verticapy.machine_learning.model_selection.hp_tuning.parameter_grid#

verticapy.machine_learning.model_selection.hp_tuning.parameter_grid(param_grid: dict) → list[dict]#

Generates a list of the different combinations of input parameters.

Parameters#

param_grid: dict: dictionary of parameters.

Returns#

list of dict: list of the different combinations.

Examples#

Its easy to generate a list of the different combinations of input parameters.

from verticapy.machine_learning.model_selection.hp_tuning.param_gen import parameter_grid

parameter_grid(
    {
        "nbins": [10, 100, 360],
        "alpha": [0.1, 0.3, 0.5],
        "solver": ["bfgs", "newton"],
    },
)

Out[2]: 
[{'nbins': 10, 'alpha': 0.1, 'solver': 'bfgs'},
 {'nbins': 10, 'alpha': 0.1, 'solver': 'newton'},
 {'nbins': 10, 'alpha': 0.3, 'solver': 'bfgs'},
 {'nbins': 10, 'alpha': 0.3, 'solver': 'newton'},
 {'nbins': 10, 'alpha': 0.5, 'solver': 'bfgs'},
 {'nbins': 10, 'alpha': 0.5, 'solver': 'newton'},
 {'nbins': 100, 'alpha': 0.1, 'solver': 'bfgs'},
 {'nbins': 100, 'alpha': 0.1, 'solver': 'newton'},
 {'nbins': 100, 'alpha': 0.3, 'solver': 'bfgs'},
 {'nbins': 100, 'alpha': 0.3, 'solver': 'newton'},
 {'nbins': 100, 'alpha': 0.5, 'solver': 'bfgs'},
 {'nbins': 100, 'alpha': 0.5, 'solver': 'newton'},
 {'nbins': 360, 'alpha': 0.1, 'solver': 'bfgs'},
 {'nbins': 360, 'alpha': 0.1, 'solver': 'newton'},
 {'nbins': 360, 'alpha': 0.3, 'solver': 'bfgs'},
 {'nbins': 360, 'alpha': 0.3, 'solver': 'newton'},
 {'nbins': 360, 'alpha': 0.5, 'solver': 'bfgs'},
 {'nbins': 360, 'alpha': 0.5, 'solver': 'newton'}]

Note

This function is essential for conducting a grid_search_cv().