furina_amh_loss_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: 0.0237
• Spearman Corr: 0.7697

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	0.9	200	0.0232	0.7664
No log	1.81	400	0.0228	0.7676
0.0039	2.71	600	0.0255	0.7675
0.0039	3.62	800	0.0235	0.7673
0.0033	4.52	1000	0.0224	0.7712
0.0033	5.43	1200	0.0243	0.7666
0.0039	6.33	1400	0.0223	0.7700
0.0039	7.24	1600	0.0244	0.7631
0.0036	8.14	1800	0.0234	0.7705
0.0036	9.05	2000	0.0224	0.7680
0.0036	9.95	2200	0.0227	0.7673
0.0032	10.86	2400	0.0225	0.7680
0.0032	11.76	2600	0.0242	0.7665
0.0029	12.67	2800	0.0233	0.7671
0.0029	13.57	3000	0.0214	0.7683
0.0027	14.48	3200	0.0217	0.7705
0.0027	15.38	3400	0.0233	0.7675
0.0025	16.29	3600	0.0239	0.7683
0.0025	17.19	3800	0.0231	0.7678
0.0023	18.1	4000	0.0234	0.7692
0.0023	19.0	4200	0.0227	0.7674
0.0023	19.91	4400	0.0230	0.7687
0.0022	20.81	4600	0.0237	0.7697

Framework versions

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