Edit model card

furina_pan_corr_2e-05

This model is a fine-tuned version of yihongLiu/furina on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0209
  • Spearman Corr: 0.7712

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: 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 0.85 200 0.0206 0.7703
No log 1.69 400 0.0207 0.7688
0.0004 2.54 600 0.0213 0.7701
0.0004 3.38 800 0.0208 0.7727
0.0003 4.23 1000 0.0209 0.7688
0.0003 5.07 1200 0.0208 0.7703
0.0003 5.92 1400 0.0212 0.7687
0.0002 6.77 1600 0.0216 0.7695
0.0002 7.61 1800 0.0209 0.7695
0.0002 8.46 2000 0.0202 0.7730
0.0002 9.3 2200 0.0203 0.7748
0.0003 10.15 2400 0.0203 0.7729
0.0003 10.99 2600 0.0212 0.7710
0.0003 11.84 2800 0.0208 0.7720
0.0004 12.68 3000 0.0204 0.7737
0.0004 13.53 3200 0.0208 0.7706
0.0004 14.38 3400 0.0207 0.7726
0.0004 15.22 3600 0.0208 0.7719
0.0004 16.07 3800 0.0209 0.7712

Framework versions

  • Transformers 4.37.2
  • Pytorch 2.2.0+cu121
  • Datasets 2.17.0
  • Tokenizers 0.15.2
Downloads last month
4
Safetensors
Model size
394M params
Tensor type
F32
·

Finetuned from