0
点赞
收藏
分享

微信扫一扫

datawhale数据分析2——数据清洗与特征处理

腊梅5朵 2022-03-17 阅读 52

Task02:数据清洗与特征处理

声明:本文主要参考DataWhale开源学习——动手学数据分析,GitHub地址:https://github.com/datawhalechina/hands-on-data-analysis

一.数据清理与特征工程

import numpy as np
import pandas as pd
path = 'D:/datawhale/hands-on-data-analysis-master/hands-on-data-analysis-master/2/train.csv'
df = pd.read_csv(path)
df.head(3)

PassengerIdSurvivedPclassNameSexAgeSibSpParchTicketFareCabinEmbarked
0103Braund, Mr. Owen Harrismale22.010A/5 211717.2500NaNS
1211Cumings, Mrs. John Bradley (Florence Briggs Th...female38.010PC 1759971.2833C85C
2313Heikkinen, Miss. Lainafemale26.000STON/O2. 31012827.9250NaNS

我们可以观察到,Cabin这一列有缺失值,这就是数据清理要做的工作之一。
数据中一般会有缺失值,一些异常点等,需要经过一定的处理才能继续做后面的分析或建模,所以拿到数据的第一步是进行数据清洗。

1.1 缺失值观察与处理

1.1.1 缺失值观察

查看每列缺失值个数

#方法一
df.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 891 entries, 0 to 890
Data columns (total 12 columns):
 #   Column       Non-Null Count  Dtype  
---  ------       --------------  -----  
 0   PassengerId  891 non-null    int64  
 1   Survived     891 non-null    int64  
 2   Pclass       891 non-null    int64  
 3   Name         891 non-null    object 
 4   Sex          891 non-null    object 
 5   Age          714 non-null    float64
 6   SibSp        891 non-null    int64  
 7   Parch        891 non-null    int64  
 8   Ticket       891 non-null    object 
 9   Fare         891 non-null    float64
 10  Cabin        204 non-null    object 
 11  Embarked     889 non-null    object 
dtypes: float64(2), int64(5), object(5)
memory usage: 83.7+ KB
#方法二
df.isnull().sum()
PassengerId      0
Survived         0
Pclass           0
Name             0
Sex              0
Age            177
SibSp            0
Parch            0
Ticket           0
Fare             0
Cabin          687
Embarked         2
dtype: int64

1.1.2 缺失值处理

df[df['Age']==None] = 0
df[df['Age'].isnull()] = 0 
df[df['Age'] == np.nan] = 0
df.head(6)

PassengerIdSurvivedPclassNameSexAgeSibSpParchTicketFareCabinEmbarked
0103Braund, Mr. Owen Harrismale22.010A/5 211717.2500NaNS
1211Cumings, Mrs. John Bradley (Florence Briggs Th...female38.010PC 1759971.2833C85C
2313Heikkinen, Miss. Lainafemale26.000STON/O2. 31012827.9250NaNS
3411Futrelle, Mrs. Jacques Heath (Lily May Peel)female35.01011380353.1000C123S
4503Allen, Mr. William Henrymale35.0003734508.0500NaNS
5000000.00000.000000
df.dropna().head(3)

PassengerIdSurvivedPclassNameSexAgeSibSpParchTicketFareCabinEmbarked
1211Cumings, Mrs. John Bradley (Florence Briggs Th...female38.010PC 1759971.2833C85C
3411Futrelle, Mrs. Jacques Heath (Lily May Peel)female35.01011380353.1000C123S
5000000.00000.000000
df.fillna(0).head(6)

PassengerIdSurvivedPclassNameSexAgeSibSpParchTicketFareCabinEmbarked
0103Braund, Mr. Owen Harrismale22.010A/5 211717.25000S
1211Cumings, Mrs. John Bradley (Florence Briggs Th...female38.010PC 1759971.2833C85C
2313Heikkinen, Miss. Lainafemale26.000STON/O2. 31012827.92500S
3411Futrelle, Mrs. Jacques Heath (Lily May Peel)female35.01011380353.1000C123S
4503Allen, Mr. William Henrymale35.0003734508.05000S
5000000.00000.000000

