Dashboard panel types
Overview of the available monitoring Panels.
What is a monitoring Panel?
A monitoring Panel is an individual plot or counter on the Monitoring Dashboard.
-
You can add multiple Panels and organize them by Tabs. You can customize Panel type, values shown, titles and legends.
-
When adding a Panel, you choose a Test or Metric with the specific value (“metric result”) inside it. Evidently pulls corresponding value(s) from all Reports in the Project and plots them on the Panel.
-
You can use Tags to filter data from specific Reports. For example, you can plot the accuracy of Model A and Model B next to each other. To achieve this, add relevant Tags to the Report.
How to add Panels. This page explains the Panel types. Check the next section on adding Panels.
Panel types. There are 3 main panel types:
-
Metric panels plot individual values from inside Reports.
-
Test panels show pass/fail Test outcomes in time.
-
Distribution panels plot distributions over time.
Metric Panels
Metric Panels (DashboardPanel) show individual values from inside the Reports in time.
For example, if you capture Data Summary Reports (include mean, max, min, etc., for each column) or Data Drift Reports (include the share of drifting columns and per-column drift score), you can plot any of these values in time.
Panel time resolution depends on Report frequency. For instance, if you log Data Drift Reports daily, you can plot the share of drifting features with daily granularity. You can also open the source Report to see feature distributions on a specific day.
Counter
Class DashboardPanelCounter.
Shows a value with supporting text or text alone. Perfect for dashboard titles.
Plot
Class DashboardPanelPlot. Shows individual values as bar, line, scatter plot, or histogram.
Line chart
PlotType.LINE shows values over time from multiple Reports.
Bar chart
PlotType.BAR shows values over time from multiple Report.
Scatter plot
PlotType.SCATTER shows values over time from multiple Reports.
Histogram
PlotType.HISTOGRAM shows the frequency of individual values across Reports.
Test Panels
Test Panels show the Test results.
As you run the same Tests repeatedly, you can visualize the pass/fail outcomes or result counts. You choose which Test results to include.
Test Panels only work with Test Suites: you must add Tests to the Metrics inside your Report to be able to render these panels.
Test counter
Class DashboardPanelTestSuiteCounter. Shows a counter of Tests with specified status.
Test plot
Class DashboardPanelTestSuite.
Detailed plot
TestSuitePanelType.DETAILED. Individual Test results are visible
Aggregated plot
TestSuitePanelType.AGGREGATE. Only the total number of Tests by status is visible.
Distribution Panel
Class DashboardPanelDistribution. Shows a distribution of values over time.
For example, if you capture Text Evals or Data Summary that include histograms for categorical values, you can plot how the frequency of categories changes.
Stacked
barmode="stack": stacked bar chart shows absolute counts in a single bar.
Grouped
barmode="group": grouped bar chart shows absolute counts in separate bars.
Overlay
barmode="overlay": overlay bar chart shows overlaying absolute counts.
Relative
barmode="relative": relative bar chart shows stacked relative frequency.
What is the difference between a Distribution panel and a Histogram? A histogram plot (DashboardPanelPlot withPlotType.HISTOGRAM) shows the distribution of the selected values from all Reports. Each source Report contains a single value (e.g., a “mean value”). A Distribution Panel (DashboardPanelDistribution) shows how a distribution changes over time. Each source Report contains a histogram (e.g. frequency of different categories).
What’s next?
How to add monitoring Panels and Tabs.