Real-life data for object-centric processing mining
View the project on GitHub LienBosmans/pystackt
Watch the demo on Youtube PyStack't Demo BPM 2025
from pystackt import *
create_statistics_views(
quack_db="./stackt.duckdb",
schema_in="main",
schema_out="statistics"
)
| Parameter | Type | Description |
|---|---|---|
quack_db |
str |
Path to the DuckDB database file containing the input data. |
schema_in |
str |
Name of the schema in the DuckDB file that contains the input data. Needs to be stored using in Stack’t relational schema. |
schema_out |
str |
Name of the schema where the output SQL views will be created. If schema already exists, it will be cleared first. |
quack_db, schema_in)The input data must be stored in a DuckDB database file using the Stack’t relational schema. quack_db defines the path to this file, and schema_in specifies the schema containing the OCED data.
schema_out)Two SQL views are created in the schema defined by schema_out. They contain some simple statistics that provide an excellent starting point to get to know the data.
event_stats: provides summary statistics for each event type, including:
first_event_timestamp / last_event_timestamp: timestamp of the earliest/latest event recorded for this typeevent_count: total number of events of this typeevent_attribute_count: number of distinct attributes defined for this event typeevent_attribute_value_update_count: number of times a value is recorded for these attributes across all eventsobject_stats: provides summary statistics for each object type, including:
first_object_update_timestamp / last_object_update_timestamp: timestamps of the earliest/latest attribute value update recorded for objects of this typeobject_count: number of distinct objects of this typeobject_attribute_count: number of distinct attributes defined for this object typeobject_attribute_value_update_count: number of attribute value updates recorded for objects of this typeIf schema_out already exists, it will be cleared before the new views are created.
The article How to view DuckDB files explains how you can view the result.