rudashka-bge-baseline
This model is a fine-tuned version of BAAI/bge-small-en-v1.5 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1698
- Precision: 0.8886
- Recall: 0.9199
- F1: 0.9040
- Accuracy: 0.9800
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: 64
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.5710 | 1.0 | 313 | 0.4853 | 0.6635 | 0.7253 | 0.6930 | 0.9466 |
| 0.3530 | 2.0 | 626 | 0.2980 | 0.8054 | 0.8570 | 0.8304 | 0.9692 |
| 0.2370 | 3.0 | 939 | 0.2348 | 0.8301 | 0.8916 | 0.8598 | 0.9738 |
| 0.1803 | 4.0 | 1252 | 0.2022 | 0.8627 | 0.9002 | 0.8811 | 0.9773 |
| 0.1528 | 5.0 | 1565 | 0.1875 | 0.8718 | 0.9079 | 0.8895 | 0.9784 |
| 0.1430 | 6.0 | 1878 | 0.1815 | 0.8759 | 0.9132 | 0.8941 | 0.9790 |
| 0.1224 | 7.0 | 2191 | 0.1772 | 0.8797 | 0.9157 | 0.8973 | 0.9791 |
| 0.0907 | 8.0 | 2504 | 0.1727 | 0.8906 | 0.9162 | 0.9032 | 0.9801 |
| 0.0955 | 9.0 | 2817 | 0.1718 | 0.8844 | 0.9197 | 0.9017 | 0.9794 |
| 0.0896 | 10.0 | 3130 | 0.1698 | 0.8886 | 0.9199 | 0.9040 | 0.9800 |
Framework versions
- Transformers 5.2.0
- Pytorch 2.10.0+cu128
- Datasets 4.8.3
- Tokenizers 0.22.2