metadata
base_model: facebook/bart-large-cnn
library_name: peft
license: mit
metrics:
- rouge
tags:
- generated_from_trainer
model-index:
- name: legal_bart_large_cnn
results: []
legal_bart_large_cnn
This model is a fine-tuned version of facebook/bart-large-cnn on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.2520
- Rouge1: 0.489
- Rouge2: 0.2433
- Rougel: 0.3262
- Rougelsum: 0.3267
- Gen Len: 137.2973
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: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 352 | 2.3127 | 0.4831 | 0.2372 | 0.3195 | 0.3198 | 136.5541 |
| 2.4898 | 2.0 | 704 | 2.2627 | 0.4883 | 0.2432 | 0.3257 | 0.326 | 136.9211 |
| 2.3583 | 3.0 | 1056 | 2.2520 | 0.489 | 0.2433 | 0.3262 | 0.3267 | 137.2973 |
Framework versions
- PEFT 0.11.1
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1