pandas简单操作（1）：创建、删除、处理等

pandas简单操作（1）：创建、删除、处理等

input:

import pandas as pd
a = pd.Series([1,3,4,5,6])
print(a)
import os
print(os.path.abspath('.'))

output:

	    0    1
	    1    3
	    2    4
	    3    5
	    4    6
	    dtype: int64
	    /Users/mac

input:

datas_1 = pd.date_range(start ='20160301',end = '20160306')
datas = pd.date_range(start ='20160301',periods = 6)
datas_1

output:

DatetimeIndex(['2016-03-01', '2016-03-02', '2016-03-03', '2016-03-04',
               '2016-03-05', '2016-03-06'],
              dtype='datetime64[ns]', freq='D')

input:

import numpy as np
data = pd.DataFrame(np.random.randn(6,4), index = datas, columns = list('ABCD'))
#data.shape
#data.values
data

output:

	A	B	C	D
2016-03-01	-0.271952	-1.217686	-0.037302	-0.179983
2016-03-02	-1.803772	0.689236	-2.532408	-0.819420
2016-03-03	-0.596284	0.747474	-2.169957	0.173793
2016-03-04	0.349248	-1.013605	0.963074	-0.306067
2016-03-05	0.981165	0.348213	0.011119	0.721957
2016-03-06	-0.576315	1.312909	-0.407155	0.793780

input:

#data.head()
data.head(2) #只查看前两行
#data.tail()
data.tail(2)#只看最后两行

output:

	A	B	C	D
2016-03-05	0.981165	0.348213	0.011119	0.721957
2016-03-06	-0.576315	1.312909	-0.407155	0.793780

input:

data.index #查看行标签
data.columns #查看列标签
data.describe() #count:几位有效数字 mean:平均值 std:方差 min:最小值 25%：4分位

output:

	A	B	C	D
count	6.000000	6.000000	6.000000	6.000000
mean	-0.319652	0.144424	-0.695438	0.064010
std	0.947295	1.026036	1.364903	0.625212
min	-1.803772	-1.217686	-2.532408	-0.819420
25%	-0.591291	-0.673150	-1.729257	-0.274546
50%	-0.424133	0.518724	-0.222229	-0.003095
75%	0.193948	0.732915	-0.000986	0.584916
max	0.981165	1.312909	0.963074	0.793780

input:

data.T #转置

output:

	2016-03-01 00:00:00	2016-03-02 00:00:00	2016-03-03 00:00:00	2016-03-04 00:00:00	2016-03-05 00:00:00	2016-03-06 00:00:00
A	-0.271952	-1.803772	-0.596284	0.349248	0.981165	-0.576315
B	-1.217686	0.689236	0.747474	-1.013605	0.348213	1.312909
C	-0.037302	-2.532408	-2.169957	0.963074	0.011119	-0.407155
D	-0.179983	-0.819420	0.173793	-0.306067	0.721957	0.793780

input:

data.sort_index(axis=1,ascending=False) #根据列标签排序（降序）
data.sort_index(axis=0,ascending=True) #根据列标签排序（升序）
data.sort_values(by='A') #根据A列的数据排序（升序）

output:

	A	B	C	D
2016-03-02	-1.803772	0.689236	-2.532408	-0.819420
2016-03-03	-0.596284	0.747474	-2.169957	0.173793
2016-03-06	-0.576315	1.312909	-0.407155	0.793780
2016-03-01	-0.271952	-1.217686	-0.037302	-0.179983
2016-03-04	0.349248	-1.013605	0.963074	-0.306067
2016-03-05	0.981165	0.348213	0.011119	0.721957

input:

data['A'] #数据选择 选择一个列 或者data.A 
data[2:4] #选择2-4行
data.iloc[2:4] #这个效率比较高，用这个，（iloc）只认数字
data['20160302':'20160305'] #按行选择
data.loc['20160302':'20160305'] #这个效率比较高，用这个，（loc）只认标签

output:

	A	B	C	D
2016-03-03	-0.596284	0.747474	-2.169957	0.173793
2016-03-04	0.349248	-1.013605	0.963074	-0.306067

input:

data.loc[:,['B','C']] #只选择B-C两列的数据
data.loc['20160302':'20160304',['B','C']] #选择B-C两列,20160302-0304之间的数据

output:

	B	C
2016-03-02	0.689236	-2.532408
2016-03-03	0.747474	-2.169957
2016-03-04	-1.013605	0.963074

data.loc['20160302','B'] #获得表格中的具体的一个值
data.at[pd.Timestamp('20160302'),'B'] #这个效率更高，用这个，但是得传入时间戳的格式

output:

0.6892360151598519

input:

data.iat[0,0]=100 #修改单个数据
data.A = range(6) #修改一列数据
data.A = 100 #修改一列数据

d = {'A':1,'B':pd.Timestamp('20201106'),'C':range(4),'D':np.arange(4)}
d

output:

{'A': 1,
 'B': Timestamp('2020-11-06 00:00:00'),
 'C': range(0, 4),
 'D': array([0, 1, 2, 3])}

input:

df = pd.DataFrame(d)
df

output:

	A	B	C	D
0	1	2020-11-06	0	0
1	1	2020-11-06	1	1
2	1	2020-11-06	2	2
3	1	2020-11-06	3	3

input:

df.dtypes
df.A

output:

0    1
1    1
2    1
3    1
Name: A, dtype: int64