Author: Josh Lospinoso Pub Date: 2019 ISBN: 978-1593278885 Pages: 792 Language: English Format: PDF Size: 10 Mb A fast-paced, thorough introduction to modern C++ written for experienced programmers. After reading C++ Crash Course, you’ll be proficient in the core language concepts, the C++ Standard Library, and the Boost Libraries. C++ is one of the most widely used languages for real-world software. In the hands of a knowledgeable programmer, C++ can produce small, efficient, and readable code that any programmer would be proud of. Designed for intermediate to advanced programmers, C++ Crash Course cuts through the weeds to get you straight to the core of C++17, the most modern revision of the ISO standard. Part 1 covers the core of the C++ language, where you’ll learn about everything from types and functions, to the object life cycle and expressions. Part 2 introduces you to the C++ Standard Library and Boost Libraries, where you’ll learn about all of the high-quality, fully-featured facilities available to you. You’ll cover special utility classes, data structures, and algorithms, and learn how to manipulate file systems and build high-performance programs that communicate over networks. You’ll learn all the major features of modern C++, including: Fundamental types, reference types, and user-defined types The object lifecycle including storage duration, memory management, exceptions, call stacks, and the RAII paradigm Compile-time polymorphism with templates and run-time polymorphism with virtual classes Advanced expressions, statements, and functions Smart pointers, data structures, dates and times, numerics, and probability/statistics facilities Containers, iterators, strings, and algorithms Streams and files, concurrency, networking, and application development With well over 500 code samples and nearly 100 exercises, C++ Crash Course is sure to help you build a strong C++ foundation.

Introduction to 3D Game Programming with DirectX 10（英文版）

过去几年里，深度学习在许多异质领域取得了令人印象深刻的成功:图像识别、游戏、生物学、自然语言处理，等等。然而，要真正理解这些成就背后的数学原理，我们还有很长的路要走。除了本身具有理论意义之外，理解为什么这些技术如此有效，肯定会提高它们的性能，扩大它们的应用领域。 几何深度学习(GDL)是一个很有前途的研究方向。GDL提出了一种统一的方法来理解为什么不同的体系结构(如CNN、LSTM、图神经网络、transformer)如此成功。强大的基本思想是，每当要逼近的函数被一组对称G不变/等变时，G不变/等变应该在体系结构中进行编码。例如，这就是为什么通过权值共享实现翻译等效的CNN在图像识别上工作得如此好，这是一个自然的翻译等效任务。 意大利第一所几何深度学习学校是位于意大利的第一所GDL强化学校。这将在意大利建立一个GDL社区，与其他欧洲国家建立联系，并最终促进GDL的研究。 一门有针对性的密集课程是使学生能够解决GDL的理想设置，否则这一领域将需要几个不同技能的单独训练，从群论和微分几何，到图论、概率论和深度学习。 目标受众包括计算机科学、数学、物理和统计专业的研究生，以及博士后和

ntroduction to Linear Algebra(Strang)

中文翻译Introduction to Linear Algebra, 5th Edition 1.3节 另一个视角改变至关重要。到目前为止，数 x 1 , x 2 , x 3 已知。右手边的 b 未知。我们通过将 A 与 x 相 乘来求出差分向量。现在我们设想 b 为已知然后我们找 x。 旧问题：计算线性组合 x 1 u + x 2 v + x 3 w 来求 b。 新问题：u, v, w 的哪种组合产生一个特定向量 b？ 这是逆问题——求能得出期望输出 b = Ax 的输入 x。你以前见过这个问题，它作为一个关于 x 1 , x 2 , x 3 的线性方程组。方程右手边是 b 1 , b 2 , b 3 。我现在要解方程组 Ax = b 来求 x 1 , x 2 , x 3

中文翻译Introduction to Linear Algebra, 5th Edition 1.2节 第一节放弃了讲向量相乘。现在我们继续来定义 v 与 w 的“点积” 。这个乘法包含单独的积 v 1 w 1 和 v 2 w 2 ，但它并不止于此。这两个数加起来得出一个数 v · w。 以下是几何部分 (向量长度及它们夹角的余弦)。 v = (v 1 , v 2 ) 与 w = (w 1 , w 2 ) 的点积或者说内积是数 v · w： v · w = v 1 w 1 + v 2 w 2 . (1) 例 1 向量 v = (4, 2) 与 w = (−1, 2) 点积为零： [ ] [ ] 点积为 0 4 −1 · = −4 + 4 = 0。 垂直向量 2 2 在数学中，0 总是一个特别的数。对于点积，它意味着这两个向量是垂直的。它们的夹角是 90 ◦ 。当我 们在图 1.1 中画出它们时，我们见到了一个矩形（不仅仅是任一平行四边形）。垂直向量最清晰的例子 是沿 x 轴的 i = (1, 0) 与沿 y 轴向上的 j = (0, 1)。再一次地，点积为 i · j = 0 + 0

Petre Stoica 著，chapter1 basic concepts. chapter2 non parametric methods chapter3 parametric methods for rational spectra. chapter 4 parametric method for line spectra. 5 filter bank methods. 6 spatial methods.

The Device Tree (DT), and Device Tree Overlay are a way to describe hardware in a system. An example of this would be to describe how the UART interfaces with the system, which pins, how they should be muxed, the device to enable, and which driver to use. The original BeagleBone didn't use the DT, but the recently released BeagleBone Black was released with the DT and is now using the 3.8 Linux Kernel.

An Introduction to Fluid Dynamics

Strang第五版线代教材，高清。比较不容易找，在另一个论坛找到的，别人手工制作的，效果非常好。