1. 实现几个函数方便下载数据
import hashlib
import os
import tarfile
import zipfile
import requests
#@save
DATA_HUB = dict()
DATA_URL = 'http://d2l-data.s3-accelerate.amazonaws.com/'
def download(name, cache_dir=os.path.join('..', 'data')): #@save
"""下载一个DATA_HUB中的文件,返回本地文件名"""
assert name in DATA_HUB, f"{name} 不存在于 {DATA_HUB}"# 判断变量name是否存在于DATA_HUB中,不在则抛出异常
url, sha1_hash = DATA_HUB[name]
os.makedirs(cache_dir, exist_ok=True)# cache_dir目录不存在,则创建该目录,如果目录已经存在,则什么都不做
fname = os.path.join(cache_dir, url.split('/')[-1])# 拼接成一个完整的路径
if os.path.exists(fname): # 路径存在
sha1 = hashlib.sha1() # 创建了一个哈希对象
with open(fname, 'rb') as f:
while True:
data = f.read(1048576)
if not data:
break
sha1.update(data)
if sha1.hexdigest() == sha1_hash:
return fname # 命中缓存
print(f'正在从{url}下载{fname}...')
r = requests.get(url, stream=True, verify=True)
with open(fname, 'wb') as f:
f.write(r.content)
return fname
def download_extract(name, folder=None): #@save
"""下载并解压zip/tar文件"""
fname = download(name)
base_dir = os.path.dirname(fname)
data_dir, ext = os.path.splitext(fname)
if ext == '.zip':
fp = zipfile.ZipFile(fname, 'r')
elif ext in ('.tar', '.gz'):
fp = tarfile.open(fname, 'r')
else:
assert False, '只有zip/tar文件可以被解压缩'
fp.extractall(base_dir)
return os.path.join(base_dir, folder) if folder else data_dir
def download_all(): #@save
"""下载DATA_HUB中的所有文件"""
for name in DATA_HUB:
download(name)
2. 使用pandas读入并处理数据
%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'] = ( # 将数据集的名称kaggle_house_train作为字典DATA_HUB的键
DATA_URL + 'kaggle_house_pred_train.csv', # 数据集的下载链接
'585e9cc93e70b39160e7921475f9bcd7d31219ce') # 哈希值用于验证数据的完整性
DATA_HUB['kaggle_house_test'] = (
DATA_URL + 'kaggle_house_pred_test.csv',
'fa19780a7b011d9b009e8bff8e99922a8ee2eb90')
# 从指定的数据源下载名为'kaggle_house_train'的CSV文件,
# 并使用pd.read_csv()函数将其读取为一个DataFrame对象,并将该对象赋值
train_data = pd.read_csv(download('kaggle_house_train'))
test_data = pd.read_csv(download('kaggle_house_test'))
print(train_data.shape)
print(test_data.shape)
观察特征
打印出前4行,前4列和最后3列打印出来
【使用iloc属性对train_data这个DataFrame对象进行切片操作,选取了指定行和列的数据子集】
print(train_data.iloc[0:4, [0, 1, 2, 3, -3, -2, -1]])
在每个样本中,第一个特征ID不能参与训练,所以要将其删除
saleprice作为标签在训练数据中要进行删除
all_features = pd.concat((train_data.iloc[:, 1:-1], test_data.iloc[:, 1:]))# 将train_data去除第一列ID和最后一列标签,和去除id的test_data进行合并
将所有缺失的值替换为相应特征的平均值,通过将特征重新缩放到零均值和单位方差来标准化数据
【.fillna(0)对选择的数值型特征进行了填充操作,将缺失值(NaN值)填充为0。fillna()是一个DataFrame对象的方法,用于填充缺失值】
numeric_features = all_features.dtypes[all_features.dtypes != 'object'].index # all_features.dtypes != 'object'-》数值型数据
"""-》将数值特征均值设为0,方差设为1"""
all_features[numeric_features] = all_features[numeric_features].apply(
lambda x: (x - x.mean()) / (x.std())) # 将(数值特征 - 均值)/方差
all_features[numeric_features] = all_features[numeric_features].fillna(0) # 对选择的数值型特征进行了填充操作,将缺失值(NaN值)填充为0
处理离散值,用一次独热编码替换它们
all_features = pd.get_dummies(all_features, dummy_na=True)
all_features.shape
从pandas格式中提取NumPy格式,并将其转化为张量表示
【.values将该列数据转换为一个Numpy数组。
.reshape(-1, 1)改变数组的形状,将其变为一个列向量(具有一列)。】
n_train = train_data.shape[0]
all_features = all_features.astype(float) # 进行强制类型转化否则会报错
train_features = torch.tensor(all_features[:n_train].values, # 之前将train_data和 test_data结合,现在进行下标分开
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),# SalePrice列数据提取出来,并将其转换为一个列向量(具有一列)
