search

yjxiong / caffe

A fork of Caffe with OpenMPI-based Multi-GPU (mainly data parallel) support for action recognition and more. More documentation please see the original readme.
http://caffe.berkeleyvision.org/
Other
551 stars 154 forks source link

I found only " the split layer gather layer reshape layer and flatten layer" are optimization.How do i optimize the layer like convolution? #211

Closed GeorgeBohw closed 6 years ago

GeorgeBohw commented 6 years ago

I found only " the split layer gather layer reshape layer and flatten layer" are optimization.How do i optimize the layer like convolution?

In my network,it just save 1G memory which is from optimization of "split layer gather layer reshape layer and flatten layer".

GeorgeBohw commented 6 years ago

In forward process,i end up with saving 1G memory.