def test_statistical_inefficiency(self, dhdl):
'''Test if extract_u_nk assign the attr correctly'''
dataset = load_benzene()
dhdl = extract_dHdl(dataset['data']['Coulomb'][0], 310)
new_dhdl = statistical_inefficiency(dhdl)
assert new_dhdl.attrs['temperature'] == 310
assert new_dhdl.attrs['energy_unit'] == 'kT'
def test_subsample_bounds_and_step(self, gmx_ABFE):
"""Make sure that slicing the series also works
"""
subsample = statistical_inefficiency(gmx_ABFE,
gmx_ABFE.sum(axis=1),
lower=100,
upper=400,
step=2)
assert len(subsample) == 76
def slicer(self, *args, **kwargs):
return statistical_inefficiency(*args, **kwargs)
def test_duplication_on_series_noseries(self, gmx_ABFE):
duplicated = alchemlyb.concat([gmx_ABFE, gmx_ABFE])
subsample = statistical_inefficiency(duplicated.sum(axis=1),
None,
drop_duplicates=True)
assert len(subsample) == 1001
def test_duplication_on_dataframe(self, gmx_ABFE):
duplicated = alchemlyb.concat([gmx_ABFE, gmx_ABFE])
subsample = statistical_inefficiency(duplicated,
duplicated.sum(axis=1),
drop_duplicates=True)
assert len(subsample) < 1000
def test_duplication_off(self, gmx_ABFE):
duplicated = alchemlyb.concat([gmx_ABFE, gmx_ABFE])
with pytest.raises(KeyError):
statistical_inefficiency(duplicated,
duplicated.sum(axis=1),
drop_duplicates=False)
def test_sort_on_noseries(self, gmx_ABFE):
unsorted = alchemlyb.concat([gmx_ABFE[-500:], gmx_ABFE[:500]])
subsample = statistical_inefficiency(unsorted,
None,
sort=True)
assert subsample.reset_index(0)['time'].is_monotonic_increasing
def test_sort_off(self, gmx_ABFE):
unsorted = alchemlyb.concat([gmx_ABFE[-500:], gmx_ABFE[:500]])
with pytest.raises(KeyError):
statistical_inefficiency(unsorted,
unsorted.sum(axis=1),
sort=False)
def test_multiindex_duplicated(self, gmx_ABFE):
subsample = statistical_inefficiency(gmx_ABFE,
gmx_ABFE.sum(axis=1))
assert len(subsample) == 501