Skip to content

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

DataRobot workflow overview

To build models in DataRobot, you first create a project by importing a dataset, selecting a target feature, and clicking Start to begin the modeling process. A DataRobot project contains all of the models built with the imported dataset.

Begin this process by logging in to DataRobot.


If your organization is using an external account management system for single sign-on:

  • If using LDAP, note that your username is not necessarily your registered email address. Contact your DataRobot administrator to obtain your username, if necessary.
  • If using a SAML-based system, on the login page, ignore the entry box for credentials. Instead, click Single Sign-On and enter credentials on the resulting page.

Model data with DataRobot

The following steps provide a quick overview of how to begin modeling data with DataRobot. Links within the steps point to the full documentation if you need assistance.


See the file type reference for information about file size limitations.

  1. Create a new DataRobot project by importing a dataset using any one of the methods on the new project page:

    DataRobot displays a progress indicator while the file is processing:

  2. To begin modeling, type the name of the target and configure the optional settings described below:

    Element Description
    What would you like to predict? Type the name of the target feature (the column in the dataset you would like to predict) or click Use as target next to the name in the feature list below.
    No target? Click to build an unsupervised model.
    Secondary datasets Optionally, add a secondary dataset by clicking + Add datasets. DateRobot performs Feature Discovery and creates relationships to the datasets.
    Feature list Displays the feature list to be used for training models.
    Optimization Metric Optionally, select an optimization metric to score models. DataRobot automatically selects a metric based on the target feature you select and the type of modeling project (i.e., regression, classification, multiclass, unsupervised, etc.).
    Show advanced options Specify modeling options such as partitioning, bias and fairness, and optimization metric (click Additional).
    Time-Aware Modeling Build time-aware models based on time features.
  3. Scroll towards the bottom of the screen to see the list of available features. Optionally, select a Feature List to be used for model training. Click View info in the Data Quality Assessment area on the right to investigate the quality of features.

  4. After specifying the target feature, select a Modeling Mode. (For large datasets, see the section on early target selection.) Click Start to begin modeling:

    DataRobot prepares the project (EDA2) and starts running models. A progress indicator for running models is displayed in the Worker Queue on the right of the screen. Depending on the size of the dataset, it may take several minutes to complete the modeling process.

    The results of the modeling process are displayed in the model Leaderboard, with the best performing models (based on the chosen optimization metric) being ranked at the top of the list.

  5. Click a model to display the model blueprint and access the many tabs available for investigating model information and insights.

  6. You can test and generate predictions from any model manually without deploying to production via Predict > Make Predictions. Provide a dataset by drag-and-dropping a file onto the screen or use a method from the dropdown. Once data upload completes, click Compute Predictions to generate predictions for the new dataset and Download, when complete, to view the results in a CSV file.

  7. Once you have selected a model to use for generating predictions, navigate to the Predict > Deploy tab. If it is not the recommended model, DataRobot recommends preparing the model for deployment. If it is the recommended model, click Deploy Model.

  8. Complete the required information on the deployment information page. Once complete, click Deploy model.

  9. You can then monitor model health and prediction statistics from the Deployments inventory.

Updated February 1, 2023
Back to top