<!-- 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. -->
multibert_1210seed7
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: 0.5019
- Precisions: 0.8874
- Recall: 0.7790
- F-measure: 0.8105
- Accuracy: 0.9107
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: 16
- eval_batch_size: 16
- seed: 7
- 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 |
|---|---|---|---|---|---|---|---|
| 0.6032 | 1.0 | 236 | 0.4733 | 0.8608 | 0.6507 | 0.6853 | 0.8645 |
| 0.3527 | 2.0 | 472 | 0.3790 | 0.8098 | 0.7259 | 0.7383 | 0.8826 |
| 0.2198 | 3.0 | 708 | 0.4191 | 0.8209 | 0.7632 | 0.7816 | 0.8936 |
| 0.1359 | 4.0 | 944 | 0.4433 | 0.8430 | 0.7344 | 0.7590 | 0.8924 |
| 0.0862 | 5.0 | 1180 | 0.5207 | 0.8067 | 0.7697 | 0.7838 | 0.8947 |
| 0.0637 | 6.0 | 1416 | 0.5019 | 0.8874 | 0.7790 | 0.8105 | 0.9107 |
| 0.0454 | 7.0 | 1652 | 0.5048 | 0.8049 | 0.8135 | 0.8070 | 0.9058 |
| 0.0318 | 8.0 | 1888 | 0.5969 | 0.8135 | 0.7710 | 0.7845 | 0.9003 |
| 0.024 | 9.0 | 2124 | 0.6388 | 0.8295 | 0.7999 | 0.8057 | 0.9048 |
| 0.0138 | 10.0 | 2360 | 0.6448 | 0.8304 | 0.7727 | 0.7949 | 0.9033 |
| 0.0084 | 11.0 | 2596 | 0.6589 | 0.8216 | 0.7756 | 0.7936 | 0.9017 |
| 0.0091 | 12.0 | 2832 | 0.6471 | 0.8340 | 0.7683 | 0.7952 | 0.9045 |
| 0.005 | 13.0 | 3068 | 0.6817 | 0.8600 | 0.7662 | 0.8034 | 0.9073 |
| 0.0045 | 14.0 | 3304 | 0.6774 | 0.8397 | 0.7680 | 0.7976 | 0.9077 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1
