深度学习&PyTorch 之 DNN-回归中使用HR数据集进行了实现，但是HR数据集中只有一个变量，这里我们使用多变量在进行模拟一下

流程还是跟前面一样

1.1 数据导入

我们使用波士顿房价预测数据，这是个开源的数据集，所以通用性更强

data = pd.read_csv('./boston_house_prices.csv')
data

1.2 数据拆分

from sklearn.model_selection import train_test_split
train,test = train_test_split(data, train_size=0.7)

train_x = train[['CRIM', 'ZN', 'INDUS', 'CHAS', 'NOX', 'RM', 'AGE', 'DIS', 'RAD', 'TAX', 'PTRATIO', 'B', 'LSTAT']].values
test_x = test[['CRIM', 'ZN', 'INDUS', 'CHAS', 'NOX', 'RM', 'AGE', 'DIS', 'RAD', 'TAX', 'PTRATIO', 'B', 'LSTAT']].values

train_y = train.MEDV.values.reshape(-1, 1)
test_y = test.MEDV.values.reshape(-1, 1)

1.3 To Tensor

train_x = torch.from_numpy(train_x).type(torch.FloatTensor)
test_x = torch.from_numpy(test_x).type(torch.FloatTensor)
train_y = torch.from_numpy(train_y).type(torch.FloatTensor)
test_y = torch.from_numpy(test_y).type(torch.FloatTensor)

1.4 数据重构

from torch.utils.data import TensorDataset
from torch.utils.data import DataLoader

train_ds = TensorDataset(X, Y)
train_dl = DataLoader(train_ds, batch_size=batch, shuffle=True)

train_ds = TensorDataset(train_x, train_y)
train_dl = DataLoader(train_ds, batch_size=batch, shuffle=True)

test_ds = TensorDataset(test_x, test_y)
test_dl = DataLoader(test_ds, batch_size=batch * 2)

与之前是一样的

1.5 网络定义

class LinearModel(nn.Module):
    def __init__(self):
        super(LinearModel, self).__init__()
        self.linear = nn.Linear(13, 1)
        
    def forward(self, inputs):
        logits = self.linear(inputs)
        return logits

我们这里有13个特征变量

1.6 训练

model = LinearModel()
loss_fn = nn.MSELoss()
opt = torch.optim.SGD(model.parameters(), lr=lr) # 定义优化器

train_loss = []
train_acc = []

test_loss = []
test_acc = []


for epoch in range(epochs+1):
    model.train()
    for xb, yb in train_dl:
        pred = model(xb)
        loss = loss_fn(pred, yb)

        loss.backward()
        opt.step()
        opt.zero_grad()
    if epoch%10==0:
        model.eval()
        with torch.no_grad():
            train_epoch_loss = sum(loss_fn(model(xb), yb) for xb, yb in train_dl)
            test_epoch_loss = sum(loss_fn(model(xb), yb) for xb, yb in test_dl)
        train_loss.append(train_epoch_loss.data.item() / len(train_dl))
        test_loss.append(test_epoch_loss.data.item() / len(test_dl))

        template = ("epoch:{:2d}, 训练损失:{:.5f}, 验证损失:{:.5f}")
    
        print(template.format(epoch, train_epoch_loss.data.item() / len(train_dl), test_epoch_loss.data.item() / len(test_dl)))
print('训练完成')

epoch: 0, 训练损失:469.15608, 验证损失:440.95737
epoch:10, 训练损失:101.80890, 验证损失:109.48333
epoch:20, 训练损失:91.18239, 验证损失:100.17014
epoch:30, 训练损失:100.83169, 验证损失:97.70323
epoch:40, 训练损失:89.96843, 验证损失:97.37273
epoch:50, 训练损失:94.20027, 验证损失:96.82300
…
epoch:480, 训练损失:74.97700, 验证损失:81.29946
epoch:490, 训练损失:74.74702, 验证损失:80.76858
epoch:500, 训练损失:89.31947, 验证损失:83.06767
训练完成

1.7 结果展示

import matplotlib.pyplot as plt

plt.plot(range(len(train_loss)), train_loss, label='train_loss')
plt.plot(range(len(test_loss)), test_loss, label='test_loss')
plt.legend()