Kot5 News Summarization

🚀 KoT5_news_summarization

This model is designed for news summarization, offering efficient and accurate summaries.

🚀 Quick Start

Basic Usage

# Python Code
from transformers import AutoTokenizer
from transformers import AutoModelForSeq2SeqLM

tokenizer = AutoTokenizer.from_pretrained("noahkim/KoT5_news_summarization")
model = AutoModelForSeq2SeqLM.from_pretrained("noahkim/KoT5_news_summarization")

✨ Features

This model is a lcw99/t5-base-korean-text-summary finetuned on the daekeun-ml/naver-news-summarization-ko.

📚 Documentation

Model Description

<<20221021 Commit>>

To create a model specialized for news summarization for the project, the lcw99's t5-base-korean-text-summary model was further fine-tuned with the naver-news-summarization-ko dataset provided by daekeun-ml.

I plan to conduct additional training with the news data I currently have. I will continuously improve it to implement a model with good performance. Thank you.

Execution Environment

• Google Colab Pro
• CPU: Intel(R) Xeon(R) CPU @ 2.20GHz
• GPU: A100 - SXM4 - 40GB

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: 4
• mixed_precision_training: Native AMP

Training Results

Framework Versions

• Transformers 4.23.1
• Pytorch 1.12.1+cu113
• Datasets 2.6.1
• Tokenizers 0.13.1