Deploying the Model

It is now time to deploy our machine learning model as a REST API. First, however, you need to install the Flask micro‐framework.

---

TIP

Flask is a micro‐framework for Python that allows you to build web‐based applications. Micro‐frameworks in Python have little to no dependencies to external libraries and are thus very lightweight. Flask is particularly useful for developing REST APIs. For more information on Flask, check out its documentation at http://flask.pocoo.org/docs/1.0/.

---

Type the following in Terminal or Command Prompt to install Flask:

$ pip install flask 

Once Flask is installed, create a text file named REST_API.py, and enter the following code snippet:

import pickle
from flask import Flask, request, json, jsonify
import numpy as np
 
app = Flask(__name__)
 
#---the filename of the saved model---
filename = 'diabetes.sav'
 
#---load the saved model---
loaded_model = pickle.load(open(filename, 'rb'))
 
@app...