Latest Research


This content is currently locked.

Your current Info-Tech Research Group subscription does not include access to this content. Contact your account representative to gain access to Premium SoftwareReviews.

Contact Your Representative
Or Call Us:
1-888-670-8889 (US/CAN) or
+1-519-432-3550 (International)

Comprehensive software reviews to make better IT decisions

Dessa Launches Atlas 2.0, a Foundations Suite of Tools for Building ML at Scale

Dessa, a Canadian artificial intelligence (AI) start-up previously known as Deeplearni.ng, announced today the availability of Atlas 2.0, one half of its Foundations suite of tools for building, deploying, and maintaining enterprise-grade machine learning (ML) models at scale.

Foundations Atlas is a suite of tools for building and deploying ML models. Its complement, Foundations Orbit, is a suite of tools for monitoring and maintaining ML models in production.

Atlas was developed to “transform the time-consuming and frustrating process of managing [ML] experiments into something that’s easy and even enjoyable, also allowing for insights you might not have otherwise,” says Alex Krizhevsky, the company’s principal ML architect. (He is one half of the duo of graduate students at the University of Toronto who, together with Geoffrey Hinton, cracked the ImageNet competition in 2012 and ushered in a new era of AI and ML with deep learning/neural networks.)

Specifically, Atlas addresses the following five ML frustrations:

  • Time spent on infrastructure-related work: Atlas sits on top of your infrastructure and has a multi-node, multi-GPU job orchestrator. You can run and schedule jobs remotely with one command.
  • Reproducibility of ad hoc experiments and speed to value: Only three lines of code are needed to record and reproduce your experiments with Atlas, alongside other features like tagging jobs, recording data artifacts, and more.
  • Compute costs: Auto-scaling of clusters and the ability to use pre-emptive/spot instances with Atlas's built-in retry mechanism allow you to save on GPU costs.
  • Lack of collaboration: Collaborate across your organization with Atlas Projects, project-specific discussion forums, multi-tenancy, and user access controls.
  • Lack of flexibility with existing tools that requires dramatic changes to workflow: Atlas was built to be augmentations of existing workflows as opposed to replacements.

Source: “Foundations Atlas 2.0 Is Now LIVE, & Also On Product Hunt!”

In addition to fixes and improvements, version 2.0 of Atlas includes two new features:

  • One can now launch jobs to a remote machine exactly as the jobs were submitted to a local scheduler.
  • A more intuitive object to allow for advanced hyperparameter searches (in beta).

Read more on the company’s Medium page.

Atlas 2.0 is available as a community edition on Product Hunt.

Visit our IT Cost Optimization Center
Over 100 analysts waiting to take your call right now: 1-519-432-3550 x2019