def test_prophet_template_default():
"""Tests prophet_template with default values, for limited data"""
# prepares input data
num_days = 10
data = generate_df_for_tests(freq="D",
periods=num_days,
train_start_date="2018-01-01")
df = data["df"]
template = ProphetTemplate()
config = ForecastConfig(model_template="PROPHET")
params = template.apply_template_for_pipeline_params(df=df, config=config)
# not modified
assert config == ForecastConfig(model_template="PROPHET")
# checks result
metric = EvaluationMetricEnum.MeanAbsolutePercentError
pipeline = params.pop("pipeline", None)
expected_params = dict(
df=df,
time_col=cst.TIME_COL,
value_col=cst.VALUE_COL,
date_format=None,
freq=None,
train_end_date=None,
anomaly_info=None,
# model
regressor_cols=None,
lagged_regressor_cols=None,
estimator=None,
hyperparameter_grid=template.hyperparameter_grid,
hyperparameter_budget=None,
n_jobs=COMPUTATION_N_JOBS,
verbose=1,
# forecast
forecast_horizon=None,
coverage=None,
test_horizon=None,
periods_between_train_test=None,
agg_periods=None,
agg_func=None,
# evaluation
score_func=metric.name,
score_func_greater_is_better=metric.get_metric_greater_is_better(),
cv_report_metrics=CV_REPORT_METRICS_ALL,
null_model_params=None,
relative_error_tolerance=None,
# CV
cv_horizon=None,
cv_min_train_periods=None,
cv_expanding_window=True,
cv_periods_between_splits=None,
cv_periods_between_train_test=None,
cv_max_splits=3)
assert_basic_pipeline_equal(pipeline, template.pipeline)
assert_equal(params, expected_params)
def test_silverkite_template():
"""Tests test_silverkite_template with default config"""
data = generate_df_for_tests(freq="D", periods=10)
df = data["df"]
template = SilverkiteTemplate()
config = ForecastConfig(model_template="SK")
params = template.apply_template_for_pipeline_params(
df=df,
config=config
)
assert config == ForecastConfig(model_template="SK") # not modified
pipeline = params.pop("pipeline", None)
metric = EvaluationMetricEnum.MeanAbsolutePercentError
expected_params = dict(
df=df,
time_col=TIME_COL,
value_col=VALUE_COL,
date_format=None,
freq=None,
train_end_date=None,
anomaly_info=None,
# model
regressor_cols=None,
estimator=None,
hyperparameter_grid=template.hyperparameter_grid,
hyperparameter_budget=None,
n_jobs=COMPUTATION_N_JOBS,
verbose=1,
# forecast
forecast_horizon=None,
coverage=None,
test_horizon=None,
periods_between_train_test=None,
agg_periods=None,
agg_func=None,
# evaluation
score_func=metric.name,
score_func_greater_is_better=metric.get_metric_greater_is_better(),
cv_report_metrics=CV_REPORT_METRICS_ALL,
null_model_params=None,
relative_error_tolerance=None,
# CV
cv_horizon=None,
cv_min_train_periods=None,
cv_expanding_window=True,
cv_periods_between_splits=None,
cv_periods_between_train_test=None,
cv_max_splits=3
)
assert_basic_pipeline_equal(pipeline, template.pipeline)
assert_equal(params, expected_params)
def test_prophet_template_custom():
"""Tests prophet_template with custom values, with long range input"""
# prepares input data
data = generate_df_with_reg_for_tests(freq="H",
periods=300 * 24,
remove_extra_cols=True,
mask_test_actuals=True)
df = data["df"]
time_col = "some_time_col"
value_col = "some_value_col"
df.rename({
cst.TIME_COL: time_col,
cst.VALUE_COL: value_col
},
axis=1,
inplace=True)
# prepares params and calls template
metric = EvaluationMetricEnum.MeanAbsoluteError
# anomaly adjustment adds 10.0 to every record
adjustment_size = 10.0
anomaly_df = pd.DataFrame({
cst.START_DATE_COL: [df[time_col].min()],
cst.END_DATE_COL: [df[time_col].max()],
