<!-- 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-pre-finetune-div-2-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: 0.4278
- Accuracy: 0.9056
- F1: 0.8701
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 | 0.4720 | 0.8056 | 0.7662 |
| No log | 0.58 | 200 | 0.3207 | 0.8730 | 0.8336 |
| No log | 0.87 | 300 | 0.3667 | 0.8841 | 0.8114 |
| No log | 1.16 | 400 | 0.3267 | 0.9008 | 0.8691 |
| 0.3894 | 1.45 | 500 | 0.2668 | 0.9167 | 0.8846 |
| 0.3894 | 1.74 | 600 | 0.3569 | 0.9048 | 0.8532 |
| 0.3894 | 2.03 | 700 | 0.3204 | 0.8992 | 0.8617 |
| 0.3894 | 2.33 | 800 | 0.3048 | 0.9008 | 0.8664 |
| 0.3894 | 2.62 | 900 | 0.3945 | 0.8968 | 0.8622 |
| 0.204 | 2.91 | 1000 | 0.2661 | 0.9127 | 0.8734 |
| 0.204 | 3.2 | 1100 | 0.3420 | 0.9111 | 0.8725 |
| 0.204 | 3.49 | 1200 | 0.3142 | 0.9135 | 0.8799 |
| 0.204 | 3.78 | 1300 | 0.3165 | 0.9103 | 0.8742 |
| 0.204 | 4.07 | 1400 | 0.3072 | 0.9103 | 0.8696 |
| 0.1278 | 4.36 | 1500 | 0.4094 | 0.9063 | 0.8717 |
| 0.1278 | 4.65 | 1600 | 0.4570 | 0.9095 | 0.8669 |
| 0.1278 | 4.94 | 1700 | 0.3694 | 0.9095 | 0.8703 |
| 0.1278 | 5.23 | 1800 | 0.4550 | 0.9111 | 0.8766 |
| 0.1278 | 5.52 | 1900 | 0.3854 | 0.9135 | 0.8777 |
| 0.0913 | 5.81 | 2000 | 0.4278 | 0.9056 | 0.8701 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3
