furina_hau_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: 0.7736

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	0.95	200	0.0214	0.7723
No log	1.91	400	0.0225	0.7716
0.0012	2.86	600	0.0211	0.7695
0.0012	3.82	800	0.0207	0.7718
0.0011	4.77	1000	0.0214	0.7723
0.0011	5.73	1200	0.0209	0.7753
0.001	6.68	1400	0.0210	0.7710
0.001	7.64	1600	0.0204	0.7721
0.0009	8.59	1800	0.0217	0.7731
0.0009	9.55	2000	0.0216	0.7692
0.0009	10.5	2200	0.0206	0.7724
0.0009	11.46	2400	0.0213	0.7734
0.0009	12.41	2600	0.0208	0.7725
0.0009	13.37	2800	0.0207	0.7760
0.0008	14.32	3000	0.0209	0.7724
0.0008	15.27	3200	0.0208	0.7729
0.0007	16.23	3400	0.0212	0.7732
0.0007	17.18	3600	0.0209	0.7746
0.0007	18.14	3800	0.0209	0.7745
0.0007	19.09	4000	0.0202	0.7759
0.0007	20.05	4200	0.0206	0.7750
0.0007	21.0	4400	0.0209	0.7736

Framework versions

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