Apache Ignite

Machine Learning

Continuously train, execute and update your machine learning
models at scale and in real time

Ignite Machine Learning APIs Overview

Ignite Machine Learning (ML) is a set of simple, scalable, and efficient tools that allow building predictive machine learning models without costly data transfers.

How does Apache Ignite support ML APIs?

You have two options:

Use built-in ML APIs for some of the typical ML and deep learning (DL) tasks, such as:
— Classification— Regression — Clustering— Recommendation— Preprocessing
Use external ML and DL libraries that use Apache Ignite as scalable and high-performance distributed data storage:
— TensorFlow— Scikit— Spark— And more

Benefits of Apache Ignite Machine Learning APIs

Expedite the training process
with horizontally scalable cluster

You can distribute your training data set over an unlimited number of cluster nodes and train your models with the speed of memory.
With built-in Ignite ML APIs, you:

Avoid, or minimise ETL
Load all your training data sets in the same cluster
Minimise network utilization during the training process

Execute your ML models with in-memory speed from your application code

Once the model is trained, deploy it on the cluster and execute it with in-memory speed. Use built-in Ignite APIs or 3rd party libraries.

Continue updating your models with new data in real time

Data and user behavior change rapidly, so you must constantly update your models. With Apache Ignite, you can update your already deployed ML models with new data sets.