09.10.2023 Views

Advanced Data Analytics Using Python_ With Machine Learning, Deep Learning and NLP Examples ( 2023)

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Chapter 1

Introduction

if f == 'client_time':

feature_value = get_

hour(int(feature_value))

predicted = predicted + get_value(g.

session, g.scores, f, feature_value)

return str(math.exp(predicted + g.b)-1)

app.run(debug = True, host ='0.0.0.0')

This code exposes a deep learning model as a Flask web service.

A JavaScript client will send the request with web user parameters such

as the IP address, ad size, ad position, and so on, and it will return the

price of the ad as a response. The features are categorical. You will learn

how to convert them into numerical scores in Chapter 3. These scores

are stored in an in-memory database. The service fetches the score from

the database, sums the result, and replies to the client. This score will be

updated real time in each iteration of training of a deep learning model. It

is using MongoDB to store the frequency of that IP address in that site. It is

an important parameter because a user coming to a site for the first time

is really searching for something, which is not true for a user where the

frequency is greater than 5. The number of IP addresses is huge, so they

are stored in a distributed MongoDB database.

High-Performance API and Concurrent

Programming

Flask is a good choice when you are building a general solution that is

also a graphical user interface (GUI). But if high performance is the most

critical requirement of your application, then Falcon is the best choice. The

following code is an example of the same model shown previously exposed

by the Falcon framework. Another improvement I made in this code is that

I implemented multithreading, so the code will be executed in parallel.

17

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!