furina_seed42_eng_esp_kin_basic

This model is a fine-tuned version of yihongLiu/furina on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0183
• Spearman Corr: 0.7765

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	1.81	200	0.0242	0.6836
0.0889	3.62	400	0.0204	0.7530
0.025	5.43	600	0.0197	0.7779
0.019	7.24	800	0.0179	0.7805
0.0153	9.05	1000	0.0177	0.7877
0.0121	10.86	1200	0.0170	0.7842
0.0102	12.67	1400	0.0175	0.7797
0.0087	14.48	1600	0.0177	0.7753
0.0077	16.29	1800	0.0172	0.7800
0.007	18.1	2000	0.0175	0.7792
0.0063	19.91	2200	0.0176	0.7791
0.0063	21.72	2400	0.0175	0.7754
0.0057	23.53	2600	0.0175	0.7791
0.0055	25.34	2800	0.0183	0.7765

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.1