Credit: Dreamstime
Amazon Web Services (AWS) has made its Amazon Forecast solution available in its Sydney region, giving Australians direct access to generate machine learning (ML)-powered forecasts without prior experience.
Through Amazon Forecast, users can produce private and custom models, with the tech giant claiming its models are up to 50 per cent more accurate than traditional methods.
“Using machine learning, Amazon Forecast automatically discovers how variables such as product features, seasonality, and location affect each other," AWS said.
“These complex relationships can be difficult to spot using traditional forecasting methods, but Amazon Forecast uses the machine learning developed at Amazon to quickly recognise complex patterns in data, to improve forecast accuracy."
Amazon Forecast is able to automatically set up a data pipeline, ingest data, train multiple ML forecasting models, choose the best performing model for a user’s data and provides a private application programming interface (API) to download forecasts from.
The solution can also be set up in less than five clicks and establish models in a few hours, as opposed to months, AWS claimed.
Amazon Forecast is available in AWS’ free tier and in a paid tier. In the free tier, users have up to 10,000 time series forecasts per month, up to 10GB per month in data storage and up to 10 training hours per month.
In the paid tier, generated forecasts cost U.S.$0.60 per 1,000 forecasts, data storage runs at U.S.$0.088 per GB and training hours are priced at U.S.$0.24 per hour.
In addition to being added to the Asia Pacific (Sydney) region, Amazon Forecast has also been added to the Asia Pacific (Bombay) and EU (Frankfurt) regions.
