skcyto.AsinhTransformer

class skcyto.AsinhTransformer(t: float = 262144, m: float = 4.5, a: float = 0)[source]

Parametrized Arcsinh transform

Implemented according to GatingML2.0 specification Note that this is equivalent to a logicle transform with w = 0

Parameters:
tfloat

Upper bound of the linear scale, by default 262144

mfloat

Number of decades for the logarithmic scale, by default 4.5

afloat

Number of negative decades, by default 0

__init__(t: float = 262144, m: float = 4.5, a: float = 0)[source]
fit(X: ndarray[Any, dtype[ScalarType]], y: ndarray[Any, dtype[ScalarType]] | None = None)[source]

Fit only validates parameters and data, as everything is defined by the user.

XNDArray

Input data

yIgnored.

Not used, only present for API conventions.

fit_transform(X: ndarray[Any, dtype[ScalarType]], y=None) → ndarray[Any, dtype[ScalarType]][source]

Fit and transform data

As fit does not doo anything, this is identical to just running transform and is only implemented for API conventions.

Parameters:
XNDArray

Input data

yIgnored.

Not used, only present for API conventions.

Returns:
NDArray

Transformed data

inverse_transform(X: ndarray[Any, dtype[ScalarType]]) → ndarray[Any, dtype[ScalarType]][source]

Perform inverse asinh transform

Parameters:
XNDArray

Input data

Returns:
NDArray

inversely transformed data

transform(X: ndarray[Any, dtype[ScalarType]]) → ndarray[Any, dtype[ScalarType]][source]

Transform according to specified asinh transform

Parameters:
XNDArray

Data

Returns:
NDArray

Transformed data