Litgu.ru - Литературный Гуру

Название: Machine Learning from Weak Supervision : An Empirical Risk Minimization Approach
Автор: Masashi Sugiyama, Han Bao, Takashi Ishida
Издательство: The MIT Press
Год: 2022
Страниц: 325
Язык: английский
Формат: pdf
Размер: 37.1 MB

Fundamental theory and practical algorithms of weakly supervised classification, emphasizing an approach based on empirical risk minimization. Standard Machine Learning (ML) techniques require large amounts of labeled data to work well. When we apply machine learning to problems in the physical world, however, it is extremely difficult to collect such quantities of labeled data. In this book Masashi Sugiyama, Han Bao, Takashi Ishida, Nan Lu, Tomoya Sakai and Gang Niu present theory and algorithms for weakly supervised learning, a paradigm of machine learning from weakly labeled data. Emphasizing an approach based on empirical risk minimization and drawing on state-of-the-art research in weakly supervised learning, the book provides both the fundamentals of the field and the advanced mathematical theories underlying them. It can be used as a reference for practitioners and researchers and in the classroom.

Название: Tidy Modeling with R
Автор: Max Kuhn and Julia Silge
Издательство: O’Reilly Media
Год: 2022
Формат: True PDF
Страниц: 384
Размер: 21,3 Mb
Язык: English

Models can be used in almost any domain for purposes including prediction, inference, or simply describing data. In all these cases, the predictive capacity of a model can be used to evaluate it, and we can build better, more useful models by adhering to good statistical practice. The tidymodels framework harmonizes the heterogeneous model interfaces in R and offers a consistent, flexible framework for modeling suitable for beginners as well as the very experienced.

Название: Ansible: Up and Running, 3rd Edition
Автор: Bas Meijer, Lorin Hochstein, and Rene Moser
Издательство: O’Reilly Media
Год: 2022
Формат: True PDF
Страниц: 473
Размер: 15,3 Mb
Язык: English

Among the many configuration management tools available, Ansible has some distinct advantages: It's minimal in nature. You don't need to install anything on your nodes. And it has an easy learning curve. With this updated third edition, you'll quickly learn how to be productive with this tool, whether you're a developer deploying code to production or a system administrator looking for a better automation solution.

Название: The Complete Linux Coding & Programming Manual - 15th Edition, 2022
Автор: Black Dog Media Limited
Издательство: Papercut Limited
Год: 2022
Страниц: 149
Язык: английский
Формат: pdf
Размер: 65.4 MB

Linux присутствует повсюду. Он оживляет Интернет как основная операционная система, стоящая за серверами Сети, он поддерживает космические корабли, это операционная система для самых быстрых суперкомпьютеров в мире, и он используется в умных телевизорах и мобильных устройствах. Почему? Потому что он сверхстабилен, молниеносен и совершенно бесплатен. С помощью этой книги вы тоже сможете стать частью открытого сообщества пользователей Linux. Учебные пособия на этих страницах помогут Вам освоить Linux, покажут, как он работает, что Вы можете делать с ним и как Вы можете программировать на Linux (C++ и Python), чтобы поднять Ваш опыт работы с Linux на еще большую высоту.

Название: Python in a Nutshell, 4th Edition (Early Release)
Автор: Alex Martelli, Anna Martelli Ravenscroft, Steve Holden, and Paul McGuire
Издательство: O’Reilly Media
Год: 2022-09-22: Fifth Release
Формат: True ePUB, MOBI
Размер: 10 Mb
Язык: English

Python was recently ranked as today's most popular programming language on the TIOBE index, thanks to its broad applicability to design and prototyping to testing, deployment, and maintenance. With this updated fourth edition, you'll learn how to get the most out of Python, whether you're a professional programmer or someone who needs this language to solve problems in a particular field.
Carefully curated by recognized experts in Python, this new edition focuses on version 3.10, bringing this seminal work on the Python language fully up to date on five version releases, including preview coverage of upcoming 3.11 features.

