bart-base-wsd-finetuned-cve-reason-2
This model is a fine-tuned version of mgkamalesh7/bart-base-wsd-finetuned-cve-reason on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4159
- Rouge1: 91.1806
- Rouge2: 87.5256
- Rougel: 91.1424
- Rougelsum: 91.1124
- Gen Len: 8.6492
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 200
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
---|---|---|---|---|---|---|---|---|
No log | 1.0 | 56 | 0.4140 | 89.658 | 85.4214 | 89.5042 | 89.4542 | 8.7377 |
No log | 2.0 | 112 | 0.3450 | 89.6585 | 86.3159 | 89.5529 | 89.537 | 8.7344 |
No log | 3.0 | 168 | 0.4252 | 89.2531 | 85.5599 | 89.1241 | 89.1115 | 8.7311 |
No log | 4.0 | 224 | 0.4278 | 89.4207 | 85.3854 | 89.1996 | 89.2622 | 8.7443 |
No log | 5.0 | 280 | 0.4023 | 90.0865 | 86.5253 | 89.8973 | 89.9373 | 8.7475 |
No log | 6.0 | 336 | 0.3831 | 89.6788 | 86.8093 | 89.5032 | 89.5405 | 8.6557 |
No log | 7.0 | 392 | 0.4100 | 90.7802 | 87.4674 | 90.6112 | 90.6915 | 8.6721 |
No log | 8.0 | 448 | 0.4425 | 90.4749 | 87.3615 | 90.4286 | 90.3849 | 8.6459 |
0.0544 | 9.0 | 504 | 0.4098 | 90.3948 | 86.9721 | 90.4295 | 90.3163 | 8.5541 |
0.0544 | 10.0 | 560 | 0.4289 | 89.8096 | 85.6718 | 89.6744 | 89.6666 | 8.6787 |
0.0544 | 11.0 | 616 | 0.4338 | 90.9849 | 87.3715 | 90.9871 | 90.9153 | 8.6787 |
0.0544 | 12.0 | 672 | 0.4159 | 91.1806 | 87.5256 | 91.1424 | 91.1124 | 8.6492 |
Framework versions
- Transformers 4.42.3
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1
- Downloads last month
- 4
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.