Noam Ross comments

Results 237 comments of


                                            Noam Ross

Seemingly unrelated / Multivariate Probit Regression

Thank you. I'm not in a rush so look forward to that work, but I will look and see if `custom_family` can accommodate that approach.

Implement remaining raster::rasterize options

Note that the current `getCover` option just rasterizes at 10x resolution and then aggregates (see [here](https://github.com/cran/raster/blob/4d218a7565d3994682557b8ae4d5b52bc2f54241/R/rasterizePolygons.R#L415-L419)). Future implementation should avoid this so that one never needs 100x memory. Shouldn't be...

Upcoming sf breaks fasterize

Thanks @rsbivand. @mdsumner that would be great. I've been meaning to get to this but have been traveling and will be on vacation for the next week. Will review and...

Return appropriate types

Factors should be handled as well, with transferring of appropriate levels and types. Errors should be thrown when factors are used with inappropriate functions(sum, min, max, etc.)

cell-abstraction mode

So I understand, your intention here is another separate function, that would return a data frame? Within C++ it would use the Edge/Edgelist types, and the parts of the C++...

Outputs are not identical to raster::rasterize()

Thanks! Geospatial stuff is fairly new to me, so glad to have the input.

Pre-submission Inquiry - tidyhte

Closing this as we haven't expanded our standards categories. Will return and re-open if we do.

Possibility to use "own" posterior simulation object in confint

Similarly, I would like to be able to use `gam.mh` outputs for gratia's `draw.*` methods, so as to plot all the model smooths (as well as confidence intervals) using the...

Interaction constraints

Thanks! I will give this a try.

galamm: Generalized Additive Latent and Mixed Models

@osorensen, thanks for following up. My apologies, I think this package fell between cracks in our editor hand-off. I'll follow up later today.