MindSpore

Last updated

MindSpore
Developer(s) Huawei
Initial releaseMarch 31, 2020;3 years ago (2020-03-31)
Repository github.com/mindspore-ai/mindspore
Written in C++, Rust, Julia, Python, ArkTS, Cangjie, Java (Lite)
Platform Linux, Microsoft Windows, EulerOS, openEuler, OpenHarmony, Oniro OS, HarmonyOS, Android
Type Machine learning library
License Apache License 2.0
Website www.mindspore.cn/en

MindSpore is a open-source software framework for deep learning, machine learning and artificial intelligence developed by Huawei. [1] [2]

Contents

Overview

It has support for custom OpenHarmony-based HarmonyOS NEXT core system, includes an AI system stack that comes with Huawei's built LLM model called PanGu-Σ with full MindSpore framework support. [3] Alongside, OpenHarmony Native device-side AI support for training interface and ArkTS programming interface for its NNRt (Neural Network Runtime) backend configurations via MindSpore Lite AI framework codebase introduced in API 11 Beta 1 of OpenHarmony 4.1. MindSpore platform runs on Ascend AI chips and Kirin alongside other HiSilicon NPU chips. [4]

CANN (Compute Architecture of Neural Networks), heterogeneous computing architecture for AI developed by Huawei. With CANN backend in OpenCV DNN, giving developers ability to run created AI models on the Ascend, Kirin and other HiSilicon NPU enabled chips. [5]

See also

Related Research Articles

<span class="mw-page-title-main">HiSilicon</span> Chinese fabless semiconductor manufacturing company, fully owned by Huawei

HiSilicon is a Chinese fabless semiconductor company based in Shenzhen, Guangdong province and wholly owned by Huawei. HiSilicon purchases licenses for CPU designs from ARM Holdings, including the ARM Cortex-A9 MPCore, ARM Cortex-M3, ARM Cortex-A7 MPCore, ARM Cortex-A15 MPCore, ARM Cortex-A53, ARM Cortex-A57 and also for their Mali graphics cores. HiSilicon has also purchased licenses from Vivante Corporation for their GC4000 graphics core.

Eclipse Deeplearning4j is a programming library written in Java for the Java virtual machine (JVM). It is a framework with wide support for deep learning algorithms. Deeplearning4j includes implementations of the restricted Boltzmann machine, deep belief net, deep autoencoder, stacked denoising autoencoder and recursive neural tensor network, word2vec, doc2vec, and GloVe. These algorithms all include distributed parallel versions that integrate with Apache Hadoop and Spark.

Zeroth is a platform for brain-inspired computing from Qualcomm. It is based around a neural processing unit (NPU) AI accelerator chip and a software API to interact with the platform. It makes a form of machine learning known as deep learning available to mobile devices. It is used for image and sound processing, including speech recognition. The software operates locally rather than as a cloud application.

<span class="mw-page-title-main">MulticoreWare</span>

MulticoreWare Inc is a software development company, offering products and services related to HEVC video compression, machine learning, compilers for heterogeneous computing, and software performance optimization services. MulticoreWare's customers include AMD, Microsoft, Google, Qualcomm and Telestream. The company was founded in 2009 and has offices in the United States, China and India.

<span class="mw-page-title-main">TensorFlow</span> Machine learning software library

TensorFlow is a free and open-source software library for machine learning and artificial intelligence. It can be used across a range of tasks but has a particular focus on training and inference of deep neural networks.

<span class="mw-page-title-main">Tensor Processing Unit</span> AI accelerator ASIC by Google

Tensor Processing Unit (TPU) is an AI accelerator application-specific integrated circuit (ASIC) developed by Google for neural network machine learning, using Google's own TensorFlow software. Google began using TPUs internally in 2015, and in 2018 made them available for third-party use, both as part of its cloud infrastructure and by offering a smaller version of the chip for sale.

An AI accelerator, deep learning processor, or neural processing unit (NPU) is a class of specialized hardware accelerator or computer system designed to accelerate artificial intelligence and machine learning applications, including artificial neural networks and machine vision. Typical applications include algorithms for robotics, Internet of Things, and other data-intensive or sensor-driven tasks. They are often manycore designs and generally focus on low-precision arithmetic, novel dataflow architectures or in-memory computing capability. As of 2024, a typical AI integrated circuit chip contains tens of billions of MOSFET transistors.

spaCy Software library

spaCy is an open-source software library for advanced natural language processing, written in the programming languages Python and Cython. The library is published under the MIT license and its main developers are Matthew Honnibal and Ines Montani, the founders of the software company Explosion.

<span class="mw-page-title-main">Huawei Mate series</span> Android phablet smartphone line by Huawei

The Huawei Mate series, formally the Huawei Ascend Mate series, is a line of high-end HarmonyOS-powered phablet smartphones produced by Huawei, and is one of their flagship products along with the P series.

The Open Neural Network Exchange (ONNX) [] is an open-source artificial intelligence ecosystem of technology companies and research organizations that establish open standards for representing machine learning algorithms and software tools to promote innovation and collaboration in the AI sector. ONNX is available on GitHub.

