AWS Machine Learning Services
This is a short refresher of the 7 AWS machine learning services announced at Re:invent 2018 which will cover:
- Amazon SageMaker Ground Truth
- Amazon Forecast
- Amazon Comprehend Medical
- Amazon Textract
- Amazon Personalize
- Amazon SageMaker RL
- AWS DeepRacer
- It aims to provide an awareness of what each of the ML services is used for and the benefit that they can bring to you within your organization
- This course would be beneficial to anyone who is responsible for implementing, managing, and securing machine learning services within AWS
- You should have a basic understanding of Machine learning concepts and principles to help you understand how each of these services fit into the AWS landscape
Related Training Content
In the Back to the Future movie franchise Doc invents a time machine that allows both him and Marty McFly to time travel. Regardless of whether it is in the past or future in which they travel, the consistent narrative throughout is that they gained a significant advantage over anyone who hadn't time traveled. In general being able to predict the future gives you and your business the same advantage. So how can you leverage time travel without rebuilding the DeLorean? You use Amazon Forecast.
Amazon Forecast is a managed service which provides you with the same time series forecasting capability i.e. being able to predict the future and therefore gain and execute an advantage over your competitors who can't. Who is this service aimed at? Well anyone or any business that wants to predict the future outcomes based on historical data and any other relevant metadata they have in their possession. For example consider an e-commerce retailer who wants to predict how much summer stock they should maintain for the next summer period based on purchasing behaviors of customers from past summers. Using Amazon Forecasting you can go from zero to hero without the prerequisite of existing machine learning experience, simply load up your existing historical data in Amazon Forecast complement it with other related datasets, configure the various Amazon Forecasting dials and Amazon Forecast will train and build a time series forecasting model for you mostly hands-off. Before training your historical data is typically loaded into an S3 bucket. Next, you instruct Amazon Forecasting to import your historical data once it has been loaded. Amazon Forecasting will by default inspect the data and automatically determine the correct time series machine learning algorithm using its AutoML feature. Of course you have the ability to override this and leverage a custom approach via SageMaker. When the training phase completes, you can visualize the time series forecasting model using the console.
The model can now be queried to predict future values and to do so you leverage the Amazon Forecast API using either the AWS CLI, and or the SDK. Amazon Forecast uses the concept of a recipe to specify the specific time series algorithm to be used during the training phase. For starters, the following time series algorithms are supported: ARIMA, DeepAR+, ETS which is Exponential Smoothing, MDN Mixture Density Network, MQRNN Multiquantile Recurrent Neural Network NTPS Non-Parametric Time Series, Prophet and SQF Spline Quantile Forecaster. A key feature when building Amazon Forecasting models is the ability to assess the accuracy of predictions. Amazon Forecasting provides accuracy metrics which should be used to evaluate and determine the overall accuracy of made predictions. With this you can fine tune and optimize your time series forecasting. This can be used in combination with the embedded console visualization tools to drill down and further understand time based patterns that exist in your data and therefore your business. Amazon Forecasting takes the pain out of building and training a time series forecasting model. If you need to predict what is ahead either today tomorrow or well into the future then try out Amazon Forecasting.
Stuart has been working within the IT industry for two decades covering a huge range of topic areas and technologies, from data center and network infrastructure design, to cloud architecture and implementation.
To date, Stuart has created 80+ courses relating to Cloud reaching over 100,000 students, mostly within the AWS category and with a heavy focus on security and compliance.
Stuart is a member of the AWS Community Builders Program for his contributions towards AWS.
He is AWS certified and accredited in addition to being a published author covering topics across the AWS landscape.
In January 2016 Stuart was awarded ‘Expert of the Year Award 2015’ from Experts Exchange for his knowledge share within cloud services to the community.
Stuart enjoys writing about cloud technologies and you will find many of his articles within our blog pages.