张量并行(Tensor Parallelism)

参考文献: Shoeybi M, Patwary M, Puri R, et al. Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism[J]. ArXiv, 2020 张量并行(

Nvidia GPU与Huawei NPU

1. Nvidia GPU 参考文献: HeKun-NVIDIA. CUDA-Programming-Guide-in-Chinese[EB/OL]. https://github.com/HeKun-NVIDIA/CUDA-Programming-Guide-in-Chinese. Nvidia.

数据并行(Data Parallelism)

参考文献: Team D, Majumder R, President V, et al. DeepSpeed: Extreme-scale model training for everyone[J]. Microsoft, 2020. Rajbhandari S, Rasley J, Ruwas

通用LaTex数学公式语法手册

MathJax是一款运行于 Web 浏览器当中的开源 JavaScript 数学符号渲染引擎,通过它可以方便的在现代 Web 浏览器当中显示数学公式,目前已经能够解析 LaTex、MathML 等标记语言。MathJax 项目发源于 2009 年,目前由 NumFOCUS 基金会主持,并且得到了 M
Your browser is out-of-date!

Update your browser to view this website correctly. Update my browser now

×