metadata
license: apache-2.0
base_model: facebook/bart-base
tags:
- generated_from_trainer
model-index:
- name: pubmed-abs-noise-05
results: []
pubmed-abs-noise-05
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: 0.6496
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_steps: 10
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.092 | 0.11 | 500 | 0.9325 |
| 0.9641 | 0.21 | 1000 | 0.8608 |
| 1.0013 | 0.32 | 1500 | 0.8162 |
| 0.9596 | 0.43 | 2000 | 0.7818 |
| 0.8598 | 0.54 | 2500 | 0.7740 |
| 0.7675 | 0.64 | 3000 | 0.7710 |
| 0.7932 | 0.75 | 3500 | 0.7411 |
| 0.766 | 0.86 | 4000 | 0.7280 |
| 0.7806 | 0.96 | 4500 | 0.7130 |
| 0.7571 | 1.07 | 5000 | 0.7091 |
| 0.6968 | 1.18 | 5500 | 0.7011 |
| 0.6832 | 1.28 | 6000 | 0.6986 |
| 0.674 | 1.39 | 6500 | 0.6935 |
| 0.6475 | 1.5 | 7000 | 0.6829 |
| 0.6876 | 1.61 | 7500 | 0.6798 |
| 0.6355 | 1.71 | 8000 | 0.6745 |
| 0.6788 | 1.82 | 8500 | 0.6717 |
| 0.62 | 1.93 | 9000 | 0.6692 |
| 0.5702 | 2.03 | 9500 | 0.6697 |
| 0.6105 | 2.14 | 10000 | 0.6665 |
| 0.5658 | 2.25 | 10500 | 0.6646 |
| 0.5826 | 2.35 | 11000 | 0.6630 |
| 0.5342 | 2.46 | 11500 | 0.6558 |
| 0.5624 | 2.57 | 12000 | 0.6560 |
| 0.5659 | 2.68 | 12500 | 0.6526 |
| 0.5328 | 2.78 | 13000 | 0.6514 |
| 0.5268 | 2.89 | 13500 | 0.6501 |
| 0.5536 | 3.0 | 14000 | 0.6496 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0
- Datasets 2.14.7
- Tokenizers 0.14.1