---
license: mit
base_model: facebook/mbart-large-50
tags:
- generated_from_trainer
datasets:
- tiagoblima/du-qg-squadv1_pt
model-index:
- name: mbart50-qg-ap-peft-oficial
  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. -->

# mbart50-qg-ap-peft-oficial

This model is a fine-tuned version of [facebook/mbart-large-50](https://huggingface.co./facebook/mbart-large-50) on the tiagoblima/du-qg-squadv1_pt dataset.
It achieves the following results on the evaluation set:
- Loss: 2.5230

## 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: 0.0001
- train_batch_size: 32
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-06
- lr_scheduler_type: linear
- num_epochs: 5.0

### Training results

| Training Loss | Epoch | Step  | Validation Loss |
|:-------------:|:-----:|:-----:|:---------------:|
| 2.8113        | 1.0   | 2367  | 2.6544          |
| 2.5839        | 2.0   | 4734  | 2.5805          |
| 2.4789        | 3.0   | 7101  | 2.5482          |
| 2.4232        | 4.0   | 9468  | 2.5294          |
| 2.3971        | 5.0   | 11835 | 2.5230          |


### Framework versions

- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.15.0
- Tokenizers 0.15.1