def test__all_SetupSteps__subsequent_TreatmentPeriod(
fixture__population_slice_setup_steps,
fixture__treatment_period_setup_steps,
fixture__treatment_period_expected_columns,
):
ps = PopulationSlice(
id=PopulationSliceID(date=pd.Timestamp("2016-07-01")),
setup_steps=fixture__population_slice_setup_steps,
)
first_tpid = TreatmentPeriodID(
population_slice_id=ps.id, time_period=pd.Period("2016-07")
)
first_tp = TreatmentPeriod(
id=first_tpid,
setup_steps=fixture__treatment_period_setup_steps,
init_data=ps.data,
)
second_tpid = TreatmentPeriodID(
population_slice_id=ps.id, time_period=pd.Period("2016-08")
)
results = TreatmentPeriod(
id=second_tpid,
setup_steps=fixture__treatment_period_setup_steps,
init_data=first_tp.data.copy(),
)
assert set(results.data.columns) == set(fixture__treatment_period_expected_columns)
len_expected = len(first_tp.data[first_tp.data["eligible_population"]]) - len(
first_tp.data[first_tp.data["evaluation_group"] == "T"]
)
assert len(results.data) <= len_expected
assert len(results.data[results.data["eligible_population"]]) < len(results.data)
def test__StartingPopulation():
data = pd.DataFrame(
{
"original_population": [True] * 20,
"eligible_population": [True] * 8 + [False] * 12,
"evaluation_group": ["T"] * 2 + ["C"] * 6 + [0] * 12,
}
)
any_data_id = TreatmentPeriodID(
population_slice_id=PopulationSliceID(date="2016-01-01"),
time_period=pd.Period("2016-01"),
)
results = TreatmentPeriod(
id=any_data_id,
setup_steps=SetupSteps(
steps=[
StartingPopulation(
eligible_from_previous_period_col="evaluation_group",
starting_pop_label="C",
)
]
),
init_data=data,
)
print(results.data)
# Should be 6 records in new population
# ...corresponding to the 6 records in the original control group
assert len(results.data) == 6
def test__TreatmentPeriodGenerator(fixture__treatment_period_generator,
fixture__population_slice):
treatment_period_generator = fixture__treatment_period_generator
population_slice = fixture__population_slice
results = {
treatment_period.id: treatment_period
for treatment_period in treatment_period_generator.run(
population_slice)
}
assert results[TreatmentPeriodID(
population_slice_id=population_slice.id,
time_period=pd.Period("2016-02"))].data.shape == (8, 5)
def test__JobPathStarts__jobpath_operational(fixture__population_slice_setup_steps):
ps = PopulationSlice(
id=PopulationSliceID(date=pd.Timestamp("2016-01-01")),
setup_steps=fixture__population_slice_setup_steps,
)
tpid = TreatmentPeriodID(
population_slice_id=ps.id, time_period=pd.Period("2016-01")
)
results = TreatmentPeriod(
id=tpid, setup_steps=SetupSteps(steps=[JobPathStarts()]), init_data=ps.data,
)
# Apparently 1336 people started JobPath in Jan 2016
assert len(results.data[results.data["jobpath_starts"]]) == 1336
def test__TreatmentPeriod(
fixture__population_slice,
fixture__SampleFromPopulation,
):
setup_steps = SetupSteps([
fixture__SampleFromPopulation(0.9),
])
results = TreatmentPeriod(
id=TreatmentPeriodID(
population_slice_id=fixture__population_slice.id,
time_period=pd.Period("2016Q1"),
),
setup_steps=setup_steps,
init_data=fixture__population_slice.data,
)
assert results.data.shape == (9, 5)
def test__EvaluationGroup():
data = pd.DataFrame(
{"eligible": [True] * 8 + [False] * 2, "treatment": [True, False] * 5,}
)
any_data_id = TreatmentPeriodID(
population_slice_id=PopulationSliceID(date="2016-01-01"),
time_period=pd.Period("2016-01"),
)
results = TreatmentPeriod(
id=any_data_id,
setup_steps=SetupSteps(
steps=[EvaluationGroup(eligible_col="eligible", treatment_col="treatment")]
),
init_data=data,
)
