def test_accum_cols_multikey(self):
num_rows = 12
data = rt.Dataset({
'Symb':
rt.Cat(['A', 'B'] * int(num_rows / 2)),
'Exch':
rt.Cat(['X', 'Y', 'Y', 'X'] * int(num_rows / 4)),
'Count':
rt.full(num_rows, 1.0),
'PlusMinus': [1.0, -1.0] * int(num_rows / 2),
})
data.MultiKeyCat = rt.Cat([data.Symb, data.Exch])
accum = rt.accum_cols(data.MultiKeyCat, [data.Count, data.PlusMinus],
['Count', 'PlusMinus'])
accum_expected = rt.Dataset({
'Symb': ['A', 'B', 'A', 'B'],
'Exch': ['X', 'Y', 'Y', 'X'],
'Count': [3.0, 3.0, 3.0, 3.0],
'PlusMinus': [3.0, -3.0, 3.0, -3.0],
})
accum_expected.footer_set_values('Total', {
'Exch': 'Total',
'Count': 12.0,
'PlusMinus': 0.0
})
self.assertTrue((accum == accum_expected).all(axis=None))
def test_accum_cols_noncat(self):
num_rows = 10
pointer = rt.FA([0, 1] * int(num_rows / 2))
count = rt.full(num_rows, 1.0)
accum = rt.accum_cols(pointer, count)
accum_expected = rt.Dataset({'YLabel': [0, 1], 'col0': [5.0, 5.0]})
accum_expected.footer_set_values('Total', {
'YLabel': 'Total',
'col0': 10.0
})
self.assertTrue((accum == accum_expected).all(axis=None))
def test_accum_cols(self):
num_rows = 10
data = rt.Dataset({
'Symb': rt.Cat(['A', 'B'] * int(num_rows / 2)),
'Count': rt.full(num_rows, 1.0),
'PlusMinus': [1.0, -1.0] *
int(num_rows / 2), # Added to handle edge case of zero footer
})
accum = rt.accum_cols(data.Symb, [data.Count, data.PlusMinus],
['Count', 'PlusMinus'])
accum_expected = rt.Dataset({
'Symb': ['A', 'B'],
'Count': [5.0, 5.0],
'PlusMinus': [5.0, -5.0]
})
accum_expected.footer_set_values('Total', {
'Symb': 'Total',
'Count': 10.0,
'PlusMinus': 0.0
})
self.assertTrue((accum == accum_expected).all(axis=None))
class TestNanmin:
"""Tests for the rt.nanmin function."""
@pytest.mark.parametrize(
"arg",
[
pytest.param(math.nan, id='scalar'),
pytest.param([math.nan, math.nan, math.nan, math.nan, math.nan], id='list-float'),
pytest.param({np.nan}, id='set-float', marks=pytest.mark.skip("Broken in both numpy and riptable.")),
pytest.param(np.full(100, np.nan, dtype=np.float32), id='ndarray-float'),
pytest.param(rt.full(100, np.nan, dtype=np.float32), id='FastArray-float')
# TODO: Ordered Categorical with all invalids
]
)
def test_allnans(self, arg):
# Call rt.nanmin with the test input.
# It should raise a RuntimeWarning when given an input which
# has all NaNs **on the specified axis**.
with pytest.warns(RuntimeWarning):
result = rt.nanmin(arg)
# If given a scalar or 1D array (or some collection converted to such)
# the result should be a NaN; for higher-rank arrays, the result should
# be an array where one of the dimensions was collapsed and if there were
# all NaNs along the selected axis there'll be a NaN there in the result.
# TODO: Need to fix this to assert correctly for when rt.nanmin called with a higher-rank array.
assert rt.isnan(result)
@pytest.mark.parametrize(
"arg",
[
pytest.param([], id='list-float'),
pytest.param({}, id='set-float', marks=pytest.mark.skip("Broken in both numpy and riptable.")),
pytest.param(np.array([], dtype=np.float32), id='ndarray-float'),
pytest.param(
rt.FastArray([], dtype=np.float32), id='FastArray-float',
marks=pytest.mark.xfail(
reason="RIP-417: The call to riptide_cpp via the ledger returns None, which then causes the isnan() to raise a TypeError. This needs to be fixed so we raise an error like numpy (either by checking for this and raising the exception, or fixing the way the function punts to numpy."))
