Skip to main content
Version: 0.21.2

Prepare Data

Before you begin, ensure your data is as ready as possible to make it easier for a model to identify meaningful trends. We recommend you use the data cleaning tools in DataChat to make sure your data satisfies our best practices before attempting to train a model.

The Visualize skill helps you get a picture of your data. It uses a generative chart builder algorithm based on the columns in your dataset to create charts that illustrate various trends in your data.

To use Visualize, load data into your session and select ML > Visualize in the sidebar.

ML Visualize button

The Visualize form appears.

ML Visualize form

  1. Select at least one key performance indicator (KPI) as a dependent column. Optionally, select another KPI.
  2. Optionally, select at least one variable as the independent column to use to visualize the dependent column. Optionally, select up to two other variables.
  3. Optionally, click the with ML recommendations checkbox to generate machine learning recommended features. When this option is enabled:
    • If you didn't specify any independent columns in step 2, DataChat uses machine learning to identify which features have the most impact on your chosen KPI. Then, Visualize creates several charts using the most impactful features as independent columns, in order from most impactful to least impactful.
    • If you did specify independent columns in step 2, DataChat uses machine learning to identify which features have the most impact on your chosen KPI along with your specified independent columns to create two groups of charts:
      • Charts based on your KPI and the independent variables you chose in step 2.
      • Charts based on the most impactful features identified by the machine learning process, in order from most impactful to least impactful.
  4. Click Submit.

One of the charts appears as a popup and, once closed, moves to the chart panel. If other charts are available, click on the links in the chat box to display the chart. The chat box also shows how the chart was generated.

You can also build a DataChat sentence in the chat box:

Visualize <dependent column>

See Visualize for more options and information.

Give us feedback