Building machine learning systems in Python
Please sign in so that we can notify you about a reply
The book is designed for programmers writing on Python and wanting to learn about the construction of machine learning systems using open source libraries. We consider the main models of machine learning on examples taken from real life. This book will also be useful for machine learning specialists who want to use Python to create their own systems.
In chapter 1 "Introduction to machine learning in Python", the reader gets acquainted with the main idea of machine learning on a very simple example. But, despite simplicity, in this example there is a danger of retraining.
In chapter 2 "Classification in Real Life", we use real data to demonstrate the classification and teach the computer to distinguish between various classes of colors.
In chapter 3 "Clastorization - Search for interconnected messages" We will learn about the effectiveness of a model of a set of words with which we can find similar messages without "understanding" their meaning.
In chapter 4 "Thematic Modeling", we will not limit ourselves to classifying the message only to one cluster, but we will connect several topics with it, since political evidence is characteristic of real texts.
In Chapter 5 "Classification - Identification of Bad answers", we will learn how to apply the dilemma of displacement -dispersion to debugging machine learning models, although this chapter is devoted mainly to the use of logistics regression to evaluate whether the user"s response to the question is good or bad .
In chapter 6 "Classification II - Analysis of emotional coloring" explains the principle of the naive Bayesian classifier and describes how to use it to find out whether the tweet is a positive or negative emotional charge.
In Chapter 7 "Regression" it is explained how to use the classic but not relevant method - regression - when processing data. You will learn about more complex methods of regression, in particular Lasso and elastic networks.
In chapter 8 "Recommendation" we will build a recommendation system based on assessments set by consumers. We will also learn how to form recommendations, having only purchases data, without any assessments (which users do not always set) In chapter 9 "Classification by musical genres", we assume that someone deliberately introduced chaos into our huge collection of musical works, and the only hope of restoring order is to entrust their classification by the car. As it turns out, sometimes it is better to trust someone else"s experience than create signs yourself.
In chapter 10 "machine vision", we apply the methods of classification to image processing, highlighting signs from the data. We will also see how similar images in the set can be found using these methods.
From the chapter 11 "Reducing the dimension", we learn about methods to reduce the amount of data so that machine learning algorithms can cope with them.
In chapter 12 "When the data is more than the data, we will consider some approaches that allow you to successfully process large data sets, using several nuclei or computing clusters. We will also get acquainted with the basics of cloud computing (for the example of Amazon Web Services).
In the application "where to get additional information about machine learning" listed numerous useful resources devoted to this topic
In chapter 1 "Introduction to machine learning in Python", the reader gets acquainted with the main idea of machine learning on a very simple example. But, despite simplicity, in this example there is a danger of retraining.
In chapter 2 "Classification in Real Life", we use real data to demonstrate the classification and teach the computer to distinguish between various classes of colors.
In chapter 3 "Clastorization - Search for interconnected messages" We will learn about the effectiveness of a model of a set of words with which we can find similar messages without "understanding" their meaning.
In chapter 4 "Thematic Modeling", we will not limit ourselves to classifying the message only to one cluster, but we will connect several topics with it, since political evidence is characteristic of real texts.
In Chapter 5 "Classification - Identification of Bad answers", we will learn how to apply the dilemma of displacement -dispersion to debugging machine learning models, although this chapter is devoted mainly to the use of logistics regression to evaluate whether the user"s response to the question is good or bad .
In chapter 6 "Classification II - Analysis of emotional coloring" explains the principle of the naive Bayesian classifier and describes how to use it to find out whether the tweet is a positive or negative emotional charge.
In Chapter 7 "Regression" it is explained how to use the classic but not relevant method - regression - when processing data. You will learn about more complex methods of regression, in particular Lasso and elastic networks.
In chapter 8 "Recommendation" we will build a recommendation system based on assessments set by consumers. We will also learn how to form recommendations, having only purchases data, without any assessments (which users do not always set) In chapter 9 "Classification by musical genres", we assume that someone deliberately introduced chaos into our huge collection of musical works, and the only hope of restoring order is to entrust their classification by the car. As it turns out, sometimes it is better to trust someone else"s experience than create signs yourself.
In chapter 10 "machine vision", we apply the methods of classification to image processing, highlighting signs from the data. We will also see how similar images in the set can be found using these methods.
From the chapter 11 "Reducing the dimension", we learn about methods to reduce the amount of data so that machine learning algorithms can cope with them.
In chapter 12 "When the data is more than the data, we will consider some approaches that allow you to successfully process large data sets, using several nuclei or computing clusters. We will also get acquainted with the basics of cloud computing (for the example of Amazon Web Services).
In the application "where to get additional information about machine learning" listed numerous useful resources devoted to this topic
Author:
Author:Коэльо Луис Педро
Cover:
Cover:Soft
Category:
- Category:Computer & Technology
- Category:Science & Math
ISBN:
ISBN:978-5-97060-330-7
No reviews found