cwe-parent-vulnerability-classification-roberta-base-roberta-base

This model is a fine-tuned version of roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.6490
• Accuracy: 0.6452
• F1 Macro: 0.4825

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: 1e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Macro
3.1744	1.0	234	2.9202	0.1655	0.0682
2.3863	2.0	468	2.2049	0.4179	0.2748
1.9917	3.0	702	1.9277	0.5131	0.3250
1.617	4.0	936	1.7494	0.5762	0.3687
1.27	5.0	1170	1.7202	0.5798	0.3744
1.1845	6.0	1404	1.7314	0.6024	0.4028
1.1197	7.0	1638	1.6490	0.6452	0.4825
1.0453	8.0	1872	1.7107	0.6381	0.4659
0.6282	9.0	2106	1.6758	0.6536	0.5108
0.7391	10.0	2340	1.6929	0.6452	0.5256
0.5555	11.0	2574	1.7681	0.6762	0.5248
0.4857	12.0	2808	1.8233	0.6940	0.5267
0.4891	13.0	3042	1.9212	0.7131	0.5488
0.3272	14.0	3276	2.0065	0.7202	0.5296
0.2221	15.0	3510	1.9993	0.7190	0.5335
0.2426	16.0	3744	2.0301	0.7048	0.5495
0.1999	17.0	3978	2.1874	0.6833	0.5283
0.131	18.0	4212	2.2069	0.7345	0.5826
0.1219	19.0	4446	2.2270	0.7036	0.5364
0.0942	20.0	4680	2.4053	0.7083	0.5590
0.114	21.0	4914	2.4296	0.7333	0.5790
0.0691	22.0	5148	2.5488	0.7381	0.5546
0.0575	23.0	5382	2.4609	0.7274	0.5631
0.0665	24.0	5616	2.6766	0.7440	0.5625
0.0386	25.0	5850	2.7689	0.7440	0.5480
0.0688	26.0	6084	2.7388	0.7107	0.5382
0.0522	27.0	6318	2.9133	0.7464	0.5615
0.0559	28.0	6552	2.9099	0.7452	0.5591
0.0303	29.0	6786	2.9052	0.7595	0.5707
0.0277	30.0	7020	3.0239	0.75	0.5754
0.0292	31.0	7254	3.1133	0.7440	0.5590
0.013	32.0	7488	3.1130	0.7536	0.5769
0.0095	33.0	7722	3.2587	0.7429	0.5691
0.0303	34.0	7956	3.2025	0.7536	0.5728
0.0199	35.0	8190	3.1846	0.7512	0.5651
0.0106	36.0	8424	3.1951	0.7488	0.5478
0.0149	37.0	8658	3.2673	0.7512	0.5680
0.01	38.0	8892	3.3173	0.7440	0.5643
0.0076	39.0	9126	3.3449	0.75	0.5667
0.0075	40.0	9360	3.3469	0.7464	0.5647

Framework versions

• Transformers 4.57.1
• Pytorch 2.9.0+cu128
• Datasets 4.3.0
• Tokenizers 0.22.1