import gradio

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer


MODEL_NAME = "arnir0/Tiny-LLM"

tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
model = AutoModelForCausalLM.from_pretrained(MODEL_NAME)

def generate_text(prompt, model, tokenizer, max_length=4096, temperature=0.8, top_k=50, top_p=0.95):
    inputs = tokenizer.encode(prompt, return_tensors="pt")

    outputs = model.generate(
        inputs,
        max_length=max_length,
        temperature=temperature,
        top_k=top_k,
        top_p=top_p,
        do_sample=True
    )


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


def my_inference_function(text):
    prompt = f"Summary the context below\n\n{text}"
    generated_text = generate_text(prompt, model, tokenizer)
    
    return generated_text[len(prompt):]

gradio_interface = gradio.Interface(
    fn=my_inference_function,
    inputs="text",
    outputs="text",
    examples=[
    ["Jill"],
    ["Sam"]
    ],
    title="REST API with Gradio and Huggingface Spaces",
    description="This is a demo of how to build an AI powered REST API with Gradio and Huggingface Spaces – for free! Based on [this article](https://www.tomsoderlund.com/ai/building-ai-powered-rest-api). See the **Use via API** link at the bottom of this page.",
    article="© Tom Söderlund 2022"
)
gradio_interface.launch()