This page summarises what you’ve learned about Kedro so far.
Kedro can be used on Windows, macOS or Linux
Installation prerequisites include Python 3.7+,
You should install Kedro using
pip install kedroor
conda install -c conda-forge kedro
Kedro nodes are the building blocks of pipelines. A node is a wrapper for a Python function that names the inputs and outputs of that function.
A pipeline organises the dependencies and execution order of a collection of nodes.
Kedro has a registry of all data sources the project can use called the Data Catalog. There is inbuilt support for various file types and file systems.
Kedro projects follow a default template that uses specific folders to store datasets, notebooks, configuration and source code.
You can create a Kedro project:
with just the basic code:
or you can populate a new project with pre-built code, e.g.
kedro new --starter=pandas-irisfrom a range of starter projects
Once you’ve created a project, you need to navigate to its project folder; you can then install its dependencies:
pip install -r src/requirements.txt
To run the project:
To visualise the project:
Next, you should work through the spaceflights tutorial for the hands-on experience. It illustrates how to build a working project, which includes creating nodes, registering pipelines, setting up the Data Catalog and adding dependencies.