bert_key_extractor_finetune
This model is a fine-tuned version of facebook/bart-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.0069
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: 0.0001
- train_batch_size: 8
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 32
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 12
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 11.322 | 0.99 | 14 | 10.9937 |
| 9.3954 | 1.98 | 28 | 8.8487 |
| 6.5458 | 2.97 | 42 | 5.9401 |
| 5.1036 | 3.96 | 56 | 4.6196 |
| 3.8587 | 4.96 | 70 | 3.4009 |
| 2.7987 | 5.95 | 84 | 2.7571 |
| 2.6306 | 6.94 | 98 | 2.5074 |
| 2.3636 | 8.0 | 113 | 2.3132 |
| 2.2169 | 8.99 | 127 | 2.2248 |
| 2.1732 | 9.98 | 141 | 2.1092 |
| 2.0377 | 10.97 | 155 | 2.0351 |
| 1.9973 | 11.89 | 168 | 2.0069 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.0.0+cu117
- Datasets 2.14.5
- Tokenizers 0.14.1