furina_arq_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: 0.0214
• Spearman Corr: 0.7725

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	0.92	200	0.0215	0.7689
No log	1.84	400	0.0221	0.7652
0.0024	2.76	600	0.0215	0.7726
0.0024	3.69	800	0.0223	0.7706
0.0022	4.61	1000	0.0218	0.7726
0.0022	5.53	1200	0.0219	0.7717
0.002	6.45	1400	0.0211	0.7684
0.002	7.37	1600	0.0216	0.7698
0.0017	8.29	1800	0.0222	0.7721
0.0017	9.22	2000	0.0228	0.7731
0.0017	10.14	2200	0.0217	0.7710
0.0017	11.06	2400	0.0221	0.7739
0.0016	11.98	2600	0.0220	0.7708
0.0016	12.9	2800	0.0224	0.7732
0.0016	13.82	3000	0.0216	0.7750
0.0015	14.75	3200	0.0215	0.7774
0.0015	15.67	3400	0.0219	0.7754
0.0014	16.59	3600	0.0222	0.7751
0.0014	17.51	3800	0.0216	0.7751
0.0013	18.43	4000	0.0209	0.7755
0.0013	19.35	4200	0.0217	0.7728
0.0013	20.28	4400	0.0210	0.7731
0.0013	21.2	4600	0.0212	0.7737
0.0012	22.12	4800	0.0214	0.7725

Framework versions

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