assert results.data[results.data["evaluation_group"] == "T"].shape == (4, 3)
def test__JobPathStartedEndedSamePeriod(fixture__population_slice_setup_steps):
ps = PopulationSlice(
id=PopulationSliceID(date=pd.Timestamp("2016-07-01")),
setup_steps=fixture__population_slice_setup_steps,
)
tpid = TreatmentPeriodID(
population_slice_id=ps.id, time_period=pd.Period("2016-07")
)
results = TreatmentPeriod(
id=tpid,
setup_steps=SetupSteps(steps=[JobPathStartedEndedSamePeriod()]),
init_data=ps.data,
)
print(results.data["jobpath_started_and_ended"].value_counts())
print(results.data)
# Apparently 2 people started *and* ended JobPath in July 2016
assert len(results.data[results.data["jobpath_started_and_ended"]]) == 2
def test__EvaluationModel__add_periods(
fixture__population_slice_generator, fixture__treatment_period_generator
):
evaluation_model = EvaluationModel(
population_slice_generator=fixture__population_slice_generator,
treatment_period_generator=fixture__treatment_period_generator,
)
evaluation_model.add_population_slices()
evaluation_model.add_treatment_periods()
results = evaluation_model.treatment_periods[
TreatmentPeriodID(
population_slice_id=PopulationSliceID(
pd.Timestamp("2016-01-01", freq="QS-JAN")
),
time_period=pd.Period("2016-06", "M"),
)
]
assert results.data.shape == (48, 5)
def test__all_SetupSteps__first_TreatmentPeriod(
fixture__treatment_period_setup_steps,
fixture__population_slice_setup_steps,
fixture__treatment_period_expected_columns,
):
ps = PopulationSlice(
id=PopulationSliceID(date=pd.Timestamp("2016-07-01")),
setup_steps=fixture__population_slice_setup_steps,
)
tpid = TreatmentPeriodID(
population_slice_id=ps.id, time_period=pd.Period("2016-07")
)
results = TreatmentPeriod(
id=tpid, setup_steps=fixture__treatment_period_setup_steps, init_data=ps.data,
)
print(f"Results columns: {results.data.columns}")
assert set(results.data.columns) == set(fixture__treatment_period_expected_columns)
assert len(results.data) == len(ps.data[ps.data["eligible_population"]])
assert len(results.data[results.data["eligible_population"]]) < len(results.data)
def test__all_SetupSteps_for_EvaluationModel_treatment_periods(
fixture__population_slice_setup_steps,
fixture__treatment_period_setup_steps,
fixture__treatment_period_expected_columns,
tmpdir,
):
data_handler = ModelDataHandler(
database_type="sqlite", location=tmpdir, name="jobpath_evaluation",
)
population_slice_generator = PopulationSliceGenerator(
setup_steps_by_date={
pd.Timestamp("2016-01-01"): fixture__population_slice_setup_steps
},
start=pd.Timestamp("2016-07-01"),
end=pd.Timestamp("2016-09-30"),
)
treatment_period_generator = TreatmentPeriodGenerator(
setup_steps_by_date={
pd.Timestamp("2016-07-01"): fixture__treatment_period_setup_steps
},
end=pd.Period("2016-09"),
)
evaluation_model = EvaluationModel(
data_handler=data_handler,
population_slice_generator=population_slice_generator,
treatment_period_generator=treatment_period_generator,
)
evaluation_model.add_population_slices()
evaluation_model.add_treatment_periods()
results_id = TreatmentPeriodID(
population_slice_id=PopulationSliceID(date=pd.Timestamp("2016-07-01")),
time_period=pd.Period("2016-09"),
)
results = evaluation_model.treatment_periods[results_id]
assert set(results.data.columns) == set(fixture__treatment_period_expected_columns)
# Manually check how many people are on LR and eligible
assert len(results.data[results.data["eligible_population"]]) < len(results.data)
assert len(results.data[results.data["eligible_population"]]) == 64333
assert len(results.data[results.data["jobpath_starts"]]) == 4273