FlowerPower 🌸 — Build & Orchestrate Data Pipelines
A simple, configuration-driven workflow framework built on Hamilton
FlowerPower is a Python framework for building, configuring, and running data processing pipelines. You write your logic as plain Python functions using Hamilton; FlowerPower takes care of the project structure, layered configuration, execution, the CLI, a web UI, and pluggable I/O.
Hamilton = the DAG, FlowerPower = everything around it. You describe nodes as functions; FlowerPower wires up config, runs, retries, visualization, and orchestration.
✨ Features
- Modular pipelines — Define transforms as Python functions; Hamilton assembles them into a directed acyclic graph (DAG) from their signatures.
- Configuration-driven — Separate logic from settings with layered YAML, env
overlays (
FP_PIPELINE__*),${VAR}interpolation, and runtime kwargs. - Unified interfaces — Drive everything from the Python API
(
FlowerPowerProject), the CLI (flowerpower), or the web UI (Hamilton UI). - Executors — Run locally, in a thread pool, or distributed with Ray.
- Adapters — Track lineage with the Hamilton Tracker, experiments with MLflow.
- Extensible I/O — CSV, JSON, Parquet, DeltaTable, DuckDB, PostgreSQL, MySQL,
MSSQL, Oracle, SQLite, MQTT via the
flowerpower-ioplugin. - Filesystem abstraction — Local, S3, GCS, and more, through fsspeckit.
📦 Installation
FlowerPower requires Python 3.11+. We recommend uv:
uv venv && source .venv/bin/activate
uv pip install flowerpower
Optional extras:
uv pip install 'flowerpower[io]' # I/O plugins
uv pip install 'flowerpower[ui]' # Hamilton UI
uv pip install 'flowerpower[ray]' # distributed execution
uv pip install 'flowerpower[io,ui,ray]' # combine the extras you need
🚀 Quick start
Create a project, a pipeline, and run it:
flowerpower init --name hello-flowerpower
cd hello-flowerpower
flowerpower pipeline new hello
flowerpower pipeline run hello
Or in Python:
from flowerpower import FlowerPowerProject
# 1. create the project
FlowerPowerProject.new(name="hello-flowerpower")
# 2. load it and create a pipeline
project = FlowerPowerProject.load("hello-flowerpower")
project.pipeline_manager.creator.create_pipeline(name="hello")
# 3. (edit pipelines/hello.py + conf/pipelines/hello.yml, then run)
result = project.run("hello")
print(result)
A pipeline module
Write your DAG as functions in pipelines/hello.py. Parameters come from YAML
and are wired in with Hamilton's @parameterize:
from pathlib import Path
from hamilton.function_modifiers import parameterize
from flowerpower.cfg import Config
PARAMS = Config.load(
Path(__file__).parents[1], pipeline_name="hello"
).pipeline.h_params
@parameterize(**PARAMS["greeting_message"]) # PARAMS is a dict — use ["..."]
def greeting_message(message: str) -> str:
return f"{message},"
@parameterize(**PARAMS["target_name"])
def target_name(name: str) -> str:
return f"{name}!"
def full_greeting(greeting_message: str, target_name: str) -> str:
return f"{greeting_message} {target_name}"
# conf/pipelines/hello.yml
params:
greeting_message:
message: "Hello"
target_name:
name: "World"
run:
final_vars: [full_greeting]
project.run("hello") → {'full_greeting': 'Hello, World!'}.
📖 The full, step-by-step walkthrough is in the Tutorial.
⚙️ Running pipelines
FlowerPowerProject.run() (and PipelineManager.run()) accept a RunConfig and/or
keyword overrides — kwargs always win:
from flowerpower.cfg.pipeline.run import RunConfig
# simple
result = project.run("hello")
# with a RunConfig
result = project.run("hello", run_config=RunConfig(log_level="DEBUG"))
# with kwargs overrides
result = project.run(
"hello",
inputs={"greeting_message": {"message": "Hi"}},
final_vars=["full_greeting"],
)
Async runs use Hamilton's async driver via PipelineManager.run_async():
from flowerpower.pipeline import PipelineManager
pm = PipelineManager(base_dir="hello-flowerpower")
result = await pm.run_async("hello")
See the docs for RunConfig & builders, additional modules, and adapters.
🖥️ Web UI
Launch the Hamilton UI to visualize and inspect pipeline runs:
flowerpower ui
📖 Documentation
Full documentation — installation, tutorial, how-to guides, concepts, CLI, and API reference — lives at https://legout.github.io/flowerpower/.
🤝 Contributing
Contributions are welcome. See the contributing guide and open issues or PRs against legout/flowerpower.
📜 License
MIT — see LICENSE.
1 maintainer
Metadata
1 maintainer
Release files for FlowerPower 0.37.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| flowerpower-0.37.0.tar.gz | 91.0 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| flowerpower-0.37.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 199.6 kB
Release files / flowerpower-0.37.0.tar.gz
| Download URL | flowerpower-0.37.0.tar.gz |
|---|---|
| Size | 91.0 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
37d0f121c703d891cbe0b6a6222540b8417e94d9c59d7299ed6a97d5eb0c3dfa
|
|
BLAKE2b-256 checksum How to use checksums |
913f790f0b44d26be78422889c4a909d5c0a81d6aca3fddc996ee07164cfca5d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.11.28 {"installer":{"name":"uv","version":"0.11.28","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}
|
Release files / flowerpower-0.37.0-py3-none-any.whl
| Download URL | flowerpower-0.37.0-py3-none-any.whl |
|---|---|
| Size | 108.6 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
2ccf5b57730d3cc5d8c2a4290c8a528406df8823c55c05d4c482bdbe2108746f
|
|
BLAKE2b-256 checksum How to use checksums |
ea6e6294ffbfd88fc0644d7d74865eea7cf474b478b6ba80adb2677c4caa5842
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.11.28 {"installer":{"name":"uv","version":"0.11.28","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}
|
