@@ -57,55 +61,42 @@
## Introduction
-[English](/README.md) | 简体中文
+MMPreTrain 是一款基于 PyTorch 的开源深度学习预训练工具箱，是 [OpenMMLab](https://openmmlab.com/) 项目的成员之一
-MMClassification 是一款基于 PyTorch 的开源图像分类工具箱，是 [OpenMMLab](https://openmmlab.com/) 项目的成员之一
-
-主分支代码目前支持 PyTorch 1.5 以上的版本。
-
-
-
-
+`主分支`代码目前支持 PyTorch 1.8 以上的版本。
### 主要特性
- 支持多样的主干网络与预训练模型
-- 支持配置多种训练技巧
+- 支持多种训练策略（有监督学习，无监督学习等）
+- 提供多种训练技巧
- 大量的训练配置文件
- 高效率和高可扩展性
-- 功能强大的工具箱
+- 功能强大的工具箱，有助于模型分析和实验
## 更新日志
-2022/12/30 发布了 v1.0.0rc5 版本
+🌟 2023/4/6 发布了 v1.0.0rc6 版本
-- 支持了**EVA**, **RevViT**, **EfficientnetV2**, **CLIP**, **TinyViT** 和 **MixMIM** 等骨干网络结构
+- 整和来自 MMSelfSup 的自监督学习算法，例如 `MAE`, `BEiT`, `MILAN` 等
+- 支持 t-SNE 可视化
+- 重构数据管道可视化
+
+之前版本更新内容
+
+- 支持了 **LeViT**, **XCiT**, **ViG**, **ConvNeXt-V2**, **EVA**, **RevViT**, **EfficientnetV2**, **CLIP**, **TinyViT** 和 **MixMIM** 等骨干网络结构
- 复现了 ConvNeXt 和 RepVGG 的训练精度。
-- 支持了 **多任务** 训练和测试，详见 [#1229](https://github.com/open-mmlab/mmclassification/pull/1229)
-- 支持了测试时增强（TTA），详见 [#1161](https://github.com/open-mmlab/mmclassification/pull/1161)
-
-2022/12/06 发布了 v1.0.0rc4 版本
-
-- 更新了主要 API 接口，用以方便地获取 MMClassification 中预定义的模型。详见 [#1236](https://github.com/open-mmlab/mmclassification/pull/1236)。
+- 支持混淆矩阵计算和画图。
+- 支持了 **多任务** 训练和测试。
+- 支持了测试时增强（TTA）。
+- 更新了主要 API 接口，用以方便地获取 MMClassification 中预定义的模型。
- 重构 BEiT 主干网络结构，并支持 v1 和 v2 模型的推理。
-2022/11/21 发布了 v1.0.0rc3 版本
+这个版本引入一个全新的，可扩展性强的训练和测试引擎，但目前仍在开发中。欢迎根据 [文档](https://mmclassification.readthedocs.io/zh_CN/1.x/) 进行试用。
-- 添加了 **Switch Recipe Hook**，现在我们可以在训练过程中修改数据增强、Mixup设置、loss设置等
-- 添加了 **TIMM 和 HuggingFace** 包装器，现在我们可以直接训练、使用 TIMM 和 HuggingFace 中的模型
-- 支持了检索任务
-- 复现了 **MobileOne** 训练精度
+同时，新版本中存在一些与旧版本不兼容的修改。请查看 [迁移文档](https://mmclassification.readthedocs.io/zh_CN/1.x/migration.html) 来详细了解这些变动。
-2022/8/31 发布了 v1.0.0rc0 版本
-
-这个版本引入一个全新的，可扩展性强的训练和测试引擎，但目前仍在开发中。欢迎根据[文档](https://mmclassification.readthedocs.io/zh_CN/1.x/)进行试用。
-
-同时，新版本中存在一些与旧版本不兼容的修改。请查看[迁移文档](https://mmclassification.readthedocs.io/zh_CN/1.x/migration.html)来详细了解这些变动。
-
-新版本的公测将持续到 2022 年末，在此期间，我们将基于 `1.x` 分支进行更新，不会合入到 `master` 分支。另外，至少
-到 2023 年末，我们会保持对 0.x 版本的维护。
-
-发布历史和更新细节请参考 [更新日志](https://mmclassification.readthedocs.io/zh_CN/1.x/notes/changelog.html)
+发布历史和更新细节请参考 [更新日志](https://mmclassification.readthedocs.io/zh_CN/1.x/notes/changelog.html)。
## 安装
@@ -115,8 +106,8 @@ MMClassification 是一款基于 PyTorch 的开源图像分类工具箱，是 [O
conda create -n open-mmlab python=3.8 pytorch==1.10.1 torchvision==0.11.2 cudatoolkit=11.3 -c pytorch -y
conda activate open-mmlab
pip3 install openmim
-git clone -b 1.x https://github.com/open-mmlab/mmclassification.git
-cd mmclassification
+git clone https://github.com/open-mmlab/mmpretrain.git
+cd mmpretrain
mim install -e .
```
@@ -126,79 +117,136 @@ mim install -e .
我们为新用户提供了一系列基础教程：
-- [使用现有模型推理](https://mmclassification.readthedocs.io/zh_CN/1.x/user_guides/inference.html)
-- [准备数据集](https://mmclassification.readthedocs.io/zh_CN/1.x/user_guides/dataset_prepare.html)
-- [训练与测试](https://mmclassification.readthedocs.io/zh_CN/1.x/user_guides/train_test.html)
- [学习配置文件](https://mmclassification.readthedocs.io/zh_CN/1.x/user_guides/config.html)
-- [如何微调模型](https://mmclassification.readthedocs.io/zh_CN/1.x/user_guides/finetune.html)
-- [分析工具](https://mmclassification.readthedocs.io/zh_CN/1.x/user_guides/analysis.html)
-- [可视化工具](https://mmclassification.readthedocs.io/zh_CN/1.x/user_guides/visualization.html)
-- [其他工具](https://mmclassification.readthedocs.io/zh_CN/1.x/user_guides/useful_tools.html)
+- [准备数据集](https://mmclassification.readthedocs.io/zh_CN/1.x/user_guides/dataset_prepare.html)
+- [使用现有模型推理](https://mmclassification.readthedocs.io/zh_CN/1.x/user_guides/inference.html)
+- [训练](https://mmclassification.readthedocs.io/zh_CN/pretrain/user_guides/train.html)
+- [测试](https://mmclassification.readthedocs.io/zh_CN/pretrain/user_guides/test.html)
+- [下游任务](https://mmclassification.readthedocs.io/zh_CN/pretrain/user_guides/downstream.html)
+
+关于更多的信息，请查阅我们的 [相关文档](https://mmclassification.readthedocs.io/zh_CN/pretrain/)。
## 模型库
-相关结果和模型可在 [model zoo](https://mmclassification.readthedocs.io/zh_CN/1.x/modelzoo_statistics.html) 中获得
+相关结果和模型可在 [模型库](https://mmclassification.readthedocs.io/zh_CN/1.x/modelzoo_statistics.html) 中获得。
-
-支持的主干网络
-
-- [x] [VGG](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/vgg)
-- [x] [ResNet](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/resnet)
-- [x] [ResNeXt](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/resnext)
-- [x] [SE-ResNet](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/seresnet)
-- [x] [SE-ResNeXt](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/seresnet)
-- [x] [RegNet](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/regnet)
-- [x] [ShuffleNetV1](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/shufflenet_v1)
-- [x] [ShuffleNetV2](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/shufflenet_v2)
-- [x] [MobileNetV2](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/mobilenet_v2)
-- [x] [MobileNetV3](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/mobilenet_v3)
-- [x] [Swin-Transformer](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/swin_transformer)
-- [x] [Swin-Transformer V2](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/swin_transformer_v2)
-- [x] [RepVGG](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/repvgg)
-- [x] [Vision-Transformer](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/vision_transformer)
-- [x] [Transformer-in-Transformer](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/tnt)
-- [x] [Res2Net](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/res2net)
-- [x] [MLP-Mixer](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/mlp_mixer)
-- [x] [DeiT](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/deit)
-- [x] [DeiT-3](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/deit3)
-- [x] [Conformer](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/conformer)
-- [x] [T2T-ViT](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/t2t_vit)
-- [x] [Twins](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/twins)
-- [x] [EfficientNet](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/efficientnet)
-- [x] [EdgeNeXt](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/edgenext)
-- [x] [ConvNeXt](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/convnext)
-- [x] [HRNet](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/hrnet)
-- [x] [VAN](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/van)
-- [x] [ConvMixer](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/convmixer)
-- [x] [CSPNet](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/cspnet)
-- [x] [PoolFormer](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/poolformer)
-- [x] [Inception V3](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/inception_v3)
-- [x] [MobileOne](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/mobileone)
-- [x] [EfficientFormer](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/efficientformer)
-- [x] [MViT](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/mvit)
-- [x] [HorNet](https://github.com/open-mmlab/mmclassification/tree/master/configs/hornet)
-- [x] [MobileViT](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/mobilevit)
-- [x] [DaViT](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/davit)
-- [x] [RepLKNet](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/replknet)
-- [x] [BEiT](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/beit) / [BEiT v2](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/beitv2)
-- [x] [EVA](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/eva)
-- [x] [MixMIM](https://github.com/open-mmlab/mmclassification/tree/1.x/configs/mixmim)
-
-
+
+ 概览
+
+
+
+
+ |
+ 支持的主干网络
+ |
+
+ 自监督学习
+ |
+
+ 其它
+ |
+
+
+ |
+
+ |
+
+
+ |
+
+ 图像检索任务：
+
+ 训练和测试 Tips:
+
+ |
+
+
## 参与贡献
-我们非常欢迎任何有助于提升 MMClassification 的贡献，请参考 [贡献指南](https://mmclassification.readthedocs.io/zh_CN/1.x/notes/contribution_guide.html) 来了解如何参与贡献。
+我们非常欢迎任何有助于提升 MMPreTrain 的贡献，请参考 [贡献指南](https://mmclassification.readthedocs.io/zh_CN/1.x/notes/contribution_guide.html) 来了解如何参与贡献。
## 致谢
-MMClassification 是一款由不同学校和公司共同贡献的开源项目。我们感谢所有为项目提供算法复现和新功能支持的贡献者，以及提供宝贵反馈的用户。
-
+MMPreTrain 是一款由不同学校和公司共同贡献的开源项目。我们感谢所有为项目提供算法复现和新功能支持的贡献者，以及提供宝贵反馈的用户。
我们希望该工具箱和基准测试可以为社区提供灵活的代码工具，供用户复现现有算法并开发自己的新模型，从而不断为开源社区提供贡献。
## 引用
-如果你在研究中使用了本项目的代码或者性能基准，请参考如下 bibtex 引用 MMClassification。
+如果你在研究中使用了本项目的代码或者性能基准，请参考如下 bibtex 引用 MMPreTrain。
```BibTeX
@misc{2020mmclassification,
@@ -219,7 +267,7 @@ MMClassification 是一款由不同学校和公司共同贡献的开源项目。
- [MMCV](https://github.com/open-mmlab/mmcv): OpenMMLab 计算机视觉基础库
- [MIM](https://github.com/open-mmlab/mim): MIM 是 OpenMMlab 项目、算法、模型的统一入口
- [MMEval](https://github.com/open-mmlab/mmeval): 统一开放的跨框架算法评测库
-- [MMClassification](https://github.com/open-mmlab/mmclassification): OpenMMLab 图像分类工具箱
+- [MMPreTrain](https://github.com/open-mmlab/mmpretrain): OpenMMLab 深度学习预训练工具箱
- [MMDetection](https://github.com/open-mmlab/mmdetection): OpenMMLab 目标检测工具箱
- [MMDetection3D](https://github.com/open-mmlab/mmdetection3d): OpenMMLab 新一代通用 3D 目标检测平台
- [MMRotate](https://github.com/open-mmlab/mmrotate): OpenMMLab 旋转框检测工具箱与测试基准
@@ -243,7 +291,7 @@ MMClassification 是一款由不同学校和公司共同贡献的开源项目。
扫描下方的二维码可关注 OpenMMLab 团队的 [知乎官方账号](https://www.zhihu.com/people/openmmlab)，加入 OpenMMLab 团队的 [官方交流 QQ 群](https://jq.qq.com/?_wv=1027&k=aCvMxdr3) 或联络 OpenMMLab 官方微信小助手
我们会在 OpenMMLab 社区为大家
diff --git a/docs/en/index.rst b/docs/en/index.rst
index 75092f08..b8a00932 100644
--- a/docs/en/index.rst
+++ b/docs/en/index.rst
@@ -1,7 +1,7 @@
Welcome to MMPretrain's documentation!
============================================
-MMPretrain is a newly upgraded open-source framework for visual pre-training.
+MMPretrain is a newly upgraded open-source framework for pre-training.
