Spaces:
Running
on
Zero
Running
on
Zero
First commit
Browse files- app.py +69 -0
- requirements.txt +8 -0
app.py
ADDED
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
|
3 |
+
import torch
|
4 |
+
model_name = "ruslanmv/Medical-Llama3-8B"
|
5 |
+
device_map = 'auto'
|
6 |
+
# Check if GPU is available
|
7 |
+
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
8 |
+
print(f"Using device: {device}")
|
9 |
+
|
10 |
+
if device.type == "cuda":
|
11 |
+
bnb_config = BitsAndBytesConfig(
|
12 |
+
load_in_4bit=True,
|
13 |
+
bnb_4bit_quant_type="nf4",
|
14 |
+
bnb_4bit_compute_dtype=torch.float16,
|
15 |
+
)
|
16 |
+
model = AutoModelForCausalLM.from_pretrained(
|
17 |
+
model_name,
|
18 |
+
quantization_config=bnb_config,
|
19 |
+
trust_remote_code=True,
|
20 |
+
use_cache=False,
|
21 |
+
device_map=device_map
|
22 |
+
)
|
23 |
+
else:
|
24 |
+
model = AutoModelForCausalLM.from_pretrained(
|
25 |
+
model_name,
|
26 |
+
trust_remote_code=True,
|
27 |
+
use_cache=False
|
28 |
+
)
|
29 |
+
tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
|
30 |
+
tokenizer.pad_token = tokenizer.eos_token
|
31 |
+
|
32 |
+
def askme(symptoms, question):
|
33 |
+
sys_message = '''
|
34 |
+
You are an AI Medical Assistant trained on a vast dataset of health information. Please be thorough and
|
35 |
+
provide an informative answer. If you don't know the answer to a specific medical inquiry, advise seeking professional help.
|
36 |
+
'''
|
37 |
+
content = symptoms + " " + question
|
38 |
+
messages = [{"role": "system", "content": sys_message}, {"role": "user", "content": content}]
|
39 |
+
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
|
40 |
+
inputs = tokenizer(prompt, return_tensors="pt").to(device)
|
41 |
+
outputs = model.generate(**inputs, max_new_tokens=200, use_cache=True)
|
42 |
+
response_text = tokenizer.batch_decode(outputs)[0].strip()
|
43 |
+
answer = response_text.split('<|im_start|>assistant')[-1].strip()
|
44 |
+
return answer
|
45 |
+
|
46 |
+
# Example usage
|
47 |
+
symptoms = '''
|
48 |
+
I'm a 35-year-old male and for the past few months, I've been experiencing fatigue,
|
49 |
+
increased sensitivity to cold, and dry, itchy skin.
|
50 |
+
'''
|
51 |
+
question = '''
|
52 |
+
Could these symptoms be related to hypothyroidism?
|
53 |
+
If so, what steps should I take to get a proper diagnosis and discuss treatment options?
|
54 |
+
'''
|
55 |
+
|
56 |
+
examples = [
|
57 |
+
[symptoms, question]
|
58 |
+
]
|
59 |
+
|
60 |
+
iface = gr.Interface(
|
61 |
+
fn=askme,
|
62 |
+
inputs=["text", "text"],
|
63 |
+
outputs="text",
|
64 |
+
examples=examples,
|
65 |
+
title="Medical AI Chatbot",
|
66 |
+
description="Ask me a medical question!"
|
67 |
+
)
|
68 |
+
|
69 |
+
iface.launch()
|
requirements.txt
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
torch==2.2.1
|
2 |
+
torchvision
|
3 |
+
torchaudio
|
4 |
+
xformers
|
5 |
+
bitsandbytes
|
6 |
+
accelerate
|
7 |
+
gradio
|
8 |
+
transformers
|