furina_seed42_eng_kin_hau_cross_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: 0.0268
• Spearman Corr: 0.7372

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	0.53	200	0.0336	0.6022
No log	1.06	400	0.0303	0.6548
No log	1.6	600	0.0329	0.6851
0.0491	2.13	800	0.0288	0.7186
0.0491	2.66	1000	0.0258	0.7170
0.0491	3.19	1200	0.0272	0.7286
0.0491	3.72	1400	0.0285	0.7289
0.0229	4.26	1600	0.0264	0.7193
0.0229	4.79	1800	0.0303	0.7334
0.0229	5.32	2000	0.0257	0.7393
0.0229	5.85	2200	0.0260	0.7466
0.0159	6.38	2400	0.0251	0.7402
0.0159	6.91	2600	0.0256	0.7396
0.0159	7.45	2800	0.0266	0.7453
0.0159	7.98	3000	0.0268	0.7395
0.0114	8.51	3200	0.0266	0.7433
0.0114	9.04	3400	0.0261	0.7459
0.0114	9.57	3600	0.0260	0.7410
0.0087	10.11	3800	0.0274	0.7428
0.0087	10.64	4000	0.0268	0.7372

Framework versions

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