furina_seed42_eng_esp_hau_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: 0.0221
• Spearman Corr: 0.7567

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	1.45	200	0.0291	0.6127
0.0714	2.91	400	0.0240	0.7147
0.0265	4.36	600	0.0235	0.7297
0.0265	5.82	800	0.0266	0.7353
0.0199	7.27	1000	0.0225	0.7407
0.016	8.73	1200	0.0233	0.7521
0.0127	10.18	1400	0.0232	0.7541
0.0127	11.64	1600	0.0237	0.7530
0.0107	13.09	1800	0.0219	0.7558
0.0088	14.55	2000	0.0233	0.7486
0.0078	16.0	2200	0.0219	0.7540
0.0078	17.45	2400	0.0224	0.7545
0.007	18.91	2600	0.0230	0.7599
0.0062	20.36	2800	0.0217	0.7575
0.0062	21.82	3000	0.0225	0.7569
0.0058	23.27	3200	0.0220	0.7563
0.0054	24.73	3400	0.0226	0.7551
0.0052	26.18	3600	0.0223	0.7564
0.0052	27.64	3800	0.0222	0.7565
0.0049	29.09	4000	0.0221	0.7567

Framework versions

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