T5 Base Fr Sum Cnndm

🚀 French T5 Abstractive Text Summarization

This project offers a fine - tuned T5 model in French for abstractive text summarization, with improved performance in version 2.0.

🚀 Quick Start

Version 2.0 of the model is now available, with improved performance compared to version 1.0. The model was trained on 13 times more data than v1, achieving ROUGE - 1: 44.5252, ROUGE - 2: 22.652, and ROUGE - L: 29.8866.

✨ Features

• Fine - tuned T5 model in French for abstractive text summarization.
• Trained on a large amount of data to improve summarization performance.

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

from transformers import T5Tokenizer, T5ForConditionalGeneration
tokenizer = T5Tokenizer.from_pretrained("plguillou/t5-base-fr-sum-cnndm")
model = T5ForConditionalGeneration.from_pretrained("plguillou/t5-base-fr-sum-cnndm")

To summarize an ARTICLE, just modify the string like this : "summarize: ARTICLE".

📚 Documentation

Model description

This model is a T5 Transformers model (JDBN/t5-base-fr-qg-fquad) that was fine - tuned in French for abstractive text summarization.

Training data

The base model used is JDBN/t5-base-fr-qg-fquad (it can perform question generation, question answering and answer extraction). The "t5 - base" model from the transformers library was used to translate the CNN / Daily Mail summarization dataset into French.