inference.py

import os
import sys
import fire
import torch
from peft import PeftModel
from transformers import GenerationConfig, LlamaForCausalLM, LlamaTokenizer
from utils.prompter import Prompter

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

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


def main(
    load_8bit: bool = False,
    base_model: str = "",
    lora_weights: str = "DSMI/LLaMA-E/7b",
    prompt_template: str = "",
):
    print("lora_weights: " + str(lora_weights))
    base_model = base_model or os.environ.get("BASE_MODEL", "")

    prompter = Prompter(prompt_template)
    tokenizer = LlamaTokenizer.from_pretrained(base_model)
    if device == "cuda":
        model = LlamaForCausalLM.from_pretrained(
            base_model,
            load_in_8bit=load_8bit,
            torch_dtype=torch.float16,
            device_map="auto",
        )
        model = PeftModel.from_pretrained(
            model,
            lora_weights,
            torch_dtype=torch.float16,
        )
    elif device == "mps":
        model = LlamaForCausalLM.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 = LlamaForCausalLM.from_pretrained(
            base_model, device_map={"": device}, low_cpu_mem_usage=True
        )
        model = PeftModel.from_pretrained(
            model,
            lora_weights,
            device_map={"": device},
        )

    model.config.pad_token_id = tokenizer.pad_token_id = 0  # unk
    model.config.bos_token_id = 1
    model.config.eos_token_id = 2

    if not load_8bit:
        model.half()  # seems to fix bugs for some users.

    model.eval()
    if torch.__version__ >= "2" and sys.platform != "win32":
        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=256,
        **kwargs,
    ):
        prompt = prompter.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 prompter.get_response(output).split("</s>")[0]

    print()
    instruction = "Where can I buy the handmade jewellery?"
    print("Instruction:", instruction)
    print("Response:", evaluate(instruction))
    print()

    instruction = "Create an attractive advertisement for the Christmas sale of the following product."
    input = "Custom Photo Music Plaque,Personalized Photo Frame,Album Cover Song Plaque,Music Photo Name Night Lamp,Photo and Music Gift, Music Prints"
    print("Instruction:", instruction)
    print("Input:", input)
    print("Response:", evaluate(instruction, input))
    print()


if __name__ == "__main__":
    fire.Fire(main)