20231005-5-bert-base-multilingual-cased-new
This model is a fine-tuned version of bert-base-multilingual-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Accuracy: 0.5942
- Loss: 1.8864
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 |
|---|---|---|---|---|
| 2.957 | 1.82 | 200 | 0.425 | 2.5896 |
| 2.5005 | 3.64 | 400 | 0.4635 | 2.2940 |
| 2.2966 | 5.45 | 600 | 0.5036 | 2.2117 |
| 2.1629 | 7.27 | 800 | 0.5118 | 2.0541 |
| 2.0612 | 9.09 | 1000 | 0.4856 | 2.2364 |
| 1.9424 | 10.91 | 1200 | 0.5486 | 1.9967 |
| 1.9329 | 12.73 | 1400 | 0.5419 | 1.8949 |
| 1.8239 | 14.55 | 1600 | 0.5849 | 1.7429 |
| 1.7774 | 16.36 | 1800 | 0.6025 | 1.6645 |
| 1.756 | 18.18 | 2000 | 0.5817 | 1.8334 |
| 1.7246 | 20.0 | 2200 | 0.5942 | 1.8864 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1