ironita_finetuned_no_correlations
This model is a fine-tuned version of lupobricco/irony_classification_ita_base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2605
- F1: 0.4
- Roc Auc: 0.6578
- Accuracy: 0.8041
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: 8
- eval_batch_size: 8
- 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 | F1 | Roc Auc | Accuracy |
---|---|---|---|---|---|---|
No log | 1.0 | 291 | 0.2678 | 0.0183 | 0.5046 | 0.8282 |
0.247 | 2.0 | 582 | 0.2605 | 0.4 | 0.6578 | 0.8041 |
0.247 | 3.0 | 873 | 0.3821 | 0.3636 | 0.6261 | 0.8265 |
0.1398 | 4.0 | 1164 | 0.4139 | 0.2895 | 0.5914 | 0.8247 |
0.1398 | 5.0 | 1455 | 0.4639 | 0.2270 | 0.5660 | 0.8230 |
0.0811 | 6.0 | 1746 | 0.4310 | 0.3253 | 0.6103 | 0.8162 |
0.0526 | 7.0 | 2037 | 0.5051 | 0.2597 | 0.5803 | 0.8162 |
0.0526 | 8.0 | 2328 | 0.5112 | 0.2710 | 0.5849 | 0.8179 |
0.029 | 9.0 | 2619 | 0.5245 | 0.2727 | 0.5854 | 0.8213 |
0.029 | 10.0 | 2910 | 0.5348 | 0.2745 | 0.5859 | 0.8230 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.3.0+cu118
- Datasets 2.19.0
- Tokenizers 0.19.1
- Downloads last month
- 5