app.py

import torch
from peft import PeftModel
import transformers
import gradio as gr

from transformers import AutoModelForCausalLM, AutoTokenizer, GenerationConfig

tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-v0.1")

BASE_MODEL = "mistralai/Mistral-7B-v0.1"
LORA_WEIGHTS = "./qlora-out.mistral.0.9978/"

if torch.cuda.is_available():
    device = "cuda"
else:
    device = "cpu"

try:
    if torch.backends.mps.is_available():
        device = "mps"
except:
    pass

if device == "cuda":
    from transformers import BitsAndBytesConfig

    nf4_config = BitsAndBytesConfig(
        load_in_4bit=True,
        bnb_4bit_quant_type="nf4",
        bnb_4bit_use_double_quant=True,
        bnb_4bit_compute_dtype=torch.bfloat16
    )

    model = AutoModelForCausalLM.from_pretrained(BASE_MODEL, quantization_config=nf4_config)

    model = PeftModel.from_pretrained(
        model, LORA_WEIGHTS, torch_dtype=torch.float16, force_download=True
    )
elif device == "mps":
    model = AutoModelForCausalLM.from_pretrained(
        BASE_MODEL,
        device_map={"": device},
        torch_dtype=torch.float16,
    )
    model = PeftModel.from_pretrained(
        model,
        LORA_WEIGHTS,
        device_map={"": device},
        torch_dtype=torch.float16,
    )
else:
    model = AutoModelForCausalLM.from_pretrained(
        BASE_MODEL, device_map={"": device}, low_cpu_mem_usage=True
    )
    model = PeftModel.from_pretrained(
        model,
        LORA_WEIGHTS,
        device_map={"": device},
    )


def generate_prompt(instruction, input=None):
    if input:
        return f"""Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.

### Instruction:
{instruction}

### Input:
{input}

### Response:"""
    else:
        return f"""Below is an instruction that describes a task. Write a response that appropriately completes the request.

### Instruction:
{instruction}

### Response:"""

if device != "cpu":
    model.half()
model.eval()
if torch.__version__ >= "2":
    model = torch.compile(model)


def evaluate(
    instruction,
    input=None,
    temperature=0.1,
    top_p=0.75,
    top_k=40,
    num_beams=4,
    max_new_tokens=128,
    **kwargs,
):
    prompt = generate_prompt(instruction, input)
    inputs = tokenizer(prompt, return_tensors="pt")
    input_ids = inputs["input_ids"].to(device)
    generation_config = GenerationConfig(
        temperature=temperature,
        top_p=top_p,
        top_k=top_k,
        num_beams=num_beams,
        **kwargs,
    )
    with torch.no_grad():
        generation_output = model.generate(
            input_ids=input_ids,
            generation_config=generation_config,
            return_dict_in_generate=True,
            output_scores=True,
            max_new_tokens=max_new_tokens,
        )
    s = generation_output.sequences[0]
    output = tokenizer.decode(s)
    return output.split("### Response:")[1].strip()


g = gr.Interface(
    fn=evaluate,
    inputs=[
        gr.components.Textbox(
            lines=2, label="Utasítás", placeholder="Mesélj kicsit a szürkemarháról!"
        ),
        gr.components.Textbox(lines=2, label="Input", placeholder="üres"),
        gr.components.Slider(minimum=0, maximum=1, value=0.1, label="Temperature"),
        gr.components.Slider(minimum=0, maximum=1, value=0.75, label="Top p"),
        gr.components.Slider(minimum=0, maximum=100, step=1, value=40, label="Top k"),
        gr.components.Slider(minimum=1, maximum=4, step=1, value=4, label="Beams"),
        gr.components.Slider(
            minimum=1, maximum=512, step=1, value=128, label="Max tokens"
        ),
    ],
    outputs=["text"],
    title="szürkemarha-mistral-v1",
    description="A szürkemarha-mistral egy fejlesztés alatt álló 7 milliárd paraméteres Mistral-0.1 alapú model LoRA finomhangolva instrukciókövetésre.",
)
g.queue(concurrency_count=1)
g.launch()