Mobile Deep Learning with TensorFlow Lite, ML Kit and Flutter

Hosting the custom authentication validation model

In this section, we will create an API that is used to authenticate users when they submit their login request to the model. The request headers will be parsed as a string and the model will use it to predict whether the login is valid:

1. We begin by importing the modules needed to create the API server:

from sklearn.externals import joblibfrom flask import Flask, request, jsonifyfrom string import digitsimport sysimport osimport jsonimport pandasimport numpyimport optparsefrom keras.models import Sequential, load_modelfrom keras.preprocessing import sequencefrom keras.preprocessing.text import Tokenizerfrom collections import OrderedDict

2. Now, we instantiate a Flask app object. We also load ...