Singa API

Tech Used

• Python
• Tensorflow
• Mediapipe
• FastAPI

API Documentation

Setup Instructions

1. Clone this repository

git clone https://github.com/Signa-Lingua/sing-predict-api.git

2. Change directory

cd singa-predict-api

Build, Push the backend to Artifact Registry and deploy to Cloud Run

Prerequisites

For Google Cloud Environment

1. You have a Google Cloud Platform account, Google Project, and have installed the Google Cloud SDK on your local machine.

2. Setup ADC (Application Default Credentials) on your local machine.

   gcloud auth application-default login

3. You have created a Google Cloud Storage Bucket to store the logs.

4. You have created a service account to use for the Cloud Run and Cloud Build. It should have the following permission

   Click to expand

   • Cloud Run Admin
   • Secret Manager Secret Accessor
   • Service Account User
   • Cloud Build Service Account

5. You have created a Cloud Storage bucket to store the logs.

Deploy using the following command

Do it manually from gcloud CLI

gcloud builds submit --substitutions _VPC_CONNECTOR=<your-vpc-connector>,_MODEL_NAME=<your-model-name>,_SERVICE_ACCOUNT=<your-service-account>,_LOGBUCKET_NAME=<your-log-bucket-name>

Infrastructure Overview