metadata

license: apache-2.0
base_model: jackaduma/SecBERT
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: dnrti_secbert
    results: []

dnrti_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: 0.2274
Precision: 0.7405
Recall: 0.7780
F1: 0.7588
Accuracy: 0.9389

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
0.6137	0.76	500	0.3731	0.5348	0.5951	0.5633	0.8842
0.2993	1.52	1000	0.2853	0.6684	0.6665	0.6674	0.9131
0.2157	2.28	1500	0.2624	0.6685	0.7282	0.6971	0.9212
0.152	3.04	2000	0.2414	0.6923	0.7619	0.7254	0.9308
0.1047	3.81	2500	0.2274	0.7405	0.7780	0.7588	0.9389
0.0725	4.57	3000	0.2563	0.7262	0.7964	0.7597	0.9370
0.0589	5.33	3500	0.2615	0.7489	0.8024	0.7747	0.9411
0.0442	6.09	4000	0.2638	0.7543	0.8061	0.7793	0.9434
0.0344	6.85	4500	0.2671	0.7635	0.8088	0.7855	0.9448
0.0282	7.61	5000	0.2861	0.7584	0.8111	0.7839	0.9439
0.0226	8.37	5500	0.2849	0.7693	0.8093	0.7888	0.9456
0.0207	9.13	6000	0.2932	0.7643	0.8185	0.7905	0.9456
0.0181	9.89	6500	0.2952	0.7665	0.8167	0.7908	0.9459

Framework versions

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