icemelon9 commented on a change in pull request #4644: [Relay][AutoTVM] Relay
op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r382400455
##########
File path: python/tvm/autotvm/record.py
##########
@@ -119,20 +124,29 @@ def decode(row, protocol='json'):
Parameters
----------
- row: str
+ row : str
a row in the logger file
- protocol: str
+
+ protocol : str
log protocol, json or pickle
Returns
-------
- input: autotvm.tuner.MeasureInput
- result: autotvm.tuner.MeasureResult
+ ret : tuple(autotvm.tuner.MeasureInput, autotvm.tuner.MeasureResult), or
None
+ The tuple of input and result, or None if input uses old version log
format.
"""
# pylint: disable=unused-variable
+ global _old_version_warning
+
if protocol == 'json':
row = json.loads(row)
- tgt, task_name, task_args, task_kwargs, workload, config = row['i']
+ if 'v' in row and row['v'] == 0.1:
+ if _old_version_warning:
+ logger.warning("AutoTVM log version 0.1 is no longer
supported.")
Review comment:
I think it might be useful in case people haven't updated the tophub, and
wonder why the performance is bad. But yes, after we upgraded the tophub, this
log message is useless.
----------------------------------------------------------------
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:
[email protected]
With regards,
Apache Git Services