It has set out to provide multiple powerful pre-trained backbones and
support different pre-training strategies. MMPretrain originated from the
famous open-source projects
diff --git a/docs/zh_CN/index.rst b/docs/zh_CN/index.rst
index bdbb486d..7f23c013 100644
--- a/docs/zh_CN/index.rst
+++ b/docs/zh_CN/index.rst
@@ -5,7 +5,7 @@ MMPretrain 是一个全新升级的预训练开源算法框架，旨在提供各
并支持了不同的预训练策略。MMPretrain 源自著名的开源项目
`MMClassification
`_
和 `MMSelfSup `_，并开发了许多令人兴奋的新功能。
-目前，预训练阶段对于视觉识别至关重要，凭借丰富而强大的预训练模型，我们嫩能够改进各种下游视觉任务。
+目前，预训练阶段对于视觉识别至关重要，凭借丰富而强大的预训练模型，我们能够改进各种下游视觉任务。
我们的代码库旨在成为一个易于使用和用户友好的代码库库，并简化学术研究活动和工程任务。
我们在以下不同部分中详细介绍了 MMPretrain 的特性和设计。
diff --git a/resources/xiaozhushou_weixin_qrcode.jpeg b/resources/xiaozhushou_weixin_qrcode.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..873a0ba40a5af1baec49c11b16f86edd79714eab
GIT binary patch
literal 42538
zcmd443qVZy+XsFUa?7l15)E<>A*qxOYsoFS=RRvgk|aq*&B?7+XpqYijeD%ipc|Er
zE=s6D=%%KcE~u%QF4Igi=gj~4p4e4+fA8A&{lEWzyJ6Qfb1vWS^L(Dq^Lai;lcOoZ
zewsUL&MZtz3&Uo^KbWQ%n?KXrc@u`&*D0MPS9n3xPgpxGZS8jLwL5fZ4;0e<^_FA-FXNY_6iF=P<
z>SXlu{lq?tUrUFY{9XYx
zD^{*@+^}(z(`M%_9y>jEd3pQn-hbd=K;Z9(f=+~-JaziaA7{^9zH;^2_3#@v|9lV;
z85JG#@KJ11a>~=xv}fth-(<>rBtJXm>m)ram{lacd}
z*{b4po>&q)rw@0jMW#J=4y)f{T*OlBHmNe!U^7iM*vMF+1{=IcgRy2(2Z@i@i88&F
z8mx3QrlNxs+O*+eRTmBRdcFpGJ}yFoJzGj7zb4jmnnv-JGv5=+F8($e>~aF8!2&PS
zNo4&Jy!P;I4OZKSEv@I|IM6{(^de6_{qsc@`BC7H7k?!038>ye_2o)jeb81tRZrdr
ze*UZmJGs_LgB`n}UDIS-LGI;8v7V55_9_>;x8M+2Y^1>scvXqqG}zDak+t01awRTn
zY;P6gW724PG+t~pwecxC$VNvop=hmri9^a`dzC@*Im1!1Hrl7V9Uq{o6gaax{w;N)x*s0n9YI92~!knRX*J3gudG2y4pn^q&$r%v=lm={>Yk&K|CG*}mh
z6p{3}21~M%yTxd*OY2to;9Ki_`3lv(f%)-gRA*BQ-H(TRb!p_~AZ#SM
zepi_5*U{SJrmvael$c*d^xRC_=1I`?gTwvlS_%;#6LrEg*77IyUH9{b(|c>l$Ns=d
z@YQOwPg8*4IuB`F32!fLs?eLSI3>+mLgz}$?`=|z6&*Jt&&qwmI^$m3k3XNB
zZea3qzs0kn{p*fS>(HkH@68>lU>BwB^~4WxdN$xyZrwMRv!m83d#&*yHlH2DYuxo|
zX!y<(qt!Ru&xdK5)ta|mz_5Qt3W5_x(uJG{{6yY>&m3LuFz*T#JLMdHTpq`%%9tyD
zq55T%JG*AAWW$l_;9qd#^G-U)W@3JaXa2LZ<6x~q0Iy9nS*Rnb|EG)cpKrt73$#vVK{Uj>Hh(Zyw%1_$t`O-BY6x==&FQnZAr6TYgL&%rLe*&v);Nqq5Q&S3(iD2Q
z2K!*3!R#k5ld{x5rO_@<)C)p&LO_y5H^|!Gh+agst_BMa9gWe`KEkS`NxbL`2mYcP
zSq;H9JTy2`
zgGG4-bNl;FmZizf1L}#YEN2($?DY?@!?VV28%mWuy?x@so7olR9!H9{V}fHE402|S
zJ=I&JzWh<74zz2%RBzQa`Qy`Czl=$(pCGk{qSOla;#g%}EnHL$_MO#!t4Sp&?L~Jv
ziWZ&B6z!W$uy07BC;+cpjaRRRMSseVf>kq9gN5``zw(<5&$=
zJ(pZZLeAq$-w^H8^P~=tynBcgvlE`ue7Rec5_gObsG)4nc9#?`5GG8Bvlh44Rs>!q
z39>j9Z&;1j=4-H;Q(Wwo7KaqIcw5Y+zN_4TdW07rA&Munr3ZivL8KFR4ab?F3m%nq9K)EEQ3l@hfC4dm9%>qjf`NbZvq7
zr0WSYw^OO^`xA#G%!16nmJ0v|B0z1_ZTQ8NiH5LE{)BCE2EO-fhSyXO?|Nyl7aI}q#)&`}r_4tr`-CvgoW{DN@s1;`bpV7I
zS|}!n?=T=_3B>f>_yzfDa)<#+ZSg(1)x|@9uMVEYIosXz__L1jg!oXp^)&Gv#5Gx7tK%lE}-8OT%@!2A}vjg#%t|
zun2CtOGSoz8Fj@j>S0LZ8P?D{v2z9{I9X)sU|kf|2?wYf2>^AH=lJfFnWyrrI0{>$
zc!)}`Nhgn+TS^6I8rV0`Yv8P`=%Umrk0X386`;Y^lZvf13Ok;9I4KPTM8c+EiH#RY
zSdkiRvtKwJ<{JN)ze$6+?5fTw)q;qM!|v|}YYsx{b(cKY{xX$(9qf#y?A#b}ctFA=
zUJzS)oJg7~7uB<>>>v1uwnm)GqKN$K`7xOVjosaYO^>UzQ2-!p$$LcAK-DPU&D2R>
zU71OV?OqKQZ9V5U;oLP&AelN>^ipLL?&-#^mQ-K)-8Fa;dvrJPh!b{%#WD(F#yAqP
zfw#ofoN0e7jIrAsv}$iqYnfR8oW}IiGue=Et_Mh*WrHqZ;KGNb|<4;FsaB{JO`Y
z8U@OQGMAN5q~s|ZncIGgpEnQP_O7gh8cZ4tM>qp+yatnyDDFg@gSeBkz923=zcJM`
z%zZ!H`3}S&zI3UZ(uO?5XVRkCDT3qKCVwr}YOMJ-b!7?ilS=}F>2BQThL!;BL`&#Y
znPKow(Sx-<<53KMd&jjDih()3wguaM*@xtW|1OR$)G1=1T#I=9vCYoqLdLpCKwXs|
zP6xTsi*~xvGX&Xu#V}jR3|Y1uV$#7d?j%Jh-O~hi-F=%xI?&0fl!}lf5bE92yHkbp
z31ZiT)QKMI#OxLs8A5$)_(D&sY*NF9x~##{52+#P04xgIdP}K&kyTS0H1c9Q_-1Lc
z{s$?sO~MffKT`m&TLLkZ-MWnYv9#X`x{>b`5Cz-A-a3~H06;{S-fDz5rKgj{*&N8M
zy-d`1HjC6}IoTP#wbVn^ixeC&Z@=M`w{(B?yf+%`zS{PR65kS8SNyP+e%-|$kaOYe
z;E$!tm21Qn*87Y}g+6SKooI7J!~U47aH4zS3R`Z5T!Yo;TiK70gt@`JVS6OUd7T8r
zWs#pb(X^)Z>hn{zJjNT^MfR`*Bw4TnUZ5Rd4m)5Gvjbvb2M}d0kg`gj(+Qga;2O{a
zJykoQOrHp4`g=&@>8EP-p0W?pOuDcy7Nk5jVbVniN`XwexQbF>DWr>;Zl^L$_dgp#
z#%*Z}tS~9Qm;a!V$y07niw6+j)Z%tbo?>coYl!If&9%4<-~`yF-yWo@rUitoX~2j<
zA_!Nojt`xomq)tM^RKi$1IqXHGR32fb`*}cLv|V<%K-8wNfg8|@CZVU@xY&*q2o=M!p%O_p
zVGwk=oIEzAEwaUws5%kAYVl-3=?d1qLVXDNhXE)_NC5xWSLQ15tt!3a%yJ>ULPmTj
zIO1{S2t5NO?QQ&~;T+|v*Q9bHDL6M$TiFpp#9A&S_xWUP6eRbJ1ZNV&h)g|6u}X*g
zj_i+bsTrW2r*JJIBz)aUcJAdjl`4C?m^vHb^;P-ls<^{9$_zFfO=~VfIfEf@f9F$@
zsQnu3*P6VTB2u!p$n6l()0YJ-z+<**_Syo%`BMxqER$Z`^44HpZl0(<9i&_+wzO{N
z(liwcp#S&aW&)s=R`KbEwo)5$Z~{?Hi|WnaOa-g*+YkT+UQMgwvH6a*DLJ%NU#UH9
z&D(z&@9J5IkU<+1BI)ySutA^g)PnWRdG-(DJ_dKyp;nh!d9bUlL-Nx@dw&XI9=@pz
zrm?d#+GaJYHlOT?LIXmQLh&7jEnA{;Oxc|Vp#hL2BAKotPR9tKjv41vNB5BQ;r{qG
z{Krax|Ki>>*L-l_fGQ==o8%|UWz;QhSNdI{rNUX7pT^E9GDPue9shWoyh`xYNaIOB?y6=Hip&qEMHsOnY4{Seu4-y^03LAo(s=@?ngckEutl>nBqr)9l?+)4
z0O-^m_$9=pmL}-
z?i%bjzx~w2T4B1%Dc=`fRv$J{75@iKS2-TpP^vAXgJ>S%#>%mhFIJeSMk*?RNhKgL
zJ1RnK>5RT-IC;BILA`Wz!TUn{f)8uw^e7k*Y_q{4({@mQdixDbPzfm16A?So7l;+{
zd3=qDel1Ei?sWPsJ4!FjJ||7gKSXX4aZ5;^39NWuM;pV^z}R}lep1!c5w
zUrD)Lqj}GCRr=~}l3NG5U^?}U95|!abcqJiL;D-l4ek{B4IzyXrlOPrm2jsT`G_3+
zRaKRmAMFTK5Ek2V
z28gISX5SL``y)Wmnu6h^*di)g9A1@6LEdAi=wP_0%}n&l0;R~YnTd9d^UYTMky-7r
zmj@M0%To470%ITbe#BZr3LQl%XF>|?j#x{Cu4clu8yO3~41T3kBUJa%qvzX~xKW8@
zeceXRvw?nxWw!EKh*G8sgCzuw@Y%Y%EDD0fBU%0Hj2&20Ja%
zV7`E5gD^GV=0IS%-oc%I0W4PnbZL$dpQDK~xVWOxl;w{T<9(6<4Yca?z&`j7p8^23
zZQG$0RqSu`pWlO`{~dX~owz4Lo1hz9sm?cT%wQ`d)4f<#B4vOEbD~03xCF1S^C)*y
z*{iH}H43}7jrs|`-C&D%fGwWSkzVtKMBSOKHX=8vh^9Ve;URLNc{=Ryhl8PvBp#i2
ztP#(U$%1K~&mt2iIoIeeE|
zZkm&Bc#{T;nGxp|5?9^LX@Gdd6G->n*bOU);_bk~krc$IPH>^SS!*y{HG@7|RSQQ&
zq^sP1u;sw{gOn~Ka`%V@k`b^0pZNAgH9ZFv&#n|nP0U1bJSU!~P@yn>)~3A5@%)Aj
z+C_L?Al>CI?rgG`J6th_tM`@Q$RS}bMbn7msWl)Q@b!VaYDPGT`epj0vcwb@IF**z
zlGmoVS*X)chiK(K20$y9IH4Vwu@PwHwwMMpsWMmXAd+FlCvC-xp+EWcZXY7P-fYR=
zH=E5RlHt|qxVS6x;)6brwe)C5GLHj$*pYsw!4Ac!eW-K7j5=;w86mI>ZB*y)Vn6B6
z)T3*loN7>;LJ2IS&k8o9&VW0QS}s&WwWlhxRpVCl4L-ewPe}x#&uyy8X>30@bwF?E
z^Cv-l2y64p0sP9zCI31Ado|bZI&%Q!8tjH$sN6T->lARe{P=9
z7j&W5=r{L$33E-y<%oAF&pY)SaBQTuA`PLCV{UD>phHdecUp>4R;$*4pk-%(`x{+|
z_qOgT
z?S)z$PIg6`jEoV)DFwx>ztJqoFj-a#+28&t3lYtO5z#Ckk*DV0&rNowNy#f%e$}e|
z`e_#FOON&o{
zUB9(N%BFq23
zV^bYIb^snv>orGI)>{k8J4m^)OxtRs26N|A)`i*d0;Z>@^}Te1em46ft5tWC3(?Ml
zEI$M}7+}D|yf!`Bi>%6O#EE{7)&x+{{5Suwt@~eL>t>+W8cWaYjX6kl;A{zOC)TRf
z3cPO1?lXm5WI?6U61X6;>T6@%n4x{w62STbnFuK_%ngb|Z2{c6mObJG1nVn^j9$=K
z4S_2gCDm(r&47<6=nY2(q&41(YUdITsPJxkUobca=C8g}le6J|E}GT>zS)}WFNhj|
ztuziGlD@j@O;F67xlU|Vc&r#t{h1{^2YnmsbZgLicq5W)^%rD5OAv-ocrYQ&;6s~I
zc1xwl@a5d6^j
zGh0$;nY!Hqb4c^7A0e436Vm;mvvDW^ki&0)cB}xE>gABHY_jCS5mmdk=MjS1VnC##
z#*XHI&s2>7j1bmvB9gWe2x=U_>7SrJWNejcTockHji|lRuZ>g{$Lt+|4zbrCj6G&$_2^lO@E4*4-z3NL>)-zE4a2v{gL|l<@b;z&i3X1g$a%T7`L75$?vyPm|vP0>;6+qBuO+Z2;lu
z?FRhI3l@zOJ$@ed*aKQjZs@twU&2d%71#;lP*PC(1iep_D+5
zSMd}lfqtvV+ln{`u3OWAT+%5c#-~NP!M3yS5fZU#xq35oD_&QcRbn5ENBf#X-_-!=
zlqRFT>se1msN@K(`boI)5wXs`nS^Z=>gRNWYOi~~20H?<3TThMY_+jKxt9QVwZSh!
