TAPT_data-V2_Bioformer-16L_LR-0.0001

This model is a fine-tuned version of bioformers/bioformer-16L on the Mardiyyah/TAPT_data_V2_split dataset. It achieves the following results on the evaluation set:

• Loss: 2.3000

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: 16
• eval_batch_size: 16
• seed: 3407
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-06 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.06
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss
2.1948	1.0	609	2.2790
1.9627	2.0	1218	2.3013
1.7865	3.0	1827	2.3833
1.6375	4.0	2436	2.3146
1.5212	5.0	3045	2.3126
1.4298	6.0	3654	2.3001
1.361	7.0	4263	2.3267
1.2688	8.0	4872	2.3044
1.2074	9.0	5481	2.2634
1.1643	10.0	6090	2.2897

Framework versions

• Transformers 4.48.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.2
• Tokenizers 0.21.0