20231008-10-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: 0.6026
- Loss: 1.8350
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 |
|---|---|---|---|---|
| 3.0584 | 1.82 | 200 | 0.3419 | 2.8695 |
| 2.5725 | 3.64 | 400 | 0.4167 | 2.5994 |
| 2.3538 | 5.45 | 600 | 0.4714 | 2.3281 |
| 2.2078 | 7.27 | 800 | 0.4617 | 2.4474 |
| 2.1174 | 9.09 | 1000 | 0.5062 | 2.2134 |
| 2.0354 | 10.91 | 1200 | 0.5208 | 2.1015 |
| 2.0026 | 12.73 | 1400 | 0.4932 | 2.0366 |
| 1.9121 | 14.55 | 1600 | 0.5892 | 1.7582 |
| 1.903 | 16.36 | 1800 | 0.5332 | 2.0146 |
| 1.8417 | 18.18 | 2000 | 0.5442 | 1.9231 |
| 1.8791 | 20.0 | 2200 | 0.6026 | 1.8350 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1