dtype=torch.float32)
训练
loss = nn.MSELoss()
in_features = train_features.shape[1]
def get_net():
net = nn.Sequential(nn.Linear(in_features, 1)) # 使用单层线性回归,输入特征数:in_features,输出特征数:1
return net
为解决误差的影响,可以使用相对误差 (真实房价-预测房价/真实房价),其中一种方法是用价格预测的对数来衡量差异
【torch.clamp()函数会将输出结果中小于下界的值替换为下界,将大于上界的值替换为上界,因此它可以用来对输出结果进行范围限制】
def log_rmse(net, features, labels): # log可以将除法转化为减法
clipped_preds = torch.clamp(net(features), 1, float('inf'))# 对输出进行截断,将小于1的值设置为1,大于float('inf')的值保持不变
rmse = torch.sqrt(loss(torch.log(clipped_preds), torch.log(labels))) # 对预测和实际标签进行log,然后传入损失函数后取根号
return rmse.item()# 返回 张量rmse中的值提取为一个标量
训练函数将借助Adam优化器
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)
optimizer = torch.optim.Adam(net.parameters(), lr=learning_rate,# 使用Adam【对学习率不太敏感】进行优化
weight_decay=weight_decay) # 权重衰减(weight decay)参数【lamdb】,用于控制模型参数的正则化
"""训练"""
for epoch in range(num_epochs):
for X, y in train_iter:
optimizer.zero_grad() # 优化器梯度清0
l = loss(net(X), y) # 计算损失
l.backward() # 反向传播计算梯度
optimizer.step() # 更新优化器参数
train_ls.append(log_rmse(net, train_features, train_labels)) # 更新数据
if test_labels is not None:
test_ls.append(log_rmse(net, test_features, test_labels))
return train_ls, test_ls
K则交叉验证
def get_k_fold_data(k, i, X, y):
assert k > 1
fold_size = X.shape[0] // k # 每一折的大小是样本数/k
X_train, y_train = None, None
for j in range(k):
idx = slice(j * fold_size, (j + 1) * fold_size) # 计算每个切片的起始和终止位置,根据切片索引idx取出相应位置上的数。
X_part, y_part = X[idx, :], y[idx] # 取出相应位置
if j == i: # 如果此时j==i,当前迭代的fold为验证集,则将切片X_part和y_part赋值给X_valid和y_valid。
X_valid, y_valid = X_part, y_part
elif X_train is None: # 如果训练集为空,则将切片X_part和y_part赋值给X_train和y_train
X_train, y_train = X_part, y_part
else: # 否则,将切片X_part和y_part与之前的训练集进行拼接,使用torch.cat()函数进行行拼接,将结果重新赋值给X_train和y_train。
X_train = torch.cat([X_train, X_part], 0)
y_train = torch.cat([y_train, y_part], 0)
# 返回训练集和验证集
return X_train, y_train, X_valid, y_valid
返回训练和验证误差的平均值
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): # 做k次交叉验证
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'fold {i + 1}, train log rmse {float(train_ls[-1]):f}, '
f'valid log rmse {float(valid_ls[-1]):f}')
return train_l_sum / k, valid_l_sum / k # 返回平均测试集和验证集的损失
模型选择
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}')
需要关注valid验证集的损失,需要不断的调整参数实现最小的损失
提交Kaggle预测
def train_and_pred(train_features, test_feature, 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) # 返回训练过程中的训练误差列表train_ls和验证误差列表valid_ls,但在这个函数调用中用下划线 _ 代替了后者
# 绘制并显示训练误差的变化情况
d2l.plot(np.arange(1, num_epochs + 1), [train_ls], xlabel='epoch',
ylabel='log rmse', xlim=[1, num_epochs], yscale='log')
print(f'train log rmse {float(train_ls[-1]):f}')
# 使用训练好的模型net对测试特征进行预测,得到预测结果preds
preds = net(test_features).detach().numpy()
# 预测结果转换为Numpy数组,并将其赋值给测试数据集test_data的'SalePrice'列。
test_data['SalePrice'] = pd.Series(preds.reshape(1, -1)[0])
# 将预测结果和对应的'Id'列组合成一个DataFrame submission
submission = pd.concat([test_data['Id'], test_data['SalePrice']], axis=1)
# 将submission保存为CSV文件submission.csv
submission.to_csv('submission.csv', index=False)
# 调用了train_and_pred()函数,传入相应的参数,执行整个训练和预测的过程
train_and_pred(train_features, test_features, train_labels, test_data,
num_epochs, lr, weight_decay, batch_size)