%OgJzdUUG#H()4+8No?7y) z4=uM2<-LhO#!9AiJmDLZPmOl8@~76ti{ql(0ga* zOq_l8{%xp6_(nficGxLn`Iz7Hth?_O9~csQ<30ua%*vRZPX;#T&$7q*s;a6S$fRa< z%#Aqm74IfK8U|=AZZ-|V#}=n-FDd^KEn_}Duy0@R<=aiS?+mZh_OssC=qzaql&vE< zm7d17o(O`WyTUf~KCt1p`AVu{!RquYe?}+NIaITG@_Xe)`w$afy*Jg0!oF_H-@_G^TQT?yF81 z=osoD1O~j{n|}`dB4QpsW?Dp}!^>~Cgr^0jX@O&<%+bgQJ|aCo`GV^ac1cW# zoFfh8jL~E;VbOc~>vk!em4U;f+8|2ec8c_N#G+9$h;Tk-M*Y^0RwXJWJ%9r17UL3> znaU&&5%o_ibjNtu`fy@=x~-@FBV%r+SPkxLKl^sj=N#Q&m1dlxHG_EZQ4IX8Qk5h$ z!j+%A`%dX@djIsory+%Ywp2Tj#l-B|M>G8$qk1~J8j*I@B!p9r5Es&_my%fa9akER^SF+B!6@;06{O~9kT4(hL2)F@%DW=kW1O&sxtH?Dn1ZJ{q5Bx za6(SzB(SJME*qF^@QTHD_={@$Izu^_SZ zGFZW5>QcxeNyNyZfjp`EOzMjkXJc2jE$g*!8nUwY-EUlVJC|0(nD#eU-l<^LYn>LF z>G7#jKbUi0OI_W4CDML~Ij(gF0={L7E>a|8I91JA2{ZWJpxQgAfwXppE4m z9^xg5NzLs)rQdyfY6Pi0_v_xu8I!k$F8QzP&(1uNCXky_e=zQ0_3P;m!sMRi3m%Z~ zg=dRRKChAs|EnWAc~*<@2)5%G zl3Gutk-y=NzKqfNEd^2otyY+dZP_4Xuzq9WNc!`|hl{*qqs-<%_LeqL~MzSL%gP6btH4oULhz2&k!djX*LBR%5^?XXEQ+{(2Y; z3;uZkYec@UOz9YL}yePW8!9No+AGh8lb1qk7)^Cly-R zMoLq=vGD39Iuv_J;XE)bma29M@}Ud7N=vMyBuCi>v|z@LBz$ic{#CW`HDd-ul}tS#6$@N-LTuQ6e~6$1|Jl81{$ITj=%xYqs1)BQYqymq&hw!CUJW3W;{)VdFn zn5)_0YfVo1O2}H9JAR&$q|NV1LeH&KVE>hi$qjrly)4)j=X-8Qw(&@?swcR#xIC3q zP48LXo*nx~%3M{I6sJ9tRR(QJXH9-pZ(`)(slg<_$RpWh6YTNrlToBgy|PVB6xsV{ zh8E=Z6O?h)lk9?_PI(}77k_b-_Kp%MNItc4fl+3QHwVvCFPb@1R1{g8 zi1sd&QPxJiKQ%Jsqx0_-haL8+M2K9y2^umoK1xBnB!ld|G{3A@6q<>$;iqrAou(4A zE4M789+zS@oF%PkUzLMN7td&XFLW*_e>3g|!`^?as)=l%ve;YudA&~ZbpUu}DrToU z92LZ$Svmd0$g!i}9hf^3vA}L(j1<$C$(Q@p<-4f?3?FvZtU~+Kc+08dcxtUnYgbw+ zk&>z$YWupmn>PcbJnF7E?OAGZQEBt&L1*4BDIU`?PfeMKxA6|FVU15#j?A}PXM?LP zts!0{@NeME>H&b$}vQV^O6z&D6OR=ySS=IW%O_| z5;yg(#_A6DqMC*C@O1FaaUGW~qtCNN3YhvJoHB_CUCjkbkvA6UIUh``H_FQA_K(sQ zRp0`1*(vtb$5x}3bC|MHi_6KyK&kz89MDEAsvGeT51tMn;sUsr#(fB@FVRHbBiBF6 z?g|>ljS*kDJIc02>az}n+6>+0&9-%F8N#GH6*ki$1s&-OrQ2>#sW=-XTQi`P^V)?Qe9yYuSqztX8$Y@UMQx%@4t39`P29j8RD5 zG!-3FnN2S7qY)*7yAqfbSFy$&Z+aCRsTCn8<|*@gm^1Wof4vQ!B@d+(+w?GDDe?W) z7>NsuZaqMi)@a~OLop`(f{(XWA9*zqX)-H-VGr#XsaR&E9qa_XGQB0zQ1?!escJS| zdZt3#wPM-2w5`;I)&J>VZhc36mO-C;1$t^5ZKum}t@A=#nxp+QYU;r*oK*#nC_{`I zdU1fQB(>5s%KVsr?KyOyGxQLTjP1|1K7O#p^yW1ai^X;W+S>CG#@(akYplV7oqVN1 zkpQh~=+{iu`dIDTn@^7($mrx$^&nD`!?}NjGfq2G$9^!kq2bu%&e1VWn2CezWT9*K!!;rE?^j`Jfo95?$Ol^HPz+*txn$A=ETv>pplEWc~ggXP0I=OdpG6RC>Mm z<&z=p^f1-Xj*yY$J+@Bk;@{ciSUa3@ma#r4EZuq6ru;qJmCr{Dg3EuLZ2k5TpZ@my z_o7_evAr91(>zl!A6`CI@) zuH41o$L;X(SmUXbv&KTo22shuKl9ah9MO~KTSBU0y*0d4Z9SQ-9|oy)<}@e%GQ&Zn z32*@+Q!4TtiR_x|YujYxmiH zn_>VfhWS*Ui)XDYmP)29K>VY73+q_s`trPqJ6=*odr@np-&i3efI#`ixAl2_+&Fy) z7YT_-NcvpfTmq=G-IHXuA!C)EU*=lBkV`C zDcg1o9v1$J)104^S=@iO$Rd((xM(yyDbk&C>e?e&(Ed-Da~J*T+^kr$5^&yh2FbB@ zB~zTayhtLV_3O$3LQ{_S$zOoCb1>6Ucs1w4J_N+gemq@_m#Z_%D=6eQdxUdRsKIRkbIO3%ncC>YkmWX1*~anBvd*w)L*0v@~#yfmDO; zQgZAGXlmI6jvb}UliZ{!c^Kt&r~My+^`|{vPImU0?Tu(0h!3J!B8}~OX9nO;6h;x2 zO4;tC@Z`Ik;Pjl!T2FTKi$R9_!~&(rzl#eHZ2f`(8=|EJf7MG-|BUYVO0$hM7ZGUJ zl2JO#$QG3~q+gWa;eA+9083&~}VU+X?XDF-2(HT z`24fs%r`>+t3siVxDF;59cPEAllp2uUX-c#djJ^H-CHSfDvJT~T`c8H6!6J!QTUfF z)|ViF3CufZErw~HI~<|fjriCw&axfJR5PMEhLu53GX2Tu>#;WnrBR^0I#J36DI*}D zlLM?Pf1p?4fKZa)i)Ee`{d>3DGMiqbR-o+ZCpGI;v zRt)1(XiUiQOCGaS2Y=g&&y)fG@=BbW~8ums}~C{Fdhm0Kaj&MK4>7rseI`t zBD|Aj5%8WJ`Fu@Bg8Dor{<2}qsAo2NvW}Fk zH&UaOUvgdd*_O^hU*sKVOPVX}bb?&Q_%lhzW3On*B6GK$v8y-`JM)+FX=o{bVlyg4 zQ$-)HHJ#GEJOuXa_8mMQ>DGs-3D~q6eh&Lb&$=nKpGP+1C~`q$gRGx}I#(_f)rG6*#gN$`e`g7Q~Y5r&~DD$S}tKq=?K16i}VW zEDXR#7~m4CF7|1nW#5j#_)E!C!Jr8nT-lQ8c6jza%2R)E?zYS&bBg)9wBzoeThdoU zMVlPS!dV(5T;0n4mAP(m*%{Q#s*2j3t;X6dMsLdDX00RviHZ^xmJ)ozAh_hq!ja(E zH{By~W`Ybk4bY58`e~38DRyJ+Jn|HyRT#Bh4M38mTKQ$#gD5xfCu=tnFB@UCkel7O zy7q<5&};uG8{#Z01i2Y8oI?_XEO_YWDVKW7y4MaTpr?0Tje$+pEW4zD5j-cHtz>lO z@exaCbVK-_Q20+*!uNo9#)t?jEz^UUe}?VHvDVQE1b$vAvT3Jblq|OJ*OZZw8TBEwhknzbWZ5+TIv%xnhikO|DEWo@ zv!ux`idE!AqkFC^E4@n}w;O!h{q$%4k1S)0OuA+{l&lW;Lg*sc>){bL&}rA#m10vx zwh*`JYie)coE%?ATf%P^Uv9Zo1!dCc=J}3YE7AWxi~%|XcQ@+ZIKlX$>u>Y zxOfMOe}e+rotay?Uk=A>A!fA?nB#LY$<9T1JowQvyxd0;yTA4SWLnA@UK`c; zcprV;2pY>hu}a;QhWNr38jaN6ButKUN$$3WyRCGeoN-L&1k>rIKz$D`s`0GJQW69M zwW=gPuwJ^wP^e0Bipf6#J$7xpMg<1#T|&|3)k3!57OqmDB`0sFb_28!fW+> zeSUo(Coepda)SKQ8s(qF6!be!Jz$O@SfiV2&{(C@ICn4gea}>`?LDTqThzSvfm1e4 z`>;#@p~%$3CP^xLME1MDy`W@E!Q-=}JJ0oJ)>oA2`J7@>SLGtEKQyO0&t!fl^B&G6 z0x_9J&ukJ0Vk^=y2y%a%d=4&%f4t=PY%P)+S>95T;l;NcfQ`Dnc=!(;crS~~$j$d> z0bMWN2kpV0K$2f5{tz9m&7t1l_SFDs=Vj36gF2bX-{U`gI_l*(yA{bs%B0h)TS|8sD32BxQHqMjlc3?lxNY%NMrXT-H6v#R;hOgYZV^jeLx(iGP^Wo%SS0f^?#WnMfC_?pvcQ-bZlak82ukbywfc zpmi)s770nZmkQ2(G&6+DIvc_<9KN5tMEc1%HSnE(G(g<9YVEgt0>g zjgR;jdi%#YpH$Y@fBqLfkdO=?@*jSGvX1`d`a3gW@a9^q4j7ukGHA5Gd*5#*@m@}r z_Drrj)=3!I&gEAog)$r_x^?akv^vUK*uCzfL>|$BY2N8RqJ`6uf2>ItWA~6HELq+h}As3Mp?% zTt&$~NoQy>;(@dNfF==(eX@9m%(E8Y0ymnQMMMS*!;@q5PrCvD-?^*nD`8GY;GjU@yDZ7c@xA;fCJttU+l9)@;Ue*m>o)(fXcw#mhBLJcl1#7SF=DOmSS9_l zcd0a#m?n6jcD|!dI@ZZh7ob%Ix<@8n)l*7*CgVV4U_W7c_JqRY3X#;=Xhwx<+d7l% zlBdSpKZs&v2j@cZgSA>AdO~#p$?fW^Q(p4HeqKV0ef4Sj0icYNC8lv*cOc1m<76#v zXky&NLu4OQdz?sE)9pmG@iizYf}jlf05^wfxw%klFnYCHoVfa!Alz)fSJjsOI;XDf z1XZ}Idq35gAN8C+E7slCGdgWy4t)c(XD#eb7uoZD>gkS$&E-0e&3pR_#d?o|yXcS= zyC#B7QNyr87*aF&MK!Z(kMQd2LV`lf<$r1Y1+ZL-;BitenGVps3VJ5{7m-35uYUfp zA*Ru-dZp8ysbi5mVx>71B42{*Xli(ye`P!#_8nQnTVXnxTtA#A56ced6m^nsu0Nt4 zGZTfWvCkcu34luyK`@odiuu{%#i3naGmqU7f&00mixmz!#o4LW_qys!><-4&U)ItD zzh@M;k=V=o4#Evmd_Up~@p4pel2=0==ZS^5T`w@lY0XrKey9iON$_`33;{B?(I-W7 zbz|i-)lVdzPEAcWeNf-{57WjAP@{rN)N3xc_zc;))Sb<#8z@`k4t=~1%OqS|RM)NX z+w`mw2c}s7|IG5g6+VoAc4m^KJ(z*9O*+eg+D}>|!8wCN1+z4gh>|2|)!e{KS7ql* zzi!sGqW~m3mgZdT!npF*4}26;TS;L! zNatUC9m1Vtqf9#{}6VEp>$B4!D&PzXLsJqpHA*BBzLK z1b#}*NEmRf#$O!?K`15%|5u0^)OmQ{>6LrRRZluB}QL`1|pfL-I~tdl$qH0{N>Ms>=v zsjAK@5rh(KOT|PAR2#KCVDN9W`HW47|N9ELfYLn*f*ZbgE^Uge#?%AElwrnQ z-*=vX6Ai_5b-1rUE(JALT2fNw6JVYJH@rwhXIy=agaYgU0@;eM$8r4N%OUaajsIvM zu>qxbZFkv7jaRY}&hZ-v1|EFzSK$(@xGC~jqdoe7KfPMB5~2$s$yT%!+2x2W?9oz6 zJ`;zlTYL!6A^8UByg=*6w7JYFe4$2_0PeSoYOtQ0zmMH-$CF5?sDWQnvk}>%vJkS5 zgpPG-n~?K+yrc~ykWYX;yXPGG7W{YVv~Co+JAQ4#4da!e)!~uHps~)y8d#KpMWo|! zg&DTYxpVL_o4&1s)AycT&|hBrgv}?uD!9UlREnk>Hec@Yw*iIE2|}c+J2M+8%}@~` zhE+dRy*seBIV$3iQl3{5kYD9xGKX>3Zi~2cZ!T~lh@OUs3aC3@PER{&4 zfeR$BkKIlOR+j*fm!C6(6eb*Cg;T9uXn-*69%qku%0vJNCF|Bz0Q<(9gq4uwx6DXJ zrMK|4c~~Y5eZXm3e_h2adQ{+_v8w>YC0rq%NB3wgF~AEPwxjeS>rGx9jKPbdHjWs$ zxK0o?r&FzY-)!{A{nprYlf9n+12kRRd;o0o0f3|c*gjUa%7<}PV=r__8g|FiE>U{U zP&xs2&Va6nL&t`b`;=gW6Nx1nEvkg!{g~F5%dy?=_}d&ZMauW3N;Z=VT1JzA)rk|H z`tqchGI9cxg2*qEcoyT}6AUnD{YNh}m|H`X!ySAN_1b_LeM7G;<-7j)an>F%hR-G} z@F=bNJwhPSx;mKNC(*m83uI5pAAW_EsWT2Zf#LrgyL(%;0Y~qxF^i@BUjg$b7QGea z4gm0dBk=}r$BE?Fw`Pr&&zsZ>;R1y-YhH)dPyeJuV0_rsdrWOYZ8+Ex70u+_1e{Zz zVJ{sU&n)CGXJL)j7P^p11vClsGk~de8&MDr=z|2I%OgK21`?KjIBf?tXD1eheQlP- zgnRQv-S)gbAFZ}xoE3J*d@eja0t17pX|Zz)uQqSS>S|=?hhRzan4mrDk4>0$K+`bq z+BK!d5286gZT;iej|NmcKEo6{6((;bt113{YinNk%58HQl@P*k?nwnhrKp!kBQGz1 z2GkuLK9`<^@ikg!ei?I1cX6qr90D}J#^oKL==eXg+1{pJ$LocM>dSMeEO?))F#0B6 zeAXiRE!npV;LSJzN1F@G?rENDNx6UH>zg+6R)n6OGo{6%!ia%CG+Cih3G5#=04pCM z4dII&u3-zJ*er-**#dq5H)cD|FgtJoyv~yWJnVj>mtCU^z|EHq>u!jSq z7r5yFfk+eld5;c*pKSV=l1Am~Kei&7-tfnbPcoux%t%mg+Aly611jIhC?gBxV`-Mq z5an5AT*bLH?=$;wKmgkwessVejr~Ujuo&Pke+g&PuK70N1Yz>|(nzB%98H!l+OFnuNQ-PmRS4&`!l+CjN%Jds=eA3~0 z(E%`=0op6|HTJjcWAJHpibq&vHW4kYBw&-g_R0qEa;%4_StwJxr--<5A&!aJmrNrI zhi-4#-nZKJB#ZG>i|!5aB}8kf+}9^a+e!Q3H-HbvA=rX>EIoF-6~(#-oct-Dgu$^p zIiDO{yey=jAlUUded-`qA#Pev9o+10(`9rud}jmyeiC)oFx?T_yLbc$nEWCse-f~{ z@_(PLq?1;v*VjuXdZWOsxUAr3nLBbue13syFSBWk#>a<=RROF(WPG2>?uuO z{iFwNIQ(&${NW^8VchyhLj73TjnZa~h{zrWH@)rK0Tg~yC=${nT++~)cWF-?B+=kV zMh}e3VkUY0?O-2!7OKkmyNjpJj{sFiS8scEOr1lh4q9!A6pD@m|> zUqY1y9lyuBJJId+U6}~i-#dJJ)<|pkZb37Jjk~wQ{GA;y+3Dz?ZJ}97`LM#2zB_&( zN}X2;#D`@mV_y*2mcUr&=lL~l6%qwPCcO7PuU4?5u#PG^aYT^++TSrO9W+8 zQmiE9eUyUF34)lRc=m)E8ikc5mkcs*Dg!NbMq@Py(6;z*($C{jq>O!)ItxP;>6UC- zoGhQw{U9eBkQ<@bsqV5UIc`P`tHZjn#3{Fi7n1X?O-!0tDfefj#PnkOceO??@MGFl zGJTViIPTEe@OUJ2A?>j5i`475vf+hXMBwu^xRmeC!(k-~%i5qy(Rh9>^(KCmA}ZoL zMzRa}ywX~IMgqi^@En7L$SPe4;oclm6A2uc|9v&6D%Ctwu(Njc41uZXi{13?{nDS; zY(&90Q#BmPEN~MF66X=?ao5R_vhnOE@$jOFJmLJ2D^70{zaLB(n)UokCWP(l zh3U1V{hSd5a{AyY{5-pH*5nriI*#A^qWMV9ia(ybJk@f7Que97y*+fhAE$B3i+kIZ zoEQW{+~DPmx0HdGyN3;*@}}o)CmYbe9d!vumovKNMo?+pV!}Tbwf6pD@+aK6YN5{O zqSiW&k4XXXf1_?EbCZ*ohBk5Y%az#|m?s~WB48Ym0n%0wIv8FY<$~&S{23DcP z#oCL{kSj>*>ucRB_k}_IM~h-~$P*)Uzj?u{;e~&B?aL$b^KUUjBdXzk;Ht&JHksEQ z8+3Ubw^Vgy8oO zOZH3%{_eo~t$)H!9yuM~Q#e+it_bOTk`vy2EDY&eA2JcYfdMfJRlBu(xo0!ikvK~+ ztKN{*NS!76?&4>u1<`NGX=L|QhDi_z5ZHV%+_jN#|Yx?pND$>tpI^WQ=m` z%Ofe{&i$-3&9SP;uhoD^S$pB|=n?E()wrOwRoxzIZI$jFUf}q&(_jqk2%i5m-o=Bv zp9*v~AP5y2M@1Lk5j$b90k!H@Cvbqg&2i_9qc`3gYj+8zA&L(^$!3N%vwDWE{U+$4 zhuYaE4<%r>`2frpyfty>x(G0rou*hhE8V&u&%v|%z4@3J>a>`&crDDnvHl7@qyGC( z@8<_5X8FuM3x+sDE7P}-suBk?AA4RcmFvi;%cc`yXPH(F6>rY}qZKMv2nXK~KR8j@ zNxLRBMQu3*$G1>HB5RVGrtt5k5p6EJw7`dd^qrS3X?!K0mFG)Q#F}*{c_fk>KnDlZ z83nN)9%xbOhbMdzobd_qRyl}GJ@RhJE<9Ln`7n2KF7!QbC^pP7XYm2+6JS6~YAlTS z3!DKeh38OX>&zR_zKN_PBMnl}#4<3ab1VdVL7WGfBQ=wk>MTi7XO#y=F)ak;k)k%n zi57GpxjiJ8@ltY-yxtt|G%ouw_>nQr%E5r+&w?WN@p9fD1Jdb~)}S(Wn=A-{?H1cT z25TVftt)!_o@WXs6(#Yc7`^>8Qf- z4pXZoGB98gF9D7+XIo*#A!YqBw%Uun*>WMn&7TN1>i`9G`77qTV`{D{pl0Tlx^ zu{IxoH?O%sDU>bS$Pdm_wH+c^u_E&Tj)Of^wm7l4_0EF;SVnVZ=3vh9j?pWrzRnJK zzLE7AFu7p3krR%Jo-9r}l{TwO^C18ZI1nZ^k??GATggpS(5bG8s6p1sspZ_tJG1%iz4eOYcv-fw-1s!!^fZ4@YDq51RtSt7PT zHQt4$M|QWPCvt+o3@M1h#Wwr$vqEyLi|4eSii!;X7-vyV)07lZ| zXdZ#COx6gv$cbRqo7I|W>?^y$bNIxd6F8;;zVFOXSZu`@R}u(w01XQ_YbMj2L#Icp z_N~BbtvX}MgeRf<r%4~0pK&XA&+0- z`K$+fPFrB|kd{Rcl8MEWyx$3_J+msBPxfHx4*aeqa@^>Ge`^u?QB!2}Th!7SV=g-x zC@TJ8*n}qdtF#TOO*&!3kj@i%?4IpTm-QDJFDZB!jt0uC9hsNNc1L|_Jsfhe9NO+t zHY253D$m5H@XimR0tD}aV6LRrx>~&O^?GNJU#fHtc<@FwnH1y7B%aW-3MV|REslPRy9|j1u|G*!x-Lwfre$_SO(2sOu@Sf}H zocAn$Ei!`%`#BYjv=k{R2;T??12`}MR8(`D8pE|v#h)p}V%!v2c(aSxpaNl%5SM$9 zWaUDEW~BxC5=aQR-?uNc`Or{B79$AlWK znyF*5lYbhKq>$tr#6G|QtlCIuhG_zJ%+UyM@}GtqYGObUkU-MC02z?DDheP*Y`i!$lbw} z?1>L)9_o9}Xge0v&;dTYRX*_iA_~t|Svwx-M}gD8@~3&pkKK0Ddo=M`U^F#ObD+r_ zJ_mRAOA>_YD6`m=Xm4D)t=sVIOyC{z=hq|r+Eo9m38V)U`Q^=}U4KrwBBu?__ajUi zGprCL2Yng@AtJzAkqC6eNSKgRz0)#<{kMcJviwMZ9;Ff5s{v?T8Lp7i&I`%t!GwJ#gEWL@kl7ZIR# z-F_)7O9GO8H-E$_S`=Y|fEjYfpQ)?+ALF~qX3?j(*uD&&*rd$_7Z82nX7yQm9!c8) zisXyaUozfEFd~c}Ziy`7_M+~voJas~#4=D7$JXH!*_GVMoi)kd96}w_XLw;E`n~WU zmu%eDTXz?bl-s}~E2prg*rI^$Jdbf;&H-o>5)F$|QPrAJ9VdY4ThQa?Nf`Z8xs(ck z9v=59%JDiV&$$YBykX+nzh@hIBQw9%WT!Ig%+8hVJej)sfx^vRIWmg+bvGoqgoGx6 zhN#A-D4f+emoC-;jY_&s}JSjJxglD;ynm@ zr|&IhCUQTj#38A9WD)ed3x5IiWv)a4>~k>+Cnnmun%&sY{jVU3BZL*i?KK80PKS|s zg(H&t)X{IoH--1F$Eij(+x&KZxrbP8NGAA~ep%xF+3t4uV=vRDp;x!$5qG?WoPC0K z{hOKlD=0D@3_dFF=%1`H==gRqN)O~-pg#!c%9eX79g6gkPP67-+b6u(I6GK z>9vkl%j1Hu2&RcHEJ3KcpC&%|Y#wb$9YVPVX?VSOyON5f^X8J6X3}E* z(sfy}EpJUhoBh@*O<;&XN-zV$vZmI)b z+r_|TT-Pog9w%#~VZv9GJPzw?9$#Yl#0s>x`M&&VcLM``WwuLo`L9=~PRz|WE_pHR z0B1bu&6c+2I;ZelO3sxiJgV1vX!PUARUm^?8&_5)i%)UT1@KSZKlZ$OzeOO)5Pi~> z$&m*r(_E#=LQGd8BdbIv3y$6hU5zIF2o(#Ax)3$@BKgR`;7=#knHNC%FYQg_1GFes zPSZi>WKq$R;^rx@B$|K(E{2qzZ-xVHKGHz!@k*&}GYF+2ZSf(O7dhcK8!801Q&MFB z06xjQ{3Um;pKEH=*kX=K5LbiUz%T zU=`qrJDknL`6UBN9+2)#k+l$8NpSJQ{jc1de>YhTO2nC?!Ue{e)DKW&8i11lpaV;C zBdIC3f9h^Hc>v7-PJmyj0dOKsoP?tZBx0>PzxkKKfQC1kh+G6sl51^LTsN=GhXnO^ z=%JME1N_6S-Pa$w0m?fzoL_|v?uQAv4O!3Ev$x~@7B;Tbs-;7EVwBg#_>E(Va?)`s z4BW{WfF8yA7qwXAFSl^$l#mGq$|>_8>3mtec`!X z1iA|*igUU5@j8QqCs8RLCy29S44#57yO6eQ^$x>61ydLgM&juR;T@_S=n&N>&j5F^ zDSn^<2kk1hK$?QR3=Ge(vBj@CWSYFo%l=uG0SH}&y0BzD3Lg+YqbBy7;O^=rTeAxl z<`572Jp{v%6aWnY;>AyEUosx`%+oJnUUU|9fzYi~;+?e(@W3r;?2&f=5#MmgI=4LY z3u4;?Y0BNTn6nAb!V@R{3oft_v|V6zyshcD7wNh6Z=?pB*6!yuNzlf|qR*@(Ab4BX zhZ&4UJ(&;V(5S`!fU#2G)>O|HvsS3UTcDvB#6%Pfhm~TQ|2W^BrnO)gtf4WXLWaZ# z3En8(F1>#W&b82`o!>^1f)}I$4)(KA?Qd@{is3)R{Ky%h0u`izN;bIn@2^WsN&rvm zjkUUWE(X=3YE4p7Qt#P&M)(;@T^nHc5$>Q&^D(`9oOj12&T;iTnCODo!S87LwGJ*2o`N1X88Sai5o+4{f6Vyezwb!4pofl1)zbazu-$f%` z#mcKX5{30|c?Yx}d4?i-h#;rnS~S-t6;5cssVU5G|CcR;EB@k9Q0)#voEo$;D;ri& ztN_aogIrPP1oqng>AdVry@ov|Aug$`N^Xi*Av;!}jKv8Qeckho%g_pT+H8ux)A(YF zEm!fRvm%zrLQQL;WbsA)r}YWAJ-2ePJC?hOjs`yZ^%&HLRj406hi*eUMs%`%4TW*W z$8A&J4n4MIMDV{ZE=j`w0(m5GCI_V(1MRZZ6^?$60vei)m4bsf)l&D<9zFr3%VTme z4A^*u*@a{hwp5M?`bg>lu(^Bh+tE3(V1Gb?|E6|rP<%6fm-}tkXtp{;ouv04gOe0C zV|fA`!y?_kBJoxC`w42^`i|v-)J19a0IHz>)s!+){`>R(b=mupLN4~;AnE&W0)|JdB=i^0_ndkgWd&fH#3x2N ztM>P&PnYnl1kbRnPq)rhf=*!ivn}vrjB2(NX(lY$b%%X$MRMP-Pyexan%*uVeuFU#$tdR6+Dh5;t>gYzDg1we3AN`>H-XJ%M!m)Je zGZIu$A5sQ9RUuXk_e3BwP?l<8k`F~Z{;3y1<)9Zok!Qn`kpo&4?rzr3=;tX*ZlsjG z2o9rWypX;)+zpjf)yi9L5W$sE~3 zOAD%DsheMFglbkJUQw;TWLFlw&`Gpv6Z_?wp39DjpvvA)W%NhpVCQl-zjlGg#0nmW z+oJ5NL>mJ7Q^o>YKOv3VQgSq%16~)mS zp1b8Ga$ z9`|~;edpvzywXn~5R)Ijwr^OVf#{chh|502*e7NGgv+whg|dBimKx@Z;#P3?`cb5F zG5uZ%R|MDUahZHp@%~wXPaIRb%RTd>4B?Ex27)^o zfL+_)aWFb*K^}ZkMvQt*dsa4j>U*3S+WUrnII!}uVjqPo<_jUNI8X9vQ8IUCGsmhZ zH2Y_&9G%p*$M-QJ9awPATwV*&QabC95QG3yv_n_T;eOXBH&I$vzu~~u&zBu342vOg zs`uTi0Y`1eg56Qy?KWFsSHE$ho1K&5^+bRCSZ!PDSiI92ycGbZ|EeX{wflVZ>&xiz z7v+VM6!|v0F{vzZWYH{wR&M(~DXE@;)}A|9{hNW;AlbP71|GH%C|g4=ufOI9_yG%| zrXOyyxb+{RYXA0&yh>FKN^mtIMRXQ9Y3DIB6|WHzN_H`u+MPU=mXo0G`nOu8$1x}M zGsPr~SYpEAgnd+`M%wX#7Z94@oR%(MZm$Czt=838k~0%LaRos8YrDkeUVZbTZuGDj z{q4)9%Hc9}ZoecZTzJ91yrdsMuUz*wSA#sdlf~6<{GVv$C4m*-i4(yjYLSD!SI zS+Io*Ox}AY>Z~{zV`FCHs=;QPjsCFOS6mLEPn?YeOm5ni{7tF5n8^y&mmrf2VQr>h zL_@j9-jU-1h8yaDCAt!$)Dm_ML1ChpcpB#|{$Qf{(adTTL|@k(>()wJmUvc+Vz6!& zV)~e}tR}G|H@ryUkzFW8v*ACRb7i?bQqXUmgLmU|R$Obn`4xsI(Y6_>rjnfE!ent+ z|KXVfMy=Z87L%wAi_I0zfy^h9RG+O6a!512jg3wCHi4{6ktOGOt`Y7hWHPqZ8;eiJ>h5?j0N);8{))lABBE@9hcR>f zF$6ttcM#azKvO%Rh<(6hM|U02&_#H*EXx3ZPV+ZT`w1c(Q?G&A>MAjU(D3agNfpa% zKmA5-07(v%sR{~^#brwvEhtH;z?so$8Crb^Os}9oMNI3S6CsO}+dY&9##df5oKX@@ z-@yUn%5jSxj}Th1d0kEN)hW>2DU(PMYgT04Ztdkwv%~=l4d^_x>ajVo1xUetP=zyA zrn19fu+X2SQn14SqXbvhq-@|-Hm@j26@WPhYTPr|y-cJ;A@NwM6me9SL>r@ewRQLV zQ1_@PuO*I`0wM}TaEw2iIk;`@y1BmB6*8!m6yhyvrOJHj2Fd|r%^)Pd2($<{uH#EM zfFyZlcFTre)PQbA;N~W^f*QbK2GpQ20kqeMo;0+fn;7DbGn%(W>DY)EXA)uG;XUA_rh9yxQO`mrUNIq zqS=81|KZ+-C(9nnL0*X3(K6tfeKl&zUSoC8DZ0;i(YIR@WtpT11sG=GMwL`mB#zdMvK=MGZ@NQWynByASY(K2&Pth`bCpvrT zDzF47NPzHPUQdo5_ExS25V*7dqV{6~P<}VAIkurqVCj5FL(5=SxRHl~skc6MQDBL`EH=E8+{jI{^$v3^ORtY&Q z^7HbT(H{X&5dK?b`dhg;FYzDyR;FOZfTd7Cg4*%`$Wq{S^e~A5C;bWxVgRTf_Hl>u z8&LWU)U#Ve0u*!=+!O)MMP1Tk>Ho6Fc}Cl`>Tv!K zUvC)~RrkFQgQQ3*N{4{bDbgttB3;rU3?bbi4T7|Ew}3PPLr6Ew(C8o~(k&fBK6~!( z@Bj9h5APgs=A6CvS?gL?+zS+H0v6HLPN4Br8NN+fx*|Al9`if+CNGYsIJ7%u$|Llx zT5J-S(A&>7ETN+0zk{5n?=52N;m6w_-_g^_r23n2S-yI4wsoFQW(~e=hn3 zpZxnx?#2}4D{f3428@aPgC1}-p$RVf)=}XmC!eYf*7~*y(Xdv&)X~v7=q76?25w4R z!BU78B;m9D{KC>Cj>K(`7B65;9fES)<|6p}HwZp~58qA|hV!-V80~}Hww%w#bu?CB z`{gC!BUtt`ZPGx_9Lh53Kf+7`lXS|EC_kW60samDUd#L%7pl`gMCG%qlY(SVy*Q29 z-jZP(eFKA@kAIG7Q90%gNMftnCq+~Pv~R5mG|}(`f6pBREF3(~ItW>DT9V=J9Bq-n z^(2T55uvX$KsOna+rpiGF4_6E-Cg3WVS2v5BuIN)4jVmOlOZWVugnqH-Ftym8v$@{ zu*s;`*-+p1D7Awc*n~x966D<&@$cf~9)vRgy*?h3ryc-l_2hADG}(OD{_zv0XBaOH z=thercFl--3>GhSP)1XF&-ay6WxKD{I*BM452?IZA0X z1?p+8N;YB9d6s7(8L9!7Qz-BBusO#Ln*PC$5|GpG+sqei)VS0!cIiQZ8aVRzSuH?Z z;G^6tHC%YE9I%Hbq2X{Cn@MagUaTh%q7@rIxf3 z*+i!MAeOS<+?&;opYkdEi1SfN7X|nqAc6*D1K+0_%g!HsgZHn_@!q4F`edFiEhhJ( zFDM;7#Nl|lNq8>A^Pdlqas4sUYyG~e>{mw2GGCk+#vmxjFai8KsMBLTMp*%ef^>96 z74t6(p#tx&iC)=8;GOx1pIrG(JEHyi@f!q@7uT8-+q0+swZq;eTM_C9OFb{CH5^B4 zqYn32zwXq`L}hLcWZiVA|5`r@BLDg(!J?apHRM(Xn6;Jv>IPr+L74qBgKmGeSaa}a zH-5&+6{bT@u4cT!7E-8O`mZ~pCJ2cWn&3fWO3TUr)xyqa?-cO^bK3ADmc(gKK+k^8 zMT^jivDNY+FnJbwU`F*NH8Z)7{0Tb$5l@U;*;yctM5TGfHV@}R^1q#$FHOr9jim0U_`?V%{(4o^*CzQ&i;Mb5Nor?=z1dMi&>u@p_sjXd`ULVsnAttR zX4Pm3+h6)%=s!hpk$HzC{>R(SuJWtoZ=6RvT1VTB|Ha7KBa&_0ph)5}Y*GvP1dN`0 zx{0DUis*=Yr9*@q$nOdxnvSrTl4#bVUY2>3wKOx@BDB?e6jQ2t?|i8knC&Ny)$d;T+{_)l>gsdVukFq9yG8- zOVm6_Brr=BCf0g4NQ@?cp+#2;JAhjYo~3hDwX8p8QtGYXzt%TvsZf8D-`>Eeg#)j;k^CQ!MIOjAEX@qZjq!ET)dXNaQ?e*=y3x5 z@9Hjrt6LGmpWFkzULd_V^L_05G50|E)&@@Wr z?6a-y8<`zr%(vNmgHE$K?vBD|&P)PK}8BjjO(hT?_m^q7VmR zN;-XiMY@D2{`u&i5^uHP)t*pg4%z*w2_VWA5g3}Df6a0OlE4p^x#%EnFV#uM>^`Hl7cTG%;A6qn5WSdurAKn8n$ z%7DK{oxztzG|RayodN4g*G8V$PquE6+));rMcauRe(JbkgnP66Xqf>9+7)RKieM?r z;0R3hvw^)w*_|M8NWW%~-6+kvF(| z2mBQySLW+c-dM0PsU<%I>|c2QU_*Z0W%YagBtaMcAw_Xwq1P8WweSy!|&w|S@AAvF^E`8wOjE&4nKdAyAl630FSkEWW1SiRs5x80Aqv>K<;PE{9U4qlpb&ctG zwLa%Oijfmqx9?B*Z(4$adGq7aiG9h+qZUaFSJRCGJ&qqY--z5K{1(26(1mz}@W~pO zRZ{C~HOQ9m_G9H-=knmQPhl`Ao9z1<=wVlzNg(k>e4?txPP=`1Iyz7Mi}1}YTAa+k z?w6M(nxmh6Z{NcxgM4yvwEt&%d|7rC_(l@e*l`RaUrWF3t~l{+#y;Wlf@&|q-`J&gIAve5W@vKtAnlpq=XVtDFUG(9}{*~@AZ*KVcJ$81lRTw(uC z39yzH{b$=J?)QMEMTy+YwhtJ|5?}^p%^PUl)^_OaD-Hz~;B}~Y(Phun3fw#DnM-Tr zfX}0I_KE0``Y%}Ng)KQXv4)_hz*DKDP6F$Qr~aNMc~crSR7)zbEH{QATy*Jk8!1lT z=oR>1ko1^8E+p;r^VbSf!coXKOA6CQHzoKP)KCNK?#_F+tIUz7j0-_0Ai%=PmQ`a5 znvnBP!UlOF&h)TqQ}h$JV54oy>(xsGr_(WqPFq?e-n`5mcI=`D^^xEwE6t+Dolj?( zqgRuY38JReiW0(J3JI8%IuF{6i2#>i-FO|q3;a|u*wMTS2^Q;0uZS_N1v0seijqXJ zn(-y#UJxoq+^8gAI#a(}KX!VE86;Z%x^xIlHf4x{dCGJg`^&Ln0n5EFb5rJzL4qz7 z=EIW$S6WPINbW0&)tmYv8#bIwWdKbf3!;&Csv%pt>=WMW@qX$QzRz*$Z|%_Fq68gx z)A}v#qpA6Yp9;3B(t|V-V@U{zi454HcE`WS$5X$d3*+OITgt(nV$qCd^WDgTDXl2< znO@#PZP+H5*{EZ-TXCY%j(jEJ#y0#o*mrnwgCj1{T$r=G)fBVp552_lL;a1w9pO(7 zQL|3k97mD}QnLO1i;f{_?sFZDJgi>xl%h4N@i=O7k+6MXYA`WjnQ4!$ZwuHxhhEX( z$Fgo~Hjc&yJZWe?(+`R|bo?s@Ntq81O?f&&xb5j+E!RWaW9ujI`>V|M+uIeiP=zdG zi+9_9r3SVTtE-Ey?+UiG`&5hSUP^CYUkoaUeAsp+J;O2Dl_wJdp5&sHEvf+37vlK~L)Yfs=WB>iM&XtkFvM z_6lpt`c%+d7f#@=9R3!5n{;Y((GcyCB+YH=y6q}nBax&*9yg#89^@P5N9lH%lpw86 zDvKLDPjRw${v9(!p#len-jNyRxTv@2)=+-QV9VLR@_i?FCic2C;d|}vOJzu@I>jzy zY4}e%TZn)kWxs-290pDbpaL~-P;(|t8;)otJ;eE3Xj#P6En`)Jzc@kAbCCClww%)AEY%8NzhhNkhTCry+f7)^JcU)b;{?&o6FA8d{!+M?quq11~W&tNe`cMOhCd;E` z5Bq(EDD0NM0y^rQ|GgMrOSTvhg;=)63c}Y?NPgkW;Z#FPbNwcM<{kBL{m1JuA(qWq ziF$0l>|}f)LZWv3DfHz)F?g*a1yqNco{;`qqqd7&+q#`Kv4WZvi2h!1@GP^(>Wu~6 zH%R5S<~O9%b}tajx5o~ILSKS@kL8*AA2zEUryl9YR}4~@+}NSO?3I6>KG!32^pEB5 zQPBqom}v^1%nV1l{6kqj#xD@QM;M=XwP(Df{+Zu_5Nkj;kB*nJp-6g5s2|wcQ8oGW z>|K)4HmP4`8QlwYW z=I&w#-5UmVJIXrQzZ)R|ng=)`=zt;qB13;;=Ma6Ugck|7LmS^YeGc0Yy$c;Jy=z81 zgcQ~Fxfo=Qy(TvYVeaw8thGuEc-H~$?!T0qnIkw@^9xmgJ%62!$SeruNo|?OLP=im zsnfVck~zwG`6dPBLF_*)GtaH@@U*w-) z<;8y@$GOjkWuf?#e+q|q>KkQ66&ayR4^J;R&Rm1bvfkMCkx;}9fQtdrE*H{^I)1i* zpktl#H4rNCi%ofqJOnHcx7gSe4Wy{r)F7YUy28hA(0&jhVxyrN%Rf8^aH@a)=PfIM zjq(yi++#_r(Qe+e;HjyvUj_RAH_tCa&Eo<51pdl3IvOzJPOiQ8IeUX%bH{hx3n7Iz zi1Q1fco2RgOVK~r83fvg*my{-Wd~ba5~O$rfkI_!P4knk{o z9_{l}2dF-I9Aug*L|>}S0o#7qm5tT9DK^P=8)SU}ch{i-V&HjR5J7n0q? zi*0hzO?EJqd;4=O#(i^|z)T`pLkJiEIRh-?DZUKZ(DOc#Sr)yNrOCJ4eE?l>Rzp6? zenM~G1gwx-SD$e^6CQ)p5hJLUe{N5J7RZsh_0yFL9xW5CgX*hmG+Xo#fKR`L>FY!a z0dRXC2$L94&buoADzgnQwOuXLb$Y&GynV91z_p2nGIWllv&Xf4b~yur(ylio7l$kp zn81Td69bzmlA+f*I?iv`e3Wp~vR=P(c@6D#^}q@*I44lyUSpiq=EPrN5#YoUx;dFb zU$t%9h}Z-V1e9C%Z>TL9Wem1lAqizn7<^wR(8C+t7O&gm+&KS^{axqeHXbjB0O{bMX+WG1U}pd7YEL%HE%d^sR743-p?)A*Ii3mWvSp(tP>*Q3 zIh#97AM?e&I^Xxa>7D3BjBkrlwCBB;Z2R#eQq^x;^J4$*c7J(|P$bo@YKzo43@t5s z_1*m~j&G#=Y1csSbr_-QdoTC)q$D5ux^3(AWw7eNh`w`EsogN$@jCk=KO}mEoEyZ{ zAN(GWcoHMmlbmaF81QG{?)=|E6j|G*EK2MHR-(}Wjsgv!os@UoK~o^KD05_I5J#{2 zdMoMzA|O}Ljk0e-MaMON7CqK2K0zpI0m+mX)~L`5ey_6_FN>MplOb#VZd>DHN<;Tw z+M_l_nr)l)Xc0`P@Uj6ttL&HRbkY4a%HVm5u3B{PO^eMJFfoz|KAGixW_4S?G^4K4ED6=U&|d zj+86klwr{68+t8$x+w3sWCwwUR*jZh_S6GjLvM|PQHIcu_Q}=gO1gY8r6o4kp^$Sl zmm3f-l-h9(X}!DI59h`1S*qc4@nKS%<<2kre_Q~?%Rk{D@epM~9f~rcuDUJXPCRqr zrvxE*6nZ*s{5!fqd`F!)62wGaofxJ{w7cwyHnVS~7i`O(uTg?leg1f~n zO$@v_=_4&}=(2a;{~@>M>(86s>C%q^S(l>n+3w01t0jFWsVOO@(b6STLA-(Z*v6&Jf#dp|euM{m(fF2|Y31vsi{TM^Tniejgh4lfFO5)>lBid#yt{TS zr?1RK$6R&8k2#*rviPdg5I3`kko4Wz5=vs5(-+>->VLEdw~pJ8W=)^@08s9Cw0A9h z*5P0e$`ww85^%M|HK?%O_pPX?5Hi3?S`>`&E=20X30Xm*`b=qen($fMh@}v9!Ss&V zKr9V71HxCwry7P0nB_$)T7+VaECOCcFz9XM79U9rfD_2ik)nOr^(^pi?8C=xYvNTZ z!=w>0(jh**FhR-~wAG-&!U+8bAHYj-Fz36@HEy6AL%_{+Knz*OMW2Hn4h?@%8Dmd) zX2&(Wqc?N$+HyK*MAWn%u?+ZRq9_-o_sf^wunSanW5D%5Kr~2Kt;+rt^r6!EkN3tB z2wEfp9Osa;0Y>)sLeBo~gb<9&ucG16-Glz7rR?qc=pj$PMEA_>hvYFJdn6w4L0 z(8bJkCNkgd;BMJLx_8!>s@{g1;tAZK&bub3v_o--ki?9IS?=A?!?hq$XXgPHTX>`1 z*Q^Ws;e|p=XtN`;Z+a<=@yot`Uk7?~xa3a$ZoAD#kl$bs>di0LAj&0Cf{&n?!$@;1 zAt;t_8M{myNYU)*Q6oH^9jvmhGi{&LB9zQ1F;0_&z1OC5Wm+25+r=p*``K3Odo0p1 zvJo_f5Rq(67+YSP!tJ@nr$&_LHr8GR8s1&ortO0^aGuLv072_}+HCrjbo#j@7J&ttMIZI(6bIVN=78 zGMHS*sj8>+RVI|D+^X`r^1ExDKd#tdT*`KfC};{dr(zH{+*pF7Vz3v??Ej2SdT1@} z(5hzRz#4cIwJJ1 z(wSV+h;h)gAAd`7^U#O;4}bv-9#;N$(+s7>v~BfFUdwJ4G#sk)kyp0DJn-sro0_W4 zyo-BEYg|2K2%G7{k9VS!Q8v?LOLcnFf#l+#dy(r~fhw+z;u2ILF7q0W11_jLuJxYF z#~*w9Q>3YUzqt`ZP#Z4`dGMhtP>n>w)VM2>gT&{m${5Y(G%Yvlk8;Y+z_cDbPKO@z z7+LxHHrMN9F_}~$vt;twl-F~!Dd2kA%%L>@>ZSD_bEua8dX)iaX<5-_KQNDCTNn3L zHwEXuMT3-ErZt^Fhh^{aK^?X+0n&rP-u7+ML@acw&`DIPlRkM_j(SN}}RE$_z2eul)|Gv{l3fsWYXyK-TJD!!n;v_g&_-nTm2{a3tobkcWX97i4Eti}wA%yYwDrzKURrj;9Y> zWDd>87brIs0h2B%2`YYg3UW`jtz<82`i#~A#yL$Q#kv-KKighMjH(?sS;jW~{@=ww zT&d(QV5*84!3YP*y>;>F0~~PBWMtQjt8qMJV|>g9FQ&up-%oGn<@<4dEMtc0V#b-0 zY?IuzpT^z2@nrvW>U^qLgU9cWyv$x|54rRB^_5M1yx(Isfk!Fs>Z-eW@6rb#yob|JMh;U3jS3&}enn4hgHZQ?)l!AoA!zze1iS>lX9J zHrfXJIh|Ht8C~}FNbV~;)x1_aMOF~xYt=HzPeh#p6{NRUxf*k0nU*F_8YMqQiV3Oh z_c7qHmR1H?frw))#W>PmPi}a66ji2vYKy6w8S3k^857bA3iPkqj$YmX zu0?-uo+&^<4QXjGfZADskA9iC5L}|C-9@L>g!@q%h)IoMaY1 z%r7)OYHsz=Z?>6Fe^Pz;DRYrGfPeEkRxEgG-b32x*R%=&b$6M=;;)@ zM6P{0gkn;wSUmk5f($;Q`O1drV)(uO%gk^M_>*8~EN$!Ux8iunL;C#DN%evlP8n9c4?|)gV=|C1`s?Z2&i(h=uyjjZ$$wPrz}O#W_5Apw2wLla zzVm3h#Ido`821CwxL?1t$z&}I*ov|=m#NU!lSeV9R>myJyEC>fV%>=`tEN^uYlbiM z*+cZ(Po50$5g2-iM-<=!MToE%*Z_#g5b1Dl$|Qn6I~{f#JBIO0Z@v^ioT$|)_|}@j z*KQphKZ7G>0R(GSSC;uuOml?{m3A`qe((jshT)-xi3AW5S_<>3lbYS#-h4Js=TukB zsw|^Gw{;z${POJiiZd0Yfd=?F@Dnw3LxI2=xnk3FtW?Lmkec_3;c zDVja1v|#R|_Z&T(C+QjIu4>soBW`CfGumpi)>EqclL~c5bS)gN`R9`VnsVM1jrp?f6<9L9k2TxNgZd1mEM;6o`W!Zln`fk7WzOEv~~;b zeODJ0$?~ieOSwQtfx;K^e|AmPXi)-_|Li(oflDfRv!%*Q3m?Vu1)+Q|bt`?_6CZJub;;<(1UYY8M$Z7q4vRuX*5mdN7sj`8$TxG?M$zLQfX0m!)tY5- zUD*pNOAIm6Ar)T7E3MTsvTeVL=fVH(x-oiRzNqZQ5r~9p4*|Zm{tS3eF!^wei6HLx zR5zE`S=q1LTC^RSsp_ zu0qr(du2CoynFX04$pNu$(`HjPV#V;T8~FT8hSWlQvPDpmBbbjwiQO@&?(q{NB~G9 zJV|8F;BL5s@6hLg%mmeBf10ok<$$Nor-0Keexcn@6xe>;VgZ9)PoMpB0$HJPvD2TL zr~vWiqE5GQ*il*11p`1l_;Rl1H$=R@*1h}n56qd6TdDz(QCvR;XQcfbt-np9lD@f` zS04hAD5iA#s!#m5r)kF(PQU?DDP3Q6cY6K$axZGrIHp@E#H^^t`&Va>>2=BaK7zT1sBE^a)??@|1xSr!1D}6@kku4;q}_q%#HMHKh6fW(NzRJJd37gl>Ows zM~S(;qJItvH$LlT;&{RG>`f0odvmzr3%~%(gIa*o%+KBgql!HVC;}{FuJXX6F_gcxsA`m@}A@@wFT!;T_bZH!|a4f8Fmy^QnfWjr`*>{u{@U$WZD z6j_)afj*mx8{E$C0e80%w(z4;=&+MG3L&Df$_lcqyDx=}TL*@7UVky@J_7|#r9&QG zQVEo#BKh_L^}5HB5je}z7A-*TcM$;|%t-!w31it%jQF={$R7%b4&lyT`zqx^!Lmp+l941b)v zlK5s$)D?rP!P%~ujFS?}zXtUTHI>DZZhy822&2H4LcfV`fJ~S23qWz-1C&VAW0%D+ zd*y+!$udwKfy)GVZn>Ti!u|YZ{BRin3nnK&MiqM}0(hSgM5HRyO zUJmaRbVDAyzn4VT3lfWxlnT|GFDl+A6|?-F3eCc(FN<%4BKd{s$08WKmkyd%VqGzw zn7c3spQ`PC1vH*o4%egZYi#M;No2k6Ly5-ijEoTofKwLt&3VY11Tc!>WV~|NK~lF8 z(HeE4e3dVOW7*+~Ofj3Igfy`!p!HV9Jv&}?7uRKvU0?X;A;?lRuQTv7L=TD3HwVz^ zC}L!ij*lE4MKcBnj#R1dyT-DQV7vCt12P|2cksph%|dSWO!eI10c)oHDv~NA3DQY7 zTh0gfzP6!n5lsW|jxTUNcJwXvz;6&W8lCP-mS5q2;a4)%)$Vr%U`e9wOjr%#Jpd~~ zucG$h_Vk9DWB%r!C%12%*SAJ+u7yL9F7+c{Z0Pst01SNgZ(+yyq`GV%#3BfNttb1yfYmiFb|+xs6pzT+-z zdK+`6_sd}QmTGgJvp3+$(u?cX!@6Zcwp%XhTX)~%KLLz?7Ad}6Q0Fe+ecg4XR~?(Z zwhMYMtT3~2OZW((`0>KMa}mUDWRvE0jYaN#)AIK@kUN+Xe7bO8phaM*&Du~`x(!le9TypY|ql%XPG=-jZ-|}aXeRp2cxYv|Fhnl>JjE6J=eLA z{Dj~bmfd+>Uv>nmB=Kum+tH&yMBdyF9@0!K|2M<(Q}W++UAoQ0lUf>}#Rxt0oM!Fz zIrn6Lqd60@5P;zKip7Q~mHHZN@TD&izVLO>Ymanlq?}cm)q#z3+`##S!_EGjAmCmc zIF$D$X@}1%uF*Ae)tNR3Sf4e7ok1pQ2z0}D$uDYZgZ{Y6_qZuJ)qRdQH9SG&`bse6PT4K`Ad z)0d|jr$&$U-!kt}#TD?PuOg==4d2 ze9x+dj}Gb!Pc;4dz}f2C>p6%%Br4j=mjug}l4_VYrx{^mbTKv|K zlwCTnR2)jDJl+&K?UP)&_r;%I8BQc(XlyK(v0f50^@i(gEvi%Bx%);UtyTl_E^z*~ z=8p|y+2vapNi`klnYh+%1kEkbx-U1{>W+Mve)|ZED6u5RWkH|d1a=!d>*Y_H!q-Z6 zx&V&m7W{8M74fdROYFK`+kPYNiyb7JFX%oHO-Tkc*k~C&eOv-S($23PGlou$-?fH0_cXt1}UW<d{JN^L&89<>YqqlL$EYu!aA=&c6Ig)iJy$U;|CLdL z5ygZr1F#UqnfsHh6_X&H8cZVLC63%Mzy}?D`uEhCT6Dl^;v!bP{8upaeOx(pG3eMO zey;?MlCR8GS4bo$u(U%T;vYexCnjtzJRsoEtuy~^m7EV1((fPSc%3ka9;3}i@L7UK zLzJ2jh+HVIIrNs!IUdcK*5&&SR>I5gfYf4z{*a)5>8dFd$?u(GHO<(DF}vdYlTG zQ7ZYfKZYK!-LVJzXO5g9yp;+JaC|AC%FrmJ7*VLen8frjxo~{=g9W8t4u=?=_^?@v z7?1I;q|1UfP2zZHkSJ~m;lmesk6JA7yADG@pqnoBgvL|fG{TN!8hbXH>P(yzXG(tw zy)YY^23}+|LBRURgKWG8Zk{=Eoi2}Do#&JcFg5Cd(fAqsG(sAhJJZ6`lzrX4G%*o* z942MR89`dojck`nM#w3c>{6H1p4I7O@fLkGe_tXF0A7N0xsJu$w7oo~6r&DYYocTg zoa{e(+PSC}D(2U(U-xl9RaebfmX*02Q!t@4Q>xKJcQG}5!&-@tZ)%}ZB`47QEM6S?T%|xkxz_ECV-bGmd`WcS0TkRftRY zcxvjB(yjx`5T+6eL58>F{Qyf)D^xHq^WB`%Bn)VO{y z=g%VgFVY)cVjc}%l)J}ByT!v22c-~q!1(Y2b;YQ_*R5TMl6-JwYGDs1R&aOUXz0!b zOm0le5-%D01n{(9#95S}=fA(#%;W5luk@vK{@UN(p5Vqm;Qa}aQmoyLZvrTz0pf+v z97B<-+-KQ#ARH*f19sLKL!S$1!XXf+Hnn`ap1M*;7kED&@gNSDGxVCV?6W7ERknU>1DWj+ zn+!YO0BJAq_0{TJoa_$W7w)(DFThEfExv$n9S4Xhz<@HuvMz$@4QTzcm9xX3lZ#+x zY$kcr5vA5%%h3ZycPjKaaP2J$b3^g*KS!5*B}0vg04YOb;%qR`^GS-~fLr|%7VV(0 zSD`~6#0dt(6o5~y$1_FX-!^~kbBwF?(B-bbUr&Tm9x=S_D!asKcyUFVl8n*K_nWXj zU2$#ySrQcK1Maq4%9}v$HxI2j-@cb;ocU`;ne+$|0rfs8V#~FoGw+o*0ICDL*|eq0-G!1|aiwNMF1HM~-i3X5`Dgk5qf%#%FwJ-d55O%;(3Hn6*jta6S|p zqG_*T4hSZA$T{;8oN8X-yug+*^;OE+&R*218WXO?>J12?pY984@gsaTsz-DSB?Y-1y}pUdqcm|U!dOt%Qson+yQQ?`+DIFKO>C2Fy1`vjP6irD3(V8f{EcM2aE=Fo(D2Xcsk-0Di+#7}dep z-Srq(8~vcwMHwi%y+FIZ>lPzWSuut6Ri(Q9@`Itv(6dnoK*tgG`i+aw&+@tm%KQ4f+$yG*FPMriCHh8-S%sT{g&gx6dY;0`a z^R!$4o;Ct^&};GPPdG{{fy-hmN>3*RKnSrNy|Gsr>)dFmYg4nn;n1C`7Q#hfD_qmi zAW%0l{QypkiSE#PDCY~fJ_DfTL2;303-PxnP@j-r8Ehh}KNzd`y5HSLYf$agw)Qt+%Ju121xs zN*hf)CY`IZA8n6MQyLdMs8Ax7zaa2vJCyMrpe3!xj4x5gtjfML-jP{5j|qexz4}E5 zY)O`#<37$y7n*lOx4)uH-hl`jG@R-C;;%4Ek$VUjX)HbyXUA{`2)4k63X)T zaM(&>YAdJORa-4}(#-`UXr!AwguS};q2PjB;#ktIQ`Ulc?j2d*^f(A#8=jRFdM?2AEpiD1;wzxUrl z3;Jrqj|z#Hc(O>7)3SFkg7l5e|LrUya^||l#pCEvt#3e z?(UH8KWfjO2hkCMKnRs%D+w_AeTou^HGXa3$d#hySJF&>wq*L_gA?YUAKC%j1dT@7qXAFC}n~ z_WFN&gX$C@<$eqaeQw3j9{y|FvMM4OZ`N`~#Vf1IA$<+zbs{}sjWT$M7-u~NHRZ}uQjV_uiF^+s<5+Rmvh zsnYAd%`!2faG6HZFFrpg;4+_~e72|p4D3}*>A%#iWxiE7w$Bo(lU{-G9w6Lgk6^>5 z>U0nr%x~IViPwr?Pce_IudMmOKWDRx(-j?&F>UOrrcIyz-RZHEX6A}y0KVhXoSaV~ zHtf%CxOUWS^%zQi`iivBRy@XOyLjmLI4KG&Hl6ZMVALZ>&+h6~Ig-8Oopl&f0QFE2 z@=F8Lv_mgn1@PM*ker3Ud;70EsLS?zh2ZW@HfhrX^!@Y{GKoydzPp=RweIOZZ-U&3x{#`F63+BS?7Onp4zPJ;jFI z3PgItVIRvI{v0BUx7FlVB6qkv%d$V!>{Xvf2`5M5 zi&<$H&kk$ks#=At4025m-wp@n@wno_cAurRQJ$D4oq!CN;Jv*_UY8uS9YospUSb^6 z8sbz1L>daXnvwtTLwsyAKAgP2u5C#t@NF^vD)$|QyZ->^LRxzO(WlulD#-^u*LmC? zxVP@4Lbu|a;{KyMkz17|D!#?6mdQZ}9p<#@s{Z{Qej9na-kG0Fa{d2YWV*uy^1fHW zm}y?6L|w%nAIrY46#6kw8#@YUBxP>3(eZT;o5hXe{ic1GS^aj z6GHEFD#hGVX6#Fm-vy!6_5n^7hT>mx>0!R9TNsj7K8mcK4zhZ*L9NCt+`De9f{?GJ zS!uKT=lM`OX@!q}cLsH&Se*rge_MzPP{zP2&wb~LlkCceU38=~doEWxK7oTV^;0&Jke}18mZQ7E$P@aT|ryJMaR+XSyx#lEuH*-A_5>qTKY;GLX zn)ACjDzt&xxy67_P1bYO#ydd~vUWxgDb)}B_l}Ky-BYd}e zRQrW+U48vmB~$5xCffQ4{ff%UW_#Ysk(+q7bYpH`?*N>8s4wv2$gDJ{phiN1da%baoFtRSQKl&6Y zSA=i8Uot=#x}{vWoqpUo>P6|Fb1_Fwq4XGTl1!AckaqobvM0vmuEKl~silZrbo^Je z!H%&K!hJdtILe!C*dU{A^r3R%1Ju;sKaxrG{<8 zVuwgC5eCtq<8Va#-*e#>EoW$FT-sjl0Ij-WL;~!7;+hk7i$El>o zGx6E3Hi|;N;f*R4m*Vr6VUI@zmI#9*02II$>D2VwxdwbOpc~>y)O&BjUG`6pyS~}~ z#@?f?0%w+EyL9nR>oi?HLG(r{kPrmt|il26WET(WUC-k_yaXA7L%sP!k z4BDdSCG|7^!XK?qjLD=#^dv%-)%fmVuD%e0@ErB;u)OVD<(ObF^)twnFGkx!i`?7q zspXUGSsJGi>Ji8eLwNH)_!4W+pP(s!v#CR3dyLLauR^<1vbg_lJu$W(T1(0W{~A)J z7-vIa4$8tXp;(eIK3v@3?V@LP)jXsxJ}CLu`|7pS`Rc<1Rc(cttoSLrW{)5YibIoS zpc%{Rnov+_<|t2O#KQ4M=x<2rXO79GNL#rB!larawo?D+#6KY4JXLhF0+cS(Ggw7Q zt~@1Xl^it3`LsY8J1GJcp!>-~0J{%yr&MGGPMi^AlUgZa&T2dah;#&>n$pAcEpQow zGO%}ooSdfi&li?#AvL=Q*k3?1Fa=|?kgmLXV?-1bF*2{I5z}BQ57ec zd90z)4qjCO`YRaVRQBPpWG18cHjp#kSOWTz4=E;}E7aN=UWw|V#`Up-rMv?Z?V;c@ zK%vlxeV~>I>sjixqWW~Inhm^p0pRAxJ@u)F0LQO>i&_bJ+4L*T0s8N?8hNP@(9IOa zXGQArheZLfh7~7UgUvg=mY7YFfSQq;pT+mdC+fi%B7y>5?<97>X@B|D-#hu?M%Z~> zonk^S>u~+a1=8m*w~Nqj)>G=A;pN(@>i4}!@1qV;>I-kS_3{a8C46i&);TfndI35m zH1D>Ngp$+FA4S)o@&P6&s$We)`5p~?H<;z6j4=$?!9I4Z6bc!!e56j=yhaPrWf$p+ zO^EN-QxT>gjYUubR|5(Y4<`ti0injCF>*ja9`YTikFs)VKi-_3p1SDJ28prt@=c&y zUMT@i4HA!@@T&$)_~Fm&%6oeca7>3YCDFhk#P=J8V9+FRWVZE&6G{_fqQK3|S|3(_ zm9}CduD7bJdfglH2OASen_TwfTqphFNA*M%7}#IDOuke1?wb@Sq~P{_AO5C2#B%z{ zFGJ&@@A5J1#H=keKdZ9}D&%n+y{MC5TlwsB-;Y%*YU`A>7m*B}F^JijbiR*o0W&cS zF&SQKr27uCRY_)kG{Zoj8&p}FTM-Tl9n?~y|80p+8RW})25^!aPJ2Z?2_nJaROmx~ zv?G7c$TEz6`N%EAh6*-;4!I&J51eV)AvztI)9Ev=U+^M&Q;qnYjt<_tb(Un&$JDRY zhCtX!#z^rFzAsh*Z@rF4DcCBQL7MuDEV`MVrtSIphkSAjDS#qUsbl+r-Y%oXmTsJrt)7+v14ke^NrElXu)(bom_U!Nl3U9WU{c5t7rLh+oR%z4Z<134 zb=&sWYUGFutk2cp+tgAt6QgB}HGh|??c&j7CeByw<{!=bi136T>;+U82-e66C)^_9 zH3fb#uukwy-}LIKxxTrFNY}svCAK{25C`^L9fvr~HKQ2isSzy2Q2#T&%p>mxQd0&D zMAAR;r40b)Lu>2qBwdU}I_+K`zwViF^#=BU4I8|nY?3-g4)||?i8Jc+66y8J0umYp zn)~O|j?H?3d-nI~Z&)#?sGD?tS&_>A^J6nVN-t#di(}Bc$^i*-l}~^Z%<)UhqDI>K zCz1^KGGNjSegI7BoJ{y#lma?gFHs z1JFP|Ld|*OA&D=jS{=pZBFT2Vj*u{5f47_0oFv6U@T4>ba&cuOIDHG#UlEZ4ZwCN z@`dk?!D6t+WJIkBOX?JWT$*-k;m@fJi+T?$%M$sOoe%rS%=iDaIc(>Ul$Cj?DJ~hG z{~#Ncyz@m>JTI%I=}E?G`N52|9enC#K+MSeVmAAU4}c#)ei0KczH0_fjz=ndI1d53 z`5&qiw=7!!C zBc5)V*mD>8M1Cx&kBr(`TPo>{1wOvL` zw5Q+Xg$I889-RVEU(>~M6pHX5Qn;B~x9e{U5lKE(X0~_&C=ZJu4cvL?S!&ZI``aEX zU`66}oVgfnC3!Uiko;%Qf2P-30nBS2P`fSwLbkD5$ASSC6Y5JajEe4yol~5)lr6;Y zf_mH~R+Ds1)ZOl7l19_M-~Rtc*;z(K*|*W25>P2I7#anXk`|Cg6a)cDkp_n$q>+#o zP*9{BY3Y>i5K&rE>6C69y3aMf>s_bTI%}QhgU=TfxM$|R|JT0u{_T&NP{4%IumZy< zUflOpk%>kf9JERV3)U|y!u(q(KHL{`rIkk?VrqYgZC>NTO(riQ_kDFaFWCVIq949^ ztXn+S(|cF`n`=6pNe)V-IWgKL7W(mdeg%9xpYj_auD{a8OA2%Cu~lQ>5rF}+(4j;9 z89HyL9iq_Y1$DT8;b*1*-0t~XFeqYG2S&F8vmx^kqh9{?p{T=3RLXkY{*6b*BtZOl zq~}|#droWqk?1&R7~>%9{D3!Rhwo@XKJ7tudCA^n<=s+1hCyRJm6G_OF$*(^Anp1i z>5=pCMREVi`gN;=-(NiE#ghr}HwCW3U2jxP>HYh|v;MS6#v;1Y< z1}}+y+=tkJ2nQR2Q*$;h58ir)*!syS)5)XQ>+Oj$G@Yuo&z@AM|JMF&^+SO5+Yv{{ z7RLy`%wU5c6>*k-!&+p9Y&LE@D?4T}%o6|eHaz-%XLoQBWSn|V)g!u(5*-bqQP!Hw z8OHc8qZ2;@`bz4SWoOCGfy<|kLY}v)U-OEd;aA8ja$YU)aU^T~d-Kt?oT)c>*g{30 zG{ie>$=c9r8>;u7&s;cIjM)r z%B*?4*>((>95?L(GFr%3{naUhKIIa5#Tqd1_zJb${E%Vyo=57@cWkCz`J|N3u_+9q zDm;XnN~WW3&gW;(-QD&_Z1%+$C!IIy7ys^Mtqj)hC)7JTf6=$tT`_Xr94=1V`gugD zDUgBQ8=YXaEQRxHDHZWZ)te`eTjyWprHVn=3}MgDz3P&f$f^Ec+7ppqIu4w z(?m;*VAOH5%L=y>k-*YvEL9?-@npCm@yc1v&Y8r{Ie4bX{$Ul zJ1S{dQM;W2Hlv7$;};pmV?UU}O7G=54cS|@Arn9|UA;eX1eJ7hepCDQz$Ne&zS2xkw>ep|7-q;&PRI@wprr4R! zqR;vK*qasoQOL<~roTz0UylQ?jAc%wn`>8VY(10KbESQ64A>->tflqWYi5O{6wFrVdslM>_C1(VUc>z1M0y!QwL{5oW z`qh=(tSiTy+y=E?a3h8X)JCfKuyTFIN~jo*N4C4J8Wt^zN-CpYhYm_!WS3@bOnr?N$lK;!*TZe&dvlQ z9o)Aim^#I>GP}M!#T|vRjgsAs%$b2wr`n(yTeDfbvwE%GeIkGRl;klUr+NOn(b3T$ zddLX3?7VA`q-?`=JyQ;8owsvzybkwD6(-X&?hPtIGK@*VFZDN4U+20PD|1#*ZAG^| z?`uw9v`x7-p^tmK>6vu6 zGEu8QQ8L~wasLKNrP@nF3~>V6OZXS+p0Qr+F7JLy=+^^2e;7*2ZyNF!6?QO^vg4sg+|L53r)_6LbNS!%;KP#GI8(4Q z{Z(!`4X36ISsTM8+k-$JWUE$FtoxwDpegPf$`wLoWe0ujCz~PIsXbWPZ=^!GNMC$y z4a-lXq+(2bP|TCGi>_A6UwbuZFmg|FGBs5?za}ZUPp)dern_cs^3%v0*Prd){a;kW=V(S-#69y_wWShT?6V`ce~=Ec}4cn$e~#_%g% zEw~33$;fb}@hYMBTdso?V~w7T1|t^daT zOm@n2X*R6Cq2-DGu!q1q+sDQFI{~mwBgCll9W!5z`H!;*I~w3$j!C_kUXHjEx(K8E z!?3_%d*y7ldblcwhc48%x8+_uy^Reh|!FH)lW-56$p;^J`U7@I@+$zv%UCL4V z#1Ijw#;H-$Z^j>rLU&*GAbt~eb*=NUT3o>&@<$BxUD+iUNV>Vb))3dn)^EX!J4lL; zyGq@!F4LU{2pg%;p4YuSSOZ^t8E*!omhW)aa6HcWn^?VWm??~-ZjVGK{Q?IHO{Z`- z?