Bangkit Capstone Project - Edusign (Django ML API)

This is the translation service backend for the Edusign application. This service acts as the integrator between the machine learning model and main backend service.

Not accepting outside party contribution

This project is created to fulfill the requirements of Bangkit Academy 2024 batch 1's Capstone Project. We greatly appreciate the enthusiasm of those who would like to contribute. However, according to the rules, we are not allowed to accept help from outside parties. The public visibility of this repository is simply to allow the Bangkit team reviewer to review our project. Any and all pull request to this repository coming from outside parties will be rejected and closed. We thank you for your understanding.

Instruction to run locally

clone the repository
create .env file with the following content

SECRET_KEY
ENVIRONMENT ('production' if deployed)
MODEL_URL
ENCODER_URL
HAND_URL
POSE_URL
FACE_URL

run python -m venv env to create a virtual environment
run env\Scripts\activate.bat to activate the virtual environment
run pip install -r requirements.txt to install requirements in the virtual environment
run python manage.py runserver to run the application

API Documentation

The following is the documentation for our API.

/translation

Endpoint for processing video into landmark and then predicting the result with ML model
Method: POST

Body:

link: string, required

Response status 200 (Success):

message: "Success message"
result: string

Deployment link

ml-api-edusign deployment (Maybe down in the future)

Authors

This project is developed by the Cloud Computing and Machine Learning division of C241-PS015 Bangkit Capstone Team

Cloud Computing

C010D4KY1114 - Kade Satrya Noto Sadharma
C253D4KY1157 - Wahyu Fardiansyach

Machine Learning

M006D4KY2955 – Yehezkiel Stephanus Austin
M006D4KY2954 – Juan Christopher Young
M006D4KY2953 – Haikal Irfano

agung-adhinata/ml-api-edusign