From ab1cff2fc7465965221a6a445b2a19e7e1c6bf25 Mon Sep 17 00:00:00 2001 From: costero-e Date: Thu, 16 Feb 2023 14:49:02 +0100 Subject: [PATCH] semantic --- beacon/db/filters.py | 4 ++-- beacon/semantic_similarity.py | 3 ++- deploy/extract_filtering_terms_oriol.py | 11 ++++++----- 3 files changed, 10 insertions(+), 8 deletions(-) diff --git a/beacon/db/filters.py b/beacon/db/filters.py index 4424e47..096ca93 100644 --- a/beacon/db/filters.py +++ b/beacon/db/filters.py @@ -6,7 +6,7 @@ from beacon.request import ontologies from beacon.request.model import AlphanumericFilter, CustomFilter, OntologyFilter, Operator, Similarity -from beacon.semantic_similarity import semantic_similarity +#from beacon.semantic_similarity import semantic_similarity import logging @@ -91,7 +91,7 @@ def apply_ontology_filter(query: dict, filter: OntologyFilter) -> dict: cutoff = 0.7 elif filter.similarity == Similarity.LOW: cutoff = 0.5 - similar_terms = semantic_similarity(filter.id, cutoff) + similar_terms = 'semantic_similarity(filter.id, cutoff)' LOG.debug("Similar: {}".format(similar_terms)) for term in similar_terms: if query["$text"]["$search"]: diff --git a/beacon/semantic_similarity.py b/beacon/semantic_similarity.py index bb3653c..4ee891b 100644 --- a/beacon/semantic_similarity.py +++ b/beacon/semantic_similarity.py @@ -1,3 +1,4 @@ +''' #from sentence_transformers import SentenceTransformer from scipy.spatial import distance import networkx as nx @@ -192,6 +193,6 @@ def semantic_similarity(term:str, x:float): print(list_neighbours) - +''' diff --git a/deploy/extract_filtering_terms_oriol.py b/deploy/extract_filtering_terms_oriol.py index bb7c81b..ac6b1b9 100644 --- a/deploy/extract_filtering_terms_oriol.py +++ b/deploy/extract_filtering_terms_oriol.py @@ -182,19 +182,20 @@ def insert_all_alphanumeric_terms_used(): collections = client.beacon.list_collection_names() if 'filtering_terms' in collections: collections.remove('filtering_terms') + collections = ['runs'] print("Collections:", collections) for c_name in collections: terms = find_alphanumeric_terms_used(c_name) print(terms) - #if len(terms) > 0: - #client.beacon.filtering_terms.insert_many(terms) + if len(terms) > 0: + client.beacon.filtering_terms.insert_many(terms) -#insert_all_ontology_terms_used() +insert_all_ontology_terms_used() #insert_all_alphanumeric_terms_used() #terms=find_ontology_terms_used("individuals") #print(terms) #hola = get_ontology_term_label('NCIT','C173381') #print(hola) -find_alphanumeric_terms_used('analyses') - +#hola = find_alphanumeric_terms_used('analyses') +#print(hola)