<!-- 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-4-4000-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: 1.4203
- Accuracy: 0.8968
- F1: 0.8578
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 | 0.8 | 100 | 2.9174 | 0.8540 | 0.8084 |
| No log | 1.6 | 200 | 2.6630 | 0.8579 | 0.7807 |
| No log | 2.4 | 300 | 1.9436 | 0.8810 | 0.8379 |
| No log | 3.2 | 400 | 2.0654 | 0.8817 | 0.8342 |
| 2.3053 | 4.0 | 500 | 1.8668 | 0.8881 | 0.8416 |
| 2.3053 | 4.8 | 600 | 2.0437 | 0.8833 | 0.8470 |
| 2.3053 | 5.6 | 700 | 2.0103 | 0.8865 | 0.8375 |
| 2.3053 | 6.4 | 800 | 1.9602 | 0.8849 | 0.8416 |
| 2.3053 | 7.2 | 900 | 1.8075 | 0.8897 | 0.8514 |
| 0.7844 | 8.0 | 1000 | 1.9642 | 0.8833 | 0.8384 |
| 0.7844 | 8.8 | 1100 | 1.8446 | 0.8929 | 0.8530 |
| 0.7844 | 9.6 | 1200 | 1.6398 | 0.8873 | 0.8472 |
| 0.7844 | 10.4 | 1300 | 1.7704 | 0.8881 | 0.8449 |
| 0.7844 | 11.2 | 1400 | 1.8369 | 0.8992 | 0.8628 |
| 0.5562 | 12.0 | 1500 | 1.6476 | 0.8929 | 0.8476 |
| 0.5562 | 12.8 | 1600 | 1.7742 | 0.8873 | 0.8463 |
| 0.5562 | 13.6 | 1700 | 1.6897 | 0.8992 | 0.8633 |
| 0.5562 | 14.4 | 1800 | 1.7927 | 0.8921 | 0.8544 |
| 0.5562 | 15.2 | 1900 | 1.6334 | 0.8952 | 0.8561 |
| 0.4566 | 16.0 | 2000 | 1.6920 | 0.8881 | 0.8519 |
| 0.4566 | 16.8 | 2100 | 1.5482 | 0.9008 | 0.8680 |
| 0.4566 | 17.6 | 2200 | 1.6125 | 0.8960 | 0.8559 |
| 0.4566 | 18.4 | 2300 | 1.6259 | 0.8960 | 0.8597 |
| 0.4566 | 19.2 | 2400 | 1.5829 | 0.8921 | 0.8506 |
| 0.3913 | 20.0 | 2500 | 1.6328 | 0.8968 | 0.8538 |
| 0.3913 | 20.8 | 2600 | 1.5371 | 0.8984 | 0.8635 |
| 0.3913 | 21.6 | 2700 | 1.5889 | 0.8897 | 0.8526 |
| 0.3913 | 22.4 | 2800 | 1.4394 | 0.8944 | 0.8511 |
| 0.3913 | 23.2 | 2900 | 1.4062 | 0.9063 | 0.8742 |
| 0.3498 | 24.0 | 3000 | 1.4479 | 0.8992 | 0.8653 |
| 0.3498 | 24.8 | 3100 | 1.3948 | 0.8968 | 0.8615 |
| 0.3498 | 25.6 | 3200 | 1.4202 | 0.8960 | 0.8523 |
| 0.3498 | 26.4 | 3300 | 1.5816 | 0.8865 | 0.8500 |
| 0.3498 | 27.2 | 3400 | 1.6830 | 0.8889 | 0.8566 |
| 0.3183 | 28.0 | 3500 | 1.4982 | 0.8960 | 0.8544 |
| 0.3183 | 28.8 | 3600 | 1.4614 | 0.8944 | 0.8554 |
| 0.3183 | 29.6 | 3700 | 1.5068 | 0.8952 | 0.8574 |
| 0.3183 | 30.4 | 3800 | 1.5978 | 0.8873 | 0.8427 |
| 0.3183 | 31.2 | 3900 | 1.5153 | 0.8968 | 0.8591 |
| 0.2947 | 32.0 | 4000 | 1.6809 | 0.8833 | 0.8353 |
| 0.2947 | 32.8 | 4100 | 1.5216 | 0.9008 | 0.8626 |
| 0.2947 | 33.6 | 4200 | 1.4883 | 0.8937 | 0.8517 |
| 0.2947 | 34.4 | 4300 | 1.4585 | 0.8952 | 0.8554 |
| 0.2947 | 35.2 | 4400 | 1.4203 | 0.8968 | 0.8578 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3
