Multiset

class torchhd.structures.Multiset(dimensions: int, vsa: Literal['BSC', 'MAP', 'HRR', 'FHRR', 'BSBC', 'VTB', 'MCR', 'CGR'] = 'MAP', *, device=None, dtype=None)

class torchhd.structures.Multiset(input: VSATensor, *, size=0)

Hypervector multiset data structure.

Creates an empty multiset of dim dimensions or from an input tensor.

Parameters:

• dimensions (int) – number of dimensions of the multiset.

• vsa – (VSAOptions, optional): specifies the hypervector type and operations used (Default: "MAP").

• dtype (torch.dtype, optional) – the desired data type of returned tensor. Default: if None, uses a global default (see torch.set_default_tensor_type()).

• device (torch.device, optional) – the desired device of returned tensor. Default: if None, uses the current device for the default tensor type (see torch.set_default_tensor_type()). device will be the CPU for CPU tensor types and the current CUDA device for CUDA tensor types.

• input (VSATensor) – tensor representing a multiset.

• size (int, optional) – the size of the multiset provided as input. Default: 0.

Examples:

>>> M = structures.Multiset(10000)

>>> x = functional.random(1, 10000)
>>> M = structures.Multiset(x[0], size=1)

__len__() → int

Returns the size of the multiset.

Examples:

>>> len(M)
0

add(input: VSATensor) → None

Adds a new hypervector (input) to the multiset.

Parameters:

input (VSATensor) – Hypervector to add to the multiset.

Examples:

>>> letters = list(string.ascii_lowercase)
>>> letters_hv = torchhd.random(len(letters), 10000)
>>> M.add(letters_hv[0])

clear() → None

Empties the multiset

Examples:

>>> M.clear()

contains(input: VSATensor) → VSATensor

Returns the cosine similarity of the input vector against the multiset.

Parameters:

input (VSATensor) – Hypervector to compare against the multiset.

Examples:

>>> M.contains(letters_hv[0])
tensor(0.4575)

classmethod from_ngrams(input: VSATensor, n=3)

Creates a multiset from the ngrams of a set of hypervectors.

See: ngrams().

Parameters:

• input (VSATensor) – Set of hypervectors to convert in a multiset.

• n (int, optional) – The size of each \(n\)-gram, \(1 \leq n \leq m\). Default: 3.

Examples:

>>> x = functional.random(5, 3)
>>> M = structures.Multiset.from_ngrams(x)

classmethod from_tensor(input: VSATensor)

Creates a multiset from a set of hypervectors.

See: multiset().

Parameters:

input (VSATensor) – Set of hypervectors to convert in a multiset.

Examples:

>>> x = functional.random(3, 3)
>>> M = structures.Multiset.from_tensor(x)

remove(input: VSATensor) → None

Removes a hypervector (input) from the multiset.

Parameters:

input (VSATensor) – Hypervector to be removed from the multiset.

Examples:

>>> M.remove(letters_hv[0])