日本語
DataRobot's Visual AI allows users to leverage images in their models just like any other type of data. Dive in to go from Databricks image data to deployed classifier!
About this Accelerator
While there are other methods of ingesting image data into DataRobot, in this notebook you will encode the image data directly into the data frame using base64 encoding.
This methodology allows you to keep all of the relevant data in a single data frame, and works well for a Databricks environment.
This technique also extends widely to a wide variety of multimodal datasets.
What you will learn
- Import image files using Spark
- Prepare them into a data frame suitable for ingest into DataRobot
- Leverage DataRobot through code to rapidly train and deploy a powerful multiclass image classifier
