Pytorch【60天修炼计划】之第一阶段——入门：Kaggle实战-房价预测

Kaggle

kaggle是一个著名的供机器学习爱好者交流的平台。网页为https://www.kaggle.com 。
今天为入门阶段的最后一天，我们对kaggle中的房价预测进行实操练习。

项目结构图如下：

按照之前转载知乎大佬的pytorch项目结构进行布置，最终结果如下：

data /

首先从kaggle中将数据集下载下来，即将train.csv和test.csv下载下来，并放到data/dataFile/下
从数据集文件可以发现：

数据集包括每栋房子的特征，有连续值也有离散值，甚至还有缺失值na，所以我们需要将这些数据进行预处理。

因为csv并不能进行数据的操作，所以接下来我们进行数据集的读取：通过pandas读取csv文件。

import pandas as pd
# 1460 × 81 其中最后一列为价格标签
train_data = pd.read_csv('./data/dataFile/train.csv')
# 1459 × 80
test_data = pd.read_csv('./data/dataFile/test.csv')

读取完数据，我们需要进行数据的预处理。

• 将所有训练数据和测试和数据的特征连接concat起来。
• 对连续数值的特征进行标准化
• 标准化后用0代替缺失值na
• 对离散数值的特征转成指示特征
• 将数据转化为tensor类型

代码依次为：

 # 将所有训练数据及测试数据的79个特征按照样本连接
 all_features = pd.concat((train_data.iloc[:, 1:-1], test_data.iloc[:, 1:]))

 # 对连续数值的特征做标准化
 numeric_features = all_features.dtypes[all_features.dtypes != 'object'].index
 all_features[numeric_features] = all_features[numeric_features].apply(lambda x: (x - x.mean()) / (x.std()))
 # 标准化后用0代替缺失值nan
 all_features = all_features.fillna(0)

 # 对离散数值转成指示特征
 # dummy_na = True将缺失值也当作合法的特征值并为其创建指示特征
 all_features = pd.get_dummies(all_features, dummy_na=True)

 # 再将数据转化为numpy格式的数据，并转为tensor
 n_train = train_data.shape[0]
 train_features = torch.tensor(all_features[:n_train].values, dtype=torch.float)
 test_features = torch.tensor(all_features[n_train:].values, dtype=torch.float)
 train_labels = torch.tensor(train_data.SalePrice.values,dtype=torch.float).view(-1, 1)

model /

模型就很简单了，一个Logistics回归就需要一个linear即可。
代码如下：

class Linear(nn.Module):
    def __init__(self, feature_num):
        super(Linear, self).__init__()
        self.linear = nn.Linear(feature_num, 1)

    def forward(self, x):
        return self.linear(x)

训练模型

取出网络，并定义loss为平方损失函数

def get_net(feature_num):
    net = Linear(feature_num)
    return net
    
loss = nn.MSELoss()

再定义比赛用评价模型的对数均方根误差。给定预测值 $\widehat{y_1},...,\widehat{y_n}$ 和对应的真实标签$y_1,...,y_n$ ，它的定义为
$$\sqrt{\frac{1}{n}\sum _{i=1}^n(log(y_i)-log(\widehat{y_i}){)}^2}$$
代码如下：

# 求对数均方根误差
def log_rmse(net, features, labels):
    with torch.no_grad():
        clipped_preds = torch.max(net(features), torch.tensor(1.0))
        rmse = torch.sqrt(2 * loss(clipped_preds.log(), labels.log()).mean())
    return rmse.item()

然后进行训练：

def train(net, train_features, train_labels, test_features, test_labels,
          num_epochs, learning_rate, weight_decay, batch_size):
    train_ls, test_ls = [], []
    # 将数据集特征和标签结合起来
    dataset = torch.utils.data.TensorDataset(train_features, train_labels)
    # 训练集迭代器
    train_iter = torch.utils.data.DataLoader(dataset, batch_size, shuffle = True)
    # 定义优化器Adam
    optimizer = torch.optim.Adam(params=net.parameters(), lr = learning_rate, weight_decay=weight_decay)
    net = net.float()
    # 进行迭代训练
    for epoch in range(num_epochs):
        for X, y in train_iter:
            l = loss(net(X.float()), y.float())
            optimizer.zero_grad()
            l.backward()
            optimizer.step()
        train_ls.append(log_rmse(net, train_features, train_labels))
        if test_labels is not None:
            test_ls.append(log_rmse(net, test_features, test_labels))
    return train_ls, test_ls

K折交叉验证

因为数据集只有训练集和测试集，那么为了得到验证集来选择模型并调节超参数。
所以我们实现一个函数返回第 $i$ 折交叉验证时所需要的训练和验证数据

# k折交叉验证，返回第i折交叉验证时所需要的训练和验证数据
def get_k_fold_data(k, i, X, y):
    assert k > 1
    fold_size = X.shape[0] // k # 返回整数
    X_train, y_train = None, None
    for j in range(k):
        idx = slice(j * fold_size, (j + 1) * fold_size)
        X_part, y_part = X[idx, :], y[idx]
        if j == i:
            X_valid, y_valid = X_part, y_part
        elif X_train is None:
            X_train, y_train = X_part, y_part
        else:
            X_train = torch.cat((X_train, X_part), dim = 0)
            y_train = torch.cat((y_train, y_part), dim = 0)
    return X_train, y_train, X_valid, y_valid

在 $K$ 折交叉验证中我们训练 次并返回训练和验证的平均误差。

def k_fold(k, X_train, y_train, num_epochs, learning_rate, weight_decay, batch_size):
    train_l_sum, valid_l_sum = 0, 0
    for i in range(k):
        data = get_k_fold_data(k, i, X_train, y_train)
        net = get_net(X_train.shape[1])
        train_ls, valid_ls = train(net, *data, num_epochs, learning_rate, weight_decay, batch_size)
        train_l_sum += train_ls[-1]
        valid_l_sum += valid_ls[-1]
        if i == 0:
            semilogy(range(1, num_epochs + 1), train_ls,'epochs', 'rmse',range(1, num_epochs + 1), valid_ls,['train', 'valid'])
        print('fold %d, train rmse %f, valid rmse %f' % (i, train_ls[-1], valid_ls[-1]))
    return train_l_sum / k, valid_l_sum / k

模型的选择。

我们使⽤⼀组未经调优的超参数并计算交叉验证误差。可以改动这些超参数来尽可能减⼩平均测试误
差。

train_data, test_data, train_features, test_features ,train_lables= loadData()
k, num_epochs, lr, weight_decay, batch_size = 5, 100, 5, 0, 64
train_l, valid_l = k_fold(k, train_features, train_lables, num_epochs, lr, weight_decay, batch_size)
print('%d-fold validation: avg train rmse %f, avg valid rmse %f' %(k, train_l, valid_l))

预测

def train_and_pred(train_features, test_features, train_labels, test_data,
                   num_epochs, lr, weight_decay, batch_size):
    net = get_net(train_features.shape[1])
    train_ls, _ = train(net, train_features, train_labels, None, None,
                        num_epochs, lr, weight_decay, batch_size)
    semilogy(range(1, num_epochs + 1), train_ls, 'epochs', 'rmse')
    print('train rmse %f' % train_ls[-1])
    preds = net(test_features).detach().numpy()
    test_data['SalePrice'] = pd.Series(preds.reshape(1, -1)[0])
    submission = pd.concat([test_data['Id'], test_data['SalePrice']], axis = 1)
    submission.to_csv('./data/dataFile/submission.csv', index = False)