Numpy与Pytorch对比（二）随机数生成

np.random

np.random.rand()
均匀分布
Create an array of the given shape and populate it with random samples from a uniform distribution [0, 1).

np.random.random()
均匀分布
Return random floats in the half-open interval [0.0, 1.0).
Results are from the “continuous uniform” distribution over the stated interval.
Sample from [a, b), b > a : (b - a) * random_sample() + a

np.random.randn()
正态分布
Return a sample (or samples) from the “standard normal” distribution.

np.random.randint()
Return random integers from low (inclusive) to high (exclusive).

>>> np.random.randint(5, size=(2, 4))
array([[4, 0, 2, 1],
       [3, 2, 2, 0]])

Pytorch

torch.rand
均匀分布
Returns a tensor filled with random numbers from a uniform distribution on the interval [0, 1)

torch.randn
正态分布
Returns a tensor filled with random numbers from a normal distribution with mean 0 and variance 1 (also called the standard normal distribution).

torch.randint
均匀分布
Returns a tensor filled with random integers generated uniformly between low (inclusive) and high (exclusive)

>>> torch.randint(3, 5, (3,))
tensor([ 4.,  3.,  4.])

总结

1、np.random.random()和np.random.rand()的区别：前者可以无需输入参数返回单个浮点数；
2、对于randint，两个框架的操作是基本一致的。
3、np.random.rand <=> torch.rand 均匀分布
4、np.random.randn <=> torch.randn 正态分布

其实两个框架各自还有很多api，这里只选了几个比较相似的来对比哈orz