Rooba
/
spaCy
mirror of https://github.com/explosion/spaCy.git
1
1
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Don't automatically write new entities into the Doc in the Matcher. This fixes a long-standing wart, but introduces a *backwards incompatibility.*

This commit is contained in:
Matthew Honnibal 2016-09-24 01:16:45 +02:00
parent e48df859b5
commit 55f1f7edaf
1 changed files with 0 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
spacy/matcher.pyx
View File

@ -294,17 +294,6 @@ cdef class Matcher:
label = pattern[1].attrs[1].value label = pattern[1].attrs[1].value
if acceptor is None or acceptor(doc, ent_id, label, start, end): if acceptor is None or acceptor(doc, ent_id, label, start, end):
matches.append((ent_id, label, start, end)) matches.append((ent_id, label, start, end))
seen = set()
filtered = []
for ent_id, label, start, end in sorted(matches,
key=lambda m: (m[2],-(m[2]-m[3]))):
if all(i in seen for i in range(start, end)):
continue
else:
for i in range(start, end):
seen.add(i)
filtered.append((label, start, end))
doc.ents = [(e.label, e.start, e.end) for e in doc.ents] + filtered
return matches return matches
def pipe(self, docs, batch_size=1000, n_threads=2): def pipe(self, docs, batch_size=1000, n_threads=2):