91 lines
4.0 KiB
Markdown
91 lines
4.0 KiB
Markdown
---
|
|
license: cc
|
|
language:
|
|
- en
|
|
library_name: transformers
|
|
pipeline_tag: text-generation
|
|
tags:
|
|
- medical
|
|
---
|
|
# MedAlpaca 7b
|
|
|
|
|
|
## Table of Contents
|
|
|
|
[Model Description](#model-description)
|
|
- [Architecture](#architecture)
|
|
- [Training Data](#trainig-data)
|
|
[Model Usage](#model-usage)
|
|
[Limitations](#limitations)
|
|
|
|
## Model Description
|
|
### Architecture
|
|
`medalpaca-7b` is a large language model specifically fine-tuned for medical domain tasks.
|
|
It is based on LLaMA (Large Language Model Meta AI) and contains 7 billion parameters.
|
|
The primary goal of this model is to improve question-answering and medical dialogue tasks.
|
|
Architecture
|
|
|
|
|
|
### Training Data
|
|
The training data for this project was sourced from various resources.
|
|
Firstly, we used Anki flashcards to automatically generate questions,
|
|
from the front of the cards and anwers from the back of the card.
|
|
Secondly, we generated medical question-answer pairs from [Wikidoc](https://www.wikidoc.org/index.php/Main_Page).
|
|
We extracted paragraphs with relevant headings, and used Chat-GPT 3.5
|
|
to generate questions from the headings and using the corresponding paragraphs
|
|
as answers. This dataset is still under development and we believe
|
|
that approximately 70% of these question answer pairs are factual correct.
|
|
Thirdly, we used StackExchange to extract question-answer pairs, taking the
|
|
top-rated question from five categories: Academia, Bioinformatics, Biology,
|
|
Fitness, and Health. Additionally, we used a dataset from [ChatDoctor](https://arxiv.org/abs/2303.14070)
|
|
consisting of 200,000 question-answer pairs, available at https://github.com/Kent0n-Li/ChatDoctor.
|
|
|
|
| Source | n items |
|
|
|------------------------------|--------|
|
|
| ChatDoc large | 200000 |
|
|
| wikidoc | 67704 |
|
|
| Stackexchange academia | 40865 |
|
|
| Anki flashcards | 33955 |
|
|
| Stackexchange biology | 27887 |
|
|
| Stackexchange fitness | 9833 |
|
|
| Stackexchange health | 7721 |
|
|
| Wikidoc patient information | 5942 |
|
|
| Stackexchange bioinformatics | 5407 |
|
|
|
|
## Model Usage
|
|
To evaluate the performance of the model on a specific dataset, you can use the Hugging Face Transformers library's built-in evaluation scripts. Please refer to the evaluation guide for more information.
|
|
Inference
|
|
|
|
You can use the model for inference tasks like question-answering and medical dialogues using the Hugging Face Transformers library. Here's an example of how to use the model for a question-answering task:
|
|
|
|
```python
|
|
|
|
from transformers import pipeline
|
|
|
|
pl = pipeline("text-generation", model="medalpaca/medalpaca-7b", tokenizer="medalpaca/medalpaca-7b")
|
|
question = "What are the symptoms of diabetes?"
|
|
context = "Diabetes is a metabolic disease that causes high blood sugar. The symptoms include increased thirst, frequent urination, and unexplained weight loss."
|
|
answer = pl(f"Context: {context}\n\nQuestion: {question}\n\nAnswer: ")
|
|
print(answer)
|
|
```
|
|
|
|
## Limitations
|
|
The model may not perform effectively outside the scope of the medical domain.
|
|
The training data primarily targets the knowledge level of medical students,
|
|
which may result in limitations when addressing the needs of board-certified physicians.
|
|
The model has not been tested in real-world applications, so its efficacy and accuracy are currently unknown.
|
|
It should never be used as a substitute for a doctor's opinion and must be treated as a research tool only.
|
|
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
|
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_medalpaca__medalpaca-7b)
|
|
|
|
| Metric | Value |
|
|
|-----------------------|---------------------------|
|
|
| Avg. | 44.98 |
|
|
| ARC (25-shot) | 54.1 |
|
|
| HellaSwag (10-shot) | 80.42 |
|
|
| MMLU (5-shot) | 41.47 |
|
|
| TruthfulQA (0-shot) | 40.46 |
|
|
| Winogrande (5-shot) | 71.19 |
|
|
| GSM8K (5-shot) | 3.03 |
|
|
| DROP (3-shot) | 24.21 |
|