metadata
license: mit
base_model: megagonlabs/transformers-ud-japanese-electra-base-ginza-520
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: electra-ginza-based-ja-fc-classifier
results: []
electra-ginza-based-ja-fc-classifier
This model is a fine-tuned version of megagonlabs/transformers-ud-japanese-electra-base-ginza-520 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2376
- Accuracy: 0.9258
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: 3.38340974405913e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- distributed_type: tpu
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.4988 | 1.0 | 1223 | 0.3910 | 0.8418 |
| 0.3011 | 2.0 | 2446 | 0.2376 | 0.9258 |
| 0.1658 | 3.0 | 3669 | 0.2649 | 0.9336 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.0+cu118
- Datasets 2.14.5
- Tokenizers 0.14.0