tesseract_chakma_sagor_banglabert
This model is a fine-tuned version of sagorsarker/bangla-bert-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.0812
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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 5.1857 | 1.0 | 184 | 4.5550 |
| 4.5189 | 2.0 | 368 | 4.2591 |
| 4.2575 | 3.0 | 552 | 4.0445 |
| 4.0892 | 4.0 | 736 | 3.8625 |
| 3.9351 | 5.0 | 920 | 3.7626 |
| 3.8631 | 6.0 | 1104 | 3.6802 |
| 3.7813 | 7.0 | 1288 | 3.6179 |
| 3.6816 | 8.0 | 1472 | 3.5077 |
| 3.6314 | 9.0 | 1656 | 3.4359 |
| 3.5459 | 10.0 | 1840 | 3.3758 |
| 3.4738 | 11.0 | 2024 | 3.3338 |
| 3.4369 | 12.0 | 2208 | 3.2885 |
| 3.3802 | 13.0 | 2392 | 3.2130 |
| 3.3504 | 14.0 | 2576 | 3.2004 |
| 3.3323 | 15.0 | 2760 | 3.1764 |
| 3.3007 | 16.0 | 2944 | 3.1422 |
| 3.2924 | 17.0 | 3128 | 3.1056 |
| 3.2354 | 18.0 | 3312 | 3.0958 |
| 3.2205 | 19.0 | 3496 | 3.1334 |
| 3.2555 | 20.0 | 3680 | 3.0812 |
Framework versions
- Transformers 4.56.1
- Pytorch 2.8.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.0
- Downloads last month
- 13
Model tree for adity12345/tesseract_chakma_sagor_banglabert
Base model
sagorsarker/bangla-bert-base