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

Pretrained models #1

Open
VigneshSrinivasan10 opened this issue Nov 25, 2019 · 3 comments
Open

Pretrained models #1

VigneshSrinivasan10 opened this issue Nov 25, 2019 · 3 comments

Comments

@VigneshSrinivasan10
Copy link

Dear Authors,

Thank you very much for the amazing work and the code!
It would be great if you could upload the pretrained models for Image classification tasks.

Thank you in advance,

@kazukiosawa
Copy link
Collaborator

@VigneshSrinivasan10
Thank you for your suggestion!
We will upload the pertained models. Please be patient as it takes time to prepare.

@VigneshSrinivasan10
Copy link
Author

@kazukiosawa
Thank you very much for your reply.

In the meantime, I trained a model on CIFAR10 dataset as described in your repository. These are the results after training the model with configs/cifar10/lenet_vogn.json setting for 100 epochs. And these are the results I get:

Eval: Average loss: 1.8339, Accuracy: 6904/10000 (69.04%)

I find that the config file: configs/cifar10/lenet_vogn.json corresponds to VOGN+BN. However, I would like to reproduce the results from the paper of the setting VOGN+momentum+BN, where you get the best results.

It would be great if you update the config file to be compatible with that setting or point me in the direction of how to do it.

Thanks!

@jmamath
Copy link

jmamath commented Aug 25, 2021

Dear authors,

I would like to know if it is still possible to get the pretrained weights ?

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

3 participants