furina_hin_loss_0.0001

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

• Loss: 0.0480
• Spearman Corr: nan

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: 0.0001
• 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	0.85	200	0.0476	nan
No log	1.69	400	0.0476	nan
0.0476	2.54	600	0.0481	nan
0.0476	3.38	800	0.0470	nan
0.0478	4.23	1000	0.0473	nan
0.0478	5.07	1200	0.0474	0.0168
0.0478	5.92	1400	0.0478	-0.0241
0.0479	6.77	1600	0.0478	nan
0.0479	7.61	1800	0.0466	nan
0.048	8.46	2000	0.0480	nan
0.048	9.3	2200	0.0469	nan
0.0478	10.15	2400	0.0491	nan
0.0478	10.99	2600	0.0468	nan
0.0478	11.84	2800	0.0486	nan
0.0476	12.68	3000	0.0470	nan
0.0476	13.53	3200	0.0487	nan
0.0478	14.38	3400	0.0480	nan

Framework versions

• Transformers 4.37.2
• Pytorch 2.2.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.2