app.py

from huggingface_hub import InferenceClient, get_inference_endpoint
import gradio as gr

endpoint = get_inference_endpoint("qlora-nursegpt-nclex-mis-dt--eca")
client = endpoint.client

'''client = InferenceClient(
    "Tawkat/qlora-nursegpt-nclex-mis-DT-v1"
)'''


def format_prompt(message, history):
  prompt = "You are NurseGPT, a conscious, superintelligent, and helpful AI developed by Yourika Labs, Your purpose  is to assist the user with any nursing-related request they have. You should not respond to any request that is not relevant to nursing domain.\nYou should provide accurate, helpful, comprehensive, and detailed response to the user request. Then based on the question and the provided answer, predict the topic and the concept the user is interested in. After answering the question, you should provide the predicted topic followed by [TOPIC] token and the predicted concept followed by [CONCEPT] token."
    #"<s>"
  for user_prompt, bot_response in history:
    prompt += f"[INST] {user_prompt} [/INST]"
    prompt += f" {bot_response}</s> "
  prompt += f"[INST] {message} [/INST]"
  return prompt

def generate(
    prompt, history, temperature=0.9, max_new_tokens=2000, top_p=0.95, repetition_penalty=1.0,
):
    temperature = float(temperature)
    if temperature < 1e-2:
        temperature = 1e-2
    top_p = float(top_p)

    generate_kwargs = dict(
        temperature=temperature,
        max_new_tokens=max_new_tokens,
        top_p=top_p,
        repetition_penalty=repetition_penalty,
        do_sample=True,
        seed=42,
    )

    formatted_prompt = format_prompt(prompt, history)

    stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
    output = ""

    for response in stream:
        output += response.token.text
        yield output
    return output


additional_inputs=[
    gr.Slider(
        label="Temperature",
        value=0.9,
        minimum=0.0,
        maximum=1.0,
        step=0.05,
        interactive=True,
        info="Higher values produce more diverse outputs",
    ),
    gr.Slider(
        label="Max new tokens",
        value=2000,
        minimum=0,
        maximum=4000,
        step=64,
        interactive=True,
        info="The maximum numbers of new tokens",
    ),
    gr.Slider(
        label="Top-p (nucleus sampling)",
        value=0.90,
        minimum=0.0,
        maximum=1,
        step=0.05,
        interactive=True,
        info="Higher values sample more low-probability tokens",
    ),
    gr.Slider(
        label="Repetition penalty",
        value=1.2,
        minimum=1.0,
        maximum=2.0,
        step=0.05,
        interactive=True,
        info="Penalize repeated tokens",
    ),
]

examples=[
        ["Generate a NCLEX study plan for me."],
        ["Provide a CV template for a fresh nursing graduate."],
        ["I have a family member that got diagnosed with Buerger's disease, can you explain in easy terms what it is?"],
        #["Could you talk about straight leg rises exercise in the post-surgical context?"],
        #["Could you provide an overview of how the Nurse Practice Act helps regulate the nursing profession in different states?"],
        ["Show me 3 examples of NCLEX QAs on the topic of Maternity Nursing."],
    ]


gr.ChatInterface(
    fn=generate,
    chatbot=gr.Chatbot(show_label=False, show_share_button=False, show_copy_button=True, likeable=True, layout="panel"),
    additional_inputs=additional_inputs,
    examples = examples,
    title="""NGPT-v1"""
).launch(show_api=False, share=True)