watex.utils.funcutils.map_specific_columns

watex.utils.funcutils.map_specific_columns(X, ufunc, columns_to_skip=None, pattern=None, inplace=False, **kws)

Apply function to a specific columns is the dataframe.

It is possible to skip some columns that we want operation to not be performed.

Parameters:

• X (dataframe,) – pandas dataframe with valid columns

• ufunc (callable,) – Universal function that can be applying to the dataframe.

• columns_to_skip (list or str ,) – List of columns to skip. If given as string and separed by the default pattern items, it should be converted to a list and make sure the columns name exist in the dataframe. Otherwise an error with raise.

• pattern (str, default = '[#&*@!,;s]s*') –

  The base pattern to split the text in column2skip into a columns For instance, the following string coulb be splitted to:

  'depth_top, thickness, sp, gamma_gamma' ->
  ['depth_top', 'thickness', 'sp', 'gamma_gamma']
  

  Refer to str2columns() for further details.

• inplace (bool, default=True) – Modified dataframe in place and return None, otherwise return a new dataframe

• kws (dict,) – Keywords argument passed to :func: pandas.DataFrame.apply function

Returns:

X – Dataframe modified inplace with values computed using the given func`except the skipped columns, or ``None` if inplace is True.

Return type:

Dataframe or None

Examples

>>> from watex.datasets import load_hlogs
>>> from watex.utils.plotutils import map_specific_columns
>>> X0, _= load_hlogs (as_frame =True )
>>> # let visualize the  first3 values of `sp` and `resistivity` keys
>>> X0['sp'][:3] , X0['resistivity'][:3]
... (0   -1.580000
     1   -1.580000
     2   -1.922632
     Name: sp, dtype: float64,
     0    15.919130
     1    16.000000
     2    24.422316
     Name: resistivity, dtype: float64)
>>> column2skip = ['hole_id','depth_top', 'depth_bottom',
                  'strata_name', 'rock_name', 'well_diameter', 'sp']
>>> map_specific_columns (X0, ufunc = np.log10, column2skip)
>>> # now let visualize the same keys values
>>> X0['sp'][:3] , X0['resistivity'][:3]
... (0   -1.580000
     1   -1.580000
     2   -1.922632
     Name: sp, dtype: float64,
     0    1.201919
     1    1.204120
     2    1.387787
     Name: resistivity, dtype: float64)
>>> # it is obvious the `resistiviy` values is log10
>>> # while `sp` stil remains the same