Edit model card

tmvar_5e-05_0404_ES6_strict_tok

This model is a fine-tuned version of microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0372
  • Precision: 0.7742
  • Recall: 0.8528
  • F1: 0.8116
  • Accuracy: 0.9906

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: 5e-05
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • training_steps: 2000

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
0.3642 0.49 25 0.0757 0.0 0.0 0.0 0.9727
0.0672 0.98 50 0.0660 0.6397 0.4416 0.5225 0.9841
0.0347 1.47 75 0.0357 0.7129 0.7310 0.7218 0.9888
0.0292 1.96 100 0.0255 0.7630 0.8173 0.7892 0.9903
0.012 2.45 125 0.0325 0.6923 0.8223 0.7517 0.9903
0.0087 2.94 150 0.0372 0.7742 0.8528 0.8116 0.9906

Framework versions

  • Transformers 4.27.4
  • Pytorch 2.0.0+cu118
  • Datasets 2.11.0
  • Tokenizers 0.13.3
Downloads last month
8