metadata
license: apache-2.0
base_model: facebook/bart-base
tags:
- generated_from_trainer
model-index:
- name: pubmed-abs-sub-02
results: []
pubmed-abs-sub-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: 0.2005
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 |
|---|---|---|---|
| 0.6442 | 0.11 | 500 | 0.5078 |
| 0.455 | 0.21 | 1000 | 0.3945 |
| 0.4276 | 0.32 | 1500 | 0.3555 |
| 0.3951 | 0.43 | 2000 | 0.3177 |
| 0.3851 | 0.54 | 2500 | 0.3026 |
| 0.3178 | 0.64 | 3000 | 0.2766 |
| 0.294 | 0.75 | 3500 | 0.2702 |
| 0.2857 | 0.86 | 4000 | 0.2565 |
| 0.305 | 0.96 | 4500 | 0.2434 |
| 0.2224 | 1.07 | 5000 | 0.2427 |
| 0.2122 | 1.18 | 5500 | 0.2419 |
| 0.2026 | 1.28 | 6000 | 0.2402 |
| 0.2155 | 1.39 | 6500 | 0.2258 |
| 0.2277 | 1.5 | 7000 | 0.2239 |
| 0.2115 | 1.61 | 7500 | 0.2209 |
| 0.194 | 1.71 | 8000 | 0.2191 |
| 0.2329 | 1.82 | 8500 | 0.2163 |
| 0.1781 | 1.93 | 9000 | 0.2114 |
| 0.154 | 2.03 | 9500 | 0.2148 |
| 0.1722 | 2.14 | 10000 | 0.2129 |
| 0.1329 | 2.25 | 10500 | 0.2114 |
| 0.1535 | 2.35 | 11000 | 0.2076 |
| 0.1414 | 2.46 | 11500 | 0.2057 |
| 0.1618 | 2.57 | 12000 | 0.2033 |
| 0.1486 | 2.68 | 12500 | 0.2028 |
| 0.1302 | 2.78 | 13000 | 0.2023 |
| 0.1314 | 2.89 | 13500 | 0.2020 |
| 0.1392 | 3.0 | 14000 | 0.2005 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0
- Datasets 2.14.6
- Tokenizers 0.14.1