cancel
Showing results for 
Search instead for 
Did you mean: 
Technical Blog
Explore in-depth articles, tutorials, and insights on data analytics and machine learning in the Databricks Technical Blog. Stay updated on industry trends, best practices, and advanced techniques.
cancel
Showing results for 
Search instead for 
Did you mean: 
lin-yuan
Databricks Employee
Databricks Employee

Customers with high-frequency time series data often need forecasts at a finer granularity based on their data sources. For example, an energy supply company may receive energy demand data every 15 minutes, while highway speed monitoring systems update traffic data every 5 minutes. Previously, AutoML Forecast only supported single-unit time intervals (e.g., 1 second, 1 minute, 1 hour), making it difficult—or even impossible—to generate forecasts at these multi-minute intervals.

We’re excited to announce that Mosaic AI Serverless Forecasting now supports 5, 10, 15, and 30-minute interval data. To leverage this feature, simply upload your dataset with the desired time interval granularity, select the appropriate frequency from the “Forecast Frequency” dropdown menu, and start training. The rest of the process remains unchanged, and in the end, you’ll receive forecasts based on your specified data intervals.

 

linyuan_0-1741673944668.png