20230430-002-baseline-mdeberta-qa-ft-clickbait-spoiling
This model is a fine-tuned version of timpal0l/mdeberta-v3-base-squad2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.6307
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: 2e-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
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 1.0 | 200 | 2.3170 |
| No log | 2.0 | 400 | 2.2467 |
| 2.2007 | 3.0 | 600 | 2.4259 |
| 2.2007 | 4.0 | 800 | 2.6208 |
| 1.1169 | 5.0 | 1000 | 2.9482 |
| 1.1169 | 6.0 | 1200 | 3.1302 |
| 1.1169 | 7.0 | 1400 | 3.4215 |
| 0.5849 | 8.0 | 1600 | 3.4504 |
| 0.5849 | 9.0 | 1800 | 3.5371 |
| 0.3761 | 10.0 | 2000 | 3.6307 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3