app.py

import gradio as gr
from huggingface_hub import InferenceClient

# """
# For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
# """
# client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")

from transformers import AutoTokenizer, AutoConfig, AutoModelForCausalLM, TextStreamer
from peft import PeftConfig, PeftModel

config = PeftConfig.from_pretrained("jonathanjordan21/mos-mamba-6x130m-trainer")

tokenizer = AutoTokenizer.from_pretrained("jonathanjordan21/mos-mamba-6x130m-trainer", trust_remote_code=True)

# streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)

model = AutoModelForCausalLM.from_pretrained(
    "jonathanjordan21/mos-mamba-6x130m-trainer",
    eos_token_id=tokenizer.eos_token_id,
    trust_remote_code=True
)

model = PeftModel.from_pretrained(model, "jonathanjordan21/mos-mamba-6x130m-trainer",)#, adapter_name="norobots")
model = model.merge_and_unload()

print(model.config.eos_token_id)



def invoke(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    messages = [{"role": "system", "content": system_message}]

    for val in history:
        if val[0]:
            messages.append({"role": "user", "content": val[0]})
        if val[1]:
            messages.append({"role": "assistant", "content": val[1]})

    messages.append({"role": "user", "content": message})

    tokens = tokenizer.apply_chat_template(messages, return_tensors='pt', add_generation_prompt=True)

    response = model.generate(
        tokens, 
        eos_token_id=model.config.eos_token_id, 
        max_new_tokens=max_tokens, 
        # temperature=temperature
    )

    print(response)

    res = tokenizer.batch_decode(response)

    print(res)

    return res[0]
    # yield res


def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    messages = [{"role": "system", "content": system_message}]

    for val in history:
        if val[0]:
            messages.append({"role": "user", "content": val[0]})
        if val[1]:
            messages.append({"role": "assistant", "content": val[1]})

    messages.append({"role": "user", "content": message})

    response = ""

    for message in client.chat_completion(
        messages,
        max_tokens=max_tokens,
        stream=True,
        temperature=temperature,
        top_p=top_p,
    ):
        token = message.choices[0].delta.content

        response += token
        yield response

"""
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
"""
demo = gr.ChatInterface(
    invoke,
    additional_inputs=[
        gr.Textbox(value="You are a helpful assistant.", label="System message"),
        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
        gr.Slider(
            minimum=0.1,
            maximum=1.0,
            value=0.95,
            step=0.05,
            label="Top-p (nucleus sampling)",
        ),
    ],
)


if __name__ == "__main__":
    demo.launch()