def test_parse_eligibilities(self):
event = {
"classifier_set": ["a", "?"],
"age_eligibility": [],
"eligibility": [1],
"target": [1.0, 1, "A", "1", "dist1", "1", 100]
}
classifiers, classifier_values, aggregates = \
self.get_mock_classifiers()
e = self.assemble_disturbance_events_table([event])
sit_events = sit_disturbance_event_parser.parse(
e,
classifiers,
classifier_values,
aggregates,
self.get_mock_disturbance_types(),
None,
separate_eligibilities=True)
elgibilities_input = pd.DataFrame(
columns=["id", "pool_filter", "state_filter"],
data=[[1, "pool_expression_1", "state_expression_1"]])
sit_eligibilities = sit_disturbance_event_parser.parse_eligibilities(
sit_events, elgibilities_input)
self.assertTrue(
list(sit_eligibilities.columns) == [
x["name"]
for x in sit_format.get_disturbance_eligibility_format()
])
def test_undefined_sort_type_error(self):
"""check if an error is raised when an invalid sort type is specified
"""
event = {
"classifier_set": ["a", "?"],
"age_eligibility": ["True", -1, -1, -1, -1],
"eligibility": [-1] * self.get_num_eligibility_cols(),
"target": [1.0, "UNDEFINED", "A", 100, "dist1", 2, 100]
}
classifiers, classifier_values, aggregates = \
self.get_mock_classifiers()
with self.assertRaises(ValueError):
e = self.assemble_disturbance_events_table([event])
sit_disturbance_event_parser.parse(
e, classifiers, classifier_values, aggregates,
self.get_mock_disturbance_types(), self.get_mock_age_classes())
def test_incorrect_number_of_classifiers_error(self):
"""checks that the format has the correct number of columns
according to the defined classifiers
"""
event = {
"classifier_set": ["a", "?", "EXTRA"],
"age_eligibility": ["False", -1, -1, -1, -1],
"eligibility": [-1] * self.get_num_eligibility_cols(),
"target": [1.0, "1", "A", 100, "dist1", 2, 100]
}
classifiers, classifier_values, aggregates = \
self.get_mock_classifiers()
with self.assertRaises(ValueError):
e = self.assemble_disturbance_events_table([event])
sit_disturbance_event_parser.parse(
e, classifiers, classifier_values, aggregates,
self.get_mock_disturbance_types(), self.get_mock_age_classes())
def test_undefined_age_class_error(self):
"""check that an error is raised if the using_age_class is set
to true and any of the age class ids are not valid
"""
event = {
"classifier_set": ["a", "?"],
"age_eligibility": ["True", -1, -1, -1, "UNDEFINED"],
"eligibility": [-1] * self.get_num_eligibility_cols(),
"target": [1.0, "1", "A", 100, "dist1", 2, 100]
}
classifiers, classifier_values, aggregates = \
self.get_mock_classifiers()
with self.assertRaises(ValueError):
e = self.assemble_disturbance_events_table([event])
sit_disturbance_event_parser.parse(
e, classifiers, classifier_values, aggregates,
self.get_mock_disturbance_types(), self.get_mock_age_classes())
def test_differing_hw_sw_age_criteria_error(self):
"""check that an error is raised if hw age and sw age criteria
differ (CBM only has stand age)
"""
event = {
"classifier_set": ["a", "?"],
"age_eligibility": ["False", -1, 10, -1, 20],
"eligibility": [-1] * self.get_num_eligibility_cols(),
"target": [1.0, "1", "A", 100, "dist1", 2, 100]
}
classifiers, classifier_values, aggregates = \
self.get_mock_classifiers()
with self.assertRaises(ValueError):
e = self.assemble_disturbance_events_table([event])
sit_disturbance_event_parser.parse(
e, classifiers, classifier_values, aggregates,
self.get_mock_disturbance_types(), self.get_mock_age_classes())
def test_undefined_classifier_value_error(self):
"""checks that the format has values that are either wildcards or
classifier sets drawn from the set of defined classifiers values
and aggregates
"""
event = {
"classifier_set": ["UNDEFINED", "?"],
"age_eligibility": ["False", -1, -1, -1, -1],
"eligibility": [-1] * self.get_num_eligibility_cols(),
"target": [1.0, "1", "A", 100, "dist1", 2, 100]
}
classifiers, classifier_values, aggregates = \
self.get_mock_classifiers()
with self.assertRaises(ValueError):
e = self.assemble_disturbance_events_table([event])
sit_disturbance_event_parser.parse(
e, classifiers, classifier_values, aggregates,
self.get_mock_disturbance_types(), self.get_mock_age_classes())
def test_expected_value_with_numeric_classifier_values(self):
"""Checks that numeric classifiers that appear in events data
are parsed as strings
"""
event = {
"classifier_set": [1, 2.0],
"age_eligibility": ["False", -1, -1, -1, -1],
"eligibility": [-1] * self.get_num_eligibility_cols(),
"target": [1.0, "1", "A", 100, "dist1", 2, 100]
}
classifiers = pd.DataFrame(data=[(1, "classifier1"),
(2, "classifier2")],
columns=["id", "name"])
classifier_values = pd.DataFrame(
data=[(1, "1", "a"), (1, "b", "b"), (2, "a", "a")],
columns=["classifier_id", "name", "description"])
aggregates = [{
'classifier_id': 1,
'name': 'agg1',
'description': 'agg2',
'classifier_values': ['a', 'b']
}, {
'classifier_id': 1,
'name': 'agg2',
'description': 'agg2',
'classifier_values': ['a', 'b']
}, {
'classifier_id': 2,
'name': '2.0',
