stefan-it's picture
readme: add initial version of model card (#2)
120da92 verified
|
raw
history blame
4 kB
---
language: de
license: mit
tags:
- flair
- token-classification
- sequence-tagger-model
- hetzner
- hetzner-gex44
- hetzner-gpu
base_model: deepset/gbert-base
widget:
- text: Wesentliche Tätigkeiten der Compliance-Funktion wurden an die Mercurtainment
AG , Düsseldorf , ausgelagert .
---
# Fine-tuned Flair Model on CO-Fun NER Dataset
This Flair model was fine-tuned on the
[CO-Fun](https://arxiv.org/abs/2403.15322) NER Dataset using GBERT Base as backbone LM.
## Dataset
The [Company Outsourcing in Fund Prospectuses (CO-Fun) dataset](https://arxiv.org/abs/2403.15322) consists of
948 sentences with 5,969 named entity annotations, including 2,340 Outsourced Services, 2,024 Companies, 1,594 Locations
and 11 Software annotations.
Overall, the following named entities are annotated:
* `Auslagerung` (engl. outsourcing)
* `Unternehmen` (engl. company)
* `Ort` (engl. location)
* `Software`
## Fine-Tuning
The latest [Flair version](https://github.com/flairNLP/flair/tree/42ea3f6854eba04387c38045f160c18bdaac07dc) is used for
fine-tuning.
A hyper-parameter search over the following parameters with 5 different seeds per configuration is performed:
* Batch Sizes: [`16`, `8`]
* Learning Rates: [`3e-05`, `5e-05`]
More details can be found in this [repository](https://github.com/stefan-it/co-funer). All models are fine-tuned on a
[Hetzner GX44](https://www.hetzner.com/dedicated-rootserver/matrix-gpu/) with an NVIDIA RTX 4000.
## Results
A hyper-parameter search with 5 different seeds per configuration is performed and micro F1-score on development set
is reported:
| Configuration | Seed 1 | Seed 2 | Seed 3 | Seed 4 | Seed 5 | Average |
|--------------------|-----------------|--------------|--------------|--------------|--------------|-----------------|
| `bs8-e10-lr5e-05` | [0.9477][1] | [0.935][2] | [0.9517][3] | [0.9443][4] | [0.9342][5] | 0.9426 ± 0.0077 |
| `bs16-e10-lr5e-05` | [0.9214][6] | [0.9364][7] | [0.9334][8] | [0.9489][9] | [0.9257][10] | 0.9332 ± 0.0106 |
| `bs8-e10-lr3e-05` | [**0.928**][11] | [0.9248][12] | [0.9421][13] | [0.9295][14] | [0.9263][15] | 0.9301 ± 0.0069 |
| `bs16-e10-lr3e-05` | [0.918][16] | [0.9256][17] | [0.9331][18] | [0.9273][19] | [0.9196][20] | 0.9247 ± 0.0061 |
[1]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs8-e10-lr5e-05-1
[2]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs8-e10-lr5e-05-2
[3]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs8-e10-lr5e-05-3
[4]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs8-e10-lr5e-05-4
[5]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs8-e10-lr5e-05-5
[6]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs16-e10-lr5e-05-1
[7]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs16-e10-lr5e-05-2
[8]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs16-e10-lr5e-05-3
[9]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs16-e10-lr5e-05-4
[10]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs16-e10-lr5e-05-5
[11]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs8-e10-lr3e-05-1
[12]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs8-e10-lr3e-05-2
[13]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs8-e10-lr3e-05-3
[14]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs8-e10-lr3e-05-4
[15]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs8-e10-lr3e-05-5
[16]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs16-e10-lr3e-05-1
[17]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs16-e10-lr3e-05-2
[18]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs16-e10-lr3e-05-3
[19]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs16-e10-lr3e-05-4
[20]: https://hf.co/stefan-it/flair-co-funer-gbert_base-bs16-e10-lr3e-05-5
The result in bold shows the performance of the current viewed model.
Additionally, the Flair [training log](training.log) and [TensorBoard logs](../../tensorboard) are also uploaded to the model
hub.