Blog

Feature learning makes relational data usable for machine learning, unlocking a vast trove of data with business potential for companies.

The idea of storing data in relational structures dates back to the 1970s. Today, relational data forms the backbone of every modern business. Corporate data accumulates in databases, playing a pivotal role in bridging the AI gap identified by decision-makers. However, despite the enthusiasm for innovation, extracting value from relational data with machine learning (ML) is currently only possible through significant effort. This challenge limits even large companies' access to machine learning and business applications with artificial intelligence (AI).

Remember when "requires Docker to run" felt intimidating?

I guess I’ve been there too. But fast-forward a decade, and now even my dev environment runs in a container. Why? Because speed, consistency, and simplicity matter.

After 9 months of development, we are releasing getML 1.5 into the world. This is whats new.

It's a classic, around since the 70s, used it myself back in uni, but data science needs have clearly evolved.

With TSFresh, Blue Yonder's engineering team set the sails for brute-force feature engineering, combining a large set of 100+ predefined feature templates with traditional feature selection methods. This approach performs well, but their implementations limitations: computational complexity on real-world data and limited support for multivariate time series.

It's a classic, around since the 70s, used it myself back in uni, but data science needs have clearly evolved.

With TSFresh, Blue Yonder's engineering team set the sails for brute-force feature engineering, combining a large set of 100+ predefined feature templates with traditional feature selection methods. This approach performs well, but their implementations limitations: computational complexity on real-world data and limited support for multivariate time series.