@~&?x;{dV+~rW7HExn7D^wPjpBIU~lEf6YyvEi-`J3M^@dB*f4?iZIjMcXge3`!{ zOe~)xdf-fb(X(X=e#P+kkU#hCm3~m={rpIRTTsxSqcb!!hxQ(uZxBC2aLf~urMH5( zg3nhRgar=pPF!U?1Y~4*;jr_K+lyxq^8$#MPgsx1d`b(_Krl*e0FG|=r&e_8z_L0g zUt0+gxEh?|aCR~7+SPOH9H31Efms0VGI?C^A)4F?4Eri7NA5<7o6ROBLcXxLm|7mV zLxo(cIba$|n177<>vPHC&vRB(G?8dZ{zBQsHU}~Vueq_`NIUy$z1pd=iaJyLQ+-9bYQmXY)l4><`2U->eK1d-1GX4T|sEV z0Xphf*nSgAmIa_>&C-)fTt-5UsPdAbF`ieAikXFa_vzSbgvZIOf_7Qee5&6HJ#Lm z>1z3tC2^Xg7yY#bJGMYFZ^8DDPzxpi0WF<2ObiGTRDCtq`{n&(3_rn!M#vV+j%JMc zmI~9VY?k7+NB=zU6SdB>4aXa$QVP);A+hWkwnKW(Gv}v`=l%?#SG1kA>&!^Uy@*Ra}in*!BMvsR$v2A z>OZQc`=6`HML`G@|A#BqH>5(JUS`X{%z6p-9jvcmu(keTM5)_6Kjk3$>eVZZ&%{#J z>7z2;NpKG+c>hv2v-Ib~`m0X(pgMNl055{vQ&)0*_-Q|nSZW&>Dm1uf27+>lnUjyl zrYh3Dwmc?y!M9S^XEqIQVquKsW2B`$<3IvDQ?KDGf-xO2j7QIZEoyy5_+qA2i2F~z z5sGFOp_zY?*?y2y$(0%b2Qhv&%gY| zt}CI>$%SvZSOxG6M|wO3Cif~TjqL3D@b75}9O7S(coEozP88EqOL*ZrGT%uj{(nE~&{V`YJl zHSH&B5aD>}7ik`)3mhBpiL7_}%Li>P;dyf~jzT;ED5fdu4Z0~S;=xXZ`T4^XvUlV{ z661gTc!bdzzsqV&RH_JVm%_M4l;F0?JlX7d>AcjFxM(5G@ zE5^XH5#G7%tp!E&NTLm>h#3oGY;d=}CT;_VZX=fXI|?|C5{hax!VZM6z4QNpAHls` z-tzba3|U3URucG*&4W|vE>1Z$vN&*vRSM+-tUFbs_%eyUdowTOMj(@B-2iP=d zis80BHO52J;vXn@7CL&G$&t<7`GWROuI1$#(|a>y8b@=0Ggr%ED;U5Zl?L`exq2T)vHxky!QI*sj(gUKI+M;}#fg7P^n4S_o4EQh?0(>r>|A zR@VP9wbs>m>-cpq-1Gf>zHpCGZEQV_)y)klY0<6QaxKZjtDcQTMHpBa(}x5zQLDvy zDuT-XG`WGqaLEH=`TVwH$UKC&KI{7%g>Ka}XcPcKXL|SGoT<-X2 zj+2HjQCOUzW~lA((wyy1?vd>(Gof*h4KjTuX)mc+zTHT!D=SY{)IE==U_NpIx;h@| z>gY&+>JSJK|EBEom7Ay_@$bP=R8-Wr-}j{$Xr&6uCvJo<*X))}C<{U<#a+OFynm5+ zbb{qLAcj3uzWSMyKB%a9aNVh3otvzHEt#h!smwAE^eZZ`;zq@V2t`LMHglI{<-kE` zF9n5QY$9YB`+bAj)Zouj^qd;@)8}2js`dg+OnwF&($O09H`2@T2w)d{ zQF>Wfq4uiscf@i!$mlX@*hr?>yNpEAZt|NM`utvoJl`< zxumX+h`~mpT}TI(KQajUEI7-qW_ho;LG7El>=-4xWm$tDwUf1X-@)e&x?yjii~C}H z3S*;^8ak^TJ_3`Q`&O7>QCi+*IGP z?juHN)*3EGr`K3L^%X+QZ!~88MGI4Cnw%u}q|#m6w3{h!nFBgXt-J}1#*5i5E!Kk5 zfduMzhqFc&=AVK&xVDxHczAj9%3HpD!wC!v`vr~&{wOu(PFNZVgwC=izzJ7{)8LTh ziLkcwoaLY}wpO&(@gkwdwRQ(*jE)1ZCWz$aLAa|~@;$*f5MVzPr+!#nRiX9v>@Jb? zSe0PBqJwavd>p=a;NGLYrHi?<#&ROc^1`uoT%Bl@4s$7kaHbEt{@NCpSz9{Iy_b#< zcacqtxBk!6FAam^XYU4iZFv+~UJbe-78r~2hqNRumY-+lbOqW+PA#^+AMC%0k5+Mk zFbBbb<1Z`eVLSd7=w3itTX(x95AsF`BZw@Myid03Fo!9SW0bOx+X%cv4#%jw%IEMP zw;GddoEcE;2%mpC;QqGZ(!N{86C;*HzOzkV)C}Knp=OkoNE|Wy8292w1Kv>S}3gb1ogBX*Z3-bxJl346H)Ddfi&criU~RP|a~~M=%`zUB1F#u~5q4 zg)OR>fm=kGW+?w*d%ot3fiS=D*?sBYNGir4M_AUA;lZLwpoptk&*$q1wt%YELh`M7 z1!`{LfcaX_T8LY(CDO!+cAe>J8E|tR7DD$M?Qx9)R@N+j@Z#9bOa zfLZF~L^SYq<~sU%=8Fsw{@OtPiy?EQh@syIusw=KTjI zzaspel;FcO?=3Ik(HKBXzHXDI^AUZm^ge( z*p*8e>7{=zmz787EMQtVDuW}r+DgM%;Tgig$%z@)2(aY5AkGc9P{y>%M=-a;)|?nl zQF!PKsxn_KAhKc>j#Yc37JZf-8_Jg7hu?#*?IMKKZBl=~93iRanRRZt{|VOefqdOx z=}(&I1M#GEqZgztT;z7XoF9EzcZ^hvvh3&o0Zl4Ca3-O4N9G#b$ef@B=^{qYi4ukn zKxy@vwOLDR5TAyT`Q4n{$rQXPM-ESMhES4fkzo~M@X8;perd%}62s8y?GG)MZAC#T zX#xBJZoxg!k_sgDM+~lOh&!ikL0_xOZQ#s=tK86-AA6bI4C~xx`qMsosT_AL+bG?o zPMd~96Aa3+3ff++va&86)tw(EpEqHa$2b4!L?T#f%<)%GXd*hTs!>8Q;M z{$h|F7^6Dfc5JI53Wc)t543BMrbCm#a0Wds6hAX9&A30RKly_l$dmTu%`hKy#9*ir zTL3j}t9QpkGMS-5cmr7P*`vyA7Jd~JkPOxFn8?RBV5vG#^P@!;a-N%HTmv$Z`1xVaY+>yN z4v!4Y`@ytMNnKzC{u4w%g6MzGmupYX6kzP+*JI6ZeE+);LHQMMW`J8IxK(_w_{SHl zgY8yMI{#+2i&`{l0#DYi^$cGTp=N+ts8uQ^)mF%~0r(f7*Ng^HR%P?M&64_!Aap z#$!^igXX`VhU!dSh^6NLqpHGN4>Tsh>fB>kWruj3ATjWd3k*X_k>L3xbczUp+Mvs%t9#jxg<85)1&`_ttnvdI=V9zl5Nm={eUQ{mzW5x=QNh z*Jn+XGoGWQKB%H^xWRhBTep}|(tlP!ou(=Gb1cPKGShDn4g}+Xo3-p4H4iE7(d~sP z@w7|QTF=z)eQsSW>5~7T7J_t~pCX-8{|iNc|I_vUKcxlL z|JO#mji=s%%=X~fOL-0Y;SBW##i&}biK@iA2Q4Db+IG(F(^IJ0i(6LPB?Vi#{j+-? z-&nS`6_nh5C$fr_R%=^T`@Jf;_WX-~i6|wXtFDl9n{DOASyS=n^irXltK!zC-UWVW zV>)mv`%h{LM2}54@ptkP`8S6RFlD{tDS4;;w;BUhv;Uk1MSgc_vG1!ZU^E>VHLY%D zU96;Ya18UWhOh5%UDV$%cvsi}+g#7miADO_hft!|(dJOiKNZ4y{WP_uJpXY9oyZCcVz>>%wN>? z5?1tKD00PleDurPu4+Ek2?!3P=M15H{sQwG3!3=5UP6(j=1ZhNi6I#$a4w!Lv@fKM zF$yg67VW$2f>z?V51H*mAR7pPJrj^kfvuz8~%=eTnjQR!0_ zUa9FYaA0XTr^KpC-uABwEb$qduCL(iQe4t=T%UkJcsC#kg4L^G- zGs?pJCH^!PsUPM!u3buUA}uS2%UuSuLL0OoQ|D@0ZuP5-$HVO+GTFRYAkARw;z7#Y((Ncz|FV&b$FkMEmdAGmlR(D<*KXUm2cV+b(kmr#Ns|j zkn8+trK){L4k`YEuau{@Gh!j@ONZ=L7yMimB3jX;4}+etch0?OhldGWrh$fVL3s;+XJBH- zdMvCEfx7!S(&6p#X#;UZ5RX%+xvneBqv=Qm$5yUR#dwr1xR`!v``?{aI#UM?97x7F@G*4cv@3u6)>TB6VOsbTcC zeHb8NO`B4R$2`_IuZ2CK!<}R>ftaGq#hKm7m0TgM`n+qs_X$M1z~PGj z4=Z^0bGc03Z}cerssES}$7(M?BZwA?gN*26_szfctq4W;S!d2}%g*(tN;C2@4|LNT z9XFL_PtAhq4C+2G_T61q?iTeU(5Av*$N28u1-OMR5TWJ<6p>6M7n^&JQvNeyucs4b z%6d5EP?62-MqY3Xu26o*bI5?`RA-J@H*tDZ(XKp_jThvnAeCi)37L*m#*exiV7yZ_ z6v5ayf;jATb=XPj8FF=qBjcu;z7Z8elHMhEZN_8co`WlqXk3NGffPep+V@y@^{^fm z@lrN=je`!OK)hB<4!yEM@r)4O`KM%UOL{y(>wfc8MEHzB9?^%g#DQVczK+-Y?9sV8 z5(95}XL;k9!kTB6%04+(CV4#JEsQ|N@|pjYVtnP2I2$i!m2P~ad0rm z8S?`M;v1ro9_za)Xcaf*jB~K5iB}dmrUhgkQSaYNcfO=q78W~1ffh|QA82HdsYnPmVB~|q><%(5fZlGkt z_Xye=yXC$cukQ|!G_SUfJ9`J9;TV9O5lMn}J z`oR!@KC*<5yfg2&uH>QVkO~CZ?}&h5sEHT0zXP%E=C@IJfh~jsV2~a;DK5;;o3-?& z6$eN4qbs0H?khh-qEUKU(MIqVlm8I%LfpMK8M8AKM%Ev@-&+qBe(KR(;=a7Xq)hT_ zAa^HV_xlvbalXn1aWJ;o+Y{)R5O~2bqvnZ3hrKFbevJ%m@1Xam=2lNt-(ZO23>DEf zfN+<4p=ybxWZ^Gp&NY z-W`&=wHoa4P(_?Q)3JdgRX!>59kn~o;bRA~$YW;khqtj0d8F!juvvu8@r;b3$Ux$ukHz||<$8J(@8F)yr&@7wg=D@8O9 zO*mI-7fIXM5t~g^X`MqK)N(NIUe^;lo_xW(>v{RuddE(7rQEGNHr>RX3E5c9qYI;v zo8m}xH{ZbL)1cDi^*^!egop^P>lbmfk`*)bwX`U;40Q3Y=k|r(v(OEmmd@$h-Ehb^ zV(~T{apdHG!xF9n1%`%+@{}C(n{Z?MM_7P6{FT;X4W%KL?5|yS=4S=^`8_8~RRZ?= zr(-8!-E6M=M)VD5#8q>F_(3c^(`Sx7&(b2l0ovv_bpeV=xD#Fe;Za~1)(cP1Nqhp8 zYv9y*kn=On(^4RQxcb!{Sc^gELNA)^(HXM>WNrWjG1V(K;6kzLPK99P3M($w)7332 z$?cB<`Oc9;j-1H<@QF=XIj<10k_Z%?GB%BJG+hxXkTp$gainSHrvLlrG-A9`#YYkc zXGLHYT>5+M1di;T#-ZPcU*syw%e-5JBWg{}OOZJhCdTjLh&Pb73E_}{z|7E0E%4lF>^<)=D^JTb z+&4jsg0E90Nj_!GfGhzNxvZHL4Z;!6*b_1CpXv}=w%614D9oV+CLDMa+QRA*!2cuN zSk?wPmjCqvJd3hp3$-xQhq}rcUsvi@9=OD*kzE{2AH1T&&l0Ri zXBi}SnoMP-xNC@9;Uf<#V=0D2(qv+k)@0MufomXwXg^U<)2LW~7-w#Ac0-L{pKbf8 zC;F-=bdCO~Y&^AET%pO3!QJ49xMzW7i(%FGg?N^ocE6K}MUf{|j9I~TgUCFFn1?Y| zTyiF~rpwWlXI&ewzPgy~C61YLk7>8(&jww0tWH1;Al1}&VTO#dP4eo4`$a}bPl_8i zpe$1q8&Hz==CW9AvK_J3*B-$WsOs!!)o!(*^1-V6ji#s>nN*H((Vm|_XMmy4Mt8a- zJ4O;>5&Prll?wn8KVm`%pc0kYzY2;zG>wZQzq1D!15$tuXkyRp&I>ochzkmvXM#SA z^Wg0siK&H;Jv6N_AM@xsPdI|GaN!BL)&mHr)hj^*QQg)%uJ2$7LE)COgy%dL7#_gL zBN+4rhPN9kgl>kX

bz*V$s5Y8gZ?t@5!iKKN3 zgdCR0z9|oqy=wP{S=4SJ&h*baF87;ITPK0QJ2plJcHyi;a15d;i)-h<;&OLDF2Dc? zoj3L2fd`-2&lA=_9(uj*y|)LLO;U&kRhn0c*ML?O_YEVRx$SS1*H|e`O0fVr2{s9) z;@ffmU=bXhmJDg-UPCDy5&PZnB8OmFNSxX^JjR4DCVouG23Y-~GCdDZARg#e1hNFi z7a)D@k-+1;ek){p+Uup>o0uylz>rSkffgnIPw(&6ds)$M(IE%@Y2(&aGx)kL^HKl$ zXHC?cD$!Bvj}2+XFijzs^|x>^Y}uB;-~$_FtCB+bSn z4Gj5HC#K7+E;24}?afrG;-%!~DL9T|>Qo}xNDs6sUL3&(81pexx<4OzB};WL@D{4) zRicwhgKURjaS2j14X;_+M^S`fuo`dhG8+2YZeaoy zThQx+0K}yvVu3HU0laH5TG-P7*%ZY%D9hIf`GYw+w=h)CDPT5!G;kSUByyM2*urVs zg_{^gBE;yUK_22fEh(VJ&5(Ktb&DTc!VzxR^@}U`S}nUnJa-t~m!#=SxK5t6leeu3 z!Y6y_5tk3o;m0uz7G_UlKtsL;&kIQg8v8Tb!MMej`WqP2S>Nie%%3-Q`(G$B((A=H z>$bOQhU!Tmv2*g>zP*G9;KMX$vl3^`gIV!IfblE2^~Z?R4WYmDnPTScRiz^AJg&P) z^t>ef@o?)_$oWACrpxyO$F{Tb#nr>8FYe5obz2WW8Xm=tXQM|3ez(Hh5hnq{I}4wY zvc$Yx{v7JnVn$7ZO;Es2wn0;VEZng38WMh?XiWA<-m2U3v@CWa!HTTX-u?O`_aT!w zFY}hLX|)uVaJ}b)16P1d{l__ZnawbcalTwuwc^EOmek+4h`9*A^=QgA=XpY(-KVKX zJ5Bzozq{RGzw+Kxg)2AVld9_Z5AVXrG>p^K4+jl77jC5&Q)2}BP+vrKuLi|JYh~ZA$F| zn=4NbP3md~q{!6i(VrvNEWZg(MuqF~x{HWx6}UH0CTH3UU6_)0b{=Bc;NZ5_^hu-b zsC#ZL>Etsrb_7O~M;E*k9%?y}I^UWW%l}xC=<@u|34N#@d%yY~p%VFh2SwKVWSO}n zRRbSX50#Y7*<>qfbrP=Ik=9?rxv$F^Fqht?C0z8jLb4jxf$K!YQN7XS-92MiGR~`? zZWo?m9kNFKIPLmaCGgZkLw-a$x#NT&bk#Pz?WP2i+#D92f$Q7j!EODGU$AfZkdNB3 zWpJjT<5em|?P<8nTtNm-oH7T`Z}HP{Ym_<0fK^c_x>S9s$5c)bhUhI_W5-@pIkBh5>>x2~`b)fw#h!9@FKHTE)CLyZu`wBV7!dSY}+t#i(xv&qq28 z$|4r-h-m&M*hD7zq_{rydQh#x$)Tg7^r&j}Sux3YYzx8m_r20wUD=MXK^27Gl2a{g zv3z+-e(rOC$~Sm1DmW-%@Nl_)HlcKE8Y@jHV!wKSTbPRYda~a-Ri(x26So8&y7kC= z4223TS%2DNBa1EOz(jJvn%y+ea%!QYXvsFfQTdtd4>#TH%i{t;%QYk=kN>)6Fg-@b z<@Rrjz2fWD9%K0Sdy1oTt?5N67OoD1&37LL33AWTDp$azntUH0PndA(^IxStaQ-rs z_lsgk;_pK<>mfmsq2%H;a=fNtwdWJ~+N>8jT1RjgP^p~MjJOODc=%ek{V%=4(C}@x zsUF$4MUv8~w~=5wM(a>^U&=QqklmLg#TiMm9R*|oNec-*gX2}dl5%6pF2>z=xKul7 z;-~3BeKGK+(WS$)H;j1N`X=a*ZzywYy+Z7!J9%K3(u=g4{-xDAAxB9~yk-UKE)?;< zFJe8+!VV)A(TcEii_7FCGzpXfbFD4Qod+SlyB2{C-c-aLyzXh~y3<}4 zg32gV7`t)%uNJDewM5Lka!kKA?z&0q8118~;E`BSTfUYlq5iFrtzMy=afbv;cRhmU z#o9Z{{pS5x!)Z?2C0gAkH+r@A$elWg<`@~|ghu0o~RTj)5?{85~aQV8Z%ggviFk!OLSp$^ulh4TU>{Zzr zf)v8r@&i?=$2Pr637dPJu0=2<|HQTznc~RyH5YdE^d=p>w<{QoTFB0w(U1q*bV^_7 zHe4M;=z+Qw+}cA})r9c6^MOWut~N%B){&>vxaD2NjBi81bGqmu0xMS`n&9DkvgID% zixYl4sD7J@SA&^Qf5~Ykow$#>WO!m8E?V>*m29!8)+NCo1vUnrN?L|y-7lrn6-&Nt z6x;K4!0DUvN&VniO2|6LZs0uSB4Y^$y+Xj}wu--zYQO8Ly$|PJ1?A8^zXfJ-wSS%C zRDUTn9)=NZ6e-j-Ar{@IrLVnEm>&}7Y{=N{S_$X`i z#Oat7|8ln3TWebK45rYookRO0FD7W zy}nJ?n!?P0XIZytwwc6iN9C30jn=(mAfb8fjBsuPDLNE2(qg?^+xg>-IL z!xcj2ov<=)GfMK26#msR$5(cdBDrA#9c6wMUdLWaQzSSKU?&ak|vz805!+ySF?PHl0~c7ujLD zp3~{xEt6Dv{h%Pf7!LbZ&9Yu`EEswD-(ulK(DQ4fX%)cY|vFXxkl|o3Kp=3%q zUZe{DVSANSs52JgYhZ0mQ?Slpt|VS4Z$JLB+M&z!x@f0Xl~X&2pi8vJ+GC?L6ieUm zkf*3dM~dy5j4I#$CD-w8J(#p$`-(n;&1TDdgzKXTL*YXD#M>Ons{O^-OkdU{kET5y zO?8BSoV@$Nu75fCl9%`CeTcvR@0KYLf`GwA z%fH7ju<2Rxj4dSzO01;zZBwGt-W9)_e$A=08rV!JD9=vBewTT(3`_7fMe_1jQ#q~) zC70;=M2X6?tcxCa)bF_?=S?2zafWA>_{BI%u>Q=fYnudT403(N^STrRSJ2CXb;A3; z-*SWT=1sfIkP@6lv*nU>y}eEz=GcE0OB5!rrVtObc;6>m(`sSuc8ui$xHn{V#+ zit4t+E7)e1Y(dl-`|qkSQ>>Ul4`HgbzhR(5z7_=vQtn`n6ZO{V9aXp!fGZ_1uh84M=`mpA(2AC>ux;1%%gwac_L}@LWo7x z|P zF01E)_$sD;Lx20z5U8gFWcwqtR6S|5R8)zCw68#*04SOrq$7+_s2}(emeWgb5k;lN zS}>X&E+&Vx0@(nxl*%S{aIvDVSTT@Nf%nC3K1%b?S8rTU0T^jRWTYpbY=IAOD?qdE zo01Ls&{DJ$%n*}qn?E3P&=BT^)gEDUcwyH(51t_~&Ts+Eu2t(ko)jHJcL!lN7*?Rf zfd;ZBScilZrem2_(l9{~kkEVuj|pLc?36~%=Yi$v{L=1!T>%^sC}Ybqy(v{DMKw%6 zsz9jethn4K7}wzk!UO^GPi@J3J$qX*c?Jw*4}94%zvu0|->F^_oO0*GI1|Rf=w30C zUVmh0YuopTLd#8GPwxk2fwH;(F-8X@A&f%-@N&;xa3b1G*x@lN1l<~cF}0tNJ|M(d zChM0~ygIrBhQW9R#WFfJ^W(>N3qG9WuQwmP(+1IQuO0B7m4S7SDDiu-S zM!w_e*UAa+tauS@Up9DKQ7r-$ciN{NFS1!Tur^lN!;+DCg!>h%$P4svtXK?@)ct56 zSwMTWSfa}&oM{0l5!Kam;ZGF6JL00RMWu3WWD}FikHzpfGN(4jOi@}U{rvaLKhZ@m z;QsYTnN*i@@WFax;R4rd?*+rukK;l~%f7Mgo zL(<&KlA(a`^UqK@sRr6((BNm%ji<#48<0%o!=!aAM| zq-TPyQ6W=}fSweDd{{-pou7-Cy3!8=pISoVTbP#rep(b=yl>*@=$N0JwA1Q8_yUYr zU8Y@0Vs3-X@j}5re^3}$V!TsnyGI@m|F+8htua46V9|HxKkK5ZN_Leo^vynYJ6bu@ z&#v)$8+ficqc76Ft8tLznBF5s#lpTF7Fi1!svfShM*RTu>g40>BkO0Z9j~H0ihjG; zt&T^vGhN$cJmI#1`N(VHtoP4&bj0FR{0hq5No`9}_h9YJ7jdM9c+uj;dlG zH6MRHHtU7BAY34 z2j|h5nD+6hI7CLux}Ro_bHr*&qLVe&?hwsoS!J5P+`lV$TLOX0GDUx%aL$l6!}yEd zekEYtCIrVPb`75yx@gfY42(B^DON4H)7V*c!p`j0MB(gWi~8F5fv07ooNm0)>z+lnH8mqj~&p0>J7V(*)Ab;PnnZEkjnRRR8+e!31DUP zfUwJR;`Nc=`{q(24j%X(UrlJ{zPY9q7M;Zo3eipww!9B9)+!qWQ{{W)Ict95`P_{8 z#;(Sp!lpzc6Crw=9_xM?OZB;^H?NQ6UbU(C#|nYV+!ETI_la5*oXnyzd1B}i_kY~) z#FD1vEqw7YHFQR;X|e2FfNt8dL(uViUTjUwK5(=3;6T;VWha+1sY!OoHbzf+1$UZ! zbxk?#OET$;9{iX`AwU23J@w_TOV&xT21nn1nzdGZjAQ-4W8d`}%I0_uKSUB)<_hC# zFJ`sXT!+Pj%tajYxn<*=Hdv9XFlC8BNa4PD;V|KGV0^L)oi#E6WuFp7x-j6_md~BWE6^=%;TcRdS-<)cuup|S8)K>e-l~K{Km-TW#2KMNaShMf6a&M>klmR)x1mNXdn~&7KSw`qSvoIY zg^GLdTi}v394{G+aI!IbM~^jh@gLQ<>OPk1un!PVfyxv*wmHT^^%JFC%+=Q?=1#;DI)3n z<&i8=`v-@D)Goq43_|W@AEu6NeUsS^tpu$n4+ImjiIrHrq*Z0KrdxY-U21ynzIqVJ zl{jc3HSu8=c1!4ZZq$O;_Yam<1@rlfy&UGILPapI%Hu^vieCCPU9P6U7wV`W6x{bt; z?6H3#$nionu2$hvw5z`RRu!}b6i%O)Y`puzBsmwJ1b96ai|sV@5x;nSVO2^5l7 zHKkePql&OIOKQ_H-2IL%_qBfXMCOuweSb_enf7buiB*7l6mpdNsWs&t)-j=g9^=z% z^Ie-Clle#^I$RkMj5QX4OVz66#zB{(OBAA)-=;b|sNTtZ`A`+2$2}$PlplH07~i05 ze+{gYag%E7+vAO{FIDaGDvldQ9MQHG-(4+R?N$6J#pR1LdcRcV$Z6N{n#Q`Q zRplz1W*pD8pttkvwb5pGgT#V@Nb**a*8LnoY34WLIgfzYCEes{V9#eFUCREZ+9B+; zwCsTvJ#BDkSP9SeBhwzn6{RAa={wtw?`vy1Cab2=65>aHHDl0t+hf`g&LkCWrJo`S z-q_XMd)<*m8$wwxe~`WrAr> z(j$}jN~Q>V@42~{aIKIWrQidaR{!6$+>?93H_U8y26p&L)fe9kKPeq~c&$B!(h6C+ z;!&HuN2n7>ZCI5?3EDrf3~bB6N4j&WWh(Q5a*c?ari9fstxP176!pTkn( zg06Y}55CSjp2|P|`y?cxsO%BS%s6C+>^-Aw#WBj>M~)Fu_8!@Bl4Nu26&c4WgyYyV zd#_{NpY#3QzsKYL>;5;DbFOn;*SX%W_v`h1=_d!~w^>fokiR$30gi;LFvw z5F_2U@bdg$K$wLKO^E9zT*A`(T#US4W_T0l64?)I&HJp|G(+~TV))_I$ zi5sBDQGkX;bN?>^MWPlSi=(iuFfFm>7k-mZf|SI+LuoAb_K7K%{>+NBb&imQ;u`Gd zaLDs5P^>qLu}Ue|jhQE$s+5yH3sE@8KG={;P=2e>*HHoy zL|k1OhtwPi7^MuAs|Rdvpl~F3od4Fa=t;ct3;qIC5U3>+oeZRRs%;y%UQzyPVmSZr zoBzD&I2@chB z>E>*bmpfSe;8y>Jp_QyLY!7cib=E*AnGd(-1pm*)w^B{?n{O?6I{?y7gWH|qHefQA zYnIoA*zAfmKYxAH3pvZH76~I~gn%QEtgFVj|IYL9f4u;IW$=KbT8Rt#g%g5z3$N04 zkjedYDFsy5Izw}+k&slPMCj+Q%SRYz(UwD#7AUwY!(N-G4!)iiJtJ=YsFA3*4C3*k zIG?iWutz>aiIc-as3-V+U}F4ll7UhD-$^E&71S)-Pw!_P(^x4j32MElpDH;86?<=o z-&9BHvAToealSJnQ&VB*fHRCHFa$%j;yW(0?%~pl=HMu$gDbRmX>GVF0GxGzxo}Ef z39JiMqd;V(vr^Fh?%@u^nMUx*KheD#B^A^BYC;CD566_p)GrKrjsF${BN;P5+4ln} zy;u}m@Vag?8wBNWoI;R)e%&jqC0acS#?B^2O4i3!E1+MFw-_(n-kE$b~yZEzx9mdeQSrSga3-Rf7kh5N{evk;^x&4*(Z~{O`E6mkv7;?X>Itf(meP#w-6l~H5L3mrX^;6^jPu@8XSoi+l4D7Ul^wkHu6A2VX$= zv6IsDn7k|VyB8(D8t-2n>o!J1GrWwZ8f`yZlt#cf^}HAdiBm;Jh`+8b|4q@hvcB5k zO-OhiEK6=Ne--Ecaa8!Xgv=tB=F(kd>~;UG;A1+rg7yl(kpO01$>UCKeepZ!i;|Dc zSEkp@SH7AhjVh7Ii>rqORGL3(qd})+LafER+73)~x*_5MJxOqZHtB=8iZU{jeLjk;(<%68k%$^}EUb6VtF`q8zF*t<%gx|1yw}uD zSw%Uei%lAX906Yr=faF2kF#S$mB@n^>LBOo_A%=2O?THh>aPO%k&45iZoCN=Xj^5W zr6-MKoQMsQ21Np0wdnU2{0U+}?4l!Oyq)v~_#bt6+w*UxeszE%p14piCd2%pv8!J)iETVJna-N$H1}6Hz8c2@Z|xbV{2xxh~byYLRs_a(D)< z>S&{K^!X_j-B8Om&uAgt^tQcfJ@(&5m|uh%cy9Txpsw$vFvgx$XlP~Ro71w9SDs3p7?fnvvlV<6vfF7>x=llNV49xs=!Y#9pXGSYV9$dfh})>PtmJcE4uewK%nF z{7gn}Dt}|r_jt)QZ|9U(8ypHDQ7sFRYw+vY>Wovjo}!)3r#%FtI3pbY*-0# zM-JrKS*FJs#;Fdw>@Q2e5qq7Jp3Mmmdn{kVNalJfu-)PtxQ4tMlB(6R5}NC-PQM?V z;0k-}o(j#@q(|%xuS>6rNy<#T4?ViczM9Vs65wXU*p?83#O1aaJ_iY3} z+yC0o9Uy%s2Iu%%zbrftu^xRoZ1s7#vzqzRVl=vz69KU3kv+D{-z60JIv4Y1z_Ubb z)czU$W65ovCkT+SIaPK!wYT_!Jyan9(5fTl9}{%F!+nu5G7 zPbXr{w^86kZDf$$qN=HY=Q0FI4XS;b*pRvLB#ReGt9D$pex2EyI75 z>oa65U1;iHA)-)+)bCs3lqU|$Nms*Mz>|m%Pyv5acoV!yX~a$~W&2nXq3q3&ulC^p zc()!7UzzAQj-#rO;b1zg4bOWn-47_i zKs#=E^=qQU3ear^ff##mI@}GvyCkF8`7{Q$FympWcA8l^ zXpf(ld0j$org}I~3*J;e3L?am#r-3os3>=A9X}Rvcf(R&#&19pkyl4e1qekGz z3?zP!lr3qx>Dk(HOF`7XBbg$${@WVf?c}G?R@O$+;Yt^Kf9-*KOCmgV=eR%c)Wfrr z`g%R6YXVp#T)mwVv2kacQkLi3fWHPLugtuH7YEH)ee|i(xeae3%w%kYV&d*P0$}aH zqhB{@Zd8mME@Ifx4IWLTGhp3)0sS->62?`e_nfUNz)-N@LDtt!&g3g7bgUvqkNKygk!V zSs1R2n`csx7W!;EPRC}%{xc}jT>uhqC-A3b7WHy*d))Jg4hf*Qa=|#(%yGz^e!3V? zJR3Np)h0H}P5G~e8bL?Mi2QcG=TlIa5zJjyU1*T&$V7sh5 zxfdvvt(#bi=&QJe%iQ8Tj}us-3ob&1<=CBzQAU8 z0&qj2!Ne3_Kn9sQ&}>Gh6U2ngyr-bhEcQ>LCoW*U;R9?~;Z)jEcYs(2kSQWrYNzz3 zh`UU?AC+?e>E}9S)x#M${ljRh44?T=xNOLJK(ECAL;j0a6VDBPg~a|>==1Db$055e}Fj1vWT z!r=M23+yV2Wr%Wz`RG}Y9`m0#=G#D(Q@8rGLC|Nt`F5}qaRTX8FzAi}FkB`20E@ro zk_^S|l|V@DI!GyQ-c4>>DWzUs1q%pqf3T8x4IX%jw3@5%cA+2{3?9R~K5=Ay1hEOu zBx3L8#{^v?F3t`Z=G#Iv6yS)AU(@q2h$Sk<@yF!>y~=vo29TpxfitkXPZ zzbB7@W^FL?f~l)b4E!G(Ie+q3ONiHW?<(Q|LRwP03ILAOh0V8b_X6cZQs;9NUBCiq zd$<@hN}HNCq|<{g6AXMKk`A!l(gKi)n#|f~FKJMA(A(F=e0Er-cS1#{ORIRFL=W` zg@n67kmz{IoB6-)DUf%qnhK33Pb4c4gj`#pTLPj_n!8zP(0sG($HOMA5@=;Z}a1Hd4#Jf7n4bSFI84OB@k%*EZnD_UP(x zXSAmI9u6T+!IFY72}^+V9D)L{_oRHSAdoJ=DnI{D>GpLn@v_Y{RHTa7f1bAhgltzZ zepvy{@`ln_*qOsaO{f{He4q<_SpYct7;fo>^pt(Na0P0hnty~G?;P9YB>u#eo=bta zZp~3oRjL!tKt#lMuG3)8y^n%i5V27$Ej(F*qL zek?L)Ee(95G6Q?_?)i!g(5WkP&L>}J0aCs)kyLt=-c8)Q-1#cnL;_M+3R%+cAWB3x zrFViT{^P=d5$!r+af&(z3&2(`tB90&T;V)85FlkolHQ{%y?3W^RnxiRN`L12!9!Kb z|MJEF;6)_6H*oa*CgLt`m|c-RuETR{!tv*hmULr$@8SyUfkTL?wbRp!;TC(#QFZCu zLw*z}yZ{<|9c6N0vZ&#b_cb zW=vxrXI;hx`#`q<)PXTn5Ct*2~xlK3B9R>V^&>TyqvPW^4sQfq)WJ}Lp3NScccr*xvKb-& z1%-nSUH~*5TYuQ#DN8rI>5m#bPv?lRPM5aJljWA-Drjs=xU2ah88m<$9(p=BIIxor zuqFsk6ssTSiU5i#h{#)eRoJBj%J6$fEI;A&XqeKFT*?oGd$;|H+xGi7_j}T*y1X2} zwd2)HzU=(avpc~2QYBK;;zIm<{rs5eM`}rDY0RZ^|M5uwpyx8aHg;v7z<42q7hE&3~*n##=kVwSG{Y5Nu_) zHycc)!xw3t*^CV=ib3mSy*1w83|>{n&}vSINGR%0NMu5Ku8%R6Lv)_4Io-rMWluYfoE+h>|dbz3!z1iuX|l%IAJ z?Hz5RkYwDaI+L1{HSFME_d4~Y>Ppa-;JSt}`daMO3^~*$SRaxAyBh^|VHQlnTb7I5 zN@vmGF^ZX8?o^wfYMi8KzT~#4m(+Uq`}iIGCcN$jIVNp_m1x)|@}520Fy`>yvr&%q zTYq2NPv*g(NZ>cne5Wa6N@$Mw6k-AU#rMI;dy1zr)+tD(y}iEoi4V_A6uu7s^t}PK zKKA4&6ClO^Bz&SsEs@!}bjx$W4a=$$yliWfeCuXlq3wO*vfey*RmR?ZSqZy(Y?oE8 z`+oipQ{+ud&7QT*F+V@tpmqauxh0Pj=Z`rc@J|-xq%iXTBL&TkDa%xJLAWzk{*hpG zIc?H1h0pVShnuo4+;4GTks7gCL)*|zY0KY5#Ozfqn0%8)M1szl4%hm(>+-1J85r+# z3)(6zX4cvGLJGuCrWx(kg4lW9yHTsLRds&X)j$dU_yD@vSg2&gPcO9Krxm~PJ^t^` zTWgrj=iKLC_$Ckzwr1VUwaaWr$K2loB*MxW|>|agmBKvr! zXXD3#pp9ax14$f_ii7sB<$kdAYOtR?WF(qmB?cWlN4vV&kpDCg6(3c9jkW#&0hEqc)uz>Fj z0ug$npYjuiPQ-I`oQw+wt!%JjTHbNB4T~7pg{G81E+O`SPgBgr!zmNpIo)2AXHp!J zt{Hr?8$A?f95I4TQ_-yBjwng3aMV{$*de;C6b2g~X!H<4@-7|Q_oC$AR;CvnF79`1 zdWCSlA<(s+R#K;j2JBfsJAQ9B^>)I>mJ?wTq?aPqoJQK&9cchyx$^a z_%P#K^~7rBx0Nh3gY|q)FuR26Cr5*X4ywaAntG48osXt63^Y$U(|zdtXy=)k?=ir^I$f9HhKjrV{OGPkhXi7lsR@ zx(0bkRn9tPawidGekYbxj)?~)Y#%hj z7jdgx5%C9 zKZI6Bx^`noi_`C)DjCdu74U-1`?^GiL*<3(Rx2;;-TS}7 z5)p%2T(;x_kuT1-)Qy!B z%!0hSMHudwztfo5YX?9WOcg>CP=*5cTS%eA(pz42k*{DE3Gmoeu58@VTC*iH2?0G> z0-nzhD!)P#H5TwHR(&Q{7Nwk4YKYMiIh85}37fIm2&sHQ_jw^EZF4=sAs*H}J-Qg4 zB=|_mC>lT1e(AK{D?{GJjWEP9hmSKtslAbKg!oW}5nca{Zh|;inm9^g4H1QJ^v?1k zm&6Xh_V+bhS&37n#vf5Cu9l=ytfe+15O+pTOH?q1I59Af8hOy_(Dx-Om6dIZpqgVTFrD; zGu(1KAKW9fCKlzE2od%jrH5Fm{;||~f(+VAb%@zW&|n+v751_?$+S0J+HKGlZMDY2 z9-OgbsNcLb68rNy2eG+)2p5Jp|EVTsY3Mg#S<3aKyRlxoCu?Wrnz1!RoyM7>Sa@*% zu9i<7&$2vMrDX(bCA|pEExgnq?y8SbRzj-<151dj*vps9o&eb>K$lkI9lExll{vZ! zuhK(4_~y;GY*K8PD`ra(4_oyvj1kjvW?Kpw8I1|4pNC=35*Rye?5`5pu1vc>h`~!g z$_S8oDqj&+Sh*Mg2r|$JfI*GJHR_1+{fkfpZ{S9;G~efb8%lySWiF~Pc~Xi_)l2-t zU(`oo5X*Dulr&=*b0W#%i?HeMKT8W8uM_s?D7Hr~>}aKDsG#wGhY1_AWU}k`Jjh#w z8HEY|tB*YtEOYs#-4##HtccE(q!XF-mOzW9DZQm?P0j8)$%f5~HHjv6!9D+em~!?v zL^ayiy|HFVJX|KU*!rmgDC$-ET;K>qi*Z>`4TXKr`C|trZT7=ou3kU69qVqMAPtQM zEd^4)jVVnK9X@bGi>5xITYj7uawrgbItxPY?-z-MZMuQ8e>( z=mbh4aATG)W*fr%?JewYM7>MybKh;DVf5}<&kcBgw709NL$n~1EptEGE@Wm%(*npN zK$7{$Dpa^gky&)ahUHp_j0|Y5x0A1;u-BJ9yQY_J{ZtmvOWhn&UlbmAQ#ri*TNc%7 z9UvxhT83$FKDDmZQ;v_Y3&)x?NJxEhp_sFp|bE|0h*Nw!4c!{EZj@4?j(IbYYPX2UDnJE_@ zc(IE2{CE$_qX8Hee4=dO724(B2(XTKq2B-Yi9S|3g3+#WDc0S?fK6Egrh2ya84s;M)a&3SfhnauW59oL~I@N1!3Q zN_WD@{4z5d*o`LbQh5T7ezp=q$DC?}1Rql%`m9qTd^!~n;g1&!a=~W|m@PBg9PR|z z+NXs{c8;C6H|F1x9r-MGH(szi%e_29Dm58NThA&%dO(;g14>J}Y}z3sGAxHrqd6EP zgM634lG5ifH$ItxME6{Km@7b)&w{XKlE&vC`_)9?C;+54To)5abh0EoQXyWp>BUHuKP(Zw+}CPH^MwYjr0es^grYegI>qww8PA-*2@v=In}QvFP1(iz_;9Anzzm z8_aR*zl|#OTjEJ0i5%g_m_7`nU?z&qblIyciV7sr0_EmZIQ=P73C(P$+3<}M(5pIe zc4nMH2WBiaMHn+MIm8X702|=q076NB5>V?(;or{$Fr~_1@izj zr_eG}qrC(a#rPF#p?`*R|4hw>PZITf7UPt6a4}giG`NzA4S6ahW4odSSL3x))mlBT1H#hXM><_c=_iGuBMDuV3G zL_M4;8K`8n=RYa+u5Lb@$K|iVi)ZUQe{ZIP({Ze6%hLyhrqfrc){l|^m;%R##L?*R zV~_erm-r4Ty9A{V3$^EPoW_FBeclp`I7lImyN}0GB$_?oEn%GgVT93pG6=^(>tX=MtEpO-Lt`r2l}L&<`}qyD4l(W(|3N$Q#|aFcTG9P zbzb1K`hma4h6~wbPWn|^-Y-Du;^^2|@m(&+<9ndmHrbRiJK07EJ~znPt{(URd*R>t zyH3BpzW#Gv7`+O|vaU5B2kL_?;Aa9!9ExA zFiuB;e(Rt8(+ejlQ4T*Ce3K@sFD7*7Pl&4z@J*Is{yCoYf#Ba+JyHGE44x&2B=G^WBH>q1X|b{PC9>>>8lpx1 zlhuB$9vAb>d^iNZV$Vw2Lee$}&hQ^ix@YqDR)a#}!J0zpn0nUF4)=)e`9WiKOGz`P z=j2iT_o15lrPj$31rJO4QW)|h$CVpOr8owAD_$y@UziYLIM1NHuZp-Z$wvF=dLmQ` zE$d=K^Fn3L$l^ojDAgLXH=>r_WAM434ULH zw~(-T>lbZQzX@btNUa5~8DpC%>9~OC=!1e~{MQFDgJUz$d@DV1)H{fr8Si!<9_iiNJWq_FKn*Fa@8po4l(OHp@SoY$k@PNP=DO0+ zNUv#eR{ojxuU4zd}m zZnHgFRu3dY7-~Gl_PzrxR(eeJkAG76 z9l5%RE3z9SID~=;?YOiS>c;pMWb1P{REk3`^T}Ef& zz6CBbXhXCCVX0SHfL~|8oo}N4N4bZKO-DcE3wN$#6z^OlfeBt!WkfsS+sW)eS_kO!<9POmSO(Aq?V?IJ;XI)lDGrT~spu2sw zFLvdlMf4DHPAbS{6+brp-7_|<7j3p#a?|2U&U0e#_3r~j>3cCPerF|J?2i?(=%$!Y zc}>c?SbqX?C%^9+H?kf4x_W;CF;v>t3^_)85wi(s)gbqP_sDZG7#8u>_T9 zooVHk>ZAL^BcX|Rtdz`uJc2%u$rF5=XfqUT-r_ z((f#q^e^4KhK0S{W##@(J#*aaaSKL=B1DA48R7BZjL4c};Gq3BALg4pF*Fzcd*v%sipe%> zwela#U63=J6g5VX7OG;Hw(o(+h`a8dGQF>$O9SJ7riYqT^+n6#ar=~HLC+jCKdTb$H+E~$oIfWLdnOx!OU^xcjxolSLTFj zP-;%&bdw(&L6}glHbyrXebjmlje!7DdJ(mQZIc^+8)afTZ9BPP~`Etjq-ugp9vc&ABIm)hOMg3+k5cm+Li(-#f&`q4o zqBdsGy+6V%vvNmq`R4T{)V3ty!~@3&$Sa2;I;Skor0D7c;+f5_vTqs`F92NPdFh;E zh@%GNpWBe$y~$5Ie~qrwwjwI`&Um5fegigRT#TtJn^Gm2Srwa~IUMpB@chOvcu1Gh zUoaiEBhkJ)ch=-@Xt7j_Bm`(B%B#GK`Lf~OvmjP7*@F5>Qa-x+mzn(F0T3&e`y1Nl z1QH5<`~XaVv_-v=1y`(CyN=nAXdb6oAQuAS%u^C?Q<>KIi9yA?C1PBY7G%J{90S&V z=I-w(P45|}6Vk+@p0Baxld>F|6%uQOn01SF6cM}O&lU|j+5Q=?F!(JUIASADn#qZp zUhYhl6AxWvU7&$`0byt4a4CXSD4L9j0O%=^O3U676^s~KR0lGye&fR)aYenJ8eta^ zt^BxobdV?S%PigvwXG|T@V@-;20sJ(q|t;L84?+#SMJ+luO4VG&Nq9Ypj4cIsg_FD83QZybS%_D|3y%O3YD)LSDJe2EtTT?ko zb(sA?9p=VVgJ{}@hLs6`yMg)zVYL8;zGj0>3n-{d16B|K>`-M6@-r&8024ZFv+=2{ zfojoTP*f(%kWPs2eX{Q{?NpOn-8nH~{F;Bm)^1*i6r>J&q8%&k{a|pck%4!r zoCgkjT0Z_|t?_-tUBFK(tE>C;r@JvobkGpbiVAEL^ZwjeUtj;oS{dA5{%Dcn!e3@r zoNsz!(+gr#2Lwtd@y#ke^x(|}kWP>V4~h=L;pZ*s1VjB`*pJDJmf)_`y;Jc7^;U|B zg9>X{l(bIDTM8QGfiBV0TP~jHE!pZH*|luQgVpcm1K$9FyFMAGnHqKuIp${4T^}a- zmVEToqn~}e!ne}FfxZBoTY~4%fir1q7PSptT|u_L1kTy-EKnXZY=lc6gSI^IzV72o zRBF5k6&v8YJ};&Ove7u-gdecs^l8X(%CwG%G8G(0E9-`UNAC=O>?#h;Bw&eh!^;Z; zW_x_#9N!~45@r=K<~NuRn54#18pvQvKqsru6>J&*v}=W(oi5RX=Q zL@=Zb-!la?DqvD_1#W^zRN5Hl@A-gd9wu}C00-#C>*QDs2%H%MTU(Hd(86_=$pds= zrG1vE5FMi7mjM{UKpGbgL<6cGL9XHNXdL+*92wlyho^_uVKxwAiuq4O0mdTmNu_@@ z;{$Dcacs)}8KH1IC}7AjE;bxenz@IcWnp2_g%cz?!fHR>0BLGhpgt_%g5aHjbj&ri_$sh_oX7w@u) zk}pFW^lY2D^*sEzaB!0mZZFxcdVnhgyf{1Nn5onG9t^(5ahdMg^^%6N^76QoVrMz+ zr%&j}GJr+L2fXPB@HgeT?lorGrT9%!2EdQ}n*~i6caKR1^kL9-@P|bR5N49Te`!`T!xbf5?ND6Bp2U{IF?TvW>+TqK44`rnKUIoO`>^542ta*( zi)xA7;%Uez9S#5|ebPqp$lpLAtlsTPRRkOeHhky-4&4dQ&Qa|H!Y)0vz zL8}Rx(|boqYn-x&7i)V9c!lA%k(sD7hfW6h`F>u8vAg3Lju znz5V_hP{(V*G`6=9&Q}_WHi6~DZL`M?qOAo7H%2qNg%OY!`Gm*XRD%To@CH@vp1Sl z@8FmT{bOhCAA?>$jNg9HBow;q{a|Run`>}O&Qk05OGF%|&CXbkT!{2&mR$z3 zxnz6P?>TLd)!{jJddVmkQdl8j=B(75%S^I zi0htyew6H}OvG~9y)gqkdtw8&L)!&OY100$bxm*R(k52LK0Wwv3a1ww<%v1-Gyl4> zXn!Ye^EE@l$wYA?y2(kTf15{4cZIKoKYMbMC>#?u*~mK5Qrx@wv8nb|fc0bv=>$aP zEXUtFypQ_R)Q7-e7%dHf1sc2Np+h8o*5JTDh}o=Z5#k@NWO#9Uv;wQR=b7C0keREY zJ{F1*nf-}$KD%PIJhPw`*GfS0W=xjxVFLVA(s&F86&i0Oy2puVla|V=p#GKLKsLqN zujlePGP|}_)Xm04=vXvqbhR`}ASoLt&a*_(3kKI)7yn3{IrviZiJuIqZAi)1i`{rE zg(N-dRLZ=FYwnS8askB+#}L`3O3hs`QD2K^ zg{^#e`zVxsonh5q3J>f>8wb6yH`Bu@c`d9ya4do3AFC_6YIEL3eAVw!g>V zH?K7>y`rFuxVa=xAMeZkR{gF6Tm zrXASUJ$nk(HhfsG#{PJ7>C;xCMt*rsP-(JSl@_aKlToI$B-t-C#?|jvi$+&&pXHoK zbYFMDP#DIY!p6dtW#jLhvhd$N3Ze9+&o9|z#wjb7qM-f|{!%_^APr9w)3f(IK%X6*>iSQo;Haqq*1f9z z@vcpGqyin8p9w5gl@niuk&jDSm` z=Sed*QSBnN9(;qYeE^EdD3HDk@;tFfQ^e4_HxVOC>tWj`UU?e|DHVw`r_|i~>mCci zdn*f^1-T#3&9IMOTNxLSx~3%`ZKXS)9Z^NLhMdmSUMB2%FB&D(Ec$wh^V7L1GIXR* z6(rNLV1DrNEuF6ha+#{=x0zT|6$*vU6;^D*g3Osb^UkTnFxQjQ+6{@tiX>o>lI0Sr zy~DK_+X{;gWsl;rrh)eG2!ZENs+O?55ydPGaoTzBuPOM6jwxD?2Jv6PMhJ;zS&FkA zs~C+d?l~T=&UoJHW~%Z0*UMJE-wSB^uPpMquariNTz*ByHsgD@*k4Ic-9D4r|BK&i z)3Ww`r6*K9I`WnD7f$$l{kv%D>F?@EmKu$K^RB!n!kj41~c5j^U zY~VogjAOGyi~hTFifVozg87a~vxr`pVtE(-n)G6&UTKL~c>?Z4Rp4$dVE^>|tA<5N zaOhfDi@L)CDWW)Cst!9)d2w!fFe&9q_zBWx#ia7CET&1v4oV*Tg)^`B7?=+mP*vv4ZK<*8>?Tqe|f)^v4Tue+3(hc1K830rtm zA~Y#cHH=2!1#6eV)Vv5Kz@l$Whqz?Vtm`B2jX0c1l@m0EIDYzauogbL3YC2EZqpw< zvl6A3kf3waCf)q^cQayBi%sb3;64>g)1U#-6I9-tx-xP}pEqVRCE^4PiLbJFp}t?C z;xJIX!KIxXm*Q;5JuG>?VVrXJ-lc6AQv_i;r&1{5I=2-Vj#wEJ!qBw{!H*{^eIy#|m|zzW?Td zEaxm3XE-NUS_5@|+%jgEg!42Rl||j`CYp+M?e-dqUBs}i5JK13!dU@bM#10sEf_yJ z5x?J5KVnGTV~XitpGMr;*xnB;eN3!31!^EC@xagCudKh{Td)qfGV3O}5lbGTYYkY> z=y<(HA5|~Hh6d$YW|UiI22Y@w6-mf%&+B`8-J9z{Q?5j&R(w`~Y;BVURhNyo&gg3R zoywv0Ltxqu_HXF*0FD-n9WB3|j5On$;&SlR1k$^}3#*hDmzEgfV3QT8#0DPyKvDtB za2>BEV)#21bU~9X6bE}ko4meHw&v*W4(7$))7TsIoot=DSe{b$ki9p>LZtbPid}sE zb`fA)14tn7EdI$9j^u(5CBw6}?K3ZrKFkD_72_>SQFls;9~sK%XW^5NGrU~Q0ls>Q zoqu^T6FpScfmXC&D7(2_NLl;@w;Pq*6qW*euNrL*mkN-`{g0w0eY$8G%uBGu2`1#) zO(bRqR4Bjx3H!(3z`>~ZHVV8UtC*+vL2cEX0`p69jxR&6q9xWiag$q*_v@OXTP1D2 zQcMdt@!o)f{iysas#tkR07C`!A3= zpYooGYO{<@d%3uj&H=I>VDslQ;GPxyqIoX~EOPN;owI?WWQLYn=HM7;~d$)IBv~dA2a2pT6U+iT!PKTM@RTOhu zykWlco*HcRsp%rbL}6{D=7HeT$NMU^_LsO9eli#KA9)P;o9pUldpY1sU?E5g#xl3f zw+$68q(Qsmll-H0sl9%9E;!;C65~{e5>`!mS|*p*K2 zMtjfxB;~JiTdb_AO1y4%RCm*gKBk}i=^x{TKg!I%WG;qf=F}~)l;Chx@chL7d){wz zp#OAa&x3L?D^lN=_fK(Ijju!U!EXS`jiTnx`ix3pTZuFK7jM$ zvo~aeO@3Mjuobt1z5MdB&3r;&nyi5t1z_2OPI~jOUj7ya(5vwJ=j+}S;q?B+aJ>Dy zjtjUz!d(4&NjsRRKWn;gOE1X%frJBAReX^fWcd_Ztu75RF1UiOlEytXoPQjW^g`au z&aosG#q)LqhW0Uh64_tse^~S%82a_lLM46YK2R033KAkIB8K=??<(Rj=+QNR-%QoA z+P&Wb@;nr{l{{6zycjUe0j4xFZq2&iVT0lu&P}X)?~8_}%&irbc7kLGxYYo{P-C;Q zBq$XCS1s?)n(=Us`y-$$VBG}LzkG} zi9nXmJPEja+7*LSSty~GTNfooH&ESO7z`3RfZqmA0P0MwF3HsK?kHl!(1)kDr!T;o zlls#h?f-fKz*WYnINCQ#D=X|#dpGJH|)7Q+4u`A() zMa^e0Zi8z8Bm-pE4^TGUPUHMqE@@tWutcfIqcb@0Ism_9@nD#q&9!=PLj`Md1Oky? znkzkYMHGw=E{?M1X7i#J8QPb|#sE+95~v4)T@1Jv{KsxFY6|&yuf;U7VfN+q)LJ-r zQ5`eA^F?nP=95qE&$`cpF&9K8nZ15O{%ruGY=zWmT99=C?MHl}>` zQ>zQe2Cngz{n_g}(SqkD>`%LoT!PyBnHy#l%<*$>eX-lkLMn|LBKlI96<4je-~x=$ z)@C=MR^s(9{1uUnSg+k-2N5CQwb*}Pj{fqLJ4t3GUPMkzX`IxTFifvDreSAV~Z7P;mG2fG*tze zRVryZ%u@G@zQ-fZBSt<1;8^$siEj zzs+Ck{xhey=hNVWeJ|thp+#_NrSFlUZGi8x z++8RdJ>~t)T&2~eV2^2*#@q5S>quD!&vg+Z(aOzU5Iz@5x~^sK2eQqVrYquEwF8^; z%7q*Hqe^V<>N?W~)&mKtUq4PlC#!T(GRN`K3Z^18R4&R5kKZTgS?)EL>dfAp6AG;8 z-g5}PLHn1rjWOzXPUjil(lMu6b~aglLxMrCkKx{$L527@PF?`RIpR%|{XdvXR|b}h*?GN#=$Xot_# zXTQqP8njVf&ScyK)EJ0Ym`AH%q)Xe1`u2OP_mj8xPkjOCxZ0R1A|dnJ`Uj&MF1L>~ z)%=1YwEE74Z@z2IpoXqzbf(&VAn%x(1-nc_eW%ICj~JfQh>6!GZk9&zz8!@i7{>*C zDhGeOO(rx&AF0(5*o!AR->}GRUXYc*`#&^YcRbbK|F7DBSZwO!+P?&tgXJs$q#df)p#_q@;Z^;~sJMqY00MYnTN z+wYGr*6#JWpek(eoe&yZvjoUw3beW;*}Wla3J{(G1VE)s`%P|CR)%^3CDDdqlgW$l zIHR40Npt$*R=D;`J-h1u4t!>`D(IXChm)Au7agg_JAQqh1^f(c=0Dm7M-EuXXJ6*{ zZ3p>3(i9P{Gxk{W1MDW#cHR1BBKhG^PeXtZ-;VHIU0+AhyQ$DirS#LQv#MU~*`fM2 z2rHb)S&;HJ`c6)bd{KPs5rMZuhQOaN)8B5rRB0-I+g`hJ82_BXXCk^C`93L8;_rFP zAr!Zyd`}X$3F-Rq5CK8lD~sK&MeC-h8$T4i+ce)Z&_bN9v(?S}?ZXC4>T7F4>U5Zc ztVWUItF&STv}lP|)sB6CN`dhqH0~z|(i%CEbMVu#V#?!K9BUYPLIf8_&DrMIT*omY zdISnDy($#^$E{K0I?)cMB~GpWa_mQg?8n^#4h-?4!q9S;&&JZo5lN$C<;X=|Q!!1a z%Y4%JzpGv#0Q*u^S1mak2vsTJHT(sS+NpyBf*N%s`5+oRxdgnE1$6NuW;pJo3w;PlFaqB`sE45uQF{dZ>s>?}$%QbmpC~PjU?OD%s;2eNj+gfh~ z7-yQ-3IVtamUPe`zm{>n+x1Sax&#<+UfuUFWUs57WZHd#=PvvpeTm%#l~EikFrzDm z{YXvK+E^j52CYY3k=UA+^vbwHF`YtRtt7{R+S503&GJDjlfcW?{bYH+4bDEh-A{0f z(^uQh6=Zi2+?EN^duMsl=R7{p;0vn(*KMxOHvoAsGvwe*`kW}5C--C37{_<(mJTGB ziSxxK7k(7%@6wgKbI>?FEn<4XGB~G`N+zSOGl2(`s7#)2dEd}Xvd8Ec>e6}=5C2QX zH;>|?DKvR=#k;JeICS-^i46(r=hPWJX64rF?kl8g@XE27d=7GDAuMRJJwH;vpOnAw z^Q606uFFcCaGg6;bGF!gc71-yM$TSP!}=>c3%!h|&q-ku@t3ZDuCanUVz9E@8xC}f z1iQ+=Snh8<8W~HOl{%RXUz9?pks~J9`_!`z^Ax10;#>T6QtYQ2OL!(_r12@3q0M|5 z%rIx82KfD8kU%HH*mRS+TfAUgXP)jFx4KbBmNBiWYr3;Hac9evQS*;nEj!*NeL|{4 zcLGHBjmxi(h!X{duSc*|C8XCqnIc`Tb`*2$*N-fRxFYBP9n%Z_+GAF9tF6NMd<)5d z{J%*}auv;|2X;BBR=Ekp?CYZO=rqB4f!7Bc; z{}<~u*?qVIKMhCnua{jn&@e7i;Wb->V%^7SPIp5UKi2IQw2x+6-S|86*Q_TO^A)Yz zF95sKEhAuq5xhOtyZvD|cZs?(bu(EY5vCf)!N*ibR!aW692zshM1=G5C2I3wE(9t4 z)fuOb!9s;TbN-43@UIUhY#kjkD897$!OR$>AqQM<-Wn7)W&Mm!qMIM|(kQ5$koo#b zPp!Yo(KON$l&jvUBhRgq9o>jLdq1@~F(I0pQu_!GE$h-&?Kp5qzm(bR7dpjew zQYCWx6I8>BN-1*O`hs}ngX~>j&QS0-G~wUOHFI{gBEN(2T0uVl3@1!y4+xaAROJhP z`Pk}xCUL&i%mA$D;M@=R(E-*GvgH%EMrf9*oh%~>K1FaG{~cuUEoZEQ^#Ks`dkgmo zdmu?&(59(A3Lnbkno6KR5UxHi7UzQ12wOwWZyy8t3A>Q;e_C0|+rJ`r!DQ@OukYU{ z()n@Qg4A8w8d=AkcM$OJekm67@e$ztli_;it=%ODXt~a%oSEhlVnkSLN?l+_hvWUv zc>A#(DbvH?cy2w>)(MzBY<_>ss4@U&g+)XFPYW`<{u_jt=d#aGX$e5Gh=vu8pCe>8l?q}5S!vHo(5!~ZL#|0vj%DO6+V z>VSpPRt@~-M5?&y?{jdb1nohhs!Y&LByO5;Oy3U?W7Z>2c;XfJ?^lN*SF23?w7yLb zX0ySO7QlzIq^a{v+X;*5$HxTkcVXpVCWggN`=U`e^%q8c``HgQknU!9D_V(5Fvb!j zC8%`+8$O4olLT+5p@BgU26qFOJceG*g9pEHWZQ7vHwL-Pf}nC~&_X_CL?A6S2gJBy z0-VLqmO~W{WW=~&Z0WAQJ=gm0ilh1;(PqZEV-*F=TLVhci727{{6wukr)8 zypPe>;SUXe`jRz3$r!V_%iqylLm$^3hYj0EH}dI zBrVl}8Fs?PM!?2`>}N?vujOE?d#?+I!4VwPZ1K$=l>6kpnE z#GvCHj5^u`4J0AIl@vUq>ykMGS-8z zmeVJ$K|)I)m>?YVOA8(N>bBL=#`bvuM1N5E)}NS;jxKDiH0jeiXyp;dR7-&_HSC|V;ybgQUabNw+0Q@K|4NyxsgyZN4*5insI96|}BjKL2BM7qk8ty9ABs= zQL1Tx7;nd!#f$jautfM*1u$j9bYV5Grx*%A`s=9+DLa5fd>Q0(Xb32RDaid>IgAJy z)7G=!k(m5;3qTdqE*%cMwikTb3A*jUz4WwlV1BD1L5$&4LB__PH6^U=3IpDtPxmnu zNEm0q_gj^T2!ztZ&W#Tg z)wkDvKF{lSh|2PqvpWMx=0amqOh9;yx`=-3hbsjVE)vV@zyLg&=Mdw>0K~B*2Ck>O zTjcLFIej3MK%wgVn}spf4a~xIwii#P+c> z*s{R2lQSHed29ZXoYj2}HW9F!gYY{5<&#^3=9R!m12bs^jIvcBaq?{GVj#a=QZm4?Zdx7K)5T23b3?% zR9`bTr|jf+kmynXD&*#Otz&Ql&TEW%*)MXc83`>Xwzjyy|0{X%wxcCzCt0_*)6St{ z(TP3NT#n6)$IvNn0OIfc^=y4%0}NMrUoVN9Gky4Gey`lS!j?XZy1{rWR-ocl%6O5H z<&V5tbz+9tZz2Ij1sf`sYn z%J;4Pu21o4^jp_nIxJ#u%B1>N-TZ!{`nckbSrALexW$Z>g zTBS#)9OBu<(DQWE%KNDfh>9MQh#XeFn^VEsPtL2<^?k$_^4Zk5D#wC_@aFw27{jbHX;y5s@y6_JSuD4)r_@q|DL8QI?TPjrY>- zrhYN%qq>DJZsz+0k1AL_Xo-*#D7urzQ)Gm{V}Xy>R(PV%n5vXJwk1S@Kx%1uE95E; zJG&1Eh;_21;X4B7kv-i7KLujPc;4_Cc01c zEywt{-XFrMP;bCEA^aa8PWU3m86HCNHLc|#ud^kdav*>de+hrQE2n9Kpr9#FC6b|p z_AFSj_2l71*jkS?nBy9{r^R~voj)LTn+M??piSCjR&~iy8QiW^%{k&@-G;XlJ&IIU z*&fT=baC)cy9JVZAF315CIIs8rW+;!fo-yCi|$`vr;NH%$xxLX08 zQVyj*EqN7SLg@!7TAak*B~h^Sw7r~rQx?r4XvY0swQD4j>&5KwZciI*;L}$Uvm`?k zb#4lEAXcn%&8N)0&YyZ?By5LnQu{>P9Bv)dV57dklt-b73Cc8ngW&GsBJ0|pw zw+xD(e%z&x9eQcdKaG_3Ey%W7j6@V{I6Zt#f*btq1Iw@=`3za9-fV~Zs39c+QCIQQ z4K*NL^4h<3plfZIQazfqmPB6_j51E^s#yG!)@zLLW7G}BN^O|(`*+MfIK-k-kFkob z^2hqo-?`IM924tE8|z6NVkdKxY)T|&RqrX0r!|nw(C-vp0+&`s8>&1+AUAupc_#m_ z6{NLY4mH0S*)xC?Iyb&pXN`ob->>MKdd7as1IlVpoG7?S{P`ouNJyboYb1MHFoR0+ zxD(QPh%$LM@8K{*WvAt_$Xl>rlqhd1$-KP>#J`JHpSg+VO{~l-Uq4{i3oMcB1uh@- zl!cv}B!}Cqh&RUN?v-)B ztK!X2Im~}iSjR`D^=6z`@o(E+Gbs`XGO<<)IFoc!hHkvCv6K-%;q7H{j`e66!^7wL z>eLRF47*7RVl8Y#!jema_T1EBZ34Xwa3|IzLaF)Y80?}aQ7Rfji!u>*K>Q%O!Yx(s z_EY(Sx`{rMrSKI8-Y!u;?}x(A{^~ZGl0oP#-dqXJXPZqO7dsF?%P8`b_*Gi8ty&)8 z#69@4>MHe&QL(NkuL4dQ1!IP+Q;a=IvMYNR6Q6WZJqQ~bL7{mKL;08|}phzJ^FqEY9%jaRy1QPEv%cr(*Zg@d~q zNzpGrP{LeNQt{0xENr*5*_=l9r3AJqs{jt?yI79#kkd9SnHw}R3he~2=$H_fr1&?0L|S-` z?ECYl$e$xPlxy3HAy*tD%LG4IcB6J+Qf}aP_3sp|&Xx5M%P!8`1bDbHKLFlf6yUsp zCA?in)gTKC)HnbOHPC;5eO6NOSr!00P~_)DPz~8y7*Ov6LT|z6%n!X|`u>zXZ(UTy z7nw8Q0B(sy5Y_n+BybhPMm^?jn8Gyw_V&JHV>}ywdtZQ!U&l6U@rCnUxw{xGIiPO; zX7p^!KBcheTED&(I z)=vvxpCN$H88Jfy(+C%zd09EMDlZ}|j5%&M7*u{QHd0rcbkC1(e_ZiqCh&rEDSOZ@ z*F*Ckn3@$ww13nS`Xf@#jdhpr62KsnUl&tNZ2o4xe%=cC zV_hIrF}2}o1d195#F2q0ceU#^HLSE2VAoIhqTW#l9MbgSUi`IB3D;X+xd3%GM$zpD z%8D}flUeHg#v)DL1Bt--C;z#CmMo0Q0jQ*%cN?ccs}^Mfff<_$2x@l|m}LP%w@8rS zj_DRAAZJu|YWnie&++FR40Xcy3AYRYYm-f-1|A_)X#QZ}Y3Tp9djNvxVQEmqa z02sf4D-8!92|9_Qx4Kwj*RC0W%|y8>&0}Ss4N}-i7cTiSu)#yX?MraWy#{2=d!Ijo zIiEn|Hxppe)yp~zrM~jFWBk{j)ckOOIsq6^#NCWjHPWA(TXcwA4ty`ti6KhoUWRgLTcqXUpJF^Etc!$4pVXf9QgxTi@$LgX4Z z55xkPwq-3*|J}5rLEjN{AV|w>V4GP>8qj>wQ zEKoHClrIN-Hz>ZdE#1zH0_;Jv2Sln^&S(28%B^}HCWdc8{WV4l4P>|;q^(z&;C+cb zZ}KWo%4XFUelEKUZhSKaZ4W)0W{w#mGm6i;0x0ovGv}J~daL@HNQ3fmYY8v_SzG_? zjMkhryhTId<&?U44*p%lDTvJbq%sy8(SHbh!!}>=y(o68YTAi zMfDA&{?7i-TLrZVx{%_7V~&XZ{cq%}fZ5OfD3^Pw#EGeA?kP7p=GsGR!3BYK1=EdP z)p79#qrN6Ygzz(70)w<<_G#M&$XO8cv@oHn93u0jlxsf!lX-v%`~x!+XoI*fm=$4W zAI_Lr6o8Yeyv9ZI zCmzF*e#=#SpYm)iQj47c^ab_kzdjL?l+K;!08*b`(ofW_Vpd4XfLH_0(6N}I9tZ4% z7_>k}gWIgR@`HeV8theHE4!y^iB)lcI0#6(1$FuT+=pc7i({X`aw5p~Tb1w0{UN9U zMiwpdEhSE%JTMnhT52c4de<=Sg$hV)0BQ(l zFr~Bk{hs_!{ltYZ+ikQ2kn8|bel4YsrGMs!tBfP73%w1Ml^hDIh%YlUGd-GJ|J36{ z-Oec*Bhd#g3oOb+8{ludWZ;Jl_@c^Xw!8rNIgDU&WCbHO3Mu-k2PukBL-4yGio&T7 z1?(JPTSc*>lNv2^{#T@hUy043?wg?TI{0>vCZ%q$+_tmGw+1^x z)Wd%{7^YZ$aV)-P7uoQvu+Ed2=fOYQ8aaNw{w{u%J9C8)Dwx>>gu%*Do~W^wE0gf2 z#w{44IB>=_uC8uu?lMk|te-@+L%aa21s2?Kxge4rcji!spSWX*pH`evn|uanf~TJ} zH}fsi!$TK5`e3I=#p1L5^Y%vTJUEla`Q}dPD>)2;4r@4AX;F%ZUyfhge7;YiFQA4cKuU^FP*vfPPQF^}ZSb)z z(2v{W)y}7F@G57_p+S*#!R2FVev=P6(OWpwd4*?c$?p>j?*X_3T_P zHb>k7_siPSxCHE)OV+)hN;7vPv=`pPEouvw!k zEc~-xZ*8D%#;#TCSq^-&Tz6|Y`zU-1m+dXL`J8yV1Hqcb6eGKg@DM4>^{hZ8#T<2C z%IMp>>v}@uda0M7*=DWuwa?GjMW$AORb39nod$7L6ufqS4b4aCn|FNtaYbt`k;Nb0{L`=WI2$p8}EfIwzd*8g0uNPHwXk)o%p7TK^}eUDSVl??5h z+^>n4k5j@coHjaMeAKxk&x$MH{TFN=Y7?%-Jj{@ z5@NlV^yjDLe2Ow3d*j2M0Gw-|!YXi<_x8Akl-BdweLD&1XQyB`_L`+Lbw3xImsyUz_65C3VB3n*YE zQmExTN(mcgAbQ}=@5r!lXTvI`~qGG1kpAfdv=JZ_HFkb|&?f zMU#rNKJKzi<|NWT!^Y3j7_=HJP)g8(C&i&)PP6t5J<^uT!&&~H4SdRsjoI_4PcU+ zSl4*XXRyJS?PxW~I+5~}58N8{XL6Kn96V*g4v&GbxKq_3v$k_2v}NmEvLRd7z%nhu zK4;Za7wRDDmB4bW9b8fWuZ7k9Cye~vrGwgxAOB6;niOBe(PG$6;H*GVt^)(W=#WyV z?{CBV2E}TEnIF_IM#W~Om_JW_pb+v+&>?7RtSjjpBHB)eSAqErUn9)$SuQ+m7<;p5 zQpj=e6%*6bK@@a~I#nr$;EP2=vqN9OKc{qKq3JyFj2m^RmKX|j7Kfz$?S zzcgP{LSwHa$O|L(P|1&lY%gMT!=B8yY$sAyB9omABzPfkG8tj)!9HH&nym4+vb%9U z;}G94-gnQcc{)U&Lp@g@PEGT-VpGS$wspwosmnvw2kMXwUpb7H`w1m(P+Om;Eq<_D z!k<=?dT%&7%%}Rq-8eP~N6w21$13x>YQtLV@~Ls$MNd@C4CLpyXet!x)qWAJ=IyA) z8XY5p*{a*>P=nmlVbvpYW~OVr1b3Xm?b+X7tuJrN7G>hkX0zbV@bJG>7j_*jc@)s! z{XW#>r@I1Bibq6F{qbNGJ16WuLisWX%o15K*;;BcO6Jr|+%59V$_&$=Qs57&Qy>=VW;2o354r?%NOYfkdbg-kowwt9^42lwnd2>mK((%Nh$ewM%lgnr zhsSu4ogwUt-G6Eruxp)mKftU6brElO$2$)>qQPX9?>1ByA{PAXa!0?~N`sa;D1SNPSKw?Qw3P z8|WSs%Dl?#85Y?S7UXcM$^v*qO8UREBkI6=VWb^TEKi%cCH=2~+OAlW2$aki*gH5m z0l0z1A3I!;p9MmEj?DP`mDgsb#7hV^w{!h?fG3D=$b!+736>2Mj{s{YY0H16CBX;G zJO$rLIq?Z59Bk1kA5@c-UdrF%hEaX=`{sO?AxRFA7QoIvmkXio1o_C(?erruE&0{z zr$@#o^I^(Md#8f2-wTI!D1j3I8vrFEzk?s5KwY}4MJ&&NFGsM#JghhjD$an|_3aAK zSeFW#g?}z3oAI2bs(c2LeO#`Nr!fXrl{qKmfK&O@1MqJ6d}opSgEo?aE&DA{{Fbi zH~+n!_n&mEI^A1aK_`IO1v=y6hZ?8e7u?^#*%Xu~f6Tc)Dh44{65Jrx3)KF4fn2VS zW41{oKJ<99$Gt95fU#!4Ao)gxi!qN6{+L>?0}xm*nvONr;omYnr|h-Yfp)^ApM6sH zssbjDtP$^FH)mj&W+e`3YXP|tzn=l13?lYlg3>WBabTGqI6hbZDw)^cR=fs8$kuF9 zu+`AoHB3<1`9@BTGCYst($b`3ZU3N;sSi^<{5=-b#ahO$P~@E9fm3{X2MePB z*-f|c*^LZ1&Bs-jtW%?;Fi_Uw87#x5$-sUALp_7efrJCWfD<{<$MfZ`+|DmA8y#SD$QlIlMW=Q6_s!S73>Z7*~4j6Ko zv}$S>4%orEx)2QJ6*xh8H*1E6o!6a*2p}Ajt1aDC zz~l!_Sr-G+YSvz8JnN_oBZCGD{Q$@>aIOM6QP(pZh5O^F9h4)!zYc-ySL6fuvRPm5 zSOP=|DC19A*f00I&*305oB%_@0%S{sfye+_b$_~T=C=d*^)s;WJr@GdyBH8-o76Ge z3KAFaF*1P5AEuYLFvuGSRG8_j8>jvPo*Sc|@4P{G_5lvFVd@iki^T%ibM)81$IGfq zE0!L+x=-8z6b~5p06>-!ijZD9rq$Z7LkGYR@C_4k_JTZDT>3!{lh-n~k=xL-ffwz( zNdgDnz*jIERNC4Bh8p$55Ms>arWY?pQIew-%qadNW~*>3hiK-(*Ef&4{Whjg%>_8U zzl`zIX>|bCLBlF=0I!_^rTJygtwWIcFV(Yjy}Jbb5ZV0<`K86+b-`u@7-7~8j)|-g z!{76_4Qc-y(U`lt-$$SB^#NseqS%d9@XO&aEZj9`QDt5VK1i(8bu$>k1xLpy_kjRD z1-8s_VbOF#Qc9pusSxA`ne3`ApH0lo#TK2WY0YT8WR7>|o5;b-?y23^%MS_S`p>@4@$0Gs^A^pEtAcegRy$oCBRU;7QPxiR`5X#ib0c*nT> zp^|8$XGz4K_?%kIfwfv6{-<&m;mq{;;(xxTnIakkjLo0>?L|1X1yE&~q}rpV?XB6Yk``kiJdB zqo5vw!#301|MYvWF#bHYr6qfi|JUh`)_Q35zp5psA5E=AZ*`9ccLx6r4xcNOXKil& zD}%UiO^SgzN&~U>u7yE^u!j?{6|t*jfB!_meh2SgR@Wy@BCsQ4IiKw4JG_1SNO4e0 zEGK)nk;Gv>NnTr$a}NXv7&&*}nmyp6Q!sR0@;?PFA3XcUeGoW<8>8od6`%T4Ta%+H z)$@!D$w2VhV9R;c%5|16eopIr`uI?qGt8h8qYYTblvaSjj7Yr6{DnP1WJjB3@hp3m zr2m6rwH)4T9inYyM_{s36$>}h?V1n=*}=XK8df?;J>8NDA$;FwD71>U$?np*n99&! zicU2P4uz)@=m(J8liT;A?*`U${wj}JTr>twA!YeS;jDm!H2E`0#MXrF%gX4x4f3;` z<;=I+@;Q~#hZ8rro~3Jn4ohcI@xD^7KdrYncDP{>emKnZ{Df#-3==b}@kMTiY)TAf zAr-8L2d^qBO{XTMk*G?rqwfQ`OQUn*oY*L|vgQqv#GU^(ikX^!H*d^Vblb#!74N^m zllM6cCf-phjNU_jbsdH*UBiBLr_V)PcuVW2a3bq-U=ke850w`CT9tYKS1M|aLkzaY zy~^``ArE3(u}HKMn+eKNwqK1W}C!4 zGyf>^yVFL){VL%>peP-j=BOW#39I{Jl(r`)-FhoZia{MA_2$>!Y2Qnv|3#;6lE4+)Je~V9 z5*77bABakO)AneUn`w5bC44ud;1?%FkqYRMVv;s|wvbWFqF%a#SgZB+{9XQj(bgdh z*4IdMMo%XEX1@Q8gk)IvhYwvg{4?p4fzxxMLqa9+%N?>2Ns~5t+8=HuO+Qw7cMfew z{dzmnq)bLO@p-)%k*Z?N5Ig2hsTyBZOSt@j)|M;vW6j6`6}oiYOTk9+6^RvA3Q8X$$ypnh8G3(^awzroW9*PLyM>UaIZA>kYWGsP8#PK z?PlBLak0zV(RVaG>Mf?t?t*Ysrp;^?+{4!o1_}+!HU)x-L-C