<span class="mw-page-title-main">ROCm</span> Parallel computing platform: GPGPU libraries and application programming interface

ROCm is an Advanced Micro Devices (AMD) software stack for graphics processing unit (GPU) programming. ROCm spans several domains: general-purpose computing on graphics processing units (GPGPU), high performance computing (HPC), heterogeneous computing. It offers several programming models: HIP, OpenMP/Message Passing Interface (MPI), and OpenCL.

Alluxio is an open-source virtual distributed file system (VDFS). Initially as research project "Tachyon", Alluxio was created at the University of California, Berkeley's AMPLab as Haoyuan Li's Ph.D. Thesis, advised by Professor Scott Shenker & Professor Ion Stoica. Alluxio sits between computation and storage in the big data analytics stack. It provides a data abstraction layer for computation frameworks, enabling applications to connect to numerous storage systems through a common interface. The software is published under the Apache License.

<span class="mw-page-title-main">HarmonyOS</span> Distributed operating system by Huawei

HarmonyOS (HMOS) is a distributed operating system developed by Huawei for smartphones, tablets, smart TVs, smart watches, personal computers and other smart devices. It has a multikernel design with dual frameworks: the operating system selects suitable kernels from the abstraction layer in the case of devices that use diverse resources. The operating system was officially launched by Huawei in August 2019.

oneAPI (compute acceleration) Open standard for parallel computing

oneAPI is an open standard, adopted by Intel, for a unified application programming interface (API) intended to be used across different computing accelerator (coprocessor) architectures, including GPUs, AI accelerators and field-programmable gate arrays. It is intended to eliminate the need for developers to maintain separate code bases, multiple programming languages, tools, and workflows for each architecture.

The version history of the HarmonyOS distributed operating system began with the public release of the HarmonyOS 1.0 for Honor Vision smart TVs on August 9, 2019. The first commercial version of the IoT based operating system, HarmonyOS 2.0, was released on June 2, 2021 for phones, tablets, smartwatches, smart speakers, routers, and internet of things. Beforehand, DevEco Studio, the HarmonyOS app development IDE, was released in September 2020 together with the HarmonyOS 2.0 Beta. HarmonyOS is developed by Huawei. New major releases are announced at the Huawei Developers Conference (HDC) in the fourth quarter of each year together with the first public beta version of the operating system's next major version. The next major stable version is then released in the third to fourth quarter of the following year.

<span class="mw-page-title-main">OpenHarmony</span> Family of open-source operating systems based on HarmonyOS

OpenAtom OpenHarmony, or abbreviated as OpenHarmony (OHOS), is a family of open-source operating systems based on HarmonyOS derived from LiteOS, donated the L0-L2 branch source code by Huawei to the OpenAtom Foundation. Similar to HarmonyOS, the open-source distributed operating system is designed with a layered architecture, which consists of four layers from the bottom to the top, i.e., the kernel layer, system service layer, framework layer, and application layer.

HarmonyOS NEXT is a proprietary distributed operating system and an iteration of HarmonyOS, developed by Huawei to support only HarmonyOS native apps. The operating system is primarily aimed at software and hardware developers that deal directly with Huawei. It does not include Android's AOSP core and is incompatible with Android applications.

Huawei Ark Engine is a conglomerate of proprietary application programming interfaces (APIs) similar to Microsoft DirectX for handling tasks related to system and multimedia, especially game programming and video, on HarmonyOS platform such as Graphics engine, ArkGraphics 2D for 2D computer graphics, 2D Drawing high-performance, interactive graphics with ArkUI support and also low-level, low-overhead hardware-accelerated 3D graphic and compute shader ArkGraphics 3D API, multimedia engine for audio and video, memory engine, scheduling engine, storage engine and low power consumption engine.

BiSheng Compiler is an open-source compiler toolchain developed by Huawei for general-purpose processor architectures, such as Kunpeng within HiSilicon domain. It introduces and enhances multiple compilation optimization technologies and supports different programming languages, such as ArkTS, Cangjie, C, C++ and Fortran.

Huawei PanGu, PanGu, PanGu-Σ or PanGu-π is a multimodal large language model developed by Huawei. It was announced on July 7, 2023, positioned as a contender to other multimodal large language models.

References

  1. "Huawei MindSpore AI Development Framework". Springer Link. Huawei Technologies. Retrieved February 13, 2024.
  2. "Inference on the Ascend 310 AI Processor". MindSpore. MindSpore. Retrieved February 13, 2024.
  3. Bhati, Kamlesh. "Huawei HarmonyOS NEXT Official Now: Bring Spatial Design and Vivid Color Concepts". Sparrownews.com. Sparrownews.com. Retrieved February 13, 2024.
  4. "OpenHarmony 4.1 Beta1 Unleashes Cutting-Edge Features and API Advancements". World Tech. FTT World. Retrieved February 13, 2024.
  5. MSV, Janakiram. "Huawei Wants To Tackle NVIDIA And Google With A Solid AI Strategy". Forbes. Forbes. Retrieved March 11, 2024.

Bibliography