DEV Community

Cover image for Getting Started Machine Learning with AWS
Saima Javaid
Saima Javaid

Posted on • Originally published at Medium

Getting Started Machine Learning with AWS

What is Machine Learning?

Machine learning (ML) is a software development technique that enable computers to solve complex problems by using samples from real-world data. Therefore, this technique lets computer to automatically learn and improve with the experience when not being explicitly programmed to do so.

How it works?

In traditional problem solving with software engineering, engineer will gather data and analyze the real-world problem. They design a solution and write down a code to solve that particular problem. It’s a laborious task to develop a code for many real-world problems.

In Machine Learning, developer writes a code snippet that will be trained by the real-world data in such a way that it will start predicting the outcome of new data which was not supplied before. That code snippet is called Model and it will be trained by a special program called Model Training Algorithm.

Why Machine Learning with AWS?

Yes, that’s the first question arises when you’ll start reading the title. Why AWS?

AWS is offering so many easy to use services and tools, which makes it easy for everyone to dive in the machine learning journey. It abstracted out so many complex implementation from the user and provides easy functionality that any new comer can easily start playing with. The mission of AWS is to put machine learning in the hands of every developer.

AWS provides many pre-trained AI services which can be used as a ready-made intelligence to a wide range of applications. There are many example to such services, for example, personalized recommendations, modernizing your contact center, improving safety and security, and increasing customer engagement.

AWS Educational Devices

Amazon SageMaker

It’s a fully managed service which removes complexity from machine learning workflows, therefore every developer and data scientist can deploy machine learning easily.

Getting Started

As I have previously explained that AWS is offering many services that enable every developer to easily start working with ML. However, AWS is also offering many ways of learning for the ML students and developers that are new to this field.

AWS has created an educational portfolio of devices that helps to put new machine learning techniques into the hands of developers in a fun ways with AWS DeepLens, AWS DeepRacer, and AWS DeepComposer.

  1. AWS DeepLens: This is a camera device that helps in learning Computer vision.
  2. AWS DeepRacer: It’s a racing car designed to run on the physical track. It helps to learn Reinforcement Learning technique.
  3. AWS DeepComposer: It’s a composing device that powered by generative AI. It creates melody that transforms into a completely original song.

AWS DeepLens

AWS DeepLens allows you to create and deploy end-to-end computer vision–based applications. This device is integrated with several AWS machine learning services. You can connect with AWS cloud to use deployed models that are trained and test with the new data captured by the device. It enables you to learn and explore the latest artificial intelligence (AI) tools and techniques for developing computer vision applications based on a deep learning model.

AWS DeepLens Console

AWS DeepRacer

This device is an autonomous racing car designed to test reinforcement learning models by racing on a physical track. Reinforcement learning is a ML technique where a model, normally called as agent here, is trained by feedback from the environment. The agent will be rewarded in numbers for the actions it takes. Actions that help the agent achieve its goal are incentivized with higher numbers. The actions that are not helpful will result in a low reward or no reward.

There are several racing competitions held monthly by AWS DeepRacer League. You can enter in these competitions by sending your model for chances to win prizes while diving into machine learning development skills.

Note that the device is not mandatory during development, AWS offers simulator where you can test your model. There competitions also not required physical device, many competitions are held using simulators.

AWS DeepRacer Simulator

AWS DeepComposer

AWS DeepComposer is specifically design to provide a creative and easy way to get started with machine learning (ML), specifically generative AI. It consists of a USB keyboard that connects with the computer to input melody. There is an AWS DeepComposer console, which includes a Music Studio to generate music and there is a learning capsules to dive deep into generative AI models.

Here again you do not need to purchase a device to sharpen your skill, you can upload existing music and apply ML techniques. There are several AWS DeepComposer Chartbusters challenges where you can share your music generated by ML techniques and showcase your skills with other.

AWS DeepComposer Music Studio

AWS Account

You will need an account to access AWS console and use ML services. It requires payment information but that is only for verification and it won’t deduct amount until you are using free tier and services that are free. AWS offers a free tier that allows you to use several services free of cost up to specified limits for each service. Details can be find on the AWS website.

Reference & Further Readings

Top comments (0)