<!-- 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. -->
rafsankabir/Finetuned_NLI_TeacherModel
This model is a fine-tuned version of sagorsarker/bangla-bert-base on the xnli_bn dataset. It achieves the following results on the evaluation set:
- Loss: 2.5831
- Accuracy: 0.6879
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: 32
- eval_batch_size: 32
- seed: 33
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.8381 | 1.0 | 11921 | 0.7554 | 0.6701 |
| 0.7221 | 2.0 | 23842 | 0.7214 | 0.6833 |
| 0.643 | 3.0 | 35763 | 0.7164 | 0.6920 |
| 0.5614 | 4.0 | 47684 | 0.7536 | 0.6862 |
| 0.4761 | 5.0 | 59605 | 0.8104 | 0.6875 |
| 0.395 | 6.0 | 71526 | 0.9219 | 0.6891 |
| 0.3239 | 7.0 | 83447 | 1.0047 | 0.6833 |
| 0.2627 | 8.0 | 95368 | 1.0624 | 0.6900 |
| 0.2138 | 9.0 | 107289 | 1.2522 | 0.6714 |
| 0.1768 | 10.0 | 119210 | 1.2947 | 0.6763 |
| 0.1455 | 11.0 | 131131 | 1.4790 | 0.6838 |
| 0.1246 | 12.0 | 143052 | 1.5446 | 0.6813 |
| 0.1073 | 13.0 | 154973 | 1.7562 | 0.6742 |
| 0.094 | 14.0 | 166894 | 1.8442 | 0.6891 |
| 0.0822 | 15.0 | 178815 | 1.9902 | 0.6842 |
| 0.0707 | 16.0 | 190736 | 2.2021 | 0.6825 |
| 0.0589 | 17.0 | 202657 | 2.2803 | 0.6854 |
| 0.0497 | 18.0 | 214578 | 2.4199 | 0.6804 |
| 0.0407 | 19.0 | 226499 | 2.5014 | 0.6850 |
| 0.0337 | 20.0 | 238420 | 2.5831 | 0.6879 |
Framework versions
- Transformers 4.27.1
- Pytorch 2.0.0+cu117
- Datasets 2.9.0
- Tokenizers 0.13.2
