def test_interaction_cont_only(cont):
interact = Interaction(cont=cont)
assert interact.nobs == cont.shape[0]
assert_frame_equal(cont, interact.cont)
expected = cont.to_numpy()
actual = interact.sparse
assert isinstance(actual, csc_matrix)
assert_allclose(expected, actual.A)
def test_interaction_cat_only(cat):
interact = Interaction(cat=cat)
assert interact.nobs == cat.shape[0]
assert_frame_equal(cat, interact.cat)
expected = category_interaction(category_product(cat), precondition=False)
actual = interact.sparse
assert isinstance(actual, csc_matrix)
assert_allclose(expected.A, actual.A)
def interact(request):
if not request.param:
return None
rs = np.random.RandomState(0)
interactions = []
for i in range(request.param):
cat = random_cat(4, 100, frame=True, rs=rs)
cont = random_cont(100, rs=rs)
interactions.append(Interaction(cat, cont))
return interactions
def test_interaction_cat_cont(cat, cont):
interact = Interaction(cat=cat, cont=cont)
assert interact.nobs == cat.shape[0]
assert_frame_equal(cat, interact.cat)
assert_frame_equal(cont, interact.cont)
base = category_interaction(category_product(cat), precondition=False).A
expected = []
for i in range(cont.shape[1]):
element = base.copy()
element[np.where(element)] = cont.iloc[:, i].to_numpy()
expected.append(element)
expected = np.column_stack(expected)
actual = interact.sparse
assert isinstance(actual, csc_matrix)
assert_allclose(expected, interact.sparse.A)
def test_interaction_cat_cont_convert(cat, cont):
base = Interaction(cat, cont)
interact = Interaction(cat.to_numpy(), cont)
assert_allclose(base.sparse.A, interact.sparse.A)
def test_empty_interaction():
interact = Interaction(nobs=100)
assert isinstance(interact.sparse, csc_matrix)
assert interact.sparse.shape == (100, 0)
def test_interaction_cat_bad_nobs():
with pytest.raises(ValueError):
Interaction()
with pytest.raises(ValueError):
Interaction(cat=np.empty((100, 0)), cont=np.empty((100, 0)))
def test_interaction_from_frame(cat, cont):
base = Interaction(cat=cat, cont=cont)
interact = Interaction.from_frame(pd.concat([cat, cont], 1))
assert_allclose(base.sparse.A, interact.sparse.A)
def generate_data(
k=3,
const=True,
nfactors=1,
factor_density=10,
nobs=2000,
cont_interactions=1,
factor_format="interaction",
singleton_interaction=False,
weighted=False,
ncont=0,
):
rs = np.random.RandomState(1234567890)
density = [factor_density] * max(nfactors, cont_interactions)
x = rs.standard_normal((nobs, k))
if const:
x = np.column_stack([np.ones(nobs), x])
e = rs.standard_normal(nobs)
y = x.sum(1) + e
factors = []
for i in range(nfactors):
ncat = nobs // density[min(i, len(density) - 1)]
fact = rs.randint(ncat, size=nobs)
effects = rs.standard_normal(ncat)
y += effects[fact]
factors.append(pd.Series(pd.Categorical(fact)))
for i in range(ncont):
cont = rs.standard_normal(size=nobs)
factors.append(pd.Series(cont))
if factors:
factors = pd.concat(factors, 1)
if factor_format == "interaction":
if nfactors and ncont:
factors = Interaction(factors.iloc[:, :nfactors],
factors.iloc[:, nfactors:])
elif nfactors:
factors = Interaction(factors, None)
else:
factors = Interaction(None, factors)
else:
factors = None
interactions = []
for i in range(cont_interactions):
ncat = nobs // density[min(i, len(density) - 1)]
fact = rs.randint(ncat, size=nobs)
effects = rs.standard_normal(nobs)
y += effects
df = pd.DataFrame(pd.Series(pd.Categorical(fact)),
columns=["fact{0}".format(i)])
df_eff = pd.DataFrame(effects[:, None],
columns=["effect_{0}".format(i)])
interactions.append(Interaction(df, df_eff))
if factor_format == "pandas":
for i, interact in enumerate(interactions):
interactions[i] = pd.concat([interact.cat, interact.cont], 1)
interactions = interactions if interactions else None
if interactions and singleton_interaction:
interactions = interactions[0]
if weighted:
weights = pd.DataFrame(rs.chisquare(10, size=(nobs, 1)) / 10)
else:
weights = None
return AttrDict(y=y,
x=x,
absorb=factors,
interactions=interactions,
weights=weights)
