intubate (version 1.0.0)

kernlab: Interfaces for kernlab package for data science pipelines.

Description

Interfaces to kernlab functions that can be used in a pipeline implemented by magrittr.

Usage

ntbt_gausspr(data, ...)
ntbt_kfa(data, ...)
ntbt_kha(data, ...)
ntbt_kkmeans(data, ...)
ntbt_kpca(data, ...)
ntbt_kqr(data, ...)
ntbt_ksvm(data, ...)
ntbt_lssvm(data, ...)
ntbt_rvm(data, ...)
ntbt_sigest(data, ...)
ntbt_specc(data, ...)

Arguments

data

data frame, tibble, list, ...

...

Other arguments passed to the corresponding interfaced function.

Value

Object returned by interfaced function.

Details

Interfaces call their corresponding interfaced function.

Examples

## Not run: 
# library(intubate)
# library(magrittr)
# library(kernlab)
# 
# ##  ntbt_gausspr: Gaussian processes for regression and classification
# data(iris)
# 
# ## Original function to interface
# gausspr(Species ~ ., data = iris, var = 2)
# 
# ## The interface puts data as first parameter
# ntbt_gausspr(iris, Species ~ ., var = 2)
# 
# ## so it can be used easily in a pipeline.
# iris %>%
#   ntbt_gausspr(Species ~ ., var = 2)
# 
# 
# ## ntbt_kfa: Kernel Feature Analysis
# data(promotergene)
# 
# ## Original function to interface
# kfa(~ ., data = promotergene)
# 
# ## The interface puts data as first parameter
# ntbt_kfa(promotergene, ~ .)
# 
# ## so it can be used easily in a pipeline.
# promotergene %>%
#   ntbt_kfa(~ .)
# 
# 
# ## ntbt_kha: Kernel Principal Components Analysis
# data(iris)
# test <- sample(1:150,70)
# 
# ## Original function to interface
# kpc <- kha(~ ., data = iris[-test, -5], kernel = "rbfdot", kpar = list(sigma=0.2),
#            features = 2, eta = 0.001, maxiter = 65)
# pcv(kpc)
# 
# ## The interface puts data as first parameter
# kpc <- ntbt_kha(iris[-test, -5], ~ ., kernel = "rbfdot", kpar = list(sigma=0.2),
#                 features = 2, eta = 0.001, maxiter = 65)
# pcv(kpc)
# 
# ## so it can be used easily in a pipeline.
# iris[-test, -5] %>%
#   ntbt_kha(~ ., kernel = "rbfdot", kpar = list(sigma=0.2),
#            features = 2, eta = 0.001, maxiter = 65) %>%
#   pcv()
# 
# 
# ## ntbt_kkmeans: Kernel k-means
# ## Original function to interface
# sc <- kkmeans(~ ., data = iris[-test, -5], centers = 3)
# centers(sc)
# 
# ## The interface puts data as first parameter
# sc <- ntbt_kkmeans(iris[-test, -5], ~ ., centers = 3)
# centers(sc)
# 
# ## so it can be used easily in a pipeline.
# iris[-test, -5] %>%
#   ntbt_kkmeans(~ ., centers = 3) %>%
#   centers()
# 
# 
# ## ntbt_kpca: Kernel Principal Components Analysis
# data(iris)
# test <- sample(1:150,20)
# 
# ## Original function to interface
# kpc <- kpca(~ ., data = iris[-test, -5], kernel = "rbfdot",
#             kpar = list(sigma = 0.2), features = 2)
# pcv(kpc)
# 
# ## The interface puts data as first parameter
# kpc <- ntbt_kpca(iris[-test, -5], ~ ., kernel = "rbfdot",
#                  kpar = list(sigma = 0.2), features = 2)
# pcv(kpc)
# 
# ## so it can be used easily in a pipeline.
# iris[-test, -5] %>%
#   ntbt_kpca(~ ., kernel = "rbfdot",
#             kpar = list(sigma = 0.2), features = 2) %>%
#   pcv()
# 
# 
# ## ntbt_kqr: Kernel Quantile Regression
# ## Not found example using formula interface, and I am
# ## completely ignorant to construct one.
# x <- sort(runif(300))
# y <- sin(pi*x) + rnorm(300,0,sd=exp(sin(2*pi*x)))
# 
# dkqr <- data.frame(x, y)
# 
# ## Original function to interface
# set.seed(1)
# kqr(x, y, tau = 0.5, C = 0.15)
# 
# ## The interface puts data as first parameter
# set.seed(1)
# ntbt_kqr(dkqr, x, y, tau = 0.5, C = 0.15)
# 
# ## so it can be used easily in a pipeline.
# set.seed(1)
# dkqr %>%
#   ntbt_kqr(x, y, tau = 0.5, C = 0.15)
# 
# 
# ## ntbt_ksvm: Support Vector Machines
# data(spam)
# index <- sample(1:dim(spam)[1])
# spamtrain <- spam[index[1:floor(dim(spam)[1]/2)], ]
# spamtest <- spam[index[((ceiling(dim(spam)[1]/2)) + 1):dim(spam)[1]], ]
# 
# ## Original function to interface
# set.seed(1)
# ksvm(type ~ ., data = spamtrain, kernel = "rbfdot",
#      kpar = list(sigma = 0.05), C = 5, cross = 3)
# 
# ## The interface puts data as first parameter
# set.seed(1)
# ntbt_ksvm(spamtrain, type ~ ., kernel = "rbfdot",
#           kpar = list(sigma = 0.05), C = 5, cross = 3)
# 
# ## so it can be used easily in a pipeline.
# set.seed(1)
# spamtrain %>%
#   ntbt_ksvm(type ~ ., kernel = "rbfdot",
#             kpar = list(sigma = 0.05), C = 5, cross = 3)
# 
# 
# ## ntbt_lssvm: Least Squares Support Vector Machine
# data(iris)
# 
# ## Original function to interface
# set.seed(1)
# lssvm(Species ~ ., data = iris)
# 
# ## The interface puts data as first parameter
# set.seed(1)
# ntbt_lssvm(iris, Species ~ .)
# 
# ## so it can be used easily in a pipeline.
# set.seed(1)
# iris %>%
#   ntbt_lssvm(Species ~ .)
# 
# 
# ## ntbt_rvm: Relevance Vector Machine
# ## Not found example using formula interface, and I am
# ## completely ignorant to construct one.
# x <- seq(-20,20,0.1)
# y <- sin(x)/x + rnorm(401,sd=0.05)
# 
# drvm <- data.frame(x, y)
# 
# ## Original function to interface
# set.seed(1)
# rvm(x, y, tau = 0.5, C = 0.15)
# 
# ## The interface puts data as first parameter
# set.seed(1)
# ntbt_rvm(drvm, x, y, tau = 0.5, C = 0.15)
# 
# ## so it can be used easily in a pipeline.
# set.seed(1)
# drvm %>%
#   ntbt_rvm(x, y, tau = 0.5, C = 0.15)
# 
# 
# ## ntbt_sigest: Hyperparameter estimation for the Gaussian Radial Basis kernel
# data(promotergene)
# 
# ## Original function to interface
# set.seed(1)
# sigest(Class ~ ., data = promotergene)
# 
# ## The interface puts data as first parameter
# set.seed(1)
# ntbt_sigest(promotergene, Class ~ .)
# 
# ## so it can be used easily in a pipeline.
# set.seed(1)
# promotergene %>%
#   ntbt_sigest(Class ~ .)
# 
# ## ntbt_specc: Spectral Clustering
# ## Not found example using formula interface, and I am
# ## completely ignorant to construct one.
# ## End(Not run)