furina_amh_corr_5e-06
This model is a fine-tuned version of yihongLiu/furina on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0210
- Spearman Corr: 0.7684
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: 5e-06
- train_batch_size: 32
- eval_batch_size: 128
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Spearman Corr |
---|---|---|---|---|
No log | 0.9 | 200 | 0.0210 | 0.7728 |
No log | 1.81 | 400 | 0.0217 | 0.7612 |
0.0131 | 2.71 | 600 | 0.0206 | 0.7664 |
0.0131 | 3.62 | 800 | 0.0211 | 0.7672 |
0.0116 | 4.52 | 1000 | 0.0203 | 0.7709 |
0.0116 | 5.43 | 1200 | 0.0211 | 0.7678 |
0.0107 | 6.33 | 1400 | 0.0216 | 0.7706 |
0.0107 | 7.24 | 1600 | 0.0211 | 0.7679 |
0.0099 | 8.14 | 1800 | 0.0210 | 0.7684 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2
- Downloads last month
- 4
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Model tree for Shijia/furina_amh_corr_5e-06
Base model
yihongLiu/furina