nispace.stats.misc.zscore_df

nispace.stats.misc.zscore_df(df, along='cols', force_df=True)[source]

Z-standardize a DataFrame/Series/ndarray, preserving its type.

Plain numpy/pandas (via scipy.stats.zscore, ddof=0, NaN omitted per column/row). Widely used across NiSpace (api.py, core/permute.py, core/clean_y.py, core/nullmaps.py, datasets.py) for self-normalization of maps/vectors (population, not sample, convention).

Parameters:

  • df (pandas dataframe, Series, or ndarray) – input data

  • along (str, optional) – Either “cols” or “rows”. Defaults to “cols”.

  • force_df (bool, optional) – for Series/ndarray input, wrap the output in a DataFrame instead of returning a Series/ndarray. Defaults to True.

Returns:

standardized dataframe/series

Return type:

pd.DataFrame or pd.Series

Raises:

  • ValueError – if along is not “cols” or “rows”.

  • TypeError – if df is not a DataFrame, Series, or ndarray.