PowerSumRegularizer
- class PowerSumRegularizer(weight=1.0, dim=- 1, normalize=False, p=2.0, apply_only_once=False, parameters=None)[source]
Bases:
pykeen.regularizers.Regularizer
A simple x^p based regularizer.
Has some nice properties, cf. e.g. https://github.com/pytorch/pytorch/issues/28119.
Instantiate the regularizer.
- Parameters
weight (
float
) – The relative weight of the regularizationapply_only_once (
bool
) – Should the regularization be applied more than once after reset?parameters (
Optional
[Iterable
[Parameter
]]) – Specific parameters to track. if none given, it’s expected that your model automatically delegates to theupdate()
function.
Attributes Summary
The default strategy for optimizing the power sum regularizer's hyper-parameters
Methods Summary
forward
(x)Compute the regularization term for one tensor.
Attributes Documentation
- hpo_default: ClassVar[Mapping[str, Any]] = {'weight': {'high': 1.0, 'low': 0.01, 'scale': 'log', 'type': <class 'float'>}}
The default strategy for optimizing the power sum regularizer’s hyper-parameters
Methods Documentation