eazygrad.log_softmax

eazygrad.log_softmax(input: _Tensor, dim: int) → _Tensor

Compute the logarithm of the softmax along a given axis.

Parameters:

• input (_Tensor) – Input tensor.

• dim (int) – Axis along which the log-softmax is computed.

Returns:

_Tensor – Tensor containing log-probabilities.

See also

torch.log_softmax