How Amazon Forecast Works

When creating forecasting projects in Amazon Forecast, you work with the following resources:

• Importing Datasets – Datasets are collections of your input data. Dataset groups are collections of datasets that contain complimentary information. Forecast algorithms use your dataset groups to train custom forecasting models, called predictors.

• Training Predictors – Predictors are custom models trained on your data. You can train a predictor by choosing a prebuilt algorithm,or by choosing the AutoML option to have Amazon Forecast pick the best algorithm for you.

• Generating Forecasts – You can generate forecasts for your time-series data, query them using the QueryForecast API, or visualize them in the console.