|
--- |
|
language: en |
|
license: mit |
|
tags: |
|
- flair |
|
- token-classification |
|
- sequence-tagger-model |
|
base_model: dbmdz/bert-tiny-historic-multilingual-cased |
|
widget: |
|
- text: Cp . Eur . Phoen . 240 , 1 , αἷμα ddiov φλέγέι . |
|
--- |
|
|
|
# Fine-tuned Flair Model on AjMC English NER Dataset (HIPE-2022) |
|
|
|
This Flair model was fine-tuned on the |
|
[AjMC English](https://github.com/hipe-eval/HIPE-2022-data/blob/main/documentation/README-ajmc.md) |
|
NER Dataset using hmBERT Tiny as backbone LM. |
|
|
|
The AjMC dataset consists of NE-annotated historical commentaries in the field of Classics, |
|
and was created in the context of the [Ajax MultiCommentary](https://mromanello.github.io/ajax-multi-commentary/) |
|
project. |
|
|
|
The following NEs were annotated: `pers`, `work`, `loc`, `object`, `date` and `scope`. |
|
|
|
# Results |
|
|
|
We performed a hyper-parameter search over the following parameters with 5 different seeds per configuration: |
|
|
|
* Batch Sizes: `[4, 8]` |
|
* Learning Rates: `[5e-05, 3e-05]` |
|
|
|
And report micro F1-score on development set: |
|
|
|
| Configuration | Seed 1 | Seed 2 | Seed 3 | Seed 4 | Seed 5 | Average | |
|
|-------------------|--------------|--------------|--------------|--------------|------------------|-----------------| |
|
| `bs4-e10-lr5e-05` | [0.5579][1] | [0.5082][2] | [0.5434][3] | [0.4949][4] | [0.4882][5] | 0.5185 ± 0.0306 | |
|
| `bs8-e10-lr5e-05` | [0.5301][6] | [0.468][7] | [0.525][8] | [0.4989][9] | [**0.4707**][10] | 0.4985 ± 0.0292 | |
|
| `bs4-e10-lr3e-05` | [0.4972][11] | [0.427][12] | [0.4745][13] | [0.4394][14] | [0.4249][15] | 0.4526 ± 0.0319 | |
|
| `bs8-e10-lr3e-05` | [0.4402][16] | [0.3531][17] | [0.4141][18] | [0.3808][19] | [0.4073][20] | 0.3991 ± 0.0333 | |
|
|
|
[1]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1 |
|
[2]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2 |
|
[3]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3 |
|
[4]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4 |
|
[5]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5 |
|
[6]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1 |
|
[7]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2 |
|
[8]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3 |
|
[9]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4 |
|
[10]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5 |
|
[11]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1 |
|
[12]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2 |
|
[13]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3 |
|
[14]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4 |
|
[15]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5 |
|
[16]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1 |
|
[17]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2 |
|
[18]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3 |
|
[19]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4 |
|
[20]: https://hf.co/stefan-it/hmbench-ajmc-en-hmbert_tiny-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5 |
|
|
|
The [training log](training.log) and TensorBoard logs (not available for hmBERT Base model) are also uploaded to the model hub. |
|
|
|
More information about fine-tuning can be found [here](https://github.com/stefan-it/hmBench). |
|
|
|
# Acknowledgements |
|
|
|
We thank [Luisa März](https://github.com/LuisaMaerz), [Katharina Schmid](https://github.com/schmika) and |
|
[Erion Çano](https://github.com/erionc) for their fruitful discussions about Historic Language Models. |
|
|
|
Research supported with Cloud TPUs from Google's [TPU Research Cloud](https://sites.research.google/trc/about/) (TRC). |
|
Many Thanks for providing access to the TPUs ❤️ |
|
|
