文件最后提交记录最后更新时间
2 个月前
1 个月前
1 个月前
1 年前
1 年前
1 个月前
README.md

AI SIG

  • 构建AI基础设施框架,实现南向XPU(如:CPU/GPU/NPU)提供统一接入, 异构融合内存,异构融合虚拟化,算力切分,训推资源优先级配置,弹性迁移等关键能力,使能AI 应用全栈适配、协助企业在训推加速,训推混部等场景构建AI基础设施解决方案。

组织会议

  • 公开的会议时间:NA
  • 邮件讨论为主,涉及决策的内容,TC会议申报议题

成员

Maintainer列表

  • jimmy_hero
  • sinever
  • weixizhu94
  • zhushoukui
  • anhongkui

Committer列表

联系方式

项目清单

  • tensorflow: A framework for deep learning
  • libsvm: A Library for Support Vector Machines
  • ComputeLibrary: ARM Compute Library
  • arm-ml-examples: Machine learning examples used in Arm's ML developer space
  • armnn: Arm NN SDK enables machine learning workloads on power-efficient devices
  • oneDNN: Intel(R) Math Kernel Library for Deep Neural Networks
  • pytorch: python deep learning framework
  • SuperLUMT: Single precision real SuperLU routines for shared memory parallel machines
  • glpk: GNU Linear Programming Kit
  • octave: A high-level language for numerical computations
  • sundials: Suite of nonlinear solvers
  • opencl-headers:
    OpenCL (Open Computing Language) header files
  • eigen: a C++ template library for linear algebra: matrices, vectors, numerical solvers, and related algorithms.
  • opencl: OpenCL ICD Bindings
  • libxsmm: Library targeting Intel Architecture for specialized dense and sparse matrix operations, and deep learning primitives.
  • dlib: A modern C++ toolkit containing machine learning algorithms and tools for creating complex software in C++ to solve real world problems.
  • opennn: An open source neural networks library for machine learning.
  • mlpack: A fast, flexible machine learning library, written in C++
  • ensmallen: A flexible C++ library for efficient numerical optimization
  • incubator-mxnet: Apache MXNet (incubating) is a deep learning framework designed for both efficiency and flexibility.
  • opencv: OpenCV means Intel® Open Source Computer Vision Library.
  • alluxio: Alluxio (formerly known as Tachyon) is a virtual distributed storage system.
  • MindSpore: MindSpore is a new open source deep learning training/inference framework that could be used for mobile, edge and cloud scenarios.
  • python-asttokens: Module to annotate Python abstract syntax trees with source code positions
  • hearthd: HEARTH daemon - HACP/JSON-RPC capability broker that exposes hardware resources (GPIO, I2C, UART, sensors, system telemetry, files) to LLM agents in a safe, audit-friendly way. Includes the HACP / HBP / HCM / HMB / HAF specifications, board profiles, and an MCP bridge.