Moai 7B

🚀 MoAI模型

本倉庫包含了論文 MoAI: Mixture of All Intelligence for Large Language and Vision Models 中所提出模型的權重。該模型致力於解決圖像與文本處理中的複雜任務，為多模態數據的理解和生成提供了強大的支持。

🚀 快速開始

簡單運行代碼基於 MoAI-Github。你僅需以下七個步驟：

[0] 下載MoAI的Github代碼，安裝所需庫，設置必要的環境變量（README.md 中有詳細說明，無需擔心！）

git clone https://github.com/ByungKwanLee/MoAI
bash install

[1] 加載圖像

from PIL import Image
from torchvision.transforms import Resize
from torchvision.transforms.functional import pil_to_tensor
image_path = "figures/moai_mystery.png"
image = Resize(size=(490, 490), antialias=False)(pil_to_tensor(Image.open(image_path)))

[2] 指令提示

prompt = "Describe this image in detail."

[3] 加載MoAI

from moai.load_moai import prepare_moai
moai_model, moai_processor, seg_model, seg_processor, od_model, od_processor, sgg_model, ocr_model \
    = prepare_moai(moai_path='BK-Lee/MoAI-7B', bits=4, grad_ckpt=False, lora=False, dtype='fp16')

[4] MoAI的預處理

moai_inputs = moai_model.demo_process(image=image, 
                                    prompt=prompt, 
                                    processor=moai_processor,
                                    seg_model=seg_model,
                                    seg_processor=seg_processor,
                                    od_model=od_model,
                                    od_processor=od_processor,
                                    sgg_model=sgg_model,
                                    ocr_model=ocr_model,
                                    device='cuda:0')

[5] 生成

import torch
with torch.inference_mode():
    generate_ids = moai_model.generate(**moai_inputs, do_sample=True, temperature=0.9, top_p=0.95, max_new_tokens=256, use_cache=True)

[6] 解碼

answer = moai_processor.batch_decode(generate_ids, skip_special_tokens=True)[0].split('[U')[0]
print(answer)

📄 許可證

本項目採用MIT許可證。