dmlc/mshadow issues and pull requests

#376 - Improve openblas CMake logic

Pull Request - State: closed - Opened by larroy over 5 years ago - 2 comments

#375 - Doesn't install any files

Issue - State: closed - Opened by yurivict over 5 years ago - 2 comments

#374 - Update mshadow.cmake to set BLAS

Pull Request - State: closed - Opened by yinghu5 almost 6 years ago - 7 comments

#373 - [RFC] Deprecate mshadow and donate code to apache-mxnet

Issue - State: closed - Opened by szha almost 6 years ago - 13 comments

#372 - [WIP] Tensor shape overflow checking in Blas Engine

Pull Request - State: closed - Opened by larroy almost 6 years ago - 2 comments

#371 - Add compile flag to choose data type for tensor size due to performance degradation

Pull Request - State: closed - Opened by apeforest almost 6 years ago - 14 comments

#370 - Add missing semicolon

Pull Request - State: closed - Opened by TaoLv almost 6 years ago

#369 - Add except handling in TensorContainer

Pull Request - State: closed - Opened by arcadiaphy almost 6 years ago

#368 - Add round-to-nearest-even rounding to float2half().

Pull Request - State: closed - Opened by DickJC123 about 6 years ago - 2 comments

#367 - Return size_t from shape::size().

Pull Request - State: closed - Opened by ZhennanQin about 6 years ago - 6 comments

#366 - Add fp16 support for batch_dot

Pull Request - State: closed - Opened by eric-haibin-lin about 6 years ago

#365 - [WIP] Change MKL BLAS linkage from LP64 to ILP64

Pull Request - State: closed - Opened by TaoLv about 6 years ago - 6 comments

#364 - Large array support

Pull Request - State: closed - Opened by apeforest about 6 years ago

#363 - Setting cuBLAS stream

Pull Request - State: closed - Opened by ptrendx about 6 years ago

#362 - Install CMake config

Pull Request - State: closed - Opened by ruslo over 6 years ago - 2 comments

#361 - Add Turing GPU support

Pull Request - State: closed - Opened by KellenSunderland over 6 years ago - 3 comments

#360 - Initialized Tensor::dptr_ to nullptr

Pull Request - State: closed - Opened by lebeg over 6 years ago - 2 comments

#359 - Re-Revert PR #348 to support large tensor size in MXNet

Pull Request - State: closed - Opened by apeforest over 6 years ago - 4 comments

#358 - Revert fp16 batched_dot

Pull Request - State: closed - Opened by eric-haibin-lin over 6 years ago

#357 - Revert "Allow large array operation in MXNet (#348)"

Pull Request - State: closed - Opened by azai91 over 6 years ago - 6 comments

#356 - disable omp parallel for only when CUDACC

Pull Request - State: closed - Opened by szha over 6 years ago

#355 - does there need some update with guide and tests?

Issue - State: open - Opened by wxywb over 6 years ago - 1 comment

#354 - Is 'base.h:173:20:fatal error: cuda.h: No such file or directory' means I did't install cuda?

Issue - State: open - Opened by CP-here over 6 years ago - 2 comments

#353 - Add half_t support for batch_dot.

Pull Request - State: closed - Opened by eric-haibin-lin over 6 years ago - 1 comment

#352 - Add Scientific Linux to makefile's CentOS blas library exception

Issue - State: open - Opened by elofgren over 6 years ago - 1 comment

#351 - CI is failing

Issue - State: open - Opened by apeforest over 6 years ago - 1 comment

#350 - Upgrade pip before installing packages in CI

Pull Request - State: closed - Opened by asitstands over 6 years ago - 1 comment

#349 - fix shape check of complex unitary operation

Pull Request - State: closed - Opened by arcadiaphy over 6 years ago

#348 - Allow large array operation in MXNet

Pull Request - State: closed - Opened by apeforest over 6 years ago - 8 comments

#347 - Add stable reduce merging

Pull Request - State: closed - Opened by leezu over 6 years ago

#346 - Allow array size to be greater than 2^32

Pull Request - State: closed - Opened by apeforest over 6 years ago - 6 comments

#345 - Add Finalize to reducers

Pull Request - State: closed - Opened by leezu over 6 years ago

#344 - Add path_suffixes for finding cudnn.lib

Pull Request - State: closed - Opened by nehaljwani over 6 years ago - 2 comments

