LocalDatasetLoader¶

class LocalDatasetLoader(directory=None)[source]¶

Bases: chemicalx.data.datasetloader.DatasetLoader, abc.ABC

A dataset loader that processes and caches data locally.

Attributes Summary

`contexts_name`
`features_name`
`labels_name`
`structures_name`

Methods Summary

`get_context_features`()	Get the context feature set.
`get_drug_features`()	Get the drug feature set.
`get_labeled_triples`()	Get the labeled triples dataframe.
`preprocess`()	Download and preprocess the dataset.
`write_contexts`(contexts)	Write the context feature set.
`write_drugs`(drugs)	Write the drug data.
`write_labels`(df)	Write the labeled triples dataframe.

Attributes Documentation

contexts_name: ClassVar[str] = 'contexts.tsv'¶

features_name: ClassVar[str] = 'features.tsv'¶

labels_name: ClassVar[str] = 'labels.tsv'¶

structures_name: ClassVar[str] = 'structures.tsv'¶

Methods Documentation

get_context_features()[source]¶

Get the context feature set.

Return type: ContextFeatureSet

get_drug_features()[source]¶

Get the drug feature set.

Return type: DrugFeatureSet

get_labeled_triples()[source]¶

Get the labeled triples dataframe.

Return type: LabeledTriples

abstract preprocess()[source]¶

Download and preprocess the dataset.

The implementation of this function should write to all three of self.drugs_path, self.contexts_path, and self.labels_path using respectively write_drugs(), write_contexts(), and write_labels().

write_contexts(contexts)[source]¶: Write the context feature set.

write_drugs(drugs)[source]¶

Write the drug data.

Return type: None

write_labels(df)[source]¶: Write the labeled triples dataframe.