zCSV%0oORbeZ&*2olO>y{hT0w06GIRu?cTd4bE(A8FSRf&wnubR#|695HA4w@R(3b+ z+WzrRWLwGDx`(wgvRm)S(|Y*VGoJh3(}?e z`wlrN7_ z+R}lxB_qtNPi&w)qSrOS-YaG?eL_9e9p}4nN8+EoFC2e{cXVF_hj_w8>h{g{Ma9kle*+4+e|Nq2NG#`TSS~d-?3(q zBdB;y6Q|<(k88Eb!#>J>%`%-sShCwJR(;e9Rj+0Z%70)moy7*^Mxt!IB7HE6L8cuE zI0y_ab3OC$Ptq7n-EIs0FpOxJX(3de^1ch6P3P7>gRg?zZBwIR*#&LleJOiw^K39h z(e+-B7Nb#G?iBL0D#_Hl(3_<2=yFC|(@13&a`@w|@Lj*;MGF2?(h$FMb@(SmcZZ+q zTMLDmslm+ssg1DAv>&{x#1nOnVQRc^lbPu{DX#pT>~M`HEJD1`%}KI)0Xs2sPR-&S zTRX*0RMEQywvNIStEgP*ekUX7C}9_h=C=i7WRCGU4RKLh7-RoBJZV z^JXSw_7d-N_3;bk4OCwJw#WZ4SyUV(NE&wOJ8`M6FMhElIw< zdy;xOF;~&(G%ZUjzU00Hd#H$#xZ5Diz;gWL_5xq-Q+|}YBn)V30n7x4d}f-!bGO;G z!bbgD1H7xOG1$;IG5d3>qYbEx_rUEw;A&&T{%YcMym`tysP&`x6I>65_vRgmKH68< zt6KVO!S1iF;IZhVjMI&O4dJ8+NX*di=t6FmfXe_(Dz}}RSBG$E9un!;*5 z`!;_yE2SNm=JZ!dIUyxD22hY4g;I>;1njR?EI|j|>T%{k`@=VPZWaHJz348pJ_UfTD9`+QU~`^}O|w{3Hhr#S)~LSH;R_;wi571Hu5) zbA;|briu+q`NBMHuXLG3MT=g{6%Uh}+>5e~bxqc&o^}D|V3=z9RPRT=R~zyPtlb#O z)(=b~`sw4j(y?Aj?X{o&QtotEJbL;ddVfa4YsgI3kHZ749~|5C0l zVb|WQ#A&Es0(#_q&%jg9nS}*Bvz-$GQq8G1&j6glj70z_PhX&B`0T#{UFyc9ae+i; zWoo0*VDb-y5N#2IizbpA6)>-U=?ryrljf3uRsp6>R{3dmz-h#Aka74~cnvG98ULi6_`0fzVTP*mFKc$om>+tm- zFeYHR25dFq{QSJ7YE|_U{O3VgSy?*^WvWpi?cov>m2*1*61>3fI>~x=4JXa43aE{j z_Hg?{v$%@C>XffwI{yw*n%Zswi{r!w|Iyp*XT%1;wAu+w2fZgJ!sC-WZ^{~dFETWV zj(r#m4GsSu`)tfmn4ar|G2eFnabI`Yz1o)n2r8$j{9&)O4k?#0apd5CeyC-yOWbd) z_9q1%hSRhvtZSiUO)y{pKglhP5v+gO^ zfrlbz*QcF1n^DAbtAP5(_%h!5ZyUN?pKsVa*Ulf~FiFU5^#1){rX42u4l~F3v~yfm zI~9zTtXJx{F#-09tSpAlVs1swCl)U5?>yvX!=?@L$7~11<<@HI+uKRa+5+r2A{az! znT`*Q3Q^J)`lAMxo}MDymKlHC1iy`nU-0CE+2CNhAZBT;Z{y4{5$iQ z8gAOP5K2b5?CF#Ct5mDF_h@1on-`blx6!dAFa zYpdbe&sfJms!cyeJO{ydHJ>ET+(v7jI9p{oe(v)>aiH3PQ};cE!{J4!i5XS3Y;ge4 zRi*7$*IMxlgZ$p6hcgAxfO9=e+~EQ~6&xfH(G}5n-$&|z_Ejr3PhK#Cu*3OGU488R zTsOqh%s?}JNa*{l=NdK$1T4RCb4{PkC+`Ok27>I6!x2DQ&G7S;2kXV|5^YM zyK^1FDY_giwol%g8*c&)>_IAu+nUZ}x;dsP>=8?}bK@f@R{g}g!umG_b{##&waRj| zOhOK_JE=zQ!sIm0FXO#aN%+;6O9NpwOt}kZ!?*Ud%6}5+(u_}%)YURse$``-beGa;Q>7H3IBFy@Rg}jxw*NKAiIRM~%Q9nBf z=~)@w$=H$I_A4aq>QBuwI0%E%!M&$yr~ZT)O)Uuwd^mQJmEdCJ>k)XVaV@}vgsL?N zt7&geYIU-?)FJ4}rbbw&`JnJBjU`dfC0+6gk$J_#Eb``1>#cWVycAQ~8o#u?jqdLV z@VI&J_|!PYjr8t?u|8GH+cnGD9b2y-(?JOyB01}-{OZITY&@j9{OGBsgzwaleyF@% zv6C09FO{_{tOQZ41; zHwCf$q?LzvzggGoi$4h+#xpl~r<@Wdq*5PWIA$yKbK<^Xv7RRhi_!F(ECL;2bK93K zDT-?P`17;{j9(E1Yr{-ZxL>O-01QR|3r#yVKlU%sN5dhJ{SC!Hwedwcb{60NBt+m# zrCH|jX7Ll*09`KfGNY3ykFX8FT_FkjJtB2{13F)OWQQ#-S%Ocyq_E?lw&`I=%Rmdk zf?B_0$s?uo;0nOn96zZF2osKk3~d{?v>__onY-?CJ#f_#klJJKI4b!o9^O0RIRufN zogq!B8=q5-<$OR2!I4X6Opa?L&ZlD6ba? z5CMKRJcbWv{&}H^Ru%6CR)3a`09ftVg{VSVPS47#SHJIVu~aw0w#vTMt&(8*wN$!G zGFxmQ-LK0uc%Vr8`B+#`D74~K&a(J>!?slKN#%b9W z8BMhmI-@Lxq#p)G%ZFVl@(0a)ypnu}#Ne#Mzw8Y^?y|-wId0+Emr5 zw24QyWscz#{6tCr+W-7fMYsin?e5f2yJiw`XfQwx$mHo3RH-|o!`yDa`!!cRc&@~8 z3vka0^CPA8EepEuFEp=dv8h?aj*iC_mb%QHQ3f;6llmpB#SZTJ9&*OSvOd+4@Zu$< z*fV{ZUTbM$G<~NiwaA+6+Ez%W-9$d4MfbqmWXD%9R@)BM;WuYdywQ`==s%b`{Jh6c zJ6h>}EljtNJxHAsr5{dV=p;3bZvmTWtvalK9-!&`T zt$R&*0cXRF2U#0vN1X`&K9h-(I=)G{%-^T5Aoo6>2ITsL8`^KlyF3ODCuF(%pLAs3 zN96w3m{z@pT26c{>1=u37I=3UkRs%d3k)}+ugy5d{G z={XCJ@`!h!qB+frYlp=$^b;EG+wQBx!Ta4oXd$_jE^_2D<9m;v>0y|04 z+x@yW*uBw@N9v;qNtIpR6)B93zA?ULsUI?l;zjE0@QgS3aNVn}FWqvE*gP=lI0`z@ zs4NqjA~Y>t&#(A*LIHI&&0O~On6Gtz_e(uRE?v_!6}xiijp1eeyq0$RIPZPf-+C2; zjp5iQhunBHTOQB3>724Q5`=7IVYwuzzkcuqHtd|CSjt z^PyR`-lWG#OWz52zdL$Z?|3`p%cU_^O!F)ybX;JkaeQ~%_yhG!X7Kug{jVk7{Oyh( zDAfqf^Q}k9v09+Z-Y!e6zHTJ*F?skWi(-4^2s>R&ff`ElfW?deA$~FUf->Eot6^Av z->dzVsdbp$=;rajJ~OCz6?|@Jj_hWqp8vE;JSMI)S;(T?=f0=i`||@8ai3-R7P_DMHitIFlE zr-f}E_#nPSu`2eiiP`xmtz*CUAAB6^-c42(5u8*`7Stvk7K!Ud$(@2s% zVl3tDRH<`lB0_AC*-qE|r+uhy(;$JBCi-Q1yIFu8A3|Y5sQyW#0V&85QA&Soj~z%0 z5~sJ0FZ7n5zCCB6HS8`MfjTdV`#>xA!}BKXLHMnE+(*+aEm^`A{i-Z8_i<^DNsMusEY0Fdg6 zMfUmU)2kHPsYc@sfPnzCdD3yY&KFDjBhQ4AE|BKKe}T-@?SJL8-aKU-#6tn{u4a^R|+-I?fwsvo796eIT64lNT-YcanAzvtfYNEmcau18@Z+vE>JUj>p!Idh#RVzw>oAVBu}*^ct$@ z@jzwzfA!?B3q}2wt(?~BUzjxPetF3R2G1-t(V%xLVbs4dNCX%;v1r5&31`I3?_2?i zHAT+(pp6oNE8+{ue--PX773HC)*u`l`^vOI_|@PC^N;gRt)$b&o(zEM4VukiD}ri}#ul#Gkw6dlZ?*a4+@1}M-7ByTGZ-P7B$^+Gl@F?$JyYrAg)p&$`0 z7N#lyjW($KBeV>sFtglfj1rA;0g-UOF=1qR9Ed-3y4~pU+jB8lvEYjXg!M>_3p_Uo zQB@73App*-U=p?g2OQ-0^}9P65yr$1!sC|9%Bv7K+=w3bRFJ<0Mlp16+n#REP^+|? z^cAL>M2NJlkOHDeU%mYQx_j?{a&Zf=Xd}0{r+*^f8IULA35cmb6wZD-tX7z`Ff6+gqc5= zmCq3*)89wNkC6Dke=X#k&cwTZp9P6N_-2F0Lfi0{fpvB>9528ZU&ehTAObA)CPgTw zP61vnX>8+K)q+;hcLMP?o(rWP^qH#l8_v#xln%MSeu6EJsLiQcf&MqOqdO_|Nd!sr z9NJcoIeVKhfVY%hXM}w3eCjS@v`nbU1t#;H{X;I&RrpofL>S8%r_@kZDYJ{{mOT>l*PmIh{x*~!Yh3BvrpwZ}(x zr1#nTPV!%>K%+?~^v5RzCoru6#TEo1$7FrL==}Q=Bywlqbb>6yY?f&EbV|Ra z88|2dN|bii@7abeEiK_C0(p$Qq$mUGR7*l887zTEFngO`&lZ+SVAR`*0!#$j56ET+ zsAGL{Q_2{1Z$q%WnuD?nT-gpkMbFvD`*~=z$R&z}zvDa^Bh+i9UrMGbMI;L(>k`xE z=Nog6xXsBGQS%+ADBi^z4e#cndfvlpvq*W;DTTAN*o79U_1TDQ>>t{l37uE#VPB$; zL635k*lERVI`_qyU!OF5_iq~-EiRXKlXhN}Fs7QrV%2B97CS>5{`#YpohA0NNymGX)isZsOr&F4mE`cw-t$+L|M0!yt>zH)zn~Pp)P3Jb?-i@K zX^hPH=I4H!&x4EVID#SO{vqD$%v%K43xhB8L&zbM9ywz}gXb|9eg?{bwc&iX;phJL z@Ij=sf8$Rx?H(7Dl4jSv7Q1Ft`#-UVD%U^e;U5w)K&93WG!Yl7gif32;vAAXELI;AdB~W3 z`Cxu8XO%u6!Li0mILy9exo^Auz)vrtD7T?AT9zv^_}z4lP%!?nUG~fEwcjd5b9O_7G2lXScK9H|6YhYFJ)h; z$B8^#bL^?Yg&R1DM;5W+MFG5U`IhC`e#$;UW0z(CW8l^jG@XW&4XBrZhGS91R-o8^ zWXa(i9@Bzpt7p)W&IwgXAb#`FM=kp3a(9K~eUon}?;F!Coo$Z`Jubs0PFGCLPJKJ# zLM>@lbh`hWqyj)xZ=4M;dMZl< zDGR7+f&nXDzqPAoQ2e{6C0B>|{d4QDhchjahq#nW*)(4_>;wd5kTvm`qN6fVV(Y}j ziFgk%w&_`sENScT8nGX|+Ibjk-&n|-`J%c1D*Xaw+btHFZaY!lBGM*SuDIy3+~V+Z zY?_b;#s$@u{)@X!xMOy7e>gIQ&0HO~*OPRi0HmdpV$4rUHycm1MRD%t-mhwkLp&AO z^^MxtIa_2;sm0wuTcf-f zC(pJuV@->f-@9rV6T-aNW~(c>TVIB^aPD%dB6x1~m`P+R1k_1hNay2jR_+yl+x5 zVCN6n=Kbz9ysh2ZUYok1XM8XpM4PHvb({4by#nr0)*{Vj(QAY>>P206#4IjtmYOu; znPFbC=lcxz@VsB4>3oNQpPo3he(%SGaGNVug|2t!wLZxYeI~8VSkP=QCbMt$EDe#M z)56A38$KTDyZ&<9Y#-e=3*3?z`I%e5Fe%Ah^q?6z5F#g57%i`*_f3kq{w<&B93^Ey zP@Yp9Bh|> zVwikye+qe)ieGLVA2^}&&7k?slDka`A}}YH;%R3{Gc7Y;FK5Pj^5sU`YwxZ;cc?X? zeZhA(3-wHF=`eKn6P7N5o=5yhMQ!Vzmkdu^=%{p-$?&>P(85JnssBB`e~^be{|p)7 z@kAa@?Z3aBc`0)5e|cJ)`Wy{ip-^3B+m7fl?A!lrSf&t1W=IDo*VNMflY58KHLwLv zrh5Hl+_&%1VneNX?n_HZNDvCV=Yac%7YwR}?4luL_J`x)fTM{R%jJ-d`6%V6bi*Qg z>=-k~ZaWaHZ3u!hw^3L67@-FY!>$ui4-&HorNe4>gby9?G5v6}9BS+4dTo>qX9%V! zsT-fT!dh-pu1TG2eLcxQa5I5-uE!?}QZ0XGzu*)hu2XOS95F|S|0RS#kNb8GM!nsvQy|Mw=MhNCOPT-?!G5;`}AIF{f zuxN1gYqJy$!QBM5wZx8@JDByQL@lfpm{>Bfz#RtvWU5U=IgB z)vgLp`b7Y!3GOKb>E8TxVQ_5#x8^_WN59z#^AK2;+MB+vK~N<4n*KB7A%x9#FCk%V zxs~EHeT!fnEwAtXJ7{N$OXf|6QUvVS>DqZ05Gl}697;G;tv2DGJjybE?pkyVvz11u z1Oo{+u*y+ymkbSGmY!-jldYce&$5Z|UKh3Jb?FocKE<({NdR}7yf}MspM!(%1V%HX zT(LmAVfk=u`F99!;m>fKI&GYCM#%^NaU8DR)@XmYMAom?T57kVRY89cs_eX?8YaJ+ zb}C^4Tnh_Nh$aZM%dw5D+p&Ll=2aZEQM*sb2#ztU^uDX~UxA%kTv>U#E+%aE`E)j1 zS3B(Thr$7-{t~dj6TykJ|{qFWm@~Khr~~b-J8zQYh+|)7Hoj;_tJqi zb+uuUMp_21Ui^!bvSHj@77N;%x=PvN0iV9kP2=Z&$^}l zXL1D^OoFFz1BC7nPwh?T$_y;U>KML}IcSji(Vd|LrLm~&(~C+Vxm05FAYWrLjiUB3A9S+DgxlnSx+|bqW2$Q z0>f&Gv0bbiiC`eH-RZ)u^m0Y8oPjw8O5c04A8c%8S(>3l86!%aJ_h>dOM?He``m|D zKxfJlDBQ3J@}CD*@zxendr&J$QIv@q3eI1iiv?e2Q@x0$yiR^y2!|T$fFL3q+roVL zJ=GEor<3rCrPlT&tH{a}oLt&LP@|C1;&n>RyQZDp3{0@_bip7CKU9Ke=rreg$EiF6 z)_BWR!#3q-$ShvMQc8pInxDuW8XMzhOMH#C?EC}v&h1n#bPCW@Q#L&R8{X-YQQidQ z=ZA8}8z4{C>G%?P<8tWwuOZ$m`w;4=iq~`Wg)O+lW9@0o?4n zJVEebELc&xb!O+uh`~@O>dq4r2Kbkc{Wcyc@zr%i98~gfK49nsvl!<9_9Q_nsF2}k zBRQ1^YVug#>Hde^p5SnH;VEO;Blwbu`5cIWGPcQXK&T$G4&L1a5hu)*dv6UlddM+u z?0XX^in=Kj{wGOLrOHucK*0^HH*V+Ip5MqXE}Rr=O1NbSmxAnZ%M@XP1GdsIUFiZ_ z%R}V+DPz0_8b`PXVE95vH#XZWeZ5B9)WmI$e2xIPKuv4+6RRhL5SK1mKQJiv1{-D| z?f)4nF^#>lub0GeFxQ;{vnpYxOH>`bG8h39UtN1gc`@N3#doZ7iCGD^jF~A4!ZaNv zVh?Nt|yT zxN2JWRuQ9DkU#mc!B!@P2c}~5u-@H|TAhEQg@sK?SE)_-iMml@a*t_RGph8cEnN~? zSz|$l2kg%m8C{iad+V>gNytJ&{!@98hrCka^e6eBIfnf1g&Iw|nyq+G5MOp#35R4a zpDb{_=r`i3#`?VfgGQ6wpH00aS+|}Er&_rBx+58Uy)LCGMg|pAf!2aulfuGJDWL0K z+5>tGLn6IuXuY3z`z!lNh}44F8O*84Y4wFBH$D)MpfGz&sjp_G^kNeleam_U@1<5< z>XD}y5}pr7(-w#S-m8hKrBV1 z8)u#SnPSRGTeouuSQ#Tv;UN}-h3)->Xt`N4Er=I`!6|b|ma|d*pb&H(OJ&%!jZZnp z)+B+i;XT7D8lCX5nHsx4by%8BtsfS(0+Vx?{B2=rp?wN(;C$D787s-c+tcr}we!po z{O&b*lWLmht=7i$W@+pSW>SwU>D>{TZ`Z9QTpvySeKDsxve~#b@uJKk$o}oNIqQ(? z+-2qV-cuZ3H2K!6*6;ATWO4h7dn9Yj+_m-PSiB6xRk0#_VJ2PX4_EZaH+u$78&!_* z7H%ne{m$J7!n9#gT=~t$FSED~%@d9%VO!+q_7rp%)ur3#2n| zCgCd+HP!1FSSQ)Cdg2@0jR?nrHxz`1|}KtbU9;1k2|?Sjv=M`!Y=JR?8tLk zSJYe-hpZ7oJy@C`YpaxqdpajbtZl%>-${=&BF<#+X5^ymKLlNAICSqOQv$t-HFxBCFFz!(5l#7{7!s|`LphYs z!5HwScUg)s_+`)s+s|Th3)sn=k}1adq8|vnDM;(B+rlkx6USq)Tq!12S*VzLQ5%W$ z{Cx%EILNd-FM@GBG9-89q$HtwWg$M!x!JL=`9*gCg6Z$vl{p=gJ$l5zu(KgfweC+p zAQ!7jlwpdxe3$k0$Iqn&z3M$%C0GyDA){X+S1BJ5g$W4$o$3|qp3st$H#L$Fz)Kcq zn7mqi5DMF{bVzg-@Od+E}1M~nkHS7tt$*vR$To2 zK-SO~X*m}-ZOn*sBALwiEvk~hgFuDQk?mI58twY6-dXMwXzkT$()}&9^x}FfV)KQ` zY!IA#Ipct34#8f=Ys#8D>yItPSU%$H#24`#JgxdUzupvFH|<}JmVH&I(u_xvPKgWV zI3PktMRmt(4&I4{-^$j?L3uvWItt)OOKdK)cD#7P95I_p{DGGIsvSBy-Br4_*8Mow zYv{$7K-u5MfHFThWhLnU>;4WEsqdFg9tp*3H;29at%XvQsd7hihCB@WV!{r8@XJWt z&BPigmQfuk&c>&FUS@iF6vGvv)e<#(i#kiwHHBJYJJ&WrU_|+k8yjzfw8L!q&QnI0 zqa_zzcCA4yNy=C2WwqIml|Ad_TRbN5GXBQZh;w8bTS(Ry8@+C$J{d75*>puQSB^v- zKWiS?4QBIJe?0K0DWC|IU0k*PSa^W@l;g3AVbIqyXybb8(yl8|XEv`+I;~A5jIUUI z6tmAc_+xb^MulT}ME`4m<+#ZAKmQKRjx_!8UyfKtayImFrFin6?ZZ#km;StFWaU1R z;Tv*o=+1ri%2c!O#KbR}T&h%4iI34gqg(}*nMBPupCa=qdndoOXM-(1%4haC7e(Lu zR5<$QuSTEgvx2+%J&i(h#TQ4GOXWs~9(b#&JPP)TZTS?wn6K*PSec?`o9$6xz^>rY z5&G;~$ar4-%)v)hJZ7RYV)^!O<9OBR6aVcf#2!OI_}@;$_B9Mo{_{g|@c;63{!*@- z{$BygcC3eqm8g_*FP)Bo>xFPODZMH)lO63d#X5)y57=>g=O>gAo*EdtlxmIE>-FYX ze203BhM3nDA1F1Bs0rzEow^~aJm@`DM(e~skYe~#)XphIC&mN7qv<%K@NAGWnA(;* zA6lj9d)GoNQE6K-8XXAr$_-v6?|eOHZ$Q@TZ`Z6+|8k!U!&Pj`84suyHZr z_>t(!)70e(#pmaPFFDj6A`kkF4lmf0GyNIDm$hGzM!AZy(5xoUN^<3SE!HD}5p1uvCemtVZZRVX8S3@r2 z@%V|=2wzF*6@OY|Fd!UJ)3H77=~Lk$0Qxw$LZ+$8VEbvP6L+SKj?BsA1hu8|Cw1A(65yChWqo%y>vXCc4 zl~a?&Fd3CbH2vLKoh0NH2E^4KOb78^eAUST1dI8TGQD%bA8>D{1hgs z2#g^vF8dzuuR?P}A0fTRT_&0|a7ITu=mmV!z^ODM$+3p9+K|>d(Sq3j#-I=+TO3W$ z9uW~tXo2^HL`8}J>wo{c)k@?waC|n*)wB)MsrqvPTt;}o|1lwDx#uQGP**YqyifXW zDEYo5V6aS;9rQ3!bUgLfkfk|v#K}YEk%|f76K*`HeQ5;BESQ)TrpiU-9a63%`*=uQ>R8Z^N2sk~Abw9h%DCv?CYE-4=vmMmJI9 z_V4wVqQE~Su=2h_t4ip47g@&t2o_`f{is=$nBss@M~ArceCTx^)l}{yys_0BDg|3M zME>0kk=Y)ecR;#E|)RIh_~=vH%gZ7v6kAl4qllz;y^yl+-UcAZ!wzT~h% z7N4Ge{k|S@2my&HCF1K+x|dkOW+v`MapK%`MsIZilMe!YdyAA=1|7a6c^Br|Cc)VR zlz-j@CRtz+9Z*LV^=<7A?|){3kIYi3pl`qVZD*qHv^|b9=?zb2DRLMN0hSeiI-^}k zzlE}lP64QQ0BOF=Zp9Xz!Uq!FdU%%1qSyhj0$^1EKGXLbDmv{vkpZI~$2)xa_8jTu zZ?*zB#=E0q^{4akw;L4^vyXVTuMBE^=a(;WQrk-cSW)8RpK|jd?AcDs+1o9fy?0q3 zUms8C8));0%HlyHk&&D;CtqpoZd@1NHGgUH8d73ke!yJ-g~%0~*1mpicQ#{jyG+$c z^SEyWYA&JbRx2*$_Q*e7X05{b>bBZCk9PDB?EvFqs^>(mn{ic*KdM6c4B6PDXW^~c zISIjZWEM8iA>)1;F(StRm3>fPft=mSw3PTJ>%a?+rkVQhSuxDURW#8*RF_xFxrT#< z)o$S)-;u_oTBk%{ZI+g^hqi@ShcDFGkNZm+-|q#)NHC>3jeNL5xFhXXlT-4YIsSd! zqQP6u#^HsrT1I*>_wAwnTc{29WOwLyZ0&A82dEDz$`G;WCMk= zC3OtmD*&m1i|7{_*yHDBw;_?bo@Q5G(ZR zWW2*;JB=6(SnC0~3asZ?e93)<=_|z30=NPEWefD38N1!JoN_10A^#r#DYwCLxk!o` zmwWEjte;Otn+)0}&h$civr><#6E~^e8f-tGG*^a!$&f> z_ueZgjyNqX&%C?cseLx(`7%!BGOY#ORS#Eo<94 zAI>3oc-#J`9}-em|WJc{AL`RJ5Bc;nq{P+M>= z25rE`HUfR?mH3%q4l%dk2F@3$*QlA91ein$*2ABjeK`0=%4Xp@*p+hTYPV&RHafi( z7IE(I9-4+(IkWYqL34N;HhWavV3)3y2O}_|o7ZzIRcW+uSTZxg40`tV2kCv*1!_c( zw0zsIL`qC)&99ERq*^FPh*Q6%pvY4>6VLSmb6%PCbJUxqsg+;37^OoZ)nOOo-aaW$ z>M5**D>Z`~Mqlv8f^Omr;@w3#pxwFfg?z;4n8y}Ekvg3bbIeNI>NjX@Kg?#Nj;t@6 z>s;AW9{!#y=KW6X#sU)-pCO#)twvoOTmh*qEBQUx-NcrjFt_Q&5&!*IH<}k zex#;JGa^>Oy)(M1crB8xl6biOER1$IZyOZT`Q&VIlo@gc4|Of6|?LiO~Y!p4t;XAK8kZ0 z9f4%+#%Lf=o)#k{3;spb=Om<3Wt2Vp)JEPHS0keSmNrz-kWBs@FVPK(%LdhRC%9 z&!*CqBWoMQ+WB9V#rx5mE#vbEp9+dd^pYk*4@q1zRf;SdYq<6~GT$2L-Sebq9404G znO1z4LLM=492fMgxzcWTgcqqKM>=<_v~bd|vbQ9#T4+ZsR*@mT{X_f(rknd?DmN^c z&PB$x{Y6ae8;kZ0f)Wlu?6MmTov1v>69l z3o(fC74u;n1aeU-VF>ow=s`+=>EbF!S!PAX}zSr<+--(g?h!=_6iIG81aTEcHJ>44nC&pN`Bl5cz3XkiM{lD!fh-A0`8ng71Pq2q|=S z%0Dk}iR>%%+c2-cG@sOXrhl~1m!m1>z2}0tmsHI%P{w4aHv2HgO0wF z0_M^yjmac(SQKtJ{aglR1eonpK&>!um-$lpJt8S6uEKdf8wR@*wB=&^6J2_Q2}nJw+b@5^qT5-` zHRh(3qx?`-4Kxj@fA4}chzG1{-veo-UlXVe=sf+q1966D$i5zKjSwKiCFEH$Vjx3! zR;O-P=)Rd`m`!nG`%fx2Occ-?6{COeF4W-cpnLxF{;7jpwPELb9RM8_Kq#78V+or- zA$>s-#u+S6*`nIau}MI`fF$nKuLo;wkRkE+FNo#ao>bTc4On<|u!B+F4%V$E5H?-{ zrJz6QT?fe~Hzf%OzKrS+2R{($U>*LzzQS0#eBfU);qT}1H%_*0O?Fd_$pp1{)~5K=P6mqDh|FcNGv;=9wp{>Rd1#!`@l z2EDbAN6OFRy@l5$m^HjMMc1>LPX-7{0H9m5*I)Dp6x|69du#J|7jb(@3iX&Rnd5aC zK(Bv>>$EucKmQVcLC&`KCcx8xv$TIe5iVi4TNNNjweHHnd8_;ZhuSL~KW__m593uR z8Ll7_`0l=oZH>cewB|1z5K3MrJB*a2_su2Fn@JZ+&=X9X07CABjqilpH*iGI!~Iq24LB|cnT) znrnMPFy;s)B+BoAjUg00ka9=JalG@SaL+Vp!XXi6YWL5fN7gP&WYWUv+!<-jxdY)V*p}9Y^u~#N%KPCBsJ(CP|zbbEqxiM$!woEXCI1?hP#pAG_mra)pkb54@XsvtOoT5dmKi=YLUkqfbX)_;B z&seG?R9i_O?{-dRB{{U(2?VC;#&1@5G?bPsn;iGsi4p`2lHQp-%UU{=ObR0{VAlYb zj~|I;ZKxCGMJpGY?ylog1_R#&$M#K^ZLQSD0#_I8+z0E&3$h&f83=fizs8uu`XS+htx!1fDm2qAoV}eislG* zP1(h!7NJHZCpt z)B(4eA?s6xUx~GM;aa(#l3)KB5vIm-JTSa(@NPe60%@sg>EBw#ZM9=>)oz#V7sJRk z7MOA;C8gnhRD3oEkg2R~FsZa8>zvKoVgs)b+NiaUG;;0-lO`ir>7r$Ie-`h$M1xa`2$k>aJtm1d6g|&iX(<8C8~u*dRtN0(R_l-f@$(tV-8!5c=81d@o*Q| z_|1Q>A|=7335(S8j9l96?GNt}9`}y*Ovu_T>w8vPtP|5p5oV?HIa;rhB0NS2n>--y z?32Nr_IIWSK`vT$L+;GmmPS|BksE>JGb66jA5uO^neCfYy2VDUSxBu)4(@&eJ{e-k zrW}>aJn=;UK=`CT_q^J2%+RHAwzDTULWSSSK$-*>BAoD|2$f?FbwG;|Mz*I?I0XNt3V$&7~Yqmc<6wlmUx-ZAU3DT+7p{ z(_@_2aF0klEL%PK<#~E>*Ui}k(CZn*$UB$uX3?CGi7<0mO6uA&#qs+!gk~>4V}(H_ zmkZU!-)3hsUu3_jU?Q{Fh^URLH-1%rZPWHFF;`^75K6 zr|p#Fj7fwrS1S!Vn8ybBEn~wU4cIxN?tVYeF_5`|JT@Mz?!=r~5DD~Ce%MNN{knN- zK3X3s-I^|M6q9KXXD&E|*`yNO&QpmXRcW_nsT^;)j65t(RvjF)uxDZl0UvG7Sj*b<`tISkJJ(~GCl#iGr~~I$Iz{!7 zY?Que=8M#k1Jfx6v8x6bxM{`JS>EvQu5&)_&(p{vEmK;8@4g_5cBfRz#R0@>iNwe! z-n2fQg|H>nJ;Q@3saT;5G%N$-%dl=}=V?hE))jB2n4?)4X7zE!aKjQ)h!{a@7!A+n^(Sved=4NqL{CLdf;)K@eUpko%G)_uqEliiub zquGYpgh_^tcd@F?V5LU2@=HH}=0T&a!7nN=y7 zflif#dh63HJDJ`rA}}|d1z@F?KiN_^PvsF`J*sk~4dykF+ zK5K+N9bjPiGOCxfyXWU|$@y&YC`r>DPSuEM{y_VfrC&6cA$R^7!NsLf)Vd=$q+ZASCvf!M>SAdPR6GHQv)j6bdl8J51YB@MDAEauZ` zw~>jB8?(xf*e!T8Z~AUY>7-Q0mN}BQVUVBP;iy4c^8{pxVPvcP8doM z#b>HPgnUn*T{xPX`FYFLxCzEicXC4d5s0Qgg?HPG;(otAyu*IbIAw`V)@5p*D>f~1R1KnO z5OjXMlhCTVnV?k|D_3$=VGkfT69;ZrNqf5g5|OiFm4{NoT)w8ubI;Ki+M zCl*tH$(5gpo|y7Ey&ilB4)(I3AKkdOh@uS<`bsJk=vd%7&?#^^?Nx$shg(4Sq}Zaw z{-Mu|5#Q+rC2Hp!5@KZ4o|5`F)a@+IaL5mO&$1x#vUQ3%sPj-LbEnEsD_rOkr=_uz z|IxYS80uF+Ic0u>cioUNOwp8iMX*b`w*^x*PFp5=#T)1)656HSdyqX7GTU5&t9#`- zAH_v_B9jirY#?}rdT~ND6a7upzuwcAD0)os2{WevueckmX9qu;#lZk!kK146aJ@ zJZRgi5_Co!vW#4B5)3V5%u86SFRMl%*zB@Hw@gw6jqt)t>d(ilPx{Q=vq_PtXcJAO znnGOvYRkL-Sz#&~8RvQGEd{QPY(4w&kudm=IaKgVl#CbKQ^-ZCiI}nSe?0o!Yi$$2 zx%HteV_p3gWkCCTCl0uWqV?Hfq60OtUXnk1rYcb#l3SH#-&Op$%oJ3Dd|HTRbw70o zyl!?CzW)8Ft_d-DKhV(P>yCsBBDBbG{@(i&C(xfx?DUeZjW0D(#n1Vi?Yh5yq?e@h z75|u)DRb>?<{zsx#I-&~V@dAx^V2V$0SBL)bm(Yz)U`u$V$yTu12C%TT5n%j$wuqj3{ya_Qk=Fx@u4HnN9lF!MRz1FhbrxG@wsqp2 z6t8Yu6HzM)(5z?b{_wiw6;xlTaqWJVAw~Ux(t*`7AUvP|bmO~RP;IaS_;y+XeEWOHObqyXmFh$^Dr-=@%m1};Y8^??z-`| zv)?x()j)I8`LVMTX`_BXFuM1$@$QMNsyih4zEkO%23E%I4v%)Zf->4&i15Pr z@F36nVBtLJp>4iELhJi&W>&XVTOn~q?t=N}$4C54rN$9LH=mTSrVBlZpt1q=2G{ql zeAFN2-~Gd46LOKAw?Jk)3aw%L8)RjJtNvdgnEwl~`9D7`D(`rK zh=}-#@;$l7{~ILwKmX?cfB*M?X(9gwiV0`__rtYgVzMLEzsqNeX^7yDvV!`(!n?)+ F{|9loL~8&5 literal 197726 zcmeFZXEa>j|M#tvs7WMxNk|Z#B$$a7K}ZlHYP9Gr`iS0wM3jg@gi)h?5xqnkC3;On z9SqS8MmLz@-oC&8?_Sq~>sr^7`{BJ-tjJk%&YXSr+57!@z26@no@=VoUS_>aL_|cZ z_T;e+5fLfL#TOL?I5O4fRtNsN=qR2 zf$c~9Sm7ywy+Ln%AD77=Ed0kfH(Hz`l$+|`Khf6}&;I*8l`+>PGBQPmi$4}L@d^t6 zey7OL$?*Rl{{L`HKI}4$rQXxXf`{r{HgY6Wocm55@>2j4B2LlCOyugUIB~}alU1Yk zHIHcS_vQBCr4Pq_E6&sJ7nl_p>K;5-zo~YWS1nW}C%(aNFRL1Mem60y;XB;DLTqsH z;A+{44#h#l_K$17a(LBVu|1C8l+au}w#crI5T{UN;8we;d-u9sNj&eRP;Le0h-1|U z*PCgrSoR*6e*#DI1lq(r4=!2NpNac(m4=zl-@mW^ zm8LM?6d?`0HMRL5DLI*#NSm>hS@QFyLJDzl@lUsnwC-Cg9+LC5iQ;4Bdg|a@b*%}C zf5%gGE|=(0cTNzM70V^?y6v`xG~VKyja*}~dz4~vM-)-(-I>)Kn~McZs+)QvbUB@EZ6r0$ zs4*|$Sj%@7(uasIW2_%Pek{cE|Be)pU#8dF~|rCL>Tez_ijo1YB~B-*>)CGJqo8p@rj%}M_IUdQfk#h$%?%Hop% zmOp#2h8XqSQ0`S(bJeOb10$nH3}o-K3yQW3RohM!4=#lgjf{*;GzUOgVWnOsEge0${1b~}BJUXhGy`cQJp**%SNXVil<*p$6pwNb&S z1Ng~dqZ}}Dc|X#bEz_pOo~XZoMm1meeB=n_sqs5{jn_nHa^MtdlQ6VlC6$dj+JI;4d=aCR;DYwgP?W z<#61~D=Q`n&y&){-tuQCGR(nWXtw_R`OfiFyHca$YBk;SGL|M*OGG>2^k{*A4Uz{Z z{Sx$;ndBb3qPgmy(3V&d743Jna(gMbRaN<^`=6g%rD-ujPM*nYykLiTQn>`3qQs@7 zR<gFTZwtvM1{vqG8cTi0r%r!7sF#kOYIzMvjxUOHPP^nWW zZNh$gtNHZJtAK;iI&AYPkrigpUjKoly1*Zd*Q>zer8?Z-j|dLB_A(c??-Z6aUkC1d zN1plmZ+T?2gL&woUKrPu>9J>wJTQK{0!4oq(S#1j>{}myTksvxkecT@o9Eh|3!{gy z(1!kvcbN)r@;mx_aZ~9>jVm&f?8>%B6Xlk;((@BhJosEvV9{tBKBLGf_P77~8a$t5 zQ(u9zIr=lAYuBzFqb?g78MQGXamxP7k{NE2SFP7Bg+*$f@9*E6J*=sn|NK3K9`N-ueUbi&De$V2Uhg5`=2C5JUu5YZg^FtUMji2)wrkm@NmL%cxuqsuIauC zY`pE~PqG^~Zp3W*gpOiaXvj6=ZddwlHkF3ZHIO(E79@&rs&T5xHGz_Yy!OEs@vxxtCvEnKOdJ`4$Bg*!OoLCjj zZTbC4r3WPxBQtoQe^Rw2?>gS3s<_f5R)u6)Utw_a@|Vq``DTAS<(P^3^}JW)M>hti zo+&~C^O@rBNkvDxc7~1z54eb~nzLC36`Pcfo0$F3J4I1jZp@LGlQLw(O;%o!?U79! zY!rLF_lPMcR2*jWlNrN)Nc}UUc=I^u>y&k7v)qmO z{Myg;j1*eTQ-;voira(unf?Otw@=+kp05$tdA%=wXc3S#pV>-WahSz$O3`DM?V8PO z$q~<87Tb)?#7QX1P$@x`#J6ypZ9$3Z*5BAGs~N~9PrSGj65MJj)vYxjdfD9eRb{b| z&0G5*oSZc~GS99oV~KcY@!)AE##r%<9Ty?fWnxnSv1grYW}?%Hzx~*=_f!$udB&Fk zlh5uxGng*T(GG>A^1?hADMSgq-B<#5nwKsG*4CH($?AASn|T)d9v2}$aI^M`(_3zZ z@XSn|Le=WF&JX@j+#qEP%3rIaCgQEUjl9`hVmaJ>75A^Szif@W_ z_tBId6ln#PR&(_4Sr-eHHajhGjAmCok<$wNn)6uBtkWvYoUn^QE1@Q^Eb_87h;u^& zu3dL*-$}u*AOp4kk*)89BgshqN85@;JmLNy#Y<)R{(pbCJ7SBq$_Q^Zv*!}j72Q-Ib z!;@pnbUZ$}hdblI`ip&{Y?t@xvGzZHk#Pew-k1LwbpSff!mjom`!|2OP=8bWiKrtI z@8J`bo*TV{lBcSesfpV5X3fHrv_lI9#uIqerfWW_eXipOj0i#A&e541goM>F{XyYj zoqan_rISbPmtsFT)t;Bo&^{`7gHkVXq9yUMBUGg$X{fqC?%*6*rQ zcJ}oivBd-k@kD)eb^!(-rrnAyBFIHM%Y5fgbXM+%WHmZnS5Z}pe<=}TdgY$oy-p)X z-J+N8EZ*Bx${ap=>Thk+)jHU#zomHB*RbJu&B0kqet%@th$nX85hY*Hx$k^_57Li? zcnqr!Mg^UM^IJU%Js5YWPNO3Wk^Ru1SG44U_WbDXYk{$(2%EB zrp*vX;6e`7QJVQLPT0+tG#odlzS;{|bBvf)p`5FXgo(x!_vn3DZ!~Y_$+0Dn4~IqA z<6_@@<_*AjR4=0FyV=99{W?!yTN89Z8f{l`0R)^tIXwEI4)D%I05p zulf!^6ZcJuOe+#zri_nUULd2Mq!38LCdx8KM6D` z<&6^(9VR?0oolnTupm|!0OKgNYZE6v48_Wz^RtC@O~=i&61+EC_!z4#!s`aSf+x31@vI`o^qHpS-KkiOQG%vec`^_!)Iw@(`jPq`uyD(>(i%CFLdzX*33t6 zWjklmU+TF!{Eo$830iVL9Ldr!{OaFEYk7EXkn*|3&?6ui1^*j zTD=F>+(cG`Daf%mL+P@0>;Dv$xuBb;XBJ4Imso9xK*95`rpuxa=oZf3H7u&A99i-M zB|+2C>__?YJ^8M-w$O&5FHTK&V?;m2yoBbmk}VHdYnUB<~Z^%e+! zDgsDnAacN)RLf{#z~kj;@%l9h_H7q*JCv!B$o?y=NA&rI6&H})L8rArAF)B_r*8@= z!~SO55<#zd4%{D~n{q4-AFFq4f~7QuTfG1AL&nZobp=e_Z2m`#lev0fFyC&3_v{D3 z(#ETi^jxsmVKD?)j>1SO>1giLX2aWh_%fpVM3L-FuUtc@@pe4DQ*PB@{6+oeso(Y) z&;C7WHvxJ5-1+!j(fEJ0JFkS*^W4 zwtgitDi`hrn|z+APQH@nK{qI~k~V#9ff&<4PX^a9Fg5KIuATmyW{+&o26O58X~ZBF zeuIgrJm@^|;mK-_h=%2cQG@hg9yYPphH*k>bEbhQKg~VL!*gNc%oToA$|MxWMa6HO zKLyIkyu3U-3Srr`d!roGO%IQL3@I~Ca>X3;Xe)EvBYNdM<5jnmWc<8nH0@f;%2=go z#0N7v66xj(HL z1tz`u#GhMMRrQX|rSS@QWSmBjch)qaEd(sIJ4`d88#|=_kMLhsjZhmhH(eCwkA#rmCDmpToe=0_txLX+Ek#gsX%vw4&o%5^ubww zKdn?U*xB2xt9S0`EC3(SBWBe}ziVi^E{<{hgu~{Bv6DS`&Pw4g<(2Miu*h9{J$GYQ za-FEOd^2i2!M&i`q*Gd4eBP?mPB(+u(wT+Q9QUJN&K^O~D>f>D99wbPNlLNi78B^& zomTD#tr=+td?!Djd6sY>sg65*2kG}eOh_qt21xs#`*$KlR~%& zUHg@7w{Vp~DZ6`*HlE-yVG_R2N}q_^ZHKmqQ1`OEf-xT>34=~NQ=?L|f$1!E#D;_u zx<~S)*zxP=%FpqV%N?b0iQj4s&1LZQ2OOE&AtCn>w`xw+w>h>OY2fp<*Mm$MDl4M-22L1m_=kJ8#Dtx&H>6CHrgHIqlCFk^4r_zxg{C$3CF z{2x@T$gIT6@L$a>{`h}?*lGlbYgW8jhsWvtAlr8bc=ESrqq-BViNOKrgRLQZSByS8hWVy_PO!#OP_p|f` zlJg_XyW7RS2Y+hKM)8b?W>c;+cxIqIIl&ZnOdaY(t-oALh5T+%TD;&E;Wgn&P@fXZ%#0L2$(#lcG>DJg#CW~cKpxRvyDNYFqfb`W%ex-?IsERF_B(Q{mU2qBm%P6f zzW{o>0&Y({gUUY8fmv?!bR3huz|mw_u25U~wN{SopdF((xg$mufp9kMw92gjbC@0&ZWa-pJ@_Psug`zcP& z$#W;%r+C#g=XXqBc2HQoso!S>cgN9KGB3*A_=_)F2_rXpQ>o#92Arb{RWA% zOgbtd0PPMAf9Dfa8aK!va963s-0lPNy9%t|)41_UrQHT4!`p(cp~$3;sqA*c&j;1i@S9o_ab8o-P;-K0&{b*WT`zEi@q+CJ%F4acGcQ<;HU~y_PB2+nWaK@@%0-+DU z*a@bzOG9b~jtsDa3s8Luz*awpQsI_mnW5D_1g3VZy7gFJ46FRwEX{G zqYB;l;+J$k!#gA3^l*N+Pt-d?GHPS0H<0+J9v>CSLyX~&d(gAvZ&$M-i6*qA(OH5;seE!fAL=$L`pN=MVJlR5g@_ta zUromtO3t@L=Z7xm5kP43en7Bf7T;LidL8ayA$hjXeOc{h_VL zZJb^=X<>};+xk(r+bakjJXJ}x$D%!EkKa1154@QEfb+KMBTrubEGN)XV+xpzP_0?u zon4VXKdxKN{x}`{!Z3gpSQi(Xi`ooF@&-wO&dd&VWncwQ&<~le73N zZ5w=I?^x%F&vG}mgUM|A;2{!i_k}W-5wTBKfLNx6HDOu5q5$r3KW=CUUqpkZWhBIPcHAbzAiqSCn(@565SbJhFkt zIQvtIpQ0;eQwvgo3sSDLJ}X(Kr41D$6G>jL(YQ{jZS46O7RdGp@lG#)&+RDby?)7^ zWho@fh0|lxY4xM$A;dYIs1#$+JM zUIMJXlKYx+$5xGp)ukQjvVKHXsnq+_o-XxkcR+b#*0K;MKYRU)IDL*nu8txo-^P(= zUem7av?d47b@_rzo=ydf@i>1HpS;>VaJjDni2gL+^};t4DGwS7Xsc@^=_`Q+ZrE!P;?;-atY_m4e)uLEZwXOJEVxosU1aV~>Qz=vmDgw3F z`M(n@S|wFc6N3xyOT&u&a_f^Z&g8(4`{g;Of7SD(UX8^Xv0oG;CG+vWSpYVOl|n5- zB;o#gAyW4vq}5iHcwNNYcH2%LkM+xfSF8C)Jn#~1Jv=})7c>PFcx>?a#5kj=dnr#Z zk2Q!-Vmo{-o{fz!E&U_Z%Dp|dyF(*4j}VqE6buSA z%0lQNEa3`YbKJo{aQDIRj12#&#xV?5aUsrV5z|~iO1g1kwTxm$U8?xrn`znS9hqgq z^VG0!#o|?_$Qb580TVeUi>o=zi%CS^rBNwCqp5LlTy7mR0zqj ziyb_oTp+`AFhix4Q^p{u?$^5J$gMuzTpgCvNh0%p65>qbyB+AZTDQPIq|5t9i{U3Do|KadAyzOqNYbdV4oY@1JyT40K^|O_HfZnvxrdPeP90Tpg!FXISyEx;ryh7QZ?@sZEKjHKHNrln@_8`N0Ma|##Pf!M0u|o36=Wm!f7E7nHR$IK^>~5zE{)o_pZay+TM1_QCnt$_FwB2FXfpeA!BZ3@Rd}` zo~deJ35{Yy)6I?1jdXX^{#s?B_3}eqZ@s^F$?+Civ5e;%&$3zH`M@ATlL6XkN*TX6 z(vyItf31q==aoW&Yr(5{!Catzt|-S#ZYan8Gy;-B z+IQbB`Pye*HFSKTP34D_6c*qZX;gqCsWJUmiom}()=P?%+=}#) zvb{5zHlu}uLvO$#M`eY-gjm!9v)X5K0Pq0o9JCiMVw<8s-yvQ)sWpOE@ZnDmqn+i2 z2qv_FlQBIMiNT4Ja|dn7^i#Yoy?&*EvCp@v2PsTJWPuRQO>;G#-$G+J#G(ret@G^0 z5559sA>w@GS7y9ZEy@H-A+a#HLWoTmyrkMlAPPh8xxk#BNdk15ssZRJuojpTV$6Zi z7O_NVHT<$Ox3@1WauU2ii$DwtQ@>UnTl0*lc*w8uVq><64RBufSV}0qLKFDcb7!a6 z3*c>n0*XU6v#t@S;)x#~`GQ*2MjZcK(=Hm_26+QZ7;Ma|hK54mMUDd4H~aempGvRW zto1U4Ix!kyzMIF_Yg2@UJELlT%v}Q-FQDl_yi+6BLRTL+Ay=Z;H%ZkZ;+Q=hOClt{$$!eVewvA~ zt}YvT^7$v?@$8#gM9f`KT739$iBB}R+ACDdE=o9=yZDp;%A4p`C3@L+L4@(3j<&W( z`YTtIXB>G6n$I0$y10TG1sS-H6+}9Jt6p4jcd} z{XKy%jgt=_n^T{G`^i_<*tpTF(7wgCWqN`9CV&|p-4obOWTn}eneydJ%XR%%A@?-E zIJs_2h)cEUkCk6o;q6>!Wo#e5rl-Rc^$m}K{*MIAbvp6mpanA4r5YFl^E_O~?-K;` zioA}WV0>Q`>blN{GySy|^_VeY?IO zdolO>si>ExXjo;1wg=EWGE0nUfhTKXopqD>7CrqA+UoP&@jSRc`Eucg$!D7&R^g-C2H5dJu3nJqmYzq3DD5ubL|o(&vOqe;FRvOF z_9Rq{3UCz)3ffe03V=2{^W_c%E&>21IU&ainM~FQq{|{9wrtd4elO#^(WA|e#c&oY zUkk*g|6VHk=Z}?iOrFbRq2$onUN9{g2sfPV$)8_4LI<6@aXbgS^C|l-evVSUR_FCrZCV9Z2qY;1_E`sAdq>HPFYo|^ar0u3Tbp+!<9_W|1eI$(1K>-uQA zf2(m#!{s7}0I$f|HL{h@WA|i<4tXpy86>Ma^$)q2}@ZTn;#$$5cbqswERTz zvi@*W>9bYe6MLnP=~v#3O0bT-oVbcZSbVl>=)iF#x%}Lvh*5bcpRMIxG@O2AHgdEQI6fXDjH9>+yMo zr%hyG-?+=5ml6{zN7fWDPR>Q*lk~XWwARtK#Kie+$p)fy@PG~YCp3jX@vw|rwh-lsw`V>{f=t7z(|#-U`U@5jx)%|Ks35MjL!$9 zs~Z;HK+u<>Pb zi}11+uk+*22fu6y%LMRc>v#Ree5Hp@;{)*1eOZSYJ4nk}JI@r=&*!OV{kzv!X!Zis z4ZFfs8fb!4-ue7(*!AIJ^fZtr8E9Na3DI--KuQIBY)kZ%n5VdEiDwii#h$!q*dXf; zeGZ9lSNFI-d}(e=y!qOTWnLFXZAO`{4C1GT4WucW_Re8)Nil0FDHh!cNXbhhw~7jr193|?1>?DpXCU)~Utd}hlEj|Xp?%vi?REH_M7`cq3q zt)079CX$*raJvwGem;d7!E2>QVb+&P{Lwz2A9VOq5{n^b2;7;b(prT03O%tDr>Pa#;HzjtC2SIl=@Kjdw2 zFYWkr?6ZFJkko3k_VSUDZ6sOwi+zr0(?JrXR43YwJ9gZ*St9_)feP2;qkk-Ti)Ql&c_WuWN^#lN>a*Y4q8_U;<;0Ic{&8U< zccc`QFGPoP_vL3QcAd#Qf5Iyu?GO>@x0r*rAR$T8N9Z1QGWh)x$rT~04Ih?=TG>p^ zwA<%4DIe1}Mlp?`KMxakH zriEYYWf|fBTpknZ#iV=%EX-h)ClcIKJ~aIo9~za6d9~d6`wIb6YtmLUaB@yyH!*T?Ko4_Q)D{?85hi zcWYFM+J(9dUE+ZDg>t(_KKhk1->@q-JijVru6p!wiw{3b&xw1LA{n0t`w5}hED+hN zy}Ifz5t9hn1+t`?1tAU@BdmML?r=!dD6@A3U?MkX-qg-C0SRt3jy=An?8s&{{7a$} zQXUwr_fDs`8l-Z8zU+A%eju*Nu(@n=ILUblWG=%!f zAn?LM0T3X#6>AC1eSI?mWCf5r{#=ldAU3f5;A~W)_jHdf;&50dc8l+W$)gc0OI`=e z(4LTAD4Y0&qmqk}G1m~-o=V*{!&4OG)?P}KpbAbQqli)FsbW*BHuCa$*_*yw1ecX* zC`fCR(s;4bzcVFfLl)e>mK!!092{#TdJ$Scox#N^gaj|W<@W%bT*+(0Alb+fkXq&h znFXiRTOhIvrUUpv5aaMO#r$nA#_zATRLJG<-(^2GrB?hjKm`O%eo*^ABcqQ@R?gE{ zz}6|0$dnX!_;q*G?w-5%<+e`lkl-?lVd<03z|+Icy17tlvHSPKdCPekH|*BFl|*o_ znz{@Qx&oyP_1^Lvw=ec774OREWg~-mZh~!HU=V^4VjgLVF6>q_5(M&(=}p6C&G*5A zjFI{q;KP~jeI1_z0n9-|Iwh!snO^6G^=gLy%OnnL#q^?42a!@`a6^sjFH0YRC@92-rJ*rDX6$kiUM`}9FbJj`8)$(r1|YS6t~0ysv+>i(wTu14;~#=Ei_U-0^1l@< z^merVyys`pJP3*mLsH4t`7(kNtAS%nE!ueD;RA9EOdS(Vdr`O!NJij5()Aq>%F{ji zzX*=Xj!+W9hOImslbL_HpfSiXwm!QOqe9x?jioT}>+?@qN?el#f~3pQbdQzkVKHyh zs7{HV+xDE|MLH88J=m^<0|*v?%P}=?hlALL8!7Q7?6giU6kxMCAm1-)5kQ^5IY%9; z|2(+V=6ee~tea&@y-e=zE)7hiu z$l9h&=eHZ>T^HFGN{GC?8KfTRf|h#)(xwv)-lZOpk9pGp7s(ZRfV1EHl%P?n{g&0g z1LUaRHpqgI0S;%_Fqd-%Scl&LeQ-`Qmh~P6oltrF833Oy=}^*wXZf3 z>!jEWoHKxH5lJ5}ChUCoZlWFvI@<%w$CWN|5Pi9R*H;A4?(G)RI{a?m4)_4$F`ooq zo+;s~%RnUjUpfoe?uN<=C_hP~E+XW*ci+^|7))|^{Q1LZdO9`!H?h<|w=zUf&83>z zavS`ET;D97@K*Y(dIt}>Xu~t`%P`_S5JV9k#y}+LZH_6L_)7~?b8+*Z`E>Qar#4-rp@s(9K z3L=ji6RU~MkuKFn7eOZA#6;Yyu7oHY)7Qp`gXs*+t1Tmnmv?MISa)MLgFgzaEaaB; z9PW^HpMuK9I4p5g2JuFK+UsOBO-;@E{ovi|?V|9$enhR-{Ys>GHX?(O^)eIf7) zHs-Q7z3X7W7^3<@L%9t+g6>d-WgGZP5m{Co%;k)UzahK#zWAesR~{RAU!rJ&&diGm z`>zBWb72lrs6uXie)?qj?$>jKXk_4&;-}3BCI4d%ijBvsLZ=nlo$6a_;qp&Ota3^P zK_mu*tYq@rxN?O<2jxGJTXwP|?_HGoUdJ?{TU-l?xm|?;dDpsNGjiyc`$Bt9D2JIH zOMBU$wDijCb=s;9%-<}gmk$!+lRwJVN6=SBce0=RG34qkKdIpOVoR|PhSLr3z?7Gs39$V%zQr#=hKTyviJrlNiTcQ6eV3Zyqa{_*%e z$KMklHkgWu22SdQ>Wp!^J{`voiVFq|sf%}RX{r#8M%X*Iq@Tq>o~Ek!u&A(|_;%Nr z8!21wW7?6Tr9(~K`R#<=LdzTy40-p-_r%pJeM{r?D6h1v{^vCYah=8e%cV@})@S5< z%09xsBi$@Kt2{U|gMUkthsavprrHR(N1$o-Rczglxr-1pn_Pv*gmRD6ogl);CTrrG zqVJR*ADz+>p;c`?g%Ld;m~YFQv6+<0JzTttb}OlR#dq1dY-9~!ylf0)se;LA1(?u3} zztmn+V$c}x2q7+uRH7L_uWPKSqb)0VZ!u>GaiD3+5$~E@u^o@a*{ZJixnp6n1Wpqc z>tvi~48OxcpFY+1fIc*H|EFo2Au;5skeRk`(QV+!+`WEb%1|+noG)5gZhYwS)Rf@? zjFy-&j{hPh>*jXF$aduhbHbW0OJJ9>>|J$Ro=YTb3`-zhKQ`{oCD%UU8q;GF?!V&r zT&zht)}&!a*8@#Qd%1?~+zZ0tt?;t+gq2LuoKhR!OSh4oV@ifSI1q{2Aa?LP^ixbJ z*1c;|ul~cN=LaW-c+WbTb}r%PAgm%is+V%2QVm~G{{45A-?BUV&L)#}M95shr@R?T zDw2*G-)UySwU>9H@$n@8Hn+ukkf~g`EF2@iA;{`*wux>q#c@C}f4=m8Sg@U4ftGuI z!y$DJsbH-1W){c&+~)!*tOdkC9jYOWk%glUH$gB!FfU8@c%z() zNKDH!Jz9YNGokr@?54nK8Zq1T0uK=G2Ky$sfVuz(uHR#gnVwo?Y_Rto@8<(UcCL1D<$z>?;%^Ff>#b0-2>V*oyn7RkPyZ%yPV~Dl&OSsm+_2R z*d<>DjZta992Yrfv}yAjzXto-=_#5s3AR|%ohW^CA0HMYRJ<(GV_uSd$FeJE%0AzV z{9&Vu@X~8eLU7n7dd>4+4F+1$-a|W*Ist6t(C=dF*{nrxt$&Q|5ZeQ7f!if60o35M z39kv|rhC(G7h(ppj#Mp9mb`${M;{M*|Laljk4+s~U$pm$2JU=x15r{1%tkHAS%If0 z2>ZJ24MFbKidA!WcXvC@Z{lM$au=``T%&=#TB3locG{i#B2UZbXz@ij5<@-$ zCG3y5ZT(nz4c4yhS?xK{Vu%~7Z4OuE^dZx>3d#Z58S16zr6Ro|X>8-3>w01SvI~nb zm_VRz0Pgs91M;*%+IJ^ ztQ7Ki;4KO@ySz2?WR)35M-sCQG|^?~7=GoT#oOvjsThsh1(`NGTQd8U2XkQVCJ`S) zp0lGn2EVk11PvL*MCr_QjXOH4hLtw%s)Hg48LVUI+D-R0LzOu8lRI2iW3%9I8D#n* z2OG4*6cm!98kcQU0^&)w(*#)mi=xpokfq>GFlL>vxRMP5&2ymbt#pP;<7*x$W4=#^ zErV?R)eA_vAfV+FMNJmV0bN$P^bxdOb{Ws^-neMd1AP&ENjc%lL;og?lvkW~OewA? zZ%kEF2OTE|eb8^-P2BonRI(geVnME6v;+dY0%w5wGSffYCi;EPn6u?%O__o0?fQ$1 zsG`zlR=>jyMG$v=+Gj^e0eW`8#{h94kdeJDtnk2Heiiak-2yCuK6t#wt8n_~9A!0M=)PhXb$UjJUd zes=>^J0xRZ4}m)8=;LiY)yK!aKb|kJpk7gqbHjEotuX@gbf`ax>eiC_%?B!T5n* z2ISJVojEKeB_;mKtIPbf4-%s^*9sP|b+7-yz5?K+k44&j5|i>!-8u*HF4w!XZs%WTen8Uu=<|d%?<>vuaRT@MWR>$r+~^b;CgP_gK#y}n znn_8uao%$k=7@_NGXvS1dooeCJRyjzs`lw=c1_3{FViAuOrD&aZ02C`iid&n1GL{b zsa;XG?lYONp;72fU#cLFHIUGV{X_Ru3SbmoktaRL`2}*9HpP+?wgW}lS$RPRr}u(% zgbwVO4aJ>-Rcxs_I(Ywd7%(?p&?^$+g_Frtw2iDV=9R1x3?8+FXjb(hG_h#hb2^qe zBA8h2MMSLn3UdNsITpG1!LlLR{`O%YSgq{84(Mt}o=(=3M2XKJz2~_vV|?IA(_*e8 z5aX= z>Sjd-Phn~dCG?E92c~9`3YRU_eQ+}5`|a99sa(j$(fvkrDM}kG`dEm+@i(vYrKg=% z@^lnVDJwK`Ozs|ZEB4!%;inYd$+5Yd8|=rGfz8k2{_+3LXL+59 zNN}tOWnA{{s?uEI=-lJ#DyP|B!^rcod>~Z01Ly?4( z(bv(mjBAF^q$z&xjx0_ z=rfT!