<!-- 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-2-6000-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: 0.9913
- Accuracy: 0.9151
- F1: 0.8745
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 | 0.53 | 100 | 1.5706 | 0.8929 | 0.8530 |
| No log | 1.06 | 200 | 1.2589 | 0.9040 | 0.8620 |
| No log | 1.6 | 300 | 1.3527 | 0.9040 | 0.8535 |
| No log | 2.13 | 400 | 1.4937 | 0.9008 | 0.8608 |
| 1.3483 | 2.66 | 500 | 1.1956 | 0.9063 | 0.8659 |
| 1.3483 | 3.19 | 600 | 1.3803 | 0.9071 | 0.8559 |
| 1.3483 | 3.72 | 700 | 1.3785 | 0.8976 | 0.8561 |
| 1.3483 | 4.26 | 800 | 1.2104 | 0.9119 | 0.8705 |
| 1.3483 | 4.79 | 900 | 1.0962 | 0.9095 | 0.8722 |
| 0.672 | 5.32 | 1000 | 1.0363 | 0.9056 | 0.8632 |
| 0.672 | 5.85 | 1100 | 1.0125 | 0.9159 | 0.8809 |
| 0.672 | 6.38 | 1200 | 1.2072 | 0.9063 | 0.8710 |
| 0.672 | 6.91 | 1300 | 1.0776 | 0.9048 | 0.8638 |
| 0.672 | 7.45 | 1400 | 0.9799 | 0.9103 | 0.8686 |
| 0.5035 | 7.98 | 1500 | 1.0001 | 0.9127 | 0.8725 |
| 0.5035 | 8.51 | 1600 | 1.0030 | 0.9143 | 0.8768 |
| 0.5035 | 9.04 | 1700 | 1.0099 | 0.9190 | 0.8773 |
| 0.5035 | 9.57 | 1800 | 1.0064 | 0.9135 | 0.8767 |
| 0.5035 | 10.11 | 1900 | 0.9880 | 0.9127 | 0.8750 |
| 0.4353 | 10.64 | 2000 | 0.9869 | 0.9167 | 0.8764 |
| 0.4353 | 11.17 | 2100 | 0.9409 | 0.9175 | 0.8843 |
| 0.4353 | 11.7 | 2200 | 0.9905 | 0.9087 | 0.8751 |
| 0.4353 | 12.23 | 2300 | 0.9260 | 0.9135 | 0.8760 |
| 0.4353 | 12.77 | 2400 | 0.8649 | 0.9198 | 0.8843 |
| 0.3637 | 13.3 | 2500 | 1.0389 | 0.9008 | 0.8510 |
| 0.3637 | 13.83 | 2600 | 0.9714 | 0.9151 | 0.8762 |
| 0.3637 | 14.36 | 2700 | 0.9542 | 0.9119 | 0.8696 |
| 0.3637 | 14.89 | 2800 | 1.0179 | 0.9095 | 0.8646 |
| 0.3637 | 15.43 | 2900 | 0.8804 | 0.9190 | 0.8794 |
| 0.3489 | 15.96 | 3000 | 1.0735 | 0.9048 | 0.8687 |
| 0.3489 | 16.49 | 3100 | 0.8882 | 0.9119 | 0.8696 |
| 0.3489 | 17.02 | 3200 | 1.0558 | 0.9111 | 0.8602 |
| 0.3489 | 17.55 | 3300 | 0.8915 | 0.9079 | 0.8691 |
| 0.3489 | 18.09 | 3400 | 0.8256 | 0.9190 | 0.8836 |
| 0.3171 | 18.62 | 3500 | 0.9152 | 0.9198 | 0.8876 |
| 0.3171 | 19.15 | 3600 | 0.8762 | 0.9159 | 0.8824 |
| 0.3171 | 19.68 | 3700 | 0.8981 | 0.9127 | 0.8722 |
| 0.3171 | 20.21 | 3800 | 0.9151 | 0.9119 | 0.8737 |
| 0.3171 | 20.74 | 3900 | 0.9346 | 0.9159 | 0.8828 |
| 0.2863 | 21.28 | 4000 | 0.8687 | 0.9183 | 0.8837 |
| 0.2863 | 21.81 | 4100 | 0.9926 | 0.9095 | 0.8666 |
| 0.2863 | 22.34 | 4200 | 1.0249 | 0.9143 | 0.8817 |
| 0.2863 | 22.87 | 4300 | 0.9584 | 0.9135 | 0.8725 |
| 0.2863 | 23.4 | 4400 | 1.0183 | 0.9071 | 0.8656 |
| 0.2786 | 23.94 | 4500 | 1.0467 | 0.9135 | 0.8746 |
| 0.2786 | 24.47 | 4600 | 0.9518 | 0.9135 | 0.8705 |
| 0.2786 | 25.0 | 4700 | 0.9842 | 0.9063 | 0.8579 |
| 0.2786 | 25.53 | 4800 | 0.9568 | 0.9127 | 0.8742 |
| 0.2786 | 26.06 | 4900 | 0.9272 | 0.9095 | 0.8641 |
| 0.2633 | 26.6 | 5000 | 0.9913 | 0.9151 | 0.8745 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3
