ontolearn.binders

Pyhon binders of other concept learners.

Classes

PredictedConcept

DLLearnerBinder

dl-learner python binder.

Module Contents

class ontolearn.binders.PredictedConcept(**kwargs)[source]
__iter__()[source]
class ontolearn.binders.DLLearnerBinder(binary_path=None, model=None, kb_path=None, storage_path='.', max_runtime=3)[source]

dl-learner python binder.

binary_path
kb_path
name
max_runtime
best_predictions = None
config_name_identifier = None
write_dl_learner_config(pos: List[str], neg: List[str], use_sparql=False) → str[source]

Writes config file for dl-learner.

Parameters:

  • pos – A list of URIs of individuals indicating positive examples in concept learning problem.

  • neg – A list of URIs of individuals indicating negatives examples in concept learning problem.

Returns:

Path of generated config file.

Return type:

str

fit(lp: PosNegLPStandard, max_runtime: int = None, use_sparql=False)[source]

Fit dl-learner model on a given positive and negative examples.

Parameters:

  • lp – PosNegLPStandard

  • problem. (lp.neg A list of URIs of individuals indicating negatives examples in concept learning)

  • problem.

  • max_runtime – Limit to stop the algorithm after n seconds.

Returns:

self.

best_hypotheses(n: int = None) → PredictedConcept[source]
best_hypothesis()[source]

Return predictions if exists.

Returns:

The prediction or the string ‘No prediction found.’

parse_dl_learner_output(output_of_dl_learner: List[str], file_path: str) → Dict[source]

Parse the output received from executing dl-learner.

Parameters:

  • output_of_dl_learner – The output of dl-learner to parse.

  • file_path – The file path to store the output.

Returns:

…, ‘Accuracy’: …, ‘F-measure’: …}.

Return type:

A dictionary of {‘Prediction’

static train(dataset: List = None) → None[source]

Dummy method, currently it does nothing.

abstract fit_from_iterable(dataset: List = None, max_runtime=None) → List[Dict][source]

Fit dl-learner model on a list of given positive and negative examples.

Parameters:

  • dataset – A list of tuple (s,p,n) where s => string representation of target concept, p => positive examples, i.e. s(p)=1 and n => negative examples, i.e. s(n)=0.

  • max_runtime – Limit to stop the algorithm after n seconds.

Returns:

self.