导入模块¶

In [34]:
import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
import warnings
%matplotlib inline
warnings.filterwarnings('ignore')
pd.options.display.max_columns = 999

加载数据集¶

In [35]:
df = pd.read_csv('hour.csv')
df.head()
Out[35]:
instant dteday season yr mnth hr holiday weekday workingday weathersit temp atemp hum windspeed casual registered cnt
0 1 2011-01-01 1 0 1 0 0 6 0 1 0.24 0.2879 0.81 0.0 3 13 16
1 2 2011-01-01 1 0 1 1 0 6 0 1 0.22 0.2727 0.80 0.0 8 32 40
2 3 2011-01-01 1 0 1 2 0 6 0 1 0.22 0.2727 0.80 0.0 5 27 32
3 4 2011-01-01 1 0 1 3 0 6 0 1 0.24 0.2879 0.75 0.0 3 10 13
4 5 2011-01-01 1 0 1 4 0 6 0 1 0.24 0.2879 0.75 0.0 0 1 1
In [36]:
# 统计信息
df.describe()
Out[36]:
instant season yr mnth hr holiday weekday workingday weathersit temp atemp hum windspeed casual registered cnt
count 17379.0000 17379.000000 17379.000000 17379.000000 17379.000000 17379.000000 17379.000000 17379.000000 17379.000000 17379.000000 17379.000000 17379.000000 17379.000000 17379.000000 17379.000000 17379.000000
mean 8690.0000 2.501640 0.502561 6.537775 11.546752 0.028770 3.003683 0.682721 1.425283 0.496987 0.475775 0.627229 0.190098 35.676218 153.786869 189.463088
std 5017.0295 1.106918 0.500008 3.438776 6.914405 0.167165 2.005771 0.465431 0.639357 0.192556 0.171850 0.192930 0.122340 49.305030 151.357286 181.387599
min 1.0000 1.000000 0.000000 1.000000 0.000000 0.000000 0.000000 0.000000 1.000000 0.020000 0.000000 0.000000 0.000000 0.000000 0.000000 1.000000
25% 4345.5000 2.000000 0.000000 4.000000 6.000000 0.000000 1.000000 0.000000 1.000000 0.340000 0.333300 0.480000 0.104500 4.000000 34.000000 40.000000
50% 8690.0000 3.000000 1.000000 7.000000 12.000000 0.000000 3.000000 1.000000 1.000000 0.500000 0.484800 0.630000 0.194000 17.000000 115.000000 142.000000
75% 13034.5000 3.000000 1.000000 10.000000 18.000000 0.000000 5.000000 1.000000 2.000000 0.660000 0.621200 0.780000 0.253700 48.000000 220.000000 281.000000
max 17379.0000 4.000000 1.000000 12.000000 23.000000 1.000000 6.000000 1.000000 4.000000 1.000000 1.000000 1.000000 0.850700 367.000000 886.000000 977.000000
In [37]:
# 数据类型信息
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 17379 entries, 0 to 17378
Data columns (total 17 columns):
 #   Column      Non-Null Count  Dtype  
---  ------      --------------  -----  
 0   instant     17379 non-null  int64  
 1   dteday      17379 non-null  object 
 2   season      17379 non-null  int64  
 3   yr          17379 non-null  int64  
 4   mnth        17379 non-null  int64  
 5   hr          17379 non-null  int64  
 6   holiday     17379 non-null  int64  
 7   weekday     17379 non-null  int64  
 8   workingday  17379 non-null  int64  
 9   weathersit  17379 non-null  int64  
 10  temp        17379 non-null  float64
 11  atemp       17379 non-null  float64
 12  hum         17379 non-null  float64
 13  windspeed   17379 non-null  float64
 14  casual      17379 non-null  int64  
 15  registered  17379 non-null  int64  
 16  cnt         17379 non-null  int64  
dtypes: float64(4), int64(12), object(1)
memory usage: 2.3+ MB
In [38]:
# 每个特征中不重复的值
df.apply(lambda x: len(x.unique()))
Out[38]:
instant       17379
dteday          731
season            4
yr                2
mnth             12
hr               24
holiday           2
weekday           7
workingday        2
weathersit        4
temp             50
atemp            65
hum              89
windspeed        30
casual          322
registered      776
cnt             869
dtype: int64

