Hi all,
I am getting difference in model execution time calculated using time_evaluator(‘run’,ctx,1,1000) and sum of layer wise execution time(from the log while executing debug graph). with time_evaluator: 52.6 (ms) with layer wise : 60.3 (ms)
can any one explain why there is difference and which one is correct measurement of execution time?
Are you talking about debug graph runtime? If so, did you check if there have duplicated lines in the log? If the info of one layer is too long, then debug graph runtime will separate it to two lines to make the print pretty. However, the execution time of that layer will also be printed twice in this case, so it’s possible that the sum of each printed line does not match the total execution time.