This repository contains the backend services for a machine learning model designed for cancer detection. The backend provides two endpoints: /predict and /predict/histories.
The service is built using Hapi.js and is deployed on Google Cloud Run (GCP), utilizing Docker for containerization. The prediction histories are stored on Google Cloud Firestore, and the machine learning model is stored in Google Cloud Storage Buckets. The ML model receives an image input and provides an output indicating whether it is "Cancer" or "Non Cancer".
Note: This repository is created as part of the submission for the Dicoding course called "Applied Machine Learning in Google Cloud".