My Insight Project
I wanted to build a personalized
product recommender
Online shopping is hard!
There are so many choices. How can we find what we want in a timely manner?
That's why I built Pair
Pair is an image-based product recommender system that is scalable and personalizd.
Inference Pipeline
Deployment
Architecture
1. Google Cloud Platforms (GCP)
A set of globally-distributed computers, hard drives, and virtual resources that provide compute, storage, big data and machine learning services.
2. Docker
A standard unit of software that packages up code and all its dependencies so the application runs quickly, reliably, and is completely independent of computing environment.
3. Google Kubernetes Engine (GKE)
A managed environment with clusters of multiple machines for deploying, managing, and scaling containerized applications using Google infrastructure.
4. Streamlit
An open-source app framework for quickly prototyping and deploying beautiful, performant apps in Python.
A step-by-step GCP guide on how I deployed Pair