README.md

metadata

tags:
  - spacy
  - token-classification
language:
  - multilingual
model-index:
  - name: xx_fro_sigtyp_trf
    results:
      - task:
          name: TAG
          type: token-classification
        metrics:
          - name: TAG (XPOS) Accuracy
            type: accuracy
            value: 0.8626325088
      - task:
          name: LEMMA
          type: token-classification
        metrics:
          - name: Lemma Accuracy
            type: accuracy
            value: 0.8348421285
      - task:
          name: UNLABELED_DEPENDENCIES
          type: token-classification
        metrics:
          - name: Unlabeled Attachment Score (UAS)
            type: f_score
            value: 0.6967152383
      - task:
          name: LABELED_DEPENDENCIES
          type: token-classification
        metrics:
          - name: Labeled Attachment Score (LAS)
            type: f_score
            value: 0.6162015312
      - task:
          name: SENTS
          type: token-classification
        metrics:
          - name: Sentences F-Score
            type: f_score
            value: 0.8357487923

Feature	Description
Name	xx_fro_sigtyp_trf
Version	0.1.0
spaCy	>=3.6.1,<3.7.0
Default Pipeline	transformer, parser, trainable_lemmatizer, tagger
Components	transformer, parser, trainable_lemmatizer, tagger
Vectors	0 keys, 0 unique vectors (0 dimensions)
Sources	n/a
License	n/a
Author	n/a

Label Scheme

View label scheme (124 labels for 2 components)

Component	Labels
parser	ROOT, acl, acl:relcl, advcl, advmod, amod, appos, aux, aux:pass, case, case:det, cc, cc:nc, ccomp, conj, cop, csubj, dep, det, expl, flat, iobj, mark, nmod, nsubj, nummod, obj, obl, obl:advmod, parataxis, punct, vocative, xcomp
tagger	ADJcar__NumType=Card, ADJind__PronType=Ind, ADJord, ADJpos__Poss=Yes, ADJqua, ADJqua__Tense=Past|VerbForm=Part, ADJqua__Tense=Pres|VerbForm=Part, ADVgen, ADVgen.PROper, ADVgen__PronType=Ind, ADVgen__PronType=Prs,Rel, ADVgen__PronType=Rel, ADVint__PronType=Int, ADVneg, ADVneg.PROper__Polarity=Neg|PronType=Prs, ADVneg__Polarity=Neg, ADVsub, CONcoo, CONcoo__PronType=Prs,Rel, CONsub, CONsub.PROper, CONsub__PronType=Rel, DETcar__NumType=Card, DETdef__Definite=Def, DETdef__Definite=Def|PronType=Art, DETdef__PronType=Ind, DETdef__PronType=Prs, DETdem__PronType=Dem, DETdem__PronType=Prs, DETind__PronType=Ind, DETint__PronType=Int, DETndf__Definite=Ind, DETndf__Definite=Ind|PronType=Art, DETord, DETord__NumType=Ord, DETpos__Poss=Yes, DETrel__PronType=Rel, INJ, NOMcom, NOMcom__Morph=VFin, NOMcom__VerbForm=Inf, NOMpro, PONfbl, PONfrt, PONpdr, PONpga, PONpxx, PRE, PRE.DETdef__Definite=Def|PronType=Art, PRE.PROper, PRE.PROper__Definite=Def|PronType=Art, PRE__Morph=VFin, PRE__PronType=Dem, PREdetdef__PronType=Prs,Rel, PROadv, PROadv__PronType=Dem, PROcar, PROcar__NumType=Card, PROdem__PronType=Dem, PROdem__PronType=Prs,Rel, PROimp, PROimp__PronType=Prs, PROind, PROind__PronType=Ind, PROind__PronType=Rel, PROint__PronType=Int, PROord__NumType=Ord, PROper, PROper.PROper__PronType=Prs, PROper__Poss=Yes, PROper__PronType=Prs, PROpos__Poss=Yes, PROpos__Poss=Yes|PronType=Prs, PROrel, PROrel__PronType=Prs,Rel, PROrel__PronType=Rel, RED, VERcjg, VERcjg__VerbForm=Fin, VERcjg__VerbForm=Inf, VERinf__VerbForm=Inf, VERppa__Tense=Pres|VerbForm=Part, VERppe, VERppe__Tense=Past, VERppe__Tense=Past|VerbForm=Part, devenir__Tense=Past|VerbForm=Part, devenir__VerbForm=Fin, laisser__VerbForm=Fin, remanoir__VerbForm=Fin, ressembler__VerbForm=Fin, sembler__VerbForm=Fin

Accuracy

Type	Score
DEP_UAS	69.67
DEP_LAS	61.62
SENTS_P	78.82
SENTS_R	88.95
SENTS_F	83.57
LEMMA_ACC	83.48
TAG_ACC	86.26
TRANSFORMER_LOSS	44338.10
PARSER_LOSS	6886.30
TRAINABLE_LEMMATIZER_LOSS	1446.31
TAGGER_LOSS	3500.05