Skip to content

The Machine Learning framework and backend for zeta.ml. Deployed using Docker and Kubernetees. The Official Project from Team Zeta for HACKMIT2020.

License

Notifications You must be signed in to change notification settings

sumukshashidhar-archive/zeta-backend

Repository files navigation

zeta-backend

This is the official cloud backend software for Zeta. The other components of Zeta can be found

Raspberry Pi Software for Image Delivery

Authentication Server

ML Functions

Zeta is authenticated through tokens

Image Upload Guidelines

Send a POST Request to /api/image/upload

Sample Code.

import time
import requests

url = "http://.../api/upload/image"

image = open("test.png", 'rb').read()

token = "..."


files = {'image': open('test.png','rb')}
values = {"token":token}


response = requests.post(url, files=files, data=values)

Image Deletion

GET - /api/deleteImage

@params:
token - the JWT token
path - the image path to delete

For example

http://0.0.0.0/api/deleteImage?token=somejwttokenhere&path=somerandompath.png

This is a valid request

Machine Learning Routes

All Routes are prefixed with /api/ml/ to indicate their nature.

GET - /api/ml/face_recognition

Returns face data. Draws hitboxes for you.

GET - /api/ml/classifier

Returns classification data. For eg: people, cats, dogs, fruits, vehicles.

GET - /api/ml/color_recognition

Returns dominant color data

GET - /api/ml/writing_recognition

Returns OCR data

About

The Machine Learning framework and backend for zeta.ml. Deployed using Docker and Kubernetees. The Official Project from Team Zeta for HACKMIT2020.

Topics

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Contributors 3

  •  
  •  
  •