Pypete - Python Performance Test

October 28, 2016 · View on GitHub

This project is no longer maintained. Consider using much more mature pytest-benchmark for py.test.

This plugin helps writing performance tests in traditional nosetests way. To turn it use the --with-pypete argument in running nosetests in command line.

Plugin run tests number times and this experiment is repeated repeat times. So for measurement test will be ran times x repeat times. If number equals 0, plugin computes optimal number of tests so the time of each experiment is bigger than threshold.

I recommend to use PrettyTable <https://code.google.com/p/prettytable/>_ for better overview of test results. You can select file, where the results will be stored in json format. With file and prettytable you can see comparison of current, last, best and worst run of tests (best and worst are according to avg value).

You can access all source codes at my Github <https://github.com/Artimi/pypete>_.

Install

pip install pypete

Example of usage:

$ nosetests --with-pypete --pypete-prettytable --pypete-file pypete.json tests/tests.py
F..F..
======================================================================
FAIL: test_fail (tests.BasicTest)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/psebek/projects/pypete/tests/tests.py", line 18, in test_fail
    self.assertTrue(False)
AssertionError: False is not true

======================================================================
FAIL: test_timed (tests.BasicTest)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/usr/lib/python2.7/site-packages/nose/tools/nontrivial.py", line 100, in newfunc
    raise TimeExpired("Time limit (%s) exceeded" % limit)
TimeExpired: Time limit (0.001) exceeded

Pypete results:
repeat = 3 and number = 0
test_fail (tests.BasicTest):
+--------+-------------+----------+----------+-----------+
| Metric | current [s] | last [s] | best [s] | worst [s] |
+--------+-------------+----------+----------+-----------+
|  best  |   0.000034  | 0.000033 | 0.000033 |  0.000033 |
|  avg   |   0.000037  | 0.000033 | 0.000033 |  0.000036 |
| worst  |   0.000039  | 0.000034 | 0.000034 |  0.000040 |
+--------+-------------+----------+----------+-----------+
test_time (tests.BasicTest):
+--------+-------------+----------+----------+-----------+
| Metric | current [s] | last [s] | best [s] | worst [s] |
+--------+-------------+----------+----------+-----------+
|  best  |   0.001184  | 0.001174 | 0.001158 |  0.001193 |
|  avg   |   0.001186  | 0.001184 | 0.001176 |  0.001193 |
| worst  |   0.001189  | 0.001189 | 0.001196 |  0.001193 |
+--------+-------------+----------+----------+-----------+
test_time2 (tests.BasicTest):
+--------+-------------+----------+----------+-----------+
| Metric | current [s] | last [s] | best [s] | worst [s] |
+--------+-------------+----------+----------+-----------+
|  best  |   0.002216  | 0.002229 | 0.002144 |  0.002257 |
|  avg   |   0.002237  | 0.002238 | 0.002179 |  0.002276 |
| worst  |   0.002252  | 0.002245 | 0.002201 |  0.002302 |
+--------+-------------+----------+----------+-----------+
test_timed (tests.BasicTest):
+--------+-------------+----------+----------+-----------+
| Metric | current [s] | last [s] | best [s] | worst [s] |
+--------+-------------+----------+----------+-----------+
|  best  |   0.010448  | 0.010492 | 0.010442 |  0.010490 |
|  avg   |   0.010465  | 0.010541 | 0.010470 |  0.010645 |
| worst  |   0.010474  | 0.010621 | 0.010492 |  0.010751 |
+--------+-------------+----------+----------+-----------+
tests.test_arguments(0.001,):
+--------+-------------+----------+----------+-----------+
| Metric | current [s] | last [s] | best [s] | worst [s] |
+--------+-------------+----------+----------+-----------+
|  best  |   0.001188  | 0.001210 | 0.001150 |  0.001205 |
|  avg   |   0.001202  | 0.001213 | 0.001186 |  0.001219 |
| worst  |   0.001213  | 0.001217 | 0.001205 |  0.001227 |
+--------+-------------+----------+----------+-----------+
tests.test_arguments(0.002,):
+--------+-------------+----------+----------+-----------+
| Metric | current [s] | last [s] | best [s] | worst [s] |
+--------+-------------+----------+----------+-----------+
|  best  |   0.002274  | 0.002250 | 0.002183 |  0.002250 |
|  avg   |   0.002280  | 0.002262 | 0.002204 |  0.002286 |
| worst  |   0.002288  | 0.002273 | 0.002222 |  0.002318 |
+--------+-------------+----------+----------+-----------+

----------------------------------------------------------------------
Ran 6 tests in 1.619s

FAILED (failures=2)