All datasets / nicoboss

nicoboss/medra-medical-sampled

48 downloads in 30 days0 likesLast modified 12 May 2025View on Hugging Face

Unclear

The checks could not identify a recognised licence ID for at least one declared source dataset.

Declared licence: apache-2.0, from the tagEvidence: README.md at 0139fe5
Licence files: No licence file at the top level of the repositoryEvidence: file list at 0139fe5
Declared sources: 2
Access gated on Hugging Face: No

Each cell links to what was checked: a file, a file list or a Hub record.

Sources listed in the card metadata

bigbionot foundUnclearEvidence: Hub page
othernot foundUnclearEvidence: Hub page

Tasks

Citation text from the card

If you use this dataset in your research or projects, please cite the original source datasets as well as this combined dataset. A suggested citation format for this dataset would be:

```bibtex
@data{MedraMedicalReasoningDataset,
  author = {drwlf and nicoboss},
  title = {{Medra Medical Reasoning Dataset: A Curated Corpus for Training LLMs with Explicit Medical Reasoning}},
  year = {2025}, % Fill in the year of release
  publisher = {Hugging Face},
  url = {[https://huggingface.co/datasets/](https://huggingface.co/datasets/)}, 
}

Please also refer to the individual dataset cards of the source datasets for their specific citation requirements.
Limitations and Bias
 * Source Data Limitations: The quality and coverage of the dataset are directly dependent on the quality and scope of its source datasets.
 * Reasoning Quality: The reasoning included in the <think> tags is either extracted from source data or formatted from it. Its accuracy and completeness may vary and are not guaranteed to be perfect medical reasoning.
 * Token Limit: The 2048 token limit per example means that very long dialogues or detailed reasoning processes might be truncated.
 * Bias: As mentioned in the Responsible Use section, biases from source datasets are likely present.

All datasets marked Unclear | JSON

The same facts for your agent

A free MCP server at mcp.datalicense.uk/mcp. Connect it from an MCP client such as Claude or Cursor.

dataset_licence({ "id": "nicoboss/medra-medical-sampled" })
signal: "unclear"
stated_licence: "apache-2.0"
sources: 2