_zaZ%RXHs7GATQ9BvJ=9e(5{_k19o*mQ5=hnj%G_y2q2zFzcZbsX)}}X3r34 zuC3+V_vW-f>|$@drT12riRVbfKNR1rSoxCAO$Bt(--+}mTPxMkVpS-?WlXlvz^M7N zylXzPN7l_ufieZx6!Me^>1Lt;u&)ggT2pEht4OV16mHFw6_|xZfNadS;-j#kkjp=|cca~xBo=Fc$3}syV6ZUqOrNChbMn$H|1OK%At#`X=|H|ed(x>MW0Z_d zruR&KT3Bq^3muR<_T3-he>xPTDG&}4iri&`0W35U{Q907A-ntjGKZzKm9%SfC-x(5 zgP2EGe14d&d0-Sv8||EH>xCSjj*(_y0c6oDl0{2V`5&8~IwE8um0_LZq`4b8m6)E;n0wWH zEhDgKX!90HQ$Xs#U>*&b-JE@Qbq!zg@493eT4MxvcbaW1jChj&o7UOhhez#-tIWqL ziO!4p9S={&-)hMrj~#y^w*>k36yUuoRL60h8J*JSoR%A$)6F?^=#`+un57SY^3C{^ zt@~f-+U8JTM&-ld9EK@xDopSRu)eJ$i!M!E`mM-dJ7;=d@S8j~Lk3_&wNjwLY^(y>6h)_-^5y-1`K)^W#< z+sXGXMp9M?BqDAr1&EBak7nyG539z@il^3_!8xtNmdqQpYa(w$Q;BnnY4abxm@OP? z%9JBAH2MAg#@eZo*d1tO$I!Js<2xDC=ydCBJPm2mOfsM4Mqz7CHiqs0;p{D=qH6oT z;Q_2`&Asx01E!#PI61M=RzKybTTpKHUL zuI_8Dn6S4Jz`g0U(%~Fz#JkrPagGzML6w!+x<_lC;Kv?!*Z#xJy=jHqSJ>??+PFs? zV*jWRU{C>`7PWq9t2?!;`^$Zs`tDROH-1V0bNQKn$=j1k|BCWhu&xJr>b!7-iWni2 z3+u6$S#2gh{ZHYDREJ$oN~M8vPGa0d_NNqA2g8E+&${Mzp-)-GfY{QasIg*ydqPHK zaSAq}OVsAjIr4*}Q0|XdlaDez<2$m8+Cy%Es80+0N#H z2N$yIyw%q6A#d7($Ph{o@jJp`RMO?NU|=$(A}ibK+poknn}qdYg>j^ADuHF!^$CsK zSsmm@9jcH?eD2R|Ywyh&7vS+kOr(dMipda9g>jvGO;k3E5(pRHmtAYO)CoF!kFBJR zs2O~>wO)`mUzR8nwob$*Fvc^aeP=5L8=FWv(EuKdc2)7m{N`*}ZG6 zL^n_@R6s~Ggx0|^@U!-?5_fGn!r$+I<3Sum6mo_98@fEu2C@$x)zG z1gn32gU^Me(F!Yb%(dbzAx0YMe{2q&6dlH#b^ZnCWCD(U*O7ySkl+#>i?O_b0q%IT z>N#WDW|$Qdi7c3lNqO5czFU*r#j;(bZ$gSJ)Y)LzT30fRQm_pNfAG_@ou}>KONZ%C zXD8?YPp{u10%iav+t zflhrU+5#T!IcjV(B8mXFOSjhxeKkA=SV5$%R>fBVE|0E;TB!~;U$Zv-e zmePpd35W(fl3P)crjR);{Jpy4rQ1CE76B+=x#macblpgBkATFY9_j@t$F~~3E06Y@ zi6l|Yy#bJ|{it1{^7aiPNc^$#-?b8kPL_`y5&7*U#?N9{I-A7|3+{s@`}s^T<1rIt z1Xe8C66gHu6T$0VzfP{~@1OsCF8pHrViF!?>y}yl z*z|#V{hwDn52Ht0KJrNnj&n0MBFhR&QDz_#TS=Qg zgZj$!LAI&5H~f$g^2vO75rbamlBvgdi3bPaD)W? z)HWQ6ydp}}{B}cn!RX3<<0Y$k;mjo-|H*E++n|aE4%iSp4}IUby4654=aYIJDio57 zlz**9C@G{2sp=SY(F;J50Iy+ReA<kzSqt>>Jo!V(I1S$*w#=MZ5pa>Bn_m`~~-i9JC~SeWvaDv1Dde%AvuYYr2g? zSI@aG4kie#Xg8IEbI;r?jMT?x?DyWW*N0fQa-Vi~r8PTKeTMLorrR0a)aI$}0`yZ3 z;;}Z1UF7)~;?~PRglWZ|G$*2SBEo26WeV1rWZsWAr>ooQr*7%56g?LU!JEKTy=tV3 zo$VsRiE-62h6w)3Q2#C;9^Lr-T2uA&9ss7|DGJ@Cq==XzGMyO|} z(ef_0`;>BTX3x%f`^vVdk0;vENKQKlB(D7sxw^=ie&*+w@B*=^L_nQ|yNs!SHIYT{ z`F9Hqx?Ts(f$E~+Hdpc)BZv=|p{HZRr+whe|FNmxKTnQi4VW&G#ol4KSM>PKZ%NFn zqG`EpJQk%m^<34G|70tIcF@qA1Zbsyp42H3 zPgD?B3LV|@w|J3QOLeZDTyNJ_YN{30d?rqEoUP?2*-k&~efoSl$HARYtcm4+bdMbA z|M$%0|2F^pFRPG+hapP}N*O(|xv9JIV@F1hQE%h4km9)8eTJKeb8It%q%m^C%uBUv z`deU@jk~a>r16>Wfw6Su%C486_rclTm+8p>F>?~zrPmeGV$#)*uD#f2f3UhnLS$Y# znG(&yAyM3}5$W6%z0GcAFwy$Dn)b3=CW++gS&xeVX56u2zwYDwP9bAs*e3Nbw)$t6 zF3`NW<{Fi`^%NB!z-1BWWT#Q6e$^W_T%2KlzSKBwzlO2AQ>L$h)K&|wV);{*M|BIg zG*3y;Z`|MSX;X&2;=wATt;S=smlH^f>#8>GBKT5gyjyw2&PA_l=7g4$tLm=Wh1T&| zzx8O+&_8%ND?^>zP5Nj4xjL|N_wOn1GrpC7Mv+6P-O$2JRn%v5ePfp(fFssvv* zY+fGsa-FTcZm#|CeW`BiEY$25zMK?~Pg;lwBYWQmTvu_@W~zt#;OX!`a(&b}SMX~I zK5GSr^e1pu8ta>g1k8-Qy7$dcAzGkf;^6!5{VkLDG;Mbkg-X0sdj}4K!5cqJ;ukU} zX<516Y9|$(SMTPA0ekS-eN_9z8>is)RWZZ->9|3T^be#92iJq$Nxt+sh%C#FS*E?s zP(3WA3mKEKlsgwVd1I`q#!q7BmeN|cSI+20&G zQ1*Cx583LA&9B${vnpO`8?`UvFoz%zse5epePxo*le#KRqC7$CYM!j0eW!V;^LIpK zVlkp@zh97A5N(jiG7Xnq@p;9F-tx>cU#E36KNN#t!c1f zLPN<^d8nj$QD&ObZ_>;~fpj7cZYCT;myz4P5Tdk<;8U9x)OLTXxw7w~CT#m;c-PE2m>hF|hvm)3 z2W*208c)@IX?d-Eyk!(pdgS>_w( z7INLvh)=!sG-OQa-7&$>O?}Z@QaGBTBmbj_5uFS`6K_h<#L|MRTJq4qL54?fB=v{k zI#8QNY>P_;1_kc)|HN7WiQdU3*7I){3{aK&qc&T0K3+f%rp_2V%={j3ZLlfQ<@VudZ(E7Us-z@iN19)s8RPcWhsV9!ht6?jHJ>LpvV;wsuJuSJH@ z#4r>qB!kJ_h@L!CeQZi~fNJ%nb54VzJazY&WhH9Ue2@tL%5?$Qs z3iTyPrxxsftG&ys9nZN@4hr6bASOhv0R!#DuG|WV1Bg(r5m+&)KX$O*`%fvvR+U=| z!I1ErMVN$Z!O*jX-xN9zl2l`7j~M-m^;tM+JE1;gjQq%Sfb|+K8~(;ieB&Zp zI3H=BO>XL9y!d`!puyeKvPLIksH{K+Nu-^kH&6RcvdG8$;!JH!#6LYfJqURWkAHPj z&MP4|ch|<|MD=Lm{2mP~rTaIE*(3zgi+wCz8cv2xq4)xah72GSuo3?@u1mi!Di@&Ts z53P8$SHrvL(Erj5@7(V5#(`U1xNy*!wD;~hghzF$)~>KY{G#9YF>K8 zRPOR9TDV@2{Z}GqcJYf16FNzs(epfL0(mF8OpsI%RFIt9dU|qH0s&9Ze9AdBd#{Cj zD5FF<4(nT)fwMiJf>}=VQ|t*T1`Zb{R6?d?&pTz;+Ngy7Mh3SOw_$zY_hBG_%q@t% z2MRB5nuP)T-vY=3^r8}I@L(#I@rXLLr})+|ENZH{NApnh4z$h*p8noAb+8jl-U-=B zd+|Gy{Q-=8%xPc2AA^1N;o3is_nu#7@cAQC52Zcs1D&T5FyKysF~soL$8lSB^|#iO zh~^vAD?MN1bWL6(;Y^@7xvBp8V}vw@&d+DLp!Cx1a9Xor{JqWXlDqeWa4o%;rSk~- z?F38UOjz3|5PWapYm2bH;g$mKtv0FCIVtz-cDHIY|7A$7AR*Pw-!GabmKEVC2Fuua z@+~jPW-|eKbD|wfv33b_IVW|tl-5V~FQ1>DJWGKD08;Asj^#r1SMD;*suDvsMJ4D% z06}#3NZ!4{2$8|X0^c*Dw>ZQKDqYPDm2+#2nt{d8d7fO0g5JuINRC1d0dx09jDl^8 z8iqycfKKX--~B=r3~#Le?N-zZ3dmbzm?nxPbLH_kAv#NMaEUBg{j{)#w^I)$=UjRlrW!`-Y||sl$Gx!p66};r%V1wPa`DE7n~2m zT2K#Y*Ih;7QY~|DryLZiOL)k^KOx4?AB+?ZKKE#)&4|4qu9B^7sZe2h9IJfWBaM0n zjQ_p%Z*`KNtP`W;okYgFw4%YB1-(T@`X5rCD_X^q{P9c)B=FBP$3D}Pm$BUwCNiaY z1rI&WC{*@RD7GSfQF_|jvnHFXHExMH~24HcA=6h4KxCB~Gl?6K7ubUH8Z9aV$7 zE}zlVw-lGKME-MBoSZ&nQR}614N3M^dhk;RFSwQv%m{Tmc-0EnknlM&omnsot&Z@*()&-1hPS^k*eDL^PqV zoWld?1N;Vr{-M#Q*lDiKh_-*cA(kw$h0Z;CYwZTFdr|BcSn7iI<1dBXKPO*L5u2a$ zS59iVO?I4rzx4hl-GlarxxC832D~PvDVJ7nukOW3TO~YnCyQRrSy0Gp$sU@QIP&Gk z_ma$r^&U$I?_;Ou!*CH3KHXQrxL24ls~%Z3@abVVnN6oYpo~m=ovotc(}sVeck;K( z1s^r0S(ymrWHJ)cy5}Se4mrts!(D(FpWg+qI)NN#df>r;lYFh9H+(JMh0ubew_JS> z{ct6Y1f_6cO-H$Eq9RokOY0*ecH4J=l=Kc?g-B`eGcA; z9#tNq=WTas$J`a`h)O4ze;zH0_{T@4qA>!bn`jU!U4tM12Y?~5!W_PMe=g_UgVwj}uA#yKCP+y z;>&5M`QKOZZw^uX^Z$KOGRY}q{P#73pY@BF;=eDW01EoRp#Q!uz55`gk#$4fhfBu*Qr)pKl`xC zSaylBLbq2!lPx60%#v=Gu8Cv!dZJT)jCmxw5C=6o2;RaGkhovfNp`4u3xPlOH-zqu z3zmxdj&n5eulUq3d>7C_Bx`af*n15&an-ojO$NHvPzw{V^5~CjdO(b{!k}V=L`7q@ zCAFe!f$!(&8a_r;Em-t+*Le~`Nt>4+#^T}++m8O689Z3j!mz8bz8oo5?H9yh6g{6w zRC95@+ZdS}&5xR1-2sw*#*?*A{jho{xOnae1oZfnO2jBe{As;*q7jGh@0_r*dTPc)xKeSe&zZL(Reo=bll4w z`N^46@_s)y8EtH7#O)}-hYj)fjA^u`F1YRZjAhvyyPrc1czC6a-4~2dYi%LCuX3RG z!Bx`L!IPBGv@j;*=19b1LoZQ(SjS#Zh>)E}bW;Ikv?rTEvV0tA{KJ(e+8RMjRDBo5 zI8>+D!i5Fv9NS99;-YQ!if$Y0U%Ptmf4l$}@yqfqEA$36+bPD|DaNp)<*()m*)dfw zcrnC&&@PM-z^otDiTfm9uPnI}#S1rhO559|$9y>>W+?vR06Bp zOZVIUT|r(q}3>9sSb4?f%Z|0kTC?&8TR!!81-c+bh0<({7)x@{$ZP%^sC$C!M|82M^-AU8k zTX&zpyxhuMpNGL6@-KEoueWCZ_KjB^d;Ls+2K*|eZbJDV>?;h|6S@cjtCu`0{kFDL zB+R4<6ZsZ9Lzd1%?U1f`T7pK+6A4^iMbZa4xx%IQF!Eg1^iHF$P#ZUaaf7V+;1QOe zIj=lQKyR0cc&!@_O{Y=c4UeiuLqh}l$j4nO+#}fv&g)HUSy-MQO_~ogGC7EUtDU3v z1Szr4czlP{*Dk7(S!m%eqHsxOQig;jO{-5!`hTi;N$gMOGvsItlAGiAEGIN1XoU>w zD#3gxCfgY?#L{m>Wn7tsz*!oK?+73JXJ|8vD)0t=ffTxG)6KJ zgPFea>loRc0J|R_JH*+-my+7oO;(Z`AR{ICv3};|rahpL11Y3( z7QkAKr^JXyS}07x$N{2KHc0UKGu}OkG_VD)6XH-3{Pgyg(9eY11JHXp-*E+k&mSfo zT5;lrl;xiV#Idc5J;_L91RVQ+wT6)X+B;(Ixs_@Q=n)ySy145|v?x9dJsay)nPT?p zg3f(73iy)NChI7nL^wWY!bj@N zsXeaGP7p7#`RwOgo=QKdH(S%V55IT4B#@?h2$Z!&=z9?+iq4yM=cpy~iistQqbD<6 z2Q93RK4UqLbW?VBARq@RiS0eWEV6)Z0Goeq0k95r5HONq3TBUS0IIY}P#ePP)6 zARGDz5q6N&@VBv5@_zz0lB2E?$|`fzLmuegep5hJtK|~d?T^cUUr4p4Yu`trFw+OA z(=8y(tc%E)%0Iw`IYjd^rY*@Iw(>p%2@{She=alAlJ^$v4MKi|DVXx+afhJ$)1;7k1i3{yVxHjrHd^cK08h7ub9LnE$CB#W0xxG2CchcgHZ%Vrb7Ju<_mDW!%A@=y7lH3 zD8f7~uq8vFZD9=|ptBRd>0WM&^B5<(xv#IU?XNbS+9EALFc8RPelOw*MU$M!_SN9m1W1ff&M5s?c*+;{}@Z%TmzM*mMK&~JDUH3FAFw%?q;`u8x6VU+_C_$e3MJZGL*@u0>Cv`j#f zQvt|nMbypONr&Hd?}jaBvPV{DRh}MR1-)Ej78G~5Gla`Ck1%pmRx%?d#?x($GYXIc zCj%lZWhVE>uzn#()e&N-4XIQL*t`diT>x>d6H~i_NUHupgGuNTT1u{TN-rTh<0f7X z-s@Y}>hIpYYX(B8u;YyM&o?r^5`S5TJYl6JEKq9?jJSw`USVE=h8hxDi4VHRxWIs>j%jG&i1E>p$R+wKtbf&EP6yk5e8kvmzJA7BoM) z6SrnPZ-8uR(DFpr_?1C{rJITEz>K%a?&Df6K$!{qXsbKO3R^aB){tr}Syk<*l2Ik1VxpTYL827`G1{OlH(DODRiPnDoWG7XI zZA>U?D2wvgYz}X4`Z6j+&`}F(tn?b(BL3lh6^3X+l~}F!r1+MmxzihSP|N^NTDa8- z`yLAX9POEb0MeH-tmIpG&LDTu(Ngp!l=(^>oC+*v{%TXE&WDt2q?x9%I^EX;U$q}yx zDvWGV>kcXRE~`(8ctI(FbagXp52)`qE{0D?e|L z_Uy~BD4sp~J&EEVAiTADK51m$itt^4ZJ(a}sa3lohBZQsE$~o5f{K-LN(B>*w`S0P zTd=$FIl~8YlNhh)m$hFqG(}g7wl}2U384m?DS>UaXJ?*Vfnf=epJ{e&@;#ySq3Wtvs+Dk?)J3 zVp?3#;y_zfyzj|TljXpr1(n}@U7wrJGcKzmdAr;8a$>ve)>e%XJJdwdPG2On8dz9{S~EvWh3*nHh*=XPRe- z|1+1lHrk=w-*2GCPFp}U5*II+5W2KLWYS!B7npM^-o73L*a)hKV zO-9}MtB3~CXkuR@YJ{4Ka3SrO#G(sb@ovG4Nhm)|d#YatEi#~qIhU6jF*Xud`d!q& zB1SvUV*0yjGxLs_o*(x5pwbGN>Gt0;*lL{8SeK|K#x-`|o2HxX*WL9SjvJH~Ex0)65?O<^_Q393E)A10ookp^Wg-iuhHWC3R@i37FRio` z+a?4LYK?#4mBJHdWX55RNxRr4#NBB^-9(3J4~F&s3+FW*O|)IlUHtEhTl!6ew9Nm! zTFxtezrz0CS2gPY!v3qyLrfoI$D%X~uSQoryF8u^k+P4`>CHx(Jtch7LzPR#$=g4WD862EV^EM zvUTEEZ1|R}PO$^bx3FfCF$apW9A&pIuA+n=MYbzB$6d&DRqBsR-vR@(z_tj;92UoKLOoYEc1Z&ht$CHYXmeMj+8$NYe+bT-ZgbYiGKYIeP&Z|SXy1`T>riEr~ zW~@_9v1(>VdWYB2i^G4!tv{DG;DxMqInPR31t7Pt&9 z&$QvT5G-Yl5i!Zg@%_JD4}Z)APS=p=)AX18q5CayS3NAKpl7Z8u+eHw z(~Q+^IMqI7-|nE)YZ&9jggr&(h#tOd8B;-?%tnnje$#fn%x($G-{t#FlcL-EBUCK( zff5u=?NuBJCs@hlvSk>w9m6v}4IR2LfS1y+r59hQoVx8>7B9ShkEg*x-(tz^>VBjC zd%gHz&y2>%Yc`qnd%Sh$Q#3LhIMDHQarO^sj9B%!g_M%jFt9yIYV@v#aYxYU%fNfS zJG{R78XjpA9J<`uV#;-ct4Z`@HnMj9Sjw@Kjj$Wv3h!Be4Y;6V9rn0hoTz^z*;qZe z*-qiNr)4V6O#O*%Razt6eT<6AXR1}159Q*G>c)fn~?I4n8X7#fHjbw02ylOiVY#_*H$IAnSj-^4eXENM-O%K>oS zCBQM^^rh{79=~{69D|UHHs)NHlijYwjKQ$J+lHqD0bgD#3ui+dlf5Kyx50d?-4$N+ z{QZ2P<>zIExpbKwVZR763%(b|QVaBtuLY1KdTTX$E^nT$ohz0M-0X_=oXzo4Wy>R& z3$n?op!G~;(D%ax)Urn2j#}>U=1t9Cg&1P*Zw}d&zT&_Jc4vYJcnJGF-Ow!hJoHao z;P#{^`Z0#2Lq2NrAg1(HP-j%pjxE*lcAg`YHT=GuL@0(bNQQeu!Su!inH-Jl)yJ+= znIo$U4kizXKE={zKgOtiP!3jFOqxcEEc--v`0xld2v8~RHzxgqz4Q+%AyQb7`Ak)) zz)a2WDVeIqs*KgQjqVqG1@kYmjhGYceq`tGr?xx*QZ}>x*f)_IF9(M~_2GVWQw0C! z!Yq{9`zXe%hrwnCHXA+ism$Y^PznI$3bNhi))%<-V`Gxmk*_gPjhR(&{$}Gt^<=fj{4c*&%ek;US zTAhu42*syW5UuR7z`M2cG5~6kk<2hOMb&be9c16dbNDJu-b+l-rqOSO9uhO{U3C@x zwc}NJH{)bc=Ew@-3aUJxeY##%LVjZC35LT)cP_wB3G*G-ZTH+{jGYJr&s2r3G7zM{ zn_$nwL+{+M?jvmJR%h+-c==I^F;1M4tH?@E7si^y>)EOkYSW1!!6$ z8i*k?gq={75fvu#ZP|j6?TEPNvC)5fktA=1=@?M2<3c$hZkXqmojK%dHCMOAEyy zvQ^l4_)HvX0V-wh!%i!paI<3=JnN7)t`6qN(SYAD!GT5zFulfE#Bl}{)VxMPeJ+sn z;<8j}zacjS`w&o&onUftdJkA3`=ufyZmY!#_?RGX5y;#}k2_TH69^RCM=&nn#)&*Q zAMPwZSNq(dp2cTc0bozvB*UEn?ano_!{a3>M%bz2ys2p8g7et;6Pif}N+2~r;KQH6 z-%`&tCuk6W_kdK2GzrLtlIc(8jjX~1oV2Vp3~#L1k((qe?W(#8rbGB_X}-Jqs|jEF zx}(2ShF=0fs<_y6f#byPvPF}atUu%~5yoJ@(bFdu&plgP$ppXoU=hYn!?CuKiEsC@ zk=#VnFMB)+2Oae*9;LP6|HAZcXzDHlFtv>F|2z=JIe7L*^Gu@vSa=$+JJM-+ye4`)B zUe8fIJ?u~T6G-mU9Fd_#*x9U|LxgHNNN##f?!Vufw0y5GdIg9C%mSaF!w=$VA?!b7 z{C11bu{8t2l~oway>RhdF_J}~(G}&2Mq2Y8<^Cb^|3l;fIX^q|PUq;Xd~vKSfOx3i zIB(>+5J#XMNm?)*9C=^MAHGCn$2T)*s4#V9f76{Glpo0g;0;=ht>%e3$XO86YW?X> z5hGP6;Sb`GYUrcT;V%6hGad8e#k8*n)_W>AhD~E!I(awg#U$kH903*1%JFzzsh#A8 zUk-n`K&PyXzy@JD5J%@GQk#*jeE}#NCp$~ZEl~!zs1eU^fmpM|vr)L(77T!OTHx`g zhT>ct^`2#F&WDP&KmsE8zU`OZ2jf9H_K_86i<@u~FC(M_%7sy|U%Ijv_$MTqSdHry zEAhlx0zO&gWJ5xF^HB^s$;#M^y(3uj_@5gHx>G@y@Pwi}C`LmEDi_@%Wyp)D1Q)p^yfSLmXVRfIrI zJw+JMp26@-gjK*5fDk$yKfdms!-Y432MY%f-P?8E*e>)c&p-xtea8!tNqwfvzr$w| zYFM~6@eOuP8O4hOQGy-ge@~F4=ZFk$P0&zW+AhI!bdt6c)QxP?SLrZ$v7pCV%KrAH zHbZbB2m{fGa%r+S65}oxbiyKvIe5@R*c#i{+5|qkr&yfPQHn9{(GA+!x1(NRnY=8ttgCy~xFNBXdgz$eIEouco@d+6$ zD7&p_yiWqIhuP7dUzxGgD&HPs&iTTbLvAKeHDb4cS1tW$l|Pq80xO$40DVTmOXw!m zxOfw|cZ2%^`4KVv0F}se8%3KvH1$EUDqGT1EI;XWTFx)W*6Wt8@dN`Ouk0TAf&mwS z!n-eQGW5whajFV`LCFWgChdirkG#%s*sRlL?K1?r60ZBH25-FTrJp!gnJ=QT^t=&# zXu#75(X{$^ZM(P{KF6vV>1~KS3w!)w0v-m9lk-*?xC=YsAa)Ai6Ct9!YR7Sw&-big zb3R4`+1UwEvVHc8atAR3P?o6dyN}iA0>&eA1$CBKMJSg~_83nEEd}W-i{h4Q%WO0f) zgPS`f3qL*6QuPz;yV?IRm{E<(gzj)BTzcg;aRszy>T{M~Utdhl^|3?kC$%VDPR!dC zh*l44n_eX?jE;FINy_na^VF{LOrEa*Y>>Ask~K!YNU(}O{<^A zE$m!{QuKioZLBl%FRr5DoEB39ZTHT4oKXDq9+B>yS&;zS?S*@{QZqM_wYbI-v*MCj z2GD*XB1R^Z?Tb#rOonZ90i}Pj5p`7Eg%MYQ(`UNj`K)KofZT~X4C>6}DIe7dT{=aX zVs8hNO~1f?d8Dv4V@@BNxs?4PL|6H8;@FiHA~#RHgP$IMY7Cc|vkKWhtbt62N|4-0 zaGO*%8`EQOXTtTzEDtFxSNr~R`S zY;2x*Jd3A<(&tEv)YQ9NjP!BZ4qxA)=RLIq_Ji&(d~qqor=SLq@opij9?ykQw8>0{ zUdbIlmd1Yj{aKO%hF?F#Y3|69!j&R=XV%kFkkjUpl8^tND?z-fPoMFJ8x#;W52}>j z#3)B9rY5nFs4o!jvF|e3$IYKdUnT6<*8PObH1eBRQ^|c&zPyt{{Ix;KQjL> z)Q|Unq5k2A3IUwTfRu}9cH?V@E%W3&C3UqF8;ATq$mp|ZVeVF3ASwJS*8TKTF*mC0 zBmTxWY`$S3OZ2Y(@O1H!jqf6cSv+3nMop>w)_4|7p>e#vw{%-(ec2kK6%71BugF<; z$~4ESovOXDpbxM8+2=g_hS4B7&gc_)3{0~(r&v~#J8I)OU9*<0udY-NE^8GTb9Wx` zV_R2bhm-34V(XQS)#G3M__{gsMzJ;ijxmP;dErcS?1pgE-V$S%i1PbRTP&r8=loK+ zAlqv@wk%bgBF0INCc1n((3pS6wIvMchMr%i$WjLSh_j{P*siMUA@!$BA${PV;g(xe4m{~ zG*{TAaFRwp--&h{h0}7d9T;PHNta);^Dp{d4YZ?33Lj;K>Ix=w$~Zg zNmI3Xu4S?N$5e!)$Vt}N+!KUy+Njh^S>kMFD#}RssboEn0CLj}CLilQo|?M&e_%YzH>GAyBs5*+O$jnIi_A7Fp>jmQwF5fco9sk{O@c zA$N9IQFN)vrTo{Q#`b!4agbnlo5qoZz3G&fdebfXJI48R^|i%8whU%Ei{J2qD@F5+ zMBcB?V^r|c_ZlVHn6o^h&RVr8T~Wp{;!Xd4rVI4hs=?C>zVlAa+3f+k9GQ`8$0UYg z-Xbg`idfEsuecxbQfo7);9l<(UcG1;`ZHD;x@2`!S5AcuR|SLb@Vp&R-T*WhE^O<@ z#M)5QB&i>YA4-_^wcR$OGBAv$C$iy;2?)GLaD)1>rHr>q*@+qZLJ*Z_x>litAr*1| zVjPW;zn`86_67$*yxQ;r%lGm91!4DKTKxrcNqx%lBc;KXXE%2eo$*rqdmbDb4>L+F zbt({U1;j`c^#m>BIcO46O?}9m&hgj8ki5Yg+fmnJ9k1h>^9prqrdE)B@;_bxQLFmA z$ESr$0TVpWv`?*#@}DU*kNc&q&z85a|iI6gd1nS+^N#xrd_1FtY@-hc6t zAf4T>M2+Q#an8&*JD)f{b!T{6kEII}X-@L1Vb*}ZKN5p-v)jl=^+EQ$QR~F&&RT=e z^C@sol4AMsK$mLk@9%GRn}eEZhYW{ok}Z&=g*~_CrG3@FLNI2ho;7d`g@_ZP;dA`c za#ZotW!YnPNGEl!Mml25A=RP^AqX6(Mn=uexpt@HeJ4mEu%mr$tXuzBfDL@8{*Pj7 z3Y$U`ve?bXY~2Pt()h0=^B*0I8#)4;O}nZ;kZwRF2v|caQZ-ke=@|dpjfucc=&pe| zJ{c=cZ&eJq0|l@oUMl#nRC9vv&mzODHT3632vLoqN8=5Y%gBlR8im zKi!`LCWY?6ocuAzE2O~j-;Chr!rD8Yhvx|tzJ78HpsqLu+wj}z!-8oVM?F^(&U-qd ztj$@It4Ufj@Pilkez=e;XGfo;hS%6-!J-fgehJurHg&^?4a8zFvL44UYc)QX7ouje zwiiVZPYHVVK%0S&QGEDW#UCUU>f+)8<7vxTJvqIU>AXI7!oM|$hLv_Dc#3q%S9Aq~ z=D>f^CfK2zbmeh`>b%GblAGP{P>>9g&ora86@0*%cfDQf}8jfTCSt545F)zic;G`khI!v=;Cz*!+%;>`Z5I>UN`8&Png_~D6U zk?c`$_CdpK)(d=i-U@sK5)47d z5sXp}SoDnY($AycQN)CoI(zYpcEtDg^$nQm<*uZ8{Ja2elNZwNstSHq8&VPGsu~(d znA&5k%Y+cvyfS4q);}y-^Oo|X2d4C7bAjL!@9&S6NZ*lsU9UX(^99EFc5S>|?m!FT zbq5)`OI}3nqYtdg{n2*G5M;C70P4Hh5n*3#WOG?gNcdJZsi#_fX$$_5n{=?&8&W&d$8|`2NSc3>EoRkQ#lO*54W&-;aTqJTsKD|7ucU zPXJ2jXVckH)A!XRg+<}N(n9GT*TEL4Yj##sT-*a9L^IYv37O2pfkEwN2#tPkr#xQ@ELQ|2{ZU?%P+n@ct;gOq%ZQ`cAjXKc2nBj)76bR*loVuYK^_LPYK=ixIK%^L0dVDN(37pb1yXwA)I` zWd2jGYhJ>s1Vlf-2#ktkwX`g9|X5@moXzY6R z0DcM&)X!jVY7n;HQ_n9SMJvmR_4inN(d&OJ4-v=|kpDQp<HAMRf^f0=m3QNsfDM_fzE)W0i(?w(9D2`7{D-p@KF0$1gijrXEZQaeh+1 z+P;{iMwqWmtswD+*^t|r<%M=BQ>@g?Nf$@vu1&MDU<$Yyr9E@n0L z#B@1lH1h7pO0sHNW6NLGhWA=)AhSm^H<4jtm1|;Xy!^G&G|ODsPDI$vUt(j83vp3fS;xQ1Ms6)v z{4ZO%qsgQH|1$LFsVg%p=Zp%Z8*NgFrhhIP9yI@XLg9g(KUtXAX;o~zDaDn}b#1!V zDy?klO9F5D7SG%K=@4Vig>TqnZ2gvjC*x`5RS%w;($_CPGh46kUYdCtt+h~|7WG*j z#N>E~!7bv+`HRkRLd37Q`^umH51|#d%-cs#Z-U)z=tWKr?hrZBwK#svEoW^ zR=h}TpQN5{cV>P%Ona6R{)IR)lBay+j(&)wW^99vOSz~+W}n=MFFeo`aqaMWPS-Aq zO5)SZ&RPmNvl?Yyu0AM|>(}|Z38Pu;pTrAsHu2F9pZDFZVbZuU7vv{nJ0Pe+55^Y^ zFL6*5`eQ6@b(GW90S)H7qwa)Z)<(?0`5c9(E1qeSXXbidrN5J`@`P+y+}_wM851N5 zu)E}B$Cl*g``{P)9E+AGy4`XM`^`dp=VC1XG1BP?zkWc_!{C%Sp_?g>4$@sgUA*X@`C`F*Ru;*5&FtY+2s zn@VO)XG_?oN{#ncaulfQFNYIA4kl)Sj!l`gH@29px<-#?%g3Si+c958Ptexl;uGZ> zwY<`P0#&h=-DrPtZhMvsCyc9Fw&5_*a;hu>pFAeu$q4G%vx?{hJr4%AkW<4&Pbw{o zAb+e2?I-lj4J>YRs%g4O+w;T#ufgVr9?nn?$% zi>jjLC{P#W!AeM5WFf+t6ahBuxr~YlQX$gpFam)1cCR*)BQ6Gif-kW*YR}8^^K%qT zV+4fK%n5IPdOEmDMNIzf(J!6#9aem`w&l6Ed`hBTY1lc}k7+`}Z#V}O84Nj84Oa+) zjgrzWYD?>5le=TMRu51FWPdvQ%&CWHuV&ivyf){qK@*b?f zh)a!2jMK>OZoW1V$;8Wy864mOW;n~hRt)x9wyfmll)#h$;3tv&e<=IPsHnnsUrInp z#h_bSq(M5QkuvB|N~A;yhelE*q(LNFC!6&bLy73?^U@2N_}rWu#bQKtL7lwrn9dxqj{mzRz6$5 zU}92ZB@&~q#4Q{Sw9&Usk0LoMRK?Kn)CMJ=`>Up8Ko5Vvhq&r$WQg_vij0{coQ+Gud91R4Vgx1;xOSl160ey& zKKNq~RjQyqw&cl0+9z&V-m-gwS?bnmi~Sm4A7ShOyGWA|jrZ{>IbciqMGP2Nnf)*a zjQs!A=M{YaVM6HmeiMxooO%z;zY!1Lcs2SZzVn|K`;aR6=+QXP4SO(2D|Y)5E>WGpl&;ei17 z{!WHivb$gF6htus@eZ6f4z*AA}FFp?32@mafO1q0Fxh~1 z0fzO^u_@V?8VrLZKXIQrf#(!d<_t3+^Uq8U)jtP8IUp5$);NI05Yb5w5=f-4-iM8@ zg!_7Y5&j^Zzom8xZ_dWU$|I-mYv%!^0NX97ey3^R--7&??%TMvfgzwrSKuBd7f-!K z=fAYCG`EWogTNEQ8y`AN`5jb=B+tSo-HFQRy74N-V*!cH-9s1=6+w|-XBxVSN?FE_ zaFIh{*n!FfQNU)^$S5C}+QGBqXs7Ccl`y)v+)UJpaVT;H8{%8wX_~gcWW3Z9CrG*o z;cslmsBxTMdItYOJ~uM&2{EdF7J)gH(DwARfH&v*=!3gbcy1$iaOT8s8ylFqomCg-;oj?5Tr!zExGa10gTe~PHsDOs-G&ywrm^eD?z7LHp6ntLS&*&{s~oT1`)zOP zx4&EqWn4%u?Vm&2;KYLjQM`Ml)nAQ~M5oR)&slqXp@U!nk?1=Saa=D;;Z*;tz%qw~ zS;#vxl9P^r_3$Iz`&Vlag>3k}k7she<7nP{EHC4S>E=Z0d0Y0Bru-9UI$a|m`dj?d z+O22(4(35X>b(O25N@J$<=Dh0Z*gQH>;gD?&AxQA9{BO!;{I6ia4YOj7NW=2bfWw) z-vV9zyIaTa0)@<<*G?4^vJrGS7r|E_j_~2MA#Bh$}Dj^BKkYLR6 z(o`T78f7G07YL9!ZZXouhb<9Y`k}Uo93?&@oq!oE0PFu(q685s1Ra<7Ec5;&`qQeX zH~Y>z7Tt2ZT{6B+()^obuebR~6ATsa?>vcC$a`1iP5(1>?uY|Yc~8y|Q7{s`mb;}1pttoXr{V z!4)5jS$;Q?k_PgYviog>hy~)z{9CSGh5b?%t68vfX}^^S1YHY~sXJD-T6g6wmcHHV z2NIw>;MpWBAVt;Lmf`yL&702v*f^!3;5tCsQrK>Mt^nha&xRFK&}v;&zT3;i+!8cnZJQDjL8yi2mm72pOY==@;K} zvq>R5y%`{+wzkRV2Wbh0{gCso^j4Vp%)-U{+Vl#3_DMSM!bsMj?(-5Qb zK8a*ng`Bp$Xk}X($shh;@adfR>@zmlA^YADF2T>Qf*nZ9@$OtS(t8cvnU(^ABtI!8)u=-C^CqMf;DQ; z)QFp;dJT|{XTMvCh>E*=Ii;KmH~WS7^z8*{1`>=M+qv7v) zu_-q9K!1BcfgLAGaRbnIRz%IuI8Yc_R7RVZh`VKb)N$FQPuitYb*^mF4fgsGkO}yu z>FP}4b`9yYoS_8I4`EH^ek0mvP6lskE47nJf|B{z`a(}cT7jNZ;F@v_P5uK;P@9oC zu_fV8E6?qo4}Ir*F=sBLZ!roaiBq6n)toS2O^9Y7B>#*`7yl}#8MKDm(cC@Q*3nG_k)8L{fbMt1j71dK%INOW`_>K-knPgo)e-HO77y`bXGCz(+ca=xR>Qt+{ zzCC>y9bDh<@x08!+&qWgh9Ucl*vI}49t7;L*U_V?TxnGahzeLTQ&raMkuH{BrX%DP zDR`JgNIoRUn=liXjp-Wf##sC2z3&m)78hgX!@m?8+Jde0NF%)Qk}WbqCHZcs9-Im3 z>`}RB!L4%spL^L__8~m2V=My!A+h==?JT$>r^u&Y30?6`d%lCaqj?~`CdY~E8{<>yoa`)&lU<@g zXlR65-m=Hl$24TmZBa;{V^vL0&|o6K3r%997<{^Qz6(cywkS$J)KtYLjaXcdeW$(bo7xrA)46nb$rH#Xd||nG0R#YnE1>4FXyolc<%^nf36o?<}(boiF=x3-+1L z;8Uqq9-H;M+%+yZN#wUyyzf}2$92X(=`HCw-|_SS^l3=+2-FkFiWsI1K;7h zs_htk3rr6Y_2gH-r8+)~v7IBcp4SJj-O1*ECG}a&iUz1h|K+zSarZmKB|Z4CBvU?g zw`W4;7{s?g4=K5GyzCol7TVkPt;kA$eM7C7_UK)n8D_~$K!KU-_6cupUOM1TpeY@~CIRHM$ zS{bHi%_GHBDP!+quhx*@L8WX<)y3B^UpHjHV`VlI8tmgl3->I7pW8z?YAE z?xjRCe40T48)JXw94$Jn8!MAO&4|`kfin4%5P*;>Iq%}_ubUK&$v5q^!)9+C z8}}tvvZ7l=0SRZ5doa?2guC{?znc$6Jb^LQK6_}HMwmBkT#y>tCQ^D@&gCt_6xXR` z!4rJM--}z6?El|dfU9&IlV5)~yJ?7Uv`}-BMM%*;a$@|1@fH5Shy5Zl*-}<^Ki$X! zW(tm3CV~A$7Z94nL|P47VwwuA2A@U>46oD;buDjc1)avZ=fu+#SjjH$G|oM-G|$0W z`U-yqz14Vk_t1+3(DxEbmc;ws>x_lY#ezF<#=p^;(-E}Z#1zDzo%fw1C~`CnU5;aj z2QPggSU`rh8^5VI4_xkn(r*~V(P8nl^bDF82ynr{mGEw)q|o6Wp3}>6>&4NnRHObv z$!wXgxkeo8w-`rIYj!_PFs~1;J5fHaFtYDz{N{&U5BN=gK+*Gjx2HZG1?~s-!qY7wk*)GJPD4LeFJ*t)Ai}E z8Mww|Hd%GVdKI|GS+dM>_7`@}Z8rQ9N_5FeMe%uFZgTM;cCMKoV>Dy}<&;B3clOqy zhpuwRIn++4SMK+c_djww= zc#_Fikhzf!BSRUtA7mL-V_;4Ux1z$SxNogmI4wH;lGjGX0McGJd{NrKmH1S8Fi1+e zm}`c$KjA#(75(|v@Glvim?7u+&LD=E<(5M=n%jQH4$?qB14>gZ3%mXGor9YZzMw_` ziEnijj(E-+@#kQ0BSh56pvxq(TpCBGkUyJs@3nz9fw&p_c)kVK!~oRngk?qujp-^q z@mseOmEo!nMcC zH~OojROae2eA03H0^zv)-#+rqXW)UgbW2;aPSD||`kevJP_PNpSBjnJjj7W%xe!K! z%)#1o=2}wkm`k@b``I>1=(;0>03j7>dpi4vPCh$;;UFomPwE_}CB}NaLOr|y$AEF; z-*`u2-5FUpt(_;#yxZtysPnK0Fg88@Bt4!4@OOjPp|Chd;qYxrN-HRmVMIkhk+(4h zSK!u>r=6wRnz=u1Iz|}Cn1vzTJ_ZFi$p%as* zW4FJngnrEw%Ce*V{$#y7*V>a6>kMb5X3kH_4~E;j>9#KxCuGUnw*Wu_m;q^UIRV=L zx@$;~hK4d(8zn5?L1`>0Ig_j-zGcq9+j={zuDJyLXPXWduRNQuX~aquY2tJSIp3|M z15%6%09AYFgJd*+x$gIk@P_M#%6yf_;3q+uC{`rfstu`_AqY&-BUTdT>Bcd45H~Ka zTAK*D9}f7~cd)fryZ+kg%d+UZzC}dcw)Aka*sS3<)O3EP{V``G}e?`2AY-BQHVqI|#_gI0bb%k4tOE8BidD83cJp z!YV)~mgn{7ZX@P#B-k1h)kU{nH=--YYJY^%6`I{HPG6O5w@0ACB&f;Ret+dt*Fd#s z>aA%O>KZKw6Z{>U)JOmWZAE>@x5`Fx5*$P?Pb9Y&SjD-fi9rg@q3`r6Qf#Bc@UyGl z+#wQ{KPU&5e@jzalEE4^ zp=gs#2QyBasx0O_lA2hp)Tga|Z$Dnh?*M6Rg($vFVj1_4`uLmNA(3?A3oEHE9l*Wg zw-AWH8Hat<8z|nQ-COlPvGZA^n``Ngu&QYG-T6-416GJIT8aaPQ$Ny5*Jjj z31ZjE=zf0S22iliPbH?$cCuI+sotJs`|t4@wjSG}B`7jKu=q0}fgpLqWkJzK{{>JU zL8ng#0#t>BocTCzv_CS{eh#8+kcz79@}y6s$SsU5$-uwz8xjtE3Xc3>=zD6fu7OFI z|8m3nhmZf1Gl-fcT0kUiLHZCYP4Qo*gnxk#87F`80O=RYAAHX64$)ccl=#c|EsFa= z*O!cV^v5qB28r22z2?-tTu1Q|62_8*v{=tm_6I#@b6Gg>Vc0!ejurV$hZ6eZoxClZ&oFc zHJ-lP*y+&&5p?1MohBWTtEj!7s&bElZbbdowX3jZ!@iIAu2fndmEca$*!w}@FCoCI zI5Qa63wKGxfbp*am_F6sZJoA+NnD+dVGI{10Aec;oe$FR1b<3KJFT1KCeC~}XGW!c zspF7m_K(BdI1FUSQw3DqVjQfGa^E~Ge%coErSXZ%Kfw`FS5>=F-iwp0K*=~*{z@VW z!Z<04`n7_!ET&KwdNz=2aP<7#E|f=rkLKxgI[?0e^?{T1Ad^8}OqrDL@4Alv(V zr6~ZO*zK-}9508E$S=0kF9XCcQkg0w z3_NHrk-{&)m~mb+UeN&m5YQ-K6+?6+Cf;OiN1kH5hzFM(0Q7%|^#QshOy5~BiEqL= z9iDv_&|~`>*6>V+Agby4G7|Pt{OPYTj`O_K^-rVB)A(ulkX5MBN*Jo~r68lRBbhu> z=1I$^WI-OcgTj%sAQwzMbUwKHTnI2|e|Lr>*n9yldvKSm$Q^|4=ERDj;Jt@F%Y?^djAT@P&bRq~ahPu8>mhXtc&- z!+7_B$pwO6X|d87#Z@VFuw8CLl~JjWM#}2ZjjsxI);X#S zyh*))gu>Z35iLA{hf#6jR_egr2B-8NvKgT81p4N0jK2Mq#o)ic4x4uDKS4lc*HHgX z@YssrW!Y*%36p_g;TBUgDpyI9VSGxaA>)kig`HTBj6T&1Yjh8Z(vz6t@%aaSdh}=Y z2Xy0OQ_om29ejrEBsinGGq{##H?yVpP}h*TB&V;Utx%1J;?(Yjs$z~NW3N(ff`>Fo zQU6o3<#MC7R;<>1>T3H7M}pYS;U?8@mXy3{;*lXXuL*H$y-Nk-L*I`)3>1AQZOD0) z?x^+5V&8j+JG&K~t%s3VCP@|=e!M4VWYDcAYX4CEMh2O2gEkeGgfvRFy)322x+!i) zoFDwH22msI%Q5WG@t1llRkg^hMf7aEKJc<%r}`JVn6E|ht!B9Uo=sR4Or*xtiG#c&#aiKChP3yKbUnlPMrwEV5PB` z6JvIhssBVIo;kC6?SBJj{l7kq@(&*gA#BdYA4}%F-{EuFbTV)i^K@!@CF}cj3iFIa z@f4yLjU4J2SxnJUB{r_PxBQWe59hLaOrpVyLY={0^Rq8^)#Gq`J4Q({L-Ea5(O0jT z#g4F9KAQ|^eVPp9)#A7$F0OegfWGRtsEG4#}zRW~!< z;u5W$(lz(ip(#Dm=dhQI|1#O5KPzXHhiZ2{dGZOJb~FT0!(@aBs*;WOd-CFqF7VbJ zEtjGbUyTKMpV39xDCJa{3}|o<|HM$J&4p`n*YGjL2iwi39j>gcVn})ubNEL>$Tl}$^C9G!U4yLjvXS23>Nn=du_bgjDv1BPTbzic%zpI z)i*5?!~UFUbM&BpT%b7V(REN|Nxq)ak-t_% z^pYmi2ENH__v4)CQ^H?||b-AgQI7$)- z<%5>$ZuSHVE!2IC>}wtiqWh|Ad(?NTuYJT%XFK>*fk5h}(QngPqQ_Z3^bKNlJ+1xT zpW1Ep1&Q+Z9G-4I9cY+!S{djNXcHo!CwSRPOsn1#d!edwBvbHiqTo4N`~r63geTY4 zBB+;>+5l>~gt?!D zJy+6oMoqRn2G)M%Wvs5{9sXG(@6Wx5+H0C((9_6aQ^*uk>Ctp4BT&fPbUjlPq#wj*+_iZQ4ZQeea8J?_&OV3)U7-g$3bK`I7HEov?wdi5qXy z_$_;q39;%K)!+B6^W=!}{&>kawH>+S$RE4WS9Au(f5Erm#A|R)u^6*|B1iifQ#$-` zyv8X>edBZ=mD6KWb~C2#X=i`^!z)G}1jzw!CACxEN7~#+Cobkk1uW+KieU!uwHhBx z>*^iSDD&h#@FqxBXucidR&@OZOOETFhwkT&Y%bk*thn3rPwF!DhYdGyVkbJ7bIw)A zo=#^`SC?Vamfz#rbq8t>V;i-?>b_6dWGN9Yq%n2sWuZOHwrd6MG-#ZAC${&Fs(hJ* z`nc<(MbD6jtXnXO{KeW=mb33X#s;K_bW!gss+y;l={UZZUC~`$2)#9uz!$8mJvAnk zSoaWHc$;P&G7fL3$YK9D7Qhimt6oKQx)GNf7HTWUtdoYlqrG-(s!QPhstBnPD6|>o z_2vu*MDnvnAKl#=>Tg`>wh<(LCpNRhAiZ*%SakmN^3$ZC_vMpK@l7^2tjjOv>l8Ye z_*FNnv*ysqzq?i1f00#|I58MAFZy1sDk(2kCvK?nR54%dLRSr&e3cFC(9`q8QkPax zD2b?ZN1F;D^(E}o3ra`no+#!wT%oV*)f^tPS}&)}ed@`h>p?&mtiZ&Zp0%;#WhT|a z{&%~3uHa_x5B3y}xtw?yPg3^#6o~`POY%JZZT&(7Ja0HI&fjo*;@Ic3_+y_FPm1Y? zKWA!2poDQRBW)K_tnPix!3Wvn`TCfc^ANq~D?=!!9Eh2iNN)0G30@B`eHzO-3_9Ji zGC?Syxk+Rqzb!rRMMz966l(6ieba-vYUW{VVWS53zdVINk6-cqk2kg0`qxP- zV6|P}k^HPnmVP5@%IsOr&p!fOQh4~5;Z^~DGz)WE2m>p?f@gqS$((r4mp!^ABzM&`F zX_6(PQdR)tCBKYg2_aknsvk6@)hvRIFF+xKdbU3iby zHLHxjI9MDSH?D}5d>aTk{bpqCI_JCl1L~}bJ@~TKWGzk)#-SUEWNx>L5l#a{#HY>2 z>Bzp&&dWl_?Z|FeMs3V~qC;r=96DL;op)`65BlGTKI=7^ecs*OZOHird#@&S9a6eM zPyZXz8||eY$V`bzXyrEpIN^+^DnFc=|TRtZLsd(3*0uAP!=IuUj!?7#H;H~ z)KOU~nOH{_(ou#ySet;fDIK=oM|Zzqe>pNggyafd9GrQFo6i9^lVIUgMO1cX}-=2+TB(5GvqSG8`W0;0_oof2Sd`?>KTXxLEBaPEiziRR2PPU_Ww&Ejax~AaDhvu>t zq~l-~wYnDLFn%{n%T3GZ=m*mY3*)nST$v9ol6IWeUX}Q6I_x3V8|^%*;UZR^OQM6R z?w$2!{HoP}k$m@5Nj0etfNQ@IlTgxL?~bk{$9-h|OmukDV;eB6^GNyx zPIB1R{i_HOY-Q8#Zibhg#bQ_5yQ-vmCjvX>Am@&+O4KTh^Hjs9?hS7vc`oU zP}8`4ZOR2uZoKo{S0S)3AdJb68UZ137U>A9IzyO)*VUTW@0NV#HQzoblG>kKQsq0c z&HRx+J030<$U)qrK*B*m@feE!&|0z_+y8)4@m7f>jF-Qf4xKl?*HPW4W_sQI*_Hn7 z(}5fRYL3A>Qbnl<@PO5W2dt=v5nTETf;%6kH`o6FKJWWeT^;go;Tl7b0#mQGoH8*v z67K%@^e~j3zNGM($46p0-AL~c`07%BXpnUD#||u?B47Cg0yb&%L?+Vuvrop*?h;X` z%PV)7rxz32AHUtZ)(#o}#zncRFNP^uW$W8p42gvYzuVr3*U)fB_mWT9-#qGDq*s-p zYzxf@q#r!9P}8?2zyIKMrZ8iFVugx z$93c(e^3RN5Q+Q}n3I`_1r$D)V(q91*2&qiw$mpX3dDgBga85f-YRN6303P?ux+j9 z*Cpm&VQ7_C_+8=J`1$i;@nFp~L3P=Vm#^~Zj6R%-Nk>-F(Tm(YqkcL;PG3O-1k3@B zm1gBJ-%pA)s+>D%FBKfG2yEUsw%0pk4}Nn7H~(O1CQC#4mcQ(!f465|YBImq>5Y^t zmoqozB^LOD&q?k}UGnChAm?6t8youhbdykglDfXQ&P0CEnsHwy#f3409hJ67mPRpL za%bcYM-QW}(1SGgG<+<4w#nC`pQ?H})-yF(x5>>ms3+(E&$?Ik)zaIJLP=V}#%hKU zGQ4GUz3s0m%11ZdM6J+aGj-%`%ii+=8mhv)vsA9njM`G&ByTc)bLUUL1*b4gj}?er=3#qRs3IB}}X zeq@K-y9&j^T=wG&JQ7CBIz0%U8 zErN&{K2_u3a?L5VH?z~(3}v*?A(m>=Axu4ecFqlA#ddMl^1*(dkG`bK65)>7 zA_uJruC)t#PekiRL$9TVv>a3CHl^9=x(g|p(&$FMq8tj^W0a~ zAAPHPl>>J&G|7Fj5PXcgTUiO@$#6%Hb}xUfwtKM3`AXmp`o+^CCvF&BH`&o&o~Aep z9rM+?y>l|~Vfss=d(h!CB&R=zvHavZdH;aPTs)FX-zhp?|9YqT>4RyFDPCf{G(u05 zj)|9R)&1Vozjv0DP^7-H%NjJCHiK89mtFsE_nz5_aSA#1W~_F3XZ|#4i`CI2@7GnE zn4JrX*W`F|=pPz+w2Zvl#1h%d8AVHXM%}t6%(={fE9uihFck3NjxnLT#hPY^J7kA|XVm=f894EFFjWNqXN9pu@@FM49_#?wQPrYPc1?jTA+nFj^et8yl~L@tmI^ zyUE3C;sN2UA=EAG*o-qf37u*M&#y+kw{9(;-d}VtxArfODQs3yK}`@>q(37wx0@)x z&md@M6VNeU`tIcEAD)+Ya9W@66U8&(Kf`k~i}`OYfY?{%rr5L-AcMGG0*?q`**iB- zn6lpXdMhu~GO5y2&cIHUn6*lZIA&eX&>Q8NCiK~|@70`c+7c})A8-fLJUq`j>!EuE z>>baXJo<31mMCnj-wBp<>Ww@gw5C|c)kqvsT~C=|QEMAhfb0346B*U?z5MG(*9IB) zxlE5(JJLi*z!wO;hAe68F3OgoVs1_XpN(H z?)h@!NT^HURWE!K2x5ltp&)d9-6c)sAEM9vfWx79r4X`G7nSAwNf}8J2m{_+qV2WF z1;66aM|Bzv7h?&zg5g8MWR#HtC_b!Jw0T*LqZDdjhS*0 z*nBT?1-ul2{~W@SI(%%$N|&Rc@48cvtq8S@7yo@jfvfX{rx>QEQ$EW5hT3u3I!wkm zLCjwdx}##^b0c-yT>CB>PBozGjmL-P%OF!^^pR+yir&n1LY9$Ve?e9dWwmfI)<9N% z^1{}9S=I&ReNXV`14d8#S@)Lj!if0XuJBGI<^i)dlO)$4EYkO2PNCSIcOmabeR1_w z?Ige)@y=}u1srD^X^_3M;EfyhOdsxNDlZaf>H7T$xxx_@73X%KuYpAO ztq!E^?GO4pGrc)?Y1G{eY%s`HwLZ`)J@j2fV`60d1-boh9mI(JdDquzLgY!0vc{!5 z6Rm%4>&S#S*U$+A(2vW;B>AR4RUBBgMR8@c({)TNdwA@+K@uv-Ny}VH; z+-1q|+ZZNPYiojk!Ec~B|5EHSo>U~*NUz3G)&(q4ZhwA(AkT)JQu~^omY(T7o#Ap>}WKRaG%A$4^Zq;y@Tr zAr4#Wvf<I;>$W2m73>>C^Iw?lwn6&aQmI*cZw0;+u zzHr$NcQZiFWMbQnO7)8GeN2V}+-v?<2&B&nz0nv@prdlYF+mH-hJHFgblfS(kAWxG ze8zY(s6#**ULd6rpXFF#uq3~ z<|!&oY4~g+RjwrxQL0Qk*I08J4DzTxF`%D*o@ zhK%|eh^hYB^z>{f#J~&f?|wb6Xo=+Gh57FhnF3iX*+m=796mUm$A<-O3*-YxI>Nu0 z1UMZBatGxi^qsA5L>+#g>^3tYt&Dyu%;>#QMd}ZFXm|A34<{GzRK{BF=T2>i+5>r$ z#eR(1p9dN==VTq+8orx=vs9ioqLD(Sn~4VZMGsu`VDwD-i}7j`?M(Z z2W|E=>Rw|4smydc#uGaBfIq%W@Lx-+Jzz-VsV#mpW>7L-1p2T%@ zTuw#9l_l>RtEZhJA|f7o|Mg`$_LDQ{450)a+SKl}hjIKYpb(ZG9{0wTE*2=9U%5fy zLHpD5nNRva?LW9H_?_{Kvl3%-xeESjOb&5&nt77db4B+?@yp=3bDw%eo5D8jMpOz9 z6#*{PXg=M<)pb|!w)`aAf4@z--1nm5EpA7+FOn3faU+$P0(a6*{VZfo+fOm@Kz(dx zpyT^K;?;bGk`@)UV}iuED1Kl-k44RN#7l7n zl@Is#5c#j78pY_U1@3r0;$oZbqL~-W=*EFMX_PPX?6*>^bJDkg>&??UsnQ%qyVbV` zhd5d8sCJUSHsqJgTDli)>Y+|(0?{vgAuHmh*~W^x&p1-d<(RxkWMjSGBC7Ld4E9R=k@Ta90G#rZJS zG3@pYL525(__9TsZ0eRC^ghqn991X>s`LEk!iYqoXw4HbN5}o`ier<@9yHm~9=F@A zXPq{lI+?Bd8%^w&`>s9NfBEIXc-e7JTwbKvfWko8atb@zv&A&SK6}l9_KzxGTv})2 zAL7B9X1j#AHTGkc=s`(}@H{Ifi+I&eL0&SvmA8$RiFm^ry#`Y*VtuzUwaY!|?Ui8T z>iF$|H`m2*s&OZ;xA0-RY|2axoCFdJxl*kqNl>mRKK{EsHmq4)1a$WB^BC@6qkDtq z=lBb>6uE*j@wNFqFDrL?$Uf{kBBlPa+F@87GdZ3`nsUXDqkD@NH<3u7cj4QMZO^gu zXc^K`TcLOo%DGgW@LoRd#RMEI6M-#7IBRXL$Y2ir%P})^KD}%Ye8N8caxLNO&6P!d z%_Z30LFLjWpSkgJZk~)Z#%Ksd$Iq!Tbo8!CI$!Clf;p3XYv^I`AJsoPz1t6H|6IFI zy)F_KN%?5aie)4;QZIYV`1paYf6r5oa3qtD$E+4(zF+ma+A5^4BJcY8mQ3|r-~u+_S!thvp3HA*7Am+=|Q=Zot`>AMT&N&&GJvzVf` zmZ+c&N&1_)iqxk|^^a&KZ@*YhIKuhb{mr$4gYH8u`RN;f8&D}LQOr;s^jTix3&J!PYzUoy^;vu3DI zk@(TE&w@I1s(3PUGV;L%XSKGcb+T@HvZ%uNp6F7>%?>UyDVY_g*MAaU7KIgm6w~Yd zsMPT5eHX!*KPaQsYiGD`ttPv7co~uJ1IK5FAK_E$LUO7DyDFH-KdCcejh0fK8eHis z)GwAwdnil_lC2lcIOqJ3;CW5@90#<~MfWs;fON`^|IkmV=w7Izs#=x)jep3}d`~V~ zhS8%;H};8@EcY~@V>6?2Zr0t4-_PEk(rj^K#7vzXQ@WUv94ge`%efO&m_&3jrvM2D0K;ehE$wi#%hLim2qC+a+0k;Q0qgzKn;bZJ~!JIJS#IWB%q| zeRMXh)uEUhm(<2TF9uV7xbkrKwVA=@ACpR9_UL7D-|yS8k?OS7=dWKJcdufdl#;%@ zOr=X6bdEn~Ez(x>!yOIAK`Tr|%0S@2a};a@r9A0JGbN&?&c}4>w&LFHu=djDQSA`H z8P;8)Ie21~=g>PR_Q-^g^g^s|dc1P_u*unpeNrXi56Lw*&a3X$;)`!u)~!b!^v9U3 z`@cDIny9lAa6vy{961JtYSZ<#S9a0F+BI=~oDB&{-9d$jAt7U1)XgWEwuWl8` zbL?+v9cYkB{u^BXns8Zuu75F0?M${2?n7h0=khbIM&vRke18Px8`HAo2A?ml)}cD5 zZpP50*-1T`R!+aA>j5)m(-!)##K&JacPU77!M56$p`J418n3@k`0FO5>We14p#vUBheXVRLB^s1Y+j^i?D4$brc8r|hrp4uA< zd2wlfrew4yw9g4B(2((Kc8SRIL{NU{v=vV_TDtp;u{sMkeuyZre&$VrPRmll^^?Q| z)x%KJQKhh;N}4X^0GiP4wj zPYgU2{8^55$k5rxv915A(dOPy9C^*VQ!lN7Zz6nRIjQ_v=hzoThLx_I#e5ZE;t@CY z9+?-b1(|X6PSIa&((sc{A({Y(idx`x-}5Kg%|B*HnB_j@v{>E_--D)|XtEZykE27& zWjpob=vMc|VZyL8bbez*$D~eH6ZZPoXsvCOY&}`Y#hDDP=H-(t0!E@%I77U?ns$?PtLHD z4viT{opY|P|?bGZn?J^r1 zZ_imfEV#Ca5lpj0M@20l%4KBmXJF`b?dxJ8vq45K5E;rr6a%n|loY|2@rGe`JrGMq z*CP)6u|lJuB&Y%FzAd5^7s%1O_sZf3?{IfZhwM%{hAXH2>7Yy)C~t2@6ApbvF>v3a z_@rOZvOD^A4aC|=VBYJYzDVoz=kk$zU%S-`yc(LetdJ0EkIkW)?R)PRRAb(F9>Q!* z5BC}=lb)%LFDt!`H5;JkU4fon!<-2` z*OkN9#Lz?yv<;*8hG3c;$SaIJ{Dz^S{NI31V5DQz8RuN{3L!z|8?kf*xU`WfxsJ){ zn_=0e=_z?BBlo;Naf}ft~zbhA|hFB$ENQ}!vNOiN)PS@nJ*$}X+mSf#9igm z8+$4+U*-QNiwOef83Tsy&h0*R($#qJU)lr2UIA8ss4T+ZH~qG@``LFqaZKQU0xSYH z^iC9wDwpq5FKko{c0gGmZn6}fK6x{Kt>MHh6=Nb~2R1S^SVu0fY{O3Dlr-Li3{`Tu6j^iODqixDQ&;E4jaJ^U&Ag>2V?BhI= zeUEr5;KTC!#Uj~v7{UMnghAE{8WqCdTM-F|y`X$VI}BJ1f8Ejrg1~j%OI1OA5U_`E z*Z*P^goEzo;jJ0_zs(c^cy$~o0EoAMwt6JVAqC1p!vjOmA5ia+z>cDt*_Vir`-SjD zx06?2;Fe$xa1o8&+dybjzg+YEi6%xDXFI^0SXcv^99)rbejr_=-6v66LVG_v0sG5r zI^J|yl$zTI>Q|$Xt;J{Cko#^X1sBQ;ngVJMe>GGsRs@;)+3pAH4qt^B4B*W58h46Z zLbg7x3OFXR;dHkpo*iCUO-Z3adfvMAdzVMpAbD8mMFax2asTnjk)UM^G{(eW{t42fwP>*quVV-3&i29K} zJmg2E9-GE$_P^@MO!xydwvhHWGV$*)MdP!?H!oqYsn02W7HCHl3`pAkvsM!G01)ap zrTg(ma=m=~^D7G?k49w9MDk{@%5A2Fw2%x46PHhy=AazV4Qe^{vf7Pym4m0>2OrlDy?Of5%1p-aLR2=HS_ zf+nvV2FwEL-#-6qM}*}DPzV5q5Y?H2>kA)*`g#D0AacSMIrtLTKH(n{D>NZ|u!!J6 zWDcT(1<1-DWr%)M^+D>By_S~N{MUC%dxNJ(Iz9_j87=8D-t&i#1fTOAn?lA71@uIK zLWLg$>$U)8h&qhP69?Y=m>_N1wuecvOK)|mnF)oc&D)3hw?mF-&21+vz0=pO4y$R) z9slXegt_jTn#O2c7HhFd4LS0y`}(vn@mH;gE?HSwA;G8++tJ=m0B0gY!kES8NGyrr z@fFd0Mf$_X=78d8v)s|W6l#0IN5{{i75ygi)hY@(EFi36R4$K0PQWGtZd`pv z+Y@}9p6@MS0@f zgPnh`zi}U&n^F%>z_;^Lez-ae5HJv1yd#=WptKy$y){v3PX$Q%cxrO;*HgO$7m(2a z{P+dJe#oT_2Rs_mqxe8~$0kFveM>gKf+%g|lO9BFewaUSAZ>P_Dc zKwQ0quOjhiv0A||0_z-iR?8-6dPRarndWr6k!xhgqeS$>KT+L@p+0r zrk@802u{so`W@1EbUg@Sy5h-vjVdNboNjFC+b*9MU;+CRWJ>#5t(Jk+z|BqMGN<(5}-xyIE;cM%EC=hiRFIKLY zgycd!VZrx>M`7hPg z_?pz-FOv+G5xnZ?LV?WdxC~t(`otWA*q8VgQ>-d3@rLxfM2BvN4OH6dw!q);<~`Af zh#meW54|1AA4{~~#)pif{`QMRA{#c;8Xgukb^Kp8f_fz;noh2vs^D-cN-Wc3h0|Kz z@mKXFi5wQ|ePxiQpoqPvsJXN>LS1aE7equxL7_lN9wN{skn{Ig%=Fp&ec`VCidz#o z5)Wm~f1OTUn9T88Si}-pJiXs?$cD$EM{T; z@2`hTpeOmSXyVTQFV%$RpR5_lNAi__iPK)zKDYVmaa-`1p}tY=(XJ?k2tohRqE7&) ztUyco#nf_6PXEQmQBaf+hnq0-0yj?Yn+Ls5Chu>|MqO}_2%argANxFWS2OKl?@I4X zVKt65d226Eu75k)Db*$QjOIBdUDX(GR3}ER5{_KINbo&Rl31>So0@edmE-#faVgSq ztWn0rlTopxZ_m@%#+LlY*tM+Nb6DWkaXEM=cGx$2E&b ztgZy!dge<)87`MHWj8H)gfX&X+-1IIE-`=F2|N>d^-|L=)$EY{=Rf9}xahdeig?DQ zU39O4rcwR;M1pb_GX%0++yU3!7hFg$j_PhsH0svc26j>etJ-( zO=VO3ZLKkT`dS>{7lCh;9G~}AuwdakZ^XeBj~%%n%?IE-_RVBIb0m|R!ZCLvC;r^2 z1>^6`{f{w})>dzKzmGC0UB4+tv_U zt%<293m3gUkXKHXsYg2TZuPdMg4-!16yAD_ z{p2b;<6kEt82hgepiv*}vWUx5^hhE1(pXC|X?e>2>pe+QXOV36^_fZT@*5LMog##s zemt?FiOPmO#u?-Gm8CS7EUbOJal&-`#frI-?<6sC(?56C!!hSo#%AKFKer+WTF{xY z`bxw?eoD}>2kGH2MdzPU!vPF60r>u6Q?Bod#}E4YxX0eJuN4vVF*WLyZYMYF`x++h z@{c`DIwE=K+CAm;)a*S*N@J<)tyDAT)FrMTGU>9pp+Oj=1_D2;*d{e^l@vOYK-JM? zDQBeSGVA53yMMgi@-03R6V9Ntx=6f3!5prqwq-t35TJrdqyo5*fx^GQF z9tTJoNC`1jYOLxj`bKKjbCvNmzj@i^ zPN&2f$>-j+GB4>{61YGc3)6maKk}j}Sue>4w|To4E*bHSW8UuDxWmD|0l%Ny-R7Uu zQbzAy&JXt$QnvPuxg0PIb2qRxSfrQnVV(Vt04nz}PU7XgyE#{y0#yk6oIT~NWF&Fu zbJh+jRAmK3siWi5ST#gi_@(0d*?EW_sN}}}jWo%5wID;0IISVC_gP_vUa6&5x>e=Re8LBDS;?V` z?TbMj+8^H1CG_-BOz>HP}IxYgqEM3~iD zk8woZFi6w8Wlj%-oN%&a@A_WJXHJ7LrE-GRQ-AtkUr5*w8Hv5mO(MK#p?5o=@9Tcy=eIbKd2I6LHI2$^QQ<`0 z=I*JtMwVUH9+4K=#szJ;abtveJ!6U0aCb&UWAA-6czyn+lr_mW1gK$JuH_gYVI z(9(yo^>SL5wzjOD-NS8#pQ@D6Jkp-{Y9q~^p&!fCW zl-Xn7ydRW2T|GJl{Q4^MFLRaJ`#zLY-HQQ~$|~2#ET~YPHHlkga0)s|%gL&W&n6g% zTyDKsU6>fXbCyNm6klgJW%w-b0|P~O^Z|YX|Ly*~?>2s}g2wu`IRl%{M)m(pokTtF z0I92WxBzqDz>voCe;C#GR=;L4y|GRafo=n+lr|A~DBz9#AerM7wA@GtIt==VSejp3 z;H^zwnWcXm3Kf%-y(3Lu`;|8=%qj1_%B{!QftLXr_>_* zVBl3wK%UYZW3gDEraF`h00CLnyQaR^PW{D{#U!%E7}x^Bik#(FQB@8%-=qW+v3#_n z#DM`3@o6Kf_A^15DvVgZy;UG@vlP?3Pt4&!cJ6jMDqTk46|}-BM_#I&%M9o;?1^JT zO5qTA2OBj=B*c!