def test_clone_idempotent():
model = preprocessing.StandardScaler() | linear_model.LogisticRegression(
optimizer=optim.Adam(), l2=0.1)
trace = []
for x, y in datasets.Phishing():
trace.append(model.predict_proba_one(x))
model.learn_one(x, y)
clone = model.clone()
for i, (x, y) in enumerate(datasets.Phishing()):
assert clone.predict_proba_one(x) == trace[i]
clone.learn_one(x, y)
@pytest.mark.parametrize(
'lm, dataset',
[
pytest.param(lm(
optimizer=copy.deepcopy(optimizer), initializer=initializer, l2=0),
dataset,
id=f'{lm.__name__} - {optimizer} - {initializer}')
for lm, dataset in [(lm.LinearRegression, datasets.TrumpApproval()
), (lm.LogisticRegression, datasets.Bananas())]
for optimizer, initializer in itertools.product(
[
optim.AdaBound(),
optim.AdaDelta(),
optim.AdaGrad(),
optim.AdaMax(),
optim.Adam(),
optim.AMSGrad(),
# TODO: check momentum optimizers
# optim.Momentum(),
# optim.NesterovMomentum(),
optim.RMSProp(),
optim.SGD()
],
[
optim.initializers.Zeros(),
optim.initializers.Normal(mu=0, sigma=1, seed=42)
])
])
@pytest.mark.slow
def test_finite_differences(lm, dataset):
"""Checks the gradient of a linear model via finite differences.