MatteoFasulo

/

mdeberta-v3-base-subjectivity-multilingual

like 0

This PR significantly enhances the model card for mdeberta-v3-base-subjectivity-multilingual by:

• Adding pipeline_tag: text-classification to enable better discoverability on the Hugging Face Hub (e.g., via https://huggingface.co/models?pipeline_tag=text-classification).
• Updating the license to cc-by-4.0, as specified in the associated GitHub repository.
• Refining tags to include deberta-v3, subjectivity-detection, multilingual, and sentiment-analysis for more accurate categorization.
• Adding specific language tags for all languages the model was trained/evaluated on (ar, de, en, it, bg, el, pl, ro, uk).
• Adding arxiv_id and code_url to the metadata for direct, machine-readable links to the paper and codebase.
• Adding datasets to specify the source of training data.
• Populating the "Model description", "Intended uses & limitations", and "Training and evaluation data" sections with comprehensive details extracted from the paper abstract and the GitHub README.
• Providing a clear "How to use" example utilizing the transformers pipeline for easy inference.
• Adding a dedicated "GitHub Repository" section for easy access to the code.
• Including a BibTeX entry for proper citation.

These updates ensure the model card is more informative, discoverable, and adheres to best practices for documentation on the Hub.