+GXvSkO+D9)I7aITrzm711e(VX-h^(*nYaFfVNJJbXmzJ{TfB~l z%1aQA7`6oyXmJF7y;g#OOx&*f-({c`qB?#5*RNlkqN1feQqv|f zqM@|7+v7%&X!!Z=7RmsTAo~I=9!^kPSz|TfkC03!u6%J-%jktVb~pqGG+&Q+wce3O zlsm^ynCiJJhlm=Gc^h~ty2%v{dDb=HRRqY=iSM9&mE7mw+kO85?eoTa5H%Aw`81m zOcbbrs++Og~NTeUBS|EFN1-lde#ae~iGd>ZofZHsg@%$?u`(gnWZ zsfEzG9`<_tO=m&U+T2z?V^2>?Sc+d#`K#0VdOq+M=$Ac}t`^O1K`Nxs{($74An5Al>pt6;JUb{e=Of{)VTHKcsVUi=&~<4Gq@qZiu~)X&?&B3FzA z?P)wI+#8gGnmj+F)eJ!=gqTVox8n`=MP*8rbr_sU0EH)I$v>8%?UMHirLL>JFTnCm zRQg%R4%K~c5Cb}(<2s{9#}m64id)G_D+IQ#SYX9CpY2;OXz`dm)k7#2DH5r=42R@D zx_P{aOr`-^96CTg01MJ;p^0~W{~Gp@#89MV^rh~N_T`5nyI^)8OFQm0gFocNnBtTg z-;ea#BOZ3d*bcN@mR2$7mCh(tp6S)v&PGhxHd6sh$P6Nc{b31@M|Z>F`kw2cux&@u z|8{ne<`jV2a8>st1;cp;%6t^hX7^}yktZzB|BW7nY6fk%PNuILEwhAz?FgY2J|2u} z)8#y>oeEkmc$nAJ?CEj|5%lXge7ld-@7-l&qzh~TBnys_HaV!&ovY|!Z$ro|$SkCP z)9auU5$$7U#x)SBsHC*UtflrLg5~Uoe2bGVBHf}n83=J)ek{M|>kwDsmWq&T3pGhvBRR_uItyPp znYkx9tr6#i`ezmm%fef&mv8Ez=+r%WoH+^T)I-jC^v}5w0tM7T&0@`^xeV^9?^&WC zyAVD_*O|T4kVwx(`ajkBr}u&J1SKiZ5{b&4s#L*X8h`_WV%f5?tys%opy*V^xIk5`Ls8ZPu9_IZZDf!osc!`1r8Q?X-yj=0IvGLbD8mS@EBn)=z zhG!=SU2Pco`{Qp%E$)NtVTJY${WEh=J&n1ET=7RC1ve2wq;%>pse>)rFR2LknpVDs z45 zvbVMiS+SQpIyww77_nM}&Nhxt`Ju}fK;y$rbixI`a^VOPu8PV}cXM~wdT=#GiP5eK z-JdEfpkmE?3m!v(TS*0&nyXF%RB&s%=36Jo&%*<0!5hc%TE)RkYWG+$zR=hEUbPzc zOINxkL6k4}FDuYS1P>Ce5>T8#OBf5hAvhp#^&$5Phy^%$cVism12s}n1aAcnT9xke z;$OU02{)Oa4)lMwJb4Yq6^RU+BHemY30$}o;F_7{>i6G`R-50RYpbK0*KHM|s_5yv zt~mKf4@Dm+2J9Nuy_!h3x^Go*3rDg#;j@N2tD&c-(4o&QG#O%5t25Gpw5m;K2f>{Z zOKc3ryLXWXR$#{alSgR@*i-096_NrB0lJF3`q;&x^@FNuxOT2VqLS8wz4;Gd1%x6i zl*NoExLi!#nwB1c7iT~q;(a+PVS?D|+EBWVtBj)jjj`gYT@1#L^kV#4?b1oJmB}GQ z79yM%%lYC&`EJ{q+pZpcYzESQLk;Tw>Rh6@EkM(fd^4D(#Y9Yu@sU}h23THRbnSzj zxI?0yW`4moFpb=5Tc;jf?W4~mVJxVNjrTXLqRiy8Cy%J}wzRIn3|V^0)v{K@32gEd zoG{%-?-KLIH$RwPcB?>VDW+hmPcNll<gN@~O*wHa-H%uknBJ@+| z(o)wxcJ05U!-d&EA@v+b;@zKUE98IF7@_`oqNScR{<0rX?ivsnQt;nZ%#( ziF_cJ*7toyJblwW)m<#vSumeawbNW%>JYDe)AOx>nHP_~Dhb-qP-5L?<`eKvm8(Z1 zyn@(~ZYSRyA%yFsGrPz-fz)8~{o-;|kZNhtWhRtKbWLf`=ZHBg(X{7VjF{SaQCn_h zZ1U7e`_J$B#l~kv{oh|EdHPc}hCSq~%3-r8L^J zw(HcEC?Rd}`0$A74M+IWp>B2!}fl1Ij2^F zSb0`%*W=dVRfzz~X7z-QbiSCoA>w`7_|96+A~=4Xas|)4<~{Tp_l~MY@+diC z8ORrnv+lQ^%5#>v|EHPA;)+)86G$prhaJWx+>E%mg;V>Mr9YXnY=zz2q_L3#-A13< zX=crpH96v5pD;h#u^EoZA+m^!2Qc-Cw95f3*IL{O_CL+K@8Llg>d5bP)tJVD_;4bY zS1)yM0?;U_|Jy@6)noXiX}{8Kpc6OaaEwS|v-0ipn^l4>tUg~I_6Ci-VO33Kh@zMiS3O3}Hawf~?kSRi(o?bidmwS-L^k4C+s##?vQI$Q+JQm|I-f(g#p zdt|9N1R@VQxq~vaZh3t&sFxGYiM;m#ucciD{o;_-pwYC^)w=%NI?aJaW#&2=iqGFs zUF|)AZgwxVZh7n#m8%LS^AG>dHenCVo3R+(>ZQ=Hu@BB5zhNTTX_RocH05cqP*kv3 zz@-@aT2ZVWLV3NcsGt;=(-T&sYAxXojB=~42KLK(J$(VX{bEf+OLmU(MO#Z2lo1bV z6vxYVK3bdyy~2-?%WaKKw9GsDblzZn1P(Vvn@r>p47WfwT7HwBUO=lP|v5?7J#*```r~$8tB)8&Sw}r?qv&>2s&&VsMb^#UHf% z9r`{=XiKDdH6!^Mw}E{3j&YFtDB_E0Iq8y=H+h23U--0|%|kWdb5)><%NiIBGMPk) zE)BEi10u?6m9AhCmJl?>@jZC(^%I^S1Q^`f0aw7}&Ba#FG9BkUd`^+u1lVu)MTHod z3mH_iu>~n*tOv=Ya*4$~$nQ!d5L{+H<;2Ge;A#09B2J9eFC<7JF;TMYR!(cKlA_5Z zksm?Q^VPHB?CWQC9P55|BkC&6QTp|PLoYcdk0#WJ9_iB1=$%vK6&lfT;JIfGq4G+>YqXVsr&UdG$^|oUsw`x+ie@hYNw+7z^_yme_&~JG& zlN;d^Ge?F8VZXbBb;6{!`A!$dZJCFSx*vn=B>air{Al(giYwu-g1>x^W?iU|W6{!U zYc(n;^X&@QN-`JWYamS#PAB2V9hS9v|9R3)Ssv;Y}?6!!O=Si&B0oHhPjtZw@}aW_uf zC682A3M#n>50cm@EkuAWbKRTUED-&5gQohk0(c-Hzhrwew zSI6Ebe54Z=9)2f+Z@_ouQg#gBIcQT?DDt~2bl2Qf{RHMn2sB{$l@Q2Dzjp|+mQqK* zVg6-r`eV3hjPWEwNG2}CdNIPtW?3L20NK+QEf!UD7K4kQuM9eAivu zq&@h_Pl&uy+ze^UAeC^dtnUN;%=pp&nZa?uz?DPH@L{;J2Jw}UMo^y@cpYd)A zN!bEmB%`6hG5E`^uW7Hzs!Eapi!vzJK)95j{A<6T4{kA_^^mnAY$>r`;VfxV2{4Z& zA0nw?RYOWchwUxw3N^0ceokTosRJqGo8RZ!LhKMq1TaUPjcex}@F2Es zf@Gr{Jn6v10^bD5WhnW{yv*dU8 zuY)-Sft{3{`06UAAYc8}qznOs?_{VE!Tr8&P?1248o@ZM?P1{6ar`dDzg|>{B01gs zG6;E7}w^$U~kX(kTLj%9FVrMhX)d1Juf*?pT-# ze?$OiJ`h2}pG~NCj8#h$jvDrqNSlBhm_y?5 z01E8~*u`k-9oGPszc{)$r$Iy|fF#zIu`H^!OeHU&Kxp*&bmm9|Axy}qs75;X{uc4d z$jFp?9NB;o>lc6HE9dAh;6t!!*pY77FyWBQ`YuLNP_IfEt3!y`Lc}hwP!jZ9?S4gs z31JHt;gMb&Mo2m(ptPz!ldix(^ea$0fyf1R@}QdMd+?SPbUIKTdZncK+}$frO1@x; zcjIS^oqFEJle7~Oc&UV9!l$i{6>Eb4=pm|;=6pNE2}eqvXE5!K_KAk`v(*6X#5YxB zlropckaCQYrV|U02N-yJ>$&h(=Gwr`9Db6cS5f$4H*D0u9ez+~C z{Q0{q8kr>%xwdXHGHduw&1n(HcG#g$Zr1!(S-QM0kzgJ>doo(P{WIX&=c5tHKCv1q zj1OTz%9VOduWy5ehy=P>?!w;yOJ7Yq8g#ILv*+BXL4yWiCsosYN#7RdZV`rwfZpl9 zy^(;4D2_-v? zYK}L!In3i@5%`7Y6Im$HQ=xnmq8yERx8V)U!|ERLvY(t-{WwmfC%~5nNeDDTEG{yD zJcN=_BrjIUFhe#`U?<)p;VfPkhPlw6lu%lBSX z_Gtr0fCD$UXHTy2c!$%seQ4ns!XryJnIPHp zgdZk>NWG@nAx1Phgb-=6u=vD@yXY%JIvfg&3WV}NfC^X*UcY`Vq0f|~S^e521q1}I z<<>rzX;hkqkIHqb7Gf+dJ^5GOCNEX|p-st^C_mP+!kAcUjSw;ZAWENqSqZG?Zd7jA zH6WN$7CsW`b zTFDG(xUKaBxTc+)Lv{Vx0a!q4;v zUqZR(m*+YeJn@5J-7@biin3P5amw;r@Sj=J_gku9`|Zx%fB269POWOTfD-Ryjv-Er zbDOjqF4rH=U##P@*R_?bwvQ=hahSE$%*JUsXBh>Z+uFEygiAd~KD;;U79-@mcS<7O zXqK`k%pp9uo_M2$A2ZH{+FDlLY3y5{{*0SK#`hO1{Sx?mXF(OljJ{c>0gLZMAY`bb z*TJe@f!@t)<^~zVx11IM4|YbOwd9;Vf_L>9MY4Eq7yUs25;bRhoXmb7*g|63CbchUrlq6=p~sH zj|`J!-~KeYNgYYm-p_iOZW!NRd}!{X%C^2Qv~S6-WZlB2QO5-fm1AT@!_+S>@KYke zOx2b8%t((sdb2X+yg;G}JO01!lbGZvKnDMpad;{_ah>VEiU`f1|MfRnXYFzD7_4J` zzX&@h`8l#r>4E8DO8ZYhhlUH#7F zLtne~C%3vRbEu*l2z@H|(wt|?rbu(v{Xfy+**2*xOVtP5a+9`gv?LC#nTUYy8PW^iwYNyqRG31tz}6HzD;BAtTN*LIBgx}I+LzOKZkezh6LkL$YEt@ZR5T?%v&({>Fjlr5~&B{@!Ep0E)^P7rhEe=+?l~L=AS)D(8<&#nT&+o2^K{Cb4xGUJ) zkeA>2`gZk*G@LPxS->T1BPoQs~dF|DQ06&Davl-JbP-T_yyzCi{?;2`)uWMFC*%jG*eEwB(h6ucOIBJB+prk zxp5ozv>mbqOxg}AZ#$62JX}u!v%4T&&ZP1en~R69)Gy)akKe5;ndC8l_*U)mQp{tL zf`t5h&j=%Ig&cJ(&VZP3kym^Q+Dgtf&oBcB{MhK#KKzW7e}q=$8Zx+po8io|38n6Q zNw|{t4HI)@)$Ji;B=!(ox&EyYvbX+%g^4h##wZu_8gX1dpEvVUTC==6CPFU=B&*HI zZoHLG#dn6G=g+wKq+xf9=iM73_#CzkJsPL0u!Iz~CWt9>~6)_pI1vDe5-*QC)D+A|ylez=X;y-5*z&6ZnYjOUpIo`;RQMV zd`Hjj-MGOiZUk$im6es2tIb3!i^mh^DUI%Qifx=<&FVV|$#V4MXWVSG(p%GK?=nlC z-wj-S#o6yXllUAvQ!c7v{{E)W#Q1S~>V(o)_L~Qm3ES_T7jQ_%f`tEp2quPr2oXG` zA+-Cj8Fq~3HXobyw>UhCY+e|#ZYatRUGJrc6_|mTMg?iADkq-DW4N|{Qx&~YooeM2 z1i7u)cN98@49GRFN%^&Qq1HJq<-5}>Gz+q;!+YkUajj!jyzvh!7Q6PW^haOt!pMAZ zp*g->KE3TFI+j<3J95ljWSG12kT><0_<`oMn=jLoXsE@?En#P?|AewJNBT)Xb-Frhg&$7WP&a)$m1Y(m;=HyFfY2z`_@-3;QHI)I(cG z4pb9{6ScN!qq+h@%%RJaSLwE4tOdPTv3&m%2UytH*ofri=0bEJA2M~Sv4rTta>l0C ziIFaxr(fvb!42j)9I+)XTufN78S~+z%E0Rn{|QlLotS;Gn*ayL`NtmpR(j7a2)(VDNgdrF$(WThS(L7vtv$*5~vm*tcl^7&4 zY57XnK47opega zU^5RG+VM0gD8@qX0vTFM^-B+;j=u)DLezpcc%mEd&P6X1b z^y`v}Q-IGITn(j7lBw~^HCJHlwz9&Hj*iBaJm@7@NQ#vxzZ7@b5}(UG1C zP=5m<2F`LPqZBE!V8k{8HAm@Nr(9kdR}fr>EZ^-QlliAqw)}uy6>|XScFoOrCxCs!VMS%?uy_ zLE3+^L{I2-A&!MQxIgM!7lL7si1ojtwr9_Zta?;^BybU%jMvfhANW+>FsZp{?_%Sl zM#O=-ir4pkc)rnpj*lQ2l*zrOEwj`ueinYXRFS&?;8KXXte({<9uOG=xsFTYx~}W= zx?Tuup=Aq5pL9hrrhEQIdNmXKJ+UvKz@r359TGc0rau27D28PesEsd3u7qXDCu}(3 z2bykk2>0)C!Wo3n5|EIo#S!X14S)^IMz+xR@Mk|yG#m-%B)m5D2L1WxS^T-Ehb?t> zm+Fjb(<4dIhUlgJM)~w4#x(o&qg1 zKBw#4%4e@}ei_@U+~9WaEu*k671$vd6mlz=e&79yj1s=T_`Ul3sHppgsC|>>@o2-- zttsIxL9ZpVOrx3FneOg&FitV@+}rcVd%0d*J5Nt@j0-9I&KqQq&{9#WtuX`ck9Ih& zAyYQ>1b~5qmaZiHWzInyOAUJ|TkVKFL!BkoZ>yXgXJPddDTc7m?+zYF7XbDHR5r^j zm^p(5lD7g;h%E(a#Q6hUL(CQQSLdrR0z(4`0;-XEf{lsFyk(d9TM@jZK#(JT6=kor zC!coMZv1*J%|tM7|0yiVwu$_vz9;l|zImEylb6b{dLqvb z$ym?iBWIZWNB|#SIfYjqZoUWs1W}gbHP!~(OlhOJST{ef{Bz4vaZ3w`(s|J5P^0M6 ziewBtt^X)yRClzu2kIKO)#MGkNrhbaq=Jko?K9lAfH6?~Ehp_V|+B25k(xz`}(Heokxfy^w;iw`)nnFait z-)OI*O$UJ<0-Gel&O%I(_`2i6$D&N34IzTga?G1QZKxfI1XBpl7evpofZ?dMSBdG& zQG>!Bi)tc}nSuNWI85cF|3H_Q^ZQORqCCHYR;>d@(W~V)-%P7BL4OW+n z8#sULa(1^lH-FS=XEjz`Z})NO(T{$yx`tIn+Q39^M|9$i^0Km*@`+d8352;Zyyj}x z)tUI3rNgIi-M5M`n!$^%5x0gzr1?dM;T_sXbrjE^iJZ={?9*~i9faeFh-!}oOqLLt zzP@bGJxib;Q#%}ixZ7&J}chS zk8n(5&S{7C4&Rx>YCW{3{lKfbf_!Qe&r8r)d8|c6JXM7iiVHmnt?uqqb)Vn1CDy!O zl?0B4OOh)9pAfVoE-Zpz1MCK5JhcXm%E+j_h36aTTKO zf8Qa9`QL8@+W+6aF_vn0pEiB$Z`*mz|7h{7S>w&;od-v3cE4Y#kk}A;&a4Wxeevq! zT>8G1YV{s_q>~e3B}`Vl`@CJgyKhHaaWhUE;dngZA$zE&*W-l+BRcq77DcDweD92)m1m)!_pz46 zr|n7p0Rih$O-&9%omKLi6Mv1HBcrU`2wCpc5>_ng;wZ)5JGv0JD=~S~ZACP}$7!Yc zYXseF=~lFOqJ5nQ^M}5UrGtEm+83NwTy%|on2`e$SsNbu@bkB}vy{AR*D6z@)qd?u z=8ymj8A#+T*d#v1gxUD!cKzc5qP*YFQ_q^nVz_ZyP^;yOKS&jED~ppx^Qdut+hD|n zizfXNH(sZsh?{!BWm8M|F`TStbXv)3fG<%B|H&^Yld|zGsgG0Qbr+F%Hq|}5ic!Y- zoAT94*P`rtOjS;GSc)9h{291Ba<9F)hcB57eIsIZ;dIyTuzU``CcAf^sfj0iN}4(E z$Goiy6`-<&mtq$Y_6cQHrxLgx`2*>kg|CH#xfH3%`D_I{RHGP7X}s5l(M#{OW$WDQ z^7VDuVHFu8e^f?ubQq>e0wYB|TehVKzJ#TNGrrRW=f`7oGrP%7Tcfgv;pt^B-nmTAYEh{bVRLlc*c#Rt@-_!Cx+dI6Y2DbGS~E zMpxY~|K9u4>ETKxOAyA%8S->RsPL~{KfjiZ z!&l7kW8Zk{v8t9PwT(7*u){{B9cF>m%`bAA=#v)>AFWO9OOl(U>d;r=))GQes9SQ! zeUtZ_U&DP=gkQc5KPw|L{aKhl)6c7wCy;!McJI_Iks#CEjI%vdOkmZbm};Nq@JVpw zw(X^vn&EWr)ROL=@%lTFi(M^V+v|40S*OBm^q_nOl_iPu1t&qkvHz#RpIGT~h6r*# z9Tw?lP{>6!Zl`kPlDNCSIZ;Bi+TQ1#L^;EVOpaLjL+{)i8Y|cRjd}hW!cNXr^=>jg z5~g4@we;nlw@a%Mmt)Qrym-Twb*-9Oyu}|BK z)ezJ57U>g_b@s+qvi9N`46-i0%%kZne`p@xZ!+^jsF;N5RExq3XW!Y^4IYneCq}Mu zcGqn_?T|O+mf*eDocJEeKD6H!_y<3z*Kzye6JML6n?kkS91&e-lk3H2=NzBV;kYDjVExEt zAVapq&#~oeqpb>Ax(g?ZT<=efLwC`N%?PDV`Jzq}HmO4o*1r^rpG*>CMMZt2_Uejj`6xtVg@svn|Y6kga7{``I(; z%QxbQI?(9kqO?rWTXPNi>pS7b+`1*7joXy?zle^bMw3&HM?k;WjBs{~CAj28?7J9GHO{K45vmyUn2`8@h z)RXQ_d>^Jc#WGE4vHJK$ow*ZLrRsdg+&RJE11l^_$AQU}Gb@Uowh^DUs_AN_nlrhF zKcYC-oNEYjdRf15mX&DOwpCqu5Z<=7v6QKCFN~b@L;XzZdZIprT5oK|(Q?uksAK;1 zhGZ{f=GV~B_!`Jr_Pydsp*jmt(BSCDiJ%dUfDy_pKqH1b*zuE&G=M;&OGu;v;9i(T zYBo0ElBewC%I3{#(a8g$Hy8ARO2``RH4H zOh+qJg7VEaq}jh`xB}{SKrq$&EI~j1twUMHI$=pb0S)7m>#sz!GV`IksJrLT zWW~kxJ-V3{2H;1HdNP$0k0_Il>`**yKfAx@GsVig9!{89e^OzG6Tke+wH8(a4^OvzBIfvdk+gsP)j0-gR=kd+%AaWVhRnrLU##K z>pysa(-KokBuzsQHNqeV*tFbkL7FL+{S(x2qjpoS)IJiBjU2OtyN`O5;&k?UjjgOEK7Nh5}qLR7mP~DV(>QL7}EcG$(LTDK{~j8&^d=B;{8Wp4}OOU+BgO| zm>)s7ZTI(&9fO=dl5U6yv0wH*TsF4Fd0TRk^ZhA%Jj`gmaD09Os%)EfnN1&0V-O&M(5uWf_$d}>s~$!@6oBWmah5K4G7XDc#X@GfaN zpSV$yhZ!x9eddSaO)w5C_Dzy>arkuz zEk!^O;_!IJWzxeKIeSt(`x1UXg+MhDG)raH?6_k6yGC|o$_hqf1_Na_Rf zQ8?DGp3;dmaa}{=ijb~>?WUu~Ye@enN4&$pUEGt%mOFC($YJ@j9U>YApUpTS@6^om zds>#@k6D1Q|IdYb2P{a9y;dH_{4&6qa-+^|2tEw?)92C@0DZE1B2)QGkN@pzMBpc+ zN`!81eI%Dr^@Ps>F&iM;b*+aEr?IiVqaDnj0Dj(k8Rzfolrk&#;-BwP~{gvioJ7v==&X=LR@rmK7iH*ESf(ky`4AD z)74oBHu~rJrR%3bu-^vae1rId>M85`0&&s&dIZuy`{J4FP}0mL=A2kRp%cBM$KkfYN)DGU>Jfx2HU}LHnBoWbIs=^B?b})1XX$F0QFD8s?1%Fq99%J-dN_l3=qaiG`Ak+RLALBm z-wNHRYPV#*da++r42ahQ_Cs!ck5q>jwoTIJI-VD(Th;xjn?K477Qwe6S1`}Da@m{= zQs8K;GGoNd>3vl=n%A@d$~O=SAix@ty%#?#%%jB40VWbru+#`ew{QFwm$4fL`O(d5 zB7bP%!9d9TIHWJ}Rc6&yG`@&H5O0jRO2n-k5(wqoLdlv;#J&^yZ(#%&7QRC-9(&Re zZ8qf(E>y>M!HC7Pn^(A8uY5^-gzJa&cOn&rChNvzh+|1RuT-rd{rii>8tr)4kt3QR zq~y|{Rt#vSYAvZ+OVhnKVb__zrOVV>&Kw(r;Nsg*6#8-E+pXrr&%-5OZU1YnK!}h^ zAQ0-_fUe*s`(2ZWK|Qk!T#bEp>UfZk~>ZI1Btb>o~wC zIO*YNri5D2mc96tb5o~+LZa+>zA|b%B&vxU&`LW_9Hi0KJq^3 zO+-1XPXxymH-EYPXl><%^=MRZhNiz-@iULq6gH0T1R_)3%n=J1s0W$T$L1Ei+(0XX zlI`)e3%6u=KXfX#rdZDY#yI>yN%OMebVR7t%rrL@_VbXB!O!$uyA#)OOWO%Yt#PdA z+a&U2nJW*tI2-0x*B*8#c*Q@@75lJWr8Jv?p+u#k8E!m4-Zq6DGz5DVwglM%-$gRKUJhhP{DGt;m8aeZ*nwjoxgzB?G@;uXVLYIeJXCrJBf!d+*}cW~McYG>+Bti?awhH& zH?Ds99PRc9tYVK^Bc~xz!m3f{o`%YW<6!#T28_z7SYxh+O{wLN##eE zo?hqA82B(DQ_mo`gMGayWYQ-dS=@= ztB#k187$M&*uw%MaEDs<>&pI4_uHIQ_TR!`#T(`5Y0SqGq{`oa!5Jy9_w1@x9wGC3 z__0p^ixg`6FG^P9buLqcQ>UD}1x}?y1=Xu!A+H_o|6IdOuMHcMr;hh0`n5rr$|J5h zM?0FI@jI6=#|8^~-Db^m)PyJ@;BhB9q0$!Zu1Tuc<{q^BnL96;$)t~&1yDKd-M4FZ z4`-#|`$T(dis!+*KfP?GIATXfuk?Sp^Sk_`L<{y~4Ay@bo@?kV@D_&o4yP}f9SNc} zt}+f`B%kA!PJK+U$@SaW;kh=9X56}U)@}`wLep$s>66tKdn51Z-G%$PU1z_~w}QKJ zwby47BUH1hRgGD%9i6U@qD8ZL^K7xqXF9I;+qYtEM71X$c1fE4AaoVZR2QiH^5pr% z;g7=~*SbRrajKM*AAAeo?u{0ln5bkQ_5V@+W4r{;GQH>O?CTgVoBvQ* z_X?hbuHBCKMwcyCYSden6>Nw5y?&!A|Ew4%|A|hcgf3V9`NvyroBg4mQMul3vMf}X z24!J&2ahJuGy;TfnuMd$>>Y|V*Pp!fn<2)oG!_)O`j|4WN5%JvOjO{C?&y3+Y~JFD zE=8P`7MV-!R5d{ub|Kz^@$r5=VYZ&2rp1EYP~sW8i2z>9`fpoyMZEP8s>_JG=wVCe^TPZmkycmBu;fUGT*Q#Y;t$}IOQ4I8=r341O zED}WpyNQi0c1KMAmR`*5)7Qj|FeMa;Cs?C2gMQ8Mz#c*}gb;eowIheE$rSvk?o$!) z#es4Vo{+1ntErwm!eNo<{IKd#@bF0>_$mH<4LW@X)}Y7}R0}8A+%&J<-Pk>d2%>Pr zs>{ppDiYjWORJ&OfegIYk+{L zTtt5~t$>!04}a)EL@U|h87PHosEVJb#7J2#@Y;ZxQ!B|NaCpwG>|zG&)hHf%Aw{>~ z47W}-%W$xzUV#Dg-#CODP4u@P)zyLY4HG19qs|Myofc-Un@tfJQAD3zgGmPAFgBkQ zHh(q~J=Huqr2Y}yucT@j-so{O*?;yj+@dAWwpt76cza#jxM^KeKG4Ol8xWCIpq868 zCxHv8_EbLS%RV+T3Kgs0Sk&lx>xut^q)>`_e{|BHlf&)D2hgz3p_xxG40%|bcK_O( zp;_vMGG&s9qD?1>)OGOjnaUjJ8~e6eK928}5j|1$TvgDQAbY%pwO+07c?`BeS*)hmwW#?U_=Or{5C?!aQrcz_FXsAS@nM_Qd zIYW6l^gckd#;RixUESnft$M;14z(+)`OvZ%JBI{TJ>Wj3dP{ys&d~IhGt8Iz7YF*C zpi(Ai8NTLZL9JzpERB#jxSi=O?=bg5mDi76mw-qSI=O!k_cQ_xy*O}`Ly11)^VdvmX7s38^7zkM{ePUDWmHsO z{Pl+(ke2T5lJ3p{DN&@mLt45)I)_p^1O%i*DQTqz=>`EQl`etj%>OFaI;T4YZH^%;S$`sUG?)%fgL%Vn&>8 zgCFpFThu8eBZl8xHd^sexG*jVOKX&OQ4q+#%oC`fnnSreTL5N}fM7}?SPr@7F!MG@ z0!>ld&c7k$engIDNNFd&M1o%CUBei;Aoi8HHu*Q0sh7`_DNX*`{ete2Cx=V79f4a4 zx7F`&?!U<$Is=24|JIwcVygsy|LyzE?8aG?GL^J54hkSH^8`xGt?|ZEO8pX$OP$!vu5NS14Lr_b(+y0=sKIF7I+N zchRZh1#NhjO+|!8@+}#CKDN6&yi6daAN8qUzE9wngwE>Yqxwjw&3eliNSG9VrD7Ox zvHjG2S?A(huPq(%C0B6R#e9fhZVw1rfI9zbsRF)}B9-}_&->HMwoaPXWRp>!04MUy z=5}r`sv#iDw74qA&f0S7*9d_*uapMZ@}1A@^5|IODq#1Wh4#CDYnxrn|M3Cd;NpkR zB`|zaF+GEKiA@QawI#c}n1~O%a|SCYnP%ZPhV{l6Gk2EX4VOF$gFyyr#Z@HWKy`BQ z@J{R6IvCv2U}d@=r3Np)h?fTS{82jMRU_o$;zHc%L;XnlNHWUXql6D4&*cL?(pqHa z-zxf9Ns!yBZ_j*b5JwgiUPwLd2fWotaQouKcI|@^dcI{ttF+fL*LRpCn-Z7eW960B z=VMnjP>{gGV##1Xg_W(Fz7Ba^Ya>HC;P?6XlMP*@4#P*J-S>>O9gkrOM$( zyaqs8^M*(9Vbp8W41z+Fdg{>G9U`MIxyO7-xRpTloPHJY<&Z#gD091sa9wSD#&Jg6 z-7CAn8_=v&wPQDbQKt~4G>K{)g8-GC7GnOsUX!zjEk$5d= zm}a?EpSIa4Rf_weoheTi*rPg!Y+L}_n*yw{Rs2?QICF_aB^LkvE39zD6Lt~8wUxoJ zWLO@}IFe@bYkyFur)^VIdVro%KL={zETZ4Sp5^B;7Qk-Q)P3@M0wf5KvUeN1())9T zO;B&&00RW@ZjX+Q;b0F**4H$w?1+g~rG9rxRgz^U7^*+2*ql!oCmn+_sabZaVIg<6 z%f+Q4lSLOLQu0lV#wRDLLrdz!JmrvvAE}UO#{8h07|aSn??Vj}LlaR;sHam5u{jj> zpooi)#Z%oHz0Z99d-S!=)^P&OZ0D@<_x81T;9!(G0Rva-wu;0&+yo3&ysi>{Y7D`+ z`B*-l;f^KY%*W$#OtE{Ag|4us#Kbf{!SZZfENtziMR#V)&Q@Y!eHzA zx?qC`uXT()?Th$o-d380-9*BDOge9I2%~z0HQWQ|3|IT-xj}ivtgW?)Q=>T0kqkPw zTB{6go(!MWy+I*vGYoVqLzpToEOr@B0(W+mZwiVKLx7 zvDAP2R~onj1jI)Mz!$b!^nQZ*QJF3#D8}gpy|2GkUB54q2@qX?&!wpFfkb>JIstot z%4iuJWU|upTsasheS(9f1Jrrs$xt+)7IT@O!ZqkiQW~~^ZXW?kY5_we@|R4;a_x7W zYtRq_abSe_XNXWLUHB6z0cS-^$L+v|m;i7kfz|LL0a;<|#}h=iDv-~t|7`!FQ8`k6 z3T}6E2PsotSZZlJ?dxAx*edlnL<6tfCWPomW5(KSr84EynC9y&P}2HG?Hh10#I6vZ zC~l`py?yc?3UBIaWzaE`%BZ4UsR!yf5`pLK#X}O|f5U|8$AiIfnuexTbMl5 zs&0H@V(ZuU5|B%QA`83|uxIkGuS)_`6#BBJKeCQ4*mLf(0i`pJMGMZ0zVSb!xKoHVLa&16XJt^=}N9%K!R*ZM_+$&A>DQNUIULwx727S~eXm z4XBZ@SUEE@JP?{FqP)t(Y#qNh!EwO-COmJJ0&(?#Vx9e|UB=z{eH2pT>fx->tAUAs zqZ8J@`L%n#`)>O81LbZa=(p$9T*%#=a|%tiG6yi_z#4BD$aej-gxOZj{PV2 zvfX_vQDkxN*(`>{iX=joM}TANj$#`!n64Rx%lmAyOrjCmu+%uu+QYZ>g=q^AkGV>h zxqb}7qNuPOW)a!X*7jw9Snjr)_ZoqtMN8?@lj4EvBD#WR3rQ)+sQEJ@kMJlHL$ui`^TXKmVkmb@C5*2PT65ECPS!|Z!J+8QxC9! zc3lndZUS#ulR)^FUs>q&_f4}1ps*bVZnLoRWyN+BSZgBA!_`z)S0or`$vFv}?%_(e z*}K@HBC32p?{joqkG(?dA4;h@xRP5Y9`SaBP6S^xB4%u0H-Nq`+>(VuDZHL((Y6#*38b>z)V2*GP$Q$WA>{z+kgb6N4IsPsu}U9fF3EXB z+UN5AL8#k6({DTwtLzYC5Qia>ObB!{7(~S|oK6|Y_?ci$uSrM*>@{6+1|oeSZO9rL z39;s_rcLVDygp=xD6=4_Twpu01z1BYp6OL!JAVqzI_ff3$N%(VdGC2i6cds&v8Ks3F6eqcjRM^<#R?j3Z>! z{+h1zklJKjSXXJ4&H`=90vg`KYivVIM`@V)7gA2Kuea=4Vy&=#`pcQ`DUcQ0o@#)S zk*P{ck>J*iQ~$JV0<)w;$lf^d!qrWx{>OjPY!CV0d&Nrv>;CoC`NDF~sfTuy0`;W_ z<;dP%1}>A%Myh6Q5SMwgvz2M{WEFYe@UxJ68(S&iJ#pkfe+@5p8}<~Udy!P%M5XGG zSp#_=6NH`Q zizN9=P@nAkc zgn#J}Kq?J-M$;7W)o_9T2eZuO|C6A{8ma{lzO2SHa=x#EmeDcJF-Drql&S^(-w?j7 z1HwO0uust9v75A<9hlxg$HDfb{4To&IS6z}K{&gx|Jys{3xp~=raj`Wj+eWY5%qhj zj)sH7Ex1{TSs5n#oW2QF5jiTI*F#>eBAvt7@mIR0e4)HO8(U6xd!#H}Rh7F4AA&C& z9CVTwA^kBN@mpt+{^!@ammqaL*kbHy`9n5;(uFWs*(3GqGKPiEB5(gjho;K1P{2^s zQs{m#N5QVS1>9fXJ_hdW7d5HW$t8M zM;sc_e}^`aO#mrGVoqQzwD57n7ySfNaiUYj|Ak-XL$0fino^-cz~V(M!MSV6|G}zp zH9E1G9Ik}cT!&lv*xugjIcZ*7gimV}?E$S7kHHtunofHr9Ijs^mO^;k#Gi{XD>)&k zgsBiH9o?w&{~Q?gD#1%8AQQmY1KvJq?HZ0)Dt8~@$j4lFeW>vE?{C4kEm)-E3t;f9 z(sX^boj>}P6rgZ|3a@1@xqPPOeO;Ohy+1QowG)>G;P4rxySKy2MlY)WJ!SX&MRX83 z;WhHk2B_99*Yz6Y$wqH<-$rz{Wf%Ywyv~;+8KlNLfLvAKD=#ZwXyRCurdL-l6oq4n)tK?rcO{e3T{DQ4CyiHu`Z?T&y161+J4>68 zz|tX~h*4Q&k3k1>@mNxGT{M4IYwJsI{Jh8|W8l`8K412?eesW3wOyiB6-jF7ZQDn{>d%uuC^fLw8wk8Fe(l$Je3H!M=55w&{bxhDd+|Z(bK?2XZ90=&r)le* zfzHc)-ix&bX*#?M5g-|1D}Y_se6S7GO|Y1`bej#=MEF-u(#+N>Uhp5vbjfd9%IvU) zrH4oqK4!xFcRGRahagV0iiksABzj8Vz=fg za^$l+-?CH5mj(zwTR47O<+gmg{NRTtBhaaqMsu!-WRi%L;Ac}}Egi+T%TQPJ`O?-R zhj8}Er7-$bXWYsf(lgdP&7oPGkn&V6#pAAi^#P%@m59dBCTSJ5_LbkMYWbLaU)cc7 zt;L2Eb+1Ro-8k^CL7A>x>(WuGnmD79$0B%-;0X_$GL+>;q<%KwK?U5fh0P8l*y??# zq2#@0$c9jfM|2RT-s%OBgF7VqL`9BKpNT=?PiVV8S~hAPcgIjq+iloQJow1C1_ z(HMpQf)oJI+O?z;`1{LU6Rk{l?JgIp&_pIi@kYY8HPFT;u!sJgkYo2bsaYy_ljjfej!bYX zeEETqs6tV2II4_e^Yq~COe>fM|!=rvVJQHwnaEF{5Z zoU`2BS=sib43)v5Py1}>o=UkDuB=Cf_(+z6*-adI^bBj(ioHFPwmjn`q0kTdcKTq~ z9TsB|Pn-+Ai9XlEXgE?NlwJv@oLu`HIgKUdR<}{JSZ!WQP5rxNeR=t0Yp}p z>)=(ZEDI|K4f_1rAJd^%^p|BA2_jn=?0UrKAF-!SC_rJNNcJ{8540Q0L$_-MWh$ zHazgMW{P7#?WrtT!j^Ep(MNX8UkR*pop-Panb`TxZC94yIe;6!-LDyIv2v2mFC*Q-{L|=y7DldybTT3s#{eXM zBbgP~*7EyPHwnpAm5A^OLk^wd6jd~4LR5JN6YHf z9)ykpng*3uiHRUVC5j{=a|q8Q_xa03`dVzsKgz`zDgae_si|}nT&exW_S`=;RY_`0 zrY4bI2CC267Q~hpD;cyg*cXsyEr}vM67~*{-apGm2k%nB$!faJ`DL=KV|=xb{Q-N% zqIJg$y@wyXT#Pk|Rz^&7BLnN1TlK0Al5+IATYifPK! z=9`6+N;=jm5*fO{DA!W5=^0x^wr~a5VaVXo0c=^)MlgX78x#svQPuLcrloM7J+&7| z^*s%_p+6i{(@?=rP+U}}13N`|-pa{SLIDnNi{%p&ixm}a6sDpce)hI}1n9Cg#@d<7 zH#9z6rk-$bWA?+@$Or|dhT z;Zc|em8`?ONvubl=L|q~AgY(LO2HXg!x&D_X~sRJ0jg8Lrf86egE3h9-iv+(;@Kl_ z^d+bN_T?|Kobtvw&LaGo)d<1<9{gFtm|WPyxCL{pLl61PSh!Q)I+}(lVtY*`21p-Z z7mnr4d)}do1^lt&SbtlaA*YOyUwV6y)B3S@bXQLb$7B~Q4L&Lcj7T8I_fsoHbW0E9( z*YU=qh54~Y$H^pvtRvXka{@)i`%A>|Ho#0?T@A@30ZfJAryrkvhiUkd&I49YX@F~x zL2Z=6v?ZdrK@2!Q5t~?0E7QcPRr%QM!zi{o!(%pfdc9DXZqLGe(!8{y?aAn>a_z6n!G+80R}w()#o~AHVZ$Qn_neYy`)_T2O*H$md+pi>O#OKj?%UJ&If^}Y4rEgUU{;NX#GJ9gs#ykA-V_h-7+;&=QxH?$Y6jQC0fh!GEh3JcKCKGp@7?A{+pKYHIVoFkf# z{{iD(6HML7>TK$2boMEc5|HP8=h|1J^t*3nq-?&PP}{!sXXjgCN*ZcitpnR`FrsUv ze!`tKg5T@4%(W#sF9&)5&*Tk9GO)b{{OKg+?Yd&K=atuXw~ll0bjz0HT8!C znSkR5W!CIBvdG-X_NoCFa(JwGtVY}b4At@5yH)`|+oU9tuTk#H{^T3P_7@b|92%GW z5c)q}0N`;0ET0Eprxgfup4eg|WoO2v5)I`IxgPO7`Nj$?!(rgRKq3w{!XNQ`c`C9J z8PD?>3TB&*pvUh*I39n{0MSj9D6(pQ?hV#7tCDbj6toGM2g(21u#Sgl-@t7L!+CRR zeJW}ZFmuW4pXsxeo9zTjld+rY@w!fSvDB|M zoW23@!c7dI=mZQc6_V$L{*pYa6oeUdlEO8fj{Ur zN-0Ioyv2WS0#**-v+<(p0!@;`kK>k zLr`@6*c6>24dv~_M4jm}S+yNz++yb=%$Cn>joHnjV1#n03X)Dye4NAcQ(Y9h6ULz- zGACrg_#t?2!mVsC8Mp~5AvZ+&=I1|?&(%LkKX&mEDr{~kdC+}7av+{}e2A& z#RX%tA!*q-+{-=1l59gwBBgLEN;*lC7q%b3bj}WqFM)OvidH%`DNm>OO!W3jIq>8| z@>bd~jYcbLk-7=pFUW`2?;V+Y&%A%4ra`m62zIsI3i#?t#+lt+w*R%<#@~T?;;R-m ztPJ7XY=eDneumPqr8r{jM?}aGmNv!a6gHMf-r;mA;bw472v5h_OhYY=rW!1HDW?cw zV1hzWbz=0_373P9T}0K7WSQ9umI-GNT0j-`|7@5UQp7_^u#iK45b6-pO=9{CNVpxe zv~E)Gc9VPS5h&4n6%y*21+Jy zaQL|fk)r3kn3j=hsIm&*$OQZ;WF$PLcw#1$*hRhHd}t2EW%9x=DrT)K3lcW%zT0*4 zX`r5I3s+=Q<^&-lBGIGL9J5<5lOb#B*wQN!420XpYcJX6+J*{b2vCSt}mC4t?duNiy|`& z7+C^m_yuLv6oI+QU_@>ko?t7|h)>-JX6wu*iJg3|Kb3!r6^r4!yTc6F1mpTe*6-Pv z<<*9Q@BFT0A~Nj`TW{?%kxoa0HgZHLlokTd0`hNHnAJw_E%_C?S~!yg}9ru-LSIq)y9IC?0hnZ z0xvI=f@<+?1N2^Vjys1e;ALXEP}WROAD#AciD(!Rm0`ryN^O2O+!a|)s&v{7Dv{?S<83)(aOCMokmZnNIc2?ykTGQ$RFkPCq?*tpl}V9^LS-Rx3`Hoi^vZ=qbZ6hD6C-Fa z6ro5^I??Od2wb7bjO>J}Hs(O$`%j^FfDNL zo(*Fw)=}em=tcdKsz^F{x{^Vu+m^bpDYycdqvp|`ry4;bM_$OgOhzz(s2r7jFUCEJd17A0^b)zMcT-ixd~KPCNK<;p9sp^ zqxBz^h{8K!?(+Jsk_Q_r(CB&I1v1+fk7W-M4L3Mr-yvPA^^)-;3lA9m#mIX2b0tIb zAak6zAEByV#F$nU%C-PFZ=;g=dfFWC12_nDIKb$37|UziC7 z1_H-sXVVISmHYR!Tv`{Lxx`=FLq1I!f`rNCPc3!~p|C(C4BA~3h+PG(ZWAWrWeMX#5_jq_}tG*kcRX0JUd>zP4MsYPSgusI)7 zrlE2a*7VpTs=lCigd;il(wcqvR@B18T1FswwDwW=rBmc@fs=hfqb%5Y(r+XdE(wQC z>G<>B+pAHX!PaFrN5g~FBqr#nsxgh4rKYyeL9;PW!DfPO(c>=-Gvd@eqnR79p`QYb zt+0*g*n)wKW=^&Y?x*gqeoco&qijOxK8x?V&-&fD#5U1jSR)N(1}Ykg;Ec7*Ralcpc{7?AjlXU9v3j|SJO zisU&oNqdvz-=w=R;9b1ZeUCz~Bv8+TJ<SX$V>FBY2`M&Ej0S)e`DhC2FLu<$k3JBynrY7wpn=Plsq#AnD0_oTGA z5J>Sq%Uvi!8n(l-c>iJreWI;t7OV<^s&^r#M-Zysj9ZA*0#A6!W1k;TB z9GKVl)siw#8=WAR3hYr}@5W|xg98D3JSoiq4@Sks31H9!kCG);RRhT{Jsl$N#`2{T zCJx@r?p2@t_yM2k21T;KBZ5(lx_sZg;;K-&x^Tr{zl4}KXTWSKE{23#K}3#@uS%Zn zmC`w+wB#0nv;-IiuzCkW_>%8mR~-p@M?HWV58*>`8ng8rXzrP&@Am#^KA-ZkQc&B} zcLOc(w`T!@F_G?!KSC!*&U{D|{w0A*QRjwm3P(R{K?g+uA`#`g2ss_b08Jn$ zV6x4@E$Bo40Pp?Z70I{n)Vt9~b`c3W5Q6gs+61JOn!+ajx+Kja5O{~dRQqV@`ZE5N z*r%PhPJeGfx3<(A@Emwgogg^A_gBdeWZ;+~8gEz>Z!|ScjXnc3#$Pbip8J@$)5%$5 z94%*$$&C#4D)Nq0Y;pe-jJD4pPiF}s6uq8up;4Ab-Q)wsD{dVD_*y>-lnFWPP0$Jp zZ7_am(&2{Iw%nuXkD4V(HRrL#lS9;h7|AQ3Q^;aGDvsd&eTTK!yp^)ZB$mo7s|J0p z^lW0I{shHW=MI&0U+TrG=N57352O#4gzjY#iH;QdFKP!KDE~<*mWL%}D~atPf9_Ur ze}6T$_HPVKg5PU`kq_cclY-JH3oVJc&#whANNa=;9uLHY@dbPQLk|=R83Ply&ro`x zO8o`+qEp`^Co?B+Z4ksIQu?%Kj0>jxjLDMWGFK5Yb$^I)toShL|4WvEBHWhb(tH#4 zy0t~Xaz%s^5tLf|-ZM2h^W0`h;5Y$_)_;9PUe$)ZWXbM_Z_V>bbrwO+WF zZYbjT2qeWoPkKl!aQse)>-T}@5r~8070x&=1iRGKc{F`x`1-;Jc)tM=76FNw0K99& zA^QiLX*T4k|0msnU}qyvnOTH|Xcm~yWSSf4 zO+!d*IxkoL^7ry$RAd~%2viw^`8OubCNbuw7c}Ug9rbo^fz$1w>(yq(FFILlBs8hF zSyY*$<{mrdx$RHhgA?|4&9XyIq|73!D^yuR!`nY)ZnhNf>uT zY&SPzu46}N((rN=tCp#07D0vr80mXlz{drt>@Ogb(9n7=+X?2MAT4kdpS{I4G99*; z3;a+(34zj?u;}vazS1miHh3fh&_S&EzQ_nzG5*lr?c)XM)hKsU@0sUU+h+GR2{`RLK!ntH*I(?N&(LO1SM+S&i*r$fP%0{RIJ+eZK)5h$<`k6h!@7L&-}{ z^0|&*o$``M+6MJOE|V-erueI?{*W6i@0wI^0YA)Z7Xog8r^ycGNHD9M4dZzNIE)#* zOI7w7fiKZA6(uz2t-1EuBQQsL-8!j|LN{Tkn)9TE9rY&N>I<1fs7@=$vy+%l2zbJQQ2V8WV;Fg9F^ z*{MemByqXNI#l$ygp%HyGRL{np?SUr7w!s-=KGv5WXYwK_}%SbHs(Smx- za#b_o_wP>2ILPK`f z7cC39o5{!uX7^e)`cg8KE`lOXTHn4DBSdH>ocKks(73{=FV6*PduQDqr$HyfqN%1P zpHIqG6VS_c^2d5`h2>|ZFbjr$#7?Qe5{{&NLSC@n!A*e#RK%wmg_%DH_EGT{J1*Tk z_)92(B0vvA6^4sIhDX=2m%i_%Kd{W8eq0L1X^9zwra>^#DQ6Lu3Eq6M(uyap}B5v*uoJBsB&T4DCM&_eD|Em1~s5h#|8e zEacp;;BoGoS-LFbm@)_rW4tkxJ%CIig%^f9Rn_)LO!!M}PL3*)=_};_J4JweV0}(@ z;JvcTzqM+6KZE+dA2po01H!2X;U@N~s4r^5%*x_bVIq3Fy=9-ZTNE`Ye+2PT+c2SY z?QcFG(ZLewC7VK#Ohkf09(xeZ9eu`aPyVe~F8IcEUcN>MrwB!G_^HM(ep90WW}l|l zviq@N0?6Pt1A$NCGlbz|hSSNJz};6tNe7#9Nk4VK*|+Y0yM74W%5$y zC%U|JqaNP^zTKb$U5oAHng=rlF@zN2)y)2+hL@Zc1<`lZj8u9 zbl%v^32~+xfsx7`gqVWJ8cdYODs%s6f+emC>UdX^%hMnFt@&d)hqEkLvSl%uxfC%S zwY52d>fhOXAw{_!ec7*CI4MtxNrYTeh#Xe&VsPiew&_Q_tz7zVro{6L?!2-yU@g zIa)X35JI2u37Z$ykJ`(9Rb6qcoG~=wZg&0iy7q%H!z%4}Gus#5HZw=AKq1Hu}c3p++yajj(28yLT8gYVb?7z zM0?^8Tgn-`2S(dzYl0u|A;6NtyGYzV@C(v$H1kARi*pjd5sS=`Z@-64Ut7Fw=D}Xu4xg;FJ;Nm%c=TDQLEE6uL5K zV?8@o!O)ncCR$7`o zvv=Lwd3`eCB5#yx6?PvnVXTBC?H9TXUax55V`p8REUmCXrVbkir|d|A7ezd_I0m?S zb>lWREC8#6J!`~4g2Qp?B&bpk@Z3Ft5(elJ{tWIqlmJ|0pXbubbw_KwLNdUrN}1)U z1n)%@1Z9hHG6|nGFN$bS+V}H>(S)+HF|&-C~^X?+&Ez zE@}PL=&aY=mNOro1K&!i%kwj7tGLz>V(VMyJq-6;>XgrE*%R3<_A#O9J^GPvp&O=A z%xIheV!$2^mfbPM722XZmx*JX0#mYao6m+Zg`8F(^m#5kWa-rJct*V+7O&~{(gWo( z^$DOU88ttjqL%8q7C2u^t$qJ_x$XQl4ZU(JQ2GRg#o7*4w%)>_X6ad*fZ41#nlF3G z1v*b++D@?Yo7vxPf(s}MWOi^^4wvv037F977Ul3h1RCr*2)_dW+j`rvw=E4(t3iu~ z>NPK{JUsXl)o*N48IX>GG{*B%E+!w<H}|B)+%B z-4O4a@8(Ib%ogM9_-xZ8Os_ER9S)IvT4;7L?}+O22K>>>P$Zuh!!V-;1ZcMT=)QEa znnSAS1}#q`Os6knQ+KE*AzvJqI2y+7hi+S0oxN)^#$GV#by&>a{j`=mdpY_;5VKD8P@ zC1$30VR(Rz{4MKjqnLjzxdg;8-6r;*vpgz$mg$?H#qqUsSjjs3rxIut`edrv|8=ZH z?{#^hd4k%+?BAf*rm4-EEVWE<4_v{$TD47Ffs6>lc%VQy7{PrU6C{W}l7ePf*;&`p znk-mZ)0mHywKbEFo6_ZQ`D^C-!{|_`!dG~$=5`UKF8iP;Zy}p|Sz18PN6J}oi&}Ui zI&;!E?sTjd9j29rt)QxZ0R6FJULL=3r0!j(q|(o6_HB!!*tmE9GH$me*x!u*IgvMY z+Kn|;6(^1RS$Wz%u0Bjx+dtenX*zq|K)6u{N0wgOsrHVx`CiBqb4j~6O%rWl#%L-1 zRW_g0R4r}Hlqi+7?b@yLM&Qfgq4Nj6>#lcx_+blGHE41v{*pCgi`7p$@fOn0)Qpu} z`W2pC9>AYtCiJ+}Pnrx?TQ^7=thY&O$E84QzNo)lmO7@s-&Aw?%)wHGkFt6#^v7XXp;D34#Gdh?Vr;~6u##VNaxJ!AW@H6!* zc@f*htaud4vXt7UZH;0qJA+zoqwGz3bIxOxn;}bvpb0}(-qHt5rE8Sk!l^}awd>$k zy%=X*j(0V`cC}MGV!6_9Mn)NU_z4b< z9TLybilP|q8=bKFWd@Sa-x)8){0Qp4+7x90sj*k9Y%kTgoZ?ES)i-veFgqzsO@FD9 zVe)1eM|v67P^OSmX9t+=WR$z5Y`@CV9(Hy}eo-7%}36-TUC*r)QGuE#4152GUaH^P_iXpg;))Hv$j ze4JgFd^u3IPZ507VYYj?Pwrf0syg!n6AF1Gj#O4ZBvK&2p*F_&9Jz|iwged^HV*Uo zW9k4&5}6*;#)2W1f}LktA#uu6=I{B{N*M@=pW$Y$qOiMYOZxNgoH>;+JhR3uNT9z< zv5TX6KK#wDNZ-gskitPmBh6V%ffqh(?Vl!UC9)<%LRrQi1snWf>Nm{D|8!V)*p59# zobSDfL8fT@@Svd`S6r7x!aI|_ZTpwFrOTyH<2G22+V%Zya^Ou*Q!CYYK7L%LOO`Y`U zpZz-Z8@W7Zfa_MiQQb^y*pGTLZocO{k!E&TR!q+3IM~-Hy;!p;)5-Vb1!uyPZ;@q- z^EMHRC;ju_+PPz6Eas1PDRT)4xn15RiB7yd(kF5M_>=Z}C%S97{sr#}5*ezUm?vYa-5^r3B@rjYh$&u(9 z#x48OtZiJ?GV8SwhabJe-eFbV!|wORP6bx0dUKZ@FGCF`+?o|%u5&j2?g~kI$_FXl zcl5+4E|tS))~Ws%L6`U0vkTb4wZe`y(-CkFET77st4~uO^eL|0RTp8k|1kMyBLbqE zHDBebOQhf1X|&{4=!>R4zP=6kmbqQ}@<4QeZ5!Rwi5tM!cK~3(!o#ECRG98GE=Crv z*#W==2!4~>QS+P+I+8I}{%*4Yyz9^7XnGJPN0ga4+u^S(B9%8kTLfVhMAH+(g!22h zk7Bk%4KUPIZr^~qbU=^NggahhbEEFz`;{gD?rk5yVP^{ucgT)1ZW9hY$$&+9AIu&% zWOXJuHZRV8FMYAIZ|=8U20(Sf;sGosy->OK;=1{2v+jJe+h#LRLd3z%IpN8R zbNS%`)Z!pYp#hG2y5;Ehca-E_)e49=0=4E9sG^b_OMjB{riE=8^g4qieA%)r&-pAT zb{lz2YsWVC;aWGXn*;ow&El2NS`Q26(`h<;vT>?lv&sNR6K2koN{+o@uEk7@AV2wDV& zZg!lkAl~wI1KYM2q68Y@p#c`2l=st%t4TFXY9QqH?@a4K5Y%DW-c)Bq>Mjh&QA_yL z*-hkob^OVC!Ttwi!QKGaH%p$}uje5nVNE&BPIDgMlhy$61x-Bu&7@_yNn!#52)5i1 zNayV=XMxB36i9>(0&4tp@WO~Hj8Nn;z-!HS1xXl6l)Prvt@h4EE_D~8NNm!Mpfd*r z-OIjPi3sZtH9Zxfrjy-FyxlH9$L}Iw4W!0yaw-ZD& zhf%G8^l3Z&HuCT78P1Pw2v6c$k59`isymBenG{ZWn{)PpSB)~a0yKU<2{z~1ai{Y8 zJyV8GbLkB#^{#H_y2k-)e$iGVlm8V$BD`AOm7d!n*GRjl9CAgt5jQ(d$4g#Rl^tob z&%ACey7mA~=WZZE_uj>9GyR_Rt}s2|_op|nK~4%-m`aZn7fXIT6=`u^=v}*8S#zH? zuq%QG&}*=ZG%~M#V2F;6p2{I~%9CRjz*gD-Ujx5~+qHrnyhXcfxNhKiYJ45pTeXPE zre&v=@{A|u$M&)>6?80L@sEWBY@Gf}9_s#f>XffA#2m~LNM75%-rV7FH?lCh7)Ofs z((G}7Iq_9&^wu{0+&6|JS?Na>>$aR9zCn%7@a<}))7RR6US8b@to<`8RWFVyyN@Y>N?NPdM?Srz z@T$hf$0sv6_^jj~LfG67!$od&uC3r{ngt5_D(KBaBp_3ney@ZV-;&1tIAYE+?xnr^ zP76?LJ3kmXth~UtXFT^+$nBCOa25ihUU{4|2!@J1>&6ET(g{21Q*yxL`%@->;=UX( z)~XwKIxdln7xYWPd9jH-hWGbd)v~$=_Q9UmSDFYfaD=M%>Es*^H-?$ z%`O@=0P+dgPCgMAUnFNz;!t2O54`R=9lU7BETdPI)J^q_)xuXACQ~TWRU>vP)pKuR-tmo^aN+hZ$^xMu`9emr8Xjt$Rv24%}G(+vDo& zTUSmQ@?-^I%`)tti-V2C-iVRz>K*+3@$qq!jM|FYOEzK2xL^7_h7z^9uiyfxn=E6c zu|gzg-^#2$rF0##qp9^0NcEJzUK?#?dA0j>(n5JtD7PSYB7irDcS&W!F3&^>AI_K0 zEH3@R5AxwBcG#B0ztOBf6Ra0b!^&aD_IgB(862-FI+F@2@`h-oCBHF#f2P5ov}E_o zH{1B?ngOf%5Xx_6ePIuxl&Z~!Hs|{@m4PI2`4%SBZG{F z#@5SA@Tpa;m*GLexPNEif(be@0TeqKXSyIA(eQPb%=tFP4<^zz)-!IjtNMv^pfjW zuxsM@fxbQ!CG^KpR8I%5dOto4rmuR!H!V!1k72={(-NYs!Sv0#)NA_oH6#PmUG{8B zE47VXpOX$+I-}>Zs|=RrWi6!$g(4aVGA8?y@C3`z1oyQfjO(u@BJP*_E6&+9$KTZNIeBaDn7l)X7i+Dm z2l`5tWy?>vo~^k|nDy_u`MKWbz{1J}2OVJ_`Y#4wxOVM(N)kwaD=+vbDbm`dnxG|{ zGyli|&hWwFkBvFvc=-w?IGsf)|GOYjE#^%&AqQv-*wts`Vf}<}BWX1rV!DGiM8GV-e+ER>oHm&%Lkh6 z^oe0Vha59X$SgQq^wF$Mg8vR2eVsgNfZu{8RAe{Rj!-c=OIc@2kh61CAF{Yk-4xg8(h*ebs|{?o8&1h(iiH~ z2r2&;O=lff#rJk?kQ6EDlI||)Mi36tNGhGu-67q1C@Eh1I^#{Sn;k0=0Et)6tR$1-in@L=;;vA1)it4D!nh+JJSmQhI??o={;N4 zGN`M8vQ3(L|I#-i2(&Rlol~8T$7pM5GhJC2E7l$ODT^@kc88L>K=9PIORO@XEcO zW-}KijG=y^bUyqn zB@L6t+87!=5q}aHIzxvpp9Y^3J!uCs>+}0Jk(kte)qWV;wIgd|Cw9iRzk8?~ShBZY zB1$%io&aPFkSljC`d=hiN5#PAh{+1oO3FD1Z8twKHzpaKIx%vXL~W=Og)(fkgK`LP z&j+n$F|t8mv?W$-8NCM`X26{(q2(wwYf&A1r9!t*>n42=a-qZTsju1)LV2D_aT+sF zx5zH#S3g~3q0z6y8WmF@4KCa7_W?8d_Z$^DL?+?X?%LVG256x0+^gfuneZHWJ*pMO zwF{e!4m+}f`X`3ZY9mCbU>HUt8oOSb7sc62i~!Cg3L)Ejn*L&G#ENat9eB2T65Han zy<0yo3AKt!28D;~>8AVq!qvg_XIW;Spq?qNa`9XzQ-C{8>_{Rs*(L+;wnTmQYcu1BdxeV%KednyI6uC;r>vEG-( zK)p#XWyN$m6!qMiO;(4R7yf0A=YvA#Yae=h{mUpuJx2wzGF>-eiH$9r$v;?bKH*C= z-X6)mbdL{Bk4aiL;xTGsnhY+4*O7==LB_xKHc3(1hW-8iDW{Lu&}Qe7ycf)eMJTu3 zR5W(4j4;v=O=b?mCD6na4AhMI*ik4sWgsaVSW8od%FEp1yVJe_AR>kHqzV3+GX`Q)oGPHGLE(0C% zC-taY7gQ`k>FMdUhlKn72JEQ|cv=o9Oaq@4TLFn}4I?lBctq!Y*- zPIzuuA)Z)0?@QIyHWtg_VC0(7Ed{b-1!aEgl?gn9D&1=bjvZW=zVsEp$mtC7l4VL19N zSBnp$o97D0LEO_(s5aV_y#_|ds+MkK^PRe7A;Rnos%gkadt^m)AutYwm%EkW zK3?5l@D9VY!=2dZbuH9=;yt-*-D+@+_g_E+?fv-`rcURT!!(#rTR0GzMn|~uhlF$7 zjx*i8IyT6eChjg1HLE%ZrNqjw6$WL-=`UT2YH$aX+c3MMCSH_iSt+RT-G}Ef#wf4J z;nYJvN!X*SRQI(CewQNRy~-%5IQrm|mXhA1tee)e-^L*G%n&DW_T=cS{U{d2Kdh3# zQQunFUVL|@YTf{YPTFutdoZ2{Y5YP#3X!&I6QTLRo}6MSW{RND*yoi|E!~-Po5zNE z+ud2Jz^Axe+rv6$B4a4hLZEP8Tc3f*sa4{*!1-?)IXv>0V(=W%BE1H!NAhjgcQZ}9lfwpqR{?+&{Hwl$w`@MTM>GqE#Lnu zge%=$i)qTO+N%CVwHYd;(GTpQfDz2MwTPO7c7zlC`OkRTffOk>{k2uhpYi53@`?vf*aT3zqRgdn*nX{9K1rR_!E2;z zoNiJtk1~%VfRt!JEZ|H+a}^DYh44xXcH}iDkx~*IZ_AYX+ZzL5_Oq z6xe28wdIWX4V(QEn_?}cqQqzE(ngcB<~O4cY&=bjO!Y1Z@KzjS!z--bu~MCXqjDj2 zg1SHP+0#td&j6iP%zA0E(1qKozc=$6uih?nNa+pH21h^Qtq;C@$xo#dGrGs^gr@x# z<#}anM5`}^QFfz)sp+sewStioCaGPUnEZlzo{A+ooXdkSpHLfOEg(KF_Z<7yw=f#y zAViLR{9)w_y~PxlVEK|;!s|smG?o}}Z22Ae@>Nrg=vaPPq;<=u^>e7}t5S4G&Xv9* zxh)1s)ccn_?|5*3sYK$psWO)S*389;V|xA5$T?VE7^$v)5g~SgOiY&P9bDKK6k?87 zQVj5t8Zvq@bR7yJ(U`;>iZu*AZ5FznGUaf#xv~*#Qy68j(>!{C5F3@Ez-ki3X)R%n z1me6i2*n4hdnu;kGnP&0+l0pPL*dh8W&fHvT6F0$*J?;_x!*NlC;*Qibg);5sba+A z6Ke5+$)$dF2Ogpe?w$A;Kzt7d!+YZ%BG}Lt3@LcNb zt@yLfQ`)-?IeP`gXhGV|UQBy8g-l*lfp|b*Hob1dCDM5R)~;khjd=!QCiK0a0*ou! z2*+NYrnXpAj%zoi2%wxmV(`ABk~5vzY3VxzS-+t!+sY%#Vg+KVl2uDGWHVUyOFN?V zf(2Xw*EVUtF}0(@I=*DQGDyqsq*N>7T#`#X)Q>1tVV_v34L+yf+ADY!j-7l#y_v3D zIz1tD4&jK7Y1 z(aR@_sEtexV(bDqzwvL5>!pdEm2|)Va0{gF~49sf#q7 zoL>F4Ov);*!SGlu8!@IXK-02R2Vbkv^e;|JWpFitnMkVJgfG4F&G%o_+qdc##N~+H z;i)bNUajXWXyIx-XsxjHAR2Ko`Lw6=_n9=R+;v272c>l_>6R`AN4BdR zOeM{JxZXDT6=ouQ=P_{ltY`AWbt^c*U5&td5hZnm){zB&<2uVs*JVI}DW~wR@D&@< zwZPUFCF`2hbfl7;La8$b>yH%i+&g85cXUSG9?+Y={39GFLtm&%G#sA2NkIVBn6WvJ z*Etq8Hf1!yG#o84j`lGug->!5n9gO@a}Zet1a_}j@-|#9doTFyhN_Klr@o?y=6Wg* ziu(Q)Es@}wROX=k5mffX1MAvwt!}x-SRLO&)!T#A8*CXzK+4z|;WoxQ8#?sC4!Ho; zP(Js&1{S8Kh$fkrZ=yx|gQ2LH#EYOvy&GUL#zo}C2ym?zQsIFv_mND*gVVOpaZlfC zG}VHi-2cw8%$A=y_p##=v_LuzvDFuN0+m#=)K%(sZzGy#;sB{aw9!EABo5x0t8QE*xt|78gpYztc}g=2M?RDY$O47a2H_X!-S z3g%FY$!&$>Mkm(;(`=^-cTk-e!LS~1vW@(as4F`K!~2U^*ysweOGvp;2y{p^&UT|jZbu} zE&e{qG!xy_Ewna?zwEOb(V0(J%X#c)K=;%29~AL45^8SB{(Ei}IjJfvz52uu^t@UPQCt1B54d0h2R zBc)VEhON!t5@lkURqRXD%*Qu>k{53g80gzsmXB{A&Zqr*mIn~XAO~-Eu?vlLC5fX( zcV+!ktYs^@CgKfm=Lq2#MJ%vADq&vik6U|Mj)yNUxjtrh8XqTe_v((+Z6Ei>RBUFR zG)fdf-D%)d&_+L%A+3|LpVd3K-g6(s9tO?QjM+{9Dax?>hmP5>eYj8`2N*rh05MU- ztz2TmkOUYSUv$fRIqteRj)3>`^W5B=3%#x%2UuT60Irz% zg_RND-vnBdj5UAng>H~-xpVLnMRn0-LJ{B~;wT*!YAzu8*r8NlYrFGUsK%7M&4Ogj z2AX6ywleIuUKm_Seb?qEiRlu72v6(s2nx>eX|f`5vtwalo%Zs#CCmVSr)~Q=r1-C3 z^^c{~l<7t)v`4@?1v>aDMYZ4k4Upu`P~{e|z&q2};N?jGfv&*qhTJ0K_ra@o1|sujAn;iX27|0!IQcqCk(uGF8vet>;kvCtL60dix^@*dol$Bi-LK z83WVgsPis!`7m;k-!peOB4^{)ZlNMFCi6Rl#j!hYG4CUmyG0%QKfk7Bqp@^}Z<S%K7Sn>>m5<>*g{N-JT_m#o?M&qN>}@aBHm|Ncnvz#Pbb$y1C^;_GssR6_V1HSEwIhZHNUeWB zi-7MsPSb#xqgxqmnX3{MR9)wAiVZ3?T(bS2sY_xhxM zjwbhnTX1DNaUe z@RQZhE}1{csVkrK6x*8&|Nit~druhWy=nqI)vFKDv0;s8Oyf)n-lR7ptsJu}Fb|XX z%ulHJsv}?;``Ksy?(ps7<#I4~I2K46V9f|pordzODu&AYa1P@9+B91wgD^#-)FDlTsx^>il~wj_8H=sqi;QC*#cwo_e;Y+3pw;93*51E4s7+RpRe+}RuxKUkd5yM< zf?oY$Q^MixX@V+Fn?b|K?KX~Vnzg>>v+LRsx+R&*QMx@JZ8?T;859z-i zsS^Q;tuH^Q2}m9^D$7FuQ^m?`l<=141L*<^Le8n(=;BU}Df}X}+PBQmc_20zes+o@ zYM5|?HDp%gc{{@@G5eCJA#&qE;cnk4Z_ms7$Is0ShsHuSMz}E$ym{?tZd&J$W#Vt4 zYL-ZTHglp6oZT-=_c$w-uOLthPW@+Yc3&#bI~f^`j&Z)}R{BXjmtG?I-Ir5L$r(w4 z#!unsWQm8rYXS>9e%;GYJD+&PR$G~n2TB5aI zq)gJatUm~vM2Ko1mcPZIZ~E`$djuW*^V65a*mcqF@_`z4myD6(kMPaH-9Ie9prf3&V&Y3vqGy#jl!g}RbqibMTim|b)2U0oy?i}>Ca!BW6b!VccT_LeUu zN)fLyNoX_8(*)g!2uJ3?Wq71E>(^^SDRFvqsrttv3{L%ynkio&r^}V&h_~W#F?t$G zPN`XfMv}xx&$ZFN+wP3SQOq}8BAztIxlMz29KCW)uN<2fS%*AJ@hheo*48!q9)$*O z&4_bR1{=|27nA92A^K9~nhLkATnCwn`g%NQ*Hs!f@%=q{EU_lK;IE6~$n?!