invert4geom.optimal_spline_damping

optimal_spline_damping(coordinates, data, weights=None, **kwargs)

Find the best damping parameter for a verde.SplineCV() fit. All kwargs are passed to the verde.SplineCV class.

Parameters:

• coordinates (tuple[Series | ndarray[tuple[Any, ...], dtype[TypeVar(_ScalarT, bound= generic)]], Series | ndarray[tuple[Any, ...], dtype[TypeVar(_ScalarT, bound= generic)]]]) – easting and northing coordinates of the data

• data (Series | ndarray[tuple[Any, ...], dtype[TypeVar(_ScalarT, bound= generic)]]) – data for fitting the spline to

• weights (Series | ndarray[tuple[Any, ...], dtype[TypeVar(_ScalarT, bound= generic)]] | None) – if not None, then the weights assigned to each data point. Typically, this should be 1 over the data uncertainty squared, by default None

• kwargs (Any)

Keyword Arguments:

• dampings (float | None) – The positive damping regularization parameter. Controls how much smoothness is imposed on the estimated forces. If None, no regularization is used, by default None

• force_coords (bool) – The easting and northing coordinates of the point forces. If None (default), then will be set to the data coordinates.

• cv (None | cross-validation generator) – Any scikit-learn cross-validation generator. If not given, will use the default set by verde.cross_val_score.

• delayed (bool) – If True, will use dask.delayed.delayed to dispatch computations and allow dask to execute the grid search in parallel (see note above).

• scoring (None | str | Callable) – The scoring function (or name of a function) used for cross-validation. Must be known to scikit-learn. See the description of scoring in sklearn.model_selection.cross_val_score for details. If None, will fall back to the verde.Spline.score method.

Returns:

the spline which best fits the data

Return type:

Spline