基于Tensorflow2_goemetric（tfg）的Graph Autoencoder（GAE）实现

import os
import tensorflow as tf
from tensorflow import keras
import tf_geometric as tfg
from tf_geometric.utils import tf_utils
from tf_geometric.utils.graph_utils import edge_train_test_split, negative_sampling
# from sklearn.metrics import roc_auc_score
# import numpy as np
import time


os.environ["CUDA_VISIBLE_DEVICES"] = "0"  # 设置使用GPU 0

# 自己构建数据集的方法
# graph = tfg.Graph(
#     x=np.random.randn(5, 20),  # 5 nodes, 20 features,
#     edge_index=[[0, 0, 1, 3],
#                 [1, 2, 2, 1]]  # 4 undirected edges
# )
graph, (train_index, valid_index, test_index) = tfg.datasets.CoraDataset().load_data()
print("特征矩阵:\n", graph.x.shape)  # 矩阵，包含所有节点的BOW特征
print("边矩阵:\n", graph.edge_index.shape)  # 矩阵，包含所有的边信息

# 训练测试分离时，我们将边转换为无向边，也即只在右上角的矩阵中采样
undirected_train_edge_index, undirected_test_edge_index, _, _ = edge_train_test_split(
    edge_index=graph.edge_index,
    test_size=0.15,
)