Automatic Data Visualizations in Driverless AI
- Predictive AI
- H2O Driverless AI
This lesson walks through the Autoviz tab in Driverless AI, which automatically prepares nine visualizations for a dataset, including correlated scatterplots, skewed histograms, disparate boxplots, an outliers view, a correlation graph, a parallel coordinates plot, a radar plot, and a data heatmap. It shows how each visualization row tracks its status, creation date, and processing time, and how a completed set can be searched, opened, or deleted like the Datasets tab. The lesson closes by adding a custom histogram of annual income across the applicant dataset, showing most applications come from incomes below 40 to 50 thousand dollars.
Key takeaways
- What nine automatic visualizations Autoviz generates for a dataset, from scatterplots to a data heatmap
- How to add a custom graph, such as a histogram, and choose a transformation or bar count for it
- Why the type of automatically generated graphs can vary depending on the feature types in a dataset
- How the annual income histogram shows most applicants earn below roughly 40 to 50 thousand dollars
Start owning your AI