基本数据类型## PyTorch ## 基本数据类型 主讲人:龙良曲 ## All is about Tensor |python|PyTorch| |---|---| |Int|IntTensor of size()| |float|FloatTensor of size()| |Int array|IntTensor of size \[d1, d2, ...]| |Float array|FloatTensor0 码力 | 16 页 | 1.09 MB | 2 年前3
PostgreSQL 9.3 DocumentationValues.....149 8.13.2. Encoding Handling.....150 8.13.3. Accessing XML Values.....151 8.14. JSON Type.....151 8.15. Arrays ..... 152 8.15.1. Declaration of Array Types ..... 152 8.15.2. Array xml_is_well_formed ..... 256 9.14.3. Processing XML ..... 257 9.14.4. Mapping Tables to XML ..... 258 9.15. JSON Functions and Operators ..... 262 9.16. Sequence Manipulation Functions ..... 265 9.17. Conditional 1. ALTER ..... 2210 E.26.3.4.2. VIEWs ..... 2210 E.26.3.5. Data Types ..... 2211 E.26.3.5.1. JSON ..... 2211 E.26.3.6. Functions ..... 2211 E.26.3.7. Server-Side Languages ..... 2212 E.26.30 码力 | 3183 页 | 7.53 MB | 2 年前3
C++高性能并行编程与优化 - 课件 - 10 从稀疏数据结构到量化数据类型2; } bate::timing("main"); return 0; } ## 使用 int64_t :每个占据 8 字节 • 如果用更大的数据类型,用时会直接提升两倍! 这是因为 i%2 的计算时间,完全隐藏在内存的超高延迟里了。 - 可见,当数据量足够大,计算量却不多时,读写数据量的大小唯一决定着你的性能。 - 特别是并行以后,计算量可以被并行加速,而访存却不行。 :每个占据 1 字节 - 因此我们可以把数据类型变小,这样所需的内存量就变小,从而内存带宽也可以减小! - 对于右边这种内存瓶颈的循环体,从 4 字节的 int 改成 int8_t,理论上可以增加 4 倍速度! - 这就是量化数据类型的思想,把占空间大的数据类型转换成较小的(损失一定精度,换来性能)。 因此如果你的程序不需要那么高精度,可以考虑用小点的数据类型。main: 0.120654s int0 码力 | 102 页 | 9.50 MB | 2 年前3
MoonBit月兔编程语言 现代编程思想 第五课 数据类型:树、二叉树、二叉搜索树、AVL树0 码力 | 29 页 | 1015.26 KB | 2 年前3
PlantUML 1.2020.23 Справочное руководство по языкуДиаграмма развёртывания • Диаграмма синхронизации Также поддерживаются следующие не-UML-диаграммы: • JSON Data • Wireframe graphical interface • Archimate diagram • Specification and Description Language [Image](/uploads/documents/1/c/e/4/1ce46ed71f9fcafdd140c9d283dbb6db/p182_1.jpg) ## 13 Display JSON Data JSON format is widely used in software. You can use PlantUML to visualize your data. To activate |fruit|Apple| |---|---| |size|Large| |color|Red| ### 13.1 Complex example You can use complex JSON structure. @startjson { "firstName": "John", "lastName": "Smith"0 码力 | 306 页 | 3.12 MB | 2 年前3
pandas: powerful Python data analysis toolkit - 1.2.3files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 2.4.2 JSON . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 936 3.1.5 JSON . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 943 3 pandas supports the integration with many file formats or data sources out of the box (csv, excel, sql, json, parquet,...). Importing data from each of these data sources is provided by function with the prefix0 码力 | 3323 页 | 12.74 MB | 2 年前3
Apache OFBiz Developer ManualIntegration Tests 36 10.3. UI Tests 36 11. Deployment 37 12. Security 38 12.1. Passwords and JWT (JSON Web Tokens) usage 38 12.1.1. How are set and used passwords and JWT in Apache OFBiz 38 12.2. Impersonation les requêtes de récupération des informations d'affichage. Le viewHandler retourne, au format json deux ensemble (map) d'élément ( viewScreen et viewEntities), la première le/s affichage/s et la le jeu de données nécessaire à son utilisation (ViewEntities). Actuellement le format utilisé est json.  ## V iewScreen0 码力 | 65 页 | 1.22 MB | 2 年前3
angular2 mock test iithe application. D - All of the above. ### Q 4 - Which of the following is correct about tconfig.json? A - This file is used to give the options about TypeScript used for the Angular JS project. B - JS application. D - All of the above. ### Q 5 - Which of the following is correct about tconfig.json? A - The target for the compilation is generally es5 because most browsers can only understand ES5 will not compile. D - All of the above. ### Q 6 - Which of the following is correct about package.json? A - This file is used to give the options about TypeScript used for the Angular JS project. B -0 码力 | 4 页 | 84.49 KB | 2 年前3
Beyond Compilation Databases to Support C++ Modules: Build Databasesmodules change the status quo? 3. Build databases ## Compilation Databases: Overview • JSON document • Array of JSON objects • Each object describes a single command ☐ Working directory ☐ Input file -std=c++23 in importer Q ☐ Different BMIs for different flags! ## Tomorrow: Build Databases • Still JSON ☐ Simple format ☐ Parsers are widely available ☐ Mainly tooling-oriented (humans are unlikely to cmake_build_database [0-1->6/6@122.4] Combining all module command databases $ cat build_database.json | json4slides ## Build Databases: Creating With CMake 3 { "version": 1, "revision":0 码力 | 31 页 | 585.05 KB | 1 年前3
Why is my Build so SlowUpdated 5 hours ago - Can build and run server locally as well - Chrome Event Tracing Format JSON - ... and more! ## Perfetto Slice CMakeFiles/WhyIsMyBuildSoSlow.dir/main.cpp.o ## V isualization dy> ## Clang -ftime-trace - Generates JSON file based on output filename - Detailed information on where the compiler spent time main.cpp.json 












