News

Dataiku Update Boosts Containerization, Deep Learning and More

• By David Ramel
• September 12, 2018

Data science and advanced analytics specialist Dataiku announced version 5 of its flagship platform, with new capabilities to enable scalable enterprise AI.

The company highlighted three main new features in its release of Dataiku 5: expanded containerization capabilities; a new deep learning library; and more comprehensive documentation.

Dataiku introduced Docker and Kubernetes containerization functionality in a previous release, and in v5 have taken it further, enabling in-memory processing of Python and R recipes for optimized resource usage.

"When individuals need to run in-memory jobs, Dataiku will create a Docker image containing the code and required libraries or packages, and automatically deploy it to a Kubernetes cluster," said the company's Pauline Brown in a Sept. 12 blog post. "Kubernetes will then take care of job orchestration and execution where cluster resources are available, thus bringing computation elasticity (and thus saved resources) and power to both the users ... and to the machines."

On the deep learning front, Keras, an open source neural network library written in Python, has been added to the platform's libraries and technologies. Also, the TensorFlow library has been added to the platform's Visual AutoML Interface, which reportedly allows both beginner and expert users to more easily leverage deep learning.

"Users can now define the architecture of their deep learning models directly from the Visual Machine Learning Interface," Brown said. "From there, Dataiku will automatically handle data preprocessing, and model training, deployment, versioning, rollback, and monitoring."

Platform documentation received a boost with the addition of group discussions, wikis and personalized project home pages, which the company said takes collaboration from a team-centric focus to a large-scale, organization-wide approach.

More information on the above and many other new features can be found in the release notes.