<!-- This model card has been generated automatically according to the information Keras had access to. You should probably proofread and complete it, then remove this comment. -->
hemagamal/mdeberta_quran_qa_model
This model is a fine-tuned version of timpal0l/mdeberta-v3-base-squad2 on an unknown dataset. It achieves the following results on the evaluation set:
- Train Loss: 11.9013
- Train End Logits Loss: 5.9506
- Train Start Logits Loss: 5.9506
- Train End Logits Sparse Categorical Accuracy: 0.0582
- Train Start Logits Sparse Categorical Accuracy: 0.0426
- Validation Loss: 11.9013
- Validation End Logits Loss: 5.9506
- Validation Start Logits Loss: 5.9506
- Validation End Logits Sparse Categorical Accuracy: 0.0459
- Validation Start Logits Sparse Categorical Accuracy: 0.0917
- Epoch: 15
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', 'weight_decay': None, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': True, 'is_legacy_optimizer': False, 'learning_rate': 2e-05, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
- training_precision: float32
Training results
| Train Loss | Train End Logits Loss | Train Start Logits Loss | Train End Logits Sparse Categorical Accuracy | Train Start Logits Sparse Categorical Accuracy | Validation Loss | Validation End Logits Loss | Validation Start Logits Loss | Validation End Logits Sparse Categorical Accuracy | Validation Start Logits Sparse Categorical Accuracy | Epoch |
|---|---|---|---|---|---|---|---|---|---|---|
| 12.4236 | 6.1795 | 6.2441 | 0.0724 | 0.0895 | 11.9013 | 5.9506 | 5.9506 | 0.0459 | 0.0917 | 0 |
| 11.9013 | 5.9506 | 5.9506 | 0.0469 | 0.0469 | 11.9013 | 5.9506 | 5.9506 | 0.0459 | 0.0917 | 1 |
| 11.9013 | 5.9506 | 5.9506 | 0.0369 | 0.0398 | 11.9013 | 5.9506 | 5.9506 | 0.0459 | 0.0917 | 2 |
| 11.9013 | 5.9506 | 5.9506 | 0.0369 | 0.0554 | 11.9013 | 5.9506 | 5.9506 | 0.0459 | 0.0917 | 3 |
| 11.9013 | 5.9506 | 5.9506 | 0.0483 | 0.0455 | 11.9013 | 5.9506 | 5.9506 | 0.0459 | 0.0917 | 4 |
| 11.9013 | 5.9506 | 5.9506 | 0.0554 | 0.0412 | 11.9013 | 5.9506 | 5.9506 | 0.0459 | 0.0917 | 5 |
| 11.9013 | 5.9506 | 5.9506 | 0.0241 | 0.0398 | 11.9013 | 5.9506 | 5.9506 | 0.0459 | 0.0917 | 6 |
| 11.9013 | 5.9506 | 5.9506 | 0.0369 | 0.0412 | 11.9013 | 5.9506 | 5.9506 | 0.0459 | 0.0917 | 7 |
| 11.9013 | 5.9506 | 5.9506 | 0.0426 | 0.0426 | 11.9013 | 5.9506 | 5.9506 | 0.0459 | 0.0917 | 8 |
| 11.9013 | 5.9506 | 5.9506 | 0.0511 | 0.0426 | 11.9013 | 5.9506 | 5.9506 | 0.0459 | 0.0917 | 9 |
| 11.9013 | 5.9506 | 5.9506 | 0.0426 | 0.0469 | 11.9013 | 5.9506 | 5.9506 | 0.0459 | 0.0917 | 10 |
| 11.9013 | 5.9506 | 5.9506 | 0.0440 | 0.0341 | 11.9013 | 5.9506 | 5.9506 | 0.0459 | 0.0917 | 11 |
| 11.9013 | 5.9506 | 5.9506 | 0.0412 | 0.0398 | 11.9013 | 5.9506 | 5.9506 | 0.0459 | 0.0917 | 12 |
| 11.9013 | 5.9506 | 5.9506 | 0.0440 | 0.0440 | 11.9013 | 5.9506 | 5.9506 | 0.0459 | 0.0917 | 13 |
| 11.9013 | 5.9506 | 5.9506 | 0.0426 | 0.0412 | 11.9013 | 5.9506 | 5.9506 | 0.0459 | 0.0917 | 14 |
| 11.9013 | 5.9506 | 5.9506 | 0.0582 | 0.0426 | 11.9013 | 5.9506 | 5.9506 | 0.0459 | 0.0917 | 15 |
Framework versions
- Transformers 4.29.1
- TensorFlow 2.12.0
- Datasets 2.12.0
- Tokenizers 0.13.3
