<!-- 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. -->
madatnlp/rob-large-krmath2
This model is a fine-tuned version of klue/roberta-large on an unknown dataset. It achieves the following results on the evaluation set:
- Train Loss: 0.0707
- Validation Loss: 0.2571
- Epoch: 17
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': 'SGD', 'learning_rate': 0.01, 'decay': 0.0, 'momentum': 0.9, 'nesterov': False}
- training_precision: float32
Training results
| Train Loss | Validation Loss | Epoch |
|---|---|---|
| 1.7479 | 1.5215 | 0 |
| 1.1286 | 0.7513 | 1 |
| 1.2498 | 0.9242 | 2 |
| 1.0213 | 0.7140 | 3 |
| 0.8002 | 0.6080 | 4 |
| 0.5895 | 0.3721 | 5 |
| 0.4699 | 0.3392 | 6 |
| 0.3064 | 0.2599 | 7 |
| 0.2803 | 0.2360 | 8 |
| 0.2162 | 0.3075 | 9 |
| 0.1878 | 0.2652 | 10 |
| 0.1635 | 0.1618 | 11 |
| 0.1342 | 0.1061 | 12 |
| 0.1058 | 0.2906 | 13 |
| 0.0869 | 0.3535 | 14 |
| 0.0704 | 0.2090 | 15 |
| 0.0608 | 0.1777 | 16 |
| 0.0707 | 0.2571 | 17 |
Framework versions
- Transformers 4.19.2
- TensorFlow 2.8.2
- Datasets 2.2.2
- Tokenizers 0.12.1
