metadata
license: apache-2.0
base_model: facebook/bart-base
tags:
- generated_from_trainer
model-index:
- name: pubmed-abs-noise-01
results: []
pubmed-abs-noise-01
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: 0.2094
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 |
|---|---|---|---|
| 0.3418 | 0.11 | 500 | 0.3102 |
| 0.3315 | 0.21 | 1000 | 0.2811 |
| 0.3479 | 0.32 | 1500 | 0.2585 |
| 0.308 | 0.43 | 2000 | 0.2609 |
| 0.2716 | 0.54 | 2500 | 0.2549 |
| 0.2845 | 0.64 | 3000 | 0.2442 |
| 0.2781 | 0.75 | 3500 | 0.2379 |
| 0.2893 | 0.86 | 4000 | 0.2314 |
| 0.2581 | 0.96 | 4500 | 0.2297 |
| 0.2269 | 1.07 | 5000 | 0.2334 |
| 0.2274 | 1.18 | 5500 | 0.2272 |
| 0.2053 | 1.28 | 6000 | 0.2305 |
| 0.2062 | 1.39 | 6500 | 0.2246 |
| 0.241 | 1.5 | 7000 | 0.2215 |
| 0.1625 | 1.61 | 7500 | 0.2239 |
| 0.2179 | 1.71 | 8000 | 0.2181 |
| 0.2372 | 1.82 | 8500 | 0.2187 |
| 0.2116 | 1.93 | 9000 | 0.2115 |
| 0.1625 | 2.03 | 9500 | 0.2168 |
| 0.187 | 2.14 | 10000 | 0.2170 |
| 0.159 | 2.25 | 10500 | 0.2163 |
| 0.1741 | 2.35 | 11000 | 0.2144 |
| 0.1964 | 2.46 | 11500 | 0.2111 |
| 0.1679 | 2.57 | 12000 | 0.2117 |
| 0.1662 | 2.68 | 12500 | 0.2096 |
| 0.1436 | 2.78 | 13000 | 0.2107 |
| 0.1875 | 2.89 | 13500 | 0.2099 |
| 0.1656 | 3.0 | 14000 | 0.2094 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0
- Datasets 2.14.7
- Tokenizers 0.14.1