metadata
license: apache-2.0
base_model: facebook/bart-base
tags:
- generated_from_trainer
metrics:
- rouge
model-index:
- name: bart-base-finetuned-CNN-DailyNews
results: []
bart-base-finetuned-CNN-DailyNews
This model is a fine-tuned version of facebook/bart-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.8594
- Rouge1: 0.1876
- Rouge2: 0.1056
- Rougel: 0.1659
- Rougelsum: 0.1757
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: 5.6e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|---|---|---|---|---|---|---|---|
| 2.6078 | 1.0 | 63 | 1.9681 | 0.1706 | 0.0988 | 0.1529 | 0.1607 |
| 2.0494 | 2.0 | 126 | 1.8818 | 0.1908 | 0.1085 | 0.17 | 0.1792 |
| 1.8102 | 3.0 | 189 | 1.8500 | 0.1888 | 0.1084 | 0.1665 | 0.175 |
| 1.627 | 4.0 | 252 | 1.8500 | 0.185 | 0.1033 | 0.1631 | 0.1718 |
| 1.4668 | 5.0 | 315 | 1.8453 | 0.1793 | 0.1072 | 0.1612 | 0.1689 |
| 1.3781 | 6.0 | 378 | 1.8412 | 0.1815 | 0.1018 | 0.159 | 0.1697 |
| 1.292 | 7.0 | 441 | 1.8545 | 0.1868 | 0.1067 | 0.1667 | 0.175 |
| 1.253 | 8.0 | 504 | 1.8594 | 0.1876 | 0.1056 | 0.1659 | 0.1757 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0