furina_seed42_eng_kin_amh_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: 0.0229
• Spearman Corr: 0.7406

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	1.75	200	0.0718	0.0554
0.1419	3.51	400	0.0366	0.4543
0.0605	5.26	600	0.0293	0.6111
0.0327	7.02	800	0.0270	0.6638
0.0263	8.77	1000	0.0265	0.6874
0.0228	10.53	1200	0.0254	0.6896
0.0204	12.28	1400	0.0240	0.7026
0.0188	14.04	1600	0.0228	0.7227
0.0188	15.79	1800	0.0248	0.7165
0.0176	17.54	2000	0.0249	0.7234
0.0169	19.3	2200	0.0231	0.7327
0.0163	21.05	2400	0.0228	0.7329
0.0153	22.81	2600	0.0227	0.7384
0.0146	24.56	2800	0.0227	0.7387
0.0144	26.32	3000	0.0233	0.7432
0.0139	28.07	3200	0.0229	0.7406

Framework versions

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