uncertainty_curve builtin

uncertainty_curve(
    spacing,
    uncertainty,
    *,
    bins=None,
    quantiles=Ellipsis,
    threshold=0.15,
    min_count=8,
    data=None,
)

Uncertainty quantiles and the share within a threshold per bin of data spacing.

Parameters:

Name	Type	Description	Default
spacing	array_like or str	Data spacing per row (e.g. data_spacing), or its column in data.	required
uncertainty	array_like or str	Uncertainty per row (e.g. relative error), or its column in data.	required
bins	int or array_like	Number of equal bins over the spacing range, or bin edges; default Freedman–Diaconis widths.	None
quantiles	sequence of float	Probabilities of the uncertainty quantiles.	Ellipsis
threshold	float	Uncertainty counted in share.	0.15
min_count	int	Bins with fewer rows are dropped.	8
data	(PointSet, BlockModel, Table or dict)	Where column names are looked up.	None

Returns:

Type	Description
Table	One row per bin: spacing (mean in the bin), n, one column per quantile named P50, P90, ..., non-decreasing in spacing, and share, the fraction of rows with uncertainty at or below threshold. Rows with a null or NaN are skipped.

used in

• Drill-hole spacing from virtual grids
• Drill-hole spacing from a learning curve