Edit model card

biobert-base-cased-v1.2-bc2gm-ner

This model is a fine-tuned version of dmis-lab/biobert-base-cased-v1.2 on the bc2gm_corpus dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1528
  • Precision: 0.7988
  • Recall: 0.8243
  • F1: 0.8114
  • Accuracy: 0.9772

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: 10

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
0.057 1.0 782 0.0670 0.7446 0.8051 0.7736 0.9738
0.0586 2.0 1564 0.0689 0.7689 0.8106 0.7892 0.9755
0.0123 3.0 2346 0.0715 0.7846 0.8076 0.7959 0.9750
0.0002 4.0 3128 0.0896 0.7942 0.8199 0.8068 0.9767
0.0004 5.0 3910 0.1119 0.7971 0.8201 0.8084 0.9765
0.0004 6.0 4692 0.1192 0.7966 0.8337 0.8147 0.9768
0.013 7.0 5474 0.1274 0.7932 0.8266 0.8095 0.9773
0.0236 8.0 6256 0.1419 0.7976 0.8213 0.8093 0.9771
0.0004 9.0 7038 0.1519 0.8004 0.8261 0.8130 0.9772
0.0 10.0 7820 0.1528 0.7988 0.8243 0.8114 0.9772

Framework versions

  • Transformers 4.23.1
  • Pytorch 1.12.1+cu113
  • Datasets 2.6.1
  • Tokenizers 0.13.1
Downloads last month
5
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Dataset used to train chintagunta85/biobert-base-cased-v1.2-bc2gm-ner

Evaluation results