---
license: apache-2.0
base_model: rujengelal/nepali_t5
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: nepali_t5
  results: []
---

<!-- 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. -->

# nepali_t5

This model is a fine-tuned version of [rujengelal/nepali_t5](https://huggingface.co/rujengelal/nepali_t5) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.9595
- Bleu: 5.529
- Gen Len: 15.9474

## 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: 10
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step   | Validation Loss | Bleu   | Gen Len |
|:-------------:|:-----:|:------:|:---------------:|:------:|:-------:|
| 3.5866        | 1.0   | 17734  | 3.3249          | 4.351  | 16.0208 |
| 3.4502        | 2.0   | 35468  | 3.2435          | 4.6878 | 16.0742 |
| 3.4103        | 3.0   | 53202  | 3.1736          | 4.9997 | 16.024  |
| 3.306         | 4.0   | 70936  | 3.1176          | 5.0627 | 16.1368 |
| 3.1849        | 5.0   | 88670  | 3.0681          | 5.2125 | 16.0203 |
| 3.1681        | 6.0   | 106404 | 3.0311          | 5.2869 | 15.7844 |
| 3.1283        | 7.0   | 124138 | 3.0028          | 5.3816 | 16.1057 |
| 3.0484        | 8.0   | 141872 | 2.9804          | 5.3871 | 15.9089 |
| 3.0153        | 9.0   | 159606 | 2.9638          | 5.5117 | 15.8761 |
| 3.0429        | 10.0  | 177340 | 2.9595          | 5.529  | 15.9474 |


### Framework versions

- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2