z#i57cGNn9JX#8B|6iQ!etjeveF8p?jPGh%gbphKF0@)(s+pNp`7WYrn`Ek<4slKAv
ztI9w%O5IPE87ea_7VIVOtLI#SP`DwM&U%X!n?K8Ixz#0~XJ>vab#JEavdeMxrTz3R
zmUB?U{`KtyUl&K4vry(UOehtH;#I!jWQ-NU~~frdZ)4B>d)PN++-CT1S=Ye(;14
zej6@=tPNl_-NvO4iE0iF$>#u}!8#feeCiT3kL5l@8d-{|8e7S>1f|`PGz3H+rF_{z
zJ`#Pv(HuOi!6t|#CKBR4Tr(mipSGqmoXVA?V=b9|Jwrvi#&vE*klfI2Z2H$+a~!0~
zzr#A`7zyMeu5dO94jZn>qTpb@?$c>1dlYpT)cFGKQ{qFvf?)R(S~b8Yw!8}hr14;O
z;cM812F2ki`~i@*+*Z(U%y%Kx2#UHBnW&qW2BOlV2NCjw+J$wZ`07Ig_^Fh0?>a=%dOPJWytpTvvN%kv4@If8jP6p;r#L0Dk|ZtU}}%>I@C{DP=TG1OL)A-Bq=1NfhBxdJq6NiO_`|ivnP0J+%?p8~}6F
zkUf%701VwGvT_nx=7*W?t$T-YMRX#NB|OP$*%kR+sd=*fAsY*jS{B}?u5JU#=cIh-
zW;AVl$7nu-7e&v0rM_!=#ISQ8$}n5jLp4tA0g?+>#iwHaQ9Ora7B-TLL{hv=)yDJh
z`IYKNAk&DUk}o_ITe1bXR9(|_nNaQ^8l#cwxor;opV7~VeQr&EOf%F0MbRb2C)FGP
z07TMfnYnr>Aq}R&q0_N^R~RfhffYS<5UjXSHne-pkqs51Hu4z2(U*vgA_RHJf!3z7
z@UM_3A3>h`$}ifOmXJr~g%l($P!AG*ew$7x`XZ~m*Z-&mVy)A5Ei}n~xVBRVfI19B
zM2>95&-O?}^an_2D4-X{i>e@CB2tJB+vW?!xHHg!+X|IM+WQzv^c|_M)w89de
zZlxik6=oI=BN8ioZ`m=|Xh!YB%<rddin5-yusw8cVuPng&;H>1%=AcbCo`2B(}FmYFon+GOC!=!
z`AAvBph47w$Qn~)U2~sedJuEUb7L}jHIhv4%^|yZxUTtWKg@po
zusEIU0V>*WScmVB4h!1>d~U&egFp`M+Q&vY3ly9iP4+=_ie5ocN(hgv7`T
z6r4BLAoWl6T_e;AI6m<8v%oS?aMpGbWRX8Q&48y&<+j!YQ&WDN=rLA4XJS-KfxtqY
z*3$xlyIeirMN&MKX!i*%$Sp*^&4BmaL>uF3ha;_kY2t|L>YkRfa*Z}Bk2&cN_{l{v
z=X;eoP}4k)_&>!EUryA-|v0KDa>sis8`P56k>L_ro2=;zkiAHLHd{M`DA;=f`d2vrfJrjm*
zf$k`hFGeP0dzRWU^2OK~Mo$m&#U5WqW9VR_<$Ma~S)bloO*(D#EGRCv77i&GqTvSh
zueRvrmMW|IcHfbEgP^>!3376)IU8co(uF!X{21O-Ry(y5AZZW4l?mZU@am{CuR61i
z8BwXvU+u%p9K>xnnq5Bq{D$q*?`ju0RXvtv$wO#>(Fcvuy#ll_P-PDPnYC47&5!m;
zuvhyDL9n}odMS3ZcqSBrdKv=JgukU!KO$3VJ^sVVG?$2FGobiv?s(2sxZMOs7z%hg
zLp3a$nU(OH%(R1tqVm)8%jy%yBK_yq|DBTp9^c>q?HC{D?YpNJT+!i)MUVqwOF
z9=|;SP-TDFj3cGe&Vnp{3mr+DDS)L~>b=>GN)G5qiIZBg?5FJf7xE>b(Y-JaL|1|V
z&F$@4o9wR<;^4UE0CC~(0FlB|j3y;BzI4>rLo;SIg4w8}jyf^jnT|TsiP`seC&uYX
zYYn`bZWiTzBhULXC~-MRQ6rZsE{Vzg><59t++n`vZ$Z`;Z+qneS|Iby0j8N5GYjOz
zFZoxFXP3jt@6duwHBtc5vewEKPs;a|PHkuCpv{8KTB$h8sKHHFI#e{E@~b!Tt0>oA
zj0by~z!XcoE_J`e&heRQZMHn@8shkSqkY%7D0OmLe^#@+iemEPjW_FZYadEeUDdaaR(zdgKarw(vXmM-=ZkJow5rVwg=&h`vw>B~DYj
zms~JNT-Q%=P@dt$c^s2f^(&qJ+Nf^Z-Le(yc{@)B9e$M4En@VM+G&DMyYo|aXD8MK
za1FEQp)Mhciqw#l)2T}X-skx_EZR76_Mo7CMFAEe<@u)!hAet~p;owX|F#!9w%AR#
zoRe!*@a(C{x?F+q!Vk_PO<^wVQP*iGB7(Sq#90=Bv{i*ikNdx5Jv@GtQ1g;#Y}E%EI`
z{Tbg&W>tb8xco+Kb%k*D`?J{7K>xYll?9Jrw{2fg?J+dSVt~`s4#doq?D7mj8Yh3U
zvVHWrzS1h$@nbJ~AK<#&Fk1~-AVcMlfQ8x1*jp0!(Df$Nj57fqo84GaVo`ddkA79@YH!L|l_om5;IJS4iq$sJIgkQox
z`blLel*P!G%Wl4@wl;SiF4%F%E&N0N=P5s@B@{z;x_Inr^vCPVI;9^Q`tu1(o>Li=
zLyUF?DIT?bCn-Y2JVx=DH>(;cXBfrf+BLSS7kgG9GMM#TY>`-WeBHRTS555K5~y~W
z&B&0;h+zJ|k$+8XY%vM;gd?5D2SIuO!6?qmB;M>a=(dZ)VO;YNEWukSDe003Rh2Tea6(9!cCpt})&SprviD~pTwcFP@k*w$GtuCoE@-QMUxx+{
zPyFI#Yh>~=Q2$veOh6Pw!Pz5>1G?50>O!>|bge{-CY_fC`**Vv+u1QA)Gki+PXfhn
zB1H@#y^lwdF+{nETv)}6@Z-@xi{M-Y(f0JMO%e@eBe~_z5`8vi1gA_Yh6oJcSVvUzd>9JcSXuNi+o~V=ToM7QlJ~eEM~2RYWdZHVQN;!pK}@f*YvL8ahXCg`wN(U|Q;R4YN<5
zb2@0Sjcl}0kAubyW^^@J^G;oiG;TKwn4P+Eh}D(jtmhT_iKRWJ4bV|rq|vvW=+y*O
zd!>aU%Aqp*hbu?^rSA!=TF_lCKpTFi23e@%&wa}QU{`!KVTyOC*d7MFjetDzJ4|VV
z((W$}&p6nXkT|wG?y^j^C%MTWP*P9!QV&l7~m0ZY4*Cl&TNnx4T0OfM?xa0(T2K&TqiR|YGfDVb|J2M
zfEaL-ykq?T0yBX5eo{>$5=XicPGm4>6Ow&ogRPga0}9KhhtIjxb^Vx-)$`m8d{X6p
zM|(YcaI;qU@=8Bd@<)@StFDy^C5MW1`>yP|@>7%5hm-@-2L)4~&rU8l81E<*kblrJ
zoL$BVBl=lV(eVQwllKOJpgQ*8o)1>NWqQO5lXSl2(}^zK?z~^5e%&zs)2Bs3>s|Nk
zJa&}y`}Jt4m*c^u>#m+FZ#X(V<7vTx12fzIJIqEQBNyf=>EH#243GQGfGM@ieKHeb9wN}
zA*RXl!#;Q3d!IZwaa!%=+k1Cd&TqK;tZ?O@W3GEA#ksb>muPdUwt^S0E#~Pdw)FFz
zVHVjqWhr~({eZ%W{4G9)lTNSnN{>8v!2H?KOnajvYv1Bd1v3`@vDWWFDr=&1!R=rD
z@9gV*IcSKC7eDmn#vBQ4^lY0;)u>;`_Zd5Ec_Ae35Y-zTzZ1IYWa$3nI|ouYfPV9%bl*g_ehv^sXU>F;=!7ex?ba~+OpgUgv0@#WqJmP(5iGBCgwqCbC*|7I74H}Wz-`M-@
z0xstd$*hds>_FeeP5AyvH9fqaHyruoIN(gFptJS95qUjy4Cd|S6$`q)+Q~6g6554R
zySN-RQ1x7EwQl*^uwT6TzFt45_-tSohs=QI4Aa<*}&;{
z=2@KWo1b~O>Oj?0>p#*KxbOxpI4Ll=v)W4;<@RVz!;a@Gf_9daToCuZleMA7?MY+$
zxUt1r*H1Zih(7wV;{wyYmG2*3*t7PH8Lx{=<@)Upq9XN<#a6Vp9{F6odsI!M%VC*Q
zvRfbTq}kg%57j>^4}M=QeBt7}u5eOpzN)^P>cF$5weE?MHR}}<{EF|4t6Y4d{?$4ID?>W(O)|^;yr-c12d)dxHFXkuY;c;K{YNFy
z-po~J>9Y8HQSSs}-gjPV&2c){@4EMM^XJRLoIX6ZapatLr<28W=cN7e_lD3_J_z8g
zOC9|yo=fJkx6;|5g+o%b%Qeqy6LQO;ix*du8*wy0?PGRT4-Gc-76&S@eFQZ$O7J_r
zjm+oES1BH;MtN5jbCPQepqfxzSu$lyK7i#-pU1mRj=gZZmX`0?Tw>sOR5;G`NVgLm
ze5-ffIdbF7uHEO&&Bt^Yt}=oHXfb@+Lb|Hf`kt0@Sy@=RI+=dU(Z-8Ixe$hnV&vJ7
z34s_tOyN}^9)|d)D{A6R1pTmt?_06WJ5whC#92QQgrEE`sG3@6YFhLUIpkXy`(u5p
z9(=~hJ1pj_CxJS&m#;JZ6ymiPXcO)e7yZH|VfHyy#gQ6}QwC#XA!K#8ZJ=pVR)g<_
z;1ahZ_l0O6?u<)F#o`1=_6K+{J`mwYQj;^YK*kwb3it7Yw7Q@S|ByMk?#PiQou}*^
zbLaz4C~`$3AOdZ|QQ>9_JcTC}&dj1SZWE1*XM4>V!un$DVKW7d)qEp;n2MMAITV
z&5H{#P3Mj^C=O)laGY2_wDriiFHK6sM;d1J``^L#FTlb%IR<&I}5GRTfG*AME@M@d1F
zKrzfs9o7gnZDqiHm}+|wM7>v#AXJSg0?90vsKKt7b!pte7L!u)nG(!O+7oH)48tf{
zyM{BfW!-ER-wRITma#1dIE6LTbqsYH92x!S(3`RQeYalRPZhaT61|^z_Kkm)ne&Fu
z$a>lJ^nd}SjjE_(_m8Jzm!03WWNyglk!voAHP{_ak5#&HBFBdKfwTWy?y{3NL&|;3
z3W|)%@*K6y?c(6&zgSu}2$p5JEiQersVddw=RSu$x8*~*yR7#4y5w-qQ;*G)pWARw
z`qel_KW5EW9CdZvnc>lO-WdOOIhJYJ)#vIrzc1UecDFKY+r*%A+dpMJYdkXEHGJQm
zCDCTvxdYyfbGzH^v=`TvEZkg%n`f-;bN%McIp!gr;gK`iSL+y+-TiEF^+fFb@^?=n
