Softplus operation. Computes the element-wise softplus of x:
op_softplus(x, steepness = 1, name = "")
| x | matrix or CNTK Function that outputs a tensor |
|---|---|
| name | (str) the name of the Function instance in the network |
softplus(x)=log(1+exp(x))softplus(x)=log(1+exp(x))
The optional steepness allows to make the knee sharper (steepness>1) or softer, by computing softplus(x * steepness) / steepness. (For very large steepness, this approaches a linear rectifier).
The output tensor has the same shape as x.