torch.nn.functional.dropout

torch.nn.functional.dropout(input, p=0.5, training=True, inplace=False) [source]

During training, randomly zeroes some elements of the input tensor with probability p.

Uses samples from a Bernoulli distribution.

See Dropout for details.

Parameters

• p (float) – probability of an element to be zeroed. Default: 0.5
• training (bool) – apply dropout if is True. Default: True
• inplace (bool) – If set to True, will do this operation in-place. Default: False

Return type

Tensor