mallikrao2/qa-finetuned-swag
This model is a fine-tuned version of mallikrao2/sQuad_bertmodel1_ on an unknown dataset. It achieves the following results on the evaluation set:
- Train Loss: 0.0020
- Train Accuracy: 0.9994
- Validation Loss: 1.9750
- Validation Accuracy: 0.7508
- Epoch: 19
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:
- optimizer: {'name': 'Adam', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 5e-05, 'decay_steps': 66860, '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}
- training_precision: float32
Training results
Train Loss | Train Accuracy | Validation Loss | Validation Accuracy | Epoch |
---|---|---|---|---|
0.8405 | 0.6630 | 0.6156 | 0.7596 | 0 |
0.4697 | 0.8233 | 0.6102 | 0.7643 | 1 |
0.2479 | 0.9087 | 0.7102 | 0.7573 | 2 |
0.1571 | 0.9439 | 0.8434 | 0.7482 | 3 |
0.1139 | 0.9599 | 1.0923 | 0.7453 | 4 |
0.0881 | 0.9698 | 1.0614 | 0.7421 | 5 |
0.0705 | 0.9758 | 1.1311 | 0.7412 | 6 |
0.0577 | 0.9802 | 1.1761 | 0.7387 | 7 |
0.0453 | 0.9845 | 1.3310 | 0.7446 | 8 |
0.0379 | 0.9869 | 1.3076 | 0.7361 | 9 |
0.0301 | 0.9898 | 1.3147 | 0.7434 | 10 |
0.0228 | 0.9923 | 1.6641 | 0.7388 | 11 |
0.0195 | 0.9932 | 1.6168 | 0.7397 | 12 |
0.0165 | 0.9948 | 1.6042 | 0.7458 | 13 |
0.0118 | 0.9960 | 1.6922 | 0.7426 | 14 |
0.0098 | 0.9970 | 1.7052 | 0.7449 | 15 |
0.0059 | 0.9982 | 1.8137 | 0.7453 | 16 |
0.0040 | 0.9986 | 1.9369 | 0.7504 | 17 |
0.0032 | 0.9991 | 1.9089 | 0.7498 | 18 |
0.0020 | 0.9994 | 1.9750 | 0.7508 | 19 |
Framework versions
- Transformers 4.29.2
- TensorFlow 2.8.0
- Datasets 2.12.0
- Tokenizers 0.13.3
- Downloads last month
- 2
Inference API (serverless) does not yet support transformers models for this pipeline type.