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 2020/11/15 19:51:59 UTC
[GitHub] [incubator-tvm] merrymercy commented on a change in pull request #6909: Make AutoScheduler handling of errors during measure consistent with AutoTvm
merrymercy commented on a change in pull request #6909:
URL: https://github.com/apache/incubator-tvm/pull/6909#discussion_r523804433
##########
File path: src/auto_scheduler/measure.cc
##########
@@ -267,7 +269,11 @@ Array<MeasureResult> ProgramMeasurerNode::Measure(const SearchTask& task,
}
if (error_ct > max_continuous_error) {
- LOG(FATAL) << "Too many errors happened during tuning";
+ LOG(WARNING) << "Too many errors happened during tuning. Switching to debug mode."
+ << std::endl;
+ verbose = 1;
Review comment:
```suggestion
verbose = 2;
```
In this PR (https://github.com/apache/incubator-tvm/pull/6882), we changed the verbosity level. Now the debug model is `verboes=2`
----------------------------------------------------------------
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