已有训练集和测试集的.csv文件，写一个ARMA卷积图卷积神经网络的python的torch代码

首先，需要导入必要的库：

import pandas as pd
import numpy as np
import torch
import torch.nn as nn
import torch.optim as optim
from torch.utils.data import DataLoader, Dataset

接下来，读取.csv文件：

train_data = pd.read_csv('train_data.csv')
test_data = pd.read_csv('test_data.csv')

定义自定义数据集类，继承自Dataset：

class CustomDataset(Dataset):
    def __init__(self, data):
        self.data = data.to_numpy()
        self.len = data.shape[0]
        
    def __getitem__(self, index):
        sample = {'x': self.data[index, :-1], 'y': self.data[index, -1]}
        return sample
    
    def __len__(self):
        return self.len

接下来，定义ARMA卷积图卷积神经网络模型：

class ARMAConvNet(nn.Module):
    def __init__(self):
        super(ARMAConvNet, self).__init__()
        self.conv1 = nn.Conv1d(in_channels=1, out_channels=32, kernel_size=5, stride=1, padding=2)
        self.conv2 = nn.Conv1d(in_channels=32, out_channels=64, kernel_size=5, stride=1, padding=2)
        self.fc1 = nn.Linear(in_features=64*8, out_features=128)
        self.fc2 = nn.Linear(in_features=128, out_features=1)
        self.relu = nn.ReLU()
        
    def forward(self, x):
        x = self.relu(self.conv1(x))
        x = self.relu(self.conv2(x))
        x = x.view(-1, 64*8)
        x = self.relu(self.fc1(x))
        x = self.fc2(x)
        return x

定义训练函数：

def train(model, optimizer, criterion, train_loader):
    model.train()
    train_loss = 0.0
    for i, data in enumerate(train_loader):
        inputs, labels = data['x'], data['y']
        inputs = inputs.float().unsqueeze(1)
        labels = labels.float().unsqueeze(1)
        optimizer.zero_grad()
        outputs = model(inputs)
        loss = criterion(outputs, labels)
        loss.backward()
        optimizer.step()
        train_loss += loss.item() * inputs.size(0)
    train_loss /= len(train_loader.dataset)
    return train_loss

定义测试函数：

def test(model, criterion, test_loader):
    model.eval()
    test_loss = 0.0
    with torch.no_grad():
        for i, data in enumerate(test_loader):
            inputs, labels = data['x'], data['y']
            inputs = inputs.float().unsqueeze(1)
            labels = labels.float().unsqueeze(1)
            outputs = model(inputs)
            loss = criterion(outputs, labels)
            test_loss += loss.item() * inputs.size(0)
    test_loss /= len(test_loader.dataset)
    return test_loss

初始化模型、优化器和损失函数：

model = ARMAConvNet()
optimizer = optim.Adam(model.parameters(), lr=0.001)
criterion = nn.MSELoss()

初始化训练集和测试集：

train_dataset = CustomDataset(train_data)
test_dataset = CustomDataset(test_data)

初始化训练集和测试集的DataLoader：

train_loader = DataLoader(train_dataset, batch_size=64, shuffle=True)
test_loader = DataLoader(test_dataset, batch_size=64, shuffle=False)

开始训练：

num_epochs = 100

for epoch in range(num_epochs):
    train_loss = train(model, optimizer, criterion, train_loader)
    test_loss = test(model, criterion, test_loader)
    print(f'Epoch {epoch+1}: Train Loss = {train_loss:.4f}, Test Loss = {test_loss:.4f}')

最后，根据需要保存模型：

torch.save(model.state_dict(), 'model.pth')