sprofiler 0.1.0

Lightweight profiler with checkpoints

sProfiler

Python script profiler/timer supporting code checkpoints and reporting.

Installation

pip install sprofiler

Usage

Use the Profiler to create named checkpoints throughout your code. Checkpoints need a 'start' and a 'stop', and multiple iterations are combined to summarize how long it takes to complete each leg. The report() function prints the results from all checkpoints.

import sprofiler as sp
from time import sleep

pr = sp.Profiler()

pr.start('program')
print('Code outside loop')
sleep(1)

for _ in range(10):
    pr.start('sleep_1s')
    print('Code in loop')
    sleep(1)
    pr.stop('sleep_1s')
pr.stop('program')

pr.report()

The printed report appears as:

program | 11.0 s ± 0.0 s per iteration, n = 1
sleep_1s | 1.0 s ± 0.0 s per iteration, n = 10

Logging

sProfiler automatically logs results by default. You can change the destination filename, as well as the level of verbosity: 0 - no logging, 1 - only elapsed times, 2 - start and stop times. Defaults are logname='profiler.log' and verbose=2.

import sprofiler as sp
from time import sleep

pr = sp.Profiler(logname='my_log.log', verbose=1)

Function Decorators

sProfiler also supports timing functions with decorators, via Profiler.time_func. Demonstrating on the first example:

pr = sp.Profiler()

@pr.time_func
def sleep_1s():
    print('Code in loop')
    sleep(1)

@pr.time_func
def my_func():
    for _ in range(10):
        sleep_1s()

pr.report()

The printed report appears as:

my_func | 10.0 s ± 0.0 s per iteration, n = 1
sleep_1s | 1.0 s ± 0.0 s per iteration, n = 10

Future Directions

• Potential support for more complex profiling