z>%GkKsdK^idZoZ*9$y_>^z)s%-L-oh
z0=z5I`rmnf@aCsN{h1FU_V_eDN*rSy@#eySd7hUae9XW85^MxRoO*)>R$Hr;toY7
zO*MDHMC5u?oCnz$8y5b2d$#JJ$uHbpNRZd3H0(9V>sm}71A~vo!Em_lZHDju(Zc_C
z1}C@BTp!kUiDGLuA(M1b_F$QXghP00Qtnm62l
zj1g=Usa17Heqxmyx20ZH4NNnl#SA={bhR1~X;H^))~?f&%M_d<&VDvyJ)$bU&YZht|KzK!y@wmXOgKE#5^)RJ_q>{$i;T>e)@g1oo*O%kjBG&uPW>-1LvOLDxj4Z0
znfv+RX=dtTq=_^ifwW5FBM=4AtW~n2u6@xeg%WA}+}n&)$)Tw}mn1S`*0+=@`5`K}
zwfBH}KXn7&ViL^1&r8F5aWgm(9H3RZE8HN+-6u9W1FibjCWE@;4jcu*+WDMv4GFc??_9
zVrmaA#7vaBT8Imf3sOr%8UX=B^HM-M?MO8D2^4B9qCh(TBxUCCvP{POHHSB+`0qJ9
zKNa22jWz^~pdSve7O0JdiWOvCJ352TRbZotPBNM67AnWhsh_usBFgjQDbsxqd#blE2oDWY?-q5q`
zcK#=Qu5G6Km^C+&jFn>Ll}{PV-AtB71Z?08FJ7x3w2$jHgVS%-lS-elPfwTIhRq!9
z{bWf@lfz^0kQBXhr*AAy-rnQnnAOMGHw^4oRgA5#beUk3#P#W3vZ6cZkp|n_U)f1j
z>$;VDVeX&T>}>BG7tVA6H-+fKtbUbQc*V7)agQ$yow+%9s{Nf^XL`*rJo&=d|H;PH
zcU?}zAH|Ce;@)jr(=ph5Gxbp1wEV-7q!70y9cxy++j+}As&?EQE1Ru;)%#+Pc719%
zW#7BIp03lwbnr3LJhHCKmA?xN&pbW;xhdh)l>*%u1ILLsN_94l(4pS;$#On=)BBCS
z*=py!{JB{L#YY_iLY?!^%LgpVlz+A|x)qojd;0FCh{s-bw|CvN7SkP;J(WLp+`iH0
z@X8L)k_QBBI_2oRD`^8&>hJmiU$VB_=^EFLRX#TYhplkJ54u*JU39|Uv5Vch&x)I;
z3|-H6|M?hR^z%cqS7p*vL*=-!NtXhabxSVKOfHO7nyp^ZHN*I@#oh{|8g8HCB`c~x
zx5X0KoO|_hapB_6x6S>Mc_B5+BY!PeQF$l3JowM7Bb%3I&BN2f0B^NdS}JA;h#W3J
zpgCh*q<}Q5&8S+
zUTnn*wsi08H^i@?m;tuVPoUZWzd1@o&at&>L9`Dkt5(UkD!sNDMtjnYiz(qtg%g#;
zyNnZ+ZF$=^j3_XEEk+5UqRRzt75pH9mi*ha>La|Y1p$oSsKxx_jaF~8RI7;d0QxyI
zptGH+s3_*=-jbgL)zolYSG`^?lB_kZTE3geBO+$QWGVz+r;OGnJE^Yz`vkVGgHYPLP@bC39IpJo{V07>>DA
zaSn=N&PY~+S!bS8GAAIxl1`Hr@+u$;f0^zz6*7&;V}CayeLXq%*bf@jJuhL7m45eBGOd@OF^ubP=K7ZK(apJyvLDaEXG$!yg`J7f630o(-cVH
zULenfcru;`i^PQrm+)<JtCCR4}U{PF|&)T)mY(4}MnriH-PvLjW|xe9ffhjK&48
z`Z}Ne`?}`wZ2iB-vmxKDCv)8Be!y)m_61d8=Royvb}m_`OABHA9WM5#@`{%tIbXg@
zQWG|{5o)#y5b)+ALb^YBA7BuwbkPu`vueZz`HN?+c(&35gyL?hKr8pu9mpG#glU*>
zr@mkXFVy1jAPJz+HL;*77Y#A0`@)?@Y%IZa&NJ&WRqW=@xofEJC!6vQ73Yf_<)E%0
zx7ou18(7glUABe;?nI~uqxubw_?60D9+*phD$jqB3qxcv+a$r$a4q0#TX-#e$6ifD
zY=IGpX;dgokA;foC@#TuKm(+}j8CIeRO~M&<`;`72_S|m`hlpBSI_jRdt-}c_SVvT
znG!gALb&2-`H(49Ms
zd>NC=KDBIADx=l`Q!2AN#G}NC6qfk}fcp6njDwSj5p7d}9GvVQ#r1>vR>~h_IOk1r
zaC!GHE)5+OXJlw1wb`a2@?Zfp^_y)P7&T4fVz6oW$`_DV@<29K)SbK{f_IQnClk8T
z&uuy?oMlGf^FR-U^cZ=x0DK;?wB+^qr;W`)tS^4ldJ`lz<1P6$-SAES4|+Qc~%RB^@1#Y
z^1W>X{w%wUl*|r-@lip*GP*`>^d165(aeTDYjCv_tx8Q_P-kG#Zj<31(sx7&XetttsqYKxFtNkBot8$whT<&<^Rn7DC
zIeYzxdt~gi+S&r5W4yr0)w4ie`jxaqcU#DBR{eKnSN
zGTC~p%C>S>|FXM>N^a#u=}xJgu9r7hIe%N)0AA=L!JpQg68*thLtZc37VLgx@7wL7
zHzeEI?DFTP;M=Dc8JXvm&&0Y1zqD-1t_;3)?eL8;i-vZWziPipC!wSLCx@V`m#L)9
z@7+AoCpp`ZK_g@XB$25bK6{Y6I_eakYhouiz3orz&$f?R`*2AuZ+T@?_R-|x{1=A<
zVsaw~9-Je7?9T2N?{(9Aai@58@yhlNwW;fzc`w%=wh$Kx3vO>Mt)H?pDpyVfm<^nq
zZ8!Z<`%IRsd%~%k1|Kq2iV3y0M(xO+^|(Q0mLhUhhG5;;?9h8w3sQRJ50F&S8Qgtp
zBSn-TR}^y~e&fin*(a^djo%&j-5ZonEZ%bN>&z8Ud`J2-22!1H-=?O&#a
zO)G!srt56%;rZw1j3IyKt;igk#_z>fe!w=z&285b8t7$S^jYUoW$gfAl4+eC%VseA|_dbuKsageEmkQNe@3
zr?^&Z@g#ka1JEk>^b4e+>)h6I8Tz=i>k6=?$k1SvR|t3>n&qBHz6A|7y&yxCa%^Ag
zn^6$t+ek5u)2Bv!M+WyVe1rKVhOPSB8219ym=IiHl69d>2r0VBNiO`oVgGeV(U}2g
zAmv(fp=@zxt9zVpCgZ@YF8iK26PN(}YsqLu7i53hT8zl13)wNes%s#2U&I_7Z9(VA
zhTo^oxqkJBZo0q>k~LrWF9y*_Tnl;zvL*W{DE;1NFlnvNw}#xelC%v%LxFb-knKpCsIzo$vT$q5n&a#8)*+S3VuEcB9fxX(HLanwqlr#KES
zS74XT&ilC?Bfr7ji(|#RJ8+F0ka_#|+{wP<(%pp#37WM!w!BX2i}Jpl0Y7
zFxd_L$M#TWXkwM!20T|bR-$at1-hh=?w`Mp05wx2*4o#I>+b-dPqDU$*l*YYd6-UW
zriad=seI)0dD)4X%15rW8DHk_%1xT7cSeeE5AP6f_eF}(oRyoU2uxN+QUv=)F_sk;
z{V>22Fo?r|fm_^3Fbs*nz=$+}G#f023HJ3oknL
zfzB_~7OJK1@%mhM&V6>&^()=asFo^r%ePAGS$Ug+Ryz2m8}sQL-_
zRYlMIsoif*_rJV*CM8q{I|N^f+;rzN`BEMCCZWG~l=bp~JLi4c_S?ycA>H=ct5gZA
zC=873Dmg4;7w(u7rnd_JG@5m%ywh*I%e@*s&Y!-OwaS^?_EW)@!$TLFKlfEjbn`9C
zV-0k*e~Ek;ALF-Nv@tf2*LWzB|G0AXZ&R;ioq`WNa;W7~bs%apSUQ|I*1@q;Q}C>!NUSsI__LOod)g1hwsk
zFuZWO%YZ9ek6$WgoQ|V1K94H-bFS4|{)ptrjrQ7edoFi25`S)n!N7dSXXSJ;RwhyL
zxEs6K@(+9^_`31#&6aX)&5ra)~@
z%&kpSrx`MH;LFmg^3h;+hrlG5pqvfm^FWvcC*&eD%ZK{G;Z>$k;BQq`SP1o>P92jVQSB%&h9hnqV@*
zNj*j;3s1%wC*Wr1&lgsIJAeB8?fmJ_x(GV@7RLQBW}&k9Um3Pxbb$Yntb%b0&OqNq
zjDH9)E^ib1tLGvIahNJHlAu{o&NnX81
zp$9HH$PV`i*AiTGPJxRKfR9dI$;~c0%T%smjEj!t?Xl$RdS&&v4I&Ts-bCD?=;{wS
z(^5dW_KnaFT7%zt1kJKW3ghBn04ZJOp_;7T3H>xjDap7vAhUTYMs9BUMlET(kTThV?lPhxFGngRAysm7$CF%_?iu_zI5EJ-def?F9}TpJ7c+3(z!k_3EiDA&#J-tM_F<7*ClTnQ`6A-
zHl5z_vQhD{(>mvfam%&l2Xgx_T-$EL5Z@v&cI4l?)s20KubhRR<8v>{L@~;#YOwEB
z^OTDom6t2o>VvbMysLNey2(xCcZf0$kGuEQ^TOo9C65M0#~vz~R?tqr_>`={W2vpW
zS7Y&zyI}S2I|FF_zI^pz2X1%YH40tT{0puLofU_hZ%u}riuO7U=A70d;A<`(-c>W&
zzS&6aQ^H6hsjJlONgh9if5@T27Ff|z>d%kKlV}(qo(H;muP*56J-^3~syr9;_SW8ZG|;N^v88tf-0Ixdf>
z9qX$CC2X#$8}S-m`+NcLpwE`6(}}GvcpX@nz|b9Ay#+>!@AK>A9B=a7R+9ELd58oc
z#UipZ{JjypeP;=->XyZQYC4r}KZPz?LGI}g|GP)8!C$PSKk*gYggK6M5OmJXKm~n8`D;O%&dGSk-xlLxjJr1%AD<cm_}_nk5ptg%!F=3w)9*j`+jzE
z!M2Z+-h3dol;w|EP_SGwCFaekHR3*IFGk-jydG)cr9F4RonC$Xnps3*m%7NA2~yNtaTk3
zy1mrzE;2zzGM$Ojk*WVxpxb|8H$NB3$LSt0?xHegB4FY7rkoc4n2Kx%XlGiu%(wOV
z77-h36vL(iaXiJ_MLb)9h-08;MXt@nF=E6m@8H=KLmZ2mdE&sqUE*x%-%Y%6~L<^8Y8@`1Be+G$s11q2F)@
zTAN8%_?k2)6gEmIXl_jaU5WGA#Almi)Rhn=OgLY6gx&7{i-bbh??ewrXz?_%?>!&u
z-?=lA-#TddA(=ocfABVshk_ycZ>%|6=nVeR&l9g<29{|T{PhqxARQpqU2ZW%dPAEr
z-9k^FX~&F6eD6hH{QTQ^wgO+?mdV(xUBv#+(~!%Ww7%n?3uf=l<2A6a;LrcvhPVY!
