piglot.optimisers.botorch.dataset.BayesDataset

class BayesDataset(n_dim: int, n_outputs: int, export: str | None = None, dtype: dtype = torch.float64, std_tol: float = 1e-06, pca_variance: float | None = None, device: str = 'cpu')[source]

Bases: object

Dataset class for multi-outcome data.

Methods

`load`	Load data from a given input file.
`min`	Return the minimum objective value of the dataset.
`numel_latent_space`	Return the number of components of the latent space.
`push`	Add a point to the dataset.
`save`	Save all dataset data to a file.
`to`	Move the dataset to a given device.
`transform_outcomes`	Transform outcomes to the latent standardised space.
`untransform_outcomes`	Transform outcomes back to the original space.
`update_stats`	Update the statistics of the dataset.

classmethod load(filename: str) → T[source]

Load data from a given input file.

Parameters

filenamestr: Path to the file to read from.

Returns

BayesDataset: Dataset loaded from the file.

min() → Tuple[ndarray, ndarray][source]

Return the minimum objective value of the dataset.

Returns

Tuple[np.ndarray, np.ndarray]: Parameters and objective value for the minimum point.

numel_latent_space() → int[source]

Return the number of components of the latent space.

Returns

int: Number of components of the latent space.

push(params: ndarray, results: ndarray, covariance: ndarray, objective: float | None) → None[source]

Add a point to the dataset.

Parameters

paramsnp.ndarray: Parameter values for this observation.
resultObjectiveResult: Result for this observation.

save(filename: str) → None[source]

Save all dataset data to a file.

Parameters

filenamestr: Output file path.

to(device: str) → BayesDataset[source]

Move the dataset to a given device.

Parameters

devicestr: Device to move the dataset to.

Returns

BayesDataset: The dataset in the new device.

transform_outcomes(values: Tensor | None = None, covariances: Tensor | None = None, diagonalise: bool = True) → Tuple[Tensor, Tensor][source]

Transform outcomes to the latent standardised space.

Parameters

valuestorch.Tensor: Values to transform.
covariancestorch.Tensor: Variances to transform.
diagonalisebool: Whether to diagonalise the covariance matrix (default: True).

Returns

Tuple[torch.Tensor, torch.Tensor]: Transformed values and variances.

untransform_outcomes(values: Tensor) → Tensor[source]

Transform outcomes back to the original space.

Parameters

valuestorch.Tensor: Values to transform.

Returns

torch.Tensor: Transformed values.

update_stats() → None[source]: Update the statistics of the dataset.