The official codes for "PMC-LLaMA: Towards Building Open-source Language Models for Medicine".
We prove that medical LLM should be first pretrained with domain corpus, and then tuned with instructions following dataset.
We have released The latest model PMC_LLaMA_13B finetuned on our instructions the following dataset. It has shown a better ability to follow user instructions than MedLLaMA_13B.
Similarly, it can be easily loaded with:
import transformers
import torch
tokenizer = transformers.LlamaTokenizer.from_pretrained('axiong/PMC_LLaMA_13B')
model = transformers.LlamaForCausalLM.from_pretrained('axiong/PMC_LLaMA_13B')Hereby we present PMC_LLaMA's versions and briefs.
MedLLaMA_13B is pretrained on medical corpus, and PMC_LLaMA_13B is further finetuned based on that.
We have released a new report genration metrics RaTEScore. We strongly believe to promote the develop a generative-based medical foundation models, developing a robust and reliable metric is a critical and foundation step.
Simply set up the required environment as following:
conda install pytorch==1.13.0 torchvision==0.14.0 torchaudio==0.13.0 pytorch-cuda=11.6 -c pytorch -c nvidia
pip install transformers=4.28.1, sentencepiece, datasetsCheck simple_test.py for quickly use PMC-LLaMA or you can follow this folowing simple sample.
import transformers
import torch
tokenizer = transformers.LlamaTokenizer.from_pretrained('axiong/PMC_LLaMA_13B')
model = transformers.LlamaForCausalLM.from_pretrained('axiong/PMC_LLaMA_13B')
model.cuda() # move the model to GPU
prompt_input = (
'Below is an instruction that describes a task, paired with an input that provides further context.'
'Write a response that appropriately completes the request.\n\n'
'### Instruction:\n{instruction}\n\n### Input:\n{input}\n\n### Response:'
)
example = {
"instruction": "You're a doctor, kindly address the medical queries according to the patient's account. Answer with the best option directly.",
"input": (
"###Question: A 23-year-old pregnant woman at 22 weeks gestation presents with burning upon urination. "
"She states it started 1 day ago and has been worsening despite drinking more water and taking cranberry extract. "
"She otherwise feels well and is followed by a doctor for her pregnancy. "
"Her temperature is 97.7°F (36.5°C), blood pressure is 122/77 mmHg, pulse is 80/min, respirations are 19/min, and oxygen saturation is 98% on room air."
"Physical exam is notable for an absence of costovertebral angle tenderness and a gravid uterus. "
"Which of the following is the best treatment for this patient?"
"###Options: A. Ampicillin B. Ceftriaxone C. Doxycycline D. Nitrofurantoin"
)
}
input_str = [prompt_input.format_map(example)]
model_inputs = tokenizer(
input_str,
return_tensors='pt',
padding=True,
)
print( f"\033[32mmodel_inputs\033[0m: { model_inputs }" )
topk_output = model.generate(
model_inputs.input_ids.cuda(),
max_new_tokens=1000,
top_k=50
)
output_str = tokenizer.batch_decode(topk_output)
print('model predict: ', output_str[0])The training process can be divided as two phases: pretrain and instruction-tuning.
Pre-training
The script for pretraining locates at Pretrain/training.sh.
Our pretraining dataset sources from S2ORC. Only those papers with PubMed IDs are deemed as medical-related and used during pretraining.
The book is listed in this repo as MedicalBook.xlsx, due to licenses, we cannot release raw content. For reproducing, pls buy and process the books.
More details about how to fine-tune LLaMA can refer to Finetune_LLAMA
Instruction Tuning
We also provide instruction tuning script at SFT/train.py.
And you can find our instruction dataset at PMC LLaMA Instructions.
| Method | Model Size | USMLE | MedMCQA | PubMedQA |
|---|---|---|---|---|
| Human (pass) | - | 50.0 | -- | 60.0 |
| Human (expert) | - | 87.0 | 90.0 | 78.0 |
| ChatGPT | 175B | 57.0 | 44.7 | 63.9 |
| LLaMA-2 | 13B | 42.73 | 37.41 | 68.0 |
| LLaMA-2 | 70B | 43.68 | 35.02 | 74.3 |
| Med-Alpaca | 13B | 30.85 | 31.13 | 53.2 |
| Chat-Doctor | 7B | 33.93 | 31.10 | 54.3 |
PMC_LLaMA_13B  |
13B | 56.36 | 56.04 | 77.9 |
Note that, the manual and zero-shot results with * are referred from LMFLow.
We demonstrate PMC_LLaMA_13B's responses with out of domain queries.
Note that, due to train on the papers, MedLLaMA_13B may generate some citation numbers (LLaMA somtimes will do this as well) and we dismiss them in the cases to show the main contents. While for PMC_LLaMA_13B, it's much easier to extract the correct answer as the output result is structured.
Minimal LLaMA -- https://github.com/zphang/minimal-llama
alpaca -- https://github.com/tatsu-lab/stanford_alpaca
LLaMA: Open and Efficient Foundation Language Models -- https://arxiv.org/abs/2302.13971
If you have any question, please feel free to contact [email protected].