batch inference
The process of running machine learning predictions on a large group of collected data all at once on a schedule.
Unlike real-time inference, which processes data continuously item by item, batch inference runs periodically. This reduces hardware costs because computing resources can be turned off when not processing data.
