MiniCPM-V 2.6 训练指南
输入“/”快速插入内容
🔥MiniCPM-V 2.6 训练指南
MiniCPM-V 2.6 训练指南
飞书用户309
飞书用户77842024年9月9日修改
🙌
如果您有任何问题，请在GitHub上给我们提issue，技术同学会集中解答
1.
笔者的训练环境介绍MiniCPMV训练的pip list
2.
获取minicpmv的github代码：
代码块
git clone https://github.com/OpenBMB/MiniCPM-V.git
3.
安装依赖包
代码块
cd MiniCPM-V
pip install -r requirements.txt
4.
处理数据集
1.
处理数据集成以下形式：
a.
id值不可重复
b.
<image>\n应该出现在每个数据集对话数据的开头
c.
"image"对应的地址需要存在图片
d.
每个conversations对应的列表中是一个多轮对话，content代表对话内容，role对应user代表用户输入，role对应assistant代表模型输出
单图（一轮对话中仅一张图）
代码块
[
{ "id": "0",
"conversations": [
{
"content": "<image>\nWho are they?",
"role": "user"
},
{
"content": "They're Kane and Gretzka from Bayern Munich.",
"role": "assistant"
},
{
"content": "What are they doing?",
"role": "user"
},
{
"content": "They are celebrating on the soccer field.",
"role": "assistant"
}
],
"image":
"/root/ld/ld_project/LLaMA-Factory/data/mllm_demo_data/1.jpg"
}
...以上是单条数据，列表中可存在多个相同格式的数据
]
多图（一次对话包含多张图片）：
代码块
[
{
"id": "0",
"image": {
"<image_00>": "path/to/image_0.jpg",
"<image_01>": "path/to/image_1.jpg",
"<image_02>": "path/to/image_2.jpg",
"<image_03>": "path/to/image_3.jpg"
},
"conversations": [
{
"role": "user",
"content": "How to create such text-only videos using CapCut?\n<image_00>\n<image_01>\n<image_02>\n<image_03>\n"
},
{
"role": "assistant",
"content": "To create a text-only video as shown in the images, follow these steps in CapCut..."
}
]
}
]