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

Could you me give more demos on simnets. #22

Open
CuriousCat-7 opened this issue Jun 4, 2018 · 2 comments
Open

Could you me give more demos on simnets. #22

CuriousCat-7 opened this issue Jun 4, 2018 · 2 comments

Comments

@CuriousCat-7
Copy link

CuriousCat-7 commented Jun 4, 2018

I use your keras example in README.md. And get only 0.9 acc in MNIST, which is equal to random guess.

What's the best way to train the plain MNIST classifier using SimNet?

Could you give more demos like experiments in.
In Deep SimNet.

BTW, the keras example has some mistakes, fixed in there

Thank you,
Neo

@orsharir
Copy link
Member

orsharir commented Jun 5, 2018

Hi,

We'll try to upload new examples based on our older works in the near future. Meanwhile, you can either follow the architecture we've discussed in the paper you've linked to, or examine the slightly different take on SimNets as a generative model that can be found in this repository (based on Caffe instead of TF): https://github.com/HUJI-Deep/Generative-ConvACs

Or

@CuriousCat-7
Copy link
Author

@orsharir
Thank you for your reply.
I'm looking for your new examples. 😶

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