'description': 'agg1',
'classifier_values': ['a']
}]
e = self.assemble_disturbance_events_table([event])
result = sit_disturbance_event_parser.parse(
e, classifiers, classifier_values, aggregates,
self.get_mock_disturbance_types(), self.get_mock_age_classes())
self.assertTrue(list(result.classifier1) == ["1"])
self.assertTrue(list(result.classifier2) == ["2.0"])
def test_parse_eligibilities_error_on_missing_id(self):
event = {
"classifier_set": ["a", "?"],
"age_eligibility": [],
"eligibility": [2], # missing
"target": [1.0, 1, "A", "1", "dist1", "1", 100]
}
classifiers, classifier_values, aggregates = \
self.get_mock_classifiers()
e = self.assemble_disturbance_events_table([event])
sit_events = sit_disturbance_event_parser.parse(
e,
classifiers,
classifier_values,
aggregates,
self.get_mock_disturbance_types(),
None,
separate_eligibilities=True)
elgibilities_input = pd.DataFrame(
columns=["id", "pool_filter", "state_filter"], data=[[1, "", ""]])
with self.assertRaises(ValueError):
sit_disturbance_event_parser.parse_eligibilities(
sit_events, elgibilities_input)
def parse(sit_classifiers,
sit_disturbance_types,
sit_age_classes,
sit_inventory,
sit_yield,
sit_events=None,
sit_transitions=None,
sit_eligibilities=None):
"""Parses and validates CBM Standard import tool formatted data including
the complicated interdependencies in the SIT format. Returns an object
containing the validated result.
The returned object has the following properties:
- classifiers: a pandas.DataFrame of classifiers in the sit_classifiers
input
- classifier_values: a pandas.DataFrame of the classifier values in the
sit_classifiers input
- classifier_aggregates: a dictionary of the classifier aggregates
in the sit_classifiers input
- disturbance_types: a pandas.DataFrame based on the disturbance types in
the sit_disturbance_types input
- age_classes: a pandas.DataFrame of the age classes based on
sit_age_classes
- inventory: a pandas.DataFrame of the inventory based on sit_inventory
- yield_table: a pandas.DataFrame of the merchantable volume yield curves
in the sit_yield input
- disturbance_events: a pandas.DataFrame of the disturbance events based
on sit_events. If the sit_events parameter is None this field is None.
- transition_rules: a pandas.DataFrame of the transition rules based on
sit_transitions. If the sit_transitions parameter is None this field
is None.
- disturbance_eligibilities: a pandas.DataFrame of the disturbance event
eligibilities based on sit_eligibilities. If the sit_events parameter
is None this field is None.
Args:
sit_classifiers (pandas.DataFrame): SIT formatted classifiers
sit_disturbance_types (pandas.DataFrame): SIT formatted disturbance
types
sit_age_classes (pandas.DataFrame): SIT formatted age classes
sit_inventory (pandas.DataFrame): SIT formatted inventory
sit_yield (pandas.DataFrame): SIT formatted yield curves
sit_events (pandas.DataFrame, optional): SIT formatted disturbance
events
sit_transitions (pandas.DataFrame, optional): SIT formatted transition
rules. Defaults to None.
sit_eligibilities (pandas.DataFrame, optional): SIT formatted
disturbance eligibilities. Defaults to None.
Returns:
object: an object containing parsed and validated SIT dataset
"""
s = SimpleNamespace()
classifiers, classifier_values, classifier_aggregates = \
sit_classifier_parser.parse(sit_classifiers)
s.classifiers = classifiers
s.classifier_values = classifier_values
s.classifier_aggregates = classifier_aggregates
s.disturbance_types = sit_disturbance_type_parser.parse(
sit_disturbance_types)
s.age_classes = sit_age_class_parser.parse(sit_age_classes)
s.inventory = sit_inventory_parser.parse(sit_inventory, classifiers,
classifier_values,
s.disturbance_types,
s.age_classes)
s.yield_table = sit_yield_parser.parse(sit_yield, s.classifiers,
s.classifier_values, s.age_classes)
if sit_events is not None:
separate_eligibilities = False
if sit_eligibilities is not None:
separate_eligibilities = True
s.disturbance_events = sit_disturbance_event_parser.parse(
sit_events, s.classifiers, s.classifier_values,
s.classifier_aggregates, s.disturbance_types, s.age_classes,
separate_eligibilities)
if sit_eligibilities is not None:
s.disturbance_eligibilities = \
sit_disturbance_event_parser.parse_eligibilities(
s.disturbance_events, sit_eligibilities)
else:
s.disturbance_eligibilities = None
else:
s.disturbance_events = None
s.disturbance_eligibilities = None
if sit_transitions is not None:
s.transition_rules = sit_transition_rule_parser.parse(
sit_transitions, s.classifiers, s.classifier_values,
s.classifier_aggregates, s.disturbance_types, s.age_classes)
else:
s.transition_rules = None
return s
def create_sit_rule_based_processor(
sit,
cbm,
random_func=np.random.rand,
reset_parameters=True,
sit_events=None,
sit_disturbance_eligibilities=None,
sit_transition_rules=None,
event_sort: EventSort = EventSort.disturbance_type):
"""initializes a class for processing SIT rule based disturbances.
