metadata
tags:
- autotrain
- conversational
- meta-llama
- meta-llama/Llama-2-7b-hf
inference: false
widget:
- text: >
instruction: "If you are a doctor, please answer the medical questions
based on the patient's description."
input: "Hi, I had a subarachnoid bleed and coiling of brain aneurysm last
year. I am having some major bilateral temple pain along with numbness
that comes and goes in my left arm/hand/fingers. I have had headaches
since the aneurysm, but this is different. Also, my moods have been
horrible for the past few weeks."
response: ''
library_name: peft
!huggingface-cli login
_| _| _| _| _|_|_| _|_|_| _|_|_| _| _| _|_|_| _|_|_|_| _|_| _|_|_| _|_|_|_|
_| _| _| _| _| _| _| _|_| _| _| _| _| _| _| _|
_|_|_|_| _| _| _| _|_| _| _|_| _| _| _| _| _| _|_| _|_|_| _|_|_|_| _| _|_|_|
_| _| _| _| _| _| _| _| _| _| _|_| _| _| _| _| _| _| _|
_| _| _|_| _|_|_| _|_|_| _|_|_| _| _| _|_|_| _| _| _| _|_|_| _|_|_|_|
To login, `huggingface_hub` requires a token generated from https://huggingface.co./settings/tokens .
Token: <your-hf-access-token>
from peft import PeftModel, PeftConfig
from transformers import AutoModelForCausalLM
from transformers import AutoTokenizer
import torch
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
config = PeftConfig.from_pretrained("Ashishkr/llama2_medical_consultation")
model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-2-7b-hf")
model = PeftModel.from_pretrained(model, "Ashishkr/llama2_medical_consultation").to(device)
tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-2-7b-hf")
def llama_generate(
model: AutoModelForCausalLM,
tokenizer: AutoTokenizer,
prompt: str,
max_new_tokens: int = 128,
temperature: float = 0.92):
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
inputs = tokenizer(
[prompt],
return_tensors="pt",
return_token_type_ids=False,
).to(
device
)
with torch.autocast("cuda", dtype=torch.bfloat16):
response = model.generate(
**inputs,
max_new_tokens=max_new_tokens,
temperature=temperature,
return_dict_in_generate=True,
eos_token_id=tokenizer.eos_token_id,
pad_token_id=tokenizer.pad_token_id,
)
decoded_output = tokenizer.decode(
response["sequences"][0],
skip_special_tokens=True,
)
return decoded_output[len(prompt) :]
prompt = """
instruction: "If you are a doctor, please answer the medical questions based on the patient's description." \n
input: "Hi, I had a subarachnoid bleed and coiling of brain aneurysm last year.
I am having some major bilateral temple pain along with numbness that comes and
goes in my left arm/hand/fingers. I have had headaches since the aneurysm,
but this is different. Also, my moods have been horrible for the past few weeks.\n
response: """
response = llama_generate(
model,
tokenizer,
prompt,
max_new_tokens=100,
temperature=0.92,
)
print(response)