Simplifying Distributed AI for Big Data

BigDL makes it easier for data scientists and data engineers to build end-to-end, distributed AI applications. The BigDL release provides the following features:

• DLlib: A distributed deep learning library for Apache Spark* (that is, the original BigDL framework with a Keras*-style API and Spark machine learning pipeline support)
• Orca: Seamlessly scales out TensorFlow* and PyTorch* pipelines for distributed big data
• Friesian: A large-scale, end-to-end recommender framework
• Chronos: Scalable time-series analysis using AutoML
• PPML: Privacy preserving big data analysis and machine learning

Features

• DLlib: Distributed deep learning library with a Keras-style API and Spark ML pipeline support.
• Orca: Seamlessly scales out TensorFlow and PyTorch pipelines for distributed big data.
• Friesian: Comprehensive framework for large-scale recommendation systems.
• Chronos: Scalable time-series analysis using AutoML.
• PPML: Privacy-preserving big data analysis and machine learning.