Data Analytics and Machine Learning (ML Ops) on the Cloud - 专业及持续教育

注册接收电子邮件关于开放大学的机器学习项目.

Course Description:
Machine Learning as a field is now incredibly pervasive, with applications in areas including business intelligence, homeland security, 生化相互作用分析, 基础设施监控, and astrophysics. Deep learning is a branch of machine learning concerned with the development and application of modern neural networks. Deep learning algorithms extract layered high-level representations of data in a way that maximizes performance of a give task. Deep learning is behind many recent advances in AI, 包括Siri的语音识别功能, Facebook的标签建议, machine language translation and self-driving cars. This course is an introduction to 使用TensorFlow的机器学习.0, which is a very popular framework for building predictive models. The course will provide a step by step approach to building complex machine learning models starting from the very basics concepts of machine learning and the TensorFlow 2.Google的0框架. We will be using a variety of tools and platforms such as Python, TensorFlow/Keras, and Google Collaboratory Notebooks for building, testing, 部署机器学习模型.

This 7-week program contains 42 contact hours of online, synchronous instruction and is broken into 3 modules and covers fundamental topics exposing students to Artificial Intelligence and Machine Learning. The program is ideal for graduating and working engineers new to the Artificial Intelligence and Machine Learning world.

This program contains specializations for Retail, Healthcare, Financial Services and 工业/制造业. You can select one or more specializations as part of the course (each specialization is 3-5 weeks long). You will understand the use cases defined below and implement one use case end-to-end as a part of your project.

A PACE Certificate of Achievement will be awarded upon successful completion of the program.

REGISTER

Learning Outcomes
By the end of the course, students will be able to:

解释机器学习模型是如何工作的
将任务构建为机器学习问题
Use machine learning toolkits to implement the designed models
Justify when and why specific machine learning techniques work for specific problems
Build, test, and deploy complex machine learning models to solve specific problems

暂定课程大纲

Each module contains corresponding hands-on labs covering module topics.

模块1:Python基础- 1周

Python教程，包括:
- Data Types & Strings
- 用户自定义函数
- Pandas Series
- Lambda & Map
- 类和对象导论

This module covers the fundamentals of Python programming. 完成本模块后, students will be able to write reasonably complex Python code for working with data. The following table shows the topics covered in this module.

Python Basics

Topics	Details
Python Basics	入门，数据类型 & Strings, List & Tuples, Sets & 字典，If Else语句，For & While循环，用户定义函数，读取 & Write a File, Pandas Series, Data Frames, Lambda & Map
Data structures	列表，元组，字典. Using Built-in modules and functions for strings, computations and dates. Object-Oriented Programming (OOP) principles.
Using Modules	创建和使用函数. Creating a Module in class; Calling a Module; Returning value from a Module; Adding a Method that takes parameters;
类和对象导论	Creating a Class; Creating an Object; Using an Object; Adding Instance variables; Controlling accessibility; Naming conventions for class members. Inner Classes. Class Constructors; Parameterized Constructors. Inheritance. Overload.
Files, streams, database connectivity and API	Open, Traverse, Read and Create Files: databases, csv, txt and Json Files. 连接到数据库, create Database, drop a database, create a table, alter tables, drop a table, insert, delete, update records, 查询数据库并显示结果. 连接到不同的api

Module 2: Machine Learning using TensorFlow - 4 weeks

机器学习概论
什么是机器学习?
介绍TensorFlow
构建TensorFlow模型
扩展和模型部署

This module focuses on the fundamentals of machine learning and the commonly used ML and Deep Learning models on the Google Cloud platform. 使用TensorFlow构建模型, training and assessing their performance using TensorBoard, and deploying the models will be discussed.

使用TensorFlow的机器学习.0

Topics	Details
机器学习概论	什么是机器学习? Machine Learning vs. Data Science 所有内容如何融入AI & ML world?
什么是机器学习?	机器学习入门设置云环境(GCP) 了解Google Cloud的基础知识 Developing first Machine Learning model - Regression Developing Machine Learning model - Classification
介绍TensorFlow	TensorFlow是如何工作的? 神经网络简介开始使用TensorFlow TensorFlow与TensorFlow的区别.x and TensorFlow 2.x
构建TensorFlow模型	开始使用TensorFlow ML模型理解数据集准备训练模型训练一个深度神经网络模型理解TensorBoard 什么是特征工程? Improving Model Performance - Feature Engineering
扩展和模型部署	云部署规划设置存储桶 Enable API & Services 创建服务帐户密钥打包代码在Google Cloud上运行ML作业部署ML模型

Labs for Module-2
Lab1: Implement a Linear Regression and KNN Model.
Lab2: Create a model using TensorFlow - Feature Engineering for a DNN Model
Lab3: Improve the model performance using Feature Engineering.
Lab4: Deploy the TensorFlow model using Flask API.

模块3:行业焦点- 3周

This module is industry specialization for Retail, Healthcare, Financial Services and 工业/制造业. Students can select one or more specializations as part of the course. This module will focus on the use cases defined below, and students will implement one use case end-to-end as part of the capstone project.

Industry Focus	Use Cases
Retail	Demand Forecasting Retail Search 产品推荐库存优化
Healthcare	远程保健/虚拟保健互操作性的加速器医院影响预测生物医学数据分析
Financial Services	反洗钱(AML) 了解客户(KYC) 数字社会安全网外借文件处理
工业/制造业	工业自适应控制制造目视检查物流优化连接操作

Labs for Module-3
Lab1: Understand a business problem and implement an exploratory data analysis using Python.
Lab2: Create a machine learning model using TensorFlow
Lab3: Improve the model performance using Feature Engineering
Lab4: Deploy the TensorFlow model using Flask API