def init_od_prophet(state_dict: Dict) -> OutlierProphet:
"""
Initialize OutlierProphet.
Parameters
----------
state_dict
Dictionary containing the parameter values.
Returns
-------
Initialized OutlierProphet instance.
"""
od = OutlierProphet(cap=state_dict['cap'])
od.model = state_dict['model']
return od
def test_prophet(prophet_params):
import fbprophet
growth, return_instance_score, return_forecast = prophet_params
od = OutlierProphet(growth=growth)
assert isinstance(od.model, fbprophet.forecaster.Prophet)
assert od.meta == {'name': 'OutlierProphet', 'detector_type': 'offline', 'data_type': 'time-series',
'version': __version__}
if growth == 'logistic':
df_fit['cap'] = 10.
df_test['cap'] = 10.
od.fit(df_fit)
forecast = od.score(df_test)
fcst_cols = list(forecast.columns)
check_cols = ['ds', 'yhat', 'yhat_lower', 'yhat_upper', 'score', 'y']
assert all(check_col in fcst_cols for check_col in check_cols)
od_preds = od.predict(df_test,
return_instance_score=return_instance_score,
return_forecast=return_forecast)
assert od_preds['meta'] == od.meta
assert (od_preds['data']['is_outlier']['ds'] == df_test['ds']).all()
assert od_preds['data']['is_outlier']['is_outlier'].shape[0] == df_test['ds'].shape[0]
if not return_instance_score:
assert od_preds['data']['instance_score'] is None
if not return_forecast:
with pytest.raises(KeyError):
od_preds['data']['forecast']
OutlierAEGMM(threshold=threshold,
gmm_density_net=gmm_density_net,
n_gmm=n_gmm,
**kwargs),
OutlierVAE(threshold=threshold,
latent_dim=latent_dim,
samples=samples,
**kwargs),
OutlierAE(threshold=threshold, **kwargs),
OutlierVAEGMM(threshold=threshold,
gmm_density_net=gmm_density_net,
n_gmm=n_gmm,
latent_dim=latent_dim,
samples=samples,
**kwargs),
OutlierProphet(threshold=.7, growth='logistic'),
SpectralResidual(threshold=threshold, window_amp=10, window_local=10),
OutlierSeq2Seq(input_dim,
seq_len,
threshold=threshold,
threshold_net=threshold_net,
latent_dim=latent_dim)
]
n_tests = len(detector)
@pytest.fixture
def select_detector(request):
return detector[request.param]