Permalink
Commits on Jan 23, 2017
  1. fix (#206)

    piiswrong committed on GitHub Jan 23, 2017
  2. fix (#205)

    piiswrong committed on GitHub Jan 23, 2017
  3. fix openmp for windows (#204)

    piiswrong committed on GitHub Jan 23, 2017
Commits on Jan 22, 2017
  1. add parallism in function MapPacketPlan, MapReduceKeepLowest (#203)

    * add parallism in function MapPacketPlan, MapReduceKeepLowest and MapReduceKeepHighDim
    
    * Update packet-inl.h
    
    * Update tensor_cpu-inl.h
    piiswrong committed on GitHub Jan 22, 2017
Commits on Jan 19, 2017
Commits on Jan 7, 2017
  1. Add post kernel check + add more known gpu archs + fix warning (#196)

    * Add post kernel check
    
    * Revise
    
    * Revise message
    
    * Add more known gpu archs
    
    * Add 52
    
    * Fix warning
    sxjscience committed on GitHub Jan 7, 2017
Commits on Jan 6, 2017
  1. fix the -G in cuda same effect in release (#198)

    use set CUDA_NVCC_FLAGS_DEBUG
    yajiedesign committed with tqchen Jan 6, 2017
  2. Typo (#199)

    * Fix typo
    
    * Fix typo
    
    * typo
    mzhang001 committed with tqchen Jan 6, 2017
Commits on Jan 1, 2017
Commits on Dec 30, 2016
  1. fix (#193)

    piiswrong committed on GitHub Dec 30, 2016
Commits on Dec 24, 2016
Commits on Dec 19, 2016
Commits on Dec 18, 2016
  1. Add MSHADOW_FORCE_STREAM preprocessor define (#186)

    This change brings the CMake build closer to parity with the upstream
    Makefile, which unconditionally enables the MSHADOW_FORCE_STREAM preprocessor define at
    the MXNet level. This is kept within the USE_CUDA condition based on the
    assumption that it only impacts GPU training.
    alextnewman committed with piiswrong Dec 18, 2016
  2. fix compile error with cudnn v4 (#183)

    add cuda debug with debug
    yajiedesign committed with piiswrong Dec 18, 2016
Commits on Dec 17, 2016
  1. Softmax omp support (#184)

    * add omp support for softmax
    
    * fix lint issue
    
    * remove omp for mapplan
    
    * remove omp for mapplan
    
    * remove MKL guard
    zhenlinluo committed with tqchen Dec 17, 2016
Commits on Nov 23, 2016
  1. Fix compile error (#181)

    The Eval(...) function takes a pointer to Tensor as the first argument, not the Tensor itself.
    yllan committed with sxjscience Nov 23, 2016
  2. Merge pull request #182 from sxjscience/revision

    Includes:
    1. transpose_indices
    2. Fix a bug in SortByKey on CPU
    3. Use float for start/stop/step for the range OP
    4. index_fill
    sxjscience committed on GitHub Nov 23, 2016
  3. Add transpose_indices

    Fix SortByKey on CPU
    
    Use float for start/stop/step
    
    Add index_fill
    
    Remove debugging output
    
    remove debugging info
    
    fix error
    
    fix doc + lint
    
    fix lint
    
    update comment
    sxjscience committed Nov 15, 2016
Commits on Nov 22, 2016
  1. fix pip install

    tqchen committed on GitHub Nov 22, 2016
Commits on Nov 20, 2016
  1. lint fix (#178)

    piiswrong committed on GitHub Nov 20, 2016
Commits on Nov 3, 2016
  1. Add layout flags (#169)

    * Add layout flags
    
    * fix
    piiswrong committed with tqchen Nov 3, 2016
Commits on Nov 1, 2016
  1. Fix warning in clang (#175)

    tqchen committed on GitHub Nov 1, 2016
Commits on Oct 27, 2016
Commits on Oct 17, 2016
  1. Merge pull request #172 from dmlc/pluskid-crop-ex

    Allow crop to be on lhs
    pluskid committed on GitHub Oct 17, 2016
  2. Allow crop to be on lhs

    Allow `crop(a, ...) = b` expression.
    pluskid committed on GitHub Oct 17, 2016
Commits on Oct 13, 2016
  1. Merge pull request #170 from Lorrainexun/master

    Add Trinary op template in mshadow
    winstywang committed on GitHub Oct 13, 2016
Commits on Oct 12, 2016
  1. fix last whitespace

    Lorrainexun committed Oct 12, 2016
  2. fixed llllint

    Lorrainexun committed Oct 12, 2016
  3. Merge branch 'master' into master

    Lorrainexun committed on GitHub Oct 12, 2016
  4. fixed lint problems

    Lorrainexun committed Oct 12, 2016