@k0ZKI zTRZk?3`#WS2f1R``2=S$#_&z`{)%KByYy<6PN_BO0Xb{Xi>r!HxJAv{T|XMSzOk|K z+^qdWDhSey*D#;>TG0U~>BDRJMWet?C&B#}S?5{pv^mv}{>5TP#)di;h-JgE`ze^B z!onE`?)omzGGIzom3+b&%9zl~i#KK!xYw;KzC#g=1`rs9g>AI3yx?Q=;^FPX^FUjs zojp{wi)+kDPpn_Iq6Idqszz_bvS5FwS;F1}JlSB6ejUF?S^qisFW+Du20$l)-W zaQ}tQr}2XlO1jAs>lt#!F!>~+ysnQUYqsAI}4*cE341keGF6WnviRGvBHSwW_7(t=>LF{fnn9&nlK5 zWXgHVL}N88vk8NSOR(~Ujqk+!LCiCo2s4q-1Rbmt z3=X->X`xO^#{qX>q>(u|>(`gNp~X)X+9RQ+P0zhMP#^<;))XkvW8>8PR7d?o(Q(5_-N78)z_!b|%4F$|=@hE{x1! zJo9W+ee*OK4tHt5hrFG=tPV61wb(h}w(M@+e3|q=%#YrV{=uI5OyjM*+rME9*p|xIY&%| zCGx6^bnrVde`>rphWJz%_ZsnW3PSCZ3Vtf_ovIJ)vLwx-(rbwql>1B-cT#-eatlfA zz<(M_1eb{QoW=jNooQ2qAE!d3RO^+brat&h3fBm!Afc&YKPh!;s7u|el_^$Nj%O{?wf8gI~zAkk~6^ZpS~;j6uKK+L1~De}E^6lQA_rxyw;8de4NOCP*qKv9DDP!YHa+KdcEGo+3)D!nV8+?jaVDd} z5!f`MWTmN-;-GRbQbKd7{{CN=qQ=+m#mlG*)tPH z?n2dNp%QJm;p0@hzr3EmE#xj#XRQT6P_&)9>VOG+D;LjBy3kl&(`)@f6T#r1>GW|H zD}v9I&sYo$u5Y)3L}+p@@9eqZm#-+Z_^Pm8d3J&;`t7qEW8V?F@HhYUT<6KM|Ty72iZ)#Jo8t4?b87 z3fK2C+e;;Odigbx0Ei%+>>w-^J)Pwgv%*JHYr4wyuTQic{Cn*)m{6pXnh~vbNY!)o z0X&op6k+?Ds*XI)J`2Qm17FJn$wT29;M+2|f3rk0w&+?6*7lPmFvx((KY))~TYTHX zAJumpL9t$%odq|3E}TrCe>0D7au1-Zy_pR~`q*=H?L*s)gDF+)^ojK4S;w2Vb`i;H zNNl2XXcwMH^$qK#4YczW~E|^5S3c{UCtbcuojwPUw1EfM=pDMoJ znuzydAly_Z*`1vwYIZ*aq;}YI3X^K4A@vX!#*(*Ne~Xa9041R5X=#5z|FB)n4V86{Q#&f(z~%S%rB7HIVCAy} zIV{h@yeIgJ2mtu)Nbt=_ntFJo)r<&5f!oiO&`>_4BHmx>H34$~X9d{; zqEINLkk|hD&aL{%`$eqc7r@r!pX-s`2ZFPS)6Lt0ad-lam6aabgmuQx*tMOXTTJpj zBPkmc9W*ZMWR4>A>K59vbVAQ~qY*^%G?n*YM5>BZArQg)TJ`sgiy*|p$fOc;U3!Ao z)9n)$L69nW+iUIQrAK2~y>OLx877RSD%ujqz=r6${Dqs3FNImth^USUQN~vGW)v97 zz|#;&7-~TGm2M=oo-SJj$z6b145 zdu4%yxVqg?gL$;kN}I2h-`Y|YqP_&=iG?VTu_Kz`UE)j;H-$={@&@Yt}klgrs=u@a@JZe zsr?2>i0?Nn?P^M;hVlwRQJ z!Ae;0Bm6w6jLpt>-T8}X`zdk)tc8VLYU8x7Srw;Axq3u}nOen~rN_noH^rIm%br%= z*hWl_BuT{29QTUYW4L^-eo~l&l02*1q!Yr<9DYO*j)m8M031jh7(_8QozfYdgeN8Z zk#6%T?iuTs-1IWIDR*{zDjshtP66a8-K$lJs0lVqZ4-se8325mU~>Pbg%~0@>e&u& zGny^hvMqXOWNuEcXi+)^Ne-k2!51)H{i$tkHZ zik&-iyG$eg<3skA&o$?|=wh05j5r;AUs20@vDoTkR|d7*mbymL-wNthO@0xpAa4pR z9cfCl-5uQy9N(P0qY7Rp%78Xnee#lS_1%+1ljIJjhr(a_ck(8PYX z_5K-JU~s%6KsJJ!UCqbC3hRuK~GIHbp3WeQ7~`!Y8^Q{JIf;#N~9=%4Uj?do3;HH`p3j3 zfXje;=sD``(I>nNXwo1D1Ka1EiRcFFVe|<<75r2%=WtS+iqnW)JIzai@l@{qRJ>kb zG66OOqjy=Kd{zW0v4AK&If(}`t2tqLK6)CS7ye|715P>Ll6*)eyiC7<)8F4zVwa7h zjtjQ+Z0SbLAMmVMis7sn*IvhmF5B?U0fdyY$nl%Es9~s|UfNIx3~f3!#mw$RqdNAd zjP?KOd%CFH_O!hYA+jebXYOrs_d^EOOK!=ao?!Ef^$&Ufq!%ADjlXTIivz)PRP>Te z*lo{}<&v`BE%3Kc1~s9qtnAZg)zbGgyUFTiI3|4JKm)GTjL#Wbz1zO-kG7kCXODNS ztzEA9sef>R*6aB`2($aOYRvA>f4!L3j`;zr7AI(9o&6? z@$V|xv9q7A?uryOF@Q)qTG`O2k|bo)(kH&Jm}0&C;fTW{vh|%-hNX2n2HvBudFG*r zBgMHzXoK6CuD;qr_tZ~99-PUNt5WU4P{v)fY<6`eso1|1M(b6lEf_c9&;FoDS zEb=@U_v$LB45m(bR9HX_5cV(Ke0vT=jpj> zhU=Eu4YMKv11~`%N^jLO0LVp|?@OYl^?wqYfU#}okz4i*P1vCO`s^3aXu*oSt`>7* z-N>;M-B;+s4v2zR0VMfZv!5##y8%$648fmtJ6>Q8M};hlZsAQ-a0?~K&Rk#ibPc5R zK}<@>r%BAG7?0bF|)U`+EM&<9VRIxz`7d;k~H4oJU)xNZ4syDd^l4Ii1gjMJ!MuWf#mBC$G6w(kLL_z&tthZU#rgN z87ruxSlY_pF`1{Eiima8yv?Q*Gmn4-Ohaj8=DZloyui_7!%(|4Gf6Ye{0*mAuIXWxZJKHw7x^;uSt@%^HCme>hi+`*ke}RyO_xk z|EGaeR{!O=DpxarOpM*4_cyAYIb=$i zvm=a%@qZUm!IxHPquUi7mJImreepntJRh{wV&E8;^nQ}A2$KHK&xOP|We~wPpU0|A znkyRjNyEZN{VY+n6H2KmXIXqED+$4aKOOH;$Ge zj=Sx1uTJVKAqpCMt%7{nb@V?GtEc0CMn)uTTDxn-f}zPSoYY%Arkc;k(_GnUzR`~! zhZI$*gaX>avSrk@Fb$gU7VI(XV~m1{x7Pp#5NpKc4^l6Z0p8txGN zPaIR}_@F-$702;bi_cUl4^5)dfA7jq{jYyma#%CHm>U%s1y2nx zp&tcWgUS~|DP9Tpf#t3!)4<*gL?q=|G{vrdg3RlEX5SSHAoGJa^|NNB6IVJSlW~Md zO*E%)q6S`t=EU*|p&_BTTHz90L5C8(dp5lCrRPPu77NdV6jm19?sbTkkJW9cIuML2 zWQ?gezfex5t@jj^Bbik(R?_jmtdxKXwdKHum0qAmzm4v5&wj+5LAPDEkgbqT#`{6< z+p?HzT8uDmOgk(a=Hb|()S9%A(;gx^kWWjF*voK&?l*qhy%$31D{<3e^PWl1`9i3< zD{W81At6T+yMSShop1aih*DyA4?VuohDONnI6kYe%oU~Vb%r9ap)=!yhtBrC#4Qn3 zn9e|>yb;kwS^M3AhU6d2A%yYcGE(XlfS89cfjtb+2bStuw0uq_PZ+#_c`7cx{12gU zTL|?|QdD!t@Ot~jiA`a362WZ` z!$!{1wP-CVt|v`?lHKKyMFh>GC;PE{`o<}+CjwqQk-L7t;cW@^3-GkLr7K&c+5E_! z`ur#7(yqb5?mK1uIi&jHS}uCg_ccHXKsViaDQQ4kkouxy_VowoN{v^4)l{oUcxL8@d8Pn!>T zD=*u>AI}etStlybno&eYWeU6Voi;}fMWwW6dlGt7XR1e2>xsyn;*+@!44d_>eo<)9 zM^>Tk>aQxj1`|Ct9xk{78!x7_a!nRC7UEw}Cs~>Hn+z;S$(hD)#>j9J_dg2t#Cb_{ zD;G+U?43Fv($XLb zkwckE_~N)Ol2zm14_ykhKVfKd=9zB21U^f7q-%Yq zVkjhxm;SVETi8#_gY^xMCi$MtAn(%>-VECs1dhtL{-X#fQ{5)C6OfSR>#Hnwmo6MxZN!*{vs#b`1V^m;^ z9!Mb$_tAY<<~l^$IY7KqGcK|m5$%X1@gT3rqhjt*HH!jQ^gjxteZ2fWpIGoro*;+0qQW8S8507{5&M{b-eNz3!%{9`HSsxJ@NEx~V#H=?!q=skddPPW; znI~E(;g*iwdP(=uXMV%~dWp!wrr)_R(=@)04>$OlHv&7^qQ9EZvrVLe0;JzYoq15N zn-~Yzov8G&iGyivopm%pB^GCIc|^)xgx063mnMTm+sPsCu6L%hm5}|7Iu`_Fl&tXx zY;il@;uy`i^?)=D zliGYxY(CHey|^EP% zHZ2_aMB;Jo=2K&1LsV9(JR*qnjF5MUIySbBDKFmDv?{)nX=lw<$mK8g+cjZ^3S5P?kt#3g#?pS_T9g-# zSyq0UVtyjoei=5`wV+1q(WBbo?IpI9#OCDU>iqGs8CXzk19cAAjGvT*=6>;dxdskG zNosQHX*CB@kg{34%54kY=+O4|j#PcU7Q8RKJK6rpyZ{czVP zS*EpwX(c{i&qIF;{ck8NCI4+)H9+6#pCqP&WdfG6H{*#TSHgB| zObh^G!)P7R|Hi4MPuQ&Y#+2#QAe&@5i4qovPk#aLEly{D#{>spK0aU~{bZ7HjvYO% zc7OgX$^S&MB@MR;s6u^ahkGu}dyKWhN2>W{0`2puPs!jl>IsIN`V=19IXxv>FMAw;2V_tu;IBG< zYP~C2r5bGMo>Ali4xOiT*pupDrM^K+o_qR8i7Eh?$Je{<57s~)xiQY_o*KsH>;>bY z*>->N_H(^bK4AX-ns>8a8iWyjmJ7n}v9`u^uy z{>drru#tRJ26o9I@LGro`s^;bL`Acf+o4a9S5o_ym*q<%jFNx!eX~!mX`U@~)fQX**QTnJ4{p)zMi!#Fe{QM^;)LPb|kA+?i4vjI_nC^rP z-*q>GSSN6wyav%IbMqpL_E(j>MC*Lg0YJ6i@K9Z`+*KpB|P_hy^= zgl{rmd?LbeN*{&usT`DvlKY;iLJHO9<;~y*{HNy?UquEIhfgz#;74HF2C&S8~nauv&jK&_??{^{-`IkfCkqx@Si&>k|($UfkUF z^?u(2ZiajL_PUmTu)?(dh_1aT^GsbT>5lTWuC8NLSpLVB_1pHLs(Fp3;}Q$}c>x+j z$M=7iwRbm;MzVw_9*yjJx8`Bio`6!&1B^n_iM5A&+^I(lL*Vd16&)q&Gf=0Z^69Ah z$SD_U3@+6Vjy?F$q0P1vmH8>qDOkVaC;X>(s>Sa2i$c67_47%FX(B5(@R=6cyaOsT z^4B?=x)@IoAxY6J?fHy=BH++5+*08|F!cYu0POESyr5d?E6mil0h=@RXkM4X%x!I< zZO|c_A*-i0b-A+ATKX^gayeU4wv!+e3H@uQ#9lYmH&#_t#9z2RNrYkpUdF!)1j^D_ zPo;*%Sv)_;Do&b~1b6=p3Xn&vkq9FC%xIKSWR;$g8z!9D_~~%Fr=q?2qD1WdHs-e5 z%E6COUH2vL(UOm_u$9`d`30@=lcpTj%DXA;g>t8pA&VeVSz9}A5w+I zT{{rSzm^@Z$W4KRD4E1B1!GplZU?goCto%rmIbKS(879 z!czwAJpWg&C|BEwHHGDP=qujSF6MwSB6e((={(+eHRWnAN%!2SG(I&WS%!iDp0~vl z0vE7E@j1>Z8vV^%Bd!4WiBI+pO#=C`8*yT1dsDdIv<&F(~crHGId(H z4|mDp_X}^7^Q?fU&qhF-sZP>KbqG(ey>4MW&6efy|Ir-Neg zM6%@=IJP7C?6B2a<1l05Zp5AzlJF4WKwGOUFE2kpRq($L zz2RWm@t0iT4VAti}^>Z5BK#FGa{_Z#fTq zw)~GDMbB-2{aX@fbb7`vw7S$_x!dERb7kylVDy+$6!bOhvSkhDZYG<;(A}Fu*Xxm* z+6>07Xz&1UI%_6 zyzRI%;ujPi^I{>I0#UTz4ZorbT`-9>o;o*J9b#PF3Fh=D(f0RlX4LgGq1p))0S{^cGUmg-~5RgtaEnYsl$vd+5C4T?kzWqKLP0 zKS1pK2;Zfm6g#Oq^5W0(&&3yM!dgfC?l>y{IaOMu%H zvZHNLhEfN31YMJp8W&C|^~}7on6kNKX1n51Fap3!1OW1neUDLYU;m8Y`h-q9X_K91 zqHLy|jNDXdDj8?-E?zF_3Z74;omC3{TNsmR+MT2q6E(fW0~%<#v{82=b8UV{(`reB66_Tu}m8Z1^N_ZCk#BHhvQ4#t0|D9vr@ zwCXdp7{GLq;>ni0>enl9qJeV=PM0C2sw{;`0T{(cn_+sHNHw&3KMF@O;RQwEkt?bfIl#5^6)FX4dBK0g@RD$< z*k6uLnFBNMjB|pT(IDDgwg-bQia{KWQHN`Ku&pJ4G5FAw2s)-c@u$ZznlW-QB3akk zqJLc*vw0A$*?+~aJ&rLk$+hI2DjvWLH0%=Dj7ePe23;EYz4cW-!Hb)*82V+cqmP`* zi#*3BVin*+!~4%Os;8a*^zHWH49o6fd#pc?=#b3M9=NQ1+jqx_NaAVyy#h4a(Abkd zLy<57Fs%wX??O;Xq;=dh7K^IXL>IxRfjJcgucYLMug~sFz6y2v@ob@*@gvb-$#31E zPD7WIvR!%U>l)2x&#cQ|OkR1a`o)=EShJOJbB3#Oy{yHEw?xRtsV(KQfgwpq<|81* zG%=`Y>JgcZ%^+*S6Tp{Eh2OZ&QOW6a&r2}e*2BM-$b&Qa?AsKSlb&@o@p$8WH_x&X zT2FpzEzs)7tn`~<_4TfPkpJAhZQ8vsVL77pCN>=mk zo+gQm!SA1(Z-@p_#e<}fz2p-OIpcNUnF+G+Dhj!e{7LAU5xmc*ix|jm(GcbQ?s(f) z2lN&0xbd0n`9vSLPwyVB;;#iCr@bCE12v(dhdMIR%t3L}Kf~I_N=#3c?waZ}|K;pk zMrU4fvgha)hh;F)(F8rn4KhL#)yb3I$X+GA5h5px83u&EOx3m3`^IE^{M5)@=8}n~9~jzPydyp~Eaug#i#6U-S|Z5P&Og)5SUQ zjcPDx*0nbK4cf?-*HnN^6WjPk~Ry1rA$pgTe-PVosKGl zgf($MUe?Gz-({X_k!h--UZ;xCmP#A>#1ZW(9u*CsmVO0J4$acWSo{gXLX(U!rJAJo zfcGXGdzR{hZ0na$t_8Uh334}33Z%@!AP-U7p~{8rU)5;TCE)zM0>8aCuwF_iC8DxVcUF>%Ey+-&Dl(ngtl1LlZmkZX!g4 z^Cyp3ITu3CHO}5GCuO~IBkD_bY^;p&@Q*#HSInoc^*mCCD;fG%rWA?A@s zt8Z7*Kc~Uvjz>ZX`Ln`b8r`ao)k_4ts|mdckEdcTa(B67xlSh!KIAcV2C^&t+>~3b zSN<1m=y~hfDQ!ny$eZA#oXX_mM1e-tyA0VR%wSjeg7c)FQSo^8)76!-N9x)XiH-{C z6k)8oR*H_B^NcXHHQFFb2bxpJfQwB6?s~s?Jh{ESZPQjmIIv92WEZHFZ2YhVe#XG+ zj`kR0@kH2%JHA7=<`GZ)se!Ss&oJLV3_86Aj+FW1S29DuQp2K|T1en=-DD+a4bs z7B0zU8x$kaD>1V8Cbm2^d5h_9Ah#6lJciEi3MaQ-Kp8(``0~U2)^_rKo9>NEfx1`A z`)+;>UP|-_P~asfa?(!&Foe>9%Ikf*pEVkGobQ`Ti!KZV&)=&Bd5&DK1*)l&Cb1oM$VElo3dIyL+WhS7gNfU%}ui^G_zf$Cp8&Ej`^a^qX4A*NGiGl7Ehj$>cDlpGM_9MT!}fc1uE9b|eH3SK1lKix8%# ze-8K-KOO>Vvudn-x3;&dtY_G@R@TuefV0R0Ox-^;Uypoc^eES0768ZbF91N&k}CjT)y*lH7G*@%Oz`1qJ9m6=yo3eS(n)WyR`|w(9P;0b9V6 zzT$DSQw(a=0f&4-h+R&P84}U{TX%g@UD>shM)YB1HK8gI>D3!Qu*(8WmZ+gcQIVQxX>oKte=@uV~4ad2Cc@ro^p=Cs9 zLNQqJZQ7^6c|#%We&E7^#N+V}M<+HlL@ubZpMQN6>y1272C>iCu)CdAx8(%8zHt4Zvms zI0Ko{l^Br%Kp_J%ln(&2{)CC1>U0C&jKleq78cT3tNy6#+KB-0z8zcmUs!0@+r3JB zBfE`^;Q6_tU(mAL@fTNL!w8j`4VT#mOi*OU`T)wx@SeV+S5AXr-dhw{#Y}=%jHO1C zex$JR{lE*8=*|>2y`zWAc8cvRcXIF|{@aEx%)E^rSZq&EPa7U?c8`-Qe2^qa5J}Zr z-q<%HN(MX?#XVJcg#iFE#Oz9DQC>=~4alNT7?vH*eRuN7n&!hc(wzJMN7-9PMcIY# z!gNU}B3&Zj&?%iFH7F%7ozjkgAT`pUpfpHINeN2F(2aD148nl42t#-C?RnqxTi^NP z`_5Wt77N!xVrHKG?0w(YeZ{6dSoK;U2sRtF1)QV){7D><2Cij$?G*mE%lx)r*m=1# z@X=#g@N!PmTF)3Tv>!ngm9hxmTL;ryo)~tU;pge!WwvpN)Qv0&Y$Ce5r1p2XfZw_X zz|XM8>)0q}rU86L(|RD?8Uh^NPWT5T5_Ic!S}u0aMgo{m=Db(}yM{OuA~`LTG&pg{ z=%SvMJw(OG0$3EY4&6djxIYT5e}E#4yBrHLLNUJz@%feH8sSb$r7@2p)jbbD5X4;0 zxu4Z6ySu<>4!%6^pL)6Dr_^9S!n`?M^|01g;iYLRz*=7e_hN#OA zW{3L2jB1Fal?mu@GNip8bFIEecXgC>fCFja;%3!Iw@?q`GC^wi z_Y>)3n`> z^1L%G9NRR~SGa%YPuA3_$lc3D58fS#wQ|bt#4dY|Gec+N)>7&TyLU&M$pRMv%q>By zM+q4+#fe7DQf90}jWq%8;~GzO#&lz(EW&Mc zDYYV-WSj-<`(kptmxK@&an*!a*Bw@$8XJR7-+^#(arNDYpZFf<#{aVfXiPs^uhFeb zV3!BQ=BvMTZ7bW2JKex!FmwxKl;RUivsM69DsM$$ztN22|9z zV1O$CUA_NbI2s0HijU8YRXVRM*R7-x&=Op|^}>~r#<_2`RYzV#B&5DxvUW}IWQGaf zLT>p@mGs|Yf+hj4_P*^mX@30kU)B)NzG#Z;ZH`M+z@2E1WOlb+5_DW+^=kq9n-bp^ zKZ?Aw1zFYAfg4kZB>L&?Do4!^bzRdN6|;Od)&qybQ&b0=UVpE+3Z!o zPUhcgk)qnK$2g$q*5Z3_U5U0IPB)8^_R;k;m-|-s^Pd1h+L`K3rVX}QWpFGiUHo2~ zB(pL@qK}j_j$f8Ez1!Sz)w<{t;pO}&)Xn5eh1VHoelIbqQpM+4n*O)m_22M|zKhCY z#H)YkL)Ym*ehf*bn$|p!qkyI|2T&!0h1G4e{5J5nAn36&dzN!6xq*{6%H_<7@y|aS ziCaNXSZmFene6L*oxB==x7|2ro_K&|=-)3+QHmSr*|=5XDfg4sI#e^>zFC(h+=hxa zT3w|}@>AXvn(WMWUW~eZE2#dTa)evTnneAy^VsU~8k28wZR0#IBgn*7I>>XbAguXPZ)CIQO(M8ZZ5^T&bb~4jCm`;>}s1 zz3ymrl9B>JNRY^}m4q*SvA~xWg$dU2Z{!4S!h-0D8Qc*@tZ~^S)m4pei_LomRm0)E zpNLsYtYKY!RWfZsCIx50j|-Hiv#YSkpBe~ff^C@AdUWR`V=*F!2&BS*0CNNRs|B!! zchF*16DY7}McZCK{6#CtSg0Fky^}~kr;azy8nFdZx~s_4*@A{3-atXLd(b4RKwrJ* zbsO(P4$+l=>*$?3GdweKt|QX3AC_rB@fV19ND7H(ztQI2kit#zY$Rhv*usy#C#2gA z1Y*Vd?8Ua`D8MC8kJD@D)o_2v(Ai?e$v#VJcVSKtGFNKW{q8Lf2PLrmT1XalXV5e1 zlKd(m+TO+IzDS9SzMZT6cUFJNSq7zTlcd0!o0Sb#AZQC;;UKluE&(Nyg<`4^ncqG- zO(<+9YYI?fytL_-1#21+XDB$mbvej3o~o}9i@FuQ@|FXD0U!X8lmOQtxx1IuJhKHi zM+xN5PJ5o;`cfw{Y#3bmY+jiEdT+TIU^?I&wyRnPWe~BRnrTYFHr@f#<+s$P>+tBG%?+N*#U?iAOX2TOweMD_?i~d6ac3 zv1RIIThi(3(N3}f#b8+0{G{$-eypDyQ@oH{c!e(IuL~yO7fw<<3aJZMznnQV9;#D< z=LzO={=L~$_>>A89~=vFA+nK$YDn(@a^}|w^3k@O_q8k_ER2s3#B^Dx9ld!K1W1Z~t!M|8f2#CHZ=HmpV}+)`$k6e)Rdh6opz~kGrz%11J}G-)SGn zWmPf#?2;*zhpMY8bEVD=d)9DbA9N{5l6KB&_GgE_{UP=>itD9UJ{2egcUB_YaPpX{ zCIJkvklYx(&g{N~qbftrykgOOM``Dgc=M(RNkih>s%r zFnR{WVL`YrSVI)P8v5PmeaT2KNq_|+QHdvBLLO$V-J)YkD>H2xR^_70uGarLu)cdn z7sWC#G_THNXJgZ#))>j%-K=9)P5`OxZ7#gS+@p^71^&YE(fZ=pd=*fDcx^Sb{TGvn=# zZ0p(+)Rmtek@QpUP~-(sq8+IQX?8}$7{d_jSa*}z-&?(Hcx4{gEma|Qw8mSoW1^a^ z*WjznwO0jR4v;gWgZGE_u|H+OjftV^gk0)*4>Tpwo{jZPk~ILl10QtXPXrS*&+@XL z8uDDrh1-;?e;X^JBa*&Ics{e_>Qss#<=55Z5|3E_t4&y)MEh#?C@kF9Ns?$ZyCPWP zafbpc1vf7P(qJ@DTSJl&5j zcyE~H;8Q{@(>oNL*9TW(2r-X@%joc(o6{4^*e;*beZ5L|i42yWN`iTp2)4+$)N=MX zI>vrC8D-|4RN5+1n%S>0ixHuEDq=M$-Dw^^u1XR)+@&xfak!(F>?0N4Q zTJvFxCA9LDUHTvvTfw2^mH#}xi1N{k$U)KF-;9bSQ)o?9m8IW5Mac`dzUrnKKg@jqdt+6QG-h36X}#GVpe4T z7Y(rccoZH5?I=7W$scD;uh7vaP-%}&5h$vrQP$oL`WYHiC7{qC6LY4=?I-S0qGO6n zh)rMLu8`EBEeYEDFkppkS4!>Roda=u{sdqx=&7OHa)i#cm9uEpzyCDHTUMTI!k%z52|0OAFh$$8_18G{g%;$WKC0-momT_0Nl9p zHkM~z`d`lEGxQ23T!*uJdzncun=1K?EWqAUa$-X`V(glkn-H@Pfc?1{599;X%Mv7_SH1>qy z8y~hz%BBGOfvn^afV>*!_!B67BuK1DG4Bt^fjoRGi!m!1b0C%6RIh?fmELb%j`_Q` z{$7{zNu~3W1n9xt2PF2(k-$;6-zld(uvSehNR;&}U;f~Gt35YKg>e;~ci;e~NjZ}% z{QwCshacF17-?~lyUJ$*JL7*`X6x>NSVRDHq-14`+ymZD7O^wr}%(lzgm4Fw-)ui3r9AzG3$1z>9H45^_h?KpZMO zVYi{sw_QBuK{0-t)}W)CX2;T9&50}6^S?e@+j}qIdtaO>+ZBJ|YZdTVFCF}cCk#qK z+R8ZlO3W2FCNK8$l)Xc)|7{TX;jSV&Xv$m-*d>>k*f(@Dp<=_6ppm_aleezNQCj-KK)rBhYy=brZVzsy0#SFfFt{Vu3$Wsmir!N zxj+8}UJBVS$Nht6UC}v2ELVS3U6(#Sl;X$_8M?(rC`r!#KU@H4ID92eb$2YrAKa)K z$m%3pJ|oE!z_h@{NMsgMtfO5OD-p}8&GsNLuusdM612Q^DDS|LBb1B=fHFizWR`JY z1P*Nuk`JO(a9giXtx=#Mor>aqUKvnt) zPN?)iPn8mk@j%WYFIZOJC%>DhSU=tSyN>B=F<#K+`YLmdBh|w@-t-Rysw{xD^XYza ztBm{B3ma!iaOMMPjqLVsMb;p{u_}-l33N03&4;=`%mw^VZUH)lLVPd8!Ms^}#QbW* zpz+H`F+l+`*8N71GYZ_FvBhU!<8t4FyL6}QWnjbjhJo!q#!To}o$G@oLxSdbo6DzH zARh9i?_n;BGB`-U9eUhY=nNtzK}@1^%bA_2&lW-(da19oK%yoMc%92Y+R@WRV*v(V z>9bU_F7>1HE$5SEo>C)GGDu1!v}qs6vWVtSyTviJmix&@$HteN_VOVB6kTMiMuG#b1M|0Po41@w zKS}eOfow_rB5$xV7v*@y-z#<9AG-7Y={NVoGUJjZ0AzGw@=kBY?_c&8?88cc3y#+U z@VCJM@;IxVjj{5*l0*d`)3Vo++H~Flf25OxiP+Ej7X@|* zs%9eO$DO+MCXD0>bIp|cnL_s~8OsjPcVpu|14pPU?>-IlrP8$%W%4Tathq+7<#TkI z*FIj}eqXDq>we$E<(TWgF`CN{-6y3DjAGo zXS$3e3+NsLG2T?#FTDpkb3_AsFVEJpV!S}c<+e!wNnLdqgL^47Dyxz(j~RobVg4h6 z#Ml>u^Imq!U*h@$ogMi!@s~QL1zv%0-*h6V z#_BD=O6GRvqbK+y?!LutYPUFYP~&zl?=vKJF^8g^d2u2a)bsD_(3YZ+*;X{x-cEDMnjsIiAMIBYO*^Ma_UsH%mEY5gHwxDW zzCKV$uJW$`;iHK6C$>FV^6Uax3L89sLkwV?3$H_J-fd~7H6S;-YvIbSFQ{P?9IH=T zY3cCUOLQr%L$fy*@Tmr-7oYgkwEhp!0Dy=O`TtQx{(qo`{|GIB(*hoDRJxvc7#hTX zkLbtbaMiY%!`&2#q-M)A>76ha5EL&F(33Iw3%l2N8WGa?^z;n!b~*jbhPf)N@%2N3 z00ZXVBI+v&)X4Xs@fcKg`FIe_b^nw<&wIw&FMJ)URT- zwJQcQQd#YqS-cx`MjBurC{uSA%S-hOscTy6Du}FNue_IvSobW6nC5n^e-g(yXdRhO zKeOd&6}g$4IaGVIywm$7A^nm*G?&6&+O$IdFs>DbFwSnm>Rd~HZ6C!>C&%aelj11Z zL&J!PV?xd-t<94csswCDq}Qiur^##tV<(g%c6UaYo&>rlm&jDwn)k;5(1E1u=Bz5I z90}l6srtU{?@5FdGL@c&wZ%;H(S*19-+%&48cq8^x`fIpwb0{jIY;(jE(UiyI12c z7NX5w+0WFuZfUlnwmdV)%!bavxjyLMy0e5EK$tCKD2F7Z;9#J6ulXL5vD^k?GrgmOhUbY7zp82x@om_Q@neo+?DXC`)Fp z-{MIv6YcIj&h}sH7LSGAbD>eoFZd!-I6Pooa~Fq{5E~huyX<#)x!dA9{NjB&eDE=T z|8*9-+eVRvR<0oD96`d=*;Z0lnMP}Lazo-AWPgJpR*Hp`pwzmzb0KiaeSkxK#ZnT+ z@rx`N>&ic16A1$fR!=wPP`5y{c(sRih{&V%AIWA41k{0|YP}&+in6O*#em&ljZQdO zRSb}sOUkH}xm*K41qj3joWE#<*I*37(H<^TIfNB^u3#2pEy?~}jUqWh*R8)6vGn8< zyQ|Mw**!ynLzA|@?H_eW2`cYB#5L%jfBAFOjkxo+;%x~Q4HMbY<-Q%?)=%t|3xtZX z(xZz7))}J)p;irMFOwva_^SN^07bx5m6Kw|bKU>}6sR0F=5j{V>=)HPS=bA;tsr5U zpiTro2hf?#e5KQNb3jrF--_XWN+~5rpr}CaPOW(VOWcIiT#q@7xD=jIhIRdj%256` z7CEDBHY*PYIjYr*aR9Oe@n#{B^l1=Hxqfp>m-vtgd$JQA=Z>`j5>@VV zHRMA-(Kf`@oovrLh!HI;v28HYGWcKw}208%U(_|#<`-Qzb zAWLHLhps$2%@1o_Dc{$myPdrM3=*j>P5xB+(Ifo)@DL;`5ny1My}dRx@%_Bj-fc{L zX(SG~z0iDrQHqjBn6tCpGEgw#AHr8IC1E3U``L4c=yySH%;_;Ne9KcvTpSE(wqC?c zlEgCrbpGV}f4ac>}rlP{=O^lUtvJ~-D#sN^ngghetmNm5ZIudueYm7=yDqeU0K zK8^u6&3Fa2|9x6fQ{-1BeCu(`V4qtm6f>#H4`z?k zd}7`pn4r_+Tnl{Q030#F673uvJOaGT1-rq-mzBuobvA^^YAr^porOS0S=*v=IQpdb zlM5`uLxzIj`X|U_I<{Pk$enq7Ws>HuXz|(HLzJdX9g3=)IL0K{otu0RwT`v6qQ194pC7tZn_Y zq&89?qobqZ?ePNLiNg@oCHe!zH!tzHx*6;2;ySFIT9$KTaou7hl5`$?UG#H=z)P~a zTPdR4ev6bm3-P7LQAmf=LItuM{jBu3%}gjrV{|z_M#kbg`kw}5KHts4%^TUzQnXey z^!L}qmHuw->rGK&`!kk_^G{;G1PPJ`^`3k%;M%PZ_CTR!`7b{!`A>nHJUUVWer;Pq zR+F1t%eK3C!gqhfz4>aUepQU05O4XA)K*0}+V*%jc|u1U39%>b?uk02{WNT-LlC$J z`xc^*q+dRl1Mw@kUj#FP47BjsT598=6~DRL#5!p;r?acGDF)4_u2uOlye`Zh+;xG4 z*}sk0?-R3}E`ByYZ%o!~!H*2zDHrogejh>&FW=;CML34VozUU8imMQ~okp@kGKK!E z75fZVj%wa(-#^>`n(%`Kz6&1|z_m~~tcB=H{u&*|deOO#doa^hFy;Wpo|0GkP&?!B zmCtg)TfGfV2;=vj0MCAj*eJDPmJ|_n{N{OEA?jEclwW}N=7}%TR zvNL^0gtOJ!w@@OF37h_w6jDMnxqNVBgocELb$$Iwl+dqQ7YpanT1sDBRn@_@SHH^J zTiuDc2^i*#BLJ8HwqF!%Hl_4V) zn$3H*=Nf^a1>ZVU)ghT#Cn}3x5o^H-2tuFybNd) zK|dO}7!-uSL%~2q7~mL`NPmr9u`0pq^f}&CY@PSY&G$Dcy%#A2X84A!{>`_(FB_`J zAM8WVvOa&NuM|L%k8Imz%Z4x0f-c}qg#!2qe^;tvGkyAVXU3zS5yN)P&dsc~<8#kXNHFPrMb00+Q1XW-e# zJs)cu_E5(TZ51`}tO1w>re6MwH2&@QcQ#9&zdA()ID_v8Vz1qGt$-Q6_x(8@=WZ1@ zCt!FR0>Y)R08qxyM250tMZb5f!=~HQhFSsNl{#R4{4Z$6!9>ftm3ZL7{YX9yG~x{N zrz%Zlk36>@%(M%}_v~cHhI!;QN z@IQ2y!2Opt-oko3dOc4PU^^I->4k*nnso7g7Tqd)r^_+cpsV#h98(ROVd&Mu#lm2a zVRDD})7{6_OD8_DuUGzPs#cF*xW3rwxvM3!XfBh>NS*q;q0G3V-~hDMgIz-z-HI)nfku?R*%9=F&sqzl&=hrf!~r!zRAtGk>w$ zFx5QV%np{rZRIoRSWppyVgSRsI0Zmb@JmFvvjLKe2HzOjPx)`aSKY?)SC2{{{ z@z;P?@%Y=en2+(cP>0 zQd#7-IvZ{*7zF|>O#*W*p2U$7&?HvOCnpD^kEHmUb{_))*Ck0pywQ8MX*sH=Ej7tR zxOekyt}2oA00!tH6Q2zP%0PA>$v~?Iy{%`By@mWVK)VwG$$!0FzqSOR2d^w!E9P&s z&_4PV%w5t0jLpT}W`+#duEkVva@$!ma*oQZnfBj|qx){@#mJpQJhx}pH+-nBng+CjbDE02*85@0Ae%qntMOue_REC&+Ti{@xs_2H{dDF)d8&2 zbhL#w0Xomq$6atEpS3(+JLkL%mO&;OYs~8GNYYX6tWOIdMWwV2@@V%*b!UIl{zVkQ zHmfv3OEWs5rW<-S5`zce+}~%zUQ5*zns0DP{3Q=&S|?vF0)bW+pwVn>*aMWhb0=VyuCH4|^@Bm@{_x zUf!xPil=h9lg<%sTjKyUVBSD2_8tUB)~&5f@?<@tCEbm3@x`^cT%y?nh$GERmAYj~ z7Y3Wc2r1hE(zVksS!$l3>kWLa_P%NPkzjr|`3Dh|_VeEW+%!KQPhJ{w@qc3#wyf>e zD~%&3N&>o>=2DLwwXcLdib}ST#Syqqf+Q+;ZmqnCbQwFD-f^JqZf__MrhlLqK-ITR zt4AH3Zl9Pm&+y;@yV2_YLE_K@kQT$)n!Dh4u&s3}JK!T|J7kPj@eJyk9DsgcgHQ8B zPavX^U!S)t^VY~w7yzvI(?9dqSZB7(q~pmDzsu@eBAnEJmWe)7plqpJd3VZrkKnJY zH|__}Dv@aaAL;ZWI6?@;OU(yp?gwRN{MYb^ctO>=m$;AXNe|djSpd4@InD znzQL`yvoZDXQojyIF1Tg?3FTMBG=G+BW^g~hD~@fy!thD%#8##Hz#_Em63L({{#Mj zfAtgpkKj;$v;vSK{rBk8QTvZ71K2X)AxR<$YO9v%-&>w@6 zZxr6XU4|4VGQF;ArVFTpDcTDMWVcF^a z+ZVMuVhK;;HeKe|-M=JkHaa`KrGI-Y$#b=S^p-Gv5Vdn+wRL*6Lcv*WkNkR#y6Nk@ z88)tf6@egMDC)WS%vQ}rE680?Xt`qfx_n#+hH+b-(ib0acaPFNQM6UuCI)us&oxdz zy*`1!64A70kVNM&nQi#&5Rsa-)>qjX$vjlpcw4C`jVAa&f(*(Pb;Gy zHLR;w_(Qp;_buLe2cZ%R&YL$`laP~_pZI93!iK&2r(yeFLXK)F&7L;oQ{n|lw|zk6 z?SHAyS=+;Vbq#m{uf)dI-I03QuOa1PbDG2RJxzsjCC_j{Uk!#K;N?m?))7e=F#iRS zg(3Pyt2Ph{xe!#IF%10FtXKeil-hthF}qEWbhRl;r^Y7#M@70yX#()+4MDm|eM78& z8OiQ8u9yd;jkD~owzfnQcASW+5Lezcs@+JYnpEAQssLGO<^(SJ!pLQB9TQs?Qs025 z%`YC-qqhFcJS&I4eZ}jgW9&b&56t-3 z)f@14*>#|rv-wOpez++9gNYQ;1W_?iZS6+oW8J6-(#>{aSE{XlLv&kk( z%lvopt%sZK$S5jprrlCm~9;3g*bTG3FA3HL8ECNdXm#09Ud;4$erZn$@9wJUx=B z{JOfM;qTE~tS6eZ3imQpvlAl!?gwTWmxKaumcLCO{?(pcwWODI7-g~U1hOkza!yCYZN3&OGQWy1Tv{Q)E366OcbEG|E_;GHVnhkO*F$a$^9mO|kREQ0hvqN6dQvvUK!2GoiZljo3i+9^yqsYa zR@#w5eIAoJX*H}{&A#P{M4rAr`Y>-6s!Z;2{@2VAyNun0L+zvMPp`e=se-Qs!F-^aR zjyVOVI-7>C!$`r`ujw~x@t2=WrpK}~Cw3~*;ZADP-}Sfw&7TOzDEtSxJ@~0SLptMv z2MyInfv0f*r?PmfwrBf*)VA+IvX*Pj8>)mwNc^fZ9x7&XnrA1Zt|8*uP)? z(t%CfIRT9g(^balEn5g&4T`{<-^ zM|=xN3UhQkkaQ4oOo;a~{G~{cH~48gk1~JUd(e2DQHMO-^4gov&nS*b#yG|IXT_6W zM5N)r3c=VKlvo3%eC6$2yHrH^RNx*j4LT$%MF`V(!lueBvUt>*l1QE@V+8imFI=@a z{UNczkZY91lyk}HY>wxYvA!{2#RB*UFNpVGU$yO_I&J^c2lhRI?&VgGzW9MlJ*~I# zPUtK|&k7=kT7p%fZ#9lZDm>?MDK)MIgD3u4FBvYuew0L42BOI7~ zX$8QSQ;;K+{%pz8n7-6p%lxw_#7L1@3{@(yCSP^R*CH4=zD zH8`|OF6=C|S0n^k3 zpQ;jp8TKLoUfMBPPgvtf_4v}&VV_JSJrLSRVbP5pB@?gY@6`ig!!Aw09*ySLPHm}s zZT*{i%;2`^#BkkT<;njlPhva*fYHCL_|5gsO*H85&9#!wq9*2mFqx3q8_T#rx+h8_ z7ts{s)=4J5&`!wWp5~xU7_njpx-kG4xdL`SrYjKvwSD6tEY3_iWt4!$Yu2@HC0=6} zWG;=Lu_3o0-a{D7M&M>&z` zInYW9o1nSs(J7nS1O3@u45@#P6L`y9|G^sQGTysTyuASIYVtU8UKHjvTn>co;%7cs0xl&Ts&e~xFbK<_`SyYc*BAGf zfB;A5DrIkO!9J+GYh+B(dP>zJRO37c`sKQg5(Aq#`>dXu^}y}AWpdy1iL7p*A2XHv zwgPwHdqCOLUFLOjvWSzxA7H{GqVM8|HgO%6T&vThLED+0$z3i|EW<1=K3@a~IqG+S z#uFpuv3~pZtr~L=-G+&CJ-)?Rzd>m5?Bw9u;$yI4rRUFvIciM+2m)f><$O;X=RF(| z%<~oBCtse9C%3$-ZN1t8G>ZEu4gjeKzMr~>s0CSw5?j?1yGt~!-VWMUk1F(WST@FA zMw)8|gUJWLQKql7(pqRQjS!m(eW0r`s+Znrx(lw@RP5kA#D$xXiXE<|p?2gH1T9(!u{rkOtTK+`do1Q9`Z^{3jqjH?KE2g1%GVeUcX!c+a0n1m z5a1qOREGSN@F`1hr(PcbLc#If3k1^qQ}EGOM6uj1Aw~30;W9DXlEnWQ68~R%jQ@*1 zxcC0tqrcki8E(EyhUE9?@V%Z{N`2h|>TQ?rZwx~6JA9)1wZ@{cjR*9DW-Vst_woT1 z53v1F_9>2+ZWkw3Ejen@juzb4S658jy@kT(j^tvw8`@mc`$*&Ivk+-%Lmt%lpFTWR z)bDd?=}44@Z@vQ`s$Q2L{57^EI4%6OiTUzhRWG3%L48PmZ)GB`5Avz*e8=I`<7^_(H_t zM~|n4RXoXZOS>x4jf6B+5J@G+z=S{;_zff}==djx2z4jMHJZ3SyluHJ{RNChK!3F2 zlWxlV@>h}(L#zJ&Gq@n9Y)PWqv(&y$TFEYB*j`Bs-_V58y!cpV4H$E$4OJW`wCJF5 zy#{#VVfibh)Hn1nThthvAwdmO?0HMh+L;HmFh+Sj>6R_2nvR&1J3Z6M2E)c|ax2H7^4!9SM*zmjI>U z%<25<#+BHbn2w+v#jWgmm@UF|PjFX~m*HN-NM0wlu05~>c+F`W%jp^uLJSlnEZ0cm z;CvmAXVpnq5$imO;Ua;bk~l0SV(qY4@UTJl>n~p%T}@BQX)AsxVIGRdt@Tg4rC`m3 z8ePFA=f`wA({FmF8BxJmRofkc?qS1JlEfP>j04AYv*FQFLXN*$5-5=Lj>&|;9a1pc)$$>lu>}!}4qEdhzUF@Zx0>YRrSm&}BK;j6_yG;@t8CW+e-* zkH`r`y1!sV7eCr_+7)V~`DRp7LgcEQ6lTV_I%p+r{lIVo*{?u~$}611u7Ld{jyFi= zSf(rCQ-z2Mw`6DMV4roUcG=H>5qEmMuH)^DE^#r9_LS;mX>RZSyWaD$wq@CO?Je7O zqT`ZphvQ4WTzz*^WdGPUDW$4Ti9V3-V)Hr$cnlK& z!X@DP(j<5_ngJ5XYy)vH79ga6W(i1L0N*l0^P$f)HY71KA4pG3O~4Wd1VCgQ-@Of1 zl+m|C%QT(0ngBrtoi7l1nidoqCw_YMtt=PtKnwL&-kB?GsB=xH$&lLO2~b#!tbd7$ zOMUOU-^}Rn2^8JHL?P|li}D1TUnTEAuM9X2LqO01is^DscvW2yKB{Iuxw8kf6EiBg z*w!|juS(jQ&MPUE<0ig=A~Xf4llN2l+q1^?fVYW+TC3SO>faKwk)+VYE+ZSi7)+6^&YgZc=g4?k-~;R z#J*$6X#&8K*4G`y3-5#q^BCIr7~M39-?-GG>Y0Y)UvRpZO$kylb31_nVtjA4SKQd* z8b-3;+lw1zaJ!dCpBa9_7?u@H&O%E*sG9xrZ)TmYaQtoMIFTGZL)Hvu-0`>tUcFZY z)Ov0C9cwlL2mZ=O_4?&vTTk~~=xbRa%tEVwiE+)YThkj%ffHvo82y`<)y>ibpqEO!g=6o;({g4`nk!36B7w_d1bBhrKC00`+uAPT0)AEE3Q^z-7l8C&x!W(n8V}Vtp%p9&nHe zSUM^K)6}FtqKq;^d>keqpDtSq=L~SRt}R&!n<``_j2n|)#R_!}u%zrD@I2;#mqegg zNR2a9%PUvW8Dq4&m#Osb?@jmeWhhHx7i|bA8$pAx$$-{HFXc;F(Kh%BkvUnLt4Xa;%wYmQ0z?yB3M3On zz5fDBsBn1Z8WmUJK)W%qcFIHf9r!LiJD@p(IUu^b<^(GrUl4vKoe zYgg6r58us@&gwgVL5C#C#km9H_`7p*;avUs=oXb17EdBcaiR{dJl3HtEZ%`H*RZKo zcfFA=w_hb8-lH}lsEdH5N!%lJt8X6e#DFI(Jik*g5LBEv_N_@Yuv!^tTB!P{9%U2- z&>usJ?-eN7|A>41r=rj~DI`FChP=j0r}>GF)c2Q9r9yf`TlgvJi|JQt6D zB83efsnS1eFfU^fEb99r86%i5ZL%RKzubGpZY4)IdLU9`eJ3%Tm!Vp03vYj?$#);_ z%3C#TJiz-4!PU1RZj8tR_XA!dK== zdkv49+)D2y#OIT`vx@o-1~9uDif! zLA>zn<~|;V#VatSqDs+9@{1q67D^(v=+2eEug3lqxGUU(q+dx7YB$ju(?_v9Q;}GLm;YMTKi!5MUGH9!G$? z&ls>v7_eX=cjxb`l4q*_to`}3yW2^zj7dKQ-dzV91v^J?Pp3!xLtD>_TIo0TcoIGF ze3hId@fN{*05&enkT{l|`1?Ob$%E+irsQzM`cqq1pMGCNMA}PA80@BUoMcp;sVVj4XMoK+Z0iZ3&;G`+9B!8DWU5`*{(>~NoHg>qQ z+*wr!K=fdk++dW^wc^SFriot7Xfy9Qv^ho~bMn+W?@dy&Rz6jY%X~D@+djw2YfAs8 zQcV7fRy-bNB)$7pyqOrW z;5$H*qAkGrES??X!0`JZ^JD?=>bp8RzzlrmGbtw+AOQLG^(~C{-AvVz??q%gKyvm# ziUek0W)Oi~-Rz5tezLkFPCG+~UN4|T`Ykn)KP`IF2`P4-M%`uN`w=5U<;fp}u1@k&D{eQY8KRvM004`;WHa3Q{})7hAQ~ z{{GB>M>hF#viYYp0OjH?MTRACK||KUAi@1zgLMnO9v4b_fja&>@r?wt@z~A zbM^g()CeA=|_$K@o<0&J(x~KPc z^NBeoRh%ezkq(=H@^Y)71X!hjQSM&-h#xs1T6p!ZGla1zWe^T1S0}|Cva1972E{@s zgEPj#W%f}KT{>T=&WLj@>hyf7HTva0NGgH%pt8qNNRl>r_-Uo0hq)6IID<9*$5&I| zuJVWLrbb8_2K%*hna0)U~VpR{L5O&@$^y#bDc5C zZ*ty9E?O`QJIMo11SxE{R$=4w*z+Qs+S?Csdh2mx~IUZ)+=zBW)hg+ z?F8Ihfb0-41CUEY+XXIZ6b24?0=2AU1Ah!$@drELw*Xg^y3NV~fYPZ4 zBVe#RrdYELz|MLNpru??;)zRP1PmWJof{(;juG`>Y+V4A$+KmI4wO!Sq0-qSf9;y? zyWcmg8F`Bc+j4C<*E?9cIuG37*Z#V?*Sc@7y#v$CWp$57tASePi7=v+v09lL= z^P&$4D4*{R3>QiEqdtjbz%jco5Rj?vQfoS*Wjn4;6jzyFG zzG)pWTGP>AVEha0xX2j9Y;>QJbAJ*NW-=;?_KzDA(4fw@;S51r94yFQ?t|t6!&e5_ zz;?Z8onCMLl02?j#s?t+urXUnH2sM`YN7rT1f`sw%>JAImCZYAb7!)RHUz{|Z}vn4 z3oFv$g-Va-jv8A7S{q2g@thL&i?dG6$iUpK;O!-M0Gjc@&3?=l0ys}(vYi27#=_d< zY`R-P2_$yOR}Q>7>;gXasB*P6Y{!DlcpU#0+W&xPw8|eY!lJIt&Eg`T{O+Bq7D_`J zKwL_7L4wvoTKHO8y{2{>yCrztSj-iScqZ3Z z{mT4h&1k*MB=CTTQz&%Gt?kE0g>LL1^+49tA;XU!4({JpW3T*T`QkBUp5+HBZ0I|D zh$~Jz&Qlz7wYCA;o_24#7}*;8?q#FxX8WD*CraxUW@fcT7hqGWA9#9`SLCCa^}X~v zaJ1R@_I;Zh3Cd&eta-sAcsAcrY1i%z4qb4Lfrxdr{)%A@k%*j{Ckf-I(9#5}3Eaa# z+ecqYU3(B#C{`zrvMCV{mY%aB;XuupkPiSV7MV@~3!Ez7V9EY$od4g!`oEv1cN1{` zwO#$sE@C;>jTK```rpHiQJ4IG*KjEH2mNnjde`2DUymTsM~YOl(u{@;Rgq*~1dI0D z4`4kf?LG}IofK8LA&pE)x8I@#=fy0W6>AJ#B1$j(3yyl$>x`6K{4ReeaZE$0|7t4tgjsY%(3`_#Jhh zG)!8?QO{SnOuxuHBZbO&u3no!@e9TJ;`Ow$+c&*KBSKH3h~rl3$1K$b8gYeWcG|D4a;-18Yy7DYw+xcQJIe6B-oxv2T>Z%;KF?n&uzD=C94d9Ur>|z@ z@ZtX98$Ad4#PMqyCo3J}SK_ZUs`F=q{beh@;s&)_T1rhAKVxYfQi6m_rD+y~fUiW8 z7(lc1XnOiu+W1BDM1R!mUo2s%tG<#JZ5j^!gr7(_s-oL(gO!^Q5W)Lrf?qyS2>m(g z!q2eMhV{wPlR8J5#Xl2b{0cJ?y8O`mnABR^uKCLH)wT-}KGc0WajTR~BffXcR5*@= zFPq!$GNb}oT_~{Mtjp)M?GY2BB_&yPMBnPv=RW^2B10=i`u}kD-SJfZao<)bvnYF& z;@DC4E~7)_q!8g`Rz`#zTSUp8$&N$HCS+!3gp90EWRtzu^SQe3`*q*X{onKaUcG*Q zG@Ns;^S!>;_w$~cC!=I1;f9{)Nu$=WpAEnM{Yjuz$a7lAR3i4#${+7ipn7_d7cX;W z*kAGo^{$9h<^8Ac6Kr{;zP9;|+;x#=n@n^lm|h@R66d@wYK=X((bx2-XPGBZduZ+U z2Ul7~5;qyk(hyGd#PO;cpCuT=mC#FlBc+$aU{rRTey<_FBQwI;{?%irEh}1H*ym7k zy|rk1sDYC_ST?>3uGZdmNHAv3OM@JHNpNk^nwA7OY|*WFXL9aRk3QQj^u~Ua=xToR$#%K`Hw^G9L~s&y#_kSTT1%||`FuC$ zVQ9U`_4Mx|qyb4@5x)r{NiX6X4-TrN^v`!a8B`5F9+H}7(mm>|Hm zW3MIFuzF&fexr%7CqSNjBemhKO8^zovBP4+gRFKB@`#R8E0v=sKeI(1Ru3$2kw?mn zE{Vhh84vyPMDb)=jVq9Eda2HhZp9Uo^qnh&N<}pP{J+_6Gh>}l3biKq_B7R z>w!S|d%D`UFKuMb5yUk3lJqg_E2+`O-`bvUz1?%!nAL&uSVeMg1v2ivdf&+VU79i! z^Ve(hc~Bs7g#_0;S@APUsbo@%wb4C?4|HV0G0ZGWk$N=g@1l(_n~&R!6usMk2jUz@ z{VnY`MYZvbram9rqOptt&12m{p>)F);7dSlaubkg;<~AcZm`es=Sci;;#5rQULmNs zhQGYYM8m8l%_3V@GHM6ov-{+C&LSeIbSsY3h!Zk8;^gwDzBDgs(5UyAOZi^+HPAA) z;%O5J*7&*X;yN}ypv3Tdtc8byvsJj=V&)ZjniX;J-*nx*RSuc1twYApkk&Y1y}Z(f z_KMCR386RIGTti9Pk++SampO=_Oxd?X%@|guF&6>o>5%W56>6!G3V@Fn2VnQrm2n- zN8b@R`id7%rfbBM>KBQyZluJAUmG?yuIzbMZ?uKl_KE+#hJHxO(kI46GQ5OTHQ3z` zh7c_kKxWEDuFf?%9JSs4*biB^2JcdG+*4$I|2U{hN0v^;XNgs1ExgBAg`VJj%Fg@H zWW8U`!hWO?j4y!Ag%ERpoL;n_)d#d*XdorZ<7@hunEnJ=37ws9@6Z@tQgQpX5`AL5YD|CKm*AMp zm;Ke&zGrX5zBH>b3{wa^XC_Cd)G{XSC9afe#jvERGHJmdO%Wm~a~pLL1_g1*%oEw* zNXgUPi^nO>d=o!K*)sk@U)tR+i>alxs47Y5Y>(a+wqU`p(P58`pzZmen9(B|ZVR&u z(^nN|D#r(km38O^nAZi zVs4bn^Y33rpMhgaCUu2T?`{e#dI|9tV!`5)_DD3SKlZb z9|$AeU6Z8t?+#UBT~PCTKO_H~Q$%CdhT))N+_TYPtxCeGVp6}h<;2eBw>%jxCpB_z zTl4hvXtdmsvUQfHMYJeRm$ZAk(3W-T&xG_DpC80qQSzSckA(X;P`%4uL*!{>{3T2! z{e4$Wc6}O?DM{T<(5f#MCNR`5unLIW=h_H5Rt;QBKP*{_&!>J#d;TiZwa+`|pU=D@ zUDwEZ>o))K*}QtSu>}8p#Wk-n?$oi~A1gAO_CHs2g!XB=EAHzy*C{HD{a9X5+n4OV zXK!wpa_4y`^VzVZ6G_e~;Y+8WQ#>4lu+u)`iMVt<)aD~-y}3+$QnAZxFSRHzEjjlO zkO9MulZd@#ehN%9Qg2vpnmwzsbHqXo^wjW+Ot2w(JdhNda zmlj~T?+NJz@(mx!t_T7r!*`Rd(zDx}=~VJbWF=yQOD+&;qv?rZ@NgsH)_O_XDkD|u zwY$dWpB$ox&i5cu67`Aqf2yUHHL~v%g z=MUd#vuhUL>*wdU3F~`mB6Yv=3gc|tYlCAJHvC4$oyt)y=QYI%O^dW><2w(<6Ud21 z+2>5@zc0+GSw8;6ymwgRFl{R{|1R2#BnMtfiEm$bOVFUeqr)DLQK}qOrM2Eq(32nb zoTuXXxi+6Xce%>*P~XSacdCSk|Meo4mOE4`JmV#O{} z@r0X5DRB8hUgeXW6)aClz*#tT4Piq|@zFaW<2zLOsV~$muYhU-s8u-qQnO0>93ONY zH$wX5wCd%8x}wS4yC3z9j2eMqwPl6khLgEEe24n9F)qLrSS2~1gY1M4(P;%Tx0>q?}CZn$MBH zmLOpd6}uv&;T9WnzwRONBLJ@)1mbK$iv%LuVM2XAP z=3k%$sl*@~lS~UFGLixq)XqjO!p=mw@%q6(r3Wziy{F>8IQgq9ppmK|L>_=dLUh?FRV?k;+P$+;6kGG(5u zf0Hn?v+JZJAh2b$Wph5Q3>m(|8Fd9Jq)%X4M}ikZqORc>nnoaP25}`k0I!3p)tfDD z5Gf#B>bJ@KdWJ)(# z6=yCA3laPS4&@=7a{}xKs_+nP4I&lrNtR1``}TQsR6@Cd2FO(V^-My>A>TvXk3Z!p z)vQ}MW@7exdihkWA%fB-Y^7ugr`PXq?tA|9hR61%{oUs2FC<2Jo>Q-6p=+_1whB#pI9^Fy+EB{J1FZwDQ zIl-%_6WpD4tSypGv*8Z&MvQ@~L8Oly98Lo`0I=R`mL^kvp2_JD*W!YZNth?rA3w*u zO0#t3B7tT-I7fVreL*{a z{{@=6g23*EBq*!8Y}sTO&S)lRJh?(zy_uSyYw?$q||6~b^;T9j$ z;^i=dsUn~~5rq&~%HsO;Xw7L(*)A$Aihyhl93-FPC=B%VzrYb5*+9isE~wyGPt;VM z%&?HUfn<(qwAUv;VY-yBJiDZ`%iG|Hpp~S4e=Z)%%9Ncxt^5n#C}1Gu^&-_g-pMvE zx;+zUTT{!wDoroL{fi_+du<|znljrF%>hfR?{*6z5&&JO#JF*mN8TZuziT{Rbmn~*2F)t( z!5B!*COQzb;YP0?D$*x&jQExMKX~ano|(%%;dN8szkwAgOB5N=e`j{Ka7`)xkIYWV z4Ay@|GyfT+5%|AAnBS%AZ)h>|bkv{CXrAqV9X{Ws^3uG8?yqKD?x!`ESHBbXwnnz2 z+(u-{dw7_hE;OINYxazQp~kt3?IJJF-#r3unb2EXKX-jr0q&dnwD7{Q(0fKMp6sZ( z3b^S$Kq|zi+qJrC7-+8imS5)g-F9Aov~1oexb{AJokUqDwPd;iZ@{$CJ;$loX;>I8Lb_{@FJ3^)lfoRcFh}6&&_eMORD6FN8X*zclmg z#?Y&1xV-0kzLc0Uq#88uyV-m!z_Q?+*qw~R)4X`#26Df}q~sU2ErTu(-CL4cK0V*+ zE#0j8>sNZNTl8Is)hNRpOLzta_huAX?**0LdZ)UAPAl(Y5B?5Y#EJA7diAIdEpZ96 z+vW(y-KZCQ9)s@GBPvmd%o>3dvBj0IMTOt;RBPh=YUw{JGS2(y_;*v2y||8la`r7R zUfTd3wBU1)WY(dda(RM5!&9H$kbU~WvAA6{dU>D!deKsLZeuBCr8u85id$I0z&XO1 zNde=Lrh#IYf{GuAB%stRr!;B?GcCEyPK%rlL6EYw6nGi*$Iift6X^KbB`WMzW36JBZX`; zs3@Krc(*S-vMM@9Cm~92w1h>mEEn)Q`RpDFlo7y^;iHp7X5=O$jf<8qJA=YCx7;qCVVn8b$Ejvi^z*H8 zeQJD9@yz5jKF(N~mINfH?{)bAnth?ltlo7!AkDb$w~hAqAcAno+W-}ws!=avsu@{g zZt09Iq?z}P?rA)${ETnQ$JD616OL^cf;QyGvG3WmOXD{oK34V9y{lrC(G+APi%t@v zrIX-y4~6_Y&#;cpebY#ZO%>@O|Tc$(=8llrI5ps)jec)5!4~N{$Aa_2N1+2FnMsh=KxS!tls>{Tk$R z`%Wc#@`GASHF35d-^;7|9$pF86lru=ZF@mG*M9-OVJezGHV_0)uI|qS8?&fTGXU`# z3zL7ZGD-1KstwAjzaO%Jsg6$~nbysmjl}n1>Z0T&U)rHiP1@O^cJxK5()%yXn@r_Q zXWFYfAD3b$?YYS}C{Lw^_&p{-v5cVb4fU?*x%93fou}N;YG{-@e@*W>ry~7a%X6e~ zDK+juE!%{dG1DThH|AuxQ@~<@r+->96RdGd%lO^(xxeT^sN;jH2uJ-#!&Gt7>pc`i z)o<%1ZSuD|9-jkVPVy+RDQI%gW2@UdopR%Xg#2K@3w0V}DVZ3CXY~5X@RXd~ z@W)zHe`=4Kd-Q-qR-G*tBRo3O-=M18*0^gfgW+?S^>kLMHZ>ubjlsVow5!e6i{4EU ze@;hTD){s_p)D|u2qD81{09%v838y0xcM`MT4?fNUsFi#>$pXabPzC`=H z|JPiFEo?NgLRckrbKP5Nl9%@*fUpF4i|VYre|MOUQ{|rPHg8bJ>JRy;*JjlM*Zi78 zeL~ECl1ESyy;t;c!93@Tmn&IsSL%LLzY%Yu=Y)YBPro zTY-IV74GUpWh`~Itlf?n*sD@E1D!c&@qn~Q#=;bI(B z$0DI$->>W*c-7iV2fPy}ou#|!2YA3)DgqnpjG>pU#`ptK33&sT0n*6vP6WA=0drh-rQLVKTP$9+$t18-d`qHjtJGZe(R{>Kxi3y2#1Y0gW*>9X=e#xBd%poZUxpw8JyMzP!Y3)g8s! zR^RKV)5ba6C2D^lvs(IX^>n~ot4)B#$bIdchj#CO=In^{iQ%21J*Zx&-pKxQ-?ynr zCH+=SKs~7v9$v45myB*Q07nvuw`BfmtgWDDBjL!GxI~ao>LTx(Vf=Z^@~%eI3;$Hx52XbBx=)!O2&o*z}Tm70E>P_Dw$uco*mbay`Z?J zY>;fnj~~g1u85lA!=A>zk>y|~*^4*Am&mnh*z5XtcWt9lP{yD+V^KD_Nnp|9LHJ6j zn9bJ7n$CvBZsc6$h0+L&g6a<&@zW%}2H##y;=2w<0K$5^>Ba`X{Cvk4eLA(-dwYBJ z_4R!)r(iRKCuulFTY2ge549s;CmDVG93`(On`8AD=UWWkbjc1CRIBKiZ-kaQR7=UF z10kYI=4(7>S=+xnqwtJ_q;ES?;P>7*#N$xA^S{>9OIa{OBZMro_?xuDR(_TV(vhM6b- zs>-FNw`I2s3)9uR=2L(?b_H2MsOSFH=Q={`ru1vd7bs_E>S{N0?Q#4qcIrOHAMP6& zfxYdYAiju?A`rY12U0pu6*icnQb@RFTZ%78D zk;{<9B|*imB0av=NxgQZ>_6W*=c$*AANF3d87gX$tn6#WQXc#R3fNu0I7I@eB0Dbm zN~;)5@N&yI8lOK7Rh->gUdC=v+F3d`XscH95+Y$IARt0ys?egLqWyuTB`FeFSO`B7 zo+FdiFR(|iQ-Cd6c5RV?@SU4)uhc|wQeHy>BzB-Ef*KWd|J`C~F*`_>pkJeLD$c~t zDcaVo`i}#ioNEP`*WZB(2VBYCN4VSas85q`2N+^ zj}XC^;6F_c-M+uDLg0EH(R9hK2Y-K{-}pY_JqOUmf`t*$ zARwxJM#>F%EDpa~Z4GeqhC&7;y8|x5*CYM!SMl`7r3E+sZ|Fo3C1xKkmxW>T^GtHW zhqCiox5(vwKY{7&Jo$#{r-5GWn}6nBWA?ts0OCRmUI}H^g%;7u(NzTn8g9Rq;x2iX z+BN>#=l5*O&vMQmgUx>xsvo}1Y8bl{oJm=Qn8h!zg=?($x~nAlef_{=^a9WkkU1H% z3C}O=dsf$*Wa~azZDK(5A!w@ZA9PLCdnAQPL|%FQk6d@#O+q1f)6>{@h*~a5)}@o< zc&DXWtJ=MmQ{D1C4V!fIo-OMjX?Vc%=1EXHT!!$>-$p9;Z2i@1Q#CF)O~vSCCa-3g zbxJnobwz2*Gf9kl6hFLfkVM$~+a3;mNS29R`EOEIDaXdIL0TDXV|neIEPVBjeACVIMnflGRyzO}uK=O>Ci(S1Az1A4DLo^H6Lz1c&H*~a zNQR=^?091AhJBwhYJOBNf;%Y@hIHhl>~>f^`SbO3RA{u@9aMgU$56X*08|ElLY~Dp zE(8UoycmPn6}yVLOa9n3Bhm1j`j7-rDOHmQ^0=X*X2W#yi zi?CBSw?eP%$}{inX9xyB$i1mm{S7h}=(HpWkq#Fvg+r(Hm(Qb{AUXfDR2AI2Dy%$V zTvPZHvsDNXiCk2Nf&0(mB^N6_N>x!Wz{-SVtTq`~C0X2^@n<})6jq$*Ho;BB{7XK7 zKNj2J$P9)lX)sup$?l4E8Km+)IouqW+|2i`{vIdm{?+Xc8Vfg2|D!c(on!&RLw88s z$Uhv;|NY@(-g8W=j67}+pf$>+;*>)HcHE!SyjO?Rh~^kgoW84+&t)bc<=rGoN9M%q z;9It;?v(9#*y{f{B#q@Xs z@h2^EmyGH6`V#Q}%yVT<0yH50u}}^pzQ;YZtE1!B%I^kOSa(am79wF$g0$7x__n_~ z4G9n#m6r?G9U3YI$W4OV>$Qm0pZ!GAOb~qbaDO`iSWYi&pH}Z7hsx>9%*-c%O2qG1 z;F9UvZvvibc6@YriTzmn!@3aw53t>Jh}%ZTN6$UG>d`$&MHdTY=p4wKryjgJ0jeKGN> z8BICi{nXn^+NGkZmTl}?7{<;0iea;2kw&y|7{Bjs;W`<2WrQY?`wR_gbL;7rH%F<^ z;HD9MzPYFALTv6cos6zWD|4mGdoWixzOBH_MSbS+#kAgi;WKu02VHHEYfpAsbgpWM zJYw+a->(hgy3t)a|KVmXS&!+Pd&G$!YmOkxr7{%G*#{(Gm+wQK|7JTRj8w zA=mwTr%OjIhNHUUViXMsD(q8@eDI!sC7(Qd`7ZcJf{|*IoS(0=v3akJoGvysMPM8lGRs_E_I60JgcnZRHxQ*TsJO~FWgL#mhZqyrvI=n9=hVz zBJ^B!CRM0$<@YTD3V&LJ@7b-m6AN)ze|4=1!5|)gMNk=; zrqWn+jP!a7!Z?S5`LcxT#|&^UaC6P88eDcU>9Z>fvALu(^Vrw@T;!`KW{}K64g0_Q!1Esc~sf?w8ITN{IyvJO`d6lOI58Z;bFVYZszNDvm z^_q59Y{N!zweEzzm6>6bl-dbG%SZ%X4FnN9`~fMT^R3@ zXYHdQPTGepp5C~umSQ)i@ z;%Sxh8$}?jFmqB;HLLe>@t;0IuVnM#cCMiF;1km2m>XUei0+|&Pg~LT&L&Wb_KB{m z32Jk0%*A`Lm!(Y8M}Ad`d;?V`zv%XkUE2|Fc9EP5qP~FDqYX$ia46}bXJW{Ku?Jk4 zU;yE{H}ja#qPZs4A~R%7$;48GF%<@WmfGZD8K&B};!rGc=4ApWGWr0n3CT`pWkR-; z3l*~IzpuUA5amAJ|LRB`o37Y?Ls&x!ubHnpxXSdSnuiK+du3_ifRg*6Xw}H-y0k+t zGY@9IPLY{3t774u z-2F({EoKNXS4V2aE)!XwnoG0;D@#Ir+1mGo&4TiBNSAzkS7i3rYVzcD^eYew!R}2- z+In)c1czfg;C=E%0ggW+l-~p{^scsAeEJXANtG0aK%mI7fT-Cr7^J&YDp<7t+(B99 zs{($*_pag^J-bzbod6vni7)#iI-Be);5i{hm*6Zu4r1j@xk91rr<2Xb zXW`WRC;o2nP>G+N`YZz`G{?J?f}^AoUIzP~BWAMt#|Yl-4aGynTSWI4;zxu9K_WtMO z^M|72PNzJA-)7Y*b{Y3CIQKq;{}t~|_-RGvQaB_YCq8K1^Gza5l_5fXN{2ox_s`uC z>9>Xak*GR?