使用Llamafactory训练多轮functional call数据构造请教(类openmanus支持模型) #7722
Unanswered
jingtian11
asked this question in
Q&A
Replies: 1 comment
-
|
hello 解决了吗 |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
在类Agent的数据训练时,请教训练数据构造方法
发现数据流程是(使用clauld的input格式)
如果适配sharegpt格式,不知道是否可以理解为
human(原始问题)->funcion_call(模型对于使用function的回复)-> observation(调用工具的结果)-> human(agent对于下一步的定义)
然后希望模型回复
但是这与sharegpt格式不一致
sharegpt中要求observation之后要有gpt(模型的回答),但是训练agent的时候需要有一个【agent对于下一步的定义】的操作,这种情况下
方法(1)加一个空的gpt的部分,凑齐sharegpt的格式
方法(2)把下一步的操作放在observation的后面~
请问有没有有类似训练经验的大佬帮忙建议一下
【或者有没有其他的多轮对话的数据格式可以参考构造】感谢🙏
这里附带一组原始的数据,希望各位老师可以帮忙讲解一下适配的训练和推理格式~!感谢
A_test_example.txt
Beta Was this translation helpful? Give feedback.
All reactions