- 真实数据通常混合了不同的数据类型,需要进行预处理。
- 常用的预处理方法:将实值数据重新缩放为零均值和单位方法;用均值替换缺失值。
- 将类别特征转化为指标特征,可以使我们把这个特征当作一个one-hot向量来对待。
- 我们可以使用K折交叉验证来选择模型并调整超参数。
- 对数对于相对误差很有用。
书中为此特意设置了几个函数,方便数据集的下载:download函数用来下载数据集缓存在本地
数据每条记录都包括房屋的属性值和属性,如街道类型、施工年份、屋顶类型、地下室状况等。 这些特征由各种数据类型组成,以及一些缺失值“NA”。
%matplotlib inline
import numpy as np
import pandas as pd
import torch
from torch import nn
from d2l import torch as d2l
DATA_HUB['kaggle_house_train'] = ( #@save
DATA_URL + 'kaggle_house_pred_train.csv',
'585e9cc93e70b39160e7921475f9bcd7d31219ce')
DATA_HUB['kaggle_house_test'] = ( #@save
DATA_URL + 'kaggle_house_pred_test.csv',
'fa19780a7b011d9b009e8bff8e99922a8ee2eb90')
train_data = pd.read_csv(download('kaggle_house_train'))
test_data = pd.read_csv(download('kaggle_house_test'))
#如果已经下载好了可以直接读取
#train_data = pd.read_csv('C:/Users/data/kaggle_house_pred_train.csv')
#test_data = pd.read_csv('C:/Users/data/kaggle_house_pred_test.csv')
print(train_data.shape)
print(test_data.shape)
print(train_data.iloc[0:5, [0, 1, 2, 3, -3, -2, -1]])
#拼接训练数据和测试数据(不包括标签列),同时删除Id列
all_features = pd.concat((train_data.iloc[:, 1:-1], test_data.iloc[:, 1:]))
# 若无法获得测试数据,则可根据训练数据计算均值和标准差
numeric_features = all_features.dtypes[all_features.dtypes != 'object'].index #获得类型不是'object'的特征的index(列名),即获取所有的数据列
all_features[numeric_features] = all_features[numeric_features].apply(
lambda x: (x - x.mean()) / (x.std())) # 对获得到的所有数据列进行规范化
# 在标准化数据之后,所有均值都是0,因此我们可以将缺失值设置为0
all_features[numeric_features] = all_features[numeric_features].fillna(0)
pandas软件包会自动为我们实现这一点。将原本all_features.shape由(2919,79)增加到了(2919,331)
# 接下来用独热编码替换处理离散值,也就是类型是'object'的列,如'MSZoning'
# “Dummy_na=True”将“na”(缺失值)视为有效的特征值,并为其创建指示符特征
all_features = pd.get_dummies(all_features, dummy_na=True)
all_features[['MSZoning_RL']],all_features[['MSZoning_RM']],all_features[['MSZoning_FV']]
通过values属性,我们可以从pandas格式中提取NumPy格式,并将其转换为张量表示用于训练。
n_train = train_data.shape[0]
train_features = torch.tensor(all_features[:n_train].values, dtype=torch.float32)
test_features = torch.tensor(all_features[n_train:].values, dtype=torch.float32)
train_labels = torch.tensor(
train_data.SalePrice.values.reshape(-1, 1), dtype=torch.float32)
- 显然线性模型很难让我们在竞赛中获胜,但线性模型提供了一种健全性检查, 以查看数据中是否存在有意义的信息。
- 如果我们在这里不能做得比随机猜测更好,那么我们很可能存在数据处理错误。
- 如果一切顺利,线性模型将作为基线(baseline)模型, 让我们直观地知道最好的模型有超出简单的模型多少。
loss = nn.MSELoss()
in_features = train_features.shape[1]
def get_net():
net = nn.Sequential(nn.Linear(in_features,1))
return net
def log_rmse(net, features, labels):
# 为了在取对数时进一步稳定该值,将小于1的值设置为1
clipped_preds = torch.clamp(net(features), 1, float('inf')) #torch.clamp(input, min, max)将input控制在[1,inf]内
rmse = torch.sqrt(loss(torch.log(clipped_preds),
torch.log(labels)))
return rmse.item()
