|
--- |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: greek_legal_bert_v2-finetuned-ner |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# greek_legal_bert_v2-finetuned-ner |
|
|
|
This model is a fine-tuned version of [alexaapo/greek_legal_bert_v2](https://huggingface.co/alexaapo/greek_legal_bert_v2) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0900 |
|
- Precision: 0.8424 |
|
- Recall: 0.8638 |
|
- F1: 0.8530 |
|
- Accuracy: 0.9775 |
|
|
|
## Model description |
|
|
|
More information needed |
|
|
|
## Intended uses & limitations |
|
|
|
More information needed |
|
|
|
## Training and evaluation data |
|
|
|
More information needed |
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 2e-05 |
|
- train_batch_size: 16 |
|
- eval_batch_size: 16 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 10 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| No log | 0.64 | 250 | 0.0839 | 0.7859 | 0.8539 | 0.8185 | 0.9737 | |
|
| 0.1127 | 1.29 | 500 | 0.0783 | 0.8092 | 0.8569 | 0.8324 | 0.9759 | |
|
| 0.1127 | 1.93 | 750 | 0.0743 | 0.8284 | 0.8446 | 0.8364 | 0.9766 | |
|
| 0.0538 | 2.58 | 1000 | 0.0816 | 0.8243 | 0.8597 | 0.8416 | 0.9774 | |
|
| 0.0538 | 3.22 | 1250 | 0.0900 | 0.8424 | 0.8638 | 0.8530 | 0.9776 | |
|
| 0.0346 | 3.87 | 1500 | 0.0890 | 0.8401 | 0.8597 | 0.8498 | 0.9770 | |
|
| 0.0346 | 4.51 | 1750 | 0.0964 | 0.8342 | 0.8576 | 0.8457 | 0.9768 | |
|
| 0.0233 | 5.15 | 2000 | 0.1094 | 0.8336 | 0.8645 | 0.8488 | 0.9768 | |
|
| 0.0233 | 5.8 | 2250 | 0.1110 | 0.8456 | 0.8549 | 0.8502 | 0.9777 | |
|
| 0.0161 | 6.44 | 2500 | 0.1224 | 0.8408 | 0.8535 | 0.8471 | 0.9769 | |
|
| 0.0161 | 7.09 | 2750 | 0.1281 | 0.8347 | 0.8624 | 0.8483 | 0.9770 | |
|
| 0.0114 | 7.73 | 3000 | 0.1268 | 0.8397 | 0.8573 | 0.8484 | 0.9773 | |
|
| 0.0114 | 8.38 | 3250 | 0.1308 | 0.8388 | 0.8549 | 0.8468 | 0.9771 | |
|
| 0.0088 | 9.02 | 3500 | 0.1301 | 0.8412 | 0.8559 | 0.8485 | 0.9772 | |
|
| 0.0088 | 9.66 | 3750 | 0.1368 | 0.8396 | 0.8604 | 0.8499 | 0.9772 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.23.1 |
|
- Pytorch 1.12.1+cu113 |
|
- Datasets 2.6.1 |
|
- Tokenizers 0.13.1 |
|
|
