ontochem_biobert_half

This model is a fine-tuned version of dmis-lab/biobert-base-cased-v1.1 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0778

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: 1e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 14

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	24	0.9696
No log	2.0	48	0.8620
No log	3.0	72	0.6842
No log	4.0	96	0.4193
No log	5.0	120	0.1765
No log	6.0	144	0.1210
No log	7.0	168	0.0996
No log	8.0	192	0.0849
No log	9.0	216	0.0770
No log	10.0	240	0.0739
No log	11.0	264	0.0739
No log	12.0	288	0.0731
No log	13.0	312	0.0751
No log	14.0	336	0.0778

Framework versions

• Transformers 4.39.3
• Pytorch 2.2.2+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2