Studio: Optimize more iterations without starting from beginning

[medakk]

After completing an optimization run, it would be useful to be able to run a few more optimization iterations(with different hyperparameters?) on the obtained shape model instead of starting from the beginning everytime.

[sheryjoe]

Akin to fine-tuning in deep learning.

[akenmorris]

Yeah, the only problem is that your end result is probably not reproducible. Unless we are suggesting storing different stages of parameters.