def test_quantile_broadcast_1():
broadcast_tester(qt.quantile,
"(n),()->()",
otype="float64",
excluded=(1, ),
dtype=np.float_,
elements=[floats_, probs])
def test_linear_rescale_broadcast():
def clean_up(args):
X, lb0, ub0, lb1, ub1, enforce_bounds = args
enforce_bounds = enforce_bounds >= 0
# Ideally, hypothesis should be able to handle constraints like this
lb0, ub0 = pair_sort(lb0, ub0)
lb1, ub1 = pair_sort(lb1, ub1)
assume(np.all(lb0 < ub0))
assume(np.all(lb1 <= ub1))
if enforce_bounds:
X = np.clip(X, lb0, ub0)
return X, lb0, ub0, lb1, ub1, enforce_bounds
broadcast_tester(
np_util.linear_rescale,
"(),(),(),(),(),()->()",
"float64",
excluded=(5, ),
map_=clean_up,
min_value=-1000,
max_value=1000,
)
def test_decode_broadcast_float():
broadcast_tester(
sp.decode,
"(m,n),(n),()->(m)",
otype=CAT_DTYPE,
excluded=(1, 2),
dtype=[np.float_, CAT_DTYPE, np.bool_],
elements=[floats(), from_dtype(np.dtype(CAT_DTYPE)), booleans()],
unique=[False, True, False],
min_side={"n": 1},
map_=decoder_gen_broadcast,
)
def test_encode_broadcast_float():
broadcast_tester(
sp.encode,
"(),(n),(),(),()->(n)",
otype=float,
excluded=(1, 2, 3, 4),
dtype=[np.int_, CAT_DTYPE, np.bool_, object, np.bool_],
elements=[integers(0, INT_MAX), from_dtype(np.dtype(CAT_DTYPE)), booleans(), just("float"), booleans()],
unique=[False, True, False, False, False],
min_side={"n": 1},
map_=encoder_gen,
)
def test_biexp_broadcast():
broadcast_tester(sp.biexp, "()->()", otype=float, min_value=-10, max_value=10)
def test_bilog_broadcast():
broadcast_tester(sp.bilog, "()->()", otype=float)
def test_order_stats_broadcast():
broadcast_tester(order_stats_trim,
"(n)->(n)",
otype="float64",
dtype=np.float_,
elements=floats_)
def test_isclose_broadcast():
broadcast_tester(np_util.isclose_lte,
"(),()->()",
otype="bool",
min_value=-1000,
max_value=1000)
