metadata
license: apache-2.0
base_model: facebook/bart-large
tags:
- generated_from_trainer
datasets:
- eur-lex-sum
model-index:
- name: BART_no_extraction_V2
results: []
BART_no_extraction_V2
This model is a fine-tuned version of facebook/bart-large on the eur-lex-sum dataset. It achieves the following results on the evaluation set:
- Loss: 2.0427
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: 5e-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
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 40
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.4492 | 1.0 | 69 | 2.2924 |
| 2.2834 | 2.0 | 138 | 2.1161 |
| 2.0586 | 3.0 | 207 | 2.0469 |
| 1.9241 | 4.0 | 276 | 2.0229 |
| 1.8083 | 5.0 | 345 | 2.0068 |
| 1.7006 | 6.0 | 414 | 1.9867 |
| 1.6168 | 7.0 | 483 | 1.9871 |
| 1.5344 | 8.0 | 552 | 2.0058 |
| 1.4678 | 9.0 | 621 | 2.0054 |
| 1.3988 | 10.0 | 690 | 2.0284 |
| 1.3369 | 11.0 | 759 | 2.0427 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.2.1+cu121
- Datasets 2.17.1
- Tokenizers 0.19.1