README.md

metadata

language:
  - fr
multilinguality:
  - monolingual
task_categories:
  - token-classification

m0_qualitative_analysis_ref_cmbert_io

Introduction

This dataset was used to perform qualitative analysis of Jean-Baptiste/camembert-ner on flat NER task using Flat NER approach [M0]. It contains 19th-century Paris trade directories' entries.

Dataset parameters

• Approach : M0
• Dataset type : ground-truth
• Tokenizer : Jean-Baptiste/camembert-ner
• Tagging format : IO
• Counts :
  • Train : 6084
  • Dev : 676
  • Test : 1685
• Associated fine-tuned model : nlpso/m0_flat_ner_ref_cmbert_io

Entity types

Abbreviation	Description
O	Outside of a named entity
PER	Person or company name
ACT	Person or company professional activity
TITRE	Distinction
LOC	Street name
CARDINAL	Street number
FT	Geographical feature

How to use this dataset

from datasets import load_dataset

train_dev_test = load_dataset("nlpso/m0_qualitative_analysis_ref_cmbert_io")