<!-- 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. -->
mongolian-xlm-roberta-large-ner
This model is a fine-tuned version of xlm-roberta-large on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1256
- Precision: 0.9361
- Recall: 0.9423
- F1: 0.9392
- Accuracy: 0.9824
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 |
|---|---|---|---|---|---|---|---|
| 0.1837 | 1.0 | 477 | 0.0939 | 0.8524 | 0.8895 | 0.8705 | 0.9745 |
| 0.0736 | 2.0 | 954 | 0.0731 | 0.9318 | 0.9370 | 0.9344 | 0.9809 |
| 0.0525 | 3.0 | 1431 | 0.0724 | 0.9244 | 0.9311 | 0.9278 | 0.9795 |
| 0.036 | 4.0 | 1908 | 0.0807 | 0.9312 | 0.9409 | 0.9361 | 0.9819 |
| 0.0248 | 5.0 | 2385 | 0.0855 | 0.9314 | 0.9407 | 0.9360 | 0.9814 |
| 0.0163 | 6.0 | 2862 | 0.1014 | 0.9327 | 0.9397 | 0.9362 | 0.9815 |
| 0.0112 | 7.0 | 3339 | 0.0997 | 0.9354 | 0.9433 | 0.9393 | 0.9822 |
| 0.0064 | 8.0 | 3816 | 0.1171 | 0.9384 | 0.9432 | 0.9408 | 0.9824 |
| 0.0049 | 9.0 | 4293 | 0.1237 | 0.9355 | 0.9418 | 0.9387 | 0.9822 |
| 0.0024 | 10.0 | 4770 | 0.1256 | 0.9361 | 0.9423 | 0.9392 | 0.9824 |
Framework versions
- Transformers 4.29.2
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
