Easier readable profile statistics -- especially for caller and callee analysis
I started the implementation of this module in order to improve the profiler’s calibration using statistical methods. Unfortunately, variance proved to be far too high and my calibration trials did not converge even for large sample sets. I had to abandon this approach.
The code is still there (in case anyone would like to look what I have tried) but I do not use it.
Beside the profiler’s not so good calibration, I was not satisfied with “pstats” format especially for the caller and callee analysis and with its insistence to write to “stdout”.
Therefore, I derived a new class Stats from pstats.Stats and added new methods showStats, showCallers and showCallees which correspond to print_stats, print_callers and print_callees. They write to a file set with setOutputFile (default sys.stdout) and use a different output format. For showCallers and showCallees it is much more readable than the format used by print_callers and print_callees. I am not sure that this is also the case for the format used for showStats versus print_stats.
When you are using “Stats.print_callers” or “Stats.print_callees” and find their output unreadable, then a switch to “dm.profile.Stats” may be profitable for you.
The method showHeader outputs header information (with involved files, top level functions, number of calls and profiled time).
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
|Filename, size||File type||Python version||Upload date||Hashes|
|Filename, size dm.profile-1.0.tar.gz (5.8 kB)||File type Source||Python version None||Upload date||Hashes View hashes|