The files are available on HuggingFace.
The easiest way to download and use it is to use Python (see below).
ROUGE-1: 44.5252
ROUGE-2: 22.652
ROUGE-L: 29.8866
This model is a T5 Transformers model (JDBN/t5-base-fr-qg-fquad) that was fine-tuned in french for abstractive text summarization.
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".
The base model I used is JDBN/t5-base-fr-qg-fquad (it can perform question generation, question answering and answer extraction).
I used the "t5-base" model from the transformers library to translate in french the CNN / Daily Mail summarization dataset.