# In case `shapenet.cs.stanford.edu` is offline, try to download the data
# from Kaggle instead (requires login):
# https://www.kaggle.com/datasets/mitkir/shapenet/download?datasetVersionNumber=1
下载成功后修改文件名为
‘shapenetcore_partanno_segmentation_benchmark_v0_normal.zip’
在代码中给出压缩包所在的目录即可自动解压和处理。
from torch_geometric.datasets import ShapeNet
dataset = ShapeNet(root='E:/tmp/ShapeNet/', categories=['Airplane'])
dataset[0]
>>> Data(pos=[2518, 3], y=[2518])
参考github issuce地址:https://github.com/pyg-team/pytorch_geometric/issues/1630
图上的学习算法
在学习了 PyG 中的数据处理、数据集、加载程序和转换之后,是时候实现我们的第一个图形神经网络了!
data handing datasets、loader and transforms in pyG. It’s time to implement our first graph neural network work!
from torch_geometric.datasets import Planetoid
dataset = Planetoid(root='/tmp/Cora', name='Cora')