transform_ddts()

ShiftScaleTransformer.transform_ddts(ddts, inplace: bool = False)

Apply the learned transformation to snapshot time derivatives.

Denoting the transformation by \(\mathcal{T}(\q) = \alpha(\q - \bar{\q}) + \beta\), this is the function \(\mathcal{T}'(\z) = \alpha\z\). Hence, \(\mathcal{T}'(\ddt q) = \ddt \mathcal{T}(q)\).

Parameters:

ddts(n, …) ndarray

Matrix of n-dimensional snapshot time derivatives, or a single snapshot time derivative.

inplacebool

If True, overwrite ddts during the transformation. If False, create a copy of the data to transform.

Returns:

ddts_transformed(n, …) ndarray

Transformed n-dimensional snapshot time derivatives.

Raises:

AttributeError

If the transformer is not ready because fit() or fit_transform() have not been called.

ValueError

If the ddts do not align with the state_dimension.