metadata
license: mit
base_model: sagorsarker/bangla-bert-base
tags:
- generated_from_trainer
metrics:
- f1
- accuracy
model-index:
- name: bangla-bert-base-MLTC-1
results: []
bangla-bert-base-MLTC-1
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: 0.3627
- F1: 0.8553
- Roc Auc: 0.8521
- Accuracy: 0.5707
- Hamming Loss: 0.1478
- Jaccard Score: 0.7473
- Zero One Loss: 0.4293
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: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy | Hamming Loss | Jaccard Score | Zero One Loss |
|---|---|---|---|---|---|---|---|---|---|
| 0.3717 | 1.0 | 146 | 0.3740 | 0.8447 | 0.8438 | 0.5398 | 0.1562 | 0.7312 | 0.4602 |
| 0.3812 | 2.0 | 292 | 0.3627 | 0.8373 | 0.8420 | 0.5476 | 0.1581 | 0.7201 | 0.4524 |
| 0.2373 | 3.0 | 438 | 0.3830 | 0.8450 | 0.8386 | 0.5476 | 0.1613 | 0.7316 | 0.4524 |
| 0.1688 | 4.0 | 584 | 0.3610 | 0.8555 | 0.8534 | 0.5758 | 0.1465 | 0.7475 | 0.4242 |
| 0.153 | 5.0 | 730 | 0.3627 | 0.8553 | 0.8521 | 0.5707 | 0.1478 | 0.7473 | 0.4293 |
Framework versions
- Transformers 4.41.1
- Pytorch 2.1.2
- Datasets 2.19.1
- Tokenizers 0.19.1