Nested Cross Validation for the Relaxed Lasso and Other Machine Learning Models

---

Documentation for package ‘glmnetr’ version 0.4-6

• DESCRIPTION file.
• User guides, package vignettes and other documentation.

Help Pages

aicreg	Identify model based upon AIC criteria from a stepreg() putput
ann_tab_cv	Fit an Artificial Neural Network model on "tabular" provided as a matrix, optionally allowing for an offset term
ann_tab_cv_best	Fit multiple Artificial Neural Network models on "tabular" provided as a matrix, and keep the best one.
best.preds	Get the best models for the steps of a stepreg() fit
bsint	Construct the bias terms for going from model layer to layer to carry forward an offset to mimic a linear model
calceloss	calculate cross-entry for multinomial outcomes
calplot	Construct calibration plots for a nested.glmnetr output object
calplot0	Construct a single calibration plot for a nested.glmnetr output object
cox.sat.dev	Calculate the CoxPH saturated log-likelihood
cv.glmnetr	Get a cross validation informed relaxed lasso model fit.
cv.stepreg	Cross validation informed stepwise regression model fit.
devrat_	Calculate deviance ratios for CV based
diff_time	Output to console the elapsed and split times
diff_time1	Get elapsed time in c(hour, minute, secs)
dtstndrz	Standardize a data set
factor.foldid	Generate foldid's by factor levels
get.foldid	Get foldid's with branching for cox, binomial and gaussian models
getlamgam	get numerical values for lam and gam
glmnetr	Fit relaxed part of lasso model
glmnetr.cis	Calculate performance measure CI's and p's
glmnetr.cis0	Calculate performance measure CI's and p's
glmnetr.compcv	Compare cross validation fits from a nested.glmnetr output.
glmnetr.compcv0	Calculate agreement differences with CI and p
glmnetr.simdata	Generate example data
glmnetrll_1fold	Evaluate fit of leave out fold
glmnetr_devratio	Get Deviance ratio.
glmnetr_seed	Get seeds to store, facilitating replicable results
myaxis	Un-log the log(HR)'s for plotting
myrug	A customized rug
nested.glmnetr	Using (nested) cross validation, describe and compare some machine learning model performances
plot.cv.glmnetr	Plot cross-validation deviances, or model coefficients.
plot.glmnetr	Plot the relaxed lasso coefficients.
plot.nested.glmnetr	Plot results from a nested.glmnetr() output
plot_perf_glmnetr	Plot nested cross validation performance summaries
predict.cv.glmnetr	Give predicteds based upon a cv.glmnetr() output object.
predict.cv.stepreg	Beta's or predicteds based upon a cv.stepreg() output object.
predict.glmnetr	Get predicteds or coefficients using a glmnetr output object
predict.nested.glmnetr	Give predicteds based upon the cv.glmnet output object contained in the nested.glmnetr output object.
predict_ann_tab	Get predicteds for an Artificial Neural Network model fit in nested.glmnetr()
predict_nested_rf	Get predicteds for a rf fit from nested.glmnetr() output object
predict_nested_xgb	Get predicteds for a XGB fit from a nested.glmnetr() output object
prednn_tl	predicted values from an ann_tab_cv output object based upon the model and its lasso model used for generating an offset
preds_1	Get predictors form a stepwise regression model.
print.nested.glmnetr	A redirect to the summary() function for nested.glmnetr() output objects
print.rf_tune	Print output from rf_tune() function
rederive_rf	Rederive Random Forest models not kept in nested.glmnetr() output
rederive_xgb	Rederive XGB models not kept in nested.glmnetr() output
rf_tune	Fit a Random Forest model on data provided in matrix and vector formats.
rf_xbhat	get XBeta from an rfsrc output object
roundperf	round elements of a summary.glmnetr() output
rpart_xbhat	get XBeta from an rpart output object
stepreg	Fit the steps of a stepwise regression.
summary.cv.glmnetr	Output summary of a cv.glmnetr() output object.
summary.cv.stepreg	Summarize results from a cv.stepreg() output object.
summary.nested.glmnetr	Summarize a nested.glmnetr() output object
summary.rf_tune	Summarize output from rf_tune() function
summary.stepreg	Briefly summarize steps in a stepreg() output object, i.e. a stepwise regression fit
wtlast	Construct the weights for going from the last hidden layer to the last layer of the model, not counting any activation, to carry forward an offset to mimic a linear model
wtmiddle	Construct the weights for going between two hidden layers, carrying forward an offset term to mimic a linear model
wtzero	Construct the weights for going from the observed data with an offset in column 1 to the first hidden layer
xgb.simple	Get a simple XGBoost model fit (no tuning)
xgb.tuned	Get a tuned XGBoost model fit
xgb_xbhat	get XBeta from an XGB.train output object