Chart Builder
Deephaven makes it easy to visualize your data. You can plot data with code by using the built-in plotting API. You can also plot data directly in the Deephaven IDE user interface. This guide focuses on the Chart Builder feature, which creates line, bar, scatter, and pie charts from a table without writing any code. Charts you build from a ticking table update as new data arrives.
Open the Chart Builder
To open the Chart Builder, click the Table Options icon at the right of a table (circled in the screenshot below). Then select Chart Builder from the Table Options menu. Tree tables and rollup tables don't offer the Chart Builder.


The menu then shows the Chart Builder form, where you configure a chart of the table's data. Click Back to return to the menu.

Configure and create a chart
You can choose from four different chart types:
Each chart type has its own fields. Click in a field to see the table's column names.
- Line, Bar, and Scatter charts take an X-Axis column and one or more Series columns. The X-Axis column sets each point's horizontal position. Each Series column sets the values plotted against it, and appears as its own line, set of bars, or set of points.
- Pie charts take a Labels column, which names the slices, and a Values column, which sets each slice's size.
Each field starts with one of the table's columns already selected, so you only need to change the fields you want to be different.
To plot another column in a line, bar, or scatter chart, click Add Series. The new Series field starts with the table's first column selected, so change it to the column you want.
To remove a series, click the trash can icon to the right of its Series field. The icon appears only when the chart has more than one series.
To start over, click Reset. The Chart Builder returns to a line chart with its default columns and Sync State.
The Chart Builder plots the table as it currently appears, with any active filters, sorts, and custom columns applied. If the table has no filters, it plots all of the data.
The table you open the Chart Builder from is the parent table. Below the column fields, choose whether the chart follows later filters and other changes you make to the parent table:
- Sync State: When selected, the chart updates to match any filters or other user modifications made to the parent table.
- Freeze State: When selected, the chart ignores filters and other user modifications made to the parent table after you create it. If the parent table is a ticking table, the chart still updates when new data arrives.
Once you make your selections, click Create. The chart opens as a new panel in your workspace with a title built from the column names, such as Avg_Temp by Year. Each time you click Create, the Chart Builder opens another new panel.
Caution
You can't edit an existing chart configuration. To change a chart, create a new one in the Chart Builder.
Examples
The next sections show an example of each chart type, using two tables of Seattle weather data:
hiLoByYearhas one row for each year from 2000 through 2020. ItsAvg_Temp,Lo_Temp, andHi_Tempcolumns hold the average, lowest, and highest of Seattle's daily mean temperatures for that year. The line, bar, and scatter examples use this table.tempBandscounts the days in 2020 that fell into each temperature band. The pie example uses this table.
Run this script if you'd like to follow along:
Line, bar, and scatter
Open the Chart Builder from the hiLoByYear table to chart the average temperature for each year. Select Line, then click in the X-Axis field to see the table's column names:

Select Year for the X-Axis and Avg_Temp for the Series. Click Create to generate a single-series chart.

Let's make a multi-series bar chart. This time, select Bar, set the X-Axis to Year, and select Lo_Temp for the first Series field. Then click Add Series, select Hi_Temp, and click Create:


To make a scatter chart, select Scatter and set the X-Axis and Series fields the same way as in the line chart.
Pie
To chart the tempBands table, open the Chart Builder from that table, select Pie, select Band for Labels and Days for Values, and click Create:


Interact with charts
The following features work on any chart made with the Chart Builder or the built-in plotting API.
Plot hover menu
Hovering your cursor over a chart brings up the following menu:

- Download plot as a png: Save the plot locally as a PNG file.
- Zoom: Drag across the part of the chart you want to magnify. Double-click the chart to return it to its original view.
- Pan: Click and drag the chart to move other areas of it into view. Double-click the chart to return it to its original view.
- Zoom in: Zoom in on the chart.
- Zoom out: Zoom out on the chart.
- Autoscale: Rescale the axes so that all of the data fits in view.
- Reset axes: Return the axes to their original ranges.
For a large chart, Deephaven may downsample the data, plotting fewer points than the table holds so that the chart stays responsive. The menu then adds buttons for this:
- Downsampling status: Click to turn downsampling off or back on.
- Downsampling failed or Error: A warning button that appears when downsampling fails or the chart reports an error. Click it to see the error message.
While downsampling is in progress, or while a warning is showing, the menu stays visible without hovering.
Tooltips
If you hover your cursor over the data shown in a chart, a tooltip shows the exact value of the data under the cursor.

Toggle series visibility
When a chart plots multiple series, you might want to temporarily hide one or more of them. A multi-series chart made with the built-in plotting API shows a legend. Click a series name in the legend to hide that series. Click it again to show it.
Charts made with the Chart Builder don't show a legend.

Related documentation
- Deephaven's table types
- Real-time plots
- Built-in plotting API
- Navigate the GUI
- Filter without writing queries
- Use table tools in the UI
- Create a hierarchical tree table
- Create a hierarchical rollup table programmatically
- Read CSV or other delimited files
- Multi-aggregation
readCsvviewyeartimeZonewhereupdatecountByaggByAggAvgAggMinAggMaxformatColumnsTableJavadocAggregationJavadoc