ehrapy.preprocessing.offset_negative_values

ehrapy.preprocessing.offset_negative_values#

ehrapy.preprocessing.offset_negative_values(edata, layer=None, copy=False)[source]#

Offsets negative values into positive ones with the lowest negative value becoming 0.

This is primarily used to enable the usage of functions such as log_norm that do not allow negative values for mathematical or technical reasons.

Supports both 2D and 3D data:

  • 2D data: Standard offset across observations

  • 3D data: Applied to all elements across samples and timestamps

Parameters:

  • edata (EHRData | AnnData) – Central data object.

  • layer (str, default: None) – The layer to offset.

  • copy (bool, default: False) – Whether to return a modified copy of the data object.

Return type:

EHRData | AnnData | None

Returns:

None if copy=False and modifies the passed edata, else returns an updated object.

Examples

>>> import ehrdata as ed
>>> import ehrapy as ep
>>> import numpy as np
>>> edata = ed.dt.physionet2012(layer="tem_data")
>>> np.nanmin(edata.layers["tem_data"])
-17.8
>>> ep.pp.offset_negative_values(edata, layer="tem_data")
>>> np.nanmin(edata.layers["tem_data"])
0.0