mirror of
https://github.com/explosion/spaCy.git
synced 2025-07-04 20:03:13 +03:00
* Adjust NER rules, so that U entries in gazetteer don't become B moves to the model
This commit is contained in:
parent
6e68b344c1
commit
5623242b3e
|
@ -173,6 +173,10 @@ cdef class Begin:
|
||||||
# If the next word is B or O, we can't B now
|
# If the next word is B or O, we can't B now
|
||||||
elif st.B_(1).ent_iob == 2 or st.B_(1).ent_iob == 3:
|
elif st.B_(1).ent_iob == 2 or st.B_(1).ent_iob == 3:
|
||||||
return False
|
return False
|
||||||
|
# If the current word is B, and the next word isn't I, the current word
|
||||||
|
# is really U
|
||||||
|
elif preset_ent_iob == 3 and st.B_(1).ent_iob != 1:
|
||||||
|
return False
|
||||||
# Don't allow entities to extend across sentence boundaries
|
# Don't allow entities to extend across sentence boundaries
|
||||||
elif st.B_(1).sent_start:
|
elif st.B_(1).sent_start:
|
||||||
return False
|
return False
|
||||||
|
|
Loading…
Reference in New Issue
Block a user