IBM launches open ML dataset hub curated for the enterprise

The IBM Data Asset eXchange provides free and open datasets for machine learning, with quality and licensing term checks that should make it simpler for businesses to use.

IBM on Tuesday launched the IBM Data Asset eXchange (DAX), a repository of free and open datasets for machine learning. While there are plenty of options online for finding open datasets, IBM says DAX is uniquely curated for the enterprise. 

executive guide

What is AI? Everything you need to know about Artificial Intelligence

A guide to artificial intelligence, from machine learning and general AI to neural networks.

Read More

DAX will only offer high-quality datasets with clearly-defined open data licenses in standardized formats, according to IBM. Will also provide metadata and supplementary content such as tutorials, making it easier for corporations to get started with the datasets. 

The repository will also provide unique access to various IBM and IBM Research datasets. For example, IBM is releasing the Finance Proposition Bank and Contracts Proposition Bank datasets. These datasets are part of an active research program aimed at improving the natural language understanding technologies underlying some IBM products like Watson Natural Language Understanding.

IBM designed the DAX repository to complement the IBM Model Asset eXchange (MAX), which helps data scientists and developers find free and open source machine learning and deep learning models.

The goal, IBM said in a blog post is "to make it straightforward to use DAX and MAX assets in conjunction with IBM AI products as well as other hybrid, multicloud AI tooling, both proprietary and open source."

In other open-source news, IBM on Tuesday launched three new open source projects — Kabanero, Appsody, and Codewind — that should make it easier for developers to build and deploy cloud-native apps for Kubernetes. The move lines up with IBM's focus on helping businesses adopt hybrid cloud strategies -- containers are increasingly a part of hybrid cloud deployments.

Appsody provides pre-configured stacks and templates for popular open source runtimes and frameworks, which developers can use to build applications for Kubernetes and Knative deployments.

Kabanero integrates popular runtimes and frameworks with a Kubernetes-native DevOps toolchain. It will incorporate Appsody stacks and templates into its overarching framework.

Meanwhile, Codewind is a project managed by the Eclipse Foundation that provides extensions to popular integrated development environments (IDEs) like VS Code, Eclipse and Eclipse Che. IBM made the first major contribution to Codewind.