Skip to main content

Sessions Overview

In this section, we'll cover how to get oriented with a DataChat session.

Launch a Session​

All sessions in DataChat start at the homepage. To start a session, go to the homepage and select New > Session. By default, you can have a maximum of five sessions open at a time.

Name a Session​

You can name your session to make them easier to find and to differentiate on the homepage. By default, new sessions are titled "Unnamed". Double-click on the name of the session at the top left of the screen and enter a new name:

Name session

DataChat sessions have three main components:

tabs

Data Tab​

The Data tab displays all of datasets loaded into the session. From here, you can:

Data tab

Select a Dataset​

While working with your data, you might generate different versions of modified datasets. By default, DataChat uses the most recently loaded or versioned dataset. However there are a few ways to switch which dataset set you're using:

  • You can filter for a dataset by hovering over the Data tab and simply entering the parameters to filter for. Note that the filters only apply to dataset names.

    dataset filter

  • From the Data tab, the version list is an easy way for you to reference different versions of your work on a given dataset. You can select a versioned dataset by clicking version number > Versions.

    dataset versions

    note

    Datasets with a lock next to them are read-only. Performing actions on them first copies the selected dataset to "[Dataset] v2" and then performs the action on that dataset.

Rename a Dataset​

Renaming your datasets can be a useful tool to make navigating your work much easier. To rename a dataset:

  1. Hover the Data tab button.
  2. From the list of datasets, click the more options menu of the dataset you'd like to rename.
  3. Click Duplicate.
  4. Enter your new name for the duplicated dataset.

Optionally, you can also click the dataset's version number > Duplicate.

rename data

This creates a copy of the dataset with the new name. Optionally, you can remove the original dataset.

Hide Datasets​

To hide a dataset from focus:

  1. Hover the Data tab button to view your datasets.
  2. Click the checkbox of the menu the dataset you'd like to hide. To unhide the dataset, simply click the checkbox again.

Optionally, you can also click the dataset's version number > Hide.

hide data

Export Datasets​

note

Exporting large datasets are limited to 10 million cells.

To export a table from the Data tab as a .csv file, click Export table in the top right of the table header.

export table

Remove Datasets​

To delete a loaded dataset:

  1. Hover the Data tab button.
  2. From the list of datasets, click the more options menu of the dataset you'd like to delete.
  3. Click Delete.

Optionally, you can also click the dataset's version number > Delete.

remove data

Chart Tab​

The Chart tab allows you to create charts with the Chart Builder, and displays other charts created within a session.

Chart tab

Export Charts​

Charts provide a couple of options to export through the chart's More options menu:

  • Export slide or all slides (for charts with sliders)
  • Export a chart or chart's data

Refer to Use Charts for more information.

Data Assistant Tab​

The Data Assistant tab offers a simple, chat-based interface where you can enter freeform questions about your data and leverage generative AI to produce insights.

DA tab