🧐 Openlayer TL;DR

It is currently impossible to have the foresight required to uncover every bug, bias, and inconsistency in your model before you ship it to customers. The process of iterating on training and testing to find these bugs manually is also time-consuming because models are often treated as black boxes. It's hard to get concrete, actionable insights without a lot of guesswork and head-scratching.

Openlayer is the debugging workspace for machine learning (ML), where companies are able to track and version models, uncover errors, and make informed decisions on data collection and model re-training. We make collaborative error analysis easy and intuitive.


Reach out

Openlayer currently only supports models that work with tabular or textual data. If you work with another data type, please reach out so we can prioritize accommodating your business needs.

How the documentation is organized

There is a lot you can do with Openlayer. This high-level overview of the documentation will help you find the information you need.

  • Tutorials carefully guide you through a small project that will introduce you to the most important aspects of the platform. This is the place to start if it’s your first time with us;
  • How-to guides present the step-by-step of how to use our features. They are more advanced than tutorials and are organized by overarching theme;
  • The API reference contains the technical description of our API and references on how to operate it. This is where you should go if you want to know more about how to call our API when uploading models and datasets to the platform;
  • Topic guides clarify particular topics from a higher level. These guides are understanding-oriented and serve the purpose of expanding the documentation’s coverage on a certain topic.

If you are interested in learning more about error analysis and Openlayer in general, feel free to check out our white paper and our blog.

Did we miss something?

In the quest toward building the future of ML, we are constantly shipping new features and perfecting Openlayer. Therefore, we apologize in advance if something is missing from the documentation.

If this is the case, please, do not hesitate to contact us. We will do our best to update everything as soon as possible.