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.

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

Jump to bottom

Fix CPU implementation of roi_align_op backward #18728

Merged
merged 1 commit into from
Jul 22, 2019
Merged

Fix CPU implementation of roi_align_op backward #18728

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
Fix CPU implementation of roi_align_op backward 
test=develop
  • Loading branch information
@qingqing01
qingqing01 committed Jul 22, 2019
commit 34894799090e0998c90dc71b703d46d14739aed6
6 changes: 6 additions & 0 deletions paddle/fluid/operators/roi_align_op.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -154,6 +154,8 @@ class CPUROIAlignOpKernel : public framework::OpKernel<T> {
int width = in_dims[3];
int rois_num = rois->dims()[0];

if (rois_num == 0) return;

auto in_stride = framework::stride(in_dims);
auto roi_stride = framework::stride(rois->dims());
auto out_stride = framework::stride(out->dims());
Expand Down Expand Up @@ -278,6 +280,10 @@ class CPUROIAlignGradOpKernel : public framework::OpKernel<T> {
const T* out_grad_data = out_grad->data<T>();
T* in_grad_data = in_grad->mutable_data<T>(ctx.GetPlace());

auto& dev_ctx = ctx.template device_context<DeviceContext>();
math::SetConstant<DeviceContext, T> set_zero;
set_zero(dev_ctx, in_grad, static_cast<T>(0));

auto in_stride = framework::stride(in->dims());
auto roi_stride = framework::stride(rois->dims());
auto out_stride = framework::stride(out_grad->dims());
Expand Down