zBWm&B2rvm_bZ}rS+lti(q8peY0M%`u^3w>bv(kqcI9?FiwM~TDA4y{U^`u0b@PzgmH;QaEJc3PD2)I-?!g8
zxdDLT0a;J
zCW8O}DLDF6O|nVUq?}Hn*_AWRO-eLEg#Xr@YtS^11$1AjSTwyN+U{skp+BfA8YBVn
z+W*x3`X>GZU1GiO?U`w<+prw|DKwDYvc}SuvBBR0rr?}p@0n*7-h>h4sgCKYWiSPY
z{1_K^0zkR#cnVA)3%-)V9O+zM?w)1~)UG(KXAe)hKo+QIO>JZ>P@C5jvM%_tri5gR
zNUm!9V$TB)sWKap`o)7J^cxQnc#cs^4-yC+z^ktVUcDcv0?j4Hz^lWMoCf=yW+7tz
zmi$G=j4Dx{-m#oDpqc&L>%;0f33(;ZH}UD^TJUIl@oa`!Zy}H%7W}A}*nT9(ZgmN3
zplbgIkKa+z%`0m3BL=;kDY~!Vd_ga;qU^yy7NHl5Z7^Tv7l8WSq;G}3*aBluZ}kwW
z1V7Nyq30&u=RF~Nao>;;9Jm221+R$F(rqEP#xq(vAKin(CYwuBpGJmC?hfNM9-my0
zf2MnHUOT`w-#nTt3jBmv08bdrgq?JS8o?8!nYd3d=!DTsm=mA53_5|QkpVjSiG|*z
z#kJ&e7olg7eQT0s1ip^OrQY!xy0?_GQRD
z;b%zEw`inZ+ay#hXG^c49(<>dGCi=%1yPQP336~b{0t`N9w7&pkZRXVpUjRp7hM)U
zV^(O!JW~xl+=rWwo~g!|6{2SwokeDa!UT^PYbWhmy>Ev~UxDbSRh<8SdOQ2DsOq$j
zk5RE&E~WYS?x>i$OG>*YXd1_ITYSmTTC?kFb255FMI8+jUtkWVVwT&OZfmRQtZh%)
zV>T%YCCWkFbu<%1Z4m|>6c>~jXA~F)&dh1Q_wNkO%Nf!^+dsPcM}>j&a^Lswdw)Kp
zIr^RMq|ggD6P27vY=vzA?M~#-3LD#tp%-+FcMSfv9~>wM64BP|
z(z$o?75`uD0A=Zt=50_AUHC$J?8L>Y?*HvDOC;2MW7=uWC-Lxcps{lKde#l!8Cla
z|C$Er3l|7&0_nN_$ub4Z6|{$Ap#r7~VxahG))ua#w>3HIOe3g3S
zN7X&9nZO!!#KOeync*%`oM>H~mqt$7dh@P#@R*o?{aX+}dV1KGSE?1~LHn^rbEnq=
zQLh7O6c}py^5}|WU|+<5o4~#<)vm`hGO#b%trsIbz&^y}F;J4W$7oCH=3N+r)KpPk
zSBNf5hxNS~EoqM7ycAjFvh6KzdZ-M@oz)dAYaPGn&ELA@z;8pf^#p@pq60qCNv2>D
z4Bi^ugy{%SP&6U)g~9i^4=gb7lrlFE3+!zR`0ogLLq=e!2Rw=etf-Y4eLQvkjm7EL
z|5Q@lt#2Q^b^9z?9wWRH&hVkujzIXll)nDyO86Ya!b_U^=MdQUqo?6nb%
z5|q2{GK`I_8KWP?LJyJsHHiU}%=VjhhMuM0Vjj@TnDkJ0C1>Js0UwijZ
zLdrmcQk^>u;1jIe1RmF;_GC|-7o0`pPPu42$^#m>WFt^H4Bbhf>GbDibf@VdFKHU;
z6w*^EO;)XZo#l|35d-~G04M~w%pjBF@V;R#v#66dU1kgLgxb*JGINdfU+AC)14GzLEC^K1$Px*USY)vmTcwSL`@e;^5v^UVx$zB0CCzH9An
zx|QtX#oSed87$Gg(fwHby**sV$oaxx*Fh50^4#xsd>1i%A?X9{%rDvyW=5O9yJ>I1Y#sI5|
z1izQ4QgbCroPLj$5ys_B0Hkj2HJ5Eq)AUr=vIGFDOeYpgGAWAz|&6y*|5c8*2nD69DPSl^}5sJ}Nh
zU!(ZePm`Gi{12a#^m_w)rSwLsxsII>h&rBm1{sI7NMzf91bIN3p-#lEO@f)b+!;GI
zr^H_BA5yI)H|CX<1b3ZdN)dr(TvMA!DcOC-+G0FL#!8V+j#9ElbM)|tEPmGXu`^O6
z{mz7i^9G8^YZKj^u_tDjn$jAmF%NDgJ+xww;I9ukmRmRqChue^rd6_`t1>8X!
zWr2o#2Q7p^19wnT_t6_>h1E_7I!$ej+jr#?dIJac)2HtVum|tA`+*JbxwG}M3~8hU
zM;ITd<5DSt6Hxb%rYHcU6aF(stuIzC|c`}%oxQTwf;21T3+u5}zrd48q3b4&~PddGj
z>%rtplul=4o3U~_o&3Umef+}Bb!}tXk)G+aGNw{&D^@`LXOU`JbK`aXL4i!uJGPYi
zKl+PT=3JI6u;wMDaA0x0LKq+CG)1t?Na0>xO}A%hdX9c*mtT3|rfzFrYomF
zEpK`7bzwv=+OWn^MtuA`0&{mSU0R*4XB*Z}$DGPr#Pjo0jxAWwP6QsCO)EA!yJDs$
z+?-ZKvuOn#yk5p=(|20ItrvWUn0a(u7=V!tsQz$7)j@a-qethjzh)avNflqQ(JpN?
z(I5b$UD{|m@g+vPZZ_HlWd);MY&1;{$MJ`ER%mR;W`LEB4fuYwYzD{%dTa(r$s=1o
z%m80>bd#qm>#9NA;EGUB)3ID}gQz%BRYCf`9k=Ws)Fgdht9!!mu9^*O#yXvbb>W(rw*D3=&aUaL~@TfRZnjzq5XC^JGxkr~ZZIRAO
z1?7k8ObxR`_32^U=KECP`U#;S*CTPc_>RnqFL6g~WblTjcNqHpNzqKxyVdau4Bn*a
zJqRtz#Y$*;V?4mRyhHf4`o-*TBY_s|YTv22JXEyGH25`mCy#z4ownMkip`!q8oZ(Z0bn83AZY3;R($)GjMt+OjS$QL!TaZ!P
z?839Jz5XcQE&V+)iZW^jhoM#+F?T3m)X?5~EwLm;`xt+4ZR)(TZ=U&7w2}Iw2d*30
zw>re)dMttYR^M}HiIE=PD#-j%*7f}El3e~oZt`nEebXewg$be!Lmixm8%?&-HB$S@
zJPqm|b)L~7L^TQ@BLaRPZciLRB?@{bG$2?>K8n<`ZUrNcC7jmWG6|s~R$)&R@xQ?b
zJZI>E{#qV0l4n95Gr}|B9W$=Rz4(b=_Tclov$}$#XvS_BaTEM!%keQ
zJ^GGmU09lt_UPp_BU_{&h-b=_DlF%1&9Z%LE$p6ek2Y^)p9&5Bs=<2tRBD{R@N9T0
zBN#C}^&%Lz)wKahgoCtz_=ieK=kObp?jj`&LmwqU31ct5XrvkK_%D`1Z$j)Nnk*e@
zWZ>Ts$Ie%#c(jP;DHiR&H5!72zI-u^oe=jC(QCXSW<`UOnTty|h!
znd!lj*-!8KSE2Y92xvvp2FL9T
zbioVfep(Kf@7w1J{7$BHyG?Oha9oWv<_diiUe&XvU0^8GCx{Xr%#7yuP37Sp2v{=V?>@*^W%Fe@u;l&Py$msmkQhK%Q53#amZ+k8jJlv}8-3xLet~_8Fr5qehs&+E
zwD3_6b!z-KgtPA7t@oPZj=EuyVClvt_g@eK}IkL^(gEXVZna
z0nSQHsg{`>ALQa$#}1^fZkonGZ)BTj(XP}ZYc{Jyks?oMrfOI(8UN6TS-$ygKDDX(V@R1%Rz1(5d{2lgc9(1(mUG`|15
zqT@e!j9qmbfsZLOkP3E7g!37}Q*FjduWLNjW`cXvAHCJ4#bC#uhQ#(rA^mBO#Mb86
z)FQD($=!#aDNAfMKb)P3)5nQS;$Suv(NyFX`BR;|T4zuep_V}XYH)6$$uZpD61?oanzJVbT?^ry7@$M%(8DK+)=
zYOkC%?_v$K4J(vgno|Prf7G>pcfJZOL5LF={&}YF*=sng?{!s>^*!o;Y^G&SkC5PZ
z(Xx+Vo!3X8OGMhd8&~rcs*a1evUrh`9!B@qZK=dN*0_h2pq}o7Nr;7YT;H&GR?+yTJUFar)t$`{Q3?1bVqycCOocTn
zI4V}+bQ0TQmGMMG%rtaLTL}Nip+s?9ITlT3&d#Pl!buiL(8{}LEh5RMjF%#D2vv=E
z0C^+19()T9gh0i!S&vEKXI~r$H{Blv67(qDy@aF(lqkwh%@-b&&ItQODyt@P`PHd^88BbrdAxu@8iFIpNO
z=V$Rpp2i~#n+nGX)bQCBS7Aj--41d!j?#KujW6@|F9qgm{J>FYyL*?s@NHdv&D(bM6LO)PNy$mH;wGF$w<0#?e$MAa#dhgf>}wh#1`N%yI)
z6@YgJ>zfr_KXVHOnZQU31+DtsKKD5bNh2yF5OLaMa%fA29%P5%SF^LBvAh`h;o=iL
zjo+;talB%srgstEJ{J2}b<=J`BkqiA69j%2=DEJhk0?l~fJ}P}uAUq2KT8iTseZwk
zd_Bs<5i!Q4-%rnzInyAl={FGrm!;a|0ZEJAY+0wd9@#cnj3REXfW>$Uz##eazy(N6
zzqBb%O{IQipC`{rJzNuU61K+`|Mwfv~s#>K5q*mdKgUV?T^NZ4
zBTvLB=b`B-q0aw~Od?S3L}Sx$xJ(T*c$~7bv6)}A0W8j6I=6$x$?Z>Fk@slYxO2EZ
z7>si^_L7dvFwVEM*h@M?FwXw2oeMwS)^2$7TmB@WWJDymAP<2liUEW)1|94kEOWQv
zetlN8EcT9PkimZ7LFI18HIL}JIxBH@!D_>Lp_YI6ij+{gd$g1MN6%1HEtSZgUGn5D
zSjoDp>0Z1({ALw9!z`ZT^L%4Tmh?ZoYp4O`?USSBJ>XF%y?emzy?elVW1A;&p~S0N
zdm0znST85DOnN~_sfE9WSSBT(odFw8W|@?hjTcY8_cBANe+Q_}&U2n3e^f#BZ3-9rd&!5tFZ-CY6%4<0;Na5^NoySux)H`+i$Pv_k8
zoqPT}Gtb;}|GD>>sqWfUy?6KSy-L=*)?2G!^RRUQj+~UN6aWDM0FZ&d0I&^!wxp+(
zIRK!j2w(;P0M7x42zUS_cntym0)$%ukpET(0O;W*0D$xz;oq-*NBrw5GXHm^zv`%+
zf4&Gi0q}pYbZ~WWv2<{xDSQ-
zVeJ6i=ct{?{74AY07P5_BwPeoH-G~EL?{S
znB)~5Jp&^XGdB+}AHRU0#Cu68X&G5LHFXV5Eo~iL6H_yD3rj0&7gslT4^J=ekk6rE
z;a?&m6B3h>Q&PXBrRU`r6#ghGE-C%>ySk>fuD+qMv#YzOx37O-aB^yTW_E6VVR2(~
zYkOyRZ~x#Bd~tboeRB)ByZ<8>ygmO-Eco%?B>N9?;lkxYL`FtJM*Slf0;0zs!EupM
zs5qYCiL0P~a>A$P40=xRE?_X9Rlk7hySnz*IvVREn
z?{X~xBmjtiOGt?DUP3~G_Y^Wbp`fDtNvO|J|CXNrJ)!?Cz4((b|B+zun;^jN0srwc
z_!|=q7409_{?irM3Vc|;hb;mykPzU52?-Y<26&9liv(UFU7qJfG6MefY{QwCVoZQl
zkco~^ojiZI>vo2(yAS4MJF^f#FLmZRJJ3J-0yVoao=yLR^@tJj6*#nNbxLQo=X|Im7m|1bULY*O(MGw%t9jgsL5m5Ti|6LW2u!z
zQkt}qF7lu>V_U-r9nv9u)Dn&>u6T24B&2JwJU`6JqTjn!e`c7U?>RVoUeU6DVJ6nC
z+^>8E%IMi~DLRu-_YX-1V+vNOl%LSFDka^AF{pB4iEpdV?9}HdGr6NP=0#DV5U5Cg
z{r~tOxt%dD@=f4>?x6V=E|C!BAL0ISNPHUr8~@)JFQ%d0wk9MWvBrI}rx7q`rg|4>L|!WaVwI_qAiM<3KjI>9IxhS;g#;+5#6^fe24-@xSXq1?@YT
z@fy~^HMPs~TahwV2W(90YJB9!Q!C4Z*~sh3W8X&jBY*3RWc-f!CnpkAf0((#=Bga`
zRrC0GfHLKk@+feazu02lr&8A%H(cj2`R8$*!MERE-$XDE>|%L14qgW;GVTw65SB3e^6+zO1E@&
zVtuuWq(#Fdy!qJSj}rLboD=`_(_ep-{HJq2!cPF;83<63B>2|^_=!I!
z%Mm0A)c!u#nP~s%!`~DD5^UmsqmLPpw%8)MG$6IxdA3pQm{*}?On3l%wxCa>#
zHl-vQl%5|L^oDMc*e!gVj)mN#o{H9Vs(!TK8TryBRc<>vOfFjcvZy-mcr3sC%np=?
