groupby

import pandas as pd
import numpy as np

df = pd.DataFrame({"key1":["a","a","b","b","a"],
                  "key2":["one","two","one","two","one"],
                  "data1":np.random.randint(1,10,5),
                 "data2":np.random.randint(1,10,5)})
df

	data1	data2	key1	key2
0	3	3	a	one
1	1	4	a	two
2	9	4	b	one
3	4	4	b	two
4	7	4	a	one

df["data1"].groupby(df["key1"]).mean()

key1
a    3.666667
b    6.500000
Name: data1, dtype: float64

key = [1,2,1,2,3]
df["data1"].groupby(key).mean()

1    6.0
2    2.5
3    7.0
Name: data1, dtype: float64

df["data1"].groupby([df["key1"],df["key2"]]).sum()

key1  key2
a     one     10
      two      1
b     one      9
      two      4
Name: data1, dtype: int32

df["data1"].groupby([df["key1"],df["key2"]]).size()

key1  key2
a     one     2
      two     1
b     one     1
      two     1
dtype: int64

df.groupby("key1").sum()

	data1	data2
key1
a	11	11
b	13	8

mean = df.groupby(["key1","key2"]).sum()["data1"]
mean

key1  key2
a     one     10
      two      1
b     one      9
      two      4
Name: data1, dtype: int32

mean.unstack()

key2	one	two
key1
a	10	1
b	9	4

for name,group in df.groupby("key1"):
    print(name)
    print(group)

a
   data1  data2 key1 key2
0      3      3    a  one
1      1      4    a  two
4      7      4    a  one
b
   data1  data2 key1 key2
2      9      4    b  one
3      4      4    b  two

df.groupby(df.dtypes, axis = 1).sum()

	int32	object
0	6	aone
1	5	atwo
2	13	bone
3	8	btwo
4	11	aone

df = pd.DataFrame(np.random.randint(1,10,(5,5)),
                 columns=["a","b","c","d","e"],
                 index=["Alice","Bob","Candy","Dark","Emily"])
df.ix[1,1:3] = np.NaN
df

	a	b	c	d	e
Alice	4	4.0	5.0	4	6
Bob	6	NaN	NaN	8	1
Candy	7	9.0	6.0	8	8
Dark	3	6.0	1.0	3	9
Emily	2	3.0	6.0	9	1

mapping = {"a":"red","b":"red","c":"blue","d":"orange","e":"blue"}
grouped = df.groupby(mapping, axis=1)
grouped.sum()

	blue	orange	red
Alice	11.0	4.0	8.0
Bob	1.0	8.0	6.0
Candy	14.0	8.0	16.0
Dark	10.0	3.0	9.0
Emily	7.0	9.0	5.0

grouped.count()

	blue	orange	red
Alice	2	1	2
Bob	1	1	1
Candy	2	1	2
Dark	2	1	2
Emily	2	1	2

df = pd.DataFrame(np.random.randint(1,10,(5,5)),
                 columns=["a","b","c","d","e"],
                 index=["Alice","Bob","Candy","Dark","Emily"])
df

	a	b	c	d	e
Alice	7	9	8	9	8
Bob	8	4	2	9	1
Candy	9	6	1	1	1
Dark	6	6	8	1	1
Emily	3	5	8	8	6

def _group_key(idx):
    print(idx)
    return(idx)
df.groupby(_group_key).size()

Alice
Bob
Candy
Dark
Emily





Alice    1
Bob      1
Candy    1
Dark     1
Emily    1
dtype: int64

def _group_key(idx):
    print(idx)
    return len(idx)
df.groupby(_group_key).size()

Alice
Bob
Candy
Dark
Emily





3    1
4    1
5    3
dtype: int64

df.groupby(len).size()

3    1
4    1
5    3
dtype: int64

df.groupby(len).sum()

	a	b	c	d	e
3	8	4	2	9	1
4	6	6	8	1	1
5	19	20	17	18	15

columns = pd.MultiIndex.from_arrays([["china","usa","china","usa","china"],
                                    ["A","A","B","C","B"]], names = ["country","index"])
df = pd.DataFrame(np.random.randint(1,10,(5,5)),columns=columns)
df

country	china	usa	china	usa	china
index	A	A	B	C	B
0	6	5	5	5	9
1	9	5	2	9	2
2	1	6	5	9	2
3	8	9	1	3	3
4	2	2	7	9	6

df.groupby(level="country", axis=1).sum()

country	china	usa
0	20	10
1	13	14
2	8	15
3	12	12
4	15	11

df.groupby(level="index", axis=1).sum()

index	A	B	C
0	11	14	5
1	14	4	9
2	7	7	9
3	17	4	3
4	4	13	9