Finetuning dataset?

#2
by ismaelR - opened

Hello,

I'm currently working on french dialogue summarization and I'm doing experiments finetuning mBarthez. I noticed that you have recently uploaded this model and was hoping you could provide me with some additional information.

Could you tell me the specific dataset that you used for the finetuning? For my part, I'm using a French-translated version of the SamSum dataset. Nevertheless, I'm curious as to whether you've made use of a different dataset?

Also, did you initiate your training with the basic mBarthez model, or did you instead start with a version of mBarthez that had already been trained on the OrangeSum Abstract dataset?

Thank you for your time!
Ismaël

Sign up or log in to comment