README.md

---
language:
- en
library_name: peft
pipeline_tag: text-generation
tags:
- medical
license: cc-by-nc-3.0
---

# MedFalcon 40b LoRA


## Model Description

### Architecture
`nmitchko/medfalcon-40b-lora` is a large language model LoRa specifically fine-tuned for medical domain tasks.
It is based on [`Falcon-40b-instruct`](https://huggingface.co/tiiuae/falcon-40b-instruct/) at 40 billion parameters.

The primary goal of this model is to improve question-answering and medical dialogue tasks.
It was trained using [LoRA](https://arxiv.org/abs/2106.09685), specifically [QLora](https://github.com/artidoro/qlora), to reduce memory footprint. 

> This Lora supports 4-bit and 8-bit modes.

### Requirements

```
bitsandbytes>=0.39.0
peft
transformers
```

Steps to load this model:
1. Load base model using QLORA
2. Apply LoRA using peft

```python
# 
from transformers import AutoTokenizer, AutoModelForCausalLM
import transformers
import torch

model = "tiiuae/falcon-40b-instruct"
LoRA = "nmitchko/medfalcon-40b-lora"

tokenizer = AutoTokenizer.from_pretrained(model)

model = AutoModelForCausalLM.from_pretrained(model,
    load_in_8bit=load_8bit,
    torch_dtype=torch.float16,
    trust_remote_code=True,
)

model = PeftModel.from_pretrained(model, LoRA)

pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
    torch_dtype=torch.bfloat16,
    trust_remote_code=True,
    device_map="auto",
)

sequences = pipeline(
   "What does the drug ceftrioxone do?\nDoctor:",
    max_length=200,
    do_sample=True,
    top_k=40,
    num_return_sequences=1,
    eos_token_id=tokenizer.eos_token_id,
)

for seq in sequences:
    print(f"Result: {seq['generated_text']}")
```