[PYTHON] I wrote a book that allows you to learn machine learning implementations and algorithms in a well-balanced manner.
I wrote a book that teaches machine learning implementations and algorithms in a well-balanced manner using the scikit-learn library.
[Introduction to machine learning for those who aim to become AI engineers Learn the flow of algorithms while implementing (Takuya Shimizu, Yutaro Ogawa, Gijutsu-Hyoronsha)](https://www.amazon.co.jp/ dp / 4297112094 /) https://www.amazon.co.jp/dp/4297112094/
It is already on sale.
Machine learning ・ Implementation of various algorithms ・ Mechanism of operation of each algorithm
I wrote it for those who want to learn these.
We hope you will take advantage of it.
Books I wrote last year
Learn while making! Development deep learning by PyTorch (Yutaro Ogawa, Mynavi Publishing)
It is positioned like a machine learning version of.
In this article ・ Motivation to write this book ・ Outline of this book ・ Table of contents of this book I will introduce.
Motivation to write this book
This book is co-authored by my colleague Mr. Shimizu and myself (Ogawa).
We were doing business such as "educational support for machine learning and AI utilization" in-house.
The dissatisfaction I had at that time was
"Machine learning, implementation of scikit-learn, and explanation of the operating principles of various algorithms, I don't have a book that writes in a well-balanced manner. "
It was that.
Also, "It's sad that there are no books written by Japanese people that cover scikit-learn to some extent." I also thought.
At that time
Was mainly used.
The rest is as an aid [2nd Edition] Python Machine Learning Programming Theory and Practice by Expert Data Scientists
is.
● For new graduates and mid-career hires newly assigned to their team (AI Technology Department) Write a book that you can pass, "Machine learning: Please read this book first to learn the knowledge and skills of scikit-learn"!
● I will write a book that allows people involved in machine learning in the field of business to implement various algorithms and understand the operating principle, which I would like them to hold down as an AI engineer!
I wrote this book with the motivation.
Outline of this book
The machine learning algorithms described in this book are as follows.
In Chapter 2, Linear regression, regularization, logistic regression, SVC（Support Vector Machine Classification） Decision tree, random forest Naive Bayes
In Chapter 3, Principal component analysis, k-means, Gauss mixed model
In Chapter 4, Gradient boosting decision tree, elbow method and silhouette analysis, t-SNE, Anomaly detection Novelty Detection, Anomaly detection Outlier Detection
And in Chapter 5, Machine learning system construction flow and performance evaluation
Explains about.
About 6 pages of the image on the paper are excerpted. The atmosphere is as follows.
- 2 pages are posted, but they are not 2 consecutive pages.
For each algorithm, an implementation example is shown with a simple subject, and then the operating principle of the algorithm (algorithm feelings, mind) is explained.
Table of contents of this book
The contents and table of contents of this manual are as follows.
Basically, for each algorithm
・ Basic flow and outline of the algorithm -Implementation and execution of the program ・ Detailed explanation of the algorithm (●● The heart of the algorithm) ・ At the end
It is a configuration that repeats.
table of contents
Chapter 1 Outline of machine learning and how to proceed with this book 1.1 Purpose and outline of this chapter 1.2 Overview of machine learning and three categories Differences between traditional artificial intelligence systems and machine learning systems Three categories of machine learning (supervised learning, unsupervised learning, reinforcement learning) 1.3 Significance of learning the operating principle (algorithm) of each machine learning method 1.4 How to study machine learning Knowledge and skills required for machine learning Tips for Efficiently Studying Machine Learning 1.5 Library and execution environment used in this manual How to create an environment for implementing machine learning
Chapter 2 Learning Algorithms While Implementing / Supervised Learning 2.1 Purpose and outline of this chapter 2.2 Linear regression by least squares method Basic flow and outline of the algorithm Program implementation and execution Detailed explanation of the algorithm (heart of linear regression model) At the end (Notes on using the linear regression model)
- The following items are omitted. 2.3 Suppression of overfitting by L1 regularization and L2 regularization 2.4 Classification by logistic regression 2.5 Classification by SVC 2.6 Classification by decision tree 2.7 Classification by random forest 2.8 Classification of text data by Naive Bayes
Chapter 3 The Heart of Algorithms Learned While Implementing / Unsupervised Learning 3.1 Purpose and outline of this chapter 3.2 Dimensional compression by principal component analysis 3.3 Clustering and data preprocessing by k-means 3.4 Clustering by Gauss mixed model (GMM)
Chapter 4 Heart and Development of Algorithms Learned While Implementing 4.1 Purpose and outline of this chapter 4.2 Gradient boosting decision tree classification 4.3 Search for the number of clusters by elbow method and silhouette analysis 4.4 Dimensional compression by t-SNE (manifold learning) 4.5 Anomaly detection (Novelty Detection, Outlier Detection)
Chapter 5 Machine learning system construction flow and model performance evaluation 5.1 Purpose and outline of this chapter 5.2 Business understanding 5.3 Data processing 5.4 Modeling 5.5 Deployment and operation
appendix A.1 How to use Google Colaboratory A.2 How to set up a machine learning implementation / execution environment on a local PC
at the end
It may not be enough for those who are doing crunchy machine learning.
but, For those who want to properly learn the implementation and algorithms of AI / machine learning from now on! I wrote this book with Mr. Shimizu so that I can recommend it with confidence.
We hope you will take advantage of it.
[Introduction to machine learning for those who aim to become AI engineers Learn the flow of algorithms while implementing (Takuya Shimizu, Yutaro Ogawa, Gijutsu-Hyoronsha)](https://www.amazon.co.jp/ dp / 4297112094 /)
https://www.amazon.co.jp/dp/4297112094/
● In the end
In my case, books are basically written outside business hours.
From time to time, I'm asked, "How do you get the time to write a book?"
I started writing my work style around that in a Qiita article with my real name.
https://qiita.com/Yutaro_Ogawa
We would appreciate it if you could read this article along with this book.
● Addition
With this alone, it will be like a promotion of a book, so according to this book, ** Summary of Python implementation know-how and tips that AI engineers want to be careful of ** Will be added.
The article will be too long, so I will divide it and post it below. I would appreciate it if you could see it together.
Qiita: Summary of Python implementation know-how and tips that AI engineers want to be careful about
Thank you for reading the above.
Recommended Posts