kosmos.ml.config.factories.encoding

Classes

class EncodingConfig

Bases: abc.ABC

Encoding configuration.

Methods

get_instance(input_dim: int, output_dim: int) → kosmos.ml.models.vqc.encoding.VQCEncoding

Get the encoding instance.

Parameters:

• input_dim (int) – Model input dimension.

• output_dim (int) – Model output dimension.

Returns:

Encoding instance.

Return type:

VQCEncoding

---

class AngleEmbeddingConfig(rotation: Literal['X', 'Y', 'Z'] = 'X')

Bases: EncodingConfig

Angle embedding configuration.

Initialize the angle embedding configuration.

Parameters:

rotation (Literal["X", "Y", "Z"]) – The rotation to use for the angle embedding. Defaults to “X”.

Methods

get_instance(input_dim: int, output_dim: int) → kosmos.ml.models.vqc.encoding.AngleEmbedding

Get the angle embedding instance.

Parameters:

• input_dim (int) – Model input dimension.

• output_dim (int) – Model output dimension.

Returns:

Angle embedding instance.

Return type:

AngleEmbedding

---

class AmplitudeEmbeddingConfig(pad_with: complex = 0.3, *, normalize: bool = True)

Bases: EncodingConfig

Amplitude embedding configuration.

Initialize the amplitude embedding configuration.

Parameters:

• pad_with (complex) – The input is padded with this constant to size \(2^n\).

• normalize (bool) – Whether to normalize the features. Defaults to True.

Methods

get_instance(input_dim: int, output_dim: int) → kosmos.ml.models.vqc.encoding.AmplitudeEmbedding

Get the amplitude embedding instance.

Parameters:

• input_dim (int) – Model input dimension.

• output_dim (int) – Model output dimension.

Returns:

Amplitude embedding instance.

Return type:

AmplitudeEmbedding