How to properly run the 3Dgan Keras model?

[akrzgc]

Hello,

I'd like to know how to properly run the Keras based 3Dgan model? Which of the branches is recommended? It seems that master branch was not developed for a long time.
I've tried to run the model on Nvidia V100 with TF2 and Keras2 but I'v got some syntax errors.
Please, let me know if there are any guidelines within repository about how to run the model in proper way.

Thanks and regards,
Adam

[svalleco]

Hello Adam This repo is developing in different branches Master branch does not work on TF2. You can try https://github.com/svalleco/3Dgan/tree/frehm/Conv2D <https://github.com/svalleco/3Dgan/tree/frehm/Conv2D> But the network architecture is different. We still solve a 3D problem, but use 2D convolutions to do so Cheers Sofia

…

On 7 Dec 2021, at 19:46, akrzgc ***@***.***> wrote: Hello, I'd like to know how to properly run the Keras based 3Dgan model? Which of the branches is recommended? It seems that master branch was not developed for a long time. I've tried to run the model on Nvidia V100 with TF2 and Keras2 but I'v got some syntax errors. Please, let me know if there are any guidelines within repository about how to run the model in proper way. Thanks and regards, Adam — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#10>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACABGKXH6RHAYIX5VEJH4DLUPZI7DANCNFSM5JR6MSRA>. Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>.

[akrzgc]

Hello Sofia,

Thank you for such a quick answer!
It is not exactly my requirement to run the model with TF2. I'd like to know how to run the model in proper way - if it needs TF1 it is not a problem :)

Ideally I'd like to know:

• which branch is the recommended one?
• what are the packages versions needed for running the training on the selected branch?
• which of the scripts runs the 3Dgan model (there are few scripts with "Train" in the name on different branches)?
• how can I get the original dataset?
• if the dataset is not public - what should be the shape and size of synthetic random generated dataset?

Thanks and regards,
Adam