Shijia's picture
End of training
d7e7fe3 verified
metadata
base_model: yihongLiu/furina
tags:
  - generated_from_trainer
model-index:
  - name: furina_hin_corr_2e-05
    results: []

furina_hin_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: 0.0201
  • Spearman Corr: 0.7767

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 0.0199 0.7732
No log 1.69 400 0.0206 0.7716
0.0008 2.54 600 0.0206 0.7734
0.0008 3.38 800 0.0204 0.7767
0.0007 4.23 1000 0.0214 0.7723
0.0007 5.07 1200 0.0206 0.7760
0.0007 5.92 1400 0.0205 0.7718
0.0007 6.77 1600 0.0208 0.7735
0.0007 7.61 1800 0.0200 0.7767
0.0006 8.46 2000 0.0204 0.7749
0.0006 9.3 2200 0.0209 0.7749
0.0007 10.15 2400 0.0201 0.7767

Framework versions

  • Transformers 4.37.2
  • Pytorch 2.2.0+cu121
  • Datasets 2.17.0
  • Tokenizers 0.15.2