mongolian-roberta-large-mnli-ner
This model is a fine-tuned version of roberta-large-mnli on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1941
- Precision: 0.7734
- Recall: 0.8488
- F1: 0.8094
- Accuracy: 0.9582
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 |
|---|---|---|---|---|---|---|---|
| 0.3433 | 1.0 | 477 | 0.2252 | 0.6196 | 0.7338 | 0.6719 | 0.9288 |
| 0.2067 | 2.0 | 954 | 0.1859 | 0.6981 | 0.7908 | 0.7416 | 0.9381 |
| 0.165 | 3.0 | 1431 | 0.1776 | 0.7308 | 0.8112 | 0.7689 | 0.9455 |
| 0.1362 | 4.0 | 1908 | 0.1639 | 0.7513 | 0.8265 | 0.7871 | 0.9520 |
| 0.109 | 5.0 | 2385 | 0.1703 | 0.7524 | 0.8302 | 0.7894 | 0.9517 |
| 0.0873 | 6.0 | 2862 | 0.1690 | 0.7643 | 0.8396 | 0.8002 | 0.9552 |
| 0.0697 | 7.0 | 3339 | 0.1754 | 0.7696 | 0.8442 | 0.8052 | 0.9557 |
| 0.0552 | 8.0 | 3816 | 0.1793 | 0.7687 | 0.8468 | 0.8059 | 0.9572 |
| 0.0434 | 9.0 | 4293 | 0.1878 | 0.7842 | 0.8507 | 0.8161 | 0.9580 |
| 0.0354 | 10.0 | 4770 | 0.1941 | 0.7734 | 0.8488 | 0.8094 | 0.9582 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3