Modelpublic

ajrayman/Vulnerability_binary

sourceHugging Facemitupdated 19d agoView on Hugging Face

0likes149downloads

Model Card

Vulnerability_binary

This model is a fine-tuned version of microsoft/deberta-v3-base on the None dataset. It achieves the following results on the evaluation set:

—Loss: 1.0736
—Accuracy: 0.6712
—Precision: 0.6927
—Recall: 0.6169
—F1: 0.6526
—Auc: 0.7228

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
—trainbatchsize: 32
—evalbatchsize: 32
—seed: 1234
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—lrschedulerwarmup_ratio: 0.06
—num_epochs: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Auc
No log	1.0	118	0.6261	0.6687	0.7024	0.5871	0.6396	0.7236
No log	2.0	236	0.5943	0.6936	0.7549	0.5746	0.6525	0.7619
No log	3.0	354	0.6206	0.6936	0.7042	0.6692	0.6862	0.7537
No log	4.0	472	0.6992	0.6899	0.7192	0.6244	0.6684	0.7458
0.5353	5.0	590	0.8100	0.6849	0.6822	0.6940	0.6880	0.7303
0.5353	6.0	708	0.9372	0.6712	0.6835	0.6393	0.6607	0.7217
0.5353	7.0	826	1.0736	0.6712	0.6927	0.6169	0.6526	0.7228

Framework versions

—Transformers 4.44.1
—Pytorch 1.11.0
—Datasets 2.12.0
—Tokenizers 0.19.1