+!0)bIG+P6EV;Ks>cd3trQ#Hw*9R7yqa+5XgX!7=lCF~d;m#ioBT6Eb zriQ1Iru7*MLsb8xs}~s*M=ChG`FRWq@?awFz>co(m7i?r?wBT?t5aexNm5GHt9p2i zKL(v=rZ!r-%%SqV;_b_{D!}yO@hK@O+Em>wH|bSW%U*O7M{w#_Z+l*%+BX``kYy{x zDs-I;7omi8NU`};P*RADNOA29mB}yON8Gw3)I>H@zvMj2Lvxm_sfk(6uIu%EDe5Vn zw8D-@r4$Ft6!Z}SJ+Sf1uXcoVcspg=%2i@naqsA_J^u@5?JSd*CA~2NzGwUWXoM0# zAOF%as^3O^~Fzp|(hvUmXFg zkv`uMnn4ljg`4Q=2d(|74h`--q^B+$9e*ut6glV+pYwCp&7Pg9iRlStTb_w(f0Q1* zr2~;k-z8vZdICl(J@NFOIj0UEb<~o;Q`71(%NjMV(9LvhAG;QT3woQL*CWPweJhm4 zdGo_e_?yS+a?BQ+IRaW2e}W!HOJ?Z`!h5(pA!fNvJL^j`OA zE5CVw6XPVfaa~E5JLo(@-Q)ukh#4q0&B};YhJ7{Sf*KN`RZ&OqCrbkI2_U(T$9eQp zd-x5M1JJT>LVZ-bw0J1vgogO>DL1_!uP@kdY5LU)drkt9UDkgKf;qQFt#mV@k>)$7 zUi$WRRLDO0cSq!q2w)CXTc7ezgbPTlc^mJ7XcUF;oPcXxvhP@SyCQV)A&_eER>`k= zwPdKOlfB}@#2Qp8#7`-ua!l<&xt|Ycb9jS-qvZdVsI#^V0(pmc;1hAv@t}nnX;bpm zuEj^YA_royi=j4semZxz+yK!4@bcg;?{JsW$wm(P`dY+T5r#UN~8k{uH9%6TEEC@Zt}B zL1PFPwk4-~C8EfTa4!gsM2K$zVXuIHX2ZEWueFDx6+{ibE%am6&EfeGx244>PRe!t z5zZ3yB(|=ZO)3=4g`)3EQ<<%#AJl;{5i8!yhxRt^%&yEE4WefgN%&Griq%IHmcNc#@ zu#s>XQzQ39lKag#_8CK2r5@Dpu-?rCstu%A=NK3Wxrs$WD~j8s1?<>FzwPeg4*E?F z&m!@67Z{E@m*v;I+9z%8-C5+pvA|+>hNSgJD9&U4c1pd%VXp$A`d#Xu(!2(>u+vrn z!QTQYgeUPEW|RP7c4SP?-5#^YA$Iykn0T>pGhP(aje?;CaT{>#bq5xB?z^bR8*_C$ zZiqpHSI|}?#<|atAuYNdrhXu8k6>;%i$;n0NuQjhUmUBJvCy;G%^{2Kq0>6vvOf+0 zlfu9t{`~&+Lr5<)d5y#Z!iWF-TQ7*#?MWg0aE}!~jw{Q04qpt+>9!Y#{J=h5`z6e4 z<3eNM*n_h7RLu^-5EF3vc!lQJC*N!H<0TqaqkIN`T0qxbCt1c{{%4P;<$2S{ic%N{ zLE6R4G~McWLFxY6 z>lDAqLo)3&-3j&PLgO{0c@~EICDNN%iBnu>THr{(Bnh|CH%i8k%p_qi24xv{*o}@4 z$B&^!bFMhJN}?CbI=v5#nT^nc8E9D?FY0RNh7$>88@no?=?*dqh5ezf0$_LPGBMx7 ziYd)Fe~OeYs$OjGhRPD?l_aB)+o1LNbJ>^k%}=fKNO=S@X$xP5Jc=+Fhovqm#I``F z{10LDL!RBL^-ki{5q2y8*YjSwNPz)zCtM225f`xo#2RtvLyTg>ERV?Wso{*ZHd#4n zr0o>`NW6f>i`$+7gacv$e!;l^#CxsYF%ilrGdm@H-dtxIGKReYk~n@JExy3TCnt+9 zgQv_JgNFClQeBltt6WRr1i1sh;q@E(`gGe`^aU^VyXdjZgH!TfuPHLh<%F{FTNvH{ zhSVDV=YIwa|BK1|f02{F{ma}~R`|J+>RCE!xHZ83QKd{b%$vW(4w5RF1}k&_RO<-i zZ#jD85bWL zR4V*w#WB8-4<9@IwI(_4SYVOO+h+7vO?5-YZ!d*&sd?z^E+_J9)hYYvg>E&*dC&Rm zZJ=$8ru8Fo?IlouE&lRnKhQW4mNQ5X0!X`5%Md-A1S%X zBX65G+*&S{XQT|K0;%`CFY_fA?GDugP%Pp8=|(;o9A_Sg%1DL~O${w?-!vrZkty4~ zg^m|A8x`GcllUv0KANUg&Kr{$s+E-P)v{{7Wn}d>8cUl|w$-Ve?c~jSy&|l`Zwkw- zDIuKaX+OO1PWDUy@rLK1T{`bt^QVPZ4{uv}IK1R%;cWHGA3M1))@jH%&GC1>?OSL1 z{z$oL$|Va&edhe#TYR}AP&W5As&aq=gjuONW@L|!r<{Xz&y2i;Cp>Ho%1vesw93nY z+D(NlZz)(#G^{)6nU<@z226;bYyPypI6Y3?E^E$p-EIF?nNoqrL6-<++?~*z$KrO( z>84rO0Tp&T$=HTGy7_?nT85;5Q#3$E)~%X_`JRO(dL@rWeLI?)3<$JX9mBrYUY-7Q zzWUxWZwzH8;r$w~+7!oV&%w1!fe-!jkuJ+;T3QMG+-?~D(zuQCEZ}H(%CX9+FJ17g za2aJW{Hpb}#^?i{)&c|Zhpgmj+jFH0$}cy3cqpXblqRpS?HbV25xjc-&8dy-(FE-FiR{$OX?j|41^N=I#qS^ucVYu4vV6jis~d9F-< zUh3CJXxa$2LSlc)F7E3@jh zR9#2?*U_!P;^Dl~sVv$NPuund)P{Z*7?Acfg%lfkh!XcS3-Plt6N!HX>_3)LJN*mw zi==505xM!cLYVqgfNHEZ)@HiNT^$j zyFkLMjUhm$xU+IQ1Y8sizX;gSQ8OL#}inB(qvzzb>3q(5kqr4kyh zp2PF&YMv|-HZpMjldJ-!v3KrIsmT_)#~-|O|d+HZYSeD_Rfv7J|j zYl905k1QzT+aVJ|&nam^x*ujZpZxhmjV8HV>({D6b^>qqR(Qk$GNdVeV3J`4=Xc|L zB+Pm`_4e?aD%a8jec{ra<`7J)Pta8R+AE{QV$%0AFjU_+w%-wK#U`!*m} zuh3<(KusU|b$~EHsT8I?AXPFAjiDj3t@2-CIYwVJDE!5#h#vh;xLbUVJ7{Bd6EID{ zczFZvpNmaOF_7oHOsOIgJeoQrGE_K{RFo9!>8*EkAgjq9RIuE>|M}s1WrFaL-L-4C zvUM?9FAa^ocuF;&bINJVXBI~ME?o8z(t{a*gU2`8ZMwmQ8-|02%Peg3r^218Y=m)c z4KZg%UeR2urtI7?xK&G-TdYGL9~w1K1jDd+_YV~ve;|mB^R3nv8%piJ^xR?NfGYti zbubGAZJfFe+I|?SJwu~-gG4~2248j0yWk0Z?GkgPN}4Z&X2xUz#d9A|?5lBYU}Iri z;rHW==WYYuG5#ompHPP$3ruCDgEG#!^8R}K)|LFu7BZOjynaidYZgNAG=CK>vG zG$!7&@SINiAT~c>DzcKP2A=$EFv;QH(XekKY`|m~7@Fz83`BLuyZy!5uj|E0WwKex zxTgyqaufDx9|@==!pTbpdNo{1(_Bs~XOPky82u}3OZr%)D~h@ZOGAW_vZKf=*&%q+ zJOGEq)=A#JVr|n^cFr&OAjZLbYa-1rO0J<%RDOKi?ak^3S zH&YQs#Ejoo^0+kLOeqO3_BD zE~5=fFZXH$u=v>D@quhV4QwzY$zX~A&2B^>t=nBk_|zt5>m5-#C z`kEZ>)}IxOU9=#NpU0XmdOIPSs`o^R(3Dh_54&bByQ+@)USidTzI@&ws9XSAnR9Y< zUIA%W_Oi0IVXTC!KFa0zPWJ3sdsLL#rQ-K@H!q|J@GNE0zB)7(Hg_NXNOCr@{=tP; z=c{vOZMlDpSA>&|!Wv_pX>U z*-jF~kJ?%K#`F~TlrFXR_Up8DIq|yBm*)~2sK?9330X|cih9?`O9>z2U~Z!P@sxGp zl?Z+#2ge@GHl^l?VkQAbZ!N;XHCKHA(R817X70v)eOwmp4PfNd16HmBKXZJEcb)qP zPLP>IgnGTz5rwzCd5)5Sq*d$))Qwks9ltgFuxN7>QMYFo6OtSf@ZaJv7=sOmRLfw) zt@Q(M1ZF>PA>`_}2wTpHKI7U2b0w0Frpzs7kjJXOtoV+?u*|LrOy0hCKU%bL`@8f>ieckowq;#cQVnht zv|GeIOBaI)Hs#kv1u27Vb@ZW+?g&gBsim{KsT-){s3fs2k<)#7u0EgJ;C)m9u>PrDZQqz#qz3zFCf!Z%iNh%jvbM)`C3 zqy&3oa06x~I-Wuz%m}1uaMWDu=kbleRlbJridu0ZdT$Ml0~tPI4?yqSEwsA}Q zND??wS-{)FA1BXtVZ#6hySO5TK0*#?6;l4{adOF(hE)MF24~Y0;f+D)y9^8DI4kICE2m-<9v$Iy> zym#Y}RLZl`N-rDZ(^Mn)2=_xRVAaVCU9oq%pBqd#~Z7EDfb+k=7npTvuBO&@J zJB>WhBlTE1Qh+RZb;lcoS!F*RDG*NE?9JlA?phlmcQ~~fO{2IpFvX%3mT?bRo!_Wr z_0~;HJos?*Q@k<)FN2c*&L1f>9xO9zWKxO+cp-d96kD@ebhK8?jG>$3?xuh!4Ud) z%4k*-i6TI}^O~cmn6cB34hl*kk~3@w#r!YMAENa& z?2_^?YE;+X@efi=!$21cO_=N&ITC{@bNNc}h3msNY_2dP;sgh1F{sd?)V;zLIkXKv z3U=s9WMOyxMJOTZlc7_j>3MM=*xZ-D>Y0a{xpGorLasbFI?aPD-6 z42!3(@S8}YCb4=Vp0m3>NoCD(-AMp)x2r1TP>XJO*|3YBlOa0cFFt~IUXy@r8p(M8 zU=()j)6-*a~m! zm&U-W&1I#8!i_r#fXH>mJ&ybf(Ib!>aJ#7EvA6%WJj600uao}WnqGRKY8h}_AiGO1 z4y>ELfz7Eihnvz5;}*FNgzi;mkq*w}J=c|AN=R>paLvxdEzdlR+ln(%Tnn3f^9`OP z^B(@lkmx-Jo|@O@`(cv-Q{kBWPTBp%Vy%}pZ&KhV|Ni1joT+ILRrol>Wca^T;ul*4 z{$FI`|2Mc&l$kHQzGQFrxY-u9+g!0M_yoW4_v}HEdoj_j3>7bad~JqhX)ED)>#)dl zr)uBo_gK{G>oLc!ytZ~5U#kdg%1>H|E4|iSZs_xS zh)&*(o(w2e%>GlrWEc0QtoON~wjH7ReR87XyWtLrtcy+6?}midCvwGqiiU62$=>B_ zMa{|cvKV>^JTu<4*#&u4xBR}7neN~@n;;wRg1+gqFL+?Y%G2YM@(A``$QI;_+0etx zH@wUckf#r}gPi#{UclERByq7CZ0a(LyyKTcI ztfn(1ARN0C6lCW4MfY6ZWwQ#(<;T&%Y9~r$jG6T~Fqg8(z^Lw|KAnd7?k-a-;8T-Q zy&3CipLw?BDC)$kkEysdl1R}r(Mzq5v+e^Z3z49B9gg%?)?a!qU|4_H)qf__!sU-YnXY93Dz zCa&zU$qGWJ4159d?-B+LQDfj%9Ia|gs{=)j1FNvq<=#%e5Za33sbyJ*uB&cBSERpS z@H%LSUK4D`X6kH5Y$sscrXNdz^U|L;2vaoWtDd*|XBo?QKIpt>)E$4~bKAp?=2gt( z5u=w|9NeZ|atlZNFs&bk95=1?>z~XOuENd;Y02H$&yo!FYGBqi5Fg>{cwIJ39^ta( z$0-tc=JVd)`c~4BSGs7L@Ym*Tb)>1w*IGJU&a{LqZ4?Ntry3O{w!U+}6*m+XKE=Kx z{%}!L#ALC-!!s7s+IY5>xI)XeWuPg!gJA<9X*n)xeOajMiTVZ`{<@z7V>3Bgm%DpkM-GAx;4X(093raLxMR zSYWZEy!G`Z_uie$>{Of$BW)Ie*m7AjQwi$*+C#&0ImDbx@GQ=vfCoy=yG5o z9kBDq9(Bks4A^PpxSO!nOXurG-k{D*;YsMfp1m<&7~MnJ9kj@@*(`J$l*lQb1yrX; zVyBymz3I+Bto>^*v}M2em>?-nyKB)=kfHh0K851KTj7)(KZp+ozyi$cT3KSzHDYXS z5l5uq!VHDvf z!4oro#}{;4a9*@GGR`W*nz2%n0@g96dXre)OcM`vf1lMSst0_z#88sQ_UV8(+ zg1(h5t&kKJR?Ba#fUVGe$rO#Ye=uEyaCJzs*V z3V7oDf$Gi91;_(2*nLMWCO_4_Q{!g317hcXY_AM^>fyT@f_tG#BJpPlhfj=yWyeO) z=%Rka711kqmAQqbEtH6fX9Jp;MDxhaL&nHu5`iD&#BsI6olEUCI|5jzO>%2rgF~$l zkPaIvionhp& zVI#U!EL`i5DA^A$N?4&jaF>p&WRf{xZs?Jr-fQyb=AVN7&?#QGx0Ex;%(s>s>Iq|_ z-IN6=<~3M(5_sTbdpiq9N!R?s!is59Qry$QIH6$bfmu5r4Rwt^jZ&(~udUV->Ko<8tWG zQMe`0bBbcSUb1HBG4E8uIdf^K=Rc=%a-=t z`VHYa_xdwX-C^P61&UZoP&hbaJ@+w# zTjK?+OviJ#!G+Bgpu8vGvndr1X%58;yfFqC9!&iAyEK+vmB{68d~#m63uD~{i|rOT zLe51NC(#=3lac{~gWZO;xcGkZoI z%+YcgNvxm{uA)Vv^AZ`~<}~YpcNiMq)i^xq$04&EQ@d+glg?WY-~H-pn73@Nde8ec z&<~Rm|4Rd*Hgv+|DsQ4a5^CkmS`q(<0!AV5i4*DnhZfiH) zlo<5~`wYR2!=VasEr8K#pu=tg02&QaA{@xZPk8SHdcTB)Mlct@QYG67z{Ty9J$j@R z5L|6ygNj}=HN${6K;$_xq}}3+|5{*s?Hs&21g1bT_Ui}Auh`EWs3Ru@JSFLM#fb3b7cD6M` z@&_He%9N2HJYWhCzk3w}0VU6jWaz)3HpCsCUHPse4x{aO9~l`TiBUvMEQ%$0s6-z=QM z(Gq2_J7fMAZkT_S07&!#=ae~ngb_dNK%Ce2hLptLLva5DRKKCH%YSuL zlq7GGM~8~o;l4q#_z=rBqAdAr67iiRvM!3H`LM?5jlI_;*z{Z>_0PU0->Uq!|NV1b zCfbMq#~#;ih#@94+?woMhG6>;h5cfM21p@{o%I^)X#Y<)J4~#{oV+fa3=9l#4Lyjp z6%I=rBAK$jZslBpoxl}D62VBlA8^8vt?`XjZlGXE+5b7S0!nH&A8LNQvC^FnuBy8r z_ZCJH8JJjXTR5Ht^5YwcX=HssgD&Vy^_8RhBp09s|55u0_xnl3Ks~gKL5(XMMcRcu z=QI#WPw)l8j7^VR>|wLRo(}oZ^flB{xjaNIy6*fAyy>YUUQ@xK487$OaK|))oMR3) z>}`OdW}d*pFomkkOt!tAqR-Gm=z!EkKpUXg)S6T)zb|g+|Il>SaZR@G`=Ju8dQ)jk&uoF2&0+QNCD|Ck&qPW8q(b$Aw4>!B}R?-U3)&i??1jCeQ+{%?cRMJ z=leL0N=uUe4oEYIxE$D$y-pV(ev21Xv;gfDMn z-fTLCt_d+wniqF`9ox~2FC4HiMGTBxrT9XYjMJHL}6Hu;fjB>t-hl-Kwys;j^Tj0H%PqT5*3?nhrW zdrg>$P|gv1o}_lcs{!stPeQj!E!;uUAjc@*4o+v(-y-rYdRHhlJ#Lj`UD{LiA|J3SIi;WL`%P^ou`gdbt z_n)jIXTK?ZYjqo9r~|fhlu34%JeYNMpbb_Q8;59c~R{otH&CzDH`D zQz=G9JOB2J1=e@1QS3F>MCGUhKMOTIV-IP;pZQ11tdz_tYo_Tmkx!3Lo}-UAI1+?j z^F*aFELpDefgePAd`+X-Qh=#WoCc(FhG?aunBd34Gq+2%E+`)S>92-F6m8j;_1dZx7tPk6FN{YalFwkE`b4hTmACagp;Xge?l#E@<+yrgcro6 z{a}Wlc>m+_me8U-fe#Q0oJW#6sk#B>(6Q`nrac;V1()>yPD-9qt0zrgM>^D~BLbO3d% z$$URfK(gityuJGZxBkQc1m>P9x4=F%ZK;Lk~H^3t+@pv85uX5#rW z2pGmyw&L`uE}_)akvB)Hf|19O5Dr3pwzx8-v3rvr7?fPgK-3KE*Dheyn8EH9qI|!VUon?)eCfH^C5J_B=eF!7goefB>N%dU&vpWeO_S&P3fA6!GAx*d{GRnOy2^tQ>gjWpLq%oQq^p1s}XIfYxPU`FEd1w9wkJbjY; zC5wMWb>EH2UKTELA;j}deDERN#`uDq*&qQ9f=eZEs(ye^SLr>w-5~dK0?$|to&dztN(?IGpi$eh-OG!=f3Z6 z`*g30)lKhH)5&^LrPFxE6Up}21f9%Z=J|8^UL0}EX%G*)lo!K#d{ZcWtp%a}JKHba zdD$f)8??VD#&78qZn|(jpJj3&ce`PxFLy*)=u*Q(D)8jgz$vuLD11bVxHXB-pj31& z-J%oo3=GW7XI3g;&3bJ8B9tpsF7}21w-GmT=-)qY1Ww)KSB#f~#IGRB60=p-)HI3b z2E=x5d2tDrd94cuwNINZrjkVF&{Jo9Yu=2MVmI6L`sqcVLJgRgRR2d1}?~zc2?| zHrqz?FDcOFe=09u(_nea)N4b^qH3p67mv+2ZZf~3yT7|vnUMYRYAB1Nbbx9*U9h+R z@gH+tv52hWuE5kie)92Z5*_2C(dwg4X9~BuVA9Qv%ejiRT+8nA-a3S4yrR^yOxltq z;~2YV=a;3h!=7KtpOTNUptx@o?5mgXin!lCI%KTOHZ0#x*%d6^eSq+MfY%u;ABj`5 zTa98<>J}EfHgck%e)rvQsZ#TYcoTXip$C5xR-yOrlUS2V`lypghV|VT@qm=TRQ%>6 zAES1h2OBMcf)z)tL%y&*8Zd`K`oZKr8T0(kNH~S+ ztJAd~=BVTq;V(mdpx!xa89jORS@(iYe62mQli+9-M&y2;<4yQIjXH4<=3VK4`nr^y z$Y(_Pk|kewV=cjH(MNiJZ&`BNwmqrQ-PoHqmNi;6%EP3eldm*8YiyLvKE*gABA0$S zhy-RhRpGR>kMP1WWOR3Jez4<`ESd7!LnxkXEty@d4+-1RBXI}*5CQ!+p?z{!II(3g ziAdT+P+gKDczC?i;GN>`!ugQ|T(o!5%FK<`%!R*$I2ncAu}4K$9~aK?&$FyP2=Yka zB0_~ypeb@MPaxSbhDxlQf1^n!bLp#=jRGUcpLpeP)+4CPpMkkJPTk z)nNY2Q1<8dl!vxf6c3>(TI1tp%Vd`-Nl$5IHA`e%uhxtwqO;*}s;n?1Pw z;i4Hv>eELNs{U{hO`AQmwYY4wP+&u7t9ui;jJ4NSbXmP zSMohiBQ-}ulS^V|-4e-R=dx3AfcD^~rMIrhM!0elZV^ZS_0*t~?p`S(yeBHQD-L&d zZPXm$wv+wc(5NGt`Df!%pt~a{Vf~Xkj{y1bHov(}ozev_M|s}nuoK(mP0Dk!1mT`X zw(W=}kZDoex=uMBC<35rqZNp5U+0M?Cm)$1>-7Vb^X4=RTH(n4(P(k?iAq&$r^sQd=)@JE-}?$xJK zsEciWWKc9dv=LBUU>R$rnw?hyQXUG9#)CANxOQ)r+9Ka^ zLPV*4XKk#-1>Adt2UKN8uQv5k{H@{%Pap)YNahH4QCECdYWd-HZ1pP6~|B zzQVu@dXBPPtxqDmQ<(b-ooZm^*uD>gLZhGd!PaewcO`m{U z@jqO~@n!z~m!}~JXlCQ#cm{l~gwXqR95?!Njq7X9#_=r=zl%0+KT*|lO8N;ZS&waF zeW5C-q(O>?oTlPaQvucG7gt8)+u53&>9#W*%3VJh1E%gJ%b6rcj#x5z(IXX21Dy7U zf?npQpT`QK14UkD+jg5;E^{Ll9mO0U+99#*EkHC(q(oWOO(_+K2_3OtPg)2gV!Fpp zubTM0)SRS!RfBvBON_E*ArT)J{`HsUU7&sa4UwSk1=_ZAtTnL2xLdsG-{)0iO_N$% zM#S*^wbEp>^=#2(Z)J;&BU9hZqyR2ZH$~R`HcHn%q1GGA{Q>i5B7NwaV(%f5(=t8J zPvsHqo4ERC()dJ1(sY8353rY=ibyTE2|}1iaYE#MZ?h_9G2+RxFcah5{`}%K?DLlw zwsieMq_$-{Jg~CC@2I_H=iNCGu(5(K>|I=Gc-3c=Z2sIk{d)?x{3JPQ^`Wl5?q_;o z`p+i+Rcux7h;O1B?r-j(W!*LN=#CBRu)>=_g*;cHfPiD~7opa=DdFjTjrKeDcJnk3 zjNKpNPk*|20O`Fj1Zx5CMvq@Ovk zc*W*p-=6CkD`=OYzRp-~VcJR%JTz-{d?@}Mc`f4wL~2Tn=B)wb_6kRI2Q!WS8x_)I z-HvNiHhR!sDr}B9NUm?;8y=96Ooq}W4u`3crq(Zj<_IA6u$rg`6Zu-A<@+w87#}VO z5GS+W5FoPEi~Lnp_1Ul<_Fenw>7S3#*r)iiRA8roncy2mlK6=ZK$us)|7|iRW0PrR z9$wxqM(M+Ql5rp4aC%_41TF~RnEj}0q?s9ykx(#?%P!WmqChT-$tDCIkrixz8EOYL zJPhu@^W@1A(7g&)c4d(3;6E^7HGeX$U98f?U+c4sNO&A`fOM`b4!BKP6h|mVr2%Dq z1E3cCH}D2y{%u@#SB;z|AQJA()eE0Y^FJ%p7f%XP>@50AFX{E7DYj)*RP82!RQ5i2 z7~l=1T)5k%yJC~Cc^8atp{0NEWGc8eZL%lW+%d9XIRI?1*nSUpy-@+luBDS6u;N9zP`Wwd%{tnPH-vFKKyuP$ZbA4-UYl{oK@E1Sr zzYQl+QBs!KO!B^t@zIox7mTYoKR&^g0vU=oY=2pq0J9SGUc3 zIh}Du^Raq0EP{y3g*o+q784L5H}^v;R#<)*d&jo}VN-kn9M41;j-srh;=z7?W&aAx zdm9W`@}I#Ogw&t?;&{05=0jI27{WsU?*T1~GIf>u$l&!)=CafHQr4Ds)X1yO4RlNd zFPx_>gfQa2?SpqJQ~))p0l30DFD^v0IIhIQ_yoYx;jE3S$rq%&bCs}{5}CwS7M9fA z1q4et`c`u0kdS3L0ucl{3Yxn?Py%6bT!Pcd(7O`Nf<=t+0|SC+aji=lbLvw8x$=D+Yds-hMpc*@{hU^oni?c$oQnn9Jv;E#Z7L_qSNABhPRde1i|Oex)$6Bo1P-e}g{-L0i<S_Ji;%>FzAu&Y9 zS-#H&J9ro5{Z2|!1|bG`H|H5Q*7oU&2JoI0jGv;=2=%jogaw0oH67>jVas3iE%1ur z`X+(joy{#=8De!tvV$}~pxe%qvU3aZx-$a6%jZbgCjgg#*j8Q_4kCa+jpX%dGvkn3 z*Hb~AK0_3}gvYZ)!f0i!vzjn~BsoFU(KfH|&iYo8W-)!8l+J(BKdrP&~>#)H$ExQ^oOTx{|Ig z1(Yp)sr#S}scfhHpNH`rp_udpO#Ma_lCwKPW0DGQU8so)Fhl=|Xy7UqZ)YGFk+rHw z1bNT)J~iZc$miO57BDG}f;x$5BfP&~tPmM^pRfIwDj}C0;~6Mt^EAZ!@AqI#>?E?j zu}lqvc3@PtPMuP(!||C}qsqFvP~bMA!T>M6BgkG3lUc8c%JjgeL*h>p6DHPw>Qjmb z)g@N+q>`$izKLS!1m&Xly!Jg8UR8gzOFHgE%a|&gX0-9wnu$kTP=B_40@a|-js5KX zb;P4AI2=%%TPTl9X%Z8^MPJk!>Yje17ZOBG7Qzz0rDJ4b=C6X)lGi5p%s+P99%fZxBePWig8Wx_G$HX zPQIauGw41Wo?%&?vlG6aHgf=1(8rlp%cf6DWU&9*Cxod!#On=kaq9f+Q~#ieLvhb% zFUfnRR<`z7Y)T`*!9fvr*E{=C%lW3qI&Qwl357g=sgCbWo?UjmG8PYPyDwqip}JW`DxdcUJ5trpX@|9t1lE|ERvglDo1^_qE@HP#Z}^ux9et*z zLer1#D?ORG8}8?4-Q}TG&S3}qKX@zpDCou4GTFUV33M_ChQb@Re zgCZZKkYb(VCVg&the5@JO( zeM%7fBF?6{EDf>apZU!)!=LF(>FVL#__6xJD5t|)EP{h^8$V3oEZlc!6FCG)zV-zM zH@y2)92Y`C)rBuJ<45-vno|riMk23y>eJY)jCb-&eGo*sgA{??qfvcjR4J!)rM4<~ zaxIzoqfxCS{YaCv^cY2@cC$KICWKhN)iV54zV|$XyGBar6VRX zJOkSnP5-qcMN)4tayUw|3zhWRINgow5u)04jzO$Fio3}|9VRoQEYtXG0Pc)d|6usp zwPyh)W?OMW6|<2!V@rZ1z-#FyTuACunV(&8rWPpt%syVH~O zu(%9gR~SgGRoeExleFRIGV^J8ew)8F9b3wNJOW{qL(B+suP5n{w4=7DH$-8FC`0C& zRzO6eCYSqS5-vQ^n{`wUYOq!j`J}QG#;B`9p5VL>C(o4{y_K^&@h;!yRt=9=9j;s{ z0TVndvCxQiD{Ae+nhLU~!~xy8$fG!kC@Ypd-g<9cb5nEB%W8uxo7&~fj#O}dsftH^ zJJ)=sk2jQ$xMe&XBHyYhAav+lxR9Z4(#awoq2C2xKVwi2J<#=Wx9-M7H;xM z;CJ@`ys{B(7AI+ff$}RN-m5H>NiXkil24^tW4gN_=S$s1o6v!qb!F4_GvhZ5rR?LmQv9&#a$JLu!yaN|HoNY8ZoTEz0 zm{zCq6-iPec)h(E!6n|Txjyl1A&j76CU&yMGuvG(pw!A%3A&-qYXLu+S( z=~~5&*EkJ?#z^4hj+T$;*o~d7a(+*vl$}YutpB1$dhw*6{E6iX_r{i(tCY$&buw&g zhf$e}=UG`(0bgfJ;%}g@Y@^bc01kF_F54ex4Felf?ZppQ zQFY{%V*+k04AsvB@qJyRnJ;S_4cR$_2+Gq}Fq%^5?aWR}wKOkO4d$D9ahd&RaY%d&D;#S~#v1Hiwlj)ySvfZiScGW`t1t-KT>RwY* zKg)O`=OTh;%Mbn?ZUZ&*D@^N{WTOg*|9&GIRlUrW&1<(NeBu--(E^X>K7 z7OpZQuwwukxo*zRRv3K3hx6KsaS~cTZf6?OSoK28a`ZDp{BpcB&@M~dZ;VaoSH_f@ zzd4X$iA<`Qg3oTvJ>9UE=3hiqqZ6+i{*P$}Z$j)Y9_A6OEGZCBSMbgMc?0<6? zjDfPaL$G|(Bsp-e*8@I+%s7yyCOJz8RaW(9i-@O3@xo2VtkS&1!(oK0+rtP)1fN=B zQPh>0pjxxbBM3H2FD!T2h~^u)j0u|#xS`tK-Y(d!n&X)w+1%JEEhLKW_`qfr@0dQQ zzORY6gPUzkM&1kAn7Ej5nwCkH-eG4vT1lx&siU>KSu6xQmb4k*TC0)1~ z+lGz`WE^@2f{*q367Pt@M`TNr8c<Ckniw)taUtnmXI+_!onC7-lTT~ZR&RZr2J;yB|H8IIFNu2;QxYU|`x+Z< zbmJ9crPu@{ImEh%Px$OJ$3<%){zBQV6>f{8A;(`vH?#ha1-P@NXH$9m5mJ)NqS^O) znI&rJK(N=jHAjQF|E#Z4lC2E9Ql8?`6T&y8+DCuz`aZk%oz@{Y=x$Yo$(ZP=woI=# z2x)VDFcSyGSKyukR^KQitjC|%bEfq=Pg<}9X)JKESw6W9onplM5Q{)HT~+O@u6`u- zy2#%zDJ|{prU)U6%S?XG!0uoY6SVlyL=IToOy1?FB1T|VwZ(sZ`8nEEd3FxlzlD*D z5I|JiHy#nhGjFgrF)f;dyJ?O1q8(RY2NlB3A!9_i4#VPQf%_j!abQj~r~Qi>U_HxX z6xs@w%f3%=$ggZy)2k$j+=YGgU0d*)%L)U-D7zyl08`{FDlUF>dtmkVEeKX}9e@sY zc5I8fHTogczybnhBft#V7Ame8Lwn!T?GW~7x>0QV5iJlI z0-;o12k)kz;oEMsClOcK#kH&Sc-Ww&M4WS@?%Yb`o$C&!-J6B%3 z;ZwM`fNIXpR9->?1t7UW9GLE#{1H*Or_O!dpbCpwlev;R*lUz{y&aWCrP=R`&7jah zN%6pb9kpWx-zxV_A4S%3pZMWxsPV>A(?72f6zNREGJjo4GDSJSfgpY8-^fQ&ENpZy zbLr40J=KSbUYBK5m$jlq#>9PE?wVn>K@S=~6-i(l|2;xHakuZIGo>s6ZYr4FbqQpA zKUC68CL}dD=qEk3x)(kZAzMAfOD%%hfO&?Mu~Cu_iDl&2(WpO z*Oqc{K+J4OS#(}+K6Z!F$*6yEtf*Ao5FFy|sLgNL);hN26VEOy-f$bWZE5&KU$=LVjUF5&8 zwqu}I3(bD8cw0INatxP41Tw6LXC!|Ak+elddpl+^cQah1t6j@FY+}nF;`SL_j8!?d zso_!2f;rz2OcUtnm{$bA!ZXzbuWzK4UalH-uo4IMA0TDCEF4#)ArEkvS5%{fMMuln z0j~XUxDC9j7#3UFNs_Hj7C)QEEu9CPD@?ZB4PO-MmUJ$;s_16UnLVQT%b$yxxNc0X zU@1|L@6sPeoftUuD?YgA$FMxA`M69H2b|fv!2i;9c<8jl9&efM4aP^H&j7~%GZTMd zkFtrE4lnpl1;oU{0AN55Y^-36kNRIVn;V#0mdD@BHW?ax|1caOA4Vx^E8$_9mMqd#F?~^B> zTd5lsm*AxUUS$v$;ScaVCVnz=)p)PNaOS54X@IW*s?E=LM_rGKE2-?|Nt0!Rz&QVk zGw$DTo*@(yqFh*5sH?9ZhDy=F=dS~C(NpE-1X(@e+Zt>ETSEUMy!!jW_U^B``1y{D z=x@uaDj~8jK;Oyyv)OUj^9L7N0}2=bFNI+eRtkMJEatx?1cE86wXg3k2#-A3n+qEs zAAfm}Av6R0!v~wMgPjV2@{^XPvT3UBRd@T9v+D)$4zz)E=6FHJG2t$?P+M+s>_sNr zWTk*fV;>2OPP6@EDF*fL23W4B8vn}vm-hg!9wabnQGrF00xd2~q~bqd3L>4e`xJ2h z` z#l$cFwHGu2IZRXDP(BU(#U%i}V5GR?GXwD9_pT|2*x^(v3Pjp1XER=Lh^7&PPL>Cv z=+$NP!i}VFRfIajpN=S>OY=)$5P1%Gb^WZv3td(D2_=8L*TLSX+bO`6-m?H7iNJx52PYAOUmU z58!AW$bLEp9UYy6PCoDZr*of-xhOAeFA~E5i;D5PRlmO&uWF(m@SgQL8pPmT346a*h%uReI3^aeyLI*>;mnmLzkrs7c|sG9kW1ScnsKFrU2|OGKVJuz3RFPu*vDby!>~|G8LF@=^ zsn6B+LBocGy;bcQM#MSeTD(PNp#EO`;Vo}&an%^rocP-2)4Wu>KG#0)7{9NrwEWzd zl)UB>o90B3`-@Cg!0he-N?{g3dud5y;j{Rev#)P;%2-?w!p-M{CnYg0ENQr3Cmv!h z0vNkcFrzoMuQe|26P>(@Z{8q3vD9;J+X19^%51Mb62zZvXd-T7yg6W8nXv(BJhSgf zgVqrb81>?$&^OkL|8<*6o{C`6#$oASXZw2KUfyX?0Ux5xSIFj0X#DPRnkCRtjlt{S z2;x1v$gnO7;S9Z_qYvhUni!?9Lx15y8}qa9jmg#A zL_UOqs~vFU1Q-L`Quk7O@LMFj3FNk5@<}j>DCKm9-I1X{s#%E;UIvc{JXT>>nq6+R zh!7ZS)KRq#r*pzTU~TCH;wFvMr*^@TvknA*P;F%8$c*yr9tM6p2s7>g7ew2YlBWB) zGb8&(Ad3hDva8FjJWK*e*pj;lI6q;(cBze0(SG^DJG6;0f7wdzK`rWBiS5!(@ut? zPhK_}MePf0h;S#Y&W&d0AyxM7t?e@_n`8B6Wpp@_TbxrL^zwwxp5mc1BmSiu7hiE# zevZ@LGA_8h(~`63J6`HlB^2V8TXE3pkilCe6nA1&O4S?~f`oHG>;g>uwX2$BOwA^N z3wP6Hx%tRtvqEDgqS26fjWZ+gpq1rqK?K>Xcr6fSCg$vFO2;i20Jjb@l{JlEm5=c8*g?l6iX!Kc{naZ_m~gW`ycA-#tuPbWsMdC(V9>3|F=WJd4kL<2xKf>L zna>}5UaOed2vvK}uAoiIQrbUIUu`pM1vT?3wKO7r>0Ys0;>e=pBU=Ws%6P2CtC=TV z(PNgO!;I=?te+>l57tJHAb-OI0@{wA3MoI;C-XDr|B+1 zInUWl6I}>xVCIO^NwQ5I?l-5F@)E^~J>r5qU`XMQ62+eudWqfd zG1oMzbtH_Cqt9k*5)i~Btn{u z(%#=z&r@bMA{55%qlWId-*3#|P#?NnI+Ob7mP;z%{UZw@s0u_lR&xU4TDW?}ZymM( zM0Ge!GaWX}kjE>=Z;{rlY#>yT5+h`O3RU-XYCY;_vYAYgO`kQ~Ez%w=l&gLJ1h-+x zu1ZzApVht4tK>)phjW|9;%0Y2cIiM83%*qqStzCrUkZ zSe4p#hg|Jfdt@PDdd6W4S*`Ww-s+BPP(pKQm}BhTeYO>6u?c!*PaK)sK4(a2oBTbB`b7VL*H46gEurifhZ>9Su$X z+hY+uRJ=NU-yHkqxBh*+3wvX|Rn$Fs?X21dde8PWcq<}{phDre6~yOHKUy%4kieR~ z;IXe`Y9j&<@kZe=RQbZm{oGiBhn2t6sw43HYD4?<7s0zPPX?(;DZ6TEj?~p~v(tVe2)^2{9xGIBG`29K@ z7lwcd0wN<=G>p$=^J2x(aHYen`V-Oe5pP_;avN_j0P6Dwd7jff8^WRz9ss=X&$aA|30k@f1w zkhp|GW754i!-+SBI9HqYTDz#U;N6(rLwdETHw9A5+mBDn#Zl`@@|Bl)3nwQ5`?F0= z3RQc2^L3Gfg=*U_wo6m-I+52*?6`uQgD=zLP>P3Jba?u_JNXmqejChXhA_{{=nYGe z;s{=J`3R5L;L9%YiT0cPh!9o0is&}iI|JIE>2%?DQA5rZ*okVY2j<2&VRS&% zLeqAyTgz-k_CEJHtxwi{Er%$MW`^WmLy!iJK^NPw`Y&0>v{|7O1T2!WDZlKTX~+>| z|B$~(63yq?hf;2_|7$SI)vD-G&+!&3Cutu>Qv<+NKhI%+Gr zeWUwaJGbW29-S|q^cIJ0cnTWJt+>*$!IF9_H}=%k*|(Pv%e}`hOh|$KMBs0IFgL}~ z$5DP|hIR8{)cEF>xgNqIL%0;)&b;;rE{DG;UIp_BHrIL`<38??|0iVI6luX|+crEq z|7ayi^PJl&_09#PbW*ADO{(`+=%(QPY;m`Sp*=UrlwVlWr#|*_)X27BmQUf>?v7D{ zae~n+*-Ao+DFm8pxYe2@yv9PzQ8P7x;mP2x!Ok!Y%YGdYo&6a3P*3KL7YfH~Db%Dl=qC4IXDw zrksbjgSGY)-`zH5<&B?sbFo&2fV?@YQyDG0B>OZE*fp>5p`?@-Ui@2y(o)c!Rn5CW zS}c`V`mg)T*wNoAj64^9Ec(u>(VsZRn@Vl`B>*GsjtIx93K4{cDJ_|vuZJb}Ftga1 z$@LLLX0G>FN}%S4K(+ZKZ}HxDoJVuiCnKtV4FtGC2P~8$qYzh+R;SAe73w^l=kNp0 z+p+tl{L{Y+qska7!Zr6oy7_yW{`vi7MhJ1{?1lFHnW(-r?BLT)-&kD8mOGJ+d~(&| zbh%rp#uQRp(Uj$?C_ZT}W`n>$c{i0`Txy&~s+F;zfIh-&F%z{I+d zi|T3s)4!P+mdq!?riWNf6Eo-7D#e2Ygy-$vn!)b>bo&Z43&7B2^5H|5qzlKC>K!Q4 zkbUlxxAs-83l@=lE{o^x>Ef%7FnZp4d{Upx#H-~gbuFqD3obSsO^#W7mXI%m)>YC@ zoOwjH+25TYp7|+*!(CbG>lJG4$3v1j8jZwT14QeeH+l(GWlSEd6nT@(P%;Xp^#Al6 z<4-0l7M9G8zI9p|(bQ@}?$88nV&KtD=^Ife<$B1<%$uI0i>;VpzSRTeF@rKmk%MSR=opeOA++{m zCpV9@6LS99OLu%3r~Yg;UIW|3q@j?B&Ijrf-5s^lwjSHG~2yu`O9)*Afq6m{ps5zr9;KG+#dCC-gSpJsEo%bB>kj!RbYgXyi{P=*xz z=-p6vvA}?HlJ^qqu-#?K+b9oRL&FGkQWB}=hf7i# zNb7FN>(2`Bs^4R(;XN9S=xnf?;oOiep$jI|o)W^1i2c5k*-oL2I(0L+nFPYxcW)J( zf>IO;b#En(uvjI%^;!Y_4;6#164G%5Ujd{H5F5L40E@SRmaHBigd+nK*P^(`^hohQ z4m#%c{0NIt9HdbzM^HE!Q_gnJYOSa6O5u#|T1dTl(ybvd5`vY$Alv)K9cyvq4Pz%J zNC6Yw&lSTK8k>Mu{2U{kfRL6^Nn0x`{!zWE3Cxn6+xT1&lc%8_@(cmeRNfhQ9dy!u z=r~yjh^CkV(+pV5W1Ja}y?xb~(FmI1nSoVnVJrtQTZ^AenICv>HdK7WOi`Hm3~Api zF`uV(VQR`ro1P7*0SqVs+n=Z3#|*OlOlmS2}cJFrJcZ`5T~e9!%6UVbVm)K1DP{tb!tn7837& z0ZkS_FpdN&!O3{dpd57&MjCP4t)2u%F@ZmJ24ITE)O-e7C{#^gW+ZWuIOJwnY6{r< z1r8XOQSNKyjRWPkBTrx0yx==$*%4K5?4qtg9ir#EG9v0Ff-^mP6@U*NXvh;nmg8y+8p8#-Eec$8DglgYnFO2{9bl&4vr> z6$HVNh`9qN#ltjkM5@8J-eM{_%@{#43NxF*d-hWeN{wMAbw5Q`wKU;WmH$c?=g+!? zXM#7tMb$IhuPTn%hJ^@EJZj>DbOVqN4nU$RQL#36B2Dycq=0Q>oScpGyOu=OG!47B z=E;c(kWk|FXg(t_{{9*m8HE~lAbjmWQ@G0{VjMFbhlyvY=;saF)m{Gc3Y-}b+lt*n z2SA012gWI;m?({dH$KO2z_A4KOhOxzd~>5&JSdjMGCuFq0UgpKy6gWjUHez_`%A!L zL#uRH1JE%{!6e2JYR(gG5AG|_!8=jRD>y8Eqt zE5N(9i{Am5dF*OvjlcEf(OaD1ZHI5g4sZx?@jbEQ1HWP} zI{?&sD=>%f3{hT{Tg&)!+gJxl0=~w^#-<=JT|7n884A;}$$`|VwR z)%uBekFcE-Jh-lr(VmhdAZNg4&eY0?W2w+ab5*I~_0D1|^w~9YPYYXunBwaB>Vo2$ z)KXI?rH~%XxFg8QhDYk7Sk0aG)%KT&s2QXj_Wt*s(}u%z6$H0ZSGIo`h3|+0IA!LE zYV3OG9k+_=j~=6^nP>$yrIL(2-IMCXCbej0V`b;#GT6 zGT~XP?{Z35BA_uFiKul`w4c}GNe)v+4gy5&fSUl%`&2jm0u~aDr`J;)xywmZ!|Z=Q ziK|f}E;nC^dyjcwDf=4~f3@Vv9g>T=F@0~{Ke}EuF|WUo+|b|DD_IwPavyI!^aXN5 z|MGk3)DXehF+XQTqldD3h(`RZj-i$a*xu67kRJ6BhZd1cB`2el^mb>>>P*|1QQ(bW>bkxN5j_BzS+4>Q`LodzomCbnC~5vl4>QGBkJZqVcO1 zdGTPV@Y;)#;7*f%FpT9x`dBuSh63SJJ{RBlBOnC`K9S;$ zeNGzjUTQhvLO7f6wEJ+Uk{f2^8*x|0*M4Ox=h*&Dl{~lTR4;f#rI_;TL;v&A*$=I( zY9~V!!O12lkM)FVO$NzYbsatL+ZNk*Y!w(xSk;Hj=0Bb~jkH3Z7AslTd4SECs^zcny4uXqf@gyZlMb;IX z?dUZFb3^i6#@qRPbq}=LB>z>n?x@9F-wS<0h;aQHy<184TfCN^5@*_6=YGyE2V~yp z*+=Rc5$u5@M`jf33ee`~^#gJSrkb>-v;I{S>$>%&kN8!!)gyt^0lKitao*!;p zaWVnuuE9*#52$(`5W4s~r@_Bv_TzFaDwBwd%!I#|=zT^hFK0G~V)OGJ24&w>19xco zDH4=D+}$}2)F2D$)p4HwG%>E0>gU2}2+f7!Y_dp}@*>EleueIEbEDfMt9;+TQYPig)uzd+VV5N zKbTREM`jw!)<&^o;7cTHOND%4;i|ijIfCVfXRa&I`2$JWIpQ z^60$ejnKX>BGM(0%^Utn08TUuZv1J0q2N zqK}dRJZS?p9jP)QpqN|wz(OV?n>b|K4eV8)C@Uj=&F$Hw{ED><2}kC9nGX&;Vcms9 zCudBmE_fTm8*m==qkprSW@K@=qddt|3VnHZ&X{^Dg*A^~FwMzXFTgUOf1lF5(#tO$ z$Z=5eBpRMRf1L0*U??Z%Zc^XsUGxWE{L#e;>HlK^cmnR4eB@~OkV&!Ywax3?40+FOf*BlwN0TBIJF?rWS8F9sFyzi^ z`@2Zd_42XP7duhV-&;q|q@}9XTyQ0zgrTXF}r=Sv}ux4$OA9SWHp#IWOn=up* zo3?l^N$7|=r^)n()__Fh1F_6=4reAB)}1hA$u;SO;E8Lzkg2A3>F=7onuW>c!a{Y9oy#k4z)=0nOPMC1wHII zm!4l2b=8b>87W6_U3)S#&^sK-e_yvSe1ZC@&^k$*Eek{_&Y5{lUDObS^5j7Ku^}^y zCg}+l=!@iWs+8MxeVhpf)qSGDB*jPjmeorK(ujG3T_T(#(GHN^4A-^Q0-~J3j*i@l zbYDDf{@fl<^6eP}WzhP0ZkFg1S~k3sDF`z1-YGIfd}t*$YGS=rT`zchN<&&IRzZ(^ z&4=cz<4Z)u(8!qyyrM{R4gSVNJ$~lJqd55Bwq-NWbLljxs*)T#(z2#b=ilcaF_fc2s;%OjxI z)MEO4Gy^iOJl_w`6v-VPooK{25Y8us3vz1X}6m_a&sDb``bCGj(HEld2fGru$6M0xsyj#4`4d z5jDUK*Im%R2V4dWt>o)`>j?a_hYf*w+GoS;zIhKTHDHCje^wP(M$>r{yy{&TB4-Q08w= zeVlIn`t$x^^ZzBAjUd6Y#u+RoL1=1aXG|P>zaP*wK!Ba&;Pc6jjOj%5P7^AOzN&h{ zZ8hc%CSovi{9|W#Z=G(}2Vf<@Br`-|Y!hBC9Lz(l3m9ns`eOUwY$d9@8@qR{UgKD; z0!+}`rOgZrH`k{tP{o2X!K|qN3L07ojj&pdBm7Kfa5`E82o4I=;9TCq&KhDOnAF2g z3LPwYe2?72xK?0LS=_bv5laV+nD;JF3C6&91HhuClNdnyTz30h5&)>ZWgh99d#0m`K<7>q92kVc;S)OPXpobmY=9~71MI-^O1{EP)slI4 z1M;y33*YH*(}_CE?B_16K|FqSZ(obQXZAJsbC1dvF6 z0g0tPAkvc=OaK!#*8VVR+551)*$^kF?le#wiGsIJ25F$w~>4yhNg z`h4at`Nh0PhO$8B4Fc-rdXeOs$(ePUJ(Y=H=e~QxzAk2ME_rhy&wl{`i!Se_V=p-tW`4iiUATre zcu>lolh8!b&QMWl*5l~8pZGli`kh0Cg$G74~09r=# zkJ~k}s#DzlD?h{pxtd#Bb0S1cR?f_h>LweBsSI821F>({&?xQ9gi$gaL!N-JoLuAkWAC%I3zy!NdQf<;2z%oKzzFNmm0z_=l_9+m5Hk)#5~t*K zf_y8j7_H3b=G5nh_FxT*)xw{b)L(0c(l9Ntr}YaFnD|*oeR$nQ6Zk2)dYSj-q=fQl z->uXXNNU|iU1-sWo9d^y*iwQH@r70qk^Z2 zC*KrG)u+ z>ypSsGs6E)f8N{&;uUFoJbD>fMwe`o(SAp^@x2D{kt)60$?c9X&-?LC+E`&=*molJY!?S+iqYkZChX z`EI2NjefJccUK_=QH6VV9Ef4B8wISTImBV8u*Ue;GQRC`gmr+>Ah)4K^a)qO&6S_y z)4wftXwrVCz=cl`fRt{1t$uHmv*&%=KcZABQHcF&CQnmf=$69tF?i#BEe-c_jT;I4Y9)MyfRxNF3Kw0)lirTt9+KVO3+5+U36|8$*fr+O z;;2m^b1N!D`iAox6CTjj;#l+sARRnM{A^1heWcq>0mMUjpfryX0+Y24%#xyChuL5=Ylkj;uO?adT#<6jJ3t5MIGNic6)Fw?%C{V1c|xST3) z7+N|`8Ar$*%3z+_cLQ=ATxPmM)J2{AL%r{ATL?P}!nP(J9x~QSjWX}oU^99;IG;R* zQH8I~+R}CPH<}uFuR^JH zVdfW=6z>_f6LjP4<}E6`b0X)H#e!7Fqh9e@CaZ?=&hknXrpSh{RTDQ#A){g|?5j99 z^$|m6(nHU0=f!GVCe4YGB#ceI4}#qaRO^`0U?i95EDStXGL~ICO+kiT7+FJhtoD2X1lE}5QZ4uC~94AAryvrj+Djl z6N0&8WspQzt0+;nH^0jbh%>_a(dfwW@-2#oh8^kNkC-`wZvEZPG1)*elsn72B)d*L z7arJ;k~ZrkfQOtX!kQ&WQrO%$ERA>0{jF!hwKtv$$j8IE?iecG^;>jhj2hyXZ9`2n z(N!ig#`VAcm4kXQv$m1QmwIv5q){Z^G&xA!WR0~F;ugLssWKND0wWUW40?P*CfdTma$+k&U9;CVhpGkhINFO($AZ)b zHwf9SZv9&;LCagHl8q*=&3wB~0{pK5NLXIdg&^Lo!~o*$QM2;e{kyv#jH5@pPNP`0Hc?FAD3&a&$_uZIdVt9TBfN|mGovl#H9*6nORge~44@A-i$G>sf*iK`%WM+@LS zKMSy()uXJ%5v=-w1`}$x_zo;a0~z7mGYjJxvvWhr9h&?08ns!lvv;aEj6QspnQuQ} zT9F~Oqy)FIW33ERynCOCS;jXCIZihY2(!Q$?`-OCN|2xht<&4i>(!YtzE8$RI~c?t zM<-MWo}IJ6 zb6G8;#1E; zR25*_n;{oIA1Bv39OzV-oLej}8I4*E{`3>AGB39{7-yT5Sg{3~FO dV<>TXh!grv)IpDAuB0D*IA13Rz<C`k10HHO9P5MwlnPx8AqQEle&U{#z3!Sakx z4y|#@#j4rzunn|q-e8QwVAkqmRVY8m>r>l4R-q&Vohw@qQek?ihPLS+v$h z-XEXs?x?^q{pJP9@NRYD(S99LtD7LkW~G^rR8x8>;;c8UDo(#@;r>)$xPl*t@x*w3 z%uxTLpd;veHd1!>!{IGvS^4w9R7lzME98xYM(e z|Ei^6k$ivY?^=1nwM zAM0jGxA@jvfT)kqzK-j`CPU1@0wKbylW%mpnl|H}Ko_x!V9_wPsh2>{7Y9dY3t;~F zD7~L;lM~cLL8rn77BFF<1K^kfRel?*k;b#>#0V*Qk4zYA}SA&q z7N{uwCK>>V!E!qAIt$p(zm0?h`|^W)D$VkAZg0F`!%{YPPvxqB7Ch01hgQVQ4!Z(p zN1*fvdt*?bV~i=zMt@Y!2U+VttXX~E)!2|PFP7&n=IE{W)q!CB(#|e*?>kP9fz#Z; z^tY$tMl4N;=&zL@mD?&+olZ95Z^`{PGyw8_Fl2DOFoE?zJD?El+mOUxfC`XS_gYMF zWJvqXX9WBD?1R5lO!1qOag{Z(s#JOH9kV#$&yit%JN zPAzUuU7B7@~Z#jZRGH0a56*m+6lO$tr;#ov70P z$OS3bkhB3T1r8Ku4~1jsM{*(^-7M_mNhz|4mUEcKTi z#~Yl1)-)4x%WK!!8|*(uL218$@^9ic$duk++pZe(c4&5%%Bgt?4{YoiJ=@@O0{;M; zfYiVo{wzpN!!3yJnG{%Cx&qxWmeSPvD89(sfAo?E=_r6C#BxDT7W_0?2A{7;!A#Os z8`$*+-3sL1(2gA-W8cGG+Q5z8K*{O=4C!YQQSr38V_oXz?CszSv)@Nk%76|40G%HV z+Bfy4Tik8uCi&2)UHF)1Tb*fBH?%#{y6(=$o3V$hrBx+aRYlg?>SfefWv$`$7WpHq?!((o}M!G zwgl)->IaW%En0w40VJ1DL+u>MN6-;;|2zhUMNq>JT$(H^2`q8nq2Lu0D8~liLx10z z{BIr?$>3fp_a?D~R@8%l?&h?A4_FVzB|SdyN@Ej;U~u~kzW6tq+HoH!Rs=0JKlD!n z+5nW#2{{_JO)*9X2|S6@4*tszsa!bc6hHryzvtSP&?RHWym8qMa&-IwBDAk%J)0W; z8$Ce{n6LmebK#Jp`OB1b-^;!Jy->PK3_vf5yeMukw9jb7IE^xB!-DMp%h&O3G7g$f ztviCyZr9Z`=@vebK^CNehoBPUhkVF)g+$hP}E+`-Q z{>f{O;d4}d6Ac6zv~&O>YekpXk=p-k$|Mw$U~oP6|4Juxcy-co(hg<8{&u|lf|lga zy1xWw0{zec1N6QO1VgW#vo$c7G)>8~9CqtIaoldNJ zy4k`6WMP22w3I?jjVDFoUl;c2biZUch^XoHMUG`@--hrAE`5}?5Ekaxi3MnD0Eh*! zJ~*6!OroN?`r*F!>@0+c-9|T&fEKYSdV2x5TCTyQf(B4BElSBSL4xqVVdns{DKkG# zNR)9FqP^x-4k0_hWe21?Yq~L^iJt)-1p;kC3VWyb#fEl{ZpqMeN@CZCWJKb`q#C6T zRbvIqKQ;3Q{G-~ze+L?l-+wp};pcIK-w=o>@d6&8j0eD)V;c~ zC8S!ukA5gWcH*)noR>w)rjY|571}6vGvUxV35}zb?+f!_&Nb+F>GV$#(I^?wf!ncy1M=+rOh}js`cosHr)vIumxutlkeXPype(h*h znu#B~{JdHI-~n^UAq#!_$JAR_`iMb%mRrEi?vhCsLNUkAgozzzhl8z5UAgj#+}d zo5$MOoHcwOW2}Qd5M3}84J>Men4b&(N zD9TMllO(&W(sIvM5Z79^`@-foNqi@M$YZc*`Jr}4CH*N#Ag67}IFrP>SWUbG>-laH zL&8T_fYpTSmf&R{5u>>C!iNPl;#qp+rfV?y&T^M3Ja&2kLH{+gDtr?GQPpd}9Aa%i7D&tOgq`q60Za&qKK8bWWV zR?L~gMtx2QkBq#xwNvc0IX<~lkk?V7NT1bwn($5L=Pcv+rJCt^?(X#cpSU~W9sLx8 zdNy>m^9(yQK)*B*s6Nob&UvvFvWyE;f@55NiP+a?Z^4b^JD$K#`=YcNb=10PMYk;< z#9>T9UYkv?q;U$rD%VZGod&M$?ryTX8{jGvXV}$$T3LL6ekus=MN(rWJ{Lz`NeX{c znF8~3A}_)f58)YNk;kB<4>q@d>c$jFJaFw&L%~d$Az&Erv-;yjx3|-RBz;!UjMp=@ zfZ@g;#MZ3`=hNKcLXMG*e6`>>IN~Kpm?GCcHtG78j6D?ah+n;ZMsk;dAmsFc1cFf`C2YOeKcwrUfjk1M zNWA|ekAOkS0fHUYNA$@2`$(DzvQdl6=ZGjoyjY^IWiy{}rQqGlk@MU%U;WKr{8Vv= z+U<0~#gIfCu9Dg|u=)i{M0zCUAC6FVuMZpjUKY$sjrq+12fG|%mI=Mx$)VV&r>H_* zw+pf+^m?WOz}Vxj-f4#A;I1Yx-H6xDPHCLE(xZFiWM{`q(B-=3?#fOSNEx?*UN4DZh#0aKe!su@1ttziU%Eve0u*rcJ$@m8FyTYa(pAv z?JY|`Yg6E+K6}+xBw6Rs72eN3h!=Gf-_MhmiZ(qO{k!`1F>H4ypbHYiznu?HwVTVy~CQ*5p>3 zGYK4+`yAtzy|zCe!HDz?YX@f){4Zi2Mj_mj1~BCj!W#35c3Tg62gYdGeCJBOI2J2u z>#U>=-(G&D?`CtOVXd0niIm-1X;@I*E={cPdSP;5=ruloeh=zyB0F%NS}bykW+tr{ z99jq8oX{s0xF*8L=2D8Kx@g39;50)0zvQ~eb=5tXEl;_MlN{q^l2gH6e7>59QGEoN zd6`2zyd3LovOm;grp!S_z-F*1d!CHykNV{IVgD9ZT?UxG3Ay|X4DzQv+hh4~=O)g4zJXEP#aL5Sqf&nJ zobY!$g9$*TXgg4Ak+UKfF}M!jZhdTJeu)grw z1d#uY)Xs`X(cUHlD}+icn#;_nd0+!|0^uFZvrYJ}Id>34DG?emJ<>D$>AQ*#4V8

HIHDd~2^(VlE!wQZp-NEo#I z{iuwO(xpK?P%#(1)8`z&$O|fZU|W`nYhw|O8j4~HbS|0-B93~I?ep&*?+^1I%LdG= zQGA~3V*QakY-mCr-uysbz|p=ecm1GGX&qn32*wEmCxYC znFZg)rV)+BJtdz96BoHZBq@5%5VncCoIBM%>*GU zM2gyf$Lwqm{Q^>D*lwDLF`hg;Mm^#$y1S{WE(t#e(v-|5J9nQit8%Z#G;j*YxHvc# zKY!r*_WSo|&+*7m!h*}n1_+Q3<-yzXiP+I6%E;oyfEDW^kP=I`{)#zq&mlsCO&!4X zv;C)A1(y5Gm)B}=_dYcLl%b^M zR3(Rnp$_C=5V8$u4e{dUGLrkPVkN?N15R3x*EsuBNpapD*X2Y*3$%dUHX07NAQ|X* z0Med;Fp=chPcQ#fJb5-i>%xV4L-1y65mW=+G@#Z;f<97Q8y3Lw*rh zQ4RK{z!a(gREuP=!nvk2h4{0opMQU~x&p;<|9ZN@pL~?hi2J`ygHF-xKQ6D*-c3@0 zRRu8rL!>)@n*LP#kewM|V9&m}T(<(%^1D}Lj;=mFlouzv{Xh(D_*PYHex{xbGyTVp zU=6Vyfwr5g-G2A%1DcC}WO&4+gbsu+A;4VB#cn4ZSie);51p=wp$~nWrTHcrn81rq ze?&-zfye_6$s%oK{bQd`96)95o^xSxIbnW94xstusX$n}7h!G!tmn38s z?D=KnN@RJ>+W)tQJU~fFnbo*#kSH-@k%<4pfIfQ(h_q{Cgk5Kn$S}n)6tu!Xoe!E7 zD6RA@Xk@cz0~z91{{hAXkQDqb+1~&WHXwH|JQ5?3l)CW(VhRAx=9f29Rj0yLHfuxy z$38w4oUy|Qz;}oN5rH_;_nL#{s<^V>{=Nl9OojSJtBdQe!0Hg_)iDx;95stNBKtug zAqda>0FFC}_*ATVo>QS81cig19+OcyX0lg*sPG0?hOdR|d4y4>-YgU%3gRl>@90C1~l`fX@-S$iZYo?tXeN79#Q10;{F5zJbA)uU~IlEDP_1 zz+fhbv5VY)pTM^$#jj3Abb2cENj36n;3s~barkUj4FxSmU95kl4#JT+X6UXzir)dz z?{8Q=0RmFkT}s4o5@xzFNcSY2`2QSe+4TJUZBdT$=QKRh*l;>h?52@6 z&M|o>Wo3uOEvLL{n2?NDYe^*UuZ%0`N>#(v5G|3#P`{D+0le zVM=+50}K9-HbA^HqPSCV4II$!*?;o~=57OQP3d7~n*z~JsN?78wT5R=qa^U!ENKz4 z6xSzL;qc{M@OKYvW!ljvb?&S690hGN`lUG>(~nXb02}aglJcTY;kU-SQ{>iS*?Z!1 zf7yj@6$};bVEldhNSd$XwIlO~%LN0Qdv=cd5`6@PLyU?55b*uoAvq%A>M3$w@D4ay z3GLO`=jK?K2q*{bNICVE3++ z7aRp%8mmXBFX>;be6(0Y!IAymzjiJ}C~y1oP8H)fjuj29a1bPvE2b^4FxJ3j&%cxh2K~gncnm{*d23{GV`rd}{t(Y@2Xz%SuqX%9QOB_>Q1yM9eAteJ z;l>Sz5IO|KN1$X7-_%7*pE%V8Vk;D@V*IH|jxzk2gvoX1?+I;P`Vu=qe86JnA1&rj z90!8IZ9x>k_4JvB(4iF}i)T`pFo++ED);{I^RX49=b$m%pI2s6Y0Kl4Y0))BYr*h- zAYDCgO0~Yw>W@ZD_4#IDtO7C#x^Z}eQM7>y%!HQ7(6LN+i24I?1?8LkAohu2RWV

