如何在Pytorch中访问CNN中卷积层的权重和L2范数？

import torch
import torch.nn as nn

# 定义一个简单的CNN模型
class SimpleCNN(nn.Module):
    def __init__(self):
        super(SimpleCNN, self).__init__()
        self.conv1 = nn.Conv2d(in_channels=3, out_channels=16, kernel_size=3, stride=1, padding=1)
        self.conv2 = nn.Conv2d(in_channels=16, out_channels=32, kernel_size=3, stride=1, padding=1)

    def forward(self, x):
        x = self.conv1(x)
        x = self.conv2(x)
        return x

# 创建模型实例
model = SimpleCNN()

# 访问第一个卷积层的权重
conv1_weights = model.conv1.weight.data
print("Conv1 Weights Shape:", conv1_weights.shape)

# 访问第二个卷积层的权重
conv2_weights = model.conv2.weight.data
print("Conv2 Weights Shape:", conv2_weights.shape)

# 计算第一个卷积层权重的L2范数
l2_norm_conv1 = torch.norm(conv1_weights, p=2)
print("L2 Norm of Conv1 Weights:", l2_norm_conv1)

# 计算第二个卷积层权重的L2范数
l2_norm_conv2 = torch.norm(conv2_weights, p=2)
print("L2 Norm of Conv2 Weights:", l2_norm_conv2)

# 使用Xavier初始化第一个卷积层的权重
nn.init.xavier_uniform_(model.conv1.weight)

# 使用Kaiming初始化第二个卷积层的权重
nn.init.kaiming_uniform_(model.conv2.weight, nonlinearity='relu')