#343 - fix mkl batch gemm

Pull Request - State: closed - Opened by TaoLv over 6 years ago - 9 comments

#342 - Improve batch gemm performance using MKL

Pull Request - State: closed - Opened by xinyu-intel over 6 years ago - 2 comments

#341 - Added a check for already set compilers and flags environment variables

Pull Request - State: closed - Opened by lebeg over 6 years ago

#340 - Fix broken CI

Pull Request - State: closed - Opened by asitstands almost 7 years ago - 2 comments

#339 - error: no instance of function template "mshadow::Random<mshadow::gpu, DType>::SampleUniform

Issue - State: closed - Opened by gongqiang almost 7 years ago

#338 - Improved random number generation

Pull Request - State: closed - Opened by asitstands almost 7 years ago - 2 comments

#337 - check and close fp16 with arm

Pull Request - State: closed - Opened by yajiedesign almost 7 years ago - 8 comments

#336 - Fix F16C build for cmake with certain versions of nvcc and gcc

Pull Request - State: closed - Opened by rahul003 almost 7 years ago - 1 comment

#335 - F16C build flag changes

Pull Request - State: closed - Opened by rahul003 almost 7 years ago - 2 comments

#334 - Add missing Cuda arches, Volta, TX2

Pull Request - State: closed - Opened by larroy almost 7 years ago - 1 comment

#333 - fix fp16 with msvc

Pull Request - State: closed - Opened by yajiedesign almost 7 years ago - 4 comments

#332 - Remove Fermi

Pull Request - State: closed - Opened by cjolivier01 almost 7 years ago

#331 - Remove Fermi Arch

Pull Request - State: closed - Opened by cjolivier01 almost 7 years ago - 1 comment

#330 - Leveraging F16C x86 instruction set architecture extension for half datatype

Pull Request - State: closed - Opened by rahul003 almost 7 years ago - 10 comments

#329 - dot_engine-inl.h MSHADOW_STAND_ALONE BLASEngine<cpu, float>::gemm error

Issue - State: open - Opened by semyonc almost 7 years ago - 1 comment

#328 - update to use MKLDNN option

Pull Request - State: closed - Opened by jinhuang415 almost 7 years ago - 6 comments

#327 - Update README.md / Fix a bug in README demo code

Pull Request - State: closed - Opened by erichuang0771 about 7 years ago - 3 comments

#326 - May want to define NOMINMAX at a higher level

Pull Request - State: closed - Opened by cjolivier01 about 7 years ago

#325 - Revert "Add half_t hash specialization (#324)"

Pull Request - State: closed - Opened by eric-haibin-lin about 7 years ago

#324 - Add half_t hash specialization

Pull Request - State: closed - Opened by eric-haibin-lin about 7 years ago - 1 comment

#323 - add MSHADOW_REAL_TYPE_SWITCH_WITH_HALF2

Pull Request - State: closed - Opened by Caenorst about 7 years ago - 2 comments

#322 - Fix warning, add informative error messages when cuda drivers are not configured propperly

Pull Request - State: closed - Opened by larroy about 7 years ago - 2 comments

#321 - [Clarification] Provision for AMD HIP backend for MXNet

Issue - State: open - Opened by sriharikarnam about 7 years ago - 1 comment

#320 - fix bad Plan name exp

Pull Request - State: closed - Opened by zhreshold about 7 years ago

#319 - Introduce CUDA 9 support in CMake

Pull Request - State: closed - Opened by cbalioglu about 7 years ago - 5 comments

#318 - Row wise normalization of tensor

Issue - State: open - Opened by rayjs about 7 years ago - 1 comment

#314 - Support for Complex numbers

Issue - State: open - Opened by aodhan-domhnaill about 7 years ago - 2 comments

#312 - fix the '--start-group' error on macOS

Pull Request - State: closed - Opened by lightingghost about 7 years ago - 3 comments

