数据分析可视化pandas

Series

import pandas as pd
1

obj=pd.Series([1,2,-3,-4])
print(obj)
1
2

0 1
1 2
2 -3
3 -4
dtype: int64

i=["a","c","d","a"]
v=[2,4,5,7]
t=pd.Series(data=v,index=i,name="col")
print(t)
1
2
3
4

a 2
c 4
d 5
a 7
Name: col, dtype: int64

val = [2, 4, 5, 6]
idx1 = range(10, 14)
idx2 = "hello the cruel world".split()
s0 = pd.Series(val)
s1 = pd.Series(val, index = idx1)
t = pd.Series(val, index = idx2)
print(s0)
1
2
3
4
5
6
7

0 2
1 4
2 5
3 6
dtype: int64

print(s1)
1

10 2
11 4
12 5
13 6
dtype: int64

print(t)
1

hello 2
the 4
cruel 5
world 6
dtype: int64

print(s0.index)
1

RangeIndex(start=0, stop=4, step=1)

print(s1.index)
1

RangeIndex(start=10, stop=14, step=1)

print(t.index)
1

Index([‘hello’, ‘the’, ‘cruel’, ‘world’], dtype=‘object’)

print(s0[0])
1

2

print(s1[10])
1

2

print('default:',t[0],'label:' ,t["hello"])
1

default: 2 label: 2

sdata = {'Ohio': 35000, 'Texas': 71000, 'Oregon': 16000, 'Utah': 5000}
obj3 = pd.Series(sdata)
print(obj3)
1
2
3

Ohio 35000
Texas 71000
Oregon 16000
Utah 5000
dtype: int64

通过字典创建

sdata = {"a" : 100, "b" : 200, "e" : 300}
letter = ["a","b","c","e" ]
obj =  pd.Series(sdata, index = letter)
print(obj)
1
2
3
4

a 100.0
b 200.0
c NaN
e 300.0
dtype: float64

键值和指定的索引不匹配

sdata = {'Ohio': 35000, 'Texas': 71000, 'Oregon': 16000, 'Utah': 5000}
obj1 = pd.Series(sdata)
states = ['California', 'Ohio', 'Oregon', 'Texas']
obj2 = pd.Series(sdata, index = states)
print('obj1:\n',obj1)
print('obj2:\n',obj2)
1
2
3
4
5
6

obj1:
Ohio 35000
Texas 71000
Oregon 16000
Utah 5000
dtype: int64
obj2:
California NaN
Ohio 35000.0
Oregon 16000.0
Texas 71000.0
dtype: float64

不同索引数据的自动对齐

print(obj1+obj2)
1

California NaN
Ohio 70000.0
Oregon 32000.0
Texas 142000.0
Utah NaN
dtype: float64

Series 索引的修改

obj = pd.Series([4,7,-3,2])
print("修改前：\n",obj)
obj.index = ['Bob', 'Steve', 'Jeff', 'Ryan']
print("修改后：\n",obj)
1
2
3
4

修改前：
0 4
1 7
2 -3
3 2
dtype: int64
修改后：
Bob 4
Steve 7
Jeff -3
Ryan 2
dtype: int64

dataFrame

dataFrame 的创建

data = {
    'name':['张三', '李四', '王五', '小明'],
    'sex':['female', 'female', 'male', 'male'],
    'year':[2001, 2001, 2003, 2002],
    'city':['北京', '上海', '广州', '北京']
}
df=pd.DataFrame(data)
print(df)
1
2
3
4
5
6
7
8

name sex year city
0 张三 female 2001 北京
1 李四 female 2001 上海
2 王五 male 2003 广州
3 小明 male 2002 北京

索引

df1=pd.DataFrame(data,columns=['year','name','sex','city'])
print(df1)
1
2

.
. year name sex city
0 2001 张三 female 北京
1 2001 李四 female 上海
2 2003 王五 male 广州
3 2002 小明 male 北京

缺失值

df2=pd.DataFrame(data,columns=['year','name','sex','city','address'])
print(df2)
1
2

.
. year name sex city address
0 2001 张三 female 北京 NaN
1 2001 李四 female 上海 NaN
2 2003 王五 male 广州 NaN
3 2002 小明 male 北京 NaN

指定列名

df3=pd.DataFrame(data,columns=['year','name','sex','city'],index=['a','b','c','d'])
print(df3)
1
2

.
. year name sex city
a 2001 张三 female 北京
b 2001 李四 female 上海
c 2003 王五 male 广州
d 2002 小明 male 北京

