DeepSea Overview

DeepSea is a data science platform for creating interactive dashboards, which can be connected to python functions and/or a variety of sources of data.

With DeepSea, you can:

  • create interactive dashboards that end users can collaborate with you on in a no-code interface
  • share your dashboards privately with users in your organization, or share a link with anyone
  • allow viewers to run python functions and see the resulting data in the dashboard immediately
  • create forms to collect survey data that connects to dashboards or models
  • easily update & maintain your dashboards, data, and code
  • bring a data science project to production within a couple hours, no DevOps or front-end development required

How can I get access to DeepSea?

Once you/your organization has signed up for DeepSea & an admin has set up your account:

You should be receiving an email from us to confirm your email address and activate your account. Once you click the link in this email, you can set up your account by entering the temporary password.

If you haven’t received an email, please contact the person in your organization in charge of DeepSea access, or contact our support email.

If you did receive an email but it has been over 48 hours, you may need to follow the same contact steps as above because the activation link you received is no longer valid.

If you/your organization has not signed up for DeepSea:

Please get in touch with us to discuss setting up a trial for you or your team.

Navigating DeepSea

Projects

After you log in, you will be brought to the projects page. All assets in DeepSea belong to projects. Each project is shared with one or more users, allowing you to organize your work. Projects are like folders or workspaces for DeepSea content. From this page, you can create a new project or navigate into an existing one by clicking on its name.

Projects can contain a variety of DeepSea content. Once inside a project, there are four main sections of DeepSea available: Dashboards, Datasets, Models, and Forms.

Dashboards

Dashboards are the end product of most work in DeepSea, and the first place to look when exploring an existing project. A dashboard typically contains several charts, tables, or numbers (KPIs), along with a few other types of cards. From this section, you can view & edit existing dashboards or create new ones.

Datasets

Datasets are the source of any data in DeepSea, and are always in the form of a table. They will also be linked to models or forms if that’s the source of the data. From this section, you can create new datasets from a CSV file, or view & edit your existing datasets. 

The pages for a dataset show you a preview of your data, and information about each column, which we call features. Each feature has a name and a type: the type determines how DeepSea will format and visualize the data. 

Models

Models are interfaces for python functions that DeepSea users can run. Each model has a custom interface that you set up to show your users. DeepSea takes in the user input, runs the function, and stores the output data. This allows end users to run your code and see the results without interacting with python, in a way that is tailored to their needs. 

Each model is created from source code in a package, which is one or more python files uploaded to DeepSea in a folder. From the main page in this section, you can see all existing packages and models, click on them to view & edit the contents of each one, and create a new package or model.

Forms

Forms are surveys that allow you to collect data from people that can then be used in DeepSea or exported elsewhere. From this section, you can view & edit existing forms, or create a new form of your own.

Navigation Bar

To view another project, click the home button in the top left corner to return to the projects overview page.

There is also an account settings menu at the top right of the screen, where you can view your account settings or log out. On the account settings page, you can change your password, name, or profile picture. 

If you are an organization administrator, you will also be able to see some additional options for managing your organization’s accounts.

Up Next:

2. Interact With an Existing Dashboard

Learn how to find dashboards, filter data, and run models.