README.md

metadata

license: mit
base_model: PhucMap/BARTphoVi-VSL
tags:
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: mBART-vi-news-summarization
    results: []
datasets:
  - vietgpt/news_summarization_vi
language:
  - vi
pipeline_tag: summarization

mBART-vi-news-summarization

This model is a fine-tuned version of PhucMap/BARTphoVi-VSL on vietgpt/news_summarization_vi. It achieves the following results on the evaluation set:

• Loss: 0.8166
• Rouge1: 70.8598
• Rouge2: 51.2255
• Rougel: 51.8606
• Rougelsum: 65.5856
• Gen Len: 152.869

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: 0.0002
• train_batch_size: 8
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Framework versions

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