|
--- |
|
license: mit |
|
base_model: papluca/xlm-roberta-base-language-detection |
|
tags: |
|
- Italian |
|
- legal ruling |
|
- generated_from_trainer |
|
metrics: |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: ribesstefano/RuleBert-v0.4-k4 |
|
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. --> |
|
|
|
# ribesstefano/RuleBert-v0.4-k4 |
|
|
|
This model is a fine-tuned version of [papluca/xlm-roberta-base-language-detection](https://huggingface.co/papluca/xlm-roberta-base-language-detection) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3517 |
|
- F1: 0.5190 |
|
- Roc Auc: 0.6864 |
|
- Accuracy: 0.0 |
|
|
|
## 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: 0.0005 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 64 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- training_steps: 4000 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|:--------:| |
|
| 0.3447 | 0.12 | 250 | 0.3402 | 0.4810 | 0.6688 | 0.0 | |
|
| 0.3501 | 0.24 | 500 | 0.3548 | 0.4884 | 0.6786 | 0.0 | |
|
| 0.3433 | 0.36 | 750 | 0.3596 | 0.4946 | 0.6885 | 0.0 | |
|
| 0.3521 | 0.48 | 1000 | 0.3762 | 0.4861 | 0.6648 | 0.0 | |
|
| 0.3466 | 0.6 | 1250 | 0.3496 | 0.4861 | 0.6648 | 0.0 | |
|
| 0.3285 | 0.72 | 1500 | 0.3519 | 0.4861 | 0.6648 | 0.0 | |
|
| 0.333 | 0.84 | 1750 | 0.3550 | 0.4861 | 0.6648 | 0.0 | |
|
| 0.3268 | 0.96 | 2000 | 0.3436 | 0.5190 | 0.6864 | 0.0 | |
|
| 0.3376 | 1.08 | 2250 | 0.3637 | 0.4978 | 0.6891 | 0.0 | |
|
| 0.3319 | 1.19 | 2500 | 0.3459 | 0.5190 | 0.6864 | 0.0 | |
|
| 0.3169 | 1.31 | 2750 | 0.3430 | 0.4810 | 0.6688 | 0.0 | |
|
| 0.3293 | 1.43 | 3000 | 0.3480 | 0.4861 | 0.6648 | 0.0 | |
|
| 0.3293 | 1.55 | 3250 | 0.3517 | 0.5190 | 0.6864 | 0.0 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.2 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.16.1 |
|
- Tokenizers 0.15.0 |
|
|
