Downloads · 30 days
8
19% of all-time downloads
SecBERT-APTNER
Cyber-ThreaD/SecBERT-APTNER
SecBERT-APTNER is a token classification model from Cyber-ThreaD. Use it when you need labels on individual words, such as names. It is set up for transformers. The card lists the license as apache-2.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
43
Public
Parameters
82.9M
332 MB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors332 MB · 99%
At a glance
- Task
- Token Classification
- Library
- transformers
- License
- apache-2.0
- Model type
- bert
- Access
- Public
- Created
- Dec 6, 2023
- Updated
- Dec 6, 2023
- SHA
- 49fbb41a
Try a prompt
Base models
Model card
From the Hugging Face model README
<!-- 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. -->
aptner_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: 0.3230
- Precision: 0.5124
- Recall: 0.5356
- F1: 0.5237
- Accuracy: 0.9142
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 |
|---|---|---|---|---|---|---|---|
| 0.6662 | 0.59 | 500 | 0.3587 | 0.4744 | 0.4743 | 0.4744 | 0.9113 |
| 0.3128 | 1.19 | 1000 | 0.3230 | 0.5124 | 0.5356 | 0.5237 | 0.9142 |
| 0.2374 | 1.78 | 1500 | 0.3429 | 0.4750 | 0.5714 | 0.5188 | 0.9083 |
| 0.1904 | 2.37 | 2000 | 0.3650 | 0.4945 | 0.5598 | 0.5251 | 0.9090 |
| 0.1521 | 2.97 | 2500 | 0.3765 | 0.4713 | 0.5783 | 0.5193 | 0.9055 |
| 0.1101 | 3.56 | 3000 | 0.4023 | 0.4727 | 0.5744 | 0.5186 | 0.9067 |
| 0.1019 | 4.15 | 3500 | 0.4322 | 0.4726 | 0.5571 | 0.5114 | 0.9056 |
| 0.0764 | 4.74 | 4000 | 0.4595 | 0.4592 | 0.5897 | 0.5163 | 0.9039 |
| 0.0619 | 5.34 | 4500 | 0.4755 | 0.4740 | 0.5783 | 0.5210 | 0.9062 |
| 0.059 | 5.93 | 5000 | 0.4514 | 0.5055 | 0.5649 | 0.5335 | 0.9126 |
| 0.0429 | 6.52 | 5500 | 0.5036 | 0.474 | 0.5666 | 0.5162 | 0.9065 |
| 0.0425 | 7.12 | 6000 | 0.5249 | 0.4767 | 0.5726 | 0.5203 | 0.9064 |
| 0.0349 | 7.71 | 6500 | 0.5537 | 0.4634 | 0.5744 | 0.5129 | 0.9038 |
| 0.0338 | 8.3 | 7000 | 0.5301 | 0.4839 | 0.5672 | 0.5223 | 0.9089 |
| 0.0255 | 8.9 | 7500 | 0.5545 | 0.4731 | 0.5735 | 0.5185 | 0.9059 |
| 0.0253 | 9.49 | 8000 | 0.5526 | 0.4789 | 0.5702 | 0.5206 | 0.9074 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1
- Task
- Token Classification
- Library
- transformers
- Type
- bert
- License
- apache-2.0
- Created
- Dec 6, 2023
- Updated
- Dec 6, 2023