Ket5 News Summarizer

🚀 News Article Summarization Model

This is a news article summarization model fine - tuned based on the lcw99/t5 - base - korean - text - summary created by lcw99. It aims to provide high - quality text summarization services.

🚀 Quick Start

This section provides a quick overview of the news article summarization model. For detailed usage, please refer to the following documentation.

✨ Features

• Based on Pretrained Model: Fine - tuned from the lcw99/t5 - base - korean - text - summary model.
• Using Specific Training Data: Trained on news articles from the 'Document Summarization Text' provided by AIHub.
• Performance Improvement: Continuously improving the model's performance.

📦 Installation

No installation steps are provided in the original document, so this section is skipped.

💻 Usage Examples

No code examples for usage are provided in the original document, so this section is skipped.

📚 Documentation

Model Description

This 'news article summarization model' is created by fine - tuning based on the lcw99/t5 - base - korean - text - summary developed by lcw99.

The training data uses news articles from the 'Document Summarization Text (https://aihub.or.kr/aihubdata/data/view.do?currMenu = 115&topMenu = 100&aihubDataSe = realm&dataSetSn = 97)' provided by AIHub.

We will continuously improve the performance and upgrade the model to a better one.

Training Arguments

training_args = Seq2SeqTrainingArguments(
    evaluation_strategy="epoch",                    
    save_strategy="epoch",                          
    save_total_limit=2,                             
    warmup_steps=1000,                             
    learning_rate=5e-5,                             
    per_device_train_batch_size=16,                
    per_device_eval_batch_size=16,               
    num_train_epochs=3,                           
    weight_decay=0.01,                             
    predict_with_generate=True,                    
    fp16=True                               
)

Training Progress

Execution Environment

• Operating System: Window 10
• GPU: NVIDIA GeForce RTX 3070, 8192 MiB

Framework Versions

🔧 Technical Details

No in - depth technical details are provided in the original document, so this section is skipped.

📄 License

The model is licensed under the Apache 2.0 license.