Examples

The FlowerPower repository includes ready-to-run example projects that demonstrate common pipeline patterns. You can find them in the examples/ directory on GitHub.

hello-world/

The canonical starter project. It defines a minimal hello_world pipeline in pipelines/hello_world.py and pairs it with a setup.py companion module. The companion module is loaded as an additional_module, so it shows how to split a pipeline into a main DAG and shared helper code.

data-etl-pipeline/

Demonstrates a configuration-driven ETL workflow: loading a raw CSV, validating the data, cleaning it, and producing a summary report. Use it to see how params and run.config can change pipeline behavior without editing code.

ml-training-pipeline/

Shows an end-to-end machine-learning workflow covering data preprocessing, feature engineering, model training, and evaluation. It illustrates how a Hamilton DAG maps cleanly onto ML stages and how model artifacts can be saved from a pipeline run.

pipeline-only-example/

A lightweight project that uses FlowerPower's core pipeline features with no optional extras. It is useful when you want a small, self-contained DAG that runs synchronously without additional infrastructure.

web-scraping-pipeline/

Demonstrates concurrent web scraping and content processing as a FlowerPower pipeline. It covers parallel HTTP requests, rate-limiting configuration, and structured content extraction.