Closed rachlobay closed 2 months ago
To address your main points 1. sounds good to me. For 2., I did have a version of the vignette with this, but I ended up removing n_training
to keep it as simple as we can for the arguments used... This is a good point though, so I noted in the vignette that the reader can use n_training
in place of the filter command. On 3., for now I think that the vignette is sufficient without more (we could aways add more to it later on).
This branch has the ARX classifier vignette we went over (I removed the one heatmap that you didn't like that ended up looking more like a bar chart).