def test_estimate(self, jhu_data, population_data, country):
warnings.simplefilter("ignore", category=UserWarning)
# Setting
snl = Scenario(jhu_data, population_data, country)
snl.first_date = "01Apr2020"
snl.last_date = "01Aug2020"
with pytest.raises(ValueError):
snl.estimate(SIR)
snl.trend(include_init_phase=True, show_figure=False)
snl.disable(phases=["0th"])
with pytest.raises(AttributeError):
snl.estimate_history(phase="1th")
# Parameter estimation
with pytest.raises(KeyError):
snl.estimate(SIR, phases=["30th"])
with pytest.raises(ValueError):
snl.estimate(model=SIR, tau=1440)
snl.enable(phases=["0th"])
with pytest.raises(TypeError):
snl.estimate(model=SIR, phases="1st")
with pytest.raises(ValueError):
snl.estimate(model=SIR, phases=["0th"])
snl.clear(include_past=True)
snl.trend(show_figure=False)
snl.estimate(SIR)
# Estimation history
snl.estimate_history(phase="1st")
# Estimation accuracy
snl.estimate_accuracy(phase="1st")
# Get a value
snl.get(Term.RT)
with pytest.raises(KeyError):
snl.get("feeling")
def test_estimate(self, jhu_data, population_data, country):
warnings.simplefilter("ignore", category=UserWarning)
# Setting
snl = Scenario(jhu_data, population_data, country)
snl.first_date = "01Apr2020"
snl.last_date = "01Aug2020"
with pytest.raises(ValueError):
snl.estimate(SIR)
snl.trend(show_figure=False)
with pytest.raises(AttributeError):
snl.estimate_history(phase="last")
# Parameter estimation
with pytest.raises(KeyError):
snl.estimate(SIR, phases=["30th"])
with pytest.raises(ValueError):
snl.estimate(model=SIR, tau=1440)
snl.estimate(SIR, timeout=1, timeout_iteration=1)
# Estimation history
snl.estimate_history(phase="last")
# Estimation accuracy
snl.estimate_accuracy(phase="last")
# Get a value
snl.get(Term.RT)
with pytest.raises(KeyError):
snl.get("feeling")
def test_analysis(self, jhu_data, population_data):
scenario = Scenario(jhu_data, population_data, country="Italy")
with pytest.raises(KeyError):
scenario.simulate(name="Main", show_figure=False)
with pytest.raises(ValueError):
scenario.estimate(model=SIRF)
# S-R trend analysis
scenario.trend(show_figure=False)
warnings.filterwarnings("ignore", category=UserWarning)
scenario.trend(show_figure=True)
# Parameter estimation of SIR-F model
with pytest.raises(ValueError):
scenario.param_history(targets=["Rt"], show_figure=False)
with pytest.raises(ValueError):
scenario.estimate(model=SIRF, tau=1440)
scenario.estimate(model=SIRF)
# History of estimation
scenario.estimate_history(phase="1st")
with pytest.raises(KeyError):
scenario.estimate_history(phase="0th")
# Accuracy of estimation
scenario.estimate_accuracy(phase="1st")
with pytest.raises(KeyError):
scenario.estimate_accuracy(phase="0th")
# Prediction
scenario.add(name="Main", days=100)
scenario.simulate(name="Main", show_figure=False)
scenario.simulate(name="Main", show_figure=True)
scenario.param_history(targets=["Rt"], show_figure=False)
scenario.param_history(targets=["Rt"], divide_by_first=False)
scenario.param_history(targets=["Rt"], show_box_plot=False)
with pytest.raises(KeyError):
scenario.param_history(targets=["Rt", "Value"])
with pytest.raises(KeyError):
scenario.param_history(targets=["Rt"], box_plot=False)
# New scenario
sigma_new = scenario.get("sigma", phase="last") * 2
with pytest.raises(KeyError):
scenario.get("value")
warnings.filterwarnings("ignore", category=DeprecationWarning)
scenario.add_phase(name="New medicines", days=100, sigma=sigma_new)
# Summarize scenarios
summary_df = scenario.summary()
assert isinstance(summary_df, pd.DataFrame)
desc_df = scenario.describe()
assert isinstance(desc_df, pd.DataFrame)
# Estimation errors
with pytest.raises(TypeError):
scenario.estimate(SIRF, phases="1st")
with pytest.raises(KeyError):
scenario.estimate(SIRF, phases=["100th"])