huggingface-datasets

Explore Hugging Face dataset metadata, splits, and row previews via the Dataset Viewer API.

Updated Dec 17, 2024
One-click install
npx skills add https://github.com/LAF-US/IDAHO-VAULT --skill huggingface-datasets
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: huggingface-datasets
Source: https://github.com/LAF-US/IDAHO-VAULT/tree/main/.codex/plugins/cache/openai-curated/hugging-face/314574a046f21938025ae443f9c6dbbd0c2c9b7a/skills/datasets
Command: npx skills add https://github.com/LAF-US/IDAHO-VAULT --skill huggingface-datasets

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables read-only exploration of Hugging Face datasets through the Dataset Viewer API, enabling quick discovery of dataset configurations, metadata, and available subsets.

Core Features & Use Cases

  • Subset discovery: Resolve available splits and configurations for a dataset and preview structure.
  • Paginated previews: Retrieve first rows and page through data with safe limits.
  • Search & filter: Perform text matches and predicates across dataset content, then narrow results.
  • Parquet access: List and access parquet shard URLs and associated metadata for scalable analytics.

Quick Start

Query a dataset's metadata and preview its first rows using the Dataset Viewer API for a selected dataset.

Frequently Asked Questions about huggingface-datasets

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I preview rows from a Hugging Face dataset without downloading it?

You can preview Hugging Face dataset rows by querying the Dataset Viewer API to retrieve first-rows and paginate through data with safe limits. This enables quick read-only exploration of splits and configurations without downloading the entire dataset.

How can I search and filter text within Hugging Face datasets?

To search and filter Hugging Face datasets, the Dataset Viewer API provides endpoints for text matches and predicates. You can perform searches across dataset content and narrow results based on specific filter conditions.

Can I access Parquet file URLs for Hugging Face datasets for scalable analytics?

Yes, you can access Parquet resources by listing and retrieving Parquet shard URLs and associated metadata through the Dataset Viewer API. This provides access to Parquet files for scalable analytics workflows.

How do I find the available splits and configurations of a Hugging Face dataset?

To find available splits and configurations, you can resolve dataset subsets through the Dataset Viewer API. This enables discovery of dataset metadata, structures, and available subsets across namespaces.

What are the limitations of using the Dataset Viewer API for Hugging Face datasets?

The primary limitation is that the Dataset Viewer API only supports read-only exploration of Hugging Face datasets. You cannot modify, upload, or write data, as interactions are strictly limited to safe, URL-encoded queries for metadata, previews, and Parquet access.