Deeplearning4j

Deeplearning4j is one of the few machine learning frameworks written natively in Java that targets the Java Virtual Machine (JVM). The framework was developed by a group of machine learning developers based in San Francisco and is commercially supported by startup Skymind. Deeplearning4j was donated to the Eclipse Foundation in October 2017. The library is compatible with Clojure and Scala. For cluster and distributed training, Deeplearning4j integrates with Apache Spark and Apache Hadoop. It also integrates with the NVIDIA CUDA runtime to perform GPU operations and distribute training across multiple GPUs. Deeplearning4j includes an n-dimensional array class using ND4J that allows scientific computing in Java and Scala, similar to the functions provided by NumPy to Python. It can be effectively used as a library to perform linear algebra and matrix operations, for both training and inference.
Deeplearning4j can be used to train models that can perform image classification, object detection, image segmentation, natural language processing, and time series prediction.