gena-datasets

Python client for browsing public datasets and submitting a dataset to Gena for review.

Install

python -m pip install gena-datasets

Read public datasets

from gena_datasets import GenaDatasetsClient

gena = GenaDatasetsClient()
for item in gena.list_datasets():
    print(item["id"], item["title"], item["license"])

dataset = gena.get_dataset("tulueng100-reference")
print(dataset["content"])

Public catalog access does not require sign-in. download_dataset(id) returns the original dataset file contents as text.

Submit a dataset

id_token = current_firebase_user.get_id_token()
submission = gena.submit_dataset(
    {
        "title": "Example collection",
        "description": "A short description of the data and how it was collected.",
        "language": "English",
        "license": "CC BY 4.0",
        "fileName": "examples.jsonl",
        "content": '{"text":"Example"}\n',
        "rightsConfirmed": True,
    },
    id_token=id_token,
)

print(submission["datasetId"], submission["status"])

Submissions are private while pending review. A designated Gena moderator must approve a submission before it becomes public. The contributor must have permission to share the data and choose a license that allows publication. Current uploads are limited to 900,000 bytes and .csv, .json, .jsonl, or .txt files.

API endpoints

• GET /api/gena/datasets
• GET /api/gena/datasets/{id}
• POST /api/gena/datasets with a signed-in account token

License

MIT