def test_predict():
x, y, theta = create_test_data()
train_percentage = 90
split_data = int(len(x) * (train_percentage / 100))
X_train = x[0:split_data, 0]
X_test = x[split_data::, 0]
y1 = y[0:split_data, 0]
y_test = y[split_data::, 0]
y_train = y1.copy()
y_train = np.reshape(y_train, (len(y_train), 1))
X_train = np.reshape(X_train, (len(X_train), 1))
y_test = np.reshape(y_test, (len(y_test), 1))
X_test = np.reshape(X_test, (len(X_test), 1))
model = PolynomialNarmax(
non_degree=2,
n_terms=5,
extended_least_squares=False,
ylag=[1, 2],
xlag=2,
estimator="least_squares",
)
model.fit(X_train, y_train)
yhat = model.predict(X_test, y_test)
assert_almost_equal(yhat, y_test, decimal=10)
def test_model_prediction():
x, y, theta = create_test_data()
train_percentage = 90
split_data = int(len(x) * (train_percentage / 100))
X_train = x[0:split_data, 0]
X_test = x[split_data::, 0]
y1 = y[0:split_data, 0]
y_test = y[split_data::, 0]
y_train = y1.copy()
y_train = np.reshape(y_train, (len(y_train), 1))
X_train = np.reshape(X_train, (len(X_train), 1))
y_test = np.reshape(y_test, (len(y_test), 1))
X_test = np.reshape(X_test, (len(X_test), 1))
model = PolynomialNarmax(
non_degree=2,
n_terms=5,
extended_least_squares=False,
ylag=[1, 2],
xlag=2,
estimator='least_squares',
)
model.fit(X_train, y_train)
assert_raises(Exception, model.predict, X_test, y_test[:1])
def test_fit_without_information_criteria():
x, y, theta = create_test_data()
model = PolynomialNarmax(
non_degree=2,
n_terms=15,
extended_least_squares=False,
)
model.fit(x, y)
assert 'info_values' not in dir(model)
def test_fit_with_information_criteria():
x, y, theta = create_test_data()
model = PolynomialNarmax(
non_degree=2,
n_terms=15,
order_selection=True,
extended_least_squares=False,
)
model.fit(x, y)
assert "info_values" in dir(model)
def test_least_squares():
x, y, theta = create_test_data()
model = PolynomialNarmax(
non_degree=2,
n_terms=5,
extended_least_squares=False,
ylag=[1, 2],
xlag=2,
estimator='least_squares',
)
model.fit(x, y)
assert_almost_equal(model.theta, theta, decimal=2)
def test_least_mean_squares_mixed_norm():
x, y, theta = create_test_data(n=30000)
model = PolynomialNarmax(
non_degree=2,
n_terms=5,
mu=0.05,
extended_least_squares=False,
ylag=[1, 2],
xlag=2,
estimator="least_mean_squares_mixed_norm",
)
model.fit(x, y)
assert_almost_equal(model.theta, theta, decimal=2)
def test_least_mean_squares_sign_error():
x, y, theta = create_test_data(n=5000)
model = PolynomialNarmax(
non_degree=2,
n_terms=5,
mu=0.01,
extended_least_squares=False,
ylag=[1, 2],
xlag=2,
estimator='least_mean_squares_sign_error',
)
model.fit(x, y)
assert_almost_equal(model.theta, theta, decimal=2)
def test_recursive_least_squares():
x, y, theta = create_test_data()
model = PolynomialNarmax(
non_degree=2,
n_terms=5,
delta=0.00001,
lam=0.99,
extended_least_squares=False,
ylag=[1, 2],
xlag=2,
estimator="recursive_least_squares",
)
model.fit(x, y)
assert_almost_equal(model.theta, theta, decimal=2)
def test_affine_least_mean_squares():
x, y, theta = create_test_data()
model = PolynomialNarmax(
non_degree=2,
n_terms=5,
mu=0.01,
offset_covariance=0.2,
extended_least_squares=False,
ylag=[1, 2],
xlag=2,
estimator='affine_least_mean_squares',
)
model.fit(x, y)
assert_almost_equal(model.theta, theta, decimal=2)
def test_least_mean_squares_normalized_sign_error():
x, y, theta = create_test_data()
model = PolynomialNarmax(
non_degree=2,
n_terms=5,
n_info_values=6,
extended_least_squares=False,
ylag=[1, 2],
xlag=2,
info_criteria='aic',
estimator='least_mean_squares_normalized_sign_error',
mu=0.005,
)
model.fit(x, y)
assert_almost_equal(model.theta, theta, decimal=2)
def test_default_values():
default = {
'non_degree': 2,
'ylag': 2,
'xlag': 2,
'order_selection': False,
'info_criteria': 'aic',
'n_terms': None,
'n_inputs': 1,
'n_info_values': 10,
'estimator': 'least_squares',
'extended_least_squares': True,
'aux_lag': 1,
'lam': 0.98,
'delta': 0.01,
'offset_covariance': 0.2,
'mu': 0.01,
'eps': np.finfo(np.float).eps,
'gama': 0.2,
'weight': 0.02
}
model = PolynomialNarmax()
model_values = [
model.non_degree, model.ylag, model.xlag, model._order_selection,
model.info_criteria, model.n_terms, model._n_inputs,
model.n_info_values, model.estimator, model._extended_least_squares,
model._aux_lag, model._lam, model._delta, model._offset_covariance,
model._mu, model._eps, model._gama, model._weight
]
assert list(default.values()) == model_values
def test_information_criteria_lilc():
x, y, theta = create_test_data()
model = PolynomialNarmax(
non_degree=2,
n_terms=5,
extended_least_squares=False,
order_selection=True,
info_criteria="lilc",
