A unified analytics + AI platform for distributed TensorFlow, Keras and BigDL on Apache Spark
Analytics Zoo provides a unified analytics + 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.
1. Data wrangling and analysis using PySpark
2. Deep learning model development using TensorFlow or Keras
3. Distributed training/inference on Spark and BigDL
4. All within a single unified pipeline and in a user-transparent fashion!
In addition, Analytics Zoo also provides a rich set of analytics and AI support for the end-to-end pipeline, including:
1. Easy-to-use abstractions and APIs (e.g., transfer learning support, autograd operations, Spark DataFrame and ML pipeline support, online model serving API, etc.)
2. Common feature engineering operations (for image, text, 3D image, etc.)
3. Built-in deep learning models (e.g., object detection, image classification, text classification, recommendation, anomaly detection, text matching, sequence to sequence etc.)
4. Reference use cases (e.g., anomaly detection, sentiment analysis, fraud detection, image similarity, etc.)