[转载] python numpy.random.randn()与numpy.random.rand()的区别 (正态分布公式)（标准正态分布 standard normal distribution

参考链接： Python中的numpy.random.randn

引用文章: numpy.random.randn()与numpy.random.rand()的区别 https://www.cnblogs.com/hezhiyao/p/8065528.html 

 总的来书就是, numpy.random.randn()生成的numpy矩阵元素值符合标准正态分布, 而numpy.random.rand()生成的是[0, 1)之间的均匀随机数.  [附]标准正态分布: 标准正态分布（英语：standard normal distribution， 德语Standardnormalverteilung），是一个在数学、物理及工程等领域都非常重要的概率分布，在统计学的许多方面有着重大的影响力。期望值μ=0，即曲线图象对称轴为Y轴，标准差σ=1条件下的正态分布，记为N(0，1)。  标准正态分布又称为u分布，是以0为均数、以1为标准差的正态分布，记为N（0，1）。  正态分布公式: 普通正态分布如何转换到标准正态分布 https://blog.csdn.net/bitcarmanlee/article/details/86440851 见numpy.random.randn() 官方doc结尾有应用案例.  

- numpy.random.randn() 官方doc: 

Docstring:

randn(d0, d1, ..., dn)

 

Return a sample (or samples) from the "standard normal" distribution.

 

If positive, int_like or int-convertible arguments are provided,

`randn` generates an array of shape ``(d0, d1, ..., dn)``, filled

with random floats sampled from a univariate "normal" (Gaussian)

distribution of mean 0 and variance 1 (if any of the :math:`d_i` are

floats, they are first converted to integers by truncation). A single

float randomly sampled from the distribution is returned if no

argument is provided.

 

This is a convenience function.  If you want an interface that takes a

tuple as the first argument, use `numpy.random.standard_normal` instead.

 

Parameters

----------

d0, d1, ..., dn : int, optional

    The dimensions of the returned array, should be all positive.

    If no argument is given a single Python float is returned.

 

Returns

-------

Z : ndarray or float

    A ``(d0, d1, ..., dn)``-shaped array of floating-point samples from

    the standard normal distribution, or a single such float if

    no parameters were supplied.

 

See Also

--------

standard_normal : Similar, but takes a tuple as its argument.

 

Notes

-----

For random samples from :math:`N(\mu, \sigma^2)`, use:

 

``sigma * np.random.randn(...) + mu``

 

Examples

--------

>>> np.random.randn()

2.1923875335537315 #random

 

Two-by-four array of samples from N(3, 6.25):

 

>>> 2.5 * np.random.randn(2, 4) + 3

array([[-4.49401501,  4.00950034, -1.81814867,  7.29718677],  #random

       [ 0.39924804,  4.68456316,  4.99394529,  4.84057254]]) #random

Type:      builtin_function_or_method

 

- numpy.random.rand() 官方doc: 

Docstring:

rand(d0, d1, ..., dn)

 

Random values in a given shape.

 

Create an array of the given shape and populate it with

random samples from a uniform distribution

over ``[0, 1)``.

 

Parameters

----------

d0, d1, ..., dn : int, optional

    The dimensions of the returned array, should all be positive.

    If no argument is given a single Python float is returned.

 

Returns

-------

out : ndarray, shape ``(d0, d1, ..., dn)``

    Random values.

 

See Also

--------

random

 

Notes

-----

This is a convenience function. If you want an interface that

takes a shape-tuple as the first argument, refer to

np.random.random_sample .

 

Examples

--------

>>> np.random.rand(3,2)

array([[ 0.14022471,  0.96360618],  #random

       [ 0.37601032,  0.25528411],  #random

       [ 0.49313049,  0.94909878]]) #random

Type:      builtin_function_or_method