Crate caffe2op_swish
source ·Structs
- | Swish takes one input data (Tensor) | and produces one output data (Tensor) | where the swish function, y = x / (1 + exp(-x)), | is applied to the tensor elementwise. |
- | SwishGradient takes X, Y and dY and uses | this to update dX according to the chain | rule and derivatives of the swish function. |