Skip to content

Optimize MPNCOV.py for less memory usage and faster inference #29

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.

Sign up for GitHub

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

Open
csmliu wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Optimize MPNCOV.py for less memory usage and faster inference #29

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
dfe9f98
Optimize MPNCOV.py for less memory usage and faster inference
csmliu Feb 26, 2020
4378317
Fix backward function
csmliu Mar 19, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 5 additions & 4 deletions TestCode/code/model/MPNCOV/python/MPNCOV.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,9 +21,10 @@ def forward(ctx, input):
w = x.data.shape[3]
M = h*w
x = x.reshape(batchSize,dim,M)
I_hat = (-1./M/M)*torch.ones(M,M,device = x.device) + (1./M)*torch.eye(M,M,device = x.device)
I_hat = I_hat.view(1,M,M).repeat(batchSize,1,1).type(x.dtype)
y = x.bmm(I_hat).bmm(x.transpose(1,2))
I_hat = torch.empty(M, M, device=x.device).fill_(-1./M/M)
I_hat_diag = I_hat.diagonal()
I_hat_diag += (1./M)
y = x @ I_hat @ x.transpose(1,2)
ctx.save_for_backward(input,I_hat)
return y
@staticmethod
Expand All @@ -37,7 +38,7 @@ def backward(ctx, grad_output):
M = h*w
x = x.reshape(batchSize,dim,M)
grad_input = grad_output + grad_output.transpose(1,2)
grad_input = grad_input.bmm(x).bmm(I_hat)
grad_input = grad_input @ x @ I_hat
grad_input = grad_input.reshape(batchSize,dim,h,w)
return grad_input

Expand Down