---
license: mit
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: ViT5_2048
  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. -->

# ViT5_2048

This model is a fine-tuned version of [VietAI/vit5-base-vietnews-summarization](https://huggingface.co/VietAI/vit5-base-vietnews-summarization) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.492714524269104
- Rouge-1: 0.3152
- Rouge-2: 0.171
- Rouge-4: 0.1004
- Rouge-l: 0.271
- Rouge-w-1.2: 0.1115
- Rouge-s4: 0.1296
- Rouge-su4: 0.1612
- R: 0.2638
- P: 0.3914
- Bleu: 14.786

## 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: 1e-05
- train_batch_size: 1
- eval_batch_size: 1
- 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 | Rouge-1 | Rouge-2 | Rouge-4 | Rouge-l | Rouge-w-1.2 | Rouge-s4 | Rouge-su4 | R      | P      | Bleu    |
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:-------:|:-----------:|:--------:|:---------:|:------:|:------:|:-------:|
| 0.578         | 1.0   | 629  | 0.5057          | 0.1566  | 0.1075  | 0.074   | 0.1438  | 0.0597      | 0.0896   | 0.1009    | 0.119  | 0.2286 | 2.2078  |
| 0.5241        | 2.0   | 1258 | 0.4915          | 0.3766  | 0.2443  | 0.1622  | 0.342   | 0.1442      | 0.1993   | 0.2293    | 0.3074 | 0.486  | 19.4445 |
| 0.4722        | 3.0   | 1887 | 0.4928          | 0.3677  | 0.2403  | 0.16    | 0.3336  | 0.1439      | 0.1963   | 0.2254    | 0.3038 | 0.4656 | 19.0032 |


### Framework versions

- Transformers 4.26.1
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3