Skip to content

Click in-app to access the full platform documentation for your version of DataRobot.

Use applications

On the Applications tab, click Open next to the application you want to launch—from here you can test different application configurations before sharing it with users.

Note

End-users must sign in with a DataRobot account or access the application via a link that can be shared with users outside of DataRobot.

  Widget Description
Application name Displays the application name. Click to return to the app's Home page.
Pages Navigates between application pages.
Build Allows you to edit the application.
Share Share the application with users, groups, or organizations within DataRobot.
Add new row Opens the Create Prediction page, where you can make single record predictions.
Add Data Upload batch predictions—from the AI Catalog or a local file.
All rows Displays a history of predictions. Select a row to view prediction results for that entry.

Batch predictions

To make multiple prediction requests at once from the Home page, click choose file or drag the files into the box.

After adding new files, the application processes your predictions and displays them in the All Rows widget on the Home page. Click on any record to view the prediction results.

Single record predictions

To make a new prediction:

  1. Click Add new row, bringing you to the Create Prediction page with the Add New Row widget, which displays the features available to make a prediction.
  2. Fill in the feature fields—the only field required for making predictions is the association ID if one was added for the deployment. If a field is left blank, the feature field displays N/A on the prediction results page. Alternatively, you can click Populate averages to fill in the fields with the average value for each feature.

    Selecting location features

    If the dataset contains a location feature, a globe icon appears in the feature field. You can manually enter a feature value in the field, or click the globe icon to view a visual representation of the training data.

    The geometry type of the location feature determines the appearance of the training data on the map and affects which draw tool—Point, Polygon, or Path—you can use to highlight your prediction. In the example below, the location feature uses point geometry, so use the Point tool to add a new point to the map. With the point selected, click Save selected location; the point is then converted to a geojson string to make your prediction.

Click Add. After DataRobot completes the request, the prediction results page opens.

To add or remove feature fields, click Build and navigate to the Create Prediction page.

Prediction results

The prediction results page displays prediction information and insights based on the values entered for an individual prediction. This page automatically opens after making a single record prediction, but you can also view prediction results by selecting a row in the All rows widget.

The General Information widget displays helpful values for features that don't necessarily impact the prediction results, and the Prediction Information widget displays the values for features likely to impact the prediction, as well as Prediction Explanations.

Add or remove fields

To add or remove feature fields, click Build and navigate to the Prediction Details page. Once there, select the widget and click Manage in the Data tab. For more information, see the instructions for configuring chart widgets.

Prediction Explanations

The Prediction Explanations widget displays a chart with your predictions results, as well as a table with Prediction Explanations—a qualitative measure of how features in the model impact a prediction based on their relationship to the target.

Element
The predicted value, based on the values represented as a percentage.
The basis of the prediction classification, determined by the dataset the app deployed from.
A table displaying the top 10 Prediction Explanations as well as a visualization representing how the feature is distributed in the training data.

Note

You must compute Prediction Explanations for the model before making a prediction.

Prediction Explanation table

The Prediction Explanation table displays the top 10 features with the largest impact on a prediction based on their relationship in the training data.

See the table below for a description of each field:

Field Description
Impact The measured impact of a given feature on the prediction. For a description of the icons used, see "qualitativeStrength" indicator.
Feature The feature in the training data impacting the prediction.
Value The feature value causing the feature's measured impact on the prediction.
Distribution A histogram that represents the distribution of a given feature in the training data. Hover over the visualization for additional information that can then be used to add context to the Prediction Explanation.

What-if and Optimizer

The What-if and Optimizer widget displays various scenarios in a chart or a table view. To learn how to configure this widget, see the documentation on building applications

  Description
Opens the Add scenario pane.
Displays only actual, optimal, and all manually added scenarios.
Displays prediction insights in chart or table view.
Adds new scenarios to the widget.
Selects features for the x- and y-axis.

Tip

Check Show only manually added, optimal, and actual scenarios to more easily find your simulations.

Create simulations

If the What-if functionality is enabled, you can manually add scenarios to the widget's display.

To create a simulation, click Add scenario.

Provide values for each variable selected on the Build page. When you have entered values for each feature, click Add. This triggers a prediction request to a DataRobot prediction server and returns a predicted value for your target feature.

Selecting a point on the chart (data point) opens a pane on the right that displays prediction results and feature values for the selected scenario. If Prediction Explanations are available for your prediction request, they appear in the Prediction Explanations tab in the same pane.

You can continue to make predictions by updating the variable inputs with new values and repeating this process. After making several predictions, click Table view. This view allows you to drag-and-drop scenarios for side-by-side comparison.

Optimization simulation results

If the Optimizer functionality is enabled, the chart displays an optimal scenario, as well as different outcomes based on the flexible features added to the widget's configuration.

