Datasetpublic

nlpso/m1_qualitative_analysis_ocr_cmbert_iob2

m1_qualitative_analysis_ocr_cmbert_iob2 Introduction This dataset was used to perform qualitative analysis of Jean-Baptiste/camembert-ner on nested NER task using Independant NER layers approach [M1]. It contains Paris trade directories entries from the 19th century. Dataset parameters Approach : M1 Dataset type : noisy (Pero OCR) Tokenizer : Jean-Baptiste/camembert-ner Tagging format : IOB2 Counts : Train : 6084 Dev : 676 Test : 1685… See the full description on the dataset page: https://huggingface.co/datasets/nlpso/m1_qualitative_analysis_ocr_cmbert_iob2.

sourceHugging Faceupdated 4y agoView on Hugging Face

0likes29downloads

Dataset Card

m1qualitativeanalysisocrcmbert_iob2

Introduction

This dataset was used to perform qualitative analysis of Jean-Baptiste/camembert-ner on nested NER task using Independant NER layers approach [M1]. It contains Paris trade directories entries from the 19th century.

Dataset parameters

—Approach : M1
—Dataset type : noisy (Pero OCR)
—Tokenizer : Jean-Baptiste/camembert-ner
—Tagging format : IOB2
—Counts :
—Train : 6084
—Dev : 676
—Test : 1685
—Associated fine-tuned models :
—Level-1 : nlpso/m1_ind_layers_ocr_cmbert_iob2_level_1
—Level 2 : nlpso/m1_ind_layers_ocr_cmbert_iob2_level_2

Entity types

Abbreviation	Entity group (level)	Description
O	1 & 2	Outside of a named entity
PER	1	Person or company name
ACT	1 & 2	Person or company professional activity
TITREH	2	Military or civil distinction
DESC	1	Entry full description
TITREP	2	Professionnal reward
SPAT	1	Address
LOC	2	Street name
CARDINAL	2	Street number
FT	2	Geographical feature

How to use this dataset

python

from datasets import load_dataset

train_dev_test = load_dataset("nlpso/m1_qualitative_analysis_ocr_cmbert_iob2")