multibert_testrun
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: 0.4300
- Precisions: 0.8488
- Recall: 0.7908
- F-measure: 0.8172
- Accuracy: 0.9404
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: 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 |
|---|---|---|---|---|---|---|---|
| 0.4196 | 1.0 | 269 | 0.3190 | 0.8426 | 0.7090 | 0.7230 | 0.9078 |
| 0.2111 | 2.0 | 538 | 0.2981 | 0.7730 | 0.7491 | 0.7551 | 0.9190 |
| 0.1275 | 3.0 | 807 | 0.2666 | 0.8158 | 0.7744 | 0.7915 | 0.9346 |
| 0.0868 | 4.0 | 1076 | 0.2929 | 0.8276 | 0.7891 | 0.8050 | 0.9349 |
| 0.0608 | 5.0 | 1345 | 0.3253 | 0.8370 | 0.7803 | 0.8043 | 0.9353 |
| 0.0353 | 6.0 | 1614 | 0.3723 | 0.8153 | 0.7999 | 0.8051 | 0.9360 |
| 0.0254 | 7.0 | 1883 | 0.4149 | 0.8266 | 0.7688 | 0.7934 | 0.9339 |
| 0.0203 | 8.0 | 2152 | 0.4399 | 0.8356 | 0.7755 | 0.8028 | 0.9357 |
| 0.0146 | 9.0 | 2421 | 0.4413 | 0.8295 | 0.7845 | 0.8045 | 0.9349 |
| 0.0108 | 10.0 | 2690 | 0.4300 | 0.8488 | 0.7908 | 0.8172 | 0.9404 |
| 0.0054 | 11.0 | 2959 | 0.4428 | 0.8317 | 0.7858 | 0.8062 | 0.9357 |
| 0.004 | 12.0 | 3228 | 0.4681 | 0.8403 | 0.7861 | 0.8095 | 0.9375 |
| 0.0019 | 13.0 | 3497 | 0.4725 | 0.8409 | 0.7901 | 0.8123 | 0.9386 |
| 0.0013 | 14.0 | 3766 | 0.4839 | 0.8437 | 0.7895 | 0.8137 | 0.9404 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1