SecBERT-CyNER / README.md
Anonymous
Upload folder using huggingface_hub
3566be9
metadata
license: apache-2.0
base_model: jackaduma/SecBERT
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: cyner_secbert
    results: []

cyner_secbert

This model is a fine-tuned version of jackaduma/SecBERT on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1212
  • Precision: 0.7047
  • Recall: 0.5517
  • F1: 0.6189
  • Accuracy: 0.9723

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
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10.0

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
0.165 1.42 500 0.1212 0.7047 0.5517 0.6189 0.9723
0.04 2.84 1000 0.1647 0.6924 0.5147 0.5905 0.9705
0.0156 4.26 1500 0.1803 0.6769 0.5351 0.5977 0.9714
0.0087 5.68 2000 0.1866 0.6574 0.5415 0.5938 0.9713
0.0036 7.1 2500 0.2020 0.6740 0.5492 0.6052 0.9719
0.0024 8.52 3000 0.2036 0.6697 0.5568 0.6081 0.9720
0.0018 9.94 3500 0.2084 0.6682 0.5504 0.6036 0.9715

Framework versions

  • Transformers 4.36.0.dev0
  • Pytorch 2.1.0+cu118
  • Datasets 2.14.6
  • Tokenizers 0.14.1