[GitHub] [tvm] tkonolige commented on a change in pull request #8807: [GRAPH EXECUTOR,VM] Add benchmarking function to graph executor and vm

[GitBox <gi...@apache.org>]

tkonolige commented on a change in pull request #8807:
URL: https://github.com/apache/tvm/pull/8807#discussion_r694164619



##########
File path: python/tvm/driver/tvmc/model.py
##########
@@ -390,29 +391,15 @@ def format_times(self):
         This has the effect of producing a small table that looks like:
         .. code-block::
             Execution time summary:
-            mean (ms)   max (ms)    min (ms)    std (ms)
-            0.14310    0.16161    0.12933    0.01004
+            mean (ms)  median (ms) max (ms)    min (ms)    std (ms)
+            0.14310      0.14310   0.16161     0.12933    0.01004
 
         Returns
         -------
         str
             A formatted string containing the statistics.
         """
-
-        # timestamps
-        mean_ts = np.mean(self.times) * 1000
-        std_ts = np.std(self.times) * 1000
-        max_ts = np.max(self.times) * 1000
-        min_ts = np.min(self.times) * 1000
-
-        header = "Execution time summary:\n{0:^10} {1:^10} {2:^10} {3:^10}".format(
-            "mean (ms)", "max (ms)", "min (ms)", "std (ms)"
-        )
-        stats = "{0:^10.2f} {1:^10.2f} {2:^10.2f} {3:^10.2f}".format(
-            mean_ts, max_ts, min_ts, std_ts
-        )
-
-        return "%s\n%s\n" % (header, stats)
+        return str(times)

Review comment:
       This is just the printing code in TVMC, it is not modifying time_evaluator at all. The only changes to the printing code is that it now outputs median.




-- 
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