metadata
base_model: phiyodr/bert-base-finetuned-squad2
tags:
- generated_from_trainer
model-index:
- name: test-bert-1
results: []
test-bert-1
This model is a fine-tuned version of phiyodr/bert-base-finetuned-squad2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 8.3972
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: 3e-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
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 8.3538 | 0.09 | 5 | 9.3652 |
| 8.271 | 0.18 | 10 | 9.3025 |
| 8.2794 | 0.27 | 15 | 9.2420 |
| 8.0966 | 0.36 | 20 | 9.1837 |
| 8.1925 | 0.45 | 25 | 9.1277 |
| 8.1126 | 0.55 | 30 | 9.0740 |
| 7.711 | 0.64 | 35 | 9.0230 |
| 7.5978 | 0.73 | 40 | 8.9749 |
| 8.0276 | 0.82 | 45 | 8.9287 |
| 7.7477 | 0.91 | 50 | 8.8839 |
| 7.5803 | 1.0 | 55 | 8.8422 |
| 7.6439 | 1.09 | 60 | 8.8022 |
| 7.4937 | 1.18 | 65 | 8.7649 |
| 7.6905 | 1.27 | 70 | 8.7294 |
| 7.7922 | 1.36 | 75 | 8.6950 |
| 7.7377 | 1.45 | 80 | 8.6623 |
| 7.8642 | 1.55 | 85 | 8.6316 |
| 7.5909 | 1.64 | 90 | 8.6025 |
| 7.4629 | 1.73 | 95 | 8.5758 |
| 7.5195 | 1.82 | 100 | 8.5512 |
| 7.664 | 1.91 | 105 | 8.5282 |
| 7.3092 | 2.0 | 110 | 8.5074 |
| 7.015 | 2.09 | 115 | 8.4889 |
| 7.4112 | 2.18 | 120 | 8.4719 |
| 7.6664 | 2.27 | 125 | 8.4563 |
| 7.5634 | 2.36 | 130 | 8.4430 |
| 7.4502 | 2.45 | 135 | 8.4314 |
| 7.3085 | 2.55 | 140 | 8.4212 |
| 7.2558 | 2.64 | 145 | 8.4128 |
| 7.4077 | 2.73 | 150 | 8.4066 |
| 7.6318 | 2.82 | 155 | 8.4017 |
| 7.3355 | 2.91 | 160 | 8.3988 |
| 7.4367 | 3.0 | 165 | 8.3972 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1