bert-base-multilingual-uncased-mar-MICRO
This model is a fine-tuned version of bert-base-multilingual-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1625
- F1: 0.8438
- Roc Auc: 0.8963
- Accuracy: 0.8080
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy |
|---|---|---|---|---|---|---|
| 0.1477 | 1.0 | 551 | 0.1620 | 0.8244 | 0.8746 | 0.7895 |
| 0.0698 | 2.0 | 1102 | 0.1553 | 0.8395 | 0.8918 | 0.8065 |
| 0.0458 | 3.0 | 1653 | 0.1625 | 0.8438 | 0.8963 | 0.8080 |
| 0.0313 | 4.0 | 2204 | 0.2110 | 0.8164 | 0.8970 | 0.7639 |
| 0.0193 | 5.0 | 2755 | 0.2086 | 0.8353 | 0.8969 | 0.7966 |
| 0.0181 | 6.0 | 3306 | 0.2413 | 0.8258 | 0.8972 | 0.7852 |
| 0.0102 | 7.0 | 3857 | 0.2349 | 0.8335 | 0.8924 | 0.7980 |
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0