Learn R Programming

ck37r (version 1.0.3)

Chris Kennedy's R Toolkit

Description

Toolkit for statistical, machine learning, and targeted learning analyses. Functionality includes loading & auto-installing packages, standardizing datasets, creating missingness indicators, imputing missing values, creating multicore or multinode clusters, automatic SLURM integration, enhancing SuperLearner and TMLE with automatic parallelization, and many other SuperLearner analysis & plotting enhancements.

Copy Link

Version

Install

install.packages('ck37r')

Monthly Downloads

16

Version

1.0.3

License

MIT + file LICENSE

Issues

0

Pull Requests

0

Stars

19

Forks

5

Maintainer

Chris Kennedy

Last Published

February 6th, 2020

Functions in ck37r (1.0.3)

SL.xgboost_cv

XGBoost SuperLearner wrapper with internal cross-validation for early-stopping
auc_table.CV.SuperLearner

Table of cross-validated AUCs from CV.SuperLearner result
auc_table

Generate a table of AUCs by learner
cvsl_auc

Calculate cross-validated AUC from CV.SuperLearner result
auc_table.SuperLearner

Table of cross-validated AUCs from SuperLearner result
cvsl_auc_table

Deprecated, use auc_table() instead.
h2o_init_multinode

Starts an h2o cluster on multiple nodes
import_csvs

Import all CSV files in a given directory and save them to a list.
parallelize

Setup parallel processing using snow.
plot_roc.SuperLearner

Plot a ROC curve from cross-validated AUC from SuperLearner
plot_tree

Plot rpart decision tree with pretty good defaults.
sl_auc_table

Deprecated, use auc_table() instead.
SL.mgcv

Wrapper for mgcv's gam implementation
SL.h2o_auto

Automatic machine learning using h2o
plot.SuperLearner

Plot estimated risk and confidence interval for each learner
categoricals_to_factors

Convert categorical numerics to factors
prauc_table

Generate a table of PR-AUCs by learner
create_dataset

Create a dataset
sl_plot_roc

Deprecated, please use plot_roc() instead.
prauc_table.SuperLearner

Table of PR-AUCs from SuperLearner result
cvsl_plot_roc

Deprecated, please use plot_roc() instead.
cvsl_weights

Create a table of meta-weights from a CV.SuperLearner
factors_to_indicators

Convert factors to indicator variables.
gen_superlearner

Setup a SuperLearner() based on parallel configuration.
load_all_code

Load all R files in a library directory.
Mode

Compute the mode of a vector (can be multiple results).
SL.glmnet2

Elastic net regression, including lasso and ridge
impute_missing_values

Impute missing values in a dataframe and add missingness indicators.
load_packages

Load a list of packages.
plot_roc.CV.SuperLearner

Plot a ROC curve from cross-validated AUC from CV.SuperLearner
setup_parallel_tmle

Setup TMLE to run in parallel
rf_count_terminal_nodes

Count the terminal nodes in each tree from a random forest
plot_roc

Plot the ROC curve for an ensemble object.
prauc_table.CV.SuperLearner

Table of cross-validated PR-AUCs from CV.SuperLearner result
prauc

Calculate PR-AUC from cross-validation results
standardize

Rescale variables, possibly excluding some columns
vim_corr

Correlation analysis
missingness_indicators

Return matrix of missingness indicators for a dataframe or matrix.
sl_stderr

Calculate the SE of individual SL learners
stop_cluster

Stop the cluster if snow::makeCluster() was used, but nothing needed if doMC was used.
tmle_parallel

Modify TMLE to support parallel computation for g and Q.