You are viewing a plain text version of this content. The canonical link for it is here.

Posted to commits@tvm.apache.org by GitBox <gi...@apache.org> on 2021/09/02 00:33:20 UTC

[GitHub] [tvm] shingjan commented on pull request #8851: [AutoTVM] Use popenpool in local_executor

shingjan commented on pull request #8851:
URL: https://github.com/apache/tvm/pull/8851#issuecomment-910952022


   On 64-core linux machine, each did 5 runs:
   |tune_conv2d_cuda.py, n_trial = 192|This PR|Main|
   |---|---|---|
   |Local builder + Local runner|||
   |Build time|  11.16/12.47/7.88/8.40/12.28  |   11.95/7.60/10.26/2.61/8.63   |
   |Run time|  26.59/31.71/26.06/30.53/22.37  |   36.93/25.95/24.43/17.92/26.01 |
   ||||
   |Local builder + RPC runner|||
   |Build time|  10.07/10.71/10.14/11.72/8.88 |  5.33/8.73/8.57/6.19/10.40  |
   |Run time|  27.84/30.02//22.41/21.98/23.24  |  30.90/32.89/24.43/22.64/27.36  |


-- 
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.

To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org