pytorch|图卷积神经网络(GCN)与图注意力(GAT)在cora数据集的应用

关注

原创最新推荐文章于 2026-07-06 18:15:05 发布 5.3k 阅读 AI 写同款· GEO 优化 ›

pytorch简单实现GCN 论文：SEMI-SUPERVISED CLASSIFICATION WITH GRAPH CONVOLUTIONAL NETWORKS 前言：学习图卷积刚入门，看了一篇GNN开山之作，而后看了最开始的四代GCN，每一代都在上代上有所更新，大致明白了GCN的发展以及原理，这篇论文是这个系列最后一篇，也是这篇论文中出现了现在GCN的逐层传播公式，后续最新的图网络都是在这个基础上进行更新，个人觉得这四篇论文对图网络感兴趣的朋友入门还是很有帮助的。这篇文章主要是针对最终得出来的公式进行翻译，简单测试。 1. 阅读详情

本文实现的两层GCN与GAT在cora数据集上效果，并可视化展示。

1.导入的相关包

import matplotlib.pyplot as plt
import networkx as nx
import numpy as np
import pandas as pd
import torch
from torch._C import parse_ir
import torch_geometric
from torch_geometric import datasets
from torch_geometric.data import Data, InMemoryDataset
from torch_geometric.datasets import Planetoid
import torch.nn.functional as F
from torch_geometric.nn import GCNConv,GATConv # GCN
from sklearn.manifold import TSNE

2.数据处理部分

path = "E:\\download\\cora\\cora\\"
cites = path + "cora.cites"
content = path + "cora.content"

# 索引字典，将原本的论文id转换到从0开始编码
index_dict = dict()
# 标签字典，将字符串标签转化为数值
label_to_index = dict()

features = []
labels = []
edge_index = []

with open(content,"r") as f:
    nodes = f.readlines()
    for node in nodes:
        node_info = node.split()
        index_dict[int(node_info[0])] = len(index_dict)
        features.append([int(i) for i in node_info[1:-1]])
        
        label_str = node_info[-1]
        if(label_str not in label_to_index.keys()):
            label_to_index[label_str] = len(label_to_index)
        labels.append(label_to_index[label_str])

with open(cites,"r") as f:
    edges = f.readlines()
    for edge in edges:
        start, end = edge.split()
        # 训练时将边视为无向的，但原本的边是有向的，因此需要正反添加两次
        edge_index.append([index_dict[int(start)],index_dict[int(end)]])
        edge_index.append([index_dict[int(end)],index_dict[int(start)]])

# 转换为Tensor
labels = torch.LongTensor(labels)
features = torch.FloatTensor(features)
# 行归一化
features = torch.nn.functional.normalize(features, p=1, dim=1)
edge_index =  torch.LongTensor(edge_index)

device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu') # 本电脑只有一个GPU

mask = torch.randperm(len(index_dict)) # 随机打乱顺序
train_mask = mask[:140]
val_mask = mask[140:640]
test_mask = mask[1708:2708]


cora = Data(x = features, edge_index = edge_index.t().contiguous(), y = labels).to(device)

3.GCN模块

class GCNNet(torch.nn.Module):
    def __init__(self, num_feature, num_label):
        super(GCNNet,self).__init__()
        self.GCN1 = GCNConv(num_feature, 16)
        self.GCN2 = GCNConv(16, num_label)  
        self.dropout = torch.nn.Dropout(p=0.5)
        
    def forward(self, data):
        x, edge_index = data.x, data.edge_index
        
        x = self.GCN1(x, edge_index)
        x = F.relu(x)
        x = self.dropout(x)
        x = self.GCN2(x, edge_index)
        
        return F.log_softmax(x, dim=1)

