get_doc not storing POS properly

[svlandeg]

How to reproduce the behaviour

    words = ["This", "is", "a", "sentence"]
    pos_s = ["DET", "VERB", "DET", "NOUN"]
    spaces = [" ", " ", " ", ""]
    deps_s = ["dep", "adj", "nn", "atm"]
    tags_s = ["DT", "VBZ", "DT", "NN"]

    nlp = English()
    strings = nlp.vocab.strings

    for w in words:
        strings.add(w)
    deps = [strings.add(d) for d in deps_s]
    pos = [strings.add(p) for p in pos_s]
    tags = [strings.add(t) for t in tags_s]

    attrs = [POS, DEP, TAG]
    array = numpy.array(list(zip(pos, deps, tags)), dtype="uint64")

    doc = Doc(nlp.vocab, words=words, spaces=spaces)
    doc.from_array(attrs, array)
    print("1", [(token.text, token.pos_, token.tag_) for token in doc])

    doc2 = get_doc(nlp.vocab, words=words, pos=pos_s, deps=deps_s, tags=tags_s)
    print("2", [(token.text, token.pos_, token.tag_) for token in doc2])

output:

1 [('This', 'DET', 'DT'), ('is', 'VERB', 'VBZ'), ('a', 'DET', 'DT'), ('sentence', 'NOUN', 'NN')]
2 [('This', 'DET', 'DT'), ('is', 'AUX', 'VBZ'), ('a', 'DET', 'DT'), ('sentence', 'NOUN', 'NN')]

For the second token, the POS should be VERB, not AUX.

Will be submitting a PR soon to fix this!

[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.