app.py

import os
import logging
from logging.handlers import RotatingFileHandler

import gradio as gr
from transformers import AutoTokenizer, BitsAndBytesConfig
from langchain_huggingface import ChatHuggingFace
from langchain.prompts import PromptTemplate
from langchain.chains import LLMChain

# Настройка логирования
log_file = '/tmp/app_debug.log'
logger = logging.getLogger(__name__)
logger.setLevel(logging.DEBUG)
file_handler = RotatingFileHandler(log_file, maxBytes=10*1024*1024, backupCount=5)
file_handler.setFormatter(logging.Formatter('%(asctime)s - %(levelname)s - %(message)s'))
logger.addHandler(file_handler)

logger.debug("Application started")

MODEL_ID = "Qwen/Qwen2.5-Coder-7B-Instruct"
MODEL_NAME = MODEL_ID.split("/")[-1]

template = """<|im_start|>system\n{system_prompt}\n<|im_end|>\n{history}<|im_start|>user\n{human_input}\n<|im_end|>\n<|im_start|>assistant\n"""
prompt = PromptTemplate(template=template, input_variables=["system_prompt", "history", "human_input"])

def format_history(history):
    return "".join([f"<|im_start|>user\n{h[0]}\n<|im_end|>\n<|im_start|>assistant\n{h[1]}\n<|im_end|>\n" for h in history])

def predict(message, history, system_prompt, temperature, max_new_tokens, top_k, repetition_penalty, top_p):
    logger.debug(f"Received prediction request: message='{message}', system_prompt='{system_prompt}'")
    
    chat_model.temperature = temperature
    chat_model.max_new_tokens = max_new_tokens
    chat_model.top_k = top_k
    chat_model.repetition_penalty = repetition_penalty
    chat_model.top_p = top_p

    chain = LLMChain(llm=chat_model, prompt=prompt)
    
    try:
        formatted_history = format_history(history)
        for chunk in chain.stream({"system_prompt": system_prompt, "history": formatted_history, "human_input": message}):
            yield chunk["text"]
        logger.debug(f"Prediction completed successfully for message: '{message}'")
    except Exception as e:
        logger.exception(f"Error during prediction: {str(e)}")
        yield "An error occurred during processing."

tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
chat_model = ChatHuggingFace(
    model_name=MODEL_ID,
    tokenizer=tokenizer,
    model_kwargs={
        "device_map": "auto",
        "quantization_config": BitsAndBytesConfig(load_in_4bit=True),
    }
)

logger.debug("Model and tokenizer loaded successfully")

gr.ChatInterface(
    predict,
    title=f"🤖 {MODEL_NAME}",
    description=f"This is the {MODEL_NAME} model designed for coding assistance and general AI tasks.",
    examples=[
        ["Can you solve the equation 2x + 3 = 11 for x in Python?"],
        ["Write a Java program that checks if a number is even or odd."],
        ["How can I reverse a string in JavaScript?"],
        ["Create a C++ function to find the factorial of a number."],
        ["Write a Python list comprehension to generate a list of squares of numbers from 1 to 10."],
    ],
    additional_inputs=[
        gr.Textbox("You are a code assistant.", label="System prompt"),
        gr.Slider(0, 1, 0.3, label="Temperature"),
        gr.Slider(128, 4096, 1024, label="Max new tokens"),
        gr.Slider(1, 80, 40, label="Top K sampling"),
        gr.Slider(0, 2, 1.1, label="Repetition penalty"),
        gr.Slider(0, 1, 0.95, label="Top P sampling"),
    ],
    theme=gr.themes.Soft(primary_hue="blue"),
).queue().launch()

logger.debug("Chat interface initialized and launched")