FAILED tests/test_summary.py::test_get_height
Closed this issue · 3 comments
esqew commented
One test out of the suite is currently failing:
FAILED tests/test_summary.py::test_get_height - AssertionError: assert ['Louisville',\n 'ACC',\n '76.8',\n '181',\n '+0.6',\n '87',\n '+1.3',\n '44',...
Full test log:
================================================================= test session starts ==================================================================
platform darwin -- Python 3.8.9, pytest-7.1.2, pluggy-1.0.0 -- /Applications/Xcode.app/Contents/Developer/usr/bin/python3
cachedir: .pytest_cache
rootdir: <redacted>/kenpompy
collected 17 items
tests/test_fanmatch.py::test_fanmatch PASSED [ 5%]
tests/test_misc.py::test_get_pomeroy_ratings PASSED [ 11%]
tests/test_misc.py::test_get_trends PASSED [ 17%]
tests/test_misc.py::test_get_refs PASSED [ 23%]
tests/test_misc.py::test_get_hca PASSED [ 29%]
tests/test_misc.py::test_get_arenas PASSED [ 35%]
tests/test_misc.py::test_get_gameattribs PASSED [ 41%]
tests/test_misc.py::test_get_program_ratings PASSED [ 47%]
tests/test_summary.py::test_get_efficiency PASSED [ 52%]
tests/test_summary.py::test_get_fourfactors PASSED [ 58%]
tests/test_summary.py::test_get_teamstats PASSED [ 64%]
tests/test_summary.py::test_get_pointdist PASSED [ 70%]
tests/test_summary.py::test_get_height FAILED [ 76%]
tests/test_summary.py::test_get_playerstats PASSED [ 82%]
tests/test_summary.py::test_get_kpoy PASSED [ 88%]
tests/test_team.py::test_get_valid_teams PASSED [ 94%]
tests/test_team.py::test_get_schedule PASSED [100%]
======================================================================= FAILURES =======================================================================
___________________________________________________________________ test_get_height ____________________________________________________________________
browser = <mechanicalsoup.stateful_browser.StatefulBrowser object at 0x12b014340>
def test_get_height(browser):
expected = ['Louisville', 'ACC', '76.8', '181', '+0.6', '87', '+1.3', '44', '-0.1', '190', '-0.1', '189', '-0.9',
'272', '-0.5', '233', '1.92', '102', '33.3', '97', '36.7', '273']
df = kpsum.get_height(browser, season = '2019')
> assert [str(i) for i in df[df.Team == 'Louisville'].iloc[0].to_list()] == expected
E AssertionError: assert ['Louisville',\n 'ACC',\n '76.8',\n '181',\n '+0.6',\n '87',\n '+1.3',\n '44',\n '-0.1',\n '190',\n '-0.1',\n '189',\n '-0.9',\n '272',\n '-0.5',\n '233',\n '2.18',\n '29',\n '33.3',\n '97',\n '36.7',\n '273'] == ['Louisville',\n 'ACC',\n '76.8',\n '181',\n '+0.6',\n '87',\n '+1.3',\n '44',\n '-0.1',\n '190',\n '-0.1',\n '189',\n '-0.9',\n '272',\n '-0.5',\n '233',\n '1.92',\n '102',\n '33.3',\n '97',\n '36.7',\n '273']
E At index 16 diff: '2.18' != '1.92'
E Full diff:
E [
E 'Louisville',
E 'ACC',
E '76.8',
E '181',
E '+0.6',
E '87',
E '+1.3',
E '44',
E '-0.1',
E '190',
E '-0.1',
E '189',
E '-0.9',
E '272',
E '-0.5',
E '233',
E - '1.92',
E + '2.18',
E - '102',
E ? --
E + '29',
E ? +
E '33.3',
E '97',
E '36.7',
E '273',
E ]
tests/test_summary.py:66: AssertionError
esqew commented
After further research, it seems KenPom somewhat recently changed the methodology on how Experience is quantified - from his (undated) blog post:
So a solution is needed. And that solution is to measure a player’s experience by minutes played in previous seasons. I have more or less scaled this new measure to the previous measure. The new measure can be thought of as average number of full D-1 seasons played by the current roster, weighted by minutes played.
Seems it's likely this test is failing since it's using data from the old methodology to test - it's probably just a matter of updating the constants in the test file to be asserted against.
j-andrews7 commented
Yeah I saw this one during my tests as well. You can correct the test
assertion or I will before release.
…On Mon, Aug 1, 2022, 3:12 PM Sean Quinn ***@***.***> wrote:
After further research, it seems KenPom somewhat recently changed the
methodology on how Experience is quantified - from his (undated) blog post
<https://kenpom.com/blog/are-you-experienced/>:
So a solution is needed. And that solution is to measure a player’s
experience by minutes played in previous seasons. I have more or less
scaled this new measure to the previous measure. The new measure can be
thought of as average number of full D-1 seasons played by the current
roster, weighted by minutes played.
Seems it's likely this test is failing since it's using data from the old
methodology to test - it's probably just a matter of updating the constants
in the test file to be asserted against.
—
Reply to this email directly, view it on GitHub
<#21 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACOAQNB7ZHBZYK7NZNPL2QLVXAVUBANCNFSM55IW7BPQ>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
j-andrews7 commented
Latest release (v0.3.2) fixes this, available via pip now. Thanks for the help.