Discovery

discover_erg runs the complete mining pipeline and returns an ERG object.

import ergminer

log = ergminer.read_csv("data/event_log.csv")

erg = ergminer.discover_erg(
    log,
    case_id_col="case_id",
    activity_col="activity_name",
    timestamp_col="timestamp",
    resource_col="resource_id",
    verbose=True,
)

erg.print_summary()

Tuning delay analysis

Pass a DelayConfig to control how immediate vs delayed arcs are classified and how distributions are fitted:

config = ergminer.DelayConfig(
    immediate_threshold=0.01,
    min_sample_size=10,
    min_p_value=0.05,
)

erg = ergminer.discover_erg(log, delay_config=config)

See ergminer.discovery and DelayConfig for full details.