bert-base-multilingual-cased-Confusion-mlm-20230604
This model is a fine-tuned version of bert-base-multilingual-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Accuracy: 0.8295
- Loss: 0.6294
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Accuracy | Validation Loss |
|---|---|---|---|---|
| 0.732 | 1.0 | 130 | 0.8361 | 0.5171 |
| 0.6489 | 2.0 | 260 | 0.8421 | 0.6181 |
| 0.7594 | 3.0 | 390 | 0.8474 | 0.5202 |
| 0.7408 | 4.0 | 520 | 0.8212 | 0.7330 |
| 0.7802 | 5.0 | 650 | 0.8562 | 0.6039 |
| 0.7458 | 6.0 | 780 | 0.7885 | 0.7175 |
| 0.6875 | 7.0 | 910 | 0.8247 | 0.5894 |
| 0.6826 | 8.0 | 1040 | 0.7943 | 0.7404 |
| 0.6979 | 9.0 | 1170 | 0.8370 | 0.5288 |
| 0.6011 | 10.0 | 1300 | 0.8227 | 0.7400 |
| 0.5925 | 11.0 | 1430 | 0.8333 | 0.6346 |
| 0.6168 | 12.0 | 1560 | 0.8509 | 0.5866 |
| 0.5798 | 13.0 | 1690 | 0.8012 | 0.7777 |
| 0.5861 | 14.0 | 1820 | 0.8108 | 0.6529 |
| 0.53 | 15.0 | 1950 | 0.8297 | 0.5617 |
| 0.5992 | 16.0 | 2080 | 0.8505 | 0.5977 |
| 0.5424 | 17.0 | 2210 | 0.8359 | 0.5167 |
| 0.567 | 18.0 | 2340 | 0.8585 | 0.4843 |
| 0.4832 | 19.0 | 2470 | 0.8799 | 0.4312 |
| 0.4914 | 20.0 | 2600 | 0.8295 | 0.6294 |
Framework versions
- Transformers 4.29.2
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3