metadata

base_model: SynamicTechnologies/CYBERT
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: Cyber-ThreaD/CyBERT-CyNER
    results: []

Cyber-ThreaD/CyBERT-CyNER

This model is a fine-tuned version of SynamicTechnologies/CYBERT on the CyNER dataset. It achieves the following results on the evaluation set:

Loss: 0.2405
Precision: 0.4671
Recall: 0.2810
F1: 0.3509
Accuracy: 0.9568

It achieves the following results on the prediction set:

Loss: 0.2747
Precision: 0.5442
Recall: 0.3483
F1: 0.4248
Accuracy: 0.9471

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
0.2304	1.42	500	0.2405	0.4671	0.2810	0.3509	0.9568
0.1092	2.84	1000	0.2575	0.5426	0.2848	0.3735	0.9601
0.0797	4.26	1500	0.2454	0.4701	0.3308	0.3883	0.9576
0.0615	5.68	2000	0.2669	0.4902	0.3180	0.3857	0.9586
0.0504	7.1	2500	0.2687	0.4885	0.3525	0.4095	0.9580
0.0379	8.52	3000	0.2752	0.4656	0.3627	0.4078	0.9573
0.0339	9.94	3500	0.2828	0.4991	0.3499	0.4114	0.9586

Framework versions

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

Citing & Authors

If you use the model kindly cite the following work

@inproceedings{deka2024attacker,
  title={AttackER: Towards Enhancing Cyber-Attack Attribution with a Named Entity Recognition Dataset},
  author={Deka, Pritam and Rajapaksha, Sampath and Rani, Ruby and Almutairi, Amirah and Karafili, Erisa},
  booktitle={International Conference on Web Information Systems Engineering},
  pages={255--270},
  year={2024},
  organization={Springer}
}