megengine.random.RNG¶

class RNG(seed=None, device=None)[源代码]¶

RNG exposes a number of methods for generating random numbers.

参数

seed (Optional[int]) – random seed used to initialize the pseudo-random number generator. Default: None
device (Optional[str]) – the device of generated tensor. Default: None

实际案例

import megengine.random as rand
rng = rand.RNG(seed=100)
x = rng.uniform(size=(2, 2))
print(x.numpy())

Outputs:

[[0.84811664 0.6147553 ]
 [0.59429836 0.64727545]]

方法

`beta`(alpha, beta[, size])	Random variable with Beta distribution $\operatorname{Beta}(\alpha, \beta)$.
`gamma`(shape[, scale, size])	Random variable with Gamma distribution $\Gamma(k, \theta)$.
`normal`([mean, std, size])	Random variable with Gaussian distribution $N(\mu, \sigma)$.
`permutation`(n, *[, dtype])	Generates a random permutation of integers from $0$ to $n - 1$.
`poisson`(lam[, size])	Random variable with poisson distribution $\operatorname{Poisson}(\lambda)$.
`uniform`([low, high, size])	Random variable with uniform distribution $U(0, 1)$.

megengine.random

megengine.random.RNG.beta

`beta`(alpha, beta[, size])	Random variable with Beta distribution \(\operatorname{Beta}(\alpha, \beta)\).
`gamma`(shape[, scale, size])	Random variable with Gamma distribution \(\Gamma(k, \theta)\).
`normal`([mean, std, size])	Random variable with Gaussian distribution \(N(\mu, \sigma)\).
`permutation`(n, *[, dtype])	Generates a random permutation of integers from \(0\) to \(n - 1\).
`poisson`(lam[, size])	Random variable with poisson distribution \(\operatorname{Poisson}(\lambda)\).
`uniform`([low, high, size])	Random variable with uniform distribution $U(0, 1)$.