zuO~yRwye>-+l(qVd&V6`zw+zzPSO?6VLC;`n`=q5X4<||{87KMU9dLVDtAr0-}Aq>@t=17|3gNAP5y87_U2h63~(W8O>x=-1AL#$*2;xy!2t1>
zAO5X_n=jgwhuT)G?%DdGi4;#@wV5Z=iA&8-jSbal=5@{m!hQgl`y~XN7Gb>$8wg9v_TZYL56H37{-#D{BYmFqgazrKeFJ^UkZ{_H!T72_LQ#(st}$^VD4NRQsHFXIdkAnrt~8$IT0$`32K|pMQVm
z;Qw6y@tc{+gGxi3HXDH0N;~N<+glMj7S5dOfMNUta
z&3xCnyN@2kQ!odDoVpvw+(u|R9S{*&1Oo&y+;JUP1Gjw6miRh;Zd={I%yrv0l-YQS
zF&OIzh#LdD7*1b>PlSC-`9KR695y#ciTnx!FbDc^71KhKVZ`NnfvP@qe@p!3R{>)c^2w`y@%AWz|iv
z-kBju{?_{7K5I6IAq~mYL#?0ZK+wLytT;4=)UO9*>p@O1lq@q
zUo7v2alR6gwNER83rcky`cFvk&uZYmt4_`Xe(2fiXnVnvF~4Rua}}^K+Y*=Iik={X
z-upVd9yA;fwuV;&3cVk`KMiuQ;hDbKUE-6`r{)6UeP{0wa>`Mdn`)~bE!@QZ0t4W1
zuZ59gSEoY{BwUBk6MmdJzOZ*_d`a6*o0F6>;umlZYOnGYJQq#f_%QJ1X5X|iSsG7Y
z&O=7J82Lo)afa}rRwex-zftH0h8>~q52i~&-`QK1)4Ker_mj2FW530SMmU!paKW@Y
zA*BauU91{HMr*Al)-^Lf&We~%<3idtezr2e0AoLrVE}&$7$8;!2EYsji#_>qgC5Rd
z0H|}-4gw5d?To4nk&)g<@h`;T^gJ^Re`x}7pu4Yw(sugKq6ZLG_~*Q6tIQ6kz*zST
z;k)mHGL4cC+EZYj*|*`<{U{2C2yOaPR)Ze>2x3R;2}v
z7KN({5UJ54@yyP77?7nXao20Al}l8T
zZc@IRd&&tzQ=Q9Lz=oy&l-T8*0SwTN3Il9u;P~g%1<(wGVWPbu3
zOBEG?Qi9uJ*zBMZ>ms`HKA*pcyL`OG#!zKW(0or52;B%;k9Ou58)CR|gj{-1J-mM%
z5UD6BBbJzGRQ%g3kAZhX*P!U5ODIb$0wo#7H7MM*ZPBM!lmZgN7uw2rkfQlgWtr3}
z=9dh}R`!ne&1t7hKV@R*`nE}--yV{a7~U9)FujIS=3}&j6`TJ`a{6xF9cw?Nm2+d1
zB9@c5e9(>R!~jHeYBYMV2m|asWUCMzr-J>@oxyb!02
zG+#B4fmOm{7p)4X?htSwO`EaZ@~a6$w#{`clSw*v`b%uGND>_-2__O`ig&<0w#B+9
znH?D5(QhGOf2VSl46G;8XAZF?w3kF1<*<^@clT3Lc<2&NMNQG_QQdX^OgpnP!zQu0
z7E1B^sR67fnWH~-EwA^SO4n=;XmwA4z$2!;uvrpT``5;ibgZX)=%x9z3Elv~SzoJ7Wt{cjo(?XrwX&r-~a7v5^
zJ)Ata2)X+CPs9NdZ2JFQh|BZ+Uo6Vg|C2@ecZmDr_B2o@L;}8E&Ni|zbQ1}>*d}tl
z5BQudJ=xGQy$%C}nrNn5Vq5;4uDE`VJV9nrn%G}~vD+*+d+#HQ{I!7?BsJ+PZR1l*
z65zqKl^xc6_|3iMeYGnb#$8w>KjuZgbCT?4Aqd<}`8Ti5tJqL71|!5==Qw_swUcda`v_MA7x>8358G!;oh{ljB&#b4_>=&5Mu3$G@s5*4h^n
zGz{jcnx1~0%K1>~H%UaMqC9FCJ;O_tGzPREx4oHJHWzD1xyUor5|`^Q{ijV+zjj#-a$OJ2EnUBC1T;+H0YhMMHg
zRc{u0-%`CL++x^hJCXTb6|N1yNPH~Bc6jgvO*yY&|D11T5mvMvji#f`0VdcP<=>{X
z%qhl=aX?+ZbZ%ON=cj698ZYjuxVs;=tTcHno5zr}JeI~2?J{(8`8>Ao-D@3+YmKdl
zCHpgay2yZQ+cv=_8-?{b-J-UY7%t3;yY%m4c7HRD>1Qp^)xZFqHIF3$dtp^@#myLS
zN8P5db&A<#PWW@8BCb$&s)%^ej{YRT%bJvD$SsiS`o9ood0ohdL8H7Y2$_1yQA&1CHyDZthmR}To_=&S!%JmzsEI>;;`S>dg8}+O2%m`V@t*qDE)-bXj+w-s0+X-Je2X3^0tVO{
zVSt|nFu*;c){KHS6Eo-Ab^PcY5O=fqpF5B|@rpi&JAcpcq0ilXLmt$j^bYuzQClZi
zIBu`0IqRzOc8aQxnwEP8U;x(~xq;~bTOT|6r#<;;UlKQznF0{TM9qDeEDfDD#j-nf
zvFwA@{zdPE{QBL>Yq?(;S5!&SIMVUsk@vB$)>qjEou+@kp?`Vv^e809<_5N%+aJr=
z#A4adA45@N>oK+qI2}Ex$mwbf*oI^1FOOJXiA^>%#f=r?Sud7qyAq4ypO4G&WKY{w
zhZ8rw{)Sl3_X-}%0=8p~V1T)m)Tdb83l^c9p7YswiKcm#+sAldHl^tH)f;h8P~P@pC}(H67?$`vm|1#%HnM0r7t-&3ZHmVGJvw(hOtYhwhWRb!
z{hYbb=gUTY11*6tq$Sn8UD;9edmxXOeRk>ob_ZBWSu6BlhqL#AIkVRPIICvH>UH|W
zLU!^4uP+e{@cD%lISUwBd#*-Cs(;zV!XA3OR}jl?H16A4b4};V9$*)xw2GYy13Z(5
zkFV*82b=fDcx+B37hi5o9v4nng22*1o}KGUwM)`#uxjbah@luQnu
z5Y8$zLuP)q}93Dr^rHD|$2N_n*u})U$z+m89h#$eW_mt%%Dj__S-=
zKI6O-a~qSJQGE_#Ro^ajm(V2io+{4gzQG^I)J)$cer9G%-bF%Rc1UzS-)Wj0d~e>|
zrZG#`QosB;sn7d{bGXto!bQ<1JT0bg>c^d6dSh=J%eaqc*Bh@snrHPLS$&I@z|^Db
z(ow-2H6y2xu6MyZL6`iIr}h`Xk0tBz=|@mpN~4!nq>ZHY$*s}tuXb!o&UM4UQt8Sz
zpFdrrQa4@bhu{1k
z$%Mn6a|R(@qByf|55b7q3(6b=y>hwjgN#2sLz|=OTOL0!Mfe
zD(|qh-?mduUXF^9bf?a1g*D#Xpx}w>ogKpRoE69dIV%1h&fvx1{6LGU`(@$;Tidy)
z>ti-lV`B`(^S26kE~>J*zT7s8CPwP6DElC=+h+PJLE=2ph<|~l?rHL-0ri~01S{U*
z-~|V*YHmN}{XOn@Y;T*`6OsSu?2{-&$faB#*Q&y3x%yuHSX708@<%TWV653cX3V7z
z>IjIupX*R2)Q52Oa>%92FMqRN_dWfU!_P%|S_BE)2p<+WdJrQsC03#XMRuEMkilt=
z!HvR9)d!L5wlSqSXhz8IW-aUwT_1)IiZ{wC%|u)i18jC2Qd&;#Nx(EuvbPap$0Tr#
zn>XUC_COfW-?iy*4~E0)M2wLY^nHRr>ub+2?Qepj$sgr`{X@A56cM1<02?uU
zh>bSH!^En
z&jpE2i}$ClMpd`a#226kbY(c-N$00y#PlDK(qt1Gn&deK{_Xz<3$fr>czK2cWrR3w
zwusj_;6VIGFTm~CLQJw0NJvia3crwQOsbC;)38?lE&qbC8wHc8Lw-_Xe<
zLP72HNT6vEZEBvvb?J7Rn`~n;`?<4E9dYfGyKu-80a3@;Jz+Tva8?ok+3_?wAv(|B
znETvHY-nRhI$m+imfuh#na)M^5yV`Eo%Us$%oS99vLo`_jgxYF!;XXx!7(=~Lt2wS|z4)WBw|^^8a99zpI`F~eFX
zc(P26@9a=1fQ&~*aDTNk7$N`zysOE1yk>&|$UqE_u>spzhekK}4vzq>+3Tskg&Pa$
zFHWw-n-;_+mPy+Q6Sj#0o(yzDrQ6!p$&TkqnQrzRc8B-XUZ_w;wve1FAe=f}Q-uM*
zgH=}Nd>eEz(|on9#J^>~yCv|7n%KDspki#3QK%S#Bz=AB<9Myd2vyg%Qtxuqf00!Q
zmGcO}0KsD0aIN9lSN_ksBLFu0UqA*EnUdRBk$rqCVAZvu*DS%Z425pb^mY9S*;40=
zfk1hri|9GZ+uOPeF7D5Pj(kpWMKuoSSi0GoK=xq$pWZ3gOG^>+xHPm*a68CCtm1=m
zngsrcB=hK!rZFih!|T36@tbO?=bU-Nr6AF&wd
z_CASBurF7v=q=XLA%6-iT?(RXBfhTK@T|ZnB%~`@_oqKMIE|vpo+mlM7HWUU?yb-J
zMl4cx{-pQ4Rpaw9P1KPo-5phght`s^5bs!n$nlQi0uzDMGkPvB8xN0EjP$mVCVB$
zMC8xg%B^?cSM?7f4}mLLjjz^aiHn#v8w#a!oJ8WBbn;yBP*q#MgLP89t6%2Etx`Io
zx)@s61h~DbDdjLrBqn*NDVW1prxL0s+ihZR&^M!FI`*8;G>NKGR?GGo5Pk^bH|b3L
z3KW!yca&ZMV0X{VQZzsB6FAV=O*Mb}wrly^=CDMUNeU>@UYuxeFo;syfOH@O=_iQo
zS#w`$lasi_HJJF#Ux@4OgL_?>*ijj=Y4eOBNN^VLHvY5%gjQnb-X33_5xPYQ&TiJq
z$moUWEL9!sph*)xd&u!BK{iz@vGzN(Zx<>JGDi8=O|6FV99ho}zsW08K;ONuT*qR1y9s65b(M$xTA`6lBb
z(>LYj@U2N~kcMM%+gson8-f1_IXSp3q6o|uA5|e=5joxi=N{J;h^~>^9el&tv`S6+Lq|
zN=>Y7t;%Z?>4$`nB%9+7|D3#?nuP(ZM5a!2ps&XoM>&OA&4N;lzQonfrabU2fRrJm
z%A5tZk
zG(E@K$HEqKI@g{JY@8S#zGzBwplx8WCmllHcuVpTP1vO!WS$oZ-QpUy*#c{D)Terl
ztX$2FKS0|Mq*21c@UbeM16dnUG@F}ircUVFTH-&|)k!7jkcPfjN$8u$Qq=j;NLxH1
zmpjp+sR<}!BBl)KKz1YSBQ+D?}x|83q9GgA$y2O&~E3lvxt08JZBU0F-i<
zs`%=Mkq#P$1nPJR--iA2XQ<#;Tg^NVq@KSv?|H$76=r6pv}VxwN6*xK)uf(o@VmwoUFmZ&BSdL2cNdwEThhxU%W6
zN&hBi{%O!BAWTb3wM4yfx{zb9Uh;$T6n$%Xxe*q`NA~@`n~Q8Jj-j%ktse4dQ>$
zAlR_Ee+{j^eto7)7*S^9-e`T|j`C$xXiCMwgn4E|*3fpmsU>WJPsrIDecMe@)us@)
z_g$q|pDJ_oT$#HE2W@O{$;*!3DT6CZx-!uK+;oB3{GTq>Ap<-*>)ARcxQWYn$^eZo
