app.py

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

peft_model_id = "ambrosemcduffy/bloom-1b7-lora-ads"
config = PeftConfig.from_pretrained(peft_model_id)
base_model = AutoModelForCausalLM.from_pretrained(
    config.base_model_name_or_path,
    return_dict=True,
    load_in_8bit=False,
    device_map="auto",
)
tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path)

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

def make_inference(question):
    input_text = "### This is your question {}\n".format(question)
    batch = tokenizer(input_text, return_tensors='pt')

    with torch.cuda.amp.autocast():
        output_tokens = model.generate(**batch, max_length=50, num_return_sequences=1)

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

if __name__ == "__main__":
    import gradio as gr

    gr.Interface(
    make_inference,
    gr.inputs.Textbox(lines=2, label="Question"),
    gr.outputs.Textbox(label="Answer"),
    title="BlackQA",
    description="Generated Text of Black heros",
).launch()