torch.nn.functional.gumbel_softmax

torch.nn.functional.gumbel_softmax(logits, tau=1, hard=False, eps=1e-10, dim=-1)

Sample from the Gumbel-Softmax distribution (Link 1 Link 2) and optionally discretize.

Parameters

• logits (Tensor) – […, num_features] unnormalized log probabilities

• tau (float) – non-negative scalar temperature

• hard (bool) – if True, the returned samples will be discretized as one-hot vectors, but will be differentiated as if it is the soft sample in autograd

• dim (int) – A dimension along which softmax will be computed. Default: -1.

Returns

Sampled tensor of same shape as logits from the Gumbel-Softmax distribution. If hard=True, the returned samples will be one-hot, otherwise they will be probability distributions that sum to 1 across dim.

Return type

Tensor

Note

This function is here for legacy reasons, may be removed from nn.Functional in the future.

Note

The main trick for hard is to do y_hard - y_soft.detach() + y_soft

It achieves two things: - makes the output value exactly one-hot (since we add then subtract y_soft value) - makes the gradient equal to y_soft gradient (since we strip all other gradients)

Examples::

>>> logits = torch.randn(20, 32)
>>> # Sample soft categorical using reparametrization trick:
>>> F.gumbel_softmax(logits, tau=1, hard=False)
>>> # Sample hard categorical using "Straight-through" trick:
>>> F.gumbel_softmax(logits, tau=1, hard=True)