预处理数据集¶

In [39]:
# 检查是否有空值
df.isnull().sum()
Out[39]:
instant       0
dteday        0
season        0
yr            0
mnth          0
hr            0
holiday       0
weekday       0
workingday    0
weathersit    0
temp          0
atemp         0
hum           0
windspeed     0
casual        0
registered    0
cnt           0
dtype: int64
In [40]:
df = df.rename(columns={'weathersit':'weather',
                       'yr':'year',
                       'mnth':'month',
                       'hr':'hour',
                       'hum':'humidity',
                       'cnt':'count'})
df.head()
Out[40]:
instant dteday season year month hour holiday weekday workingday weather temp atemp humidity windspeed casual registered count
0 1 2011-01-01 1 0 1 0 0 6 0 1 0.24 0.2879 0.81 0.0 3 13 16
1 2 2011-01-01 1 0 1 1 0 6 0 1 0.22 0.2727 0.80 0.0 8 32 40
2 3 2011-01-01 1 0 1 2 0 6 0 1 0.22 0.2727 0.80 0.0 5 27 32
3 4 2011-01-01 1 0 1 3 0 6 0 1 0.24 0.2879 0.75 0.0 3 10 13
4 5 2011-01-01 1 0 1 4 0 6 0 1 0.24 0.2879 0.75 0.0 0 1 1
In [41]:
df = df.drop(columns=['instant', 'dteday', 'year'])
In [42]:
# 将 int 列更改为类别
cols = ['season','month','hour','holiday','weekday','workingday','weather']

for col in cols:
    df[col] = df[col].astype('category')
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 17379 entries, 0 to 17378
Data columns (total 14 columns):
 #   Column      Non-Null Count  Dtype   
---  ------      --------------  -----   
 0   season      17379 non-null  category
 1   month       17379 non-null  category
 2   hour        17379 non-null  category
 3   holiday     17379 non-null  category
 4   weekday     17379 non-null  category
 5   workingday  17379 non-null  category
 6   weather     17379 non-null  category
 7   temp        17379 non-null  float64 
 8   atemp       17379 non-null  float64 
 9   humidity    17379 non-null  float64 
 10  windspeed   17379 non-null  float64 
 11  casual      17379 non-null  int64   
 12  registered  17379 non-null  int64   
 13  count       17379 non-null  int64   
dtypes: category(7), float64(4), int64(3)
memory usage: 1.0 MB

探索性数据分析¶

In [43]:
fig, ax = plt.subplots(figsize=(20,10))
sns.pointplot(data=df, x='hour', y='count', hue='weekday', ax=ax)
ax.set(title='工作日和周末的自行车数量')
Out[43]:
[Text(0.5, 1.0, '工作日和周末的自行车数量')]
In [44]:
fig, ax = plt.subplots(figsize=(20,10))
sns.pointplot(data=df, x='hour', y='casual', hue='weekday', ax=ax)
ax.set(title='工作日和周末的自行车数量：未注册用户')
Out[44]:
[Text(0.5, 1.0, '工作日和周末的自行车数量：未注册用户')]
In [45]:
fig, ax = plt.subplots(figsize=(20,10))
sns.pointplot(data=df, x='hour', y='registered', hue='weekday', ax=ax)
ax.set(title='工作日和周末的自行车数量：注册用户')
Out[45]:
[Text(0.5, 1.0, '工作日和周末的自行车数量：注册用户')]
In [46]:
fig, ax = plt.subplots(figsize=(20,10))
sns.pointplot(data=df, x='hour', y='count', hue='weather', ax=ax)
ax.set(title='不同天气下的自行车数量')
Out[46]:
[Text(0.5, 1.0, '不同天气下的自行车数量')]
In [47]:
fig, ax = plt.subplots(figsize=(20,10))
sns.pointplot(data=df, x='hour', y='count', hue='season', ax=ax)
ax.set(title='不同季节下的自行车数量')
Out[47]:
[Text(0.5, 1.0, '不同季节下的自行车数量')]
In [48]:
fig, ax = plt.subplots(figsize=(20,10))
sns.barplot(data=df, x='month', y='count', ax=ax)
ax.set(title='不同月份下的自行车数量')
Out[48]:
[Text(0.5, 1.0, '不同月份下的自行车数量')]
In [49]:
fig, ax = plt.subplots(figsize=(20,10))
sns.barplot(data=df, x='weekday', y='count', ax=ax)
ax.set(title='不同天的自行车数量')
Out[49]:
[Text(0.5, 1.0, '不同天的自行车数量')]
In [50]:
fig, (ax1,ax2) = plt.subplots(ncols=2, figsize=(20,6))
sns.regplot(x=df['temp'], y=df['count'], ax=ax1)
ax1.set(title="气温与用户数量的关系")
sns.regplot(x=df['humidity'], y=df['count'], ax=ax2)
ax2.set(title="湿度与用户数量的关系")
Out[50]:
[Text(0.5, 1.0, '湿度与用户数量的关系')]
In [51]:
from statsmodels.graphics.gofplots import qqplot
fig, (ax1,ax2) = plt.subplots(ncols=2, figsize=(20,6))
sns.distplot(df['count'], ax=ax1)
ax1.set(title='用户数量分布')
qqplot(df['count'], ax=ax2, line='s')
ax2.set(title='理论分位数与样本分位数的比较（QQ图）')
Out[51]:
[Text(0.5, 1.0, '理论分位数与样本分位数的比较（QQ图）')]
In [52]:
df['count'] = np.log(df['count'])

