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

Accuracy 76% on MNIST -> MNIST-M #1

Open
charliezon opened this issue Jul 14, 2020 · 1 comment
Open

Accuracy 76% on MNIST -> MNIST-M #1

charliezon opened this issue Jul 14, 2020 · 1 comment

Comments

@charliezon
Copy link

charliezon commented Jul 14, 2020

Thank you very much for sharing the implementation of d-SNE.

I tried to run your code without changing any config. But I can only achieve 76% compared to the 81% stated by you and 87% stated by the paper.

Do you have any idea?

@joshuacwnewton
Copy link
Owner

joshuacwnewton commented Jul 17, 2020

Hello @charliezon! Thank you for taking the time to try out my implementation and point out issues with it. (Here, and in #2 and #3.)

Right now, I am preoccupied with personal matters, so I can't actively maintain this repository. I do agree that more thorough testing and improvements are necessary, though, so I will leave the issues open.

Thanks for your understanding, and sorry I can't be of more help right now.

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