ngraph.hard_sigmoid

ngraph.hard_sigmoid(data: _pyngraph.Node, alpha: Union[_pyngraph.Node, int, float, numpy.ndarray], beta: Union[_pyngraph.Node, int, float, numpy.ndarray], name: Optional[str] = None) _pyngraph.Node

Perform Hard Sigmoid operation element-wise on data from input node.

Parameters
  • data – The node with data tensor.

  • alpha – A node producing the alpha parameter.

  • beta – A node producing the beta parameter

  • name – Optional output node name.

Returns

The new node performing a Hard Sigmoid element-wise on input tensor.

Hard Sigmoid uses the following logic:

y = max(0, min(1, alpha * data + beta))