Название: Разработка интерфейсов. Паттерны проектирования. 3-е изд
Автор: Тидвелл Д., Брюэр Ч., Валенсия Э.
Издательство: Питep
Год: 2022
Формат: True PDF
Страниц: 558
Размер: 24,5 Mb
Язык: Русский

Спроектировать хороший интерфейс приложения не так просто. Все заказчики хотят, чтобы пользовательский опыт (user experience) был гладким, чтобы пользователь быстро и безболезненно получал то, что ему нужно, на любом устройстве и пользуясь любым каналом взаимодействия. Главная «фишка» книги — паттерны проектирования. Вы научитесь использовать паттерны при разработке интерфейсов мобильных, настольных и веб-приложений. Для каждого паттерна даны наглядные примеры и подробное описание -— от реализации до сценариев использования. Вы сразу перейдете от теории к практике!

Название: Machine Learning with Python Cookbook: Practical Solutions from Preprocessing to Deep Learning, 2nd Edition (Early Release)
Автор: Kyle Gallatin, Chris Albon
Издательство: O’Reilly Media, Inc.
Год: 2022-08-24
Язык: английский
Формат: epub (true), mobi
Размер: 10.2 MB

This practical guide provides more than 200 self-contained recipes to help you solve machine learning challenges you may encounter in your work. If you're comfortable with Python and its libraries, including pandas and scikit-learn, you'll be able to address specific problems all the way from loading data to training models and leveraging neural networks. Each recipe in this updated edition includes code that you can copy, paste, and run with a toy dataset to ensure it works. From there, you can adapt these recipes according to your use case or application. Recipes include a discussion that explains the solution and provides meaningful context. Go beyond theory and concepts by learning the nuts and bolts you need to construct working Machine Learning applications. NumPy is a foundational tool of the Python machine learning stack.

Название: Analyzing Social Networks Using R
Автор: Stephen P. Borgatti, Martin G. Everett, Jeffrey C. Johnson, Filip Agneessens
Издательство: SAGE Publications Ltd
Год: 2022
Формат: PDF
Страниц: 392
Размер: 48,7 Mb
Язык: English

This approachable book introduces network research in R, walking you through every step of doing social network analysis. Drawing together research design, data collection and data analysis, it explains the core concepts of network analysis in a non-technical way. The book balances an easy to follow explanation of the theoretical and statistical foundations underpinning network analysis with practical guidance on key steps like data management, preparation and visualisation.

Название: Learning Git: A Visual Approach for Using Git Successfully (Third Early Release)
Автор: Anna Skoulikari
Издательство: O’Reilly Media, Inc.
Год: 2022-09-22
Страниц: 131
Язык: английский
Формат: epub (true), mobi
Размер: 10.3 MB

Learn the basics of Git in a simple, visual, and tangible manner. With this practical book, author Anna Skoulikari helps you build a mental model of how Git version control works so that you can get the most out of this powerful tool. Whether you're a developer, data professional, technical writer, designer, or product manager, you'll learn the basics of Git in a straightforward and approachable way. This book is for anyone that wants to learn the basics of how to use Git. It is especially designed for individuals that are newly gaining technical skills or individuals that work in non-technical roles but still need to use Git to collaborate with their technical counterparts. Some examples of individuals that may benefit from this book include but are not limited to: coding bootcamp students, computer science students, technical writers, product managers, designers, junior developers, data scientists, and self-taught programmers.

Название: Enhancing Deep Learning Performance Using Displaced Rectifier Linear Unit
Автор: David Macedo
Издательство: Editora Dialetica
Год: 2022
Страниц: 121
Язык: английский
Формат: epub
Размер: 10.2 MB

Recently, Deep Learning has caused a significant impact on computer vision, speech recognition, and natural language understanding. In spite of the remarkable advances, Deep Learning (DL) recent performance gains have been modest and usually rely on increasing the depth of the models, which often requires more computational resources such as processing time and memory usage. To tackle this problem, we turned our attention to the interworking between the activation functions and the batch normalization, which is virtually mandatory currently. In this work, we propose the activation function Displaced Rectifier Linear Unit (DReLU) by conjecturing that extending the identity function of ReLU to the third quadrant enhances compatibility with batch normalization. Moreover, we used statistical tests to compare the impact of using distinct activation functions (ReLU, LReLU, PReLU, ELU, and DReLU) on the learning speed and test accuracy performance of VGG and Residual Networks state-of-the-art models.