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

通过 pipeline 发布训练好的模型,预测返回的 score 总是为 0 #138

Open
hqsh opened this issue Mar 5, 2022 · 4 comments
Open

Comments

@hqsh
Copy link

hqsh commented Mar 5, 2022

fate 1.7.1.1 是通过 kube fate 部署的多方。

按如下链接的文档,pipeline mini demo 的基础上,pipeline.online.load、pipeline.online.bind 发布模型到 fate serving。

https://github.com/FederatedAI/FATE/blob/master/doc/api/fate_client/pipeline.md

训练结果在 fate broad 上看,完全正常。

但使用训练集进行预测,返回的 score 总是为 0。

代码就如上链接里面的这些,也没找到任何其他关于模型发布的代码。

不知有人操作成功过吗?能否分享一下您的经验,或者相关代码?

感谢!

@luyehui
Copy link

luyehui commented Mar 29, 2022

遇到一样的问题。。。有人分享一下成功的经验吗

@bitdaocao
Copy link

我遇到了差不多的问题,请问楼主问题解决了吗?

@KAgBMan
Copy link

KAgBMan commented Dec 5, 2022

dsl文件如果是v2,在load之前需要先deploy,deploy时需要将预测模型dsl传入,dsl获取方式为
result = client.model.get_predict_dsl(model_id,model_version)

@Timo9Madrid7
Copy link

Timo9Madrid7 commented Dec 26, 2022

"need_run" should be set to false for evaluation module in the host side.

You can firstly try this demo step by step

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

5 participants