elselse's picture
CIRCL/cwe-parent-vulnerability-classification-roberta-base
9b52994 verified
|
raw
history blame
4.41 kB
metadata
library_name: transformers
license: mit
base_model: roberta-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: cwe-parent-vulnerability-classification-roberta-base
    results: []

cwe-parent-vulnerability-classification-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: 1.3447
  • Accuracy: 0.7727
  • F1 Macro: 0.4235

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
3.263 1.0 25 3.2313 0.2614 0.0345
3.1726 2.0 50 2.9923 0.1818 0.0431
3.1008 3.0 75 2.8828 0.0 0.0
2.9805 4.0 100 2.8858 0.0114 0.0019
3.0021 5.0 125 2.8257 0.4432 0.0626
2.8775 6.0 150 2.7730 0.0795 0.0462
2.8805 7.0 175 2.6421 0.2841 0.1362
2.6602 8.0 200 2.6462 0.3864 0.1366
2.5303 9.0 225 2.5584 0.3523 0.1461
2.5236 10.0 250 2.4933 0.4205 0.1209
2.3221 11.0 275 2.3458 0.5909 0.2232
2.1446 12.0 300 2.2679 0.625 0.2521
1.9937 13.0 325 2.1932 0.625 0.2736
1.8521 14.0 350 2.0372 0.6477 0.2881
1.7899 15.0 375 1.9494 0.6364 0.2679
1.5273 16.0 400 1.8457 0.6705 0.3205
1.4178 17.0 425 1.8276 0.6477 0.2931
1.335 18.0 450 1.7690 0.6591 0.3004
1.2685 19.0 475 1.6681 0.6705 0.3577
1.112 20.0 500 1.6399 0.6818 0.3152
1.01 21.0 525 1.5561 0.6932 0.3255
0.9637 22.0 550 1.5008 0.7159 0.4218
0.9571 23.0 575 1.5387 0.7045 0.3385
0.8213 24.0 600 1.5366 0.7159 0.4043
0.7538 25.0 625 1.4691 0.75 0.3942
0.7228 26.0 650 1.4826 0.7273 0.3872
0.7244 27.0 675 1.4789 0.7386 0.3915
0.6746 28.0 700 1.4439 0.7727 0.4322
0.5959 29.0 725 1.4202 0.7614 0.3942
0.5788 30.0 750 1.4339 0.7727 0.4002
0.5718 31.0 775 1.3723 0.7955 0.4431
0.5358 32.0 800 1.4186 0.7727 0.3812
0.5094 33.0 825 1.3722 0.7841 0.4579
0.5003 34.0 850 1.3955 0.7614 0.3786
0.4973 35.0 875 1.3733 0.8068 0.4635
0.4721 36.0 900 1.3447 0.7727 0.4235
0.4457 37.0 925 1.3622 0.7955 0.4573
0.4232 38.0 950 1.3736 0.7614 0.3986
0.4405 39.0 975 1.3683 0.7727 0.4235
0.437 40.0 1000 1.3642 0.7614 0.3986

Framework versions

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