Notebooks and examples on how to onboard and use various features of Amazon Personalize
Open the getting_started/ folder to find a CloudFormation template that will deploy all the resources you need to build your first campaign with Amazon Personalize. The notebooks provided can also serve as a template to building your own models with your own data.
This repository is cloned into the environment so you can explore the more advanced notebooks with this approach as well.
If you just want a simple walkthrough to explore later you can execute personalize_sample_notebook.ipynb, it works well inside the same Jupyter environments.
Open the advanced_examples/ folder to see detailed descriptions of the following typical use cases.
- Collaborative filtering based on user-item interaction tables. The intuition behind is that similar users like similar items.
- Hybrid recommendation also considering user, item, and event meta-data. The result is to extrapolate to out-of-sample users and items, based on their meta-data features.
Open the diagnose/ folder to have a visualization of the key properties of your input datasets. The key components we look out for include: missing data, duplicated events, and repeated item consumptions; power-law distribution of categorical fields; temporal drift analysis for cold-start applicability; and an analysis on user-session distribution.
This is a project to showcase how to quickly deploy a Personalize Campaign in a fully automated fashion using AWS Step Functions.
To get started navigate to the ml_ops folder and follow the README instructions.
This sample code is made available under a modified MIT license. See the LICENSE file.