PM4Py — Process Mining for Python Tutorials
View more detailed examples in the PM4Py repository on GitHub
View Examples on GitHubJump to category
New to PM4Py?
Event Data
Import, convert and filter event logs, also object-centric ones (OCEL).
Process Modeling & Discovery
Discover models from logs and work with DFGs, Petri nets, process trees and BPMN.
Process Analysis
Statistics, resource networks, conformance checking, model quality and simulation.
- Throughput Time
- Case Arrival/Dispersion Ratio
- Performance Spectrum
- Cycle Time and Waiting Time
- Activity Service Time (Sojourn Time)
- Concurrent Activities
- Eventually Follows Graph
- Displaying Graphs
- Dotted Chart
- Events Distribution
- Detection of Batches
- Rework (Activities)
- Rework (Cases)
- Query Structure - Paths over Time
- Process Cube
- Partially Ordered Variants
Machine Learning
Turn logs into features, then predict, detect anomalies and mine decisions.
- Automatic Feature Selection
- Manual Feature Selection
- Calculating Useful Features
- Event-based Feature Extraction
- One-hot Trace Encoding
- Count2Vec Trace Encoding
- N-gram Trace Encoding
- TF-IDF Trace Encoding
- Doc2Vec Trace Encoding
- Word2Vec Trace Encoding
- BERT Trace Encoding
- Token Replay Trace Encoding
- Alignment Trace Encoding
- Feature Extraction on DataFrames
- Temporal Feature Extraction
Miscellaneous
Process mining on event streams and the configuration of PM4Py.
- Streaming Package General Structure
- Streaming Process Discovery (Directly-Follows Graph)
- Streaming Conformance Checking (TBR)
- Streaming Conformance Checking (Footprints)
- Streaming Conformance Checking (Temporal Profile)
- Streaming Importer (XES trace-by-trace)
- Streaming Importer (XES event-by-event)
- Streaming Importer (CSV event-by-event)
- OCEL Streaming
