furina_seed42_eng_kin_hau_basic_5e-06

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

• Loss: 0.0266
• Spearman Corr: 0.7441

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: 5e-06
• 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	1.59	200	0.0747	0.1638
0.1417	3.19	400	0.0459	0.4336
0.0661	4.78	600	0.0364	0.6237
0.0367	6.37	800	0.0314	0.6619
0.0293	7.97	1000	0.0302	0.6957
0.0293	9.56	1200	0.0298	0.7101
0.0253	11.16	1400	0.0286	0.7235
0.0229	12.75	1600	0.0276	0.7291
0.0212	14.34	1800	0.0271	0.7317
0.02	15.94	2000	0.0263	0.7372
0.02	17.53	2200	0.0279	0.7384
0.0188	19.12	2400	0.0279	0.7397
0.018	20.72	2600	0.0274	0.7390
0.0172	22.31	2800	0.0257	0.7430
0.0165	23.9	3000	0.0270	0.7419
0.0165	25.5	3200	0.0270	0.7409
0.0164	27.09	3400	0.0270	0.7449
0.0155	28.69	3600	0.0266	0.7441

Framework versions

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