metadata
license: apache-2.0
base_model: facebook/bart-base
tags:
- generated_from_trainer
model-index:
- name: pubmed-abs-noise-02
results: []
pubmed-abs-noise-02
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: 0.3305
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 |
|---|---|---|---|
| 0.5324 | 0.11 | 500 | 0.4918 |
| 0.4891 | 0.21 | 1000 | 0.4361 |
| 0.4725 | 0.32 | 1500 | 0.4166 |
| 0.5121 | 0.43 | 2000 | 0.4054 |
| 0.4325 | 0.54 | 2500 | 0.3945 |
| 0.4185 | 0.64 | 3000 | 0.3897 |
| 0.4125 | 0.75 | 3500 | 0.3782 |
| 0.3891 | 0.86 | 4000 | 0.3704 |
| 0.3846 | 0.96 | 4500 | 0.3627 |
| 0.3211 | 1.07 | 5000 | 0.3664 |
| 0.3084 | 1.18 | 5500 | 0.3548 |
| 0.3552 | 1.28 | 6000 | 0.3543 |
| 0.3409 | 1.39 | 6500 | 0.3513 |
| 0.3258 | 1.5 | 7000 | 0.3520 |
| 0.3635 | 1.61 | 7500 | 0.3515 |
| 0.2881 | 1.71 | 8000 | 0.3420 |
| 0.3454 | 1.82 | 8500 | 0.3428 |
| 0.3194 | 1.93 | 9000 | 0.3391 |
| 0.3092 | 2.03 | 9500 | 0.3426 |
| 0.2572 | 2.14 | 10000 | 0.3448 |
| 0.2803 | 2.25 | 10500 | 0.3403 |
| 0.2375 | 2.35 | 11000 | 0.3367 |
| 0.2657 | 2.46 | 11500 | 0.3361 |
| 0.2782 | 2.57 | 12000 | 0.3325 |
| 0.2713 | 2.68 | 12500 | 0.3308 |
| 0.2386 | 2.78 | 13000 | 0.3317 |
| 0.2428 | 2.89 | 13500 | 0.3318 |
| 0.2702 | 3.0 | 14000 | 0.3305 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0
- Datasets 2.14.7
- Tokenizers 0.14.1