def apply(ocel: OCEL, parameters: Optional[Dict[Any, Any]] = None) -> OCEL:
"""
Explode an OCEL: an event associated to N objects is "split" to N events, each one associated to one object.
Parameters
-----------------
ocel
Object-centric event log
parameters
Possible parameters of the algorithm
Returns
-----------------
ocel
Exploded object-centric event log
"""
if parameters is None:
parameters = {}
ocel = deepcopy(ocel)
ocel.relations[ocel.event_id_column] = ocel.relations[ocel.event_id_column] + "_" + ocel.relations[ocel.object_id_column]
ocel.events = ocel.relations.copy()
del ocel.events[ocel.object_id_column]
del ocel.events[ocel.object_type_column]
return ocel
def apply(ocel: OCEL,
values: Collection[Any],
parameters: Optional[Dict[Any, Any]] = None) -> OCEL:
"""
Filters the object-centric event log on the provided event attributes values
Parameters
----------------
ocel
Object-centric event log
values
Collection of values
parameters
Parameters of the algorithm, including:
- Parameters.ATTRIBUTE_KEY => the attribute that should be filtered
- Parameters.POSITIVE => decides if the values should be kept (positive=True) or removed (positive=False)
Returns
----------------
ocel
Filtered object-centric event log
"""
if parameters is None:
parameters = {}
attribute_key = exec_utils.get_param_value(Parameters.ATTRIBUTE_KEY,
parameters, ocel.event_activity)
positive = exec_utils.get_param_value(Parameters.POSITIVE, parameters,
True)
ocel = copy(ocel)
if positive:
ocel.events = ocel.events[ocel.events[attribute_key].isin(values)]
else:
ocel.events = ocel.events[~ocel.events[attribute_key].isin(values)]
return filtering_utils.propagate_event_filtering(ocel,
parameters=parameters)
def filter_end_events_per_object_type(
ocel: OCEL,
object_type: str,
parameters: Optional[Dict[Any, Any]] = None) -> OCEL:
"""
Filters the events in which an object for the given object type terminates its lifecycle.
(E.g. an event with activity "Pay Order" might terminate an order).
Parameters
------------------
ocel
Object-centric event log
object_type
Object type to consider
parameters
Parameters of the algorithm, including:
- Parameters.EVENT_ID => the attribute working as event identifier
- Parameters.OBJECT_ID => the attribute working as object identifier
- Parameters.OBJECT_TYPE => the attribute working as object type
Returns
------------------
filtered_ocel
Filtered object-centric event log
"""
if parameters is None:
parameters = {}
event_id = exec_utils.get_param_value(Parameters.EVENT_ID, parameters,
ocel.event_id_column)
object_id = exec_utils.get_param_value(Parameters.OBJECT_ID, parameters,
ocel.object_id_column)
object_type_column = exec_utils.get_param_value(Parameters.OBJECT_TYPE,
parameters,
ocel.object_type_column)
evs = set(
ocel.relations[ocel.relations[object_type_column] == object_type].
groupby(object_id).last()[event_id].unique())
ocel = copy(ocel)
ocel.events = ocel.events[ocel.events[event_id].isin(evs)]
return filtering_utils.propagate_event_filtering(ocel,
parameters=parameters)
def sample_ocel_events(ocel: OCEL,
parameters: Optional[Dict[Any, Any]] = None) -> OCEL:
"""
Keeps a sample of the events of an object-centric event log
Parameters
------------------
ocel
Object-centric event log
parameters
Parameters of the algorithm, including:
- Parameters.EVENT_ID => event identifier
- Parameters.NUM_EVENTS => number of events
Returns
------------------
sampled_ocel
Sampled object-centric event log
"""
if parameters is None:
parameters = {}
event_id_column = exec_utils.get_param_value(Parameters.EVENT_ID,
parameters,
ocel.event_id_column)
num_entities = exec_utils.get_param_value(Parameters.NUM_ENTITIES,
parameters, 100)
events = list(ocel.events[event_id_column].unique())
num_events = min(len(events), num_entities)
random.shuffle(events)
picked_events = events[:num_events]
ocel = copy(ocel)
ocel.events = ocel.events[ocel.events[event_id_column].isin(picked_events)]
return filtering_utils.propagate_event_filtering(ocel,
