ce-MiniLM-L6layer
This model is a fine-tuned version of cross-encoder/ms-marco-MiniLM-L-6-v2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1559
- Accuracy: 0.7273
- Precision: 0.9091
- Recall: 0.6349
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: 5e-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: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall |
|---|---|---|---|---|---|---|
| 12.9679 | 1.0 | 56 | 20.2827 | 0.6970 | 0.7797 | 0.7302 |
| 9.2483 | 2.0 | 112 | 12.1491 | 0.6465 | 0.7188 | 0.7302 |
| 1.9612 | 3.0 | 168 | 1.7406 | 0.6667 | 0.8409 | 0.5873 |
| 0.5046 | 4.0 | 224 | 0.4060 | 0.6061 | 0.8158 | 0.4921 |
| 0.3575 | 5.0 | 280 | 0.2410 | 0.6667 | 0.7885 | 0.6508 |
| 0.244 | 6.0 | 336 | 0.1860 | 0.6263 | 0.9062 | 0.4603 |
| 0.2324 | 7.0 | 392 | 0.1706 | 0.6970 | 0.9231 | 0.5714 |
| 0.1958 | 8.0 | 448 | 0.1873 | 0.7172 | 0.7869 | 0.7619 |
| 0.1687 | 9.0 | 504 | 0.1742 | 0.7778 | 0.8868 | 0.7460 |
| 0.1581 | 10.0 | 560 | 0.1559 | 0.7273 | 0.9091 | 0.6349 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.1
- Datasets 2.14.6
- Tokenizers 0.15.1