Concept | Charts#

Watch the video

Visualization is a key tool in the data exploration and discovery process. To meet this need, the Charts tab of a Dataiku dataset houses a drag-and-drop interface for visual exploration. Many different types of charts are natively available. Bar charts, line graphs, pivot tables, and scatter plots are just a few examples.

A Dataiku screenshot showing a visualization created in the Charts tab of a dataset.

Building a chart#

To get started, all you have to do is select a chart type, and drag variables from the Data panel on the left onto the desired axis.

A Dataiku screenshot showing the first steps to create a chart.

Customizing a chart#

Once you’ve dragged and dropped the relevant variables to build your chart, you can customize it in many ways.

Configuring the variables#

The interface lets you configure the chart variables. For example, you can:

  • Select the correct aggregation to display.

    A Dataiku screenshot showing how to edit the aggregation.
  • Adjust the sorting of values in the chart.

    A Dataiku screenshot showing how to change the sorting of values.
  • Change the value and label formatting of the variables. These changes affect the tooltips and the legend.

    A Dataiku screenshot showing how to edit a variable label.

The Chart builder has many other features to assist in the exploration of your data. For example, with time series, you can zoom in on different periods, change the aggregated date interval, explore multiple series within the same chart, examine them side-by-side in subcharts, or create basic animations.

Filtering the data#

When working with large numbers of groups of categorical data, you can easily control the number of displayed values by selecting them for filtering. Also, to handle less-prevalent categories, you can group them into an Others bucket.

You can also drill down into a dataset by clicking the filter icon from a tooltip.

This will open the filter’s dropdown where you can apply changes.

A Dataiku screenshot showing how to filter charts.

Configuring the sampling and execution engine#

The Sampling & Engine panel on the left allows you to configure the sampling and execution engine used in charts.

A Dataiku screenshot showing the Sampling & Engine panel.

While the default execution engine is set to the DSS engine, you can select another engine to improve performance. Indeed, certain data sources that support SQL queries can be executed in database for greater speed.

Regarding the sampling, by default, charts in Dataiku use the same sample as the one found in the Explore tab (the Use same sample as explore option is checked). However, you can change it by deselecting the option and adjusting your sampling method.

What’s next?#

You just learned how to use the Charts tab of a Dataiku dataset for visualization. Get more practice with this and other features of Dataiku by visiting the Explore the data article.