Pandas - 数据的聚合与分组计算

---

1 数据分组 - groupby()

from pandas import DataFrame,Series
import pandas as pd
import numpy as np
from numpy import nan as NA

df = DataFrame({
  "key1":list("aabba"),
               "key2":["one","two","one","two","one"],
               "data1":np.random.randn(5),
               "data2":np.random.randn(5)})
df

.dataframe tbody tr th:only-of-type { vertical-align: middle; } .dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; }

	data1	data2	key1	key2
0	0.008908	0.652712	a	one
1	0.438874	0.423774	a	two
2	0.299105	-1.279888	b	one
3	-0.191032	0.429504	b	two
4	-0.395208	0.523417	a	one

# 对数据按照key1进行分组
# groupby 只进行分组，不会进行任何的计算操作
grouped = df["data1"].groupby(df["key1"]) 

grouped.count()  # 根据分组结果，计算每个分组下元素的个数

    key1
    a    3
    b    2
    Name: data1, dtype: int64

grouped.max()   # 根据分组结果，计算每个分组下的最大值

    key1
    a    0.438874
    b    0.299105
    Name: data1, dtype: float64

grouped.size()

    key1
    a    3
    b    2
    Name: data1, dtype: int64

grouped.mean()

    key1
    a    0.017525
    b    0.054036
    Name: data1, dtype: float64

grouped.describe()

.dataframe tbody tr th:only-of-type { vertical-align: middle; } .dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; }

	count	mean	std	min	25%	50%	75%	max
key1
a	3.0	0.017525	0.417108	-0.395208	-0.193150	0.008908	0.223891	0.438874
b	2.0	0.054036	0.346579	-0.191032	-0.068498	0.054036	0.176571	0.299105

df

.dataframe tbody tr th:only-of-type { vertical-align: middle; } .dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; }

	data1	data2	key1	key2
0	0.008908	0.652712	a	one
1	0.438874	0.423774	a	two
2	0.299105	-1.279888	b	one
3	-0.191032	0.429504	b	two
4	-0.395208	0.523417	a	one

1.1 按照多列进行分组的两种方法

# 按照多列进行分组
# 先获取某一列的所有值，再按照key1和key2进行分组
grouped1 = df["data1"].groupby([df["key1"],df["key2"]]) 
grouped1.mean()

    key1  key2
    a     one    -0.193150
          two     0.438874
    b     one     0.299105
          two    -0.191032
    Name: data1, dtype: float64

# 先将所有的数据按照key1和key2进行分组，再获取分组结果data1或者data2的分组结果
group_all = df.groupby(["key1","key2"])

# 获取data1的分组结果
group_all["data1"].mean()

    key1  key2
    a     one    -0.193150
          two     0.438874
    b     one     0.299105
          two    -0.191032
    Name: data1, dtype: float64

# 获取data2的分组结果
group_all["data2"].mean()

    key1  key2
    a     one     0.588065
          two     0.423774
    b     one    -1.279888
          two     0.429504
    Name: data2, dtype: float64

df

.dataframe tbody tr th:only-of-type { vertical-align: middle; } .dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; }

	data1	data2	key1	key2
0	0.008908	0.652712	a	one
1	0.438874	0.423774	a	two
2	0.299105	-1.279888	b	one
3	-0.191032	0.429504	b	two
4	-0.395208	0.523417	a	one

# 按照key1将所有的数据进行分组
group_key1 = df.groupby(["key1"])

1.2 对分组进行迭代

# 通过循环遍历已分组对象中的每一个组别数据
for name,group in group_key1:
    print("组别：",name)
    print("数据：\n",group)

    组别： a
    数据：
           data1     data2 key1 key2
    0  0.008908  0.652712    a  one
    1  0.438874  0.423774    a  two
    4 -0.395208  0.523417    a  one
    组别： b
    数据：
           data1     data2 key1 key2
    2  0.299105 -1.279888    b  one
    3 -0.191032  0.429504    b  two

1.3 将列数据按照数据类型进行分组

# 查看DataFrame中每一列的数据类型
df

.dataframe tbody tr th:only-of-type { vertical-align: middle; } .dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; }

	data1	data2	key1	key2
0	0.008908	0.652712	a	one
1	0.438874	0.423774	a	two
2	0.299105	-1.279888	b	one
3	-0.191032	0.429504	b	two
4	-0.395208	0.523417	a	one

df.dtypes

    data1    float64
    data2    float64
    key1      object
    key2      object
    dtype: object

# 将列数据按照数据类型进行分组
group_types = df.groupby(df.dtypes,axis=1)