fc-binary-unfrozen-model
This model is a fine-tuned version of line-corporation/line-distilbert-base-japanese on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3561
- Accuracy: 0.9219
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: 0.0001
- train_batch_size: 64
- 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: 6
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
No log | 1.0 | 306 | 0.2484 | 0.9199 |
0.2878 | 2.0 | 612 | 0.2141 | 0.9180 |
0.2878 | 3.0 | 918 | 0.2149 | 0.9316 |
0.144 | 4.0 | 1224 | 0.2708 | 0.9180 |
0.065 | 5.0 | 1530 | 0.3119 | 0.9238 |
0.065 | 6.0 | 1836 | 0.3561 | 0.9219 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.0+cu118
- Datasets 2.14.5
- Tokenizers 0.14.0
- Downloads last month
- 2