<!-- 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-non-kd-from-post-finetune-div-4-data-smsa-model-haryoaw-scenario-normal
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: 1.2567
- Accuracy: 0.9
- F1: 0.8559
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 | 0.29 | 100 | 2.7460 | 0.8484 | 0.7872 |
| No log | 0.58 | 200 | 2.4746 | 0.8667 | 0.8332 |
| No log | 0.87 | 300 | 2.1068 | 0.8698 | 0.8198 |
| No log | 1.16 | 400 | 2.0103 | 0.8794 | 0.8369 |
| 2.613 | 1.45 | 500 | 1.9245 | 0.8841 | 0.8360 |
| 2.613 | 1.74 | 600 | 1.9021 | 0.8921 | 0.8495 |
| 2.613 | 2.03 | 700 | 1.8269 | 0.8833 | 0.8354 |
| 2.613 | 2.33 | 800 | 1.7552 | 0.8897 | 0.8430 |
| 2.613 | 2.62 | 900 | 1.9049 | 0.8841 | 0.8432 |
| 1.3691 | 2.91 | 1000 | 1.4065 | 0.9024 | 0.8594 |
| 1.3691 | 3.2 | 1100 | 1.5714 | 0.9032 | 0.8564 |
| 1.3691 | 3.49 | 1200 | 1.4925 | 0.9040 | 0.8671 |
| 1.3691 | 3.78 | 1300 | 1.4045 | 0.9 | 0.8535 |
| 1.3691 | 4.07 | 1400 | 1.4670 | 0.8992 | 0.8547 |
| 0.8956 | 4.36 | 1500 | 1.4870 | 0.9008 | 0.8647 |
| 0.8956 | 4.65 | 1600 | 1.4043 | 0.9040 | 0.8611 |
| 0.8956 | 4.94 | 1700 | 1.3597 | 0.9063 | 0.8707 |
| 0.8956 | 5.23 | 1800 | 1.3316 | 0.9040 | 0.8609 |
| 0.8956 | 5.52 | 1900 | 1.3629 | 0.8992 | 0.8529 |
| 0.6887 | 5.81 | 2000 | 1.1770 | 0.9103 | 0.8725 |
| 0.6887 | 6.1 | 2100 | 1.3610 | 0.9095 | 0.8696 |
| 0.6887 | 6.4 | 2200 | 1.2546 | 0.9095 | 0.8691 |
| 0.6887 | 6.69 | 2300 | 1.2785 | 0.9063 | 0.8728 |
| 0.6887 | 6.98 | 2400 | 1.2374 | 0.9063 | 0.8642 |
| 0.5938 | 7.27 | 2500 | 1.2526 | 0.9056 | 0.8630 |
| 0.5938 | 7.56 | 2600 | 1.2068 | 0.9135 | 0.8731 |
| 0.5938 | 7.85 | 2700 | 1.2689 | 0.9048 | 0.8667 |
| 0.5938 | 8.14 | 2800 | 1.2339 | 0.9040 | 0.8645 |
| 0.5938 | 8.43 | 2900 | 1.1751 | 0.9095 | 0.8698 |
| 0.5161 | 8.72 | 3000 | 1.2023 | 0.9087 | 0.8682 |
| 0.5161 | 9.01 | 3100 | 1.2265 | 0.9056 | 0.8496 |
| 0.5161 | 9.3 | 3200 | 1.2097 | 0.9095 | 0.8713 |
| 0.5161 | 9.59 | 3300 | 1.3023 | 0.9063 | 0.8723 |
| 0.5161 | 9.88 | 3400 | 1.1319 | 0.9087 | 0.8666 |
| 0.4863 | 10.17 | 3500 | 1.1575 | 0.9071 | 0.8675 |
| 0.4863 | 10.47 | 3600 | 1.2294 | 0.9063 | 0.8681 |
| 0.4863 | 10.76 | 3700 | 1.0715 | 0.9119 | 0.8726 |
| 0.4863 | 11.05 | 3800 | 1.1397 | 0.9095 | 0.8709 |
| 0.4863 | 11.34 | 3900 | 1.2156 | 0.9079 | 0.8577 |
| 0.4467 | 11.63 | 4000 | 1.2038 | 0.9056 | 0.8600 |
| 0.4467 | 11.92 | 4100 | 1.2567 | 0.9 | 0.8559 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3
