furina_seed42_eng_esp_kin_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: 0.0287
• Spearman Corr: 0.7243

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	1.63	200	0.0273	0.6199
0.0785	3.27	400	0.0237	0.6780
0.0231	4.9	600	0.0289	0.7095
0.0171	6.53	800	0.0221	0.7230
0.0135	8.16	1000	0.0317	0.7194
0.0135	9.8	1200	0.0266	0.7238
0.0108	11.43	1400	0.0290	0.7264
0.0089	13.06	1600	0.0279	0.7195
0.0075	14.69	1800	0.0317	0.7258
0.0065	16.33	2000	0.0295	0.7245
0.0065	17.96	2200	0.0271	0.7201
0.0058	19.59	2400	0.0287	0.7243

Framework versions

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