xlm-roberta-base-Confusion-mlm-20230603
This model is a fine-tuned version of xlm-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:
- Accuracy: 0.8799
- Loss: 0.4204
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Accuracy | Validation Loss |
|---|---|---|---|---|
| 0.6139 | 1.0 | 130 | 0.9435 | 0.2902 |
| 0.6053 | 2.0 | 260 | 0.925 | 0.3066 |
| 0.6449 | 3.0 | 390 | 0.8884 | 0.4615 |
| 0.694 | 4.0 | 520 | 0.9058 | 0.3164 |
| 0.6768 | 5.0 | 650 | 0.9173 | 0.3421 |
| 0.6429 | 6.0 | 780 | 0.8968 | 0.3863 |
| 0.5945 | 7.0 | 910 | 0.8920 | 0.4843 |
| 0.632 | 8.0 | 1040 | 0.8984 | 0.3297 |
| 0.6015 | 9.0 | 1170 | 0.9219 | 0.3380 |
| 0.5431 | 10.0 | 1300 | 0.8967 | 0.4618 |
| 0.4813 | 11.0 | 1430 | 0.8885 | 0.4543 |
| 0.5598 | 12.0 | 1560 | 0.8885 | 0.4255 |
| 0.557 | 13.0 | 1690 | 0.8861 | 0.4283 |
| 0.4345 | 14.0 | 1820 | 0.8777 | 0.4792 |
| 0.5099 | 15.0 | 1950 | 0.9402 | 0.2961 |
| 0.5013 | 16.0 | 2080 | 0.8755 | 0.5784 |
| 0.5079 | 17.0 | 2210 | 0.9288 | 0.3294 |
| 0.557 | 18.0 | 2340 | 0.9167 | 0.3471 |
| 0.4574 | 19.0 | 2470 | 0.8969 | 0.4371 |
| 0.3938 | 20.0 | 2600 | 0.8799 | 0.4204 |
Framework versions
- Transformers 4.29.2
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3