Tour the H2O Driverless AI Interface
- Predictive AI
- H2O Driverless AI
This lesson walks through a first tour of the H2O Driverless AI interface after connecting to an Aquarium instance. It covers the Datasets tab where imported data lives, the Autoviz tab for automatic visualizations, and the Experiments tab where models get trained, viewed, and organized into projects. It introduces the Diagnostics and Machine Learning Interpretability tabs for deeper model insights, the Recipes tab for uploading custom Python code, and the Deployments tab for managing models in production. It closes by pointing out the Resources tab's documentation and the option to work through Python or R clients instead of the UI.
Key takeaways
- What lives in the Datasets tab versus the Projects tab for organizing experiments and collaborating with peers
- How the Autoviz tab automatically generates visualizations based on a dataset's feature types
- Why the Diagnostics and MLI tabs matter for getting deeper insight into a trained model's predictions
- What the Recipes tab is for, letting you upload custom Python code as runtime plugins
- The Deployments tab is where trained models get managed once they move toward production
Start owning your AI