matrix-pygmalion-bot/matrix_pygmalion_bot/ai/pygmalion_helpers.py

import asyncio
import os, tempfile
import logging

import json
import requests

from transformers import AutoTokenizer, AutoConfig
from huggingface_hub import hf_hub_download

import io
import base64
from PIL import Image, PngImagePlugin

logger = logging.getLogger(__name__)


async def get_full_prompt(simple_prompt: str, bot, chat_history):

    # Prompt without history
    prompt = bot.name + "'s Persona: " + bot.persona + "\n"
    prompt += "Scenario: " + bot.scenario + "\n\n"
    prompt += "<START>" + "\n"

    for dialogue_item in bot.example_dialogue:
        dialogue_item = dialogue_item.replace('{{user}}', 'You')
        dialogue_item = dialogue_item.replace('{{char}}', bot.name)
        prompt += dialogue_item + "\n\n"
    prompt += "<START>" + "\n"
    #prompt += bot.name + ": " + bot.greeting + "\n"
    #prompt += "You: " + simple_prompt + "\n"
    #prompt += bot.name + ":"

    MAX_TOKENS = 2048
    max_new_tokens = 200
    total_num_tokens = await num_tokens(prompt)
    total_num_tokens += await num_tokens(f"You: " + simple_prompt + "\n{bot.name}:")
    visible_history = []
    current_message = True
    for key, chat_item in reversed(chat_history.chat_history.items()):
        if current_message:
            current_message = False
            continue
        if chat_item.message["en"].startswith('!begin'):
            break
        if chat_item.message["en"].startswith('!'):
            continue
        if chat_item.message["en"].startswith('<ERROR>'):
            continue
        #if chat_item.message["en"] == bot.greeting:
        #    continue
        if chat_item.num_tokens == None:
            chat_item.num_tokens = await num_tokens("{}: {}".format(chat_item.user_name, chat_item.message["en"]))
        # TODO: is it MAX_TOKENS or MAX_TOKENS - max_new_tokens??
        logger.debug(f"History: " + str(chat_item) + " [" + str(chat_item.num_tokens) + "]")
        if total_num_tokens + chat_item.num_tokens < MAX_TOKENS - max_new_tokens:
            visible_history.append(chat_item)
            total_num_tokens += chat_item.num_tokens
        else:
            break
    visible_history = reversed(visible_history)

    if not hasattr(bot, "greeting_num_tokens"):
        bot.greeting_num_tokens = await num_tokens(bot.greeting)
    if total_num_tokens + bot.greeting_num_tokens < MAX_TOKENS - max_new_tokens:
        prompt += bot.name + ": " + bot.greeting + "\n"

    for chat_item in visible_history:
        if chat_item.is_own_message:
            prompt += bot.name + ": " + chat_item.message["en"] + "\n"
        else:
            prompt += "You" + ": " + chat_item.message["en"] + "\n"
    prompt += "You: " + simple_prompt + "\n"
    prompt += bot.name + ":"

    return prompt


tokenizer = None


async def num_tokens(input_text: str):
#    os.makedirs("./models/pygmalion-6b", exist_ok=True)
#    hf_hub_download(repo_id="PygmalionAI/pygmalion-6b", filename="config.json", cache_dir="./models/pygmalion-6b")
#    config = AutoConfig.from_pretrained("./models/pygmalion-6b/config.json")
    global tokenizer
    if not tokenizer:
        tokenizer = AutoTokenizer.from_pretrained("PygmalionAI/pygmalion-6b")
    encoding = tokenizer.encode(input_text, add_special_tokens=False)
    max_input_size = tokenizer.max_model_input_sizes
    return len(encoding)


async def estimate_num_tokens(input_text: str):
    return len(input_text)//4+1