app.py

import os
from typing import Optional, Tuple, Any
from functools import partial
import gradio as gr
import spaces
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
from dataclasses import dataclass

torch.set_grad_enabled(False)
model_name = "TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ"
token = os.environ['hf_token']
pipe = pipeline("text-generation", model=model_name, device="cuda")
generate_kwargs = {'max_new_tokens': 20}


system_prompt = '''You are given a partial input text for another AI chat interface. 
Propose auto-completion to the text. You have several roles:
- Fight under-specification.
- Complete text to save the user time.

Don't suggest anything if there are no good suggestions. 
Make sure the suggestions are valid completions of the text! Suggest only up to 5 words ahead. The scheme of your answer should be "answer1;answer2;answer3" (return between 0 to 4 answers).
Answers should be only the completions themselves. 
You will now get a blank message from the user and then after your answer, the user will give you the text to complete.

'''


extra_prompt = '''
Examples: 
(1)
User: "Help me write a sentiment analysis pipeline"
Assistant: "using huggingface;using NLTK;using python"

(2)
User: "My name is"
Assistant: "" (nothing much to contribute at this point. return nothing)

(3)
User: "Help me find a present for my"
Assistant: "girlfriend;mother;father;friend"
'''


start_messages = [
    {'role': 'system', 'content': system_prompt}, 
    {'role': 'user', 'content': '  '},
    {'role': 'assistant', 'content': '<Waiting for text>'}
]


# functions
@dataclass
class PastKV:
    past_key_values: Any = None

past_key_values = PastKV()


def past_kv_to_device(past_kv, device):
    return tuple((k.to(device).detach(), v.to(device).detach()) for k, v in past_kv)


@spaces.GPU
def set_past_key_values():
    model, tokenizer = pipe.model, pipe.tokenizer
    tokenized = tokenizer.apply_chat_template(start_messages, return_tensors='pt')    
    
    # Check that this is indeed a prefix of the entire message
    test_messages = [*start_messages, {'role': 'user', 'content': 'Hello World!'}]
    tokenized_test = tokenizer.apply_chat_template(test_messages, return_tensors='pt')    
    assert (tokenized_test[:, :tokenized.shape[1]] == tokenized).all().cpu().item()
    past_key_values.past_key_values = model(tokenized.to(model.device)).past_key_values
    return True
    

@spaces.GPU
def generate(text, past_key_values):
    messages = [
        *start_messages,
        {'role': 'user', 'content': text}
    ]
    response = pipe(messages, 
                    past_key_values=past_key_values.past_key_values, #past_kv_to_device(past_key_values, pipe.model.device), 
                    **generate_kwargs)[0]['generated_text']
    return response[-1]['content']

    
if __name__ == "__main__":
    with torch.no_grad():
        set_past_key_values()
        print(past_key_values)
        demo = gr.Interface(partial(generate), 
                            inputs="textbox", outputs="textbox")
        demo.launch()