Work with data (Classic)¶

DataRobot knows that high-quality data is integral to the ML workflow—from importing and cleaning data to transforming and engineering features, from scoring with prediction datasets to deploying on a prediction server—data is critical. DataRobot provides tools to help you seamlessly and securely interact with your data.

Import data from various sources, including from external data sources to minimize data movement and control data governance across your cloud data warehouses and lakes.

Explore patterns and insights in your data; automate the discovery, testing, and creation of hundreds of valuable new features.

1: Import data¶

Import data into the DataRobot platform from the AI Catalog, directly from a connected data source, or as a local file.

2: Explore data¶

After importing your data, DataRobot performs exploratory data analysis, a process that analyzes the datasets, summarizes their main characteristics, and automatically creates feature transformations —the results of which are displayed on the Data page of your project.

Once EDA1 completes, you can use the Data Quality Assessment to find and address quality issues surfaced in your dataset.

3: Prepare data¶

Now that you've explored your dataset and identified areas for improvement, you can:

Perform manual feature transformations.

Prepare your data using Spark SQL.

Add secondary datasets and then define those relationships to the primary in Feature Discovery projects.

Next steps¶

Now that your data is where it needs to be, you're ready to start modeling.