Mitchell O'Hara-Wild

> The `parameters()` method is really nice (seems to be new?) thanks for showing me that. Yes, Like you say, we'd want to do something similar for transformed. Also it's...

Any suggested function names for obtaining the name of a distribution? Comparable to `parameters()`, but obviously can't be `name()`. I also don't like `dist_name()` as the `dist_*()` functions are used...

I had also considered `family()` but was concerned about confusion with related topics like the 'exponential family'. However `stats::family(object, ...)` is used to identify specific distribution names, so it seems...

Closing as serialisation of fables is mostly a question of the vctrs classes they contain. The `format()` method of time vectors from `{tsibble}` should be sufficient to reconstruct the time...

Yep - looks like that's suitable for our usage. I'd probably prefer the RHS to be bare, as this is more consistent with our support for bare formulae being the...

I've thought about this some more, and I prefer introducing a `model_when()` function, rather than vectorising the formula. This will allow you to apply different model specifications (and accordingly, different...

This option would require a model specific implementation, of which none of the currently implemented methods are. However as far as implementation is concerned, there is nothing preventing a newly...

Thanks for the mention - useful update indeed.

Hard to say. Control parameters for models have not yet been implemented, but this will be done for the first release of the package (https://github.com/tidyverts/fable/issues/73). I'm hoping to release the...

Unfortunately not, this was pushed back a lot since it involves a rewrite of the underlying model code on an individual basis. When I first raised this issue, I was...