You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Nov 17, 2023. It is now read-only.
All dmlc projects are designed to be portable. If you see related project like xgboost, they are readily run on YARN, which means on most spark cluster which runs on YARN.
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.
iblislin
pushed a commit
to iblislin/incubator-mxnet
that referenced
this issue
Mar 18, 2018
Is runing mxnet on top spark a thing that we can dream of ?
The text was updated successfully, but these errors were encountered: