cwe-parent-vulnerability-classification-microsoft-codebert-base

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

• Loss: 2.6351
• Accuracy: 0.6517
• F1 Macro: 0.3129

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 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
3.2599	1.0	25	3.2873	0.0225	0.0037
3.1436	2.0	50	3.3011	0.0562	0.0216
3.1168	3.0	75	3.3439	0.0449	0.0113
3.0315	4.0	100	3.3314	0.1461	0.0645
3.0604	5.0	125	3.3334	0.0899	0.0581
2.9746	6.0	150	3.3430	0.1124	0.0546
2.9773	7.0	175	3.3535	0.4157	0.0990
2.8666	8.0	200	3.2720	0.4831	0.2052
2.8196	9.0	225	3.2289	0.4270	0.1442
2.6704	10.0	250	3.1301	0.2584	0.1440
2.6964	11.0	275	3.0508	0.2809	0.1197
2.5442	12.0	300	2.9618	0.3596	0.1644
2.4519	13.0	325	2.9271	0.3596	0.1637
2.4064	14.0	350	2.8342	0.3933	0.2154
2.2469	15.0	375	2.7950	0.3596	0.2097
2.1662	16.0	400	2.7928	0.3596	0.1926
2.126	17.0	425	2.6786	0.4157	0.2223
2.0579	18.0	450	2.7615	0.3820	0.1987
1.8908	19.0	475	2.6469	0.4157	0.2015
1.8119	20.0	500	2.7396	0.4157	0.2097
1.8234	21.0	525	2.7319	0.3933	0.2101
1.6483	22.0	550	2.7024	0.4607	0.2504
1.7195	23.0	575	2.6693	0.4944	0.2345
1.5326	24.0	600	2.6387	0.5169	0.2341
1.5649	25.0	625	2.6509	0.6180	0.2934
1.4294	26.0	650	2.7232	0.6292	0.3175
1.4872	27.0	675	2.6745	0.6404	0.3005
1.3451	28.0	700	2.6499	0.6517	0.3100
1.296	29.0	725	2.6788	0.6517	0.3290
1.2962	30.0	750	2.6351	0.6517	0.3129
1.2969	31.0	775	2.6432	0.6742	0.3226
1.1886	32.0	800	2.6496	0.6742	0.3226
1.1426	33.0	825	2.6603	0.6742	0.3230
1.1833	34.0	850	2.6660	0.6742	0.3253
1.14	35.0	875	2.6588	0.6854	0.3477
1.0947	36.0	900	2.6501	0.6854	0.3477
1.0714	37.0	925	2.6654	0.6854	0.3477
1.0678	38.0	950	2.6454	0.6854	0.3477
1.0535	39.0	975	2.6375	0.6854	0.3481
1.0273	40.0	1000	2.6422	0.6854	0.3481

Framework versions

• Transformers 4.55.4
• Pytorch 2.7.1+cu126
• Datasets 4.0.0
• Tokenizers 0.21.2