<!-- 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-scratch-silver-data-smsa-model-xlm-roberta-base
This model is a fine-tuned version of xlm-roberta-base on the smsa dataset. It achieves the following results on the evaluation set:
- Loss: 1.9375
- Accuracy: 0.8714
- F1: 0.8161
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 | 5.3026 | 0.7175 | 0.4819 |
| No log | 0.58 | 200 | 4.1689 | 0.7770 | 0.5325 |
| No log | 0.87 | 300 | 4.2795 | 0.7548 | 0.5057 |
| No log | 1.16 | 400 | 4.1872 | 0.7817 | 0.7052 |
| 4.5426 | 1.45 | 500 | 2.7055 | 0.85 | 0.8017 |
| 4.5426 | 1.74 | 600 | 2.6983 | 0.8563 | 0.8130 |
| 4.5426 | 2.03 | 700 | 3.4019 | 0.8254 | 0.7965 |
| 4.5426 | 2.33 | 800 | 2.5624 | 0.8579 | 0.8061 |
| 4.5426 | 2.62 | 900 | 2.3574 | 0.8659 | 0.8153 |
| 2.3412 | 2.91 | 1000 | 2.4085 | 0.8683 | 0.8316 |
| 2.3412 | 3.2 | 1100 | 2.6235 | 0.8595 | 0.8022 |
| 2.3412 | 3.49 | 1200 | 2.7817 | 0.8516 | 0.8023 |
| 2.3412 | 3.78 | 1300 | 2.3318 | 0.8683 | 0.8306 |
| 2.3412 | 4.07 | 1400 | 2.3613 | 0.8587 | 0.7911 |
| 1.6124 | 4.36 | 1500 | 2.2787 | 0.8683 | 0.8194 |
| 1.6124 | 4.65 | 1600 | 2.2896 | 0.8587 | 0.8017 |
| 1.6124 | 4.94 | 1700 | 2.2348 | 0.8659 | 0.8238 |
| 1.6124 | 5.23 | 1800 | 2.2386 | 0.8627 | 0.7989 |
| 1.6124 | 5.52 | 1900 | 2.2097 | 0.8778 | 0.8308 |
| 1.2907 | 5.81 | 2000 | 2.1637 | 0.8825 | 0.8378 |
| 1.2907 | 6.1 | 2100 | 2.0826 | 0.8865 | 0.8346 |
| 1.2907 | 6.4 | 2200 | 2.1444 | 0.8810 | 0.8376 |
| 1.2907 | 6.69 | 2300 | 2.1018 | 0.8698 | 0.8231 |
| 1.2907 | 6.98 | 2400 | 2.0959 | 0.8730 | 0.8274 |
| 1.1356 | 7.27 | 2500 | 2.2466 | 0.8667 | 0.8292 |
| 1.1356 | 7.56 | 2600 | 2.4450 | 0.8619 | 0.8156 |
| 1.1356 | 7.85 | 2700 | 2.2658 | 0.8778 | 0.8244 |
| 1.1356 | 8.14 | 2800 | 2.2488 | 0.8714 | 0.8309 |
| 1.1356 | 8.43 | 2900 | 2.0180 | 0.8762 | 0.8268 |
| 0.9637 | 8.72 | 3000 | 2.0077 | 0.8865 | 0.8477 |
| 0.9637 | 9.01 | 3100 | 2.1380 | 0.8706 | 0.8175 |
| 0.9637 | 9.3 | 3200 | 2.1575 | 0.8730 | 0.8162 |
| 0.9637 | 9.59 | 3300 | 2.1116 | 0.8683 | 0.8255 |
| 0.9637 | 9.88 | 3400 | 1.9180 | 0.8778 | 0.8301 |
| 0.8516 | 10.17 | 3500 | 2.1591 | 0.8714 | 0.8139 |
| 0.8516 | 10.47 | 3600 | 2.0185 | 0.8857 | 0.8388 |
| 0.8516 | 10.76 | 3700 | 1.9748 | 0.8810 | 0.8409 |
| 0.8516 | 11.05 | 3800 | 1.8923 | 0.8897 | 0.8518 |
| 0.8516 | 11.34 | 3900 | 2.2109 | 0.8722 | 0.8004 |
| 0.7572 | 11.63 | 4000 | 2.1431 | 0.8722 | 0.8245 |
| 0.7572 | 11.92 | 4100 | 2.2153 | 0.8706 | 0.8026 |
| 0.7572 | 12.21 | 4200 | 2.1741 | 0.8754 | 0.8358 |
| 0.7572 | 12.5 | 4300 | 2.0556 | 0.8770 | 0.8253 |
| 0.7572 | 12.79 | 4400 | 2.3891 | 0.8627 | 0.8146 |
| 0.7121 | 13.08 | 4500 | 2.3681 | 0.8683 | 0.8359 |
| 0.7121 | 13.37 | 4600 | 2.1352 | 0.8722 | 0.8078 |
| 0.7121 | 13.66 | 4700 | 1.8919 | 0.8817 | 0.8397 |
| 0.7121 | 13.95 | 4800 | 1.7781 | 0.8817 | 0.8376 |
| 0.7121 | 14.24 | 4900 | 2.0605 | 0.8825 | 0.8373 |
| 0.6564 | 14.53 | 5000 | 1.8521 | 0.8770 | 0.8375 |
| 0.6564 | 14.83 | 5100 | 1.8985 | 0.8817 | 0.8396 |
| 0.6564 | 15.12 | 5200 | 1.9006 | 0.8730 | 0.8263 |
| 0.6564 | 15.41 | 5300 | 1.9375 | 0.8714 | 0.8161 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3
