20231008-4-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: 0.6336
- Loss: 1.6836
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 |
|---|---|---|---|---|
| 2.9921 | 1.82 | 200 | 0.4005 | 2.6163 |
| 2.5551 | 3.64 | 400 | 0.4392 | 2.3236 |
| 2.3073 | 5.45 | 600 | 0.4470 | 2.5807 |
| 2.2161 | 7.27 | 800 | 0.4988 | 2.2248 |
| 2.1575 | 9.09 | 1000 | 0.5249 | 2.0541 |
| 2.0626 | 10.91 | 1200 | 0.4914 | 2.2755 |
| 1.9948 | 12.73 | 1400 | 0.4894 | 2.1453 |
| 1.9417 | 14.55 | 1600 | 0.5041 | 2.0137 |
| 1.8983 | 16.36 | 1800 | 0.5310 | 1.8913 |
| 1.8788 | 18.18 | 2000 | 0.5364 | 1.8624 |
| 1.8485 | 20.0 | 2200 | 0.6336 | 1.6836 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1