soft.fuzzy.unsupervised.granulation.offline package
Submodules
soft.fuzzy.unsupervised.granulation.offline.expert module
Implements the Expert Partitioning algorithm. Basically, Expert Partitioning finds the minimum and maximum values possible in some dimension, and then creates (reasonable) fuzzy sets across it such that each fuzzy set has reasonable spread or coverage of the domain space.
- soft.fuzzy.unsupervised.granulation.offline.expert.apply_expert_partitioning(dataset: SupervisedDataset, config: Config) LinguisticVariables
Expert partitioning, is an unsupervised learning algorithm that produces a fixed number of fuzzy sets for the given dataset.
- Parameters:
dataset – The dataset.
config – The configuration settings. Expects config.fuzzy.partition.max to exist,
make. (which is the number of partitions to)
- Returns:
A list of fuzzy sets that inherit from ContinuousFuzzySet.
- soft.fuzzy.unsupervised.granulation.offline.expert.expert_partitioning(tensor: Tensor, config: Config, minimums: Tensor, maximums: Tensor, terms: List[Lorentzian] | None = None) List[Lorentzian]
Expert partitioning, is an unsupervised learning algorithm that produces Lorentzian fuzzy sets.
- Parameters:
tensor – The selected tensor from the dataset.
config – The configuration settings.
minimums – The minimum values in each dimension.
maximums – The maximum values in each dimension.
terms – A list of terms to start from, if any.
- Returns:
A list of Lorentzian fuzzy sets along the tensor space.