QwenVL系列多模态模型学习笔记-第3篇

第二代 Qwen2-VL 应用案例

模型下载 —— 基于魔塔社区

本次 Qwen2-VL 开源了两个尺寸的模型，Qwen2-VL-2B-Instruct 和 Qwen2-VL-7B-Instruct，以及其 GPTQ 和 AWQ 的量化版本。

模型链接：

Qwen2-VL-2B-Instruct：https://www.modelscope.cn/models/qwen/Qwen2-VL-2B-Instruct

Qwen2-VL-7B-Instruct：https://www.modelscope.cn/models/qwen/Qwen2-VL-7B-Instruct

推荐使用 ModelScope CLI 下载模型

modelscope download --model=qwen/Qwen2-VL-7B-Instruct --local_dir ./Qwen2-VL-7B-Instruct

模型推理 —— 基于 transformers

安装依赖：

pip install git+https://github.com/huggingface/transformers

pip install qwen-vl-utils

单图推理

from PIL import Image
import torch
from transformers import Qwen2VLForConditionalGeneration, AutoTokenizer, AutoProcessor
from qwen_vl_utils import process_vision_info
from modelscope import snapshot_download


model_dir = "/mnt/workspace/Qwen2-VL-2B-Instruct"
min_pixels = 256*28*28
max_pixels = 1280*28*28

# Load the model in half-precision on the available device(s)
model = Qwen2VLForConditionalGeneration.from_pretrained(
    model_dir, device_map="auto", torch_dtype = torch.float16
)
# Load the processor
processor = AutoProcessor.from_pretrained(
	model_dir, min_pixels=min_pixels, max_pixels=max_pixels
)

messages = [
    {
        "role": "user", 
        "content": [
            {"type": "image", "image": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-VL/assets/demo.jpeg"}, 
            {"type": "text", "text": "Describe this image."}
        ]
    }
]

# Preparation for inference
text = processor.apply_chat_template(
    messages, tokenize=False, add_generation_prompt=True
)
image_inputs, video_inputs = process_vision_info(messages)
inputs = processor(
	text=[text], images=image_inputs, videos=video_inputs, 
    padding=True, return_tensors="pt"
)
inputs = inputs.to('cuda')

# Inference: Generation of the output
generated_ids = model.generate(**inputs, max_new_tokens=128)
generated_ids_trimmed = [
    out_ids[len(in_ids):] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
]

output_text = processor.batch_decode(
	generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
)

print(output_text)

把输入文本编程对话形式：输入文本被处理成对话模板的样式，图片或者视频暂时被 <|image_pad|> 这样的占位符替代。

 <|im_start|>system
You are a helpful assistant.<|im_end|>
<|im_start|>user
<|vision_start|><|image_pad|><|vision_end|>Describe this image.<|im_end|>
<|im_start|>assistant

多图推理

# Messages containing multiple images and a text query
messages = [
    {
        "role": "user", 
        "content": [
            {"type": "image", "image": "file:///path/to/image1.jpg"}, 
            {"type": "image", "image": "file:///path/to/image2.jpg"}, 
            {"type": "text", "text": "Identify the similarities between these images."}
        ]
    }
]

视频理解

# Messages containing a video and a text query
messages = [
    {
        "role": "user", 
        "content": [
            {
                "type": "video", 
                "video": "file:///path/to/video1.mp4", 
                'max_pixels': 360*420, 
                'fps': 1.0
            }, 
            {"type": "text", "text": "Describe this video."}
        ]
    }
]

模型推理 —— 基于 vllm

安装依赖

pip install git+https://github.com/fyabc/vllm.git@add_qwen2_vl_new

启动 OpenAI 接口服务

python -m vllm.entrypoints.openai.api_server --served-model-name Qwen2-VL-7B-Instruct --model model_path

