app.py

import os
from threading import Thread
from typing import Iterator

import gradio as gr
import spaces
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer

MAX_MAX_NEW_TOKENS = 2048
DEFAULT_MAX_NEW_TOKENS = 1024
MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))

DESCRIPTION = """\
# Machine Mindset

MM (Machine_Mindset) series models are developed through a collaboration between FarReel AI Lab(formerly known as the ChatLaw project) and Peking University's Deep Research Institute. These models are large-scale language models for various MBTI types in both Chinese and English, built on the Baichuan and LLaMA2 platforms.
"""

LICENSE = """

---
* Our code adheres to the Apache 2.0 open-source license. Please refer to the [LICENSE](https://github.com/PKU-YuanGroup/Machine-Mindset/blob/main/LICENSE) for specific details of the open-source agreement.

* Our model weights are subject to an open-source agreement based on the original weights, with specific details provided in the Chinese version under the baichuan open-source license. For commercial use, please refer to [model_LICENSE](https://huggingface.co/JessyTsu1/Machine_Mindset_zh_INTP/resolve/main/Machine_Mindset%E5%9F%BA%E4%BA%8Ebaichuan%E7%9A%84%E6%A8%A1%E5%9E%8B%E7%A4%BE%E5%8C%BA%E8%AE%B8%E5%8F%AF%E5%8D%8F%E8%AE%AE.pdf) for further information.

* The English version follows the open-source agreement under the [llama2 license](https://ai.meta.com/resources/models-and-libraries/llama-downloads/).
"""

if not torch.cuda.is_available():
    DESCRIPTION += "\n<p>Running on CPU 🥶 This demo does not work on CPU.</p>"


if torch.cuda.is_available():
    model_id = "FarReelAILab/Machine_Mindset_en_INTJ"
    model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.float16, device_map="auto")
    tokenizer = AutoTokenizer.from_pretrained(model_id)
    tokenizer.use_default_system_prompt = False
    
    model_id_zh = "FarReelAILab/Machine_Mindset_zh_INTJ"
    model_zh = AutoModelForCausalLM.from_pretrained(model_id_zh, torch_dtype=torch.float16, device_map="auto")
    tokenizer_zh = AutoTokenizer.from_pretrained(model_id_zh)
    tokenizer_zh.use_default_system_prompt = False


@spaces.GPU
def generate(
    select_model: str,
    message: str,
    chat_history: list[tuple[str, str]],
    system_prompt: str,
    max_new_tokens: int = 1024,
    temperature: float = 0.6,
    top_p: float = 0.9,
    top_k: int = 50,
    repetition_penalty: float = 1.2,
) -> Iterator[str]:
    if select_model=="INTJ-en":
        conversation = []
        if system_prompt:
            conversation.append({"role": "system", "content": system_prompt})
        for user, assistant in chat_history:
            conversation.extend([{"role": "user", "content": user}, {"role": "assistant", "content": assistant}])
        conversation.append({"role": "user", "content": message})

        input_ids = tokenizer.apply_chat_template(conversation, return_tensors="pt")
        if input_ids.shape[1] > MAX_INPUT_TOKEN_LENGTH:
            input_ids = input_ids[:, -MAX_INPUT_TOKEN_LENGTH:]
            gr.Warning(f"Trimmed input from conversation as it was longer than {MAX_INPUT_TOKEN_LENGTH} tokens.")
        input_ids = input_ids.to(model.device)

        streamer = TextIteratorStreamer(tokenizer, timeout=10.0, skip_prompt=True, skip_special_tokens=True)
        generate_kwargs = dict(
            {"input_ids": input_ids},
            streamer=streamer,
            max_new_tokens=max_new_tokens,
            do_sample=True,
            top_p=top_p,
            top_k=top_k,
            temperature=temperature,
            num_beams=1,
            repetition_penalty=repetition_penalty,
        )
        t = Thread(target=model.generate, kwargs=generate_kwargs)
        t.start()

        outputs = []
        for text in streamer:
            outputs.append(text)
            yield "".join(outputs)

    if select_model=="INTJ-zh":
        conversation = []
        if system_prompt:
            conversation.append({"role": "system", "content": system_prompt})
        for user, assistant in chat_history:
            conversation.extend([{"role": "user", "content": user}, {"role": "assistant", "content": assistant}])
        conversation.append({"role": "user", "content": message})

        input_ids = tokenizer_zh.apply_chat_template(conversation, return_tensors="pt")
        if input_ids.shape[1] > MAX_INPUT_TOKEN_LENGTH:
            input_ids = input_ids[:, -MAX_INPUT_TOKEN_LENGTH:]
            gr.Warning(f"Trimmed input from conversation as it was longer than {MAX_INPUT_TOKEN_LENGTH} tokens.")
        input_ids = input_ids.to(model_zh.device)

        streamer = TextIteratorStreamer(tokenizer_zh, timeout=10.0, skip_prompt=True, skip_special_tokens=True)
        generate_kwargs = dict(
            {"input_ids": input_ids},
            streamer=streamer,
            max_new_tokens=max_new_tokens,
            do_sample=True,
            top_p=top_p,
            top_k=top_k,
            temperature=temperature,
            num_beams=1,
            repetition_penalty=repetition_penalty,
        )
        t = Thread(target=model_zh.generate, kwargs=generate_kwargs)
        t.start()

        outputs = []
        for text in streamer:
            outputs.append(text)
            yield "".join(outputs)

chat_interface = gr.ChatInterface(
    fn=generate,
    additional_inputs=[
        gr.Dropdown(choices=["INTJ-en", "INTJ-zh"], value="INTJ-en", label="Select Model"),
        gr.Textbox(label="System prompt", lines=6),
        gr.Slider(
            label="Max new tokens",
            minimum=1,
            maximum=MAX_MAX_NEW_TOKENS,
            step=1,
            value=DEFAULT_MAX_NEW_TOKENS,
        ),
        gr.Slider(
            label="Temperature",
            minimum=0.1,
            maximum=4.0,
            step=0.1,
            value=0.6,
        ),
        gr.Slider(
            label="Top-p (nucleus sampling)",
            minimum=0.05,
            maximum=1.0,
            step=0.05,
            value=0.9,
        ),
        gr.Slider(
            label="Top-k",
            minimum=1,
            maximum=1000,
            step=1,
            value=50,
        ),
        gr.Slider(
            label="Repetition penalty",
            minimum=1.0,
            maximum=2.0,
            step=0.05,
            value=1.2,
        ),
    ],
    stop_btn=None,
    examples=[
        ["Hello there! How are you doing?"],
        ["Can you explain briefly to me what is the Python programming language?"],
        ["Explain the plot of Cinderella in a sentence."],
        ["How many hours does it take a man to eat a Helicopter?"],
    ],
)

with gr.Blocks(css="style.css") as demo:
    gr.Markdown(DESCRIPTION)
    gr.DuplicateButton(value="Duplicate Space for private use", elem_id="duplicate-button")
    chat_interface.render()
    gr.Markdown(LICENSE)

if __name__ == "__main__":
    demo.queue(max_size=20).launch()