PaddleCloud [Feature] Add serve command to start a serverless predict serving port

[Feature] Add serve command to start a serverless predict serving port

Open typhoonzero opened this issue 8 years ago • 2 comments

Can run `paddlecloud serve -model-path xxx -scale 100 -cpu 1 -memory 8Gi -entry "infer.py" to start a serverless URL endpoint for serve the model.

Sep 14 '17 12:09 typhoonzero

Which inferance type we support? Online or offline? If we setup an online serve, we also need to add an ingress rule for the submited serving.

Sep 15 '17 02:09 Yancey0623

Online of course.

Offline inference can use the same method as training.

Sep 15 '17 03:09 typhoonzero