Welcome to the FeatureByte tutorials. These tutorials to learn our SDK at your own pace, in any order you wish.
Where to Start¶
If you are a data scientist who wants to jump straight to feature engineering, we recommend you start with the quick start feature engineering tutorial.
If you are a data engineer who wants to learn more about data modeling in FeatureByte, we recommend that you start with the deep dive data modeling tutorial.
If you’d like to learn the end-to-end workflow, we recommend you start with the quick start end-to-end tutorial.
Before You Start¶
Please initialize the playground environment before trying the tutorials.
This will start a local Spark service with preloaded datasets used in our tutorials, and create a feature store so you can dive into the tutorial right away.
See installation for more details.
The tutorial notebooks requires a helper script to work.
Download the script here and save it in the same directory as the tutorial notebooks.
Our quick-start tutorials take you directly to the tasks you want to learn:
- Quick Start Feature Engineering
- Quick Start Reusing Features
- Quick Start Feature Management
- Quick Start Model Training
- Quick Start End-to-End
Deep Dive Tutorials¶
Once you’ve mastered the basics, you will probably want to learn more about the power of FeatureByte:
The tutorials use the French grocery dataset, and we encourage you to use your creativity to engineer amazing new features on that dataset.
We have also provided another two built-in datasets for you to test your skills and be creative:
Download all the tutorials here.