# TODO: Empty ordered Categorical (create with some categories but an empty backing array).
# TODO: Empty Date array (representing a FastArray subclass)
]
)
def test_empty(self, arg):
# Call rt.nanmin with an empty input -- it should raise a ValueError.
with pytest.raises(ValueError):
rt.nanmin(arg)
def test_unordered_categorical_disallowed(self):
"""Test which verifies rt.nanmin raises an exception if called with an unordered Categorical."""
# Create an unordered Categorical.
cat = rt.Categorical(["PA", "NY", "NY", "AL", "LA", "PA", "CA", "IL", "IL", "FL", "FL", "LA"], ordered=False)
assert not cat.ordered
with pytest.raises(ValueError):
rt.nanmin(cat)
@pytest.mark.xfail(reason="RIP-417: nanmin does not (yet) support returning a category scalar.")
def test_ordered_categorical_returns_scalar(self):
"""
Test which verifies rt.nanmin returns a scalar (Python object or numpy scalar) representing the min Category given an ordered Categorical.
"""
# Create an ordered Categorical (aka 'ordinal').
cat = rt.Categorical(
["PA", "NY", "", "NY", "AL", "LA", "PA", "", "CA", "IL", "IL", "FL", "FL", "LA"], ordered=True)
assert cat.ordered
result = rt.nanmin(cat)
# The result should either be a Python string, a numpy string scalar, or a Categorical scalar (if we implement one).
is_py_str = isinstance(result, (bytes, str))
is_np_scalar = isinstance(result, np.str)
is_rt_cat = isinstance(result, rt.Categorical)
assert is_py_str or is_np_scalar or is_rt_cat
# Check the result is correct.
assert result == "PA"
def test_accum_ratiop(self):
num_rows = 12
data = rt.Dataset({
'Symb':
rt.Cat(['A', 'A', 'A', 'B'] * int(num_rows / 4)),
'Exch':
rt.Cat(['Z', 'Z', 'X', 'X'] * int(num_rows / 4)),
'Count':
rt.full(num_rows, 1.0),
})
# Invalid input
with self.assertRaises(
ValueError,
msg=f'Failed to raise an error when passing invalid norm_by arg'
):
rt.accum_ratiop(data.Symb, data.Exch, data.Count, norm_by='z')
# Ratio within total
accum = rt.accum_ratiop(data.Symb, data.Exch, data.Count, norm_by='T')
accum_expected = rt.Dataset({
'Symb': ['A', 'B'],
'X': [25.0, 25.0],
'Z': [50.0, 0.0],
'TotalRatio': [75.0, 25.0],
'Total': [9.0, 3.0],
})
accum_expected.footer_set_values(
'TotalRatio',
{
'Symb': 'TotalRatio',
'X': 50.0,
'Z': 50.0,
'TotalRatio': 100.0
},
)
accum_expected.footer_set_values('Total', {
'Symb': 'Total',
'X': 6.0,
'Z': 6.0,
'Total': 12.0
})
self.assertTrue((accum == accum_expected).all(axis=None))
# Ratio within columns
accum = rt.accum_ratiop(data.Symb, data.Exch, data.Count, norm_by='c')
accum_expected = rt.Dataset({
'Symb': ['A', 'B'],
'X': [50.0, 50.0],
'Z': [100.0, 0.0],
'TotalRatio': [75.0, 25.0],
'Total': [9.0, 3.0],
})
accum_expected.footer_set_values(
'TotalRatio',
{
'Symb': 'TotalRatio',
'X': 100.0,
'Z': 100.0,
'TotalRatio': 100.0
},
)
accum_expected.footer_set_values('Total', {
'Symb': 'Total',
'X': 6.0,
'Z': 6.0,
'Total': 12.0
})
self.assertTrue((accum == accum_expected).all(axis=None))