def main(argv):
argv = docopt.docopt(__doc__, argv=argv)
print(argv)
scan.scan_document(source=argv['SOURCE'],
directory=argv['--out'],
suffix='.negbio.xml',
fn=clean_sentences,
non_sequences=[])
def main(argv):
argv = docopt.docopt(__doc__, argv=argv)
print(argv)
ptb2dep = Ptb2DepConverter(universal=True)
lemmatizer = Lemmatizer()
scan.scan_document(source=argv['SOURCE'], directory=argv['--out'], suffix='.ud.xml',
fn=convert, non_sequences=[ptb2dep, lemmatizer])
def main(argv):
argv = docopt.docopt(__doc__, argv=argv)
print(argv)
parser = Bllip(model_dir=os.path.expanduser(argv['--model']))
scan.scan_document(source=argv['SOURCE'],
directory=argv['--out'],
suffix='.bllip.xml',
fn=parse,
non_sequences=[parser])
def main(argv):
argv = docopt.docopt(__doc__, argv=argv)
print(argv)
neg_detector = Detector(argv['--neg-patterns'],
argv['--uncertainty-patterns'])
scan.scan_document(source=argv['SOURCE'],
directory=argv['--out'],
suffix='.neg.xml',
fn=detect,
non_sequences=[neg_detector])
def main(argv):
argv = docopt.docopt(__doc__, argv=argv)
print(argv)
splitter = NltkSSplitter(newline=argv['--newline_is_sentence_break'])
scan.scan_document(source=argv['SOURCE'],
directory=argv['--out'],
suffix='.ss.xml',
fn=ssplit,
non_sequences=[splitter])
def main(argv):
argv = docopt.docopt(__doc__, argv=argv)
print(argv)
splitter = ssplit.NltkSSplitter(newline=True)
parser = parse.Bllip(model_dir=argv['--model'])
ptb2dep = ptb2ud.Ptb2DepConverter(universal=True)
lemmatizer = ptb2ud.Lemmatizer()
neg_detector = negdetect.Detector(argv['--neg-patterns'],
argv['--uncertainty-patterns'])
scan.scan_document(
source=argv['SOURCE'],
directory=argv['--out'],
suffix='.neg.xml',
fn=pipeline,
non_sequences=[splitter, parser, ptb2dep, lemmatizer, neg_detector])
"""
Clean up sentences
Usage:
negbio_pipeline cleanup [options] --output=<directory> <file> ...
Options:
--suffix=<suffix> Append an additional SUFFIX to file names. [default: .negbio.xml]
--verbose Print more information about progress.
--output=<directory> Specify the output directory.
"""
from negbio.cli_utils import parse_args
from negbio.pipeline.cleanup import clean_sentences
from negbio.pipeline.scan import scan_document
if __name__ == '__main__':
argv = parse_args(__doc__)
scan_document(source=argv['<file>'],
directory=argv['--output'],
suffix=argv['--suffix'],
fn=clean_sentences)
"""
Usage:
negbio_pipeline normalize [options] --output=<directory> <file> ...
Options:
--output=<directory> Specify the output directory.
--suffix=<suffix> Append an additional SUFFIX to file names. [default: .normalized.xml]
--verbose Print more information about progress.
"""
from negbio.cli_utils import parse_args
from negbio.ext.normalize_mimiccxr import normalize
from negbio.pipeline.scan import scan_document
if __name__ == '__main__':
argv = parse_args(__doc__)
scan_document(source=argv['<file>'],
verbose=argv['--verbose'],
suffix=argv['--suffix'],
directory=argv['--output'],
fn=normalize)
"""
Split text into sentences
Usage:
negbio_pipeline ssplit [options] --output=<directory> <file> ...
Options:
--newline_is_sentence_break Whether to treat newlines as sentence breaks. True means that a newline is always a
sentence break. False means to ignore newlines for the purpose of sentence
splitting. This is appropriate for continuous text, when just the non-whitespace
characters should be used to determine sentence breaks. [default=False]
--suffix=<suffix> Append an additional SUFFIX to file names. [default: .ssplit.xml]
--output=<directory> Specify the output directory.
--verbose Print more information about progress.
"""
from negbio.pipeline.scan import scan_document
from negbio.pipeline.ssplit import NegBioSSplitter
from negbio.cli_utils import parse_args
if __name__ == '__main__':
argv = parse_args(__doc__)
splitter = NegBioSSplitter(newline=argv['--newline_is_sentence_break'])
scan_document(source=argv['<file>'],
directory=argv['--output'],
suffix=argv['--suffix'],
fn=splitter.split_doc,
non_sequences=[])
negbio_pipeline neg [options] --output=<directory> <file> ...
Options:
--neg-patterns=<file> Specify negation rules [default: negbio/patterns/neg_patterns.txt]
--uncertainty-patterns=<file> Specify uncertainty rules [default: negbio/patterns/uncertainty_patterns.txt]
--suffix=<suffix> Append an additional SUFFIX to file names. [default: .neg.xml]
--verbose Print more information about progress.
--output=<directory> Specify the output directory.
"""
import os
from negbio.cli_utils import parse_args, get_absolute_path
from negbio.neg.neg_detector import Detector
from negbio.pipeline.negdetect import detect
from negbio.pipeline.scan import scan_document
if __name__ == '__main__':
argv = parse_args(__doc__)
argv = get_absolute_path(argv,
'--neg-patterns',
'negbio/patterns/neg_patterns.txt')
argv = get_absolute_path(argv,
'--uncertainty-patterns',
'negbio/patterns/uncertainty_patterns.txt')
neg_detector = Detector(os.path.realpath(argv['--neg-patterns']),
os.path.realpath(argv['--uncertainty-patterns']))
scan_document(source=argv['<file>'], directory=argv['--output'], suffix=argv['--suffix'],
fn=detect, non_sequences=[neg_detector])
negbio_pipeline section_split [options] --output=<directory> <file> ...
Options:
--suffix=<suffix> Append an additional SUFFIX to file names. [default: .secsplit.xml]
--output=<directory> Specify the output directory.
--verbose Print more information about progress.
--pattern=<file> Specify section title list for matching.
"""
import re
from negbio.cli_utils import parse_args
from negbio.pipeline.scan import scan_document
from negbio.pipeline.section_split import split_document
def read_section_titles(pathname):
with open(pathname) as fp:
return re.compile('|'.join(fp.readlines()), re.MULTILINE)
if __name__ == '__main__':
argv = parse_args(__doc__)
if argv['--pattern'] is None:
patterns = None
else:
patterns = read_section_titles(argv['--pattern'])
scan_document(source=argv['<file>'], verbose=argv['--verbose'], suffix=argv['--suffix'],
directory=argv['--output'], fn=split_document, non_sequences=[patterns])