furina_seed42_eng_basic_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: 0.0170
• Spearman Corr: 0.7981

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
0.0996	2.33	200	0.0152	0.7264
0.0264	4.65	400	0.0147	0.7717
0.0174	6.98	600	0.0151	0.7851
0.0138	9.3	800	0.0145	0.7920
0.0115	11.63	1000	0.0164	0.7917
0.0083	13.95	1200	0.0163	0.7924
0.0077	16.28	1400	0.0168	0.7967
0.0067	18.6	1600	0.0171	0.7994
0.0062	20.93	1800	0.0170	0.7984
0.0058	23.26	2000	0.0175	0.8021
0.0057	25.58	2200	0.0168	0.7985
0.0052	27.91	2400	0.0170	0.7981

Framework versions

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