ribesstefano/RuleBert-v0.1-k2
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: 0.3049
- F1: 0.5103
- Roc Auc: 0.6747
- 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: 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 |
---|---|---|---|---|---|---|
0.3592 | 0.14 | 250 | 0.3131 | 0.5179 | 0.6796 | 0.0 |
0.3369 | 0.27 | 500 | 0.3063 | 0.5109 | 0.6758 | 0.0 |
0.3352 | 0.41 | 750 | 0.3087 | 0.5110 | 0.6750 | 0.0 |
0.3283 | 0.54 | 1000 | 0.3042 | 0.5105 | 0.6749 | 0.0 |
0.3246 | 0.68 | 1250 | 0.3068 | 0.5101 | 0.6747 | 0.0 |
0.3264 | 0.82 | 1500 | 0.3028 | 0.5152 | 0.6771 | 0.0 |
0.3365 | 0.95 | 1750 | 0.3051 | 0.5103 | 0.6747 | 0.0 |
0.3269 | 1.09 | 2000 | 0.3042 | 0.5103 | 0.6747 | 0.0 |
0.3173 | 1.22 | 2250 | 0.3059 | 0.5103 | 0.6747 | 0.0 |
0.3127 | 1.36 | 2500 | 0.3053 | 0.5110 | 0.6750 | 0.0 |
0.3211 | 1.49 | 2750 | 0.3067 | 0.5103 | 0.6747 | 0.0 |
0.3155 | 1.63 | 3000 | 0.3067 | 0.5103 | 0.6747 | 0.0 |
0.319 | 1.77 | 3250 | 0.3051 | 0.5103 | 0.6747 | 0.0 |
0.3286 | 1.9 | 3500 | 0.3042 | 0.5103 | 0.6747 | 0.0 |
0.3243 | 2.04 | 3750 | 0.3051 | 0.5103 | 0.6747 | 0.0 |
0.3111 | 2.17 | 4000 | 0.3049 | 0.5103 | 0.6747 | 0.0 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0
- Downloads last month
- 9
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Model tree for ribesstefano/RuleBert-v0.1-k2
Base model
FacebookAI/xlm-roberta-base