distilbert-base-multilingual-cased-hyper-matt
This model is a fine-tuned version of distilbert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4924
- Accuracy: 0.88
- Recall: 0.7967
- Precision: 0.8099
- F1: 0.8033
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: 1.0247874543132884e-05
- train_batch_size: 4
- eval_batch_size: 16
- seed: 13
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Recall | Precision | F1 |
|---|---|---|---|---|---|---|---|
| 0.1205 | 1.0 | 400 | 0.3566 | 0.875 | 0.7967 | 0.7967 | 0.7967 |
| 0.1525 | 2.0 | 800 | 0.5043 | 0.855 | 0.8618 | 0.7211 | 0.7852 |
| 0.341 | 3.0 | 1200 | 0.4811 | 0.865 | 0.8211 | 0.7594 | 0.7891 |
| 0.1254 | 4.0 | 1600 | 0.4924 | 0.88 | 0.7967 | 0.8099 | 0.8033 |
Framework versions
- Transformers 4.46.3
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.20.3