nnopinf.operators.LinearAffineSkewTensorOperator#

class nnopinf.operators.LinearAffineSkewTensorOperator(acts_on, depends_on, skew=True, name='LinearAffineSkewTensorOperator')[source]#

Bases: Module

\(f: (v,x) \mapsto [S_{ijk} - S_{jik}] x_k v_k\)

Constructs a linear affine skew-symmetric tensor operator

Parameters:

  • acts_on (nnopinf.Variable) – The state the operator acts on, i.e., the \(x\)

  • depends_on (tuple of nnopinf.Variable) – The affine variables the operator depends on, i.e., the \(v\)

skewbool

Constructs a skew-symmetric operator if true, constructs a symmetric operator if false

namestring

Operator name. Used when saving to file

forward(inputs, return_jacobian=False)[source]#

Forward pass of operator

Warning

return_jacobian is currently not implemented

Parameters:

  • inputs (dict(str, np.array)) – Dictionary of input data in the form of arrays referenced by the variable name, i.e., inputs[‘x’] = np.ones(3)

  • return_jacobian (bool, optional) – If True, return the (approximate) Jacobian in addition to the output.

set_scalings(input_scalings_dict, output_scaling)[source]#

Store input, state, and output scaling factors for the affine skew operator.

Parameters:

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

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

Show Source