?DwLi5NJRsPgRzFu!$ijuUHn&D2p zQHb~9$~CL*M;DkV5W&ru4ci#Ll?kr)#fCdH|+X z*yO4y6EzMOp=)=I<>eF7H1LcErvT6fQQ@Z~~oquM?KIZc0LP|DbJ&!)Rq^H^gt(SRjx@~q@s zl=GC+N+s`}Vku#z5OdL36lxk)W6LusYKWW8==HDQIZ~P>Q4I)jyH@W+$;4I2#)isl zS&CgW2#FAU5y#guH@21#mn2Y?1t{UTlqPUdTz~VL2|R#L!Yq(=N)kpap&UD!TiX}b zj(DZcnn^0=^R}|Gw9T7qUVeeuBaYJ=g$Wb#QzCYmURm;sXCgmq@V7W#YD~IHQq$t) z41I{3DEhpfc)a<1HhgkTg}tf8gXIe>)gY(6GpDu(;Z9oci-&n_Fx>Sqa?O8L;g^uH zkm2+d_N60ELZmEd!9#e|c(2vULa65V2iBQMDpmBW$%zIE+k)qa{b>eI2Yta+Ig1a= z)S_Q$U1Dt&Imi(1Ci{uOzBJaj#|+aaCl8XNwKvF7wejbfUM4%%wa;D`j97o7m-jTf z-x??`#bUz+#LVpL=DX!99l`X1#1QSk`f%1Do-d*H+^Xkz=9;JKJfd*O%Vx&C({~%cax8H}=G+FIIr;ay zKZP|q4ac>WyJe`ByLBOx+<4OK)p{Z_moQO)Z!jhC)`fnvG6T%MNqM$88qQ%!X&_?Z zLi_XYs&JYRFJA4gw($IP2Ek>BxK$eSHhJ_ik)LO9lPu8vJV z7L<%7{JRkSM=WN}+Q8#$4q(LrLePb3jf=d>mfZQB17KLbUaVWo%$}7jm2t$Wo9x%b*4?1$ zmSik0SR(=)B6gml8~dZHZ(xVdkaR(>8rjI+v^vToVn0+|V_v`AWq-bZX&HAt=I!$i zS7}?kT9SUJZd57;A8;l*rMKMlu@UsU8#lvXCre(kf0VaxA80^|-^cyXuUC~%4Ku{q z`4!ev11C!}u zZigGMG)uTE6CX0vbnsa}m-vBq&P0$fp)+%=g;XeAdGZ3h8)BmOg9jlT7ier~LXmc8 z#`ygdgDkdKt^OWXKId74Z>ZVb=@FogxtM7Ft$x?~V*8rhF<*OQN)7w^6q~)9LhN@B3T+;#-O2<015*k?MoojnF=%6b^gU^!)QtA3ki5 z()Lqu-H7O$p2U5l)Wq>*%%fagQQmQ*|8nbAQ;eaE=VoV1J4wDiJ};53HtaLdob*8^ zG2ch}T!uS1L7*Ug2w=vbvQsRR0@7fw;a<=brDyy}8Ya^a4~E>MK+TC$<0|hrJ_ne@ zr8jJJ#^L#|yu2Qynz$)nW;?gk?q<#jQ4#&;n8YR#j~jKTC^9#<*D0TH>BfC{d2>83 zIaym?brV>9^GJYC)$2eiLjdtKXJ*Splwhjl$C zk+A7vw%o1^;uc)#*K=eO}gYWJE$ObcX0z9Ji2anQe=l%m6+b zQWqbI;3fSMEpZ2XI$Dq+p5gK1KpZ3(&iQUtg2F8l8 zxTEuP)m8o*GjJ&VlE=LM0Gd6Kc@ca7r|Uc~?~DRxC9C{C?0!#eIuFwJ8vgtH zt+tC_yOTGWKQuFGS+~|8OnDB#d_(@ubEC)OO9qW&c#6SUIDy!RGa$JYgF^G@k^At} zM8=;_Ts@%;rG43?UmeHT3gnrFaO)s4l2JJTW32wq$;CTj%z;&Rs*S$@Y6rFOt{`lj zAWuPdCURB*HS$R$6rwr-!!CZDxB8lrB{}FMPX8!MMXigki=5Zlr*Nz3_-|X$^?-G& zkCLG{seRp&&m(<({hH?Hm#6+lSw%_riq*5eLHT+t*>AD5*Mez0y%J740PFI0FotI{ zTf13x#KPt9!uNxm8~?Z25x}8Mv69B9U6wx4dHSKV3v{mPocxv)o1NvkYNjt={t)GW zxr!}T5~A0?y}NTdpePvv0lGoJY?;^e!MMdTn1_Jjz*rOjj{hUM2@1TL3Y_pcX&XMu zj@!LZO27ddaxl1<^tE{U}6ZJ5zM-U34Ho(Rdq$Np>lIWj5aO*IRf<5S0i-@%fwO+atV=pmO$ z*WPk}ba}Bx*1u%jkGIKjFh{Z_%B2=!3hbjm*I#kT8}zwOPf)Pvz{aefE>nvt6$%ms zw(!}mx7o$mnCSP>NoPwe+7A>T2aCmAe@1)%h#VKo$TTuIFbqKX{42odoW!Z{I>v%Iyaa zeJyFVrG{^U2k}*82?!&ef0FQf0HPiLm3gobZT+J8gsE`HV|GW7m_|4VsKxECigoxm z+HOCt*YR2#q6De55~n`|NTt))7NpH2(QsqT&F%etKJ^^QXz03krP)HUM3>%KMP-u8 z(nm2sdIDkZzn?m6sRaE@UP@MJvLe-hTzRNoV!F;Y_U}Rw%t|nXmYe`Xp2hgRKU$R_+R{IE z3B3I9Dt*dkW^)WH_$Se*aN3T+UnPf~?{DnohEc`ANw4p>m9`@CNS`TX+9@>o&+R6;YpDxO>bn9_R+=VW7(3so5OfW`%QSF2}`8!LSMxN zV8U*As2Hp%8N!u}65F|(I6WUD`NAw;o{lgg$K5z2b9VZ!ZNx09TO1V(rICndl`(C% z*)pfnzVO4$if1ex9yqK%rB^RdK_5T^fbFr#XrMk z;yMXM%Ca+S^=zno+QzNZjc_flou3{svLYoM?}x^*4{g1BDUnt?VVf_!+w=WwC}!+_2_Lhk{kvHRd5q#pLe zH#JI&`K{Sk693MM5(C(Bt6mt2SCkVsKK6_a6pFTvd7i)sGCcwFDi?_Q@R4+OoC^@(98`8kM_*V;6BkxL z@p@wmykTf(3i8^pdq+43pYJd_JlNcHF!Hg^{L&uw68&}QNVeUPoemZr%y59nZL6B^ z>phv!4i`}4r6F#}$6Lv-)=Qhh=*3J?UU$vUrA^}V#);^8TszcZJS?n~j`~_VtQfqH zwc;4Z`1+2vg>R!@r4m~M5pGoeU$uf#U#;xMO!(U@hkJF-uq|?6w@E?C_u`TmQS-}@0be$QTby6zzYOU!$UUrriNnd7Xh2>ouPAsh1R_ec* zNQmRm*&=fE;awd`>STW2=dZh3$sQZ#HY`AZooa6}S(=ou(T+xF;cw}v6!vJZ>9lXN zmgNTH>$WmQW^9sw>$vU!=fvw(fCTD!6Z#oGtl5!BhLi zRYb@I+Znqz$1IU*eX6WE_GM;58-X2O!n=mcorY-(jVRZ#}hYwqga(wd{mEv}2Ojm3-OJls;vChm_&J1Q>dUY-M;!*LtO8<&f zcHH~2(HOOA!(S6>n;TJWsd(j=s{C}TyQhYxzJ`Lc?*%lszx$g<&!KbkVQW6)^@s|V zw^Z-k;IRq%Emex`5uL+@qH3OOunsitn80ctapN=e%uwAQ>aj{s|VsO=kqD(Bnmjb(C;cp9kNE(}AlR*VguvFYlIjLQN%N#}@en zopKzvN>gYy>=!hr{v)z7^o(tNEIg?kU*@G*rb+4;-&V=rY&nGJhU7CJTFo_MzD;vl2CTA@9xte@pL{DDPU_? zej@k9-TEu~-BD7+Nwi<572ikWn{xYzk8g>qn-L?5f4j9KUJ%Si}ry_YeiU2RL*Y9cny)eI=QIvpwh z4vPGpywdBFQAJH#?@;6n<_x;Tgv{?mcq}A4RUW7|e`HmgxL~x&Zo0!o>f*repkgoT zBkfdRC#Slex5>{gu$$+WN<`AR{>*=bbwkN-j<02()VvhlUlKmz zw|K4MUcLx@McMb~KWW%&*2xwBU>2^d9lvhlQr^$++Gw*)r-BdhHR!04mcR=mlkQnk z7>zBf-&#E!^*DL}OAHAo7|3G8!)p}CRCpXG z^u;6lvc^_x^h5(^9x*z5XZlieq4XlibX~1Aj`7xK5T#wCY*wr4cx3zxV_eQu(h;@S z=3kUHwSPWrmov1_8#p$~-KdwavTnvxQr;u3*mY@N_Cm*|R+W~cAGK*Fjkhn;z2x$; z=9$CPjx>CuiFeyt(ESNttc{{*^riZ8dcJ1rU$EMMLUcgeH04E5K~_iP*Zv6Vn|{x= zxX->Ph~M4*cFws*HC8zL;*U7!w3Tm1@tD zW?-K{F(-D+(R1S#fO$3vygE9!D6V~C6uu8dn1K%Q_5y(9F8Q2^a0&YC!Rer;=G!;f z9Zy(VDj;P;?oz*#znsG5o-yS~XS4gzpDotNc_cFY;uex6tP7xW_v(u!)K0p!XEo+& zA0B4{;c6F@Sp}d<1VOKjGp$C>GXag~Xv#bmtbVovnn3-MV(IF77sA~F+ZcVAQIvIS ztZ_cH_&wYhJ9nQP_R8&O6vAL{AI4Cn(!ms@)|IAe*TVVLXTw&Uj{i8yn86I7b9L^k zW&W=g;Iv2T@;QyDhf->C{;^?^!tIjg3OY6>sL_MOlBb}xq;r%T6_{XAurcgt|CGK4 zfIyHOb*YcqS&pVZPMh=b?>}-P5^Gn#dq4cgl?Hs}Kc%WvHTJ`|+t!{{cLNQlB_Uqu zw~z8VtpOn3{JZh|{sTfhY=S&Ox~(p+g+d!ACt@hF-DR;06SO)cLs;9eLva$oT*rz- zChnXceZzQ%DjpK<*JurbOAl_C0ML>c>O5R(4Hg-0Zk(? z;eL_|Gz(<@e!P+EeOqY$?UTP97f#TDqCMT|b5&DH({lkC>>i{Q-CGPLkyY`0`%q)Q zGQ=xyVb{viGS32AEvrgaB@)_-I{S`g{wGeZ)PJ$%?-R+IM+0UHmxDQ%x30klKLT}P z*BZ4}5v?EO6op96)185VG9)a~Nt+d6(4gitU+Vk`_(mN++3@#evZ0;zf9DFov~^B) z1??2ov&AD%I&ZF^VD@@*%O`3&zp7CvlyjM2=KVxx(+DeCAfp5=7dck2a7=@al43Ot z0q{G1l7v&SXk?F1S*MH6eZUWSJHN>!`QpV3VYd|mpi^eQGI^X-r`q$ab2#QV_qZ1+Z+rz{Ti#`fo|Lb;eqkAQ+mMh27eUD>^Q&%myk_NK1gxE3s3)$d6QS=A@;kOUeXzByc~t{_>pKkj~vO zTR3sOzVrcD{TFDa*`^{tZt@{Tity6{ zv9}ouI&5d18_fr;>{jObHdR+-3?zJ~QbdGf{0F2suZcWN!KsR|GF8B%{zO?r*$sjE zcNq<8G~U><|KhNHTip*-lI{XtaOB9$fHcw1wIIgAxK zu5tZOtba0;k6zAc3<>(2<(-Q&Jkl(6#ZGQb{2ebteFYiBB_Y=JkD|%d#FM14-|Lw~ zDacDP#;c zCzn&P&ykr`X_oH#!p58Y@iyz~W;{*6$QE+ptiz}_9@fi=t5!%HB26t?q>PPS5p(Id z;gg*)nvDm(jaa+3{%(k_e8z*=f`<$w1%io>w>(5eYb=d3i7jcb0?eD9Y;}7m(Ng1Y zb(YeGLIcpr9Ta`dU*A_|Z)DeVD{^9n-;wrqKe*PQHk!JwHy?M$pdgMYckm?FnQ z9iP`{PKi|qevP+s@#uqojz00^t4($d4ze1P!M473rpn)UrrIm}^_&9dpEJD>fwb$f zG#CTiR?HkyA5=x1OmNAs9+ruwD|a6j=*nQwns?9};EQn@cL-sFp@f!Y9oPjm9O#%9 z&un-cdRGKaYW>>kBF=0^Aiox-d7!Xt&*A{LYyHJlFSctRax~nCulr~^VzS#1M|zO? zVC@3E-d^Lo+r2c6QJ{!$edTl%)|YL)G*2f)^zvvAv9B-P;70*kzcQQ=Y3U=ZX>J@a$qzE?EySDHu<7mRg49LXxAV>o z$Ewt+8nrTtKzDv(s|L3tE%CfsQ@Ub3Qc1^j%aJ8`#%lK>wj;vc76IST@5jZ8aSSa@ z5_m26mM9;t^V6{}D`|{>*IC7n#_p}%;U9tMga=}e4=DBWX{c`@Wu7>S5(rWdPF(Fa zd(`wC!W;Z$-}|>@5j8$Uf1Ud)NXZ>I#mF5q9GY|{8o23~ysY1{GDEk`Hz}*FU(~SitjFOfB&zqz1W9W2#7vav zkYuo1|N8W3d98WlRue;2lZP-3aYWCOi?A0x=m-AK|8Sb3@0VM0A^{^~XuJnx{D;OT zYcV*|*J~0|VBqa{yxm=@Qw%1>B@6rO>OyOo=d$;`pQMAA1}xPy{7rTH1N$;!?5DeU zn)6Ju|5bB$9&#<_MLc5PzHcp=BPO4A_R`R>8*l>VzXXY?e602BE-d>0eh08FV0CC( zyEw+y3h#NWo+YX`ot$^+O)f*a2u7^so1g#sx}E;B{3Pyk&#oIgm{#TBcnCQ_tj%5r+@JYUx97ssZ2d4EH`5SlR&=pVt^N`zYd@AE) z3XGz|K#Xk&Z6>p}3IMJ^B!_pXOax2ecfhNF4%n`fZ^@qP! zU5O2FLNLLLpBT_e-gi`2U11c4f)fUePLNZls%0%B-iwIAJ_9QG0ki=t4!yV&bO(oK z>vP3h-BD1>ECGv3tY8W(LI>ebEy!+15iacd8|NCNf>Ho?$JfvYvW`Ukl|!odDaEp2 z`u?!mfAy*BpC6AIl|MIesV3lnSD}5Gx{9?es{Q>sz@>sr4PvMf@wOoV zauk%#zM7arc0_es1aZSRUYs5lOGSe}pkAE6+NUq1oUo`Ub$%DaGi$e8C)@D^W3%9w z*xzPsp*fdO$%T-Sd#i0x0&05z#(E2i1*Tt}RYxDe>A2r^-}%=k)Nxn43_C;Bv_S$Q zjR3IZPhTmOEE~iB!YUxNCZy&h$ZPChy%L-efu9RNs1ghfiKC^&c6e-Atn?owMt+Nw%;n7Txu~2%rE22EaRxT3|>Ee#F_k;6Vke+;OhHCQtq(hhhoU^@#G#-O?pJMu_zgtl?q zzCS;J`}#WMi9t#tUwVDoBI;pCoFh`!((k>Hv21`ubKDWScutcdg@)Q;(${u!mV54N>3wYWkdOep=Q#B1d2I z6cV;fltP{=6gPu{|8?Ix092cEvq?2tn}C6Z=yFzr?Nky+W?uB9?d~_~+ zvTxfg3y1Hn7u&^PWeuwr>--=?)rrNAKbed*jPN|VwQ~NP7E$XjDD(cD%%+Y5%>0$c zE;ZS&r$hKq2OjGvk2D1xzmM2C{d^hc7NMRBaEn&#VNAxY$H@IeBg4+baFp7ZzYmc{ zOv9jk%4(gED@{Inhe7$C1hw(hUumaKAKEeE=bi|Ea@D8G_p-baO0dg<&M8e1kFj|+ z84n!m_>D0qZ`oc?gA~NgS!anK9e&vntWov9D&JuL_Brcz&n_lTUX~E_S$bAa?(jT7 zKO#Mbw3n%8990$K#Es04+GJ{SGn?3e0 zWfvxDaxVK^YE$&@{<9*?x?kR2g(tJ5#Ox57zphM}{FRYh(z8GA6yfhY%Hy;}Jja*Z{Kf~P(^EN+U~`cY zjp8fq+gT}y^+)@{88wApQQ5JO6tS-71n9$}{v zIsFUUhhw_97^}9-%GXX9w7C4VUiGWH!#|tY)9|SZhE?v%lQMX(CCEiL?-e$QPHWY= z7w#1-!Tgda3Gr|v;x}eKa^|MYfhX&{mI7WV{ST?6a#rmv_|`tbiaL3mHFP*UGxMNfL5}B3t}kx6k+c{r&s>^Y_QE|L*CWdz||oujlhI;qK0l+qGu^weq?SdcO41`{=mH9Y;LpQ`} zrUX`^V^&5nf^LQdQ=+qLy)<9#t~GEWYjV9BZE4#(F(plEEd9?%6wi*CWMZ;{F>JVx zF7(BKC}aP+=AukzKU;h(RjrPKEddp%xfJ7!7GFyPwnq$8aL zVlD#)pCM8C;^Iyf$m`1^tZF_F6?%PV*AmDx0CWoGmo!1j{j+JXBtb{nTtUxigW|DT zEbaG*cy#OaNwki}U>S{O9wZ)c^lO1CUH(jrZCp;ke+@@g5M@ukNS4aUie@ys%oLn~JK`DB;SS zViaXw@JDib))KHYUgqXPW$w)rH|e#$$8){H(;%F<{AtowGIs_iFD&$g3ckoxy2wPw z@I*ziI#Nat(y-F1XCq9(E@x`BDz)JfVw398S{WHES3Mv5El@-M1;DT<%Hj>ez77|H zrI~!nft3SGt;=tbcmH(Nk677U0F8hk>=~)}iL(+=OZ5-%^6c4C7e$M9vnzV95nZLx zM9Ox{AjL_|6c_D+Z^7k+SaZ$DJjGEi8qZc1X=Pb(a?nCw-ai^Ff`&>|56iGq{0c!0 zZ+OKF{lSeq_r9Qr8&t^{0!$kKc>u9|MQTYGVLwgDu2FODj8){Cj9vDAfr+So`I&5I;@K z(tp>a0l;(SD7MN%v$+1J6RqafA+4(eu3#IyuzrYIKdzv+;}&w|MoafN2#4zr2mF{` zH{3Pt`etpof2%ar;CC1HkiZCkYj7Q>0$K-+C+XMD5p$3OA!V2GR+8`RZe>-_swEGp zrvdbRj)wYMrUyIcWRV`1JL8IM612vN z4ZhBY;*yds!NE4GmCL1?)OMXRQ#9lLV3?$oleCfFHeiSVnFbX4*ARbcsMUte@d<-@ zsCJV|47l^BVsA`q&Ns?Y9p&m9UrDWNxe=Ya^#$gxOEJNpvSl?SJf;xIf4!H-CO7Z?f>-)vb>^voe^{Kb{U3&wAi7X_Z)?3k#hKzfAa_#Hj{K&VQK zOb{I~>P2S>srYQ0T}e%>vyX-f|J23#*+gex**sU`*Bk<4;eWY`s9lt^Am2qi;>{i_ zM3poPTaNq|EckL2-c=D!>C|c{tY_I-qo$*&zbGgvDS7{Ifvj!x^p}o!KbW}t)e>s0 zDGZ!+c1tcMjv{95bD_VbNmqo*E6x zPBr@m7L68CS@{6ISbxlJFZ2-OQ%3EnlVaZHzkgt>xj>wciG~c9ryY zr)(wvo88mK1G7;Bb5ZS-WF1k0+JQSi?@`vUo9$63$g?AB6X+v0t2jEEvR&C|T7K5y z!DeZ`7f(pTd-?tTrhTX7LH=fI(;=|@_#W-~b|N!^D*;<<1LSgK8sH^|Bx&{`;IVo= ziL>M^5JJ0XjYK&P1}#KC{8`%%z}v?j>m+CVLI_wtMZIj4;@alif!@j1_Dw;8T8veB zqi1yHpv6+X+weDXlt}lS2k#d-P(bpG%snTfWm z3kG`gGuXXuV4K=Fx+9gd9HMl5dmEj&t%7{;9xP>lC{3|^M#Sfqy)wU|i6N9d zeXQ!S>VL=7-vNE)_D*(JwNb|6$&x?#rTrB-nWWU!S^53?Ke<}(odDS+xfs^-fb&FVYwTJA-PDqN1-J1+&5%lAfPG4We}nJc*pl- z#tS^y|M;LCZ3TF664^;eCP(<9W|m$dldzsAfBo!Ufk=pzHksxC%muJkwt!*|vM@1> zE`n_z+C|3av%vxyp2Unp+J|{*N3pd4kJ;}>-Mf!~B6LozdAhvbj)%NIfu@NLO2Bi_ zM>=zQ+E=(pael=+9?KlBZyd9L=thvRV}4&}ZWnY1lUpOdkx8!9(sLfI2oHNTAmQZb z`o><}Q`!8b?2vlr6mlfznB2wM1u{AL47dE{C-3?zT-7S+O(pk!G^19Vf5p0~TqlN^ zTj6V}rvYV!`X-Fha?Hh+s%1PJ1$zKs3d!y@vO$uWe>@`G^|36}x=S(s_w+F80=WZ3 zij(Q4kucNsjvv~RZ>~E0W+d-IT|TVVA};~5GIzW-_pABwVKcb@wD#-tIUmje2!Q)k_YcPK}PvT9CNc4CK^V z3ZyDxATZC?t_kW|FlPEEj2F;wZw9Z;#oGCFHU{+nsX0}-4`F96`SLPE0D6`6B+U$C zwnuJz1v_x$^5Wrenoivd8&?NMEJ^Kbzw|l?gwY~esTMH8Sfelo=6t?j)m|W1tEMhS z<7>{9@%l#G(20tz za))hEaw-jNXXB!&{tIxclHHBuV56^x`+SUR77#K}*4+IU0gxZ?eH26;t0D_(S#szP zYRB#I1xeKJm)pxZZgb0y`io*)J~NFTQ;g7wV0LiMUz-g$hawfD^dA$@!@jS~_QAb< zdT%|b+wRAsX}9s-}|e1FUFzzirlO$h$>dAQ9hdgM3l0$6gX$ozPaLOLN?*q9W- zuCZ9~@jZ%9dU zfHu5%w}87z54x|qWzdUh_-)N9m*;AeKJ3S3uyDVxBbP8kV*Jp}aXD9;i`_peHu%!w zv7l*jNhTd;mw4|5OG>ch)`*3X?87`?`k%XTW~riJ)N?p^yUa>QO$I7=mT#KU44VsJ z#_QXL?se)yx6juq`J3h}W}Mecau-fssq34S&@F{M#FenUHc3Nn3c>#xEX(Ds(?q}$ ztr+BIJZCVXU{w%h%+KZ>v!Nqx@Nn?^xy3Z~29Q?2?qcgwlX#>iUwg{%4#G; z3ed(lPy5bIf84Fq(q0Yzw-#P#wpO3{LzMH)_o>6a`Nxg6m2Z;NW}JD}0`2fGH_EBI z9zM2LgSr5AS8|vt6aT(x0eEF>mctbGzjubWpx^-ppT-E+tjzvO z!eW|H9UD$}EPRG^Pql`6S4AF&gMF-Pq>Umi0w##1W?Wp-nfEPhxl?P(SmP4M3_V|v2DsNb z%=AI>>nLRPy8a8Wa>1v>?vK?;&6r9|>y24B4ljNHBdIhe6G$Jw!!S^iU|9p1>ZWKl z3dwQ9WMpA56PrGCZQxR@-ezBH*o$W%zGlH$r^27YD4U`y=>x6#oM6J?G9~9joM1!f zI2ixyO{&(2vi86N>D5@=V0XrbU+1LKxy2o6b^1(V@llxl0o=VM%KA}_Wmf{EN_kbk#bDBQe?~c=^g>l)_|j!v?TFDuGr0b z(f(i<;DK5fVxRBZF^;}U6Q$Wtr(70kT;wdb+&UzOHKVV*Ko^h zss0U;A(3^au+&!T!!Tmyo$s9=X*%L_R(Ex1sWjFLbu)BuS(VPyN@z%x9^mw~^%s9- z>faM(*QsZ~n@e!4c73K>sxKRrfzIKFaURnm>IHfdq3XQ8-#7Sxj~%gF8)oQs$=Y5L ziPk~7tE_Tvw^}~vL^cyAhV_yP1K%>680lUtU%3kt*Z8EDB?xrDDamMT>H6p5h|*}i z*A9Z9APU?I!#iHcC=(3UrunDFei(blj+v(YYXP9R`g;+^jx?%Ro4mF`!JVtCYePlm?kC?FT#e5w8QxZGpS^IF22%Zv z&5%b=pi3JzdwNqz0?29{uwQ>~4cD5rXczg>jB_EMC^5O8_H}CD0G6=%Ejw6bi)0r zj+t;AlvPVa*hs%rr;J2`U+q0OuT_bmvuL$0M#7v)Jw@Tcg%Tr_ug%TVfI;R~3|J;# zfuW7AG;)Dt0I5)xjgzvd|J2lw&XFphH_pWKn1rK0PkiN(!Scl_I`C<`Hf8HN!8w)q zlgDeff`jG4UYG5`8&u?p@!4T>S!099&2Pb;T-kAAc-(0eX#D^S1vaoguV1NlTtmB9y(DcWS1G)tHH`fvF}y7mQ(qDX*0Aqc16Sd zfr4z-?0mI34S$)k(^w6~&f$yFDJ0?Cm`1FpHEu3NIY!jFOU_b&O%9D*l;|^hN!Ucm zo>-!DRm(K%xO8}2`zgLh*k?{AG@MQSZT5pp`4ZH>KbmE@L2C7L0J{0KE5s^g;(g1e zf&}(@e@3~`vNprESuK}=`J@i=-sBVXzU@6an){6$;f!^JCurRK_kC(2b3G=DszAn2 z!RstT{u8OK$3!nup%n7f!KWLUOUhN0xZ}gFus|M@ZFf@a^27#<^}7;6V%e%r?KW1N zr*VVE&(6B5$pboSb@$MQ?!Wi4J|MuR`7d(yVm#!3`QZb0M!uxuPzSQOG4KdP4l*8& zt>3Wb2qj|ZKc=_ePbXAyY-O00iIad$t(wsH`Exd^AlEK%fba%aPyM7tF!A3(Snv^+z?Uw9Tr#Ms>!=E8x z`mD{B(hVFSX$FP*-41>a#s?iFuQyIk@$)ceCA|S3N)68$Tec?(TLoe{AIkWIXBvt8})Yt6K0eje332 zyGJc4zd^30^Zg0f9%R@VxPxwTmYqW){Dk1Bt`AO9Be}+7uCnS{44wTaa!9rh#K^8w z>O_fgJDed4k0V2)Dgi)*=EkZ|$)OyI0Y6Ok{U~_G9*mtGfR(Vxd1)~K_VJ85s2KPY z!P(($Rs+#Z5+p0O98L}h9PuzW)D%lV&y?v?=jUX014Mvi2MRu)4zwgDuYu{G%#!#n zrrf?TikD$W$5N;8ns{tF>fV>#uG>>uucCc? zS(jiryB*8(O+@eO7t7Cu06{&FOeP6(wZ|#|!j_vMAJ!)Lk<~=Hxv%^iIT~?UZC5F* zhI28JI5vH6BMsC%oY78Pf7QaeBiHiq_hoY70l-WF>+we6p{tV=6nvY*@n#x%8wa`g z6;%mjW_#kcBeTX@GVil}hhxKCatY-_BJs;Ja8!X)Qqd~t3hqgO<%O=?a(n}{1fX{n-1|u-^^~M_9`MdvhJ>!&ckClq7H|Z7u#yb-3+M)N zLjHe63c&AVUyUU8F!F zJ-^bMCt4_$v%5*IW>{;2!DlBgn3c)WgP*WYtdHz~bDyo-mh6yZp!y7!I@m9-X5#b9 zjrQw>XS6Upet>)bj5=Hcp(J^!K3PKvr`HJJsW|>ScLD+Fk$g!vAgQppO$5zEW&6i` zGw^2!H2hc7$hq9TC`&yF`LQ2fh=(~bZ>RNYlIt+p*Pzzj;GO{;AuYf)n20S;U`{hr znhLl(<xi;n9>`Ufzg&%JUBXvHu{p$F=Q&*f=Fz>v) z5`5|4ubRV7UD`ao`X1#sYl!>hC~JAo+fO&jlTGmT967~Q*)Y^R2a3`MNJ8fiJ-fZQ zgPdS~u7OMg<+k3Bw-oLJkXk^Jj>N?|6N;Y@+ZakPiukO24^v>xH}M= z)cHv}pLgQTpQ-p`mc_=4jUf0qq&AVLz__Ya%7H4nlL)>OY`(8M2-P_~&XQ=H%`@ z1t>aD0f6RNdU9P#DZ(&ydiW!t{YZcy)iQYttBl;84J0C37UGpb%}RrW!n7$J)F2Jf zlEJ1my-9|4azRe)F978jF3ZNep4rqZJxC<;;{EZEKF~&6OmwPrQ?8(8QT$6z3<6gm z{9p?auEIvvW1yCUQ1m{Jh{OO2aChG)Ue`xz)hq|^jTfu^6Vw7D-bHYoMs64DXsGRp zz)@11^x>pYJJzlZfp`}(15K7+{QkVG2GHr(a~mA4g+k!00K}!D{}L0JXhDjqdrC^> zhg2pS_s|7v47y(f5TX#ypB7X6waILka}~1aEFn&8G&2v<1x^h_uj~P15R`#RmHW<+ zzoo5^R!ONAtUA=&oZc!LP5Q+fc*!G^cR=fUO+y1%AVC0O`I#_T_Yy4zWJxR+M47`0*ld`d;4z5}{VWTFwVEgn=Fm>fCM3W&o@Dr#==62?HO@enL$kjnA1EGY~ zEpcOtOyc9p23yYy&*JAED2`L=*z5?=u9}fYXnOw5#M!pZPhX^S(bp%-4VyeTRtjs+ zH7d2Q=3itmC)hlLsTgBuX2BKeOM>qYXxq~8+8Dz3LgWA9a2u&$x{WF zkG9*D8%`7*?y4+bo=tSa|4?r0B&Dz*uubLx^j zIimefA)V;f1tV%jQLX1%Y?=jt#MtM2>NlP4v-DDC`OxzeHH0d)AYRG>|AdofB*R^1 zlj)M8U*BP9_m>xUShc?2M`hgRtET4};9bS-T0VHIQ2GSI7Re3c@U3uF1Mg}M=)}RM zoMpsBL%s?;gYkiiB=2>iR1fhpT?ZvOWQR1_=(}c?mb1(rNtYUjsK1aU&nT9akt=2% zp}trg?}xdSF&sDz^n0sirTtznMbp)Hu4(4odFkUPXE4Js|ZWuTV z#1mX!+257~n!~F^2i!w;bg!Hxwf6WaJK`DWR!`&szy*53SJIpT?Llu=X2Z^J5q_^6 zzun^|{DND>CD%9uBWI?MF#l#f_^M{M`LrF3jbi+k#d#F~n4}|rwcUD(-nW=8dM(i@ z*O;J?a5_%MX2s+DRH=SOkp59_x5vE__r=iE4as8K^O(S`R2wjD7h(nA9VemtNI{F`DnXz|XsVois$ ztjJ3t_~mPczIisV+E8qUUyYym5zG!WC2+X5^4+d#SL?EiJD^%Dg5^nRq-o)o&f4D4 z^lk8by>-yHd01P|ICnde32+EAEK}LMO}u&Q>{>tIdSKKm!@EsV!3!qLWV1{QCmX1P zM=W(Ru_3`Mi+y6dg8AzsL+0_xH^hU>BNy>9@-IOl`s*}L`A)l6SX?hNm3$j&{j(By zRfDL3&BbyNeDbuqQKp!=CeLclCRe{_xYy_&>Uce!4cDwYHo;Tv1#h~?2mVS2Sv-p} zo>VM;iswQ5o1x2kgogT9Z$Q{F%Sj9LudV;ioo zezCY^ZROyGH7C?2P;CvmMdgQ>a+r8XsgCL0Pm<*y-{j*yufY|Z(zK31AAHBjA%V^Ugrt)nfZ_@y{v4m98l3O$FvGudMjg&w3%GS%9Ns`p4Vg99x!T z_|7bvqGBfQ=}=jvLy4@HI_hR{?X0o{nwg}wrPppDOl^au0)%S;=BDh4LZOAMo|rL^ zcDGwi^lC)hiUDFLPi!fp;wtGG(#U?u$1L>*)dJRw27A zVc}EYul#{e(`$k4Gl_`T6K0Z{vZ^?4OrIXGa2B77AF91$le)oTMGFrG)&PJSb>()m zng+4HN(kRP5E1oOy%Qt}uK)^!@a#a{=b@?CShOeuD1~-T-}oY@ZwEv@2#Mc$5>$hM z7=@GR>mB~I1X;@FIXe558KE+&ru_1FfO=d36RaQgESydm39Qkli`lRjFqh5vD zk2v7?Hl`ODKbbdKJFrO8;oC@dzYT_0I_kz`$l?S!FQFEs!c**@wIZsT>38QU(ETi;kC9fY|kT#X0x5cc-Z;JiL+q1r(ULQ8ZJ_wG~Xh z3vn*Kdu;4B@pdcbK2vE!N6wsHYsjK zO0xJ@aNHThlSLW4PUg+kNxW^&u<3`o1vj3LmqVQi73%R(6OFwsf_iGnm!mSJU6hSr z;(S52Hb6*YW`%fwX5~a^^o?uJNe*4ISXmLAXVNgwV5nxlJjbJgx2Pea338V8R`>1W zHMnTe-ly_!9#K3x^#90BeYIY)ZuS~W=p}>zq|X$>W=^jZrBu4&=6cej$C1%I0stti&R3w^8$C4L6IZHo3Y zR>k9{4&4_p-Q!;Fe0))`2k*DWt)7GPjjVTn@Rb6!MnEwGA+E*#g?@DY$fKN@oi(s0 z6CiR=bLd<<8B7D3k_3D6uSfsNBy`;5?ReWEhim{EXV8Oi^St&!)zd2+)c zg=f}h(^2a2PW4@X^=P0w0)Y0s;pfQ*Nr0`gUL!YW&fl<|bTcPk(|w@ce+Y7O;{2C~ zJUZQ^%xE{10g?hV_7ob3<#f2Hd)@SPsAbL8W?J_}uqTa0+V{=Z1eTj)gMcoV%;CQ> z!w7?c`y~;kWXFn*BYL+pgPpyW6xa@iMQ?*@z@B2Wys7f&tb?0gR{jNxby*mo3EmIKT`7V+kHXx|Tq-_M!X}`DZwz)*PJ9|^6<5yc=o%`m zcg8w@$rXID?&Xet$h_rh&c{Fg<)_y7WrJT1#IXCl7HJ|ef1Gd2)g7U@>wvl!fNBlW zr=DnRwEM{Fre(?*l2!b(rHuJI!wD{p8zGnhAY9bhn{o}HL+r+#CMJx$;7c}=jCTQl zGQ=HFqJl4GI8fZ1b!SXA$hcww?yZ#(EI+*L5p*kzxcVn;<&%;OlDEWI5F2oD$S(G*b-&g#9RYbu6 zHyKj!x!p?S&`gvU8Z&;XTfv@U^KeQ%T_yudlIjypk7wHt7X1lCB!GmIa^}id#!@`E zQ)nS9R?qs<1=+(1Z-@ur6X=M25?AbPc-5UNjhUIoAkxrHH0H9P-jl%g&eA9!;Qkano&j z1;LEPeC#prYcv1e2WuIG?&lCICZ`dr!!=mT@-nLTH0kAOmNjws-9n0Wqrzx8^mF zG+cGdnT$l^l;|z&r@p^52@>JO9zAsU8%{lTmmBTD*Bw-^VdFdSnhx%0VJt@xf z2mGXH5Bf5x#Wz1sM|^^>nMnt)*i)Z}DO^sOmaB_jF|S&#la^8rH0yG!WH9A@pxYY4 z3`wbB;Z_yBh>30sxS8+8w4aqbMv*~~$!+S0KdrpLXR)0;4sP*=8+E06$2B}BKqRB^Y>lHB0jN!gRd zYpc2Av()DyoU*F0U?fDodKnoW_7)9-7+$5XqI96SY^~>dPn5pG>%m4n8ATmj!}iRx z)jkDtM^cOxPgl~Ru(BSK-|3oZle%(H3RD5A`a_Ex7vD9}UKQinPC%>&d+yW@4UzSv zCYabu&*nd@#GPKuT&v~-c0`PF$H1kEfwyr&}?DL>fccy%DlLuda) zPHHv24TtJ;Z5~9$?oLMl2hiAz@gcN!$Q##*#Vx_(cM*Q8Jt^c zDPK0>w^Fi^&z(90ar@IxLiM~;7X>~kh;WQUm&EmwY?WkJGZ$q%^ta5Cup|t{!4=cO z)HRb0VP@(Z^+qMUhTB^EA7mB<(wjVFYx5&Ck_3Td8yI`Q`3M*hf0}R zalY=@@0*@Q1I3ac$m;WR={Ua`g}sYvku2_<@y#z2G8CjwmPXGmVIWkHe7)-nE-;;e zVa3I)3wKuW@v8Xo%vr2T-Is^<=S$^}z{zdSl~~wu4+t88k!2y7E7Txa4DT3|^Gvq( z447TzEDI{cShwBJXXRi;A1sR~>O`%QTsv2MEo`2DdnAI~m6?)4;A}@1&WiAFaxvwp zb+&E3s|`8hYYPNV_QK(eeg;{p&^oclHV*oWF@kzsLS5e9mE$LD*MQJOm86fUOlId} zr2+EujVSQS-l)uq&`r9L@OCQ%_fovm*V#V3TM6_J>S(MC1N0a4Y-+xbVgqmD#)yl+ zm}r+Ci1G3k^rEgG?BCYO{Mt@2PD$!5SsMzucPZkaD4gfrGdik5QLQwYAQs42|3`QN zl7fQjm);Fci`|J2GI25~xpY(?bZ$hYJMcMSHH&1o#s#Zvte9w?I2|<-#M(=LEzKF= zD|j{Ho*TFI(V!ET-MLny1!Bi^eN=&__BVNQ;Fi1`2LiFa@+_i}vXx)0aVi>jA!hW51LDRm2%@Lw^D!P<@)A Date: Mon, 29 Nov 2021 11:19:33 +0800 Subject: [PATCH 29/31] delete repeat doc delete repeat doc --- code_and_doc.md | 324 ------------------------------------------------ 1 file changed, 324 deletions(-) delete mode 100644 code_and_doc.md diff --git a/code_and_doc.md b/code_and_doc.md deleted file mode 100644 index 1c006740c3..0000000000 --- a/code_and_doc.md +++ /dev/null @@ -1,324 +0,0 @@ -# 附录 - -本附录包含了Python、文档规范以及Pull Request流程，请各位开发者遵循相关内容 - -- [附录1：Python代码规范](#附录1) - -- [附录2：文档规范](#附录2) - -- [附录3：Pull Request说明](#附录3) - - - -## 附录1：Python代码规范 - -PaddleOCR的Python代码遵循 [PEP8规范](https://www.python.org/dev/peps/pep-0008/)，其中一些关注的重点包括如下内容 - -- 空格 - - - 空格应该加在逗号、分号、冒号前，而非他们的后面 - - ```python - # 正确： - print(x, y) - - # 错误： - print(x , y) - ``` - - - 在函数中指定关键字参数或默认参数值时, 不要在其两侧使用空格 - - ```python - # 正确： - def complex(real, imag=0.0) - # 错误： - def complex(real, imag = 0.0) - ``` - -- 注释 - - - 行内注释：行内注释使用 `#` 号表示，在代码与 `#` 之间需要空两个空格， `#` 与注释之间应当空一个空格，例如 - - ```python - x = x + 1 # Compensate for border - ``` - - - 函数和方法：每个函数的定义后的描述应该包括以下内容： - - - 函数描述：函数的作用，输入输出的 - - - Args：每个参数的名字以及对该参数的描述 - - Returns：返回值的含义和类型 - - ```python - def fetch_bigtable_rows(big_table, keys, other_silly_variable=None): - """Fetches rows from a Bigtable. - - Retrieves rows pertaining to the given keys from the Table instance - represented by big_table. Silly things may happen if - other_silly_variable is not None. - - Args: - big_table: An open Bigtable Table instance. - keys: A sequence of strings representing the key of each table row - to fetch. - other_silly_variable: Another optional variable, that has a much - longer name than the other args, and which does nothing. - - Returns: - A dict mapping keys to the corresponding table row data - fetched. Each row is represented as a tuple of strings. For - example: - - {'Serak': ('Rigel VII', 'Preparer'), - 'Zim': ('Irk', 'Invader'), - 'Lrrr': ('Omicron Persei 8', 'Emperor')} - - If a key from the keys argument is missing from the dictionary, - then that row was not found in the table. - """ - pass - ``` - - - -## 附录2：文档规范 - -### 2.1 总体说明 - -- 文档位置：如果您增加的新功能可以补充在原有的Markdown文件中，请**不要重新新建**一个文件。如果您对添加的位置不清楚，可以先PR代码，然后在commit中询问官方人员。 - -- 新增Markdown文档名称：使用英文描述文档内容，一般由小写字母与下划线组合而成，例如 `add_new_algorithm.md` - -- 新增Markdown文档格式：目录 - 正文 - FAQ - - > 目录生成方法可以使用 [此网站](https://ecotrust-canada.github.io/markdown-toc/) 将md内容复制之后自动提取目录，然后在md文件的每个标题前添加 `` - -- 中英双语：任何对文档的改动或新增都需要分别在中文和英文文档上进行。 - -### 2.2 格式规范 - -- 标题格式：文档标题格式按照：阿拉伯数字小数点组合 - 空格 - 标题的格式（例如 `2.1 XXXX` ， `2. XXXX`） - -- 代码块：通过代码块格式展示需要运行的代码，在代码块前描述命令参数的含义。例如： - - > 检测+方向分类器+识别全流程：设置方向分类器参数 `--use_angle_cls true` 后可对竖排文本进行识别。 - > - > ``` - > paddleocr --image_dir ./imgs/11.jpg --use_angle_cls true - > ``` - -- 变量引用：如果在行内引用到代码变量或命令参数，需要用行内代码表示，例如上方 `--use_angle_cls true` ，并在前后各空一格 - -- 补充说明：通过引用格式 `>` 补充说明，或对注意事项进行说明 - -- 图片：如果在说明文档中增加了图片，请规范图片的命名形式（描述图片内容），并将图片添加在 `doc/` 下 - - - -## 附录3：Pull Request说明 - -### 3.1 PaddleOCR分支说明 - -PaddleOCR未来将维护2种分支，分别为： - -- release/x.x系列分支：为稳定的发行版本分支，会适时打tag发布版本，适配Paddle的release版本。当前最新的分支为release/2.0分支，是当前默认分支，适配Paddle v2.0.0。随着版本迭代，release/x.x系列分支会越来越多，默认维护最新版本的release分支，前1个版本分支会修复bug，其他的分支不再维护。 -- dygraph分支：为开发分支，适配Paddle动态图的dygraph版本，主要用于开发新功能。如果有同学需要进行二次开发，请选择dygraph分支。为了保证dygraph分支能在需要的时候拉出release/x.x分支，dygraph分支的代码只能使用Paddle最新release分支中有效的api。也就是说，如果Paddle dygraph分支中开发了新的api，但尚未出现在release分支代码中，那么请不要在PaddleOCR中使用。除此之外，对于不涉及api的性能优化、参数调整、策略更新等，都可以正常进行开发。 - -PaddleOCR的历史分支，未来将不再维护。考虑到一些同学可能仍在使用，这些分支还会继续保留： - -- develop分支：这个分支曾用于静态图的开发与测试，目前兼容>=1.7版本的Paddle。如果有特殊需求，要适配旧版本的Paddle，那还可以使用这个分支，但除了修复bug外不再更新代码。 - -PaddleOCR欢迎大家向repo中积极贡献代码，下面给出一些贡献代码的基本流程。 - -### 3.2 PaddleOCR代码提交流程与规范 - -> 如果你熟悉Git使用，可以直接跳转到 [3.2.10 提交代码的一些约定](#提交代码的一些约定) - -#### 3.2.1 创建你的 `远程仓库` - -- 在PaddleOCR的 [GitHub首页](https://github.com/PaddlePaddle/PaddleOCR)，点击左上角 `Fork` 按钮，在你的个人目录下创建 `远程仓库`，比如`https://github.com/{your_name}/PaddleOCR`。 - -![banner](/Users/zhulingfeng01/OCR/PaddleOCR/doc/banner.png) - -- 将 `远程仓库` Clone到本地 - -``` -# 拉取develop分支的代码 -git clone https://github.com/{your_name}/PaddleOCR.git -b dygraph -cd PaddleOCR -``` - -> 多数情况下clone失败是由于网络原因，请稍后重试或配置代理 - -#### 3.2.2 和 `远程仓库` 建立连接 - -首先查看当前 `远程仓库` 的信息。 - -``` -git remote -v -# origin https://github.com/{your_name}/PaddleOCR.git (fetch) -# origin https://github.com/{your_name}/PaddleOCR.git (push) -``` - -只有clone的 `远程仓库` 的信息，也就是自己用户名下的 PaddleOCR，接下来我们创建一个原始 PaddleOCR 仓库的远程主机，命名为 upstream。 - -``` -git remote add upstream https://github.com/PaddlePaddle/PaddleOCR.git -``` - -使用 `git remote -v` 查看当前 `远程仓库` 的信息，输出如下，发现包括了origin和upstream 2个 `远程仓库` 。 - -``` -origin https://github.com/{your_name}/PaddleOCR.git (fetch) -origin https://github.com/{your_name}/PaddleOCR.git (push) -upstream https://github.com/PaddlePaddle/PaddleOCR.git (fetch) -upstream https://github.com/PaddlePaddle/PaddleOCR.git (push) -``` - -这主要是为了后续在提交pull request(PR)时，始终保持本地仓库最新。 - -#### 3.2.3 创建本地分支 - -可以基于当前分支创建新的本地分支，命令如下。 - -``` -git checkout -b new_branch -``` - -也可以基于远程或者上游的分支创建新的分支，命令如下。 - -``` -# 基于用户远程仓库(origin)的develop创建new_branch分支 -git checkout -b new_branch origin/develop -# 基于上游远程仓库(upstream)的develop创建new_branch分支 -# 如果需要从upstream创建新的分支，需要首先使用git fetch upstream获取上游代码 -git checkout -b new_branch upstream/develop -``` - -最终会显示切换到新的分支，输出信息如下 - -``` -Branch new_branch set up to track remote branch develop from upstream. -Switched to a new branch 'new_branch' -``` - -#### 3.2.4 使用pre-commit勾子 - -Paddle 开发人员使用 pre-commit 工具来管理 Git 预提交钩子。 它可以帮助我们格式化源代码（C++，Python），在提交（commit）前自动检查一些基本事宜（如每个文件只有一个 EOL，Git 中不要添加大文件等）。 - -pre-commit测试是 Travis-CI 中单元测试的一部分，不满足钩子的 PR 不能被提交到 PaddleOCR，首先安装并在当前目录运行它： - -``` -pip install pre-commit -pre-commit install -``` - - > 1. Paddle 使用 clang-format 来调整 C/C++ 源代码格式，请确保 `clang-format` 版本在 3.8 以上。 - > - > 2. 通过pip install pre-commit和conda install -c conda-forge pre-commit安装的yapf稍有不同的，PaddleOCR 开发人员使用的是 `pip install pre-commit`。 - -#### 3.2.5 修改与提交代码 - - 假设对PaddleOCR的 `README.md` 做了一些修改，可以通过 `git status` 查看改动的文件，然后使用 `git add` 添加改动文件。 - -``` -git status # 查看改动文件 -git add README.md -pre-commit -``` - -重复上述步骤，直到pre-comit格式检查不报错。如下所示。 - -[![img](https://github.com/PaddlePaddle/PaddleClas/raw/release/2.3/docs/images/quick_start/community/003_precommit_pass.png)](https://github.com/PaddlePaddle/PaddleClas/blob/release/2.3/docs/images/quick_start/community/003_precommit_pass.png) - -使用下面的命令完成提交。 - -``` -git commit -m "your commit info" -``` - -#### 3.2.6 保持本地仓库最新 - -获取 upstream 的最新代码并更新当前分支。这里的upstream来自于2.2节的`和远程仓库建立连接`部分。 - -``` -git fetch upstream -# 如果是希望提交到其他分支，则需要从upstream的其他分支pull代码，这里是develop -git pull upstream develop -``` - -#### 3.2.7 push到远程仓库 - -``` -git push origin new_branch -``` - -#### 3.2.7 提交Pull Request - -点击new pull request，选择本地分支和目标分支，如下图所示。在PR的描述说明中，填写该PR所完成的功能。接下来等待review，如果有需要修改的地方，参照上述步骤更新 origin 中的对应分支即可。 - -![banner](/Users/zhulingfeng01/OCR/PaddleOCR/doc/pr.png) - -#### 3.2.8 签署CLA协议和通过单元测试 - -- 签署CLA 在首次向PaddlePaddle提交Pull Request时，您需要您签署一次CLA(Contributor License Agreement)协议，以保证您的代码可以被合入，具体签署方式如下： - - 1. 请您查看PR中的Check部分，找到license/cla，并点击右侧detail，进入CLA网站 - - 2. 点击CLA网站中的“Sign in with GitHub to agree”,点击完成后将会跳转回您的Pull Request页面 - -#### 3.2.9 删除分支 - -- 删除远程分支 - - 在 PR 被 merge 进主仓库后，我们可以在 PR 的页面删除远程仓库的分支。 - - 也可以使用 `git push origin :分支名` 删除远程分支，如： - - ``` - git push origin :new_branch - ``` - -- 删除本地分支 - - ``` - # 切换到develop分支，否则无法删除当前分支 - git checkout develop - - # 删除new_branch分支 - git branch -D new_branch - ``` - - - -#### 3.2.10 提交代码的一些约定 - -为了使官方维护人员在评审代码时更好地专注于代码本身，请您每次提交代码时，遵守以下约定： - -1）请保证Travis-CI 中单元测试能顺利通过。如果没过，说明提交的代码存在问题，官方维护人员一般不做评审。 - -2）提交Pull Request前： - -- 请注意commit的数量。 - - 原因：如果仅仅修改一个文件但提交了十几个commit，每个commit只做了少量的修改，这会给评审人带来很大困扰。评审人需要逐一查看每个commit才能知道做了哪些修改，且不排除commit之间的修改存在相互覆盖的情况。 - - 建议：每次提交时，保持尽量少的commit，可以通过git commit --amend补充上次的commit。对已经Push到远程仓库的多个commit，可以参考[squash commits after push](https://stackoverflow.com/questions/5667884/how-to-squash-commits-in-git-after-they-have-been-pushed)。 - -- 请注意每个commit的名称：应能反映当前commit的内容，不能太随意。 - - -3）如果解决了某个Issue的问题，请在该Pull Request的第一个评论框中加上：fix #issue_number，这样当该Pull Request被合并后，会自动关闭对应的Issue。关键词包括：close, closes, closed, fix, fixes, fixed, resolve, resolves, resolved，请选择合适的词汇。详细可参考[Closing issues via commit messages](https://help.github.com/articles/closing-issues-via-commit-messages)。 - -此外，在回复评审人意见时，请您遵守以下约定： - -1）官方维护人员的每一个review意见都希望得到回复，这样会更好地提升开源社区的贡献。 - -- 对评审意见同意且按其修改完的，给个简单的Done即可； -- 对评审意见不同意的，请给出您自己的反驳理由。 - -2）如果评审意见比较多: - -- 请给出总体的修改情况。 -- 请采用`start a review`进行回复，而非直接回复的方式。原因是每个回复都会发送一封邮件，会造成邮件灾难。 \ No newline at end of file From 7c15d1b7902d7b38f1246d7de587b505cc3585a9 Mon Sep 17 00:00:00 2001 From: WenmuZhou <572459439@qq.com> Date: Mon, 29 Nov 2021 08:38:15 +0000 Subject: [PATCH 30/31] change is to does --- ppocr/utils/save_load.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/ppocr/utils/save_load.py b/ppocr/utils/save_load.py index a5326edd69..f6013a4066 100644 --- a/ppocr/utils/save_load.py +++ b/ppocr/utils/save_load.py @@ -57,7 +57,7 @@ def load_model(config, model, optimizer=None): if checkpoints.endswith('.pdparams'): checkpoints = checkpoints.replace('.pdparams', '') assert os.path.exists(checkpoints + ".pdparams"), \ - "The {}.pdparams is not exists!".format(checkpoints) + "The {}.pdparams does not exists!".format(checkpoints) # load params from trained model params = paddle.load(checkpoints + '.pdparams') @@ -106,7 +106,7 @@ def load_pretrained_params(model, path): if path.endswith('.pdparams'): path = path.replace('.pdparams', '') assert os.path.exists(path + ".pdparams"), \ - "The {}.pdparams is not exists!".format(path) + "The {}.pdparams does not exists!".format(path) params = paddle.load(path + '.pdparams') state_dict = model.state_dict() From 4eb970488e45474705b27830c0904c43abbe5ef0 Mon Sep 17 00:00:00 2001 From: littletomatodonkey Date: Tue, 30 Nov 2021 10:51:37 +0800 Subject: [PATCH 31/31] add infer doc (#4791) * add infer doc * fix trt sub size --- doc/doc_ch/inference.md | 126 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 126 insertions(+) diff --git a/doc/doc_ch/inference.md b/doc/doc_ch/inference.md index 4e0f1d131e..c964d23117 100755 --- a/doc/doc_ch/inference.md +++ b/doc/doc_ch/inference.md @@ -34,6 +34,8 @@ inference 模型（`paddle.jit.save`保存的模型） - [1. 超轻量中文OCR模型推理](#超轻量中文OCR模型推理) - [2. 其他模型推理](#其他模型推理) +- [六、参数解释](参数解释) + ## 一、训练模型转inference模型 @@ -394,3 +396,127 @@ python3 tools/infer/predict_system.py --image_dir="./doc/imgs_en/img_10.jpg" --d 执行命令后，识别结果图像如下： ![](../imgs_results/img_10_east_starnet.jpg) + + + + +# 六、参数解释 + +更多关于预测过程的参数解释如下所示。 + +* 全局信息 + +| 参数名称 | 类型 | 默认值 | 含义 | +| :--: | :--: | :--: | :--: | +| image_dir | str | 无，必须显式指定 | 图像或者文件夹路径 | +| vis_font_path | str | "./doc/fonts/simfang.ttf" | 用于可视化的字体路径 | +| drop_score | float | 0.5 | 识别得分小于该值的结果会被丢弃，不会作为返回结果 | +| use_pdserving | bool | False | 是否使用Paddle Serving进行预测 | +| warmup | bool | False | 是否开启warmup，在统计预测耗时的时候，可以使用这种方法 | +| draw_img_save_dir | str | "./inference_results" | 系统串联预测OCR结果的保存文件夹 | +| save_crop_res | bool | False | 是否保存OCR的识别文本图像 | +| crop_res_save_dir | str | "./output" | 保存OCR识别出来的文本图像路径 | +| use_mp | bool | False | 是否开启多进程预测 | +| total_process_num | int | 6 | 开启的进城数，`use_mp`为`True`时生效 | +| process_id | int | 0 | 当前进程的id号，无需自己修改 | +| benchmark | bool | False | 是否开启benchmark，对预测速度、显存占用等进行统计 | +| save_log_path | str | "./log_output/" | 开启`benchmark`时，日志结果的保存文件夹 | +| show_log | bool | True | 是否显示预测中的日志信息 | +| use_onnx | bool | False | 是否开启onnx预测 | + + +* 预测引擎相关 + +| 参数名称 | 类型 | 默认值 | 含义 | +| :--: | :--: | :--: | :--: | +| use_gpu | bool | True | 是否使用GPU进行预测 | +| ir_optim | bool | True | 是否对计算图进行分析与优化，开启后可以加速预测过程 | +| use_tensorrt | bool | False | 是否开启tensorrt | +| min_subgraph_size | int | 15 | tensorrt中最小子图size，当子图的size大于该值时，才会尝试对该子图使用trt engine计算 | +| precision | str | fp32 | 预测的精度，支持`fp32`, `fp16`, `int8` 3种输入 | +| enable_mkldnn | bool | True | 是否开启mkldnn | +| cpu_threads | int | 10 | 开启mkldnn时，cpu预测的线程数 | + +* 文本检测模型相关 + +| 参数名称 | 类型 | 默认值 | 含义 | +| :--: | :--: | :--: | :--: | +| det_algorithm | str | "DB" | 文本检测算法名称，目前支持`DB`, `EAST`, `SAST`, `PSE` | +| det_model_dir | str | xx | 检测inference模型路径 | +| det_limit_side_len | int | 960 | 检测的图像边长限制 | +| det_limit_type | str | "max" | 检测的变成限制类型，目前支持`min`, `max`，`min`表示保证图像最短边不小于`det_limit_side_len`，`max`表示保证图像最长边不大于`det_limit_side_len` | + +其中，DB算法相关参数如下 + +| 参数名称 | 类型 | 默认值 | 含义 | +| :--: | :--: | :--: | :--: | +| det_db_thresh | float | 0.3 | DB输出的概率图中，得分大于该阈值的像素点才会被认为是文字像素点 | +| det_db_box_thresh | float | 0.6 | 检测结果边框内，所有像素点的平均得分大于该阈值时，该结果会被认为是文字区域 | +| det_db_unclip_ratio | float | 1.5 | `Vatti clipping`算法的扩张系数，使用该方法对文字区域进行扩张 | +| max_batch_size | int | 10 | 预测的batch size | +| use_dilation | bool | False | 是否对分割结果进行膨胀以获取更优检测效果 | +| det_db_score_mode | str | "fast" | DB的检测结果得分计算方法，支持`fast`和`slow`，`fast`是根据polygon的外接矩形边框内的所有像素计算平均得分，`slow`是根据原始polygon内的所有像素计算平均得分，计算速度相对较慢一些，但是更加准确一些。 | + +EAST算法相关参数如下 + +| 参数名称 | 类型 | 默认值 | 含义 | +| :--: | :--: | :--: | :--: | +| det_east_score_thresh | float | 0.8 | EAST后处理中score map的阈值 | +| det_east_cover_thresh | float | 0.1 | EAST后处理中文本框的平均得分阈值 | +| det_east_nms_thresh | float | 0.2 | EAST后处理中nms的阈值 | + +SAST算法相关参数如下 + +| 参数名称 | 类型 | 默认值 | 含义 | +| :--: | :--: | :--: | :--: | +| det_sast_score_thresh | float | 0.5 | SAST后处理中的得分阈值 | +| det_sast_nms_thresh | float | 0.5 | SAST后处理中nms的阈值 | +| det_sast_polygon | bool | False | 是否多边形检测，弯曲文本场景（如Total-Text）设置为True | + +PSE算法相关参数如下 + +| 参数名称 | 类型 | 默认值 | 含义 | +| :--: | :--: | :--: | :--: | +| det_pse_thresh | float | 0.0 | 对输出图做二值化的阈值 | +| det_pse_box_thresh | float | 0.85 | 对box进行过滤的阈值，低于此阈值的丢弃 | +| det_pse_min_area | float | 16 | box的最小面积，低于此阈值的丢弃 | +| det_pse_box_type | str | "box" | 返回框的类型，box:四点坐标，poly: 弯曲文本的所有点坐标 | +| det_pse_scale | int | 1 | 输入图像相对于进后处理的图的比例，如`640*640`的图像，网络输出为`160*160`，scale为2的情况下，进后处理的图片shape为`320*320`。这个值调大可以加快后处理速度，但是会带来精度的下降 | + +* 文本识别模型相关 + +| 参数名称 | 类型 | 默认值 | 含义 | +| :--: | :--: | :--: | :--: | +| rec_algorithm | str | "CRNN" | 文本识别算法名称，目前支持`CRNN`, `SRN`, `RARE`, `NETR`, `SAR` | +| rec_model_dir | str | 无，如果使用识别模型，该项是必填项 | 识别inference模型路径 | +| rec_image_shape | list | [3, 32, 320] | 识别时的图像尺寸， | +| rec_batch_num | int | 6 | 识别的batch size | +| max_text_length | int | 25 | 识别结果最大长度，在`SRN`中有效 | +| rec_char_dict_path | str | "./ppocr/utils/ppocr_keys_v1.txt" | 识别的字符字典文件 | +| use_space_char | bool | True | 是否包含空格，如果为`True`，则会在最后字符字典中补充`空格`字符 | + + +* 端到端文本检测与识别模型相关 + +| 参数名称 | 类型 | 默认值 | 含义 | +| :--: | :--: | :--: | :--: | +| e2e_algorithm | str | "PGNet" | 端到端算法名称，目前支持`PGNet` | +| e2e_model_dir | str | 无，如果使用端到端模型，该项是必填项 | 端到端模型inference模型路径 | +| e2e_limit_side_len | int | 768 | 端到端的输入图像边长限制 | +| e2e_limit_type | str | "max" | 端到端的边长限制类型，目前支持`min`, `max`，`min`表示保证图像最短边不小于`e2e_limit_side_len`，`max`表示保证图像最长边不大于`e2e_limit_side_len` | +| e2e_pgnet_score_thresh | float | xx | xx | +| e2e_char_dict_path | str | "./ppocr/utils/ic15_dict.txt" | 识别的字典文件路径 | +| e2e_pgnet_valid_set | str | "totaltext" | 验证集名称，目前支持`totaltext`, `partvgg`，不同数据集对应的后处理方式不同，与训练过程保持一致即可 | +| e2e_pgnet_mode | str | "fast" | PGNet的检测结果得分计算方法，支持`fast`和`slow`，`fast`是根据polygon的外接矩形边框内的所有像素计算平均得分，`slow`是根据原始polygon内的所有像素计算平均得分，计算速度相对较慢一些，但是更加准确一些。 | + + +* 方向分类器模型相关 + +| 参数名称 | 类型 | 默认值 | 含义 | +| :--: | :--: | :--: | :--: | +| use_angle_cls | bool | False | 是否使用方向分类器 | +| cls_model_dir | str | 无，如果需要使用，则必须显式指定路径 | 方向分类器inference模型路径 | +| cls_image_shape | list | [3, 48, 192] | 预测尺度 | +| label_list | list | ['0', '180'] | class id对应的角度值 | +| cls_batch_num | int | 6 | 方向分类器预测的batch size | +| cls_thresh | float | 0.9 | 预测阈值，模型预测结果为180度，且得分大于该阈值时，认为最终预测结果为180度，需要翻转 |