Machine learning models require thousands of data points to train models. This session will look at a number of mechanisms druid offers to speed up model training. The tuple sketch can be used to train models where many different metrics are required for training. The sessionization extension can be used to speed up training for use cases where the need is to forecast viewership or session experience. Druid’s ability to output metrics at different time granularities can be used to smoothen out time forecasts.
Demo Video: Imply Polaris Overview
Imply Polaris is a cloud-native database-as-a-service that simplifies the developer experience by providing all the advantages of Apache Druid, plus auto-scaling for seamless data ingestion, enhanced security,...
Watch nowImply Polaris + Natural Intelligence: Powering Real-time Analytics & Observability
Imply Polaris provides the fastest path to real-time analytics by offering seamless data ingestion, auto-scaling capabilities, built-in visualization, and a secure, fully managed infrastructure. Watch this...
Watch nowPhysical Hardware, Digital Analytics: IoT Challenges, Best Practices, and Solutions
From supporting ad-hoc queries across massive time series datasets to ensuring data freshness and scalability, our panelists will share insights into selecting the right database solutions that can deliver...
Watch now