Modeling¶

The Modeling section provides information to help you easily navigate the modeling process. Modeling begins after you have loaded, transformed, performed feature discovery, and analyzed your data.

Workflow overview¶

The UI-based Workflow overview: AutoML provides a quick overview of steps from data ingest to model deployment. Each step links back to the full documentation for more complete detail.

Build models¶

Topic	Describes...
Advanced options	Setting advanced modeling parameters prior to building.
Build models	Elements of the basic modeling workflow.
Manage projects	Managing models and projects and export data.

Model insights (Leaderboard tabs)¶

Topic	Describes...
Evaluate tabs	Viewing key plots and statistics needed to judge and interpret a model’s effectiveness.
Understand tabs	Explaining what drives a model’s predictions.
Describe tabs	Viewing model building information and feature details.
Predict tabs	Making predictions in DataRobot using the UI or API.
Compliance tabs	Compiling model development documentation that can be used for regulatory validation.
Comments tab	Adding comments or host a discussion around any item in the AI Catalog.
Bias and Fairness tabs	Identifying if a model is biased and why the model is learning bias from the training data.
Other model tabs	Comparing models across a project.

Specialized workflows¶

Topic	Describes...
Date/time partitioning	Building models with time-relevant data (not time series).
Unsupervised learning	Working with unlabeled or partially labeled data and detect abnormalities in your dataset.
Visual AI	Using image-based datasets.
Location AI	Using geospatial datasets.

Time series modeling¶

Topic	Describes...
Workflow overview	Basic time series modeling workflow.
Date/time partitioning	Building models with time-relevant data.
Time series modeling	Building time series forecasting models.
Multiseries modeling	Building models with datasets that contain multiple time series based on a common set of input features.
Nowcasting	Predicting the current value of a target based on past and present.
External prediction comparison	Using the output of predictions made by non-DataRobot models as a baseline to compare against DataRobot predictions.
Modeling dataset	The modeling dataset that is a result of the feature derivation process.
Time series reference	Provides deep-dive reference material for DataRobot time series modeling.

Business operations¶

Topic	Describes...
AI App Builder	Building and configuring AI-powered applications using a no-code interface.
AI applications (deprecated)	Enabling core services without building models and evaluating performance in DataRobot.
Partner applications	DataRobot applications developed for partner integrations.
Use case tracking	Measuring success by defining expected, and tracking actual, model value in real time.

Modeling reference¶

Topic	Describes...
Data and sharing	Partitioning, feature derivation, and sharing roles.
Modeling details	The Leaderboard and the processes that drive model building.
Eureqa advanced tuning	Tuning Eureqa models by modifying building blocks, customizing the target expression, and modifying other model parameters.