Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

ROC-AUC on Mvtec #16

Open
emathian opened this issue Apr 19, 2022 · 2 comments
Open

ROC-AUC on Mvtec #16

emathian opened this issue Apr 19, 2022 · 2 comments

Comments

@emathian
Copy link

Hello Alessio,
I'm Emilie, a PhD student in deep learning. I'm really interested in the anomaly detection model, and FastFlow seems really promising. I'm trying to develop my own version to make it work on medical images inspired by the following repo:

@AlessioGalluccio
Copy link
Owner

AlessioGalluccio commented Apr 19, 2022

Hi Emilie!
I haven't managed to reproduce them yet, but others had some progress,
I'm currently working on my Master thesis and, since I couldn't reproduce the results of the paper, I had to use another neural network, so I can't work on this project at this moment. The neural network I'm using is CS-Flow. It was created by the same author of Differnet, which I used in this repository as a starting point. I'm getting good results, so I advice you to use this one if you want to use Normalizing Flows.
The great problem with FastFlow is that there are ambiguities in the paper, so it's hard to implement. Look at the Q&A in the issues of this repo if you want to know more. There are some great guys who did some progress in solving these ambiguities,
Good luck for you PhD! Working on medical images is a noble mission, and contributing to OpenAI is really cool. I'm almost envious! ;)
Best,
Alessio

@emathian
Copy link
Author

The Q&A gives great input I did not see this! Thank you and good luck with your thesis :)
Best
Emilie

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants