judithrosell's picture
End of training
750d654
metadata
license: mit
base_model: microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: PubMedBERT_BioNLP13CG_NER_new
    results: []

PubMedBERT_BioNLP13CG_NER_new

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: 0.1724
  • Precision: 0.8806
  • Recall: 0.8773
  • F1: 0.8789
  • Accuracy: 0.9595

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: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
No log 1.0 191 0.2269 0.8404 0.8521 0.8462 0.9468
No log 2.0 382 0.1772 0.8728 0.8710 0.8719 0.9574
0.362 3.0 573 0.1724 0.8806 0.8773 0.8789 0.9595

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.1.0+cu121
  • Datasets 2.16.1
  • Tokenizers 0.15.0