cst.ADJUSTMENT_DELTA_COL: [adjustment_size],
cst.METRIC_COL: [value_col]
})
anomaly_info = {
"value_col": cst.VALUE_COL,
"anomaly_df": anomaly_df,
"start_date_col": cst.START_DATE_COL,
"end_date_col": cst.END_DATE_COL,
"adjustment_delta_col": cst.ADJUSTMENT_DELTA_COL,
"filter_by_dict": {
cst.METRIC_COL: cst.VALUE_COL
},
"adjustment_method": "add"
}
metadata = MetadataParam(
time_col=time_col,
value_col=value_col,
freq="H",
date_format="%Y-%m-%d-%H",
train_end_date=datetime.datetime(2019, 7, 1),
anomaly_info=anomaly_info,
)
evaluation_metric = EvaluationMetricParam(
cv_selection_metric=metric.name,
cv_report_metrics=[
EvaluationMetricEnum.MedianAbsolutePercentError.name
],
agg_periods=24,
agg_func=np.max,
null_model_params={
"strategy": "quantile",
"constant": None,
"quantile": 0.8
},
relative_error_tolerance=0.01)
evaluation_period = EvaluationPeriodParam(test_horizon=1,
periods_between_train_test=2,
cv_horizon=3,
cv_min_train_periods=4,
cv_expanding_window=True,
cv_periods_between_splits=5,
cv_periods_between_train_test=6,
cv_max_splits=7)
model_components = ModelComponentsParam(
seasonality={
"yearly_seasonality": [True],
"weekly_seasonality": [False],
"daily_seasonality": [4],
"add_seasonality_dict": [{
"yearly": {
"period": 365.25,
"fourier_order": 20,
"prior_scale": 20.0
},
"quarterly": {
"period": 365.25 / 4,
"fourier_order": 15
},
"weekly": {
"period": 7,
"fourier_order": 35,
"prior_scale": 30.0
}
}]
},
growth={"growth_term": "linear"},
events={
"holiday_lookup_countries":
["UnitedStates", "UnitedKingdom", "India"],
"holiday_pre_num_days": [2],
"holiday_post_num_days": [3],
"holidays_prior_scale": [5.0]
},
regressors={
"add_regressor_dict": [{
"regressor1": {
"prior_scale": 10.0,
"mode": 'additive'
},
"regressor2": {
"prior_scale": 20.0,
"mode": 'multiplicative'
},
}]
},
changepoints={
"changepoint_prior_scale": [0.05],
"changepoints": [None],
"n_changepoints": [50],
"changepoint_range": [0.9]
},
uncertainty={
"mcmc_samples": [500],
"uncertainty_samples": [2000]
},
hyperparameter_override={
"input__response__null__impute_algorithm":
"ts_interpolate",
"input__response__null__impute_params": {
"orders": [7, 14]
},
"input__regressors_numeric__normalize__normalize_algorithm":
"RobustScaler",
})
computation = ComputationParam(hyperparameter_budget=10,
n_jobs=None,
verbose=1)
forecast_horizon = 20
coverage = 0.7
config = ForecastConfig(model_template=ModelTemplateEnum.PROPHET.name,
metadata_param=metadata,
forecast_horizon=forecast_horizon,
coverage=coverage,
evaluation_metric_param=evaluation_metric,
evaluation_period_param=evaluation_period,
model_components_param=model_components,
computation_param=computation)
template = ProphetTemplate()
params = template.apply_template_for_pipeline_params(df=df, config=config)
pipeline = params.pop("pipeline", None)
# Adding start_year and end_year based on the input df
model_components.events["start_year"] = df[time_col].min().year
model_components.events["end_year"] = df[time_col].max().year
expected_params = dict(
df=df,
time_col=time_col,
value_col=value_col,
date_format=metadata.date_format,
freq=metadata.freq,
train_end_date=metadata.train_end_date,
anomaly_info=metadata.anomaly_info,
# model
regressor_cols=template.regressor_cols,
estimator=None,
hyperparameter_grid=template.hyperparameter_grid,
hyperparameter_budget=computation.hyperparameter_budget,
n_jobs=computation.n_jobs,
verbose=computation.verbose,
# forecast
forecast_horizon=forecast_horizon,
coverage=coverage,
test_horizon=evaluation_period.test_horizon,
periods_between_train_test=evaluation_period.
