Pandas Learning

pandas网站链接

Object Creation

import pandas as pd
import numpy as np

s = pd.Series([1,3,5, np.NAN,6,8 ])

Series is just like the list in python

s

0    1.0
1    3.0
2    5.0
3    NaN
4    6.0
5    8.0
dtype: float64

s[1]

3.0

Creating a DataFrame by passing a NumPy array, with a datetime index and labeled columns:

dates = pd.date_range('20130101', periods=6)
dates

DatetimeIndex(['2013-01-01', '2013-01-02', '2013-01-03', '2013-01-04',
               '2013-01-05', '2013-01-06'],
              dtype='datetime64[ns]', freq='D')

df = pd.DataFrame(np.random.randn(6, 4), index=dates, columns=list('ABCD'))
df

	A	B	C	D
2013-01-01	-1.815560	-2.066970	0.083446	0.457541
2013-01-02	0.878679	-1.279713	-2.170213	-0.317302
2013-01-03	-2.504659	0.727344	-0.062560	1.473940
2013-01-04	0.245661	1.158579	0.431938	-0.511382
2013-01-05	0.159338	1.616634	-0.567605	-0.221172
2013-01-06	-0.758477	-1.364958	-0.398516	-0.241077

df2 = pd.DataFrame({'A': 1.,
                     'B': pd.Timestamp('20130102'),
                     'C': pd.Series(1, index=list(range(4)), dtype='float32'),
                     'D': np.array([3] * 4, dtype='int32'),
                     'E': pd.Categorical(["test", "train", "test", "train"]),
                     'F': 'foo'})
df2

	A	B	C	D	E	F
0	1.0	2013-01-02	1.0	3	test	foo
1	1.0	2013-01-02	1.0	3	train	foo
2	1.0	2013-01-02	1.0	3	test	foo
3	1.0	2013-01-02	1.0	3	train	foo

df2.dtypes

A           float64
B    datetime64[ns]
C           float32
D             int32
E          category
F            object
dtype: object

View Data

df.head()

	A	B	C	D
2013-01-01	-1.815560	-2.066970	0.083446	0.457541
2013-01-02	0.878679	-1.279713	-2.170213	-0.317302
2013-01-03	-2.504659	0.727344	-0.062560	1.473940
2013-01-04	0.245661	1.158579	0.431938	-0.511382
2013-01-05	0.159338	1.616634	-0.567605	-0.221172

df.tail(2)

	A	B	C	D
2013-01-05	0.159338	1.616634	-0.567605	-0.221172
2013-01-06	-0.758477	-1.364958	-0.398516	-0.241077

df.index

DatetimeIndex(['2013-01-01', '2013-01-02', '2013-01-03', '2013-01-04',
               '2013-01-05', '2013-01-06'],
              dtype='datetime64[ns]', freq='D')

df.columns

Index(['A', 'B', 'C', 'D'], dtype='object')

df.to_xarray

<bound method NDFrame.to_xarray of                    A         B         C         D
2013-01-01 -1.815560 -2.066970  0.083446  0.457541
2013-01-02  0.878679 -1.279713 -2.170213 -0.317302
2013-01-03 -2.504659  0.727344 -0.062560  1.473940
2013-01-04  0.245661  1.158579  0.431938 -0.511382
2013-01-05  0.159338  1.616634 -0.567605 -0.221172
2013-01-06 -0.758477 -1.364958 -0.398516 -0.241077>

df.describe()

	A	B	C	D
count	6.000000	6.000000	6.000000	6.000000
mean	-0.632503	-0.201514	-0.447252	0.106758
std	1.311665	1.550118	0.915211	0.746014
min	-2.504659	-2.066970	-2.170213	-0.511382
25%	-1.551289	-1.343647	-0.525333	-0.298246
50%	-0.299569	-0.276185	-0.230538	-0.231124
75%	0.224080	1.050770	0.046944	0.287863
max	0.878679	1.616634	0.431938	1.473940

df.T

	2013-01-01 00:00:00	2013-01-02 00:00:00	2013-01-03 00:00:00	2013-01-04 00:00:00	2013-01-05 00:00:00	2013-01-06 00:00:00
A	-1.815560	0.878679	-2.504659	0.245661	0.159338	-0.758477
B	-2.066970	-1.279713	0.727344	1.158579	1.616634	-1.364958
C	0.083446	-2.170213	-0.062560	0.431938	-0.567605	-0.398516
D	0.457541	-0.317302	1.473940	-0.511382	-0.221172	-0.241077

df.sort_values(by = 'B')

	A	B	C	D
2013-01-01	-1.815560	-2.066970	0.083446	0.457541
2013-01-06	-0.758477	-1.364958	-0.398516	-0.241077
2013-01-02	0.878679	-1.279713	-2.170213	-0.317302
2013-01-03	-2.504659	0.727344	-0.062560	1.473940
2013-01-04	0.245661	1.158579	0.431938	-0.511382
2013-01-05	0.159338	1.616634	-0.567605	-0.221172

df.sort_index(axis=1, ascending=False)

