def test_contribution(self):
df = DataFrame({
DTTM_ALIAS: [
datetime(2020, 7, 16, 14, 49),
datetime(2020, 7, 16, 14, 50),
],
"a": [1, 3],
"b": [1, 9],
})
# cell contribution across row
row_df = proc.contribution(df,
PostProcessingContributionOrientation.ROW)
self.assertListEqual(df.columns.tolist(), [DTTM_ALIAS, "a", "b"])
self.assertListEqual(series_to_list(row_df["a"]), [0.5, 0.25])
self.assertListEqual(series_to_list(row_df["b"]), [0.5, 0.75])
# cell contribution across column without temporal column
df.pop(DTTM_ALIAS)
column_df = proc.contribution(
df, PostProcessingContributionOrientation.COLUMN)
self.assertListEqual(df.columns.tolist(), ["a", "b"])
self.assertListEqual(series_to_list(column_df["a"]), [0.25, 0.75])
self.assertListEqual(series_to_list(column_df["b"]), [0.1, 0.9])
def test_contribution():
df = DataFrame(
{
DTTM_ALIAS: [
datetime(2020, 7, 16, 14, 49),
datetime(2020, 7, 16, 14, 50),
datetime(2020, 7, 16, 14, 51),
],
"a": [1, 3, nan],
"b": [1, 9, nan],
"c": [nan, nan, nan],
}
)
with pytest.raises(QueryObjectValidationError, match="not numeric"):
contribution(df, columns=[DTTM_ALIAS])
with pytest.raises(QueryObjectValidationError, match="same length"):
contribution(df, columns=["a"], rename_columns=["aa", "bb"])
# cell contribution across row
processed_df = contribution(
df, orientation=PostProcessingContributionOrientation.ROW,
)
assert processed_df.columns.tolist() == [DTTM_ALIAS, "a", "b", "c"]
assert_array_equal(processed_df["a"].tolist(), [0.5, 0.25, nan])
assert_array_equal(processed_df["b"].tolist(), [0.5, 0.75, nan])
assert_array_equal(processed_df["c"].tolist(), [0, 0, nan])
# cell contribution across column without temporal column
df.pop(DTTM_ALIAS)
processed_df = contribution(
df, orientation=PostProcessingContributionOrientation.COLUMN
)
assert processed_df.columns.tolist() == ["a", "b", "c"]
assert_array_equal(processed_df["a"].tolist(), [0.25, 0.75, 0])
assert_array_equal(processed_df["b"].tolist(), [0.1, 0.9, 0])
assert_array_equal(processed_df["c"].tolist(), [nan, nan, nan])
# contribution only on selected columns
processed_df = contribution(
df,
orientation=PostProcessingContributionOrientation.COLUMN,
columns=["a"],
rename_columns=["pct_a"],
)
assert processed_df.columns.tolist() == ["a", "b", "c", "pct_a"]
assert_array_equal(processed_df["a"].tolist(), [1, 3, nan])
assert_array_equal(processed_df["b"].tolist(), [1, 9, nan])
assert_array_equal(processed_df["c"].tolist(), [nan, nan, nan])
assert processed_df["pct_a"].tolist() == [0.25, 0.75, 0]
def test_contribution(self):
df = DataFrame({
DTTM_ALIAS: [
datetime(2020, 7, 16, 14, 49),
datetime(2020, 7, 16, 14, 50),
],
"a": [1, 3],
"b": [1, 9],
})
with pytest.raises(QueryObjectValidationError, match="not numeric"):
proc.contribution(df, columns=[DTTM_ALIAS])
with pytest.raises(QueryObjectValidationError, match="same length"):
proc.contribution(df, columns=["a"], rename_columns=["aa", "bb"])
# cell contribution across row
processed_df = proc.contribution(
df,
orientation=PostProcessingContributionOrientation.ROW,
)
self.assertListEqual(processed_df.columns.tolist(),
[DTTM_ALIAS, "a", "b"])
self.assertListEqual(processed_df["a"].tolist(), [0.5, 0.25])
self.assertListEqual(processed_df["b"].tolist(), [0.5, 0.75])
# cell contribution across column without temporal column
df.pop(DTTM_ALIAS)
processed_df = proc.contribution(
df, orientation=PostProcessingContributionOrientation.COLUMN)
self.assertListEqual(processed_df.columns.tolist(), ["a", "b"])
self.assertListEqual(processed_df["a"].tolist(), [0.25, 0.75])
self.assertListEqual(processed_df["b"].tolist(), [0.1, 0.9])
# contribution only on selected columns
processed_df = proc.contribution(
df,
orientation=PostProcessingContributionOrientation.COLUMN,
columns=["a"],
rename_columns=["pct_a"],
)
self.assertListEqual(processed_df.columns.tolist(),
["a", "b", "pct_a"])
self.assertListEqual(processed_df["a"].tolist(), [1, 3])
self.assertListEqual(processed_df["b"].tolist(), [1, 9])
self.assertListEqual(processed_df["pct_a"].tolist(), [0.25, 0.75])