This is an automated email from the ASF dual-hosted git repository.
junrushao pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git
The following commit(s) were added to refs/heads/main by this push:
new 2708b6ca02 [MetaSchedule][Minor] Fix EvaluatorConfig Argument
Description (#11766)
2708b6ca02 is described below
commit 2708b6ca024d8e328151d9beb237b9852093cff6
Author: Xiyou Zhou <[email protected]>
AuthorDate: Fri Jun 17 12:47:30 2022 -0700
[MetaSchedule][Minor] Fix EvaluatorConfig Argument Description (#11766)
Pointed out by @sunggg that the description of `number` and `repeat` for
evaluator configuration is not accurate, updated to a version more consistent
with `TimeEvaluator`.
---
python/tvm/meta_schedule/runner/config.py | 9 +++++++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/python/tvm/meta_schedule/runner/config.py
b/python/tvm/meta_schedule/runner/config.py
index 585b88ed99..e3b5364dee 100644
--- a/python/tvm/meta_schedule/runner/config.py
+++ b/python/tvm/meta_schedule/runner/config.py
@@ -28,9 +28,14 @@ class EvaluatorConfig(NamedTuple):
Parameters
----------
number: int
- The number of runs.
+ The number of times to run this function for taking average.
+ We call these runs as one `repeat` of measurement.
repeat: int
- The number of times to repeat in each run.
+ The number of times to repeat the measurement.
+ In total, the function will be invoked (1 + number x repeat) times,
+ where the first one is warm up and will be discarded.
+ The returned result contains `repeat` costs,
+ each of which is an average of `number` costs.
min_repeat_ms: int
Minimum repeat time in ms. if the execution latency is too short,
increase the number of runs to the given time (in ms) to reduce the
measurement error.