periods_between_train_test,
agg_periods=evaluation_metric.agg_periods,
agg_func=evaluation_metric.agg_func,
# evaluation
score_func=metric.name,
score_func_greater_is_better=metric.get_metric_greater_is_better(),
cv_report_metrics=evaluation_metric.cv_report_metrics,
null_model_params=evaluation_metric.null_model_params,
relative_error_tolerance=evaluation_metric.relative_error_tolerance,
# CV
cv_horizon=evaluation_period.cv_horizon,
cv_min_train_periods=evaluation_period.cv_min_train_periods,
cv_expanding_window=evaluation_period.cv_expanding_window,
cv_periods_between_splits=evaluation_period.cv_periods_between_splits,
cv_periods_between_train_test=evaluation_period.
cv_periods_between_train_test,
cv_max_splits=evaluation_period.cv_max_splits)
assert_basic_pipeline_equal(pipeline, template.pipeline)
assert_equal(params, expected_params)
def test_silverkite_template_custom(model_components_param):
""""Tests simple_silverkite_template with custom parameters,
and data that has regressors"""
data = generate_df_with_reg_for_tests(
freq="H",
periods=300*24,
remove_extra_cols=True,
mask_test_actuals=True)
df = data["df"]
time_col = "some_time_col"
value_col = "some_value_col"
df.rename({
TIME_COL: time_col,
VALUE_COL: value_col
}, axis=1, inplace=True)
metric = EvaluationMetricEnum.MeanAbsoluteError
# anomaly adjustment adds 10.0 to every record
adjustment_size = 10.0
anomaly_df = pd.DataFrame({
START_DATE_COL: [df[time_col].min()],
END_DATE_COL: [df[time_col].max()],
ADJUSTMENT_DELTA_COL: [adjustment_size],
METRIC_COL: [value_col]
})
anomaly_info = {
"value_col": VALUE_COL,
"anomaly_df": anomaly_df,
"start_date_col": START_DATE_COL,
"end_date_col": END_DATE_COL,
"adjustment_delta_col": ADJUSTMENT_DELTA_COL,
"filter_by_dict": {METRIC_COL: VALUE_COL},
"adjustment_method": "add"
}
metadata = MetadataParam(
time_col=time_col,
value_col=value_col,
freq="H",
date_format="%Y-%m-%d-%H",
train_end_date=datetime.datetime(2019, 7, 1),
anomaly_info=anomaly_info
)
evaluation_metric = EvaluationMetricParam(
cv_selection_metric=metric.name,
cv_report_metrics=[EvaluationMetricEnum.MedianAbsolutePercentError.name],
agg_periods=24,
agg_func=np.max,
null_model_params={
"strategy": "quantile",
"constant": None,
"quantile": 0.8
},
relative_error_tolerance=0.01
)
evaluation_period = EvaluationPeriodParam(
test_horizon=1,
periods_between_train_test=2,
cv_horizon=3,
cv_min_train_periods=4,
cv_expanding_window=True,
cv_periods_between_splits=5,
cv_periods_between_train_test=6,
cv_max_splits=7
)
computation = ComputationParam(
hyperparameter_budget=10,
n_jobs=None,
verbose=1
)
forecast_horizon = 20
coverage = 0.7
template = SilverkiteTemplate()
params = template.apply_template_for_pipeline_params(
df=df,
config=ForecastConfig(
model_template=ModelTemplateEnum.SK.name,
metadata_param=metadata,
forecast_horizon=forecast_horizon,
coverage=coverage,
evaluation_metric_param=evaluation_metric,
evaluation_period_param=evaluation_period,
model_components_param=model_components_param,
computation_param=computation
)
)
pipeline = params.pop("pipeline", None)
expected_params = dict(
df=df,
time_col=time_col,
value_col=value_col,
date_format=metadata.date_format,
freq=metadata.freq,
train_end_date=metadata.train_end_date,
anomaly_info=metadata.anomaly_info,
# model
regressor_cols=template.regressor_cols,
estimator=None,
hyperparameter_grid=template.hyperparameter_grid,
hyperparameter_budget=computation.hyperparameter_budget,
n_jobs=computation.n_jobs,
verbose=computation.verbose,
# forecast
forecast_horizon=forecast_horizon,
coverage=coverage,
test_horizon=evaluation_period.test_horizon,
periods_between_train_test=evaluation_period.periods_between_train_test,
