Edit model card

scibert-finetuned-ner

This model is a fine-tuned version of allenai/scibert_scivocab_uncased on the jnlpba dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4717
  • Precision: 0.6737
  • Recall: 0.7757
  • F1: 0.7211
  • Accuracy: 0.9226

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: 5

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
0.1608 1.0 2319 0.2431 0.6641 0.7581 0.7080 0.9250
0.103 2.0 4638 0.2916 0.6739 0.7803 0.7232 0.9228
0.0659 3.0 6957 0.3662 0.6796 0.7624 0.7186 0.9233
0.0393 4.0 9276 0.4222 0.6737 0.7771 0.7217 0.9225
0.025 5.0 11595 0.4717 0.6737 0.7757 0.7211 0.9226

Framework versions

  • Transformers 4.21.1
  • Pytorch 1.12.1+cu113
  • Datasets 2.4.0
  • Tokenizers 0.12.1
Downloads last month
15

Dataset used to train siddharthtumre/scibert-finetuned-ner

Evaluation results