[GitHub] [incubator-mxnet] tuskiomi opened a new issue #15304: Mxnet fails to properly parallelize with a ryzen CPU.

[GitBox <gi...@apache.org>]

tuskiomi opened a new issue #15304: Mxnet fails to properly parallelize with a ryzen CPU.
URL: https://github.com/apache/incubator-mxnet/issues/15304
 
 
   
   ## Description
   With a Rysen 7 2700 CPU, the libmxnet DLL fails to properly parallelize. At any one time, it's using only one core of CPU and no more, while in an exclusive CPU configuration
   
   ## Environment info (Required)
   ----------Python Info----------
   ('Version      :', '2.7.15')
   ('Compiler     :', 'MSC v.1500 32 bit (Intel)')
   ('Build        :', ('v2.7.15:ca079a3ea3', 'Apr 30 2018 16:22:17'))
   ('Arch         :', ('32bit', 'WindowsPE'))
   ------------Pip Info-----------
   ('Version      :', '18.1')
   ('Directory    :', 'C:\\Python27\\lib\\site-packages\\pip')
   ----------MXNet Info-----------
   No MXNet installed.
   ----------System Info----------
   ('Platform     :', 'Windows-10-10.0.17134')
   ('system       :', 'Windows')
   ('node         :', 'DESKTOP-33G95HI')
   ('release      :', '10')
   ('version      :', '10.0.17134')
   ----------Hardware Info----------
   ('machine      :', 'AMD64')
   ('processor    :', 'AMD64 Family 23 Model 8 Stepping 2, AuthenticAMD')
   Name
   AMD Ryzen 7 2700 Eight-Core Processor
   
   ----------Network Test----------
   Setting timeout: 10
   Timing for MXNet: https://github.com/apache/incubator-mxnet, DNS: 0.0060 sec, LOAD: 1.1130 sec.
   Timing for PYPI: https://pypi.python.org/pypi/pip, DNS: 0.0350 sec, LOAD: 0.6090 sec.
   Error open FashionMNIST: https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/dataset/fashion-mnist/train-labels-idx1-ubyte.gz, <urlopen error [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed (_ssl.c:726)>, DNS finished in 0.039999961853 sec.
   Timing for Conda: https://repo.continuum.io/pkgs/free/, DNS: 0.0360 sec, LOAD: 0.2240 sec.
   Timing for Gluon Tutorial(en): http://gluon.mxnet.io, DNS: 0.0750 sec, LOAD: 0.6050 sec.
   Timing for Gluon Tutorial(cn): https://zh.gluon.ai, DNS: 0.0400 sec, LOAD: 0.4570 sec.
   
   Package used (Python/R/Scala/Julia):
   Python
   
   ## Error Message:
   N/A
   
   ## Minimum reproducible example
   I'm using Deap Learning Studio, I detail the problem [here](https://superuser.com/questions/1451208/deep-learning-studio-only-handling-5-images-per-second)
   
   ## Steps to reproduce
   Train on the neural net
   Observe Cpu cycles with Process Hacker 2 or similar of the python executable using the most memory.
   
   ## What have you tried to solve it?
   
   1. Gosh, I wish I knew what was going on inside the DLL. I've tried finicking with the neural net...
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services