Writing the Code

Lesson Details:

I: Introduction

A: Cloud computing for machine learning.

Cloud computing is very useful when it comes to machine learning, because it is possible to use the machines in the cloud with a limited budget. Cloud computing is an affordable way of providing the computational power required for machine learning. There are many types of machine learning, including supervised, unsupervised and semi-supervised. In supervised machine learning, there is a teacher who tells the student what is right or wrong. The teacher is usually human, but it can also be another computer program. Unsupervised machine learning does not have a teacher, so all of the tasks must be set up by the learner. Semi-supervised machine learning is a combination of both supervised and unsupervised machine learning. Cloud computing can allow users to do many different things, including writing code for machine learning.

II: Body

A: Writing the code

There are many options for writing the code for cloud computing, including using your own computer, using Google’s App Engine, using Microsoft’s Azure or using Amazon Web Services (AWS). Google’s App Engine is an easy option for people who do not know how to write their own code, because they provide all of the services needed to use machine learning on their platform. Microsoft Azure provides similar tools to Google’s App Engine, however, there is some functionality that is missing, such as writing custom functions in Python. AWS is an excellent choice for people who already know how to write their own code, because they provide good documentation, which allows developers to understand how to use their service effectively. AWS provides each customer with three “virtual machines” (VMs) for free. They also provide a great deal of flexibility, because customers can choose from a variety of operating systems and programming languages. Some of the features that AWS offers include tools for deploying code, complex analytics and data warehousing. Customers can also make use of the Relational Database Service (RDS) to store their data.

III: Conclusion