<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
japanese-roberta-base-finetuned-rinna
This model is a fine-tuned version of rinna/japanese-roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.9637
- Precision: 0.8062
- Recall: 0.8107
- F1: 0.8071
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: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 |
|---|---|---|---|---|---|---|
| 0.7188 | 1.0 | 617 | 0.5825 | 0.7584 | 0.7945 | 0.7724 |
| 0.5951 | 2.0 | 1234 | 0.7466 | 0.7489 | 0.7799 | 0.7628 |
| 0.5206 | 3.0 | 1851 | 0.9570 | 0.7814 | 0.7767 | 0.7747 |
| 0.4061 | 4.0 | 2468 | 0.9637 | 0.8062 | 0.8107 | 0.8071 |
| 0.1879 | 5.0 | 3085 | 1.0516 | 0.8028 | 0.8026 | 0.8021 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0
- Datasets 2.8.0
- Tokenizers 0.13.2