【思考】dropna和fillna有哪些参数,分别如何使用呢?

dropna:https://blog.csdn.net/qq_43188358/article/details/108335776

fullna:https://blog.csdn.net/weixin_45456209/article/details/107951433

1.2 重复值观察与处理

1.2.1 查看重复值

df[df.duplicated()]

PassengerIdSurvivedPclassNameSexAgeSibSpParchTicketFareCabinEmbarked
17000000.00000.000
19000000.00000.000
26000000.00000.000
28000000.00000.000
29000000.00000.000
.......................................
859000000.00000.000
863000000.00000.000
868000000.00000.000
878000000.00000.000
888000000.00000.000

176 rows × 12 columns

1.2.2 处理重复值(去重)

df = df.drop_duplicates()
df[df.duplicated()]

PassengerIdSurvivedPclassNameSexAgeSibSpParchTicketFareCabinEmbarked

1.3 特征观察与处理

1.3.1 对离散化数据(年龄)进行分箱操作

pd.cut和pd.qcut

#将连续变量Age平均分箱成5个年龄段,并分别用类别变量12345表示
df['AgeBand'] = pd.cut(df['Age'], 5,labels = [1,2,3,4,5])
df.head()

PassengerIdSurvivedPclassNameSexAgeSibSpParchTicketFareCabinEmbarkedAgeBand
0103Braund, Mr. Owen Harrismale22.010A/5 211717.2500NaNS2
1211Cumings, Mrs. John Bradley (Florence Briggs Th...female38.010PC 1759971.2833C85C3
2313Heikkinen, Miss. Lainafemale26.000STON/O2. 31012827.9250NaNS2
3411Futrelle, Mrs. Jacques Heath (Lily May Peel)female35.01011380353.1000C123S3
4503Allen, Mr. William Henrymale35.0003734508.0500NaNS3
#将连续变量Age划分为(0,5] (5,15] (15,30] (30,50] (50,80]五个年龄段,并分别用类别变量12345表示
df['AgeBand'] = pd.cut(df['Age'],[0,5,15,30,50,80],labels = [1,2,3,4,5])
df.head()

PassengerIdSurvivedPclassNameSexAgeSibSpParchTicketFareCabinEmbarkedAgeBand
0103Braund, Mr. Owen Harrismale22.010A/5 211717.2500NaNS3
1211Cumings, Mrs. John Bradley (Florence Briggs Th...female38.010PC 1759971.2833C85C4
2313Heikkinen, Miss. Lainafemale26.000STON/O2. 31012827.9250NaNS3
3411Futrelle, Mrs. Jacques Heath (Lily May Peel)female35.01011380353.1000C123S4
4503Allen, Mr. William Henrymale35.0003734508.0500NaNS4
#将连续变量Age按10% 30% 50 70% 90%五个年龄段,并用分类变量12345表示
df['AgeBand'] = pd.qcut(df['Age'],[0,0.1,0.3,0.5,0.7,0.9],labels = [1,2,3,4,5])
df.head()

PassengerIdSurvivedPclassNameSexAgeSibSpParchTicketFareCabinEmbarkedAgeBand
0103Braund, Mr. Owen Harrismale22.010A/5 211717.2500NaNS2
1211Cumings, Mrs. John Bradley (Florence Briggs Th...female38.010PC 1759971.2833C85C5
2313Heikkinen, Miss. Lainafemale26.000STON/O2. 31012827.9250NaNS3
3411Futrelle, Mrs. Jacques Heath (Lily May Peel)female35.01011380353.1000C123S4
4503Allen, Mr. William Henrymale35.0003734508.0500NaNS4

1.3.2 对文本变量进行转换

#查看类别文本变量名及种类
#方法一: value_counts
df['Sex'].value_counts()
male      453
female    261
0           1
Name: Sex, dtype: int64
#方法二: unique
df['Sex'].unique()
array(['male', 'female', 0], dtype=object)
df['Sex'].unique()
array(['male', 'female', 0], dtype=object)

unique()是以数组形式(numpy.ndarray)返回列的所有唯一值(特征的所有唯一值)

nunique()是返回unique返回的类型个数(male,female,0)

