furina_original_kin-amh-eng_train_spearman_corr

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

• Loss: 0.0232
• Spearman Corr: 0.7603

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	1.75	200	0.0327	0.6413
0.0853	3.51	400	0.0219	0.7158
0.0217	5.26	600	0.0220	0.7493
0.0159	7.02	800	0.0199	0.7703
0.0122	8.77	1000	0.0192	0.7628
0.0101	10.53	1200	0.0216	0.7542
0.0085	12.28	1400	0.0206	0.7665
0.0075	14.04	1600	0.0214	0.7578
0.0075	15.79	1800	0.0215	0.7601
0.0065	17.54	2000	0.0213	0.7618
0.0059	19.3	2200	0.0208	0.7621
0.0055	21.05	2400	0.0232	0.7603

Framework versions

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