Dutch tag map not updated with new models in spaCy v2.2

[BLKSerene]

Hi, the tag map for the Dutch language hasn't been updated with new models in spaCy v2.2, so all universal POS tags become X.

>>> import spacy
>>> nlp = spacy.load('nl_core_news_sm')
>>> doc = nlp('Het Nederlands is een West-Germaanse taal en de moedertaal van de meeste inwoners van Nederland, België en Suriname.')

>>> for t in doc:
	print(t.tag_, t.pos_)

LID|bep|stan|evon X
N|eigen|ev|basis|onz|stan X
WW|pv|tgw|ev X
LID|onbep|stan|agr X
ADJ|prenom|basis|met-e|stan X
N|soort|ev|basis|zijd|stan X
VG|neven X
LID|bep|stan|rest X
N|soort|ev|basis|zijd|stan X
VZ|init X
LID|bep|stan|rest X
VNW|onbep|grad|stan|prenom|met-e|agr|sup X
N|soort|mv|basis X
VZ|init X
N|eigen|ev|basis|onz|stan X
LET X
N|eigen|ev|basis|onz|stan X
VG|neven X
N|eigen|ev|basis|onz|stan X
LET X

• Operating System: Windows 10 x64
• Python Version Used: 3.7.4 x64
• spaCy Version Used: 2.2.0

[honnibal]

Thanks! Just pushed an updated model for Dutch. If you do spacy download nl_core_news_sm it should get you v2.2.1 of the model, which should resolve the problem.

[lock]

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.