Hi,
I can foresee a situation where it would be useful to separate the capture and training sections in a model.
Imagine that I am in a studio and i have got the tone I want and have tracked it for the album and I want to take that sound out with me on tour. It would be useful to do the capture there and then, but leave the training until later to save on expensive studio time. I know that this would stop me from doing the A/B testing at the end, but there are situations where I would be willing to sacrifice this.
