safe_nanmedian_filter

grizli.utils.safe_nanmedian_filter(data, filter_kwargs={}, filter_footprint=None, axis=1, clean=True, cval=0.0)

Run nanmedian filter on data

Parameters

dataarray-like

The 2D data to filter

filter_kwargsdict

Arguments to make_filter_footprint to make a 1D filter

filter_footprintarray-like

Specify the filter explicitly. If 1D, then will apply the filter over axis=1 (i.e., x) of data

axis0 or 1

Axis over which to apply the filter (axis=1 filters on rows)

clean, cvalbool, scalar

Replace nan in the output with cval

Returns

filter_dataarray-like

Filtered data

filter_namestr

The type of filter that was applied: nbutils.nanmedian if nanmedian was imported successfully and median_filter for the fallback to scip.ndimage.median_filter otherwise.