numpy.random.choice

• Novice tutorial
• • describe
  • grammar
  • Parameters
  • Return value
  • example
• Official documents
• • Introduce
  • Parameters
  • example
• Extended reading
• • First look at numpy in choice()
  • Look again random in choice() and choices()
  • The last to see random.sample()

Novice tutorial

describe

choice() Method to return a list , Random items of tuples or strings .

grammar

Here are choice() The grammar of the method :

import random
random.choice( seq )

Be careful ：choice() It's not directly accessible , Import required random modular , And then through random Static objects call the method .

Parameters

seq -- It could be a list , Tuples or strings .

Return value

Return random item .

example

Here's how to use choice() Examples of methods ：

#!/usr/bin/python
import random
print "choice([1, 2, 3, 5, 9]) : ", random.choice([1, 2, 3, 5, 9])
print "choice('A String') : ", random.choice('A String')

The output of the above example is ：

choice([1, 2, 3, 5, 9]) : 2
choice('A String') : n

Official documents

Introduce

random.choice() function ： From the given 1 Random sampling function in dimension group .

Parameters

numpy.random.choice(a, size=None, replace=True, p=None)

• a : If it's a one-dimensional array , It means random sampling from this one-dimensional array ; If it is int type , It means the from 0 To a-1 Random sampling in this sequence .

• size : Number of sampling results , The default is 1. It can be an integer , Indicates the quantity to be sampled ; It can also be for tuple, Such as (m, n, k), Then the quantity to be sampled is m *n *k, size by (m, n, k).

• replace : boolean type , Whether the sampled sample should be replaced ？ I don't quite understand this place , We measured it and found that replace Designated as True when , The sampled elements will repeat ; When replace Designated as False when , Sampling does not repeat .

• p : A one-dimensional array , Formulated the a The probability of sampling each element in , If default None, be a Each element in the has the same probability of being sampled .

Be careful ：
replace The representative means whether to put it back after sampling , If it is False Words , Then the three numbers are different , If it is True Words , There may be repeated , Because the front one was put back .
p The number of median values must be equal to a The number of data in is consistent .

Return value ：samples : single item or ndarray（The generated random samples）

ValueError：
If a is an int and less than zero, if a or p are not 1-dimensional, if a is an array-like of size 0, if p is not a vector of probabilities, if a and p have different lengths, or if replace=False and the sample size is greater than the population size

example

From size to 3 Of np.arange(5) Generate a uniform （p=None） A random sample of ：

>>> np.random.choice(5, 3)
array([0, 3, 4])
>>> #This is equivalent to np.random.randint(0,5,3)

From size to 3 Of np.arange(5) Generate a Nonuniformity （p Valuable ） A random sample of ：

>>> np.random.choice(5, 3, p=[0.1, 0, 0.3, 0.6, 0])
array([3, 3, 0])

From size to 3 Of np.arange(5) Generate a uniform A random sample of , There is no replacement （ repeat ）：

>>> np.random.choice(5, 3, replace=False)
array([3,1,0])
>>> #This is equivalent to np.random.permutation(np.arange(5))[:3]

From size to 3 Of np.arange(5) Generate a Nonuniformity A random sample of , There is no replacement （ repeat ）：

>>> np.random.choice(5, 3, replace=False, p=[0.1, 0, 0.3, 0.6, 0])
array([2, 3, 0])

The first parameter in the above example can be replaced by an arbitrary array , Not just integers . for example ：

>>> aa_milne_arr = ['pooh', 'rabbit', 'piglet', 'Christopher']
>>> np.random.choice(aa_milne_arr, 5, p=[0.5, 0.1, 0.1, 0.3])
array(['pooh', 'pooh', 'pooh', 'Christopher', 'piglet'],
dtype='|S11')

In practice , First create a mask Variable , And then through mask To sample the data to be sampled ：

 mask = np.random.choice(split_size, batch_size)
captions = data['%s_captions' % split][mask]
image_idxs = data['%s_image_idxs' % split][mask]

Extended reading

python in choice contrast （ stay numpy and random There have been ） Plus sample() Function comparison

First look at numpy in choice()

Look again random in choice() and choices()

The last to see random.sample()