nnopinf.operators.CompositeOperator#

class nnopinf.operators.CompositeOperator(state_operators, name='CompositeOperator')[source]#

Bases: Module

\(f: (f_1,\ldots,f_K) \mapsto \sum_{i=1}^K f_k\)

Constructs an operator composed of other NN-OpInf operators

Parameters:

  • state_operators (list of nnopinf.operators.Operator) – List of individual operators

  • name (string) – Operator name. Used when saving to file

Examples

>>> import nnopinf
>>> import nnopinf.operators
>>> x_input = nnopinf.Variable(size=3,name="x")
>>> mu_input = nnopinf.Variable(size=2,name="mu")
>>> MatrixMlp = nnopinf.operators.MatrixOperator(n_outputs=5,acts_on=x_input,depends_on=(x_input,mu_input,),n_hidden_layers=2,n_neurons_per_layer=2)
>>> MatrixMlp2 = nnopinf.operators.MatrixOperator(n_outputs=5,acts_on=x_input,depends_on=(x_input,mu_input,),n_hidden_layers=2,n_neurons_per_layer=2)
>>> CompositeMlp = nnopinf.operators.CompositeOperator([MatrixMlp,MatrixMlp2])
forward(inputs, return_jacobian=False)[source]#

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

set_scalings(input_scaling, output_scaling)[source]#

Apply input and output scaling factors to each component operator.

Parameters:

  • input_scaling (dict) – Mapping from variable name to the corresponding feature-wise input scaling vector.

  • output_scaling (tensor-like) – Feature-wise scaling vector for the composite output.

Show Source