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
snntorch multi GPU training issue #154
Comments
Hi, I notice you use this line for distributed training: device_ids = [0, 1] #your GPU index
model = torch.nn.DataParallel(model, device_ids=device_ids) |
Hi, thank you for your help. I tried with your
code and same error appeared.
aslo I udated the enviroment info aobove just in case thank you again. |
Hi, I debug with the Leaky class. You are right, The value of ...
if self.init_hidden:
self._leaky_forward_cases(mem)
self.reset = self.mem_reset(self.mem)
self.mem = self._state_fn(input_)
... You can alter them with this version: ...
if self.init_hidden:
self._leaky_forward_cases(mem)
self.reset = self.mem_reset(self.mem)
self.mem = self._build_state_function_hidden(input_)
... This works correct in my Dual-GPU workstation. |
Hi, I've pushed a PR for fixing this issue, after merging into the master branch, you can clone it and the problem will be solved! #156 |
Made the same fix for other neurons too. #161 |
Description
I'm trying to train NMNIST with snntorch using multi GPU. since snntorch is based on torch package, I thought data parrallel from torch nn should work.
here's whole code.
and here's error
I rerun this code after removing snn.Leaky layer in CNN and it worked fine. (of course the cost doesn't converge and accuracy was 0% but still it runs)
So I assume that the reason of this error is snn.Leaky layer.
I think changing
The text was updated successfully, but these errors were encountered: