佳木斯网站制作,wordpress 获得用户信息,海南中小企业网站建设,建设银行 网站无法打开深度学习#xff08;37#xff09;—— 图神经网络GNN#xff08;2#xff09;
这一期主要是一些简单示例#xff0c;针对不同的情况#xff0c;使用的数据都是torch_geometric的内置数据集 文章目录 深度学习#xff08;37#xff09;—— 图神经网络GNN#xff08…深度学习37—— 图神经网络GNN2
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 文章目录 深度学习37—— 图神经网络GNN21. 一个graph对节点分类2. 多个graph对图分类3.Cluster-GCN:当遇到数据很大的图 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_components2).fit_transform(h.detach().cpu().numpy())plt.figure(figsize(10, 10))plt.xticks([])plt.yticks([])plt.scatter(z[:, 0], z[:, 1], s70, ccolor, cmapSet2)plt.show()# 加载数据
dataset Planetoid(rootdata/Planetoid, nameCora, transformNormalizeFeatures()) # transform预处理
print(fDataset: {dataset}:)
print()
print(fNumber of graphs: {len(dataset)})
print(fNumber of features: {dataset.num_features})
print(fNumber of classes: {dataset.num_classes})data dataset[0] # Get the first graph object.
print()
print(data)
print()# Gather some statistics about the graph.
print(fNumber of nodes: {data.num_nodes})
print(fNumber of edges: {data.num_edges})
print(fAverage node degree: {data.num_edges / data.num_nodes:.2f})
print(fNumber of training nodes: {data.train_mask.sum()})
print(fTraining node label rate: {int(data.train_mask.sum()) / data.num_nodes:.2f})
print(fHas isolated nodes: {data.has_isolated_nodes()})
print(fHas self-loops: {data.has_self_loops()})
print(fIs 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, p0.5, trainingself.training)x self.conv2(x, edge_index)return xmodel GCN(hidden_channels16)
print(model)# 训练模型
optimizer torch.optim.Adam(model.parameters(), lr0.01, weight_decay5e-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 lossdef test():model.eval()out model(data.x, data.edge_index)pred out.argmax(dim1)test_correct pred[data.test_mask] data.y[data.test_mask]test_acc int(test_correct.sum()) / int(data.test_mask.sum())return test_accfor epoch in range(1, 101):loss train()print(fEpoch: {epoch:03d}, Loss: {loss:.4f})test_acc test()
print(fTest Accuracy: {test_acc:.4f})
model.eval()
out model(data.x, data.edge_index)
visualize(out, colordata.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(rootdata/TUDataset, nameMUTAG)
print(fDataset: {dataset}:)
print()
print(fNumber of graphs: {len(dataset)})
print(fNumber of features: {dataset.num_features})
print(fNumber of classes: {dataset.num_classes})data dataset[0] # Get the first graph object.
print(data)
print()# Gather some statistics about the first graph.
# print(fNumber of nodes: {data.num_nodes})
# print(fNumber of edges: {data.num_edges})
# print(fAverage node degree: {data.num_edges / data.num_nodes:.2f})
# print(fHas isolated nodes: {data.has_isolated_nodes()})
# print(fHas self-loops: {data.has_self_loops()})
# print(fIs undirected: {data.is_undirected()})train_dataset dataset
print(fNumber of training graphs: {len(train_dataset)})# 数据用dataloader加载
train_loader DataLoader(train_dataset, batch_size8, shuffleTrue)
for step, data in enumerate(train_loader):print(fStep {step 1}:)print()print(fNumber 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, p0.5, trainingself.training)x self.lin(x)return xmodel GCN(hidden_channels64)
print(model)# 训练
optimizer torch.optim.Adam(model.parameters(), lr0.01)
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 0for data in loader: # Iterate in batches over the training/test dataset.out model(data.x, data.edge_index, data.batch)pred out.argmax(dim1) # 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(fEpoch: {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, ClusterLoaderdataset Planetoid(rootdata/Planetoid, namePubMed, transformNormalizeFeatures())
print(fDataset: {dataset}:)
print()
print(fNumber of graphs: {len(dataset)})
print(fNumber of features: {dataset.num_features})
print(fNumber of classes: {dataset.num_classes})data dataset[0] # Get the first graph object.
print(data)
print()# Gather some statistics about the graph.
print(fNumber of nodes: {data.num_nodes})
print(fNumber of edges: {data.num_edges})
print(fAverage node degree: {data.num_edges / data.num_nodes:.2f})
print(fNumber of training nodes: {data.train_mask.sum()})
print(fTraining node label rate: {int(data.train_mask.sum()) / data.num_nodes:.3f})
print(fHas isolated nodes: {data.has_isolated_nodes()})
print(fHas self-loops: {data.has_self_loops()})
print(fIs undirected: {data.is_undirected()})# 数据分区构建batch构建好batch1个epoch中有4个batch
torch.manual_seed(12345)
cluster_data ClusterData(data, num_parts128) # 1. 分区
train_loader ClusterLoader(cluster_data, batch_size32, shuffleTrue) # 2. 构建batch.total_num_nodes 0
for step, sub_data in enumerate(train_loader):print(fStep {step 1}:)print()print(fNumber of nodes in the current batch: {sub_data.num_nodes})print(sub_data)print()total_num_nodes sub_data.num_nodes
print(fIterated 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, p0.5, trainingself.training)x self.conv2(x, edge_index)return xmodel GCN(hidden_channels16)
print(model)# 训练模型
optimizer torch.optim.Adam(model.parameters(), lr0.01, weight_decay5e-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(dim1)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 accsfor epoch in range(1, 51):loss train()train_acc, val_acc, test_acc test()print(fEpoch: {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
