def test_gscv(forecaster, param_grid, cv, scoring):
y, X = load_longley()
gscv = ForecastingGridSearchCV(forecaster,
param_grid=param_grid,
cv=cv,
scoring=scoring)
gscv.fit(y, X)
param_grid = ParameterGrid(param_grid)
_check_cv(forecaster, gscv, cv, param_grid, y, X, scoring)
def test_evaluate_no_exog_against_with_exog():
"""Check that adding exogenous data produces different results."""
y, X = load_longley()
forecaster = ARIMA(suppress_warnings=True)
cv = SlidingWindowSplitter()
scoring = MeanAbsolutePercentageError(symmetric=True)
out_exog = evaluate(forecaster, cv, y, X=X, scoring=scoring)
out_no_exog = evaluate(forecaster, cv, y, X=None, scoring=scoring)
scoring_name = f"test_{scoring.name}"
assert np.all(out_exog[scoring_name] != out_no_exog[scoring_name])
def test_evaluate_no_exog_against_with_exog():
# Check that adding exogenous data produces different results
y, X = load_longley()
forecaster = ARIMA(suppress_warnings=True)
cv = SlidingWindowSplitter()
scoring = sMAPE()
out_exog = evaluate(forecaster, cv, y, X=X, scoring=scoring)
out_no_exog = evaluate(forecaster, cv, y, X=None, scoring=scoring)
scoring_name = f"test_{scoring.name}"
assert np.all(out_exog[scoring_name] != out_no_exog[scoring_name])
def test_rscv(forecaster, param_grid, cv, scoring, n_iter, random_state):
"""Tests that ForecastingRandomizedSearchCV successfully searches the
parameter distributions to identify the best parameter set
"""
y, X = load_longley()
rscv = ForecastingRandomizedSearchCV(
forecaster,
param_distributions=param_grid,
cv=cv,
scoring=scoring,
n_iter=n_iter,
random_state=random_state,
)
rscv.fit(y, X)
param_distributions = list(
ParameterSampler(param_grid, n_iter, random_state=random_state))
_check_cv(forecaster, rscv, cv, param_distributions, y, X, scoring)
def test_ARIMAForecaster_multivariate(fh):
X, y = load_longley(return_X_y=True)
# get data in required format
max_fh = np.max(fh)
m = len(y.iloc[0])
cutoff = m - max_fh
y_train = pd.Series([y.iloc[0].iloc[:cutoff]])
y_test = pd.Series([y.iloc[0].iloc[cutoff:]])
X_train = pd.DataFrame(
[pd.Series([X.iloc[0, i].iloc[:cutoff]]) for i in range(X.shape[1])]).T
X_train.columns = X.columns
X_test = pd.DataFrame(
[pd.Series([X.iloc[0, i].iloc[cutoff:]]) for i in range(X.shape[1])]).T
X_test.columns = X.columns
m = ARIMAForecaster()
m.fit(y_train, X=X_train)
y_pred = m.predict(fh=fh, X=X_test)
assert y_pred.shape[0] == len(fh)
assert m.score(y_test, fh=fh, X=X_test) > 0
def test_ForecastingStrategy_multivariate():
longley = load_longley(return_X_y=False)
train = pd.DataFrame([
pd.Series([longley.iloc[0, i].iloc[:13]])
for i in range(longley.shape[1])
]).T
train.columns = longley.columns
test = pd.DataFrame([
pd.Series([longley.iloc[0, i].iloc[13:]])
for i in range(longley.shape[1])
]).T
test.columns = longley.columns
target = "TOTEMP"
fh = np.arange(len(test[target].iloc[0])) + 1
task = ForecastingTask(target=target, fh=fh, metadata=train)
estimator = ARIMAForecaster()
s = ForecastingStrategy(estimator=estimator)
s.fit(task, train)
y_pred = s.predict(data=test)
assert y_pred.shape == test[task.target].iloc[0].shape
#!/usr/bin/env python3 -u # -*- coding: utf-8 -*- # copyright: sktime developers, BSD-3-Clause License (see LICENSE file). """Unit tests of DateTimeFeatures functionality.""" import pytest from sktime.datasets import load_airline, load_longley from sktime.forecasting.model_selection import temporal_train_test_split from sktime.transformations.series.date import DateTimeFeatures # Load multivariate dataset longley and apply calendar extraction y, X = load_longley() y_train, y_test, X_train, X_test = temporal_train_test_split(y, X) # Test that comprehensive feature_scope works for weeks pipe = DateTimeFeatures(ts_freq="W", feature_scope="comprehensive") pipe.fit(X_train) test_full_featurescope = pipe.transform(X_train).columns.to_list() # Test that minimal feature_scope works for weeks pipe = DateTimeFeatures(ts_freq="W", feature_scope="minimal") pipe.fit(X_train) test_reduced_featurescope = pipe.transform(X_train).columns.to_list() # Test that comprehensive feature_scope works for months pipe = DateTimeFeatures(ts_freq="M", feature_scope="comprehensive") pipe.fit(X_train) test_changing_frequency = pipe.transform(X_train).columns.to_list()
y_multi = get_examples(mtype="pd-multiindex", as_scitype="Panel")[0]
# y Train will be univariate data set
y_train, y_test = temporal_train_test_split(y)
# Create Panel sample data
mi = pd.MultiIndex.from_product([[0], y.index],
names=["instances", "timepoints"])
y_group1 = pd.DataFrame(y.values, index=mi, columns=["y"])
mi = pd.MultiIndex.from_product([[1], y.index],
names=["instances", "timepoints"])
y_group2 = pd.DataFrame(y.values, index=mi, columns=["y"])
y_grouped = pd.concat([y_group1, y_group2])
y_ll, X_ll = load_longley()
y_ll_train, _, X_ll_train, X_ll_test = temporal_train_test_split(y_ll, X_ll)
# Get different WindowSummarizer functions
kwargs = WindowSummarizer.get_test_params()[0]
kwargs_alternames = WindowSummarizer.get_test_params()[1]
kwargs_variant = WindowSummarizer.get_test_params()[2]
def count_gt100(x):
"""Count how many observations lie above threshold 100."""
return np.sum((x > 100)[::-1])
# Cannot be pickled in get_test_params, therefore here explicit
kwargs_custom = {
