astromodule.hp.HyperParameter#

class HyperParameter[source]#

This abstract class is the generic representation of a hyperparmaeter and can be used on any model of Machine Learning or Deep Learning. This data model is designed to be a optuna wrapper, In this way, it is possible to implement a model training function without worry about how hyperparimeters are obtained.That is, it is possible Get the hyperparimeters automatically by the optuna library or predefined (manual) form with the same code.

Methods

clear_last_value()

Sets last_value attribute to None

from_dict(params)

Creates a concrete instance of hyperparmeter from a dictionary of the python

set_attr(key, value)

set_trial(trial)

suggest([trial])

Used to recover the value of the hyperparmeter as defined by the concrete class.The concrete class can implement a logic based in Auto-ML using the optuna library or predefined values.

to_dict([show_name])

A hyperparmeter can be defined from various attributes.
classmethod __new__(*args, **kwargs)#