metadata
license: cc-by-sa-4.0
base_model: cl-tohoku/bert-base-japanese
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: bert-base-japanese-ghost_rate
results: []
bert-base-japanese-ghost_rate
This model is a fine-tuned version of cl-tohoku/bert-base-japanese on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.7155
- Accuracy: 0.1667
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-07
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
No log | 1.0 | 275 | 1.7696 | 0.1858 |
1.7385 | 2.0 | 550 | 1.7574 | 0.1821 |
1.7385 | 3.0 | 825 | 1.7461 | 0.1785 |
1.7262 | 4.0 | 1100 | 1.7364 | 0.1758 |
1.7262 | 5.0 | 1375 | 1.7280 | 0.1712 |
1.7105 | 6.0 | 1650 | 1.7221 | 0.1703 |
1.7105 | 7.0 | 1925 | 1.7183 | 0.1667 |
1.6983 | 8.0 | 2200 | 1.7163 | 0.1667 |
1.6983 | 9.0 | 2475 | 1.7156 | 0.1667 |
1.697 | 10.0 | 2750 | 1.7155 | 0.1667 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1