python.query()函数

#使用python.query()函数对数据框进行（挑选行）的操作

import pandas as pd
d={
    'name':['a','n','c','d','e','f'],
    'Gender':['male','female','male','male','female','female'],
    'age':[23,24,24,22,21,20],
    'hight':[173,174,164,172,161,160],
    'weight1':[53,74,44,62,71,60],
    'weight2':[53,64,54,66,81,50]
}
df=pd.DataFrame(d)
df

	name	Gender	age	hight	weight1	weight2
0	a	male	23	173	53	53
1	n	female	24	174	74	64
2	c	male	24	164	44	54
3	d	male	22	172	62	66
4	e	female	21	161	71	81
5	f	female	20	160	60	50

df[df.age==24]

	name	Gender	age	hight	weight1	weight2
1	n	female	24	174	74	64
2	c	male	24	164	44	54

df[(df.age==24 )&( df.hight ==174)]

	name	Gender	age	hight	weight1	weight2
1	n	female	24	174	74	64

df.query("age==24")

	name	Gender	age	hight	weight1	weight2
1	n	female	24	174	74	64
2	c	male	24	164	44	54

df.query("age==24").query('hight==174')

	name	Gender	age	hight	weight1	weight2
1	n	female	24	174	74	64

df.query('index > 2')

	name	Gender	age	hight	weight1	weight2
3	d	male	22	172	62	66
4	e	female	21	161	71	81
5	f	female	20	160	60	50

df.query('Gender =="male" and name =="a"')

	name	Gender	age	hight	weight1	weight2
0	a	male	23	173	53	53

df.query('Gender =="male" and age<24')

	name	Gender	age	hight	weight1	weight2
0	a	male	23	173	53	53
3	d	male	22	172	62	66

import numpy as np

grade = np.array([1, 3,  4, 5, 0,  2, -1])

grade

array([ 1,  3,  4,  5,  0,  2, -1])

grade1 = np.where(grade > 3)

grade1#注意它返回的是数据的索引位置。

(array([2, 3], dtype=int32),)

grade1 = np.where(grade > 3, 'high', 'low')

#返回最小值的索引，返回最大值的索引，排序后的索引。

grade.argmin()

6

 grade.argmax()

3

grade.argsort()

array([6, 4, 0, 5, 1, 2, 3], dtype=int32)

grade = np.array([1, 3,  4, 5, 0,  2, -1])

grade1 = np.array([1, 3,  4, 5])

np.intersect1d(grade, grade1)#返回实际值

array([1, 3, 4, 5])