Ctrl+K

mloptimizer.hyperparams

mloptimizer.hyperparams#

Submodules#

Package Contents#

Classes#

Hyperparam

Class to define a hyperparam to optimize. It defines the name, min value, max value and type.

HyperparameterSpace

This class represents the hyperparameter space for a scikit-learn classifier. It contains the fixed hyperparameters
class mloptimizer.hyperparams.Hyperparam(name: str, min_value: int, max_value: int, hyperparam_type: str, scale: int = 100, values_str: list = None)[source]#

Bases: object

Class to define a hyperparam to optimize. It defines the name, min value, max value and type. This is used to control the precision of the hyperparam and avoid multiple evaluations with close values of the hyperparam due to decimal positions.

name#

Name of the hyperparam. It will be used as key in a dictionary

Type:

str

min_value#

Minimum value of the hyperparam

Type:

int

max_value#

Maximum value of the hyperparam

Type:

int

hyperparam_type#

Type of the hyperparam (‘int’, ‘float’, ‘nexp’, ‘x10’)

Type:

str

scale#

Optional param in case the type=float

Type:

int, optional (default=100)

values_str#

List of string with possible values (TODO)

Type:

list, optional (default=[])

correct(value: int)[source]#
Returns the real value of the hyperparam in case some mutation could surpass the limits.

  1. Verifies the input is int

  2. Enforce min and max value

  3. Apply the type of value

Parameters:

value (int) – Value to correct

Returns:

ret – Corrected value

Return type:

int, float

class mloptimizer.hyperparams.HyperparameterSpace(fixed_hyperparams: dict, evolvable_hyperparams: dict)[source]#

This class represents the hyperparameter space for a scikit-learn classifier. It contains the fixed hyperparameters and the evolvable hyperparameters. The fixed hyperparameters are just a dictionary with the hyperparameters that are not going to be optimized and their value. The evolvable hyperparameters are a dictionary with the hyperparameters that are going to be optimized. The keys are the hyperparameter names and the values are instances of the Hyperparam class.

fixed_hyperparams#

Dictionary with the fixed hyperparameters

Type:

dict

evolvable_hyperparams#

Dictionary with the evolvable hyperparameters of Hyperparam instances

Type:

dict

default_hyperparameter_spaces_json#
classmethod from_json(file_path)[source]#

This method creates a HyperparameterSpace object from a JSON file.

Parameters:

file_path (str) – Path to the JSON file

Return type:

HyperparameterSpace

Raises:

  • FileNotFoundError – If the file does not exist

  • json.JSONDecodeError – If the file is not a valid JSON file

to_json(file_path, overwrite=False)[source]#

This method saves the hyperparameter space as a JSON file.

Parameters:

  • file_path (str) – Path to the JSON file

  • overwrite (bool, optional (default=False)) – If True, the file will be overwritten if it exists. If False, a FileExistsError will be raised if the file exists

Raises:

  • ValueError – If the file path is None

  • FileExistsError – If the file exists and overwrite is False

static get_default_hyperparameter_space(estimator_class)[source]#

This method returns a dictionary with the default hyperparameters for the scikit-learn classifier. It reads the default_hyperparameter_spaces.json file and returns the hyperparameters for the classifier

Parameters:

estimator_class (class) – The scikit-learn classifier class

Returns:

The hyperparameter space for the classifier

Return type:

HyperparameterSpace

Contents