def test_cross_validation_folds_return_length(train_size, min_train_size,
horizon, step, expected):
y_train = np.arange(train_size)
cv = ms.rolling_forecast_origin(y_train, min_train_size, horizon, step)
result = ms.cross_validation_folds(b.Naive1(), cv)
assert len(result) == expected
def test_cross_validation_score_return_length(train_size, min_train_size,
horizon, step, expected):
y_train = np.arange(train_size)
cv = ms.rolling_forecast_origin(y_train, min_train_size, horizon, step)
metric = metrics.mean_error
result = ms.cross_validation_score(b.Naive1(), cv, metric=metric, n_jobs=1)
assert len(result) == expected
def test_rfo_test_length(train_size, horizon, expected):
'''
test the length of test is correct in the rolling origin
generator
'''
train = np.arange(train_size)
cv = ms.rolling_forecast_origin(train, min_train_size=1, horizon=horizon)
_, test_cv = next(cv)
assert expected == len(test_cv)
def test_rfo_min_train_length(train_size, min_train_size, horizon, expected):
'''
check that the minimum training size is correct
'''
train = np.arange(train_size)
cv = ms.rolling_forecast_origin(train,
min_train_size=min_train_size,
horizon=horizon)
train_cv, _ = next(cv)
assert expected == len(train_cv)
def test_forecast_accuracy_length(n_horizons, expected):
model = b.Naive1()
train = np.arange(10000)
metric = metrics.mean_absolute_error
horizons = np.arange(1, n_horizons + 1).tolist()
cv = ms.rolling_forecast_origin(train, min_train_size=100, horizon=100)
train_cv, test_cv = next(cv)
result_h = ms.forecast_accuracy(model,
train_cv,
test_cv,
horizons=horizons,
metric=metric)
assert len(result_h) == expected
def test_rfo_second_fold_train_size(train_size, min_train_size, horizon, step,
expected):
'''
check that the second fold size = minimum training size + step
'''
train = np.arange(train_size)
cv = ms.rolling_forecast_origin(train,
min_train_size=min_train_size,
horizon=horizon,
step=step)
train_cv, _ = next(cv)
# second fold min_train_size + step == len
train_cv, _ = next(cv)
assert expected == len(train_cv)
def test_mase_cv_number_of_folds(train_size, min_train_size, horizon, step,
expected):
'''
check that the number of folds returned from rolling origin
is as expected when data source is a pandas.DataFrame
'''
train = np.arange(train_size)
cv = ms.rolling_forecast_origin(train,
min_train_size=min_train_size,
horizon=horizon,
step=step)
scores = ms.scaled_cross_validation_score(b.Naive1(), cv)
print(expected, len(scores))
print(scores)
assert expected == len(scores)
def test_rfo_number_of_folds_pd(train_size, min_train_size, horizon, step,
expected):
'''
check that the number of folds returned from rolling origin
is as expected when data source is a pandas.DataFrame
'''
train = pd.DataFrame(np.arange(train_size))
cv = ms.rolling_forecast_origin(train,
min_train_size=min_train_size,
horizon=horizon,
step=step)
actual = 0
# number of folds found
for _, _ in cv:
actual += 1
assert expected == actual
def test_rfo_number_of_folds(train_size, min_train_size, horizon, step,
expected):
'''
check that the number of folds returned from rolling origin
is as expected
'''
train = np.arange(train_size)
cv = ms.rolling_forecast_origin(train,
min_train_size=min_train_size,
horizon=horizon,
step=step)
actual = 0
# number of folds found
for _, _ in cv:
actual += 1
assert expected == actual
