Transform Composite

composite

CLASS	DESCRIPTION
CompositeTransformer	Combine multiple transformers in sequence.

Classes

CompositeTransformer(*transformers: Transformer)

Bases: Transformer

Combine multiple transformers in sequence.

Arguments

transformers : tuple of Transformer Transformers to apply in sequence.

Examples:

>>> from spectre.transform import StandardizationTransformer, L2Normalizer
>>> composite = CompositeTransformer(
    StandardizationTransformer(mean, std),
    L2Normalizer(),
)
>>> composite.transform(X)

METHOD	DESCRIPTION
transform	Sequentially apply transformers.
inverse_transform	Inverse transform.

Source code in spectre/transform/composite.py

def __init__(self, *transformers: Transformer):
    super().__init__()
    self.transformers = torch.nn.ModuleList(transformers)

Functions

transform(X: torch.Tensor) -> torch.Tensor

Sequentially apply transformers.

PARAMETER	DESCRIPTION
X	Input data to normalize. TYPE: Tensor

RETURNS	DESCRIPTION
Tensor	Transformed data.

Source code in spectre/transform/composite.py

def transform(self, X: torch.Tensor) -> torch.Tensor:
    """
    Sequentially apply transformers.

    Parameters
    ----------
    X : torch.Tensor
        Input data to normalize.

    Returns
    -------
    torch.Tensor
        Transformed data.
    """
    for transformer in self.transformers:
        X = transformer.transform(X)
    return X

inverse_transform(X: torch.Tensor) -> torch.Tensor

Inverse transform.

PARAMETER	DESCRIPTION
X	Normalized input data. TYPE: Tensor

RETURNS	DESCRIPTION
Tensor	Inverse transformed data.

Source code in spectre/transform/composite.py

def inverse_transform(self, X: torch.Tensor) -> torch.Tensor:
    """
    Inverse transform.

    Parameters
    ----------
    X : torch.Tensor
        Normalized input data.

    Returns
    -------
    torch.Tensor
        Inverse transformed data.
    """
    for transformer in reversed(self.transformers):
        X = transformer.inverse_transform(X)
    return X