Shijia's picture
End of training
27fb92e verified
metadata
base_model: yihongLiu/furina
tags:
  - generated_from_trainer
model-index:
  - name: furina_seed42_eng_esp_hau_basic_0.0001
    results: []

furina_seed42_eng_esp_hau_basic_0.0001

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

  • Loss: 0.0312
  • Spearman Corr: 0.7194

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: 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 1.45 200 0.0282 0.6099
0.0652 2.91 400 0.0303 0.6702
0.0263 4.36 600 0.0301 0.7175
0.0263 5.82 800 0.0333 0.7147
0.0178 7.27 1000 0.0346 0.7279
0.0123 8.73 1200 0.0274 0.7307
0.0089 10.18 1400 0.0290 0.7331
0.0089 11.64 1600 0.0302 0.7243
0.0064 13.09 1800 0.0292 0.7243
0.0052 14.55 2000 0.0297 0.7253
0.0044 16.0 2200 0.0335 0.7158
0.0044 17.45 2400 0.0313 0.7319
0.0038 18.91 2600 0.0289 0.7169
0.0032 20.36 2800 0.0312 0.7194

Framework versions

  • Transformers 4.37.2
  • Pytorch 2.1.0+cu121
  • Datasets 2.17.0
  • Tokenizers 0.15.2