<!-- 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. -->
multibert1110_lrate7.5b4
This model is a fine-tuned version of bert-base-multilingual-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.7163
- Precisions: 0.8864
- Recall: 0.8013
- F-measure: 0.8374
- Accuracy: 0.9059
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: 7.5e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 14
Training results
| Training Loss | Epoch | Step | Validation Loss | Precisions | Recall | F-measure | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.7304 | 1.0 | 942 | 0.4905 | 0.8049 | 0.6436 | 0.6549 | 0.8554 |
| 0.4336 | 2.0 | 1884 | 0.6035 | 0.8585 | 0.6334 | 0.6863 | 0.8477 |
| 0.3238 | 3.0 | 2826 | 0.5094 | 0.8668 | 0.7014 | 0.7232 | 0.8882 |
| 0.249 | 4.0 | 3768 | 0.5951 | 0.8798 | 0.7110 | 0.7609 | 0.8770 |
| 0.191 | 5.0 | 4710 | 0.4988 | 0.8304 | 0.7761 | 0.7816 | 0.8975 |
| 0.1513 | 6.0 | 5652 | 0.5998 | 0.8351 | 0.7917 | 0.8062 | 0.8962 |
| 0.1088 | 7.0 | 6594 | 0.5874 | 0.8427 | 0.7953 | 0.8158 | 0.9003 |
| 0.0914 | 8.0 | 7536 | 0.5529 | 0.8580 | 0.7885 | 0.8087 | 0.9069 |
| 0.0682 | 9.0 | 8478 | 0.6882 | 0.8371 | 0.7773 | 0.8024 | 0.8958 |
| 0.0487 | 10.0 | 9420 | 0.7163 | 0.8864 | 0.8013 | 0.8374 | 0.9059 |
| 0.0319 | 11.0 | 10362 | 0.7020 | 0.8724 | 0.7867 | 0.8235 | 0.9007 |
| 0.0305 | 12.0 | 11304 | 0.6886 | 0.8689 | 0.8002 | 0.8311 | 0.9079 |
| 0.0184 | 13.0 | 12246 | 0.6994 | 0.8680 | 0.8089 | 0.8357 | 0.9085 |
| 0.0138 | 14.0 | 13188 | 0.7183 | 0.8677 | 0.8105 | 0.8362 | 0.9093 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1