#311 - --start-group not supported by macOS

Issue - State: open - Opened by lightingghost about 7 years ago - 2 comments

#307 - add USE_CUSOLVER config

Pull Request - State: closed - Opened by crazy-cat over 7 years ago - 2 comments

#306 - Shall we add half_t support to Random<cpu, DType>

Issue - State: open - Opened by KiddoZhu over 7 years ago - 2 comments

#305 - Error on compiling basic_stream.cu with CUDA 9.0

Issue - State: open - Opened by vomarazabo over 7 years ago - 3 comments

#299 - enable padding with custom value, default to 0

Pull Request - State: closed - Opened by mtin over 7 years ago - 2 comments

#298 - mac make errors

Issue - State: open - Opened by 5idaidai over 7 years ago - 1 comment

#279 - why there is no kDim inference check in ExpInfo for TenaryMapExp？

Issue - State: open - Opened by zhangxiao-ustc over 7 years ago - 1 comment

#278 - how to reduce a one-dimension tensor (vector) ?

Issue - State: open - Opened by buaasun over 7 years ago - 1 comment

#259 - About getting the shape of a tensor

Issue - State: open - Opened by AIROBOTAI over 7 years ago - 1 comment

#253 - New operator defined with mshadow

Issue - State: open - Opened by yijiezh almost 8 years ago - 1 comment

#252 - "Cannot find CUDA device" error

Issue - State: open - Opened by yuconglin almost 8 years ago - 1 comment

#251 - complex tensor support

Issue - State: open - Opened by Roger-luo almost 8 years ago - 1 comment

#248 - feature request: add Tensor contraction

Issue - State: open - Opened by AuroraDysis almost 8 years ago - 1 comment

#225 - Update mshadow.mk to support custom CUDNN_PATH

Pull Request - State: closed - Opened by ghost almost 8 years ago - 2 comments

#224 - How to choose a GPU to run

Issue - State: open - Opened by Zzzians almost 8 years ago - 2 comments

#218 - update for ATLAS

Pull Request - State: closed - Opened by jli05 almost 8 years ago - 2 comments

#217 - change CXX to clang++ for use of C++1x

Pull Request - State: closed - Opened by jli05 almost 8 years ago - 2 comments

#216 - add -std=c++11

Pull Request - State: closed - Opened by jli05 almost 8 years ago - 2 comments

#213 - Random<>::SampleUniform() has inconsistant behaviours on CPU, GPU and docs

Issue - State: open - Opened by cairijun about 8 years ago - 1 comment

#208 - Fixes for Android build.

Pull Request - State: closed - Opened by lygstate about 8 years ago - 2 comments

#207 - link error when enable DEBUG in mxnet/make/config.mk

Issue - State: open - Opened by glingyan about 8 years ago - 3 comments

#195 - fix bug in fedora building, use pkg to find find the right library name

Pull Request - State: closed - Opened by sethbrin about 8 years ago - 2 comments

#187 - Disabling OpenMP parallel pragma for CPU tensors causes performance regression

Issue - State: open - Opened by alextnewman about 8 years ago - 1 comment

#185 - MSDHADOW_USE_PASCAL in shadow.mk

Issue - State: open - Opened by ungil about 8 years ago - 2 comments

#179 - Make error in mshadow/guide/neuralnet

Issue - State: open - Opened by blazerygy over 8 years ago - 5 comments

#177 - Apple, cuDNN cmake build tweaks

Pull Request - State: closed - Opened by alextnewman over 8 years ago - 2 comments

#160 - How can embedding work under cuda version 6.5

Issue - State: open - Opened by quanzongfeng over 8 years ago - 2 comments

#145 - Dot Product Implementation:missing?

Issue - State: open - Opened by marcoippolito over 8 years ago - 1 comment

#125 - half_t multiplication

Issue - State: open - Opened by vchuravy over 8 years ago - 1 comment

#99 - Some misleading syntax

Issue - State: open - Opened by Jokeren almost 9 years ago - 1 comment

#88 - Record locations of pooling region.

Issue - State: open - Opened by MonPower about 9 years ago - 2 comments

#79 - [question] why just replace the inline keyword with predefined macro?

Issue - State: open - Opened by dzhwinter about 9 years ago - 1 comment

GitHub / dmlc/mshadow issues and pull requests