Saulius Lukauskas
Saulius Lukauskas
I have to admit I've never tested this on windows. It looks like the problem lies within syntactic issues in variable declaration in `cdtw.c` that are not compatible with visual...
Hm.. this looks strange. Which compiler you are using?
Also dead: https://www.dropbox.com/s/gnbnkuup3pdfenk/Bottomly_results.zip?dl=1 https://www.dropbox.com/s/j4bznighrbtj02i/Boj_results.zip?dl=1
My fork should now support pasting in OSX through `pngpaste`: https://github.com/lukauskas/imagepaste
Currently there's no good solution for categorical data in decision trees. `sklearn` is working on something (see [here](https://github.com/scikit-learn/scikit-learn/pull/4899)) but so far it hasn't been merged. It is generally accepted that...
These 0.5 splits come from `sklearn` not from `rulefit`. It's the implementation of their trees that create the floats. As far as I am aware `sklearn` does not work with...
As an aside, if my intuition is true, the following formulations are also equivalent where a and b are two rule-transformed columns and ws are the weights. ``` y =...
My concern is that simplest rule might overgeneralise on the training dataset. In theory there is a small chance that the additional rules that are redundant in the training dataset...
Sure: https://support.bioconductor.org/p/p132561/
So this issue has been resolved by adding a warning message. This sort of makes sense for the "wide" format of input, but any particular reason duplicates are not allowed...