def train(net, train_features, train_labels, test_features, test_labels,
num_epochs, learning_rate, weight_decay, batch_size):
train_ls, test_ls = [], []
train_iter = d2l.load_array((train_features, train_labels), batch_size)
# 这里使用的是Adam优化算法
optimizer = torch.optim.Adam(net.parameters(),
lr = learning_rate,
weight_decay = weight_decay)
for epoch in range(num_epochs):
for X, y in train_iter:
optimizer.zero_grad()
l = loss(net(X), y)
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
def get_k_fold_data(k, i, X, y):
assert k > 1
fold_size = X.shape[0] // k
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], 0)
y_train = torch.cat([y_train, y_part], 0)
return X_train, y_train, X_valid, y_valid
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()
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:
d2l.plot(list(range(1, num_epochs + 1)), [train_ls, valid_ls],
xlabel='epoch', ylabel='rmse', xlim=[1, num_epochs],
legend=['train', 'valid'], yscale='log')
print(f'折{i + 1},训练log rmse{float(train_ls[-1]):f}, '
f'验证log rmse{float(valid_ls[-1]):f}')
return train_l_sum / k, valid_l_sum / k
有了足够大的数据集和合理设置的超参数,折交叉验证往往对多次测试具有相当的稳定性。 然而,如果我们尝试了不合理的超参数,我们可能会发现验证效果不再代表真正的误差。
通过K折交叉验证确定合适的超参数:
k, num_epochs, lr, weight_decay, batch_size = 5, 100, 5, 0, 64
train_l, valid_l = k_fold(k, train_features, train_labels, num_epochs, lr,
weight_decay, batch_size)
print(f'{k}-折验证: 平均训练log rmse: {float(train_l):f}, '
f'平均验证log rmse: {float(valid_l):f}')
请注意:
def train_and_pred(train_features, test_features, train_labels, test_data,
num_epochs, lr, weight_decay, batch_size):
net = get_net()
train_ls, _ = train(net, train_features, train_labels, None, None,
num_epochs, lr, weight_decay, batch_size)
d2l.plot(np.arange(1, num_epochs + 1), [train_ls], xlabel='epoch',
ylabel='log rmse', xlim=[1, num_epochs], yscale='log')
print(f'训练log rmse:{float(train_ls[-1]):f}')
# 将网络应用于测试集。
preds = net(test_features).detach().numpy()
# 将其重新格式化以导出到Kaggle
test_data['SalePrice'] = pd.Series(preds.reshape(1, -1)[0])
submission = pd.concat([test_data['Id'], test_data['SalePrice']], axis=1)
submission.to_csv('submission.csv', index=False)
如果测试集上的预测与倍交叉验证过程中的预测相似, 那就是时候把它们上传到Kaggle了。 下面的代码将生成一个名为submission.csv的文件。
train_and_pred(train_features, test_features, train_labels, test_data,
num_epochs, lr, weight_decay, batch_size)
提交预测到Kaggle上,即可查看在测试集上的预测与实际房价(标签)的比较情况。
# 打印所有列
all_features.drop_duplicates(subset=['MSZoning'])
# 只打印所筛选列,即在上一个基础上再索引所需列
all_features.drop_duplicates(subset=['MSZoning'])[['MSZoning']]
使用duplicated()all_features[['MSZoning']].duplicated() == True/False
计算重复次数value_counts():all_features[['MSZoning']].value_counts()
注意!NAN这种缺失值是不被计入的,所以显示只有5行。
torch.clamp(input, min, max)将input控制在[1,inf]内