应用对数变换（如np.log(df['count'])），然后重新绘制分布和 QQ（分位数-分位数）图可能很有用，原因如下：

数据的正态性：许多统计技术假设数据服从正态分布。对数变换有助于标准化严重倾斜的变量分布。

稳定方差：对数变换可以稳定数据集的方差。在方差随平均值增加的情况下，应用对数变换可以产生更加同方差的数据集。

线性化关系：转换可以线性化关系，使数据中的模式更易于解释并适合线性建模。

减少异常值的影响：它还可以减少异常值的影响，因为对数转换显着缩小了数据的范围。

In [53]:
fig, (ax1,ax2) = plt.subplots(ncols=2, figsize=(20,6))
sns.distplot(df['count'], ax=ax1)
ax1.set(title='Distribution of the users')
qqplot(df['count'], ax=ax2, line='s')
ax2.set(title='Theoritical quantiles')
Out[53]:
[Text(0.5, 1.0, 'Theoritical quantiles')]

相关矩阵¶

In [54]:
corr = df.corr()
plt.figure(figsize=(15,10))
sns.heatmap(corr, annot=True, annot_kws={'size':15})
Out[54]:
<Axes: >

独热编码¶

In [55]:
pd.get_dummies(df['season'], prefix='season', drop_first=True)
Out[55]:
season_2 season_3 season_4
0 0 0 0
1 0 0 0
2 0 0 0
3 0 0 0
4 0 0 0
... ... ... ...
17374 0 0 0
17375 0 0 0
17376 0 0 0
17377 0 0 0
17378 0 0 0

17379 rows × 3 columns

In [56]:
df_oh = df

def one_hot_encoding(data, column):
    data = pd.concat([data, pd.get_dummies(data[column], prefix=column, drop_first=True)], axis=1)
    data = data.drop([column], axis=1)
    return data

cols = ['season','month','hour','holiday','weekday','workingday','weather']

for col in cols:
    df_oh = one_hot_encoding(df_oh, col)
df_oh.head()
Out[56]:
temp atemp humidity windspeed casual registered count season_2 season_3 season_4 month_2 month_3 month_4 month_5 month_6 month_7 month_8 month_9 month_10 month_11 month_12 hour_1 hour_2 hour_3 hour_4 hour_5 hour_6 hour_7 hour_8 hour_9 hour_10 hour_11 hour_12 hour_13 hour_14 hour_15 hour_16 hour_17 hour_18 hour_19 hour_20 hour_21 hour_22 hour_23 holiday_1 weekday_1 weekday_2 weekday_3 weekday_4 weekday_5 weekday_6 workingday_1 weather_2 weather_3 weather_4
0 0.24 0.2879 0.81 0.0 3 13 2.772589 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
1 0.22 0.2727 0.80 0.0 8 32 3.688879 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
2 0.22 0.2727 0.80 0.0 5 27 3.465736 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
3 0.24 0.2879 0.75 0.0 3 10 2.564949 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
4 0.24 0.2879 0.75 0.0 0 1 0.000000 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
In [57]:
X = df_oh.drop(columns=['atemp', 'windspeed', 'casual', 'registered', 'count'], axis=1)
y = df_oh['count']

