20231008-2-distilbert-base-multilingual-cased-new
This model is a fine-tuned version of distilbert-base-multilingual-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Accuracy: 0.5483
- Loss: 1.8877
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Accuracy | Validation Loss |
|---|---|---|---|---|
| 2.9855 | 1.82 | 200 | 0.3753 | 2.6688 |
| 2.5641 | 3.64 | 400 | 0.436 | 2.2769 |
| 2.385 | 5.45 | 600 | 0.4970 | 2.3584 |
| 2.2444 | 7.27 | 800 | 0.5095 | 1.9800 |
| 2.1422 | 9.09 | 1000 | 0.5072 | 2.1302 |
| 2.0398 | 10.91 | 1200 | 0.5717 | 1.9107 |
| 2.0215 | 12.73 | 1400 | 0.5435 | 1.9803 |
| 1.9611 | 14.55 | 1600 | 0.5678 | 1.9622 |
| 1.928 | 16.36 | 1800 | 0.5253 | 1.9510 |
| 1.8712 | 18.18 | 2000 | 0.5711 | 1.6620 |
| 1.8322 | 20.0 | 2200 | 0.5483 | 1.8877 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1