CatarAct API

Overview

API Server for serving CatarAct APK requests. This repository is part of CatarAct Google Bangkit Capstone Project.

How to Run Locally

Prerequisites

• Linux OS (prefered)
• nodejs
• yarn package manager
• PostgreSQL database
• A Google Cloud Project with a public Google Cloud Storage Bucket configured
• A machine learning model in Tensorflow.js format, preferably stored in Google Cloud Storage Bucket

Step by Step

• Clone this repository.
• Create a .env file in the root folder. Fill it with:

APP_PORT= # port to run the server (e.g. 3001)
DATABASE_URL= # postgresql database connection url
JWT_SECRET= # a secret text for signing JWT
MODEL_PUBLIC_URL= # a url to ML model JSON file in Tensorflow.js format
GCS_BUCKET= # Google Cloud Storage Bucket name
PROJECT_ID= # Google Cloud Project ID
GCS_CREDENTIALS= # Google Cloud Storage Service Account credentials key in stringified JSON format.

• Run yarn to install dependencies.
• Run yarn start:dev to run the development server.

Deployment

A Dockerfile is available on the root folder for containerization. The docker image can be used for deployment such as Cloud Run or a remote VM.

API

API documentation available on GDocs.

Acknowledgement

This project is developed by Bangkit CatarAct team, specifically bonaventuragal and aidahputri.