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__all_SetupSteps_for_EvaluationModel_population_slices(
fixture__population_slice_setup_steps,
fixture__population_slice__expected_columns,
tmpdir,
):
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"),
)
data_handler = ModelDataHandler(
database_type="sqlite", location=tmpdir, name="jobpath_evaluation",
)
evaluation_model = EvaluationModel(
data_handler=data_handler,
population_slice_generator=population_slice_generator,
)
evaluation_model.add_population_slices()
results = evaluation_model.population_slices[
PopulationSliceID(date=pd.Timestamp("2016-07-01"))
]
assert set(results.data.columns) == set(fixture__population_slice__expected_columns)
# Manually check how many people are on LR and eligible
assert len(results.data) == 315654
assert len(results.data[results.data["eligible_population"]]) == 86240
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__EvaluationModel__add_slices(fixture__population_slice_generator,):
evaluation_model = EvaluationModel(
population_slice_generator=fixture__population_slice_generator,
)
evaluation_model.add_population_slices()
results = evaluation_model.population_slices[
PopulationSliceID(pd.Timestamp("2016-07-01 00:00:00", freq="QS-JAN"))
]
assert results.data.shape == (90, 5,)
def test__AgeEligible__ge_min(fixture__date_of_birth_df):
ge_min = AgeEligible(date_of_birth_col="date_of_birth", min_eligible={"years": 25})
results = ge_min.run(
PopulationSliceID(date=pd.Timestamp("2016-01-01")),
data=fixture__date_of_birth_df,
)
# 22 out of 30 records have date_of_birth more than 60 years before date
# Should be 2 columns in results df (date_of_birth and age_eligible)
assert results.loc[results["age_eligible"]].shape == (25, 2)
def test__ClaimDurationEligible__ge_min(fixture__claim_duration_df):
eligible = ClaimDurationEligible(
claim_start_col="clm_comm_date", min_eligible={"years": 1}
)
results = eligible.run(
PopulationSliceID(date=pd.Timestamp("2016-01-01")),
data=fixture__claim_duration_df,
)
assert results.loc[results["claim_duration_eligible"]].shape == (28, 2)
def test__ClaimCodeEligible():
data = pd.DataFrame(
{"lr_code": ["UA", "UB", "UC", "UD", "UE", "UA2", "UB2", "UC2", "UD2", "UE2"]}
)
eligible = ClaimCodeEligible(code_col="lr_code", eligible_codes=["UA", "UB"])
results = eligible.run(
PopulationSliceID(date=pd.Timestamp("2016-01-01")), data=data
)
assert results.loc[results["claim_code_eligible"]].shape == (2, 2)
def test__OnJobPath():
"""Test basic case using just JobPath operational data and not ISTS flag.
"""
eligible = OnJobPath(assumed_episode_length={"years": 1})
psid = PopulationSliceID(date=pd.Timestamp("2016-02-01"))
lrp = LiveRegisterPopulation(
columns_by_type={"JobPath_Flag": "boolean", "JobPathHold": "boolean",}
)
results = eligible.run(data_id=psid, data=lrp.run(psid))
# Manually check number of people on JobPath at start of Feb 2016 == 1441
assert results.loc[results["on_jobpath"]].shape == (1441, 3)
def test__PopulationSlice(fixture__RandomPopulation,
fixture__SampleFromPopulation):
setup_steps = SetupSteps([
fixture__RandomPopulation(),
fixture__SampleFromPopulation(0.1),
])
results = PopulationSlice(
id=PopulationSliceID(date=pd.Timestamp("2016-01-01")),
setup_steps=setup_steps,
)
assert results.data.shape == (10, 5)
def fixture__population_slice(fixture__RandomPopulation,
fixture__SampleFromPopulation):
setup_steps = SetupSteps([
fixture__RandomPopulation(),
fixture__SampleFromPopulation(0.1),
])
population_slice = PopulationSlice(
id=PopulationSliceID(date=pd.Timestamp("2016-01-01")),
setup_steps=setup_steps,
)
return population_slice
def test__all_SetupSteps_for_PopulationSlice(
fixture__population_slice_setup_steps, fixture__population_slice__expected_columns
):
results = PopulationSlice(
id=PopulationSliceID(date=pd.Timestamp("2016-07-01")),
setup_steps=fixture__population_slice_setup_steps,
)
assert set(results.data.columns) == set(fixture__population_slice__expected_columns)
# Manually check how many people are on LR and eligible
assert len(results.data) == 315654
assert len(results.data[results.data["eligible_population"]]) == 86240
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__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__ModelDataHandler__run__new(fixture__setup_steps_by_date,
fixture__population_slice_generator,
tmpdir):
data_path = f"sqlite:///{tmpdir}/test.db"
data_handler = ModelDataHandler(data_path)
population_slice_generator = fixture__population_slice_generator
results = {
population_slice.id: population_slice
for population_slice in population_slice_generator.run(data_handler)
}
key = PopulationSliceID(date=pd.Timestamp("2016-07-01", freq="QS-JAN"))
assert results[key].data.shape == (
90,
5,
)
def test__PopulationSliceGenerator(fixture__setup_steps_by_date,
fixture__population_slice_generator):
population_slice_generator = fixture__population_slice_generator
results = {
population_slice.id: population_slice
for population_slice in population_slice_generator.run()
}
key = PopulationSliceID(date=pd.Timestamp("2016-07-01", freq="QS-JAN"))
assert results[key].data.shape == (
90,
5,
)
assert population_slice_generator.date_range.equals(
pd.date_range(start=pd.Timestamp("2016-01-01"),
end=pd.Timestamp("2017-12-31"),
freq="QS"))
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__ModelDataHandler__run__existing(fixture__setup_steps_by_date,
fixture__population_slice_generator,
tmpdir):
data_path = f"sqlite:///{tmpdir}/test.db"
data_handler = ModelDataHandler(data_path)
population_slice_generator = fixture__population_slice_generator
# First iteration should run setup_steps then write to storage
first_population_slices = {
population_slice.id: population_slice
for population_slice in population_slice_generator.run(data_handler)
}
# Second iteration should just read from storage
second_population_slices = {
population_slice.id: population_slice
for population_slice in population_slice_generator.run(data_handler)
}
key = PopulationSliceID(date=pd.Timestamp("2016-07-01", freq="QS-JAN"))
assert (len(first_population_slices[key].data) == len(
second_population_slices[key].data))
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