internal: Random KNN Internal Functions In rknn: Random KNN Classification and Regression

Description

Some internal and under-development functions

Usage

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13``` ``` rbyb(p, m, eta) rbyp(p, m, eta) rbyv(p, m, nu) rbyz(p, m) rbyz.sim(p, m, nsim=1000) rbyz.geo(p, m=floor(sqrt(p)), rmax=p) rbylambda(p, m, lambda=1) knn(train, test, cl, k=1) knn.cv (train, cl, k=1) knn.reg(train, test = NULL, y, k = 3) pressresid(obj) ```

Arguments

 `m` Number of elements in a subset to be drawn. `p` Total number of available features. `mtry` Number of features to be drawn for each KNN. `eta` Coverage Probability. `nu` mean mutiplicity of a feature `rmax` number of series terms for independent geometric approximation `nsim` number of simulations for geometric simulation. `lambda` mean number of silient features. `samples` A vector of indice for a set of observations. `cl` A factor for classification labels. `train` A data matrix. `test` A data matrix. `y` A vector of responses. `k` Number of nearest neighbors. `cl` A vector of class labels. `K` Number of folds for cross-validation. `pk` A real number between 0 and to indicate the proportion of the feature set to be kept in each step. `r` Number of KNN to be generated. `seed` An integer seed. `criterion` either uses mean_accuracy or mean_support for best. `obj` A linear model.

Author(s)

Shengqiao Li<[email protected]>

rknn documentation built on May 30, 2017, 3:33 a.m.