bert-base-multilingual-uncased-hin-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: 0.2864
- F1: 0.7690
- Roc Auc: 0.8690
- Accuracy: 0.7053
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 |
|---|---|---|---|---|---|---|
| 0.1586 | 1.0 | 574 | 0.1943 | 0.7398 | 0.8388 | 0.6971 |
| 0.0729 | 2.0 | 1148 | 0.1984 | 0.7539 | 0.8476 | 0.7121 |
| 0.0427 | 3.0 | 1722 | 0.2212 | 0.7685 | 0.8524 | 0.7326 |
| 0.0269 | 4.0 | 2296 | 0.2522 | 0.7598 | 0.8611 | 0.7080 |
| 0.014 | 5.0 | 2870 | 0.2864 | 0.7690 | 0.8690 | 0.7053 |
| 0.0133 | 6.0 | 3444 | 0.3065 | 0.7673 | 0.8757 | 0.6999 |
| 0.0066 | 7.0 | 4018 | 0.3189 | 0.7619 | 0.8499 | 0.7149 |
| 0.007 | 8.0 | 4592 | 0.3584 | 0.7384 | 0.8529 | 0.6794 |
| 0.0031 | 9.0 | 5166 | 0.3422 | 0.7651 | 0.8611 | 0.7094 |
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0