Skip to content

Getting started#

Fondant provides a couple of example pipelines which can help you get started.

For demonstration purposes, we'll use the example pipeline to download and filter images from the fondant-cc-25m creative commons image dataset.

Clone the Fondant GitHub repository#

git clone https://github.com/ml6team/fondant-usecase-filter-creative-commons.git

Install the requirements#

Install the requirements.txt

pip install -r requirements.txt

And make sure that Docker Compose is installed.

Run the pipeline#

navigate into the src folder:

cd src

And execute the pipeline locally using the fondant cli:

fondant run local pipeline.py

IMPORTANT

For local testing purposes, the pipeline will only download the first 100 images.

Inspect the results#

Congrats, you just ran your first Fondant pipeline! To visually inspect the results between every pipeline step, you can use the fondant explorer:

fondant explore start --base_path ./data-dir

Building your own pipeline#

To learn how to build your own pipeline, you can: - Check out the pipeline.ipynb notebook in the example repository which runs through the steps to build the pipeline one by one. - Continue to the next guide on building your own pipeline