Impetus Technologies Unveils New, TensorFlow-Based Deep Learning Feature on Apache Spark for StreamAnalytix

Print Friendly, PDF & Email

Impetus Technologies, a big data software products and services company, announced integration of a new, deep learning capability for its StreamAnalytix™ platform. Based on the TensorFlow™ open source software library for machine learning, this new capability demonstration showcases an image recognition application running on an Apache Spark Streaming pipeline on StreamAnalytix. The combination of streaming analytics and deep learning enables a new breed of applications and machine capabilities in industrial IoT, voice analytics and anomaly detection.

StreamAnalytix is a real-time stream processing and machine learning platform that offers a Visual Spark Studio for development and life-cycle management of Apache Spark applications in both streaming and batch mode. By leveraging TensorFlow’s data flow graphs, the integration allows data and analytics professionals to deploy deep learning models on real-time data sources.

Deep learning is a powerful new domain in artificial intelligence with a diverse set of applications in various industry verticals. As an example, image recognition is one of those universal deep learning use cases relevant to everything from the web and consumer social media to our clients in the industrial sector,” said Dr. Nitin Agarwal, head of data science at Impetus Technologies. “Recently, we helped one of our large oil and gas clients automatically extract and process valuable intelligence and metadata from millions of complex engineering drawings using GPU deep learning in TensorFlow on Spark. This provides a powerful and fast new capability for reducing total cost of ownership, and improving configuration management, uptime and equipment performance.”


Sign up for the free insideBIGDATA newsletter.

Speak Your Mind