sentiment-distilbert-multilingual-5class
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: 1.5419
- Accuracy: 0.375
- Precision: 0.4133
- Recall: 0.37
- F1: 0.3804
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 1.6071 | 1.0 | 7 | 1.6086 | 0.2174 | 0.0435 | 0.2 | 0.0714 |
| 1.5880 | 2.0 | 14 | 1.6053 | 0.2174 | 0.0526 | 0.2 | 0.0833 |
| 1.5669 | 3.0 | 21 | 1.5993 | 0.2174 | 0.1471 | 0.2100 | 0.1394 |
| 1.5483 | 4.0 | 28 | 1.5949 | 0.2174 | 0.1783 | 0.24 | 0.2041 |
| 1.5114 | 5.0 | 35 | 1.5905 | 0.2609 | 0.2067 | 0.2800 | 0.2305 |
| 1.4738 | 6.0 | 42 | 1.5824 | 0.2609 | 0.2686 | 0.2800 | 0.26 |
| 1.4567 | 7.0 | 49 | 1.5781 | 0.3913 | 0.4971 | 0.4 | 0.4000 |
| 1.4533 | 8.0 | 56 | 1.5765 | 0.3913 | 0.4971 | 0.4 | 0.4000 |
Framework versions
- Transformers 5.10.2
- Pytorch 2.11.0+cu128
- Datasets 4.0.0
- Tokenizers 0.22.2