Would an e.g. bm_dict.py in [1] be a good place for a few benchmarks of dict; or is there a more appropriate project for authoritatively measuring performance regressions and optimizations of core {cpython,} data structures?
[1] https://github.com/python/pyperformance/tree/master/pyperformance/benchmarks
(pytest-benchmark looks neat, as well. an example of how to use pytest.mark.parametrize to capture multiple metrics might be helpful: https://github.com/ionelmc/pytest-benchmark )
Its easy to imagine a bot that runs some or all performance benchmarks on a PR when requested in a PR comment; there's probably already a good way to do this?
On Wed, Sep 16, 2020, 10:44 PM Wes Turner wes.turner@gmail.com wrote:
That sounds like a worthwhile optimization. FWIW, is this a bit simpler but sufficient?:
python -m timeit -n 2000 --setup "from uuid import uuid4; \ o = {uuid4().hex: i for i in range(10000)}" \ "dict(**o)"
Is there a preferred tool to comprehensively measure the performance impact of a PR (with e.g. multiple contrived and average-case key/value sets)?
On Wed, Sep 16, 2020, 7:07 PM Marco Sulla Marco.Sulla.Python@gmail.com wrote:
Well, it seems ok now: https://github.com/python/cpython/compare/master...Marco-Sulla:master
I've done a quick speed test and speedup is quite high for a creation using keywods or a dict with "holes": about 30%:
python -m timeit -n 2000 --setup "from uuid import uuid4 ; o = {str(uuid4()).replace('-', '') : str(uuid4()).replace('-', '') for i in range(10000)}" "dict(**o)"
python -m timeit -n 10000 --setup "from uuid import uuid4 ; o = {str(uuid4()).replace('-', '') : str(uuid4()).replace('-', '') for i in range(10000)} ; it = iter(o) ; key0 = next(it) ; o.pop(key0)" "dict(o)"
Can I do a PR? _______________________________________________ Python-ideas mailing list -- python-ideas@python.org To unsubscribe send an email to python-ideas-leave@python.org https://mail.python.org/mailman3/lists/python-ideas.python.org/ Message archived at https://mail.python.org/archives/list/python-ideas@python.org/message/QWXD2D... Code of Conduct: http://python.org/psf/codeofconduct/