Analytics Zoo provides a unified analytics and AI platform that seamlessly unites Spark, TensorFlow, Keras, and BigDL programs into an integrated pipeline. The entire pipeline can then transparently scale out to a large Hadoop/Spark cluster for distributed training or inference.

 

This session would offer the tech details behind Analytics Zoo and walk through its key capabilities. It would explore how to build and productionize deep learning applications for big data (transfer learning-based image classification, sequence-to-sequence prediction for precipitation nowcasting, neural collaborative filtering for recommendations, unsupervised time series anomaly detection, etc.) with Analytics Zoo, using real-world use cases from JD.com,  Midea/KUKA , Microsoft, Office Depot etc.

 

Who is this presentation for?

Big Data Developers, Data Scientists, Data Engineers

 

Prerequisite knowledge:

Familiarity with big data and machine learning

 

What you'll learn?

Explore Analytics Zoo, a unified analytics and AI platform for distributed TensorFlow, Keras, and BigDL on Apache Spark

Gain experience using innovative application pipelines and architecture for the new class of deep learning applications on big data platforms