def test_resample(self):
df = timeseries_df.copy()
df.index.name = "time_column"
df.reset_index(inplace=True)
post_df = proc.resample(
df=df,
rule="1D",
method="ffill",
time_column="time_column",
)
self.assertListEqual(post_df["label"].tolist(),
["x", "y", "y", "y", "z", "z", "q"])
self.assertListEqual(post_df["y"].tolist(),
[1.0, 2.0, 2.0, 2.0, 3.0, 3.0, 4.0])
post_df = proc.resample(
df=df,
rule="1D",
method="asfreq",
time_column="time_column",
fill_value=0,
)
self.assertListEqual(post_df["label"].tolist(),
["x", "y", 0, 0, "z", 0, "q"])
self.assertListEqual(post_df["y"].tolist(),
[1.0, 2.0, 0, 0, 3.0, 0, 4.0])
def test_resample_should_raise_ex():
with pytest.raises(InvalidPostProcessingError):
pp.resample(
df=categories_df,
rule="1D",
method="asfreq",
)
with pytest.raises(InvalidPostProcessingError):
pp.resample(
df=timeseries_df,
rule="1D",
method="foobar",
)
def test_resample():
post_df = pp.resample(df=timeseries_df, rule="1D", method="ffill")
"""
label y
2019-01-01 x 1.0
2019-01-02 y 2.0
2019-01-03 y 2.0
2019-01-04 y 2.0
2019-01-05 z 3.0
2019-01-06 z 3.0
2019-01-07 q 4.0
"""
assert post_df.equals(
pd.DataFrame(
index=pd.to_datetime(
[
"2019-01-01",
"2019-01-02",
"2019-01-03",
"2019-01-04",
"2019-01-05",
"2019-01-06",
"2019-01-07",
]
),
data={
"label": ["x", "y", "y", "y", "z", "z", "q"],
"y": [1.0, 2.0, 2.0, 2.0, 3.0, 3.0, 4.0],
},
)
)
def test_resample():
df = timeseries_df.copy()
df.index.name = "time_column"
df.reset_index(inplace=True)
post_df = resample(
df=df,
rule="1D",
method="ffill",
time_column="time_column",
)
assert post_df["label"].tolist() == ["x", "y", "y", "y", "z", "z", "q"]
assert post_df["y"].tolist() == [1.0, 2.0, 2.0, 2.0, 3.0, 3.0, 4.0]
post_df = resample(
df=df,
rule="1D",
method="asfreq",
time_column="time_column",
fill_value=0,
)
assert post_df["label"].tolist() == ["x", "y", 0, 0, "z", 0, "q"]
assert post_df["y"].tolist() == [1.0, 2.0, 0, 0, 3.0, 0, 4.0]
def test_resample_zero_fill():
post_df = pp.resample(df=timeseries_df, rule="1D", method="asfreq", fill_value=0)
assert post_df.equals(
pd.DataFrame(
index=pd.to_datetime(
[
"2019-01-01",
"2019-01-02",
"2019-01-03",
"2019-01-04",
"2019-01-05",
"2019-01-06",
"2019-01-07",
]
),
data={
"label": ["x", "y", 0, 0, "z", 0, "q"],
"y": [1.0, 2.0, 0, 0, 3.0, 0, 4.0],
},
)
)
def test_resample_linear():
df = pd.DataFrame(
index=to_datetime(["2019-01-01", "2019-01-05", "2019-01-08"]),
data={"label": ["a", "e", "j"], "y": [1.0, 5.0, 8.0]},
)
post_df = pp.resample(df=df, rule="1D", method="linear")
"""
label y
2019-01-01 a 1.0
2019-01-02 NaN 2.0
2019-01-03 NaN 3.0
2019-01-04 NaN 4.0
2019-01-05 e 5.0
2019-01-06 NaN 6.0
2019-01-07 NaN 7.0
2019-01-08 j 8.0
"""
assert post_df.equals(
pd.DataFrame(
index=pd.to_datetime(
[
"2019-01-01",
"2019-01-02",
"2019-01-03",
"2019-01-04",
"2019-01-05",
"2019-01-06",
"2019-01-07",
"2019-01-08",
]
),
data={
"label": ["a", np.NaN, np.NaN, np.NaN, "e", np.NaN, np.NaN, "j"],
"y": [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0],
},
)
)
def test_resample_with_groupby():
"""
The Dataframe contains a timestamp column, a string column and a numeric column.
