<!-- 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-3-4000-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: 1.1602
- Accuracy: 0.9071
- F1: 0.8659
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 | 2.6064 | 0.8603 | 0.8031 |
| No log | 1.6 | 200 | 1.9752 | 0.8802 | 0.8384 |
| No log | 2.4 | 300 | 1.9469 | 0.8825 | 0.8374 |
| No log | 3.2 | 400 | 2.3416 | 0.8810 | 0.8469 |
| 1.9113 | 4.0 | 500 | 1.7481 | 0.8952 | 0.8548 |
| 1.9113 | 4.8 | 600 | 1.6816 | 0.8937 | 0.8525 |
| 1.9113 | 5.6 | 700 | 1.7184 | 0.8960 | 0.8447 |
| 1.9113 | 6.4 | 800 | 1.5429 | 0.8992 | 0.8614 |
| 1.9113 | 7.2 | 900 | 1.6045 | 0.9008 | 0.8621 |
| 0.7029 | 8.0 | 1000 | 1.4493 | 0.9008 | 0.8627 |
| 0.7029 | 8.8 | 1100 | 1.3875 | 0.9008 | 0.8493 |
| 0.7029 | 9.6 | 1200 | 1.4209 | 0.9040 | 0.8633 |
| 0.7029 | 10.4 | 1300 | 1.4195 | 0.9056 | 0.8693 |
| 0.7029 | 11.2 | 1400 | 1.3910 | 0.9032 | 0.8607 |
| 0.4942 | 12.0 | 1500 | 1.3877 | 0.9048 | 0.8651 |
| 0.4942 | 12.8 | 1600 | 1.2617 | 0.9 | 0.8623 |
| 0.4942 | 13.6 | 1700 | 1.3888 | 0.8968 | 0.8534 |
| 0.4942 | 14.4 | 1800 | 1.3091 | 0.9056 | 0.8668 |
| 0.4942 | 15.2 | 1900 | 1.1946 | 0.9087 | 0.8739 |
| 0.4141 | 16.0 | 2000 | 1.2246 | 0.9056 | 0.8702 |
| 0.4141 | 16.8 | 2100 | 1.2220 | 0.9008 | 0.8619 |
| 0.4141 | 17.6 | 2200 | 1.2171 | 0.9040 | 0.8668 |
| 0.4141 | 18.4 | 2300 | 1.1826 | 0.9040 | 0.8699 |
| 0.4141 | 19.2 | 2400 | 1.3287 | 0.8992 | 0.8588 |
| 0.3603 | 20.0 | 2500 | 1.2944 | 0.9008 | 0.8642 |
| 0.3603 | 20.8 | 2600 | 1.1306 | 0.9071 | 0.8724 |
| 0.3603 | 21.6 | 2700 | 1.1615 | 0.9087 | 0.8677 |
| 0.3603 | 22.4 | 2800 | 1.1964 | 0.9 | 0.8661 |
| 0.3603 | 23.2 | 2900 | 1.2759 | 0.9048 | 0.8684 |
| 0.3245 | 24.0 | 3000 | 1.2212 | 0.9056 | 0.8652 |
| 0.3245 | 24.8 | 3100 | 1.2150 | 0.8984 | 0.8536 |
| 0.3245 | 25.6 | 3200 | 1.2183 | 0.9032 | 0.8573 |
| 0.3245 | 26.4 | 3300 | 1.0802 | 0.9167 | 0.8775 |
| 0.3245 | 27.2 | 3400 | 1.2012 | 0.9008 | 0.8624 |
| 0.2977 | 28.0 | 3500 | 1.3696 | 0.9 | 0.8549 |
| 0.2977 | 28.8 | 3600 | 1.2059 | 0.9024 | 0.8535 |
| 0.2977 | 29.6 | 3700 | 1.2707 | 0.9016 | 0.8645 |
| 0.2977 | 30.4 | 3800 | 1.1167 | 0.9103 | 0.8743 |
| 0.2977 | 31.2 | 3900 | 1.1569 | 0.9048 | 0.8633 |
| 0.2826 | 32.0 | 4000 | 1.4307 | 0.9 | 0.8511 |
| 0.2826 | 32.8 | 4100 | 1.2148 | 0.9024 | 0.8619 |
| 0.2826 | 33.6 | 4200 | 1.2538 | 0.9 | 0.8613 |
| 0.2826 | 34.4 | 4300 | 1.0693 | 0.9103 | 0.8746 |
| 0.2826 | 35.2 | 4400 | 1.2773 | 0.9024 | 0.8622 |
| 0.2647 | 36.0 | 4500 | 1.1964 | 0.9016 | 0.8591 |
| 0.2647 | 36.8 | 4600 | 1.1228 | 0.9111 | 0.8716 |
| 0.2647 | 37.6 | 4700 | 1.1870 | 0.9048 | 0.8622 |
| 0.2647 | 38.4 | 4800 | 1.1602 | 0.9071 | 0.8659 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3
