import hashlib
import os
import tarfile
import zipfile
import requests
#这一段是用来下载数据集的————不用看
DATA_HUB = dict()
DATA_URL = 'http://d2l-data.s3-accelerate.amazonaws.com/'
def download(name, cache_dir=os.path.join('..', 'data')):
"""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}."
url, sha1_hash = DATA_HUB[name]
os.makedirs(cache_dir, exist_ok=True)
fname = os.path.join(cache_dir, url.split('/')[-1])
if os.path.exists(fname):
sha1 = hashlib.sha1()
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
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):
"""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():
"""下载DATA_HUB中的所有文件。"""
for name in DATA_HUB:
download(name)
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'] = (
DATA_URL + 'kaggle_house_pred_train.csv',
'585e9cc93e70b39160e7921475f9bcd7d31219ce')
DATA_HUB['kaggle_house_test'] = (
DATA_URL + 'kaggle_house_pred_test.csv',
'fa19780a7b011d9b009e8bff8e99922a8ee2eb90')
#代码从这里开始阅读
#使用pandas读入并处理数据——读取数据并且预处理
train_data = pd.read_csv(download('kaggle_house_train'))
test_data = pd.read_csv(download('kaggle_house_test'))
#可以查看下数据的基本结构
#iloc可以查看csv文件具体内容
print(train_data.iloc[0:4, [0, 1, 2, 3, -3, -2, -1]])
#数据预处理
#pd.concat见函数解析:
all_features =pd.concat((train_data.iloc[:, 1:-1], test_data.iloc[:, 1:]))
#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
#将非字符串的特征取出来放在numeric_features
numeric_features = all_features.dtypes[all_features.dtypes != 'object'].index
#将数值特征的均值变成0方差变成1
#lambda函数见详解
all_features[numeric_features] = all_features[numeric_features].apply(
lambda x: (x - x.mean()) / (x.std()))
#将缺失值全部fill(填充)为0
all_features[numeric_features] = all_features[numeric_features].fillna(0)
#字符串特征的处理——“用one-hot编码”( pd.get_dummies)
all_features = pd.get_dummies(all_features, dummy_na=True)
# dummy_na对缺失值进行一个处理——如果有缺失值就再增加一个特征,缺失就填入1,不缺失就填入0
all_features.shape
#最后展现出来的是有331个特征,特征增加的主要原因是进行了one-hot编码(原本只有81个)
#前面是把训练集与数据集进行了统一的处理,现在我们要将其分开并且转换成tensor
n_train = train_data.shape[0] #取的是训练集中样本的个数————shape[0]取得是csv的行数
train_features = torch.tensor(all_features[:n_train].values, #all_features[:n_train].values是一个numpy的结构,现在要将其变成一个tensor
dtype=torch.float32)
test_features = torch.tensor(all_features[n_train:].values,
dtype=torch.float32)
print(train_data.SalePrice.iloc[0:2])
#reshape(-1,1)见函数详解
train_labels = torch.tensor(train_data.SalePrice.values.reshape(-1, 1),
dtype=torch.float32)
loss = nn.MSELoss() #在这里我们是一个回归的问题所于用的是均方误差
in_features = train_features.shape[1] #去特征数
#用了一个很简单的线性回归,这是一个单层的网络
def get_net():
net = nn.Sequential(nn.Linear(in_features, 1))
return net
我们更关心“相对误差” (𝑦−𝑦̂)/ 𝑦 , 解决这个问题的一种方法是用价格预测的对数来衡量差异
def log_rmse(net, features, labels):
#clamp见函数详解
clipped_preds = torch.clamp(net(features), 1, float('inf')) # float('inf')代表的是正无穷
rmse = torch.sqrt(loss(torch.log(clipped_preds), torch.log(labels)))
return rmse.item()
#定义训练的
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,
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
#k折交叉验证
#这个函数做的就是将训练集与测试集分开,比如取出第二份做测试集的话就将其他四份cat(拼接)起来
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):
#slice见函数详解
idx = slice(j * fold_size, (j + 1) * fold_size)
#训练集被分成k折,每折的大小是fold_size,我们要将每折取出来就要用到slice
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:
#torch.cat见函数详解
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'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
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}')
函数解析:
1、iloc[,]
选中csv中你所要的东西
2、pd.concat
concat是将两个矩阵以行相加的形式进行拼接的
79怎么来的 81-2:训练集减去id和目标值 80-1:测试集减去目标值
将我们需要的数据用iloc选中,并且用concat拼接起来
3、lambda :函数
这个就相当于函数的简单定义
exmaple:
def sum(x,y):
return x+y
print(sum(1,2))
sum = lambda x,y : x+y
print(sum(1,2))
:前面的是输入,后面的是你要做的事(函数)
4、对于图像来说shape[0],shape[1],shape[2]分别代表的是图像的垂直高度,水平宽度,通道数
对于matrix来说shape[0],shape[1]分别代表的是matrix的行与列
5、torch.tensor(xxxx.values)——这样就可以将matrix转换成为一个矩阵
6、
reshape(-1, 1):将“整个”matrix变成列
reshape(1, -1):将“整个”matrix变成行
7、torch.clamp torch.clamp(input, min, max, out=None) → Tensor 将输入input张量每个元素的夹紧到区间 [min,max][min,max],并返回结果到一个新张量。 在这里就是将net的输出夹紧到[1,正无穷],不然等下求log的时候会出现问题
8、slice
slice() 函数实现切片对象,主要用在切片操作函数里的参数传递。
example:
>>>myslice = slice(5) # 设置截取5个元素的切片
>>> myslice
slice(None, 5, None)
>>> arr = range(10)
>>> arr
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
>>> arr[myslice] # 截取 5 个元素
[0, 1, 2, 3, 4]
>>>
9、torch.cattorch.cat是将两个张量(tensor)拼接在一起,cat是concatenate的意思,即拼接,联系在一起。
C=torch.cat((A,B),0)#按维数0(行)拼接——A、B的列数必须相同
C=torch.cat((A,B),1)#按维数1(列)拼接——A,B的行数必须相同