Can a morphologizer (Tagger architecture) use static vectors as features?

[thjbdvlt]

Hello :-)

I am currently training a morphologizer (for french). It works well but i was wondering if it could use word vectors, because i've produced some word2vec vectors with gensim.

I've read in the documentation that "Many neural network models are able to use word vector tables as additional features, which sometimes results in significant improvements in accuracy" (here) but it only shows an config example for spacy.MultiHashEmbed.v2 architecture and implies using the parameter include_static_vectors.

But the Tagger architecture (used for pos/morphologizer) doesn't seems to accept include_static_vectors parameter (if i've correctly understood). So, is there any way to make my morphologizer (Tagger model) use my vectors? (without training a tok2vec component, if it's possible.)

Thanks for reading :)