z-^da;GzwZ|XuB+sEvQ`LDqLw3g&TON6jCX;{l0Fwb1HI}B01LyU-vS1uDl1IsU`C$
za4ySDTCT)ZT*Tih;@#6<+YSL3~+t4?`0mkJy=%T6&;&w{66?_=6eIu
zpnq`VIK)&YY
zz%8y8mK)!8r-4(U-~0c9L#C}l&Pc0>vUG3j@c1NWrx0CCij=~iM6a7>3|_yh2F1!E
zHlmqeAoxGk9DOpsv-N%AV4#m5%X`eTWnV;0OfRF&$YoY%``EQud-{PIm%_d#y)t;d
z8-m#;AewxjmQp7?7&+L(KeKI+B`L|$^j={78p?cdk<9ND+efHHxhK$9QrTbE$iMSa
z^Ab(5!T>u|=G9D?|7~{jqv9ude9Rxq+f{wfYPqFs{H#YzQ9V<*?jv{mdU@C{qmDm%1$F
zp9s#33&y4NcKXM1#b&=N)61B?f3wu8R&P=?S&(dK(V3I+bCgS^hpAS`^h>bIs52v~
zXA`Ftm81;xC&5QVkpn|~ywkV}aSx6J8FEp87&{?bIzi7Reoo8S3vt@`Cc%?;z1vpS#!?M%eSV9izHf!M6lmtj0KBWu-HC$;G{*MyZyxJij~!q^`~n5bO8G?8Hz
zVP;P^<)f?bipOG|hF_GQ(aSFPeorQg&)L%$3~ItV^@vX)_1o_YQr%W@0M0k
z)`KqKG^}-#bW(Mcu#m1ueTQ99h&oB7#LMWD0ftH&U7n(6g{W1TB~o%D6g<}622XKf
z3pLMm>7QPTx%Cy7AKvR}Or5wDzg+qdWAtg`%O1njYO_zbC|1)Xy+)toh1e*kvX@QyXEug*o1(>SgGI%50N$@AMM*jL&r8M#
zp(lRIhtJvE^yL`yg5fO7zeUgeOC>N391v+MOp!ZMzUI5cFLR5++V`DeF0#=vX*(Ci
zefV%DhF@T0fyU8rF(coYPQsDME1KdPTcwCVnZ}*$=2`SA?Rj0d4I7!R(!icf@9vua
zb^^7d`0JZufa`(L4^a%TdFoaG<~$_leUhSuA6|b$yCBM8^<
zkc+o=9b=yxB)>E-DcY?7^ya$rUlQYw=;^Cy_C`aY^0;$Ig{~=$0A$jXQ6H?Co`^oR
zyc!F8QZ|sm;`5yuolzV-bP{9
zL^zGjOiB@)Cw=}ncNgaKs3jLnpjTTmsmQ@s^#
zCep;Q@+(n!jz3-MYP`vg26O1pQztm(%s>!Wg7qR+Wvy8KjnWW
z?l0JBw!~ny&xU+!Km@gdyYl^^6!UStj#dgk;gGfNMe&6l!m^${3?Mh!GGot5i|qSF
z5J7v=#71DR#Z9$q&84H)952x+ETsT7{=wVubk}I)@iIXUav$o
zQXjfV$ftMiBCtF`QXI0{U<2t(<}th7VLE(@H!)3~?P8&pw!u6Xb4ZqlG9$PJ3c}UI
zA3RJ}tIXpIn=F?GNq>a|>XBdR)i*S#{GhO0G$p>`A3IR_mYaARru@x)L_Mq3?$cX>xq`N@1VzOYuYB@CC(Z)3m!x(vUB
zUXsun_e9X%kk260)kcIb6tFNI#b;_5ZUa+xZHg!*$UBxZjlZYpeB=*2aqr!V#bW01
z%1^vacs4d+Ep7EaupU66i>{aYd|0mQOom(lhqQ^
ztYPvYtktd8R+1|VF^8IP3%g0iY@sSlf=PHb{VR5g6vrBMsU3Doy_D9dGMUMdiXP$^
z(^tgYpJ^o2Srt`$bMG#t5^VX90Yx0eM>(oi6MCK5rlovdfx+5)Vp%fOD9`PBs(OiD
zd|*l-4WP;D2pX%(yN^{l^@2x*V>&p$+EbziCUAa8tZB_o-l}_fu{b=gPZcx09!3l*
zD||e>3oK^Xxr-Il!~$)saOu=?7%{C)%aV`=-OrVT4(}-zkEuIrb^NIG9rm8o60d-G
z!vK=?W?5*Tt8fPv!fo$gbu1kiNy}AR1G~X{Z
zD(ncZOWY}*m!W8OLLz7;n!9yYdIPB0f%lI#b=b5u!pjN@gqPpHHs~kby2mHVQt}lH
zixaQ!74sB{ROVvMC{~IjQ5@;)s+naCwpR_b_oE&eCRrP4XR9A9IhW%-TaJoja@LM*
z_EsnTk-}-_5}y?+y^E=Q6Cm*U_-%+k8tQBRWX=VhKIUO_=j`yw8V%erW-W7F>9DTt
z9A!|2QEk{N(K&Aqq7A9BT2YDS4a-m8{~
zTJmhu$ufb|PNS%MN7qDr3!XI0%NC+_--=KxKO$>HLXrBPQo}n){{*i8msI`pAL&5P
zY_D&$PGv$|WB>xw?Q{s+=3KXcuvJ-xj}F3=?Lx|mR=x#7W&(mx)oM$?R$R9jwSX^4
zf$YQF2R0*1eu-cli(LLun{5f4Xh(if{f2D`lb?OwT<%>Rk%elqbV$MDOk3J|4)JiC
z_0M^%34A>1x!IP>8z5vf`}K?&#A*`_+9aa%_U6qa-gaUV>O`;LImco%!{%`%T@A|o
zJ1XO$Y_u_b@41MmJJ*s5H$}mBznJi(siQdww{d)M4}p`T?BEJdnTubUMl;LSmEl+R
z^P^npa(BOP%pgNHV25`16A#A(n?>vAHN^4gSo}H!91K>(0|UFQiJ(#IN!RTSnrnS>
zX5vV>@;_mozXjp`8XPx(TQC68)1<%X`C258B9b!1BA{f+>Q+^+$BXQOdGw1T8G>#5
zv!m%po27n;&+Clo96BQPK~#Gv;2
zF*!d{nKPd~l{`)l^rFI}`hctc
zoPt^4qsxGaQ9#&v9Iuy~ttJ(n3V*4S29vJt!^lwjB1`g_yYGl0Ux%M{)+@z&od!{z
z7585?*Z!Jf8vCGdv5m4;S~%BFH{EtlZS*9W|MaS9Q~43k0KV*pNC%Jm)&&P_5<3w=
z@57%k;VYa2+2QwV$uNL$+c_FMikM;9EffUF>B=Vf{VF`dO~`CFl&=nG>x%6_9)(nD`<$iDmGh0jyspOR1TPtY&h3i=2tUUIA(+!
z<=>ToX(0gY5n(HQbbn~f-i%Gi8zM!OMo3QD=FApHhT9~Vg00ez$-Y{f`zLOoP*H(k
zmw@%V6k>Z?P3D{JuQaPtSj&i6gY3sBriXCU7qb>>(Y!cPgrr&6S-rUs#!1QG=pB|T
z>8T>a5=_6fadZ
zA?n(8k3Lc}xqIU#B82fWsTu|HPjclYooY^TA}53%HXUr;eUBz3=wCy9->etXk+$j?
zoK&aX&X5p<=Yocef}dW%mnZTaHsJ+@A!
z(dU;b@jLjF)#-z43Nl)54(igb{Rr>W#DX09dZQe|c-siwWV!367B9B6ZHhD^-~M9%
z!nP<^>;8T8+}}P^YnlS1^*JaUYGSfUY806s7a^6cE6ct2wb{9hNjc|{UIC!wrPg0I
z>PgBv^z*R#@pH;N4>zY+lZ#02WCYMN6BnC`H)B`cJ4RDKxj|N~(tTd26*WPzCpk0Aj<6TiApxAE
zHrnOmw7^(VljZbGXRi`8%n?4ax&G0DrRJ=$x+3CE;VMmmr~p#WcSqfInV%bnumGrhZi4T*<_@>WHiC(5VtS}kWiXtU%w73D>~xkuwG=@Cc4pj(MTF)
zrD$6DKEy)Qjrf{=@CBUNk@`%;QLH+D`OAYW^P1;=)M!K0E^cF3)Gc)~OD@@h
zJD;FD_q9Mq%ju45qmDj=%!p#0J*`K-j!2dLGYygZ%5rwCp&V~p+C9P4o;5k~RnhZ;
zzF~2R4w5~>i@mOZ#r$@$m%i$)#OF_`Ws8O{tWGuM=S^Z?|73DT(;?mt>Lc7-R6Yp@
z+jQ487Kma&G@L-PyJhAy$88)ktc)|6uc>Y(mv6NsBYUzi|6m^p
zU7TrUewqbtod|8jTpoC{HkEZJE)_Vv-KTLrByQp!6o{!822Z75G5MqWPi0d_+k&LeS9NYU0HqblQWSBPqW9h-+5sx><%XoE$O?HH
z;jBX4(;w_E0hC-KP`tA_7=#QTp?vP#%pxPS&}8ye!g_VB4>Axd@LBMuZn0+mL7hX}
zX$=EtBy9K=hxLdkdGR=f
zSg_I%zOr6XBwbMz1L8nO^qUg~guD{675TWo#{x@n<#X4KnFk%5`QA#tn}{}P75Mc63X07-vm1UFqU-*3Tm&Q*9N>PR3k4Uo1vgfyBT7#
zy#N>j-=AJ{I8~R6q*3z^4BYS+Q`SXVWlL*VsP&@gQ@@u&4hF0nSB0lnYUh6RJ`wV)
zz<;l-IAW?EeW#D~IFWOtUeuf*0Ieu4i%YT)KrCeM3@V7)b>yerdseno>_W3F6|A}a
zaW_FhNwEz3a}UkkFe(jF;%dS{UJQFNY7vP*aS8)6emyyaOl_4z=#AI91$_mz3IV5H
z{~EKy)~@-TdyWEm^+3Td4U~uhVT8rBRfKDdb9c^7>1`*QhkYd*Ct^
z1HfE5qo&B^cS^tlQnvN;(3b8=%%x-iu+QY^Ojcyz?!(7Gz_xiQTL#(ZQk?STy~(eY
zQ5WK5%dwHq3FQWvK1p2yMr98cD-|6pckIw(@>zmbqYN&?E=g%Wa57;e;6#28>0=UYFew7rjN1P9-&VcM38_$eJqx}auSz3___0off9MV3$tI*nD
zbGc`YJKx%e(~dMB^8UpQ*&s1U#8wm=ie&|NEapz5
zRbndD(NjkN#oWBJ&E;9&Dn~Oo_Sb#V8m*j6-RI2D3JCXGVlh4f^jdGs%y(P*&z_`5
zov@sLCJ8XM`DMoDmgpa*{QAvoP1CQrjXVbmAwf>w7C}i5Um=m
zkswA_e7C*Fvap^*^1;4(4Bgmv9e=f%)vm(2)G3shjBKtIYWfK|g`e^)CMo)Iwi#v0
zQAk<&$dHxF(2i@ibAfhD0fG-UepsLDCN;uqS4j75*R#fXxtQ>-k3^WFA**W{{c&U5
ztp(WKC{o#!x6Cq#bgK7N|aod>$95Iqz@fc{AP7%3_!4xm8uRdl35CYt<_
zt$?^7luNSx+tltgJ4cUirAwea`66ll5hQ>shN+Q<*N37oNi75^9v*x)H>Cd%^tYvk
zIQB5!6dhivNS(A;LUm_Zk-$RG*jVHmAyf}M)tvx+3E{%%>>K8=zR$&cRc*Osio8y|
zuGWQrDdYcvqsxyC!?Bk}pqpR3ST(psc4*_XPn+nF`;(oQK&yj1f~=hEpz|b$-oTAf
z*}fM=UrzLm*vlvS13|1D6R&blqqmJ8S>8BLSt~u~`~@K0p>i+jxO4WIo5b;-^-eCw
zUDESq0t-96i;U094=l|GAl&!|YK1|Hy1dzrH^6c2HX9?HNBRj}$_YN-PJ$jZuDgSh
zGXbDLuFgQm{6m{n7+|A<$LOB!i5>=Mg>U$WR-C_GBq
zR}2Oqt_-+gX|a=Y__CPz?q>H)C)Ncc$DfNStCkOGSAQKmt<`U#&lFnpb&F%hc_utd
z%wllHLKQkCdoVmX)#kF&ZsBH;_Mxkb@Wh1GrU*+~zwGtmg+skiNd4Du(!
