app.py

import gradio as gr
import os
from dotenv import load_dotenv
from text_generation import Client

load_dotenv()

API_TOKEN = os.environ.get("API_TOKEN", None)
MODEL_URL = os.environ.get("MODEL_URL", None)

def evaluate(hotel_request: str, **kwargs):
    hf_client = Client(
        MODEL_URL,
        headers={"Authorization": f"Bearer {API_TOKEN}"},
    )

    stream = hf_client.generate_stream(
        hotel_request,
        **kwargs,
    )

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

    return output

gr.Interface(
        fn=evaluate,
        inputs=[
        #     gr.components.Textbox(
        #         lines=2,
        #         label="Instruction",
        #         placeholder="Tell me about alpacas.",
        #     ),
            gr.components.Textbox(lines=2, label="Input", placeholder="Request for the Hotel"),
        #     gr.components.Slider(
        #         minimum=0, maximum=1, value=0.1, label="Temperature"
        #     ),
        #     gr.components.Slider(
        #         minimum=0, maximum=1, value=0.75, label="Top p"
        #     ),
        #     gr.components.Slider(
        #         minimum=0, maximum=100, step=1, value=40, label="Top k"
        #     ),
        #     gr.components.Slider(
        #         minimum=1, maximum=4, step=1, value=4, label="Beams"
        #     ),
        #     gr.components.Slider(
        #         minimum=1, maximum=2000, step=1, value=128, label="Max tokens"
        #     ),
        #     gr.components.Checkbox(label="Stream output"),
        ],
        outputs=[
            gr.inputs.Textbox(
                lines=1,
                label="Guest Name",
            ),
            gr.inputs.Textbox(
                lines=1,
                label="Hotel",
            ),
            gr.inputs.Textbox(
                lines=1,
                label="Location",
            ),
            gr.inputs.Textbox(
                lines=1,
                label="Date",
            )
        ],
        allow_flagging="never",
        title="Falcon-LoRA",
        description="Falcon-LoRA is a 1B-parameter LLM finetuned to follow instructions. It is trained on the [Hotel Requests](https://huggingface.co/datasets/MichaelAI23/hotel_requests) dataset.",  # noqa: E501
    ).queue().launch() #server_name="0.0.0.0", server_port=8080)