Machine Learning for Java Developers
Deep learning and AI is mostly constituted with Python. The community, support, and resources are abundant when it comes to AI+Python but what about other languages?
PyTorch, Tensorflow, and Keras are already ruling the industry but which framework to use when the only language you code in is Java?
There are many Java developers who are interested in AI but might feel isolated due to a lack of AI+Java support.
But there’s good news!
LinkedIn open-sourced one of the powerful frameworks called ‘Dagli’.
Dagli, a new framework simplifies the implementation of machine learning models in JVM-based languages. LinkedIn strongly believes in Java and has most of their infrastructure set up with that. Keeping that in mind, how can it stay behind in the race when it comes to the most powerful technologies?
Dagli enables efficient, production-ready models easier to write, revise, and deploy, avoiding the technical debt and long-term maintenance challenges.
It already supports common models including K-means Clustering, Gradient Boosted Decision Trees (XGBoost), Logistic Regression (liblinear), Isotonic Regression, FastText (an enhanced Java port), and Neural Networks.
But with the popularity it’s attaining each day, it’s highly likely to see more features from Dagli soon.
LinkedIn explained all the thoughts and concepts in this blog https://engineering.linkedin.com/blog/2020/open-sourcing-dagli and you can also find the official repository here https://github.com/linkedin/dagli.