A Fault-Tolerant, Elastic, and RESTful Machine Learning Framework
Rich, query-able access the visual world at unprecedented scales.Try an Example
Stream or serve Terabytes of data through state of the art OCR engines and Image Classifiers.Try an Example
Understand the meaning behind your database of free-form text.Try an Example
Identify faces in thousands of image with just a handful of lines.Try an Example
Understand any image classifier with a distributed implementation of Local Interpretable Model Agnostic Explanations (LIME).Learn More
Serve any spark computation with sub-millisecond latency in Python, Scala, or your language of choice.Learn More
The Snow Leopard Trust partners up with Microsoft to automate their Snow Leopard Recognition and detection and monitoring pipeline. Learn more about how to use LIME, Bing Image Search, and Deep Learning to create a custom, distributed object detection system without any human labeled datasets.
Spark is well known for it's ability to switch between batch and streaming workloads by modifying a single line. We push this concept even further and enable distributed web services with the same API as batch and streaming workloads.
MMLSpark provides powerful and idiomatic tools to communicate with any HTTP endpoint service using Spark. Users can now use Spark as a elastic micro-service orchestrator.
We use CNTK on Spark to distribute a Faster RCNN object detection network and deploy it as a web service with MMLSpark Serving for use on Unmanned Aerial Vehicals (UAVs)Watch Now
We use CNTK on Spark and deep transfer learning to create a real-time geospacial application for conservation biology in 5 minutesWatch Now