def sample_log(log, no_traces=100):
"""
Randomly sample a fixed number of traces from the original log_skeleton
Parameters
-----------
log
Log
no_traces
Number of traces that the sample should have
Returns
-----------
newLog
Filtered log_skeleton
"""
new_log = EventLog(attributes=log.attributes, extensions=log.extensions, globals=log._omni,
classifiers=log.classifiers)
new_log._list = random.sample(log, min(no_traces, len(log)))
return new_log
def filter_log_by_relevance(topK, log, relevance_scores):
log_new = EventLog()
for label in relevance_scores:
topK = len(relevance_scores[label]['scores']) if len(
relevance_scores[label]['scores']) < topK else topK
relevance_scores[label]['scores'] = dict(
sorted(relevance_scores[label]['scores'].items(),
key=lambda x: x[1],
reverse=True)[:topK])
log_dummy = filter_log_by_caseid(log,
relevance_scores[label]['traces'])
log_dummy = attributes_filter.apply_events(
log_dummy,
relevance_scores[label]['scores'].keys(),
parameters={
attributes_filter.PARAMETER_CONSTANT_ATTRIBUTE_KEY:
"concept:name",
"positive": True
})
log_new._list = log_new._list + log_dummy._list
return relevance_scores, log_new
