bert-f1-durga-muhammad
This model is a fine-tuned version of google-bert/bert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0079
- Accuracy: 0.999
- Precision: 0.999
- Recall: 0.999
- F1: 0.999
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: 4
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.1978 | 0.24 | 60 | 0.1764 | 0.968 | 0.968 | 0.968 | 0.968 |
| 0.1657 | 0.48 | 120 | 0.0619 | 0.981 | 0.981 | 0.981 | 0.981 |
| 0.1155 | 0.72 | 180 | 0.0475 | 0.989 | 0.989 | 0.989 | 0.989 |
| 0.0675 | 0.96 | 240 | 0.0143 | 0.997 | 0.997 | 0.997 | 0.997 |
| 0.0009 | 1.2 | 300 | 0.0148 | 0.997 | 0.997 | 0.997 | 0.997 |
| 0.0006 | 1.44 | 360 | 0.0151 | 0.997 | 0.997 | 0.997 | 0.997 |
| 0.0267 | 1.6800 | 420 | 0.0083 | 0.999 | 0.999 | 0.999 | 0.999 |
| 0.0335 | 1.92 | 480 | 0.0080 | 0.999 | 0.999 | 0.999 | 0.999 |
| 0.0315 | 2.16 | 540 | 0.0073 | 0.999 | 0.999 | 0.999 | 0.999 |
| 0.0056 | 2.4 | 600 | 0.0076 | 0.999 | 0.999 | 0.999 | 0.999 |
| 0.0004 | 2.64 | 660 | 0.0078 | 0.999 | 0.999 | 0.999 | 0.999 |
| 0.0004 | 2.88 | 720 | 0.0079 | 0.999 | 0.999 | 0.999 | 0.999 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.5.0+cu121
- Datasets 3.0.2
- Tokenizers 0.19.1