phelan

btw, other modesl such as maskrnn, dbnet also have the same problem

@xinnjie it seem go-git could not clone lfs file. I tried but failed. btw, my environment already install git lfs and could clone successfully with git clone command

BTW, why you do learn.freeze_to(-2) could you share reason why freeze for 2 first group layer?

yes, I also prefer go with K8S. Recently, I really surprised with this one https://render.com/ super easy to deploy within second at scale Maybe there something interesting inside One super...

@kthui yes, I run triton inference server inside docker and could increase the `shm-size` but when I deploy more and more Python backend models, It reach to maximum of `shm`....

@kthui @Tabrizian any update for this issue

@Tabrizian thanks for your information.

@liranbg `kubectl version` Client Version: version.Info{Major:"1", Minor:"20", GitVersion:"v1.20.4", GitCommit:"e87da0bd6e03ec3fea7933c4b5263d151aafd07c", GitTreeState:"clean", BuildDate:"2021-02-18T16:12:00Z" , GoVersion:"go1.15.8", Compiler:"gc", Platform:"linux/amd64"} Server Version: version.Info{Major:"1", Minor:"16+", GitVersion:"v1.16.15-gke.7800", GitCommit:"cef3156c566a1d1a4b23ee360a760f45bfbaaac1", GitTreeState:"clean", BuildDate:"2020-12-14 T09:12:37Z", GoVersion:"go1.13.15b4", Compiler:"gc", Platform:"linux/amd64"}

>Can you try using helm charts to install Nuclio? I will try it, thanks for quick response

@liranbg I tried with helm helm install nuclio --set registry.pushPullUrl asia.gcr.io/mlpipeline-284807 nuclio/nuclio => error: Error: expected at most two arguments, unexpected arguments: asia.gcr.io/mlpipeline-284807