You can deploy this model and start earning money today!

stefan-it/
hmbench-icdar-nl-hmbert-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3

Name: hmbench-icdar-nl-hmbert-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3
Availability: InStock

Views

No views yet

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

This Flair model was fine-tuned on the Dutch 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.

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.
bs8-e10-lr5e-05	0.8191	0.8086	0.8237	0.8318	0.8235	82.13 ± 0.76
bs8-e10-lr3e-05	0.8056	0.8183	0.8241	0.8431	0.8155	82.13 ± 1.24
bs4-e10-lr5e-05	0.8055	0.822	0.8243	0.8093	0.8144	81.51 ± 0.72
bs4-e10-lr3e-05	0.8039	0.8122	0.8073	0.8246	0.8132	81.22 ± 0.7

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 ❤️