الأحد، 30 أبريل 2023
Show HN: LibRapid -- A high performance multi-dimensional array library https://ift.tt/LeSOlsh
Show HN: LibRapid -- A high performance multi-dimensional array library tl;dr I'm developing an incredibly fast library for arrays and mathematics, and I've implemented a few new features and made some improvements. I'd love for you to check it out! Links: GitHub: https://ift.tt/GWcZLes Documentation: https://ift.tt/DdE8BSc Discord: https://ift.tt/fYMj8R1 Hey everyone! I am the lead developer of LibRapid (https://ift.tt/GWcZLes), a high-performance C++ library for array manipulation and mathematics. I've been working hard to bring you some significant updates and improvements to the library, and I'm excited to share them with you! These changes will make it even easier for you to use LibRapid in your projects and enable you to do even more with it. Here's a rundown of the most notable changes: CUDA improvements: We've fixed several bugs and made performance enhancements for the CUDA implementation. Expanded support for BLAS libraries: LibRapid now has greater support for BLAS libraries, including Intel's MKL. Test suite and documentation updates: We've updated our test suite and overhauled the documentation for clarity and ease of use. Matrix Transposition: The library now supports matrix transposition with highly optimised, architecture-specific SIMD implementations and generic implementations for non-trivial types. New math utility functions: We've added various new utility functions for complex numbers, multi-precision arithmetic, and more. Matrix transposition and SIMD: LibRapid now supports matrix transposition and SIMD matrix transpose for SSE and AVX2, improving performance and memory alignment. Random number generation: We've started implementing a new random number generation library. Fast Fourier Transforms: LibRapid now supports fast Fourier transforms on 1D arrays, with more features coming soon! (We're using FFTW, so you can be sure you are getting the best possible performance) Array BLAS operations: We're implementing BLAS functions that operate on our high-level array type, allowing you to access the high-performace, specialised routines without the tedious interface. (Many operations already use BLAS functions internally) You can find more detailed information about the library and example code and usage guides in the documentation (https://ift.tt/DdE8BSc). We're always looking for feedback and contributions, so if you're interested in using LibRapid or helping out, please feel free to create a pull request or contact us via our Discord Server (https://ift.tt/fYMj8R1). I'm putting in a lot of effort to enhance LibRapid and would greatly appreciate any sponsorships to support my work. If you or your organisation is interested in sponsoring LibRapid, please feel free to reach out to me to discuss your ideas. We hope you enjoy the latest updates to LibRapid, and look forward to seeing what you build with it! April 30, 2023 at 10:55PM
الاشتراك في:
تعليقات الرسالة (Atom)
������ �����
خدمات طبيه https://www.cut-titles.com/Y4ZR
-
Медицинскийуризм в Турции Турция занимает 3е место в мире по количеству иностранных пациентов согласно отчету Глобального ал...
-
Show HN: Polka Jot v1.0 (just a scratch pad) https://ift.tt/3JwqCDN December 31, 2021 at 11:12PM
-
Show HN: WebMIDI Enabled Editors and Tools https://studiocode.dev/ March 1, 2021 at 02:33AM
-
Show HN: A Free Gatsby.js/Tailwind Theme for Business Websites https://ift.tt/2XfQIVb July 31, 2020 at 07:21AM