2019-07-17 HOW POWERFUL ARE GRAPH NEURAL NETWORKS? Graph Convolution 3D HOW POWERFUL ARE GRAPH NEURAL NETWORKS?GNN 可以有邻接元素聚合方法,本文讨论了很多的 图变形和图的方法 import torch from torch_geometric.nn.conv impo
2019-07-16 SEMI-SUPERVISED CLASSIFICATION WITH GRAPH CONVOLUTIONAL NETWORKS Graph Convolution 3D SEMI-SUPERVISED CLASSIFICATION WITH GRAPH CONVOLUTIONAL NETWORKS本文提出了一种,直接作用于图的卷积神经网络. INTRODUCTION作者发现可以用过 图拉普拉斯正则项来优化损
2019-07-15 SplineCNN:Fast Geometric Deep Learning with Continuous B-Spline Kernels Graph Convolution 3D SplineCNN: Fast Geometric Deep Learning with Continuous B-Spline Kernels本文提出条样方法的卷积神经网络,处理无规则和无序的空间数据,如mesh.所谓的条样卷积:就是使用