<!-- 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_1210seed85
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: 0.4043
- Precisions: 0.8689
- Recall: 0.8339
- F-measure: 0.8498
- Accuracy: 0.9067
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: 85
- 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 |
|---|---|---|---|---|---|---|---|
| 0.5968 | 1.0 | 236 | 0.4307 | 0.8843 | 0.6749 | 0.7165 | 0.8690 |
| 0.3238 | 2.0 | 472 | 0.3849 | 0.8827 | 0.7215 | 0.7489 | 0.8916 |
| 0.2021 | 3.0 | 708 | 0.4067 | 0.8540 | 0.7640 | 0.7969 | 0.8983 |
| 0.1335 | 4.0 | 944 | 0.3857 | 0.8227 | 0.8002 | 0.8071 | 0.8983 |
| 0.0886 | 5.0 | 1180 | 0.4043 | 0.8689 | 0.8339 | 0.8498 | 0.9067 |
| 0.0654 | 6.0 | 1416 | 0.4734 | 0.8847 | 0.8016 | 0.8359 | 0.9089 |
| 0.0451 | 7.0 | 1652 | 0.5312 | 0.8215 | 0.7826 | 0.7996 | 0.8980 |
| 0.031 | 8.0 | 1888 | 0.5520 | 0.8730 | 0.7873 | 0.8222 | 0.9074 |
| 0.0248 | 9.0 | 2124 | 0.4954 | 0.8896 | 0.8145 | 0.8454 | 0.9149 |
| 0.0149 | 10.0 | 2360 | 0.5595 | 0.8717 | 0.8101 | 0.8354 | 0.9104 |
| 0.0086 | 11.0 | 2596 | 0.5703 | 0.8814 | 0.8051 | 0.8348 | 0.9112 |
| 0.0061 | 12.0 | 2832 | 0.5855 | 0.8655 | 0.8138 | 0.8356 | 0.9103 |
| 0.006 | 13.0 | 3068 | 0.6068 | 0.8578 | 0.8137 | 0.8329 | 0.9105 |
| 0.0039 | 14.0 | 3304 | 0.6147 | 0.8656 | 0.8129 | 0.8356 | 0.9112 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1
