app.py

import torch
from peft import PeftModel, PeftConfig
from transformers import AutoModelForCausalLM, AutoTokenizer

HUGGING_FACE_USER_NAME = "pfh1976"
peft_model_id = f"{HUGGING_FACE_USER_NAME}/missionGenPFH-dataset"
config = PeftConfig.from_pretrained(peft_model_id)
model = AutoModelForCausalLM.from_pretrained(config.base_model_name_or_path, return_dict=True, load_in_8bit=True, device_map='auto')
tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path)

# Load the Lora model
model = PeftModel.from_pretrained(model, peft_model_id)

def make_inference(company_name, company_description):
    batch = tokenizer(
        f"### Company name and Service Description:\n{company_name}: {company_description}\n\n### Mission:",
        return_tensors="pt",
    )

    batch = {key: value.to("cuda:0") for key, value in batch.items()}

    with torch.cuda.amp.autocast():
        output_tokens = model.generate(**batch, max_new_tokens=50)

    return tokenizer.decode(output_tokens[0], skip_special_tokens=True)

if __name__ == "__main__":
    # make a gradio interface
    import gradio as gr

    gr.Interface(
        make_inference,
        [
            gr.Textbox(lines=2, label="Company Name"),
            gr.Textbox(lines=5, label="Company Service"),
        ],
        gr.Textbox(label="Mission"),
        title="missionStatementGenerator",
        description="missionStatementGenerator-AI is a generative model that generates mission statements for service companies.",
    ).launch()