<!-- 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. -->
scenario-kd-from-post-finetune-gold-silver-div-3-8000-data-smsa-model-haryoaw-sc
This model is a fine-tuned version of haryoaw/scenario-normal-finetune-clf-data-smsa-model-xlm-roberta-base on the smsa dataset. It achieves the following results on the evaluation set:
- Loss: 0.9617
- Accuracy: 0.9159
- F1: 0.8815
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6969
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| No log | 0.4 | 100 | 2.7237 | 0.8627 | 0.8226 |
| No log | 0.8 | 200 | 1.9013 | 0.8905 | 0.8594 |
| No log | 1.2 | 300 | 1.7793 | 0.8944 | 0.8548 |
| No log | 1.6 | 400 | 1.7672 | 0.8921 | 0.8588 |
| 2.2175 | 2.0 | 500 | 1.5303 | 0.9016 | 0.8583 |
| 2.2175 | 2.4 | 600 | 1.5394 | 0.9040 | 0.8627 |
| 2.2175 | 2.8 | 700 | 1.3986 | 0.9056 | 0.8639 |
| 2.2175 | 3.2 | 800 | 1.5774 | 0.8984 | 0.8492 |
| 2.2175 | 3.6 | 900 | 1.4396 | 0.8952 | 0.8609 |
| 0.9672 | 4.0 | 1000 | 1.2810 | 0.9071 | 0.8673 |
| 0.9672 | 4.4 | 1100 | 1.1932 | 0.9103 | 0.8718 |
| 0.9672 | 4.8 | 1200 | 1.1350 | 0.9119 | 0.8742 |
| 0.9672 | 5.2 | 1300 | 1.2363 | 0.9135 | 0.8780 |
| 0.9672 | 5.6 | 1400 | 1.1641 | 0.9071 | 0.8626 |
| 0.6841 | 6.0 | 1500 | 1.1281 | 0.9079 | 0.8719 |
| 0.6841 | 6.4 | 1600 | 1.1603 | 0.9159 | 0.8763 |
| 0.6841 | 6.8 | 1700 | 1.2870 | 0.9087 | 0.8738 |
| 0.6841 | 7.2 | 1800 | 1.0976 | 0.9111 | 0.8760 |
| 0.6841 | 7.6 | 1900 | 1.1423 | 0.9143 | 0.8816 |
| 0.5322 | 8.0 | 2000 | 1.3210 | 0.9040 | 0.8668 |
| 0.5322 | 8.4 | 2100 | 1.1357 | 0.9151 | 0.8804 |
| 0.5322 | 8.8 | 2200 | 1.1500 | 0.9111 | 0.8800 |
| 0.5322 | 9.2 | 2300 | 1.0416 | 0.9103 | 0.8704 |
| 0.5322 | 9.6 | 2400 | 1.1862 | 0.9040 | 0.8637 |
| 0.4735 | 10.0 | 2500 | 1.1656 | 0.9063 | 0.8702 |
| 0.4735 | 10.4 | 2600 | 1.0991 | 0.9190 | 0.8843 |
| 0.4735 | 10.8 | 2700 | 1.0836 | 0.9119 | 0.8730 |
| 0.4735 | 11.2 | 2800 | 1.1783 | 0.9079 | 0.8632 |
| 0.4735 | 11.6 | 2900 | 1.0525 | 0.9119 | 0.8772 |
| 0.4363 | 12.0 | 3000 | 1.0336 | 0.9183 | 0.8834 |
| 0.4363 | 12.4 | 3100 | 1.0902 | 0.9175 | 0.8792 |
| 0.4363 | 12.8 | 3200 | 1.0225 | 0.9214 | 0.8798 |
| 0.4363 | 13.2 | 3300 | 1.1369 | 0.9056 | 0.8592 |
| 0.4363 | 13.6 | 3400 | 1.0268 | 0.9087 | 0.8697 |
| 0.4078 | 14.0 | 3500 | 1.0688 | 0.9127 | 0.8707 |
| 0.4078 | 14.4 | 3600 | 0.9677 | 0.9119 | 0.8770 |
| 0.4078 | 14.8 | 3700 | 1.0233 | 0.9135 | 0.8772 |
| 0.4078 | 15.2 | 3800 | 0.9663 | 0.9206 | 0.8830 |
| 0.4078 | 15.6 | 3900 | 0.9963 | 0.9198 | 0.8838 |
| 0.3809 | 16.0 | 4000 | 1.0626 | 0.9087 | 0.8715 |
| 0.3809 | 16.4 | 4100 | 0.9617 | 0.9159 | 0.8815 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3
