提交 8641831e 编写于 作者: Y Yelrose

Update README.md

上级 aa855dee
<img src="./docs/source/_static/logo.png" alt="The logo of Paddle Graph Learning (PGL)" width="320"> <img src="./docs/source/_static/logo.png" alt="The logo of Paddle Graph Learning (PGL)" width="320">
[![PyPi Latest Release](https://img.shields.io/pypi/v/pgl.svg)](https://pypi.org/project/pgl/)
[![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](./LICENSE)
[文档](https://pgl.readthedocs.io/en/latest/) | [快速开始](https://pgl.readthedocs.io/en/latest/quick_start/instruction.html) | [English](./README.md) [文档](https://pgl.readthedocs.io/en/latest/) | [快速开始](https://pgl.readthedocs.io/en/latest/quick_start/instruction.html) | [English](./README.md)
## 最新消息
- **ERNIESage**是PGL团队最新提出的模型,可以用于建模文本以及图结构信息。你可以在[这里](./examples/erniesage)看到详细的介绍。
- PGL现在提供[Open Graph Benchmark](https://github.com/snap-stanford/ogb)的一些例子,你可以在[这里](./ogb_examples)找到。
- 新增了图级别的算子包括**GraphPooling**以及[**GraphNormalization**](https://arxiv.org/abs/2003.00982),这样你就能实现更多复杂的图级别分类模型。
Paddle Graph Learning (PGL)是一个基于[PaddlePaddle](https://github.com/PaddlePaddle/Paddle)的高效易用的图学习框架 Paddle Graph Learning (PGL)是一个基于[PaddlePaddle](https://github.com/PaddlePaddle/Paddle)的高效易用的图学习框架
<img src="./docs/source/_static/framework_of_pgl.png" alt="The Framework of Paddle Graph Learning (PGL)" width="800"> <img src="./docs/source/_static/framework_of_pgl.png" alt="The Framework of Paddle Graph Learning (PGL)" width="800">
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册