Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

clueai微调,结果与微调内容的预期不一致 #23

Open
ampyang opened this issue Mar 21, 2023 · 4 comments
Open

clueai微调,结果与微调内容的预期不一致 #23

ampyang opened this issue Mar 21, 2023 · 4 comments

Comments

@ampyang
Copy link

ampyang commented Mar 21, 2023

已经成功上传了微调数据,调用训练好的微调模型也是成功的,但是结果不是微调预期的结果。
上传的json内容如下:
{"id": 1, "type": "qa", "question": "用户:闵行区图书馆地址是?\n小元:", "answer": "闵行区名都路85号"}
{"id": 2, "type": "qa", "question": "用户:开放时间是\n小元:", "answer": "平日开放时间:【一楼】借还处\n周二至周日 9:00–20:00\n办证处\n周二至周日 9:00–16:00\n报刊阅览室\n 周一 12:00–17:00\n 周二至周日 9:00–20:00\n城市书房闵图馆\n周一至周日 9:00–17:00"}

然后测试微调结果代码如下:
response = cl.start_finetune_model(
engine_key=engine_key,
base_model_name="ChatYuan")
print(response)
prompt= '''用户:闵行区图书馆地址是
小元:'''

prediction = cl.finetune_generate(
engine_key=engine_key,
base_model_name='ChatYuan',
prompt=prompt)
result = prediction.generations[0].text

返回的结果跟不使用微调模型相比,没有变化

@joytianya
Copy link
Collaborator

可能需要更多的训练,后续这个微调会开放更多参数给用户,后面会开发开放

@ampyang
Copy link
Author

ampyang commented Mar 21, 2023

可能需要更多的训练,后续这个微调会开放更多参数给用户,后面会开发开放

是需要更多的微调训练数据的意思吗?上传的微调数据太少了是吗?

@joytianya
Copy link
Collaborator

后面也会升级微调算法

@ampyang
Copy link
Author

ampyang commented Mar 22, 2023

好的,谢谢

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants