sys.path.insert(0, '../..')
from lxml import etree
from rdflib import Graph
from nltk.corpus import framenet as fn
import FrameNetNLTK
from FrameNetNLTK import add_annotations_from_naf_31
from FrameNetNLTK import load, get_luid
from rdflib.term import URIRef
naf_path = 'test_naf_files/predicate_in_compound.naf'
corpus_name = 'HDD'
corpus_description = 'HistoricalDistanceData'
my_fn = load(folder='test_lexicon', verbose=2)
# load Dutch FrameNet in Lemon
path_dfn_in_lemon = 'stats/dfn_0.1.ttl'
dfn_in_lemon = Graph()
dfn_in_lemon.parse(path_dfn_in_lemon, format='ttl')
# update NAF file with correct LU URIs
parser = etree.XMLParser(remove_blank_text=True, strip_cdata=False)
doc = etree.parse(naf_path, parser)
# ext_ref_el for predicate annotation
ext_ref_els = doc.findall('srl/predicate/externalReferences/externalRef')
assert len(ext_ref_els
) == 1, f'expected 2 externalRef elements, found {len(ext_ref_els)}'
ext_ref_el = ext_ref_els[0]
verbose=2)
add_lu_to_info(your_fn=fn,
language='en',
premon=LexicalDataD2TAnnotationTool.premon,
namespace='http://rdf.cltl.nl/efn/',
major_version=1,
minor_version=7,
output_folder=out_dir,
verbose=2)
add_lemma_to_pos_to_lu_urls(output_folder=out_dir,
language='en',
verbose=2)
fn_nl = load('res/DutchFrameNet-0.1')
add_lu_to_info(your_fn=fn_nl,
language='nl',
premon=LexicalDataD2TAnnotationTool.premon,
namespace='http://rdf.cltl.nl/dfn/',
major_version=0,
minor_version=1,
output_folder=out_dir,
verbose=2)
add_lemma_to_pos_to_lu_urls(output_folder=out_dir,
language='nl',
verbose=2)
import sys
import pytest
sys.path.insert(0, '../..')
from FrameNetNLTK import load, remove_lu, get_luid
fn = load('../test/test_lexicon')
lu_id, reason = get_luid(my_fn=fn,
frame_label='People_by_origin',
lemma='Fransman',
pos='N')
remove_lu(your_lexicon_folder='test_lexicon',
lu_id=lu_id,
verbose=2)
lu_id, reason = get_luid(my_fn=fn,
frame_label='Appellations',
lemma='president',
pos='N')
try:
remove_lu(your_lexicon_folder='test_lexicon',
lu_id=lu_id,
verbose=2)
except AssertionError:
print(f'AssertionError was correctly raised for lu_id {lu_id} since it is part of an endocentric compound.')
import sys
sys.path.insert(0, '../..')
from FrameNetNLTK import load, get_frame_stats_df, get_lu_stats_df, get_lu_per_pos_stats_df, get_lexeme_stats_df, get_ambiguity_df
from FrameNetNLTK import get_stats_html
my_fn = load('test_lexicon')
frame_df = get_frame_stats_df(my_fn)
print(frame_df)
lu_stats_df = get_lu_stats_df(my_fn)
print(lu_stats_df)
lu_per_pos_stats_df = get_lu_per_pos_stats_df(my_fn)
print(lu_per_pos_stats_df)
lexeme_stats_df = get_lexeme_stats_df(my_fn)
print(lexeme_stats_df)
ambiguity_df = get_ambiguity_df(my_fn)
print(ambiguity_df)
get_stats_html(your_fn=my_fn, html_path='stats/descriptive_statistics.html')