metadata
license: apache-2.0
base_model: facebook/bart-base
tags:
- generated_from_trainer
model-index:
- name: bart-noised-with-gcd-dist-0.3
results: []
bart-noised-with-gcd-dist-0.3
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: 1.0832
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 |
|---|---|---|---|
| 1.4395 | 0.11 | 500 | 1.3691 |
| 1.4568 | 0.21 | 1000 | 1.3068 |
| 1.3702 | 0.32 | 1500 | 1.2510 |
| 1.2945 | 0.43 | 2000 | 1.2319 |
| 1.3022 | 0.54 | 2500 | 1.2092 |
| 1.1775 | 0.64 | 3000 | 1.2022 |
| 1.1657 | 0.75 | 3500 | 1.1822 |
| 1.1756 | 0.86 | 4000 | 1.1823 |
| 1.2103 | 0.96 | 4500 | 1.1521 |
| 1.1467 | 1.07 | 5000 | 1.1535 |
| 1.0617 | 1.18 | 5500 | 1.1443 |
| 1.0667 | 1.28 | 6000 | 1.1339 |
| 1.0906 | 1.39 | 6500 | 1.1369 |
| 1.0492 | 1.5 | 7000 | 1.1213 |
| 1.0203 | 1.61 | 7500 | 1.1207 |
| 0.9431 | 1.71 | 8000 | 1.1183 |
| 1.0522 | 1.82 | 8500 | 1.1052 |
| 1.0673 | 1.93 | 9000 | 1.0996 |
| 0.9521 | 2.03 | 9500 | 1.1079 |
| 0.9615 | 2.14 | 10000 | 1.0990 |
| 0.996 | 2.25 | 10500 | 1.0976 |
| 0.9458 | 2.35 | 11000 | 1.0997 |
| 0.9375 | 2.46 | 11500 | 1.0939 |
| 0.8869 | 2.57 | 12000 | 1.0896 |
| 0.8934 | 2.68 | 12500 | 1.0895 |
| 0.8561 | 2.78 | 13000 | 1.0839 |
| 0.9795 | 2.89 | 13500 | 1.0834 |
| 0.8886 | 3.0 | 14000 | 1.0832 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1