bengali_pos_v1_400000
This model is a fine-tuned version of mHossain/bengali_pos_v1_300000 on the pos_tag_100k dataset. It achieves the following results on the evaluation set:
- Loss: 0.5609
- Precision: 0.7830
- Recall: 0.7856
- F1: 0.7843
- Accuracy: 0.8402
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.5908 | 1.0 | 22500 | 0.5513 | 0.7688 | 0.7698 | 0.7693 | 0.8289 |
0.4642 | 2.0 | 45000 | 0.5415 | 0.7799 | 0.7822 | 0.7810 | 0.8382 |
0.3773 | 3.0 | 67500 | 0.5609 | 0.7830 | 0.7856 | 0.7843 | 0.8402 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0
- Downloads last month
- 6
Finetuned from
Evaluation results
- Precision on pos_tag_100kvalidation set self-reported0.783
- Recall on pos_tag_100kvalidation set self-reported0.786
- F1 on pos_tag_100kvalidation set self-reported0.784
- Accuracy on pos_tag_100kvalidation set self-reported0.840