调用服务

curl http://localhost:8000/v1/chat/completions \    
	-H "Content-Type: application/json" \    
	-d '{    
	"model": "Qwen2-VL-7B-Instruct",    
	"messages": [    
		{"role": "system", "content": "You are a helpful assistant."},    
		{"role": "user", "content": [        
			{"type": "image_url", "image_url": {"url": "https://modelscope.oss-cn-beijing.aliyuncs.com/resource/qwen.png"}},        
			{"type": "text", "text": "What is the text in the illustrate?"}    
		]}    
	]    
	}'

调用服务

from openai import OpenAI

# Set OpenAI's API key and API base to use vLLM's API server.
openai_api_key = "EMPTY"
openai_api_base = "http://localhost:8000/v1"

client = OpenAI(
	api_key=openai_api_key, base_url=openai_api_base
)

chat_response = client.chat.completions.create(
	model="Qwen2-7B-Instruct",    
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": [
            {"type": "image_url", "image_url": {"url": "https://modelscope.oss-cn-beijing.aliyuncs.com/resource/qwen.png"}},
            {"type": "text", "text": "What is the text in the illustrate?"},        
        ]},
	]
)

print("Chat response:", chat_response)

模型微调 —— 基于 swift

在开始微调之前，请确保您的环境已准备妥当。

git clone https://github.com/modelscope/swift.gitcd swift

pip install -e .[llm]

pip install pyav qwen_vl_utils

图像描述微调

# 默认会将 lora_target_modules 设置为 llm 的所有linear
CUDA_VISIBLE_DEVICES=0,1,2,3 NPROC_PER_NODE=4 swift sft \  
	--model_type qwen2-vl-7b-instruct \  
    --model_id_or_path qwen/Qwen2-VL-7B-Instruct \  
    --sft_type lora \  
    --dataset coco-en-mini#20000 \  
    --deepspeed default-zero2
    --dataset train.jsonl \  
    --val_dataset val.jsonl \  # 自定义数据集

以下是自定义数据集的样例：

{"query": "<image>55555", "response": "66666", "images": ["image_path"]}

{"query": "eeeee<image>eeeee<image>eeeee", "response": "fffff", "history": [], "images": ["image_path1", "image_path2"]}

{"query": "EEEEE", "response": "FFFFF", "history": [["query1", "response2"], ["query2", "response2"]], "images": []}

微调后推理脚本如下：

CUDA_VISIBLE_DEVICES=0 swift infer \    
	--ckpt_dir output/qwen2-vl-7b-instruct/vx-xxx/checkpoint-xxx \    
	--load_dataset_config true 
	--merge_lora true

图像 grounding 微调

# swift 跨模型通用格式
{
    "query": "Find <bbox>", 
    "response": "<ref-object>", 
    "images": ["/coco2014/train2014/COCO_train2014_000000001507.jpg"], 
    "objects": "[{\"caption\": \"guy in red\", \"bbox\": [138, 136, 235, 359], \"bbox_type\": \"real\", \"image\": 0}]" 
}

{
    "query": "Find <ref-object>", 
    "response": "<bbox>", 
    "images": ["/coco2014/train2014/COCO_train2014_000000001507.jpg"], 
    "objects": "[{\"caption\": \"guy in red\", \"bbox\": [138, 136, 235, 359], \"bbox_type\": \"real\", \"image\": 0}]" 
}

# qwen2-vl-chat 特定格式，注意特殊字符的存在
{
	"query": "Find <|object_ref_start|>the man<|object_ref_end|>", 
	"response": "<|box_start|>(123,235),(324,546)<|box_end|>", 
	"images": ["/coco2014/train2014/COCO_train2014_000000001507.jpg"]
}

视频微调

NFRAMES=24 MAX_PIXELS=100352 CUDA_VISIBLE_DEVICES=0,1,2,3 NPROC_PER_NODE=4 swift sft \  
	--model_type qwen2-vl-7b-instruct \  
	--model_id_or_path qwen/Qwen2-VL-7B-Instruct \  
	--sft_type lora \  
	--dataset video-chatgpt \  
	--deepspeed default-zero2