4.GAT模块

class GATNet(torch.nn.Module):
    def __init__(self, num_feature, num_label):
        super(GATNet,self).__init__()
        self.GAT1 = GATConv(num_feature, 8, heads = 8, concat = True, dropout = 0.6)
        self.GAT2 = GATConv(8*8, num_label, dropout = 0.6)  
        
    def forward(self, data):
        x, edge_index = data.x, data.edge_index
        
        x = self.GAT1(x, edge_index)
        x = F.relu(x)
        x = self.GAT2(x, edge_index)
        
        return F.log_softmax(x, dim=1)

5.画图及显示模块

def gcn_apply():
    model = GCNNet(features.shape[1], len(label_to_index)).to(device)
    optimizer = torch.optim.Adam(model.parameters(), lr=0.01, weight_decay=5e-4)

    acc_record = []
    loss_record = []

    for epoch in range(500):
        optimizer.zero_grad()
        out = model(cora)
        loss = F.nll_loss(out[train_mask], cora.y[train_mask])
        loss_record.append(loss.item())
        print('epoch: %d loss: %.4f' %(epoch, loss))
        loss.backward()
        optimizer.step()
        
        # if((epoch + 1)% 10 == 0):
        model.eval()
        _, pred = model(cora).max(dim=1)
        correct = int(pred[test_mask].eq(cora.y[test_mask]).sum().item())
        acc = correct / len(test_mask)
        acc_record.append(acc)
        print('Accuracy: {:.4f}'.format(acc))
        model.train()
            


    ts = TSNE(n_components=2)
    ts.fit_transform(out[test_mask].to('cpu').detach().numpy())

    x = ts.embedding_
    y = cora.y[test_mask].to('cpu').detach().numpy()

    xi = []
    for i in range(7):
        xi.append(x[np.where(y==i)])

    colors = ['mediumblue','green','red','yellow','cyan','mediumvioletred','mediumspringgreen']
    plt.figure(figsize=(8, 6))
    for i in range(7):
        plt.scatter(xi[i][:,0],xi[i][:,1],s=30,color=colors[i],marker='+',alpha=1)
    plt.show()

    plt.figure(figsize=(14, 9))
    x = range(500)

    plt.plot(x,acc_record,label='acc')
    plt.plot(x,loss_record,label='loss')
    plt.legend()
    plt.grid()
    plt.show()


def gat_apply():
    seed = 1
    torch.manual_seed(seed)
    torch.cuda.manual_seed(seed)
    torch.cuda.manual_seed_all(seed)  
    np.random.seed(seed)  # Numpy module.
    # random.seed(seed)  # Python random module.
    torch.manual_seed(seed)
    torch.backends.cudnn.benchmark = False
    torch.backends.cudnn.deterministic = True
    model = GATNet(features.shape[1], len(label_to_index)).to(device)

    optimizer = torch.optim.Adam(model.parameters(), lr=0.01, weight_decay=5e-4)

    acc_record = []
    loss_record = []


    for epoch in range(500):
        optimizer.zero_grad()
        out = model(cora)
        loss = F.nll_loss(out[train_mask], cora.y[train_mask])
        loss_record.append(loss.item())
        print('epoch: %d loss: %.4f' %(epoch, loss))
        loss.backward()
        optimizer.step()
        
        # if((epoch + 1)% 10 == 0):
        model.eval()
        _, pred = model(cora).max(dim=1)
        correct = int(pred[test_mask].eq(cora.y[test_mask]).sum().item())
        acc = correct / len(test_mask)
        acc_record.append(acc)
        print('Accuracy: {:.4f}'.format(acc))
        model.train()
            


    ts = TSNE(n_components=2)
    ts.fit_transform(out[test_mask].to('cpu').detach().numpy())

    x = ts.embedding_
    y = cora.y[test_mask].to('cpu').detach().numpy()

    xi = []
    for i in range(7):
        xi.append(x[np.where(y==i)])

    colors = ['mediumblue','green','red','yellow','cyan','mediumvioletred','mediumspringgreen']
    plt.figure(figsize=(8, 6))
    for i in range(7):
        plt.scatter(xi[i][:,0],xi[i][:,1],s=30,color=colors[i],marker='+',alpha=1)
    plt.show()

    plt.figure(figsize=(14, 9))
    x = range(500)

    plt.plot(x,acc_record,label='acc')
    plt.plot(x,loss_record,label='loss')
    plt.legend()
    plt.grid()
    plt.show()

