--- library_name: transformers base_model: dmis-lab/biobert-base-cased-v1.1 tags: - generated_from_trainer metrics: - accuracy - precision - recall - f1 model-index: - name: Biobert_combo_v1 results: [] --- # Biobert_combo_v1 This model is a fine-tuned version of [dmis-lab/biobert-base-cased-v1.1](https://huggingface.co/dmis-lab/biobert-base-cased-v1.1) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.4421 - Accuracy: 0.769 - Auc: 0.867 - Precision: 0.745 - Recall: 0.887 - F1: 0.81 - F1-macro: 0.757 - F1-micro: 0.769 - F1-weighted: 0.763 ## 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: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - num_epochs: 5 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | Auc | Precision | Recall | F1 | F1-macro | F1-micro | F1-weighted | |:-------------:|:------:|:-----:|:---------------:|:--------:|:-----:|:---------:|:------:|:-----:|:--------:|:--------:|:-----------:| | 0.6198 | 0.1028 | 500 | 0.5619 | 0.693 | 0.769 | 0.676 | 0.858 | 0.757 | 0.671 | 0.693 | 0.681 | | 0.5401 | 0.2057 | 1000 | 0.5167 | 0.717 | 0.807 | 0.717 | 0.81 | 0.761 | 0.708 | 0.717 | 0.713 | | 0.5141 | 0.3085 | 1500 | 0.4933 | 0.731 | 0.827 | 0.75 | 0.772 | 0.761 | 0.726 | 0.731 | 0.73 | | 0.4886 | 0.4114 | 2000 | 0.4942 | 0.736 | 0.827 | 0.727 | 0.841 | 0.78 | 0.726 | 0.736 | 0.732 | | 0.4882 | 0.5142 | 2500 | 0.4814 | 0.743 | 0.838 | 0.712 | 0.901 | 0.795 | 0.724 | 0.743 | 0.732 | | 0.4734 | 0.6170 | 3000 | 0.4718 | 0.746 | 0.843 | 0.724 | 0.878 | 0.794 | 0.733 | 0.746 | 0.739 | | 0.4659 | 0.7199 | 3500 | 0.4767 | 0.748 | 0.843 | 0.733 | 0.859 | 0.791 | 0.737 | 0.748 | 0.743 | | 0.4632 | 0.8227 | 4000 | 0.4617 | 0.755 | 0.852 | 0.724 | 0.901 | 0.803 | 0.739 | 0.755 | 0.746 | | 0.4602 | 0.9255 | 4500 | 0.4639 | 0.752 | 0.852 | 0.717 | 0.915 | 0.804 | 0.734 | 0.752 | 0.742 | | 0.457 | 1.0284 | 5000 | 0.4613 | 0.752 | 0.85 | 0.729 | 0.881 | 0.798 | 0.739 | 0.752 | 0.745 | | 0.4468 | 1.1312 | 5500 | 0.4541 | 0.757 | 0.855 | 0.731 | 0.891 | 0.803 | 0.743 | 0.757 | 0.75 | | 0.4421 | 1.2341 | 6000 | 0.4591 | 0.755 | 0.853 | 0.727 | 0.892 | 0.801 | 0.74 | 0.755 | 0.747 | | 0.4373 | 1.3369 | 6500 | 0.4537 | 0.759 | 0.856 | 0.739 | 0.874 | 0.801 | 0.748 | 0.759 | 0.753 | | 0.4402 | 1.4397 | 7000 | 0.4552 | 0.755 | 0.855 | 0.74 | 0.863 | 0.797 | 0.745 | 0.755 | 0.751 | | 0.4296 | 1.5426 | 7500 | 0.4545 | 0.76 | 0.857 | 0.742 | 0.87 | 0.801 | 0.749 | 0.76 | 0.755 | | 0.4407 | 1.6454 | 8000 | 0.4458 | 0.762 | 0.86 | 0.742 | 0.877 | 0.804 | 0.751 | 0.762 | 0.757 | | 0.4225 | 1.7483 | 8500 | 0.4472 | 0.761 | 0.86 | 0.735 | 0.889 | 0.805 | 0.748 | 0.761 | 0.754 | | 0.4327 | 1.8511 | 9000 | 0.4485 | 0.758 | 0.858 | 0.741 | 0.867 | 0.799 | 0.747 | 0.758 | 0.753 | | 0.4311 | 1.9539 | 9500 | 0.4479 | 0.76 | 0.859 | 0.742 | 0.869 | 0.801 | 0.749 | 0.76 | 0.755 | | 0.4288 | 2.0568 | 10000 | 0.4527 | 0.761 | 0.859 | 0.742 | 0.873 | 0.802 | 0.75 | 0.761 | 0.756 | | 0.4124 | 2.1596 | 10500 | 0.4477 | 0.762 | 0.861 | 0.736 | 0.891 | 0.806 | 0.749 | 0.762 | 0.756 | | 0.4181 | 2.2624 | 11000 | 0.4569 | 0.759 | 0.857 | 0.741 | 0.87 | 0.8 | 0.748 | 0.759 | 0.754 | | 0.4178 | 2.3653 | 11500 | 0.4469 | 0.762 | 0.861 | 0.741 | 0.879 | 0.804 | 0.751 | 0.762 | 0.757 | | 0.4127 | 2.4681 | 12000 | 0.4448 | 0.764 | 0.863 | 0.742 | 0.881 | 0.806 | 0.753 | 0.764 | 0.759 | | 0.419 | 2.5710 | 12500 | 0.4454 | 0.764 | 0.864 | 0.734 | 0.902 | 0.809 | 0.75 | 0.764 | 0.757 | | 0.4232 | 2.6738 | 13000 | 0.4394 | 0.766 | 0.864 | 0.747 | 0.873 | 0.805 | 0.756 | 0.766 | 0.761 | | 0.4226 | 2.7766 | 13500 | 0.4404 | 0.766 | 0.864 | 0.747 | 0.873 | 0.805 | 0.756 | 0.766 | 0.761 | | 0.4196 | 2.8795 | 14000 | 0.4477 | 0.765 | 0.862 | 0.758 | 0.846 | 0.8 | 0.757 | 0.765 | 0.762 | | 0.408 | 2.9823 | 14500 | 0.4497 | 0.763 | 0.862 | 0.745 | 0.871 | 0.803 | 0.752 | 0.763 | 0.758 | | 0.4054 | 3.0852 | 15000 | 0.4404 | 0.765 | 0.865 | 0.749 | 0.865 | 0.803 | 0.755 | 0.765 | 0.76 | | 0.4155 | 3.1880 | 15500 | 0.4466 | 0.764 | 0.863 | 0.74 | 0.885 | 0.806 | 0.752 | 0.764 | 0.758 | | 0.4155 | 3.2908 | 16000 | 0.4417 | 0.765 | 0.864 | 0.744 | 0.879 | 0.806 | 0.754 | 0.765 | 0.76 | | 0.4104 | 3.3937 | 16500 | 0.4407 | 0.766 | 0.866 | 0.742 | 0.887 | 0.808 | 0.755 | 0.766 | 0.761 | | 0.4081 | 3.4965 | 17000 | 0.4406 | 0.765 | 0.866 | 0.753 | 0.859 | 0.802 | 0.756 | 0.765 | 0.762 | | 0.4046 | 3.5993 | 17500 | 0.4384 | 0.768 | 0.868 | 0.742 | 0.891 | 0.81 | 0.756 | 0.768 | 0.762 | | 0.4065 | 3.7022 | 18000 | 0.4443 | 0.766 | 0.866 | 0.742 | 0.888 | 0.808 | 0.754 | 0.766 | 0.76 | | 0.4028 | 3.8050 | 18500 | 0.4438 | 0.768 | 0.866 | 0.746 | 0.882 | 0.808 | 0.757 | 0.768 | 0.763 | | 0.4035 | 3.9079 | 19000 | 0.4453 | 0.766 | 0.865 | 0.753 | 0.862 | 0.804 | 0.758 | 0.766 | 0.763 | | 0.4003 | 4.0107 | 19500 | 0.4426 | 0.767 | 0.865 | 0.75 | 0.87 | 0.806 | 0.757 | 0.767 | 0.763 | | 0.4011 | 4.1135 | 20000 | 0.4423 | 0.767 | 0.867 | 0.741 | 0.892 | 0.81 | 0.755 | 0.767 | 0.761 | | 0.3924 | 4.2164 | 20500 | 0.4394 | 0.768 | 0.867 | 0.75 | 0.874 | 0.807 | 0.758 | 0.768 | 0.764 | | 0.4043 | 4.3192 | 21000 | 0.4421 | 0.768 | 0.867 | 0.744 | 0.887 | 0.809 | 0.757 | 0.768 | 0.762 | | 0.395 | 4.4220 | 21500 | 0.4450 | 0.767 | 0.865 | 0.75 | 0.87 | 0.805 | 0.757 | 0.767 | 0.763 | | 0.4072 | 4.5249 | 22000 | 0.4392 | 0.769 | 0.867 | 0.751 | 0.874 | 0.808 | 0.759 | 0.769 | 0.765 | | 0.4055 | 4.6277 | 22500 | 0.4439 | 0.768 | 0.866 | 0.745 | 0.885 | 0.809 | 0.757 | 0.768 | 0.762 | | 0.3987 | 4.7306 | 23000 | 0.4435 | 0.768 | 0.866 | 0.747 | 0.881 | 0.808 | 0.758 | 0.768 | 0.763 | | 0.396 | 4.8334 | 23500 | 0.4430 | 0.768 | 0.867 | 0.745 | 0.885 | 0.809 | 0.757 | 0.768 | 0.763 | | 0.4017 | 4.9362 | 24000 | 0.4421 | 0.769 | 0.867 | 0.745 | 0.887 | 0.81 | 0.757 | 0.769 | 0.763 | ### Framework versions - Transformers 4.53.1 - Pytorch 2.6.0+cu124 - Datasets 2.14.4 - Tokenizers 0.21.2