metadata
base_model: csebuetnlp/banglat5
tags:
- generated_from_trainer
model-index:
- name: BConvQA
results: []
BConvQA
This model is a fine-tuned version of csebuetnlp/banglat5 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.1670
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: 5e-05
- train_batch_size: 3
- eval_batch_size: 3
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.9883 | 0.15 | 10000 | 1.7429 |
| 1.6691 | 0.3 | 20000 | 1.4494 |
| 1.5742 | 0.45 | 30000 | 1.3819 |
| 1.46 | 0.59 | 40000 | 1.3138 |
| 1.4805 | 0.74 | 50000 | 1.2829 |
| 1.4381 | 0.89 | 60000 | 1.2696 |
| 1.3504 | 1.04 | 70000 | 1.2356 |
| 1.3064 | 1.19 | 80000 | 1.2144 |
| 1.2817 | 1.34 | 90000 | 1.2073 |
| 1.2709 | 1.48 | 100000 | 1.2058 |
| 1.2348 | 1.63 | 110000 | 1.1814 |
| 1.1828 | 1.78 | 120000 | 1.1848 |
| 1.324 | 1.93 | 130000 | 1.1670 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.1+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1