Fine-tuned Flair Model on French ICDAR-Europeana NER Dataset

This Flair model was fine-tuned on the French ICDAR-Europeana NER Dataset using hmBERT as backbone LM.

The ICDAR-Europeana NER Dataset is a preprocessed variant of the Europeana NER Corpora for Dutch and French.

The following NEs were annotated: PER, LOC and ORG.

Results

We performed a hyper-parameter search over the following parameters with 5 different seeds per configuration:

And report micro F1-score on development set:

Configuration	Run 1	Run 2	Run 3	Run 4	Run 5	Avg.
bs4-e10-lr3e-05	0.7731	0.7696	0.7666	0.7823	0.7714	77.26 ± 0.53
bs4-e10-lr5e-05	0.774	0.7571	0.7685	0.7694	0.7704	76.79 ± 0.57
bs8-e10-lr5e-05	0.7675	0.7698	0.7601	0.7657	0.7641	76.54 ± 0.33
bs8-e10-lr3e-05	0.7596	0.7697	0.7711	0.7628	0.7574	76.41 ± 0.54

The training log and TensorBoard logs (only for hmByT5 and hmTEAMS based models) are also uploaded to the model hub.

More information about fine-tuning can be found here.

We thank Luisa März, Katharina Schmid and Erion Çano for their fruitful discussions about Historic Language Models.

Research supported with Cloud TPUs from Google's TPU Research Cloud (TRC). Many Thanks for providing access to the TPUs ❤️