Designing-Machine-Learning-Systems-with-Python.pdf

文件名称: Designing-Machine-Learning-Systems-with-Python.pdf.pdf

所属分类: 其它

开发工具:

文件大小: 2mb

下载次数: 0

上传时间: 2019-09-14

提供者: weixin_********

下载 (2mb)

不能下载？报告错误

详细说明：Designing-Machine-Learning-Systems-with-Python.pdfDesigning Machine Learning Systems with Python Copyright o 2016 Packt Publishing All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, without the prior written permission of the publisher, except in the case of brief quotations embedded in critical articles or reviews rt has been made in the preparation of this book to ensure the accuracy of the information presented However the information contained in this book is sold without warranty, cither express or implied. Neither the author nor Pacl Publishing and its dealers and distributors will be held liable for any damages caused or alleged to be caused directly or indirectly by this book Packt Publishing has endeavored to provide trademark information about all of the companies and products mentioned in this book by the appropriate use of capitals However, Packt Publishing cannot guarantee the accuracy of this information First published: April 2016 Production reference: 1310316 Published by Packt Publishing ltd Livery place 35 Livery street Birmingham b3 2PB UK IsBN978-1-78588-295-1 www.packtpub.com Credits Author Project Coordinator David julian Suzanne coutinho Reviewer Proofreader Dr. Vahid mirialili Safis Editing Commissioning Editor Indexer Veena pagare Rekha nair Acquisition Editor Graphics Tushar Gupta Disha har Jason monteiro Content Development Editor Merint thomas mathew Production coordinator Aparna bhagat Technical editor Abhishek r Kotian Cover work Aparna Bhagat c。 py edito Angad Singl About the author David Julian is currently working on a machine learning project with Urban EcologicalSystemsLtdandBlueSmartFarms(http://www.bluesmartfarms com. au) to detect and predict insect infestation in greenhouse crops. He is currently collecting a labeled training set that includes images and environmental data (temperature, humidity, soil moisture, and pH), linking this data to observations of infestation(the target variable), and using it to train neural net models. The aim is to create a model that will reduce the need for direct observation be able to anticipate insect outbreaks and subsequently control conditions. There is a brief outline of theprojectathttp://davejulian.net/projects/ues.Davidalsoworksasadata analyst, I.t. consultant, and trainer I would like to thank Hogan Gleeson, James Fuller, Kali Mclaughlin and Nadine miller. This book would not have been possible without the great work of the open source machine learning community About the reviewer Dr. Vahid mirjalili is a data scientist with a diverse background in engineering, mathematics, and computer science. With his specialty in data mining he is very interested in predictive modeling and getting insights from data. Currently, he is working towards publishing a book on big data analysis, which covers a wide range of tools and techniques for analyzing massive data sets Furthermore, as a python developer, he likes to contribute to the open source community. He has developed Python packages for data clustering, such as Py Clust. A collection of his tutorials and programsondatasciencecanbefoundinhisGithubrepositoryathttp://github com/mirjalil/DataScience. For more information, please visit his persona websiteathttp://vahidmirjalili.com. Www.Packtpub.com eBooks, discount offers and more Did you know that Packt offers e Book versions of every book published, with PDF andepuBfilesavailableYoucanupgradetotheebookversionatwww.packtpub.Com and as a print book customer, you are entitled to a discount on the eBook copy. Get in touchwithusatcustomercarepacktpub.comformoredetails Atwww.packtPub.com,youcanalsoreadacollectionoffreetechnicalarticlessignup for a range of free newsletters and receive exclusive discounts and offers on Packt books and ebooks PACKTLIB https://www2.packtpub.com/books/subscription/packtlib Do you need instant solutions to your it questions? PacktLib is Packt's online digital book library. Here, you can search, access, and read Packt's entire library of books Why subscribe? Fully searchable across every book published by Packt Copy and paste, print, and bookmark content On demand and accessible via a web browser Table of contents Preface Chapter 1: Thinking in Machine Learning The human interface Design principles 5 Types of questions 6 Are you asking the right question? Tasks 8 Classification Regression Clustering 10 Dimensionality reduction 10 Errors Optimization Linear programming Models 15 Features 23 Unified modeling language Class diagrams Object diagrams Activity diagrams State diagrams 31 Summary 33 Chapter 2: Tools and Techniques 2 Python for machine learning 36 IPython console 36 Installing the SciPy stack 37 NumPY 38 Constructing and transforming arrays 41 Mathematical operations 42 Matplotlib 44 Table of contents Pandas 48 SciP 51 Scikit-earn 54 Summary 61 Chapter 3: Turning Data into Information 63 What is data? 64 Big da 64 Challenges of big data 65 Data volume Data velocity Data variet 66 Data models Data distributions 68 Data from databases Data from the web 73 Data from natural language Data from images 78 Data from application programming interfaces 78 Signals 80 Data from sound 81 Cleaning data 82 Visualizing data 84 Summary 87 Chapter 4: Models-Learning from Information 89 Logical models 89 Generality ordering Version space 93 Coverage space 94 PAC learning and computational complexity 96 Tree models Purity 100 Rule models 101 The ordered list approach 103 Set-based rule models 105 Summary 108 Chapter 5: Linear Models 109 Introducing least squares 110 Gradient descent 111 The normal equation 116 Table of Contents ogistic regression 118 The Cost function for logistic regression 122 Multiclass classification 124 Regularization 125 Summary 128 Chapter 6: Neural Networks 129 Getting started with neural networks 129 Logistic units 131 Cost function 136 Minimizing the cost function 136 Implementing a neural network 139 Gradient checking 145 Other neural net architectures 146 Summary 147 Chapter 7: Features- How Algorithms See the World 149 Feature types 150 Quantitative features 150 Ordinal features 151 Categorical features 151 Operations and statistics 151 Structured features 154 Transforming features 154 Discretization 156 Normalization 157 Calibration 158 Principle component analysis 163 Summary 165 Chapter 8: Learning with Ensembles 167 Ensemble types 167 Bagging 168 Random forests 169 Extra trees 170 Boosting 174 Adaboost 177 Gradient boosting 179 Ensemble strategies 181 Other methods 182 lary 84

(系统自动生成,下载前可以参看下载内容)