How to do it...

The steps for this recipe are as follows:

1. On your compute device, download the machine learning model files for YOLO:

wget https://pjreddie.com/media/files/yolov3.weights
wget https://raw.githubusercontent.com/microshak/AI_Benchtest_Device/yolov3.txt
wget https://raw.githubusercontent.com/microshak/AI_Benchtest_Device/yolov3.cfg

2. Create a CPU folder and create an __init__.py file inside it:

from flask import Flask
cpu = Flask(__name__)
 
from CPU.Yolo import yolo
from CPU.manifest import manifest
cpu.register_blueprint(yolo)
cpu.register_blueprint(manifest)

3. Create a manifest.py file that will send the capabilities of the compute server to a centralized server:

from flask_apscheduler import APScheduler
from flask import Blueprint, request, jsonify, session
import requests
import socket
import json
import os
manifest = Blueprint('manifest','manifest',url_prefix='/manifest')
scheduler = APScheduler()
 
def set_manifest():
    f = open("manifest_cpu...