sentency package

Submodules

sentency.logs module

sentency.logs.get_console_handler() → logging.StreamHandler

Get console handler.

Returns: logging.StreamHandler, StreamHandler which logs into stdout

sentency.logs.get_logger(name: str = 'sentency.logs', log_level: Union[str, int] = 40) → logging.Logger

Get logger.

name: Text, logger name log_level: Text or int, logging level; can be string name or integer value Returns: logging.Logger, logger instance

sentency.regex module

sentency.regex.regexize_keywords(keyword_str, keyword_delimiter=' ', line_delimiter='\n', case_insensitive=True)

Convert a string of keywords into a regular expression that can be used as input for the sentenCy sentex component. You can separate keywords individually or into groups using keyword and line delimiters.

Example usage: >>> from sentency.regex import regexize_keywords >>> keyword_str = “abdominal aortic aneurysmnaneurysm abdominal aorta” >>> regexize_keywords(keyword_str) (?i)((abdominal.*aortic.*aneurysm)|(aneurysm.*abdominal.*aorta))

keyword_str: str, The keyword string to be converted into a regular expression. keyword_delimiter: str, The delimiter separating individual keywords in keyword_str. Default is ‘ ‘ line_delimiter: str, The string separating lines into a regular expression. Default is ‘n’ case_insensitive: bool, Should the regular expression be case-insensitive? RETURNS: str, The regular expression

sentency.sentency module

class sentency.sentency.Sentex(nlp: spacy.language.Language, sentence_regex: str, ignore_regex: str, annotate_ents: bool, label: str)

Bases: object

Sentex is a spaCy pipeline component that adds spans to the list Doc._.sentex based on regular expression matches within each sentence of the document. If an ignore_regex is given, sentences matching that regular expression will be ignored.

nlp: Language,

A required argument for spacy to use this as a factory

sentence_regex : str,

A regular expression to match spans within each sentence of the document.

ignore_regex : str,

A regular expression to identify sentences that should be ignored.

annotate_ents: bool,

Write/overwrite matches to Doc.ents

label: str,

If annotate_ents == True, the label for the matched entity

set_annotations(doc)

Modify the document in place. Logic taken from spacy.pipeline.entityruler.EntityRuler

class sentency.sentency.Size(nlp: spacy.language.Language, size_regex: str, sentex_only: bool, annotate_ents: bool, label: str)

Bases: object

Size is a spaCy pipeline component that adds spans to the list Doc._.size based on regular expression matches within each sentence of the document.

nlp: Language,

A required argument for spacy to use this as a factory

size_regex : str,

A regular expression to match spans within each sentence of the document.

sentex_only : bool,

Only match in sentences with Sentex-matched entities

annotate_ents: bool,

Write/overwrite matches to Doc.ents

label: str,

If annotate_ents == True, the label for the matched entity

parse_size(text: str) → float

process(doc: spacy.tokens.doc.Doc) → spacy.tokens.doc.Doc

set_annotations(doc)

Modify the document in place. Logic taken from spacy.pipeline.entityruler.EntityRuler

sentency.sentency.create_sentex_component(nlp: spacy.language.Language, name: str, sentence_regex: str, ignore_regex: str, annotate_ents: bool, label: str)

sentency.sentency.create_size_component(nlp: spacy.language.Language, name: str, size_regex: str, sentex_only: bool, annotate_ents: bool, label: str)

Module contents

Top-level package for sentency.