electra_base_generator_AGGRO_V2
This model is a fine-tuned version of google/electra-base-generator on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 5.8414
- Exact Match: 0.0
- F1 Score: 7.2863
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: 1e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 3407
- gradient_accumulation_steps: 16
- total_train_batch_size: 64
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- training_steps: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Exact Match | F1 Score |
|---|---|---|---|---|---|
| 5.9896 | 0.0053 | 1 | 5.9890 | 0.0 | 4.4594 |
| 5.9717 | 0.0107 | 2 | 5.9786 | 0.0 | 4.5055 |
| 5.9601 | 0.0160 | 3 | 5.9580 | 0.0 | 4.3989 |
| 5.9488 | 0.0214 | 4 | 5.9278 | 0.0 | 4.7080 |
| 5.94 | 0.0267 | 5 | 5.8890 | 0.0 | 5.8695 |
| 5.9032 | 0.0321 | 6 | 5.8433 | 0.0 | 7.2872 |
| 5.8764 | 0.0374 | 7 | 5.8002 | 0.0 | 6.1300 |
| 5.8194 | 0.0428 | 8 | 5.7590 | 0.0 | 6.0529 |
| 5.7899 | 0.0481 | 9 | 5.7189 | 0.0 | 6.1531 |
Framework versions
- Transformers 4.46.3
- Pytorch 2.4.0
- Datasets 3.1.0
- Tokenizers 0.20.3