Zeroth¶

class hdnnpy.training.loss_function.Zeroth(model, properties, **_)[source]¶

Bases: hdnnpy.training.loss_function.loss_functions_base.LossFunctionBase

Loss function to optimize 0th-order property.

Parameters:	model (HighDimensionalNNP) – HDNNP object to optimize parameters. properties (list [str]) – Names of properties to optimize.

eval(**dataset)[source]¶

Calculate loss function from given datasets and model.

Parameters:	*dataset (ndarray*) – Datasets passed as kwargs. Name of each key is in the format ‘inputs/N’ or ‘labels/N’. ‘N’ is the order of the dataset.
Returns:	A scalar value calculated with loss function.
Return type:	Variable

name = 'zeroth'¶

Name of this loss function class.

Type:	str

order = {'descriptor': 0, 'property': 0}¶

Required orders of each dataset to calculate loss function.

Type:	dict