<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
SloBertAA_Top5_WithoutOOC_NEW
This model is a fine-tuned version of EMBEDDIA/sloberta on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3464
- Accuracy: 0.9419
Other related models
| Base model | Includes the OOC class? | 5 classes | 10 classes | 20 classes | 50 classes |
|---|---|---|---|---|---|
| SloBERTa | Yes | x | x | x | x |
| SloBERTa | No | x | x | x | x |
| BERT Multilingual | Yes | x | x | x | x |
| BERT Multilingual | No | x | x | x | x |
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: 2e-05
- train_batch_size: 12
- eval_batch_size: 12
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.2787 | 1.0 | 8757 | 0.2899 | 0.9120 |
| 0.2197 | 2.0 | 17514 | 0.2314 | 0.9294 |
| 0.1483 | 3.0 | 26271 | 0.3004 | 0.9349 |
| 0.0997 | 4.0 | 35028 | 0.3182 | 0.9405 |
| 0.0699 | 5.0 | 43785 | 0.3464 | 0.9419 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.8.0
- Datasets 2.10.1
- Tokenizers 0.13.2
