search

ranyus / FedFS

Federated Feature Selection as in the paper https://arxiv.org/abs/2109.11323
MIT License
5 stars 2 forks source link

Dataset #3

Open AomineKKL opened 1 year ago

AomineKKL commented 1 year ago

What kind of dataset used? And can i used heart disease dataset from https://www.kaggle.com/datasets/sulianova/cardiovascular-disease-dataset.

ranyus commented 1 year ago

You can use any dataset you want. Just keep in mind that the code performs feature selection. Therefore the features must be already extracted from the dataset.

AomineKKL commented 1 year ago

Hi, thank you for your time and attention. When I ran the code (main_fedfs.py) I couldn't find accuracy in the output. Can you clarify this? What output should I expect.

From: Lorenzo @.> Sent: Tuesday, June 20, 2023 3:09 PM To: ranyus/FedFS @.> Cc: KONG KOK LUNG @.>; Author @.> Subject: Re: [ranyus/FedFS] Dataset (Issue #3)

You can use any dataset you want. Just keep in mind that the code performs feature selection. Therefore the features must be already extracted from the dataset.

— Reply to this email directly, view it on GitHubhttps://github.com/ranyus/FedFS/issues/3#issuecomment-1598232468, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AURLKPU434UWA3J3W6LERZ3XMFEEPANCNFSM6AAAAAAZKJS6BY. You are receiving this because you authored the thread.Message ID: @.***>

ranyus commented 1 year ago

The output of main_fedfs.py is, among other stats, the features selected that should be used to train an ML model. The training of the ML model is not part of the library since it is not in the scope of this code.

AomineKKL commented 10 months ago

Thank you, Sir, for your response.