paws_x_m_bert_only_ja
This model is a fine-tuned version of bert-base-multilingual-cased on the paws-x dataset.
It achieves the following results on the evaluation set:
- Loss: 0.7244
- Accuracy: 0.831
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: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 10
Training results
| Training Loss |
Epoch |
Step |
Validation Loss |
Accuracy |
| 0.4839 |
1.0 |
386 |
0.4311 |
0.791 |
| 0.2888 |
2.0 |
772 |
0.4328 |
0.8115 |
| 0.2095 |
3.0 |
1158 |
0.4820 |
0.8255 |
| 0.1604 |
4.0 |
1544 |
0.4594 |
0.832 |
| 0.1251 |
5.0 |
1930 |
0.5119 |
0.829 |
| 0.0956 |
6.0 |
2316 |
0.5388 |
0.8335 |
| 0.0791 |
7.0 |
2702 |
0.5696 |
0.834 |
| 0.064 |
8.0 |
3088 |
0.6405 |
0.837 |
| 0.0546 |
9.0 |
3474 |
0.6916 |
0.833 |
| 0.0464 |
10.0 |
3860 |
0.7244 |
0.831 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.13.0
- Datasets 2.6.1
- Tokenizers 0.13.1
