metadata
license: apache-2.0
base_model: facebook/bart-base
tags:
- generated_from_trainer
model-index:
- name: bart-noised-with-babylon-dist
results: []
bart-noised-with-babylon-dist
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: 0.2462
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
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.543 | 0.11 | 500 | 0.3991 |
| 0.4213 | 0.21 | 1000 | 0.3544 |
| 0.3907 | 0.32 | 1500 | 0.3328 |
| 0.4101 | 0.43 | 2000 | 0.3178 |
| 0.2998 | 0.54 | 2500 | 0.3148 |
| 0.3549 | 0.64 | 3000 | 0.2948 |
| 0.3401 | 0.75 | 3500 | 0.2861 |
| 0.3304 | 0.86 | 4000 | 0.2802 |
| 0.3404 | 0.96 | 4500 | 0.2749 |
| 0.2548 | 1.07 | 5000 | 0.2730 |
| 0.2725 | 1.18 | 5500 | 0.2696 |
| 0.2305 | 1.28 | 6000 | 0.2755 |
| 0.2424 | 1.39 | 6500 | 0.2647 |
| 0.2638 | 1.5 | 7000 | 0.2601 |
| 0.2276 | 1.61 | 7500 | 0.2622 |
| 0.2299 | 1.71 | 8000 | 0.2587 |
| 0.2817 | 1.82 | 8500 | 0.2519 |
| 0.2252 | 1.93 | 9000 | 0.2505 |
| 0.2022 | 2.03 | 9500 | 0.2554 |
| 0.1722 | 2.14 | 10000 | 0.2558 |
| 0.1878 | 2.25 | 10500 | 0.2546 |
| 0.2093 | 2.35 | 11000 | 0.2521 |
| 0.1656 | 2.46 | 11500 | 0.2513 |
| 0.1921 | 2.57 | 12000 | 0.2478 |
| 0.1754 | 2.68 | 12500 | 0.2468 |
| 0.2081 | 2.78 | 13000 | 0.2469 |
| 0.1707 | 2.89 | 13500 | 0.2472 |
| 0.2068 | 3.0 | 14000 | 0.2462 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1