技术背景
MindSpore Graph Learning是一个基于MindSpore的高效易用的图学习框架 。得益于MindSpore的图算融合能力，MindSpore Graph Learning能够针对图模型特有的执行模式进行编译优化，帮助开发者缩短训练时间 。MindSpore Graph Learning 还创新提出了以点为中心编程范式，提供更原生的图神经网络表达方式，并内置覆盖了大部分应用场景的模型，使开发者能够轻松搭建图神经网络 。【MindSpore Graph Learning】这是一个关于mindspore-gl的官方介绍，其定位非常接近于dgl，而且从文章（参考链接3）中的数据来看，mindspore-gl的运算效率还要高于dgl 。
在传统的机器学习中，我们可以对各种Tensor进行高效的运算、卷积等 。但是如果是一个图结构的网络 ， 除了把图结构转换成Tensor数据，再对Tensor进行处理之外，有没有可能用一种更加便捷的运算方式，能够直接在图的基础上去计算呢？在这里mindSpore-gl也给出了自己的答案 。我们可以一起来看一下mindspore-gl是如何安装和使用的 。
mindspore-gl的安装虽然官方有提供pip的安装方法，但是在库中能够提供的软件版本是非常有限的，这里我们推荐使用源码编译安装，这样也可以跟自己本地的MindSpore的版本更好的对应上 。首先把仓库clone下来，并进入到graphlearning目录下：
$ git clone https://gitee.com/mindspore/graphlearning.git正克隆到 'graphlearning'...remote: Enumerating objects: 1275, done.remote: Counting objects: 100% (221/221), done.remote: Compressing objects: 100% (152/152), done.remote: Total 1275 (delta 116), reused 127 (delta 68), pack-reused 1054接收对象中: 100% (1275/1275), 1.41 MiB | 316.00 KiB/s, 完成.处理 delta 中: 100% (715/715), 完成.$ cd graphlearning/$ ll总用量 112drwxrwxr-x 12 dechin dechin4096 11月9 17:19 ./drwxrwxr-x 10 dechin dechin4096 11月9 17:19 ../-rwxrwxr-x1 dechin dechin1429 11月9 17:19 build.sh*drwxrwxr-x2 dechin dechin4096 11月9 17:19 examples/-rwxrwxr-x1 dechin dechin3148 11月9 17:19 FAQ_CN.md*-rwxrwxr-x1 dechin dechin4148 11月9 17:19 faq.md*drwxrwxr-x8 dechin dechin4096 11月9 17:19 .git/-rwxrwxr-x1 dechin dechin1844 11月9 17:19 .gitignore*drwxrwxr-x2 dechin dechin4096 11月9 17:19 images/drwxrwxr-x3 dechin dechin4096 11月9 17:19 .jenkins/-rw-rw-r--1 dechin dechin 11357 11月9 17:19 LICENSEdrwxrwxr-x 11 dechin dechin4096 11月9 17:19 mindspore_gl/drwxrwxr-x 11 dechin dechin4096 11月9 17:19 model_zoo/-rwxrwxr-x1 dechin dechin52 11月9 17:19 OWNERS*-rwxrwxr-x1 dechin dechin3648 11月9 17:19 README_CN.md*-rwxrwxr-x1 dechin dechin4570 11月9 17:19 README.md*drwxrwxr-x4 dechin dechin4096 11月9 17:19 recommendation/-rwxrwxr-x1 dechin dechin922 11月9 17:19 RELEASE.md*-rwxrwxr-x1 dechin dechin108 11月9 17:19 requirements.txt*drwxrwxr-x2 dechin dechin4096 11月9 17:19 scripts/-rwxrwxr-x1 dechin dechin4164 11月9 17:19 setup.py*drwxrwxr-x5 dechin dechin4096 11月9 17:19 tests/drwxrwxr-x5 dechin dechin4096 11月9 17:19 tools/然后执行官方提供的编译构建的脚本：$ bash build.shmkdir: 已创建目录 '/home/dechin/projects/mindspore/graphlearning/output'Collecting Cython>=0.29.24Downloading Cython-0.29.32-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux_2_24_x86_64.whl (2.0 MB)|████████████████████████████████| 2.0 MB 823 kB/sCollecting ast-decompiler>=0.6.0Downloading ast_decompiler-0.7.0-py3-none-any.whl (13 kB)Collecting astpretty>=2.1.0Downloading astpretty-3.0.0-py2.py3-none-any.whl (4.9 kB)Collecting scikit-learn>=0.24.2Downloading scikit_learn-1.1.3-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (30.8 MB)|████████████████████████████████| 30.8 MB 2.6 MB/sRequirement already satisfied: numpy>=1.21.2 in /home/dechin/anaconda3/envs/mindspore16/lib/python3.9/site-packages (from -r /home/dechin/projects/mindspore/graphlearning/requirements.txt (line 5)) (1.23.2)Collecting networkx>=2.6.3Downloading networkx-2.8.8-py3-none-any.whl (2.0 MB)|████████████████████████████████| 2.0 MB 4.6 MB/sRequirement already satisfied: scipy>=1.3.2 in /home/dechin/anaconda3/envs/mindspore16/lib/python3.9/site-packages (from scikit-learn>=0.24.2->-r /home/dechin/projects/mindspore/graphlearning/requirements.txt (line 4)) (1.5.3)Collecting threadpoolctl>=2.0.0Downloading threadpoolctl-3.1.0-py3-none-any.whl (14 kB)Collecting joblib>=1.0.0Downloading joblib-1.2.0-py3-none-any.whl (297 kB)|████████████████████████████████| 297 kB 2.2 MB/sInstalling collected packages: threadpoolctl, joblib, scikit-learn, networkx, Cython, astpretty, ast-decompilerSuccessfully installed Cython-0.29.32 ast-decompiler-0.7.0 astpretty-3.0.0 joblib-1.2.0 networkx-2.8.8 scikit-learn-1.1.3 threadpoolctl-3.1.0running bdist_wheelrunning buildrunning build_pycreating buildcreating build/lib.linux-x86_64-3.9...removing build/bdist.linux-x86_64/wheelmindspore_gl_gpu-0.1-cp39-cp39-linux_x86_64.whl------Successfully created mindspore_gl package------
推荐阅读
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
- PGL Paddle Graph Learning 关于图计算&图学习的基础知识概览：前置知识点学习
- 论文笔记 - GRAD-MATCH: A Gradient Matching Based Data Subset Selection For Efficient Learning
- 论文笔记 - SIMILAR: Submodular Information Measures Based Active Learning In Realistic Scenarios
- 谣言检测《Rumor Detection with Self-supervised Learning on Texts and Social Graph》
- RDCL 谣言检测——《Towards Robust False Information Detection on Social Networks with Contrastive Learning》
- Learning Records JavaScript进阶
- GLA 论文解读《Label-invariant Augmentation for Semi-Supervised Graph Classification》
- Learning Records 计算机网络
- Nebula Graph介绍和SpringBoot环境连接和查询
- 谣言检测《Data Fusion Oriented Graph Convolution Network Model for Rumor Detection》
