Skip to content

Share convolution buffers to reduce memory usage #2016

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
shelhamer wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Share convolution buffers to reduce memory usage #2016

Show file tree
Hide file tree
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
4 changes: 3 additions & 1 deletion include/caffe/layers/base_conv_layer.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -165,7 +165,9 @@ class BaseConvolutionLayer : public Layer<Dtype> {
int col_offset_;
int output_offset_;

Blob<Dtype> col_buffer_;
// N.B. sharing the col buffer reduces memory but interferes with
// ND conv and parallelism
static Blob<Dtype> col_buffer_;
Blob<Dtype> bias_multiplier_;
};

Expand Down
3 changes: 3 additions & 0 deletions src/caffe/layers/base_conv_layer.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,6 +8,9 @@

namespace caffe {

template <typename Dtype>
Blob<Dtype> BaseConvolutionLayer<Dtype>::col_buffer_;

template <typename Dtype>
void BaseConvolutionLayer<Dtype>::LayerSetUp(const vector<Blob<Dtype>*>& bottom,
const vector<Blob<Dtype>*>& top) {
Expand Down