def _run_aggr_dist_test(x, y):
# default parameters
default_params = AggrDist(transformer=ScipyDist())
default_params_transformation = np.around((default_params.transform(x, y)),
decimals=3)
assert np.array_equal(
np.array([
[1.649, 1.525, 1.518, 1.934, 1.636],
[1.39, 1.395, 1.313, 1.617, 1.418],
[1.492, 1.489, 1.323, 1.561, 1.437],
[1.559, 1.721, 1.544, 1.589, 1.642],
]),
default_params_transformation,
), "Error occurred testing on default parameters, result is not correct"
for transformer in PAIRWISE_TRANSFORMERS_TAB:
for aggfunc in AGGFUNCS:
aggfunc_params = AggrDist(transformer=transformer(),
aggfunc=aggfunc)
aggfunc_params_transformation = aggfunc_params.transform(x, y)
assert isinstance(
aggfunc_params_transformation,
np.ndarray), (f"Error occurred testing on following parameters"
f"transformer={transformer}, aggfunc={aggfunc}")
def _run_aggr_dist_test(x, y):
# default parametersc
default_params = AggrDist(transformer=ScipyDist())
default_params_transformation = np.around((default_params.transform(x, y)),
decimals=3)
assert np.array_equal(
np.array([
[1.714, 1.49, 1.53, 1.699, 1.849],
[1.479, 1.36, 1.358, 1.476, 1.471],
[1.553, 1.476, 1.354, 1.523, 1.425],
[1.641, 1.704, 1.603, 1.698, 1.37],
]),
default_params_transformation,
), "Error occurred testing on default parameters, result is not correct"
for transformer in PAIRWISE_TRANSFORMERS_TAB:
for aggfunc in AGGFUNCS:
aggfunc_params = AggrDist(transformer=transformer(),
aggfunc=aggfunc)
aggfunc_params_transformation = aggfunc_params.transform(x, y)
assert isinstance(
aggfunc_params_transformation,
np.ndarray), (f"Error occurred testing on following parameters"
f"transformer={transformer}, aggfunc={aggfunc}")
def _run_scipy_dist_test(x, y):
# default parameters
default_params = ScipyDist()
default_params_transformation = np.around((default_params.transform(x, y)),
decimals=3)
assert np.array_equal(
np.array([
[2.318, 1.657, 1.582, 1.502, 1.461],
[1.79, 1.249, 1.715, 1.656, 1.449],
[2.424, 2.083, 2.28, 1.735, 1.73],
[1.602, 1.012, 1.658, 1.167, 0.901],
[2.219, 1.643, 1.373, 1.005, 1.216],
]),
default_params_transformation,
), "Error occurred testing on default parameters, result is not correct"
for metric in METRIC_VALUES:
for p in P_VALUES:
for colalign in COLALIGN_VALUES:
metric_params = ScipyDist(metric=metric,
p=p,
colalign=colalign)
metric_params_transformation = metric_params.transform(x, y)
assert isinstance(metric_params_transformation, np.ndarray), (
f"Error occurred testing on following parameters"
f"metric={metric}, p={p}, colalign={colalign}")
},
HampelFilter: {
"window_length": 3
},
OptionalPassthrough: {
"transformer": BoxCoxTransformer(),
"passthrough": False
},
FeatureSelection: {
"method": "all"
},
ColumnwiseTransformer: {
"transformer": Detrender()
},
AggrDist: {
"transformer": ScipyDist()
},
PyODAnnotator: {
"estimator": ANOMALY_DETECTOR
},
ClaSPSegmentation: {
"period_length": 5,
"n_cps": 1
},
ClaSPTransformer: {
"window_length": 5
},
}
# We use estimator tags in addition to class hierarchies to further distinguish
# estimators into different categories. This is useful for defining and running
},
Prophet: {
"n_changepoints": 0,
"yearly_seasonality": False,
"weekly_seasonality": False,
"daily_seasonality": False,
"uncertainty_samples": 1000,
"verbose": False,
},
PartialAutoCorrelationTransformer: {"n_lags": 1},
AutoCorrelationTransformer: {"n_lags": 1},
Imputer: {"method": "mean"},
HampelFilter: {"window_length": 3},
OptionalPassthrough: {"transformer": BoxCoxTransformer(), "passthrough": True},
ColumnwiseTransformer: {"transformer": Detrender()},
AggrDist: {"transformer": ScipyDist()},
PyODAnnotator: {"estimator": ANOMALY_DETECTOR},
}
# We use estimator tags in addition to class hierarchies to further distinguish
# estimators into different categories. This is useful for defining and running
# common tests for estimators with the same tags.
VALID_ESTIMATOR_TAGS = tuple(ESTIMATOR_TAG_LIST)
# These methods should not change the state of the estimator, that is, they should
# not change fitted parameters or hyper-parameters. They are also the methods that
# "apply" the fitted estimator to data and useful for checking results.
NON_STATE_CHANGING_METHODS = (
"predict",
"predict_proba",
"decision_function",