Datasets:

---

wlsdzyzl

/

MedPointS-cls

like 0

This PR enhances the dataset card for MedPointS-CLS to provide more comprehensive information and improve discoverability on the Hugging Face Hub.

Key changes include:

• Metadata Updates:
  • Added task_categories: ['graph-machine-learning'] to align with the dataset's nature and Hub guidelines.
  • Expanded tags to include completion and segmentation, reflecting all the tasks the dataset supports as mentioned in the paper's abstract.
• Structured Links:
  • Added a direct link to the paper: Hierarchical Feature Learning for Medical Point Clouds via State Space Model.
  • Included a link to the associated project page: https://flemme-docs.readthedocs.io/en/latest/medpoints.html.
  • Provided a direct link to the GitHub repository where the code for using the dataset is available: https://github.com/wlsdzyzl/flemme.
• Paper Abstract: Included the full abstract of the paper to give users a quick overview of the dataset's context and methodology.
• Sample Usage: Added a dedicated "Sample Usage" section with bash code snippets directly from the GitHub README, demonstrating how to train and evaluate models for classification, completion, and segmentation using the flemme framework.
• Citation: Preserved the existing BibTeX entry with the arXiv URL.
• The dataset_info block and its delimiters (---) have been preserved in their original position at the end of the Markdown content.