Datameer announces SmartAI to democratize data science in enterprises

Big data analytics platform, Datameer, in a bid to democratize data access within the enterprise, has unveiled SmartAI. The company claims this solution to be an industry-first, for operationalizing deep learning models directly within enterprise data lake environments securely and at scale.

What does it do?

SmartAI combines the rich data management and pipelining capabilities of Datameer, with Google TensorFlow, for an end-to-end deep learning analytic cycle. It also claims to reduce the time, effort and cost to deploying deep learning insights. According to the company the combined solution with TensorFlow delivers the fastest time to deeper, business-ready insights through agile creation of rich analytic pipelines that feed deep learning models through Datameer’s advanced integration, preparation and feature engineering features. It also helps in creating deep learning models using TensorFlow’s advanced algorithms and performance architecture. This connection between the model and the data lake will allow scalable runtime execution of TensorFlow models with Datameer jobs, and easy integration of deep learning results to downstream applications and tools. The addition of SmartAI into Datameer’s Hadoop-native data preparation and analytics platform means all of Datameer’s security, governance, monitoring and other mission-critical operational requirements apply, satisfying stringent IT requirements and finally allowing deep learning to be applied to enterprise data lake environments. The company states that whether deployed on-premises or in the cloud, private data is kept well secured and governed in Datameer as always, while it is leveraged for smarter, deep learning-driven predictive analytics.

How will it help?

Datameer claims that SmartAI will allows enterprises to democratize data science, taking the deep learning work of data scientists from the lab to the business in production-ready scenarios that meet the big data security, governance and management standards IT requires. This would eventually help business analysts to apply and execute trusted deep learning models against massive datasets from enterprise data lakes to drive better business outcomes. AI-enriched insights can be executed at scale on big data, business analysts can easily plug AI models into their analytic workflows and proper enterprise-grade security and governance can be applied. Analyst can also access trusted deep learning models within the Datameer function library for easy one-click application of the model during analysis and integrate models with scale, performance and governance using direct operationalization of TensorFlow models into Datameer data pipelines. “Today, we’re only seeing the tip of the iceberg in terms of what can be accomplished in the world of deep learning and artificial intelligence,” said Peter Voss, CTO, Datameer.

“AI is only as good as the data that feeds it. We’re thrilled to connect the dots by allowing enterprises to bring together massive amounts of disparate data, prepare and design the data pipeline, and now ultimately feed the data into models that have the potential to radically optimize business models.”

The combination of Datameer and TensorFlow claims to also give enterprises a faster end-to-end AI analytic cycle to speed the time to insight. According to the company, the client can shape and organize big data faster with rich data preparation features while locking down the data with extensive security features. It can also rapidly create deep learning models with as little as a few lines of code using TensorFlow and train them quickly using its performance architecture. It can add TensorFlow models to the Datameer function library and into the organization’s Datameer workbooks via the TensorFlow plug-in.