instancelib.ingest.qrel module

class instancelib.ingest.qrel.Qrel(topic, doc_id, relevancy)

Bases: object

Parameters:

• topic (str) –

• doc_id (str) –

• relevancy (int) –

doc_id: str

relevancy: int

topic: str

class instancelib.ingest.qrel.TrecDataset(docids, texts, qrels, topics, pos_label='Relevant', neg_label='Irrelevant')

Bases: object

Parameters:

• docids (Mapping[str, FrozenSet[str]]) –

• texts (Mapping[str, Mapping[str, Mapping[str, str]]]) –

• qrels (Mapping[str, DataFrame]) –

• topics (DataFrame) –

• pos_label (str) –

• neg_label (str) –

classmethod from_path(base_dir)

Parameters:

base_dir (Path) –

get_document(topic_key, doc_id)

Parameters:

• topic_key (str) –

• doc_id (str) –

Return type:

str

get_documents(topic_key)

Parameters:

topic_key (str) –

Return type:

FrozenSet[str]

get_env(topic_key)

Parameters:

topic_key (str) –

Return type:

TextEnvironment[str, ndarray[Any, dtype[Any]], str]

get_envs()

Return type:

Mapping[str, TextEnvironment[str, ndarray[Any, dtype[Any]], str]]

get_labels(topic_key, document)

Parameters:

• topic_key (str) –

• document (str) –

Return type:

FrozenSet[str]

get_topicqrels(topic_key)

Parameters:

topic_key (str) –

Return type:

DataFrame

instancelib.ingest.qrel.build_doc_map(topic_docs)

Parameters:

topic_docs (Mapping[str, Mapping[str, Mapping[str, str]]]) –

Return type:

Mapping[str, Set[str]]

instancelib.ingest.qrel.hidden(p)

Parameters:

p (Path) –

Return type:

bool

instancelib.ingest.qrel.read_docids(docid_file)

Parameters:

docid_file (Path) –

Return type:

FrozenSet[str]

instancelib.ingest.qrel.read_doctexts(doctext_file)

Parameters:

doctext_file (Path) –

Return type:

Optional[Mapping[str, Mapping[str, str]]]

instancelib.ingest.qrel.read_qrel(qrel_file)

Parameters:

qrel_file (Path) –

Return type:

DataFrame

instancelib.ingest.qrel.read_qrel_dataset(base_dir)

Parameters:

base_dir (Path) –

instancelib.ingest.qrel.read_topics(topic_dir)

Parameters:

topic_dir (Path) –

Return type:

DataFrame