Graph neural networks are a versatile machine learning architecture that received a lot of attention recently due to its wide range of applications. In this technical report, we present an implementation of graph convolution and graph pooling layers for TensorFlow-Keras models, which allows a seamless...
TensorFlow GNN is a library to build Graph Neural Networks on the TensorFlow platform. It provides... a tfgnn.GraphTensor type to represent graphs with a heterogeneous schema, that is, multiple types of nodes and edges; tools for data preparation, notably a graph sampler to convert a huge ...
Spektral is a Python library for graph deep learning, based on the Keras API and TensorFlow 2. The main goal of this project is to provide a simple but flexible framework for creating graph neural networks (GNNs). You can use Spektral for classifying the users of a social network, predictin...
定义tensorflow的palceholder用于数据输入。 def convergence(a, state, old_state, k): with tf.variable_scope('Convergence'): # assign current state to old state old_state = state # 获取子结点上一个时刻的状态 # grub states of neighboring node gat = tf.gather(old_state, tf.cast(a[:, 0]...
GRAPH ATTENTION NETWORKS 文章来源:ICLR2018 下载地址:GRAPH ATTENTION NETWORKS 论文源码 Tensorflow:github.com/PetarV-/GAT Pytorch:github.com/Diego999/pyG 论文摘要 本文提出了图注意网络(GATs),它对图结构数据进行操作,并且使用了masked self-attentional layer。网络中的图注意力层的计算效率非常高(不需要矩阵计...
Graph neural networks in tensorflow and keras with spektral [application notes]. IEEE Comput. Intell. Mag. 16, 99–106 (2021). Article Google Scholar Kipf, T. N. & Welling, M. Semi-supervised classification with graph convolutional networks. In Proceedings of the 5th International Conference...
同样,如果你更熟悉 TensorFlow 和 Keras,Spektral可能会更有价值。如果你想使用新兴的 JAX 生态系统进行开发,那么Jraph可能非常适合你的 GNN 项目。 当然,如果你的团队更喜欢 Julia 而不是 Python,你可能更希望着眼于GeometricFlux.jl或GraphNeuralNetworks.jl,它们都基于 Flux.jl 机器学习生态系统。
在理论分析的基础上,我们提出了一个新的基线框架gfNN(graph filter neural network)来实证分析顶点分类问题。gfNN包括两个步骤:1。用图过滤矩阵相乘的方法过滤特征。利用机器学习模型学习顶点标签。我们使用下图中的一个简单实现模型来演示我们的框架的有效性。
同样,如果你更熟悉 TensorFlow 和 Keras,Spektral可能会更有价值。如果你想使用新兴的 JAX 生态系统进行开发,那么Jraph可能非常适合你的 GNN 项目。 当然,如果你的团队更喜欢 Julia 而不是 Python,你可能更希望着眼于GeometricFlux.jl或GraphNeuralNetworks.jl,它们都基于 Flux.jl 机器学习生态系统。
论文:网页链接;TensorFlow代码:网页链接;PyTorch代码:网页链接; ICCV 2019 Oral,继Google的ClusterGCN工业风之后,成功将GCN从可训练3/4层拓展到56层,训练超级深层的图卷积神经网络。作者在 CNN 结构的启发下成功将 GCN 的可训练深度从 3/4 层拓展到了 56 层,大幅度提高了图卷积网络的性能,并开源了源代码。