Once DataRobot completes running simulations, the chart populates the results. The y-axis measures the values of the target feature, and the x-axis indicates the simulation iteration. Each point on the graph represents the predicted value for each simulation run.

The orange data point represents your prediction and the green data point represents the optimal scenario—the values of the feature that most often produce the optimal result for your target (the minimal or maximal, based on your settings). The selected values are those you selected for a given iteration, allowing you to compare the selected values for each iteration to the overall most optimal values determined by the app.

Time Series Forecasting widget

When a time series deployment is used to create a Predictor application, DataRobot automatically configures the Time Series Forecasting widget to surface insights relevant to time series use cases, including charts for Prediction vs Actual and Prediction Explanations over time.

Note that there are a few key differences between time series and non-time series Predictor applications:

  • The default configuration for time series applications only includes a Home page and the Add Data widget for batch predictions.
  • You cannot add or remove features from the Times Series widget, you can only modify its appearance, including the chart name and line colors, in the Properties tab.
  • Non-time series applications cannot add a Time Series Forecasting widget in Build mode.

Score predictions

Initially, the Time Series Forecasting widget is empty unless the deployment has already scored predictions. To score new predictions:

  1. Drag-and-drop a prediction file into the Add Data widget. A prediction line appears on both charts and Prediction Explanations are calculated (Predicted vs Actual chart shown here).

    In this example, the file contains sales predictions for 6/14/2014 to 6/20/2014.

    Association IDs in prediction files

    If the association ID is not configured properly in the prediction file, the Time Series Forecasting widget will not display prediction information. Consider the following before uploading a prediction file:

    • For single series projects, the association ID entered for the deployment must match the name of a dataset column containing dates.
    • For multiseries projects, the association ID entered for the deployment must match the name of a "combined" dataset column—a column with values that are a combination of the series name and date, for example, Boston_2014_09_12. The "combined" column is only required in prediction files, not the training dataset.
  2. Click Deployments and select your time series deployment in the Deployment inventory.

  3. Click Settings > Data. Scroll down to Actuals and upload the dataset containing actuals. In this example, the file contains actuals for 6/14/2014 to 6/20/2014. The actuals file must contain the association ID, which you can also find on the Settings page.

    Note

    If the forecast file contains actuals for the range of the initial prediction file, you do not need to upload actuals to the deployment and can proceed to step 5.

    In this example, the forecast file would need to contain actuals for 6/14/2014 to 6/20/2014, the range of the initial prediction, in addition to predictions for 6/21/2014 to 6/27/2014.

    The application displays an Actuals line and calculates Residuals—the difference between the prediction and the actuals for a given range—in the Time Series Forecasting widget.

  4. Navigate back to your time series application.

  5. Drag-and-drop a second prediction file, or forecast file, into the Add Data widget. A forecast line appears on both charts. This forecast file contains predictions for 6/21/2014 to 6/27/2014.

Predicted vs Actual chart

Similar to Accuracy Over Time, the Predicted vs Actual chart helps you visualize how predictions change over time using predicted and actual vs time values based on forecast distances.

Setting Description
Filter data Hide target (actual) and residual information from the chart. Prediction information cannot be hidden.
Resolution View the results by day, week, month, quarter, or year.
Prediction line Represents scored predictions.
Actual line Represents actual values for the target.
Forecast line Represents the prediction based on time, into the future using recent inputs to to predict future values.
Residuals Represents the difference between prediction and actual values.
Date range Drag the handles on the preview panel to bring specific areas into focus on the main chart.

Hover over any point to view the date, prediction values, and top 3 Prediction Explanations.

Prediction Explanations chart

To view Prediction Explanations over time for your scored predictions, click the Prediction Explanations tab. Every point on the chart represents a separate prediction, and therefore has its own set of Prediction Explanations. Every Prediction Explanation also has its own unique color, allowing you to explore trends in the data.

Setting Description
Fade explanations Allows you to hide either all positive or negative Prediction Explanations. Select the box next to Fade explanations and select an option from the dropdown.
Highlight explanations Highlights specific features in the Prediction Explanations based on its unique color. Click Highlight explanations and select features from the dropdown.
Resolution View the results by day, week, month, quarter, or year.
Enable segment analysis Creates the specified number of additional rows for the forecast value. Select Enable segment analysis and choose a Forecast distance from the dropdown.
Prediction Explanations Each point represents a separate prediction and each prediction has its own set of explanations, which are grouped by color.
Prediction line Represents scored predictions.
Forecast line Represents the prediction based on time, into the future using recent inputs to to predict future values.
Date range Drag the handles on the preview panel to bring specific areas into focus on the main chart.

Updated July 15, 2022
Back to top