def test_and_dict_of_lists():
"""
Test AndAggregator with input as a dict of lists of time stamps or time
stamp 2-tuples
"""
lists = {
"A": [
(Timestamp("2017-1-1"), Timestamp("2017-1-2")),
(Timestamp("2017-1-5"), Timestamp("2017-1-8")),
Timestamp("2017-1-10"),
],
"B": [
Timestamp("2017-1-2"),
(Timestamp("2017-1-3"), Timestamp("2017-1-6")),
Timestamp("2017-1-8"),
(Timestamp("2017-1-7"), Timestamp("2017-1-9")),
(Timestamp("2017-1-11"), Timestamp("2017-1-11")),
],
}
assert aggt.AndAggregator().aggregate(lists) == [
Timestamp("2017-1-2"),
(Timestamp("2017-01-05 00:00:00"), Timestamp("2017-01-06 00:00:00")),
(Timestamp("2017-1-7 00:00:00"), Timestamp("2017-1-8 00:00:00")),
]
lists = {
"A": [
(Timestamp("2017-1-1"), Timestamp("2017-1-2")),
(Timestamp("2017-1-5"), Timestamp("2017-1-8")),
Timestamp("2017-1-10"),
],
"B": [],
}
assert aggt.AndAggregator().aggregate(lists) == []
def test_and_df():
"""
Test AndAggregator with input as a DataFrame
"""
df = pd.DataFrame(
[[1, 1], [1, 0], [0, 1], [0, 0], [float("nan"), 1], [0, float("nan")]],
index=pd.date_range(start="2017-1-1", periods=6, freq="D"),
)
pd.testing.assert_series_equal(
aggt.AndAggregator().aggregate(df),
pd.Series(
[1, 0, 0, 0, float("nan"), 0],
index=pd.date_range(start="2017-1-1", periods=6, freq="D"),
),
)
def test_detector_return_intermediate():
s = pd.Series(
[0, 1, 2, 3, 2, 1] * 10,
index=pd.date_range(start="2017-1-1", periods=60, freq="D"),
)
my_pipe = Pipenet([
{
"name": "deseasonal_residual",
"model": (transformer.NaiveSeasonalDecomposition(freq=6)),
"input": "original",
},
{
"name": "abs_residual",
"model": transformer.CustomizedTransformer1D(transform_func=abs),
"input": "deseasonal_residual",
},
{
"name": "iqr_ad",
"model": detector.InterQuartileRangeAD(c=(None, 3)),
"input": "abs_residual",
},
{
"name": "sign_check",
"model": detector.ThresholdAD(high=0.0, low=-float("inf")),
"input": "deseasonal_residual",
},
{
"name": "and",
"model": aggregator.AndAggregator(),
"input": ["iqr_ad", "sign_check"],
},
])
result = my_pipe.fit_detect(s, return_intermediate=True)
assert set(result.keys()) == {
"original",
"deseasonal_residual",
"abs_residual",
"iqr_ad",
"sign_check",
"and",
}
def test_skip_fit():
s = pd.Series(
[0, 1, 2, 3, 2, 1] * 10,
index=pd.date_range(start="2017-1-1", periods=60, freq="D"),
)
deseasonal_residual = transformer.NaiveSeasonalDecomposition(freq=6)
my_pipe = Pipenet([
{
"name": "deseasonal_residual",
"model": deseasonal_residual,
"input": "original",
},
{
"name": "abs_residual",
"model": transformer.CustomizedTransformer1D(transform_func=abs),
"input": "deseasonal_residual",
},
{
"name": "iqr_ad",
"model": detector.InterQuartileRangeAD(c=(None, 3)),
"input": "abs_residual",
},
{
"name": "sign_check",
"model": detector.ThresholdAD(high=0.0, low=-float("inf")),
"input": "deseasonal_residual",
},
{
"name": "and",
"model": aggregator.AndAggregator(),
"input": ["iqr_ad", "sign_check"],
},
])
with pytest.raises(RuntimeError):
my_pipe.fit_detect(s, skip_fit=["deseasonal_residual"])
my_pipe.fit_detect(s)