Foundations Atlas Team Edition (Azure)¶

Atlas

Overview¶

Atlas Team Edition (Azure), part of Dessa's Foundations suite on Azure cloud, is a machine learning development toolkit that helps data scientists and machine learning engineers to more effectively and efficiently perform model development as part of their machine learning development workflow.

Core features of Atlas Team Edition (Azure) include:

Experiment management:
Automatically record and track model experiments including hyperparameters, metrics, code, and artifacts (e.g. images, GIFs, audio clips) in the Atlas GUI.

GUI

Job scheduling:
Launch and run hundreds to thousands of model experiments at once to fully utilize compute resources.

GUI

Collaboration:
Invite team members and stakeholders to collaborate and contribute thoughts, ideas, and comments on various aspects of each project.

Reproducibility:
Maintain a centralized view of every project and model experiment and audit trails with complete metadata, code and saved items.

Key Components¶

Atlas consists of the following core modules:

SDK & CLI - A set of programmatic and command-line interface protocols for implementing Atlas features.
GUI - A Dockerized web application for users to view projects and model experiments.
Scheduler - An orchestration mechanism that enables Atlas features such as experiment management and job scheduling.

See CLI and SDK reference documentation for details.

To learn more about Atlas for Enterprise, please reach out to us here.

License¶

Unauthorized copying, distribution, reproduction, publication, use of this library, via any medium is strictly prohibited. Proprietary and confidential.