fine_tuned_bart_on_vnx_1751605261_
This model is a fine-tuned version of facebook/bart-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 5.4274
- Rouge1: 0.3939
- Rouge2: 0.1079
- Rougel: 0.2477
- Rougelsum: 0.2471
- Gen Len: 20.2068
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: 128
- eval_batch_size: 128
- 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: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 21 | 6.1676 | 0.377 | 0.106 | 0.2407 | 0.2405 | 19.3735 |
| No log | 2.0 | 42 | 6.0141 | 0.3919 | 0.1147 | 0.2513 | 0.2512 | 19.8889 |
| No log | 3.0 | 63 | 5.7388 | 0.3782 | 0.1072 | 0.2433 | 0.243 | 19.0278 |
| No log | 4.0 | 84 | 5.5987 | 0.3849 | 0.1082 | 0.2462 | 0.2461 | 19.4012 |
| No log | 5.0 | 105 | 5.5299 | 0.3867 | 0.1129 | 0.2492 | 0.2488 | 19.4198 |
| No log | 6.0 | 126 | 5.4911 | 0.3932 | 0.1123 | 0.2491 | 0.2488 | 19.8735 |
| No log | 7.0 | 147 | 5.4498 | 0.3998 | 0.1165 | 0.2523 | 0.2519 | 20.0802 |
| No log | 8.0 | 168 | 5.4455 | 0.3897 | 0.1068 | 0.2465 | 0.246 | 19.966 |
| No log | 9.0 | 189 | 5.4099 | 0.393 | 0.108 | 0.2468 | 0.2466 | 20.2006 |
| No log | 10.0 | 210 | 5.4274 | 0.3939 | 0.1079 | 0.2477 | 0.2471 | 20.2068 |
Framework versions
- Transformers 4.50.3
- Pytorch 2.6.0+cu126
- Datasets 3.5.0
- Tokenizers 0.21.1