mindspore.ops.bernoulli

mindspore.ops.bernoulli(input, p=0.5, seed=None)[source]

Generates Bernoulli random values (0 or 1).

\[out_{i} \sim Bernoulli(p_{i})\]

Parameters

input (Tensor) – The input Tensor.
p (Union[Tensor, float], optional) – The probability of setting 1 for the corresponding position of the returned tensor. The value of p must be in the range [0, 1]. Default 0.5 .
seed (Union[int, None], optional) – The random seed. Default None means using the current timestamp.

Returns

Tensor

Supported Platforms:: GPU CPU

Examples

>>> import mindspore
>>> input = mindspore.tensor([1, 2, 3])
>>> mindspore.ops.bernoulli(input, p=1.0)
Tensor(shape=[3], dtype=Int64, value= [1, 1, 1])
>>> p = mindspore.tensor([0.0, 1.0, 1.0])
>>> mindspore.ops.bernoulli(input, p)
Tensor(shape=[3], dtype=Int64, value= [0, 1, 1])