def test_elmo_offline():
mentions = get_embedd_mentions()
embeder = ElmoEmbedding()
for mention in mentions:
embeder.get_head_feature_vector(mention)
elmo_embeddings = WordEmbeddingRelationExtraction(EmbeddingMethod.ELMO_OFFLINE, elmo_file=None)
elmo_embeddings.embedding.embeder = embeder.cache
assert (
elmo_embeddings.extract_all_relations(mentions[0], mentions[0]).pop()
== RelationType.WORD_EMBEDDING_MATCH
)
assert (
elmo_embeddings.extract_all_relations(mentions[0], mentions[1]).pop()
== RelationType.WORD_EMBEDDING_MATCH
)
assert (
elmo_embeddings.extract_all_relations(mentions[0], mentions[2]).pop()
== RelationType.NO_RELATION_FOUND
)
assert (
elmo_embeddings.extract_all_relations(mentions[0], mentions[3]).pop()
== RelationType.WORD_EMBEDDING_MATCH
)
def test_elmo_online():
mentions = get_embedd_mentions()
elmo_embeddings = WordEmbeddingRelationExtraction(EmbeddingMethod.ELMO)
assert elmo_embeddings.extract_all_relations(
mentions[0], mentions[0]).pop() == RelationType.WORD_EMBEDDING_MATCH
assert elmo_embeddings.extract_all_relations(
mentions[0], mentions[1]).pop() == RelationType.WORD_EMBEDDING_MATCH
assert elmo_embeddings.extract_all_relations(
mentions[0], mentions[2]).pop() == RelationType.NO_RELATION_FOUND
assert elmo_embeddings.extract_all_relations(
mentions[0], mentions[3]).pop() == RelationType.WORD_EMBEDDING_MATCH
def run_example():
logger.info("Running relation extraction example......")
computed = ComputedRelationExtraction()
ref_dict = ReferentDictRelationExtraction(
ref_dict=str(LIBRARY_ROOT / "datasets" / "coref.dict1.tsv"))
vo = VerboceanRelationExtraction(
vo_file=str(LIBRARY_ROOT / "datasets" /
"verbocean.unrefined.2004-05-20.txt"))
wiki = WikipediaRelationExtraction()
wn = WordnetRelationExtraction()
embed = WordEmbeddingRelationExtraction(method=EmbeddingMethod.ELMO)
mention_x1 = MentionDataLight(
"IBM",
mention_context=
"IBM manufactures and markets computer hardware, middleware and software",
)
mention_y1 = MentionDataLight(
"International Business Machines",
mention_context="International Business Machines Corporation is an "
"American multinational information technology company",
)
computed_relations = computed.extract_all_relations(mention_x1, mention_y1)
ref_dict_relations = ref_dict.extract_all_relations(mention_x1, mention_y1)
vo_relations = vo.extract_all_relations(mention_x1, mention_y1)
wiki_relations = wiki.extract_all_relations(mention_x1, mention_y1)
embed_relations = embed.extract_all_relations(mention_x1, mention_y1)
wn_relaions = wn.extract_all_relations(mention_x1, mention_y1)
if RelationType.NO_RELATION_FOUND in computed_relations:
logger.info("No Computed relation found")
else:
logger.info("Found Computed relations-%s",
str(list(computed_relations)))
if RelationType.NO_RELATION_FOUND in ref_dict_relations:
logger.info("No Referent-Dict relation found")
else:
logger.info("Found Referent-Dict relations-%s",
str(list(ref_dict_relations)))
if RelationType.NO_RELATION_FOUND in vo_relations:
logger.info("No Verb-Ocean relation found")
else:
logger.info("Found Verb-Ocean relations-%s", str(list(vo_relations)))
if RelationType.NO_RELATION_FOUND in wiki_relations:
logger.info("No Wikipedia relation found")
else:
logger.info("Found Wikipedia relations-%s", str(wiki_relations))
if RelationType.NO_RELATION_FOUND in embed_relations:
logger.info("No Embedded relation found")
else:
logger.info("Found Embedded relations-%s", str(list(embed_relations)))
if RelationType.NO_RELATION_FOUND in wn_relaions:
logger.info("No Wordnet relation found")
else:
logger.info("Found Wordnet relations-%s", str(wn_relaions))
def run_example():
logger.info('Running relation extraction example......')
computed = ComputedRelationExtraction()
ref_dict = ReferentDictRelationExtraction(OnlineOROfflineMethod.ONLINE,
LIBRARY_ROOT + '/datasets/ref.dict1.tsv')
vo = VerboceanRelationExtraction(OnlineOROfflineMethod.ONLINE,
LIBRARY_ROOT + '/datasets/verbocean.unrefined.2004-05-20.txt')
wiki = WikipediaRelationExtraction(WikipediaSearchMethod.ONLINE)
wn = WordnetRelationExtraction(OnlineOROfflineMethod.ONLINE)
mention_x1 = MentionDataLight(
'IBM',
mention_context='IBM manufactures and markets computer hardware, middleware and software')
mention_y1 = MentionDataLight(
'International Business Machines',
mention_context='International Business Machines Corporation is an '
'American multinational information technology company')
computed_relations = computed.extract_all_relations(mention_x1, mention_y1)
ref_dict_relations = ref_dict.extract_all_relations(mention_x1, mention_y1)
vo_relations = vo.extract_all_relations(mention_x1, mention_y1)
wiki_relations = wiki.extract_sub_relations(mention_x1, mention_y1,
RelationType.WIKIPEDIA_REDIRECT_LINK)
embed = WordEmbeddingRelationExtraction(
EmbeddingMethod.ELMO)
embed_relations = embed.extract_all_relations(mention_x1, mention_y1)
wn_relaions = wn.extract_sub_relations(mention_x1, mention_y1,
RelationType.WORDNET_DERIVATIONALLY)
if RelationType.NO_RELATION_FOUND in computed_relations:
logger.info('No Computed relation found')
else:
logger.info('Found Computed relations-%s', str(list(computed_relations)))
if RelationType.NO_RELATION_FOUND in ref_dict_relations:
logger.info('No Referent-Dict relation found')
else:
logger.info('Found Referent-Dict relations-%s', str(list(ref_dict_relations)))
if RelationType.NO_RELATION_FOUND in vo_relations:
logger.info('No Verb-Ocean relation found')
else:
logger.info('Found Verb-Ocean relations-%s', str(list(vo_relations)))
if RelationType.NO_RELATION_FOUND in wiki_relations:
logger.info('No Wikipedia relation found')
else:
logger.info('Found Wikipedia relations-%s', str(list(wiki_relations)))
if RelationType.NO_RELATION_FOUND in embed_relations:
logger.info('No Embedded relation found')
else:
logger.info('Found Embedded relations-%s', str(list(embed_relations)))
if RelationType.NO_RELATION_FOUND in wn_relaions:
logger.info('No Wordnet relation found')
else:
logger.info('Found Wordnet relations-%s', str(list(wn_relaions)))