Using custom tag_map in prodigy?

,
1

I've trained a model for Luxembourgish using a tag_map oriented on the German Tiger tagset:

TAG_MAP = {
    "$(": {POS: PUNCT, "PunctType": "brck"},
    "$,": {POS: PUNCT, "PunctType": "comm"},
    "$.": {POS: PUNCT, "PunctType": "peri"},
    "ADJA": {POS: ADJ},
    "ADJD": {POS: ADJ},
    "ADV": {POS: ADV},
    "APPO": {POS: ADP, "AdpType": "post"},
    "APPR": {POS: ADP, "AdpType": "prep"},
    "APPRART": {POS: ADP, "AdpType": "prep", "PronType": "art"},
    "APZR": {POS: ADP, "AdpType": "circ"},
    "ART": {POS: DET, "PronType": "art"},
    "CARD": {POS: NUM, "NumType": "card"},
...
}

Now, when I use prodigy to further annotate and correct the POS, I am only offered the Universal commons tag set:

python3 -m prodigy pos.correct mypos_new model-december2020/model-best/  rtl_radio.jsonl
ℹ Using universal coarse-grained POS tags: ADJ, ADP, ADV, AUX, CONJ,
CCONJ, DET, INTJ, NOUN, NUM, PART, PRON, PROPN, PUNCT, SCONJ, SYM, VERB, X,
SPACE

Is this the intended behaviour or is it possible to access also the custom POS tags in prodigy?
Thanks!

2

The pos recipes aren't currently set up to make updating an existing fine-grained tagger as straightforward as it could be. Here's some more info: https://github.com/explosion/spaCy/issues/6283#issuecomment-718788016

3

Thanks. I could solve my problem using the --fine-grained option.