Jacob

Hi @andcastillo and sorry I haven't been helping with this library lately (I had planned to do so months ago). Would you mind elaborating on how the current implementation differs...

> This could be a good chance to learn some ts for me, do you mind if I do this? No objections from me. I also don't see any reason...

Wow, I didn't realize that the damping parameter wasn't being adjusted during the algorithm. That explains a lot! I would suggest that in order to avoid breaking the API (i.e....

Sure, I'm just saying that rather than (1) make a breaking change to 1.x or (2) bumping to 2.x, just add support to 1.x in a non-breaking way and document...

@m93a Thanks for sharing. Something like that might be well-suited for deployment to GH pages. My concern with that was that I didn't think it'd be easy to allow live...

FYI I am taking a stab at this also. https://github.com/jacobq/levenberg-marquardt/tree/support-error

@jobo322 @m93a Regarding the question about [divergence in the `Math.sin` test](https://github.com/mljs/levenberg-marquardt/pull/19#issuecomment-426658687), this family of functions seems to have a deep canyon near the solution but a broad plateau for low...

You can also see how holding the frequency constant could lead toward the sub-optimal solution of zero amplitude: 

Since there is a lot of work in this PR I'd like to break it down into smaller bites (each as their own commit), and possibly even merge them separately....

> Should I push all the minor fixes we talked about before you start rebasing? Or will you take care of it yourself? I'm talking specifically: ... Yes, please. I...