Args:
sit (object): sit instance as returned by :py:func:`load_sit`
cbm (object): initialized instance of the CBM model
random_func (func, optional): A function of a single integer that
returns a numeric 1d array whose length is the integer argument.
Defaults to np.random.rand.
reset_parameters (bool): if set to true,
cbm_vars.parameters.disturbance_type and
cbm_vars.parameters.reset_age will be reset prior to computing
new disturbances and transition rules.
sit_events (pandas.DataFrame, optional): if specified the returned rule
base processor is based on the specified sit_events input. The
value will be parsed and validated (sit_classifiers,
sit_disturbance_type etc.) based on the values in the specified sit
object.
sit_disturbance_eligibilities (pandas.DataFrame, optional): SIT
formatted disturbance eligibilities. Cannot be specified without
also specified sit_events using the disturbance-eligibility
formatting. Defaults to None.
sit_transition_rules (pandas.DataFrame, optional): if specified the
returned rule base processor is based on the specified
sit_transition_rules input. The value will be parsed and validated
(sit_classifiers, sit_disturbance_type etc.) based on the values in
the specified sit object. Note if the sit_disturbance_events
parameter is set, but this parameter is not set the
transition_rules (if any) attached to the specified sit object will
be used by default. If null transition rules are required with
non-null sit_events set this parameter to a dataframe with zero
rows `pandas.DataFrame()`. Defaults to None.
event_sort (EventSort): one of the EventSort values, which determines
the order in which the supplied sit_events are applied within a
given timestep.
Raises:
ValueError: cannot specify sit_disturbance_eligibilities with no
specified sit_events
Returns:
SITRuleBasedProcessor: an object for processing SIT rule based
disturbances
"""
separate_eligibilities = sit_disturbance_eligibilities is not None
disturbance_events = None
disturbance_eligibilities = None
transition_rules = None
if sit_events is not None:
disturbance_events = sit_disturbance_event_parser.parse(
sit_events,
sit.sit_data.classifiers,
sit.sit_data.classifier_values,
sit.sit_data.classifier_aggregates,
sit.sit_data.disturbance_types,
sit.sit_data.age_classes,
separate_eligibilities=separate_eligibilities)
if sit_disturbance_eligibilities is not None:
disturbance_eligibilities = \
sit_disturbance_event_parser.parse_eligibilities(
sit_events, sit_disturbance_eligibilities)
else:
disturbance_events = sit.sit_data.disturbance_events
disturbance_eligibilities = sit.sit_data.disturbance_eligibilities
if separate_eligibilities:
raise ValueError(
"cannot specify sit_disturbance_eligibilities with no "
"specified sit_events")
if sit_transition_rules is not None:
if len(sit_transition_rules.index) == 0:
transition_rules = None
else:
transition_rules = sit_transition_rule_parser.parse(
sit_transition_rules, sit.sit_data.classifiers,
sit.sit_data.classifier_values,
sit.sit_data.classifier_aggregates,
sit.sit_data.disturbance_types, sit.sit_data.age_classes)
else:
transition_rules = sit.sit_data.transition_rules
classifiers_config = get_classifiers(sit.sit_data.classifiers,
sit.sit_data.classifier_values)
tr_constants = SimpleNamespace(
group_err_max=sit_transition_rule_parser.GROUPED_PERCENT_ERR_MAX,
classifier_value_postfix=sit_format.get_tr_classifier_set_postfix(),
wildcard=sit_classifier_parser.get_wildcard_keyword())
return sit_rule_based_processor.sit_rule_based_processor_factory(
cbm=cbm,
random_func=random_func,
classifiers_config=classifiers_config,
classifier_aggregates=sit.sit_data.classifier_aggregates,
sit_events=_initialize_events(disturbance_events, sit.sit_mapping,
sit.sit_data.disturbance_types,
event_sort),
sit_transitions=_initialize_transition_rules(transition_rules,
sit.sit_mapping),
tr_constants=tr_constants,
sit_disturbance_eligibilities=disturbance_eligibilities,
reset_parameters=reset_parameters)