Openshift autoscaler based on cpu usage and request

[pax94]

While training some labs for the EX280 I was asked to scale an application automatically based on the cpu usage and resource request.
What is not clear to me is how to define an autoscaler based on the resource request. 
oc autoscale -h only shows the --cpu-percent option but doesn't mention anything regarding requests.
Any idea?

[mamin]

I think that it is a best practice to define resource requests for applications that need to be autoscale.

For example, scale an application to a number of replicas, without defining resource requests in the deployment, to something that Openshift can't handle or guarantee that many replicas.

All I am saying is that is that has nothing to do with autoscale command