<!-- 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. -->
xnli_xlm_r_only_th
This model is a fine-tuned version of xlm-roberta-base on the xnli dataset. It achieves the following results on the evaluation set:
- Loss: 0.8277
- Accuracy: 0.7498
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: 2e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.7122 | 1.0 | 3068 | 0.6337 | 0.7398 |
| 0.5821 | 2.0 | 6136 | 0.5995 | 0.7610 |
| 0.5175 | 3.0 | 9204 | 0.5998 | 0.7671 |
| 0.4643 | 4.0 | 12272 | 0.6067 | 0.7627 |
| 0.4148 | 5.0 | 15340 | 0.6037 | 0.7755 |
| 0.3709 | 6.0 | 18408 | 0.6939 | 0.7550 |
| 0.3323 | 7.0 | 21476 | 0.7132 | 0.7530 |
| 0.2989 | 8.0 | 24544 | 0.7651 | 0.7530 |
| 0.272 | 9.0 | 27612 | 0.8219 | 0.7494 |
| 0.2527 | 10.0 | 30680 | 0.8277 | 0.7498 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.13.0
- Datasets 2.6.1
- Tokenizers 0.13.1
