metadata

license: mit
base_model: classla/xlm-roberta-base-multilingual-text-genre-classifier
tags:
  - Italian
  - legal ruling
  - generated_from_trainer
metrics:
  - f1
  - accuracy
model-index:
  - name: ribesstefano/RuleBert-v0.1-k3
    results: []

ribesstefano/RuleBert-v0.1-k3

This model is a fine-tuned version of classla/xlm-roberta-base-multilingual-text-genre-classifier on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.3285
F1: 0.4638
Roc Auc: 0.6576
Accuracy: 0.0714

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: 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
0.3423	0.13	250	0.3539	0.4497	0.6562	0.0670
0.3231	0.27	500	0.3425	0.4596	0.6594	0.0670
0.3248	0.4	750	0.3364	0.4495	0.6541	0.0714
0.3283	0.54	1000	0.3351	0.4529	0.6555	0.0714
0.3237	0.67	1250	0.3315	0.4600	0.6581	0.0625
0.325	0.81	1500	0.3313	0.4681	0.6624	0.0312
0.3316	0.94	1750	0.3290	0.4595	0.6564	0.0714
0.3239	1.08	2000	0.3310	0.4592	0.6572	0.0625
0.3085	1.21	2250	0.3280	0.4614	0.6567	0.0670
0.3161	1.35	2500	0.3303	0.4623	0.6574	0.0670
0.314	1.48	2750	0.3289	0.4613	0.6566	0.0714
0.3187	1.62	3000	0.3293	0.4594	0.6554	0.0714
0.3145	1.75	3250	0.3295	0.4629	0.6569	0.0714
0.3128	1.89	3500	0.3285	0.4629	0.6569	0.0714
0.3135	2.02	3750	0.3285	0.4615	0.6566	0.0714
0.3171	2.16	4000	0.3285	0.4638	0.6576	0.0714

Framework versions

Transformers 4.36.2
Pytorch 2.1.0+cu121
Datasets 2.16.1
Tokenizers 0.15.0