Add openarm-dataset-upload#53
Conversation
| parser.add_argument( | ||
| "--licence", | ||
| default="apache-2.0", | ||
| help="The licence to associate with the dataset on the Hugging Face Hub. " | ||
| "Defaults to Apache-2.0.", | ||
| ) |
There was a problem hiding this comment.
Ah, let's add license: in metadata.yaml and Metadata.license.
Can we rename this to --default-license that is used when metadata misses license information?
We can work on this as a follow-up task.
| {{ dataset_description | default("", true) }} | ||
|
|
||
| - **Homepage:** {{ url | default("[More Information Needed]", true)}} | ||
| - **Paper:** {{ paper | default("[More Information Needed]", true)}} |
There was a problem hiding this comment.
Let's add these information to metadata.yaml as a follow-up task.
Co-authored-by: Sutou Kouhei <kou@cozmixng.org>
| ) | ||
| parser.add_argument( | ||
| "--licence", | ||
| default="apache-2.0", |
There was a problem hiding this comment.
In general, we should use SPDX ID ( https://spdx.org/licenses/ ) for license ID:
| default="apache-2.0", | |
| default="Apache-2.0", |
There was a problem hiding this comment.
huggingfaced api only accepts a small letter.https://huggingface.co/docs/hub/repositories-licenses
There was a problem hiding this comment.
Oh... Hugging Face doesn't use SPDX ID...
Then we should change the "Defaults to Apache-2.0." text ("Apache-2.0" -> "apache-2.0"). Anyway, let's work on license related thing as a follow-up task.
| parser.add_argument( | ||
| "--repo-id", | ||
| required=True, | ||
| help="Target Hugging Face dataset repository id, e.g. username/dataset-name", | ||
| ) |
There was a problem hiding this comment.
It's better that we use our recommended naming convention (we should describe it...) as the default. Let's work on it as a follow-up task.
3 participants
Fix GH-51
Summary
Adds an
openarm-dataset-uploadCLI that publishes an OpenArm dataset directory to the Hugging Face Hub, creating the dataset repo if needed, generating a dataset card, and tagging the upload with the dataset version.sample: https://huggingface.co/datasets/k1000dai/fixture