Bangkit 2024 Batch 2 Capstone Team C242-PS189

DermaSeer API - Cloud Computing

Table of Contents 📝

1. Overview
2. Getting Started
3. Setup Guide
   • Clone the repo
   • Deploy Machine Learning Model API & VertexAI Models
   • Database PostgreSQL
   • Cloud Storage
   • Firebase Setup
   • Running the Application
4. env Configuration
5. API Endpoints

Overview 🌐

• DermaSeer API is an app that analyzes skin condition images using machine learning that provide diagnoses and use AI-powered to get product recommendations.

Contributor

Name	Bangkit-ID	Github-Profile	LinkedIn
Rojak Kurniawan	C318B4KY3957	@rojakkurniawan
Muhammad Rizqi Faadhilah	C318B4KY3071	@mr-fadh

Getting Started 🛠️

Tech stack

Setup Guide 🧑‍💻

1. Clone the Repo 📩

• Ensure you are on the right or your desired directory.

git clone https://github.com/DermaSeer/DermaSeer-api.git

2. Deploy Machine Learning Model API & VertexAI Models 🤖

• Machine Learning Model API

  • Ensure that you already have the machine learning model deployed. You can follow the instructions in the DermaSeer-Model-API.
  • Save the deployed model URL to the .env file.

  # example: https://dermaseer-ml-api-id.asia-southeast2.run.app
  MODEL_API_URL=""

• VertexAI Models

  • Ensure that you have the VertexAI models deployed. You can deploy our selected HuggingFace models on VertexAI.
  • Save the VertexAI model endpoint details to the .env file.

  # Vertex AI Model Endpoint
  VERTEX_AI_PROJECT_ID=""
  VERTEX_AI_LOCATION=""
  VETEX_AI_ENDPOINT_ID=""

3. Database PostgreSQL 🗄️

• PostgreSQL Installation

  • Ensure you have PostgreSQL installed on your machine. You can download it from PostgreSQL official website.

• Create Database

  • Create a new PostgreSQL database for the project.

  CREATE DATABASE [db_name]

• Configure .env file

  • Use PostgreSQL
  • Create Database and setup DATABASE_URL in .env

  DATABASE_URL=postgresql://username:password@localhost:5432/[db_name]

4. Cloud Storage ☁️

• Create a Storage Bucket
  • Create a storage bucket on Google Cloud Platform.
  • Save the bucket name to the .env file.

  # Google Cloud Storage
  GOOGLE_CLOUD_PROJECT_ID=""
  GOOGLE_CLOUD_BUCKET_NAME=""

• Fine-Grained Access Control
  • Ensure that you enable Fine-grained access control for your bucket.
• Service Account Permissions
  • Create a Service Account in Google Cloud IAM with Storage Object Admin permissions.
  • Download the Service Account JSON key file
• Save Credentials
  • Save the downloaded Service Account credentials file to:

  ./src/application/credentials/cloudStorage.json

5. Firebase Setup 🔥

• Firebase Admin
  • Go to your Firebase project settings.
  • Under Service Accounts, generate a new private key.
  • Save the Firebase Admin Service Account JSON file to:

  ./src/application/credentials/firebaseAdmin.json

• Firebase App Web
  • In Firebase project settings, click General then add a new Web App.
  • Copy the Firebase configuration object.
  • Save the Firebase configuration object to ./src/application/firebaseAuth.js Example of firebaseAuth.js :

  const firebaseConfig = {
    apiKey: "",
    authDomain: "",
    projectId: "",
    storageBucket: "",
    messagingSenderId: "",
    appId: ""
  };

6. Running the Application 🚀

• Local Development
  • Install dependencies

    npm install

  • Run database migrations

    npx prisma migrate

  • Start the application

    # Production mode
    npm run start
    
    # Development mode
    npm run dev

• Docker Deployment
  • Build docker image

    docker build -t dermaseer-api:1.0.0 .

  • Run the container

    docker run -d -p 5001:5001 --name backend dermaseer-api:1.0.0

• Google Cloud Run Deployment
  • Create Artifact Registry Repository

    gcloud artifacts repositories create dermaseer \
    --repository-format=docker \
    --location=asia-southeast2

  • Set Up Environment Variables

    # Set your project ID
    export GOOGLE_CLOUD_PROJECT=your-project-id
    
    # Set your region
    export REGION=asia-southeast2
    
    # Set your service name
    export SERVICE_NAME=dermaseer-api
    
    # Setu your Artifact Registry Repository
    export ARTIFACT_REPO=dermaseer

  • Build and push docker image to Artifact Registry

    gcloud builds submit --tag asia-southeast2-docker.pkg.dev/${GOOGLE_CLOUD_PROJECT}/${ARTIFACT_REPO}/dermaseer-api:1.0.0

  • Deploy to Cloud Run

    gcloud run deploy ${SERVICE_NAME} \
    --image asia-southeast2-docker.pkg.dev/${GOOGLE_CLOUD_PROJECT}/${ARTIFACT_REPO}/dermaseer-api:1.0.0 \
    --platform managed \
    --region ${REGION} \
    --allow-unauthenticated \
    --port 5001

Once the installation is complete:

• Access the application at http://localhost:5001 or the provided Cloud Run URL.
• Take a look at the API documentation to interact with the endpoints.

env Configuration ⚙️

Variable	Description
DATABASE_URL	Connection string for the PostgreSQL database
PORT	Port number for the Express.js server
GOOGLE_CLOUD_PROJECT_ID	Google Cloud project ID
GOOGLE_CLOUD_BUCKET_NAME	Name of Google Cloud Storage bucket
MODEL_API_URL	URL for the external model API (e.g., deployed on Google Cloud Run)
VERTEX_AI_PROJECT_ID	Google Cloud project ID for Vertex AI
VERTEX_AI_LOCATION	Location for the Vertex AI model
VERTEX_AI_ENDPOINT_ID	Endpoint ID for the deployed Vertex AI model

API Endpoints 📡

You can also explore and interact with these endpoints directly using our Postman collection: Postman API Documentation