torch.Tensor.bernoulli_#

Tensor.bernoulli_(p=0.5,*,generator=None)→Tensor#

Fills each location ofself with an independent sample from$\text{Bernoulli}(\mathtt{\text{p}})$Bernoulli(p).self can have integraldtype.

p should either be a scalar or tensor containing probabilities to beused for drawing the binary random number.

If it is a tensor, the${\text{i}}^{th}$ith element ofself tensorwill be set to a value sampled from$\text{Bernoulli}(\mathtt{\text{p\_tensor[i]}})$Bernoulli(p_tensor[i]). In this casep must havefloating pointdtype.

See alsobernoulli() andtorch.bernoulli()