File size: 3,392 Bytes
b4af5c1 1022b9a b4af5c1 1022b9a b4af5c1 1022b9a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 |
---
license: bigscience-openrail-m
base_model: ehsanaghaei/SecureBERT
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: Cyber-ThreaD/SecureBERT-DNRTI
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# Cyber-ThreaD/SecureBERT-DNRTI
This model is a fine-tuned version of [ehsanaghaei/SecureBERT](https://huggingface.co/ehsanaghaei/SecureBERT) on the [DNRTI](https://github.com/SCreaMxp/DNRTI-A-Large-scale-Dataset-for-Named-Entity-Recognition-in-Threat-Intelligence) dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2427
- Precision: 0.7694
- Recall: 0.7854
- F1: 0.7773
- Accuracy: 0.9382
It achieves the following results on the prediction set:
- Precision: 0.8346
- Recall: 0.8403
- F1: 0.8374
- Accuracy: 0.9554
## 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 |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
| 0.7821 | 0.76 | 500 | 0.4215 | 0.5219 | 0.5919 | 0.5547 | 0.8745 |
| 0.3559 | 1.52 | 1000 | 0.3152 | 0.6272 | 0.6587 | 0.6426 | 0.9008 |
| 0.2807 | 2.28 | 1500 | 0.2952 | 0.6445 | 0.7232 | 0.6816 | 0.9084 |
| 0.2272 | 3.04 | 2000 | 0.2793 | 0.6682 | 0.7513 | 0.7073 | 0.9161 |
| 0.1837 | 3.81 | 2500 | 0.2489 | 0.7151 | 0.7526 | 0.7334 | 0.9258 |
| 0.1497 | 4.57 | 3000 | 0.2511 | 0.7254 | 0.7826 | 0.7529 | 0.9286 |
| 0.1371 | 5.33 | 3500 | 0.2496 | 0.7425 | 0.7757 | 0.7587 | 0.9331 |
| 0.1135 | 6.09 | 4000 | 0.2554 | 0.7289 | 0.8075 | 0.7662 | 0.9325 |
| 0.1018 | 6.85 | 4500 | 0.2427 | 0.7694 | 0.7854 | 0.7773 | 0.9382 |
| 0.0899 | 7.61 | 5000 | 0.2516 | 0.7583 | 0.8167 | 0.7864 | 0.9378 |
| 0.0809 | 8.37 | 5500 | 0.2459 | 0.7717 | 0.8176 | 0.7940 | 0.9406 |
| 0.0763 | 9.13 | 6000 | 0.2553 | 0.7518 | 0.8217 | 0.7852 | 0.9392 |
| 0.0687 | 9.89 | 6500 | 0.2534 | 0.7621 | 0.8204 | 0.7902 | 0.9407 |
### 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}
}
```
|