-
Notifications
You must be signed in to change notification settings - Fork 270
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
How to implement ResNet? #1076
Comments
The PyTorch interface only supports sequential networks, but ResNet contains an addition and thus isn't sequential. We have implemented ResNet-50 inference, which you can run as follows from the MP-SPDZ root directory:
You can change the last line to the |
I have implemented a simple training code for residual blocks in ml.py, and I hope it may bring some motivations for you. If anyone has implemented a complete ResNet training, I am really looking forward to it being open source. class SimpleRes_Linear(DenseBase):
|
I try to write Resnet18 as this:
The error is
CompilerError: unknown PyTorch module: ResNet18.
It seems I can't pass a self-defined module to the Compiler. Is there any example of ResNet18 inference in MP-SPDZ?
The text was updated successfully, but these errors were encountered: