|
--- |
|
base_model: google/pegasus-x-base |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- eur-lex-sum |
|
model-index: |
|
- name: PegasusX_no_extraction_V1 |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# PegasusX_no_extraction_V1 |
|
|
|
This model is a fine-tuned version of [google/pegasus-x-base](https://huggingface.co/google/pegasus-x-base) on the eur-lex-sum dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.6795 |
|
|
|
## 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: 2 |
|
- eval_batch_size: 1 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- num_devices: 4 |
|
- gradient_accumulation_steps: 2 |
|
- total_train_batch_size: 16 |
|
- total_eval_batch_size: 4 |
|
- 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 | |
|
|:-------------:|:-------:|:----:|:---------------:| |
|
| 4.9176 | 0.9927 | 68 | 3.9239 | |
|
| 3.8267 | 2.0 | 137 | 3.2236 | |
|
| 3.2452 | 2.9927 | 205 | 2.6649 | |
|
| 2.7272 | 4.0 | 274 | 2.2625 | |
|
| 2.4546 | 4.9927 | 342 | 2.0656 | |
|
| 2.2504 | 6.0 | 411 | 1.9579 | |
|
| 2.1713 | 6.9927 | 479 | 1.8934 | |
|
| 2.0563 | 8.0 | 548 | 1.8536 | |
|
| 2.023 | 8.9927 | 616 | 1.8237 | |
|
| 1.9452 | 10.0 | 685 | 1.8021 | |
|
| 1.9365 | 10.9927 | 753 | 1.7839 | |
|
| 1.8701 | 12.0 | 822 | 1.7746 | |
|
| 1.8756 | 12.9927 | 890 | 1.7641 | |
|
| 1.8261 | 14.0 | 959 | 1.7505 | |
|
| 1.827 | 14.9927 | 1027 | 1.7454 | |
|
| 1.7861 | 16.0 | 1096 | 1.7353 | |
|
| 1.7943 | 16.9927 | 1164 | 1.7280 | |
|
| 1.7501 | 18.0 | 1233 | 1.7276 | |
|
| 1.7606 | 18.9927 | 1301 | 1.7176 | |
|
| 1.7264 | 20.0 | 1370 | 1.7119 | |
|
| 1.7371 | 20.9927 | 1438 | 1.6997 | |
|
| 1.7008 | 22.0 | 1507 | 1.7067 | |
|
| 1.7101 | 22.9927 | 1575 | 1.7002 | |
|
| 1.6865 | 24.0 | 1644 | 1.6997 | |
|
| 1.6967 | 24.9927 | 1712 | 1.6914 | |
|
| 1.6648 | 26.0 | 1781 | 1.6915 | |
|
| 1.6761 | 26.9927 | 1849 | 1.6893 | |
|
| 1.6432 | 28.0 | 1918 | 1.6918 | |
|
| 1.6688 | 28.9927 | 1986 | 1.6863 | |
|
| 1.6289 | 30.0 | 2055 | 1.6858 | |
|
| 1.6475 | 30.9927 | 2123 | 1.6878 | |
|
| 1.6176 | 32.0 | 2192 | 1.6838 | |
|
| 1.6435 | 32.9927 | 2260 | 1.6835 | |
|
| 1.6139 | 34.0 | 2329 | 1.6802 | |
|
| 1.638 | 34.9927 | 2397 | 1.6806 | |
|
| 1.6099 | 36.0 | 2466 | 1.6830 | |
|
| 1.6359 | 36.9927 | 2534 | 1.6778 | |
|
| 1.6056 | 38.0 | 2603 | 1.6813 | |
|
| 1.6281 | 38.9927 | 2671 | 1.6789 | |
|
| 1.6132 | 39.7080 | 2720 | 1.6795 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.40.1 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.17.1 |
|
- Tokenizers 0.19.1 |
|
|