6.主函数

if __name__ == "__main__":
    gcn_apply()
    gat_apply()

7.整个网络的代码

import matplotlib.pyplot as plt
import networkx as nx
import numpy as np
import pandas as pd
import torch
from torch._C import parse_ir
import torch_geometric
from torch_geometric import datasets
from torch_geometric.data import Data, InMemoryDataset
from torch_geometric.datasets import Planetoid
import torch.nn.functional as F
from torch_geometric.nn import GCNConv,GATConv # GCN
from sklearn.manifold import TSNE

path = "E:\\download\\cora\\cora\\"
cites = path + "cora.cites"
content = path + "cora.content"

# 索引字典，将原本的论文id转换到从0开始编码
index_dict = dict()
# 标签字典，将字符串标签转化为数值
label_to_index = dict()

features = []
labels = []
edge_index = []

with open(content,"r") as f:
    nodes = f.readlines()
    for node in nodes:
        node_info = node.split()
        index_dict[int(node_info[0])] = len(index_dict)
        features.append([int(i) for i in node_info[1:-1]])
        
        label_str = node_info[-1]
        if(label_str not in label_to_index.keys()):
            label_to_index[label_str] = len(label_to_index)
        labels.append(label_to_index[label_str])

with open(cites,"r") as f:
    edges = f.readlines()
    for edge in edges:
        start, end = edge.split()
        # 训练时将边视为无向的，但原本的边是有向的，因此需要正反添加两次
        edge_index.append([index_dict[int(start)],index_dict[int(end)]])
        edge_index.append([index_dict[int(end)],index_dict[int(start)]])


class GCNNet(torch.nn.Module):
    def __init__(self, num_feature, num_label):
        super(GCNNet,self).__init__()
        self.GCN1 = GCNConv(num_feature, 16)
        self.GCN2 = GCNConv(16, num_label)  
        self.dropout = torch.nn.Dropout(p=0.5)
        
    def forward(self, data):
        x, edge_index = data.x, data.edge_index
        
        x = self.GCN1(x, edge_index)
        x = F.relu(x)
        x = self.dropout(x)
        x = self.GCN2(x, edge_index)
        
        return F.log_softmax(x, dim=1)


class GATNet(torch.nn.Module):
    def __init__(self, num_feature, num_label):
        super(GATNet,self).__init__()
        self.GAT1 = GATConv(num_feature, 8, heads = 8, concat = True, dropout = 0.6)
        self.GAT2 = GATConv(8*8, num_label, dropout = 0.6)  
        
    def forward(self, data):
        x, edge_index = data.x, data.edge_index
        
        x = self.GAT1(x, edge_index)
        x = F.relu(x)
        x = self.GAT2(x, edge_index)
        
        return F.log_softmax(x, dim=1)


# 为每个节点增加自环，但后续GCN层默认会添加自环，跳过即可
# for i in range(2708):
#     edge_index.append([i,i])
  
# 转换为Tensor
labels = torch.LongTensor(labels)
features = torch.FloatTensor(features)
# 行归一化
features = torch.nn.functional.normalize(features, p=1, dim=1)
edge_index =  torch.LongTensor(edge_index)

device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu') # 本电脑只有一个GPU

mask = torch.randperm(len(index_dict)) # 随机打乱顺序
train_mask = mask[:140]
val_mask = mask[140:640]
test_mask = mask[1708:2708]


cora = Data(x = features, edge_index = edge_index.t().contiguous(), y = labels).to(device)

def gcn_apply():
    model = GCNNet(features.shape[1], len(label_to_index)).to(device)
    optimizer = torch.optim.Adam(model.parameters(), lr=0.01, weight_decay=5e-4)

    acc_record = []
    loss_record = []

    for epoch in range(500):
        optimizer.zero_grad()
        out = model(cora)
        loss = F.nll_loss(out[train_mask], cora.y[train_mask])
        loss_record.append(loss.item())
        print('epoch: %d loss: %.4f' %(epoch, loss))
        loss.backward()
        optimizer.step()
        
