Converting dataset metadata from OpenML to Croissant format.
1.0K
docker run --rm openml/croissant-converter generate_croissants.py --help
docker run --rm openml/croissant-converter upload_datasets_to_minio.py --help
Make sure you have a .env file to give to the uploader:
MINIO_ACCESS_KEY=[KEY]
MINIO_SECRET_KEY=[SECRET]
Example:
docker run --rm \
-v ./data:/output \
-v ~/.cache/openml:/home/unprivileged-user/.cache/openml \
openml/croissant-converter:[VERSION] generate_croissants.py \
--all \
-o /output
docker run --rm \
-v ./data:/input \
--env-file .env \
openml/croissant-converter:[VERSION] upload_datasets_to_minio.py \
-i /input
Update the version in pyproject.toml. Use the same version in the following script:
docker build -f docker/Dockerfile --tag openml/croissant-converter:[VERSION] .
docker push openml/croissant-converter:[VERSION]
Content type
Image
Digest
sha256:31dd71c30…
Size
377.6 MB
Last updated
over 1 year ago
docker pull openml/croissant-converter