Kristian Agasøster Haaga

> YOu shouldn't expect to access them after the PR is merged Yep, agreed. > Is this issue existing while a PR is open? Yes.

> Seems to me like this should be a different estimator. Maybe. I'm not sure yet. > In any case this exact case presented here seems low priority because it...

@Datseris In CausalityToolsBase, I define `RectangularBinning`. It basically provides different pre-defined ways of partitioning the state space. I use this for the binning based transfer entropy estimators. The current implementation...

To get the (unordered) probabilities for my marginal `x`, I call ```julia probabilities(x, est::ProbabilitiesEstimator) ``` I am understanding you correctly?

I think we're speaking of the same thing. `x` is typically some `Dataset` (which may be the entire multidimensional dataset, or some subset (marginal) of the entire dataset) (which is...

Simply linking these estimators under a section "Compatible estimators" in the doctoring for `TEShannon` should suffice here with the new docs.

For the `OCE` algorithm, we can perhaps manipulate the output string for the `Progress` bar instead of manually printing.

Thanks for opening the issue! Partial cross mapping should be part of the library, but is not implemented currently. I've got a rudimentary implementation of the partial cross mapping that...

> Found another method to construct causal graphs here: Nice. Good catch! We should implement this too. I'm not too worried of the "quality" of the methods when it comes...

Yep, `te_embed(::EmbeddingTE, ::AbstractVector{T}, ::AbstractVector{T}) where T` (and the conditional variant) requires that all input timeseries have the same element type. This shouldn't be necessary.