模型训练¶

In [58]:
from sklearn.linear_model import LinearRegression, Ridge, HuberRegressor, ElasticNetCV
from sklearn.tree import DecisionTreeRegressor
from sklearn.ensemble import RandomForestRegressor, GradientBoostingRegressor, ExtraTreesRegressor

models = [LinearRegression(),
         Ridge(),
         HuberRegressor(),
         ElasticNetCV(),
         DecisionTreeRegressor(),
         RandomForestRegressor(),
         ExtraTreesRegressor(),
         GradientBoostingRegressor()]
In [59]:
from sklearn import model_selection
def train(model):
    kfold = model_selection.KFold(n_splits=5, shuffle=True, random_state=42) 
    pred = model_selection.cross_val_score(model, X, y, cv=kfold, scoring='neg_mean_squared_error')
    cv_score = pred.mean()
    print('Model:',model)
    print('CV score:', abs(cv_score))

代码中的内容cv_score就像您正在使用的机器学习模型的成绩单。它告诉您模型的表现如何。它的工作原理如下：

交叉验证 (CV)：将您的数据视为一个大馅饼。代码将这个馅饼切成 5 片（因为n_splits=5）。然后，它使用 4 个切片来训练模型，并使用 1 个切片来测试模型。这样做 5 次，每次使用不同的切片进行测试。

评分：每次测试后，模型根据其错误（均方误差）获得分数。但在代码中，这些分数是负数。

平均分数 ( cv_score)：这cv_score是这些测试分数的平均值。我们将负分改为正分（使用abs(cv_score)）以使它们更容易理解。较低的分数意味着模型犯的错误较少，这很好！

因此，cv_score平均分数表明您的模型的预测效果如何。它的值越低越好。

In [60]:
for model in models:
    train(model)

Model: LinearRegression()
CV score: 0.44849511159541205
Model: Ridge()
CV score: 0.4484090089563206
Model: HuberRegressor()
CV score: 0.46596807512124105
Model: ElasticNetCV()
CV score: 0.45614918135359145
Model: DecisionTreeRegressor()
CV score: 0.44255199359646225
Model: RandomForestRegressor()
CV score: 0.23279282002190094
Model: ExtraTreesRegressor()
CV score: 0.23485168754583902
Model: GradientBoostingRegressor()
CV score: 0.35702811006978274

线性回归：基本回归模型，CV分数为0.4485，表示平均误差。

岭回归：与线性回归类似，但经过正则化，误差稍低，为 0.4484。

Huber 回归器：一个对异常值具有鲁棒性的模型，CV 得分为 0.4660，表明它对此数据集可能不那么有效。

ElasticNetCV：结合L1和L2正则化，CV得分为0.4561。

决策树回归器：非线性模型，CV 得分为 0.4426。

随机森林回归器：决策树的集合，显示出明显更好的 CV 分数 0.2328。

Extra Trees Regressor：与随机森林类似，但 CV 分数稍好，为 0.2349。

Gradient Boosting Regressor：一个专注于纠正其上一个子模型错误的集成模型，CV 得分为 0.3570。

CV 分数越低表明模型性能越好。RandomForest 和 ExtraTrees 回归器显示了这些模型中的最佳结果。

In [61]:
from sklearn.model_selection import train_test_split
x_train, x_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=42)
In [65]:
model = RandomForestRegressor()
model.fit(x_train, y_train)
y_pred = model.predict(x_test)
In [66]:
# 绘制误差差
error = y_test - y_pred
fig, ax = plt.subplots()
ax.scatter(y_test, error)
ax.axhline(lw=3, color='black')
ax.set_xlabel('Observed')
ax.set_ylabel('Error')
plt.show()
In [67]:
from sklearn.metrics import mean_squared_error
np.sqrt(mean_squared_error(y_test, y_pred))
Out[67]:
0.48527134611361483