Foundations Atlas

Atlas

Platform Support Python Support Signups to download File issues Follow Dessa

What is Atlas?

Atlas was built by our machine learning engineers to dramatically reduce the model development time, from the experimentation workflow to production.

Here are some of the core features:

Experiment Management & Tracking:
Tag experiments and easily track hyperparameters, metrics, and artifacts such as images, GIFs, and audio clips in a web-based GUI to track the performance of your models

GUI

Job queuing & scheduling:
Launch and queue thousands of experiment variations to fully utilize your system resources

GUI

Collaboration & Bookkeeping:
Keep a journal of thoughts, ideas, and comments on projects

Reproducibility:
Maintain an audit trail of every single experiment you run, complete with code and any saved items

How does Atlas Work?

Atlas CE consists of the following core modules:

  • GUI - A Dockerized web application to view job status for various projects.
  • Foundations SDK & CLI - A programmatic and command-line interfaces for Atlas. CODE
  • Local Scheduler - A scheduler which is used for job orchestration and management.

You can also check out the CLI and the SDK reference documentation to find further information.

To learn more about the entirety of the Atlas platform, including Atlas Enterprise Edition, and how we can accommodate your needs, please reach out to us here.

License

Copyright (C) DeepLearning Financial Technologies Inc. - All Rights Reserved

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