I asked to deploy my AutoML model over an hour ago but it is processing the request… How long should this take?
Answering my own question: it took almost 2 hours to deploy a very simple model.
Yes. I also have seen such behaviour. Not sure whether this is because in most of the cloud services there is a concept of most used services will work fast. They are already in memory and works - probably currently the vertexAPI usage is low now, so the delay.