深度学习（37）—— 图神经网络GNN（2）

深度学习（37）—— 图神经网络GNN（2）

这一期主要是一些简单示例，针对不同的情况，使用的数据都是torch_geometric的内置数据集

1. 一个graph对节点分类

from torch_geometric.datasets import Planetoid  # 下载数据集用的
from torch_geometric.transforms import NormalizeFeatures
from torch_geometric.nn import GCNConv
import matplotlib.pyplot as plt
from sklearn.manifold import TSNE
import torch
from torch.nn import Linear
import torch.nn.functional as F


# 可视化部分
def visualize(h, color):
    z = TSNE(n_components=2).fit_transform(h.detach().cpu().numpy())

    plt.figure(figsize=(10, 10))
    plt.xticks([])
    plt.yticks([])
    plt.scatter(z[:, 0], z[:, 1], s=70, c=color, cmap="Set2")
    plt.show()


# 加载数据
dataset = Planetoid(root='data/Planetoid', name='Cora', transform=NormalizeFeatures())  # transform预处理
print(f'Dataset: {dataset}:')
print('======================')
print(f'Number of graphs: {len(dataset)}')
print(f'Number of features: {dataset.num_features}')
print(f'Number of classes: {dataset.num_classes}')

data = dataset[0]  # Get the first graph object.
print()
print(data)
print('===========================================================================================================')

# Gather some statistics about the graph.
print(f'Number of nodes: {data.num_nodes}')
print(f'Number of edges: {data.num_edges}')
print(f'Average node degree: {data.num_edges / data.num_nodes:.2f}')
print(f'Number of training nodes: {data.train_mask.sum()}')
print(f'Training node label rate: {int(data.train_mask.sum()) / data.num_nodes:.2f}')
print(f'Has isolated nodes: {data.has_isolated_nodes()}')
print(f'Has self-loops: {data.has_self_loops()}')
print(f'Is undirected: {data.is_undirected()}')


# 网络定义
class GCN(torch.nn.Module):
    def __init__(self, hidden_channels):
        super().__init__()
        torch.manual_seed(1234567)
        self.conv1 = GCNConv(dataset.num_features, hidden_channels)
        self.conv2 = GCNConv(hidden_channels, dataset.num_classes)

    def forward(self, x, edge_index):
        x = self.conv1(x, edge_index)
        x = x.relu()
        x = F.dropout(x, p=0.5, training=self.training)
        x = self.conv2(x, edge_index)
        return x


model = GCN(hidden_channels=16)
print(model)

# 训练模型
optimizer = torch.optim.Adam(model.parameters(), lr=0.01, weight_decay=5e-4)
criterion = torch.nn.CrossEntropyLoss()


def train():
    model.train()
    optimizer.zero_grad()
    out = model(data.x, data.edge_index)
    loss = criterion(out[data.train_mask], data.y[data.train_mask])
    loss.backward()
    optimizer.step()
    return loss


def test():
    model.eval()
    out = model(data.x, data.edge_index)
    pred = out.argmax(dim=1)
    test_correct = pred[data.test_mask] == data.y[data.test_mask]
    test_acc = int(test_correct.sum()) / int(data.test_mask.sum())
    return test_acc

for epoch in range(1, 101):
    loss = train()
    print(f'Epoch: {epoch:03d}, Loss: {loss:.4f}')

test_acc = test()
print(f'Test Accuracy: {test_acc:.4f}')
model.eval()
out = model(data.x, data.edge_index)
visualize(out, color=data.y)

2. 多个graph对图分类

• 图也可以进行batch，做法和图像以及文本的batch是一样的
• 和对一张图中的节点分类不同的是：多了聚合操作 将各个节点特征汇总成全局特征，将其作为整个图的编码

import torch
from torch_geometric.datasets import TUDataset  # 分子数据集：https://chrsmrrs.github.io/datasets/
from torch_geometric.loader import DataLoader
from torch.nn import Linear
import torch.nn.functional as F
from torch_geometric.nn import GCNConv
from torch_geometric.nn import global_mean_pool

