def get_kde_numeric_attribute(log, attribute, parameters=None):
"""
Gets the KDE estimation for the distribution of a numeric attribute values
Parameters
-------------
log
Event stream object (if log, is converted)
attribute
Numeric attribute to analyse
parameters
Possible parameters of the algorithm, including:
graph_points -> number of points to include in the graph
Returns
--------------
x
X-axis values to represent
y
Y-axis values to represent
"""
if type(log) is EventLog:
event_log = transform.transform_event_log_to_event_stream(log)
else:
event_log = log
values = [event[attribute] for event in event_log if attribute in event]
return attributes_common.get_kde_numeric_attribute(values,
parameters=parameters)
def get_kde_numeric_attribute(df, attribute, parameters=None):
"""
Gets the KDE estimation for the distribution of a numeric attribute values
Parameters
-------------
df
Pandas dataframe
attribute
Numeric attribute to analyse
parameters
Possible parameters of the algorithm, including:
graph_points -> number of points to include in the graph
Returns
--------------
x
X-axis values to represent
y
Y-axis values to represent
"""
values = list(df.dropna(subset=[attribute])[attribute])
return attributes_common.get_kde_numeric_attribute(values,
parameters=parameters)
def get_kde_numeric_attribute(df, attribute, parameters=None):
"""Gets the KDE estimation for the distribution of a numeric attribute values
"""
values = df.select(attribute).rdd.map(lambda row: row[0]).collect()
return attributes_common.get_kde_numeric_attribute(values,
parameters=parameters)
def get_numeric_attribute(self, attribute_key, parameters=None):
if parameters is None:
parameters = {}
url = self.get_url("getNumericAttributeValues", parameters={"attribute_key": attribute_key})
r = requests.get(url)
ret_text = r.text
ret_json = json.loads(ret_text)
ret = ret_json["points"]
x, y = attributes_common.get_kde_numeric_attribute(ret)
return x, y
def get_numeric_attribute(self, attribute_key, parameters=None):
if parameters is None:
parameters = {}
list_logs = self.get_list_logs()
for key in self.init_parameters:
if key not in parameters:
parameters[key] = self.init_parameters[key]
parameters["filters"] = self.filters
parameters["attribute_key"] = attribute_key
ret = parquet_handler.get_case_duration(".", self.distr_log_path, list_logs, parameters=parameters)
x, y = attributes_common.get_kde_numeric_attribute(ret)
return x, y