__timestamp city val
0 2022-01-13 Chicago 6.0
1 2022-01-13 LA 5.0
2 2022-01-13 NY 4.0
3 2022-01-11 Chicago 3.0
4 2022-01-11 LA 2.0
5 2022-01-11 NY 1.0
"""
df = DataFrame({
"__timestamp":
to_datetime([
"2022-01-13",
"2022-01-13",
"2022-01-13",
"2022-01-11",
"2022-01-11",
"2022-01-11",
]),
"city": ["Chicago", "LA", "NY", "Chicago", "LA", "NY"],
"val": [6.0, 5.0, 4.0, 3.0, 2.0, 1.0],
})
post_df = resample(
df=df,
rule="1D",
method="asfreq",
fill_value=0,
time_column="__timestamp",
groupby_columns=("city", ),
)
assert list(post_df.columns) == [
"__timestamp",
"city",
"val",
]
assert [str(dt.date()) for dt in post_df["__timestamp"]
] == (["2022-01-11"] * 3 + ["2022-01-12"] * 3 + ["2022-01-13"] * 3)
assert list(post_df["val"]) == [3.0, 2.0, 1.0, 0, 0, 0, 6.0, 5.0, 4.0]
# should raise error when get a non-existent column
with pytest.raises(QueryObjectValidationError):
resample(
df=df,
rule="1D",
method="asfreq",
fill_value=0,
time_column="__timestamp",
groupby_columns=(
"city",
"unkonw_column",
),
)
# should raise error when get a None value in groupby list
with pytest.raises(QueryObjectValidationError):
resample(
df=df,
rule="1D",
method="asfreq",
fill_value=0,
time_column="__timestamp",
groupby_columns=(
"city",
None,
),
)
def test_resample_after_pivot():
df = pd.DataFrame(
data={
"__timestamp": pd.to_datetime(
[
"2022-01-13",
"2022-01-13",
"2022-01-13",
"2022-01-11",
"2022-01-11",
"2022-01-11",
]
),
"city": ["Chicago", "LA", "NY", "Chicago", "LA", "NY"],
"val": [6.0, 5.0, 4.0, 3.0, 2.0, 1.0],
}
)
pivot_df = pp.pivot(
df=df,
index=["__timestamp"],
columns=["city"],
aggregates={
"val": {"operator": "sum"},
},
flatten_columns=False,
reset_index=False,
)
"""
val
city Chicago LA NY
__timestamp
2022-01-11 3.0 2.0 1.0
2022-01-13 6.0 5.0 4.0
"""
resample_df = pp.resample(
df=pivot_df,
rule="1D",
method="asfreq",
fill_value=0,
)
"""
val
city Chicago LA NY
__timestamp
2022-01-11 3.0 2.0 1.0
2022-01-12 0.0 0.0 0.0
2022-01-13 6.0 5.0 4.0
"""
flat_df = pp.flatten(resample_df)
"""
__timestamp val, Chicago val, LA val, NY
0 2022-01-11 3.0 2.0 1.0
1 2022-01-12 0.0 0.0 0.0
2 2022-01-13 6.0 5.0 4.0
"""
assert flat_df.equals(
pd.DataFrame(
data={
"__timestamp": pd.to_datetime(
["2022-01-11", "2022-01-12", "2022-01-13"]
),
"val, Chicago": [3.0, 0, 6.0],
"val, LA": [2.0, 0, 5.0],
"val, NY": [1.0, 0, 4.0],
}
)
)
def test_resample_should_not_side_effect():
_timeseries_df = timeseries_df.copy()
pp.resample(df=_timeseries_df, rule="1D", method="ffill")
assert _timeseries_df.equals(timeseries_df)