# 加载数据
dataset = TUDataset(root='data/TUDataset', name='MUTAG')
print(f'Dataset: {dataset}:')
print('====================')
print(f'Number of graphs: {len(dataset)}')
print(f'Number of features: {dataset.num_features}')
print(f'Number of classes: {dataset.num_classes}')

data = dataset[0]  # Get the first graph object.
print(data)
print('=============================================================')

# Gather some statistics about the first graph.
# print(f'Number of nodes: {data.num_nodes}')
# print(f'Number of edges: {data.num_edges}')
# print(f'Average node degree: {data.num_edges / data.num_nodes:.2f}')
# print(f'Has isolated nodes: {data.has_isolated_nodes()}')
# print(f'Has self-loops: {data.has_self_loops()}')
# print(f'Is undirected: {data.is_undirected()}')

train_dataset = dataset
print(f'Number of training graphs: {len(train_dataset)}')

# 数据用dataloader加载
train_loader = DataLoader(train_dataset, batch_size=8, shuffle=True)
for step, data in enumerate(train_loader):
    print(f'Step {step + 1}:')
    print('=======')
    print(f'Number of graphs in the current batch: {data.num_graphs}')
    print(data)
    print()


# 模型定义
class GCN(torch.nn.Module):
    def __init__(self, hidden_channels):
        super(GCN, self).__init__()
        torch.manual_seed(12345)
        self.conv1 = GCNConv(dataset.num_node_features, hidden_channels)
        self.conv2 = GCNConv(hidden_channels, hidden_channels)
        self.conv3 = GCNConv(hidden_channels, hidden_channels)
        self.lin = Linear(hidden_channels, dataset.num_classes)

    def forward(self, x, edge_index, batch):
        # 1.对各节点进行编码
        x = self.conv1(x, edge_index)
        x = x.relu()
        x = self.conv2(x, edge_index)
        x = x.relu()
        x = self.conv3(x, edge_index)

        # 2. 平均操作
        x = global_mean_pool(x, batch)  # [batch_size, hidden_channels]

        # 3. 输出
        x = F.dropout(x, p=0.5, training=self.training)
        x = self.lin(x)

        return x

model = GCN(hidden_channels=64)
print(model)

# 训练
optimizer = torch.optim.Adam(model.parameters(), lr=0.01)
criterion = torch.nn.CrossEntropyLoss()
def train():
    model.train()
    for data in train_loader:  # Iterate in batches over the training dataset.
        out = model(data.x, data.edge_index, data.batch)  # Perform a single forward pass.
        loss = criterion(out, data.y)  # Compute the loss.
        loss.backward()  # Derive gradients.
        optimizer.step()  # Update parameters based on gradients.
        optimizer.zero_grad()  # Clear gradients.

def test(loader):
    model.eval()
    correct = 0
    for data in loader:  # Iterate in batches over the training/test dataset.
        out = model(data.x, data.edge_index, data.batch)
        pred = out.argmax(dim=1)  # Use the class with highest probability.
        correct += int((pred == data.y).sum())  # Check against ground-truth labels.
    return correct / len(loader.dataset)  # Derive ratio of correct predictions.

for epoch in range(1, 3):
    train()
    train_acc = test(train_loader)
    print(f'Epoch: {epoch:03d}, Train Acc: {train_acc:.4f}')