        # if((epoch + 1)% 10 == 0):
        model.eval()
        _, pred = model(cora).max(dim=1)
        correct = int(pred[test_mask].eq(cora.y[test_mask]).sum().item())
        acc = correct / len(test_mask)
        acc_record.append(acc)
        print('Accuracy: {:.4f}'.format(acc))
        model.train()
            


    ts = TSNE(n_components=2)
    ts.fit_transform(out[test_mask].to('cpu').detach().numpy())

    x = ts.embedding_
    y = cora.y[test_mask].to('cpu').detach().numpy()

    xi = []
    for i in range(7):
        xi.append(x[np.where(y==i)])

    colors = ['mediumblue','green','red','yellow','cyan','mediumvioletred','mediumspringgreen']
    plt.figure(figsize=(8, 6))
    for i in range(7):
        plt.scatter(xi[i][:,0],xi[i][:,1],s=30,color=colors[i],marker='+',alpha=1)
    plt.show()

    plt.figure(figsize=(14, 9))
    x = range(500)

    plt.plot(x,acc_record,label='acc')
    plt.plot(x,loss_record,label='loss')
    plt.legend()
    plt.grid()
    plt.show()


def gat_apply():
    seed = 1
    torch.manual_seed(seed)
    torch.cuda.manual_seed(seed)
    torch.cuda.manual_seed_all(seed)  
    np.random.seed(seed)  # Numpy module.
    # random.seed(seed)  # Python random module.
    torch.manual_seed(seed)
    torch.backends.cudnn.benchmark = False
    torch.backends.cudnn.deterministic = True
    model = GATNet(features.shape[1], len(label_to_index)).to(device)

    optimizer = torch.optim.Adam(model.parameters(), lr=0.01, weight_decay=5e-4)

    acc_record = []
    loss_record = []


    for epoch in range(500):
        optimizer.zero_grad()
        out = model(cora)
        loss = F.nll_loss(out[train_mask], cora.y[train_mask])
        loss_record.append(loss.item())
        print('epoch: %d loss: %.4f' %(epoch, loss))
        loss.backward()
        optimizer.step()
        
        # if((epoch + 1)% 10 == 0):
        model.eval()
        _, pred = model(cora).max(dim=1)
        correct = int(pred[test_mask].eq(cora.y[test_mask]).sum().item())
        acc = correct / len(test_mask)
        acc_record.append(acc)
        print('Accuracy: {:.4f}'.format(acc))
        model.train()
            


    ts = TSNE(n_components=2)
    ts.fit_transform(out[test_mask].to('cpu').detach().numpy())

    x = ts.embedding_
    y = cora.y[test_mask].to('cpu').detach().numpy()

    xi = []
    for i in range(7):
        xi.append(x[np.where(y==i)])

    colors = ['mediumblue','green','red','yellow','cyan','mediumvioletred','mediumspringgreen']
    plt.figure(figsize=(8, 6))
    for i in range(7):
        plt.scatter(xi[i][:,0],xi[i][:,1],s=30,color=colors[i],marker='+',alpha=1)
    plt.show()

    plt.figure(figsize=(14, 9))
    x = range(500)

    plt.plot(x,acc_record,label='acc')
    plt.plot(x,loss_record,label='loss')
    plt.legend()
    plt.grid()
    plt.show()

if __name__ == "__main__":
    gcn_apply()
    gat_apply()