使用pytorch对房价进行回归预测

好的，下面是一个使用PyTorch进行房价回归预测的示例代码：

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
from torch.autograd import Variable
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler

然后读取数据集，进行预处理和划分训练集和测试集：

# 读取数据集
data = pd.read_csv('housing.csv')

# 分离特征和标签
X = data.iloc[:, :-1].values
y = data.iloc[:, -1:].values

# 数据标准化
sc = StandardScaler()
X = sc.fit_transform(X)

# 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)

接下来定义一个简单的神经网络模型，包含两个全连接层和一个ReLU激活函数：

class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.fc1 = nn.Linear(13, 64)
        self.fc2 = nn.Linear(64, 1)
        self.relu = nn.ReLU()

    def forward(self, x):
        x = self.fc1(x)
        x = self.relu(x)
        x = self.fc2(x)
        return x

然后定义训练函数和测试函数：

def train(model, criterion, optimizer, X_train, y_train):
    model.train()
    train_loss = 0.0
    for i in range(len(X_train)):
        x = Variable(torch.FloatTensor(X_train[i]))
        y_true = Variable(torch.FloatTensor(y_train[i]))
        optimizer.zero_grad()
        y_pred = model(x)
        loss = criterion(y_pred, y_true)
        loss.backward()
        optimizer.step()
        train_loss += loss.item()
    return train_loss / len(X_train)

def test(model, criterion, X_test, y_test):
    model.eval()
    test_loss = 0.0
    with torch.no_grad():
        for i in range(len(X_test)):
            x = Variable(torch.FloatTensor(X_test[i]))
            y_true = Variable(torch.FloatTensor(y_test[i]))
            y_pred = model(x)
            loss = criterion(y_pred, y_true)
            test_loss += loss.item()
    return test_loss / len(X_test)

最后定义训练和测试过程，使用Mean Squared Error作为损失函数，使用Adam优化器进行优化：

# 定义模型、损失函数和优化器
model = Net()
criterion = nn.MSELoss()
optimizer = torch.optim.Adam(model.parameters(), lr=0.001)

# 训练和测试
for epoch in range(100):
    train_loss = train(model, criterion, optimizer, X_train, y_train)
    test_loss = test(model, criterion, X_test, y_test)
    print('Epoch [{}/{}], Train Loss: {:.4f}, Test Loss: {:.4f}'.format(epoch+1, 100, train_loss, test_loss))

训练完成后，可以使用模型进行预测：

# 预测
x = Variable(torch.FloatTensor(X_test[0]))
y_true = Variable(torch.FloatTensor(y_test[0]))
y_pred = model(x)
print('True Value: {:.2f}, Predicted Value: {:.2f}'.format(y_true.item(), y_pred.item()))

以上就是一个使用PyTorch进行房价回归预测的示例代码。