Flask

In this section, we will use the Flask microserver framework provided by Python to make a web application that provides predictions. We will get a RESTful API that we can query to get our results. Before commencing, we need to install Flask (use pip):

1. Let's begin by importing the packages:

   import re

   import pickle

   import numpy as np

   from flask import Flask, request, jsonify

   from keras.models import load_model

   from keras.preprocessing.sequence import pad_sequences

2. Now, let's write a function that loads the trained model and tokenizer:

   def load_variables():

   global model, tokenizer

   model = load_model('trained_model.h5')

   model._make_predict_function() #https://github.com/keras-team/keras/issues/6462

   with open('trained_tokenizer.pkl', 'rb') as f:

   tokenizer = pickle.load(f)

   The make_predict_function() is a hack that allows using keras models with Flask.

3. Now, we'll define preprocessing functions similar to the training code...