BioLinkBERT-Large-LitCovid-1.4

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

BioLinkBERT-Large-LitCovid-1.4

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

• Loss: 0.5976
• Hamming loss: 0.0604
• F1 micro: 0.6804
• F1 macro: 0.5425
• F1 weighted: 0.7357
• F1 samples: 0.6807
• Precision micro: 0.5509
• Precision macro: 0.4271
• Precision weighted: 0.6552
• Precision samples: 0.5921
• Recall micro: 0.8895
• Recall macro: 0.8221
• Recall weighted: 0.8895
• Recall samples: 0.9063
• Roc Auc: 0.9165
• Accuracy: 0.1370

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Framework versions

• Transformers 4.28.0
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.13.3