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 2020/11/29 16:54:27 UTC

[GitHub] [tvm] merrymercy opened a new pull request #6994: [AutoScheduler] Use a smaller iteration number for GA to acclerate the search

merrymercy opened a new pull request #6994:
URL: https://github.com/apache/tvm/pull/6994


   The default value of `evolutionary_search_num_iters` is 10.
   I did some experiments and found this value is too large.
   I changed it to 3 or 4 and got almost the same results on resnet-50.
   So I think we can change it to a smaller number to save some search time.
   
   cc @comaniac @jcf94 Can you also verify this for your workloads?
   


----------------------------------------------------------------
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

[GitHub] [tvm] comaniac merged pull request #6994: [AutoScheduler] Use a smaller iteration number for GA to acclerate the search

Posted by GitBox <gi...@apache.org>.
comaniac merged pull request #6994:
URL: https://github.com/apache/tvm/pull/6994


   


----------------------------------------------------------------
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

[GitHub] [tvm] merrymercy edited a comment on pull request #6994: [AutoScheduler] Use a smaller iteration number for GA to acclerate the search

Posted by GitBox <gi...@apache.org>.
merrymercy edited a comment on pull request #6994:
URL: https://github.com/apache/tvm/pull/6994#issuecomment-735465425


   @comaniac I added a new argument to control the parameters of search policy from the task scheduler.
   
   I tested n-iter =3 for resnet-50 on both CPU and GPU. I can confirm that it is enough. It can accelerate the search by 1.5x - 2x.


----------------------------------------------------------------
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

[GitHub] [tvm] merrymercy commented on pull request #6994: [AutoScheduler] Use a smaller iteration number for GA to acclerate the search

Posted by GitBox <gi...@apache.org>.
merrymercy commented on pull request #6994:
URL: https://github.com/apache/tvm/pull/6994#issuecomment-735465425


   @comaniac I added a new argument to control these parameters from the task scheduler.
   
   I tested n-iter =3 for resnet-50 on both CPU and GPU. I can confirm that it is enough. It can accelerate the search by 1.5x - 2x.


----------------------------------------------------------------
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

[GitHub] [tvm] merrymercy edited a comment on pull request #6994: [AutoScheduler] Use a smaller iteration number for GA to acclerate the search

Posted by GitBox <gi...@apache.org>.
merrymercy edited a comment on pull request #6994:
URL: https://github.com/apache/tvm/pull/6994#issuecomment-735465425


   @comaniac I added a new argument to control the parameters of search policy from the task scheduler.
   
   I tested `evolutionary_search_num_iters=3` for resnet-50 on both CPU and GPU. I can confirm that it is enough. It can accelerate the search by 1.5x - 2x.


----------------------------------------------------------------
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