From ccac69d00d2b670b70358d5507571c01167acfb6 Mon Sep 17 00:00:00 2001
From: Vico Chu <30412827+vicoooo26@users.noreply.github.com>
Date: Wed, 13 Oct 2021 19:48:54 +0800
Subject: [PATCH] Create README.zh-cn.md and fix typo in chapter 8/2 (#399)

Co-authored-by: chubo <chubo@megvii.com>
---
 3-Web-App/translations/README.zh-cn.md | 22 ++++++++++++++++++++++
 8-Reinforcement/2-Gym/README.md        |  2 +-
 2 files changed, 23 insertions(+), 1 deletion(-)
 create mode 100644 3-Web-App/translations/README.zh-cn.md

diff --git a/3-Web-App/translations/README.zh-cn.md b/3-Web-App/translations/README.zh-cn.md
new file mode 100644
index 00000000..f6d8505a
--- /dev/null
+++ b/3-Web-App/translations/README.zh-cn.md
@@ -0,0 +1,22 @@
+# 构建一个 Web 应用程序来使用您的机器学习模型
+
+课程的本章节将为您介绍机器学习的应用：如何保存您的 Scikit-learn 模型为文件以便在 Web 应用程序中使用该模型进行预测。模型保存后，您将学习如何在一个由 Flask 构建的 Web 应用程序中使用它。首先，您将会使用一些 UFO 目击事件的数据去创建一个模型！然后，您将构建一个 Web 应用程序，这个应用程序能让您输入秒数，经度，纬度来预测哪个国家会报告 UFO 目击事件。
+
+![UFO Parking](../images/ufo.jpg)
+
+图片由 <a href="https://unsplash.com/@mdherren?utm_source=unsplash&utm_medium=referral&utm_content=creditCopyText">Michael Herren</a> 拍摄，来自 <a href="https://unsplash.com/s/photos/ufo?utm_source=unsplash&utm_medium=referral&utm_content=creditCopyText">Unsplash</a>
+
+## 教程
+
+1. [构建一个 Web 应用程序](../1-Web-App/translations/README.zh-cn.md)
+
+## 作者
+
+"构建一个 Web 应用程序" 由 [Jen Looper](https://twitter.com/jenlooper) 用 ♥ 编写️
+
+测验由 Rohan Raj 用 ♥️ 编写
+
+数据集来自 [Kaggle](https://www.kaggle.com/NUFORC/ufo-sightings)
+
+Web 应用程序的架构一部分参考了 Abhinav Sagar 的[文章](https://towardsdatascience.com/how-to-easily-deploy-machine-learning-models-using-flask-b95af8fe34d4)和[仓库](https://github.com/abhinavsagar/machine-learning-deployment) 
+
diff --git a/8-Reinforcement/2-Gym/README.md b/8-Reinforcement/2-Gym/README.md
index 7f8df688..6331cfc6 100644
--- a/8-Reinforcement/2-Gym/README.md
+++ b/8-Reinforcement/2-Gym/README.md
@@ -121,7 +121,7 @@ To see how the environment works, let's run a short simulation for 100 steps. At
 
 ## State discretization
 
-In Q=Learning, we need to build Q-Table that defines what to do at each state. To be able to do this, we need state to be **discreet**, more precisely, it should contain finite number of discrete values. Thus, we need somehow to **discretize** our observations, mapping them to  a finite set of states.
+In Q-Learning, we need to build Q-Table that defines what to do at each state. To be able to do this, we need state to be **discreet**, more precisely, it should contain finite number of discrete values. Thus, we need somehow to **discretize** our observations, mapping them to  a finite set of states.
 
 There are a few ways we can do this: