20231008-3-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: 0.5585
- Loss: 1.8969
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 |
|---|---|---|---|---|
| 3.0355 | 1.82 | 200 | 0.3649 | 2.5361 |
| 2.5432 | 3.64 | 400 | 0.4264 | 2.4729 |
| 2.3986 | 5.45 | 600 | 0.4748 | 2.1543 |
| 2.2755 | 7.27 | 800 | 0.5533 | 1.9309 |
| 2.1743 | 9.09 | 1000 | 0.5165 | 2.1052 |
| 2.1152 | 10.91 | 1200 | 0.5307 | 2.0024 |
| 1.9368 | 12.73 | 1400 | 0.5820 | 1.7487 |
| 1.989 | 14.55 | 1600 | 0.5507 | 1.9093 |
| 1.8819 | 16.36 | 1800 | 0.5835 | 1.6382 |
| 1.8822 | 18.18 | 2000 | 0.5873 | 1.7373 |
| 1.8628 | 20.0 | 2200 | 0.5585 | 1.8969 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1