xlm-roberta-base-mongolian-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: 0.1166
- Precision: 0.9251
- Recall: 0.9335
- F1: 0.9293
- Accuracy: 0.9787
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 |
|---|---|---|---|---|---|---|---|
| 0.2013 | 1.0 | 477 | 0.0958 | 0.8951 | 0.9124 | 0.9037 | 0.9731 |
| 0.0846 | 2.0 | 954 | 0.0825 | 0.9155 | 0.9240 | 0.9197 | 0.9774 |
| 0.0622 | 3.0 | 1431 | 0.0844 | 0.9109 | 0.9235 | 0.9172 | 0.9766 |
| 0.0456 | 4.0 | 1908 | 0.0940 | 0.9174 | 0.9266 | 0.9220 | 0.9767 |
| 0.0347 | 5.0 | 2385 | 0.1015 | 0.9184 | 0.9284 | 0.9234 | 0.9770 |
| 0.0253 | 6.0 | 2862 | 0.1117 | 0.9174 | 0.9254 | 0.9214 | 0.9764 |
| 0.0203 | 7.0 | 3339 | 0.1147 | 0.9225 | 0.9310 | 0.9267 | 0.9780 |
| 0.0152 | 8.0 | 3816 | 0.1129 | 0.9229 | 0.9316 | 0.9272 | 0.9779 |
| 0.0129 | 9.0 | 4293 | 0.1150 | 0.9245 | 0.9324 | 0.9285 | 0.9784 |
| 0.0102 | 10.0 | 4770 | 0.1166 | 0.9251 | 0.9335 | 0.9293 | 0.9787 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3