#将类别文本转换为12345
#方法一: replace
df['Sex_num'] = df['Sex'].replace(['male','female'],[1,2])
df.head()

PassengerIdSurvivedPclassNameSexAgeSibSpParchTicketFareCabinEmbarkedAgeBandSex_num
0103Braund, Mr. Owen Harrismale22.010A/5 211717.2500NaNS21
1211Cumings, Mrs. John Bradley (Florence Briggs Th...female38.010PC 1759971.2833C85C52
2313Heikkinen, Miss. Lainafemale26.000STON/O2. 31012827.9250NaNS32
3411Futrelle, Mrs. Jacques Heath (Lily May Peel)female35.01011380353.1000C123S42
4503Allen, Mr. William Henrymale35.0003734508.0500NaNS41
#方法二: map
df['Sex_num'] = df['Sex'].map({'male': 1, 'female': 2})
df.head()

PassengerIdSurvivedPclassNameSexAgeSibSpParchTicketFareCabinEmbarkedAgeBandSex_num
0103Braund, Mr. Owen Harrismale22.010A/5 211717.2500NaNS21.0
1211Cumings, Mrs. John Bradley (Florence Briggs Th...female38.010PC 1759971.2833C85C52.0
2313Heikkinen, Miss. Lainafemale26.000STON/O2. 31012827.9250NaNS32.0
3411Futrelle, Mrs. Jacques Heath (Lily May Peel)female35.01011380353.1000C123S42.0
4503Allen, Mr. William Henrymale35.0003734508.0500NaNS41.0
#方法三: 使用sklearn.preprocessing的LabelEncoder
from sklearn.preprocessing import LabelEncoder
feat = 'Sex'
lbl = LabelEncoder()  
label_dict = dict(zip(df[feat].unique(), range(df[feat].nunique())))
df[feat + "_labelEncode"] = df[feat].map(label_dict)
df[feat + "_labelEncode"] = lbl.fit_transform(df[feat].astype(str))
df.head()

PassengerIdSurvivedPclassNameSexAgeSibSpParchTicketFareCabinEmbarkedAgeBandSex_numCabin_labelEncodeTicket_labelEncodeSex_labelEncode
0103Braund, Mr. Owen Harrismale22.010A/5 211717.2500NaNS21.01354092
1211Cumings, Mrs. John Bradley (Florence Briggs Th...female38.010PC 1759971.2833C85C52.0744721
2313Heikkinen, Miss. Lainafemale26.000STON/O2. 31012827.9250NaNS32.01355331
3411Futrelle, Mrs. Jacques Heath (Lily May Peel)female35.01011380353.1000C123S42.050411
4503Allen, Mr. William Henrymale35.0003734508.0500NaNS41.01353742

1.3.3 任务三(附加):从纯文本Name特征里提取出Titles的特征(所谓的Titles就是Mr,Miss,Mrs等)

df['Title'] = df.Name.str.extract('([A-Za-z]+)\.', expand=False)
df.head()

PassengerIdSurvivedPclassNameSexAgeSibSpParchTicketFareCabinEmbarkedAgeBandSex_numCabin_labelEncodeTicket_labelEncodeSex_labelEncodeTitle
0103Braund, Mr. Owen Harrismale22.010A/5 211717.2500NaNS21.01354092Mr
1211Cumings, Mrs. John Bradley (Florence Briggs Th...female38.010PC 1759971.2833C85C52.0744721Mrs
2313Heikkinen, Miss. Lainafemale26.000STON/O2. 31012827.9250NaNS32.01355331Miss
3411Futrelle, Mrs. Jacques Heath (Lily May Peel)female35.01011380353.1000C123S42.050411Mrs
4503Allen, Mr. William Henrymale35.0003734508.0500NaNS41.01353742Mr

[A-Za-z]代表任一字母(不分大小写)

+号表示+号前的字符出现几次

.表示点(到.就结束)

举报

相关推荐

0 条评论