索引对象 属性

print(df3)
print(df3.index)
print(df3.columns)
1
2
3

.
. year name sex city
a 2001 张三 female 北京
b 2001 李四 female 上海
c 2003 王五 male 广州
d 2002 小明 male 北京
Index([‘a’, ‘b’, ‘c’, ‘d’], dtype=‘object’)
Index([‘year’, ‘name’, ‘sex’, ‘city’], dtype=‘object’)

print('name' in df3.columns)
print('a' in df3.index)
1
2

True
True

print(df3.shape)
1

(4, 4)

print(df3.ndim)
1

2

print(df3.size)
1

16

print(df3.values)
1

[[2001 ‘张三’ ‘female’ ‘北京’]
[2001 ‘李四’ ‘female’ ‘上海’]
[2003 ‘王五’ ‘male’ ‘广州’]
[2002 ‘小明’ ‘male’ ‘北京’]]

df
1

插入索引值

df.index
1

RangeIndex(start=0, stop=4, step=1)

df.index.insert(1,'w')
1

Index([0, ‘w’, 1, 2, 3], dtype=‘object’)

索引操作

重建索引

obj = pd.Series([7.2,-4.3,4.5,3.6],index = ['b', 'a', 'd', 'c'])
print(obj)
1
2

b 7.2
a -4.3
d 4.5
c 3.6
dtype: float64

# 重建索引
obj.reindex(['a','b','c','d','e'])
1
2

a -4.3
b 7.2
c 3.6
d 4.5
e NaN
dtype: float64

# 重建索引时填充缺失值
obj.reindex(['a','b','c','d','e'],fill_value=4)
1
2

a -4.3
b 7.2
c 3.6
d 4.5
e 4.0
dtype: float64

import numpy as np
obj1=pd.Series(['blue','red','black'],index=[0,2,4])
obj1
1
2
3

0 blue
2 red
4 black
dtype: object

obj1.reindex(np.arange(6))
1

0 blue
1 NaN
2 red
3 NaN
4 black
5 NaN
dtype: object

obj1.reindex(np.arange(6),method='ffill')
1

0 blue
1 blue
2 red
3 red
4 black
5 black
dtype: object

obj1.reindex(np.arange(6),method='bfill')
1

0 blue
1 red
2 red
3 black
4 black
5 NaN
dtype: object

df4=pd.DataFrame(np.arange(9).reshape(3,3),index=['a','c','d'],columns=['one','two','four'])
df4
1
2

print(df4)
1

.one two four
a 0 1 2
c 3 4 5
d 6 7 8

df4.reindex(index=['a','b','c','d'],columns=['one','two','three','four'])
1

df4.reindex(index=['a','b','c','d'],columns=['one','two','three','four'],fill_value=3)
1

df4
1

df4.reindex(index=['a','b','c','d'],method='ffill')
1

df4.reindex(index=['a','b','c','d'],method='bfill')
1

df1
1

df5=df1.set_index('city')# 索引为city
df5
1
2

df5.reset_index()
1

df1
1

df5
1

df5[['name','year']]
1

# 选择非‘int64’类型的列显示
display(df5.select_dtypes(exclude='int64').head())

1
2
3

# 选择‘int64’类型的列显示
df5.select_dtypes(include=['int64']).head()
1
2

# 选择‘int64’类型的列显示
df5.select_dtypes(include=['int64']).head(3)
1
2

df
1

df[:2]
1

df[1:3]
1

df[1:]
1

df.head(3) # head(3)前3行
1

df.head()  # head() 前5行
1

df.tail(3)
1

df.sample(3)
1

df.sample(frac=0.8) # 前80%行
1

df5
1

# .loc根据行和列筛选
df5.loc[:,['name','year']]  
1
2

df5.loc[['北京','上海'],['name','year']]
1

df5.loc[df5['year']<=2002,['name','year']]
1

df5
1

df5.iloc[:,2]
1

city
北京 female
上海 female
广州 male
北京 male
Name: sex, dtype: object

df5.iloc[[1,3]]
1

df5.iloc[[1,3],:2]
1

df5.iloc[[1,3],[0,2]]
1

df5
1

df5.query('year>2001')
1

df5.query('year>2001 & year<2003')
1

df5[df5['year']==2001]
1

df5.query('year==2001')
1

df
data1 = {'city':'兰州','name':'李红','year':2005,'sex':'female'}
# df.append(datal,ignore_index=True)
1
2
3

df
1

# 增加列
df['score']=[56,85,98,78]
df
1
2
3

# 指定索引位置的后面插入列
df.insert(1,'No',['001','002','003','004'])
df
1
2
3

# 删除数据
print(df5)
print(df5.drop('广州'))
print(df5)  # 并没有在原数据修改
1
2
3
4

.
year name sex
city
北京 2001 张三 female
上海 2001 李四 female
广州 2003 王五 male
北京 2002 小明 male
year name sex
city
北京 2001 张三 female
上海 2001 李四 female
北京 2002 小明 male
year name sex
city
北京 2001 张三 female
上海 2001 李四 female
广州 2003 王五 male
北京 2002 小明 male

1

# 删除列
# df.drop('score',axis=1,inplace=True)
# df
1
2
3

# 修改列(直接修改原数据)
df['year']=2001
df
1
2
3

# 替换值
df.replace('张三','李明',inplace=True)
df
1
2
3

# 替换多个值
df.replace({'李明':'张山山','王五':'小五'},inplace=True)
df
1
2
3

# 修改列名,要指定
df.rename(columns={'No':'number'},inplace=True)
df
1
2
3