parameters=parameters)
def apply_timestamp(ocel: OCEL,
min_timest: Union[datetime.datetime, str],
max_timest: Union[datetime.datetime, str],
parameters: Optional[Dict[Any, Any]] = None) -> OCEL:
"""
Filters the object-centric event log keeping events in the provided timestamp range
Parameters
-----------------
ocel
Object-centric event log
min_timest
Left extreme of the allowed timestamp interval (provided in the format: YYYY-mm-dd HH:MM:SS)
max_timest
Right extreme of the allowed timestamp interval (provided in the format: YYYY-mm-dd HH:MM:SS)
parameters
Parameters of the algorithm, including:
- Parameters.TIMESTAMP_KEY => the attribute to use as timestamp
Returns
-----------------
filtered_ocel
Filtered object-centric event log
"""
if parameters is None:
parameters = {}
timestamp_key = exec_utils.get_param_value(Parameters.TIMESTAMP_KEY,
parameters,
ocel.event_timestamp)
min_timest = get_dt_from_string(min_timest)
max_timest = get_dt_from_string(max_timest)
ocel = copy(ocel)
ocel.events = ocel.events[ocel.events[timestamp_key] >= min_timest]
ocel.events = ocel.events[ocel.events[timestamp_key] <= max_timest]
return filtering_utils.propagate_event_filtering(ocel,
parameters=parameters)
def propagate_object_filtering(ocel: OCEL,
parameters: Optional[Dict[Any,
Any]] = None) -> OCEL:
"""
Propagates the filtering at the object level to the remaining parts of the OCEL structure
(events, relations)
Parameters
----------------
ocel
Object-centric event log
parameters
Parameters of the algorithm, including:
- Parameters.EVENT_ID => the column to be used as case identifier
- Parameters.OBJECT_ID => the column to be used as object identifier
- Parameters.OBJECT_TYPE => the column to be used as object type
Returns
----------------
ocel
Object-centric event log with propagated filter
"""
if parameters is None:
parameters = {}
event_id = exec_utils.get_param_value(Parameters.EVENT_ID, parameters,
ocel.event_id_column)
object_id = exec_utils.get_param_value(Parameters.OBJECT_ID, parameters,
ocel.object_id_column)
selected_object_ids = set(ocel.objects[object_id].unique())
ocel.relations = ocel.relations[ocel.relations[object_id].isin(
selected_object_ids)]
selected_event_ids = set(ocel.relations[event_id].unique())
ocel.events = ocel.events[ocel.events[event_id].isin(selected_event_ids)]
return ocel
def apply(ocel: OCEL,
min_num_obj_type: Dict[str, int],
parameters: Optional[Dict[Any, Any]] = None) -> OCEL:
"""
Filters the events of the object-centric logs which are related to at least
the specified amount of objects per type.
E.g. apply(ocel, {"order": 1, "element": 2})
Would keep the following events:
ocel:eid ocel:timestamp ocel:activity ocel:type:element ocel:type:order
0 e1 1980-01-01 Create Order [i4, i1, i3, i2] [o1]
1 e11 1981-01-01 Create Order [i6, i5] [o2]
2 e14 1981-01-04 Create Order [i8, i7] [o3]
Parameters
------------------
ocel
Object-centric event log
min_num_obj_type
Minimum number of objects per type
parameters
Parameters of the filter, including:
- Parameters.EVENT_ID => the event identifier
- Parameters.OBJECT_ID => the object identifier
- Parameters.OBJECT_TYPE => the object type
Returns
-----------------
filtered_event_log
Filtered object-centric event log
"""
if parameters is None:
parameters = {}
event_id = exec_utils.get_param_value(Parameters.EVENT_ID, parameters,
ocel.event_id_column)
object_id = exec_utils.get_param_value(Parameters.OBJECT_ID, parameters,
ocel.object_id_column)
object_type = exec_utils.get_param_value(Parameters.OBJECT_TYPE,
parameters,
ocel.object_type_column)
num_obj = objects_ot_count.get_objects_ot_count(ocel,
parameters=parameters)
filt_evs = set()
for evid, evobjs in num_obj.items():
is_ok = True
for k, v in min_num_obj_type.items():
if not k in evobjs:
is_ok = False
break
elif evobjs[k] < v:
is_ok = False
break
if is_ok:
filt_evs.add(evid)
ocel = copy(ocel)
ocel.events = ocel.events[ocel.events[event_id].isin(filt_evs)]
return filtering_utils.propagate_event_filtering(ocel,
parameters=parameters)