写一个基于pytorch的多对多回归预测，特征和标签都存在data.csv中，前四列为特征，其余五十列为标签，将真实值与预测值输出到名为1.xlsx的文档中

假设数据集文件名为"data.csv"，读入数据如下：

import pandas as pd

# 读入csv文件
data = pd.read_csv("data.csv")

# 分离特征和标签
X = data.iloc[:, :4]
y = data.iloc[:, 4:]

然后可以使用 PyTorch 建立模型并进行训练：

import torch
import torch.nn as nn
from sklearn.model_selection import train_test_split

# 划分训练集和测试集
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)

# 将数据转换为 PyTorch tensor
X_train = torch.tensor(X_train.values, dtype=torch.float32)
y_train = torch.tensor(y_train.values, dtype=torch.float32)
X_test = torch.tensor(X_test.values, dtype=torch.float32)
y_test = torch.tensor(y_test.values, dtype=torch.float32)

# 定义网络模型
class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.fc1 = nn.Linear(4, 100)
        self.fc2 = nn.Linear(100, 50)
        
    def forward(self, x):
        x = self.fc1(x)
        x = self.fc2(x)
        return x

# 初始化网络模型
net = Net()

# 定义损失函数和优化器
criterion = nn.MSELoss()
optimizer = torch.optim.Adam(net.parameters())

# 训练模型
for epoch in range(200):
    # 前向传播
    outputs = net(X_train)
    loss = criterion(outputs, y_train)
    # 反向传播并更新权重
    optimizer.zero_grad()
    loss.backward()
    optimizer.step()

# 测试模型
with torch.no_grad():
    test_outputs = net(X_test)
    test_loss = criterion(test_outputs, y_test)

# 输出测试结果
print("Test loss: ", test