agg_periods=evaluation_metric.agg_periods,
agg_func=evaluation_metric.agg_func,
relative_error_tolerance=evaluation_metric.relative_error_tolerance,
# evaluation
score_func=metric.name,
score_func_greater_is_better=metric.get_metric_greater_is_better(),
cv_report_metrics=evaluation_metric.cv_report_metrics,
null_model_params=evaluation_metric.null_model_params,
# CV
cv_horizon=evaluation_period.cv_horizon,
cv_min_train_periods=evaluation_period.cv_min_train_periods,
cv_expanding_window=evaluation_period.cv_expanding_window,
cv_periods_between_splits=evaluation_period.cv_periods_between_splits,
cv_periods_between_train_test=evaluation_period.cv_periods_between_train_test,
cv_max_splits=evaluation_period.cv_max_splits
)
assert_basic_pipeline_equal(pipeline, template.pipeline)
assert_equal(params, expected_params)
def test_apply_forecast_config(df_config):
"""Tests `apply_forecast_config`"""
df = df_config["df"]
config = df_config["config"]
model_template = df_config["model_template"]
reg_cols = df_config["reg_cols"]
# The same class can be re-used. `df` and `config` are taken from the function call
# to `apply_forecast_config`. Only `model_template_enum` and
# `default_model_template_name` are persistent in the state.
forecaster = Forecaster()
# no config
with warnings.catch_warnings():
warnings.simplefilter("ignore")
pipeline_params = forecaster.apply_forecast_config(df=df)
template_class = SimpleSilverkiteTemplate # based on `default_model_template_name`
expected_pipeline_params = template_class(
).apply_template_for_pipeline_params(df=df)
assert_basic_pipeline_equal(pipeline_params.pop("pipeline"),
expected_pipeline_params.pop("pipeline"))
assert_equal(pipeline_params, expected_pipeline_params)
assert forecaster.config is not None
assert forecaster.template_class == template_class
assert isinstance(forecaster.template, forecaster.template_class)
assert forecaster.pipeline_params is not None
# custom config
with warnings.catch_warnings():
warnings.simplefilter("ignore")
pipeline_params = forecaster.apply_forecast_config(df=df,
config=config)
template_class = ModelTemplateEnum[
model_template].value.template_class # SimpleSilverkiteTemplate
expected_pipeline_params = template_class(
).apply_template_for_pipeline_params(df, config)
expected_pipeline = expected_pipeline_params.pop("pipeline")
assert_basic_pipeline_equal(pipeline_params.pop("pipeline"),
expected_pipeline)
assert_equal(pipeline_params, expected_pipeline_params)
# Custom `model_template_enum`. Same result, because
# `MySimpleSilverkiteTemplate` has the same apply_template_for_pipeline_params
# as `SimpleSilverkiteTemplate`.
forecaster = Forecaster(model_template_enum=MyModelTemplateEnum)
pipeline_params = forecaster.apply_forecast_config(df=df,
config=config)
assert_basic_pipeline_equal(pipeline_params.pop("pipeline"),
expected_pipeline)
assert_equal(pipeline_params, expected_pipeline_params)
# `model_component` of config is incompatible with model_template
forecaster = Forecaster()
config = ForecastConfig(model_template=ModelTemplateEnum.PROPHET.name,
model_components_param=ModelComponentsParam(
regressors={"regressor_cols": reg_cols}))
with pytest.raises(ValueError) as record:
forecaster.apply_forecast_config(df=df, config=config)
assert "Unexpected key(s) found: {\'regressor_cols\'}. The valid keys are: " \
"dict_keys([\'add_regressor_dict\'])" in str(record)
# metadata of config is incompatible with df
df = df.rename(columns={
TIME_COL: "some_time_col",
VALUE_COL: "some_value_col"
})
with pytest.raises(ValueError, match="ts column is not in input data"):
forecaster.apply_forecast_config(df=df, config=config)