	D	C	B	A
2013-01-01	0.457541	0.083446	-2.066970	-1.815560
2013-01-02	-0.317302	-2.170213	-1.279713	0.878679
2013-01-03	1.473940	-0.062560	0.727344	-2.504659
2013-01-04	-0.511382	0.431938	1.158579	0.245661
2013-01-05	-0.221172	-0.567605	1.616634	0.159338
2013-01-06	-0.241077	-0.398516	-1.364958	-0.758477

Selection

df['A']

2013-01-01   -1.815560
2013-01-02    0.878679
2013-01-03   -2.504659
2013-01-04    0.245661
2013-01-05    0.159338
2013-01-06   -0.758477
Freq: D, Name: A, dtype: float64

df[0:3]

	A	B	C	D
2013-01-01	-1.815560	-2.066970	0.083446	0.457541
2013-01-02	0.878679	-1.279713	-2.170213	-0.317302
2013-01-03	-2.504659	0.727344	-0.062560	1.473940

df['20130102':'20130104']

	A	B	C	D
2013-01-02	0.878679	-1.279713	-2.170213	-0.317302
2013-01-03	-2.504659	0.727344	-0.062560	1.473940
2013-01-04	0.245661	1.158579	0.431938	-0.511382

df.loc[dates[0]]

A   -1.815560
B   -2.066970
C    0.083446
D    0.457541
Name: 2013-01-01 00:00:00, dtype: float64

 df.loc['20130102':'20130104', ['A', 'B']]

	A	B
2013-01-02	0.878679	-1.279713
2013-01-03	-2.504659	0.727344
2013-01-04	0.245661	1.158579

df.loc[dates[0], 'A']

-1.8155597588741252

df.iloc[3]

A    0.245661
B    1.158579
C    0.431938
D   -0.511382
Name: 2013-01-04 00:00:00, dtype: float64

df[df.A > 0]

	A	B	C	D
2013-01-02	0.878679	-1.279713	-2.170213	-0.317302
2013-01-04	0.245661	1.158579	0.431938	-0.511382
2013-01-05	0.159338	1.616634	-0.567605	-0.221172

Missing data

df1 = df.reindex(index=dates[0:4], columns=list(df.columns) + ['E'])
df1

	A	B	C	D	E
2013-01-01	-1.815560	-2.066970	0.083446	0.457541	NaN
2013-01-02	0.878679	-1.279713	-2.170213	-0.317302	NaN
2013-01-03	-2.504659	0.727344	-0.062560	1.473940	NaN
2013-01-04	0.245661	1.158579	0.431938	-0.511382	NaN

df1.loc[dates[0]:dates[1], 'E'] = 1
df1

	A	B	C	D	E
2013-01-01	-1.815560	-2.066970	0.083446	0.457541	1.0
2013-01-02	0.878679	-1.279713	-2.170213	-0.317302	1.0
2013-01-03	-2.504659	0.727344	-0.062560	1.473940	NaN
2013-01-04	0.245661	1.158579	0.431938	-0.511382	NaN

df1.fillna(value=5)

	A	B	C	D	E
2013-01-01	-1.815560	-2.066970	0.083446	0.457541	1.0
2013-01-02	0.878679	-1.279713	-2.170213	-0.317302	1.0
2013-01-03	-2.504659	0.727344	-0.062560	1.473940	5.0
2013-01-04	0.245661	1.158579	0.431938	-0.511382	5.0

Operation

df.mean()

A   -0.632503
B   -0.201514
C   -0.447252
D    0.106758
dtype: float64

df.mean(1)

2013-01-01   -0.835386
2013-01-02   -0.722137
2013-01-03   -0.091484
2013-01-04    0.331199
2013-01-05    0.246799
2013-01-06   -0.690757
Freq: D, dtype: float64

df.apply(lambda x: x.max() - x.min())

A    3.383339
B    3.683604
C    2.602150
D    1.985322
dtype: float64

Grouping

df = pd.DataFrame({'A': ['foo', 'bar', 'foo', 'bar',
                       'foo', 'bar', 'foo', 'foo'],
                    'B': ['one', 'one', 'two', 'three',
                          'two', 'two', 'one', 'three'],
                    'C': np.random.randn(8),
                    'D': np.random.randn(8)})
df

	A	B	C	D
0	foo	one	0.445369	-0.374983
1	bar	one	1.147522	0.078017
2	foo	two	0.411960	-2.394559
3	bar	three	-1.254990	-0.817442
4	foo	two	0.421721	-0.667637
5	bar	two	-1.153841	0.159370
6	foo	one	0.170444	-0.078069
7	foo	three	-0.302964	0.584654

df.groupby('A').sum()

	C	D
A
bar	-1.261309	-0.580056
foo	1.146529	-2.930594

df.groupby(['A', 'B']).sum()

		C	D
A	B
bar	one	1.147522	0.078017
	three	-1.254990	-0.817442
	two	-1.153841	0.159370
foo	one	0.615813	-0.453052
	three	-0.302964	0.584654
	two	0.833681	-3.062196

另外，pandas还可以读入csv和excel文件，感觉非常方便，就是python版本的Excel