map-apply-applymap

In [1]:

import warnings
import math
import pandas as pd
import numpy as np
import matplotlib

warnings.filterwarnings('ignore')
pd.options.display.max_rows = 100
pd.options.display.max_columns = 100
pd.set_option('max_colwidth', 500)

get_ipython().magic(u'matplotlib inline')
matplotlib.style.use('ggplot')

from matplotlib import pyplot as plt
plt.rcParams['font.sans-serif'] = ['SimHei']
plt.rcParams['axes.unicode_minus'] = False

myfont = matplotlib.font_manager.FontProperties(fname=u'simsun.ttc', size=14)

In [11]:

data = pd.DataFrame({
    'age' : np.random.randint(15, 100, 100),
    'height':np.random.randint(140, 180, 100),
    'weight':np.random.randint(40, 80, 100),
    'gender':np.random.randint(0,2, 100),
    'salary':np.random.randint(3000, 30000, 100)
})
data.head()

Out[11]:

 

	age	height	weight	gender	salary
0	68	140	58	0	15988
1	22	160	59	0	5807
2	90	160	60	0	21741
3	71	178	45	0	14133
4	61	165	59	1	14346

In [12]:

# series map:值转换
data['gender'] = data.gender.map({0:'man', 1:'women'})
data.head()

Out[12]:

 

	age	height	weight	gender	salary
0	68	140	58	man	15988
1	22	160	59	man	5807
2	90	160	60	man	21741
3	71	178	45	man	14133
4	61	165	59	women	14346

In [14]:

# series map: 也可以使用字符串format
data.gender.map('i am a {} '.format).head()

Out[14]:

0      i am a man 
1      i am a man 
2      i am a man 
3      i am a man 
4    i am a women 
Name: gender, dtype: object

In [19]:

# series apply: 对序列的每个值做相应操作
data.salary = data.salary.apply(lambda x: x* 1.3)
data.head()

Out[19]:

 

	age	height	weight	gender	salary
0	68	140	58	man	35125.636
1	22	160	59	man	12757.979
2	90	160	60	man	47764.977
3	71	178	45	man	31050.201
4	61	165	59	women	31518.162

In [33]:

# series apply: 对序列的每个值做相应操作，可传递复杂参数和函数
def fun1(x, bias):
    return x * bias
data.salary = data.salary.apply(fun1, args=(1.3,))
data.head()

Out[33]:

 

	age	height	weight	gender	salary
0	68	140	58	man	45663.3268
1	22	160	59	man	16585.3727
2	90	160	60	man	62094.4701
3	71	178	45	man	40365.2613
4	61	165	59	women	40973.6106

In [36]:

# series apply: 对序列的每个值做相应操作，可传递复杂参数和函数
def fun2(x, **kwargs):
    for month in kwargs:
        x += kwargs[month] * 100
    return x 
data.salary = data.salary.apply(fun2, june=30, july=20, august=25)
data.head()

Out[36]:

 

	age	height	weight	gender	salary
0	68	140	58	man	53238.3268
1	22	160	59	man	24160.3727
2	90	160	60	man	69669.4701
3	71	178	45	man	47940.2613
4	61	165	59	women	48548.6106

In [28]:

# dataframe apply:根据函数对行或者列应用对应操作
data[['age','height', 'weight']].apply(np.sqrt).head()

Out[28]:

 

	age	height	weight
0	8.246211	11.832160	7.615773
1	4.690416	12.649111	7.681146
2	9.486833	12.649111	7.745967
3	8.426150	13.341664	6.708204
4	7.810250	12.845233	7.681146

In [37]:

# dataframe apply:根据函数对行或者列应用对应操作
data[['age','height', 'weight']].apply(np.sum).head()

Out[37]:

age        5818
height    16225
weight     6081
dtype: int64

In [29]:

# dataframe apply:根据函数对行应用对应操作，做比较复杂计算
def BMI(series):
    height = series['height']/100.0
    weight = series['weight']
    return weight / height ** 2

data.apply(BMI, axis = 1).head()

Out[29]:

0    29.591837
1    23.046875
2    23.437500
3    14.202752
4    21.671258
dtype: float64

In [43]:

data[['age','height', 'weight']].head()

Out[43]:

 

	age	height	weight
0	68	140	58
1	22	160	59
2	90	160	60
3	71	178	45
4	61	165	59

In [44]:

# dataframe applymap:对每个元素操作
data[['age','height', 'weight']].applymap(lambda x:x-2).head()

Out[44]:

 

	age	height	weight
0	66	138	56
1	20	158	57
2	88	158	58
3	69	176	43
4	59	163	57