3.Cluster-GCN:当遇到数据很大的图

• 传统的GCN，层数越多，计算越大
• 针对每个cluster进行GCN计算之后更新，数据量会小很多

但是存在问题：如果将一个大图聚类成多个小图，最大的问题是如何丢失这些子图之间的连接关系？——在每个batch中随机将batch里随机n个子图连接起来再计算

• 使用torch_geometric的内置方法

  • 首先使用cluster方法分区
  • 之后使用clusterloader构建batch

【即】分区后对每个区域进行batch的分配

# 遇到特别大的图该怎么办？
# 图中点和边的个数都非常大的时候会遇到什么问题呢？
# 当层数较多时，显存不够

import torch
import torch.nn.functional as F
from torch_geometric.nn import GCNConv
from torch_geometric.datasets import Planetoid
from torch_geometric.transforms import NormalizeFeatures
from torch_geometric.loader import ClusterData, ClusterLoader

dataset = Planetoid(root='data/Planetoid', name='PubMed', transform=NormalizeFeatures())
print(f'Dataset: {dataset}:')
print('==================')
print(f'Number of graphs: {len(dataset)}')
print(f'Number of features: {dataset.num_features}')
print(f'Number of classes: {dataset.num_classes}')

data = dataset[0]  # Get the first graph object.
print(data)
print('===============================================================================================================')

# Gather some statistics about the graph.
print(f'Number of nodes: {data.num_nodes}')
print(f'Number of edges: {data.num_edges}')
print(f'Average node degree: {data.num_edges / data.num_nodes:.2f}')
print(f'Number of training nodes: {data.train_mask.sum()}')
print(f'Training node label rate: {int(data.train_mask.sum()) / data.num_nodes:.3f}')
print(f'Has isolated nodes: {data.has_isolated_nodes()}')
print(f'Has self-loops: {data.has_self_loops()}')
print(f'Is undirected: {data.is_undirected()}')

# 数据分区构建batch，构建好batch，1个epoch中有4个batch
torch.manual_seed(12345)
cluster_data = ClusterData(data, num_parts=128)  # 1. 分区
train_loader = ClusterLoader(cluster_data, batch_size=32, shuffle=True)  # 2. 构建batch.

total_num_nodes = 0
for step, sub_data in enumerate(train_loader):
    print(f'Step {step + 1}:')
    print('=======')
    print(f'Number of nodes in the current batch: {sub_data.num_nodes}')
    print(sub_data)
    print()
    total_num_nodes += sub_data.num_nodes
print(f'Iterated over {total_num_nodes} of {data.num_nodes} nodes!')

# 模型定义
class GCN(torch.nn.Module):
    def __init__(self, hidden_channels):
        super(GCN, self).__init__()
        torch.manual_seed(12345)
        self.conv1 = GCNConv(dataset.num_node_features, hidden_channels)
        self.conv2 = GCNConv(hidden_channels, dataset.num_classes)

    def forward(self, x, edge_index):
        x = self.conv1(x, edge_index)
        x = x.relu()
        x = F.dropout(x, p=0.5, training=self.training)
        x = self.conv2(x, edge_index)
        return x

model = GCN(hidden_channels=16)
print(model)

# 训练模型
optimizer = torch.optim.Adam(model.parameters(), lr=0.01, weight_decay=5e-4)
criterion = torch.nn.CrossEntropyLoss()

def train():
    model.train()
    for sub_data in train_loader:
        out = model(sub_data.x, sub_data.edge_index)
        loss = criterion(out[sub_data.train_mask], sub_data.y[sub_data.train_mask])
        loss.backward()
        optimizer.step()
        optimizer.zero_grad()

def test():
    model.eval()
    out = model(data.x, data.edge_index)
    pred = out.argmax(dim=1)
    accs = []
    for mask in [data.train_mask, data.val_mask, data.test_mask]:
        correct = pred[mask] == data.y[mask]
        accs.append(int(correct.sum()) / int(mask.sum()))
    return accs

for epoch in range(1, 51):
    loss = train()
    train_acc, val_acc, test_acc = test()
    print(f'Epoch: {epoch:03d}, Train: {train_acc:.4f}, Val Acc: {val_acc:.4f}, Test Acc: {test_acc:.4f}')

这个还是很基础的一些，下一篇会说如何定义自己的数据。还有进阶版的案例。
所有项目代码已经放在github上了，欢迎造访