mongolian-xlm-roberta-base-ner
This model is a fine-tuned version of xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1298
- Precision: 0.9227
- Recall: 0.9298
- F1: 0.9262
- Accuracy: 0.9770
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: 16
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.203 | 1.0 | 477 | 0.0961 | 0.8798 | 0.8986 | 0.8891 | 0.9708 |
| 0.0807 | 2.0 | 954 | 0.0912 | 0.8989 | 0.9173 | 0.9080 | 0.9734 |
| 0.0581 | 3.0 | 1431 | 0.0860 | 0.9087 | 0.9219 | 0.9152 | 0.9754 |
| 0.0433 | 4.0 | 1908 | 0.0954 | 0.9133 | 0.9255 | 0.9194 | 0.9763 |
| 0.0316 | 5.0 | 2385 | 0.1010 | 0.9183 | 0.9265 | 0.9224 | 0.9767 |
| 0.0234 | 6.0 | 2862 | 0.1077 | 0.9178 | 0.9286 | 0.9232 | 0.9770 |
| 0.0178 | 7.0 | 3339 | 0.1195 | 0.9223 | 0.9291 | 0.9257 | 0.9765 |
| 0.0142 | 8.0 | 3816 | 0.1263 | 0.9154 | 0.9280 | 0.9216 | 0.9767 |
| 0.0108 | 9.0 | 4293 | 0.1284 | 0.9204 | 0.9297 | 0.9250 | 0.9769 |
| 0.0088 | 10.0 | 4770 | 0.1298 | 0.9227 | 0.9298 | 0.9262 | 0.9770 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3