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

Fix initialization of token mappings in new align (#4640) 

Initialize all values in `a2b` and `b2a` since `numpy.empty()` otherwise
result unspecified integers.
This commit is contained in:
adrianeboyd 2019-11-13 21:22:18 +01:00 committed by Matthew Honnibal
parent 3ac4e8eb7a
commit d67b0f196a
1 changed files with 2 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
spacy/gold.pyx
View File

@ -129,6 +129,8 @@ def align(tokens_a, tokens_b):
cost = 0 cost = 0
a2b = numpy.empty(len(tokens_a), dtype="i") a2b = numpy.empty(len(tokens_a), dtype="i")
b2a = numpy.empty(len(tokens_b), dtype="i") b2a = numpy.empty(len(tokens_b), dtype="i")
a2b.fill(-1)
b2a.fill(-1)
a2b_multi = {} a2b_multi = {}
b2a_multi = {} b2a_multi = {}
i = 0 i = 0
@ -138,7 +140,6 @@ def align(tokens_a, tokens_b):
while i < len(tokens_a) and j < len(tokens_b): while i < len(tokens_a) and j < len(tokens_b):
a = tokens_a[i][offset_a:] a = tokens_a[i][offset_a:]
b = tokens_b[j][offset_b:] b = tokens_b[j][offset_b:]
a2b[i] = b2a[j] = -1
if a == b: if a == b:
if offset_a == offset_b == 0: if offset_a == offset_b == 0:
a2b[i] = j a2b[i] = j