# pm4py.algo.discovery.batches package¶

## pm4py.algo.discovery.batches.algorithm module¶

class pm4py.algo.discovery.batches.algorithm.Variants(value)[source]

Bases: enum.Enum

An enumeration.

LOG = <module 'pm4py.algo.discovery.batches.variants.log' from 'C:\\Users\\berti\\FRAUNHOFER\\pm4py-core\\pm4py\\algo\\discovery\\batches\\variants\\log.py'>
PANDAS = <module 'pm4py.algo.discovery.batches.variants.pandas' from 'C:\\Users\\berti\\FRAUNHOFER\\pm4py-core\\pm4py\\algo\\discovery\\batches\\variants\\pandas.py'>
pm4py.algo.discovery.batches.algorithm.apply(log: Union[pm4py.objects.log.obj.EventLog, pandas.core.frame.DataFrame], parameters: Optional[Dict[str, Any]] = None) → List[Tuple[Tuple[str, str], int, Dict[str, Any]]][source]

Provided an event log / dataframe, returns a list having as elements the activity-resources with the batches that are detected, divided in: - Simultaneous (all the events in the batch have identical start and end timestamps) - Batching at start (all the events in the batch have identical start timestamp) - Batching at end (all the events in the batch have identical end timestamp) - Sequential batching (for all the consecutive events, the end of the first is equal to the start of the second) - Concurrent batching (for all the consecutive events that are not sequentially matched)

The approach has been described in the following paper: Martin, N., Swennen, M., Depaire, B., Jans, M., Caris, A., & Vanhoof, K. (2015, December). Batch Processing: Definition and Event Log Identification. In SIMPDA (pp. 137-140).

Parameters
• log – Event log / dataframe object

• parameters – Parameters of the algorithm: - ACTIVITY_KEY => the attribute that should be used as activity - RESOURCE_KEY => the attribute that should be used as resource - START_TIMESTAMP_KEY => the attribute that should be used as start timestamp - TIMESTAMP_KEY => the attribute that should be used as timestamp - CASE_ID_KEY => the attribute that should be used as case identifier - MERGE_DISTANCE => the maximum time distance between non-overlapping intervals in order for them to be

considered belonging to the same batch (default: 15*60 15 minutes)

• MIN_BATCH_SIZE => the minimum number of events for a batch to be considered (default: 2)

Returns

A (sorted) list containing tuples. Each tuple contain: - Index 0: the activity-resource for which at least one batch has been detected - Index 1: the number of batches for the given activity-resource - Index 2: a list containing all the batches. Each batch is described by:

# The start timestamp of the batch # The complete timestamp of the batch # The list of events that are executed in the batch

Return type

list_batches