ai-forever / ner-bert

BERT-NER (nert-bert) with google bert https://github.com/google-research.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Mistake in main_metrics

brickee opened this issue · comments

current_tag = current_token.split('', 1)[-1] # replace '-' with ""
This bug will totoally change the evaluation result.

Maby you are right. You can do pull request with changes or try to use from modules.analyze_utils.plot_metrics import get_bert_span_report for mesure results.
Code from modules/analyze_utils/main_metrics.py was used for evaluation of FactRuEval dataset.