Skip to main content

Browse the archive

https://github.com/Microsoft/CNTK
10 March 2023, 01:01:58 UTC
Revision 40af7bb68fa3ca245676b81938568aedefcb5ece authored by Bowen Bao on 18 September 2018, 01:10:34 UTC, committed by Bowen Bao on 18 September 2018, 01:10:34 UTC 
Improve conv performance with freedimension
 
* CPU.
  - Short-circuit the call to ComputeConvolveGeometryExplicit() when
    MKL is enabled.
  - Replace div/mod inside ComputeConvolveGeometryExplicit with
    fast_divmod.
* GPU.
  - Instead of creating a new CuDnnConvolutionEngine for every batch,
    just update the geometry related info, and try to reuse the
    workspacememory from previous run.
1 parent 1f28c7d
Raw File
Tip revision: 40af7bb68fa3ca245676b81938568aedefcb5ece authored by Bowen Bao on 18 September 2018, 01:10:34 UTC
Improve conv performance with freedimension
Tip revision: 40af7bb
CMakeLists.txt 
cmake_minimum_required(VERSION 3.8)

project(CNTK)

include(Tools/cmake/options.cmake REQUIRED)
include(Tools/cmake/common.cmake REQUIRED)
include(Tools/cmake/cntk_common.cmake REQUIRED)

add_subdirectory(bindings/csharp/swig)

include(Tools/cmake/debug.cmake REQUIRED)
back to top