|
library(readr)
|
library(dplyr)
|
|
library(ggplot2)
|
library(broom)
|
|
library(caret)
|
library(rpart)
|
|
library(splines)
|
library(party)
|
|
library(leaps)
|
library(glmnet)
|
|
library(MASS)
|
library(class)
|
|
as.array(x)
|
as.character(x)
|
|
as.data.frame(x)
|
as.factor(x)
|
|
as.logical(x)
|
as.numeric(x)
|
|
folds <- crossv_kfold(data, k = 5) [k is the number of folds]
|
|
lambdas_to_try <- 10seq(-3, 5, length.out = 100)^
|
|
ridge_cv <- cv.glmnet(X, y, alpha = 0, lambda = lambdas_to_try,standardize = TRUE, nfolds = 10) [Setting alpha = 0 implements ridge regression]
|
|
lasso <- glmnet(data_x, data_y, alpha = 1.0)
|
|
randomForest(formula, data) [formula is a formula describing the predictor and response variables. data is the name of the data set used]
|
|
read_csv("path/nhanes.csv") [Read nhanes.csv in the path/ folder (readr)]
|
|
View(df) [View tabular data frame df in a graphical viewer]
|
|
mean, median, range [Descriptive stats. Remember na.rm=TRUE if desired]
|
|
filter(df, ..,) [ Filters data frame according to condition ... (dplyr)]
|
|
factor(x, levels=c("wt", "mutant")) [Create factor specifying level order]
|
|
relevel(x, ref="wildtype") [Re-level a factor variable]
|
|
t.test(y~grp, data=df) [T-test mean y across grp in data df]
|
|
lmfit <- lm(y~x1+x2, data=df) [Fit linear model y against two x’s]
|
|
anova(lmfit) [Print ANOVA table on object returned from lm()]
|
|
summary(lmfit) [Get summary information about a model fit with lm()]
|
|
TukeyHSD(aov(lmfit)) [ANOVA Post-hoc pairwise contrasts]
|
|
wilcox.test(y~grp, data=df) [Wilcoxon rank sum / Mann-Whitney U test]
|
|
xt <- xtabs(~x1+x2, data=df) [Cross-tabulate a contingency table]
|
|
addmargins(xt) [Adds summary margin to a contingency table xt]
|
|
addmargins(xt) [Adds summary margin to a contingency table xt]
|
|
chisq.test(xt) [Chi-square test on a contingency table xt]
|
|
fisher.test(xt) [Fisher’s exact test on a contingency table xt]
|
|
mosaicplot(xt) [Mosaic plot for a contingency table xt]
|
|
power.t.test(n, power, sd, delta) [T-test power calculations]
|
|
power.prop.test(n, power, p1, p2) [Proportions test power calculations]
|
|
tidy() augment() glance() [Model tidying functions in the broom package]
|
|
is.na(x)
|
is.null(x)
|
|
is.nan(x)
|
is.array(x)
|
|
is.data.frame(x)
|
is.numeric(x)
|
|
is.complex(x)
|
is.character(x)
|
|
head(x)
|
tail(x)
|
|
summary(x)
|
str(x)
|
|
length(x)
|
dim(x)
|
|
dimnames(x)
|
attr(x,which)
|
|
nrow(x)
|
ncol(x)
|
|
NROW(x)
|
NCOL(x)
|
|
class(x)
|
unclass(x)
|
|
createDataPartition(y,p=0.8) [createDaIt splits a vector 'y' with 80 percent data in one part and 20 percent in other parttaPartition(y,p=0.8)]
|
|
trainControl( summaryFunction=<Rfunction>,classProbs=<logical>) [It is used for controlling training parameters like resampling, number of folds, iteration etc.]
|
|
densityplot.rfe(x,data,...) [Lattice functions for plotting resampling results of recursive feature selection]
|
|
featureplot(x,y,plot...) [A shortcut to produce lattice plots]
|
No comments yet.
Cheatography n. The art of mapping complex subjects into clear, instantly navigable reference guides.