zHp<}<>jEBd58TL^pL>$XA!s}1u!HYQT-6#nUxOx+^c7-13VJVqpt0=aHDXsG^P)^o
z6RoHa@lAsBy)SRh%o^(JORWUHCMLJqFUigNV&ajedfl5!vv~T|W!N?ipuRJf4(2q~V{C1MV?u
zj;99B7qoEA@BDFjl!(1c3|64HKG0Fr*s_?xpZuk8r?2$g4LmwedI>cb)R*(Q;z8KU
z`xeTo%)S<2Nl|s@Z!VmZ;qLBqTQ%FJ(HsCeGJJU+mJjmukV!O{wtGqQEDLM%GBp`D
zaFL`{!+04hxb;mFq+&~6o4&Ob@{Br?&WbKYUI(!>R_EI{exwqfv5m|s*K_{*@u(Ut
zlo9ml14Vv=gRwH@G#0`-sZ8BsdHi=8ubK(63YFBNRHu#)3Y&67p*ju=Tjp@50Nr62
z&a^KAZ^pQW`x29vZXpi~2S)Nwlda^FP#v)J246T&lGU3UJi9%0yK)ncXdKe2N*i&=
z!X`GJwsbkfVck|RG=l@p#Rh4J#8h}A1&XZGJ+ny!o1
z1tnImoIl(CNHutUwp`?QZnTb>+G{7YzU1kk<`W~WY$xszmj5MQgWfJojLk``fkxm=
zPYh8QI}ZZ9$p)u>7Up=-V`5eG{%3hyOQ2u0^Rh*v9)8Y=(USYX#z8*BYtq)c89yEd
z$WLhu_-sN^%hrvj6<^p9{ZQ_9;nnIA`!v|C+O3tD^pm56j%iRnTZ0;8zS6Q}cc1;F
zld@52a^S2MZ5QFFNZT@l`Xk#z^t!e-c9tNRs`K_#QO7|eHalpqP$YA!Jo9a@2~7xv^f_3gz>bv$^{OOpscU2e&BxD#1hEXs`o;l
z;#*KGa9M0)gW`GHK4E#l8ZJDp&u!RKYIX7G1`5vfP@x@o%nY}UB(^wy+46JE8V~+PW^bFQRNy!lZ>(_nCa&1wR-I(r~28adURN@F#4Jc9W7;
zhzEM5TgEZnjETGBI6g
zva;n;(=4>AHbt{e{#s0Ou-If@)rP&Ybnu>sT#t3OsjcMvHqf8IbE;3QS~^0M3HoWu
zf8<%@klptM*{CQh`1bYK@YCe?$eD4ikNkwazIKyWO>Ijw8(}dvR`y>DvzkUQ9P{!C
z(+^d3-gex*Jp@+IJ|R9$-G|i_AHFi`eNiR^%xKuHt<$l0!r+t{Av3{xhAU&pzu78N
z1+fE%^cYoVt%ZC7Z_gPrpReQ(`o243^2Fm-ca>=J?8q59TAsT~#gxX3?VTNl?_W%a
z#dG}d*}8Z#j)B_$%3U>iB_q*6J|}9=zLgYP`Pokmfl^}{1}1(!zL60-3Ny1jSr+Aa
zY=~5~kT}muR!lYi9c!!UhXT7`ft|OY@PJQVpCvX0DH?s>i75*FwTgNw_u|<2
z^FuORHWgp3k|yUU#qTjOB@n6ey8|x-F#Asoy&kYv^%j?)GWaEQP^s@YTrhx0*(Jnm
zn%$QpbFyo5>x<&29B%al<;9_u+er`2Haq3=lBkL=qrVQ5iv?M)Q-M2e?KQW0MwOwU
zrN;L@%Hh#1*9)@HTu>ujL&7tbSEpmv5Yn5Sw*$>$n>_KegWv|&WJx+fRsGZx{((!QcL6A5Zjl43-4@HI2I_@{U40IWmMa3yX_0biWexZ
zMN5$4E~OMN4nd1k+}+(>i%W3{9^4_gyB7%V?(poq?>=X(z0MeWo%JCZ`J9p5_w}FG
zoWJ=N-BxYzTH;GR2uu5J*}MSHV^sBkv(^c|Fn_TmoxQ-8F|CiLM&H}#*Ss4RC)ni{
z*2&aYs4rn!s3mSgmzn(hJ%)6mze+@xlFJ1K&B$SXEfK+d*t?dS6H?v$sqzGs$kNr4
zo`^KE7k8zhNYchL8W+N_*C){=ldKD#hZ7vb|k241FfB7ZIK3!c7vBoJk-5HqZMiiEl3t4E!iYAs=wtXScx|7R*k_-}&zdDCKtUnyv+A67O-
zA-^+NRtyc;Z5(>~!$FhVR>hI-C1SMJ#F{1_ob-jfflk;;x91L90zX2HMF!ES#4iIHwkDqHv>XT*e^ay=N*xRA7^7io*Zw8
zQ;xeV=Hz{;S1M?xd10)13@?;k-pN2=6kPtuYUVGqJ5Efe18~TwwFw=??k&L}4Qwt!
z3Aa}iNwKNWjBR3E_sD9n)VyHN^5s@>S2x5RL;t`Ya(b?sVJToTjDfVKRm
zps{${cw8)DiJz{h#0ip`7|>$N^1=uj>U|5^xoTRDUc;D9->?-B*aqX|Op-z_>l2cC
zZL>?69?=XH+71?6aiN}it@N-4)n=}CB*C&(=T42OYWL2sLVnrV+Bmc}b!?bEVNu&>
zzi*$Qg(9|5sFTYs#uSrh<_ds**XFn^Md{)L8PG>gqRD}s9@Qj#C%>eS@=VX+B`CqtUix`=fhCwd0DPy2nJEH9rEh;IybiU(bWy#-yN~!40
zW<S$k&52zg31n^tky6ShGtMEoFAWs00u99wP{pKjsrCWT1!BFhO{$Sr;161ORnp`z9tO=OQ3_*Jz`AcJ1>$#wIc;7sr74p$i%?Ph(W
zoMr3^XHU9B*kPsk@Pu!*%x7CIOOJPSj^Jrta6ywb-_a4<4y13mujAn{=?jQvn{)vT
zEx%7kWFu2uwp6=a@|=Wkmax#|bR?_=;fqly$eG+%K8FN;*`lomI80dKX}m8Sq29vruaj%4oP@
z&*AuhlV!^we9W^;qHeq!n-_|LXnbU)vUhd1U`$e)$p@DZ4fJgPts$a&VJ4KS2R8`v
zbS^w~^>A5VPo#a_{vK&aO8B4=aJ!e8w3`mp|HAMP>X)KrabNbcS(yK>RHWxi3T-s}
z-+LmmQ<3JPiA+u_Is7pV+)r#H(H+tQoQk#TvEQ-pqY#=7EDU4K=VGd1Q#e6Sx5(No
zP$G7Z?MQtT4*qzd(Jq7ab69~krP-lPn?jT5&A4BRJ`6JUCYJj57WUbEaVrU!q-XsW
zlK;Z@MD-^UApgJjjQ>M>PJBE(LtfFi`#q$KTr}jSw#1^}r}a{H=L(mzM^rV3$(o0v
z$g@0!yMn(qEzR&y08C!#K{@axyg+SkGSnF~X#@Xyk{3iOEi8DM@%Z#cY|tMIKeQQ{
zY86wTPWu?a&v2|>{RL?7(af&Hw4~9?;CIbu}Ch!G<^={|J-(>>t`G2)LL815inlp41OpE
zO4eih50xNQ?0Y>Id$^q$Ecf)0kzCd)%jTO&{0&@x+&qKW;~YkTECOlWh1m5L*wJiL^EYaZv7-AN-=5!u5n?s
z{Kz$}X^x5*ng8qNW}BqlU)L7+Hi7=Efh*>OD~}f&9m`SZBjGJ#8kB<8bZ{psKI{+)#z(%DD2J)P*DGj{
znRfoX12^9L43S(rGm$dUm1>hvwiuzMfqR;=!6TU?89_B*LIoeBm!DHtFHjN+HOtBC^U7v8M+s{dj
zYRZ0@jP5qLbaBvZQR?UWVt%t5^~}Dg3Z;Co^do{Tb(t3ZF3;b%HFIlCp+}tFl5=z+
zpp9d#NY{|J8~IP^krY{Fe4?`VxkPW%e(3Ee)q3t`{VQ2_tucuOGKBFbMA89^R#;f#
z*n^Q5=S!WgbBo)A;f9eC^>eq7jP*5Rdt=UooZwQ2ENA)-wB*AoyfLQP(Hs&^H5&#-
zFxOs2lSl@O^O=*RF%C1fI^m?X~_%+6FhyPOUt&iaE6Zx*s
zn5RY`Wo`9Km<);wR8PXiU@xZGYbaK
z7kJrV%QL@g)-Px^>uKwIbLruB@MG0)(B%r@hXVbO{
zo}AKfi6LGKQHKR{;g~k#0&+i`wcBi-M;=#L{U(VSC9S9slxx`l1`aP+6QvBVq#L}~
zLq-AnAUsPyoaw`R)6bt<&q#I5{eEU$E>Q7qVeMm#kH=cCNG!bdDd=#|$$lnzNcSZ)
zCm9Sp_p23h!i4+S3Z}M<${9UR_QuAXoC@5i;;tQ(sNX;^f_9H}HZgno0akyf6
z_I~HDI>DOJ%F5HB)64`=q|*>h6V29l0ylZ)sN}4$mAAk`_{1ZwTQQ1~(3|06VN&WM
zu21eS?&0*LU~xXe2+n#qa~>_Lh(5?i{?e@94^>Fe?vJ}&n6dbSFFU+~v-aZ$TdbuYWcrtgZ-+ug{rMksw{lMOX
zt@{sLMCX^cF_F)gOAkR%?DtmtDjy{9bQC_W3ikC5Q=7`M>9TU5kOk@6W$b_*9&dGh
zFfdZkC-i19x%q)A%c2|H_;g+ksi!(k0YQ*K%v33C4LnK<^D5a|2PLbfmI{T1w!@a$
z7b@6j4)Jm98>FP<%I3h^?!pVqzyu$YcbTrCCnI7meV5*bg%)QM;j3+5(|xJDMJ^|!
zz}a%z)7W9Y09q!i%0aHW>XjL0Mt8%6k*b*jPb7?XN3n4y0k_l13lRv7^v$a_J6ic4
zx^^+8&)sk+Fgz`sGyb^6KX5Q<@Yf{Si}l5Q40F41MJ~P^v!_q{NZCh4mcI8hTI%|H
z=5q4kM3GshC`Vi2%N}!{jqLW5ip#VGIn5(F*R=tP?XJy
z4w$Y83prkK`$mObR76-!hj^;xL1{ByPw5IYwE5RPIicm3dvcP6WfO%}nTq!=6sxaE
zf;|YLO)y3DAGi~et|u`I{rs3NI=omu?uXZWZ@Guug10G>#lbHRK$4RJjYfih;FkaX
zYN61hip1P7OPGABeBF4c&i^-7>3l;3rWvM)TE61Ny$DV^0Uj7Pw@?A@mLGSY
zS7c{{ZRjc&Cd;R9%!V#Xh)vxE+)KCV=q9xb7ti+2iDeHDDY`uFoAIdefA85Tt0~EF
z)&;V0U<>Va!fY9#oew!xW=WfO?L<4y&8tBtHm!^Euw&T8?-Hda^^Jc1hBOZIudWe-a~E@}
zY8$o5<<>_m8Oq=SpTFZ2rWbxtI%RdZ7jMmu?zQ9P9OMQ7+%Ob3$1`mrT7GL5C|MC8h#v2h-6~bZZ@o
zB`Jv}O33ODl#CGXkc)~2SyF{EkwAx`6h%vi&xCfqbyuc9GXmhCXUgL6{*rtXCh
zb7AD#lZe@<=eqH*;Wt-{Ue?IbIM2Nq7AJmIAL;M;{;hS{xpmD>U}EB~
zjKaC}HH-CWxrMWE_1$iiO(q
z^2JRjQCssRD+rn_5YK`zNP{l%EqM4^5bCDb+og_QoIwee>oE4kI)gSc?ep!0!ooSr
zl7t8mF)oyL9uj=#d>zJ2GTsY~ty8xTMve51c5&4ae-2x`SV$Gk!L6-+4Ed(E%&70y
z&TRMj@gvI0&o=-Mp$nhqK@X~WjF+l+{tqq}b$6>@FDrki-t#W}wP6O>a{vRy5%p7a<)I#7-VUfvthu4adt(;e}gdnsT$U@;cq5ATTr<6fAEaP)+
zeAN4YILfjA8{ARe&7Vkk*y_n7<4ly|NV@R=D27EhNsEY(<5Mw~&k-Ii&-bZ?9sbp-
z?O~dc630b)Iabv5>&@%RVw$9P!}B*Dzu;)6B>fswE)V;Lzlkq#yDxhc?oWsRa?J6#
zWnf2gvml!!_R5!cdGZiTlJ~spOuU$kakl&LJqo?jP4_H3ntiYY
z(KGl?WDUg!Lo`Ox=$ydNm?v3
zT&#PwXY#f}
zPetb1fb9fRCBx&-CoQ#5A*8)B!Nlzo3DzVtd$-Xxkq*vCXFWC9OEy~gsVC35?v;se
ztU#rLN{vu;uJ0DV+!d&_|9FUBZdKvNerYHlx?M10|*rq@ywH
zvx$R6e68QxZ@DvVpYVq=I2ck7Y2h`4iHP759_MC;VsI<#l8yM3H>y#PcuG-(#cs-H
zwaRga@LNdLvf>4EwWo1D*Cj7pYsXxMBAC-A#mn536*b0e5Su<`BTHg`^YO8i*87z<
zNX?syf_U}$Gh+JiWY>C$kBqW0Z=G&D^j-o64Y_27`^}WmXp`!K-s0(^XW6yB`ofz8KdNyV=B$W(!
zjD9`{=}MtlQ`|FB1EM(=^2|<_XJvp5KZ9p&hN$YljI`{fb4NSL?WH_UV4WnyEoAdE
z9etl|3~`IGrN*@}
+
+English | [简体中文](/README_zh-CN.md)
+
@@ -59,55 +63,43 @@
## Introduction
-English | [简体中文](/README_zh-CN.md)
+MMPreTrain is an open source pre-training toolbox based on PyTorch. It is a part of the [OpenMMLab](https://openmmlab.com/) project.
-MMClassification is an open source image classification toolbox based on PyTorch. It is
-a part of the [OpenMMLab](https://openmmlab.com/) project.
-
-The `1.x` branch works with **PyTorch 1.6+**.
-
-
+
