Dataset · medalpaca
MedQA medical question-answering dataset
medalpaca/medical_meadow_medqa
Medical Q&A from exams in English and Chinese for QA training and evaluation.
What's inside
The dataset contains medical question-answer pairs and, according to the card, textbooks; questions are sourced from exams in the United States and mainland China. Languages are English and Chinese, with a size ranging from 10k to 100k examples. This includes data and baseline code from the paper by Jin et al. on MedQA.
What you can do with it
- Fine-tune a medical QA model
- Train LoRA for exam question answering
- Evaluate LLMs on MedQA
- Tasks
- Question answering
- Languages
- English, Chinese
- Size
- 10K–100K rows
- Downloads this month
- 2.5K
- Likes
- 118
Models trained on it
- MohamedAhmedAE/Llama-3.1-8B-Instruct-Medical-Finetuned-mergedText generation 1.5K
- MohamedAhmedAE/Llama-3.2-3B-Instruct-Medical-Finetuned-mergedText generation 821
- mradermacher/Llama-3.2-3B-Instruct-Medical-Finetuned-merged-GGUFQuestion answering 374
- MohamedAhmedAE/Llama-3.2-1B-Instruct-Medical-Finetuned-mergedText generation 340
- NewSonnet/triage-qwen2.5-1.5b-ggufText generation 78
- llSourcell/medllama2_7bText generation 56
- garcianacho/MedLlama-2-7B-GGUF 33
- ae-aydin/Llama-3-8B-Instruct-Medical-QLoRA 33
How to load it
from datasets import load_dataset
ds = load_dataset("medalpaca/medical_meadow_medqa")