paumena/QA-Electra_Base
This model is a fine-tuned version of google/electra-base-discriminator on an unknown dataset. It achieves the following results on the evaluation set:
- Train Loss: 0.3170
- Epoch: 4
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
{'exact_match': 84.66414380321665, 'f1': 91.59502747403307}
Training procedure
GPU V100 Standard RAM 2443s 439ms/step
Training hyperparameters
The following hyperparameters were used during training:
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 27660, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
- training_precision: mixed_float16
Training results
| Train Loss | Epoch |
|---|---|
| 1.0689 | 0 |
| 0.6768 | 1 |
| 0.5085 | 2 |
| 0.3933 | 3 |
| 0.3170 | 4 |
Framework versions
- Transformers 4.30.2
- TensorFlow 2.12.0
- Datasets 2.13.1
- Tokenizers 0.13.3