<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
bert_sm_gen1
This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.1391
- Accuracy: 0.829
- Precision: 0.5241
- Recall: 0.4270
- F1: 0.4706
- D-index: 1.6122
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: 4
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 8000
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | D-index |
|---|---|---|---|---|---|---|---|---|
| 0.4764 | 1.0 | 1000 | 0.4123 | 0.828 | 0.5455 | 0.2022 | 0.2951 | 1.5328 |
| 0.5783 | 2.0 | 2000 | 0.6138 | 0.83 | 0.5690 | 0.1854 | 0.2797 | 1.5295 |
| 0.5735 | 3.0 | 3000 | 0.7900 | 0.801 | 0.4525 | 0.5618 | 0.5013 | 1.6205 |
| 0.4248 | 4.0 | 4000 | 0.9244 | 0.84 | 0.6875 | 0.1854 | 0.2920 | 1.5429 |
| 0.2873 | 5.0 | 5000 | 1.0765 | 0.815 | 0.4774 | 0.4157 | 0.4444 | 1.5899 |
| 0.2717 | 6.0 | 6000 | 1.1807 | 0.814 | 0.4661 | 0.3090 | 0.3716 | 1.5518 |
| 0.2166 | 7.0 | 7000 | 1.2673 | 0.821 | 0.4970 | 0.4607 | 0.4781 | 1.6131 |
| 0.1294 | 8.0 | 8000 | 1.5151 | 0.808 | 0.4628 | 0.4888 | 0.4754 | 1.6054 |
| 0.0485 | 9.0 | 9000 | 1.6610 | 0.823 | 0.504 | 0.3539 | 0.4158 | 1.5794 |
| 0.0522 | 10.0 | 10000 | 1.8193 | 0.802 | 0.4519 | 0.5281 | 0.4870 | 1.6106 |
| 0.0307 | 11.0 | 11000 | 1.7044 | 0.828 | 0.5211 | 0.4157 | 0.4625 | 1.6071 |
| 0.0196 | 12.0 | 12000 | 1.8297 | 0.818 | 0.4873 | 0.4326 | 0.4583 | 1.5996 |
| 0.0048 | 13.0 | 13000 | 1.9419 | 0.827 | 0.5188 | 0.3876 | 0.4437 | 1.5962 |
| 0.0098 | 14.0 | 14000 | 2.0127 | 0.828 | 0.5211 | 0.4157 | 0.4625 | 1.6071 |
| 0.0082 | 15.0 | 15000 | 2.0195 | 0.833 | 0.5420 | 0.3989 | 0.4595 | 1.6079 |
| 0.0 | 16.0 | 16000 | 2.0748 | 0.827 | 0.5161 | 0.4494 | 0.4805 | 1.6172 |
| 0.0 | 17.0 | 17000 | 2.0948 | 0.831 | 0.5319 | 0.4213 | 0.4702 | 1.6129 |
| 0.0 | 18.0 | 18000 | 2.1141 | 0.831 | 0.5338 | 0.3989 | 0.4566 | 1.6053 |
| 0.0 | 19.0 | 19000 | 2.1411 | 0.828 | 0.5205 | 0.4270 | 0.4691 | 1.6109 |
| 0.0 | 20.0 | 20000 | 2.1391 | 0.829 | 0.5241 | 0.4270 | 0.4706 | 1.6122 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
