Note

This documentation is for a development version. Click here for the latest stable release (v1.1.0).

nengo_spa.semantic_pointer

Classes

AbsorbingElement(n_dimensions[, vocab, algebra])

Absorbing element.

Identity(n_dimensions[, vocab, algebra])

Identity element.

SemanticPointer(data[, vocab, algebra, name])

A Semantic Pointer, based on Holographic Reduced Representations.

Zero(n_dimensions[, vocab, algebra])

Zero element.

class nengo_spa.semantic_pointer.SemanticPointer(data, vocab=None, algebra=None, name=None)[source]

A Semantic Pointer, based on Holographic Reduced Representations.

Operators are overloaded so that + and - are addition, * is circular convolution, and ~ is the inversion operator.

Parameters
  • data (array_like) – The vector constituting the Semantic Pointer.

  • vocab (Vocabulary, optional) – Vocabulary that the Semantic Pointer is considered to be part of. Mutually exclusive with the algebra argument.

  • algebra (AbstractAlgebra, optional) – Algebra used to perform vector symbolic operations on the Semantic Pointer. Defaults to CircularConvolutionAlgebra. Mutually exclusive with the vocab argument.

  • name (str, optional) – A name for the Semantic Pointer.

v

The vector constituting the Semantic Pointer.

Type

array_like

algebra

Algebra that defines the vector symbolic operations on this Semantic Pointer.

Type

AbstractAlgebra

vocab

The vocabulary the this Semantic Pointer is considered to be part of.

Type

Vocabulary or None

name

Name of the Semantic Pointer.

Type

str or None

normalized()[source]

Normalize the Semantic Pointer and return it as a new object.

If the vector length is zero, the Semantic Pointer will be returned unchanged.

The original object is not modified.

unitary()[source]

Make the Semantic Pointer unitary and return it as a new object.

The original object is not modified.

A unitary Semantic Pointer has the property that it does not change the length of Semantic Pointers it is bound with using circular convolution.

copy()[source]

Return another semantic pointer with the same data.

length()[source]

Return the L2 norm of the vector.

bind(other)[source]

Return the binding of two SemanticPointers.

rbind(other)[source]

Return the binding of two SemanticPointers.

get_binding_matrix(swap_inputs=False)[source]

Return the matrix that does a binding with this vector.

This should be such that A*B == dot(A.get_binding_matrix(), B.v).

dot(other)[source]

Return the dot product of the two vectors.

compare(other)[source]

Return the similarity between two SemanticPointers.

This is the normalized dot product, or (equivalently), the cosine of the angle between the two vectors.

reinterpret(vocab)[source]

Reinterpret the Semantic Pointer as part of vocabulary vocab.

The vocab parameter can be set to None to clear the associated vocabulary and allow the source to be interpreted as part of the vocabulary of any Semantic Pointer it is combined with.

translate(vocab, populate=None, keys=None, solver=None)[source]

Translate the Semantic Pointer to vocabulary vocab.

The translation of a Semantic Pointer uses some form of projection to convert the Semantic Pointer to a Semantic Pointer of another vocabulary. By default the outer products of terms in the source and target vocabulary are used, but if solver is given, it is used to find a least squares solution for this projection.

Parameters
  • vocab (Vocabulary) – Target vocabulary.

  • populate (bool, optional) – Whether the target vocabulary should be populated with missing keys. This is done by default, but with a warning. Set this explicitly to True or False to silence the warning or raise an error.

  • keys (list, optional) – All keys to translate. If None, all keys in the source vocabulary will be translated.

  • solver (nengo.Solver, optional) – If given, the solver will be used to solve the least squares problem to provide a better projection for the translation.

distance(other)[source]

Return a distance measure between the vectors.

This is 1-cos(angle), so that it is 0 when they are identical, and the distance gets larger as the vectors are farther apart.

mse(other)[source]

Return the mean-squared-error between two vectors.

class nengo_spa.semantic_pointer.Identity(n_dimensions, vocab=None, algebra=None)[source]

Bases: nengo_spa.semantic_pointer.SemanticPointer

Identity element.

Parameters
  • n_dimensions (int) – Dimensionality of the identity vector.

  • vocab (Vocabulary, optional) – Vocabulary that the Semantic Pointer is considered to be part of. Mutually exclusive with the algebra argument.

  • algebra (AbstractAlgebra, optional) – Algebra used to perform vector symbolic operations on the Semantic Pointer. Defaults to CircularConvolutionAlgebra. Mutually exclusive with the vocab argument.

class nengo_spa.semantic_pointer.AbsorbingElement(n_dimensions, vocab=None, algebra=None)[source]

Bases: nengo_spa.semantic_pointer.SemanticPointer

Absorbing element.

If \(z\) denotes the absorbing element, \(v \circledast z = c z\), where \(v\) is a Semantic Pointer and \(c\) is a real-valued scalar. Furthermore \(\|z\| = 1\).

Parameters
  • n_dimensions (int) – Dimensionality of the identity vector.

  • vocab (Vocabulary, optional) – Vocabulary that the Semantic Pointer is considered to be part of. Mutually exclusive with the algebra argument.

  • algebra (AbstractAlgebra, optional) – Algebra used to perform vector symbolic operations on the Semantic Pointer. Defaults to CircularConvolutionAlgebra. Mutually exclusive with the vocab argument.

class nengo_spa.semantic_pointer.Zero(n_dimensions, vocab=None, algebra=None)[source]

Bases: nengo_spa.semantic_pointer.SemanticPointer

Zero element.

Parameters
  • n_dimensions (int) – Dimensionality of the identity vector.

  • vocab (Vocabulary, optional) – Vocabulary that the Semantic Pointer is considered to be part of. Mutually exclusive with the algebra argument.

  • algebra (AbstractAlgebra, optional) – Algebra used to perform vector symbolic operations on the Semantic Pointer. Defaults to CircularConvolutionAlgebra. Mutually exclusive with the vocab argument.