pandas.DataFrame.opti.to_paramdict#

DataFrame.opti.to_paramdict()[source]#

Cast a single-column DataFrame into a ParamDict1D/ParamDictND.

Note: The opti-extensions package has to be imported first to use this method with pandas.

Returns:

ParamDict1D or ParamDictND

A single-index, single-column DataFrame will be cast into a ParamDict1D having index label as keys and column values as values. The index name will be set as the ParamDict1D.key_name attribute (as str). The column name will be set as the ParamDict1D.value_name attribute (as str).
A multi-index, single-column DataFrame will be cast into a ParamDictND having tuple of index labels as keys and column values as values. The multi-index names will be set as the ParamDictND.key_names attribute (as list[str]). The column name will be set as the ParamDictND.value_name attribute (as str).

Raises:

ValueError: If the DataFrame is empty.
ValueError: If the DataFrame has multiple columns.
ValueError: If the DataFrame has duplicate index label(s).
TypeError: If the DataFrame has index label(s) that are not scalar (any iterable except string).
TypeError: If the DataFrame column values are not int or float.

See also

pandas.Series.opti.to_paramdict: Cast a Series into a ParamDict1D/ParamDictND.

Examples

>>> import pandas as pd
>>> import opti_extensions  # required to access this method

Casting a single-index, single-column DataFrame into a ParamDict1D

>>> inventory = pd.DataFrame({'PRODUCT': ['chair', 'desk'], 'UNITS': [200, 500]})
>>> inventory
  PRODUCT  UNITS
0   chair    200
1    desk    500
>>> inventory_units = inventory.set_index('PRODUCT')
>>> inventory_units
         UNITS
PRODUCT
chair      200
desk       500

>>> inventory_units.opti.to_paramdict()
ParamDict1D: PRODUCT -> UNITS
{'chair': 200, 'desk': 500}

Casting a multi-index, single-column DataFrame into a ParamDictND

>>> routes = pd.DataFrame(
...     {
...         'ORI': ['Delhi', 'Delhi', 'Seattle', 'Tokyo'],
...         'DES': ['Seattle', 'Tokyo', 'Tokyo', 'Delhi'],
...         'DIST': [11303, 5836, 7695, 5830],
...     }
... )
>>> routes
       ORI      DES   DIST
0    Delhi  Seattle  11303
1    Delhi    Tokyo   5836
2  Seattle    Tokyo   7695
3    Tokyo    Delhi   5830
>>> routes_dist = routes.set_index(['ORI', 'DES'])
>>> routes_dist
                  DIST
ORI     DES
Delhi   Seattle  11303
        Tokyo     5836
Seattle Tokyo     7695
Tokyo   Delhi     5830

>>> routes_dist.opti.to_paramdict()
ParamDictND: (ORI, DES) -> DIST
{('Delhi', 'Seattle'): 11303,
 ('Delhi', 'Tokyo'): 5836,
 ('Seattle', 'Tokyo'): 7695,
 ('Tokyo', 'Delhi'): 5830}