n_info_values=5,
ylag=[1, 2],
xlag=2,
estimator="least_squares",
)
model.fit(x, y)
info_values = np.array([-1767.926, -2326.183, -2985.514, -4474.072, -72860.973])
assert_almost_equal(model.info_values[:4], info_values[:4], decimal=3)
def test_information_criteria_bic():
x, y, theta = create_test_data()
model = PolynomialNarmax(
non_degree=2,
n_terms=5,
extended_least_squares=False,
order_selection=True,
info_criteria="bic",
n_info_values=5,
ylag=[1, 2],
xlag=2,
estimator="least_squares",
)
model.fit(x, y)
info_values = np.array([-1764.885, -2320.101, -2976.391, -4461.908, -72845.768])
assert_almost_equal(model.info_values[:4], info_values[:4], decimal=3)
def test_error_reduction_ration():
piv = np.array([4, 2, 7, 11, 5])
model_code = np.array([[2002, 0], [1002, 0], [2001, 1001], [2002, 1002],
[1001, 1001]])
x, y, theta = create_test_data()
model = PolynomialNarmax(
non_degree=2,
n_terms=5,
order_selection=True,
n_info_values=5,
info_criteria="aic",
extended_least_squares=False,
ylag=[1, 2],
xlag=2,
estimator="least_squares",
)
model.fit(x, y)
# assert_array_equal(model.pivv, piv)
assert_array_equal(model.final_model, model_code)
def test_information_criteria():
x, y, theta = create_test_data()
model = PolynomialNarmax(
non_degree=2,
n_terms=5,
extended_least_squares=False,
order_selection=True,
info_criteria='aic',
n_info_values=5,
ylag=[1, 2],
xlag=2,
estimator='least_squares',
)
model.fit(x, y)
info_values = np.array([
-1769.7907939, -2329.9129067, -2991.1078461, -4481.5306496, -71792.6413
])
assert_almost_equal(model.info_values, info_values)
def test_information_criteria_fpe():
x, y, theta = create_test_data()
model = PolynomialNarmax(
non_degree=2,
n_terms=5,
extended_least_squares=False,
order_selection=True,
info_criteria="fpe",
n_info_values=5,
ylag=[1, 2],
xlag=2,
estimator="least_squares",
)
model.fit(x, y)
info_values = np.array(
[-1769.7907932, -2329.9129013, -2991.1078281, -4481.5306067, -72870.296884]
)
assert_almost_equal(model.info_values[:4], info_values[:4], decimal=3)
def test_information_criteria_bic():
x, y, theta = create_test_data()
model = PolynomialNarmax(
non_degree=2,
n_terms=5,
extended_least_squares=False,
order_selection=True,
info_criteria='bic',
n_info_values=5,
ylag=[1, 2],
xlag=2,
estimator='least_squares',
)
model.fit(x, y)
info_values = np.array([
-1764.8850406, -2320.1014001, -2976.3905863, -4461.9076365,
-71768.1125336
])
assert_almost_equal(model.info_values, info_values)
def test_unbiased_estimator():
x, y, theta = create_test_data()
mu, sigma = 0, 0.2
nu = np.random.normal(mu, sigma, 1000).T
e = np.zeros((1000, 1))
lag = 2
for k in range(lag, len(x)):
e[k] = 0.2 * nu[k - 1] + nu[k]
y = y + e
model = PolynomialNarmax(
non_degree=2,
n_terms=5,
extended_least_squares=True,
ylag=[1, 2],
xlag=2,
estimator='least_squares',
)
model.fit(x, y)
assert_almost_equal(model.theta, theta, decimal=2)
def test_information_criteria_lilc():
x, y, theta = create_test_data()
model = PolynomialNarmax(
non_degree=2,
n_terms=5,
extended_least_squares=False,
order_selection=True,
info_criteria='lilc',
n_info_values=5,
ylag=[1, 2],
xlag=2,
estimator='least_squares',
)
model.fit(x, y)
info_values = np.array([
-1767.9260841, -2326.1834872, -2985.5137169, -4474.0718106,
-71783.3177513
])
assert_almost_equal(model.info_values, info_values)
def test_results():
x, y, theta = create_test_data()
model = PolynomialNarmax(
non_degree=2,
n_terms=5,
extended_least_squares=False,
order_selection=True,
info_criteria='lilc',
n_info_values=5,
ylag=[1, 2],
xlag=2,
estimator='least_squares',
)
model.fit(x, y)
results = model.results(err_precision=8, dtype='dec')
assert isinstance(results, list)
assert_raises(ValueError, model.results, theta_precision=-1)
assert_raises(ValueError, model.results, err_precision=-1)
assert_raises(ValueError, model.results, dtype='DEC')
def test_default_values():
default = {
"non_degree": 2,
"ylag": 2,
"xlag": 2,
"order_selection": False,
"info_criteria": "aic",
"n_terms": None,
"n_inputs": 1,
"n_info_values": 10,
"estimator": "least_squares",
"extended_least_squares": True,
"aux_lag": 1,
"lam": 0.98,
"delta": 0.01,
"offset_covariance": 0.2,
"mu": 0.01,
"eps": np.finfo(np.float).eps,
"gama": 0.2,
"weight": 0.02,
}
model = PolynomialNarmax()
model_values = [
model.non_degree,
model.ylag,
model.xlag,
model._order_selection,
model.info_criteria,
model.n_terms,
model._n_inputs,
model.n_info_values,
model.estimator,
model._extended_least_squares,
model._aux_lag,
model._lam,
model._delta,
model._offset_covariance,
model._mu,
model._eps,
model._gama,
model._weight,
]
assert list(default.values()) == model_values
