[ceph-users] diskprediction_local module and trained models

Hi everyone, We have recently noticed diskprediction_local module only works for a set of manufacturers. Hence we have the following questions: Are there any plans to support more manufacturers in the near future? Can we contribute to the process of training new models and how? Can the existing models be used with disks of other vendors (with hackish methods)? Is it that It just doesn't work or it would not be reliable? Basically how does the trained model and the vendor relation work? Say we have a disk for an existing model, what is the best way to test the trained model and the module itself? Thanks in advance