Kinto
Adding an R package
@baslat @MattCowgill Here's what I've got. A few thoughts right off the bat: * I think the addition of `labelled` as a dependency is well-justified. Storing the labels in additional...
Hi, Thanks for your excellent package. One thing I find myself needing to do often enough is to create a single labeled vector out of a vector of codes and...
Hi, Fantastic piece of software! Thanks for your work I would like to use nomnoml to produce ERDs. However, it is not currently possible to refer to weak entities, identifying...
Stellar package! I think it would be handy to have an `autoAttendantBar()` which behaves similarly to `autoWaiter()`, i.e. it is shown on `shiny:recalculating` and closed on `shiny:error` or `shiny:value`. Also,...
The lack of colour channels in MNIST (and MNIST-like) datasets means we get errors. It looks like the network implementations aren't set up to handle single channel inputs? Could you...
It would be useful to also see the performance of each SSL model against the purely supervised backbone run on the labeled data. For example, TSVM vs pure SVM: ```python...
Hey! Just adding a link to a make-like R package I wrote. STAT545 is what put me onto the idea in the first place. Thanks for making the course notes...
Relates to #755. I didn't grab the output earlier (is there a way to print without re-benchmarking?) but this makes no difference to the performance. Submitting a PR for record...
Relates to #755 ``` $ cargo bench --bench specializations unique/fold unique/fold time: [11.672 µs 11.710 µs 11.753 µs] change: [-5.4377% -4.4270% -3.6741%] (p = 0.00 < 0.05) Performance has improved....